A molecular phylogenetic and modeling approach towards understanding the transmission dynamics and genetic diversity of HIV-1 in the fishing communities of Lake Victoriaa, Uganda

Abstract

Fishing communities (FCs) in Uganda have a greater HIV burden relative to other high-risk groups and the general population (GP). As a result of the very high HIV-1 incidence rates, they have been hypothesized to act as potential reservoirs for new HIV-1 infections in the GP and as such, they have become an important target population in planning effective epidemic control that is aimed at controlling disease spread. However, there has been lack of enough evidence to prove whether these hard-to-reach communities are indeed sources of HIV-1 transmission to the GP. This was mainly due to an inadequate understanding of the inter-population viral transmission dynamics between the FCs and other groups. Also, of interest in this study was the HIV-1 subtype diversity in the FCs and the GP and how this compares to the viral genetic diversity in Uganda and globally. This is because, the extensive genetic diversity of HIV has since the inception of the AIDS epidemic played a key role in viral transmission and dissemination in populations and continues to present challenges for vaccine development and molecular epidemiological investigations. Furthermore, several studies done in Uganda have observed different rates of HIV 1 transmission between subtypes and different levels of clustering within identified HIV transmission networks. The main objective of this research was to understand the HIV-1 transmission dynamics and viral genetic diversity in the FCs and determine how this group contributes to the HIV epidemic in the GP. This research was composed of four sub-studies. In the first sub study, a Bayesian phylogeographic analysis was used to investigate the geospatial viral migration patterns between the FCs and the neighbouring GP to determine the direction of viral transmission between the two populations. This was a source-sink analysis that was aimed at determining whether the FCs were sources of or sinks for HIV transmission in the GP. The major finding in this study was that the FCs were a sink xi

for and not a source of HIV from the GP. In the second sub study, a larger HIV-1 nucleotide sequence dataset (pol region) was analysed using a combination of both phylogenetic and modeling techniques. The main goal of this study was to understand the underlying HIV transmission network structure in the FCs and estimate parameters that influence network formation. Major findings showed that the underlying network structure in the FCs was scale-free and shaped by preferential attachment in contrast to the GP. In the third substudy, a novel agent based model was implemented in the R programming language to simulate and assess the impact of targeted combination interventions (TCIs) in the FCs and other high-risk groups on HIV-1 incidence in the GP. Major findings from this study showed that targeted interventions in the FCs would not directly result in a reduction in the number of new HIV infections in the GP. The fourth substudy aimed at determining the HIV subtype diversity in the FCs and the GP. Results based on the HIV-1 pol gene revealed that Subtype A1 was the most prevalent strain in both the FCs and the GP. Nonetheless, there was an increase in the prevalence of HIV-1 recombinant forms based on near full-length viral genomes. In conclusion, findings from both the phylogenetic and modelling studies revealed localized epidemics in the FCs of Lake Victoria with viral introductions from the GP. The preferential attachment mechanism inferred from networks of FCs suggests that epidemic control in that population could prove difficult without a prior understanding of the HIV transmission dynamics. Model simulations found the GP and FSWs to be direct sources of HIV infections to the FCs and GP respectively. However, although the FCs acted as transmission sinks for HIV infections from the GP, they were found to play a key role as an indirect source of HIV transmission through the female sex workers (FSWs) who act as a bridging population for viral spread between the FCs and the GP. Targeted interventions in the FCs should therefore not negate the roll out of preventative measures in the neighboring GP. Such interventions should be implemented as early as possible in areas identified as hotspots for HIV-1 transmission.