MicroRNA-378 regulates oocyte maturation via the suppression of aromatase in porcine cumulus cells

Abstract

We sought to investigate whether miR-378 plays a role in cumulus cells and whether the manipulation of miRNA levels in cumulus cells influences oocyte maturation in vitro. Cumulus-oocyte complexes (COCs) from ovarian follicles had significantly lower levels of precursor and mature miR-378 in cumulus cells surrounding metaphase II (MII) oocytes than cumulus cells surrounding germinal vesicle (GV) oocytes, suggesting a possible role of miR-378 during COC maturation. Overexpression of miR-378 in cumulus cells impaired expansion and decreased expression of genes associated with expansion (HAS2, PTGS2) and oocyte maturation (CX43, ADAMTS1, PGR). Cumulus cell expression of miR-378 also suppressed oocyte progression from the GV to MII stage (from 54 ± 2.7 to 31 ± 5.1%), accompanied by a decrease of growth differentiation factor 9 (GDF9), bone morphogenetic protein 15 (BMP15), zona pellucida 3 (ZP3), and CX37 in the oocytes. Subsequent in vitro fertilization resulted in fewer oocytes from COCs overexpressing miR-378 reaching the blastocyst stage (7.3 ± 0.7 vs. 16.6 ± 0.5%). miR-378 knockdown led to increased cumulus expansion and oocyte progression to MII, confirming a specific effect of miR-378 in suppressing COC maturation. Aromatase (CYP19A1) expression in cumulus cells was also inhibited by miR-378, leading to a significant decrease in estradiol production. The addition of estradiol to IVM culture medium reversed the effect of miR-378 on cumulus expansion and oocyte meiotic progression, suggesting that decreased estradiol production via suppression of aromatase may be one of the mechanisms by which miR-378 regulates the maturation of COCs. Our data suggest that miR-378 alters gene expression and function in cumulus cells and influences oocyte maturation, possibly via oocyte-cumulus interaction and paracrine regulation.

micro-rnas (miRNAs) are small RNAs that diversely regulate gene expression through interaction with the 3′-untranslated region (3′-UTR), inducing mRNA destabilization and suppressing translation of their target genes (16, 24, 57). The functions of these noncoding RNAs are emerging as important regulators controlling diverse physiological processes, including cell division and differentiation in various cell types (16, 24, 57). Expectedly, important roles for miRNA-mediated regulation have also been demonstrated in the ovary (reviewed in Ref. 3). Oocyte-specific deletion of DGCR8, a key enzyme in miRNA biogenesis, had no significant adverse effect on oocyte maturation (51), indicating that the role of miRNA in the oocyte is limited. Increasing evidence supports the notion that miRNAs regulate ovarian function primarily through their actions in ovarian somatic cells, such as granulosa cells (42, 55, 58, 61). Granulosa miRNAs have been shown to regulate proliferation, apoptosis, and steroidogenesis (8, 46, 59, 60). Variations in granulosa miRNA expression, and thus their cellular activity, likely affect biological processes taking place in the oocyte, although few studies to date have examined this phenomenon.

Mural granulosa cells that make up the follicle wall and cumulus granulosa cells that surround the oocyte are essential in supporting the growth and maturation of the oocyte in preparation for the release of a fertilizable egg (7, 14). Oocytes and cumulus cells, functioning together as a cumulus-oocyte complex (COC), interact through direct contact mediated by gap junctions and through paracrine signaling. This bidirectional communication is crucial for the acquisition of oocyte developmental competence (15, 19). Steroidogenesis by cumulus cells is critical for successful oocyte maturation (18, 25) and depends on specific enzymes that modify cholesterol and its derivatives.

We have shown previously that miR-378 is expressed in mural granulosa cells in an inverse manner compared with the expression of aromatase, a key enzyme in estrogen biosynthesis (58). In vitro miR-378 overexpression and knockdown experiments revealed that aromatase expression, and therefore, estradiol production, by granulosa cells is posttranscriptionally downregulated by miR-378. These findings were supported by a recent in vivo study showing that miR-378 expression in granulosa cells is dependent on the stage of follicular development and is inversely correlated to the levels of estradiol in the follicular fluid (41). The expression of microRNAs in ovarian cumulus cells has been reported recently (1, 35); however, the specific roles in this cell type have only begun to be explored. In the current study, we developed a method to overexpress miR-378 exclusively in the cumulus cells of COCs. We then used this system to investigate whether miR-378 also regulates cumulus function and whether altering cumulus miRNA expression influences oocyte maturation.

