Characterization of microRNAs in spent culture medium associated with human embryo quality and development.

BACKGROUND: Given implantation failure limits the improvement of in vitro fertilization (IVF) success rates, there is an urgency to identify potential biomarkers for embryo quality and predict the outcomes of IVF-embryo transfer (IVF-ET).
METHODS: Using RNA-sequencing, we identified the expression profiles of 16 spent culture medium (SCM) collected from embryos at the cleavage on day 3 (D3 cleavage) and blastocyst stages on day 5 (D5 blastocyst) during IVF cycles. Differentially expressed miRNAs (DEmiRNAs) were then identified, and microRNA (miRNA)-messenger RNA (mRNA) interaction networks were constructed. Finally, quantitative real-time polymerase chain reaction (qRT-PCR) confirmation and validation in the Gene Expression Omnibus (GEO) database were performed.
RESULTS: Compared with the pregnant group, 29 DEmiRNAs were detected in the non-pregnant group at D3 cleavage, and 26 were detected in the non-pregnant group at D5 blastocyst. Among them, a total of six known miRNAs, including hsa-miR-199a-3p>hsa-miR-199b-3p, hsa-miR-199a-5p, hsa-miR-379-5p, hsa-miR-432-5p, hsa-miR-99a-5p, and hsa-miR-483-5p, were identified. The results of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis indicated that these target genes of DEmiRNAs were associated with various biological processes (BPs). The results of validation in qRT-PCR and the GEO database suggested the reliability of our RNA-sequencing results.
CONCLUSIONS: In conclusion, we identified three miRNAs, including hsa-miR-199a-5p, hsa-miR-483-5p, and hsa-miR-432-5p, which may serve as biomarkers for embryo quality during IVF cycles. 2021 Annals of Translational Medicine. All rights reserved.

Introduction

In vitro fertilization (IVF) is an effective and powerful assisted reproductive technology (ART), and has been commonly used to treat about 16% of couples who suffer from infertility worldwide (1). ARTs are well developed and have brought solutions to some fertility problems. IVF to produce embryos is advantageous because it bypasses challenges associated with fertilization, and offers an opportunity to better study early embryonic development (2). However, as a common means to measure embryo quality and potential of in vivo and IVF embryos in humans, morphological assessment often fails to indicate whether an embryo can successfully establish a pregnancy (3). Whereas embryos are often of similar morphology, the underlying genetics, such as transcriptomes, can be strikingly different. In addition, implantation failure after embryo transfer (ET) is also an obstacle to ARTs and has fettered the further improvement of the IVF success rate. Thus, a need exists for a better noninvasive method to efficiently select embryos based on their genetic potential and ability to establish a pregnancy.

MicroRNAs (miRNAs) are evolutionarily conserved single-stranded non-coding RNA molecules and important regulators in the post-transcriptional regulation of gene expression in various biological and physiological processes (4). miRNAs can regulate the levels of many target genes at the same time by partially complementary sequences and subsequent interference with messenger RNA (mRNA) stability and/or protein translation (4,5). Since they are stable in the extracellular environment, and their abnormal expressions are associated with physiological or pathological status, it is possible to treat miRNAs as new non-invasive biomarkers (6). It has also been widely reported that miRNAs were involved in embryo implantation, and associated with implantation failure (7,8). Kresowik et al. suggested that miR-31 plays a role in regulating the immune system during implantation (9), while Ioannidis et al. indicated that miR-26 might be a candidate biomarker for very early pregnancy in cows (10). miRNAs have attracted attention in several areas of medicine, including assisted reproduction.

An ideal approach to predict the implantation outcome should not compromise embryo implantation. It was believed that human embryos secrete miRNAs in easily collectable embryonic culture media, making it possible to use them as non-invasive biomarkers for embryo reproductive competence. The development of non-invasive biomarkers has driven extensive research in this area, with an overall aim to improve the success rate of implantation and IVF treatment. Abu-Halima et al. analyzed the miRNAs secreted from pre-implantation embryos into the embryonic culture media and identified that miR-634 was correlated with a positive pregnancy outcome (11). Then, in 2019, Abu-Halima and his colleagues reported that the abundance of miR-19b-3p was significantly lower in spent culture medium (SCM) samples associated with positive pregnancy (12). Fang et al. indicated that hsa-miR-26b-5p and hsa-miR-21-5p could serve as potential biomarkers for reproductive outcomes by detecting the miRNA profiles in the human embryo culture media (13).

