| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Obstetrics and Gynecology, Department of Cell Biology, Duke University Medical Center, Durham, North Carolina 27710
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
fertilization, gamete biology, oocyte development, ovary, ovum
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This endopeptidase family was identified initially by the Saccharomyces cerevisiae protein Kex2, a calcium-dependent protease involved in cleavage of yeast 
-mating factor [4, 5], and consists of related proteins in many different organisms, including Caenorhabditis elegans, Drosophila, and mammals [6]. The seven identified mammalian proteins proteolyze and subsequently activate substrates such as serum proteins, prohormones, and receptors [7]. The best-characterized member of this family, furin, is a ubiquitously expressed endopeptidase of the constitutive secretory pathway. Although it is predominantly localized in the trans-Golgi region, furin is also found in endosomes, at the cell surface, or outside the cell, having been shed by cleavage upstream of its transmembrane domain [8].
The possibility of zp protein processing by a convertase of the furin family suggests an attractive hypothesis to describe assembly of the zp. As proposed [9, 10], this mechanism describes extracellular proteolysis at the furin consensus cleavage site of each of the three zp proteins to facilitate matrix formation and progressive expansion of the growing zp. Support for this hypothesis is found in the observation that mouse ZP1 and ZP2 are localized at the plasma membrane when zp assembly is prevented by knockout of the zp3 gene [11]. Furthermore, experimental evidence for C-terminal proteolytic processing has recently been reported for native mouse ZP2 and ZP3 [12]. To expand and extend these findings to human zp, we have utilized a recombinant expression system that offers a novel experimental approach. Using a mammalian expression system, we provide evidence that the three human zp proteins are cleaved by a member of the furin convertase family during their synthesis and secretion. Secreted recombinant zp proteins have a migration pattern in SDS-PAGE similar to that of native human zp proteins. Inhibition of furin cleavage by chemical treatment or consensus site mutation retards the migration of zp proteins, indicating that they are not processed normally. Exogenous recombinant furin efficiently cleaves human ZP3 in vitro, confirming that ZP3 is a furin substrate. Cleavage of native human ZP3 was also compared to its recombinant counterpart to investigate whether proteolysis at the furin cleavage site is a mechanism used by the growing oocyte to expedite zona matrix formation in vivo.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Molecular biology reagents were purchased from Stratagene (La Jolla, CA) or Boehringer-Mannheim (Indianapolis, IN). Polymerase chain reaction (PCR) amplification was performed using pfu polymerase and a Robocycler (Stratagene). Synthetic oligonucleotides were purchased from Sigma-Genosys (The Woodlands, TX). Expression vector DNA was prepared using BIGGESTprep kits (5 Prime
3 Prime, Boulder, CO). All chemical reagents were purchased from Sigma (St. Louis, MO), with the exception of pefabloc (Boehringer-Mannheim) and decanoyl-Arg-Val-Lys-Arg-chloromethylketone (RVKR; Bachem, King of Prussia, PA). All media and transfection reagents were purchased from Gibco-BRL (Gaithersburg, MD).
ZP Constructs
A cDNA encoding human ZP1 (ZPB) was generously provided by J.D. Harris (Zonagen, The Woodlands, TX). Human ZP2 and ZP3 and mouse ZP3 cDNAs were kindly donated by J. Dean (NIDDK, NIH, Bethesda, MD). These cDNAs were PCR-amplified from plasmid DNA and subcloned using BglII and ClaI into a modified version of the mammalian expression vector pCMV4 [13]. A hemaggluttinin antigen (HA) epitope tag (PYDVPDYA) and a 6-Histidine (6xHis) sequence were engineered at the carboxyl termini of the expressed zp proteins. All constructs were verified by restriction analysis and nucleotide sequencing.
ZP furin- Constructs
Human ZP1furin- was constructed using the primer set 5'-CGATGGTGACCTGTCCGGACC-3' (ZP11367–1384) and 5'-GGCCTCTAGACACCATCATGATGA-3' (vector sequence, reverse compliment [rc]) to amplify an internal, 300-base pair BstEII-Xba1 human ZP1 fragment containing a mutated furin consensus cleavage site (SRRR to SANA) and an engineered AccIII site. This fragment was subcloned back into the human ZP1 expression vector.
