Skip to main content
NCBI home page
Molecular Human Reproduction logoLink to Molecular Human Reproduction
. 2015 Apr 15;21(6):527–534. doi: 10.1093/molehr/gav017

A role for retinoids in human oocyte fertilization: regulation of connexin 43 by retinoic acid in cumulus granulosa cells

Monica W Best 1, Juanjuan Wu 1, Samuel A Pauli 1, Maureen A Kane 2, Keely Pierzchalski 2, Donna R Session 1, Dori C Woods 3, Weirong Shang 1, Robert N Taylor 4, Neil Sidell 1,*
PMCID: PMC4447995  PMID: 25877907

Abstract

Retinoids are essential for ovarian steroid production and oocyte maturation in mammals. Oocyte competency is known to positively correlate with efficient gap junction intercellular communication (GJIC) among granulosa cells in the cumulus-oocyte complex. Connexin 43 (Cx43) is the main subunit of gap junction channels in human cumulus granulosa cells (CGC) and is regulated by all-trans retinoic acid (ATRA) in other hormone responsive cell types. The objectives of this study were to quantify retinoid levels in human CGC obtained during IVF oocyte retrievals, to investigate the potential relationship between CGC ATRA levels and successful oocyte fertilization, and to determine the effects of ATRA on Cx43 protein expression in CGC. Results showed that CGC cultures actively metabolize retinol to produce ATRA. Grouped according to fertilization rate tertiles, mean ATRA levels were 2-fold higher in pooled CGC from women in the highest versus the lowest tertile (P < 0.05). ATRA induced a rapid dephosphorylation of Cx43 in CGC and granulosa cell line (KGN) cultures resulting in a >2-fold increase in the expression of the functional non-phosphorylated (P0) species (P < 0.02). Similar enhancement of P0 by ATRA was shown in CGC and KGN cultures co-treated with LH or hCG which, by themselves, enhanced the protein levels of Cx43 without altering its phosphorylation profile. Correspondingly, the combination of ATRA+hCG treatment of KGN caused a significant increase in GJIC compared with single agent treatments (P < 0.025) and a doubling of GJIC from that seen in untreated cells (P < 0.01). These findings indicate that CGC are a primary site of retinoid uptake and ATRA biosynthesis. Regulation of Cx43 by ATRA may serve an important role in folliculogenesis, development of oocyte competency, and successful fertilization by increasing GJIC in CGC.

Keywords: all-trans retinoic acid, connexin 43, cumulus granulosa cells, oocyte competency

Introduction

Emerging evidence indicates important effects of vitamin A compounds (retinoids) on oocyte maturation and function (Zheng et al., 1999; Livera et al., 2000; Brown et al., 2003; Ikeda et al., 2005). To this end, in vivo administration of retinoids in cattle, sheep and pigs have been shown to enhance oocyte fertilization competence (Eberhardt et al., 1999; Whaley et al., 2000; Ikeda et al., 2005) while in vitro studies identified all-trans retinoic acid (ATRA) as the active vitamin A metabolite in this activity (Alminana et al., 2008; Nasiri et al., 2011; Tahaei et al., 2011). Given the demonstrated importance of ATRA in oocyte competency in animals, we have been interested in providing evidence of its role and mechanism of action in human oocyte development. Our previous work demonstrated that high follicular fluid (FF) ATRA concentrations at the time of oocyte retrieval were associated with oocytes giving rise to the highest quality (grade I) embryos in our in vitro fertilization (IVF) program (Pauli et al., 2013). In that investigation, follicular fluid ATRA concentrations also were shown to correlate with patient plasma levels. As the relative contribution of plasma to follicular fluid ATRA is unknown, we evaluated active intra-follicular biosynthesis of ATRA. In this regard, as far as we know, active synthesis of ATRA by ovarian follicle cells has not been reported.

The mechanism(s) by which retinoids may affect oocyte development has received only scant attention. It was suggested that ATRA may promote cytoplasmic maturation of bovine oocytes via its modulatory effects on the gene expression of gonadotrophin receptors, midkine (neurite outgrowth-promoting factor 2), cyclooxygenase-2, and/or nitric oxide synthase in cumulus granulosa cells (CGC) (Ikeda et al., 2005). We have previously shown that ATRA can rapidly up-regulate gap junction intercellular communication (GJIC) in human endometrial stromal cells through a post-translational mechanism of action that involves rapid dephosphorylation of gap junction alpha-1 protein (GJA1) also known as connexin 43 (Cx43) (Wu et al., 2013). In ovarian granulosa cells, Cx43 is the most widely expressed Cx, and plays an important role in regulating the micronutrient environment of the oocyte by allowing transfer of ions, metabolites, and small molecules up to 1 kDa (Goldberg et al., 2004; Wang et al., 2009). Multiple animal studies have demonstrated the regulation of ovarian Cx43 by luteinizing hormone (LH); exposure of pre-ovulatory follicles to LH deactivates gap junctions through induced phosphorylation of Cx43 (Sela-Abramovich et al., 2005; Norris et al., 2008). This activity prevents the intercellular transport of cAMP, a meiosis inhibitor, leading to resumption of meiosis (Dekel, 2005). The importance of this activity was highlighted by showing that Cx43 knockout mice have impaired folliculogenesis as the follicles are unable to proceed beyond the pre-antral follicular stage (Ackert et al., 2001). Mechanisms that reestablish GJIC in the corpus luteum, an action that is necessary for progesterone production, are still unresolved although cAMP and oxytocin are thought to co-operate with LH/hCG signaling in this process. However, the rapid increase of GJIC during the early postovulatory luteal phase (Grazul-Bilska et al., 1996b, c) suggests the involvement of other, as yet unknown, regulatory co-factors that can quickly modulate Cx43 function. In light of the importance of retinoids in oogenesis and the recognized action of ATRA on Cx43, we were interested in the possibility that ATRA functions in the post-meiotic follicle through its ability to enhance GJIC in human cumulus granulosa cells. As such, our objectives for this study were to (i) quantify retinoid levels in human CGC; (ii) investigate the potential relationship between CGC ATRA levels and successful oocyte fertilization; and (iii) determine the effects of ATRA on Cx43 in CGC and in the immortalized KGN granulosa cell line. The findings reported herein provide evidence of a novel role for ATRA in regulating CGC-oocyte interactions during human ovarian folliculogenesis.

