Sperm DNA integrity status is associated with DNA methylation signatures of imprinted genes and non-imprinted genes

Abstract

Purpose

To evaluate the association between the DNA methylation of specific genes and sperm DNA integrity status in human sperm samples.

Methods

A total of 166 semen samples were evaluated (86 controls and 80 cases with impaired sperm DNA integrity). We detected the methylation status of 257 CpG sites among two imprinted genes (H19 and SNRPN) and four non-imprinted genes related to male infertility (MTHFR, GSTM1, DAZL, and CREM) by using a targeted next-generation sequencing method.

Results

Differential methylation was found in 43 CpG sites of the promoters of the six candidate genes. H19, SNRPN, MTHFR, DAZL, GSTM1, and CREM contained 22, 12, 1, 4, 0, and 4 differentially methylated CpG sites (P<0.05), respectively. The imprinting genes were associated with relatively higher rates of differentially methylated CpG sites (28.21% in H19 and 41.38% in SNRPN) than the non-imprinting genes. One CpG site in H19 remained significant after performing strict Bonferroni correction.

Conclusion

In this study, we found that different site-specific DNA methylation signatures were correlated with sperm DNA integrity status. Further studies are needed to investigate the specific mechanisms leading to the epigenetic modifications.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10815-021-02157-6.

Introduction

Assisted reproductive techniques (ARTs) provide an opportunity for infertile couples to conceive children, however, they also show a potential risk of transmission of genetic/epigenetic alterations [1]. Many studies have reported enhanced risks of congenital imprinting disorders in children conceived via ARTs, especially via intra-cytoplasmic sperm injection (ICSI) [2, 3]. Studying the sperm epigenetic aberrations in male infertility can help us determine whether these syndromes arise from the use of ARTs themselves or from epigenetic abnormalities that already exist in the gametes [4–6].

DNA methylation is the most studied epigenetic mechanism, which is characterized by reversible and heritable modifications of cytosine residues in eukaryotic organisms [7, 8]. Differential methylation of genes is closely related to gene regulation, and the methylation patterns differ between paternal and maternal alleles. The imprinted genes are crucial for embryo development, fetal growth, and post-natal behavior. Additionally, the aberrant DNA methylation of sperm can be transmitted to the developing embryo and offspring, which is likely to affect the embryonic development [9–11].

Several studies have shown that global DNA methylation is significantly correlated with sperm parameters and sperm chromatin status [12, 13]. Global DNA methylation level in human sperm could even influence the IVF results [14]. It has been also proved that aberrant DNA methylation patterns of spermatozoa were found in men with unexplained infertility and idiopathic recurrent pregnancy loss [15, 16]. It was recently proposed that the DNA methylation level is a new candidate biomarker of male fertility [17]. Early studies on sperm DNA methylation were specifically performed on imprinted genes, due to the increased risk of the development of congenital imprinting diseases in children conceived via ART [4, 18–20]. These studies showed the aberrant methylation patterns of imprinted genes present in poor-quality sperm [21]. Altered sperm DNA methylation patterns have also been detected in non-imprinted genes associated with spermatogenic impairment, such as methylenetetrahydrofolate reductase (MTHFR), cAMP-responsive element modulator (CREM), deleted in azoospermia-like (DAZL), and other genes involved in spermatogenesis [22–25]. Glutathione S-transferase mu 1 (GSTM1) is a major member of the glutathione transferase (GST) gene family, which plays a role as an important antioxidant in the testes and seminiferous tubules [26]. These studies highlighted that the defective sperm DNA methylation patterns of both imprinted and non-imprinted genes have been associated with subfertility and conventional poor sperm parameters.

