Ribozymes in The Gene

COMPANY
R.G.LANDES
MEDICAL INTELLIGENCE UNIT
Genetic Mechanisms in Multiple Endocrine Neoplasia Type 2 c-Myc Function in Neoplasia
MEDICAL INTELLIGENCE UNIT 4
Barry D. Nelkin, Johns Hopkins University Chi V. Dang and Linda A. Lee, Johns Hopkins University
Functional Heterogeneity of Liver Tissue: From Cell Lineage Endothelins
Diversity to Sublobular Compartment-Specific Pathogenesis David J. Webb and Gillian Gray, University of Edinburgh
Fernando Vidal-Vanaclocha, Universidad del Pais Vasco
SCANLON • KASHANI-SABET
p53 B Cells and Autoimmunity
Host Response to Intracellular Pathogens Christian Boitard, Hôpital Necker-Paris
Stefan H.E. Kaufmann, Institute für Mikrobiologie und
Immunologie der Universität Ulm Hyperacute Xenograft Rejection Kevin J. Scanlon • Mohammed Kashani-Sabet
Jeffrey Platt, Duke University
Myocardial Injury: Laboratory Diagnosis
Johannes Mair and Bernd Puschendorf, Universität Innsbruck Transplantation Tolerance
J. Wesley Alexander, University of Cincinnati
Cellular Inter-Relationships in the Pancreas: Implications for
Islet Transplantation
Lawrence Rosenberg and William P. Duguid, McGill University
Anti-HIV Nucleosides: Past, Present and Future

Hiroaki Mitsuya, National Cancer Institute
Premalignancy and Tumor Dormancy
Eitan Yefenof, Hebrew University - Hadassah Medical School
Richard H. Scheuerman, University of Texas Southwestern
Myocardial Preconditioning
Ribozymes in the Gene
Therapy of Cancer
Cherry L. Wainwright and James R. Parratt,
Heat Shock Response and Organ Preservation University of Strathclyde
George Perdrizet, University of Connecticut
Cytokines and Inflammatory Bowel Disease
Glycoproteins and Human Disease Claudio Fiocchi, Case Western Reserve
Inka Brockhausen, Hospital for Sick Children—Toronto
Bone Metastasis
Exercise Immunology F. William Orr and Gurmit Singh, University of Manitoba
Bente Klarlund Pedersen, Rigshospitalet—Copenhagen
Cancer Cell Adhesion and Tumor Invasion
Chromosomes and Genes in Acute Lymphoblastic Leukemia
Lorna M. Secker-Walker, Royal Free Hospital-London
Pnina Brodt, McGill University
MIU
Surfactant in Lung Injury and Lung Transplantation
James F. Lewis, Lawson Research Institute
Cutaneous Leishmaniasis
Felix J. Tapia, Instituto de Medicina-Caracas 4
Richard J. Novick, Roberts Research Institute Molecular Basis of Autoimmune Hepatitis
Ruud A.W. Veldhuizen, Lawson Research Institute Ian G. McFarlane and Roger Williams, King’s College Hospital
Ribozymes in the Gene Therapy of Cancer

Gamma Interferon in Antiviral Defense Estrogen and Breast Cancer
Gunasegaran Karupiah, The John Curtin School of Medical W.R. Miller, University of Edinburgh
Research—The Australian National University Molecular Mechanisms of Hypercoagulable States
Management of Post-Open Heart Bleeding Andrew I. Schafer, University of Texas-Houston
Rephael Mohr, Jacob Lavee and Daniel A. Goor, Organ Procurement and Preservation for Transplantation
The Chaim Sheba Medical Center Luis Toledo-Pereyra, Michigan State University
The Glycation Hypothesis of Atherosclerosis Liver Stem Cells
Camilo A.L.S. Colaco, Qandrant Research Institute Stewart Sell and Zoran Ilic, Albany Medical College
Cytokines in Reproduction: Molecular Mechanisms of Fetal Skin Substitute Production by Tissue Engineering
Allograft Survival Mahmoud Rouabhia, Laval University
Gary W. Wood, University of Kansas
HIV and Membrane Receptors
Computers in Clinical Medicine Dimitri Dimitrov, National Institutes of Health
Eta Berner, University of Alabama Christopher C. Broder, Uniformed Servives University of the
Cellular & Molecular Biology of Airway Chemoreceptors Health Sciences
Ernest Cutz, University of Toronto Interferon-Inducible Genes
Immunology of Pregnancy Maintenance in the First Trimester Ganes Sen and Richard Ransohoff, Case Western Reserve
Joseph Hill, Harvard University University
Peter Johnson, University of Liverpool Artificial Neural Networks in Medicine
Inherited basement Membrane Disorders Vanya Gant and R. Dybowski, St. Thomas Medical School—
Karl Tryggvaso, Karolinska Institute London
Cartoid Body Chemoreceptors von Willebrand Factor

Constancio Gonzalez, Universidad de Madrid Zaverio M. Ruggeri, Scripps Research Institute
Molecular Biology of Leukocyte Chemostasis Immune Mechanisms in Atherogenesis

Antal Rot, Sandoz Forschungsinstitut—Vienna Ming K. Heng, UCLA
Breast Cancer Screening The Biology of Germinal Centers in Lymphoral Tissue

Ismail Jatoi, Brook Army Medical Center G.J. Thorbecke and V.K. Tsiagbe, New York University
R.G. LANDES R.G. LANDES

C OM PA N Y C OM PA N Y
MEDICAL
INTELLIGENCE
UNIT 4

Therapy of Cancer
Kevin J. Scanlon, Ph.D.

Berlex Biosciences
Cancer Research Department
Richmond, California, U.S.A.
Mohammed Kashani-Sabet, M.D.

U.C.S.F. Cancer Center
University of California San Francisco
San Francisco, California, U.S.A.
R.G. LANDES
COMPANY
AUSTIN, TEXAS
U.S.A.
R.G. LANDES COMPANY

Austin, Texas, U.S.A.
Copyright © 1998 R.G. Landes Company
All rights reserved.
No part of this book may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopy, recording, or any information storage and
retrieval system, without permission in writing from the publisher.
Printed in the U.S.A.
Please address all inquiries to the Publishers:
R.G. Landes Company, 810 South Church Street, Georgetown, Texas, U.S.A. 78626
Phone: 512/ 863 7762; FAX: 512/ 863 0081
ISBN: 1-57059-552-6
While the authors, editors and publisher believe that drug selection and dosage and the specifica-
tions and usage of equipment and devices, as set forth in this book, are in accord with current
recommendations and practice at the time of publication, they make no warranty, expressed or
implied, with respect to material described in this book. In view of the ongoing research, equipment
development, changes in governmental regulations and the rapid accumulation of information
relating to the biomedical sciences, the reader is urged to carefully review and evaluate the informa-
tion provided herein.
Library of Congress Cataloging-in-Publication Data
Ribozymes in gene therapy of cancer / [edited by] Kevin J. Scanlon, Mohammed

Kashani-Sabet.
p. cm.--(Biotechnology intelligence unit)
Includes bibliographical references and index.
ISBN 1-57059-552-6
1. Cancer--Gene therapy. 2. Catalytic RNA--Therapeutic use. I. Scanlon, Kevin
J., 1947- . II. Kashani-Sabet, Mohammed. III. Series.
[DNLM: 1. Neoplasms--therapy. 2 RNA, Catalytic--therapeutic
use. 3. Gene Therapy--methods. QZ 266 R486 1998]
RC271.G45R53 1998
616.99'4042--dc21
DNLM/DLC 98-28767
for Library of Congress CIP
MEDICAL
INTELLIGENCE
UNIT 4
PUBLISHER’S NOTE
Landes Bioscience produces books in six Intelligence Unit series:
Medical, Molecular Biology, Neuroscience, Tissue Engineering,
Therapy of Cancer
Biotechnology and Environmental. The authors of our books are
acknowledged leaders in their fields. Topics are unique; almost
without exception, no similar books exist on these topics.
Our goal is to publish books in important and rapidly changing

areas of bioscience for sophisticated researchers and clinicians. To
achieve this goal, we have accelerated our publishing program to
conform to the fast pace at which information grows in bioscience.
Most of our books are published within 90 to 120 days of receipt of
Berlex Biosciences
the manuscript.Cancer
We would
ResearchtoDepartment
like thank our readers for their
continuing interest and welcome any comments
Richmond, California, U.S.A. or suggestions they
may have for future books.
Judith Kemper
U.C.S.F. Cancer CenterProduction Manager
R.G. Landes Company
R.G. LANDES
COMPANY
AUSTIN, TEXAS
U.S.A.
CONTENTS
Section I: Biochemistry of Ribozymes
1. The Biochemistry of the Hammerhead Ribozyme .................................. 3
Philip C. Turner
The Hammerhead Motif ......................................................................... 3
Mutagenesis and Sequence Requirements ............................................. 4
Hammerhead Ribozyme Structure ......................................................... 5
Reaction Mechanism and Kinetics ......................................................... 5
Specificity ................................................................................................. 7
Modified Hammerhead Ribozymes ....................................................... 8
Pharmacokinetic and Cellular Uptake Studies ...................................... 9
Making Hammerhead Ribozymes Work In Vivo .................................. 9
Target Site Selection .............................................................................. 10
Hammerhead Ribozyme Design for In Vivo Expression .................... 10
Conclusion ............................................................................................. 11
2. Biochemistry of the Hairpin Ribozyme ................................................. 15
Andrew Siwkowski and Arnold Hampel
Introduction ........................................................................................... 15
Biochemistry and Mechanism of the Reaction .................................... 15
Conclusions ........................................................................................... 21
3. Biochemistry of Hepatitis Delta Virus Catalytic RNAs ........................ 23
N. Kyle Tanner
Introduction ........................................................................................... 23
Properties of HDV ................................................................................. 23
Properties of the Catalytic Domains ..................................................... 24
Strain Comparisons ............................................................................... 27
Data Summary ....................................................................................... 28
Self-Cleavage Reaction .......................................................................... 30
Biological Relevance .............................................................................. 30
Trans-Cleaving Activity ......................................................................... 32
Applications ........................................................................................... 34
Concluding Remarks ............................................................................. 34
Section II: Expression and Delivery of Ribozymes ......................................... 39

4. Exogenous Delivery of Ribozymes ......................................................... 41
Mark A. Reynolds
Introduction ........................................................................................... 41
Tissue Culture Studies ........................................................................... 41
Localized Delivery In Vivo .................................................................... 45
Systemic Delivery ................................................................................... 50
Future Applications ............................................................................... 54
5. Novel RNA Motif (Dimeric Minizyme) Capable of Cleaving L6
BCR-ABL Fusion (b2a2) mRNA with High Specificity ........................ 61
Tomoko Kuwabara, Masaki Warashina and Kazunari Taira
Introduction ........................................................................................... 61
Current Research ................................................................................... 62
Future Prospects .................................................................................... 74
6. Using Ribozymes to Attenuate Gene Expression

in Transgenic Mice .................................................................................. 79
Shimon Efrat
Introduction ........................................................................................... 79
Examples of Ribozyme Applications in Transgenic Mice ................... 79
Future Directions ................................................................................... 85
7. Retroviral Delivery of Ribozymes .......................................................... 87

Lun-Quan Sun and Geoff Symonds
Retroviral Properties and Life Cycle ..................................................... 87
Retroviral Vectors and Their Design .................................................... 87
Cell Type Specific Promoters ................................................................ 90
Packaging Cell Lines .............................................................................. 90
Targeting of Retroviruses ...................................................................... 90
Ribozyme Action ................................................................................... 91
Murine Model Systems for Retroviral-Ribozyme Delivery
and Inhibition of Gene Expression .................................................. 92
Anti-HIV Retroviral Ribozyme Constructs ......................................... 93
General Aspects of Retroviral and Other Delivery Systems ................ 95
Clinical Application of Retroviral Delivery of Ribozymes .................. 96
Concluding Remarks ............................................................................. 97
8. Adeno-Associated Virus (AAV) Mediated Ribozyme Expression
in Mammalian Cells .............................................................................. 101
Piruz Nahreini and Beth Roberts
Introduction ......................................................................................... 101
Generation of RecombinantAAV ....................................................... 104
rAAV Infection and Transgene Expression ........................................ 109
Concluding Remarks ........................................................................... 116
Section III: Ribozyme Targets for Gene Therapy .......................................... 123
9. Applications of Anti-Oncogene Ribozymes for the Treatment

of Bladder Cancer .................................................................................. 125
Akira Irie and Eric J. Small
Overview of Bladder Cancer ............................................................... 125
Molecular Genetics of Bladder Cancer ............................................... 126
H-ras Oncogene in Bladder Cancer .................................................... 127
Anti-Oncogene Ribozymes Against Bladder Cancer ......................... 128
Delivery of Ribozymes to Bladder Tumors ........................................ 130
Conclusions ......................................................................................... 131
10. Gene Therapy of Breast Cancer ............................................................ 135

T. Suzuki and B. Anderegg
Introduction ......................................................................................... 135
Anti-c-erbB-2 Ribozyme Strategy ....................................................... 135
Anti-c-erbB-2 Ribozyme Expression .................................................. 136
Future Prospects .................................................................................. 138
Conclusions ......................................................................................... 139
11. Therapeutic Application of an Anti-ras Ribozyme in Human
Pancreatic Cancer .................................................................................. 143
Hiroshi Kijima and Kevin J. Scanlon
Abstract ................................................................................................ 143
Introduction ......................................................................................... 143
Results .................................................................................................. 144
Discussion ............................................................................................ 146
Conclusion ........................................................................................... 147
12. The Use of Ribozymes for Gene Therapy of Lung Cancer .................. 151
Alex W. Tong, Yu-An Zhang, David Y. Bouffard, John Nemunaitis
Summary .............................................................................................. 151
Introduction ......................................................................................... 152
Growth Inhibition by Antisense Oligonucleotides (AS-ODN) ........ 152
Growth Inhibition by Ribozyme ......................................................... 153
Future Considerations ......................................................................... 158
13. Ribozymes in Gene Therapy of Prostate Cancer ................................. 165

Dale J. Voeks, Gary A. Clawson, and James S. Norris
Introduction ......................................................................................... 165
Ribozymes in Prostate Cancer ............................................................ 166
Conclusions ......................................................................................... 170
14. Ribozymes in Targeting Tumor Suppressor Genes ............................. 175
Tapas Mukhopadhyay and Jack A. Roth
Introduction ......................................................................................... 175
General Considerations for Ribozyme Function ............................... 176
Targets for Ribozymes in Cancer ........................................................ 176
Cellular Genes Targeted by Ribozyme Molecules ............................. 177
Conclusion ........................................................................................... 180
15. Inhibition of the Multidrug Resistance Phenotype
by Different Delivery Systems of an Anti-mdr Ribozyme .................. 183
Per Sonne Holm, David T. Curiel and Manfred Dietel
Introduction ......................................................................................... 183
Catalytic Activity of the mdr Ribozyme ............................................. 184
Delivery Systems of the Anti-mdr Ribozyme ..................................... 186
Conclusion ........................................................................................... 191
16. Anti-BCR-ABL Ribozymes ................................................................... 195
Lance H. Leopold, Scott K. Shore and E. Premkumar Reddy
Introduction ......................................................................................... 195
Current Research ................................................................................. 196
Future Directions ................................................................................. 206
17. Potential Design and Facilitation of Hammerhead Ribozyme

Turnover by Cellular Proteins .............................................................. 213
Mouldy Sioud
Introduction ......................................................................................... 213
General Aspects for Ribozyme Design ............................................... 214
Effect of Proteins on Ribozyme Cleavage ........................................... 215
Ribozymes as Gene Therapy in Cancer Treatment ........................... 216
18. Human Papillomaviruses ..................................................................... 223
E.J. Shillitoe
Introduction ......................................................................................... 223
Papillomaviruses .................................................................................. 223
The Viral Genome ............................................................................... 224
Mechanism of Oncogenesis ................................................................ 224
Anti-Papillomavirus Gene Therapy .................................................... 226
Antisense Inhibition of Expression of HPV-Genes ........................... 226
Inhibition of HPV by Ribozymes ....................................................... 227
Delivery Methods for Anti-HPV Ribozymes ..................................... 228
Future Directions for Research ........................................................... 229
Index ................................................................................................................ 235
EDITORS
Cancer Research Department
Berlex Biosciences
Chapter 11

U.C.S.F. Cancer Center
CONTRIBUTORS
B. Anderegg, Ph.D. Shimon Efrat
Department of Cancer Research Department of Molecular
Berlex Biosciences Pharmacology
Richmond, Califonia, U.S.A. Albert Einstein College of Medicine
Chapter 10 Bronx, New York, U.S.A.
Chapter 6
David Y. Bouffard, Ph.D.
Berlex Biosciences Arnold Hampel, Ph.D.
Richmond, California, U.S.A. Department of Biological Sciences
Chapter 12 Northern Illinois University
DeKalb, Illinois, U.S.A.
Gary A. Clawson Chapter 2
Penn State University
Hershey, Pennsylvania, U.S.A. Per Sonne Holm, Ph.D.
Chapter 13 Universitätsklinikum Charité
Institute of Pathology
David T. Curiel, M.D. Berlin, Germany
Gene Therapy Program Chapter 15
The University of Alabama at
Birmingham Akira Irie, M.D.
Birmingham, Alabama, USA Department of Cancer Research
Chapter 15 Berlex Biosciences
Manfred Dietel, Prof., M.D. Chapter 9
Universitätsklinikum Charité
Institute of Pathology Hiroshi Kijima, M.D.
Berlin, Germany Department of Pathology
Chapter 15 Tokai University School of Medicine
Bohseidai, Isehara, Kanagawa, Japan
Chapter 11
Tomoko Kuwabara, Ph.D. Mark A. Reynolds, Ph.D.
National Institute for Advanced Ribozyme Pharmaceuticals, Inc.
Interdisciplinary Research Boulder, Colorado, U.S.A.
National Institute of Bioscience and Chapter 4
Human Technology; and
Institute of Applied Biochemistry Beth Roberts, M.S.
University of Tsukuba Ribozyme Pharmaceuticals Inc.
Tsukuba Science City, Japan Boulder, Colorado, U.S.A.
Chapter 5 Chapter 8
Lance H. Leopold, M.D. Jack A. Roth, M.D.

The Fels Institute for Cancer Research Section of Thoracic Molecular Oncology
and Molecular Biology Department of Thoracic and
Temple University School of Medicine Cardiovascular Surgery and
Philadelphia, Pennsylvania, U.S.A. Department of Tumor Biology
Chapter 16 The University of Texas M.D. Anderson
Cancer Center
Tapas Mukhopadhyay, Ph.D. Houston, Texas, U.S.A.
Section of Thoracic Molecular Chapter 14
OncologyDepartment of Thoracic
and Cardiovascular Surgery E.J. Shillitoe
The University of Texas M.D. Anderson Department of Microbiology and
Cancer Center Immunology
Houston, Texas, U.S.A. SUNY College of Medicine
Chapter 14 Syracuse, New York, U.S.A.
Chapter 18
Piruz Nahreini, Ph.D.
Ribozyme Pharmaceuticals Inc. Scott K. Shore
Boulder, Colorado, U.S.A. The Fels Institute for Cancer Research
Chapter 8 and Molecular Biology
Temple University School of Medicine
John Nemunaitis, M.D. Philadelphia, Pennsylvania, U.S.A.
Mary Crowley Cancer Research Program Chapter 16
Baylor Research Institute
Baylor University Medical Center Mouldy Sioud
Dallas, Texas, U.S.A. Department of Immunology
Chapter 12 Institute for Cancer Research
The Norwegian Radium Hospital
James S. Norris, Ph.D. Montebello, Oslo, Norway
Medical University of South Carolina Chapter 17
Charleston, South Carolina, U.S.A.
Chapter 13 Andrew Siwkowski, M.S.
Department of Biological Sciences
E. Premkumar Reddy, Ph.D. Northern Illinois University
Department of Biochemistry DeKalb, Illinois, U.S.A.
Temple University Chapter 2
Philadelphia, Pennsylvania, U.S.A.
Chapter 16
Eric J. Small Alex W. Tong, Ph.D.
Department of Medicine and Urology Cancer Immunology Research
University of California at San Francisco Laboratory
San Francisco, California, U.S.A. Baylor-Sammons Cancer Center
Chapter 9 Baylor University Medical Center
Dallas, Texas, U.S.A.
Lun-Quan Sun, Ph.D. Chapter 12
Johnson & Johnson Research
Laboratories Philip C. Turner, Ph.D.
Sydney, Australia School of Biological Sciences
Chapter 7 University of Liverpool
Liverpool,U.K.
T. Suzuki, M.D. Chapter 1
Department of Cancer Research
Berlex Biosciences Dale J. Voeks
Richmond, Califonia, U.S.A. Medical University of South Carolina
Chapter 10 Charleston, South Carolina, U.S.A.
Chapter 13
Geoff Symonds, Ph.D.
Johnson & Johnson Research Masaki Warashina, Ph.D.
Laboratories National Institute for Advanced
Sydney, Australia Interdisciplinary Research
Chapter 7 National Institute of Bioscience and
Kazunari Taira, Prof. Ph.D. Institute of Applied Biochemistry
National Institute for Advanced University of Tsukuba
Interdisciplinary Research Tsukuba Science City, Japan
National Institute of Bioscience and Chapter 5
Institute of Applied Biochemistry Yu-An Zhang
University of Tsukuba Cancer Immunology Research
Tsukuba Science City, Japan Laboratory
Chapter 5 Baylor-Sammons Cancer Center
Baylor University Medical Center
N. Kyle Tanner, Ph.D. Dallas, Texas, U.S.A.
Département de Biochimie Médicale Chapter 12
Centre Médical Univeritaire
Genéve, Switzerland
Chapter 3
FOREWORD
T he field of ribozymes has come a long way since the initial observations of
cis-acting self-cleaving molecules in Tetrahymena and RNase P. Along with
the discovery of different types of catalytic RNAs in diverse biological systems
has come the demonstration of the biological utility of ribozyme-mediated
genetic manipulation. Ribozymes have clearly found a place alongside anti-
bodies and antisense oligonucleotides in the armamentarium used to disrupt
target-specific gene expression. Beyond target validation, ribozymes are being
increasingly investigated for their therapeutic potential.
The time is therefore ripe to review some of the recent developments in
the field of ribozyme biochemistry and to examine their use in biological sys-
tems related to cancer. Advances in gene therapy offer the promise of a signifi-
cant impact on the management of intractable diseases. Cancer represents one
of the most important targets of gene therapy studies and, not surprisingly,
much of the biological investigations using ribozymes have occurred in the
realm of cancer.
In order to conduct a comprehensive review of the relevant aspects of
ribozyme technology, this Medical Intelligence Unit volume is composed of
three sections. The first section serves as an overview of the biochemistry of
hammerhead, hairpin and hepatitis delta ribozymes. This is followed by five
chapters discussing the ways in which optimal delivery and expression of
ribozymes may be achieved. The issue of adenoviral delivery of ribozymes has
been left to the individual chapters in which this modality has been utilized.
The final section contains manuscripts discussing the multiple targets of
ribozyme action with relevance to cancer research. We hope that the reader will
enjoy these up-to-date discussions on the latest developments in ribozyme tech-
nology which will pave its path to the clinical arena.

Section I
Biochemistry of Ribozymes
CHAPTER 1
The Biochemistry
of the Hammerhead Ribozyme
Philip C. Turner
E xperiments in the early nineteen eighties led to the discovery of catalytic RNAs
(ribozymes), including self-cleaving introns1,2 and the RNA component of the tRNA
processing enzyme RNase P.3,4 Later, examples of the class of ribozymes known as hammer-
head ribozymes were discovered during studies of the replication of plant viruses and
virusoids.5-7 These small, pathogenic plant RNAs undergo a self-catalyzed cleavage reaction
to generate monomeric genomes, and this reaction can be replicated in vitro at neutral pH
in the presence of Mg2+.8,9 Other types of ribozymes have been discovered, including the
hairpin ribozyme10,11 and the hepatitis delta virus ribozyme12 which are discussed in chap-
ters 2 and 3.
The Hammerhead Motif

By comparing sequences of the hammerhead catalytic motif from a variety of sources,
and combining it with information from mutagenesis studies, the conserved and important
features of the motif have been elucidated; they are shown in Figure 1.1. For a recent de-
tailed review, see ref. 13. It was found to consist of 3 base paired helices, or stems, (I-III),
with helices I and III flanking the cleavage site which is on the 3' side of an unpaired, or
bulged, nucleotide. Helix II is connected to the other helices by two single-stranded regions
that contain most of the conserved nucleotides in the hammerhead motif. The numbering
system of Hertel et al14 is useful for descriptive and comparative purposes and is used in
Figure 1.1. As shown, of the nucleotides in the single stranded regions which presumably
form the hammerhead’s catalytic core, only the U7 position shows variations in nature.15
The sequences of helices I-III show little conservation except at the ends of helix II, where a
purine, R10.1, pairs with a pyrimidine Y11.1, and helix III where A15.1 base pairs with
U16.1.
Hammerhead ribozymes, as they exist in nature, self-cleave and hence work in cis. This
is achieved by the RNA folding to form loops at the ends of two of the helices. In vitro,
however, it is possible to assemble the catalytically active motif shown in Figure 1.1 in a
variety of ways, using either 1, 2 or 3 RNA molecules.16-20 If only one RNA molecule is
involved, then two of the helices are joined by loops and cleavage is in cis (self-cleavage). If
two RNA molecules are used, then only one loop is required between a pair of helices and
cleavage occurs in trans.20 Finally, using 3 RNA molecules there are no loops connecting the
ends of the helices and cleavage is also in trans. In the case of two RNA molecules, there are
three ways of dividing the hammerhead motif, which have been called I/II, II/III and I/III
Ribozymes in the Gene Therapy of Cancer, edited by Kevin J. Scanlon and Mohammed Kashani-Sabet.
©1998 R.G. Landes Company.
4 Ribozymes in the Gene Therapy of Cancer
Fig. 1.1. Diagram of the hammerhead ribozyme. The bond cleaved is indicated by an
arrow. N stands for any nucleotide, H is A, C or U, but not G. Y (pyrimidine) is C or
U and R (purine) is A or G. Numbering is according to ref. 14. In nature one of the
helices is not terminated by a loop. Conserved nucleotides are all those other than N,
with the exception of U7. Domain I of the catalytic core is nucleotides C3-A6 and
domain II is U7-A9 and G12-A14.
depending on which helices are used to bind to the substrate, or cleaved, strand. The design
called I/III, in which the remaining loop terminates helix II, contains almost all the con-
served residues in the enzyme strand.17 Hence, the substrate strand has only a minimal
sequence requirement (5'-U-H-3'). This design has been widely utilized because it is not
only ideal for use in attempts to downregulate gene expression, but also permits relatively
easy chemical synthesis of the short enzyme strand of RNA. In this design the substrate
strand is commonly referred to as the target RNA which is to be cleaved, and the enzyme
strand is generally called the ribozyme.
Mutagenesis and Sequence Requirements

In the single stranded regions of the hammerhead ribozyme’s catalytic core, nucleotide
substitutions at all positions except U7 destroy catalytic activity and this fact can be utilized
to create non-cleaving ribozymes as negative controls, for example, to determine antisense
effects. Ruffner et al15 have shown that catalytic rates in vitro are best with U at position 7,
but G>A>C, which is only 20% as active.
Conserved nucleotides are not present in helix I, but helix II contains two immediately
adjacent to the catalytic core. These are most commonly a G-C base pair at R10.1 - Y11.1,
and this combination is the most active in vitro. Although only one nucleotide pair is con-
served in helix II, its absolute length has an effect on ribozyme activity in vitro.7,15,21,22 If
The Biochemistry of the Hammerhead Ribozyme 5
helix II is shorter than 2 bp, activity is severely reduced. The loop terminating helix II, which
is usually 4 nucleotides long, has little apparent influence on activity in vitro and can even
be replaced by non-nucleoside linkers23-25 or deoxynucleotides26 without total loss of activity.
Helix III contains the conserved A15.1 - U16.1 base pair and the U16.1 is found as part
of the cleaved triplet, which is most often 5'-GUC-3', although the triplets 5'-GUA-3' and
5'-AUA-3' are seen as cleavage sites in nature, though rarely.27,28 In vitro analysis of system-
atic mutations in the GUC sequence in the target have shown that cleavage can occur after
any NUH triplet15,29-31 (H means any nucleotide but G), though efficiency is usually much
reduced compared to GUC. Shimayama et al30 showed that, in vitro, the efficiency of cleav-
age at the various triplet combinations depended on the relative concentrations of the
ribozyme and substrate. If either were saturating, cleavage depended on kcat and GUC,
AUC>GUA, AUA, CUC were much more efficient than other versions of NUH. If ribozyme
or substrate was limiting, then GUC>CUC and all other combinations were much less effi-
cient. The extent to which these observations can be applied to cleavage in vivo, particularly
of long target RNAs by either chemically modified, exogenously applied ribozymes or
ribozymes transcribed in vivo which contain additional sequences, is not clear and at least
one report suggests they are not applicable.32
Hammerhead Ribozyme Structure

The structure of the hammerhead ribozyme has been studied both in solution32-35 and
in crystalline form.36,37 Two X-ray diffraction crystal structures show that the hammerhead
ribozyme has a y-shape in which helices II and III stack colinearly with helix I adjacent to
helix II (Fig. 1.2). General features of the crystal structures agree with the proposed solution
structures. In the colinear stacking of helices II and III, nucleotides in part of the catalytic
core, called domain II (consisting of G12, A13, A14 and U7, G8, A9) form non-Watson-
Crick base pairs and result in a pseudocontinuous helix. Domain I of the catalytic core (C3,
U4, G5, A6) forms a uridine turn motif, allowing the sugar phosphate backbone of helix II
(at U7) to turn and form the bottom of helix I (after C3).
In trying to account for the hammerhead ribozyme’s catalytic requirement for Mg2+, at
least two Mg2+ binding sites have been proposed.37 One of these is thought to be mainly
structural and involves a pentahydrated Mg2+, binding mainly to the 5' phosphate of A9
with additional contacts to G8, G10.1 and G12. Another site, in which Mg(H2O)62+ may
bind to C3 and C17 as well as other groups in the vicinity of the catalytic pocket, is closer to
the cleavage site (3' to C17) and is thought to be involved in the catalytic mechanism by
facilitating deprotonation of the 2'-hydroxyl of C17.37,38 Electrophoretic mobility studies39
indicate that there is a Mg2+ threshold above which helix I is adjacent to helix II (as in the
crystal and catalytically active form) and below which helix I is adjacent to helix III.
Reaction Mechanism and Kinetics

Magnesium ions, or certain other divalent metal ions, are essential for hammerhead
ribozyme catalysis. As suggested by the structural studies, the divalent metal ions are re-
quired firstly to promote correct folding of the catalytic core, and secondly as a reaction
cofactor.40 Zn2+ and Cd2+ can only perform this second, cofactor function and therefore
require spermine to help fold the RNA. By studying variation in reaction rates with pH and
with pKa values of hydrated metal ions, it was concluded that the metal ion acts as a base in
the reaction mechanism and that only one deprotonation event was involved.41 Although
some details of the reaction mechanism are still unclear (such as a second metal ion re-
quirement), the overall cleavage reaction is outlined in Figure 1.3. A hydrated Mg2+ acts as a
base to attack the essential 2'-OH group of C17, which then acts as a nucleophile attacking
the scissile phosphate. After cleavage, C17 has a 2',3' cyclic phosphate and N1.1 has a 5'-OH.
Fig. 1.2. Diagram of the hammerhead ribozyme based on the X-ray crystal structure.
For explanation of symbols see Fig. 1.1. Note how helix III stacks colinearly with helix
II. Reversed-Hoogsteen base pairs form between A9 and G12 and between G8 and A13
and non-Watson Crick base pairs are also present between U7 and A14 as well as be-
tween A15.1 and U16.1.
The hammerhead ribozyme catalytic reaction can be fitted to the Michalis-Menten

equation as long as:
1. the ribozyme concentration is much less than the substrate concentration;
2. the step R + S R.S is rapid and reversible; and
3. the rate determining step is R.S R.P (where R = ribozyme, S = substrate and P =
products).42-45
Under these multiple turnover conditions, the values derived from the Michalis-Menten
equation for Km (which is a measure of the ribozyme’s affinity for substrate) are usually
around 50-500 nM and for kcat (which is the rate of product production) are about 1 min–1,
which is much lower than for protein enzymes (≈500 min–1). These conditions are usually
only met in vitro, and for short ribozyme and target molecules, where alternative (probably
inactive) conformations for ribozyme and substrate are few or nonexistent.46,47 When these
conditions are not met, Km and kcat can often be determined under single turnover condi-
tions where the ribozyme concentration is made to exceed the substrate concentration.48
Fig. 1.3. Proposed reaction mechanism for phosphodiester bond cleavage by the
hammerhead ribozyme. A hydrated Mg2+ ion deprotonates the 2' OH of N1 and the
2' oxygen acts as a nucleophile to attack the scissile phosphate. A 2',3' cyclic phos-
phate is formed on N1.
Specificity
Ribozymes bind to target RNAs according to the rules of RNA duplex formation, at a
rate of around 5 x 108 M, the association being largely independent of length and sequence.49
The specificity of a ribozyme is its ability to cleave at one particular site and is usually a very
important consideration when one intends to cleave only one target RNA species in a com-
plex mixture, in which some RNAs could be similar to the target. Most studies on the speci-
ficity of hammerhead ribozymes have been performed in vitro with simple ribozyme and
substrate molecules and in these conditions specificity is determined by the rate of cleavage
compared to substrate dissociation.50 Higher specificity will be achieved with a slower cleavage
step or an increased rate of substrate dissociation.
Hertel et al examined the effects of shortening helices I and III, combined with intro-
ducing single mismatches in helix III, and concluded that the total target recognition length
(of helices I and III) could be 12 nucleotides without a reduction in specificity.51 These
experimental conditions differ from those expected in vivo, and hence it is likely that high
specificity can be retained in vivo with duplex regions longer than 12 bp, especially if the
number of G-C base pairs is not excessive. Mismatches close to the cleavage site can also be
used to give high specificity such as when targeting a mutant oncogene transcript which
may differ by only one nucleotide from the wild type mRNA.52
Modified Hammerhead Ribozymes

A considerable variety of modifications to the basic, all-RNA type I/III hammerhead
ribozyme have been carried out, mostly with the intention of improving the efficiency of
target cleavage by the ribozyme as a way of developing these molecules as therapeutic com-
pounds for exogenous delivery. Some modifications, such as the asymmetric hammerhead
ribozyme design, can also be applied when ribozymes are to be transcribed within cells
from gene constructs.
Chemical modifications of the hammerhead ribozyme that have been tried include
replacement of some of the 2'-OH moieties with allyl, amino, deoxy, fluoro or O-methyl
groups.53-55 In addition, the modification of parts of the backbone of the ribozyme from
phosphates to phosphorothioates has the dramatic effect of increasing the resistance of the
ribozyme to the action of cellular and serum nucleases.56 The key nucleases are pyrimidine-
specific endonucleases, which will attack C3, U4 and U7 of the unmodified core, as well as
3'-exonucleases. Most of these chemical modifications do not significantly increase the cata-
lytic activity; in fact reductions are more common. However, significant increases in stabil-
ity and/or useful pharmacological properties outweigh these drawbacks and make modi-
fied hammerhead ribozymes an exciting class of potential therapeutic reagents. The positions
U4 and U7 have been modified and stabilized to pyrimidine-specific endonucleases by both
2'-C-allyl and 2'-amino nucleotides without loss of catalytic activity,54,56 and there is one
report of improved catalytic activity by replacing U7 with pyrimidine-4-one.57 Some very
encouraging results have been obtained by direct injection of such chemically protected
ribozymes into mouse jaws58 and rabbit knee joints59 in which specific target mRNA levels
were transiently reduced or eliminated. Both chemically modified and unmodified ribozymes
have also been delivered to cells encapsulated in cationic liposomes60-62 and have resulted in
the desired phenotypic changes, such as restoration of drug sensitivity,63 cell proliferation60
and secretion of TNFα.64 Liposomes do have some drawbacks and other delivery systems
are being evaluated.65
Modifications to reduce helix II and its connecting loop have established a minimum
version of the hammerhead ribozyme called a minizyme which has no base pairs in helix II,
such that A9 is connected to G12 only by a loop that can be DNA or RNA.26,66 When helix II
still retains 1 bp the ribozyme is called a miniribozyme. These kinds of modifications can
reduce the costs of chemical synthesis, but generally yield ribozymes with decreased chemi-
cal cleavage rates as measured using short targets in vitro. However, against long target RNAs
in vivo, this may not be a problem, as cleavage is not likely to be rate limiting in this situa-
tion.66,67 The introduction of DNA, or DNA combined with some of the other modifica-
tions listed above, into helices I and III, but not helix II, can not only make the ribozyme
more nuclease resistant, but may also stimulate the overall reaction several fold. This may
work by enhancing either the cleavage rate or the turnover rate.26
Asymmetric hammerhead ribozymes are a modification of the basic hammerhead
ribozyme in which either helix I or helix III is significantly longer. Where the long helix is
greater than about 30 residues, these molecules are best produced by transcription. Shorter
types can be made by either chemical synthesis or transcription. In vitro experiments with
relatively short substrates have shown that asymmetric hammerhead ribozymes can cleave
their target when helix I is as short as 2 or 3 bp.51 Interestingly, by varying helix I and III
lengths, it has also been observed that, in vitro, longer helix III ribozyme constructs could
cleave one to two orders of magnitude more rapidly than longer helix I constructs with the
same total base pairing capacity.48 This would be an important design feature if generally
true and applicable in vivo.
An additional form of modification of the hammerhead ribozyme that may be gener-
ally beneficial in vivo has been reported by Sioud and Jespersen.68 A region of a TNFα
ribozyme that is bound by the enzyme GAPDH in cells can be linked to other ribozymes
and causes their catalytic activity to be increased in vitro and in vivo.
Pharmacokinetic and Cellular Uptake Studies

For chemically modified hammerhead ribozymes to be used as therapeutic compounds,
it is important to study their pharmacokinetic properties and the first comprehensive study
has recently appeared.69 A symmetric 38-mer ribozyme against a cytochrome P450 con-
taining 2'-O-allyl ribonucleotides at all positions outside the catalytic core and at positions
C3, U4 and U7 was administered as a single intravenous injection of 0.25 mg into adult
male rats. This ribozyme binds to serum albumin at binding site I, as do linear
phosphorothioate oligonucleotides. The plasma elimination half life, at 6.5 hours, was shorter
than that of these more stable phosphorothioates, but intact ribozyme was still detectable
after 48 hours. Metabolism is via endonuclease attack on the internal residues that are not
2'-O-allyl protected, and a 27-mer metabolite was found to accumulate in brain tissue, which
does not happen with phosphorothioates. The main organs of accumulation were kidney
and liver. Renal excretion of the ribozyme was minor compared to phosphorothioates, per-
haps due to high reabsorption in the proximal tubule. As no tissue toxicity was observed,
the 2'-O-allyl modified ribozyme seemed promising as a therapeutic.
The cellular uptake properties of a 2'-O-methyl modified hammerhead ribozyme con-
taining 2'-amino groups on residues U4 and U7 have recently been examined.70 This sym-
metric ribozyme against EGFR had 7 bp arms and was taken up by glioma cells in a tem-
perature, energy and pH dependent fashion which could be competed with a variety of
other oligonucleotides and polyanions. The ribozyme had a punctate, extranuclear local-
ization. Overall, ribozyme uptake seemed to be via a similar endocytic mechanism to that
for other oligodeoxynucleotides and also showed some cell-type specific differences.70,71
Making Hammerhead Ribozymes Work In Vivo

There are two distinct parts to this problem. The first involves selection of a suitable
and efficient site for cleavage in the target molecule, which is usually an mRNA molecule of
several hundred nucleotides in length. The second problem involves the optimal design of
the ribozyme. A significant part of this second problem concerns how the ribozyme will be
administered, i.e., will it be transcribed from a gene that is introduced into cells (endog-
enous expression), or will the ribozyme be synthesized and administered as a drug (exog-
enously delivered)? The engineering of hammerhead ribozymes for exogenous delivery has
recently been reviewed55 and the mechanics of delivery are considered in chapter 4 of this
text and hence will not be further discussed. As chapters 5-8 cover various aspects of ex-
pressing ribozymes endogenously, only general points regarding hammerhead ribozyme
design for endogenous expression will be made here.
Target Site Selection

Naturally occurring RNA molecules that may be chosen for cleavage by hammerhead
ribozymes are sufficiently long that they can fold into complex secondary and tertiary struc-
tures. This means that not all of the 5'-UH-3' sequences along the target RNA will be equally
accessible to ribozyme attack. Initial attempts to predict accessible 5'-UH-3' sites using com-
puter programs72-74 did demonstrate some successes, but this method does not guarantee
that any particular 5'-UH-3' sequence would be efficiently cleavable in vitro, let alone in
vivo. A more empirical method of selecting accessible target sites is to determine experi-
mentally which parts of a target RNA molecule are able to bind short antisense
oligodeoxynucleotides and thus be sites of nuclease (RNase H) sensitivity.60,75,76 Usually these
sites at which oligodeoxynucleotides can anneal are also accessible to ribozymes, at least in
vitro. Sites chosen in this way might well be cleavable in vivo, but may not be the best pos-
sible sites for in vivo cleavage.
A method which permits selection of efficient ribozyme cleavage sites from a large
number of possibilities in an in vivo system would clearly be superior. Lieber and Strauss77
have developed one such system in which a library of ribozymes, containing a constant
ribozyme core sequence but random hybridizing arm sequences, is mixed with target RNA
in cellular extracts. The various ribozymes will cleave the target RNA in different positions
and to different degrees, but the most efficient will produce cleavage products that are the
most abundant. The cleavage products are cloned by making cDNA and using PCR, and the
sequences of the cleavage sites determined by sequencing the clones. With this knowledge
of the cleavage site, efficient ribozymes can be designed that work well in vivo.78
Some other in vivo methods of selecting efficient ribozyme target sites are being devel-
oped, but as yet do not permit exhaustive screening of all possible sites. Luciferase has been
used as a reporter in which target sequences have been linked upstream of the luciferase
gene. Ribozymes introduced, or co-expressed, in cells expressing the construct should cause
a reduction in luciferase activity if they are active.32,79 An E. coli based system utilizing posi-
tive selection with trimethoprim for hammerhead ribozymes cleaving in cis80 and a yeast
system that requires cell cycle arrest81 may also be worth considering if they are further
developed.
Hammerhead Ribozyme Design for In Vivo Expression

For hammerhead ribozymes expressed in vivo, there seems little point in changing the
length of helix II and its terminating loop as, in vivo, extensions to helix II seem to inhibit
activity.82 Maintaining the standard catalytic core, except when creating an inactive ribozyme
control, is also prudent. Altering the length of the arms that anneal to the target RNA is an
important factor, but it is not yet clear what the best design is for in vivo activity. Conflicting
reports exist in the literature, some suggesting that short arms of 6-8 nucleotides are more
efficient77,82,83 and others that hybridizing arms need to be in excess of 25 nt.84,85 These dif-
ferences in efficiency are most likely due to differences in the accessibility of the various
target sites used, but differences in the sequences adjacent to the ribozyme, as well as the cell
system and cellular compartment in which cleavage was being tested, will also play a part. A
recent study looking at both subcellular localization and the length of helix III of asymmet-
ric hammerhead ribozymes concluded that shorter ribozymes were superior when injected
into the cytoplasm, whereas the helix III region needed to be greater than 51 nt to be effec-
tive when injected into the nucleus.86 Clearly, efforts should be made to express ribozymes
in a way that will ensure an appropriate subcellular localization for the target in question.
Conclusion
Since their discovery less than 15 years ago, hammerhead ribozymes have become the
most studied type of catalytic RNA. Armed with this accruing knowledge, which includes
the recent crystal structure and the wealth of information on mutagenesis and chemical
modification, researchers in the field of medicine have engineered hammerhead ribozymes
which are now poised to become essential therapeutic agents.87 They hold promise for ex-
ogenous administration in a variety of disorders, and will be invaluable tools for gene thera-
pists in their efforts to combat genetic disease and cancer, as is described in Section III.
References
1. Kruger K, Grabowski PJ, Zuang AJ et al. Self-splicing RNA: Autoexcision and autocyclization
of the ribosomal RNA intervening sequence of Tetrahymena. Cell 1982; 31:147-157.
2. Cech TR. Self-splicing of group I introns. Annu Rev Biochem 1990; 59:543-568.
3. Guerrier-Takada C, Gardiner K, Marsh T et al. The RNA moiety of ribonuclease P is the
catalytic subunit of the enzyme. Cell 1983; 35:849-857.
4. Altman S. Ribonuclease P. J Biol Chem 1990; 265:20053-20056.
5. Buzayan JM, Gerlach WL, Bruening G. Non-enzymic cleavage and ligation of RNAs comple-
mentary to a plant virus satellite RNA. Nature 1986; 323:349-353.
6. Forster AC, Symons RH. Self-cleavage of plus and minus RNAs of a virusoid and a struc-
tural model for the active sites. Cell 1987; 49:211-220.
7. Symons RH. Small catalytic RNAs. Annu Rev Biochem 1992; 61:641-671.
8. Hutchins CJ, Rathjen PD, Forster AC et al. Self-cleavage of plus and minus RNA tran-
scripts of avocado sunblotch viroid. Nucleic Acids Res 1986; 14:3627-3640.
9. Prody GA, Bakos JT, Buzayan JM et al. Autolytic processing of dimeric plant virus satellite
RNA. Science 1986; 231:1577-1580.
10. Haseloff J, Gerlach WL. Sequences required for self-catalyzed cleavage of the satellite RNA
of tobacco ringspot virus. Gene 1989; 82:43-52.
11. Hampel A, Tritz R. RNA catalytic properties of the minimum (-)sTRSV sequence. Bio-
chemistry 1989; 28:4929-4933.
12. Wu H-N, Lin Y-J, Lin F-P et al. Human hepatitis delta virus RNA subfragments contain
an autocleavage activity. Proc Natl Acad Sci USA 1989; 86:1831-1835.
13. Birikh KR, Heaton PA, Eckstein F. The structure, function and application of the ham-
merhead ribozyme. Eur J Biochem 1997; 245:1-16.
14. Hertel KJ, Pardi A, Uhlenbeck OC et al. Numbering system for the hammerhead ribozyme.
Nucleic Acids Res 1992; 20:3252.
15. Ruffner DE, Stormo GD, Uhlenbeck OC. Sequence requirements of the hammerhead RNA
self-cleavage reaction. Biochemistry 1990; 29:10695-10702.
16. Uhlenbeck OC. A small catalytic oligoribonucleotide. Nature 1987; 328:596-600.
17. Haseloff J, Gerlach WL. Simple RNA enzymes with new and highly specific endoribonuclease
activity. Nature 1988; 334:585-591.
18. Jeffries AC, Symons RH. A catalytic 13-mer ribozyme. Nucleic Acids Res 1989; 17:1371-1377.
19. Odai O, Hiroaki H, Tanaka T et al. Properties of a hammerhead-type RNA enzyme system
that consists of 3 RNA oligomer strands. Nucleosides and Nucleotides 1994; 13:1569-1579.
20. Clouet-D’Orval B, Uhlenbeck OC. Kinetic characterisation of I/II format hammerhead
ribozymes. RNA 1996; 2:483-491.
21. Tuschl T, Eckstein F. Hammerhead ribozymes: Importance of stem-loop II for activity.
Proc Natl Acad Sci USA 1993; 90: 6991-6994.
22. Nakamaye KL, Eckstein F. AUA-Cleaving hammerhead ribozymes: Attempted selection for
improved cleavage. Biochemistry 1994; 33:1271-1277.
23. Benseler F, Fu D-J, Ludwig J et al. Hammerhead-like molecules containing non-nucleoside
linkers are active RNA catalyzts. J Am Chem Soc 1993; 115:8483-8484.
24. Thomson JB, Tuschl T, Eckstein F. Activity of hammerhead ribozyme containing non-
nucleotidic linkers. Nucleic Acids Res 1993; 21:5600-5603.
25. Fu D-J, Benseler F, McLaughlin LW. Hammerhead ribozymes containing non-nucleoside

linkers are active RNA catalysts. J Am Chem Soc 1994; 116:4591-4598.
26. Hendry P, McCall MJ, Santiago FS et al. In vitro activity of minimised hammerhead
ribozymes. Nucleic Acids Res 1995; 23:3922-3927.
27. Keese P, Bruening G, Symons RH. Comparative sequence and structure of circular RNAs
from two isolates of lucerne transient streak virus. FEBS Lett 1983; 159:185-190.
28. Miller WA, Hercus T, Waterhouse PM et al. A satellite RNA of barley yellow dwarf virus
contains a novel hammerhead structure in the self-cleaving domain. Virology 1991;
183:711-720.
29. Perriman R, Delves A, Gerlach WL. Extended target-site specificity for a hammerhead
ribozyme. Gene 1992; 113:157-163.
30. Shimayama T, Nishikawa S, Taira K. Generality of the NUX rule—kinetic analysis of the
results of systematic mutations in the trinucleotide at the cleavage site of hammerhead
ribozymes. Biochemistry 1995; 34:3649-3654.
31. Zoumadakis M, Tabler M. Comparative analysis of cleavage rates after systematic permu-
tation of the NUX’ consensus target motif for hammerhead ribozymes. Nucleic Acids Res
1995; 23:1192-1196.
32. Kawasaki H, Ohkawa J, Tanishige N et al. Selection of the best target site for ribozyme-
mediated cleavage within a fusion gene for adenovirus E1A-associated 300 kDA protein
(p300) and luciferase. Nucleic Acids Res 1996; 24:3010-3016.
33. Amiri KMA, Hagerman PJ. Global conformation of a self-cleaving hammerhead RNA. Bio-
chemistry 1994; 33:13172-13177.
34. Tuschl T, Gohlke C, Jovin TM et al. A three-dimensional model for the hammerhead
ribozyme based on fluorescence measurements. Science 1994; 266:785-789.
35. Bassi GS, Mollegaard NE, Murchie AIH et al. Ionic interactions and the global conforma-
tions of the hammerhead ribozyme. Nat Struct Biol 1995; 2:45-55.
36. Pley HW, Flaherty KM, McKay DB. Three-dimensional structure of a hammerhead
ribozyme. Nature 1994; 372:68-74.
37. Scott WG, Finch JT, Klug A. The crystal structure of an all-RNA hammerhead ribozyme—
a proposed mechanism for RNA catalytic cleavage. Cell 1995; 81:991-1002.
38. Murray JB, Adams CJ, Arnold JRP et al. The roles of the conserved pyrimidine bases in
hammerhead ribozyme catalysis—evidence for a magnesium ion binding site. Biochem J
1995; 311:487-494.
39. Bassi GS, Murchie AIH, Lilley DMJ. The ion-induced folding of the hammerhead ribozyme:
Core sequence changes that perturb folding into the inactive conformation. RNA 1996;
2:756-768.
40. Dahm SC, Uhlenbeck OC. Role of divalent metal ions in the hammerhead ribozyme cleav-
age reaction. Biochemistry 1991; 30:9464-9469.
41. Dahm SC, Derrick WB, Uhlenbeck OC. Evidence for the role of solvated metal hydroxide
in the hammerhead cleavage mechanism. Biochemistry 1993; 32:13040-13045.
42. McConnell T. Theoretical considerations in measuring reaction parameters. In: Turner PC,
ed. Ribozyme Protocols. Totowa: Humana Press Inc., 1997:187-198.
43. McConnell T. Experimental approaches for measuring reaction parameters. In: Turner PC,
ed. Ribozyme Protocols. Totowa: Humana Press Inc., 1997:199-208.
44. DeYoung MB, Siwkowski A, Hampel A. Determination of catalytic parameters for hairpin
ribozymes. In: Turner PC, ed. Ribozyme Protocols. Totowa: Humana Press Inc.,
1997:209-220.
45. Hendry P, McCall MJ, Lockett TJ. Characterising ribozyme cleavage reactions. In: Turner
PC, ed. Ribozyme Protocols. Totowa: Humana Press Inc., 1997:221-229.
46. Fedor MJ, Uhlenbeck OC. Kinetics of intermolecular cleavage by hammerhead ribozymes.
Biochemistry 1992; 31:12042-12054.
47. Hertel KJ, Herschlag D, Uhlenbeck OC. A kinetic and thermodynamic framework for the
hammerhead ribozyme reaction. Biochemistry 1994; 33:3374-3385.
48. Hendry P, McCall MJ. Unexpected anisotropy in substrate cleavage rates by asymmetric
hammerhead ribozymes. Nucleic Acids Res 1996; 24:2679-2684.
49. Nelson JW, Tinoco I Jr. Comparison of the kinetics of ribo-oligonucleotide, deoxyribo-
oligonucleotide and hybrid oligonucleotide double-strand formation by temperature-jump
kinetics. Biochemistry 1982; 21:5289-5295.
50. Herschlag D. Implications of ribozyme kinetics for targeting the cleavage of specific RNA
molecules in vivo: More isn’t always better. Proc Natl Acad Sci USA 1991; 88:6921-6925.
51. Hertel KJ, Herschlag D, Uhlenbeck OC. Specificity of hammerhead ribozyme cleavage.
EMBO J 1996; 15:3751-3757.
52. Funato T, Shitara T, Tone T et al. Suppression of H-ras-mediated transformation in NIH3T3
cells by a ras ribozyme. Biochem Pharmacol 1994; 48:1471-1475.
53. Pieken WA, Olsen DB, Benseler F et al. Kinetic characterisation of ribonuclease-resistant
2'-modified hammerhead ribozymes. Science 1991; 253:314-317.
54. Heidenreich O, Benseler F, Fahrenholz A et al. High activity and stability of hammerhead
ribozymes containing 2'-modified pyrimidine nucleosides and phosphorothioates. J Biol
Chem 1994; 269:2131-2138.
55. Usman N, Beigelman L, McSwiggen JA. Hammerhead ribozyme engineering. Curr Opin
Struct Biol 1996; 1:527-533.
56. Beigelman L, McSwiggen JA, Draper KG et al. Chemical modification of hammerhead
ribozymes: Catalytic activity and nuclease resistance. J Biol Chem 1995; 270:25702-25708.
57. Burgin AB, Gonzalez C, Matulic-Adamic J et al. Chemically modified hammerhead
ribozymes with improved catalytic rates. Biochemistry 1996; 35:14090-14097.
58. Lyngstdaas SP, Risnes S, Sproat BS et al. A synthetic, chemically modified ribozyme elimi-
nates amelogenin, the major translation product in developing mouse enamel in vivo. EMBO
J 1995; 14:5224-5229.
59. Flory CM, Pavco PA, Jarvis TC et al. Nuclease-resistant ribozymes decrease stromelysin
mRNA levels in rabbit synovium following exogenous delivery to the knee joint. Proc Natl
Acad Sci USA 1996; 93:754-758.
60. Jarvis TC, Alby JA, Beaudry AA et al. Inhibition of vascular smooth muscle cell prolifera-
tion by ribozymes that cleave c-myb mRNA. RNA 1996; 2:419-428.
61. Castanotto D, Bertrand E, Rossi JJ. Exogenous cellular delivery of ribozymes and ribozyme
encoding DNAs. In: Turner PC, ed. Ribozyme Protocols. Totowa: Humana Press Inc.,
1997:429-439.
62. Brown SA, Jarvis TC. Optimization of lipid-mediated ribozyme delivery to cells in culture.
In: Turner PC, ed. Ribozyme Protocols. Totowa: Humana Press Inc., 1997:441-449.
63. Kiehntopf M, Brach MA, Licht T et al. Ribozyme-mediated cleavage of the MDR-1 tran-
script restores chemosensitivity in previously resistant cancer cells. EMBO J 1994;
13:4645-4652.
64. Kisich KO, Stecha PF, Harter HA et al. Inhibition of TNF-alpha secretion by murine mac-
rophages following in vivo and in vitro ribozyme treatment. J Cell Biochem 1995; 19A:221.
65. Leopold HL, Shore SK, Newkirk TA et al. Multiunit ribozyme-mediated cleavage of bcr-
abl mRNA in myeloid leukemias. Blood 1995; 85:2162-2170.
66. McCall MJ, Hendry P, Lockett TJ. Minimized hammerhead ribozymes. In: Turner PC, ed.
Ribozyme Protocols. Totowa: Humana Press Inc., 1997:151-159.
67. Sioud M. Effects of variations in length of hammerhead ribozyme antisense arms upon the
cleavage of longer RNA substrates. Nucleic Acids Res 1997; 25:333-338.
68. Sioud M, Jespersen L. Enhancement of hammerhead ribozyme catalysis by glyceraldehyde-
3-phosphate dehydrogenase. J Mol Biol 1996; 257:775-789.
69. Desjardins JP, Sproat BS, Beijer B et al. Pharmacokinetics of a synthetic, chemically-modi-
fied hammerhead ribozyme against the rat cytochrome-P-450 3A2 messenger-RNA after
single intravenous injections. J Pharmacol Exp Ther 1996; 278:1419-1427.
70. Fell PL, Hudson AJ, Reynolds MA. Cellular uptake properties of a 2'-amino/2'-O-methyl-
modified chimeric hammerhead ribozyme targeted to the epidermal growth factor recep-
tor mRNA. Antisense and Nucleic Acid Drug Development 1997; 7:319-326.
71. Hawley P, Gibson I. Interaction of oligodeoxynucleotides with mammalian cells. Antisense
and Nucleic Acid Drug Development 1996; 6:185-195.
72. Christoffersen RE, McSwiggen J, Konings D. Application of computational technologies to

ribozyme biotechnology products. J Mol Struct 1994; 311:273-284.
73. Sczakiel G, Tabler M. Computer-aided calculation of the local folding potential of target
RNA and its use for ribozyme design. In: Turner PC, ed. Ribozyme Protocols. Totowa:
Humana Press Inc., 1997:11-15.
74. James W, Cowe E. Computational approaches to the identification of ribozyme target sites.
75. Frank BL, Goodchild J. Selection of accessible sites for ribozymes on large RNA transcripts.
76. Birikh KR, Berlin YA, Soreq H et al. Probing accessible sites for ribozymes on human
acetylcholinesterase RNA. 1997; 3:429-437.
77. Lieber A, Strauss M. Selection of efficient cleavage sites in target RNAs by using a ribozyme
expression library. Mol Cell Biol 1995; 15:540-551.
78. Lieber A, Kay MA. Adenovirus mediated expression of ribozymes in mice. J Virol 1996;
70:3153-3158.
79. Scherr M, Grez R, Ganser A. Specific hammerhead ribozyme-mediated cleavage of mutant
N-ras mRNA in vitro and ex vivo—oligoribonucleotides as therapeutic agents. J Biol Chem
1997; 272:14304-14313.
80. Fujita S, Koguma T, Ohkawa J. Discrimination of a single base change in a ribozyme using
the gene for dihydrofolate reductase as a selective marker in Escherichia coli. Proc Natl
Acad Sci USA 1997; 94:391-396.
81. Ferbeyre G, Bratty J, Chen H et al. Cell-cycle arrest promotes trans-hammerhead ribozyme
action in yeast. J Biol Chem 1996; 271:19318-19323.
82. Homann M, Tabler M, Tzortzakaki S et al. Extension of helix II of an HIV-1-directed
hammerhead ribozyme with long antisense flanks does not alter kinetic parameters in vitro
but causes loss of inhibitory potential in living cells. Nucleic Acids Res 1994; 22:3951-3957.
83. Bertrand EL, Rossi JJ. Facilitation of hammerhead ribozyme catalysis by the nucleocapsid
protein of HIV-1 and the heterogeneous nuclear ribonucleoprotein A1. EMBO J 1994:
13:2904-2912.
84. Crissell P, Thompson S, James W. Inhibition of HIV-1 replication by ribozymes that show
poor catalytic activity in vitro. Nucleic Acids Res 1993; 21:5251-5255.
85. Beck J, Nassal M. Efficient hammerhead ribozyme-mediated cleavage of the structured
hepatitis B virus encapsidation signal in vitro and in cell extracts, but not in intact cells.
Nucleic Acids Res 1995; 23:4954-4962.
86. Hormes R, Homann M, Oelze I et al. The subcellular localization and length of hammer-
head ribozymes determine efficacy in human cells. Nucleic Acids Res 1997; 25:769-775.
87. James HA, Gibson I. The therapeutic potential of ribozymes. Blood 1998; 91:371-382.
CHAPTER 2
Biochemistry of the Hairpin Ribozyme

Andrew Siwkowski and Arnold Hampel
Introduction
T he observation that the 359 nt negative strand of the satellite RNA of tobacco ringspot
virus [(-)sTRSV] was autocatalytic1 led to identification of the minimal catalytic center
consisting of a 50 nt enzyme-like RNA and a 14 nt substrate.2 This structure was named the
hairpin ribozyme.3 The ribozyme/substrate consisted of 4 helices, helices 1, 2, 3, 4, and five
loops, loops 1, 2, 3, 4, and 5 (Fig. 2.1). Helix 1, between the ribozyme and substrate, can vary
in length and have a variable sequence as long as base pairing is maintained. Helix 2, also
between the ribozyme and substrate is fixed at 4 bp; however it also can vary in sequence as
long as base pairing is maintained. Helix 3 is a 4 bp helix found in the ribozyme separated
from helix 2 by a single unpaired A15, which serves as a hinge.4-6 Helix 4, in the native
sequence, contains three Watson-Crick base pairs and one non-canonical A:G base pair.7
Thus a total of 18 bp exist in the two-dimensional structure of the hairpin ribozyme.
The helices are separated by five single stranded loop regions. Loop 5 is dispensable
and can be replaced by other structures as long as a strong helix 4 is maintained. Loops 1, 2,
4, and 5, however, contain required bases. Thus the hairpin ribozyme consists of essentially
two domains—domain I (helices 1 and 2; loops 1 and 5) and domain II (helices 3 and 4 and
loops 2 and 4).8
Cleavage takes place in the substrate at loop 5 by breakage of the phosphodiester bond
at ApG to produce a 5' cleavage fragment with a 2',3'-cyclic phosphate terminus and 3'
cleavage fragment with a 5'-OH terminus. The hairpin ribozyme can also produce a ligated
product.9 The ligation reaction uses the same termini as produced with cleavage, and thus
appears to be a simple reversal of the forward (cleavage) reaction.10
Biochemistry and Mechanism of the Reaction

The (-)sTRSV ribozyme supports multiple cleavage events, has a temperature opti-
mum of 37°C, and an energy of activation of 19 kcal/mol.2 Recent reports of kinetic param-
eters for the cleavage event, determined by several groups,11-14 fall within a range of 19-96 nM
for Km and 0.12-0.36/min for kcat. Substrate dissociation rates are so much slower than cleav-
age rates that virtually every substrate that binds is cleaved, and the rate of ligation was
found to be 10X faster than cleavage—indicating that the hairpin ribozyme is truly a unique
catalytic system among known catalytic RNAs.12
The hairpin ribozyme catalyzes cleavage of substrate containing an Rp phosphorothioate
substitution at ApG of the cleavage site.15 The cleavage reaction proceeds with inversion of
configuration of the phosphorus in the product with respect to that of the substrate, sug-
gesting an in-line attack mechanism.16 A low sulfur effect was associated with catalysis at an
Fig. 2.1. The hairpin ribozyme/substrate complex. The hairpin ribozyme complexed
with its substrate forms a structure with four helical regions interspersed by five loop
regions. These exist as two domains, domain I and domain II.
Rp phosphorothioate substitution at the cleavage site.17,18 Efficient cleavage of the Sp

phosphorothioate isomer in the presence of Mg2+ suggests that it too is not directly coordi-
nated with a metal cofactor during cleavage;18 however, outer sphere coordination with ei-
ther Rp or Sp phosphate oxygen is still possible.
The classic model of RNA catalysis is supplied by bovine pancreatic RNase. The nu-
merous similarities between RNA catalysis by this RNase and the hairpin ribozyme suggest
a common basic mechanism. This protein cleaves RNA in a sequence specific manner using
an in-line attack mechanism.19 The 2'-OH is deprotonated by His-12, and then is available
for nucleophilic attack on the phosphorus. The resulting trigonal bipyramid contains both
attacking and leaving groups at axial positions. The cleavage results in products with 5'-OH
(leaving group) and 2',3'-cyclic phosphate termini. In the case of RNase, the trigonal
bipyramidal intermediate is thought to be stabilized by interactions between Lys-41 and a
non-bridging phosphate oxygen.20 This particular role may be filled in the hairpin ribozyme
by a hexahydrated Mg2+ or Co(NH3)63+ complex.18 Hexahydrated Mg2+ has been shown to
coordinate to phosphate oxygens in yeast tRNAPhe,21 and cobalt hexaammine has been shown
to coordinate to a non-bridging phosphate oxygen in the stabilization of a Z-DNA struc-
ture.22 The fact that the 2'-OH of A10 is directly across from the cleavage site and that it is
Biochemistry of the Hairpin Ribozyme 17
Fig. 2.2. Mechanism of catalysis by

the hairpin ribozyme. Direct in-
volvement of metals in the catalytic
step is not the case. Rather, func-
tional groups on the ribozyme/sub-
strate complex itself are likely to
participate in the catalytic step.
Table 2.1. Unimolecular rate

constants and pKa values for
cleavage of the hairpin
ribozmyme
Metal k (min–1) pKa
Mg2+ 1.8 ± 0.2 11.4

Ca2+ 3.6 ± 0.7 12.8
Sr2+ 2.2 ± 0.4 13.3
Ba2+ 2.2 ± 0.7 13.5
critical to catalysis, is particularly significant in light of this possible role for the metal.
Given the fact that As5 can accommodate any base identity and retain substantial catalytic
activity,23 it seems likely that this base is oriented outside of the base stacking system of
helix 2. The cleavage site could therefore be positioned closer to nucleotide A10, facilitating
the formation of a binding pocket for a hydrated metal. The pocket would consist of outer
sphere coordination sites, including the 2'-OH of A10, the cleavage site non-bridging phos-
phate oxygen, and possibly N3 of A10. Additional coordination sites are likely to be supplied
by functional groups in loops 2 and 4 in the folded ribozyme-substrate structure (Fig. 2.2).
Since deprotonation of the 2'-OH in the case of RNase A is carried out by a histidine,
and obviously no such histidine exists in the ribozyme, the question arises: What deprotonates
the 2'-OH in the hairpin ribozyme? In the hammerhead, it has been suggested that
deprotonation is mediated by a partially hydrated metal cofactor in the reaction. This same
hypothesis is not supported in the hairpin ribozyme, since a variety of metals support high
cleavage rates regardless of the differing pKa values of their hydrated complexes (Table 2.1).
The answer to this question is even more difficult to ascertain given the fact that the pH
optimum of the reaction has not been reached. Consequently, no significant pKa has been
determined. The deprotonation may occur by action of an RNA functional group or a sol-
vent molecule whose pKa has been perturbed by the molecular environment created by
folding of the hairpin ribozyme.
It appears that at least two metal binding sites are used for cleavage by the hairpin
ribozyme in its active structure. Evidence suggesting this is as follows:
1. two cofactors support cleavage when either one alone could not;24 and
2. the sigmoidal shape of the curve showing cleavage reaction rate as a function of
Mg2+ concentration, indicative of multiple binding sites (see ref. 25 and Siwkowski,
unpublished data).
To date, however, it has not been definitively determined exactly how many binding
sites exist.
Generally, metals have been proposed to serve two different roles in ribozyme-medi-
ated catalysis; structural and catalytic.26,27 The role the metal plays in catalysis by the hairpin
ribozyme has been suggested to be structural rather than catalytic. The finding that
hexaammine cobalt chloride supports cleavage by the hairpin ribozyme in the absence of
other metals suggests that the metal normally carries out its role as a fully hydrated com-
plex, thereby working through outer sphere interactions with its coordinated waters rather
than through an inner sphere mechanism.18 Outer sphere coordination between Mg2+ and
a phosphodiester is strongly favored over inner sphere complex formation from a thermo-
dynamic standpoint.28 The Co(NH3)63+ complex stabilizes an RNA helix junction structure
and, albeit to a lesser degree than Mg2+, tertiary structure.29 Given the high pKa of the coor-
dinated amine groups with hexaammine cobalt chloride, as well as the slow ligand exchange
rate, it seems unlikely that the complex can promote the deprotonation of the 2'-OH, which
is likely to be the first step of the cleavage reaction pathway. These findings point to the
metal serving a structural role in the formation of the transition state structure; however, a
possible catalytic role, wherein the metal coordinates to a phosphate oxygen to prepare the
phosphorus for nucleophilic attack, cannot be entirely excluded.
Using a cis-cleaving ribozyme (Fig. 2.3b), kinetic cleavage rate constants supported by
Mg2+, Ca2+, Sr2+, and Ba2+ were all very similar (Table 2.1). These results show rate of cleav-
age is independent of the ionic radius or coordination number of the metal cofactor. Fur-
thermore, the cleavage rate constant is not dependent on the pKa of the hydrated metal.
While Mn2+ supports cleavage, Co2+ does not, nor does Li+, Na+, K+, or Cs+. This pattern is
very similar to that reported for a class of proposed structural sites in the Tetrahymena
ribozyme,27 further supporting the concept that the metal serves to stabilize structure in the
hairpin ribozyme.
The importance of inter-domain interactions for catalysis is particularly clear when
reviewing the evidence obtained from several groups—all showing that catalytic rate is de-
pendent on the distance between the 5' end of the substrate and the 3' end of the ribozyme.
A series of linkers, each consisting of a different number of bases, when joining the 3' end of
the ribozyme to the 5' end of the substrate (Fig. 2.3c), corresponded directly to increased
levels of circularization (i.e., ligation) with increasing linker length.4 When 1,3-propanediol
phosphate units for nucleotide residues were used as linker units, similar results were ob-
tained.6 When the method was revised to retain the linker between the 3' end of the ribozyme
and the 5' end of the substrate and the removal of the bond between ribozyme positions
A15 and C16, along with the addition of a single additional base 5' to C16 (Fig. 2.3d), the
results were consistent wherein catalysis was dependent on linker length.14
A hairpin ribozyme was constructed in which the two domains were attached in the
opposite manner as that found in the native structure, where the two domains are joined by
formation of a new helix containing variable-length linkers between the 5' end of the
ribozyme and the 3' end of the cytidine immediately preceding loop 3 (Fig. 2.3e).8 In this
particular construct, the substrate region and ribozyme region containing loop 2 are sepa-
rate RNAs, so that the final cleavage reaction is trimolecular. The same pattern associating
increasing linker length with increasing cleavage rate was observed. The domains can be
totally separated (Fig. 2.3f) and, when the reaction has high concentrations of the RNA
comprising the domains, cleavage rates were obtained which were similar to those of the
standard bimolecular reaction between hairpin ribozyme and substrate (Fig. 2.3a).30
Using a slightly different construct (Fig. 2.3g), this ability to obtain cleavage activity
when physically separated domains were combined was again demonstrated.31 When link-
ers of varying lengths were inserted between A14 and A15, a different trend from the previ-
ous studies was observed. With the exception of an increase in activity accompanying the
insertion of a single nucleotide, the cleavage levels associated with remaining linker lengths
followed the basic trend of increasing linker length causing lower cleavage activity. This
result suggested that close restraint of one domain to the other facilitated cleavage rather
than decreasing it as was seen in the previous studies.
Fig. 2.3. Forms of the hairpin ribozyme used for structure/function studies. (a) conven-
tional form for bimolecular trans reactions;5 (b) cis-cleaving form with 3' end of sub-
strate linked to 5' end of ribozyme;7 (c) 5' end of substrate linked to 3' end of ribozyme;6
(d) 5' end of substrate linked to 3' end of ribozyme with break at A15;14 (e) construct of
Komatsu et al, 1995;8 (f) domain I and domain II are separate;30,31 (g) with a linker
placed at A15;31 (h) Tripartite construct used for functional group studies.13,25,33
Complexes between 3' cleavage products and ribozymes are much stronger than pre-
dicted from simple helix association,12 suggesting that the remainder of the molecule is
contributing to stabilization of the binding—perhaps by folding over at the hinge region.32
Such a folding at A15 would allow helices 2 and 4 to interact and perhaps contribute to the
stabilization. The exact nature of these interactions is unknown; however, it is reasonable to
expect that critical tertiary interactions could also occur between any or all of the required
loop regions. It is very likely that components of loops 1, 2, 4, and 5 interact in some way.
The interactive groups could be any of the six required positions in these loops.7,23
The study of specific functional group requirements has done much to explain the
catalytic mechanism, as well as structure. When the role of 2'-OH groups was analyzed,
there were four positions, A10, G11, A24, and C25, where substitution with either a 2'-H or
2'-O-methyl resulted in drastic reductions in cleavage rate.33 Phosphorothioate substitu-
tions on the hairpin ribozyme revealed that, while there are three positions which accom-
pany a modest reduction in cleavage (5' to A7, A9, and A10), no phosphorothioate substitu-
tion within the ribozyme appears to prevent cleavage.34
Another form of functional group substitution study, base substitution, determined
the secondary structure of the hairpin ribozyme. This method has been employed using
two different schemes, each with success. These are mutational analysis and in vitro selec-
tion. Mutation analysis, wherein base substitutions are made precisely, has the advantage of
being able to directly interrogate the molecule regarding the importance of a specific base
identity. The second scheme, in vitro selection, has the potential advantage of allowing one
to rapidly scan for more globally significant interactions, but its interpretation is often dif-
ficult. It has greater potential utility for co-variation of multiple sites wherein each site is
randomly mutagenized and then selected by an in vitro method. However, randomization
of approximately 15 bases in a given experiment is the upper limit, since it is necessary to
produce a reasonable number of each sequence variant in a reaction. Additionally, with
increasing numbers of bases randomized, larger numbers of selected sequences have to be
analyzed, particularly when selection conditions are stringent. This is true because, under
very strong selection pressures, only the most active sequences will be retained to a high
degree. Both of these schemes have been used to locate the base pairs in the helical regions
of the ribozyme-substrate complex. The seventeen Watson-Crick base pairs in the structure
were all originally identified by mutagenesis3,5 while the 18th base pair found, the A:G pair
in helix 4, was originally identified by in vitro selection and then confirmed by mutagenesis.7,35
Both direct mutagenesis and in vitro selection have been successful in identifying key
bases in the loop regions.7,23,35 However, no covariations have been identified to date between
any of the four required loops. Mutational analysis data show that there are six base positions
in the ribozyme-substrate complex which, when mutated to any of the non-native variants,
result in cleavage rates below the lower limit of detection. These are ribozyme base posi-
tions G8, A22, A23, C25, and A38, as well as substrate base position Gs6.7,23 The remaining
base positions show varying sensitivities to change with respect to their supporting cleavage.
The five required ribozyme bases are involved either in structure or in the catalytic
event itself—exactly which is at present unknown. The required substrate base, Gs6, the
base immediately 3' of the cleavage site, has an exocyclic amino group which is absolutely
required for cleavage and may have particular significance to the catalytic structure at the
cleavage site.36
With the advent of the more exotic RNA phosphoramidites, the determination of re-
quirements for specific functional groups other than merely base or 2'-OH moieties was
possible. Two studies that exemplify this manner of investigation examined functional groups
in the loop regions of the hairpin ribozyme.13,25 These studies have been helpful in suggest-
ing important sites within nucleotide residues of the RNA where key interactions may oc-
cur. The use of propyl linkers as well as abasic substitutions at specific positions within the
molecule have also been used to demonstrate the relative unimportance of several positions
as well. When combined with mutational analysis data, these studies have gone far in sug-
gesting sites used in secondary and possible tertiary interactions.
Conclusions
The hairpin ribozyme, while in the class of ribozymes which carry out cleavage to yield
2',3'-cyclic phosphate and 5'-OH termini, has a structure and mechanism unique among all
known ribozymes. It alone is capable of facile ligation and it has excellent catalytic param-
eters. Substrate recognition occurs by formation of two helices, helix 1 and helix 2. Helix 1
is of variable length and helix 2 is fixed at 4 bp. The scissile phosphate is at the ApG in the
substrate where the G is required and is part of a preferred BN*GUC sequence.5 The cata-
lytic reaction itself likely occurs without the direct involvement of a multivalent cation—
again a unique aspect of the hairpin ribozyme.
References
1. Gerlach WL, Buzayan JM, Schneider IR et al. Satellite tobacco ringspot virus RNA: Bio-
logical activity of DNA clones and their in vitro transcripts. Virology 1986; 151:172-185.
2. Hampel A, Tritz R. RNA catalytic properties of the minimum (-)sTRSV sequence. Bio-
chemistry 1989; 28:4929-4933.
3. Hampel A, Tritz R, Hicks M et al. ‘Hairpin’ catalytic RNA model: Evidence for helices and
sequence requirement for substrate RNA. Nucleic Acids Res 1990; 18:299-304.
4. Feldstein P, Bruening G. Catalytically active geometry in the reversible circularization of
‘mini-monomer’ RNAs derived from the complementary strand of tobacco ringspot virus
satellite RNA. Nucleic Acids Res 1993; 21:1991-1998.
5. Anderson P, Monforte J, Tritz R. Mutagenesis of the hairpin ribozyme. Nucleic Acids Res
1994; 22:1096-1100.
6. Komatsu Y, Koizumi M, Nakamura H et al. Loop-size variation to probe a bent structure
of a hairpin ribozyme. J Am Chem Soc 1994; 116:3692-3696.
7. Siwkowski A, Shippy R, Hampel A. Analysis of hairpin ribozyme base mutations in loops
2 and 4 and their effects on cis-cleavage in vitro. Biochemistry 1997; 36:3930-3940.
8. Komatsu Y, Kanzaki I, Ohtsuka E. Enhanced folding of hairpin ribozymes with replaced
domains. Biochemistry 1996; 35:9815-9820.
9. Buzayan JM, Gerlach WL, Bruening G. Non-enzymatic cleavage and ligation of RNAs
complementary to a plant virus satellite RNA. Nature 1986; 323:349-353.
10. Buzayan JM, Hampel A, Bruening G. Nucleotide sequence and newly formed phosphodiester
bond of spontaneously ligated satellite tobacco ringspot virus RNA. Nucleic Acids Res 1986;
14:9729-9743.
11. DeYoung MB, Siwkowski AM, Lian Y et al. Catalytic properties of hairpin ribozymes de-
rived from chicory yellow mottle virus and arabis mosaic virus satellite RNAs. Biochemis-
try 1995; 34:15785-15791.
12. Hegg LA, Fedor MJ. Kinetics and thermodynamics of intermolecular catalysis by hairpin
13. Schmidt S, Beigelman L, Karpeisky A et al. Base and sugar requirements for RNA cleavage
of essential nucleoside residues in internal loop B of the hairpin ribozyme: Implications
for secondary structure. Nucleic Acids Res 1996; 24:573-581.
14. Komatsu Y, Kanzaki I, Koizumi M et al. Modification of primary structures of hairpin
ribozymes for probing active conformations. J Mol Biol 1995; 252:296-304.
15. Buzayan JM, Feldstein PA, Bruening G et al. RNA mediated formation of a phosphoro-
thioate diester bond. Biochem Biophys Res Commun 1988; 156:340-347.
16. van Tol H, Buzayan JM, Feldstein PA et al. Two autolytic processing reactions of a satellite
RNA proceed with inversion of configuration. Nucleic Acids Res 1990; 18:1971-1975.
17. Chowrira BM, Burke JM. Binding and cleavage of nucleic acids by the ‘hairpin’ ribozyme.
Biochemistry 1991; 30:8518-8522.
18. Hampel A, Cowan JA. A unique mechanism for RNA catalysis: The role of metal cofactors
in hairpin ribozyme cleavage. Chemistry and Biology 1997; 4:513-517.
19. Usher DA, Erenrich ES, Eckstein F. Geometry of the first step in the action of ribonuclease
A. Proc Nat Acad Sci 1972; 69:115-118.
20. Fersht A. The structures and mechanisms of selected enzymes. In: Enzyme Structure and
Mechanism. 2nd ed. New York: W.H. Freeman and Co., 1984:389-452.
21. Saenger W. Principles of Nucleic Acid Structure. Cantor CR, ed. New York: Springer-Verlag,
1984:331-349.
22. Gessner RV, Quigley GJ, Wang AH-J et al. Structural basis for stabilization of Z-DNA by
cobalt hexaammine and magnesium cations. Biochemistry 1985; 24:237-240.
23. Shippy R, Siwkowski A, Hampel A. Mutational analysis of loops 1 and 5 of the hairpin
ribozyme. Biochemistry 1998; 37:564-578.
24. Chowrira BM, Berzal-Herranz A, Burke JM. Ionic requirements for RNA binding, cleav-
age, and ligation by the hairpin ribozyme. Biochemistry 1993a; 32:1088-1095.
25. Grasby JA, Mersmann K, Singh M et al. Purine functional groups in essential residues of
the hairpin ribozyme required for catalytic cleavage of RNA. Biochemistry 1995;
34:4068-4076.
26. Guerrier-Takada C, Haydock K, Allen L et al. Metal ion requirements and other aspects of
the reaction catalyzed by M1 RNA, the RNA subunit of ribonuclease P from Escherichia
coli. Biochemistry 1986; 25:1509-1515.
27. Grosshans CA, Cech TR. Metal ion requirements for sequence-specific endoribonuclease
activity of the tetrahymena ribozyme. Biochemistry 1989; 28:6888-6894.
28. Cowan JA. Biological chemistry of magnesium ion with physiological metabolites, nucleic
acids, and drug molecules. In: Cowan JA, ed. The Biological Chemistry of Magnesium.
New York: VCH Publishers, Inc., 1995:185-209.
29. Laing LG, Gluick TC, Draper DE. Stabilization of RNA structure by Mg ions: Specific and
non-specific effects. J Mol Biol 1994; 237:577-587.
30. Butcher SE, Heckman JE, Burke JM. Reconstitution of hairpin ribozyme activity following
separation of functional domains. J Biol Chem 1995; 270:29648-29651.
31. Shin C, Choi JN, Song SI et al. The loop B domain is physically separable from the loop A
domain in the hairpin ribozyme. Nucleic Acids Res 1996; 24:2685-2689.
32. Walter NG, Burke JM. Real-time monitoring of hairpin ribozyme kinetics through base-
specific quenching of fluorescein-labeled substrates. RNA 1997; 3:392-404.
33. Chowrira BM, Berzal-Heranz A, Keller CF et al. Four ribose 2'-hydroxyl groups essential
for catalytic function of the hairpin ribozyme. J Biol Chem 1993b; 268:19458-19462.
34. Chowrira BM, Burke JM. Extensive phosphorothioate substitution yields highly active and
nuclease-resistant hairpin ribozymes. Nucleic Acids Res 1992; 20:2835-2840.
35. Siwkowski A, Humphrey M, DeYoung MB, Hampel A. Screening for important base iden-
tities in the hairpin ribozyme by in vitro selection for cleavage. BioTechniques 1998;
24:278-284.
36. Chowrira BM, Berzal-Herranz A, Burke JM. Novel guanosine requirement for catalysis by
the hairpin ribozyme. Nature 1991; 354:320-322.
CHAPTER 3
Biochemistry of Hepatitis Delta Virus

Catalytic RNAs
N. Kyle Tanner
Introduction
H epatitis delta virus (HDV) is a unique human pathogen with a world-wide distribution.
It is associated with a high incidence of fulminant hepatitis and premature death, al-
though the severity of the disease varies widely depending on the geographic location (re-
viewed in refs. 1-5). HDV is a satellite virus of hepatitis B (HBV), but HDV is completely
unrelated to its helper virus. It requires the coat proteins encoded by HBV for encapsulation
and formation of infectious particles, but infection and replication occur independently.
HDV is a unique mammalian virus, but it does share a number of features with certain
pathogens found in plants: the viroids and viroid-like satellite viruses (reviewed in refs. 6,
7). These features include a single-stranded RNA genome, an RNA-dependent rolling-circle
mode of replication, and the ability of the isolated RNA to self-cleave in vitro. The self-
cleaving reaction is probably used in vivo to process the intermediates generated during
rolling-circle replication into unit-length progeny (Fig. 3.1). However, unlike the hammer-
head and hairpin (paper clip) motifs found in the plant viruses, the sequences constituting
the HDV self-cleaving domains are unique, and they form an entirely different catalytic
motif. This review will summarize the characteristics of the HDV self-cleaving domains,
show how they can be converted into trans-cleaving ribozymes, and discuss how the prop-
erties of the catalytic domains can be exploited for developing therapeutic or prophylactic
agents.
Properties of HDV
To better understand the catalytic domains, it is first useful to briefly discuss some
general properties of the virus (reviewed in refs. 1-5). A cartoon of the viral life cycle is
shown in Figure 3.1. HDV consists of a circular, single-stranded RNA strand of about 1700
nucleotides (nt). The RNA has a high content of guanines and cytidines (60%), and it is
capable of making extensive intramolecular base pairs (~70%) to form a long rod-shaped
structure that is visible by electron microscopy (Fig. 3.2).8,9 The infectious genomic (-) strand
is found in large excess over the complementary antigenomic (+) strand, which is a replica-
tive intermediate. Small quantities of linear dimer and trimer molecules are also found,
which is consistent with a rolling-circle mechanism of replication. No DNA intermediates
are evident, and replication is most likely due to an RNA-dependent activity of the host’s
RNA pol II.10-12 Unlike the plant viruses, HDV expresses a protein (HDAg) from a single
open reading frame (ORF). Early in infection, this is a 195 amino acid (aa) protein (small
Fig. 3.1. Cartoon for HDV replication. The virion enters the cell as a circular genomic
strand within a nucleoprotein complex (not shown). This is transported to the nucleus,
and the circular RNA serves as a template for a host-derived, RNA-dependent, RNA
polymerase (most likely RNA pol II). The small antigen protein, and perhaps other
host-specific factors, is required for replication, and it is perhaps associated with the
polymerase. Linear multimers are generated that are site-specifically cleaved at the cata-
lytic domains (shown as diamonds). These are then ligated by an unknown mecha-
nism to form the closed-circular antigenomic strand, which similarly acts as a template
for rolling-circle replication. The genomic strand is used to express the HDAg mRNA.
Late in infection, the closed-circular genomic strand, as a nucleoprotein complex with
the HDAg proteins, is encapsulated with the coat proteins of HBV. The catalytic do-
mains are not shown within the circular forms for clarity.
HDAg) that is required for replication (reviewed in ref. 4). Later in infection, a specific RNA
editing event eliminates a stop codon and extends the ORF by 19 aa (refs. 13, 14 and refer-
ences therein). The large HDAg inhibits replication and promotes encapsulation. Both the
large and small HDAg will stimulate self-cleavage in transfected cells, but they are not re-
quired for activity (ref. 15 and references therein). The small HDAg has some homology to
a human transcription factor,16 although there is some dispute over the significance of this.17
The genomic and antigenomic RNAs have self-cleaving activity. The antigenomic cata-
lytic domain is located just downstream of the HDAg ORF (Fig. 3.2), and it is close to one
end of the predicted rod-shaped structure. The genomic self-cleaving domain is located in
a position that is juxtaposed to the antigenomic domain in this structure; that is, the antisense
of the antigenomic domain has extensive complementarity to the genomic domain sequence
(Fig. 3.2). Hence, the two catalytic domains are similar—but not identical—in sequence.
The integrity of both the genomic and antigenomic ribozymes is required for replication in
transfected cells (see below). However, some mutations within the catalytic domains, which
otherwise do not affect catalytic activity, block replication of HDV in transfected cells.18
Thus, sequences within the catalytic domains are important for other roles in the viral life
cycle besides self-cleavage. An intact antigenomic catalytic domain stabilizes the 3' tran-
script after cleavage at the polyadenylation site (ref. 19 and references therein).
Properties of the Catalytic Domains

