Leptin is an important hormone influencing reproductive function. However, the mechanisms underpinning the role of leptin in the regulation of reproduction remain to be completely deciphered. In this study, our objective is to understand the mechanisms regulating the expression of leptin receptor (Lepr) and its role in ovarian granulosa cells during ovulation. First, granulosa cells were collected from superovulated mice to profile mRNA expression of Lepr isoforms (LeprA and LeprB) throughout follicular development. Expression of LeprA and LeprB was dramatically induced in the granulosa cells of ovulating follicles at 4 h after human chorionic gonadotropin (hCG) treatment. Relative abundance of both mRNA and protein of CCAAT/enhancer-binding protein β (Cebpβ) increased in granulosa cells from 1 to 7 h post-hCG. Furthermore, chromatin immunoprecipitation assay confirmed the recruitment of Cebpβ to Lepr promoter. Thus, hCG-induced transcription of Lepr appears to be regulated by Cebpβ, which led us to hypothesise that Lepr may play a role during ovulation. To test this hypothesis, we used a recently developed pegylated superactive mouse leptin antagonist (PEG-SMLA) to inhibit Lepr signalling during ovulation. I.p. administration of PEG-SMLA (10 μg/g) to superovulated mice reduced ovulation rate by 65% compared with control treatment. Although the maturation stage of the ovulated oocytes remained unaltered, ovulation genes Ptgs2 and Has2 were downregulated in PEG-SMLA-treated mice compared with control mice. These results demonstrate that Lepr is dramatically induced in the granulosa cells of ovulating follicles and this induction of Lepr expression requires the transcription factor Cebpβ. Lepr plays a critical role in the process of ovulation by regulating, at least in part, the expression of the important genes involved in the preovulatory maturation of follicles.
Leptin, a hormone of adipose tissue origin, is primarily responsible for regulating energy and food intake. However, there is also clear evidence of its role in reproduction (Barash et al. 1996, Wauters et al. 2000). There are multiple isoforms of the leptin receptor (Lepr, A–F), of which LeprA (short isoform) and LeprB (long isoform) are known to transduce leptin signals in multiple cell types (Lee et al. 1996, Murakami et al. 1997). Although there has been significant progress in our understanding of the mechanisms of Lepr signalling in the hypothalamus, such clearer understanding is still lacking at the level of the ovary, specifically in granulosa cells.
Majority of studies examining direct effects of leptin on the ovary and granulosa cells have been in vitro, of which very few have addressed multiple Lepr isoforms. Expression of Lepr (mainly LeprB) has been demonstrated in the granulosa cells of many species including the mouse (Ryan et al. 2002), pig (Ruiz-Cortes et al. 2003), cattle (Spicer & Francisco 1997) and human (Karlsson et al. 1997), clearly suggesting a direct effect of leptin on these cells. Expression of LeprA and LeprB was shown to be induced by preovulatory LH stimulus in rat ovaries (Ryan et al. 2003), although it was not measured in pure population of granulosa and luteal cells. Though numerous studies demonstrate the expression of Lepr in granulosa cells, none have addressed the mechanisms of regulation of Lepr expression. Potential transcription factors that could regulate Lepr expression in ovulating follicles include nuclear receptor 5a2 (Nr5a2; Duggavathi et al. 2008), progesterone receptor (Pgr; Lydon et al. 1995) and CCAAT/enhancer-binding protein β (Cebpβ; Sterneck et al. 1997), as they all play an indispensable role in ovulation. Further, it was recently reported that Lepr mRNA abundance was reduced in conditional knockout mice lacking Cebpα/β in granulosa cells (Fan et al. 2011). However, whether Cebpβ plays a role in Lepr expression in granulosa cells has not been investigated.
Regarding the role of Lepr in granulosa cells, the available data from in vitro studies have revealed contradicting results. Most studies have focused on their role in steroidogenesis of granulosa cells. In the rat and bovine, leptin was shown to suppress oestradiol (E2) production (Zachow & Magoffin 1997, Spicer & Francisco 1997, Duggal et al. 2000), whereas in the rabbit it showed augmented E2 production (Sirotkin et al. 2009). Leptin was shown (Accili et al. 1996) to have suppressive effects on progesterone synthesis in bovine granulosa cells (Spicer & Francisco 1997, Spicer 2001, Sirotkin et al. 2009); whereas dose-dependent bimodal effects were demonstrated in porcine granulosa cells (Ruiz-Cortes et al. 2003). Reported variations on the effect of leptin on granulosa cells appear to be due to wide range of dose of leptin (1–100 ng/ml) used in different studies. Although in vitro approach allows for investigation of direct effects, it is not suitable for testing the role of Lepr in processes like follicular rupture, cumulus expansion and corpus luteum (CL) formation.
