| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ovary; |
Academy of Sciences of the Czech Republic,3 Institute of Animal Physiology and Genetics, 277 21 Libechov, Czech Republic
Department of Public Health and Cell Biology,4 University of Rome Tor Vergata, 00133 Rome, Italy
Wellcome Trust Centre for Cell-Matrix Research,5 Faculty of Life Sciences, University of Manchester, M13 9PT Manchester, United Kingdom
ABSTRACT
We have previously shown that the heavy chains (HCs) of inter-alpha-trypsin inhibitor (IalphaI) become covalently linked to hyaluronan (HA) during in vivo and in vitro expansion of porcine oocyte-cumulus cell complexes (OCCs). We have now studied by immunoblotting the synthesis of tumor necrosis factor alpha-induced protein 6 (TNFAIP6), which is essential for catalyzing this reaction in expanding mouse OCCs. Expanding OCCs were collected from preovulatory follicles of naturally cycling pigs and also after in vitro culture (24 or 42 h) in medium supplemented with FSH and pig serum. After isolation, OCCs were treated with Streptomyces hyaluronidase or Chondroitinase ABC. Matrix, cell pellet, and total extracts were analyzed by Western blotting. A band of about 35 kDa and a doublet of about 120 kDa, corresponding to the molecular weight of the native and HC-linked forms of TNFAIP6, respectively, were detected by a rabbit anti-human TNFAIP6 polyclonal antibody in matrix extracts of expanded cumuli. Moreover, we found by using a cell-free assay that porcine follicular fluid collected from follicles at 24 h after hCG stimulation contains HC-HA coupling activity. This activity was abolished by the rat anti-human monoclonal antibody A38, which has an epitope within the Link module domain of TNFAIP6. These experiments suggest that free TNFAIP6 protein was present in follicular fluid aspirated from porcine follicles 24 h after hCG stimulation. In contrast to mouse, we show that the A38 monoclonal antibody does not affect in vitro cumulus expansion of porcine OCCs.
cumulus cells, extracellular matrix, hyaluronan, oocyte-cumulus complex, ovary, pig, TNFAIP6
In most mammals, before ovulation, cumulus cells synthesize a large amount of the polysaccharide hyaluronan (HA), which is organized in a highly hydrated muco-elastic matrix [1]. This process is called cumulus expansion or mucification. It has been shown that regulation of porcine and bovine cumulus expansion differs in some aspects from that of mouse cumulus expansion. Thus, although porcine and bovine oocytes [2–5], as mouse oocytes [6, 7], produce soluble factors enabling expansion and HA synthesis by FSH-stimulated mouse cumulus cells, pig and bovine FSH-stimulated cumulus cells do not require the presence of oocytes to expand and synthesize HA in vitro [4, 8, 9]. In vivo and in vitro studies demonstrated that the heavy chains (HCs) of serum-derived inter-alpha-trypsin inhibitor (I
I) become covalently linked to HA during mouse cumulus expansion and significantly contribute to cumulus matrix organization [10, 11].
Nevertheless, experiments with mice suggest that the incorporation of such proteins in the cumulus matrix involves LH/hCG-induced changes in the blood-follicle barrier [12, 13] and functional cooperation between oocytes, cumulus cells, and granulosa cells within the follicle [14, 15]. Our recent work provided evidence that HCs of II are covalently linked to HA in pig oocyte-cumulus cell complexes (OCCs) expanded in vivo and in vitro but, in contrast to mouse, pig cumulus cells produce the factors required for completing this reaction independently from the oocyte [16]. Moreover, II molecules could freely cross the blood-follicle barrier, and follicular fluid (FF) collected before the LH surge was successfully used for stabilizing in vitro-expanded cumulus oophorus extracellular matrix [16].