The catalytic domains of HDV have been the subject of extensive analyses. This mate-
rial is reviewed in detail elsewhere,20-22 so I will limit myself here to summarizing the char-
acteristics and incorporating some recent observations. The minimum sequence required
a
b
Biochemistry of Hepatitis Delta Virus Catalytic RNAs
Fig. 3.2. Features within the HDV RNA. (a) shows the genomic RNA as a closed-circular, rod-shaped structure. Intramolecular base pairing is not shown for
clarity. The genomic RNA is the template for transcribing the HDAg mRNA. Early in infection, a 195 aa ORF is expressed. Late in infection, an RNA editing
event eliminates a stop codon and extends the ORF by 19 aa. The position of the polyadenylation signal is as indicated. The antigenomic catalytic domain is
just downstream of the ORF. (b) is an enlargement of one end of the rod-shaped structure showing the sequence details of the catalytic domains and
flanking elements. This folding was done according to ref. 20 and the HDV numbering is according to ref. 9. Note that other variations of this folding are
possible and that there is no experimental evidence for this particular structure. The catalytic domains are in juxtaposition to each other within this
structure; that is, the genomic domain has extensive complementarity to the antisense sequence of the antigenomic domain.
25
for optimum catalytic activity consists of one nt 5' and 84 nt 3' to the cleavage site for both
the genomic and antigenomic domains. Slightly shorter constructs retain activity, but they
are less reactive and less stable to denaturants. Constructs with longer 3' extensions are also
often less reactive, apparently because they form competing interactions that are inhibitory.
High temperatures and/or denaturants are needed to melt out these interactions and par-
tially restore activity. However, strain comparisons and experiments with trans-cleaving
forms indicate that a small number of additional 3' sequences may be beneficial (see below).
Sequences 5' to the +1 nt (relative to the cleavage site) will enhance the activity of some
constructs, although the mechanism for this is still unclear.23,24 The minimum sequence
requirement 5' to the cleavage site is a characteristic also shared by the catalytic domain
isolated from the Neurospora VS RNA, but it is clear that the catalytic domains are not the
same (ref. 25 and references therein).
A number of secondary structure models have been proposed (reviewed in refs. 20,
22), but it now generally agreed that the pseudoknot model, first presented by Been and
coworkers,26,27 best fits the experimental data for both the genomic and antigenomic do-
mains (Fig. 3.3), and it will form the basis for my further discussions. This structure consists
of two helical regions (I, II) and two hairpins (III, IV; nomenclature of ref. 26). The se-
quences forming helix II are discontinuous and they constitute the pseudoknot interaction.
For comparison, other secondary structure models generally agree with the general charac-
ter of helix I and hairpin IV. Moreover, many of the models form hairpin III with some
minor alterations of the folding. However, none of the other models form helix II. Rather,
these sequences are generally shown interacting with sequences 5' to the cleavage site.
The secondary structure models of HDV have been extensively tested by site-specific
mutagenesis, chemical probing, enzymatic digestion and analog substitution (reviewed in
refs. 20-22). Unfortunately, these data are often very difficult to analyze and compare be-
cause of differences in the constructs and methods of analysis. As admonished by
Uhlenbeck,28 mutations within structured RNAs can have unanticipated effects. For HDV,
many mutations affect both the reaction rate and the ability of the catalytic domains to fold
into active structures. Analyses that measure the time it takes for half the material to react
inevitably incorporate the rate of the reaction plus the time it takes for misfolded molecules
to assume an active conformation. The catalytic domains of HDV seem to be particularly
prone to misfolding, even with constructs containing the wild type sequence, and widely
different interpretations of some of the data have resulted. My laboratory distinguished
between these two different phenomena by measuring the reaction rates at early times and
by comparing these rates at different reaction temperatures.29 Higher temperatures increase
the conformational flexibility of RNA and thereby facilitate the formation of active mol-
ecules. This is reflected in the reaction profiles at the different temperatures. Nevertheless,
despite these difficulties, the various analyses are largely consistent with the pseudoknot
model.
Additional studies were made to determine the spatial relationships of the different
nucleotides. Bravo et al30 used deoxy-4-thiouracil-substituted substrates in a trans-cleaving
construct of the antigenomic catalytic domain (see Fig 3.5a). The 4-thiouracil is activated
by long UV light, and it forms crosslinks with nucleotides in close proximity. They found
crosslinks between the -1 and -2 positions and C24, G28 and C76 (numbering for cis-cleav-
ing construct). Likewise, Rosenstein and Been31 used a photo-activatable azidophenacyl group
tethered to the phosphate at the cleavage site in a similar system and in a cis-cleaving sys-
tem. They obtained crosslinks within junction IV/II, the 5' half of helix III and within loop
III, showing that these regions are in close proximity. Oddly, the in trans form gave a some-
what different crosslinking pattern than the in cis form. Lead-cleavage studies also indicate
that these regions are close together in tertiary space.32 Recent mutagenic data indicate that
Biochemistry of Hepatitis Delta Virus Catalytic RNAs 27
Fig. 3.3. The pseudoknot secondary-structure models of the catalytic domains. The minimal
sequence requirements for optimal activity are shown. Numbering is relative to the cleavage site
and nomenclature is according to refs. 26, 27. The cleavage sites are indicated by arrowheads.
Nucleotides that are important for catalytic activity are indicated in bold.
G41 and G75 in the antigenomic domain are probably stacked at the base of hairpin IV, and
they may form non Watson-Crick base pairs.33 Similar interactions are possible for G40 and
G74 in the genomic domain, but the data for this are still inconclusive.33 Probing with
Fe(II)-EDTA, which is sensitive to exposed ribose functionalities, shows that the junction
regions I/IV and IV/II, and loop III are largely protected from the solvent.31 This result is
consistent with phosphorothiolate-interference studies.34 Ethidium bromide and time-re-
solved fluorescence spectroscopy indicate that the secondary structure of the antigenomic
domain is highly structured even in 95% formamide at 25°C.35 In the same vein, NMR
studies of a model RNA designed to mimic the antigenomic hairpin III, with helix II coaxially
stacked at its base, show that the structure is particularly stable, and that it forms unusual
interactions.36
Strain Comparisons
Phylogenetic comparisons are a powerful technique for elucidating important sequences
involved in secondary and tertiary interactions in RNA.6,37,38 However, since the genomic
and antigenomic HDV RNAs are the sole examples of the pseudoknot catalytic motif, this
methodology might be expected to be of limited use. Fortunately, this is not the case. Like
many RNA viruses, HDV has a very high frequency of mutations during replication, and
there is significant sequence variability between different isolates. Currently, there are three
classified genotypes of the virus (ref. 39 and reference therein). Genotype I is the most
widespread, and it is associated with a range of clinical symptoms, from very mild to severe.
Genotype II is less widespread, and it is associated with a mild form of the disease. It is 88.5
to 95.6% similar to genotype I.40 In contrast, genotype III, which also has a limited distribu-
tion, is associated with a particularly severe form of the disease.41 It is more divergent from
the other two genotypes and shares only 61 to 80% similarity.40,41 A phylogenetic tree was
derived for much of this sequence data.42
Sequence alignments of the catalytic domains were made previously,20,22 and they sup-
port the pseudoknot model as shown in Figure 3.4. However, additional sequences were
added to the databases since these publications. These data largely confirm the previous
interpretations, but they provide some additional information. This material is summa-
rized in Figure 3.4. However, I should state that the events leading to the changes within
highly variable regions are unknown. Sequence alignments are therefore based on the sim-
plest interpretation of the data that involve the fewest alterations. Recent data provide evi-
dence for slightly different interpretations of the variability within hairpin IV and the 3'
flanking sequences than previously shown.20,22 Nevertheless, despite the high degree of iso-
late variability, it is clear that the core sequences of the catalytic domains are strongly con-
served. Most of the changes occur within hairpin IV, which mutagenic data indicate to be
very malleable. These changes rearrange but do not alter the general characteristics of the
structure.
For convenience in the further discussions, genomic sequences will be indicated as (-)
and the antigenomic as (+). The equivalent nucleotides G76(-) and U77(+) were shown to
be alterable by site-specific mutagenesis and the substitution of an A or C at this position is
not surprising.43,44 The A78G(+) change is unexpected since changing this position to a
uridine, or the equivalent A77U(-), significantly reduces activity in vitro.43,44 However, a
purine substitution may be better tolerated. U23C(-) and the equivalent U26C(+) maintain
the pyrimidine character of the loop, which is consistent with the mutagenic data (reviewed
in refs. 20, 22). The more extensive alterations of the genomic loop III sequence in the Lai et
al isolate45 (k) are somewhat surprising, and they may reflect an artifact; nevertheless, they
make the spatial relationship of the sequence more like that found in the antigenomic loop.
As previously noted,20 the sequence variability of the different isolates is consistent with
stem II forming and with its elongation by one nucleotide for the genomic domain (gener-
ally) and five nt for the antigenomic domain (always). Finally, the 3' flanking sequence gives
a suggestion for the enormous potential for variability within the HDV sequence. The fact
that the 5' flanking sequences show less variability could reflect the importance of these
sequences in catalytic activity as previously proposed.23,24 However, it is probable that they
are important for other aspects of the viral life cycle.
Data Summary
The experimental data and sequence comparisons are consistent with the pseudoknotted
structure as shown in Figure 3.3. Helix I, helix II and hairpin IV are mostly structural ele-
ments in which the sequences can be altered as long as the structural features are main-
tained. There are some sequence requirements around the cleavage site in helix I, and this
will be discussed in more detail in the section on trans-cleaving ribozymes. Helix III is also
a structural element, but it has sequence-specific requirements. It is most likely coaxially
stacked with helix II, rather than with helix I; as discussed below, this is needed to maintain
the spatial relationship of the loop III sequences relative to the catalytic site. The single-
stranded regions loop III, junction I/IV, and junction IV/II have specific sequence require-
ments and together with helices I and III they probably form the catalytic core of the
ribozyme. They may be involved in chelating the divalent cation(s) and orienting the scis-
sile linkage.
Fig. 3.4. Sequence variability of HDV isolates. The primary sequences are those most commonly
found. These sequences were isolated from woodchuck (which had passed through chimpanzee
and human),78 human,79 chimpanzee,80 Central African Republic isolate passed through wood-
chuck,81 North American (14 clones),82 Italian isolate passed through chimpanzee (origin of
HDV used to infect woodchucks),78,9 and from several French patients.78,83 Note that while the
sequence of isolates was the same in the region shown, they varied at other positions. Positions of
sequence variability are as shown: ◊ indicates insertions; ∆ indicates deletions. However, in highly
variable regions (helix IV and the 3' flanking sequence) of some isolates, the nature of the changes
are unclear. The changes shown represent the simplest interpretation that is consistent with the
alignment of all the strains. The parentheses indicate the isolate: (a) human US-2;41 b) northern
South American Peru-1;41 (c) South Pacific island of Nauru;84 (d) Taiwan;85 (e) Japan, patient M
(9/20/86);86 (f) Japan, patient M (7/6/89);86 (g) Japan, patient S (7/18/83);86 (h) Japan, patient S
(8/6/87);86 (i) Japan;87 (j) woodchuck;88 (k) human;45 (l) Lebanon;89 (m) Taiwan strain Taiwan
3;40 (n) central China;90 (o) patient from southern California;82 and (p), (q), (r) three patients
from Los Angeles.85 Hairpin IV is redrawn in the case of the Peru-1 isolate for the genomic
sequence to clarify the nature of the changes (indicated by the wedges). Microheterogeneity (quasi-
species) within clones is not shown because of uncertainty as to the source of the difference.
A three-dimensional model of the genomic ribozyme was derived from these data with
an interactive graphical computer.44 This model brings helix I, loop III, junction I/IV and
junction VI/II into close proximity, and it has helix II and hairpin IV pointing off from the
catalytic core. A similar model was also made for the antigenomic strand.30 A different model
was proposed for the antigenomic model based on the axehead secondary structure.46 This
model is similar to the previous two models, but it is more open and junction IV/II follows
a different pathway. However, recent data are largely consistent with the pseudoknot three-
dimensional model. Crosslinking studies with photo-activatable agents close to the scissile
linkage show the close proximity of loop III and junction IV/II to the cleavage site.30,31 Fe(II)-
EDTA protection31 and phosphorothiolate interference34 experiments also indicate that these
regions are protected from the solvent, as predicted by the model. However, details of the
model need to be refined. The crosslinking studies of Been and Rosenstein31 and biophysical
studies of Kolk et al36 indicate that the loop III sequence has a slightly different orientation
with respect to the cleavage site. Moreover, junction I/IV is more protected than expected
from the model.31 Clearly additional studies are needed to further clarify the spatial rela-
tionships of the different elements.
Self-Cleavage Reaction
The catalytic requirements for HDV were reviewed in detail previously;20-22 I will sim-
ply summarize that information here and add some recent information. Like all catalytic
RNAs, HDV has an absolute requirement for a divalent cation, such as Mg2+, Ca2+ or Mn2+.
Very low concentrations (< 0.1 mM) of these cations are sufficient to cleave optimized con-
structs. Monovalent cations do not support the reaction. Circular dichroism measurements
indicate that there are three bound Mg2+, although this was done with a three-strand con-
struct (Fig. 3.5f) that required 100 mM Mg2+ for optimum activity.47
Like the other self-cleaving RNAs, the reaction generates products with 2',3-cyclic phos-
phates and 5' hydroxyls. A 2' deoxyribose at the scissile linkage blocks cleavage.48 Previous
work showed that the reaction was independent of pH from 5.0 to 9.0, but recent experi-
ments show a linear increase of the reaction rate with increasing pH from 4.0 to 6.0, with a
pH optimum around 7.0 to 7.5.47,49 This latter result is more consistent with what one would
expect if the observed reaction rate actually corresponds to the chemical cleavage step. The
temperature optimum is around 55°C to 65°C for constructs near the optimal size.20
Phosphorothiolate substitutions have also been made at the scissile linkage.34,49 A pro-Rp
phosphorothiolate is poorly cleaved, and it is not recovered by using Mn2+. The pro-Sp
phosphorothiolate was slightly less reactive than the normal linkage. These results suggest
that Mg2+ does not interact directly with the pro-R oxygen.49 A similar result was also re-
cently obtained for the hammerhead ribozyme.50
Previous studies indicated that the catalytic domains also catalyze a ligation reaction,
which might be used to generate the circular templates used in replication and encapsula-
tion (Fig. 3.1).51,52 However, in one case this was shown to be an artifact of the assay condi-
tions53,54 and in the other the reaction conditions were not physiological and an equal pro-
portion of both 2',5' and 3',5' linkages was formed.52 Currently, the best indication is that
the RNAs are circularized by a host-encoded ligase.55,56
Biological Relevance
Recently, a number of experiments were conducted to determine the biological rel-
evance of the catalytic domains in the life cycle of the virus. Mutations that disrupt catalytic
activity in vitro will similarly block replication of the virus in cell culture.56,18 Moreover,
shortening the rod-shaped structure to within a few nt of the 3' ends of the catalytic do-
mains reduces replication to barely detectable levels.57 Additional studies have also looked
at the cleavage activity independent of viral replication.18 These studies indicate that cata-
lytic activity is essential for replication, but that sequence alterations that maintain self-
cleavage activity are not always sufficient to maintain replication. Hence, the sequences within
the catalytic domains play other roles in the life cycle of the virus besides self-cleavage. This
may account for the high degree of sequence conservation within the catalytic domains,
which is in contrast to the flexibility revealed by the mutagenesis experiments. Finally, mu-
tations were made and tested in cell culture to distinguish among three common HDV
secondary structure models.18 These experiments show that the sequences defining the
pseudoknot structure are required for activity and that this structure is used in vivo.
Since the self-cleaving domains are regenerated with the ligation of the linear progeny,
the catalytic activity of HDV must be regulated in some fashion during the viral life cycle to
prevent inappropriate cleavage of the template or encapsulated RNA. A model was pro-
Fig. 3.5. Trans-cleaving forms of the catalytic domains. (a) The catalytic domain is sepa-
rated within junction I/II. This is the most commonly used form, and it was originally
designed by Been and coworkers.26,27 (b) is a variant where the transcriptional start and
stop have been circularly permuted.48,66 (c) is a completely closed circular ribozyme that
was constructed with the aid of a permuted catalytic intron.66 (d) This construct was
originally designed based on the ax head secondary structure,23 but it is equally appli-
cable to the pseudoknot model. (e) is a variant of (d) where the transcriptional start and
stop sites are circularly permuted.64 (f) is a trans-cleaving form made from three sepa-
rate strands.63,47 Genomic, antigenomic and synthetic sequences are used (see text and
Table 3.1).
posed (ref. 58 and references therein) where the rod-shaped structure of the HDV RNA
(Fig. 3.2) sequesters the catalytic domain sequences and prevents formation of the active
structure. However, during replication the catalytic domain rapidly forms and cleaves the
precursor before these interactions occur. According to this model, the dimer and trimer
molecules found in vivo are dead-end replication products that are blocked for self-cleav-
age. Evidence for this model is that constructs containing sequences that are complemen-
tary to the catalytic domain (attenuators), which are either HDV-derived 3' to the catalytic
domain or artificial (5' or 3' to the catalytic domain), are processed to circular products in
human hepatoma cells.55 They are not processed (self-cleaved or ligated) in vitro. The HDV-
encoded HDAg enhances cleavage, but it is not necessary.15 These results indicate that host-
encoded proteins, in conjunction with the attenuator sequence, participate in regulating
the catalytic activity. Moreover, constructs containing attenuators are properly processed in
mammalian cells but not in yeast or Escherichia coli, which give results similar to those
obtained in vitro.55 Thus, the factor(s) is (are) probably mammalian-specific. An intriguing
observation in this light is the sequence similarity between the genomic catalytic domain
and human 7S RNA.59
Trans-Cleaving Activity
As with the other catalytic RNAs, the cis-cleaving activity of HDV can be converted
into a trans-cleaving activity by dividing the catalytic domain into separate “substrate” and
“ribozyme.” I put these in quotations because both are required for the formation of the
active structure and because there are many ways of making such constructs. Hence, the
substrate is defined as the molecule that is cleaved and the ribozyme is the unaltered, reus-
able component. Figure 3.5 shows different variants of the HDV ribozymes, which are ap-
plicable for both the genomic and the antigenomic forms. It is clear that a number of vari-
ants are possible. The catalytic domains have been successfully divided within junction I/II
(Fig. 3.5a,b,c)48,60-62 and within loop IV (Fig. 3.5d,e).23,62 It is also possible to make a con-
struct, consisting of three separate strands, that is divided in both regions (Fig. 3.5f).47,63 In
some instances, at least, molecules can be separated at the 3' extremity of loop III,64 al-
though cleaving within loop III would normally be expected to eliminate activity.63,65 It is
not possible to separate the molecule within junctions II/III, III/I, I/IV and IV/II.63-65 Sepa-
ration within helix IV is possible, but constructs separated within helix I or II are less effec-
tive (Lescure and Tanner, unpublished data).64 The transcription starts and stops can be
circularly permuted within these structures (Fig. 3.5b,e).60,64,66 A particularly intriguing vari-
ant takes advantage of permuted exon-intron sequences within a self-splicing group I in-
tron to generate a completely closed-circular molecule (Fig. 3.5c).66 This molecule is resis-
tant to nucleases found in serum66 and recently it was successfully generated within E. coli.67
A comparison of the catalytic properties of these constructs is shown in Table 3.1. Care
must be exercised in evaluating this table because the details of the individual constructs
vary substantially. Many of the constructs have an artificial hairpin IV, some have an abbre-
viated stem II and others are hybrid constructs containing sequence elements from both the
genomic and antigenomic domains. Moreover, no distinction is made between single turn-
over (ribozyme saturating) and multiple turnover (substrate saturating) conditions. The
latter can be 10-fold less than the former, which probably reflects rate-limiting product
release.48 Association constants are shown, where available, but they are consistent with the
values expected for Watson-Crick base pairing between the substrate and ribozyme. Up to
12-fold turnover was obtained with various hybrid constructs at physiological temperatures.66
However, these were obtained using Ca2+ as the divalent cation; Mg2+ was less effective.
The most frequently used designs of the trans-cleaving ribozymes are variants of con-
structs separated within junction I/II (Fig. 3.5a,b,c), where the substrate consists of an 8
nucleotide sequence that binds to the ribozyme to form helix I. There are minimal sequence
requirements for this construct (see below), and product release is not expected to be as
limiting as in the other constructs. On the other hand, a target site of 7 to 8 nt is of limited
specificity. Unfortunately, as discussed below, it is not yet possible to extend helix I. More-
over, making base pairs between sequences 5' to the -1 nucleotide and junction I/IV are
strongly inhibitory (Lescure and Tanner, unpublished data).22 In contrast, ribozymes sepa-
rated within loop IV contain extensive base pairs (Fig. 3.5d). However, in this case the sub-
strate has many of the conserved nucleotides, and these are unlikely to be found in a tar-
geted RNA. The circularly permuted variant (Fig. 3.5e) reduces the number of conserved nt
within the substrate, but it imposes other limitations.
Table 3.1. Comparison of trans-cleaving ribozymes
Construct* Cleavage Rate Turnover Conditions

(kobs)
(a) Genomic65 0.022 min–1 ND pH 7.1, 37°C, 1 mM MgCl2

(1.7 min–1)‡
(a) Genomic49,68 ~1.90 min–1 ND pH 7.4, 37°C, 10 mM MgCl2
(a) Genomic63 ND ND pH 8.0, 37°C, 10 mM MgCl2
(a) Antigenomic48 ~2.8 min–1 6-fold pH 8.0, 55°C, 10 mM MgCl2
(a) Antigenomic61 ND ND 55°C, 10 mM MgCl2
(a) Hybrid60 1.5 min–1 6-fold pH 8.0, 37°C, 10 mM MgCl2
(a) Hybrid66 ~0.25 min–1 ~12-fold pH 8.0, 37°C, 10 mM CaCl2
(~5-fold)¤
(a) Selected68 2.45 min–1 ND pH 7.4, 37°C, 10 mM MgCl2
(b) Hybrid60 0.52 min–1 3-fold pH 8.0, 37°C, 10 mM MgCl2
(b) Hybrid66 ~0.25 min–1 ~12-fold pH 8.0, 37°C, 10 mM CaCl2
(~3.5-fold)¤
(c) Hybrid66 0.29 min–1 ~12-fold pH 8.0, 37°C, 10 mM CaCl2
(~3-fold)¤
(d) Genomic62 0.011 min–1 ND pH 7.2, 50°C, 12 mM MgCl2
(d) Genomic/ ND ND pH 7.5, 50°C, 5 mM MgCl2
Antigenomic23
(e) Hybrid64 0.66 min–1 6-fold pH 7.5, 50°C, 12 mM MgCl2
(e) Hybrid91 0.17 min–1 8-fold pH 7.5, 50°C, 12 mM MgCl2
(0.98 min–1)‡ (3-fold)‡
(f) Genomic63 ND ND pH 8.0, 37°C, 10 mM MgCl2
(f) Hybrid47 ~0.09 min–1 ND pH 8, 37°C, 10 mM MgCl2
(0.38 min–1)#
* the letters in parentheses refer to the constructs shown in Figure 3.5. ‘Hybrid’ is a combination of
genomic and antigenomic sequences, and it generally contains a synthetic hairpin IV. ‘Selected’ is
an in vitro selected sequence.
‡, longer helix II; ¤, turnover in the presence of 10 mM MgCl ; #, reaction in 100 mM MgCl .
2 2
In general, the antigenomic sequence is more effective than the genomic sequence as a
trans-cleaving ribozyme. The reasons for this are unclear, but it appears to reflect the greater
propensity of the genomic catalytic domain to misfold or for the substrate to bind in a
nonproductive fashion (Lescure and Tanner, unpublished data). Interestingly, in vitro se-
lection experiments of a randomly mutagenized genomic sequence yields a construct that is
very similar to the antigenomic sequence.68 For this reason, many workers combined differ-
ent sequence elements from the two domains to make HDV hybrids. These normally have
the antigenomic helix I and loop III, and the genomic helix II, helix III, junction I/IV, and
junction IV/II. Hairpin IV often consists of a shorter artificial sequence. Extending helix II
will enhance the catalytic rate of at least some constructs (Table 3.1).65,68
While not all possible basepair combinations have been tested, it appears there are few
sequence requirements for helix I other than that basepairing is maintained and that the
helix is 7 base pairs long. There is little evidence for a base pair at the -1 position, although
it is reasonable to expect that the -1 nt is constrained in some fashion for cleavage to occur.
Extending the helix or shortening it drastically reduces activity.44,69 Watson-Crick base pairs
are preferred and G-U base pairs at positions other than the +1 position are inhibitory.20,69
The -1 nucleotide preference is C = U>A>>G for the antigenomic43 and U>C>A>>G for
the genomic,70 although except for G the differences are not large. The +1 position is prefer-
ably a purine with a pyrimidine base pair (G-U>A-C>G-C>A-U for the antigenomic43 and
G-U>G-C>A-U for the genomic).70 Interestingly, a G-C base pair was obtained at the +1
position in the in vitro selection experiments.68 In the same set of in vitro selection experi-
ments, a variant catalytic domain was isolated that cleaves between +1 and +2 when there is
a U-G, G-G or A-G base pair at the +1 position.70 This phenomenon is unique for the
selected construct, so its significance is unclear. The mechanism for this aberrant cleavage
site is unknown, but this construct may be useful in elucidating the characteristics of the
catalytic core.
Applications
The applications of the trans-cleaving forms of HDV are rather limited to date. Re-
searchers generally do not publish their failed experiments, so it is unclear how often the
trans-cleaving ribozymes have been tested. However, to my knowledge there is no successful
application of the various trans-cleaving forms in vivo or in cell culture. This may reflect
the difficulty of obtaining effective ribozymes at physiological temperatures. Effective, in
this case, means both high activity and turnover. The major problem seems to be maintain-
ing the ribozyme element in a conformation that is both competent for binding the sub-
strate and functional for catalysis. Fortunately, people are devising clever ways of reducing
these problems, and future constructs may suffer less from these problems. This bodes well
for their future use in vivo.
However, because of the minimum sequence requirements 5' to the cleavage site, the
cis-cleaving form has found a number of applications. It is used to trim the ends of in vitro
transcribed RNAs to generate discrete products for biophysical studies.71 Likewise, it is used
to release ribozymes that are synthesized as part of a long transcript, and the HDV ribozyme
was compared with the hammerhead and hairpin as releasing elements for processing
ribozyme cassettes generated in vitro and in cell culture.72 Finally, it seems to be a popular
tool amongst negative-strand RNA virologists for generating the discrete ends of cDNA
transcripts in vivo that are required for replication.73-76
Moreover, since the HDV pathogen is a serious health problem and since the catalytic
domains are essential for replication, the HDV ribozymes are potential targets themselves
for therapeutic agents. Some antibiotics are effective inhibitors of catalytic activity in vitro
at µmolar concentrations.32 Although there is little evidence that these agents are equally
effective in vivo, they nevertheless provide the basis for combinatorial screenings to isolate
forms that may prove effective.77 Antisense oligonucleotides are also effective inhibitors of
catalytic activity in vitro (Thill, Crain-Denoyelle and Tanner, unpublished data) and in cell
culture (Sureau, unpublished data).
Concluding Remarks
Catalytic RNAs are widely found in plants, bacteria, bacteriophage and lower eukary-
otes. However, HDV is the only ribozyme found in man, and its catalytic activity is probably
optimized for this cellular environment. Indeed, as described above, the self-cleaving reac-
tion of HDV is regulated only within mammalian cells. Moreover, the HDV virion can be a
vector for delivering the ribozyme to the intended cell. While earlier constructs were
uninspiring, recent trans-cleaving ribozymes show high activity and turnover under physi-
ological conditions. With the aid of the three-dimensional model, it may be possible to
further optimize the activity and to design constructs with greater sequence specificity. There
is still much to learn about the HDV catalytic domains.
Acknowledgments
I am grateful to Patrick Linder and to members of the Department of Medical Bio-
chemistry for their support. I thank Josette Banroques for checking through the text. This
work was supported, in part, with a grant from the Roche Research Foundation.
References
1. Taylor JM. Human hepatitis delta virus. Curr Top Microbiol Immunol 1991; 168:141-166.
2. Taylor JM. The structure and replication of hepatitis delta virus. Annu Rev Microbiol 1992;
46:253-276.
3. Lai MM. Molecular biologic and pathogenetic analysis of hepatitis delta virus. J Hepatol
1995; 22:127-131.
4. Lai MMC. The molecular biology of hepatitis delta virus. Annu Rev Biochem 1995;
64:259-286.
5. Poisson F, Roingeard P, Goudeau A. The hepatitis delta virus: A singular replication. M S-
Med Sci 1995; 11:1379-1387.
6. Symons RH. Small catalytic RNAs. Annu Rev Biochem 1992; 61:641-671.
7. Symons RH. Plant pathogenic RNAs and RNA catalysis. Nucleic Acids Res 1997;
25:2683-2689.
8. Kos A, Dijkema R, Arnberg AC et al. The hepatitis delta (δ) virus possesses a circular
RNA. Nature 1986; 323:558-560.
9. Wang KS, Choo QL, Weiner AJ et al. Structure, sequence and expression of the hepatitis
delta (δ) viral genome [published erratum appears in Nature 1987; 328:456]. Nature 1986;
323:508-514.
10. MacNaughton TB, Gowans EJ, McNamara SP et al. Hepatitis delta antigen is necessary for
access of hepatitis delta virus RNA to the cell transcriptional machinery but is not part of
the transcriptional complex. Virology 1991; 184:387-390.
11. Fu TB, Taylor J. The RNAs of hepatitis delta virus are copied by RNA polymerase II in
nuclear homogenates. J Virol 1993; 67:6965-6972.
12. Beard MR, MacNaughton TB, Gowans EJ. Identification and characterization of a hepatitis
delta virus RNA transcriptional promoter. J Virol 1996; 70:4986-4995.
13. Casey JL, Gerin JL. Hepatitis D virus RNA editing: Specific modification of adenosine in
the antigenomic RNA. J Virol 1995; 69:7593-7600.
14. Polson AG, Bass BL, Casey JL. RNA editing of hepatitis delta virus antigenome by dsRNA-
adenosine deaminase. Nature 1996; 380:454-456.
15. Jeng KS, Su PY, Lai MM. Hepatitis delta antigens enhance the ribozyme activities of hepa-
titis delta virus RNA in vivo. J Virol 1996; 70:4205-4209.
16. Brazas R, Ganem D. A cellular homolog of hepatitis delta antigen: Implications for viral
replication and evolution. Science 1996; 274:90-94.
17. Long M, de Souza SJ, Gilbert W. Delta-interacting protein A and the origin of hepatitis
delta antigen. Science 1997; 276:824-825.
18. Jeng KS, Daniel A, Lai MM. A pseudoknot ribozyme structure is active in vivo and re-
quired for hepatitis delta virus RNA replication. J Virol 1996; 70:2403-2410.
19. Hsieh SY, Yang PY, Chu CM et al. Role of the ribozyme of hepatitis delta virus on the
transcription after polyadenylation. Biochem Biophys Res Commun 1994; 205:864-871.
20. Tanner NK. The catalytic RNAs from hepatitis delta virus: Structure, function and appli-
cations. In: Dinter-Gottlieb G, ed. The Unique Hepatitis Delta Virus. Austin: R.G. Landes,
1995:11-31.
21. Been MD. Cis- and trans-acting ribozymes from a human pathogen, hepatitis delta virus.
Trends Biochem Sci 1994; 19:251-256.
22. Been MD, Wickham GS. Self-cleaving ribozymes of hepatitis delta virus RNA. Eur J Biochem
1997; 247:741-753.
23. Branch AD, Robertson HD. Efficient trans cleavage and a common structural motif for the
ribozymes of the human hepatitis delta agent. Proc Natl Acad Sci USA 1991; 88:10163-10167.
24. Gottlieb PA, Prasad Y, Smith JB et al. Evidence that alternate foldings of the hepatitis delta
RNA confer varying rates of self-cleavage. Biochemistry 1994; 33:2802-2808.
25. Beattie TL, Collins RA. Identification of functional domains in the self-cleaving Neuro-
spora VS ribozyme using damage selection. J Mol Biol 1997; 267:830-840.
26. Perrotta AT, Been MD. A pseudoknot-like structure required for efficient self-cleavage of
hepatitis delta virus RNA. Nature 1991; 350:434-436.
27. Rosenstein SP, Been MD. Evidence that genomic and antigenomic RNA self-cleaving ele-
ments from hepatitis delta virus have similar secondary structures. Nucleic Acids Res 1991;
19:5409-5416.
28. Uhlenbeck OC. Keeping RNA happy. RNA 1995; 1:4-6.
29. Thill G, Vasseur M, Tanner NK. Structural and sequence elements required for the self-
cleaving activity of the hepatitis delta virus ribozyme. Biochemistry 1993; 32:4254-4262.
30. Bravo C, Lescure F, Laugaa P et al. Folding of the HDV antigenomic ribozyme pseudoknot
structure deduced from long-range photocrosslinks. Nucleic Acids Res 1996; 24:1351-1359.
31. Rosenstein SP, Been MD. Hepatitis delta virus ribozymes fold to generate a solvent-inac-
cessible core with essential nucleotides near the cleavage site phosphate. Biochemistry 1996;
35:11403-11413.
32. Rogers J, Chang AH, von Ahsen U et al. Inhibition of the self-cleavage reaction of the
human hepatitis delta virus ribozyme by antibiotics. J Mol Biol 1996; 259:916-925.
33. Been MD, Perrotta AT. Optimal self-cleavage activity of the hepatitis delta virus RNA is
dependent on a homopurine base pair in the ribozyme core. RNA 1995; 1:1061-1070.
34. Jeoung YH, Kumar PK, Suh YA et al. Identification of phosphate oxygens that are impor-
tant for self-cleavage activity of the HDV ribozyme by phosphorothioate substitution in-
terference analysis. Nucleic Acids Res 1994; 22:3722-3727.
35. Duhamel J, Liu DM, Evilia C et al. Secondary structure content of the HDV ribozyme in
95% formamide. Nucleic Acids Res 1996; 24:3911-3917.
36. Kolk MH, Heus HA, Hilbers CW. The structure of the isolated, central hairpin of the
HDV antigenomic ribozyme: Novel structural features and similarity of the loop in the
ribozyme and free in solution. EMBO J 1997; 16:3685-3692.
37. Kirsebom LA. RNase P—a ‘Scarlet Pimpernel.’ Mol Microbiol 1995; 17:411-20.
38. Saldanha R, Mohr G, Belfort M et al. Group I and group II introns. FASEB J 1993; 7:15-24.
39. Shakil AO, Hadziyannis S, Hoofnagle JH et al. Geographic distribution and genetic vari-
ability of hepatitis delta virus genotype I. Virology 1997; 234:160-167.
40. Lee CM, Changchien CS, Chung JC et al. Characterization of a new genotype II hepatitis
delta virus from Taiwan. J Med Virol 1996; 49:145-154.
41. Casey JL, Brown TL, Colan EJ et al. A genotype of hepatitis D virus that occurs in north-
ern South America. Proc Natl Acad Sci USA 1993; 90:9016-9020.
42. Krushkal J, Li WH. Substitution rates in hepatitis delta virus. J Mol Evol 1995; 41:721-726.
43. Perrotta AT, Been MD. Core sequences and a cleavage site wobble pair required for HDV
antigenomic ribozyme self-cleavage. Nucleic Acids Res 1996; 24:1314-1321.
44. Tanner NK, Schaff S, Thill G et al. A three-dimensional model of hepatitis delta virus
ribozyme based on biochemical and mutational analyses. Curr Biol 1994; 4:488-498.
45. Lai ME, Mazzoleni AP, Porru A et al. Unpublished 1995; accession number X85253.
46. Branch AD, Polaskova JA. 3-D models of the antigenomic ribozyme of the hepatitis delta
agent with eight new contacts suggested by sequence analysis of 188 cDNA clones. Nucleic
Acids Res 1995; 23:4180-4189.
47. Sakamoto T, Tanaka Y, Kuwabara T et al. Properties of hepatitis delta virus ribozyme,
which consists of three RNA oligomer strands. J Biochem Tokyo 1997; 121:1123-1128.
48. Perrotta AT, Been MD. Cleavage of oligoribonucleotides by a ribozyme derived from the
hepatitis delta virus RNA sequence. Biochemistry 1992; 31:16-21.
49. Fauzi H, Kawakami J, Nishikawa F et al. Analysis of the cleavage reaction of a trans-acting
human hepatitis delta virus ribozyme. Nucleic Acids Res 1997; 25:3124-3130.
50. Zhou DM, Kumar PKR, Zhang LH et al. Ribozyme mechanism revisited: Evidence against
direct coordination of a Mg2+ ion with the pro-R oxygen of the scissile phosphate in the
transition state of a hammerhead ribozyme-catalyzed reaction. J Am Chem Soc 1996;
118:8969-8970.
51. Wu HN, Lai MM. Reversible cleavage and ligation of hepatitis delta virus RNA. Science
1989; 243:652-654.
52. Sharmeen L, Kuo MY, Taylor J. Self-ligating RNA sequences on the antigenome of human
hepatitis delta virus. J Virol 1989; 63:1428-1430.
53. Rosenstein SP, Been MD. Self-cleavage of hepatitis delta virus genomic strand RNA is en-
hanced under partially denaturing conditions. Biochemistry 1990; 29:8011-8016.
54. Wu HN, Lai MM. RNA conformational requirements of self-cleavage of hepatitis delta
virus RNA. Mol Cell Biol 1990; 10:5575-5579.
55. Lazinski DW, Taylor JM. Intracellular cleavage and ligation of hepatitis delta virus ge-
nomic RNA: Regulation of ribozyme activity by cis-acting sequences and host factors. J
Virol 1995; 69:1190-1200.
56. Macnaughton TB, Wang YJ, Lai MM. Replication of hepatitis delta virus RNA: Effect of
mutations of the autocatalytic cleavage sites. J Virol 1993; 67:2228-2234.
57. Wu TT, Netter HJ, Lazinski DW et al. Effects of nucleotide changes on the ability of hepa-
titis delta virus to transcribe, process, and accumulate unit-length, circular RNA. J Virol
1997; 71:5408-5414.
58. Lazinski DW, Taylor JM. Regulation of the hepatitis delta virus ribozymes: To cleave or
not to cleave? RNA 1995; 1:225-233.
59. Negro F, Gerin JL, Purcell RH et al. Basis of hepatitis delta virus disease? Nature 1989;
341:111.
60. Been MD, Perrotta AT, Rosenstein SP. Secondary structure of the self-cleaving RNA of
hepatitis delta virus: Applications to catalytic RNA design. Biochemistry 1992; 31:
11843-11852.
61. Prasad Y, Smith JB, Gottlieb PA et al. Deriving a 67-nucleotide trans-cleaving ribozyme
from the hepatitis delta virus antigenomic RNA. Antisense Res Dev 1992; 2:267-277.
62. Wu HN, Wang YJ, Hung CF et al. Sequence and structure of the catalytic RNA of hepati-
tis delta virus genomic RNA. J Mol Biol 1992; 223:233-245.
63. Lescure F, Blumenfeld M, Thill G et al. Trans cleavage of RNA substrates by an HDV-
derived ribozyme. Prog Clin Biol Res 1993; 382:99-108.
64. Lai YC, Lee JY, Liu HJ et al. Effects of circular permutation on the cis-cleavage reaction of
a hepatitis delta virus ribozyme: Application to trans-acting ribozyme design. Biochemistry
1996; 35:124-131.
65. Kawakami J, Yuda K, Suh YA et al. Constructing an efficient trans-acting genomic HDV
ribozyme. FEBS Lett 1996; 394:132-136.
66. Puttaraju M, Perrotta AT, Been MD. A circular trans-acting hepatitis delta virus ribozyme.
67. Puttaraju M, Been MD. Circular ribozymes generated in Escherichia coli using group I self-
splicing permuted intron-exon sequences. J Biol Chem 1996; 271:26081-26087.
68. Nishikawa F, Kawakami J, Chiba A et al. Selection in vitro of trans-acting genomic human
hepatitis delta virus (HDV) ribozymes. Eur J Biochem 1996; 237:712-718.
69. Wu HN, Lee JY, Huang HW et al. Mutagenesis analysis of a hepatitis delta virus genomic
ribozyme. Nucleic Acids Res 1993; 21:4193-4199.
70. Nishikawa F, Fauzi H, Nishikawa S. Detailed analysis of base preferences at the cleavage
site of a trans-acting HDV ribozyme: A mutation that changes cleavage site specificity.
71. Ferre-D’Amare AR, Doudna JA. Use of cis- and trans-ribozymes to remove 5' and 3' het-
erogeneities from milligrams of in vitro transcribed RNA. Nucleic Acids Res 1996;
24:977-978.
72. Chowrira BM, Pavco PA, McSwiggen JA. In vitro and in vivo comparison of hammerhead,
hairpin, and hepatitis delta virus self-processing ribozyme cassettes. J Biol Chem 1994;
269:25856-25864.
73. Bridgen A, Elliott RM. Rescue of a segmented negative-strand RNA virus entirely from
cloned complementary DNAs. Proc Natl Acad Sci USA 1996; 93:15400-15404.
74. Conzelmann KK, Schnell M. Rescue of synthetic genomic RNA analogs of rabies virus by
plasmid-encoded proteins. J Virol 1994; 68:713-719.
75. Sidhu MS, Chan J, Kaelin K et al. Rescue of synthetic measles virus minireplicons: Measles
genomic termini direct efficient expression and propagation of a reporter gene. Virology
1995; 208:800-807.
76. Pattnaik AK, Ball LA, LeGrone AW et al. Infectious defective interfering particles of VSV
from transcripts of a cDNA clone. Cell 1992; 69:1011-1020.
77. Mei HY, Cui M, Sutton ST et al. Inhibition of self-splicing group I intron RNA: High-
throughput screening assays. Nucleic Acids Res 1996; 24:5051-5053.
78. Deny P, Zignego AL, Rascalou N et al. Nucleotide sequence analysis of three different
hepatitis delta viruses isolated from a woodchuck and humans. J Gen Virol 1991; 72:735-739.
79. Houghton M, Wang K-S, Choo Q-L et al. Hepatitis delta diagnostics and vaccines, their
preparation and use. Patent 1988; EP 0251575-A1.
80. Kos T, Molijn A, van Doorn LJ et al. Hepatitis delta virus cDNA sequence from an acutely
HBV-infected chimpanzee: Sequence conservation in experimental animals. J Med Virol
1991; 34:268-279.
81. Langon T, Pichoud C, Jamard C et al. Analysis of a hepatitis delta virus (HDV) isolate
from the Central African Republic. Unpublished 1997; accession number AJ000558.
82. Makino S, Chang MF, Shieh CK et al. Molecular cloning and sequencing of a human hepa-
titis delta (δ) virus RNA. Nature 1987; 329:343-346.
83. Saldanha JA, Thomas HC, Monjardino JP. Cloning and sequencing of RNA of hepatitis
delta virus isolated from human serum. J Gen Virol 1990; 71:1603-1606.
84. Chao YC, Chang MF, Gust I et al. Sequence conservation and divergence of hepatitis delta
virus RNA. Virology 1990; 178:384-392.
85. Chao YC, Lee CM, Tang HS et al. Molecular cloning and characterization of an isolate of
hepatitis delta virus from Taiwan. Hepatology 1991; 13:345-352.
86. Imazeki F, Omata M, Ohto M. Heterogeneity and evolution rates of delta virus RNA se-
quences. J Virol 1990; 64:5594-5599.
87. Imazeki F, Omata M, Ohto M. Complete nucleotide sequence of hepatitis delta virus RNA
in Japan. Nucleic Acids Res 1991; 19:5439.
88. Kuo MY, Goldberg J, Coates L et al. Molecular cloning of hepatitis delta virus RNA from
an infected woodchuck liver: Sequence, structure, and applications. J Virol 1988;
62:1855-1861.
89. Lee CM, Bih FY, Chao YC et al. Evolution of hepatitis delta virus RNA during chronic
infection. Virology 1992; 188:265-273.
90. Liu S, Yi Y, Cong X et al. Molecular cloning and sequencing of a hepatitis delta antigen-
coding cDNA from China. Ping Tu Hsueh Pao 1994; 9:15-22.
91. Lee CB, Lai YC, Ping YH et al. The importance of the helix 2 region for the cis-cleaving
and trans-cleaving activities of hepatitis delta virus ribozymes. Biochemistry 1996;
35:12303-12312.
Section II
Expression and Delivery of Ribozymes
CHAPTER 4
Exogenous Delivery of Ribozymes

Mark A. Reynolds
Introduction
H ammerhead ribozymes are trans-acting, RNA-based molecules that specifically cleave

target mRNA transcripts in the absence of protein-based cofactors or enzymes.1,2 Over
the past several years, considerable progress has been made in our laboratories toward the
development of these molecules as therapeutic agents.3 Chemically-modified (stabilized)
hammerhead ribozymes are highly resistant to nuclease degradation while retaining cata-
lytic cleavage rates that are similar to their parent, unmodified (all-ribo) versions.4 These
stabilized ribozymes specifically inhibit smooth muscle cell proliferation in culture,5 pre-
vent cytokine-stimulated (interleukin-1) expression of stromelysin mRNA in a rabbit-knee
osteoarthritis model6 and inhibit VEGF-stimulated angiogenesis in a rat cornea model.7
For the two in vivo studies just listed, a single administration of the ribozyme in a saline
vehicle permitted sufficient tissue trafficking and cellular uptake for activity to be observed
in a dose-dependent manner.
The ability to observe pharmacodynamic effects with free (saline) ribozyme formula-
tions in vivo is very important, since it provides us with a baseline for comparing alternative
delivery technologies. Although the field of exogenous ribozyme delivery is still relatively
new, there is a wealth of information pertaining to the delivery of antisense oligonucle-
otides and DNA that may be directly applicable to ribozymes. While this chapter will focus
primarily on the exogenous delivery of synthetic hammerhead ribozymes, many of the drug
delivery systems described below may also be applicable to plasmid-based ribozyme expres-
sion vectors.8-10
While local delivery of ribozymes is now emerging as clinically viable, sustained-re-
lease delivery systems offer the potential for enhanced potency and/or less frequent dosing.
Furthermore, the development of drug carriers for systemic delivery would bring this tech-
nology to a much greater number of human diseases, including cancer.
Tissue Culture Studies

The study of cellular uptake in tissue culture is probably not applicable to in vivo
pharmacokinetic modeling. For example, cell culture studies typically require drug carriers
such as cationic lipids for uptake and efficacy,11,12 whereas in vivo studies have shown that
antisense oligonucleotides13-19 and ribozymes6,7,20 are efficacious in the absence of any car-
rier. Nevertheless, cell culture studies are an important first step toward identifying lead
ribozymes against a given target sequence. Such studies have indicated that oligonucleotides
enter cells primarily via pinocytotic or endocytotic mechanisms.21-24 When fluorescent-tagged
oligonucleotides are used, they can be observed inside the cell in the form of punctate
Fig. 4.1. Schematic for the identification of lead cationic lipid formulations for ribozyme delivery
to a given cell type. A similar scheme can be used with other carriers (e.g., polycations, positively
charged nanoparticles, etc.).
(presumably endosomal) bodies. Since the targets for antisense oligonucleotides and
ribozymes reside in the cytoplasm and nucleus, many laboratories have focused on drug
carriers that can mediate endosome release.
In order to identify lead formulations for ribozyme delivery in tissue culture, we have
developed a screening method using fluorescent-tagged ribozymes that examines total cel-
lular uptake, subcellular trafficking and cytotoxicity (Fig. 4.1). Microinjection experiments
with fluorescent-tagged ribozyme show fairly rapid trafficking from the cytoplasm to the
nucleus of the cell (typically within about 5 min after injection, unpublished data). Similar
observations have been reported for antisense oligonucleotides.25,26 Thus, we use nuclear
fluorescence as a measure of the ability of a formulation to promote endosomal release.
Total cellular uptake can be estimated by fluorescence activated cell sorting (FACS) analysis,
using standardized fluorescent microbeads for calibration. This is a convenient method for
estimating the efflux of ribozymes over time following a single treatment (Fig. 4.2). Finally,
the cytotoxicity of the drug carrier/ribozyme complex is an important parameter that can
be conveniently measured by colorimetric metabolic activity analysis (e.g., MTS assay). We
find that drug carrier/ribozyme formulations can usually be optimized to reduce cytotoxic-
ity (e.g., by varying the charge ratios in the case of cationic carriers).
Cationic Lipids
Cationic lipids are amphiphiles that contain a positively charged head group, for bind-
ing polynucleotides through charge interactions, and a hydrophobic domain capable of as-
sociating with membranes or lipid bilayers. A considerable number of these compounds
have been described for the transfection of DNA plasmids,27-29 oligonucleotides11,12,30,31 and
ribozymes.5 Typically, cationic lipids are mixed with a fusogenic lipid such as dioleoyl phos-
phatidylethanolamine (DOPE), which is thought to promote endosome release.32 It ap-
pears that cationic lipid complexes must enter the endosomal pathway for fusion to occur,
and that binding to the cell surface is not sufficient for cytoplasmic release.33,34 Perhaps one
reason why so many different cationic lipids have been described is that lipids which work
well for one cell type do not always work for another. Panels of cationic lipids have been
described that enable the researcher to identify an optimal formulation for a given cell type.29
Exogenous Delivery of Ribozymes 43
Fig. 4.2. Ribozyme delivery to human aortic smooth muscle cells in culture. Cells were treated
with cationic lipid formulations as indicated, containing carboxyfluorescein-labeled
ribozyme, for 4 h in serum-free media. For the t = 4 h timepoint, the cells were washed and
assayed immediately for ribozyme internalization (based on fluorescence) using a FACScan
instrument (Bectin-Dickinson). For the t = 24 h timepoint, the cells were washed and then
incubated in serum-containing media for 20 h prior to analysis. The mean number of
ribozymes internalized per cell was determined using fluorescent microbead standards for
calibration (Molecular Probes, Inc.).
In our hands, the identification of lead cationic lipid reagents and the method by which
they are complexed with ribozymes are equally important. For example, the charge ratio
(+/-, molar ratio of cationic lipid to phosphodiester bonds) can dramatically affect the physi-
cal properties of the complex, as can the choice of cell culture media to use (Table 4.1).
Cationic lipid/ribozyme complexes have a tendency to aggregate in solution, particularly in
the presence of some types of cell culture media, which may adversely affect cellular uptake.
The charge ratio can also affect the ability of the complex to get taken up by cells and to
release the packaged ribozymes into the cytoplasm and nucleus (Fig. 4.3). Methods for char-
acterizing and optimizing cationic lipid formulations have been described.35 In particular,
it is important to determine the size distribution of the complexes over time, since this is a
measure of the tendency for a given lipid:ribozyme combination to aggregate during the
treatment regimen.
Polycations
Polycations that have been used for oligonucleotide and DNA plasmid delivery include
polylysines, polyethylenimine and polyamidoamine (PAMAM) dendrimers. These com-
pounds condense polynucleotides to form small particles that are taken up by cells either
through pinocytotic or endocytotic mechanisms.
Polylysines are known to promote cellular uptake, but lack an endosome-release mecha-
nism. Various ligands such as transferrin,36,37 glycosyl moieties38 and folate39,40 have been
conjugated to polylysine to promote receptor-mediated uptake. In this case, it is important
that the polylysine/oligonucleotide complexes are small (i.e., <120 nm) in order to enter
clathrin-coated pits on the cell surface. Folate-conjugated polylysine has been used to de-
liver ribozyme multimers with some success.40 Adenovirus41,42 and fusogenic peptides43 have
Table 4.1. Optimization of LipofectAMINE:ribozyme formulationsa
a) Effect of different culture mediab

T = 15 min T=2h
Media Mean Diam. (nm) Polydispersity Mean Diam. (nm) Polydispersity
IMDM 250 0.152 311 0.191
DMEM 639 0.392 3601 0.464
OptiMEM 694 0.432 3676 0.292
b) Effect of different charge ratiosc

Charge Ratio Mean Diam. (nm) Polydispersity
2:1 1503 0.364

4:1 154 0.127
6:1 107 0.110
8:1 106 0.158
a LipofectAMINE was purchased from Life Technologies Inc. A stabilized hammerhead ribozyme was
used having the sequence 5'-ususususcccu GAuGaggccgaaaggccGaaAuucucB-3', where lowercase

letters refer to 2'-O-methyl modified bases; capital letters refer to unmodified ribonucleotides;
s = phosphorothioate linkages; u = 2'-C-allyl modified uridine; B = inverted abasic residue. Samples
were analyzed in a Malvern Zetasizer 4 laser light scattering device.
b 20 µM lipid concentration, room temperature.
c IMDM media, 20 µM lipid concentration, T = 15 min.
been coupled with polylysine formulations to enhance endosome release. A combination of

receptor-ligands and fusogenic molecules may be required for optimal delivery. Major dis-
advantages of such formulations include their complexity and high cost.
Polyethylenimine (PEI) is an inexpensive reagent that has been shown to transfect plas-
mid DNA into cells with high efficiency.44 Since endosomes are known to be acidic, it has
been proposed that PEI becomes further protonated inside the endosome, resulting in an
osmotic pressure gradient that promotes endosome release. PAMAM dendrimers are well-
defined, uniform polymers that have also been shown to mediate DNA transfection and
deliver oligonucleotides in cell culture.45,46 We have tested both of these compounds with
fluorescently-labeled ribozymes according to the protocol outlined in Figure 4.1, and have
observed high levels of nuclear fluorescence for each reagent (typically >70% of the cells in
a field for a given cell type, unpublished results).
pH-Sensitive Liposomes
A number of investigators have employed liposomes for the delivery of antisense oligo-
nucleotides47-52 and ribozymes53,54 in cell culture. Most of this work has been conducted
with small unilamellar vesicles (SUVs), which can be prepared in a size range sufficient for
cellular uptake via endocytosis (<200 nm). Two principal drawbacks to this approach are:
1. low encapsulation efficiencies (typically <20%); and
2. the absence of an endosome-release mechanism.
Thus, the majority of internalized liposomes deliver their nucleic acid payload to the
lysosome, rather than enabling release into the cytoplasm and nucleus.51 pH-sensitive lipo-
somes were developed to enhance endosome release by incorporating lipids with protonatable
Fig. 4.3. Sub-cellular localization of internalized ribozymes. HeLa cells were treated with
cationic lipid formulations, containing carboxyfluorescein-labeled ribozyme, for 2 h in se-
rum-free media. The cells were washed and then examined by epifluorescence microscopy
using a Nikon N200 microscope (40x objective) hooked up to a Hitachi HV-C20 CCD cam-
era. The data was stored and processed using Adobe Photoshop.
head groups (e.g., oleic acid51 and cholesterylhemisuccinate, or CHEMS55) together with a
fusogenic lipid such as phosphatidylethanolamine. This approach enhances the cellular
uptake and efficacy of antisense oligonucleotides in tissue culture.48-50,52,53,55-61 Such studies
have not yet been reported for ribozyme delivery.
Devices
Electroporation has been used in selected cases to deliver DNA plasmids and oligo-
nucleotides into cells in culture.62,63 A biolistic device (Gene Gun, Bio-Rad Corporation)
has recently been made available for research applications. When fully optimized, such de-
vices could eventually offer some advantages compared to cationic lipids, e.g., by providing
more general delivery protocols that can be applied to a variety of different cell types. At the
present time, however, their utility is somewhat limited, since it is very difficult to achieve
high levels of delivery without considerable cell death.
Localized Delivery In Vivo

There is considerable evidence in the literature that antisense oligonucleotides are ac-
tive in vivo when administered as free (saline vehicle) formulations. Studies conducted in
our laboratories have shown that stabilized hammerhead ribozymes are also active when
administered locally as saline solutions. For example, a 2'-O-methyl modified ribozyme
containing 2'-amino substitutions at U4 and U7 and five unmodified ribonucleotides was
designed against stromelysin mRNA and evaluated in a rabbit knee arthritis model.6
Intraarticular injections gave a dose-dependent inhibition of new stromelysin mRNA syn-
thesis, stimulated by introduction of IL-1. A similar ribozyme targeted to the flt-1 VEGF
Fig. 4.4. Structure of a stabilized hammerhead ribozyme used in our laboratory for
biodistribution experiments. As indicated in the figure, the ribozyme can be modified with
either (a) an internal [32P]-label (contains a 2'-O-methyl A residue in place of S); or (b) a
fluorescent-label (non-radioactive).
receptor inhibited VEGF-stimulated neovascularization in a rat corneal implant model.7 In

this case, the ribozyme was delivered via an intracorneal implant and also by sub-conjuncti-
val injection into the limbus. In one report from another laboratory, a 2'-O-allyl-modified
hammerhead ribozyme containing five unmodified ribonucleotides showed good inhibi-
tion of amelogenin expression following a single injection into the jaw of newborn mice.20
These data suggest that the rate limiting steps for cellular uptake and trafficking of
ribozymes in vivo are different from those observed in tissue culture. We have employed
radiolabeled and fluorescently-labeled ribozymes to measure biodistribution and cellular
uptake in vivo (Fig. 4.4). For radiolabeling, we typically use a kinase/ligation reaction that
inserts an internal 32P-phosphodiester bond in the Stem II region.6 For fluorescent-labeling,
we insert an aminolinker-modified base in the Stem II region and couple with N-
hydroxysuccinimide-activated fluorescent probes (Molecular Probes, Inc.). Our data indi-
cate that hammerhead ribozymes can transport into tissue and enter cells following local-
ized administration as free (saline) solutions.
Trafficking of Free Ribozyme After Localized Administration
Intraarticular
The biodistribution of [32P]-labeled ribozyme following a single intraarticular injec-
tion in rabbit knees has been described.6 Tissues were harvested at 4 and 24 h post-admin-
istration in one experiment and at 1,3 and 7 d post-administration in a second experiment.
Synovial tissue was harvested and the amounts of total ribozyme accumulation were deter-
mined by scintillation counting. The percentage of intact (non-catabolized) ribozyme was
determined by polyacrylamide gel electrophoresis. Peak concentrations of ribozyme accu-
mulation were observed at 24 h, with approximately a 4-fold drop in synovial radioactivity
at 7 d. The 4 h samples showed fully intact ribozyme (no ribozyme catabolites were ob-
served); and the 24 h samples showed 80-90% intact ribozyme. Autoradiographic analysis
of synovial tissue sections taken at 24 h indicated most of the ribozyme to be intracellular.
This was confirmed subsequently using tetramethylrhodamine-labeled ribozyme (Fig. 4.5).
Epifluorescence images showed most of the intracellular fluorescence to be in the form of
punctate (presumably endosomal) bodies at 4 h, with more diffuse cytoplasmic fluores-
cence after 24 h. This data indicates that endosomal release of ribozymes in vivo may occur
more rapidly than is typically observed in cell culture.
Intraocular
We have developed ribozymes targeting the VEGF receptors flt-1 and KDR as potential
anti-angiogenic agents. A rat cornea implant model was developed to demonstrate the anti-
angiogenic potential of these ribozymes.7 A tetramethylrhodamine-labeled ribozyme was
used to observe trafficking from the implant toward the targeted vascular endothelial cells
of the pericorneal plexus. As shown in Figure 4.6, the ribozyme can indeed passage from the
implant toward the targeted tissue and enter cells surrounding blood vessels. We conclude
that tissue distribution and cellular uptake are not rate-limiting for hammerhead ribozymes
in this model.
Delivery Systems for Localized Administration

Although very limited data is available at this time, it appears that ribozymes can traffic
into tissue and enter target cells when administered locally. Drug delivery systems can po-
tentially enhance ribozyme efficacy following localized administration. Such systems pro-
tect the encapsulated ribozyme from catabolic breakdown and provide sustained-release
into surrounding tissues. The selection of lead drug delivery systems for preclinical devel-
opment depends on the indication and route of administration. Again, biodistribution studies
can be used for this screening process using fluorescent and/or radiolabeled ribozymes,
provided that an adequate animal model is available.
Biodegradable polymers
Biodegradable polymers have been employed extensively for protein and peptide de-
livery.64 Poly(L-lactic acid) (PLA) matrices have been shown to provide sustained-release
for antisense oligonucleotides.65 These materials also protect the encapsulated oligonucle-
otide from degradation in the presence of serum. Microbeads of PLA and poly(L-lactic-co-
glycolic) acid (PLGA) can be prepared for localized administrations via a variety of differ-
ent routes (e.g., intraarticular, intraocular, intratumoral, subcutaneous, etc.).
In our laboratory, we have found that modified hyaluronic acid and carboxymethylcel-
lulose (Sepragel Bioresorbable Gel, Genzyme Corporation) is well-suited for intraarticular
delivery of ribozymes.66 Radiolabeled ribozyme was formulated in Sepragel and tested for
biodistribution in the hind legs of male New Zealand rabbits following a single intraarticular
injection. In vitro release studies showed an initial burst of ribozyme release over the first
several hours (8-15%), followed by a more linear sustained release over the next three days
(20-45%). A comparison of ribozyme accumulation in synovial tissue for free (saline) ver-
sus Sepragel formulation is shown in Table 4.2.
Nanoparticles
Polyalkylcyanoacrylate nanoparticles have been described for the delivery of antisense
oligonuclotides via subcutaneous injection.67 The oligonucleotides are adsorbed onto the
surface of these particles via ionic interactions. The resulting complexes offer significant
protection from nuclease degradation, even for unmodified oligonucleotides. 68
Biodistribution studies have shown that these particles rapidly accumulate in the liver
Fig. 4.5. Biodistribution of tetramethylrhodamine-labeled ribozyme in the

rabbit knee synovium. Female New Zealand rabbits were treated with the
ribozyme (50 µM in saline vehicle) by intraarticular injection. At the times
indicated, the tissue was harvested and analyzed by epifluorescence mi-
croscopy. The intraarticular space is shown at the top right. Internalized
ribozyme occurs in synoviocytes and is shown to passage below the sur-
face layer and around fat cells. At t = 4 h, the intracellular fluorescence is
predominantly punctate (presumably endosomes and lysosomes); at
t = 24 h the fluorescence is more diffuse (presumably cytoplasmic).
Fig. 4.6. Biodistribution of

tetramethylrhodamine-la-
beled ribozyme in the rat
cornea. A sterile nitrocel-
lulose filter disc (0.5 mm
diameter) was soaked in a
solution containing the
ribozyme and then surgi-
cally implanted into the
cornea of a male Sprague-
Dawley rat, approximately
3 mm from the pericor-
neal plexus. The tissue was
then examined by epi-
fluorescence microscopy
approximately 24 post-
implantation.
following intravenous administration in mice.69 Thus, these drug carriers may be less well-
suited for systemic delivery than for local delivery.
A novel type of nanoparticle that gives high encapsulation of oligonucleotides has re-
cently been described, the SupraMolecular BioVector (SMBV).70 These particles contain a
positively charged polysaccharide core surrounded by a lipid bilayer consisting of phos-
phatidylcholine and cholesterol. Encapsulation is achieved by incubating the oligonuceotide
with SMBV approximately 10°C below the phase-transition temperature of the outer mem-
brane bilayer. SMBV nanoparticles protect oligonucleotides from nuclease degradation and
enhance cell uptake in tissue culture.
Iontophoresis
Iontophoretic devices are available through a variety of different manufacturers for
topical delivery of small molecules, peptides and proteins.71-73 One such device is also being
developed for gene transfer (E-TRANS, Alza Corporation).
A recent study showed that a 2'-O-methyl modified hammerhead ribozyme contain-
ing a 2'-C-allyl modification at U4 and five unmodified ribonucleotides can be delivered
locally into pig coronary arteries using an iontophoretic porous balloon catheter (e-Med,
Table 4.2. Biodistribution of [32P]-labeled stabilized hammerhead ribozyme to

rabbit knee synovium following intraarticular administration of saline or
Sepragel vehiclea
Formulation Tissue # Knees Ribozyme delivered

(ng/mg tissue)*
T = 24 Hrs
Saline Synovium 8 146 ± 51
Sepragel (Med) Synovium 8 266 ± 79
T = 72 Hrs
Saline Synovium 6 82 ± 29
Sepragel (Med) Synovium 6 190 ± 54
a Male New Zealand White Rabbits (3-4 kg) were anesthetized with ketamine-HCl and injected
intraarticularly with about 200 x 106 cpm of [32P]-internally labeled ribozyme together with unlabeled
ribozyme corresponding to a total dose of 100 µg in a volume of 250 µl. At the indicated timepoints,
tissues were harvested and total ratioactivity was determined by scintillation counting.
Inc.).74 [3H]-labeled ribozyme was prepared by exchange with tritiated water.75 Arteries were
treated according to the iontophoretic catheter manufacturer’s recommendations and then
sacrificed at various timepoints. Tissues were harvested and the amount of tritium accumu-
lation in artery tissue was determined by scintillation counting (Table 4.3). Tetramethyl-
rhodamine-labeled ribozyme was delivered similarly and shown to occur intracellularly based
on epifluorescence and confocal microscopy measurements (data not shown).
Systemic Delivery
To date, there are no published pharmacodynamic studies of stabilized hammerhead
ribozymes administered systemically. However, recent advances in large-scale oligonucle-
otide synthesis have made it possible to initiate such experiments. It is anticipated that this
work will eventually enable stabilized ribozymes for the treatment of systemic diseases, in-
cluding inflammatory diseases and cancers.
Plasma Pharmacokinetics of Stabilized Hammerhead Ribozymes

Plasma pharmacokinetic parameters were recently described in a rat model for a stabi-
lized hammerhead ribozyme, containing 2'-O-allyl ribonucleotides except for eight unmodi-
fied ribonucleotides in the catalytic core.76 5'-carboxyfluorescein-labeled ribozyme was ad-
ministered via a single tail vein injection at approximately 1.25 mg/kg. Intact ribozyme and
ribozyme catabolites were quantitated after tissue digestion and processing by polyacryla-
mide gel electrophoresis. The tissue biodistribution data showed almost complete endo-
nucleolytic cleavage after 48 h, presumably within the unmodified GAAA region. A biphasic
plasma clearance profile was observed, with a distribution half life of 12 min and an elimi-
nation half life of 6.5 h.
Plasma pharmacokinetic studies have been conducted in our laboratory with a 2'-O-
methyl modified hammerhead ribozyme containing a 2'-C-allyl sugar at the U4 position
and five unmodified ribonucleotides in the catalytic core.77 Plasma clearance profiles have
been determined in mice following single bolus intravenous doses of up to 30 mg/kg. The
plasma elimination half life in the mouse for this modified ribozyme is around 30 min. The
major routes of elimination appear to be via renal filtration and catabolism. Thus, although
Table 4.3. Iontophoretic Delivery of [3H]-labeled stabilized ribozyme to pig

coronary arteriesa
Rz Remaining in the Tissue (per Wet Tissue Weight)

Artery 30 min 1 day 3 days
LAD 48 ng/mg 4.4 ng/mg 2.2 ng/mg

LCX Untreated 3.2 ng/mg 2.4 ng/mg
RCA 37 ng/mg Untreated Untreated
a Juvenile domestic pigs (~25 kg) were anesthetized and the coronary arteries were instrumented under aseptic
conditions via a femoral artery approach, under fluoroscopic guidance. Immediately after balloon angioplasty
in each epicardial arterial segment the balloon catheter was removed and exchanged for the iontophoretic
porous balloon catheter (e-Med, Inc.). The device was used according to the manufacturer’s specifications to
deliver the test oligonucleotide to each specified artery. One of the three arteries per heart (LAD, LCX or RCA)
was left untreated for use as a control. Tissues were harvested at the times indicated and total radioactivity was
determined by scintillation counting.
chemically modified ribozymes are highly stable in vitro in the presence of human serum,
they are catabolized in vivo to some extent.
In both studies, stabilized ribozymes appear to be eliminated from plasma at much
faster rates compared to phosphorothioate (PS) modified antisense oligonucleotides. Rat
plasma half lives for PS oligonucleotides have been determined to be in the range of
34-53 h.78-80 PS oligonucleotides are known to bind plasma proteins,81 which presumably
leads to longer plasma clearance profiles and protection from nuclease degradation. It is
possible that ribozymes bind plasma proteins to a lesser extent and/or are more exposed to
catabolic enzymes in plasma.
Drug Delivery Systems for the Systemic Administration of Ribozymes

A variety of approaches currently being investigated for the systemic delivery of antisense
oligonucleotides may also be applicable to hammerhead ribozymes. Additionally, technolo-
gies being studied for non-viral gene transfer may also be suitable for ribozyme expression
vectors. While many different routes of administration have been considered (i.e., intra-
peritoneal, oral, pulmonary, intravenous), most of our studies have been done via the intra-
venous route, since this gives complete and nearly instantaneous plasma exposure.
Bioconjugates
One potential method for increasing plasma circulation half life of a ribozyme is to
increase its lipophilicity. For example, aliphatic moieties such as cholesterol have been con-
jugated to PS oligonucleotides and examined for biodistribution in the mouse.82 Such modi-
fications promote binding to low-density lipoproteins in the plasma, thereby preventing
renal filtration and enhancing delivery to certain target organs such as the liver.83 In one
study, conjugation of cholesterol to an antisense PS oligonucleotide enhanced its biological
activity in vivo.84 One of the potential drawbacks to this approach is that the lipophilic
moiety can adversely affect intracellular trafficking, causing the oligonucleotide to become
sequestered in membrane compartments and preventing it from finding its intended mRNA
target. Cleavable linkers have been proposed to release the oligonucleotide from its lipophilic
carrier once inside the cell. For example, disulfide bonds can be cleaved after intracellular
delivery with disulfide reductases or glutathione.85 Biodegradable ester bonds have also been
employed for release of a lipophilic 5'-palmitoyl moiety from an antisense oligonucleotide.86
Cationic lipids
Cationic lipids have been studied for systemic gene transfer.31,50,51,87-91 As in the case of
tissue culture studies, these compounds are thought to enhance transfection by promoting
cell uptake and endosome release. However, blood components such as the complement
system can associate with (opsonize) cationic lipid/DNA complexes, leading to rapid clear-
ance by the mononuclear phagocytotic system (MPS).92 Thus, many studies have shown
high levels of exogenous gene expression in the lung, liver and spleen, yet it is unclear whether
these genes are capable of reaching their intended target cells (e.g., Kupfer cells versus hepa-
tocytes in the liver). The detailed physical analysis and optimization of cationic lipid/DNA
complexes is leading toward improved transfection efficiencies, particularly for pulmonary
delivery, (e.g., for the treatment of cystic fibrosis93). Nevertheless, further improvements
will be useful to provide cationic lipid formulations that can be more effectively adminis-
tered systemically and accumulate in target tissues such as metastatic cancers.
We have examined a simple cationic lipid formulation of 1,2-dioleoyl-3-trimethyl-
ammonium propane (DOTAP) with a 2'-O-methyl-modified hammerhead ribozyme con-
taining a 2'-C-allyl modification at U4 and five unmodified ribonuclesides in the catalytic
core.94 One-to-one charge ratio formulations (positively charged lipid to phosphodiester
bond) were made with [32P]-labeled ribozyme and administered to BALB/c mice via a tail
vein injection. Major organs were dissected at various timepoints, digested and quantitated
by liquid scintillation counting. Aliquots were also analyzed by polyacrylamide gel electro-
phoresis to determine intact ribozyme and ribozyme catabolites. Representative data are
shown in Figure 4.7. Unlike plasmid biodistribution data that is based on expression, these
data indicate directly the tissue exposure of the ribozyme. Thus, cationic lipid formulations
are capable of enhancing the tissue biodistribution of stabilized hammerhead ribozymes
when administered systemically. However, ribozyme catabolism is observed in the tissues,
although more rapid catabolism is seen in animals treated with the ribozyme in saline ve-
hicle alone (no drug carrier). The plasma clearance profile for this DOTAP/ribozyme com-
plex is shown in Figure 4.8. As anticipated, plasma clearance is fairly rapid, consistent with
the tendency for cationic lipid/polynucleotide particles to become opsonized and removed
by MPS tissues.
Long-circulating liposomes
Surface-modified (long-circulating) liposomes have emerged as powerful tools for
modulating the pharmacokinetic profiles, resulting in increased therapeutic windows and/
or new label indications for existing drugs such as doxorubicin (e.g., Kaposi’s sarcoma) and
amphotericin B (e.g., systemic fungal infections).95-98 These modified liposomes are resis-
tant to opsonization, thereby enabling them to circulate for long periods of time in the
blood. This property enhances the ability of the encapsulated drug to get delivered to its
intended target tissue. In particular, long-circulating liposomes have been found to selec-
tively accumulate by extravasation into solid tumor tissue.99-101 Thus, long-circulating lipo-
somes may be particularly well-suited for use with novel, cytostatic antitumor drugs that
are designed to inhibit angiogenesis.
Since the early 1970s, biodistribution studies of conventional liposomes following in-
travenous administration have shown rapid elimination from plasma into MPS tissues.102
The various types of modifications that have been applied to enhance liposome circulation
times have been reviewed elsewhere.103,104 Examples include the incorporation of monosialo-
ganglioside (GM1), sphingomyelin (SM) and phosphatidylinositol (PI). More recently, the
Fig. 4.7. Biodistribution of [32P]-labeled ribozyme in Balb/c mice. A DOTAP-

ribozyme formulation (1:1 charge ratio, containing 1-5 x 106 cpm of [32P]-
labeled ribozyme, 3 µmole total lipid dose) was administered via a single bo-
lus tail vein injection. At the times indicated, animals were euthanized by CO2
asphyxiation and then perfused with normal saline through the heart until the
liver was cleared of blood (5-10 ml). The tissue samples were then pulverized
or homogenized and digested with proteinase K-containing buffer (0.5% SDS/
100 mM NaCl/10 mM Tris (pH 8.0)/25 mM EDTA/135 µg/ml proteinase K) at
50°C for 1 hour. Aliquots were removed and assayed for radioactivity using a
Packard Model 2500 scintillation analyzer. Then, an equal volume of gel load-
ing buffer (95% formamide/0.05% bromophenol blue/20 mM EDTA) was
added and samples were electrophoresed on a 20% polyacrylamide/7 M
urea/1 x TBE gel. Bands corresponding to intact ribozyme and ribozyme
catabolites were detected and quantitated using a Molecular Dynamics Model
425E PhosphorImager.
incorporation of polymer-modified lipids have been used, particularly poly(ethylene gly-

col) (or PEG). PEG-modified lipids have been extensively studied in numerous liposome
formulations and are now commercially available for research use. Theoretical calculations
have shown that PEG polymers form a “brush” or steric barrier on the surfaces of lipo-
somes.103,105 PEG-modified liposomes are resistant to protein binding, and thereby are re-
sistant to opsonization and removal from the blood by RES tissues.
The application of long-circulating liposome technology to gene transfer is currently
being investigated in a number of laboratories, although few accounts have been described
in the literature. Considerations for optimal delivery include methods to enhance encapsu-
lation. Encapsulation efficiency is particularly important for ribozymes, which are still fairly
costly to synthesize.
We have investigated several prototype liposome formulations for their effects on the
biodistribution of stabilized hammerhead ribozymes in a murine Lewis lung carcinoma
model.106 [32P]-labeled ribozyme was encapsulated by the reverse-evaporation method,107
extruded to approximately 100 nm diameter using a Lipex extrusion device (Vancouver,
Fig. 4.8. Plasma distribution profiles for lipid-ribozyme formulations in Balb/c mice. DOTAP =
simple cationic lipid complex of DOTAP and ribozyme (1:1 charge ratio); DSPE-PEG2000 =
ribozyme encapsulated in liposomes containing egg yolk phosphatidylcholine, cholesterol, DOTAP
and 1,2-disteroyl-phosphatidylethanolamine-PEG2000. Animals received a single dose of ap-
proximately 1 mg/kg ribozyme (containing 1-5 x 106 cpm [32P]-labeled ribozyme, 3 µmol total
lipid) via tail vein injection. At the indicated timepoints, animals were euthanized by CO2 as-
phyxiation and blood was sampled from the heart. Total radioactivity was determined by scintil-
lation counting and intact ribozyme was determined by PAGE analysis as described in Figure 4.7.
Each data point represents the standard mean of n = 5 animals.
Canada), and purified by size-exclusion chromatography through Sepharose CL-4B. Intact

ribozyme was determined by polyacrylamide gel electrophoresis, as described above. Typi-
cal plasma profiles are given in Figure 4.8. The most favorable formulation, based on plasma
clearance data, was a liposome composition containing EYPC/Chol/DOTAP/DSPE-PEG2000.
The cationic lipid DOTAP was included to enhance ribozyme encapsulation. According to
our data, PEG-lipids are capable of masking the surface to protein binding, even in the
presence of cationic lipids, thereby permitting enhanced encapsulation and long-circulat-
ing properties to be achieved simultaneously.
Bioerodible polymers
In a recent study, bioerodible polymer microspheres were shown to offer an oral deliv-
ery route for plasmid DNA.108 The microbeads, consisting of poly(fumaric acid:sebacic acid),
poly(FA:SA), were formulated with a DNA plasmid containing a bacterial gene for β-galac-
tosidase. Rats fed the DNA-containing microspheres showed β-galactosidase expression in
the small intestine and liver. Micrographs showed that the microparticles can traverse through
the mucosal epithelium and follicle-associated epithelium. Such technologies may allow the
oral delivery of oligonucleotides and ribozymes.
Future Applications
The technologies just described are only a small subset of the available drug delivery
systems that may be applied to stabilized hammerhead ribozymes. It is anticipated that
more data will become available in the near future as the cost of ribozyme production de-
creases due to current efforts in large-scale synthesis. Some of these systems may also be
applied to stabilized ribozyme expression vectors.
Although the results from our preliminary pharmacokinetic studies are indicative of a
fairly rapid plasma clearance profile for stabilized hammerhead ribozymes, it is anticipated
that continuous infusion may lead to improved plasma and tissue biodistributions. Studies
are currently in progress using sustained-delivery devices for intravenous, intraperitoneal,
and intratracheal administration.
Targeted liposomes (immunoliposomes) are being developed for more selective deliv-
ery to diseased tissues.109 For example, sterically stabilized anti-HER2 immunoliposomes
have been designed for targeting to human breast cancers.110 While such approaches have
been shown to work in vitro, selective targeting in vivo remains to be demonstrated.
I am particularly intrigued by the possibility of applying long-circulating liposome
technologies to ribozyme delivery. Based on similar studies, it appears likely that long-cir-
culating liposomes may enhance ribozyme delivery to tumor tissue, and possibly to other
neovascularized tissues such as occur in various inflammatory diseases. One of the key is-
sues currently being addressed is ribozyme release, i.e., its ability to escape from the drug
carrier and traffic into surrounding tissue and cells. As illustrated above, we have already
generated data indicating that ribozymes can enter cells in vivo in the absence of any drug
carrier. The utility of a drug carrier for ribozyme delivery relies on its ability to enhance
tissue exposure and also to promote cellular uptake either directly or indirectly through
sustained release. There is some evidence that liposomes are more rapidly degraded in tu-
mor tissue compared to normal tissue, resulting in greater drug release in the former case.111
Thus, long-circulating liposomes may be particularly well suited for exogenous ribozyme
delivery in treating disseminated cancers. We are currently investigating this possibility in
our laboratory using fluorescent-labeled ribozymes and histological examination with con-
focal microscopy.
References
1. Usman N, Beigelman L, McSwiggen JA. Hammerhead ribozyme engineering. Curr Opin
Struct Biol 1996; 6:527-533.
2. Usman N, Stinchcomb DT. Design, synthesis, and function of therapeutic hammerhead
ribozymes. Nucleic Acids Mol Biol 1996; 10:243-264.
3. Christoffersen RE, Marr JJ. Ribozymes as human therapeutic agents. J Med Chem 1995;
38(12):2023-37.
4. Beigelman L, McSwiggen JA, Draper KG et al. Chemical modification of hammerhead
ribozymes. J Biol Chem 1995; 27:25702-25708.
5. Jarvis TC, Alby LJ, Beaudry AA et al. Inhibition of vascular smooth muscle cell prolifera-
tion by ribozymes that cleave c-myb mRNA. RNA 1996; 2:419-428.
6. Flory CM, Pavco PA, Jarvis TC et al. Nuclease-resistant ribozymes decrease stromelysin
mRNA levels in rabbit synovium following exogenous delivery to the knee joint. Proc Natl
Acad Sci USA 1996; 93(2):754-8.
7. Cushman C, Escobedo J, Parry TJ et al. Ribozyme inhibition of VEGF-mediated endothe-
lial cell proliferation in cell culture and VEGF-induced angiogenesis in a rat corneal model.
Angiogenesis Inhibitors and other novel therapeutic strategies for Ocular Diseases of
Neovascularization. Boston, MA: IBC USA Conferences, 1996.
8. Thompson JD, Macejak D, Couture L, Stinchcomb DT. Ribozymes in gene therapy. Na-
ture Med 1995; 1(3):277-8.
9. Ohkawa J, Taira K. Possibility for gene therapy by ribozyme RNA. Igaku no Ayumi 1995;
175(9):638-41.
10. Wong-Staal F. Ribozyme gene therapy for HIV infection intracellular immunization of lym-
phocytes and CD34+ cells with an anti-HIV-1 ribozyme gene. Adv Drug Delivery Rev 1995;
17(3):363-8.
11. Bennett CF, Chiang M-Y, Chan H et al. Cationic lipids enhance cellular uptake and activ-
ity of phosphorothioate antisense oligonucleotides. Mol Pharmacol 1992; 41:1023-33.
12. Bennett CF, Chiang M-Y, Chan H, Grimm S. Use of cationic lipids to enhance the biologi-
cal activity of antisense oligonucleotides. J Liposome Res 1993; 3(1):85-102.
13. Dean NM, McKay R. Inhibition of protein kinase C-α expression in mice after systemic
administration of phosphorothioate antisense oligodeoxynucleotides. Proc Natl Acad Sci
USA 1994; 91:11762-11766.
14. Dean N, McKay R, Miraglia L et al. Inhibition of growth of human tumor cell lines in
nude mice by an antisense oligonucleotide inhibitor of protein kinase C-α expression.
Cancer Res 1996; 56:3499-3507.
15. Haller H, Dragun D, Miethke A et al. Antisense oligonucleotides for ICAM-1 attenuate
reperfusion injury and renal failure in the rat. Kidney Interntl 1996; 50:473-480.
16. Monia BP, Johnston JF, Geiger T, Muller M, Fabbro D. Antitumor activity of a
phosphorothioate antisense oligodeoxynucleotide targeted against C-raf kinase. Nature Med
1996; 2:668-675.
17. Nyce JW, Metzger WJ. DNA antisense therapy for asthma in an animal model. Nature
1997; 385:721-725.
18. Rubenstein M, Mirochnik Y, Chou P, Guinan P. Antisense oligonucleotide intralesional
therapy for human PC-3 prostate tumors carried by athymic nude mice. J Surgical Oncol
1996; 62:194-200.
19. Yazaki T, Ahmad S, Chahlavi A et al. Treatment of glioblastoma U-87 by systemic admin-
istration of an antisense protein kinase C-α phosphorothioate oligodeoxynucleotide. Mo-
lecular Pharmacol 1996; 50:236-242.
20. Lyngstadaas SP, Risnes S, Sproat BS, Thrane PS, Prydz HP. A synthetic, chemically stabi-
lized modified ribozyme eliminates amelogenin, the major translation product in develop-
ing mouse enamel in vivo. EMBO J 1995; 14:5224-5229.
21. Nakai D, Seita T, Tetsuya T et al. Cellular uptake mechanism for oligonucleotides: In-
volvement of endocytosis in the uptake of phosphodiester oligonucleotides by a human
colorectal adenocarcinoma cell line, HCT-15. J Pharmacol and Exp Ther 1996; 278:
1362-1372.
22. Beltinger C, Saragovi HU, Smith RM et al. Binding, uptake, and intracellular trafficking of
phosphorothioate-modified oligodeoxynucleotides. J Clin Invest 1995; 95:1814-1823.
23. Shoji Y, Akhtar S, Periasamy A, Herman B, Juliano RL. Mechanism of cellular uptake of
modified oligodeoxynucleotides containing methylphosphonate linkages. Nucleic Acids Res
1991; 19(20):5543-5550.
24. Stein CY, Tonkinson JL, Zhang L-M et al. Dynamics of the internalization of phosphodiester
oligodeoxynucleotides in HL60 cells. Biochemistry 1993; 32:4855-4861.
25. Fisher TL, Terhorst T, Cao X, Wagner RW. Intracellular disposition and metabolism of
fluorescently-labeled unmodified oligonucleotides microinjected into mammalian cells.
26. Leonetti JP, Mechit N, Degols G, Gagnor C, Lebleu B. Intracellular distribution of micro-
injected AS oligonucleotides. Proc Natl Acad Sci USA 1991; 88:2702-2706.
27. Remy J-S, Sirlin C, Vierling P, Behr J-P. Gene transfer with a series of lipophilic DNA-
binding molecules. Bioconjugate Chem 1994; 5:647-654.
28. van der Woude I, Wagenaar A, Meekel AAP et al. Novel pyridinium surfactants for effi-
cient, nontoxic in vitro gene delivery. Proc Natl Acad Sci USA 1997; 94:1160-1165.
29. Griffiths T, Russell M, Froning K et al. The PerFect(TM) lipid optimizer kit for maximiz-
ing lipid-mediated transfection of eukaryotic cells. BioTechniques 1997; 22(5):982-988.
30. Capaccioli S, DiPasqual G, Mini E, Mazzei T, Quattrone A. Cationic lipids improve antisense
oligonucleotide uptake and prevent degradation in cultured calls and in human serum.
Biochem Biophys Res Commun 1993; 197:818-25.
31. Lappalainen K, Urtti A, Jääkeläinen I, Syrjänen K, Syrjänen S. Cationic liposomes medi-
ated delivery of antisense oligonucleotides targeted to HPV 16 E7 mRNA in CaSki cells.
Antiviral Res 1994; 23:119-30.
32. Farhood H, Serbina N, Huang L. The role of dioleoyl phosphatidylethanolamine in cat-

