REVIEWS

Protecting Group Strategies in Organic Synthesis

Michael Schelhaas and Herbert Waldmann"
Dedicated to Projessor Leopold Horner on the occasion of his 85th birthday
f

'

The choice of protecting groups is one of

the decisive factors in the successful realization of a complex, demanding synthetic project. The protecting groups
used influence the length and efficiency
of the synthesis and are often responsible for its success or failure. A wide
range of blocking groups are currently
available for the different functional
groups; however, an overall strategy
combining these different masking techniques in an advantageous and reliable

manner has never been proposed or at

best only for individual cases. This review attempts to make a contribution to
filling this gap. First a very short
overview of the most commonly used
protecting groups will be given, in which
they are classified according to their
lability and not according to the functional group they protect. This classification clarifies coherent concepts for the
development of blocking strategies. On
the basis of this brief summary reliable

Today, organic synthesis has reached a remarkable level of

competence and even the most complex molecules are accessible.['] The prerequisites for this success are both the availability
of a wide range of efficient synthetic methods['] and reagents,f3f
and the fact that "retrosynthetic analysis"[41 can provide a
framework for the design of a synthetic strategy leading to the
desired product in the most efficient and logical way.
These strategies include tactics for the construction of the
molecular framework, for the establishment of the absolute configuration of any stereocenters present, for the efficient formation of rings, and for the reduction of the number of synthetic
steps. The complex synthetic intermediates and products contain, in general, a multiplicity of functional groups, most of
which must be blocked and, at an appropriate point in the
synthesis, liberated. The correct choice of protecting groups is
often decisive for the realization of the overall operation. This
is exemplified by the recently published total synthesis of the
mitomycin analog FR-900482. In model studies Danishefsky et
a]. had shown that the basic skeleton of this natural product can
be accessed by FeC1,-mediated opening of the epoxide in 1
(Scheme 1 ).Is1However, the synthesis could not be concluded
successfully. because cleavage of the methyl ether was accompa[*] Prof. Dr. H. Wdldmann. DipLChem. M. Schelhads
lnstitut fur Orgdnische Chemle der Universitdt Karlsruhe
Richard-Willstiitter-Allee2. D-76128 Karlsruhe (Germany)
F a x : I n t . code +(721)608-4825
e-mail . waldmannkr ochhddes.chemie.uni-karlsruhe.de

\T:

Keywords: protecting groups * retrosynthetic analyses synthetic methods total syntheses

1. Introduction

Angrit.. c'hcwi. I n t . Ed. Eiigl 1996. 35. 2056-2083

strategies will then be illustrated and

developed with selected examples from
the recent literature by which protecting
groups may be combined successfully
and advantageously in synthetic projects
of differing degrees of complexity and
difficulty.

"

Me02C

mMe0

N-CO,Me

MeO&

CHO

"

N-C02Me

OMOM

!???#.+

Me02C

2
Scheme 1. Protecting group problems in the synthesis of mitomycin analog
FR-900482 by S. Danishefsky et al. [ 5 ] . MOM = methoxymethyl, Bzl = benzyl.

nied by decomposition of the aziridine. To avoid this problematic situation, the hydroxyl groups in 2 were not protected as
methyl ethers but rather as benzyl and methoxymethyl ethers.
Unfortunately, after this seemingly innocuous"I alteration in
the protecting group pattern, the selective FeC1,-mediated
cleavage of the oxirane failed owing to the higher lability of the
newly introduced ether functions. The authors were forced to
find an entirely new reagent to complete the synthesis.'6]
Another unexpected protecting group problem appeared during the total synthesis of taxol, again by Danishefsky et aL1'] The
advanced intermediate 3 was converted into the epoxide 4, but
the robust tot-butyldimethylsilyl (TBDMS) ether could not be
removed in the next steps (Scheme 2). Thus the TBDMS in 3
was replaced by the more labile triethylsilyl (TES) ether. Compound 5 was converted into the corresponding epoxide 6, which
was then successfully transformed into the complex natural

VCH Ver.lu~sgesellsr/7u/tinhH. 0-69451 Weinheim, 1996

0570-0833:Y6~3518-2057$ I S OO+ .25/0

2057

M . Schelhaas and H . Waldinann

REVIEWS

OBzl

OBzl

iEtL
0
OAc

OBzI

>

0
'

Schcmc 2 . I'l-otcc[tn~group prohieins ~n :lie \ ~ i i i h c s t s01'

c: a1 171.

tiixol

h> S l);inlslieS\k!

product. The TES group was removed only after construction of

the correctly functionalized polycycle.
These examples make it obvious that a successful synthesis
requires both basic retrosynthetic planning and a separate protecting group strategy that takes into account different requirements such as the lability of the intermediates and of the reagents
to be used.
As a consequence of the great importance of protecting
groups in organic chemistry. a multitude of blocking techniques
have been developed for a wide range of functional groups. Here
the chemist has numerous choices. However. when giving ;I
detailed description of a successfully completed total synthesis.
authors rarely comment on why they selected particular protecting group patterns.[81 Similarly. in the monographs"). '01 and
reviews" ' I concerning protecting group chemistry, the emphasis lies in the presentation of the various possibilities that exist
for the blocking and deprotection of the function i n question.
Strategies that can be used to combine protecting groups
in appropriate ways and that have proved their capability and
reliability in complex syntheses have never been published or

only for isolated

The protecting groups will be classificd according t o their lability (for ii inore comprehensive treatincnt sec rets. [9 1 I ] ) instead according to the functional group
they block. This has the advantage that the sensitivity of thc
compounds to be protected and the required reaction conditions
can be accountcd for in the planning of 21 synthesis.
Experience shows that the critical paramcters ;ire generally
the stability and thecleavage ofthe protecting group rather than
its introduction. For most of the typically required functional
groups, protecting groups are known that are labile under different. often alternative. conditions. Furthermore. unified concepts for the development of new blocking possibilities become
clear a s it consequence of this approach. As an example, the
N-fri.t-butyloxycarbony1 (Boc) gro~ip.the trrt-butyl ester. and
the twt-butyl ether can be cleaved under acidic conditions. the
trimethylsilylethoxycarbonyl (Tcoc) group and the trimethylsilylethyl (TMSE) ether are cleaved by fluoride ions through
/Mimination. and allyloxycarbonyl (Aloc) urethane, ally1 (All)
ester. ally1 ether, and 5-methylene-1.3-dioxane acetal,['3Jwhich
is used for the protection ofcarbony1 groups, can all be removed
by noble metal catalyzed reactions. Following this short
overview. selected examples from the irecent literature will be
used to illustrate proven strategies for the successful and advantageous combination of protecting groups in synthetic projects
of varying complexity and difficulty. This review should provide
simple guidelines for the the development of successful tactics
and strategies foi- organic synthesis.

2. Lability and Cleavage of Protecting Groups

For 21 protecting group to find wide application in organic
synthesis, it must fulfill several criteria. In particular. it must
be introduced into the molecule to be protected under mild
conditions in a selective manner and in high yield; functional
groups other than that to be protected must not be attacked.

REVIEWS

Protecting Group Chemistry

be stable under all the conditions used during the synthesis,
including rhose of the purification steps, up to the step in
which the protecting group is removed, it should, as Par as
possible, have a stabilizing effect on the molecule and should
suppress racemization or epimerization.
be cleavable under very mild conditions in a highly selective
manner and in high yield; other protecting groups present in
the molecule and unprotected functionalities should not be
affected by the cleavage conditions.

Protecting Groups for Diols

n=1,2

isopropylidene

cyclopentylidene

cyclohexylidene

In addition to these minimum requirements, the protecting

group should also
~

be introduced and removed with the help of readily available

reagents. such that in both transformations the products can
be easily purified.
introduce no additional stereocenters.
lend the protected intermediates advantageous physical properties; for example the compounds should be easily crystallized and/or readily soluble.

Only a few protecting groups meet all of these demands, and

in most cases a compromise must be found, in which the most
important criteria are addressed. In most cases guaranteeing that
the protecting group is very stable and, at the same time, readily
liberated (an apparent contradiction) is the crucial problem and
overshadows the requirements for efficient introduction and the
provision of desirable physical and chemical properties. In the
next sections the most important protecting groups will be presented briefly according to their lability and their ease of cleavage.

2.1. Acid-Labile Protecting Groups['-

benzylidene

pmethoxybenzylidene

opdirnethoxybenzylidene

Protecting Groups for Carbonyls

0
x
R R'

O x 0

R R '

1,3-dioxane

1,3-dioxolane

Scheme 3. Examples of cyclic 0.0-acetals as protecting groups for 1.2-diols,

1.3-diols. and carbonyl compounds.

the methoxymethyl (MOM), the benzyloxymethyl (BOM), and

the methoxyethoxymethyl (MEM) protecting groups (Scheme 4).
They can be cleaved readily under mildly acid conditions, and
their complexing ability can also be used to control the stereochemical course of a transformation (see Section 3.5).

The cleavage of protecting groups by acid-mediated hydrolysis is one of the best established methods in protecting group
chemistry and forms one of the central pillars of the subject.
Nevertheless. only those protecting groups that can be removed
under sufficiently mild (but not too mild) conditions find widespread use.

2.1.1. Acetal Protecting Groups

Acetals are formed and cleaved under acid catalysis. They are
generally stable towards nucleophiles and bases, and are thus
suitable for the protection of alcohols, diols, and carbonyl compounds. Furthermore, N,N- and N,O-acetals are utilized for the
protection of amino groups.
Cyclic 0,O-acetals can be considered as protecting groups for
either carbonyl groups or diols, depending on which component
is of interest. Carbonyl groups are normally protected as the
1,3-dioxolane or -dioxane by condensation with ethylene glycol
or propylene glycol (Scheme 3). Similarly, 1.2- and 1.3-diols are
masked by reaction with acetone, (substituted) benzaldehyde,
and cyclic ketones (cyclopentanone, cyclohexanone, cycloheptanone) as rsopropylidene, (substituted) benzylidene, and cycloalkylidene acetals. Recently, the dispiroketal (DISPOKE)1'41
and the cyclohexane-I ,2-diacetal (CDA) groups1151have been
proposed as two promising new acetal protecting groups (see
Section 5 ) .
Acyclic 0.0-acetals are normally used for the protection of
alcohols. Most commonly used are the tetrahydropyranyl (THP),

tetrahydropyranyl
(ThP)

rnethoxymethyl

(MOM)

methoxyethoxyrnethyl
(MEW

benzyloxyrnethyl

rnethylthiomethyl

(BOW

(MTM)

Scheme 4. Examples of acyclic 0.0-acetals and 0.S-acetals ns protecting groups

for alcohols.

2.1.2. Formation of Stabilized Cations

The second large class of acid-labile protecting groups is characterized by the formation of stabilized cations upon cleavage.
The tert-butyl ether along with the corresponding ester and
urethane (Boc protecting group) are examples of this class,
which find use in the protection of alcohols, thiols, amines, and
carboxylic acids (Scheme 5 ) . In each case, their cleavage liberates the tert-butyl cation. By analogy, benzyl cations are generated on the cleavage of benzyl protecting groups, which are also
common blocking groups for hydroxyl groups, thiols, esters,
and amino groups. By introduction of additional substituents,

2059

REVIEWS

M. Schelhaas and H. Waldmann

pGiq

horvaeetate can be removed selectively in the presence of acetate, and acetate can be cleaved in the presence of pivalate.
Basic cleavage of amides to liberate amines is only seldom
used on account of the generally harsh conditions required.
The phthaloyl group is one exception; it can be cleaved with
hydrazine under mildly basic conditions. Again, it should be
noted that the silyl ethers can be removed by basic hydrolysis
(however, see Section 2.3).

X=O,NH
PG =

felt-butyl
0 Bu)

felt-butyloxycarbonyl

adarnantyloxycarbonyl
(Adoc)

tO K /
NO*
benzyloxycarbonyl
(Z, Cbz)

para-methoxybenzyloxycarbonyl

para-nitrobenzyloxycarbonyl

(MOZ)

(4-NO,-Z)

Scheme 5. Examples of acid-labile protecting groups that are cleaved with formation of stabilized cations.

the stability of the cation can be increased or reduced, and

thereby the acid sensitivity of the protecting group can be finetuned. Thus, the adamantyloxycarbonyl (Adoc) and the paramethoxybenzyloxycarbonyl (MOZ) groups are more acid-labile
than the respective parent Boc and Z groups. The 4-N02-Z
group is, however, significantly more stable than the Z-urethane. It should be noted that the silyl ether protecting groups,
which are usually removed by treatment with fluoride ion (see
Section 2.3), are also labile under mild acidic conditions.

2.2.2. Cleavage by Base-Induced 8-Elimination

Protecting groups such as the fluorenylmethoxycarbonyl
(Fmoc) urethane, which has established itself as one of the standard protecting functions for amino groups in solution-phase
and solid-phase peptide synthesis, and the phenylsulfonylethyl
group are cleaved by the abstraction of an acidic proton and
subsequent 8-elimination, with the formation of a vinyl system
(Scheme 7). Direct attack of the base on a carbonyl function is
thus avoided. This type of protecting group has found many
applications.

X=O,NH
PG =

9-fluorenylmethoxycarbonyl

2-(phenylsulfonyl)ethoxycarbonyl

(Fmoc)

2.2. Base-Labile Protecting GroupsTg-''I

The removal of protecting functions under basic conditions is
also one of the tried-and-true methods in protecting group
chemistry. On the basis of mechanistic considerations, two categories can be distinguished : basic hydrolysis and basedinduced p-elimination.

2.2.1. Basic Hydrolysis

This method applies to practically all esters, with the exception of the tert-butyl ester mentioned in Section 2.1.2. Carboxylic acids are typically protected as (substituted) alkyl esters.
Alcohols are often esterified with acetic acid, benzoic acid, or
pivalic acid (Scheme 6). The rate of hydrolysis can be modified
by adjusting steric and electronic factors. For example, tri-

R-0-PG

PG =

acetyl

benzoyl
(Bz)

pivaloyl
(Piv)

Scheme 6. Examples of base-labile acyl protecting groups.