Materials and Methods

Experimental participants and collection of CGC

CGC were obtained from patients undergoing fresh IVF at the Emory Reproductive Center. Written informed consent was obtained from all participants prior to the procedure. Retinoid levels in CGC were measured in samples collected from participants of a previous study conducted between August 2008 and July 2009 (Pauli et al., 2013). Subsequent CGC used for the in vitro studies of ATRA effects on Cx43 were obtained from new patients enrolled between November 2011 and March 2013. Each participant's cycle parameters were collated from clinical records as previously reported (Pauli et al., 2013). All patients underwent gonadotrophin stimulation administered by subcutaneous injection. Serial E2 levels and 2-dimensional transvaginal ultrasound follicle measurements were performed until at least two follicles reached 18 mm or more in diameter. Final oocyte maturation was induced by administering hCG (10 000 IU intramuscular) followed by transvaginal follicle aspiration and oocyte retrieval 35 h later.

Primary human CGC were collected from women undergoing IVF by gently removing them from the cumulus-oocyte complex following egg retrieval. CGC from all cumulus-oocyte complexes retrieved from each patient were pooled in order to obtain enough material for retinoid analysis.

Human endometrial and mammary tissues were used for comparison of retinoid levels with CGC. Endometrial biopsies were obtained from patients (n = 42) undergoing surgery for infertility or pelvic pain as part of a previously published study (Pierzchalski et al., 2014). Normal mammary tissue was obtained from the Marlene and Stewart Greenebaum Cancer Center tissue bank (n = 9).

Cell cultures and chemicals

The human granulosa cell line KGN, originally provided to us by Drs Nishi and Yanase (Graduate School of Medical Sciences, Kyusyu University, Fukuoka, Japan), was derived from a patient with invasive ovarian granulosa cell carcinoma (Nishi et al., 2001). This cell line has been used as a model for human granulosa cell steroidogenesis, cell growth and apoptosis (Nishi et al., 2001; Woods et al., 2008). Primary CGC and KGN cells were cultured in 6 well plates in Dulbecco's Modified Eagle Medium/F12 media (Cellgro, Manassas, VA, USA) containing 1% insulin-transferrin-selenium, 1% penicillin/streptomycin, 10% fetal bovine serum (complete medium) for 24 h prior to treatment. For evaluating Cx43 expression, cells were treated for up to 72 h with all-trans retinoic acid (ATRA), LH, hCG (all from Sigma Chemical Co., St. Louis, MO, USA) or in combinations as indicated. A 50 mM stock of ATRA was made in dimethyl sulphoxide (DMSO), aliquoted, and stored frozen at −20°C for up to 2 months for future use. For experiments, ATRA stock was further diluted in complete medium to the concentrations indicated. Final solvent concentrations (<0.02% DMSO) were contained in the vehicle controls. To assess the ability of CGC to synthesize ATRA, cells were washed with phosphate-buffered saline (PBS) and then cultured in serum-free medium in the presence of 2 µM retinol (ROL, Sigma Chemical Co., St. Louis, MO, USA) or solvent control for 16 h. ATRA concentration was then quantified in the culture supernatant and cell pellets (below).

Western blot analysis

Western blot analysis was performed as previously described (Wu et al., 2013). Briefly, whole cell extracts of CGC and KGN were derived by dissolving cell pellets in Cell Extraction Buffer (Invitrogen, Grand Island, NY, USA) and protein concentrations determined using a bicinchoninic acid (BCA) protein assay kit (Sigma Chemical Co.). Protein (20 µg) from cells treated with medium (control), ATRA, LH, hCG, and combinations were loaded on 10% SDS–PAGE gel, transferred to nitrocellulose membrane and blocked with 4% bovine albumin in PBS. Rabbit polyclonal anti-Cx43 antibody (1:500, Zymed, Grand Island, NY, USA) and sequential secondary antibody linked with horseradish peroxidase was used to detect total Cx43. The non-phosphorylated form of Cx43 was detected using mouse monoclonal anti-Cx43 antibody (1:1,000, Invitrogen, Cat. No. 13-8300). Enhanced Chemiluminescence System (Amersham Biosciences Piscataway, NJ, USA) was utilized to visualize immunoreactive bands. Even loading was ensured by washing, stripping and reprobing blots with anti-β-actin antibody.

Scrape loading/dye transfer

Relative GJIC in control and treated cultures was determined by the Scrape Loading/Dye Transfer (SL/DT) technique previously described (Tanmahasamut and Sidell, 2005; Wu et al., 2013) by using the gap junction-permeable fluorescent dye Lucifer Yellow (LY) (Molecular Probes, Life Technologies, Grand Island, NY, USA). KGN cells were cultured as described above and were washed with PBS before performing the assay. The SL was done by making a razor blade cut across the cell monolayer and then 1 mg/ml LY was added to the culture at the area of the defect. Dye was rinsed after 10 min. Cells were washed three times with PBS and fixed with 4% paraformaldehyde. Fluorescence emission with an inverted microscope equipped with a camera detected the lateral diffusion of LY into cells with GJIC. Cells that received LY from an adjacent scrape-loaded cell were considered communicating and the numbers of communicating (fluorescent) cells in the untreated and treated cultures were counted. For each condition, five randomly selected fields were counted. GJIC was reported as the number of LY positive cells.

