| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,1,**,2
* School of Pharmacy and the
Department of Clinical Biochemistry, Hebrew University-Hadassah Medical Centers, Jerusalem, Israel IL-91120;
Departments of Pathology and Laboratory Medicine,
§ Environmental Medicine and
** Medicine, University of Pennsylvania, Philadelphia, Pennsylvania 19104, USA; and the
Department of Biology, Angstrom Pharmaceuticals Inc., San Diego California 92121, USA
2Correspondence: University of Pennsylvania, Dept. Pathology and Laboratory Medicine, 513A Stellar-Chance, 422 Curie Blvd., Philadelphia, PA 19104, USA. E-mail: higazi{at}mail.med.upenn.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: smooth muscle cell • intracellular calcium • vascular contractility • rat aorta
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3
4)
, angiogenesis
(5
6
7
8)
, neointima and aneurysm formation
(9
10
11
12
13)
, chemotaxis (14)
, and wound healing
(15)
. uPA-/- mice are also more
susceptible to lethal pulmonary infection than are wild-type controls
(16)
. Certain of these activities may require the
proteolytic activity of uPA (17
, 18)
, whereas others
involve intracellular signaling through the urokinase receptor (uPAR)
(14
, 19)
or additional, as yet undefined receptors
(15
, 20
21
22)
.
Urokinase is synthesized as a single-chain molecule (scuPA) that
expresses little or no intrinsic enzymatic activity
(23
24
25)
. scuPA is a multidomain protein composed of a
carboxyl-terminal protease domain, termed low molecular weight scuPA
(LMW-scuPA) and an amino-terminal fragment (ATF). ATF is composed of
two subdomains: a growth factor domain (GFD) that binds to uPAR and a
single kringle, the function of which is unknown. scuPA can be cleaved
by plasmin at the
Lys158-Ile159 position to
generate an enzymatically active, disulfide-linked two-chain molecule
(tcuPA).
Between the ATF and the protease domain is a region designated the
‘connecting peptide’ (amino acids 136–158). Within this region is an
eight amino acid sequence (136–143, KPSSPPEE in single-letter code)
that can be generated by the combined activities of tcuPA, which
cleaves uPA between K135 and
K136 (26)
, and matrix
metalloproteinases, such as matrilysin (MMP-7) and stromelysin (MMP-3),
which digest the bond between E143 and
L144 (27
, 28)
. It has been reported
that phosphorylation of Ser-138 or its mutation to glutamic acid
abolishes scuPA-induced adhesion and motility of myelomonocytic cells
without affecting receptor binding (29
, 30)
. However, this
conclusion appears to conflict with other studies showing that cell
adhesion and motility are mediated by the interaction between uPAR and
ATF (31
32
33)
.
One potential explanation for these divergent findings is that
urokinase can assume more than one conformation, which differ in
biological activity. scuPA and tcuPA differ in their intrinsic
enzymatic activity (34)
, susceptibility to plasminogen
activator inhibitors 1 and 2 (35
36
37)
and regulation by
the urokinase receptor (38
, 39)
. Mutations in the
connecting peptide may modulate the behavior of each form of uPA in
ways that are not evident in the adhesion assay but are operative
in vivo.
To address this possibility, we examined the effect of uPA and its
fragments on vascular contractility. Vascular reactivity is mediated in
part by changes in intracellular Ca2+
concentration in smooth muscle cells (SMC). uPA is known to prime
neutrophil production of superoxide by increasing intracellular calcium
(40)
and to mediate signal transduction in monocytic cells
(41
, 42)
and vascular SMC (20)
, but the
effect on calcium flux in this latter cell type has not been described.
The results of the present study indicate that urokinase and its fragments modulate vascular contractility in vitro and in vivo. This activity is mediated by two independent and interdependent epitopes, one residing within the kringle, the other a peptide corresponding to amino acids 136–143. The expression of both epitopes is modulated by conformational changes that occur during the conversion of scuPA to tcuPA.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Uncapped peptides had no effect on vascular function.
|
Contraction response
Male Sprague Dawley rats (250–275 g) were killed by
exsanguination. The thoracic aortas were removed with care to avoid
damage to the endothelium, dissected free of fat and connective tissue,
and cut into transverse rings 5 mm in length (43
44
45
46)
. The
tissues were kept in an oxygenated (95% O2, 5%
CO2) solution of Krebs-Henseliet (KH) buffer (144
mM NaCl, 5.9 mM KCl, 1.6 mM CaCl2, 1.2 mM
MgSO4, 1.2 mM
KH2PO4, 25 mM
NaHCO3, and 11.1 mM D-glucose). The rings were
mounted to record isometric tension in a 10 ml bath containing KH
solution under continuous aeration. The rings were equilibrated for
1.5 h at 37°C and maintained under a resting tension of 2 g
throughout the experiment. Each aortic ring was then contracted by
adding phenylepherine (PE) in stepwise increments (from
10-10 M to 10-5 M). In
other experiments, various concentrations of scuPA, tcuPA, or uPA
fragments were added 15 min before adding PE. In every experiment,
rings exposed to KH buffer alone were analyzed in parallel. Isometric
tension was measured with a force displacement transducer and recorded
online using a computerized system (ExperimentiaÆ, Budapest, Hungary).