Human ZP2furin- was constructed using an N-terminal 5'-primer and a ZP21898–1930 3'-primer (5'-CGTAGCCCCTGTGGCTTGTGCATGCGCAGAGGACACA-3', rc) containing a mutated furin consensus cleavage site (RHRR to AHAQ) and an engineered SphI site. The amplified 1.9-kilobase (kb) BglII-BglI fragment was used to recreate a human ZP2 expression vector.
Human ZP3furin- was similarly constructed using an N-terminal 5'-primer and a ZP31026–1065 3'-primer (5'-CGCGCACATGCGCGGCGTTAGCGGAAGCAGACCTGGACCACTG-3', rc) to amplify a 1.0-kb fragment containing a mutated furin consensus cleavage site (RNRR to ANAA). This fragment was partially digested with BglII and BglI, resulting in digestion of the ends and not the internal BglI site, and the gel-purified, 1.0-kb fragment was used to recreate the human ZP3 expression vector.
All ZPfurin- constructs were verified by sequencing, and for human ZP1 and ZP2, the newly engineered restriction enzyme sites were confirmed by restriction analysis.
Expression of Recombinant zp Proteins
Human embryonic kidney 293 cells (supplied by A.M. Pendergast, Duke University, Durham, NC) were cultured in Dulbecco modified Eagle medium (DMEM) containing 10% (v/v) fetal calf serum (FCS) and transiently transfected with an expression construct using the polycationic lipid lipofectamine (Gibco-BRL). Cells were transfected with ZP expression constructs to express recombinant zp proteins or mock-transfected using expression vectors lacking a cDNA insert. Transfections were carried out in an enriched serum-free media (Opti-MEM). After 6 h, an equal volume of DMEM containing 20% FCS was added to the transfection and incubated overnight. Transfected cells were then incubated in Opti-MEM (Gibco BRL, Grand Island, NY); secreted proteins were collected after 48–72 h. Supernatant was filtered through a 0.2-µm membrane and concentrated 10-fold using a Centricon 30 (Millipore, Bedford, MA) concentrator. Cell-associated zp proteins were collected by washing transfected cells with PBS and incubating for 1 h on ice with lysis buffer (1% Nonidet P-40, 50 mM Tris [pH 7.4], 2 mM EDTA, 100 mM NaCl, and protease inhibitors [1 µg/ml of leupeptin, 2 µg/ml of antipain, 10 µg/ml of benzamidine, 1 µg/ml of chymostatin, 1 µg/ml of pepstatin, and 24 µg/ml of pefabloc]). The presence of sulfonyl fluoride protease inhibitors does not inhibit furin protease activity [14]. Cells were scraped off the dishes, and the cell suspension was sonicated (three times for 20 sec each). After removal of particulate cell debris by centrifugation at 10 000 x g for 10 min, lysates were stored at -70°C.
Inhibitor Treatment
Experiments designed to inhibit cleavage by furin family members were carried out using transfected cells that had been cultured for 48 h. These cells were pretreated in Opti-MEM containing 5 µg/ml of Brefeldin A or 5–50 µM RVKR for 2 h and then treated with fresh media containing the inhibitor for 6 h. Secreted zp proteins and cell lysates were collected at 6 h and prepared as described above.
Preparation of Native zp
Human oocytes were obtained from frozen ovarian tissue as described previously [15, 16]. The oocytes were pooled and solubilized by boiling in electrophoresis sample buffer [17] containing 100 mM ß-mercaptoethanol.
Preparation of Human ZP3 from Mouse Ovaries
Mice that lack mouse ZP3 but transgenically express human ZP3 ("ZP3 rescue mice" [18]) were generously supplied by T. Rankin and J. Dean (NIDDK, NIH). Native mouse zp containing mouse ZP1, mouse ZP2, and human ZP3 were isolated from ovaries of 3-wk-old female mice and prepared as described previously [19]. Zonae were solubilized in PBS/polyvinylpyrrolidone by incubation at 60°C for 60 min. The protein concentration was determined by the DC protein assay (Bio-Rad, Richmond, CA).
