Abstract
The aim of this study was to test the hypothesis that the high ovulation rate in ewes (BB) homozygous for a mutation in the bone morphogenetic protein receptor type 1B (BMPR1B) gene is linked to lower BMP15 and/or GDF9 mRNA in oocytes compared with those in wild-type (++) ewes. Cumulus cell–oocyte complexes (COC) and granulosa cells (GC) were recovered from ≥1 mm diameter follicles of BB and ++ ewes during a prostaglandin-induced follicular phase. Expression levels of GDF9 and BMP15 were measured by multiplex qPCR from individual COC. The gonadotropin-induced cAMP responses of the GC from each non-atretic follicle were measured following treatment with FSH or human chorionic gonadotropin. In a separate validation experiment, GDF9 and BMP15 expression was present only in oocytes and not in cumulus cells. There was no effect of follicular diameter on oocyte-derived GDF9 or BMP15 mRNA levels. The mean expression levels of BMP15, but not GDF9, were significantly lower in all non-atretic follicles, including the subsets containing either FSH- or LH-responsive GC in BB, compared with ++, ewes. No genotype effects were noted for FSH-induced cAMP production by GC either with respect to dose of, or number of follicles responding to, FSH. However, ovaries from BB ewes contained significantly more follicles responsive to LH, with respect to cAMP production in GC. We propose that these findings are consistent with the hypothesis that the higher ovulation rate in BB sheep is due, at least in part, to lower oocyte-derived BMP15 mRNA levels together with the earlier onset of LH-responsiveness in GC.
Introduction
Ewes that inherit a mutation in the bone morphogenetic protein receptor type 1B (BMPR1B; also known as activin-like kinase 6) gene have significantly higher ovulation rates compared with their wild-type contemporaries (see reviews by McNatty et al. (2005) and Fabre et al. (2006)). These BMPR1B-mutant ewes are commonly referred to as Booroola ewes. It has been proposed that the Booroola mutation influences the level of oocyte-derived signalling, possibly via altered protein expression of BMP15 (McNatty et al. 2005) or via altered BMP15 signalling (Mulsant et al. 2001, Campbell et al. 2006). However to our knowledge, this has not been proven. In ewes heterozygous for an inactivating mutation in BMP15, such as the Inverdale (FecXI) or Hanna (FecXH) genotype, it would be reasonable to expect lower levels of oocyte-secreted protein for BMP15 (Galloway et al. 2000). A physiological relationship between lower biologically active concentrations of BMP15 and higher ovulation rates has been demonstrated by immunising wild-type ewes against the entire mature BMP15 protein or small peptide sequences of the mature BMP15 region (Juengel et al. 2002, McNatty et al. 2007). Collectively, the evidence from both the immunisation and mutation studies indicates that as the dose of biologically available BMP15 decreases, the ovulation rate increases until BMP15 concentrations are too low to support follicular growth (McNatty et al. 2004, 2007, Fabre et al. 2006). At this point, animals become infertile, as is the case in homozygous BMP15-mutant animals (Galloway et al. 2000, Scaramuzzi et al. 2011). The notion that altered BMP15 synthesis is in some way linked to the BMPR1B mutation is consistent with the observation that ewes with both an inactivating BMP15 mutation and BMPR1B mutation have ovulation rates that are more than additive (Davis et al. 1999, McNatty et al. 2005).
Recently, the higher ovulation rates in Inverdale ewes were found to be associated with an earlier acquisition of LH (but not FSH) responsiveness by granulosa cells (GC) in a greater proportion of developing antral follicles (McNatty et al. 2009). In homozygous mutant Booroola (BB) ewes, ovarian follicles ovulate at smaller diameters (i.e. >2.5–4.5 mm) compared with those in the wild-type ewes (i.e. >4.5 mm), with the ovulating follicles in BB ewes each containing fewer GC (McNatty et al. 1986, Henderson et al. 1987). Moreover, similar to that which occurs in Inverdale ewes, GC in BB ewes develop a responsiveness to LH earlier in follicular development, consistent with the reduced size of the ovulating follicles (Henderson et al. 1987). It should be noted that although the GC in small (i.e. 1–2.5 mm) antral follicles of BB ewes also showed an increased sensitivity to FSH with respect to cAMP production (Henderson et al. 1987), the FSH preparations used in this experiment contained >1% contamination of LH (McNatty et al. 2009).
The aim of this study was to test the hypothesis that the increased ovulation rate in BB ewes, compared with those in wild-type (++) ewes, is linked, at least in part, to lower the mRNA expression levels of BMP15 and/or GDF9 in oocytes of follicles, particularly in those containing GC that are responsive to LH with regard to cAMP production.
Results
Effect of genotype on ovulation rate
The mean (and range of) ovulation rates based on the number of corpora lutea (CL) present in the wild-type (++) and homozygous mutant (BB) ewes were 2.8 (2–5) and 9.7 (8–11) respectively.
Effect of genotype on number of GC with respect to follicular diameter
With respect to the number of GC present within follicles, there were significant main effects of follicular diameter (P<0.001) and genotype (P<0.001), but had no interactions (Fig. 1).
Effect of genotype and follicular diameter on number of granulosa cells (GC) present in non-atretic follicles. The numbers of GC are represented by black bars for wild-type (++) ewes and white bars for homozygous Booroola mutant (BB) ewes. Values are expressed as geometric means and error bars denote 95% confidence limits. Numbers of follicles are shown in parentheses.