Determination of retinoid levels

Biopsy specimens, cultured cell pellets or culture supernatant were prepared for retinoid analysis under yellow lights. In the case of CGC, cells from all cumulus-oocyte complexes retrieved from each patient were pooled in order to obtain enough material for analysis. Retinoids were extracted with a two-step liquid-liquid extraction that has been described in detail (Kane et al., 2008a, Kane and Napoli, 2010). Total protein concentrations were determined using the Bradford method (Bio-Rad Laboratories, Hercules, CA, USA). Retinoic acid was quantified by liquid chromatography-tandem mass spectrometry (LC–MS/MS) with atmospheric pressure chemical ionization in positive-ion mode on a 5500 QTRAP hybrid triple quadrupole linear ion trap mass spectrometer operated in MRM mode (AB Sciex, Foster City, California) (Kane et al., 2008b). Retinol and total retinyl ester (RE) were quantified by high-performance liquid chromatography with ultraviolet detection (HPLC-UV) on an Aquity H-Class UPLC (Waters, Milford, MA, USA) (Kane et al., 2008a).

Statistical analysis

Statistical analysis was performed using GraphPad Software (San Diego, CA, USA). Data are presented as mean ± standard error of the mean (SEM). Differences between groups as indicated were analyzed by t-test (unpaired, two-tailed), where P < 0.05 was considered statistically significant. Each experiment was replicated a minimum of three times.

Results

Retinoid levels in cumulus granulosa cells

To quantify retinoid levels in primary CGC, the cells were obtained from 77 IVF participants that were part of a group of 79 IVF patients whose peripheral plasma and follicular fluid were analyzed in a previous study (Pauli et al., 2013). Insufficient cumulus granulosa cells were obtained from two patients in the previous work to allow for analysis. Mean ATRA, ROL and RE levels are shown in Table I. The concentrations of ATRA and ROL were determined to be higher than those found in primary human endometrial and mammary cells used as sources of hormone sensitive tissue known to actively synthesize ATRA from ROL (Pierzchalski et al., 2014). 13-cis RA was not detected (ND) in 66 of the samples and had an average concentration of 94.5 ± 38.4 pmol/g protein in the remaining 11 samples. Neither was 9-cis RA detected in any of the patient samples. Limits of detection for both 13-cis and 9-cis RA in tissue were ∼12 pmol/g protein (Kane et al., 2008b). As expected, ROL and RE were more abundant than ATRA, falling in the nmol/g protein range. ATRA levels ranged from 127 to 7552 pmol/g protein while ROL levels ranged from 23.2 to 3915.2 nmol/g protein. RE levels ranged from 49.0 to 337.3 nmol/g protein. CGC ATRA concentrations from the same women correlated weakly with ROL levels (r2 = 0.41) but did not correlate with RE levels (r2 = 0.00). There was also no correlation between granulosa ATRA and plasma ATRA levels (r2 = 0.02).

Table I.

Mean retinoid levels in cumulus granulosa cells, endometrial and mammary tissue. Data are mean (SEM).

Granulosa cells
N = 77		 Endometrial tissue
N = 42		 Mammary tissue
N = 9
All-trans retinoic acid (pmol/g protein) 1140 (176) 330 (43) 880 (96)
Retinol (nmol/g protein) 603 (108) 55 (12) 171 (26)
Retinyl esters (nmol/g protein) 216 (81) 56 (10) 213 (32)
Open in a new tab

Oocyte fertilization rate

Previous studies showed the positive effects of vitamin A supplementation on bovine fertility in artificial insemination while in vitro oocyte studies provide evidence of a role for retinoids in oocyte cytoplasmic maturation and fertilization competence (Ikeda et al., 2005; Alminana et al., 2008; Nasiri et al., 2011; Tahaei et al., 2011). In our previous work, we determined that follicular fluid from follicles yielding high quality (grade I) had higher mean levels of ATRA but not ROL than fluid from follicles that yielded nongrade I embryos (Pauli et al., 2013). In the present study, we sought to assess the potential relationship between retinoid levels in CGC and the fertilization competence of that subject's oocytes. However, a direct evaluation of this relationship in individual follicles was not possible due to insufficient cumulus material necessary for retinoid measurements in single-follicle CGC. Thus, to address this question, retinoid levels in CGC pooled from all retrieved oocytes from each patient were measured and results stratified by tertiles according to the percent of successfully fertilized oocytes (i.e. the number of oocytes that were successfully fertilized compared to the total retrieved from each patient; #2PN/total retrieved × 100). Figure 1 shows that the mean ATRA level was significantly (>2-fold) higher in the CGC pool from patients in the highest fertilization tertile compared with those in the lowest tertile. In contrast, there were no significant differences in ROL concentrations across the different fertilization tertiles. The clinical characteristics of known confounding factors between the highest and lowest fertilization groups were not significantly different, and both groups had similar stimulation parameters and response to gonadotrophins (Table II). Furthermore, there were no differences in CGC ATRA levels in patients with varying infertility diagnosis (unexplained, tubal or male factor infertility, polycystic ovarian syndrome, endometriosis, diminished ovarian reserve) although this stratification lacked adequate power with the sample size studied.

Figure 1.

Figure 1

Open in a new tab

Retinol and all-trans retinoic acid (ATRA) concentrations in patient pooled cumulus granulosa cells (CGC) stratified by tertiles according to percent successfully fertilized (2 pro-nuclei; 2PN) oocytes (i.e. #2PN/total oocytes retrieved from each patient × 100). Values represent mean ± SEM of retinoid concentrations of CGC in the indicated tertile. *Significant difference compared with ATRA levels in the lowest tertile, P < 0.05. Numbers of patients (range of fertilization rates) in each tertile: n = 27 (18–53) in lowest; n = 24 (54–69) in middle; n = 26 (70–100) in highest.