2060

Scheme 7. Examples of protecting groups that are cleaved by a-elimination

2.3. Fluoride-Labile Protecting Groups['

'I

Silyl protecting groups can be cleaved by treatment with fluoride ion under conditions that affect nearly no other functionalities. They thus make up a n essential chapter in protecting
group chemistry (Scheme 8). Variation of the substituents on
silicon allows modification of their stability towards acids and
bases, as well as the selectivity of the cleavage with fluoride ion
(see Section 3.2.2 and Scheme 37). A simple rule of thumb
holds-the greater the steric demand, the higher the stability.
Alcohols are typically protected as trialkylsilyl ethers; the order
of stability for the ethers is TMS < TES < TBDMS < TIPS[161
z thexyl. For particularly challenging cases, a further level of
fine-tuning can be achieved with, for example, isopropyldimethylsilyl,[171 diisopropylmethylsilyl,['81 and diethylisopropylsilyl (DEIPS)[191groups. I,3-Diols can be masked as
silanediyl derivatives (see also Scheme 44) or by introduction of
the tetraisopropyldisiloxane-l,3-diyl(TIPDS) group. Silyl groups
are rarely used for the protection of esters and amines due to the
high lability of the resulting derivatives (the STABASE protecting group[201for amines is an exception, see Scheme 8). For
this purpose, trialkylsilylethyl esters and carbamates have been
Angew. Chenr. In!. Ed. Engl. 1996. 35, 2056-2083

Protecting Gram Chemistry

REVIEWS

R', R', R3 = Me: TMS

R ' , R', R3 = Et: TES
R', R', R3 = i Pr: TIPS
R' = t Bu, R', R3 = Me: TBDMS
R'

R-X

X=O,NH

= peptide, nucleoside,

nucleotide. carbohvdrate.
plactarn antibiotic [2'-231
1

IDenicillin G acvlase I

t Bu, R', R3 = Ph: TBDPS

0
R.oz/Si,

Me
1 ,Me

peptide, lipopeptide P41

peptide, lipopeptideI'[

Me
trirnethylsilylethox ycarbonyl
(Teoc)
X=NH,O

trirnethylsilylethyl
(TMSE)

trirnethylsilylethoxyrnethyl
( S W

acetyl esterase

0
0
RKO-~(~e)3

I
0
~r

butyrylcholine esterase
M,e ,Me
,Si

R-N\

Si
I \
Me Me

RKO-

R = pepttde, glycopeptide

llipasej

STABASE

di-t-butylsilanedryl
(DTBS)

1 ,I ,3,3-tetraisopropyldisiloxane-l,3-diyl
(TIPDS)

R = peptide, glycopeptide P7I

R = carbohydrate [212']
I

Scheme 8. Examples of protecting groups that are cleaved with fluoride ions
R'OKMe

developed, which can be removed by fluoride ion promoted

fragmentation analogous to the previously described p-elimination (Section 2.2.2; Scheme 8). The same principle is also behind
the trimethylsilylethyl (TMSE) and trimethylsilylethoxymethyl
(SEM) protecting groups for alcohols.

2.4. Enzyme-Labile Protecting Groups"'.

"I

In many cases the selective removal of different acyl protecting groups from amines and alcohols and the targeted deblocking of carboxylic acids under mild conditions is most readily
achieved with biocatalysts. Enzymes typically function in the
p H range 5 - 9 and at room temperature; they may display a
high specificity for the structures they recognize and the reactions they catalyze, and at the same time they may tolerate a
wide range of substrates. Thus enzymes enable the targeted removal of protecting groups (which may in principle also be
removed by classical chemical methods) under mild conditions
and with a chemo- and regioselectivity that is hardly, if at all,
possible by classical chemical techniques. Although the development of enzymatic protecting group techniques has been intensively studied only in the last few years. a number of interesting
enzyme-labile protecting groups have been developed for organic synthesis (Scheme 9). Thus, for example. the phenylacet
(PhAc) amide['31 and the 4-acetoxybenzyloxycarbonyl (AcOz)
have been used as enzyme-labile protecting functions for amino groups, and carboxylic acids can be selectively
dernasked through enzyme-mediated cleavage of heptyl[251and
choline esters.[261Hydroxyl groups protected as the acetate,
benzoate. butyrate, and even pivalate can be liberated enzymatically.["
221 In this way such conversions can be carried out in

carbohydrate, steroid,
nucleoside, phenol ['12'2J

Scheme 9. Selected enzyme-labile protecting groups

polyfunctional carbohydrates, nucleosides, steroids, alkaloids,

and phenolic natural products, often with astonishingly high
regioselectivity not achievable with classical chemical techniques. The ability of enzymes to function in organic solvents
opens up new possibilities for the introduction of protecting
groups.

2.5. Reduction-Labile Protecting Groups['

Reductive conditions can be used to cleave a variety of protecting groups that are used more o r less frequently in organic
synthesis.

2.5.1. Cleavage by Hydrogenolysis

Benzyl groups, which can be present as ethers, esters, urethanes, carbonates, or benzylidene acetals, and are used for the
protection of alcohols, c a r b o x y k acids, amines, and diols, can
be removed under mild conditions by hydrogenolysis. The rate
of hydrogenolysis can be influenced by introducing electrondonating or -accepting substituents in the aromatic ring. Benzyl
groups are frequently used.

2.5.2. Cleavage by Reductive Elimination

Protecting groups of the 2-haloethyl type can be cleaved by a
mechanism involving the donation of electrons into the carbon halogen bond, leading to a fragmentation corresponding to the
p-elimination discussed in Sections 2.2.2 and 2.3 (Scheme 10).
2061

M . Schelhaas and H. Waldmann

REVIEWS

OMe

R - 0 = ether, ester, urethane

eQ = Zn", electrochemical
X

Br, CI

Y = H, Br. CI

ScI1c111c 10 '-Haloelll!l

p r o l c c l l l l ~gr<Iup, 111'11

i.11,

IhC

cIc;I\cd h\

rcdUCIIOI1

DDQ

or

CAN

Zinc is the preferred electron donor. but electi-ochemical methods are also successfully used.
The reduction of. for instance. ester protecting groups with
complex hydrides. reagents that also will attack ii pivalate ester.
is rarely used. owing to thc poor selectivity ofthese reagents. A n
exception worth mentioning is the regioselcctive opening of benxylidene acetals in carbohydi-ate
(Scheme 1 1 ). The

OMe

,OAc

(OAc

,OAc

,OAc

2.7. Cleavage of Protecting Groups Assisted by

Heavy Metal Salts or Complexes"' ' I
The activation of protecting groups with noble metals often
o ffe r s a n ad van t a geo us a 1t e I-na t Ive tor the t a rge ted deblock i ng
of functional groups. Thus. carbonyl compounds masked a s

reductive cleavage of disulfides with thiols o r complex hydi-ides

i s also seldom used. However. an interesting application of this
deblocking technique for the libel-ation of amino groups from
dithiasuccinimides (Dts protecting group) i n saccharide synthesis was I-ecently published['"' (Scheme I I ) .

2.6. Oxidation-Labile Protecting GroupsLU'

1 ..i-dithiancs can be hydrolyticallq regenerated easily by treatnicnt with stoichiomcti-ic amounts of Hg". Ag'. Cu". or TI"' salts
o r alternativelq bq 1-caction with other electi-ophiles o r by oxidution o f thc sulfur (see Section 7.6).
Catalytic iunotints of Khl. I]-". and Pd" complexes and even
Pd -C suffice for the removal of protecting groups containing
the ally1 group. Thus. ally1 cthel-s in carbohydrates and peptides
can be selectively isomerized to the acid-labile prop-l -enyl system through the action ofcalalytic quantities of Pd C 01- Kh'.
o r Ir" complexes. Hydrolysis of this g r o ~ i pis achieved under
inild conditions. (Schemc 1 3 ) . T h t allyloxycarbonyl ( Aloc)

or Pd C

The choice of oxidation-labile protecting groups is vci-y l i m i t ed. The 4-methoxybenzyl ( M p m ) and the more labile 2.4dimethoxybenzyl (Dmpm) ethers have proved their worth 21s
reagents in the synthesis o f complex molecules. Both o f these
functions are easily removed under mild conditions with
dichlorodicyanoquinone ( D D Q ) or with cerium(rv) ammonium
nitratc (CAN) (Scheme 12). or under acidic conditions (sec SCCtion 2.1.1).
I t is also worth mentioning that S.S-acetals of carbonyl compounds, such as 1.3-dithianes. can he cleaved after oxidation of
the sulfur (Scheme 12).
2062

R = alkyl, acyl

cod=

Protecting Group Chemistry

group. allyl esters, and ally1 phosphates can be cleaved under
mild conditions by Pdo-mediated allyl transfer to a variety of
nucleophiles such as amines and C-H-acidic compounds
(Scheme 13). As a consequence of its ready removal and its
compatibility with a range of other functional groups, the allyl
protecting group has become established in protecting group
chemistry and is finding increasing application in the synthesis
of complex natural products (see Sections 3.1.3, Schemes
23-27). 1,3-Dithianes and allyl protecting groups can also be
viewed two-stage protecting groups (see Section 2.9).

2.8. Photolabile Protecting Groupsr9-

Stable Form

oxidation

REVIEWS

f
~9Labile Form

OdR base

,s=o

Me0

Cleavage

Me

Me,s=o
\\

rnethylthioethyl (Mte)
HN~

HN~

(1,3-dithiane-2-yl)methoxycarbonyl (Dmoc)

Photolabile protecting groups contain a chromophore of high

chemical stability that can be selectively activated by irradiation
with light of a suitable wavelength. Of the many known photolabile protecting groups, the o-nitrobenzyl
has been
used repeatedly in the form of ethers, esters, carbonates, carbamates. and acetals (Scheme 14, see also Section 3.1.5). Two
pyridylethyl (Pyet, Pyoc)

R-0-PG

PG =

2-brornoethylester

onitrobenzyl
(OW

o-hydroxystyrydirnethylsilyl

choline ester
(Cho)

Z-oxo-l,2-diphenylethyl
(Desyl)

Scheme 14. Examples of photolabile protecting groups.

acetoxypropyl

further interesting photolabile protecting groups are the 2-0x01.2-diphenylethyl (Desyl)

and the o-hydroxystyryldimethylsilyl
(Scheme 14). Photolabile protecting
groups can be cleaved under mild conditions with light of a
suitable wavelength. and highly selective cleavage in the presence of other functional groups is possible by the choice of an
appropriate chromophore. Despite these advantages, photolabile protecting groups are used much less frequently than
other types of protecting groups.

2.9. Two-Stage and Safety-Catch Protecting Groups

The inherent (and deliberately employed) reactivity of, for
example, acid- and base-labile protecting groups is often the
source of undesired side reactions. Thus, masking can be (partially) lost at the wrong point in the synthesis. If several protecting groups of comparable reactivity are present in one molecule,
selective removal of a single protecting group can be difficult.
This problem can be avoided by the use of protecting functions
that initially exist in a chemically stable form and can be converted into a labile group when their cleavage is required; these
are the two-stage protecting groups.[331Examples include the
methylthioethyl (Mte) and the 1,3-dithianylmethoxycarbonyl
(Dmoc) groups, which are insensitive towards acid and base, but
which, after oxidation of the sulfur to sulfone. fragment by

Scheme 15. Selected two-stage protecting groups

base-induced ,&elimination (Scheme 15). The same principle

applies to the 2- and 4-pyridylethyl protecting groups, which,
after alkylation of the pyridine nitrogen, can be cleaved even
with morpholine (Scheme 15). 2-Bromoethyl esters can be
transformed into salts of the corresponding choline esters by
reaction with trimethylamine. The protecting group can then be
cleaved either with base or under milder conditions with the
enzyme butyrylcholine esterase (see Sections 2.4 and 3.1.4).
Removal of another interesting protecting group, the acetoxypropyl protecting group for amines, requires hydrolysis of the
acetate, subsequent oxidation of the liberated hydroxyl group
to give the aldehyde, and finally base-induced elimination
(Scheme 15). This proved to be a useful alternative to classical
acyl protecting groups in the total synthesis of mitomycin C.[341
Other examples of this method of protecting group cleavage are
1) the removal of 1,3-dithianes by reaction with electrophiles (or
oxidation of sulfur) and subsequent hydrolysis (see Sections 2.6
and 2.7), and 2) the removal of allyl groups by a ) isomerization
of the allyl system followed by acid hydrolysis o r by b) formation of a ~-allylpalladium complex and subsequent reaction
with a nucleophile (see Sections 2.7 and 3.1.3).
Closely related to the two-stage protecting groups are the
safety-catch protecting groups.[351In these systems a chemi2063

M . Schelhaas and H. Waldmann

REVIEWS
cally stable precursor is introduced, which is converted into an
activated intermediate directly before cleavage. In this case no
additional reagent, such as a base or a nucleophile, is required
in order to remove the protecting groups. The activated intermediate itself carries a reactive functional group, typically a nucleophile, which intramolecularly attacks the bond to the
blocked function and thus causes the cleavage of the protecting
group. Several examples of this technique are presented in
Scheme 16. Thus, reduction of the nitro-substituted aromatic

3-(oNitrophenyl)acetyl
or -propionyi

X=O,NH

n=1,2

4-azidobutvrvl

4-oxoacyl

protecting group strategies. For example, if a fragment contains

a C - C double bond, one would not consider using benzyl ethers

or esters as protecting groups, which must be removed by hydrogenolysis. After a successful analysis, concrete reactions, the
required reagents, and protecting groups are chosen simuItaneously. Functional groups, which, after they are introduced or
formed, should remain blocked throughout the entire synthesis,
will be protected with permanent protecting groups, whereas
those that will be liberated at an early stage in the synthesis will
be blocked with temporary protecting groups. In practice, it is
often neccessary to try out alternative approaches in the course
of the synthesis. One must remain alert to the fact that in such
cases other protecting groups may be required. For example, if
an aldol reaction is planned and strong base is required to deprotonate the carbonyl compound, any protecting groups required should be base-stable but may be acid-labile. However,
if this aldol reaction does not prove satisfactory, for example
because of low or undesired stereoselectivity, an alternative
would be Mukaiyama's variant of the aldol reaction, in which
the corresponding silyl enol ether is activated by a Lewis acid
(such a case has been described by Reetz et al.[36]). Now the
requirements for successful protecting groups would be completely reversed, and exchange of the previously used protecting
groups for others might thus be required. It should be noted that
the type of blocking can also affect the course of the planned
synthesis (see Section 3.5). This example illustrates that each
synthetic alternative (here an aldol reaction under basic or
acidic conditions) can require its own pattern of protecting
groups (base-stable or acid-stable as the case may be).
In the development of a protecting group strategy two fundamental concepts are available (Scheme 17):
-

2-chloracetyl

0 NH,

Scheme 16. Selected "safely-catch" protecting groups.

precursor generates the aniline derivative, which then cyclizes to

give the amide with liberation of the deprotected functional
group. The same principle applies for azidobutyrates. Crotyl
and 4-oxoacyl protecting groups react with hydrazine by addition to the a,fl-unsaturated system and by formation of a hydrazone, respectively, thereby placing the nucleophile in the correct
position. Analogously, chloroacetates are converted to substituted a-thioacetates by reaction with thiourea; subsequent cyclization releases the unprotected functional group.

