So far, studies have shown that circRNA and miRNA have regulatory potency, especially in sheep's growth and reproduction [18–23]. However, little research has been done on circRNA and miRNA in sheep oviduct. Here, we performed RNA-Seq to analyze the circRNA and miRNA of sheep oviduct tissue between follicular phase and luteal phase in MM sheep and WW sheep, and host genes of DE circRNAs and predicted target genes of miRNAs associated with fecundity were identified. We also analyzed the distribution of circRNAs in the genome regions and length distribution of circRNAs. Most of the circRNAs genome composition in sheep uterus was intron [28], while the majority of oviduct was composed of exons. The circRNAs with a length of less than 20,000 nt account for the majority, which was consistent with circRNAs of sheep mammary gland. The length distribution of miRNAs in pituitary and ovary of sheep was similar to our results [29, 30], with 22nt miRNAs accounting for the majority. Thus, circRNAs may be tissue-specific and miRNAs are conservative in different tissues.
We identified 15 DE circRNAs in MF VS. ML, but no DE circRNAs were found in WF VS.WL. This result indicated a difference in the expression profiles of MM and WW sheep. Among the DE circRNAs, the top two with the highest expression levels were novel_circ_0009938 and novel_circ_0007904, whose host gene were PAWR and SMC6. Studies have found pro-apoptotic WT1 regulator (PAWR) conducted cell apoptosis, which inhibited the growth of prostate cancer cells [31]. Also, PAWR regulates apoptosis in rat follicles in the ovary but suppressed by FSH by activating PKCζ-dependent anti-apoptotic pathway [32]. But, the expression of PAWR was observed up-regulated in granulosa cells, indicating the increased susceptibility of GCs to undergo apoptosis [33]. In this study, the expression level of novel_circ_0009938 was low at follicular phase and increased at luteal phase, which may be explained by these studies, implying the ovary and oviduct coordinate with each other and stay in sync during the estrous cycle [34, 35]. Studies have shown structural maintenance of chromosomes 6 (SMC6) is essential for DNA repair and maintenance of genomic integrity [36]. And, SMC6 plays a key role in spermatogenesis and oocyte meiosis [37, 38], which indicated that SMC6 may maintain the genomic integrity of sperm and embryo to ensure fertility.
GO analysis showed that most of the host genes of circRNAs were related to response to stimulus, positive/negative regulation, binding and protein activity. The top enriched GO terms in MF VS.ML (at BP level), whose host gene were both XPR1, were response to external stimulus, MAPK cascade, regulation of response to food. Xenotropic and polytropic retrovirus receptor 1 (XPR1) is a gene encoding cellular inorganic phosphate export protein, and its mutation can cause primary familial brain calcification. The normal development of the fetus is inseparable from phosphorus. This nutrient is mainly transported from the maternal blood to the fetus via the placenta. Xu et al. [39] found that XPR1 was expressed at a high level in the murine placenta, but the placenta of the murine that knocks out this gene was severely calcified. Another top enriched host gene in GO analysis (at MF level) is SLC7A11. Soluble carrier family 7 member 11 (SLC7A11) gene is a target of p53-mediated transcriptional repression, and p53 can inhibit the uptake of cystine by repressing the expression of SLC7A11. Studies in mutant mice have found that p53 plays an important role in embryonic development [40]. In addition, SLC7A11 exists in the sperm of stallions and regulates the oxidation–reduction status of sperm by exchanging extracellular cystine (Cyss) for intracellular glutamate [41]. Moreover, host gene lysophosphatidic acid receptor 3 (LPAR3) is involved in Rap1 signaling pathway, PI3K-Akt signaling pathway and neuroactive ligand-preceptor interaction. Rap1 combined with GTP activates the PI3K-Ark signaling pathway, and the PI3K-Ark signaling pathway is widely involved in various important processes of mammalian ovarian development [42], and is related to the survival and activation of primitive follicles [43], hormone secretion, and so on. In addition, neuroactive ligand-preceptor interaction is related to the effect of GnRH and GnRHR. LPAR3 was found to be expressed in mouse oviduct, placenta, and uterus, and its essential role in the female reproductive system was reported [44]. Studies have found that progesterone is likely to have a direct effect on LPAR3, and progesterone treatment can increase the expression of LPAR3 mRNA in endometrium [45]. Besides, dynamic changes that occur in the organization of luminal and glandular epithelia in endometrium during the estrous cycle are necessary to modulate the appropriate environment for the developing embryo and to allow implantation of the conceptus [46]. The differentiation of oviductal epithelial cells is also affected by progesterone. Given the key role of progesterone, we suppose that LPAR3 may play crucial roles in sheep reproduction. Therefore, three novel circRNAs—novel_circ_0012086, novel_circ_0001794 and novel_circ_0014274, whose host genes are XPR1, LPAR3 and SLC7A11, respectively, may have key roles in reproduction. However, it remains to be validated.