Citation: REPRODUCTION 142, 1; 10.1530/REP-10-0485
Effects of dose, follicular diameter and genotype on FSH- or LH-induced cAMP responses in GC
Follicles described as FSH- or LH-responsive contained GC that resulted in the production of cAMP of >10 pmol/106 cells after treatment with either FSH or LH respectively. Those follicles that were defined as LH-responsive also responded to FSH in 11/11 follicles from ++ ewes and 25/26 follicles from BB ewes. The GC from the one (1/26) follicle from a BB ewe that was defined as LH-responsive, but not FSH-responsive, elicited a cAMP response of 7 pmol/106 cells after treatment with 1000 ng/ml of FSH.
The effects of FSH dose, follicular diameter and genotype on FSH-induced cAMP response of GC are shown in Fig. 2. There were highly significant effects of dose (10, 100 and 1000 ng/ml; P<0.001) and follicular diameter (1–2.5 and >2.5 mm; P<0.001), but not genotype (Fig. 2), on FSH-induced cAMP production in GC and no interactions were observed. The proportion of non-atretic follicles containing FSH-responsive GC was similar (P>0.05) between ++(36/48; 75%) and BB (34/50; 68%) ewes.
Effect of follicular diameter and dose of FSH on cAMP production by granulosa cells from follicles of wild-type (++; black bars) and homozygous Booroola mutant (BB; white bars) ewes combined. Follicles 1–2.5 mm in diameter are represented in the top graph and >2.5 mm diameter are represented in the bottom graph. Values are expressed as geometric means and error bars denote 95% confidence limits. The number of observations across the doses for the 1–2.5 mm follicles is between 39 and 58 and for those >2.5 mm is between 31 and 33.
Citation: REPRODUCTION 142, 1; 10.1530/REP-10-0485
There were significant effects of follicular diameter (P<0.01) and genotype (P<0.05) on LH-induced cAMP production in GC, but no interactions were observed. The geometric mean (and 95% confidence limits) levels for the LH-induced cAMP levels for 1–2.5 and >2.5–4.5 mm diameter follicles in BB ewes were 4 (2, 10) and 16 (6, 46) pmol/106 GC, respectively (no follicles >4.5 mm diameter were present). For the ++ ewes, the cAMP values were 0.8 (0.4, 1.4), 3.2 (0.6, 17.0) and 29 (6, 132) pmol/106 GC for the 1–2.5, >2.5–4.5 and >4.5 mm diameter follicles respectively. The proportion of non-atretic follicles containing LH-responsive GC was higher (P<0.001) in BB (26/50; 52%), compared with ++(11/48; 23%) ewes. This difference in the proportion of LH-responsive follicles was due to responsiveness being attained in more smaller-sized follicles in BB ewes, compared with ++ ewes. Specifically, 41% of small (1–2.5 mm) and 54% of large (>2.5 mm) follicles were LH-responsive in BB ewes compared with only 16 and 44%, respectively, of those in ++ ewes.
Specificity testing of GDF9 and BMP15 in dissociated cumulus cells and denuded oocytes
The localisation of mRNA expression of GDF9 and BMP15, together with gap junction protein alpha 4 (GJA4, also known as connexin 37 (CX37) a specific marker of oocytes), GJA1 (CX43) and FSHR (specific markers of granulosa and cumulus cells), within ovine cumulus cell-oocyte complexes (COC) is shown in Fig. 3. Expression of GDF9, BMP15 and GJA4 mRNA was present only in oocytes, whereas expression of GJA1 and FSHR mRNA was present only in the cumulus cells.
Relative gene expression levels of GDF9, BMP15, GJA4, GJA1 and FSHR in cumulus cells and denuded oocytes from individual cumulus cell–oocyte complexes (n=6). Values are corrected for RPL19 (housekeeping gene) and expressed as means (and errors) relative to the mean value of oocyte-derived GDF9 mRNA expression.
Citation: REPRODUCTION 142, 1; 10.1530/REP-10-0485
Effect of follicular health, genotype and gonadotropin-responsiveness of GC on relative expression levels of GDF9 and BMP15 mRNA in oocytes
These data are summarised in Table 1. The mRNA values for GDF9 and BMP15 with respect to follicular health, genotype and gonadotropin-responsiveness of GC were corrected against a control (calibrator) sample containing differing amounts of GDF9 and BMP15 mRNA, and are referred to herein as relative mRNA levels. Hence, comparisons of the mRNA expression levels of GDF9 within and between genotypes and of BMP15 within and between genotypes were made with relative mRNA expression levels. Also given in Table 1 are the ratios for GDF9/BMP15 mRNA with respect to follicular health, genotype and gonadotropin-responsiveness of GC. These ratio values were obtained by comparing cycle threshold (Ct) values, and were not corrected against the calibrator sample.
Relative expression levels of GDF9 and BMP15 mRNA, and GDF9/BMP15 ratios of cumulus cell–oocyte complexes, with respect to genotype, follicular health and cAMP responsiveness of granulosa cells (GC) to either FSH or LH, from homozygous Booroola mutant (BB) and wild-type (++) ewes.