3. Strategies for the Choice of Protecting Groups

When a synthesis is planned, the choice of suitable protecting
groups i s inseparable from the consideration of how one will
construct the basic structure of the target molecule and how the
stereochemistry will be established. The first step in planning a
synthesis is the retrosynthetic analysis of the target molecule,
which designates the fragments containing the groups to be
protected. On the basis of the inherent lability of these intermediates one can, even at this early stage, already exclude certain
2064

the use of orthogonally stable protecting groups

the use of protecting groups with modulated lability

The principle of orthogonal stability[371requires that only

those protecting functions should be used that can be cleaved
under (as a rule totally) different reaction conditions without
affecting the other functions present (Scheme 17). For example,
three protecting groups A, B, and C, which are attacked by acid,
by base, and by hydrogenolysis, respectively, could therefore be
selectively removed in any sequence at any point in a synthesis.
In following the principle of modulated lability, the protecting groups are all sensitive to one set of conditions, but to
differing extents. Consider three protecting groups A', A" and
A"', all of which are acid-labile. The most acid-sensitive group,
A', can be removed without affecting the other two. However,
the least labile A"' cannot be removed selectively in the presence
of one or both of the others (Scheme 17).
A strategy based upon the concept of orthogonal stability
guarantees a great degree of flexibility in the execution of a
synthesis. However, it is often not possible to find and use the
requisite number of orthogonally stable protecting groups, particularly in the case of multifunctional molecules, since cleavage
of one protecting group requires not only the stability of all the
other protecting groups but also of the masked molecule itself
under a wide range of reaction conditions. Additionally, in the
last steps of a synthesis it is often not necessary that all protecting groups have strictly orthogonal stability, since only a few
groups must be selectively liberated and a series of selective
A n p v . Chern In!. Ed. Engl. 1996, 35,2056-2083

REVIEWS

Protecting Group Chemistry

groups of the nucleobases, and base-labile phosphate esters

block the phosphate groups (Scheme 18). Similarly, in solid-

Orthogonal Stability

YA BY C?

-1base-labileJ

A
/VNH

Modulated Lability

A"

ASA

mNH

mNH

WWLT-rf
X l f

c,

. I

'I

DMTo@
CN

w
Y Y ?

A' A" A"'

x y z

A"

y v z

'

Scheme 18. Oligonucleotide synthesis with acid-labile alcohol protecting groups

and base-labile linkers and amino protecting groups. R = phenyl. isobutyl. etc.

Scheme 17. Orthogonal stability and moldulated lability as basic concepts for the
development of protecting group strategies.

deblockings extend the synthesis significantly. These disadvantages do not exist if one uses a set of protecting groups of similar
but modulated stability. In practice, a combination of the two
strategies is often employed, in which the earlier stages rely more
heavily on orthogonal stability and in the latter stages the concept of similar lability is more important.
These two fundamental strategies can be augmented by auxiliary strategies, including:
- the earliest possible unification of a protecting group pattern
- the introduction of "stand-ins"
- the use of protecting groups to direct reactions.

phase peptide synthesis the base-labile Fmoc group is frequently

used to block the N-terminus, while an acid-sensitive function
links the C-terminal carboxylic acid and the polymeric carrier
(that is, an acid-labile, polymeric protecting group). The orthogonal stability of all three types of protecting groups has, for
example, been employed in the synthesis of a precursor of the
dolastatin cytostatics by R. C. Kelly et al. (Scheme 19).[38a1

,Cost

BU

Pd-C,
MeOH/ THF

+
+

vb-

%HI\,

Fmoc

, ..,
MI\,

5"C, quant.

Fmoc

3.1. The Use of Orthogonally Stable Protecting Groups

The protecting groups discussed in Section 2 of differing lability can be used in a wide variety of combinations. In this case
again. there are numerous possibilities. A general answer to the
question of the most convenient combinations is not available.
However. from the many published syntheses one can sift out
some combinations that have proved effective time and again.
3. I . I . Acid Hydrolysis/Base HydvolysislHydvogenolysis
The combination of acid-labile, base-labile, and hydrogenolytically labile protecting groups is a frequently used protecting group strategy. In oligonucleotide chemistry, for instance, the acid-labile dimethoxytrityl ether is typically used to
protect hydroxyl functions, base-labile amides mask the amino

I Dolastatin 3

] = 11

Scheme 19. The Combination of acid-labile, base-labile, and reduction-labile protecting groups in the synthesis of a precursor of dolastatin 3 (11) by R. C Kelly
et al. [38a]

2065

M. Schelhaas and H. Waldmann

REVIEWS
selectively protected glutamic acid derivative 7 the a-benzyl ester
was selectively removed by hydrogenolysis and then the desired
bisthiazole unit was constructed. Acid-mediated cleavage of the
y-tert-butyl ester in 9 opened up the possibility of converting the
y-carboxyl group of glutamic acid into the amide. In both steps
the Fmoc group remained intact. This group can be removed
under basic conditions, if required, and the amino acid N-terminus can be
Among the classical combinations of protecting groups is the
simultaneous use of acid-labile acetals for the protection of carbony1 groups or 1,2-diols, base-labile esters, and reductively
cleavable benzyl groups. This strategy is often used, particularly
in the field of carbohydrate chemistry, as illustrated by the synthesis of the dimeric Lewis" antigen described by R. R. Schmidt
et al. (Scheme 20).[391In this case, the isopropylidene group of
R
CF3C02H
CHzC12
98%

/
p
A
\

O-CO2Me
OBz

OBzHo

12

AcO

I OAc

3.1.2. Fluoride-Labile Protecting Croups in Combination with

Oxidation-Labile Functions

,&$-O$O
H
HO

OBzHo

ride 15 by acidic hydrolysis. This made possible repeated glycosylation with the trisaccharide unit 14, leading to the complete
octasaccharide 16. Finally, all the protecting groups were successively removed from 16.
As demonstrated in this example, acetals are commonly employed as temporary protecting groups, whereas benzyl and acyl
protecting functions are preferred as permanent protecting
groups. When all such groups are removed to give the target
compound, the acyl groups are typically removed before the
benzyl ethers, in order to minimize steric hindrance in the final
heterogeneous catalytic hydrogenolysis. For this reason, it can
be advisable to exchange benzyl for acyl protecting groups as
early in the synthesis as possible (see Section 3 . 3 ) .
Examples of alcohols protected as acyclic 0,O-acetals are
given in Schemes 40,45, and 5O.Oxidation-labile and acid-labile
protecting groups can also be advantageously combined. In the
example in Scheme 41 the oxidatively cleavable (but also acidlabile) Mpm and Dmpm ethers were combined with an acidlabile dimethyl acetal.

OBz

BF3.Etz0, CH&/

hexane, 71%

13
AcO

15

The 4-methoxybenzyl (Mpm) and 2,4-dimethoxybenzyl

(Dmpm) ether groups, which are cleaved by oxidation to
quinone methides, can be favorably combined with a range of
different protecting groups. Their combination with fluoridelabile silyl ethers has proved to be particularly effective. This
combination rarely gives problems, since the substituted benzyl
ethers are completely inert towards the fluoride ion. The

64%

AcO

I OAc

M e p O B z l
OAc
AcO

16

MeO.,,
OTBC

1) CF3COzH,CH,CI,
95%
2) NaOMe, MeOH; Ac,O, pyr, 87%
3) HS-(CH&-SH. py / Hz0; AcZO, py, 93%
4)Pd-C, H, MeOH: NaOMe, MeOH, quant

Galp(l-4)GlcNAcS(1-3)Galp(l-4)GlcNAc~(l-3)Gal~(1-4)Glc~1-O)-R

Fuca(1-3)

17

Fuca(1-3)

Scheme 20. Synthesis of the dimeric Lewis" antigen by R. R. Schmidt et al. [39].

the lactoside 12 was initially removed by acid hydrolysis. Of the

two liberated hydroxyl groups present in 13, only the equatorial
reacts in the subsequent glycosylation with the activated trisaccharide 14. A combination of base-labile esters (acetate and
benzoate), a benzyl ether, and, again an acid-labile acetal are
present in the resulting pentasaccharide 15. The benzyl group in
the fucose unit of 15 ensures that the correct stereochemistry is
produced in the introduction of this monosaccharide (see also
Section 3.5); the acetates increase the stability of the 0-fucosidic
bond towards acidic
The acetal protecting group
was again selectively removed from the complex pentasaccha-

2066

HFI CH CN

Scheme 21. Final transformations in the synthesis of the immunosuppressant

FK506 (21) by S . L. Schreiber et al. [41].
Anger,,. Chem. Int. Ed. Engl. 1996.35, 2056-2083

Protecting Group Chemistry

REVIEWS

oxidants D D Q and CAN (see Section 2.3), which are used for
the oxidative cleavage of the Mpm and Dmpm groups, can in
the worst case attack only the most labile silyl ethers such as the
T M S group. which is seldom used in any case due to its high
hydrolytic lability.
An illustration of the use of Mpm ethers in concert with
various silyl ethers is provided by the synthesis of the
immunosuppressant FK506 (21) by S. L. Schreiber et al.
(Scheme 21 ) . I 4 ' ' Two hydroxyl groups are masked as substituted benzyl ethers and four others as different silyl ethers in the
advanced intermediate 18. Oxidation with DDQ cleaved the
alkoxybenzyl ethers selectively, and the hydroxyl group thus
liberated was then oxidized with Dess- Martin periodinane to
give the corresponding diketone 19. Of the various silyl groups
in 19 (for a discussion of their relative stabilities see Section 3.2.2). the most labile was removed by treatment with acid
and the selectively demasked hydroxyl group was then oxidized
to give triketone 20 (see Section 3.4 for the strategy of masking
carbonyl groups by the temporary introduction of hydroxyl
groups as "stand-ins"). The synthesis was completed by the
simultaneous cleavage of the remaining silyl groups.
The combination of Mpm and silyl ethers also proved its
worth in the synthesis of the immunosuppressant rapamycin by
K. C. Nicolaou et al. (Scheme 22).[421In this case, the hydroxyl

-+ '+
0

MpmO

MpmO

OTBDMS

Me

Me

OMe

Me

22

Me

23

group in the product obtained from the Cr"-mediated addition

of vinyl iodide 23 to aldehyde 22 was blocked as a TIPS ether.
The TBDMS group was cleaved selectively (see Section 3.2.2)
and the deblocked hydroxyl group was oxidized to give aldehyde 25. which enabled a subsequent aldol reaction (25 -+ 26).
Intermediate 26carries both Mpm groups and a TIPS ether, and
the methoxybenzyl ethers were selectively removed by oxidation.
The resulting diol 27 was oxidized to provide the dicarbonyl
compound 28. The TIPS ether remained intact in this reaction
sequence and was removed only at the end of the synthesis.

3.1.3. Ally1 Croups with Acid- and Base-Labile Protecting Croups

Ally1 groups can be profitably utilized in combination with a
wide variety of other protecting groups, in particular acid- and
base-labile blocking groups, as they are readily cleaved by
means of Pd-, Rh-, and Ir-catalyzed reactions (see Section 2.7).
Carboxylic acids, amines, and alcohols can be blocked as allyl
esters, allyl urethanes, allyl ethers, and allyl carbonates. In the
past ten years the increasing number of examples in which allylic
groups are removed in crucial deblocking steps proves that this
protecting group has assumed a permanent position in protecting group chemistry.
One of the first demonstrations of the effectiveness of allyl
protecting groups in the synthesis of complex molecules was
provided by the synthesis of glycopeptide 32 by H. Kunz et al.
(Scheme 23).L431The allyl group in the glycosylated allyl ester of
asparagine 29 was transferred to morpholine in almost quantitative yield in a Pdo-catalyzed reaction. Neither the acid-labile
tert-butylurethane, nor the glycosidic bond, nor the base-labile

TIPS

Morn0

Mom0

OTBDMS

1) CrCI2, DMSO, 83%

2) TIPSOTf, 98%
t

24
TIPS

29

I ) HFI Py.97%
2) (COCI),, DMSO
NEt,, 97%

H-Phe-Thr-OAII

25

EEDQ, 81%

TIPS

30

____)

26
TIPS
quant.

OH

DDQ, CHCI, I HO
,
25"C, 1h, 98%
Me

Me

6MeMe

31

Me

27

__I)

TIPS
0

*
\

1 -

Me

Me

OMeMe

Me

28
Scheme 22. Sections of the synthesis of the immunosuppressant rapamycin by
K. C. N~colaouet iil. [42].
Angrpi . Clioi? Inr Ed En,ql 1996, 35, 2056-2083

32
Scheme23. Application of allyl esters in a glycopeptide synthesis by H. Kunz
et a]. [43]. EEDQ = ethyl-2-ethoxy-I .2-d1hydroquinoline-l-carboxylate.

2067

M. Schelhaas and H. Waldmann

REVIEWS
acetates of the carbohydrate moiety were attacked under these
conditions.
The especially mild conditions for the cleavage of allyl esters
were also used to advantage by S. Danishefsky et al. in the
synthesis of the immunosuppressant r a p a m y ~ i n [(Scheme
~~I
24).
Attempts to activate the carboxyl group in the advanced intermediate 33 prior to a condensation reaction resulted in lactonization at the tertiary OH group of the tetrahydropyranyl

H
[Pd2(dba)J, PPh,, CHCI,
n C4H9NH2I HC02H
50"C, l h

K
11

33

Me.,,&

OMe

Me

0'
polymer-bound DNA
All

Scheme 25. Application of allyl protectlng groups in the synthesis of oligonucleotides by R. Noyori et al. [45]. dbd = dibenzyhdeneacetone.

35

Scheme 24. Section of the synthesis of rapamycin by S. Danishefsky et

dl.

[44]

ring. The OH group therefore had to be protected. To avoid side

reactions, it was necessary to mask the carboxylic acid as an allyl
ester and then to silylate the remaining free hydroxyl group in
34, leading to the doubly protected 35. At this point the carboxyl
group could be deprotected. The Pdo-mediated removal of the
allyl ester in 35 proceeded under such gentle conditions that
neither the complex molecular framework (including the z-ketoamide) nor the acid- and base-labile T M S ether were lost or
damaged.
R. Noyori et al. relied on ally1 groups to protect amines
and phosphates in the preparation of oligonucleotide^^^^^
(Scheme 25). They employed 0-allyl-masked phosphoroamidites as condensation reagents and allyloxycarbonyl (Aloc)
protected nucleobases for the construction of oligonucleotides
37 on the solid phase. When the synthesis was complete, all of
the allyl protecting groups in the polymer-bound oligonucleotide were removed in a single step. The D N A oligomers 38 thus
obtained, which are still bound to the polymer carrier, can be
used directly for biological tests or can be cleaved from the solid
phase. In one example, the simultaneous cleavage of 104 allyl
groups led to the 60mer 39 in an overall yield of 25 YOand in high
purity. (When conventional protecting groups were used an
overall yield of 6 % was obtained.)
Blechert et al. incorporated the targeted cleavage of an allyl
carbonate into the synthesis of a biologically active analog of
the antitumor alkaloid taxol r461 (Scheme 26). The ally1 carbonate in 40 was cleaved in almost quantitative yield without attack
at the ketal or ring expansion by fragmentation of the baselabile B-hydroxyketone unit.
2068

completely deprotected DNA

41
Scheme 26. Cleavage of an ally1 carbonate in the synthesis of a taxol analog by
S . Blechert et 81. 1461.

The selective cleavage of an allyl ether was employed by D. A.