Various studies have confirmed that sperm DNA integrity plays an important role in sperm maturation, fertilization, embryo development, and normal pregnancy [27, 28]. Sperm DNA integrity is calculated using the DNA fragment index (DFI), which assesses male fertility combined with conventional semen analysis [28]. Therefore, it is reasonable to presume that sperm DNA damage is potentially associated with defects in sperm DNA methylation, given that both defects are related to defective spermatogenesis. Studies have shown that there is a negative association between global sperm DNA methylation levels and sperm DNA integrity status [13]. A study described the use of reduced representation bisulfite sequencing to investigate DNA methylation in boar sperm cells showing different levels of DFI [29], the authors found that boar sperm cells that show varying levels of DNA fragmentation exhibit different site-specific DNA methylation signatures. However, epigenetic studies evaluating the relationship between DNA methylation aberrations of specific genes and human sperm DNA integrity status in a larger population are still required.

This study was designed to further evaluate the association between DNA methylation of specific genes and sperm DNA integrity status by conducting a targeted next-generation sequencing (NGS) analysis of a large cohort of semen samples. Combined with previous literature research, the two most-studied imprinted genes, H19 imprinted maternally expressed transcript (H19) and small nuclear ribonucleoprotein polypeptide N (SNRPN), and four non-imprinted genes related to male infertility (MTHFR, GSTM1, DAZL, and CREM) were analyzed in the study.

Materials and methods

Patient samples

A total of 166 semen samples were obtained (86 normal controls and 80 cases showing impaired sperm DNA integrity). The samples in this study were all from patients who were seeking for fertility help in our center. None of the patients had impregnated their female partners in the last year. The time of ejaculatory abstinence is 3–7 days before sample collection. All semen samples were collected via masturbation into a sterile container. Semen parameters, including semen volume, sperm concentration, and motility, were assessed in the laboratory of The First Affiliated Hospital of Anhui Medical University, Anhui, China, according to the World Health Organization guidelines [30]. The participants included in this study reported no exposure to environmental pollutants (e.g., carbon monoxide, toxic metals, radioactive pollutants, and heavy metals) recently. This study was approved by the Ethical Review Board of The First Affiliated Hospital of Anhui Medical University (PJ2020-03-13). Written informed consent was obtained from all enrolled patients.

Sperm chromatin structure assay (SCSA)

At the same time of semen parameter analysis, sperm DNA integrity and chromatin condensation were assessed by performing SCSAs and were reported as DFI values [31, 32]. Briefly, the spermatozoa were exposed to an acidic detergent solution and stained with acridine orange. The fluorescence patterns were analyzed using a FACScan flow cytometer (Becton Dickinson, San Jose, CA, USA). Red fluorescence represented single or double DNA strand breaks, whereas green fluorescence corresponded to normal intact DNA. The ratio of red to total fluorescence intensity was used to calculate the percentage of spermatozoa showing DNA fragmentation. Each sample was measured in duplicate, and the mean of the values was calculated. Sperm samples were categorized into two groups: normal control group (DFI≤15%) and impaired group (DFI >15%), based on their DFI values. The cut-off value was recommended by consensus of Chinese experts from Chinese Andrology Association [33] and the instruction of the testing kit (CellPro Biotech, Ningbo, China).

Sperm DNA isolation and sodium bisulfite treatment

Semen samples were centrifuged using the density gradient method (1000rpm for 5min) and washed twice with phosphate-buffered saline to remove as many somatic cells as possible. Each sample was assessed for the presence of round cell contaminants. If somatic cell contamination was confirmed, the entire sample was treated using the swim-up technique and subjected to genomic DNA isolation. Genomic DNA was isolated from the sperm samples using the DNA kit (Qiagen Inc., Germany), according to the manufacturer’s protocol. The concentration and purity of extracted DNA were estimated using the NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). Genomic DNA (500 ng) from each sample was treated with sodium bisulfite using the EZ DNA Methylation-Gold Kit (Zymo Research, Irvine, CA, USA), by following the manufacturer’s standard protocol.

