The frequency of HIV-I drug resistance mutations among treatment-naïve individuals at a tertiary care centre in south India

Abstract

Antiretroviral treatment (ART) use in India requires information on baseline drug resistance mutations and polymorphisms in the protease (Pr) and reverse transcriptase (RT) genes of HIV-1 strains from treatment-naïve individuals. We report resistance predictor mutations and polymorphisms in the Pr and the RT sequence of non-clade B HIV-1 strains from ART naïve individuals. The genotypic resistance assay was done on 93 treatment-naïve individuals. The sequences were analysed by Stanford HIV drug resistance data for genotypic drug resistance analysis and REGA HIV-1 subtyping tool. Phylogenetic tree was generated with MEGA 4 for quality control. Ninety-two strains belonged to clade C and one to clade A (A1). Amino acid substitutions were seen at positions associated with drug resistance in Pr gene – 10, 24, 74 (each 3%) and position 82 (11%). Substitutions were seen at positions 41 (1%), 100 (1%), 101 (6%), 103 (2%), 179 (6%) and 181 (1%) of the RT sequence known to confer drug resistance in clade B.

Polymorphisms in HIV-1 pol gene among treatment-naïve individuals were similar when compared with previous data. One strain each had Y181C substitution, T74S and E35G substitutions in the Pr and one had A98G, K101R and L210FL substitutions in RT.

Keywords

non-clade B drug resistance genotyping testing HIV-1 India

INTRODUCTION

India has 2–3.1 million HIV-infected individuals as per the latest National AIDS Control Organization (NACO) surveillance reports with a prevalence of 0.36% in the general population.¹ The government health agencies such as NACO and other non-governmental organizations (NGOs) are taking a number of steps to control the spread of HIV infections in India. These steps have helped in bringing about a decline in the number of HIV infections in the country.^2,3 One of the steps taken to tackle HIV infections is to increase the accessibility of antiretroviral drugs by providing them free of cost through the National AIDS Control Program Phase III (NACP III) by NACO (www.naco.org). Antiretroviral treatment (ART) is known to reduce morbidity and mortality associated with HIV infection but this benefit is hampered by the emergence of drug resistance. Drug resistance cannot only contribute to reduction in the efficacy of ART but if unchecked can also lead to spread of drug resistance strains among the population.⁴ Numerous polymorphisms and mutations have been reported, which are known to contribute to drug resistance in clade B strains.⁵ There are a few reports from India on such changes in the clade C strains.^6,7 Previous reports from ART naïve individuals in India have shown protease (Pr) to be more polymorphic when compared with reverse transcriptase (RT).^6,7 The present report is based on findings of a study on baseline drug resistance mutations and polymorphisms in the Pr and RT genes of HIV-1 strains from infected individuals.

MATERIALS AND METHODS

Blood samples were collected from 93 HIV-1-infected treatment-naïve individuals. An institutional review board clearance was given to use archived samples and prospectively collected samples for all the analysis. Information regarding treatment was obtained from the patients and their medical charts were reviewed by the collaborating physicians for other clinical information. The sample size used in our study was to estimate a prevalence of 5% at a precision of ±5%,⁸ this was achieved by convenient sampling from patients visiting the infectious disease clinic of our hospital. The HIV-1-infected individuals had come to the clinical virology department of a tertiary care hospital in Vellore, south India for CD4+ T-cell estimation and/or HIV-1 viral load estimation during the year 2005–2007. Two milliliters of blood was collected in K₃EDTA vacutainer tubes for CD4 + T-cell estimation. For HIV-1 viral load estimation, blood was collected in Na EDTA tubes. Plasma was separated from these tubes after they were centrifuged at 1500 rpm for 10 minutes at 4°C. The plasma was stored as multiple aliquots at −80°C until the time of testing. Informed written consent was obtained from all the individuals before collection of the samples. The Guava^® Easy CD4™ System (Guava Technologies, Hayward, CA, USA), as described earlier, was used for CD4 estimation.⁹ The viral load estimation for all the individuals was done using Artus RealArt™ HIV-1 kits (Qiagen Hamburg Gmbh, Hilden, Germany) using Rotor-Gene™ 3000, as described earlier.¹⁰

