Expression, structure and function analysis of the sperm-oocyte fusion genes Juno and Izumo1 in sheep (Ovis aries)

Background JUNO and IZUMO1 are the first receptor-ligand protein pairs discovered to be essential for sperm-oocyte fusion; their interaction is indispensable for fertilization. Methods PCR was used to clone the full-length DNA sequence of the Juno gene in sheep. The single nucleotide polymorphism (SNP) loci of Juno were genotyped by Sequenom MassARRAY®. PCR combined with rapid amplification of cDNA Ends were used to clone the full-length cDNA sequence of Juno and Izumo1. Reverse transcriptase-PCR (RT-PCR) and real time-quantitative-PCR (RT-qPCR) were used to analyze the genes’ expression in tissues of sheep, and single cell RNA-seq was used to analyze the genes’ expression in oocytes, granulosa cells and follicular theca of polytocous and monotocous Small Tail Han ewes. Bioinformatics was used to analyze advanced structure and phylogeny of JUNO and IZUMO1 proteins. Results The full-length DNA sequence of the Juno gene in sheep was cloned and nine SNPs were screened. We found a significant association between the g.848253 C > A locus of Juno and litter size of Small Tail Han sheep (P < 0.05). The full-length cDNA sequence of Juno and Izumo1 genes from Small Tail Han sheep were obtained. We found a new segment of the Izumo1 CDS consisting of 35 bp, and we confirmed the Izumo1 gene has 9 exons, not 8. RT-qPCR showed that Juno and Izumo1 genes were highly expressed in ovarian and testicular tissues, respectively (P < 0.01). Single cell RNA-seq showed Juno was specifically expressed in oocytes, but not in granulosa cells or follicular theca, while Izumo1 displayed little to no expression in all three cell types. There was no difference in expression of the Juno gene in oocyte and ovarian tissue in sheep with different litter sizes, indicating expression of Juno is not related to litter size traits. Bioinformatic analysis revealed the g.848253 C > A locus of Juno results in a nonconservative missense point mutation leading to a change from Phe to Leu at position 219 in the amino acid sequence. Conclusions For the first time, this study systematically analyzed the expression, structure and function of Juno and Izumo1 genes and their encoded proteins in Small Tail Han sheep, providing the basis for future studies of the regulatory mechanisms of Juno and Izumo1 genes. Supplementary Information The online version contains supplementary material available at 10.1186/s40104-021-00548-4.


Introduction
Sexual reproduction results in completely new individuals and ensures continuous reproduction for most eukaryotes. Fertilizationthe encounter and combination of two morphologically distinct male (sperm) and female (oocyte) germ cellsis the most important part of mammalian sexual reproduction. These two haploid germ cells meet in the reproductive tube of the mother (e.g., the oviduct in mammals), recognize each other, fuse and become a diploid that develops into a genetically unique organism.
Until recently, sperm-oocyte fusion has been a mysterious phenomenon. In 2005, Inoue et al. identified a sperm surface protein, IZUMO1, that is essential for binding to oocytes [1]. Each of the four members of the gene family, Izumo1, Izumo2, Izumo3, and Izumo4, share a homologous N-terminal "IZUMO" domain [2].
It took more than 9 years to find the cognate receptor of IZUMO1 on the oocyte cell surface. In the meantime, several receptor proteins were explored as potential recognition factors [3], such as CD9 [4][5][6][7]. Although CD9 and IZUMO1 are both essential for fertilization to occur, CD9 was ruled out because IZUMO1 is capable of binding to both wild-type and CD9 deficient oocytes. In 2014, Bianche et al. devised a new technique to identify low-affinity extracellular interactions that allowed them to uncover the IZUMO1 receptora GPI-anchored protein called folate receptor 4 (Folr4). Because Folr4 turned out to be the sole receptor for IZUMO1 (and was not able to bind folate after all), it was aptly renamed JUNO, after the Roman goddess of marriage [8]. While the IZUMO1-JUNO adhesion complex is the first ligand-receptor pair known to be essential for fertilization, a recent study found a new protein involved in sperm-oocyte membrane fusion during fertilization, Fertilization Influencing Membrane Protein (FIMP) [9].
The recognition and binding of IZUMO1 and JUNO is an adhesion event, conserved in both human and mouse oocytes [10]. Within 40 min of the first adhesion event between JUNO and IZUMO1, JUNO is completely absent from the oocyte cell surface. During this time, JUNO is packed into vesicles and released outside of the oocyte membrane where it quickly neutralizes other activated sperm. This shedding mechanism ensures the oocyte can only fuse with one sperm, as polyspermy results in a nonviable embryo [11]. Because of the key roles of IZUMO1 and JUNO, their sensitivity is likely to affect the pregnancy rate and sperm-oocyte recognition efficiency of mammals, and overall affect the fecundity of mammals.
The interaction between IZUMO1 and JUNO is conserved in both mammals (e.g., human, mouse, and pig) and nonmammals (e.g., opossum) [8]. Most research on IZUMO1 and JUNO has been conducted in human [12] and mouse, whereas their relevance in sheep reproduction is rarely investigated. There are some research about Izumo1 in pig [13] and sheep [14]. However, research to date has not addressed JUNO in sheep. In this study, the single nucleotide polymorphisms (SNP) of the Juno gene in sheep were identified and genotyped; the association between each SNP and litter size was analyzed. The fulllength cDNA sequence of Juno and Izumo1 were cloned from the ovary of Small Tail Han sheep using PCR and RACE technologies, and bioinformatic analyses were carried out to characterize the genes and their hypothetical protein products. The expression of Juno and Izumo1 genes in various tissues were measured with RT-PCR and RT-qPCR. This study provides the basis for further revealing the function and expression regulation of the Juno and Izumo1 genes in sheep, and for further studying the molecular mechanisms regulating reproduction in sheep.

