BIOLOGY OF REPRODUCTION (2015) 92(5):116, 1–10 Published online before print 25 March 2015. DOI 10.1095/biolreprod.114.127274

Activation of Mouse Cumulus-Oocyte Complex Maturation In Vitro Through EGF-Like Activity of Versican1 Kylie R. Dunning,2 Laura N. Watson, Voueleng J. Zhang, Hannah M. Brown, Adrian K. Kaczmarek, Rebecca L. Robker, and Darryl L. Russell The Robinson Research Institute, School of Paediatrics and Reproductive Health, The University of Adelaide, Adelaide, South Australia, Australia

In vitro maturation of oocytes is suboptimal to in vivo maturation with altered gene expression and compromised oocyte quality. The large proteoglycan versican is abundant in mouse cumulus-oocyte complexes (COCs) matured in vivo but is absent in cultured COCs. Versican is also positively correlated with human oocyte quality. Versican contains an epidermal growth factor (EGF) motif, and based on EGF-like activities in other systems we hypothesized that versican acts as an EGF-like signaling factor during COC maturation. Here, we purified recombinant versican and compared its function with that of EGF during in vitro maturation (IVM). Versican significantly increased cumulus expansion and induced cumulus-specific genes Ptgs2, Tnfaip6, and Has2, which was blocked by EGF receptor antagonist. Microarray analysis revealed that versican has overlapping function with EGF; however, a subset of genes was uniquely altered following 6 h of IVM with either treatment. Following 6 h of IVM, both Areg and Ereg were significantly increased by both treatments, whereas Egln3, Nr4a1, Nr4a2, Nr4a3, and Adamts5 were significantly higher following versican treatment compared with EGF. In contrast, Sprr1a and Aqp3 were increased after 6 h of EGF but not versican treatment. To determine whether there were temporal differences, COCs were cultured with EGF or versican for 0–12 h. Versican-induced expression occurred later but remained elevated for longer compared with EGF for Ptgs2, Ereg, and Nr4a3. The unique expression profiles of Aqp3 and Nr4a3 during IVM were similarly regulated in vivo. These data indicate that versican has EGF-like effects on COC gene expression, but with distinct temporal characteristics. cumulus cell mucification, cumulus cells, cumulus expansion, EGF-like, oocyte maturation, versican

INTRODUCTION Mammalian oocyte growth depends on the functions of ovarian follicle and cumulus cells, which provide a specialized environment to maintain oocytes in meiotic arrest and provide appropriate nutritional and developmental signals for the growing oocyte [1]. Activation of meiotic maturation occurs only in mature preovulatory follicles which respond to the ovulation-inducing surge of luteinizing hormone (LH) [1]. The ovarian granulosa and cumulus cells produce factors that 1 Supported by grants from the Australian Research Council and the National Health and Medical Research Council. 2 Correspondence: E-mail: [email protected]

Received: 3 December 2014. First decision: 23 December 2014. Accepted: 16 March 2015. Ó 2015 by the Society for the Study of Reproduction, Inc. eISSN: 1529-7268 http://www.biolreprod.org ISSN: 0006-3363

1

Article 116

Downloaded from www.biolreprod.org.

promote meiotic arrest during oocyte growth in the preovulatory follicle, and then the LH surge induces a periovulatory environment in which oocyte maturation, cumulus matrix expansion, and ovulation are coordinately activated. The LHresponsive granulosa cells produce a range of factors that impinge on cumulus cells, including prostaglandin E2 [2–4] and C-type natriuretic peptides [5, 6]. Of particular importance are the epidermal growth factor (EGF)-like family of ligands Epiregulin (EREG), Amphiregulin (AREG), and betacellulin (BTC), which are essential to activating maturation of the cumulus-oocyte complex (COC) [7]. Thus, the granulosa cells play a key role, transmitting the simple LH surge through a more complex system of intrafollicular signals that may be important to amplify and add contextual control to the cumulus response. As a result, the COC governs a range of physiological processes, including oocyte maturation [8] and cumulus expansion [9], as well as temporally controlled adhesive and invasive capacity [10] and ovulation [1]. Each of the EGF-like ligands binds EGF receptors (EGFRs) on cumulus cells and activates the canonical EGF signaling pathway, leading to the activation of the MAPK1/3 (ERK1/2) kinases and the induction of a large network of genes in cumulus cells, including those involved in cumulus matrix assembly; hyaluronan synthase (Has2), tumor necrosis alphainduced protein 6 (Tnfaip6), pentraxin 3 (Ptx3) and others, and prostaglandin generation; prostaglandin-endoperoxide synthase 2 (Ptgs2) [11]. Subsequently, the EGF-like genes themselves in cumulus cells are considered important to sustain an autocrine signaling loop and the progress of oocyte maturation and ovulation of the COC [12, 13]. The same gene expression and functional changes can also be induced in isolated COCs in vitro by treatment with EGF [14, 15], any of the EGF-like ligands [7], or follicle-stimulating hormone (FSH) [16]. However, the temporal pattern of response is altered in vitro, including abbreviated gene induction in cumulus cells, loss of synchronization with oocyte meiotic progression, and consequently the developmental potential of oocytes matured in vitro is compromised [17–19]. A range of studies have investigated whether one or a combination of the EGF-like family of factors better recapitulate in vivo COC response patterns [7, 20–22]. Overwhelmingly, the conclusion has been that the gene expression response is truncated in COCs treated in vitro [18–21, 23–25] and that this aberrant response contributes to poor oocyte competence in vitro. Versican is a complex matricellular proteoglycan that is also rapidly induced in granulosa cells by the LH surge [19, 26]. Versican protein has a complex structure, including a highaffinity hyaluronan binding N-terminal domain, a densely chondroitin sulfate substituted midsection, and cell surface receptor-interacting functions, including reported abilities to activate toll-like receptors [27–29] and the EGFR [30–32]. Versican is widely expressed and mainly involved in tissue

