Discovery and full genome characterization of two highly divergent simian immunodeficiency viruses infecting black-and-white colobus monkeys (Colobus guereza) in Kibale National Park, Uganda

Abstract

Background: African non-human primates (NHPs) are natural hosts for simian immunodeficiency viruses (SIV), the zoonotic transmission of which led to the emergence of HIV-1 and HIV-2. However, our understanding of SIV diversity and evolution is limited by incomplete taxonomic and geographic sampling of NHPs, particularly in East Africa. In this study, we screened blood specimens from nine black-and-white colobus monkeys (Colobus guereza occidentalis) from Kibale National Park, Uganda, for novel SIVs using a combination of serology and "unbiased" deep-sequencing, a method that does not rely on genetic similarity to previously characterized viruses.

Results: We identified two novel and divergent SIVs, tentatively named SIVkcol-1 and SIVkcol-2, and assembled genomes covering the entire coding region for each virus. SIVkcol-1 and SIVkcol-2 were detected in three and four animals, respectively, but with no animals co-infected. Phylogenetic analyses showed that SIVkcol-1 and SIVkcol-2 form a lineage with SIVcol, previously discovered in black-and-white colobus from Cameroon. Although SIVkcol-1 and SIVkcol-2 were isolated from the same host population in Uganda, SIVkcol-1 is more closely related to SIVcol than to SIVkcol-2. Analysis of functional motifs in the extracellular envelope glycoprotein (gp120) revealed that SIVkcol-2 is unique among primate lentiviruses in containing only 16 conserved cysteine residues instead of the usual 18 or more.

Conclusions: Our results demonstrate that the genetic diversity of SIVs infecting black-and-white colobus across equatorial Africa is greater than previously appreciated and that divergent SIVs can co-circulate in the same colobine population. We also show that the use of "unbiased" deep sequencing for the detection of SIV has great advantages over traditional serological approaches, especially for studies of unknown or poorly characterized viruses. Finally, the detection of the first SIV containing only 16 conserved cysteines in the extracellular envelope protein gp120 further expands the range of functional motifs observed among SIVs and highlights the complex evolutionary history of simian retroviruses.

Figure 1

Distribution of black-and-white colobus (Colobus guereza) across equatorial Africa. A: C. guereza distribution is shown in grey and distribution of the C. guereza occidentalis subspecies is overlaid in black dots. The newly identified SIVkcol-1 and SIVkcol-2 were isolated from Kibale National Park, Uganda whereas the previously identified SIVcol was isolated from Yaounde, Cameroon. Map and distribution adapted from [44]. B: Location of Kibale National Park within Uganda. C: Location of the study site within Kibale National Park. Site of sample collection is indicated with a black star and encompasses an area of approximately 15 km².

Figure 2

Location of cysteine pairs in the extracellular envelope glycoprotein gp120 of primate lentiviruses. Signal peptide, variable regions 1–4 (V1-V4) as well as the CD4-binding site are shown in relation to their position within gp120. Black bars indicate the position of the 18 conserved cysteine residues while blue bars indicate additional cysteines. The position of the missing cysteines for SIVkcol-2 is highlighted with a dashed red line.

Figure 3

Sliding window similarity plots of concatenated protein sequences of SIVkcol-1 and SIVkcol-2 against other Colobine SIVs. Dashed vertical lines indicate start positions of viral proteins Gag, polymerase (Pol), Vif, envelope (Env), and Nef.

Figure 4

Phylogenetic relationships of newly discovered SIVkcol-1 and SIVkcol-2 to other SIVs. Separate Bayesian Markov Chain Monte Carlo phylogenies were constructed for gag, polymerase (pol), envelope (env) and nef proteins. Posterior clade probabilities are shown on branches. The scale bar below the phylogenetic trees represents substitutions per site. SIVkcol-1 and SIVkcol-2 are highlighted in bold.

Figure 5

Time to most recent common ancestor (TMRCA) for SIVkcol-1 and SIVkcol-2 and other representative SIVs. Bayesian Markov Chain Monte Carlo phylogenies were generated in order to estimate TMRCA. The scale bar below the phylogenies represents years before present and the black arrow represents the Bioko calibration point used in this analysis. Posterior clade probabilities above 0.7 are shown on branches. SIVkcol-1 and SIVkcol-2 are highlighted in bold.