In this study, we comprehensively analyzed the expression profile of miRNAs secreted by human pre-implantation embryos into SCM at different stages between a pregnant and non-pregnant group undergoing IVF treatment to investigate potential predictive markers of clinical outcomes of IVF-ET. By doing this, it would have a major impact on biomarker development and clinical practice for reproductive clinicians and scientists. Improved diagnosis of embryo implantation could have a profound effect on psychological and financial well-being on women and couples undergoing IVF treatment. We present the following article in accordance with the MDAR reporting checklist (available at https://dx.doi.org/10.21037/atm-21-5029).

Methods

Embryo evaluation

Embryos at the cleavage on day 3 (D3 cleavage) and blastocyst stages on day 5 (D5 blastocyst) were evaluated using the ALPHA/ESHRE guidelines, and based on the number of blastomeres, rate of fragmentation, and symmetry. Cleavages with eight to 10 blastomeres, fragmentation <10%, or with mild unevenness, were considered of high-quality, and those with better morphology scores were maintained in culture until the blastocyst stage (at the 5th day of culture). Blastocysts that were scored as 4BB, 4BA, 4AB, and 4AA were regarded as good-quality blastocysts.

ET

In the case of a single blastocyst ET, only the most viable embryo was transferred on day 5. Serum or urinary ®-hCG was measured 14 days after ET, and its rise (>20 IU) across two consecutive blood tests or a positive urinary test indicated pregnancy. Clinical pregnancy was confirmed when a gestational sac with fetal heartbeat was revealed by ultrasound examination after 7 weeks of pregnancy.

SCM collection

The research described in the study was approved by the ethics committee of The Affiliated Drum Tower Hospital of Nanjing University Medical School (2019-198-01) and in accordance with the Declaration of Helsinki (as revised in 2013). Written informed consent for use of their samples was provided by every participant. All embryos were obtained after controlled stimulation of the ovaries using gonadotropin-releasing hormone agonists and IVF techniques, and G1-PLUSTM/G2-PLUSTM (Vitrolife, Gothenburg, Sweden) was used for embryo culture. The SCM used for individual culture was collected individually from every embryo at D3 cleavage and D5 blastocyst during IVF cycles and stored at −80 °C until the analysis was performed.

RNA isolation and sequencing

Fifty-µL SCM from eight female/single-ETs, including three individuals in a non-pregnant group and five in a pregnant group were used for miRNA profiling. The media was collected from the same embryos on days 3 and 5. According to the manufacturer’s protocol, total RNA was extracted respectively from 16 SCM using miRNeasy Serum/Plasma Kit (Qiagen, Germany). Table S1 shows the characteristics of all these individuals. Small RNA sequencing was conducted by OE Biotech Co., Ltd. (Shanghai, China), while NEBNext Multiplex Small RNA Library Prep Set for Illumina was used to generate smRNA-seq libraries, with 20M reads per library obtained. The starting amount of input RNA was 1 ng. RNA sequencing was performed based on Illumina X-ten platform and the RNAs were aligned with bowtie and then subjected to the BLAST search against Rfam v.10.1 and GenBank databases. The known miRNAs were determined through aligning against miRBase v.21 database, and unannotated small RNAs were identified with mirdeep2 to predict novel miRNAs.

Identification of differentially expressed miRNAs (DEmiRNAs) and miRNA target prediction

The expression of miRNAs into SCM from the non-pregnant group was compared with those from the pregnant group. With DESeq, DEmiRNAs were determined with P value <0.05 and |log2FC| >1 between the groups at D3 cleavage and D5 blastocyst. The DEG algorithm was used to calculate P value in the R package. The targets of known DEmiRNAs were predicted using Miranda, with the parameter as follows: S ≥150, ΔG ≤−30 kcal/mol and demand strict 5’ seed pairing. Cytoscape soft (http://www.cytoscape.org/) was used to construct miRNA-mRNA interaction networks.

