MicroRNA Dysregulation and Steroid Hormone Receptor Expression in Uterine Tissues of Rats with Endometriosis during the Implantation Window

Endometriosis is a common gynecological disease and is characterized by ectopic growth of functional endometrial glands and stroma outside the uterine cavity.[ 1 ] Endometriosis is estrogen dependent and is benign hyperplasia of the ectopic endometrium.[ 2 3 ] About 25–50% of women with infertility suffer from endometriosis and 30–50% of women with endometriosis show infertility.[ 4 ] The incidence of endometriosis has been estimated as 2.37–2.49/1000 person per year.[ 5 ] The strong associations between endometriosis, infertility, and impaired daily activities, career, and sexual life have a significant economic burden estimated to US 1.8 billion dollars in Canada[ 6 ] and US 18–22 billion dollars in the USA.[ 7 ] While it is acknowledged that an association exists between endometriosis and infertility, little is known about how endometriosis causes fertility. Endometriosis can be caused by various factors including hormonal status, genetic susceptibility, and environmental factors.[ 8 ] The working theory is that endometrial glandular epithelium and mesenchymal cells grow back to the ovaries and pelvic peritoneum along with menstrual blood flow during menstrual period.[ 1 9 ] Successful pregnancy is achieved by appropriate implantation, which is dependent on endometrial receptivity and embryo functionality.[ 10 ] In a regular menstrual cycle, successful implantation only occurs during a short limiting period from day 20 to day 24,[ 11 ] i.e., 7–10 days after ovulation, termed as window of implantation (WOI). For rats, the WOI is generally from 4 to 6 days of pregnancy.[ 12 ] Disorder of WOI is a major cause for female infertility. Dysregulation of molecular and cellular biomarkers related to endometrial receptivity may contribute to the correlation observed between infertility and endometriosis.[ 10 13 ] MicroRNAs (miRNAs) are noncoding RNAs of 18–24 nt. miRNAs can recognize and combine to the 3' untranslated region (3'UTR) of its target gene, inducing degradation of the mRNA and blocking translation.[ 14 ] miRNAs participate in the process of embryo implantation and gene regulation.[ 15 ] miRNAs in the endometrium affects the occurrence and development of endometriosis,[ 13 ] but most of the existing studies focused on a few specific miRNAs, whereas few studies have interrogated the expression profile changes of miRNA in the endometrium with endometriosis during the implantation window. The estrogen receptors (ERs) and progesterone receptor (PR) are involved in the pathogenesis of endometriosis.[ 16 ] Estrogen binds to the ER present on the endometrial glandular epithelial cells and stromal cells to enhance endometrial growth.[ 17 ] Studies have shown that ectopic endometrium contains ER and PR.[ 16 ] PR is an important factor in pregnancy decidual formation, and its expression during the implantation window is related to the activity of decidual-specific genes.[ 18 ] We hypothesized that there is a high correlation between the changes in miRNA expression profile and the changes in ER and PR expression in endometriosis versus no endometriosis during the implantation stage. Therefore, the aim of the present study was to compare the expression profiles of some fertility-related proteins (such as ER and PR) and miRNA in the endometrium with and without endometriosis during the implantation window using Sprague-Dawley rats as an experimental animal model. Due to high incidence of endometriosis-related infertility, it is of importance to investigate endometriosis during the implantation window. The results could provide new insights to understand endometriosis-related infertility and offer basis for improving pregnancy for patients with endometriosis.