Evans in the synthesis of van~ornycin[~']
(Scheme 27). The Boc
group in peptide 42 was specifically removed with retention of

OAll

1) CF~COZH
thioanisole

HO.,

0
81-90%
I

42

OBzl

45

OBzl

OBzl

44

Scheme 27. Section of the synthesis of vancomycin by D. A. Evans et al 1471.

DIC = diisopropylcarbodiimide. HOBt = 1-hydroxybenzotriazole
Angel<..Chem. h i , Ed. Enai. 1996. 35. 2056-2083

Protectin& Group Chemistry

the phenolic allyl ether, thus enabling selective extension of the

amino acid chain to give 43. Removal of the allyl group under
mild conditions provided phenol 44, and subsequent oxidative
ring closure yielded the cyclic product 45.
3.1.4. Enzyme- Labile Protecting Gvoups with Acid- and
Base-Labile Groups
Owing to the often distinct chemo-, regio-, and stereoselectivity o f enzyme-mediated transformations, enzyme-labile protecting groups can be combined with other protecting groups that
can be removed by classical chemical techniques (see Section
2.4). Enzyme-catalyzed reactions are invaluable especially when
- the substrate is so labile that only the most gentle deblocking

methods are tolerated.

- the chemo-. regio-. and stereoselectivity of the .enzyme im-

parts orthogonality to protecting groups which, under classical deblocking methodology, would have similar or even inverted lability.
Both principles were utilized in the chemoenzymatic synthesis
of the S-palmitoylated and S-funesylated C-terminal lipo-

@
choline
A l o ~ - C y s - M e t - G l y - O - ~ ( ~ ~ ) ~esterase
I
Bra
S
-

Aloc-Cys-Met-Gly-OH
I

S-pal

"---Y-46

Pal

47

H-Leu-Pro-Cys-OMe

48 F./'

REVIEWS

hexapeptide of the human N-Ras protein 49[24,261 (Scheme 28).

This type of modified lipoprotein cannot be deblocked under
basic conditions because the labile palmitic acid thioester group
would be preferentially hydrolyzed. Under acidic conditions the
double bonds of the farnesyl moiety are attacked. However, the
C-terminus of the peptide chain was successfully deprotected by
selective enzymatic hydrolysis of the choline ester in 46 without
attack on the palmitic acid thioester. The observed chemoselectivity here is exactly opposite to that found in the nonenzymatic
conversion. Extension of the peptide chain in 47 with the farnesylated tripeptide 48 yielded the desired lipopeptide 49. Alternatively, the N-terminal AcOZ protecting group in 52 could be
removed by lipase-initiated spontaneous fragmentation under
neutral conditions and without attack on the C-C double
bonds. After extension of the peptide chain in 48 with a further
AcOZ-protected dipeptide and enzymatic deprotection, the
lipopentapeptide 51 was obtained, which afforded the acid- and
base-labile Ras peptide 49 after treatment with the S-palmitoylated cysteine 50. In these enzyme-catalyzed deprotections of
urethane protecting groups, the biocatalyst differentiates between the C-terminal methyl ester and the aryl acetate incorporated in the urethane group.
The advantageous properties of the enzyme-labile protecting
groups also makes possible the chemoenzymatic synthesis of
base- and acid-sensitive O-gly~opeptides.[~*~
491 Thus, the C-terminal heptyl ester of the glycosylated serine derivative 53
could be selectively cleaved by lipase-mediated hydrolysis[481
(Scheme 29). Glycopeptide 54 was condensed with an N-terminal deblocked glycodipeptide to give 55, and enzymatic removal
of the heptyl ester was still efficient. Compound 56, accessible in
this way in high yield, was subsequently extended to provide the
glycohexapeptide 57. This compound represents a characteristic

49

Aloc-Cys-OH
I
50 '-Pal

H-Met-Gly-Leu-Pro-Cys-OMe
I

51

.F
' ar

53

1) AcOZ-Met-Gly-OH

AcO

carbodamide
2) lipase
I

48

54

Hep

AcO
A

base M
PH
37%,
7.076%

Z-ker-Thr-Ala-OHep
I

'-Far

2-Ser-OH

OAC

4~0%
AcHN

H-Leu-Pro-Cys-OMe

OAC

cAcHN
O q

?I
Z-Ser-Thr-Ala-OH
1

___t

F
.' ar

4
iipase
0

M e P o e o d

AcOZ

A c O B o

Act) 'OAc

Acb b A c

O O e O d L e u - P r o - C y s - IO M e
0

A c O S o

55

AcO

56

OAc

AcO+
AcHN
H-Pro-Pro-Ala-OHep

Leu-Pro-Cys-OMe
I
52 S

Angric Clicrn Inr Ed Engl 1996. 35. 2056- 2083

I
Z- Ser -Thr-Ala-Pro-Pro-Ala-OHep
I
AcHN

57

Far

Scheme 28. ChemoenLymatic synthesis of the S-palm~toylatedand S-farnesylated

C-terminal Iipohexapeptide of the human N-Rds protein by H. Waldmann et al.
[24. 261

carbodiirnide, 62%

AcO

OAc

Scheme 29. C-terminal enzymatic deprotection

H. Waldmann and H. Kunz et al. [48].

of glycopepfide heptyl esters by

2069

M. Schelhaas and H. Waldmann

REVIEWS
fragment of a tumor-associated antigen that appears on the
surface of human breast cancer cells.
Selective N-terminal cleavage of a phenylacetoxybenzyloxycarbonyl (PhAcOZ) urethane group (analogous to the AcOZ
group) in 0-glycopeptides was accomplished by action of penicillin G a c y i a ~ e(Scheme
[ ~ ~ ~ 30). The enzymatic cleavage of this

pow

penicillin G acylase
pH 7.5, RT, 78%

Nd-orBu

v
58

PhAcOZ

3.1.5. Photolabile Protecting Groups with

AcO

Other Protecting Groups

PhAcOZ-Ser-Pro-OH
H-Ser-Of Bu

(Ac),GlcNAc

PhAcOZ-Ser-Pro-Ser-Of
Bu
I
I
* (Ac)~GIcNAc GlCNAC(AC),

(Ac)3GlcNAc

condensation

59

60
H-Ser-Pro-Ser-Of
Bu
I
I
(Ac)~GIcNAc G~cNAc(Ac)~

penicillin G acylase
pH 7.5, RT, 74%

61

Aloc-Ser-Pro-Thr-Ser-Pro-Ser-Of

__t
___)

O=P-OAII

Bu

62

Scheme 30. Use of the enzyme-labile PhAcOZ group in the synthesis of a hexdpeptide sequence of a partially processed RNA polymerase by H. Wdldmann et al. [49].
RT = room temperature.

urethane protecting group in 58 yielded 59, which was condensed with a further glycosylated dipeptide to give 60. The
N-terminal protecting group in 60 was removed enzymatically
in a selective manner, and the peptide 61 thus liberated was
finally converted into the phosphorylated and glycosylated peptide conjugate 62, a characteristic sequence of a partially processed RNA polymerase. In the enzymatic transformations described in Schemes 29 and 30, no undesired attack on the acetate
esters in the carbohydrates was observed, and neither b-elimination of the carbohydrate moiety nor anomerization of the glycosidic bonds take
The stereoselectivity of enzyme-mediated reactions can also
be used advantageously for the preparation of specifically functionalized and protected synthetic intermediates. An example of
this is the desymmetrization of the meso diacetate 63 in the
enantioselective synthesis of the indole alkaloid (-)-anthirin
(66) by G. Lesma et al.[511(Scheme 31). Diester 63 was convert-

O-Ac

lipase from
~porcine
~
pancreas
96%; 99% ee

OH
0-Ac

---+

ti

u
65

66

Scheme 31. Example of an enzymatic conversion in the synthesis of ( - )-anthirin

(66) by G. Lesma et al. [Sl].

2070

The cleavage of protecting groups by irradiation with light of

a suitable wavelength lends itself well to an orthogonal protection strategy.[641Many other protecting groups are not affected
by these conditions, and thus no particularly advantageous
combinations have emerged. One of the most impressive applications of this technique is found in the synthesis of the enediyne
antitumor antibiotic calicheamicin 7: by K. C. Nicolaou et
aLf6'](Scheme 32). To activate the complex oligosaccharide of

(AC)~GICNAC GlCNAC(AC),

OAll

ed into the monoprotected diol 64 with essentially complete

selectivity by treatment with porcine pancreatic lipase. The configuration remained intact throughout the construction of the
target compound 66 via intermediate 65. Further applicat i o n ~of[ such
~ ~ ~biocatalyzed transformations include syntheses
of s h o w d ~ m y c i nbiotin,[541
~ ~ ~ ~ an analog of thromboxane
A, ,i551 r h i ~ o x i n , [ ~neplanocin
~]
A,r571ni~ardipin,[~*]
chrysanthemic acid, permethric acid and caronic acid,[591rifamycin,[601
compactin,[6'1 ( +)-disparlure,r621and a p h a n ~ r p h i n . ' ~ ~ ]

67

1I

hv, 82%

Scheme 32. Application of a photolabile o-nitrobenzyl protecting group in the synthesis of calichearnicin y: by K. c. Nicolaou et al. [65].

the natural product towards the coupling reaction with the aglycone, the anomeric center of the terminal carbohydrate part of
the fully protected intermediate 67 had to be liberated. The
o-nitrobenzyl ether function proved to be a suitable protecting
group. Due to its high stability, it could be introduced at an
early point in the synthesis and remained intact even under
drastic conditions such as treatment with methoxide, tetrabutylammonium fluoride, bromine, and diisobutylaluminum hydride. However, it was easily cleaved from 67 in high yield by
photolytic conditions, under which the oxime group, the glycosidic bond, the thioester, the Fmoc group, and the silyl ether all
remained intact.
The rate and selectivity of the photochemical deblocking is
also an integral part of the "caging" technique used in biological
studies[661(Scheme 33). In this context a inactive substrate such
as 69 having a photolabile protecting group is introduced into a
biological system, for example, a cell. The protecting group is
removed by an intense flash of light and the active compound,
in this case 70, is thereby liberated. The advantage of this
method is the precise triggering of the deprotection and liberaAnpen-. Chem. Int. Ed. E n d 1996, 35. 2056-20X3

Protecting Group Chemistry

R. Ellison et al. in the synthesis of a prostaglandin precursor[7o1

(Scheme 34). The selective removal of the dimethyl acetal in 71
proceeded in almost quantitative yield and provided the liberated aldehyde 72, which was used in chain extension. After the
removal of the dithiane group, 73 was ready for ring closure,
which yielded the desired functionalized prostaglandin precursor 74.
m
\

OH

HO-P11.0

Ic-AMP

70

:93
Me-0

Scheme 3 3 . Application of a photolabile protecting group in the "caging" method.

A = adenine.

tion of the active species, making pharmacokinetics measureable, even for fast processes. The prerequisite for the use of the
"caging" technique is that the photodecomposition products
have minimal toxicity. The required excitation wavelength and
intensity should also not damage the cell. The photolabile protecting group most frequently employed in synthesis is the o-nitrobenzyl group. Since the photochemical cleavage of this species leads to the production of the toxic o-nitrobenzaldehyde,
other photolabile groups such as l-pyrenylethyl[671and the benzoin
are commonly used for biological applications
(see Scheme 33). An exception was described by W. Mantele et
al.,[691who monitored the progress of a reaction by following
the decomposition of an mi-nitro intermediate by IR spectroscopy.
Although photochemically cleavable protecting groups offer
many advantages, their use is restricted since the substance to be
deblocked may not itself contain any photoexcitable functional
groups. For this reason this protecting group strategy was not
widely applied for a long time. In more recent developments, for
example the establishment of anchor groups in combinatorial
chemistry (see Section 5), photolabile protecting groups may
play an increasingly important role.

5% CF&O2H I
CHC13,0C,

72

71
1 ) OsO, NalO,
2) NaOH, H20 I
dioxane
AcO

HO

74

73

Scheme 34. Use of acid-labile protecting groups with modulated lability

synthesis of a prostaglandin precursor by R. Ellison et al. [70]

In

the

The principle of using protecting groups with modulated

acid-lability has found repeated application in peptide chemistry (see Sections 2 and 4). This is exemplified by the synthesis
of human insulin carried out by P. Sieber et al." and sketched
in Scheme 35. Compound 75 bears three protecting groups of

0
m O ' E - O t
\

CF3CH2OHI
HCI. pH 3.5

Bu

lllr

75 Trt

Bpoc
Bpoc-Ot

Ot Bu

Bu

76

s
'

coupling

___t

'I O f B u

i Or Bu

p Ot Bu

BU

s?

S-Acm
Boc

77
Bu

H-------rOt

'

S-Acm
Boc

I
S-Acm
S-Acm
I

Boc

90% aq.
CF3CHzOH
60C

coupling

78

s-s
Boc

OtBu

.
1 ) 95% aq.
CF3C02H

2) 12/AcOH

Ot Bu

79

s-s
I

In many cases acetals have different labilities towards acids.

For instance, acyclic acetals are more sensitive than their cyclic
equivalents to acidic conditions. This property was utilized by

__t

O H

3.2. Protecting Groups with Modulated Lability

3.2.1. Modulated Acid- or Base-Lability

0-Me

Bpoc~-Ot

Protecting groups with modulated lability are often utilized in

organic synthesis, although they d o not provide the same degree
of safety offered by orthogonally stable functions. In particular,
this strategy is applied when several of the same type of functional group are present in the molecule to be protected for
which an insufficient number of orthogonal groups is available,
or when the removal of the most stable protecting group could
lead to undesired side reactions. In principle, protecting groups
of modulated lability can be developed for each of the classes
discussed in Section 2. Acid-labile, base-labile, and fluoridelabile protecting groups have proved to be particularly suitable;
oxidatively cleavable functions have more limited scope. The
modulated lability of many protecting groups can be referred to
in, for example, ref. [9].

:v3
-

?
?

s-s
'

OH

OH

80

Scheme 35. Use of acid-labile protecting groups with modulated lability in the
synthesis of human insulin by P. Sieber et al. [71].

