According to international expert panels, in industrialized countries, drug resistance testing is recommended for women who are pregnant or those who require a treatment change because of treatment failure.1,2 In resource-limited settings, antiretroviral therapy has become more accessible but genotypic drug resistance testing is either unaffordable or hindered by on-site maintenance.
Plasma virion genotype analysis is considered the reference method to determine the drug resistance-related mutations. Mutations arise in the genes encoding reverse transcriptase (RT) and protease leading to amino acid substitutions associated with resistance to RT and protease inhibitors (http://hivdb.stanford.edu/hiv/). The interpretation of genotypic resistance tests depends on knowledge of the phenotypic and clinical impact of drug resistance mutations. HIV is divided into 2 types, HIV-1 and HIV-2, and HIV-1 is subdivided into 3 groups, M (main), O (outlier), and N (non-M/non-O). The pandemic group M viruses comprise 9 major subtypes3 and at least 16 circulating recombinant forms (http://www.hiv.lanl.gov/content/hiv-db/CRFs/CRFs.html), the genetic diversity resulting from intersubtype recombination is in constant increase. Genotypic algorithms have not yet been defined for highly divergent HIV-1 group O and HIV-2 and may be difficult to interpret for complex mutation patterns.
Recombinant virus assay (RVA) phenotype, where an infectious virus is generated by recombination of patient derived protease/RT sequences with a protease/RT-deleted HIV-1 backbone,4,5 has a very high cost. Phenotypic analyses measuring inhibition of viral enzymatic activity by specific drugs are easier to interpret, well suited to assess complex mutation patterns that may arise from combination therapy, and capable of bypassing the high diversity of HIV.
Nonnucleoside HIV-1 RT inhibitors (NNRTIs) are structurally diverse compounds that do not require intracellular activation by phosphorylation and bind noncompetitively to conserved residues of the RT p66 subunit. NNRTIs share a common mechanism of action, altering the conformational structure and modifying the active site of the enzyme.6
Nevirapine is a powerful tool to reduce mother-to-child HIV transmission in countries in the process of health care development,7-9 but the high frequency of appearance of consequent mutations requires careful supervision for secondary treatment.9-11
We have previously demonstrated that RT activity quantification in plasma samples correlated significantly with RNA copies and was considered an alternative assay for monitoring viral loads.12 We evaluated a basic assay, easy to perform without sophisticated technology, to monitor NNRTI resistance. We investigated the susceptibility of RT directly recovered from plasma to 3 NNRTIs, efavirenz (EFV), nevirapine (NVP), and delavirdine (DLV).
We also compared RT phenotype of samples from patients infected by HIV-1 group M, group O, and HIV-2 with genotype analyses. For some patients, RT-based phenotype was compared with RVA phenotype.
Plasma and Cell Sample Collection
Preserved blood samples, collected and transported at ambient temperature, were centrifuged within 6 hours of collection; aliquots of plasma were stored at −80°C. Peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll-Hypaque gradients and eventually stored in liquid nitrogen.
Population and Viruses
Primary isolates studied included 22 HIV-1 group M, 9 HIV-1 group O, and 7 HIV-2 samples.
The 14 HIV-1 group M patients included in the study were treated with antiretroviral drugs including an NNRTI (EFV or NVP), except for Patient 10, who did not receive an NNRTI. Eighteen plasma and 4 cell culture supernatants of viral amplification were assayed. Viral amplification was performed on HIV-seronegative donor PBMCs cocultured with fresh patient PBMCs for Patients 07 and 10.2, or infected with plasma for Patients 04.3 and 05.2.
Six HIV-1 group O previously isolated from treatment-naive Cameroonian patients13 were amplified in PBMC culture. YBF26, YBF32, YBF38, and YBF39 are group O:A viruses. YBF35 is a group O:B/:A recombinant virus. YBF37 is an unclassified group O virus close to the prototypic VAU isolate.14 RBF125 group O:B virus in a patient from Cameroon15 was isolated by coculture with patient PBMCs obtained before introduction of antiretroviral treatment (RBF125.1) and 8 months after the last administration of a 10-month antiretroviral regimen including EFV (RBF125.2, RBF125.3).
Seven HIV-2 viruses from treatment-naive patients were isolated and amplified in PBMC culture.