Furthermore, it has been shown that TNFAIP6 is produced by cumulus and granulosa cells after an ovulatory stimulus [17–20], in parallel to HA synthesis by OCCs [6, 21, 22]. In this regard, TNFAIP6 is an inflammation-associated protein with the ability to bind HA, II, and other ligands and participate in extracellular matrix formation and remodeling [23–25]. More importantly, covalent transfer of HCs to HA does not occur in OCCs of TNFAIP6-null mice, indicating that TNFAIP6 is critically involved in this process [26]. Recent biochemical experiments using purified human II and recombinant human TNFAIP6 have confirmed that the TNFAIP6 protein has a critical role in HC transfer, where covalent complexes formed between TNFAIP6 and HCs (i.e., HC-TNFAIP6) act as intermediates in this reaction [27]. A further demonstration that TNFAIP6 is critical for the formation and stabilization of the expanded mouse OCC came from the study of Ochsner et al. [28], which used a rat monoclonal antibody against TNFAIP6, termed A38, which had been shown previously to inhibit the interaction of TNFAIP6 with HA [29]. It was found that, in addition to A38 inhibiting the formation of HC-TNFAIP6 complexes by human purified II and recombinant TNFAIP6 in vitro, this monoclonal antibody could also inhibit the formation of HC-TNFAIP6 complexes in OCCs expanded ex vivo and impair the organization of the cumulus matrix in 80% of mouse OCCs exposed to forskolin [28]. A related rat anti-TNFAIP6 monoclonal antibody (Q75), which does not inhibit either HA binding or HC-TNFAIP6 complex formation [28, 29], did not alter forskolin-stimulated OCC expansion: both A38 and Q75 recognize distinct epitopes within the TNFAIP6 Link module.
Therefore, to determine whether the mechanism underlining the formation of covalent complexes between HCs of II and HA in the cumulus matrix is common for mouse and pig, we have investigated the expression of the TNFAIP6 protein in porcine OCCs matured in vivo and in vitro, as well as its activity in FF. This study also assessed the effect of rat monoclonal antibody against TNFAIP6, termed A38, on in vitro cumulus expansion, HA synthesis, and HC-HA complex formation in vitro.
In vivo-expanded porcine OCCs were obtained from preovulatory follicles (>10 mm) of naturally cycling pigs at a local abattoir. In vivo-expanded mouse OCCs were isolated from 22-to 24-day-old CD-1 mice injected with 5 IU eCG (Folligon; Intervet), and after 48 h with 5 IU hCG (Corulon; Intervet). Ovulated OCCs were collected from the oviducts at 14 h from hCG injection. Unexpanded porcine OCCs were obtained from ovaries of prepubertal gilts (Landrace; Large White) at a local abattoir. The complexes were aspirated from antral follicles about 5–7 mm in diameter. After isolation, 30–40 OCCs were washed and cultured in four-well plates containing 300 µl medium M-199 (Sevac, Prague, Czech Republic) supplemented with 20 mM NaHCO3, 6.25 mM HEPES, 0.91 mM sodium pyruvate, 1.62 mM calcium lactate, and antibiotics, and containing 100 ng/ml recombinant FSH (Puregon; Organon, Oss, The Netherlands) and 10% porcine serum (Sigma-Aldrich, Prague, Czech Republic). In some experiments 30 µg/ml rat anti-human TNFAIP6 monoclonal antibody A38 or monoclonal antibody Q75 was also added to the culture medium. The complexes were cultured for 24 and 42 h at 38.5°C in an atmosphere of 5% CO2 in air.
