Human papillomavirus infection in oral fluids of HIV-1-positive men:prevalence and risk factors

Human papillomavirus is one of the most common sexually transmitted diseases worldwide. The natural history of oral HPV infection is unclear, and its risk factors have not been explored. Immunocompromised individuals, as exemplified by HIV patients, are at high risk for HPV-related diseases. The mean of this study is to determine the prevalence ofHPV in the oral tract of HIV-1-positive male subjects and its association with risk factors. A total of 283 oral wash samples from HIV-1-positive men were tested. The oral fluid samples were used for DNA extraction and conventional PCR amplification; HPV genotyping was performed by hybridization. HPV genotyping revealed that nine samples (3.5%) were positive for HPV DNA; the major high-risk HPV types identified were 51 and 66. Worldwide studies have shown a variable prevalence of oral HPV. The diversity of genotypes and the high prevalence of multiple infections in HIV-infected subjects can be better explained by the effects of HIV-induced immunosuppression. The most important risk factors are unprotected sexual intercourse, but other factors for this infection have been described elsewhere including smoking, age and HIV-positive serostatus. In this study, smoking was the most important risk factor for acquiring oral HPV in HIV-1-infected subjects in Brazil.

To date, limited information is available regarding the natural history of oral HPV infectionand which risk factors are associated with it. Similarly, little is known about the association between HPV and HIV infections in men 18 . The aim of this study was to measure the prevalence of HPV in the oral tract of HIV-1-seropositive men and to evaluate major risk factors for transmission.

Methods
Study population. Two-hundred and eighty-three HIV-1-seropositive male patients referred to the HIV-Out Clinic ADEE3002, Dermatology Department of hospital das Clinicas, University of São Paulo medical school -HC-FMUSP and the Institute of Infectious Diseases ''Emilio Ribas'' from December 2011 to May 2013were included in the study. All patients appeared to have healthy oral mucosa. The mean age was 43.81 years (22-72 years). After signing an informed consent form, patients answered a questionnaire containing information on their social-behavioral characteristics. The study was approved by the respective ethical boards from the Institute of Tropical Medicine of São Paulo and the from Hospital das Clinicas of FMUSP.
Sample collection. Oral washes were performed as follows. Patients wereinstructed to performance oral rinse with 10 ml of sterile saline solution (0,9% sodium chloride -Baxter International), and the rinses were submitted to sample collection. After collection, samples went through a process of washing and centrifugation. They were then aliquoted and stored at 220uC for later analysis.
Precautions for contamination prevention. Strict procedures were followed to avoid false-positive reactions due to contamination. DNA extraction, reagent preparation and addition of sample DNA were carried out separately. Each area has its own dedicated equipment and every test included a negative control amplification containing distilled water without a DNA sample. All working surfaces were decontaminated with sodium hypochlorite and alcohol (70%), before and after use.
Laboratory methods. Samples were subjected to a DNA extraction process using a commercially available kit (Illustra Tissue and Cells GenomicPrep Mini Spin kit, Easton Turnpike, Fairfield, USA) and, run according the manufacturer's instructions. After this procedure, the DNA was stored at 280uC for later analysis.
Beta-globin PCR was performed for all samples as a control for DNA quality. The protocol used was previously standardized in the laboratory where the research was conducted. PCR was performed in a final reaction volume of 50 ul, containing 5 ul of template DNA, 10 3 PCR buffer, 50 mM of MgCl 2 , 10 mM of dNTPs, 5 U/ml of Taq DNA polymerase and 10 pmol of each primer. The PCR conditions were as follows: pre-heated at 95uC for 4 minutes followed by 35 cycles of 1 minute at 94uC, 1 minute at 55uC and 1 minute at 72uC, with a final extension of 7 minutes at 72uC.
HPV detection was carried out with a conventional PCR using the primers MY09/ 11. The protocol used was previously standardized in the laboratory where the research was conducted. PCR was performed in a final reaction volume of 50 ul, containing 5 ul of template DNA, 0.05 U/ul of Taq DNA polymerase, 4 mM of MgCl 2 , 0.4 mM of each dNTP, of PCR Mastermix(Thermo Scientific/Walthan -MA) and a final concentration 4 pmol of the MY09/11 primers. The PCR conditions were as follows: preheatedat 95uC for four minutes followed by 40 cycles of 1 minute at 94uC, 1 minute at 55uC and 1 minute at 72uC, with a final extension of 7 minutes at 72uC. The presence of HPV DNA was confirmed using electrophoresis on a 2% agarose gel containing ethidium bromide.
Finally, a commercially available kit was used for HPV genotyping (

