Reviewing the history of HIV-1: spread of subtype B in the Americas

Abstract

The dispersal of HIV-1 subtype B (HIV-1B) is a reflection of the movement of human populations in response to social, political, and geographical issues. The initial dissemination of HIV-1B outside Africa seems to have included the passive involvement of human populations from the Caribbean in spreading the virus to the United States. However, the exact pathways taken during the establishment of the pandemic in the Americas remain unclear. Here, we propose a geographical scenario for the dissemination of HIV-1B in the Americas, based on phylogenetic and genetic statistical analyses of 313 available sequences of the pol gene from 27 countries. Maximum likelihood and bayesian inference methods were used to explore the phylogenetic relationships between HIV-1B sequences, and molecular variance estimates were analyzed to infer the genetic structure of the viral population. We found that the initial dissemination and subsequent spread of subtype B in the Americas occurred via a single introduction event in the Caribbean around 1964 (1950-1967). Phylogenetic trees present evidence of several primary outbreaks in countries in South America, directly seeded by the Caribbean epidemic. Cuba is an exception insofar as its epidemic seems to have been introduced from South America. One clade comprising isolates from different countries emerged in the most-derived branches, reflecting the intense circulation of the virus throughout the American continents. Statistical analysis supports the genetic compartmentalization of the virus among the Americas, with a close relationship between the South American and Caribbean epidemics. These findings reflect the complex establishment of the HIV-1B pandemic and contribute to our understanding between the migration process of human populations and virus diffusion.

Figure 1. Bayesian tree of 263 HIV-1 subtype B sequences of the pol gene from 25 American countries.

Majority-rule Bayesian consensus tree of 268 HIV-1 subtype B pol sequences isolated in 25 countries in the Americas. The outgroups are subtypes C and D. Branches are colored according to the sample origin. Orange branches represent isolates from the Caribbean, the green branches represent South American isolates, and the blue branches represent isolates from North America. Posterior probabilities are shown for the key nodes. The tips of the tree contain isolate information regarding the subtype, country, year of isolation, and GenBank accession number. Abbreviations of countries are as follows: AR, Argentina; AG, Antigua and Barbuda; BH, Bahamas; BR, Brazil; BW, Botswana; CA, Canada; CD, Democratic Congo; CM, Cameroon; CO, Colombia; CU, Cuba; DO, Dominican Republic; EC, Ecuador; ET, Ethiopia; GD, Grenada; GY, Guyana; HT, Haiti; JM, Jamaica; LC, Santa Lucia; SR, Suriname; TT, Trinidad and Tobago; TZ, Tanzania; UG, Uganda; US, United States; VC, St Vincent; VE, Venezuela.

Figure 2. Part of the Bayesian tree of 263 HIV-1 subtype B sequences of the pol gene (Pandemic Clade B).

This is the full version of the collapsed clade presented in the Figure 1. The orange branches represent the isolates from the Caribbean, the purple branches represent Central American isolates, the green branches represent the South American isolates, and the blue branches represent the isolates from North America. Abbreviations of countries are as follows: AR, Argentina; AG, Antigua and Barbuda; BH, Bahamas; BR, Brazil; CA, Canada; CO, Colombia; CU, Cuba; EC, Ecuador; HN, Honduras; HT, Haiti; JM, Jamaica; MX, Mexico; PA, Panama; PE, Peru; TT, Trinidad and Tobago; US, United States; UY, Uruguay; VE, Venezuela.

Figure 3. Part of the Bayesian tree of 263 HIV-1 subtype B sequences of the pol gene (Pandemic Clade C).

Part of the 50% majority rule consensus tree constructed from the Bayesian MCMC (BEAST) analysis. This is the full version of the collapsed group shown in Figure 1, indicating the evolutionary relationships among the sequences in the pandemic clade. The orange branches represent the isolates from the Caribbean, purple branches represent isolates from Central America, the green branches represent the South American isolates, and the blue branches represent the isolates from North America. The branches are not drawn to scale. Abbreviations of countries are as follows: AR, Argentina; AG, Antigua and Barbuda; BH, Bahamas; BR, Brazil; BW, Botswana; CA, Canada; CD, Democratic Congo; CM, Cameroon; CO, Colombia; CU, Cuba; DO, Dominican Republic; EC, Ecuador; ET, Ethiopia; GD, Grenada; GY, Guyana; HT, Haiti; JM, Jamaica; LC, Santa Lucia; SR, Suriname; TT, Trinidad and Tobago; TZ, Tanzania; UG, Uganda; US, United States; VC, St Vincent; VE, Venezuela.

Figure 4. Genetic Structure of 308 HIV-1 subtype B sequences from American Countries.

Synthetic map illustrating the distributions and geographic origins of strains isolated in the Americas and the genetic structure among continents and countries. (a) Countries of sample isolation are colored according to geopolitical regions, comprising South America, Central America (including the Caribbean), and North America. No isolates from the gray-colored countries were included in this study. Countries located in Central America are represented by numbers: (1) Bahamas, (2) Puerto Rico, (3) Antigua and Barbuda (4) Santa Lucia, (5) St Vincent, (6) Grenada, (7) Jamaica, (8) Honduras, and (9) Panama. The red dotted lines represent Φ_ST estimates between continents. South American sequences are genetically intermediate between those of Central America and North America. (b) Nonmetric multidimensional scaling plot of the Φ_ST estimates among 22 South American countries. Dimension 1 separates the 308 isolates by country.