2071

M. Schelhaas and H. Waldmann

REVIEWS
varying lability-the tert-butyl ester and the Bpoc and trityl (Trt)
groups. First the trityl group was removed selectively in the presence of the Bpoc group, which is likewise very acid-sensitive. After
extension of the B chain (76 -+77),the Bpoc group was selectively
removed. The A chain was completed by coupling 78 with a
similarly Boc-protected peptide. All the tert-butyl protecting
groups in 79 were then simultaneously cleaved. After the cysteine side chain was deblocked, ring closure provided insulin 80.
While the synthesis shown in Scheme 35 proves that this
methodology can be quite successful, it has not found widespread use in peptide chemistry due to the potential of losing
protecting groups required later in the synthesis. Owing to their
greater reliability, orthogonal protecting strategies are generally
preferred.
The modulated based-lability of two urethane protecting
groups was used by K. C. Nicolaou et al. in the synthesis of a
calicheamicin-dynemicin hybrid1721(Scheme 3 6 ) . The 2-fluo-

Relative stability of silyl ethers

R-O-SiRS3

Actdolysis:

R3Si: Me3Si (1) < Et$i (64) < t BuMe,Si (20000)

< i Pr3Si (700000)< t BuPh,Si (5000000)

Basic solvolysis:

R',Si:

Me3Si (1) < Et3Si (10-100) c t BuMe,Si

(20000) < i Pr3Si (100000)

= t BuPh,Si

Scheme 37. Relative rates of hydrolysis for silyl ethers R - 0 -SIR;

(!ir;,').

various silyl ethers under acidic and basic conditions. (The order
of the cleavage rates on treatment with fluoride ions is similar.)
As a consequence of this range of activities, silyl ethers with
modulated solvolytic activity are now commonly used in synthesis.
tert-Butyldimethylsilyl (TBDMS) ethers can be cleaved with,
for example, tetrabutylammonium fluoride (TBAF), while the
cleavage of the more stable triisopropylsilyl (TIPS) and the terfbutyldiphenylsilyl (TBDPS) protecting groups requires solutions of HF in pyridine, acetonitrile, or even water.
These clear differences in reactivity were employed by
Danishefsky et al. in the synthesis of the immunosuppressant
r a p a m y ~ i n (Scheme
~ ~ ~ ] 38). When intermediate 83, which con-

I ) TBAFI HOAc
2) Dess-Martin
periodinane
Me..,&

e
/

;
-

M
Me

OTlPS

Me OMe Me Me OTBDMS

83
Me

1) TiCl,(Oi Pr)
2 ) HF 1 PY

Me

82

Scheme 36. Use of amino protecting groups of differing lability in the synthesis of
a calicheamicin-dynemicin hybrid by K . C. Nicolaou et al. 1721.

84
Me

renylmethoxycarbonyl (Fmoc) group was removed from the

complex glycoside 81 by treatment with diethylamine, without
attack at the similarly base-labile 2-phenylsulfonylethoxycarbony1 (Psec) group. The cleavage of this group under basic
conditions later triggered the liberation of the "warhead",
which is buried in the dynemicin aglycone of 82.
3.2.2. Modulated Lability of Silyl Protecting Groups

The lability of silyl groups towards acids, bases, and fluoride

ions can be specifically tuned by varying the substituents on
silicon. Scheme 37 indicates the relative rates of hydrolysis of
2012

I Rapamycin 1 = 85
Scheme 38. Use of silyl protecting groups of modulated lability in the synthesis of
rapamycin (85) by S Danishefsky et al. [44].
Anpew. Chem. l n t . E d Enyl. 1996.35. 2056-2083

REVIEWS

Protecting Group Chemistry

tains a TMS. a TBDMS, and a TIPS ether, was treated with
TBAF, the TMS and the TBDMS ethers were cleaved, but the
TIPS ether remained intact. After oxidation of the liberated
secondary hydroxyl group, the macrocyclic ring was closed, and
finally the TIPS ether in the cyclohexane moiety in 84 was
cleaved with a solution of H F in pyridine to yield the natural
product 85.
The different stabilities of silyl ethers was also utilized by
K. C. Nicolaou et al.[421 in their synthesis of rapamycin.
Scheme 22 indicates that treatment of 24 with HF/pyridine removed only the TBDMS protecting group, while the TIPS-protected secondary hydroxyl group remained intact. A similar
strategy was used in the last steps of this synthesis (Scheme 39).

in the synthesis of FK505[411(Scheme 21). Macrocycle 19 contains TIPS, TBDMS, and TES protecting groups, of which only
the TES group was removed by treatment with aqueous trifluoroacetic acid, allowing the selective oxidation of only one of
the four hydroxyl groups.
R. A. Holton et al. took advantage of the varying lability of
silyl ethers towards both bases and different sources of
fluoride ions in the synthesis of the antitumor agent tax01"~~
(Scheme 40). Thus, treatment of 89 with acetic acid cleaved only

F c c o
TIPSO

91

92

86
TIPSO

aq. HF, CH3CN

b

94

93
0
TASF: (Et2N)$
Me

Me

OMe Me Me

.,

88
Scheme 39. Use ofsilyl protecting groups of modulated lability in the synthesis of
rapamycin by K . C. Nicolaou et al. [42].

Of the five silyl protecting groups present in 86, only the TES
ethers were cleaved by HF/pyridine. Thus the selective oxidation of the liberated secondary hydroxyl groups to give the
corresponding carbonyl groups could be carried out. The remaining, more stable silyl ethers were cleaved by treatment with
H F in CH,CN/H,O, and the synthesis was subsequently concluded by Pdo-mediated coupling of the two vinyl iodides.
The differing acid sensitivity of silyl ethers allowed S.
Schreiber et at. to selectively deblock an advanced intermediate
A n g w . (%em i n l Ed Engl. 1996. 35, 20.56 -2083

0
Me,SiF,

Scheme 40. Use of silyl protecting groups of modulated lability in the synthesis of
taxol by R. A. Holton et al. 1731.

the TMS ether; the other acid-labile groups, the BOM acetal
and the other silyl groups, remained intact. After transformation of 90 into the oxetane 91, the TES group was selectively
removed by treatment with HF/pyridine. The central eightmembered ring was oxygenated as required in the position adjacent to the free hydroxyl group. Finally, the remaining TBDMS
protecting group was removed with the reagent TASF (see
Scheme 40; this conversion failed with TBAF) to allow the attachment of the side chain.

3.2.3. Modulated Oxidation-Lability

The oxidation-labile Mpm and Dmpm ethers (see Section 2.6)
are attacked at different rates by dichlorodicyanoquinone
(DDQ) at 0-5C. This modulated lability can be employed
advantageously in the synthesis of polyfunctional target compounds, including those containing C-C double bonds. Thus,
in the course of a total synthesis of methynolide (98), 0. Yonemitsu et al. cleaved the more electron-rich, and therefore more
easily oxidized Dmpm group in 95 in the presence of an Mpm
group by treatment with DDQ at 0 "C (Scheme 41).[74,7s1
The hydroxyl group in 96 thus liberated was then oxidized to the
2013

REVIEWS

M. Schelhaas and H. Waldmann

'

-,

HO&
o>
:OH

1''

'
0

".,

96

&"'*

MpmO

"OH

DDQ
CH,CI,
RT

only a few- and under certain circumstances too few--of the

available protecting groups fulfill this criterium. Second, such a
strategy can lead to a significant increase in the number of steps
required at the end of the synthesis, even to the point that the
associated losses become crippling.
To avoid these problems, the following strategy is often followed, in particular in complex multistep syntheses. As early as
possible in the synthesis, protecting groups are used that can be
removed under the same conditions in order to unify the protecting group pattern as much as possible. For example, intermediates having a single type of protecting group can be designed and used from the beginning of a synthesis, thus
automatically simplifying the problem. This strategy cannot be
used if specific reactivities in the intermediates are eliminated in
the course of the synthesis, or the directing effect of particular
protecting groups is required (see Section 3.5). Then protecting
groups in the intermediates must be removed and replaced with
others that correspond to the type remaining in the molecule.
Under certain conditions this can increase the length of the

"0

1."

"OH

0
97
_.

(MethynolidelE 98

"OcN
0

DDQ=

CI

CN

0
Scheme 41. Selective oxidative cleavage of the Dmpm group in the presence of the
Mpm group in the synthesis of methynolide (98) by 0. Yonemitsu et al. [74,75].
PDC = pyrrdinium dichromate. RT = room temperature.

8-

TBDMSO
(Ph),F TBDMSO.,,

OTBDMS
,..OTBDMS
TBDMSO

C o \ ' " Z M S

aldehyde, which cyclized to give the hemiacetal97 following the

acid-catalyzed hydrolysis of the adjacent acetal protecting
group. Finally, the desired natural product was obtained by
oxidative cleavage of the Mpm ether at room temperature.
Such fine-tuning of lability through the introduction of substituents can also be achieved for other types of protecting
groups. Thus, for example, when a nitrophenyl substituent is
attached to the allyl system of an allyloxycarbonyl (Aloc) group,
cleavage of the resulting 4-nitrophenylallyl (Noc) urethane with
Pdo and Rh' catalysts is significantly slower than that of the
simple unsubstituted allyl protecting groups.[761As can be seen
from Scheme 42, the allyl ester in dipeptide 99 can be cleaved
selectively without attack on the Noc group.

CH(SEt),

103 OMMTr

/'

102

~ O T B D M S

'"OTBDMS
OTBDMS

OTBDMS

HO
Me0

TBDMSO

t'
0

Noc

TBDMSO
OTBDMS

Scheme42. Selective cleavage of an unsubstituted allyl ester in the presence of a

nitrophenylallyl group by H. Kunr et al. 1761.

OTBDMS
OTBDMS

M p m q oM~
:,pm
'"OTBDMS

Me0w

OAc

&

OTBDMS

' OAc

OTBDMS

3.3. Unification of Protecting Groups

M

"

'.OTBDMS
m
~

In the construction of polyfunctional molecules the tactic of

using several protecting groups that must be removed under
drastically different reaction conditions can lead to problems in

the late stages of a synthesis, for example after successful construction of the molecular framework. First, the molecule must
be stable under all of the conditions used for deblocking. This
stipulation must be made in the selection of protecting groups
for each subunit in the whole molecule. It is thus possible that
2074

1) - 5)

Scheme43. Part of the synthesis of palytoxln by Y. Kishi et al. [77]. 1 ) DDQ,

rBuOH/CH,CI,, room temperature (RT), then Ac,O. DMAP in pyridine; 2)
HCIO, (1.18 N) inTHF, 25C. 8 d ; 3 ) 0 . 0 8 ~LiOH, H,O/THF (t:2:8), 2 5 C 20h;
4) nBu,NF in THF. 22C. 2Oh; 5 ) AcOH/H,O (1:9), RT. 12h
A n p w Chem lnr. Ed. Enql 1996, 35, 2056-2083

REVIEWS

Protecting Group Chemistry

synthesis; however, the risk of losses is shifted toward earlier
stages in the synthesis with the simpler and more readily available early intermediates.
The strategy of creating the uniformly protected intermediates, which can be completely deblocked in only a few steps at
the end of the synthesis, was followed by Y Kishi et al. in
the synthesis of the marine natural product palyt~xin[~]
(Scheme 43). The size of the molecule, sheer number of groups
needing protection, and the long list of synthetic problems to be
overcome made it very difficult to use only one type of protecting
group. Intermediate 109 (Scheme 43) contains the complete and
still completely protected molecular framework of palytoxin,
which was formed by the coupling of fragments 101 -108. In 101,
102, 103. and 104 the protecting groups are almost exclusively
fluoride-labile. 105 and 108 incorporate esters of differing baselability. and in 107 the hydroxyl groups are protected as oxidation-labile Mpm ethers. In addition, it was necessary to protect
the hemiacetal at C-47 and the diol unit at C-100 and C-101 as
acid-labile acetals. This significant unification of the types of protecting groups enabled efficient deblocking in only five steps at
the end of this extremely arduous synthesis (Scheme 43).

The strategy of choosing all the permanent protecting groups

from the same class was demonstrated by D. A. Evans et al. in
the synthesis of cytovaricin (115)[781(Scheme 44). The intermediates 110-112 used to construct the protected form of the
natural product 114 carried exclusively silyl protecting groups,
all of which were removed in high yield in a single step at the end
of the synthesis. The conditions required were so mild that the
hemiacetal formed at C-17 remained stable and did not lose
water and form a diene.
S. Danishefsky et al. used the same principle in the synthesis
of the enediyne antibiotic dynemicin
(Scheme 45). Thus

MOM?

@O+
MOMO

OLi

116

117

Me

MgBr,. 0.24%.
MO#cozMoM
\
/
Me

Me

,t Bu

MOMO

OMe

/
\

36h

15% (4 steps)

OH

t Bu

OTES

I I ,

112

HO

OH
119

TBDMSO

t Bu

113
1) DDQ, CH&
2 ) Dess-Martin Deriodinane
Me
0 ,tBu
f Bu

TBDMSO

114

OH

= 115
Scheme44. Synthesis of cytovaricin (115) by D. A. Evans et al. [78]. DEIPS =
diethylisopropylsilyl. TCEM = trichloroethoxymethyl.

Scheme 45. Use of a single type of protecting group in the synthesis of dynemicin
by S. Ddnishefsky et al. [79]. MOM = methoxymethyl.

116 and 117 were fused to give compound 118 with three MOM
acetal protecting groups, from which the sensitive natural
product 119 was liberated.
The protecting group pattern of an advanced intermediate
was changed by K. C. Nicolaou et al. in the synthesis of
calicheamicin y, [651 (Scheme 46). The complex glycoside 123
was formed from the predominantly acyl-protected unit 120 and
the intermediates 121 and 122. After the coupling, the acetates
were transformed into silyl ethers, which were cleaved in the
same step as the silyl ethers originating from 121 and 122 by
treatment with HF/pyridine in almost quantitative yield.
In the synthesis of the glycosphingolipid Gb, (130). K. C.
Nicolaou et a1.[801swapped the reduction-sensitive benzyl ethers
for the base-labile acetates and thus unified the pattern of protecting groups (Scheme 47). Because the gdlactosylfluoride 125
carried benzyl groups, its coupling with the partially deblocked
glycosyl acceptor 126 yielded the trisaccharide 127 predominantly as the a-anomer. Intermediate 126 carried acyl protecting
groups, guaranteeing a high degree of P-selectivity in the introduction of the sphingosine group (see also Section 3.5). At the
stage of trisaccharide 127 the benzyl protecting groups were
2075

M. Schelhaas and H. Waldmann

REVIEWS

-v

removed by hydrogenolysis and the hydroxyl groups thus liberated were immediately protected as acetates. The complete
trisaccharide unit thus contained exclusively base-labile acyl
protecting groups, which, at the end of the synthesis, were removed simultaneously in high yield. This strategy also proved to
be very effective in the construction of the trimeric Lewis"
nonasaccharide.[*'I

AcO

01

Meofi

TESO Me

oMe

123

.4

Fmoc
Et'MeO

PY
THF/ CHzCIz (5:l)
96%

124

3.4. Introduction of "Stand-Ins"

In the synthesis of polyfunctional compounds, two factors
can make the choice of protecting group strategy particularly
difficult: the lability of the intermediates and the fact that sufficiently different enough protecting groups might not be available for a given type of functional group. Aldehydes and ketones are generally protected as thioacetals or 0,O-acetals. An
example of this is given by the synthesis of FK506 by S.
Schreiber et al.,[411a section of which is shown in Scheme 48.
The terminal aldehyde was masked as a 1,3-dithiane in 131;
however, all attempts to liberate the protected carbonyl group
by established methods afforded the desired product in very low

EfMeO

Scheme46. Change in the pattern of protecting groups in the synthesis of

calicheamicin $, by K. C. Nicolaou et al. [80] NB = nitrobenzyl.
Phl(OCOCF3)z
MeOH / CHzCIz
84%

125

126

.OMpm

127
1) NBS, HF/ py, 89%
2) Hz, Pd(OH)z/C

3) Ac~O,DMAP, py
91% (2 steps)

HC(0H)ZCOzH
AcOH I CHzCIz
70%

128

?)Me

?H
PivO

129

HNK(CHZ),&H3

Pglycoside
NaOMe, MeOH
90%

HO

"'OMe
OMe

130
Scheme 47. Unification of the protecting group pattern in the synthesis of the
glycosphingolipid Gb, (130) by K. C. Nicolaou el al. [Sl].

