Expression of microRNAs and their target genes and pathways ...

Physiol Genomics 46: 735–745, 2014. First published August 5, 2014; doi:10.1152/physiolgenomics.00036.2014.

Expression of microRNAs and their target genes and pathways associated with ovarian follicle development in cattle A. E. Zielak-Steciwko,1 J. A. Browne,2 P. A. McGettigan,2 M. Gajewska,3 M. Dzie˛cioł,4 T. Szulc,1 and A. C. O. Evans2 1

Institute of Animal Breeding, Wrocław University of Environmental and Life Sciences, Wrocław, Poland; 2School of Agriculture and Food Science, University College Dublin, Belfield, Dublin, Ireland; 3Department of Physiological Sciences, Warsaw University of Life Sciences, Warsaw, Poland; and 4Department of Reproduction and Clinic of Farm Animals, Wrocław University of Environmental and Life Sciences, Wrocław, Poland

Submitted 20 March 2014; accepted in final form 4 August 2014

Zielak-Steciwko AE, Browne JA, McGettigan PA, Gajewska M, Dzie˛cioł M, Szulc T, Evans ACO. Expression of microRNAs and their target genes and pathways associated with ovarian follicle development in cattle. Physiol Genomics 46: 735–745, 2014. First published August 5, 2014; doi:10.1152/physiolgenomics.00036.2014.—Development of ovarian follicles is controlled at the molecular level by several gene products whose precise expression leads to regression or ovulation of follicles. MicroRNAs (miRNAs) are small noncoding RNAs that regulate gene expression through sequence-specific base pairing with target messenger RNAs (mRNAs) causing translation repression or mRNA degradation. The aim of this study was to identify miRNAs expressed in theca and/or granulosa layers and their putative target genes/pathways that are involved in bovine ovarian follicle development. By using miRCURY microarray (Exiqon) we identified 14 and 49 differentially expressed miRNAs (P ⬍ 0.01) between dominant and subordinate follicles in theca and granulosa cells, respectively. The expression levels of four selected miRNAs were confirmed by qRT-PCR. To identify target prediction and pathways of differentially expressed miRNAs we used Union of Genes option in DIANA miRPath v.2.0 software. The predicted targets for these miRNAs were enriched for pathways involving oocyte meiosis, Wnt, TGF-beta, ErbB, insulin, P13K-Akt, and MAPK signaling pathways. This study identified differentially expressed miRNAs in the theca and granulosa cells of dominant and subordinate follicles and implicates them in having important roles in regulating known molecular pathways that determine the fate of ovarian follicle development. microRNA; targets; signaling; ovarian follicles IN CATTLE, FOLLICLE DEVELOPMENT beyond the early antral stage is characterized by two or three successive waves of follicular growth in each estrous cycle. During follicular wave, a single follicle is selected to continue growing and becomes dominant, while other subordinate follicles undergo atresia (18). It is well established that the fate of the growing follicles (ovulation or atresia) is regulated by the interaction of endocrine signals (e.g., gonadotropins, their receptors and steroids) and intraovarian molecules [e.g., IGF-I, transforming growth factor (TGF)-␤] produced in granulosa and theca cells (31, 34, 22). These factors are coordinated by expression of numerous genes. Any alteration in the activity of these genes might be critical in determining the survival of dominant follicles or may play a role in the demise of subordinate follicles (6, 13, 15, 30).

Recent studies have shown that microRNAs (miRNAs) have a significant impact on gene expression in a variety of tissues and biological processes, both in humans (12) and animals (14, 40). These small noncoding RNA molecules function via partial base pairing of their seed region (2- to 7-nucleotide-long region from the 5=-end of the miRNA) with complementary sequences to the 3=-untranslated region (UTR) of their target genes (1, 5). This interaction usually results in gene silencing through translation repression or direct mRNA degradation (3). Recent findings have proven that miRNAs are important regulators of development, cell proliferation, and apoptosis and are involved in many physiological processes, including reproduction (2). It is anticipated that these molecules are an important element in the mechanism of the regulation of ovarian follicle development in mammals. Several studies have reported that a number of miRNAs are involved in murine granulosa cell proliferation (43, 44) and estradiol production (45). Similar observations were made in porcine granulosa cells where miRNAs regulate estradiol production and apoptosis (26, 42). Furthermore, miRNAs have also been detected in ovine granulosa and theca cells at different stages of follicle development (28) and suggested to play a role in the regulation of cell survival, steroidogenesis, and follicle differentiation in equine ovaries (36). To date, there is little known about the role of miRNAs in bovine follicle development and in particular in dominant follicle selection. In an earlier study, miRNAs were identified in whole bovine ovaries without distinguishing specific tissue compartment and follicle classes (19). Recently two reports have indicated expression of miRNAs in bovine granulosa cells from various sized follicles (32) and in cultured bovine theca and granulosa cells (28). Considering the limited knowledge on the role of miRNAs in bovine ovarian follicle development and in dominant follicle selection the aim of the present work was: 1) to determine if miRNAs are expressed locally within bovine antral follicles and to localize expression within the theca and/or granulosa layer, 2) to determine if the pattern of miRNA expression is different between the dominant and the largest subordinate follicle from the first follicle wave of the cycle, and 3) to use a bioinformatic approach to identify the potential pathways and putative targets for the selected miRNAs involved in bovine follicle development. MATERIAL AND METHODS

Address for reprint requests and other correspondence: A. Zielak-Steciwko, Inst. of Animal Breeding, Wrocław Univ. of Environmental and Life Sciences, ul. Chełmo´nskiego 38c, 50-375 Wrocław, Poland (e-mail: anna.zielak @up.wroc.pl).

Animal study and tissue collection. To obtain ovarian follicles from the first wave, six cross-breed beef heifers were subjected to an estrus synchronization program combined with daily ultrasonography of

1094-8341/14 Copyright © 2014 the American Physiological Society

735

736

ROLE OF miRNAs IN OVARIAN FOLLICLE DEVELOPMENT

ovaries. Animals were kept and fed under the same conditions. Heifers were synchronized with an 8-day progesterone-releasing intravaginal device (CIDR, Pfizer Animal Health) and a single intramuscular injection (25 mg) of prostaglandin F2 alpha (Dinolityc, Pfizer) on the seventh day. Estrus was observed every 6 h between 24 and 48 h after CIDR withdrawal. Each animal was subjected to daily, individual ovarian ultrasound scans to monitor ovarian follicle development by real-time B mode scanner (HS-1500 Vet, Honda Electronics, Japan) with 5 MHz transrectal linear probe. Animals were slaughtered between day 2.5 and 3.5 of the synchronized estrus cycle (day 0 ⫽ onset of estrus). The two largest follicles were collected from the ovaries of each animal, and individual follicle diameter was measured. Follicular fluid was aspirated from the follicles, snap-frozen in liquid nitrogen, and stored at ⫺80°C for hormone analyses. Theca and granulosa cells were separated as previously described (13), snapfrozen, and stored at ⫺80°C prior to RNA extractions. All experimental procedures were licensed by the 2nd Local Ethics Committee at Wrocław University of Environmental and Life Sciences, Poland. Estradiol and progesterone concentrations were measured using commercially available kits Coat-A-Count Estradiol RIA and CoatA-Count Progesterone RIA kit, respectively (Siemens Healthcare Diagnostics, Tarrytown, NY). Based on the intrafollicular ratio of estradiol to progesterone, follicles were hormonally classified as estrogen active follicles (dominant) - ratios of ⬎ 1 or atretic follicles (subordinate) - ratios of ⬍ 1 (20). Follicle diameter and follicular fluid hormone concentrations between the two follicles from each individual were analyzed by ANOVA using the SAS 9.2 software. Results are shown as means ⫾ SE. Statistical significance was accepted when P ⬍ 0.05. RNA isolation, miRNA microarray, and data analysis. Total RNA, including small RNA was isolated from theca and granulosa samples using miRNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions followed by DNase treatment step using RNase-Free DNase set (Qiagen). Total RNA quantity was determined by NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE) and quality was assessed using the Agilent Bioanalyzer 2100 with RNA 6000 Nano Chip kit (Agilent Technologies, Santa Clara, CA). All RNA samples were shown to have RNA integrity number values ⬎ 7.5. To exclude cross-contamination between theca and granulosa cells, we verified the mRNA expression profiles of CYP19A1 and CYP17A1 genes by quantitative real-time PCR (qRT-PCR) using the ABI Prism 7500 FAST sequence detection system and Fast SYBR