Functional enrichment analysis and protein-protein interaction (PPI) networks construction

CPDB (http://cpdb.molgen.mpg.de/CPDB) was applied to perform Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of target genes of DEmiRNAs. GO enrichment analysis elucidated the functional annotations including biological process (BP), cellular component (CC), and molecular function (MF), and KEGG enrichment analysis was conducted to elucidate the target genes related signaling pathways, with the threshold set as P value <0.05. Online database STRING (https://string-db.org) was employed to analyze the PPI networks.

Quantitative real-time polymerase chain reaction (qRT-PCR) validation

In total, 15 individuals, including eight in the non-pregnant group and seven in the pregnant group, were recruited for qRT-PCR validation, and the media was collected from the same embryos on days 3 and 5. Total RNA extracted from 30 SCM with the miRNeasy Serum/Plasma Kit was used to validate the differential expression of miRNAs between the pregnant group and non-pregnant group by qRT-PCR analysis, with the qRT-PCR reactions performed in ABI 7300 Real-time PCR Detection System. All experiments were performed in three independent replicates. Relative gene expression was analyzed by 2−ΔΔCT method, and human hsa-U6 was used as endogenous controls for miRNA expression in analysis. The PCR primers are displayed in Table S2.

Validation in the Gene Expression Omnibus (GEO) database and receiver operating characteristic (ROC) analysis

GSE93810 was downloaded from the GEO database, which consists of 17 females with a positive pregnancy transfer and 39 females in whom no pregnancy occurred. The expression levels of selected DEmiRNAs were validated with GSE93810. Using pROC package in R language, we then performed the ROC analysis to assess the diagnostic value of DEmiRNAs. The area under the curve (AUC) was then calculated.

Statistical analysis

Statistical analyses of differential miRNAs between the non-pregnant group and the pregnant group were performed using R software. In addition, the GraphPad Prism Software, version 7.0 was used for the statistical analysis of experimental data. The results are expressed as means ± standard deviations (SDs). P<0.05 was considered to indicate a significant difference between the groups.

Results

Identification of DEmiRNAs

Compared with the pregnant group, 29 DEmiRNAs (11 up- and 18 down-regulated DEmiRNAs) were detected in the non-pregnant group at D3 cleavage with P value <0.05 and |log2FC| >1 (, ), and 26 DEmiRNAs (19 up- and 7 down-regulated DEmiRNAs) were detected in the non-pregnant group at D5 blastocyst with P value <0.05 and |log2FC| >1 (, ). In total, six known miRNAs, including hsa-miR-199a-3p>hsa-miR-199b-3p, hsa-miR-199a-5p, hsa-miR-379-5p, hsa-miR-432-5p, hsa-miR-99a-5p, and hsa-miR-483-5p, were identified in this analysis. The raw data have been deposited in the GEO database (GSE167961; https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE167961).

Figure 1

Hierarchical clustering analysis of all DEmiRNAs in SCM between non-pregnant group and pregnant groups. (A) D3 cleavage, (B) D5 blastocyst. DEmiRNAs, differentially expressed miRNAs; SCM, spent culture medium; D3 cleavage, cleavage on day 3; D5 blastocyst, blastocyst stages on day 5.

Table 1

All DEmiRNAs in SCMs of D3 cleavage between non-pregnant group and pregnant groups

miRNA ID	Log2FC	P value	Regulation
hsa-miR-483-5p	Inf	0.049484	Up
novel1009_mature	10.26512873	0.002501	Up
novel1036_mature	−9.295460431	0.000136	Down
novel1054_mature	7.728030255	0.044975	Up
novel106_mature	−6.64869016	0.017121	Down
novel1118_mature	−8.863984044	0.00074	Down
novel120_mature	8.783172155	0.020206	Up
novel1214_mature	−6.935777618	0.047552	Down
novel1238_mature	−6.332283973	0.048518	Down
novel146_mature	−5.16022631	0.000798	Down
novel187_mature	−6.914511993	0.037222	Down
novel19_mature	−4.166158994	0.00061	Down
novel245_mature	Inf	0.022999	Up
novel267_mature	−8.143688147	0.004185	Down
novel273_mature	7.160261611	0.027341	Up
novel3_mature	−8.846116416	0.007268	Down
novel437_mature	−8.865099597	0.042348	Down
novel52_mature>novel303_mature	−9.209703294	0.016312	Down
novel545_mature>novel673_mature	Inf	0.027824	Up
novel552_mature	−7.264898066	0.000599	Down
novel589_mature	−Inf	0.020123	Down
novel708_mature	7.496888895	0.006455	Up
novel76_mature	Inf	0.000883	Up
novel849_mature	−7.616105961	0.001598	Down
novel865_mature	−9.598129639	0.007133	Down
novel898_mature	−7.082397281	0.01306	Down
novel926_mature	Inf	0.027763	Up
novel940_mature>novel986_mature	−7.790471729	0.020836	Down
novel984_mature	Inf	0.043655	Up

DEmiRNAs, differentially expressed miRNAs; SCM, spent culture medium; D3 cleavage, cleavage on day 3; miRNA, microRNA.