Anti-zp Antisera
During our initial analysis of recombinant human zp proteins, expression was verified by the use of polyclonal antisera that had been previously authenticated elsewhere, and generously donated, as follows: anti-human ZP1 (ZPB) and anti-human ZP2 (ZPA) by J.D. Harris (Zonagen), and anti-human ZP3 by M. van Duin (Organon, Oss, The Netherlands). These reagents (-ZP1[Harris], -ZP2[Harris], and -ZP3[van Duin]) were used solely for the data shown in Figure 2. We subsequently developed several anti-zp protein antisera as described below. All experiments were conducted in compliance with an approved Duke University Animal Care and Use Committee animal study protocol.
|
Anti-human ZP1 antisera Polyclonal antibodies against human ZP1 were produced using a DNA immunization strategy. Human ZP1 expression vector DNA (described above), prepared using the BIGGESTprep kits, was solubilized in sterile PBS. A New Zealand white rabbit was immunized with 1 mg of DNA by i.m. injections as described by Davis et al. [20]. Three weeks after the initial DNA injection, the rabbit received a second injection. Rabbit antiserum was collected monthly and used to detect recombinant human ZP1 by Western blot analysis (described below). Specific reactivity against recombinant human ZP1 was observed 2 mo following the second injection and persisted for more than 1 yr.
Anti-glutathione-S-transferase ZP antisera Glutathione-S-transferase (GST) fusion proteins were used to generate polyclonal antibodies against human ZP2 and ZP3. The GST fusion proteins encompassing the ZP domain of human ZP2 (amino acids 344–636) or human ZP3 (amino acids 121–334) were constructed by amplification of the region from the ZP expression constructs described above using the following primers: GST-hZP2ZP (5'-GCGCGAATTCTCCTTCGGCCAGAGACAGTATCCAT-3' and 5'-GCGCGAATTCCACAGGGCAGGTCACAGAACACA-3'), and GST-hZP3ZP (5'-GCGCGAATTCGCCCCGTGGGAAACCTGTC-3' and 5'-GCGCGAATTCCCTGGAATGGCTTGGAGT-3'). These fragments were subcloned into the EcoRI site of the pGEX4T bacterial expression vector (Amersham Pharmacia Biotech, Piscataway, NJ). The GST-ZP fusion constructs were sequenced to verify their composition. These constructs were transformed into protease-deficient BL-21 cells (Stratagene), and expressed and purified, as described by Sune and Garcia-Blanco [21]. Then, 0.2 mg of GST-hZP2ZP and 0.7 mg of GST-hZP3ZP were used to immunize New Zealand white rabbits as described by Harlow and Lane [22].
Anti-human ZP3 peptide antisera A polyclonal rabbit antipeptide antibody, anti-human ZP3353–368, as well as the peptide antigen HVTEEADVTVGPLIFL, representing residues 353–368 of human ZP3, were prepared commercially by Sigma-Genosys.
Immunoblotting
Protein samples were prepared in sample buffer containing 100 mM ß-mercaptoethanol, heated for 8 min at 95°C, and separated using SDS-PAGE in 10% or 12% gels according to the method of Laemmli [17]. The gels were blotted onto nitrocellulose (Schleicher and Schuell, Keene, NH) using an Amersham Pharmacia Biotech blotting apparatus and stained with Ponceau S to verify transfer. The nitrocellulose was blocked using 1% nonfat dry milk diluted in TBST (10 mM Tris, 100 mM NaCl, and 0.1% Tween-20; pH 7.5) and probed using primary antibodies (anti-HA [mAb 12CA5; Roche, Piscataway, NJ] or anti-zp antibodies described above) diluted in the same blocking buffer. Blots were washed with TBST and probed with horse radish peroxidase (HRP)-labeled secondary antibodies (anti-mouse-HRP [KPL, Gaithersburg, MD] for anti-HA, anti-human-HRP [Zymed, San Francisco, CA] for -ZP1[Harris] antisera, or anti-rabbit HRP [Sigma] for all other antibodies) according to the manufacturer's recommendations. The blots were developed using enhanced chemiluminescence (Amersham Pharmacia Biotech). All Western blot experiments were replicated at least three times. When necessary, the membranes were stripped with a solution of 100 mM ß-mercaptoethanol, 2% SDS, and 62.5 mM Tris (pH 6.7) for 30 min at 50°C and washed with TBST before reprobing with another antibody.
Wheat Germ Agglutinin Chromatography
Wheat germ agglutinin (WGA)-Sepharose (Vector, Burlingame, CA) was packed into 2-ml columns and washed with a buffer of 10 mM Hepes, 150 mM NaCl, 1 mM CaCl2, and 1 mM MnCl2 (pH 7.5). Concentrated (10x) tissue-culture supernatant was diluted at least 1:1 (v/v) with buffer and loaded onto the WGA-Sepharose column. After washing the beads with 
50-fold excess of buffer, recombinant zp glycoproteins were eluted sequentially with five volumes of 50 mM GlcNAc followed by five volumes of 500 mM GlcNAc. In both elutions, ZP glycoproteins were present, but they were more abundant in the 500 mM elution. The eluates were concentrated in a Centricon 30 until their protein concentration was >1 mg/ml. Alternatively, small-scale experiments were conducted in a similar way using 25 µl of packed WGA-Sepharose eluted with two volumes of 2x sample buffer. The WGA-enriched samples were used for all human ZP1 Western blots, because crude material failed to elicit detectable reactivity with human ZP1-specific antisera.