ABSTRACT

DUNNING ET AL. eluted from Q-sepharose on a Poly-prep chromatography column (Bio-Rad Laboratories) using 2 M NaCl in PBS (pH 7.4), and then purified by gel chromatography on a Sephacryl S400 (15 3 650 mm column; GE Life Sciences) equilibrated with PBS (pH 7.4) at 48C. Versican-containing fractions were determined by dot blotting. Briefly, samples were applied to a nitrocellulose membrane (Amersham GE Healthcare Life Sciences), blocked with Odyssey Blocking Buffer (1:1 with PBS; Licor Bioscience), and detected using 12C5 mouse monoclonal anti-versican primary antibody (1:500; Developmental Studies Hybridoma Bank, University of Iowa) and IRDye 680RD goat anti-mouse secondary antibody (1:20 000; diluted; Licor Bioscience); then, they were scanned using the Odyssey infrared imaging system with processing in Image Studio (Licor Bioscience). Fractions containing the most versican were pooled, dialyzed against double-distilled milli-Q water, and concentrated by lyophilization in the presence of 25 mM sucrose, which was used as a stabilizer [42]. The molecular integrity and purity of the recombinant versican were determined by Western blot and silver stain (see Supplemental Methods; Supplemental Data are available online at www. biolreprod.org) analyses.

morphogenic processes, such as heart valve and limb morphogenesis [33–35], but it also promotes growth and metastasis of a range of carcinoma types [36]. In the mouse ovary, versican is expressed throughout folliculogenesis and localized to the extracellular matrix (ECM) surrounding growing follicles where pronounced structural remodeling is required [26]. In response to the LH surge, granulosa cell expression of versican increases 20- to 50-fold and the protein is secreted and localizes to the COC because of its strong hyaluronan-binding affinity [19, 26, 37]. It is clear that versican is an abundant hyaluronan cross-linking protein in the expanded cumulus matrix, which forms a complex ECM structure through protein-hyaluronan interactions as well as protein-protein binding of pentraxin 3 (PTX3) and tumor necrosis factor alpha-induced protein 6 (TNFAIP6) [9]. Each of these products is absent or reduced in cultured COCs [18, 19, 25], and as a result the in vitro COC matrix is functionally altered. We demonstrated this by showing a loss of capacity to control diffusion of small molecules and signaling factors in COCs matured in vitro [38]. Importantly, in at least three independent studies the level of versican expression in human COCs has been found to correlate with the developmental potential of the oocyte [39–41]. This supports a key role in oocyte maturation, but no specific function of versican in oocyte maturation has been described. In mouse ovarian follicles versican is almost exclusively produced by granulosa and not cumulus cells, and is completely absent in COCs stimulated with EGF, FSH, or a combination of the two during in vitro maturation (IVM) [19]. The absence of versican during oocyte maturation in vitro means that its function is lost and may account for the poor efficacy of IVM. In this study we have synthesized and purified recombinant full-length versican and investigated its function in the COC during the maturation process in vitro. Given the structural homology to EGF ligands and the evidence for EGF-like activity of versican in various cell lines, we hypothesized that versican may promote EGF-like signal in cumulus cells. Further, because of its unique structure and ability to bind hyaluronan, this may be an important component of the periovulatory signaling cascade between granulosa cells and COCs.

Isolation and Culture of Mouse COCs

Gene Expression Analysis Twenty COCs were pooled for gene expression analysis, and total RNA was isolated from frozen samples and reverse transcribed into cDNA as described previously [44]. Briefly, RNA was isolated using Trizol (Invitrogen, Life Technologies) as per the manufacturer’s instructions, with the inclusion of 7.5 lg of glycoblue (Ambion, Life Technologies) during precipitation overnight at 208C. Total RNA was then treated with 1 U of DNase (Ambion) as per the manufacturer’s instructions, and RNA concentration and purity were quantified using a Nanodrop ND-1000 spectrophotometer (Biolab Ltd.). Firststrand cDNA was synthesized from total RNA using random hexamer primers (Geneworks) and Superscript III reverse transcriptase (Invitrogen, Life Technologies). Real-time PCR reaction controls included omission of the cDNA template or reverse transcription enzyme in otherwise complete reactions. Gene expression was normalized to Rpl19. All values were then expressed relative to either the no treatment control or the 0 h time point using the 2(DDCT) [45]. Taqman gene expression assays were purchased from Applied Biosystems (Supplemental Table S1), and reactions were run in duplicate on an AB7900HT Fast PCR System using the manufacturer’s recommended amplification settings. Each reaction used 0.5 ll of taqman assay, 5 ll of taqman master mix, 5 ng of cDNA, and H2O to a final volume of 10 ll.

MATERIALS AND METHODS Materials Equine chorionic gonadotropin (eCG), human chorionic gonadotropin (hCG), and FSH were purchased from the National Hormone and Peptide Program. Culture medium was purchased from Gibco, Invitrogen Australia Pty. Ltd. All other reagents were purchased from Sigma-Aldrich Pty. Ltd. unless otherwise stated.

Animals All mice were purchased from Laboratory Animal Services and were maintained on a 12L:12D photoperiod with rodent chow and water provided ad libitum. All experiments were approved by the University of Adelaide’s ethics committee and were conducted in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes.

Production and Purification of Versican

Comparison of Global Gene Expression Changes in EGFTreated Versus Versican-Treated COCs

Full-length recombinant human versican was purified from the conditioned media of Chinese Hamster Ovary cells stably transfected to overexpress V1 versican, kindly provided by Prof. R. LeBaron (University of Texas at San Antonio). Purification involved using a combination of anion exchange and gel filtration chromatography. Briefly, conditioned media was batch-absorbed to Q-sepharose (10 ml per 100 ml of conditioned media; GE Healthcare Life Sciences) at 48C for 30 min in the presence of protease inhibitors (cOmplete protease inhibitor cocktail tablets; Roche Applied Science). Versican was

After 6 h of IVM culture, 80 COCs per treatment were pooled, and RNA was extracted and DNase treated as described above. RNA integrity was verified and quantitated on an Aligent bioanalyzer with RNA picochips (Aligent Technologies). Samples from four to five independent replicate experiments were labeled and hybridized to Affymetrix mouse Gene 1.0

2

Article 116

Downloaded from www.biolreprod.org.