MATERIALS AND METHODS

In vitro maturation of pig COCs.

All animal procedures were performed in accordance with the guidelines established by and with the approval of the Animal Care Committee at the University of Guelph. Porcine ovaries were collected from gilts in 1× PBS from a local slaughterhouse, transported to the laboratory within 1.5 h, maintained at 25–28°C, and rinsed three times with sterile 1× PBS. The COCs were aspirated from large-sized follicles (3–6 mm in diameter) with a 20-gauge needle fixed to a 20-ml disposable syringe. The COCs surrounded by a compact cumulus mass and having evenly granulated cytoplasm were picked out by mouth pipette under a microscope. Serum-free TCM-199 (Gibco, Burlington, ON, Canada), buffered with 10 mM HEPES and 26 mM bicarbonate was used for washing COCs. Pools of 40–50 COCs were then matured in vitro in 0.5 ml TCM199 (Invitrogen, Hercules, CA) containing 10% porcine follicular fluid (pooled from several ovaries and stored in aliquots at −80°C until use) and supplemented with 5 IU/ml follicle-stimulating hormone (Sioux Biochemicals), 5 IU/ml luteinizing hormone (Sioux Biochemicals), 0.1 mg/ml cysteine (Sigma), and 10 ng/ml epidermal growth factor (EGF; Sigma) in 4-well plates (NUNC) at 38.5°C in 5% CO₂ in a humidified air atmosphere (54). At given time points, COCs were pipetted into 1.5 ml tubes, and 150 IU hyaluronidase was added and incubated for 3 min at 37°C. Oocytes were then picked out using mouth pipette, and cumulus cells were collected by centrifugation at 1,000 g for 3 min at room temperature. Cumulus cells, oocytes, and spent media samples were stored at −80°C until further analysis.

The cumulus expansion diameter and area were measured after 30 h of in vitro maturation. The plates were imaged under a stereomicroscope, and ImageJ was used for analysis of cumulus expansion, which was calculated as described previously (30). Metaphase II (MII) oocytes were obtained after in vivo maturation (IVM) and assessed as described previously (23), and both the number of MII oocytes and non-MII oocytes were recorded. The rate of MII was considered as the number of oocytes reaching MII divided by all recovered oocytes.

Production of recombinant lentiviral particles.

Cloning of the lentiviral gene transfer plasmids for miR-378 overexpression, pL-SIN-Lenti-H1-miR-378-EF1A-EGFP (Lenti-miR-378), and for control virus, pL-SIN-Lenti-EF1A-EGFP (Lenti-GFP), as well as production of recombinant lentiviral particles, was carried out as described previously (58). High-concentration virus was made by ultracentrifugation; the viral supernatant was passed through a 0.45-μm filter and carefully decanted into sterilized Ultra-Clear centrifuge tubes (Beckman cat. no. 344058). For each round of ultracentrifugation, 30 ml of viral supernatant was centrifuged at 16,500 g for 90 min at 4°C in a Beckman SW28 swinging bucket rotor lined with a Beckman Ultra-Clear centrifuge tube. All concentrated viral supernatants for each virus were pooled, divided into aliquots, and stored at −80°C until use.

Viral transduction of COCs.

COCs were grouped randomly and cultured in four-well plates with 0.5 ml of IVM medium, as described above. Equal volumes of concentrated Lenti-green fluorescent protein (GFP) or Lenti-miR-378 virus were added to each well to achieve a multiplicity of infection of 1.0. Polybrene was included in the media at a final concentration of 8 μg/ml (Sigma Chemical, St. Louis, MO). Infected COCs were cultured in a humidified atmosphere of 95% air and 5% CO₂ at 38.5°C for 44 h. Trypan blue exclusion analysis revealed that transduced cumulus cells remained viable (87.7 ± 1.2 and 92.9 ± 3.6% for GFP and miR-378 viruses, respectively).