Relaxation response
Separate aortic preparations were used to measure the effect of
Å6 on vascular relaxation. After the equilibration period, the aortic
rings were contracted with PE at concentrations
(3x10-7 M and 6x10-8
M), which induced submaximal contraction (50% and 25%, respectively).
After contraction was induced, Å6 was added in stepwise increments
(10-9 M to 10-4 M) to the
same preparation and the isometric tension was measured as above.
Blood pressure recording
Rats were anesthetized by intraperitoneal (i.p.) administration
of ketamine and xylazine (50 and 10 mg/Kg-1,
respectively). One cannula was placed in the left carotid artery to
record the mean arterial blood pressure and a second was placed in the
right jugular vein through which PE (0.1 mg/kg) were administered.
Fifteen minutes later, Å6 (0.1 mg), Å6-variants, or buffer were
administered in the same manner. The blood pressure was monitored
continuously through a transducer placed in the left carotid cannula
using the CARDIOSYS computerized system (ExperimentiaÆ).
Conversion of scuPA to tcuPA
In some experiments, two-chain urokinase (tcuPA; gift of
American Diagnostica, Greenwich, Conn.) was studied. tcuPA was
documented to be free of the isolated amino-terminal fragment of uPA on
native gels. In other experiments, scuPA or scuPA variants (see below)
(20 µM each) were incubated with plasmin (0.1 µM) for 30 min at
37°C to generate tcuPA. The mixture was added to soluble recombinant
human urokinase receptor (suPAR) (see below) bound to CnBr-activated
Sepharose (Sigma) for 1 h at 4°C (39)
. The beads
were washed extensively and the tcuPA was released by adding glycine
buffer, pH 3.0. The activity of tcuPA was examined using the
chromogenic substrate S-2444 (47)
. The completeness of the
conversion of scuPA to tcuPA was verified on sodium dodecyl
sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under reducing
conditions. The preparation was free of plasmin as judged by cleavage
of its chromogenic substrate S-2251 (Chromogenics, Molndal,
Sweden).
Generation of scuPA
136–143 and
kringle-scuPA
A plasmid encoding
scuPA136–143 was
generated using a two-step polymerase chain reaction (PCR). The cDNA
encoding full-length scuPA in pcUK176 served as the template. The
primers 5'-CGCGGATCCAGCAATGAAC-3' and 5'-TGGCCACACTGAAATTTT
AATTTTCCATCTGCGCAGTCAT-3' were used to generate the 438 bp 5'-fragment
pUN121 (48)
, while 5'-TTAAAATTTCAGTGTGGCCA and 5'-CCAAGCT
CGAGGTGCCCG were used to generate the 834 bp 3' fragment. After
completion of the second PCR step, the final product was digested with
BamHI and XhoI and directionally subcloned into
pMT/Bip (Invitrogen, San Diego, Calif.) to yield
scuPA136–143/pMT/Bip.
The recombinant protein was expressed using the Drosophila Expression
System (Invitrogen) in Schneider S2 cells according to the
manufacturer’s recommendations.
scuPA136–143/pMT/Bip
was purified from S2 medium by affinity chromatography using
immobilized rabbit polyclonal anti-human uPAR antibody (ImmunoPure
Protein G IgG Plus Orientation Kit, Pierce; Rockford, Ill.). scuPA
lacking residues 47–135 (k-scuPA) was generated by a two-step PCR
procedure in the same manner. The fragment was digested with
BglII and XhoI, subcloned into pMT/BiP
(Invitrogen), and expressed Schneider 2 cells. k-scuPA was purified
from the cell supernatant using SP-Sephadex and HPLC chromatography.
The protein was characterized by SDS-PAGE under reducing and
nonreducing conditions, followed by Western blot analysis with
anti-urokinase antibody.
Purification of uPA kringle
tcuPA (7.5 mg/ml) was dialyzed against 0.1 M NaPO4
0.6 M NaCl pH 7.8. Plasmin was added to a final concentration of 1 µM
and the mixture was incubated at 37°C for 48 h. Pefablock (1
µM) was added to quench the reaction and the kringle was purified
using reverse phase HPLC (RP-HPLC) on a C8
column. Amino-terminal sequencing confirmed the
NH2 terminus as starting with Ser-47 of the
mature uPA sequence, and the mass of the kringle (determined by
MALDI-TOF mass spectrometry) was consistent with a composition
corresponding to amino acids 47–135 of uPA (10,138 Da). The kringle
was > 95% pure by SDS-PAGE and > 99% pure by analytical
C8 RP-HPLC.