Prepubertal (Days 21–23 of age) CBA 3 C57BL/6 first filial generation (F1) (CBAF1) female mice were treated with intraperitoneal (i.p.) administration of eCG (5 IU), followed by hCG (5 IU i.p.) 44 h later. Ovaries or oviducts were dissected from mice and placed in HEPES-buffered minimum essential medium alpha (aMEM) supplemented with gentamycin (50 lg/ml) and 3 mg/ ml fatty acid-free bovine serum albumin (BSA). The COCs were collected 44 h after eCG (immature COCs for IVM), or at 4, 8, or 12 h (preovulatory) or 16 h (postovulatory) after hCG. Cumulus-oocyte complexes were collected by puncture of large antral follicles. Ovulated COCs (16 h after hCG) were isolated by puncture of the ampulla of the oviduct. For IVM, COCs were cultured in drops (10 COCs per 100-ll drop) of bicarbonate buffered aMEM supplemented with gentamycin (50 lg/ml), 3 mg/ml fatty acid-free BSA, and 5 mIU/ml FSH overlaid with sterile mineral oil for up to 16 h at 378C in 20% CO2, 80% air. As outlined throughout the manuscript, some IVM treatments were supplemented with EGF (3 ng/ml), versican (12.4 or 24.8 ll per 100-ll culture drop), protein tyrosine kinase inhibitor AG1478 (5 lM), and sucrose (31 or 62 mM, to equate for the concentration of sucrose present in versican treatments). For experiments assessing cumulus expansion, fetal calf serum (5%) was used in place of BSA and versican was used at 15, 30, or 60 ll per 100 ll. The degree of cumulus expansion was determined at 16 h of culture by a blinded assessor using the standard scale previously described [43]. For each treatment a mean cumulus expansion index was calculated. Cells required for RNA analysis were snap frozen in liquid nitrogen and stored at 808C until use.

EGF-LIKE ACTIONS OF VERSICAN IN COCs

Microarrays and were washed and scanned according to the manufacturer’s instructions at the University of Adelaide Microarray Centre. Statistical analysis of the differential gene expression was compared among the three groups using the Partek Genomics Suite, version 6.4 (Partek Inc.). An ANOVA was applied with post hoc adjustment for false discovery due to multiple measures [46]. Genes accepted as significantly altered were those with q , 0.01 after FDR adjustment and 2.0-fold change in expression level.

compared with the untreated, cultured control (Fig. 1). Interestingly, at the highest dose of versican, a visible translucent pericellular coating of ECM was observed surrounding the COC and individual cumulus cells (Fig. 1, B and C), which was not observed in other treatments. Versican Induces Cumulus-Specific Gene Expression via an EGFR Signaling Pathway

Pathway Analysis Data from the microarrays were analyzed through the use of Qiagen’s Ingenuity Pathway Analysis (IPA; Qiagen) to identify pathways associated with significantly altered transcripts.

We next determined whether versican induced genes associated with cumulus expansion and compared this with EGF-induced expression. Both versican and EGF significantly induced Ptgs2, Tnfaip6, and Has2 expression in COCs compared with the untreated, cultured control (Fig. 2, A and B), with induction of Tnfaip6 at the highest dose of versican being nearly 37-fold higher than the untreated control (Fig. 2B). Similarly versican, at the top dose, resulted in a significant 13-fold induction of Has2 compared with the untreated, cultured control (Fig. 2C). Versican-induced expression levels of Ptgs2 and Tnfaip6 were significantly reduced in the presence of EGF receptor antagonist AG1478.

Statistical Analyses All results are presented as mean 6 SEM of three or more independent replicate experiments. Statistical analyses were performed as indicated in figure legends using IBM SPSS Statistics version 21. All data were assessed for equal variances and normality and were transformed where necessary, as indicated in the figure legends. One-way ANOVA and two-way ANOVA were used as described in the figure legends, and statistical significance was considered at a P value of ,0.05. Data comparing the effect of versican and EGF on gene expression in COCs over time were analyzed using a univariate analysis under the generalized linear model, with treatment and time as fixed factors (samples were independent and not repeated measures). Post hoc paired comparisons were made when appropriate to do so, using the Tukey, Dunn, and Dunnett T3 correction.

Global Pattern of Versican-Induced Gene Expression in COCs Is Similar to, but Distinct from, EGF Treatment To further characterize the effects of versican on COCs and whether it is similar to EGF signaling, we performed a microarray analysis of global gene expression changes in COCs cultured in the absence or presence of versican or EGF for 6 h. The most highly differentially expressed genes comparing EGF- or versican-treated COCs, as well as the relative induction of each gene by EGF or versican compared with untreated, cultured controls, are shown in Table 1. Principal component analysis showed that versican- and EGFtreated COCs were similar in their global gene expression changes compared with cultured, untreated controls (Fig. 3A).

RESULTS Versican Induces Expansion in COCs During IVM The purified recombinant versican appeared as a single band of ;350 kDa, the expected size of the V1 isoform protein, and a single immunoreactive band of the same size was identified by Western blot analysis (Supplemental Fig. S1). We assessed the ability of recombinant versican to induce cumulus expansion in vitro. In a dose-dependent manner, versican significantly increased the degree of cumulus expansion 3

Article 116

Downloaded from www.biolreprod.org.

FIG. 1. Versican induces cumulus expansion during IVM. Cumulus-oocyte complexes were matured in vitro for 16 h without or with three concentrations of versican. A–C) Representative images of untreated, cultured control (A, –) and versican-treated (B and C) COCs. Arrows indicate visible ECM surrounding cumulus cells. Black bar ¼ 200 lm; white bar ¼ 50 lm. Cumulus expansion index after versican treatment of 0, 15, 30, or 60 ll per drop was scored by a trained observer blinded to the treatments and was presented as mean 6 SEM (D, n ¼ 30–59 COCs from three independent experiments). Because of nonnormal data distribution a nonparametric one-way ANOVA (Kruskal-Wallis test) with Dunn multiple comparison test was used. Different superscripts signify statistical difference of P , 0.0001.

DUNNING ET AL.

Pairwise comparisons against control samples showed that 673 genes were similarly regulated by both versican and EGF (Fig. 3B), demonstrating conserved or overlapping functions. Interestingly, versican significantly altered 317 transcripts not regulated by EGF (Fig. 3B), indicating that important differences exist between the effects of versican and EGF treatments. Similarly, 319 transcripts were significantly regulated by EGF and not versican (Fig. 3B). To further examine the differences and similarities between EGF and versican signaling in the COC, Figure 3C depicts the number of transcripts similarly or differentially regulated, either up or down, by the two treatments. This further demonstrates that after 6 h of treatment versican shows similarity to EGF signaling but also differentially regulates 299 genes compared with EGF. Biological pathways altered by versican and EGF treatment of COCs were identified by IPA software. The top 10 pathways significantly affected by versican and EGF are shown in Figure 3D. ‘‘Axonal guidance signaling’’ was the pathway most impacted by EGF, and ‘‘cholecystokinin/gastrin mediated signaling’’ by versican (Fig. 3D).