Reference strains HIV-1 NL4-316 and HIV-2 (clade A) Rod17 strains were amplified in CEM cells. HIV-1 YU-218 and HIV-2 (clade B) P1102 strains (GenBank accession number AF170058) were propagated in HIV-seronegative donor PBMCs. Aliquots of culture supernatant were stored at −80°C.
Phenotypic Analysis of RT Resistance to NNRTIs
Lysate containing intraviral RT was recovered from plasma and quantified using the ExaVir load, RT activity viral load kit (Cavidi Tech AB, Uppsala, Sweden), as previously described.12 RT activity is expressed in fg/mL. The susceptibility of RT activity to NNRTI was measured using the Cavidi HS Lenti kit (Cavidi Tech AB) by running quantitative RT reactions in the presence of serial drug concentrations. Drugs were serially diluted in 5-fold steps and 25 μL added to 100 μL of RT reaction buffer and 50 μL of lysate containing 40 fg/mL RT activity. The percentage of inhibition was calculated for each drug concentration as the ratio of the RT activity level obtained in the presence of drug to that obtained in RT reactions done in the absence of drug (×100). The drug concentration resulting in 50% inhibition (IC50, expressed as μM) was calculated with the Median effect equation19 using CalcuSyn software (Biosoft, Cambridge, UK).
Two reference recombinant RTs (rWT, the standard RT derived from BH10) and its mutant form rY181C were purchased from Cavidi Tech AB.
The NNRTI drugs, DLV, NVP, and EFZ, were kindly provided by Pfizer, Inc., Boehringer-Ingelheim Pharmaceuticals, Inc., and Bristol-Myers Squibb Co., respectively.
Scales of drug used were from 32 nM up to 100,000 nM for DLV and NVP and from 6.4 nM up to 20,000 nM for EFV.
Between-Run IC50 Variation
IC50 variations were measured using 40 fg RT activity of rWT, NL4-3, and YU-2, in 9 (YU-2) to 14 (rWT, NL4-3) independent experiments. The respective means, SDs, and 95% CIs were calculated.
IC50 Variation Determined With 40 fg and 400 fg RT Activity
At least 3 experiments were performed using 40 fg and 400 fg RT activity of rWT, NL4-3, and YU-2. IC50 variations due to the 10-fold increase of the RT activity were calculated as the mean ratio of IC50 obtained with 400 fg RT activity to IC50 obtained with 40 fg RT activity. The respective SDs and 95% CIs were calculated.
DNA Sequence and Genotype Analyses
Results of the genomic analysis are reported as amino acid changes at positions along the RT gene compared with the HXB2 reference sequence (GenBank accession number K03455).
Genotypes were performed for each HIV-1 group M plasma and viral supernatant. RNA was isolated using QIAamp viral RNA mini kit (Qiagen GmbH, Germany). Reverse transcription to cDNA and sequencing of both strands were performed with the TRUGENE HIV-1 genotyping kit (Visible Genetics, Inc., Epinay-sur-Orge, France). The sequence covered the RT region coding for AAs 38 to 247. NL4-3 and YU-2 pol RT sequences were identical to original sequences described (GenBank accession numbers U26942 and M93258).
Proviral DNA was extracted from HIV-2 cell cultures using QIAamp DNA blood mini kit (Qiagen). RT genomic region was amplified by nested polymerase chain reaction (PCR) with outer (RTC and RT2) and inner (RT3 and RT4) primers as described by Rodes et al.20 The PCR products were purified using the QIAquick PCR purification kit (Qiagen) and sequenced on both strands with sense (RT3 and RT5) and antisense (RT4 and RT6) primers as described by Rodes et al.20 Sequencing reaction was carried out using the Dye Terminator kit (Applied Biosystems, Foster City, CA) and run on an automated capillary ABI Prism 310 DNA sequencer (Perkin Elmer, Wellesley, MA). The sequence covered the RT region coding for AAs 47 to 372. Sequence alignment and phylogenetic analysis were performed with Sequence Navigator software (Applied Biosystems) and ClustalW programs. HIV-2 pol RT genes newly sequenced were compared with pol RT gene sequence from the reference strain ROD (GenBank accession numbers M15390).
