Potential involvement of EGF-like growth factors and phosphodiesterases in initiation of equine oocyte maturation

Animal Reproduction Science 103 (2008) 187–192

Short communication

Potential involvement of EGF-like growth factors and phosphodiesterases in initiation of equine oocyte maturation S.M. Lindbloom, T.A. Farmerie, C.M. Clay, G.E. Seidel Jr., E.M. Carnevale ∗ Animal Reproduction and Biotechnology Laboratory, Colorado State University, Fort Collins, Colorado 80523-1683, USA Received 25 April 2006; accepted 12 April 2007 Available online 20 April 2007

Abstract Human chorionic gonadotropin (hCG) was administered to mares in estrus with large, dominant ovarian follicles to initiate follicular and oocyte maturation. Follicular contents were collected at 0, 2, 4 and 6 h after hCG. Epiregulin, amphiregulin and phosphodiesterase (PDE) mRNA contents of granulosa cells (PDE 4D) were determined by reverse transcription and real-time PCR; PDE 3A mRNA content of single oocytes was determined similarly. Copy numbers of mRNA did not increase for PDE 3A or 4D over the time interval studied. Amounts of epiregulin and amphiregulin mRNA were correlated (r = 0.98) when log transformed. Epiregulin and amphiregulin mRNA increased (P < 0.01) from controls by 4 h after hCG administration, with amphiregulin increasing (P < 0.01) by 2 h after hCG administration. Epiregulin and amphiregulin mRNA levels remained elevated (P < 0.01) at 6 h after hCG. These results indicate that EGF-like growth factors are likely paracrine mediators of the LH signal in the horse. © 2007 Elsevier B.V. All rights reserved. Keywords: Oocyte maturation; Epiregulin; Amphiregulin; Phosphodiesterase; Equine

1. Introduction Maturation of the ovarian follicle and oocyte normally occur in synchrony, resulting in ovulation of a metaphase II oocyte in most species. Horses are an excellent model to study maturation ∗

Corresponding author. Tel.: +1 970 491 8626; fax: +1 970 491 3557. E-mail address: [email protected] (E.M. Carnevale).

0378-4320/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.anireprosci.2007.04.006

188

S.M. Lindbloom et al. / Animal Reproduction Science 103 (2008) 187–192

mechanisms, as mares are monovulatory with a longer phase of follicular development than most species and follicular wave pattern similar to women (Ginther et al., 2004). Administration of human chorionic gonadotropin (hCG) to the estrous mare will induce ovulation in approximately 36 h, and induction of ovulation in women with hCG requires a similar time interval (Edwards and Steptoe, 1975). For years, the LH surge was thought to function directly on the cumulus-oocyte complex (COC) in most species; however, neither oocytes nor cumulus cells have measurable numbers of LH receptors (Peng et al., 1991). This controversy was resolved in 2004, at least in mice, when Park et al. (2004) showed that LH stimulation of granulosa cells led to production of the EGFlike growth factors, amphiregulin and epiregulin. Inhibiting release of these factors with matrix metalloprotease inhibitors suppresses LH-induced oocyte maturation (Ashkenazi et al., 2005). EGF-like growth factors bind the EGF receptor (Johnson et al., 1993) on cumulus cells and cause cumulus expansion and germinal vesicle breakdown (GVBD). Prior to the initiation of oocyte maturation, cAMP concentrations within the oocyte and cumulus cells are maintained at enhanced concentrations as compared to other times during oocyte maturation. Oocyte maturation is initiated by a general decrease in cAMP within the oocyte, which could be caused by the activation of phosphodiesterases (PDE). Thomas et al. (2004) reported that isoforms of PDE are found in cumulus cells (PDE 4D) and oocytes (PDE 3A). Administration of hCG may lead to increased transcription of PDE 4D and decreased cAMP entering the oocyte from the cumulus cells via gap junctions. Transcription of PDE 3A may also be up regulated in response to hCG. Masciarelli et al. (2004) developed PDE 3A knockout mice and showed that oocytes from these mice were permanently arrested in germinal vesicle stage. Therefore, PDE 3A is essential for oocyte maturation and may be a key step in the initiation of oocyte maturation by LH or hCG. The objectives of the present experiment were to determine if administration of hCG to estrous mares resulted in an increase in mRNA for EGF-like growth factors (amphiregulin and epiregulin) and PDE 3A and 4D in the dominant equine ovarian follicle. 2. Materials and methods 2.1. Equine-specific primer design To design equine-specific primers for epiregulin, amphiregulin, PDE 3A and PDE 4D, human, mouse and bovine sequences were aligned. Several sets of primers were designed based on conserved regions, and efficiency of each set of primers for PCR at various annealing temperatures was evaluated using the expanded high fidelity PCR system (Roche Applied Science, Indianapolis, IN). Equine genomic DNA was used for PCR reactions. Electrophoresis (1% agarose) was used to identify products of the correct size. Products were gel-isolated and ligated into pGEM-T Easy (Promega, Madison, WI). Presence of the insert was verified by restriction digest, and plasmids containing the correct insert were sequenced. Equine-specific sequence amplified by the original primers was used to design new, equinespecific primers, and these were used with equine genomic DNA to produce amplicons which were sequenced and submitted to Genbank (epiregulin, DQ2358588; amphiregulin, DQ238589, and PDE 3A, DQ238592) (Table 1). When a primer pair yielded the correct sequence from genomic DNA, that primer pair was used to amplify partial cDNA generated from granulosa cells or oocytes. Partial equine cDNAs were sequenced (Table 1), and sequences were submitted to Genbank (PDE 3A, DQ238591 and PDE 4D, DQ238590).

