In the last decade, only a few novel genital human papillomavirus (HPV) types have been cloned. The last types reported 5 years ago were HPV89, 90 and 91.1 This may reflect that the most prevalent types were isolated and characterized earlier, and not many mucosal types are left to discover. Recently cloned types 92, 93, 94, 95 and 96 were isolated from skin lesions and belong to the cutaneous Beta Papillomavirus genus which has a less important influence on human health.2, 3, 4, 5 It was thus unusual that 2 independent research teams nearly simultaneously cloned and sequenced a novel HPV type that infects the anogenital mucosa, results that were presented by both groups at the 23rd International Papillomavirus Conference and Clinical Workshop 2006 in Prague.6, 7
We were greatly interested in the following article by Chen et al. which was published on line March 9, 2007 in this journal8, on molecular characterization of HPV97 isolate cloned from a healthy Costa Rican women. Their work clearly demonstrates the relatedness of HPV97 with types 18 and 45, and classifies HPV97 in alpha 7 species. They have also demonstrated that HPV97 oncoproteins degrade p53 and bind pRb in vitro. Since HPV97 was detected in only 2 of 1,796 women tested, the authors suggested that HPV97 prevalence in the general population is probably very low. Our results complement their findings by describing a similar HPV97 isolate that was cloned in our laboratory in 2006 as well as the prevalence of HPV97 in a cohort of HIV-seropositive men having sex with men.
The individual from whom the first isolate of HPV97 was obtained in our laboratory7 was an HIV-seropositive 48-year-old male living in Montreal who was born in Peru. The initial sample tested in May 2005 was a paraffin-embedded anal biopsy with a histopathological diagnosis of low-grade anal intraepithelial neoplasia (AIN-1). The 450-bp L1 amplicon generated with MY09-MY11 PCR from this sample displayed a unique, yet unknown RsaI restriction pattern. PCR-sequencing of the amplicon revealed a new HPV-like sequence having 86% of sequence homology with types 18 and 45. As the DNA extracted from formalin-fixed, paraffin-embedded tissues is usually partially degraded, the whole genome of the novel type was cloned from DNA extracted from a cell suspension obtained with an anal swab collected 3 months before the biopsy. Besides the novel type, this sample also contained HPV types 33, 58, 70, 71 and possibly 52, as assessed by Roche's Linear Array HPV Genotyping Test. To clone the whole genome of the novel virus from a mixture of several HPV types, we amplified human genomic and episomal viral DNA by rolling circle amplification (RCA) according to a previously published protocol.9 For the details of cloning, please refer to the electronic supplementary material.
Sequencing revealed a genome of 7,843 bp coding for the complete set of HPV early and late genes (Gene Bank accession number: EF436229). The virus has 81.9% and 83.7% homology with whole genomes of HPV types 18 and 45, respectively. Our isolate was almost identical (except for 12 single nucleotide variations) to the HPV97 sequence deposited in the NCBI GenBank (Chen and Burk, May 31, 2006, HPV97, accession number: DQ080080) (Table I). Ten variations were localized in coding regions, and 2 in the long control region (LCR). Two were synonymous variations, 4 caused conservative substitutions, and 6 others were nonconservative substitutions in the amino acid sequence.
|Nucleotide substitution||Position (NT)||Amino-acid change||Viral protein|
|C ⇒ T||556||A ⇒ V||E7|
|C ⇒ A||1,341||H ⇒ N||E1|
|G ⇒ A||2,192||None (L)||E1|
|G ⇒ A||3,201||M ⇒ I||E2|
|G ⇒ T||3,461||S ⇒ I||E2|
|3,461||V ⇒ L||E4|
|G ⇒ A||3,602||R ⇒ Q||E2|
|A ⇒ G||3,754||N ⇒ D||E2|
|G ⇒ A||5,202||D ⇒ N||L2|
|G ⇒ A||5,423||None (T)||L2|
|5,423||R ⇒ H||L1|
|A ⇒ T||6,397||T ⇒ S||L1|
|C ⇒ T||7,194||LCR region|
|G ⇒ C||7,280||LCR region|
Cheng et al.6, 8 detected HPV97 in only 2 of 1,796 tested women, 1,062 of whom came from the Women's Interagency HIV Study. Therefore, they concluded that “… HPV97 is nearly extinct. It appears that HPV97 has not thrived possibly related to its carriage by human genetic isolates or by a loss of fitness, such as decreased infectivity.” Our data do not support this statement, since HPV97 was detected by a highly specific PCR-sequencing assay (for details please, see the electronic supplementary material) in 24 anal swabs taken from 12 out of 247 (4.9%) HIV-seropositive men participating in the HIPVIRG cohort; the same clone was also present in 2 out of 9 formalin-fixed, paraffin-embedded anal biopsies taken from the same 12 men. This was done as part of the HIPVIRG cohort study of anal HPV infection and AIN, details of which have been described elsewhere (Gohy et al., Clin Infect Dis under revision). HIV-seropositive men having sex with men were all residents of metropolitan Montreal (QC, Canada) and had been recruited in the cohort if they were aged 18- to 65-years old, had a history of anal sex in the last year, a nadir CD4 cell count <500/mm3 and had been treated for at most 2 years with highly active antiretroviral therapy or were scheduled to initiate therapy. The demographic and clinical characteristics of the 12 men infected with HPV97 are presented in Table II. Although both individuals from whom HPV97 was initially cloned by both teams were born in Latin America, only one of the 11 additional Latino American participants of our cohort tested positive for this HPV type. HPV97 was detected in one individual without AIN and all the other HPV97-positive participants had biopsy-confirmed HPV-related lesions graded from AIN I to AIN III (Table II). HPV97 infection was persistent, since all 12 men tested positive for HPV97 also at the second visit, which followed by 6–8 month the initial HPV97 detection.
