| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology (C.E.R., E.Y.A.), University of Maryland School of Medicine, Baltimore, Maryland 21201; Stanford University, Department of Pediatrics (P.J.F.), Palo Alto, California 94305; and Department of Pediatrics (R.G.R.), Oregon Health Sciences University, Portland, Oregon 97201
Address all correspondence and requests for reprints to: Dr. Eli Y. Adashi, Division of Reproductive Sciences, Department of Obstetrics and Gynecology, University of Utah Health Sciences Center, 546 Chipeta Way, Mailbox No. 20, Salt Lake City, Utah 84108. E-mail: eadashi{at}hsc.utah.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, TGFß1,
EGF, or endothelin-1 failed to do so. However, the concurrent provision
of GnRH, TNF, EGF, or endothelin-1 proved inhibitory to the IGFBP-5
endopeptidase-inducing property of FSH. Activin-A and
TGFß1 in turn further stimulated the FSH effect.
Sensitivity to EDTA, 1,10 phenanthroline, and high concentrations
(
0.1 mM) of Zn2+ suggested a Zn2+
metalloprotease. Insensitivity to TIMP-1 and TIMP-2 argued against a
matrix metalloprotease (MMP). Relative insensitivity to PMSF, AMPSF,
aprotinin, TPCK, and benzamidine argued against the possibility of a
serine protease. Insensitivity to pepstatin A and E64 argued against
aspartic and cysteine proteases, respectively. Insensitivity to
plasminogen activator inhibitor-1 (PAI-1) and the presumed lack of free
plasminogen in serum-free culture media argued against plasmin.
Proteolysis was completely inhibited over the acid pH range but
proceeded unencumbered at neutral and basic pH. Competition studies
using unlabeled IGFBPs (1–6) as well as cell-free proteolysis assays
of [125I]-labeled IGFBP-1, 2, 3, and 6 suggested a
significant level of specificity for the FSH-induced/IGFBP-5-directed
endopeptidase. Centricon-mediated fractionation of FSH-conditioned
media revealed the IGFBP-5 endopeptidase activity in the fraction
representing proteins of molecular weight >100K. Taken together, these
observations document a secreted, granulosa cell-derived, high
molecular weight, FSH-inducible, IGFBP-5-selective, neutral/basic
pH-favoring, non-MMP Zn2+ metalloprotease. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
10 ng/ml) was shown to inhibit the
elaboration of IGFBP-5 (1, 2, 3), an effect requiring the activation of
the A-kinase transduction pathway (4). This inhibitory FSH action is
all the more noteworthy in light of the generally stimulatory effect
exerted by FSH at the level of the granulosa cell. Conceivably then,
this apparently inhibitory action of FSH may in effect constitute a net
stimulatory gain because granulosa cell-derived IGFBP-5 is inhibitory
to IGF-I (5) and thus inevitably to gonadotropin (3) hormonal
action. In part, the ability of FSH to suppress the accumulation of IGFBP-5 may be transcriptional in nature (3). However, a posttranscriptional degradative phenomenon, i.e. the acquisition of an IGFBP-5 endopeptidase activity, may also be operational. That IGFBP-5 may in fact be degradation prone has been suggested earlier by the observation that continued progressive degradation of granulosa cell-derived IGFBPs was apparent in the face of cycloheximide-induced blockade of protein biosynthesis (6). Additional evidence for IGFBP-5 proteolysis can be derived from the observation of a granulosa cell-derived FSH-inducible/cation-dependent IGFBP-5 endopeptidase (3, 7). Although the precise nature of the IGFBP-5 endopeptidase remains unknown, identity with established ovarian proteases such as renin and plasmin appears unlikely. Renin, an aspartic protease, constitutes an extremely selective enzyme the only known substrate of which is angiotensinogen (8). Although the established FSH-mediated stimulation of plasminogen activator activity (9) raises the possibility of plasmin involvement, the presumed absence of plasminogen in serum-free culture media argues against this possibility.