Preparation of recombinant suPAR
cDNA encoding the full length of uPAR in uPAR/pGEM
(49)
was used as a template in a PCR reaction to introduce
a stop codon after amino acid residue 277 and to introduce enzyme
restriction sites. The PCR product was subcloned into pMT/Bip at the
BglII and XhoI sites. The complete sequence was
confirmed using automated fluorescence-based sequence analysis
(University of Pennsylvania DNA Sequencing Facility). suPAR was
expressed using the Drosophila Expression System in Schneider S2 cells
as described above and purified by affinity chromatography on
scuPA-Sepharose (39)
.
Binding of scuPA136–143 and
kringle-scuPA to the urokinase receptor
Binding of wild-type,
scuPA136–143, and
kringle-scuPA was measured using a BIA 3000
optical Biosensor (Biacore, AB, Sweden) (50)
. This method
detects binding interactions in real time by measuring changes in the
refractive index (RI) at a biospecific surface, enabling association
and dissociation rate constants to be calculated. For these studies,
suPAR was coupled to a CM5 research grade sensor chip flow cell
(Biacore) via standard amine coupling procedures (51)
using N-hydroxysuccinimide/N-ethyl-N'-[3-(dimethylamino)
propyl] carbodiimine hydrochloride (Pierce) at a level of 500 response
units. The sensor surface was coated with suPAR (10 µg/ml) in 10 µM
NaAc buffer, pH 5.0. After immobilization, unreacted groups were
blocked with 1 M ethanolamine, pH 8.5. A second flow cell, similarly
activated and blocked without immobilization of protein, served as a
control. All binding reactions were performed in phosphate-buffered
saline (PBS), pH 7.4, 0.005% TWEEN-20. Binding of scuPA (WT,
scuPA136–143, and
k-scuPA) was measured at 25°C at a flow rate of 60 µl/min for 2
min, followed by 2 min of dissociation. The bulk shift due to changes
in RI was measured using the blank surface and was subtracted from the
binding signal at each condition to correct for nonspecific signals.
Surfaces were regenerated with 2 x 18 s pulses of 1 M NaCl,
pH 3.5, followed by an injection of binding buffer for 1 min to remove
this high-salt solution. All injections were performed in a random
fashion using the RANDOM command in the automated method. Binding of
urokinase was measured over a range of concentrations (12.5–0.2 nM).
Data, collected at 2.5 Hz, were fit to a 1:1 Langmuir reaction
mechanism using global analysis in the BIA evaluation 3.0 software
(Biacore). Secondary plots of the data [ln(|dR/dt|) vs. t] were
also performed to unmask any contribution of mass transport to the
kinetic data.
Measurement of plasminogen activator activity
The effect of Å6 on the plasminogen activator activity of
scuPA/suPAR complexes was performed as described (52)
.
Briefly, equimolar complexes between scuPA and suPAR were generated at
50-fold final concentration. Purified human fibrinogen was radiolabeled
with 125I and resuspended to a specific activity
of 30,000 cpm/ml in plasma. Clots were formed from 0.4 ml plasma by
adding thrombin (0.4 units/ml final concentration). scuPA/suPAR (10 nM)
was incubated in the presence or absence of 50 µM Å6 for 2 h,
the clots were washed with PBS, and the release of radiolabeled soluble
fibrin degradation products into the lavage solution at 4 h was
measured.
Measurement of intracellular calcium
The effect of Å6 on PE-induced intracellular calcium
[Ca2+]i in human
umbilical vein endothelial cells (HUVEC) was measured (53
, 54)
. HUVEC were incubated in media supplemented with 1 µM
fluo-3 acetomethyl ester (Fluo-3; Molecular Probes, Eugene, Oreg.)
(55)
in the presence or absence of 5 or 50 µM Å6 for 30
min at 37°C. The cells were washed three times with Krebs’ ringer
bicarbonate solution that contained either 2 mM
Ca2+ or no Ca2+ plus 1 mM
EGTA (calcium-free medium) to remove unbound dye. One milliliter of
Ca2+-containing or
Ca2+-free medium was then added to each well.