Green Master Mix (Applied Biosystems, Warrington, UK). Primers were designed for each gene of interest (Primer Express Software v2.0, Applied Biosystems): CYP19A1 (forward: 5=-TGG TGA CCA TCT GTG CTG AT-3= and reverse: 5=-GTC AAC ACG TCC ACA TAG CC-3=) and CYP17A1 (forward: 5=-CTG GAG GTT CTG AGC AAG GA-3= and reverse: 5=-TGG CTT TGC TGG GGA AAA TC-3=). The specificity of all primers was confirmed both by meltcurve analysis and by agarose gel electrophoresis of the amplified PCR fragments. Primer efficiency was determined using a serial dilution of Bos taurus-derived cDNA (1:4 dilution series over 7 points) and shown to lie between 90 and 110%. The cDNA prepared for the miRNA study (of each sample) was also used to perform the mRNA analysis. The optimal number of reference targets for this sample set were identified using the geNorm application within the qbasePLUS software package (Biogazelle, Zwijnaarde, Belgium). The normalization factor was calculated as the geometric mean of reference targets YWHAZ (forward: 5=-TGA AGC CAT TGC TGA ACT TG-3= and reverse: 5=-TCT CCT TGG GTA TCC GAT GT-3=) and PPIA (forward: 5=-CAT ACA GGT CCT GGC ATC TTG TCC-3= and reverse: 5=-CAC GTG CTT GCC ATC CAA CC-3=). Calibrated normalized relative quantities of gene expression for each analyzed sample were generated by the qbasePLUS package and analyzed within the package using an unpaired t-test. MiRNA expression profiling was performed by Exiqon (Exiqon, Vedbaek, Denmark). The samples from each tissue were analyzed separately. Two tissue pools comprising RNA from all granulosa samples and all theca samples respectively were generated and used as the control sample on each slide. There were six samples for each follicle type (dominant and subordinate) within each tissue (theca and granulosa) for a total of 24 slides. In brief, the samples were labeled using the miRCURY LNA microRNA Hi-Power Labeling Kit, Hy3/ Hy5, and hybridized on the miRCURY LNA microRNA Array 6th Gen (Exiqon). Slides were scanned using the Agilent G2565BA Microarray Scanner System (Agilent Technologies), and the image analysis was conducted in ImaGene 9 (miRCURY LNA microRNA Array Analysis Software, Exiqon). Raw file data were analyzed using the limma R-package (37). After background correction (method ⫽ “subtract”) and normalization of samples (normalizeWithinArrays no offset), probes with no name, control probes, and spike in probes were eliminated from further analysis. One sample (granulosa subordinate follicle) exhibited an expression profile markedly different from all the other samples and was excluded from the granulosa analysis. Analysis of differential expression was performed by lmFit

Table 1. MiRNAs in theca cells that were differentially expressed in dominant compared with the largest subordinate follicles Microarray MiRNA Name

Accession Number

Sequence

Ratio

FDR P Value

hsa-miR-301b/bta-miR-301b hsa-miR-190b/bta-miR-190b hsa-miR-1301 hsa-miRPlus-I181b-2* hsa-miR-1255b-5p hsa-miR-1184 hsa-miR-let-7i-3p hsa-miR-129-2-3p/bta-miR-129-3p hsa-miR-548aa/hsa-miR-548t-3 hsa-miR-3684 hsa-miR-29b-1-5p hsa-miR-302e hsa-miR-196a-3p has-miR-1284

MIMAT0004958 MIMAT0004929 MIMAT0005797 MIMAT0017084 MIMAT0005945 MIMAT0005829 MIMAT0004585 MIMAT0004605 MIMAT0018447 MIMAT0018112 MIMAT0004514 MIMAT0005931 MIMAT0004562 MIMAT0005941

cagugcaaugauauugucaaagc ugauauguuugauauuggguu uugcagcugccugggagugacuuc cucacugaucaaugaaugcaaa cggaugagcaaagaaagugguu ccugcagcgacuugauggcuucc cugcgcaagcuacugccuugcu aagcccuuaccccaaaaagcau aaaaaccacaauuacuuuugcacca uuagaccuaguacacguccuu gcugguuucauauggugguuuaga uaagugcuuccaugcuu cggcaacaagaaacugccugag ucuauacagacccuggcuuuuc

18.44 12.84 11.59 5.11 3.98 3.74 1.24 ⫺8.68 ⫺5.88 ⫺4.65 ⫺3.97 ⫺3.09 ⫺3.04 ⫺1.31

⬍0.001 0.008 ⬍0.001 0.002 0.002 0.008 0.010 0.010 0.002 0.002 0.004 0.007 0.006 0.009

Data from heifers (n ⫽ 6) collected at dominance stage of the 1st follicular wave determined by DNA microarrays. Positive ratio values indicate greater expression in the dominant compared with the subordinate follicle and negative values vice versa. MiRNA, microRNA; FDR, false discovery rate. In Tables 1 and 2, miRNAs in boldface were rejected from the target prediction and pathway analysis as they were not available in the database of DIANA miRPath software. Physiol Genomics • doi:10.1152/physiolgenomics.00036.2014 • www.physiolgenomics.org

737


and empirical Bayes functions in Limma package of Bioconductor (38). The Benjamini-Hochberg correction was used to control for false discovery rate (FDR). The FDR was set at 1% (4). The raw data from the experiment were deposited in the Gene Expression Omnibus repository under accession number GSE55890. miRNA target gene prediction and pathway analysis. Differentially expressed miRNAs (P ⬍ 0.01), between dominant and subordinate follicles in theca and granulosa cells, recognized by microarray, were further analyzed to identify their target prediction and pathway analysis. For this purpose, the Union of Genes option, in the second version of DIANA miRPath (41) software, was used. This computational tool performs an analysis between the miRNAs and targeted genes compiled in UNION_SET. Enrichment investigation identi-

fies pathways significantly enriched with genes belonging to the UNION_SET (a priori method) in Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways (21a). Prediction was performed by using DIANA-microT-CDS with MicroT threshold set to 0.8 score. Benjamini and Hochberg’s FDR was applied with significant threshold set at P ⬍ 0.01 (except for hsa-miR-18a-5p/bta-miR-18a for which significant threshold was set at P ⬍ 0.05, because no predicted pathways were found at P ⬍ 0.01). qRT-PCR analysis of miRNAs. Bovine miRNA sequences are not integrated with the current version of DIANA web server. Therefore, four potentially differentially expressed miRNAs (hsa-miR-301b/btamiR-301b, hsa-miR-129-2-3p/bta-miR-129-3p, hsa-miR-18a-5p/btamiR-18a, and hsa-miR-582-5p/bta-miR-582), which were human and

Table 2. MiRNAs in granulosa cells that were differentially expressed in dominant compared with the largest subordinate follicles Microarray MiRNA Name