It is therefore the objective of this communication to establish in some detail the key operational characteristics, the class, the specificity, the approximate pH optimum, the approximate size, and the hormonal dependence of the FSH-associated IGFBP-5 endopeptidase of rat ovarian origin.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Reagents/hormones
Chloramine-T, KI, EDTA, 1,10 phenanthroline (0-phenanthroline),
E64 (trans-epoxysuccinyl-L-leucylamido-(4-guanidine
butane), PMSF (phenylmethylsulfonyl fluoride), aprotinin, pepstatin A,
chymostatin, benzamidine, ZnCl2, PMA (phorbol 12-myristate
13-acetate), succinic acid, MES
(2-[N-morpholino)ethanesulfonic acid), HEPES, CHES
(2-[N-cyclo-hexylamino)ethanesulfonic acid), and BSA were
from Sigma Chemical Co. (St. Louis, MO). TPCK
(N-tosyl-L-phenylalanine chloro-methyl
ketone) and APMSF ([4-Amidinophenyl]methylsulfonyl fluoride) were
from Boehringer Mannheim (Indianapolis, IN). Acetonitrile and TFA were
from J. T. Baker (Phillipsburg, NJ). McCoy’s 5a medium (modified;
without serum), penicillin-streptomycin solution,
L-glutamine, and trypan blue stain (0.4%) were
obtained from GIBCO-BRL Life Sciences (Grand Island, NY). Sodium
[125I] (3.7 Gbq/ml) was obtained from Amersham Life
Sciences (Arlington Heights, IL).
Ovine FSH (NIH-oFSH-S18; FSH potency = 65 NIH FSH-S1 U/mg; LH
activity = 0.04 NIH LH-S1 units/mg; PRL activity < 0.1% by
weight) was generously provided by the National Pituitary Agency,
Pituitary Hormone Distribution Program, NIADDK (Bethesda, MD). PMSG
(pregnant mare serum gonadotropins) was from Sigma. Recombinant hIGF-I
and EGF were from Bachem Laboratories (Torrance, CA). Purified hIGFBP-1
was generously provided by Markku Seppälä, Helsinki
University (Helsinki, Finland). Recombinant hIGFBP-2 was generously
provided by Dr. J.-L. Mary, University of Basel (Basel, Switzerland).
Recombinant (nonglycosylated) hIGFBP-3 was generously provided by Dr.
Christopher A. Maack, Biogrowth, Inc. (Richmond, CA). Recombinant
hIGFBP-4 was generously provided by Dr. Michael C. Kiefer, Chiron Corp.
(Emeryville, CA). Recombinant hIGFBP-5 and hIGFBP-6 were from Austral
Biologicals (San Ramon, CA). Recombinant human FSH was generously
provided by Dr. James Hutchinson, Serono Laboratories (Norwell, MA).
Recombinant hActivin-A, TNF, and TGFß1 were generously
provided by Dr. Jennie P. Mather, Genetech, Inc. (South San Francisco,
CA). GnRH and endothelin-1 were from Peninsula Laboratories (Belmont,
CA). Plasminogen activator inhibitor-I (PAI-I) was the generous gift of
Dr. Adrienne Racanell of DuPont-Merck Pharmaceutical Co. (Wilmington,
DE). TIMP-1 was kindly provided by Dr. John L. Fowlkes of Duke
University (Durham, NC) courtesy of Dr. Hideaki Nagase of the
University of Kansas (Kansas City, KS). TIMP-2 was generously
provided by Dr. Philippe Monget of URA CNRS 1291 (Nouzilly, France)
In vitro studies
Granulosa cells [(5 x 105 viable cells/dish)
unless indicated otherwise], obtained by repeated follicular puncture
as previously described (10), were inoculated into 35 x 10 mm
tissue culture dishes containing 1 ml McCoy’s 5a medium (modified;
without serum) supplemented with L-glutamine (2
mmol/liter), penicillin (100 U/ml), and streptomycin sulfate (100
µg/ml). Cellular viability was assessed by trypan blue exclusion.
Cell cultures were maintained for up to 72 h at 37 C under a
water-saturated atmosphere of 95% air and 5% CO2. All
agents were dissolved in sterile culture medium and deployed in 50-µl
aliquots. At the end of each experiment, the media were collected for
further processing as described below.
