Integrity of human sperm DNA assessed by the neutral comet assay and its relationship to semen parameters and clinical outcomes for the IVF-ET program.

OBJECTIVE: To explore potential relationships between sperm DNA integrity and both semen parameters and clinical outcomes.
METHODS: Semen analysis of 498 samples was performed according to the 2010 criteria of the World Health Organization. The sperm DNA fragmentation Index (DFI) of the semen samples was assessed using a neutral comet assay.
RESULTS: Sperm DFI showed a significant correlation with semen parameters, including the patient's age, sperm viability, motility, morphology, and number of leukocytes (p<0.05). The sperm DFI values for asthenozoospermic (15.2%), oligoteratozoospermic (18.3%), asthenoteratozoospermic (17.5%), and oligoasthenoteratozoospermic semen samples (21.3%) were significantly higher than that observed in normozoospermic semen samples (10.5%, p<0.05). A sperm DFI value of 14% was used as a threshold of sperm DFI in assessing whether DNA was highly damaged. In 114 IVF-ET cycles, the fertilization rate of the sperm DFI <14% group (70 cycles, 61.7%) was significantly higher than that observed for the ≥14% group (44 cycles, 55.3%), but there was no difference in the other clinical outcomes between the two groups. In the ≥14% group, the pregnancy rates of the ICSI cycles (40.0%) and half-ICSI (44.0%) were higher than conventional IVF cycles (30.7%), but the difference was not statistically significant.
CONCLUSION: Along with the conventional semen analysis, the sperm DFI assessed using the comet assay was shown to improve the quality of the semen evaluation. To evaluate the precise effect of ICSI on pregnancy rates in the patients who demonstrate high sperm DFI values, further study is necessary.

Introduction

Traditionally, the diagnosis of male infertility is based on microscopic assessment, including the concentration, motility, and morphology of sperm, but the results of this conventional semen analysis are insufficient as a diagnostic tool in male infertility. Recently, sperm DNA integrity has been regarded as a complementary diagnostic tool and biologic marker of male reproductive health and infertility.

It is well known that DNA damage in spermatozoa occurs during late spermatogenesis as a consequence of endogenous factors present in the testis/epididymis, or due to exogenous factors present after ejaculation. Potential mechanisms for generating DNA damage in sperm have been proposed, and include incomplete chromatin packaging [1], abortive apoptosis [2], oxidative stress by reactive oxygen species (ROS) [3], and an imbalance in endogenous hormone levels [4]. Unlike other cells, spermatozoa are more vulnerable to DNA damage because they do not have the capacity for DNA repair. Therefore, if sperm with damaged DNA are able to successfully fertilize an oocyte, it may result in pathologies such as infertility [3], childhood cancer [5], and imprinting diseases [6,7]. During fertilization, the oolemma acts as a selective barrier and prevents the penetration of anomalous spermatozoa. In ICSI cycles, the sperm selection mechanism of the oocytes is bypassed, which may increase the risk of transmitting damaged DNA. However, there is some debate on the actual risk of ICSI in regard to the selective barrier of the oolemma for the sperm with damaged DNA.

The most widely used methods to detect DNA integrity in individual spermatozoa are the sperm chromatin structure assay (SCSA) [8,9], the terminal deoxynucleotidyl transferase-mediated dUDP nick-end labeling (TUNEL) assay [10-12], and the alkaline [13,14] and neutral comet assay [15-17]. Each of these tests determines different aspects of DNA damage. The SCSA and TUNEL assays require the use of expensive instrumentation, such as flow-cytometry equipment for pre-cise assessment. Alternatively, the neutral comet assay is technically simpler to use, is more cost efficient, and has a comparably higher sensitivity for detecting DNA damage. Thus, the latter technique may be more suitable for routine measurements involving the quantification of double-strand (ds) break DNA damage in sperm cells [17].

