Sperm-borne miR-202 targets SEPT7 and regulates first cleavage of bovine embryos via cytoskeletal remodeling

With the birth of the cloned sheep Dolly, somatic cell nuclear transfer (SCNT) has made great progress to some extent. However, compared with in vitro fertilization (IVF) for generating embryos, there has been no significant improvement in the in vitro development of SCNT embryos and their health after birth in livestock production. As the male gamete, sperm is responsible not only for the transmission of the paternal genome, but also for factors necessary for fertilization and preimplantation embryo development (Trigg et al., 2019). As mature sperm has only about 0.5 fg of long-chain RNA and 0.3 fg of small non-coding RNA (sncRNAs) (Goodrich et al., 2013), the RNA was long considered to be only a residue of sperm maturation, with no effect on fertilization or embryo development. With the deepening of research on RNAs in recent years, especially sncRNAs (including microRNAs, endo-siRNAs, piwi-interacting RNAs and a large number of small RNAs formed by tRNA) (Krawetz et al., 2011; Song et al., 2011; Peng et al., 2012), the role of these RNAs carried by sperm is no longer being ignored in mammalian fertilization and early embryo development, or even in the development of newborns (Jodar et al., 2013; Yuan et al., 2016; Guo et al., 2017; Conine et al., 2018; Trigg et al., 2019). Our previous research found that microinjection of sperm-borne small RNAs can change the abnormal pronuclear-like structures of SCNT embryos, delay the time of first cleavage, and improve the developmental potential of embryos by regulating the expression of H3K9me3 and acetylated α-tubulin (Qu et al., 2019); however, the underlying mechanism is unclear.

MicroRNAs (miRNAs) are a class of non-coding small RNA molecules with a length of 18-25 nucleotides that are highly conserved in evolution; they can reduce the expression of mammalian target genes by inhibiting translation (Schwarz and Zamore, 2002; Rana, 2007). As a miRNA that is expressed only in mouse sperm but not in oocytes, miR-34c is crucial for the first cleavage of mouse embryos (Liu et al., 2012). In related studies on the early embryonic development of bovine SCNT embryos, researchers increased the expression of miR-34c or miR-449b in bovine fetal fibroblasts through the Tet-On system and then used the fibroblast as donor cells to prepare SCNT embryos (Wang et al., 2014a, 2017). The developmental potential of SCNT embryos was improved and the epigenetic state of the embryos was changed, suggesting that sperm-borne miRNAs play a key role in regulating early embryonic development.

In light of the crucial roles of miRNAs in embryonic development, we can no longer ignore the association between miRNAs and embryo morphogenesis. The cytoskeleton is the predominant component of embryonic morphogenesis. It is a complex network of microtubules, microfilaments and intermediate filaments, which drives the alteration of cell shape and embryogenesis (Goldstein and Theriot, 2001). Post-translational modifications (PTMs) of cytoskeletal proteins confer different biological functions to ensure that their activity and dynamics are strictly regulated in time and space.

Acetylation is a reversible PTM that can crosslink with other PTMs such as phosphorylation, ubiquitylation and methylation, and plays an important role in the modification of tubulin (Yang and Seto, 2008). For example, acetylation of α-tubulin at lysine 40 (K40) has received special attention because it is the only PTM that occurs in microtubule lumens (Janke and Montagnac, 2017) and it involves cell adhesion and migration (Aguilar et al., 2014; Boggs et al., 2015), dendritic and axonal growth (Ferreira and Cáceres, 1989; Tapia et al., 2010; Jenkins et al., 2017), and intracellular transport (Nakakura et al., 2016). In mouse oocytes, the acetylation pattern of tubulin varies at different stages of cell division. The degree of acetylation is extremely high at the poles of the metaphase meiotic spindle, is distributed throughout the spindle at the anaphase and is limited to the mid-body at telophase (Schatten et al., 1988). During spermatogenesis, the cytoskeleton can directly promote mitosis and meiosis, and the acetylation of axial microtubules can affect sperm maturation, movement and morphology (Kalebic et al., 2013; Bhagwat et al., 2014). Qu et al. have reported that the microinjection of sperm-borne small RNAs changed the abnormal pronuclear-like structures of SCNT embryos, delayed the first cleavage and improved the developmental potential of embryos by regulating the expression of H3K9me3 and acetylated α-tubulin (Qu et al., 2019). This evidence suggests that sperm-specific small RNAs may affect the development of early embryos through PTM of cytoskeletal proteins.