Viral RNA was extracted from HIV-1 group O cell supernatant using QIAamp viral RNA mini kit (Qiagen) and then transcribed to cDNA using the Omniscript Reverse Transcriptase kit (Qiagen). PCR and nested PCR were performed using the High Fidelity PCR Master kit (Roche Diagnostics, GmbH, Meylan, France). A first round of PCR generated a fragment of 2134 bp with forward primer GAGCAM U2 (5′-GCATGGGTAAAGGCAGTA-3′) and reverse primer POL-O-L1 (5′-CTAATTCCTTGATAGATTTGACT-3′). Two overlapping RT genomic regions were then amplified by nested PCR with respective forward and reverse primers (POL-O-F1 5′-CAGTATTRGTGGGACCTACTCCTGTTA-3′ and POL-O-R3 5′-GGGTCTCATTGTTYACACTAGGAATAG-3′) and (POL-O-U1 5′-GCAATCTGTTACAGTYTTAGATG-3′ and POL-O-L1), generating respectively 495-bp and 516-bp fragments. Primers were synthesized by Eurogentec (Seraing, Belgium). The first round of PCR had an initial denaturation of 94°C for 2 minutes, 30 cycles of denaturation at 94°C for 15 seconds, annealing at 55°C for 30 seconds, and extension at 72°C for 3 minutes followed by a last extension at 72°C for 10 minutes. The nested PCR had an initial denaturation of 94°C for 2 minutes, 10 cycles of denaturation at 94°C for 10 seconds, annealing at 55°C for 30 seconds, and extension at 72°C for 45 seconds, 17 cycles of denaturation at 94°C for 15 seconds, annealing at 55°C for 30 seconds, and extension at 72°C for 45 seconds successively incremented of 5 seconds at each cycle, followed by a last extension at 72°C for 7 minutes. Amplified products were purified on columns with the QIA quick PCR purification kit (Qiagen). The 2 DNA purified fragments were sequenced on both strands with respective sense and antisense primers (POL-O-F1 and POL-O-R3; POL-O-U1 and POL-O-L1). Sequencing reactions were carried out using the CEQ DTCS Quick Start Kit (Beckman Coulter, Roissy CDG, France) and run on an automated capillary CEQ 8000 Genetic Analysis System DNA sequencer (Beckman Coulter). The sequences covered the RT region coding for AAs 1 to 261. Sequence alignment was performed with ClustalW version 1.8 programs and compared with the ANT70 reference sequence (GenBank accession number L20587).
Determination of IC50 Cut-Off Values
Variations of the 3 NNRTI concentrations inhibiting 50% of RT activity (IC50) were determined for 2 wild-type viruses (NL4-3 and YU-2) and 1 wild-type recombinant RT (rWT) (Table 1). Mean between-run IC50 values were the lowest for rWT and the highest for YU-2; 0.09-0.63 μM for EFV, 0.51-4.9 μM for NVP, and 0.90-19 μM for DLV. Susceptibility cut-off values were then defined as 2 SD from the highest IC50 mean for each drug, ie, 1.5 μM for EFV, 11 μM for NVP, and 37 μM for DLV. Viruses with IC50 values ranging between the cut-off and 20 μM for EFVor 100 μM for NVP and DLV were considered of intermediate resistance and above as highly resistant.
Fold-change variations between IC50 were determined for 40 fg and 400 fg of RT activity of rWT, NL4-3, and YU-2 (Table 2). As the SD for NL4-3 was superior to the 95% CI, we considered that IC50 equivalence obtained with 40 fg and 400 fg RT activity could not be shown and chose to perform further experiments with 40 fg RT activity.
Plasma RT Activity Phenotype Correlated With RVA-Based Phenotype Analysis
Nine plasma samples collected from 9 NNRTI-experienced patients with viral load >4.4 log10 copies/mL were subjected to RT phenotype and RVA PhenoSense analysis (ViroLogic, Inc., South San Francisco, CA) (Table 3). Both analyses were performed based on identical samples. RT susceptibility was expressed in drug IC50 and cut-off values were calculated according to between-run variations. PhenoSense assay was expressed in fold-change in drug susceptibility compared with drug-susceptible reference virus4 and fold-changes >2.5 were considered evidence of reduced susceptibility. RT-based and RVA-based phenotypes were in complete agreement. Six plasma samples (Patients 02, 06, 11.1, 12.1, 13, and 14) were highly resistant to EFV, NVP, and DLV; the genotype analysis showed at least the 103N mutation in 5 cases and the association 101E-181C-190S in 1 case. Three plasma samples (Patients 03, 08.1, and 09) were susceptible to the 3 drugs. Genotype analysis was in agreement for Patients 08.1 and 09 with no evidence of mutation associated with NNRTI resistance; however, the mutations 101E and 106I/V in Patient 03 viral population would have predicted an NNRTI resistance with algorithm interpretation from ANRS 2003.10, HIVDB 2004.04 and Rega Institute 6.2 (algorithms available online http://hivdb.stanford.edu/asi/).
