Genotypic Variability of HIV-1 Reverse Transcriptase Gene from Long-Term Antiretroviral-Experienced Patients in Kenya

The majority of the global HIV-1 epidemic is represented by genetically diverse forms of the virus. ^1,2 The evolutionary rate of HIV has been shown to track with epidemic events or milestones, including rollout of highly active antiretroviral treatment (HAART) and changes in prevalence. ³ Of the global epidemic, subtype C contributes about 50% of all infections, while recombinants account for approximately 20%. HIV-1A and B account for about 12%, 11%, and 20% of the global HIV epidemic, respectively, with subtype B being restricted to Europe, North America, and parts of North Africa. In Africa, HIV infection exhibits distinct regional trends with varying rates and prevalence. ⁴

The Kenyan epidemic has declined significantly and steadily over the past 8 years to stabilize at low levels of 5.6%, ^5,6 although there are key demographic pockets for escalating new infection. ⁷ The overall improvements in HIV prevalence in Kenya can be attributed to gains in uptake of testing and scaled access to HAART. ⁸ But increased access to HAART presents special challenges in diagnosing new infections, monitoring of the epidemic, and managing subepidemics as new or mutant virus strains appear in circulation. ⁹

We recently reported on HIV subtype variability in Kenyan HAART patients based on envelope sequences. ¹⁰ The epidemiologic architecture of the Kenyan epidemic displays characteristic variability at both genetic and geographic levels, with HIV-A dominating, although at a declining prevalence. ^{10 –13} Here, we report the genetic diversity of the HIV-1 RT gene among patients who have sustained HAART for between 8 and 102 months.

Fifty-four patients aged 18 to 56 years were enrolled between February and September 2013 from six comprehensive care centers located in Kiambu, Kajiado, Naivasha, Malindi, Kisumu, and Homa Bay counties. Five milliliters of EDTA venous blood was drawn from each patient and fractionated to obtain peripheral blood mononuclear cells (PBMCs) and plasma as described earlier. ¹⁰ 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 and according to manufacturer's instructions. For the RNA, a 701-base pair segment of the HIV-1 pol-RT gene corresponding to nucleotides 2480–3180 of HIV-1_HXB2 was amplified in a One-Step reverse transcriptase polymerase chain reaction (RT-PCR) protocol (Qiagen) followed by a nested PCR. Primers RT18 (5′-GGAAACCAAAAATGATAGGGGGAATTGGAGG-3′) and KS104 (5′-TGACTTGCCCAATTTAGTTTTCCCACTAA-3′) were used, respectively, in the forward and reverse RT-PCR reactions. A final PCR reaction volume of 25 μl included a 1×PCR buffer, 0.4 mM dNTPs, 1.5 mM MgCl₂, 5 μl of RNA template, 1 μl enzyme mix, 0.6 μM of each primer, and 5 units of RNAseOut. Reverse transcription was performed at 50^°C for 30 min followed by amplification steps of 15 min denaturation at 95^°C and 40 cycles of denaturation at 94^°C for 30 s, annealing at 55^°C for 45 s, and extension at 72^°C for 1 min. A final extension step was performed at 72^°C for 10 min.

Second round PCR was done using HotStar taq polymerase (New England Biolabs) in a 25 μl reaction volume comprising 1×PCR buffer, 0.2 mM dNTPs, 2 mM MgCl₂, 0.5 μM forward and reverse primers, 0.625 units of the enzyme, and 3 μl of RT-PCR product as template. The thermocycling profile comprised denaturation at 95^°C for 5 min followed by 40 PCR cycles of 95^°C denaturation for 30 s, annealing at 56^°C for 45 s, and extension at 68^°C for 45 s. Afinal extension was done at 68^°C for 10 min. KS101 (5′-GTAGGACCTACACCTGTTCAACATAATTGGAAG-3′) and KS102 (5′-CCCATCCAAAGAAATGGAGGAGGTTCTTTCTGATG-3′) were used as forward and reverse nested PCR primer,s respectively. When DNA was used as the starting material, PCR procedures were similar except that the RT step was excluded. The PCR products were purified with Qiaquick purification columns and sequenced using standard Big-Dye chain terminator chemistry.