All animal protocols were approved by the Animal Care and Use Committee of the Peking Union Medical College Hospital (Permit Number: XHDW-2016-000). Pathogen-free, sexually mature, Sprague-Dawley rats were purchased from Beijing Vital River Laboratory Animal Technology Co. Ltd. (China). The rats included male rats (380 ± 10 g) and nonpregnant female rats (60–70 days old, 200 ± 10 g). The rats were kept at the Laboratory Animal Center of Peking Union Medical College Hospital in cages on bedding (light cycle 14/10 h light/dark) with adequate access to sterilized water and food at 20°C–25°C and relative humidity of 45–60%. The rats were housed five/cage and were adaptively fed for 2–3 days. The estrus cycles of female rats were observed using the vaginal smear method to be about 4–5 days. Twenty female rats were randomized in three groups: endometriosis ( n = 7), fat tissue control ( n = 6), and normal ( n = 7) groups. Rats in the endometriosis group were observed for three consecutive estrus cycles. On the fourth cycle, a piece of uterine tissue was obtained and transplanted to the peritoneal wall using the autotransplantation method described by Vernon and Wilson[ 19 ] but with slight modifications. The rats were anesthetized with 5% pentobarbital sodium (50 mg/kg) intraperitoneally and were placed in a supine position. After disinfection, a ventral midline incision of about 5 cm was made on the abdomen to open the abdomen cavity. The uterus was isolated, and a piece of uterine tissue of 1 cm was taken from the left horn. The piece was immediately washed with physiologic saline. The excess fat tissue outside the uterine serosa was removed, and the uterine cavity was longitudinally incised and trimmed into two pieces of 5 mm × 5 mm. Those pieces were stitched to the two sides of the abdominal wall near the area enriched with blood vessels. The abdominal wall was closed layer by layer. The rats in the fat tissue control group underwent the same surgery as in the endometriosis group, except that the left uterus was ligatured without taking any tissue. Two pieces of adipose tissue taken from the abdomen were stitched at the same location than the uterine tissue in the endometriosis group. For the normal control group, no surgery was performed. On the 28 th day after surgery, the abdomen of the rats was open to observe the growth, invasion, and adhesion of the endometriosis tissue, followed by layer-wise suturing. The transplanted intima was evaluated following the scoring method based on a previous study on rat endometriosis model:[ 20 ] 0, no epithelium; 1, poorly preserved epithelium; 2, moderately preserved epithelium; and 3, well-preserved epithelium. Ten male rats were housed in single-spaced cages individually. The female rats of the three groups were caged at the ratio of 1:1 at 18 o' clock every night, and trays were placed at the bottom of the cage. At 8:00–9:00 on the next morning, the trays were checked for vaginal plug. Each female rat after mating was subjected to vaginal smear for the presence of sperm. If both a vaginal plug was found in the tray and sperm was found, the rat was considered pregnant and the day was considered as the 1 st day of pregnancy (D1). Pregnant rats were no longer subjected to mating. The rats were sacrificed on day 5 of pregnancy (D5; implantation occurs on day 5 in rats),[ 12 ] and uterine tissues were obtained and divided into two pieces. One piece was placed in 10% formalin at 4°C for hematoxylin-eosin staining and immunohistochemistry. The other piece was rapidly placed into liquid nitrogen at −80°C for gene chip and quantitative polymerase chain reaction (qPCR). Uterine tissues were processed routinely for immunohistochemistry. The samples were incubated with ER (1:2000 Beijing Zhongshan Jinqiao Biotechnology Co., Ltd., Beijing, China) and PR (1:10 Beijing Zhongshan Jinqiao Biotechnology Co., Ltd., Beijing, China) for 1–2 h at room temperature. The sample were observed using microscopy. Immunohistochemistry was evaluated as follows.