Background
Germinal vesicle breakdown (GVBD) is the first physical sign that an oocyte is committed to maturation and also represents the onset of a period of transcriptional quiescence which persists until the activation of the embryonic genome. During this period, changes in mRNA and protein abundance within the oocyte can occur through interactions with the surrounding cumulus cells and/or through posttranscriptional gene regulation (PTGR) within the oocyte. MicroRNA (miRNA) represent a unique RNA class that function as potent regulators of transcription and protein abundance through PTGR [
12]. MiRNA are small (18–24 nt), non-coding RNA molecules that confer PTGR through several mechanisms, such as impairing translation efficiency and affecting mRNA stability following interaction with the 3′ untranslated region (3′UTR) of target mRNA molecules [
5,
6]. Numerous miRNA are expressed in the mouse oocyte and developing embryo and it has been demonstrated that the conditional knockout of DICER, an enzyme involved in miRNA processing, during oocyte development impairs the production of oocytes capable of being fertilized and developing normally [
38]. Estimations predict approximately 1,000 miRNA are present in the human genome having the potential ability to impact approximately 30 % of protein coding genes [
28]. Some miRNA have numerous mRNA targets [
27] while others have few predicted targets [
8,
34].
Utilizing miRNA microarray analysis and deep sequencing we have previously identified microRNA-21 (MIR21) as an up-regulated miRNA during porcine oocyte in vitro maturation [
41]. In mice it has been reported that luteinizing hormone may increase the expression of MIR21 in mouse granulosa cells and in vivo MIR21 inhibition has a negative impact on ovulation rate [
11]. MIR21 is a well characterized miRNA that has demonstrated the ability to confer PTGR oncogenic cell lines by affecting cellular proliferation through controlling apoptosis [
9,
13,
15,
42]. The MIR21 gene is transcribed via RNA polymerase II and is located in intronic regions of the transmembrane 49 gene (TMEM49; also referred to as VMP1) [
10,
17]. The mature MIR21 sequence was first identified in human HeLa cells [
22] and has since been predicted and verified to be present in the transcriptome of several other species including the pig. The anti-apoptotic capabilities of MIR21 in cancer cells are manifested through the ability to suppress critical apoptotic genes including programmed cell death 4 (PDCD4, previously referred to as neoplastic transformation inhibitor) [
3,
16,
24,
33]. MIR21 interacts with
PDCD4 through binding with complementary sequence in the 3′UTR of
PDCD4 mRNA resulting in reduced translation and subsequently reduced protein abundance in oncogenic cell lines [
3,
24]. Importantly the 3′UTR of pig
PDCD4 possesses a conserved MIR21 recognition sequence, particularly in the seed sequence. This suggests that if both MIR21 and
PDCD4 are present in the oocyte, MIR21 could impact PDCD4 protein abundance as the necessary accessory proteins for miRNA function are present in the oocyte during GVBD and progression to MII arrest.
Our working hypothesis that increased MIR21 abundance in the maturing cumulus oocyte complex of the pig is associated with posttranscriptional regulation of PDCD4 expression in the oocyte and that suppression of MIR21 function during oocyte maturation will compromise subsequent embryonic development. The objective of this study was to determine expression patterns of MIR21 and demonstrate its potential interactions with PDCD4 in the cumulus oocyte complex (COC) during oocyte maturation in the pig. Here we demonstrate PDCD4 protein down regulation is temporally associated with MIR21 abundance increase during in vitro oocyte maturation. These data indicate a reduced ability of MIR21 to suppress PDCD4 protein abundance in the presence of a MIR21 inhibitor suggesting a biological interaction between MIR21 and PDCD4 mRNA occurs during in vitro oocyte maturation in the pig.
Methods
Animal use was in accordance with the Guiding Principles for Care and Animals and procedures were approved by the Iowa State Institutional Animal Care and Use Committee.