2076

133
Scheme 48. Application of 1.3-dithiane and dimethyl acetal protecting groups in
the synthesis of the immunosuppressant FK506 by S . L. Schreiber et al. [41].
An.qew. Chem. Int. Ed. Engl. 1996, 35, 2056--2083

Protecting G r o w Chemistry
yields. Only dethioacetalization employing a hypervalent iodine
reagent as a mild oxidant gave satisfactory results. Even so, the
aldehyde thus liberated was not isolated directly in the free state.
but rather as the dimethyl acetal 132. This very sensitive compound yielded 133 with the free aldehyde group after transacetalization with glyoxylic acid under mildly acidic conditions.
This approach is practically identical to that employed by T. K.
Jones et al. in the synthesis of FK506.[821
The example just given illustrates that, despite the effectiveness of thioacetals and acetals as protecting groups, the conditions necessary for their removal often hinder their use. 1,3-Di01s can also be protected as silylene derivatives, which are useful
replacements for the acid-labile acetals (see 114 in Scheme 44).
When such choices are not available, a completely different
tactic must be used.
In one increasingly popular alternative strategy a stand-in
is introduced for the required functional group. In the case of a
carbonyl group, a masked alcohol can be used, which can be
deprotected and subsequently oxidized to give the required aldehyde o r ketone. Although a further protecting group must be
introduced in following this strategy, many more masking
groups are established for the hydroxyl function.
This methodology is illustrated by the synthesis of FK506 by
S. Schreiber et al.14]sketched out in Scheme 21. The very sensitive and reactive tricarbonyl unit in 20 was constructed right at
the end of the synthesis by the deprotection and oxidation of the
corresponding hydroxyl groups. The successive deblocking and
oxidation prevented the formation of mixtures of products resulting from incomplete oxidation. TES- and Mpm-protected
hydroxyl groups were employed as the stand-ins for the carbony1 groups. while the hydroxyl groups found in the final
product (20) were masked as TIPS and TBDMS ethers.
Similarly, in the synthesis of rapamycin by S. Danishefsky et
al.441(Scheme 38) a t a step close to the end of the synthesis, one
of the three protected hydroxyl groups in 83 is selectively deblocked and oxidized to give a carbonyl group.
D. A. Evans et al. also used this strategy in the synthesis of
cyt~varicin~]
(Scheme 44). The hydroxyl group to be oxidized
in 113 was present as the Mpm ether, whereas all other alcohols
were blocked as silyl ethers. After cleavage of the benzylic protecting group with DDQ, the required carbonyl group was
formed and finally the silyl groups removed. This approach was
necessary because the hemiacetal formed in the last step is very
sensitive (see Section 3.3.).

3.5. The Influence of Protecting Groups

on the Course of Syntheses
One factor that should always be taken into account when
planning a protecting group strategy is the potential of the protecting group, under certain circumstances, to actively influence
the course of the synthesis.
For example, the ability (or lack of) of a protecting function
to complex metal ions can be utilized to direct the stereochemical course of a reaction. This use of (non)chelating protecting
groups is one of the best established tools of asymmetric
in the
synthesis.[83]Such an effect was used by J. Mulzer et
synthesis of erythronolide B (Scheme 49). The diastereomeric

REVIEWS

Me

Me

Me

134

PG = BzI
PG = TBDMS

+
135 : 136 = 2:l
135 : 136 = 5:l

OH

PGO

Me

135

PGO
B

OH

z
Me

Me

136

137

138

Scheme 49. Stereochemical directing effects of protecting groups in the synthesis of

erythronolide B by J. Mulzer et al. [84].

ratio of 135 and 136 formed from the addition of propenylrnagnesium bromide to the aldehyde 134 was dependent upon the
type of ether function present in 134. When the secondary hydroxyl group was protected as its benzyl ether, this ratio was
2 : l ; however, the corresponding silyl ether led to a 5 : l ratio
of 135 and 136. The silyl group hindered the formation of a
1,3-chelate leading to anti-136 and favored the development of
a Felkin-Anh transition state leading to the s.vn-diol. The
diastereomeric ratio could be increased further by manipulation
of the protecting groups. The mixture of 135 and 136 was converted to the acetonides 137 and 138. An unfavorable 1,3-diaxial
interaction of the ketone with one of the methyl groups of the
acetal protecting group came into play in 138, which meant that
138 could be equilibrated to the more stable 137 by treatment
with K,CO, in methanol.
BOM, MOM, and MEM ethers are typically used as chelating
protecting groups (see Section 2.1 .I .). Thus, for example,
M. Isobe et al. employed the directing effect of a M E M group
for the stereoselective construction of a precursor to mayt a n ~ i n (Scheme
~~]
50). In the reaction of 139 with methyllithium, the oxygen atoms of the protecting group complex the
lithium cation and thus force the attack of the organometallic
reagent onto the Re side of the double bond in the vinyl sulfone.
The directing effect of protecting groups is typically used in
the synthesis of oligosaccharides, where the course of the glyco-

?+ lM:\Tp

Me3Si> S02Ph

0-07

<o!.ii-Me

Ph02S

Me

Ph

SiMe,

139
SO2Ph

Ph

Me

Scheme 50. Directing influence of the MEM protecting group in the synthesis of a
precursor of maytansin by M. Isobe et al. [85].

2077

M. Schelhaas and H. Waldmann

REVIEWS
sylation can be affected. In glycosyl donors such as 140
(Scheme 51) the ester function in the 2-position shields the underside of the glycosyl cation 142 formed during the glycosylation reaction. Thus, the glycosyl acceptor 143 attacks from
above to form the /I-glycoside 144.[861If the 2-position of the

CICH2C00

140

CICH2C00
68%

142

yo
Me

145

82%

146

BzlO ,oBZl

B z l 0 BzlO
9 ; % 4&

147
AcHN OBzl

OAc

a-glycoside
1

148
B

0O

II,

++I
1

G
OBzl

OH

Tetrasaccharide

Scheme 51. Control over zip-selectivity by choice of suitable protecting groups in

the synthesis of the blood group B tetrasaccharide by B. Fraser-Reid et al. [86].
Pn = pentenyl, NIS = N-iodosuccinimide.

glycosyl donor is not blocked by an acyl group capable of neighboring group participation, as is the case in the benzyl-protected
donor 145, then the a-glycoside, in this case 147, is preferentially
formed. This directing effect was used highly effectively by
B. Fraser-Reid et al. in the synthesis of 147, en route to the
tetrasaccharide characteristic of the human blood group B.
A further example is given in the preparation of the Gb, glycolipid by K. C. Nicolaou et al. sketched in Scheme 47.
The type of protecting group determines not only the stereochemical course of glycoside synthesis, but also affects the reactivity of the glycosyl donor. Since esters are better electron acceptors than ethers, the formation of glycosyl cations is less
favored from an acyl-protected glycosyl donor like 140 than
from an alkyl-protected donor such as 145. The nucleophilicity
2078

of the leaving group at the anomeric center is also less in the

ester case than in the alkyl-protected system. To characterize
this finding, B. Fraser-Reid et al. described these acyl-masked
glycosyl donors as disarmed with respect to activation by
electrophiles, while alkyl-masked donors were viewed as being
armed.187] This concept of armed/disarmed groups has
proved useful for a wide variety of different glycosyl donors.

4. The Best Protecting Group

Is No Protecting Group
The examples discussed in the foregoing sections illustrate the
capabilities of the protecting group techniques available today
and indicate that protecting groups can also be used to influence
the course of reactions. Nevertheless, an unavoidable consequence of the introduction and removal of protecting groups is
that the synthesis is lengthened, and that the use of protecting
groups is thus often seen as unproductive.r841
The number of protecting group manipulations can be kept to
a minimum through the use of the differences in reactivity of
similar o r identical functional groups. In certain cases this can
determine the course of the entire synthesis and the chosen protecting group pattern. This approach is chosen again and again
in the synthesis of glycosides, for example, because reliable information is available concerning the reactivity of the various
hydroxyl groups in carbohydrates. R. R. Schmidt et al. applied
this strategy in the synthesis of the dimeric Lewis determinant
illustrated in Scheme 20.r39]In 12 and 15. the hydroxyl groups
at C-3 and C-4 of the terminal carbohydrate were protected
together as an acetal. After cleavage of the isopropylidene
acetals of 12 and 15, the glycosyl donor 14 reacted regioselectively in both cases with the equatorial groups at C-3 to give
the pentasaccharide 15 and the octasaccharide 15, respectively.
The introduction of a protecting group at C-4 was thus
unneccessary. R. R. Schmidt et al.E881
also utilized the difference
in reactivity of the C-3 and C-4 hydroxyl groups in the syntheses
of the tri- and tetrameric Lewis glycosphingolipids.
Differences in the reactivity of functional groups have also
been utilized in the synthesis of oligopeptides. An example of
this tactic of minimal protection was also used in impressive
fashion by R. Hirschmann and D. F. Veber et aI.IE9in the synthesis of the S-protein of ribonuclease A. This peptide consists
of 103 amino acids and contains all of the trifunctional amino
acids with the exception of tryptophan. As indicated in Scheme
52 for the masked 21 -40 fragment 149, only the c-amino groups
of lysine and the p-mercapto group of the cysteine were masked,
while the side chains of arginine, aspartic acid, and the hydroxyamino acids serine, threonine, and tyrosine remained unprotected. This strategy stands in contrast to the concept of maximum protection, in which the maximum number of functional
groups are masked in order to avoid side reactions. E. Wiinsch
et al. followed this strategy in the synthesis of the peptide hormone glUCdg0nrgo1and protected the side chains of all the hydroxyamino acids. histidine, lysine, and aspartic acid with acidlabile protecting groups. The resulting 7- 15 intermediate 150
(Scheme 52) indicates that a distinctly greater effort must be
made to prepare such synthetic intermediates than is demanded
by the minimal protection tactic.
Angcic. Chrm. lnt. Ed Engl. 1996, 35. 2056-2083

Protecting Group Chemistry

Acm

Of Bu

Boc-Ser-Ser-Ser-Asn-Tyr-Cys-Asn-Gln-Met-Met-Lys-Ser

1) (CHzO),, 98%

21

H,N-Cys-Arg-Asp-Lys-Thr-Leu-Asn- Arg
40

149

Acm

f B u f B u O t B u f B u tBu Boc t 6 u

OfBu

NPS-Thr-Ser-Asp-Tyr-Ser-Lys-Tyr-Leu-Asp-OH
7

ffpy0

HO

2) LDA, NCC02Me

, N-NMe
\

'

ao
CN)

MeN

151

C0,Me

K0 NMe

152

15

150
0
Acm.

MepN?
H

NPS:

153

Scheme 52. The "minimal protection" tactic In the synthesis of the S-protein of
ribonuclease A (11 40 fragment 149) by R. Hirschmann and D. F. Veber el al. [89]
and the "maximum protection'. tactic in the synthesis of glucagon (7-15 fragment
150) by E. Wiinsch et al. [90].

Scheme 53. Use of the triazone protecting group in the synthcsis of strychnine by
L. E. Overman et at. [92].

H-Phe-Trp-Gly,

While the danger of side reactions is greater when one uses

minimal protection, and the number of compatible coupling
techniques is reduced, the tactic of maximum protection has the
disadvantage that losses in yield can occur during the removal
of protecting groups, and that the protected peptide fragments
are often poorly soluble or difficult to purify.

!*No
H

laccase, p~ 4,

154

H-Phe-Trp-Gly,N,N

o2

H-Phe-Trp-Gly-OH + N2

155
Scheme 54. Enzymatic cleavage of phenylhydrazides from peptides by A. N.
Semonov et al. [93].

5. New Developments
Most of the protecting groups discussed in Sections 2 through
4 have been known for some time. Their usefulness has been
demonstrated in a multitude of complex syntheses and they can
be considered to be well-established[lO. (this conclusion cannot yet be drawn for the enzyme-labile protecting groups). Nevertheless, the development of new protecting group techniques
and the improvement of existing methods continues to be the
subject of intensive research in organic synthesis. The following
trends can be distinguished from the large number of new developments.