DNA methylation analysis of the candidate genes

CpG islands were selected in the region of 2 kb upstream of the transcriptional start site to 1 kb downstream of the first exon in the promoter region of the six genes. The methylation levels of the selected genes were analyzed using MethylTarget™ (Genesky Biotechnologies Inc., Shanghai, China), an NGS-based multiple methylation-specific polymerase chain reaction (PCR) analysis. In total, 257 CpG sites were analyzed by performing targeted bisulfite sequencing, according to the manufacturer’s instructions. The EZ DNA Methylation-Gold Kit (Zymo Research) was used for bisulfite modification of genomic DNA samples, followed by the conversion of unmethylated cytosine to uracil. Multiplex PCR was performed using primers designed by MethPrimer (http://www.urogene.org/) and Primer3web (http://primer3.ut.ee/) to detect the converted sequences (Table S1). A primer containing an index sequence and a specific tag compatible with the Illumina platform was introduced into the end of the converted sequences via PCR amplification. PCR amplification products of all samples were equally mixed, and a sequencing library was created. High-throughput bisulfite sequencing was performed using the Illumina MiSeq sequencing platform (Illumina, San Diego, CA, USA), according to the manufacturer’s protocols. All data obtained from the sequencing step were processed, filtered, and aligned using the BiQ Analyser HT software by excluding all reads containing equal to or more than 10% of missing CpG sites (maximal fraction of unrecognized sites ≥ 0.1). The aligned sequences showed the absence of alterations at CpG positions.

Statistical analysis

Age and semen parameters are expressed as the mean and standard deviation. Data were analyzed using SPSS version 22.0 (IBM, Armonk, NY, USA). The independent-sample t test (Mann-Whitney test) was applied to compare the mean values of quantitative variables. P values < 0.05 were considered significant. In the subgroup analysis of each CpG site, a Bonferroni correction method was also applied for multiple comparisons, and P < 0.00019 (0.05/257) was considered statistically significant.

Results

Clinical data

The clinical features and semen parameters of the subjects recruited in this study are listed in Table 1. There were no significant differences in the mean age between subjects in the normal control and impaired groups (30.94±4.13 and 32.05±3.69, respectively, P>0.05). To exclude the possible effects of the other semen parameters studied, the semen volume, sperm concentration, and sperm motility were also compared between the two groups, which showed no significant differences (P>0.05). The mean DFI value of the control group (10.74±2.93) was significantly lower than that of the impaired group (30.18±9.93, P<0.01).

Table 1.

Clinical characteristics and semen parameters of men in the study

Group	Normal control (n=86)	Impaired group (n=80)	P
Age (years)	30.94±4.13	32.05±3.69	NS
Semen volume (mL)	3.87±1.29	3.83±1.49	NS
Sperm count (*10^6/mL)	80.29±55.04	89.72±76.43	NS
Progressive motility (%)	40.43±20.82	37.65±11.73	NS
DFI (%)	10.74±2.93	30.18±9.93	<0.05

Data are expressed as mean and standard deviations and analyzed by t test

DFI DNA fragmentation index, NS not significant (P≥0.05)

CpG site methylation of imprinted genes

The methylation of two imprinted genes (H19 and SNRPN) was measured in the semen samples at the promoter regions via sodium bisulfite treatment combined with MethylTarget sequencing. A total of 107 CpG sites were detected, and 34 CpG sites were found to be significantly associated with damage to sperm DNA integrity (Table 2).The percentages of differentially methylated sites (DMSs) of H19 and SNRPN were 28.21% and 41.38%, respectively (Table 2). Analyses of the individual CpG sites using the two-tailed Student’s t test revealed that the DMSs of H19 were all increased significantly in the impaired group compared to those in the control group. The methylation differences of one CpG site in H19 remained significant after performing strict Bonferroni correction (Table 3). A significantly lower level of DNA methylation was detected at the 12 DMSs of SNRPN. The methylation status of individual CpG sites of SNRPN showed no significant differences between the groups after performing Bonferroni correction (Table 4).

Table 2.