[ 21 ] Based on the positively stained cells, cells were divided as weakly positive (+), score 1; moderately positive (++), score 2; and strongly positive (+++), score 3. According to the number of positive cells, cells were divided as weakly positive (+, the total number of positive cells is 50%). The degree of positivity was evaluated using the following formula: (+)% × 1+ (++)% × 2+ (+++)% × 3. The result of 1.5 was denoted as +++. The samples were randomly observed at 10×. Total RNA was extracted from uterine tissues using the mirVana miRNA Isolation Kit (AM1561, Thermo Fisher Scientific, Waltham, MA, USA), followed by purification and quantification using the Qubit system (Thermo Fisher Scientific, Waltham, MA, USA). An Agilent 2100 (Agilent Technologies, Santa Clara, CA, USA) was used to check the RNA integrity. The purified RNA was subjected to Affymetrix miRNA expression profile chip, and the miRNA expression profile was analyzed by Bo Ao Bio-Tech (Beijing, China). The Affymetrix miRNA 4.0 chip was applied according to the manufacturer's instructions. The Nanodrop instrument (Affymetrix, Sacramento, CA, USA) was used for quality detection of miRNA extracted from endometrial tissue. Good-quality miRNA from six samples (three from the endometriosis group and three from the normal control group) was subjected to expression profile chip detection. Fluorescence qPCR was used to validate the expression of miRNAs, with RNA U6 as an internal reference. The conditions of PCR amplification were as follows: 95°C for 15 s; 60°C for 30 s, 40 cycles; 75°C–95°C. All reactions were performed in triplicates. The primers are listed in Table 1 . The CT (ΔΔCT) method was used to calculate the fold changes of miRNA expression. Primers for the five selected miRNAs MiRNA: MicroRNA; RT: Reverse transcription; AS: Antisense strand. The Affymetrix ® GeneChip ® Command Console ® software was used to scan the chips, and the images were saved to DAT file for analysis. The DAT file was converted to CEL file, which were preprocessed using the RMA algorithm for background correction, integration of the probe signal to probeset signal, and interwafer normalization removing intersample variation caused by nonbiological factors. Following this, the preprocessed files were subjected to correlation analysis including cluster analysis and principal component analysis to interrogate the similarity and/or difference among various groups. Three or more biological replicates were provided to analyze differentially expressed genes using the SAM (significance analysis of microarray) R package (Beijing, China). Genes were considered differentially expressed when the q -value was ≤5% and the fold change was ≥1.5 or ≤0.5. The target genes of the differentially expressed miRNA were predicted using miRWalk2.0 (PubMed). The function of the predicted target genes was analyzed using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG). All data were obtained in triplicate and the results were expressed as mean ± standard deviation (SD). Comparison between two groups was performed using the two-tailed Student's t -test. Comparison among three groups was performed using analysis of variance with post hoc Tukey's test. Two-sided P < 0.05 was considered statistically significant. The statistical analyses were performed using SPSS version 19.0 (SPSS Inc., Chicago, IL, USA).

All rats were alive 28 days after surgery. Three rats were randomly selected from each of endometriosis and fat tissue control groups and were subjected to abdominal cavity opening to observe macroscopically the growth of the implanted tissues. No changes were observed in the fat tissues in the fat tissue control group, whereas the ectopic uterine tissues in the endometriosis group were increased in volume, vascularity, and cystic changes, all three features being 2–3 more important than before surgery [ Figure 1 ]. Growth of the implanted tissues in the fat tissue control group and endometriosis group. The uterine lining, the myometrium (including the subserosal, vascular, and submucosal layers), and the inner lining of the uterus were observed under light microscopy. Figure 2 shows that the vascular layer of the endometriosis group (averagely 10 mm) was significantly thinner than those of the fat tissue control group (averagely 18 mm) and the normal control group (averagely 18 mm). Atrophy was even observed in the endometriosis group. As shown in Figure 2 , in the endometriosis group, the number of endometrial glands was decreased (≤5 in low-magnification fields) and the diameter of the glands (maximum of 5 mm) was reduced. Some glands were hyperplastic (round) with poor body secretion, decidual dysplasia, endometrial epithelial and interstitial disarrangement, and obvious tissue edema. In the fat tissue control group, compared with the endometriosis group, the diameter of endometrial glands (maximum of 10 mm) was larger and the number of endometrial gland was higher (>10 in low-magnification fields). The glandular structure was curved with strong secretion, good decidual growth, and abundant blood vessels. In the normal control group, the diameter of the endometrial gland was the largest (maximum of 10 mm). The glandular structure showed good secretion, good decidua growth, and abundant blood vessels. Hematoxylin-eosin staining of the three groups of uterine tissue in rats (original magnification, ×100). 