Extraovarian cell lines studied included U2 human osteosarcoma cells (generously contributed by Dr. Subburaman Mohan, University of Loma Linda, Loma Linda, CA), B104 rat neuroblastoma cells (generously contributed by Dr. Sharron Gargosky, Oregon Health Science University, Portland, OR), GM10 human dermal fibroblasts (generously provided by Dr. David R. Clemmons, University of North Carolina at Chapel Hill, NC) and rat pituitary GH3, mouse pituitary AtT20, human embryonic kidney hEK293, and CHO cells (generously provided by Dr. Richard E. Mains, Johns Hopkins University, Baltimore, MD).
Iodination of IGFBPs
Iodination of IGFBPs 1, 2, 3, 5, and 6 was carried out
with chloramine-T as per Mohan et al. (11). Briefly, 1 µg
of IGFBP was oxidized for 60 sec in the presence of 1 mCi of sodium
[125I]. The reaction was stopped with excess KI and the
reaction mixture separated on Sep Pak C18 as described
(12).
Cell-free IGFBP-5 protease assay
A convenient and rapid cell-free proteolysis assay was
developed, the substrate for which is [125I]-labeled
rhIGFBP-5. Whereas consideration was given to a rat IGFBP-5 substrate,
none was available to us. Moreover, human and rat IGFBP-5 are 97%
homologous (13). The IGFBP-5 proteolysis assay was predicated on a
similarly designed assay for IGFBP-3 proteolysis (14). Specifically,
serum-free media (50 µl) conditioned for 72 h by untreated or
treated granulosa cells (5 x 105 viable cells/dish)
were incubated for up to 5 h at 37 C with 30 x
103 cpm (5 µl) of [125I] rhIGFBP-5
substrate. The reaction buffer was comprised of 20 mM
HEPES, 5 mM CaCl2, and 0.1% BSA at a pH of
7.5. The reaction was stopped by the addition of SDS sample buffer and
the products subjected to SDS-PAGE fractionation (15%) and
autoradiography (2–5 days) as previously described (15). It is
recognized that the [l25I] rhIGFBP proteolysis assay is
limited by its ability to detect only [125I]-labeled
(i.e. tyrosine-endowed) fragments. Tyrosine-free fragments
will not be detected. In one experimental paradigm, cell-free assays
were carried out using [125I] IGFBP-1, 2, 3 and 6 as
substrates. Assay conditions were otherwise identical.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) or to buffer alone (TI;
see Fig. 4). In contrast, exposure to FSH-conditioned media (F; Fig. 1)
produced evidence of IGFBP-5-directed proteolytic activity as early as
15 min into the incubation. Specifically, the signal corresponding to
intact [125I] rhIGFBP-5 displayed time-dependent
diminution in intensity without achieving a formal maximum at the end
of the experiment (180 min incubation). In contrast, an apparent steady
state was reached by 120 min for the relative intensity of the novel
putative [125I]-labeled cleavage products (x1 and x2)
of rhIGFBP-5. Given the relatively rapid time course, subsequent
incubations were carried out over a 1-h period.
|
|
), cleavage of [125I] rhIGFBP-5
yielded two fragments, an estimated 19.5 kDa [125I]
IGFBP-5(x1) and a 17.5 kDa [125I]
IGFBP-5(x2). To exclude the possibility that the smaller
IGFBP-5 fragment represents a sequentially cleaved product of the
larger precursor, the order of appearance of the two fragments in the
course of a cell-free assay was monitored. Subject to the limitations
imposed by the initial 15 min lag, the apparent simultaneous appearance
of the two cleavage products supports the notion that the two fragments
are distinct and that a single cleavage site is involved in their
generation. The combined MW of the two fragments approximates that of
the parent molecule. Still, short of direct sequence data for the
fragments in question, the current data cannot exclude the possibility
of sequential cleavage and of more than one scission site. Note was
also made of the fact that the intensity of the band corresponding to
the smaller cleavage product IGFBP-5(x2) proved
consistently stronger when compared with the intensity associated with
the heavier cleavage product IGFBP-5(x1). Although the
reason(s) for the above pattern remain uncertain, it is possible that
the smaller cleavage product is endowed with a greater number of
iodinated tyrosine residues.