Validation of microarray data was performed on selected representative genes by quantitative real-time RT-PCR of COCs again matured in vitro for 6 h in the presence or absence of versican and EGF (Fig. 4). For all genes tested, differences between treatments were similar in both microarray and quantitative PCR results, thus validating the microarray data. The genes measured by quantitative PCR fell into three groups: those induced at 6 h by both EGF and versican (Fig. 4, A and B), those induced by versican and not EGF (Fig. 4, C–G), and those induced by EGF but not versican (Fig. 4, H and I). Both EGF and versican significantly increased Areg and Ereg compared with the untreated, cultured control, with versicaninduced expression significantly inhibited and reduced to control levels in the presence of EGFR antagonist AG1478 (Fig. 4, A and B). Egln3, Nr4a1, Nr4a2, Nr4a3, and Adamts5 were significantly increased in COCs cultured for 6 h in the presence of versican and not EGF (Fig. 4, C–G), whereas after treatment with versican and AG1478 no significant increase above controls was evident, indicating that the induction was EGFR dependent. EGF (3 ng/ml, 31 mM sucrose) induced expression of Sprr1a and Aqp3 in COCs by 50- and 18-fold, respectively, compared with the untreated, cultured control (Fig. 4, H and I). These data confirm the finding from microarray that versican and EGF activate similar signaling pathways but also have separate and distinct consequences at the 6-h time point. Different Temporal and Amplitude Patterns of Gene Expression in COCs Treated with Versican Compared with EGF

FIG. 2. Induction of cumulus-specific gene expression by versican and requirement for EGF receptor kinase activity. A–C) Cumulus-oocyte complexes were matured in vitro for 6 h with EGF (3 ng/ml) or versican (see Materials and Methods). Some COCs treated with the highest versican dose were also cotreated with the EGF receptor antagonist AG1478 (5 lM). Analysis of genes induced during cumulus-oocyte maturation, Ptgs2, Tnfaip6, and Has2 were normalized to Rpl19 and presented as mean 6 SEM fold change (n ¼ 4 experimental replicates). Data in A were log10 transformed to normalize and were analyzed by one-way ANOVA with Tukey post hoc test. Data in B were assessed by nonparametric one-way ANOVA (Kruskal-Wallis test) with Dunn post hoc test because of nonnormal distribution of data. Data in C were assessed by one-way ANOVA with Dunnett T3 post hoc because of unequal variance. Different superscripts signify statistical difference of P , 0.05. Sucrose was included during lyophilization of recombinant versican (as a stabilizer), thus equivalent concentrations of sucrose were used in control treatments to match each dose of versican (31 or 62 lM).

To further examine the differences in versican and EGF signaling in the COC, we measured temporal induction of Ptgs2, Tnfaip6, Areg, Ereg, Nr4a3, and Aqp3 in COCs cultured in the presence of EGF or versican during the course of 0–12 h. Treatment with EGF significantly increased the expression of Ptgs2, Tnfaip6, Areg, and Nr4a3 (Fig. 5) after 2 h. Similarly, versican treatment resulted in a significant increase in Nr4a3 following 2 h of treatment (Fig. 5E), whereas Ptgs2, Tnfaip6, and Areg were significantly increased following 4 h of versican treatment. There were no consistent differences in the 4

Article 116

Downloaded from www.biolreprod.org.

Validation of Gene Expression Differences Observed in Microarray by Quantitative PCR

EGF-LIKE ACTIONS OF VERSICAN IN COCs TABLE 1. Top 20 genes more highly expressed in 6-h EGF-treated or in 6-h versican-treated COCs. Gene symbola

Fold difference (EGF vs. versican)

q value

EGF vs. control, fold changeb

Versican vs. control, fold changeb

NM_016689 NM_009264 X00438 NM_011428 NM_031170 NM_008737 NM_010664 NM_013609 NM_026770 NM_023256 NM_011909 NM_010608 NM_177161 NM_010253 NM_007899 NM_001039676 NM_011817 NM_030127 NM_001243008 NM_018865

3.57E-05 2.28E-04 1.09E-05 3.57E-05 4.06E-04 1.16E-03 2.28E-04 1.01E-02 1.34E-03 1.46E-04 3.39E-03 1.28E-03 2.07E-03 3.25E-04 1.34E-03 2.51E-03 5.38E-03 1.73E-04 1.73E-04 5.71E-04

5.3 5.3 4.4 2.9 2.9 2.9 2.9 2.7 2.6 2.5 2.5 2.5 2.4 2.4 2.3 2.3 2.2 2.2 2.2 2.2

6.63*** 9.14*** 8.77*** 4.1*** 3.98*** 11.31*** 4.56*** 14.44*** 3.75*** 2.85*** 2.02** 2.16** 1.04NS 4.59*** 3.43*** 2.38** 6.44*** 3.29*** 3.20*** 4.06***

1.26NS 1.74NS 1.29NS 1.42** 1.39NS 3.96*** 1.63** 5.28*** 1.47NS 1.14NS 1.2NS 1.14NS 2.34** 1.89** 1.48NS 1.06NS 2.87** 1.49** 1.46** 1.88**

ENSMUST00000030025 NM_028133 NM_153171 NR_028478 NM_172523 NM_011782 NM_007810 NM_001001309 NM_018803 NM_021432 NM_001081287 NM_010050 NM_009889 NM_007913 NM_007950 NM_010733 NM_001160378 NM_027052 NM_010570 NM_011723

2.07E-04 1.37E-03 4.57E-03 6.06E-03 2.55E-03 3.80E-03 7.33E-03 4.57E-03 2.07E-03 2.98E-04 4.71E-03 1.09E-05 8.81E-04 8.81E-04 3.80E-03 1.73E-04 7.32E-03 2.04E-03 8.20E-03 3.40E-03

4.5 3.2 3.1 3.0 2.7 2.7 2.6 2.5 2.4 2.4 2.4 2.4 2.4 2.4 2.3 2.3 2.3 2.2 2.2 2.1

2.76** 4.76*** 1.29NS 1.4NS 3.29*** 1.79NS 2.02* 4.42*** 2.30** 3.66*** 1.10NS 1.50** 1.03NS 2.62*** 12.75*** 1.25NS 2.79** 3.25*** 1.08NS 1.19NS

12.37*** 15.22*** 2.4** 4.24** 1.2NS 4.78*** 1.27NS 1.79* 1.06NS 1.51** 2.17** 1.02NS 1.47* 6.17*** 29.84*** 1.73** 1.21NS 1.47NS 1.18NS 1.79**

a

Bold text indicates genes that were confirmed by quantitative PCR. NS, not significant. * P , 0.01, ** P , 0.001, *** P , 0.001 compared with no treatment control. b

h in vivo and remained high until the time of ovulation (Fig. 6B), which was more similar to the sustained expression pattern observed in response to versican in vitro.

amplitude of the peak induction, but consistent temporal differences with a later peak of expression and longer duration of significant induction over the untreated controls were observed (Fig. 5). The peak of gene expression by versican was delayed by 2 h compared with EGF, with a peak observed at 4 h of culture for all genes with the exception of Ereg, which peaked at 8 h (Fig. 5).