ionic liposome mediated gene transfer. Biochim Biophys Acta 1995; 1235:289-295.
33. Zabner J, Fasbender AJ, Moninger T, Poellinger KA, Welsh MJ. Cellular and molecular
barriers to gene transfer by a cationic lipid. J Biol Chem 1995; 270:18997-19007.
34. Wrobel I, Collins D. Fusion of cationic liposomes with mammalian cells occurs after en-
docytosis. Biochim Biophys Acta 1995; 1995:296-304.
35. Eastman S, Siegel C, Tousignant J, Smith A, Cheng S, Scheule R. Biophysical characteriza-
tion of cationic lipid: DNA complexes. Biochim Biophys Acta 1997; 1325:41-62.
36. Wagner E, Cotten M, Mechtler K, Kirlappos H, Birnstiel ML. DNA-binding transferrin
conjugates as functional gene-delivery agents: Synthesis by linkage of polylysine or ethidium
homodimer to the transferrin carbohydrate moiety. Bioconjugate Chem 1991; 2:226.
37. Wagner E, Cotten M, Foisner R, Birnstiel ML. Transferrin-polycation-DNA complexes: The
effect of polycations on the structure of the complex and DNA delivery to cells. Proc Natl
Acad Sci USA 1991; 88:4255-4259.
38. Erbacher P, Roche A, Monsigny M, Midoux P. glycosylated polylysine/DNA complexes:
Gene transfer efficiency in relation with the size and the sugar substitution level of
glycosylated polylysines and with the plasmid size. Bioconjugate Chem 1995; 6:401-410.
39. Citro G, Szcylik C, Ginobbi P, Zupi G, Calabretta B. Inhibition of leukaemia cell prolifera-
tion by folic acid-polylysine-mediated introduction of c-myb antisense oligodeoxynucleotides
into HL-60 cells. Proc Natl Acad Sci USA 1993:463-467.
40. Leopold LH, Shore SK, Newkirk TA, Reddy RMV, Reddy EP. Multi-unit ribozyme-medi-
ated cleavage of bcr-abl mRNA in myeloid leukemias. Blood 1995; 85(8):2162-70.
41. Curiel DT, Agarwal S, Wagner E, Cotten M. Adenovirus enhancement of transferrin-polyl-
ysine-mediated gene delivery. Proc Natl Acad Sci USA 1991; 88:8850-4.
42. Wagner E, Zatloukal K, Cotten M et al. Coupling of adenovirus to transferrin-polylysine/
DNA complexes greatly enhances receptor-mediated gene delivery and expression of trans-
fected genes. Proc Natl Acad Sci USA 1992; 89:6099-6103.
43. Plank C, Oberhauser B, Mechtler K, Koch C, Wagner E. The influence of endosome-dis-
ruptive peptides on gene transfer using synthetic virus-like gene transfer systems. J Biol
Chem 1994; 269:12918-12924.
44. Boussif O, Lezoualc’h F, Zanta MA et al. A versatile vector for gene and oligonucleotide
transfer into cells in culture and in vivo: Polyethyenimine. Proc Natl Acad Sci USA 1995;
92:7297-7301.
45. Bielinska A, Kukowska-Latallo JF, Johnson J, Tomalia DA, Baker JRJ. Regulation of in
vitro gene expression using antisense oligonucleotides or antisense expression plasmids trans-
fected using starburst PAMAM dendrimers. Nucleic Acids Res 1996; 24:2176-2182.
46. Haensler J, Szoka FCJ. Polyamidoamine cascade polymers mediate efficient transfection of
cells in culture. Bioconjugate Chem 1993; 4:372-379.
47. Leonetti JP, Machy P, Degols G, Lebleu B, Leserman L. Antibody-targeted liposomes con-
taining oligodeoxyribonucleotides complementary to viral RNA selectively inhibit viral rep-
lication. Proc Natl Acad Sci USA 1990; 87:2448-51.
48. Ropert C, Malvy C, Couvreur P. Inhibition of the Friend retrovirus by antisense oligo-
nucleotides encapsulated in liposomes: Mechanism of action. Pharm Res 1993; 10:1427-33.
49. Zelphati O, Zon G, Leserman L. Inhibition of HIV-1 Replication in cultured cells with
antisense oligonucleotides encapsulated in immunoliposomes. Antisense Res Dev 1993;
3:323-38.
50. Thierry AR, Dritshilo A. Intracellular availability of unmodified, phosphorothioated and
liposomally encapsulated oligodeoxynucleotides for antisense activity. Nucleic Acids Res
1992; 20(21):5691-5698.
51. Staubinger RM, Papahadjopoulos D. Liposomes as carriers for intracellular delivery of
nucleic acids. Methods Enzymol 1983; 101:512-527.
52. Thierry AR, Rahman A, Dritshilo A. Liposomal delivery as a new approach to transport of
antisense oligonucleotides. In: Izant RP, ed. Gene Regulation: Biology of Antisense RNA
and DNA. New York: Raven Press, Ltd., 1992:147-161.
53. Chonn A, Cullis PR. Recent advances in liposomal drug-delivery systems. Current Opin-
ion Biotechnol 1995; 6:698-708.
54. Sullivan SM. Liposome-mediated uptake of ribozymes. METHODS: A Companion to Meth-

ods in Enzymology 1993; 5:61-66.
55. Chu C-J, Dijkstra J, Lai M-Z, Hong K, Szoka F. Efficiency of cytoplasmic delivery by pH-
sensitive liposomes to cells in culture. Pharm Res 1990; 7(8):824-834.
56. Connor J, Yatvin M, Huang L. pH-sensitive liposomes: Acid-induced liposome fusion. Proc
Natl Acad Sci USA 1984; 81:1715-1718.
57. Collins D, Litzinger D, Huang L. Structural and functional comparisons of pH-sensitive
liposomes composed of phosphatidylethanolamine and three different diacylsuccinyl-
glycerols. Biochim Biophys Acta 1989; 1025:234-242.
58. Ropert S, Lavignon M, Dubernet C, Couvreur P, Malvy C. Oligonucleotides encapsulated
in pH sensitive liposomes are efficient toward Friend retrovirus. Biochem Biophys Res
Commun 1992; 183:879-85.
59. Straubinger R. pH-Sensitive liposomes for delivery of macromolecules into cytoplasm of
cultured cells. Methods Enzymol 1993; 221:361-376.
60. Wang S, Lee RJ, Cauchon G, Gorenstein DG, Low PS. Delivery of antisense oligonucle-
otides against the human epidermal growth factor receptor into cultured KB cells with
liposomes conjugated to folate via polyethylene glycol. Proc Natl Acad Sci USA 1995;
92:3318-3322.
61. Loke SL, Stein C, Zhang X, Avigan M, Cohen J, Neckers LM. Delivery of c-myc antisense
phosphorothioate oligodeoxynucleotides to hematopoietic cells in culture by liposome fu-
sion: Specific reduction in c-myc protein expression correlates with inhibition of cell growth
and DNA synthesis. Current Topics in Microbiol and Immunol 1988; 141:282-289.
62. Bergan R, Connell Y, Fahmy B, Neckers L. Electroporation enhances c-myc antisense
oligodeoxynucleotide efficacy. Nucleic Acids Res 1993; 21:3567-3573.
63. Rabinowitz J, Magnuson T. Independent gene targeting by coelectroporation of multiple
vectors. Anal Biochem 1995; 228:180-182.
64. Gombotz W, Pettit D. Biodegradable polymers for protein and peptide drug delivery.
Bioconjugate Chem 1995; 6:332-351.
65. Lewis KJ, Irwin WJ, Akhtar S. Biodegradable poly(L-lactic acid) matrices for the sustained
delivery of antisense oligonucleotides. J Controlled Release 1995; 37:173-183.
66. Min J, Bouhana K, Jensen K et al. Sepragel(TM) enhances the biodistribution of hammer-
head ribozymes in a rabbit knee model of osteoarthritis. 1996 (manuscript in prepara-
tion).
67. Schwab B, Chavany C, Duroux I et al. Antisense oligonucleotides adsorbed to polyalkyl-
cyanoacrylate nanoparticles specifically inhibit mutated Ha-ras-mediated cell proliferation
and tumorigenicity in nude mice. Proc Natl Acad Sci USA 1994; 91:10460-10464.
68. Chavany C, Saison-Behmoaras T, Le Doan T et al. Adsorption of oligonucleotides onto
polyisohexylcyanoacrylate nanoparticles protects them against nucleases and increases their
cellular uptake. Pharm Res 1994; 11:1370-1378.
69. Nakada Y, Fattal E, Foulquier M, Couvreur P. Pharmacokinetics and biodistribution of
oligonucleotides adsorbed onto poly(isobutylcyanoacrylate) nanoparticles after intravenous
administration in mice. Pharm Res 1996; 13:38-43.
70. Berton M, Sixou S, Kravtzoff R et al. Improved oligonucleotide uptake and stability by a
new drug carrier, the SupraMolecular Bio Vector (SMBV). Biochim Biophys Acta 1997;
1355:7-19.
71. Green PG. Iontophoretic delivery of peptide drugs. J Controlled Release 1996; 41:33-48.
72. Sage BH Jr. Insulin iontophoresis. Pharm Biotechnol 1997; 10:319-341.
73. Singh J, Bhatia KS. Topical iontophoretic drug delivery: pathways, principles, factors, and
skin irritation. Med Res Rev 1996; 16:285-296.
74. Robinson KA, Chronos NAF, Kleinman ME et al. Local delivery of c-myb ribozymes to
coronary arteries of pigs using an iontophoretic catheter. 1997 (manuscript in prepara-
tion).
75. Graham MJ, Freier SM, Crooke RM, Ecker DJ, Maslova RN, Lesnik EA. Tritium labeling
of antisense oligonucleotides by exchange with tritiated water. Nucleic Acids Res 1993;
21:3737-3743.
76. Desjardins JP, Sproat BS, Beijer B et al. Pharmacokinetics of a synthetic, chemically modi-
fied hammerhead ribozyme against the rat cytochrome P-450 3A2 mRNA after single in-
travenous injections. J Pharmacol Exp Ther 1996; 278(3):1419-1427.
77. Sandberg JA, Agrawal A, Bouhana K et al. Pharmacokinetics of an anti-FLT ribozyme
(RPI.4610) in normal and tumor bearing mice after daily bolus or continuous intravenous
infusion. Oligonucleotide & Gene Therapy Based Antisense Therapeutics with New Appli-
cation for Genomics. San Diego, CA: IBC USA Conferences, 1997.
78. Cossum PA, Truong L, Owxens SR, Markham PM, Shea JP, Crooke S. Pharmacokinetics
of a 14C-labeled phosphorothioate oligonucleotide ISIS 2105 after intradermal administra-
tion in rats. J Pharmacol Exp Ther 1994; 269:89-94.
79. Agrawal S, Temsamani J, Galbraith W, Tang JY. Pharmacokinetics of antisense oligonucle-
otides. Clin Pharmacokinet 1995; 28:7-16.
80. Iversen PL, Mata J, Tracewell WG, Zon G. Pharmacokinetics of an antisense phos-
phorothioate oligodeoxynucleotide against rev from human immunodeficiency virus type
1 in the adult male rat following single injections and continuous infusion. Antisense Res
Devel 1994; 4:43-52.
81. Bennet CF, Dean N, Ecker DJ. Pharmacology of antisense therapeutic agents. In: Agrawal
S, ed. Methods in Molecular Medicine: Antisense Therapeutics. Totowa, NJ: Humana Press,
1994:13-46.
82. Crooke S, Graham M, Zuckerman J et al. Pharmacokinetic properties of several novel oli-
gonucleotide analogs in mice. Journal Pharmacol Exp Ther 1996; 277(2):923-937.
83. Firestone RA. Low-density lipoprotein as a vehicle for targeting antitumor compounds to
cancer cells. Bioconjugate Chem 1994; 5:105-113.
84. Desjardins J, Mata J, Brown T et al. Cholesterol-conjugated phosphorothioate oligodeoxy-
nucleotides modulate CYP2B1 expression in vivo. J Drug Targeting 1995; 2:477-485.
85. Boutorine AS, Kostina EV. Reversible covalent attachment of cholesterol to oligodeoxy-
nucleotides for studies of the mechanisms of their penetration into eucaryotic cells.
Biochimie 1993; 75:34-41.
86. Polushin NN, Cohen JS. Antisense pro-drugs: 5'-ester oligodeoxynucleotides. Nucleic Ac-
ids Res 1994; 22:5492-5496.
87. Thierry AR, Rabinovich P, Peng B et al. Characterization of liposome-mediated gene de-
livery: Expression, stability and pharmacokinetics of plasmid DNA. Gene The 1997;
4:226-287.
88. Liu Y, Mounkes LC, Liggitt HD et al. Factors influencing the efficiency of cationic lipo-
some-mediated intravenous gene delivery. Nature Biotechnol 1997; 15:167-173.
89. Liu Y, Liggitt D, Zhong W et al. Cationic liposome-mediated intravenous gene delivery. J
Biol Chem 1995; 270(42):24864-24870.
90. Thierry AR, Lunardi-Iskandar Y, Bryant JL et al. Systemic gene therapy: Biodistribution
and long-term expression of a transgene in mice. Proc Natl Acad Sci USA 1995;
92:9742-9746.
91. Stewart MJ, Plautz GE, Del Buono L et al. Gene transfer in vivo with DNA-liposome com-
plexes: Safety and acute toxicity in mice. Human Gene Ther 1992; 3:267-275.
92. Plank C, Mechtler K, Szoka FJ, Wagner E. Activaton of the complement system by syn-
thetic DNA complexes: A potential barrier for intravenous gene delivery. Human Gene
Ther 1996; 7:1437-1446.
93. Lee E, Marshall J, Siegel C et al. Detailed analysis of structures and formulations of cat-
ionic lipids for efficient gene transfer to the lung. Human Gene Ther 1996; 7:1701-1717.
94. Sandberg JA, Min JJ, Jensen KL et al. Lipid-based carriers enhance biodistribution and
efficacy of an anti-angiogenic ribozyme in a murine Lewis lung carcinoma model. 1998
(submitted).
95. Hartsel S, Bolard J. Amphotericin B: New life for an old drug. Trends Pharmacol Sci 1996;
17:445-449.
96. Janknegt R. Liposomal formulations of cytotoxic drugs. Support Care Cancer 1996;
4:298-304.
97. Leenders AC, de Marie S. The use of lipid formulations of amphotericin B for systemic
fungal infections. Leukemia 1996; 10:1570-1575.
98. Working PK, Dayan AD. Pharmacological-toxicological expert report CAELIX(TM)
(Stealth(TM) Liposomal Doxorubicin HCl). Human & Experimental Toxicol 1996;
15:752-785.
99. Wu N, Da D, Rudoll T, Needham D, Whorton A, Dewhirst M. Increased microvascular
permeability contributes to preferential accumulation of stealth liposomes in tumor tissue.
Cancer Res 1993; 53:3765-3770.
100. Unezaki S, Murayama K, Hosoda J-I et al. Direct measurement of the extravasation of
polyethyleneglycol-coated liposomes into solid tumor tissue by in vivo florescence micros-
copy. Intl J Pharm 1996; 144:11-17.
101. Lasic DD, Papahadjopoulos D. Liposomes revisited. Science 1995; 267:1275-1276.
102. Hwang KJ, Beaumier PL. Partial interaction of liposomes with the biological milieu. In:
Gregoriadis G, ed. Liposomes as Drug Carriers—Recent Trends and Progress. New York:
Wiley, 1988:19-34.
103. Lasic DD, Needham D. The “stealth” liposome: A prototypical biomaterial. Chemical Re-
views 1995; 95(8):2601-2627.
104. Woodle M. Surface-modified liposomes: Assessment and characterization for increased sta-
bility and prolonged blood circulation. Chem Phys Lipids 1993; 64:249-262.
105. Torchilin VP, Omelyanenko VG, Papisov MI et al. Poly(ethylene glycol) on the surface:
On the mechanism of polymer-coated liposome longevity. Biochim Biophys Acta 1994;
1195:11-20.
106. Min J, Jensen K, Bouhana K et al. Long-circulating liposomes enhance the pharmacokinet-
ics of hammerhead ribozymes in a murine Lewis lung cancer model. 1997 (manuscript in
preparation).
107. Szoka F, Papahadjopolous D. Procedure for preparation of liposomes with large internal
aqueous space and high capture by reverse-phase evaporation. Proc Natl Acad Sci USA
1978; 75(9):4194-4198.
108. Mathiowitz E, Jacob JS, Jong YS et al. Biologically erodible microspheres as potential oral
drug delivery systems. Nature 1997; 386:410-414.
109. Allen T, Moase E. Therapeutic opportunities for targeted liposomal drug delivery. Adv
Drug Del Rev 1996; 21:117-133.
110. Kirpotin D, Park JW, Hong K et al. Sterically stabilized anti-HER2 immunoliposomes:
Design and targeting to human breast cancer cells in vitro. Biochemistry 1997; 36:66-75.
111. Wu NZ, Braun RD, Gaber MH et al. Simultaneous measurement of liposome extravasa-
tion and content release in tumors. Microcirculation 1997; 4(1):83-101.
CHAPTER 5
Novel RNA Motif (Dimeric Minizyme)

Capable of Cleaving L6 BCR-ABL
Fusion (b2a2) mRNA with High
Specificity
Tomoko Kuwabara, Masaki Warashina and Kazunari Taira
Introduction
T he hammerhead ribozyme is one of the smallest RNA enzymes. Because of its small size
and potential utility as an antiviral agent, it has been extensively investigated in terms of
the mechanism of its action and possible applications in vivo. It was first recognized as the
sequence motif responsible for self-cleavage (cis action) in the satellite RNAs of certain
viruses.1 The putative consensus sequence required for activity has three duplex stems and
a conserved “core” of two non-helical segments, plus an unpaired nucleotide at the cleavage
site. The trans-acting hammerhead ribozyme consists of an antisense section (stem I and
stem III) and a catalytic domain with a flanking stem/loop II section.2,3 Such RNA motifs
can cleave oligoribonucleotides at specific sites (most effectively at GUC).4-8 Because of its
small size and potential utility as an anti-virus agent, this ribozyme has been extensively
investigated in terms of the mechanism of its action and possible applications in vivo.9-26
For such applications, it is clearly necessary to direct the ribozyme specifically to the cellular
RNA target of interest.
Chronic myelogenous leukemia (CML) is a clonal myeloproliferative disorder of he-
matopoietic stem cells associated with the Philadelphia chromosome.27 The reciprocal chro-
mosomal translocation t(9; 22) (q34; q11) can be subdivided into two types: K28 transloca-
tions and L6 translocations. They result in the formation of the BCR-ABL fusion gene, which
encodes two types of mRNA: b3a2 (consisting of bcr exon 3 and abl exon 2) and b2a2 (con-
sisting of the bcr exon 2 and abl exon 2) (Fig. 5.1).28-33 Both of these mRNAs are translated
into a protein of 210 kDa (p210BCR-ABL) which is unique to the malignant cell phenotype.34
For the design of ribozymes that will disrupt chimeric RNAs, it is necessary to target
the junction sequence. Otherwise, normal mRNAs that share part of the chimeric RNA
sequence would also be cleaved by the ribozyme, with resultant damage to the host cells. In
the case of the BCR-ABL chimeric RNA sequence b3a2, a potential ribozyme-cleavage site,
a GUU triplet, is located three nucleotides upstream from the chimeric junction. A conven-
tionally designed hammerhead ribozyme might be expected to specifically cleave the ab-
normal mRNA generated from K28 translocations. Indeed, several such examples have been
Fig. 5.1. BCR-ABL transloca- K28 Translocation

tions and fusion mRNAs. The
BCR ABL
two types of chromosomal 1 2 3 2
translocation—K28-type
(upper panel) and L6-type
2 3 2 K28 b3a2 mRNA
(lower panel)—that are asso- L6 b2a2 mRNA
2 2
ciated with CML and the cor-
responding fusion mRNAs are
depicted. Dotted lines con-
necting bcr and abl exons in- K28 b3a2 mRNA (BCR exon 3 - ABL exon 2)
b3 3 nts a2
dicate alternative splicing
[
BCR ABL
pathways.
GUU triplet
L6 Translocation
BCR ABL
1 2 2
2 2 L6 b2a2 mRNA
L6 b2a2 mRNA (BCR exon 2 - ABL exon 2)

45 nts
BCR ABL
b2 a2 GUC triplet
reported.35-43 By contrast, in the case of the b2a2 sequence, which results from L6 transloca-
tions, as well as some K28 translocations, there are no triplet sequences that are potentially
cleavable by hammerhead ribozymes within two or three nucleotides from the BCR-ABL
junction.44 In designing ribozymes that might cleave b2a2 mRNA, we must be sure to avoid
cleavage of the normal abl mRNA itself.
Recently, we discovered a novel motif in a minizyme, a hammerhead ribozyme with
short oligonucleotide linkers instead of stem/loop II.45 Our previous study demonstrated
that a minizyme with high-level activity forms a dimeric structure with a common stem II.
Because of their dimeric structure, heterodimeric minizymes are capable of recognizing
two independent sequences. We wondered whether it would be possible to design a novel
heterodimeric minizyme that would form a catalytically competent structure only in the
presence of the L6 BCR-ABL (b2a2) mRNA junction, by the use of one of the substrate-
recognition sequences as the recognition arm of an active heterodimeric minizyme.
Since we were interested in cleaving b2a2 mRNA, we compared the specificity and cata-
lytic activity of conventional hammerhead ribozymes and our novel heterodimeric minizyme
with respect to the cleavage of BCR-ABL chimeric L6 (b2a2) mRNA.
Current Research
Minizymes
The hammerhead ribozyme is a small and versatile nucleic-acid molecule which can
cleave RNA at specific sites. In its most useful form, it consists of a substrate-binding region
(stem I and stem III) and a catalytic domain with a flanking stem-loop II region (Fig. 5.2A).
In attempts to identify functional groups and to elucidate the role of the stem II region,
Novel RNA Motif Capable of Cleaving L6 BCR-ABL Fusion mRNA with High Specificity 63
A B Cleavage site
Substrate
Cleavage site 5' GCCGUCCCCCG 3'
Substrate CGGCA GGGGC 5'
5' GCCGUCCCCCG 3' 3' A C
UG
A A
CGGCA GGGGC G U
3' A C 5' AG
Stem III U GA Stem I Homodimeric C G
A
G U minizyme A
G C
AG G G
C G U
A A
Ribozyme C G GU
G C C A
G C Stem-loop II 5' CGGGG ACGGC 3'
A G region
A A GCCCCCUGCCG
3' 5'
Substrate
Cleavage site
C
MzL ( Minizyme left ) Heterodimeric minizyme
3' CGGCA 5' Cleavage site
CAGUAGC Substrate
A CUG 5'GCCGUCCCCCG 3'
A A
U CGGCA GGGGC 5'
G AG 3'A CUG
CG A A
G U
A G
MzL C G MzR
MzR ( Minizyme right ) G C
A G
3' CCUUGCA G
GGGGC 5' A
U A
A CU GU
C A
G 5' CGAUGAC ACGUUCC 3'
A A
G G U GCUGCUGdCUGCAAGG
A 3'
Pseudosubstrate 5'
CG
Fig. 5.2. Secondary structures of (A) a hammerhead ribozyme, (B) the

homodimeric minizyme and (C) the heterodimeric minizyme used in a previ-
ous study.45 In the case of the heterodimeric minizyme, the heterodimer (MzL-
MzR) can generate two different binding sites: One is complementary to the
sequence of a substrate, the other is complementary to an uncleavable
pseudosubstrate. Only after the formation of the heterodimer could the sub-
strate be cleaved.
various modifications and deletions have been made in this region.46-52 Such minizymes are
smaller versions of hammerhead ribozymes in which the stem-loop II region has been re-
placed by a short linker. They can, therefore, be synthesized more economically and chemi-
cal modifications can be made more easily. For the application of such enzymes as thera-
peutic agents for the treatment of infectious diseases and cancer, minizymes seem to be
particularly attractive. However, activities of originally synthesized minizymes were two to
three orders of magnitude lower than those of the parental hammerhead ribozymes, a re-
sult that led to the suggestion that minizymes might not be suitable as gene-inactivating
reagents.51 Thus, original minizymes were considered to be crippled structures and attracted
minimal interest because of their extremely low activity, as compared to that of the full-
sized ribozyme.
Fig. 5.3. The effect of the concen-

trations of the pseudosubstrate A 20
Relative amount of product (%)

on the cleavage activity of the
heterodimeric minizyme. (A)
Time courses of cleavage activ-
15
ity of the heterodimeric mini-
zyme in various concentrations
of the pseudosubstrate (PsS):
open square, 0 M; closed circle, 10
200 nM; open circle, 500 nM; tri-
angle, 750 nM; closed square,
1 µM. (B)The effect of the con- 5
centrations of the pseudosub-
strate on V0 (min–1).
0
0 5 10 15 20 25
Time (min)
B 0.010
0.008
V0 ( min-1 )
0.006
0.004
0.002
0.00 0.25 0.50 0.75 1.00 1.25
Concentration of pseudosubstrate ( µM )
We later found that some minizymes have cleavage activity nearly identical to that of
the wild type hammerhead ribozyme.45 In the case of the active minizymes, the linker se-
quences were palindromic, so that two minizymes were capable of forming a dimeric struc-
ture with a common stem II (Fig. 5.2B). The activity of the homodimeric minizyme (a
dimer with two identical binding sequences) depends on Mg2+ ions, and interactions with
the substrates also stabilize the dimeric structures.45,53,54 Figure 5.3 shows the dependence
of cleavage activities of the MzL-MzR heterodimeric minizyme (a dimer with two different
binding sequences, Fig. 5.2C) on concentrations of a pseudosubstrate. Since the dimer for-
mation is essential for the cleavage reaction of the minizyme, the cleavage activity increases
linearly by the addition of a pseudosubstrate. This kinetic study of the heterodimeric
minizyme indicates that the active form of the minizyme is clearly a dimer. Since this novel
RNA motif, a dimeric minizyme, has two substrate-binding regions and two catalytic do-
mains, it was possible to construct dimeric minizymes that would cleave a target substrate
at two sites simultaneously.55 The cleavage activity and the stability of dimeric minizymes
increased with increases in number of G-C pairs in the common stem II region of the dimeric
minizyme.55
Cleavage of BCR-ABL mRNA by Conventional Ribozymes

In the sequence of BCR-ABL chimeric L6 (b2a2) mRNA, there are no triplet sequences
that are potentially cleavable by hammerhead ribozymes within two or three nucleotides
from the BCR-ABL junction. In the sequence of b2a2, ribozyme-cleavage sites in the vicin-
ity of the BCR-ABL junction are located 7, 8, 9, and 19 nucleotides away from the junction.
A GUC triplet, which is generally most susceptible to cleavage by hammerhead ribozymes,
is also located 45 nucleotides from the junction. For the design of ribozymes that will dis-
rupt chimeric RNAs, it is necessary to target the junction sequence. If such a GUC triplet
was selected as a cleavage site of ribozymes, normal abl mRNA that shares part of the abnor-
mal BCR-ABL RNA sequence would also be cleaved by the ribozyme, with resultant damage
to the host cells (Fig. 5.4A). In designing ribozymes that might cleave b2a2 mRNA, we must
be sure to avoid cleavage of the normal abl mRNA itself.
Previous attempts have involved a combination of a long antisense arm and the ribozyme
sequence.56,57 In the previous studies, long antisense sequences of about 10 to 30 nucle-
otides in length, which could bind to and cover the junction region for some distance from
the cleavage sites, were connected to one of the binding sites of hammerhead ribozymes
(81-mer, 41-mer and 52-mer Rzs) (Fig. 5.4B). The lengths of annealing arms are important
for the activity of ribozymes because they influence the efficiency, as well as the specificity,
of the cleavage reaction. In the case of a ribozyme that is directed against two non-contigu-
ous sequences, the specificity is particularly important if we are to avoid nonspecific cleav-
age of normal mRNAs. Among the conventional ribozymes depicted in Figure 5.4B, which
were designed to cleave L6 BCR-ABL chimeric mRNA specifically, the 52-mer Rz and the
41-mer Rz had long binding arms, in the stem III region, of 20 and 12 nucleotides, respec-
tively. In the case of the 81-mer Rz, the binding arm was 50 nucleotides in length in the stem
III region and was connected to the ribozyme sequence by a 13-nucleotide spacer sequence
that was non-complementary to the substrate, to achieve greater flexibility of binding.56
The 52-mer Rz was designed to cleave the L6 BCR-ABL mRNA at the UUC triplet located
9 nts 3' of the junction.57 The 41-mer Rz was designed to cleave the substrate at the CUU
triplet located 8 nts 3' of the junction. The 81-mer Rz was designed to cleave the substrate at
the GUA triplet located 19 nts 3' of the junction. According to a published report, the 81-
mer and 41-mer Rzs should have enhanced specificity for the chimeric b2a2 mRNA sub-
strate. However, according to other studies, it seems that hammerhead ribozymes have cleav-
age ability even if the binding arm is as little as three nucleotides in length. As can be seen
from Figure 5.4B, the binding region of these antisense-type ribozymes to the normal abl
mRNA sequence consisted of at least six base pairs. Therefore, we could not exclude the
possibility that such ribozymes might bind non-specifically to and cleave the normal abl
mRNA just as they specifically cleave the BCR-ABL (b2a2) mRNA. Moreover, longer sub-
strate-binding arms might lower the rate of dissociation from the substrate, with a resultant
reduction in the ribozyme-turnover rate. Therefore, we synthesized ribozymes (81-mer,
41-mer, and 52-mer Rzs in Fig. 5.4B) that were identical to those in the literature and re-
examined their specificities.
Comparison of the Specificities of Conventional Hammerhead Ribozymes and

Novel Dimeric Minizymes with Respect to the Cleavage of BCR-ABL Chimeric
L6 (b2a2) mRNA
Since a minizyme with high-level activity forms a dimeric structure with a common
stem II,45,53-55 it has two independent substrate-binding regions. Therefore, we decided to
66
Fig. 5.4. Cleavage of normal and/or
abnormal RNAs by conventional Conventional hammerhead ribozymes
ribozymes. The control ribozyme
A
GUC triplet
(Rz37), which targets the same site 45 nts
as the dimeric minizyme on L6 BCR exon 2 ABL exon 2
b2a2 mRNA, is expected to cleave
not only the abnormal chimeric 3' 5'
BCR-ABL mRNA but also the nor-
mal abl mRNA since the cleavage
site is located far from the BCR-ABL
GUC triplet
junction (upper panel). Nucleotide
sequences of the conventional Normal ABL mRNA
antisense-type ribozymes and Rz37
are shown in the lower panel. The Damage to
3' 5' host cells
sequence of L6 BCR-ABL near the
junction is expanded. The bcr exon
2 sequence near the junction is de-
picted by capital letters and that of B Target b2a2 substrate
the abl exon 2 sequence is shown in 5' BCR exon 2 ABL exon 2 3'
lower-case letters. The sites of cleav- CACAGCAUUCCGCUGACCAUCAAUAAGGAAG a a g c c c u u c a g c g g c c a g u a g c a u c u g a c u u guc
age by antisense-type ribozymes

Recognition site for dimeric minizyme 41mer-Rz 52-mer Rz 81mer-Rz
(81-mer Rz, 41-mer Rz and 52-mer
Rz) and the control ribozyme, Rz37, 3 ' - GU G U C G U A A G G C G A C U G G U A G U U A U U C C U U C C GGUC A CGUAGACUG-5'
Spacer sequence A C UG
are indicated. The site of cleavage AA A AACCCAAGAA A A
G U
81-mer Rz
C - G AG
by the dimeric minizyme is identi- 3'-CCUU C UUCGGGA GUCGCCG-5' A - U
A C UG G - C
cal to that of Rz37 and the 29 nt A A G - C
41-mer Rz G U A G GGAGUCCCA ACUCAC
recognition site for the dimeric Conventional ribozymes C - G AG G U 3' A C UG 5'
A - U A A
minizyme is also indicated by un- G - C G U
G - C AG
A G Rz37 C G
derlining. G U C G
G C
3'-AGUUAUU C CUU C UUCGGGAA UC GCC GGUCA-5' G C
A C UG A G
A A AA
52-mer Rz G U
C - G AG
A - U
G - C
G - C
A G
G U
Formation of active heterodimeric minizymes

in the presence of L6 b2a2 mRNA
5'
5'
BCRexon
BCR exon2 2 ABL exon 2
BCR exon
BCRexon 2 2 ABL exon 2
3' 5'
3' 5'
Super
Dimeric dimeric
minizyme minizyme
5 3' 5' 3'
' CUG
CUG
3'
3'
Cleavage site Cleavage site
Fig. 5.5. Cleavage of L6 b2a2 mRNA by dimeric minizymes. Minizyme left (MzL) and minizyme
right (MzR) form a dimeric structure with a common stem II in the presence of L6 b2a2 mRNA.
One unit of the substrate-recognition sequences is used to recognize the abnormal BCR-ABL
junction. One of the catalytic cores of the heterodimeric minizyme can be deleted completely to
yield a “super dimeric minizyme” (right figure).
use one of the substrate-binding regions, within an active heterodimeric minizyme, as the
recognition arms for the abnormal BCR-ABL junction (Fig. 5.5). One of the catalytic cores
of the heterodimeric minizyme can be deleted completely to yield a “super dimeric minizyme”.
In order to achieve high substrate-specificity, the heterodimeric minizyme should retain its
active conformation only in the presence of the abnormal BCR-ABL junction, while its con-
formation should remain inactive in the presence of the normal abl mRNA. The novel
minizyme, MzL (minizyme left) and MzR (minizyme right), shown in Figure 5.6 should
enable such conformational changes depending on the presence or absence of the abnormal
b2a2 mRNA. One unit of the substrate-recognition sequences is used to recognize the ab-
normal BCR-ABL junction. As a control ribozyme, Rz37, designed to target the same site as
the dimeric minizyme on L6 b2a2 mRNA, was also prepared. Rz37 is expected to cleave not
only the abnormal chimeric BCR-ABL mRNA but also the normal abl mRNA since the
cleavage site is located far from the BCR-ABL junction (Fig. 5.4B).
In order to examine the specificity of cleavage reactions catalyzed by conventional
ribozymes or by our novel heterodimeric minizyme, we examined three types of RNA sub-
strate (Figs. 5.7 and 5.8), namely, the normal abl substrate, the chimeric BCR-ABL sub-
strate, and a short BCR-ABL substrate with lengths, respectively, of 92, 121 and 16 nucle-
otides. Enzymes with high specificity should cleave the chimeric BCR-ABL substrate (Fig. 5.7)
or the 16-mer short abl substrate in the presence of a pseudo-substrate that has the b2a2
junction sequence (Fig. 5.8), without cleaving the other RNAs. All kinetic measurements
were made in 25 mM MgCl2 and 50 mM Tris-HCl (pH 8.0), under enzyme-saturating (single-
turnover) conditions at 37°C (measurements of kcat or kobs), namely, our standard condi-
tions for kinetic measurements.55
Normal ABL mRNA

5'
ABL exon 1
ABL exon 2
CUC CAG CUG UUA UCU GGAAG
AAG CCC UUC AGC 3'
GCU GAC CAU CAA UAA GGAAG
5' BCR exon 2

junction
Abnormal L6 b2a2 mRNA
Formation of active dimeric minizyme

ABL exon 2 3'
CCUCAGGGUCUGAGUG
3' GGAGUCCCA ACUCAC 5'
A C
UG
A A
U Abnormal L6 b2a2 mRNA
G G
A
MzL C G MzR
U A
A G
5'GAAGGGCUUCUUUC UUAUUGAUGGUCAG 3'
CUUCCCGAA GAAGGAAUAACUACCAGUC
ABL exon 2 BCR exon 2 5'
junction
Inactive heterodimeric minizymes
ABL exon 2 3'

CCUCAGGGUCUGAGUG
GGAGUCCCA ACUCAC
3' 5'
AA C
G
5' CU U
GAAGGGCUUCUUUCA GAUGA 3'
CUUCCCGAA GAAGGU CUAUU
ABL exon 2 ABL exon 1 5'
junction
5' 3'
CCUCAGGGUCUGAGUG Normal ABL mRNA
3' GGAGUCCCA ACUCAC 5'
A C
A
GU
CG
UA
MzL A U MzR
CG
UA
UG
UA
5' CG 3'
GAAGGGCUU UUAUUGAUGGUCAG
Fig. 5.6. Formation of active or inactive heterodimeric minizyme. In order to achieve

high substrate-specificity, the heterodimeric minizyme components should retain their
active conformation only in the presence of the abnormal BCR-ABL junction (middle
panel), while their conformation should retain inactive in the presence or absence of
the normal abl mRNA (bottom panel). The novel minizyme, MzL (minizyme left)
and MzR (minizyme right) should enable such conformational changes depending
on the presence or absence of the abnormal b2a2 mRNA.
Results of cleavage of relatively long substrates by the conventional antisense-type

ribozymes and the super dimeric minizymes are shown in Figure 5.7. As expected, all the
conventional ribozymes cleaved the BCR-ABL substrate at the anticipated sites. However, in
contrast to expectations based on previous reports,56,57 not only the control Rz37 but also
the antisense-type ribozymes cleaved the normal abl substrate within 1 hour. Moreover, the
amounts of cleavage products obtained from the normal abl mRNA with each ribozyme
were almost the same as those obtained from the chimeric BCR-ABL substrate, indicating
that these conventional ribozymes, with their relatively long antisense arms, recognized not
only the abnormal BCR-ABL mRNA but also the normal abl mRNA as substrates. There-
fore, non-specific cleavage of normal abl mRNA could not be avoided when we used con-
ventionally designed ribozymes. In previous studies on these long antisense-type ribozymes,56
one part of the target site was designed to be accessible for annealing and served to direct
ribozyme nucleation, while the other part recognized the cleavage triplet in the vicinity of
the BCR-ABL junction, where specific cleavage of the hybrid mRNA occurred. We note that,
in all cases, these conventional Rzs have regions of complementary binding to the normal
abl mRNA sequences of at least 6-8 nts. Previous studies of hammerhead ribozymes dem-
onstrated that cleavage of the substrate could occur when one of the substrate-binding arms
was three nucleotides long.58,59 Thus, we would not expect substrate specificity from the
conventionally designed ribozymes shown in Figure 5.4B.
Our novel heterodimeric minizyme was expected to show high substrate specificity for
the L6 BCR-ABL substrate, if and only if it forms an active conformation in the presence of
the abnormal b2a2 mRNA, as depicted in Figure 5.6. The specificity of the dimeric minizyme
was tested by incubating the minizymes with the 5'-[32P]-labeled short 16-mer substrate
(S16) in the presence or absence of either a short 20-mer normal abl pseudo-substrate or a
short 28-mer BCR-ABL pseudo-substrate (Fig. 5.8). In this case, one part of the target site
(b2a2 mRNA junction within the BCR-ABL pseudo-sub) was designed to be accessible to
MzL and MzR for annealing and served to direct formation of the active dimeric minizyme
(Fig. 5.6), while the other part recognized the cleavage triplet in the short 16-mer abl substrate
RNA, where specific cleavage of the latter RNA occurred (Fig. 5.8). It is to be noted that the
cleavage activity of the novel heterodimeric minizyme was nearly identical to that of the
control hammerhead ribozyme Rz37. In terms of substrate specificity, no products of cleav-
age of the substrate were detected in the absence of the BCR-ABL junction, demonstrating
the expected high substrate-specificity of the heterodimeric minizyme. Since MzL or MzR
alone, in the presence or in the absence of the pseudo-substrate, did not show any cleavage
activity, the active species is clearly the heterodimeric form of the minizyme. A similar study
demonstrated high substrate-specificity for the longer L6 BCR-ABL substrate (Fig. 5.7).
Activities and Specificities of tRNA-Embedded Minizymes

Ribozymes have been shown to be potent inhibitors of gene expression and viral func-
tion. There are two basic strategies for ribozyme delivery into cells, endogenous delivery in
that a gene for the ribozyme is transcribed in cells, and exogenous delivery in that a pre-
synthesized ribozyme is supplied to the cells by means of carriers. For the endogenous
ribozyme expression, the gene encoding the ribozyme is inserted into a vector which can be
delivered into target cells. Efficacy of ribozyme-mediated repression of a target gene in liv-
ing cells depends on the level of expression and also the stability of transcribed ribozymes.
For the repression of BCR-ABL chimeric mRNA whose products cause CML diseases, our
dimeric minizyme should be particularly useful because of its high cleavage activity and
specificity. For the endogenous delivery of minizymes, we chose a promoter for RNA
polymerase III, which naturally drives tRNA and snRNA synthesis with high levels of
expression.
Fig. 5.7. Gel electrophoresis showing cleavage of normal and/or abnormal long substrate by con-
ventional ribozymes and dimeric minizyme. It revealed the non-specific cleavage of chimeric
BCR-ABL mRNA, as well as of normal abl mRNA, by conventional ribozymes. In contrast, the
dimeric minizyme showed high cleavage specificity. Specificity was examined with the normal
abl substrate (92-mer) and the chimeric BCR-ABL substrate (121-mer). Each enzyme (1 µM)
and 2 nM 5'-[32P]-labeled substrate were incubated at 37°C for 60 min in a solution that con-
tained 50 mM Tris-HCl (pH 8.0) and 25 mM MgCl2 and then the mixture was subjected to
electrophoresis on an 8% polyacrylamide/7M urea gel.
Cleavage products from the normal abl substrate (92-mer) were as follows. Non-specific
cleavage, at the UUC triplet located 9 nts 3' of the junction, by the 52-mer Rz generated a [32P]-
labeled 5'-fragment of 43 nts in length. Similarly, non-specific cleavage, at the CUU triplet lo-
cated 8 nts 3' of the junction, by the 41-mer Rz generated a visible fragment of 42 nts. Non-
specific cleavage, at GUA located 19 nts 3' of the junction, by the 81-mer Rz generated a fragment
of 54 nts. Non-specific cleavage, at GUC located 45 nts 3' of the junction, by Rz37 generated a
fragment of 79 nts.
Cleavage products from the BCR-ABL substrate (121-mer) were as follows. Cleavage by the
52-mer Rz generated a visible 5'-fragment of 72 nts in length. Similarly, cleavage by the 41-mer
Rz generated a fragment of 71 nts. Cleavage by the 81-mer Rz generated a fragment of 83 nts.
Cleavage by Rz37 and dimeric minizyme generated a fragment of 108 nts in length.
Fig. 5.8. Gel electrophoresis showing cleavage by dimeric minizymes. The specificity of
dimeric minizyme-mediated cleavage was tested by incubating the minizymes with the
5'-[32P]-labeled short 16-mer substrate S16 (Sub) in the presence or absence of either a
short 20-mer normal abl pseudo-substrate (abl pseudo-sub) or a short 28-mer BCR-ABL
pseudo-substrate (BCR-ABL pseudo-sub). Minizymes (MzL and MzR) were incubated at
0.1 µM with 2 nM 5'-[32P]-labeled S16 substrate (Sub). When applicable, the concentra-
tion of pseudosubstrate, such as abl or BCR-ABL, was at 1 µM. Reactions were usually
initiated by the addition of 25 mM MgCl2 to a buffered solution that contained 50 mM
Tris-HCl (pH 8.0) and enzyme together with the substrate, and each resultant mixture
was then incubated at 37°C for 60 min. The reaction mixture was subjected to electro-
phoresis on an 8% polyacrylamide/7M urea gel.
In order to achieve high expression of these dimeric minizymes in vivo for future gene
therapy, we embedded the dimeric minizyme portion (MzL and MzR) downstream of a
tRNAVal promoter sequence which could be recognized by RNA polymerase III (Fig. 5.9).
For the tRNAVal promoter-driven minizymes to be active, the attached tRNAVal portion should
not cause severe steric hindrance during the formation of dimeric minizymes. Therefore,
we first examined the effect of the tRNA portion of the pol III-derived ribozyme transcripts.
Results of cleavage by the tRNA-embedded dimeric minizyme are shown in Figure 5.10. In
order to examine the specificity of cleavage reactions catalyzed by tRNA-embedded dimeric
minizymes, we examined both chimeric BCR-ABL substrate (121 mer) and the normal abl
substrate (92 mer). Template DNAs which encode the T7 promoter and tRNA-embedded
dimeric minizyme sequence depicted in Figure 5.9 were prepared. In the T7 transcription
reaction solution, purified [32P]-labeled substrate was added. As can be seen in Figure 5.10,
Secondary structure of tRNAVal-MzL U

CA U
C
U G 20
CG UC A
U A U AA
U A
60 U
G A CG CC
C AC G CU G
A G U CG C G
A GC G C C A A
GAA CC G
A G
U C G
G G
C G A G
1
40 CU MzL 29mer
U
G CA
C CU
A G G U A C A 80 U
C U A U GA A 100 U UC
U G 20 UG AA C A
A G A U
UU C U CC A AG CU G CU
C U G A
UG A AA AGG A
GU C A
C CA
UG UC
C G 120
AG
C A UU GA
UU
1 U AA
tRNAVal-MzL 128mer
Secondary structure of tRNAVal-MzR G

A U GA
20
GA UA U A U G
U
C A CU GU U G U G
CA AC
C G
1
MzL 30mer
100
CG A GA
U A U U
U A C G
60 G C A A
G C C G
C G U A
G UCC G 80 U C GUAU
G U C U
C G U A U G
A A
C CG C GA A CACUAC
AA
A AA C C AA C A U G G U CA
120
A
A C G CU U GU GA U G U UU G G UU G U G C C A AG
UC G C G U 20 CC C A U AU
A C
40 A U U UU
C G 1
U G
AU U
tRNAVal-MzR 133mer
Fig. 5.9. Secondary structures of tRNA-embedded dimeric minizyme (tRNAVal-MzL and tRNAVal-
MzR) and naked dimeric minizyme (MzL and MzR; upper right figure in both panels). The
dimeric minizyme portion is outlined.
Fig. 5.10. Gel electrophoresis showing cleavage by tRNA-embedded dimeric minizyme.

Minizymes (MzL and MzR) which form a dimeric structure were embedded after the tRNAVal
promoter sequence, namely tRNAVal-MzL and tRNAVal-MzR. The tRNAVal-MzL and the
tRNAVal-MzR form an active dimer complex, namely tRNAVal-DMz. The reaction solutions
after 3 h and 6 h incubation for each condition was loaded on the gel (From lane 3 to lane 16).
The cleavage activity and specificity was examined in the T7 transcription of tRNA-embed-
ded dimeric minizyme with incubation of purified [32P]-labeled substrate. In lanes 15 and 16
(*), the reaction solutions which contained an excess of T7 RNA polymerase were loaded.
the tRNA-embedded dimeric minizyme specifically cleaved the chimeric BCR-ABL sub-
strate. No cleavage products were detected in the presence of the normal abl substrate. As
expected, tRNA-MzL or tRNA-MzR alone did not show any cleavage. Similarly to the naked
dimeric minizyme (Figs. 5.7 and 5.8), the tRNA-embedded dimeric minizyme also showed
high substrate specificity. Figure 5.10 demonstrates the formation of active dimeric tRNA-
minizymes, providing evidence that the attached tRNA portion did not prohibit the dimer-
ization process.
In order to characterize in further detail the properties of tRNA-embedded dimeric
minizymes, we determined kinetic parameters using a short 16-mer substrate (S16), and
Table 5.1. Kinetic parameters for the cleavage of short BCR-ABL substrate (S16)*
Enzyme kcat (min–1) Kd (µM)
Dimeric minizyme 0.018 0.27

tRNA-embedded dimeric minizyme 0.013 0.21
*All reaction rates were measured, in 25 mM MgCl2 and 50 mM Tris-HCl (pH 8.0) under enzyme-
saturating (single-turnover) conditions at 37˚C. In all cases, kinetic measurements were made under
conditions where all of the substrate was expected to form a Michaelis-Menten complex, with high
concentrations of enzymes. Rate constants are averages from two sets of experiments.
compared the activities between the tRNA-embedded dimeric minizyme and the naked
dimeric minizyme. Kinetic analysis carried out under single-turnover conditions and the
rate constants of the dimeric minizyme and tRNA-embedded dimeric minizyme determined
with the S16 substrate are shown in Table 5.1. Interestingly, the cleavage activity of tRNA-
embedded dimeric minizymes was almost the same as that of the naked dimeric minizymes.
This result indicates that our dimeric minizyme can fully form an active dimeric structure
even if the tRNAVal sequence was connected at the 5' end of each minizyme (MzL and MzR).
Moreover, the minizyme retains the active conformation only in the presence of the abnor-
mal BCR-ABL junction (b2a2 mRNA), while the conformation remains inactive in the ab-
sence of an abnormal BCR-ABL junction.
Future Prospects
Potential Gene Therapy for Treatment of Chronic Myelogenous Leukemia (CML)
The specific association of nucleic acid-based drugs, such as our novel heterodimeric
minizymes, with their targets via base pairing and subsequent cleavage of the RNA sub-
strate suggests that these catalytic molecules might be useful for gene therapy. There are
basically two ways to introduce ribozymes into cells. One such technique is an exogenous
delivery (drug-delivery) system (DDS) in which chemically pre-synthesized ribozymes are
encapsulated in liposomes or other related compounds and delivered to target cells. For this
exogenous delivery, chemical modifications60 to make nuclease-resistant heterodimeric
minizymes and/or DNA enzymes44,61 should be useful. Another way to introduce ribozymes
into cells is by transcription from the corresponding DNA template (gene therapy). Current
gene-therapy technology is limited primarily by the necessity for ex vivo manipulations of
target tissues and the technology is practical for endogenous delivery systems.62 Ribozymes
with natural components but not chemically modified counterparts can be transcribed in
vivo. In this context, our novel heterodimeric minizymes driven by a pol III promoter are
superior to the other nucleic acid-based drugs, because of their extremely high substrate
specificity and high cleavage activity, for the treatment of chronic myelogenous leukemia
(CML), especially in the case of L6 translocations.
References
1. Symons RH. Self-cleavage of RNA in the replication of small pathogens of plants and ani-
mals. Trend Biochem Sci 1989; 14:445-450.
2. Haseloff J, Gerlach WL. Simple RNA enzymes with new and highly specific endonuclease
activities. Nature 1988; 334:585-591.
3. Uhlenbeck OC. A small catalytic oligonucleotide. Nature 1987; 328:596-600.
4. Koizumi M, Iwai S, Ohtsuka E. Cleavage of specific sites of RNA by designed ribozymes.

FEBS Lett 1988; 239:285-288.
5. Ruffer DE, Stormo GD, Uhlenbeck OC. Sequence requirements of the hammerhead RNA
ribozyme. Gene 1992; 113:157-163.
7. Shimayama T, Nishikawa S, Taira K. Generality of the NUX rule: Kinetic analysis of the
results of systematic mutations in the trinucleotide at the cleavage site of hammerhead
8. Zoumadakis M, Tabler M. Comparative analysis of cleavage rates after systematic permu-
tation of the NUX consensus tartget motif for hammerhead ribozymes. Nucleic Acids Res
1995; 23:1192-1196.
9. Sarver N, Cantin E, Chang P et al. Ribozymes as potential anti-HIV-1 therapeutic agents.
Science 1990; 247:1222-1225.
10. Homann M, Tzortzakaki S, Rittner K et al. Incorporation of the catalytic domain of a
hammerhead ribozyme into antisense RNA enhances its inhibitory effect on the replica-
tion of human immunodeficiency virus type 1. Nucleic Acids Res 1993; 21:2809-2814.
11. Mulligan RC. The basic science of gene therapy. Science 1993; 260:926-932.
12. Altman S. RNA enzyme-directed gene therapy. Proc Natl Acad Sci USA 1993; 90:
10898-10900.
13. Marschall P, Thomson JB, Eckstein F. Inhibition of gene expression with ribozymes. Cell
Mol Neurobiol 1994; 14:523-538.
14. Sullivan SM. Development of ribozymes for gene therapy. J Invest Dermatol 1994; 103:85-89.
15. Cameron FH, Jennings PA. Multiple domains in a ribozyme construct confer increased
suppressive activity in monkey cells. Antisense Res Dev 1994; 4:87-94.
16. Ohkawa J, Yuyama N, Takebe Y et al. Importance of independence in ribozyme reactions:
Kinetic behavior of trimmed and of simply connected multiple ribozymes with potential
activity against human immunodeficiency virus. Proc Natl Acad Sci USA 1993;
90:11302-11306.
17. Sun LQ, Pyati J, Smythe J et al. Resistance to HIV-1 infection conferred by transduction of
human peripheral blood lymphocytes with ribozyme, antisense or polyTAR constructs. Proc
Natl Acad Sci USA 1995; 92:7272-7276.
38:2023-2037.
19. Ferbeyre G, Bratty J, Chen H et al. Cell cycle arrest promotes trans-hammerhead ribozyme
action in yeast. J Biol Chem 1996; 271:19318-19323.
20. Kiehntopf M, Brach MA, Licht T et al. Ribozyme-mediated cleavage of the MDR-1 tran-
script restores chemosensitivity in previously resistant cancer cells. EMBO J 1994;
13:4645-4652.
21. Kiehntopf M, Esquivel EL, Brach MA et al. Ribozymes: Biology, biochemistry, and impli-
cations for clinical medicine. J Mol Med 1995; 73:65-71.
22. Thompson JD, Ayers DF, Malmstrom TA et al. Improved accumulation and activity of
ribozymes expressed from a tRNA-based RNA polymerase III promoter. Nucleic Acids Res
1995; 23:2259-2268.
23. Thompson JD, Macejak D, Couture L et al. Ribozymes in gene therapy. Nature Med 1995;
1:277-278.
24. Tuschl T, Thomson JB, Eckstein F. RNA cleavage by small catalytic RNAs. Curr Opin Struct
Biol 1995; 5:296-302.
mediated cleavage within a fusion gene for adenovirus E1A-associated 300kDa protein
26. Eckstein F, Lilly DMJ eds. In: Catalytic RNA, Nucleic Acids and Molecular Biology. Vol
10. Berlin:Springer-Verlag, 1996.
27. Nowell PC, Hungerford DA. A minute chromosome in human chronic granulocytic leuke-
mia. Science 1960; 132:1497-1499.
28. Rowley JD. A new consistent chromosomal abnormality in chronic myelogenous leukaemia
identified by quinacrine fluorescence and Giemsa staining. Nature 1973; 243:290-293.
29. Bartram CR, de Klein A, Hagemeijer A et al. Translocation of c-abl oncogene correlates
with the presence of a Philadelphia chromosome in chronic myelocytic leukaemia. Nature
1983; 306:277-280.
30. Heisterkamp N, Stephenson JR, Groffen J et al. Localization of the c-abl oncogene adja-
cent to a translocation break point in chronic myelocytic leukaemia. Nature 1983;
306:239-242.
31. Groffen J, Stephenson JR, Heisterkamp N et al. Philadelphia chromosomal breakpoints are
clustered within a limited region, bcr, on chromosome 22. Cell 1984; 36:93-99.
32. Schtivelman E, Lifshitz B, Gale RP et al. Fused transcript of abl and bcr genes in chronic
myelogenous leukaemia. Nature 1985; 315:550-553.
33. Shtivelman E, Lifschitz B, Gale RP et al. Ribozyme-mediated inhibition of bcr-abl gene
expression in a Philadelphia chromosome-positive cell line. Cell 1986; 47:277-284.
34. Konopka JB, Watanabe SM, Witte ON. An alteration of the human c-abl protein in K562
leukemia cells unmasks associated tyrosine kinase activity. Cell 1984; 37:1035-1042.
35. Shore SK, Nabissa PM, Reddy EP. Ribozyme-mediated cleavage of the BCR-ABL oncogene
transcript: in vitro cleavage of RNA and in vivo loss of P210 protein-kinase activity.
Oncogene 1993; 8:3183-3188.
36. Snyder DS, Wu Y, Wang JL et al. Ribozyme-mediated inhibition of bcr-abl gene expres-
sion in a Philadelphia chromosome-positive cell line. Blood 1993; 82:600-605.
37. Lange W, Cantin EM, Finke J et al. In vitro and in vivo effects of synthetic ribozymes
targeted against BCR/ABL mRNA. Leukemia 1993; 7:1786-1794.
38. Wright L, Wilson SB, Milliken S et al. Ribozyme-mediated cleavage of the bcr/abl tran-
script expressed in chronic myeloid leukemia. Exp Hematol 1993; 21:1714-1718.
39. Lange W, Daskalakis M, Finke J et al. Comparison of different ribozymes for efficient and
specific cleavage of BCR/ABL related mRNAs. FEBS Lett 1994; 338:175-178.
40. Kearney P, Wright LA, Milliken S et al. Improved specificity of ribozyme-mediated cleav-
age of bcr-abl mRMA. Exp Hematol 1995; 23:986-989.
41. Leopold LH, Shore SK, Newkirk TA et al. Multi-unit ribozyme-mediated cleavage of bcr-
42. Kronenwett R, Haas R, Sczakiel G. Kinetic selectivity of complementary nucleic acids: bcr-
abl-directed antisense RNA and ribozyme. J Mol Biol 1996; 259:632-644.
43. Leopold LH, Shore SK, Reddy EP. Multi-unit anti-BCR-ABL ribozyme therapy in chronic
myelogenous leukemia. Leuk Lymphoma 1996; 22:365-373.
44. Kuwabara T, Warashina M, Tanabe T et al. Comparison of the specificities and catalytic
activities if hammerhead ribozymes and DNA enzymes with respect to the cleavage of
BCR-ABL chimeric L6 (b2a2) mRNA. Nucleic Acids Res 1997; 25:3074-3082.
45. Amontov S, Taira K. Hammerhead minizymes with high cleavage activity: A dimeric struc-
ture as the active conformation of minizymes. J Am Chem Soc 1996; 118:1624-1628.
46. Goodchild J, Kohli V. Ribozymes that cleave an RNA sequence from human immunodefi-
ciency virus: The effect of flanking sequence on rate. Arch Biochem Biophys 1991;
284:386-391.
47. McCall MJ, Hendry P, Jennings PA. Minimal sequence requirements for ribozyme activity.
Proc Natl Acad Sci USA 1992; 89:5710-5714.
48. Tuschl T, Eckstein F. Hammerhead ribozymes: Importance of stem-loop II activity. Proc
Natl Acad Sci USA 1993; 90:6991-6994.
49. Thomson JB, Tuschl T, Eckstein F. Activity of hammerhead ribozymes containing non-
50. Fu DJ, Benseler F, Mclaughlin LW. Hammerhead ribozymes containing non-nucleoside
linkers are active RNA catalysts. J Am Chem Soc 1994; 116:4591-4598.
51. Long DM, Uhlenbeck OC. Kinetic characterization of intramolecular and intermolecular
hammerhead RNAs with stem II deletions. Proc Natl Acad Sci USA 1994; 91:6977-6981.
52. Hendry P, McCall MJ, Santiago FS et al. In vitro activity of minimised hammerhead
ribozymes. Nucleic Acids Res 1995; 23:3922-3927.
53. Amontov S, Nishikawa S, Taira K. Dependence on Mg2+ ions of the activities of dimeric
hammerhead minizymes. FEBS Lett 1996; 386:99-102.
54. Sugiyama H, Hatano K, Saito I et al. Catalytic activities of hammerhead ribozymes with a
triterpenoid linker instead of stem/loop II. FEBS Lett 1996; 392:215-219.
55. Kuwabara T, Amontov S, Warashina M et al. Characterization of several kinds of dimer
minizyme: Simultaneous cleavage at two sites in HIV-1 tat mRNA by dimer minizymes.
56. Pachuk CJ, Yoon K, Moelling K et al. Selective cleavage of bcr-abl chimeric RNAs by a
ribozyme targeted to non-contiguous sequence. Nucleic Acids Res 1994; 22:301-307.
57. James H, Mills K, Gibson I. Investigating and improving the specificity of ribozymes di-
rected against the bcr-abl translocation. Leukemia 1996; 10:1054-1064.
EMBO J 1996; 15:3751-3757.
59. Birikh KR, Heaton PA, Eckstein F. The hammerhead ribozyme—structure, function and
application. Eur J Biochem 1997; 245:1-16.
60. Jarvis TC, Wincott FE, Alby LJ et al. Optimizing the cell efficacy of synthetic ribozymes.
Site selection and chemical modifications of ribozymes targeting the proto-oncogene c-
myb. J Biol Chem 1996; 271:29107-29112.
61. Santoro SW, Joyce GF. A general purpose RNA-cleaving DNA enzyme. Proc Natl Acad Sci
USA 1997; 94:4292-4266.
62. Morgan RA, Anderson WF. Human gene therapy. Annu Rev Biochem 1993; 62:191-217.
CHAPTER 6
Using Ribozymes to Attenuate

Gene Expression in Transgenic Mice
Shimon Efrat
Introduction
T he development of efficient methods for downregulation of cell-specific gene expression

in vivo is of great value for gaining new insights into complex biological systems, and for
advancing gene therapy for human diseases. Recent progress in gene targeting approaches
in transgenic mice utilizing the Cre-loxP DNA recombination system opened the way for
disrupting gene function in a cell-specific and time-specific manner.1 By placing the Cre
recombinase under control of an inducible regulatory system, such as that of the bacterial
tetracycline operon,2 one can control the timing of elimination of a certain gene function
from a particular cell type in vivo.
Many applications, however, require partial attenuation of gene expression, rather than
the all-or-none effect of gene disruption. For such purposes the transcript-specific
downregulation of RNA levels and activity using antisense RNA techniques represents a
promising approach. The use of antisense RNA in vivo has met with some success in a
number of transgenic experiments (reviewed in ref. 3), while in many other cases it failed to
affect a significant change in gene expression. One major obstacle has been the need to
maintain relatively high intracellular levels of the antisense RNA, which interacts stoichio-
metrically with the target transcript.
The discovery of ribozymes opened the way for incorporating catalytic RNA elements
into antisense RNA. This allows one antisense RNA molecule to interact with and inactivate
multiple target RNA molecules, thus reducing the constraint to overexpress a vast excess of
the antisense RNA in the target cells. In principle this advantage appears very attractive. It is,
therefore, surprising that only a handful of publications have reported the successful appli-
cation of ribozymes in transgenic animals.
The choice of target may be an important factor. Ribozymes, and antisense RNA ap-
proaches in general, may be effective against constitutively expressed genes that are not
regulated at the transcriptional level. In contrast, a gene with a capacity to sense the ribozyme-
induced downregulation of expression by some feedback mechanism, and respond by
upregulating its transcription, may be much harder to downregulate with this approach.
Examples of Ribozyme Applications in Transgenic Mice

Our laboratory has chosen to apply ribozymes to downregulate the expression of the
enzyme glucokinase (GK) in pancreatic β cells in transgenic mice.4 GK is a high-Km member
of the hexokinase family of enzymes. It is expressed specifically in β cells and hepatocytes,
where it represents the major activity responsible for phosphorylation of glucose to glu-
cose-6-phosphate.5 This is the first and rate-limiting step in glycolysis, which in β cells gen-
erates signals for insulin secretion.6 Thus GK has been denoted the “glucose sensor” of β
cells, in charge of coupling extracellular glucose levels to the correct amounts of secreted
insulin.5 Reduced glucokinase activity may result in decreased sensitivity of β cells to glu-
cose, leading to abnormal insulin secretion and diabetes. DNA polymorphism studies have
established a linkage between the GK locus and diabetes in patients with a non insulin-
dependent diabetes (type II) form termed MODY (maturity-onset diabetes of the young).7,8
This disease is characterized by an early age of onset and an autosomal dominant inherit-
ance. Sequencing of the GK gene from MODY patients has detected a number of nonsense
and missense mutations which are associated with regions of the enzyme molecule involved
in glucose and ATP binding.7,8 The molecular mechanism which makes these mutations
dominant remains unknown, but a gene dosage effect has been suggested. The inheritance
pattern of the disease suggests that the patients’ β cells contain normal enzyme molecules
encoded by the wild type allele. The mutant proteins manifest drastically reduced enzy-
matic activities.9,10 Since GK expression in β cells is not transcriptionally regulated,11,12 the
wild type allele can not compensate for this reduction. The decreased GK activity may be
sufficient to shift the threshold for glucose sensing, thereby resulting in impaired insulin
secretion at physiological glucose levels. However, it remains unclear whether abnormal
glucose metabolism in the liver contributes to the disease. Glucose uptake into the liver,
where it is stored as glycogen, is an important component in maintaining normal blood
glucose levels. This function depends on normal glucokinase activity. The GK gene is tran-
scribed in hepatocytes through a specialized promoter that was shown to be upregulated by
insulin.11,13 Thus, it is possible that the hepatocytes of MODY patients can correct the re-
duced GK activity by increased transcription of the wild type allele. Yet, as this regulation
depends on insulin, it may not be effective when insulin secretion is impaired, as is the case
in MODY patients.
Understanding of the human disease can benefit from an animal model, which will
allow detailed biochemical and physiological studies. We aimed at generating transgenic
mice in which GK activity is specifically impaired in β cells, leaving the liver activity intact,
to dissect the relative contribution of these two cell types to the MODY phenotype. To this
end a GK-specific ribozyme was expressed in β cells in transgenic mice under control of the
insulin promoter.4
A synthetic DNA fragment was generated consisting of two 12 bp fragments of mouse
GK gene exon 3 sequence in antisense orientation that flank a hammerhead ribozyme cata-
lytic element14,15 (Fig. 6.1a). The length of the regions complementary to the target repre-
sents a compromise between the need to provide target specificity on one hand and cata-
lytic turnover on the other hand. The region of the transcript chosen as target should be
unique to assure target specificity. The only target RNA sequence requirements for cleavage
by the ribozyme is a GUX element (X = A, C or U). The ribozyme cleaves the target RNA
immediately 3' to the X (Fig. 6.1a). A number of ribozymes directed against different re-
gions of the transcript may need to be tested for each target in cultured cells, in order to
choose the most efficient one for in vivo expression.
The hybrid DNA fragment was placed downstream of the rat insulin II promoter and
an intron element, and upstream of the SV40 late polyadenylation site (Fig. 6.1b). Stable
transfection of this construct, denoted RIP-GKRZ, into a murine β cell line (βTC6) resulted
in a 45% reduction in GK mRNA levels; however no cleavage products could be detected
(Fig. 6.2b). This likely results from the fact that the two cleavage products lack either a cap
site or a poly-A tail, both of which are needed for RNA stability. In their absence the RNA is
rapidly degraded. In the absence of demonstrable degradation products it is hard to prove
Using Ribozymes to Attenuate Gene Expression in Transgenic Mice 81
Fig. 6.1. Design of the RIP-

GKRZ construct. (a) The GKRZ
transcript hybridized with the
target GK mRNA. The con-
served ribozyme nucleotides are
shown in bold letters. The arrow
marks the putative cleavage
site. (b) The RIP-GKRZ hybrid
gene consists of a synthetic
DNA fragment encoding the
ribozyme flanked by GK anti-
sense sequences (GKRZ), an
upstream intron element and a
downstream polyadenylation
site. Reproduced with permis-
sion from Efrat S et al, Proc
Natl Acad Sci USA 1994; 91:
2051-2055.
that the reduction in mRNA levels occurs as a result of ribozyme activity, as opposed to a
simple antisense effect, which could lead to degradation of the target GK transcript through
the formation of a duplex RNA. It is not known whether the ribozyme cleavage activity and
the subsequent RNA degradation take place in the nucleus or in the cytoplasm.
Immunoblotting analysis of the transfected cells revealed a 2- to 3-fold reduction in GK
protein levels, compared to untransfected cells (Fig. 6.2c). This suggests that in addition to
RNA degradation, attenuation may be achieved through reduced translational activity of
the GK mRNA, presumably by the formation of double-stranded RNA hybrids with the
GKRZ transcripts, as has been observed in other antisense RNA experiments.16
The RIP-GKRZ construct was microinjected into mouse embryos, and two indepen-
dent transgenic lines were shown to express the transgene by RT-PCR analysis of islet RNA.
Immunohistochemical analysis of pancreas sections with a GK antiserum revealed a re-
duced staining intensity in transgenic islets, compared with normal controls (Fig. 6.3).
Glucose phosphorylation activity at various glucose concentrations was assayed in is-
lets isolated from the transgenic mice. GK activity in RIP-GKRZ islets was reduced by 70%,
compared to normal islets, while the activity of the related low-Km hexokinases remained
essentially unaffected (Fig. 6.4). This finding demonstrates the sequence specificity of the
approach. The incomplete inhibition of expression obtained with the antisense approach
may represent an advantage for studying its consequences in vivo, since it mimics the situ-
ation in MODY. In addition, total shutoff of GK expression in β cells is lethal, as demon-
strated by gene disruption experiments.17-19
Although no data on islet GK activity is available from MODY patients, it is assumed
that the wild type GK allele produces half of the normal activity. Therefore islet GK activity
in the RIP-GKRZ mice is likely to be as low or lower than that of MODY patients. Neverthe-
less, the RIP-GKRZ mice maintained normal fasting plasma glucose and insulin levels and
manifested normal glucose tolerance. In contrast, analysis of glucose-induced insulin secre-
tion from in situ-perfused pancreas (Fig. 6.5) revealed a markedly reduced response, for the
Fig. 6.2. Analysis of pRIP-GKRZ effect

on GK expression in transfected β cells.
(a) RT-PCR analysis of GKRZ expres-
sion in the transfected (+) vs. untrans-
fected (-) cells. Expression is manifested
by the 200 bp band that results from
spliced transcripts. P represents the size
of an unspliced fragment amplified
from pRIP-GKRZ DNA. (b) Northern
blotting analysis of GK mRNA. Ribo-
somal RNA bands serve as size mark-
ers. An α-tubulin probe was used to
correct for loading. (c) Immunoblot-
ting analysis of GK protein. Size mark-
ers are in kilodaltons. Reproduced with
permission from Efrat S et al, Proc Natl
Acad Sci USA 1994; 91:2051-2055.
Fig. 6.3. Immunohistochemical analysis of RIP-GKRZ pancreas. Tissue sections were

incubated with a sheep-anti-GK serum followed by a biotinylated second antibody,
and visualized with horseradish peroxidase-conjugated avidin. (A) transgenic islet
stained with pre-immune sheep serum; (B) transgenic islet stained with anti-GK
serum; (C) normal islet stained with pre-immune serum; (D) normal islet stained
with anti-GK serum. Magnification is x200. Reproduced with permission from Efrat S
et al, Proc Natl Acad Sci USA 1994; 91:2051-2055.
Fig. 6.4. Glucose phosphorylation activ-

ity in RIP-GKRZ islets. The great differ-
ence in Km, 0.05 and 8 mM for hexoki-
nase and glucokinase, respectively, allows
distinction between the two enzymatic
activities. Solid bar, normal islets; dot-
ted bar, RIP-GKRZ islets, lineage 2;
hatched bar, RIP-GKRZ islets, lineage 4.
Values are expressed in units per gram
of islet protein. 1 U = 1 µmol product
per minute. Values are mean ± SEM
(n = 3). Transgenic and normal GK ac-
tivity differences are statistically signifi-
cant (p <0.0001 by t test). Reproduced
with permission from Efrat S et al, Proc
Natl Acad Sci USA 1994; 91:2051-2055.
Fig. 6.5. Insulin secretion from in situ-per-

fused control (open circle) and transgenic
(closed circle) mouse pancreas. Values are
mean ± SEM (n = 5). Reproduced with
permission from Efrat S et al, Proc Natl
Acad Sci USA 1994; 91:2051-2055.
glucose concentration range of 75-200 mg/dl, in the transgenic pancreas compared to that
of normal controls.
Thus, in spite of the considerable reduction in β cell GK activity, below the level that
gives rise to diabetes in MODY patients, and the reduced insulin secretory response to glu-
cose, in a manner similar to that observed in MODY patients,20 the RIP-GKRZ mice did not
develop overt diabetes. It is possible that in the mouse other insulin secretagogues can com-
pensate for the reduction in glucose-induced secretion. Alternatively, these results indicate
that an impaired liver function, in addition to that of β cells, is required for the induction of
overt diabetes by GK deficiency, as suggested by the gene disruption experiments.17-19 Such
liver impairments have been documented in MODY patients.21 The finding that partial
attenuation of expression of the normal GK protein is sufficient to impair the sensitivity of
β cells to glucose supports the interpretation of the dominance of the GK mutations in
MODY as a gene dosage effect, rather than a gain-of-function negative dominant effect of
the mutant protein.
The reduced β cell GK activity in the RIP-GKRZ mice may cause a predisposition to
diabetes. This may develop into overt disease in certain physiological conditions, such as
those caused by age, sex, weight, diet, and genetic background differences. Thus, these mice
provide an experimental system for studying the effect of such factors on the development
of type II diabetes.
A second example of an effective ribozyme application in transgenic mice is the target-
ing of β2-microglobulin (β2M) mRNA.22 This protein is required for antigen peptide pre-
sentation by class I major histocompatibility molecules on the surface of most mammalian
cells. A hammerhead ribozyme directed to exon 2 was expressed in transgenic mice under
the cytomegalovirus promoter. Expression was detected in lung, kidney and spleen, with
the greatest reduction (>90%) in β2M mRNA levels observed in the lungs, although consid-
erable variation was noted among individual mice in the same lineage. Unfortunately, the
phenotype of these mice has not been described.
Recently, mice were generated carrying a transgene that encodes a hammerhead

ribozyme directed against bovine α-lactalbumin (α-lac) under control of the mouse mam-
mary tumor virus long terminal repeat.23 The ribozyme was targeted to the 3' untranslated
region of the transcript and was flanked by 12-nucleotide segments of complementary se-
quence. The mice were crossed to a transgenic line over-expressing a bovine α-lac transgene
in the mammary gland. Expression of the ribozyme resulted in up to 78% reduction in
target mRNA and protein. The endogenous murine transcript and protein levels were not
affected, demonstrating the specificity of the approach.
Future Directions
These successful studies demonstrate the feasibility of employing ribozymes in the at-
tenuation of gene expression in vivo. Although they represent an encouraging beginning,
much work remains to be done to render this approach more efficient and widely appli-
cable. Better understanding of the ribozyme mechanism of action and the subcellular site of
action is needed to design improved targeting constructs. Ways to improve ribozyme syn-
thesis in the cells and increase ribozyme stability will result in enhanced function. The pa-
rameters guiding the choice of the optimal cleavage site within the target transcript need to
be established. Improved in vitro assays, in which the cleavage products can be preserved,
may assist in some of these tasks. In addition, a rigorous comparison of antisense and
ribozyme approaches is needed to establish their relative efficiency in individual systems.
In spite of the seemingly slow progress, this methodology continues to hold a great
promise for research and therapy alike. In combination with conditional gene expression
approaches, it should be possible to effectively utilize ribozymes for cell-specific and time-
specific downregulation of gene expression in vivo.
Acknowledgments
The research in my laboratory has been supported by the Juvenile Diabetes Founda-
tion International, by a Career Scientist Award from the Irma T. Hirschl Foundation and by
an NIDDK James A. Shannon Director’s Award.
References
1. Gu H, Marth JD, Orban PC et al. Deletion of a DNA polymerase beta gene segment in T
cells using cell type-specific gene targeting. Science 1994; 265:103-106.
2. Gossen M, Bujard H. Tight control of gene expression in mammalian cells by tetracycline-
responsive promoters. Proc Natl Acad Sci USA 1992; 89:5547-5551.
3. Sokol DL, Murray JD. Antisense and ribozyme constructs in transgenic animals. Transgen
Res 1996; 5:363-371.
4. Efrat S, Leiser M, Wu Y-J et al. Ribozyme-mediated attenuation of pancreatic β-cell glu-
cokinase expression in transgenic mice results in impaired glucose-induced insulin secre-
tion. Proc Natl Acad Sci USA 1994; 91:2051-2055.
5. Meglasson MD, Matschinsky FM. Pancreatic islet glucose metabolism and regulation of
insulin secretion. Diabetes Metab Rev 1986; 2:163-214.
6. Meglasson MD, Matschinsky FM. New perspectives in pancreatic islet glucokinase. Am J
Physiol 1984; 246:E1-13.
7. Vionnet N, Stoffel M, Takeda J et al. Nonsense mutation in the glucokinase gene causes
early-onset non-insulin-dependent diabetes mellitus. Nature 1992; 356:721-722.
8. Stoffel M, Froguel P, Takeda J et al. Human glucokinase gene: Isolation, characterization,
and identification of two missense mutations linked to early-onset non-insulin-dependent
(type 2) diabetes mellitus. Proc Natl Acad Sci USA 1992; 89:7698-7702.
9. Gidh-Jain M, Takeda J, Xu LZ at al. Glucokinase mutations associated with non-insulin-
dependent (type 2) diabetes mellitus have decreased enzymatic activity: Implications for
structure/function relationships. Proc Natl Acad Sci USA 1993; 90:1932-19336.
10. Takeda J, Gidh-Jain M, Xu LZ et al. Structure/function studies of human β-cell glucoki-

nase. J Biol Chem 1993; 268:15200-15204.
11. Iynedjian PB, Pilot PR, Nouspikel T et al. Differential expression and regulation of the
glucokinase gene in liver and islets of Langerhans. Proc Natl Acad Sci USA 1989;
86:7838-7842.
12. Liang Y, Najafi H, Matschinsky FM. Glucose regulates glucokinase activity in cultured is-
lets from rat pancreas. J Biol Chem 1990; 265:16863-16866.
13. Magnuson MA, Shelton KD. An alternate promoter in the glucokinase gene is active in the
pancreatic beta cell. J Biol Chem 1989; 264:15936-15942.
15. Hasellof J, Gerlach WL. Simple RNA enzymes with new and highly specific endoribonuclease
16. Strickland S, Huarte J, Belin D et al. Antisense RNA directed against the 3' noncoding
region prevents dormant RNA inactivation in mouse oocytes. Science 1988; 241:680-684.
17. Bali D, Svetlanov A, Lee H-W et al. Animal model for maturity-onset diabetes of the young
generated by disruption of the mouse glucokinase gene. J Biol Chem 1995; 270:21464-21467.
18. Grupe A, Hultgren B, Ryan A et al. Transgenic knockouts reveal a critical requirement for
pancreatic beta cell glucokinase in maintaining glucose homeostasis. Cell 1995; 83:69-78.
19. Terauchi Y, Sakura H, Yasuda K et al. Pancreatic beta-cell-specific targeted disruption of
glucokinase gene. Diabetes mellitus due to defective insulin secretion to glucose. J Biol
Chem 1995; 270:30253-30256.
20. Byrne M, Sturis J, Clement K et al. Insulin secretory abnormalities in subjects with hyper-
glycemia due to glucokinase mutations. J Clin Invest 1994; 93:1120-1130.
21. Sakura H, Kawamori R, Kubota M et al. Glucokinase gene mutations and impaired glu-
cose uptake by liver. Lancet 1993; 341:1532-1533.
22. Larsson S, Hotchkiss G, Andang M et al. Reduced beta 2-microglobulin mRNA levels in
transgenic mice expressing a designed hammerhead ribozyme. Nuc Acids Res 1994;
22:2242-2248.
23. L’Huillier PJ, Soulier S, Stinnakre MG et al. Efficient and specific ribozyme-mediated re-
duction of bovine alpha-lactalbumin expression in double transgenic mice. Proc Natl Acad
SCi USA 1996; 93:6698-6703.
CHAPTER 7
Retroviral Delivery of Ribozymes

Lun-Quan Sun and Geoff Symonds
Retroviral Properties and Life cycle
R etroviruses (RNA tumor viruses) have been used as vectors for gene transfer since the
early 1980s. They possess structural and enzymatic properties that make them suitable
vectors. These properties are summarized in Table 7.1. The long terminal repeats (LTRs) are
involved in integration of the provirus and contain promoter/enhancer elements to drive
proviral transcription. The viral enzymes reverse transcriptase, polymerase and integrase
are present within the virions and are essential to the retroviral life cycle. The structural
genes gag and env encode internal (Gag) and external (Env) viral proteins. Retroviruses
infect cells via the relevant receptors. While there is some degree of cellular specificity, their
ability to infect cells is quite broad. The retroviral life cycle is shown in Figure 7.1 and re-
viewed by Miller in ref. 1.
Retroviral Vectors and Their Design

It was shown in the late 1970s/early 1980s that internal portions of the retroviral ge-
nome could be deleted and cDNAs for a variety of genes (oncogenes, marker genes, poten-
tially therapeutic genes) could be inserted into the viral DNA. The use of packaging cell
lines (mouse and human) which supplied the missing vector functions in trans allowed the
generation of virions from the cDNA, following transfection of the recombinant DNA into
the packaging cell line.1 The retroviruses that were thereby produced were termed recombi-
nant retroviruses and, more recently, retroviral vectors (the latter term is now utilized to
describe the viral backbone plus the specific insert). The mode of production of retroviral
particles from engineered cDNA is shown in Figure 7.2. The variety of inserts utilized ex-
perimentally is summarized in Table 7.2.
A number of different retroviral vectors have been developed for gene transfer into
human cells, generally based on the Moloney murine leukemia virus (MoMLV) backbone.1
A wide variety of vector constructs from the original vectors, such as N2 and LNL6, have
been described. These vectors include those with the long terminal repeats (LTRs) as the
promoter of gene transcription,2 those with internal promoters,3 those with the gene of
interest expressed in the reverse orientation,4 those with self-inactivating (SIN) LTRs,5 those
with pol III and pol I promoters,6,7 those with the expression cassette inserted in the LTRs
(double-copy vectors)8 and those with modified LTRs.9 However, it has gradually become
apparent that as the complexity of design increases, the corresponding vector titers usually
decrease and there is an increasing propensity for retroviral rearrangement. This has tended
to lead the field back to simple one-gene vector designs. In our laboratory, a simple design
of a retroviral vector based on LNL6 has been consistently found to be very efficient in the
Fig. 7.1. Retroviral life cycle. The

retroviral virion binds to the cell
surface generally, through an ap-
propriate receptor. It is then un-
coated and the viral genomic
RNA, in association with the vi-
ral enzymes reverse transcriptase,
polymerase and integrase, enters
the cell. The RNA is reverse tran-
scribed to a cDNA copy which
becomes double stranded and
integrates into the host genome.
This proviral DNA is transcribed
as for other cellular genes into
RNA. The RNA is either mRNA
to produce viral proteins follow-
ing translation, or genomic RNA
which is packaged into virions
that bud from the cell surface.
This is generally (except in the
case of HIV and certain other
lentiviruses) non-pathogenic to
the host cell. The resultant viri-
ons can then infect further target
cells to reinitiate the process. In
the case of replication-incompe-
tent recombinant retroviruses,
there is a requirement for so-
called “helper” or replication-
competent retrovirus to supply
those functions missing from the
vector in trans. This can be ac-
complished to produce single-hit
retroviruses by the use of appro-
priate packaging cell lines (see
text and Fig. 7.2).
Table 7.1. Properties of retroviruses
1. Ability to accommodate cDNA inserts through genetic engineering in the cDNA form.
2. Virions containing modified genomic RNA can be produced through the transfection/
infection of relevant packaging cell lines.
3. Following infection, proviruses integrate at single to several copies into the target cell
genome and hence carried to progeny cells.
4. Insert cDNA is expressed from viral long terminal repeat or from internal promoters via
cellular transcriptional machinery.
5. Envelope can be modified to be cell targeted.
6. Non-toxic and can be modified to reduce the risk of replication competent retroviruses
(RCR) generation.
Retroviral Delivery of Ribozymes 89
Fig. 7.2. Production of retroviral vectors.