|Characteristic||No. of participants (%)|
|Country of birth|
|Latin America||2 (16.7%)|
|Mean age in years ± SD (range)||44.6 ± 7.0 (28.6–51.7)|
|History of receptive anal intercourse|
|Average no. of male sexual partners per year|
|Diagnosis of AIDS|
|Mean CD4 counts ± SD (range)||522 ± 340 (101–1,280)|
|CD4 count categories|
|CD4 < 200 cells/μl||2 (16.7)|
|CD4 200–400 cells /μl||73 (47.4)|
|CD4 > 400 cells /μl||10 (83.3)|
|CD4 not tested||3 (1.9)|
|Mean HIV load ± SD (range)||6,362 ± 10,096 (0–28,833)|
|No. of samples with HPV18||2 (16.7)|
|No. of samples with HPV45||3 (25.0)|
|No. of samples HPV + for types other than 97||12 (100.0)|
|Median no. of HPV types per anal swab sample (range)|
|Any type||6.0 (3–19)|
|Unknown risk||1.0 (0–4)|
|Cytology smear results|
|Histology results (highest grade)|
|AIN unknown grade||1 (8.3)|
All 12 HPV97-positive participants had multiple type infections with 3–19 types (median, 6) identified per anal swab sample with the Line blot assay (Table II). The presence of anogenital infections with multiple HPV types is common in HIV-seropositive men (Gohy et al., Clin Infect Dis under revision).10, 11, 12 To assess if HPV97 was involved preferentially in coinfections with some of the genital types, the ratios of observed rate and expected rate of coinfection of HPV97 with each of 36 genotypes were then calculated with 95% confidence interval (95% CI). HPV70 was the only genotype which was more frequently detected with HPV97 than the rate expected from the actual prevalence of each type in the cohort with a ratio of observed to expected rates of 3.40 (95% CI 1.11–7.94). The 95% CI of the ratios calculated with the other types all included one. HPV70 is classified by phylogenetic analysis as a high-risk type and by epidemiologic studies as a low-risk type. There is no explanation for a greater rate of HPV97 and HPV70 coinfections. The determinants of HPV97 infection were then investigated in our cohort. Compared to 13 HPV-negative men in the cohort, the 12 HPV97-infected men had a higher number of sexual partners in the last year (p = 0.008). They also had a higher mean annual number of sexual partners in their lifetime, although this difference did not reach statistical significance (p = 0.09). Greater recent exposure, measured by a greater number of recent partners, was more significant for HPV97 infection than lifetime exposure. CD4 counts, HIV load, a diagnosis of AIDS, age and smoking were not associated with HPV97 infection (p > 0.10). However, the small number of HPV97-positive individuals is a limitation of our analysis. The association between the number of sexual partners in the last year and HPV97 infection remained significant after controlling by logistic regression for age and CD4 counts (odds ratio (OR) 1.9, 95% CI 1.1–3.4).
The 4.9% prevalence rate in HIV-seropositive individuals in our study demonstrates that HPV97 is not a disappearing artefact of HPV evolution. Since HPV97 was rarely detected in the cervix of HIV-seropositive women,8 it would be worthwhile to investigate its prevalence in anal samples from HIV-seropositive and seronegative women, as well as HIV-seronegative men. Additional larger studies are needed to estimate HPV97 prevalence in immunocompetent and immunocompromised men and women in different countries.
In conclusion, we have cloned and characterized a new strain of HPV97 belonging to the oncogenic Papillomavirus alpha 7 species. Our clone differs from the original strain published by Cheng et al.6, 8 by a limited number of variations. This virus seems to have the potential to cause persistent anal infection in HIV-seropositive men. It always coinfected the anal canal with other HPV types, which will make the investigation of association with AIN difficult. Because of its close homology with highly oncogenic HPV18 and 45, a closer surveillance is needed in men and women who are sexually active and especially in those who are immunocompromised, at least in the geographical area where it was found to be prevalent in HIV-seropositive men.