A single early introduction of HIV-1 subtype B into Central America accounts for most current cases

Abstract

Human immunodeficiency virus type 1 (HIV-1) variants show considerable geographical separation across the world, but there is limited information from Central America. We provide the first detailed investigation of the genetic diversity and molecular epidemiology of HIV-1 in six Central American countries. Phylogenetic analysis was performed on 625 HIV-1 pol gene sequences collected between 2002 and 2010 in Honduras, El Salvador, Nicaragua, Costa Rica, Panama, and Belize. Published sequences from neighboring countries (n = 57) and the rest of the world (n = 740) were included as controls. Maximum likelihood methods were used to explore phylogenetic relationships. Bayesian coalescence-based methods were used to time HIV-1 introductions. Nearly all (98.9%) Central American sequences were of subtype B. Phylogenetic analysis revealed that 437 (70%) sequences clustered within five significantly supported monophyletic clades formed essentially by Central American sequences. One clade contained 386 (62%) sequences from all six countries; the other four clades were smaller and more country specific, suggesting discrete subepidemics. The existence of one large well-supported Central American clade provides evidence that a single introduction of HIV-1 subtype B in Central America accounts for most current cases. An introduction during the early phase of the HIV-1 pandemic may explain its epidemiological success. Moreover, the smaller clades suggest a subsequent regional spread related to specific transmission networks within each country.

Fig 1

Maximum likelihood phylogenetic analysis of HIV-1 group M subtype B pol sequences. The data set included 625 HIV-1B strains from six Central American countries, including Honduras (366 strains), El Salvador (116 strains), Nicaragua (52 strains), Costa Rica (38 strains), Panama (37 strains), and Belize (17 strains), and 807 non-Central America control sequences downloaded from GenBank and the Los Alamos HIV database, including 67 sequences from Central America neighboring countries and 740 sequences from the rest of the world (including 21 Haitian sequences). The tree was rooted by using subtype D sequences as the outgroup. The tree showing the tip labels with the full name of each strain is given in Fig. S1 in the supplemental material. (A) The maximum likelihood tree shows five significantly supported monophyletic clades (numbers inside the monophyletic clades correspond to approximate likelihood ratio test S-H-like values) formed essentially by Central American sequences. Branches are scaled in nucleotide substitutions per site according to the bar at the bottom of the figure and are colored to indicate the regional origin of each American strain according to the legend in the figure, while black branches represent strains from the rest of the world (Europe, Asia, and Africa). (B) Distribution of the Central American sequences in each of the clades.

Fig 2

Phylogenetic analysis of HIV-1 group M subtype B pol sequences using a subsample of the reference sequences. Shown is the Bayesian maximum clade credibility tree inferred from a data set including the 21 available Haitian sequences and a random subsample of sequences from North America (42 strains), Central America (21 strains), South America (42 strains), and Europe/Oceania (42 strains). The numbers along the monophyletic branches correspond to approximate likelihood ratio test S-H-like values. Branch lengths in nucleotide substitutions per site were scaled according to the bar at the bottom of each tree. Trees from four additional data sets with randomly selected sequences are given in Fig. S2 in the supplemental material.