1: Blood vessel; 2:Decidua; 3: Gland. Taken together, these results indicated that the implantation of endometrial tissues in the abdomen of rats led to a successful model of endometriosis, as shown by changes in the uterine cavity. The control groups showed normal uterine cavities. The expression of ER in glandular epithelium was significantly different between the endometriosis and normal control groups (10.95 ± 0.30 vs. 7.82 ± 0.52, t = 4.125, P < 0.001) and between the fat tissue control and normal control groups (10.05 ± 0.45 vs. 7.82 ± 0.52, t = 2.528, P = 0.011), but not between the endometriosis and fat tissue control groups (10.95 ± 0.30 vs. 10.05 ± 0.45, t = 1.756, P = 0.079; Figure 3a and 3b ). The expression of ER in the endometrial stroma was different between the endometriosis and fat tissue control groups (7.58 ± 0.31 vs. 4.33 ± 0.66, t = 3.627, P < 0.001) and between the endometriosis and normal control groups (7.58 ± 0.31 vs. 3.05 ± 0.33, t = 4.816, P < 0.001), but no significant difference was observed between the fat tissue control and normal control groups (4.33 ± 0.66 vs. 3.05 ± 0.33, P = 0.189; Figure 3a and 3b ). Immunohistochemistry of uterine tissue in rat. (a) Expression of ER and PR in the endometriosis group, fat tissue control group, and the normal control group. 1: Glandular epithelium; 2: Endometrial stroma; 3: Uterine epithelium. The arrows represent positive staining in the nucleus (×400). (b and c) Quantification of the immunochemistry results. *P < 0.05, compared with endometriosis group using analysis of variance with Tukey's test for multiple comparisons. ER: Estrogen receptor; PR: Progesterone receptor. The expression of the PR in the uterine epithelium was different between the endometriosis and fat tissue control groups (2.52 ± 0.52 vs. 5.5 ± 1.28, t = 2.353, P = 0.019) and between the endometriosis and normal control groups (2.52 ± 0.52 vs. 7.4 ± 0.93, t = 3.539, P < 0.001), but not between the fat tissue control and normal control groups (5.5 ± 1.28 vs. 7.4 ± 0.93, t = 0.970, P = 0.332). The expression of the PR in endometrial stroma was different between the endometriosis and fat tissue control groups (0.78 ± 0.22 vs. 2.30 ± 0.74, t = 2.911, P = 0.004) and between the endometriosis and normal control groups (0.78 ± 0.22 vs. 4.66 ± 0.42, t = 4.847, P < 0.001), but there was no difference between the fat tissue control and normal control groups (2.30 ± 0.74 vs. 4.66 ± 0.42, t = 0.937, P = 0.349; Figure 3a and 3c ). Taken together, these results indicated that endometriosis affected the expression of ER and PR in uterine tissues. The microarray results showed that 475 miRNA transcripts were differentially expressed between the endometriosis and normal control groups, with 127 being upregulated and 348 being downregulated. Table 2 lists the top 20 downregulated miRNAs and top 20 upregulated miRNAs [the heat map is shown in Figure 4a ]. In the Circos plot analysis, the downregulated miRNAs were more significant than the upregulated ones [ Figure 4b ], and five downregulated miRNAs with a fold change of ≥2 and P ≤ 0.050 were selected for PCR validation, based on previous studies and clinical implication. These results indicated that endometriosis affected miRNA expression in the uterine tissues. Top 20 downregulated and 20 upregulated miRNAs in rat uterine tissues in the endometriosis group compared with paired normal control group MiRNA: MicroRNA; FC: Fold changes. (a) Heat map showing the differentially expressed microRNA between the endometriosis and control groups. Each row represents a microRNA; each column represents a sample. Yellow indicates upregulation; blue indicates downregulation; 2.0 and −2.0 indicate fold change in the corresponding expression profile; 1–3 indicate sample name; EM indicates endometriosis group; and NC indicates normal control group. (b) Circos plot showing the degree of difference in the differential genes based on their location. Green indicates downregulated genes. The length of the bar represents the multiple of the differentially expressed genes. The longer the bar, the greater the difference multiple. As shown in Figure 5a , the most common pathways involved by the dysregulated miRNAs were cancer, vasopressin-regulated water reabsorption, morphine addiction, and prostate cancer. Interestingly, the estrogen signaling pathway was also one of the common pathways involved by the dysregulated miRNAs. The GO pathway analysis showed that the dysregulated miRNAs were also involved