In recent decades, assisted reproduction technology (ART) has been widely used to treat human infertility. However, only approximately 30% of in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI) cycles result in pregnancy1. The classical semen parameters, such as sperm density, morphology, and motility, are not sufficient to effectively assess sperm fertility. Therefore, a method for distinguishing high-quality sperm samples from samples with normal semen parameters is highly desirable. Many types of small non-coding RNAs (sncRNAs), including microRNAs (miRNAs), PIWI-interacting RNAs (piRNAs), tRNA-derived small RNAs (tsRNAs) and rRNA-derived small RNAs (rsRNAs), have been found in mammalian male germ cells and play important regulatory roles in spermatogenesis2,3. Identification of the subpopulations of sperm sncRNAs responsible for sperm fertility is of great interest. However, using sncRNAs as prognostic biomarkers for evaluating sperm quality for IVF has not been performed to date. Here, we found that sncRNAs were significantly associated with sperm quality, providing useful biomarkers for improving the success rate of IVF.
In this study, 87 human sperm samples were collected from male partners of couples undergoing IVF treatment (Supplementary Table S1). Microscopic analysis of the isolated sperm by Diff-Quik staining showed that the sperm morphology was normal and the purity was high, which were appropriate for RNA extraction and deep sequencing (Supplementary Fig. S1). The sperm samples were divided into two groups according to the rate of good quality embryos: high rate of good quality embryos (H-GQE, GQE≥75%, n = 23) and low rate of good quality embryos (L-GQE, GQE≤25%, n = 64)4. The maternal influence was excluded as much as possible, and there were no significant differences in age, numbers of oocytes obtained, or MII (metaphase II stage) oocytes among the female partners. The sperm samples were all classified as normal sperm samples by a semen-parameter assessments, including sperm density, morphology, viability and progressive motility5, except that the L-GQE group had a slightly higher normal sperm morphology rate (P < 0.05) than the H-GQE group.
We extracted total RNAs from 87 human sperm samples and profiled the expression of sncRNAs by deep sequencing. We found that tsRNAs, rsRNAs, miRNAs, and piRNAs were highly abundant in the samples (Fig. 1a; Supplementary Fig. S2a), which is consistent with previous reports in mice2,3. On average, ~56% of the sncRNAs annotated to tsRNAs, 18% to rsRNAs, 6% to miRNAs, and 4% to piRNAs (Supplementary Tables S2-S3). The length distribution of these sncRNAs was similar in each sample in both groups. The peak of the tsRNA-, rsRNA-, and miRNA-length ranged from 29 to 34 nt, 17 to 40 nt and 20 to 23 nt, respectively. tsRNAs can be classified into five groups according to the region on tRNAs from which they are derived: 5′-tRNA halves, 3′-tRNA halves, 5′-tRFs (tRNA-derived RNA fragment), i-tRFs (internal tRFs), and 3′-tRFs6. Interestingly, among the five types of tsRNAs, the 5′-tRNA halves were the most abundant tsRNAs in human sperm, accounting for more than 75% of all tsRNAs (Supplementary Fig. S2b). rsRNAs can be divided into five types according to the subtypes of rRNA precursors (5S, 5.8S, 18S, 28S, and 45S) from which they are derived7. We found that the 28S rRNA precursor-derived rsRNAs (28S rsRNA) were the most abundant rsRNAs, accounting for 60% of all rsRNAs (Supplementary Fig. S2c).