Accession Number

hsa-miR-548b-3p hsa-miR-212-5p/bta-miR-212 hsa-miR-3689b-3p/hsa-miR-3689c hsa-miR-548h-5p hsa-miR-33a-3p hsa-miR-564 hsa-miR-671-5p/bta-miR-671 hsa-miR-3153 hsa-miRPlus-G1140-3p hsa-miR-582-5p/bta-miR-582 hsa-miR-1226-3p hsa-miR-597 hsa-miR-190b/bta-miR-190b hsa-miR-3175 hsa-miR-4292 hsa-miR-125a-3p hsa-miR-18a-5p/bta-miR-18a hsa-miR-569 hsa-miR-3190-3p hsa-miR-3605-3p hsa-miR-93-3p hsa-miR-548t-5p hsa-miR-3663-3p hsa-miR-216a/bta-miR-216a hsa-miR-524-5p hsa-miR-3146 hsa-miR-30d-3p hsa-miR-148b-5p hsa-miR-555 hsa-miR-887 hsa-miRPlus-C1087 hsa-miR-4254 hsa-miR-219-5p/bta-miR-219-5p hsa-miR-582-5p/bta-miR-582 hsa-miR-583 hsa-miR-520f hsa-miR-615-5p/bta-miR-615 hsa-miR-548h-3p/hsa-miR-548z hsa-miR-4314 hsa-miR-185-3p hsa-miR-26a-1-3p bta-miR-362-5p hsa-miR-3178 hsa-miR-1469 hsa-miR-4285 hsa-miR-659-3p hsa-miR-4288 hsa-miR-625-3p hsa-miR-3195

MIMAT0003254 MIMAT0022695 MIMAT0018181 MIMAT0005928 MIMAT0004506 MIMAT0003228 MIMAT0003880 MIMAT0015026 N/A MIMAT0003247 MIMAT0005577 MIMAT0003265 MIMAT0004929 MIMAT0015052 MIMAT0016919 MIMAT0004602 MIMAT0000072 MIMAT0003234 MIMAT0022839 MIMAT0017982 MIMAT0004509 MIMAT0015009 MIMAT0018085 MIMAT0000273 MIMAT0002849 MIMAT0015018 MIMAT0004551 MIMAT0004699 MIMAT0003219 MIMAT0004951 N/A MIMAT0016884 MIMAT0000276 MIMAT0003247 MIMAT0003248 MIMAT0026609 MIMAT0004804 MIMAT0022723 MIMAT0016868 MIMAT0004611 MIMAT0004499 MIMAT0009298 MIMAT0015055 MIMAT0007347 MIMAT0016913 MIMAT0003337 MIMAT0016918 MIMAT0004808 MIMAT0015079

Sequence

caagaaccucaguugcuuuugu accuuggcucuagacugcuuacu cugggaggugugauauuguggu aaaaguaaucgcgguuuuuguc caauguuuccacagugcaucac aggcacggugucagcaggc aggaagcccuggaggggcuggag ggggaaagcgaguagggacauuu N/A uuacaguuguucaaccaguuacu ucaccagcccuguguucccuag ugugucacucgaugaccacugu ugauauguuugauauuggguu cggggagagaacgcagugacgu ccccugggccggccuugg acaggugagguucuugggagcc uaaggugcaucuagugcagauag aguuaaugaauccuggaaagu uguggaagguagacggccagaga ccuccguguuaccuguccucuag acugcugagcuagcacuucccg caaaagugaucgugguuuuug ugagcaccacacaggccgggcgc uaaucucagcuggcaacuguga cuacaaagggaagcacuuucuc caugcuaggauagaaagaaugg cuuucagucagauguuugcugc aaguucuguuauacacucaggc aggguaagcugaaccucugau gugaacgggcgccaucccgagg N/A gccuggagcuacuccaccaucuc ugauuguccaaacgcaauucu uuacaguuguucaaccaguuacu caaagaggaaggucccauuac ccucuaaagggaagcgcuuucu ggggguccccggugcucggauc caaaaaccgcaauuacuuuugca cucugggaaaugggacag aggggcuggcuuuccucugguc ccuauucuugguuacuugcacg aauccuuggaaccuaggugugagu ggggcgcggccggaucg cucggcgcggggcgcgggcucc gcggcgaguccgacucau cuugguucagggagggucccca uugucugcugaguuucc gacuauagaacuuucccccuca cgcgccgggcccggguu

Ratio

18.08 15.98 14.13 12.33 11.15 10.28 9.77 6.85 4.82 3.56 1.88 1.55 1.41 1.39 1.36 1.35 1.32 ⫺13.03 ⫺11.14 ⫺10.96 ⫺10.34 ⫺10.13 ⫺9.72 ⫺9.00 ⫺8.59 ⫺8.25 ⫺7.11 ⫺6.71 ⫺5.60 ⫺4.84 ⫺4.66 ⫺4.53 ⫺4.50 ⫺4.50 ⫺3.60 ⫺3.53 ⫺3.51 ⫺3.28 ⫺2.60 ⫺2.11 ⫺1.04 ⫺1.01 ⫺1.84 ⫺1.53 ⫺1.46 ⫺1.43 ⫺1.41 ⫺1.40 ⫺1.31

FDR P Value

⬍0.001 0.005 0.006 0.007 0.004 0.001 ⬍0.001 ⬍0.001 ⬍0.001 0.002 0.006 0.009 0.009 0.002 0.007 0.009 0.009 0.007 0.001 ⬍0.001 0.009 0.009 0.009 0.010 ⬍0.001 0.001 0.001 0.005 ⬍0.001 0.001 ⬍0.001 ⬍0.001 0.004 0.007 0.007 ⬍0.001 0.002 0.005 0.001 0.009 0.006 0.002 ⬍0.001 0.007 0.005 ⬍0.001 0.008 0.001 0.008

Data from heifers (n ⫽ 6) collected at dominance stage of the first follicular wave determined using DNA microarrays. Positive ratio values indicate greater expression in the dominant compared with the subordinate follicle and negative values vice versa. N/A, not available. Physiol Genomics • doi:10.1152/physiolgenomics.00036.2014 • www.physiolgenomics.org

738


Table 3. Selected miRNAs differentially expressed between dominant and the largest subordinate follicles in theca and granulosa cells from heifers (n ⫽ 6) collected at dominance stage of the first follicular wave determined by microarrays and qRT-PCR Microarray

qRT-PCR

MiRNA name

Accession Number

Tissue Type

Ratio

P Value

Ratio

P Value

hsa-miR-301b/bta-miR-301b hsa-miR-129-2-3p/bta-miR-129-3p hsa-miR-18a-5p/bta-miR-18a hsa-miR-582-5p/bta-miR-582

MIMAT0004958 MIMAT0004605 MIMAT0000072 MIMAT0003247

TC TC GC GC

18.44 ⫺8.68 1.32 ⫺4.49

⬍0.001 0.010 0.008 0.007

2.82 ⫾ 0.36 ⫺2.06 ⫾ 0.55 1.90 ⫾ 0.11 ⫺1.35 ⫾ 0.45

0.003 0.047 0.047 0.049

TC, theca cell; GC, granulosa cell.

bovine orthologs, were validated by qRT-PCR. MiRNA expression profiling was performed using the miScript system (Qiagen, Hilden, Germany) as per manufacturer’s instructions. RNU6 was used as internal control. In brief, 1,000 ng of total RNA was reverse transcribed using the miScript II Reverse Transcriptase in a 20 ␮l reaction; this was subsequently diluted 1:40 using RNase/DNase-free water. qRT-PCR was conducted using a 7500 Fast Real-Time PCR machine (Applied Biosystem, Foster City, CA) in 96-well plates using miScript SYBER Green PCR Kit (Qiagen); each reaction (25 ␮l) contained 12.5 ␮l of 2⫻ QuantiTec SYBER Green PCR Master Mix, 2.5 ␮l of 10⫻ miScript Universal Primer, 2.5 ␮l of 10⫻ miScript Primer Assay, and 5 ␮l of the diluted template cDNA. Dissociation curves of PCR reactions were monitored to ensure a single specific PCR product, and the appropriate negative and positive controls were included. Changes in relative concentration were calculated with the qBasePLUS Software (Biogazelle), and relative expression levels were compared by ANOVA using the SAS 9.2 software. Results are expressed as means ⫾ SE. Statistical significance was accepted when P ⬍ 0.05. RESULTS

miRNA expression profiling in bovine ovarian follicles. To establish miRNA expression patterns, we collected ovarian Table 4. Significantly enriched signaling pathways (P ⬍ 0.01) associated with differentially expressed miRNAs in theca cells between dominant and subordinate follicles Pathway Name