The genotypic resistance assay was carried out by sequencing the Pr and RT genes. HIV-1 RNA was extracted from plasma using QIAamp^® viral RNA extraction kit (Qiagen GmbH, Hilden, Germany). The extracted RNA was then subjected to a one-step RT-polymerase chain reaction (PCR) using a one tube RT-PCR (Qiagen GmbH, Hilden, Germany) along with specific forward and reverse primers, HIV-1 out 1 and HIV-1 out 2, respectively (custom synthesized at Invitrogen, Carlsbad, CA, USA). The cycling conditions were as follows: 50°C for 30 minutes, 95°C for 15 minutes followed by 40 cycles of 94°C for 30 seconds, 60°C for 30 seconds, 72°C for 2 minutes and a final extension of 72°C for 7 minutes.

The first round products were further amplified using Qiagen Hot Start Master Mix (Qiagen GmbH, Hilden, Germany) with forward and reverse primers, PCR res 1 and PCR res 2, respectively. The cycling conditions were as follows: 95°C for 15 minutes followed by 23 cycles of 94°C for 30 seconds, 58°C for 30 seconds, 72°C for 2 minutes and a final extension of 72°C for 7 minutes. The amplification reactions were carried out on PTC-100 (MJ research, San Francisco, CA, USA) or MyCycler™ (BioRad, Hercules, CA, USA).

The amplified products were run on an ethidium bromide-stained 2% agarose (Sigma Aldrich Inc, St Louis, MO, USA) gel to check for the specific 1800 bp size amplicon. The agarose gel was visualized using the gel documentation system Geldoc 2000 (BioRad, CA, USA) using the software Quantity One version 4.1.1 (BioRad, CA, USA). The amplified products contained in their original amplification tubes were then purified with Millipore (Billerica, MA, USA) plates. The purified DNA was collected from the wells of the plate by pipetting out and transferred to PCR tubes which were then subjected to sequencing using the Big Dye terminator assay. The sequenced products were analysed on the ABI PRISM 310 Genetic Analyzer (PE Applied Biosystems, Foster City, CA, USA). The sequencing was performed at our institutional facility using sequencing primers that was performed using in-house primers. The complete protease gene was sequenced whereas only the first 300 amino acids were sequenced for the RT.⁸ Sites of nucleotide ambiguity (sites of mixture) were considered as mutation ‘N’ as recommended.⁸ There were no stop codons and nucleotide ambiguities were less than 5%.⁸ Consensus sequence was created using the BioEdit sequence alignment editor version 7.0.4.1 and aligned using ClustalW (http://www.ebi.ac.uk/Tools/clustalw/).¹¹ The sequences were analysed by Stanford HIV drug resistance database for genotypic drug resistance analysis (http://hivdb.stanford.edu). Sequences obtained were also submitted to REGA HIV-1 subtyping tool – v2.0 to identify the subtype of the strains (http://dbpartners.stanford.edu/RegaSubtyping/).

Phylogenetic analysis was performed for the 93 treatment-naïve sequences with Mega 4 software using a minimum evolution method with Kimura two-parameter and bootstrap value of 500 replicates.¹² The phylogenetic tree was constructed to confirm the subtyping analysis and also as a contamination control for the sequencing reactions as recommended earlier.⁸ These were compared with the HIV-1 pol sequences obtained from the Los Alamos HIV sequence data base. The sequences included were consensus sequences for M group, A1,A2,B,C, D,AE,F1,F2,G,H; Indian subtype C sequences (AF286232, AF067155, AF067157, AY713414, AY049708, DQ826595 and DQ826669); non-Indian subtype C sequences from China (AY967806), Ethiopia (AY255823), Myanmar (AB097871) and South Africa (AF544009 and DQ093604). The M group consensus sequence was used to root the tree.