| GDF9 | BMP15 | GDF9/BMP15 ratios | ||||
|---|---|---|---|---|---|---|
| Follicular type | ++ | BB | ++ | BB | ++ | BB |
| Non-atretic | 0.72±0.06a | 0.66±0.06a | 0.91±0.07a | 0.67±0.06b | 1.19±0.05a | 1.22±0.05a |
| n=56 | n=73 | |||||
| Atretic | 0.42±0.06a,* | 0.42±0.06a,* | 0.52±0.09a,* | 0.49±0.10a,* | 1.14±0.06a | 1.23±0.04a |
| n=26 | n=30 | |||||
| With FSH-responsive GC | 0.78±0.06a | 0.72±0.06a | 0.99±0.08a | 0.63±0.07b | 1.10±0.03a | 1.35±0.13a |
| n=36 | n=30 | |||||
| With LH-responsive GC | 1.02±0.12a | 0.66±0.12a | 1.06±0.17a | 0.67±0.09b | 1.24±0.08a | 1.20±0.04a |
| n=7 | n=23 | |||||
Values are expressed as means±s.e.m. The number of observations for each genotype is indicated in the ratio column. For each variable (GDF9, BMP15 or GDF9/BMP15 ratio), values in each row with different alphabetical superscripts are significantly different (P<0.05) from one another. Asterisks indicate that the values for atretic follicles are significantly (P<0.05) lower than those for the non-atretic follicles for each genotype.
There was no significant effect of follicular diameter on the mean relative expression levels of either GDF9 or BMP15 mRNA (data not shown). For both genes, relative expression levels were higher in non-atretic versus atretic follicles. The mean relative GDF9 mRNA expression levels were similar between genotypes. For the follicles containing GC that were responsive to LH, the relative GDF9 mRNA levels in BB, compared with ++, ewes tended to be lower (P=0.062) but was not significantly different. Conversely, with respect to relative BMP15 expression levels, a significant effect of genotype was noted for oocytes from non-atretic follicles and from the subsets of FSH- and LH-responsive follicles. In all instances, the relative BMP15 mRNA levels were lower in oocytes from the BB ewes (P<0.05). The mean ratios of GDF9:BMP15 irrespective of follicular health for the ++ and BB genotypes were 1.19±0.04 and 1.26±0.04 respectively. These tended to be higher in the BB ewes (P=0.05), with a greater number (P<0.05) of COC with a GDF9:BMP15 ratio of >1.1 in BB compared with ++ ewes. However, no genotype differences were noted for GDF9:BMP15 ratios when follicles were further classified into healthy, atretic or with FSH- or LH-responsive GC (Table 1).
The cycle threshold (Ct) values indicating the expression levels of GDF9 and BMP15 within individual oocytes of ++ and BB ewes were highly correlated (Fig. 4; ++, R2=0.9362, P<0.0001; BB, R2=0.9364, P<0.0001).
Regression plots of cycle threshold (Ct) values and lines of best fit for BMP15 and GDF9 genes from individual cumulus cell–oocyte complexes from wild-type (++; black circles; solid line) and homozygous Booroola mutant (BB; white circles; dashed line) ewes. Equations of lines of best fit and the coefficients of determination (R2) values are presented for both sheep genotypes.
Citation: REPRODUCTION 142, 1; 10.1530/REP-10-0485
Discussion
The key findings from this study were that GDF9 and BMP15 mRNA are expressed exclusively by oocytes in the COC; GDF9 and BMP15 mRNA levels are strongly correlated in both genotypes; and BMP15, but not GDF9, mRNA levels in oocytes of BB ewes were lower than that in oocytes of ++ ewes. The finding from our QPCR studies that GDF9 and BMP15 mRNA are expressed only in oocytes supports previously published in situ hybridisation (ISH) and immunohistochemistry data (Galloway et al. 2000, McNatty et al. 2005, 2006). Recent studies in mice, pigs and humans have suggested that BMP15 and GDF9 are present in granulosa and/or cumulus cells as well as in oocytes of large antral follicles (Margoulis et al. 2009, Sun et al. 2010). It could be argued that ISH is not sufficiently sensitive to detect the low expression levels of BMP15 or GDF9 mRNA in granulosa or cumulus cells. However, the evidence from this study shows clearly that GDF9 and BMP15 mRNA were expressed exclusively in oocytes, and not in surrounding somatic cells devoid of oocyte contamination, of antral follicles at various developmental stages in sheep. The highly correlative relationship between mRNA expression levels of GDF9 and BMP15, regardless of genotype, suggests that the regulation of these genes is very tightly regulated with expression levels higher in non-atretic than atretic follicles but with no effect of diameter of the antral follicles.
The evidence from this study that the BMP15 mRNA levels were lower in BB compared with ++ ewes supports the notion that the oocyte is implicated in influencing the ovulation rate in these animals. It is possible that the GDF9 mRNA levels in follicles with LH-responsive GC were also lower in BB than in wild-type ewes, but the sample size was not large enough for a statistical difference to be observed. It is known that ewes heterozygous for mutations in both BMP15 and GDF9 have ovulation rates that are greater than additive for each mutation alone (McNatty et al. 2005). Therefore, the possibility exists that both genes are expressed at lower levels and that both contribute to the high ovulation rates in these animals. However, until a significant effect for GDF9 is established in a larger study, only lower expression levels of BMP15 mRNA are able to be related to the higher ovulation rate in BB ewes. Currently, seven different point mutations in the BMP15 gene and three in the GDF9 gene have been identified that influence ovulation rate in a number of different sheep breeds (see McNatty et al. (2005) and Paulini & Melo (2011) for reviews). Moreover, in sheep carrying the Woodlands FecX2W mutation that also affects ovulation rate, the expression levels of BMP15, but not GDF9, mRNA are significantly lower than in the wild type. In FecX2W animals, the mutation appears to be on the X-chromosome but has not been linked to any known mutation in the BMP15 gene (Feary et al. 2007). However, as is the case with the BB ewes, a downstream consequence is a reduced expression level of BMP15 mRNA (Feary et al. 2007). Moreover, the results raise the possibility that, at least in part, one common mechanism might exist between ewes with the BMP15, Woodlands and the BMPR1B (Booroola) mutations in that all may have lower BMP15 protein levels in the developing follicles. Of interest is that Inverdale and Booroola ewes have a greater proportion of follicles developing an earlier onset of LH-responsiveness and thereby ovulating at a smaller diameter (McNatty et al. 2009; this study). However, whether the earlier acquisition of LH-responsiveness is directly due to lower concentrations of oocyte-secreted BMP15 protein will require further studies including the measurement of this growth factor in follicular fluid. Currently, no assay for the accurate measurement of BMP15 in biological fluid is available. Owing to a lack of sensitivity, quantification of BMP15 by western blotting using specific BMP15 antibodies would require pooling of follicular fluid from many follicles. Our initial attempts to quantify BMP15 in ovine follicular fluid using a western blotting procedure have been unsuccessful despite the application of an affinity-enrichment step and a highly specific monoclonal antibody. A preferred alternative approach might be the application of a more sensitive two-site sandwich-based ELISA assay and a purified biologically active reference standard; however, this methodology is not yet available.
