Abstract
Thirty-five HIV-1 infected patients showing clinical and/or immunological failure to first line antiretroviral therapy (ART) according to WHO criteria were recruited from the ART center of Lok Nayak Hospital, New Delhi to detect the presence of resistance-mutations in reverse transcriptase (RT) and protease (PR) region of pol gene of HIV-1. Plasma viral load (PVL) was estimated. HIV-1 pol gene region encoding complete protease and reverse transcriptase (codons; 1-232 to 1-242) was reverse transcribed, followed by nested PCR. The PCR product was sequenced and analyzed. Plasma samples from 94.3% of patients with PVL >log10 3.0 c/mL could be amplified and analyzed. Virologic failure was detected in 65.7% of patients according to WHO criteria (PVL >log10 4.0). All patients were found to be infected with subtype C. One or more resistance-mutations were observed among 90.9% of study sequences. Nucleoside reverse transcriptase inhibitor (NRTI) resistance mutations were seen among all patients, with M184V and thymidine analogue mutations (TAM) being most frequently detected (75.6% and 72.7%, respectively). Nonnucleoside reverse transcriptase inhibitor (NNRTI) resistance-mutations were detected in 63.6% of sequences, of which Y181C/I (47.6%), K103N (33.3%) and G190S (28.6%) are the most common. None of the sequences showed major protease inhibitors (PIs) resistance mutation. High prevalence of NRTI and NNRTI drug resistance mutations among the study participants warrants the use of genotypic resistance testing to prevent accumulation of resistance mutations, which would limit future treatment options.
Introduction
Generic antiretroviral (ARV) drugs in fixed dose combinations are being provided free of cost in the public sector in India under 3 by 5 initiative of World Health Organization–Joint United Nations Programme on HIV/AIDS (WHO-UNAIDS), resulting in a reduction in the HIV-related morbidity and mortality. United Nations member states have endorsed the global goal of universal access to antiretroviral therapy (ART) by 2010 in resource-limited countries as an international priority. 1 However, cases of treatment failure (TF), either clinical or immunologic or both, are being encountered at the ART clinics, thereby, hampering the success of ART and posing a challenge for the clinical management of the patients. Emergence of drug-resistant HIV type 1 (HIV-1) variants has been reported as the most probable cause of TF under the subtherapeutic plasma levels of ongoing ART regimen due to poor adherence , variations in pharmacodynamics between individuals, and differential penetration of drugs in sanctuary sites. 2 As a result of these obstacles, combination ART does not durably suppress HIV-replication in 50% to 70% of treatment-experienced patients. Evolution of drug resistance can limit treatment options and resistance to 1 drug can result in cross-resistance to other drugs from the same class. 3,4
Current knowledge regarding antiretroviral drug resistance (ARVDR) in HIV-1 protease (PR) and reverse transcriptase (RT) is mostly from the data published using resistance assays on subtype B of group M viruses prevalent in Europe and North America. 1 However, increasing data on the genetic mechanisms of drug resistance in nonsubtype B viruses suggest that some differences in drug-resistance evolution among HIV-1 subtypes occur. New mutations at known and new positions are emerging and novel patterns of resistance are accumulating. 5 The most common global HIV-1 variant is subtype C, responsible for more than 50% of the global HIV infection among the population. 6 This is the most common subtype in Southern Africa as well as in India. Plasma sample harbors the actively replicating viruses and is preferred for resistance testing by genotypic assay. 7 In India, as in many resource-limited settings, the use of ARVDR testing in clinical care is still uncommon, and data on ARVDR profile among TF patients are very limited.
The objective of the present study was to identify the ARVDR-associated mutation patterns in HIV-1 PR and RT region of pol gene among TF patients by genotypic assay.
Materials and Methods
Study Population
First-line ART-experienced HIV-1-infected participants, showing clinical and/or immunological TF according to WHO criteria, who were registered in ART clinic under Lok Nayak Hospital, New Delhi, and referred to the Department of Microbiology, Maulana Azad Medical College, New Delhi, for CD4 count (CD4 count) estimation, were enrolled in the study between October 2006 and September 2008. A proforma was filled after obtaining written informed consent. The study protocol was approved by Institutional Review Board of Maulana Azad Medical College, New Delhi. Treatment Failure patients were provided with fixed dose combinations of 2 nucleoside reverse transcriptase inhibitor (NRTI) and 1 nonnucleoside reverse transcriptase inhibitor (NNRTI).