Baseline pictures were taken at 490 nm excitation and 520 nm emission
using a Hamamatsu ORCA-100 cooled CCD digital camera; an inverted
Nikon-TMD Diaphot epifluorescence microscope. PE (0.1 µM) was then
added and Fluo-3 emission was measured at 5 min. Å6 had no effect on
dye uptake.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
scuPA inhibited PE-induced contraction of the aorta, increasing the
EC50 from 29 nM to 182 nM (Fig. 1
). We then converted scuPA to tcuPA by adding plasmin, which cleaves
Lys158-Ile159, generating a
molecule composed of two chains that are linked by a single disulfide
bond. tcuPA was affinity purified using immobilized soluble uPAR or
benzamidine to remove plasmin, with identical results. In direct
contrast to scuPA, tcuPA augmented PE-induced aortic contraction,
decreasing the EC50 of PE from 29 nM to 4.47 nM
(Fig. 1)
.
|
The amino-terminal fragment of uPA promotes vascular contractility
Single- and two-chain full-length uPA may have opposing effects on
vascular contractility because they differ in which biologically active
epitopes are exposed and/or because intramolecular interactions among
the domains differ. To begin to distinguish between these
possibilities, experiments were performed using isolated fragments of
urokinase.
LMWuPA (amino acids 144–411), which lacks ATF and is catalytically
active in tcuPA but nascent in scuPA, had no effect on PE-induced
contraction of aortic rings at concentration as high as 20 µM (not
shown). Therefore, we turned our attention to the ATF (amino acids
1–135), the portion of the molecule that binds to uPAR. To do so, we
first examined the effect of soluble uPAR on uPA-mediated
vasoconstriction. suPAR completely abolished the effects of scuPA and
tcuPA, as evidenced by a return of the EC50 to 29
nM (not shown). This outcome suggests that uPA-induced aortic
contraction is mediated by ATF or a portion of uPA in close enough
spatial proximity that it is no longer available when the molecule is
bound to its receptor in solution. Isolated ATF was itself
procontractile, as evidenced by a decrease in the
EC50 for PE to 0.71 nM (Fig. 2
). Indeed, ATF was sixfold more potent than tcuPA (compare Figs. 1
and 2
), and its activity was abolished by suPAR as well (Fig. 2)
.
|
The ATF of urokinase is composed of two subdomains: the GFD and the
kringle. The GFD is known to mediate the binding of uPA to its
receptor. The function of the kringle is unknown. To examine which
portion of ATF is responsible for the procontractile effect of tcuPA
and ATF, the effect of each fragment on PE-induced vasoconstriction was
measured (Fig. 3
). The kringle did not induce vasoconstriction directly, but markedly
augmented the effect of phenylepherine. Addition of 1 nM kringle
decreased the EC50 from 29 nM to 0.1 nM, whereas
10 nM GFD did not augment contraction (not shown).
|
The connecting peptide of urokinase inhibits vascular contractility
We then performed experiments to identify the portion of scuPA
that inhibited PE-induced vasoconstriction. We first examined the
effect of Å6, a capped peptide corresponding to amino acids 136–143
(Ac-KPSSPPEE-Am) that lies between LMWuPA and ATF. Å6 increased the
EC50 of PE sevenfold to 192 nM. Scrambling the
order of the amino acids (Ac-PSESPEKP-Am; designated Å16) completely
abolished the inhibitory effect (control, Fig. 4
). Activity was also lost when the carboxyl-terminal amino acid
(E143) (Å7), but not when the amino-terminal
amino acid (K136) (Å8), was deleted (Fig. 4)
. In
accordance with previous findings that mutating Ser-138 to Glu in
full-length scuPA abolished its proadhesive effect (29
, 30)
, a connecting peptide variant E138
(Å29) had no effect on vascular contractility (not shown).
|
We then asked whether Å6 would cause relaxation of precontracted
aortic rings. Å6 caused a dose-dependent relaxation of aortic rings
pretreated with PE (Fig. 5
). The tension in rings contracted to 25% of maximum was reduced 70%
by 10 µM Å6 (Fig. 5A
) and 40% when rings were first
contracted to 50% of maximal tension (Fig. 5B
). The
scrambled connecting peptide variant (Å16), a variant lacking
E143 (Å7), and a variant
S138E (Å29) did not cause vascular relaxation,
whereas the Å6 variant lacking the amino-terminal amino acid
(K136) (Å8) induced relaxation with the same
potency as wild-type Å6 (not shown).
|
A scuPA variant lacking the connecting peptide lacks vasorelaxation
activity
Taken together, these findings indicate that the procontractile
effect of tcuPA is mediated by the kringle, an effect that may be
overridden in scuPA by a signal involving the connecting peptide.
