Yin and yang of the plasminogen activator inhibitor.

Abstract

Plasminogen activator inhibitor type one (PAI-1) is involved in many biological processes, and depending on a disease its overexpression, deficiency or normal level can sometimes produce surprising outcomes. This paper reviews low, high, and normal levels of PAI-1 in relation to diseases, their incidents, and possible treatment. Also, a role of PAI-1 in cancer is discussed with attention to "PAI-1 paradox" and explanation of the possible therapeutic applications using this protein.

Full text

Page 1

TRANSLATIONAL MEDICINE  Yin and yang of the plasminogen activator inhibitor1
In general, proteolysis is defined as directed deg‑
radation of proteins by enzymes. It plays an im‑
portant role in physio logical and patho logical
functions of the organism. Proteolysis is tightly
controlled by regulation on the level of expres‑
sion, activation, and inhibition. One of the mem‑
bers of the plasminogen activation system (PAS),
plasminogen activator inhibitor type one (PAI‑1),
is involved in many bio logical processes and de‑
pending on a disease its overexpression, under‑
expression or normal level can sometimes pro‑
duce surprising outcomes. “Yin and yang” refers
to a Chinese expression describing complementa‑
ry opposites within a greater whole and it comes
to mind when describing functions of PAI‑1.
The plasminogen activation system The PAS
contains a number of elements described below.
Plasminogen Plasminogen is a pro‑enzyme that
is cleaved by urokinase plasminogen activator
(uPA) or tissue plasminogen activator (tPA) to its
active form called plasmin. Plasmin is a prote‑
olytic enzyme able to digest proteins of connec‑
tive tissue and basement membranes. It also ac‑
tivates other latent proteolytic enzymes, broad‑
ening the spectrum of proteins attacked. Pro‑
‑collagenase is activated to collagenase in this
manner. Plasmin is a key enzyme in the mech‑
anism responsible for tissue remodeling, tumor
invasion, and development of distant meta stasis
and angiogenesis.1
uPA and tPA activators Both uPA and tPA are
weak proteolytic enzymes that activate plasmi‑
nogen to plasmin by proteolytic cleavage. uPA is
involved in pericellular proteolysis during cell
migration, wound healing, and tissue remodel‑
ing under a variety of physio logical and patho‑
logical conditions. tPA mainly mediates intravas‑
cular thrombolysis.2
Inhibitors of plasminogen activators There are
four known protein inhibitors of uPA/tPA:
PAI‑1, PAI‑2, PAI‑3 and a protein called nexin.
All of them are regulatory proteins mediating
proteolysis on the activation level. Most relevant
seems to be PAI‑1, which exists in three different
forms: active, nonactive‑latent, and cleaved. PAI‑1
has a dual function; first, it plays an important
role as a direct inhibitor of the plasminogen ac‑
tivation system, and second, its inter action with
the adhesive glycoprotein vitronectin plays a role
in tissue remodeling and meta stasis. This func‑
tion is independent of its proteinase inhibitory
properties. Unique to this serpin is the close as‑
sociation between its conformational and func‑
tional properties.3
uPA receptor The binding site of uPA is called
the uPA receptor (uPAR). It is a glycoprotein that
binds uPA to the cell surface while uPA retains
its ability to activate plasminogen. High num‑
bers of uPA receptors on the surface of cancer
TRANSLATIONAL MEDICINE
Yin and yang of the plasminogen activator
inhibitor
Jerzy Jankun1,2, Ewa Skrzypczak‑Jankun1
1  Urology Research Center, Department of Urology, Health Science Campus, The University of Toledo, Toledo, OH, United States
2  Department of Clinical Nutrition, Medical University of Gdańsk, Gdańsk, Poland
Correspondence to:
Prof. Jerzy Jankun, DSc, PhD, 
Urology Research Center, 
Department of Urology, Mail Stop 
1091, Health Science Campus, 
The University of Toledo, 3000 
Arlington, Toledo, OH, USA, 
phone: + 1‑419‑383‑36‑91, 
fax: + 1‑419‑383‑37‑85, email: 
jerzy.jankun@utoledo.edu
Received: March 17, 2009.
Revision accepted: March 19, 2009.
Conflict of inter  est: none declared.
Pol Arch Med Wewn. 2009; 
119 (6): xx‑xx
Copyright by Medycyna Praktyczna, 
Kraków 2009
AbSTRACT
Plasminogen activator inhibitor type one (PAI‑1) is involved in many bio  logical processes, and depend‑
ing on a disease its overexpression, deficiency or normal level can sometimes produce surprising 
outcomes. This paper reviews low, high, and normal levels of PAI‑1 in relation to diseases, their 
incidents, and possible treatment. Also, a role of PAI‑1 in cancer is discussed with attention to PAI‑1 
para  dox and explanation of the possible therapeutic applications using this protein.
KEy wORDS
cancer, high level 
of PAI‑1, PAI‑1 
deficiency, 
plasminogen 
activator inhibitor 
(PAI)