Results
Laboratory data and behavioral characteristics of all 283 patients were analyzed. All patients in this study were male and HIV-infected (data not shown), 83% were using HAART, 82.3% had T CD4 cell counts $ 350 cell/mm 3 and 78.8% had undetected HIV viral loads (less than 50 copies). Most of the patients had a college degree (45.2%), were single (53.4%), of white ethnicity (60.8%) and did not use drugs (62.5%). We also evaluated alcohol intake and tobacco use: 54.8% ingested some alcohol and 30.4% were current smokers, 3.9% of the patients had smoked for months and 26.5% had smoked for years.
In regards to sexual exposure, 41% reported having had sex with men in the last year, 73.9% reported using a condom with a casual partner and 53.4% reported having engaged in active oral sex. Furthermore, 54.3% had passive oral sex and 52.6% had active anal sex. Approximately 75.4% had no discharge, and 80.2% had no warts or wounds on the genital site.
HPV DNA was found in ten of the 283 samples (3.5%) (data not shown). The mean age of the HPV1 group was 42 years (26-56 years), 70% were using HAART, 80% had T CD4 cell counts $ 350 cells/mm 3 and undetectable HIV viral loads. Most of the patients were single (80%), 50% had a college degree, most were of white ethnicity and did use drugs. Among these patients, 80% are smokers and 60% smoke more once a day (Table 1).
For sexual exposure, all of the HPV-positive patients reported having sex with men in the last year, with active anal sex and using a condom with casual partners, 80% had active oral sex, 40% had passive oral sex, 80% had no discharge or wounds and 90% had no warts at the genital site.
HPV DNA was genotyped, identify and classifythe subtypes in each sample (table 2). In the high-risk group, the most highly prevalent type was HPV-66 (60%) followedHPV-51 (40%). The HPV-58 type was found in three samples (30%) and the HPV-16, 56 and 69 types were each found once in different samples.
In the low-risk group, the HPV-6 and 83 types were found in two samples (20%) and the HPV-44, 62, 67, 72 and 84 types were each found once in different samples. Three samples had multiple highrisk subtypes, and two samples had multiple low-risk subtype.
Variables such as sexual exposure, alcohol consumption, cigarette smoking, HIV serostatus, HIV viral load, CD4 cell count and HAART use were assessed as possible risk factors for oral HPV (table 3). The bivariate analysis was performed comparing the distribution of factors between HIV-infected men HPV-HIV coinfected men.
The variables that had significant results based on a p value less than 0.05 were included in the multivariate logistic regression analysis to determine which factors are a risk for oral HPV infection (

Discussion
Worldwide studies have shown a variable prevalence of oral HPV in healthy individuals 19 . This variability may be due to factors such as the difference in the local mucosal immunity, saliva flow and immu-  noglobulins in the oral cavity, different methods of sample collection and analysis and propensity to micro trauma injuries 20,21 .
In HIV-infected subjects, oral HPV infection is more aggressive and more highly prevalent compared with a healthy population 16,19,22 . A comparative study including HIV-infected patients and healthy subjects foundthat 25.3% of HIV patients had an HPV infection in the oral tractcompared with 7.6% of healthy subjects 23 . In the United States, a study was conducted with 190 HIV-positive men, and found that 6.3% them had detectable HPV in their oral fluid samples 23 . On the other hand, a study conducted in theUnited States and Spain with 166 patients showed a prevalence of 21% of oral HPV in HIV-positive MSM individuals, of which 11.1% were low-risk types and 66.6% were high-risk types 24 . In our study, oral HPV infection was detected in 3.5% of HIV-positive patients, in accordance withboth previous results and with the variability found in the prevalence of oral HPV in several studies 9 .
The diversity of genotypes and the high prevalence of multiple infections in HIV-positive subjects can be better explained by the effects of HIV-induced immunosuppression 25 . Our results demonstrate that most HPV-infected patients had more than one highrisk type.
When comparing the anogenital and oral genotypes, some types could be found at both sites. Regarding the low-risk groups, HPV-6 and 11 are the most prevalent 26 . In our study, HPV-6 was found in 20% of the samples. In the high-risk group, HPV-16 and 18 were the most prevalent at the anogenital site followed by HPV-31, 33, 39, 45, 51, 52, 53, 58, 59, 66 and 73 27 . In our study, several high-risk anogenital types were found, especially HPV-66.
The most important classical risk factors are unprotected sexual intercourse and multiple sexual partners 20 . Several studies have suggested that a high number of lifetime sexual partners, and practices such as kissing, oral-penile sex and hand warts are predictive of oral HPV infection 20,28 .
Other factors for this infection have been described elsewhere, including smoking and HIV-positive serostatus 20 . Although some studies indicate that there is conflicting evidence on the relationship between cigarette smoking and HPV infection, several studies suggested that recent cigarette smoking was associated with a significantly increased risk of oral HPV 29,30 . According to Read and coworkers 14 , smoking causes oral epithelial thickening, periodontal disease and epithelial abrasion, thereby increasing the risk of HPV infection.
In this report, we examined the oral tract as a site associated with the presence of HPV infection. After controlling for potential confounding variables, the only risk factor remaining significantly associated with HPV infection in our patients was cigarette smoking (p , 0.001).
Studies suggest that HIV-infected individuals may have a higher risk of developing oropharyngeal cancer as well other HPV-associated cancers [31][32][33] . Some studies, indicate that highly active therapy (HAART) does not reduce the prevalence of oral HPV and that CD4 cell count increases the incidence, progression and persistence of HPV-induced lesion 16,34 .
In our study, there was no substantial difference between these factors, because the majority of our patients had undetectable HIV viral loads and their CD4 cell counts were greater than 350 cells/ mm 3 .
We could not determine whether HPV in the oral cavity of our patients had caused any related disease, such as head and neck squamous cell carcinoma (SCCHN) associated with cigarette smoking. However, it is known that these HPV oral carriers are potential transmitters of the virus.
Finally, it is important to acknowledge that most HPV types found in our patients are high-risk types but that, those types are unfortunately not treated by the currently available marketed vaccines. As a highlight, more studies are needed to enable a better understanding of the association between oral HPV infection and its risk factors in HIV-positive people.