The present study involved 86 infertile males (B) with abnormal semen and 86 normal healthy adult males (H) as the control. The samples were collected from the inpatient clinic of the International Peace Maternity and Child Health Hospital of the China Welfare Institute (Shanghai, China) between February and September 2010. All human materials were obtained according to consent regulations and approved by the Ethical Review Committee of the World Health Organization Collaborating Center for Research in Human Reproduction in Shanghai, China as authorized by the Shanghai Municipal Government. Due to material limitations, we could only analyze a limited number of severely abnormal sperm samples.

Semen samples were produced by masturbation, collected in sterile containers and immediately transported to the laboratory. A conventional semen profile was obtained for each sample using the procedures described by the World Health Organization (10).

Total RNA was isolated from each semen sample using the TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) according to the manufacturer’s instructions (28,29). The RNA samples were treated with DNase I (Sigma-Aldrich, St. Louis, MO, USA) and then quantified.

RNA labeling and hybridization were performed on miRNA microarray chips as previously described (25,27,30). Briefly, 50 μg of total RNA was purified using the mirVana miRNA isolation kit (Ambion, Austin, TX, USA) to enrich the small RNA fraction. The purified RNA was labeled with fluorescein and hybridized using the CapitalBio mammalian miRNA array V3.0 (CapitalBio Corporation, Beijing, China) containing 2844 mature miRNA gene oligonucleotide probes in triplicate, corresponding to 1823 human, 648 mouse and 373 rat miRNAs. Each individual’s semen RNA was analyzed on a separate chip. Finally, scanned images of the microarray were captured and the hybridization signals were quantified. The signal intensity values were normalized to per-chip mean values.

Following the detection of total RNA, we used a Poly(A) Tailing kit (Ambion) to add a poly(A) tail to the RNA products according to the kit’s instructions. The RNA samples were treated with DNase I, quantified and reverse-transcribed into cDNA using the ReverTra Ace-α First Strand cDNA Synthesis kit (Toyobo, Osaka, Japan). Notably, this reverse transcription reaction uses the oligo(dT) reverse transcription primer 5′-GCTGTCAACGATACGCTACCTAACGGCATGACAGTGTTTTTTTTTTTTTTT(C/G/A)-3′. All reaction steps were carried out according to the manufacturer’s instructions.

In accordance with the manufacturer’s instructions and as previously described (23), qRT-PCR was conducted in the realplex⁴ real-time PCR detection system from Eppendorf (Hamburg, Germany), using SYBR^® Green RealTime PCR Master mix (Toyobo) as the detection dye. The qRT-PCR amplification process comprised 40 cycles of denaturation at 95°C for 10 sec and annealing at 57°C for 20 sec. The target cDNA was quantified using a relative quantification method. A comparative threshold cycle (Ct) was used to quantify the gene expression relative to the control (calibrator). The steady-state mRNA levels were expressed as an n-fold difference relative to the calibrator. For each sample, the Ct values were normalized using the formula: ΔCt = Ct_miRNA − Ct_{18S rRNA}. To detemine relative expression levels, the following formula was used: ΔΔCt = ΔCt_B − ΔCt_H. The values used to plot the relative miRNA expression levels were calculated using the expression 2^−ΔΔCt. The miRNA levels were calibrated by 18S rRNA. The miRNA primers used in the cDNA amplification are shown in Table I.

All steps in the northern blotting process were carried out as previously described (28,29). For all samples, 20 μg good quality total RNA was analyzed on a 7.5 M urea 12% PAA denaturing gel and transferred to a Hybond-N+ nylon membrane (Amersham, Freiburg, Germany). The membranes were crosslinked using UV light for 30 sec at 1200 mJ/cm². Hybridization was performed using miRNA antisense StarFire probes to detect the 22-nt miRNA fragments, according to the manufacturer’s instructions. After washing, the membranes were exposed for 20–40 h to Kodak XAR-5 films (Sigma-Aldrich). The ethidium bromide-stained gels prior to the transfer of tRNA were used as controls to ensure equal loading of the RNA samples.

Each experiment was performed at least three times and data are the mean ± SE, where applicable. Differences were evaluated using the Student’s t-test. p<0.05 was considered to indicate a statistically significant result.

A total of 172 males were invited to participate in this study: 86 healthy males with normal semen and 86 infertile males with semen abnormalities. The only significant difference between the populations was the percentage of progressive motile (a+b) forms (p<0.001). The results of laboratory tests indicated that asthenozoospermia was the most frequent finding in the 86 infertile males. The characteristics of the study participants are presented in Table II.

A 260 to 280 nm absorbance ratio (260/280)>1.8 is usually considered to be an acceptable indicator of RNA purity for miRNA microarrays and indicates an absence of detectable protein contamination in the RNA sample (15). Following the extraction of total RNA from the samples, the 260/280 ratio of each extract was determined using a spectrophotometer (15,31). The 260/280 ratios ranged from 1.83 to 1.97. Formaldehyde denaturing gel electrophoresis was used to confirm the presence of clear 28S, 18S and 5S bands (Fig. 1) and the absence of marked RNA degradation. This analysis indicated that the purity and integrity of each RNA sample met the requirements of the miRNA microarray and qRT-PCR experiments (15).