Alternatively, the procontractile epitope may be exposed in tcuPA but
not in scuPA. To distinguish between these possibilities, we
synthesized a variant scuPA that lacks the connecting peptide
(scuPA136–143) and
studied its vasoactive properties before and after its conversion to a
two-chain molecule.
scuPA136–143 had
no effect on PE-induced aortic contraction (Fig. 6A
). This result demonstrates that the connecting peptide is
required to inhibit contraction and that the procontractile elements in
ATF are not active in scuPA. We then asked whether the lack of activity
of scuPA136–143 was due
a loss of affinity for uPAR. The binding kinetics of wild-type scuPA
and scuPA136–143 to
suPAR immobilized on a CM5 sensor chip were measured using a BIA 3000
optical biosensor (Fig. 7
). Binding was kinetically driven as examination of secondary plots
failed to reveal any limitation resulting from mass transport.
scuPA136–143 bound to
suPAR with only a small reduction in Kd compared
with WT scuPA (Table 2
), a difference that cannot account for its complete loss of vascular
reactivity. Furthermore,
scuPA136–143 expressed
the same plasminogen activator activity as wild-type scuPA in the
presence of suPAR (38
, 39)
and after cleavage by plasmin
to a two-chain molecule (data not shown). In addition, the
EC50 for PE-induced contraction in the
presence of
tcuPA136–143 and
wild-type tcuPA were almost identical (Fig. 1
and Fig. 6A
).
Taken together, this series of experiments supports the hypothesis that
the epitope(s) in the kringle that modulate vasoconstriction are
exposed only when scuPA is converted to tcuPA.
|
|
|
A tcuPA variant lacking the kringle lacks procontractile activity.
We used a similar approach to ask whether the
vasoconstrictive effect of tcuPA is due to the overriding influence of
the kringle above that of the connecting peptide or whether there may
be a difference in the exposure of two independently functioning
domains. Therefore, we examined the effect of a uPA variant lacking the
kringle (k-uPA) on PE-induced vasoconstriction. k-scuPA and WT
scuPA bound to suPAR with almost the same Kd (Table 2
and Fig. 7
). k-scuPA also inhibited PE-induced vasoconstriction to
the same extent as WT scuPA (Fig. 1
and Fig. 6B
). However,
tcuPA lacking the kringle was unable to augment PE-induced
vasoconstriction.
Intracellular pathways through which the connecting peptide
modulates vascular reactivity
Experiments were then performed to begin to identify the pathway
by which the uPA connecting peptide mediates vasorelaxation. First, we
asked whether the connecting peptide opposed the contractile activity
of PE through a direct effect on the 
-adrenergic receptor or on a
postreceptor step. To address this question, we examined the capacity
of Å6 to reverse vasoconstriction induced by endothelin-1, which
operates through an independent receptor system. Å6 caused a
concentration-dependent relaxation of rings contracted by 80 nM
endothelin-1, a concentration that induces 50% of maximal contraction
(Fig. 8
). Endothelin-induced constriction was inhibited 40% by 10 µM Å6
(Fig. 8)
, whereas the scrambled connecting peptide variant Å16 had no
effect (not shown). These data indicate that Å6 does not inhibit
PE-induced vasoconstriction by blocking the -adrenergic receptor.
|
PE exerts its effect on contraction by increasing intracellular
Ca2+, and this effect can be observed in HUVECs
used as a model system. The PE-induced rise in intracellular calcium is
biphasic, with an initial release from intracellular stores followed by
a sustained increase due to influx from the extracellular space
(56)
. To identify the locus of action of the connecting
peptide, PE-mediated Ca2+ influx in calcium-free
or calcium-containing media was measured in the presence and absence of
Å6. PE induced a 178 ± 14% (n=3; P<0.05
vs. baseline) increase in Calcium Green fluorescence in
Ca2+-containing medium, indicative of an increase
in intracellular Ca2+, and a 127 ± 11%
increase (n=3; P<0.05 vs. baseline) in
Ca2+-free medium (Fig. 9A vs. B
). The PE-induced increase in intracellular
Ca2+ was reduced 68 ± 11% by 50 µM Å6
(n=3; P<0.05 vs. PE alone) in
Ca2+-containing medium and 85 ± 9%
(n=3; P<0.05 vs. PE alone) in
Ca2+-free medium. The increase in intracellular
calcium was reduced 21 ± 6% (n=3; P<0.05
vs. PE alone) by 5 µM Å6 in Ca2+-containing
medium and by 45 ± 8% (n=3; P<0.05 vs. PE
alone) in Ca2+-free medium.
|
The capacity of Å6 to inhibit the contraction of aortic rings in
Ca2+-free media was then studied (Fig. 10
). In the absence of extracellular Ca2+, the
EC50 of PE was increased from 25 nM to 1.5 µM,
as expected. Addition of Å6 to the media increased the
EC50 to > 100 µM, an extent of inhibition
even greater than that seen in the presence of extracellular
Ca2+ (Fig. 10
vs. Fig. 1
), whereas the scrambled
peptide variant (Å16) had no effect (data not shown). This result
indicates that Å6 inhibits SMC contraction by inhibiting the release
of intracellular Ca2+, likely from the
endoplasmic reticulum.