The retroviral construct is first made in the
DNA form by using a retroviral backbone,
generally MoMLV based, and transfecting
this into an ecotropic packaging cell line
(ψ2, ψcre). The retroviral pool is then used
to infect an amphotropic packaging cell
line (PA317, ψcrip, GP&E), and clonal cells
used to produce virions with a genome
corresponding to the initial DNA con-
struct.
Table 7.2. Retroviral cDNA inserts
1. Marker genes, e.g., neo, hygromycin to select cells and follow cell fate.
2. Potentially therapeutic genes, e.g., adenosine deaminase, β-globin, to impact on cell
phenotype.
3. Oncogenes, e.g., myc, erbB, to induce cell transformation/oncogenesis.
4. Tumor suppressor genes, e.g., p53, p21, Rb, to reverse transformation.
5. Ribozymes, antisense, decoys and transdominant protein genes to suppress gene
expression.
6. Suicide genes, e.g., HSV-tk, to kill cancer cells
transduction of T lymphocyte cell lines and human peripheral blood lymphocytes (PBLs)
in terms of vector titers, transduction efficiency as measured by G418 resistance, duration
of transgene expression and stability of transgene transcripts.
Cell Type Specific Promoters

Retroviral vectors are widely used vehicles for the effective delivery of ribozyme genes
constructs into mammalian cells. The lack of regulation of these expression vectors repre-
sents an obstacle for appropriate and controlled expression of foreign genes. The large number
of well-characterized regulatory elements controlling cell specific gene expression and the
identification of responsive/inducible elements within promoter sequences has laid the foun-
dation for utilizing these elements in the construction of retroviral vectors (Table 7.3). Sec-
ond and third generation retroviral vectors which harbor cell specific promoters or ele-
ments responding to regulatory signals represent an important component for safe, selective,
and controlled expression of therapeutic genes. Furthermore, exploitation of cell type spe-
cific promoter sequences for gene expression in the appropriate cell type could augment the
efficacy and stability of gene expression.
Packaging Cell Lines

Recent generation packaging cell lines have been designed to reduce the chance of gen-
eration of replication competent retrovirus (RCR). RCR have the potential to develop fol-
lowing extensive passaging in packaging cell lines. Such RCR, also termed “helper viruses”,
are generated through recombination events. To reduce the possibility of recombination,
the PA317 packaging cell line was designed with deletion not only of the ψ packaging signal
but also of two other regions in the 5' and 3' LTR.10 Also, a modified version N2 vector,
LNL6,11 was designed. The start codon of the gag gene in the previous vector N2 was re-
placed by a stop codon; some of the MoMLV sequences of LTR in N2 vector were replaced
by Moloney murine sarcoma virus (MSV) to decrease the homology between the vector
and helper virus genome. This combination (LNL6 and PA317) has been used in most of
the clinical studies to date. Other alterations of packaging cell lines (Ψ-cre and Ψ-crip) in-
clude deletion of the packaging signal (as previously) as well as expression of the gag-pol
and env genes on two separate plasmids to decrease the possibility of recombination.12 More
recently, transient packaging cell line systems have been developed, including the PHOE-
NIX system.13
Targeting of Retroviruses
Research has been conducted into cell-specific retroviral targeting by modifications in
viral envelope protein sequences, using antibodies as specific mediators in viral infection,
and modified ligands or receptors. The strategies for modification of the target cell specific-
ity of retroviral vectors are summarized as follows. Generally this involves re-engineering
the env gene by replacing segments with epitopes that recognize receptors other than those
normally used by the virus. This is achieved by manipulation of the env gene in the proviral
construct of the packaging cell line.14
Antibody-Mediated Binding
Retroviral vector particles can be engineered that contain an antigen binding site of an
Ab molecule (single chain antigen binding proteins, scFv) in place of the natural retroviral
receptor binding peptide.15 This single chain antibody can be targeted to specific cellular
membrane proteins. An antibody (Ab) bridge can thus be engineered between a retroviral
vector and the designated target cell that does not contain a receptor for the virus. In addi-
tion, antibodies directed against a specific known protein that is located on the target cell
Table 7.3. Cell-specific promoters and their target cells.
Promoters Target cells
β-Globin promoter/LCR Erythroid cells

Immunoglobin promoters B lymphoma
CD11a, b promoters Leucocytes
Albumin promoter Hepatocytes
HIV-LTR HIV-infected T cells
Tat/Rev responsive elements CD4+ T cells
CEA promoter Colon and lung carcinomas
AFP promoter Hepatocellular carcinomas
Tyrosinase promoter Melanomas
MMTV-LTR Mammary carcinoma
Egr-1 promoter Irradiated tumors
HSP70 promoter Tumor treated with hyperthermy
surface can be linked by streptavidin to antibodies against viral Env proteins. The Ab-
streptavidin complex functions as a bridge to link the virus to its target cells via the novel,
specific receptor.
Asialoglycosylation
The retroviral Env proteins can be chemically modified so that they are recognized by
receptors expressed on the surface of the target cell. One such example is to asialate the Env
proteins by the coupling of lactose. By this means, the hepatocyte-specific asialoglycoprotein
receptors can be targeted using modified retroviral vectors.16
Pseudotyped Vectors
This is based on the concept that cell tropism is determined by the source of the viral
Env protein present on the vector virions. One such example is the recently developed pack-
aging cell line, PG13.17 This packaging cell line can pseudotype the MoMLV vector with the
gibbon ape leukemia virus (GALV) envelope, and these pseudotyped retrovirus vectors ap-
pear able to infect a larger number of human cells more efficiently than comparable MoMLV-
based amphotropic retroviruses.
Ribozyme Action
As noted in the previous chapters, ribozymes are RNA molecules with catalytic proper-
ties enabling them to cleave target RNA substrates.18-25 The two main types of ribozymes are
hairpin and hammerhead.21,22 They differ in their structure, but each has been shown to be
amenable to modifications in which the substrate and catalytic components are separated
to enable cleavage in trans.21,22 Several features of ribozymes make them attractive as poten-
tial therapeutic agents—in particular, their specificity of cleavage and their potential ability
to cleave multiple substrate molecules. Having shown their ability to cleave in vitro-derived
substrates, ribozyme DNA constructs can then be incorporated into a variety of vectors.
Such vectors include relatively simple expression vectors in which the ribozyme is incorpo-
rated downstream of a mammalian promoter, generally as a chimeric gene construct, or in
more complicated vectors which are based on the genome of retroviruses or other viruses.
The ribozyme is engineered in the DNA form by complementarity to the target se-
quence, which in the case of hammerhead ribozymes possesses a GUX or, in certain cases,
NUX (where N represents any nucleotide and X represents A, C or U) motif. The cDNA is
incorporated into the retroviral cDNA and the new genomic material is produced following
transfection into the packaging cell line. We, and others, have used this approach to intro-
duce a variety of ribozymes into retroviruses. We have previously detailed the production of
anti-HIV retroviruses.26
As to stability of the ribozyme within cells, it is likely that many factors contribute.
These include structure of the RNA transcript, the cellular compartment in which ribozyme
transcripts are produced and transported, and whether the ribozyme is embedded within
another gene. To date, tRNA, U1 or U6 snRNA have been used as expression cassettes for
ribozymes. We consistently found that the insertion of a ribozyme sequence into the 3'
untranslated region of the neomycin resistance gene within an expression vector led to a
high level expression of the neo/ribozyme chimeric RNA transcript.27,28
Ribozymes have been shown to affect the expression of a number of genes in tissue
culture systems. Various read-outs can be utilized to monitor the effectiveness of ribozyme
action. Such read-outs include modulation of gene expression (structural genes, cytokines)
monitored by Northern or Western blots, modification of cellular phenotype as measured
by microscopy or FACS, and inhibition of viral replication determined by ELISA or viral
RNA dot-blots. Once introduced into cell lines or primary cells, ribozyme-mediated inhibi-
tion can be assayed in appropriate test systems. Two major systems where ribozymes have
shown efficacy in tissue culture are:
1. the reversion of the transformed cellular phenotype; and
2. the inhibition of HIV replication.
The overall construction and testing procedure used by this group for such studies is
outlined in Figure 7.3. Reversion of the transformed cellular phenotype is based on ribozyme
mediated downregulation of oncogene expression. In inhibiting HIV replication, the
ribozymes act to cleave genomic or sub-genomic mRNA molecules of HIV. Ribozymes that
have been utilized for these two purposes are detailed at length in other chapters of this
book. The following focuses on one example of retroviral delivery of ribozymes—the one
that has been used most extensively to date—the use of ribozymes targeted to inhibition of
retroviral replication, in particular, that of HIV.
Murine Model Systems for Retroviral-Ribozyme Delivery

and Inhibition of Gene Expression
By using a Moloney murine leukemia virus (MoMLV) model system, we have shown
that the ψ packaging site, the site essential for packaging of viral genomic RNA, is an effec-
tive ribozyme target site and can be used to effectively inhibit MoMLV replication.27 The
efficiency of substrate cleavage by these ψ packaging site-targeted ribozymes correlated with
the ability of the corresponding ribozyme expression constructs to inhibit MoMLV replica-
tion.27 We have used retroviral vectors containing ribozymes to this site to demonstrate the
utility of retroviral vectors (and the specificity of the packaging site as a target) by demon-
strating that ribozymes could be designed to specifically cleave a MoMLV ψ packaging site
target, though not the MoMSV variant sequence, within the retroviral vector itself. This was
despite the high degree of homology between the two ψ packaging site sequences. Virus
could be produced from the recombinant ribozyme containing provirus (without an ap-
parent inhibition of vector titer) and cells transduced with this ribozyme-containing virus
were shown to inhibit target MoMLV replication.29 This shows the utility of retroviral-
ribozyme constructs.
Fig. 7.3. Strategy of construction and testing of anti-viral ribozyme constructs. Ribozymes are
designed based on the accessibility of their corresponding target RNA. This can be assessed by
computer modeling and/or in vitro assays such as RNAase mapping. After testing for in vitro
cleavage, the ribozymes are cloned into expression or retroviral vectors. They can then be tested
in different cell systems using the various read-outs as shown.
Anti-HIV Retroviral Ribozyme Constructs

A number of studies have employed ribozymes to show effective inhibition of HIV
replication. These studies have been previously summarized both by ourselves and other
groups and are briefly outlined here.
Sarver and colleagues used a ribozyme construct targeted to the HIV-1 gag sequence
and showed inhibition of HIV replication.30 The retroviral (MoMLV) transfer of HIV-1 (5'
leader sequence) targeted ribozymes into a stable T cell line (MT4) resulted in resistance to
HIV-1 infection.31 Other investigators used a ribozyme construct engineered to specifically
cleave the HIV-1 tat gene.32 Retrovirally transduced cells showed a delay in measurable HIV
p24 levels—15 days compared to 7 days in control cells. In addition, a hairpin ribozyme
expressed from a tRNAVal transcriptional cassette within various expression vectors, includ-
ing retroviruses, was shown to confer resistance to several HIV-1 isolates.33-38
In our own experiments we used ribozyme constructs in which the ribozyme was:
1. cloned into the 3' region of the neo gene and driven by the SV40 promoter in an
expression vector; or
2. inserted into a MoMLV-based retroviral vector.
Using ribozymes which were targeted against either the HIV tat gene or the ψ packag-
ing site of HIV-1,27,28,39 we observed protection of transfected or retrovirally transduced
T lymphocyte (SupT1) cells from HIV-1 infection—both in terms of delay of HIV-1 repli-
cation and absolute virus levels (assayed by syncytia and p24 ELISA antigen assay). Replica-
tion of the virus was inhibited by 70-95% for the laboratory adapted HIV-1 isolates (SF2,
IIIB) and by 2-4 logs for primary clinical isolates. The most effective ribozyme in both
expression systems was one directed to the first coding exon of tat, termed Rz2; within the
MoMLV vector LNL6 this is termed RRz2.27,28,39 These results were borne out in a second T
lymphocyte cell line, a pooled CEM T4 cell line system, in which the most effective con-
struct was RRz2. This construct elicited an anti-viral effect, reducing p24 antigen levels by
70-80% compared to vector/marker- alone (LNL6) controls. An antisense retroviral con-
struct termed RASH-5 with sequences complementary to a 550 base 5' region of the HIV-1
genome exhibited a similar level of inhibition (Smythe, Sun, Pyati, Gerlach, Symonds, un-
published results).
These results using retroviral-ribozyme constructs were confirmed in non-HIV in-
fected, normal peripheral blood lymphocytes (PBLs)—both total and CD4+ enriched. The
RRz2 retroviral construct, shown to be effective in the pooled T lymphocyte assay, was also
effective (70-90% inhibition) in this assay. Other ribozyme-containing constructs were some-
what less effective. For the retroviral assay systems, another recombinant retrovirus, engi-
neered to contain polyTAR—a construct similar to that shown by others to be effective in T
cell lines,40 was used as a positive control. In the PBL assay, this polyTAR retrovirus was
found to be somewhat less efficient than the Rz2 and antisense viruses.28 In addition, RRz2
and LNL6 vectors were also used to transduce PBLs from HIV-1 infected patients. Paired
analysis showed that cell viability in the ribozyme-transduced HIV-1 infected PBLs was
significantly higher than that in the vector-transduced cells. This difference in viability (be-
tween RRz2 and LNL6 transduced cells) was not observed in PBLs from non-infected
donors.40a This observation is the first evidence that a ribozyme can impact on the survival
of HIV-1 infected patient-derived PBLs in cell culture, and implies that the ribozyme-ex-
pressing cells may have a growth viability advantage over non-transduced cells within HIV-1-
infected patients.
Recent work from this group has confirmed the specificity of ribozyme action. In a
study addressing this issue, a different ribozyme, Rz1, targeted to the 5' splicing region of
the tat gene was designed to cleave GUC N, in which N is G in HIV-1 IIIB and N is A in
HIV-1 SF2. The data from both in vitro and in vivo studies showed that the ribozyme could
protect cells from challenge by only those HIV isolates whose genomic sequence was cleav-
able in vitro, and demonstrated the importance of the first base pair distal to the NUX
within helix I of the hammerhead structure for both in vitro and in vivo ribozyme activities.39
To address issues of the impact of ribozyme expression on the viral population, includ-
ing virus sequence integrity, a multiple-passage assay was developed in our group to analyze
HIV-1 sequence variation and viral replication dynamics in ribozyme-expressing cells. The
results demonstrated that Rz2 ribozyme expression in transduced human T cells yields:
1. no mutations within the ribozyme targeting region over five sequential viral pas-
sages; and
2. rapid disappearance of certain quasi-species of HIV-1.40a This further indicates the
potential for clinical use of an anti-HIV ribozyme.
Taken together, these results show that retroviral-ribozyme constructs effectively in-
hibit HIV-1 replication.
General Aspects of Retroviral and Other Delivery Systems

Probably the major hurdle to be overcome in gene therapy is the efficient delivery of
the therapeutic gene(s) into relevant cell populations within patients. At present, retroviral
vectors are generally the delivery system of choice due to their relative efficiency, observed
safety, stable integration and persistent expression.1,41,42 Well over 150 gene therapy clinical
trials have been approved in the USA to date and, of these, a large majority rely on a retroviral
vector to carry marker or therapeutic genes into the target cell genome. Most of these viral
vectors are derived from MoMLV.1 However, retroviral vectors also have some limitations.
Their host range may be restricted (to date amphotropic retroviruses have generally been
used) and, probably more importantly, they require cell division for integration of their
genomic material.
Recently, advances have been made by investigators17,43,44 who described the construc-
tion and use of gibbon ape leukemia virus (GaLV) components or vesicular somatititis vi-
rus (VSV) G-glycoprotein pseudotyped retroviral vectors. These vectors appear able to some-
what more efficiently infect a larger number of human cells than amphotropic virus, thereby
potentially yielding higher transduction efficiencies compared to MoMLV-based
amphotropic retroviruses.45 However, to a certain extent, lowering the viral harvest tem-
perature to 32°C and using concentrated virus (both increasing the effective MOI) to ini-
tially infect the cells, can mitigate this apparent advantage by acting to increase amphotropic
viral transduction efficiency.
As previously discussed, both the retroviral cDNA vector backbone and the “helper”
provirus within the various packaging cell lines have been manipulated to minimize the risk
of generating replication competent retrovirus (RCR). This has been accomplished by de-
leting portions of the two retroviral genomes (backbone and “helper”) to minimize the
chance of recombination.1,41,42 As RCR can cause insertional mutagenesis, the use of
retroviruses for gene therapy in humans is dependent on demonstration of lack of RCR.
Assays have been developed to show whether or not RCR exist within retroviral pools.
Other potential delivery systems for ribozymes include adeno-associated virus (AAV),
which has been shown to be capable of stable and efficient insertion of other anti-HIV-1
genes into hematopoietic cells.46 Chatterjee and colleagues have reported the use of AAV to
transduce an HIV-1 antisense construct into human T cells, resulting in a significant reduc-
tion of HIV-1 levels following challenge of the transduced cells.47 There are several poten-
tial advantages of AAV vectors that are relevant for HIV-1 gene therapy.48 They appear to
have no potential for homologous recombination, and have high transduction efficiency
for hematopoietic cells, including progenitor cells. In addition, AAV is non-pathogenic and
replication-incompetent, and non-dividing cells can be transduced. However, for practical
application, there are still three main problems:
1. a limit of approximately 4.7 kb for insertion of therapeutic genes (though similar to
retroviruses);
2. potential adenovirus contamination; and
3. the possible excision or instability of the integrated gene.
Another approach being investigated is the use of another retroviral system,49 namely
HIV-1 based vectors, in which the positive aspects:
1. HIV-1 target cell specific expression;
2. ability to infect non-dividing cells; and
3. possibility of tat inducible expression,
need to be balanced by potential negatives:
1. possibility of recombination with HIV-1; and
2. a present lack of robust producer cell lines.
Adenoviruses have also been used as gene delivery vectors; adenoviral delivery of
ribozymes is dealt with in other chapters.
Clinical Application of Retroviral Delivery of Ribozymes

Ex Vivo Transduction of Peripheral Blood Lymphocytes in Gene Therapy
The concepts for a genetic type of therapy were developed almost 20 years ago and
have recently been transformed (albeit with limited success) into clinical reality.1 In 1990
the first gene therapy study for the treatment of ADA deficiency began, followed by a rap-
idly growing number of clinical gene therapy trials, which covered diseases caused by ge-
netic disorders, viral infections, and malignancies.1
For the clinical application of gene therapy for AIDS, a simple approach to treating
HIV-infected patients is the infusion of transduced and “protected” CD4+ PBLs. Recent
findings about how the AIDS virus population behaves in the body suggest that the battle
between the immune system and the virus is so evenly matched that any approaches which
weaken the virus replication and give the immune system a slight but maintained edge
might be enough to show a significant effect.50,51 Therefore, if ribozyme constructs can pro-
tect CD4+ T cells from HIV-1 infection and its sequelae in patients, the decline in the num-
bers of CD4+ T cells could be halted or even potentially reversed and HIV-infected indi-
viduals could benefit clinically. It is relevant that the half life of an HIV infected and producing
CD4+ T lymphocyte is of the order of only two days.50,51 Recently, we and others proposed
and are conducting clinical trials of HIV-1 gene therapy. Although it is only Phase I studies
that are underway for ribozymes, overall these clinical trials will address such issues as safety,
efficacy and the issue of potential emergence of escape mutants in HIV-1 infected patients.
A scheme for gene transfer into PBLs in a clinical application is outlined in Figure 7.4.
Patient PBLs are collected by apheresis. After separation on a Ficoll/Hypaque gradient, cells
are depleted of the CD8+ subpopulation of T cells by binding to antibody-coated plates.
Enriched PBLs are then stimulated with OKT-3 antibody. Following stimulation, PBLs are
transduced with retroviral vectors in the presence of interleukin-2. This is the scheme that
we have employed for a ribozyme-based clinical trial in AIDS.
Currently, there are several procedures being used in the transduction and expansion
of PBLs. The simplest is to use tissue culture flasks for both transduction and expansion,
such as the Lifecell Tissue Culture Flask of Baxter.52 This method is easy to use, and rela-
tively inexpensive, but is very labor intensive. Furthermore, since this system is not a closed
one, there is a potential risk of contamination.
The second protocol is to transduce and expand PBLs in tissue culture bags (Wong-
Staal et al, Clinical protocol (Ribozyme), NIH RAC). This is very similar to that using tissue
culture flasks except that a large volume of media is used (normally about 0.5 to 1 liter) to
produce the requisite number of cells. Recently, closed Artificial Capillary Systems have
been developed for transduction of PBLs and for large scale cell manipulation for clinical
therapy.53 These systems, such as the CELLMAX system of Cellco, consist of an electroni-
cally controlled pump that perfuses culture medium through a capillary cartridge designed
for the growth of a particular cell type. The culture cartridge used for PBLs is the ‘Peripheral
Blood Lymphocyte-Moderate Pore Size’ which has been specifically designed to support the
growth of OKT-3 stimulated human mononuclear lymphocytes. Up to 5 x 109 lymphoid
cells can be obtained from each cartridge. The advantages of this system over conventional
methods are:
1. it permits optimum expansion of PBLs with minimal operator manipulation and
interference, thus having less chance of potential contamination and cell stress; and
Fig. 7.4. Retrovirus-mediated gene

transfer of CD4+ cells in a clinical
setting. Peripheral blood mono-
nuclear cells are isolated from the
patient’s or donor’s blood by
apheresis followed by Ficoll-
Hypaque. Following a purification
step to deplete CD8+ T lympho-
cytes, they are stimulated with an
agent such as OKT-3 and ex-
panded with the T cell cytokine
interleukin-2. They are then trans-
duced with the retrovirus, ex-
panded further and infused into
the recipient.
2. it allows for the concentration of retroviral vectors in the closed system, that is, a
greater effective multiplicity of infection (MOI)
The latter appears to be an important issue for PBL transduction.
Ex Vivo Transduction of Human CD34+ Cells

Gene therapy protocols targeting human stem cells (HSC) have primarily used recom-
binant retroviruses as delivery vehicles. HSC-based therapies require long-term expression
of the transgene, limiting the choice of vector to those which allow stable integration into
the host genome. The success of gene transfer in the murine system has led to in vitro and in
vivo studies with HSCs and the initiation of clinical trials. High levels of gene transfer rou-
tinely reported for murine progenitors were achieved by co-cultivation of target cells with
the retroviral producer cells. While transduction by co-cultivation has been shown to be
superior to cell-free viral supernatant,54 co-cultivation is generally not suitable for clinical
protocols due to the potential risk of contamination of cellular infusions with the producer
cells.55 Strategies to overcome this problem have included a high MOI and multiple rounds
of infection with viral supernatant, which is now standard in most clinical gene therapy
protocols. Fibronectin fragments are also under trial.
The challenge in this area is to identify culture conditions which will induce cell cy-
cling, thus rendering the cells susceptible to infection, while maintaining pluripotentiality.
Most clinical protocols include overnight pre-culture in media supplemented with the growth
factors interleukin-3, interleukin-6 and stem cell factor (SCF) to induce cell cycling, fol-
lowed by addition of retroviral supernatant daily for 2-4 days. The primary concern has
been to minimize the time in ex vivo culture to limit the degree of differentiation induced
by the recombinant cytokines. Recent studies have shown that FLT-3/FLK-2 ligand and KIT
ligand, as well as megakaryocyte growth and development factor (MGDF) are capable of
expanding progenitor cell numbers while maintaining the reconstituting ability of the
cells.56,57
Concluding Remarks
Ribozymes cleave oncogenes to suppress cell transformation and cleave genomic and
subgenomic HIV to suppress the latter’s replication. This has been shown in cell culture
systems. The key question is: Can ribozymes affect the disease course of cancer and AIDS?
In the case of AIDS, can they impact on the two surrogate markers of advancing disease,
viral load and CD4+ T cell counts? Initial approaches have been made to address these ques-
tions in Phase I safety clinical trials presently being conducted, or about to be conducted, by
a number of groups including our own.
Ribozymes are but one of several possible gene therapeutic approaches. It can be ex-
pected that continued improvement in the design of antigenes such as ribozymes will aug-
ment the potential for gene therapeutic approaches to AIDS, cancer or some genetic dis-
eases. Concomitantly, improvements in delivery systems, the use of combinations of different
strategies (multi-targeted gene therapy, drugs, immune modulation), as well as specific means
for the introduction into appropriate target tissue or cells, will result in realization of the
potential clinical applications of ribozyme-based gene therapy.
Acknowledgments
We thank Halley Hanlen for assistance with preparation of this manuscript and Janet
Macpherson for preparation of the figures. Much of the work described herein from the
authors’ laboratories was funded by contract with Gene Shears Pty Ltd, Australia.
References
1. Miller AD. Human gene therapy comes of age. Nature 1992; 357:455-459.
2. Hock RA, Miller AD, Osborne WRA. Expression of human adenosine deaminase from vari-
ous strong promoters after gene transfer into human haematopoietic cell lines. Blood 1989;
74:876-881.
3. Overell RW, Wesser KE, Cosman D. Stably transmitted triple-promoter retroviral vectors
and their use in transformation of primary mammalian cells. Mol Cell Biol 1988;
8:1803-1808.
4. Dzierzak EA, Papayannopoulou T, Mulligan RC. Lineage-specific expression of a human
β-globin gene in murine bone marrow transplant recipients reconstituted with retrovirus-
transduced stem cells. Nature 1988; 331:35-41.
5. Yee JK, Moores JC, Jolly DJ et al. Gene expression from transcriptionally disabled retroviral
vectors. Proc Natl Acad Sci USA 1987; 84:5179-5201.
6. Sullenger BA, Lee TC, Smith CA et al. Expression of chimeric tRNA-driven antisense tran-
scripts renders NIH 3T3 cells highly resistant to Moloney murine leukemia virus replica-
tion. Mol Cell Biol 1990; 10:6512-6523.
7. Palmer TD, Miller AD, Reeder RH et al. Efficient expression of a protein coding gene
under the control of an RNA polymerase I promoter. Nucleic Acids Res 1993; 21:3451-3457.
8. Hantzopoulos PA, Sullenger BA, Ungers G et al. Improved gene expression upon transfer
of the adenosine deaminase minigene outside the transcriptional unit of a retroviral vec-
tor. Proc Natl Acad Sci USA 1989; 86:3519-3523.
9. Hawley RG, Sabourin LA, Hawley TS. An improved retroviral vector for gene transfer into
undifferentiated cells. Nucleic Acids Res 1989; 17:4001.
10. Miller AD, Buttimore C. Design of retrovirus packaging cells to avoid recombination lead-
ing to helper virus production. Mol Cell Biol 1986; 6:2895-2902.
11. Bender MA, Palmer TD, Gelinas RE et al. Evidence that the packaging signal of Moloney
leukemia virus extends into the gag region. J Virol 1987; 61:1639-1646.
12. Danos O, Mulligan RC. Safe and efficient generation of recombinant retroviruses with
amphotropic and ecotropic host ranges. Proc Natl Acad Sci USA 1988; 85:6460-6464.
13. Pear WS, Nolan GP, Scott ML et al. Production of high-titer retroviruses by transient trans-
fection. Proc Natl Acad Sci USA 1993; 90:8392-8396.
14. Bushman F. Targeting retroviral integration. Science 1995; 267:1443-1444.
15. Etienne-Julan M, Roux P, Carillo S et al. The efficiency of cell targeting by recombinant
retroviruses depends on the nature of the receptor and the composition of the artificial
cell-virus linker. J Gen Virol 1992; 73:3251-3255.
16. Neda H, Wu CH, Wu GY. Chemical modification of an ecotropic murine leukemia virus
results in redirection of its target cell specificity. J Biol Chem 1991; 266:14143-14146.
17. Miller AD, Garcia JV, von Suhr N et al. Construction and properties of retrovirus packag-
ing cells based on gibbon ape leukemia virus. J Virol 1991; 65:2220-2224.
18. Cech T. The chemistry of self-splicing RNA and RNA enzymes. Science 1987; 236:1532-1539.
19. Uhlenbeck OC. A small catalytic oligoribonucleotide. Nature (London) 1987; 328:596-600.
20. Forster A, Symons R. Self-cleavage of plus and minus RNAs of a virusoid and a structural
model for the active sites. Cell 1987; 49:211-220.
21. Hampel A, Nesbitt S, Tritz R et al. The hairpin ribozyme. Meth Enzymol 1987; 5:37-42.
22. Haseloff J, Gerlach WL. Simple enzymes with new and highly specific endoribonuclease
activity. Nature (London) 1988; 334:585-591.
EMBO J 1996; 14:3751-3757.
24. Rossi JJ. Ribozymes. Curr Opin Biotech 1992; 3:3-7.
25. James W, Al-Shamkhani A. RNA enzymes as tools for gene ablation. Curr Opin Biotec
1995; 6:44-49.
26. Sun LQ, Gerlach WL, Symonds G. The design, production and validation of an anti-HIV
type 1 ribozyme. Methods in Molecular Medicine 1998; 4:51-64.
27. Sun LQ, Warrilow D, Wang L et al. Ribozyme-mediated suppression of Moloney murine
leukemia virus and human immunodeficiency virus type 1 replication in permissive cell
lines. Proc Natl Acad Sci USA 1994; 91:9715-9719.
28. Sun LQ, Pyati J, Wang L et al. Resistance to HIV-1 infection conferred by transduction of
human peripheral blood lymphocytes with ribozyme, antisense or polyTAR constructs. Proc
Natl Acad Sci USA 1995; 92:7272-7276.
29. Lowenstein P, Symonds G. Inhibition of Moloney murine leukemia virus (MoMLV) by a
retroviral vector LNL6, carrying ribozymes, targeted to the 5' non-coding sequence. J Gen
Virol 1997; 78:2587-2590.
30. Sarver N, Cantin EM, Chang PS et al. Ribozymes as potential anti-HIV-1 therapeutic agents.
Science 1990; 247:1222-1225.
31. Weerasinghe M, Liem SE, Asad S et al. Resistance to human immunodeficiency virus type
1 (HIV-1) infection in human CD4+ lymphocyte-derived cell lines conferred by using
retroviral vectors expressing an HIV-1 RNA-specific ribozyme. J Viol 1991; 65:5531-5534.
32. Lo KMS, Biasolo MA, Dehni G et al. Inhibition of replication of HIV-1 by retroviral vec-
tors expressing tat-antisense and anti-tat ribozyme RNA. Virology 1992; 190:176-183.
33. Leavitt MC, Yu M, Yamad O et al. Transfer of an anti-HIV-1 ribozyme gene into primary
human lymphocytes. Human Gene Ther 1994; 5:1115-1120.
34. Yu M, Ojwang JO, Yamada O et al. A hairpin ribozyme inhibits expression of diverse
strains of human immunodeficiency virus type 1. Proc Natl Acad Sci USA 1993;
90:6340-6344.
35. Ojwang JO, Hampel A, Looney DJ et al. Inhibition of human immunodeficiency virus type
1 expression by a hairpin ribozyme. Proc Natl Acad Sci USA 1992; 89:10802-10806.
36. Yu M, Leavitt MC, Maruyama M et al. Intracellular immunization of human fetal cord
blood stem/progenitor cells with a ribozyme against human immunodeficiency virus type
1. Proc Natl Acad Sci USA 1995; 92:699-703.
37. Yu M, Poeschla E, Yamada O et al. In vitro and in vivo characterization of a second func-
tional hairpin ribozyme against HIV-1. Virology 1995; 206:381-386.
38. Yamada O, Kraus G, Leavitt MC et al. Activity and cleavage site specificity of an anti-
HIV-1 hairpin ribozyme in human T cells. Virology 1994; 205:121-26.
39. Sun LQ, Wang L, Gerlach WL et al. Target sequence-specific inhibition of HIV-1 replica-
tion by ribozymes directed to tat RNA. Nucleic Acids Res 1995; 23:2909-2913.
40. Lisziewicz J, Sun D, Smythe J et al. Inhibition of human immunodeficiency virus type 1
replication by regulated expression of a polymeric Tat activation response RNA decoy as a
strategy for gene therapy in AIDS. Proc Natl Acad Sci USA 1993; 90: 8000-8004.
40a. Wang L, Witherington C, King A, Gerlach W, Carr A, Penny R, Cooper D, Symonds C,

Sun L-Q. Preclinical characterization of an anti-tat ribozyme for therapeutic application.
Human Gene Therapy 1998; 9:1283-1291.
41. Miller AD, Rosman GJ. Improved vectors for gene transfer and expression. BioTechniques
1989; 7:980-986.
42. Jolly D. (1994) Viral vector systems for gene therapy. Cancer Gene Ther 1994; 1:51-64.
42. Kavanaugh MP, Miller DG, Zhang W et al. Cell-surface receptors for gibbon ape leukemia
virus and amphotropic retrovirus are inducible sodium-dependent phosphate symporters.
44. Bums JC, Friedmann T, Driever W et al. Vesicular stomatititis virus G glycoprotein
pseudotyped retroviral vectors: Concentration to very high titer and efficient gene transfer
into nonmammalian cells. Proc Natl Acad Sci USA 1993; 90:8033-8037.
45. Von kalle C, Kiem HP, Goehle SO et al. Increased gene transfer into human haematopoietic
progenitor cells by extended in vitro exposure to pseudotyped retroviral vector. Blood 1994;
84:2890-2897.
46. Kuzcyzka N. Use of adeno-associated virus as a general transduction vector in mammalian
cells. Curr Top Microbiol Imuunol 1992; 158:97-129.
47. Chatterjee S, Johnson PR,Wong KK Jr. Dual-target inhibition of HIV-1 in vitro by means
of an adeno-associated virus antisense vector. Science 1992; 258:1485-1488.
48. Carter DJ. Adeno-associated virus vectors. Curr Opin Biotechnol 1992; 3:533-539.
49. Lever AML. HIV and other lentivirus-based vectors. Gene Therapy 1996; 3:470-471.
50. Ho DD, Neumann AU, Perelson AS et al. Rapid turnover of plasma virions and CD4 lym-
phocytes in HIV-1 infection. Nature (London) 1995; 373:123-126.
51. Wei X, Ghosh SK, Taylor ME et al. Viral dynamics in human immunodeficiency virus
type 1 infection. Nature (London) 1995; 373:117-122.
52. Woffendin C, Ranga U, Yang Z-Y et al. Expression of a protective gene prolongs survival
of T cells in human immunodeficiency virus-infected patients. Proc Natl Acad Sci USA
1996; 93:2889-2894.
53. Packard BS. Mitogenic stimulation of human tumor-filtrating lymphocytes by secreted fac-
tors from human tumor cell lines. Proc Natl Acad Sci USA 1990; 87:4058-4062.
54. Havenga M, Hoogerbrugge P, Valerio D et al. Retroviral stem cell gene therapy. Stem Cell
1997; 15:162-179.
55. Gordon MY. Adhesive properties of haemopoietic stem cells. Br J Haematol 1988; 68:149.
56. Yonemura Y, Ku H, Ltman SD et al. In vitro expression of hematopoietic progenitors and
maintenance of stem cells: Comparison between FLT-3/FLK-2 ligand and KIT ligand. Blood
1997; 89:1915-1921.
57. Rasko JE, O’Flaherty E, Begley CG. Mpl ligand (MGDF) alone in combination with stem
cell factor (SCF) promotes proliferation and survival of human megakaryocyte, erythroid
and granulocyte/macrophage progenitors. Stem Cells 1997; 15:33-42.
58. Amado RG, Symonds G, Jamieson BD, Zhao G, Rosenblatt JD, Zack JA. Effects of mega-
karyocyte growth and development factor on survival and retroviral transduction of T-
lymphoid progenitor cells. Human Gene Therapy 1998; 9:173-183.
CHAPTER 8
Adeno-Associated Virus (AAV)

Mediated Ribozyme Expression
in Mammalian Cells
Piruz Nahreini and Beth Roberts
Introduction
G ene therapy has increasingly become an ideal method for the treatment of many heredi-
tary and acquired human diseases.1 Efficient delivery, regulated expression, low toxicity,
lack of adverse immune responses, and in most cases long-term stable expression of a thera-
peutic gene in target cells are desired features for a successful gene therapy. Several viral and
nonviral methods of gene delivery into cells have been developed, but none has a universal
application for diseases amenable to genetic treatment.2 Each method, however, has its unique
set of advantages for a given human disorder.
For example, adenoviral vectors are very efficient in gene delivery.3 However, the ex-
pression of a therapeutic gene is sustained for only several weeks because these vectors do
not integrate stably into the chromosomal DNA of target cells. Therefore, adenovirus is a
suitable vector if transient expression of a therapeutic gene is sufficient to ameliorate a
disease state, provided that the associated immunogenicity of these vectors does not inter-
fere with the treatment process. On the other hand, retrovirus and adeno-associated virus
(AAV), by virtue of integrating into chromosomal DNA, are suitable vectors when stable,
long-term expression of a therapeutic gene is preferred.4-6 Retroviral vectors are dependent
on cell mitosis for efficient integration into chromosomal DNA, and this explains the inef-
ficient expression of a gene of interest in non-dividing cells.7 A new generation of retroviral
vectors, the lentiviruses, have recently been reported to efficiently express a transgene in
non-dividing cells.8 The strong enhancer and promoter of retroviral long terminal repeats
(LTR) can potentially be a problem when a regulated expression of a therapeutic gene is
warranted. When regulated, long-term expression in non-dividing, as well as dividing cells,
is desired, AAV may be the ideal vector.
In this chapter, we focus on the utility of AAV as a vector for gene expression, and
emphasize certain issues which have direct impact on the feasibility of its broader use in
human gene therapy. We report here some of our own observations with regard to vector
production and transgene expression, particularly hammerhead ribozyme expression. These
small, catalytic RNA molecules are potentially attractive for gene therapy for a variety of
human disorders. The expression of a ribozyme using retroviral and adenoviral vectors has
been reported in several studies;9-13 however the utility of AAV for ribozyme expression has
not been investigated extensively.
AAV Life Cycle and Genetics

Adeno-associated virus (AAV) is a small non-pathogenic human parvovirus which in-
fects a variety of mammalian cells with a broad host range.4-5 The AAV genome (4681 bases)
is a single-stranded DNA (ss-DNA) of either polarity flanked by inverted terminal repeats
(ITRs). Plus and minus viral strands are encapsidated into mature virions at the same fre-
quency. AAV can establish a lytic or latent infection in mammalian cells in the presence or
absence of a helper virus, respectively. The ITRs, which are 145 bases in length, are the sole
cis-acting elements essential for chromosomal excision (rescue), integration, replication,
and encapsidation of nascent viral DNA.4-5 The viral genome consists of two major open
reading frames (ORF). The left ORF encodes the overlapping AAV replication proteins,
Rep68 and Rep78 from the p5 promoter and Rep40 and Rep52 from the p19 promoter. The
right ORF encodes AAV capsid proteins (VP1, VP2, and VP3) driven by the p40 promoter.4-5
A schematic representation of the AAV life cycle and recombinant AAV (rAAV) pro-
duction is shown in Figure 8.1. Productive AAV infection is dependent on a helper virus,
usually adenovirus; however herpesvirus and vaccinia virus can also perform this func-
tion.4 The helper virus may primarily optimize the host milieu for productive AAV infec-
tion. In support of this, several stress-inducing agents (ultraviolet light, gamma-irradiation,
chemical carcinogens, metabolic inhibitors, and heat shock) that alter the intracellular mi-
lieu, support productive AAV infection in the absence of a helper virus, albeit less efficiently
(~ two logs lower).14-18 In the absence of a helper virus, the AAV genome preferentially
integrates into a defined region of human chromosome 19 (q13.3-qter), and establishes a
stable latent infection.19-21 The frequency of the site-specific integration of AAV is estimated
to be up to 70% in established human cell lines such as HeLa and Detroit-60.19 The presence
of viral Rep proteins (Rep68 or Rep78) in trans and ITRs in cis are essential for the site-
specific integration of the viral genome on chromosome 19.22 Efficient rescue and replica-
tion of the integrated viral genome from chromosomal DNA ensues when latently AAV
infected cells are superinfected with a helper virus.4-5
Positive and Negative Features of AAV Vectors

Attractive features of AAV for use in human gene therapy include:
1. non-pathogenicity—AAV has not been linked directly to the cause of any human
disease and has not been associated with tumorigenicity, although it is frequently
isolated from certain tissues (in fact, current data indicate that AAV may function as
an anti-oncogenic virus);23-24
2. broad host range—AAV infects a broad host range including dividing and non-
dividing cells. However, transduction of the former is more efficient;25
3. stable integration—AAV stably integrates into chromosomal DNA, allowing long-
term transgene expression;4-5
4. stable virions—AAV virions are very stable and tolerate conditions which readily
inactivate enveloped viruses, such as retroviruses;4
5. stem cell infection—AAV infects human (CD34+) and mouse (Lin–Sca-1+) hemato-
poietic progenitor cells;26-31
6. absence of promoter interference—unlike the retrovirus LTR, the AAV-ITR is de-
void of enhancer and TATA box-containing promoter elements, and it is less likely
to interfere with the transgene expression.32 However, it has an initiator sequence
characteristic of a TATA-less promoter and binding sites for the SP1 transcription
factor;33 and
Adeno-Associated Virus (AAV) Mediated Ribozyme Expression in Mammalian Cells 103
Fig. 8.1. Schematic representation of AAV life cycle, rAAV production, and wild type (wt) AAV
genome. (A) In the absence of a helper virus, the wt AAV integrates into the chromosomal DNA
site-specifically (latent infection). In the presence of a helper virus, the wt AAV undergoes a
productive infection (lytic infection). Helper plasmid (pHelper), recombinant AAV plasmid
(prAAV), and adenovirus are required for generating recombinant AAV (rAAV production). Open
squares in pHelper depict the adenovirus ITRs, and filled rectangles in prAAV depict AAV-ITRs
which flank the expression cassette for the gene of interest (Pro-Gene) (B) The wt AAV genomic
map. Inverted terminal repeats (ITRs), promoters (p5, p19, p40) with the associated gene prod-
ucts (replication {Rep} and capsid {VP} proteins), and the polyadenylation site (pA) are indi-
cated. See text for detail.
7. low immunogenicity—AAV, in contrast to adenovirus, has no observed significant

adverse immune responses to rAAV vectors.34-36 This is probably due to the absence
of AAV ORFs in most rAAV vectors.
There are, however, certain undesirable features of rAAV for use as a vector for gene
therapy, such as:
1. There is no simple method for large scale rAAV generation, as currently is possible
for recombinant adenovirus and retrovirus production;
2. AAV-mediated gene expression is dependent on the efficient conversion of its ss-
DNA genome into double-stranded form in most cells.37-40 This rate limiting step is
dependent on cell type, and can be dramatically enhanced by certain factors and
conditions, such as helper virus infection, genotoxic treatments (e.g., gamma radia-
tion), and tyrosine phosphorylation;37-40
3. Integration efficiency varies depending on cell type, and in some instances the viral
genome remains episomal in the host cell nuclei, conferring transgene expression
up to several weeks;41-42
4. The packaging capacity for foreign DNA into rAAV is limited to 4.5 kb, which ex-
ceeds the size of many human therapeutic cDNA genes;43-45 and
5. 80-90% of the human population is seropositive for AAV,46 which may pose a prob-
lem for in vivo transduction of human cells.
Generation of Recombinant AAV

Background
The molecular cloning of an intact double-stranded form of the AAV genome into a
bacterial plasmid facilitated the study of AAV genetics and paved the way for the production
of rAAV.47-48 Rescue and replication of the AAV genome from a plasmid was demonstrated
in adenovirus infected cells.47-48 Nascent viral DNAs were shown to be packaged into bio-
logically infectious virions.47-48 Genetic studies using the cloned AAV revealed that foreign
DNA, substituted for the entire AAV Rep and Cap coding region and flanked by AAV-ITRs,
can undergo rescue/replication in host cells in the presence of AAV Rep proteins and helper
adenovirus (see Fig. 8.1).49 Furthermore, replicated DNA fragments between ~2-5 kb can
be packaged into mature AAV virions, provided that AAV Cap proteins are provided in
trans.43,49 Recombinant viral genomes ranging in size between 4-5 kb are efficiently pack-
aged into virions, whereas smaller recombinant genomes are packaged inefficiently.43-45
Methods for rAAV Production

Early work on developing a method for generating rAAV made use of a two-plasmid
system. The “helper plasmid” contained an intact cloned AAV genome with a 1 kb lambda
DNA insertion just upstream of the right ITR.43 Upon rescue and replication in adenovirus
infected cells, this modified AAV genome is too large to be packaged into mature virions,
but provides AAV Rep and Cap expression. The “recombinant plasmid” (prAAV) contains a
gene of interest flanked by AAV ITRs. The rescue, replication, and packaging of the recom-
binant genome occurs in the presence of the Rep and Cap proteins provided in trans by the
helper plasmid. The major problem with this method of generating the rAAV is the produc-
tion of contaminating wild type AAV (wt AAV), which can range from 10%-50% of the
total virus produced.43 This is primarily due to the homologous AAV sequences in the helper
and recombinant plasmids.
Samulski et al constructed new helper and recombinant backbone plasmids which re-
sulted in a significant increase in rAAV titer and dramatically reduced wt AAV contamina-
tion.50-51 The new helper plasmid (pAAV/Ad) consists of the entire AAV coding region (Rep
and Cap ORFs) flanked by the first 107 bp of adenovirus type 5 terminal repeats (Ad-ITR).51
The insertion of adenovirus termini was originally intended to amplify the helper plasmid
in cells; however helper plasmid amplification was not detected. Instead, the strong en-
hancer within the Ad-ITR significantly augments AAV Rep and Cap expression. For con-
struction of the recombinant AAV plasmids, the cloned AAV backbone vector was modified
for easy replacement of the entire AAV coding region with a gene of interest. The resulting
absence of homologous AAV sequences in the helper and recombinant plasmids has re-
sulted in a dramatic reduction in the generation of wt AAV.51
Vector Constructions
We have constructed several rAAV backbone plasmids which were tailor-made for the
insertion of ribozyme expression cassettes driven from either pol II or pol III promoters
(Fig. 8.2). To allow for selection in the target cell, either the bacterial neomycin
phosphotransferase (NPT II) gene driven by the herpes simplex thymidine kinase promoter,
or a modified nerve growth factor receptor (NGFR) gene, driven by cytomegalovirus early
Fig. 8.2. Schematic drawing of recombinant AAV plasmids. Recombinant AAV plasmids pAT29
and pAT30 contain the “TRZA568” transcription unit, a hammerhead ribozyme targeted against
a site in the HIV-LTR, driven by a modified tRNA promoter.52 See text for additional detail.
promoter, was cloned into these rAAV constructs. The NGFR expressed from these con-
structs is biologically nonfunctional because of a deletion in the C terminus of the cDNA.
However, the expression of this truncated NGFR can be readily detected on the membrane
of transduced cells by fluorescence activated cell sorting (FACS). In contrast, the NPT II
selection requires the neomycin analog G418, which is toxic to uninfected cells. Due to its
rapid selection and lack of toxicity, NGFR expression is a preferred method of selection in
our laboratory. A 1.6 kb stuffer fragment originating from the bacterial LacZ gene, and con-
taining a multiple cloning site (MCS), was inserted in two of the constructs to increase the
size of the recombinant genome for efficient packaging. The pol II and III promoters used
for driving hammerhead ribozyme (Rz) expression are modified versions of mammalian
tRNA, U1, and U6, and viral (CMV) promoters. The molecular design and details of the
construction of these modified promoters are described elsewhere (Unpublished data for
U1, U6, and CMV).52
Rescue and Replication

Rescue and replication of these rAAV, with or without ribozyme transcription units,
was examined in 293 cells (Fig. 8.3). The characteristic viral replicative intermediates (mono-
mer, dimer, tetramer, etc.) were detected, similar to the rescue/replication pattern of cloned
AAV.49-50 Rescue and replication of pAT20 appears to be more efficient as compared to the
other recombinant AAV plasmids, as shown in Figure 8.3. The reproducibility of this obser-
vation has not been investigated. Rolling and Samulski examined the stability in rAAV of
exogenous sequences with secondary structures, such as the adenovirus polymerase III VA
RNA promoter, the internal ribosomal entry site (IRES) of encephalomyocarditis virus
(EMCV), and the 3' untranslated region of Factor IX (FIX).53 With the exception of the 3'
Fig. 8.3. Rescue and replication of rAAV plasmids. Adenovirus-infected 293 cells
were cotransfected with the helper plasmid (pAAV/Ad) and with pAT20, pAT22,
pAT23, or pAT24 using a calcium phosphate transfection method (Promega).
Recombinant AAV plasmids pAT23 and pAT24 contain a hammerhead ribozyme
expression cassette targeted against a metalloproteinase mRNA driven by a U1
promoter. The expression cassettes were cloned into the MCS of the pAT22 in
either the forward or reverse orientation, (pAT23 and pAT24, respectively). Low
molecular weight (LMW) DNAs were isolated 48 h post-transfection/infection
using the protocol described by Hirt.75 LMW DNAs (~20 µg) were digested with
the restriction enzyme DpnI (digests unreplicated, methyated DNA), resolved
by electrophoresis on a 1% agarose gel, transferred to a nitrocellulose mem-
brane (Genescreen plus), and probed with [32P]-labeled LacZ DNA. M and D
denote the monomeric and dimeric forms of the rAAV DNA replicative inter-
mediates, respectively.
untranslated region (UTR) of FIX, these sequences did not interfere with rescue/replication
or virus production. The 3' UTR of FIX in rAAV vectors resulted in significant reduction of
viral yield due to progressive deletion of recombinant sequence during replication.53 To
date, none of our hammerhead Rz transcription units have caused aberrant replication in
the context of rAAV plasmids. rAAV titers, with or without Rz-transcription units, are ap-
proximately the same, ranging between 7.5-8.5 x 108 total IU/ml obtained from large scale
preparations as described below.
Viral Production and Titer Determination

For large scale preparation of rAAV, 293 cells were cotransfected with pAAV/Ad and
prAAV (100 µg of each plasmid per 500 cm2 plate) using a calcium phosphate transfection
method. Transfection is carried out for 12-16 h, followed by adenovirus infection (MOI:5-10).
Cells are processed for rAAV purification when adenovirus cytopathic effect (CPE) is vis-
ible microscopically (65-72 h p.i.). The cell pellet is frozen and thawed 3-4 times at -70°C
and 37°C, respectively, and the cell lysate is centrifuged to collect the supernatant (also re-
ferred to as the “crude rAAV” preparation). The rAAV particles in this crude preparation are
further purified by CsCl equilibrium gradient centrifugations. The rAAV with a genomic
size similar to that of AAV has a density of 1.42-1.45 g/cm3 as compared to the 1.33 g/cm3
density of adenovirus in CsCl solution. We further purified the crude viral preparations by
two CsCl step gradient centrifugations (1.33/1.55 g/cm3 densities). The rAAV band is lo-
cated at the junction between 1.33 and 1.5 CsCl densities, just below the visible band of
adenovirus. The portion of the gradient containing rAAV is collected and dialyzed against
PBS containing 1% sucrose. The dialyzed rAAV preparation is concentrated by centrifuga-
tion through an Amicon Centriplus membrane system.
The titer of dialyzed rAAV preparations is routinely determined by two different meth-
ods. The copy number of rAAV genomes is determined using a previously described slot-
blot hybridization method.54 The number of infectious units per ml are determined by FACS
analysis of cell surface NGFR expression. In our experience, the use of slot-blot hybridiza-
tion for particle number determination results in a 1-2 log higher rAAV titer than the phe-
notypic assay. This discrepancy is most likely due to defective interfering particles giving
rise to artificially high apparent titers. A typical rAAV titer determination is shown in Fig-
ure 8.4.
An Update on New Strategies for rAAV Production

One of the major obstacles for rAAV application in human gene therapy has been the
lack of an efficient system to generate high titer recombinant virus, free of AAV and aden-
ovirus. A complete packaging cell line that inducibly expresses AAV Rep and Cap proteins
and adenovirus VA I and II RNAs, and constitutively expresses the remaining helper func-
tions of adenovirus (E1A, E1B, E2A, and E4), would be an ideal cell line for generating
rAAV. Progress on developing alternative methods for the production of rAAV is briefly
discussed below.
To date, at least two different cell lines that inducibly express AAV Rep have been estab-
lished.55-56 In one line, the mouse metallothionein-inducible promoter was used to drive
the expression of the entire Rep ORF in 293 cells.55 Clonal isolates showed variable levels of
Rep expression which influenced cell growth rate. Interestingly, Rep78 could be induced in
one subclone of the primary isolate to 50-70% of the level present in 293 cells coinfected
with AAV and adenovirus. Rep52 was constitutively expressed in these cell lines. No expres-
sion of Rep40 was detected in any of the isolated clones. In the second cell line, the gluco-
corticoid-responsive promoter of mouse mammary virus was used to drive the expression
of the entire Rep ORF in HeLa cells.56 Two promising clonal isolates have been found to
Fig. 8.4. rAAV titer determination. The recombinant AAV plasmid pAT30 contains a hammer-
head ribozyme expression cassette, targeted against a site in the HIV LTR, driven by a modified
tRNA promoter cloned into the MCS of pAT22. The plasmid pAT30 was used to generate a large
scale preparation of rAAV-AT30 in 293 cells as described in the text. HeLa cells were infected with
dilutions of the CsCl gradient purified rAAV-AT30. (A) The particle concentration was deter-
mined by measuring viral genome copy number using slot blot hybridization as previously de-
scribed.54 (B) The percentage of cells expressing cell surface NGFR was determined by FACS
analysis at 48 h p.i.
express Rep78 in the presence of dexamethasone, and have been shown to complement the
replication of a Rep mutant rAAV.56 In contrast to Rep-expressing 293 cells, no growth in-
hibitory effect was noted. Even though these two HeLa clones efficiently express all three
AAV capsid proteins upon trans-activation by Rep in the presence of adenovirus, neither
clone supported the packaging of a Rep-deficient rAAV. Interestingly, transient expression
of any one of the AAV Rep proteins (Rep78, 68, 52 or 40) in these clones was sufficient to
obviate the packaging problem.57 Thus, lack of packaging in these HeLa cell lines was ex-
plained to be due to low level of Rep proteins. Furthermore, high levels of Rep78 expression
can compensate for the absence of Rep52, which is critical for the accumulation of ss-DNA
and subsequent packaging.57 This suggests redundancy of some function(s) among Rep
proteins. It appears that development of a cell line which expresses all four Rep proteins to
a similar level as that in 293 cells coinfected with AAV and adenovirus is a challenging task;
however, ongoing progress is directed toward achieving this goal.
The utility of an EBV-based vector to maintain an AAV Rep and Cap expression cas-
sette, a gene of interest flanked with AAV-ITR, or a combination of both on the same episo-
mal vector, was investigated in 293 and HeLa cell lines.58-59 Results indicate that continuous
Rep expression is either toxic (e.g., 293 cells) or insufficient (e.g., HeLa cells) for efficient
rAAV production. Since the rAAV-EBV-based vectors, without an AAV Rep ORF, were shown
to be stable episomally in clonal isolates of 293 or HeLa cells, the incorporation of AAV Rep
genes driven by an inducible promoter in this system might be warranted.
Several alternative methods for the production of rAAV were presented at the VII In-
ternational Parvovirus Workshop, held in Heidelberg, Germany in September, 1997. First, a
recombinant herpesvirus containing AAV Rep and Cap expression cassettes was constructed
and tested for its use in generating rAAV.60 This recombinant herpesvirus can effectively
substitute for helper plasmid and adenovirus, and gives rise to rAAV titers equivalent to the
two-plasmid method of rAAV production.60 Second, a plasmid containing the helper func-
tions of adenovirus and AAV Rep and Cap expression cassette was constructed, which can
substitute for the helper plasmid and adenovirus in the current method of rAAV produc-
tion.61 The titers of rAAV using the latter plasmid were reported to be comparable to the
standard two-plasmid method. Third, a packaging deficient adenoviral plasmid was tested
as a substitute for adenovirus for the rAAV production.62 The major advantage here is that
the viral preparations are free of adenovirus. rAAV titers are comparable to the two-plasmid
production method.62-63 Lastly, the Cre-Lox P recombination system was used to establish
cell lines with Rep and Cap expression cassettes.64 Expression from these cassettes occurred
only in the presence of Cre, which was provided by a helper recombinant adenovirus.
In view of the ease of generating high titers (1011-1012 IU/ml) of AAV from cells
coinfected with AAV and adenovirus, why is the development of an efficient large scale
method for the production of high titer rAAV free of adenovirus still lacking? The foregoing
studies underscore the need to better understand AAV transcriptional regulation in the pres-
ence of adenovirus. It is clear that the level of expression of AAV Rep proteins (78, 68, 54,
40) and Cap (VP1, 2, 3) is essential for optimal rAAV production,65-66 as this is tightly regu-
lated in cells coinfected with AAV and adenovirus. The activity of the AAV promoters, p5,
p19, and p40, are tightly coordinated and significantly influenced by AAV-ITRs, AAV Rep
proteins, adenovirus, and cellular transcription factors.67-68 The design of an efficient method
for the production of rAAV has to utilize a transcriptional regulation system similar to one
existing in cells coinfected with AAV and adenovirus.
rAAV Infection and Transgene Expression

Role of Helper Virus and Host in AAV-Mediated Expression
The conversion of AAV ss-DNA genome into ds-DNA form is essential for initiation of
AAV replication, gene expression, and presumably for integration. Although adenovirus
helper functions are required for productive AAV infection, none of the adenovirus gene
products have been demonstrated to be directly involved in supporting AAV replication,
expression, and integration.4-5 The adenovirus E1 and E4 gene products are indirectly im-
plicated in efficient conversion of single-stranded AAV genomes to the transcriptionally
active double-stranded form, presumably via enhancing host DNA polymerase(s) activ-
ity.37-38 In support of this, helper virus-independent AAV production was demonstrated in
some hamster and human cells pretreated with hydroxyurea, ultraviolet (UV) light, or
cyclohexamide.14-15 These observations signify that the helper virus primarily induces some
host functions which support the initiation of AAV replication and expression. Consistent
with this, rAAV devoid of Rep and Cap genes and containing either NPT II or alkaline
phosphatase (ALP) genes transduces human diploid fibroblasts primarily during S-phase
of the cell cycle.25 Interestingly, factors that influence host cell DNA metabolism, such as
DNA synthesis inhibitors (e.g., aphidicolin, hydroxyurea) or topoisomerase inhibitors (e.g.,
etoposide, ectoposide, and camptothecin) significantly augment rAAV transduction (ex-
pression and/or integration) in normal human diploid fibroblasts.14-18 It appears that the
viral or non-viral factors which support AAV replication and expression induce some host
cell functions which are normally turned on in the S-phase of the cell cycle.
In an elegant report, Qing et al identified a cellular protein, named “double stranded
D-sequence-binding protein” (dsD-BP), which specifically interacts with the D sequence at
the 3' end of the AAV genome.39 Tyrosine phosphoration and dephosphorylation of dsD-BP
is hypothesized to block, or permit, second strand AAV DNA synthesis, respectively. Inter-
estingly, some of the aforementioned factors (adenovirus E4 ORF6, hydroxyurea), which
are implicated in enhancing second strand DNA synthesis, promote the dephosphorylation
of dsD-BP bound to the AAV genome.39-40 Furthermore, a specific inhibitor of tyrosine
kinases, genistein, is shown to completely substitute for adenovirus in augmenting rAAV
expression.39 The discovery of dsD-BP not only paves the way for improvising new strate-
gies to realize the utility of rAAV for gene therapy, it underscores a plausible role of dsD-BP
in the regulation of leading-strand DNA synthesis in human cells. It is reasonable to specu-
late that in the S-phase of the cell cycle tyrosine dephosphorylation of the dsD-BP triggers
initiation of leading strand DNA synthesis. In view of this, AAV replication and expression
is intimately linked to host DNA and RNA synthesis.
We were interested in testing whether adenovirus is the sole factor required for optimal
rAAV transgene expression in cells coinfected with rAAV and adenovirus. To address this,
the titer of a crude rAAV-LacZ preparation was determined on 293 cells. The preparation
was either titered directly, titered after heat treatment to inactivate the adenovirus, or titered
after heating followed by the addition of increasing amounts of exogenous adenovirus
(Fig. 8.5B). Heat treatment of the rAAV-LacZ decreased the titer from 8 x 107 to 105, a re-
duction of ~2-3 logs in LacZ expression. We tested whether addition of exogenous adenovi-
rus to the heat-treated rAAV-LacZ preparation would fully reconstitute LacZ expression.
The titer of heat-treated rAAV-LacZ increased as the MOI of adenovirus increased. How-
ever, only ~ 5% reconstitution of LacZ expression was achieved following maximum addi-
tion of adenovirus. In a separate experiment, rAAV-LacZ and adenovirus titers were deter-
mined in a preparation which was partially purified by polyethylene glycol precipitation
and further purified by CsCl gradient centrifugation (no heat treatment), to remove con-
taminating adenovirus. The rAAV-LacZ titer was determined prior to CsCl gradient cen-
trifugation, in rAAV-containing CsCl gradient fractions, and following addition of exog-
enous adenovirus to the rAAV-containing fractions (Fig. 8.5A). The addition of exogenous
adenovirus to the purified rAAV preparation reconstituted LacZ expression to near that of
the original titer. Lack of complete reconstitution of rAAV titer can be explained by loss of
rAAV particles during the CsCl purification step.
Our data suggest that, besides adenovirus, one or more heat-labile element(s) might be
critical for rAAV expression in 293 cells. It is possible that rAAV may package heat-labile
factors (rAAV Rep, dsD-BP, etc.) which may play crucial roles in vector-mediated gene ex-
pression following viral uncoating in the cells. Interestingly, AAV Rep has been detected in
the mature AAV virions, and is implicated in the growth inhibitory effect of rAAV, devoid of
Rep gene, on primary human cells.69 Whether encapsidated AAV Rep plays a role, alone or
by interacting with cellular factors such as dsD-BP, in the early stage of viral replication
remains to be investigated. In view of the critical role of dsD-BP in either inhibiting or
stimulating second strand DNA synthesis, it would be interesting to investigate whether
dsD-BP is encapsidated in mature AAV virions.
To address the possibility that the aforementioned observations might be related to a
specific transcription unit (e.g., the CMV promoter driving the LacZ gene), we tested rAAV
containing other transcription units. The “TRZ” transcription unit, containing an anti-HIV
ribozyme driven by a modified tRNA promoter,52 was cloned into two of our rAAV back-
bone plasmids as shown in Figures 8.2 and 8.6. Ribozyme expression was detected at low
levels in 293 cells infected with rAAV-AT29 and -AT30, which were heat-treated to inacti-
vate contaminating adenovirus. In contrast, the addition of exogenous adenovirus to the
heat-treated rAAV sample dramatically augmented TRZ expression. Our data is in agree-
ment with the results of other investigators who reported the effect of adenovirus on en-
hancing rAAV-mediated gene expression, independent of the nature of the transgene pro-
Fig. 8.5. Adenovirus enhance-

ment of rAAV-LacZ expression.
(A) The total IU of a rAAV-LacZ
preparation, which was partially
purified by polyethylene precipi-
tation, was determined before
CsCl gradient centrifugation (*),
to remove adenovirus, and after
purification, with the addition of
the indicated MOIs of exogenous
adenovirus (dl312). (B) Effect of
heat-treatment and addition of
exogenous adenovirus on rAAV-
LacZ titer. The total infectious
unit (IU) of a crude rAAV-LacZ
preparation was determined on
293 cells using the nonheat-
treated sample or the heat-treated
(56°C, to inactivate adenovirus)
sample with or without the ad-
dition of the indicated MOIs of
exogenous adenovirus (dl312).
moter and its coding region.37-39 In our system, adenovirus enhancement of ribozyme ex-
pression can be explained by considering the following possibilities:
1. adenovirus augments the second-strand synthesis of rAAV genomic DNA, which is
indispensable for transcription initiation;
2. contaminating wt AAV, generated during the production of rAAV preparations, can
increase the copy number of rAAV genomic templates, leading to higher expression;
and
3. adenovirus activates the tRNA promoter driving the ribozyme expression.
These possibilities are being addressed in our laboratory.
Fig. 8.6. Adenovirus enhancement of rAAV-mediated ribozyme expression. 293 cells were in-
fected with heat-treated rAAV-AT30 or rAAV-AT29 with or without the addition of exogenous
adenovirus (MOI:2). Cells were harvested at the indicated time points p.i., and RNA purification
and Northern blot analysis were carried out as described previously.52 As a positive control, 293
cells were transfected with pAT30 using a calcium phosphate transfection protocol (Promega),
and total RNA was analyzed 48 h post-transfection (labeled as pAT30). This blot was hybridized
with a [32P]-labeled T7 transcript containing the ribozyme and modified tRNA promoter se-
quences which also hybridizes to the endogenous species of tRNAs.
Host Range, Transduction, and Gene Expression

An attractive feature of rAAV as a vector for human gene therapy is its ability to infect
a broad range of mammalian cell types, such as human, mouse, and rat. In addition, rAAV-
mediated transgene expression appears to be stable for a long period of time without any
deleterious immune response. These features facilitate the testing of safety and efficacy of
rAAV in vivo, in animal models of human disease.
The transduction efficiency of rAAV varies significantly among different cell types within
a species. Several recent studies have shown that muscle, brain, and liver are among the best
tissues for rAAV-mediated transgene expression.34-36,70-73 When rAAV-LacZ was injected in-
travenously in mice, LacZ expression was predominately detected in liver hepatocytes, (cells
which are non-dividing); however moderate expression of LacZ was shown in lung and
kidney tissues.72 This finding paves the way for target-specific therapeutic application of
rAAV. Efficient transduction of non-dividing liver cells with rAAV containing the expres-
sion cassette for the human factor IX (hFIX) was recently reported with a single administra-
tion of rAAV to mice.71 This study demonstrated stable, persistent, and therapeutic levels of
factor IX expression in mice, without any significant immune response.
Efficient expression of LacZ and hFIX was demonstrated in mouse muscle cells.34 High
levels and stable LacZ expression was detected in mice following intramuscular injection of
rAAV, which integrated into chromosomal DNA of differentiated muscle fibers.34 In this
study, no humoral or cellular immune responses were elicited against the E. coli β-galactosi-
dase. Furthermore, neutralizing antibody against AAV capsid proteins did not prevent
readministration of rAAV vector. In the case of rAAV-mediated hFIX expression in mouse
muscle, the level of plasma factor IX reached therapeutic levels, and was sustained for 6
months.34 The efficient expression of LacZ or factor IX mediated by rAAV in muscle did not
induce any detectable cytotoxic T lymphocyte response; however, immune competent mice
developed circulating antibodies to hFIX.34 This is in stark contrast to adenovirus-mediated
expression of LacZ or Factor IX, which induces a massive immune response leading to com-
plete abrogation of transgene expression.3,34
These observations underscore the potential of rAAV-mediated gene therapy in muscle
tissues for the treatment of inherited and acquired human diseases. Muscle tissues are also a
good reservoir for the systemic delivery of therapeutic molecules like blood clotting factors,
growth factors, or hormones.
Several reports have demonstrated the efficient expression of transgenes in neuronal
cells.70,73 LacZ expression was detected in caudate nucleus, amygdala, striatum, and hippoc-
ampus of the adult rat brain 3 days following stereotactic microinjection of rAAV-LacZ.70
Transduction efficiency into various regions of the rat brain was estimated to be about 10%,
which closely matches that of transgene delivery with HSV or adenovirus. Neuronal expres-
sion of LacZ was detected in caudate nucleus for up to 3 months. Kaplitt et al investigated
the therapeutic benefits of expressing tyrosine hydroxylase (AAVth) in a rat model for
Parkinson’s disease.70 Tyrosine hydroxylase expression was detected in striatal neurons and
glia for up to 4 months following vector injection into the denervated striatum of unilateral
6-hydroxydopamine-lesioned rats. Significant behavioral recovery in lesioned rats was noted
with rAAVth versus AAVLacZ treated animals.
The ability of rAAV to infect primary mouse (Lin–Sca-1+) and human (CD34+) he-
matopoietic progenitor cells, capable of multilineage differentiation, has also been docu-
mented in several studies.26-31 For example, murine hematopoietic cells transduced ex vivo
with an rAAV containing a human β-globin expression cassette followed by transplantation
into lethally irradiated congenic mice resulted in long-term expression of β-globin in bone
marrow cells for up to 6 months.29 When bone marrow cells from these primary mice were
transplanted into secondary animals, the expression of human β-globin was detected in
bone marrow cells for up to 3 months.29 Although rAAV-mediated LacZ expression in hu-
man hematopoietic progenitor cells (CD34+) has been reported in vitro, transgene expres-
sion varies in CD34+ cells from different donors, ranging between 0% to 80%.28 When bone
marrow cells that are refractory to rAAV infection are induced with growth factors to un-
dergo differentiation, they become permissive to rAAV transduction and transgene expres-
sion.28 The stable expression of the transgene was detected in the differentiated lineages of
the primary infected hematopoietic cells, such as myeloid, during the course of experiments.
The observed variability of rAAV transduction of human hematopoietic cells is linked to
the differential expression of a putative AAV receptor on CD34+ cells.28
Flotte et al reported in several studies the utility of rAAV-mediated expression of the
cystic fibrosis transmembrane regulator (CFTR) gene in respiratory epithelial cells, which
are a suitable target for the treatment of cystic fibrosis (CF).33,41 Efficient expression of the
human CFTR gene has been demonstrated in an immortalized human bronchial epithelial
cell line in vitro and subsequently in rabbit and rhesus monkey respiratory epithelium in
vivo. Fiberoptic bronchoscopy was used to deliver the rAAV-CFTR to a lobe of the lung in
the aforementioned animals.41 This resulted in efficient and stable expression of the CFTR
gene in lung epithelium, with minimal toxicity and no deleterious immune responses. These
promising observations set the stage for testing the utility of CFTR-expressing rAAV in the
ongoing clinical trial in CF patients.
Table 8.1. Summary of rAAV transduction
Cell Line Species Cell Type vAT30 MOI % Heat

(µl) NGFR+ Inact. Ad
HeLa Human Cervical Carcinoma 64 10 95

H4a Human Neuroglioma 32 5 99
SK-N-SH Human Neuroblastoma 64 20 80 Yesb
MVECc Human Endothelium 50 20 93
LNCap.FGC Human Prostate Carcinoma 64 20 85 Yesb
MDA-MB-435S Human Breast Carcinoma 64 20 80
T-47D Human Breast Carcinoma 64 20 8 Yesb
HT-29 Human Colon Carcinoma 64 20 35
HS-27c Human Foreskin fibroblast 100 25 98
RSFc Rabbit Synoviocyte 100 25 95
B16 Mouse Melanoma 100 25 78
HL60 Human Promelocytic Leukemia 60 90 0
An approximately equivalent number of cells for each cell line (~5 x 105) were infected with the
indicated volume of rAAV-AT30. The percentage of cells expressing NGFR was determined by FACS
analysis 3 days post-infection. aTransduction (percentage of NGFR-expressing cells) was approximately
the same with or without heat-treatment of rAAV-AT30. brAAV-AT30 was heat-treated (56°C for
30 min) to inactivate adenovirus, as the non-heated viral inoculum was toxic to these cell lines 3 days
p.i., presumably due to lytic infection by adenovirus. cPrimary cell lines.
We were interested in comparing and evaluating the transduction efficiency of rAAV

containing a ribozyme expression cassette in a variety of mammalian cell lines. In pursuit of
this, rAAV-AT30 was used to infect a variety of mammalian cell lines. The titer of rAAV-
AT30 vector was determined on HeLa cells as shown in Figure 8.4. Table 8.1 is a summary of
our transduction data based on measuring cell surface NGFR expression by FACS.
In contrast to 293 cells, approximately equivalent transduction of the neuroglioma
(H4) cell line was achieved with either the nonheat-treated or the heat-treated rAAV-AT30.
One likely explanation might be the efficient synthesis of second strand rAAV DNA in the
H4 neuroglioma cells, independent of the adenovirus E4 ORF helper function. This obser-
vation is in agreement with others who reported that the conversion of ss-DNA of the rAAV
genome to transcriptionally active ds-DNA may vary significantly in different cell types and
tissues.40 This may explain the reason for efficient expression of rAAV transgene in liver,
muscle, and neuronal cells, and poor expression in other cell types, in vivo. In our experi-
ence, the rAAV transduction of human monolayer cell lines generally were more efficient
than suspension cell lines.
The levels of NGFR expression from three cell lines in Table 8.1 is shown in Figure 8.7.
The H4 neuroglioma cell line is highly susceptible to rAAV transduction as compared to
other cell lines tested in our laboratory so far. We also noted that, although some cell lines,
such as HeLa and H4, show very similar transduction efficiencies as measured by NGFR
expression, the level of transgene expression can vary. Differences in gene expression might
be due to several factors:
1. AAV receptor density may vary in different cell lines;
2. The conversion of rAAV ss-DNA into ds-form might be more efficient in some cell
lines as compared to others; and
3. The activity of a promoter driving the transgene expression is influenced by the cell
type.
Fig. 8.7. FACS profile
of the rAAV-AT30-
mediated NGFR
expressing cells. The
pattern of NGFR
expression for three
human cell lines (H4,
HeLa, and LNCap.FGC)
transduced with the
rAAV-AT30 (Table 8.1) is
shown. Cells were
infected at an MOI
(based on particle
concentration) of 5
(H4), 10 (HeLa), or 40
(LNCap.FCG). The
particle:IU ratio can vary
with cell line, but is
typically in the range of
50-500:1. The percentage
of cells expressing
NGFR, in mock- and
rAAV-infected cells, was
determined by FACS
analysis.
Adeno-Associated Virus (AAV) Mediated Ribozyme Expression in Mammalian Cells
115
The expression of an anti-HIV ribozyme from an rAAV vector (TRZA568) was readily
detected in several of these cell lines (Fig. 8.8). The modified tRNA promoter is highly active
in most of the cell lines tested in our laboratory. Since rAAV-mediated TRZ expression is
very efficient in the H4 neuroglioma cell line, it would be interesting to examine the inhibi-
tory effect of this on HIV infection, similarly to what has been attempted using retroviruses.
Retroviral vectors have been used to transduce either the hairpin or the hammerhead
ribozyme expression cassette into a variety of cell lines.10-13,74 The majority of these studies
focused on targeting HIV transcripts, including Tat, Tat/Rev, and Gag mRNAs. For instance,
retroviral mediated expression of a hammerhead ribozyme specific for Tat or Tat/Rev in
CD4+ cell lines (CEM, SupT) showed significant levels of resistance against HIV infection
as demonstrated by the reduction of p24 antigen.12-13 In the latter studies, the hammerhead
ribozyme was embedded in the body of a NPT II transcript driven by Molony LTR, or it was
inserted in the 3' LTR driven by a modified tRNA in a LN retroviral vector. In a separate
study, macrophage-like cells differentiated from the CD34+ cells, which were transduced
with a retrovirus containing a hammerhead ribozyme expression cassette specific for the
HIV genome, were significantly resistant to a macrophage tropic HIV-1 infection.74
The efficacy of vector-mediated ribozyme expression in cells is dependent on several
factors, including the levels of ribozyme and target RNA expression, the half-life of the
ribozyme and target RNA transcripts, absence of inhibitory secondary structure, and
colocalization of ribozyme with its target RNA transcript. Bertrand et al addressed some of
the aforementioned issues by constructing ribozyme expression cassettes specific for the
TAR region of the simian immunodeficiency virus (SIV) driven by modified RSV, human
U1, U6, and tRNA promoters cloned into retroviral or AAV backbone plasmids.32 The tran-
scripts of U1, U6, and tRNA were primarily localized in the nucleus, and were not effective
in reducing heterologous mRNA containing the TAR region of SIV. However, RSV-LTR
driven transcripts containing the SIV-TAR specific ribozyme were primarily localized in the
cytoplasm, and significantly reduced the SIV LTR driven heterologous transcript. In the
aforementioned study, the U6 and tRNA driven ribozyme expression were active in the
context of an AAV plasmid; however, these cassettes were inactive in a retroviral plasmid,
presumably due to transcriptional interference by Molony LTR.
Concluding Remarks
Progress in the basic biology of AAV has answered several key questions regarding the
potential of this vector for broad use as an efficient delivery vehicle for laboratory experi-
mentations, and ultimately for applications in human gene therapy. We would like to high-
light some of the salient features of AAV as a vector, and emphasize existing knowledge gaps
pertinent to its use for gene therapy.
First, a better understanding of AAV transcriptional regulation is essential for optimiz-
ing rAAV production or establishing an “ideal packaging cell line”. Second, we now know
that AAV infection is receptor mediated. It may be useful to further characterize this puta-
tive receptor, to facilitate an understanding of the molecular mechanisms of viral entry into
cells. It might be feasible in the near future to augment rAAV transgene delivery via a tran-
sient induction of AAV receptors on target tissues prior to viral transduction. Third, in light
of the importance of second strand DNA synthesis for AAV-mediated transgene expression,
it would be prudent to characterize associated viral and host cell factors, and formulate
strategies to enhance this in tissues in which this step is rate limiting. Fourth, stable viral
integration is indispensable for long-term therapeutic gene expression. The lack of integra-
tion reported in certain systems should be investigated, to define the underlying mecha-
nism which dictates the efficiency of viral integration. Fifth, site specific integration on
chromosome 19 is one of the hallmarks of wt AAV infection, and it would be desirable to
Fig. 8.8. Ribozyme expression profile of rAAV-AT30 transduced cells. The levels of TRZ expression in H4 (A) and HeLa (B) cell lines
Adeno-Associated Virus (AAV) Mediated Ribozyme Expression in Mammalian Cells
transduced with rAAV-AT30 (same infections as shown in Table 8.1 and Fig. 8.7). Cells were harvested 3 days p.i. for FACS and RNA
analysis. RNA purification and Northern blot analysis were carried out as described previously.52 Blots were hybridized with the same
[32P]-labeled probe as in Fig. 8.3.
117
impart this feature to rAAV vectors. This unique feature could reduce the risk of insertional
mutagenesis associated with randomly integrating vectors, such as retroviral and current
rAAV vectors. Sixth, although the overall in vivo safety and lack of deleterious immunoge-
nicity are attractive features of rAAV vectors, the reproducibility of these features in human
gene therapy awaits comprehensive clinical testing.
To end this review with a humorous note, AAV is no longer an acronym for “Almost A
Virus”; instead it stands for “All Activity Vehicle” for gene delivery, which was, ironically,
coined by the automobile industry rather than by parvovirologists.
Acknowledgments
We are grateful to Drs. Dennis Macejak, Jude Samulski, Arun Srivastava, and Jim Th-
ompson for the generous gifts of plasmids and ribozyme expression cassettes. We are in-
debted to Dr. Larry Couture for his help and guidance through the course of this study. We
also thank Drs. Larry Couture and Dennis Macejak for critical reading of this manuscript.
We thank Kristi Jensen and Tim McKenzie for excellent technical assistance. We are grateful
to Dr. A. Srivastava for communicating his unpublished data to us.
References
1. Anderson WF. Gene therapy. Sci Amer 1995; 273:124-128.
2. Mulligan RC. The basic science of gene therapy. Science 1993; 260:926-932.
3. Hitt MM, Addison CL, Graham FL. Human adenovirus vectors for gene transfer into mam-
malian cells. Adv Pharmacol 1997; 40:137-206.
4. Muzyczka N. Use of adeno-associated virus as a general transduction vector for mamma-
lian cells. Curr Top Microbiol and Immunol 1992; 158:97-129.
5. Berns KI, Giraud C. Biology of adeno-associated virus. Curr Top Microbiol Immunol 1996;
218:1-23.
6. Miller AD, Miller DG, Garcia JV et al. Methods Enzymol 217:581-599.
7. Roe T, Reynolds TC, Yu G et al. Integration of murine leukemia virus DNA depends on
mitosis. EMBO J 1993; 12: 2099-2108.
8. Naldini L, Blomer U, Gallay P et al. In vivo gene delivery and stable transduction of non-
dividing cells by a lentiviral vector. Science 1996; 272:263-266.
9. Huang S, Stupack D, Mathias P et al. Growth arrest of Epstein-Barr virus immortalized B
lymphocytes by adenovirus-delivered ribozymes. Proc Natl Acad Sci USA 1997; 94:8156-8161
10. Couture LA, Stinchcomb DT. Anti-gene therapy: Use of ribozymes to inhibit gene func-
tion. Trends in Genetics 1996; 12:510-515.
11. Sarver N, Cantin EM, Chang PS et al. Ribozymes as potential anti-HIV-1 therapuetic agents.
Science 1990; 247:1222-1225.
12. Zhou I, Bahner C, Larson GP et al. Inhibition of HIV-1 in human T-lymphocytes by
retrovirally transduced anti-tat and rev hammerhead ribozymes. Gene1994; 149:33-39.
13. Zhou C. I. Bahner, Rossi JJ et al. Expression of hammerhead ribozymes by retroviral vec-
tors to inhibit HIV-1 replication: Comparison of RNA levels and viral inhibition. Antisense
and nucleic acid drug development 1996; 6:17-24.
14. Yakobson B, Koch T, Winocour E. Replication of adeno-associated virus in synchronized
cells without the addition of a helper virus. J Virol 1987; 972-981.
15. Yakobson B, Hrynko TA, Peak MJ, Winocour E. Replication of adeno-associated virus in
cells irradiated with UV light at 254 nm. J Virol 1989; 63:1023-1030.
16. Alexander IA, Russell DW, Miller AD. DNA-damaging agents greatly increase the trans-
duction of nondividing cells by adeno-associated virus vectors. J Virol 1994; 68:8282-8287.
17. Alexander IA, Russell DW, Spence AM et al. Effects of gamma irradiation on the transduc-
tion of dividing and nondividing cells in brain and muscle of rats by adeno-associated
virus vectors. Hum Gene Therap 1996; 7:841-850.
18. Russell DW, Alexander IE, Miller AD. DNA synthesis and topoisomerase inhibitors in-
crease transduction by adeno-associated virus vectors. Proc Natl Acad Sci USA 1995; 92:
5719-5723.
19. Kotin RM, Siniscalco M, Samulski RJ et al. Site-specific integration by adeno-associated

virus. Proc Natl Acad Sci USA 1990; 87:2211-2215.
20. Kotin RM, Linden RM, Berns KI. Characterization of a preferred site on human chromo-
some 19q for integration of adeno-associated virus DNA by non-homologous recombina-
tion. EMBO J 1992; 11:5071-5078.
21. Samulski RJ, Zhu X, Xiao X et al. Targeted integration of adeno-associated virus (AAV)
into human chromosome 19. EMBO J 1991; 10:3941-3950.
22. Weitzman, MD, Kÿostio SRM, Kotin RM et al. Adeno-associated virus (AAV) rep proteins
mediate complex formation between AAV DNA and its integration site in human DNA.
23. Tobiasch E, Rabreau M, Geletneky K, Larue-Charlus S et al. Detection of adeno-associated
virus DNA in human genital tissue and in material from spontaneous abortion. J Med
Virol 1994; 44(2):215-222.
24. Hermonat PL, Plott RT, Santin AD, Parham GP et al. Adeno-associated virus Rep78 inhib-
its oncogenic transformation of primary human keratinocytes by a human papillomavirus
type 16-ras chimeric. Gynecol Oncol 1997; 66:487-494.
25. Russell DW, Miller AD, Alexander IE. Adeno-associated virus vectors preferentially trans-
duce cells in S phase. Proc Natl Acad Sci USA 1994; 91:8915-8919.
26. Zhou SZ, Broxmeyer HE, Cooper S et al. Adeno-associated virus 2-mediated gene transfer
in murine hematopoietic progenitor cells. Exp Hematol 1993; 21:928-933.
27. Chatterjee S, Lu D, Podsakoff G et al. Strategies for efficient gene transfer into hematopoi-
etic cells: The use of adeno-associated virus vectors in gene therapy. Ann NY Acad Sci
1995; 770:79-90.
28. Ponnazhagan S, Mukherjee P, Wang X-S et al. Adeno-associated virus Type 2-Mediated
transduction in primary human bone marrow-derived CD34+ hematopoietic progenitor
cells: Donor variation and correlation of transgene expression with cellular differentiation.
J Virol 1997; 71:8286-8267.
29. Ponnazhagan S, Yoder MC, Srivastava A. Adeno-associated virus type 2-mediated trans-
duction of murine hematopoietic cells with long-term repopulating ability and sustained
expression of a human globin gene in vivo. J Virol 1997; 71:3098-3104.
30. Einerhand MPW, Antoniou M, Zolotukhin S et al. Regulated high-level human beta-globin
gene expression in erythroid cells following recombinant adeno-associated virus-mediated
gene transfer. Gene Ther 1995; 2:336-343.
31. Goodman S, Xiao X, Donahue RE et al. Recombinant adeno-associated virus mediated
gene transfer into hematopoietic progenitor cells. Blood 1994; 84:1492-1500.
32. Bertrand E, Castanotto D, Zhou C et al. The expression cassette determines the functional
activity of ribozymes in mammalian cells by controlling their intracellular localization. RNA
1997; 3:75-88.
33. Flotte TR, Afione SA, Solow R et al. Expression of the cystic fibrosis transmembrane con-
ductance regulator from a novel adeno-associated virus promoter. J Biol Chem 1993;
268:3781-3790.
34. Fisher KJ, Jooss K, Alston J et al. Recombinant adeno-associated virus for muscle directed
gene therapy. Nat Med 1997; 3(3):306-312.
35. Herzog RW, Hagstrom JN, Kung SH et al. Stable gene transfer and expression of human
blood coagulation factor IX after intramuscular injection of recombinant adeno-associated
virus. Proc Natl Acad Sci USA 1997; 94(11):5804-5809.
36. Snyder RO, Miao CH, Patijn GA et al. Persistent and therapeutic concentrations of human
factor IX in mice after hepatic gene transfer of recombinant AAV vectors. Nat Genet 1997;
16(3):270-276.
37. Fisher KJ, Gao G-P, Weitzman MD et al. Transduction with recombinant adeno-associ-
ated virus for gene therapy is limited by leading-strand synthesis. J Virol 1996; 70(1):
520-532.
38. Ferrari FK, Samulski T, Shenk T et al. Second-strand synthesis is a rate-limiting step for
efficient transduction by recombinant adeno-associated virus vectors. J Virol 1996;
70:3227-3234.
39. Qing K, Wang X-S, Kube DM et al. Role of tyrosine phosphorylation of a cellular protein
in adeno-associated virus 2-mediated transgene expression. Proc Natl Acad Sci USA 1997;
94(20):10879-10884.
40. Qing K, Khuntirat B, Mah C et al. Adeno-associated virus type 2-mediated gene transfer:
Correlation of tyrosine phosphorylation of the cellular single-stranded D sequence-binding
protein with transgene expression in human cells in vitro and murine tissue in vivo. J
Virol 1998; 72(2):1593-1599.
41. Carter BJ, Flotte TR. Development of adeno-associated virus vectors for gene therapy of
cystic fibrosis. Curr Top Microbiol Immunol 1996; 218:119-144.
42. Malik P, McQuiston SA, Yu XJ et al. Recombinant adeno-associated virus mediates a high
level of gene transfer but less efficient integration in the K562 human hematopoietic cell
line. J Virol 1997; 71(3):1776-1783.
43. McLaughlin SK, Collis P, Hermonat PL, Muzyczka N. Adeno-associated virus general trans-
duction vectors: Analysis of proviral structures. J Virol 1988; 62:1963-1973.
44. Hermonat PL, Quirk JG, Bishop BM, Han L. The packaging capacity of adeno-associated
virus (AAV) and the potential for wild-type-plus AAV gene therapy vectors. FEBS Lett
1997; 407:78-84.
45. Dong JY, Fan PD, Frizzell RA. Quantitative analysis of the packaging capacity of recombi-
nant adeno-associated virus. Hum Gene Ther 1996; 7(17):2101-2112.
46. Blacklow NR, Hoggan MD, Rowe WP. Serologic evidence for human infection with aden-
ovirus-associated viruses. J Natl Cancer Inst 1968; 40:319-327.
47. Samulski RJ, Berns KI, Tan M et al. Cloning of adeno-associated virus into pBR322: Res-
cue of intact virus from the recombinant plasmid in human cells. Proc Natl Acad Sci USA
1982; 79:2077-2081.
48. Laughlin CA, Tratschin JD, Coon H, Carter BJ. Cloning of infectious adeno-associated
virus genomes in bacterial plasmids. Gene 1983; 23:65-73.
49. Nahreini P, Woody MJ, Zhou SZ et al. Versatile adeno-associated virus 2-based vectors for
constructing recombinant virions. Gene 1993; 124:257-262.
50. Samulski, RJ., Chang L-S, Shank T. A recombinant plasmid from which an infectious adeno-
associated virus genome can be excised in vitro and its use to study viral replication. J
Vriol 1987; 61:3096-3101.
51. Samulski, RJ, Chang L-S, Shenk T. Helper-free stocks of recombinant adeno-associated
viruses: Normal integration does not require viral gene expression. J Virol 1989;
63:3822-3828.
52. Thompson J, Ayers D, Malmstrom T et al. Improved accumulation and activity of ribozymes
expressed from a tRNA-based polymerase III promoter. Nucleic Acids Res 1995;
23:2259-2268.
53. Rolling F, Samulski RJ. Stability of sequence cassettes containing secondary structure in
AAV vectors. VIth Parvovirus Workshop, Montpellier, France, 1995, 16.
54. Kube DM, Srivastava A. Quantitative DNA slot blot analysis: Inhibition of DNA binding
to membranes by magnesium ions. Nucleic Acids Research 1997; 25:3375-3376.
55. Yang Q, Chen F, Trempe JP. Characterization of cell lines that inducibly express the adeno-
associated virus Rep proteins. J Virol 1994; 68:4847-4856.
56. Holscher C, Horker M, Kleinschmidt JA et al. Cell lines inducibly expressing the adeno-
associated virus (AAV) rep gene: Requirements for productive replication of rep-negative
AAV mutants. J Virol 1994; 68:7169-7177.
57. Holscher C, Kleinschmidt JA, Burkle A. High-level expression of adeno-associated virus
(AAV) Rep 78 protein is sufficient for infectious-particle formation by a rep-negative AAV
mutant. J Virol 1995; 69: 6880-6885
58. Lebkowski J, Okarma TB, Philip R. The challenges of recombinant adeno-associated virus
manufacturing: Alternative use of adeno-associated virus plasmid/liposome complexes for
gene therapy applications. Curr Top Microbiol Immunol 1996; 218:51-59.
59. Neyns B, Rijcke MD, Teugels, Hermonat P et al. Establishment of recombinant adeno-
associated virus producer cell lines that provide the rep and cap complementor functions
from an episomal plasmid. VIIth International Parvovirus Workshop, Heidelberg, Germany,