In total, 1899 tsRNAs (average RPM >10) were detected in the sperm samples (Supplementary Table S4). The ratio of total tsRNAs expressed in the L-GQE group was similar to that in the H-GQE group (Fig. 1b). Notably, most of the top 100 tsRNAs were from the 5′-end of the tRNA, including 5′-halves and 5′-tRF (Supplementary Fig. S3a). In contrast, 3′-halves, 3′-tRFs, and i-tRFs were expressed at very low levels. Previous studies showed that 5′-tsRNAs can regulate translation by different mechanisms, such as interfering with translation initiation and RNA modification, or repressing translation by miRNA-like mechanism8. Moreover, sperm tsRNAs are able to mediate the transmission of paternal traits to offspring and influence embryonic gene expression in mice9,10,11, indicating that tsRNAs might play important roles in sperm maturation as well as early embryonic development. To further investigate the relationship between tsRNAs and sperm quality, we identified ten differentially expressed tsRNAs between the two groups (Fig. 1c, d; Supplementary Table S5). Among these tsRNAs, five tsRNAs, including GlyGCC-30-1 (the first number indicates the length of the tsRNA, the second number indicates the isoform of tsRNAs with the same length), GlyGCC-30-2, ThrTGT-38, ThrTGT-39, and GluTTC-23, were downregulated in the L-GQE group. Five other tsRNAs, including ProAGG-32, ProTGG-32, ProAGG-31, AsnATT-20, and ArgCCG-33, were upregulated (Fig. 1d, e). These differentially expressed tsRNAs belong to different types of tsRNAs: three were 5′-tRNA halves, two were 3′-tRNA halves, four were 5′-tRFs and one was an i-tRF. Furthermore, we found that a principal component analysis (PCA), which is a powerful tool for exploratory data analysis and generating predictive models, could separate the H-GQE group from the L-GQE group based on these ten tsRNAs (Fig. 1f). The accuracy of the support vector machine (SVM) classifier was evaluated by the area under the ROC curve (AUC), and the results showed the reliability of the classifier for the ten tsRNAs (Fig. 1g, AUC = 0.8716). These results indicated that the ten tsRNAs have an excellent prognostic value and can be potential biomarkers for assessing human sperm quality for IVF. Recently, tRF-Gly-GCC was shown to repress genes associated with the endogenous retroelement MERVL in zygotes and later in development10. Intriguingly, we found that two types of 5′-tRF, GlyGCC-30-1, and GlyGCC-30-2, were downregulated in the L-GQE group. Thus, the mis-regulation of these tsRNAs in sperm might contribute to poor sperm quality and abnormal early embryo development, a notion which warrants future investigation.
rsRNAs are a type of rRNA-derived sncRNA abundantly expressed in mature sperm. Previous studies showed that they were sensitive to the environment and were diminished in the sperm of leukocytospermia patients3,11. Remarkably, we found that rsRNAs were highly expressed in human sperm (Fig. 1a, b; Supplementary Table S6), especially the rsRNAs derived from the 28S rRNA (Supplementary Fig. S2c, 3b). The ratio of total rsRNAs expressed in the L-GQE group was slightly lower than that in the H-LEQ group (Fig. 1b). To explore the association between rsRNAs and sperm quality, we analyzed the expression of rsRNAs and identified seven differentially expressed rsRNAs: 28S-58, 28S-34, 28S-23, 28S-20, 28S-25, 18S-37, 28S-19 (the number after the dash indicates the length of the rsRNA) (Fig. 1h; Supplementary Table S7). Only 28S-58 was upregulated in the L-GQE group, the remaining six rsRNAs were all downregulated (Fig. 1h). PCA and SVM classifier analyses showed that these seven rsRNAs could also classify the samples into two groups (Fig. 1i, j, AUC = 0.8588), indicating that these rsRNAs have comparable predictive power as tsRNAs and may be another type of useful biomarker for the clinical evaluation of sperm quality.
miRNAs were also expressed in human sperm (Fig. 1a, b), and 467 known miRNAs were detected (average RPM > 10) in 87 samples (Supplementary Table S8). Similar to tsRNAs, there was no significant difference in the ratio of total miRNAs expressed in the two groups (Fig. 1b). Nearly half of the top 100 miRNAs were upregulated in almost half of the L-GQE sperm samples, while another half of the miRNAs were downregulated (Supplementary Fig. S3c), indicating that miRNA expression in these samples is dysregulated during sperm maturation. We identified five miRNAs that were differentially expressed between the two groups; three miRNAs (miR-132-3p, miR-191-3p, and miR-520a-5p) were downregulated, and two miRNAs (miR-101-3p and miR-29a-3p) were upregulated in the L-GQE group (Fig. 1k; Supplementary Table S9). Both PCA and SVM (AUC = 0.7022) analyses showed that the miRNAs can be used for separating the H-GQE and L-GQE groups (Fig. 1l, m). Moreover, an analysis of significant gene ontology (GO)-enriched terms showed that both the downregulated and upregulated target genes were involved in the biological processes of cell development and differentiation (Supplementary Fig. S4a, b), indicating that these miRNA target genes might be important for spermatogenesis and early embryo development.
In conclusion, we investigated differentially expressed sncRNAs in human sperm as candidate markers for evaluating sperm quality during IVF. We demonstrated that differentially expressed tsRNAs, rsRNAs, and miRNAs are linked to sperm quality according to embryo quality, even though these sperm samples were all considered normal by the traditional semen-parameter assessment. Therefore, the sncRNAs, especially tsRNAs and rsRNAs, may be potential clinical biomarkers for the assessment of sperm quality in IVF.