Furin Cleavage
Purified recombinant furin (specific activity, 1650 U/ml) was generously supplied by T. Komiyama and R. Fuller (University of Michigan, Ann Arbor, MI). Cleavage reactions were performed for 1 h at 37°C using 50 µg of recombinant human ZP3 lysate and 20 U of purified furin in a reaction mixture (50 µl) of 20 mM 2-(N-morpholino)ethansulfonic acid, sodium salt (Sigma) (pH 7.0), 1 mM CaCl2, and 0.1% Triton X-100. For reactions lacking exogenous furin, a solution of 10 mM Tris and 1 mM CaCl2 was substituted. Cleavage reactions were terminated by addition of 15 µl of 4x sample buffer.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mammalian expression vectors were constructed with human ZP1, ZP2, or ZP3 cDNAs and used to transfect 293 cells. As depicted for human ZP3 in Figure 1, the C-terminal region of all cDNA sequences was modified by addition of an HA sequence and a 6xHis tag. Cells transfected with human ZP expression vectors expressed recombinant human zp proteins, which were secreted into the cell-culture media. The ZP proteins appeared to undergo processing when expressed in human 293 cells comparable to that in human oocytes, because secreted recombinant human zp proteins and native human zp proteins migrate similarly in SDS-PAGE when visualized by anti-zp immunodetection (Fig. 2). Recombinant human ZP1, ZP2, and ZP3 migrate at apparent average molecular weights of 80, 110, and 65 kDa, respectively. Whereas native and recombinant human ZP2 and ZP3 have identical migration patterns, native human ZP1 appears to have a lower molecular weight compared to recombinant human ZP1. Additional native human zonae were not available to resolve this discrepancy; therefore, it remains to be determined if this reflects a true difference in posttranslational processing for these proteins. No immunoreactivity against ZP1, ZP2, or ZP3 was detected in cell supernatants from cells transfected with vector lacking a zp cDNA insert (mock transfectants). Similar results were obtained for mouse recombinant zp proteins (data not shown).
|
Secreted Human ZP Proteins Migrate Differently from Their Cell-Associated Counterparts
The possibility of C-terminal cleavage of recombinant human zp proteins was revealed by SDS-PAGE comparison of recombinant zp proteins present in cell lysates versus those in cell supernatants (Fig. 3). Probing with an anti-HA antibody demonstrated that, although cell-associated recombinant human ZP3 (Fig. 3A, lane 2) contains the C-terminal epitope, secreted recombinant human ZP3 does not (Fig. 3A, lane 4). Stripping and reprobing this blot with anti-human ZP3 antisera confirmed the presence of similar levels of reactive recombinant human ZP3 in the cell lysate and supernatant (Fig. 3B, lanes 2 and 4). Similar results were obtained for recombinant human ZP1 and recombinant human ZP2 (data not shown). For all three recombinant human zp proteins, no zp-specific reactivity was observed when lysates or supernatants prepared from mock-transfected cells were used (Fig. 3B, lanes 1 and 3).
|
Modulation of Processing at the Cleavage Consensus Site
A proteolytic processing site for furin family convertases present upstream of the transmembrane domain in all zp protein sequences is a likely site for the C-terminal cleavage observed in the recombinant human zp proteins. To investigate whether a furin family member is responsible for C-terminal cleavage, three methods—interference with Golgi modifications, pharmacological inhibition, and consensus site mutation—were used.