In vitro fertilization.

Following IVM, denuded oocytes were washed and underwent in vitro fertilization (IVF), as described previously without modification (53). On day 7 after fertilization, oocytes were examined under a light microscope to determine cleavage and blastocyst rates.

Antisense inhibition of miRNA expression.

Specific Anti-miR miRNA Inhibitor (cat. no. AM17000; Life Technologies) for the mature sequence of mir-378-3p and Anti-miR miRNA Inhibitor Negative Control (cat. no. AM17010) were used to inhibit endogenous miRNA. COCs were transfected with either anti-miR-378-3p or negative control using the Lipofectamine 2000 reagent, following the manufacturer's protocol, at final concentrations of 20 and 40 pmol/ml. After 24 h, IVM medium was replaced to minimize cytotoxic effects. Oocytes, cumulus cells, and spent medium were collected 44 h after transfection and stored at −80°C until analysis.

Reverse transcription and quantitative PCR.

Reverse transcription and quantitative PCR (RT-qPCR) for miRNA and mRNA were performed as described previously (58). For miRNA expression normalization, U6 and sn44 were used as reference genes (54), whereas GAPDH and RPII were used to normalize mRNA expression. Relative expression was determined using the 2^−ΔΔC_T method (31). Primer information is given in Table 1.

Table 1.

Primer information

Gene Symbol	Primer Sequences	Product Size, bp	NCBI Accession No.
CX43	ACTGAGCCCCTCCAAAGACT	191	NM_001244212
	GCTCGGCACTGTAATTAGCC
ADAMTS1	CGTGAACAAGACCGACAAGA	103	DQ177331
	AACTCCTCCACCACACGTTC
PTGS2	ATGATCTACCCGCCTCACAC	279	AF207824
	GCAGCTCTGGGTCAAACTTC
HAS2	GAAGTCATGGGCAGGGACAATTC	407	NM_214053
	TGGCAGGCCCTTTCTATGTTA
PTGX3	TCAGTGCCTGCATTTGGGTC	225	GQ_412351
	CTACATGCCCTTGTTCAGAA
TNFAIP6	TCATAACTCCATATGGCTTGAAC	396	NM_001159607
	TCTTCGTACTCATTTGGGAAGCC
PGR	CCCTAGCTCACAGCGTTTCT	184	NM_001166488
	CACCATCCCTGCCAATATCT
BMP15	GGATCCAGAAAAGCACAACC	207	NM_001005155
	AGTGTCCAGGGATGAAATGC
GDF9	TAGTCAGCTGAAGTGGGACA	227	HQ687750
	ACGACAGGTGCACTTTGTAG
C-KIT	TGTATTCACAGAGACTTGGCGG	124	NM_001044525
	CGTTTCCTTTGACCACGTAA
ZP3	TGGTGTACAGCACCTTCCTG	202	NM_213893
	ATCAGGCGCAGAGAGAACAC
GAPDH	TCGGAGTGAACGGATTTGGC	147	NM_001206359.1
	TGCCGTGGGTGGAATCATAC
premir-378	CTCCTGACTCCAGGTCCTGT	60	NR_038523.1
	GCCTTCTGACTCCAAGTCCA

CX43, connexin 43; ADAMTS1, a disintegrin and metalloproteinase with thrombospondin motifs 1; PTGS2, prostaglandin-endoperoxide synthase 2; HAS2, hyaluronan synthase 2; PTGX3, pentraxin 3; TNFAIP6, TNFα-induced protein 6; PGR, progesterone receptor; BMP15, bone morphogenetic protein 15; GDF9, growth differentiation factor 9; C-KIT, kit ligand's receptor; ZP3, zona pellucida 3.

Western blotting analysis and antibodies.

The cell extracts were assessed for protein concentration using the Bio-Rad DC Protein Assay Kit. Western blotting for aromatase protein was performed as described (58). Protein (30 μg/lane) was resolved via 10% SDS-PAGE. Proteins were transferred onto polyvinylidene difluoride (Hybond-P; Amersham Pharmacia Biotech).