Data on mRNA expression in rat ovaries (Ryan et al. 2003) indicate that Lepr may be important for ovulation and/or luteinisation. Indeed, leptin is known to increase the expression of prostaglandin-endoperoxide synthase 2 (Ptgs2) in multiple cell types in vitro (Gao et al. 2009, Hsu et al. 2012, Manuel-Apolinar et al. 2013). However, direct effects of leptin, examined using superovulation models, on ovulation are also confusing with studies showing both reduced and increased number of ovulations in the rat (Almog et al. 2001, Ricci et al. 2006). This may be because of the number and dosage of leptin used in the treatments (Di Yorio et al. 2013). Thus, an alternative approach to study leptin signalling in granulosa cells would be to investigate the importance of Lepr expression in granulosa cells. Therefore, the objectives of our study were i) to examine pattern and regulation of the expression of LeprA and LeprB in granulosa cells during follicular and luteal development and ii) to examine the importance of Lepr expression in granulosa cells during ovulation.
Materials and methods
All experimental protocols were approved by the Animal Care and Use Committee of McGill University. The inbred C57BL/6NCrl mice (Charles River, Senneville, QC, Canada) were housed in standard plastic rodent cages and maintained on a 12 h light:12 h darkness cycle with ad libitum feed (Rodent Diet, Harlan Teklad, Montreal, QC, Canada) and water.
Superovulation and sample collection
Immature (23–25 days old) female mice were superovulated by administration of equine chorionic gonadotropin (eCG, 5 IU, i.p.) followed 48 h later by human chorionic gonadotropin (hCG, 5 IU, i.p.) (Duggavathi et al. 2008). Animals were killed at specific time points during the gonadotropin-stimulated follicular and luteal development. Pure populations of granulosa cells were collected by laser microdissection (Duggavathi et al. 2008) for initial profiling of genes of leptin signalling system. For further experiments involving the periovulatory period, granulosa cells were collected by follicular puncture using 27G needle. The cell suspension was passed through a cell strainer (BD Falcon, Mississauga, ON, Canada; 40 μm) to filter out cumulus–oocyte complexes. Granulosa cells from both ovaries of each mouse were pooled together. For luteal cell collection, CL from both ovaries were scraped off with a 27G needle and collected in PBS. All samples were stored at −80 °C until further analyses.
Inhibition of leptin signalling
A recently developed pegylated superactive mouse leptin antagonist (PEG-SMLA; mutant D23L/L39A/D40A/F41A; PLR Laboratories, Rehovot, Israel), which has been shown to inhibit leptin signalling (Shpilman et al. 2011), was used in this study. In line with hCG-induced Lepr expression in the granulosa cells of ovulating follicles, we administered PEG-SMLA (10 μg/g body weight, i.p.) into eCG-primed mice at the time of hCG stimulation. Control mice were treated with PBS.
Evaluation of ovulation rate and oocyte maturation
Animals were killed at 18 h post-hCG to collect oviducts. The cumulus–oocyte complexes were collected and the cumulus cells were removed by a brief exposure to hyaluronidase (1 mg/ml). Oocytes were counted for ovulation rate and fixed in paraformaldehyde in PBS for 12 min at room temperature. They were then placed in PBS with 1% BSA in four-well plates and stored at 4 °C. The following day, the oocytes were placed on slides with moviol and Hoechst 33342 stain (10 μg/ml) and observed under a stereomicroscope. The presence of a polar body and condensed chromosomes aligned on the metaphase plate indicated that the oocytes had progressed to the mature metaphase II stage (Mandelbaum 2000).