in metabolic process, development process, cell death, rhythmic process, biological regulation, cellular process, and growth [ Figure 5b ]. These results indicated that the miRNAs differentially regulated in endometriosis were involved in a number of pathways, some of them potentially playing roles in implantation and fertility. (a) KEGG signaling pathway analysis of target genes for differentially expressed microRNAs. The first 30 KEGG signaling pathways and the most significant P value. The X-axis represents the most significant P value; the Y-axis represents the KEGG signaling pathway. (b) GO-standard analyzes biological process differentially expressed microRNAs. GO: Gene Ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes. The target genes of five dysregulated miRNAs (rno-miR-29c-3p, rno-miR-34c-5p, rno-miR-141-5p, rno-miR-24-1-5p, and rno-miR-490-5p) were predicted using miRWalk2.0. The target genes were screened using six types of miRNA target genes prediction procedure: miRWalk, miRanda, miRDB, miRMap, miRNAMap, and RNAhybrid, followed by enrichment analysis using GO and KEGG pathways. Results are summarized in Supplemental Table 1 . Target genes prediction Due to the short sequence length of miRNAs, the prediction accuracy was limited, and the identified target genes were involved in a broad range of pathways including embryo implantation, blastocyst formation, decidualization, luteinizing hormone signaling pathway, sperm-egg recognition, estradiol 17-beta-dehydrogenase, estrogen signaling pathway, ER binding, steroid biosynthesis, steroid hormone receptor binding activity, and steroid dehydrogenase activity [ Supplemental Table 2 ]. Nevertheless, these results showed that these five miRNAs were potentially involved in pathways that were part of the reproductive functions. Biological processes of fertility ER: Estrogen receptor. Rno-miR-29c-3p expression was different between the endometriosis and fat tissue control groups (0.23430 ± 0.03000 vs. 0.31670 ± 0.02000, P = 0.022) and between the endometriosis and normal control groups (0.23430 ± 0.03000 vs. 0.40590 ± 0.05000, P = 0.001), without significant difference between the fat tissue control and normal control groups (0.31670 ± 0.02000 vs. 0.40590 ± 0.05000, P = 0.085). Rno-miR-24-1-5p expression was different between the endometriosis and fat tissue control groups (0.00030 ± 0.00004 vs. 0.00050 ± 0.00003, P = 0.008) and between the endometriosis and normal control groups (0.00030 ± 0.00004 vs. 0.00060 ± 0.00008, P = 0.001), but not between the fat tissue control and normal control groups (0.00050 ± 0.00003 vs. 0.00060 ± 0.00008, P = 0.101). Rno-miR-490-5p expression was different between the endometriosis and fat tissue control groups (0.00080 ± 0.00011 vs. 0.00130 ± 0.00016, P = 0.028) and between the endometriosis and normal control groups (0.00080 ± 0.00011 vs. 0.00130 ± 0.00004, P = 0.001), but not between the fat tissue control and normal control groups (0.00130 ± 0.00016 vs. 0.00130 ± 0.00004, P = 0.366). The expression of rno-miR-34c-5p was different between the endometriosis and fat tissue control groups (0.00180 ± 0.00020 vs. 0.00280 ± 0.00026, P = 0.015), between the endometriosis and normal control groups (0.00180 ± 0.00020 vs. 0.00360 ± 0.00021, P = 0.001), and between the fat tissue control and normal control groups (0.00280 ± 0.00026 vs. 0.00360 ± 0.00021, P = 0.022). Finally, the expression of rno-miR-141-5p was different between the endometriosis and fat tissue control groups (0.08140 ± 0.01164 vs. 0.12670 ± 0.01542, P = 0.011), between the endometriosis and normal control groups (0.08140 ± 0.01164 vs. 0.19240 ± 0.00422, P = 0.002), and between the fat tissue control and normal control groups (0.12670 ± 0.01542 vs. 0.19240 ± 0.00422, P = 0.032; Figure 6 ). Taken together, these results suggested that these five miRNAs were involved in endometriosis. Relative expression of rno-miR-29c-3p, rno-miR-24-1-5p, rno-miR-141-5p, rno-miR-490-5p, and rno-miR-34c-5p among the three groups, according to the ΔΔCT method. * P < 0.05, versus endometriosis group. The associations between the expression of the five miRNAs and the expression of ER (ERα and ERβ) and PR were studied using miRWalk 3.0 and TargetScan. Table 3 shows that the five miRNAs had binding sites on the coding DNA sequence and 3'UTR of ERα, ERβ, and PR. Relation between the five selected miRNAs (rno-miR-29c-3p, rno-miR-24-1-5p, rno-miR-141-5p, rno-miR-490-5p, and rno-miR-34c-5p) and steroid hormone receptors predicted by miRWalk 3.0 ER: Estrogen receptor; PR: Progesterone receptor; CDS: Coding DNA sequence; UTR: Untranslated region; miRNA: MicroRNA.