1997:33.
60. Schetter C, Heilbronn R. A recombinant herpesvirus expressing AAV Rep and Cap for
efficient production of AAV vectors. VIIth International Parvovirus Workshop, Heidel-
berg, Germany, 1997, 3.2.
61. Grimm D, Rittner K, Kleinschmidt JA. Helpervirus-free packaging of adeno-associated vi-
rus derived vectors. VIIth International Parvovirus Workshop, Heidelberg, Germany,
1997:3.3.
62. Zhou SZ, Ladner M, Escobedo JA, Dwarki V. Recombinant adeno-associated virus prepa-
ration without the use of adenovirus: A novel triple transfection protocol. VIIth Interna-
tional Parvovirus Workshop, Heidelberg, Germany, 1997, 30.
63. Ferrari FK, Xiao X, McCarty D, Samulski RJ. New developments in the generation of Ad-
free, high-titer rAAV gene therapy vectors. Nature Med 1997; 3:1295-1297.
64. Burstein H, Rogers LC. Stable split packaging cell lines for rAAV vectors based on Cre-
LoxP system. VIIth International Parvovirus Workshop, Heidelberg, Germany, 1997:3.4.
65. Li J, Samulski RJ, and Xiao X. Role for highly regulated rep gene expression in adeno-
associated virus vector production. J Virol 1997; 71(7):5236-5243.
66. Flottte TR, Barraza-Oritz X, Solow R et al. An improved system for packaging recombi-
nant adeno-associated virus vectors capable of in vivo transduction. Gene Therapy 1995;
2:29-37.
67. Pereira DJ, McCarty DM, Muzyczka N. The adeno-associated virus (AAV) Rep protein
acts as both a repressor and an activator to regulate AAV transcription during a produc-
tive infection. J Virol 1997; 71(2):1079-1088.
68. Pereira DJ, Muzyczka N. The adeno-associatedvirus type 2 p40 promoter requires a proxi-
mal Sp1 interaction and a p19 CArG-like element to facilitate Rep transactivation. J Virol
1997; 71(6):4300-4309.
69. Kube, DM, Ponnazhagan S, Srivastava A. Encapsidation of adeno-associated virus type 2
Rep proteins in wild-type and recombinant progeny virions: Rep-mediated growth inhibi-
tion of primary human cells. J Virol 1997; 71(10):7361-7371.
70. Kaplitt, MG., Leone P, Samulski RJ et al. Long-term gene expression and phenotypic cor-
rection using adeno-associated virus vectors in the mammalian brain. Nature Genetics 1994;
8:148-153.
71. Koeberl DD, Alexander IE, Halbert CL et al. Persistent expression of human clotting fac-
tor IX from mouse liver after intravenous injection of adeno-associated virus vectors. Proc
Natl Acad Sci USA 1997; 94 (4):1426-1431.
72. Ponnazhagan S, Mukherjee P, Yoder MC et al. Adeno-associated virus2-mediated gene trans-
fer in vivo: Organ-tropism and expression of transduced sequences in mice. Gene 1997;
190:203-210.
73. Rosenfeld MR, Bergman I, Schramm L et al. Adeno-associated viral vector gene transfer
into leptomeningeal xenografts. J Neurooncol 1997; 34(2):139-144.
75. Hirt, B. Selective extraction of polyoma DNA from infected mouse cell cultures. J Mol
Biol 1967; 26:365-369.
Section III
Ribozyme Targets for Gene Therapy
CHAPTER 9
Applications of Anti-Oncogene
Ribozymes for the Treatment
of Bladder Cancer
Akira Irie and Eric J. Small
Overview of Bladder Cancer
B ladder cancer is the fifth most common malignancy in the United States. Approximately
54,500 patients will be diagnosed in 1997, with nearly 12,000 deaths attributed to this
disease.1 Despite improved diagnostic and therapeutic modalities, the incidence of bladder
cancer is increasing, and the overall mortality from bladder cancer has not decreased dra-
matically over the last decade. In the US, more than 90% of bladder cancers are transitional
cell carcinoma; squamous carcinoma and adenocarcinoma constitute the remainder.2 Blad-
der tumors are graded based on the degree of the anaplasia of the tumor cells. A correlation
exists between tumor grade and tumor stage, i.e., well-differentiated tumors are more likely
to be superficial, and poorly differentiated tumors are more likely to be muscle-invasive.
Bladder cancers can be characterized into two groups with considerably different natu-
ral histories, i.e., superficial tumors confined to the mucosa and muscle-invasive tumors,
which penetrate the muscularis mucosa. Approximately 70% of bladder cancers are superficial
tumors at the time of clinical presentation.3 Superficial bladder cancers are primarily treated
by transurethral resection (TUR). Despite complete resection of visible tumors, about 60%
of patients will develop a recurrence within 5 years. Intravesical instillation of chemothera-
peutic agents such as doxorubicin, thiotepa, and mitomycin or immunological agents such
as bacillus Calmette-Guerin (BCG) has been shown to reduce the likelihood of recurrence
or progression after TUR. One third of patients will have a tumor recurrence in spite of
intravesical therapy. Progression of the disease, i.e., development of a more advanced stage,
will be recognized in 10% of patients and is of more concern in carcinoma in situ (CIS),
particularly if it is high grade or extensive and multifocal. Intravesical therapy is not with-
out complications such as pyrexia, hematuria, cystitis, and irritative voiding symptoms. Se-
vere adverse events are very uncommon but include BCG sepsis and massive hemorrhage.
Bladder cancer invades through the lamina propria into the submucosa, muscle layer,
and perivesical fat. Also, it spreads hematogenously and via lymphatics to regional lymph
nodes and distant metastatic sites. Clinical stage and depth of invasion correlate well with
the risk of subsequent relapse even if local therapy has been undertaken. Regional lymph
nodes, liver, lung, and bone are the most common sites of metastases. Approximately 30%
of bladder cancer patients are characterized by muscle invasion and/or node positive disease
at initial diagnosis. Radical cystectomy is generally recommended for treatment of muscle

invasive non-metastatic bladder cancer, with or without adjuvant therapies.4 Bladder pre-
serving therapy with chemoradiotherapy may be appropriate in selected patients.5 Recur-
rence after cystectomy is attributed to the presence of micrometastatic disease. Only a small
percentage (20% or less) of patients with metastatic disease can obtain long term disease
free survival with systemic chemotherapy.
Novel approaches are needed for the treatment of bladder cancer, both for recurrent
superficial tumors and for muscle-invasive and metastatic disease. A biologically based ap-
proach designed to reverse the invasive and/or metastatic phenotype is attractive in this
regard. Recently, molecular-based strategies have been investigated extensively for treat-
ment of bladder cancer, including anti-oncogene agents such as antisense oligonucleotides
and ribozymes, and restoration of tumor suppressor gene function.6-12
Molecular Genetics of Bladder Cancer

Many genetic alterations have been identified in bladder cancers. As in other malig-
nancies, the pathogenesis of bladder cancer appears to involve a multi-step process of car-
cinogenesis and disease progression.13-15 These genetic alterations include activation of
oncogenes, inappropriate expression of growth factors and their receptors, and loss or inac-
tivation of tumor suppressor genes. The role of these alterations with respect to tumorige-
nicity and their relation to clinical features have also been investigated. The ras family has
been extensively investigated in bladder cancers. The ras family consists of H-ras, K-ras, and
N-ras, and are important components of the signal transduction pathway.16 Mutated ras
oncogenes activate downstream proteins such as Raf, MAP kinase and MEK, resulting in
uncontrolled phosphorylation of transcription factors that regulate gene expression. Once
a Ras protein is activated, it loses its ability to return to its inactive form, and continues to
stimulate cell proliferation. Recent studies have demonstrated that the activation of the Ras
stimulates the expression of the transcription factor c-Fos. The expression of c-Fos is thought
to be one of the downstream effects of Ras-regulated signal transduction.17 Increased c-Fos
expression is crucial for cell transformation, and activation of tumor growth. The H-ras
oncogene has also been reported to stimulate tumor angiogenesis by activation of vascular
endothelial growth factor (VEGF) and inactivation of angiogenesis inhibitors. 18,19
Neovascularization permits rapid expansion of tumor growth and may increase its meta-
static ability.
The erbB-2 gene is part of the epidermal growth factor receptor family and possesses
tyrosine kinase activity. Overexpression of erbB-2 has been observed in 20-40% bladder
cancers, and the activation of erbB-2 has been correlated with the stage and grade of the
tumor.20 Three oncogenes, int-2, hst1 and bcl-1 are located on the long arm of chromosome
11, a portion of which appears to be amplified in various epithelial tumors including blad-
der cancer.21 int-2 and hst1 genes are members of the fibroblast growth factor family, and
bcl-1 is involved in the apoptotic pathway. These genes are thought to be related to bladder
cancer tumorigenesis. Bcl-2 is known to be an inhibitor of apoptosis. Overexpression of
Bcl-2 has been identified in approximately 30% of bladder tumors.22,23 The relationship
between expression of Bcl-2 and tumor grade or between Bcl-2 expression and patient sur-
vival is controversial. Epidermal growth factor (EGF) is a cellular mitogen which is excreted
into the urine at high levels, and initiates the events of cell replication through the EGF-
receptor (EGFR). Amplification or overexpression of EGFR is frequently present in bladder
cancer, and has been shown to correlate with the stage of the tumor.24
Several tumor suppressor genes such as p53, the retinoblastoma (Rb) gene, p16/CDKN2
and DCC, are thought to be related to tumorigenesis and/or progression of bladder cancers.
Altered expression of p53 has been described in up to 50% of bladder cancers.14 Inactiva-
Applications of Anti-Oncogene Ribozymes for the Treatment of Bladder Cancer 127
tion of p53 can occur by point mutation at many sites, and these mutations are accompa-
nied by the loss of the wild type allele of p53. Some p53 mutations have a transdominant
effect, even in the presence of wild type p53.25 Also, some p53 mutations are thought to have
transforming ability and to act as oncogenes.26 Mutations of the p53 gene and nuclear accu-
mulation of p53 protein are associated with grade and stage of bladder cancers. The detec-
tion of nuclear mutant p53 is associated with a higher risk of recurrence after radical cystec-
tomy, and with lower overall survival.27 p53 mutations are presumed to play an important
role in the multistep progression of bladder cancer. Rb is a nuclear phosphoprotein which
negatively downregulates transcription. Mutation of Rb has been observed in 30-40% of
invasive bladder cancers. Patients with absent or heterogeneous expression of Rb have a
decreased survival compared to that of patients with positive Rb expression.28,29 Further-
more, it has been shown that the introduction of Rb into Rb-negative cell lines decreases
tumorigenicity and tumor growth in vitro and in vivo.11,12 The p16/CDKN2 gene is located
at 9p21, and the deletion of 9p21 has been reported in up to 50% of bladder cancers. CDKN2
is an inhibitor of the cyclin/CDK complexes needed for Rb phosphorylation, and the loss of
CDKN2 results in inactivation of Rb. p16/CDKN2 has been thought to be involved in the
pathogenesis of bladder cancer.30
H-ras Oncogene in Bladder Cancer

The H-ras oncogene was initially discovered as the transforming gene of the T24 hu-
man bladder carcinoma cell line.31,32 Subsequent studies have confirmed that the H-ras
oncogene is activated by a point mutation resulting in the substitution of the encoding
amino acid.33-35 Codons 12, 13 and 61 of the H-ras gene are thought to be “hot spots” for
point mutations.36-38 Extensive studies have reported the prevalence and potential role of
activated H-ras genes in bladder cancer. Overall, the frequency of H-ras mutations has been
reported to range from 3 to 76% in primary bladder tumors, with no consensus to its rela-
tionship to tumor grade and stage.39-48 These discordant results may reflect studies with
small sample sizes, varying assay sensitivities and methodology. One of the technical prob-
lems associated with studies analyzing archived samples is the isolation of pure tumor cells
and the potential for contamination with normal tissue. In addition, the potential coexist-
ence of both wild type and mutated cells in a single tumor may further affect the reported
incidence of H-ras mutations.
Recent reports which have studied the incidence of H-ras mutations in bladder cancer
by molecular-based analyses are summarized in Table 9.1. A 45% incidence of a point mu-
tation in codon 12 of H-ras (GGC to GTC) has been reported, using polymerase chain
reaction (PCR)-based analysis.40 Specifically, there were no H-ras mutants in grade I tu-
mors, with a sharp increase to 44% in grade II, and 65% in grade III tumors. These results
suggest a reasonably high frequency of H-ras mutations in bladder cancer when molecular
based analyses are utilized, as well as a possible association of these mutations with increas-
ing aggressiveness. A second study has also demonstrated the high incidence (67%) of mu-
tations at H-ras codon 12 by PCR after the digestion with the restriction enzyme specific to
the wild type codon 12.41 In this study, mutations of H-ras were also detectable in 48% of
urine samples. Furthermore, Fitzgerald et al42 have shown that 44% (44/100) of urine samples
demonstrate mutations in exon I of the H-ras gene by single-strand conformation poly-
morphism (SSCP) and sequencing. However, not all studies have shown such a high preva-
lence of H-ras mutations in bladder cancer patients. In one study, 9 of 152 samples evalu-
ated by SSCP and digested restriction fragment length polymorphism (RFLP) were found
to have a point mutation, none of which was a GGC to GTC mutation.43 No correlation was
found between tumor grade or stage and the presence of H-ras mutations in this study. By
contrast, using SSCP and RFLP, Levesque et al44 have demonstrated that 30% of specimens
Table 9.1. Incidence of H-ras mutation in bladder cancer
Methods of Detection Number of Number of Cases GGC to GTC Refs.

Cases Examined with Mutation Mutation at
(%) Codon 12
PCR-based Hybridization 67 30 (45) 30/30 40

RFLP 21 14 (67) 14/14 41
RFLP 21* 12 (48) 12/12 41
RFLP+SSCP 152 9 (6) 0/9 43
RFLP+SSCP 111 33 (30) 26/33 44
PCR+Sequencing 50 9 (18) 8/9 45
PCR 97 46 (47) 46/46 46
SSCP+Sequencing 100* 44 (44) ND 42
SSCP+Sequencing 39 1 (3) 0/1 47
RFLP-PCR 50 6 (12) 0/6 48
PCR: polymerase chain reaction
RFLP: restriction fragment length polymorphism
SSCP: single-strand conformation polymorphism
* : urine samples
ND: not defined
harbored H-ras point mutations, with 26 out of 30 cases having a GGC to GTC mutation.
Another study demonstrated a GGC to GTC mutation at codon 12 in 8 out of 9 mutations
by a PCR-based assay.45 In another study, 46 out of 97 (47%) specimens had mutations at
H-ras codon 12, as detected by a hemi-nested, non-isotopic, allele-specific PCR.46 The
simultaneous presence of wild type and mutated H-ras codon 12 was seen in 44 cases out of
46 cases (96%) in this study. Two recent reports have demonstrated a 3% and 12% mutation
incidence of H-ras codon 12 respectively, and none of these were a GGT to GTC mutation.47,48
The exact incidence of H-ras gene mutations in bladder cancer remains controversial,
but the oncogenic role of mutated H-ras has been demonstrated. Increased tumorigenic
and metastatic properties of NIH/3T3 murine fibroblast cells and the RT4 non-invasive
bladder cancer cells have been demonstrated upon transfection of the activated H-ras gene
into these cells.49,50 The mutated H-ras oncogene has also been shown to have the ability to
activate other tumor growth regulation factors such as c-Fos, VEGF, and EGFR.51
Anti-Oncogene Ribozymes Against Bladder Cancer

Cancers develop from a multi-step process, and several genes have already been identi-
fied in this process. The reversion of the malignant phenotype by elimination of a single
activated oncogene or by restoration of a single tumor suppressor gene has been demon-
strated.6-12 Any disease in which a specific gene or protein appears to be important in its
pathogenesis can be targeted with ribozymes. In cancer, oncogenes are obvious potential
targets for ribozyme strategies. The mutated H-ras oncogene might be an appropriate tar-
get for ribozyme-mediated gene therapy, because the ribozyme might have a specific cleav-
age ability, leaving the normal proto-oncogene unaffected.52 A hammerhead ribozyme tar-
geted to the mutated H-ras oncogene has been designed, and has been studied in the bladder
cancer cell line EJ, which contains a GTC mutation at codon 12 (Fig. 9.1). Initial attempts to
examine the cellular effects of an anti-H-ras ribozyme have relied on stable transfection of
eukaryotic plasmids encoding the ribozyme in EJ cells.8 In one study, the anti-H-ras ribozyme
was cloned into the mammalian expression vector pHβ Apr-1-neo, under the expression of
a β-actin promoter. Following transfection of the vector into EJ cells, anti-H-ras ribozyme
expression resulted in the decrease of H-ras mRNA and p21 protein levels. Moreover, clones
with higher ribozyme expression exhibited greater reduction of H-ras mRNA and p21 pro-
tein. This reduction of H-ras gene expression led to altered morphology, cell growth inhibi-
tion, and decreased DNA synthesis in vitro. In vivo studies with these clones have demon-
strated suppressed tumorigenicity, decreased invasive properties, and improved survival of
nude mice. This anti-tumor efficacy was not seen in control EJ cells transfected with the
empty plasmid which did not express the ribozyme. The anti-tumor properties of this anti-
H-ras ribozyme have also been studied in NIH3T3 murine fibroblast cells transfected with
a mutated H-ras gene.53 When the anti-H-ras ribozyme vector was introduced in parent
NIH3T3 cells and in the activated H-ras transformed NIH3T3 cells, cell growth was inhib-
ited only in the mutated H-ras transfected NIH3T3 cells and not in the parent NIH3T3
cells. These findings demonstrated the ability of a ribozyme to discriminate between a nor-
mal and a mutated H-ras transcript, resulting in no deleterious effect on the normal H-ras
proto-oncogene. The anti-tumor activity of a mutant ribozyme which does not possess
cleavage ability was also examined. The results demonstrated the superior anti-tumor effi-
cacy of the active anti-H-ras ribozyme compared to the inactive mutant ribozyme.9 These
studies have established the anti-H-ras ribozyme as a viable strategy for inhibiting the ma-
lignant phenotype of bladder cancer cells.
Although the potential utility of an anti-H-ras ribozyme has been demonstrated in
cell-based assays, efficient delivery systems will be required for future clinical trials. Re-
cently, a recombinant adenovirus encoding the anti-H-ras ribozyme was constructed and
tested in EJ cells.54,55 An anti-H-ras ribozyme sequence was cloned into an E1 deleted aden-
ovirus type 5. The resultant replication-deficient recombinant adenovirus expressing the
anti-H-ras ribozyme was infected into EJ cells. Downregulation of H-ras gene expression
was demonstrated by Northern blot analysis in EJ cells infected by this recombinant aden-
ovirus compared to the control adenoviral vectors lacking expression of the ribozyme. Fur-
ther, treatment with the anti-H-ras ribozyme-expressing adenovirus resulted in growth in-
hibition and reduced viability of EJ cells, while these anti-tumor effects were not observed
in the cells infected with control adenoviruses. This decreased tumorigenicity has also been
demonstrated in mice by subcutaneous injection of EJ cells infected by adenoviruses ex-
pressing an anti-Hras ribozyme. In another study, tumor growth inhibition was demon-
strated in athymic mice after intralesional injection of adenovirus expressing an anti-H-ras
ribozyme.55
Because of its ability to activate cell proliferation and tumor growth, the c-fos oncogene
is thought to be a suitable target for the treatment of cancers. Also, the expression of c-fos is
thought to be partially regulated by Ras related signal transduction. The tumor inhibition
efficacy of an anti-c-fos ribozyme has been investigated in the EJ cell line.56 The anti-c-fos
ribozyme was cloned into the plasmid pMAMneo, which contains a dexamethasone-induc-
ible mouse mammary tumor virus promoter and Rous sarcoma virus-long terminal repeat.
EJ cells were transfected with this anti-c-fos ribozyme-expressing plasmid, and the resultant
transformants were tested with or without dexamethasone. Increased expression of an anti-
fos ribozyme in transfected cells was seen by the addition of dexamethasone. Downregulation
of c-fos RNA expression was seen in the ribozyme-expressing transfectants, along with
changes of cell morphology. The parent EJ cells exhibit a spindled shape and spread rapidly,
whereas cells with anti-c-fos ribozyme expression were rounded in shape and tended to
grow in patches. Also, the downregulation of c-fos expression decreased DNA synthesis and
inhibited cell growth. These results demonstrate the potential utility of anti-oncogene
ribozymes as therapeutics against bladder cancer.
Fig. 9.1. Hammerhead ribozyme

targeted to mutated H-ras onco-
gene. The H-ras gene is frequently
activated by a point mutation at
codon 12, which converts a GGC
sequence encoding glycine to a
GUC sequence encoding valine.
The hammerhead ribozyme has
been designed to cleave the 3' end
of the mutated codon 12.
Delivery of Ribozymes to Bladder Tumors

Efficient ribozyme delivery to a tissue target remains one of the key obstacles to the
development of ribozyme-based therapeutics. Adenoviral vectors are currently being used
in several clinical trials and hold promise as an efficient delivery system for gene therapy.57
Adenoviruses have advantages as a delivery system for urothelial cancer, because of their
tropism for the lower genitourinary epithelium, and are said to infect epithelial cells with
high frequency.58 Successful adenoviral-mediated gene transfer to the bladder epithelium
has been demonstrated.59-61 Human adenovirus type 5 is one of the most well characterized
and utilized for production of a replication-defective adenovirus. The viral genome can be
divided into 4 noncontiguous early regions (E1-E4) and the continuous late regions (L1-L5).62
Deletion of an early region causes a defect in viral DNA replication, and allows for the
accommodation of a DNA insert.63 Wild type adenovirus type 5 can accommodate 2 kb of
insert DNA, and deletion of the E1 region increases the potential insert size an additional
3 kb. Therefore, approximately 5 kb of DNA can be inserted following the deletion of the E1
region.
The efficacy of gene transfer into bladder epithelium and bladder cancer cells has been
studied with the adenoviral vector expressing a marker gene in vitro and in vivo. In one
study, the adenoviral vector expressing the Escherichia coli β-galactosidase gene was inocu-
lated into the rat bladder.59 The expression of the marker gene was seen in excised bladder
wall specimens both by histological staining and PCR based analysis of DNA and RNA. The
maximal expression of the marker gene was detected at 24 hours after the adenoviral inocu-
lation, with progressive decline thereafter. In another study, the transfection efficacy of ad-
enovirus into bladder cancer cells was tested in vitro using adenovirus expressing the firefly
luciferase marker gene.60 The adenovirus was simply added to the cultured cells and incu-
bated for 2 hours. The expression of the marker gene was seen in all 3 transitional carci-
noma cell lines tested (HT1197, HT1376, and T24), and persisted for 7 days. Other investi-
gators utilized orthotopic and peritoneal murine bladder tumors which were treated with
intravesical or intraperitoneal inoculation of the adenovirus expressing β-galactosidase, re-
spectively.61 Gene transduction into bladder mucosa and tumors was confirmed by histol-
ogy and PCR. Taken together, these results indicate that gene transfer into the bladder epi-
thelium or bladder tumor can be achieved by topical administration of an adenoviral vec-
tor, and increase the potential utility of a replication-defective adenovirus as a delivery system
of ribozymes against bladder cancer.
Conclusions
Novel therapies are clearly required both for muscle-invasive and metastatic bladder
tumors, as well as for recurrent superficial tumors. A molecular-based approach may be an
ideal strategy to reverse the malignant phenotype. Tumor specific oncogenes are suitable
targets for ribozyme strategy. To date, efficient tumor inhibition by anti-oncogene ribozymes
targeting H-ras and c-fos has been demonstrated in vitro and in vivo in bladder cancer
models. However, any gene related to the oncogenesis of bladder cancer, including other
oncogenes, growth factors, their receptors, and mutated tumor suppressor genes might also
be suitable targets for a ribozyme-based therapeutic strategy. The delivery of ribozymes
into bladder cancer cells remains a critical issue for future clinical applications. Because of
their anatomy and pathophysiology, bladder tumors are well suited for delivery of genes by
either intravesical instillation or intralesional injection at high concentrations. However,
the development of an effective delivery system will be required to develop anti-bladder
cancer therapeutics based on ribozymes.
References
1. Parker SL, Tong T, Bolden S et al. Cancer Statistics, 1997. CA Cancer J Clin 1997; 47:5-27.
2. Lamm DL, Torti FM. Bladder cancer, 1996. CA Cancer J Clin 1996; 46:93-112.
3. Raghavan D, Shipley WU, Garnick MB et al. Biology and management of bladder cancer.
New Engl J Med 1990; 322:1129-1138.
4. Holmang S, Hedelin H, Anderstrom C et al. Long-term followup of all patients with muscle
invasive (stages T2, T3 and T4) bladder carcinoma in a geographical region. J Urol 1997;
158:389-392.
5. Kaufman DS, Shipley WU, Griffin PP et al. Selective bladder preservation by combination
treatment of invasive bladder cancer. N Engl J Med 1993; 329:1377-1382.
6. Schwab G, Chavany C, Duroux I et al. Antisense oligonucleotides adsorbed to polyalkyl-
cyanoacrylate nanoparticles specifically inhibit mutated Ha-ras mediated cell proliferation
and tumorigenicity in nude mice. Proc Natl Acad Sci USA 1994; 91:10460-10464.
7. Gray GD, Hernandez OM, Hebel D et al. Antisense DNA inhibition of tumor growth in-
duced by c-Ha-ras oncogene in nude mice. Cancer Res 1993; 53:577-580.
8. Kashani-Sabet M, Funato T, Tone T et al. Reversal of the malignant phenotype by an
anti-ras ribozyme. Antisense Res Develop 1992; 2:3-15.
9. Tone T, Kashani-Sabet M, Funato T et al. Suppression of EJ cells tumorigenicity. in vivo
1993; 7:471-476.
10. Werthman PE, Drazan KE, Rosenthal JT et al. Adenoviral-p53 gene transfer to orthotopic
and peritoneal murine bladder cancer. J Urol 1996; 155:753-756.
11. Goodrich DW, Chen Y, Scully P et al. Expression of the retinoblastoma gene product in
bladder carcinoma cells associates with a low frequency of tumor formation. Cancer Res
1992; 52:1968-1973.
12. Takahashi R, Hashimoto T, Xu HJ et al. The retinoblastoma gene functions as a growth
and tumor suppressor in human bladder carcinoma cells. Proc Natl Acad Sci USA 1991;
88:5257-5261.
13. Strohmeyer TG, Slamon DJ. Proto-oncogenes and tumor suppressor genes in human uro-
logical malignancies. J Urol 1994; 151:1479-1497.
14. Reznikoff CA, Belair CD, Yeager TR et al. A molecular genetic model of human bladder
cancer pathogenesis. Semi Oncol 1996; 23:571-584.
15. Sandberg AA, Berger CS. Review of chromosome study in urological tumors. II. Cytoge-
netics and molecular genetics of bladder cancer. J Urol 1994; 151:545-560.
16. Egan SE, Weinberg RA. The pathway to signal achievement. Nature 1993; 365:781-783.
17. Ledwith BJ, Manam S, Kraynak AR et al. Antisense-fos RNA causes partial reversion of the
transformed phenotypes induced by the c-Ha-ras oncogene. Mol Cell Biol 1990;
10:1545-1555.
18. Rak J, Mitsuhasi Y, Bayko L et al. Mutant ras oncogenes upregulate VEGF/VPF expression:
Implication for induction and inhibition of tumor angiogenesis. Cancer Res 1995;
55:4575-4580.
19. Arbiser JL, Moses MA, Fernandez CA et al. Oncogenic H-ras stimulates tumor angiogen-
esis by two distinct pathways. Proc Natl Acad Sci USA 1997; 94:861-866.
20. Moriyama M, Akiyama T, Yamamoto T et al. Expression of c-erbB-2 gene product in uri-
nary bladder cancer. J Urol 1991; 145:423-427
21. Proctor AJ, Coombs LM, Cairns JP et al. Amplification at chromosome 11q13 in transi-
tional cell tumours of the bladder. Oncogene 1991; 6:789-795.
22. Glick SH, Howell LP, Devere White RW. Relationship of p53 and bcl-2 to prognosis in
muscle-invasive transitional cell carcinoma of the bladder. J Urol 1996; 155:1754-1757.
23. King J, Matteson J, Jacobs SC et al. Incidence of apoptosis, cell proliferation and bcl-2
expression in transitional cell carcinoma of the bladder: Association with tumor progres-
sion. J Urol 1996; 155:316-320.
24. Neal DE, Sharples L, Smith K et al. The epidermal growth factor receptor and the progno-
sis of bladder cancer. Cancer 1990; 65:1619-1625.
25. Harvey M, Vogel H, Morris D et al. A mutant p53 transgene accelerates tumor develop-
ment in heterozygous but not nullizygous p53-deficient mice. Nat Genet 1995; 9:305-311.
26. Dittmer D, Pati S, Zambetti G et al. Gain of function mutations in p53. Nat Genet 1993;
4:42-46.
27. Esrig D, Elmajian D, Groshen S et al. Accumulation of nuclear p53 and tumor progression
in bladder cancer. New Engl J Med 1994; 331:1259-1264.
28. Cordon-Cardo C, Wartinger D, Petrylak D et al. Altered expression of the retinoblastoma
gene product: Prognostic indicator in bladder cancer. J Natl Cancer Inst 1992; 84:1251-1256.
29. Logothetis CJ, Xu HJ, Ro JY et al. Altered expression of retinoblastoma protein and known
prognostic variables in locally advanced bladder cancer. J Natl Cancer Inst 1992;
84:1256-1261.
30. Wu Q, Possati L, Montesi M et al. Growth arrest and suppression of tumorigenicity of
bladder-carcinoma cell lines induced by the p16/CDKN2 (p16INK4A, MTS1) gene and
other loci on chromosome 9. Int J Cancer 1996; 65:840-846.
31. Goldfarb M, Shimizu K, Perucho M et al. Isolation and preliminary characterization of a
transforming gene from T24 bladder carcinoma cells. Nature 1982; 296:404-409.
32. Shih C, Weinberg RA. Isolation and transforming sequence from a human bladder carci-
noma cell line. Cell 1982; 29:161-169.
33. Tabin CJ, Bradley SM, Bargmann CI et al. Mechanism of activation of a human oncogene.
Nature 1982; 300:143-149.
34. Reddy EP, Reynolds RK, Santos E et al. A point mutation is responsible for the acquisition
of transforming properties by the T24 human bladder carcinoma oncogene. Nature 1982;
300:149-152.
35. Taparowsky E, Suard Y, Fasano O et al. Activation of the T24 bladder carcinoma trans-
forming gene is linked to a single amino acid change. Nature 1982; 300:762-765.
36. Fujita J, Srivastava SK, Kraus MH et al. Frequency of molecular alterations affecting ras
proto-oncogenes in human urinary tract tumors. Proc Natl Acad Sci USA 1985; 82:
3879-3853.
37. Fujita J, Yoshida O, Yuasa Y et al. Ha-ras oncogenes are activated by somatic alteration in
human urinary tract tumors. Nature 1984; 309: 464-466
38. Visvanathan KU, Pocock RD, Summerhayes IC. Preferential and novel activation of H-ras
in human bladder carcinomas. Oncol Res 1988; 53: 133-139.
39. Burchill SA, Lunec J, Mellon K et al. Analysis of Ha-ras mutations in primary human
bladder tumors. Br J Cancer 1991; 63 (supp):162
40. Czerniak B, Cohen GL, Etkind P et al. Concurrent mutation of coding and regulatory
sequences of the Ha-ras gene in urinary bladder carcinomas. Hum Pathol 1992;
23:1199-1204.
41. Haliassos A, Liloglou T, Likourinas M et al. H-ras oncogene mutations in the urine of
patients with bladder tumors: Description of a non-invasive method for the detection of
neoplasia. Int J Oncol 1992; 1:731-734.
42. Fitzgerald JM, Ramchurren N, Rieger K et al. Identification of H-ras mutations in urine
sediments complements cytology in the detection of bladder tumors. J Natl Cancer Inst
1995; 87:129-133.
43. Knowles MA, Williamson M. Mutation of H-ras is infrequent in bladder cancer: Confir-
mation by single-strand conformation polymorphism analysis, designed restriction frag-
ment length polymorphisms, and direct sequencing. Cancer Res 1993; 53: 133-139.
44. Levesque P, Ramchurren N, Saini K et al. Sequencing of human bladder tumors and urine
sediments for the presence of H-ras mutations. Int J Cancer 1993; 55: 785-790.
45. Burchill SA, Neal DE, Lunec J. Frequency of H-ras mutations in human bladder cancer
detected by direct sequencing. Br J Urol 1994; 73:516-521.
46. Ooi A, Herz F, Ii S et al. Ha-ras codon 12 mutation in papillary tumors of the urinary
bladder. A retrospective study. Int J Oncol 1994; 4:85-90.
47. Uchida T, Wada C, Ishida H et al. Infrequent involvement of mutations on neurofibroma-
tosis type 1, H-ras, K-ras and N-ras in urothelial tumors. Urol Int 1995; 55: 63-67.
48. Saito S, Hata M, Fukuyama R et al. Screening of H-ras gene point mutations in 50 cases of
bladder carcinoma. Int J Urol 1997; 4:178-185.
49. Greig RG, Koestler TP, Trainer DL et al. Tumorigenic and metastatic properties of “nor-
mal” and ras-transfected NIH/3T3 cells. Proc Natl Acad Sci USA 1985; 82:3698-3701.
50. Theodorescu D, Cornil I, Fernandez BJ et al. Overexpression of normal and mutated forms
of HRAS induces orthotopic bladder invasion in a human transitional cell carcinoma. Proc
Natl Acad Sci USA 1990; 87:9047-9051.
51. Theodorescu D, Cornil I, Sheehan C et al. Ha-ras induction of the invasive phenotype
results in up-regulation of epidermal growth factor receptors and altered responsiveness to
epidermal growth factor in human papillary transitional cell carcinoma cells. Cancer Res
1991; 51:4486-4491.
52. Koizumi M, Hayase Y, Iwai S et al. Design of RNA enzymes distinguishing a single base
mutation in RNA. Nucleic Acids Res 1989; 17:7059-7071.
53. Funato T, Shitara T, Tone T et al. Suppression of H-ras-mediated transformation in NIH
3T3 cells by a ras ribozyme. Biochem Pharmacol 1994; 48:1471-1475.
54. Feng M, Carbrera G, Deshane J et al. Neoplastic reversion accomplished by high efficiency
adenoviral-mediated delivery of anti-ras ribozyme. Cancer Res 1995; 55:2024-2028.
55. Irie A, Scanlon KJ. Gene therapy for experimental carcinomas using antioncogene ribozymes.
Cancer Invest 1997; 15 suppl 1:79-80.
56. Shitara T, Scanlon KJ. Suppression of the human carcinoma phenotype by an antioncogene
ribozyme. In: Robbins PD, ed. Gene therapy protocols. New Jersey: Humana press,
1997:391-401.
57. Weitzman MD, Wilson JM, Eck SL. Adenovirus vectors in cancer gene therapy. In: Sobol
RE, Scanlon KJ (eds) The internet book of gene therapy. Stamford: Appleton & Lange,
1995:17-25.
58. Mufson MA, Belshe RB. A review of adenoviruses in the etiology of acute hemorrhagic
cystitis. J Urol 1976; 115:191-194.
59. Morris BD, Drazan KE, Csete ME et al. Adenoviral-mediated gene transfer to bladder in
vivo. J Urol 1994; 153: 506-509.
60. Bass C, Cabrera G, Elgavish A et al. Recombinant adenoviral-mediated gene transfer to
genitourinary epithelium in vitro and in vivo. Cancer Gene Ther 1995; 2: 97-104.
61. Werthman PE, Drazan KE, Rosenthal JT et al. Adenoviral-p53 gene transfer to orthotopic
and peritoneal murine bladder cancer. J Urol 1996; 155: 753-756.
62. Berkner KL. Development of adenovirus vectors for the expression of heterologous genes.
Biotech 1988; 6:616-629.
63. Bett AD, Prevec L, Graham FL. Packaging capacity of human adenovirus type 5 vectors. J
Virol 1993; 67:5911-5921.
CHAPTER 10
Gene Therapy of Breast Cancer

T. Suzuki and B. Anderegg
Introduction
B reast cancer is the most common cancer among women in the United States and the
western world. It is estimated to be responsible for 15–20% of all female cancer related
deaths in 1997.1 Although the mortality rate from breast cancer has been declining since
1989, about a third of all new cancer cases diagnosed in women are localized to the breast.2
Whereas node-negative patients with tumors 1.0 to 2.0 cm in diameter have a good progno-
sis with an average five-year disease-free survival of 85%, this rate decreases dramatically to
approximately 45% in patients with four or more positive nodes.3
Among oncogenes and tumor-suppressor genes, c-erbB2 is an important prognostic
factor in breast cancer.4 In addition, it is probably linked to neoplastic transformation itself
and/or to its promotion.5 Therefore, it is a potential target gene for gene therapy of breast
cancer. In this chapter, the application of ribozymes in the field of breast cancer gene therapy
is discussed based on results regarding the anti-c-erbB-2 ribozyme. Recently, it has been
shown that the growth of breast cancer cells can be efficiently inhibited by anti-c-erbB-2
hammerhead ribozymes.6-8
Anti-c-erbB-2 Ribozyme Strategy

Anti-Oncogene Ribozymes
The utility of catalytic RNAs (hammerhead ribozymes, hairpin ribozymes, etc.) to at-
tenuate mRNA expression has been well documented in a variety of tumors and its applica-
tion to cancer gene therapy has been demonstrated.9,10 Ribozymes have the ability to cleave
RNA species in trans and prevent them from subsequent translation. It has been demon-
strated that anti-oncogene ribozymes of the hammerhead type are effective in reversing the
malignant phenotype of cancer cells in EJ human bladder cancer cells,11-13 FEM human
malignant melanoma cells,14-16 Capan-1 human pancreatic cancer cells,17 and NIH3T3 mouse
fibroblast cells transformed by mutated H-ras.18,19
A critical step in designing any ribozyme is the selection of a suitable target gene. EJ
and FEM cells have a H-ras gene mutated at codon 12 which has been implicated in their
malignant phenotype. An anti-H-ras ribozyme targeting this mutation was introduced into
EJ cells, FEM cells, and NIH3T3 cells transformed by mutated H-ras. The ribozyme was
shown to cleave only the mutated H-ras mRNA containing the GUC mutation at codon 12
without affecting the normal counterpart containing GGC at this site. The ribozyme was
shown to downregulate H-ras expression and reverse the malignant phenotype of the EJ
and FEM cells.
Mutated K-ras is another good target gene for ribozyme cleavage. An anti-K-ras
ribozyme designed against the K-ras mutation found in Capan-1 cells (GUU at codon 12)
has already been shown to downregulate K-ras expression and reverse the malignant phe-
notype effectively.17
The c-erbB-2 Gene

The c-erbB-2 proto-oncogene (also called HER-2/neu) encodes a 185 kDa tyrosine ki-
nase type transmembrane receptor homologous with the epidermal growth factor receptor
(EGFR).20,21 EGFR is known to transform mouse NIH3T3 cells only in the presence of ligand
and receptor22 while the c-erbB-2 protein is constitutively phosphorylated and demonstrates
tyrosine kinase activity without the presence of ligand when overexpressed in NIH3T3 cells.14
Thus, c-erbB-2 overexpressing NIH3T3 cells can be transformed in a ligand-independent
manner.23-25 A growth stimulatory role for c-erbB-2 in human breast and ovarian cancer has
been postulated through its amplification and overexpression.5 c-erbB-2 overexpression
occurs in approximately 30% of all tumors examined.26,27 High levels of c-erbB-2 expres-
sion and poor clinical outcome seem to be correlated, especially in patients with positive
axillary lymph nodes.28-30 c-erbB-2 also has been implicated in the metastatic phenotype.31
Since the c-erbB-2 protein is not expressed in most normal human tissues,32,33
downregulating c-erbB-2 expression at the level of DNA, mRNA, or protein may be a realis-
tic strategy for cancer gene therapy (Table 10.1). Downregulation at the DNA level has been
performed using triplex forming oligonucleotides against c-erbB-2 genomic DNA.34,35 Inhi-
bition at the protein level has been achieved by monoclonal antibodies36-39 and intracellular
expression of single chain antibodies.40-43 In fact, a monoclonal antibody targeting c-erbB-2
protein has been tested in clinical trials of breast cancer.44 Antisense oligonucleotides tar-
geting c-erbB-2 mRNA have been utilized to downregulate the c-erbB-2 mRNA level.45-47
Finally, suppression of gene expression at the mRNA level can be further optimized by uti-
lizing an anti-c-erbB-2 ribozyme.
Designing anti-c-erbB-2 Ribozymes

Since specific c-erbB-2 mutations have not been found in breast cancer,48 an anti-c-
erbB-2 ribozyme was designed to cleave normal c-erbB-2 mRNA. Two different anti-c-erbB-
2 hammerhead ribozymes targeting the GUC sequences at codon 663/664,8 and at the trans-
lation initiation site at codon 71/72,7 respectively, were designed. In vitro studies revealed
the anti-oncogenic potential of these ribozymes in ovarian8 and breast cancer cells7,8
overexpressing the c-erbB-2 gene. The ribozyme targeting codon 71/72 was even capable of
downregulating the growth of breast cancer MCF-7 cells which show low c-erbB-2 protein
levels.7
Anti-c-erbB-2 Ribozyme Expression

Plasmid-Mediated Anti-c-erbB-2 Ribozyme Expression In Vitro
The effect of plasmid-mediated anti-c-erbB-2 ribozyme expression was evaluated by
using stable transformants expressing the anti-c-erbB-2 ribozyme. The anti-c-erbB-2
ribozyme was cloned into the pHβ APr-1-neo expression plasmid and transfected into BT-474
or MCF-7 breast carcinoma cells by electroporation.7 The doubling time of the ribozyme
expressing cells was significantly longer than that of controls.6,7 Anti-c-erb-B-2 ribozyme
expression reversed the phenotype of transformed cells when the ribozyme was expressed
in stable transformants, probably by blocking aberrant growth signals from the receptor to
the nucleus. The in vitro expression of a significant amount of anti-c-erbB-2 ribozyme driven
Gene Therapy of Breast Cancer 137
Table 10.1. Mechanisms of inhibition of c-erbB-2 expression
Method of Inhibition References Comments
monoclonal antibodies 36-39 effective in breast cancer cells overexpressing

c-erbB-2;
successful immunological stimulation in vitro
single chain antibodies 40-43 selective growth inhibition in ovarian and breast
cancer cells overexpressing c-erbB-2;
intracellular expression of the antibody
triplex formation 34, 35 effective and specific transcriptional inhibition in
vitro;
polypurine:polypyrimidine target sequence or
linker molecule required
antisense strategies 45-47 successful growth inhibition of ovarian and breast
cancer cells
overexpressing c-erbB-2 in vitro;
inhibition of serum induced cell spreading
ribozyme strategies 6-8 successful growth inhibition of ovarian and breast
cancer cells independent of the c-erbB-2
expression level;
adenoviral delivery and expression demonstrated
in vitro, in vivo
by the β-actin promoter and downregulation of the steady state c-erbB-2 mRNA level were
clearly correlated.6
Adenovirus-Mediated Anti-c-erbB-2 Ribozyme Expression In Vitro

In order to achieve more efficient gene delivery and expression, a replication deficient
recombinant adenovirus was prepared encoding the anti-c-erbB-2 ribozyme driven by the
CMV promoter (rAdEB2Rz7). The growth of MCF-7 and BT-474 cells was efficiently inhib-
ited by transduction with 500 plaque forming units (PFU)/cell of rAdEB2Rz.7 The control
vector rAdCMVLacZ expressing the LacZ gene instead of the ribozyme did not suppress
tumor growth in vitro, although the transduction efficiency was virtually 100% as analyzed
by X-gal staining.7
The specificity of the growth inhibitory effect produced by the ribozyme on breast
cancer cells was confirmed by transducing several types of recombinant control adenoviruses
into BT-474 cells. Only the recombinant adenovirus expressing an anti-c-erbB-2 antisense
oligonucleotide inhibited breast cancer cell growth, although the effect was not as strong as
that of rAdEB2Rz.7 Northern blot analysis revealed that the anti-c-erbB-2 ribozyme in fact
cleaved c-erbB-2 mRNA and, thus, downregulated c-erbB-2 expression.7
In a recent study, Czubayko et al also used a recombinant adenoviral vector to drive
expression of a hammerhead ribozyme targeting c-erbB-2 at codon 1991.8 In vitro studies
suggested ribozyme expression peaked three days after adenoviral infection and was unde-
tectable by day 10. Moreover, a dose dependent increase in ribozyme expression was noted
in vitro with increasing viral titer. Recombinant adenoviral infection of SK-OV-3 ovarian
cancer cells at a dose of 100 MOI resulted in downregulation of c-erbB-2 mRNA by 75%.
Decreased c-erbB-2 protein levels were demonstrated using fluorescence activated cell sorter
analysis in MDA-MB-361 breast cancer cells. However, no effects of ribozyme expression
on in vitro cell growth of these tumor cell lines were reported.
Adenovirus-Mediated Anti-c-erbB-2 Ribozyme Expression In Vivo

The effects of rAdEB2Rz treatment were examined by subcutaneous inoculation of
BT-474 cells into the flanks of nude mice and subsequent administration of 10 MOI of
rAdEB2Rz or the control vector rAdCMVLacZ into the tumor nodules. The tumorigenicity
was significantly reduced by rAdEB2Rz treatment in comparison with rAdCMVLacZ
application.7
Repeated administration of 10 MOI of rAdEB2Rz inhibited the growth of BT-474 nod-
ules to 20% of the size of control PBS-treated tumors.7 Treatment of the nodules with a
recombinant adenovirus encoding an anti-c-erbB-2 antisense oligonucleotide also inhib-
ited tumor growth, but the effect was weaker than that achieved by rAdEB2Rz. Other aden-
oviral control vectors did not show any significant growth inhibitory effect. rAdEB2Rz in-
jection also inhibited the tumor growth of MCF-7 cells effectively and specifically and
appeared to be superior to the antisense-expressing construct, whereas none of the control
vectors showed any significant effect. The results indicate that the c-erbB-2 protein may be
a critical factor for growth of breast cancer cells in general.
Mechanisms of Ribozyme Action

The mechanisms underlying the inhibition of tumor cell growth by ribozyme-medi-
ated cleavage of c-erbB-2 mRNA remain unknown. Pathological analysis of the adenovirus-
treated tumors showed that there was neither obvious induction of differentiation nor in-
flammatory cell infiltration.7 Reduction of c-erbB-2 expression by a ribozyme may lead to a
shift to alternative signal transduction pathways which slow down cell growth or induce
apoptosis, as reported for single chain antibody-treated cells.43 However, complete responses
were not obtained even by the administration of the rather high viral dose of 500 MOI of
rAdEB2Rz.7 This is probably partially due to the presence of other members of the type I
growth factor receptor family, such as EGFR/c-erbB-1,49 c-erbB-3,50 or c-erbB-4,51 which
may complement the role of c-erbB-2 in cell growth signaling.52 Alternatively, the half life of
c-erbB-2 mRNA may be too long to achieve complete suppression of gene expression in the
model system reported here.
Future Prospects
Although treatment with rAdEB2Rz alone may not be strong enough to eradicate breast
cancer cells completely, it can be expected to enhance the antitumor efficacy of conven-
tional therapeutic modalities. It is especially intriguing to investigate the combination of
anti-c-erbB-2 treatment and cytostatic drugs such as tamoxifen or cis-diamminedichloro-
platinum (cisplatin): c-erbB-2 overexpression is known to be associated with tamoxifen
resistance,53,54 but a monoclonal antibody raised against c-erbB-2 protein was reported to
enhance cisplatin cytotoxity in nude mice.55 Similarly, downregulation of c-erbB-2 by
rAdEB2Rz may enhance the cytotoxity of tamoxifen and, possibly, cisplatin also.
As an alternative to the addition of ribozyme therapy to conventional hormonal and/or
chemotherapies, combination gene therapy should be considered. The contribution of tu-
mor suppressor genes such as BRCA1,56 BRCA2,57 and p5358 to the pathogenesis of breast
cancer has been postulated. Coexpression of these tumor suppressor genes and anti-c-erbB-2
ribozyme may lead to further reduction or even complete eradication of breast cancer cells
in vivo. Moreover, the inhibition of c-erbB-2 may be coupled with cell cycle interruption,
whether through the expression of genes such as p21waf1 or inhibition of the cyclins D1
and E.59-62
Conclusions
Expression of a hammerhead ribozyme targeting c-erbB-2 can suppress human breast
cancer cell growth effectively both in vitro and in vivo. c-erbB-2 expression seems to be
critical for breast cancer growth, especially since inhibition of cell growth can be achieved
irrespective of the basal level of c-erbB-2 expression by the tumor cells. In this respect the
ribozyme approach might be superior to the antisense, triplex, or antibody-based assays; it
remains to be shown that the latter are also sufficient for targeting cells not overexpressing
c-erbB-2. The reported results suggest ribozyme-mediated downregulation of c-erbB-2 as a
reasonable strategy for gene therapy of breast cancer.
Acknowledgments
T. Suzuki is supported by the grant from Uehara Memorial Foundation for Research of
Life Science, Japan.
References
1. Parker SL, Tong T, Bolden S, Wingo PA. Cancer Statistics, 1997. CA Cancer J Clin 1997;
47:5-27.
2. Ries LAG, Kosary CL, Hankey BF et al. SEER Cancer Statistics Review, 1973-1993: Tables
and Graphs (NIH Pub. 96-2789). Bethesda, Md. National Cancer Institute, 1995.
3. Donegan WL. Tumor-related prognostic factors for breast cancer. CA Cancer J Clin 1997;
47:28-51.
4. Marks JR, Humphrey PA, Wu K et al. Overexpression of p53 and HER-2/neu proteins as
prognostic markers in early stage breast cancer. Ann Surg 1994; 219:332-341.
5. Gusterson BA. Identification and interpretation of epidermal growth factor and c-erbB-2
overexpression. Eur J Cancer 1992; 28:263-267.
6. Suzuki T, Tsai J, Curcio LD et al. Anti-c-erbB-2 ribozyme for breast cancer. In: Scanlon
KJ, ed. Therapeutic Application of Ribozymes. Totowa: Humana Press, 1998: (in press).
7. Suzuki T, Ohkawa T, Anderegg B et al. A recombinant adenovirus encoding a ribozyme
targeting c-erbB-2 inhibits tumorigenicity and growth of human breast cancer cells in vivo.
1997; (submitted).
8. Czubayko F, Downing SG, Hsieh SS et al. Adenovirus-mediated transduction of ribozymes
abrogates HER-2/neu and pleiotrophin expression and inhibits tumor cell proliferation.
Gene Ther 1997; 4:943-949.
9. Scanlon KJ, Ohta Y, Ishida H et al. Oligonucleotide-mediated modulation of mammalian
gene expression. FASEB J 1995; 9:1288-1296.
10. Irie A, Kijima H, Ohkawa T et al. Anti-oncogene ribozymes for cancer gene therapy. In:
Gene Therapy. August JT, ed. San Diego: Academic Press, 1997:207-257.
anti-ras ribozyme. Antisense Res Dev 1992; 2:3-15.
12. Feng M, Cabrera G, Deshane J et al. Neoplastic reversion accomplished by high efficiency
adenoviral-mediated delivery of an anti-ras ribozyme. Cancer Res 1995; 55:2024-2028.
13. Tone T, Kashani-Sabet M, Funato T et al. Suppression of EJ cells tumorigenicity. IN VIVO
1993; 7:471-476.
14. Ohta Y, Tone T, Shitara T et al. H-ras ribozyme-mediated alteration of the human mela-
noma phenotype. Ann NY Acad Sci USA 1994; 716:242-253.
15. Ohta Y, Kijima H, Kashani-Sabet M et al. Suppression of the malignant phenotype of
melanoma cells by anti-oncogene ribozymes. J Invest Dermatol 1996; 106:275-280.
16. Ohta Y, Kijima H, Ohkawa T et al. Tissue-specific expression of an anti-ras ribozyme in-
hibits proliferation of human malignant melanoma cells. Nucleic Acids Res 1996; 24:938-942.
17. Kijima H, Bouffard DY, Scanlon KJ. Ribozyme-mediated reversal of human pancreatic car-
cinoma phenotype. In: Ikehara S, Takaku F, Good RA, eds. Proceedings of International
Symposium on Bone Marrow Transplantation. Berlin: Springer, 1996:153-164.
18. Kashani-Sabet M, Funato T, Florenes VA et al. Suppression of the neoplastic phenotype in
vivo by an anti-ras ribozyme. Cancer Res 1994; 54:900-902.
19. Funato T, Shitara T, Tone T et al. Suppression of H-ras-mediated transformation in NIH3T3
cells by a ras ribozyme. Biochem Pharmacol 1994; 48:1471-1475.
20. Bargmann CI, Hung M-C, Weinberg RA. The neu oncogene encodes an epidermal growth
factor receptor-related protein. Nature (Lond.) 1986; 319:226-230.
21. Yamamoto T, Ikawa S, Akiyama T et al. Similarity of protein encoded by the human c-
erbB-2 gene to epidermal growth factor receptor. Nature (Lond.) 1986; 319:230-234.
22. Di Fiore PP, Pierce JH, Fleming TB et al. Overexpression of the human EGF receptor
confers an EGF-dependent transformed phenotype to NIH 3T3 cells. Cell 1987;
51:1063-1070.
23. Lonardo F, Di Marco E, King CR et al. The normal erbB-2 product is an atypical receptor-
like tyrosine kinase with constitutive activity in the absence of ligand. New Biol 1990;
2:992-1003.
24. Hudziak RM, Schlessinger J, Ullrich A. Increased expression of the putative growth factor
receptor p185HER2 causes transformation and tumorigenesis of NIH3T3 cells. Proc Natl
Acad Sci USA 1987; 84:7159-7163.
25. Di Fiore PP, Pierce JH, Kraus MH et al. erbB-2 is a potent oncogene when overexpressed
in NIH/3T3 cells. Science (Washington, DC) 1987; 237:178-182.
26. Slamon DJ, Godolphin W, Jones LA et al. Studies of the HER-2/neu proto-oncogene in
human breast and ovarian cancer. Science (Washington, DC) 1989; 244:707-712.
27. Slamon, DJ, Clark GM, Wong SG et al. Human breast cancer: Correlation of relapse and
survival with amplification of the HER-2/neu oncogene. Science (Washington, DC) 1987;
235:177-182.
28. Perren TJ. c-erbB-2 oncogene as a prognostic marker in breast cancer (Editorial). Br J
Cancer 1991; 63:328-332.
29. Gullick WJ, Love SB, Wright C et al. c-erbB-2 protein overexpression in breast cancer is a
risk factor in patients with involved and uninvolved lymph nodes. Br J Cancer 1991;
63:434-438.
30. Muss HB, Thor AD, Berry DA et al. C-erbB-2 expression and response to adjuvant therapy
in women with node-positive early breast cancer. N Eng J Med 1994; 330:1260-1266.
31. Tan M, Yu D. Overexpression of the c-erbB-2 gene enhanced intrinsic metastasis potential
in human breast cancer cells without increasing their transformation abilities. Cancer Res
1997; 57:1199-1205.
32. Pier GN, Nicortra MR, Bigotti A et al. Expression of the p185 encoded by HER2 oncogene
in normal and transformed human tissues. Int J Cancer 1990; 45:457-461.
33. Press MF, Cordon-Cardo C, Slamon D. Expression of the HER-2/neu proto-oncogene in
normal human adult and fetal tissues. Oncogen 1990; 5:953-962.
34. Ebbinghaus SW, Gee JE, Roudu B et al. Triplex formation inhibits HER-2/neu transcrip-
tion in vitro. J Clin Invest 1993; 92:2433-2439.
35. Noonberb SB, Scott GK, Hunt CA et al. Inhibition of transcription factor binding to the
HER2 promoter by triplex-forming oligodeoxyribonucleotides. Gene 1994; 149:123-126.
36. Hudziak RM, Lewis GD, Winget M et al. p185HER–2 monoclonal antibody has antiprolif-
erative effects in vitro and sensitized human breast cancer cells to tumor necrosis factor.
Mol Cell Biol 1989; 9:1165-1172.
37. Fendly BM, Winget M, Hudziak RM et al. The extracellular domain of HER2/neu is a
potential immunogen for active specific immunotherapy of breast cancer. Cancer Res 1990;
50:1550-1558.
38. Shepard HM, Lewis GD, Sarup JC et al. Monoclonal antibody therapy of human cancer:
Taking the HER2 proto-oncogene to the clinic. J Clin Immunol 1991; 11:117-127.
39. Lewis GD, Figari I, Fendly B et al. Differential responses of human tumor cell lines to
anti-p185HER2 monoclonal antibodies. Cancer Immunol Immunother 1993; 37:255-263.
40. Beerli RR, Wels W, Hynes NE. Intracellular expression of single chain antibodies reverts
erbB2 transformation. J Biol Chem 1994; 269:23931-23936.
41. Deshane J, Loechel F, Conry RM et al. Intracellular single-chain antibody directed against
erbB2 down-regulated cell surface erbB2 and exhibits a selective anti-proliferative effect in
erbB2 overexpressing cancer cell lines. Gene Therapy 1994; 1:332-337.
42. Deshane J, Grim J, Loechel L et al. Intracellular antibody against erbB-2 mediates targeted
tumor cell eradication by apoptosis. Cancer Gene Ther 1996; 3:89-98.
43. Deshane J, Siegel GP, Alvarez RD et al. Targeted tumor killing via an intracellular anti-
body against erbB-2. J Clin Invest 1995; 96:2980-2989.
44. Baselga J, Tripathy D, Mendelsohn J et al. Phase II study of weekly intravenous recombi-
nant humanized anti-p185 HER2 monoclonal antibody in patients with HER2/neu-
overexpressiong metastatic breast cancer. J Clin Oncol 14:737-744.
45. Vaughn JP, Igelehart JD, Demirdji S et al. Antisense DNA downregulation of the ERBB2
oncogene measured by a flow cytometric assay. Proc Natl Acad Sci USA 1995; 92:8338-8342.
46. Wiechen K, Dietel M. c-erbB-2 anti-sense phosphorothioate oligodeoxynucleotides inhibit
growth and serum-induced cell spreading of p185c–erbB–2 overexpressing ovarian carcinoma
cells. Int J Cancer 1995; 63:604-608.
47. Liu X and Pogo B. Inhibition of erbB-2-positive breast cancer cell growth by erbB-2 antisense
oligonucleotides. Antisense Res Dev 1996; 6:9-16.
48. Pier GN, Nicortra MR, Bigotti A et al. Expression of the p185 encoded by HER2 oncogene
in normal and transformed human tissues. Int J Cancer 1990; 45:457-461.
49. Ullrich A, Coussens L, Hayflick JS et al. Human epidermal growth factor receptor cDNA
sequence and aberrant expression of the amplified gene in A431 epidermoid carcinoma
cells. Nature (Lond.) 1984; 309:418-425.
50. Kraus MH., Issing W, Miki T et al. Isolation and characterization of ERB-B-3, a third
member of the ERBB/epidermal growth factor receptor family: evidence for overexpression
in a subset of human mammary tumors. Proc Natl Acad Sci USA 1989; 86:9193-9197.
51. Plowman GD, Culouscou JM, Whitney GS et al. Ligand-specific activation of HER4/p180
(erbB4), a 4th member of the epidermal growth factor receptor family. Proc Natl Acad Sci
USA 1993; 90:1746-1750.
52. Kokai Y, Myers JN, Wada T et al. Synergistic interaction of p185c-neu and the EGF recep-
tor leads to transformation of rodent fibroblast. Cell 1989; 58:287-292.
53. Borg A, Baldetorp B, Ferno M et al. ErbB2 amplification is associated with tamoxifen re-
sistance in steroid-receptor positive breast cancer. Cancer Lett 1994; 81:137-144.
54. Pietras RJ, Arboleda J, Reese DM. HER-2 tyrosine kinase pathway targets estrogen receptor
and promotes hormone-independent growth in human breast cancer cells. Oncogene 1995;
10:2435-2446.
55. Hancock MC, Langton BC, Chan T et al. A monoclonal antibody against the c-erbB-2
protein enhances the cytotoxity of cis-diamminedichloroplatinum against human breast
and ovarian tumor cells. Cancer Res 1991; 51:4575-4580.
56. Miki Y, Swensen J, Shattuck-Eidens D et al. A strong candidate for the breast and ovarian
cancer susceptibility gene BRCA1. Science (Washington, DC) 1994; 266:66-71.
57. Wooster R, Bignell G, Lancaster J et al. Identification of the breast cancer susceptibility
gene BRCA2. Nature (Lond.) 1995; 378:789-792.
58. Malkin D, Li FP, Strong LC et al. Germ line p53 mutations in a familial syndrome of
breast cancer, sarcomas, and other neoplasms. Science (Wash. DC) 1990; 250:1233-1238.
59. Katayose D, Wersto R, Cowan KH et al. Effects of a recombinant adenovirus expressing
WAF/Cip1 on cell growth, cell cycle, and apoptosis. Cell Growth Differ 1995; 6:1207-1212.
60. Buckley MF, Sweeney KJE, Hamilton JA et al. Expression and amplification of cyclin genes
in human breast cancer. Oncogene 1993; 8:2127-2133.
61. Lammie GA, Fantl V, Smith R et al. D11S287, a putative oncogene on chromosome 11q13,
is amplified and expressed in squamous cell and mammary carcinomas and linked to BCL-
1. Oncogene 1991; 6:439-444.
62. Keyomarsi K, O’Leary N, Molnar G et al. Cyclin E, a potential prognostic marker for breast
cancer. Cancer Res 1994; 54:380-385.
CHAPTER 11
Therapeutic Application
of an Anti-ras Ribozyme
in Human Pancreatic Cancer
Hiroshi Kijima and Kevin J. Scanlon
Abstract
P ancreatic cancer is one of the most lethal human cancers, and development of new thera-
peutic strategies is urgently required. Point mutation in the K-ras gene is observed at a
high incidence in human pancreatic carcinomas. These alterations can be used as potential
targets for specific ribozyme-mediated reversal of the malignant phenotype. We have dem-
onstrated previously the efficacy of a hammerhead ribozyme directed against codon 12 of
the activated K-ras gene in a Capan-1 human pancreatic carcinoma cell line. To develop this
strategy into a therapeutic application, we designed a recombinant adenovirus encoding a
gene cassette for the anti-K-ras ribozyme. By using this recombinant adenovirus in a mu-
rine model system, it was possible to accomplish efficient reversion of the malignant phe-
notype in human pancreatic tumors with K-ras gene mutations. The high efficiency aden-
oviral-mediated delivery of anti-oncogene ribozyme could emerge as a significant gene
therapy strategy against human malignancies.
Introduction
Pancreatic cancer is a lethal human malignancy with a less than 10% 3-year survival.1,2
The factors which account for this poor prognosis include:
1. difficulty of early diagnosis due to anatomical location and lack of early symptoms;
2. limitation of conventional cancer therapy, including surgery, chemotherapy, radia-
tion therapy or immune therapy;
3. rapid tumor spread to the surrounding organs, causing obstructive jaundice; and
4. frequent incidence of metastasis from even a small primary tumor less than 2 cm in
diameter.3-5
Pancreatic cancer ranks fifth as a cause of cancer-related mortality in the United States,
as well as Japan. Development of a new therapeutic strategy such as gene therapy for pan-
creatic cancer represents one of the most pressing issues in medicine today.6,7
Cancer cells have been shown to have alterations of oncogene expression such as point
mutation, amplification or overexpression.8,9 A point mutation of the ras oncogene family
activates the p21 gene products affecting cancer cell growth and the malignant phenotype.10
Characteristically, K-ras gene mutations have been found in approximately 90% of human
pancreatic adenocarcinomas.11-17 Most of the tumors examined harbored activated K-ras

genes with mutations at codon 12. The K-ras mutation could occur in the early phase of
pancreatic ductal carcinogenesis because this mutation has also been found in some pan-
creatic precancerous lesions, such as mucous cell hyperplasia.18-20 However, the activated ras
oncogene products are thought to alter the cellular signal transduction pathways and to
affect neoplastic growth of the pancreatic cells.
Recent advances in the understanding of the genetic mechanisms of carcinogenesis
and the manipulation of gene expression have introduced new therapeutics for cancer in-
cluding gene therapy. Specific gene modulation using oligonucleotides have been devel-
oped and defined as a plausible strategy for suppressing activated oncogenes.21-24 Oligo-
nucleotides modulating target gene expression include triplex DNA, antisense DNA/RNA
and ribozymes (catalytic RNA).21 Antisense oligonucleotides are capable of altering the in-
termediary metabolism of mRNA and inhibiting the transfer of information from gene to
protein. Antisense-mediated gene modulation has been shown to be effective for gene
therapy.25-27 In contrast, ribozymes have been characterized as RNA molecules having site-
specific catalytic activity.28,29 Trans-acting ribozyme molecules, such as “hammerhead” and
“hairpin” ribozymes, include catalytic core and flanking sequences which bind the target
RNA. Ribozymes are modified antisense molecules with site-specific cleavage activity and
catalytic potential.22,30-34 In recent years, researchers have demonstrated the efficacy of
ribozymes against various oncogenes (e.g., ras, c-fos, BCR-ABL),35 the drug resistance genes
(e.g., mdr1)36,37 and the human immunodeficiency virus type 1.31,38,39 We have previously
demonstrated that anti-oncogene ribozymes effectively suppress the expression of targeted
genes and result in reversal of the malignant phenotype in human cancer cells.40-47 Because
of their in vitro efficacy, the anti-oncogene ribozymes have been proposed to have future
clinical utilities as anticancer agents.
In a recent study, we investigated the in vivo efficacy of a hammerhead ribozyme against
activated K-ras gene transcripts in human pancreatic carcinoma cells. For future clinical
trials of cancer gene therapy using ribozymes, we have evaluated an efficient adenoviral-
mediated delivery system of the anti-K-ras ribozyme as a therapeutic agent against human
pancreatic carcinoma.
Results
Double-stranded DNA cycle sequencing of genomic DNA isolated from Capan-1 hu-
man pancreatic cancer cells exhibited a GTT homozygous mutation of the K-ras oncogene
at codon 12 which encodes a valine (data not shown). This sequence is recognized by the
anti-K-ras hammerhead ribozyme which we have designed for this study (Fig. 11.1). For
expressing the anti-K-ras ribozyme, oligonucleotides KrasRz-1 and -2 were synthesized and
cloned into an adenoviral shuttle vector, the pACCMVpLpA plasmid. The methodology for
recombinant adenovirus construction is based on in vivo homologous recombination be-
tween the adenoviral shuttle vector pACCMVpLpA and the adenoviral packaging plasmid
pJM17. The adenoviral vector Ad-Kras Rz contains the anti-K-ras ribozyme expression cas-
sette inserted in place of the deleted adenoviral E1 region. The PCR assay of viral DNA
demonstrated the presence of the anti-K-ras ribozyme in the recombinant adenovirus (data
not shown).
In a tissue culture study, the Capan-1 cells infected with Ad-Kras Rz exhibited expres-
sion of the anti-K-ras ribozyme and decreased K-ras gene expression (data not shown). In
the Capan-1 cells infected with Ad-Kras Rz, the generation time was significantly longer by
4.3 times compared to the parental cells (Table 11.1). Also, the Capan-1 cells infected with
Ad-Kras Rz showed 47% and 83% decrease in [3H] thymidine incorporation and soft agar
colony formation, respectively.
Therapeutic Application of an Anti-ras Ribozyme in Human Pancreatic Cancer 145
Fig. 11.1. Design of the recombinant adenovirus encoding the anti-K-ras ribozyme. The ham-
merhead ribozyme against K-ras targets the GUU mutant mRNA sequence of K-ras codon 12,
which encodes valine (top of the figure). The GGU wild type sequence encoding glycine is not
cleaved by the ribozyme. The pACCMVpLpA shuttle vector of the recombinant adenovirus is
driven by the cytomegalovirus (CMV) promoter (bottom of the figure). The insert containing
the anti-K-ras ribozyme sequence is cloned between the promoter and SV40 poly A.
In an in vivo system, inoculation of 1 x 106 Capan-1 cells subcutaneously into the flanks
of athymic mice resulted in the rapid development of progressively enlarging tumors
(Fig. 11.2). The efficacy of Ad-KrasRz was investigated using a single intralesional injection
of the ribozyme-encoding adenovirus at a viral dose of 1 x 108 PFU when the tumor volume
reached 100 mm3. Overall, treatment with Ad-KrasRz resulted in profound suppression of
Capan-1 cell growth in vivo (Fig. 11.2), with the average tumor volume on day 40 approxi-
mately one-third that of control tumors. Significantly, of the 20 mouse tumors treated with
Table 11.1. Growth characteristics of the Capan-1 pancreatic carcinoma cells
Cell Line GT1 [3H]Thd2 SAC3
Capan-1 (parent) 100% (59 hrs) 100% 100% (121)

Capan-1 / Ad dl312 (E1-) 102 99 92
Capan-1 / Ad anti-K-ras Rz 432 53 17
1 GT, generation time (hours);
2 [3H]Thd, the rate of [3H]-labeled thymidine incorporation compared to parental cells
3 SAC, the rate of soft agar colony formation assay with 20% fetal bovine serum compared to
parental cells (colony number provided in parentheses).