The sequences were observed for amino acid substitutions that are known to confer drug resistance in clade B strains and for common amino acid substitutions in the Pr and RT sequences. Mutations conferring drug resistance in subtype B are defined as follows: for protease inhibitors (PIs) at 10, 24, 30, 32, 33, 36, 46, 47, 48, 50, 53, 54, 63, 71, 73, 77, 82, 84, 88, 90 and 93 (22 amino acid positions), nucleoside reverse transcriptase inhibitor (NRTI) resistance mutations at positions 41, 44, 62, 65, 67, 69, 70, 74, 75, 77, 115, 116, 118, 151, 184, 210, 215 and 219 (18 amino acid positions), and for non-nucleoside reverse transcriptase inhibitors (NNRTIs) at positions 98, 100, 101, 103, 106, 108, 179, 181, 188, 190, 225, 227, 230, 236 and 238 (15 amino acid positions) were specifically examined.⁵

For quality control randomly selected amplified products from 10 strains were sent to an external commercial centre along with the sequencing primers (1st Base, Science Park II, Singapore) for sequencing.

RESULTS

The mean HIV-1 viral load was log₁₀ 5.81 copies/mL and ranged from log₁₀ 3.04 copies/mL to log₁₀ 7.0 copies/mL with a median value of log₁₀ 5.33 copies/mL. The CD4 counts in this group ranged from 10 to 1037 cell/mm³ with a mean value of 287 cells/mm³ and with a median value of 221 cells/mm³. Among the 93 patients, there were 65 (70%) males and 28 females patients who had an age of range of 12–61 years, with a mean age of 37.4 and a median age of 36. The study population consisted of individuals from south (n = 81, 87%), east (n = 7, 8%) and central (n = 5, 5%) India. Data on modes of transmission and time of infection were not elicited for these individuals during blood collection to avoid privacy concerns.

In the phylogenetic tree, the study sequences were seen to segregate with consensus C at the major node denoting that majority of the strains were clade C. One strain was seen to closely match with the consensus A1 sequence. We obtained similar results using the REGA HIV-1 Subtyping Tool – v2.0, which identified 92 strains as clade C and the same strain as clade A1 based on Pr and RT sequence data. None of the strains were identical ruling out cross contamination.

In the protease gene, amino acid substitutions were seen at positions 10 (n = 3), 24 (n = 3), 35 (n = 12), 74 (n = 3) and 82 (n = 11). Positions known to confer drug resistance to NRTI and NNRTI also had amino acid substitutions. We found the substitutions for NRTI drug resistance at position 41 (n = 1) and for NNRTI they were at positions 100 (n = 1), 101 (n = 6), 103 (n = 2), 179 (n = 6) and 181 (n = 1). One strain had amino acid substitutions at positions 210 and 98 of the RT. The most common amino acid substitutions in the protease and RT genes at drug resistance conferring positions are listed in Table 1.

Table 1

Most common amino acid substitutions in the protease and reverse transcriptase genes at drug resistance conferring positions of the antiretroviral treatment-naïve HIV-1-infected individuals (n = 93)

Sl No.	Gene	Position	Amino acid change	No. (%)
1	Pr	L10	V (n = 1), I (n = 2)	3 (3)
2	Pr	L24	LIV (n = 1), LF (n = 2)	3 (3)
3	Pr	E35	G (n = 1), D (n = 11)	12 (12)
4	Pr	T74	A (n = 1), S (n = 2)	3 (3)
5	Pr	V82	I (n = 11)	11 (12)
6	RT	M41	L (n = 1)	1 (1)
7	RT	L100	FL (n = 1)	1 (1)
8	RT	K101	KN (n = 2), R (n = 2), Q (n = 2)	6 (6)
9	RT	K103	R (n = 2)	2 (2)
10	RT	V179	E (n = 1), V (n = 5)	6 (6)
11	RT	Y181	C (n = 1)	1 (1)