In vivo- and in vitro-expanded OCCs were washed three times in PBS containing 3 mg/ml polyvinylpyrrolidone (BDH, Poole, U.K.) and a cocktail of protease inhibitors (PBS-PI; Complete Mini; Boeringher Mannheim, Mannheim, Germany). They were then either directly solubilized in reducing Laemmli buffer containing 2% (w/v) sodium dodecyl sulphate (SDS) and 5% (v/v) β-mercaptoethanol, or digested with 1 IU Streptomyces hyaluronidase or 50 mU Chondroitinase ABC (Merck-Calbiochem, Prague, Czech Republic) in 20–30 µl PBS-PI at 37°C for 2 h before adding the reducing Laemmli buffer. In some experiments, the digested samples were centrifuged at 300 x g, 4°C, for 7 min to separate the matrix extract from the cell pellet that was washed once with PBS-PI. Reducing Laemmli buffer then was added to each fraction. All samples then were boiled at 99°C for 5 min, separated in 7.5% (w/v) acrylamide/SDS gels, and transferred to Hybond-P membranes (Amersham Biosciences). Membranes were blocked with 5% (w/v) nonfat dry milk in Tris-buffered saline for 2 h at room temperature. Primary and secondary antibodies were diluted in Tris-buffered saline containing 5% (w/v) BSA and 0.05% (v/v) Tween 20. After electrophoresis, the proteins were detected with rabbit anti-human TNFAIP6 antibody (1:5000) incubated overnight at 4°C and, after membrane stripping, with rabbit anti-human II antibody (1:2000; DAKO, Carpenteria, CA) incubated overnight at 4°C, followed by rabbit anti-actin antibody (1:500; Sigma-Aldrich) incubated 1 h at room temperature. Donkey anti-rabbit IgG horseradish peroxidase-linked F (ab') 2 fragment (1:10 000; Amersham Biosciences) was used as a second antibody for 1 h at room temperature. After primary and secondary antibody, 2 x 5 min and 3 x 10 min washes were made in PBS containing 0.1% (v/v) Tween-20. Immunoreactive products were detected by enhanced chemiluminescence (Amersham-Pharmacia). In some experiments, after the first detection, membranes were stripped in 62.5 mM Tris-HCl, 2% (w/v) SDS, 0.7% (v/v) β-mercaptoethanol, pH 6.8, for 20 min at 45°C before reprobing with another antibody following the procedure already described.
TNFAIP6-Dependent HC-HA Coupling Activity in FF
Follicular fluids were obtained from unstimulated pig ovaries (medium-size follicles) or from ovaries 24 h after hCG stimulation (follicles about 10 mm in diameter) as previously described [16]. The presence of factors able to catalyze the transfer of HCs of II to HA was determined by a modification of a method previously described [30]. Briefly, 2 volumes (20 µl) of FF were mixed with 1 volume (10 µl) of Tris buffer (45 mM Tris-HCl, pH 7.4, 150 mM NaCl, 4.5 mM CaCl2) with or without 4.5 mg/ml HA (from rooster comb; Sigma) and incubated for 24 h at 37°C. In some experiments, rat monoclonal anti-human TNFAIP6 antibody A38 was added to the mixture at a final concentration of 30 µg/ml. After incubation, aliquots (15 µl) of the reaction mixtures were digested with Streptomyces hyaluronidase (1 U/µg HA) for 2 h at 37°C. Then, undigested and digested samples were mixed with reducing Laemmli buffer and analyzed by Western blot with II antibody as described above.
The Animal Research Committee of Academy of Sciences of the Czech Republic approved all procedures.
Synthesis of HA by OCC cultures and its distribution between medium and cell matrix compartments was assessed using a modification of the procedure described previously [31]. Briefly, two OCCs were cultured for 24 or 42 h in 20-µl drops of culture medium, covered with mineral oil (Sigma-Aldrich) to prevent evaporation, in the presence of 100 µCi/ml 3H-glucosamine hydrochloride (Perkin Elmer) with or without 30 µg/ml rat anti-human TNFAIP6 monoclonal antibody A38. At the end of each culture, the incubation medium was aspirated. Medium and cell matrix fractions were then treated separately with papain (3.78 activity units/20 µl; Sigma-Aldrich) for 1 h at 65°C. The extraction was completed by adding an equal volume of 8 M guanidine HCl containing 4% (v/v) Triton X-100. Each extract was eluted on a column of Sephadex G50 (2-ml bed volume) equilibrated with 0.1 M Tris, 0.1 M sodium acetate, and 0.5% (v/v) Triton X-100, pH 7.3. The excluded volume, containing the labeled macromolecules, was recovered and counted. Each sample was then digested with Streptomyces hyaluronidase (Calbiochem) for 2 h at 37°C and then chromatographed on a column of Sephadex G-50 (8 ml bed volume). The excluded and the included fractions were counted for radioactivity to determine the proportion of the radiolabeled macromolecules digested by the enzyme.
Values presented are means ± SEM from four independent experiments performed in duplicates. Two mean values were compared by using the Student t-test. A probability of P < 0.05 was considered to be statistically significant.