1-3 Experiments were performed at least twice in duplicate. Standard deviation was less than 10%.
Fig. 11.2. Effect of adenovirus-

mediated anti-K-ras ribozyme
(Ad-Kras Rz) treatment on
growth inhibition of Capan-1
tumors in athymic nude mice.
Injection of the Ad-Kras Rz sup-
pressed tumor growth of the
Capan-1 cells, compared to the
untreated control cells or the
Capan-1 cells injected with the
irrelevant adenovirus (Ad dl312;
Ad-vector only). The graph
shows data regarding thirteen
Capan-1 tumors injected with
Ad-Kras Rz; the other seven
Capan-1 tumors completely re-
gressed after Ad-Kras Rz treat-
ment.
Ad-KrasRz, seven regressed completely. Finally, the effects of vector-mediated cytotoxicity

were investigated using an irrelevant adenovirus (Ad dl312). Treatment of Capan-1 tumors
in vivo with Ad dl312 using an identical schedule to that of Ad-KrasRz had little effect on
the rate of tumor growth (Fig. 11.2).
Discussion
In the present study, we have demonstrated the efficacy of a hammerhead ribozyme
against the mutant K-ras gene in the Capan-1 human pancreatic carcinoma cell line. Re-
cently, other groups have reported the efficacy of antisense molecules against the K-ras gene
to suppress human pancreatic tumor growth in vitro and in vivo in mouse models.48,49
Potential advantages of ribozyme-mediated gene modulation include its catalytic activity,
site-specific cleavage and ability to discriminate a single base mutation.21,22,50 For examina-
tion of ribozyme-mediated strategies, investigators should take into account determinants

of an effective ribozyme, such as the target gene, cleavage site, flanking sequences, ribozyme
stability and delivery systems.22,31,36,51
Point mutations of the K-ras gene at codon 12 are found in approximately 90% of
human pancreatic carcinomas, and activate its oncogene product.10-17 This mutant mRNA
can be a candidate for ribozyme-mediated gene modulation. The hammerhead ribozyme
targeting the mutant K-ras gene we have designed could cleave mutant but not wild type
mRNA.51 The ultimate goal of this study is therapeutic application of the anti-K-ras ribozyme
in human pancreatic cancer.6,30,52-55 Therefore, to express the anti-K-ras ribozyme in Capan-
1 tumors in a mouse model, we have cloned it into the adenoviral vector and generated Ad-
Kras Rz. The adenoviral vector is driven by a cytomegalovirus early promoter/enhancer and
polyadenylation signals from SV40. This system has been shown to express high amounts of
the inserted transgene in human cells.46,56
In a tissue culture study, we demonstrated the efficacy of Ad-Kras Rz in inhibiting
human pancreatic cancer cell proliferation.6 The Capan-1 pancreatic carcinoma cells were
effectively infected with Ad-Kras Rz, and their generation time was 4.3 times longer than
control cells. The anti-K-ras ribozyme affected not only the generation time, but also other
growth characteristics, such as thymidine incorporation into DNA and soft agar colony
formation. Thus the sequence-specific modulation of mutant K-ras mRNA by the hammer-
head ribozyme was capable of reversing the malignant phenotype of human pancreatic
carcinoma cells.
In our in vivo system, xenograft implantation of the Capan-1 cells in athymic mice
resulted in the rapid development of subcutaneous tumors. The Capan-1 tumor nodules
injected with the irrelevant adenovirus (Ad dl312) result in a similar pattern of rapid tumor
progression, suggesting lack of vector-associated toxicity. Injection of the Ad-Kras Rz sup-
pressed the rapid tumor growth effectively. These results indicated that adenovirus-medi-
ated anti-K-ras ribozyme targeted mutant K-ras mRNA transcripts and was capable of in-
hibiting neoplastic progression of pancreatic cancer in vivo.
In this study using a recombinant adenoviral vector, we demonstrated high-efficiency
in vivo delivery of the anti-K-ras ribozyme to human pancreatic carcinoma cells. However,
the adenovirus we used was driven by the cytomegalovirus early promoter/enhancer, and
not by a tissue-specific promoter.22,57,58 For future clinical application of ribozyme-medi-
ated gene modulation, tissue-specific promoters will be investigated. One potential pro-
moter is the carcinoembryonic antigen (CEA) promoter, because pancreatic carcinomas
frequently produce high amounts of CEA.59,60 When targeting a tumor-selective gene ex-
pressed in a tissue-specific manner, ribozymes may offer minimal toxicity for the applica-
tion of cancer gene therapy.
Conclusion
The anti-K-ras ribozyme is an effective modulator of mutant K-ras gene expression
because of its site-specific cleavage activity. Adenovirus-mediated ribozyme expression sup-
pressed tumor growth of the target pancreatic carcinoma in the in vivo system. Further
studies of an optimal vector with a tissue-specific promoter are required to advance the
anti-K-ras ribozyme toward clinical application of pancreatic cancer gene therapy.
Acknowledgments
This work was supported in part by Grants-in-Aid for Scientific Research from the
Ministry of Education, Science and Culture of Japan (H.K.) and Tokai University School of
Medicine Research Aid (H.K.).
References
1. Warshaw AL, Fernandez-del Castillo C. Pancreatic carcinoma. N Engl J Med 1992;
326:455-456.
2. Yamaguchi K, Enjoji E. Carcinoma of the pancreas: A clinicopathologic study of 96 cases
with immunohistochemical observations. Jpn J Clin Oncol 1989; 19:14-22.
3. Ozaki H. Improvement of pancreatic cancer treatment from the Japanese experience in the
1980's. Int J Pancreatol 1992; 12:5-9.
4. Arbuck SG. Overview of chemotherapy for pancreatic cancer. Int J Pancreatol 1990;
7:209-222.
5. Cohn I Jr. Overview of pancreatic cancer, 1989. Int J Pancreatol 1990; 7:1-11.
6. Kijima H, Bouffard DY, Scanlon KJ. Ribozyme-mediated reversal of human pancreatic car-
cinoma phenotype. In: Ikehara S, Takaku F, Good RA, eds. Bone Marrow Transplanta-
tion—Basic and Clinical Studies. Tokyo: Springer, 1996:153-163.
7. Kijima H, Bouffard DY, Scanlon KJ. Ribozymes as a novel approach for the treatment of
human pancreatic carcinoma. Methods Mol Med 1998; 11:193-208.
8. Egan SE, Weinberg RA. The pathway to signal achievement. Nature 1993; 365:781-783.
9. Slamon DJ, deKernion JB, Verma IM, Cline MJ. Expression of cellular oncogenes in hu-
man malignancies. Science 1984; 224:256-262.
10. Barbacid M. ras genes. Annu Rev Biochem 1987; 56:779-827.
11. Bos JL. ras oncogenes in human cancer: A review. Cancer Res 1989; 49:4682-4689.
12. Almoguera C, Shibata D, Forrester K, Martin J, Arnheim N, Perucho M. Most human
carcinomas of the exocrine pancreas contain mutant c-K-ras genes. Cell 1988; 53:549-554.
13. Smit VTHBM, Boot AJM, Smit AMM, Fleuren GJ, Cornelisse CJ, Bos JL. Kras codon 12
mutations occur very frequently in pancreatic adenocarcinomas. Nucleic Acids Res 1988;
16:7773-7782.
14. Grunewald K, Lyons J, Frohlich A, Feichtinger H, Weger RA, Schwab G, Janssen JWG,
Bartram CR. High frequency of Ki-ras codon 12 mutations in pancreatic adenocarcino-
mas. Int J Cancer 1989; 43:1037-1041.
15. Mariyama M, Kishi K, Nakamura, K, Obata H, Nishimura S. Frequency and types of point
mutation at the 12th codon of the c-Ki-ras gene found in pancreatic cancers from Japa-
nese patients. Jpn J Cancer Res 1989; 80:622-626.
16. Tada M, Yokosuka O, Omata M, Ohto M, Isono K. Analysis of ras gene mutations in
biliary and pancreatic tumors by polymerase chain reaction and direct sequencing. Cancer
1990; 66:930-935.
17. Nagata Y, Abe M, Motoshima K, Nakayama E, Shiku H. Frequent glycine-to-aspartic acid
mutations at codon 12 of c-Ki-ras gene in human pancreatic cancer in Japanese. Jpn J
Cancer Res 1990; 81:135-140.
18. Yanagisawa A, Ohtake K, Ohashi K, Hori M, Kitagawa T, Sugano H, Kato Y. Frequent c-
K-ras oncogene activation in mucous cell hyperplasia of pancreas suffering from chronic
inflammation. Cancer Res 1993; 53:953-956.
19. DiGiuseppe JA, Hruban RH, Offerhaus GJ, Clement MJ, van den Berg FM, Cameron JL,
van Mansfeld ADM. Detection of K-ras mutations in mucinous pancreatic duct hyperpla-
sia from a patient with a family history of pancreatic carcinoma. Am J Pathol 1994;
144:889-895.
20. Tada M, Ohashi M, Shiratori Y, Okudaira T, Komatsu Y, Kawabe T, Yoshida H, Machinami
R, Kishi K, Omata M. Analysis of K-ras gene mutation in hyperplastic duct cells of the
pancreas without pancreatic disease. Gastroenterol 1997; 110:227-231.
21. Scanlon KJ, Ohta Y, Ishida H, Kijima H, Ohkawa T, Kaminski A, Tsai J, Horng G, Kashani-
Sabet M. Oligonucleotide-meiated modulation of mammalian gene expression. FASEB J
1995; 6:1288-1296.
22. Kashani-Sabet M, Scanlon KJ. Application of ribozymes to cancer gene therapy. Cancer
Gene Ther 1995; 2:213-221.
38:2023-2037.
24. Helene C. Control of oncogene expression by antisense nucleic acids. Eur J Cancer 1994;
30A:1721-1726.
25. Georges RN, Mukhopadhyay T, Zhang Y, Yen N, Roth JA. Prevention of orthotopic hu-
man lung cancer growth by intratracheal instillation of a retroviral antisense K-ras con-
struct. Cancer Res 1993; 53:1743-1746.
26. Zhang Y, Mukhopadhyay T, Donehower LA, Georges RN, Roth JA. Retroviral vector-me-
diated transduction of K-ras antisense RNA into human lung cancer inhibits expression of
the malignant phenotype. Human Gene Ther 1993; 4:451-460.
27. Mercola D, Cohen JS. Antisense approach to cancer gene therapy. Cancer Gene Ther 1995;
2:47-59.
28. Castanotto D, Rossi JJ, Sarver N. Antisense catalytic RNAs as therapeutic agents. Adv
Pharmacol 1994; 25:289-317.
29. Symons RH. Ribozymes. Curr Opin Structural Biol 1994; 4:322-330.
30. Kijima H, Ishida H, Ohkawa T, Kashani-Sabet M, Scanlon KJ. Therapeutic applications of
ribozymes. Pharmacol Ther 1995; 68:247-267.
31. Irie A, Kijima H, Ohkawa T, Bouffard DY, Suzuki T, Curcio LD, Holm PS, Sassani A,
Scanlon KJ. Anti-oncogene ribozymes for cancer gene therapy. Adv Pharmacol 1997;
40:207-257.
32. Pley HW, Flaherty KM, McKay DB. Three-dimensional structure of a hammerhead
ribozyme. Nature 1994; 372:68-74.
34. Fujita S, Koguma T, Ohkawa J, Mori K, Kohda T, Kise H, Nishikawa S, Iwakuma M, Taira
K. Discrimination of a single base change in a ribozyme using the gene for dihydrofolate
reductase as a selective marker in Escherichia coli. Proc Natl Acad Sci USA 1997; 94:391-396.
35. Bouffard DY, Ohkawa T, Kijima H, Irie A, Suzuki T, Curcio LD, Holm PS, Sassani A,
Scanlon KJ. Oligonucleotide modulation of multidrug resistance. Eur J Cancer 1996;
32A:1010-1018.
36. Ohkawa T, Kijima H, Irie A, Horng, G, Kaminski A, Tsai J, Kashfian BI, Scanlon KJ. Oli-
gonucleotide modulation of multidrug resistance gene expression. In: Gupta S, Tsuruo T,
eds. Multidrug Resistance in Cancer Cells. England: Wiley & Sons, 1996:413-433.
37. Sarver N, Cantin EM, Chang PS, Zaia JA, Ladne PA, Stephens DA, Rossi JJ. Ribozymes as
potential anti-HIV-1 therapeutic agents. Science 1990; 247:1222-1225.
38. Yu M, Ojwang J, Yamada O, Hample A, Rapapport J, Looney D, Wong-Staal F. A hairpin
ribozyme inhibits expression of diverse strains of human immunodeficiency virus type-1.
39. Scanlon KJ, Jiao L, Funato T, Wang W, Tone T, Rossi JJ, Kashani-Sabet M. Ribozyme-
mediated cleavage of c-fos mRNA reduces gene expression of DNA synthesis enzymes and
metallothionein. Proc Natl Acad Sci USA 1991; 88:10591-10595.
40. Kashani-Sabet M, Funato T, Tone T, Jiao L, Wang W, Yoshida E, Kashfian BI, Shitara T,
Wu AM, Moreno JG, Traweek ST, Ahlering TE, Scanlon KJ. Reversal of the malignant
phenotype by an anti-ras ribozyme. Antisense Res Dev 1992; 2:3-15.
41. Tone T, Kashani-Sabet M, Funato T, Shitara T, Yoshida E, Kashfian BI, Horng M, Fodstad
O, Scanlon KJ. Suppression of EJ cells tumorigenicity. In Vivo 1993; 7:471-476.
42. Funato T, Shitara T, Tone T, Jiao L, Kashani-Sabet M, Scanlon KJ. Suppression of H-ras-
mediated transformation in NIH3T3 cells by a ras ribozyme. Biochem Pharmacol 1994;
48:1471-1475.
43. Ohta Y, Tone T, Shitara T, Funato T, Jiao L, Kashfian BI, Yoshida E, Horng M, Tsai P,
Lauterbach K, Kashani-Sabet M, Florenes VA, Fodstad O, Scanlon KJ. H-ras ribozyme-
mediated alteration of the human melanoma phenotype. Ann NY Acad Sci 1994;
716:242-253.
44. Kashani-Sabet M, Funato T, Florenes VA, Fodstad O, Scanlon KJ. Suppression of the neo-
plastic phenotype in vivo by an anti-ras ribozyme. Cancer Res 1994; 54:900-902.
45. Scanlon KJ, Ishida H, Kashani-Sabet M. Ribozyme-mediated reversal of the multidrug-
resistant phenotype. Proc Natl Acad Sci USA 1994; 91:11123-11127.
46. Feng M, Cabrera G, Deshane J, Scanlon KJ, Curiel DT. Neoplastic reversion accomplished
by high efficiency adenoviral-mediated delivery of an anti-ras ribozyme. Cancer Res 1995;
55:2024-2028.
47. Ohta Y, Kijima H, Kashani-Sabet M, Scanlon KJ. Suppression of the malignant phenotype
of melanoma cells by anti-oncogene ribozymes. J Invest Dermatol 1996; 106:275-280.
48. Carter G, Gilbert C, Lemoine NR. Effects of antisense oligonucleotides targeting K-ras ex-
pression in pancreatic cancer cell lines. Int J Oncol 1995; 6:1105-1112.
49. Aoki K, Yoshida T, Sugimura T, Terada M. Liposome-mediated in vivo gene transfer of
antisense K-ras construct inhibits pancreatic tumor dissemination in the murine perito-
neal cavity. Cancer Res 1995; 55:3810-3816.
50. Koizumi M, Hayase Y, Iwai S, Kamiya H, Inoue H, Ohtsuka E. Design of RNA enzymes
distinguishing a single base mutation in RNA. Nucleic Acids Res 1989; 17:7059-7071.
51. Cech TR, Uhlenbeck OC. Hammerhead nailed down. Nature 1994; 372:39-40.
52. Jolly D. Viral vector systems for gene therapy. Cancer Gene Ther 1994; 1:51-64.
53. Hodgson CP. The vector void in gene therapy. Biotechnology 1995; 13:222-225.
54. Anderson WF. Human gene therapy. Science 1992; 256:808-813.
55. Thompson JD, Macejak D, Couture L, Stinchcomb DT. Ribozymes in gene therapy. Na-
ture Med 1995; 1:277-278.
56. Lieber A, Kay MA. Adenovirus-mediated expression of ribozymes in mice. J Virol 1996;
70:3153-3158.
57. Ohta Y, Kijima H, Kashani-Sabet M, Scanlon KJ. Tissue-specific expression of an anti-ras
ribozyme inhibits proliferation of human malignant melanoma cells. Nucleic Acids Res
1996; 24:938-942.
58. Lan K-H, Kanai F, Shiratori Y, Ohashi M, Tanaka T, Okudaira T, Yoshida Y, Hamada H,
Omata M. In vivo selective gene expression and therapy mediated by adenoviral vectors
for human carcinoembryonic antigen-producing gastric carcinoma. Cancer Res 1997;
57:4279-4284.
59. Schrewe H, Thompson J, Bona M, Hefta LA, Maruya A, Hassauer, M, Shively JH, von
Kleist S, Zimmermann W. Cloning of the complete gene for carcinoembryonic antigen:
Analysis of its promoter indicates a region conveying cell type-specific expression. Mol
Cell Biol 1990; 10:2738-2748.
60. Hauck W, Stanners CP. Transcriptional regulation of the carcinoembryonic antigen gene.
J Biol Chem 1995; 270:3602-3610.
CHAPTER 12
The Use of Ribozymes

for Gene Therapy of Lung Cancer
Alex W. Tong, Yu-An Zhang, David Y. Bouffard, John Nemunaitis
Summary
T he adverse prognosis of patients having common oncogene mutations (ras, p53, c-erbB-2)
suggests that these genetic lesions contribute to tumor pathogenesis. Currently, the ma-
jority of lung cancer gene therapy makes use of antisense oligonucleotides (AS-ODN) or
expression vectors (such as a viral vector construct) that deliver the antisense sequence to
inactivate the mutant oncogene. The specific targeting of ras, c-myc, L-myc, bcl-2, c-kit, and
mutant p53 by AS-ODNs collaterally suppressed lung cancer cell growth in vitro by 40-90%.
Ribozymes present a viable alternative in antisense therapy by virtue of their renewable
catalytic capability for site-specific RNA cleavage. We recently examined the growth-modu-
lating effect of a hammerhead ribozyme that is specific for the K-ras codon 12 mutant se-
quence GUU, given the findings that:
1. in the United States, approximately 30% of human non-small cell lung cancers
(NSCLC) express K-ras oncogene mutations, nearly all of which reside in codon 12;
2. anti-K-ras, anti-H- as well as anti-N-ras hammerhead ribozymes are potent growth
inhibitors in various human cancers tested; and
3. in vitro and animal model studies suggest that ribozymes directed at oncoproteins
(K- and H-Ras, c-Fos, BCR-ABL) or human immunodeficiency viral proteins are
more effective than their antisense counterpart.
We examined the growth inhibitory effect of a K-ras ribozyme, using the human lung
adenocarcinoma cell line H1725 with a heterozygous GGT→GTT mutation at K-ras codon
12. H1725 cell growth was reduced by 81% following treatment with a pHb Apr-1-neo
plasmid construct with the relevant ribozyme (KRbz) sequence against GTT, whereas the
growth rate of mock-transfected cultures was reduced by less than 15%. By contrast, KRbz
did not significantly affect the growth of the NSCLC cell line H460, which lacks the relevant
K-ras mutation. Further studies with a KRbz-adenoviral (ADV) vector construct
(pACCMVpLpA) inhibited H1725 cell growth by >90%, based on enumeration of viable
cell numbers and [3H]-thymidine uptake. KRbz-ADV-treated cells had a correspondingly
lower K-ras expression. These findings indicate that KRbz-ADV is potentially useful for
controlling the growth of lung tumor cells having the relevant K-ras mutation. Additional
measures that could further improve ribozyme efficacy are discussed, including biochemi-
cal modifications of the ribozyme backbone, and the simultaneous targeting of multiple
oncogene/tumor suppressor genes.
Ribozymes in the Gene Therapy of Cancer, edited by Kevin J. Scanlon and

Mohammed Kashani-Sabet. ©1998 R.G. Landes Company.
Introduction
The increasing incidence of lung cancer is estimated to reach 2,000,000 cases world-
wide by the year 2000. Lung cancer is currently the leading cause of cancer death for men
and women in the United States,1 with a 5 year survival rate of <15% for newly diagnosed
cases. With the improved understanding in lung cancer molecular pathogenesis, numerous
studies have explored the applicability of reversing oncogenetic mutation-related defects as
a means of controlling lung tumor cell growth.
Perhaps as many as 10-20 genetic mutations have occurred by the time lung cancer
becomes clinically evident. Controlled normal cell growth and differentiation depend on
the balanced functions of two groups of cellular genes, collectively known as proto-oncogenes
and tumor suppressor genes (reviewed in refs. 2-5). Tumorigenesis may involve the activa-
tion of dominant oncogenes (such as ras, c-erbB-2, myc, raf, jun, myb, fms, fur), recessive
mutations that lead to loss of tumor suppresser/negative regulator function (such as p53 or
retinoblastoma [rb]), or both.2-4 Most genetic lesions associated with lung cancer occur
across small cell (SCLC) and non-small cell (NSCLC, including large cell, adenocarcinoma,
and squamous cell) histologic types. Exceptions are mutations in the ras gene family, which
occur in 30% of adenocarcinomas but are not present in SCLCs, and rb gene mutations that
are found in >95% of SCLC but in only 20-30% of NSCLC cancers (Table 12.1). The contri-
bution of each genetic lesion in lung cancer growth alteration is undefined, although clini-
cal data suggest that ras, p53 and c-erbB-2 mutations are associated with adverse survival
and are likely to play a role in tumor pathogenesis.5-8
Growth Inhibition by Antisense Oligonucleotides (AS-ODN)

The current armamentarium for modulating gene expression includes triplex DNA,
antisense RNA/DNA and ribozymes (catalytic RNA).9 A majority of lung cancer gene therapy
studies make use of antisense oligonucleotides (AS-ODN) or expression vectors (such as a
viral vector construct) for delivering the antisense sequence to inactivate mutant oncogenes.
AS-ODNs are short, gene-specific nucleic acid sequences, typically 15-25 bases in length.
They bind to mRNA, or single/double stranded DNA via base pairing to form double-
stranded mRNA:DNA, mRNA:RNA or triple-stranded DNA, thereby stopping the transla-
tion or transcription of targeted genes via degradation by specific RNase.
In vitro studies show that AS-ODNs are effective for inhibiting human lung tumor cell
growth (Table 12.2). AS-ODNs specific for the translation initiation codon of c-myc mRNA
reduced NSCLC cell line growth by 32-72%, and inhibited cell adhesion and c-myc expres-
sion.10 c-myc, bcl-2, and mutant p53 specific AS-ODNs were equally effective in lowering
the corresponding oncoprotein expression, and reduced the growth of NSCLC cell lines by
up to 40%.11 Other effective targets for AS-ODN inhibition of NSCLC growth include growth
factor receptors (IGF-IR and IL-8 R1)12,13 and the cell cycle protein cyclin D1 that regulates
phosphorylation of the Rb gene product.14 Similarly, L-Myc and c-Kit oncoprotein expres-
sion in SCLC cell lines were downregulated by the relevant AS-ODNs, which also decreased
cell growth by 58% and 40%, respectively.15,16
AS-ODN viral vector constructs similarly inhibit lung tumor cell growth in vitro, and
provide a highly effective means for the delivery of AS-ODN in vivo.17-19 Retroviral (LNSX)17
or adenoviral20 constructs with a 2 kb K-ras gene antisense DNA insert specifically inhib-
ited expression of the codon 61-mutated K-Ras p21 protein in NSCLC H460a cells, and
reduced in vitro growth by up to 90%. Intratracheal instillation of this construct similarly
inhibited the growth of the cell line xenograft in 87% of inoculated animals (Table 12.2).18
These observations indicate that specific targeting of oncogenetic defects can collater-
ally suppress lung cancer cell growth. The introduction of phosphorothioate linkage im-
proves the stability of AS-ODNs.10,16,21 Nevertheless, their capacity to inhibit lung tumor
The Use of Ribozymes for Gene Therapy of Lung Cancer 153
Table 12.1. Clinically relevant oncogenetic alterations in lung cancer
Aberrations NSCLC SCLC
Proto-oncogene/ oncogene K-ras (activating point mutation) c-myc (gene amplification)

c-erbB-2 (altered transcription)
Tumor Suppressor gene p53 (point mutation+deletion) rb
rb (point mutation+deletion) 3p deletion
3p deletion p53
Growth Factor/receptors Insulin-like growth factor I (IGF-I) Bombesin/gastric-releasing
Transforming growth factor- α (TGFα) peptide-neuroendocrine
c-jun, c-fos peptides
IGF-I receptor Transferrin
c-erbB-1 receptor (EGFR)
aNSCLC: non-small cell lung cancer; SCLC: small cell lung cancer; rb: retinoblastoma
cell growth ranges from 40-90%. Contributing factors may include tumor heterogeneity
with respect to the oncogene defect; incomplete penetration by the vector delivery system; a
limited role of the targeted oncoprotein in cancer pathophysiology; stoichiometric limita-
tions on the reversible interactions between the antisense reagent and the targeted genetic
message; or dependence on endogenous RNase H, which may be required to release the
bound antisense molecule while destroying the target mRNA.22
Growth Inhibition by Ribozyme

Ribozymes (catalytic RNAs, RNA enzymes) are RNAs with intrinsic site-specific RNA
cleavage or ligation activities (reviewed in refs. 23, 24). Originally discovered in the pre-
rRNA of Tetrahymena thermophilia,25 seven types of naturally occurring RNA catalytic motifs
have now been identified, and trans-cleaving variants based on these motifs have been con-
structed.26 Trans-cleaving ribozymes are appealing agents for gene therapy, in view of their
ability to irreversibly inactivate their targets and their regenerative catalytic activity. By cloning
specific antisense sequences into a basic ribozyme motif, it is feasible to design ribozymes
that discriminate substrate RNAs with a single base mutation 27 as well as closely-related
RNAs.28
The hammerhead ribozyme motif is widely utilized in gene therapy ribozyme design.23
Hammerhead ribozymes were originally identified in the plus (+) strand of satellite RNA of
tobacco ringspot virus (sTobRV). They represent the smallest ribozyme motif (30 nucle-
otides) and are best characterized with respect to target sequence, requirement for the cata-
lytic core, and kinetics. As described in detail in other chapters of this monograph, the
secondary structure of the ribozyme-substrate complex consists of three helical stems ema-
nating from the catalytic core.9 The RNA substrate binds to the hammerhead ribozyme
through two helices/stems (helix I and III). The catalytic core is comprised of an unpaired
connector loop and helix II, which is made up of 8 internally complementary nucleotides.
Based on early mutational analysis, interaction between the hammerhead ribozyme and its
substrate requires an “NUH” triplet sequence 3' of the cleavage site, with N being any nucle-
otide and H being A,C, or U. The GUC triplet is frequently chosen as a cleavage substrate
because of its wide occurrence in natural ribozyme motifs.29 Substrates containing GUA,
GUU, UUC or CUC triplets are also efficiently cleaved, whereas GAC, GUG, AUC, CGC,
154
Table 12.2. Targets of gene therapy in human lung cancer
Cell Type Targeta Agent Efficacy

(Reference) In Vitro In Vivo
NSCLCa
H460a line K-ras codon 61 mutant AS-ADV (18,20) 47-90% ↓ growth NT
H460 line K-ras codon 61 AS-RTV (17) up to 90% ↓ growth 87% of treated
(vs. 10% untreated)
mice are tumor-free
H596 line mutant p53 AS-ODN (59) potentiates apoptosis NT
A549 line c-myc 4G-AS-ODN (10) 62-72% ↓ growth NT
A427, SKMES-1, c-myc AS-ODN (11) 40% ↓ growth NT

A549 bcl-2
mutant p53
H460, SCC5 IGF-IR AS-ADV (12) 84% ↓ clonogenicity Prolonged survival
IL-8 Receptor B AS-ODN (13) Reversible growth NT
inhibition
SCLC
SBC-1 line c-kit proto-oncogene AS-ADV (15) 40% ↓ growth NT
H209, H510, H82 L-myc AS-ODN (16) 70% ↓ growth NT
Drug-resistant MRPa AS-ODN (21) >90% ↓ expression NT
SCLC lines
aNSCLC: non-small cell lung cancer; SCLC: small cell lung cancer; IGF-IR: insulin growth factor-I receptor; MRP: multidrug resistant protein; AS: antisense;
RTV:retroviral vector; ODN: oligodeoxynucleotides; ADV: adenoviral vector; NT: not tested.
GGC, AGC or UGC triplet sequences are poorly cleaved. Sense-anti-sense complementa-
tion by flanking sequences (helix I or helix III of the ribozyme but probably not both)30
ensures sequence specificity of the cleavage reaction and juxtaposes the conserved 13 nucle-
otide-catalytic core across from the cleavage site. Hammerhead ribozymes are recognized as
metalloenzymes, since divalent cations (Mg2+) are captured by the catalytic domain and
play an integral role in the multiple turnover site-specific trans-cleavage process by non-
hydrolytic transesterification.31 Efficacy of cleavage is dependent on the length of the comple-
mentary flanking sequence (optimal length, 12-13 nt), and composition and accessibility of
the cleavage site.32,33 Based on the current understanding of the ribozyme cleavage reaction,
the criteria for efficient functioning of ribozyme within a cell are:
1. the target RNA sequence must be accessible for ribozyme hybridization;
2. the ribozyme and target RNAs must be present in the same subcellular compartment;
3. the ribozyme must be stable in order to attain a high ribozyme RNA:target RNA
ratio; and
4. the ribozyme should be able to access all the target cells within a particular tis-
sue,30,34 except in circumstances where downregulation of the target RNA has a “by-
stander” effect on neighboring cells.
Ribozymes that target the expression products of mutant or amplified genes have been
used to modulate growth and/or drug-resistance activities of human cancer cells.24 The ras-
encoded p21ras protein is a guanine nucleotide binding GTPase.35,36 Activating ras muta-
tions, invariably found in the GTP binding regions of p21ras, produce a constitutively acti-
vated GTP-locked p21ras which is believed to contribute to uncontrolled malignant growth.
The mutant ras oncogene is an attractive target for ribozyme-mediated gene modulation, in
view of its high mutation incidence, the restricted localization in the ras gene, and the cru-
cial role of ras in signal transduction during neoplastic growth. Treatment with a hammer-
head ribozyme against the H-ras codon 12 mutation (GGC→GUC) inhibited H-ras mutant
tumor cell growth and partially reverted the malignant phenotype in human bladder carci-
noma, human melanoma, and murine NIH3T3 models, while normal H-ras proto-oncogene
function was unaffected.37 In vivo treatment similarly reduced tumorigenicity of the H-ras
mutant bladder carcinoma xenograft with no identifiable toxicity.38 By comparison, mutant
ribozymes that lack cleavage activity did not exert a tumor-suppressive effect, despite hav-
ing the appropriate complementary antisense sequence.37 Thus cleavage function appears
integral to the ribozyme’s ability to alter H-ras gene expression and cell growth.
Approximately 30% of NSCLCs carry a ras mutation, which occur primarily in lung
adenocarcinomas and small numbers of large cell undifferentiated and squamous cell lung
carcinomas. NSCLC patients with ras mutation and/or p21ras overexpression have a sig-
nificantly poorer prognosis (reviewed in refs. 5, 8). In US and European studies, nearly all
mutations occur in the K-ras gene, although N-ras and H-ras mutations have been detected
occasionally.8 K-ras mutations occur predominantly in codon 12 with a G→T transversion
(Table 12.3). Roth and coworkers have demonstrated that K-ras AS-viral constructs are ef-
fective for inhibiting human NSCLC cell growth in vitro and in vivo.17,20 However, these
studies are based on the H460a cell line model, which has a homozygous K-ras codon 61
mutation.
We recently examined the growth-modulating effect of a hammerhead ribozyme that
is specific for the K-ras codon 12 mutant sequence GUU, given the findings that
1. nearly all NSCLC ras mutations occur in K-ras codon 12 in the United States
(Table 12.3; also ref. 8);
2. anti-K-ras, anti-H- as well as anti-N-ras hammerhead ribozymes are potent growth
inhibitors in various human cancers tested;24,39-42 and
Table 12.3. K-ras mutations and potential ribozyme substrate sites in non-small
cell lung cancers
Mutation* Corresponding Ribozyme % of

Codon Sequence Substrate Sequence Total Mutations
12 TGT UGU 55
12 GAT GAU 18
12 GTT GUU 12
12 GCT GCU 3
12 AGT AGU 3
12 TTT UUU 3
13 TGC UGC 6
*wild type sequence is GGT for codon 12, GGC for codon 13. Mutation was determined by PCR/
gene sequencing with 103 paraffin-embedded, formalin-fixed archived specimens of histologically
proven lung cancer.6
Reprinted with permission from Tong AW et al. In: Scanlon K, ed. Therapeutic Applications of
Ribozymes. Methods in Molecular Medicine, 1997.
3. comparative studies show that ribozymes directed at oncogenes (K- and H-ras, c-fos,
BCR-ABL) or viral proteins are more effective than their antisense counterpart;43
Double-stranded DNA dideoxy end-label sequencing of the human lung adenocarci-
noma cell line H1725 genome DNA demonstrated a heterozygous GGT/GTT mutation at
codon 12 of the K-ras gene.44 This cell line is relevant in the study of mutant ras gene modu-
lation, since the point mutation resulting in a heterozygous single base pair mismatch is
expected to be more prevalent than homozygous ras mutations.
Earlier transfection studies were performed using this ribozyme construct (KRbz) cloned
into the pHβ Apr-1-neo plasmid and transfected into H1725 cells by electroporation.44,45
Expression of the transgene was demonstrated in isolated G418-resistant clones by RT-PCR.
The growth rate of KRbz-transfected H1725 cells was reduced by 81% as compared with
less than 15% decrease in growth rate for mock-transfected cultures, based on cell number
determinations at day 8 (Fig. 12.1, Table 12.4). Parent monolayer culture cells were of ir-
regular morphology with long, spindly processes that rapidly spread and overlap one an-
other. By contrast, KRbz-transfected cells were round, had fewer spindly processes and tended
to grow in patches, suggesting a reversion of the malignant phenotype. Culture doubling
time was substantially longer for KRbz-treated H1725 cells (>45 hours) as compared with
untreated H1725 cells (28 hours; Table 12.4). By contrast, KRbz did not significantly affect
the growth of the NSCLC cell line H460 (Table 12.4), which lacks the relevant K-ras muta-
tion.44 These observations suggest that KRbz treatment is effective in inhibiting the growth
of lung adenocarcinoma cells carrying the relevant heterozygous K-ras codon 12 mutation.
For potential in vivo applications, it is necessary that the ribozyme be delivered effec-
tively to the desired tissue in a non-toxic manner. Recent studies have utilized plasmids,
retroviruses, adenoviruses, adeno-associated viruses (AAV), and cationic lipids for ribozyme
delivery. A problem of lipofection (cationic liposome) is its high toxicity in currently avail-
able formulations.46 Among the various viral vectors, the adenovirus is preferred because of
its high efficacy of gene transfer in many types of human cancers and its ability to infect
both dividing and non-dividing cells. Adenoviruses are stable and can be obtained, in many
systems, in higher titer than retrovirus (1010 vs. 108 PFU/ml, respectively), which is advanta-
geous in clinical situations that require large quantities of viral particles. Adenoviral vector
Fig. 12.1. Growth inhibition of

NSCLC H1725 Cells by the
KRbz-Plasmid vector. H1725
cells were transfected with KRbz-
pHβ Apr-1 or pHβ Apr-1 empty
cassette by electroporation (200
V, 960 µ Fd). Transfected cells
were selected and cloned in pres-
ence of G418. Growth rate of
KRbz-treated and mock-trans-
fected cultures was compared by
the number of viable cells deter-
mined at a given day. Duplicate
variations are <20%.
Table 12.4. Cell culture doubling time following transfection with pHβ vector or
pHβ KRbza
Cell Line Treatment Clones Avg. Doubling % Changec

Time (hours)b
H1725 Untreated 28.0

pHβ E3 32.4 15.7
pHβ Rbz D8 45.9 63.9
pHβ Rbz D11 50.9 81.8
H460 Untreated 20.3
pHβ D4 24.3 19.7
pHβ Rbz B7 20.6 1.2
pHβ Rbz D5 20.6 1.2
aCells were transfected with pHβ K-ras or pHβ mock vector, cloned, and selected by G418 for
2 months before evaluation.

bCell generation number (N) is calculated by the formula: N = 3.32 x (log
10 Cy - log10 Cx) where
Cy is cell number on day 8 and Cx is the cell number on day 1. Doubling time = 1/N x 24 hours;
c(N
treated culture - Nuntreated culture / Nuntreated culture) x 100.
constructs are expected to have a transient (episomal) mode of expression. Nevertheless,

the recombinant adenoviral:ribozyme construct (pACCMVpLpA/KRbz) was more effec-
tive in suppressing K-ras gene expression in pancreatic cancer cells as compared with its
retroviral (pLNCX) or plasmid (pHβ) counterpart.39 In light of its natural tropism for res-
piratory tissues, the adenovirus vector is particularly appropriate for delivery to lung can-
cers. Within each vector construct, an optimal promoter/enhancer system is crucial for stable
and efficient expression of the ribozyme. Certain ribozymes could be designed with an in-
ducible and/or tissue-specific enhancer/promoter for exclusive expression in diseased tis-
sue.40 Such a sequence is currently not available for lung cancer applications.
Based on these considerations, we have constructed a replication-deficient adenoviral

vector encoding the anti-K-ras ribozyme(KRbz-ADV) under a cytomegalovirus promoter/
enhancer element and an SV40 polyadenylation signal. Oligonucleotides encoding the se-
quence of the ribozyme were cloned into the polylinker of the shuttle pACCMVpLpA.45
pACCMVpLpA contains a constitutive CMV promoter insert for high levels of ribozyme
RNA production.38,39 The resultant plasmid and the pJM17 packaging plasmid were
cotransfected into the E1 transcomplementing cell line HEK293 using a cationic liposome
(DOTAP), where the anti-K-ras ribozyme was generated via homologous recombination
between pACCMVpLpA-KRbz and pJM17 (Fig. 12.2).38 The purified adenoviral vector
preparation is predicted to contain the anti-K-ras ribozyme expression cassette inserted in
place of the deleted adenoviral E1 sequence. The presence of contaminating replication-
competent (wild type) adenoviruses was ruled out by PCR analysis, using specific primers
to the E1A region (Fig. 12.3).
The adenoviral-mediated suppression of H1725 cell growth was evaluated in absence
of any selection pressure, following infection with pretitered concentrations of KRbz-ADV.
KRbz-ADV infected H1725 cells expressed the KRbz and had an approximately 25% de-
crease in p21ras expression, based on semi-quantitative immunohistochemical analysis
(Table 12.5). The decreased Ras expression parallels a dose-dependent growth inhibition
that ranged from 54% (50 PFU/cell) to 92% (500 PFU/cell) in KRbz-ADV-treated H1725
cells at day 11 post-infection. [3H]-thymidine uptake assays showed that KRbz-ADV
(250 PFU/cell) significantly inhibited DNA synthesis of H1725 cells (Table 12.6). Signifi-
cantly prolonged generation time was observed following KRbz-ADV. The growth inhibi-
tory effect of KRbz-ADV increased with time.45 These findings suggest that the KRbz-ADV
vector can effectively inhibit H1725 lung cancer cell growth, despite the transient expres-
sion characteristics of ADV vectors. Tumor cell growth appears to be more dependent on
oncogenic Ras protein function as compared with that of normal p21ras.47 This may ex-
plain our observation that moderate decreases in overall p21ras expression (but preferential
depletion of oncogenic ras by KRbz) can lead to >90% growth inhibition in the K-ras codon
12 heterozygous mutant H1725 cells.
Future Considerations
Our observations confirm that the mutant ras gene could serve as a specific target for
cancer gene modulation.9,24,38 Ribozymes directed at N-, H- and K-ras has been shown to
selectively discriminate mutated oncogenes from proto-oncogenes and reversed the trans-
formed phenotype of human melanoma, pancreatic and bladder carcinomas.37,38,40,48 The
moderate suppression of p21ras expression in KRbz-treated cultures may be explained by
the heterozygous genotype of H1725 cells, where normal p21ras may continue to be ex-
pressed and unaffected by KRbz treatment. Nevertheless, modulation of K-ras codon 12
mutant gene expression by plasmid or ADV was effective in suppressing >90% of lung can-
cer cell growth. Thus the same treatment may be potentially applicable for the treatment of
lung cancers having the same single base pair substitution in K-ras codon 12. We plan fur-
ther studies to verify the enhanced growth-inhibitory efficacy of this KRbz-Adv vector in
comparison with K-ras AS-ODNs of the same specificity.
Based on kinetic studies, we have optimized the anti-K-ras ribozyme with a 12 base
optimal length flanking sequence to maximize the turnover rate of the ribozyme and to
enhance its efficacy. Recent studies suggest that modifications of the hammerhead ribozyme
basic structure may further enhance its efficacy. 2'-fluoro-2'-deoxyuridine/cytidine-substi-
tuted N-ras ribozymes have prolonged stability in vitro and comparable catalytic activity as
their unmodified counterpart.41 The introduction of terminal phosphorothioate groups
may further improve stability without loss of efficacy.41 Alternatively, oligonucleotide facili-
Fig. 12.2. Generation of KRbz-ADV. The K-ras ribozyme hammerhead sequence

targets the GUU mutant mRNA sequence of K-ras codon 12. The ribozyme se-
quence was subcloned into the HindIII and SalI site of pACCMVpLpARS (+) to
form the shuttle plasmid KRbz-pACCMVpLpA. This ribozyme expression shuttle
vector and the adenoviral packaging plasmid pJM17 were cotransfected into
HEK293 cells to generate the KRbz-ADV by homologous recombination. Stan-
dard protocols were used for purification of viral vectors from transfected cul-
tures, verification of KRbz insert in viral DNA and determination of viral stock
titer.60
tators that are inserted at 3' and 5' ends of the flanking sequence can potentially pre-form
the substrate for ribozyme attack,49 thereby promoting ribozyme-substrate association as
well as the cleavage rate. These facilitators produce the most pronounced improvements for
long substrates (>900 bp) cleavage, although short substrate (<40 bp) cleavage was also
improved by 4-fold.50 Minizymes, hammerhead ribozymes with short oligonucleotide link-
ers instead of stem-loop II, have been produced with high specific activity. Minizymes form
dimeric structures with a common stem II which endow homodimers with a biphasic cleav-
age activity, and allow heterodimers to simultaneously cleave a substrate at two different
sites.51,52 These mechanisms account for the higher catalytic activity of minizymes as
Fig. 12.3. PCR analysis of the A B

KRbz-ADV vector. Purity of
KRbz-ADV preparations
was determined by PCR,
using KRbz sequence-spe-
cific primers (A) and prim-
ers for detecting the pres-
ence of wt adenoviral DNA
(E1 region) sequence (B).
(A): Lane 1, DNA Marker;
Lane 2, pACCMVpLpA
KRbz plasmid (positive con-
trol); Lane 3, KRbz-ADV vector; Lane 4, water control. Primers used are: 5'-primer:
5'-GCGTGTACGGTGGGAGGTCT-3' (CMV promoter); 3'-primer: 5'-GTTTCGTCC-
TCACGGACTCAT-3' (KRbz insert); for 32 cycles of amplification. (B): Lane 1, DNA Marker;
Lane 2, wt adenovirus type 2 DNA (positive control); Lane 3, KRbz-ADV vector; Lane 4,
water control. 5'-primer: 5'-ATTACCGAAGAA-ATGGCCGC (1.64 mu); 3'-primer:
5'-CCCATTTAACACGCCATGCA (4.60 mu); for 29 cycles of amplification. Sensitivity of wt ad-
enoviral DNA (E1A) detection is ≥ 1 pg.
Fig. 12.4. Dose-dependent

growth inhibition by KRbz-
ADV. H1725 cells (4 x 103/well)
were infected with graded
doses of KRbz-ADV (25-
500 PFU/cell; 90 min). The
average (n = 2) cell numbers in
treated and control cultures
were determined through day
11 post-infection. Control cul-
tures were treated with me-
dium alone. Duplicate varia-
tions are <20%.
compared with the parent hammerhead ribozyme. Such structural modifications presum-
ably may be combined with approaches that upregulate the activity of ribozyme-activating
proteins in vivo.53,54
Finally, KRbz growth inhibition may be enhanced by concomitant targeting of other
oncogene/tumor suppressor gene candidates. Subsequent to establishing the patient’s tu-
mor oncogenetic phenotype (Table 12.1), administration of multiple relevant ribozyme/
minizyme agents of proven efficacy could conceivably produce an additive/synergistic growth
inhibitory effect. The reconstitution of wild type (wt) p53 function induced either growth
suppression or apoptosis in p53-mutant NSCLC cells.55 This p53-targeted gene modulation
approach appears more effective than ribozymes directed at endogenous mutant p53 pre-
messenger RNAs.56 Serial direct injection of a wt p53-retroviral similarly produced tumor
regression in previously chemorefractory NSCLC patients.57,58 Thus the suppression of
mutant K-ras expression by the K-ras ribozyme, together with wt p53 gene replacement,
represents a promising combination gene therapy approach for NSCLC.
Table 12.6. Decreased [3H]-Thymidine uptake in

KRbz-ADV-treated H1725 cell culturesa
Experiment no. % Inhibition ± SDb
1 60 ± 12
2 78 ± 12
3 56 ± 4
Mean 65 ± 9
aH1725 cells (1 x 104/well) were infected by KRbz-ADV (250 PFU/
cell) and cultured for 48 hours. [3H]-thymidine (0.15 µCi/well)
was added for additional 16 hours. Value represents mean of 3
to 6 replicas.
bcompared with culture with medium only; SD: standard deviation.
Table 12.5. Decrease in p21ras expression in KRbz-ADV-treated culturesa
% Total (Mean ± SD)b

Strongly Reactive Moderate/ Non-Reactive
Weakly Reactive
Untreated 74.2 ± 5.0 20.2 ± 4.8 5.6 ± 2.0

KRbz-treated 44.4 ± 9.2 ** 31.2 ± 5.9 * 24.6 ± 7.5 **
adetermined with cytospin preparations at 48 hours after transfection, using the pan-Ras
monoclonal antibody Ab-3 (Oncogene Science) and the immunoperoxidase technique (Ventana
Automated Systems).6
bdiffers statistically from untreated sample (student’s t test; * p <0.05, ** p <0.001).
Acknowledgments
This work was supported in part by the Robert Schanbaum Memorial Fund, the Ed-
ward and Ruth Wilkof Foundation and the Tri Delta Cancer Research Fund.
References
1. Ginsberg R, Vokes E, Raben A. Non-small cell lung cancer. In: DeVita VJ, Hellman S,
Rosenberg S, eds. Cancer: Principles and Practice of Oncology. 5th ed. Philadelphia:
Lippincott-Raven, 1997:858-911.
2. Bishop J. Molecular themes in oncogenes. Cell 1991; 64:235-248.
3. Hunter T. Cooperation between oncogenes. Cell 1991; 64:249-270.
4. Levin W, Press M, Gaynor R et al. Expression patterns of immediate early transcription
factors in human non-small cell lung cancer. Oncogene 1995; 11(7):1261-1269.
5. Strauss GM, Kwiatkowski DJ, Harpole DH, Lynch TJ, Skarin AT, Sugarbaker DJ. Molecu-
lar and pathologic markers in stage I non-small-cell carcinoma of the lung. J Clin Oncol
1995; 13:1265-1279.
6. Nemunaitis J, Klemow S, Tong A et al. Prognostic value of K-ras mutations, ras oncoprotein
and c-erb B-2 oncoprotein expression in adenocarcinoma of the lung. Am J Clin Oncol
1997; 21:155-160.
7. Minna J. The molecular biology of lung cancer pathogenesis. Chest 1993; 103:445S-456S.
8. Rodenhuis S. Ras oncogenes and human lung cancer. In: Pass H, Mitchell J, Johnson D,
Turrisi A, eds. Lung Cancer: Principles and Practice. Philadelphia: Lippincott-Raven,
1996:73-82.
9. Scanlon K, Ohta Y, Ishida H et al. Oligoncleotide-mediated modulation of mammalian
gene expression. FASEB J 1995; 9(13):1288-1296.
10. Saijo Y, Uchiyama B, Abe T, Satoh K, Nukiwa T. Contiguous four-guanosine sequence in
c-myc antisense phosphorothioate olognucleotides inhibits cell growth on human lung can-
cer cells: possible involvement of cell adhesion inhibition. Jpn J Cancer Res 1997;
88(1):26-33.
11. Robinson L, Smith L, Fontaine M, Kay H, Mountjoy C, Pirrucello S. c-myc antisense
oligodeoxyribonucleotides inhibit proliferation of non-small cell lung cancer. Ann Thorac
Surg 1995; 60(6):1583-1591.
12. Lee C, Wu S, Gabrilovich D et al. Antitumor effects of an adenovirus expressing antisense
insulin-like growth factor I receptor on human lung cancer cell lines. Cancer Res 1996;
56(13):3038-3041.
13. Olbina G, Cieslak D, Ruzdijic S et al. Reversible inhibition of IL-8 receptor B mRNA ex-
pression and proliferation in non-small cell lung cancer by antisense oligonucleotides. Anti-
Cancer Res 1996; 16(6B):3525-3530.
14. Schrump D, Chen A, Consoli U. Inhibition of lung cancer proliferation by antisense cyclin
D. Cancer Gene Ther 1996; 3(2):131-135.
15. Yamanishi Y, Maeda H, Hiyama K, Ishioka S, Yamakido M. Specific growth inhibition of
small cell lung cancer cells by adenovirus vector expressing antisense c-kit transcripts. Jpn
J Cancer Res 1996; 87(5):534-542.
16. Dosaka-Akita H, Akie K, Hiroumi H, Kinoshita I, Kawakami Y, Murakami A. Inhibition
of proliferation by L-myc antisense DNA for the translational initiation site in human small
cell lung cancer. Cancer Res 1995; 55(7):1559-1564.
17. Zhang W, Roth J. Anti-oncogene and tumor suppressor gene therapy-examples from a
lung cancer animal model. In Vivo 1994; 8(5):755-769.
18. Georges R, Mukhopadhyay T, Zhang Y, Yen N, Roth J. Prevention of orthotopic human
lung cancer growth by intratracheal instillation of a retroviral antisense K-ras construct.
Cancer Res 1993; 53(8):1743-1746.
19. Mukhopadhyay T, Roth J. A codon 248 p53 mutation retains tumor suppressor function
as shown by enhancement of tumor growth by antisense p53. Cancer Res 1993;
53(18):4362-4366.
20. Alemany R, Ruan S, Kataoka M et al. Growth inhibitory effect of anti-K-ras adenovirus on
lung cancer cells. Cancer Gene Ther 1996; 3(5):296-301.
21. Stewart A, Canitrot Y, Baracchini E, Dean N, Deeley R, Cole S. Reduction of expression of
the multidrug resistance protein (MRP) in human tumor cells by antisense phosphorothioate
oligonucleotides. Biochem Pharmacol 1996; 51(4):461-469.
22. Agrawal S, Jiang Z, Zhao Q et al. Mixed-backbone oligonucleotides as second generation
antisense oligonucleotidess: in vitro and in vivo studies. Proc Natl Acad Sci USA 1997;
94(6):2620-2625.
23. Cech TR, Uhlenbeck OC. Hammerhead nailed down. Nature 1994; 372:39-40.
24. Kijima H, Ishida H, Ohkawa T, Kashani-Sabet M, KJ S. Theraputic applications of
ribozymes. Pharmacol Therap 1995; 68:247-267.
25. Cech T. Self-splicing of group I introns. Ann Rev Biochem 1990; 59:543-568.
26. Uhlenbeck O. A small catalytic oligonribonucleotide. Nature 1987; 328:596-600.
27. Koizumi M, Kamiya H, Ohtsuka E. Ribozyme designed to inhibit transformation of NIH3T3
cells by the activated c-Ha-ras gene. Gene 1992; 117:179-184.
28. Bennett M, Cullimore J. Selective cleavage of closely related mRNAs by synthetic ribozymes.
Nuc Acids Res 1992; 20:831-837.
30. Kronenwett R, Haas R, Sczakiel G. Kinetic selectivity of complementary nucleic acids: bcr-
abl-directed antisense RNA and ribozymes. J Mol Biol 1996; 259(4):632-644.
31. Ohkawa J, Koguma T, Kohda T, Taira K. Ribozymes: From mechanistic studies to applica-
tions in vivo. J Biochem 1995; 118:251-258.
32. Ruffner DE, Stormo GD, Uhlenbeck OC. Sequence requirements of the hammerhead RNA
33. Bertrand E, Castanotto D, Zhou C et al. The expression cassette determines the functional
activity of ribozymes in mammalian cells by controlling their intracellular localization. RNA
1997; 3(1):75-88.
34. Lieber A, Kay M. Adenovirus-mediated expression of ribozymes in mice. J Virol 1996;
70(5):3153-3158.
35. McCormick F. Signal transduction: How receptors turn ras on. Nature 1993; 363:15-16.
36. Sun J, Qian Y, Hamilton A, Sebti S. Ras CAAX peptidomimetic FTI276 selectively blocks
tumor growth in nude mice of a human lung carcinoma with K-ras mutation and p53
deletion. Cancer Res 1995; 55:4243-4247.
48:1471-1475.
38. Feng M, Cabrera G, Desane J, Scanlon KJ, Curiel DT. Neoplastic reversion accomplished
55: 2024-2028.
39. Kijima H, Bouffard D, Scanlon K. Ribozyme-mediated reversal of human pancreatic carci-
noma phenotype. In: Ikehara S, Takaku F, Good R, eds. Proceedings of International Sym-
posium on Bone Marrow Transplantation. Tokyo: Springer-Verlag, 1996:153-163.
40. Ohta Y, Kijima H, Kashani-Sabet M, Scanlon K. Suppression of the malignant phenotype
of melanoma cells by anti-oncogene ribozymes. J Invest Dermatol 1996; 106(2):275-280.
41. Scherr M, Grez M, Ganser A, Engels J. Specific hammerhead ribozyme-mediated cleavage
of mutant N-ras mRNA in vitro and ex vivo. Oligoribonucleotides as therapeutic agents. J
Biol Chem 1997; 272(22):14304-14313.
42. Eastham J, Ahlering T. Use of an anti-ras ribozyme to alter the malignant phenotype of a
human bladder cancer cell line. J Urol 1996; 156(3):1186-1188.
43. Kashani-Sabet, M., Scanlon KJ. Application of ribozymes to cancer gene therapy. Cancer
Gene Therapy 1995; 2:213-223.
44. Zhang Y, Nemunaitis J, Mues G, Scanlon K, Tong A. In vitro suppression of human non-
small cell lung cancer (NSCLC) cell growth by an anti-K-ras ribozyme. FASEB J 1996;
10(6):A1409.
45. Zhang Y, Mues G, Nemunaitis J, Scanlon K, Tong A. Inhibition of human non-small cell
lung cancer growth in vitro by a recombinant anti-K-ras adenoviral vector. AACR Pro-
ceedings 1997; 38:6.
46. Snyder DS, Wu Y, Wang JL et al. Ribozyme-mediated inhibition of bcr-abl gene expres-
sion in a Philadelphia chromosome-positive cell line. Blood 1993; 82:600-605.
47. James G, Goldstein J, Brown M et al. Benzodizepine petidomimetics: Potent inhibitors of
ras farnesylation in animal cells. Science 1993; 260:1937-1942.
48. Kashani-Sabet M, Funato T, Tone T et al. Reversal of the malignant phenotype by anti-ras
ribozyme. Antisense Res & Develop 1992; 2:3-15.
49. Dedman R. Facilitator oligonucleotides increase ribozyme RNA binding to full-length RNA
substrates in vitro. FEBS Letters 1996; 382:116-120.
50. Jankowsky E, Schwenzer B. Efficient improvement of hammerhead ribozyme mediated cleav-
age of long substrates by oligonucleotide facilitators. Biochemistry 1996; 35(48):15313-15321.
51. Amontov S, Nishikawa S, Taira K. Dependence on Mg++ ions of the activities of dimeric
hammerhead minizymes. FEBS Lett 1996; 386(2-3):99-102.
52. Kuwabara T, Amontov S, Warashina M, Ohkawa J, Taira K. Characterization of several
kines of dimer minizyme: Simultaneous cleavage at two sites in HIV-1 tat mRNA by dimer
minizymes. Nucleic Acids Res 1996; 24(12):2302-2310.
53. Kiehntofp M, Esquivel E, Brach M, Herrman F. Ribozymes: Biology, biochemistry, and
implications for clinical medicine. J Mol Med 1995; 73:65-71.
54. Herschlag D, Khosla M, Tsuchihashi Z, Karpel R. An RNA chaperone activity of non-

specific RNA binding proteins in hammerhead ribozyme catalysis. EMBO J 1994;
13:2913-2924.
55. Fujiwara T, Kagawa S, Ogawa N et al. Recombinant virus-mediated transfer of the wild-
type p53 gene is a potent therapeutic strategy for human cancer. Hum Cell 1996; 9(1):25-30.
56. Cai D, Mukhopadhyay T, Roth J. Suppression of lung cancer cell growth by ribozyme-
mediated modification of p53 pre-mRNA. Cancer Gene Ther 1995; 2(3):199-205.
57. Roth J, Nguyen D, Lawrence D et al. Retrovirus-mediated wild-type p53 gene transfer to
tumors of patients with lung cancer. Nature Med 1996; 2(9):985-991.
58. Swisher S, Roth J, Lawrence D et al. Persistent transgene expression following repeated
injections of a recombinant adenovirus containing the p53 wildtype gene in patients with
non-small cell lung cancer. AACR Proceedings 1997; 38:342.
59. Murayama Y, Horiuchi S. Antisense oligonucleotides to p53 tumor suppressor suppress
the induction of apoptosis by epidermal growth factor in NCI-H596 human lung cancer
cells. Antisense Nucleic Acid Drug Dev 1997; 7(2):109-114.
60. Tong A, Zhang Y, Nemunaitis J, Mues G. K-ras ribozyme for lung cancer. In: Scanlon K,
ed. Theraputic Application of Ribozymes. Methods in Molecular Medicine. vol. II; Totowa
(NJ): Humana 1998:209-222.
CHAPTER 13
Ribozymes in Gene Therapy

of Prostate Cancer
Dale J. Voeks, Gary A. Clawson, and James S. Norris
Introduction
S ince the discovery of naturally occurring catalytic RNA,1,2 research in ribozyme design
and its implementation in therapy has expanded exponentially. Ribozyme-mediated in-
hibition of gene expression at the message level allows the mRNA of nearly any gene to be
targeted for destruction. Possible applications are limited only by knowledge of the disease.
The greatest part of ribozyme utility in therapy thus far has centered on HIV and cancer.
Use in HIV protocols has been widespread due to the wealth of targets offered by a well-
characterized retroviral genome. In cancer, oncogenes are often overexpressed or mutated
in the signal transduction pathway, leading to uncontrolled cell growth. Because the process
works through an RNA intermediate, targets are readily available for ribozyme activity. The
major focus of ribozymes in gene therapy of cancer has been on inhibition of specific
oncogene expression in tumors having a relatively defined genetic basis. In these instances,
ribozymes have demonstrated great therapeutic potential. Unfortunately, ribozyme proto-
cols in prostate cancer have been restricted due to an uncertainty of targets.
Prostate Cancer Background

Prostate cancer is the most frequently diagnosed cancer in US males and the second
leading cause of cancer death.3 Although already a major health problem with predicted
dramatic increases in incidence and mortality over the next 15 years, the molecular mecha-
nisms behind the disease’s initiation and progression remain largely a mystery. The overall
value of screening and treatment also remains controversial. PSA is currently the most sen-
sitive marker for screening and monitoring prostate cancer. However, questions of diagnos-
tic accuracy hinder its predictive ability.4 Therapeutic modalities are limited, mostly inef-
fective, and defined in large part by stage. Locally confined disease can be controlled with
radical prostatectomy and radiation therapy. Treatment for disease with local advancement
or distant metastasis includes watchful waiting, radiation therapy, hormonal therapy, and
limited chemotherapy. Recurrence rates and mortality with non-localized disease are very
high.4 Further, approximately one-third of patients are in advanced stages of the disease at
the time of diagnosis.5 Additional therapeutic options are clearly necessary.
Gene Therapy in Prostate Cancer

Several prostate cancer gene therapy protocols have been approved for clinical trial.
Immunotherapy trials involve either ex vivo gene delivery of interleukin-2 (IL-2) or
granulocyte-macrophage stimulating factor (GM-CSF), or in vivo gene delivery of pros-

tate-specific antigen (PSA).6 GM-CSF and IL-2 cytokine-secreting tumor cells proved to
have preclinical efficacy as therapeutic vaccines in the Dunning rat model of prostate can-
cer.7,8 A tumor-specific host immune response can also be generated by in vivo transfer of
the prostate tumor antigen gene.9 In another clinical trial, transfer of herpes simplex virus
thymidine kinase (HSV-tk) creates susceptibility to cytotoxic ganciclovir treatment.10 The
final approach targets c-myc by antisense technology for therapy of late stage androgen-
resistant cancer.11
A large body of prostate cancer therapeutic research covering a wide range of strategies
is currently in preclinical stages. Due to advances in vector biology, cytokine delivery in vivo
has emerged as a viable alternative to ex vivo approaches.7 In addition, a number of genes
such as p53, Rb, E-cadherin and androgen receptor have been found mutated or with al-
tered levels of expression in late stage prostate cancer.12-15 Gene therapy strategies directed
at these late stage changes have been developed. Delivery of cell adhesion molecules (CAMs)
or an active form of the androgen receptor into advanced stage prostate cancer cells display-
ing decreased E-cadherin expression and androgen independence induced a strong anti-
proliferative response.16,17 Introduction of the cell cycle and apoptosis regulatory genes p21
and p53 resulted in reduced tumorgenicity.18,19 Similarly, transfection of Rb inhibited pros-
tate tumor cell growth.20 As the understanding of prostate cancer grows, so too will the
number of potential gene therapy approaches.
Ribozymes in Prostate Cancer

With the explosion of ribozyme technology, their involvement in prostate cancer gene
therapy remains conspicuously absent. A sequential series of activations and inactivations
of oncogenes and tumor suppressor genes has not been elucidated in prostate cancer as they
have, for example, in colon cancer.21 Because specific genetic alterations are relatively un-
known or represent only a limited number of the tumors, ribozyme strategies cannot rely
solely on inhibition of oncogene expression. The target needs to be more universal, encom-
passing all prostate cancers rather than just a subset containing a specific oncogene or tu-
mor suppressor gene dysfunction.
Ribozyme Target
RNA polymerase I (RNA pol I) is responsible for the transcription of ribosomal genes
(rDNA). rDNA constitutes only 1% of the genome but accounts for 40% of total cellular
transcription and 80% of the RNA content of the cell.22 The ribosomal RNA (rRNA) is
cleaved to mature rRNA, which interacts with ribosomal proteins to form the ribosome
complex necessary for translation and protein synthesis. This immense undertaking is com-
pletely dependent on activated RNA pol I and its associated factors. A ribozyme targeted to
this essential player in growth and viability would exert a profound impact on the cell.
Indeed, Rb may repress cellular proliferation by disrupting RNA pol I-mediated transcrip-
tion through interaction with its transcription factor UBF, which hinders formation of the
initiation complex.23,24
A ribozyme was targeted to bind and cleave at a susceptible site in subunit AC40 of
RNA pol I (Fig. 13.1). The AC40 subunit is highly conserved among distant eukaryotes,
suggesting its importance as a core component of RNA pol I.25 AC40 was also found to be a
shared subunit with RNA polymerase III.26 Therefore, a ribozyme targeted to this subunit
may also interfere with tRNA production to further disrupt protein synthesis.
The RNA pol I targeted ribozyme was cloned between 5' and 3' autocatalytic ribozyme
sequences to form a triple ribozyme (TR) (Fig. 13.2). The two cis-acting ribozymes un-
dergo self-cleavage, releasing the internal RNA pol I targeted ribozyme as a short uniform
Ribozymes in Gene Therapy of Prostate Cancer
Fig. 13.1. Ribozyme target site (RNA polymerase I). To identify potential ribozyme cleavage sites, the sequence of mouse RNA pol I was searched for the
presence of the nucleotide triplet GUC. Secondary RNA structure of the RNA pol I transcript was then predicted using the program MFOLD. Suscep-
tible regions of the transcript sequence containing GUC were identified. The region at nucleotide 457-459 indicated by the arrow was chosen as the
167
target cleavage site.

Fig. 13.2. RNA pol I triple ribozyme. The 5' and 3' cis-acting hammerhead ribozymes were
designed to bind and undergo self-cleavage at GUC sequences by homology of 13 bases,
thereby releasing the internal RNA pol I (subunit AC40) targeted ribozyme with defined 5'
and 3' ends.
RNA to avoid compromised activity due to non-specific flanking sequences.27 Transfection

of TR into mouse fibroblast cells resulted in a 50% decrease in RNA pol I message levels
(50% transfection efficiency).27 The reduction of RNA pol I mRNA caused by the activity of
the triple ribozyme was sufficient to markedly diminish cell population compared to a mutant
non-catalytic version of the ribozyme acting in antisense fashion.27 Hence, the ribozyme is
able to decrease ribosomal gene transcription through knockout of the polymerase to a
level significant enough to realize cell killing activity. However, application of the RNA pol I
triple ribozyme to prostate cancer gene therapy requires highly restricted targeting of ex-
pression to the tissue of interest, to prevent collateral damage in other tissues.
Tissue Specificity
Tumors often sustain the ability to produce proteins specific for the tissue from which
the neoplasm arose.28 To exploit this transcriptional specificity, essential promoter regions
can be defined and used to restrict expression of an exogenous gene of interest. Although
transcriptional targeting is generally considered to be tissue-specific, a number of promot-
ers can also be thought of as tumor-specific if the normal tissues are not essential for viabil-
ity or accessible to the introduced DNA.29 Prostate tissue can be technically considered non-
essential, thereby defining prostate-specific promoters as both tissue-specific and
tumor-specific.29 The coupling of cytotoxic or apoptosis-inducing genes to prostate-spe-
cific promoters is a promising approach to prostate cancer gene therapy.
Ribozymes in Gene Therapy of Prostate Cancer 169
The expression of a number of proteins is strongly restricted to prostate tissue. PSA is

the most notable example. PSA production is nearly exclusive to both normal and, in most
cases, neoplastic luminal prostatic epithelial cells.30 Its androgen regulated promoter region
has been isolated and characterized.31 A minimal PSA promoter segment was able to drive
expression of a reporter gene in LNCaP cells (PSA-producing human prostate cancer cell
line) without showing activity in lines of non-prostatic origin.32 A larger 6 kb PSA pro-
moter fragment containing an additional upstream enhancer region demonstrated pros-
tate-specific expression of a reporter gene in transgenic mice mirroring the expression pat-
terns seen in humans.33 Research to further refine the promoter regulatory elements necessary
for tissue specificity and maximum expression continues.34,35 Nonetheless, the PSA pro-
moter has already shown significant potential in prostate cancer gene therapy. The adenovi-
ral gene E1A, DNA polymerase-α and topoisomerase IIα under transcriptional control of
the PSA promoter displayed both specificity of expression and cytotoxicity in PSA-express-
ing cancer cells.32,36
Similarly, androgen regulated rat probasin production occurs predominantly in the
dorsal and lateral regions of the prostate.37 A minimal probasin promoter region driving
expression of a reporter construct in transgenic mice displayed highly specific expression
restricted to the lateral, dorsal and ventral lobes of the prostate, with very limited expression
also observed in the anterior prostate and seminal vesicles but nowhere else in the body.38
The highest level occurred in the mouse dorsolateral prostate, which correlates to the pe-
ripheral zone of the human prostate where the majority and most invasive forms of human
cancer arise.39 A larger promoter fragment containing both androgen and zinc regulatory
regions was able to direct higher and more consistent levels of expression in subsequent
research.40 Expression patterns of the probasin promoter have yet to be determined in hu-
man tissue but, at minimum, the promoter sequence has utility in the transcriptional tar-
geting of prostate cancer gene therapy constructs for evaluation of effect in animal model
systems.
By coupling to the minimal probasin promoter, the cell killing effects of the RNA pol I
targeted triple riboyzme can be highly restricted to prostatic epithelium and its neoplastic
derivatives. The activity of the probasin-driven triple ribozyme (PBTR) construct was evalu-
ated in a recently developed mouse prostate tumor cell line (TRAMP) (kindly provided by
Norman M. Greenberg). Transient transfection of a probasin-driven Seap reporter con-
struct showed that the probasin promoter was able to drive expression of a heterologous
gene in this cell line. Initiation of maximal expression occurred between day one and two
with slightly increasing intensity thereafter.41 Transfection of PBTR into the prostate tumor
cells reduced cellular proliferation beginning between day one and two as compared to con-
trol transfected cells (Fig. 13.3).42 The span of decreased cell growth corresponded to a di-
minished level of RNA pol I (subunit AC40) as judged by Western blot analysis (unpub-
lished). As these were transient transfections at low efficiency, a marked deviation in cell
number occurred for several days only before returning to normalcy (presumably when the
transduced cells were removed from the population following expression of the triple
ribozyme).
For evaluation of therapeutic potential on tumor tissue in vivo (TRAMP xenografts),
the PBTR construct was administered via intra-tumoral injection utilizing a cationic lipo-
somal vehicle. Both single administration and repeated administration on three consecu-
tive days demonstrated a consistent reduction in tumor growth between approximately day
two and day six post-administration followed by a return to the normal tumor growth curve
(Fig. 13.4)42 compared to control injections (data not shown). Reporter construct delivery
under the same conditions indicated extremely low liposomal gene transfer (unpublished).
Still, the overall trend of reduced tumor burden displayed in a transient manner following
Fig. 13.3. Transfection assay (probasin-RNA pol I triple

ribozyme). TRAMP cells were transfected with the probasin-
driven RNA pol I targeted triple ribozyme (PBTR), the identi-
cal construct without the internal ribozyme (PBDR) or mock
transfected with liposome only (Lipo Cntrl). Cell number was
determined daily. Reprinted with permission, Cancer Gene
Therapy. Savage LM, ed. Southborough, MA: International Busi-
ness Communications, Inc. 1997.
injection of the probasin-driven triple ribozyme offers the promise of increased effective-
ness through use of a more efficient delivery vehicle.
Conclusions
Ribozymes have displayed significant therapeutic potential in a number of malignan-
cies. Due to the relatively poor understanding of the molecular events responsible for pros-
tate cancer initiation and progression, target sites for ribozyme design are not overly clear.
To overcome this and the variable genetics associated with each stage of prostate cancer,
ribozymes can be designed to have universal cell killing effects if properly targeted to the
tissue of interest. Prostate epithelial cell-specific knockout of a key gene in cellular growth
and viability is capable of localized cell killing, with the potential for tracking metastasized
tumor cells. The initial reduction in tumor burden caused by liposomal delivery of the
probasin-driven RNA pol I triple ribozyme indicates the possible translation of a prostate-
specific RNA pol I ribozyme therapy strategy to human cancer if increased levels of cell
killing can be achieved with improved delivery. However, even with stronger research em-
phasis, delivery remains the major rate-limiting step in most gene therapy protocols. Be-
sides improved efficiency, targeted delivery coupled with the expanding list of tissue-spe-
cific regulatory elements will allow both direct delivery to the tumor site and systemic delivery
to treat metastatic sites. Also, increased molecular knowledge of malignancies will identify
additional ribozyme targets. The potential of ribozymes in gene therapy, in general, and
prostate cancer therapy, in particular, will eventually be realized if advances in ribozyme
technology, delivery vectors, and the genetics of disease continue at the current rate.
Fig. 13.4. Tumor grafting studies (probasin-RNA pol1 triple ribozyme). Tumor-bearing mice
were generated by subcutaneous injection of TRAMP cells. Direct intra-tumoral injections of
PBTR-liposome complexes (PBTR-Lipo) or liposome only (Lipo Cntrl) were performed as a
single administration (1x) or a repeated administration on three consecutive days (3x). Tumor
volume was calculated from standard caliper measurement. Lipo Cntrl showed no deviation
from normal tumor growth curves. A consistent reduction in tumor growth was observed be-
tween ≈ day two and day six postadministration of PBTR.
References
1. Cech TR, Bass BL. Biological catalysis by RNA. Annu Rev Biochem 1986; 55:599-629.
2. Altman S. Ribonuclease P: an enzyme with a catalytic RNA subunit. Adv Enzymol Relat
Areas Mol Biol 1989; 62:1-36.
3. Parker SL, Tong T, Bolden S et al. Cancer statistics. CA Cancer J Clin 1997; 47(1):5-27.
4. Cersosimo RJ, Carr D. Prostate cancer: Current and evolving strategies. Am J Health Syst
Pharm 1996; 53(4):381-396.
5. Konety BR, Getzenberg RH. Novel therapies for advanced prostate cancer. Semin Urol
Oncol 1997; 15(1):33-42.
6. Sanda MG. Biological principles and clinical development of prostate cancer gene therapy.
Sem Urol Oncol 1997; 15(1):43-55.
7. Sanda MG, Ayyagari SR, Jaffee EM et al. Demonstration of a rational strategy for human
prostate cancer gene therapy. J Urol 1994; 151(3):622-628.
8. Viewig J, Rosenthal FM, Bannerji R et al. Immunotherapy of prostate cancer in the Dun-
ning rat model: Use of cytokine gene modified tumor vaccines. Cancer Res 1994;
54(7):1760-1765.
9. Xue BH, Zhang Y, Sosman JA et al. Induction of human cytotoxic T lymphocytes specific
for prostate-specific antigen. Prostate 1997; 30(2):73-78.
10. Hall SJ, Mutchnik SE, Chen SH et al. Adenovirus-mediated herpes simplex virus thymi-
dine kinase gene and ganciclovir therapy leads to systemic activity against spontaneous
and induced metastasis in an orthotopic mouse model of prostate cancer. Int J Cancer
1997; 70(2):183-187.
11. Hrouda D, Dalgeish AG. Gene therapy for prostate cancer. Gene Ther 1996; 3(10):845-852.
12. Dahiya R, Deng G, Chen KM et al. p53 tumour-suppressor gene mutations are mainly
localized on exon 7 in human primary and metastatic prostate cancer. Br J Cancer 1996;
74(2):264-268.
13. Kubota Y, Fujinami K, Uemura H et al. Retinoblastoma gene mutations in primary hu-
man prostate cancer. Prostate 1995; 27(6):314-320.
14. Umbas R, Schalken JA, Aalders TW et al. Expression of the cellular adhesion molecule E-
cadherin is reduced or absent in high-grade prostate cancer. Cancer Res 1992; 52(18):
5104-5109.
15. Tilley WD, Wilson CM, Morcelli M et al. Androgen receptor gene expression in human
prostate carcinoma cell lines. Cancer Res 1990; 50(17):5382-5386.
16. Kleinerman DI, Zhang WW, Lin SH et al. Application of a tumor suppressor (C-CAM1)-
expressing recombinant adenovirus in androgen-independent human prostate cancer
therapy: a preclinical study. Cancer Res 1995; 55(13):2831-2836.
17. Yuan S, Trachtenberg J, Mills GB et al. Androgen-induced inhibition of cell proliferation
in an androgen-insensitive prostate cancer cell line (PC-3) transfected with a human an-
drogen receptor complementary DNA. Cancer Res 1993; 53(6):1304-1311.
18. Yang C, Cirielli C, Capogrossi MC et al. Adenovirus-mediated wild-type p53 expression
induces apoptosis and suppresses tumorigenesis of prostatic tumor cells. Cancer Res 1995;
55(19):4210-4213.
19. Eastham JA, Hall SJ, Sehgal I et al. In vivo gene therapy with p53 or p21 adenovirus for
prostate cancer. Cancer Res 1995; 55(22):5151-5155.
20. Bookstein R, Shew JY, Chen PH et al. Suppression of tumorigenicity of human prostate
carcinoma cells by replacing a mutated RB gene. Science 1990; 247(4943):712-715.
21. Cho KR, Vogelstein B. Genetic alterations in the adenoma-carcinoma sequence. Cancer
1992; 70(6 suppl): 1727-1731.
22. Moss T, Stefanovsky VY. Promotion and regulation of ribosomal transcription in eukary-
otes by RNA polymerase I. Prog Nucleic Acid Res Mol Biol 1995; 50:25-66.
23. Cavanaugh AH, Hempel WM, Taylor LJ et al. Activity of RNA polymerase I transcription
factor UBF blocked by Rb gene product. Nature 1995; 374(6518):177-180.
24. Voit R, Schafer K, Grummit I. Mechanism of repression of RNA polymerase I transcrip-
tion by retinoblastoma protein. Mol Cell Biol 1997; 17(8):4230-4237.
25. Song CZ, Hanada K, Yano K et al. High conservation of subunit compostion of RNA poly-
merase I(A) between yeast and mouse and the molecular cloning of mouse RNA poly-
merase I 40-kDa subunit RPA40. J Biol Chem 1994; 269(43):26976-26981.
26. Lanzendorfer M, Smid A, Klinger C et al. A shared subunit belongs to the eukaryotic core
RNA polymerase. Gene Dev 1997; 11(8):1037-1047.
27. Clawson GA. (personal communciation).
28. Hart IR. Tissue specific promoters in targeting systemically delivered gene therapy. Semin
Oncol 1996; 23(1):154-158.
29. Deonarain MP, Spooner RA, Epenetos AA. Genetic delivery of enzymes for cancer therapy.
Gene Ther 1995; 2(4):235-244.
30. Peehl DM. Prostate specific antigen role and function. Cancer 1995; 75(7 suppl): 2021-2026.
31. Lundwall A. Characterization of the gene for prostate-specific antigen, a human glandular
kallikrein. Biochem Biophys Res Commun 1989; 161(3):1151-1156.
32. Lee CH, Liu M, Sie KL et al. Prostate-specific antigen promoter driven gene therapy tar-
geting DNA polymerase-α and topoisomerase IIα in prostate cancer. Anticancer Res
1996;16(4A):1805-1812.
33. Cleutjens KB, van der Korput HA, van Eekelen CC et al. A 6-kb promoter fragment mim-
ics in transgenic mice the prostate-specific and androgen-regulated expression of the en-
dogenous prostate-specific antigen gene in humans. Mol Endocrinol 1997; 11(9):1256-1264.
34. Zhang S, Murtha PE, Young CY. Defining a functional androgen responsive element in the
5' far upstream flanking region of the prostate-specific antigen gene. Biochem Biophys Res
Commun 1997; 231(3):784-788.
35. Pang S, Dannull J, Kaboo R et al. Identification of a positive regulatory element respon-
sible for tissue-specific expression of prostate-specific antigen. Cancer Res 1997;
57(3):495-499.
36. Rodriguez R, Schuur ER, Lim HY et al. Prostate attenuated replication competent aden-
ovirus (ARCA) CN706: A selective cytotoxic for prostate-specific antigen-positive prostate
cancer cells. Cancer Res 1997; 57(13):2559-2563.
37. Spence AM, Sheppard PC, Davie JR et al. Regulation of a bifunctional mRNA results in
synthesis of secreted and nuclear probasin. Proc Natl Acad Sci USA 1989; 86(20):7843-7847.
38. Greenberg NM, DeMayo FJ, Sheppard PC et al. The rat probasin gene promoter directs
hormonally and developmentally regulation of a heterologous gene specifically to the pros-
tate in transgenic mice. Mol Endocrinol 1994; 8(2):230-239.
39. Price D. Comparative aspects of development and structure in the prostate. In: Vollmer
EP, Kauffman G, eds. Biology of the Prostate and Related Tissues. Washington DC: US
Government Printing Office, 1963:1-28.
40. Yan Y, Sheppard PC, Kasper S et al. A large fragment of the probasin promoter targets
high levels of transgene expression to the prostate of transgenic mice. Prostate 1997;
32:129-139.
41. Voeks DJ, Clawson GA, Norris JS. Tissue-specific triple ribozyme vectors for prostate can-
cer gene therapy. Gene Ther Mol Biol 1998; 1:407-418.
42. Voeks DJ, Staley MC, Greenberg NM et al. Prostate cancer gene therapy using targeted
expression of ribozymes against the AC40 subunit of RNA po11. In: Savage L, ed. Cancer
Gene Therapy. IBC 1997:213-222.
CHAPTER 14
Ribozymes in Targeting
Tumor Suppressor Genes
Tapas Mukhopadhyay and Jack A. Roth
A promising approach to gene therapy for cancer involves decreasing the level of abnor-
mal proteins by specific interference with gene expression at the mRNA level. The ex-
pression of a specific mRNA can be suppressed using a number of antisense approaches,
including ribozymes. Ribozymes are catalytic RNA molecules that can destroy the target
RNA. Ribozymes have several advantages over other available antisense approaches.
Ribozymes can inactivate the target RNA without relying on the host cell’s machinery, and
they can cleave more than one copy of the target RNA by dissociating from the cleavage
products and binding to another target molecule. Furthermore, ribozymes are effective in
modulating gene expression because of their simple structure, site-specific cleavage activity,
and catalytic potential. The targets of ribozyme-mediated gene modulation range from cancer
cells to foreign genes that cause infectious diseases, and additional target sites are being
developed. The numerous genes against which ribozymes have been targeted include
oncogenes (ras, BCR-ABL), telomeres, and drug-resistance genes (mdr1, c-fos). The demon-
stration that target RNA can be cleaved by cis ribozymes (catalytic RNAs, RNA enzymes)
has potentially important therapeutic implications.
Introduction
Certain RNA molecules have enzymatic activity, as was first described for the autocata-
lytic removal of the intervening sequence from the large ribosomal RNA precursor of Tet-
rahymena thermophila. Catalytic RNAs have now been described in a number of systems
ranging from bacteria to human. The ubiquity of catalytic RNA has prompted intense in-
vestigation into the understanding of its macromolecular function and potential applica-
tion in the therapy of genetic diseases, including cancer.1-3
Tumor cells undergo profound genetic alterations before a malignant phenotype is
established. Genetic changes such as mutation or overexpression can lead to oncogene acti-
vation, and these events are clearly important in the etiology of cancer. Structural and func-
tional alterations in a number of vital genes have been associated with a high percentage of
human cancers. Current experimental clinical approaches involve selectively killing cells
with unique cancer-related phenotypes, such as cell surface antigens or growth factor re-
ceptors, or altering the host immune system to attack cancer cells. However, the key to spe-
cific cancer therapy is likely to be the identification and targeting of processes that are unique
to the tumor. Cells expressing mutations in the oncoproteins that lead to the development
of the cancer are therefore ideal targets for gene therapy.
The inhibition of abnormally functioning oncogenes or tumor suppressor genes, and

thus the alteration of the malignant phenotype of the cell,4 is theoretically attractive, but
such biological therapy is beset with major logistic difficulties. Mutations in the vital genes
could lead to the production of faulty proteins with altered biological functions. During the
multistep process of carcinogenesis, a number of such faulty gene products accumulate and
provide a growth advantage to the cancer cells.
There are many ways to affect the activity of gene products: Ribozyme targeting of
mRNA is one of them. Ribozymes are a group of catalytic RNA molecules that cleave tar-
geted RNAs in a sequence-specific manner that is unique in their class. Ribozymes have the
potential to cycle; that is, having cleaved one target RNA, they can move on to the next. For
this reason, catalytic RNAs and their counterparts are being assessed as potential tools for
future gene therapy.
General Considerations for Ribozyme Function

The demonstration that RNA can be cleaved by ribozymes (catalytic RNAs, RNA en-
zymes) may have important therapeutic implications. Ribozymes are suitable for the modu-
lation of gene expression because of their simple structure, site-specific cleavage activity,
and catalytic potential. Ribozymes, a class of antisense agents, are small catalytic RNAs that
were initially discovered in a group-I self-splicing intron of the Tetrahymena pre-rRNA.5
Examples of these types of catalytic RNAs have now been found in many plants and ani-
mals. Ribozymes are small oligo-ribonucleotides whose specific catalytic domain has endo-
nuclease activity.6-8 This activity can be directed against virtually any RNA target by the
insertion of an antisense region into the ribozyme, but a GUC consensus sequence in the
target RNA is required at the targeted cleavage site. A typically altered ribozyme, called a
“hammerhead” ribozyme, is shown in Figure 14.1.
Despite so many promising attributes, ribozymes have their shortcomings. A major
disadvantage is that, being an RNA, ribozymes are particularly sensitive to nuclease diges-
tion. A number of modifications in ribozyme structure and function are currently under
investigation.9 One such strategy helps ribozymes achieve their delivery by genetic means,
in the form of gene constructs. Even though ribozyme-mediated gene inhibition involves a
mechanism (target cleavage) that is different from the standard antisense RNAs, many of
the essential steps required for gene inhibition by ribozyme and antisense RNA are identi-
cal. For example, ribozymes must exist long enough (long t1/2) to find their target sites in-
side the cell, and they should locate their target through intramolecular base pairing. Fur-
thermore, the relative concentrations of target and antisense or ribozyme affect the kinetics
of target localization. Once the problem of efficient and effective targeting is solved, ribozymes
will be effective RNA-inactivating agents with potentially great therapeutic applications. A
polymerase chain reaction-based “in vitro evolution” technique has been recently shown to
produce a population of ribozymes with 100-fold enhancement in target cleavage over the
original molecule.10
Targets for Ribozymes in Cancer

The targets of ribozyme-mediated gene modulation have ranged from cancer cells to
foreign genes that cause infectious diseases. Ribozymes have been targeted against numer-
ous genes, including oncogenes (ras, c-fos, BCR-ABL) and drug-resistance genes (mdr1).
Gene therapy, which involves treating a disease by altering the patient’s natural genetic make-
up, is fast emerging as a viable molecular technique in the treatment of cancer.
In recent years, our understanding of the features that distinguish a cancer cell from its
normal tissue counterpart has increased, and a number of vital gene defects accompanying
the onset of cancer have now been documented. Many genes involved in cancer exert their
Ribozymes in Targeting Tumor Suppressor Genes 177
Fig. 14.1. A schematic diagram

showing the structure of a typi-
cal “hammerhead” ribozyme.15
effect by overexpression, temporarily inappropriate expression, or in some cases, by expres-

sion of a faulty protein.
Cancer can also occur when novel sequences are generated from within the cell by
mutational processes like chromosomal translocation or rearrangement. Some genes are
abnormally expressed after translocation followed by fusion with other genes, for example,
p210BCR-ABL, bcl-2/immunoglobulin fusion gene, and t(15;17) in promyelocytic leuke-
mia, where the retinoic acid receptor α and zinc finger protein PML become fused. Several
gene defects resulting from a single point mutation lead to activation of a proto-oncogene
producing dominantly-acting transforming proteins (for instance, the ras oncogene). Such
gene abnormalities in cancer could be potential targets for ribozyme therapy. Other genes
involved in cancer exert their effect through overexpression of their normal structural pro-
teins like c-Fos, c-Myc, N-Myc, c-erb-2, and nucleolar antigen p120. These proteins could
be potential targets for antisense therapy. In addition, downregulation of MDR-1 and c-Raf1
can increase the sensitivity of the target tissue to conventional cytotoxic drugs. We have too
limited a scope here to describe the ribozyme action on all different target genes tested.
Here we discuss a few examples which anticipated the possibility that anti-oncogene
ribozymes could be used as antineoplastic agents in gene therapy protocols.
Cellular Genes Targeted by Ribozyme Molecules

Ribozymes have been used in a number of studies to target cellular oncogenes. One
study looked at the product of the c-myc oncogene, which is an important regulator of both
cell proliferation and programmed cell death (apoptosis). Hammerhead ribozymes were
shown to specifically cleave the v-myc but not the c-myc transcript in vitro. Molecular analysis
revealed a reduction of v-myc expression in ribozyme-expressing cells, which was associ-
ated with the abrogation of hormone-induced apoptosis.11 This confirms a direct involve-
ment of v-myc in the induction of apoptosis and indicates a potential means to control
apoptosis at the molecular level using a ribozyme-targeting approach.
Another study involved the activation of signal transduction pathways by mutation or
overexpression of cellular oncogenes, which has been associated with neoplastic transfor-
mation. The therapeutic potential of ribozymes targeted against the activated H-ras oncogene
as well as against the nuclear proto-oncogenes c-fos and c-myc in the FEM human mela-
noma cell line containing an H-ras mutation was evaluated. The anti-ras ribozyme could
affect not only the proliferation but also the differentiation process of human melanoma
cells in vitro. Anti-ras ribozyme clones showed a dendritic appearance in monolayer culture
which was associated with enhanced melanin synthesis. They also reinforced the role of
anti-oncogene ribozymes as suppressors of the neoplastic phenotype of melanoma cells.12,13
A study of EJ tumors containing the anti-ras ribozyme showed a reduction in tumor
size and prolonged survival compared with standard EJ cells.14 Ribozymes have been tar-
geted against H-ras oncogenes, and the expression of c-H-ras was suppressed in cells con-
taining ribozymes.15 These ribozymes cleaved the target RNAs in vitro and altered the cellu-
lar pathology. In several systems, the ability of a ribozyme to specifically cleave the mRNA
of the activated H-ras gene, and alter the malignant phenotype of an invasive human blad-
der cancer, was evaluated. Plasmids containing the ribozyme-encoding genes were expressed
under the control of the long terminal repeats (LTR) of Rous sarcoma virus in NIH3T3 cells
transfected with the activated c-H-ras gene. These ribozymes were found to inhibit the for-
mation of foci (by about 50%) by cleaving the oncogene mRNA rather than by hybridizing
to it. Moreover, the expression of c-H-ras was suppressed in cells expressing ribozymes.16 In
an EJ cell line that contained the activated H-ras gene, the efficacy of ribozyme action was
examined in athymic (nude) mice using an orthotopic model of bladder cancer. EJ tumors
containing the anti-ras ribozyme showed a reduction in EJ tumors expressing the H-ras
ribozyme, characterized by a marked reduction in tumor size and invasion compared with
those formed by control EJ cells.14,17 These studies suggest that the invasive phenotype is
blunted with the anti-ras ribozyme, delaying but not abolishing the metastatic pheno-
type.14,18,19
The tumor inhibitory activity of anti-ras ribozymes was also evaluated by a retroviral
gene delivery system.20 Using a tissue-specific (tyrosinase) promoter in a retroviral vector to
express the anti-ras ribozyme in human melanoma cells was found to be superior in sup-
pression of the human melanoma phenotype in vitro, as characterized by changes in growth,
melanin synthesis, morphology, and H-ras gene expression. Thus, the use of tissue-specific
expression of anti-oncogene ribozymes could have a potential therapeutic application in
human cancers.21 Anti-oncogene ribozymes may be useful as suppressors of tumor cell growth
and inhibitors of cellular transformation.22 A recombinant adenovirus was designed that
encoded a gene cassette for the H-ras ribozyme. By using this virus, the phenotype in mu-
tant H-ras-expressing tumor cells reverted without the need for any selection steps. The
demonstration of the utility of adenoviral-mediated delivery of anti-ras ribozymes was very
promising.23 Currently, the therapeutic application of ribozymes to human diseases is lim-
ited by the available gene transfer systems. However, adenoviral-mediated delivery of anti-
cancer ribozymes will allow the practical development of gene therapy strategies.
In a sarcoma model using a metastatic rat embryo cell line that constitutively secretes
MMP-9, a hammerhead ribozyme directed against the rat MMP-9 mRNA sequence resulted
in the absence of detectable MMP-9 mRNA and a loss of released 92 kDa gelatinase activity.
Although these cells were no longer metastatic in a lung colonization assay, they retained
their tumorigenic potential.24 The introduction of an expression vector for a control ham-
merhead ribozyme had no effect. These data document the requirement of MMP-9 expres-
sion for metastasis in this system.
One study determined the ability of ribozymes to inhibit c-fos gene expression. The c-
fos gene product Fos has been implicated in many cellular processes, including signal trans-
duction, DNA synthesis, and resistance to antineoplastic agents. fos ribozyme (catalytic RNA)
transfectants revealed decreased c-fos gene expression concomitant with reduced expres-
sion of thymidylate synthase, DNA polymerase beta, topoisomerase I, and metallothionein
IIA mRNA. In contrast, c-myc expression was elevated after fos ribozyme action.1 This study
established a role for c-fos in drug resistance and in mediating DNA synthesis and repair
processes by modulating the expression of these genes.
Another study used hammerhead ribozymes to the RNA of human papillomavirus
type 18 (HPV-18) in an HPV-18-expressing cell line. This ribozyme affected the phenotype
of HeLa cells, causing reduced growth rates, increased serum dependency, and reduced fo-
cus formation in soft agar. An increase in the intracellular concentration of the tumor sup-
pressor protein p53 appears to indicate that this ribozyme may be effective against cancers
that express HPV-18.25,26 Moreover, p53 pre-mRNA can be modified by a specific ribozyme
in vivo, suggesting a possible role for these agents in gene therapy strategies for cancer.27
The p53 gene is the most commonly mutated gene yet identified in human cancers.28
The gene product can regulate transcription and progress through the cell cycle, facilitating
repair of DNA damage.29-31 Wild type p53 (wtp53) can mediate a dominant tumor suppres-
sor effect,32,33 but mutation of p53 may confer a gain of function which enhances features of
the malignant phenotype.34-37
The mutated p53 gene product may oligomerize with wtp53, inactivating its tumor
suppressor function, or may acquire a prolonged half life. Therefore, treatment for malig-
nancies caused by loss of p53 tumor suppressor gene function would ideally be achieved by
introducing the wtp53 gene into tumor cells while eliminating the endogenous mutant gene.
Mutations in the p53 gene span a wide range of the coding region; gene replacement strate-
gies, therefore, are best targeted at the pre-mRNA level. We constructed a retroviral vector
that would deliver wtp53 into cells while expressing a ribozyme targeted to p53 pre-mRNA,
which would specifically block the endogenous mutated p53. We evaluated this vector as a
model for p53 gene replacement.
A ribozyme gene spanning a 52 bp sequence with StuI and BglII restriction enzyme
sites was designed. The catalytic domain of this Rz5a ribozyme recognizes a sequence at
codon 187 of p53 exon 6 close to intron 5. The flanking ribozyme sequences were antisense
to the intron 5-exon 6 boundary region, determining its specificity for unspliced p53 pre-
mRNA cleavage. Two synthetic primers with little overlap (5'-CCT GAG GAG GGG CCA
CTG ATG AGT CCT TTT G-3' and 5'- TGA TTG CTC TTA GGT TTC GTC CAA AAG GAC
TCA-3') were used to synthesize the ribozymes by polymerase chain reaction (PCR) ampli-
fication. Ribozyme DNA thus synthesized was subcloned into a StuI/BgIII site of the LNSX
retroviral vector, where expression of the ribozyme was under control of the SV40 pro-
moter. A DNA fragment containing β-actin promoter and wtp53 cDNA was subcloned into
a BgIII site of the same recombinant vector with the p53 gene under control of the β-actin
promoter. This vector was designed to transcribe a catalytic RNA molecule that was antisense
to the intron 5-exon 6 junction sequence (thus cleaving only endogenous p53 pre-mRNA)
and to express exogenous wt p53 cDNA. The ribozyme’s catalytic activity has been shown
previously to be limited to unspliced endogenous p53 pre-mRNA, thus having no effect on
p53 mRNA. H358 p53-null human non-small cell lung cancer cells expressed p53 protein
after transfection with Rzp53. After human non-small cell lung cancer cells H226Br (con-
taining a mutated p53 gene) were infected with the recombinant viral supernatant, they
expressed both the ribozyme and exogenous wt p53 RNA, as detected by reverse transcrip-
tion-polymerase chain reaction using sequence-specific amplimers. The ribozyme compo-
nent of the Rzp53 vector mediated cleavage of p53 pre-mRNA in vivo. Rzp53 had a growth-
inhibitory effect on transfected cells, as demonstrated by a proliferation assay.
The ribozyme has specificity for the p53 intron 5-exon 6 boundary sequence and cleaved
p53 unspliced mRNA at codon 187 (GUC) but did not cleave mature p53 mRNA. The
ribozyme sequence was subcloned into the retroviral LNSX vector such that expression of
the ribozyme was driven by the SV40 promoter. The vector also contained human p53 cDNA,
which was efficiently transcribed by the β-actin promoter. The expression of wt p53 was
evaluated by infecting the p53-negative H358 cell line with the Rzp53 viral supernatant. The
p53 protein expression in these infected cells was demonstrated by Western blot analysis
using an α-p53 monoclonal antibody. These cells displayed a strong p53 protein band, indi-
cating the expression of exogenous p53 in virus-infected cells.
The specificity of the in vitro catalytic activity of the Rz5 ribozyme for p53 pre-mRNA
has been shown previously. We examined the effect of this ribozyme on a human NSCLC
cell line, H226Br, which has a p53 mutation at codon 254 and expresses large quantities of
mutant p53 protein. H226Br cells were infected with Rzp53 retroviral supernatant and then
analyzed for the expression of ribozyme and exogenous wt p53. Reverse-transcriptase PCR
analysis using SV40 and β-actin promoter-specific amplimers indicated that transduced
H226Br cells expressed both ribozyme and exogenous wt p53 following Southern hybrid-
ization with p53 and ribozyme-specific probes.
To detect in vivo cleavage of the p53 target RNA by the ribozyme, total RNA was ex-
tracted from the Rzp53-transduced H226Br cells 6 h or 12 h after transfection. Total RNA
containing the ribozyme was used to cleave the in vitro labeled RNA substrate. The cleavage
products were separated on polyacrylamide gels, which were autoradiographed. The results
indicated that the total RNA contained anti-p53 ribozymes that efficiently cleaved the p53
pre-mRNA substrate in vitro. However, we were unable to detect cleavage products by RNase
protection assay or primer extension analysis.
The concomitant expression of anti-p53 pre-mRNA ribozyme and wild type p53 ex-
pression in the same cell was evaluated for its effect on cell proliferation. H226Br cells were
transduced with Rzp53 viral supernatant, and their growth was monitored for 7 days. H226Br
cells infected with LNSX vector only were used as a negative control. The growth rate for the
Rzp53-transduced cells was greatly reduced compared to that of non-transduced cells or
cells transduced with the empty LNSX vector. The cleavage specificity of our ribozyme re-
lies on the unspliced p53 RNA, and therefore it can cleave the endogenous p53 pre-mRNA
independently of the site of mutation.
With the development of a retroviral vector that induces the expression of both the
wild type tumor suppressor gene and the mutant-inhibiting ribozyme, we have made an
attempt to set up a model for gene replacement therapy. Treatment for malignancies caused
by defects in tumor suppressor gene function would be ideally achieved through the deliv-
ery of wt p53 at the same time that endogenous abnormal gene function is eliminated.
The precise role of p53 mutation in the transformation process is not clear at the present
time. The site of mutation appears to play a critical role in the activity of the resultant
mutant. Some mutants have gain of function and are able to cooperate with activated ras
oncogenes to increase colony formation.34,38 However, this does not appear to be mediated
by a transdominant negative effect of the protein, as cells transduced with a vector contain-
ing both wild type and mutated p53 genes co-expressed both genes but showed no inhibi-
tion of wt p53 growth suppression by the mutated p53 gene.39 Many mutations will cause
loss of the transactivating activity of the protein, but others do not.40,41 The tumor suppres-
sor activity of wt p53 may be retained despite the presence of a mutation in codon 248.42,43
A vector system that mediates expression of an anti-p53 ribozyme and restoration of wt p53
function in the cell carrying mutated p53 may prove useful for investigating mutant and
wild type interactions and represents a novel gene replacement strategy.
Conclusion
Ribozymes are a class of RNA molecules that can perform catalytically in the absence
of protein. Specifically, they can hybridize to and cleave target RNA molecules indepen-
dently of cellular proteins. The cleaved target RNA cannot be translated, thereby preventing
the synthesis of specific protein(s). Ribozymes targeted to the mRNA of key proteins in-
volved in maintaining a disease state may result in its elimination. The ribozymes can be
chemically synthesized and delivered directly to cells, or they can be expressed from an
expression vector after either permanent or transient transfection. Future directions in
ribozyme research include optimizing the design to obtain the maximal cleavage rate, in-
creasing the specificity to prevent aberrant reactions, identifying cleavage sites within the
target RNA, and delivering the ribozymes to cells of interest both in vitro and in vivo. The
ribozyme-mediated inhibition of gene expression may have potential therapeutic applica-
tions in the future treatment of cancer.
References
1. Scanlon KJ, Jiao L, Funato T et al. Ribozyme-mediated cleavage of c-fos mRNA reduces
gene expression of DNA synthesis enzymes and metallothionein. Proc Natl Acad Sci USA
1991; 88:10591-10595.
2. Sarver N, Cantin EM, Chang PS et al. Ribozymes as potential anti-HIV-1 therapeutic agents.
Science 1990; 247:1222-1225.
3. Sioud M, Natvig JB, Forre O. Preformed ribozyme destroys tumour necrosis factor mRNA
in human cells. J Mol Biol 1992; 223:831-835.
4. Kashani-Sabet M, Scanlon KJ. Application of ribozymes to cancer gene therapy (Review).
Cancer Gene Ther 1995; 2:213-223.
5. Kruger K, Grabowski PJ, Zaug AJ, Sands J, Gottschling DE, Cech TR. Self-splicing RNA:
Autoexcision and autocyclization of the ribosomal RNA intervening sequence of tetrahy-
mena. Cell 1982; 31:147-157.
6. Kim SH, Cech TR. Three-dimensional model of the active site of the self-splicing rRNA
precursor of tetrahymena. Proc Natl Acad Sci USA 1987; 84:8788-8792.
7. Gerlach WL, Llewellyn D, Haseloff J. Construction of a plant disease resistance gene from
the satellite RNA of tobacco rinspot virus. Nature (London) 1987; 328:802-805.
9. Erickson RP, Izant JG. Biology of antisense RNA and DNA. In: Erickson RP, Izant JG, eds.
Gene Regulation. New York: Raven Press, 1992.
10. Beaudry AA, Joyce GF. Directed evolution of an RNA enzyme. Science 1992; 257:613.
11. Dolnikov A, King A, Luxford C, Symonds G, Sun LQ. Ribozyme-mediated suppression of
v-myc expression abrogates apoptosis in transformed monocytes. Cancer Gene Ther 1996;
3:289-295.
12. Ohta Y, Kijima H, Kashani-Sabet M, Scanlon KJ. Suppression of the malignant phenotype
of melanoma cells by anti-oncogene ribozymes. Journal of Investigative Dermatology 1996;
106:275-280.
13. Ohta Y, Tone T, Shitara T et al. H-ras ribozyme-mediated alteration of the human mela-
noma phenotype. Ann N Y Acad Sci 1994; 716:242-253;discussion 253-6.
14. Eastham JA, Ahlering TE. Use of an anti-ras ribozyme to alter the malignant phenotype of
a human bladder cancer cell line. J Urol 1996; 156:1186-1188.
15. Koizumi M, Hayase Y, Iwai S, Kamiya H, Inoue H, Ohtsuka E. Design of RNA enzymes
distinguishing a single base mutation in RNA. Nucleic Acids Res 1989; 17:7059-7071.
16. Koizumi M, Kamiya H, Ohtsuka E. Ribozymes designed to inhibit transformation of NIH/
3T3 cells by the activated c-Ha-ras gene. Gene 1992; 117:179-184.
anti-ras ribozyme. Antisense Res Dev 1992; 2:3-15.
18. Kashani-Sabet M, Funato T, Florenes VA, Fodstad O, Scanlon KJ. Suppression of the neo-
plastic phenotype in vivo by an anti-ras ribozyme. Cancer Res 1994; 54:900-902.
19. Tone T, Kashani-Sabet M, Funato T et al. Suppression of EJ cells tumorigenicity. In Vivo
1993; 7:471-476.
20. Li MX, Lonial H, Citarella R, Lindh D, Colina L, Kramer R. Tumor inhibitory activity of
anti-ras ribozymes delivered by retroviral gene transfer. Cancer Gene Ther 1996; 3:221-229.
21. Ohta Y, Kijima H, Ohkawa T, Kashani-Sabet M, Scanlon KJ. Tissue-specific expression of

an anti-ras ribozyme inhibits proliferation of human malignant melanoma cells. Nucleic
Acids Res 1996; 24:938-942.
48:1471-1475.
55:2024-2028.
24. Hua J, Muschel RJ. Inhibition of matrix metalloproteinase 9 expression by a ribozyme
blocks metastasis in a rat sarcoma model system. Cancer Res 1996; 56:5279-5284.
25. Shillitoe EJ, Kamath P, Chen Z. Papillomaviruses as targets for cancer gene therapy (Re-
view). Cancer Gene Ther 1994; 1:193-204.
26. Chen Z, Kamath P, Zhang S, St. John L, Adler-Storthz K, Shillitoe EJ. Effects on tumor
cells of ribozymes that cleave the RNA transcripts of human papillomavirus type 18. Can-
cer Gene Ther 1996; 3:18-23.
27. Cai DW, Mukhopadhyay T, Roth JA. Suppression of lung cancer cell growth by ribozyme-
mediated modification of p53 pre-mRNA. Cancer Gene Ther 1995; 2:199-205.
28. Hollstein M, Sidransky D, Vogelstein B, Harris CC. p53 mutations in human cancers. Sci-
ence 1991; 253:49-53.
29. Diller L, Kassel J, Nelson CE et al. p53 functions as a cell cycle control protein in osteosa-
rcomas. Mol Cell Biol 1990; 10:5772-5781.
30. Kuerbitz S, Plunkett B, Walsh W, Kastan M. Wild-type p53 is a cell cycle checkpoint de-
terminant following radiation. Nature (London) 1992; 89:7491-7495.
31. Spandau DF. Distinct conformations of p53 are observed at different stages of keratinocyte
differentiation. Oncogene 1994; 9:1861-1868.
32. Baker SJ, Markowitz S, Pearson ER, Villson JKV, Vogelstein B. Suppression of human
colorectal carcinoma cell growth by wild-type p53. Science 1997; 249:912-915.
33. Cai DW, Mukhopadhyay T, Liu YJ, Fujiwara T, Roth JA. Stable expression of the wild-
type p53 gene in human lung cancer cells after retrovirus-mediated gene transfer. Hum
Gene Ther 1993; 4:617-624.
34. Hinds P, Finlay C, Levine AJ. Mutation is required to activate the p53 gene for coopera-
tion with the ras oncogene and transformation. J Virol 1989; 63 (2):739-746.
35. Hinds PW, Finlay CA, Quartin RS et al. Mutant p53 DNA clones from human colon car-
cinomas cooperate with ras in transforming primary rat cells: A comparison of the “hot
spot” mutant phenotypes. Cell Growth Differ 1990; 1:571-580.
36. Aaronson SA. Unique aspects of the interactions of retroviruses with vertebrate cells: CP
Rhoads memorial lecture. Cancer Res 1983; 43:1-5.
37. Espevik T, Figari IS, Ranges GE, Palladino MAJ. Transforming growth factor-B1 (TGF-B1)
and recombinant human tumor necrosis factor-α reciprocally regulate the generation of
lymphokine-activated killer cell activity. J Immunol 1988; 140:2312-2316.
38. Robbins KC, Antonaides HN, Shshilkumar GD, Hunkapiller MW, Aaronson SA. Struc-
tural and immunological similarities between simian sarcoma virus gene product(s) and
human platelet derived growth factor. Nature 1983; 305:605-608.
39. Frebourg T, Sadelain M, Ng YS, Kassel J, Friend SH. Equal transcription of wild-type and
mutant p53 using bicistronic vectors results in the wild-type phenotype. Cancer Res 1994;
54:878-881.
40. Miller CW, Chumakov A, Said J, Chen DL, Aslo A, Koeffler HP. Mutant-p53 proteins
have diverse intracellular abilities to oligomerize and activate transcription. Oncogene 1993;
8:1815-1824.
41. Finkel T, Der SJ, Cooper GM. Activation of ras genes in human tumors does not affect
localization, modification, or nucleotide binding properties of p21. Cell 1984; 37:151-158.
42. Cline MJ, Slamon DJ, Lipsick JS. Oncogenes: Implications for the diagnosis and treatment
of cancer. Ann Intern Med 1984; 101:223-233.
43. Duesberg PH. Retroviral transforming genes in normal cells. Nature 1983; 304:219-226.
CHAPTER 15
Inhibition of the Multidrug Resistance