We selected miR-202 from the sperm-enriched miRNA library established previously and investigated the association between miR-202 and embryonic development and morphogenesis. We determined the effects of miR-202 on embryonic development and discovered that miR-202 regulated the acetylation of α-tubulin in zygotes via its target gene SEPT7, thus affecting the first cleavage process and developmental potential of embryos. This is an important finding that suggests a need for us to further explore the roles of the paternal regulatory factors on embryonic development and the regulation of cytoplasmic cytoskeletal modifications in early embryos.

In our previous miRNA-sequencing study, bovine miRNA expression libraries of sperm, MII oocytes and fetal fibroblasts were constructed (Fig. 1A,B, Table S2). As shown in Table 1, miR-202 was identified as an important miRNA, specifically expressed in sperm. Quantitative RT-PCR was performed to verify the sequencing results, and the dynamic expression pattern of miR-202 was investigated in early bovine embryos. Results showed that MII oocytes and fetal fibroblasts contained a hardly detectable amount of miR-202, whereas it could easily be detected in sperm and zygotes (Fig. 1C,D). These results indicated that miR-202 was highly enriched in sperm, and could be transferred into oocytes through fertilization. In IVF embryos, the expression level of miR-202 was highest at the pronuclear stage, decreased until the four-cell stage, and then reached another expression peak at the eight-cell stage (Fig. 1E).

According to predictions of target genes using the TargetScan and miRanda software, we selected 35 overlapping genes (Table S1) as candidates. The accuracy of the gene information of candidate genes was queried on NCBI to ensure that the seed sequences bound to miR-202, and we ultimately identified 13 candidate target genes. We performed dual-luciferase reporter assays on the candidate genes, and only SEPT7 (Fig. 2B) and LIN7C (Fig. S4) reduced the fluorescence intensity, and SEPT7 had a greater effect (other data not shown). In addition, functional analysis of 13 candidate genes (Table S3) was performed using the DAVID (Dennis et al., 2003) database. Our previous studies found that SCNT embryos had abnormal and earlier first cleavage times compared with IVF embryos (Qu et al., 2019). Considering that SEPT7 is involved in biological processes such as cytokinesis and regulation of embryonic cell shape, and participates in the formation of the axoneme, cleavage furrow and other structures, we chose SEPT7 for further research.

After preliminary verification of the targeting relationship between miR-202 and SEPT7 (Fig. 2A,B), we performed a luciferase reporter assay in the presence of a miR-202 inhibitor to exclude the effects of endogenous miR-202 on luciferase activity. We observed much higher luciferase activity with the SEPT7 3′UTR construct in the presence of miR-202 inhibitors compared with control (Fig. 2C). To prove the regulatory relationship between miR-202 and SEPT7 more directly, the relative level of SEPT7 mRNA in 293T cells was measured by qRT-PCR 24 h after miR-202 mimic transfection, and the results showed that miR-202 mimic significantly inhibited SEPT7 expression (Fig. 2D). miR-202 inhibitor was injected into bovine zygotes, and the protein level of SEPT7 in IVF embryos increased significantly (Fig. 2E).

In the zygotes injected with miR-202 inhibitor, the average time of the first cleavage of the embryos was earlier than that of the control group (Fig. 3A), suggesting that miR-202 was involved in the regulation of the first cleavage time of IVF embryos. To investigate whether there is a related role in SCNT embryos, we counted the average cleavage time of SCNT embryos injected with control mimic or miR-202 mimic. The results indicated that the miR-202 mimic delayed the average time of the first cleavage of SCNT embryos (Fig. 3B).

To determine whether the effect of miR-202 on the first cleavage time of bovine embryos was related to the target gene SEPT7, we reduced the expression of miR-202 by microinjecting SEPT7 siRNA into IVF embryos. After confirming knockdown efficiency (Fig. S1), we observed a small delay in the cleavage time, which was lessened by the miR-202 inhibitor (Fig. 3A). However, the first cleavage time of SCNT embryos injected with SEPT7 siRNA was significantly later than that of the control group (Fig. 3C), similar to the results of injection with miR-202 mimic. The results suggested that miR-202 could play a role in the first cleavage of bovine embryos through regulation of SEPT7.