Relative mRNA expression by quantitative PCR
All quantitative PCR (qPCR) procedures were carried out in accordance with MIQE guidelines (Bustin et al. 2009). Total RNA was extracted from granulosa and luteal cells using the PicoPure RNA isolation Kit (Life Technologies, Burlington, ON, Canada). Using 250 ng of total RNA, cDNA was synthesised using the iScript cDNA Synthesis Kit (Bio-Rad). Diluted (1:40) cDNA was subsequently used in qPCR analyses to determine mRNA levels. Primer sequences and efficiency of amplification for each set of validated primers are presented in Table 1. The qPCR was carried out with the following conditions: an initial denaturation at 95 °C for 5 min followed by 39 cycles at 95 °C for 15 s and 58 °C for 30 s for annealing, and 95 °C for 10 s. Relative mRNA expression data was analysed using the standard curve method. Data were normalised to the expression levels of two reference genes (B2m and Sdha) determined in each sample.
Primers used in real-time PCR studies.
|Gene||Forward primer||Reverse primer||Amplification efficiency (%)|
Protein extraction and immunoblot analyses
Following isolation of granulosa and luteal cells, 100 μl of lysis buffer (Laemmli, PBS, 2-mercaptoethanol) was added to each sample with 1 μl of each protease inhibitor: Mammalian Protease Arrest, Phosphatase Arrest III and EDTA (G Biosciences, St. Louis, MO, USA). The granulosa cell protein extracts were boiled at 95 °C for 10 min and were stored at −20 °C until immunoblot analysis.
Proteins were separated by electrophoresis using 10% SDS–PAGE gel, followed by transfer to nitrocellulose membrane. The membrane was blocked in 5% milk in Tris-buffered saline with 0.1% Tween-20 (TBS-T) for 45 min at room temperature, followed by overnight incubation at 4 °C with a primary antibody, mouse monoclonal anti-mouse Cebpβ (1H7) (1:2000, Abcam, Toronto, ON, Canada; cat no. ab15050). Following a TBST wash three times for 10 min, the membrane was incubated with horse anti-mouse (1:10 000, Cell Signalling, Danvers, MA, USA; cat no. 7076) secondary antibody for 1 h at room temperature. The proteins were detected by Immun-Star Western Chemi luminescent Kit (Bio-Rad). The membrane was exposed using the gel imaging ChemiDoc XRS+ System (Bio-Rad). Densitometry analyses of images were performed using Image Lab Software (Bio-Rad), in which protein quantification was done by comparison of CEBPβ protein isoforms relative to β-actin (Actb). Each of the five sets of immunoblots included samples from 0, 1, 4, 7 and 18 h post-hCG. Abundance of CEPBP was then normalised to hCG 0 h time point.
Comparative bioinformatics analysis of Lepr promoter
For comparative bioinformatic analysis, the Lepr promoter sequences of the mouse, human and rat were retrieved from UCSC Genome Browser. The genome assemblies from which the sequences obtained were Dec 2011 (GRCm38/mm10), Feb 2009 (GRCh37/hg19) and Nov 2004 (Baylor3.4/rn4) for mouse, human and rat respectively. The Lepr promoters analysed included sequences from −2000 bp to 5′-UTR. Sequences of each species were uploaded to online Patch Software (http://www.gene-regulation.com/cgi-bin/pub/programs/patch/bin/patch.cgi). The presence of putative Cebpβ binding sites was determined using the following settings: minimum length, 6; maximum mismatches, 1; mismatch penalty, 100 and lower score boundary, 87.5. The sequence of first 124 bases of 5′-UTR of Myod1 was also analysed for the presence of Cebpβ binding sites and this region was used as a negative control for chromatin immunoprecipitation (ChIP) analysis (described below).
ChIP analyses were performed as described previously (Duggavathi et al. 2008, Svotelis et al. 2011). Briefly, granulosa cells were crosslinked with 1% formaldehyde in PBS for 10 min at room temperature. The cells were then washed in PBS, resuspended in 200 μl of ChIP lysis buffer (1% SDS, 10 mm EDTA, 50 mm Tris–HCl (pH 8.0) and protease inhibitors) and sonicated to obtain chromatin fragments of 200–500 bp. The chromatin solution was diluted tenfold in ChIP dilution buffer and 5% of the diluted lysate was used for purification of total DNA (as input). The chromatin was precleared by incubating with 2 μg of salmon sperm DNA/protein A–agarose 50% gel slurry (Roche Diagnostics) for 2 h at 4 °C. Then the chromatin was divided into two parts and each part was incubated with Cebpβ antibody (ab32358) or normal rabbit IgG at 4 °C overnight. Next morning, DNA–protein crosslinks were reversed by incubating at 65 °C overnight followed by proteinase K treatment. DNA was purified with the QIAquick PCR purification column (Qiagen). QPCR was carried out with primers for positive (Lepr) and negative (Myod1) loci. Amplicon abundance was expressed as the percent of immunoprecipitated DNA relative to the input DNA.