Specimen Collection
Whole blood of 7 mL was collected by venepuncture under aseptic conditions, 2 mL of which was transferred to a sterile vial containing K3 EDTA as an anticoagulant for estimation of CD4 count and stored at room temperature until analysis but within 24 hours of blood collection. Remaining 5 mL blood in K3 EDTA vial was centrifuged at 800 to 1600g for 20 minutes at room temperature within 6 hours of collection to separate plasma. Plasma was aliquoted into sterile cryovials and stored at −70°C until used for viral load estimation and RNA extraction from cell-free HIV-1 virion to carry out genotypic resistance testing of complete PR and 5′ part of RT region of pol gene to detect ARVDR.
CD4 Count and Plasma Viral Load Estimation
Enumeration of absolute CD4 count was done by automated flow fluorocytometry on FACSCount system (Becton Dickinson, San Jose, California) to check the immune status of the participants. Plasma viral load was estimated using AMPLICOR HIV-1 MONITOR TEST, version 1.5 (Roche Diagnostics, Branchburg, New Jersey) based on RT-polymerase chain reaction (RT-PCR) technique by adopting the standard procedure with a linear dynamic range of 400 to 750 000 copies/mL (3.6-5.9 log10 copies/mL) per the kit literature provided by the manufacturer. Briefly, extracted RNA specimen from the plasma was reverse transcribed and amplified on the GeneAmp 9700 thermal cycler (Applied Biosystems, Foster City, California). Amplified fragment of HIV genome was detected colorimetrically and results were interpreted accordingly.
Genotypic Resistance Testing of HIV-1 PR and 5′ RT Regions of Pol Gene
HIV-1 RNA from plasma was extracted using QIAamp viral RNA mini kit (QIAGEN, Hilden, Germany) per the manufacturer’s instructions. RNA was reverse transcribed to make complementary DNA (cDNA) using GeneAmp RNA PCR kit (Applied Biosystems) in a 20-µL reaction volume. The reaction mixture consisted of 4 µL of RNA template preheated at 68°C for 2 minutes to denature secondary RNA structures, 4 µL of 5× RT-PCR buffer, 5 mmol/L MgCl2, 600 µmol/L deoxynucleotide triphosphates (dNTPs), 20 units RNase inhibitor, 50 units of MuLV RT, and 100 pmol/L of random hexamer. Temperature profile used for RT-PCR was 42°C for 45 minutes, 99°C for 5 minutes, and 5°C for 5 minutes. Self-designed primers were used for amplification of pol gene region. Outer PCR primers used were POL1 (sense primer, 5′-CCACCAGCAGAGAGCTTCAGGTTCG-3′) and POL2 (antisense primer, 5′-CTGTATAGGCTGTACTGTCCATTTGTCAG-3′) to cover 1111 bp region of pol gene. The reaction mixture consisted of cDNA, 2 µL of 10X PCR buffer, 1.5 mmol/L MgCl2, 200 µmol/L dNTPs, 0.5 pmol/L each of primer POL1 and POL2, and 2.5 units of Hotstar Taq DNA polymerase (Qiagen, Hilden, Germany). The temperature conditions were held at 95°C for 15 minutes, 35 cycles of amplification at 94°C for 30 seconds, 54°C for 30 seconds, 72°C for 1.5 minutes, and finally at 72°C for 10 minutes.
Inner PCR was performed using inner primer pair POL3 (sense primer, 5′-CCACCAGCAGAGAGCTTCAGGTTCG-3′) and POL4 (antisense primer, 5′-CTGTATAGGCTGTACTGTCCA-3′) to amplify 1104 bp fragment of pol gene. The inner PCR was carried out utilizing the same PCR reaction composition as used in outer PCR using 1 µL of outer PCR product and inner primer pair, 0.5 pmol/L each of primer POL3 and POL4, under the following PCR temperature conditions held at 95°C for 15 minutes, 25 cycles of amplification at 94°C for 30 seconds, 54°C for 30 seconds, 72°C for 1.5 minutes, and finally at 72°C for 10 minutes. Polymerase chain reaction product was analyzed on 1.5% agarose gel (Amresco Inc., Solon, Ohio).