Table 2

All DEmiRNAs in SCMs of D5 blastocyst between non-pregnant group and pregnant groups

miRNA ID	Log2FC	P value	Regulation
hsa-miR-199a-3p>hsa-miR-199b-3p	−7.17811	0.026131	Down
hsa-miR-199a-5p	−7.95985	0.038632	Down
hsa-miR-379-5p	−7.50241	0.033988	Down
hsa-miR-432-5p	Inf	0.023522	Up
hsa-miR-99a-5p	−4.93374	0.014704	Down
novel1052_mature	Inf	0.029945	Up
novel11_mature>novel694_mature>novel887_mature	−7.96554	0.049573	Down
novel1250_mature	Inf	0.031455	Up
novel130_mature	9.108542	0.012543	Up
novel160_mature	10.96	0.001951	Up
novel162_mature	Inf	0.029479	Up
novel238_mature	9.675487	0.002768	Up
novel24_mature	7.646165	0.029516	Up
novel292_mature	Inf	0.023266	Up
novel305_mature	Inf	0.034748	Up
novel323_mature>novel581_mature	Inf	0.01271	Up
novel32_mature	8.267443	0.020619	Up
novel387_mature	#NAME?	0.014261	Down
novel495_mature	7.035155	0.038632	Up
novel718_mature	7.509777	0.019616	Up
novel86_mature>novel741_mature	9.724586	0.033219	Up
novel884_mature	−8.42776	0.011719	Down
novel891_mature	11.00996	0.003768	Up
novel90_mature>novel1213_mature	Inf	0.029307	Up
novel989_mature>novel1088_mature	Inf	0.034597	Up
novel99_mature	8.797029	0.032628	Up

DEmiRNAs, differentially expressed miRNAs; SCM, spent culture medium; D5 blastocyst, blastocyst stages on day 5; miRNA, microRNA.

Construction of miRNA-mRNA interaction networks

It has been reported that miRNAs can act by binding mRNA targets to repress their expression, allowing prediction of the targets of known DEmiRNAs. A total of 63 miRNA-mRNA pairs, involving one miRNA and 63 mRNAs, were obtained at D3 cleavage (), and a total of 203 miRNA-mRNA pairs, involving five miRNAs and 203 mRNAs, were obtained at D5 blastocyst (). In addition, we used TargetScan (http://www.targetscan.org/vert_72/) software to further validate the targeted relationship between miRNA and mRNA, such as hsa-miR-432-5p-ASPSCR1/ZNF384/GALNT9 and hsa-miR-483-5p-ZBTB16/HDAC2 (Figure S1).

Figure 2

miRNA-mRNA interaction networks. (A) D3 cleavage, (B) D5 blastocyst. The rectangles and ellipses indicate miRNAs and mRNAs, respectively. Red and green colors represent up- and down-regulation, respectively. miRNA, microRNA; mRNA, messenger RNA; D3 cleavage, cleavage on day 3; D5 blastocyst, blastocyst stages on day 5.

Functional enrichment analysis and PPI networks

The functional enrichment of the targets of known DEmiRNAs was performed to explore the function of the known miRNAs. The significant GO terms and pathways involved with genes targeted by miRNAs enriched in embryos at D3 cleavage and D5 blastocyst are displayed in , respectively, and the results indicate these genes are associated with various BPs, including anatomical structure morphogenesis, cellular developmental process, cell communication, and cell signaling. With the STRING online database and Cytoscape software, the PPI networks of the targets of known DEmiRNAs were constructed, and as shown in Figure S2, the D3-specific PPI network included 27 proteins and the D5-specific PPI network included 92 proteins. Among them, HDAC2 (degree =5), ASPSCR1 (degree =4), ZNF384 (degree =4), GALNT9 (degree =4), and ZBTB16 (degree =3) were several hub proteins.