To investigate the molecular mechanism of miR-202-SEPT7 regulation of the first cleavage in bovine embryos, we injected IVF zygotes with miR-202 inhibitor or knocked down the expression of SEPT7 while reducing miR-202 levels. The results of qRT-PCR showed that the miR-202 inhibitor significantly reduced the expression of miR-202 in IVF embryos during the first cleavage, regardless of whether si-SEPT7 was co-injected (Fig. S2A). At the protein level, compared with the control group, miR-202 inhibitor significantly increased the level of SEPT7 in IVF embryos and downregulated acetylated α-tubulin, but the expression of HDAC6 did not change significantly (Fig. S2C). There was no significant change in expression of SEPT7, HDAC6 or acetylated α-tubulin co-injected with miR-202 inhibitor and SEPT7 siRNA.

During the first cleavage of IVF embryos, miR-202 inhibitor resulted in SEPT7 enrichment around the chromatin/chromosomes and misalignment of metaphase chromosomes (Fig. 4A-D). In addition, although SEPT7 protein was significantly increased, miR-202 inhibitor reduced the expression of acetylated α-tubulin and caused abnormal structures to form in the middle of cleavage (Fig. 4E-H). Consistent with the western blot results, the miR-202 inhibitor had no significant effect on HDAC6 expression.

In SCNT embryos, the miR-202 mimic significantly increased the expression of miR-202 (Fig. S2B), resulting in downregulation of SEPT7 but upregulation of acetylated tubulin at the protein level (Fig. 5A-H, Fig. S2D, Fig. S3). In addition, the miR-202 mimic altered the distribution of SEPT7 (Fig. 5A-D), and improved the misalignment of metaphase chromosomes (Fig. 5E-H).

As there was no direct interaction between SEPT7 and acetylated α-tubulin, we postulated that SEPT7 might recruit HDAC6, a histone deacetylase that also removes acetyl groups from α-tubulin, to the microtubule site. Here, we used PLA to determine whether SEPT7 interacted with HDAC6 during the first cleavage.

We used IVF embryos injected with a miR-202 inhibitor and replaced HDAC6 with α-tubulin-acetyl K40 antibody of the same species and at an appropriate concentration when PLA primary antibody was incubated, and used this as the negative control. The red fluorescence signal was detected in the cytoplasm from metaphase to telophase, while the negative control group did not show any fluorescence, indicating that SEPT7 and HDAC6 interacted in IVF embryos during the first cleavage (Fig. 6A-D). Compared with the scramble inhibitor, the miR-202 inhibitor significantly enhanced the fluorescence signal, suggesting that the interaction between SEPT7 and HDAC6 may be affected by miR-202 expression.

Similarly, we replaced HDAC6 with α-tubulin-acetyl K40 in a region of SCNT embryos injected with siRNA NC during PLA primary incubation and used it as the negative control. The results showed that both SEPT7 and HDAC6 were associated from metaphase to telophase during the first cleavage of SCNT embryos, and the relationship was affected by changes in expression of SEPT7 (Fig. 7A-D).

To evaluate the effect of miR-202 on development potential of SCNT and IVF embryos, the development of embryos injected with miR-202 inhibitor or mimic was monitored for 48 h and 7 days after activation or fertilization, and the quality of blastocysts in each group was assessed. In IVF embryos, the cleavage rate of embryos injected with miR-202 inhibitor was significantly lower than that of the scramble inhibitor group (58.10±4.99% versus 74.45±5.18%, P<0.05); the cleavage rate of the miR-202 inhibitor and SEPT7 siRNA co-injection group (66.25%±3.19) was lower than the control group but higher than the test group (Fig. 8A,B). There was no significant difference in blastocyst rate between the control group and the co-injection group (41.99±6.56% versus 33.23±3.61%), but both were significantly higher than the miR-202 inhibitor injection group (22.67±4.04%) (Fig. 8C,D).

The results of IVF blastocyst immunofluorescence staining with CDX-2 and DAPI showed that, compared with the control group, there was no significant change in the ratio of inner cell mass to trophoblast cells in the blastocysts from the miR-202 inhibitor injection group, but the total numbers of cells, inner cells and trophoblast cells were significantly reduced. Compared with the miR-202 inhibitor group, co-injection significantly increased the number of trophoblast cells, the total number of cells and the number of inner cell groups (Fig. 8E,F). These results suggested that knockdown of SEPT7 can rescue the abnormal development caused by miR-202 deletion to some extent.

In SCNT embryos, the cleavage rate (Fig. 9A,B) and blastocyst rate (Fig. 9C,D) in the miR-202 mimic injection group were significantly higher than those in the control mimic group (85.87±4.45% versus 75.77±2.32%; 44.24±3.95% versus 32.76±2.70%, P<0.05). Compared with the control group, the total numbers of cells, inner cells and trophoblast cells were significantly increased in the miR-202 mimic injection group (Fig. 9E,F), indicating that miR-202 could improve the developmental potential of SCNT embryos.