Interference with Golgi modifications and pharmacological inhibition Two pharmacological inhibitors were used to analyze cleavage of human zp proteins (Fig. 4A). The first inhibitor, Brefeldin A, prevents interaction of recombinant zp proteins with furin enzymes by blocking export from the endoplasmic reticulum (ER) to the Golgi complex [23]. Furin convertases have been localized to several cellular compartments downstream of the ER, including the trans-Golgi-network, the cell surface, and endosomes [8]. Because the secretory processing pathway is highly ordered, each successive step depends on completion of the preceding one. Thus, inhibiting recombinant human ZP3 transport from the ER would be expected to inhibit both furin cleavage as well as Golgi-derived carbohydrate processing (Fig. 4A). A second pharmacological inhibitor, RVKR, mimics the furin cleavage sequence and directly inhibits all furin family enzymes by binding to their catalytic sites. This inhibitor does not alter protein synthesis, secretion, or cell viability in 293 cells [24]. Human ZP3 was chosen to investigate the effects of these inhibitors because of the high sensitivity of detection reagents for this protein.
|
Cells transfected with human ZP3 were treated with either Brefeldin A or RVKR; lysates were prepared and analyzed with anti-human ZP3 antisera (Fig. 4B). The untreated recombinant human ZP3 lysate (Fig. 4B, lane 2) reflects the range of recombinant human ZP3 species from different compartments of the cell, with the predominant reactivity at 65 kDa. Brefeldin A treatment causes accumulation of a 75-kDa form of recombinant human ZP3 (Fig. 4B, lane 3), whereas RVKR treatment shifts the predominant human ZP3 species to near 80 kDa (Fig. 4B, lane 4). The intermediate gel shift demonstrated by Brefeldin A-treated recombinant human ZP3 lysates is likely due to inhibition of both cleavage at the furin consensus site and Golgi processing. In addition, migration of RVKR-treated recombinant human ZP3 (Fig. 4B, lane 4) matched the migration of recombinant human ZP3 from cells transfected with a construct lacking an intact cleavage site (Fig. 4B, lane 5; ZP3furin-, see below) suggesting that consensus site mutation and RVKR treatment were equally effective methods to selectively inhibit convertase cleavage of fully processed recombinant human ZP3.
Consensus site mutation Human ZP cleavage consensus sites were mutated in ZP1 (SRRR to SANA), ZP2 (RHRR to AHAQ), and ZP3 (RNRR to ANAA). These constructs are denoted as ZPfurin- and were used to transiently transfect 293 cells. Proteins from wild-type ZP and ZPfurin- transfections were analyzed by Western blot (Fig. 5). All ZPfurin- constructs produced recombinant zp proteins that migrated more slowly in SDS-PAGE, consistent with abnormal processing due to the absent consensus cleavage sequence, RXK/RR.
|
These constructs effectively demonstrated a shift in migration of ZPfurin- proteins consonant with elimination of their consensus cleavage sites, but they might also imply a role for convertase cleavage in zp protein secretion. To explore this possibility, a more thorough analysis of human ZP3furin- secretion was attempted. When cells expressing similar levels of wild-type human ZP3 and human ZP3furin- proteins were analyzed by Western blot to detect relative amounts of secreted recombinant human ZP3, mutation of the furin consensus cleavage site diminished, but did not abolish, the presence of recombinant human ZP3 in the cell supernatant (data not shown). Similarly, immunofluorescence with anti-ZP3 antisera did not reveal a significant increase in the amount of plasma membrane-associated human ZP3furin- (data not shown), suggesting that convertase cleavage and protein secretion are independent cellular activities.
In Vitro Furin Cleavage
To verify that recombinant human ZP3 served as a substrate for furin, cell-associated recombinant human ZP3 derived from transfected cells treated with Brefeldin A or RVKR or from cells transfected with the human ZP3furin- construct was analyzed for cleavage in vitro in the absence or presence of exogenously added purified furin (Fig. 6). Following exposure to furin, a measurable decrease in molecular weight of the predominant reactivity was observed for recombinant human ZP3 (Fig. 6A, lane 2 vs. 6). The faster migration of human ZP3 following exposure to exogenous furin mimicked the pattern seen for secreted recombinant human ZP3 (Fig. 6A, lane 10), consistent with the occurrence of furin cleavage before secretion of recombinant human ZP3. This shift in migration (from predominant reactivity at 75–80 kDa to 65 kDa) was also evident in recombinant human ZP3 acquired from cells treated with RVKR (Fig. 6A, lane 3 vs. 7) or Brefeldin A (Fig. 6A, lane 4 vs. 8); these treatments were predicted to inhibit furin-mediated cleavage of recombinant human ZP3 in vivo (see above). The intermediate shift in migration observed in lane 7 of Figure 6A with the RVKR lysates can be attributed to the presence of residual RVKR in the sample, which would be expected to inactivate the exogenous furin. As anticipated, when the furin cleavage site was absent, as in recombinant human ZP3furin-, exposure to exogenous furin did not affect electrophoretic migration (Fig. 6A, lane 5 vs. 9).