Analysis of estradiol levels by ELISA.

After 44 h of culture, spent medium was collected and centrifuged at 500 g to remove cells or debris. An aliquot of each sample was diluted 100-fold with double-distilled water, and estradiol (E₂) level was measured using an Estradiol ELISA kit (EA 70; Oxford Biomedical Research, Oxford, MI) according to the manufacturer's protocol. E₂ levels were compared between Lenti-miR-378 and Lenti-GFP, whereas an empty well was used as blank control for E₂ to assess the basal level in IVM media. Each experiment was carried out in triplicate.

Statistical analysis.

All experiments were performed at least three times from different batches of ovaries and three replicates per batch. Data are presented as means ± SE. Differences were analyzed by the Student t-test or one-way analysis of variance (ANOVA), followed by the Bonferroni post hoc test, using GraphPad Prism 5.0 (GraphPad Software, San Diego, CA). Significance was set at P < 0.05.

RESULTS

We first examined the expression of miR-378 in cumulus cells during in vitro oocyte maturation (IVM). As shown in Fig. 1, both the pre-mir-378 hairpin precursor and mature miR-378-3p were detected in cumulus cells of the COC at both germinal vesicle (GV) and MII stages. Levels of miR-378 (both precursor and mature forms) were significantly lower in cumulus cells surrounding MII stage oocytes compared with those from GV stage COCs, suggesting a potential role of miR-378 in cumulus cells during IVM. No difference in miR-378 level was observed between oocytes at these two stages (Fig. 2).

Fig. 1.

Expression of miR-378 in cumulus cells during in vitro maturation. Data represent mean expression of pre-mir-378 (A) and mature miR-378-3p (B) in pooled cumulus cells from immature germinal vesicle (GV) and metaphase II (MII) stage cumulus-oocyte complexes (COCs) ± SE of 4 independent experiments. Endogenous snU6 RNA was used to normalize expression. *Statistical differences between GV and MII groups (P < 0.05).

Fig. 2.

Overexpression of miR-378 is restricted to cumulus cells. A: relative expression levels for mature miR-378 in oocytes transduced with control lentivirus [green fluorescent protein (GFP)] or Lenti-mir-378-GFP (mir-378). Expression of the endogenous gene snU6 was used to normalize miRNA expression. B: quantitative RT-PCR detection of GFP expression, used as a lentivirus reporter, confirmed its expression only in cumulus cells. No amplification was detected in the oocyte group when the PCR products were run on a 1% agarose gel. Data represent the mean ± SE of 4 independent experiments.

To investigate whether suppression of miR-378 in cumulus cells is necessary for oocyte maturation, we next studied the feasibility of overexpressing miR-378 in cumulus cells during IVM using a lentivirus expressing the miR-378 with a GFP reporter (miR-378); as a negative control, we used a lentivirus expressing GFP alone. Transduced COCs were examined for GFP expression using dark-field fluorescence. GFP was visualized in cumulus cells of both transduced groups (Fig. 3A), and RT-qPCR analysis revealed a significant increase in cumulus cell miRNA expression in the miR-378 group compared with the GFP control group (Fig. 3B). Expression of GFP in the oocyte was not observed in any transduction (Fig. 3A). This result is not surprising considering the multiple layers of expanded oophorous and the thick zona pellucida protecting the oocyte from lentivirus infection. As expected, RT-qPCR analysis detected GFP in pooled cumulus cells but not oocytes of transduced COCs (Fig. 4), confirming cumulus cell-specific overexpression of miR-378.

Fig. 3.

A: morphology of COCs after 24 h of in vitro maturation (IVM) culture are shown; COCs transduced with control lentivirus (GFP) or Lenti-mir-378-GFP (mir-378). B: relative expression of miR-378-3p in transduced cumulus cells. C: after 30 h of IVM culture, cumulus cell expansion was assessed in both groups. A representative image is shown. D: quantified cumulus cell expansion. Data represent the mean ± SE of 3 or 4 independent experiments (n = total COC used). *Statistical differences compared with GFP controls (P < 0.05). Scale bar, 200 μm in A and C.