In vitro maturation
All chemicals were purchased from Sigma Chemical Co. (St. Louis MO) unless otherwise stated. Sow ovaries were obtained from a local abattoir for isolation of cumulus oocyte complexes (COCs) to be subjected to in vitro maturation (IVM) as previously described [
41,
44]. Briefly, follicles (3–5 mm) were aspirated and COC were collected and washed in TL-Hepes with 0.1 % polyvinyl alcohol (PVA). Cumulus oocyte complexes were cultured in maturation media (Tissue Culture Media 199 (TCM-199)) containing 0.57 mM L-cysteine, follicle stimulating hormone (0.5 μg/mL), luteinizing hormone (0.5 μg/mL), and epidermal growth factor (10 ng/mL) for 42–44 h at 39.0 °C in 5 % CO
2. Prior to in vitro maturation, an aliquot of GV stage oocytes for each replication were randomly selected from the COC pool. GV stage oocytes used for analysis were stripped of cumulus cells via vortex (6 to 8 min) in 1 % hyaluronidase in TL-Hepes-PVA. Following in vitro maturation oocytes were stripped of cumulus cells by vortexing 4–6 min in TL-Hepes-PVA supplemented with 1 % hyaluronidase, and Metaphase II arrested (MII) oocytes were identified by the presence of an extruded polar body. Cumulus cells before and after maturation and GV and MII oocytes (25 oocytes per pool) were snap frozen in liquid nitrogen and stored at −80 °C until used for quantitative reverse transcription PCR (RT-qPCR). Pools of GV and MII arrested oocytes from the same replications (50 oocytes per pool) were utilized for Western blot analysis.
MIR21 expression in oocytes with and without LH and FSH during in vitro maturation
To determine the effect of LH and FSH on MIR21 expression in oocytes during in vitro maturation COCs were matured in defined maturation media with LH and FSH as described above or containing only LH, only FSH, or lacking both. COCs were washed four times and cultured in the designated hormone treatment in groups of 80–90 COCs per well. COCs were cultured for 42 h, denuded of cumulus cells by vortexing as described above and MII oocytes were identified by the presence of a polar body. This experiment consisted of four biological replications. Maturation rates, as defined by the percentage of oocytes achieving MII arrest, were recorded and MII oocytes from each treatment and replication were collected in pools and used for MIR21 expression analysis as described above.
MIR21 expression in oocytes cultured with and without cumulus cells
To determine the effect of cumulus cell presence on MIR21 expression in oocytes during in vitro maturation we subjected COCs to one of three treatments: 1) standard in vitro maturation as described above using intact COCs, 2) in vitro maturation following cumulus cell removal and then utilization of detached cumulus cells for culture with denuded oocytes or, 3) denuded oocytes matured without the presence of cumulus cells. Cumulus cells were removed by gentle vortex in 1 % hyaluronidase in TL-Hepes, washed twice in 200 μL of maturation media and then resuspended in 200 μL of maturation media and added to the in vitro maturation culture plates which contained 300 μL maturation media and the denuded oocytes. Final volume of culture media for all plates was 500 μL and each well contained 75–85 oocytes. This experiment consisted of four biological replications. Maturation rates, as defined by the percentage of oocytes achieving MII arrest, were recorded and MII oocytes from each treatment and replication were collected in pools and used for MIR21 expression analysis as described above.
MIR21 inhibition during in vitro maturation
Peptide nucleic acids (PNA) are artificially constructed oligonucleotides with strong affinity and specificity to endogenous nucleotides while resistant to nucleases making them ideal for miRNA inhibition [
31]. We used an anti-MIR21 PNA (Panagene Inc. Daejeon, Korea) designed to specifically bind to and prevent MIR21 activity. A scrambled PNA with no predicted targets was used as a negative control (Panagene Inc., Daejeon, Korea). PNA oligonucleotides were diluted in maturation media at a stock concentration of 100 nM/μL and then added to maturation media on the day of COC collection to acquire a final concentration of 2.0 nM and 0.2 nM. A control group without PNA was used to evaluate the potential toxicity of the PNA.
Parthenogenetic activation of MII oocytes, used to remove the confounding effects of miRNA introduced by sperm, was performed with 50 oocytes from each treatment to determine developmental competence of embryos up to 60 h. MII oocytes were washed in a high calcium activation media (Mannitol 0.28 M, CaCl
2 1.0 mM, MgCl
2 0.1 mM, HEPES 0.5 mM and BSA 1 mg/mL), then placed between two electrodes covered with activation media and activated by two consecutive 30 μsec pulses at 1.2 kV/cm. Following activation, zygotes were washed and cultured in porcine zygote medium-3 (PZM3) at 39 °C in 5 % CO
2 [
21]. At 60 h embryos were evaluated for development and the number of embryos with four or more uniform blastomeres was recorded.