Although this study shows a decreased level of BMP15 expression in oocytes of BB ewes, the link between this finding and the signalling pathway via the mutant BMPR1B remains obscure. Fabre et al. (2003) examined the effects of the Booroola genotype on BMPR1B signalling activity in HEK-293 cells transiently transfected with BMPR2 and the wild type or mutant forms of BMPR1B. Their results showed an enhanced level of basal signalling activity but an absence of BMP4-stimulated signalling activity in the transfected mutant cells compared with wild-type cells. Campbell et al. (2006) reported that BMP2, 4 or 6 augmented the effects of FSH- and insulin-like growth factor 1 (IGF1)-induced oestradiol production by GC in BB compared with that in ++ ewes, indicating a direct link among BMPs, BMPR1B and follicular somatic cell maturation. However, at present, the specific ligand(s) that interacts with BMPR1B and the associated type II receptor are not known. Possible candidates include BMP15, BMP15–GDF9 heterodimeric complexes and BMP6 (McNatty et al. 2005, Juengel et al. 2006). It is unlikely that other BMPs such as 2, 4 and 7 are candidates as their expression levels in ovine follicular cells were undetectable by ISH (Juengel et al. 2006). Therefore, it is possible that the downstream consequences of the mutated BMPR1B in BB ewes are mediated solely via lower levels of expression of BMP15 mRNA, leading to lower concentrations of BMP15 protein within follicular fluid although additional ligand-induced signalling effects via BMPs other than BMP15 cannot be ruled out.
This study showed that non-atretic, antral follicles, in comparison with atretic follicles, expressed higher levels of BMP15 and GDF9 mRNA. A reduction in mRNA levels of these growth factors in atretic follicles is likely to be due to the reduced support to the oocyte through the loss of cell-to-cell contacts between the GC, and between the GC and the oocyte (Wiesen & Midgley 1994, Lenhart et al. 1998, Carabatsos et al. 2000). The cAMP responses induced by FSH were influenced by follicular diameter and dose of FSH, which is consistent with our earlier observations (Henderson et al. 1987). Previously, we also reported a genotype difference in the sensitivity of GC to FSH as well as LH (Henderson et al. 1987). However, we now know that the FSH preparation was contaminated with ∼1–2% LH (McNatty et al. 2009). In this study, the highly purified FSH preparation contained no measurable LH (i.e. <0.002%) and there was no genotype effect of FSH on cAMP production. Thus, as with heterozygous BMP15-mutant Inverdale ewes (McNatty et al. 2009), the early maturation of antral follicles observed in BB ewes is unlikely to be due to a greater sensitivity to FSH, as known previously (Henderson et al. 1987).
Despite the genotype difference in levels of BMP15, but not GDF9, mRNA, the mean ratios of GDF9:BMP15 mRNA in individual oocytes were not different between the genotypes, although the frequency of higher ratios was significantly lower in BB ewes. This is supported by a less positive slope of regression when Ct values for BMP15 and GDF9 in individual oocytes were plotted on the y and x axes, respectively, compared with that of ++ ewes. The lack of significant difference in the ratio values due to genotype is probably due to a combination of only a small difference between genotypes and substantial variability within individual oocytes. Despite this variability, the expression levels of GDF9 compared with BMP15 were similar (∼1.2) irrespective of follicular diameter, health or gonadotropin-responsiveness. Earlier studies on assessing the production of BMP15 and GDF9 proteins following co-transfection of 293T cells indicate complex intracellular interactions between BMP15 and GDF9, indicating that the processing and secretion of these genes are not independent (see review by Moore & Shimasaki (2005)). This study clearly demonstrates that their expression levels are also unlikely to be independent of one another. Thus, the expression ratios of GDF9:BMP15 are likely to have significant effects on downstream processing and in BMP15 and GDF9 protein levels. It is likely that the expression ratios of GDF9:BMP15 mRNA will vary between mono- and poly-ovulatory species. For example, the rodent is thought to have little or no BMP15 protein expressed throughout the majority of follicular development (Hashimoto et al. 2005). Our preliminary evidence suggests that, indeed, the expression ratios of GDF9 and BMP15 vary between multiple and low ovulation rate species (Crawford & McNatty 2011).