As a non-coding RNA involved in post-transcriptional regulation, research related to miR-202 has mainly focused on cancer. The expression of miR-202 is downregulated in various cancers, which led researchers to regard miR-202 as a tumor suppressor (Zhao et al., 2013; Wang et al., 2014b, 2015; Meng et al., 2016). miR-202-5p may also regulate the interaction between somatic and germ cells and human support cells during spermatogenesis (Dabaja et al., 2015), and play a crucial role in embryonic gonadal sex differentiation of chicken and spermatogenesis in mice (Bannister et al., 2011; Wainwright et al., 2013). miR-202 can be detected in bovine follicular fluid, and its expression level can be used as an indicator of steroid hormone production capacity (Santos et al., 2018). Here, we constructed and screened a bovine miRNA expression library, and verified that miR-202 was highly enriched in sperm. Therefore, we further explored the possible biological role of sperm-borne miR-202 in embryogenesis and development.

Previous studies have noted the importance of paternal small RNAs, but there has been no clear evidence showing how sperm miRNAs influence embryonic development and the offspring (Amanai et al., 2006; Liu et al., 2012; Dickson et al., 2018). Studies found that sperm-expressed miR-34c and miR-449b could improve the development of SCNT embryos (Wang et al., 2014a, 2017) by regulating H3K9me3 and α-tubulin acetylated on lysine 40 (Qu et al., 2019). Moreover, sperm miRNAs were reported to initiate a series of molecular events to reprogram embryonic development by targeting maternal mRNA in zygotes (Fullston et al., 2016) and altering PTM (Short et al., 2017). This evidence supports the idea that sperm miRNAs are epigenetic signals transmitted to offspring and playing an important role in embryonic development. In this study, we demonstrated the presence of sperm-borne miR-202 and its contribution to the development and quality of IVF and SCNT embryos.

Upon fertilization or artificial activation, oocytes undergo a second meiosis, and the block is removed so that the first cleavage can take place. Numerous studies have indicated that the first cleavage is crucial for subsequent development of an embryo, and several parameters during first cleavage, such as the starting time, blastomere size and fragments in the perivitelline space, have been used to predict the developmental potential of embryos (Yu et al., 2017; Zaninovic et al., 2017; Yang et al., 2018). Qu et al. found that the starting time of first cleavage of bovine SCNT embryos was earlier than their IVF counterparts, indicating that the first cleavage in SCNT embryos may be abnormal (Qu et al., 2019). Our research demonstrates the effect of sperm-borne miR-202 on the first cleavage of bovine embryos. The embryo cleavage rate, the blastocyst formation rate and blastocyst quality are significantly decreased in bovine IVF embryos lacking miR-202 and significantly increased in SCNT embryos injected with a miR-202 mimic, confirming that miR-202 is involved in regulating the cleavage of bovine embryos and ultimately affects the developmental potential of the embryo. It is necessary to understand the regulatory mechanism of sperm-borne miR-202 on the first cleavage of bovine embryos to improve the cloning efficiency.

This study aimed to assess the importance of miR-202 in embryonic development and investigate the underlying mechanism. We experimentally verified that miR-202 mediates the delay of the first embryonic cleavage by injecting miR-202 inhibitor into zygotes, and significantly reduced the cleavage rate, blastocyst formation and quality. The mechanism is unknown but it may have something to do with the expression and location of the cytoskeletal GTPase, SEPT7. This study did not delve into the subsequent development of IVF embryos derived from sperm lacking miR-202. On the one hand, the original intention of this research was to compare the formation of IVF and SCNT embryos under different conditions to improve the efficiency of the SCNT embryo cloning method. On the other hand, although miR-202 can be detected in SCNT embryos, its developmental potential is significantly lower than that in normal IVF embryos, indicating that sperm is indispensable in this process. Whether sperm-borne miR-202 plays a regulatory role in this process alone or in collaboration with other miRNAs is still under study.