Analyses were performed using SigmaPlot 12.3 Software, San Jose, CA, U.S.A. Significant differences between time points for relative mRNA expression data were analysed by one-way ANOVA followed by Tukey's multiple comparisons post-hoc test. Relative levels of protein were analysed by one-way ANOVA followed by Fisher's least significant difference (LSD) multiple comparisons post-hoc test. Ovulation data, ChIP data and mRNA data from PEG-SMLA experiment were analysed by unpaired Student's t-test. Normality of data was confirmed by Shapiro–Wilk test. All data are expressed as mean±s.e.m. A significance level of P<0.05 was used.
Expression profiles of LeprA, LeprB and Lep
In order to understand the regulatory mechanisms of Lepr signalling, we first profiled the expression patterns of its isoforms in purified granulosa and luteal cells during gonadotropin-induced follicular and luteal development. Of the six known isoforms, we examined the expression profile of the two isoforms that have been shown to transduce leptin signal, LeprA and LeprB (Baumann et al. 1996, Murakami et al. 1997). Relative levels of LeprA mRNA were low in granulosa cells from eCG-stimulated growing follicles. Their levels increased dramatically from 48 h post-eCG to 4 h post-hCG (P<0.01) and declined by 12 h post-hCG (P<0.01) followed by low levels through luteal phase (Fig. 1A). Relative mRNA levels of LeprB were also low during the follicular phase and increased at 4 h post-hCG (P<0.01), before declining to basal levels by 12 h post-hCG (Fig. 1B). Although LeprA and LeprB were both induced by hCG treatment, there were differences in the levels of their induction. LeprA expression was induced by a 23-fold increase, whereas LeprB by a 11-fold increase from eCG 0 to 4 h post-hCG. At 12 h post-hCG, LeprA mRNA levels were still tenfold higher relative to 0 h eCG. However, LeprB abundance at 12 h post-hCG was similar to the basal level at eCG 0 h. Taken together, these data indicate that mRNA abundance of LeprA was higher than LeprB in the granulosa cells of ovulating follicles (4 and 12 h post-hCG).
Regulation of Lepr expression: role of Cebpβ
As both isoforms of Lepr were dramatically upregulated during the periovulatory period and Cebpβ has been shown to regulate ovulatory events (Sterneck et al. 1997), we hypothesised that Cebpβ may regulate Lepr transcription. This hypothesis was tested using multiple approaches. First, comparative bioinformatics analyses of proximal promoters of the mouse, human and rat Lepr genes showed multiple potential binding sites for Cebpβ (Fig. 2). Specifically for the mouse promoter, this analysis identified three potential Cebpβ binding sites at −537, −579 and −1620 from transcription start site. Second, expression profile of Cebpβ in purified granulosa and luteal cells showed that the mRNA abundance significantly increased in granulosa cells during periovulatory period. Maximal expression occurred at 1 h post-hCG, with a significantly higher expression during the period of 1–7 h post-hCG compared with other time points (P<0.001; Fig. 3A). Third, we examined the protein levels of Cebpβ isoforms in the granulosa cells of periovulatory follicles. Cebpβ is expressed as three isoforms generated by differential initiation of translation (Cottrell & Mercer 2012), namely, a full-length isoform, a LAP (liver-enriched activator protein) isoform and a truncated LIP (liver-enriched inhibitory protein) isoform (Calkhoven et al. 2000). As LAP and LIP isoforms have been shown to have transcriptional activity (Descombes & Schibler 1991), we quantified their abundance. In line with the mRNA expression profile, the protein abundance of LAP (4 and 7 h post-hCG) and LIP (7 h post-hCG) isoforms was higher in periovulatory granulosa cells relative to granulosa cells at hCG-0 h time point (Fig. 3B).