Significant effects of genotype and follicular diameter were noted for human chorionic gonadotropin (hCG)/LH-responsive follicles, consistent with our earlier report (Henderson et al. 1987). That is, a much larger proportion of follicles ≥1 mm diameter responded to LH in BB, compared with ++, ewes. This increase in the proportion of LH-responsive follicles in BB ewes was due mainly to more follicles of a smaller (1–2.5 mm) size attaining LH-responsiveness. In this study, the number of LH-responsive follicles was 11 (N=4) in ++, and 26 (N=3) in BB, ewes. If all these follicles went on to ovulate, this would correspond to ovulation rates of 2.75 and 8.7 in ++ and BB ewes respectively. The numbers of CL observed from the previous ovulatory cycle were 2.8 and 9.7 for ++ and BB ewes respectively. Interestingly, one of the ++ ewes contained five CL, which is rare for wild-type Romney ewes. However, the ovaries from these ewes contained follicles that were typical of those in ++ ewes, i.e. characteristically larger follicles with more GC. Importantly, oocytes recovered from LH-responsive follicles of BB, compared with ++, ewes had lower BMP15 but not GDF9, mRNA levels. This finding is consistent with the overall hypothesis that the higher ovulation rate in BB ewes is linked to the timing of onset of LH-responsiveness in developing follicles and this in turn might be regulated by oocyte-derived BMP15.
In conclusion, using a qPCR approach, these studies confirm for sheep that the expression of GDF9 and BMP15 mRNA in COC is specific to oocytes and is tightly regulated and influenced by follicular health. Additionally, we confirm that a greater proportion of ovarian follicles ≥1 mm diameter in BB ewes attain LH-responsiveness earlier in development compared with that in ++ ewes. This study has established that there are no genotype effects of FSH-responsiveness in follicles. A major new finding is that oocytes from non-atretic follicles, FSH-responsive follicles and presumptive pre-ovulatory follicles (i.e. LH-responsive) in the BB ewes expressed less BMP15 mRNA compared with that in wild types. Therefore, we propose that the common pathway to higher ovulation rates in BB ewes is similar, at least in part, to that in heterozygous BMP15-mutant ewes and that the oocyte plays a role in this pathway.
Materials and Methods
All experiments were performed with the approval of the Invermay Animal Ethics Committee in accordance with the 1999 Animal Welfare Act (Part 6) of New Zealand. All animals had access to pasture and water ad libitum.
Animals
The animals in this study were 5–7-year-old Booroola Romney wild-type (++; N=4) or Booroola Romney homozygous BMPR1B-mutant ewes (BB; N=3). The ++ and BB ewes were derived from mating animals of known pedigrees. The ovaries of all animals were collected 24 h after the onset of a follicular phase, as induced by prostaglandin F2α (PGF2α) treatment on days 10–11 of the luteal phase of the oestrous cycle.
Ovarian collection and recovery of COC and GC
Ovaries were extracted, weighed and collected into saline (0.9%) containing 20 mM Hepes buffer (pH 7.4). Thereafter, all follicles ≥1 mm diameter and CL were dissected at room temperature in DMEM supplemented with 20 mM Hepes buffer, 0.2 mM 3-isobutyl-methyl-xanthine (Sigma Chemical Co.) and 0.1% (w/v) BSA (>97% pure; ImmunoChemical Products Ltd, Auckland, New Zealand). The numbers and weights of CL were recorded. Each individual follicle was transferred into a dry petri dish, and the diameter was measured using a graticule in the eyepiece of the dissecting microscope before puncture for release of intra-follicular contents. The COC was then recovered in 2–4 μl of follicular fluid, snap frozen and stored at −80 °C until total RNA extraction. The remaining follicular fluid was carefully removed from the dish and stored at −20 °C in 100 μl PBS. Thereafter, 1 ml of DMEM was added to the petri dish, and the GC were scraped from the inner wall of the follicle. The number of GC per follicle was determined by haemocytometer and the health status of each individual follicle was determined as described earlier (McNatty et al. 1986). In brief, non-atretic follicles were those defined as having a vascularised theca interna, the absence of debris in follicular fluid and a normal-looking COC and ≥25% of the maximum predetermined number of GC for a given diameter. Atretic follicles were considered to be those that fail one or more of the above criteria. In those atretic follicles, COC were collected for further analyses but GC were discarded.
For individual non-atretic follicles, the GC were centrifuged at 300 g at room temperature for 5 min, and the resulting cell pellets were resuspended in the aforementioned DMEM medium to a final concentration of 60 000 cells per incubation. Cells were incubated with or without FSH (0, 10, 100 and 1000 ng/ml) or hCG (100 ng/ml) in triplicate in a final volume of 600 μl in 48-well culture plates incubated in a water bath at 37 °C for 45 min. When insufficient numbers of cells were collected to undertake all treatments (e.g. in follicles 1–2.5 mm diameter), the minimum treatment regime was the inclusion of a control, a high dose of FSH (1000 ng/ml) and hCG (100 ng/ml) in duplicate. Under this regime, over 70% of all non-atretic follicles 1–2.5 mm diameter and >95% of all non-atretic follicles >2.5 mm were analysed for cAMP production in response to treatment. After the 45 min incubation period, samples were heated to 80 °C for 15 min and then stored at −20 °C until assayed for cAMP.
FSH and LH reagents
The ovine FSH preparation was a highly purified in-house ovine reagent (oFSH Wal; McNatty et al. 2009). This reagent had a bioactivity of 1.4×USDA-oFSH-19-SIAFP or 33 000 IU/mg when the human FSH International Reference Preparation 78/549 was used as a standard in a radio-receptor assay (Cheng 1975). The level of LH contamination was <0.002% as determined by bioassay. The LH reagent was the hCG preparation CR121 (13 450 IU/mg) supplied by the National Institutes of Child Health and Development, Bethesda, MD, USA. The reason for using hCG is that it is devoid of FSH activity, whereas most LH preparations have residual FSH contamination.