Quantitative RT-PCR of oocytes and cumulus cells
Oocytes were collected and denuded of cumulus cells as described above. Oocytes from each stage of development and treatment were collected in pools of exactly 25 oocytes in 5 μL of PBS. As before using a precise number of oocytes per reaction we were able to avoid the introduction of additional variation associated with reference genes [
41]. Both
PDCD4 and MIR21 analysis were analyzed from the same oocyte sample lysis. TaqMan™ Gene Expression Cells-to-Ct™ Kit (Applied Biosystems, Carlsbad, CA) was used to lyse oocytes and prepare samples for RT-qPCR. Lysis solution and DNase from the Cells-to-Ct kit were added to each oocyte pool at 4.95 and 0.05 μL, respectively, and incubated at RT for 5 min. Stop solution (0.5 μL) was added, incubated for an additional 2 min and placed on ice. Two μl of the sample lysis was added to each RT-qPCR reaction.
PDCD4 forward (5′-ACAGTTGGTGGGCCAGTTTATTGC-3′) and reverse (5′-CTTTGCGCCTTCCACCTTTAGACA- 3′) primers were used to determine mRNA abundance of
PDCD4 within each pool. QuantiTect® SYBR® Green RT-PCR Kit (Qiagen) was used for the RT-qPCR reaction for
PDCD4 according to manufacturer’s recommendations. The standard cycling conditions were 50 °C for 30 min, 95 °C for 15 min followed by 45 cycles of 95 °C for 15 s, 60 °C for 30 s, and 72 °C for 30 s followed by melting curve analysis.
MIR21 was quantified using TaqMan® MicroRNA Reverse Transcription kit (Applied Biosystems Carlsbad, CA) for the reverse transcription (RT) reaction and the primers and probe used were TaqMan® MicroRNA Assay for hsa-MIR21 (Applied Biosystems, Carlsbad, CA) according to manufacturer’s recommendations. The RT reaction was 20 μL consisting of 13 μL master mix, 3 μL primers, and 4 μL sample lysis. Reverse transcription conditions were 16 °C for 30 min, 42 °C for 30 min and 85 ° C for 5 min. The final volume for all RT-qPCR reactions was 20 μL which include 1.33 μL of the RT product, 1 μL TaqMan MicroRNA Assay (20x), 10 μL TaqMan 2X Universal PCR Master Mix and 7.67 nuclease free water. The thermal cycling conditions for the TaqMan MicroRNA RT-qPCR were 95 °C for 10 min, followed by 45 cycles of 95 °C for 15 s and 60 °C for 60 s. Fluorescent data acquisition was during the 60 °C extension step.
For analysis of MIR21 expression in cumulus cells total RNA was extracted from cumulus cells of GV stage and MII arrested oocytes using the mirVana RNA isolation kit (Life Technologies, Grand Island, NY), 10 ng of total RNA was utilized for the RT reaction conducted the same way as the oocyte samples. For cumulus cells, cycle threshold values were normalized to expression of another small RNA, RNU43, prior to comparison between stages and statistical analysis. All samples were assayed in duplicate. The comparative C
T method was used to calculate relative fold changes between samples as previously described [
35].
PDCD4 Western blot analysis
Pools of 50 denuded GV and MII oocytes within a replication were collected, washed in PBS, and stored at −80 °C until used for Western blot analysis. Oocyte pools were lysed in 2.5 μL of 5X SDS (total sample volume 12.5 μL) at 95 °C for 4 min followed by 1 min on ice and then centrifugation at 1000 rpm for 1 min at RT. Samples were then loaded into a 4–20 % Tris glycine gel (Lonza PAGEr® Gold Precast Gels). The BioRad Mini PROTEAN Tetra System was used to run the gel at 60 V for 30 min followed by 120 V for 90 min. The gel was transferred to a nitrocellulose membrane for 1 h at 100 V at 4 °C. In addition to utilizing an exact number of oocytes per lane, Ponceau S staining was used to confirm relative transfer efficiency between lanes and equivalent total protein loading per lane. Membrane blocking was conducted using 5 % milk in PBST (PBS with 0.5 % Tween 20) for 1 h at RT. A rabbit anti-PDCD4 monoclonal antibody (Abcam, ab79405) was added (1:1000 dilution) to the membrane in 0.5 % milk in PBST overnight at 4 °C and a negative control membrane lacking primary antibody was also conducted. Following primary antibody incubation, the membranes were washed with PBST three times at RT for 10 min each. Donkey anti-Rabbit IgG (Amersham™ ECL™ NA934) was incubated (1:2000) with the membrane for 1 h at RT. The membrane was then washed three times for 10 min each at RT. Horseradish peroxidase substrate (Millipore, Billerica, MA) was added to the membrane for 1 min in the dark. The membrane was then exposed to x-ray film and developed for visualization. Average pixel intensity for the protein corresponding to 52 kDa (PDCD4 molecular weight) was conducted using Image J [
1].