A number of studies have shown a relationship between human sperm DNA damage and various semen parameters including sperm concentration, motility, morphology, and leukocyte concentration [18-23]. Sperm DNA damage has also demonstrated a negative effect on the mitochondrial activity of sperm [22,24], the efficiency of fertilization [25-27], embryo development [25,28,29], the chances of a successful pregnancy [26,29-31], and the ability to bring the preg-nancy to term [32,33]. Interestingly, a patient's age [17,25] and endogenous hormone level [34] were associated with the integrity of sperm DNA. In contrast, other research groups did not observe a close relationship between sperm DNA integrity and sperm morphology [35], fertilization rate, embryo development [36], or pregnancy outcome for ICSI [10]. The analysis of the sperm DNA integrity is important for improving the accuracy of semen analysis and the diagnosis of male infertility. However, various threshold values of sperm DNA integrity have been reported and differ due to both the variety of methods employed and the inherent variations in results among different research laboratories. Moreover, there is debate regarding the relationship between the sperm DNA integrity and the clinical outcome.

Here we have used a modified neutral comet assay to provide quantitative measurements of DNA damage in human sperm using a large sample size to more accurately approximate the optimal threshold value of sperm DNA integrity. We also explored the relationships among the sperm DNA integrity, semen parameters, and the clinical outcome in IVF programs.

Methods

This study was approved by our Institutional Review Board (the Mizmedi Hospital Research Ethics Committee).

1. Semen samples

A total of 498 semen samples were analyzed using the comet assay after undergoing a freeze-thaw cycle. The comet assay technique was performed on 114 semen samples obtained from male patients who were subjected to the IVF-ET program. The samples were used to ascertain the presence of a possible relationship between the sperm DNA integrity and clinical outcome in the IVF-ET program.

2. Semen analysis

Semen samples were obtained via masturbation after two to five days of ejaculatory abstinence and were analyzed within one hour of collection. After seminal liquefaction, a routine semen analysis was performed using computer-aided semen analysis (CASA; Medical Supply, Seoul, Korea) according to the 2010 World Health Organization (WHO) specifications. The semen samples with a concentration ≥15×106 sperm/mL, a motility ≥40%, a viability ≥58%, a normal morphology ≥4%, and leucocytes <1×106 cells/mL were considered parameters that characterize normal semen samples. A specific nomenclature is used by andrologists to describe abnormal semen samples encountered during semen analysis. In the present study, abnormal semen samples were characterized by the nomenclature according to the 2010 WHO criteria. An oligozoospermic semen sample means a reduced sperm concentration of <15×106 sperm/mL, an asthenozoospermic semen sample means a decreased sperm motility <40%, and a teratozoospermic semen sample means a decreased number of spermatozoa with normal morphology <4%. When a semen sample shows abnormal sperm concentration, motility, and morphology, we use combinations of these terms, such as "oligoasthenoteratozoospermic" semen sample.

3. Freeze-thaw cycle of semen samples

The semen samples were collected and cryopreserved in a quantity suitable for three days worth of analysis because the comet assay was performed twice in one week in the interest of efficiency. The semen samples were frozen using ultra-rapid freezing in the absence of a cryopreservative to minimize the cryodamage to the sperm DNA as described previously [15]. The entire semen sample was placed into cryovials and was frozen by static-phase vapor cooling (10 cm above the level of liquid nitrogen [LN2]; -80℃) for 10 minutes. The samples were then submerged into LN2 (-196℃) and stored until needed. Prior to experimentation, the samples were removed from LN2 and the cryovials were left at room temperature for 1 minute before being submerged into a 37℃ water bath. Once the samples were completely thawed, an equal volume of Ca2+ and Mg2+- free phosphate buffered saline (PBS) was mixed with the sample and centrifuged at 300 g for 5 minutes to remove seminal plasma and small particles. In the preliminary test, there was no significant difference in the sperm DNA integrity between the fresh semen samples and the freeze-thawed semen samples assessed using the comet assay.