|
Successful cleavage of human ZP3 at the furin consensus cleavage site can be predicted to generate two fragments, the smaller of which would contain the C-terminus and its associated epitope (HA) tag. To visualize this cleavage product, these same samples were also assessed with an anti-HA antibody on Western blots that had been optimized for low-molecular-weight proteins (Fig. 6B). The relative intensity of this cleavage product increases as a result of in vitro furin cleavage (Fig. 6B, lanes 3 vs. 7, lane 4 vs. 8). A significant amount of the C-terminal product is also present in unmodified recombinant human ZP3 lysates (Fig. 6B, lane 2), demonstrating cleavage by endogenous convertases. However, not all the human ZP3 present in these lysates has been fully processed, because exogenous furin alters the migration of recombinant human ZP3 (Fig. 6A, lanes 2 vs. 6) and increases the relative intensity of the C-terminal product (lighter exposure of Fig. 6B; data not shown). When the furin cleavage site is eliminated, as in the human ZP3furin- constructs, exogenous furin is unable to generate the cleavage fragment (Fig. 6B, lane 5 vs. 9). No reactivity was observed for mock-transfected lysates (Fig. 6B, lane 1).
Thus, this analysis monitors both products from the cleavage reaction of the substrate, recombinant human ZP3: the N-terminal portion (Fig. 6A), and the C-terminal cleavage fragment (Fig. 6B). The retarded migration of the human ZP3 and the corresponding increase in relative intensity of the C-terminal cleavage fragment established that recombinant human ZP3 is an in vitro substrate for furin and that this cleavage requires presence of the RXK/RR consensus site.
C-Terminal Cleavage Product in Native Human ZP3
Because these experiments were conducted using recombinant zp proteins, they did not address whether this cleavage event actually occurs in oocytes. Therefore, C-terminal processing in oocytes was examined using antisera generated against a peptide immediately downstream of the human ZP3 furin cleavage site (amino acids 353–368; see Fig. 1). Anti-human ZP3353–368 antibodies were first analyzed using recombinant zp proteins. On Western blots, this probe recognizes the same recombinant human ZP3 cleavage product as that identified by the anti-HA antibody (Fig. 7A). This result verifies that this protein band represents the full-length cleavage product, because it contains both the furin cleavage site at its N-terminal and the HA epitope at its C-terminal. Because the peptide used to generate anti-human ZP3353–368 is identical to the corresponding region in mouse ZP3 for 15 of 16 amino acids, the same antiserum was assessed for reactivity with mouse zp proteins. Despite the sequence similarity, anti-human ZP3353–368 does not detect the cleavage product of recombinant mouse ZP3 (Fig. 7A, lower panel). To confirm that, as with recombinant human ZP3, the C-terminal cleavage product of recombinant mouse ZP3 is present, the blot was stripped and reprobed with the anti-HA antibody. Reactivity of the appropriately sized product was detected (Fig. 7A, upper panel).
|
Having documented that anti-human ZP3353–368 recognizes the full-length C-terminal cleavage product, we were eager to investigate whether this fragment is present in the native zp matrix. Because the region-specific antibody anti-human ZP3353–368 did not react with recombinant ZP3 from mice, only with that from humans, we purified native zonae containing human ZP3 from mice. These zonae were prepared from "human ZP3 rescue" mice that have the mouse zp3 knocked out and replaced by a human zp3 transgene [18, 25]. Mouse zp3 knockout mice do not make zonae and are infertile, but transgenic expression of human zp3 in these mice rescues zona formation and restores their fertility. Zonae from "human ZP3 rescue" mice contain mouse ZP1, mouse ZP2, and human ZP3. These mice were critical for this analysis, because they provided an opportune source of adequate human ZP3, assembled in zonae and not compromised by ethical considerations, to allow analysis of the small C-terminal cleavage product in assembled zona proteins.