DISCUSSION Although the strong induction of Vcan mRNA in mural granulosa cells of ovulating follicles and the translocation of the secreted protein to the COC has been known for more than 10 yr [26, 47], we now show a functional role of versican in COCs by demonstrating that versican can activate COC expansion in vitro and the expression of genes known to be induced by EGF-like factors and involved in the maturation of the COC required for ovulation [1]. A detailed comparison of the global transcriptional changes with 6 h of EGF versus versican treatment in vitro illustrated a strong overlap in responses, with two thirds of the regulated transcripts showing a common response. But key differences were evident, with the remaining one third of versican and EGF genes uniquely regulated by either only versican or only EGF, indicating a unique impact on cumulus cells by each of these signals. This

Similar Temporal and Amplitude Patterns of Gene Expression Observed In Vivo for Aqp3 and Nr4a3 Most of the genes we identified as EGF and/or versican induced are widely reported to be induced during maturation in vivo and/or in vitro. To confirm the in vivo patterns of gene expression for genes not previously described during COC maturation and which appeared differently regulated by EGF or versican, we measured the temporal induction of Aqp3 and Nr4a3 in COCs following hCG administration in vivo. Aqp3 was significantly induced at 4 h, then rapidly declined (Fig. 6A), which is similar to the general EGF-induced expression patterns in vitro (Fig. 5F). Nr4a3 was significantly induced at 4 5

Article 116

Downloaded from www.biolreprod.org.

Higher in EGF treated Aqp3 Sprr1a Tcrb-J Snap25 Krt8 Nrp1 Krt18 Ngf Cgref1 Krt20 Usp18 Kcnk3 P4ha3 Gal Ecm1 Slc39a2 Gadd45g Htra3 Col6a3 Wisp1 Higher in versican treated Nr4a3 Egln3 Rgs13 Snora75 Slc18a2 Adamts5 Cyp19a1 Itga8 Syt10 Nap1l5 Mpp7 Dio2 Cga Egr1 Ereg Lrrn3 Fam46a Slc38a4 Irs1 Xdh

RefSeq

DUNNING ET AL.

Downloaded from www.biolreprod.org. FIG. 3. Global pattern of versican-induced gene expression in COCs is overlapping, but distinct from, EGF treatment. Cumulus-oocyte complexes were matured in vitro for 6 h with EGF (3 ng/ml) or versican, and extracted mRNA was subjected to microarray analysis using Affymetrix mouse gene 1.0 genechips. Principal component analysis (A) shows the relationship of global gene expression in control samples (green circles) compared with those treated with versican (red) or EGF (blue circles). The Venn diagram (B) illustrates the number of transcripts with significantly altered gene expression in pairwise comparisons against control. Overlapping areas indicate the number of genes that were commonly regulated by both treatments; numbers in parentheses indicate the percentage of total significantly regulated transcripts in each set. Genes with significantly altered expression were defined as

6

Article 116

EGF-LIKE ACTIONS OF VERSICAN IN COCs

cell lines, where it promotes proliferation [30, 32, 48]. Our present results are among the first to show EGF-like signaling activity of versican in primary cells and a biological system where it is acutely regulated in vivo [26]. Notably, versican is lacking in IVM COCs [19], and hence our supplementation of IVM culture better recapitulates the in vivo COC composition. Certainly, the IVM cumulus matrix was more substantial following IVM in the presence of versican, and could even be visualized surrounding the complexes under transilluminated light microscopy. Whether the versican-supplemented IVM COC matrix has physiological characteristics more consistent with in vivo COCs, such as solute filtration [38], requires future investigation.

was further confirmed by a temporal comparison of versican and EGF treatment in vitro, with the maximal change mediated by versican treatment being later, but also more sustained than the profile of response to EGF. These results indicate that versican may contribute to the EGF-like signals emanating from granulosa cells to induce COC maturation in the periovulatory follicle, and may be partially responsible for the more sustained gene expression change that occurs in vivo and which has been proposed to be important for high-quality oocyte maturation. Versican is a complex molecule that binds hyaluronan through its N-terminal domain and has two EGF-like Cterminal domains previously shown to activate EGF receptor signaling in other cellular systems, most commonly in cancer 3

those with q , 0.01 after adjustment for multiple testing and 2.0-fold change in expression level (n ¼ 5 independent replicate experiments). C) Categories of genes that were both up-regulated by versican and EGF (category 1), both down-regulated (category 2), up-regulated by EGF but not versican (category 3), up-regulated by versican but not EGF (category 4), down-regulated by EGF but not versican (category 5), and down-regulated by versican but not EGF (category 6). Numbers indicate the number of genes that were significantly altered. D) The top 10 pathways significantly altered by either versican or EGF compared with control as identified by IPA. Yellow line indicates threshold of significance (P ¼ 0.05).

7

Article 116

Downloaded from www.biolreprod.org.

FIG. 4. Validation of gene expression differences observed in microarray by quantitative PCR; versican effect overlaps with, but is also distinct from, EGF effects on COCs matured in vitro. Confirmation of genes induced by both EGF and versican, Areg and Ereg; or induced by versican and not EGF, Egln3, Nr4a1, Nr4a2, Nr4a3, and Adamts5; or induced by EGF and not versican, Sprr1a and Aqp3, in COCs following 6 h of IVM. Messenger RNA expression was normalized to Rpl19 and presented as mean 6 SEM fold change (n ¼ 4 experimental replicates). Data in A, B, D, and E were log 10 transformed to normalize data. A one-way ANOVA was used to assess data in A, B, D, E, and G, with a Dunnett test applied to A, B, and D and a Tukey post hoc test to E and G because of unequal and equal variances, respectively. Data in C, F, H, and I were not normally distributed, thus a nonparametric one-way ANOVA (Kruskal-Wallis test) with Dunn post hoc test was applied. Different superscripts signify statistical difference of P , 0.05. Sucrose was included at equivalent concentrations to versican, as in Figure 2.

DUNNING ET AL.