Page 2

POLSKIE ARCHIWUM MEDYCYNY WEWNĘTRZNEJ  2009; 119 (6)
2
including 7 homo zygous individuals of Amish
descent with complete PAI‑1 deficiency and pro‑
longed bleeding.8‑10,16‑18 Frequency estimation
was given by Agren et al. were 4 out of 62 (6.45%)
patients undergoing treatment for benign pros‑
tatic hyperplasia where diagnosed with low level
of PAI‑1 (<1 IU/ml). Bleeding complications dur‑
ing surgery were observed in 3 out of 4 (75%) pa‑
tients with low PAI‑1 levels and in 16 out of 58
(28%) patients with PAI‑1 levels >1 IU/ml.19 Also,
study by Agren et al. reported an incidence of low
PAI‑1 activity in healthy population of 10–13%.20
There is no evidence that these data reflect inci‑
dence of entire population, nevertheless it seems
that this condition is simply under‑diagnosed
rather than extremely uncommon.
Treatment of bleeding in PAI‑1 deficient patients
Bleeding episodes in PAI‑1 deficient patients are
treated with a 5‑ to 7‑day course of oral tranexam‑
ic acid or ε‑aminocaproic acid.8,10,13 These agents
bind to plasminogen and control plasmin gener‑
ation subsequently minimizing bleeding.21 Both
are considered to be safe with some side effects
mainly related to the gastrointestinal tract and
myalgia.22‑24 Moreover, it has been reported that
they have limited efficacy for the control of lo‑
calized bleeding in the surgical setting.25 It was
also reported that menorrhagia in young pa‑
tient did not respond well to tranexamic acid
treatment.8
Other conditions related to PAI‑1 deficiency In ad‑
dition to its classic role as an inhibitor of fibrin‑
olysis, PAI‑1 has been implicated as a mediator
in other processes, including fibrosis, rheuma‑
toid arthritis, atherosclerosis, tumor angiogen‑
esis bacterial infections, and others.26‑28 PAI‑1
deficiency can affect these conditions in differ‑
ent ways. For example, PAI‑1 limits liver injury
and mortality during acetaminophen hepatotox‑
icity by preventing excessive hemo rrhage and fa‑
cilitating tissue repair.29
The best described is PAI‑1 deficiency in mice
with a deletion of the PAI‑1 gene. The impaired
host defense is manifested by enhanced lethality
and increased bacterial growth of Klebsiella pneu-
moniae.28 Overexpression of PAI‑1 in the lung
using adenoviral vector improved host defense
against this infection. Authors suggest that
PAI‑1 protects the host against Klebsiella pneu-
monia by promoting neutrophil recruitment
to the pulmonary compartment and that PAI‑1
is essential for host defense against severe Gram‑
‑negative pneumonia.28 Similar findings were
reported with a different strain of bacteria, i.e.
Streptococcus pneumoniae.27
These infections are a major cause of morbid‑
ity and mortality worldwide. Despite the wide‑
spread use of antibiotics, the mortality rate has
not decreased significantly over the past 30 years.
Patients with PAI‑1 deficiency will be most like‑
ly more susceptible to infection and should be
cells, if occupied by uPA, enhance the proteolytic
activity in the proximity of cancer cells.4
PAI‑1 half‑life PAI‑1 is a fast‑acting, highly spe‑
cific inhibitor of tPA and uPA, but it is not a stable
molecule and converts itself into the latent form
(t1/2 = 1–2 h). This conversion is associated with
partial insertion of the reactive loop (P4‑P10’)
into the central β‑sheet of the PAI‑1 molecule.
In such a conformation, P1‑P1’ and other sites are
not accessible for reaction with tPA or uPA.
PAI‑1 mutants with extended half‑life Several mu‑
tants which reduce or prevent insertion of the re‑
active loop into the PAI‑1 molecule have been
produced in the past (t½ 6–170 h).5,6 We have
developed 7 other mutants with t½ of 4–700 h
by replacing amino acids with cysteines to form
one, two or three disulfide bridges (Cys31‑Cys97,
Cys192‑Cys347, Cys197‑Cys355). Most stable was
VLHL PAI‑1 mutant (Cys197‑Cys355).7
PAI‑1 deficiency PAI‑1 deficiency is defined
by different authors as activity of PAI‑1 in blood
lower than 1 to 3 IU/ml depending on report
(normal: 7–21 IU/ml).8‑11 Fay et al. reported
that the molecular basis of PAI‑1 deficiency was
not determined with confidence; but he found
that some patients were homo zygous for a null
PAI‑1 mutation. The other individuals contained
a mutation in the PAI‑1 gene that resulted in de‑
fect in PAI‑1 function or altered PAI‑1 expression
in a tissue‑specific manner. Also, some individ‑
uals had a mutation within another factor that
controlled PAI‑1 expression or activity, such as
vitronectin.8
Symptoms of PAI‑1 deficiency Patients with PAI‑1
deficiency bleed as a result of a hyperfibrinolyt‑
ic hemo rrhage characterized by normal prima‑
ry hemo stasis but extended bleeding. In these
patients a normal thrombus is formed, but is
quickly lysed as there is no inhibitor to moderate
tPA plasmin activation.12 Spontaneous bleeding
is rarely observed, but moderate hemo rrhaging
of the knees, elbows, nose, and gums can be trig‑
gered by mild trauma. Additionally, prolonged
bleeding after surgery is common and menstrual
bleeding may be severe.10,12‑15 Moderate PAI‑1 de‑
ficiency is associated with a lifelong bleeding ten‑
dency, but severe can be life‑threatening, as de‑
scribed by Fay et al. in the case of a serious post‑
operative hemo rrhage of a 15‑year‑old patient.8
Incidents of PAI‑1 deficiency The true incidence
of PAI‑1 deficiency is unknown in large past be‑
cause of the lack of a standardized laboratory
test and a clear‑cut definition of this deficien‑
cy.10 The first case was published in 1989, when
undetectable PAI‑1 activity and antigen levels
were noted in a 76‑year‑old man with life bleed‑
ing tendency.10,16 At that time PAI‑1 deficiency
was considered an extremely rare clinical enti‑
ty.16 Since then more cases have been described