Dominant Follicles Oocyte meiosis Wnt signaling pathway TGF-beta signaling pathway Protein processing in endoplasmic reticulum RNA degradation Axon guidance Pathways in cancer Subordinate Follicles Neurotropin signaling pathway ErbB signaling pathway Pathways in cancer Insulin signaling pathway Endocytosis Chemokine signaling pathway Regulation of actin cytoskeleton HTLV-I infection Epstein-Barr virus infection

Pathway ID

FDR*

hsa04114 hsa04310 hsa04350 hsa04141 hsa03018 hsa04360 hsa05200

0.0004 0.0016 0.0016 0.0038 0.0043 0.0050 0.0084

hsa04722 hsa04012 hsa05200 hsa04910 hsa04144 hsa04062 hsa04810 hsa05166 hsa05169

5.32E-10 5.93E-09 1.02E-07 1.44E-07 6.20E-06 6.92E-06 2.57E-05 0.0044 0.0066

Dominant follicles: hsa-miR-301b/bta-miR-301b, hsa-miR-190b, hsa-miR1301, hsa-miR-1255b-5p, hsa-miR-1184. Subordinate follicles: hsa-miR-1292-3p, hsa-miR-548aa, hsa-miR-3684, hsa-miR-29b-1-5p, hsa-miR-302e, hsamiR-196a-3p. *FDR correction was calculated with Benjamini-Hochberg with a threshold of 0.01 or 0.05 as indicated.

follicles at dominance stage of the first follicular wave. Mean follicle diameter was greater in the dominant follicle (10.7 ⫾ 0.3 mm, n ⫽ 6) compared with the subordinate follicles (8.7 ⫾ 0.2 mm, P ⬍ 0.01, n ⫽ 6). Follicular fluid estradiol concentrations also were higher (P ⬍ 0.01) in the dominant (177.4 ⫾ 26.8 ng/ml, n ⫽ 6) compared with the largest subordinate follicles (17.8 ⫾ 5.6 ng/ml, n ⫽ 6). In contrast progesterone concentrations were higher (P ⬍ 0.02) in the largest subordinate (100.6 ⫾ 13.8 ng/ml, n ⫽ 6) compared with the dominant follicles (63.4 ⫾ 5.1 ng/ml, n ⫽ 6). Table 5. Significantly enriched signaling pathways (P ⬍ 0.01) associated with the differentially expressed miRNAs in granulosa cells between dominant and subordinate follicles Pathway Name

Pathway ID

Dominant Follicles Neurotropin signaling pathway hsa04722 Gap junction hsa04540 Endocytosis hsa04144 PI3K-Akt signaling pathway hsa04151 MAPK signaling pathway hsa04010 Focal adhesion hsa04510 Transcriptional misregulation in cancer hsa05202 Pathways in cancer hsa05200 Vascular smooth muscle contraction hsa04270 Regulation of actin cytoskeleton hsa04810 Chemokine signaling pathways hsa04062 Subordinate Follicles Wnt signaling pathway hsa04310 MAPK signaling pathway hsa04010 Ubiquitin mediated proteolisis hsa04120 Dopaminergic synapse hsa04728 Neurotropin signaling pathway hsa04722 Axon guidance hsa04360 mRNA surveillance pathway hsa03015 Pathways in cancer hsa05200 PI3K-Akt signaling pathway hsa04151 Osteoclast differentiation hsa04380 HTLV-I infection hsa05166 Endocytosis hsa04144 RNA transport hsa03013 Transcriptional misregulation in cancer hsa05202 Cell cycle pathways hsa04110

FDR*

1.15E-12 3.62E-11 3.20E-08 3.63E-08 1.22E-07 1.22E-07 2.41E-07 2.92E-07 2.16E-06 0.0018 0.0073 6.02E-38 4.25E-34 3.63E-30 7.13E-29 9.79E-27 1.44E-25 1.44E-21 2.37E-21 6.79E-19 2.92E-17 5.65E-14 4.73E-11 2.80E-10 5.29E-06 0.0023

Dominant follicles: hsa-miR-548b-3p, hsa-miR-212-5p, hsa-miR-3689b-3p, hsa-miR-548h-5p, hsa-miR-33a-3p, hsa-miR-564, hsa-miR-671-5p, hsa-miR3153, hsa-miR-582-5p, hsa-miR-1226-3p, hsa-miR-597, hsa-miR-3175, hsamiR-4292, hsa-miR-125a-3p, hsa-miR-18a-5p, hsa-miR-190b. Subordinate follicles: hsa-miR-4288, hsa-miR-659-3p, hsa-miR-185-3p, hsa-miR-4314, hsa-miR-548h-3p, hsa-miR-520f, hsa-miR-583, hsa-miR-582-5p, hsa-miR219-5p, hsa-miR-555, hsa-miR-148b-5p, hsa-miR-30d-3p, hsa-miR-3146, hsamiR-524-5p, hsa-miR-3663-3p, hsa-miR-548t-5p, hsa-miR-93-3p, hsa-miR3190-3p, hsa-miR-569.

Physiol Genomics • doi:10.1152/physiolgenomics.00036.2014 • www.physiolgenomics.org

739


Fig. 1. KEGG map visualization of hsa-miR-301b/bta-miR-301b involvement in prion diseases signaling pathway. Green, genes not targeted by this microRNA (miRNA); yellow, putative targets for this miRNA; red, gene especially important in ovarian follicle physiology.

Cross-contamination analysis between theca and granulosa cells revealed CYP19A1 amplicons in granulosa cells but not in theca cells. While amplicons of CYP17A1 were only detected in theca cell (data not shown).

The microarray contained a total of 1,488 bovine and human miRNAs, from miRBase version 18 (24). Comparing the dominant with the subordinate follicle our initial analysis using a threshold of P ⬍ 0.05 identified a total of 87 differentially

Table 6. Enrichment analysis of miR-301b putative gene targets in KEGG pathways, performed by using DIANA-microT with significant threshold set at P ⬍ 0.01 KEGG Pathway

Prion diseases TGF-beta signaling pathway Phosphatidylinositol signaling system mTOR signaling pathway Endocytosis Gap junction p53 signaling pathway Circadian rhythm Axon guidance Inositol phosphate metabolism Insulin signaling pathway Melanogenesis Prostate cancer Glioma RNA degradation Adipocytokine signaling pathway