High Correlation of RT Susceptibility and Genotype Mutation Profile
The 181C mutation of the recombinant RT correlated with a total resistance to NVP only, which could emphasize that recombinant RT might behave differently from wild-type viral RT. For the remainder of the HIV-1 group M viruses, RT phenotype was compared with genotype analyses. Five viruses were totally resistant to the 3 NNRTIs and the phenotype correlated with the presence of resistance mutations (103N or the association of 101E-181C-190S). The Patient 05.2 virus was amplified starting from Patient 05.1 plasma and lost 103N-190A mutations during culture. It remained totally resistant to NVP and DLV but recovered an intermediate susceptibility to EFV despite the presence of 108I-181C. Seven viruses did not comprise any mutation associated with NNRTI resistance and were susceptible to the 3 NNRTIs. We observed that for Patients 10.1 and 10.2, who never received NNRTIs, the DLV IC50 was much higher than the one obtained for references rWT RT or NL4-3 virus, in agreement with the high cut-off value that was considered for DLV.
RT Activity Phenotype of HIV-2 Viruses
The 7 HIV-2 isolates (clade A) tested were highly resistant to the 3 NNRTIs, as were the 2 reference strains ROD and P1102 (IC50 >20 μM for EFV and >100 μM for NVP and DLV). The 7 isolates genotyped harbored the 106I-181I-188L-190A amino acid substitutions naturally present in HIV-2 and commonly linked to NNRTI resistance in HIV-1 group M.20-22
RT Activity Phenotype of HIV-1 Group O Viruses
Table 4 shows RT activity phenotype investigated in 9 viruses isolated from group O-infected patients. Six isolates were highly resistant to the 3 NNRTIs and 3 isolates were susceptible to some NNRTIs. According to the IC50 cut-off values that we selected, YBF39, YBF35, and YBF37 had an intermediate resistance to EFV and were susceptible to NVP. YBF39 and YBF37 were susceptible to DLV whereas YBF35 was highly resistant to DLV. The primary structure derived from the RT sequences corresponds to GenBank accession numbers AY694423, AY694424, AY694425, AY694426, AY694427, AY694428, and AY694429. At position involved in group M resistance to NNRTIs, the mutations 98G-179E were common to the 9 isolates, irrespective of their susceptibility to NNRTIs. Mutations 135V-245Q were found in 5 isolates but were independent of the level of resistance. Only 3 of the 6 highly resistant isolates comprised the 181C mutation. For the 3 others, isolated at different times from 1 patient (RBF125) harboring the wild 181Y, the high resistance could be linked to the association of 103R and 135L. YBF35 isolate, highly resistant to DLV, harbored the 245K not found in the other isolates. Only 2 other mutations, not found in the other 8 isolates, were shown. YBF35 contained 178M and 228Q while the other isolates had 178I/L and 228L. These mutations have not yet been linked to NNRTI resistance in group M viruses.
With the constant increase in HIV diversity, attempts to amplify highly divergent sequences encounter more and more difficulties in either genotype profile or RVA phenotype analyses.23 Countries most affected by highly divergent HIV infections are also those with limited access to expensive and sophisticated technologies. Enzymatic RT phenotype analysis not only bypasses the problem of HIV diversity and technical feasibility but also provides the advantages of lower cost and easier interpretation. Although clinical cut-offs for NNRTI are not yet defined24 and thresholds must be selected arbitrarily, relying on wild-type viruses used as a reference, the test at least allows differentiation between susceptible and resistant viral populations in a large panel of samples. This type of assay has been previously proposed by Shao et al,25 showing that RT derived from virions from plasma with down to 10,000 RNA copies/mL could be used for analysis of phenotypic drug susceptibility. Plasma RT phenotype was considered reliable compared with RVA phenotype and genotype analyses to monitor resistance to NNRTIs, irrespective of HIV types or groups. Results in a large number of samples could be obtained in <5 days, including the quantification of RT activity in plasma followed by the phenotype assay. This assay may be of interest for new drug resistance profiles and for epidemiologic studies prior to a larger administration of nevirapine to prevent HIV perinatal transmission.