A sample containing 3000 "human ZP3 rescue" zonae was probed with anti-human ZP3 to ascertain the presence of human ZP3 in the "human ZP3 rescue" zona preparation (Fig. 7B, upper panel). Both the "human ZP3 rescue" zonae and recombinant human ZP3 lysates exhibit reactivity, demonstrating the presence of human ZP3 in both samples. The "human ZP3 rescue" migration pattern is much broader than that of recombinant human ZP3, because this sample has been dramatically overloaded to enable detection of any C-terminal cleavage product. Notably, when the same samples were probed with anti-human ZP3353–368 to determine presence of the furin cleavage fragment in assembled native human ZP3, no reactivity was detected in the overloaded "human ZP3 rescue" sample (Fig. 7B, lower panel). This suggests that the peptide is not present in the zona matrix and is cleaved from the ZP3 protein before zona assembly.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Recombinant human ZP1, ZP2, and ZP3 secreted in this system are similar to native human ZP1, ZP2, and ZP3. The apparent molecular mass of human ZP2 and human ZP3 correspond to previously published estimates for their native counterpart zp proteins [26, 29]. It is noteworthy, however, that human ZP1 has been reported [29] to migrate at 
150 kDa, nearly double the estimate determined here. Those authors obtained their estimate using non-disulfide reducing conditions, whereas we utilized disulfide-reducing conditions and determined an average molecular weight for recombinant human ZP1 of 80 kDa. Homodimerization of ZP1 proteins is a familiar theme, having been first identified in the mouse [30], and likely is reflected in the accumulated data for human ZP1.
Due to our use of full-length recombinant proteins, we had the opportunity to analyze both cell-associated and secreted forms of the zp proteins and observed a migration difference in SDS gels for the two forms. The apparent molecular weight of the predominant form of recombinant human ZP3 was increased by treatment of cells with two inhibitors, Brefeldin A and RVKR. Brefeldin A blocks glycoprotein transport from the ER and, consequently, inhibits both convertase cleavage and Golgi modifications. However, RVKR prevents cleavage of Golgi-processed glycoproteins and produces a larger upward shift in migration. Golgi-derived, O-linked glycosylation is critical for biological activity of ZP3 [31]; the observed intermediate shift in our RVKR-treated samples suggests that this important modification also occurs in recombinantly expressed human ZP3. Site-directed mutagenesis of the consensus cleavage sites also increased the observed molecular weight of the predominant form of each of the three recombinant zp proteins, indicating that their C-termini remained intact in the absence of furin cleavage.
Analysis of in vitro furin cleavage demonstrated that human ZP3, but not human ZP3furin-, can serve as a substrate for furin proteolysis. Coordinated use of a C-terminal anti-HA antibody and an antibody directed against the region immediately downstream of the furin site (anti-human ZP3353–368) demonstrated the production of a small cleavage product and served to monitor the cleavage reaction. In vitro furin cleavage of cell-associated recombinant human ZP3 shifted its migration to match that of secreted recombinant human ZP3, suggesting that recombinant human ZP3 is cleaved by furin before secretion.
Due to the presence of a putative transmembrane domain downstream from the consensus cleavage site, alterations in cleavage should affect zp protein secretion. When furin family convertases are inhibited, full-length zp proteins would be expected to be sequestered at the cell membrane. In the ovary, cell membrane-bound zp proteins have been proposed to be an intermediate step in zona assembly [9, 10], and they have been detected in mouse oocytes [11, 12]. Although a quantitative analysis of secretion was not possible using extracts derived from the mammalian expression system used in the present study, the treatments utilized—inhibition of furin family convertases and consensus cleavage site mutation—diminished, but did not abolish, the presence of recombinant zp in the cell supernatant. In all cases, secreted human ZP3 was cleaved upstream of its C-terminus when convertase function and consensus sequences were intact. However, when convertase cleavage was inhibited, full-length human ZP3 could be detected in the culture media by antibodies directed against the C-terminal HA epitope (data not shown). The presence of full-length recombinant zp proteins in the cell-culture media under these conditions suggests that the overexpression system described here cannot address the effect from inhibition of furin cleavage on secretion of zp glycoproteins. Thus, whereas furin cleavage does occur before secretion of zp proteins, these data do not address whether such cleavage is a requirement for secretion. More rigorous analysis of this topic requires methods to fully purify recombinant zp proteins. Quantitation of recombinant zp proteins at all stages of the secretory pathway will be necessary to determine the importance of convertase cleavage for zp protein secretion.
Collectively, these results demonstrate that recombinant human ZP3 is processed by a furin family convertase, and they suggest that recombinant human ZP1 and recombinant human ZP2 are similarly proteolyzed. The specific family member responsible for cleavage is likely to be either furin, PC6B, or PC7, because like zp proteins, these family members contain transmembrane domains upstream of their C-termini [7]. These domains could colocalize the convertase and zp proteins at membranes within the secretory pathway to facilitate cleavage. These convertases are expressed in many mammalian tissues and cell lines [32, 33].