IGFBP-5 endopeptidase: optimization of induction parameters
To establish the optimal FSH dose required for the induction of
the IGFBP-5 endopeptidase, granulosa cells were cultured for 72 h
under serum-free conditions in the absence (C) or presence of
increasing concentrations (0.02–2 mIU/ml) of FSH. As shown (Fig. 2), media conditioned by FSH-treated (but
not untreated) granulosa cells displayed progressive dose-dependent
proteolytic activity as evident by the attenuation of the signal
corresponding to [125I] rhIGFBP-5 as well as by the
appearance of labeled putative fragments thereof. These observations
suggest measurable FSH activity at concentrations as low as 0.06
mIU/ml. All subsequent conditioning was carried out with 2 mIU/ml of
FSH to ensure reproducible efficacy. Qualitatively comparable
dose-response relationship were documented for PMSG, an equine
gonadotropin with both FSH and LH-like properties, the minimal
effective dose being 0.1 IU/ml (Fig. 3).
|
|
, upper panel),
the acquisition of IGFBP-5 endopeptidase activity proved to be time
dependent, measurable activity being first noted at the 24 h time
point as assessed by a modest diminution of the [125I]
IGFBP-5 signal and the enhancement of the [125I]
IGFBP-5x2 signal. An apparent maximal effect appeared to
have been established following 48 h of incubation. Given the
above, all subsequent incubations were carried out for 72 h to
ensure optimal induction of IGFBP-5 endopeptidase activity.
To establish the optimal cell density required for the induction of
IGFBP-5 endopeptidase activity, increasing densities
(105–106 viable cells/dish) were cultured for
72 h under serum-free conditions in the absence (C) or presence of
FSH (F; 2 mIU/ml). As shown (Fig. 4, lower panel), clear cut
evidence of FSH-associated IGFBP-5 endopeptidase activity was first
noted at a cellular density of 2.5 x 105 viable
cells/dish, an apparent plateau being noted at a cellular density of
5 x 105 viable cells/dish. Consequently, all
subsequent experiments were carried out at a cellular density of 5
x 105 viable cells/dish.
Is the FSH-associated acquisition of IGFBP-5 endopeptidase an FSH
receptor-independent phenomenon?
Given the apparent ability of IGF analogs to protect IGFBP-5
from proteolysis through direct, receptor-independent interactions (7),
one must rule out the possibility that the direct association between
FSH and IGFBP-5 may enhance the susceptibility of IGFBP-5 to
proteolysis by endopeptidase(s) already present in the untreated state.
In addition, consideration must be given to the possibility of
endopeptidase activity surviving established purification protocols of
oFSH. Indeed, a contaminating serine protease has recently been
reported for hCG (16). Media conditioned by untreated granulosa cells
were assayed for their ability to proteolyze [125I]
rhIGFBP-5 under cell-free conditions as described in Materials
and Methods, in the absence or presence of FSH. As shown (Fig. 5), provision of oFSH (2 mIU/tube) or
rhFSH (2 mIU/tube) proved to have no effect on IGFBP-5 endopeptidase
activity, thereby arguing, by inference, against the possibility of an
FSH receptor-independent artifact or of contaminating pituitary-derived
contribution.
|
), the induction of IGFBP-5
endopeptidase activity proved FSH-exclusive. Although not shown,
treatment with other established granulosa cell agonists
(interleukin-1ß, TGFß1, EGF, endothelin-1, or TNF)
proved equally ineffective. However, the concurrent provision of either
PMA or GnRH proved inhibitory to the endopeptidaseinducing property of
FSH. Similar GnRH (17) and PMA (12, 18)-mediated modulation has
previously been reported for other end points of FSH. Activin-A (and
TGFß1, not shown) in turn proved modest enhancers of FSH
action in keeping with previous observations (19). The concurrent
application of IGF-I proved partially inhibitory to FSH hormonal action
reflecting in all likelihood carryover into the cell-free assay setting
and thus attenuation of IGFBP-5 endopeptidase activity in keeping with
previous reports (7).