Page 3

TRANSLATIONAL MEDICINE  Yin and yang of the plasminogen activator inhibitor3
morbidity and mortality leading to the need for
dialysis or kidney transplantation.
Additionally, genetic polymorphisms in PAI‑1
4G/5G and consequent high level of PAI‑1 are
claimed to contribute to an increased risk of ve‑
nous thromboembolism which is associated with
the occurrence of spontaneous abortions. Authors
conclude that the occurrence of PAI‑1 4G/4G or
4G/5G genotypes, respectively, is clinically sig‑
nificant for the pathogenesis of venous throm‑
boembolism in pregnancy, and possibly, sponta‑
neous early abortion.43
An association between high level of PAI‑1 and
obesity has been described, namely as contribut‑
ing to the development of secondary disorders
such as type 2 diabetes mellitus and cardiovascu‑
lar complications. However, a causal role of PAI‑1
in the development of obesity has not been estab‑
lished so far.44 It has also been shown that weight
loss has beneficial effects.45
Incidents Similarly to PAI‑1 deficiency, incidents
of PAI‑1 hyperactivity are largely unknown. Com‑
plicating factors are geographical distribution,
race, and disease. For example it has been re‑
ported that 4G/5G polymorphism in the Leba‑
nese population was found to harbor a relatively
high prevalence of mutations compared to oth‑
er ethnic communities.46
Treatment Most common treatment is anticoag‑
ulation therapy which in some cases can last for
life.47 Others include indirect treatment related
to a disease. For example antidiabetic drugs, such
as metformin, reduce plasma PAI‑1 levels in hu‑
mans with type 2 diabetes.48
However, it was reported that administration
of tiplaxtinin (PAI‑039), an orally active synthet‑
ic inhibitor of PAI‑1, reduced its activity and sig‑
nificantly reduced weight in a diet‑ induced obe‑
sity model in mice. Crandall et al. reported sim‑
ilar findings.44 Although mechanistic aspects
were not addressed in these studies, the poten‑
tial mechanisms of the effect of PAI‑1 inhibition
by tiplaxtinin on nutritionally induced obesity
were most probably multifactorial including re‑
duction of triglycerides but incised of low‑densi‑
ty lipoproteins (LDL).44,49,50
An inter esting approach to treatment of renal fi‑
brosis was proposed by Huang et al. using inactive
PAI‑1R (a new human mutant PAI‑1).51 They ex‑
ploit the finding that PAI‑1 is localized by vitronec‑
tin found at site of injured renal tissue. Further‑
more, PAI‑1 complexed with vitronectin stabilizes
PAI‑1 in its active conformation. However, when
PAI‑1 binds to uPA or tPA activators, it is cleaved
at reactive center loop inducing a rapid confor‑
mational change in PAI‑1 that results in a reduc‑
tion in PAI‑1 affinity to vitronectin. It is followed
by partitioning of the PAI‑1 complex from vit‑
ronectin. As the complex is removed, the vitronec‑
tin becomes available to bind another PAI‑1, which
creates a localized area of high anti‑proteolytic
treated with PAI‑1, but not tranexamic acid or
ε‑aminocaproic acid.
Rationale for using PAI‑1 to limit bleeding Tissue
PA has high affinity to fibrin, fibrin‑bound plas‑
minogen, and an increased activity in presence
of fibrin. These properties enhance tPA fibrinolyt‑
ic potential and localize its activity at site of fi‑
brin deposition.30 Most of the known PA inhib‑
itors are non‑specific in their inhibitory activity
toward tPA, uPA, and other members of the ser‑
ine protease family.31
As opposed to the small molecule inhibitors,
PAI‑1 is more specific for PAs and acts by making
a complex, followed by the formation of a cova‑
lent bond between the active site of the protease
and its reactive center. PAI‑1 is a critical regula‑
tor of the fibrinolytic system through inhibition
of tPA. PAI‑1 binds to the fibrin but not to fibrin‑
ogen and thus can localize itself at sites of inju‑
ry.32 PAI‑1 accumulated within thrombi retains
its complete tPA inhibitory activity protecting
clot from premature dissolution33 making PAI‑1
an attractive antifibrinolytic agent.
Therapeutic application of PAI‑1 Wild‑type PAI‑1
(wPAI‑1) inhibits fibrinolysis in a dose‑responsive
manner by intravenous bolus injection of active
PAI‑1, whereas latent, inactive PAI‑1 has no ef‑
fect. PAI‑1 markedly inhibits fibrinolysis in vivo
and delivering of PAI‑1 to a developing throm‑
bus is an important physio logical mechanism
for subsequent thrombus stabilization.19 PAI‑1
is a fast‑acting, highly specific inhibitor of tPA.
But its very short half‑life presents an obstacle
in using wPAI‑1 as a hemo static drug.34,35 PAI‑1
with extended half‑life can answer this problem
and it was successfully used to shorten total time
of bleeding and total blood loss.36,37
High level of PAI‑1
is present in increased levels and activity in var‑
ious disease states such as cancer, obesity, renal
disease and the meta bolic syndrome. Depending
on a report, high PAI‑1 activity is defined as >10–
20 IU/ml but it can be as high as 68 IU/ml.38,39 It
has been found that an insertion (5G), deletion
(4G) polymorphism at position 675 of the PAI‑1
gene promoter direct transcriptional activity
of PAI‑1.40 Increases of concentration or activity
other than that is less studied and understood.
Hyperactivity of PAI‑1 PAI‑1
Symptoms of hyperactivity of PAI‑1 High levels
of PAI‑1 have been associated with an increased
risk for coronary artery disease and myocardial
infarction due to inhibition of fibrinolysis.41
However, PAI‑1 activity plays an important
role in renal fibrosis as well. In fibrotic renal dis‑
eases, PAI‑1 is increased and localizes to areas
of glomerulosclerosis. PAI‑1 as the major inhibitor
of urokinase in kidneys downregulates degrada‑
tion of fibrin by plasminogen, and lack of activa‑
tion of meta lloproteinases by plasmin potentiates
this process.42 Renal fibrosis causes significant