Table II.

Clinical factors associated with fertilization rates.1

Risk factors Highest2, N = 26 Lowest2, N = 27 P-value3
ATRA (pmol/g protein) 1693.3 (466.7)4 715.8 (102.4) 0.04
Age (years) 34.4 (0.9) 34.2 (0.9) 0.92
BMI (kg/m2) 26.1 (1.3) 26.2 (1.5) 0.97
Peak E2 (pg/ml) 1702.6 (157.0) 2141.3 (204.5) 0.10
Total gonadotrophins (IU) 3659.6 (395.8) 2862.5 (305.7) 0.12
Antral follicle count (AFC) 19.0 (2.4) 25.9 (3.7) 0.13
Open in a new tab
1

Defined as the number of oocytes that were successfully fertilized (#2PN) compared to the total retrieved from each patient (#2PN/total retrieved).

2

Highest and lowest tertile groups according to percent 2PN.

3

Highest versus lowest tertile.

4

Mean (SEM).

Synthesis of ATRA by CGC

ATRA production was quantitated in CGC cultures by the addition of 2 µM ROL to the culture medium under serum-free conditions. After an additional 16 h (overnight) culturing, the medium and cells were harvested and analyzed for ATRA (Fig. 2). While little ATRA was detected in the absence of added ROL, treatment of the cultures with ROL induced a >3-fold increase in ATRA concentration in both the cell pellets and supernatant. These findings indicate that CGC both synthesize and actively secrete ATRA.

Figure 2.

Figure 2

Open in a new tab

All-trans retinoic acid (ATRA) production in cumulus granulosa cells (CGC). ATRA levels in supernatant (open bars; pmol/ml) and cell pellets (gray bars; pmol/g protein) were assessed by liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis following addition of 2 µM retinol (ROL) for 16 h under serum-free conditions. Values represent the mean ± SEM of ATRA concentrations in CGC cultures (n = 3 patient samples) where (*) indicates a significant increase compared with ROL-free conditions, P < 0.025.

Effects of ATRA on Cx43

We examined the effects of ATRA on Cx43 mRNA levels and protein expression in primary CGC. Dose–response experiments showed only a marginal (<50%) increase in mRNA levels at concentrations up to 10 µM ATRA and incubation times up to 3 days (data not shown). Figure 3 indicates that CGC exhibited three distinct immunoreactive Cx43 protein bands (41, 42 and 45 kDa; designated P0, P1 and P2, respectively). As shown by us and others in various human tissues, the lower band (P0) corresponds to the non-phosphorylated and most biologically active form of Cx43 (Tanmahasamut and Sidell, 2005; Solan and Lampe, 2007; Wu et al., 2013). Treatment of these cells for 2 days with 10 µM ATRA induced a change in the relative levels of the Cx43 species; P0 is enhanced, whereas P2 shows a dramatic decrease compared with control cultures (Fig. 3A, inset). Using the human granulosa cell line KGN as a model of primary CGC (Nishi et al., 2001; Woods et al., 2008), we observed similar results. The shift in band intensities induced by ATRA to the biologically active non-phosphorylated species was quantified by expressing the relative intensity of the three Cx43 species as the ratio (P1 + P2)/P0 (Fig. 3A). Using a non-phosphorylated Cx43 specific antibody, we confirmed that this P0 isoform of Cx43 was increased >2-fold by ATRA treatment (Fig. 3B).

Figure 3.

Figure 3

Open in a new tab

Effect of all-trans retinoic acid (ATRA) on Cx43 phosphorylation. Total cellular protein was isolated from cumulus granulosa cells (CGC) and human granulosa cell line (KGN) cells and assayed for Cx43 expression by western blotting. (A) Representative experiments of primary CGC from two patients and KGN cells treated with 10 µM ATRA (RA) or vehicle control (C) as indicated for 48 h show changes in the band intensity and distribution of non-phosphorylated (P0) and phosphorylated (P1 and P2) species of Cx43 (inset). Figure in the lower panel represents the relative intensity of the three Cx43 phosphorylation species expressed as the ratio (P1 + P2)/P0 in order to quantify the sensitivity of Cx43 dephosphorylation to ATRA treatment. Columns represent the mean (±SEM) from seven CGC patient samples (black bars) and three independent experiments with KGN cells (gray bars). Significant difference when compared with untreated controls, *P < 0.02; **P < 0.0001. (B) Representative western blot analysis using a non-phosphorylated Cx43 (Non-P Cx43)-specific antibody showing the effects of ATRA on Non-P Cx43 in CGC from three patients. Culture conditions were as in part (A). Human endometrial stromal cells (HESC) were previously determined to show up-regulation of non-phosphorylated Cx43 by ATRA (Tanmahasamut and Sidell, 2005; Wu et al., 2013) and were used as positive control cells for this effect. Quantification of six patient samples showed a mean (±SEM) fold increase of 2.31 ± 0.39 induced by ATRA (P < 0.02).

Combined effects of ATRA and LH/hCG on Cx43 expression

Animal studies have demonstrated modulation of Cx43 in granulosa cells by LH or hCG; exposure of the ovarian follicle to LH leads to an increase in the phosphorylation of Cx43 on serines 255, 262, and 279/282, resulting in an increase of Cx43 with mobility corresponding to the P2 position and inhibitory effects on GJIC (Norris et al., 2008). In our human CGC, LH was found to time-dependently enhance the overall protein levels of Cx43 without significant changes in the relative distributions of the three phosphorylation species (Fig. 4A). Similar results were obtained when CGC were stimulated with hCG, consistent with the fact that both gonadotrophins utilize the same transmembrane receptor (the luteinizing hormone/choriogonadotropin receptor, LhCGR) and activate identical signaling pathways (Menon et al., 2004). The CGC often showed intense P2 Cx43 bands even under control culture conditions, such that increased expression by LH and hCG was not always apparent (Fig. 4B). This may be due to the fact that all the patients were treated with hCG 35 h before retrieval as part of their IVF protocol, so that Cx43 expression in CGC may have been near maximally stimulated in vivo. Addition of ATRA treatment, both in the absence and presence of LH or hCG, induced a marked shift in the relative levels of the phosphorylated Cx43 species, as evidenced by a dramatic enhancement of P0 with a corresponding decrease in P2 (Fig. 4B). Similar results were obtained with KGN cells although up-regulation of Cx43 by hCG alone was more modest than that seen with primary CGC (Fig. 4C). Whether this difference is due to a relative paucity of LhCGR on KGN cells is unknown.