Cyclic AMP RIA
The cAMP assay was performed as described previously (McNatty et al. 2009), and had a detection limit of 0.2 pm/106 cells and the intra- and inter-assay coefficients of variation were <9%. The in-house primary antibody had a cross-reactivity of 9% with dibutyryl-cAMP, <0.001% with cGMP and ≤0.0001% with AMP, ADP or ATP.
Analyses of GDF9 and BMP15 mRNA
All reagents used and procedures performed were according to the manufacturers' instructions. The expression levels of GDF9, BMP15 and RPL19 (housekeeping gene) mRNAs and of GJA4, GJA1 and RPL19 were determined in triplex reactions using a Taqman qPCR method. The expression levels of FSHR and RPL19 were determined in singleplex reactions using a SYBR green qPCR method.
Total RNA was extracted from frozen individual COC using reagents from the ArrayPure Nano-Scale RNA Purification kit (Epicentre Biotechnologies, Madison, WI, USA). To remove any genomic DNA, the kit protocol included the incubation of each sample with DNase I for 30 min. Following total RNA extraction, each sample was re-suspended in 10 μl of UltraPure DNase/RNase-free distilled water (Invitrogen). Owing to limited template, the entire DNase-treated total RNA sample was reverse-transcribed using reagents including oligo(dT)20 from the SuperScript VILO cDNA Synthesis kit (Invitrogen).
Primers and Taqman probes for all genes were designed for multiplex (Taqman) and singleplex (SYBR green) qPCR using the computer package ‘Beacon Designer’ (Premier Biosoft International, Palo Alto, CA, USA) and are listed in Table 2. Taqman probes and primers were manufactured by Sigma–Proligo (supplied by Proligo-France SAS 1, Paris France and Proligo-Singapore Pte Ltd, Helios, Singapore) and Invitrogen respectively.
Sequence information and final concentrations (nM) of primers and Taqman probes for sheep GDF9, BMP15, GJA4, GJA1, FSHR and RPL19 genes for use in either triplex Taqman (λGDF9, BMP15 and RPL19; ψGJA4, GJA1 and RPL19) or singleplex SYBR green (δFSHR and RPL19) quantitative PCR reactions.
| Gene | Taqman probe (5′–3′) | nM | Primers (5′–3′) | nM |
|---|---|---|---|---|
| GDF9 | (6FAM)AGTCTCAGCCTCAGATTCCAACGCAGTCCTA(BHQ1) | 50λψ | F-ATTAGCCTTGATTCTCTGCCTTCTAG | 200λ |
| R-GTGTCTCCCACCTAAATGGTTCAG | 200λ | |||
| BMP15 | (HEX)AGAATGTCCCTCAGCCTTCCTGTGTCCCT(BHQ1) | 50λψ | F-AACCTTGTCAGTGAGCTGGTG | 200λ |
| R-AGATACTCCCATTTGCCTCAATCAG | 200λ | |||
| GJA4 | (6FAM)TCCTCTTCGTCAGCACGCCCACCC(BHQ1) | 100ψ | F-TTCCCCATCTCCCACATCCG | 200ψ |
| R-TCGCGCCGAGACAGGTAG | 200ψ | |||
| GJA1 | (HEX)CGGCACTCAAGCTGAATCCATAGATGTACCACT(BHQ1) | 50ψ | F-TCTTCAAGTCTGTCTTCGAGGTG | 200ψ |
| R-CTGATGCGGGCAGGGATC | 50ψ | |||
| FSHR | F-AGGACAGCAAGGTGACAGAGATG | 100δ | ||
| R-GTAGTTTGGGCAGGTTGGAGAAC | 100δ | |||
| RPL19 | (CY5) TTCTCATCCTCCTCATCCACGTTACCTTCTCGG(BHQ3) | 50λψδ | F-TAAGCGAAAGGGTACTGCCAATG | 200λψδ |
| R-TTCTTAGATTCACGGTATCGTCTGAG | 200λψδ |
HEX, 6FAM, CY5 are fluorophores, and BHQ1 and 3 are quenchers from Sigma–Proligo.
A control experiment was performed to determine the localisation of gene expression of GDF9 and BMP15 in different cell types of the COC (i.e. cumulus cells and denuded oocytes). Single COC were extracted from sheep ovaries obtained from a nearby abattoir. Cumulus cells were separated from oocytes by repeated pipetting. Single cumulus cell masses and denuded oocytes were separately washed three times in PBS, and total RNA was extracted and cDNA synthesised as described earlier. For quantification of the expression levels of GDF9 and BMP15 mRNA, or of GJA4 and GJA1 mRNA, triplex reaction mixes were prepared containing primers and Taqman probes for ovine GDF9, BMP15 and RPL19 genes or ovine GJA4, GJA1 and RPL19 genes, respectively, at optimised concentrations (Table 2) and reagents supplied in the Brilliant Multiplex QPCR Master Mix kit (Stratagene, La Jolla, CA, USA). For quantification of the expression levels of FSHR mRNA, singleplex reaction mixes were prepared containing primers for ovine FSHR or RPL19 genes at optimised concentrations (Table 2) and reagents supplied in the ‘Brilliant SYBR Green QPCR Master Mix' kit (Stratagene). Samples were prepared in duplicate by adding an aliquot of neat cDNA (3.12 μl) into the prepared reaction mix (total volume of 52 μl) and then transferring two 25 μl aliquots to adjacent 0.1 ml strip tubes (Corbett Research Ltd, Mortlake, NSW, Australia). The amplification reaction was run on a Rotor-Gene 6000 multiplexing system (Corbett Research Ltd) using the following conditions: 1 cycle of 95 °C for 10 min; 40 cycles of 95 °C for 15 s and 60 °C for 60 s. Before analysis, serial dilutions (1:1 to 1:64) of two samples were made in appropriate singleplex or triplex reactions to validate PCR reaction efficiency for each gene. This included the calculation of the line of best fit (slope±0.1) for all genes tested when ΔΔCT was plotted against log (input total RNA), as well as comparing CT values (<0.5 cycles different) for identical samples for all mRNA transcripts in singleplex and multiplex reactions.