4. Comet assay

The sperm DNA integrity was assessed using the modified neutral comet assay as described previously [37]. Comet assay was perfor-med using a comet assay kit (Trevigen, Gaithersburg, MD, USA). A mixture of semen (5 µL) and low melting agarose (45 µL) was layered onto a comet slide. The slide was transferred to a 4℃ refrigerator for 10 minutes. The slide was then submerged into a pre-cooled lysing solution (containing 2.5 M NaCl, 100 mM EDTA, 10 mM Tris, 1% Triton X-100, and 10 mM DL-Dithiothreitol; Sigma, St. Louis, MO, USA) for 2 hours prior to being placed in an electrophoresis buffer (500 mM NaCl, 100 mM Tris, 1 mM EDTA, and 0.2% DMSO; Sigma) for 20 minutes. After electrophoresis for 60 minutes at 10 V and 250 mA, the slide was washed with PBS and a neutralizing solution (50% Ethanol, 20 mM Tris, and 1 mg/mL Spermine; Sigma). After washing with PBS, the slide was stained with SYBR staining solution containing 10 mM Tris, 1 mM EDTA, and 0.01% SYBR Green (Trevigen). Images (400× magnification) were captured using a digital camera imaging system (Nikon digital sight DS-U1) attached to a fluorescence-inverted microscope (Figure 1).

Figure 1

Procedure of neutral comet assay. LMAgarose, low melting agarose.

5. IVF-ET program

Ovulation induction was achieved after several days of GnRH agonist suppression and was initiated through the administration of recombinant follicle stimulating hormone for a period of 9 to 12 days, followed by 10,000 U of hCG generally administered on the 10th day, when the largest follicular diameter was approximately 18 mm. Unlike the Comet assay, fresh semen was used for the IVF-ET program and motile sperm was prepared using a density gradient. The sperm washing buffer, fertilization medium, and embryo culture medium used for all protocols were GIII series medium (Vitrolife, Göteborg, Sweden). The embryo transfer step was performed on day three and the clinical pregnancy status was confirmed by observation of the uterine sac with a beating heart using ultrasonographic detection. A total of 114 IVF-ET cycles (36 conventional IVF cycles, 23 half-ICSI cycles, and 55 ICSI cycles) were involved in the present study. ICSI was used for the patients who showing male factors, a low number of mature eggs, or a low fertilization rate in the previous conventional IVF cycle. Half-ICSI was used for the patients who underwent the first conventional IVF cycle to prevent the total failure of fertilization. The IVF-ET cycles were divided into two groups according to the threshold of the sperm DNA integrity: 1) the <14% group and 2) ≥14% group.

6. Statistical analysis

The association between the sperm DFI and the semen parameters was explored using Pearson's correlation and a two-tailed Student's t-test. These statistical procedures were performed using the SPSS ver. 16.0 (SPSS Inc., Chicago, IL, USA). The clinical outcomes between the experiment and control groups were analyzed using the chi-square test; a p-value <0.05 was considered statistically significant.

Results

1. Correlation between sperm DFI and semen parameters

A correlation between sperm DFI and semen parameters was investigated in the 498 semen samples (Figure 2). The degree of correlation was assessed with the correlation coefficient (r) and p-value. Sperm DFI decreased significantly when the sperm viability (r=-0.605, p<0.001), motility (r=-0.395, p<0.001), rapid motility (r=-0.216, p<0.001) or morphology (r=-0.219, p<0.01) increased, but sperm DFI increased as the patient's age (r= 0.134, p<0.01) or leukocyte concentration in the semen increased (r=0.138, p<0.01). However there was no significant correlation between sperm DFI and sperm count (r=0.004, p=0.91) or semen volume (r=0.028, p=0.53).

Figure 2

Correlation between semen parameters and sperm DNA fragmentation index (DFI) assessed using neutral comet assay in 498 semen samples. p<0.05 was considered statistically significant, r=coefficient. NS, not significant.

2. Sperm DFI of subfertile semen samples

The sperm DFI of subfertile semen samples was compared with that of normal semen samples (Figure 3). The sperm DFI in asthenozoospermic (15.2±9.5%), oligoteratozoospermic (18.3±12.2%), asthenoteratozoospermic (17.5±13.4%), and oligoasthenoteratozoospermic semen samples (21.3±11.2%) were significantly higher than the percentage in normozoospermic semen samples (10.5±7.1%, p<0.05).