Fig. 4.

Oocyte expression of miR-378 is constant during maturation. Relative expression levels for mature miR-378 in oocytes at the GV and MII stages. Data represent the mean ± SE of 3 independent experiments.

Transduced COCs showed decreased expansion of the cumulus oophorus with over expression of miR-378 (Fig. 3, C and D). To study whether the expression of genes involved with cumulus cell expansion (10, 17, 37, 56) responded to miR-378, RT-qPCR was performed in cumulus cells following IVM. Overexpressing miR-378 decreased expression of prostaglandin-endoperoxide synthase 2 (PTGS2) and hyaluronan synthase 2 (HAS2) transcripts, whereas no change in the expression of pentraxin 3 (PTX3) or tumor necrosis factor-α-induced protein 6 (TNFAIP6) was detected (Fig. 5A). Genes involved with progesterone-mediated COC maturation (39, 43, 44), specifically progesterone receptor (PGR), a disintegrin and metalloproteinase with thrombospondin-like motifs 1 (ADAMTS1), and connexin 43 (CX43), were also downregulated in response to miR-378 (Fig. 5B).

Fig. 5.

Influence of miR-378 on the 24-h expression of cumulus cell genes associated with expansion (A) and oocyte maturation (B) by RT-qPCR. Relative expression normalized using the endogenous reference gene GAPDH. Data represent the mean ± SE of 3 independent experiments. *Statistical difference between Lenti-miR-378 and Lenti-GFP groups, P < 0.05.

We next studied whether the overexpression of miR-378 in cumulus cells altered oocyte gene expression and oocyte function. We observed no difference in C-KIT expression, a receptor tyrosine kinase essential for periovulatory maturation (Fig. 6B) (62). However, the oocyte-specific genes GDF9, BMP15, CX37, and ZP3 all showed lower expression in oocytes transduced with miR-378 compared with GFP controls (Fig. 6A). Moreover, overexpression of miR-378 lowered rates of oocyte meiotic progression to the MII stage and impaired embryo development (Fig. 6, B and C). To verify that the changes observed in cumulus cells and oocytes were specific to miR-378, a loss-of-function approach was utilized. Inhibition of miR-378-3p (miR-378-3p In; Fig. 7A) resulted in increased cumulus cell expansion and oocyte maturation compared with transfected negative controls (Fig. 7, B and C), confirming a specific role for miR-378 in inhibiting COC expansion and maturation.

Fig. 6.

A: after 44 h of IVM culture, total RNA was isolated from pooled, denuded oocytes, and relative expression of genes known to be important for oocyte competence was determined by RT-quantitative PCR and normalized to the endogenous reference gene GAPDH. B: oocyte progression to the MII stage was counted and quantified for GFP (n = 188) and miR-378 oocytes (n = 185). C: similarly, blastocysts were counted 7 days after in vitro fertilization of GFP (n = 150) and miR-378 (n = 190) oocytes. Data represent the mean ± SE of 3 independent experiments for A and 4 independent experiments for B and C. * and **Statistical difference between miR-378 and GFP groups (*P < 0.05; **P < 0.01). GDF9, growth differentiation factor 9; BMP15, bone morphogenetic protein 15; C-KIT, kit ligand's receptor; ZP3, zona pellucida 3; CX37, connexin 37.

Fig. 7.

Expression of miR-378 (A), cumulus expansion (B), and oocyte maturation (C) following transfection with inhibitor (miR-378-3p In; n = 194 and n = 193 for 20 and 40 pmol/ml, respectively) or negative control (NC; n = 191). Cumulus expansion and oocyte progression to MII were measured and quantified for 30 and 44 h, respectively. Data represent the mean ± SE of 3 (A and B) or 4 (C) independent experiments. Bars with different letters are significantly different from each other, P < 0.05.