DNA extraction and detection
DNA was extracted with a blood genomic DNA extraction kit (Tiangen Biotech Co., LTD, Beijing), according to the manufacturer's instructions. The purity and concentration of DNA were detected with a NanoDrop 2000 spectrophotometer (Thermo Scientific), and genomic DNA integrity was detected by 1.5% agarose gel electrophoresis. Samples were only considered acceptable if the 260 nm to 280 nm ratio was in the range of 1.8-2.0, the concentration was above 30 ng/mL and the electrophoretic bands were qualified (Fig. S1).

Primer design and synthesis
According to the published genome sequence of the sheep Juno gene (GenBank accession: NC_019472.2), three primers were designed using Primer Premier 5.0 software. To improve the accuracy of sequencing, primers were designed in segments. The primers were synthesized by Beijing Tianyihuiyuan Biotechnology Co., Ltd. Primer sequence, product size and annealing temperature are shown in Table S1.

DNA amplification
Using the extracted blood genomic DNA as templates, the target gene DNA was amplified using the TaKaRa high fidelity LA Taq enzyme system. Each 20 μL reaction contained 0.25 μL TaKaRa LA Taq (5 U/μL), 10 μL 2× GC Buffer I (5 mmol/L Mg 2+ Plus), 3 μL dNTPs (2.5 mmol/L of each dNTP), 4.25 μL RNase-free ddH 2 0, 0.5 μL forward primer (Table S1), 0.5 μL reverse primer and 1.5 μL blood genomic DNA. The PCR program started with denaturation at 95°C for 5 min, and was followed by 35 cycles of denaturation at 95°C for 30 s, annealing at 66°C for 45 s and extension at 72°C for 2 min, followed by a final extension at 72°C for 8 min.
After the reaction, the PCR product was detected by 1.5% agarose gel electrophoresis, and the positive amplification products were cloned and sequenced (Sangon Biotech Co., Ltd., Shanghai, China).

SNP loci screening
DNA sequence comparisons and analysis were conducted with DNAMAN 6.0, and peak map judgment was conducted with Chromas Pro 1.5. The DNA sequence of Juno was compared with re-sequencing data from our laboratory consisting of 100 individual sheep from 10 different breeds [15]. A loci was considered an SNP if the proportion of different bases at the same locus were more than 30% SNP loci were screened to identify synonymous and non-synonymous mutations.

Genotyping
Three pairs of primers were used to screen SNP loci of the Juno gene in 60 samples of Small Tail Han sheep and Sunite sheep. Single bands, signifying good specificity in PCR amplification were sequenced bidirectionally. The loci with clear sequence peaks were used for SNP statistics. The SNP loci of the Juno gene were genotyped in the blood DNA samples of Small Tail Han sheep,  Sunite sheep, Tan sheep, Suffolk sheep, Dorper sheep  and Prairie Tibetan sheep with the iPLEX™ assay using  the MassARRAY® genotyping system (Sequenom, San Diego, CA, USA). The experimental steps were carried out following the instrument operating guidelines.