|
Nitrous oxide (NO) is known to mediate vasorelaxation by decreasing
intracellular free Ca2+. To examine whether Å6
works through NO synthesis, the NO synthase inhibitor
L-nitro-arg-methyl ester (L-NAME, 1 µM final concentration) was added
to the rings (57)
. L-NAME did not affect relaxation
induced by 10 µM Å6 (not shown). This result indicates that Å6 does
not require synthesis of NO to inhibit Ca2+
release from intracellular stores. Prostacyclin also mediates
vasorelaxation by decreasing intracellular free
Ca2+ (58)
. To examine the
involvement of the prostaglandin pathway, aortic rings were first
contracted with PE. Indomethacin (1 µM final concentration) was added
for 10 min and Å6 was then added in progressively higher
concentrations. Indomethacin alone had no effect on PE-induced
vasoconstriction. However, indomethacin inhibited vasorelaxation
induced by Å6 over the entire range of peptide concentrations tested
(Fig. 11
). This experiment indicates that Å6 stimulates the production of
prostaglandin derivatives with vasorelaxation activity.
|
Effect of the connecting peptide in vivo.
Last, we examined the effect of the connecting peptide on the
blood pressure of rats in vivo. Å6 itself had no effect on
ambient blood pressure at concentrations as high as 1 mg, estimated to
yield a concentration 10-fold greater than is required to inhibit
PE-induced contraction (not
shown). However, Å6 markedly attenuated the hypertensive effect of PE.
Injection of 0.1 mg of Å6 caused the blood pressure to fall below
basal levels in rats injected with PE (Fig. 12A
). In
contrast, no effect on blood pressure was observed when the same amount
of a connecting peptide variant that lacks E143
(Å7) was injected (Fig. 12B
).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 59)
. Urokinase has been proposed to induce a conformational
change in uPAR that promotes its interaction with vitronectin
(52
, 60
, 61)
, complements receptor 3 (62)
,
and modulates ß2-integrin function through
intracellular signaling (63
, 64)
. However, the
determinants in urokinase and within the uPA receptor that promote
signal transduction remain incompletely defined. Moreover, auxiliary
pathways may exist by which uPA transduces signals independent of uPAR,
as suggested by the finding that the concentration of urokinase
required to induce mitogenesis in a melanoma cell line was considerably
above its Kd for this receptor
(21)
.
Phosphorylation of Ser-138 or its mutation to Glu abolishes capacity of
uPA to promote uPAR-mediated cell adhesion and motility and impairs the
motility of uPAR in the cell membrane (13
, 14)
. The
mechanism by which Ser-138 modulates the function of uPAR with no
apparent change in affinity has not been explained. It has been
postulated that phosphorylation of Ser-138 impairs a productive
interaction with uPAR that is not evident in binding assays or induces
novel associations between the ligand and other portions of the
receptor that negatively regulate cell adhesion (13
, 14)
.
These findings imply that all the information required to induce signal
transduction is contained within the connecting peptide. It is
difficult to reconcile this explanation with the findings of several
groups that the amino-terminal fragment of urokinase; indeed, its
growth factor subcomponent suffices to induce uPAR-dependent cell
adhesion (31
32
33
, 65)
.
Our studies reconcile some of these discrepant observations in urokinase-mediated signal transduction by providing novel insights into the nature of intramolecular interactions within the urokinase molecule itself. The finding that single-chain urokinase and two-chain urokinase exert opposing effects on smooth muscle contractility provides not only the first described effect of this plasminogen activator on contractility, but also offers compelling evidence for the existence of more than one functionally active site within the molecule.
One epitope, located within the connecting peptide (amino acids
136–143), inhibits vasoconstriction induced by phenylepherine and
endothelin by blocking the release of calcium from intracellular stores
as well as its entry from the extracellular space. The effect of the
connecting peptide on vasorelaxation is sequence specific and requires
both Ser-138 and Glu-143. Further support for the concept that the
connecting peptide functions as a biologically active unit derives from
its capacity to inhibit cell migration and tumor cell invasiveness (see
ref 66
).
A second epitope, within the kringle of uPA, augments PE-induced vasoconstriction. This procontractile epitope and the epitope within the connecting peptide that promotes vasorelaxation are not exposed simultaneously in either single-chain or two-chain urokinase. Only the connecting peptide is expressed in scuPA. This conclusion is established by several findings. First, a variant of scuPA lacking only these eight amino acids causes neither vasorelaxation nor vasoconstriction. On the other hand, the epitope within the kringle is exposed only in tcuPA. This conclusion is based on two observations. 1) Deleting the kringle did not inhibit the effect of scuPA on vasorelaxation, but the deletion abolished its procontractile activity as a two-chain molecule. 2) Plasmin-mediated activation of scuPA exposes a procontractile epitope within the amino-terminal fragment coincident with the loss of the inhibitory epitope within the connecting peptide. This conclusion derives from the observation that cleavage of the scuPA variant lacking the connecting peptide by plasmin generates a two-chain variant molecule possessing the same procontractile potency as wild-type tcuPA, but does not inhibit PE-induced vasoconstriction as a single-chain molecule.
