Worldwide genomic diversity of the human papillomaviruses-53, 56, and 66, a group of high-risk HPVs unrelated to HPV-16 and HPV-18.

Among more than 200 human papillomavirus (HPV) types presumed to exist, 18 "high-risk" HPV types are frequently found in anogenital cancer. The best studied types are HPV-16 and 18, which are only distantly related to one another and form two separate phylogenetic branches, each including six closely related types. HPV-30, 53, 56, and 66 form a third phylogenetic branch unrelated to HPV-16 and 18. Worldwide comparison of HPV-16 and 18 isolates revealed a distribution of variant genomes that correlated with the geographic origin and the ethnicity of the infected cohort and led to the concept of unique African, European, Asian, and Native American HPV-16 and 18 variants. Here, we address the question whether similar phylogenies are found for HPV-53, 56, and 66 by determining the sequence of the long control regions (LCR) of these HPVs in samples from Europe, Asia, and Africa, and from immigrant societies in North and South America. Phylogenetic trees calculated from point mutations and a few insertions/deletions affecting 2-4.2% of the nucleotide sequences were distinct for each of the three HPVs and divergent from HPV-16 and 18. In contrast to the "star-phylogenies" formed by HPV-16 and 18 variants, 44 HPV-53 isolates represented nine variants, which formed two deep dichotomic branches reminiscent of the beginning split into two new taxa, as recently observed for subtypes of HPV-44 and 68. A total of 66 HPV-56 isolates represented 17 variants, which formed three branches preferentially containing European, Asian, and African variants. Variants of a fourth branch, deeply separated from the other three, were characterized by a 25 bp insertion and created a dichotomy rather than star-like phylogeny. As it contained isolates from cohorts in all continents, it may have evolved before the spread of humans into all continents. 18 of 31 HPV-66 isolates represented the prototype clone, which was found in all parts of the world, while the remaining 13 clones formed 11 branches without any geographic association. Our findings confirm the notion of a quantitatively limited genomic diversity of each HPV type with some correlation to the geographic origin of the sample. In addition, we observed in some variants of these three HPV types mutations that affect the amino acid sequence of the E6 oncoproteins and the L1 capsid protein, supporting the possibility of immunogenic and oncogenic diversity between variants of any HPV type.

Genomic diversity of human papillomavirus-16, 18, 31, and 35 isolates in a Mexican population and relationship to European, African, and Native American variants.

Cervical cancer, mainly caused by infection with human papillomaviruses (HPVs), is a major public health problem in Mexico. During a study of the prevalence of HPV types in northeastern Mexico, we identified, as expected from worldwide comparisons, HPV-16, 18, 31, and 35 as highly prevalent. It is well known that the genomes of HPV types differ geographically because of evolution linked to ethnic groups separated in prehistoric times. As HPV intra-type variation results in pathogenic differences, we analyzed genomic sequences of Mexican variants of these four HPV types. Among 112 HPV-16 samples, 14 contained European and 98 American Indian (AA) variants. This ratio is unexpected as people of European ethnicity predominate in this part of Mexico. Among 15 HPV-18 samples, 13 contained European and 2 African variants, the latter possibly due to migration of Africans to the Caribbean coast of Mexico. We constructed phylogenetic trees of HPV-31 and 35 variants, which have never been studied. Forty-six HPV-31 isolates from Mexico, Europe, Africa, and the United States (US) contained a total of 35 nucleotide exchanges in a 428-bp segment, with maximal distances between any two variants of 16 bp (3.7%), similar to those between HPV-16 variants. The HPV-31 variants formed two branches, one apparently the European, the other one an African branch. The European branch contained 13 of 29 Mexican isolates, the African branch 16 Mexican isolates. These may represent the HPV-31 variants of American Indians, as a 55% prevalence of African variants in Mexico seems incomprehensible. Twenty-seven HPV-35 samples from Mexico, Europe, Africa, and the US contained 11 mutations in a 893-bp segment with maximal distances between any two variants of only 5 mutations (0.6%), including a characteristic 16-bp insertion/deletion. These HPV-35 variants formed several phylogenetic clusters rather than two- or three-branched trees as HPV-16, 18, and 31. An HPV-35 variant typical for American Indians was not identifiable. Our research suggests type specific patterns of evolution and spread of HPV-16, 18, 31, and 35 both before and after the worldwide migrations of the last four centuries. The high prevalence of highly carcinogenic HPV-16 AA variants, and the extensive diversity of HPV-18, 31, and 35 variants with unknown pathogenic properties raise the possibility that HPV intra-type variation contributes to the high cervical cancer burden in Mexico.