Statistical analysis
Popgene 32 (version 3.2) and PIC-CALC were used to process the results of genotyping and calculate the heterozygosity (HE) and the polymorphism information content (PIC). The genotype and gene frequencies of SNP loci in polytocous and monotocous breeds were subjected to a chi-square test using SPSS 18.0, and the correlation between genotype frequencies and litter size was analyzed. The relevance analysis model was: Where, y is the phenotypic value of the litter size, μ is the population mean, G i is the genotype effect, P j is parity effect, Sk is the field effect, and e ijkl is the random residual effect of each observation. Assuming each observation was independent, we used a N (0, σ 2 ) distribution. cDNA cloning and differential expression analysis of Juno and Izumo1 genes in sheep Materials and main reagents Three male Small Tail Han sheep, three female Small Tail Han sheep, three male Sunite sheep, and three female Sunite sheep in good health and with the same bodyweight and sexual maturity were selected for euthanasia,sheep and Sunite sheep for euthanasia. Sixteen types of tissues, including gonadal (testis, epididymis, sperm duct/ovary, uterus, oviduct), pineal gland, hypothalamus, pituitary, cerebellum, heart, liver, spleen, lung, kidney, skeletal muscle, duodenum, subcutaneous abdominal fat and adrenal gland were collected quickly and put into a pre-labeled 1.8 mL RNase-Free cryopreservation tube. The tubes were immediately placed in liquid nitrogen and transferred to a cryogenic refrigerator (− 80°C) for storage.
TRIZOL® Samples from different parts of each sheep were ground in liquid nitrogen, and then total RNA was extracted with the Total RNA Extraction Kit for Animal Tissues (DP431). RNA integrity was assessed by 1% agarose gel electrophoresis, and RNA concentration and purity were measured by nanodrop spectrophotometry. Only RNA samples displaying a 260 nm/280 nm ratio between 1.8 and 2 were used in this study. The product was stored in a − 80°C cryogenic freezer until further use.
Coding sequence (CDS) cloning of Juno and Izumo1 genes Total RNA of the ovaries and testes of Small Tail Han sheep was subjected to RT-PCR with the PrimeScript™ RT reagent kit. The reverse transcription products were stored at − 20°C. According to the predicted Juno and Izumo1 genes sequences (NCBI reference sequences: XM_015100756.1 and NM_001164492.1), the amplification primers of Juno and Izumo1 were designed with Primer Premier 5.0 software (Table S2).
The cDNA from ovaries and testes of Small Tail Han sheep obtained by RT-PCR was used as templates to clone the Juno and Izumo1 genes using the TaKaRa LA Taq® with GC Buffer (RR02AG) enzyme system. Each 20 μL reaction contained 1 μL cDNA, 0.5 μL forward primer (Table S2), 0.5 μL reverse primer, 10 μL 2× Taq PCR Master Mix and 8 μL ddH 2 O. The PCR program started with denaturation at 95°C for 5 min, followed by 35 cycles of denaturation at 95°C for 30 s, annealing at 62°C for 30 s and extension at 72°C for 5 s, followed by a final extension at 72°C for 8 min. The amplified products were detected by 1.5% agarose gel electrophoresis. The target fragments were recovered using the TaKaRa MiniBEST Agarose Gel DNA Extraction Kit Ver.4.0 (9762), ligated into a pMD18-T (TaKaRa) vector, and transformed into DH5α competent cells. Ampicillin resistant colonies were screened by colony PCR and DNA fragments were sent to Sangon Biotech Co., Ltd. (Shanghai) for sequencing.
3′ /5′ Rapid amplification of cDNA ends (RACE) of Juno and Izumo1 cDNA Total RNA of ovaries and testes of Small Tail Han sheep were used as templates, and a SMARTer® RACE 5′/3′Kit was used for reverse transcription. According to the sequence of the CDS region obtained by sequencing, 5′ and 3′ RACE specific primers were designed according to the requirements of the SMARTer® RACE 5′/3′ Kit. Using the reversed first chain of 3′ RACE and 5′ RACE as templates, the 3′ and 5′ ends of the Juno and Izumo1 genes from sheep were amplified by Touch Down PCR using specific primers 3'GSP and 5'GSP, and a universal primer UPM (Table S2).

RT-PCR of Juno and Izumo1 genes
Using cDNA obtained from 16 tissues of Small Tail Han sheep and Sunite sheep as templates, Juno-RT primers, Izumo1-RT primers and reference gene (βactin) primers, PCR amplification was conducted. Each 20 μL reaction consisted of 10 μL 2× Taq PCR Master Mix, 8 μL ddH 2 O, 0.5 μL forward primer, 0.5 μL reverse primer and 1 μL cDNA. The PCR program began with denaturation at 95°C for 5 min, followed by 25 cycles of denaturation at 95°C for 30 s, annealing at 62°C for 30 s and extension at 72°C for 5 s, followed by a final extension at 72°C for 8 min. The amplified products were analyzed by agarose gel electrophoresis. The quality of the cDNA obtained from reverse transcription was amplified with the primers of the reference gene β-actin to verify that it could be used for subsequent experiments.

RT-qPCR of the Juno and Izumo1 genes in sheep
Using cDNA obtained from 16 tissues of Small Tail Han sheep and Sunite sheep as templates, RT-qPCR was conducted. Each 20 μL reaction contained 10 μL SYBR Premix Ex Taq II, 6.4 μL RNase-Free ddH 2 O, 0.8 μL forward primer, 0.8 μL reverse primer and 2 μL cDNA. The RT-qPCR reaction started with denaturation at 95°C for 5 s, followed by 40 cycles of denaturation at 95°C for 5 s and annealing at 60°C for 30 s. Each sample was run in triplicate and β-actin was used as the reference gene. The relative expression of each target gene was calculated with the 2 −△△Ct method. The data were analyzed with the statistical software SPSS 18.0, and the statistical data were reported as the mean ± the standard deviation. One-way ANOVA was used to detect differences in expression among different tissues, and the Student's t-test was used to compare the differences in expression among the same tissues. P < 0.05 indicated a significant difference.