Sequences were manually inspected, joined in pairwise alignment (BioEdit ver. 7.2.5), and subjected to multiple sequence alignment using Clustal W (Supplementary Data; Supplementary Data are available online at www.liebertpub.com/aid). Two reference sequences were included in the alignment for each subtype query sequence. Phylogenetic analyses were done using the neighbor-joining method with 1,000 bootstraps. Evolutionary distances were computed using the Kimura two-parameter method, with gaps partially deleted at 50% cutoff. Phylogenetic trees were generated using MEGA version 6.06 and used to assign subtypes (Fig. 1). Alternate subtype assignment was done using the JPHMM tool for comparison. All the sequences reported in this article have been deposited and accessioned at the GenBank. Ethical approval was granted by the KEMRI Scientific and Ethics Review Unit protocol #2477. Corresponding patient and genotypic data are provided in Table 1. Using the JPHMM subtyping tool, 57.4% of the 54 sequences were determined to be subtype A1 while subtypes C, D, and recombinants were 5.6%, 14.8%, and 22.2%, respectively. The majority (n=6, or 50%) of the RT recombinants were AC (A1C, n=5 and A2C, n=1). Others included A1A2 (n=2) and A1D, A2B, A2D, and CD (n=1, each). No isolate was identified as pure subtype A2 by this method.

OPEN IN VIEWER

FIG. 1.

Phylogenetic tree of HIV pol-RT sequences from 54 highly active antiretroviral treatment (HAART) patients. The RT gene was partially sequenced using primers amplifying a 701-base pair region. 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 45% bootstrap values are hidden to minimize clutter. Reference sequences are italicized and shown by filled diamonds.

The phylogenetic method assigned more viruses as subtype A (76%), with the majority being A1 (70.4%). Seven of the 38 subtype A1 isolates were classified as A1A2 (n=2) or A1C (n=5) recombinants by JPHMM. The rest of the subtypes based on phylogenetics were HIV-1C at 9.3% and HIV-1D at 9.3%. Two sequences, MAD021 and MLD185, were substantially shorter than the rest, running just above half the length of the amplified RT gene segment. These isolates were identified by both methods as subtype D and A1, respectively, and are thus deemed to be accurately classified. Three isolates, KMB077, KAH004, and MLD548, that were determined as subtypes D, A1D, and D by JPHMM, respectively, did not cluster distinctly with any reference sequences on the phylogenetic tree and were “not classified” by this method.

Six (or 50%) of the recombinants were from the coastal region of Malindi. Of the 54 sequences, 27.8% contained reverse transcriptase inhibitor (RTI) drug resistance-associated mutations (DRAMs). Seventy-three percent of these mutations conferred resistance to both nucleoside RTIs (NRTIs) and nonnucleoside RTIs (NNRTIs), with the common ones being M184V/I and K103N, respectively. A few studies have recently endeavored to highlight differences in efficiency of different genotyping tools used for HIV subtyping. ^14,15 We recently showed slight but useful differences in detecting HIV-1 subtypes based on the envelope C2V3 gene when using both the JPHMM and phylogeny. ¹⁰ This article reaffirms the same observations using the pol-RT gene, with both JPHMM and phylogenetic methods concurring on subtype assignment in 72.2% (39/54) of the isolates.

While subtype A HIV-1 is still predominant in Kenya, recombinant viruses may be on the rise. Whereas the phylogenetic method is robust at classifying HV strains based on evolutionary distances, it is not powered to detect recombinant viruses. Hence efficient detection tools should be developed or existing methods harnessed to sensitively identify small genetic differences that might influence the overall molecular pattern of the circulating epidemic. Such efforts will empower routine genetic surveillance to better inform public health and vaccine or drug discovery efforts.