Pr = protease, RT = reverse transcriptase

All the 93 strains sequenced for the protease gene showed the amino acid substitution H69K. Ninety-two strains had an additional amino acid substitution at position 36. Other common sites of amino acid substitutions seen in the strains were at positions 15 (n = 85), 19 (n = 88), 36 (n = 93), 41 (n = 84), 63 (n = 86) and 93 (n = 89). The most common amino acid substitutions in the protease gene are listed in Table 2.

Table 2

Common amino acid substitutions seen in protease sequence of the antiretroviral treatment-naïve HIV-1-infected individuals (n = 93)

Sl No.	Position	Amino acid change	No. (%)
1	T12	S (n = 56), P (n = 8), A (n = 3), TAPS (n = 4), TIKR (n = 1)	72 (77)
2	I15	I → V	85 (91)
3	L19	I (n = 59), T (n = 25), V (n = 1), LIV (n = 3)	88 (96)
4	R41	R → K	84 (90)
5	M36	I (n = 74), V (n = 15), L (n = 2), ILV (n = 1)	92 (99)
6	L63	P (n = 54), T (n = 13), S (n = 8), A (n = 3), H (n = 3), APST (n = 2), Q, N, V (n = 1 each)	86 (93)
7	K69	H → K	93 (100)
8	I93	I → L	89 (96)

Amplification and sequencing of the RT gene revealed 93 strains with amino acid substitutions at positions 35, 39, 245 and 272. Amino acid substitutions were also seen at positions 60 (n = 87), 122 (n = 88), 173 (n = 90), 200 (n = 87), 207 (n = 90), 291 (n = 87) and 293 (n = 87). The most common amino acid substitutions observed in the RT are listed in Table 3.

Table 3

Common amino acid substitutions seen in reverse transcriptase sequence of the antiretroviral treatment-naïve HIV-1-infected individuals (1–300 amino acid positions) (n = 93)

Sl No.	Position	Amino acid change	No. (%)
1	V35	T (n = 87), K (n = 3), M (n = 2), I (n = 1)	93 (100)
2	T39	D (n = 62), E (n = 16), N (n = 12), I (n = 1), DN (n = 1), DE (n = 1)	93 (100)
3	S48	T (n = 81)	81 (87)
4	V60	I (n = 87)	87 (94)
5	D121	Y (n = 59), H (n = 12), C (n = 2), DHNY (n = 1)	74 (80)
6	K122	E (n = 87), A (n = 1)	88 (95)
7	K173	A (n = 62), T (n = 19), E (n = 3), V (n = 2), L (n = 2), IKRT (n = 1), APST (n = 1)	90 (97)
8	D177	E (n = 78), Q (n = 2), G (n = 2)	82 (88)
9	T200	A (n = 84), E (n = 2), V (n = 1)	87 (94)
10	Q207	E (n = 72), G (n = 9), K (n = 4), A (n = 3), N (n = 2)	90 (97)
11	V245	Q (n = 80), H (n = 4), E (n = 4), R (n = 1), K (n = 1), EK (n = 1), HQ (n = 1), V → *EKQ (n = 1)	93 (100)
12	A272	P (n = 93)	93 (100)
13	E291	D (n = 86), T (n = 1)	87 (94)
14	V292	I (n = 84), ILV (n = 2)	86 (92)
15	I293	V (n = 86), D (n = 1)	87 (94)

*Stop codon

The sequencing data for the 10 strains from the external centre were 100% concordant with sequencing data from our centre with regard to the mutations.