Single cell expression of genes by single cell RNA-seq
Polytocous and monotocous Small Tail Han ewes 3 years of age and with three or higher parity records were selected for single cell gene expression analysis. There were five ewes in each group. A controlled internal drug release (CIDR) device (InterAg company, New Zealand) was used to synchronize the estrus cycles of ewes by administering 300 mg progesterone. After 12 d, the CIDR device was removed. Forty-five h after the CIDR device was removed, the ewes were operated on to collect oocytes, granulosa cells and follicular theca.
The Smart-Seq2 method was used to directly measure gene expression in single cells [16]. A Qubit® 3.0 Fluorometer (Life Technologies, CA, USA) and an Agilent 2100 High Sensitivity DNA Assay Kit (Agilent Technologies, CA, USA) were used to detect the amplified cDNA concentration and fragment distribution. After that, single-cell cDNA libraries were constructed, and the length distribution of the libraries was detected using an Agilent 2100/LabChip GX Touch. After meeting the length distribution requirements, qPCR was used to accurately quantify the effective concentration of the cDNA in the libraries (more than 10 nmol/L). After the effective concentration of the libraries was qualified, the PE150 double terminal sequencing program was run on the HiSeq sequencing platform for single cell RNA-seq analysis.
Raw reads from Illumina sequencing were filtered to get clean reads for subsequent analyses. With the Hierarchical Graph FM index (HGFM) as the core retrieval method, HiSAT2 was used to filter the RNA-seq data of each sample with the sheep reference genome (Ovis_aries Oar_ v3.1pep_all.fasta). Using StringTie software, RNA-seq comparison results were quickly assembled into transcripts and gene expression levels were predicted.

Bioinformatic analysis of Juno and Izumo1 cDNAs
The sequencing results were subjected to a BLAST search (NCBI, http://blast.ncbi.nlm.nih.gov/Blast.cgi) to retrieve homologous gene sequences, and DNAMAN software was used to splice the sequences into the fulllength cDNA sequence of each gene. Open reading frame (ORF) Finder (http://www.ncbi.nlm.nih.gov/ projects/gorf/) was used to determine the ORF and predict the amino acid sequence. ProtParam (http://web. expasy.org/protparam/) was used to predict the physical parameters of each protein such as molecular weight. The hydrophilicity and hydrophobicity of JUNO and IZUMO1 of Small Tail Han sheep were analyzed using ProtScale (https://web.expasy.org/protscale/). Prediction of the secondary structure of each protein and its variants was carried out using PredictProtein (https://www. predictprotein.org/). Phyre2 was used to predict the protein signal peptide, protein transmembrane region and protein tertiary structure. Multiple amino acid sequences were compared using DNAMAN software, and a phylogenetic tree was constructed using the Neighbor-Joining method in MEGA7 software (reference).

SNPs identification based on whole genome re-sequencing results
Based on whole genome re-sequencing data from our laboratory consisting of 100 individual sheep from 10 different breeds [15], all SNPs near this region of Juno gene were checked, nine SNP loci of Juno were identified: g.844598A > G, g.846206C > T, g.846268A > G, g.847219G > C, g.847810C > G, g.847814C > T, g.847830T > C, g.847927A > C and g.848253C > A. Among them, eight were in exons and one was in a 5′ UTR. Of the 8 exonic SNPs, five were missense mutations and 3 were synonymous mutations. There were three T/C transitions, two A/G transitions, two C/A transversions and two G/C transversions.

Association between SNP loci of Juno and litter size in sheep
According to the data obtained from genotyping, nine SNP loci were preliminarily screened, and SNP loci with no difference in the allele frequencies between monotocous and polytocous populations were excluded. Seven SNP loci were further analyzed (Table 1). Genotyping showed that there were three genotypes at Juno gene loci g.844598A > G, g.847830T > C, g.847927A > C, g.846268A > G and g.847810C > G in monotocous and polytocous breeds, two genotypes (GG and CG) at g.847219G > C in the monotocous breed and three genotypes (GG, CG and CC) in polytocous breed (Table 1). There were CC and CA genotypes in monotocous breeds and CC, CA and AA genotypes in polytocous breeds at the g.848253C > A locus. Among them, the genotype frequencies of Juno gene loci g.847830T > C, g.846268A > G and g.847219G > C were significantly different between monotocous and polytocous sheep populations (P < 0.01). The allele frequencies of Juno gene on loci g.847927A > C, g.846268A > G and g.847219G > C were significantly different between monotocous and polytocous sheep populations (P < 0.01).
We analyzed the polymorphism distributions of Juno gene in six sheep breeds ( Table 2). The polymorphic information contents (PIC) of loci g.847219G > C and g.848253C > A were low in each of the six sheep breeds (PIC < 0.25), whereas the PIC in loci g.844598A > G, g.847830T > C, g.847927A > C, g.846268A > G and g.847810C > G were both low (PIC < 0.25) and moderate (0.25 < PIC < 0.5). Among seven polymorphic loci, only g. 847810C > G was under Hardy-Weinberg equilibrium in all six sheep breeds ( Table 2).
Association analysis between each of the seven polymorphic loci and litter size of Small Tail Han sheep revealed a significant association between locus g.848253C > A and litter size in different parities (P < 0.05) ( Table 3). Overall, the litter size with the CC genotype was greater than that with AA genotype at different parities, but there was no significant difference on allele frequencies between the CC and CA genotypes. Small Tail Han sheep is a polytocous breed with an average litter size of 2.67. Taking the third parity of Small Tail Han sheep as an example, the average litter size of the CC genotype was 2.88 (±0.10), while that of the AA genotype was 1.50 (±0.33).
cDNA cloning and expression profiles of Juno and Izumo1 genes in sheep CDS cloning of Juno and Izumo1 The RT-PCR products of the Juno and Izumo1 genes from Small Tail Han sheep were 463 bp and 895 bp (Fig. S2), respectively. The Juno gene from Small Tail Han sheep was 98% identity to the gene from goat (Capra hircus, Gen-Bank accession: XM_018058935.1) and 95% identity to the gene from Texel sheep (Ovis aries, GenBank accession: XM_015100756.1). The Izumo1 gene from Small Tail Han sheep was 95% identity to the gene from goat (Capra hircus, GenBank accession: NM_ 001287235.1) and 99% identity to the gene from Texel sheep (Ovis aries, GenBank accession: NM_ 001164492.1). The high similarity of Small Tail Han sheep Izumo1 and Juno gene sequences indicated that those genes are very conservative among Bovidae.