Our observations may explain why modification of Ser-138 in scuPA by
phosphorylation or mutation to Glu abolishes its effect on cell
adhesion and motility. The signal-transducing epitope in kringle may
not be expressed in scuPA, but is exposed when it is converted to tcuPA
or when ATF is released from the remainder of the molecule. In
accordance with this hypothesis, both the ATF (67)
and the
connecting peptide inhibit tumor cell invasiveness (see ref
66
). In contrast, the finding that ATF and the connecting
peptide exert opposing effects on vascular reactivity demonstrates that
the two domains are capable of exerting independent effects on signal
transduction. This dichotomy is consistent with the finding that cell
motility is stimulated by ATF (19)
, but is inhibited by
the connecting peptide (66)
. Thus, our results suggest
that the domain in ATF that promotes cell adhesion or other
signal-transducing events is not available in scuPA, whereas the
connecting peptide is functional. In contrast, the signal-transducing
effect of ATF becomes evident when scuPA is converted to tcuPA or the
fragment is isolated from the remainder of the molecule.
The mechanism by which uPAR initiates signal
transduction—specifically, whether it requires uPAR—was not resolved
by this study. Several findings suggest that the vasorelaxation and the
procontractile effects of scuPA and tcuPA, respectively, require
binding to uPAR. First, the isolated connecting peptide inhibits the
binding of scuPA to its receptor, and this inhibition is not
competitive (66)
. Second, the connecting peptide inhibits
the fibrinolytic activity of scuPA/suPAR, which requires complex
formation, but not that of tcuPA, which is unaffected by equimolar
concentrations of suPAR (38
, 39)
. Third, the effects of
tcuPA and scuPA on vascular reactivity were abrogated by suPAR. On the
other hand, several compelling observations argue against the direct
involvement of uPAR. tcuPA and scuPA bind with the same affinity to
uPAR but differ in their effects on vasoconstriction. The same
reasoning applies to the differences in the behavior of
scuPA136–143 and
k-scuPA, both of which bind to uPAR with essentially the same
Kd. Moreover, the amino-terminal fragment of human
urokinase does not bind to murine uPAR (68)
. Whether the
isolated connecting peptide, which differs in sequence from its murine
analog, shows cross-reactivity is unknown. The observation that the
scuPA variants lacking the connecting peptide or the kringle bound
normally to human uPAR but lack vasoreactivity, may indicate that
vascular cells express additional, uncharacterized binding sites for
scuPA that mediate signal transduction (21)
. Similar
studies in uPAR-/- mice will be required to
determine whether the signal generated by the isolated connecting
peptide is delivered through uPAR or through a novel receptor.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication August 23, 1999.
Revision received December 3, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
vß5-directed cell migration. J. Biol. Chem. 271,29393-29399This article has been cited by other articles:
|
|
 |
|
  J. W. Kiessling, D. B. Cines, A. A.-R. Higazi, and W. M. Armstead Inhibition of integrin {alpha}V{beta}3 prevents urokinase plasminogen activator-mediated impairment of cerebrovasodilation after cerebral hypoxia/ischemia Am J Physiol Heart Circ Physiol, March 1, 2009; 296(3): H862 - H867. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Stepanova, T. Lebedeva, A. Kuo, S. Yarovoi, S. Tkachuk, S. Zaitsev, K. Bdeir, I. Dumler, M. S. Marks, Y. Parfyonova, et al. Nuclear translocation of urokinase-type plasminogen activator Blood, July 1, 2008; 112(1): 100 - 110. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X.-Q. Wang, K. Bdeir, S. Yarovoi, D. B. Cines, W. Fang, and E. Abraham Involvement of the Urokinase Kringle Domain in Lipopolysaccharide-Induced Acute Lung Injury J. Immunol., October 15, 2006; 177(8): 5550 - 5557. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Akkawi, T. Nassar, M. Tarshis, D. B. Cines, and A. A.-R. Higazi LRP and {alpha}vbeta3 mediate tPA activation of smooth muscle cells Am J Physiol Heart Circ Physiol, September 1, 2006; 291(3): H1351 - H1359. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-H. Kwak, S. Mitra, K. Bdeir, D. Strassheim, J. S. Park, J. Y. Kim, S. Idell, D. Cines, and E. Abraham The kringle domain of urokinase-type plasminogen activator potentiates LPS-induced neutrophil activation through interaction with {alpha}V{beta}3 integrins J. Leukoc. Biol., October 1, 2005; 78(4): 937 - 945. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. M. Armstead, D. B. Cines, and A. A.-R. Higazi Plasminogen Activators Contribute to Impairment of Hypercapnic and Hypotensive Cerebrovasodilation After Cerebral Hypoxia/Ischemia in the Newborn Pig Stroke, October 1, 2005; 36(10): 2265 - 2269. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Pluskota, D. A. Soloviev, K. Bdeir, D. B. Cines, and E. F. Plow Integrin {alpha}M{beta}2 Orchestrates and Accelerates Plasminogen Activation and Fibrinolysis by Neutrophils J. Biol. Chem., April 23, 2004; 279(17): 18063 - 18072. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Nassar, S.'e. Akkawi, A. Shina, A. Haj-Yehia, K. Bdeir, M. Tarshis, S. N. Heyman, and A. A.-R. Higazi In vitro and in vivo effects of tPA and PAI-1 on blood vessel tone Blood, February 1, 2004; 103(3): 897 - 902. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Bdeir, A. Kuo, B. S. Sachais, A. H. Rux, Y. Bdeir, A. Mazar, A. A.-R. Higazi, and D. B. Cines The kringle stabilizes urokinase binding to the urokinase receptor Blood, November 15, 2003; 102(10): 3600 - 3608. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Shetty, K. Bdeir, D. B. Cines, and S. Idell Induction of Plasminogen Activator Inhibitor-1 by Urokinase in Lung Epithelial Cells J. Biol. Chem., May 9, 2003; 278(20): 18124 - 18131. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Nassar, S.'e. Akkawi, R. Bar-Shavit, A. Haj-Yehia, K. Bdeir, A.-B. Al-Mehdi, M. Tarshis, and A. A.-R. Higazi Human alpha -defensin regulates smooth muscle cell contraction: a role for low-density lipoprotein receptor-related protein/alpha 2-macroglobulin receptor Blood, December 1, 2002; 100(12): 4026 - 4032. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Nassar, A. Haj-Yehia, S.'e. Akkawi, A. Kuo, K. Bdeir, A. Mazar, D. B. Cines, and A. A.-R. Higazi Binding of Urokinase to Low Density Lipoprotein-related Receptor (LRP) Regulates Vascular Smooth Muscle Cell Contraction J. Biol. Chem., October 18, 2002; 277(43): 40499 - 40504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Guo, A. P. Mazar, J.-J. Lebrun, and S. A. Rabbani An Antiangiogenic Urokinase-derived Peptide Combined with Tamoxifen Decreases Tumor Growth and Metastasis in a Syngeneic Model of Breast Cancer Cancer Res., August 15, 2002; 62(16): 4678 - 4684. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Falkenberg, C. Tom, M. B. DeYoung, S. Wen, R. Linnemann, and D. A. Dichek Increased expression of urokinase during atherosclerotic lesion development causes arterial constriction and lumen loss, and accelerates lesion growth PNAS, August 6, 2002; 99(16): 10665 - 10670. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-X. Wang, B. Martin-McNulty, A. D. Freay, D. A. Sukovich, M. Halks-Miller, W.-W. Li, R. Vergona, M. E. Sullivan, J. Morser, W. P. Dole, et al. Angiotensin II Increases Urokinase-Type Plasminogen Activator Expression and Induces Aneurysm in the Abdominal Aorta of Apolipoprotein E-Deficient Mice Am. J. Pathol., October 1, 2001; 159(4): 1455 - 1464. [Abstract] [Full Text]
|
|
|
|
 |
|
 S. IDELL, A. P. MAZAR, P. BITTERMAN, S. MOHLA, and A. L. HARABIN Fibrin Turnover in Lung Inflammation and Neoplasia Am. J. Respir. Crit. Care Med., February 1, 2001; 163(2): 578 - 584. [Full Text]
|
|
|
|
 |
|
 Y. GUO, A. A.-R. HIGAZI, A. ARAKELIAN, B. S. SACHAIS, D. CINES, R. H. GOLDFARB, T. R. JONES, H. KWAAN, A. P. MAZAR, and S. A. RABBANI A peptide derived from the nonreceptor binding region of urokinase plasminogen activator (uPA) inhibits tumor progression and angiogenesis and induces tumor cell death in vivo FASEB J, July 1, 2000; 14(10): 1400 - 1410. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
)
or presence of 10 nM scuPA (
) or tcuPA (
). In this and in each
figure, the molar concentration of PE is displayed in logarithmic
units. The mean ± SD of 6 experiments is shown.