The NNRTI IC50 values obtained with the recombinant wild-type RT (rWT) were equivalent to those previously reported.25 We observed, however, higher IC50 values with wild-type virion RT (NL4-3 and YU-2), which could reflect differences in dimerization with recombinant RT forms. Our choice to use wild-type virion RT, rather than recombinant RT, as a reference was consolidated by the phenotype obtained with the rY181C RT. We showed that this recombinant RT was resistant to NVP only, and not to DLV, even though previous reported studies have shown that clinical isolates containing the single amino acid substitution 181C were also resistant to DLV.4,26,27
To our knowledge, the susceptibility of the wild-type YU-2 strain has not yet been previously reported. The YU-2 strain may reflect the natural variability of susceptibility in the absence of NNRTI resistance-associated mutation. We found higher IC50 values than those obtained with NL4-3, particularly for DLV, which could be related to the 142T and 178L polymorphisms associated with elevated IC50 values for DLV.24 The high DLV IC50 of primary viruses 04.1 and 10 without NNRTI-associated mutation could also account for the natural variability of susceptibility to DLV. These results are consolidated by the increase of original cut-off values defined for NNRTIs when taking into account the natural variability in drug susceptibility. RVA phenotype fold-change in the IC50 cut-offs has now been established at 6-, 8-, and 10-fold for AntiVirogram5 (Virco Lab, Inc., South Raritan, NJ) and at 3-, 4.5-, and 6.2-fold for PhenoSense24 for EFV, NVP, and DLV, respectively.
Among the 22 group M isolates studied, only 1 discrepancy was observed between phenotype analyses and genotype interpretations. The 03 virus involved emphasizes that complex mutation patterns may be difficult to interpret with genotype analyses and are dependent on the algorithm used.28-30 The loss of 103N-190A mutations found in Patient 05.1 plasma virus concomitant to 05.2 virus isolation in culture strengthens the utility of RT phenotype performed directly from plasma samples. While 05.1 virus was highly resistant to the 3 NNRTIs, 05.2, harboring 108I-181C mutations, was highly resistant to NVP and DLV but showed only intermediate resistance to EFV.
In highly divergent viruses, it was not surprising that all HIV-2 samples investigated were highly resistant to the 3 NNRTIs. These results are in agreement with genotype analyses and previous studies reporting that NNRTIs do not have a significant antiviral effect on HIV-2.20,21
Studies on HIV-O phenotype are scarce. Due to the constant 98G and frequent 181C natural polymorphisms conferring low levels and high NNRTI resistance, the use of this class of drug is not recommended in this group. However, NNRTI resistance cannot be solely linked to these mutations, as we found that RBF125 isolates were highly resistant to the 3 NNRTIs despite the absence of the 181C mutation. In addition to constant 98G-179E and frequent 245Q mutations found in group O viruses, RBF125 harbored 103R-135L mutations. The 103R polymorphism has already been reported in isolates susceptible to NNRTIs31; however, 135L polymorphism has not yet been reported in group O31-33 nor in any of the group M viruses we investigated but could be involved in NNRTI resistance. In fact, the 135M/L polymorphism in conjunction with 283I has been shown to confer a decreased 4- to 5-fold susceptibility to EFV, NVP, and DLV.34
In contrast, this study shows that, for some group O viruses not harboring 181C, a susceptibility to some NNRTIs could be demonstrated at least in vitro. Three group O 181Y viruses (YBF39, YBF37, YBF35) showed a susceptibility to NVP and an intermediate resistance to EFV. Two of them were also susceptible to DLV. The YBF35 virus, highly resistant to DLV, harbored the seldom-found 245K mutation.31,32 The 245K has only been previously reported in 2 related isolates that harbored 181C and consequently the role of 245K could not be shown.35 The only other polymorphism distinguishing YBF35 from the others31-33 and our study was the 178M mutation. Polymorphism 245K or 178M may represent DLV resistance-conferring mutation. Although not investigated in the present work, a possible involvement of the 318F mutation in DLV resistance cannot be excluded.36
In summary, RT phenotype to NNRTIs is easily adaptable to laboratories as only microplate reader is required. Although clinical evaluations are still required, the future launch of an Exavir Drug kit25 should decrease IC50 variations and improve cut-off values. It would also allow for fundamental studies to better understand the relationship between the structure and the RT function. RT phenotype could be useful in the determination of new resistance profiles and would be particularly useful in the choice of therapeutic and prophylactic options during pregnancy.
