Impact of drug resistance-associated amino acid changes in HIV-1 subtype C on susceptibility to newer nonnucleoside reverse transcriptase inhibitors

Abstract

The objective of this study was to assess the phenotypic susceptibility of HIV-1 subtype C isolates, with nonnucleoside reverse transcriptase inhibitor (NNRTI) resistance-associated amino acid changes, to newer NNRTIs. A panel of 52 site-directed mutants and 38 clinically derived HIV-1 subtype C clones was created, and the isolates were assessed for phenotypic susceptibility to etravirine (ETR), rilpivirine (RPV), efavirenz (EFV), and nevirapine (NVP) in an in vitro single-cycle phenotypic assay. The amino acid substitutions E138Q/R, Y181I/V, and M230L conferred high-level resistance to ETR, while K101P and Y181I/V conferred high-level resistance to RPV. Y181C, a major NNRTI resistance-associated amino acid substitution, caused decreased susceptibility to ETR and, to a lesser extent, RPV when combined with other mutations. These included N348I and T369I, amino acid changes in the connection domain that are not generally assessed during resistance testing. However, the prevalence of these genotypes among subtype C sequences was, in most cases, <1%. The more common EFV/NVP resistance-associated substitutions, such as K103N, V106M, and G190A, had no major impact on ETR or RPV susceptibility. The low-level resistance to RPV and ETR conferred by E138K was not significantly enhanced in the presence of M184V/I, unlike for EFV and NVP. Among patient samples, 97% were resistant to EFV and/or NVP, while only 24% and 16% were resistant to ETR and RPV, respectively. Overall, only a few, relatively rare NNRTI resistance-associated amino acid substitutions caused resistance to ETR and/or RPV in an HIV-1 subtype C background, suggesting that these newer NNRTIs would be effective in NVP/EFV-experienced HIV-1 subtype C-infected patients.

FIG 1

Prevalence of NNRTI resistance-associated amino acid substitutions among first-line treatment failures. Sequences from 1,433 HIV-1 subtype C-infected patients failing first-line therapy with NNRTI resistance-associated mutations were obtained from the SATuRN (n = 766; light gray bars) and Stanford (n = 667; dark gray bars) databases. Amino acid substitutions are grouped according to prevalence: >10%, 5% to 10%, 1% to 5%, and <1%.

FIG 2

Phenotypic resistance of viruses with single amino acid substitutions associated with resistance to the NNRTIs ETR, RPV, EFV, and NVP. Single NNRTI resistance-associated mutations (n = 34) were introduced into the p8.MJ4 subtype C HIV-1 expression plasmid by site-directed mutagenesis. Phenotypic susceptibilities to ETR (A), RPV (B), EFV (C), and NVP (D) were determined using a single-cycle nonreplicative phenotypic assay. The lower cutoff values determined for this assay are indicated for each drug. Columns indicate mean values, while error bars indicate the standard errors of the means (SEMs). Each mutant was tested in duplicate in at least two independent experiments.

FIG 3

Phenotypic resistance to ETR, RPV, EFV, and NVP of viruses with double resistance-associated amino acid substitutions. Positively associated double mutations (n = 14) were introduced into the p8.MJ4 subtype C HIV-1 expression plasmid. Phenotypic susceptibilities to ETR (A), RPV (B), EFV (C), and NVP (D) were determined with a single-cycle nonreplicative phenotypic assay. The lower cutoff values determined for this assay are indicated for each drug. Columns indicate mean values, while error bars indicate the standard errors of the means (SEMs). Each mutant was tested in duplicate in at least two independent experiments.

FIG 4

Impact of E138K and M184IV on phenotypic susceptibility to ETR, RPV, EFV, and NVP. The phenotypic susceptibility to ETR, RPV, EFV, and NVP of site-directed mutants containing amino acid substitutions M184I, M184V, E138K, and their combinations were determined. The lower cutoff values determined for this assay are indicated for each drug. Columns indicate mean values, while error bars indicate the standard errors of the mean (SEMs). Each mutant was tested in duplicate in at least two independent experiments. n.s., not significant; **, P < 0.01 (2-way analysis of variance); ***, P < 0.001 (2-way analysis of variance).

FIG 5

Phenotypic susceptibility of mutants with single and double NNRTI resistance-associated mutations. The in vitro phenotypic responses of mutants with site-directed single and double NNRTI resistance-associated mutations, ranked according to the prevalence of the mutation among the 1,433 sequences with NNRTI resistance-associated mutations used in this analysis, are summarized. Some mutations (*) were not observed among the sequences but were included, as they are implicated in ETR and RPV resistance. Light gray shading, FCs of >3.6 (ETR), >2.6 (RPV), >3.8 (EFV), and >2.8 (NVP); dark gray shading, FC of >10.

FIG 6

Genotypic and phenotypic analysis of samples from EFV/NVP-experienced patients. Patient samples with NNRTI resistance-associated NNRTI resistance-associated amino acid substitutions were screened for NNRTI susceptibility in an in vitro phenotypic assay. The Stanford HIV Drug Resistance Database genotypic resistance scores are included for reference purposes. Light gray shading, FCs of >3.6 (ETR), >2.6 (RPV), >3.8 (EFV), and >2.8 (NVP); dark gray shading, FC of >10. Pt. patient number; STD, standard deviation.

FIG 7

Contribution of T369I to NNRTI resistance. Site-directed mutants containing the T369I amino acid substitution, an amino acid substitution found in the connection domain of patient 12 (Fig. 6), in combination with Y181C were tested for susceptibility to ETR, RPV, EFV, and NVP. The lower cutoff values are indicated for each drug. Columns indicate mean values, while error bars indicate the standard errors of the means (SEMs). Each mutant was tested in duplicate in at least two independent experiments. ***, P < 0.001 (2-way analysis of variance).

FIG 8

Contribution of N348I to NNRTI resistance. Site-directed mutants containing the N348I amino acid substitution, an amino acid substitution in the connection domain of some patients (Fig. 6), in combination with K103N or Y181C were tested for susceptibility to ETR, RPV, EFV, and NVP. The lower cutoff values are indicated for each drug. Columns indicate mean values, while error bars indicate the standard errors of the mean (SEMs). Each mutant was tested in duplicate in at least two independent experiments. n.s., not significant; *, P < 0.05 (2-way analysis of variance); ***, P < 0.001 (2-way analysis of variance).