The cumulus cells play an integral role in response to the LH surge by undergoing cumulus matrix expansion, which is critical for ovulation [1] and mediating meiotic resumption in the oocyte (reviewed in Conti et al. [8]). An appropriate temporal response in the cumulus cells is critical to the acquisition of developmental potential in the oocyte, and it has been widely demonstrated that the temporal pattern of gene expression is suboptimal in COCs matured in vitro [18–21, 23– 25]. A number of recent studies have investigated the relative impact on oocyte quality of stimulating IVM with the different EGF-like factors. A suggestion that epiregulin produces a stronger activation of some metabolic surrogate markers after

12 or 17 h of treatment was recently reported [21], but this showed only a trend toward an increase compared with EGF. In the current study, recombinant versican maintained the expression of several key oocyte competence genes above control (0 h treatment) for 4–6 h longer than EGF in vitro and induced a higher overall level of induction more than 12 h after treatment. Further, regulation of Egln3, Itga8, Nap1l5, Lrrn3, and Ereg by versican in vitro occurs similarly in vivo (raw data in Hernandez-Gonzalez et al. [49]), pointing to the importance of versican for inducing an in vivo-like expression profile during IVM. Importantly, Ereg mRNA showed a very pronounced extension in induction by versican treatment in

FIG. 6. Similar temporal and amplitude patterns of gene expression observed in vivo for Aqp3 and Nr4a3. Cumulus-oocyte complexes were isolated from ovaries (0–12 h) or oviducts (16 h) at 0, 4, 8, 12, or 16 h following hCG administration (see Materials and Methods). Aqp3 and Nr4a3 (A and B, respectively) expression in COCs was normalized to Rpl19 and presented as mean 6 SEM fold change (n ¼ 3 experimental replicates, asterisk indicates statistical difference of P , 0.05 between treatments within a time point). Data were analyzed using a one-way ANOVA with Tukey post hoc test following log 10 transformation.

8

Article 116

Downloaded from www.biolreprod.org.

FIG. 5. Different temporal and amplitude patterns of gene expression in COCs treated with versican compared with EGF. Cumulus-oocyte complexes were matured in vitro for 0, 2, 4, 6, 8, 10, or 12 h with EGF (3 ng/ml) or the high dose of versican (see Materials and Methods). Both treatments contained 62 mM sucrose. Ptgs2, Tnfaip6, Areg, Ereg, Nr4a3, and Aqp3 (A–F, respectively) expression in COCs was normalized to Rpl19 and presented as mean 6 SEM fold change (n ¼ 3 experimental replicates). Data were analyzed using a two-way ANOVA with Sidak multiple comparisons test. Asterisks indicate statistical difference from time 0 h within a treatment (P , 0.05).

EGF-LIKE ACTIONS OF VERSICAN IN COCs

component of the expanded cumulus matrix and important to the overall in vivo context in which oocytes mature.

COCs, and the resulting sustained elevation in Ereg may be expected to have further beneficial influence on the temporal change occurring in the COC and is more consistent with the in vivo pattern of Ereg expression in COCs [12]. In considering the actions of EGF-like factors it should also be acknowledged that each is induced at slightly different times in the granulosa cells after the LH surge in vivo [7, 12]. Versican is among the last factors with EGF-like activity to be induced, peaking at around 4 h after hCG treatment in vivo [26]. This further suggests that in addition to incorporating into the COC matrix through hyaluronan binding, versican may have the physiological role of sustaining the EGF receptor signal. A range of evidence demonstrates that different EGF-like ligands have different consequent actions after binding their cognate receptors. In the MCF10A breast cancer cell line, amphiregulin stimulates cell motility and proliferation, whereas EGF has little effect [50, 51]. This is thought to be elicited partly through variations in the conformational change caused by ligand-receptor binding and partly by variations in interactions mediated by different ligands between the EGF receptor and other cell surface molecules, including integrins, followed by internalization and recycling of receptors to the plasma membrane (reviewed in Wilson et al. [52]). Consistent with this, the heparan sulfate proteoglycan-binding EGF-like peptides were shown to be the only forms that promote proliferation in myeloma cells [53]. Versican’s complex and chondroitin sulfate-rich structure is atypical among EGF signaling factors and could be expected to have a unique effect on receptor conformation, but as a very large matricellular proteoglycan that interacts with ECM, it also may assemble a unique macromolecular complex at the cell surface to promote unique intracellular signaling. A distinct temporal response induced by versican was responsible for most of the differential expression observed by microarray analysis at the 6-h time point, but minor sets of unique genes specifically regulated by either versican or EGF also exist, as illustrated by the example of Aqp3. Interestingly, among the genes highly induced by EGF but unaffected by versican at 6 h were five encoding ECM proteins most closely associated with rigid filamentous epithelia (keratins, Krt8, Krt18, and Krt20; small proline rich protein 1a, Sprr1a; and Extracellular matrix-1, Ecm1). There have been no reports of these proteins in COCs either in vivo or in vitro. There was less functional conservation among genes more highly regulated by versican and unaffected by EGF, but interestingly Adamts5 was induced by versican but not EGF at 6 h, and this induction appeared insensitive to EGF receptor antagonist, suggesting that versican activates additional signaling pathways. Some other versican-induced genes also showed insensitivity to EGF receptor antagonist (e.g., Rgs13; data not shown). Together, these results suggest that EGF has a wider influence on gene transcriptional changes through the canonical EGF receptor transactivation pathway than versican, whereas versican may also activate EGFR independent pathways. In conclusion, these results demonstrate that the induction of Vcan mRNA and the localization of versican protein to the hyaluronan-rich cumulus matrix in ovulating ovarian follicles has important consequences for gene expression in the COC. Whereas the maximum gene induction of key oocyte competence genes was similar after EGF or versican treatment, a slower but more sustained response to versican shows strong similarity to the in vivo temporal pattern, and more sustained gene induction is widely regarded as important for highcompetence IVM. Hence, versican is a unique factor with a sustained EGF-like signaling activity, as well as a key

REFERENCES

9

Article 116

Downloaded from www.biolreprod.org.