Full-length amplification of Juno and Izumo1 cDNA
Using the obtained partial RT-PCR product of the Juno gene of Small Tail Han sheep as template, a 572-bp fragment was obtained by 3′ RACE, and 658 bp and 518 bp fragments were obtained by 5′ RACE (Fig. S2). The full sequence of the Juno gene was 1131 bpwhich contained a 50-bp 5′ UTR, a 349-bp 3′ UTR and a 732-bp CDS. The gene contained 4 exons and 3 introns, encoding a protein comprised of 243 amino acids.
Using the partial RT-PCR product of the Izumo1 gene of Small Tail Han sheep as the template, 290 bp, and 1010 bp were obtained by 3′ RACE and 5′ RACE, respectively (Fig. S2). The Izumo1 gene was 1565 bp, consisting of 9 exons and 8 introns. It contained a 433-bp 5′ UTR, a 97-bp 3′ UTR and a 1035-bp CDS. The initial codon ATG began at nt 434 and the termination codon TAG began at nt 1466, encoding a protein 344 amino acids in length.
We previous reported that the DNA sequence of Izumo1 gene from Texel sheep (Oar_v4.0: NC_019471.2 and Oar_v3.1: NC_019471.1) was 57 shorter than the corresponding genes in Small Tail Han sheep [17]. A 35bp sequence 5'-ATGAGGCCACACTGGAAAAGGCAT CCTGGAGTTTG-3' derived from this 57 bp DNA sequence. The 35 bp insertionfrom 239-273 bp of the CDS sequence of the Izumo1 gene in Small Tail Han sheep did not exist in the CDS of Texel sheep (Fig. 1).
After sequence alignment, it demonstrated that 35 bp of the extra 57 bp belongs to exon 2 (Fig. 1c, Exon 2   : Different lowcases indicate statistically significant differences (P < 0.05) belongs to intron 1 (Fig. 1c) in Small Tail Han sheep. We further compared the CDS of Izumo1 in Texel Sheep (Ensembl: ENSOART00000013198.1 IZUMO1-201 cDNA:protein_coding) and found that the second half of exon 2 (exon 2 in Texel sheep-Part 2) was the same as that of Small Tail Han sheep (exon 2 in Small Tail Han sheep-Part 2). But the first half sequence of exon 2 of Texel Sheep (exon 2 in Texel sheep-Part 1) was in the intron 1 region of the Izumo1 genome of Small Tail Han sheep (Fig. 1b, c, d). We speculated that the 57 bp DNA sequence of Small Tail Han sheep may lead to different transcript splicing patterns. In addition, we compared the Izumo1 DNA sequence of Rambouillet sheep to that from Small Tail Han sheep (Fig. 1a, GenBank accession: NC_040265.1). The Izumo1 DNA sequence of Rambouillet sheep was highly similar to that of Small Tail Han sheep and Sunite sheep and it contained a similar insert. We suspect the CDS of Rambouillet sheep may also be similar to that of Small Tail Han sheep and Sunite sheep. In addition, we found that there were only eight exons in the sheep Izumo1 CDS in the NCBI database (GenBank accession: NM_001164492.1), and the CDS region of the sequence terminated early at exon 8. The full-length Izumo1 cDNA sequence of Small Tail Han sheep we obtained had nine complete exons (Fig. 1e). The cDNA sequence of Izumo1 of Texel Sheep provided in the Ensembl database (Ensembl: ENSOART00000013198.1 IZUMO1-201 cDNA: protein_coding) was incomplete, and the position of the terminator was not indicated. Therefore, this study provides the first full-length Izumo1 cDNA sequence of sheep.