Page 4

POLSKIE ARCHIWUM MEDYCYNY WEWNĘTRZNEJ  2009; 119 (6)
4
high amounts of uPA and its receptor. Binding
of proteolytically inactive ligand to uPA receptor
reduces amount of uPA on the surface of capillary
endothelial cells, and reduces tumor growth.62
Also, our studies have shown that uPA inhibitors
reduce angiogenesis in chick embryo model, and
reduce the length and number of sprouts of hu‑
man umbilical vascular endothelial cells. In both
cases inhibition of uPA activity on tip of capil‑
lary vessel or sprout prohibits cell migration and
reduces their growth.34,35,63 Goodson et al. has
shown that binding of proteolytically inactive
uPAR ligands prevents cell surface plasminogen
activation, and consequently prevents angiogen‑
esis in the mouse model.64 They emphasize that
the uPAR focuses uPA and initiates proteolytic ac‑
tivity on the vascular capillary cell surface, which
is required for angiogenesis. Ignjatovic et al. ob‑
served inhibition of angiogenesis in the rabbit
cornea while treating animals with amiloride, one
of competitive inhibitors of uPA.65
Urokinase regulates activity of hepatocyte growth
factor/scatter factor Prostate cancer meta‑
stasizes preferentially to the skeleton. Its abili‑
ty to invade and grow in bone marrow stroma is
thought to be due in part to degradative enzymes.
The formation of prostate skeletal meta stases has
been reproduced in vitro by growing co‑cultures
of prostatic epithelial cells in bone marrow stro‑
ma. Expression of urokinase plasminogen was
identified to be responsible for this process.59 It
has been proposed that osseous meta static pros‑
tate cancer cells must be osteomimetic in order
to meta stasize, grow, and survive in the skeleton.
The reciprocal inter action between prostate cancer
and bone stromal growth factors, including he‑
patocyte growth factor/scatter factor (HGF/SF)
among others, initiates bone tropism and is en‑
hanced by uPA.61 HGF/SF bears sequence and
structural homo logy with plasminogen. HGF/SF
exists in both an inactive single‑chain form and
an active two‑chain form. It has been proposed
that plasminogen activators could properly cleave
single‑chain of hepatocyte growth factor to gen‑
erate the active two‑chain. It has been suggested
that uPA is a natural bio logical regulator of HGF.66
Moreover, in a positive feedback manner HGF
stimulation of cancer cells results in overproduc‑
tion of proteases, including uPA, stimulating fur‑
ther activation of HGF.67‑71
Molecular basis of PAs inhibition Most
of the known PA inhibitors are non‑specific
in their inhibitory activity toward tPA, uPA, and
other members of serine protease family.72,73
Based on X‑ray structure analysis and molecular
modeling, it seems that these inhibitors are main‑
ly inserted into the specificity pocket (residues:
87–197; 212–229) of uPA and tPA, and block rec‑
ognition site preventing the binding of plasmino‑
gen activators with their substrate plasmin. This is
the case of inhibition by small molecules such as
benzamidine, p‑benzamidine, and others.74
activity. The noninhibitory PAI‑1R has the same
affinity to vitronectin as native PAI‑1 but cannot
bind to uPA or tPA and binds to vitronectin lon‑
ger than native PAI‑1. This process increases plas‑
min‑driven proteolytic activity at that site. Au‑
thors determine that short‑term administration
of PAI‑1R has slowed the progression of disease
in the mouse model.51
It was found in a randomized clinical trial that
the effect of regular physical exercise on PAI‑1 ac‑
tivity has not changed significantly during 3 years.
However, in the 4G polymorphism group the ex‑
ercise reduced PAI‑1 activity by 36%.52
PAI‑1 in cancer To achieve tissue penetration,
cancer cells stimulate their proteinase machinery,
overproduce and bind proteases, which allows cell
migration through a degraded extracellular ma‑
trix.53 One of these proteases is uPA of plasmin
activation pathway.54,55 It was shown that inhibi‑
tion of uPA activity reduced meta stasis in in vitro
and in vivo models.56,57 Also, during carcinogen‑
esis advancing tips of capillary angiogenic vessels
express a high activity of uPA. Inhibition of uPA
activity results in a reduction in angiogenesis and
cancer size as shown in in vivo models of breast,
colon, prostate, and many other cancers.34,35,57
Thus, inhibitors of urokinase activity could be
used as anticancer agents.
Inhibitors of uPA and meta stasis The urokinase
PAS (uPAS) is commonly overexpressed by many
different human cancers.58 The ability of hu‑
man carcinoma cells (expressing uPA) to invade
the choriallantoic membrane and meta stasize
from it to the embryo, while treated with the an‑
tibody against the active site of uPA, was dramat‑
ically reduced in comparison to non‑treated cells
in the chicken embryo model.2 Cells transfected
with a plasmid, causing an overexpression of uPA
in prostate cancer cells, showed a marked increase
in meta stasis, in comparison with the parental
cell phenotype in the rat model. From the same
phenotype, the cells underexpressing uPA were
selected and these cells displayed drastically de‑
creased meta stasis. In this model prostatic PC3
cancer cells were used and a decreased num‑
ber of meta stasis including skeletal meta stasis
was observed.59 An increased amount or activi‑
ty of uPA, or uPAR per cell, has been found in hu‑
man cancer cell lines with meta static behavior.60
Moreover, animals injected with PC3 prostatic
cancer cells expressing higher amounts of uPA
and/or uPAR develop meta static lesions including
skeletal meta stasis earlier and more frequently
than animals injected with the same cell express‑
ing lower amounts of uPA/uPAR.59 Additionally,
it has been reported that uPA activity is increased
in meta static tumors compared with primary tu‑
mors in experimental animals.61
Urokinase inhibitors reduce angiogenesis The tip
of neovascular advancing capillary vessels sur‑
rounding tumors has been reported to contain

