Urokinase-A Strong Plasminogen Activator PDF
Urokinase-A Strong Plasminogen Activator PDF
Urokinase-A Strong Plasminogen Activator PDF
net/publication/228652697
CITATIONS READS
4 653
5 authors, including:
63 PUBLICATIONS 2,891 CITATIONS
BMS College of Engineering
14 PUBLICATIONS 163 CITATIONS
SEE PROFILE
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Saisha Vinjamuri on 27 May 2014.
Standard Review
Urokinase (UK) is a serine protease, which specifically cleaves the proenzyme/zymogen plasminogen to
form the active enzyme plasmin. It specifically catalyzes the cleavage of the Arg-Val bond in
plasminogen. The active plasmin is then able to break down the fibrin polymers of blood clots.
Clinically, UK is given to patients suffering from thrombolytic disorders. Among the plasminogen
activators, UK provides a superior alternative for the simple reasons of its being more potent as
compared to tissue-plasminogen activator and non-antigenic by virtue of its human origin unlike
streptokinase. Based on these observations, UK is a strong plasminogen activator. Hence, UK, as one
of the most potent plasminogen activators is attracting a great deal of attention. The mechanism of
action, physico-chemical properties, in vitro production, cloning and expression, and clinical
applications of UK are reviewed in this paper.
Key words: Urokinase (UK), plasminogen activators, fibrinolysis, strong plasminogen activator, production,
cloning and expression, physico-chemical properties and clinical applications.
INTRODUCTION
Urokinase (UK) is given to patients suffering from throm- 1988). Therapy in thrombosis has been directed towards
bolytic disorders like deep vein thrombosis, thrombosis of interference with coagulation mechanism, activation of
the eye, pulmonary embolism, and myocardial infarction. fibrinolytic system, interference with platelet aggregation
This enzyme is a strong plasminogen activator which or combination of these. In addition, surgical intervention
specifically cleaves the proenzyme/zymogen plasmino- to prevent embolization or to remove thrombi and restore
gen to form the active enzyme plasmin (Lesuk et al., blood flow is of critical importance.
1967; Kunamneni et al., 2008). It specifically catalyzes Anticoagulant therapy includes the use of either cou-
the cleavage of the arg–val bond in plasminogen. The marin drugs, heparin, or a combination of both. The cou-
active plasmin is then able to break down the fibrin poly- marin drugs (warfarin, coumarin and dicoumarol) act by
mers of blood clots (Bennart and Francis, 1976). The ini- antagonizing vitamin K and result in depression of the
tial main source of UK was urine as described by Wil- concentration of clotting factors [prothrombin (factor II),
liams (1951). Later UK was prepared from cultures of factors VII, IX and X].
human embryonic kidney cells (Bernick and Kwaan, Thrombolytics are the drugs used to lyse thrombi to
1967). recanalyze occluded blood vessels. They are curative
In recent years, thrombolytic therapy has revolutionized rather than prophylactic. They work by activating the
the treatment of these diverse circulatory disorders such natural fibrinolytic system.
as pulmonary embolism, myocardial infarction, deep vein
thrombosis, thrombosis of the eye etc., (Collen et al., The plasma fibrinolytic system
Fibrinolysis
*Corresponding author. E-mail: adikunamneni@rediffmail.com. Fibrinolysis refers to the dissolution of the fibrin blood clot
Tel: +34 915855479. Fax: +34 91-5854760. by an enzyme system present in the blood of all mamma-
Kunamneni et al. 059
lian species (Castillino 1981). The fibrinolytic system con- probably megakaryocyte-derived and is stored in platelet
sists of the plasma zymogen, plasminogen; its activated and granules.
product, the proteolytic enzyme, plasmin; plasminogen The activator in urine, UK, differs structurally from t-PA
activators (PAs); inhibitors of both plasmin and PAs and and is produced primarily in the kidneys and excreted into
fibrinogen and fibrin. the urine where it may help to maintain urinary tract po-
The basic reaction of the plasma fibrinolytic system is tency. Endothelial UK probably contributes a small pro-
the conversion of a plasminogen to the active proteolytic portion of plasma activator activity. Factor XIIa not only
enzyme plasmin, by a limited proteolytic cleavage medi- intimates coagulation, but also accelerates the conver-
ated by different PAs (Castellino, 1984). The PAs are sion of plasminogen to plasmin via a proactivator, kalli-
synthesized and released from endothelial cells and other krein.
tissues. Plasmin has the capacity to hydrolyse fibrin and When clotting occurs, a small amount of plasminogen
various plasma coagulation proteins, including fibrinogen. is trapped in the fibrin strands. PA released locally from
The activity of the fibrinolytic system is modulated by the vascular endothelium or traumatized tissues, binds to
inhibitors that inhibit both PAs and the proteolytic effect of the fibrin of the thrombus and converts plasminogen to
plasmin. The main players of the fibrinolytic system are plasmin, itself bound to its surface fibrin and in this
plasminogen itself, the zymogen of a trypsin like serine conformation protected from its otherwise highly effective
protease, two activators of plasminogen and three prote- inhibitor α2 antiplasmin. Fibrin is then digested. There is
ase inhibitors (Francis and Marder, 1994). little or no plasma fibrinolytic activity because plasmin
The major two activators that occur in the circulating that is formed in the blood stream from activation of
blood are: the tissue type plasminogen activator (t-PA) plasma plasminogen is rapidly inactivated by circulating
and the urinary type plasminogen activator α2 antiplasmin. The α2 macroglobulin also acts as a
(u-PA) also called urokinase (UK). secondary plasmin inhibitor in the presence of excess
plasmin.