DISCUSSION

The most commonly occurring substitution in the protease amino acid sequence we observed were at position 69 (100%) followed by positions 19 (96%), 36 (99%) and 93 (96%). In the RT, amino acid substitutions were observed at positions 35, 39, 245 and 272 in all the 93 strains. Ninety (97%) strains showed amino acid substitutions at positions 173 and 207. Six strains had mutations at position 101 (6%), two strains showed mutations at position 103 (2%), six at 179 (6%) and one strain at position 181. When compared with subtype B, the protease sequences (99 amino acids) of the study strains varied at 28 positions, i.e. frequency of variation was 28%. The RT sequences (first 300 amino acids) on comparison had variations at 73 positions, i.e. frequency of variation was 24%.

In the absence of any drug exposure, protease sequences from B and non-B HIV-1 were shown to be polymorphic at 30% of the protease gene coded amino acid positions.¹³ In the RT about 40% of the first 240 amino acids were polymorphic.¹³ Some of these amino acid substitutions may occur at high rates in non-subtype B viruses at positions associated with drug resistance in subtype B. Such positions are 10, 20, 36, 63, 71, 77 and 93 for Pr and 69, 75, 98, 106, 118 and 179 for RT.¹³ It is not clear whether this may reduce the activity of NRTIs, NNRTIs and PIs or if this is a basis for the evolution of different primary resistance and cross-resistance pattern.¹⁴ In our study, over 90% of the strains showed substitutions at positions 36, 63 and 93 in the protease genes. It has been reported in clade B that the prevalence of these substitution increases in patients harbouring strains with multiple PI resistance mutations.¹⁵ At position 36 of the protease gene 80% showed the M36I substitution and 16% showed the M36V substitution. Ode et al. ¹⁶ have shown that the M36I, which is a non-active site mutation, decrease the volume of the binding cavity of the protease enzyme. M36V also reduces the volume of the binding cavity but not to the extent that M36I causes. It is shown that when the M36I occurs along with the D30 N in subtype B, it reduces the interaction with nelfinavir.¹⁶ Similar studies on the significance of mutations at positions 63 and 93 have not been published. At this point of time, we do not know which of these are likely to be responsible for treatment failure subsequently.

Reduced susceptibility to more than one PI is most likely to be associated with amino acid substitutions at six positions 10, 46, 54, 82, 84 and 90.¹⁷ Among our patients 11% had an amino acid change at position 82 (V82I). The V82I change has not been associated with any level of resistance to PIs,¹⁸ in contrast to V82 A/T/F/S. Reports from other centres in India have shown similar level of prevalence of this mutation.^6,7,19 Amino acid change at position 10 of the protease was seen in 3% of the study sequence and this included the sequence from the clade A1 strain. According to the Stanford HIV drug resistance database, the L10I/V, which were seen in our strains, is associated with resistance to each of the PIs when present with other mutations.

Data obtained by genotypic analyses of strains from different centres in India have shown several similar mutations.^6,7,19,20 In one study from India, wherein phenotypic analyses was carried out on treatment-naïve strains, primary resistance to PIs was 2.5% and RT inhibitors was 6.7%.²¹

The well-recognized E35D substitution was seen in 12% of the protease gene sequences in our study. This E35D substitution affects the conformational equilibrium between the closed and semi-open conformations of the free protease and also causes a significant reduction in its binding free energy of the protease for its substrate and amprenavir.²² It has also been shown that the E35D mutation reduces interaction with the HLA B44 molecule impeding cellular immune response.²² It is thus postulated to favour escape from the immune system in addition to conferring drug resistance.²² Based on the report from the Stanford HIV drug resistance database one of our strains had a E35G amino acid change, which according to the database is slightly more common in viruses from PI treated (particularly nelfinavir treated) compared with untreated persons. This particular strain also had a T74S amino acid change, which is associated with reduced susceptibility to nelfinavir. In total, 3% of the sequenced strains had this amino acid change at position 74.