For quantification of the expression levels of GDF9 and BMP15 mRNA in experimental samples, a triplex reaction mix was prepared using 1.04 μl of neat COC-derived cDNA and amplified under the same conditions as described earlier. Controls were incorporated in every run and included randomly picked samples that underwent reverse transcription-PCR with the exclusion of Superscript III/RNaseout enzyme mix to examine the effectiveness of DNase treatment, and reactions that omitted addition of template.
Quantification of samples was calculated using the ΔΔCT method (Livak & Schmittgen 2001). The expression levels of GDF9, BMP15, GJA4, GJA1 and FSHR from samples from the control experiment (i.e. cumulus cells and denuded oocytes) were corrected for RPL19 mRNA levels. The expression levels of GDF9 and BMP15 in the experimental samples (i.e. COC) were relative to a calibrator sample (cDNA from ovine ovary) and not corrected for RPL19 mRNA levels as expression of this gene was not specific to oocytes. Therefore, GDF9 and BMP15 mRNA levels are per oocyte and not relative to total RNA input. The ratios of GDF9:BMP15 mRNA in individual COC was calculated by 
Statistical procedures
Levels of mRNA are expressed as means, and the variance range is estimated by evaluating the 
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
The work was supported by funding from the New Zealand Foundation for Research Science and Technology (C10X0308 and UOAX0814) and in part by a Marsden Grant (08-VUW-010).
Acknowledgements
We acknowledge Dr Lloyd Moore, AgResearch Invermay for the supply of the ovine FSH preparation, the National Institute of Child Health and Development, Bethesda, USA for the supply of hCG, and the Invermay farm staff for animal management.
References
Campbell BK, Souza CJH, Skinner AJ, Webb R & Baird DT 2006 Enhanced response of granulosa and theca cells from sheep carriers of the FecB mutation in vitro to gonadotrophins and bone morphogenic protein-2, -4, and -6. Endocrinology 147 1608–1620 doi:10.1210/en.2005-0604.
Carabatsos MJ, Sellitto C, Goodenough DA & Albertini DF 2000 Oocyte–granulosa cell heterologous gap junctions are required for the coordination of nuclear and cytoplasmic meiotic competence. Developmental Biology 226 167–179 doi:10.1006/dbio.2000.9863.
Cheng K-W 1975 A radioreceptor assay for follicle-stimulating hormone. Journal of Clinical Endocrinology and metabolism 41 581–589 doi:10.1210/jcem-41-3-581.
Crawford JL & McNatty KP 2011 The expression level of oocyte-derived BMP15 relative to GDF9 may be a key determinant in regulating species differences in ovulation-rate. Proceedings of the 44th Annual Meeting for the Society for the Study of Reproduction, Portland, OR, USA. Abstract 636.
Davis GH, Dodds KG & Bruce GD 1999 Combined effect of the Inverdale and Booroola prolificacy genes on ovulation rate in sheep. Proceedings of the Association for the Advancement of Animal Breeding and Genetics 13 74–77.
Fabre S, Pierre A, Pisselet C, Mulsant P, Lecerf F, Pohl J, Monget P & Monniaux D 2003 The Booroola mutation in sheep is associated with an alteration of the bone morphogenetic protein receptor-IB functionality. Journal of Endocrinology 177 435–444 doi:10.1677/joe.0.1770435.
Fabre S, Pierre A, Mulsant P, Bodin L, Di Pasquale E, Persani L, Monget P & Monniaux D 2006 Regulation of ovulation rate in mammals: contribution of sheep genetic models. Reproductive Biology and Endocrinology 4 20 doi:10.1186/1477-7827-4-20.
Feary ES, Juengel JL, Smith P, French MC, O'Connell AR, Lawrence SB, Galloway SM, Davis GH & McNatty KP 2007 Patterns of expression of messenger RNAs encoding GDF9, BMP15, TGFBR1, BMPR1B and BMPR2 during follicular development and characterization of ovarian follicular populations in ewes carrying the Woodlands FecX2W mutation. Biology of Reproduction 77 990–998 doi:10.1095/biolreprod.107.062752.
Galloway SM, McNatty KP, Cambridge LM, Laitinen MP, Juengel JL, Jokiranta TS, McLaren RJ, Luiro K, Dodds KG & Montgomery GW et al. 2000 Mutations in an oocyte-derived growth factor gene (BMP15) cause increased ovulation rate and infertility in a dosage-sensitive manner. Nature Genetics 25 279–283 doi:10.1038/77033.
Hashimoto O, Moore RK & Shimasaki S 2005 Posttranslational processing of mouse and human BMP-15: potential implication in the determination of ovulation quota. PNAS 102 5426–5431 doi:10.1073/pnas.0409533102.
