Background: The role of migration in shaping genetic diversity of viral populations has been well established. We show results of a large collaborative effort to assess epidemiological networks and dispersion routes between the HIV epidemics in the United States, Mexico and Central America, including Guatemala, Belize, Honduras, El Salvador, Nicaragua and Panama.
Methods: We analyzed 7943 HIV-1 pol sequences collected from Mexican and Central American national HIV sequence databases, and local databases located at the University of North Carolina and the University of California, San Diego. These sequences were aligned to the reference sequence HXB2. We then estimated Tamura-Nei (TN) 93 pairwise genetic distances among all sequences. Sequences that were ≤1.5% TN93 distance divergent were deemed similar enough to be potential transmission partners. We then constructed transmission clusters by linking sequences through shared potential transmission partners.
Results: In total, 2349 (30%) of sequences had a potential transmission partner in the network, comprising 846 distinct transmission clusters (range 2-53 members). Twenty five (3.0%) of these clusters included sequences isolated from at least two individuals in different countries, and six clusters included sequences isolated from both San Diego and North Carolina. Fourteen clusters included linkages between sequences isolated in San Diego and in Mexico. North Carolina sequences were linked to sequences from Guatemala in three clusters, Panama in two clusters and Mexico in two clusters; nearly all these sequences were from Latino immigrants. Below is a figure depicting all of the clusters identified in this analysis with nodes shaded by country of origin.
Conclusions: Our work underlines the importance of political and cultural barriers in HIV transmission among the USA, Mexico and Central America, with most transmission events occurring within countries. Nevertheless, the presence of some international connections was observed, reflecting possible important events influencing HIV diversity in the region. Continued network analyses have the potential to inform transnational HIV prevention efforts.
Figure 1. HIV transmission clusters in the USA, Mexico and Central America. Tamura-Nei 93 pairwise genetic distances were estimeted among 7943 pol sequences and clusters were constructed by linking sequences through shared potential transmission partners. Only sequences forming clusters are shown. Each node represents a single sequence colored by geographic region.