ER and PR are involved in endometriosis,[ 2 8 16 18 22 ] but the involvement of miRNAs is unknown. Therefore, this study aimed to explore the correlation between miRNA and ER/PR in the uterine tissues of rats with endometriosis during the implantation window. The results suggested that miRNAs, ER, and PR could play important roles in the implantation period of rats with endometriosis. These miRNAs might play a role in endometrial receptivity in endometriosis. The rat model of autologous endometrial implantation is considered to be representative of human endometriosis.[ 19 ] The implantation failure observed in endometriosis is possibly related to the upregulation of ER and downregulation of PR, and the high expression of ER might be closely connected with the inflammatory characteristics of endometriosis.[ 8 18 ] The present study suggested that multiple differentially expressed miRNAs were detected in the uterine tissue of rats' endometriosis during the implantation window. Changes in the expression of particular miRNAs might play a role in the modulation of endometrial receptivity in endometriosis, ultimately resulting in implantation or failure.[ 23 ] Since Buyalos and Agarwal[ 24 ] first proposed their concept of endometriosis-related infertility in 2000, the features of endometriosis infertility gathered great attention from medical workers and researchers. Thus, the pregnancy outcomes of patients with endometriosis obviously improved, mainly with the help of preparative drug regimens and assisted reproduction techniques. Nevertheless, the incidence of endometriosis and endometriosis-related infertility did not really change, mainly because the pathogenesis of endometriosis remains poorly understood and because of the complex process of reproduction, on which endometriosis may affect multiple stages. The present study focused on the embryo implantation and endometrial receptivity, more particularly the miRNAs that could be involved. In the present study, the Gene Chip technology was used to detect the miRNA array in the uterus of rats with endometriosis during the WOI. We found that 475 miRNA transcripts were differently expressed between the endometriosis and normal control groups, with 127 being upregulated and 348 being downregulated. Among the dysregulated miRNAs, downregulated genes were more significant than the upregulated. We selected five obviously downregulated miRNAs. According to miRNA target gene prediction tools (miRWalk, miRanda, miRDB, miRMap, miRNAMap, and RNAhybrid), we predicted the target genes of rno-miR-29c-3p, rno-miR-34c-5p, rno-miR-141-5p, rno-miR-24-1-5p, and rno-miR-490-5p. MiR-29c-3p is involved in the regulation of the vascular endothelial growth factor (VEGF)-A mRNA translation in endometrial cells, inhibition of VEGF-A protein expression, and angiogenesis.[ 25 ] The present study showed that the expression of miR-29c was increased in eutopic endometrium of baboons and humans with endometriosis, where it might damage the progesterone response by reducing the levels of FKBP4.[ 26 ] Removal of the deep infiltrative endometriosis could reverse the progesterone resistance in women with endometriosis.[ 26 ] MiR-29c might inhibit endometriosis by inhibiting endometrial cell proliferation and invasion and by promoting apoptosis. Jeon et al .[ 27 ] showed that the dysregulated expression of miR-24 might be associated with infertility caused by polycystic ovary syndrome. Indeed, miR-24 inhibits the expression of the MYC gene, which is a key factor of multiplication involved in embryo implantation.[ 27 ] In the present study, rno-miR-24-1-5p was downregulated in the uterus of rats with endometriosis, supporting this role of miR-24. A study of miR-24 transfection showed that the expression of CDKN1b, which is a target gene of miR-24, was suppressed in the process of embryonic development.[ 28 ] MiR-24 might thus regulate the concentration of estradiol and progesterone during follicular fluid formation.[ 28 ] The excessive expression of miR-24 might restrain transforming growth factor signaling pathways and inhibit the secretion of estradiol.[ 28 ] MiRNA-24 is downregulated in the development of porcine embryos; because miR-24 expression was influenced by the culture conditions, miRNA-24 could be used as a quality index for embryos.[ 29 ] Another study showed that the expression of miR-24 was upregulated in the plasma of patients with severe preeclampsia during the gestational period.[ 30 ] MiR-490-3p may increase CDDP sensitivity in ovarian cancer cells by reducing the expression of ABCC2, indicating that the increase of miR-490-3p might be a potential treatment strategy against cisplatin-resistant ovarian cancer.[ 31 ] In addition, miR-490-3p mRNA expression was lower in endometrial carcinoma tissue than in normal endometrial tissue.[ 31 ] MiR-490-3p expression negatively regulates the infiltration depth and lymph node metastasis in endometrial cancer.[ 32 ] MiR-490 expression was significantly lower in ovarian cancer and borderline tumors than that of benign tumor.