TNFAIP6 Protein Is Expressed by Porcine OCCs Expanded In Vivo
In vivo unexpanded (compact) and expanded porcine OCCs were digested with Streptomyces hyaluronidase. Hyaluronidase extracts were resolved by SDS-PAGE under reducing conditions and probed with an antibody specific for TNFAIP6. Proteins extracted from in vivo-expanded mouse OCCs were analyzed in parallel for comparison. Western blot analysis of total proteins from unexpanded porcine OCCs did not show any immunoreactivity with TNFAIP6 antibody (Fig. 1A). Conversely, the antibody recognized four bands in total protein extract from expanded OCCs. The same bands were revealed in the matrix but not in the cell fraction, indicating the extracellular location of the immunoreactive proteins. The major positive band had an apparent molecular weight of 35 kDa that correlates well with the size of the free TNFAIP6 protein. A doublet at 
120 kDa was also immunoreactive with the anti-TNFAIP6 antibody. Previous characterization of a similarly sized band present in mouse cumulus matrix extracts and generated by mixing purified human TNFAIP6 and II revealed that it contained TNFAIP6 linked to the HC1 and HC2 of II molecules [20, 27, 32]. The membrane was thus stripped and reprobed with anti-II antibody (Fig. 1B). In addition to a broad band of 80 kDa, corresponding to the HCs of II, and of a band of 160 kDa, likely representing HA-linked pairs of HCs resistant to hyaluronidase digestion [14, 27], a faint band of 120 kDa was recognized by II antibody in pig matrix, as in mouse. In addition, the 120-kDa TNFAIP6 species was also found in pig cumulus matrix obtained from OCCs treated with chondroitinase ABC, supporting previous findings that chondroitin sulfate chain of II is not involved in cross-linking HC and TNFAIP6 in the complex [20, 27]. A band at 90 kDa was also recognized by TNFAIP6 antibody in pig matrix extract. The identity of this species is not known.
|
HC Transfer Activity in Porcine FF Is TNFAIP6 Dependent
Recent findings suggest that TNFAIP6 catalyzes the covalent linkage of HCs from II to HA [27]. Therefore, we investigated whether a factor with such capacity is present in the FF during hormone-stimulated OCC expansion. Follicular fluids collected either before or after hCG injection were incubated with high-molecular weight HA. Then, aliquots of each sample were either directly denaturated in SDS/β-mercaptoethanol loading buffer or digested with Streptomyces hyaluronidase followed by denaturation in SDS/β-mercaptoethanol loading buffer, and protein extracts were analyzed by Western blot with anti-II antibody (Fig. 2, A and B). In agreement with previous observations [16], serum-derived II family molecules (i.e., II [220 kDa], pre--trypsin inhibitor [130 kDa], and inter--like inhibitor [120 kDa]), were found in both FF. Free HCs (80-kDa molecular weight) were almost undetectable in FF from medium-size follicles, and the signal did not change when HA was digested with hyaluronidase (Fig. 2A). Conversely, a large increase in HC immunopositivity was observed after digestion of HA incubated with FF from hCG-stimulated follicles (Fig. 2B). This result suggests that HCs were displaced from II family molecules and covalently linked to HA by a factor present in the FF after hormone stimulation of the follicles. To determine whether TNFAIP6 is responsible for such activity, we incubated FF from hCG-stimulated follicles with HA either in the absence or in the presence of monoclonal antibody A38, previously shown to block TNFAIP6 binding to HA [29] and inhibit the formation of TNFAIP6-HC complexes [28]. The intensity of the 80-kDa band in hyaluronidase-treated samples was greatly reduced by the presence of A38 antibody (Fig. 2C). These results suggest that TNFAIP6 protein plays a role in coupling HC to HA in pig preovulatory follicles, similarly to what was observed in mouse.