|
As previously reported (7) for the murine IGFBP-5 substrate, the
concurrent presence of EDTA (5 mM) completely inhibited
proteolytic activity (Fig. 7). Although
this observation suggests the possibility of a metalloprotease
(cation-dependence), one cannot rule out a cation-activated or
stabilized nonmetalloprotease. 1,10-Phenanthroline (10 mM)
proved equally effective. This latter observation is virtually
diagnostic for a Zn2+ metalloprotease because
1,10-phenanthroline is known (20) to possess a much higher stability
constant for zinc (2.5 x 10-6 M) than
for calcium (3.2 x 10-1 M). To further
establish the metalloprotease nature of the activity in question, its
sensitivity to inhibition by high concentrations of Zn2+
was assessed. Inhibition by Zn+2 is generally attributable
to the formation of zinc monohydroxide that bridges the catalytic zinc
ion to a side chain in the active site of the enzyme, thereby producing
competitive inhibition with substrate (16). As shown (Fig. 8), the addition of increasing
concentrations of ZnCl2 produced complete inhibition of the
IGFBP-5 endopeptidase activity at concentrations >0.1 mmol. In
contrast, the addition of TIMP-1 (10–100 µg/ml) or TIMP-2 (100
µg/ml) proved ineffective (not shown), thereby arguing against an
MMP.
|
|
). However, because
PMSF decays relatively rapidly in aqueous solution (t1/2 =
55 min at 25 C and pH 7.5), the 6-h reaction was replenished twice at 2
and 4 h with an additional PMSF dose (1 mM). Relative
insensitivity to PMSF, APMSF (1–2 mM), aprotinin (19; 2
µg/ml), TPCK (100 µg/ml), and benzamidine (3 mM) argued
against the possibility of a serine protease. Chymostatin (100
µM), a putative inhibitor of chymotrypsin-like serine
proteases and some cysteine proteases proved either ineffective or
partially inhibitory (not shown). Although an aspartic protease must be
considered, the demonstration of IGFBP-5 proteolysis at pH 7.5 argues
against such a possibility because most aspartic proteases are inactive
above pH 6.0. Expectedly, pepstatin A (1 µM), an
ace-tylated pentapeptide established as a highly selective
reversible inhibitor of aspartic peptidases (22), proved ineffective
(Fig. 7). Cysteine proteases, largely acid-favoring and intracellular
are likewise less likely candidates. Indeed, E-64 (100
µM), a peptide epoxide, established as a highly specific,
irreversible inhibitor of cysteine proteases (23), was without effect
on the IGFBP-5 endopeptidase activity (Fig. 7). Insensitivity to
plasminogen activator inhibitor-1 (PAI-1; 6.7 µg/ml) and the presumed
lack of free plasminogen in serum-free culture media argue against
plasmin.
IGFBP-5 endopeptidase: IGFBP specificity
To determine if the proteolytic activity under study is
IGFBP-5-exclusive, media conditioned by FSH (2 mIU/ml)-treated
granulosa cells were assayed for their ability to proteolyze
[l25I] rhIGFBP-5 substrate in the presence of excess
unlabeled IGFBPs (1, 2, 3, 4, 5, 6). As expected (Fig. 9), provision of rhIGFBP-5 completely
inhibited the proteolysis of its labeled counterpart. Unexpectedly
however, comparable inhibition was noted for purified hIGFBP-1, an
IGFBP not expressed in the rat ovary (24, 25). rhIGFBP-6, a
pituitary-dependent IGFBP with limited antigonadotropic activity
(26) also proved partially effective in attenuating
[125I] rhIGFBP-5 proteolysis (not shown). In contrast,
IGFBPs 2, 3, and 4 proved without effect.
|
),
[125I]-labeled IGFBP-1 and IGFBP-6 proved resistant to
media conditioned by FSH-treated granulosa cells.