S.M. Lindbloom et al. / Animal Reproduction Science 103 (2008) 187–192

189

Table 1 Equine-specific primer pairs for partial equine cDNAs or genomic sequences, PCR product lengths and annealing temperatures for the four genes of interest Gene

Primer pairs

Epiregulin

5 -caataacgaagtgcagctctga3 ,5 -gcattgtccatgcaaacagt-3 5 -agtgctgatgggtttgaggt-3 ,5 ggatatttgtggttcgttgtca-3 5 -atcctgacgacgcagcctatt3 ,5 -ggcagggatatttccagaca-3 5 -cgcatcatggaggagttctt-3 ,5 ccctcctcctggtcatca-3

Amphiregulin PDE 3A PDE 4D

Annealing temperature (◦ C)

Intron spanning

Source

53

58

No

132

58

No

173

58

Yes

Genomic DNA Genomic DNA cDNA

269

58

Yes

cDNA

Product length (bp)

2.2. Collection of COCs and granulosa cells Cumulus-oocyte complexes and granulosa cells were recovered from light-horse mares (Equus caballus) between 450 and 600 kg and three to ten years of age, using procedures approved by the institutional animal care and use committee at Colorado State University. Reproductive tracts were evaluated daily by transrectal ultrasound until the following criteria were observed: (1) growing follicle 35 ± 2 mm in diameter (average of length and width); (2) uterine edema indicative of estrus; and (3) no corpus luteum. Mares were randomly assigned to undergo follicular aspiration at 0, 2, 4 or 6 h after administration of human chorionic gonadotropin (hCG, 2000 iu, i.v.). Follicle cells from three different mares were collected for each time point; follicle cells and oocytes were not always collected from the same mare for a time point. Follicles were aspirated as previously described (Carnevale and Ginther, 1993). The COCs and granulosa-cell sheets were identified and rinsed. Each COC was placed in an Eppendorf tube (0.6 mL) with 200 units of hyaluronidase, vortexed for 3 min to dissociate cells, and then centrifuged for 10 s in a microcentrifuge at maximum speed to pellet the oocyte and cumulus cells. The denuded oocyte was placed into a 0.6 mL Eppendorf tube with 25 ␮L of RNAlater (Qiagen, Valencia, CA) and stored at −80 ◦ C. The number of cumulus cells was too few to count reliably; therefore, PDE 4D mRNA was evaluated only in granulosa cells. Several sheets of granulosa cells from each follicle were placed in an Eppendorf tube with 200 units of hyluronidase, vortexed for 3 min, pelleted by centrifugation, and re-suspended in 25 ␮L of RNA later. Concentration and total number of granulosa cells were determined by hemacytometer counting. Granulosa cells were stored at −80 ◦ C until RNA isolation. 2.3. Generation of DsRed RNA To serve as an exogenous standard for tracking the efficiency of RNA isolation and cDNA synthesis, dsRed RNA in pBluescript 11 KS vector (Stratagene, LaJolla, CA) was added prior to isolation of RNA. The DsRed sequence was transcribed into RNA in vitro (RiboMAXTM Large Scale RNA production Systems, Promega, Madison, WI).