Page 5

TRANSLATIONAL MEDICINE  Yin and yang of the plasminogen activator inhibitor5
the vitronectin pathway. PAI‑1 acts then as a po‑
tent inhibitor of angiogenesis by utilizing pri‑
marily its inhibitory properties toward protei‑
nase activity.34,81
Non serpin activity of PAI‑1 It was reported that
PAI‑1 in vitro treatment of cancer cells induced de‑
tachment of cells from vessel surface.26,80 The pos‑
tulated mechanism is pointing to disruption of vit‑
ronectin and integrins complexes by PAI‑1.80 Our
own study of this phenomenon showed that treat‑
ment of cancer cells with highly stable PAI‑1 down‑
regulated nucleophosmin, while all forms of PAI‑1
(active and nonactive) downregulated fortilin.
These two proteins are implicated in important cel‑
lular processes (cell growth, cell cycle, malignant
transformation).26 This suggests that PAI‑1, in ad‑
dition to its well‑known anticancer properties, plays
an important role in cell signaling. This finding might
lead to the development of more effective therapeu‑
tic strategies in cancer treatment.
word of caution Knowledge on complex inter‑
action of proteins in disease is derived mostly
from animal models. Proteins of different organ‑
isms frequently differ from each other in a sur‑
prising way. For example, mouse uPA and human
uPA are very similar to each other while rat uPA
differs significantly from those of the mouse and
human. Thus, conclusions derived from animal
study must take into consideration limitation
of such model and its applicability to humans as
illustrated below.
Cross species reactivity of urokinase system PAI‑1
binds to uPA and consequently inactivates plas‑
min driven proteolysis. PAI‑1 binds also to the re‑
ceptor bound urokinase‑type plasminogen acti‑
vator and this complex is inter nalized via LDL re‑
ceptor‑related protein. Interaction of vitronec‑
tin with PAI‑1/uPA/uPAR and inter nalization
of that complex regulates cell migration.83 How‑
ever, binding affinity of uPA to uPAR could be
species specific.84 The human uPA fails to bind
to uPAR of murine cells, and murine uPA does
not bind to human cells, but binding is not affect‑
ed in the human‑ or mouse‑bovine systems.85‑87
The other elements of this system inter act with
each other from different species. For example,
the angiogenic activity of purified human uPA
exerts a dose‑dependent angiogenic response
in the chicken chorioallantoic membrane assay
(CAM).85 Our study has shown that human PAI‑1
inhibits angiogenesis in CAM model.34 However,
human PAI‑1 binds to chicken uPA or uPA/uPAR
complexes with lower affinity than to all human
proteins.88 Also, there are structural differences
between human and animal uPA. The most dis‑
tant to human uPA is rat uPA while baboon and,
surprisingly, mouse structure of uPA are very
similar to human uPA.89
Acknowledgments This work was supported
in part by grants from: American Diagnostica
Inhibition of uPA and tPA by PAI‑1 is more spe‑
cific. PAI‑1 is a representative of serpins that are
the members of the superfamily of serine pro‑
tease inhibitors. Inhibitors of serine proteases act
by making a 1:1 stoichiometric complex, followed
by the formation of a covalent bond between
the hydroxyl group of the reactive‑site serine
of the protease and the carboxyl group of the P1
residue at the reactive center of the serpin. Upon
cleavage of an inhibitory serpin, the N‑terminal
end of the reactive‑site loop of PAI‑1 inserts into
β‑sheet of serine protease forming a stable serine‑
‑serpin complex.75,76
PAI‑1 in cancer treatment PAI‑1 with extended
half‑life reduces angiogenesis in vitro and in vivo.
In our study we have observed a reduction in tu‑
mor size while PAI‑1 with extended half‑life was
used but not for control PAI‑1 inactive mutant
(administered by multiply tail vain injection).63,77
Additional experiments with LNCaP xenografts
in severe combined immunodeficient mice using
an osmotic pump to ensure continuous delivery
of PAI‑1 corroborate our previous findings.34 Sim‑
ilar results were shown by others.78,79
PAI‑1 para dox Surprisingly PAI‑1 deficient mice
showed lower proliferation, higher apoptosis and
different morphology of subcutaneously implant‑
ed tumors than its wild counterparts. Further‑
more, PAI‑1 is a predictive factor of poor prog‑
nosis in primary invasive breast cancer.80 These
findings strongly contradict our previous state‑
ment, since it has been demonstrated that PAI‑1
is a potent regulator of angiogenesis and tumor
growth.34,78 However, it has been demonstrated
also that when PAI‑1 is administrated in low con‑
centrations it could increase meta static poten‑
tial and angiogenesis. It has been suggested that
vitronectin and PAI‑1 act together to either pro‑
mote or inhibit angiogenesis. Vitronectin pres‑
ent in the matrix might enhance angiogenesis
by promoting vascular cell migration, and PAI‑1
might regulate this process by controlling access
to the integrin adhesion site on vitronectin.81
Vitronectin is a multifunctional glycoprotein
present in plasma, platelet, and the extracellular
matrix. PAI‑1 is the primary vitronectin binding
protein. The PAI‑1/vitronectin and uPA/uPAR
complexes could also inter act with each other.
When they bind, uPA/uPAR/PAI‑1/vitronectin
complex binds LDL receptor‑related protein. This
weakens PAI‑1/vitronectin inter action and trig‑
gers PAI‑1/uPA/uPAR inter nalization. PAI‑1 and
uPA are degraded but uPAR is recycled to the cell
surface. Subsequently, the receptor can bind uPA
again, and uPA/uPAR, PAI‑1/vitronectin can form
a new complex in the last step of this cyclic pro‑
cess. Since PAI‑1 in the immediate vicinity of cell
is exhausted, the cell migrates toward increased
concentration of PAI‑1.82
However, it has been shown in our previ‑
ous studies and by others, that PAI‑1 added
at supra‑physio logic concentrations suppresses