Figure 4.

Figure 4

Open in a new tab

Western blot analysis showing the combined effects of all-trans retinoic acid (ATRA) and LH or hCG on Cx43 phosphorylation. (A) Cumulus granulosa cells (CGC) were cultured for up to 72 h with vehicle control or LH (2.2 IU/ml) added at different time points before the end of the culture period as indicated (representative experiment of four patient samples). Similar enhancements of Cx43 protein levels without alteration of relative intensities of the three phosphorylation species were seen in cultures treated with LH for >24 h (not shown). (B) Representative experiment of five patient samples where CGC were treated with 10 µM ATRA (RA), 80 IU/ml hCG, 2.2 IU/ml LH, or combinations as indicated for 48 h. Western blot analysis was performed for both total Cx43 and non-phosphorylated Cx43 (Non-P Cx43) using a Non-P Cx43 specific antibody. (C) Representative analysis of five independent experiments showing total Cx43 in KGN cells treated for 48 h with 80 IU/ml hCG, 10 µM ATRA (RA), or hCG + ATRA as indicated.

Modulation of GJIC

Previous studies in endometrial stromal cells showed that ATRA increases GJIC through dephosphorylation of Cx43 as reflected by a similar decrease in the intensity of the P2 species and a reciprocal increase in P0 (Tanmahasamut and Sidell, 2005; Wu et al., 2013). To determine the extent to which Cx43 expression and phosphorylation status is related to GJIC in CGC, we performed SL/DT assays in KGN cells using the gap junction-permeable fluorescent dye LY. As assessed by this technique, serum-starved cells were communication-competent and transferred LY to numerous cells distant to the wound edge (Fig. 5). Figure 5E shows that ATRA- or hCG- treatment of KGN caused a 50% increase in LY positive cells compared with controls while combination of ATRA + hCG caused a further significant increase from single agent treatments, yielding a doubling of the number of LY positive cells compared with controls.

Figure 5.

Figure 5

Open in a new tab

All-trans retinoic acid (ATRA) and hCG effects on gap junction intercellular communication (GJIC). Human granulosa cell line (KGN) cells were treated with (A) vehicle, (B) 10 µM ATRA, (C) 80 IU/ml hCG or (D) ATRA + hCG for 48 h, after which GJIC was assessed by the Scrape loading (SL)/dye transfer (DT) technique as described in Materials and Methods. (E) Mean (±SEM) of five determinations that quantified cell numbers showing positive dye staining.

Discussion

Although concentrations of retinol and expression of retinoid-metabolizing enzymes in granulosa cells from animals have previously been assessed (Zheng et al., 1999; Ikeda et al., 2005; Kipp et al., 2011), a determination of retinoids or its signaling or metabolic pathway in human granulosa cells has not, to our knowledge been reported. Our results showed high levels of retinoids in CGC and active production of ATRA from its retinol substrate. As a comparison, we also quantified retinoid levels in human endometrium and mammary glands, tissues previously shown to actively synthesize ATRA from ROL (Pierzchalski et al., 2013, 2014). In the endometrium, the action of ATRA is essential for proper decidualization (Zheng and Ong, 1998) while in the mammary gland, ATRA signaling plays an essential role in ductal morphogenesis (Wang et al., 2005). Our results showing higher concentrations of retinoids in CGC compared with these other ROL metabolizing tissues (Table I) support the contention that CGC also are a primary target of retinoid uptake and an important source of ATRA biosynthesis. Our previous work demonstrated that high follicular fluid ATRA concentrations at the IVF retrieval were associated with oocytes giving rise to higher quality (grade I) embryos (Pauli et al., 2013). We now show that those patients yielding the highest percentage of successfully fertilized oocytes had significantly higher mean levels of ATRA in their CGC. Our analysis did not show significant differences in retinol levels between the groups supporting the contention that increased retinoid biosynthesis, rather than uptake of the retinol precursor, may be important for influencing oocyte fertilization competency. The finding that ATRA is actively produced and secreted from CGC suggests that the concentrations previously determined in follicular fluid at the time of oocyte retrieval may, at least partially, reflect retinoid biosynthesis by CGC, which, as shown here, is positively associated with the most successful fertilization rates. This conclusion proposes that high follicular fluid ATRA levels associated with grade I embryos is a consequence of efficient retinoid metabolism within that follicle's CGC, which in turn is important for production of competent oocytes.