[125I]-labeled IGFBP-2 and IGFBP-3 proved equally
resistant (Fig. 11).
|
|
), none of the above, with the
exception of sera or seminal plasma (not shown) produced a cleavage
pattern comparable to that associated with FSH-treated granulosa cells.
These findings suggest a measure of tissue specificity and possibly the
existence of a family of endopeptidases concerned with the cleavage of
IGFBP-5.
|
), centricon-mediated
fractionation of FSH-conditioned media revealed IGFBP-5 endopeptidase
activity only in the fraction representing proteins of molecular weight
>100K. These findings suggest that the activity in question
constitutes a high molecular weight protein.
|
), proteolysis of
[125I] rhIGFBP-5 was completely inhibited over the low pH
range. However, proteolysis proceeded unencumbered at neutral and basic
pH as created by HEPES or CHES. It is recognized that the above
experiments are unlikely to yield a true pH optimum given the absence
of pure material and the consequent variability in substrate
concentration. Still, these findings support earlier conclusions (Fig. 7) in that aspartic and cysteine proteases display acidic pH optimums,
whereas serine and metalloproteases favor an alkaline environment
(23).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although the preceding studies suggest that the IGFBP-5 endopeptidase is a Zn2+ metalloprotease, the relative rather than absolute specificity of the inhibitors in question must be borne in mind. In addition, Zn2+ serine proteases have not been identified. Moreover, the IGFBP-5 endopeptidase activity may be represented by more than one protease. The implications of a mixed population of proteases on the interpretation of inhibitor sensitivity data are self-evident.
The current findings suggest that although unlabeled IGFBP-1 and IGFBP-6 are capable of blocking IGFBP-5 endopeptidase activity, they are unable to serve as substrate. Although the precise reason(s) for this apparent discrepancy remain unknown, consideration must be given to the following possibilities: 1) the blocking properties of IGFBP-1 and IGFBP-6 are noncompetitive in nature in that they do not involve competition for the active enzymatic site of the endopeptidase. IGFBP-3-mediated inhibition of IGFBP-4 endopeptidase activity has been previously described (31); 2) the iodination of IGFBP-1 and IGFBP-6 produces conformational changes that preclude their potential cleavage by the endopeptidase; and 3) the extent of IGFBP-1 and/or IGFBP-6 cleavage is modest or gives rise to noniodinated (tyrosine-free) residues.
Subject to the clarification of the above issues, the above findings suggest a significant level of specificity for the FSH-associated IGFBP-5-directed endopeptidase. Note is also made of the fact that we (7), unlike others (3), failed to document IGFBP-4-directed endopeptidase activity in media conditioned by either control of FSH-treated cultured granulosa cells. A high degree of endopeptidase specificity favors the existence of a unique IGFBP-5 motif not present in other IGFBPs. Although less than absolute specificity might be suggested by the substantial structural homology displayed by the six known IGFBP species, existing homologies are limited to the amino and carboxy termini of the IGFBPs (18). Accordingly, and in keeping with the size of the IGFBP-5 cleavage products observed (17.5 and 19.5 kDa), it would appear more likely that the putative cleavage site is situated within the central (unique) portion of the IGFBP-5 molecule for which little or no homology has been detected in the other known IGFBP species (18). The above notwithstanding, the present findings cannot exclude the acquisition of protease activity against the other IGFBPs (e.g. inappropriate assay conditions or the generation of nontyrosine-bearing fragments).
The notion that IGFBP-5 may be a prohormone is a new and speculative
one prompted by the observation that IGFBP-5 is endowed with
dibasic (x2) and tribasic (x2) potential scission motifs as well as a
stretch of basic amino acids comprising a tetrabasic arginine-flanked
motif similar (albeit not identical) to the complex recognition signal
characteristic of exported precursors. That this in fact may be the
case has recently been further suggested by the observation of a
(carboxy-truncated) form of IGFBP-5 derived from osteoblast-like cells
for which intrinsic mitogenic and IGF-I-enhancing activities have been
noted (32). What’s more, a recombinant version of the
carboxy-truncated IGFBP-5 in question has been shown to specifically
bind to mouse osteoblasts (32). These findings suggest that
cell-surface attachment may in fact mediate the intrinsic mitogenic
activity of the truncated IGFBP-5 under study. In this respect,
IGFBP-5 may be likened to a growing number of prohormones the
maturation and diversification of which depends on endoproteolytic
processing (10, 33, 34, 35, 36) by the so-called prohormone convertases (PCs).