Differential expression analysis of Juno and Izumo1 genes in tissues of sheep
Semi-quantitative RT-PCR results showed that the Juno gene was expressed in ovaries and oviducts both in Small Tail Han sheep and Sunite sheep, but not in the other 14 tissues tested (Fig. 2a). The Izumo1 gene was expressed only in the testes and epididymis of Small Tail Han sheep and Sunite sheep, but not in the other 14 tissues tested. The expression level of Izumo1 in the testes of the two sheep breeds was higher than that in the epididymis (Fig. 2b).
The expression of Juno and Izumo1 in different tissues of the reproductive gonad axis of sheep was measured by RT-qPCR (Fig. 3). Juno expression was significantly higher in the ovaries than all other tissues (P < 0.01), but there was no significant difference between Small Tail Han sheep and Sunite sheep (P > 0.05). Expression of Izumo1 was higher in the testes than all other tissues (P < 0.01), but there was no difference between expression in the other tissues. Unlike Juno, the expression of Izumo1 in the testes of Small Tail Han sheep, a polytocous breed, was significantly higher than that in Sunite sheep, a monotocous breed (P < 0.01). Although the expression of Izumo1 in other tissues of Small Tail Han sheep was higher than that of Sunite sheep, the difference was not significant (P > 0.05).

Gene expression profiles at single cell levels
There are many types of cells in ovarian tissue, and there is heterogeneity between cells. To further explore the expression and function of the Juno gene in different types of cells in ovarian tissue, we collected single cells, including oocytes, granulosa cells and follicular theca of polytocous and monotocous Small Tail Han ewes, and sequenced them by single cell RNA-seq. The expression levels of Juno, BMP15 and GDF9 were very high in oocytes, but very low in granulosa cells and follicular theca (Fig. 4). However, expression of BMPR1B and CD9 was high in granulosa cells and follicular theca but very low in oocytes. At the same time, we found that Izumo1 had barely expression in all three cell types. The results showed that in ovarian tissue, Juno was specifically expressed in oocytes, while Izumo1 was specifically expressed in testes, consistent with the function of the two genes. In addition, there was no difference in the expression of Juno in oocytes or ovaries between polytocous and monotocous Small Tail Han sheep, suggesting the difference in litter size is not caused by the level of Juno expression, but is possibly due to some other factor.

Bioinformatic analysis of JUNO and IZUMO1 proteins Physicochemical properties of JUNO and IZUMO1 proteins
ProtParam revealed that molecular weights of JUNO and IZUMO1 proteins are 27.93 kDa and 38.89 kDa, respectively. As shown in Figure S3, the signal peptide region at the N-terminus of JUNO and the anchoring region at the C-terminus had strong hydrophobic regions, among which leucine at position 235 was the most hydrophobic (the highest score is 2.5); in the mature peptide region, hydrophobic amino acids and hydrophilic amino acids were evenly distributed. The signal peptide region at the N-terminus and the transmembrane region at the Cterminus of IZUMO1 had strong hydrophobic regions, of which the IIe at position 319 had the strongest hydrophobicity (the highest score was 3.567); hydrophobic and hydrophilic amino acids were also evenly distributed in the mature peptide region.

Advanced structural prediction of JUNO and IZUMO1 in Small Tail Han sheep
The secondary structures of Small Tail Han sheep JUNO and IZUMO1 proteins were predicted (Fig. 5). Mutation of the Juno gene at the g.848253C > A locus caused an amino acid change from Phe to Leu at position 219. The predicted secondary structures showed there were 14 protein binding regions in the JUNO protein before the mutation at g.848253C > A. After the mutation, three protein binding regions (143, 171, and 228) were lost. There were only five protein binding regions in the predicted secondary structure of IZUMO1, based on the full-length Izumo1 cDNA of Small Tail Han sheep. But there were two disulfide bonds in the IZUMO1 protein.
Phyre2 was used to analyze the tertiary structures of JUNO and IZUMO1 from Small Tail Han sheep (Fig. 6). The tertiary structure of JUNO was clustered with human oocyte surface protein JUNO, having 75% similarity, whereas it had only 62% similarity to folate receptor α. The tertiary structure of Small Tail Han sheep IZUMO1 was rod-shaped, and had 74% similarity with human sperm IZUMO1.

Sequence alignment and phylogenetic analysis of JUNO and IZUMO1 proteins
The homology between JUNO and IZUMO1 protein orthologs was analyzed. The results showed that the amino acid sequence of JUNO and IZUMO1 in Small Tail Han sheep were conservative among sheep, goat and wild yak. The similarity was much lower with wild boar, human, rhesus monkey, rabbit, rat and mouse (Table 4). A phylogenetic tree was constructed for JUNO and IZUMO1 using the Neighbor-Joining method (Fig. 7). Small Tail Han sheep and sheep (Ovis aries) first gathered into one branch, then gathered together with goat (Capra hircus).
We also compared the amino acid sequences of JUNO proteins from different species. We found that compared with human and mouse JUNO proteins, the mutated amino acid (F219L) in Small Tail Han sheep was located in α5 region of human JUNO protein [12] andα7 region of mouse JUNO protein [18], which was closed to GPI signal region, but not belong to residues that interact with Izumo1 and the other three flexible regions.