Nucleotide Sequencing and Sequence Analysis
Amplicons were cut and purified from agarose gel using Zymoclean Gel DNA recovery kit (Zymo Research, Irvine, CA) per the manufacturer’s instructions. Purified amplicons were subjected to big dideoxy terminator cycle sequencing on ABI 3100 Genetic Analyzer system (Applied Biosystems). Bidirectional sequencing was performed using inner primer pair. The region sequenced included complete PR (codon 1-99) and 5′ RT (1-232 to 241) region of pol gene. Nucleotide sequences were edited manually wherever required. The derived nucleotide sequences were submitted to “HIV db Program: Sequence Analysis” in the Stanford University HIV drug-resistance database (http://hivdb.stanford.edu/hiv) for quality assessment and drug-resistance mutations interpretation. Protease and RT mutations were defined as difference from the consensus B, according to the Stanford database. Nucleotide mixtures of wild type and mutations were classified as mutation. Subtyping of the study sequences was done by comparing them to reference sequence sets using the Basic Local alignment Search Tool (BLAST) available at Los Alamos HIV Sequence Database (http://hiv-web.lanl.gov/). Phylogenetic tree was inferred by neighbor-joining method from matrix-distance calculated with a Kimura 2-parameter algorithm and bootstrap value of 500 replicates and rooted with group M consensus sequences using Molecular Evolutionary Genetic Analysis (MEGA) software for Windows and with bootscanning (Simplot, Baltimore, MD, http://sray.med.som.jhmi.edu/RaySoft/Simplot/) to rule out recombination. The derived nucleotide sequences were submitted to GenBank to assign the accession numbers.
Statistical Analysis
Descriptive analyses of the baseline demographics of the population were performed. Categorical variables were summarized using percentages and compared using the chi-square test or Fisher exact test and continuous variables were summarized using medians and interquartile ranges (IQRs). A 2-sided P < .05 was considered significant during all analysis.
Results
A total of 35 TF patients were enrolled in the study. The median age of the study participants was 33.5 (IQR: 30-37.8). The majority of the patients were male compared with female (85.7% vs 14.3%). Predominant mode of HIV transmission was heterosexual promiscuity as compared to intravenous drug use (94.9% vs 5.1%). The median CD4 count was found to be 128 cells/mm3 (IQR: 71-108). The evidence of virologic failure, according to WHO criteria (PVL >4.0 log10copies/mL), was found among 65.7% (23 of 35) of participants and among all cases per Centers for Disease Control and Prevention (CDC). The median duration of ART among TF participants was 28.7 months (IQR: 14-38), and the majority of the individuals experienced fixed dose combinations of lamivudine (3TC), stavudine (d4T), and nevirapine ([NVP] 66.7%; 14 of 21). All patients reported >95% adherence to ART. Single drug substitution was done in 8.6% (3 of 35) of patients due to side effects and in 20.0% (7 of 35) due to initiation of antituberculosis therapy (ATT).
Of the 35 plasma samples, 33 (94.3%) samples with PVL >3.0 log10 copies/mL could be amplified and were mentioned during all further analyses. All patients were found to be infected with HIV-1 subtype C, identified on the basis of nucleotide sequence analysis of protease and RT region of pol gene.
Mutations in RT
Table 1 is showing the resistance pattern observed among plasma samples from TF individuals. In all, 90.9% (30 of 33) of the study sequences showed one or more resistance mutation, at least NRTI resistance mutations among all. All 23 patients, showing virologic failure per WHO, were found to harbor ARVDR-associated mutations. Mutated amino acids associated with ARVDR at different RT and PR positions have been illustrated in Table 2. M184V/I was the most frequently detected (25 of 33; 75.6%) mutation, conferring resistance to NRTI, 3TC. In thymidine analogue mutations (TAMs) category, which are known to confer high-level resistance mainly to thymidine analogue ARV drugs, zidovudine (ZDV) and d4T and low-level resistance to other ARV drugs also in the NRTI category were seen in 72.7% (24 of 33) sequences. Different TAMs observed were M41L (10 of 24; 41.7%), D67N/G (15 of 24; 62.5%), K70R (11 of 24; 45.8%), L210W (2 of 24; 8.3%), F215Y/F (17 of 24; 70.8%), and K219Q (8 of 24; 33.3%). None of the HIV-1 sequences among TF individuals showed major protease inhibitor (PI) resistance mutation. TAM profiles observed among individual TF patient’s samples have been shown in Table 3. Of the 24 samples from TF individuals with TAM, 22 (91.7%) HIV-1 sequences were seen with different combinations of TAMs: 5 TAMs in 1 sequence, 4 TAMs in 3 sequences, 3 TAMs in 11 sequences, and 2 TAMs in 7 sequences. One TAM was seen alone in 2 sequences. TAMs in combination with M184V were observed in 54.5% (18 of 33) of the sequences. Limited evolution of other NRTI resistance mutations; V75I (3 of 33; 6.1%), F116Y (2 of 33; 6.1%), V118I (5 of 33; 15.2%), and Q151M (2 of 33; 6.1%) was also observed. V75I, F116Y, and Q151M were found to be associated with each other in 2 (6.1%) sequences. In all, 63.6% (21 of 33) sequences showed NNRTI resistance mutation, of which Y181C/I being the most common (10 of 21; 47.6%), followed by K103N (7 of 21; 33.3%), G190S (6 of 21; 28.6%), and other mutations, perhaps in lower frequency. Polymorphic mutations were detected in varying frequencies at many positions in RT and PR regions, reaching a substitution rate of 100% at some codon.