Page 6

POLSKIE ARCHIWUM MEDYCYNY WEWNĘTRZNEJ  2009; 119 (6)
6
25  Athanasiadis T, Beule AG, Wormald PJ. Effects of topical antifibrinolyt‑
ics in endoscopic sinus surgery: a pilot randomized controlled trial. Am J 
Rhinol. 2007; 21: 737‑742.
26  Jankun J, Aleem AM, Specht Z, et al. PAI‑1 induces cell detachment, 
downregulates nucleophosmin (B23) and fortilin (TCTP) in LnCAP prostate 
cancer cells. Int J Mol Med. 2007; 20: 11‑20.
27  Lim JH, Stirling B, Derry J, et al. Tumor suppressor CYLD regulates 
acute lung injury in lethal Streptococcus pneumoniae infections. Immuni‑
ty. 2007; 27: 349‑360.
28  Renckens R, Roelofs JJ, Bonta PI, et al. Plasminogen activator inhib‑
itor type 1 is protective during severe Gram‑negative pneumonia. Blood. 
2007; 109: 1593‑1601.
29  Bajt ML, Yan HM, Farhood A, et al. Plasminogen Activator Inhibitor‑1 
Limits Liver Injury and Facilitates Regeneration after Acetaminophen Over‑
dose. Toxicol Sci. 2008.
30  Rijken DC, Lijnen HR. New insights into the molecular mechanisms 
of the fibrinolytic system. J Thromb Haemost. 2009; 7: 4‑13.
31  Katz BA, Mackman R, Luong C, et al. Structural basis for selectivity 
of a small molecule, S1‑binding, submicromolar inhibitor of urokinase‑type 
plasminogen activator. Chem Biol. 2000; 7: 299‑312.
32  Smolarczyk K, Boncela J, Szymanski J, et al. Fibrinogen contains cryp‑
tic PAI‑1 binding sites that are exposed on binding to solid surfaces or limit‑
ed proteolysis. Arterioscler Thromb Vasc Biol. 2005; 25: 2679‑2684.
33  Reilly CF, Hutzelmann JE. Plasminogen activator inhibitor‑1 binds to fi‑
brin and inhibits tissue‑type plasminogen activator‑mediated fibrin dissolu‑
tion. J Biol Chem. 1992; 267: 17128‑17135.
34  Swiercz R, Keck RW, Skrzypczak‑Jankun E, et al. Recombinant PAI‑1 
inhibits angiogenesis and reduces size of LNCaP prostate cancer xeno‑
grafts in SCID mice. Oncol Rep. 2001; 8: 463‑470.
35  Swiercz R, Skrzypczak‑Jankun E, Merrell MM, et al. Angiostatic ac‑
tivity of synthetic inhibitors of urokinase type plasminogen activator. On‑
col Rep. 1999; 6: 523‑526.
36  Jankun J, Aleem AM, Selman SH, et al. Highly stable plasminogen 
activator inhibitor type one (VLHL PAI‑1) protects fibrin clots from tissue 
plasminogen activator‑mediated fibrinolysis. Int J Mol Med. 2007; 20: 
683‑687.
37  Jankun J, Aleem AM, Struniawski R, et al. Accelerated thrombus ly‑
sis in blood of plasminogen activator inhibitor deficient mice is inhibited 
by very long half‑life PAI‑1. Pharmacol Rep. [In press].
38  Mlynarska A, Waszyrowski T, Kasprzak JD. Increase in plasma plasmi‑
nogen activators inhibitor type 1 concentration after fibrinolytic treatment 
in patients with acute myocardial infarction is associated with 4G/5G poly‑
morphism of PAI‑1 gene. J Thromb Haemost. 2006; 4: 1361‑1366.
39  Stegnar M, Uhrin P, Peternel P, et al. The 4G/5G sequence polymor‑
phism in the promoter of plasminogen activator inhibitor‑1 (PAI‑1) gene: re‑
lationship to plasma PAI‑1 level in venous thromboembolism. Thromb Hae‑
most. 1998; 79: 975‑979.
40  Dawson SJ, Wiman B, Hamsten A, et al. The two allele sequences 
of a common polymorphism in the promoter of the plasminogen activator 
inhibitor‑1 (PAI‑1) gene respond differently to inter  leukin‑1 in HepG2 cells. 
J Biol Chem. 1993; 268: 10739‑10745.
41  Jood K, Ladenvall P, Tjarnlund‑Wolf A, et al. Fibrinolytic gene polymor‑
phism and ischemic stroke. Stroke. 2005; 36: 2077‑2081.
42  Ma LJ, Fogo AB. PAI‑1 and kidney fibrosis. Front Biosci. 2009; 14: 
2028‑2041.
43  Schenk JF, Stephan B, Zewinger S, et al. Comparison of the plasmino‑
gen activator inhibitor‑1 4G/5G gene polymorphism in females with venous 
thromboembolism during pregnancy or spontaneous abortion. Clin Hemor‑
heol Microcirc. 2008; 39: 329‑332.
44  Schalkwijk  CG,  Stehouwer  CD.  PAI‑1  inhibition  in  obesity  and 
the meta  bolic syndrome: a promising therapeutic strategy. Thromb Hae‑
most. 2006; 96: 698‑699.
45  Sola E, Vaya A, Espana F, et al. Plasminogen activator inhibitor‑1 
levels in severe and morbid obesity. Effect of weight loss and influence 
of 4G/5G polymorphism. Thromb Res. 2008; 122: 320‑327.
46  Shammaa DM, Sabbagh AS, Taher AT, et al. Plasminogen Activator In‑
hibitor‑1 (PAI‑1) gene 4G/5G alleles frequency distribution in the Lebanese 
population. Mol Biol Rep. 2008; 35: 453‑457.
47  Glueck CJ, Goldenberg N, Wang P, et al. Ramifications of four concur‑
rent thrombophilic mutations and one hypofibrinolytic mutation. Clin Appl 
Thromb Hemost. 2004; 10: 365‑371.
48  De Taeye B, Smith LH, Vaughan DE. Plasminogen activator inhibitor‑1: 
a common denominator in obesity, diabetes and cardiovascular disease. 
Curr Opin Pharmacol. 2005; 5: 149‑154.
49  Crandall DL, Quinet EM, El Ayachi S, et al. Modulation of adipose 
tissue development by pharmaco  logical inhibition of PAI‑1. Arterioscler 