'

phenylhydrazide at the C-terminus, are oxidized by the enzyme

laccase to give the diimide, in this case 155, which spontaneously
hydrolyzes and liberates the desired p e ~ t i d e .Phenylhydrazides
~~~]
can be cleaved by classical chemical methods only under much
more drastic conditions.

5.2. Development of Protecting Groups that

Induce Alternative Selectivities in Reactions

New developments on protecting groups with modulated reactivity have been described for silyl ethers, in particular (see
Section 2.3). but new acid-labile groups have also been successfully developed and applied. An example is the triazone protecting group for amines developed by s. Knapp et al.,[911which
blocks two N -H positions and can be easily cleaved with dilute
acid. This protecting group was essential in studies by L. E.
Overman et
on the synthesis of the strychnine alkaloids,
as illustrated by the synthesis of strychnine itself (Scheme 53). In
the precursor 151 the amino group of the aniline was masked as
a triazone. which, after a domino sequence consisting of Mannich reaction and aza-Claisen rearrangement, could be removed
from 152, enabling the subsequent ring closure to give 153.If the
tert-butylurethane (Boc) group or a pivalinamide is used in
place of the triazone, the selective cleavage is not successful.
The application of biocatalysts has given the area of orthogo-

The number of blocking/deblocking steps in a synthesis can

be reduced by employing protecting groups which, upon introduction, can distinguish between several functional groups of
similar reactivity. Examples of such protecting groups are the
recently developed dispiroketal (DISPOKE) and cyclohexane1,2-diacetal (CDA) protecting groups,[14. '. 941 which selectively mask diequatorial 1 ,2-diols. In the synthesis of the trisaccharide unit 161 of a streptococcal antigen, the two OH groups in
the 3- and 4-positions of the glycosyl acceptor 156 were selectively blocked and thus the axial hydroxyl group at C-2 was
made available for the coupling reaction with the glycosyl donor
158 (Scheme 55).*15' The disaccharide 159 thus formed was
ready for a further glycosylation; its coupling with 160 and
subsequent cleavage of the CDA and the benzyl groups completed an efficient synthesis of the trisaccharide 161.
The development of reagents for the introduction of protecting groups under alternative conditions is of great interest. Recent resuits in this area include the introduction of allyl, benzyl,
and tert-butyl ethers via the corresponding trichloroacetimidate[951under acidic, rather than the usual basic, conditions.
The formation of benzylidene acetals, normally introduced un-

nally stable protecting groups new impetus. An example of this

is given in Scheme 54. Peptides such as 154, which carry a

der acidic conditions, with the help of benzylidene-l,2-dibromide under basic conditions[961is also a valuable alternative.

5.1. Development of Protecting Groups with

New Lability/Stability

A i i , q m . (%cni In,. Ed Enxi. 1996, 35. 2056-2083

2079

M. Schelhaas and H. Waldmann

REVIEWS
OMe

OM.

EtS

E'S

EtS

Me0

Hod
Ho OH
156

HC(OMe)3

OBzl

:;,LZCO2H*

OMe

157
EtS

one would have a very effective tool for the svnthesis of

oligomers. T. Ogawa et al.[991have recently described such a
system for the construction of oligosaccharides. They exploited
the possibility of orthogonally activating thioglycosides and glycosy1 fluorides (Scheme 57). The thioglycosides 165 and 169

Me?

Me0

HO&
-(' F

Roosph+
I
166

OMe

1) NIS, CF,CO,H

NIS /TfOH (AgOTf)

2) AcOH / HzO
3) HP, Pd-C

Ho OH
161

159

Scheme 55. Synthesis of a rhamnotrisaccharide by S. V. Ley et al. [15] using the

CDA protecting group. NIS = N-iodosuccinimide.

R O - O167
O O F

5.3. Development of Protecting Groups that

Can Be Converted into Activated Groups
If it is possible to use a protecting group that can be converted
into a form that actually activates the molecule for the next F
transformation, then protecting groups become more than just
a "necessary evil" and can be seen to play a n active, positive
role.
An example of the successful realization of such a principle is
given by the use of protecting groups based on o-phenylenedia m i t ~ e ' in
~ ~peptide
]
synthesis (Scheme 56). Under basic condi-

K k!JNQ

a
0

pH=9-10

HN-0

162

165

~ o ~ o ~ o J -

-l"."c

HOG;:

AgC104

~
169

S
---t

h
NISP/ TfOH
(AsO-rf)

- o ~
171

Scheme 57. Construction of oligsaccharides by T. Ogawa et al. [99] using orthogonal activation. NIS = N-iodosuccinimde, TfOH = trifluoromethanesulfonic acid.

were converted into reactive glycosyl donors with N-iodosuccinimide and trifluoromethane sulfonic acid o r with silver(1) trifluoroacetate. The configuration of the anomeric center of the
fluorosugar remained stable under these conditions. However,
when bis(cyclopentadieny1)hafnium dichloride and silver(1)
perchlorate were used, only the glycosyl fluorides 167 and 171
were activated. This switchable activation and direct glycosylation allows a rapid and elegant synthesis of oligosaccharides.
The strategy is highly efficient, and it can be expected that further combinations of protecting groups capable of orthogonal
activation will find use.

5.4. Development of New Anchor Groups

for Solid-Phase Synthesis
164
Scheme 56. Use of protecting groups capable of activation based on o-phenylenediamine in the synthesis of peptides by R. Pascal et a]. [97].

tions peptide derivatives such as 162 react at the urethane function to form 1-acylbenzimidazolin-2-ones such as 163. These
activated compounds can then be coupled directly with amino
acid esters and peptide esters. A further example is the use of
n-pentenyl groups in carbohydrate chemistry'981 (see Section
3.5, Scheme 51). In 140 and 145 the anomeric center is protected
as a pentenyl ether. Treatment with N-iodosuccinimide and triethylsilyl triflate converts the stable, protected 140 and 145 into
the activated glycosyl donors 141 and 146, respectively.
If the principle of converting protecting groups to activating
groups can also be applied to doubIy orthogonal systems, then

2080

In the synthesis of oligopeptides and nucleotides on polymeric

supports, the terminal monomer is attached to the solid phase
by an anchor group. The anchor groups are generally derived
from protecting groups developed for solution-phase applications. Solid-phase synthesis can thus be understood as synthesis
with polymer-modified protecting groups.
Synthesis on polymeric supports has found particular interest
due to the development of combinatorial chemistry."001 The
anchor groups used in solid-phase oligopeptide and nucleotide

chemistry often cannot be directly applied to the synthesis of

other types of compounds. With the growing application of
combinatorial techniques, the demand for new anchor groups
will increase, which will be defined by the requirements of combinatorial chemistry in general, as well as by the properties of
the target compounds. A representative example is given by the
Angel,.. Chem. In!. <I. Engl. 1996. 3s. 205C-2083

Protecting Group Chemistry

REVIEWS

work of S. Fodor et al.,['o'l who have combined modern protecting group chemistry with photolithographic methods for the
parallel synthesis of oligopeptides and oligonucleotides on spatially well-defined regions of a solid support (Scheme 58). The

-lhVl

YG YG YG YG
NH NH NH NH

\\\\\\\\

photodeprotection

YG YG
NHz NHZ NH NH

\\\\T\\\-

A-PG

The development of more efficient protecting group techniques, relying, for example, on differences in reactivity between
similar functional groups (see Section 4) and on new reagents
with alternative reactivities, stabilities, and selectivities, continues to be the focus of attention (see Section 5 ) and will certainly
lead to new protecting group strategies. Protecting group chemistry will thus continue to provide organic synthesis with great
impetus in the future and open up new possibilities.
The basis and motivation ,for this review was research carried
out at the Universities of Mainz, Bonn, und Kurlsruhe aimed at
developing and establishing enzymatically cleavable protecting
groups (see re6 / 2 1 ] ) . Active participants in this work were (in
alphabetic order) :Peter Braun, Alain CottP, Stephanie Gabold,
Simone Glonisda, Axel Heuser, Yolker Jungmunn, Thomas
Kappes, Bruno Klaholz, Karsten Kuhn, Edgar Niigele, Torsten
Pohl, Armin Reidel, Jorg Sander, Bernd Sauerbrei, Michael
Schelhaas, Dagmar Sebastian, and Paul Stiiber. We are grateful
to the Deutsche Forschungsgemeinschaft, the Fonds der Chemischen Industrie, the Bundesministeriurnfur Bildung und Forschung,
BASF AG, Boehringer Mannheim GmbH, and Degussa AG ,for
funding our research.
Received: December 15. 1995
Revised version March 11. 1996 [A14SIE]
German version: Angew. Chem 1996, 108, 2192-2219
Translated by Dr. D . Macqudrrie, York (UK)
[i] K. C . Nicolaou, E. J. Sorensen, Classics in Total Sinth
gies, Method.$, VCH, Weinheim, 1996.
[2] Comprehensive Organic Spnthesis, Vol. 1-8 (Eds.: B. M Trost. I. Fleming),
Pergamon, Oxford, 1991.
[3] Encyclopedia of Reagenrs for Orgunw C/remi.s~r? ( E d . : L. A. Paquette).
Wiley. New York. 1995.
[4] E. J. Corey. X.-M. Cheng. The Logic of Chemical Synrhesi\. Wiley. New York.
1989.
[S] K. F. McClure. S . J. Danishefsky, .
I
Am. Chem. Soc. 1993. 115, 6094-6100.
[6] J. M. Schkeryantz. S . J. Danishefsky, J. Am. Chem. Soic 1995. 117. 4722-

Scheme 58. Light-directed parallel synthesis of oligomers developed by S P. A

Fodor et al. [ l o l l .

surface of the support was initially derivatized with photolabile

protected amino functions. Certain areas of the surface were
then covered with a mask and the whole was then irradiated.
Only the protecting groups in uncovered areas of the surface
were removed, leaving free amino functions. These were chemically coupled with photolabile animo acid derivatives. By using
another mask, other regions can be selectively deblocked and
converted. This process allows a defined sequence to be prepared on a given x e d of the surface.

6. Outlook
The continuous development of new protecting group techniques over the last decades resulted in most of the strategies
discussed in this review and has contributed substantially to the
impressive achievements in organic synthesis today. The successful applications of the silyl and ally1 protecting groups in
complex syntheses are particularly good examples. The authors
are of the opinion that the potential of biocatalysts is equally
vast. Basic methods relying on these systems have been developed in many areas and all that is lacking is definitive proof of
their performance in demanding synthesis.
An,?tw Chiwf. lnr E d EngI. 1996, 35. 2056-2083

4723.
[7] J. J. Masters, J. T. Link. L. B. Snyder, W. B. Young. S. J. Danishefsky, Angeir.
Chem. 1995. 107. 1886-1888; Angew. Chern. In!. Ed. EngI. 1995. 34. 17231726.
[XI Peptide and nucieotide chemistry are, however, exceptions The most advantageous combinations of protecting groups and strategies for the synthesis of
oligopeptides and oligonucleotides have often been commented on in the
relevant publications and textbooks (for examples see Schemes 35 and 52).
Because the types of functions requiring protection in these classes of compounds is limited and the number of reactions to be carried out is restricted,
the problem IS more easily defined in this area of synthesis than in the stereoselective synthesis of a complex natural product. For discussions of protecting
group strategy in peptide synthesis see for examples refs. [35, 371, and references therein, or J. Jones, The Chemical Svnrhrsis of Pi,ptii/cs, Clarendon,
Oxford. 1992; Meihod in Moleculur Eiolog!, Vol 35 P q t i d e Synrhesis
Prorocols (Eds.. M. W. Pennington. B. M. Dunn). Humana Press, Totowa,
Scimcc,) 1994. 37,
NJ (USA), 1994: S. Sakakibara. 5iopolymer.s lPepIid~~
17-28.
191 T. W. Greene, P. G. M. Wuts, Prorrcriw Groirps in Orgunrc Cliemrsrry,
2nded.. Wiley, New York, 1991.
[lo] P. J. Kocienski. Prorecring Groups. 1st ed., Thieme, Stuttgart, 1994.
[ l l ] H. Kunz. H. Waldmann in Comprehensive Organic Synrhe.si.s, Vol. 6 (Eds.:
B. M. Trost. I. Fleming, E. Winterfeldt). Pergamon, Oxford, 1991. pp. 631 701.
[I21 In the introduction in ref. [9] the choice of protecting groups in the total
