Differentially Expressed miRNAs in Mouse Uterus during Embryo Implantation

Embryo implantation is a crucial event in the early pregnancy. In the case of implantation, a highly coordinated process is set into motion where by specialized cells of the embryo, establish contact with a specialized tissue of the mother, the uterus (Carson et al., 2000). It refers to the process in which both the embryo and maternal uterus accept each other through a series of signal factors and hormone regulation. The uterus of almost all mammals is allowed to implant only at a specific time, which is called the ‘window’ period of embryo implantation. In mice, embryo implantation occurs around midnight on D4 (D1 =vaginal plug) while the embryo develops to the blastocyst stage (Matsuo and Hiramatsu, 2017). In this process, a large number of genes need to be regulated to participate in the regulation of the uterine state.

MiRNA is an endogenous, non-coding small RNA with a length of 18-24 nucleotides. It silences the expression of target mRNA by binding to the 3' end and ultimately inhibits protein translation (Behrens et al., 1996). MiRNAs regulate more than 30% of chromosomes in the human genome (Bartel, 2004). The research on miRNA has gradually expanded from human and model animals to other common livestock (Basang et al., 2018), new miRNAs were often reported (Chang et al., 2018). As one of the most important regulators, miRNA can affect cell proliferation, apoptosis, adhesion, angiogenesis and migration directly or indirectly (Chen et al., 2016; Paramasivam et al., 2017).

To achieve a mutually acceptable state between the maternal uterus and the embryo in the process of implantation, it is necessary to regulate many related genes to make the maternal body reach an optimally receptive state. In this process, miRNA plays an important role in posttranscriptional regulation (Liang et al., 2017). Here, we collected the uteri tissue of D1, D4, D5IMS and D5IIS after pregnant for small RNA sequencing and screened out the differentially expressed miRNAs to find out which miRNAs play regulatory roles in the window period of embryo implantation. Then the target genes regulated by miR-183-5p were further analyzed by transcriptome sequencing. This study tried to focus on miRNA changes occurring during early gestation and at the implantation site, which provided a basis for revealing the dynamic expression and regulation of miRNA.
 

The experiment was conducted in Zhejiang University in 2018 and 2019. The whole experiment flow is shown in Fig 1. Mice were housed in a temperature- and humidity-controlled room with a 12/12h light/dark cycle. Adult ICR females were mated with fertile males. The morning of recovery of a vaginal plug was designated as D1. The implantation sites (IMS) were visualized as blue bands by intravenous injection of 1% Trypan Blue dye solution (0.1 mL/mouse) and the regions between the blue bands were defined as inter-implantation sites (IIS). Uterus on D1, D4, D5IMS and D5IIS was collected for small RNA sequencing analysis.


 
Small RNA sequencing analysis
 
Small RNA was isolated from D1, D4, D5IMS and D5IIS pregnant uteri, each sample repeat three times. The filtered reads were compared with the authoritative miRNAs/rRNA/tRNA/snRNA (Small nuclear ribonucleic acids)/snoRNA (Small nucleolar RNAs) database and the known ncRNAs were annotated. The microRNAs database is from microRNABase version 21 (www.mirbase.org). The databases of rRNA, tRNA, snRNA and snoRNA are from Rfam12.1 (rfam.xfam.org). The sequencing data have been deposited in the NCBI SRA database with the BioProject number of PRJNA562942.
 
Differentially expressed miRNA analysis
 
The miRNAs with differential expression in six paired samples were screened by the DEGseq package (Wang et al., 2010) with the threshold value of |log2 (fold change)|>2 and P-value < 0.01. The six paired samples group are D1 versus (vs) D4, D1 vs D5IMS, D1 vs DIIS, D4 vs D5IMS, D4 vs D5IIS, D5IMS vs DIIS. The differentially expressed miRNAs were collected for further analysis. The overlap of differentially expressed miRNAs among groups was obtained through the Venn map making website (http://bioinformatics.psb.ugent.be/webtools/Venn/).
 
Function analysis of miRNAs
 
The target genes of miRNA were predicted based on miRWalk2.0 (Wang et al., 2010), miRanda (http://www.microrna.org/microrna/getGeneForm.do) and Targetscan database (http://www.targetscan.org). The target genes were subjected to KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway analysis by using the cluster profiler package in R (Yu et al., 2012). Finally, the adjusted P-value<0.01 was set as the cutoff value.
 