MIR21 in situ hybridization in the developing follicle
Ovaries were preserved in 4 % paraformaldehyde and utilized for in situ hybridization to determine MIR21 expression. Ovary sections (5 μm) were then mounted on slides for analysis. Each section was subjected to CitriSolv (Fisher Scientific) twice for five minutes rehydrated in two changes each of 100 % ethanol for 3 min, followed by 95 % ethanol for one minute and finally 80 % ethanol for one minute. Slides were immersed in heated citrate buffer (95 °C) for 30 min and then cooled to RT. Once at RT slides were blocked for 30 min with 5 % BSA. Slides were then placed in hybridization solution for 1 h at 65 °C. The 5′ fluorescein labeled miRCURY LNA detection probe (Exiqon) was added and slides were incubated in high humidity overnight at 65 °C. Slides were then washed in saline-sodium citrate (SSC) solution and PBST at RT. Anti-fade DAPI was added and a cover slip was placed over each section. Primary and secondary follicles were imaged at 400X and tertiary follicles were imaged at 200X with a Leica microscope.
Statistical analysis
PROC MIXED of the Statistical Analysis System was used to determine statistical differences of all data including percentage maturation and differences in CT value for RT-qPCR data. Significance (P < 0.05) was determined for the model and least-square means was used to determine significant differences between samples. The effect of oocyte stage on MIR21 and PDCD4 expression (CT value) was determined. Treatment effect for the PNA inhibitor of MIR21, presence or absence of gonadotropins, and the presence or absence of cumulus cells during in vitro maturation on MIR21 expression was evaluated. Replication was included as a covariate. Graphs depicting percent change were adjusted to reflect the GV stage as 100 % and all other treatments are relative to GV. Data are displayed as mean ± SEM.
Discussion
While abundant in the developing oocyte and mouse ovary [
11,
14] miRNA expression and function during pig oocyte maturation and early embryo development is only in the initial stages of characterization [
18,
41]. The potential ability of MIR21 to interact with
PDCD4 leading to posttranscriptional gene regulation of PDCD4 protein expression in the maturing pig oocyte as demonstrated herein has also been described in several types of cancer cells [
3,
15,
16,
29,
33,
42]. The interaction between MIR21 and PDCD4 is likely conserved in pigs as the MIR21 target recognition sequence in the 3′UTR of human and pig PDCD4 is 97 % similar with 100 % similarity in nucleotides responsible for recognition by the MIR21 seed sequence.
Although a multitude of reports demonstrating the biological activity of miRNA in somatic cells exists miRNA function in maturing oocytes has been debated, primarily the result of the conditional knock-out during oocyte maturation of enzymes needed for both canonical and non-canonical miRNA biogenesis pathways. Loss of DICER function, essential for both siRNA and canonical miRNA, in developing oocytes results in the production of non-viable oocytes [
38]. Alternatively, using the same ZP3-cre conditional knock-out approach with a floxed DGCR8 allele, mice lacking the ability to produce miRNA through the canonical biogenic pathway still produced viable oocytes, despite reduced fecundity [
25,
36]. This suggests that while miRNA activity in mature mice oocytes may be suppressed, miRNA may still contribute to the developmental competency of the subsequently produced embryo. This study tested the hypothesis that increased MIR21 abundance in the maturing cumulus oocyte complex of the pig is associated with posttranscriptional regulation of
PDCD4 expression in the oocyte and that suppression of MIR21 function during oocyte maturation would compromise subsequent embryonic development.
