The emergence of HIV/AIDS in the Americas and beyond

Abstract

HIV-1 group M subtype B was the first HIV discovered and is the predominant variant of AIDS virus in most countries outside of sub-Saharan Africa. However, the circumstances of its origin and emergence remain unresolved. Here we propose a geographic sequence and time line for the origin of subtype B and the emergence of pandemic HIV/AIDS out of Africa. Using HIV-1 gene sequences recovered from archival samples from some of the earliest known Haitian AIDS patients, we find that subtype B likely moved from Africa to Haiti in or around 1966 (1962-1970) and then spread there for some years before successfully dispersing elsewhere. A "pandemic" clade, encompassing the vast majority of non-Haitian subtype B infections in the United States and elsewhere around the world, subsequently emerged after a single migration of the virus out of Haiti in or around 1969 (1966-1972). Haiti appears to have the oldest HIV/AIDS epidemic outside sub-Saharan Africa and the most genetically diverse subtype B epidemic, which might present challenges for HIV-1 vaccine design and testing. The emergence of the pandemic variant of subtype B was an important turning point in the history of AIDS, but its spread was likely driven by ecological rather than evolutionary factors. Our results suggest that HIV-1 circulated cryptically in the United States for approximately 12 years before the recognition of AIDS in 1981.

Fig. 1.

The abridged majority-rule consensus tree summarizing the results from the MrBayes analysis of complete env genes. The branch lengths represent the mean value observed for that branch among the postburnin sampled trees. Posterior probabilities are indicated for each node. As expected under a “Haiti-first” model, the non-Haitian subtype B strains are phylogenetically nested within an older and more diverse range of Haitian viral variants. The 11 sequences of the Trinidad and Tobago clade and the 96 sequences of the pandemic clade are schematically represented by the blue and yellow triangles, respectively. Haitian or Haitian-linked sequences are shown in green, with the archival sequences labeled in larger bold text. The unabridged tree is available as SI Fig. 4.

Fig. 2.

The consensus tree of the relaxed molecular clock analysis, with the Haitian archival sequences bulleted. The tips of the tree correspond to year of sampling, and the branch lengths reflect the mean of the posterior probability density. The posterior probability density for the TMRCA for subtype B in Haiti is depicted in dark green, and the 95% highest probability density (HPD) is shown by the horizontal bar and light-green shading. The TMRCA means and 95% HPDs for the other key nodes were as follows: subtype B/D ancestor = 1954 (1946–1961); subtype D ancestor = 1966 (1961–1971); Trinidad and Tobago subtype B ancestor = 1973 (1970–1976); and U.S./Canada subtype B ancestor = 1969 (1966–1972). This analysis resolves the position of the archival sequence H6 as basal to the pandemic clade. Under the relaxed molecular clock, P_{non-Haitian-origin} = 0.00003, P_{simultaneous-origin} = 0.0021, and P_{Haitian-origin} = 0.9979.