Fibrinolytic system
Fibrinolytics
Extensive studies have been made over the last 25 years
to understand the physiology of the fibrin-clot formation As in the inflammation reaction, the clotting of blood is an
(Wu and Thaigarajan, 1996). example of a defense mechanism, which can overreact
The fibrinolytic system and its constituents directly and require therapeutic intervention. It is now known that
responsible for the dissolution of the fibrin clot are briefly in normal tissues, there is a constant dynamic equilibrium
described in Figure 1 (Bick, 1992).
between blood coagulation (clotting) and fibrinolysis (the
process of dissolving the clotted blood). The maintenance
Plasminogen-plasmin system of proper balance in this equilibrium is extremely impor-
tant. If fibrinolysis is increased by a pathological cause, a
Plasminogen is a glycoprotein of molecular weight 90- predisposal to excessive bleeding results. On the other
KDa, which is synthesized in the liver. It is converted hand, if fibrinolysis is weakened so that clot formation is
enzymatically by PAs to the fibrinolytic enzyme, plasmin. favoured, conditions occur that are called thromboses
cuts away its covalently cross-linked α- chain protube- (clots formed in and remaining in blood vessels) and
rances. The rather open mesh like structure of a blood embolisms (sudden blockages of blood vessels caused
clot gives plasmin relatively free access to the polymer- by circulating fragments of clots). These can be life
rized fibrin molecules thereby facilitating clot lysis. threatening. In chronic cases, cholesterol and other fatty
Plasmin, a plasma serine protease that specifically materials may aggregate around clotted deposits in blood
cleaves fibrins tripple stranded coiled coil segment and vessels. When this pathological condition is well esta-
Plasmin is formed through the proteolytic cleavage of the blished, it is called atherosclerosis, hardening of the
86-KDa zymogen plasminogen, a protein that is homolo- arteries.
gous to the zymogens of the blood-clotting cascade. Although there are chemical anti-coagulants (such as
PA is present in the tissue (t-PA), in plasma and in urine heparin) available as well as corrective surgical techni-
(UK). t-PA is localized in the vascular endothelium of ques, acute thrombosis and embolisms (which are lump-
veins, capillaries and pulmonary arteries and in the ed together as acute thromboembolic vascular diseases)
microsomal fraction of cells. t-PA is released into the are still the largest single cause of death and disease in
blood stream in response to a number of stimuli, include- the middle-aged and elderly populations of the Western
ing ischaemia, vasoactive drugs and exercise. Released World.
activator is inactivated rapidly in the blood stream by The mechanism used by the body for controlling the
complexing to tissue plasminogen activator inhibitors equilibrium between clot formation and dissolution is a
(PAIs) and has a half-life of about five minutes. complex one involving a series of enzymes, proenzymes,
The major tissue PAIs are synthesized in the liver and activators, and proactivators. A brief outline of the high-
in vascular endothelium, but about 30% of the total is pro- lights of the process will be helpful in understanding the
060 Biotechnol. Mol. Biol. Rev.
Extrinsic Intrinsic
rt-PA t-PA
Fibrin (insoluble)
Plasminogen Plasmin
(Pro fibrinolysin)
Extrinsic Intrinsic
Aprotnin α2-macroglobulin
EACA
Fibrin (insoluble)
Plasminogen Plasmin
(Pro fibrinolysin)
cance in disease (Sherry and Troll, 1954), and potential plate method for the estimation of UK and non-UK fibrino-
for thrombolytic therapy (Fletcher et al., 1962). lytic activity in protease-inhibitor-deprived plasma. In this
Existing assay methods can be divided into three major method fibrin clots, with a suitable dye incorporated, were
groups: formed in wells of standard high adsorption micro-titer
plates.
(1). The first group consists of indirect assay of UK activity Roche et al. (1983) presented a rapid and highly sensi-
with protein as substrates eg., the fibrin plate method tive solid-phase radio assay for the measurement of PAs.
125
(Haverkate and Bradman, 1975), clot lysis method The method employs a convenient and stable I – fibri-
(Lassen 1958) and caseinolytic methods (Kline, 1971). nogen – latex bead product and can reproducibly detect
PAs cannot be assayed directly by these methods but 0.25 milli PU/ml of UK. This represents a 100-fold
only via their activating action on plasminogen present in increase in sensitivity of UK over radio isotopic solid-
or added to the substrate. With the fibrin plate method phase technique and a 120-fold increase over the sensiti-
extremely low levels of enzymatic activity can be deter- vity of the fibrin plate method.
mined but long incubation times are required, and the
responses found depend on the quality of the fibrin sub-
strate (Haverkate and Bradman, 1975). Esterolytic assays
(2) A second group of assays, suitable for kinetic studies Sherry et al. (1966) investigated the ability of UK to
is based on the determination of hydrolysis of synthetic hydrolyze a variety of alpha amino substituted Arg and
peptide esters (Bell et al., 1974). The disadvantage of the Lys esters [acetyl -arg methyl ester, benzoyl –arg methyl
use of these esters is that esterolytic rather than amido- ester, tosyl –arg methyl ester, lys methyl ester, acetyl –lys
lytic activity of a proteolytic enzyme is measured. methyl ester, benzoyl –lys methyl ester and tosyl –lys
(3) In the third group of synthetic substrates e.g., Acetyl L- methyl ester]. Their observations indicated that UK
lysine p-nitroanilines, amidolysis of the peptide amides is catalyses a more rapid hydrolysis of lysine esters and its
measured (Petkov et al., 1973). derivatives than the corresponding esters of arginine.
Substitution of alpha amino group of lysine methyl esters
Fibrinolytic assays increases the sensitivity of the ester to hydrolysis. They
further reported that acetyl -lys methyl ester is the most
The standard system used for measuring PA in cells is an sensitive substrate among the various esters tested.
indirect, two-step assay in which plasminogen is incu- A convenient and highly sensitive colorimetric assay for
bated with a source of PA and the plasmin activity gene- various proteases such as trypsin, chymotrypsin, plasmin,
rated is quantitated by using fibrin, casein or protamine as thrombin and UK was reported (Ninobe et al., 1980). The
substrates (Unkeless et al., 1974; Goldberg, 1974; substrates used were naphthyl ester derivatives of N-tosyl
Kessner and Troll, 1976). L-lysine, N-acetyl glycyl L-lysine and N-acetyl L-tyrosine.
Marsh and Gaffney (1977) developed a rapid fibrin plate Activity was assayed by colorimetric determination of
method for plasminogen activator assay. Their study was naphthol released. They reported that this method was
carried out to investigate plasminogen-enrichment as a more sensitive than the use of corresponding methyl or
means of shortening the incubation period, which is asso- ethyl ester derivatives.
ciated with the fibrin plate method. Fibrin plates were Barlow and Marder (1980) reported the use of a chro-
made up to contain 2 casein units of added plasminogen. mogenic substrate L–pyroglutamyl glycyl L-arginine p-
Each was opaque, firm, did not lyse spontaneously and nitroanilide [S-2444] for assay of plasma UK levels of
yielded biometrically valid parallel-line assays for SK and patients treated with urinary source or tissue culture
UK. source of UK. The p-nitroaniline released was measured
Jespersen and Astrup (1983) described the reproduce- in a spectrophotometer at 405nm. A linear response rela-
bility, precision and required conditions of the firbin plate tionship between UK concentration and optical activity
method for determination of fibrinolytic agents. was obtained, indicating that the method detects UK in
Under optimal conditions the assay is sensitive and pre- quantitative manner.
cise method for the quantitative determination of firbino- Kulseth and Helgeland (1993) developed a simple and
lytic agents. highly sensitive chromogenic microplate assay for quanti-
Millar and Smith (1983) compared the rapid and highly fication of rat and human plasminogen in plasma samples
sensitive solid phase assay with the fibrin plate method and subcellular fractions. The assay is based on a con-
for the measurement of UK, SK and the PA in human version of plasminogen to plasmin, using UK as an acti-
euglobulin fractions. The solid phase assay was run using vator, and a subsequent cleavage of chromogenic plas-
Glu- or Lys- plasminogen, and significant differences were min substrate D-alanyl-L-cyclohexylanyl-L-lysine-p-nitro-
observed in the activation of the plasminogen by UK and anilide-dihydroacetate.