1. Russell DL, Robker RL. Molecular mechanisms of ovulation: coordination through the cumulus complex. Hum Reprod Update 2007; 13:289–312. 2. Davis BJ, Lennard DE, Lee CA, Tiano HF, Morham SG, Wetsel WC, Langenbach R. Anovulation in cyclooxygenase-2-deficient mice is restored by prostaglandin E2 and interleukin-1beta. Endocrinology 1999; 140:2685–2695. 3. Hizaki H, Segi E, Sugimoto Y, Hirose M, Saji T, Ushikubi F, Matsuoka T, Noda Y, Tanaka T, Yoshida N, Narumiya S, Ichikawa A. Abortive expansion of the cumulus and impaired fertility in mice lacking the prostaglandin E receptor subtype EP(2). Proc Natl Acad Sci U S A 1999; 96:10501–10506. 4. Wong WY, Richards JS. Evidence for two antigenically distinct molecular weight variants of prostaglandin H synthase in the rat ovary. Mol Endocrinol 1991; 5:1269–1279. 5. Lee KB, Zhang M, Sugiura K, Wigglesworth K, Uliasz T, Jaffe LA, Eppig JJ. Hormonal coordination of natriuretic peptide type C and natriuretic peptide receptor 3 expression in mouse granulosa cells. Biol Reprod 2013; 88:42. 6. Zhang M, Su YQ, Sugiura K, Xia G, Eppig JJ. Granulosa cell ligand NPPC and its receptor NPR2 maintain meiotic arrest in mouse oocytes. Science 2010; 330:366–369. 7. Park JY, Su YQ, Ariga M, Law E, Jin SL, Conti M. EGF-like growth factors as mediators of LH action in the ovulatory follicle. Science 2004; 303:682–684. 8. Conti M, Hsieh M, Zamah AM, Oh JS. Novel signaling mechanisms in the ovary during oocyte maturation and ovulation. Mol Cell Endocrinol 2012; 356:65–73. 9. Russell DL, Salustri A. Extracellular matrix of the cumulus-oocyte complex. Semin Reprod Med 2006; 24:217–227. 10. Akison LK, Alvino ER, Dunning KR, Robker RL, Russell DL. Transient invasive migration in mouse cumulus oocyte complexes induced at ovulation by luteinizing hormone. Biol Reprod 2012; 86:125. 11. Fan HY, Liu Z, Shimada M, Sterneck E, Johnson PF, Hedrick SM, Richards JS. MAPK3/1 (ERK1/2) in ovarian granulosa cells are essential for female fertility. Science 2009; 324:938–941. 12. Shimada M, Hernandez-Gonzalez I, Gonzalez-Robayna I, Richards JS. Paracrine and autocrine regulation of epidermal growth factor-like factors in cumulus oocyte complexes and granulosa cells: key roles for prostaglandin synthase 2 and progesterone receptor. Mol Endocrinol 2006; 20:1352–1365. 13. Downs SM, Chen J. EGF-like peptides mediate FSH-induced maturation of cumulus cell-enclosed mouse oocytes. Mol Reprod Dev 2008; 75: 105–114. 14. Dekel N, Sherizly I. Epidermal growth factor induces maturation of rat follicle-enclosed oocytes. Endocrinology 1985; 116:406–409. 15. Downs SM. Specificity of epidermal growth factor action on maturation of the murine oocyte and cumulus oophorus in vitro. Biol Reprod 1989; 41: 371–379. 16. Eppig JJ. FSH stimulates hyaluronic acid synthesis by oocyte-cumulus cell complexes from mouse preovulatory follicles. Nature 1979; 281:483–484. 17. Gilchrist RB. Recent insights into oocyte-follicle cell interactions provide opportunities for the development of new approaches to in vitro maturation. Reprod Fertil Dev 2011; 23:23–31. 18. Kind KL, Banwell KM, Gebhardt KM, Macpherson A, Gauld A, Russell DL, Thompson JG. Microarray analysis of mRNA from cumulus cells following in vivo or in vitro maturation of mouse cumulus-oocyte complexes. Reprod Fertil Dev 2013; 25:426–438. 19. Dunning KR, Lane M, Brown HM, Yeo C, Robker RL, Russell DL. Altered composition of the cumulus-oocyte complex matrix during in vitro maturation of oocytes. Hum Reprod 2007; 22:2842–2850. 20. Richani D, Ritter LJ, Thompson JG, Gilchrist RB. Mode of oocyte maturation affects EGF-like peptide function and oocyte competence. Mol Hum Reprod 2013; 19:500–509. 21. Richani D, Wang X, Zeng HT, Smitz J, Thompson JG, Gilchrist RB. Prematuration with cAMP modulators in conjunction with EGF-like peptides during in vitro maturation enhances mouse oocyte developmental competence. Mol Reprod Dev 2014; 81:422–435. 22. Prochazka R, Petlach M, Nagyova E, Nemcova L. Effect of epidermal growth factor-like peptides on pig cumulus cell expansion, oocyte maturation, and acquisition of developmental competence in vitro: comparison with gonadotropins. Reproduction 2011; 141:425–435.

