Abstract
Major mutations associated with HIV-I integrase inhibitors (INI) resistance are rare in INI-naive patients. However, polymorphisms at positions that may influence the genetic barrier and/or drive the selection of specific INI resistance pathways are common in HIV non-B subtypes. The aim was to evaluate the presence of natural polymorphisms and/or INI resistance mutations in HIV-1 non-B subtype samples obtained from INI-naive patients living in rural west Cameroon. Thirty-three HIV-1 non-B samples were obtained from INI-naive African women and, as controls, 15 samples of HIV-1 subtype B were obtained from antiretroviral-naive Italian patients. The integrase gene was amplified and sequenced using Trugene Core Reagents. Several amino acid positions in B and non-B subtypes were found to be polymorphic. Interestingly, two patients infected with the CRF02_AG subtype had the resistance mutations N155H and E157Q/E and 12% of African samples had an amino acid substitution at position 143. Silent mutations leading to a higher increment of genetic barriers were detected at 140 and 151 positions in non B-subtypes. Although most polymorphisms may have little effect on INI susceptibility, the IN gene variations found in the present study should be taken into consideration as they may facilitate or delay the emergence of variants fully resistant to INIs.
HIV-1
Reduced in vitro and in vivo efficacy of IN inhibitors (INI) has been associated with the emergence of mutated IN variants. RAL in vivo failure has been linked to mutations that mainly occur through two genetic pathways: N155H in combination with the secondary mutations L74M, E92E/Q, T97A, Y143H, V151I, G163R, or D232N, and Q148K/R/H with the secondary mutations L74M, E138K/A, or G140A/S. 2 –4 Other mutations have been reported, namely Y143H/R/C plus L74A/I, E92Q, T97A, E157Q, I203M, and S230R. 2 –5 EVG resistance in vitro has been associated with primary mutations T66I, associated with the secondary mutations F121Y, S153Y, R263K, and E92Q together with secondary mutations H51Y, S147G, and E157Q. 6 Like RAL, in vivo EVG resistance generally follows two profiles: E92Q with N155H, and N155H with E138K, S147G, Q148R, G140S/C, or Q148R/H. 7
In addition to resistance mutations, some natural IN polymorphisms described in INI-naive patients may hamper INI efficacy as they disrupt the binding of the drug to the region adjacent to the catalytic core. 8 These mutations are H51Y, F121Y, T125K, Q146K/R, S147G, V151I, S153A/Y, M154I, G163R, V165I, V201I, and S230N.
The aim of the present study was to investigate whether INI-related genetic polymorphisms and/or INI resistance mutations were present in the IN gene of HIV-1 isolates obtained from women living in a rural area in Menoua Department, West Cameroon, where non-B subtypes of HIV-1 are prevalent.
Thirty-three HIV-1 plasma samples were from INI-naive African women who had participated in a public health pilot program for the prophylaxis of mother-to-child transmission of HIV-1. An additional 15 samples were collected from antiretroviral (ARV)-naive Italian patients attending the “Policlinico Umberto I” University Hospital in Rome and used as controls. At the time of analysis, African patients had a median viral load of 243,963 copies/ml (range: 390 to >500,000 copies/ml), 24 of whom had been treated with single-dose nevirapine (NVP), five with zidovudine+lamivudine, one woman with zidovudine only, and three women were untreated.
The median viral load of ARV-naive Italian patients was 12,705 copies/ml (range: 2,850–72,696 copies/ml).
HIV-1 reverse transcriptase (RT), protease (PR) and IN genes sequences were obtained from 140 μl of plasma extracted by the QIAmp Viral RNA kit (Qiagen, Milan, Italy) in accordance with the manufacturer's instructions. HIV-1 RT and PR were amplified and sequenced using a TruGene HIV-1 Genotyping kit (Siemens Healthcare Diagnostics, Deerfield, IL).
9
The IN gene was amplified and sequenced using a TruGene Core Reagents (Siemens Healthcare Diagnostics Inc., Deerfield, IL) as described.