Figure 3

Comparison of sperm DNA fragmentation index (DFI) between subfertile semen samples and normal semen samples. aThe values (DFI) of subfertile semen samples were significantly higher than the value of normozoospermic samples. Values are presented as mean±SD. p<0.05 was considered statistically significant. N, normozoospermic; O, oligozoospermic; A, asthenozoospermic; T, teratozoospermic; OA, oligoasthenozoospermic; OT, oligoteratozoospermic; AT, asthenoteratozoospermic; OAT, oligoasthenoteratozoospermic.

3. Threshold of sperm DFI

On the basis of the potential correlation of sperm DNA integrity with semen analysis, we used a DFI of 14% as a threshold of the sperm DFI in assessing whether DNA was highly damaged. When the 498 semen samples were divided into two groups (<14% and ≥14% groups) according to the threshold value of DFI (14%), the mean values of all the semen parameters in the <14% group were significantly higher than those of the ≥14% group (p<0.05), with the exception of the sperm count (Table 1).

Table 1

Characteristics of 498 semen samples divided by the threshold value (14%) of sperm DFI

Values are presented as mean±SD.

p<0.05 was considered statistically significant.

DFI, DNA fragmentation index.

4. Association between sperm DFI and clinical outcome in IVF-ET cycles

Unlike with semen parameters, the sperm DNA integrity did not demonstrate a close correlation with the clinical outcome in 114 IVF-ET cycles (Table 2). The fertilization rate (61.7%) of the sperm DFI <14% group was significantly higher than the rate of the ≥14% group (55.3%, p<0.05). However, there were no differences in the percentage of good embryos and the pregnancy rate between the <14% group (32.2% and 35.7%) and the ≥14% group (33.6% and 38.6%). The abortion rate in the <14% group (5.7%) was lower compared to the ≥14% group (11.3%), but a statistically significant difference was not observed due to the small number of IVF-ET cycles.

Table 2

Association between the sperm DFI and clinical outcomes in 114 IVF-ET cycles

Values are presented as mean±SD, or number (%).

p<0.05 was considered statistically significant.

IVF-ET, in vitro fertilization-embryo transfer; DFI, DNA fragmentation index.

5. Association between sperm DFI and conventional IVF and ICSI cycles

Of the 114 IVF-ET cycles, 36 conventional IVF cycles, 23 half-ICSI cycles and 55 ICSI cycles were performed according to the semen quality, oocyte maturity, and previous IVF outcome of the patients (Figure 4). In the sperm DFI <14% group, the pregnancy rates of the ICSI, the half-ICSI, and the conventional IVF cycles were 40.0%, 35.7%, and 30.4%, respectively, whereas the pregnancy rates of the ≥14% group were 40.0%, 44.4%, and 30.7%, respectively. There was no significant difference in the pregnancy rates between the two groups due to the small number of the IVF-ET cycles.

Figure 4

Association between the sperm DFI and conventional IVF and ICSI cycles. There was no significant difference in the pregnancy rates between the two groups. DFI, DNA fragmentation index; IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection.

Discussion

Although many methodologies are available to assess DNA fragmentation in spermatozoa, we conclude that the neutral comet assay is a technically simple and cost-efficient method to measure the ds DNA breaks in sperm cells [17]. The neutral comet assay seems to be more sensitive to and reliable for the detection of ds DNA breaks than the alkaline comet assay because the alkaline conditions inherently induce the formation of DNA damage, a phenomenon that is not observed with the neutral comet assay [38].

In the present study, we had to freeze the semen samples for the comet assay because the comet assay was performed only twice a week. The semen samples were frozen using an ultra-rapid freezing method in the absence of a cryopreservative to minimize the cryodamage to the sperm DNA integrity as described previously by Duty et al [15]. They reported that flash-freezing in LN2 in the absence of a cryoprotectant most closely reproduced the results obtained using fresh semen samples. In our preliminary test, there was also no difference in the sperm DNA integrity between the fresh and the freeze-thawed semen samples assessed using the comet assay. This result is supported by the observation that DNA of the spermatozoa obtained from fertile men was found to be unaffected by the cryopreservation process [21].