In this context, the 133–139 stretch of basic amino acids (not shared
with other IGFBPs) is of particular interest (Fig. 15) in that its tetrabasic component
KXRR closely approximates the one cleaved at the proinsulin B chain/C
peptide junction (KTRR) and in the bone morphogenic protein-3 precursor
(KARR). In addition, this very same tetrabasic component closely
approximates the consensus cleavage sequence
(RXK/R R) of furin, the mammalian homologue of
the yeast Kex2 protease. Because deviations from that consensus motif
are still cleavable (e.g. EFKR 
EL), it is conceivable
that the KDRR stretch may be a major scission site giving rise to
cleavage products in reasonably good agreement with the size of IGFBP-5
fragments as estimated by PAGE (Fig. 1). Clearly, whether furin or a
related enzyme may be involved remains to be seen. Recent studies
indicate that PC4 (a new member of the PC family) is not expressed in
the mammalian ovary (36).
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received September 26, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. J. Spicer Proteolytic Degradation of Insulin-Like Growth Factor Binding Proteins by Ovarian Follicles: A Control Mechanism for Selection of Dominant Follicles Biol Reprod, May 1, 2004; 70(5): 1223 - 1230. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. M. Rivera and J. E. Fortune Proteolysis of Insulin-Like Growth Factor Binding Proteins -4 and -5 in Bovine Follicular Fluid: Implications for Ovarian Follicular Selection and Dominance Endocrinology, July 1, 2003; 144(7): 2977 - 2987. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Firth and R. C. Baxter Cellular Actions of the Insulin-Like Growth Factor Binding Proteins Endocr. Rev., December 1, 2002; 23(6): 824 - 854. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Chamoun, D. Choi, A. B. Tavares, L. C. Udoff, E. Levitas, C. E. Resnick, R. G. Rosenfeld, and E. Y. Adashi Regulation of Granulosa Cell-Derived Insulin-Like Growth Factor Binding Proteins (IGFBPs): Role for Protein Kinase-C in the Pre- and Posttranslational Modulation of IGFBP-4 and IGFBP-5 Biol Reprod, September 1, 2002; 67(3): 1003 - 1012. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Hourvitz, A. Kuwahara, J. D. Hennebold, A. B. Tavares, H. Negishi, T. H. Lee, G. F. Erickson, and E. Y. Adashi The Regulated Expression of the Pregnancy-Associated Plasma Protein-A in the Rodent Ovary: A Proposed Role in the Development of Dominant Follicles and of Corpora Lutea Endocrinology, May 1, 2002; 143(5): 1833 - 1844. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Shalev, S. Goldman, and I. Ben-Shlomo The balance between MMP-9 and MMP-2 and their tissue inhibitor (TIMP)-1 in luteinized granulosa cells: comparison between women with PCOS and normal ovulatory women Mol. Hum. Reprod., April 1, 2001; 7(4): 325 - 331. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Poretsky, N. A. Cataldo, Z. Rosenwaks, and L. C. Giudice The Insulin-Related Ovarian Regulatory System in Health and Disease Endocr. Rev., August 1, 1999; 20(4): 535 - 582. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. A. Booth, M. Boes, B. L. Dake, and R. S. Bar Isolation and characterization of plasmin-generated bioactive fragments of IGFBP-3 Am J Physiol Endocrinol Metab, March 1, 1999; 276(3): E450 - E454. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. T. Overgaard, H. B. Boldt, L. S. Laursen, L. Sottrup-Jensen, C. A. Conover, and C. Oxvig Pregnancy-associated Plasma Protein-A2 (PAPP-A2), a Novel Insulin-like Growth Factor-binding Protein-5 Proteinase J. Biol. Chem., June 8, 2001; 276(24): 21849 - 21853. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