With both comparative bioinformatic analysis and expression pattern supporting our hypothesis that Cebpβ could regulate Lepr expression, we used ChIP–qPCR assay to test if Cebpβ is recruited to the Lepr promoter in granulosa cells. We chose 2 h post-hCG time point based on the maximal Lepr expression in granulosa cells (Fig. 1A and B). As shown in Fig. 3C, immunoprecipitation of granulosa cell chromatin using Cebpβ antibody showed significant enrichment of the Lepr promoter region containing putative Cebp-binding sites as compared with normal rabbit IgG control. Furthermore, there was no enrichment of the 5′-UTR of Myod1 gene, which does not contain a Cebpβ binding site (Fig. 3C). These data confirm that Cebpβ is recruited to the Lepr promoter during the time of its peak mRNA abundance in the granulosa cells of ovulating follicles.
Inhibition of Lepr signalling results in reduced ovulation
Based on the expression pattern of Lepr and its induction by hCG through Cebpβ, we hypothesised that Lepr may be necessary for normal ovulation. To test this hypothesis, we used a recently developed Lepr antagonist, PEG-SMLA, to inhibit Lepr signalling during hCG-stimulated ovulation in superstimulated mice. The number of oocytes ovulated in oviducts collected at 18–20 h post-hCG in inhibitor-treated mice was decreased by 65% as compared with PBS-treated mice (P<0.05; Fig. 4A). Microscopic analyses showed that the oocytes of both PBS and PEG-SMLA-treated mice were in metaphase II stage (Fig. 4B).
Effect of PEG-SMLA on granulosa gene expression
In order to determine the molecular basis for reduced ovulation in PEG-SMLA-treated mice, we collected granulosa cells at 4 h post-hCG from mice treated with PEG-SMLA or PBS. This time point was chosen to investigate molecular perturbations caused by inhibition of Lepr signalling at its peak expression. Relative levels of hyaluronan synthase 2 (Has2) and Ptgs2 were significantly decreased in granulosa cells of PEG-SMLA-treated mice as compared with those of PBS-treated animals (P<0.05; Fig. 5B).
In this study, we showed that Lepr gene expression is dramatically induced at 4 h post-hCG in the granulosa cells of ovulating follicles. The increased mRNA expression of LeprA and LeprB isoforms in granulosa cells at 4 h post-hCG is in agreement with their increased expression at 9 h post-hCG in rat ovaries (Ryan et al. 2003). The pattern observed in the rat study using the whole ovary may therefore be a result of Lepr gene expression in granulosa cells. We also found that mRNA abundance of Lep was modestly but significantly induced by hCG, which is also in agreement with a previous study on rats (Ryan et al. 2003). Overall, these data on Lepr and Lep expression in granulosa cells suggest that leptin signalling is enhanced within the microenvironment of ovulating follicles in mice.
The data of this study also indicate that LeprA mRNA abundance is higher than LeprB the in granulosa cells of ovulating follicles. Although not many of the ovarian/granulosa cell studies consider both LeprA and LeprB isoforms, Ryan et al. (2003) showed that the LeprA (Ob-Ra) isoform was slightly higher as compared with the LeprB (Ob-Rb) isoform in rat ovaries. The bigger difference between the two isoforms in our study could be due to purified populations of granulosa. In fact, LeprA mRNA abundance has been shown to be dramatically higher than LeprB mRNA in multiple non-hypothalamic tissues including the ovary, uterus and pituitary in cattle (Thorn et al. 2007). Similarly, several other studies in mice have shown that LeprA is highly expressed with very minimal expression of LeprB in all tissues except the hypothalamus (Ghilardi et al. 1996, Fei et al. 1997, Chen et al. 1999). Thus, our data suggest that LeprA may have a predominant role in granulosa cells, especially during the periovulatory period.
