History and origin of the HIV-1 subtype C epidemic in South Africa and the greater southern African region

Abstract

HIV has spread at an alarming rate in South Africa, making it the country with the highest number of HIV infections. Several studies have investigated the histories of HIV-1 subtype C epidemics but none have done so in the context of social and political transformation in southern Africa. There is a need to understand how these processes affects epidemics, as socio-political transformation is a common and on-going process in Africa. Here, we genotyped strains from the start of the epidemic and applied phylodynamic techniques to determine the history of the southern Africa and South African epidemic from longitudinal sampled data. The southern African epidemic's estimated dates of origin was placed around 1960 (95% HPD 1956-64), while dynamic reconstruction revealed strong growth during the 1970s and 80s. The South African epidemic has a similar origin, caused by multiple introductions from neighbouring countries, and grew exponentially during the 1980s and 90s, coinciding with socio-political changes in South Africa. These findings provide an indication as to when the epidemic started and how it has grown, while the inclusion of sequence data from the start of the epidemic provided better estimates. The epidemic have stabilized in recent years with the expansion of antiretroviral therapy.

Figure 1. The estimated time to the most recent common ancestor for the southern African region as well as for several southern African countries, which were co-estimated in BEAST from the four different data sets.

The dot represents the mean estimated date and the bars represent the 95% confidence interval (95% HPD).

Figure 2. Dynamics growth of the southern African and South African subtype C HIV-1 epidemic as was inferred from Bayesian Skyline Plot reconstruction.

On the top are the BSP reconstructions from gag p24 sequence data and at the bottom are the BSP reconstructions from pol sequence data. Blue graphs represent sequence data from the entire southern African region while red graphs represents data containing only South African isolates. Solid lines depict the mean estimated Ne through time while shaded areas correspond to the 95% highest posterior density intervals.

Figure 3. Maximum Likelihood tree topology constructed from pol sequence data.

The tree contains a total of 2572 taxa and was constructed in phyML, with the use of the HKY85+G (alpha = 0.8) method of nucleotide substitution. The genetic distance is shown in the bottom line and corresponds to the length of the branches. Each of the branches has been colour coded corresponding to the place of origin of each of the taxa. In the first tree southern African countries excluding South Africa has been coloured in blue, while in the second tree individual countries are coloured with their own unique colours.