SK. Very good agreement was obtained between the p-Nitroaniline being released by the cleavage is then
fibrin plate and solid phase methods in all cases. measured at 410 nm with a microplate reader. The assay
Fossum and Hoem (1996) developed a firbin micro- includes an acidification step to make plasminogen more
Kunamneni et al. 063
readily activated to plasmin. The method is suitable for ing the CM with antibodies to rhHGF, and which was
analyses of a large number of samples, measuring plas- mimicked by exogenously added rhHGF (31.3 fold
minogen in the nanogram range (0.5 - 50ng/50 l of sam- increase). These results demonstrated that HGF, which
ple). induces thrombogenesis by MDCK cells in vitro, also
increases u-PA and u-PA receptor expression in these
cells. This suggests that the resulting increase in extra-
Fluorimetric assays cellular proteolysis, appropriately localized to the cell
Kessner and Troll (1976) reported a new method for surface, is required for epithelial morphogenesis.
determining plasminogen activator levels. The assay is By the combined use of zymographies on tissue secre-
based on the digestion of N-terminal blocked protamine tions in situ hybridizations, Sappino et al. (1991) explain-
and subsequent measurement of the exposed amino ed the cellular distribution of u-PA and t-PA and of their
groups using the flurogenic amine reagent, Fluram. mRNAs in developing adult mouse kidneys. In 17.5-day-
Nieuwenhuizen et al. (1978) reported fluorigenic subst- old embryos, renal tubules synthesized u-PA while S-
rates for sensitive and differential estimation of UK and t- shaped bodies produced t-PA. In the adult kidney, u-PA
PA. Two fluorigenic peptide amides have been synthe- is synthesized and released in urine by the epithelial cells
sized, i.e. Boc L-valyl-glycyl-L-arginine-L-naphthylamide during lining the straight parts of both proximal and distal
and L-valyl-glycyl-L-arginine-2-naphthylamide. The kinetic tubules. In contrast t-PA is produced by glomerular cells
parameters of plasmin, UK and human uterine tissue and by epithelial cells lining the distal parts of collecting
plasminogen activator on substrates 1 and 11 have been ducts. The precise segmental distribution of PAs sug-
determined. gested that both enzymes might be implicated in the
Zimmerman et al. (1978) developed a simple, sensitive, maintenance of tubular potency, by catalyzing extracel-
direct assay that allowed both rapid measurement and lular proteolysis to prevent or circumvent protein precipi-
kinetic analysis of PA, independent of plasmin generation. tation.
The method employed a synthetic flurogenic peptide sub- Valinskey et al. (1981) studied the association of con-
strate 7-(N-CbZ-glycylglycyl argininamido)- 4-methyl trolled extracellular proteolysis mediated by PA with
coumarin trifluoro acetate. The assay correlated well with embryonic tissue remodeling and cell migration in the
the standard
125
I-labeled fibrin plate assay using highly developing Bursa fabricius of quail and chick embryos.
purified UK. Wojta et al. (1989) stated that vascular origin deter-
mines PA expression in human endothelial cells and
renal endothelial cells produce large amounts of scu-PA.
In vitro production of UK
Roychoudhury et al. (1999) carried out studies in T-
The presence in urine of an activator substance capable flasks and bioreactor to produce UK using HT 1080 hu-
of effecting transformation of plasminogen to plasmin was man kidney cell line. While growing the cell line it has
first described by Williams (1951) and in the following been observed that the lag phase is reduced consi-
year by Sobel et al. (1952). The latter group has assigned derably in the bioreactor as compared to T-flask culture.
the name UROKINASE to this activator. The HT 1080 cell adhesion rate and UK production were
Pepper et al. (1992) demonstrated that fibroblast – observed to be the function of serum concentration in the
-4
conditioned medium induces Madin-Darby canine kidney medium. The maximum UK activity of 3.1 x 10 PU/ml
(MDCK) epithelial cells to form branching tubules when was achieved in the bioreactor at around 65 h of batch
grown in three dimensional collagen or fibrin gels, and culture. Since HT 1080 is an anchorage dependent cell
that this morphogenetic effect is mediated by hepatocyte line, therefore, the hydrodynamic effects on the cell line
growth factor (HGF), also known as scatter factor. In were investigated.
fibrin gels, this effect is inhibited by addition of exogenous Podorolskaia et al. (1999) studied correlative intercom-
serine protease inhibitors, which suggests a role for PAs nections between PA activity (fibrin plate method) and
in the matrix remodeling required for thrombogenesis. level of UK antigen (Ag uAP) and tissue PA antigen (Ag
They investigated the effect of fibroblast-conditioned me- tAP) in urine and blood (ELISA) in 60 patients with chro-
dium (CM) and HGF on the production of PAs by MDCK nic glomerulonephritis (CGN) and 38 patients with amy-
cells. They found that u-PA activity and mRNA were loidosis. High degree of correlation r = +0.84 and P
increased 4.9 fold by CM from human Detroit-550 fibro- <0.001 was found between blood Ag uAP and urine Ag
blasts, which lacks thrombogenic activity. The u-PA tAP in amyloidosis only.