Henderson KM, McNatty KP, O'Keeffe LE, Lun S, Heath DA & Prisk MD 1987 Differences in gonadotrophin-stimulated cyclic AMP production by granulosa cells from booroola x merino ewes which were homozygous, heterozygous or non-carriers of a fecundity gene influencing their ovulation rate. Journal of Reproduction and Fertility 81 395–402 doi:10.1530/jrf.0.0810395.
Livak KJ & Schmittgen TD 2001 Analysis of relative gene expression data using real-time quantitative PCR and the method. Methods 25 402–408 doi:10.1006/meth.2001.1262.
Juengel JL, Hudson NL, Heath DA, Smith P, Reader KL, Lawrence SB, O'Connell AR, Laitinen MP, Cranfield M & Groome NP et al. 2002 Growth differentiation factor 9 and bone morphogenetic protein 15 are essential for ovarian follicular development in sheep. Biology of Reproduction 67 1777–1789 doi:10.1095/biolreprod.102.007146.
Juengel JL, Reader KL, Bibby AH, Lun S, Ross I, Haydon LJ & McNatty KP 2006 The role of bone morphogenetic proteins 2, 4, 6 and 7 during ovarian follicular development in sheep: contrast to rat. Reproduction 131 501–513 doi:10.1530/rep.1.00958.
Lenhart JA, Downey BR & Bagnell CA 1998 Connexin 43 gap junction protein expression during follicular development in the porcine ovary. Biology of Reproduction 58 583–590 doi:10.1095/biolreprod58.2.583.
Margoulis S, Abir R, Felz C, Nitke S, Krissi H & Frisch B 2009 Bone morphogenetic protein 15 expression in human ovaries from fetuses, girls and women. Fertility and Sterility 92 1666–1673 doi:10.1016/j.fertnstert.2008.08.119.
McNatty KP, Lun S, Heath DA, Ball K, Smith P, Hudson NL, McDiarmid J, Gibb M & Henderson KM 1986 Differences in ovarian activity between booroola X merino ewes which were homozygous, heterozygous and non-carriers of a major gene influencing their ovulation rate. Journal of Reproduction and Fertility 77 193–205 doi:10.1530/jrf.0.0770193.
McNatty KP, Moore LG, Hudson NL, Quirke LD, Lawrence SB, Reader K, Hanrahan JP, Smith P, Groome NP & Laitinen M et al. 2004 The oocyte and its role in regulating ovulation rate: a new paradigm in reproductive biology. Reproduction 128 379–386 doi:10.1530/rep.1.00280.
McNatty KP, Smith P, Moore LG, Reader K, Lun S, Hanrahan JP, Groome NP, Laitinen M, Ritvos O & Juengel JL 2005 Oocyte-expressed genes affecting ovulation rate. Molecular and Cellular Endocrinology 234 57–66 doi:10.1016/j.mce.2004.08.013.
McNatty KP, Lawrence S, Groome NP, Meerasahib MF, Hudson NL, Whiting L, Heath DA & Juengel JL 2006 Meat and Livestock Association Plenary Lecture 2005. Oocyte signalling molecules and their effects on reproduction in ruminants. Reproduction, Fertility and Development 18 403–412 doi:10.1071/RD05104.
McNatty KP, Hudson NL, Whiting L, Reader KL, Lun S, Western A, Heath DA, Smith P, Moore LG & Juengel JL 2007 The effects of immunizing sheep with different BMP15 or GDF9 peptide sequences on ovarian follicular activity and ovulation rate. Biology of Reproduction 76 552–560 doi:10.1095/biolreprod.106.054361.
McNatty KP, Heath DA, Hudson NL, Lun S, Juengel JL & Moore LG 2009 Gonadotrophin-responsiveness of granulosa cells from bone morphogenetic protein 15 heterozygous mutant sheep. Reproduction 138 545–551 doi:10.1530/REP-09-0154.
Moore RK & Shimasaki S 2005 Molecular biology and physiological role of the oocyte factor, BMP15. Molecular and Cellular Endocrinology 234 67–73 doi:10.1016/j.mce.2004.10.012.
Mulsant P, Lecerf F, Fabre S, Schibler L, Monget P, Lanneluc I, Pisselet C, Riquet J, Monniaux D & Callebaut I et al. 2001 Mutation in bone morphogenetic protein receptor-IB is associated with increased ovulation rate in Booroola Mérino ewes. PNAS 98 5104–5109 doi:10.1073/pnas.091577598.
Paulini F & Melo EO 2011 The role of oocyte-secreted factors GDF9 and BMP15 in follicular development and oogenesis. Reproduction in Domestic Animals 46 354–361 doi:10.1111/j.1439-0531.2010.01739.x.
Scaramuzzi RJ, Baird D, Campbell B, Driancourt MA, Dupont J, Fortune J, Gilchrist R, Martin G, McNatty K & McNeilly A et al. 2011 Regulation of folliculogenesis and the determination of ovulation rate in ruminants: a review. Reproduction, Fertility, and Development 23 444–467 doi:10.1071/RD09161.
Sun RZ, Lei L, Cheng L, Jin ZF, Zu SJ, Shan ZY, Wang ZD, Zhang JX & Liu ZH 2010 Expression of GDF-9, BMP-15 and their receptors in mammalian ovary follicles. Journal of Molecular Histology 41 325–332 doi:10.1007/s10735-010-9294-2.
Wiesen JF & Midgley AR Jr 1994 Expression of connexin 43 gap junction messenger ribonucleic acid and protein during follicular atresia. Biology of Reproduction 50 336–348 doi:10.1095/biolreprod50.2.336.