Ovarian folliculogenesis and the production of fertilizable oocytes depend on GJIC within both the developing and the mature follicle (Khan-Dawood et al., 1998; Johnson et al., 2002; Borowczyk et al., 2007). Many connexins are expressed within the oocyte-granulosa cell complex with Cx43 being the primary connexin in CGC (Wang et al., 2009) as well as in mural granulosa cells (Kalma et al., 2004). Work over the last 30 years has suggested a well-orchestrated sequence of events mediated by GJIC between the CGC and oocyte that results in successful oocyte maturation and ovulation (Kidder and Mhawi, 2002; Senbon et al., 2003; Kalma et al., 2004; Kidder and Vanderhyden, 2010). During follicle growth, gap junction coupling between CGC and oocytes play an essential role in transferring nutrients to the oocyte necessary for its development and to maintain the differentiated state of granulosa cells by preventing them from premature luteinization. In the pre-ovulatory follicle, the LH surge then triggers deactivation of gap junctions through Cx43 phosphorylation, resulting in resumption of meiosis, extrusion of the first polar body and creation of a mature oocyte. In contrast to this negative regulation of Cx43 in the pre-ovulatory ovarian follicle, LH has been reported to have a stimulatory effect on Cx43 production in the postovulatory follicle upon the development of mural granulosa cells into granulosa lutein cells where an increase in non-phosphorylated Cx43 is noted 1–5 days after the midcycle LH surge (Khan-Dawood et al., 1998). In the developed corpus luteum, gap junctions mediated by Cx43 are critical for efficient progesterone production by luteal cells (Borowczyk et al., 2007). Animal studies have indicated that cumulus cells, prior to ovulation express little if any LhCGR (Diaz et al., 2007). However, studies of material from human subjects undergoing IVF have shown a similar regulation of LhCGR expression in CGC as occurs in luteinised granulosa cells (Haouzi et al., 2009; Jeppesen et al., 2012; Maman et al., 2012; Papamentzelopoulou et al., 2012). It has also been demonstrated that, while in vivo, CGC ovulate and have no steroidogenic function, post-IVF cumulus cells in culture spontaneously luteinize and produce large amounts of progesterone (Hughes et al., 1990). These cells also respond to estrogenic agents in a manner previously seen with mural granulosa cells (Hughes et al., 1990). Thus, it appears that regardless of their origin, stimulation of CGC by an LH or an hCG surge during ovulation induction leaves them with the same capability as mural granulosa cells for steroidogenesis. Based on the timing of our in vitro experiments with relation to the hCG administration before egg retrieval (cultures evaluated 3–5 days post ‘pharmacologic’ LH surge), we conjecture that the primary CGC evaluated in our studies correspond to an early luteal phase phenotype (1–5 days post LH surge). Our results showed that ATRA induced a dephosphorylation of Cx43 in both primary CGC and KGN cells, which was associated with an increase in GJIC as assessed by SL/DT experiments. Similar effects on Cx43 phosphorylation by ATRA have been observed in human endometrial stromal cells and were shown to be a post-translational effect mediated through increased interaction of Cx43 with its primary phosphatase, PP2A (Wu et al., 2013). The rapidity of this mechanism of action suggests that ATRA may serve an important regulatory role in hormone-responsive tissue to increase quickly GJIC in cooperation with the hypothalamic-pituitary-ovarian axis. Our data showing a further dramatic increase in the non-phosphorylated Cx43 species when ATRA treatment was combined with LH/hCG support this hypothesis. This mechanism may help mediate the rapid increase in GJIC noted in postovulatory granulosa cells in the face of comparatively slower changes in gonadotrophins and steroid hormone levels (Grazul-Bilska et al., 1996a, Khan-Dawood et al., 1998). In this way, regulation of Cx43 by ATRA may play an important role in the luteinization of antral follicles and oocyte competence. This hypothesis predicts that reduced follicle ATRA production may contribute to reduced fecundity in certain reproductive disorders, such as endometriosis, via adverse effects on luteinization and oocyte competency. Support for this contention has now been provided by our recent finding that follicular fluid from women with endometriosis undergoing IVF showed significantly lower mean concentrations of follicular fluid ATRA than control subjects (Pauli et al., 2013).

Taken together, these results present a mechanism by which retinoid metabolism in CGC may play a dynamic role in oogenesis and successful fertilization. Together with an increased understanding of the role of retinoids in ovarian follicle processes, this information may lead to a new appreciation of the importance of micronutrient repletion in normal reproductive physiology and potential cause(s) of impaired oocyte development in some women.

Authors' roles

M.W.B., J.W., S.A.P., R.N.T. and N.S.: study concept and design; M.W.B., J.W., S.A.P., M.A.K. and K.P.: acquisition of data; M.W.B., J.W., S.A.P., M.A.K., D.C.W., W.S., R.N.T. and N.S.: data analysis and interpretation; M.W.B., S.A.P., D.R.S., D.C.W. and W.S.: provision of clinical and culture samples; M.W.B., J.W., D.R.S., D.C.W., R.N.T. and N.S.: writing the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by Eunice Kennedy Shriver/National Institute of Child Health grants RO1HD55379 (N.S.), UO1HD66439 (N.S., R.N.T.), U54HD55787 (R.N.T.), R01HD077260 (MAK), the National Institute of Allergy and Infectious Diseases contract HHSN272202000046C (MAK) and the University of Maryland School of Pharmacy Mass Spectrometry Center #SOP1841-IQB2014 (MAK).

Conflict of interest

None declared.

Acknowledgements

The authors thank the physicians, fellows and nursing staff of the Emory Reproductive Center for their contributions to this study. We also acknowledge Grant A. Walter for excellent technical assistance.

