To the Editor:
Increasing evidence suggests that given the extraordinary degree of genetic diversity of HIV-1, HIV subtypes and circulating recombinant forms probably can evolve in different patterns and pathways when they are under selective pressure. Viruses from some subtypes or geographic regions may have a greater propensity to develop resistance against certain drugs than do other viral variants.
In South America a high prevalence of BF intersubtype recombinant virus has recently been reported. 1,2 The main objective of our exploratory study was to analyze the frequency of drug resistance–associated mutations in 583 samples from Argentinean patients under treatment failure, this being the largest number of samples from patients under treatment failure ever studied in our country in search for drug resistance data. However, we decided to go further, performing a phylogenetic analysis of the pol gene sequences in order to classify them according to the subtypes. We found that mutation patterns in the studied samples were different, especially at the protease level, between samples belonging to different subtypes circulating in Argentina (B or BF recombinant subtypes).
Sequences were obtained using an Applied Biosystems 3700 DNA Sequencer for the resolution of the samples, as described elsewhere, 3,4 and assembled using Sequencher version 4.0.5 (Gene Codes Co., Ann Arbor, MI). The obtained sequences were analyzed using Drug Resistance Interpretation beta test (HIV RT and Protease Sequence Database, Stanford University, CA) (http://hivdb.stanford.edu/). 5 Phylogenetic analysis of the sequences was performed by Clustal X and visually corrected with the BioEdit version 5.0.9. 6 Phylogenetic trees were constructed by neighbor-joining method and recombination analysis was performed by bootscanning. Statistical differences were calculated by the χ2 test with Yates correction.
The overall frequency analysis showed that mutations most frequently detected in the reverse transcriptase (RT) coding region were the zidovudine (AZT)-specific mutation T215F/Y (54%), M41L (39%), the lamivudine (3TC)-specific mutation M184V (48%), and mutations D67N (32%) and L210W (25%) (Fig. 1A).
Prevalences of zalcitabine (ddC)-specific mutation T69D, stavudine (d4T)-specific mutation V75M/T, and didanosine (ddI)-specific mutation M74V were 5, 4, and 15%, respectively. Multidrug-resistant mutation Q151M was found in 3% of the samples. Five of 583 samples (0.85%) presented the multidrug-resistant mutation pattern A62V/V75I/F77L/F116Y/Q151M.
Among nonnucleoside reverse transcriptase inhibitor (NNRTI) resistance–associated mutations, frequencies of K103N, Y181C, and G190A were 28, 21, and 19%, respectively.
In the protease coding region, the most frequently found mutations were M36I (62%), L63P (47%), L10I/V (48%), I93L (45%), M46I/L (25%), I54V/T/L/M (30%), K20M/R (26%), V82A/F/T (31%), and L90M (30%) (Fig. 1B).
Phylogenetic analysis of pol gene sequences showed that 299 samples were from subtype B and 284 from BF intersubtype recombinant. After subtype assignment, statistically significant differences (P ≤ 0.05) (indicated in the figure with an asterisk) were found in the frequencies of RT amino acid positions E44D, A98G/S (A98G has been associated with low-level resistance to each of the NNRTIs), L210W (associated with resistance to AZT and d4T, and to a lesser extent, abacavir, ddI, ddC, and tenofovir (TDF)) and P225H (causes low-level resistance to efavirenz and possibly nevirapine. By itself, P225H causes increased susceptibility to delavirdine (DLV)) (Fig. 1A). These mutations were more frequently found in B than in BF samples (9.7% vs. 4.9%, 11.4% vs. 5.6%, 27.75% vs. 11.26%, and 3.68% vs. 0.7%, respectively).
Twelve of 20 PR positions analyzed (Fig. 1B) showed statistically significant differences between B and BF samples. On the one hand, mutations L10I/V, K20R/M, L24I, M36I, M46I (associated with resistance to indinavir), I54V/L (contribute to resistance to each of the 6 approved protease inhibitors), and V82A/F/T were more frequently found in BF samples. On the other hand, D30N, L63P/S/T/V/A/C, A71V/T, V77I, and I84V (causes phenotypic and clinical resistance to saquinavir, ritonavir, indinavir, nelfinavir (NVF), and amprenavir (APV) and contributes to lopinavir (LPV) resistance) were more prevalent in B subtype samples. It is important to highlight that I84V substitution, which is selected by all approved protease inhibitors, was found in a very low frequency, and it was lower in subtype BF samples (4.9% vs. 10% in B samples), even though these patients have received protease inhibitors as part of highly active antiretroviral therapy.