|
TNFAIP6 Protein Is Expressed by Porcine OCCs Expanded In Vitro
To analyze whether TNFAIP6 protein was expressed during in vitro expansion of pig OCCs, compact OCCs were cultured with FSH and pig serum for 24 and 42 h, times corresponding to the middle and final stages of expansion process in vivo. At the end of culture, OCCs were digested with Streptomyces hyaluronidase, and protein extracts from cells and matrix were analyzed separately. Western blot analysis with anti-TNFAIP6 antibody (Fig. 3A) of matrix extracts from 24-h cultured OCCs showed intense immunopositive signals at 120 kDa and 35 kDa, corresponding to the HC-TNFAIP6 complex and the free TNFAIP6, respectively. At 42 h, both bands were much less intense, and other species were detected at 55 kDa and 65 kDa, perhaps representing degradation products of the TNFAIP6 complexed form. The same blot probed with II (Fig. 3B) showed that the amount of HCs (–80 kDa) incorporated into the matrix was the same at 24 and 42 h, thereby suggesting that transfer of HCs to HA did not increase after 24 h. All of the other bands of higher molecular weights likely represent clusters of HCs resistant to hyaluronidase digestion [14].
|
Effect of TNFAIP6 Monoclonal Antibody A38 on In Vitro Cumulus Expansion
It has been previously shown that the anti-TNFAIP6 antibody A38 impairs matrix assembly by forskolin-stimulated mouse OCCs, leading to their disaggregation [28]. Thus, we investigated the effect of this antibody on in vitro FSH-stimulated OCCs (Fig. 4A). In contrast to mouse, the A38 antibody produced no apparent difference in the expansion of porcine OCCs. Cumulus cells were embedded in a very abundant and elastic matrix, and OCCs were visibly expanded at 24 h either in the absence or in the presence of the A38 antibody. Similarly, OCCs were further expanded and cumulus cells more loosely associated at 42 h in both culture conditions. Accordingly, the amount of HA retained in the matrix was the same in OCCs cultured with or without the antibody, accounting in both cases to 90% and 50% of the total synthesized HA at 24 h and 42 h of culture, respectively (Fig. 4B).
|
To determine whether the monoclonal A38 antibody efficiently blocked the interaction of porcine TNFAIP6 with HA and II during culture, protein extracts from OCCs stimulated for 24 h with FSH plus serum in the presence of A38 antibody were analyzed by Western blot with anti-TNFAIP6 polyclonal antibody (Fig. 5A). The absence of any positive immunoreactive signal for TNFAIP6 in the matrix extracts clearly indicates that A38 antibody prevented TNFAIP6 retention and TNFAIP6-HC complex formation. To verify that covalent transfer of HC to HA synthesized by cumulus cells was also prevented, the membrane was stripped and probed with anti-II antibody. The results show that only a low amount of HCs was present in the matrix (Fig. 5B). A related rat monoclonal antibody Q75, shown to have no activity in inhibiting the interaction of mouse and human TNFAIP6, either with HA or II, had no effect on TNFAIP6 retention, TNFAIP6-HC complex formation, and HC accumulation in the porcine cumulus matrix (Fig. 5). In addition, not surprisingly, this antibody had no effect on cumulus expansion (data not shown).
|
Hyaluronan is a ubiquitous matrix component that is believed to form a variety of different matrix architectures, depending on its association with specific hyaluronan-binding proteins [24]. This polymer is the major structural element that determines the viscoelastic properties of the expanded matrix of mammalian OCCs. It has been well established that members of the II family circulating in the blood are pivotal for the formation of the mouse cumulus matrix, where covalent translocation of HCs from II onto HA has been demonstrated [10, 11]. Another essential protein for mouse cumulus matrix assembly is TNFAIP6, which can bind to HA and interact with II [24–27]. It has been shown for TNFAIP6 that, in addition to its free form (with an apparent molecular weight of 35 kDa), it is also present in the matrix of expanded mouse OCCs as a species of about 120-kDa molecular weight under denaturating and reducing conditions [19, 20]. Characterization of this 120-kDa band by protein sequence analysis and mass spectrometry demonstrated that it also contains HCs of II [20, 27]. This result, together with the observation that HCs were missing from the cumulus matrix in TNFAIP6-null mice, led to the conclusion that TNFAIP6 forms a complex with HC that facilitates its transfer to HA [26], a finding that has been confirmed by mixing purified molecules [27].