inductive property of MRC-5 CM was completely inhibited Iwamoto et al. (1990) studied the effects of thrombin
by preincubation with antibodies to recombinant human interleukin (IL-1), tumor necrosis factor (TNF) and
HGF (rhHGF). Exogenously added rhHGF also increased gamma interferon (gamma-IFN) on the release of PA and
u-PA activity and mRNA 5.9 fold in MDCK cell, with an inhibitor (PAI) using cultivated human glomerular epithet-
optimal effect at approximately 10 mg/ml. MRC-5 CM lial cells (GEC's). Their findings indicate that the GEC's
also increased u-PA receptor mRNA 34.9 fold in MDCK participate in the regulation of extracapillary fibrinolysis in
cells, an effect which was inhibited by 71% by preincubat- the glomerular environment being modulated by thrombin
064 Biotechnol. Mol. Biol. Rev.
and cytokines IL-1 and TNF. Suzuki et al. (1989) reported the enhanced UK produc-
Lee et al. (1993) presented a method for determining tivity of 1956 IU/ml when compared to 294 IU/ml in the
the plasminogen activation rate by UK via a cascade controls by human normal diploid fibroblasts which were
enzymatic reaction system. A procedure of parameter cultured in serum free medium containing phospholipase-
estimation has been proposed for the determination of A2, phospholipase-C, bradikinin, coenzyme-A and phyto-
the activity of UK and the kinetic constants. haemagglutinin as inducers.
Schnyder et al. (1992) developed a spectrophotometric Chen et al. (1996) have formulated serum-free media
method to quantify and discriminate UK and t-PAs. by using orthogonal experiments for the growth of geneti-
Leprince et al. (1989) developed a colorimetric assay cally engineered Chinese hamster ovary (CHO) cell line
for the simultaneous measurement of PAs and PAIs in 11G and reported an increase of 80% in UK production.
serum-free conditioned media from cultured cells. Deutheux et al. (1997) examined the production of UK
Ambrus et al. (1979) reported that Plasminogen-rich by culturing human diploid fibroblasts in a serum-free me-
and plasminogen-poor radiolabeled human fibrin clots dium supplemented with peptones (protease peptone)
+
were inserted into large veins of baboons and stump- and K ions for efficient expression of human gene in
tailed monkeys. The thrombolytic effects of PAs (UK, Escherichia coli. In cultures of 3T3 fibroblasts, the UK
SK), and plasmin preparations with activator activity (SK- activity increased by 20 fold when compared to control
activated human plasmin) and without activator activity cells within 24 h when 100 M of sulphur mustard (SM)
(trypsin-activated porcine plasmin, Lysofibrin) were stu- was added. Also ryanodine (10 M) amplified the UK
died. Plasminogen-free and plasminogen-rich clots lysed upregulation by two fold and dexamethasone (1 M) add-
at equal rates. Preparations with and without activator ed directly after SM treatment almost completely pre-
activity were equally effective as thrombolytic agents. vented the induction of UK at both the protein and mRNA
Endogenous activation of plasminogen in the clot thus levels.
appears not to be the essential mechanism of thrombo- Jo et al. (1998) developed a serum-free perfusion cul-
lysis. The exogenous pathway of enzyme adsorption to ture for the production of UK. The cell-growth profile
fibrin fibers seems to represent an important thrombolytic showed a continuous increase in cell density, reaching
7
mechanism. Clot lysis was achieved with doses of fibrino- 5.1 × 10 cells/ml and the production of UK remained
lytic enzymes, which produced little or no significant stable throughout the culture (1586 ± 247 IU/ml).
hematologic changes including hypofibrinogenemia and Gomes et al. (2000) reported a 10-fold increase in UK
decreases of other blood coagulation factor levels. activity with simultaneous supplementation of three
Jamet et al. (1978) reported that UK and SK transform amino acids, based on their repeated occurrence in the
plasminogen into plasmin by rupture of a Arg-Val bond normal pro-UK produced by human kidney cell line HT
and the liberation of a peptide with a molecular weight of 1080 culture.
6 to 8-KDa. UK is a physiological activator with a direct Recently Chen et al. (2004) developed a serum-free
action. By contrast, SK is an enzyme of bacterial origin medium for the production of UK by adding insulin, a
and two hypotheses may be advanced to explain its trace element mixture, a lipid mixture, ascorbic acid and
mechanism of action: the formation of a SK-plasminogen pluronic F68 to dulbecco´s modified eagle´s medium
complex capable of activating new molecules of plasmi- (DMEM)/F12 (1:1, v/v).
nogen or the formation of a SK-plasminogen complex Usually UK production by mammalian cells depends on
within which plasminogen is transformed to plasmin. the following factors: (i) regulation of UK expression (ii)
Kang et al. (1990) reported better pro-UK production at supply of amino acid building blocks for UK synthesis.
5% serum as compared to 10% serum supplemented Moreover some amino acids like glycine have been
medium from a human kidney CAKI-1 cell line when cul- known to bring about stabilization of proteins
tured with cytodex microcarriers in a perfusion bioreactor. (Chainiotakis, 2004), while arginine is known to induce
The medium can be supplemented with BSA, insulin, UK by acting as precursor of nitric oxide, which induces
transferrin and selenium for preservation of viability in low UK production (Ziche et al., 1997).
serum for prolonged culture duration. Therefore the opti- For UK induction, preferable compounds are saccha-
mal serum concentration for UK production depends on rides such as glucose, inositol, ribose and deoxyribose,
cell type and media additives. hormones such as adrenaline (Bansal and Roychoud-
Khaparde and Roychoudhury (2004) reported that a hury, 2006; Bansal et al., 2007).
1% of serum is optimum for UK production as well as for The enhanced production of PA activity was shown to
viability of human kidney HT 1080 cells. Similarly, Tao et be a characteristic of many malignant cell types. The
al. (1987) have used 1% of serum as optimum for UK intracellular and extracellular levels of PA were demon-
production. strated to be substantially elevated in malignant cells
Lacroix and Fritz (1982) reported that the rate of pro- (Christman et al., 1975), cells treated with a tumor pro-
duction and release of UK were greatly influenced by a moter (Wigler and Weinstein, 1976), activated macro-
variety of factors including cell density, presence of hor- phages (Unkeless et al., 1974), established cell lines
mones, incubation temperature and duration of culture. (Mott et al., 1974; Rifkun and Pollack, 1977), granulosa
Kunamneni et al. 065
cells during ovulation (Strickland and Beers, 1976), em- as the core-glycosylated forms, which were mainly accumu-
bryonic cells during differentiation (Topp et al., 1976) and lated within yeast endoplasmic reticulum (6667 PU/ml of cul-
hormone treated cells (Katz et al., 1977). ture medium). But these accumulated pro-UKs were inactive
in their native state and needed to be converted to a
biologically active form by a denaturation-refolding proce-
Recombinant studies dure.