Thromb Vasc Biol. 2006; 26: 2209‑2215.
50  Lijnen HR, Alessi MC, Frederix L, et al. Tiplaxtinin impairs nutritionally 
induced obesity in mice. Thromb Haemost. 2006; 96: 731‑737.
51  Huang Y, Border WA, Yu L, et al. A PAI‑1 mutant, PAI‑1R, slows pro‑
gression of diabetic nephropathy. J Am Soc Nephrol. 2008; 19: 329‑338.
Inc., the Frank D. Stranahan Endowment Fund
for Onco logical Research, and Translational Re‑
search Stimulation Award from the University
of Toledo.
REfERENCES:
1  Iishi H, Tatsuta M, Baba M, et al. Suppression by amiloride of bombe‑
sin‑enhanced peritoneal meta  stasis of intestinal adenocarcinomas induced 
by azoxymethane. Int J Cancer. 1995; 63: 716‑719.
2  Ossowski L. In vivo invasion of modified chorioallantoic membrane 
by tumor cells: the role of cell surface‑bound urokinase. J Cell Biol. 1988; 
107: 2437‑2445.
3  Ngo TH, Hoylaerts MF, Knockaert I, et al. Identification of a target site 
in plasminogen activator inhibitor‑1 that allows neutralization of its inhibi‑
tor properties concomitant with an allosteric up‑regulation of its antiadhe‑
sive properties. J Biol Chem. 2001; 276: 26243‑26248.
4  Kwaan HC, Keer HN, Radosevich JA, et al. Components of the plasmi‑
nogen‑plasmin system in human tumor cell lines. Semin Thromb Hemost. 
1991; 17: 175‑182.
5  Berkenpas MB, Lawrence DA, Ginsburg D. Molecular evolution of plas‑
minogen  activator  inhibitor‑1  functional  stability.  EMBO  J.  1995;  14: 
2969‑2977.
6  Tucker HM, Mottonen J, Goldsmith EJ, et al. Engineering of plasmino‑
gen activator inhibitor‑1 to reduce the rate of latency transition. Nat Struct 
Biol. 1995; 2: 442‑445.
7  Chorostowska‑Wynimko  J,  Swiercz  R,  Skrzypczak‑Jankun  E,  et  al. 
A novel form of the plasminogen activator inhibitor created by cysteine 
mutations extends its half‑life: relevance to cancer and angiogenesis. Mol 
Cancer Ther. 2003; 2: 19‑28.
8  Fay WP, Parker AC, Condrey LR, et al. Human plasminogen activator in‑
hibitor‑1 (PAI‑1) deficiency: characterization of a large kindred with a null 
mutation in the PAI‑1 gene. Blood. 1997; 90: 204‑208.
9  Lee MH, Vosburgh E, Anderson K, et al. Deficiency of plasma plasmino‑
gen activator inhibitor 1 results in hyperfibrinolytic bleeding. Blood. 1993; 
81: 2357‑2362.
10  Mehta R, Shapiro AD. Plasminogen activator inhibitor type 1 deficien‑
cy. Haemophilia. 2008; 14: 1255‑1260.
11  Morimoto Y, Yoshioka A, Imai Y, et al. Haemostatic management 
of intraoral bleeding in patients with congenital deficiency of alpha2‑plas‑
min inhibitor or plasminogen activator inhibitor‑1. Haemophilia. 2004; 10: 
669‑674.
12  Largent V, Deneys V, Brichard B, et al. Bleeding diathesis in a child 
with normal screening tests: think about fibrinolysis. Eur J Pediatr. 2005; 
164: 587‑588.
13  Kuhli C, Luchtenberg M, Scharrer I, et al. Massive subhyaloidal hemo‑
rrhage associated with severe PAI‑1 deficiency. Graefes Arch Clin Exp Oph‑
thalmol. 2005; 243: 963‑966.
14  Takahashi Y, Tanaka T, Minowa H, et al. Hereditary partial deficiency 
of plasminogen activator inhibitor‑1 associated with a lifelong bleeding ten‑
dency. Int J Hematol. 1996; 64: 61‑68.
15  Zhang ZY, Wang ZY, Fu JX, et al. [Report of a case of congenital plas‑
minogen activator inhibitor‑1 deficiency]. Zhonghua Xue Ye Xue Za Zhi. 
2004; 25: 129‑131. Chinese.
16  Schleef RR, Higgins DL, Pillemer E, et al. Bleeding diathesis due to de‑
creased functional activity of type 1 plasminogen activator inhibitor. J Clin 
Invest. 1989; 83: 1747‑1752.
17  Agren A, Kolmert T, Wiman B, et al. Low PAI‑1 activity in relation 
to the risk for perioperative bleeding complications in transurethral resec‑
tion of the prostate. Thromb Res. 2007; 119: 715‑721.
18  Minowa H, Takahashi Y, Tanaka T, et al. Four cases of bleeding diathe‑
sis in children due to congenital plasminogen activator inhibitor‑1 deficien‑
cy. Haemostasis. 1999; 29: 286‑291.
19  Agren A, Wiman B, Schulman S. Laboratory evidence of hyperfibrinol‑
ysis in association with low plasminogen activator inhibitor type 1 activity. 
Blood Coagul Fibrinolysis. 2007; 18: 657‑660.
20  Agren A, Wiman B, Stiller V, et al. Evaluation of low PAI‑1 activity 
as a risk factor for hemo  rrhagic diathesis. J Thromb Haemost. 2006; 4: 
201‑208.
21  Santamaria A, Borrell M, Mateo J, et al. What is the clinical impact 
of low plasminogen activator inhibitor‑1 (PAI‑1) activity? A and study 
of the incidence of low PAI‑1 antigen in a healthy population. J Thromb 
Haemost. 2007; 5: 1565‑1566.
22  Ogston D. Current status of antifibrinolytic drugs. Blood Rev. 1989; 
3: 1‑4.
23  Sharifi R, Lee M, Ray P, et al. Safety and efficacy of intravesical amin‑
ocaproic acid for bleeding after transurethral resection of prostate. Urolo‑
gy. 1986; 27: 214‑219.
24  Verstraete M. Clinical application of inhibitors of fibrinolysis. Drugs. 
1985; 29: 236‑261.