Our data along with others (Duggal et al. 2002, Ryan et al. 2003) clearly demonstrate that Lepr gene expression is induced in granulosa cells of ovulating follicles. Thus, it is very interesting to study the mechanisms of transcriptional regulation of Lepr in granulosa cells of ovulating follicles. Our hypothesis was that Cebpβ could be one of the transcription factors regulating Lepr expression. This hypothesis was based on the reports that overexpression of Cebpβ the in Hep3B cells increased Lepr expression (Saint-Auret et al. 2011) and conditional deletion of Cebpα/β in granulosa cells abrogated hCG-induced expression of Lepr (Fan et al. 2011). Indeed, we found that Cebpβ was significantly induced at 1 h post-hCG, before Lepr induction at 4 h post-hCG, and comparative bioinformatic analyses predicted multiple Cebpβ binding sites on the Lepr promoter for the rat, human and mouse. Presence of the Cebpβ binding site in three species indicates that Cebpβ-driven regulation of Lepr is a well-conserved mechanism. Confirming our hypothesis, ChIP analysis using the chromatin from granulosa cells of ovulating follicles provides strong evidence for Cebpβ binding to the Lepr promoter during its peak expression. Therefore, the data in this study indicate that leptin signalling could be one of the pathways that Cebpβ regulates in granulosa cells of ovulating follicles and thereby ovulation (Sterneck et al. 1997).
Intriguingly, our immunoblot analyses showed that protein abundance of LAP isoform was significantly higher at 4 and 7 h post-hCG and LIP isoform at 7 h post-hCG compared with hCG-0 h time point. These isoforms are generated by alternative translation start sites (Calkhoven et al. 2000), and LAP and LIP isoforms are shown to have opposing transcriptional activity (Descombes & Schibler 1991). Overexpression of the LAP isoform activated Lepr expression and overexpression of LIP isoform inhibited Lepr expression in cultured Hep3B cells (Saint-Auret et al. 2011). In light of these results and our ChIP analysis, it is plausible that the LAP isoform of Cebpβ induces Lepr expression by 4 h post-hCG and the LIP isoform inhibits its expression by 7 h post-hCG. Unfortunately, it was impossible for us to test this hypothesis using ChIP assay as there is no commercially available ChIP-grade and isoform-specific antibody against Cebpβ. The one used in this study identifies both LAP and LIP isoforms.
Dramatic induction of Lepr isoforms by preovulatory hCG through Cebpβ and the importance of Cebpβ in ovulation (Sterneck et al. 1997, Fan et al. 2011) led us to hypothesise that Lepr could be essential for ovulation. The role of leptin in ovulation is still unresolved because of contrasting evidence with both reduced ovulation rate (Duggal et al. 2000, Ricci et al. 2006) and increased ovulation rate (Almog et al. 2001) in leptin-treated rats. Therefore, we took a complementary loss-of-function approach to inhibit Lepr signalling in the ovary. We used a recently developed leptin antagonist (Shpilman et al. 2011) to antagonise leptin action in vivo in immature mice during the superstimulation protocol. Administration of PEG-SMLA has been shown to increase body weight in αMUPA mice, which are otherwise resistant to obesity due to high leptin concentrations (Chapnik et al. 2013). Therefore, the PEG-SMLA was ideal for our in vivo model. The antagonist at a dose of 10 μg/g, administered at hCG stimulation, significantly reduced ovulation rate in immature superovulated animals. Previously, similar doses were found to induce a more significant increase in weight gain in mice (Shpilman et al. 2011).
Although ovulation rate was decreased by over 65%, meiotic maturation of the ovulated oocytes was unaltered as evidenced by a similar number of oocytes in metaphase II stage in PBS and PEG-SMLA-treated mice. However, leptin treatment of bovine oocytes in vitro (1–10 ng/ml) increased the rate of development in metaphase II (Paula-Lopes et al. 2007, Jia et al. 2012). Leptin treatment was found to increase the rate of meiotic resumption in the murine ovary (Barash et al. 1996, Ryan et al. 2002). In our study, inhibition of Lepr using PEG-SMLA impaired only follicle rupture without affecting meiotic maturation of oocytes. While we are unable to explain the reason for the lack of effect of PEG-SMLA on meiosis, these data warrant further in vitro analysis of oocyte maturation using PEG-SMLA.