Human UK can be used to treat acute thromboembolic CHO cells are considered ideal hosts for recombinant
events such as venous and arterial thrombosis, pulmo- UK production (Warner, 1999). These cells offer the
nary embolism, intracardiac thrombosis, and systemic advantages that they can be easily genetically manipu-
embolism. However, the high cost of isolation of UK from lated, can be adapted for large-scale suspension culture
either tissue culture cells or urine limits the use of this and it can give rise to proteins with glycans which are si-
enzyme as a therapeutic agent. If UK could be obtained milar, although not identical and to those found on human
from microorganisms by recombinant DNA technology, glycoproteins.
one might have a more economical method of production. Interestingly, Kim and Swartz (2004) have illustrated
In order to improve UK production, UK has been that UK could be efficiently synthesized in E. coli based
expressed in bacteria (Tang et al., 1997; Fahey and cell-free systems using glutathione redox buffer coupled
Chaudhuri, 2000; Sun et al. 2003; Zhong et al., 2007; with the disulphide isomerase to facilitate formation of
Beaton et al., 2005; Gurskii et al., 2005; Ratzkin et al., disulphide bonds. Recently, Roychoudhury et al. (2006)
1981; Deutheux et al., 1997), fungi (Hiramatsu et al., reported that the CHO cell line is known for its unstable
1991), yeast (Wang et al., 2000; Hiramatsu et al., 1989, karyotype. Loss of recombinant gene copy number and
1991; Turner et al., 1991), mammalian cells (Nelles et al., appearance of non-producing populations of cells were
1987; Hu et al., 2006; Innis and Scott, 2001), insect cells predominant causes for instability of production.
(Innis and Scott, 2001) and in plants (Oishi and Zhou,
2000). Different recombinant cell lines that have been Immobilization and bioreactor studies
used for production of UK are listed in Table 1. Large-
scale culture of bacteria or fungi is relatively easy but the Wagner et al. (1990) studied the production of pro-UK by
main drawback of using prokaryotic system for UK a human kidney tumor cell line in long term cultures.
production is the absence of post-translational modifica- Cells were grown on microcarriers, which were retained
tion machinery in these organisms. Therefore, the non- inside the reactor by sedimentation, or with a spin filter.
glycosylated UK so obtained does not have the same Two modes of operation were compared: feed harvest
efficacy and pharmacodynamic properties as that of at an average medium exchange rate of 0.3 per day and
native UK. Mammalian cell lines are hence, preferred for continuous perfusion at a higher dilution rate of 1.5 per
production of UK. A considerable number of mammalian day. In the two systems a stable production of pro-UK
cell lines have been reported to date for UK production. could be maintained for more than 400 h. Continuous
Tang et al. (1997) developed a system to produce perfusion yielded a higher cell density than feed harvest
recombinant urokinase-type plasminogen activator (ru- resulting in a 2-fold increase in the reactor productivity.
PA) in E. coli. The u-PA was produced with 6 His-tag at But higher final enzyme activities were obtained with har-
the C-terminus, which was shown to have the same acti- vest recovered medium than in the perfusion medium.
vity, after refolding, as the wild- type protein. The cumulative medium consumption for mass of product
Kohno et al. (1984) reported the establishment of a was the same in the repeated batch and in the conti-
permanent cell line, TCL-598, which produces and se- nuous operation mode.
cretes UK into the medium in smaller quantities. Kang et al. (1990) investigated kinetics of formation of
Hiramatsu et al. (1991) described the production of human UK from pro-UK in CM and worked for a possibility of
pro-UK and its deletion mutants in yeast. They succeeded in lowering the conversion to UK cultivating human cell lines
producing large amounts of human pro-UK and its mutants under perfusion operations. A human kidney cell line,
066 Biotechnol. Mol. Biol. Rev.
CAKI-1, was cultivated in DMEM with FBS, glutamine added (Eaton et al. 1984).
2+
and gentamicin without Ca to prevent clumping. Cells One of main problems in the cultivation of human and
were then inoculated into a 2L bioreactor with micro- animal cells is the fact that most of these except a few
6
carriers when cell density reached 1×10 viable cells/ml. malignant continuous cell lines are anchorage dependent
It was observed that better production of pro-UK was for their growth in vitro. This means that they require for
obtained with 5% serum containing media than 10% or growth in vitro a suitable solid surface to which they can
serum free medium on cytodex II microcarrier under per- attach and spread. Hence standard fermentors cannot be
fusion chemostat operations. Conversion of pro-UK was applied for cultivation of these cells. Production of a va-
reduced in the serum containing media. riety of biomolecules required anchorage-dependent cells
Senatore and Bernath (1986) had immobilized UK to such as primary cells or diploid cell lines. Van Wezel
the inner surfaces of fibrocollagenous tubes (FCT) in an (1967) discussed the various cultivation systems for
attempt to develop a fibrinolytic biomaterial, which may these anchorage-dependent cells with special attention to
be suitable for use as a small diameter vascular prosthe- the microcarrier culture system. In these systems, cells
sis. The enzyme was bound by adsorption followed by are grown on small particles suspended in culture me-
glutaraldehyde cross-linking. An in vitro kinetic study of dium by stirring. Cells attach and spread upon carriers
immobilized UK was conducted by employing the tubular and grow out gradually to a confluent monolayer. DEAE
material as a flow through reactor operated in a batch sephadex was used as a microcarrier.
recycle mode in which the esterolysis of the model sub- Smidsrod and Skjak-Braek (1990) stated that in recent
2+
strate, N-α-acetyl-L-lysine methyl ester (ALME), was years, entrapment of cells within spheres of Ca alginate
monitored as a function of substrate concentration, re- has become the most widely used technique for immobi-
cycle flow rate, and temperature. Results were compared lizing living cells. This versatile method includes applica-
with data from the soluble enzyme reaction, which was tions ranging from immobilization of living or dead cells in
conducted in the presence and absence of 10% swine bioreactors, immobilization of plant protoplasts for micro-
skin gelatin, in order to identify the specific effects of a propagation and immobilization of hybridoma cells for
collagenous microenvironment. Observed rates for the production of monoclonal antibodies, to entrapment of
UK-FCT catalyzed reaction were observed to be depen- animal cells for implantation of artificial organs. This re-
dent on recycle flow rates below 12 ml/min (Re = 107). view evaluates the potential of this method on the basis
Apparent Michaelis-Menton rate parameters were deter- of the current knowledge of structural and functional
mined by nonlinear search technique for two flow rates: relationships in alginate gels.