Gene Name

FDR*

KEGG Pathway ID

PRNP ROCK1, INHBB, SMURF2, INHBA, ACVR1, SKP1, ZFYVE9, SMAD5, TGFB2, TGFBR2, BMPR2 DGKE, CALM2, PLCB1, PIKFYVE, PIK3C2A, PTEN, ITPK1, PLCB4 TSC1, RRAGD, PRKAA2, PDK1, PRKAA1, EIF4E2, PTEN, ULK2 RNF41, ARAP2, SMURF2, PSD, CLTC, CHMP4B, ASAP1, RAB5A, CBLB, EPS15, ZFYVE9, TGFB2, LDLR, TGFBR2 ADCY1, SOS2, PLCB1, GJA1, ADCY4, PLCB4 ZMAT3, CDS1, GADD45A, MDM4, CDKN1A, SESN3, PTEN PRKAA2, BHLHE41, SKP1, PRKAA1, CLOCK EFNB2, ROCK1, ARHGEF12, ROBO2, EPHB4, DPYSL2, CFL2, NRP1, PAK6, ROBO1 PLCB1, PIKFYVE, PIK3C2A, PTEN, ITPK1, PLCB4 TSC1, SOS2, PRKAA2, CALM2, CBLB, PDK1, PRKAA1, EIF4E2, PPARGC1A, PPP1CB ADCY1, TCF4, WNT2B, CALM2, PLCB1, EDN1, ADCY4, PLCB4 SOS2, E2F2, TCF4, TGFA, PDK1, CDKN1A, PTEN SOS2, E2F2, TGFA, CALM2, CDKN1A, PTEN CNOT6, CNOT4, DDX6, DCP2, PAN3, BTG1 ACSL4, PRKAA2, TNFRSF1B, PRKAA1, PPARGC1A

1.078737e-27 6.359019e-07

hsa05020 hsa04350

2.181797e-05 2.181797e-05 2.282662e-05

hsa04070 hsa04150 hsa04144

8.052829e-05 0.00026 0.00049 0.00094

hsa04540 hsa04115 hsa04710 hsa04360

0.0011 0.0011

hsa00562 hsa04910

0.0057 0.0068 0.0086 0.0098 0.0098

hsa04916 hsa05215 hsa05214 hsa03018 hsa04920


740


Table 7. Enrichment analysis of miR-129-2-3p putative gene targets in KEGG pathways, performed by using DIANA-microT with significant threshold set at P ⬍ 0.01 KEGG Pathway

Gene Name

FDR*

KEGG Pathway ID

Valine, leucine, and isoleucine degradation Ubiquitin-mediated proteolysis SNARE interactions in vesicular transport Fatty acid metabolism GnRH signaling pathway Fc epsilon RI signaling pathway Nicotinate and nicotinamide metabolism Ubiquinone and other terpenoid-quinone biosynthesis

HIBADH, HMGCS1, ACADM WWP1, RHOBTB1, MAP3K1, KLHL13, RCHY1 VTI1A, STX6 ACADM MAP3K1, MAP2K6, MAP2K4 PRKC, MAP2K6, MAP2K4 NMNAT2 COQ3

0.00025 0.00036 0.00090 0.0019 0.0019 0.0030 0.0047 0.0056

hsa00280 hsa04120 hsa04130 hsa00071 hsa04912 hsa04664 hsa00760 hsa00130

expressed miRNAs in theca and 116 in granulosa cells (see Supplementary Table S1).1 However, to focus our analysis and discussion we increased the stringency by using a threshold of P ⬍ 0.01. In theca cells seven miRNAs were more expressed (P ⬍ 0.01) in dominant than subordinate follicles, while seven miRNAs were more expressed (P ⬍ 0.01) in subordinate than dominant follicles (Table 1). In the granulosa cells 17 miRNAs were more expressed (P ⬍ 0.01) in dominant than subordinate follicles, whereas expression of 32 miRNAs were more expressed (P ⬍ 0.01) in subordinate than dominant follicles (Table 2). To confirm the results obtained from the miRNA microarray, the expression of four miRNAs was analyzed by qRTPCR. In theca cells miRNA hsa-miR-301b/bta-miR-301b (miR-301b) expression was greater (P ⬍ 0.05) in the dominant compared with the largest subordinate follicle, while expression levels of hsa-miR-129-2-3p/bta-miR-129-3p (miR-129-21

The online version of this article contains supplemental material.

3p) were greater (P ⬍ 0.05) in the subordinate compared with dominant follicles. Additionally, in the granulosa cells, hsamiR-18a-5p/bta-miR-18a (miR-18a-5p) expression was greater (P ⬍ 0.05) in the dominant compared with the largest subordinate follicle, whereas hsa-miR-582-5p/bta-miR-582 (miR582-5p) expression was greater (P ⬍ 0.05) in the subordinate compared with dominant follicles. Furthermore, to determine whether validation by RT-PCR correlated with miRNA microarray results, we performed a Pearson correlation analysis. The Pearson correlation coefficient obtained was 0.85, confirming positive correlation between the miRNA levels determined by microarray and RT-PCR. Microarray and qRT-PCR ratios comparing dominant and the largest subordinate follicles for theca and granulosa cells are shown in Table 3. miRNA target prediction and pathway analysis. The second version of DIANA miRPath (41) software, which is based on miRNAs obtained from miRBase 18 (24), was used to identify potentially regulated biological pathways in ovarian follicles for differentially expressed miRNAs. From the total number

Fig. 2. KEGG map visualization of hsa-miR-129-2-3p/bta-miR-129-3p involvement in gonadotropin-releasing hormone (GnRH) signaling pathway. Green, genes not targeted by this miRNA; yellow, putative targets for this miRNA; red, gene especially important in ovarian follicle physiology. Physiol Genomics • doi:10.1152/physiolgenomics.00036.2014 • www.physiolgenomics.org


(n ⫽ 63) of differentially expressed miRNAs (P ⬍ 0.01), one in theca and three in granulosa cells were excluded from the analysis as they were not available in the database (hsamiRPlus-I181b-2*, hsa-miRPlus-G1140-3p, bta-miR-362-5p, hsa-miRPlus-C1087). In theca cells, five out of the six miRNAs whose expression was greater in dominant compared with the largest subordinate follicle were involved in oocyte meiosis, Wnt signaling, TGF-␤ signaling, protein processing in endoplasmic reticulum, RNA degradation, axon guidance pathways, and pathways in cancer (Table 4). Also in theca cells, the miRNAs whose expression was greater in subordinate than dominant follicles were involved in signaling pathways such as: neurotropin, ErbB, pathways in cancer, insulin, endocytosis, chemokine, regulation of actin cytoskeleton, human T cell lymphotropic virus type 1 (HTLV-I) infection, and EpsteinBarr virus infection (Table 4). Additionally, in granulosa cells, for 15 out of 16 miRNAs with greater expression in dominant than subordinate follicles, the most significantly targeted pathways that were predicted were neurotropin signaling, gap junction, endocytosis, phosphatidylinositol-3 kinase (PI3K)-Akt signaling, MAPK signaling, focal adhesion, transcriptional misregulation in cancer, pathways in cancer, vascular smooth muscle contraction, regulation of actin cytoskeleton, and chemokine signaling pathways (Table 5). Whereas the miRNAs with greater expression in granulosa cells of subordinate than dominant follicles 20 out of 29 were involved in following signaling pathways: Wnt, MAPK, ubiquitin-mediated proteolysis, dopaminergic synapse, neurotropin, axon guidance, mRNA surveillance, pathways in cancer, PI3K-Akt, osteoclast differentiation, HTLV-I infection, endocytosis, RNA transport, transcriptional misregulation in cancer, and cell cycle (Table 5).