10
Nucleotide (nt) sequences were checked for ambiguities, misreads, and stop codons. Sequences were then aligned to the consensus sequences of 10 subtype prototype strains and 10 recombinant circulating forms (CRFs) retrieved from the Los Alamos HIV Sequence Database (
The RT-PR alignment, originally 1029 nt in size, was shortened to 918 nt residues by removing an internal portion not covered by the TruGene HIV-1 Genotyping kit. IN alignment encompassing all sequences was 222 nt long. Genetic distances and phylogenetic relationships were determined with methods available in DAMBE (version 5.2.13,
The RT-PR nt sequences of the 48 patients were first examined for subtype and genetic and phylogenetic relationships. Nt alignment was phylogenetically informative and provided reliable segregation of African isolate sequences with subtype reference strains. All 15 samples from Italian patients belonged to subtype B (data not shown). In contrast, the 33 samples from the West Cameroonians were all non-B and distributed as follows: 19 CRF02_AG (57%), 4 D (12%), 4 CRF01_AE (12%), 3 G (9%), 2 F2 (6%), and 1 CRF11_cpx (3%) (Fig. 1). This result was observed using either NJ or ML, which reconstruct phylogenetic evolution using different approaches, extent of sequence bootstrapping, and nt or amino acid sequences. Subtype classification was confirmed by submitting sequences to the Los Alamos HIV Sequence Database. Mutations associated with NVP and nucleoside analogs resistance were found in 25% and 6% of African women, respectively.
Molecular phylogenetic analysis of reverse transcriptase-protease (RT-PR) nt sequences. Evolutionary relationship of HIV strains was inferred by means of neighbor-joining
Concerning RAL or EVG resistance-associated primary mutations, these were absent in all HIV-1 subtype B samples. One subtype B sample harbored M154I and V165I mutations and another had glutamic acid replaced by aspartic acid at position 92, linked to INI resistance. In contrast, in some African samples mutations associated with resistance to INI were detected. Specifically, N155H and E157Q/E mutations were observed in two CRF02_AG subtype samples. The presence of the E157Q/E mutation in INI-naive African patients has been previously reported, 18,19 particularly in isolates from Cameroon. 20 On the contrary, to our knowledge, the presence of N155H in naive patients had never been observed. Moreover, five samples had an amino acid substitution at position 143 (Y143H/R/C), a mutation found to be associated with lower sensitivity to RAL in several studies. 3 –5,21 The presence of this mutation is of some interest as this residue has been described as critical for HIV-1 IN function at the preintegration stage. 21 At the same position, the presence of the polymorphism Y143S was detected in one CRF11_cpx sample.
Among secondary mutations that have not been directly associated with RAL and EVG resistance, 22,23 L74M/A/I, T97A, V151I, and G163R were present at various prevalences. (Table 1 shows the non-B samples in which mutations and polymorphisms were observed.) According to other authors, 18,24 CRF02_AG and subtype B samples bear different codons at positions 140 and 151. Specifically, at position 140 95% of CRF02_AG samples had a GGA codon. In contrast, 86% of subtype B isolates showed GGC at same position; position 151 had GTG and GTA in 36% of CRF02_AG and 100% of subtype B samples, respectively. Silent mutations were also detected at the 140 position in all CRF01_AE samples, as well as in D (75%), CRF11_cpx, and F2 subtype samples. The presence of different codons at these positions in the non-B subtype is of interest since it led to a high number of transitions (ts: A↔G or C↔T) and transversions (tv: A↔C, A↔T, G↔T, G↔C) required for evolution to RAL resistance. In fact, as previously reported, 24 polymorphisms at positions 140 and 151 led to a higher genetic barrier for non-B subtypes to acquire mutations G140S, G140C, and V151.
Primary mutations associated with integrase inhibitors resistance are in bold.
Previous reports showed that F227L, T215Y, and L210W and other mutations correlated with RT resistance 22,25 as well as ARV therapy enhance IN divergence. 26 In our study, RT mutations were found in two samples: one harbored M184V, K103N, V179E, and M230L and the other had F116Y, Q151M, M184V, K103N, V90I, A98G, and V108I. In neither sample were there mutations and/or polymorphisms correlated with IN resistance, suggesting that coevolution of RT and IN genes under selection pressure put forward in several reports did not stand out in our study.
Our study confirms that natural genetic polymorphisms may occur in INI-naive patients. The study has been performed on a limited number of samples; however, interestingly, in these patients, who live in a rural area of Menoua Department, a relatively high frequency of the Y143H/R/C mutation was found. Although most polymorphisms may have little effect on INI susceptibility, 22,27 the IN gene variations found in the present study should be taken into consideration as they may facilitate or delay the emergence of variants fully resistant to INIs. It would be advisable, therefore, to also extend IN genotypic analysis to INI-naive individuals, particularly if infected with non-B subtypes. This information may be useful not only to elucidate natural IN genetic variability but also to study the efficacy of INIs in these HIV-1 subtypes, about which very little is known at this time.
Footnotes
Acknowledgments
This work was supported by a “Sapienza” University grant to O.T. and to G.A.
Author Disclosure Statement
No competing financial interests exist.