Figure 3

Functional annotation of target genes of DEmiRNAs at D3 cleavage. (A) GO terms, (B) KEGG pathways. DEmiRNAs, differentially expressed miRNAs; D3 cleavage, cleavage on day 3; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes.

Figure 4

Functional annotation of target genes of DEmiRNAs at D5 blastocyst. (A) GO terms, (B) KEGG pathways. DEmiRNAs, differentially expressed miRNAs; D5 blastocyst, blastocyst stages on day 5; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes.

QRT-PCR confirmation and validation in the GEO database

The expressions of six known miRNAs, including hsa-miR-199a-3p>hsa-miR-199b-3p, hsa-miR-199a-5p, hsa-miR-379-5p, hsa-miR-432-5p, hsa-miR-99a-5p, and hsa-miR-483-5p, were tested by qRT-PCR analysis and GEO database, and the qRT-PCR results indicated that except for hsa-miR-432-5p, the expression of the other miRNAs exhibited the same trend as the results of RNA sequencing (). In addition, hsa-miR-199a-3p>hsa-miR-199b-3p, hsa-miR-199a-5p, hsa-miR-432-5p, hsa-miR-99a-5p, and hsa-miR-483-5p in GSE93810 exhibited the same trend as the results of RNA sequencing, which was statistically significant (). ROC analysis indicated that hsa-miR-99a-5p (0.792) and hsa-miR-199a-5p (0.786) were with relatively high diagnostic value (Figure S3).

Figure 5

qRT-PCR results of known DEmiRNAs. (A) hsa-miR-483-5p, (B) hsa-miR-199a-5p, (C) hsa-miR-199a-3p>hsa-miR-199b-3p, (D) hsa-miR-379-5p, (E) hsa-miR-432-5p, and (F) hsa-miR-99a-5p. qRT-PCR, quantitative real-time polymerase chain reaction; DEmiRNAs, differentially expressed miRNAs.

Figure 6

Validation of known DEmiRNAs in the GEO database. * indicates P<0.05; *** indicates P<0.001. DEmiRNAs, differentially expressed miRNAs; GEO, Gene Expression Omnibus.

Discussion

Given the limitations of the low success rate of IVF caused by implantation failure after IVF-ET treatment, the investigation of biomarkers for embryo quality and predicting the outcomes of IVF-ET could contribute to the development of ARTs (14). Recently, the role of a variety of biomarkers including miRNAs, such as proteomics and hormones in SCM, has been extensively studied. Borges et al. suggested that miR-142-3p may be a potential biomarker of blastocyst implantation failure (15), and increased levels of apolipoprotein A1 were present in SCM containing blastocysts of higher morphologic grade, which may suggest a role for lipoproteins in early embryologic development (16). Chen et al. indicated secreted beta-human chorionic gonadotropin (HCG) as a useful biomarker for embryo selection in IVF-ET procedure (17), while Wang et al. reported that the concentration of HCG in SCM is positively correlated with the status of early embryo development and implantation rate, thus serves as a useful marker for embryo selection in the IVF-ET procedure (18). We analyzed the profiles of miRNAs secreted from pre-implantation embryos into SCM between a pregnant and non-pregnant group undergoing IVF treatment as biomarkers to predict the outcomes of IVF-ET. A total of six known miRNAs were identified in SCM of the pregnant group, and among them, hsa-miR-199a-5p, hsa-miR-483-5p, and hsa-miR-432-5p were three miRNAs with abundant target genes.

In this study, hsa-miR-199a-5p was detected to present an increased trend in SCM of the pregnant group. Considerable evidence has demonstrated that miR-199a-5p was involved in various cancers (19), and mmu-miR-199a-5p was reported to silence the self-renewal of mouse embryonic stem cells (20), and Tan et al. reported that, compared to an in vivo-fertilized group, down-regulated miR-199a-5p in IVF blastocysts was responsible for the lower developmental potential and subsequent viability (19). Li et al. suggested that ssc-miR-199a-5p may play a crucial role for implantation by investigating miRNAome of the endometrium of pregnancy and non-pregnancy groups in pigs (21). Compared with fresh mouse blastocysts, up-regulated mmu-miR-199a-5p was detected in the vitrified blastocysts, suggesting vitrification, a method commonly used in the cryopreservation of mammalian blastocysts, may decrease the implantation potential of vitrified blastocysts (22). In summary, we conclude hsa-miR-199a-5p is closely related to embryo implantation, but the underlying mechanism of hsa-miR-199a-5p in perimplantation embryos requires further exploration.