Number of differentially methylated sites in the six candidate genes between the two groups

Gene	Chr	Number of CpG island	Number of detected CpG sites	Number of DMS	Percentage of DMS (%)
H19	11	5	78	22	28.21
SNRPN	15	2	29	12	41.38
MTHFR	1	2	45	1	2.22
GSTM1	1	1	11	0	0
DAZL	3	1	21	4	19.05
CREM	10	2	73	4	5.48

Percentage of DMS: number of DMS/detected CpG sites × 100%

Chr chromosome number; DMS differentially methylated site

Table 3.

CpG sites of H19 with significant differences in methylation between patients and matched controls

Target	Position	Case	Control	β	P value	P-Adj
H19_1	43	0.576±0.036	0.553±0.046	0.023	0.000	NS
H19_2	53	0.979±0.009	0.976±0.009	0.003	0.040	NS
H19_3	55	0.571±0.034	0.546±0.051	0.024	0.000	NS
H19_4	66	0.976±0.020	0.965±0.041	0.010	0.042	NS
H19_5	77	0.982±0.007	0.979±0.008	0.003	0.008	NS
H19_6	100	0.566±0.038	0.539±0.048	0.027	0.000	0.036
H19_7	124	0.974±0.020	0.959±0.042	0.015	0.005	NS
H19_8	128	0.973±0.021	0.959±0.045	0.014	0.015	NS
H19_9	141	0.974±0.020	0.961±0.041	0.013	0.014	NS
H19_10	146	0.976±0.020	0.963±0.041	0.012	0.018	NS
H19_11	149	0.571±0.035	0.550±0.049	0.022	0.001	NS
H19_12	151	0.568±0.035	0.544±0.049	0.024	0.000	NS
H19_13	152	0.949 ±0.028	0.936±0.048	0.013	0.034	NS
H19_14	167	0.575 ±0.035	0.552±0.049	0.023	0.001	NS
H19_15	172	0.966±0.021	0.952±0.042	0.014	0.010	NS
H19_16	184	0.976 ±0.020	0.966±0.040	0.010	0.039	NS
H19_17	194	0.975±0.019	0.962±0.038	0.013	0.006	NS
H19_18	205	0.977±0.016	0.963±0.038	0.013	0.005	NS
H19_19	206	0.574±0.036	0.550±0.050	0.024	0.001	NS
H19_20	231	0.952±0.024	0.940±0.043	0.013	0.019	NS
H19_21	237	0.952±0.025	0.941±0.037	0.012	0.021	NS
H19_22	243	0.976±0.018	0.966±0.037	0.010	0.031	NS

Data are expressed as mean and standard deviations and analyzed by t test. Bold means with a significant difference

β methylation differences between the cases and controls, P values:x2 test were analyzed by R package, P-Adj: t test P values adjusted for all individual CpG sites after Bonferroni correction (P < 0.00019 (0.05/257)), NS not significant

Table 4.

CpG sites of SNRPN with significant differences in methylation between patients and controls

Target	Position	Case	Control	β	P value	P-Adj
SNRPN_1	29	0.049±0.097	0.132±0.190	−0.084	0.006	NS
SNRPN_2	34	0.057±0.096	0.144±0.173	−0.087	0.002	NS
SNRPN_3	42	0.045±0.093	0.131±0.176	−0.087	0.002	NS
SNRPN_4	53	0.049±0.088	0.135±0.179	−0.086	0.003	NS
SNRPN_5	74	0.043±0.076	0.107±0.156	−0.064	0.008	NS
SNRPN_6	103	0.051±0.094	0.127±0.168	−0.076	0.004	NS
SNRPN_7	120	0.105±0.112	0.186±0.191	−0.081	0.012	NS
SNRPN_8	127	0.144±0.112	0.225±0.183	−0.080	0.007	NS
SNRPN_9	147	0.044±0.086	0.097±0.157	−0.053	0.028	NS
SNRPN_10	156	0.050±0.092	0.113±0.159	−0.063	0.012	NS
SNRPN_11	161	0.058±0.098	0.125±0.165	−0.068	0.008	NS
SNRPN_12	196	0.060±0.102	0.130±0.176	−0.070	0.012	NS