DUNNING ET AL.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

10

and extracellular signals to cumulus cells and the oocyte. Biol Reprod 2012; 87:89. Gebhardt KM, Feil DK, Dunning KR, Lane M, Russell DL. Human cumulus cell gene expression as a biomarker of pregnancy outcome after single embryo transfer. Fertil Steril 2011; 96:47–52.e2. Wathlet S, Adriaenssens T, Segers I, Verheyen G, Van de Velde H, Coucke W, Ron El R, Devroey P, Smitz J. Cumulus cell gene expression predicts better cleavage-stage embryo or blastocyst development and pregnancy for ICSI patients. Hum Reprod 2011; 26:1035–1051. Ekart J, McNatty K, Hutton J, Pitman J. Ranking and selection of MII oocytes in human ICSI cycles using gene expression levels from associated cumulus cells. Hum Reprod 2013; 28:2930–2942. Crowe JH, Crowe LM, Carpenter JF, Aurell Wistrom C. Stabilization of dry phospholipid bilayers and proteins by sugars. Biochem J 1987; 242: 1–10. Vanderhyden BC, Caron PJ, Buccione R, Eppig JJ. Developmental pattern of the secretion of cumulus expansion-enabling factor by mouse oocytes and the role of oocytes in promoting granulosa cell differentiation. Dev Biol 1990; 140:307–317. Watson LN, Mottershead DG, Dunning KR, Robker RL, Gilchrist RB, Russell DL. Heparan sulfate proteoglycans regulate responses to oocyte paracrine signals in ovarian follicle morphogenesis. Endocrinology 2012; 153:4544–4555. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 2001; 25:402–408. Benjamini Y, Hochberg Y. Controlling the false discovery rate - a practical and powerful approach to multiple testing. J R Stat Soc Series B Stat Methodol 1995; 57:289–300. Russell DL, Doyle KM, Ochsner SA, Sandy JD, Richards JS. Processing and localization of ADAMTS-1 and proteolytic cleavage of versican during cumulus matrix expansion and ovulation. J Biol Chem 2003; 278: 42330–42339. Sheng W, Wang G, Wang Y, Liang J, Wen J, Zheng PS, Wu Y, Lee V, Slingerland J, Dumont D, Yang BB. The roles of versican V1 and V2 isoforms in cell proliferation and apoptosis. Mol Biol Cell 2005; 16: 1330–1340. Hernandez-Gonzalez I, Gonzalez-Robayna I, Shimada M, Wayne CM, Ochsner SA, White L, Richards JS. Gene expression profiles of cumulus cell oocyte complexes during ovulation reveal cumulus cells express neuronal and immune-related genes: does this expand their role in the ovulation process? Mol Endocrinol 2006; 20:1300–1321. Willmarth NE, Ethier SP. Autocrine and juxtacrine effects of amphiregulin on the proliferative, invasive, and migratory properties of normal and neoplastic human mammary epithelial cells. J Biol Chem 2006; 281: 37728–37737. Wilson KJ, Mill C, Lambert S, Buchman J, Wilson TR, HernandezGordillo V, Gallo RM, Ades LM, Settleman J, Riese DJ II. EGFR ligands exhibit functional differences in models of paracrine and autocrine signaling. Growth Factors 2012; 30:107–116. Wilson KJ, Gilmore JL, Foley J, Lemmon MA, Riese DJ II. Functional selectivity of EGF family peptide growth factors: implications for cancer. Pharmacol Ther 2009; 122:1–8. Mahtouk K, Cremer FW, Reme T, Jourdan M, Baudard M, Moreaux J, Requirand G, Fiol G, De Vos J, Moos M, Quittet P, Goldschmidt H, et al. Heparan sulphate proteoglycans are essential for the myeloma cell growth activity of EGF-family ligands in multiple myeloma. Oncogene 2006; 25: 7180–7191.

Article 116

Downloaded from www.biolreprod.org.

23. Jones GM, Cram DS, Song B, Magli MC, Gianaroli L, Lacham-Kaplan O, Findlay JK, Jenkin G, Trounson AO. Gene expression profiling of human oocytes following in vivo or in vitro maturation. Hum Reprod 2008; 23: 1138–1144. 24. Wells D, Patrizio P. Gene expression profiling of human oocytes at different maturational stages and after in vitro maturation. Am J Obstet Gynecol 2008; 198:455.e1–9; discussion 455.e9–11. 25. Tesfaye D, Ghanem N, Carter F, Fair T, Sirard MA, Hoelker M, Schellander K, Lonergan P. Gene expression profile of cumulus cells derived from cumulus-oocyte complexes matured either in vivo or in vitro. Reprod Fertil Dev 2009; 21:451–461. 26. Russell DL, Ochsner SA, Hsieh M, Mulders S, Richards JS. Hormoneregulated expression and localization of versican in the rodent ovary. Endocrinology 2003; 144:1020–1031. 27. Kim S, Takahashi H, Lin WW, Descargues P, Grivennikov S, Kim Y, Luo JL, Karin M. Carcinoma-produced factors activate myeloid cells through TLR2 to stimulate metastasis. Nature 2009; 457:102–106. 28. Li D, Wang X, Wu JL, Quan WQ, Ma L, Yang F, Wu KY, Wan HY. Tumor-produced versican V1 enhances hCAP18/LL-37 expression in macrophages through activation of TLR2 and vitamin D3 signaling to promote ovarian cancer progression in vitro. PLoS One 2013; 8:e56616. 29. Hu F, a Dzaye OD, Hahn A, Yu Y, Scavetta RJ, Dittmar G, Kaczmarek AK, Dunning KR, Ricciardelli C, Rinnenthal JL, Heppner FL, Lehnardt S, et al. Glioma-derived versican promotes tumor expansion via gliomaassociated microglial/macrophages Toll-like receptor 2 signaling. Neuro Oncol 2015; 17:200–210. 30. Zhang Y, Cao L, Yang BL, Yang BB. The G3 domain of versican enhances cell proliferation via epidermial growth factor-like motifs. J Biol Chem 1998; 273:21342–21351. 31. Xiang YY, Dong H, Wan Y, Li J, Yee A, Yang BB, Lu WY. Versican G3 domain regulates neurite growth and synaptic transmission of hippocampal neurons by activation of epidermal growth factor receptor. J Biol Chem 2006; 281:19358–19368. 32. Du WW, Yang BB, Shatseva TA, Yang BL, Deng Z, Shan SW, Lee DY, Seth A, Yee AJ. Versican G3 promotes mouse mammary tumor cell growth, migration, and metastasis by influencing EGF receptor signaling. PLoS One 2010; 5:e13828. 33. Kern CB, Wessels A, McGarity J, Dixon LJ, Alston E, Argraves WS, Geeting D, Nelson CM, Menick DR, Apte SS. Reduced versican cleavage due to Adamts9 haploinsufficiency is associated with cardiac and aortic anomalies. Matrix Biol 2010; 29:304–316. 34. Dupuis LE, McCulloch DR, McGarity JD, Bahan A, Wessels A, Weber D, Diminich AM, Nelson CM, Apte SS, Kern CB. Altered versican cleavage in ADAMTS5 deficient mice; a novel etiology of myxomatous valve disease. Dev Biol 2011; 357:152–164. 35. McCulloch DR, Nelson CM, Dixon LJ, Silver DL, Wylie JD, Lindner V, Sasaki T, Cooley MA, Argraves WS, Apte SS. ADAMTS metalloproteases generate active versican fragments that regulate interdigital web regression. Dev Cell 2009; 17:687–698. 36. Ricciardelli C, Sakko AJ, Ween MP, Russell DL, Horsfall DJ. The biological role and regulation of versican levels in cancer. Cancer Metastasis Rev 2009; 28:233–245. 37. Brown HM, Dunning KR, Robker RL, Boerboom D, Pritchard M, Lane M, Russell DL. ADAMTS1 cleavage of versican mediates essential structural remodeling of the ovarian follicle and cumulus-oocyte matrix during ovulation in mice. Biol Reprod 2010; 83:549–557. 38. Dunning KR, Watson LN, Sharkey DJ, Brown HM, Norman RJ, Thompson JG, Robker RL, Russell DL. Molecular filtration properties of the mouse expanded cumulus matrix: controlled supply of metabolites