741

We next used DIANA miRPath v2.0 software to identify target genes and pathways for single miRNAs, which were validated by qRT-PCR. The most affected pathway (P ⫽ 1.08e-27) by miR-301b was the Prion diseases pathway in which the prion protein (PRNP) gene was the only target (Fig. 1). This miRNA was also associated with TGF-␤ signaling (11 genes, including TGFBR2 and SMAD5), mechanistic target of rapamycin signaling (8 genes, including IGF1), p-53 signaling pathway, and a few others (Table 6). miR-129-2-3p was involved in eight signaling pathways, among them only gonadotropin-releasing hormone (GnRH) signaling pathway (P ⫽ 0.0019) is known to be involved in ovarian physiology (Table 7). The targeted gene in this pathway was MAP3K1 (Fig. 2). Similarly, MAP3K1 is targeted by miR-18a-5p (Fig. 3) in GnRH signaling pathway, which is one of seven pathways affected by this miRNA (Table 8). It was the only miRNA for which the significant threshold was set at P ⬍ 0.05, because there were no affected pathways for significant threshold set at P ⬍ 0.01. miR-582-5p affected 14 pathways (Table 9), among them targets the highest number of genes (24) in P13K-Akt signaling pathway, including myeloid cell leukemia sequence 1 (MCL1) gene (Fig. 4). DISCUSSION

Four differentially expressed miRNAs identified in this study (miR-301b, miR-129-3p, miR-18a-5p, miR-582-5p) have not been previously described in regard to ovarian follicle physiology, and our results suggest their involvement in bovine ovarian follicle development. Among the most highly expressed miRNAs in theca cells was miR-301b (18.44-fold higher expression in dominant compared with subordinate

Fig. 3. KEGG map visualization of hsa-miR-18a-5p/bta-miR-18a-5p involvement in GnRH signaling pathway. Green, genes not targeted by this miRNA; yellow, putative targets for this miRNA; red, gene especially important in ovarian follicle physiology. Physiol Genomics • doi:10.1152/physiolgenomics.00036.2014 • www.physiolgenomics.org

742


Table 8. Enrichment analysis of miR-18a-5p putative gene targets in KEGG pathways, performed by using DIANA-microT with significant threshold set at P ⬍ 0.05 (predicted pathways were not found at P ⬍ 0.01) KEGG Pathway

Gene Name

FDR*

KEGG Pathway ID

Endocytosis GnRH signaling pathway Bile secretion Glycerolipid metabolism GABAergic synapse Inositol phosphate metabolism Dorso-ventral axis formation

SMAP2, IQSEC3, RAB5A, PSD3, RAB11FIP2, CDC42, PARD6B MAP3K1, CDC42, ADCY4, PRKACB ABCC2, ABCC3, ADCY4, PRKACB GK, PNLIPRP3, MBOAT2 GABRA4, ADCY4, PRKACB INPPL1, IPMK, PIK3C2A ETV6, NOTCH2

0.010 0.010 0.010 0.020 0.024 0.034 0.034

hsa04144 hsa04912 hsa04976 hsa00561 hsa04727 hsa00562 hsa04320

follicles), which has been found to affect PRNP gene in the prion disease pathway (Fig. 1). This agrees with a previous study in which PRNP mRNA expression and levels of the protein for PrPC were found to be higher in bovine theca cells of dominant compared with subordinate follicles, signifying its role in promotion of dominant follicle development (16). Another miRNA that was differently expressed in theca cells was miR-129-2-3p, which our analysis revealed targets MAP3K1 gene in GnRH signaling pathway (Fig. 2). This target is involved in two MAPK cascades important for follicular growth and development (JNK and p38MAPK). The MAPK signaling pathways are associated with regulation of steroidogenic acute regulatory protein expression and steroidogenesis (27). We have shown that miR-129-2-3p had enhanced expression in subordinate compared with dominant

follicles. This may indicate that this miRNA leads in theca cells to increase progesterone production and also to inhibition of androgens that are required for estradiol synthesis in granulosa cells. Thus, these findings suggest that miR-129-3p may be associated with follicle regression. Another miRNA affecting the MAP3K1 gene in the GnRH signaling pathway is miR-18a-5p (Fig. 3), whose expression was greater in granulosa cells of dominant compared with subordinate follicles. Reduced levels of FSH prevent further follicle wave emergence until the dominant follicle undergoes either ovulation, or regression and atresia (17, 31). Based on the obtained results, we suggest that miR-18a-5p is an important factor in mediating the effects of FSH action on granulosa cells. Another identified miRNA expressed in granulosa cells was miR-582-5p, whose seed sequence binds to the 3=-UTR of MCL1 gene and reduces

Table 9. Enrichment analysis of miR-582-5p putative gene targets in KEGG pathways, performed by using DIANA-microT with significant threshold set at P ⬍ 0.05 KEGG Pathway

Vasopressin-regulated water reabsorption Glutamatergic synapse Wnt signaling pathway Dopaminergic synapse Endocrine and other factor-regulated calcium reabsorption Axon guidance Transcriptional misregulation in cancer Gap junction MAPK signaling pathway

Melanogenesis Retrograde endocannabinoid signaling Pancreatic secretion Arrhythmogenic right ventricular cardiomyopathy PI3K-Akt signaling pathway

Gene Name

FDR*

KEGG Pathway ID

CREB1, ARHGDIB, DCTN2, RAB5A, AQP4, RAB11A, DYNC1LI1, AQP2, PRKACB GRM5, GNAI3, GNB1, CPD, PPP3CB, PPP3R1, GNAQ, GRIA4, HOMER1, GNAO1, GRIK2, CACNA1D, SLC1A2, PRKACB GSK3B, PPP2R5E, TCF4, VANGL1, PPP2R5A, PRICKLE1, NLK, FZD4, SENP2, PPP3CB, AXIN2, MAPK8, CSNK1A1, SFRP2, PPP3R1, CXXC4, WNT2, MAP3K7, TBL1XR1, PRKACB GSK3B, PPP2R5E, MAPK14, CREB1, GNAI3, GNB1, PPP2R5A, PPP3CB, MAPK8, PPP2R2A, SCN1A, GNAQ, GRIA4, GNAO1, CACNA1D, KIF5B, PRKACB, PPP1CB KL, ATP1B2, ATP1B1, ATP2B1, RAB11A, GNAQ, DNM3, PRKACB EFNB2, PLXNA2, GSK3B, PAK7, ROBO2, GNAI3, NCK1, PTK2, RASA1, EFNB3, PPP3CB, DPYSL5, RHOD, NRP1, PPP3R1, SEMA3E CCNT2, TMPRSS2, HPGD, MLLT3, RUNX2, PTK2, PBX3, ETV1, KDM6A, WHSC1, EWSR1, SIX4, FOXO1 GRM5, GNAI3, LPAR1, GNAQ, GJA1, PRKG1, PRKACB RASA2, RAP1A, MAPK14, MAP3K1, MAP3K13, TAOK1, NLK, RASA1, PPP3CB, RASGRP1, MAPK8, FGF9, FGF2, PPP3R1, NF1, STMN1, CACNB2, CACNA1D, MAP3K7, FGF7, MAP3K5, PRKACB GSK3B, TCF4, CREB1, GNAI3, MITF, FZD4, GNAQ, GNAO1, WNT2, PRKACB GABRA1, GRM5, MAPK14, GNAI3, GNB1, MAPK8, GNAQ, GRIA4, GNAO1, CACNA1D, PRKACB ATP1B2, RAP1A, SLC4A4, ATP1B1, ATP2B1, PLA2G12A, RAB11A, SLC12A2, GNAQ, ATP2A2 ITGB8, TCF4, DMD, ITGA2, GJA1, ATP2A2, CACNB2, ITGA6, CACNA1D PRLR, GSK3B, RBL2, PPP2R5E, ITGB8, PIK3CB, MCL1, CREB1, GNB1, CDK6, PPP2R5A, PTK2, EIF4E, LPAR1, PPP2R2A, COL5A1, ITGA2, FGF9, PRKAA1, FGF2, ITGA6, SGK3, FGF7, COL5A2

7.828949e-10

hsa04962

1.545603e-08

hsa04724

7.05523e-08

hsa04310

7.05523e-08

hsa04728

1.534934e-06

hsa04961

1.697383e-06

hsa04360

0.00034

hsa05202

0.00093 0.00098

hsa04540 hsa04010

0.00098

hsa04916

0.0013

hsa04723

0.0013

hsa04972

0.0013

hsa05412

0.0057

hsa04151



743

Fig. 4. KEGG map visualization of hsa-miR-582-5p/bta-miR-582 involvement in phosphatidylinositol-3 kinase (PI3K)-Akt signaling pathway. Green, genes not targeted by this miRNA; yellow, putative targets for this miRNA; red, gene especially important in ovarian follicle physiology.