We have previously shown that HCs of II are covalently linked to HA during porcine in vivo cumulus expansion as well as in the matrix of in vitro-expanded OCCs when pig serum or pig FF was added to the medium as a source of II molecules [16].
The present study documents for the first time the presence of TNFAIP6 protein in porcine expanding OCCs and provides several lines of evidence that TNFAIP6 catalyzes the transfer of HC to HA in pig cumulus matrix as well. First, a 120-kDa TNFAIP6 species was revealed by Western blot in pig in vivo and in vitro OCC expanded matrix that was immunoreactive with anti-II antibodies, and was not susceptible to degradation by chondroitinase, as was found previously in mouse cumulus matrix [20]. Second, pig FF collected after hCG injection promotes the HC transfer to HA in a cell-free system, where this activity is abolished by the anti-human TNFAIP6 monoclonal antibody A38, previously shown to block binding of human TNFAIP6 to HA [29] and to inhibit the formation of human and mouse TNFAIP6-HC complexes [28]. Direct evidence for such a role for TNFAIP6 in pig OCCs was provided by stimulating pig OCCs with FSH and pig serum in the presence of the A38 antibody. Under these culture conditions, neither the TNFAIP6-HC species nor HCs were present in the matrix extract, indicating that A38 antibody successfully blocked TNFAIP6/II interaction and, consequently, HC linkage to HA. A related anti-TNFAIP6 monoclonal antibody (Q75), which recognizes a distinct epitope within the same TNFAIP6 domain but does not inhibit HC-TNFAIP6 complex formation [29], did not produce any effect. These results, together with the high sequence similarity found among human, rabbit, murine, rat, bovine, and equine TNFAIP6 [17, 18, 33] and the expression of this protein and/or the respective gene in preovulatory follicles of all of the examined species [17–19, 30, 33], strongly supports the concept that TNFAIP6-mediated HC covalent transfer onto HA synthesized by cumulus cells is a mechanism common to mammalian OCCs.
The importance played by HC-mediated HA cross-linking in the mouse cumulus matrix is supported by the evidence that mice with disrupted II production or deficient in TNFAIP6 are both unable to support the transfer of HC onto HA and exhibit impaired cumulus matrix assembly and female infertility [11, 26]. In agreement, anti-TNFAIP6 A38 antibody impairs matrix assembly by forskolin-stimulated mouse OCCs, leading to their disaggregation [28]. However, in contrast to the situation observed in the mouse, this antibody clearly did not prevent cumulus expansion or HA accumulation in the matrix of pig OCCs stimulated with FSH for 24 h. Since ovulation in the pig occurs around 40–44 h after hormone stimulation, we extended the analysis to 42 h of culture. Again, no significant differences in morphology and HA retention were detected between OCCs cultured with or without the blocking antibody.
The lack of immunoreactivity for the free TNFAIP6 (35 kDa) in the Western blot analysis of A38 antibody-treated OCCs indicates that this antibody also efficiently prevented the binding of TNFAIP6 to HA in the matrix, ruling out the possibility that the hypothesized interactions of TNFAIP6 with PTX3 [34] and thrombospondin [35], or any other matrix components, could compensate for the lack of HCs in cross-linking HA strands.