References

  1. Ackert CL, Gittens JE, O'Brien MJ, Eppig JJ, Kidder GM. Intercellular communication via connexin43 gap junctions is required for ovarian folliculogenesis in the mouse. Dev Biol 2001;233:258–270. [DOI] [PubMed] [Google Scholar]
  2. Alminana C, Gil MA, Cuello C, Caballero I, Roca J, Vazquez JM, Gomez E, Martinez EA. In vitro maturation of porcine oocytes with retinoids improves embryonic development. Reprod Fertil Dev 2008;20:483–489. [DOI] [PubMed] [Google Scholar]
  3. Borowczyk E, Johnson ML, Bilski JJ, Bilska MA, Redmer DA, Reynolds LP, Grazul-Bilska AT. Role of gap junctions in regulation of progesterone secretion by ovine luteal cells in vitro. Reproduction 2007;133:641–651. [DOI] [PubMed] [Google Scholar]
  4. Brown JA, Eberhardt DM, Schrick FN, Roberts MP, Godkin JD. Expression of retinol-binding protein and cellular retinol-binding protein in the bovine ovary. Mol Reprod Dev 2003;64:261–269. [DOI] [PubMed] [Google Scholar]
  5. Dekel N. Cellular, biochemical and molecular mechanisms regulating oocyte maturation. Mol Cell Endocrinol 2005;234:19–25. [DOI] [PubMed] [Google Scholar]
  6. Diaz FJ, Wigglesworth K, Eppig JJ. Oocytes determine cumulus cell lineage in mouse ovarian follicles. J Cell Sci 2007;120:1330–1340. [DOI] [PubMed] [Google Scholar]
  7. Eberhardt DM, Will WA, Godkin JD. Retinol administration to superovulated ewes improves in vitro embryonic viability. Biol Reprod 1999;60:1483–1487. [DOI] [PubMed] [Google Scholar]
  8. Goldberg GS, Valiunas V, Brink PR. Selective permeability of gap junction channels. Biochim Biophys Acta 2004;1662:96–101. [DOI] [PubMed] [Google Scholar]
  9. Grazul-Bilska AT, Redmer DA, Johnson ML, Jablonka-Shariff A, Bilski JJ, Reynolds LP. Gap junctional protein connexin 43 in bovine corpora lutea throughout the estrous cycle. Biol Reprod 1996a;54:1279–1287. [DOI] [PubMed] [Google Scholar]
  10. Grazul-Bilska AT, Reynolds LP, Kirsch JD, Bilski JJ, Redmer DA. Gap junctional intercellular communication of bovine luteal cells from several stages of the estrous cycle: effects of prostaglandin F2 alpha, protein kinase C and calcium. Prostaglandins 1996b;52:285–302. [DOI] [PubMed] [Google Scholar]
  11. Grazul-Bilska AT, Reynolds LP, Kirsch JD, Redmer DA. Gap junctional intercellular communication of bovine luteal cells from several stages of the estrous cycle: effects of cyclic adenosine 3′,5′-monophosphate. Biol Reprod 1996c;54:538–545. [DOI] [PubMed] [Google Scholar]
  12. Haouzi D, Assou S, Mahmoud K, Hedon B, De Vos J, Dewailly D, Hamamah S. LH/hCGR gene expression in human cumulus cells is linked to the expression of the extracellular matrix modifying gene TNFAIP6 and to serum estradiol levels on day of hCG administration. Hum Reprod 2009;24:2868–2878. [DOI] [PubMed] [Google Scholar]
  13. Hughes SF, Haney AF, Hughes CL., Jr Use of human cumulus granulosa cells for in vitro screening of reproductive toxicants. Reprod Toxicol 1990;4:11–15. [DOI] [PubMed] [Google Scholar]
  14. Ikeda S, Kitagawa M, Imai H, Yamada M. The roles of vitamin A for cytoplasmic maturation of bovine oocytes. J Reprod Dev 2005;51:23–35. [DOI] [PubMed] [Google Scholar]
  15. Jeppesen JV, Kristensen SG, Nielsen ME, Humaidan P, Dal Canto M, Fadini R, Schmidt KT, Ernst E, Yding Andersen C. LH-receptor gene expression in human granulosa and cumulus cells from antral and preovulatory follicles. J Clin Endocrinol Metab 2012;97:E1524–E1531. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Johnson ML, Redmer DA, Reynolds LP, Bilski JJ, Grazul-Bilska AT. Gap junctional intercellular communication of bovine granulosa and thecal cells from antral follicles: effects of luteinizing hormone and follicle-stimulating hormone. Endocrine 2002;18:261–270. [DOI] [PubMed] [Google Scholar]
  17. Kalma Y, Granot I, Galiani D, Barash A, Dekel N. Luteinizing hormone-induced connexin 43 down-regulation: inhibition of translation. Endocrinology 2004;145:1617–1624. [DOI] [PubMed] [Google Scholar]
  18. Kane MA, Napoli JL. Quantification of endogenous retinoids. Methods Mol Biol 2010;652:1–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kane MA, Folias AE, Napoli JL. HPLC/UV quantitation of retinal, retinol, and retinyl esters in serum and tissues. Anal Biochem 2008a;378:71–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kane MA, Folias AE, Wang C, Napoli JL. Quantitative profiling of endogenous retinoic acid in vivo and in vitro by tandem mass spectrometry. Anal Chem 2008b;80:1702–1708. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Khan-Dawood FS, Yang J, Dawood MY. Hormonal regulation of connexin-43 in baboon corpora lutea. J Endocrinol 1998;157:405–414. [DOI] [PubMed] [Google Scholar]
  22. Kidder GM, Mhawi AA. Gap junctions and ovarian folliculogenesis. Reproduction 2002;123:613–620. [DOI] [PubMed] [Google Scholar]
  23. Kidder GM, Vanderhyden BC. Bidirectional communication between oocytes and follicle cells: ensuring oocyte developmental competence. Can J Physiol Pharmacol 2010;88:399–413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kipp JL, Golebiowski A, Rodriguez G, Demczuk M, Kilen SM, Mayo KE. Gene expression profiling reveals Cyp26b1 to be an activin regulated gene involved in ovarian granulosa cell proliferation. Endocrinology 2011;152:303–312. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Livera G, Rouiller-Fabre V, Valla J, Habert R. Effects of retinoids on the meiosis in the fetal rat ovary in culture. Mol Cell Endocrinol 2000;165:225–231. [DOI] [PubMed] [Google Scholar]