Septins are guanosine triphosphate-binding proteins that are highly conserved in eukaryotic animals. Mammalian septins can interact with actin, phospholipid membranes and microtubules in the cytoskeleton. They are found in specific locations in cells where they coordinate changes in membrane and cytoskeletal organization, and act as a scaffold in the process of protein recruitment (Kinoshita and Noda, 2001; Surka et al., 2002; Nagata et al., 2003; Kremer et al., 2005; Joo et al., 2007; Spiliotis et al., 2008; Tanaka-Takiguchi et al., 2009; Bowen et al., 2011; Sellin et al., 2011). Among the members of the septin gene family, SEPT7 forms a single-member subgroup that cannot be replaced by other septins when SEPT7 is knocked out or inactivated. For example, knockout of SEPT7 can cause embryonic lethality in mice (Ageta-Ishihara et al., 2013).

Studies have suggested that SEPT7 protein provides a physical scaffold for HDAC6 to ensure efficient deacetylation of microtubules (Ageta-Ishihara et al., 2013), and that SEPT7 plays an important role in the formation of mitotic spindles and the expression of the contractile ring protein (Menon et al., 2014). The dynamic PTM changes of tubulin are very important for mitosis, and acetylation of microtubules can increase their stability and alter their function (Hubbert et al., 2002). Despite these promising results and findings, many questions still remain. The analysis of protein-protein interactions in early mammalian embryos is hampered by the small number of embryonic cells. Bedzhov and Stemmler used the proximity ligation assay (PLA) to overcome this limitation and clarify the interactions between proteins by in situ visualization of mouse preimplantation embryos (Bedzhov and Stemmler, 2015). In our study, PLA analysis confirmed that, in both SCNT and IVF embryos, there was an interaction between SEPT7 and HDAC6 during the first cleavage, which continued from metaphase to telophase.

It is interesting that changes in SEPT7 do not affect the expression of HDAC6, but do alter the expression and morphology of acetylated α-tubulin. This is consistent with previous findings (Ageta-Ishihara et al., 2013), suggesting that SEPT7 regulates the stability of α-tubulin by affecting the activity of HDAC6. In IVF embryos injected with miR-202 inhibitor, the protein level of SEPT7 was increased, which enhanced the interaction between the septin protein complex and the microtubule cytoskeleton, thus accelerating cell division. On the other hand, the increase of SEPT7 enhanced the deacetylating activity of HDAC6, reducing the stability of microtubules. The microtubule cytoskeleton may be depolymerized early in the cell division process, thus affecting the location of the contraction ring and the initiation of cytoplasmic division, preventing cells from dividing smoothly, and reducing the cleavage rate of embryos. Injection of exogenous miR-202 mimic into SCNT embryos decreased SEPT7 expression, which delayed the first cleavage but increased the development potential of the embryos. These results confirmed the association between SEPT7 and the preimplantation development of embryos.

Based on earlier observations in which the first cleavage in SCNT embryos was abnormal (Qu et al., 2019), our study reinforced the importance of sperm-borne miR-202 in the first cleavage of SCNT embryos. miR-202 may be associated with the formation of the septin protein complex and its interaction with the cytoskeleton through the microtubules, causing premature depolymerization during cell division, affecting the localization of the contractile ring and altering the initiation of cytokinesis. This research provides more evidence of the indispensable role of sperm-borne small RNAs in early embryonic development, expands our understanding of the effects of paternal regulatory factors on reproduction and helps to improve the efficiency of SCNT techniques.

The procedures were approved by the Animal Care and Use Committee of Northwest A & F University and performed in accordance with animal welfare and ethics.

Bovine ovaries were collected from the Xi'An slaughterhouse, and stored in normal saline at 18-22°C for 5-6 h until they were transferred to the laboratory. The tissue covering the follicles on the surface of the ovary was removed along with the remaining blood. Oocytes were collected and mature cultured in vitro according to previously described methods (Wang et al., 2011). Briefly, surface-visible follicles with diameters between 2 and 8 mm were aspirated. Cumulus oocyte complexes (COCs) with at least three layers of cumulus cells and uniform cytoplasm were selected, and washed three times in PBS containing 5% (v/v) FBS. The COCs were cultured in tissue culture medium-199 (TCM-199, Gibco) supplemented with 10% (v/v) FBS, 1 μg/ml 17β-estradiol and 0.075 IU/ml human menopausal gonadotropin (HMG) in a humid atmosphere of 5% CO₂ at 38.5°C for 20 h, followed by using 0.1% bovine testicular hyaluronidase to discard the cumulus cells. Oocytes with the first polar body (Pb) and uniform cytoplasm were selected for later experiments.