GO analysis of predicted target gene of miRNAs showed that most terms were related to lipid, enzymatic activity, junction, and binging in MF VS.ML, including lipid transport, protein binding, lipid binding, serine hydrolase activity, adherenss junction and so on. In addition, KEGG enrichment analysis implied that the target genes were mainly involved in the Insulin secretion, cAMP signaling pathway, cGMP-PKG signaling pathway, Rap1 signaling pathway and Calcium signaling pathway. And the DE miRNAs may affect sheep reproduction by modulating target genes associated with the above signaling pathways and biological processes. In MM sheep, these pathways were enriched for miR-370-3p and its target genes (RAB3A, PLCB3, CREB3L4, LPAR1, LPAR2, FGF18, TACR3). RAB3A, with evolutionarily conserved proteins, is essential regulator of membrane trafficking [47]. Researches have found that RAB3A is indispensable in human sperm acrosome reactions and may regulate the quality of oocytes [48, 49]. Phospholipase C beta 3 (PLCB3) was considered to be a candidate gene for litter size in Finnsheep, and playing key roles in the development of folliculogenesis and LH signaling [50]. PLCB3 was also differentially expressed in endometrial of heifers with high and low fertility [51]. CAMP responsive element binding protein (CREB) was found to be expressed in follicular granulosa cells [52]. CREB protein concentration increases during sexual maturation and ovarian follicular development. CREB3L4 was detected in different stages of embryogenesis [53]. Same as LPAR3, LPAR1 and LPAR2 play crucial roles in female reproductive system. Studies have found that LPA medium can improve the survival and development potential of follicles, and can stimulate the cell function and E2 synthesis of mouse ovarian tissue [54]. In addition, oviduct is an important part where gamete transport and fertilization happened. LPA was found to be involved in gamete transport, fertilization, and cell signal transmission between oviduct tissue and cumulus oocyte complex [55, 56]. And LPAR2 was found to be abundantly expressed in the oviduct of cattle, suggesting that the oviduct is an important target of LPA [57]. Fibroblast growth factor 18 (FGF18) inhibits the secretion of estradiol and progesterone, and is a candidate factor that regulated the steroidogenesis during ovarian development [58]. Moreover, FGF18 is likely to cause granulosa cell apoptosis, thereby affecting follicular atresia [59, 60]. Tachykinin receptor 3 (TACR3) plays a key role in regulating gonadotropin secretion and sex hormone feedback regulation of the reproductive axis [61]. Tacr3 may also be related to the regulation of granulosa cell function and changes in ovarian function [62]. Besides, the expression of TACR3/TAC3 can promote the secretion of GnRH [63], which may affect sheep reproduction. These genes are likely to participate in the reproductive process of MM sheep, but specific molecular mechanism needs further verification.
GO analysis of predicted target gene of miRNAs in WF VS. WL showed that most terms were related to lipid, enzymatic activity, junction, and binging, which is similar to MM sheep. KEGG enrichment analysis indicated that the target genes were mainly involved in the TGF-β signaling pathway, Insulin secretion, Protein processing in endoplasmic reticulum, cGMP-PKG signaling pathway, Rap1 signaling pathway. In WW sheep, in addition to target genes above (RAB3A, PLCB3, CREB3L4, LPAR1, LPAR2, FGF18, TACR3), miR-370-3p and its target genes (BMP6, SMAD4, SKP1) were enriched in TGF-β signaling pathway as well. Bone morphogenetic proteins 6 (BMP6) is a member of the transforming growth factor-β (TGF-β) superfamily and was found to be highly expressed in mammalian oocytes and granulosa cells [64, 65]. Studies have found that BMP6 is involved in primary/secondary follicle transition, dominant follicle selection, ovarian steroid production, follicular atresia, prevention of luteinization, and luteolysis [66–68]. Besides, mice genetically missing BMP6 was characterized by reduced ovulation rate, impaired oocyte quality and impaired embryo implantation, resulting in reduced litter size [69]. Seven TGF-1 receptor subtypes and five type 2 receptor subtypes associated with signal transduction ligand of the TGF-β superfamily have been found in mammals [70]. BMP ligand activates cellular activity by binding to two types of receptors and phosphorylates the responsive SMAD proteins SMAD1/5/8 (R-SMADs) [71]. Then, R-SMADs bind to SMAD, SMAD4, translocating into the nucleus, to regulate the transcription of target genes by combining with other transcription factors [72]. Mothers against decapentaplegic homolog 4 (SMAD4) is a key signal transduction molecule in the TGFβ/SMAD signaling pathway, which plays an important role in the development of mammalian follicles and the proliferation and differentiation of granulosa cells. [73]. Studies have found that specifically knocking out the SMAD4 gene in ovaries led to premature failure of mouse follicles, premature luteinization of granulosa cells, and decreased fertility [74]. In addition, mice died in the embryonic stage after knocking out SMAD4 [75]. S-phase kinase association protein 1 (SKP1), as a downstream regulator of the TGF-β/SMAD signaling pathway, regulates follicle formation and ovulation in mammals [76]. SKP1 is a key skeleton protein in Skp1-Cull-F-box protein (SCF), which mediates the ubiquitination and degradation of different cyclins [77], thereby promoting the cell cycle [78]. SCF has also been found to be crucial for oocyte division and maturation [79], as well as the process of fertilization and implantation [80]. Here, the results imply that these target genes are probably related to sheep reproduction process, but the molecular mechanism through which they affect litter size is still unclear. Further experiments are needed to verify these target genes.