The M41L amino acid substitution was seen in the RT of one of the strains and is known to cause high-level resistance when it occurs along with T215Y.²³ When M41L is present alone it causes low-level resistance to zidovudine and stavudine. The combination of mutations at M41L, L210W and T215F are also known as thymidine analog mutations (TAMs).²⁴ The strain which showed the M41L substitution in the RT also showed T74S substitution in its protease gene. One of our patients had the Y181C mutations, which is a mutation that is most frequently selected by nevirapine.²⁴ The Y181C mutation has been shown to reduce resistance to Zidovudine.²⁵ Two other NNRTI resistance conferring mutations were seen in one patient each. One of them was A98G, which is known to confer low level of resistance to NNRTI.²⁶ The strain with the A98G substitution also had L210FL substitution. The L210W substitution is a part of the TAMs. The other was V179E which causes potential low-level resistance to all the NNRTIs based on the scores given by the Stanford HIV drug resistance database. Six strains had amino acid substitution at position 101 of which two had K101KN and two strains each showed K101Q and K101R. K101N is known to cause resistance to nevirapine and delavirdine.²⁶ However, K101Q/R does not affect susceptibility to NNRTIs based on the report from the Stanford drug resistance database.

Amino acid substitutions were also seen at positions conferring resistance. These substitutions were, however, different from those known to confer drug resistance in clade B, e.g. L100FL (n = 1), K103R (n = 2) and V179I (n = 5). The L100I substitution confers intermediate resistance to the NNRTIs.²⁴ The codons for phenylalanine are UUU, UUC while for isoleucine one of the codons is AUU. Thus, a change in one nucleotide in a specific codon (U A) can lead to a phenylalanine by isoleucine substitution. This subsequent change in amino acid can lead to resistance among the NNRTIs. L100F could be an intermediate to L100I substitution in RT of this particular strain. The situation is analogous to the M184I intermediate change, which appears before the emergence of M184V strains.²⁷ The K103R substitution does not confer resistance by itself but in association with V179D/E can cause resistance to NNRTIs.²⁸ Five percent of the RT sequenced showed 179I substitution and 1% had V179E. The K103R mutation was not seen in combination with V179D/E. We could cautiously assume that the requirement for multiple mutations required for the development of resistance (genetic barrier) would be lesser in the strain with these mutations.

Based on data obtained from the Stanford HIV genotypic drug resistance database and REGA HIV-1 subtyping tool, the 92 sequences belonged to clade C and one belonged to A1. This is consistent with genotypic data for the strains from different regions of India, including the south where the predominant subtype is clade C.²⁹

While doing the genotyping-based drug resistance assay, we could postulate that the predominant circulating strain in a given patient (quasi-species) would have been amplified and sequenced.^24,30 It is reported that the drug resistance mutations could lower replication fitness on the virus.³¹ Hence, there is a biological possibility that a strain which has different mutations but with a lower degree of fitness may be not amplified. If certain mutations confer replicative advantage to the virus in the presence of the drug to which the virus has become resistance that particular quasi-species will outgrow the others.²⁴ It has been reported that amplification is of the predominant quasi-species that constitutes 30% or more of the circulating virus population.⁸

In countries where the standard initial and alternate antiretroviral regimens are restricted, it is important to evaluate if transmitted resistance has reached a population level that could affect the effectiveness of ART. Studies from India have not revealed any high prevalence of drug resistance conferring mutation.^6,7,19,20,32 These are still initial days in the treatment of HIV-infected individuals in India, but with plans to scale up ART in India it is necessary to have a HIV drug resistance surveillance network as per World Health Organization (WHO) protocols. As per WHO protocols, the drug resistance prevalence in a geographical area can be categorized into three categories <5%, 5–15% and >15%.³³ Surveillance for drug resistance and use of efficacious combination prevent the transmission of drug-resistant HIV-1 strains.