one above the critical point for external diffusion effects Avgerinors et al. (1990) used a 20 liter stirred tank fer-
(Re = 282) and one with in the mass-transfer-limited mentor, equipped with a 127 mesh ethylene tetraflouro
region (Re = 71). When the later data were corrected for ethylene rotating screen for cell recycle, for the conti-
external diffusion by applying the Graetz correlation for nuous production of recombinant single chain urokinase-
laminar flow in tubes to estimate the mass transfer type plasminogen activator (rscu-PA) from CHO cells.
coefficient, the corrected Km of 6.45 ± 0.38 mM agreed Viable cell densities between 60 and 74 millions per ml
very closely with the diffusion free parameter (that is, were maintained at medium perfusion rates of 3 to 4 fer-
6.13 ± 0.63). Furthermore, this value was observed to be mentor volumes per day. Cells were retained by the 120-
an order of magnitude higher than that of the soluble microns nominal opening filter through the formation of
enzyme but approximately equal to the Km of the soluble "clumped" cell aggregates of 200 to 600 microns in size,
enzyme in a 10% gelatin environment (8.13 ± 1.53 mM). which did not foul the filter. The rscu-PA produced over
It is postulated that the difference in kinetic parameters the course of this continuous culture was purified and
between soluble and collagen immobilized UK is due to characterized both in vitro and in vivo and shown to be
an inherent interaction between collagen and enzyme comparable to natural scu-PA produced from the trans-
rather than to mass transfer effects. Such an interaction formed human kidney cell line, TCH-598.
is supported by the effects of collagen on thermal stability Tamponnet et al. (1992) immobilized primary cultivated
and energy of activation. rabbit articular chondrocytes in calcium alginate beads.
Human fetal cells (HF) from explants of neonatal fore Both free and entrapped cells were allowed to grow
skins were cultured in DMEM containing calf serum and under normal conditions. After long-term immobilization,
antibiotics. Microcarrier cultures of these cells were pre- the cells still exhibited metabolic activities, patterns of
pared and seeded into microcarrier beads and incubated division, synthesis and secretion of extracellular matrix
with medium to grow to confluence (9 - 15 days). Serum macromolecules such as type II collagen and proteo-
free cultures were prepared by rinsing the beads with glycans. After 38 days, immobilized rabbit articular chon-
PBS, and adding medium containing ovalbumin and epi- drocytes predominantly expressed type II but not type I
dermal growth factor. After 48 h, phorbol myristate ace- collagen. Thus, they maintained their cartilage pheno-
tate was added and the cultures were incubated for an type. After bead lysis, harvested cells showed normal
additional 24 h. The culture medium was then collected growth patterns when resuspended in culture medium.
and fresh medium containing 10% fetal calf serum was On the basis of these results, long-duration storage and
Kunamneni et al. 067
large-scale production of extracellular matrix components connected to a pAAm cryogel column carrying Cu(II)-
are being investigated. iminodiacetic acid (Cu(II)-IDA-pAAm cryogel), which had
Kuo and Bjornsson (1993) developed a simple and been optimized for the capture of UK from the
sensitive method for the simultaneous determination of conditioned medium of the cell lines. Thus an automated
free t-PA and u-PA concentration in biological fluids using system was built, which integrated the features of a hol-
a solid-phase immuno assay. Microtiter plates were coat- low fibre reactor with a chromatographic protein sepa-
ed with polyclonal goat antibodies and incubated with PA ration system. The UK was continuously captured by the
standards or unknown samples. The absorbed PA's were Cu (II)-IDA-pAAm cryogel column and periodically reco-
then assayed by incubation with a mixture of plasmino- vered through elution cycles. The UK activity increased
gen, poly-L-lysine, and the chromogenic substrate H-D- from 280 PU/mg in the culture fluid to 1300 PU/mg after
norleucylherahydrotyrisyllysine-P-nitroanilide. Free t-PA recovery from the capture column which gave about 4.5
and free uPA were detectable in human plasma and urine fold purification of the enzyme. The integrated bioreactor
and in conditioned media from different endothelial cell system operated continuously for 32 days during which
cultures. no backpressure was observed because of the porous
Recently Kumar et al. (2006a) developed a novel type structure of the cryogel matrix. The enzyme eluted from
of cell culture device based on supermacroporous poly- the Cu(II)-IDA-pAAm cryogel capture column was further
acrylamide cryogel support with immobilized gelatin for purified on benzamidine-Sepharose affinity column which
continuous production of UK from human fibro sarcoma gave a preparation of different forms of UK with activity of
HT 1080 and human colon cancer HCT 116 cell lines. 13550 PU/mg of protein. This is one of the very few
The anchorage dependent cells attached to the matrix successful UK production strategies using mammalian
within 4 - 6 h of inoculation and grew as a tissue sheet cell lines.
inside the cryogel matrix. Continuous UK secretion into
the circulating medium was monitored as a parameter of
growth and viability of cells inside the bioreactor. A high Clinical applications of UK
9
yield of viable cell count (3.0 × 10 cells/ml) was obtained UK is used clinically as a thrombolytic agent in the treat-
after continuously running the cell culture reactor for 32 ment of severe or massive deep venous thrombosis, pu-
days, during which 152,600 PU of UK was obtained from monary embolism, myocardial infarction, and occluded
500 ml of culture medium. No morphological changes intravenous or dialysis cannulas. Recently, Alteplase has
were observed on the cells eluted from the gelatin-cryo- replaced UK as a thrombolytic drug in infarctation.
gel support andre-cultured in normal cell culture flasks. The clinical use of u-PA is as a thrombolytic agent.
While the cryogel matrix itself is biocompatible with cells, However, it is also a prognostic marker for tumors and its
coupling of gelatin makes it particularly suitable for grow- structure is the basis for the design of new anti-cancer
ing anchorage dependent cell lines. drugs and is used for the targeting of cytotoxic agents to
The potential of a cryogel bioreactor as a tool for pro- u-PA receptor expressing cells (Alfano et al., 2005; Blasi
cess development of mammalian cell culture has many and Carmeliet, 2002; Dano et al., 2005; Sidenius and
advantages. By using cryogel scaffolds as disposables, Blasi, 2003).
one can get rid of the problem of contamination and addi-
tional expenditures on securing sterile safe guards at
considerable extra costs, particularly for scale up pro- Conclusions
cesses for continuous runs.