Data are expressed as mean and standard deviations and analyzed by t test. Bold means with a significant difference

β methylation differences between the cases and controls, P-Adj t test P values adjusted for Bonferroni correction (P < 0.00019), NS not significant

CpG site methylation of non-imprinted genes

Four non-imprinted genes related to male infertility (MTHFR, GSTM1, DAZL, and CREM) were selected for the differential methylation analyses. Among the 150 detected CpG sites, the methylation levels of nine CpG sites of three genes (MTHFR, DAZL, and CREM) were significantly different between the two groups (Table 5). Compared to the methylation in the control group CpG sites, the methylation of eight sites was decreased and that of only one site was increased in the genes of the impaired group. The average rates of DMSs in MTHFR, GSTM1, DAZL, and CREM were 2.2%, 0%, 19.05%, and 5.48%, respectively (Table 2). After performing Bonferroni correction for multiple testing, the methylation status of all individual CpG sites showed no significant differences between the groups (Table 5).

Table 5.

CpG sites of non-imprinting genes with significant differences in methylation between patients and matched controls

Target	Position	Case	Control	β	P value	P-Adj
MTHFR_1	116	0.005±0.005	0.008±0.008	−0.002	0.035	NS
DAZL_1	32	0.080±0.107	0.123±0.157	−0.043	0.043	NS
DAZL_2	37	0.086±0.107	0.128±0.152	−0.042	0.046	NS
DAZL_3	75	0.015±0.014	0.022±0.020	−0.007	0.009	NS
DAZL_4	84	0.133±0.102	0.175±0.150	−0.041	0.042	NS
CREM_1	29	0.0077±0.008	0.0113±0.0083	−0.0036	0.007	NS
CREM_2	39	0.009±0.007	0.012±0.0112	−0.003	0.031	NS
CREM_3	182	0.006±0.006	0.0034±0.004	0.0028	0.0087	NS
CREM_4	227	0.006±0.006	0.0087±0.0087	−0.00255	0.038	NS

Data are expressed as mean and standard deviations and analyzed by t test. Bold means with a significant difference

β methylation differences between the cases and controls, P-Adj t test P values adjusted for Bonferroni correction (P < 0.00019), NS not significant

Discussion

Normal methylation of sperm DNA is essential for fertilization and early embryo viability and affects the health of offspring in mammals [10, 11, 34]. Although defective sperm DNA methylation patterns have been associated with alterations in conventional semen parameters, there is a lack of studies that explore the correlation between site-specific DNA methylation and sperm integrity status. An improved understanding of sperm epigenetics is necessary not only for studying the potential causes of male infertility but also for raising concerns related to the progeny security associated with ART [17, 18, 35].

In this study, two imprinted genes, H19 and SNRPN, were selected for the DNA methylation study. We found that the imprinting genes were associated with relatively higher rates of DMSs among all the studied genes. After performing strict Bonferroni correction, there is still one significant difference observed in the methylation of H19 between the sperm samples showing damage to DNA integrity and the normal group. These results were consistent with those of previous studies regarding aberrant methylation patterns of imprinted genes and male infertility [19–21, 36]. SNRPN is located on chromosome 15q11.2 and is expressed by the paternal allele. Aberrant DNA methylation on the promoters of H19 and SNRPN has been associated with genetic imprinting disorders, such as Angelman and Prader-Willi syndromes [36]. The altered methylation levels of H19 and SNRPN present in samples showing sperm DNA damage indicated that they may act as a marker of further imprinting errors occurring during fertilization, which possibly lead to the onset of these syndromes. This may explain the relatively high incidence of imprinting-related genetic diseases in children conceived via ICSI. These data are consistent with recent studies suggesting that abnormal methylation of specific genes is associated with sperm DNA damage in human and boar sperms [29, 37].