Resistance Mutations Profile in 33 TF Patients with PVL >1000 Copies/mL
Abbreviations: PVL, plasma viral load; 3TC, lamivudine; NRTIs, nucleoside reverse transcriptase inhibitors; NNRTIs, nonnucleoside reverse transcriptase inhibitors; PIs, protease inhibitors.
Mutations at Positions Associated with Resistance in Reverse Transcriptase Region of pol Gene among 33 TF Patients
Abbreviations: NRTIs, nucleoside reverse transcriptase inhibitors; NNRTIs, nonnucleoside reverse transcriptase inhibitors; TF, treatment failure.
Thymidine Analogue Mutations Detected in 24 HIV-1 Sequences from TF Participants
Abbreviation: TF, treatment failure.
Discussion
As international involvement in global HIV/AIDS care and access to lifesaving combination ART are progressively increasing, the number of non-B viruses that are exposed to ART will increase and reveal its consequences with regard to the evolution of drug resistance. Evolution of ARVDR can limit treatment options and reduce probability of complete viral suppression. Significant prevalence of ARVDR was found in settings where ART has been around for a significant period of time. 3
The present genotypic resistance study was performed among 35 TF individuals by home-brew RT-PCR technique in plasma specimen. Analysis of pol gene sequences of HIV-1 by Stanford Drug Resistance Database confirmed subtype C infection among all of the study participants. Other resistance studies in pol gene from India have also reported the predominance of HIV-1 subtype C, along with subtype A, CRF01 AE, and A/C intersubtype recombinant among few cases. 8 –10
In the present study, HIV-1 sequences have been found to show one or more resistance mutation in 90.9% (30 of 33) in the RT region of TF patients. M184V/I was the most frequently observed resistance-associated mutations to NRTIs (75.6%). In vitro, it causes high-level resistance to 3TC and emtricitabine (FTC), low-level resistance to didanosine (ddI) and abacavir (ABC), and increased susceptibility to ZDV, d4T, and tenofovir (TDF). 11 Few earlier reports, particularly from central India, have also documented very high prevalence of resistance mutations in the patients with evidence of TF and are consistent with our findings. Two reports were published by Sen et al from Pune in 2007, and they showed resistance-associated mutations in 80.6% of study samples estimated from peripheral blood mononuclear cells (PBMCs) and 81.8% in the plasma samples. 9 M184V was the most commonly observed DR mutation among all these studies. A study by Pillay et al among South African patients virologically failing ARV therapies, who had received at least 2 NRTIs and either an NNRTI or PI, showed that 91% of patients harbored resistance mutations; the most frequent NRTI mutations was M184V/I (37%), although, in lower frequency compared with our study. 12 Another study from Georgia among patients with virologic failure showed resistant mutations in 75.5% participants, with M184/V/I in 68.9%, which supported the fact of detection of high prevalence of resistant mutations among ART failing patients. 13
Thymidine analogue mutations were seen in 72.7% (24 of 33) of HIV-1 sequences. Thymidine analogue mutations decrease susceptibility to these NRTIs and to a lesser extent to ABC, ddI, and TDF. 11 A report has claimed that TAMs are common in low-income countries in which fixed-dose combinations containing thymidine analogues are the mainstays of therapy. In our study, 91.7% (22 of 24) of the HIV-1 sequences showed TAM in combinations. Thymidine analogue mutations accumulate in 2 distinct but overlapping patterns. 14 –16 The type I pattern includes the mutations M41L, L210W, and T215Y. The type II pattern includes D67N, K70R, T215F, and K219Q/E. Mutation D67N also occurs commonly with type I TAMs. 15 Our study sequences were also observed with overlapping pattern between types I and II cluster of TAMs. Two of the present study sequences were found with type I pattern and 6 sequences were found in combination of M41L, D67N/G, and T215Y/F with or without other TAMs. T215Y was present in 62.5% (15 of 24) of sequences found with TAM, whereas T215F was detected in 8.3% (2 of 24) sequences. HIV-1 variants with T215Y have been reported with increased fitness and improved replication kinetics over T215F. K70R and L210W occur rarely together 17 ; one of our study sequences, although, was found with this combination along with other TAMs. Type I TAM causes higher levels of phenotypic and clinical resistance to the thymidine analogues and cross-resistance to ABC, ddI, and TDF than do the type II TAM. 15,18,19 However, prevalence of TAMs reported by Sen et al was low (47.1%, 16 of 33) among NRTI-experienced patients when compared with our observations. They found T215Y in 5, T215F in 5, and overlapping between types I and II TAMs among 4 sequences. 9 Deshpande et al from Mumbai have also revealed high level of resistance to NRTIs (72.3%) among drug-experienced (DE) patients showing TF. 8 Pillay et al from South Africa also reported TAM; D67N (32%), T215Y/F (25%), K70R (21%), M41L (20%), K219Q/E (14%), and K65R (14%), which are antagonistic to TAM, were also observed and reflected in the frequent use of 3TC and ZDV. These data indicated that HIV-1 drug resistance develops in subtype C-infected patients failing ARV therapy with mutations comparable to those found among patients infected with subtype B viruses. 12
Multinucleoside resistance mutation Q151M was observed in 6.1% of the study sequences. The similar prevalence (6.1%) was also reported by Sen et al. 9 However, amino acid insertions at codon 69, generally occurring in the presence of multiple TAMs, were not observed. Furthermore, K65R, which interferes with TAM-mediated primer unblocking, was also not observed in any study sequence.