After discarding head and viscera of a 35-day-old fetus, the remaining tissues were washed thrice with PBS and cut into small pieces (∼1 mm³) as described previously (Wang et al., 2014a). The explants were placed in a culture dish with Dulbecco's modified Eagle's medium (DMEM, Gibco) supplemented with 10% FBS, 250 ng of amphotericin B, 10 ng/ml epidermal growth factor and 100 U of penicillin, and incubated in a humid atmosphere of 5% CO₂ at 37°C for 24 h. When the cells reached 80-90% confluence, they were treated with TE (0.25% trypsin/0.05% EDTA) and passaged at a ratio of 1:2 or 1:3. Fetal fibroblasts at passage 3 to 10 were used as donor cells. Prior to SCNT, they were starved for 3-5 days in fresh medium containing 0.5% FBS.

The methods of bovine sperm collection and purification were from a previous publication (Du et al., 2014). After removing the epididymal cortex, pre-warmed PBS was added and the cauda epididymis was cut into 1 cm³ pieces and passed through a sieve to separate the sperm. The sperm suspension was transferred to the bottom of a 50 ml centrifuge tube containing 30 ml Brackett and Oliphant (BO) medium supplemented with 6 mg/ml BSA and 20 μg/ml heparin pre-warmed 8 h in advance, at a 45° incline in a humid atmosphere of 5% CO₂ at 38.5°C for 30 min. The supernatant was pipetted into a new 15 ml centrifuge tube and centrifuged for 10 min at 300 g. The supernatant was discarded leaving ∼50 μl of liquid to resuspend the sperm for IVF.

Total RNA was extracted using Trizol reagent (Invitrogen, U.S.). Two hundred mature oocytes with uniform cytoplasm were selected and transferred to 500-600 μl Trizol. Bovine fetal fibroblasts in six-well plates were washed twice with pre-warmed PBS, digested with EDTA-trypsin followed by DMEM to stop the digestion, then added to 1 ml of Trizol. For isolation of sperm RNA, the sperm were pelleted by centrifugation, 500-600 μl of Trizol (warmed to 60°C in advance) was added and total RNA was isolated according to the standard procedure. Following the protocol of the TruSeq Small RNA Sample Preparation Kit (Illumina), ∼1 μg of total RNA was used to prepare a small RNA library. Briefly, the RNA 3′ adapter and 5′adapter were ligated in sequence to both ends of the small RNA sample by T4 RNA ligase. Reverse transcription was performed using the RNA RT primer provided in the kit, and the cDNA sequence was then amplified by PCR. The target fragment was purified on a 6% polyacrylamide gel to complete preparation of the miRNA library. For the built library, single-end sequencing (36 bp) on an Illumina Hiseq 2500 at the LC-BIO (Hangzhou, China) was performed following the recommended protocol.

Small RNA isolation, reverse-transcription and qRT-PCR of bovine matured oocytes, sperm and fetal fibroblast cells were performed with the miScript Single Cell qPCR Kit (Qiagen). After washing with neutral buffer PBS, 1 μl buffer FCPL/FCPM and 3 μl gDNA wipeout buffer were added to the centrifuge tube containing the sample, and incubated at room temperature for 5 min followed by 75°C for 5 min and held at 4°C. Next, 3′ ligation and 5′ ligation were respectively performed.

The procedure for reverse-transcription was based on the manufacturer's instructions. Briefly, 8 μl miScript SC 5×RT buffer, 4 μl miScript SC 10×RT Nucleics, 7 μl nuclease-free water and 1 μl miScript SC reverse transcriptase were mixed and added into 5′ ligation reaction products. The mix was incubated at 37°C for 2 h and 95°C for 5 min, then held at 4°C for at least 5 min. Afterwards, cDNA cleanup was performed by ‘beads+bind’ mixture, and the product was used as template for preamplification.

For qRT-PCR analysis, 27 μl of the preamplification product was diluted to 127 μl by adding 100 μl of nuclease-free water. Then, 12.5 μl 2×QuantiTect SYBR green PCR master mix, 2.5 μl 10×miScript Universal Primer, 2.5 μl nuclease-free water, 5 μl diluted preamplification product and 2.5 μl of primers were mixed. The PCR cycle conditions were as follows: 94°C for 30 s followed by 40 cycles of 94°C for 5 s and 60°C for 30 s. The following forward primers were used to detect the relative expression of miR-202: miR-202-F, 5′-GGGGGTTCCTATGCATA TACTTCT-3′; U6-F, 5′-CGCTTCGGCAGCACATATACTA-3′.