Phenotype by Different Delivery
Systems of an Anti-mdr Ribozyme
Per Sonne Holm, David T. Curiel and Manfred Dietel
Introduction
I n the industrial countries of the world, roughly one person in five will die of cancer; only
heart diseases appear more often. Although considerable effort has been taken to improve
diagnosis and treatment of cancer it becomes more and more obvious that the traditional
approaches, i.e., surgery, radiation and chemotherapy, will not be able to cure all types of
cancer especially after systemic distribution. A major reason for the limited success of can-
cer therapy was found to be the enormous functional heterogeneity of cancer cells exhibit-
ing a variety of different characteristics, such as the potency for infiltrative growth, the pos-
sibility to metastasize, the activation of proteolytic or detoxifying enzymes and the
overexpression of membrane-bound pump proteins to decrease intracytoplasmic levels of
toxic drugs.
In cases where local measures, such as surgery or radiation, are no longer possible,
systemic chemotherapy is the matter of choice. In many cases repeated treatments with
drugs that are selectively toxic to dividing cells will kill the majority of neoplastic cells.
However, therapy with cytostatic drugs often shows only limited effectiveness. Those cells
that are exposed to one drug often develop a resistance not only to the specific cytostatic
drug with which they have been treated and to which they were initially sensitive, but also to
other drugs to which they have never been exposed.1 In addition, the application of anti-
cancer drugs often has the disadvantage that it unspecifically hampers normal tissue and
cells, inducing severe side effects. Among the relatively great number of resistance-associ-
ated mechanisms, such as multidrug resistance-related proteins,2,3 alteration of topoisomarase
II,4 and increased activity of glutathione-S-transferase, considerable attention in cancer re-
search has focused on the so-called multidrug resistance phenotype (MDR).5-7 Multidrug
resistance describes the simultaneous expression of cellular resistance to a wide range of
structurally and functionally unrelated lipophilic drugs. The usual pattern of cross-resis-
tance includes a large variety of cytotoxic agents that do not have a common structure or a
common intracellular target, such as anthracyclines (adriamycin), alkylating agents
(melphalam), heavy metal compounds (cisplatin) and alkaloids (vinblastine). The develop-
ment of this form of resistance is clinically recognized by a short period or a total lack of
drug effectiveness.
Generally, MDR cells appear to have a reduced ability to accumulate the different drugs,
which is often accompanied by the overexpression of a membrane-bound transport pro-
tein, a P-170 kDa glycoprotein.8,9 The elevation in the expression of this protein is mediated
by elevated levels of the corresponding mdr1 4.5 kb mRNA.10,11 P-170 is encoded by a small
family of genes. In humans two MDR genes are described, while in hamsters and mice three
genes could be identified.12,13 The gene coding for P-glycoprotein is localized on chromo-
some 7.14 By means of transfection experiments it was shown that only the mdr1 gene is
responsible for the clinically relevant MDR.15 P-glycoprotein consists of 1280 amino acids16
and contains several transmembrane regions which form a pore-like structure (Fig. 15.1).
P-glycoprotein has two ATP binding sites17,18 and functions as a membrane efflux pump
which reduces the intracellular concentration of cytostatic drugs.19 In addition, it could be
shown that an increased synthesis of P-glycoprotein correlates with an elevated degree of
resistance.20 Although other models for the mechanism of P-glycoprotein-mediated resis-
tance have been proposed, the possibility that P-gp acts as a multidrug transporter has not
been excluded.21-23 At present the molecular mechanisms underlying the development of
resistance are the subject of many investigations. For example, Chin et al24 were able to show
that the expression of the mdr1 gene is influenced by Ha-ras and p53.
Several types of strategies have been utilized to modulate the MDR phenotype both in
vivo and in vitro. The reversing drugs are biochemically completely different from each
other, including drug analogs (N-acetyl daunomycin), calcium channel blockers (verapamil),
calmodulin modulators (trifluoperazine) and immunosuppressants (cyclosporine A, FK 506,
PSC 833).25-27 However, these substances are still acting in the conventional way, by limiting
more or less directly the function of P-glycoprotein as an energy-dependent pump. Unfor-
tunately, clinical studies using these compounds have been impeded by the significant tox-
icity and limited specificity of these chemosensitizers. The use of the anti-MDR1 mono-
clonal antibody MRK-16 presents another selective approach for the modulation of the
multidrug resistance phenotype.28 However, there is a considerable need to find new ways
to reverse multidrug resistance.
In recent years researchers have extensively studied another alternative approach to
eliminate the function of P-glycoprotein and, consequently, to reverse the resistant pheno-
type in tumor cells, by using antisense RNA/DNA or ribozymes.29 Particular hammerhead
ribozymes which can be applied against different target RNA sequences30,31 provide a genu-
ine alternative to the conventional antisense strategies32,33 and are useful tools to inhibit
unwelcome gene expressions. Compared with the RNA antisense method, hammerhead
ribozymes have the advantage of cleaving the target RNA after base pairing, to subsequently
emerge unchanged from the reaction and cleave several more substrate RNA molecules.30,34
Based on these findings it seems that ribozymes are simply an advance form of antisense
molecule.35
It is the goal of this chapter to describe the construction of a hammerhead ribozyme
that is directed against the mdr1 mRNA which abrogates P-glycoprotein expression by tu-
mor cells and to summarize the results obtained in vitro and in cell culture experiments.
Further, this chapter will also focus on different delivery systems of the anti-mdr ribozyme.
Catalytic Activity of the mdr Ribozyme

In order to reverse the resistant phenotype in cell culture, we designed a hammerhead
ribozyme recognizing the GUC triplet at position +2637 to +2639 in exon 21 of the mdr1
mRNA. For the in vitro analyses, it consisted of a 43 base long RNA molecule. The 5' and 3'
end sequences of the ribozyme complementary to the substrate were 10 and 11 bases, re-
spectively (Fig. 15.2). This guarantees a high specificity of the ribozyme, since a sequence of
16 nucleotides statistically is represented only once in the human genome of approximately
Inhibition of the mdr Phenotype by Different Delivery Systems" 185
Fig. 15.1. Structural properties of the channel-forming membrane spanning P-glycopro-

tein. Total length is 1280 amino acids; the two halves have similar sequence and an ATP
binding site each. The major classes of transported drugs and the efflux-blocking agents
are indicated. The binding sites of the monoclonal antibody C-219 are marked by stars.
Fig. 15.2. Schematic structures of the

anti-mdr ribozyme annealed to a
part of the 259 nt of exon 21 of the
mdr1 mRNA. The transcription plas-
mid derived from pBLUESCRIPT
IISK was obtained as described.47
4 x 109 bp. The choice of the specific target site was based on the two ATP binding sites,
which are suggested to be important for the function of P-glycoprotein. Before testing the
ribozyme in tissue culture experiments, we investigated its ability to cleave the target se-
quence in a cell-free system.
Cleavage of the target RNA (259 nt transcript and a 25-mer RNA substrate) is shown
in Figure 15.3A and 3B. Optimal conditions for cleavage of the 259 nt transcript by the
ribozyme were approximately pH 8.0, 12 mM magnesium chloride and 52°C (data not
shown). Often rather low cleavage efficiencies have been reported with large substrates (be-
tween 60 and 954 nucleotides).36,37 For example, with a 50-fold molar excess of a hammer-
head ribozyme, only about 61% of a 261 nucleotide fragment of calretinin mRNA was cleaved
after incubation for 90 min at 37°C. In contrast, using only a 4-fold excess of our mdr
ribozyme, about 70% of the 259 nucleotide fragment of mdr1 mRNA was cleaved after
60 min at 37°C (not shown). Further, we could observe multiple turnover. In cases when
4 nM ribozyme and 400 nM substrate were used, one ribozyme could cleave up to 12 sub-
strate molecules within 60 min. Not only temperature, but also the sequence of hybridiza-
tion (G-C content to be considered) or the length of a ribozyme influence cleavage activity.
While constructing longer sequences of hybridization the Km values remain nearly con-
stant, whereas a clear diminution of the kcat values can be observed. This is due to the fact
that there is a decrease in the ribozyme’s dissociation rate from the cleavage products.38 For
the time being it is difficult to evaluate whether such in vitro results are of importance in
vivo.
The in vitro kinetics of cleavage reactions by hammerhead ribozymes has been studied
in detail.39 A severe limitation is the poor cleavage efficiency of large substrates, in contrast
to the high activities observed with small substrates (oligoribonucleotides). In multiple turn-
over reactions the values of kcat/Km using short substrates are about 400-1100 mM–1sec–1.30,40,41
Using a long substrate as target (985 nucleotides) Heidenreich and Eckstein42 determined
the value of kcat/Km at four orders of magnitude lower. Thus, it is assumed that large RNA
molecules can form complex secondary and tertiary structures which interfere with the
recognition and cleavage by hammerhead ribozymes, and might therefore contribute to the
rate-limiting step.37 In contrast, the kcat values of the anti-mdr ribozyme for small and large
RNAs were identical, whereas Km was increased only 5-fold with the large substrate.43 As
studies of antisense oligonucleotides have already shown,44 not all mRNA areas are equally
suitable as target sequence. For reasons still unknown, the sequence here selected for the
docking of the anti-mdr ribozyme appears to be particularly well available or recognizable
to the ribozyme. Our data suggest that access of the ribozyme to the target sequence was
almost as easy as with the oligonucleotide, and is in agreement with some of the computer
predicted structures (not shown). However, it is not certain that in vitro results are compa-
rable with the situation in vivo. Also, according to the latest investigations, the “targeting”,
i.e., the ribozyme’s assignment to the same cellular compartment as the mRNA to be cut,
and the RNA binding proteins play a decisive role for the ribozyme’s efficacy.45,46 At present
there is no good way to predict optimal target sites, and testing of several different ribozymes
with a specific mRNA may be necessary to identify suitable target sequences for biologically
long substrate RNAs.
Delivery Systems of the Anti-mdr Ribozyme

Endogenous Delivery
The strategy for introducing antisense or catalytic RNA molecules into cells has be-
come an important issue in cancer gene therapy. Two general mechanisms have been devel-
oped for introducing antisense or catalytic RNA molecules into target cells selectively and
Fig. 15.3. Examples of the

1 min 2 min 5 min
A 20 40 80 200 20 40 80 200 20 40 80 200
cleavage of the 259 nt substrate
(A) and 25-mer substrate (B).
4 nM ribozyme (6 nM for
S 259 nt substrate) were incu-
bated with 20, 40, 80 and
200 nM substrate for 1, 2 and
5 min. The products were ana-
lyzed on a 10% (A) or 20% (B)
denaturing polyacrylamide gel.
3'–P(142 nt) R, ribozyme; S, substrate; P,
product.
5'–P(117 nt)
1 min 2 min 5 min

B 20 40 80 200 20 40 80 200 20 40 80 200
efficiently: endogenous expression of a ribozyme from a DNA template delivered by differ-

ent vector systems, and exogenous delivery. Several experiments performed during the last
5 years have shown that by using different strategies the application of antisense oligonucle-
otides or ribozymes can induce a specific inhibition of the mdr1 mRNA and the P-glycopro-
tein, resulting in a reduced multidrug resistance.47-51
The use of plasmids or viral vectors is one of the most promising strategies for gene
delivery and endogenous expression of ribozymes in cells. For this method, however, suit-
able vectors are needed, including plasmids, retroviruses, adenoviruses and adeno-associ-
ated viruses (see also Section II, Expression and Delivery of Ribozymes). The ribozyme
gene is placed downstream of a strong promoter/enhancer system, so that it is highly ex-

pressed in the cells. The ribozyme can be expressed either by the polymerase II or by the
polymerase III promoter-system. In spite of some advantages of the pol II promoter (poly
(A) tail may prolong half life of the ribozyme), it seems that only under the control of
pol III can the expression of ribozymes reach a 100- to 1000-fold excess of the ribozyme
over the target RNA, which is necessary to obtain a detectable decrease in RNA levels.52-54
To reverse the MDR phenotype, a mammalian expression vector containing the 43 bp
DNA sequence that encodes the mdr1 ribozyme was transfected into the human pancreatic
carcinoma cell line EPP85-181RDB, which expresses the MDR phenotype. This cell line was
established previously by growing the parental cells EPP85-181P in increased concentra-
tions of daunorubicin, resulting in a 1600-fold daunorubicin resistance. After transfection
and selection two cell clones that expressed the ribozyme could be isolated (EPP85-181RDB-
Rb1 and EPP85-181RDB-Rb2). In these cell clones the amount of mdr1 mRNA expression
and P-glycoprotein formation was reduced to such an extent that they were no longer de-
tectable (Figs. 15.4 and 15.5). With reference to the IC50 values, the resistance was reduced
almost completely, whereas the clone which contained only the expression vector
(EPP85-181RDB-Vec) continued to be as resistant to daunorubicin as the original resistant
cell line EPP-181RDB (Fig. 15.6 and ref. 47). The almost complete reversion of the
daunorubicin resistance by transfection of the mdr1 ribozyme excludes mechanisms of re-
sistance other than the typical MDR, which means that the increased expression of P-glyco-
protein in the EPP85-181RDB cell line is the main factor of resistance.
Adenovirus vectors are probably the most efficient tool for delivering foreign genes
into mammalian cells both in vitro and in vivo. Several advantages of this vector include the
infection of a wide variety of cell types, replicating and non-replicating, produced and pu-
rified in a high titer up to 5 x 1011 plaque forming units (PFU/ml) and the high levels of
transgene expression.
To determine the feasibility of employing the adenoviral vector for anti-mdr ribozyme
delivery, we constructed a recombinant adenovirus. The expression of the ribozyme was
directed by a polymerase III promoter (Va I promoter). Forty-eight hours after infection of
MCF-7Ad cells with 200 PFU AdVaIRib (>95% transduction frequency) and 200 PFU of
the control virus AdCMVLacZ (recombinant adenovirus encoding the β-galactosidase re-
porter gene), 100-1000 ng/ml vincristine was added to the medium; cell viability was as-
sayed after a further 72 hours (Fig. 15.7). Untransfected MCF-7Ad cells and cells infected
with the control virus did not demonstrate enhanced sensitivity to vincristine, whereas the
cells infected with the anti-mdr ribozyme induced strong sensitivity to vincristine. These
results demonstrate that it is possible to downregulate mdr expression without any kind of
selection. Together with further studies55,56 they demonstrate the potential of an adenovi-
ral-mediated delivery of ribozymes for cancer gene therapy. On the basis of these results, we
are currently in the process of evaluating this concept in a murine model.
Despite several advantages of Ad vectors for gene therapy, there exists one drawback in
using adenovirus as delivery system in vivo: its immunogenic potential.57 To achieve the
goal of using adenovirus in gene therapy, it will be necessary in the future to design Ad
vectors that will prevent the induction of immune response.
Exogenous Delivery
Although much improvement in transfection has been achieved with newly developed
cationic lipids,58,59 several limitations still exist with this exogenous nonviral delivery sys-
tem of ribozymes. This includes low stability, since they can easily be attacked and destroyed
by the nucleases contained in the serum. Kinetic investigations, however, show that a spe-
cific insertion of 2'-modified ribonucleotides, such as 2'-fluoro, 2'-amino, 2'-O-methyl and
1 2 3 4 5 6 7
4.5 kb mdr1 mRNA
Fig. 15.4. Northern blot analysis of mdr1 mRNA using a 785 bp probe. The auto-
radiograph demonstrates no detectable signal with 10 µg of total cellular RNA of
EPP85-181RDB-Rb1/2 (lanes 2 and 3) and EPP85-181P (lane 4) grown in
0.0125 µg ml-1 daunorubicin. The resistant cell line (EPP85-181RDB) and the
cell line containing only the vector (EPP85-181RDB/Vec), when grown in
2.5 µg ml-1 daunorubicin, show a clear signal (lanes 1 and 7). The resistant cell
line shows a weaker signal when grown in 0.025 µg ml-1 daunorubicin (lane 6),
and almost no signal when grown in the absence of daunorubicin (lane 5).
Fig. 15.5. Immunocytochemistry of P-glycoprotein in the human pancreatic

cell lines EPP85-181RDB (a), EPP85-181RDB-Rb1 (b), EPP85-181P (c),
EPP85-181RDB/Vec (d). The monoclonal antibody C-219 (α-P-glycopro-
tein) was used to label the cells; labeling of the cell membrane is seen in (a),
the resistant line, and (d), the resistant line with introduced vector only; (b),
with the introduced ribozyme, restores the sensitivity seen in the parental
line (c). The cell clones in (a) and (d) were grown in 2.5 µg ml-1 daunorubicin
and those in (b) and (c) in 0.0125 µg ml-1 daunorubicin.
growth in % control
Fig. 15.6. Determination of the daunorubicin-induced IC50 for parental and resistant cell lines
(EPP85-181P, EPP85-181RDB (K1-3)), the two cell lines containing the ribozyme
(EPP85-181RDB-Rb1, EPP85-181RDB-Rb2), and the cell line containing only the expression
vector (EPP85-181RDB/Vec). Data points represent cell counts of cultures.
2'-deoxynucleotides within one ribozyme will increase its stability many times.40,41,60-62 Fur-
ther, the substitute must be chosen very carefully, since modifications can also decrease the
cleavage activity. Another major barrier to the development of ribozymes for gene therapy
is their very low permeability to cellular membranes compared to viral vector delivery. A
further drawback of exogenous delivery is the fact that repeated administration is required
since the delivery is transient and inefficient. Another question is how ribozymes are able to
escape being trapped in the endosome. One approach to facilitate the release of DNA mol-
ecules from endosomes in order to avoid a prolonged exposure to this environment is the
use of a replication-defective adenovirus.63 Raja-Walia et al64 reported up to 1000-fold en-
hanced gene transfer by using a complex of plasmid DNA/cationic lipids with a replication-
deficient adenovirus. This method still remains to be performed with ribozymes.
Another strategy to increase the acceptance rate is to combine molecules at the 5' end
or 3' end with cholesterol or other lipophil groups.65 A recently proposed possibility for
achieving better biological availability in the cytoplasm is the synthesis of ribozymes with
hexaethylene glycol as a linker. It can be shown that hexaethylene glycol linkers can be in-
serted not only terminally but also in loop II of ribozymes without a significant loss of
cleavage activity.66 In what way it will influence stability and acceptance rate has still to be
investigated. Based on recent findings,67,68 several mdr ribozymes have been chemically syn-
thesized which contain both hexaethylene glycol linker and 2'-modified ribonucleotides;
work is in progress to determine whether an exogenous application is able to modulate the
MDR phenotype.
Fig. 15.7. Adenovirus-mediated delivery of an anti-mdr ribozyme enhances sensitivity to vinc-

ristine. MCF-7Ad cells were infected with either 200 PFU/cell of an adenovirus expressing an
anti-mdr ribozyme (AdVaIrib) or 200 PFU/cell AdLacZ. Two days after infection cells were
treated with vincristine. (a) Cells without any drugs or virus; (b) Cells treated with 1000 ng/ml
vincristine; (c) AdlacZ +1000 ng/ml vincristine; (d) AdVaIrib + 100 ng/ml vincristine
Conclusion
The extremely rapid development of molecular biology and the discovery of ribozymes,
combined with improvements in delivery systems, presumably opens the way to introduce
ribozymes for specific antitumor therapy in vivo. The catalytic efficiency of ribozymes com-
bined with the high efficiency of gene transfer and transgene expression in mammalian
cells mediated by recombinant adenovirus may in the near future result in the development
of an effective in vivo strategy for modulating the MDR phenotype in human cancer. How-
ever, there exist several basic issues which have to be solved prior to clinical trials: e.g., how
can tumor cells be reached “specifically” to avoid side effects, and how is it possible to in-
crease the rate of cellular uptake when administered by an exogenous pathway? These ques-
tions, however, reflect not only the difficulties of molecular antitumor approaches but the
fundamental and general difficulties of drug-oriented treatment of malignant tumors.
Acknowledgments
We wish to thank Frau Stemmler for editorial help.
References
1. Dietel M. What’s new in cytostatic drug resistance and pathology. Path Res Pract 1991;
187:892-905.
2. Cole SPC, Bhardwaj G, Gerlach JH et al. Overexpression of a transporter gene in a
multidrug-resistant human lung cancer cell line. Science 1992; 258:1650-1654.
3. Grant CE, Valdimarsson G, Hipfner DR et al. Overexpression of multidrug resistance-as-

sociated protein (MRP) increases resistance to natural product drugs. Cancer Res 1994;
54:357-361.
4. Beck WT, Kim R, Chen M. Novel action of inhibitors of DNA topoisomerase II in drug
resistant tumor cells. Cancer Chemoth Pramacol 1994; 34:14-18.
5. Roninson IB. The role of the mdr1 (P-Glycoprotein) gene in multidrug resistance in vitro
and in vivo. Biochemical Pharmacology 1992; 43:95-102.
6. Gottesman MM, Pastan I. Biochemistry of multidrug resistance mediated by the multidrug
transporter. Ann Rev Biochem 1993; 62:385-427.
7. Nielsen D, Skovsgaard T. P-Glycoprotein as multidrug transporter: A critical review of
current multidrug resistant cell lines. Biochimica et Biophysica Acta 1992; 1139:169-183.
8. Ueda K, Cardarelli C, Gottesman MM, Pastan I. Expression of a full length cDNA for the
human mdr1 gene confers resistance to colchicine, doxorubicin and vinblastine. Proc Natl
Acad Sci USA 1987; 84:3004-3008.
9. Pastan I, Gottesman M. Multiple-drug resistance in human cancer. N Engl J Med 1987;
316:1388-1393.
10. Roninson IB, Chin JE, Choi K et al. Isolation of human mdr DNA sequences amplified in
multidrug resistant KB carcinoma cells. Proc Natl Acad Sci USA 1986; 83:4538-4542.
11. Fojo AT, Ueda K, Salmon DJ et al. Expression of a multidrug resistance gene in human
tumors and tissues. Proc Natl Acad Sci USA 1987; 84:265-269.
12. Riordan JR, Deuchars K, Kartner N et al. Amplification of P-glycoprotein genes in
multidrug-resistant mammalian cell lines. Nature 1985; 316:817-819.
13. Gros P, Croop J, Roninson I et al. Isolation and characterization of DNA sequences ampli-
fied in multidrug resistant hamster cells. Proc Natl Acad Sci USA 1986; 83:337-341.
14. Callen DF, Baker E, Simmers RN et al. Localization of the human multiple drug resistance
gene, mdr1, to 7q21.1. Hum Genet 1987; 77:142-144.
15. Pastan I, Gottesman MM, Ueda K et al. A retrovirus carrying an MDR1 cDNA confers
multidrug resistance and polarized expression of P-glycoprotein in MDCK cells. Proc Natl
Acad Sci USA 1989; 85:4486-4490.
16. Kartner N, Ling V. Vielfach-Resistenz von Krebszellen. Spektrum der Wissenschaft 1989;
5:64-71.
17. Gros P, Neriah YB, Croop JM et al. Isolation and expression of a complementary DNA
that confers multidrug resistance. Nature 1986; 323:728-731.
18. Cornwell M, Tsudo T, Gottesman M et al. ATP-binding properties of P-glycoprotein from
multidrug-resistant KB cells. FASEB J 1987; 1:51-54.
19. Fojo A, Akiyama SI, Gottesman MM et al. Reduced drug accumulation in multiply drug-
resistant human KB carcinoma cell lines. Cancer Res 1985:47:3002-3007.
20. Scotto KW, Biedler JL, Melera PW. Amplification and expression of genes associated with
multidrug resistance in mammalian cells. Science 1986; 232:751-755.
21. Endicott J, Ling V. The biochemistry of P-glycoprotein mediated multidrug resistance. Annu
Rev Biochem 1989:58:137-171.
22. Willingham MC, Cornwell MM, Cardarelli CO et al. Single cell analysis of daunomycin
uptake and efflux in multidrug-resistant and -sensitive KB cells: Effects of verapamil and
other drugs. Cancer Res 1986; 46:5941-5946.
23. Hoffman MM, Wie LY, Roepe PD. Are altered pHi and membrane potential in hu MDR1
transfectants sufficient to cause MDR protein-mediated multidrug resistance. J Gen Phyiol
1996; 108:295-313.
24. Chin, K-V, Ueda K, Pastan I et al. Modulation of activity of the promoter of the human
MDR1 by Ras and p53. Science 1992; 25:459-462.
25. Deuchars KL, Ling V. P-Glykoprotein and multidrug resistance in cancer chemotherapy.
Semin Oncol 1989; 16:156-165.
26. Reymann A, Looft G, Woermann C et al. Reversal of multidrug resistance in Friend leuke-
mia cells by dexniguldipine-HCL. Cancer Chemotherapy Pharmacology 1994; 32:25-30.
27. Dietel M, Herzig I, Reymann A et al. Secondary combined resistance to the multidrug-
resistance-reversing activity of cyclosporin A in the cell line F4-6RADR-CsA. J Cancer Res
Clin Oncol 1994; 120:263-271.
28. Fogler WE, Pearson JW, Volker K et al. Enhancement of recombinant human interferon
alfa of the reversal of multidrug resistance by MRK-16 monoclonal antibody. J Natl Can-
cer Inst 1995; 87:94-103.
29. Bouffard DY, Ohkawa T, Kijima H et al. Oligonucleotide modulation of multidrug resis-
tance. European J of Cancer 1996; 32:1010-1018.
30. Uhlenbeck OA. A small catalytic oligoribonucleotide. Nature 1987; 328:596-600.
32. Stein CA, Cohen JS. Oligonucleotides as inhibitors of gene expression: A review. Cancer
Res 1988; 48:2659-2668.
33. Tidd DM. Synthetic oligonucleotides as therapeutic agents. Br J Cancer 1991; 63:6-8.
34. Sarver N, Cantin EM, Chang PS et al. Ribozymes as a potential anti-Hiv-1 therapeutic
agents. Science 1990; 247:1222-1225.
35. Rossi JJ, Sarver N. Catalytic Antisense RNA (Ribozymes): Their potential and use as anti-
HIV therapeutic agents. In: Block T et al, ed. Innovations in Antiviral Development and
the Detection of virus infection. 1992:95-109.
36. Heidenreich O, Eckstein F. Hammerhead ribozyme-mediated cleavage of the long terminal
repeat RNA of human immunodeficiency virus type 1. J Biol Chem 1992; 267:1904-1909.
37. Bertrand E, Pictet R, Grange T. Can hammerhead ribozymes be efficient tools to inactivate
gene function? Nucleic Acids Res 1994; 22:293-300.
38. Bertrand E, Grange T, Pictet R. Transacting Hammerhead Ribozymes In Vivo. Present Limits
and Future Directions. Gene Regulation, Biology of Antisense RNA and DNA. Vol. 1. New
York: Raven Press, Ltd., 1992:71-81.
39. Hertel KJ, Hershlag D, Uhlenbeck OC. A kinetic and themodynamic framework for the
hammerhead ribozyme reaction. Biochemistry 1994; 33:3374-3385.
40. Paolella G, Sproat BS, Lamond AI. Nuclease resistant ribozymes with high catalytic activ-
ity. EMBO J 1992; 11:1913-1919.
41. Pieken WA, Olsen DB, Benseler F et al. Kinetic characterization of ribonuclease-resistant
2'-modified hammerhead ribozymes. Science 1991; 253:314-317.
42. Heidenreich O, Eckstein F. Hammerhead ribozyme-mediated cleavage of the long terminal
repeat RNA of human immunodeficiency virus type 1. Biol Chem 1992; 267:1904-1909.
43. Holm PS, Dietel M, Krupp G. Similar cleavage effiencies of an oligoribonucleotide sub-
strate and an mdr1 mRNA segment by hammerhead ribozyme. Gene 1995; 167:221-225.
44. Yen TJ, Machlin PS, Cleveland DW. Autoregulated instability of beta-tubulin mRNAs by
recognition of the nascent amino terminus of beta-tubulin. Nature 1988; 334:580-585.
45. Sullenger BA, Cech TR. Tethering ribozymes to a retroviral packaging signal for destruc-
tion of viral RNA. Science 1993; 262:1566-1569.
46. Bertrand E, Rossi JJ. Facilitation of hammerhead ribozyme catalysis by the nucleocapsid
protein of HIV-1 and the heterogeneous nuclear ribonucleoprotein A1. EMBO J 1994;
13:2904-2912.
47. Holm PS, Scanlon K, Dietel M. Reversion of the multidrug resistance in the P-glycopro-
tein positive human pancreatic cell line (EPP85-181RDB) by the introduction of a ham-
merhead ribozyme. Br J Cancer 1994; 70:239-243.
49. Cucco C, Calabretta B. In vitro and in vivo reversal of multidrug resistance in a human
leukemia-resistant cell line by mdr1 antisense oligodeoxynucleotides. Cancer Res 1996;
56:4332-4337.
50. Alahari SK, Dean NM, Fischer MH et al. Inhibition of expression of the multidrug resis-
tance associated P-glycoprotein by phosphorothioate and 5'cholesterol-conjugated
phosphorothioate antisense oligonucleotides. Molecular Pharmacology 1996; 50:808-819.
51. Kobayashi H, Dorai T, Holland JF et al. Reversal of drug sensivity in multidrug-resistant

tumor cells by an MDR1 (PGY1) ribozyme. Cancer Res 1994; 54:1271-1275.
52. Cotten M, Birnstiel ML. Ribozymes mediated destruction of RNA in vivo. EMBO J 1989;
8:3861-3866.
53. Lieber A, Strauss M. Selection of efficient cleavage sites in target RNAs by using a ribozymes
expression library. Molecular and Cellular Biology 1995; 15:540-551.
54. Cagnon L, Cucchiarini M, Lefebre JC et al. Protection of a T-cell line from human immu-
nodeficiency virus replication by the stable expression of short antisense RNA sequence
carried by a shuttle RNA molecule. J of Acquired Immune Defiency Syndromes and Hu-
man Retrovirology 1995; 9:349-358.
55. Feng M, Cabrera G, Deshane J et al. Neoplastic reversion accomplished by high efficiency
adenoviral-mediated delivery of an anti-ras ribozyme. Cancer Res 1995; 55:2024-2028.
56. Lieber A, Kay MA. Adenovirus-mediated exoression of ribozymes in mice. J Virology 1996;
70:3153-3158.
57. Yang Y, Nunes FA, Berensci K et al. Cellular immunity to viral antigens limits E1-deleted
adenoviruses for gene expression. Proc Natl Acad Sci USA 1994; 91:4407-4411.
58. Huang L, Gao X. Cationic liposome-mediated gene transfer. Gene Therapy 1995; 2:710-722.
59. Lewis JG, Lin KY, Kothavale A et al. A serum-resistant cytofectin for cellular delivery of
antisense oligodeoxynucleotides and plasmid DNA. Proc Natl Acad Sci USA 1996;
93:3176-3181.
60. Shimayama T, Nishikawa F, Nishikawa S et al. Nuclease-resistant chimeric ribozymes con-
taining deoxyribonucleotides and phosphorothioate linkages. Nucl Acids Res 1993;
21:2605-2611.
61. Taylor NR, Kaplan BE, Swiderski P et al. Chimeric DNA-RNA hammerhead ribozymes
have enhanced in vitro catalytic efficiency and increased stability in vivo. Nucl Acids Res
1992; 20:4559-4565.
62. Williams DM, Pieken WA, Eckstein F. Function of specific 2'-hydroxyl groups of gua-
nosines in a hammerhead ribozyme probed by 2' modifications. Proc Natl Acad Sci USA
1992; 89:918-921.
63. Curiel DT, Agarwal S, Wagner E. Adenovirus enhancement of transferrin-polylysine-medi-
ated gene delivery. Proc Natl Acad Sci USA 1991; 88:8850-8854.
64. Raja-Walia R, Webber J, Naftilan J et al. Enhancment of liposome-mediated gene transfer
into vascular tissue by replication deficient adenovirus. Gene Therapy 1995; 2:521-530.
65. Mackellar C, Graham D, Will DW et al. Synthesis and physical properties of anti-HIV
antisense oligonucleotides bearing terminal lipophilic groups. Nucleic Acids Res 1992;
20:3411-3417.
66. Thomson JB, Tuschl T, Eckstein F. Activity of hammerhead ribozymes containing non-
67. Lyngstadaas SP, Risnes S, Sproat BS et al. Asynthetic, chemically modified ribozyme elimi-
nates amelogenin, the major translation product in developing mouse enamel in vivo. EMBO
J 1995; 14:5224-5229.
68. Snyder DS, Wu Y, Wang JL, Rossi JJ et al. Ribozyme-mediated inhibition of bcr-abl gene
expression in a Philadelphia chromosome-positive cell line. Blood 1993; 82:600-605.
CHAPTER 16
Anti-BCR-ABL Ribozymes
Lance H. Leopold, Scott K. Shore and E. Premkumar Reddy
Introduction
N ormal cell growth is a highly regulated process during which three important families of
genes, oncogenes, growth suppressor genes, and apoptotic genes, converge and cooper-
ate to maintain homeostasis. Of these three families, oncogenes promote cell growth while
growth suppressor genes and apoptotic genes provide negative growth regulation. Carcino-
genesis is a multi-step process, often involving activation of oncogenes and deletion or inac-
tivation of growth suppressor and apoptotic genes. Despite the multi-step nature of trans-
formation, in general, restoring the function of key regulators of these cell processes can
restore normal growth and differentiation in transformed cells. Among the various strate-
gies employed to accomplish this goal, the inhibition of activated oncogenes can be accom-
plished by specifically targeting these sequences with complementary nucleic acid sequences
that inhibit transcription and translation. The use of antisense and ribozyme molecules to
inhibit transforming oncogenes has broad clinical and research applications.
Chronic myelogenous leukemia (CML) is a model disease in which to study the effects
of oncogene activation, as it is characterized by the presence of the t(9,22) translocation.
CML is a clonal myeloproliferative disorder which involves the hematopoietic stem cell1
effecting the myeloid, erythroid, magakaryocytic, B lymphoid and occasionally T lymphoid
blood elements.2 In 1960, Nowell and Hungerford3 discovered a chromosomal abnormality
consistently associated with human CML, a shorter long arm of chromosome 22, and named
this the Philadelphia chromosome (Ph +). More than 95% of cases of CML, as well as 15-25%
of cases of ALL (acute lymphocytic leukemia), harbor the Philadelphia chromosome.4 Mo-
lecular studies have demonstrated that during the formation of the Philadelphia chromo-
some, a portion of the c-abl gene is translocated from chromosome 9q34 to chromosome
22q11, into the bcr gene.5,6 This causes the disruption of two genes and results in the genera-
tion of a new fused gene comprising portions of the bcr and c-abl genes.7,8 The transcript of
this gene either includes (b3a2) or excludes (b2a2) exon 3 of the bcr gene. This chimeric
gene, termed BCR-ABL, produces an abnormal 8.5 kb RNA that encodes a 210 kDa (p210)
fusion protein with increased tyrosine kinase activity compared to the normal Abl protein.9
In addition, the p210BCR-ABL protein is found predominantly in the cytoplasm compared to
the normal Abl protein, which is a nuclear protein. BCR-ABL mediated transformation is
also associated with a number of alterations in cellular signal transduction pathways, which
include increased binding of the Ras regulatory protein Grb-2, a block in apoptosis, and
altered c-Myc regulation.10 This new oncoprotein can transform hematopoietic cells, abro-
gate the growth factor dependence of myeloid cells and block the ability of these cells to
differentiate.11-14 When murine stem cells are infected with a retrovirus encoding the cDNA
for the BCR-ABL gene and inoculated into a mouse, a disease similar to CML results.15
Thus, BCR-ABL protein plays a crucial role in the transformation process in CML, and
disruption of its synthesis is expected to result in the loss of the malignant phenotype. CML
is an ideal disease for therapy targeting the BCR-ABL fusion gene; as the chimeric gene is the
etiologic agent in CML, it is only present in leukemic cells and it occurs in over 95% of CML
cases.
There are no conventional therapies that have resulted in cures in CML.16 Up to 25% of
patients receiving alpha interferon injection therapy may durably suppress the expression
of the Ph-positive clones in CML.14 Combination therapy with alpha interferon and
cytarabine appears to improve the hematologic and cytogenetic responses and the survival
rate in chronic phase CML.17 However, the ability of interferon to cure CML is not proven.18,19
Allogeneic bone marrow transplantation (BMT) using HLA identical siblings, following
high dose myeloablative chemoradiotherapy is curative in up to 85% of carefully selected
patients.20,21 Notably, patients do not need to be rendered Ph-negative to have benefit or
cure of CML.22,23 Unfortunately, less than 30% of CML patients will have a normal alloge-
neic HLA matched donor. Current use of matched unrelated donors has resulted in high
mortality due to graft-versus-host disease (GVHD) and infections.24 Thus, the develop-
ment of an autologous BMT program using Ph-negative stem cells would provide an alter-
native for patients without other curative options. Unpurged autologous BMTs in CML
have not been successful, since Ph-positive stem cells are re-infused into the patient.25
Preclinical and clinical studies in humans have shown preliminary success in CML
employing purged autologous marrow. Pharmacological purging with mafosphamide or 4
hydroperoxycyclophosphamide,26 ex vivo treatment of CML marrow with gamma inter-
feron,27 long term culture of CML marrow,28 removal of CD34 HLA DR positive cells from
stem cells (since they appear to express the Philadelphia chromosome whereas DR negative
CD34 positive cells may be largely Ph-negative),29,30 and recent data showing that stem cells
collected after chemotherapy and/or colony stimulating factor treatment have a reduction
in BCR-ABL positive progenitor cells 31 all have resulted in transient Ph-negative
hematopoeisis following high dose autografting therapy. While all of these methods have
shown modest success, none of these methods are specific. The goal of antisense and
ribozyme-based studies is to demonstrate the ability of these molecules to specifically sup-
press the BCR-ABL transformed phenotype while simultaneously preserving a Ph-negative
stem cell pool suitable for use in autologous bone marrow transplantation.
Current Research
In the past decade, the use of antisense and ribozyme molecules to block the transla-
tion of mRNA has been developed as a strategy to inhibit viral and malignant diseases.
Several groups have reported an inhibition of leukemic cell proliferation by anti-BCR-ABL
antisense oligodeoxyribonucleotides (ODNs);32-48 however, a specific inhibition of BCR-ABL
gene expression has not yet been uniformly demonstrated. Of note, Vaerman et al49 have
questioned the mechanism of ODN growth inhibition in various models and suggest that
the effects of ODNs are not due to an antisense effect but are sequence specific. Examina-
tion of the RNA sequence in the region of the b2a2 and b3a2 BCR-ABL translocations reveal
several ribozyme cleavage sites in close proximity to the fusion sites (Fig. 16.1). Further, the
b3a2 translocation contains a ribozyme cleavage site immediately 5' to the breakpoint re-
gion, suggesting that a ribozyme targeting this cleavage site could be specific. Recently, stud-
ies utilizing ribozymes to specifically target the BCR-ABL oncogene product have demon-
strated in vitro, in vivo, and animal model success.50-54 The characteristics of these first
generation ribozymes are summarized in Table 16.1.
Anti-BCR-ABL Ribozymes
Fig. 16.1. (A) Nucleotide sequence of the b2a2 BCR-ABL fusion gene. (B) Nucleotide sequence of the b3a2 BCR-ABL fusion gene. The breakpoints are
indicated by the asterisks. Potential ribozyme cleavage sites are NUX sequences, where N is any nucleotide and X = U, A or C. Cleavage sites targeted by
ribozymes discussed are in bold. Adjacent cleavage sites result in some of the bold sequences containing four bases.
197
198
Table 16.1. Characteristics of first generation anti-BCR-ABL ribozymes
Reference Ribozyme Length of Efficacy in Cell-Free Specificity Transfection Effects

Target Binding Regions Cleavage Reactions bcr vs. abl Method
Snyder50 b3a2 5': 14 bases 25% (3hr): 69% (12 hr) N/A ; N/A liposomes decrease BCR-ABL mRNA,
fusion site 3': 15 bases protein, cell growth
Shore51 b3a2 5': 9 bases 45% (2 hr) N/A ; N/A retroviral decrease BCR-ABL kinase
fusion site 3': 8 bases activity and cell number
Lange53 b3a2 5': 13 bases + (6 hr) 0 ; N/A liposomes decrease BCR-ABL mRNA
fusion site 3': 14 bases decrease cell proliferation
over antisense controls
5': 22 bases 0/+ (6 hr) 0 ; N/A liposomes decrease BCR-ABL mRNA
3': 22 bases
Wright54 b3a2 5': 12 bases 77% (4 hr) 20%; N/A N/A N/A
fusion site 3': 12 bases
Data provided is for equimolar reactions carried out at 37°C. The time points were chosen to make comparisons meaningful. When no quantification of
cleavage was performed, estimates were made as follows: 0/+ = minimal, + = moderate, and ++ = extensive cleavage.
Anti-BCR-ABL Ribozymes 199
Snyder and coworkers50 reported in 1993 on their experience with a single unit anti-
BCR-ABL ribozyme. Their ribozyme targeted the b3a2 BCR-ABL mRNA sequence and over-
lapped the joining region. The 5' and 3' flanking arms contained 14 and 15 complementary
bases respectively. In cell free experiments, this ribozyme cleaved BCR-ABL mRNA sequences
containing the joining region. After 12 hours, they reported 69% cleavage of target RNA. It
was unclear as to whether this ribozyme cleaved the normal bcr RNA, as no bcr RNA control
was included. In subsequent in vivo experiments they used a DNA-RNA hybrid ribozyme
which resisted nuclease cleavage. In an RNAse protection assay, their anti-BCR-ABL ribozyme
inhibited detectable BCR-ABL mRNA by 49% in EM-2 cells (a human CML blast crisis cell
line) treated with the anti-BCR-ABL ribozyme by lipofection for 48 hours. Antisense se-
quences targeting the same region only inhibited BCR-ABL mRNA by 25% in similar assays.
The co-isolation of ribozyme and antisense sequences with the total RNA subsequently
probed could have inhibited the probe during processing but not represented a true de-
crease in BCR-ABL mRNA. However, in an immunoprecipitation/protein kinase assay, their
anti-BCR-ABL hybrid ribozyme eliminated the p210BCR-ABL signal in EM-2 cells after trans-
fection of the ribozyme by lipofection for 72 hours. Antisense controls reduced but did not
eliminate the p210BCR-ABL signal. In liquid culture, their hybrid ribozyme inhibited cell growth
by 84% in EM-2 cells treated daily with ribozyme for 72 hours. This was marginally better
than cells similarly treated with antisense control oligonucleotides. Cells plated in methyl-
cellulose in the absence of growth factors after 72 hours of ribozyme treatment showed
significant inhibition of colony formation. No antisense control data was provided for the
colony studies. Additional colony studies performed with growth factors may have pro-
vided insight into the effect of ribozyme treatment on cell growth. Similarly, analysis of
colonies growing after ribozyme treatment for BCR-ABL mRNA and protein kinase func-
tion would have been interesting. Despite the absence of certain controls which would have
shed light on the specificity of their ribozyme and the omission of peripheral details as
discussed, their study provided data that an anti-BCR-ABL ribozyme could cleave BCR-ABL
mRNA in vitro and inhibit molecular and physiologic endpoints in cell culture conditions.
Our group had similar results with a single-unit anti-BCR-ABL ribozyme in cell free
conditions.51 Our ribozyme also targeted the joining regions of the b3a2 BCR-ABL mRNA
and contained 5' and 3' complementary sequences of 9 and 8 bases respectively. In in vitro
cleavage experiments, efficient cleavage of target BCR-ABLmRNA was demonstrated. Cleav-
age of bcr and abl RNA sequences was not initially reported. Subsequently, our single unit
ribozyme showed evidence of cleavage activity of the bcr target RNA.52
To test the effect of constitutive expression of this ribozyme, the ribozyme cDNA se-
quence was inserted into several different retroviral expression vectors. Both polymerase II
and III expression systems were tested. K562 cells (a human CML blast cell line) were in-
fected with these retroviruses and subclones analyzed for BCR-ABL mRNA, protein kinase
activity, and biological characteristics. Several subclones infected with a retrovirus express-
ing a tRNA/ribozyme hybrid molecule demonstrated inhibition of p210BCR-ABL kinase activ-
ity. RT-PCR confirmed that the ribozyme was being expressed in these cells. Furthermore,
relative levels of ribozyme expressed by these subclones correlated with BCR-ABL protein
inhibition (data not shown). In addition, K562 cells infected with retrovirus containing the
gene for the tRNA/ribozyme construct demonstrated a 3-fold decrease in cell growth in IL-
3 deficient conditions. This suggests reversal of the growth factor independence which char-
acterizes K562 cells. The effect of this retrovirus on non-BCR-ABL containing cells and the
growth pattern in IL-3-supplemented K562 subclones infected with the tRNA/ribozyme-
containing retrovirus was not reported. While this data provided important evidence that
anti-BCR-ABL ribozymes could be produced constitutively, it was clear that not all cells
infected with a retroviral vector expressed the same levels of ribozyme. Further, high level
expression of the ribozyme is necessary to obtain significant reduction in BCR-ABL tran-

scripts. The lack of consistent BCR-ABL inhibition has led us to consider modifications to
the single-unit ribozyme strategy, including a modified tRNA vector and the use of a multi-
unit ribozyme (discussed below).
Lange and coworkers also published data on the effect of single-unit anti-BCR-ABL
ribozymes targeting the b3a2 joining region as a strategy to affect CML biology.53 They
synthesized anti-BCR-ABL ribozymes with long and short oligonucleotide flanking arms
and initially studied cleavage in a cell free system. Their short ribozyme contained 5' and 3'
complementary regions of 13 and 14 bases respectively, while their long ribozyme con-
tained complementary regions of 22 bases. While both ribozymes cleaved BCR-ABL mRNA,
the degree of cleavage was not quantitated (although it appears to be in the range of similar
single-unit ribozymes). The length of the oligonucleotide flanking arms did not apprecia-
bly affect cleavage efficiency and, despite the inclusion of numerous controls, they did not
include a bcr substrate control. Thus, the specificity of their ribozymes can not be assessed.
In vivo ribozyme transfection studies were performed in K562 cells using liposomes. Both
fluorescence-labeled and [32P]-labeled ribozymes could be detected in K562 cells and re-
vealed ribozyme stability for up to 24 hours. Despite demonstrating ribozyme stability in
their system, they performed six transfections in 72 hours with short or long anti-BCR-ABL
ribozymes. K562 cell proliferation was reduced by 60% and 46% respectively, and only the
short ribozyme improved the inhibition compared to antisense control oligonucleotides. In
addition, the effect of their ribozymes were not assessed in other cell lines to insure the
specificity of the observed effects. Quantitative PCR demonstrated a three to 5-fold reduc-
tion in intracellular BCR-ABL mRNA in cells similarly transfected with active ribozymes.
The explanation for the increased efficiency of the ribozyme in vivo offered by the authors
is that cleavage products are removed intracellularly by nucleases. Any activity of the ribozyme
during the PCR reaction was not excluded (discussed further below). Further, the authors
were unable to demonstrate a reduction in p210 level by Western blot or kinase assays due
to very small numbers of cells used in their system, and no attempt to repeat these tests
using larger quantities of cells was provided. Overall, their data confirms the potential for
using ribozymes to inhibit BCR-ABL mRNA; however, ribozyme specificity (both in vitro
and in vivo) was not explored.
Wright and coworkers published their experience with an anti-BCR-ABL single-unit
ribozyme in cell free conditions.54 Their ribozyme targeted the b3a2 joining region and
contained 5' and 3' flanking arms of 12 bases each. At physiologic temperature, they dem-
onstrated efficient cleavage of BCR-ABL from both plasmid constructs and total RNA from
K562 cells. BCR-ABL RNA sequences from plasmid constructs were cleaved 77% and 91%
by 1X and 10X ribozyme, respectively, in 4 hour cleavage reactions at 37°C. Cellular BCR-ABL
RNA was cleaved 66% in an 8 hour cleavage reaction by ribozymes in similar concentra-
tions. However, they also documented substantial cleavage of bcr RNA in similar cleavage
reactions. No in vivo molecular or physiologic data is provided demonstrating the ability of
their ribozyme to function in cells.
Taken together, these initial studies demonstrated the potential for ribozyme-based
strategies to be useful in BCR-ABL transformed cells. They highlighted in vitro cleavage
activity and raised appropriate concerns for ribozyme specificity. In addition, they demon-
strated evidence for in vivo activity using both molecular and physiologic end points and,
in specific instances, they also demonstrated improved inhibition of BCR-ABL transformed
cells compared to antisense therapy. Despite these findings, anti-BCR-ABL ribozymes did
not fulfill the initial enthusiasm for use in purging programs in CML. Difficulties with sta-
bility in vitro, intracellular degradation, reliable transfection techniques, and specificity
needed to be overcome before clinical testing could proceed. The studies which followed
these initial reports investigated these problems using several strategies. Variations in single-
unit ribozyme construction,55-58 constructing multi-unit ribozymes,52 targeting neighbor-
ing sequences,59 employing novel transfection techniques,52 and improving retroviral pack-
aging60 have all been reported. The characteristics of these modified ribozymes are
summarized in Table 16.2. Whether or not these modifications will result in a therapeutic
window in which to employ a ribozyme-based purging strategy remains to be demonstrated.
Lange and coworkers synthesized four anti-BCR-ABL ribozymes differing in their 5'
and 3' complementary regions and compared their in vitro and in vivo properties.55 Their
ribozymes targeted the b3a2 BCR-ABL joining region and differed from 7 to 13 bases in the
5' flanking region and from 5 to 14 3' bases in the 3' flanking region. They included a non-
functional ribozyme with a mutated catalytic region as an antisense control. Maximal
ribozyme efficiency was 68% in a 16 hour equimolar cleavage reaction and was demon-
strated for a ribozyme with 5' and 3' flanking regions of 9 and 7 bases respectively. Ribozymes
with shorter or longer flanking regions were less efficient. Despite the authors’ stated con-
cern for specificity, no data was provided on bcr cleavage. When their most efficient ribozyme
was transfected into K562 cells by lipofection, they obtained a 77% inhibition of BCR-ABL
kinase activity as measured by autophosphorylation. This was 2-fold better than their
antisense control. In a follow-up study, it was shown that transfections employing lipo-
somes inhibited K562 cell growth approximately 50% by these ribozymes.61
Pachuk and coworkers used two separate strategies to create ribozymes targeting non-
contiguous sequences of BCR-ABL RNA.56 They chose the b2a2 BCR-ABL mRNA in order
to attempt to overcome the lack of a ribozyme cleavage site in the immediate region of the
chimeric mRNA. Their ribozymes contained anchor sequences complementary to bcr exon
2 and a hammerhead ribozyme targeting a more distant triplet in abl exon 2. The only
sequences that should be cleaved would contain both bcr and abl sequences. Specifically, the
anchor sequences are complementary to bases in exon 2 of bcr just 5' to the joining region
and are 31, 21, or 11 bases in length. The anchor sequences are connected to a 13 base spacer
sequence, which has no complementarity to either bcr or abl RNA, and are connected to a
ribozyme targeting a GUA site in abl exon 2. The 5' flanking region contained 9 bases and
the 3' flanking region 6 bases complementary to abl exon 2. In cell free cleavage reactions,
the ribozyme with the shortest anchor sequence was the most efficient. Of note, there was
no cleavage of abl or bcr RNA by the ribozymes containing anchor sequences, and in gel
shift experiments, ribozymes containing anchor sequences failed to bind to abl or bcr RNA,
indicating that the specificity was due to lack of binding efficiency. These difficulties could
be overcome by first denaturing the RNA and then renaturing in the presence of ribozymes,
suggesting that inefficient binding was due to target RNA secondary structure. The optimal
length of the anchoring sequence was not determined and may differ for each target RNA
sequence, and the importance of the spacer region remains speculative.
Using a second strategy, this group constructed ribozymes which targeted the b2a2
chimeric RNA in the immediate region of the joining region by modifying the flanking
arms to target a CUU sequence. The 5' and 3' flanking sequences contained 7 and 12 bases
respectively, and the 3' region spanned the joining region and contained sequences
complementary to both bcr and abl exons 2. In addition, a series of ribozymes with mis-
matches in the 3' abl sequences were constructed. Ribozymes with greater than 2 mismatches
were unable to cleave target b2a2 BCR-ABL RNA. Notably, ribozymes with no or 2 mis-
matches were able to cleave both b3a2 and b2a2 BCR-ABL RNA. These ribozymes similarly
showed no cleavage of bcr RNA sequences. Taken together, this data represents novel strat-
egies for targeting chimeric RNA. It is noteworthy that these strategies greatly improved the
specificity of these ribozymes for cleavage of the chimeric RNA. In addition, the gel shift
assays and modified in vitro cleavage reactions after denaturing and renaturing the target
202
Table 16.2. Characteristics of modified anti-bcr-abl ribozymes
Reference Ribozyme Length of Binding Additional Efficacy in Cell-Free Specificity

Target Regions (Bases) Modifications Cleavage Reactions bcr vs. abl
Lange55,56 b3a2 fusion 5' : 7 ; 3' : 9 none 68% (16 hr) N/A : N/A
b3a2 fusion 5' : 7 ; 3' : 5 none 30% (16 hr) N/A : N/A
b3a2 fusion 5' : 8 ; 3' : 10 none 51% (16 hr) N/A : N/A
b3a2 fusion 5' : 13; 3' : 14 none 62% (16 hr) N/A : N/A
Pachuk56 b2a2 abl exon 2 5' : 9 ; 3' : 31 3' arm anchors to bcr 0/+ 0/+ : 0/+
b2a2 abl exon 2 5' : 9 ; 3' : 21 3' arm anchors to bcr + 0/+ : 0/+
b2a2 abl exon 2 5' : 9 ; 3' : 11 3' arm anchors to bcr ++ 0/+ : 0/+
Kearney57 b3a2 fusion 5' : 12; 3' : 12 no mismatches 77% (4 hr) 20% : N/A
b3a2 fusion 5' : 12; 3' : 12 mismatch 1st bcr base 0% (4 hr) 0% : N/A
b3a2 fusion 5' : 12; 3' : 12 mismatch 2nd bcr base 86% (4 hr) 4% : N/A
b3a2 fusion 5' : 12; 3' : 12 mismatch 3rd bcr base 85% (4 hr) 13% : N/A
Kronenwett58 b3a2 fusion 5' : 52; 3' : 3 none + 0/+ : 0/+
b3a2 abl exon 2 5' : 3 ; 3' : 69 none + 0/+ : 0/+
Leopold52 b3a2; three sites 63 bases total restriction sites 95% (2 hr) 45%: 87%
mismatch
James59 b3a2 abl exon 2 5' : 10; 3' : 20 none 11% (2 hr) 0%: 0%
b3a2 abl exon 2 5' : 4 ; 3' : 20 none 24% (2 hr) 0%: 0%
b3a2 fusion 5' :10 ; 3' : 10 none 6% (2 hr) 0%: 0%
b2a2 abl exon 2 5' : 10; 3' : 20 none 21% (2 hr) 0%: 7%
Data provided is for equimolar reactions carried out at 37°C. The time points were chosen to make comparisons meaningful. When no quantification of
cleavage was performed, estimates were made as follows: 0/+ = minimal, + = moderate, and ++ = extensive cleavage.
RNA in the presence of ribozymes provided insight into the importance of RNA secondary
structure in ribozyme binding. No data was presented regarding the in vivo efficacy of these
ribozymes.
Using a similar approach, Kearney and coworkers synthesized four modified single-
unit anti-BCR-ABL ribozymes and assessed their specificity and efficiency.57 Their ribozymes
targeted the immediate region of the b3a2 chimeric RNA and their initial ribozyme con-
tained 5' and 3' flanking sequences of 12 bases each. Three modified ribozymes were syn-
thesized containing single base mismatches in the 3 bcr bases 5' of the joining region. A
fourth ribozyme targeted the b3a2 chimeric RNA, one base 3' to the joining region. The
ribozyme containing a mismatch in the second base 3' to the cleavage site maintained its
efficiency for cleaving BCR-ABL RNA but had greatly reduced catalytic activity for bcr tar-
get RNA. The ribozyme targeting the alternative cleavage site was also specific for BCR-ABL
though it was less efficient. While this data demonstrates a novel approach to improve the
specificity of a ribozyme targeting a chimeric RNA, no in vivo data was presented for these
ribozymes.
An extensive kinetic study of antisense and ribozyme sequences targeting BCR-ABL
sequences was recently published.58 Ribozymes were studied which contained long (>50
bases) 5' or 3' flanking oligonucleotides, although the length of complementarity varied
from 20 to 80 bases. One ribozyme, targeting a cleavage site in the immediate region of the
b3a2 BCR-ABL junction and containing a 52 base 5' flanking arm, demonstrated efficient
cleavage of BCR-ABL and limited cleavage of bcr or abl RNA. Notably, kinetic probing of the
local folding potential of the BCR-ABL fusion region indicates that this region is not easily
accessible for complementary DNA and RNA sequences. The authors postulate that this
observation explains the lack of antisense-mediated BCR-ABL gene inhibition observed in
human cells. This observation has been predicted by RNA secondary structure computer
programs and demonstrated in other ribozyme systems.62,63 The authors stress the impor-
tance of targeting regions of RNA that are accessible to nucleic acids.
Our group has approached the problems of efficacy, specificity, and ribozyme delivery
by constructing multi-unit ribozymes and developing novel receptor-mediated transfec-
tion vehicles.52 Because examination of the region of the b3a2 joining region reveals several
GUX sequences in close proximity, we constructed single, double, and triple-unit ribozymes
targeting these sequences. The intervening sequences were composed of complementary
bases to the target BCR-ABL RNA with the exception of the restriction enzyme sites used for
directional cloning. In in vitro cleavage reactions, the triple-unit ribozyme greatly improved
the cleavage efficiency compared to single and double-unit ribozymes (Fig. 16.2). In fact,
this ribozyme appeared to have the best cleavage efficiency of any published sequence. Be-
cause this ribozyme also targets cleavage sites in normal abl and bcr genes, it was not sur-
prising to observe cleavage of these RNAs as well. However, the greatly improved cleavage of
BCR-ABL may provide an opportunity to block BCR-ABL without significantly effecting
normal gene function. Preliminary work employing liposomes to transfect ribozymes into
K562 cells resulted in minimal evidence of cleavage of BCR-ABL transcripts, with no effect
on transformed cell growth, growth-factor independence or block in differentiation.
Because of concerns about the multiple genetic abnormalities in K562 cells, we devel-
oped a simpler model of BCR-ABL transformation. By infecting 32D cells (a murine myelo-
blast cell line) with a retrovirus encoding the cDNA for BCR-ABL, we created a BCR-ABL
transformed cell model containing no other genetic abnormalities. In addition, because of
growing concerns about the ability of liposomes to deliver nucleic acids to the proper sub-
cellular compartments in reasonable quantities, new transfection vehicles were investigated.
By covalently linking a polylysine chain to folic acid, we created a heterobifunctional re-
agent capable of binding nucleic acids and cellular receptors.
Fig. 16.2. Autoradiogram of ribozyme mediated cleavage of BCR-ABL mRNA. [32P]-labeled BCR-
ABL mRNA was synthesized from a plasmid vector containing a 499 bp segment of the b3a2
BCR-ABL chimeric gene. Substrate and ribozyme reactions were carried out in 10 µl volumes. All
RNA concentrations were 100 nM. Transcribed RNAs were resuspended in 50 mM Tris-HCl, pH
7.5, containing 1 mM EDTA and heated to 95°C for 5 min, and immediately chilled on ice. Reac-
tions were initiated by adding 1 µl of 200 mM MgCl2 and stopped after 2 hours by adding 2 µl of
stop solution containing 95 % formamide, 20 mM EDTA, and 2% bromophenol blue. Cleavage
products were separated by electrophoresis on a 6 % denaturing gel. Lane S is a control and
contains substrate without ribozymes. Lanes A, B, C, and D show cleavage products from reac-
tions containing single-unit ribozymes targeting bcr exon 3, the BCR-ABL junction, bcr exon 3,
and abl exon 2 respectively. Ribozymes A and C differ in the length of their annealing arms.
Lanes E and F show cleavage products from reactions containing double-unit ribozymes target-
ing the BCR-ABL junction and either bcr exon 3 or abl exon 2, respectively. Lane G shows cleav-
age products from a reaction containing a triple-unit ribozyme targeting all three sites. Cleavage
with double and triple-unit ribozymes releases small fragments which run at the bottom of the
gel and are not shown. P1 through P6 are 314, 286, 263, 235, 212, and 184 base fragments respec-
tively. Modified from Leopold LH et al, Blood, 1995, 85: 2162-2170.
Fig. 16.3. Autoradiogram of a Southern blot of RT-PCR amplified BCR-ABL

mRNA from transformed 32D cells transfected with ribozymes via liposomes
or folic acid-polylysine vectors. From 1 to 1 x 104 BCR-ABL transformed 32D
cells were added to 1 x 106 untransformed 32D cells. Cells were transfected with
ribozymes or vectors containing no ribozymes (controls) and after 24 hours
total cellular RNA was extracted. RT-PCR was performed with primers that
amplified the BCR-ABL chimeric gene and the β-actin gene (an internal con-
trol). Southern blotting was performed with a [32P]-kinased probe that detects
the BCR-ABL breakpoint. (+) PCR control: RNA transcribed from a plasmid
containing the BCR-ABL breakpoint region. (–) PCR control: no template addi-
tion. Modified from Leopold LH et al, Blood, 1995, 85: 2162-2170.
Transfection of the triple-unit ribozyme by folate-polylysine into BCR-ABL transformed

32D cells resulted in a 3 log reduction of BCR-ABL mRNA measured by RT-PCR and South-
ern blotting (Fig. 16.3). There was no evidence of ribozyme activity in the PCR amplifica-
tion step when ribozymes were deliberately added to cellular RNA prior to analysis. In addi-
tion, BCR-ABL transformed 32D cells demonstrated growth inhibition in growth factor
deficient medium when transfected daily with ribozymes by folate-polylysine uptake.64 There
was no growth inhibition seen in these cells by antisense control sequences, transfection
vehicle alone, or when growth factors were provided. This model does not allow appropri-
ate testing of the ribozyme for specificity, as murine bcr and abl sequences are not targeted
by this ribozyme. Preliminary work employing clonogenic assays and normal human stem
cells has failed to show any added toxicity due to the triple-unit ribozyme over that seen by
transfection vehicles alone. We are continuing to investigate improved expression systems
as has been recently reported.60 These observations will be further tested using a SCID mouse
model of CML.
James and coworkers have constructed ribozymes targeting the b3a2 and b2a2 BCR-
ABL translocations with modified complementary flanking oligonucleotides.59 They ob-
served, in in vitro cleavage reactions, that ribozymes targeting 9 bases 3' and 3 bases 5' to the
joining region of the b3a2 BCR-ABL sequence were specific at physiologic temperature. The
ribozyme targeting 9 bases 3' of the junction had a 3' flanking sequence containing 20 bases
and a 5' sequence with 10 bases. The ribozyme targeting the immediate joining region had
5' and 3' flanking arms of 10 bases each. By reducing the length of the 5' flanking region of
the ribozyme targeting 9 bases 3' of the joining region to 4 bases, efficacy improved from
11% to 24% cleavage without cleavage of normal abl sequences. They also observed that
similar ribozymes with short noncomplementary bases on the ends of otherwise identical
ribozymes (resulting from the cloning strategy used) slightly improved efficacy of cleavage.
In subsequent reports,65,66 they have shown that transfection of this modified ribozyme
into BCR-ABL transformed 32D cells failed to affect BCR-ABL mRNA levels by RT-PCR or
RNAse protection assay. Despite these observations, cell growth was inhibited in a dose-
related manner and survival of SCID mice transplanted with ribozyme-treated 32D BCR-ABL
cells had prolonged survival compared to control treated mice.
Other recent data demonstrates the potential effectiveness of constitutive expression
of ribozymes targeting BCR-ABL mRNA. Liu and coworkers recently reported the results of
transfection studies employing retroviral expression systems for anti-BCR-ABL ribozymes
in K562 cells.67 They developed ribozymes which target the immediate joining region and
the bcr translation initiation site, and expressed these ribozymes in pLXSN vector contain-
ing the SV promoter connected to a rabbit hemoglobin intron. They report high infection
efficiency of K562 cells and inhibition of K562 cell growth in culture, soft agar, and a SCID
mouse model by both ribozyme-expressing retroviruses.
Snyder and coworkers have reported on the effectiveness of an anti-BCR-ABL hairpin
ribozyme which targets the p190 mRNA found in acute lymphoblastic leukemia patients
(ALL).68 Despite lack of ribozyme specificity data, transfection studies employing liposomes
in p190-expressing cells demonstrated decreased levels of p190 protein and growth inhibition.
Our group has also recently reported preliminary results with tRNA/ribozyme hybrid
molecules expressed from the retroviral vector DCt5T' (Fig. 16.4).69 When the triple-unit
anti-BCR-ABL ribozyme is cloned into the 3' tRNA tail and expressed in in vitro systems
fully functional ribozymes are produced. Electroporation of this vector into K562 cells re-
sulted in the isolation of several subclones expressing ribozymes and containing reductions
in BCR-ABL transcripts. Experiments are currently in progress to study the effects of these
vectors on p210Bcr-abl protein levels and growth factor independent growth of transfected
cells.
In summary, these initial experiences with anti-BCR-ABL ribozymes have demonstrated
the challenges that are faced by researchers applying antisense and ribozyme strategies to
inhibit oncogenes. Despite initial success in in vitro cleavage reactions, the issues of efficacy
and specificity remain key problem areas in ribozyme development. Delivering ribozymes
to cells and subcellular locations are critical issues being explored to improve the efficacy of
ribozyme-based therapies. Delivery methods employing ribozymes as drugs and first gen-
eration models employing constitutive expression strategies have not demonstrated reliable
inhibition of BCR-ABL transformed cells, whether assayed by molecular or biologic end
points. Whether or not modifications to these approaches will improve on current results
will determine the role that antisense and ribozyme strategies play in treating malignant
diseases. These issues are discussed further below.
Future Directions
As our knowledge of RNA trafficking improves, additional strategies for ribozyme de-
sign and delivery will be developed. The problem of improving ribozyme efficacy, specific-
ity, cellular delivery, co-localization with target RNA, and efficient expression are critical
areas of ongoing research. Strategies for improving these properties of anti-BCR-ABL
ribozymes are discussed.
Modifications in ribozyme design continue to be relevant issues to improve efficacy
and specificity. Although targeting the joining region has been the most common strategy
for ribozyme design, this has not guaranteed specificity. Considering the relative inaccessi-
bility of this site due to RNA secondary structure,70 targeting other cleavage sites remains an
attractive option for improving efficacy. The use of RNA-folding programs can assist in the
Fig. 16.4. RT-PCR analysis

from K562 cells infected
with retroviral constructs
encoding triple-unit ribo-
zymes. K562 cells were elec-
troporated with retroviruses
encoding tRNA/ribozyme
hybrid molecules from the
DC t5T' vector. Cells were
plated and grown in the pres-
ence of neomycin to select
for clones containing DNA
coding for ribozymes. hrIL-3
(5 pmol/µl) was provided in
the media to prevent nega-
tive selection, as cells with
low levels of BCR-ABL ty-
rosine kinase may become
growth factor dependent.
Single cells were isolated, ex-
panded and screened for the
production of ribozymes by
RT-PCR employing primers
antisense to the multi-unit
ribozyme. Southern blots
were performed with [32P]-
labeled oligonucleotides spe-
cific for ribozymes, BCR-
ABL, and actin DNA. Panel
A shows the results of PCR
analysis for the presence of
the triple-unit ribozyme in
cells electroporated with the
DCt5T' retroviral construct.
Lanes 1, 2, 3, and 8 demon-
strate the presence of ribo-
zymes in these clones. K562
parental cells contain no
ribozymes. The positive plas-
mid control contains ribo-
zyme bands with prolonged
exposure. In Panel B, BCR-
ABL PCR analysis is shown.
Clones 1 and 8 inhibited BCR-ABL mRNA production. K562 cells and plasmid controls also
demonstrated the presence of BCR-ABL mRNA sequences. Panel C shows the results of β-actin
gene amplification and is presented to show that similar quantities of mRNA were used in RT
and PCR reactions.
identification of accessible cleavage sites. Further, employing anchoring sequences, varying

the length of the oligonucleotide flanking arms, targeting multiple cleavage sites, and in-
cluding extra or mismatched sequences have had unpredictable effects on ribozyme efficacy
and specificity.71 These modifications have largely been a process of trial and error, and
combinations of these strategies may optimize efficacy and maintain specificity.
Other aspects of ribozyme design impact on ribozyme efficacy and specificity. Muta-
tions in the target RNA may be a strategy which malignant cells employ to develop resis-
tance to ribozyme-based therapy. Analogous to the application of combination therapies
for various malignant and viral diseases, targeting several sites for ribozyme cleavage may
be necessary to prevent the emergence of resistant clones. Further, the role of intracellular
RNA binding proteins is beginning to be explored. The heterogeneous nuclear-ribonuclear
protein hnRNP A1 is associated with RNAs in the nucleus and cytoplasm, and in in vitro
cleavage reactions facilitate ribozyme cleavage by enhancing product release from ribozyme
flanking oligonucleotides. 72,73 This enhancement may be limited by the degree of
complementarity between ribozyme and target RNA and may be unavailable for ribozymes
employing mismatching or anchoring strategies. Empirical testing is required to answer
these questions.
Ribozymes can be delivered to cells as therapeutic agents or as genes encoding these
sequences. Further, gene therapy can be accomplished through both viral and non-viral
methods. There are various physico-chemical methods to facilitate RNA uptake into large
amounts of cells including liposomes and cationic lipids, and heterobifunctional reagents
binding to cell surface molecules and nucleic acids. Liposomal strategies are limited by be-
ing nonspecific, only transiently affecting target RNA levels, and delivering the majority of
their contents to endosomes where RNA degradation may occur.74 As there appears to be a
requirement for a large excess of ribozymes in in vivo applications, liposomal strategies may
be limited.75 Receptor-mediated uptake and employing heterobifunctional reagents can
improve the intracellular effectiveness of ribozymes by improving intracellular trafficking
and half life.52 The specificity of such reagents can be improved by employing antibody-
mediated uptake. Such a vector would contain an antibody (which binds to cells) attached
to liposomes or polylysine (which binds to RNA).76 The CD34 antigen would be an ideal
target for such a vector since the transformed stem cell in CML is CD34 positive. In addi-
tion, an antibody targeting the HLA-DR antigen could be upregulated by interferon and
may allow increased ribozyme uptake. These strategies may target the malignant clone which
is HLA-DR positive, while sparing the normal stem cell compartment which is HLA-DR
negative. Further, targeting multiple receptors may optimize ribozyme delivery.
Ribozyme gene therapy can employ both viral and non-viral methods. Gene therapy
has the potential to improve ribozyme co-localization with target RNA as the ribozyme is
produced by the same cellular transcription machinery. However, concerns about releasing
competent viruses, transcription silencing, low infection/transduction rates, disrupting
normal cellular genes during viral integration, and the need for the target cell to divide to
allow viral integration are some of the concerns facing the development of retroviral gene-
based therapies. Retroviral vectors employ RNA polymerase II or III based systems with
constitutive, tissue-specific, and inducible promoters. Packaging strategies must consider
sequences for termination, capping, RNA processing, and RNA transport to subcellular com-
partments. Including bacterial operator-repressor systems or tissue-specific promoters may
add expression control. In addition, variations in the viral expression system may reduce
competition between inserted promoters and viral long terminal repeats (LTRs).
Alternative viral vectors and non-viral techniques have also been developed to over-
come several of the problems associated with retroviral systems. Adeno-associated virus has
a higher transduction efficiency than traditional retroviral vectors, and its integration can
be directed. Adenovirus systems can deliver genes to cells for transient expression, as inte-
gration is not necessary for gene activation. In addition, gene guns, liposomes, and
heterobifunctional reagents can all deliver genes of various sizes to cells. While these non-
viral methods generally have higher transduction efficiency and avoid the issue of release of
competent retroviruses, random insertion, transcription repression, and toxicity remain
ongoing concerns.
Despite improvements in viral and non-viral expression systems, comparatively little is
known about intracellular RNA transport. mRNA has a heterogeneous distribution in cells
and trafficking may be accomplished by sequences in the 3' untranslated region of mRNAs.
Including these localization signals in ribozyme expression systems may improve the co-
localization of ribozymes with target RNA. In addition, mRNA processing involves small
nuclear RNA in the cell nucleus. These sequences could be included in ribozyme transcrip-
tion units to improve nuclear ribozyme and target interactions.
CML remains a model disease in which to consider using ribozyme-based therapy to
inhibit the disease-causing oncogene. Despite advances, only allogeneic BMT is curative in
CML. As the properties that affect ribozyme efficacy and specificity become better defined
and the intracellular signals that govern RNA trafficking become understood, the improve-
ments in gene therapy and RNA delivery strategies can be applied to helping patients over-
come this otherwise fatal disease.
References
1. Golde DW, Chaplin RE. Chronic myelogenous leukemia: Recent advances. Blood 1985;
65:1039.
2. Fialkow PJ, Jacobson RJ, Papayannopoulou T. Chronic myelocytic leukemia. Clonal origin
in a stem cell common to the granulocyte, erythrocyte, platelet and monocyte/macroph-
age. Am J Med 1977; 63:125.
3. Nowell PC, Hungerford DA. A minute chromosome in human chronic granulocytic leuke-
mia. Science 1960; 132:1497-1499.
4. Kurzrock R, Gutterman JU, Talpaz M. The molecular genetics of Philadelphia chromo-
some-positive leukemias. New Engl J Med 1988; 319:990-998.
5. Rowley JD. A new consistent chromosomal abnormality in chronic myelogenous leukemia
identified by quinacrine flourescence and Giemsa staining. Nature 1973; 243:290-293.
6. deKlein A, Geurts van Kessel A, Grosveld G et al. A cellular oncogene is translocated to
the Philadelphia chromosome in chronic myelogenous leukemia. Nature 1982; 300:765-767.
7. Heisterkamp N, Stephenson JR, Geoffen J et al. Localization of the c-abl oncogene adja-
cent to a translocation breakpoint in chronic myeocytic leukemia. Nature 1983; 306:239-42.
8. Schtivelman E, Lifshitz B, Gale RP et al. Fused transcript of abl and bcr genes in chronic
myelogenous leukemia. Nature 1985; 315:550-553.
9. Konopka JB, Watanabe SM, Witte ON. An alteration of the human c-abl protein in K562
cells unmasks associated tyrosine kinase activity. Cell 1984; 37:1035-1042.
10. Kantarjian HM, Deisseroth A, Kurzrock R et al. Chronic myelogenous leukemia: A concise
update. Blood 1993; 82,3:691-703.
11. McLaughlin J, Chianese E, Witte ON. In vitro transformation of immature hematopoietic
cells by the P210 BCR-ABL oncogene product of the Philadelphia chromosome. Proc Natl
Acad Sci USA 1987; 84:6558.
12. Daley GQ and Baltimore D. Transformation of an interleukin 3-dependent hematopoietic
cell line by the chronic myelogenous leukemia-specific p210bcr-abl protein. Proc. Natl. Acad.
Sci. USA 1988; 85:9312-9316.
13. Cannistra S. Chronic myelogenous leukemia as a model for genetic basis for cancer. Hematol
Oncol Clin North Am 1990; 4:337.
14. Kantarjian HM, Deisseroth A, Kurzrock R, Estrov Z, and Talpaz M. Chronic myelogenous
leukemia: A concise update. Blood 1993; 82:691-703.
15. Daley GQ, Van Etten RA, Baltimore D. Induction of chronic myelogenous leukemia in
mice by the p210bcr-abl gene of the Philadelphia chromosome. Science 1990; 247:824-830.
16. Lichtman MA. Chronic myelogenous leukemia and related disorders. In: Williams WJ et
al, eds. Hematology. 4th ed. New York: McGraw-Hill, 1990:205-212.
17. Guilhot F, Chastang C, Michallet M et al. Interferon alpha-2b combined with cytarabine
versus interferon alone in chronic myelogenous leukemia. N Engl J Med 1997; 337:223-9.
18. Talpaz M, Kantarjian H, McCredie K et al. Hematologic remission and cytogenetic im-
provement induced by recombinant human interferon alpha in chronic myelogenous leu-
kemia. N Eng J Med 1986; 314:1065-1069.
19. The Italian Cooperative Study Group on Chronic Myeloid Leukemia. Interferon alfa-2a as
compared with conventional chemotherapy for the treatment of chronic myeloid leuke-
mia. N Eng J Med 1994; 330:820-25.
20. Butturini A, Keating A, Goldman J et al. Autotransplants in chronic myelogenous leuke-
mia: Strategies and results. The Lancet 1990; 335:1255-1258.
21. Gale RP, Butturini A. How do transplants cure chronic myelogenous leukemia? Bone Mar-
row Transplantation 1992; 9:83-85.
22. Pichert G, Alyea EP, Soiffer RJ et al. Persistence of myeliod progenitor cells expressing
BCR-ABL mRNA after allogeneic bone marrow transplantation for chronic myelogenous
leukemia. Blood 1994; 84,7:2109-2114.
23. Miyamura K, Tahara T et al. Long persistent bcr-abl positive transcript detected by poly-
merase chain reaction after marrow transplant for chronic myelogenous leukemia without
clinical relapse: A study of 64 patients. Blood 1993; 81,4:1089-1093.
24. Marks DI, Cullis JO, Ward KN et al. Allogeneic bone marrow transplantation for chronic
myeloid leukemia using sibling and volunteer unrelated donors. Ann Int Med 1993;
119:207-214.
25. Reiffers J, Trouette R, Marit G et al. Autologous blood stem cell transplantation for chronic
granulocytic leukemia transformation: A report of 47 cases. Brit J Haematology 1991;
77:339-345.
26. Marcus RE, Goldman JM. Autografting in chronic granulocytic leukemia. Clinics in
Haematology 1986; 15:235-247.
27. McGlave PB, Arthur D, Miller WJ et al. Autologous transplantation for CML using mar-
row treated ex vivo with recombinant human interferon gamma. Bone Marrow Transplan-
tation 1990; 6:115-120.
28. Barnett MJ, Eaves CJ, Phillips GL et al. Successful autografting in chronic myeloid leuke-
mia after maitenance of marrow in culture. Bone Marrow Transplantation 1989; 4:345-351.
29. Berenson R. Transplantation of stem cells enriched by immunoadsorption. Third Interna-
tional Symposium on Bone Marrow Purging and Processing, San Diego, Oct 4, 1991 (Ab-
stract).
30. Verfaillie CM, Miller WJ, Boylan K et al. Selection of benign primitive hematopoietic pro-
genitors in chronic myelogenous leukemia on the basis of HLA-DR antigen expression.
Blood 1992; 79:1003-1010.
31. O’Brien SG, Goldman JM. Current approaches to hematopoietic stem-cell purging in
chronic myeloid leukemia. Journal of Clinical Oncology 1995; 13,3:541-546.
32. Taj AS, Martiat P, Dhut S et al. Inhibition of P210BCR-ABL expression in K562 cells by
electroporation with an antisense oligonucleotide. Leuk Lymphoma 1990; 3:201.
33. Skorski T, Szczylik C, Malaguarnera L et al. Gene-targeted specific inhibition of chronic
myeloid leukemia cell growth by BCR-ABL antisense oligodeoxynucleotides. Folia Histochem
Cytobiol 1991; 29:85.
34. Mahon FX, Belloc F, Reiffers J. Antisense oligomers in chronic myeloid leukaemia. Lancet
1993; 341:566.
35. Kirkland MA, O’Brien SG, McDonald C et al. BCR-ABL antisense purging in chronic my-
eloid leukaemia. Lancet 1993; 342:614.
36. Maekawa T, Murakami A, Kimura S et al. Clonal suppression of human leukemia cell
growth by antisense oligodeoxynucleotide phosphorothioates. J Cell Biochem 1993; 197
(suppl 17E).
37. Skorski T, Nieborowska-Skorska M, Barletta C et al. Highly efficient elimination of Phila-