Nonnucleoside RT inhibitor resistance mutations were seen in 63.6% (21 of 33) of our study sequences, in which Y181C (10 of 21; 47.6%) was the most frequently observed mutation followed by K103N (33.3%) and G190S (28.6%) with other NRTI mutations in lower frequency. However, Sen et al reported higher prevalence of NNRTI resistance mutations (80.65%), although prevalence of Y181C and G190A was found comparable to our findings (35.35% each). 9 Pillay et al also reported almost comparable prevalence of NNRTI resistance mutations K103N (25%), V106M (20%), and G190A (17%) among patients failing NVP or efavirenz (EFV)-based regimens. 12 However, G190S/A was the most frequently detected NNRTI resistance mutation (42.2%) followed by K103N (28.9%) by Chkhartishvili et al. 13 Undoubtedly, the NNRTI class has a low-resistance barrier (single mutations may cause a significant decrease in drug susceptibility) and NNRTI-based highly active treatment combinations are a popular initial regimens in the country as well as in the WHO treatment program. 20 However, the reason for comparatively low prevalence of NNRTI resistance mutations in our study may be that numerous nucleoside (or nucleotide) analogue RT inhibitor (nRTI) mutations, such as the M41L, L210W, and T215Y, may lead to viral hypersusceptibility to NNRTIs in NRTI-treated individuals. The viruses that contained less RT replicated less efficiently than those with wild-type levels of RT. 21
None of our study sequence showed PR major resistance mutations among TF individuals which was consistent with the findings reported by Deshpande et al. 8 The reason for such findings may be that none of the patients in both the studies have experienced PI in the first-line ART, and PI-containing regimens are being provided only to the patients found eligible for second-line ART. However, Sen et al reported multidrug PR major resistance (M46I) mutations for the first time in India among 2 of 4 patients who experienced PI. 9 According to Pillay et al,12 among the patients who received PIs, the most common mutations were V82A/T (12%), M46I (11%), and L90M (8%). These data indicated that ARVDR in HIV-1 subtype C-infected TF patients is comparable with those found infected with subtype B viruses.
No ARVDR-associated mutation was observed in 3 of our study sequences in spite of showing clinical and immunological TF. Minority quasi-species (ie, <20% prevalence) of drug-resistant HIV-1, which are undetectable by population sequencing, has been reported to play an important role with regard to early virological failure despite good adherence level. 22 Metzner et al reported drug-resistant viruses at low frequencies in patients failing to treatment, with a frequency range of 0.07% to 2.0%. A range of 1 to 4 mutations was detected in viruses from each patient. 23
Our preliminary study detected a high prevalence of drug-resistance-associated mutations among TF patients showing virologic failure or detectable viremia. Genotypic resistance testing should be recommended for such patients to prevent accumulation of resistance mutations, which would limit future treatment options. Furthermore, the transmission of resistance can possibly compromise therapy outcome among drug-naïve (DN) individuals to be initiated on ART harboring these resistant variants. In the future, ART has to be adjusted to baseline or prevalent ARVDR prophylaxis to avoid the use of ineffective ARV agents and the further evolution of drug resistance.
Footnotes
All the HIV-1 PR and RT nucleotide sequences reported in the present study have been submitted in GenBank and are available under the accession numbers FJ907462 to FJ907498 and GQ906404 to GQ906417.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Financial assistance was providing to S.S. in the form of Senior Research Fellowship (HIV/ Fellowship/18/2003/ECD-II) by Indian Council of Medical Research.