In order to identify miRNAs which are differentially expressed between the abnormal semen of infertile males and the normal semen of healthy males, we prepared a miRNA microarray containing 2844 oligonucleotide probes (1823 human, 648 mouse and 373 rat) complementary to known mammalian miRNAs (23,24,32). All probes were repeated three times in each microarray and each microarray contained 16 controls (Zip5, Zip13, Zip15, Zip21, Zip23, Zip25, Y2, Y3, U6, New-U2-R, tRNA-R, has-let-7a, has-let-7b, has-let-7c, 50% DMSO and Hex). In order to increase the reliability of the results, each miRNA microarray assay was repeated twice (24) and the scatter plots for all spots indicated that a high reproducibility and reliability were achieved (Fig. 2A).

The miRNA expression patterns for abnormal semen from infertile males (B) and normal semen from healthy males (H) were compared. Significance analysis of microarray (SAM) and a fold change criterion (B/H ratio) >1.50 or <0.667 and p<0.001 were used to identify significant differences (32,33). Using these criteria, we identified 52 miRNAs which were differentially expressed between the semen of infertile males and normal males. Analysis of the microarray expression levels confirmed that 21 miRNAs (mi-574-5p, let-7b, miR-297, miR-122, miR-1275, miR-1281, miR-373, miR-185, miR-193b, miR-145, miR-214, miR-574-3p, miR-483-5p, miR-324-3p, miR-372, miR-484, miR-933, miR-663, miR-1268, miR-923 and miR-1234) were significantly overexpressed in the abnormal semen compared with the normal semen. Conversely, 31 miRNAs (miR-1826, miR-493, miR-371-5p, miR-516a-5p, miR-512-5p, miR-498, miR-30a, miR-23a, miR-130a, miR-103, miR-30b, miR-27a, miR-18a, miR-525-3p, miR-517c, miR-199b-3p, miR-517b, miR-107, miR-199a-3p, miR-1323, miR-515-5p, miR-31, miR-455-3p, miR-100, miR-517a, miR-512-3p, miR-16, miR-523, miR-19b, miR-23b and miR-26a) were significantly underexpressed in the abnormal semen compared with the normal semen (Table III).

Following common procedures for the confirmation of microarray analysis (23,24,32–34), qRT-PCR was used to confirm the results of the miRNA microarray analysis. Of the 11 miRNAs identified by the microarray as being the most overexpressed in the abnormal semen of infertile males compared with normal semen (miR-574-5p, let-7b, miR-297, miR-122, miR-1275, miR-1281, miR-373, miR-185, miR-193b, miR-145 and miR-214), qRT-PCR confirmed that seven (miR-574-5p, miR-297, miR-122, miR-1275, miR-373, miR-185 and miR-193b) were overexpressed. Of the ten miRNAs identified as being underexpressed in abnormal semen by the microarray (miR-31, miR-455-3p, miR-100, miR-517a, miR-512-3p, miR-16, miR-523, miR-19b, miR-23b and miR-26a), the qRT-PCR analysis confirmed that six of these (miR-100, miR-512-3p, miR-16, miR-19b, miR-23b and miR-26a) were underexpressed.

Scatter plot analysis of the qRT-PCR results confirmed that seven miRNAs (miR-574-5p, miR-297, miR-122, miR-1275, miR-373, miR-185 and miR-193b) were overexpressed and six miRNAs (miR-100, miR-512-3p, miR-16, miR-19b, miR-23b and miR-26a) were underexpressed in the semen of infertile males compared with the normal semen (Fig. 2B).

A Venn diagram (Fig. 2C) was used to depict the correlation between the results of the miRNA microarray and the 21 miRNAs tested by qRT-PCR. The differential expression of 13 miRNAs (miR-574-5p, miR-297, miR-122, miR-1275, miR-373, miR-185, miR-193b, miR-100, miR-512-3p, miR-16, miR-19b, miR-23b and miR-26a) in the abnormal semen of the infertile males was confirmed by qRT-PCR (indicated by the overlap in the diagram). The expression levels of the other miRNAs correlated in some or other methods. Overall, the qRT-PCR analysis indicated that the miRNA microarray results had some small errors, however, it confirmed that a significant number of miRNAs are differentially regulated in the abnormal semen of infertile males.

The expression levels of the 13 miRNAs which were confirmed to be differently expressed by qRT-PCR were further investigated by northern blotting of the RNA isolated from the abnormal semen of three infertile males and the normal semen of three healthy adult males. Anti-sense miRNA locked nucleic acid probes were used for each miRNA (Fig. 3). The northern blotting hybridization signals for miR-574-5p, miR-297, miR-122, miR-1275, miR-373, miR-185 and miR-193b were weaker in the semen of healthy adult controls than that of the infertile males, confirming that these miRNAs are upregulated in the abnormal semen. The miR-100, miR-512-3p, miR-16, miR-19b, miR-23b and miR-26a hybridization bands were barely detectable and, therefore, we could not confirm the differential regulation of these miRNAs using northern blotting.