Cleavage of recombinant human zp proteins implies that these proteins are processed similarly during zona formation. Proteolytic processing has been experimentally calculated for native mouse zp proteins [12] and is shown here for human ZP3 assembled in zonae from "human ZP3 rescue" mice. The data presented here have extended the experimental evidence supporting the zona assembly paradigm by more precisely investigating enzymatic proteolysis and its reliance on consensus site sequences. These studies provide experimental evidence for cleavage of ZP proteins by a furin convertase family member; however, the question of whether such processing is a regulatory mechanism in zona assembly remains open. Further investigation is required to determine the pertinent convertase family member, the cellular localization and regulation of proteolysis, and the consequences of cleavage for zona assembly.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by NIH RO1 HD18201. 
2 Correspondence: Patricia Saling, 234 Sands Building, Research Drive, Duke University Medical Center, Durham, NC 27710. FAX: 919 681 6494; patricia.saling{at}duke.edu
3 Current address: Washington University School of Medicine, Box 8126, Renal Division, 660 S. Euclid Avenue, St. Louis, MO 63110.
Accepted: September 24, 2001.
Received: April 24, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. E. Swain and T. B. Pool ART failure: oocyte contributions to unsuccessful fertilization Hum. Reprod. Update, September 1, 2008; 14(5): 431 - 446. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. C. Darie, M. L. Biniossek, M. A. Gawinowicz, Y. Milgrom, J. O. Thumfart, L. Jovine, E. S. Litscher, and P. M. Wassarman Mass Spectrometric Evidence That Proteolytic Processing of Rainbow Trout Egg Vitelline Envelope Proteins Takes Place on the Egg J. Biol. Chem., November 11, 2005; 280(45): 37585 - 37598. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Yonezawa, K. Kudo, H. Terauchi, S. Kanai, N. Yoda, M. Tanokura, K. Ito, K.-i. Miura, T. Katsumata, and M. Nakano Recombinant Porcine Zona Pellucida Glycoproteins Expressed in Sf9 Cells Bind to Bovine Sperm but Not to Porcine Sperm J. Biol. Chem., May 27, 2005; 280(21): 20189 - 20196. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Quesada, L. M. Sanchez, J. Alvarez, and C. Lopez-Otin Identification and Characterization of Human and Mouse Ovastacin: A NOVEL METALLOPROTEINASE SIMILAR TO HATCHING ENZYMES FROM ARTHROPODS, BIRDS, AMPHIBIANS, AND FISH J. Biol. Chem., June 18, 2004; 279(25): 26627 - 26634. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Jovine, H. Qi, Z. Williams, E. S. Litscher, and P. M. Wassarman A duplicated motif controls assembly of zona pellucida domain proteins PNAS, April 20, 2004; 101(16): 5922 - 5927. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Hoodbhoy and J. Dean Insights into the molecular basis of sperm-egg recognition in mammals Reproduction, April 1, 2004; 127(4): 417 - 422. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Zhao, L. Gold, H. Dorward, L.-f. Liang, T. Hoodbhoy, E. Boja, H. M. Fales, and J. Dean Mutation of a Conserved Hydrophobic Patch Prevents Incorporation of ZP3 into the Zona Pellucida Surrounding Mouse Eggs Mol. Cell. Biol., December 15, 2003; 23(24): 8982 - 8991. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Sasanami, A. M. Hanafy, M. Toriyama, and M. Mori Variant of Perivitelline Membrane Glycoprotein ZPC of Japanese Quail (Coturnix japonica) Lacking Its Cytoplasmic Tail Exhibits the Retention in the Endoplasmic Reticulum of Chinese Hamster Ovary (CHO-K1) Cells Biol Reprod, October 1, 2003; 69(4): 1401 - 1407. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. S. Boja, T. Hoodbhoy, H. M. Fales, and J. Dean Structural Characterization of Native Mouse Zona Pellucida Proteins Using Mass Spectrometry J. Biol. Chem., September 5, 2003; 278(36): 34189 - 34202. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Sasanami, M. Toriyama, and M. Mori Carboxy-Terminal Proteolytic Processing at a Consensus Furin Cleavage Site Is a Prerequisite Event for Quail ZPC Secretion Biol Reprod, May 1, 2003; 68(5): 1613 - 1619. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