As over-expressed miR-483-5p suppressed glioma cell proliferation and induced a G0/G1 arrest, it is considered to be a tumor suppressor (23). It was reported that miR-483-5p regulated the progesterone concentration in a steroidogenic human granulosa-like tumor cell line (24). Down-regulated miR-483-5p was observed in cumulus cells of metaphase II oocytes from women with polycystic ovary syndrome (25), and an exploratory study uncovered that dysregulated miR-483-5p was associated with pregnancy-related adverse outcomes (26). It has been reported that miR-483-5p was a human-specific regulator of MeCP2 levels in human fetal brains (27), while decreased miR-483-5p was detected in placentas from patients with preeclampsia and preterm labor as compared to normal term pregnancies, indicating which patient’s altered expression and function may associate with pregnancy complications (28). It has been reported that ZBTB16 is essential for decidualization, which supports embryo implantation and placentation, as well as subsequent events that together ensure a successful pregnancy (29), and that HDAC2 was poorly expressed in placental tissue samples from pregnant women with preeclampsia (30,31). In this study, down-regulated hsa-miR-483-5p was detected in SCM of a pregnant group and ZBTB16/HDAC2 were targets of hsa-miR-483-5p. These findings may indicate that hsa-miR-483-5p plays a unique role via regulating ZBTB16/HDAC2 in early embryo development.

In recent years, the role of miR-432-5p in cancers has been extensively studied. A previous study revealed that decreased miR-432-5p was reported to promote breast cancer cell growth and migration by targeting E2F3 (32). It has also been reported that miR-432-5p was implicated in facilitating cell progression by activating Wnt/β-catenin signaling and promoting cell growth and metastasis in human hepatocellular carcinoma (33,34). Yang et al. reported that lower levels of miR-432-5p were observed in both glioma tissues and cells (35), and it has been extensively reported that translocation t(X;17)(p11;q25) of ASPSCR1 with transcription factor TFE3 gene results in a ASPSCR1-TFE3 fusion protein in alveolar soft part sarcoma and in renal cell carcinomas (36,37). ZNF384-related fusion genes were detected in acute lymphoblastic leukemia (38), and GALNT9 was demonstrated to be well conserved in zebrafish during embryonic development (39). To the best of our knowledge, there is no study linking miR-432-5p and ASPSCR1/ZNF384/GALNT9 to embryo development. In the current study, the expression of hsa-miR-432-5p was reduced in SCM of the pregnant group, and ASPSCR1/ZNF384/GALNT9 were targets of hsa-miR-432-5p, which indicates the role of hsa-miR-432-5p-ASPSCR1/ZNF384/GALNT9 in embryo development should be studied further.

Firstly, improving embryo quality and selecting high-quality embryos for transfer is a critical step to improve the success rate of embryo implantation. Secondly, couples should make lifestyle adjustments, including quitting smoking and alcohol, and try to maintain a normal body mass index (BMI) by gaining or losing weight. It has been suggested that extending embryo culture to day 5 or 6 in order to transfer the embryo at the blastocyst stage increases the implantation rate (40). In addition, assisted hatching has been proposed as one technique to improve implantation and pregnancy rates following IVF Improving embryo quality and selection (41).

Conclusions

In conclusion, the three miRNAs mentioned above may be involved in early embryo development by regulating abundant target genes. These findings indicate the possibility of using miRNAs secreted by embryos into SCM as non-invasive biomarkers to predict the outcomes of IVF. Our present study lays the foundation for developing biomarkers for predicting embryo quality and exploring the roles of embryonic miRNAs found in SCM. The limitations of present study are the low abundance of miRNAs, small sample size, and lack of experimental verification. Although the results of qRT-PCR and validation in the GEO database suggested the reliability of our RNA-sequencing results, ex-vivo experiments are required to confirm the role of miRNAs identified in this study, and clinical validation with larger sample sizes is mandatory to confirm its effectiveness.

The article’s supplementary files as