Utilizing miRNA microarray and small RNA sequencing, the expression of numerous miRNA in the maturing porcine cumulus oocyte complex have been previously demonstrated [
41]. The current study further characterizes the temporal relationship between increased mature MIR21 abundance and decreased PDCD4 protein abundance during in vitro maturation of pig oocytes. In cumulus cells, MIR21 abundance increases approximately 25-fold during in vitro maturation. This is consistent with reports in humans [
4] and in mice demonstrating increased MIR21 expression in granulosa cells in response to LH [
11]. Because MIR21 abundance changes appear to occur rapidly during in vitro maturation, it is of interest if MIR21 expression in the pig cumulus oocyte complex is also responsive to the gonadotropins. To examine this, the ability of both LH and FSH to affect MIR21 expression during in vitro maturation was examined. While exclusion of LH from maturation media negatively impacted oocyte maturation, as demonstrated by others [
45], a significant effect of the gonadotropins on MIR21 abundance during in vitro maturation was not detected.
Because changes in MIR21 abundance in the oocyte did not appear to be greatly influenced by LH and FSH other mechanisms that may contribute to changes in MIR21 abundance were also investigated. A potential mechanism explaining the observed increase of MIR21 abundance in IVM derived MII arrested oocytes compared to GV oocytes is that MIR21 could be transported into the oocyte from the cumulus cells during maturation. This is plausible as other factors have been demonstrated to be translocated between the maturing oocyte and the surrounding cumulus oophorus [
19]. Oocyte - cumulus cell communication is bi-directional and required for normal cumulus cell gene expression and oocyte maturation [
20] and gap junctions allow the transit of molecules less than 1000 Da including ATP, sodium, chloride, calcium ion and cAMP [
2,
26]. Therefore it is feasible that miRNA could also be transported into the oocyte from surrounding cumulus cells. In addition to gap junctions, small molecule transport between cells has also been demonstrated to occur through exosomes and other microvesicles that can be secreted and taken up by neighboring cells [
39,
40]. The potential for these mechanisms to contribute to MIR21 abundance increasing in the oocyte was examined by culturing denuded oocytes during IVM in the presence or absence of cumulus cells. While a reduction in maturation rate was observed as has been previously demonstrated [
43], MIR21 abundance in the oocytes that did achieve MII arrest was not affected by the presence of cumulus cells suggesting the increased MIR21 observed is at least in part, the result of oocyte specific mechanisms.
Taken together neither gonadotropins nor the presence of cumulus cells during IVM is demonstrably responsible for increased MIR21 abundance consistently observed in MII arrested oocytes compared to their GV stage counterparts. It has also been observed that GVBD does not occur in the majority of porcine oocytes until approximately 16–24 h of in vitro maturation [
30]. Therefore, it remains possible that the increased abundance of MIR21 in the MII arrested oocyte occurs as a transcriptional response in the oocyte during the initial phases of in vitro maturation prior to GVBD in the oocyte.
These findings taken together suggest the potential for MIR21 and miRNA in general, to impact protein expression in the maturing oocyte having implications regarding the developmental ability of subsequently produced embryos. Oocyte growth and development begins prior to antral follicle development and it is unclear when activation of MIR21 expression occurs or what mechanism is primarily responsible for this observation. The cumulus oocyte complexes collected for this study were from 3 to 5 mm antral follicles, prior to GVBD, and may be capable of transcribing primary MIR21 (pri-MIR21) transcript that can be further processed into mature MIR21 during maturation. Future studies will include an analysis of pri-MIR21 abundance prior to and during oocyte maturation which may yield insight into the mechanisms contributing to the current observations.
Several promoters have been identified upstream to the pri-MIR21 transcription start site containing predicted consensus sequences for binding of AP-1 and signal transducer and activator of transcription 3 (STAT3) [
17]. In certain cancers, the interaction between MIR21 and PDCD4 is necessary for maximal AP-1 activation. AP-1 activation is suppressed by PDCD4, however, AP-1 induction of MIR21 and subsequent posttranscriptional regulation of PDCD4 allows further and more sustained AP-1 activation [
17,
37]. In addition to AP-1, STAT3 is another transcription factor that has also been documented to induce pri-MIR21 transcription [
23]. It is possible that STAT3 could be related to the expression of MIR21 during oocyte maturation and early embryo development as leptin (an adipokine) has been demonstrated to increase STAT3 expression during bovine embryo culture and is associated with reduced apoptosis in bovine blastocysts [
7]. In addition, leptin has been shown to increase STAT3 expression in both cumulus cells and oocytes during in vitro maturation and enhance the oocytes ability to complete meiosis [
32].
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
ECW and BJH conducted experiments in the manuscript. CY and JGN assisted with completing experiments. ECW, BJH and JWR were responsible for experimental design and analysis of results. ECW, BJH and JWR prepared the manuscript. All authors read and approved the final manuscript.