its expression (25). Our pathway and gene target analysis confirmed that this miRNA regulates MCL1 gene in PI3K-Akt signaling pathway (Fig. 4). Furthermore, miR-582-5p had higher expression in subordinate compared with dominant follicles. These results are consistent with other studies in which an association between growth of the bovine ovarian dominant follicle and enhanced expression of MCL1 in granulosa cells was reported (13). We speculate that miR-582-5p plays a key role in follicle development, by decreasing the expression of MCL1 in granulosa cells of subordinate follicles undergoing apoptosis. Several miRNAs have previously been shown to play an important role in ovarian follicle development (9), and some of them have been identified in the present study. Expression of four miRNAs (miR-125b, miR-145, miR-21, miR-34a) have been reported in cultured bovine theca and granulosa cells, suggesting their involvement in the follicular-luteal transition (28). Our research confirmed the expression of these miRNAs in bovine theca and granulosa cells, but there were no significant differences between dominant and subordinate follicles with an exception of miR-145. The level of miR-145 was significantly higher (P ⬍ 0.04) in granulosa cells of dominant compared with subordinate follicles, but with no significant differences in theca cells. It is likely, that miR-145 is involved

in both follicular-luteal transition and subordinate follicle regression, while the other miRNAs are only involved in follicular-luteal transition. The results of our study concerning miR-503 are coherent with other authors who found this miRNA mostly in theca and luteal cells of ovine preovulatory follicles (28). We have shown that miR-503 had significantly higher expression (P ⬍ 0.05) in theca cells of subordinate compared with dominant follicles, suggesting its function in programmed cell death of subordinate follicles in cattle. Another miRNA, miR-383, has been reported as a positive regulator of estradiol production in mural granulosa cells (45). Our miRNA array showed its significantly higher expression in dominant compared with subordinate follicles in bovine granulosa cells (P ⬍ 0.05). It is well known that increased estradiol production is considered as a key characteristic of dominant follicle. Thus, these findings suggest that miR-383 might be an important factor controlling growth and proliferation of the dominant follicle. Using DIANA mirPath v. 2.0 software we predicted key molecular pathways important in ovarian follicle development, that are potentially targeted by the miRNAs described in this paper. These pathways included: oocyte meiosis, Wnt, TGF-␤, ErbB, and insulin signaling pathways in theca cells and PI3KAkt, MAPK, and Wnt signaling pathways in granulosa cells.


744


These are all intracellular communication networks that involve different signaling pathways, which can interact with pro- and antiapoptotic factors that appear to determine the fate of ovarian follicles. It has been reported that signals from theca cells augment the meiosis-arresting activity of granulosa cells in bovine oocytes (23). Furthermore, potential role of Wnt signaling pathway in bovine ovarian steroidogenesis and follicular growth has been indicated (7). TGF-␤ signaling pathway has a significant function in the regulation of ovarian follicle development at different stages (22). ErbB signaling pathway is likely to be the main coordinator of LH mechanisms (21). It has been suggested that EGF may play a significant role in bovine theca cells steroidogenesis (39). Concerning the insulin signaling pathway, insulin receptor mRNA has been found in theca and granulosa cells of human developing antral follicles (11). Also, in cultured theca cells, insulin has been shown to stimulate androgen production (29). The PI3K-Akt signaling pathway is a critical regulator of follicle growth, differentiation, and survival (8, 10). It has been demonstrated that regulation of the PI3K-Akt pathway activity is correlated with bovine dominant follicle selection and development (35). MAPK signaling pathway regulates cell proliferation, differentiation, and apoptosis. Its signaling protein levels were greater in bovine dominant compared with subordinate follicles (35). Moreover, in granulosa cells, the loss of trophic hormonal support is translated into a decrease of MAPK signaling pathway, and this result in the decreased phosphorylation of the proapoptotic BCL2-associated agonist of cell death (33). In conclusion, our study identified miRNAs that are likely to be regulators of bovine ovarian follicles development through global regulation of multiple targets and signaling pathways. Altogether, 14 miRNAs in theca and 49 miRNAs in granulosa cells were differentially expressed between dominant and subordinate follicles. The predicted targets for these miRNAs were enriched for pathways involving oocyte meiosis, Wnt, TGF-␤, ErbB, Insulin, PI3K-Akt, and MAPK signaling pathways. GRANTS This work was supported by National Science Centre Poland (N N311 324136). The publication was financed by the Faculty of Biology and Animal Science, Leading National Research Center (KNOW) from the Wrocław University of Environmental and Life Sciences. DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS Author contributions: A.E.Z.-S. and A.C.O.E. conception and design of research; A.E.Z.-S., J.A.B., and M.D. performed experiments; A.E.Z.-S. and P.A.M. analyzed data; A.E.Z.-S., M.G., and T.S. interpreted results of experiments; A.E.Z.-S. prepared figures; A.E.Z.-S. drafted manuscript; A.E.Z.-S., J.A.B., and A.C.O.E. edited and revised manuscript; A.E.Z.-S., J.A.B., P.A.M., M.G., M.D., T.S., and A.C.O.E. approved final version of manuscript. REFERENCES 1. Ambros V, Chen X. The regulation of genes and genomes by small RNAs. Development 134: 1635–1641, 2007. 2. Baley J, Li J. MicroRNAs and ovarian function. J Ovarian Res 5: 1757–2215, 2012. 3. Bartel DP. MicroRNAs: target recognition and regulatory functions. Cell 136: 215–233, 2009. 4. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J Roy Stat Soc B 57: 289 –300, 1995.