Discussion
The cloning of Juno and Izumo1 genes Research on the Juno gene has mainly focused on its role in mouse and human tumors [19][20][21], whereas research The phylogenetic tree showed that the Small Tail Han sheep JUNO protein had the closest relationship with goat and sheep; the similarity was over 96%, indicating the JUNO protein has been highly conserved throughout evolution.
After identification of the IZUMO1 protein by Inoue et al. [1], many studies have been done on the mammalian Izumo1 gene, but the cloning and analysis of its cDNA sequence have mostly focused on human and mouse, with a couple of exceptions. Xing et al. [14] cloned and obtained the cDNA sequence of the Izumo1 gene of the cashmere goat and sheep (breed unknown). Evaluation of the sequences revealed a typical ORF of 963 bp that could be spliced into 8 exons. Kim et al. [13] cloned and obtained the complete ORF sequence of the porcine Izumo1 gene using RACE technology. We found that most of the currently published Izumo1 cDNA sequences contained a 963-bp ORF, encoding a 320-amino acid protein, which is consistent with the results of this study, showing that the amino acid sequence of IZUMO1 is relatively conserved among species. Although the Izumo1 cDNA sequence has been reported in sheep, the breed is unknown and there is no full-length cDNA sequence for Small Tail Han sheep. Previously, we were the first to clone the full-length Izumo1 DNA sequence of Small Tail Han sheep [17], and in this study, we obtained the first full-length Izumo1 The amino acid homology analyses showed that Small Tail Han sheep IZUMO1 had 95% or higher similarity with sheep, goat and cattle. The phylogenetic tree showed that the IZUMO1 protein of Small Tail Han sheep was closely related to that of sheep, goat and