In the present study, the methylation levels of nine CpG sites in three genes related to male infertility (MTHFR, DAZL, and CREM) were found to be significantly different between normal control and DNA damage groups. Several previous studies have reported a strong correlation between variation in sperm DNA methylation levels and male infertility [15, 37]. In general, our results are in agreement with those of previous studies that show a variation in sperm DNA methylation in individuals showing poor sperm integrity compared to that in normal individuals [13, 29]. These correlations suggest that epigenetic factors may influence sperm quality parameters, and DNA methylation may contribute to the defects observed in men with poor sperm integrity.

Sperm chromatin integrity is closely related to lifestyle and environmental factors [28]. Environmental and occupational exposure to bisphenol A, polycyclic aromatic hydrocarbons, ionizing radiation, organophosphate, and carbamate pesticides has been associated with increased sperm DFI [28]. Many studies have reported modifiable lifestyle risk factors, including smoking and obesity, have a detrimental effect on semen quality and sperm DNA integrity [38, 39]. In this study, the participants included in this study reported no exposure to environmental pollutants recently, but it is not easy to track back the real history of patients. The lifestyle and environmental exposure might result in reactive oxygen species (ROS) and defective sperm function, including impaired sperm DNA integrity [40]. Epigenetics provides a functional platform to evaluate the possible interconnections between sperm DNA damage and environmental factors [41]. The degree of sperm DNA fragmentation is positively correlated with ROS levels [42, 43]. Ejaculated sperms are susceptible to oxidative damage because their plasma membranes are rich in polyunsaturated fatty acids and lack the capacity to repair DNA damage [43]. Sperm DFI acts as an indirect indicator of oxidative stress levels in semen. Excessive levels of ROS can lead to nuclear and mitochondrial DNA damage, telomere shortening, and epigenetic alterations [43]. A similar phenomenon was observed in tumor cells. ROS-induced oxidative stress is associated with both aberrant hypermethylation of tumor suppressor gene promoter regions and global hypomethylation [44]. DNA oxidation structures, including 8-hydroxy-2'-deoxyguanosine and 5-hydroxymethylcytosine, cause DNA hypomethylation by inhibiting DNA methylation at nearby cytosine bases and via direct demethylation, respectively. Therefore, we inferred that the abnormal DNA methylation patterns in patients with sperm DNA damage might have been caused by the ROS mechanism. This result could also be physiologically explained in consideration with these reports and considering the role of MTHFR and GSTM1 in the defense against ROS. Imprinting genes may be more susceptible to the ROS modulation of sperm DNA methylation, which increases the epigenetic risk of male infertility. Notably, an advanced age is associated with high levels of oxidative stress and DNA fragmentation and a high risk of transmitting epigenetic dysfunctions to their offspring due to the altered status [45, 46].

This study has a few limitations. The study was mainly descriptive, and we were unable to conduct a functional study to verify the real relationship between DNA methylation patterns and sperm DNA integrity status. Currently, there is no reliable method to alter methylation at specific CpG sites. Studying the DNA methylation status of sperm before and after antioxidant treatment may be helpful for subsequent diagnosis and treatment. Further studies focusing on the specific mechanisms leading to epigenetic modifications are required.

In summary, we observed the aberrant methylation in the promoters of genes corresponding to imprinting, spermatogenesis, and the antioxidant defense system present in the sperm of subjects with impaired sperm DNA integrity status. Further studies are required to investigate the specific mechanisms leading to the epigenetic modifications and whether the alterations in DNA methylation in imprinting genes be transmitted to the offspring.

Funding

This work was supported by the Foundation of the Education Department of Anhui Province (KJ2019A0286) and Key Research and Development Plan of Anhui Province (202004j07020032).

Declarations

Ethics approval and consent to participate

This study was approved by the Ethical Review Board of The First Affiliated Hospital of Anhui Medical University and was conducted according to the Declaration of Helsinki principles (PJ2020-03-13). Written informed consents were obtained from all enrolled patients.

Competing interests

The authors declare no competing interests.