E₂ is known to be important for Has2 expression and cumulus cell expansion in COCs (50), and aromatase protein (CYP19A) is necessary for synthesis of E₂. We asked whether expression of aromatase, a target of miR-378 in mural granulosa cells (58), was affected by miR-378 overexpression in cumulus cells. Western blot revealed that aromatase levels were decreased by miR-378 during IVM (Fig. 8, A and B). Moreover, E₂ levels were also lower in spent IVM media from miR-378 transduced COCs than in COCs transduced with the control GFP lentivirus (Fig. 8C). If decreased E₂ production was one mechanism by which miR-378 suppressed oocyte maturation, then the addition of E₂ was expected to reverse the miR-378-induced suppression. As shown in Fig. 9, A and B, the suppression of cumulus expansion and oocyte maturation by miR-378 was reversed by exogenous estradiol.

Fig. 8.

A: representative Western blot for aromatase protein in cumulus cells transduced with miR-378-expressing lentivirus (miR-378) or a GFP control virus (GFP). B: normalized Western blots were quantified by densitometry. C: after 44 h of IVM culture, COC culture medium was collected from the GFP and miR-378 groups, and levels of estradiol were analyzed by ELISA. Data represent the mean ± SE of 4 independent experiments. *Statistical differences between the GFP and miR-378 groups (P < 0.05).

Fig. 9.

Cumulus expansion (A) and MII progression (B) were measured after cells transduced with GFP lentivirus (GFP; n = 118) or miR-378 lentivirus (miR-378; n = 115) were matured for 30 h. Estradiol (E₂) was included in the medium at 1 (n = 115) or 5 ng/ml (n = 119) E₂. Data represent the mean ± SE of 3 independent experiments. Bars with different letters are significantly different from each other, P < 0.05.

DISCUSSION

It is well established that oocyte meiotic progression is influenced by the unique local microenvironment formed by somatic cells, which consist of surrounding cumulus cells and mural granulosa cells. Signals released from ovarian somatic cells direct oocyte reentry into the meiotic cell cycle and promote maturation (9). Our study provides evidence that miRNA expression in cumulus cells influences gene expression, meiotic progression, and developmental competence in the oocyte. Lentivirus infection is specific to the cumulus cells of the COCs without infecting the oocyte. This conclusion is based on the fact that neither GFP florescence nor GFP transcript was detected in the oocytes of the transduced COCs. As well, no change in oocyte miR-378 levels was detected between the control and miR-378 groups (Fig. 4). Thus the observed changes in oocyte activity, discussed below, were mediated by miR-378 overexpression in cumulus cells.

When cumulus cells were infected with lentivirus overexpressing miR-378 at the start of IVM, by 30 h expansion of the cumulus oophorus was considerably less than those cells transfected with a control virus (Fig. 3C). Cumulus expansion is triggered by an increase in the transient synthesis and accumulation of hyaluronic acid, a major component of the COC extracellular matrix that is mediated by the enzyme HAS2. In addition to hyaluronic acid production by HAS2, several other transcripts are also known to be essential to cumulus expansion, as null mutations have a direct negative impact on this process. These include Tnfaip6, Ptx3, and Ptgs2 (also known as Cox-2) (17, 37, 56). We observed decreased expression of PTGS2 and HAS2 in the miR-378 group, which is consistent with the suppression of cumulus cell expansion in the COCs transduced with the miR-378 virus. We did not observe changes in the expression levels of TNFAIP6 or PTX3, two hyaluronan-binding proteins that are not affected by changes in Has2 expression (49). Cumulus cell expansion has long been one of the primary measures of oocyte quality, creating a microenvironment favorable for nutrients and paracrine factors to reach the oocyte to alter gene expression and allow meiotic progression (52). During ovulation, the expansion of cumulus facilitates the release of a COC from the follicle and helps to maintain the viability of the oocyte during its passage in the oviduct (22, 40). Accordingly, suppression of cumulus cell expansion by miR-378 attenuated the ability of the oocyte to mature to the MII stage and subsequently reduced the blastocyst rate following IVF (Fig. 6, B and C).