The identification of target genes for bta-miR-202 was performed as described previously (Wang et al., 2017). Briefly, the cDNA obtained from bovine fetal fibroblasts was used as a template to amplify the 3′UTR of SEPT7 using primers with NotI/XhoI sites, and primers with EcoRI/BamHI sites were used to amplify the bta-miR-202 precursor. The sequences were inserted downstream of the renilla luciferase (hRluc) gene of the psiCheck-2 vector and the multiple cloning site of the pCD513B-1 vector, respectively, and correct vector was verified by sequencing. The mutant vector (psi-mut SEPT7) was obtained by mutating the binding site of the 3′UTR of SEPT7 using gene splicing by overlap extension PCR with the psiCheck-2-SEPT7 vector (psi-SEPT7) as a template. Next, the pCD513B-1 vector was transfected together with either the psi-SEPT7 or psi-mut SEPT7, and the pCD513B-1-miR-202 with the same combination. In addition, the miR-202 inhibitor (which can competitively bind to mature endogenous miR-202, prevent the complementary pairing of miR-202 and its target mRNA, and inhibit the regulation of miRNA on downstream genes) or scrambled inhibitor was transfected together with either the psi-SEPT7 or psi-mut SEPT7 vector, respectively.

The dual luciferase reporter assay was performed after 36 h using the TransDetect double-luciferase Reporter Assay Kit (TransGen Biotech), according to the manufacturer's instructions on the preparation of luciferase reaction reagent and the dilution of cell lysates. After cell lysis, the supernatants were transferred to a 96-well Microlon ELISA plates (Corning) and luciferase reaction reagent was added. The activity of the firefly luciferase reporter gene was detected by a Victor X5 multi-label plate reader (Perkin Elmer). Luciferase reaction reagent II was then added to the assay mixture and the activity of the renilla luciferase reporter gene was measured. Based on the detection results of the two luciferase activities, the ratio of renilla luciferase to firefly luciferase was calculated to determine whether there was a targeting relationship between bta-miR-202 and SEPT7.

Cumulus cells were removed from COCs by treating with 0.1% bovine testicular hyaluronidase in PBS. Oocytes with first Pb and even cytoplasm were selected and stained with 10 μg/ml Hoechst 33342 for 10 min prior to micromanipulation. Enucleation was performed using a 20 μm glass pipette by aspirating the first Pb and a small amount of the surrounding cytoplasm in a 100 μl microdrop of PBS supplemented with 7.5 μg/ml cytochalasin B (CB) and 10% FBS. The extracted cytoplasm was examined under ultraviolet illumination in another microdrop to confirm successful enucleation. The starved fetal bovine fibroblasts were used as donor cells and injected into the perivitelline space of enucleated oocytes. The reconstructed embryo was transferred into the BTXpress cytofusion medium C, and the oocyte-cell couplet was fused with a double electrical pulse of 35 V for 10 μs (Wang et al., 2017). The reconstructed embryos were incubated in modified synthetic oviduct fluid (mSOF) containing 5 μg/ml CB for 2 h, then treated with 5 μM ionomycin for 4 min and 1.9 mM dimethylaminopyridine for 4 h. Finally, the reconstructed embryos were cultured in BO-IVC medium (ivf Bioscience) in humid atmosphere of 5% CO₂ at 38.5°C for 7 days.

IVF was performed as described previously (Wang et al., 2011). Briefly, excess cumulus cells were removed mechanically from around the oocytes in PBS, and COCs with uniform cytoplasm and about three layers of cumulus cell were selected and transferred into a 50 μl microdrop of BO-IVF medium (ivf Bioscience) covered with mineral oil. Then, 2×10⁶ spermatozoa/ml of sperm suspension was added into the microdrop. After fertilization for 8 h, the zygotes were treated with 0.1% bovine testicular hyaluronidase and washed thrice in mSOF medium to remove the cumulus cells and redundant spermatozoa. Oocytes with a second Pb were considered successfully fertilized and were transferred into BO-IVC medium, covered with mineral oil and incubated in a humid atmosphere of 5% CO₂ at 38.5°C.

The miR-202 mimic, miR-202 inhibitor (RiboBio) and SEPT7 specific siRNA (GenePharma) were diluted in nuclease-free water to final concentrations of 20 μM. The SCNT embryos were injected with 10 pl of miR-202 mimic or SEPT7 siRNA between fusion and activation. In addition, the one-cell zygotes were injected with miR-202 inhibitor or with a mixture of miR-202 inhibitor and SEPT7 siRNA, respectively.