Based on our findings, we have observed no significant rise in drug resistance mutations or prevalence of known amino acid substitution in the treatment-naïve population from India.^6,7,19,21,32 We, however, suggest the implementation of resistance testing to monitor treatment-naïve population for early detection of drug resistance. In this study, we observed a Pr sequence with T74S and E35G substitution and a RT sequence with A98G, K101R and L210FL substitutions. If a strain with major mutations associated with resistance to drugs used routinely in the country is found two years in a row with no evidence of previous treatment, sentinel surveillance should be considered as per the WHO recommendation.⁸ This will help to keep a tab on transmitted drug resistance and prevent transmission of drug-resistant strains as has been reported from the west.⁵ Further studies are needed to confirm if mutation causing resistance in clade B have the same effects/mechanism in clade C HIV-1 strains.

Footnotes

ACKNOWLEDGEMENTS

We would like to thank the CMC fluid research fund for the partial financial support for the study through their grant.

GenBank Accession Numbers

The sequenced strains were submitted to the GenBank and have the following accession numbers:

EU030409, EU030410, EU030412-EU030418, EU106117- EU106126, EU110053-EU110081, EU110083, EU683752-EU683760, EU683763-EU683777, EU683779-EU638787, EU683789, EU683791-EU683793, EU683795, EU683797, EU683799-EU683800, EU744912-EU744914.

References

Technical Report India HIV estimates 2006. http://www.nacoonline.org/facts (last accessed 27 May 2008)

Kannangai

, Ramalingam

, Vijayakumar

, David

, Sridharan

. Is the human immunodeficiency virus (HIV) epidemic slowing down in India? Natl Med J India 2006;19:38–9

Kumar

, Jha

, Arora

, Trends in HIV-1 in young adults in south India from 2000 to 2004: a prevalence study. Lancet 2006;367:1164–72

Little

, Holte

, Routy

, Antiretroviral – drug resistance among patients recently infected with HIV. N Engl J Med 2002;347:385–94

Kantor

, Katzenstein

, Efron

, Impact of HIV-1 subtype and antiretroviral therapy on protease and reverse transcriptase genotype: Results of a global collaboration. PLoS Med 2005;2:e112

Balakrishnan

, Kumarasamy

, Kantor

, HIV type 1 genotypic variation in an antiretroviral treatment-naive population in southern India. AIDS Res Hum Retroviruses 2005;21:301–5

Arora

, Gupta

, Toor

, Singla

. Drug resistance-associated genotypic alterations in the pol gene of HIV type 1 isolates in ART-naive individuals in North India. AIDS Res Hum Retroviruses 2008;24:125–30

Guidelines for surveillance of HIV drug resistance (2003). www.who.int/3by5/publications/guidelines/en/resisguide12_12.pdf (last accessed December 2007)

Kandathil

, Kannangai

, David

, CD4+ and CD8+ T cell estimation by a micro-capillary cytometry technology compared with flow cytometry. Clin Diagn Lab Immunol 2005;12:1006–9

10.

Kannangai

, Kandathil

, Ebenezer

, Evidence for lower CD4+ T cell and higher viral load in asymptomatic HIV-1 infected individuals of India (south): Implications for therapy initiation. Indian J Med Microbiol 2008;26:217–21

11.

Hall

. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp Ser 1999;41:95–8

12.

Tamura

, Dudley

, Nei

, Kumar

. MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0. Mol Biol Evol 2007;24:1596–9

13.

Spira

, Wainberg

, Loemba

, Turner

, Brenner

. Impact of clade diversity on HIV-1 virulence, antiretroviral drug sensitivity and drug resistance. J Anti Chemother 2003;51:229–40

14.

Kantor

, Katzenstein

. Drug resistance in non-subtype B HIV-1. J Clin Vir 2004;29:152–9

15.

, Schiffer

, Gonzales

, Mutation patterns and structural correlates in human immunodeficiency virus type 1 protease following different protease inhibitor treatments. J Virol 2003;77:4836–47

16.