Thrombolytic diseases are today a major cause of morbi-
dity and mortality world wide. Fibrinolytic enzymes have
Integrated production and purification of UK from apparent significance in thrombosis therapy. Therefore
bioreactor great attention has been directed towards a search for
thrombolytic agents of various origins with particular
One of the most successful approaches in improving the reference to agents with more specificity and less toxicity.
economy of bioprocesses is to reduce the number of UK is one such enzyme without immunogenicity and
steps involved, as each successive step amounts to cross reactivity when given to patients. Because of its
considerable loss of the product. clinical importance, UK has eluded the interest of
Khaparde and Roychoudhury (2005) successfully researchers as obvious from the lack of reports relating to
developed a two-step integrated process for producing any kind of technological advancements for its production
UK from HT 1080 cell line in a hollow fiber reactor to from in vitro cell culture. Developments in cell lines and
which a sterile 50 ml benzamidine-Sepharose affinity bioprocess technology have made it possible to produce
column was coupled for on-line separation of UK. Appro- UK from in vitro cell culture. The main source of UK is
4
ximately 4.5×10 PU of UK was harvested per day from from cell cultures of lung, heart and kidney tissues. These
the integrated set-up continuously for more than 20 days. tissues are normally derived from mammalian source.
Similarly, Kumar et al. (2006b) developed an integrated Cloning of the UK genes followed by heterologous ex-
set-up of gelatin-pAAm cryogel bioreactor was further pression provide higher enzyme yields. Therefore, im-
068 Biotechnol. Mol. Biol. Rev.
proving the productivity and reducing the production cost Chen Z, Wu B, Jia X, Xiao C (1996). Study on serum-free media for
genetically engineered CHO cells producing prourokinase. Chin. J.
are the major goals for the current studies on the UK
Biotechnol. 12: 169–175.
production. Chen Z, Wu B, Liu H, Liu X, Huang P (2004). Temperature shift as a
New bioreactor designs based on supermacroporous process optimization step for the production of pro-urokinase by a
cryogel matrices promise major advantages. In the inte- recombinant Chinese hamster ovary cell line in high-density perfusion
culture. J. Biosci. Bioeng. 97: 239–243.
grated set-up the coupling of purification column with the
Christman JK, Sulgi S, Newcomb EW, Silverstein SC (1975). Correlated
production bioreactor provides efficient strategy for the suppression by 5-bromodeoxyuridine of tumorigenicity and plasminogen
production and purification of UK by reducing the process activator in mouse melanoma cells. Proc. Natl .Acad. Sci. USA 72: 47-50.
steps and leading to significant improvement in produc- Collen D, Stump JM, Gold HK (1988). Thrombolytic therapy. Ann. Rev.
Med. 39: 405-423.
tion. Dano K, Behrendt N, Hoyer-Hansen G, Johnsen M, Lund LR, Ploug M,
Romer J (2005). Plasminogen activation and cancer. Thromb. Haemost.
93: 676–681.
ACKNOWLEDGMENTS Deutheux M, Jijaksi H, Lisson D (1997). Effect of sulfur mustard on the
expression of urokinase in cultured 3T3 fibroblasts. Arch. Toxicol. 71:
Indian Council of Medical Research (ICMR) is thanked for 243-249.
the senior research fellowship of Dr. R. Bhavani Devi and Eaton DL, Scott RW, Baker JB (1984). Purification of human fibroblast
urokinase proenzyme and analysis of its regulation by proteases and
for the Council of Scientific and Industrial Research
protease nexin. J. Biol. Chem. 259: 6241-6247.
(CSIR), India of Drs. V. Saisha and A. Kunamneni. Fahey EM, Chaudhuri PJB (2000). Binding, refolding and purification of
a urokinase plasminogen activator fragment by chromatography. J.
Chromatogr: B. 737: 225-235.
REFERENCES Fletcher AP, Alkjaersig N, Sherry S (1962). Fibrinolytic mechanisms and
the development of thrombolytic therapy. Am. J. Med. 33: 738-752.
Abloundi FB, Hagan JJ (1957). Comparison of certain properties of Fossum S, Hoem NO (1996). Urokinase and non-urokinase fibrinolytic
human plasminogen and proactivator. Proc. Soc. Exper. Biol. 95: activity in protease-inhibitor-deprived plasma, assayed by a fibrin
195-200. micro-plate method. Immuno. Pharmacol. 32: 119-121.
Alfano D, Franco P, Vocca I, Gambi N, Pisa V, Mancini A, C a p u t i M , Francis CW, Marder VJ (1994). Physiological regulation and pathologic
C a r r i e r o M V , I a c c a r i n o I , S t o p p e l l i M P (2005). The disorders of fibrinolysis, Hemostasis and Thrombosis. In R.W.
urokinase plasminogen activator and its receptor: Role in cell growth Colman, J. Hirsh, V.J. Marder, W. Edwin, B. Salzman (eds) J.B.
and apoptosis. Thromb. Haemost. 93: 190–191. Lippincott Company, Philadelphia, p. 1077.
Alkjaersig N, Fletcher AP, Sherry S (1958). The activation of human Goldberg AR (1974). Increased protease levels in transformed cells: a
plasminogen. II. A kinetic study of activation with trypsin, urokinase, casein overlay assay for the detection of plasminogen activator
and streptokinase. J. Biol. Chem. 233: 86-90. production. Cell 2: 95-102.
Ambrus JL, Weber FJ, Ambrus CM (1979). Mechanism of action of Gomes G, Gayatri, Roychoudhury PK (2000). Urokinase production
fibrinolytic enzymes in vivo. J. Med. 10: 99-119. using human kidney cell line—effect of amino acid supplementation.
Avgerinors GC, Drapeau D, Socolow JS, Mao J, Hsiao K, Broeze RJ (1990). Chemcon, p. 101–104 BIO.
Spin filter perfusion system for high density cell culture: production of Gurskii YG, Belogurov AA, Bibilashvili RS, Grigoreva NV, Delver EP,
recombinant urinary type plasminogen activator in CHO cells. Biotechnol. Minashkin MM (2005). Cloning and recombinant expression of a
8: 54-58. human urokinase -type plasminogen activator variant with reduced
Bansal V, Roychoudhury PK, Ashok KA (2007). Urokinase separation PAI-1 sensitivity for the use as a thrombolytic. Russian Patent
from cell culture broth of a human kidney cell line. Int. J. Biol. Sci. 3: RU2247777 C2.