In the present study, the sperm DFI was significantly increased in the subfertile semen samples compared to normal semen samples (Figure 3). Shamsi et al. [23] reported a similar statistically significant increase in DNA damage in oligozoospermic (20%), asthenozoospermic (24%), teratozoospermic (28%), and oligoasthenoteratozoospermic (OAT) semen samples (43%) when compared to control samples (8%). In addition, other authors reported that oligozoospermic [28], teratozoospermic [39], leukocytospermic [22], and OAT semen samples [14] showed a significant increase in the amount of DNA dam-aged sperm.

Various threshold values of sperm DFI have been proposed according to the various protocols and criteria used to determine them. The SCSA method has defined a 27% to 30% DFI as the point at which an individual is placed into the sub-fertile group [40], and the neutral comet assay method has suggested that ds DNA breaks are present in 15% to 25% of native sperm [16]. It is certainly reasonable that the threshold value of the sperm DNA integrity will be adjusted by ongoing studies. In the present study, the mean value of sperm DNA integrity in 498 semen samples was 14.1%, and 14% as a cutoff value of sperm DFI showed a significantly closer relationship between the sperm DFI and various semen parameters compared to any other cutoff values. Thus we decided to use 14% as a threshold of sperm DFI to assess whether DNA was highly damaged. Moreover, a previous study of the sperm chromatin dispersion test used 15% as the threshold of fertile sperm DFI [41], which supports our decision to use a threshold of 14%. The 498 semen samples were divided by the threshold into two groups (a <14% group and a ≥14% group). We observed that age correlates with an increase in the percentage of sperm with highly damaged DNA, as indicated in previous reports [17,19,25]. The mean values of sperm motility, morphology, and viability in the <14% group were significantly higher than the values in the ≥14% group. These observations were similar to the reports that lower levels of semen quality are characterized by higher levels of DNA damage [10,39].

Extensive sperm nuclear DNA fragmentation has been positively correlated with lower fertilization rates [41], poorer embryo development [25], lower pregnancy rates and an increased incidence of spontaneous abortion [31,34]. In contrast, some authors reported that there was no correlation between sperm DNA integrity and the clinical outcomes in ICSI cycles [1,10,36]. In the present study, we divided 114 IVF-ET cycles into the <14% (70 cycles) and the ≥14% group (44 cycles) based on the sperm DFI threshold value. Although the fertilization rate of the <14% group was significantly higher than the rate of the ≥14% group, there were no differences in the other clinical outcomes between the two groups (Table 2). The comparison of the clinical outcomes between the two groups may be an incomplete analysis because the IVF-ET cycles were divided into two groups by the threshold value that was only based on the semen parameters. Moreover, we could not perform a proper statistical comparison due to the small number of IVF-ET cycles. Therefore, many additional IVF-ET cycles are needed to more thoroughly explore the relationship between sperm DNA integrity and the clinical outcome, with a threshold based on the clinical criteria.

In ICSI cycles, the sperm with damaged DNA can be used for fertilization by bypassing the sperm selection mechanism of the oocytes. Nevertheless there was no correlation between the sperm DFI and the ICSI outcomes [1,10,36]. Furthermore, it was suggested that ICSI might help overcome the diminished pregnancy prognosis with high sperm DFI [40]. In the present study, although there was no statistical significance, the pregnancy rates of ICSI cycles were higher than conventional IVF cycles (Figure 4). This result suggests the possibility that ICSI may be better than conventional IVF for maintaining a stable pregnancy rate even in the high sperm DFI group. ICSI may be a better method to exclude sperm with high DFI values because most embryologists strive to select only those sperm showing sufficient motility and normal morphology for injection. As mentioned above, these semen parameters demonstrated a close relationship with the sperm DNA integrity. In conclusion, we confirmed that the modified neutral comet assay is a simple, valuable, and reliable method for assessing sperm DNA integrity. Sperm DNA integrity was correlated with semen parameters and 14% was the optimal threshold of the sperm DFI on the basis of these aforementioned characteristics, at least in this study. Therefore, sperm DNA integrity may be a good biomarker to evaluate semen quality and to diagnose male infertility in conjunction with semen analysis. Although we observed a relationship between sperm DNA integrity and the fertilization rate in the IVF-ET cycles, we did not observe a relationship with other clinical outcomes. Further studies on the relationship between sperm DNA integrity and clinical outcome are necessary.