Prevalence of human papillomavirus (HPV) in the semen of patients submitted to assisted reproductive technology treatment in a private clinic in Brazil

Objective: The aim of our study was to identify the prevalence of HPV in the semen of men submitted to ART treatment and look into the possible impacts of the virus on sperm parameters. Methods: Thirty-five patients treated for infertility from March to August 2016 were invited to join the study. Samples with a minimum concentration of 40x106 spermatozoa per milliliter were included in the study. After the evaluation of semen parameters, DNA extraction and PCR were performed to verify the presence of HPV by electrophoresis in 8% polyacrylamide gel. Results: Patient age ranged from 27 to 68 years (mean 39.2 years). Semen analysis showed a mean volume of 2.5mL; mean concentration of 58.9x106; and mean motility of 51.8%. HPV DNA was identified in seven semen samples from 25 patients (28%). Ten samples with DNA concentrations below 10ng/µL were excluded from the study due to poor amplification quality. There was no statistical difference in sperm concentration when HPV-negative and HPV-positive samples were compared (65.9x106 vs. 62.3x106; p=0.70). However, sperm motility was significantly higher in HPV-positive semen (65% vs. 46.6%; p=0.02). Conclusions: HPV prevalence was 28% in the semen of patients submitted to ART treatment. HPV-positive samples had statistically increased motility compared to negative samples (65% vs. 46.6%; p=0.02).

Semen analysis is the most important element in the assessment of male factor infertility. Complementary tests for infectious diseases such as Syphilis, HIV 1 and 2, Hepatitis B and C, HTLV I and II, Chlamydia trachomatis, Ureaplasma urealyticum, Mycoplasma hominis, Neisseria gonorrhoeae, aerobic bacteria, and Zika virus are performed routinely before ART treatment in Brazil. Nevertheless, more specific tests such as HPV are not performed.
The prevalence of HPV in semen has been estimated and varies according to the subjects and the country of study. Chan et al. (1994) found infection by HPV in 35.7% of the analyzed sperm samples. Foresta et al. (2010) found the lowest prevalence among studies involving infertile patients, with 10.2%. Perino et al. (2011) looked into a cohort of 199 sperm samples from infertile couples and found that 9.5% were infected by HPV. Schillaci et al. (2013) also studied infertile couples and reported a prevalence of 7.8% of infection by HPV in a set of 308 samples analyzed. Yang et al. (2013) described the presence of HPV in 107 of 615 (17.4%) infertile patients. Laprise et al. (2014) described a prevalence of 16% of HPV in the semen of men with unexplained infertility. Garolla et al. (2016) described the presence of HPV in 23.9% (54 of 226 patients) of infertile patients submitted to ART treatment.
Among the possibilities to treat infected sperm samples, Garolla et al. (2012) proposed various sperm washing techniques and showed that a modified swim-up plus heparinase III successfully removed HPV DNA form sperm surface. So far, there is no consensus as to whether HPV truly impairs sperm function. Moreover, the actual prevalence of HPV on semen is still unknown. Therefore, our study aimed to identify the prevalence of HPV in the semen of men submitted to ART treatment and to identify the possible impacts of the virus on sperm parameters.

MATERIALS AND METHODS
We performed a prospective study analyzing semen samples of men submitted to ART treatment at the ORI-GEN Center for Reproductive Medicine from March to August 2016. A total of 35 patients agreed and gave informed consent to join the study.
Samples were collected by masturbation after an ejaculatory abstinence period of two to five days. After 60-minute liquefaction, the samples were analyzed for macroscopic (volume, pH, viscosity) and microscopic (concentration, motility, and morphology) parameters. Concentration and motility analysis were performed using phase contrast microscopy (Nikon -Diaphoto -Japan), a differential cell counter, and a Makler chamber (Origio, Dinamarca).
After ICSI, DNA extraction from the semen samples was performed using a commercially available Wizard ® Genomic DNA Purification Kit (Promega) according to the manufacturer's protocol. Flushing was performed initially to remove cells that were not targeted by the present study. Approximately 40x10 6 spermatozoa from each patient were used.
The total extracted DNA was quantified in an automatic NanoVue Plus Spectrophotometer (GE) and the total DNA concentration was estimated considering that an optical density unit (OD) corresponds to a concentration of 0.050µg/µL of double stranded DNA (dsDNA) (Moore et al., 1997).
The detection of HPV genome in semen samples was performed with polymerase chain reaction (PCR) using primers GP5+ (5'-TTTGTTACTGTGGTAGATACTAC-3') and GP6+ (5'-GAAAAATAAACTAACTGTAAATCATATTC-3'), which amplify approximately 150 bp fragments of the L1 viral gene and allow the amplification of all HPV subtypes. PCR reactions were performed using the GoTaq ® Green kit (Promega) according to the manufacturer's protocol without modifications, following the concentrations and conditions for PCR.
PCR products were observed on 8% polyacrylamide gel impregnated with silver nitrate, developed and documented with images. Samples that amplified the fragment of 150 HPV base pairs were considered positive for HPV.