190

S.M. Lindbloom et al. / Animal Reproduction Science 103 (2008) 187–192

2.4. RNA isolation and cDNA generation A single denuded oocyte or 2.5 × 104 granulosa cells were used for RNA isolations. Approximately 4 × 104 copies of DsRed RNA were added to each sample. Equine RNA was isolated (RNeasy® Micro Kit, Qiagen, Valencia, CA), and following DNase treatment, half of the RNA isolated from each sample was used to generate cDNA (Sensiscript® Reverse Transcription Kit, Qiagen, Valencia, CA). The remaining RNA was treated as described, except that reverse transcriptase (RT) was omitted from the RT to serve as a negative control for DNA contamination. 2.5. Quantitative real-time PCR analysis Half of each RT reaction (10 ␮L) was used in the real-time PCR reaction for the gene of interest, while the other half was used for the quantification of the exogenous standard, DsRed. Primers used to amplify a 137 bp segment of DsRed were 5 -gaacgtcatcaccgagttca-3 and 5 ccttggtcaccttcagcttc-3 . Exogenous standard data was used to normalize results for genes of interest. Real-time PCR was performed using the QuantiTectTM SYBR® Green PCR Kit (Qiagen, Valencia, CA) and the iCycler iQTM (Bio Rad, Hercules, CA). Quantitation was based on seven-point standard curves (10−16 to 10−22 mol) generated for epiregulin, amphiregulin, PDE 3A, PDE 4D and DsRed. Melting curves were conducted for each sample to confirm specificity of amplification. Increasing numbers of granulosa cells from one follicle were used as starting material to determine reliability of the RNA isolation, cDNA synthesis, and real-time PCR protocols. For each gene of interest, except PDE 3A, 1 × 104 , 5 × 104 or 1 × 105 cells resulted in increasing copy number ratios. Because of the difficulty in obtaining large numbers of oocytes, this procedure was not performed for PDE 3A mRNA, as it is an oocyte-specific gene. 2.6. Statistics Quantile-quantile plots were used to determine normality of the data sets. Residuals were plotted against predicted values to assess homogeneity of variances. Data, for endpoints other than PDE 4D, were log transformed to obtain normality and homogeneity of variances. PDE 4D data were normally distributed with homogeneous variances without transformation. For each gene, mean mRNA copy numbers for each time point were compared by analysis of variance (statistical analysis system software, GLM procedure, Cary, NC) and Tukey’s test. A correlation coefficient was generated between log epiregulin and log amphiregulin mRNA copy numbers. 3. Results 3.1. Epiregulin and amphiregulin mRNA content in granulosa cells Epiregulin mRNA concentrations (Table 2) were greater in granulosa cells at 2, 4, and 6 h after hCG administration than before hCG administration at 0 h (P < 0.02, P < 0.002 and P < 0.002, respectively). Amphiregulin mRNA concentration (Table 2) was greater (P < 0.01) at 2 h after hCG administration than at 0 h and greater (P < 0.01) at 4 and 6 h than at 2 h. Therefore, systemic hCG administration resulted in increased epiregulin and amphiregulin mRNAs in equine granulosa cells. Epiregulin and amphiregulin mRNA copy numbers were highly correlated (r = 0.98) when log transformed.

S.M. Lindbloom et al. / Animal Reproduction Science 103 (2008) 187–192

191

Table 2 Means ± S.E.M. for log transformed epiregulin, amphiregulin and PDE 4D mRNA copy number per 1000 equine granulosa cells and log transformed PDE 3A mRNA copy numbers per equine oocyte Hours after hCG 0 2 4 6

Epiregulin 2.03 4.65 6.27 6.25

a b c d e

± ± 0.83b ± 0.14b ± 0.06b 0.45a

Amphiregulin 1.09 2.71 4.36 4.34

± ± 0.39d ± 0.22e ± 0.03e 0.18c

PDE 4D 4.21 4.66 4.38 4.46

± ± ± ±

4.10 3.26 3.98 3.50

PDE 3A 4.43 4.55 4.86 5.34

± ± ± ±

0.41 0.22 0.09 0.60

Means within a column with different superscripts differed (P < 0.02). Means within a column with different superscripts differed (P < 0.02). Samples from three mares were used for each time point (P < 0.01). Samples from three mares were used for each time point (P < 0.01). Samples from three mares were used for each time point (P < 0.01).