Overall, these results suggest that porcine OCCs have evolved supplemental mechanisms for cross-linking HA and incorporating it into the matrix. Actually, cumulus expansion in pig differs in several aspects from that in mouse, including the time interval required completing this process as well as the morphological and physical characteristics of the matrix. Specifically, expanding pig OCCs synthesize and deposit matrix for a much longer time than mouse OCCs (from 10 to 36 h after hormonal stimulation in pig vs. from 3 to 10 h in mice) and reach a much bigger volume, indicating that a greater amount of HA is synthesized and retained. Indeed, metabolic labeling of HA performed under the same stimulating conditions and with the same specific activity of the precursor in the culture medium suggests that pig OCC synthesizes about 20 times more HA than mouse OCC (40 000 and 2000 cpm for pig and mouse OCC, respectively), although the number of cumulus cells is only approximately five times higher in the pig than in the mouse (5000 in pig vs. 1000 in mouse; data not shown). In addition, the cumulus cells appear more scattered in fully expanded OCCs from pig compared to mouse, and although the matrix in both species shows elastic properties, that of pig stretches more extensively when pulled with a micropipette, suggesting some differences in HA aggregation. Finally, paracrine mechanisms regulating synthesis and deposition of matrix also appear to be different in the pig. In fact, in contrast to mouse cumulus cells, those in the pig do not require the presence of oocytes to undergo expansion [8, 9]. Recent findings strongly suggest that the induction of cumulus expansion by LH is mediated in the mouse preovulatory follicle by epidermal growth factor-like factors synthesized by mural granulosa cells [36]. In agreement, mouse OCCs treated in vitro with epidermal growth factor express genes required for matrix formation, such as Has2 and Tnfaip6, and undergo full expansion [37]. Instead, pig OCCs isolated from large preovulatory follicles (6–7 mm in diameter and eCG-primed follicles) synthesize HA and undergo expansion when stimulated in the same culture conditions for 24 h but rapidly dissociate thereafter, showing no sign of expansion at 32 h [38]. Thus, other growth factors must be involved in promoting the production of components essential for the cross-linking of HA in the pig extracellular matrix. A good candidate is insulin-like growth factor 1 (IGF1). We have recently shown that this growth factor significantly enhances retention of HA in porcine OCCs cultured in vitro with FSH in the absence of serum, via PIK3/AKT- and MAPK3/1-dependent pathways [39]. IGF1 is a potent inducer of proteoglycan synthesis in other tissues [40], and some types of proteoglycans (lecticans) can bind to HA, an interaction that is stabilized by link proteins. This leads in other tissue (e.g., cartilage) to the formation of large aggregates, which has a significant effect on the characteristics of the matrix. Proteoglycan synthesis is indeed induced during mouse [21, 31, 41] and pig [42] OCC expansion, and the most prominent type is the chondroitin sulfate proteoglycan versican, which is an important structural component in the organization of matrices in both physiological and pathological conditions. This proteoglycan can bind to HA through its N-terminal domain and to several matrix proteins via the C-terminal region [43]. In this regard, versican (a member of the lectican family) shows highly selective binding to the hexameric extracellular matrix glycoprotein tenascin C, leading to the possibility of an alternative mechanism for cross-linking hyaluronan through versican-tenascin-versican ionic interactions [44]. In this regard, tenascin C has been found in human and mouse cumulus matrix [45, 46]. Thus, it is reasonable to hypothesize that different relative levels of HA, TNFAIP6, versican, and tenascin, as well as other matrix components, are synthesized by cumulus cells of different species, leading to different mechanisms of matrix organization and stabilization.
In conclusion, data reported in the present paper suggest that the postulated role of HCs for cross-linking HA and stabilizing the cumulus matrix, although essential in mouse OCCs, might be redundant in porcine OCCs. Clearly, much more needs to be learned about the composition and structure of pig cumulus matrix to elucidate the mechanisms underlying its assembly and stability.
FOOTNOTES
1Supported by grant 305/050960 from Grant Agency of the Czech Republic; by grant 52/ZV2 from Ministry of School, Youth, and Education, Program of Scientific and Technological Co-operation between Czech Republic and Italy; and by grant PRIN 2005 from the Italian Ministry of Education. 
Correspondence: 2Eva Nagyova, Academy of Sciences of the Czech Republic, Institute of Animal Physiology and Genetics, Rumburska 89, 277 21 Libechov, Czech Republic. FAX: 420 315 639 510; e-mail: nagyova{at}iapg.cas.cz
Received: 14 August 2007.
First decision: 10 October 2007.
Accepted: 29 January 2008.
REFERENCES
I interaction promotes a transesterification cleaving the protein-glycosaminoglycan-protein (PGP) cross-link. J Biol Chem 2005; 280:11936–11942.
e
V, Nagyova E, Miyano T, Flechon JE. Developmental regulation of effect of epidermal growth factor on porcine oocyte-cumulus cell complexes: nuclear maturation, expansion, and F-actin remodeling. Mol Reprod Dev 2000; 56:63–73.[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