Regulation of gene expression by miR-378 is likely multifaceted given the critical bidirectional communication among cumulus cells and the oocyte as well as the multitarget nature of miRNA (21). Recently, we have shown that miR-378 directly targets and downregulates PGR at the mRNA and protein level in mural granulosa cells (54). We found a similar decrease in PGR mRNA expression when miR-378 was overexpressed in cumulus cells, suggesting that PGR is a direct target in both cell types. PGR is a nuclear transcription factor that initiates a signaling cascade during COC maturation and is required for cumulus expansion and oocyte meiotic progression (36). Transcriptional targets of PGR, ADAMTS1, and peroxisome proliferator-activated receptor-γ (PPARG) were also downregulated by miR-378-mediated inhibition of PGR, although they were not directly targeted by the miRNA (54). It is probable that, as in mural granulosa cells, miR-378 directly targets PGR in cumulus cells and indirectly suppresses ADAMTS1 (Fig. 5), a metalloprotease involved in ECM remodeling during cumulus expansion (5, 43). Inflammatory pathways mediated by PTGS2, which is necessary for ovulation (34), have also been proposed as interacting with PGR and PPARG networks (reviewed in Ref. 28). Because PTGS2 has no predicted binding sites for miR-378, downregulation of PGR or additional uncharacterized interactions taking place in cumulus cells could be responsible for the transcriptional changes observed with overexpression of miR-378. Thus, the targeting of specific mRNA transcripts by miR-378 is likely to alter gene expression both directly and indirectly, suggesting that in cumulus cells miR-378 may downregulate a network that promotes expansion and supports oocyte maturation.

Altered cumulus gene expression by miR-378 may have impacted the metabolic processes required to supply the oocyte with endocrine signals and nutrients, ultimately impairing oocyte maturation. To this end, we observed decreased expression of the oocyte specific genes BMP15, GDF9, and ZP3 in the miR-378 group. Both GDF9 and BMP15 are from the transforming growth factor-β superfamily, which is expressed in an oocyte-specific manner, and play key roles in promoting follicle growth beyond the primary stage (13, 26, 29). Because we saw no evidence to suggest increased miR-378 expression in the oocyte, impaired expression of GDF9 and BMP15 likely resulted from direct miRNA targeting taking place in the cumulus cells. Expression of ZP3 is necessary for oocyte maturation and fertilization. Female mice lacking ZP3 are completely infertile and exhibit a severe reduction in ovulated eggs (38). Despite the lower expression of ZP3 in the miR-378 group, we did not observe a difference in oocyte fertilization (percentage of cleaved embryos; data not shown) compared with GFP controls. However, decreased oocyte GDF9, BMP15, and ZP3 expression may account partially for the impaired oocyte maturation and blastocyst development observed in the miR-378 group.

Our observation that cumulus cell aromatase protein levels and activity, measured via E₂ production, decreased in response to miR-378 overexpression suggests that this is one of the pathways regulating cumulus expansion and oocyte maturation. Supplementing E₂ has been shown to enhance cumulus expansion of preantral COCs and Has2 transcript levels in vitro (50). Similarly, we could reverse the suppressive effects of miR-378 on cumulus expansion and oocyte maturation by supplying exogenous E₂ to miR-378 groups during IVM. Partly through suppression of E₂ production in cumulus cells, miR-378 may act to prevent cumulus cell differentiation and expansion. The actions of E₂ in regulating cumulus expansion are mediated by oocyte-secreted GDF9 and BMP15 (50). Not surprisingly then, COCs isolated from Gdf9^+/− Bmp15^−/− double-mutant mice have impaired cumulus cell expansion and Has2 expression (48), and suppression of GDF9 by RNAi is sufficient to inhibit expression of key cumulus cell extracellular matrix genes (20). Given the bidirectional cumulus-oocyte communication, our observed changes in oocyte expression of GDF9 and BMP15 could have contributed further to the suppression of cumulus expansion.