[ 33 ] Its expression in metastatic ovarian cancer (omentum) was lower than that of primary ovarian cancer.[ 33 ] It is also negatively associated with FIGO staging and differentiation.[ 33 ] MiR-490-3p may inhibit tumor occurrence and development of ovarian epithelial carcinoma by regulating its target gene CDK1.[ 33 ] MiR-490-3p might be associated with the occurrence of drug resistance ovarian cancer. MiR-490 is also obviously abnormally expressed in uterine leiomyoma.[ 34 ] In the present study, rno-miR-490-3p was downregulated in the uterus of rats with endometriosis. In maternal plasma, the upregulation of miR-34c expression was directly related to good-quality embryos in D3.[ 35 ] There was a strong correlation between miR-34c expression level in sperm and the results of intracytoplasmic sperm injection (ICSI) treatment in cattle.[ 35 ] MiR-34c-positive patients are more likely to have good-quality embryo, implantation, pregnancy, and live birth.[ 35 ] Paternal miR-34c levels may play a role in embryonic development.[ 35 ] MiR-34c levels might be used as quality index of ICSI outcome in the human sperm.[ 35 ] MiR-34c participates in the regulation of follicular and two-cell stages of embryonic development in bovine.[ 36 ] MiR-34c is involved in cervical reconstruction after spontaneous term labor and delivery.[ 37 ] MiR-34c is important for the first cellular division in the early embryo development.[ 38 ] Upregulation of miR-34c in breast tumor is associated to downregulation of CYP 1A1 and ERα gene expression.[ 39 ] A study in mice found that miRNA-141 downregulated the expression of PTEN (phosphatase and tensin homolog deleted on chromosome ten) and might affect cell proliferation and apoptosis of endometrial cells, playing an important role in embryo implantation.[ 15 ] A previous study showed that miR-34c and miR-141 were specific to the epithelial cells and they were highly expressed in abdominal endometriosis tissue.[ 40 ] MiR-141 was upregulated in the placenta of patients with preeclampsia, and miR-141 inhibits the trophoblast cells.[ 41 ] MiR-141 was expressed in the placenta and was increased in maternal plasma during pregnancy. MiR-141 was downregulated in endometriosis.[ 23 ] Taken together, these results suggested that rno-miR-29c-3p, rno-miR-34c-5p, rno-miR-141-5p, rno-miR-24-1-5p, and rno-miR-490-5p played a number of roles in the fertility and reproduction processes, including the endometrial receptivity. Nevertheless, additional studies are still necessary to determine these roles exactly. In this study, rno-miR-34c-5p, rno-miR-141-5p, and ER were upregulated in endometriosis. There might be an intimate relationship between miRNA and ER in rat endometriosis-related infertility during the WOI. MiRNA-141 is one of the miRNA-200 family members, and progesterone can reduce miRNA-141 expression to promote stem-like breast cancer cells.[ 42 ] MiRNA-141 may directly regulate PR, which is an important transcription factor for the expansion of mammary gland stem cell.[ 42 ] The decrease of miR-141 can increase the protein levels of PR and the expression of PR is estrogen dependent on a cellular level.[ 42 ] MiR-200a/miR-141 might be related to hormone receptor status in endometrial cancer and may affect prognosis.[ 43 ] MiR-34c regulates the reduction of ERα and influences the formation of breast cancer.[ 39 ] According to GO and KEGG-PATHWAY analysis, miRNAs are involved in many relating to fertility activities that molecular functions and biological processes, for example, response to estrogen, response to hormone, luteinizing hormone signaling pathway, ER overload response, ER-nucleus signaling pathway, response to estradiol, response to steroid hormone, protein targeting to ER, ERBB2 signaling, ER-associated ubiquitin-dependent protein catabolic process, regulation of ER to Golgi vesicle-mediated transport, estrogen metabolic process, luteinization, ER to Golgi vesicle-mediated transport, estrogen biosynthetic process, steroid metabolic process, steroid hormone secretion, steroid biosynthetic process, steroid catabolic process, steroid hormone receptor binding, steroid hormone binding, hormone binding, estradiol 17-beta-dehydrogenase activity, steroid dehydrogenase activity, ER binding, steroid hormone receptor activity, steroid binding, and steroid hydroxylase activity. All these pathways may potentially be involved in the regulation of endometrial receptivity, but the exact effects of the miRNAs identified in the present study need to be verified. The results suggested that dysregulation of miRNAs and abnormal expression of ER, and PR could be observed in rats with endometriosis. These changes might influence the embryo implantation of rats with endometriosis during the implantation window. Supplementary information is linked to the online version of the paper on the Chinese Medical Journal website . This study was supported by a grant from National Nonprofit Industry Scientific Research Fund of China (No. 201502004). There are no conflicts of interest.