  26. Maman E, Yung Y, Kedem A, Yerushalmi GM, Konopnicki S, Cohen B, Dor J, Hourvitz A. High expression of luteinizing hormone receptors messenger RNA by human cumulus granulosa cells is in correlation with decreased fertilization. Fertil Steril 2012;97:592–598. [DOI] [PubMed] [Google Scholar]
  27. Menon KM, Munshi UM, Clouser CL, Nair AK. Regulation of luteinizing hormone/human chorionic gonadotropin receptor expression: a perspective. Biol Reprod 2004;70:861–866. [DOI] [PubMed] [Google Scholar]
  28. Nasiri E, Mahmoudi R, Bahadori MH, Amiri I. The effect of retinoic acid on in vitro maturation and fertilization rate of mouse germinal vesicle stage oocytes. Cell J 2011;13:19–24. [PMC free article] [PubMed] [Google Scholar]
  29. Nishi Y, Yanase T, Mu Y, Oba K, Ichino I, Saito M, Nomura M, Mukasa C, Okabe T, Goto K, et al. Establishment and characterization of a steroidogenic human granulosa-like tumor cell line, KGN, that expresses functional follicle-stimulating hormone receptor. Endocrinology 2001;142:437–445. [DOI] [PubMed] [Google Scholar]
  30. Norris RP, Freudzon M, Mehlmann LM, Cowan AE, Simon AM, Paul DL, Lampe PD, Jaffe LA. Luteinizing hormone causes MAP kinase-dependent phosphorylation and closure of connexin 43 gap junctions in mouse ovarian follicles: one of two paths to meiotic resumption. Development 2008;135:3229–3238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Papamentzelopoulou M, Mavrogianni D, Partsinevelos GA, Marinopoulos S, Dinopoulou V, Theofanakis C, Anagnostou E, Loutradis D. LH receptor gene expression in cumulus cells in women entering an ART program. J Assist Reprod Genet 2012;29:409–416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Pauli SA, Session DR, Shang W, Easley K, Wieser F, Taylor RN, Pierzchalski K, Napoli JL, Kane MA, Sidell N. Analysis of follicular fluid retinoids in women undergoing in vitro fertilization: retinoic acid influences embryo quality and is reduced in women with endometriosis. Reprod Sci 2013;20:1116–1124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Pierzchalski K, Yu J, Norman V, Kane MA. CrbpI regulates mammary retinoic acid homeostasis and the mammary microenvironment. FASEB J 2013;27:1904–1916. [DOI] [PubMed] [Google Scholar]
  34. Pierzchalski K, Taylor RN, Nezhat C, Jones JW, Napoli JL, Yang G, Kane MA, Sidell N. Retinoic Acid biosynthesis is impaired in human and murine endometriosis. Biol Reprod 2014;91:84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Sela-Abramovich S, Chorev E, Galiani D, Dekel N. Mitogen-activated protein kinase mediates luteinizing hormone-induced breakdown of communication and oocyte maturation in rat ovarian follicles. Endocrinology 2005;146:1236–1244. [DOI] [PubMed] [Google Scholar]
  36. Senbon S, Hirao Y, Miyano T. Interactions between the oocyte and surrounding somatic cells in follicular development: lessons from in vitro culture. J Reprod Dev 2003;49:259–269. [DOI] [PubMed] [Google Scholar]
  37. Solan JL, Lampe PD. Key connexin 43 phosphorylation events regulate the gap junction life cycle. J Membr Biol 2007;217:35–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Tahaei LS, Eimani H, Yazdi PE, Ebrahimi B, Fathi R. Effects of retinoic acid on maturation of immature mouse oocytes in the presence and absence of a granulosa cell co-culture system. J Assist Reprod Genet 2011;28:553–558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Tanmahasamut P, Sidell N. Up-regulation of gap junctional intercellular communication and connexin43 expression by retinoic acid in human endometrial stromal cells. J Clin Endocrinol Metab 2005;90:4151–4156. [DOI] [PubMed] [Google Scholar]
  40. Wang YA, Shen K, Wang Y, Brooks SC. Retinoic acid signaling is required for proper morphogenesis of mammary gland. Dev Dyn 2005;234:892–899. [DOI] [PubMed] [Google Scholar]
  41. Wang HX, Tong D, El-Gehani F, Tekpetey FR, Kidder GM. Connexin expression and gap junctional coupling in human cumulus cells: contribution to embryo quality. J Cell Mol Med 2009;13:972–984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Whaley SL, Hedgpeth VS, Farin CE, Martus NS, Jayes FC, Britt JH. Influence of vitamin A injection before mating on oocyte development, follicular hormones, and ovulation in gilts fed high-energy diets. J Anim Sci 2000;78:1598–1607. [DOI] [PubMed] [Google Scholar]
  43. Woods DC, Liu HK, Nishi Y, Yanase T, Johnson AL. Inhibition of proteasome activity sensitizes human granulosa tumor cells to TRAIL-induced cell death. Cancer Lett 2008;260:20–27. [DOI] [PubMed] [Google Scholar]
  44. Wu J, Taylor RN, Sidell N. Retinoic acid regulates gap junction intercellular communication in human endometrial stromal cells through modulation of the phosphorylation status of connexin 43. J Cell Physiol 2013;228:903–910. [DOI] [PubMed] [Google Scholar]
  45. Zheng WL, Ong DE. Spatial and temporal patterns of expression of cellular retinol-binding protein and cellular retinoic acid-binding proteins in rat uterus during early pregnancy. Biol Reprod 1998;58:963–970. [DOI] [PubMed] [Google Scholar]
  46. Zheng WL, Bucco RA, Sierra-Rievera E, Osteen KG, Melner MH, Ong DE. Synthesis of retinoic acid by rat ovarian cells that express cellular retinoic acid-binding protein-II. Biol Reprod 1999;60:110–114. [DOI] [PubMed] [Google Scholar]

Articles from Molecular Human Reproduction are provided here courtesy of Oxford University Press

RESOURCES