Background: HIV-1 is subdivided into four groups: M, O, N and P. Group M is responsible for the pandemic, N and P are found in very few patients, and group O is endemic in Cameroon where it represents ~1% of HIV infections. Group O genetic diversity and evolution remain poorly characterised. Previous studies estimated its emergence to be as early as group M and have proposed different nomenclature systems. Here we fully investigate group O evolution, using an extensive dataset comprising sequences sampled from France, Cameroon and Gabon.
Methods: 190 HIV-O patients were sampled in France, Cameroon and Gabon between 1987 and 2012. Viral sequences from three regions of the genome were concatenated (total: 2012 base pairs) and analysed using phylogenetic and Bayesian population genetic methods to characterize evolutionary history.
Results: The tree topology was atypical, with a predominant clade emerging from a broad and genetically diverse base population. The presence of the Y181C mutation, naturally conferring resistance to non-nucleoside reverse transcriptase inhibitors, was significantly associated to the emergent clade.
The year of the most recent common ancestor (MRCA) was estimated to be around 1930, close to that of group M and consistent with previous estimates. Bayesian skyline analysis indicates group O diversification has gone through two exponential growth phases. The first phase, around 1950, resulted in widespread genetic diversity, while the second, from the late 1970s to the early 1990s, gave rise to the emergent clade.
According to the historical context in Cameroon, the first wave of group O expansion might have been favoured by iatrogenic routes of transmission between 1940 – 1960, while the second wave would rather be associated to the development of urbanization.
Conclusions: While groups O and M share a similar age, group O has remained largely confined to Cameroon. This has resulted in broad genetic diversity and intermixing of sub-populations, making it impossible to classify subtypes using a similar system as is used for group M. However, two subgroups can be observed corresponding to two successive phases of diversification. The viral properties of the two subgroups need to be investigated, to better understand why the now predominant clade was the most recent to emerge.