RESULTS
A total of 35 semen samples were studied. Patient ages ranged from 27 to 68 years (mean=39.2±8.36 years). All patients were evaluated for sexually transmitted diseases and one was positive for HIV. Ten samples with DNA concentration below 10ng/µL of DNA were excluded from the study due to losses in amplification quality. Twenty-five samples with concentration greater than 10ng/µL were submitted to PCR. Seven of the 25 semen samples were positive for HPV (Table 1, Figure 1).   Mean sperm concentration was 58.9x10 6 (16x10 6 to 153x10 6) . Mean progressive motility (AB) was 51.8% (26% to 81%).
When we compared the sperm concentrations of the HPV-positive and HPV-negative samples, we noticed that both presented similar results, with 62.3% and 65.9%, respectively (p=0.70). However, when we compared them for sperm motility, the mean motility of HPV-positive samples was statistically higher than the mean motility of HPV-negative samples (65% vs. 46.6%) (p=0.02) ( Table 2).
The differences in sperm motility and concentration between the HPV-positive and HPV-negative groups were evaluated through Fisher's exact test on SPSS (version 22.0.0.0).

DISCUSSION
Our study revealed a high prevalence of HPV (28%) in the semen of infertile couples submitted to ART treatment. Moreover, the HPV-positive samples did not present abnormal parameters according to the WHO (2010) criteria. Therefore, we could not confirm the existence of a correlation between HPV and sperm quality. To our knowledge, this is the first study held in Brazil about HPV prevalence in patients submitted to ART treatment.
The prevalence observed in our study was higher than the prevalence reported by Yang et al. (2013), who described the presence of HPV in 107 of 615 (17.4%) infertile patients, and by Laprise et al. (2014), in a meta-analysis that reported 16% of infection by HPV in patients submitted to ART treatment. Our results were similar to the numbers described by Garolla et al. (2016), in a study that reported the presence of HPV in 23.9% (54 of 226 patients) of infertile patients submitted to ART treatment.
The absence of a difference in sperm concentration between HPV-positive and HPV-negative samples was also described in previous studies (Foresta et al., 2010;Schillaci et al., 2013;Garolla et al., 2016). Our study demonstrated a significant increase in sperm motility in the HPV-positive group. Similar results were described by Connelly et al. (2001), where higher levels of motility were observed in sperm exposed in vitro to HPV types 16 (48±0.2%), 18 (47.5±0.1%), 31 (55±0.5%), and 33 (48.5±0.7%) compared to samples not exposed to HPV (37.5±0.3%). Other studies did not find differences between HPV-positive and HPV-negative groups. Rintala et al. (2004) analyzed the sperm parameters of 65 sperm donors and found no significant difference in sperm motility. Schillaci et al. (2013) showed that non-infected and infected sperm samples had similar progressive motility. Golob et al. (2014) studied 340 patient samples and did not find significant differences in motility or sperm concentration.
Differently from our study, Foresta et al. (2010) reported a statistical decrease in motility in HPV-positive semen (37.7%±16.8 versus 53.7%±18.2%, p<0.05), while other sperm parameters were similar. Other studies by Foresta et al. (2011a;2015) also described decreased sperm motility in HPV-positive samples. Yang et al. (2013) observed that even among controls (fertile patients) HPV-positive samples had significant impairments in sperm motility (32.25%±10 vs. 39.22%±12.15, p<0.01) and morphology (8.51±4.21 vs. 13.01±4.50, p<0.01). Foresta et al. (2015) performed a meta-analysis that included 30 studies and 5325 patients. The authors described a higher prevalence of HPV with decreased semen motility among patients with unexplained infertility, in addition to a strong relationship between asthenozoospermia and HPV regardless of virus subtype. Garolla et al. (2016) also reported decreased sperm motility in patients with HPV compared to patients without HPV (25.9±16.2% vs. 34.3±17.7%). The authors suggested this was due to the presence of antibodies on sperm surface, since 40.7% of HPV-positive samples had antisperm antibodies versus 10.5% in non-infected samples. Lai et al. (1997) were the first to describe asthenozoospermia in sperm samples of HPV-positive patients with HPV types 16 and 18. Lee et al. (2002) also observed that HPV types 6/11, 16, 18, 31 caused decreases in sperm motility. Some HPV types may have direct impact on sperm motility. Lee et al. (2002) showed that type 33 did not affect semen parameters. Therefore, it is important to find the type of HPV present in sperm samples.
In the present study, the causes of infertility in the HPV-positive group were ovarian factor and unexplained infertility. Due to the limited size of our sample, we were unable to find an association between the presence of HPV and unexplained infertility. Previous studies attempted to correlate the presence of HPV and sperm function impairment. Several described increased sperm fragmentation (Connelly et al., 2001;Lee et al., 2002;Hong et al., 2013). Foresta et al. (2011a) reported that sperm infected with HPV were less able to penetrate oocytes since the virus was located on sperm surface. Embryo development may also be affected by HPV (Foresta et al., 2011b;Gomez et al., 2008). Garolla et al. (2012) proposed various sperm washing techniques to treat infected sperm samples and showed that a modified swim-up plus heparinase III successfully removed HPV DNA form sperm surface. Garolla et al. (2013) also showed that after 24 months antisperm antibodies disappeared and sperm motility improved.
Further studies are required with greater numbers of sperm samples to identify HPV virus subtypes and quantify the viral load in order to establish whether HPV might affect sperm parameters or impair fertility. Moreover, it is important to elucidate the mechanism by which HPV might affect sperm parameters and the possible implications in IVF outcomes.