Mutation D30N, associated with resistance to NFV, which was more frequent in samples from the B group (8.7% vs. 3.8% of the BF samples) has been reported at a low frequency in samples from Argentinean drug-naive patients. 7
M36I was the predominant accessory substitution among BF recombinant strains (95.4%), and it was found in 29.4% of the B subtype samples. Changes in codon 63, the most polymorphic protease position, were found in 55.6% of the BF samples vs. 80.6% of the B subtype samples. It has been reported that approximately 45% of the B subtype isolates have the substitution L63P, whereas nearly 10% have other residues at this position. However, our results showed that 63% of the B and 30% of the BF samples were L63P, and 18.4 and 25% have other residues distinct to L at this position, respectively.
BF recombinant intersubtype samples showed a high frequency of mutations at position 20 of the PR (41.2% vs. 11.7% of the B subtype samples). Mutations at this position are considered accessories and do not cause drug resistance by themselves but do contribute to drug resistance when present with other protease mutations.
Also, changes at position 82 were more frequently found in BF samples (37.2% vs. 23% in B samples). This mutation causes resistance to indinavir, ritonavir, and LPV/r; however, when it is present with other protease inhibitor mutations, it contributes to phenotypic and clinical resistance to all the available protease inhibitors. 8–10
The overall frequency data showed here are very useful as a baseline characterization of the resistance genetic pool present in our population, which could be transmitted to healthy people.
Moreover, when samples were analyzed taking into account clustering into subtypes, we observed that several drug resistance–associated mutations in the protease region are more frequently found in samples from patients infected with BF recombinant variants. These differences are mainly due to secondary mutations, i.e., positions 10, 36, 63, 71, 77, and 93.
One consequence of preexisting accessory mutations might be the faster emergence of viruses resistant to protease inhibitors. Significant differences between subtype B and BF viruses may be associated with the development of distinct resistance patterns during therapy and may affect drug utility in patients infected with any of these subtypes.
It is remarkable that mutations V82A/F/T, which confer high levels of resistance to indinavir, ritonavir, and LPV, were found in a higher frequency in BF than in B samples. The relevance of this finding lies in the possibility of transmission of resistance variants than can evade the selective pressure imposed by these commonly used protease inhibitors.
In conclusion, we have found differences in the frequencies of drug resistance–associated mutations present in the pol region of samples belonging to B or BF intersubtype recombinant. Although not conclusive, these data provide a snapshot view of the HIV-1 pol gene diversity in treated patients from Argentina and strongly suggest that baseline polymorphisms in the BF intersubtype recombinant genome (especially in the protease coding region) could, at least in part, influence the mutational patterns that develop in patients in whom therapy fails. We believe this provides a rationale for performing clinical trials comparing efficacy and development of resistance in patients with subtype B and BF receiving identical therapies.
Mauricio Guillermo Carobene, MSc
Andrea Elena Rubio, MSc
Manuel Gómez Carrillo, PhD
Guillermo E. Maligne, MD
Gustavo Hernán Kijak, PhD
Jorge F. Quarleri, PhD
Horacio Salomón, PhD
1. Carr JK, Avila M, Gomez Carrillo M, et al. Diverse BF recombinants have spread widely since the introduction of HIV-1 into South America. AIDS. 2001; 15:F41–47.
2. Avila MM, Pando MA, Carrion G, et al. Two HIV-1 epidemics in Argentina: different genetic subtypes associated with different risk groups. J Acquir Immune Defic Syndr. 2002; 29:422–426.
3. Larder BA, Kohli A, Kellam P, et al. Quantitative detection of HIV-1 drug resistance mutations by automated DNA sequencing. Nature. 1993; 365:671–673.
4. Larder BA, Kellam P, Kemp SD. Convergent combination therapy can select viable multidrug-resistant HIV-1 in vitro. Nature. 1993; 365:451–453.
5. Kantor R, Machekano R, Gonzales MJ, et al. Human Immunodeficiency Virus Reverse Transcriptase and Protease Sequence Database: an expanded data model integrating natural language text and sequence analysis programs. Nucleic Acids Res. 2001; 29:296–299.
6. Hall TA. Bioedit: a user-friendly biological sequence alignment editor and analysis program for windows 95/98/nt. Nucleic Acids Symp Ser. 1999; 41:95–98.
7. Kijak GH, Pampuro SE, Avila MM, et al. Resistance profiles to antiretroviral drugs in HIV-1 drug-naive patients in Argentina. Antivir Ther. 2001; 6:71–77.
8. Condra JH, Holder DJ, Schleif WA, et al. Genetic correlates of in vivo viral resistance to indinavir, a human immunodeficiency virus type 1 protease inhibitor. J Virol. 1996; 70:8270–8276.
9. Molla A, Korneyeva M, Gao Q, et al. Ordered accumulation of mutations in HIV protease confers resistance to ritonavir. Nat Med. 1996; 2:760–766.
10. Sham HL, Kempf DJ, Molla A, et al. ABT-378, a highly potent inhibitor of the human immunodeficiency virus protease. Antimicrob Agents Chemother. 1998; 42:3218–3224.