3.2. Phosphodiesterase content Concentrations of PDE 3A in oocytes or 4D mRNA in granulosa cells did not increase in response to hCG administration within the time period studied. Individual values for the three oocytes at 6 h after administration of hCG (12, 555, 357, 427, 436 and 124, 078 PDE 3A mRNA copies) appeared greater compared to individual values at 0 h (4, 952, 30, 588 and 129, 298 PDE 3A mRNA copies), but the extent of variation made interpretation problematic. 4. Discussion The paracrine mediation of the LH/hCG signal by EGF-like growth factors may be a mechanism common to many mammals; the research reported here indicates that this mechanism likely is functioning in the horse as well as in rodents, where it was first discovered (Park et al., 2004). The mRNA for EGF-like growth factors, epiregulin and amphiregulin, increased in equine granulosa cells in response to systemic administration of hCG. The increase in these two genes was highly correlated (r = 0.98) indicating coordinated regulation of gene expression. In rodents, amphiregulin and epiregulin can each cause GVBD as effectively as LH or hCG, while inhibition of the EGF receptor within the follicle blocks the COC response to LH stimulation (Park et al., 2004). This evidence supports the theory that the actual mechanism of LH stimulation is via paracrine factors (EGF-like growth factors in this case) released by granulosa cells and subsequently bound to EGF receptors on cumulus cells. In mares, epiregulin and amphiregulin mRNA concentrations in granulosa cells increased from 0 to 2 h after hCG administration. Amphiregulin increased again from 2 to 4 h after hCG treatment. No further increase was observed between 4 and 6 h after hCG. In the mouse, elevated EGF-like growth factor concentrations were observed by 1 h after hCG treatment, and they declined by 6 h (Park et al., 2004). The interval to EGF-like growth factor expression may be longer in the horse, resulting in PDE gene expression after the time period evaluated in this experiment. Because epiregulin and amphiregulin mRNA concentrations increased between 2 and 4 h after systemic hCG administration, the final samples may have been obtained too soon after administration of hCG to identify significant increases in PDE transcription. In conclusion, between 2 and 4 h after systemic hCG administration to mares, the mRNA content of epiregulin and amphiregulin increased in granulosa cells of the dominant ovarian

192

S.M. Lindbloom et al. / Animal Reproduction Science 103 (2008) 187–192

follicle of horses. More numbers and time points are needed to characterize this response more completely, as well as that of PDE mRNA in oocyte and follicular cells of mares. Acknowledgments This work was supported by the benefactors for the preservation of equine genetics and the Abney Foundation. We thank Jake Cox for managing mares, and Joanne Stokes for technical assistance with oocyte collections. References Ashkenazi, H., Cao, X., Motola, S., Popliker, M., Conti, M., Tsafriri, A., 2005. Epidermal growth factor family members: endogenous mediators of the ovulatory response. Endocrinology 146, 77–84. Carnevale, E.M., Ginther, O.J., 1993. Use of a linear ultrasonic transducer for the transvaginal aspiration and transfer of oocytes in the mare. J. Equine Vet. Sci. 13, 25–27. Edwards, R.G., Steptoe, P.C., 1975. Induction of follicular growth, ovulation and luteinization in the human ovary. J. Reprod. Fertil. 22 (Suppl), 121–163. Ginther, O., Gastal, E., Gastal, M., Bergfelt, D., Baerwald, A., Pierson, R., 2004. Comparative study of the dynamics of follicular waves in mares and women. Biol. Reprod. 71, 1195–1201. Johnson, G., Kannan, B., Shoyab, M., Stromberg, K., 1993. Amphiregulin iduces tyrosine phosphorylation of the epidermal growth factor receptor and p185erbB2. Evidence that amphiregulin acts exclusively through the epidermal growth factor receptor at the surface of human epithelial cells. J. Biol. Chem. 268, 2924–2931. Masciarelli, S., Horner, K., Liu, C., Park, S.H., Hinckley, M., Hockman, S., Nedachi, T., Jin, C., Conti, M., Manganiello, V., 2004. Cyclic nucleotide phosphodiesterase 3A-deficient mice as a model of female infertility. J. Clin. Invest. 114, 196–205. Park, J.Y., Su, Y.Q., Ariga, M., Law, E., Jin, S.L., Conti, M., 2004. EGF-like growth factors as mediators of LH action in the ovulatory follicle. Science 303, 682–684. Peng, X.R., Hsueh, A., LaPolt, P., Bjersing, L., Ny, T., 1991. Localization of Lutenizing hormone receptor messenger ribonucleic acid expression in ovarian cell types during follicle development and ovulation. Endocrinology 129, 3200–3207. Thomas, R.E., Armstrong, D.T., Gilchrist, R.B., 2004. Bovine cumulus cell-oocyte gap junctional communication during in vitro maturation in response to manipulation of cell-specific cyclic adenosine 3 ,5 -monophosophate levels. Biol. Reprod. 70, 548–556.