All data used to obtain the conclusions in this paper are presented in the paper and Supplementary Information. The deep sequencing data have been deposited in the National Center for Biotechnology Information Expression Omnibus (GEO) (http://www.ncbi.nlm.nih.gov/geo/) under accession number GSE110190.
Adamson, G. D. et al. International Committee for Monitoring Assisted Reproductive Technology: world report on assisted reproductive technology, 2011. Fertil. Steril. 110, 1067–1080 (2018).
Chen, Q., Yan, W. & Duan, E. Epigenetic inheritance of acquired traits through sperm RNAs and sperm RNA modifications. Nat. Rev. Genet. 17, 733–743 (2016).
Chu, C. et al. A sequence of 28S rRNA-derived small RNAs is enriched in mature sperm and various somatic tissues and possibly associates with inflammation. J. Mol. Cell Biol. 9, 256–259 (2017).
Alpha Scientists in Reproductive Medicine, E. S. I. G. o. E. The Istanbul consensus workshop on embryo assessment: proceedings of an expert meeting. Hum. Reprod. 26, 1270–1283 (2011).
World Health Organization. WHO laboratory manual for the examination and processing of human semen. (WHO Press, 2010).
Loher, P., Telonis, A. G. & Rigoutsos, I. MINTmap: fast and exhaustive profiling of nuclear and mitochondrial tRNA fragments from short RNA-seq data. Sci. Rep. 7, 41184 (2017).
Shi, J., Ko, E. A., Sanders, K. M., Chen, Q. & Zhou, T. SPORTS1.0: a tool for annotating and profiling non-coding RNAs optimized for rRNA- and tRNA-derived small RNAs. Genom., Proteom. Bioinform. 16, 144–151 (2018).
Shi, J., Zhang, Y., Zhou, T. & Chen, Q. tsRNAs: the Swiss army knife for translational regulation. Trends Biochem. Sci. 44, 185–189 (2018).
Chen, Q. et al. Sperm tsRNAs contribute to intergenerational inheritance of an acquired metabolic disorder. Science 351, 397–400 (2016).
Sharma, U. et al. Biogenesis and function of tRNA fragments during sperm maturation and fertilization in mammals. Science 351, 391–396 (2016).
Zhang, Y. & Zhang, X. Dnmt2 mediates intergenerational transmission of paternally acquired metabolic disorders through sperm small non-coding RNAs. Nat. Cell Biol. 20, 535–540 (2018).
We thank Yanfei Ru, Yihua Gu, Jufen Zheng, Fudong Yu, and Tiancheng Zhang for help with this work. We thank Miao Liu, Yao Yuan, Aijie Xin, Ronggui Qu, Xiaofeng Tang, Yulin Liu, Hua Xu, and Xi Dong for useful clinical suggestions. This work was supported by the Ministry of Science and Technology of China (2014CB943100 to H.S. and 2017YFA0504400 to L.W.), the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB19040102 to L.W.), the Shanghai Municipal Commission for Science and Technology (17JC1420100 to H.S. and 12JC1409400 to L.W.), the National Natural Science Foundation of China (91440107 and 31470781 to L.W.) and the State Key Laboratory of Molecular Biology. This work was also supported by the Merck Serono China Research Fund for Fertility Experts, the Shanghai Municipal Health and Family Planning Commission (201640369 Y.G.), and the National Natural Science Foundation of China (31801252 to Y.R.).
G.C., H.S., and L.W. conceived and designed the study. M.H. performed the data analysis. W.L. and F.Z. performed the computational analysis. Y.C. performed the sample collection and information register. B.X. designed the initial protocol for sequencing small RNAs. S.L. contributed to the sample information validation. G.C. designed the IVF studies. M.H., W.L., H.S., and L.W. wrote the manuscript.
Conflict of interest
The authors declare that they have no conflict of interest.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Hua, M., Liu, W., Chen, Y. et al. Identification of small non-coding RNAs as sperm quality biomarkers for in vitro fertilization. Cell Discov 5, 20 (2019). https://doi.org/10.1038/s41421-019-0087-9
MicroRNA expression profile analysis in sperm reveals hsa-mir-191 as an auspicious omen of in vitro fertilization
BMC Genomics (2020)
Journal of Internal Medicine (2020)
A comprehensive overview of bull sperm-borne small non-coding RNAs and their diversity across breeds
Epigenetics & Chromatin (2020)
Histone Post-Translational Modifications and CircRNAs in Mouse and Human Spermatozoa: Potential Epigenetic Marks to Assess Human Sperm Quality
Journal of Clinical Medicine (2020)
PLOS Biology (2019)