64-70. Haverkate F, Bradman P (1975). In, Progress in chemical fibrinolysis
Bansal V, Roychoudhury PK (2006). Production and purification of and thrombolysis, Vol. 1, P. 151, ed. Raven Press, NY.
urokinase. Protein Exp. Purif. 45: 1-14. Hiramatsu R, Horinouchi S, Beppu T (1991). Isolation and
Barlow GH, Marder VJ (1980). Plasma urokinase levels measured by characterization of human pro-urokinase and its mutants
chromogenic assay after infusions of tissue culture or urinary source accumulated within the yeast secretory pathway. Gene 99: 235-241.
material. Throm. Res. 18: 431-437. Hiramatsu T, Kasai S, Hirose M, Tsukada M, Tanabe T, Kamimura Y,
Beaton M, Mark S, Pineda-Lucena A (2005). Purification and refolding Kawabe H, Arimura H (1989). Purification of human urokinase
of human recombinant urokinase -type plasminogen activator for analog from cultures of recombinant Saccharomyces cerevisiae.
structure-based inhibitor design by protein NMR using a redox pair- Japanese Patent JP 01252283 A.
containing refolding buffer. PCT International Application Hu X, Gao L, Chen H, Pan S, Yin L, Yang B, Zhang Z (2006). Human
WO2005087917 A2. low-molecular weight urokinase mutant and its expression vector.
Bell PH, Dziobkowski CT, Englert ME (1974). A sensitive fluorometric Chinese Patent CN1880449 A.
assay for plasminogen, plasmin and streptokinase. Anal. Biochem. Innis M, Scott E (2001). Transfection system and expression vectors
61: 200-208. used for recombinant production of heterologous proteins (such as
Bennart NV, Francis JC (1976). Mechanism of the urokinase catalyzed CAB-2, CAB-4, uPAR, VEGF-D or viral glycoprotein) in mammalian
activation of human plasminogen. J. Biol. Chem. 251: 3906-3912. and/or insect cells. PCT International Application WO2001032901
Bernick MB, Kwaan HC (1967). Origin of fibrinolytic activity in cultures A1.
of human kidney. J. Lab. Clin. Med. 70: 650-661. Iwamoto T, Nakshima Y, Suseishi K (1990). Secretion of plasminogen
Bick RL (1992). In: Disorders of thrombosis and hemostasis: Clinical activator and its inhibitor by glomerular epithelial cells. Kidney Int. 37:
and Laboratory Practice. ASCP Press, USA. 1466-1476.
Blasi F, Carmeliet P (2002). uPAR: A versatile signalling orchestrator. Jamet M, Granthil C, Levy G (1978). Enzymatic fibrinolytic agents. Ann.
Nat. Rev. Mol. Cell Biol. 3: 932–943. Anesthesiol. Fr.19: 693-696.
Castellino FJ (1984). Biochemistry of human plasminogen. Semin. Jespersen J, Astrup T (1983). A study of the fibrin plate assay of
Thromb. Hemost. 10: 18-23. fibrinolytic agents. Optimal conditions, reproducibility and precision.
Castillino FJ (1981). Recent advances in the chemistry of the fibrinolytic Haemost. 13: 301–315.
system. Chem. Rev. 81: 431-446.Chainiotakis NA (2004). Enzyme Jo EC, Yun JW, Jung KH, Chung SI, Kim JH (1998). Performance study
stabilization strategies based on electrolytes and polyelectrolytes for of perfusion cultures for the production of single-chain urokinase-type
biosensor applications. Anal. Bioanal. Chem. 378: 89–95. plasminogen activator (sc-uPA) in a 2.5 L spin-filter bioreactor. Bio-
Kunamneni et al. 069
lack (1976). In vitro differentiation of teratomas and the distribution of Wojta J, Hoover RL, Daniel TO (1989). Vascular origin determines
creatine phosphokinase and plasminogen activator in terato- plasinogen activator expression in human endothelial cells: renal
carcinoma-derived cells. Cancer Res. 36: 4217-4223. endothelial cells produce large amounts of scu-PA?. J. Biol. Chem.
Troll W, Sherry S, Wachman J (1954). The action of plasmin on 264: 2846-2852.
synthetic substrates. J. Biol. Chem. 209: 80. Wu KK, Thaigarajan P (1996). Role of endothelium in thrombosis and
Turner BG, Avgerinos GC, Melnick LM, Moir DT (1991). Optimization of hemostasis. Annu. Rev. Med. 47: 315-331.
pro-urokinase secretion from recombinant Saccharomyces Zhong J, Liang X, Huan L (2007). Method for expressing nattokinase
cerevisiae. Biotechnol. Bioengg. 37: 869-875. with specific expression vector and engineering bacteria. Chinese
Unkeless J, Gardon S, Reich E (1974). Secretion of plasminogen Patent CN 1900289 A.
activator by stimulated macrophages. J. Expt. Med. 139: 834-850. Ziche M, Parenti A, Ledda F, Dell'Era P, Granger HJ, Maggi CA, Presta
Valinskey E, Reich E, Le Douarin NM (1981). Plasminogen activator in M (1997). Nitric oxide promotes proliferation and plasminogen
the bursa of fabricius: correlations with morphogenetic remodeling activator production by coronary venular endothelium through
and cell migrations. Cell 25: 471-476. endogenous bFGF. Circ. Res. 80: 845–852.
Van Wezel AL (1967). Growth of cell strains and primary cells on Zimmerman M, Quigley JP, Ashe B, Dron C, Goldfarb R, Troll W (1978).
microcarriers in homogeneous cultures. Nat. 216: 64-65. Direct fluorescent assay of urokinase and plasminogen activators of
Wagner A, Marc A, Engasser JM (1990). Continuous production of normal and malignant cells: Kinetics and inhibitor profiles. Proc. Nalt.
prourokinase in feed harvest and perfusion cultures. Biotechnol. Acad. Sci. 75: 750-753.
Bioengg. 36: 623-629.
Wang P, Sun Z, Chen H, Liu JN (2000). Glycosylation of pro-urokinase
by Pichia pastoris impairs enzymatic activity but not secretion.
Protein Expr. Purif. 20: 179-185.
Warner TG (1999). Enhancing therapeutic glycoprotein production in
Chinese hamster ovary cells by metabolic engineering endogenous
gene control with antisense DNA and gene targeting. Glycobiol. 9:
841-850.
Wigler M, Weinstein IB (1976). Tumor promoter induces plasminogen
activator. Nat. 259: 232–233.
Williams JRB (1951). The fibrinolytic activity of urine. Br. J. Exp. Pathol.
32: 530-537.