The authors thank Dr. David Loeber for help with clinical files and the technicians of the Virology Laboratory, CHU, Rouen. The authors thank Richard Medeiros, Rouen University Hospital Medical Editor, for his valuable advice in editing the manuscript. HIV-2 P1102 strain was kindly provided by Gilles Collin, Bichat University Hospital, Paris.
1. EuroGuidelines. Clinical and laboratory guidelines for the use of HIV
-1 drug resistance testing as part of treatment management: recommendations for the European setting. The EuroGuidelines Group for HIV
2. Yeni PG, Hammer SM, Carpenter CC, et al. Antiretroviral treatment for adult HIV
infection in 2002: updated recommendations of the International AIDS Society-USA Panel. JAMA.
3. Thomson MM, Perez-Alvarez L, Najera R. Molecular epidemiology of HIV
-1 genetic forms and its significance for vaccine development and therapy. Lancet Infect Dis.
4. Petropoulos CJ, Parkin NT, Limoli KL, et al. A novel phenotypic drug susceptibility assay for human immunodeficiency virus type 1. Antimicrob Agents Chemother.
5. Harrigan PR, Montaner JS, Wegner SA, et al. World-wide variation in HIV
-1 phenotypic susceptibility in untreated individuals: biologically relevant values for resistance testing. AIDS.
6. Tantillo C, Ding J, Jacobo-Molina A, et al. Locations of anti-AIDS drug binding sites and resistance mutations in the three-dimensional structure of HIV
-1 reverse transcriptase: implications for mechanisms of drug inhibition and resistance. J Mol Biol.
7. Ayouba A, Tene G, Cunin P, et al. Low rate of mother-to-child transmission of HIV
-1 after nevirapine intervention in a pilot public health program in Yaounde, Cameroon. J Acquir Immune Defic Syndr.
8. Lallemant M, Jourdain G, Le Coeur S, et al. A randomized, double-blind trial assessing the efficacy of single-dose perinatal nevirapine added to a standard zidovudine regimen for the prevention of mother-to-child transmission of HIV
-1 in Thailand. Paper presented at: 11th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2004; San Francisco, CA. Abstract 40LB.
9. Chalermchokcharoenkit A, Asavapiriyanont S, Teeraratkul A, et al. Combination short-course zidovudine plus 2-dose nevirapine for prevention of mother-to-child transmission: safety, tolerance, transmission, and resistance results. Paper presented at: 11th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2004; San Francisco, CA. Abstract 96.
10. Jourdain G, Ngo-Giang-Huong N, Tungyai P, et al. Exposure to intrapartum single-dose nevirapine and subsequent maternal 6-months response to NNRTI-based regimens. Paper presented at: 11th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2004; San Francisco, CA. Abstract 41LB.
11. Martinson N, Morris L, Gray G, et al. HIV
resistance and transmission following single-dose nevirapine in a PMTCT cohort. Paper presented at: 11th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2004; San Francisco, CA. Abstract 38.
12. Braun J, Plantier JC, Hellot MF, et al. A new quantitative HIV
load assay based on plasma virion reverse transcriptase activity
for the different types, groups and subtypes. AIDS.
13. Mauclere P, Loussert-Ajaka I, Damond F, et al. Serological and virological characterization of HIV
-1 group O infection in Cameroon. AIDS.
14. Roques P, Robertson DL, Souquiere S, et al. Phylogenetic analysis of 49 newly derived HIV
-1 group O strains: high viral diversity but no group M-like subtype structure. Virology.
15. Gueudin M, Plantier JC, Damond F, et al. Plasma viral RNA assay in HIV
-1 group O infection by real-time PCR. J Virol Methods.