Though leptin has previously been shown to affect the ovulation process both negatively and positively (Almog et al. 2001, Ricci et al. 2006), through its actions at the ovarian level, the molecular mechanisms have never been explored. In light of the peak expression of LeprA and LeprB at 4 h post-hCG, which is a very important time point for the expression of ovulation-related genes, we hypothesised that Lepr may play an important role in hCG (LH)-driven gene expression. Indeed, our data show that PEG-SMLA treatment at the time of hCG stimulation resulted in reduced expression of Ptgs2 and Has2. It is very well established that both Ptgs2 and Has2 play indispensable roles in terminal differentiation of granulosa cells of the ovulating follicle (Duggavathi & Murphy 2009, Richards & Pangas 2010). In line with our data, a previous in vivo study showed that leptin induces Ptgs2 expression in the rat brain (Inoue et al. 2006). Also several in vitro studies have shown that leptin increases Ptgs2 expression in endothelial cells (Manuel-Apolinar et al. 2013), renal tubular cells (Hsu et al. 2012) and endometrial cells (Gao et al. 2009). Notwithstanding this, other studies have shown either no effect (Wazir et al. 2012) or inhibition (Tsai et al. 2006) of Ptgs2 expression in response to leptin treatment. These studies show that when the role of leptin signalling is examined using leptin-treatment approach, the results appear to depend on the dose of leptin used. Indeed it has been shown that high-dose acute treatment with leptin inhibits (Duggal et al. 2000), while low-dose chronic treatment enhances (Almog et al. 2001) ovulation rate in rats. Therefore, at basal levels of leptin, Lepr may positively regulate hCG-induced expression of Ptgs2. Further support for Lepr regulation of Has2 comes from a study (van Tol et al. 2010) that showed a positive correlation between Lepr and cumulus expansion genes HAS2 and PTX3 in human cumulus cells. Taken together, our data suggest that Lepr plays an important role in LH-induced ovulatory process, at least in part, through regulation of Ptgs2 and Has2.
Overall, the data in this study contribute to our further understanding of leptin regulation of ovarian function. Ovaries of the db/db mouse, which expresses all isoforms except LeprB, are normal (Bahary et al. 1990), demonstrating that direct actions of leptin on the ovary may involve other isoforms of Lepr. This inference is further supported by the fact that neuron-specific replacement of LeprB rescues fertility in db/db mice lacking only LeprB (de Luca et al. 2005) but not in db3j/db3j mice lacking all Lepr isoforms in all tissues including the ovary (Kowalski et al. 2001). In light of these reports, our data indicating a more drastic increase in expression of the LeprA isoform in granulosa cells and disruption of normal ovulation due to leptin antagonist treatment suggest that LeprA isoform may be more important for the regulation of ovulation by leptin signalling. While our pharmacological approach revealed the potential role of leptin signalling in ovulation, additional studies are needed for further understanding.
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.
This research was supported by funds from NSERC (Rgpin 371850-09), FQRNT (2011-NC-137832) and McGill-CSR (seed grant) to R Duggavathi. CSR graduate student fellowship for L Dupuis and RQR scholarship for Y Schuermann, A Kalaiselvanraja and D Siddappa.
The authors thank Dr Silvana Obici (University of Cincinnati) for donating LeprB primers and Dr Timothy Keiffer (University of British Columbia) for help with procurement of PEG-SMLA.
Manuel-ApolinarLLopez-RomeroRZarateADamasioLRuizMCastillo-HernandezCGuevaraGMera-JimenezE2013Leptin mediated ObRb receptor increases expression of adhesion intercellular molecules and cyclooxygenase 2 on murine aorta tissue inducing endothelial dysfunction. International Journal of Clinical and Experimental Medicine6192–196.
Saint-AuretGDananJLHironMBlacheCSulpiceETendilSDaveauMGidrolXSalierJP2011Characterization of the transcriptional signature of C/EBPβ isoforms (LAP/LIP) in Hep3B cells: implication of LIP in pro-survival functions. Journal of Hepatology541185–1194. (doi:10.1016/j.jhep.2010.09.021)
TsaiEMChanTFChangYChiangPHChuangCYLongCYChaiCYLeeJN2006Leptin suppresses human chorionic gonadotropin-induced cyclooxygenase-2 expression and prostaglandin production in cultured human granulose luteal cells. Journal of the Society for Gynecologic Investigation13551–557. (doi:10.1016/j.jsgi.2006.09.004)
ZachowRJMagoffinDA1997Direct intraovarian effects of leptin: impairment of the synergistic action of insulin-like growth factor-I on follicle-stimulating hormone-dependent estradiol-17β production by rat ovarian granulosa cells. Endocrinology138847–850. (doi:10.1210/en.138.2.847)