Cell culture and transfection
 
HEC-1-A cells were purchased from Cell Bank of the Chinese Academy of Sciences in Shanghai. HEC-1-A was cultured in McCoy’s5A medium (Invitrogen) supplemented with 10% fetal bovine serum (Gibco) and 1% penicillin/streptomycin solution (Gibco). HEC-1-A cells were seeded in 6-well plates (1×10⁶ cells per well) and transfected with 75 pmol (Promega) of miR-183-5p mimics with Lipofectamine 2000 reagent (Invitrogen). NC mimics (75 pmol) were transfected with Lipofectamine2000 reagent as control. Three repeats per group. MiR-183-5p mimics and NC mimics sequences are provided in Table 1. Cells were harvested 36h after transfection and cell lysates were used for parametric transcriptome analysis.


 
Library preparation and transcriptome sequencing
 
Sequencing libraries were generated using the TruSeq RNA Sample Preparation Kit (Illumina). After adenylation of the 32  ends of the DNA fragments, Illumina PE adapter oligonucleotides were ligated to prepare for hybridization. The sequencing library was then sequenced on a Hiseq platform (Illumina). The sequencing data have been deposited in the NCBI SRA database (BioProject number: PRJNA563052).
 
Statistical analysis
 
The tool used for the differential expression analysis was DESeq, which is an R package to analyze count data from high-throughput sequencing assays such as RNA-Seq and test for differential expression. The data were presented as the mean ± SEM from multiple samples and each experiment was conducted at least three times. In the statistical analysis, two-tailed, unpaired Student’s t-test was used in the analysis of the data with the significance of P<0.01.

Classification and annotation of ncRNA
 
After sequencing, we obtained 130.6 million ncRNA reads including 33.0 million in D1 group, 32.9 million in D4 group, 33.0 million in D5IIS group and 31.7 million in D5IMS group. More than 50% of ncRNAs were classified and annotated as miRNAs.
 
Differentially expressed miRNA identification
 
In 12 samples, the expressions of 614 miRNAs were detected. Referring to the sequencing results, we got the frequency and number of miRNA in each group by the method of the Venn diagram (Fig 2). We screened 11 miRNAs with significant differences during embryo implantation among D1, D4, D5IMS and D5IIS. The 11 miRNAs included seven down-regulated miRNAs (miR-192-5p, miR-375-3p, miR-183-5p, miR-182-5p, miR-135b-5p, miR-21a-5p and miR-210-3p) and four up-regulated miRNAs (miR-335-3p, miR-199b-5p, miR-126a-3p and miR-218-5p).
 


Function of differentially expressed miRNAs
 
KEGG enrichment analysis of the 11 differential regulated miRNAs showed that most miRNAs were closely related to pathways in binding and signaling pathways such as focal adhesion, tight junction and Wnt signaling pathway (Fig 3). In organismal systems pathways, immune system and endocrine system were significantly regulated. In human diseases pathways, cancers, endocrine and infectious diseases were dramatically affected. In particular, miR-183-5p was significantly associated with key pathways during embryo implantation.


 
Pathway analysis of miR-183-5p
 
Then we focused on the expression of miR-183-5p. The expression of miR-183-5p at the implantation site was lower than that at the inter-implantation site, indicating that the decrease of miR-183-5p expression was more concentrated at the implantation site. Through GO enrichment analysis, we found that miR-183-5p was mainly involved in the biological process, cellular components and molecular function (Fig 4). In detail, the target genes were enriched in the binding processes such as protein binding, protein domain specific binding and macromolecular complex binding.



Transcriptome sequencing analysis

Through transcriptome sequencing analysis, we found the regulatory network of miR-183-5p in HEC-1-A cells. With [(fold change)]> 2 and P-value < 0.01, sequencing results showed that there were 6 up-regulated genes and 8 down-regulated genes in the miR-183-5p mimics group compared with the NC group (Table 2). We compared the expression consistency of different genes in the miR-183 group and NC group and the results of the thermogram showed that the same gene had good uniformity in the same group (Fig 5).