5. Bushati N, Cohen SM. microRNA functions. Annu Rev Cell Dev Bio 23: 175–205, 2007. 6. Canty MJ, Boland MP, Evans AC, Crowe MA. Alterations in follicular IGFBP mRNA expression and follicular fluid IGFBP concentrations during the first follicle wave in beef heifers. Anim Reprod Sci 93: 199 –217, 2006. 7. Castañon BI, Stapp AD, Gifford CA, Spicer LJ, Hallford DM, Hernandez Gifford JA. Follicle-stimulating hormone regulation of estradiol production: possible involvement of WNT2 and ␤-catenin in bovine granulosa cells. J Anim Sci 90: 3789 –3797, 2012. 8. Datta SR, Brunet A, Greenberg ME. Cellular survival: a play in three Akts. Genes Dev 15: 2905–2927, 1999. 9. Donadeu FX, Schauer SN, Sontakke SD. Involvement of miRNAs in ovarian follicular and luteal development. J Endocrinol 215: 323–334, 2012. 10. Dupont J, Reverchon M, Cloix L, Froment P, Ramé C. Involvement of adipokines, AMPK, PI3K and the PPAR signaling pathways in ovarian follicle development and cancer. Int J Dev Biol 56: 959 –967, 2012. 11. el-Roeiy A, Chen X, Roberts VJ, LeRoith D, Roberts CT Jr, Yen SS. Expression of insulin-like growth factor-I (IGF-1) and IGF-II and the IGF-I, IGF-II, and insulin receptor genes and localization of the gene products in the human ovary. J Clin Endocrinol Metab 77: 1411–1418, 1993. 12. Erdmann K, Kaulke K, Thomae C, Huebner D, Sergon M, Froehner M, Wirth MP, Fuessel S. Elevated expression of prostate cancer-associated genes is linked to down-regulation of microRNAs. BMC Cancer 2014, doi:10.1186/1471-2407-14-82. 13. Evans AC, Ireland JL, Winn ME, Lonergan P, Smith GW, Coussens PM, Ireland JJ. Identification of genes involved in apoptosis and dominant follicle development during follicular waves in cattle. Biol Reprod 70: 1475–1484, 2004. 14. Fatima A, Waters S, O’Boyle P, Seoighe C, Morris DG. Alterations in hepatic miRNA expression during negative energy balance in postpartum dairy cattle. BMC Genomics 2014, doi:10.1186/1471-2164-15-28. 15. Fayad T, Lévesque V, Sirois J, Silversides DW, Lussier JG. Gene expression profiling of differentially expressed genes in granulosa cells of bovine dominant follicles using suppression subtractive hybridization. Biol Reprod 70: 523–533, 2004. 16. Forde N, Rogers M, Canty MJ, Lonergan P, Smith GW, Coussens PM, Ireland JJ, Evans ACO. Association of the prion protein and its expression with ovarian follicle development in cattle. Mol Reprod Dev 75: 243–249, 2008. 17. Ginther OJ, Bergfelt DR, Kulick LJ, Kot K. Pulsatility of systemic FSH and LH concentrations during follicular-wave development in cattle. Theriogenology 50: 507–519, 1998. 18. Ginther OJ, Wiltbank MC, Fricke PM, Gibbons JR, Kot K. Selection of the dominant follicle in cattle. Biol Reprod 55: 1187–1194, 1996. 19. Hossain MM, Ghanem N, Hoelker M, Rings F, Phatsara C, Tholen E, Schellander K, Tesfaye D. Identification and characterization of miRNAs expressed in the bovine ovary. BMC Genomics 10: 443, 2009. 20. Ireland JL, Good TE, Knight PG, Ireland JJ. Alterations in amounts of different forms of inhibin during follicular atresia. Biol Reprod 50: 1265–1276, 1994. 21. Jamnongjit M, Gill A, Hammes SR. Epidermal growth factor receptor signaling is required for normal ovarian steroidogenesis and oocyte maturation. Proc Natl Acad Sci USA 102: 16257–16262, 2005. 21a.Kanehisa M, Goto S, Sato Y, Furumichi M, Tanabe M. KEGG for integration and interpretation of large-scale molecular data sets. Nucleic Acids Res 40: D109 –D114, 2012. 22. Knight PG, Glister C. TGF-␤ superfamily members and ovarian follicle development. Reproduction 132: 191–206, 2006. 23. Kotsuji F, Kubo M, Tominaga T. Effect of interactions between granulosa and thecal cells on meiotic arrest in bovine oocytes. J Reprod Fertil 100: 151–156, 1994. 24. Kozomara A, Griffiths-Jones S. miRBase: integrating microRNA annotation and deep-sequencing data. Nucleic Acids Res 39: D152–D157, 2011. 25. Lam LT, Lu X, Zhang H, Lesniewski R, Rosenberg S, Semizarov D. A microRNA screen to identify modulators of sensitivity to BCL2 inhibitor ABT-263 (Navitoclax). Mol Cancer Ther 9: 2943–2950, 2010. 26. Lin F, Li R, Pan ZX, Zhou B, Yu de B, Wang XG, Ma XS, Han J, Shen M, Liu HL. miR-26b promotes granulosa cell apoptosis by targeting ATM during follicular atresia in porcine ovary. PLoS One 7: e38640, 2012.


ROLE OF miRNAs IN OVARIAN FOLLICLE DEVELOPMENT 27. Manna PR, Stocco DM. The role of specific mitogen-activated protein kinase signaling cascade in the regulation of steroidogenesis. J Signal Transduct 2011, doi:10.1155/2011/821615. 28. McBride D, Carre W, Sontakke S, Hogg CO, Law AS, Donadeu FX, Clinton M. Identification of miRNAs associated with the follicular-luteal transition in the ruminant ovary. Reproduction 144: 221–233, 2012. 29. McGee EA, Sawetawan C, Bird I, Rainey WE, Carr BR. The effect of insulin and insulin-like growth factors on the expression of steroidogenic enzymes in a human ovarian thecal-like tumor cell model. Fertil Steril 65: 87–93, 1996. 30. Mihm M, Baker PJ, Fleming LM, Monteiro AM, O’Shaughnessy PJ. Differentiation of the bovine dominant follicle from the cohort upregulates mRNA expression for new tissue development genes. Reproduction 135: 253–265, 2008. 31. Mihm M, Bleach EC. Endocrine regulation of ovarian antral follicle development in cattle. Anim Reprod Sci 78: 217–237, 2003. 32. Miles JR, McDaneld TG, Wiedmann RT, Cushman RA, Echternkamp SE, Vallet JL, Smith TP. MicroRNA expression profile in bovine cumulus-oocyte complexes: possible role of let-7 and miR-106a in the development of bovine oocytes. Anim Reprod Sci 130: 16 –26, 2012. 33. Peter AT, Dhanasekaran N. Apoptosis of granulosa cells: a review on the role of MAPK-signaling modules. Reprod Domest Anim 38: 209 –213, 2003. 34. Rivera GM, Fortune JE. Proteolysis of insulin-like growth factor binding proteins -4 and -5 in bovine follicular fluid: implications for ovarian follicular selection and dominance. Endocrinology 144: 2977–2987, 2003. 35. Ryan KE, Casey SM, Canty MJ, Crowe MA, Martin F, Evans AC. Akt and Erk signal transduction pathways are early markers of differentiation in dominant and subordinate ovarian follicles in cattle. Reproduction 133: 617–626, 2007. 36. Schauer SN, Sontakke SD, Watson ED, Esteves CL, Donadeu FX. Involvement of miRNAs in equine follicle development. Reproduction 146: 273–282, 2013.

745

37. Smyth GK. Limma: linear models for microarray data. In: Bioinformatics and Computational Biology Solutions using R and Bioconductor, edited by Gentleman R, Carey V, Dudoit S, Irizarry R, Huber W. New York: Springer, 2005. 38. Smyth GK. Linear models and empirical Bayes for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol 3: Article 3, 2004. 39. Spicer LJ, Stewart RE. Interactions among basic fibroblast growth factor, epidermal growth factor, insulin, and insulin-like growth factor-I (IGF-I) on cell numbers and steroidogenesis of bovine thecal cells: role of IGF-I receptors. Biol Reprod 54: 255–263, 1996. 40. Sun J, Wang S, Li C, Ren Y, Wang J. Novel expression profiles of microRNAs suggest that specific miRNAs regulate gene expression for the sexual maturation of female Schistosoma japonicum after pairing. Parasit Vectors 2014, doi:10.1186/1756-3305-7-177. 41. Vlachos IS, Kostoulas N, Vergoulis T, Georgakilas G, Reczko M, Maragkakis M, Paraskevopoulou MD, Prionidis K, Dalamagas T, Hatzigeorgiou AG. DIANA miRPath v.2.0: investigating the combinatorial effect of microRNAs in pathways. Nucleic Acids Res 40: W498 – W504, 2012. 42. Xu S, Linher-Melville K, Yang BB, Wu D, Li J. Micro-RNA378 (miR-378) regulates ovarian estradiol production by targeting aromatase. Endocrinology 152: 3941–3951, 2011. 43. Yan G, Zhang L, Fang T, Zhang Q, Wu S, Jiang Y, Sun H, Hu Y. MicroRNA-145 suppresses mouse granulosa cell proliferation by targeting activin receptor IB. FEBS Lett 586: 3263–3270, 2012. 44. Yao G, Yin M, Lian J, Tian H, Liu L, Li X, Sun F. MicroRNA-224 is involved in transforming growth factor-beta-mediated mouse granulosa cell proliferation and granulosa cell function by targeting Smad4. Mol Endocrinol 24: 540 –551, 2010. 45. Yin M, Lü M, Yao G, Tian H, Lian J, Liu L, Liang M, Wang Y, Sun F. Transactivation of microRNA-383 by steroidogenic factor-1 promotes estradiol release from mouse ovarian granulosa cells by targeting RBMS1. Mol Endocrinol 26: 1129 –1143, 2012.