delphia leukemic cells by exposure to bcr-abl antisense oligodeoxynucleotides combined
with mafosfamide. J Clin Invest 1993; 92:194.
38. Vaerman JL, Lewalle P, Martiat P. Antisense inhibition of p210 bcr-abl in chronic myeloid
leukemia. Stem Cells 1993; 11:89 (suppl 2).
39. DeFabritiis P, Amadori S, Calabretta B et al. Elimination of clonogenic Philadelphia-posi-
tive cells using BCR-ABL antisense olgodeoxynucleotides. Bone Marrow Transplant 1993;
12:261.
40. Smetser TFCM, Skorski T, Van de Locht LTF et al. Antisense BCR-ABL oligonucleotides
induce apoptosis in the Philadelphia chromosome positive cell line BV173. Leukemia 1994;
1:129.
41. Skorski T, Nieborowska-Skorska M, Nicolaides NC et al. Suppression of Philadelphia leu-
kemia cell growth in mice by bcr-abl antisense oligodeoxynucleotides. Proc Natl Acad Sci
USA 1994; 91:4504.
42. Okabe M, Kunieda Y, Miyagishima T et al. BCR-ABL oncoprotein-targeted antitumor ac-
tivity of antisense oligonucleotides complementary to BCR-ABL mRNA and herbimycin A,
an antagonist of protein tyrosine kinase: Inhibitory effects on in vitro growth of Phl-posi-
tive leukemia cells and BCR-ABL oncoprotein-associated transformative cells. Leuk Lym-
phoma 1993; 10:307.
43. Bedi A, Zehnbauer BA, Barber JP et al. Inhibition of apoptosis by BCR-ABL in chronic
myeloid leukemia. Blood 1994; 83:2038.
44. Thomas M, Kosciolek B, Wang N et al. Capping of bcr-abl antisense oligonucleotides en-
hances antiproliferative activity against chronic myeloid leukemia cell lines. Leuk Res 1994;
18:401.
45. McGahon A, Bissonnette R, Schmitt M et al. BCR-ABL maintains resistance of chronic
myelogenous leukemia cells to apoptotic cell death. Blood 1994; 83:1179.
46. Tari AM, Tucker SD, Deisseroth A et al. Liposomal delivery of methylphosphonate antisense
oligonucleotides in chronic myelogenous leukemia. Blood 1995; 84:601.
47. O’Brien SG, Kirkland MA, Melo JV et al. Antisense BCR-ABL oligomers cause non-spe-
cific inhibition of chronic myeloid leukemia cell lines. Leukemia 1994; 8:2156.
48. Smetsers TFCM, Van de Locht LTF, Pennings AHM et al. Phosphorothioate BCR-ABL
antisense oligonucleotides induce cell death, but fail to reduce cellular bcr-abl protein lev-
els. Leukemia 1995; 9:118.
49. Vaerman JL, Lammineur C, Moureau P et al. BCR-ABL antisense oligodeoxyribonucleotides
suppress the growth of leukemic and normal hematopoietic cells by a sequence-specific
but nonantisense mechanism. Blood 1995; 86:3891-3896.
50. Snyder DS, Wu Y, Wang JL et al. Ribozyme-mediated inhibition of bcr-abl gene expression
in a Philadelphia chromosome-positive cell line. Blood 1993; 82:600-605.
51. Shore SK, Nabissa PM, Reddy EP. Ribozyme-mediated cleavage of the BCRABL oncogene
transcript: in vitro cleavage and in vivo loss of P210 protein-kinase activity. Oncogene
1993; 8:3183-3188.
52. Leopold LH, Shore SK, Newkirk TA et al. Multi-unit ribozyme-mediated cleavage of bcr-
53. Lange W, Cantlin EM, Finke J et al. In vitro and in vivo effects of synthetic ribozymes
targeted against BCR-ABL mRNA. Leukemia 1993; 7:1786-1794.
54. Wright L, Wilson SB, Milliken S et al. Ribozyme-mediated cleavage of the bcr-abl tran-
script exprressed in chronic myelogenous leukemia. Exp Hematology 1993; 21:1714-8.
55. Lange W, Daskalakis M, Finke J et al. Comparison of different ribozymes for efficient and
specific cleavage of BCR-ABL related mRNAs. FEBS Letters 1994; 338:175-178.
56. Pachuk CJ, Yoon K, Moellling K et al. Selective cleavage of bcr-abl chimeric RNAs by a
ribozyme targeted to non-contiguous sequences. Nucleic Acids Research 1994; 22:301-307.
57. Kearney P, Wright LA, Milliken S et al. Improved specificity of ribozyme-mediated cleav-
age of bcr-abl mRNA. Experimental Hematology 1995; 23:986-989.
58. Kronenwett R, Haas R, Sckakiel G. Kinetic selectivity of complementary nucleic acids: bcr-
abl-directed antisense RNA and ribozymes. J Mol Biol 1996; 259:632-644.
59. James H, Mills K, Gibson I. Investigating and improving the specificity of ribozymes di-
rected against the bcr-abl translocation. Leukemia 1996; 10:1054-1064.
ribozymes expressed from a tRNA-based RNA polymerase III promotor. Nucleic Acids
Research 1995; 23:2259-2268.
61. Lange W. Cleavage of BCR-ABL mRNA by synthetic ribozymes—effects on the prolifera-
tion rate of K562 cells. Klin Padiatr 1995; 207:222-224.
ribozyme. Gene 1992; 113:157.
63. Fedor JJ, Uhlenbeck OC. Substrate sequence affects “hammerhead” RNA catalytic efficiency.
Proc Natl Acad Sci USA 1990; 87:1668.
64. Leopold LH, Shore SK, Reddy EP. Multi-unit anti-BCR-ABL ribozyme therapy in chronic
myelogenous leukemia. Leukemia and Lymphoma 1995; 18:179-184.
65. James HA, Twomey CM, Mills KI et al. Specificity of ribozymes against the bcr-abl mRNAs
in vitro. Biochemical Society Transactions 1996; 24:409S.
66. Mills KI, Walsh V, Gilkes AF et al. In vitro ribozyme treatment of 32D cells expressing a
bcr-abl construct prolongs the survival of SCID mice. Proceedings of Amer Soc Hematolgy
1996; 2296, Abstract.
67. Liu JH, Cheng SC, Chu CJ et al. Retrovirally transduced ribozymes suppress the growth of
chronic myelogenous leukemic cell line. Proceedings of Amer. Society of Hematology 1996;
816, Abstract.
68. Snyder DS, Wu Y, McMahon R et al. Ribozyme-mediated inhibition of a Philadelphia-
chromosome positive acute lymphoblastic leukemia cell line expressing the p1‘90 bcr-abl
oncogene. Proceedings of the American Society of Hematology 1996; 821, Abstract.
69. Leopold LH, Shore SK, Krishnaraju J et al. Retroviral expression of a triple-unit anti-bcr-
abl ribozyme inhibits bcr-abl RNA expression in K562 cells. Proceedings American Society
of Oncology 1997; 1956, Abstract.
70. Mahone FX, Ripoche J, Pigeonnier et al. Inhibition of chronic myelogenous leukemia cells
harboring a bcr-abl b3a2 junction by antisense oligonucleotides targeted at the b2a2 junc-
tion. Exp Hematol 1995; 23:1608-1611.
71. Zoumadakis M, Neubert WJ, Tabler M. The influence of imperfectly paired helices I and
III on the catalytic activity of hammerhead ribozymes. Nucleic Acids Res 1994; 22:5271-5278.
72. Herschlag D, Khosla M, Tsuchihashi Z et al. An RNA chaperone activity of non-specific
RNA binding proteins in hammerhead ribozyme catalysis. EMBO J. 1994; 13:2913-2924.
73. Bertrand E, Rossi JJ. Facilitation of hammerhead ribozyme catalysis by the nucleocapsid
protein of HIV-1 and heterogeneous nuclear ribonuclear protein A1. EMBO J. 1994;
13:2904-2912.
74. Cameron F, Jennings P. Specific gene suppression by engineered ribozymes in monkey
cells. Proc Natl Acad Sci USA 1989; 86:9139-9143.
75. Cotten M, Birnstiel ML. Ribozyme mediated destruction of RNA in vivo. EMBO J 1989;
8:3861-3866.
76. Renneisen K, Leserman L, Matthes E et al. Inhibition of human immunodeficiency virus-1
in vitro by antibody-targeted liposomes containing antisense RNA to the env region. J Biol
Chem 1990; 265:163337-42.
CHAPTER 17
Potential Design and Facilitation

of Hammerhead Ribozyme Turnover
by Cellular Proteins
Mouldy Sioud
Introduction
P henotypic inhibition of gene expression is an important experimental method to deter-

mine the function of genes and could have a great impact upon the treatment of diseases
in which an undesirable overproduction of a protein is implicated in the pathogenesis. Some
of the major problems of the currently used drugs include both their specificity and cyto-
toxicity. In the case of cancer chemotherapy, for example, the ratio of the toxic dose to the
therapeutic dose is relatively low, indicating that a large number of cellular targets may be
affected by any chemotherapeutic drug. Therefore, there is a need for increasing the speci-
ficity of any medication, since it would be desirable to affect only proteins causing diseases
without harming normal cells.
Because of the specificity of Watson-Crick base pairing, in principle strategies based
upon nucleic acids targeted at a gene involved in diseases such as cancer should interfere
only with the function of this gene. In this area three major approaches have been used to
downregulate the expression of genes. The first is the use of antisense DNA or RNA,1 the
second approach involves the formation of a triple helix2 and the third is ribozyme-medi-
ated RNA cleavage.3-5 For therapeutic application, cleavage of mRNA by ribozymes is likely
to be more promising than the two other strategies mentioned above. Ribozymes not only
complex with target sequences via complementary antisense arms, but also hydrolyze the
target RNA. Thus, ribozymes have the potential to work like molecular scissors to snip a
messenger RNA that contains the genetic code for synthesizing a disease-causing protein.
Ribozyme application has revealed a number of important results in vitro, in tissue
culture and in vivo.6-11 However, despite this significant progress, there is still a lot to learn
in order to optimize ribozyme function in the complex intracellular environment. The in-
tracellular ribozyme activity can be affected by a number of potential intracellular factors
such as the ability of the ribozyme to co-localize with target RNA,12 the accessibility of the
target sites13 and the effect of endogenous proteins upon ribozyme catalysis.14-17 Studies
that address these current obstacles are likely to lead to successful applications of ribozymes.
This chapter addresses some of these problems and summarizes our experimental approaches
towards designing active ribozymes against mRNAs coding for proteins involved in the patho-
genesis and/or perpetuation of autoimmunity and cancer.
General Aspects for Ribozyme Design

In principle, ribozymes can be designed to cleave any RNA sequence as long as the
target molecule contains the nucleotide triplet NUX, where N and X can be any nucleotide,
except G for X. Cleavage occurs most efficiently after the GUC triplet in vitro. However,
considerable disparity appears to exist between in vitro and in vivo results.18 This could be
due, in part, to differential intracellular structural properties of ribozymes and/or their sub-
strates.
An important feature of synthetic ribozymes is the length and base composition of
their antisense arms. For turnover reasons, this length is usually chosen to be 6 to 8 nucle-
otides (nt) on either side of the cleavage site, so dissociation and a complete catalytic cycle
can occur. In general, the free energy for RNA duplex formation between a ribozyme and its
target RNA should be less than -16 kcal/mol for efficient catalytic cycling to occur. Thus, it
would appear that an AU-rich ribozyme target complex is more desirable than, for example,
a GC-rich complex, since the latter will have a high binding energy with only few base-pairs.
Despite these assumptions, there are cases where hammerhead ribozymes with long antisense
arms were found to be more active in the cell as compared to shorter derivatives.19 In accor-
dance with this observation, an IL-2 ribozyme with 16 nt as antisense arms was found to be
less active when compared with ribozymes having 21 or 27 nt antisense arms.16 Since all
ribozymes had the same catalytic core and were targeted to the same site, the differences in
the cell activity most likely reflect the rate of the ribozyme/substrate association. Further
analysis of the target site suggested that an important part of the chosen IL-2 site (11 nt) is
present within a double-stranded RNA region, which limited the binding of the short
ribozymes.20 Interestingly, extension of the hammerhead ribozyme antisense arms with
nucleotides which base pair with the single-stranded regions facilitated its binding to longer
RNA substrates. In another example, a ribozyme directed to the protein kinase A (PKA) RI-
α subunit with antisense arms covering 18 nucleotides of the target RNA cleaved the in vitro
transcribed PKA mRNA less efficiently when compared to a similar ribozyme with antisense
arms covering 21 nucleotides (unpublished results). In our opinion it seems that ribozymes
with long arms are more effective in binding longer target mRNAs than shorter ones.
Another problem for ribozymes having short antisense arms is reduced intracellular
specificity. Therefore, for each site it is essential to define the minimal length of ribozyme
antisense arms. Such length should be tested for binding to in vitro transcribed mRNA in a
cell-free system representative of those that would be generated in vivo, since in vivo activ-
ity of ribozymes seems to differ from predictions derived from in vitro experiments. In this
connection, it was recently shown that a minimum length of 51 nt in the antisense arm was
required for both antisense and ribozyme-mediated inhibition of HIV-1 replication.21
A current difficulty with in vivo use of ribozymes arises in part from insufficient knowl-
edge about intracellular ribozyme structures and stability. In many cases ribozymes were
transcribed with additional long sequences which would dramatically increase the number
of potentially inactive ribozyme conformations. Ribozymes flanked by a minimal length
such as short stable hairpins have been expressed in the cell.8 Such 3' stem-loop structures
were found to increase the lifetime of the ribozyme in the cell by protecting the transcripts
from 3' exonucleases.22
Our initial idea of adding hairpin structures to ribozymes (e.g., T7 terminator) was
recently adapted by Gavin and Gupta,23 where ribozymes with 3' hairpins targeted to the
polycistronic Sendai virus P/C mRNA were found to be both more active and more resis-
tant to intracellular nucleases. The major advantage of incorporating nucleotides which
form stable secondary structures at the 3' end of ribozymes is that they can be synthesized
as part of any ribozyme using conventional DNA/RNA synthesis chemistry.
Potential Design and Facilitation of Hammerhead Ribozyme Turnover by Cellular Proteins 215
In general ribozymes can be delivered to the cells either endogenously as genes coding
ribozymes or exogenously8,16 by transfection. In the latter case both in vitro transcribed
ribozymes and chemically synthesized ribozymes have been used. In order to reduce the
cost of chemically synthesized ribozymes, a smaller version of the hammerhead ribozyme,
in which stem-loop II has been replaced by a short linker, has been developed and shown to
be active both in vitro and in cells.24,25 Due to their reduced size, these ribozymes may be
useful as pharmaceutical agents.
Effect of Proteins on Ribozyme Cleavage

The simplified catalytic cycle of a hammerhead ribozyme as shown in Figure 17.1 con-
sists mainly of:
1. sequence-specific binding to the target RNA via complementary antisense sequence;
2. site-specific hydrolysis of the cleavable motif of the target strand; and
3. release of the cleavage products, which gives rise to another catalytic cycle (turn-
over).
In contrast to in vitro experiments, the turnover of the ribozyme in the cell seems to be
weak, since the ratio of ribozyme vs. target RNA was found to be relatively high (>100) for
activity to occur in the cell.8,11 Thus, there is a need to search for trans-acting factors that
could assist the ribozyme catalysis intracellulary.
We have been interested in developing ribozymes against cytokines such as the tumor
necrosis factors (TNF-α), as well as searching for cellular proteins with potential facilitation
of RNA catalysis and/or stability. The pleiotropic activity of TNF-α is mediated by its ability
to activate some transcription factors, in particular NF-κB, which controls the transcription
of many inflammatory cytokines and growth factors26 such as the granulocyte macrophage
colony stimulating factor (GM-CSF).
During our studies with ribozymes, we have observed that a hammerhead ribozyme
directed against human TNF-α specifically binds to a cellular protein as compared to other
ribozymes and RNA molecules tested.15 The ribozyme maintained its in vitro cleavage ac-
tivity despite its strong binding to the protein. Thus, neither the binding of the ribozyme to
its substrate nor the cleavage step seems to be hampered by the protein. Further analysis
identified this major protein as the glyceraldehyde-3-phosphate dehydrogenase (GAPDH).16
This protein strongly enhances ribozyme catalysis by increasing the ribozyme annealing to
their target RNAs (association) as well as by increasing the product dissociation as summa-
rized in Figure 17.2. The effect of GAPDH on RNA was attributed to its RNA chaperone
activity.16 Furthermore, our results indicated that GAPDH also binds to RNAs in a nonspe-
cific and a specific manner. Interestingly, the nonspecific binding of GAPDH to ribozymes
in general was found to be adequate for the cleavage enhancement to occur in vitro. How-
ever, tissue culture and in vivo experiments suggest that a high affinity RNA binding site for
GAPDH is required for its effect upon ribozyme catalysis. This is not a major problem, since
the facilitation by GAPDH upon ribozyme catalysis could be achieved by including a cis-
appended GAPDH high-affinity binding RNA site to any ribozyme.9 Since ribozymes act by
recognizing and annealing to one RNA target sequence, it is conceivable that proteins able
to affect RNA conformations can enhance their activity and specificity in vivo. This obser-
vation is further supported by the finding that the heterogeneous nuclear ribonucleopro-
tein (hnRNP) A1 can enhance the rate of the hammerhead ribozyme catalysis in vitro.17
As mentioned above, the binding of IL-2 ribozymes with short arms to in vitro tran-
scribed full length IL-2 mRNA or longer targets (e.g., 500 nt) was found to be weaker when
compared to ribozymes with long antisense arms and directed to the same site.20 These
results have suggested that IL-2 mRNA may adopt a conformation that prevents the binding
Fig. 17.1. Schematic representation of the catalytic cycle of a hammerhead

ribozyme. The ribozyme binds the substrate (1) and cleaves it at a specific site
determined by its antisense arms (2). Following cleavage the products can disso-
ciate from the ribozyme (3). Thus the ribozyme can perform another catalytic
cycle.
of the ribozyme with short arms (e.g., 14 nt), but not ribozymes with longer arms (e.g.,
18 nt). Interestingly, the binding problem arising from the IL-2 mRNA secondary and/or
tertiary structure was overcome in vitro by the addition of GAPDH (Fig. 17.3). The ques-
tion remains as to why GAPDH did not assist the IL-2 ribozymes with short antisense arms
in the cell.20 The answer is not surprising, since our data have suggested that a high affinity
binding RNA site for GAPDH is required for its activity in the cell and in vivo.9,16 Such a
high affinity RNA binding site for GAPDH would localize the protein near the ribozyme
and ensure its interaction with the ribozyme and/or targeted RNA.
Ribozymes as Gene Therapy in Cancer Treatment

The use of ribozymes in cancer therapy has focused mainly on the inhibition of tumor-
specific oncogenes and multi-drug resistance genes (see ref. 27 for review). These early studies
have suggested that there are many appropriate targets for ribozyme strategies in cancer
therapy. As mentioned above, TNF-α seems to regulate the expression of GM-CSF, which is
expressed by a wide range of cells following activation.28 However, in some pathological
conditions such as certain myeloid leukemias and inflammations, GM-CSF is constitutively
expressed and may play a pathological role.29,30 Ribozymes targeted to human TNF-α mRNA
were found to inhibit both the expression of GM-CSF in primary leukemic cells from pa-
tients with juvenile myelomonocytic leukemia and GM-CSF dependent colony formation
by JMML cells (Iversen and Sioud, submitted for publication).
GAPDH
GAPDH
GAPDH
GAPDH
Fig. 17.2. Potential effect of GAPDH on ribozyme catalysis. The binding of

GAPDH to the ribozyme (A) and/or the substrate (B) may modulate their sec-
ondary structures and direct the annealing process (C). Following cleavage,
GAPDH would also facilitate the dissociation step (D). These two activities seem
to depend on the length of the RNA duplexes. For details see ref. 16.
Fig. 17.3. In vitro facilitation of

RNA catalysis by GAPDH. Phos-
phorImager printout of 6% poly-
acrylamide denaturing gel show-
ing the in vitro cleavage activity
of the 500 nt target IL-2 mRNA
by an IL-2 ribozyme (+) with
14 nt as antisense arms in the
absence (lane 2) or in the pres-
ence of various concentrations of
GAPDH (lanes 3 to 8) for 60 min
at 37°C. Lane 1 contains only the
500 nt substrate. Maximum ef-
fect of GAPDH was seen at con-
centrations of 70 to 150 ng/µl
(lanes 6 and 7, respectively). For
more details see refs. 16 and 20.
Since 1986, we have been interested in investigating the molecular mechanism by which
anti-tumor drugs and newly discovered prokaryotic antibiotics kill cells.31 The molecular
target of, for example, the epipodophyllotoxins (e.g., VP16, VM26) was identified as DNA
topoisomerase II, an enzyme involved in DNA replication repair and transcription. Interest-
ingly, the cleavage activity of VP16 was found to be similar to ciprofloxacin, an inhibitor of
bacterial DNA gyrase.32 More recently, epipodophyllotoxins and other drugs known to inter-
act specifically with actin, tubulin, DNA or DNA topoisomerase II were found to also acti-
vate the apoptosis machinery of eukaryotic cells.33 However, despite their promising activ-
ity in cell death, their toxicity remains the major problem in cancer chemotherapy. Thus,
there is a need for developing novel tools that specifically induce apoptosis in cancer cells.
The susceptibility of cells to apoptosis seems to be determined by the relative levels and
interactions between the Bcl-2 related proteins, such as Bax, Bcl-xL, Bad, Bag, Bak, and Bik,
some of which protect from cell death (Bcl-xL) while other members of the family, such as
Bax, promote apoptosis.34 The interactions and the expression of these proteins can be af-
fected by different activation pathways, for example those involving the protein kinase C
(PKC).
In order to determine which PKC isoform is involved in apoptosis we have used as a
model a rat glioma cell line called BTCn.35 This cell line has the properties that seem rel-
evant for a model of aggressive human gliomas. The BTCn cell line gives both subcutaneous
and brain tumors in syngeneic BD IX rats. As shown in Figure 17.4C, a hammerhead ribozyme
targeted to the PKC-α isoform induced apoptosis in all BTCn cells as detected by the TUNEL-
method.36 This method is based on the fact that apoptotic cells contain free 3' ends of double-
stranded DNA due to the endonuclease digestion of genomic DNA at the nucleosomal in-
tervals. Such 3' ends can be used as substrate by the terminal deoxynucleotidyl transferase.
In this assay apoptotic cells would show nuclear staining as shown in Figure 17.4. By intro-
ducing 2'-amino pyrimidine residues into a catalytically active protein kinase Cα ribozyme,
we have designed a ribozyme that is over 14,000-fold more stable than its unsubstituted
version yet retains most of its biological activity. A single injection of the modified ribozyme
into a glioma solid tumor almost completely inhibited tumor growth.37
The growth of new blood vessels from an existing vascular supply in the tissue is a
process called angiogenesis. This process is an essential component of tumor growth.38 The
newly formed blood vessels, in addition to their nutritional role, also provide an exit route
Fig. 17.4. Induction of apoptosis by a

PKC-α specific ribozyme directed to the
CUC at codons 10 and 11. BTCn cells
were transfected with only DOTAP (li-
posomal transfection reagent Chemical
name N-[1-(2,3-Dioleoloxy)propyl]-N,
N, N-trimethylammonium methyl-
sulfate) (A), the mutant ribozyme (B) or
A the ribozyme (C) for 48 hours in slide
flasks. Following transfection, cells were
fixed with ETOH for 5 min and then
permeabilized with 0.1% saponin in PBS
containing 1% BSA fraction V for
10 min. Following washing with PBS, the
apoptotic cells were detected using a
commercially available in situ cell death
fluorescein detection kit (Boehringer,
Mannheim, Germany) based on termi-
nal deoxynucleotidyl transferase (TdT)-
mediated dUTP-FITC nick end labeling
(TUNEL). Briefly, permeabilized cells
were incubated in the TUNEL reaction
for 30 min. Following washing with PBS,
samples were covered with antifade and
mounting medium for fluorescence mi-
B croscopy analysis.
C
for metastasizing tumor cells into the systemic circulation. Given the importance of such a
process in tumor progression, inhibition of the factor promoting angiogenesis may lead to
the development of new anti-cancer therapy. Many factors, such as the vascular endothelial
growth factor (VEGF), basic fibroblast factor (BFF) and TNF-α, were found to stimulate the
angiogenesis process.39 We are currently studying the effect of synthetic ribozymes against
VEGF on tumor growth and vascularization. Our ongoing experiments indicate that tumor
vascularization can be inhibited by ribozymes. Using a similar strategy, it was suggested
recently that pleiotropin (PTN) is a factor that could be responsible for melanoma angio-
genesis and metastasis.40 This study supports a direct link between these two processes;
however it remains to be demonstrated that such a link exists in other tumors.
In conclusion, the described set of experiments indicate: firstly, that both synthetic
ribozymes and genes encoding ribozymes can be delivered to cells in an active form; sec-
ondly, that endogenous cellular proteins can provide protection and catalysis enhancement
for hammerhead ribozymes; and thirdly, that both the machinery of apoptosis and angio-
genesis in cancer cells can be targeted specifically by ribozymes. Furthermore, our data
emphasize the importance of detailed structural investigations of ribozyme target RNAs.
Thus, we need to learn more about the traffic of RNA inside the cell, as well as the target
secondary structures and the cellular factors that can facilitate ribozyme catalysis and sta-
bility. In addition, we should address the problem of delivery of preformed ribozymes. Pres-
ently, cationic lipids seem to be the most versatile method.
Acknowledgments
This work was supported in part by grants from Gene Shears Pty. Inc, The Norwegian
Research Council, The Norwegian Womens Public Health Organization, The European Union
(Biotech program) and currently by the Norwegian Radium Hospital.
References
1. Wagner RW. Gene inhibition using antisense oligonucleotides. Nature 1994; 372:333-335.
2. Helene C, Thuong NT, Harel-Bellan A. Control of gene expression by triple helix-forming
oligonucleotides: The antigen strategy. Ann New York Acad Sci 1992; 660:27-36.
3. Cech T. The chemistry of self-splicing RNA and RNA enzymes. Science 1987; 236:1532-1539.
5. Haseloff J, and Gerlach WL. Simple RNA enzymes with new and highly specific
endoribonuclease activities. Nature 1988; 334:585-591.
6. Cameron FH, Jennings PA. Specific gene suppression by engineered ribozymes in monkey
cells. Proc Natl Acad Sci USA 1989; 83:8859-8862.
7. Sarver N, Cantin EM, Chang PS et al. Ribozymes as potential anti HIV-1 therapeutic agents.
Science 1990; 247:1222-1225.
8. Sioud M, Drlica K. Prevention of HIV-1 integrase expression in E.coli by a ribozyme. Proc
Natl Acad Sci USA 1991; 88:7303-7307.
9. Sioud M. Ribozyme modulation of lipopolysaccharide-induced tumor necrosis factor-α
production by peritoneal cells in vitro and in vivo. Eur J Immunol 1996; 26:1026-1031.
11. Cotten M, Birnstiel ML. Ribozyme mediated destruction of RNA in vivo. EMBO J 1989;
8:3861-3866.
12. Sullenger BA, Cech TR. Tethering ribozymes to a retroviral packaging signal for destruc-
tion of viral RNA. Science 1993; 262:1566-1569.
13. Taylor NR, Rossi JJ. Ribozyme mediated cleavage of an HIV-1 gag RNA: The effects of
non-targeted sequences and secondary structure on ribozyme cleavage activity. Anti Res
Dev 1991; 1:173-186.
14. Tsuchihashi Z, Khosla M, Herschlag D. Protein enhancement of hammerhead ribozyme

catalysis. Science 1993; 262:99-102.
15. Sioud M. Interaction between tumor necrosis factor α ribozyme and cellular proteins. J
Mol Biol 1994; 242:619-629.
16. Sioud M, Jespersen L. Enhancement of hammerhead ribozyme catalysis by glyceraldehyde-
3-phosphate dehydrogenase. J Mol Biol 1996; 257:775-789.
17. Bertrand EL, and Rossi JJ. Facilitation of hammerhead ribozyme catalysis by the nucleo-
capsid protein of HIV-1 and the heterogeneous nuclear ribonucleoprotein. EMBO J 1994;
13:2904-2912.
mediated cleavage within a fusion gene for adenovirus E1A-associated 300 kDa protein
19. Crisell P, Thompson S, James W. Interaction of HIV-1 replication by ribozymes that show
poor activity in vitro. Nucleic Acids Res 1993; 21:5251-5255.
20. Sioud M. Effects of variations in length of hammerhead ribozyme antisense arms upon the
cleavage of longer RNA substrates. Nucleic Acids Res 1997; 25:333-338.
21. Hormes R, Homann M, Oelze I et al. The subcellular localization and length of hammer-
head ribozymes determine efficacy in human cells. Nucleic Acids Res 1997; 25:769-775
22. Sioud M, Opstad A, Zhao JQ et al. In vivo decay kinetic parameters of hammerhead
ribozymes. Nucleic Acids Res 1994; 22: 5571-5575.
23. Gavin DK, Gupta KC. Efficient hammerhead ribozymes targeted to the polycistronic Sendai
virus P/C mRNA. J Biol Chem 1997; 272:1461-1472.
24. McCall MJ, Hendry P, Jennings PA. Minimal sequence requirements for ribozyme activity.
25. Sioud M, Opstad A, Hendry P et al. A minimised hammerhead ribozyme with activity
against interleukin-2 in human cells. Biochem Biophys Res Com 1997; 231: 397-402.
26. Baeuerle PA, Henkel T. Function and activation of NF-κB in the immune system. Annu
Rev Immunol 1994; 12:141-179.
27. Poeschla E, Wong-Staal F. Antiviral and anticancer ribozymes. Curr Opin Oncology 1994;
6:601-606.
28. Gasson JC. Molecular physiology of granulocyte-macrophage colony stimulating factor.
Blood 1991; 77:1131-1145.
29. Young DC, Wagner K, Griffin JD. Constitutive expression of the granulocyte-macrophage
colony stimulating factor gene in acute myeloblastic leukemia. J Clin Invest 1987;
79:100-106.
30. Xu W, Firestein GS, Taetle R et al. Cytokines in chronic inflammatory arthritis. II. GM-
CSF in rheumatoid synovial effusions. J Clin Invest 1989; 83:876-882.
31. Sioud M, Baldacci G, Forterre P et al. Antitumor drugs inhibit the growth of halophilic
archaebacteria. Eur J Biochem 1987; 69:231-236.
32. Sioud M, Forterre P. Ciprofloxacin and etoposide (VP16) produce a similar pattern of
DNA cleavage in a plasmid of an Archaebacterium. Biochemistry 1989; 28:3638-3641.
33. Hannun YA. Apoptosis and the dilemma of cancer chemo-therapy. Blood 1997;
89:1845-1853.
34. Kroemer G. the proto-oncogene Bcl-2 and its role in regulating apoptosis. Nature Med
1997; 3:614-620.
35. Mella O, Bjerkvig R, Schem BC et al. A cerebral glioma model for experimental therapy
and in vivo invasion studies in syngeneic BD IX rats. J Neuro-Oncology 1990; 9:93-104.
36. Gavrieli Y, Shermen Y, Ben-Sasson SA. Identification of programmed cell death in situ via
a specific labelling of nuclear DNA fragmentation. J Cell Biol 1989; 119:493-498.
37. Sioud M, Børensen R. A nuclease-resistant protein kinase Cα ribozyme blocks glioma cell
growth. Nature Biotech 1998; 16:556-561.
38. Folkman J. Fighting cancer by attacking its blood supply. Scient American 1996;
275:116-119.
39. Levis CE, Leek R, Harris A et al. Cytokine regulation of angiogenesis in breast cancer: The
role of tumour-associated macrophages. J Leuk Biol 1995; 57:747-751.
40. Czubayko F, Schulte AM, Berchem GM et al. Melanoma angiogenesis and metastasis modu-
lated by ribozyme targeting of the secreted growth factor pleiotrophin. Proc Natl Acad Sci
USA 1996; 93:14753-14758.
CHAPTER 18
Human Papillomaviruses
E.J. Shillitoe
Introduction
S everal viruses are susceptible to the effects of ribozymes. These include HIV,1,2 bovine
leukemia virus,3 tobacco mosaic virus,4 and lymphocytic choriomeningitis virus.5 How-
ever, all of these are RNA viruses. Thus both the viral genome and its RNA transcripts could
serve as targets. For DNA viruses the susceptibility to ribozymes could be very different, and
indeed very little work has been done in that area. However there is one example of a DNA
virus in which several investigators have explored a role for ribozymes, and that is the
papillomavirus group.
Papillomaviruses
Human papillomaviruses (HPVs) are one of the largest groups of viruses. Over 70
types are now recognized. They are the focus of a very large research effort, and new infor-
mation is continually summarized and made available through the Los Alamos online data-
base (http://hpv-web.lanl.gov/). The HPVs can be classified in several ways, but it is useful
to consider some of them as being “low risk” and some as being “high risk” viruses. The low
risk HPVs are responsible for warts and similar superficial lesions, while the high risk HPVs
are found in dysplastic and malignant lesions. They include HPV-16, HPV-18 and HPV-33,
and are found in cervical cancers and some oral cancers. Other HPVs, such as types -6 and
-11 seem to be of an intermediate risk and occur in conditions, such as laryngeal papilloma-
tosis, which are generally benign but can occasionally become malignant.
An important question is whether the high risk HPVs are the cause of the malignancy
in which they are found, or whether they are simply present in a latent or passenger state.
The question is made difficult by the fact that HPVs are sometimes found in normal tissues.
However it does seem clear that in cervical cancer the HPVs are largely responsible for the
malignancy. The viruses are present in over 90% of cervical cancers.6 Several large case-
control epidemiological studies have confirmed the association of high risk HPV types and
cervical cancer.7,8 The malignant tissues continually express high levels of the viral RNA.9 It
is of course likely that other factors are also involved in the development of the cancer, but
since the virus is at least partly responsible, then any anti-HPV techniques could protect
patients against cervical cancer, or might be helpful as part of its treatment.
In the case of oral cancer the role of the virus is not quite so clear. Many studies have
shown the presence of HPV DNA in the tumors.10 For example, one study found HPV DNA
in 49% of patients and 8% of control subjects.11 Another study found HPV in 29% of oral
cancers and no control tissues.12 Only one epidemiological study has used a carefully matched
control group and, although it found a low incidence of HPV generally, it did find a signifi-
cantly higher prevalence of HPV in the cancer as compared to the control subjects.13 No
studies have reported that the viral genes are actually being expressed in oral cancer, and
this is a significant defect in the virological studies of that disease.
Studies in vitro have confirmed the clinical suspicion that the high risk HPVs are onco-
genic. These viruses can transform normal human keratinocytes in vitro to a malignant
phenotype, in a step-wise fashion. Initial infection of the cells leads to a cell line which
expresses viral genes and has an immortal phenotype in which the cells can not make tu-
mors in nude mice.14 They do, however, have an increased frequency of mutations, which
makes them more susceptible to the effects of chemical carcinogens.15 If these cells are then
exposed to carcinogens they undergo changes that do make them tumorigenic. Theses
changes include amplification of the HPV sequences.16 This sequential development of the
malignant phenotype may be an in vitro equivalent of the changes that happen in vivo as
HPV-associated cancers develop.
The progression of HPV-immortalized cells to cancer may involve genetic factors that
are not well understood. Two studies have failed to find changes in the expression level of
HPV genes or the E7 protein during the progression of cells from immortal to malignant.17,18
Thus other genetic changes could have been occurring. Indeed, one study found that ex-
pression of the myc oncogene was increased in one HPV-immortalized cell line14 and possi-
bly other genes were activated at the same time.
The Viral Genome

The genome of all HPVs is very similar, and is illustrated in Figure 18.1. It is composed
of about 8 kb of double stranded DNA, and all genes are on the same strand. Some genes are
designated ‘E’ since they have similarities to the early genes of other small DNA viruses, and
some are designated ‘L’, and are similar to the late, structural genes of other viruses. The
remainder of the genome is designated as the ‘long control region’ (LCR) or ‘upstream regu-
latory region’ (URR). This region is the major enhancer and promoter region for the viral
genes, although minor promoters do exist elsewhere.
Mechanism of Oncogenesis
The molecular mechanism by which HPVs produce cancer has been elucidated in some
detail. The most important genes are the E6 and E7 genes. These encode small proteins
which together are necessary and sufficient for oncogenesis.19 When cells are transformed
by HPVs the viral RNA must be expressed continuously for them to stay transformed.20 The
E6 and E7 genes are expressed as a single transcript, which is spliced (Fig. 18.2). The effect
of the splice is to produce two different E6 proteins, one of which is shorter and has a frame
shift beyond the splice acceptor site. The significance of the splice is not certain, although it
may allow more efficient initiation of translation of the E7 protein.21,22
The E6 gene product binds to the p53 tumor-suppressor protein,23 and the E7 gene
product binds to the Rb tumor-suppressor protein.24 The effect of the binding of E6 to p53
is to activate an enzyme pathway, the ubiquitin system, which then digests the p53 protein.
Thus the half life of p53, which is normally 3 h or more, is reduced to around 15 min in
tumor cells which express HPVs.25,26 The effect of binding E7 is to inactivate the Rb protein
by preventing its normal interactions with other cell proteins.19,24 These effects of E6 and E7
are not unique to HPVs, but are in fact the mechanisms that are used by several of the small
DNA tumor viruses.27
Although the high risk HPVs are found in many persons, the cancers that are associ-
ated with them are relatively rare. Thus there must be other factors that determine the de-
velopment of HPV-associated cancers. One critical factor appears to be the level of expres-
Human Papillomaviruses 225
Fig. 18.1. Gene map of

HPV-18, which is one of the
high risk papillomaviruses.
This map shows the features
that are common to HPV-16,
HPV-18 and other HPVs that
are associated with some hu-
man carcinomas. The ge-
nome is around 8 kbp. The
upstream regulatory region
(URR) controls expression of
the small E6 and E7 proteins,
whose continual expression is
necessary for malignancy.
The other genes are probably
involved in viral replication
and viral structure, but are
not associated with human
cancers. They are often lost
when the viral genome be-
comes integrated into the
chromosome of human can-
cer cells.
Fig. 18.2. Partial map of the HPV-18 genome, showing the open reading frames (ORFs) for the
E6 and E7 genes, together with the RNA transcript with a splice in the E6 region.70 The vertical
arrows indicate the three target sites at nucleotides 123, 309 and 671 at which ribozymes can be
effective in inhibiting the growth of HPV-18-expressing cancer cells.
sion of the transforming genes.28 Expression levels may be increased in cervical cancers by
mutation or other modifications to the viral genome.29 In many cervical cancer cell lines,
the viral DNA is integrated into the cellular chromosome and the integration routinely
truncates the E2 gene of the virus. The E2 gene product is a repressor of high risk HPV gene
expression,30 although regions of the protein have the potential to act as transcriptional
activators.31 Thus, in cancers it is possible that the function of the integration at the E2 gene
site is to produce a truncated E2 gene product which has a stimulating effect on the expres-
sion of the E6 and E7 genes.29,32 Another mechanism of overexpression can happen in la-
ryngeal papillomas which have become malignant. The HPV that is present in those lesions
has been reported to show duplications in the URR which thereby increase its strength as a
gene expression enhancer.33 In some cervical cancers other changes in the URR can lead to
increased expression of viral genes. The binding sites for the transcriptional repressor YY-1
are mutated in some cases, and this leads to over-expression of viral genes.34 In oral cancer
cell lines that contain DNA of HPVs there is frequent deletion of sequences from suppres-
sor regions of the URR, which leads to greater activity.35 Thus HPV-associated cancers often
show over-expression of the transforming genes because of mutations or other changes in
the viral genome.
Anti-Papillomavirus Gene Therapy

Since the molecular mechanism by which HPVs stimulate the growth of tumor cells
are known, several targets for new types of therapy have been revealed. These could include
use of mutated URRs for expression of suicide genes, or development of proteins that could
interrupt the effects of the E6 and E7 proteins.36 In the present context, however, the impor-
tant targets are the RNA transcripts that encode the E6 and E7 proteins.
Antisense Inhibition of Expression of HPV-Genes

One of the earliest ways of blocking gene expression specifically was by the use of
antisense RNA. These molecules, transcribed from the opposing strand of DNA from the
one used in gene expression, can hybridize to the sense strand and inhibit translation. Sev-
eral workers have shown that antisense RNA can inhibit the expression of the transforming
genes of papillomaviruses.
The first study of this type used cDNA sequences of HPV-16, which were cloned into
expression vector plasmids and then transferred to the HPV-16-expressing cervical cancer
cell line, C4-1.37 Expression of antisense RNA was then induced by dexamethasone. Expres-
sion of sense RNA was shown to be reduced, as was expression of the HPV-proteins, and
growth of the cells was reduced. The interpretation of the experiments was complicated by
the fact that the dexamethasone by itself had effects on the C4-1 cells, but nevertheless the
work showed that HPV RNA was a reasonable molecule to target for future research into
HPV-associated cancers.
In a follow-up study it was reported that the tumorigenicity of the cells was reduced by
the effects of the antisense sequences.38 Antisense-transduced cell lines were implanted into
nude mice so as to form tumors, and some mice were given dexamethasone in the drinking
water so as to induce expression of the antisense sequences. Those mice developed tumor
nodules which were significantly smaller. Surviving tumor cells were shown to be express-
ing the antisense RNA, which therefore was not completely inhibitory to the tumors.
Another group showed that synthetic antisense oligonucleotides could also inhibit ex-
pression of HPV in cancer cells.39 They made phosphorothioate oligonucleotides that rep-
resented the start codon regions of the E6 and E7 genes of HPV-18. These were shown to
inhibit the growth of cervical and oral cancer cells that carry DNA of HPV-18 but without
effects on other cells. Either the blocking of E6 or E7 had some effect, but blocking both
together was the most effective. As might be expected, the synthetic oligonucleotides were
rapidly degraded, despite the protection of the phosphorothioate modification, and repeated
application was necessary. Once the oligonucleotides were degraded the cells recovered and
grew as rapidly as before.
As with HPV-16, it is possible to express antisense RNA to HPV-18 from expression
vector plasmids, and another study introduced such plasmids into the HeLa cell line, which
expresses HPV-18.40 Cell lines were developed that expressed antisense sequences to either
the E7 gene, or to both E6 and E7. The antisense-expressing cells showed slower growth,
reduced formation of colonies in soft agar and increased serum requirements, which all
indicate a loss of the transformed phenotype. These effects were not seen in tumor cells that
lacked HPV sequences.40
These early results have been confirmed by other workers. It has been shown that syn-
thetic antisense sequences to HPV-16 E6 and E7 could inhibit the transformed phenotype
of the cervical cancer cell lines CaSki and SiHa. Both in vitro parameters of transformation
and tumorigenicity in nude mice were reduced.41 Others have investigated the effects of
synthetic oligonucleotides directed against the start codon of the E7 gene of HPV-16.42 In a
rabbit reticulocyte assay system the antisense oligonucleotide did reduce the translation of
the E7 protein. When the gene that encoded the antisense sequence was introduced into
CaSki cells by lipofection, the HPV RNA transcript was lost from the cells and was replaced
by two shorter transcripts. This was interpreted as meaning that RNase H degradation had
been activated. This interesting observation seems to be unique in antisense studies and
deserves further attention. Curiously, no change was found in the protein levels of E7 in the
cells and it was not reported whether any changes occurred in the phenotype of the cells.
The level of the receptor for epidermal growth factor has been found to fall in HeLa cells
that are transfected with antisense genes for E6 and E7, although the exact pathway has not
been investigated.43
Although confirmatory studies on the effects of antisense to E6 and E7 have been re-
ported,44,45 not all workers have been able to detect anti-HPV effects from the use of antisense
RNA. One group found significant nonspecific effects from the use of synthetic antisense
oligonucleotides,46 and in fact nonspecificity of phosphorothioate-modified antisense mol-
ecules is a critical problem with numerous targets.47
Inhibition of HPV by Ribozymes

Since antisense RNA shows some effects in the inhibition of papillomaviruses, it seems
very likely that ribozymes would also show an effect, and possibly be more potent. The first
demonstration that this could be possible came from a study of the cottontail rabbit
papillomavirus.48 It was found that a ribozyme directed at the E6/E7-containing RNA tran-
script of that virus could cleave the transcript in vitro. This early work on rabbit
papillomaviruses has not been continued, and more recent interest has centered on the
human high-risk papillomaviruses. Their DNA sequence shows a very large number of po-
tential target sites, as might be expected.36
HPV-16 RNA transcripts were first shown to be cleavable by in vitro studies, using very
short synthetic targets. One study examined the use of a self-trimming construct that could
cut itself out of a longer transcript.49 This anti-HPV ribozyme could then cut various length
HPV-16 transcripts in vitro, but the efficiency depended on the length of the target. A very
short target of 171 nucleotides (nt) was cut efficiently, with 64% reduction of the target
under the conditions of the assay. However, a target of 360 nt fragment was reduced by only
25%, and a target of 686 nt was not cut at all. The natural HPV RNA transcripts in cancer
cells are much longer than any of these, being, for example, 3.4 and 1.6 kb in HeLa cells,40
which appears to raise doubts about the use of ribozymes in cancer cells. Another study
showed that ribozymes directed to the region of the start codons of the E6 and E7 genes of
HPV-16 did cleave the target RNA in vitro when it was in the form of a partial-length 202
base transcript.50 Simultaneous cleavage by both ribozymes was also demonstrated.
Ribozymes in this case were expressed from a plasmid which contained adeno-associated
virus sequences and which could presumably be used to produce a virus vector.
Ribozymes that can cleave the RNA transcript of HPV-18 have been examined in some
depth.51,52 Two ribozymes were designed so as to be directed at the first possible target sites
in the E6 and E7 coding regions respectively, while a third target site was selected as being in
a region of minimal secondary structure, according to computer predictions. The full-length
RNA transcript of HPV-18 was then expressed in E. coli cells, and a ribozyme was expressed
simultaneously from a different plasmid. The results showed that the most effective ribozyme
was the one directed against the region of minimal secondary structure, named Rz309.51 As
well as being most effective in the E. coli assay, Rz309 was also the most effective when used
to cut HPV RNA in vitro that had been extracted from HeLa cells. The ability to cut the full-
length transcript was considered to be significant and implied a potential use for inhibiting
the tumorigenicity of the cells. The fact that the most effective ribozyme was the one se-
lected from a computer prediction is of course interesting, but is not a conclusive demon-
stration that secondary structure issues can be avoided in this simple way. It is also notewor-
thy that Rz309 is directed to a region of the transcript that is removed by splicing (Fig. 18.2).
This implies that either the cleavage by the ribozyme happens early after transcription, be-
fore splicing has occurred, or else that the unspliced E6 product is essential for tumor cell
growth.
The same three ribozymes were then cloned into an eukaryotic shuttle vector and trans-
ferred to HeLa cells. Each of the ribozymes showed inhibitory effects on the cells, but again
it was Rz309 which was most effective. It caused a decrease in the intracellular concentra-
tion of the HPV-18 RNA transcript and a corresponding increase in the intracellular con-
centration of the p53 gene product. The cells showed reduced growth rates, increased se-
rum dependency, and reduced focus formation in soft agar.52 All of these changes suggest a
reduced potential to be malignant. Thus Rz309, and other ribozymes, were judged to have
potential as anti-tumor agents for treatment of HPV-associated tumors.
Another group has cloned an anti-HPV-16 ribozyme into a plasmid vector, under con-
trol of the RSV-LTR promoter.53 The construct was transfected into C4-1 cervical cancer
cells. By the use of an RNase protection assay it was found that the level of HPV-16 E7 RNA
was reduced by around 90%. In contrast, an antisense-expressing plasmid reduced expres-
sion only around 20%. Although no measurements were taken of the effects of this on cel-
lular phenotype, these results also seem to be promising.
Delivery Methods for Anti-HPV Ribozymes

Any method for delivery of anti-HPV ribozymes will have to take account of the type
of tissue and tumor in which HPV is found. In the early stages of cervical or oral cancer, the
virus is found in epithelium which is relatively thin and accessible, and thus it might seem to
be a simple matter to transduce the affected cells. In practice, no effective method has emerged.
The most straightforward technique is the use of particle bombardment, in which a he-
lium-powered ‘gene gun’ is used to fire DNA-coated gold beads into the epithelium. Al-
though this is simple and fast, the penetration of the beads does not reach the basal layers of
the epithelium, which is where cell division takes place.54 Thus the transduced cells are lost
within a few days. Improvements to the gene gun technology might allow more effective
transfer of DNA, in which case it could become a very useful technique.
An alternative method for transduction of tissues is the use of virus vectors. Retroviruses
can be used to transduce oral and cervical cells in culture and, by careful selection, trans-
duced cell lines can be developed. Many workers have shown that retrovirus vectors can
introduce functional ribozymes into cells so as to inhibit the effects of HIV,55-58 SIV,59 the ras
oncogene60 or the BCR/ABL fusion gene.61 However there have been no in vivo experiments
to suggest that retroviruses will be used in treatment of HPV-associated human carcinomas.
Adenovirus vectors provide an alternative to retroviruses. Adenovirus vectors have been
used to introduce ribozymes to cells so as to inhibit growth hormone,62 transferase genes,63
the hepatitis C virus64 or the ras oncogene.65 Adenoviruses readily transduce oral or cervical
cancer cells in vitro, and will diffuse through solid tumors to some extent.66,67 However, in
vivo their efficiency is rather limited, since they do not replicate in the tumor and are lim-
ited to the area where they were deposited. They have not been able to transduce the basal
cell layer of epithelia in an effective way.66 Our recent experience suggests that expression of
anti-HPV ribozymes from adenovirus vectors can be obtained readily, yet they are not al-
ways functional for cleavage of the target transcript. This is probably associated with sub-
compartmental distribution within the cancer cell, secondary structure of the ribozyme, or
stability of the ribozyme, and this problem must be solved before further progress is pos-
sible (Shillitoe et al, in preparation).
In addition to a suitable delivery method, suitable gene promoters and enhancers will
be needed for proper expression of ribozymes in cancers that express HPV. The desirable
properties of such promoters are known in some detail. Presumably powerful expression
will be necessary, so as to obtain a suitable excess of ribozyme sequences relative to target
sequences, but the level of expression has not been determined accurately. Presumably speci-
ficity of expression will be less important. Since ribozymes show specific actions rather than
nonspecific, they would not be expected to produce untoward effects if they become ex-
pressed in other cells apart from the tumor cells. Thus strength of a promoter will be more
important than cell type specificity.
Tissue specific promoters for expression of anti-ras ribozymes in melanomas has been
attempted using the tyrosinase promoter.68 Tissue-specific promoters for cervical or oral
cancer have not been identified—however one candidate is the secretory leukoprotease in-
hibitor gene promoter which is active in many carcinomas.69 Alternatively, papillomavirus
promoters might be expected to serve as tumor-specific promoters in cells that express
papillomavirus genes. Indeed one group has demonstrated that in oral cancer cells that
contain HPV DNA, the promoter regions of the virus have generally undergone mutations
that make them more active in HPV-containing carcinomas.35 These promoters can now be
examined for their ability to direct tumor-specific expression of ribozymes.
Future Directions for Research

Although encouraging progress has been made in the development of ribozymes that
will target expression of HPVs in human cancer cells, there are still many issues to be re-
solved before human therapy will be possible. In the case of oral cancer it is not entirely
clear that the presence of HPV DNA is a cause, or partial cause, of the tumor. It has not been
shown that the DNA is expressed as RNA, and of course unless this occurs then there will be
no role for ribozymes. In the case of cervical cancer, a critical issue is the mode of delivery of
ribozyme genes. Injection of adenovirus vectors into solid tumors seems to be the best avail-
able method so far, but it has not been shown to produce regression of HPV-associated
tumors that are grown in nude mice. The long pre-malignant phase that is seen in many
cervical cancers does provide an opportunity for the use of ribozyme therapy, but so far
there are no reports of how the dysplastic tissues could be transduced. Another important
question is the way that expression of ribozymes can be optimized within a tumor cell. The
sub-cellular localization, the length of flanking arms, the expression level that is needed and
the role of post-transcriptional processing will all need to be addressed. Although the use of
ribozymes in management of HPV-containing cancers is under investigation, there are sev-
eral questions that must be answered before their future role will be clear.
Acknowledgment
Original work referred to in this review was supported by PHS grant R01 DE10842.
References
1. Homann M, Tzortzakaki S, Rittner K, Sczakiel G, Tabler M. Incorporation of the catalytic
domain of a hammerhead ribozyme into antisense RNA enhances its inhibitory effect on
the replication of human immunodeficiency virus type 1. Nucl Acids Res 1993;
21:2809-2814.
2. Ventura M, Wang P, Ragot T, Perricaudet M, Saragosti S. Activation of HIV-specific
ribozyme activity by self-cleavage. Nucl Acids Res 1993; 21:3249-3255.
3. Cantor GH, McElwain TF, Birkebak TA, Palmer GH. Ribozyme cleaves rex/tax mRNA and
inhibits bovine leukemia virus expression. Proc Natl Acad Sci USA 1993; 90:10932-10936.
4. de Feyter R, Young M, Schroeder K, Dennis ES, Gerlach W. A ribozyme gene and an
antisense gene are equally effective in conferring resistance to tobacco mosaic virus on
transgenic tobacco. Mol General Genet 1996; 250:329-338.
5. Xing Z, Whitton JL. Ribozymes which cleave arenavirus RNAs: Identification of suscep-
tible target sites and inhibition by target site secondary structure. J Virol 1992; 66:1361-1369.
6. zur Hausen H. Papillomaviruses in anogenital cancer as a model to understand the role of
viruses in human cancers. Cancer Res 1989; 49:4677-4681.
7. Lorincz AT, Reid R, Jenson AB, Greenberg MD, Lancaster W, Kurman RJ. Human
papillomavirus infection of the cervix: Relative risk associations of 15 common anogenital
types. Obstet Gynecol 1992; 79:328-337.
8. Munoz N, Bosch FX, DeSanjose S et al. The causal link between human papillomavirus
and invasive cervical cancer: A population-based case-control study in Colombia and Spain.
Int J Cancer 1992; 52:743-749.
9. Stoler MH, Rhodes CR, Whitbeck A, Wolinsky SM, Chow LT, Broker TR. Human
papillomavirus type 16 and 18 gene expression in cervical neoplasias. Human Pathol 1992;
23:117-128.
10. Miller CS, White DK. Human papillomavirus expression in oral mucosa, premalignant
conditions, and squamous cell carcinoma: A retrospective review of the literature. Oral
Surg Oral Med Oral Pathol Oral Radiol and Endo 1996; 82:57-68.
11. Yeudall WA, Campo MS. Human papillomavirus DNA in biopsies of oral tissues. J Gen
Virol 1991; 72:173-176.
12. Brandsma JL, Abramson AL. Association of papillomavirus with cancers of the head and
neck. Arch Otolaryngol- Head Neck Surg 1989; 115:621-625.
13. Maden C, Beckmann AM, Thomas DB et al. Human papillomaviruses, herpes simplex vi-
ruses, and the risk of oral cancer in men. Am J Epidemiol 1992; 135:1093-1102.
14. Park NH, Min BS, Li SL, Huang MZ, Cherick HM, Doniger J. Immortalization of normal
human oral keratinocytes with type 16 human papillomavirus. Carcinogenesis 1991;
12:1627-1631.
15. Shin KH, Tannyhill RJ, Liu X, Park NH. Oncogenic transformation of HPV-immortalized
human oral keratinocytes is associated with the genetic instability of cells. Oncogene 1996;
12:1089-1096.
16. Park NH, Gujuluva CN, Baek JH, Shin KH. Combined oral carcinogenicity of HPV-16
and benzo(a) pyrene: An in vitro multistep carcinogenesis model. Oncogene 1995;
10:1061-1067.
17. Inoue H, Kondoch G, Kamakura CR, Yutsudo M, Hakura A. Progression of rat embryo
fibroblast cells immortalized with transforming genes of human papillomavirus type 16.
Virology 1991; 180:191-198.
18. Hurlin PJ, Kaur P, Smith PP, Perez-Reyes N, Blanton RA, Mc Dougall JK. Progression of
human papillomavirus type 18-immortalized human keratinocytes to a malignant pheno-
type. Proc Natl Acad Sci USA 1991; 88:570-574.
19. Munger K, Phelps WC, Bubb V, Howley PM, Schlegel R. The E6 and E7 genes of the
human papillomavirus type 16 together are necessary and sufficient for transformation of
primary human keratinocytes. J Virol 1989; 63:4417-4421.
20. Crook T, Storey A, Almond N, Osborn K, Crawford L. Human papillomavirus type 16
cooperates with activated ras and fos oncogenes in the hormonally dependent transforma-
tion of primary mouse cells. PNAS 1988; 85:8820-8824.
21. Sedman SA, Barbosa MS, Vass WC et al. The full-length E6 protein of human
papillomavirus type 16 has transforming and trans-activating activities and cooperates with
E7 to immortalize keratinocytes in culture. J Virol 1991; 65:4860-4866.
22. Myers F, Androphy E. The E6 protein. In: Myers G, Bernard HU, Delius H, Baker C,
Icenogel J, Halpern A, eds. Human Papillomaviruses. Los Alamos NM:Los Alamos Na-
tional Laboratories, 1995:III-47—III-57.
23. Scheffner M, Werness BA, Huibregtse JM, Levine AJ, Howley PM. The E6 oncoprotein
encoded by human papillomavirus types 16 and 18 promotes the degradation of p53. Cell
1990; 63:1129-1136.
24. Dyson N, Howley PM, Munger K, Harlow E. Human papillomavirus-16 E7 oncoprotein is
able to bind to the retinoblastoma gene product. Science 1989; 243:934-936.
25. Hubbert NL, Sedman SA, Shiller JT. Human papillomavirus type 16 E6 increases the deg-
radation rate of p53 in human keratinocytes. J Virol 1992; 66:6237-6241.
26. Band V, Dalal S, Delmolino L, Androphy EJ. Enhanced degredation of p53 protein in HPV-
6 and BPV-1 E6-immortalized human mammary epithelial cells. EMBO J 1993;
12:1847-1852.
27. Nevins JR, Vogt PK. Cell transformation by viruses. In: Fields BN, Knipe DM, Howley
PM, eds. Fundamental Virology. Philadelphia PA: Lippincott-Raven, 1996:267-310.
28. Liu TJ, El-Naggar AK, McDonnell TJ et al. Apoptosis induction mediated by wild-type p53
adenoviral gene transfer in squamous cell carcinoma of the head and neck. Cancer Res
1995; 55:3117-3122.
29. Johnson MA, Blomfield PI, Bevan IS, Woodman CBJ, Young LS. Analysis of human
papillomavirus type 16 E6-E7 transcription in cervical carcinomas and normal cervical
epithelium using the polymerase chain reaction. J Gen Virol 1990; 71:1473-1479.
30. Bernard BA, Bailly C, Lenoir MC, Darmon M, Thierry F, Yaniv M. The human
papillomavirus type 18 (HPV18) E2 gene product is a repressor of the HPV18 regulatory
region in human keratinocytes. J Virol 1989; 63:4217-4324.
31. Bouvard V, Storey A, Pim D, Banks L. Characterization of the human papillomavirus E2
protein: evidence of trans-activation and trans-repression in cervical keratinocytes. EMBO
J 1995; 13:5451-5459.
32. Schneider-Maunoury S, Croissant O, Orth G. Integration of human papillomavirus type
16 DNA sequences: A possible early event in the progression of genital tumors. J Virol
1987; 61:3295-3298.
33. DiLorenzo TP, Tamsen A, Abramson AL, Steinberg BM. Human papillomavirus type 6a
DNA in the lung carcinoma of a patient with recurrent laryngeal papillomatosis is charac-
terized by a partial duplication. J Gen Virol 1992; 73:423-428.
34. May M, Dong X, Beyer-Finkler E, Stubenrauch F, Fuchs PG, Pfister H. The E6/E7 pro-
moter of extrachromosomal HPV16 DNA in cervical cancers escapes from cellular repres-
sion by mutation of target sequences for YY1. EMBO J 1994; 13:1460-1466.
35. Chen Z, Storthz KA, Shillitoe EJ. Mutations in the long control region of human
papillomavirus DNA in oral cancer cells, and their functional consequences. Cancer Res
1997; 57:1614-1619.
36. Shillitoe EJ, Lapeyre JN, Adler-Storthz K. Gene therapy—its potential in the management
of oral cancer. Oral Oncol-Eur J Cancer B 1994; 30B:143-154.
37. Von Knebel Doeberitz M, Oltersdorf T, Schwarz E, Gissman L. Correlation of modified
human papilloma virus early gene expression with altered growth properties in C4-1 cervi-
cal carcinoma cells. Cancer Res 1988; 48:3780-3786.
38. Von Knebel Doeberitz M, Rittmuller C, zur Hausen H, Durst M. Inhibition of tumorige-
nicity of cervical cancer cells in nude mice by HPV E6-E7 anti-sense RNA. Int J Cancer
1992; 51:831-834.
39. Steele C, Cowsert LM, Shillitoe EJ. The effects of human papillomavirus type-18-specific
antisense oligonucleotides on the transformed phenotype of human carcinoma cell lines.
Cancer Res 1993; 53:2330-2337.
40. Steele C, Sacks PG, Adler-Storthz K, Shillitoe EJ. Effect on cancer cells of plasmids that
express antisense RNA of human papillomavirus type 18. Cancer Res 1992; 52:4706-4711.
41. Tan TMC, Ting RCY. In vitro and in vivo inhibition of human papillomavirus type 16 E6
and E7 genes. Cancer Res 1995; 55:4599-4605.
42. Lappalainen K, Pirila L, Jaaskelainen I, Syrjanen K, Syrjanen S. Effects of liposomal antisense
oligonucleotides on mRNA and protein levels of the HPV 16 E7 oncogene. Anticancer Res
1996; 16:2485-2492.
43. Hu G, Liu W, Mendelsohn J et al. Expression of epidermal growth factor receptor and
human papillomavirus E6/E7 proteins in cervical carcinoma cells. J Natl Cancer Inst 1997;
89:1271-1276.
44. Hu G, Liu W, Hanania EG, Fu S, Wang T, Deisseroth AB. Suppression of tumorigenesis
by transcription units expressing the antisense E6 and E7 messenger RNA (mRNA) for the
transforming proteins of the human papillomavirus and the sense mRNA for the retino-
blastoma gene in cervical carcinoma cells. Cancer Gene Ther 1995; 2:19-32.
45. Madrigal M, Janicek MF, Sevin BU et al. In vitro antigene therapy targeting HPV-16 E6
and E7 in cervical carcinoma. Gynecol Oncol 1997; 64:18-25.
46. Storey A, Oates D, Banks L, Crawford L, Crook T. Anti-sense phosphorothioate oligo-
nucleotides have both specific and non-specific effects on cells containing human
papillomavirus type 16. Nucl Acids Res 1991; 19:4109-4114.
47. Stein CA. Phosphorothioate antisense oligodeoxynucleotides: questions of specificity. Trends
Biotechnol 1996; 14:147-149.
48. Wisotzkey JD, Krizenoskas A, DiAngelo S, Kreider JW. Cleavage of cottontail rabbit
papillomavirus E7 RNA with an anti-E7 ribozyme. Biochem Biophys Res Commun 1993;
192:833-839.
49. He YK, Lu CD, Qi GR. In vitro cleavage of HPV16 E6 and E7 RNA fragments by synthetic
ribozymes and transcribed ribozymes from RNA-trimming plasmids. Fed Eur Biochem Soc
1993; 322:21-24.
50. Lu D, Chatterjee S, Brar D, Wong KK. Ribozyme-mediated in vitro cleavage of transcripts
arising from the major transforming genes of human papillomavirus type 16. Cancer Gen
Ther 1994; 1:267-277.
51. Chen Z, Kamath P, Zhang S, Weil MM, Shillitoe EJ. Effectiveness of three ribozymes for
cleavage of an RNA transcript from human papillomavirus type-18. Cancer Gen Ther 1995;
2:263-271.
52. Chen Z, Kamath P, Zhang S, St John L, Adler Storthz K, Shillitoe EJ. Effects on tumor
cells of ribozymes that cleave the RNA transcripts of human papillomavirus type 18. Can-
cer Gene Ther 1996; 3:18-23.
53. Huang Y, Kong Y, Wang Y, Qi G, Lu C. Stable expression of anti-HPV 16 E7-ribozyme in
CV-1 cell lines. Chinese J Biotechnol 1996; 12:215-220.
54. Shillitoe EJ, Hinckle CC, Noonan S, Marini F, Kellman R. Use of particle bombardment
for transduction of normal and malignant oral epithelium. Cance Gene Ther 1997;
5:176-182.
55. Zhou SZ, Li Q, Stamatoyannopoulos G, Srivastava A. Adeno-associated virus 2-mediated
transduction and erythroid cell-specific expression of a human beta-globin gene. Gene Ther
1996; 3:223-229.
ribozymes expressed from a tRNA-based RNA polymerase III promoter. Nucl Acids Res
1995; 23:2259-2268.
58. Dropulic B, Lin NH, Martin MA, Jeang KT. Functional characterization of a U5 ribozyme:
Intracellular suppression of human immunodeficiency virus type 1 expression. J Virol 1992;
66:1432-1441.
59. Heusch M, Kraus G, Johnson P, Wong-Staal F. Intracellular immunization against SIVmac
utilizing a hairpin ribozyme. Virology 1996; 216:241-244.
60. Li M, Lonial H, Citarella R, Lindh D, Colina L, Kramer R. Tumor inhibitory activity of
anti-ras ribozymes delivered by retroviral gene transfer. Cancer Gene Ther 1996; 3:221-229.
61. Shore SK, Nabissa PM, Reddy EP. Ribozyme-mediated cleavage of the BCRABL oncogene
transcript: In vitro cleavage of RNA and in vivo loss of P210 protein-kinase activity.
Oncogene 1993; 8:3183-3188.
62. Lieber A, Kay MA. Adenovirus-mediated expression of ribozymes in mice. J Virol 1996;
70:3153-3158.
63. Hayashi S, Nagasaka T, Katayama A et al. Adenovirus-mediated gene transfer of antisense
ribozyme for alpha (1,3)galactosyltransferase gene and alpha (1,2)fucosyltransferase gene
in xenotransplantation. Transplant Proc 1997; 29:2213-.
64. Lieber A, He CY, Polyak SJ, Gretch DR, Barr D, Kay MA. Elimination of hepatitis C virus
RNA in infected human hepatocytes by adenovirus-mediated expression of ribozymes. J
Virol 1996; 70:8782-8791.
by high efficiency adenoviral-mediated delivery of an anti-ras ribozyme. Cancer Res. 1995;
55:2024-2028.
66. Clayman GL, Trapnell BC, Mittereder N et al. Transduction of normal and malignant oral
epithelium by an adenovirus vector: The effect of dose and treatment time on transduc-
tion efficiency and tissue penetration. Cancer Gen Ther 1995; 2:105-111.
67. Mitchell MF, Hamada K, Sastry KJ et al. Transgene expression in the rhesus cervix medi-
ated by an adenovirus expressing beta-galactosidase. Am J Obstet Gynecol 1996;
174:1094-1101.
68. Ohta Y, Kijima H, Ohkawa T, Kashani-Sabet M, Scanlon KJ. Tissue-specific expression of
an anti-ras ribozyme inhibits proliferation of human malignant melanoma cells. Nucl Ac-
ids Res 1996; 24:938-942.
69. Garver RJ, Goldsmith KT, Rodu B, Hu PC, Sorscher EJ, Curiel DT. Strategy for achieving
selective killing of carcinomas. Gene Ther 1994; 1:46-50.
70. Schneider-Gadicke A, Schwarz E. Different human cervical carcinoma cell lines show simi-
lar transcription patterns of human papillomavirus type 18 early genes. EMBO J 1986;
5:2285-2292.
Index
A D
Adeno-associated virus (AAV) 95, 101-118, Dimeric minizyme 62-74
156, 162, 193, 212, 231
Adenoviral vector 101, 102, 129-131, 137, E
144, 148, 151, 155, 158-160, 191
Adenovirus 43, 95, 101-114, 129-131, 137, Electroporation 45, 136, 156, 157, 206
138, 143-147, 156-158, 160, 178, 199- Endonuclease 8, 9, 176, 218
201, 208, 215, 225 Exonucleases 214
Antisense 4, 10, 24, 27, 34, 41, 42, 44, 45, 47,
51, 52, 62, 67, 69, 70, 81, 83, 85-87, 91,
92, 98, 99, 107, 129, 140, 141, 143, 145,
F
148, 149, 156, 158, 159, 160, 162, 169, Folate-polylysine 43, 205
172, 173, 177, 179, 180, 183, 192, 193, Folding 3, 5, 17, 19, 25, 26, 203, 208
200, 201, 203, 206, 208, 209-211, 215,
216, 220, 221, 223, 227, 229, 230, 232
Apoptosis 126, 138, 155, 160, 166, 168, 177, G
195, 218, 219, 220 G418 90, 105, 156, 157
Asymmetric hammerhead ribozyme 8-10 Glioma 9, 114, 116, 218
Glyceraldehyde-3-phosphate dehydrogenase
B (GAPDH) 9, 215-218
β-actin promoter 129, 137, 179, 180
BCR-ABL 61, 62, 65-71, 73, 74, 177, 195, 199 H
Bioconjugates 51 Hairpin ribozyme 3, 15-21, 93, 135, 206
Bioerodible polymer 54 Helper virus 23, 90, 102, 103, 109
Bladder cancer 125-131, 135, 178 HIV 88, 91-97, 105, 108, 110, 116, 165, 214,
Breast cancer 55, 135-139 223, 228
BT-474 cell 137, 138 Host range 95, 102, 112
C I
C-erbB-2 135-139, 151, 152, 153 Interleukin-2 (IL-2) 96, 97, 165, 166, 214-
Catalytic activity 4, 8, 9, 17, 24, 26-28, 30, 32, 216, 218
34, 62, 144, 146, 153, 158, 159, 179, 180, Intraarticular 45-48, 50
184, 203 Intraocular 47
Catalytic RNA 3, 11, 15, 30, 32, 34, 79, 101, Intravenous 9, 49-52, 55, 112
135, 144, 152, 153, 165, 175, 176, 178, Iontophoresis 49
179, 186
Cationic lipid 41-45, 52, 54, 156, 188, 190,
208, 220 K
CD34 97, 102, 113, 116
Kinetic analysis 74
Cell culture 30, 34, 41, 43, 44, 47, 94, 97, 157,
161, 184, 199
Cervical cancer 223, 225-229 L
Chronic myelogenous leukemia (CML) 61,
62, 69, 74, 195, 196, 199, 200, 205, 208, L6 61, 62, 65-67, 69, 74, 87, 90, 94, 114
209 Liposome 8, 44, 52-55, 74, 156, 158, 170, 198,
Cis-acting ribozyme 166 200, 201, 203, 205, 206, 208, 209
Clinical trial 95-98, 113, 129, 130, 136, 144, Lung cancer 151-153, 155-158, 179
165, 166, 191
M Structure 5, 6, 10, 11, 15-21, 23-32, 46, 62-65,

67, 72-74, 91, 92, 94, 105, 116, 153, 158,
Metal ion requirement 5 159, 167, 175-177, 183-186, 201, 203,
Minizyme 8, 62-74, 159, 160 206, 214, 216, 217, 220, 225, 227-229
Multidrug resistance (MDR) 177, 183, 184, Sustained release 47, 55
187, 188, 190, 191
T
N
Tat 91, 93-95, 116
Nanoparticle 42, 47, 49 Tissue-specific expression 178
Titer 87, 90, 92, 104, 107-111, 114, 137, 156,
O 159, 188
TNF-α 215, 216, 220
Oncogene. See BCR-ABL; C-erbB-2; Ras Transduction 90, 95-97, 102, 104, 109,
Oral cancer 223, 224, 226, 228, 229 112-114, 116, 126, 129, 131, 137, 138,
144, 155, 165, 177, 178, 188, 195, 208,
P 209, 228
Transitional cell carcinoma 125
P-glycoprotein 184-189 Triple ribozyme 166, 168-170
p53 89, 126, 127, 138, 151-153, 155, 160, 166, tRNA 3, 69, 71-74, 92, 105, 108, 110-112,
179, 180, 184, 224, 228 116, 166, 199, 200, 206, 207
Packaging 87-92, 94, 95, 104, 105, 107-109, tRNAVal promoter 71-74, 93
116, 144, 158, 159, 201, 208 Tumor suppressor gene 89, 126, 128, 131,
Papillomavirus 180, 223-227, 229 138, 151-153, 160, 166, 176, 179, 180
PHb Apr-1-neo 129, 151, 156
Polycations 42, 43 V
Probasin 169, 170
Promoter 69, 71, 73, 74, 80, 84, 87-91, 93, Vascular endothelial growth factor (VEGF)
101-112, 114, 116, 129, 137, 145, 147, 41, 45-47, 126, 128, 220
157, 158, 160, 168, 169, 178-180, 188,
206, 208, 224, 228, 229
Prostate cancer 165, 166, 168-170
Protein kinase C (PKC) 218, 219
R
Ras (ras) 126-131, 135, 136, 143-147,
151-153, 155-160, 175-178, 180, 184,
195, 228, 229
H-Ras 151
K-Ras 151, 152
Required bases 15
Retroviral vector 87, 89-93, 95-97, 101, 116,
155, 178-180, 199, 206, 208, 209
Ribozyme transfection 200
RNA polymerase I 87, 166-170
RNA polymerase II 23, 24, 104, 105, 188, 208
RNA polymerase III 69, 71, 74, 87, 104, 105,
166, 188
RT-PCR 81, 82, 156, 199, 205-207
S
Southern blot 205, 207
COMPANY
R.G.LANDES
Genetic Mechanisms in Multiple Endocrine Neoplasia Type 2 c-Myc Function in Neoplasia
MEDICAL INTELLIGENCE UNIT 4
Barry D. Nelkin, Johns Hopkins University Chi V. Dang and Linda A. Lee, Johns Hopkins University
Functional Heterogeneity of Liver Tissue: From Cell Lineage Endothelins
Diversity to Sublobular Compartment-Specific Pathogenesis David J. Webb and Gillian Gray, University of Edinburgh
Fernando Vidal-Vanaclocha, Universidad del Pais Vasco
SCANLON • KASHANI-SABET
p53 B Cells and Autoimmunity
Host Response to Intracellular Pathogens Christian Boitard, Hôpital Necker-Paris
Stefan H.E. Kaufmann, Institute für Mikrobiologie und
Immunologie der Universität Ulm Hyperacute Xenograft Rejection Kevin J. Scanlon • Mohammed Kashani-Sabet
Jeffrey Platt, Duke University
Myocardial Injury: Laboratory Diagnosis
Johannes Mair and Bernd Puschendorf, Universität Innsbruck Transplantation Tolerance
J. Wesley Alexander, University of Cincinnati
Cellular Inter-Relationships in the Pancreas: Implications for
Islet Transplantation
Lawrence Rosenberg and William P. Duguid, McGill University
Anti-HIV Nucleosides: Past, Present and Future

Hiroaki Mitsuya, National Cancer Institute
Premalignancy and Tumor Dormancy
Eitan Yefenof, Hebrew University - Hadassah Medical School
Richard H. Scheuerman, University of Texas Southwestern
Myocardial Preconditioning
Therapy of Cancer
Cherry L. Wainwright and James R. Parratt,
Heat Shock Response and Organ Preservation University of Strathclyde
George Perdrizet, University of Connecticut
Cytokines and Inflammatory Bowel Disease
Glycoproteins and Human Disease Claudio Fiocchi, Case Western Reserve
Inka Brockhausen, Hospital for Sick Children—Toronto
Bone Metastasis
Exercise Immunology F. William Orr and Gurmit Singh, University of Manitoba
Bente Klarlund Pedersen, Rigshospitalet—Copenhagen
Cancer Cell Adhesion and Tumor Invasion
Chromosomes and Genes in Acute Lymphoblastic Leukemia
Lorna M. Secker-Walker, Royal Free Hospital-London
Pnina Brodt, McGill University
MIU
Surfactant in Lung Injury and Lung Transplantation
James F. Lewis, Lawson Research Institute
Cutaneous Leishmaniasis
Felix J. Tapia, Instituto de Medicina-Caracas 4
Richard J. Novick, Roberts Research Institute Molecular Basis of Autoimmune Hepatitis
Ruud A.W. Veldhuizen, Lawson Research Institute Ian G. McFarlane and Roger Williams, King’s College Hospital

Gamma Interferon in Antiviral Defense Estrogen and Breast Cancer
Gunasegaran Karupiah, The John Curtin School of Medical W.R. Miller, University of Edinburgh
Research—The Australian National University Molecular Mechanisms of Hypercoagulable States
Management of Post-Open Heart Bleeding Andrew I. Schafer, University of Texas-Houston
Rephael Mohr, Jacob Lavee and Daniel A. Goor, Organ Procurement and Preservation for Transplantation
The Chaim Sheba Medical Center Luis Toledo-Pereyra, Michigan State University
The Glycation Hypothesis of Atherosclerosis Liver Stem Cells
Camilo A.L.S. Colaco, Qandrant Research Institute Stewart Sell and Zoran Ilic, Albany Medical College
Cytokines in Reproduction: Molecular Mechanisms of Fetal Skin Substitute Production by Tissue Engineering
Allograft Survival Mahmoud Rouabhia, Laval University
Gary W. Wood, University of Kansas
HIV and Membrane Receptors
Computers in Clinical Medicine Dimitri Dimitrov, National Institutes of Health
Eta Berner, University of Alabama Christopher C. Broder, Uniformed Servives University of the
Cellular & Molecular Biology of Airway Chemoreceptors Health Sciences
Ernest Cutz, University of Toronto Interferon-Inducible Genes
Immunology of Pregnancy Maintenance in the First Trimester Ganes Sen and Richard Ransohoff, Case Western Reserve
Joseph Hill, Harvard University University
Peter Johnson, University of Liverpool Artificial Neural Networks in Medicine
Inherited basement Membrane Disorders Vanya Gant and R. Dybowski, St. Thomas Medical School—
Karl Tryggvaso, Karolinska Institute London
Cartoid Body Chemoreceptors von Willebrand Factor

Constancio Gonzalez, Universidad de Madrid Zaverio M. Ruggeri, Scripps Research Institute
Molecular Biology of Leukocyte Chemostasis Immune Mechanisms in Atherogenesis

Antal Rot, Sandoz Forschungsinstitut—Vienna Ming K. Heng, UCLA
Breast Cancer Screening The Biology of Germinal Centers in Lymphoral Tissue

Ismail Jatoi, Brook Army Medical Center G.J. Thorbecke and V.K. Tsiagbe, New York University
R.G. LANDES R.G. LANDES

C OM PA N Y C OM PA N Y

Ribozymes in The Gene

Uploaded by

Copyright:

Available Formats

Ribozymes in The Gene

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ribozymes in The Gene

Uploaded by

Copyright:

Available Formats

COMPANY

Anti-HIV Nucleosides: Past, Present and Future

Ribozymes in the Gene Therapy of Cancer

Cartoid Body Chemoreceptors von Willebrand Factor

Molecular Biology of Leukocyte Chemostasis Immune Mechanisms in Atherogenesis

Breast Cancer Screening The Biology of Germinal Centers in Lymphoral Tissue

R.G. LANDES R.G. LANDES

Ribozymes in the Gene

Kevin J. Scanlon, Ph.D.

Mohammed Kashani-Sabet, M.D.

R.G. LANDES COMPANY

Ribozymes in gene therapy of cancer / [edited by] Kevin J. Scanlon, Mohammed

Our goal is to publish books in important and rapidly changing

Section II: Expression and Delivery of Ribozymes ......................................... 39

6. Using Ribozymes to Attenuate Gene Expression

7. Retroviral Delivery of Ribozymes .......................................................... 87

9. Applications of Anti-Oncogene Ribozymes for the Treatment

10. Gene Therapy of Breast Cancer ............................................................ 135

13. Ribozymes in Gene Therapy of Prostate Cancer ................................. 165

17. Potential Design and Facilitation of Hammerhead Ribozyme

Mohammed Kashani-Sabet, M.D.

Lance H. Leopold, M.D. Jack A. Roth, M.D.

Mohammed Kashani-Sabet, M.D.

The Hammerhead Motif

Mutagenesis and Sequence Requirements

Hammerhead Ribozyme Structure

Reaction Mechanism and Kinetics

The hammerhead ribozyme catalytic reaction can be fitted to the Michalis-Menten

Modified Hammerhead Ribozymes

Pharmacokinetic and Cellular Uptake Studies

Making Hammerhead Ribozymes Work In Vivo

Target Site Selection

Hammerhead Ribozyme Design for In Vivo Expression

25. Fu D-J, Benseler F, McLaughlin LW. Hammerhead ribozymes containing non-nucleoside

72. Christoffersen RE, McSwiggen J, Konings D. Application of computational technologies to

Biochemistry of the Hairpin Ribozyme

Biochemistry and Mechanism of the Reaction

Rp phosphorothioate substitution at the cleavage site.17,18 Efficient cleavage of the Sp

Fig. 2.2. Mechanism of catalysis by

Table 2.1. Unimolecular rate

Mg2+ 1.8 ± 0.2 11.4

Biochemistry of Hepatitis Delta Virus

Properties of the Catalytic Domains

Table 3.1. Comparison of trans-cleaving ribozymes

Construct* Cleavage Rate Turnover Conditions

(a) Genomic65 0.022 min–1 ND pH 7.1, 37°C, 1 mM MgCl2

Exogenous Delivery of Ribozymes

H ammerhead ribozymes are trans-acting, RNA-based molecules that specifically cleave

Tissue Culture Studies

Table 4.1. Optimization of LipofectAMINE:ribozyme formulationsa

a) Effect of different culture mediab

b) Effect of different charge ratiosc

2:1 1503 0.364

used having the sequence 5'-ususususcccu GAuGaggccgaaaggccGaaAuucucB-3', where lowercase

been coupled with polylysine formulations to enhance endosome release. A combination of

Localized Delivery In Vivo

receptor inhibited VEGF-stimulated neovascularization in a rat corneal implant model.7 In

Trafficking of Free Ribozyme After Localized Administration