Abstract
There is a continuous need to genetically characterize the HIV strains in circulation in order to assess interventions and inform vaccine discovery. We partially sequenced the envelope C2V3 gene from a total of 59 Kenyan patients on highly active antiretroviral treatment (HAART) and determined HIV subtypes using both the JPHMM subtyping tool and the phylogenetic method. HIV-1 subtype A1 was the predominant strain in circulation, representing 65.5% and 74.5% of all isolates as determined by JPHMM and phylogenetic methods, respectively. Subtypes C and D were the next most prevalent pure strains at 9.1% each by both methods. JPHMM identified 9.1% of the isolates as recombinant. Four isolates had short sequences not covering the entire C2V3 region and were thus not subtyped. From this study, subtype A viruses are still the predominant HIV-1 strains in local circulation in Kenya. Constant surveillance is needed to update molecular trends under continuing HAART scale-up.
T
At the national level, subtype A remains dominant, but emerging data suggest that other subtypes, including HIV-1C and HIV-1D, are rising in prevalence. 8 –11 In two previous studies in Northern Kenya, we reported 50–57% of all HIV-1 in that region as subtype A, followed by subtypes C and D at 27–39% and 9–11%, respectively. 8,10 This pattern appeared to be influenced by cross-border population dynamics as we observed. We later reported a different pattern at the Kenyan coast where the prevalence of subtype A1 was much higher at over 85%. 9 These rates may continue to vary as the evolutionary forces of host immunity and antiretroviral pressure exert influence. 5,12,13 Indeed, slow rates of evolution of subtype B in the western setting were found to track with the escalation of antiretroviral treatment (ART), while an increase in evolutionary rates of HIV-1C was associated with declining prevalence among Ethiopians. 12 We assessed the genetic diversity and prevalence of HIV-1 subtypes in the context of continuing HAART scale-up.
Fifty-nine patients aged 22–67 years were enrolled between April and August 2013 from the six comprehensive clinical care facilities in the counties of Homa-Bay, Kisumu, Kajiado, Nakuru, Kiambu, and Malindi, which represented five geographic provinces of Kenya. Five milliliters of venous blood was drawn into EDTA vacutainer tubes from each patient and processed to separate peripheral blood mononuclear cells (PBMCs) and plasma. PBMCs were obtained from plasma-free blood by lysing the blood with 0.84% ammonium chloride. RNA was extracted from plasma of patients with viral load (VL) above 1,000 RNA copies/ml using the QiaAmp Viral RNA Mini Kit while DNA was isolated from PBMCs of patients with VL up to 1,000 HIV RNA copies using the QiaAmp DNA Mini Kit. All protocols were strictly followed as outlined in the manufacturer's manual.
The RNA samples were subjected to One-Step reverse transcriptase polymerase chain reaction (RT-PCR) amplification using envelope primers M5-5′-CCAATTCCCATACATTATTGTGCCCCAGCTGG-3′ (forward) and M10-5′-CCAATTGTCCCTCATATCTCCTCCTCCAGG-3′ (reverse). PCR conditions were reverse transcription at 50°C for 30 min, denaturation at 95°C for 15 min, and 38 PCR cycles of denaturation at 94°C for 30 s followed by annealing for 45 s at 58°C and extension at 72°C for 1 min. First-round PCR for DNA samples followed similar conditions with the exception that the RT step was excluded and HotStar Taq DNA polymerase was used instead of the RT enzyme. Nested PCR was carried out using HotStar Taq polymerase under conditions similar to the first round PCR with the following exceptions: the initial denaturation was at 95°C for 5 min followed by 38 cycles of denaturation at 95°C for 30s, annealing at 56°C, and extension at 68°C for 45 s with a final extension at 68°C. Primers for nested PCR were M3-5′-GTCAGCACAGTACAATGCACACATGG-3′ (forward) and M8-5′-CCTTGGATGGGAGGGGCATACATTGC-3′ (reverse).
PCR and sequencing targeted a 546-base pair envelope C2V3 corresponding to nucleotides 6975–7520 on the HIV-1 HXB2. The PCR products were purified and sequenced using standard Big-Dye chain terminator chemistry (Macrogen, Netherlands). Sequences were manually inspected, pairwise aligned (BioEdit platform), and subjected to multiple sequence alignment using Clustal W. At least two reference sequences per subtype were included in the alignment. Phylogenetic analyses were done using the neighbor-joining method with 1,000 bootstraps and evolutionary distances were computed using the Kimura two-parameter method. Gaps were treated with partial deletion and 50% coverage cutoff. Subtype assignments were done using both the jpHMM tool (
Representative phylogenetic tree of envelope C2V3 sequences. A total of 55 isolates with sequences covering the C2V3 region of the HIV-1 envelope are shown. Branches corresponding to partitions reproduced in less than 50% bootstrap replicates are collapsed and all positions with less than 50% site coverage were eliminated. Node statistics corresponding to less than 50% bootstrap values are hidden to minimize clutter. Reference sequences are in italics.
NR, not reported; HAART, highly active antiretroviral therapy; ND, not determined—sequence not covering the entire C2V3 regions hence deemed unreliable for subtype assignment. TDF, tenofovir; 3TC, lamivudine; NVP, nevirapine; AZT, zidovudine; EFV, efavirenz; d4T, stavudine.
Four of the total of 59 sequences (accession numbers KM853055, KM853066, KM853068, and KM853072) did not cover the entire C2V3 region and were deemed too short for subtype determination. Of the remaining 55 sequences, 65.5% were subtype A1, 7.3% subtype A2, 9.1% subtype C, 9.1% subtype D, and 9.1% recombinants by the jpHMM subtyping method. The recombinant forms were A1A2 (n=1), A1D (n=3), and A1A2D (n=1). Using the phylogenetic method, subtype A1 viruses were 74.5%, subtype C viruses were 9.1%, subtype D viruses were 9.1%, while subtype A2 viruses represented 7.3% of the total virus isolates. Three isolates, KHC064, MLD011, and KAH010, which were determined to be A1D recombinants by jpHMM clustered with subtype A1 using phylogenetics. Both JPHMM and phylogenetic methods concurred on the subtype assignment for pure strains.
Although this article does not assess the comparative accuracies of different HIV-1 genotyping tools, varying sensitivities inherent in these tools affect their concordance in subtype assignment. 14 Since smaller regions of recombination are particularly unlikely to be detected by most of these tools, we cannot rule out the circulation of more recombinant strains locally than we demonstrate here. Our data agree with existing reports from the HIV database on the predominance of HIV-1 subtype A among the Kenyan population. 8 –10 Taken together, these data may suggest the temporal evolutionary stability of the Kenyan HIV-1 epidemic, but analysis of larger and multiple viral genomic regions is necessary to validate patterns and trends under the aggressively continuing HAART campaign.
GenBank Accession Numbers
The sequences reported in this study have been deposited in GenBank and assigned the following Accession numbers: KM853037–KM853149.
Footnotes
Acknowledgments
This study was supported by the Consortium for National Health Research, Kenya, with funds from the Wellcome Trust (UK), grant RCDG-2012-005. Dr. Ochieng was also affiliated with Harvard University at the time of this study, although Harvard was not involved in the study.
Author Disclosure Statement
No competing financial interests exist.