synthesis of palytoxin by Kishi et al. [77] is discussed (see also Section 3.3).
The synthesis of dihydroerythronolide by R. W. Hoffmann et al (R. Stiirmer,
K. Ritter. R. W. Hoffmann, Angen Chem. 1993. 105, 112- 114; Angew Chem.
h r . Ed. Engl. 1993.32, 101-103) was also analyzed from this point ofview in
ref. [lo].
.
3775-3778.
[13] E. J. Corey. J. W Suggs, Tetrahedron L ~ I I1975.
[14] S. V. Ley, M. Woods, A. Zanossi-Gerosa, Svnrhe.si.5 1992. 52-54.
[lS] a) S . V. Ley, H. M. W. Priepke, S. L. Warriner, Angel!. Chem. 1994, 106,
2410-2412; Angew Chew. In:. Ed Engl. 1994. 33. 2290, b) S.V. Ley,
H. M. W Priepke, ihid. 1994. 106, 2412-2414 and 1994. 33. 2292.
f161 C. Riicker. Chern. Rev. 1995. 95, 1009-1064
[171 E. J. Corey, R . K. Varma. ./. Am. Chem. Soc. 1971, 93. 7319 7320.

208 1

REVIEWS
(181 K. K. Ogilvie, E. A. Thompson, M. A. Quilliam, J. B. Westinore. Gtru/tlcdron
Lett. 1974. 2865-2868.
[19] K. Toshima, S. Mukaiyama. M. Kinoshtta, K. Tatsuta. E,rruhchwi Lett
1989,30.6413-6416
[20] S. Djuric, J. Venit, P. Magnus. Tetrahedron Leu. 1981. 22, 1787-1790. R. P
Bonar-Law, A. P. Davis, B. J. Dorgan, ihid 1990. 31, 6721 -6728.
(211 a) H.Waldmdnn, D. Sebastian, Chem. R P Y .1994. Y4.911-937; b) A. Reidel.
H. Wdldmann. J Prukt. Chem. 1993, 335, 109-127; c) T. Pohl, E. Nigele.
H. Waldmann, Coru/i.sis T o d q 1994. 22. 407-426; d ) H. Waldmann in 122).
Vol. 2, pp. 851 -886.
[22] Enzjnte Cura(v.si.s in Organic Clicmisrrj (Eds: K. Drauz. H. Waldmann),
VCH. Weinheim. 1995.
[23] H. Wdldmann. Tetruhedron Lelt. 1988. 29. 1131 -1 134; Liebigs Ann. Cltem.
1988, 1175-1180.
[24] H. Wdldmdnn. E. Nigele. Angew. Cliem. 1995, 107, 2425 2428. Angeir.
Chem. Inr. Ed. Engl. 1995. 34, 2259-2262.
1251 a) P. Braun, H. Waldmdnn, W. Vogt, H. Kunz. Sj/tlerr 1990. 105-107:
b) Liehigs Ann. Cltem. 1991, 165-170.
[26] M. Schelhaas, S. Glomsdd. M. Hinder, H.-D. Jakubke, H. Waldmann,
A n g w . Chem. 1996, 108, 82-85; Angew. Chem. Inr. E d Engl. 1996, 35. 106 109.
127) H. Kunz, D. Kowalczyk, P. Braun. Angei~. C h n . 1994. 106, 353-355:
Angels. Chen. I n l . Ed. Engl. 1994, 33, 336-339.
1281 a) P. Fugedi, A. Liprdk. P. Nanasi. Curbohvdr. Rrs. 1982. 104. 55-67, b) P. J.
Garegg, H . Hultberg, S. Wallin, ibid. 1982. 108, 97-101
1291 a) G. Barany. R. B. Merrifield, J Am. Cheni. Soc. 1977, 99.7363-7365: b) E.
Meinjohanns. M. Meldal, H. Paulsen. K. Bock. 1 Cliem. Soc. Perkin Triins. 1
1995. 405-415.
[30J J. A. Barltrop, P. J. Plant. P. Schofield, Chem. Commun. 1966, X22-823.
[31] a) J. C. Sheehdn, R. M. Wilson, A. W. J. Oxford, J. A m C/tem. Soc. 1971. 93,
7222-7728; b) R. S. Givens, Ph. S. Athey. B. Matuszewski. L. W. Kueper 111.
J.-Y. Xue, T. Fister. hid. 1993, f15. 6001-6012.
[32] M C. Pirrung, Y. R. Lee, J. Org. Cliem. 1993, 58, 6961 -6963.
1331 a) A. T. Tesser, 1. C Bahert-Geers. In,. J Pepf. Prorein Res. 1975. 7.295-305:
b) G. G . J. Verhart, G . I. Tesser, Reel. Trav. Chini. Paw-Bas 1988. 107. 621 626.
[34] a) F. Nakatsubo, T. Fukuyama. A. 3. Cocuzzd. Y. Kishi, J. Am. Chent. Sor.
1977, 99, 8115-8116; b) Terruhedron Left. 1977,4295-4298.
1351 J. Rudinger. Pure Appl. Clteni. 1963, 7,335-362; recent review: M. Patek, Inr.
J Pepr. Protein Res. 1993, 42. 97-117.
(361 M. T Reetz. M. W. Drewes, A. Schinitz. Angeic. Chem. 1987, 99. 1186- 1188;
Angerr.. Chem. Inr. Ed. Engl. 1987. 26. 1141-1143.
[37] R. B. Merrifield, Pepl., Proc.. Am. Pepr. S.ymp., 5rh 1977. 488.
[38] a) R. C. Kelly. I. Gebhard, N. Wicnienski. J Or:. Chem. 1986. Sf. 45904594; b) G . R. Pettit, Y. Kamano, C. W. Holzdpfel. W. J. van Zyl. A. A. Tuinman, C. L. Herald, L. Baczynskyj, J. M. Schmidt. J A m Chem. Soc. 1987,
109, 7581 -7582.
[39] R. Windmuller. R. R. Schmidt, Tetrahedron Lert. 1994. 35, 7927-7930.
[40] H. Kunz, C. Unverzagt. Angeir. Cliem. 1988, 100. 1763 - 1765; Angeit.. Chem.
Int. Ed. Engl. 1988. 27. 1697- 1699.
[41] M. Nakatsuka, J. A. Ragan, T. Sammakia. D. B. Smith, D E. Uehling, S. L.
Schreiber, J Am. C/tem. Soc. 1990, 112, 5583-5601.
(421 a) A. D. Piscopio, N. Minowa. T. K . Chakraborty, K. Koide. P. Bertinato,
K. C. Nicolaou, J Chem. Soc. Chem. Commun. 1993, 617-618; ibid. 1993,
619-622; b) K. C. Nicolaou. T. K. Chakraborty, A. D. Piscopio, N. Minowa.
P. Bertinato, J. Am. Chem. So<. 1993, 115, 4419-4420.
1431 H. Kunz. H . Waldmann. Angels. Chem. 198597,885-887; Angeii'. Chem. lnr.
Ed. Engl. 1985, 24, 883-885.
(441 C. M. Hayward, D. Yohannes. S. J. Danishefski, 1 Am. Ciimi. Suc. 1993, /IS,
9345-9346.
[451 Y. Hayakawa. S. Wdkabayashi. H. Kato, R. Noyori, J. An?. Chem Soc. 1990.
112. 1691 -1696.
1461 S. Blechert, A. Kleine-Klausing, Angew. CItem. 1991, 103, 428-429; Angeit.
C/iem. Inr. Ed Engl. 1991, 30. 412-414.
1471 D . A. Evans, J. A. Ellman, K . M . DeVries. J. A m . Chem. Soc. 1989. ffi,
8912-8914.
[48] P. Braun, H. Waldmann. H. Kunz, S d e t r 1992. 39-40; Bionrg. Med. Cliem.
1993, 1 , 197-207.
[49] T Pohl. H. Waldmann, Angew. Chent. 1996, 108. 1829-1832, A n g w . Chem.
Inr. Ed. Engl. 1996, 35. 1720-1723.
[SO] H. Kunz, An'qew. Chem. 1987, 99, 297-311; Angew. Clieni. In/. Ed. EngI.
1987,26,294-308.

[51] B. Danieli, G . Lesma, M.Mauro, G. Palmisano, D. Passarella, Terruhedrow

1994,50, 8837-8852.
(521 A. L. Margolin. Enzyme Mieroh. Technol. 1993.5.266-280; K. Mori. Syn/arr
1995, I097 - 1109.
[53] M. Ohno, Y 110, M. Arita, T. Shibdtd, K. Adachi, H. Sawai. Terruhd-on
1984. 40. 145-152.
[54] S. Iruchijima, K. Hasegdwa. G. Tsuchihashi, Agric. B i d Chem. 1982. 46.
1907- 1910.

2082

M. Schelhaas and H. Waldmann

(551 J. Fried. V. John, M. J. Szwedo, J r , C.-K. Chen, C. O'Yang. T A. Morinelfi.
A. K. Okwu. P. V. Halushka. J A m . Clrem. Soc. 1989, I f f , 4510-4511.
[56] S. Kobayashi, M. Nakada. M. Ohno. Pure Appl. Clzem. 1993, 64, 11211124.
[57] M. Arita, K. Adachi, Y. Ito. H. Sawai. M. 0 h n o . J Am. Cliem. Soc. 1983, 105.
4049 - 405 5.
1581 X. K Holdgrun. C. J. Sih, Terruhedron Lerf. 1991. 32, 3465-3468.
[59] M. Schneider. M. Engel. H. Boensmann. Angew. Cliem. 1984. 96, 52-54;
A n p w . C/iiwi In!. Ed. En:/. 19&, 23, 64-66.
[60] P. Mohr, N. Waespe-Sarceivic. C. Tarnm. K. Gawronska. J. K . Gawronski,
Hr11~.C h i . Aclo 1983. 66, 2501 -251 1.
[61] K. J. Harris, Q.-M. Gu. Y. E. Shih. G Girdaukas, C . J. Sih, Tetruhrdron Lett.
1991,32. 3941 -3944.
[62] JLL. Breve[. K. Mori, Syzrhesis 1992. 1007- 1012.
[63] K. 0. Hallinan. T. Honda, Tetrahedron 1995, 51. 12211-12216.
1641 a) V. R. N. Pillai, S.lnrhesrs 1980, 1-26; b) Org. Pholochem. 1987, 225; c) U.
Zehavi, Adr. C'urhdtwlr. Cheni. Biodiem. 1988. 46. 179-204.
1651 a) R. D. Groneberg. T. Miyazaki, N. A. Stylianides, T. J. Schulze. W. Stahl,
E. P. Schremer, T Suzuki. A. L Schmlth. K. C. Nicolaou, J Am. Chem. Soc.
1993. 115. 7593-7611; b) A. L. Schmith. E. N. Pitsinos, C:H. Hwang. Y.
Mizuno. H. Sdttonlo. G . R. Scarlato, T. Suzuki. K. C Nicolaou, rbid. 1993,
115, 7612-7624: c) K C. Nicolaou, M. Nakada, K. Shibayama. E. N. Pitsinos, H. Saimoto. Y. Mizuno. K:U. Baldenius, A. L. Schmith. ibid. 1993, 115,
7625 - 7635.
[66] S. R. Adams. J. P. Y. Kao, R. Y. Tsien. J A m . C/iem. So?. 1989, 111, 7957.7968.
1671 M. C. Pirrung. S. W. Shuey, J Org. Cliem. 1994, 59, 3890-3897
1681 R. S. Givens, P. S. Athey. L. W. Kueper 111, B. Matuszewski. J.-Y. Xue, J Am.
C/iem. Soc. 1992, 114. X708-X710.
[69] A. Barth. K. Hauser. W. MHntele. J. E. T. Corrie. D. R. Trentham, J A m
Chem. Soc. 1995. 117, 10311-10316.
[70] R. A. Ellison, E. R. Lukenbach. C. W Chiu. li.truhedron Lert. 1975. 499502.
1711 P. Sieber. B. Kamber, A. Hartmann. A Johl, B. Riniker. W. Rittel. H d v .
Cliim. Actu 1974, 57, 2617-2621
[72] K. C. Nicolaou, E. P. Schreiner. Y. Iwabuchi, T. Suzuki. Angeiv. Chem. 1992,
104. 317-319. Angecr. Chem. Inr. Ed. Engl. 1992, 31. 340-342.
I731 a) R. A. Holton, C. Somoza, H -B. Kim. F. Liang, R. J. Biedinger, P. D.

Boatman. M . Shindo, C C. Smith. S. Kim, H. Nadizadeh. Y Suzukl, C. Tao,

P. Vu. S. Tang, P. Zhang. K. K. Murthi. L. N. Gentile, J. H. Liu, J Am. Chem.
Sw. 1994. 116. 1597-1598, 1599-1600.
[74] K Horita. T. Yoshioka. T. Tanaka. Y Otkawa, 0. Yonemitsu, Tcrruhedron
1986. 42. 3021 - 3028.
[75] T. Tdnaka, Y. Oikawd. T Hamada, 0. Yonemitsu, Terruhedron Lell. 1986,27.
3651 -3654.
1761 H. Kunz. J. MBrz, Angew. Chem. 1988, 100. 1424-1425; Angrw. Chem. Inr.
Ed. Engl. 1988, 27. 1375--1376.
[77] a) R. W. Armstrong, J. M . Beau, S. H . Cheon, W. J. Christ, H. Fujiokd, W. H.

Ham. L. D. Hawkins. H. Jin. S. H Kang. Y Kishi, M. J. Martinelli. W. D.

McWhorter. Jr., M. Mizuno. M. Nakata, A. E. Stutz, F. X. Talamas. M.
Taniguchi. J. A. Tino. K. Uedd, J. I. Uenishi, J. B. White, M . Yonaga. J Am.
Chem. Soc. 1989. 111. 7525-7530. 7530-7533.
1781 D. A. Evans, S. W. Kaldor, T. K. Jones, J. Cardy, T. J. Stout, 1 Am. Chrm.
Soc. 1990. i12, 7001 -7031
[79] M. D Shair. T-Y. Yoon. S. J. Ddnishefsky. Angeii. Chem. 1995, 107, 18831885: Angea. C h m . I n [ . Ed. EngI. 1995, 34. 1720-1723.
1801 a) K. C. Nicolaou. T. J. Caulfield, H. Kataoka, T. Kumdzawd, J Am. Chen?.
Soc. 1988. 110. 7910-7912; b) K . C. Nicolaou. C/temrrarts .- Org. Chem
1991, 4. 181-198.
[81] K. C. Nicolaou. T. J. Caulfield. H. Kataoka, N. A. Stylianides, J Am. Chem.
Soc. 1990. 112, 3693-3695.

(821 T. K . Jones, S . G. Mills. R. A. Reamer, D. Askin. R . Desmond. R . P. Volante,

1. Shinkai. J Am. C / z m . Sot. 1989. 111. 1157-1159.
1831 For example. see: M. Nogrady. Srereoselecrrar S i n t h e m , 2nd ed., VCH,
Weinheim. 1994.
[84] J. Mulzer, H. M. Kirstein, J Buschmann. C. Lehmann. P. J. Luger, J. Am.
Chem. Soc. 1991, 113. 910-923.
[85] M. Isobe, M. Kitamura. T. Goto. Tetruhedron Lrrt. 1979. 3465-3468.
(861 U . E. Udodong. C. S. Rao, B. Fraser-Reid. Tetrahedron 1992, 48, 4713-

4724.
(871 a) D. R. Mootoo, P. Konradsson, U. Udodong. B. Fraser-Reid. J A m . Chem.
Soc. 1988, 110, 5583-5584: b) B. Fraser-Reid, U. E. Udodong, 2. Wu. H.
Ottoson. R. Merrit. C. S. Rao. C. Roberts. Svnlert 1992. 927-942.
1881 A. Toepfer, W. Kinzy, R. R. Schmidt, Liebigs Ann. Cltem. 1994, 449-464.
[89] R. G . Denkewalter. D. E Veber. F. W. Holly, R. Hirschmann, J A m . Chrm.
Soc. 1969. 91. 502 --503.
1901 a ) E. Wunsch. G . Wendlberger. Chem. Ber. 1968, 101,341 -345, b) ihid. 1968.
101, 3659-3663.
1911 a) S. Knapp. J. J. Hale. M. Bastos, F. S. Gibson, Tcrrahedron LeIr. 1990, 31,
2109-2112; b)S. Knapp, J. J. Hale. M. Bast0s.A. Molina, K. Y Chen.J Org.
C/iem. 1992. 57. 6239-6256.

A n g e x Chem. I n t . Ed. Engl. 1996. 35, 2056-2083

REVIEWS

Protecting Group Chemistry

[97] a ) S. R. Angle. J. M. Fevip, S. D. Knight, R. W. Marquis. Jr., L. E. Overman.
J. A m . Chein. Soc. 1993, 115,3966-3976; b) S. D. Knight. L. E. Overman. G.
Pairaudeau. ihid. 1993, 115. 9293-9294, c) ihiri. 1995. I17. 5776-5788.
[93] A. N . Srmeno\. 1. V. Lomonosova. V. I. Berezin. M. 1. Titov. Biotech. Biueng
1993.42. 1137 1141.
1941 Short recicw. T Ziegler. Angeir. Chr.171.1994. 106.2362-2365; Angen. Chem.
I n / . f,X Erigl. 1994. 33. 2212 -2275.
[95] a ) T. Iverson. 1). R. Bundle, J. Chen?. Sor. Cheni. Commun. 1981, 1240- 1241;
b) H.-P Wessel, T Iverson. D. R. Bundle, J. Chem. Sor. Perkin Trans. 1 1985.
1247 2250.
[9h] a ) H . G. Fletcher. Jr.. Meihods Curbohvdr. Chem. 1963, 2. 307; b) M. E.
Evans. i h d 1980. 8. 313-315; c ) P. J. Caregg. C. G. Swahn. Acta Chem.
Scoiul. 1972. 26. 3895 -3901

[97] R. Pascal, D. Chauvey, R. Sola. Tefrahedron Let!. 1994. 35, 62916294.

[98] a) B. Fraser-Reid. P. Konradsson, D. R. Mootoo. U. Ud0dong.L C h e ~ ~SOC.
i.
Chem. Comniun. 1988. 823 - 825.
1991 0 . Kame, Y Ito, T. Ogawa. J. Am. Chen7. Soc. 1994, 116. 12073-12074; H.
Paulsen. Angew. Chern. 1995. 107. 1562- 1564; AnRew. Chon. I n I . Ed. &I.
1995,34, 1432-1434.
[loo] a) M. A. Gallop. R. W. Barrett, W. J. Dower, S . P. A. Fodor. E. M. Gordon,
J. Med. Chern 1994. 37, 1233-1251; b) ihid 1994, 37. 1385-1401
[lo11 S P. A. Fodor. J. L. Read. M. C. Pirrung, L. Strycr. A. Tsa~ Lu.
D. Solas, Science 1994, 251. 767-773, see also: G. v. Kiedrowski,
Angea.. Chern 1991. 103. 839- 840; Angebr Cheni. Int. Ed. Engl. 1991,
30, 822.

Time is money...
Save both by starting a personal subscription
to Angewandte Chemie
(now with Chemistry - A ,European Journal).
At 9e per page you can't go wrong.
With your own private copy you can read the
"hot" articles before they cool, and waste
less time in front of the copier.

Order now!
An order form can be found on
the last page of this issue.
L

A ~ R P HChi~ni.fn/ E d Engl. 1996, 35, 2056-2083

2083