Page 7

TRANSLATIONAL MEDICINE  Yin and yang of the plasminogen activator inhibitor7
80  Binder BR, Mihaly J. The plasminogen activator inhibitor „paradox“ 
in cancer. Immunol Lett. 2008; 118: 116‑124.
81  Stefansson S, Petitclerc E, Wong MK, et al. Inhibition of angiogene‑
sis in vivo by plasminogen activator inhibitor‑1. J Biol Chem. 2001; 276: 
8135‑8141.
82  Noel A BK, Masson V, Devy L, Frankenne F, Rakic JM, Lambert V, 
Carmeliet P, Foidart JM. Regulation of Cancer Invasion and Vascularization 
by Plasminogen Activator Inhibitor‑1. Fibrynolysis and Proteolysis. 1999: 
220‑225.
83  Degryse B, Sier CF, Resnati M, et al. PAI‑1 inhibits urokinase‑induced 
chemotaxis by inter  nalizing the urokinase receptor. FEBS Lett. 2001; 505: 
249‑254.
84  Quax PH, Grimbergen JM, Lansink M, et al. Binding of human uroki‑
nase‑type plasminogen activator to its receptor: residues involved in spe‑
cies specificity and binding. Arterioscler Thromb Vasc Biol. 1998; 18: 
693‑701.
85  Ribatti D, Leali D, Vacca A, et al. In vivo angiogenic activity of uroki‑
nase: role of endogenous fibroblast growth factor‑2. J Cell Sci. 1999; 112 
(Pt 23): 4213‑4221.
86  Olson D, Pollanen J, Hoyer‑Hansen G, et al. Internalization of the uroki‑
nase‑plasminogen  activator  inhibitor  type‑1  complex  is  mediated 
by the urokinase receptor. J Biol Chem. 1992; 267: 9129‑9133.
87  Estreicher A, Wohlwend A, Belin D, et al. Characterization of the cel‑
lular binding site for the urokinase‑type plasminogen activator. J Biol Chem. 
1989; 264: 1180‑1189.
88  Sipley JD, Alexander DS, Testa JE, et al. Introduction of an RRHR mo‑
tif into chicken urokinase‑type plasminogen activator (ch‑uPA) confers sen‑
sitivity to plasminogen activator inhibitor (PAI)‑1 and PAI‑2 and allows 
ch‑uPA‑mediated extracellular matrix degradation to be controlled by PAI‑1. 
Proc Natl Acad Sci U S A. 1997; 94: 2933‑2938.
89  Jankun J, Skrzypczak‑Jankun E. Binding site of amiloride to uroki‑
nase plasminogen activator depends on species. Int J Mol Med. 2001; 
8: 365‑371.
52  Vaisanen SB, Humphries SE, Luong LA, et al. Regular exercise, plas‑
minogen activator inhibitor‑1 (PAI‑1) activity and the 4G/5G promoter poly‑
morphism in the PAI‑1 gene. Thromb Haemost. 1999; 82: 1117‑1120.
53  Hornebeck W, Emonard H, Monboisse JC, et al. Matrix‑directed regu‑
lation of pericellular proteolysis and tumor progression. Semin Cancer Biol. 
2002; 12: 231‑241.
54  Leake D, Doerr TD, Scott G. Expression of urokinase‑type plasmino‑
gen activator and its receptor in keloids. Arch Otolaryngol Head Neck Surg. 
2003; 129: 1334‑1338.
55  Weidle UH, Konig B. Urokinase receptor antagonists: novel agents for 
the treatment of cancer. Expert Opin Investig Drugs. 1998; 7: 391‑403.
56  Stephens RW, Brunner N, Janicke F, et al. The urokinase plasminogen 
activator system as a target for prognostic studies in breast cancer. Breast 
Cancer Res Treat. 1998; 52: 99‑111.
57  Duffy MJ. Proteases as prognostic markers in cancer. Clin Cancer Res. 
1996; 2: 613‑618.
58  Conese M, Blasi F. The urokinase/urokinase‑receptor system and can‑
cer invasion. Baillieres Clin Haematol. 1995; 8: 365‑389.
59  Achbarou A, Kaiser S, Tremblay G, et al. Urokinase overproduction re‑
sults in increased skeletal meta  stasis by prostate cancer cells in vivo. Can‑
cer Res. 1994; 54: 2372‑2377.
60  Festuccia C, Vincentini C, di Pasquale AB, et al. Plasminogen activa‑
tor activities in short‑term tissue cultures of benign prostatic hyperplasia 
and prostatic carcinoma. Oncol Res. 1995; 7: 131‑138.
61  Wilson MJ, Sinha AA. Plasminogen activator and meta  lloprotease ac‑
tivities of Du‑145, PC‑3, and 1‑LN‑PC‑3‑1A human prostate tumors grown 
in nude mice: correlation with tumor invasive behavior. Cell Mol Biol Res. 
1993; 39: 751‑760.
62  Pepper MS, Sappino AP, Stocklin R, et al. Upregulation of urokinase 
receptor expression on migrating endothelial cells. J Cell Biol. 1993; 122: 
673‑684.
63  Jankun J, Keck RW, Skrzypczak‑Jankun E, et al. Inhibitors of uroki‑
nase reduce size of prostate cancer xenografts in severe combined immu‑
nodeficient mice. Cancer Res. 1997; 57: 559‑563.
64  Goodson RJ, Doyle MV, Kaufman SE, et al. High‑affinity urokinase re‑
ceptor antagonists identified with bacteriophage peptide display. Proc Natl 
Acad Sci U S A. 1994; 91: 7129‑7133.
65  Ignjatovic Z, Nikolic L. [Inhibition of angiogenesis in the cornea with 
amiloride]. Srp Arh Celok Lek. 1996; 124: 120‑123.
66  Mars WM, Zarnegar R, Michalopoulos GK. Activation of hepatocyte 
growth factor by the plasminogen activators uPA and tPA. Am J Pathol. 
1993; 143: 949‑958.
67  Hart CA, Scott LJ, Bagley S, et al. Role of proteolytic enzymes in hu‑
man prostate bone meta  stasis formation: in vivo and in vitro studies. Br J 
Cancer. 2002; 86: 1136‑1142.
68  Billstrom A, Hartley‑Asp B, Lecander I, et al. The urokinase inhibitor 
p‑aminobenzamidine inhibits growth of a human prostate tumor in SCID 
mice. Int J Cancer. 1995; 61: 542‑547.
69  Kermorgant S, Aparicio T, Dessirier V, et al. Hepatocyte growth fac‑
tor induces colonic cancer cell invasiveness via enhanced motility and pro‑
tease overproduction. Evidence for PI3 kinase and PKC involvement. Car‑
cinogenesis. 2001; 22: 1035‑1042.
70  Koeneman KS, Yeung F, Chung LW. Osteomimetic properties of pros‑
tate cancer cells: a hypothesis supporting the predilection of prostate can‑
cer meta  stasis and growth in the bone environment. Prostate. 1999; 39: 
246‑261.
71  Sparks RL, Pool TB, Smith NK, et al. Effects of amiloride on tumor 
growth and intracellular element content of tumor cells in vivo. Cancer Res. 
1983; 43: 73‑77.
72  Dano K, Behrendt N, Hoyer‑Hansen G, et al. Plasminogen activation 
and cancer. Thromb Haemost. 2005; 93: 676‑681.
73  Jedinak A, Maliar T. Inhibitors of proteases as anticancer drugs. Neo‑
plasma. 2005; 52: 185‑192.
74  Jankun J, Skrzypczak‑Jankun E. Molecular basis of specific inhibition 
of urokinase plasminogen activator by amiloride. Cancer Biochem Biophys. 
1999; 17: 109‑123.
75  Shin JS, Yu MH. Kinetic dissection of alpha 1‑antitrypsin inhibition 
mechanism. J Biol Chem. 2002; 277: 11629‑11635.
76  Ding X, Rasmussen BF, Demuth HU, et al. Nature of the inactivation 
of elastase by N‑peptidyl‑O‑aroyl hydroxylamine as a function of pH. Bio‑
chemistry. 1995; 34: 7749‑7756.
77  Lawrence DA, Olson ST, Palaniappan S, et al. Serpin reactive center 
loop mobility is required for inhibitor function but not for enzyme recogni‑
tion. J Biol Chem. 1994; 269: 27657‑27662.
78  McMahon GA, Petitclerc E, Stefansson S, et al. Plasminogen activator 
inhibitor‑1 regulates tumor growth and angiogenesis. J Biol Chem. 2001; 
276: 33964‑33968.
79  Stefansson S, McMahon GA, Petitclerc E, et al. Plasminogen activa‑
tor inhibitor‑1 in tumor growth, angiogenesis and vascular remodeling. Curr 
Pharm Des. 2003; 9: 1545‑1564.