Ode

, Matsuyama

, Hata

, Computational characterization of structural role of the non-active site mutation M36I of human immunodeficiency virus type I protease. J Mol Biol 2007;370:598–607

17.

Turner

, Schapiro

, Brenner

, Wainberg

. The influence of protease inhibitor resistance profiles on selection of HIV therapy in treatment-naïve patients. Antivir Ther 2004;9:301–14

18.

King

, Winslow

, Garber

, Identification of a clinical isolate of HIV-1 with an isoleucine at position 82 of the protease which retains susceptibility to protease inhibitors. Antivir Res 1995;28:13–24

19.

Deshpande

, Recordon-Pinson

, Deshmukh

, Molecular characterization of HIV type 1 isolates from untreated patients of Mumbai (Bombay), India, and detection of rare resistance mutations. AIDS Res Hum Retroviruses 2004;20:1032–5

20.

Hira

, Panchal

, Parmar

, Bhatia

. High resistance to antiretroviral drugs: the Indian experience. Int J STD AIDS 2004;15:173–7

21.

Kandathil

, Kannangai

, Abraham

, Sudarsanam

, Pulimood

, Sridharan

. Genotypic resistance profile of HIV-1 protease gene: a preliminary report from Vellore, South India. Indian J Med Microbiol 2008;26:151–4

22.

Meiselbach

, Horn

, Harrer

, Sticht

. Insights into amprenavir resistance in E35D HIV-1 protease mutation from molecular dynamics and binding free-energy calculations. J Mol Model 2007;13:297–304

23.

Kellam

, Boucher

, Larder

. Fifth mutation in human immunodeficiency virus type 1 reverse transcriptase contributes to the development of high-level resistance to zidovudine. Proc Natl Acad Sci USA 1992;89:1934–8

24.

Clavel

, Hance

. HIV drug resistance. N Engl J Med 2004;250:1023–35

25.

Larder

. 3′-Azido-3′-deoxythymidine resistance suppressed by a mutation conferring human immunodeficiency virus type 1 resistance to nonnucleoside reverse transcriptase inhibitors. Antimicrob Agents Chemother 1992;36:2664–9

26.

Sahfer

. Genotypic testing for human immunodeficiency virus type 1 drug resistance. Clin Microbiol Rev 2002;15:247–77

27.

Keulen

, Back

, van Wijk

, Boucher

, Berkhout

. Initial appearance of the 184Ile variant in lamivudine-treated patients is caused by the mutational bias of human immunodeficiency virus type 1 reverse transcriptase. J Virol 1997;71:3346–50

28.

Parkin

, Gupta

, Chappey

, Petropoulos

. The K101P and K103R/V179D mutations in human immunodeficiency virus type 1 reverse transcriptase confer resistance to nonnucleoside reverse transcriptase inhibitors. Antimicrob Agents Chemother 2006;50:351–4

29.

Kandathil

, Ramalingam

, Kannangai

, David

, Sridharan

. Molecular epidemiology of HIV. Indian J Med Res 2005;121:333–44

30.

Domingo

, Martinez-Salas

, Sobrino

, The quasispecies (extremely heterogeneous) nature of viral RNA genome populations: biological revelance – a review. Gene 1985;40:1–8

31.

Dykes

, Demeter

. Clinical significance of human immunodeficiency virus type 1 replication fitness. Clin Microbiol Rev 2007;20:550–78

32.

Sen

, Tripathy

, Chimanpure

, Patil

, Bagul

, Paranjape

. Human immunodeficiency virus type 1 drug resistance mutations in peripheral blood mononuclear cell proviral DNA among antiretroviral treatment-naive and treatment-experienced patients from Pune, India. AIDS Res Hum Retroviruses 2007;23:489–97

33.

Protocol for evaluation of transmitted HIV drug resistance using specimens from HIV sentinel serosurveys in resource-limited settings. www.who.int/entity/hiv/drug_resistance/protocol_HIVDRevaluation.pdf (last accessed December 2007)