16. Adachi A, Gendelman HE, Koenig S, et al. Production of acquired immunodeficiency syndrome-associated retrovirus in human and nonhuman cells transfected with an infectious molecular clone. J Virol.
17. Clavel F, Guyader M, Guetard D, et al. Molecular cloning and polymorphism of the human immune deficiency virus type 2. Nature.
18. Li Y, Hui H, Burgess CJ, et al. Complete nucleotide sequence, genome organization, and biological properties of human immunodeficiency virus type 1 in vivo: evidence for limited defectiveness and complementation. J Virol.
19. Chou TC, Talalay P. Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul.
20. Rodes B, Holguin A, Soriano V, et al. Emergence of drug resistance mutations in human immunodeficiency virus type 2-infected subjects undergoing antiretroviral therapy. J Clin Microbiol.
21. Isaka Y, Miki S, Kawauchi S, et al. A single amino acid change at Leu-188 in the reverse transcriptase of HIV
-2 and SIV renders them sensitive to non-nucleoside reverse transcriptase inhibitors. Arch Virol.
22. Bacheler L, Jeffrey S, Hanna G, et al. Genotypic correlates of phenotypic resistance to efavirenz in virus isolates from patients failing nonnucleoside reverse transcriptase inhibitor therapy. J Virol.
23. Spira S, Wainberg MA, Loemba H, et al. Impact of clade diversity on HIV
-1 virulence, antiretroviral drug sensitivity and drug resistance. J Antimicrob Chemother.
24. Parkin NT, Hellmann NS, Whitcomb JM, et al. Natural variation of drug susceptibility in wild-type human immunodeficiency virus type 1. Antimicrob Agents Chemother.
25. Shao XW, Malmsten A, Lennerstrand J, et al. Use of HIV
-1 reverse transcriptase recovered from human plasma for phenotypic drug susceptibility testing. AIDS.
26. Casado JL, Hertogs K, Ruiz L, et al. Non-nucleoside reverse transcriptase inhibitor resistance among patients failing a nevirapine plus protease inhibitor-containing regimen. AIDS.
27. Demeter LM, Shafer RW, Meehan PM, et al. Delavirdine susceptibilities and associated reverse transcriptase mutations in human immunodeficiency virus type 1 isolates from patients in a phase I/II trial of delavirdine monotherapy (ACTG 260). Antimicrob Agents Chemother.
28. Dunne AL, Mitchell FM, Coberly SK, et al. Comparison of genotyping and phenotyping methods for determining susceptibility of HIV
-1 to antiretroviral drugs. AIDS.
29. Shafer RW, Gonzales MJ, Brun-Vezinet F. Online comparison of HIV
-1 drug resistance algorithms identifies rates and causes of discordance interpretations. Antivir Ther.
30. Ravela J, Betts BJ, Brun-Vezinet F, et al. HIV
-1 protease and reverse transcriptase mutation patterns responsible for discordances between genotypic drug resistance interpretation algorithms. J Acquir Immune Defic Syndr.
31. Descamps D, Collin G, Letourneur F, et al. Susceptibility of human immunodeficiency virus type 1 group O isolates to antiretroviral agents: in vitro phenotypic and genotypic analyses. J Virol.
32. Quinones-Mateu ME, Albright JL, Mas A, et al. Analysis of pol gene heterogeneity, viral quasispecies, and drug resistance in individuals infected with group O strains of human immunodeficiency virus type 1. J Virol.
33. de Baar MP, Janssens W, de Ronde A, et al. Natural residues versus antiretroviral drug-selected mutations in HIV
type 1 group O reverse transcriptase and protease related to virological drug failure in vivo. AIDS Res Hum Retroviruses.
34. Brown AJ, Precious HM, Whitcomb JM, et al. Reduced susceptibility of human immunodeficiency virus type 1 (HIV
-1) from patients with primary HIV
infection to nonnucleoside reverse transcriptase inhibitors is associated with variation at novel amino acid sites. J Virol.
35. Quinones-Mateu ME, Soriano V, Domingo E, et al. Characterization of the reverse transcriptase of a human immunodeficiency virus type 1 group O isolate. Virology.
36. Harrigan PR, Salim M, Stammers DK, et al. A mutation in the 3′ region of the human immunodeficiency virus type 1 reverse transcriptase (Y318F) associated with nonnucleoside reverse transcriptase inhibitor resistance. J Virol.