Expression of Juno and Izumo1 genes
Oocytes are produced in the ovary and mature there. They next enter the oviduct where they await fertilization by the sperm, and time is limited. Once fertilized, oocytes travel to the uterus for implantation. Therefore, the oocyte content in oviduct and uterus is relatively low and fluctuates greatly with time. Transcriptomic data of 40 different sheep tissues showed that expression of Juno in lymph node mesenteric, lymph node prescapular and peyer's patch was the highest, followed by that in ovarian follicles [22]. Juno expression in the ovary was slightly lower than in ovarian follicles, while expression in the uterus and corpus luteum was very low [22]. The quantitative results of the present study showed that Juno mRNA was highly expressed in ovary, but weakly or not at all expressed in the oviduct, uterus and other tissues, consistent with transcriptomic data. From the RT-PCR result the Juno had a highly expressed in oviduct, but we believed it might be because of the accuracy of this assay. In order to solve this problem, qPCR and single cell RNA-seq were performed to quantify the expression of Juno precisely.
In this study, single cell RNA-seq data showed that expression of Juno was very high in oocytes, but very low in granulosa cells and follicle theca. Juno was specifically expressed in oocytes in ovarian tissue, which is consistent with the high expression of Juno gene in ovarian follicles in tissue transcriptomic data [22]. It can also be inferred that Juno is specifically expressed in oocytes, but not in other ovary cell types. The specific expression of Juno in oocytes is related to its function as a receptor during fertilization.
BMPR1B, BMP15 and GDF9 are all family members of TGF-β pathway. They are also major regulatory genes for litter size in sheep [23]. GDF9 and BMP15 produced by oocytes bind to the BMPR1B receptor located on granulosa cells and act through BMPR1B to jointly stimulate proliferation of granulosa cells [24], which is critical for normal follicular development in sheep. Whether in human or sheep, GDF9 and BMP15 have the same expression pattern in follicles [25,26]. Our single cell RNA-seq data also proved that BMP15 and GDF9 are highly expressed in oocytes, whereas BMPR1B was highly expressed in granulosa cells. Our RNA-seq data was very consistent with previous results. Like BMP15 and GDF9, Juno was specifically expressed in oocytes. However, in both single cells and tissue samples, there was no difference in the expression of Juno in sheep differing in litter size, indicating that expression of Juno is not related to litter size.
Satouh et al. has firstly described the exposure of IZUMO1 precisely. They visualized the dynamics of diffusion of IZUMO1 at the time of the acrosome reaction using a fluorescent protein tag. IZUMO1 localized in the equatorial segment of the sperm surface after the acrosome reaction [27]. Inoue et al. used a polyclonal antibody prepared by recombinant IZUMO protein for immunolabeling and found that the size of IZUMO in mice is 56.4 kDa, and that it is only expressed in the testes of mice [1]. Kim et al. also confirmed that IZUMO1 is only exposed on the sperm head during sperm-oocyte fusion [13]. Sperm is produced in the testes and then stored in the epididymis. Only when ejaculating will sperm pass through the sperm duct; thus, sperm content in sperm duct should be relatively low. Our data showed that there is a low level of expression in the epididymis, which may be due to the storage of some mature sperm in epididymis.
The structure and function of JUNO and IZUMO1 proteins JUNO and IZUMO1 proteins play an important role in the process of sperm and oocyte fusion. The JUNO-IZUMO1 recognition system has been deciphered by Xray crystallography [12,28]. After JUNO-IZUMO1 recognition, the dimer IZUMO1 may also specifically recognize a second unknown receptor on the oocyte surface. Sperm-oocyte fusion is a multi-molecule event, and there may be many other molecules involved. Research showed that recognition and binding of JUNO and IZUMO1 is a relatively weak adhesion [29]. After the sperm enters the zona pellucida, IZUMO1 adheres to JUNO, and CD9 on the oocyte accumulates between cells before sperm and oocyte fusion. CD9 was considered to be a protein cooperating with JUNO [30]. This adhesion is conserved in both human and mouse oocytes [10]. Moreover, this kind of adhesion may be transient, needs a sensitive detection method, such as highly sensitive interaction assay [8]. It is also well established that after the first adhesion event between JUNO and IZUMO1, JUNO is released out of the oocyte membrane. Forty minutes after sperm and oocyte fusion, JUNO can longer be detected on the surface of the oocyte membrane. The shedding of JUNO protein is a gradient reaction, which is completed within 40 min. The shedding mechanism of JUNO is closely related to the prevention of multiple sperm from entering the oocyte, ensuring the oocyte will only fuse with one sperm under natural conditions [11].
The tertiary structure prediction of JUNO showed that it has a compact structure with a pocket in the center, which may play an important role in the recognition and binding of ligands or receptor proteins [18]. The tertiary structure prediction of IZUMO1 from Small Tail Han sheep had two stable β-hairpins in the central region of the tertiary structure. Moreover, the α-helix domain in the front end and the immunoglobulin domain in the back end are firmly bound by a disulfide bond to form a slender rod-shaped protein, similar to the tertiary structure of human IZUMO1 [12,28]. Ohto et al. established that the disulfide bond formed in the β-hairpin region of the IZUMO1 protein is very important to maintain its structural stability [28].
The mutation of the Juno gene at locus g.848253C > A will lead to a change from Phe to Leu at site 219 in the amino acid sequence. This single amino acid difference causes a loss of three protein binding sites in JUNO. Interestingly, there was a significant association between the g.848253C > A locus and litter size of Small Tail Han sheep: the litter size of the AA genotype was significantly smaller than the litter size of the CC and CA genotypes. In the 3rd parity, the litter size of sheep with the AA genotype decreased by nearly 1.38 offspring compared with the CC genotype, and there was no difference in litter size between CC and CA genotypes. The latest research showed that four SNPs of the Juno gene were found in women with fertilization failure and polyspermy, and variation of Juno may have played a role in this pathogenesis [31]. If compare with human and mouse, this mutated amino acid (F219L) in Small Tail Han sheep was not belong to motifs that interact with Izumo1. At present, we don't know the specific function of this region, but it can be seen that this amino acid is essential for the functionality of JUNO protein in sheep. More functional analysis will be performed to make our doubt clearly in our future research.

Conclusions
In this study, we successfully obtained the full-length DNA sequence of the Juno gene, screened nine SNPs of Juno and genotyped these loci. The association analysis showed a significant association between different genotypes of the g.848253C > A locus and litter size of Small Tail Han sheep. The full-length cDNA sequences of Juno and Izumo1 were obtained and 35 bp of new Izumo1 CDS sequence of in Small Tail Han sheep was found, and we confirmed the Izumo1 gene has nine exons, not eight exons. Expression analysis showed that the Juno gene was highly expressed in ovary, whereas the Izumo1 gene was highly expressed in testes. In addition, expression of the Izumo1 gene in the testes of polytocous Small Tail Han sheep was significantly higher than that in monotocous Sunite sheep, indicating that expression of Izumo1 in testes positively correlates to fecundity. Single cell RNA-seq showed Juno is specifically expressed in oocytes, but not in granulosa cells and follicular theca, while Izumo1 was barely expressed in all three cell types. There was no difference in the expression of Juno in oocyte and ovarian tissue in sheep with different litter size, indicating expression of Juno is not related to litter size traits. Bioinformatic analysis revealed the g.848253C > A locus of Juno results in a nonconservative missense point mutation leading to a change from Phe to Leu at position 219 in the amino acid sequence. For the first time, this study systematically analyzed the expression, structure and function of Juno and Izumo1 genes and their encoded proteins in Small Tail Han sheep, providing the basis for further research, unravelling the functional mechanisms of Juno and Izumo1 genes in sheep.

Supplementary Information
The online version contains supplementary material available at https://doi. org/10.1186/s40104-021-00548-4.  Table S1: Primers used for amplifying DNA of the Juno gene in sheep, Table S2: Primers used for amplifying cDNA and RT-qPCR of the Juno and Izumo1 genes in sheep. Table S3. SNP loci screening.