IVF or SCNT embryos at different stages (n=200) were rinsed thrice in PBS, and lysed in radio-immunoprecipitation assay buffer (RIPA; Beyotime Biotechnology) according to the experimental design. Proteins were separated on 6-10% polyacrylamide gels at 120 V for 90 min, and transferred onto PVDF membranes (Millipore) at 250 mA for 90-120 min. Afterwards, the membranes were treated with blocking buffer (Beyotime Biotechnology) at room temperature for 30 min, followed by incubation with diluted primary antibodies against SEPT7 (1:500; Abcam, 186021), HDAC6 (1:1000; Sigma, SAB1406911), α-tubulin (acetyl K40) (1:1000; Abcam, ab24610) or GAPDH (1:1000; TransGen Biotech, HC301-01) overnight at 4°C. The membranes were washed with Tris-buffered saline containing 0.5% (v/v) Tween 20 and incubated with horseradish peroxidase (HRP)-conjugated goat anti-mouse/rabbit IgG(H+L) (1:1000; TransGen Biotech, HS201-01/HS101-01) at room temperature for 2 h. Western Lumax Light Superior HRP substrate (ZETA LIFE) was used for visualizing target proteins.

After treating in Tyrode's solution to remove the zona pellucida and washing in PBS-polyvinyl alcohol (PVA), embryos were fixed in 4% paraformaldehyde fixing solution (Beyotime), permeabilized in PBS supplemented with 1% Triton X-100 and incubated in blocking buffer for immunostaining (Beyotime, China) at room temperature. The embryos were incubated overnight at 4°C with the mixed antibodies of SEPT7 (1:100, Proteintech, 13818-1-AP)+HDAC6 and SEPT7+α-tubulin-acetyl K40 diluted according to the manufacturer's instructions, followed by three rinses in PBS-PVA and incubation with a mixture of Alexa Fluor 488-labeled goat anti-rabbit IgG (1:500; Beyotime, A0423) and Alexa Fluor 555-labeled donkey anti-mouse IgG (1:500, Beyotime, A0460). The DNA was stained with 4′6-diamidino-2-phenylindole (DAPI; Beyotime). Glass slides with embryos were examined under the Nikon Eclipse Ti-S microscope equipped with a 198 Nikon DS-Ri1 digital camera. The cell count of blastocysts was conducted by co-staining CDX-2 with DAPI, and the method was as described above except that the zona pellucida was not removed. The blastocysts were incubated with primary antibodies against CDX-2 (BioGenex), followed by Alexa Fluor 555-labeled goat anti-mouse IgG.

The proximity ligation assay (PLA) is a new DNA ligation technology established in recent years for protein modification, quantification, localization and interaction studies. The technique allows detection of the interaction with the target protein through conversion into a fluorescent signal. PLA of SEPT7 and HDAC6 in bovine embryos was performed using Duolink In Situ Detection Reagents Red (Sigma) as described previously (Bedzhov and Stemmler, 2015). After removal of the zona pellucida, fixation, permeabilization and blocking, the samples were co-incubated overnight at 4°C with SEPT7 and HDAC6 primary antibodies, which were diluted with 1×antibody diluent, and followed by incubation with PLA probes (PLUS and MINUS) at 37°C for 2 h. As the PLA probe is labeled with a single-stranded DNA, if the two target proteins are associated, the plus and minus probes are close to each other. A 1×wash buffer A was used to remove unbound probes, and the embryos were subjected to ligation and amplification reactions, sequentially. Next, 1× and 0.01× wash buffer B were used to wash samples, and mounting medium with DAPI to label DNA. Embryos on glass slides were analyzed using a Nikon Eclipse Ti-S microscope equipped with a 198 Nikon DS-Ri1 digital camera.

Statistical analyses were conducted using the SPSS 22.0 software package. All experiments were repeated at least three times, and the data were expressed as mean±s.e.m. For qRT-PCR analysis, the value of 2^−ΔΔCt was calculated to compare the relative expression of the experimental and control groups. The two-tailed Student's t-test was used for pairwise comparisons, and the one-way ANOVA was applied for multiple group comparisons. Differences were considered statistically significant when P<0.05.

We thank Younan Wang for providing the ovaries, epididymal cauda and fetuses of Qinchuan cattle. In addition, we thank Rui Cheng for bioinformatics analysis.

The peer review history is available online at https://dev.biologists.org/lookup/doi/10.1242/dev.189670.reviewer-comments.pdf