Oocyte-cumulus communication also takes place via gap junctions, which are known to interact with the GDF9-BMP15 network (reviewed in Ref. 27). Made from hexametric connexin subunits, gap junctions are the key channels mediating local and endocrine signals to the oocyte from cumulus cells and are formed between cytoplasmic processes penetrating the zona pellucida and the oocyte as well as between cumulus cells. Connexin 37 (CX37) forming the oocyte-cumulus gap junction subunit is critical for oocyte maturation, as deletion arrests oocyte maturation and causes female sterility in mice (45). Loss of CX43 also inhibits folliculogenesis (2) in part by reducing responsiveness to GDF9 signaling (19). We found that overexpression of miR-378 decreased expression of both CX37 and CX43. Interestingly, both CX37 and CX43 have potential binding sites for miR-378 in their 3′-UTR (data not shown) and could be direct targets of this miRNA. Their decreased expression could also be a result of transcription factor targeting by miR-378. In periovulatory granulosa cells, the Gα-binding protein transcription factor α-subunit (GABPA), along with PGR, plays a role in up regulating Rhox5 transcription (6, 33). PGR is a confirmed target of miR-378 in the ovary, and GABPA is targeted and suppressed by miR-378 in cancer cells (12). Additional transcription factors necessary for connexin expression may be similarly targeted in cumulus cells. Evaluating the contribution of miR-378 on connexin expression in the COC by direct targeting, indirect modulation of transcriptional regulators, or some combination of the two warrants a full investigation.

Our finding that suppressing miR-378 during IVM leads to improved cumulus cell expansion and oocyte maturation suggests that this miRNA could be manipulated to improve in vitro reproductive technologies. Barbato et al. (4) observed decreasing miR-92 expression during neuron differentiation, which was similar to our observed decrease in miR-378 expression during COC in vitro maturation. Lentivirus-mediated delivery of miR-92 inhibitors was able to accelerate the shift to a differentiated phenotype (4). When miR-378 was overexpressed in cumulus cells, decreased cumulus expansion and oocyte maturation were observed compared with the GFP control group. However, downregulating miR-378 with transfected inhibitors promoted cumulus expansion and increased oocyte progression to the MII stage in a dose-dependent manner. These “gain-” and “loss-of-function” studies further confirm the specific effects of miR-378 on COC maturation. Future efforts to titer miR-378 as well as other potential negative regulators of oocyte maturation may offer novel therapeutic approaches to improve IVM and other assisted reproductive technologies.

Here, we have provided the first investigation into the role of miR-378 cumulus cells during IVM, further supporting a role for miR-378 in regulating ovarian function. microRNA-378 has previously been identified as being enriched in the exosome fraction of follicular fluid (11, 47), and previously we have shown that it regulates the expression of PGR and CYP19A1 in pig mural granulosa cells (54, 58). Ma et al. (32) reported the inversed expression between miR-378 and the interferon-γ receptor 1 (IFNGR1) gene, a predicted target of miR-378 during corpus luteum development. Highly upregulated miR-378 expression in nonregressed bovine corpus luteum suggested a potential role for miR-378 in suppressing luteal cell apoptosis via IFNGR1 suppression (32). Whether miR-378 specifically targets IFNGR1 transcripts in cumulus cells remains unknown, but these studies provide evidence that this miRNA may perform important functions in ovarian cell differentiation and development.

In summary, our work has demonstrated that increased miR-378 levels in cumulus cells suppress cumulus expansion and oocyte meiotic progression. This results in decreased expression of several expansion- and maturation-associated genes in both the oocyte and cumulus cells and reduces oocyte developmental competence. One mechanism for miR-378 acts via decreased E₂ levels after miRNA-mediated inhibition of aromatase expression. Our finding offers new insights into how oocyte maturation can be impacted by miRNA expression in its associated somatic cells.

GRANTS

This work was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC), the Canadian Institutes of Health Research, the NSERC EmbryoGENE Research Network, and the Ontario Ministry of Agriculture, Food, and Rural Affairs.

DISCLOSURES

The authors have nothing to disclose.

AUTHOR CONTRIBUTIONS

B.P. and D.T. performed experiments; B.P. and D.T. analyzed data; B.P., D.T., W.S., and J.L. interpreted results of experiments; B.P. and D.T. prepared figures; B.P., D.T., and J.L. edited and revised manuscript; B.P., D.T., W.S., and J.L. approved final version of manuscript; D.T. drafted manuscript; J.L. conception and design of research.