Comparing the data on D4 (before implantation) with the data on gestation D5 (after implantation), it can be concluded that which miRNAs are most involved in the regulation of the implantation window period of the maternal uterus when receiving embryos. Through screening, we found that 18 of these miRNAs were up-regulated and 18 down-regulated. MiR-182-5p, miR-183-5p and miR-375-3p were the three most significant down-regulated miRNAs. The latest published article identified differentially expressed miRNAs among blastocysts, non-outgrowth and outgrowth embryos in mice. Ten miRNA candidates were identified as significantly differentially expressed miRNAs of outgrowth embryos by in silicoanalysis (Kim et al., 2019).Another recent literature shows that miR-183 can promote embryo implantation by targeting CTNNA2 (Akbar et al., 2019). MiR-183-5p was shown to be significantly differentially expressed both in their study and ours. This suggests that miR-183-5p plays an important role in both maternal and embryonic regulation.

A comparison between implantation site and inter-implantation site can best reflect which miRNAs are involved in the regulation of embryo implantation. By comparing D5IIS with D5IMS, we found 23 up-regulated miRNAs and 22 down-regulated miRNAs. There was one miRNA microarray study to examine the differential expression of miRNAs in the mouse uterus between implantation sites and inter-implantation sites. Compared with inter-implantation sites, there were 13miRNAs up-regulated at least 2-fold and 2 down-regulated at least 2-fold at implantation sites (Hu et al., 2008). Let-7g was shown to be significantly up-regulated both in their research and ours. As hypothesized, the number of miRNAs differentially expressed between D5IIS and tissues on D1 was the lowest (n=3). We speculate that the main reason is that the embryo implantation does not occur in this area and the embryo implantation has been completed. Therefore the inter-implantation site has returned to the normal state of pregnant uterine tissue. This argument is supported by the mechanism of embryo implantation (Fukui et al., 2019).

However, when screening differentially expressed miRNAs for further quantitative research, we synthesized the differences between each group and then selected the most promising twelve miRNAs by referring to the actual reads number. GO analysis revealed that miR-183-5p, miR-182-5p, miR-199b-5p and miR-218-5p were involved in regulating many pathways related to embryo implantation. Recent studies have confirmed that miR-183-5p can regulate the PI3K singling pathway (Meng and Zhang, 2019) and Wnt/beta-catenin signaling pathway (Leung et al., 2015) while miR-182-5p can regulate Rap1/MAPK pathway (Pan et al., 2018). These pathways also play an important role in embryo implantation (Nayeem et al., 2016). Moreover, miR-183-5p and miR-182-5p belong to the same miRNA cluster. MiRNAs cluster is a common phenomenon in animals (Hertel et al., 2006). In humans, xenopus laevis and amphioxus, nearly 50% of miRNAs are found to exist in the form of gene clusters (Tang and Maxwell, 2007). Many studies have shown that the miRNA gene cluster plays an important role in the network system of gene expression regulation in cells, which is more complex and more efficient than the single miRNA regulation mode (Friggi-Grelin et al., 2008). MiR-183-5p, miR-182-5p and miR-96-5p are the three members of the miR-183 family. Sequencing results showed that they had the same down-regulated trend during the process of embryo implantation and the expression level at implantation sites was lower than that at inter-implantation sites. In recent years, many studies have shown that the abnormal expression of miR-183-5p is presumed to be associated with proliferation, apoptosis, migration and invasion of cells. At the same time, in the process of embryo implantation, cell proliferation, apoptosis, migration and invasion are important for successful embryo implantation. This indicates that the miR-183 family may have a more complex and comprehensive regulatory function in the network system of gene expression regulation during embryo implantation.

All differentially expressed miRNAs in early pregnancy (D1, D4, D5IMS and D5IIS) were identified. We screened 11 miRNAs with the widest diversity. Pathway analysis found that the function of the 11 differentially expressed miRNAs were closely related to the pathways associated with binding and signaling molecules, such as focal adhesion, tight junction and Wnt signaling pathways. MiR-183-5p was transfected into HEC cell line and 14 genes were confirmed to be influenced by miR-183 at cell level. This work laid a foundation for the study of miRNA in early pregnancy.

This research was supported by grants from Department of Science Technology of Tongxiang City (201901002), Zhejiang Team Science and Technology Commissioner Project and Leading Innovation Team Project of South Taihu Elite Program.