Evolution of the gp41 env region in HIV-infected patients receiving T-20, a fusion inhibitor

Abstract

T-20 belongs to a new family of antiretroviral drugs that inhibits HIV entry. Resistance may developin vitroas result of changes within the GIV motif of the HIV-1 gp41. However, none of four individuals who failed treatment with T-20 selected those changes. One developed a change (G←V) at position 36 not previously described. Therefore, other changes in theenvgene might be responsible for the loss of susceptibility to T-20in vivo. T-20 is a 36 amino acid synthetic peptide that mimics the HR2 region of the HIV-1 gp41. Its mechanism of action is little understood but it seems to interact competitively with HR1 to prevent the fusion of viral and cellular membranes [1]. Few data are available on the evolution of the env gene in patients undergoing treatment with T-20. In-vitro studies have defined a highly conserved three amino acid motif, GIV, within the HR1 region of gp41 to be critical for the loss of susceptibility to T-20 [2]. In particular, two changes in this motif at positions 36 (G←S) and 38 (V←M) have been associated in vitro with resistance to T-20. The purpose of this study was to analyse the possible changes in the gp41 env region potentially associated with resistance to T-20 in clinical samples collected from HIV-positive patients failing therapy with T-20. Four heavily pre-treated patients, with multiple mutations associated with resistance to both protease and reverse transcriptase (RT) inhibitors, who received T-20 were included in the study. Genotypic and phenotypic resistance tests (Virologic, San Francisco, CA, USA) were performed at baseline. The antiretroviral drugs planned to be used along with T-20 were chosen individually on the basis of the results of resistance testing. T-20 (Roche Pharmaceuticals, Nutley, NJ, USA) was administered subcutaneously twice a day at doses of 100 mg every 12 h. One patient (no. 2) discontinued T-20 at month 5 as a result of adverse events. Genetic analyses were performed on HIV RNA extracted from the plasma from each individual. Specimens collected at baseline and at 2–3 month intervals while under T-20 were examined. A gp41 fragment (567 base pairs) that included HR1 and HR2 genomic regions was amplified by RT-nested polymerase chain reaction using E41ext1 5′–GAGAAGAGTGGTGCA GAGAG–3′ and E41ext2 5′–ATTCCTTCGGGCCT GTCGGG–3′ as outer primers, and E41int1 5′–GCAGCAGGAAGCACTATGGGCG–3′ and E41int2 5′–GGTGARTATCCCTGCTAACTC–3′ as inner primers. The amplicon was sequenced in both directions using the dRhodamine Terminator Cycle Sequencing kit (Applied Biosystems, Foster City, CA, USA), and sequences were aligned using Clustall X. Plasma HIV-RNA levels and CD4 cell counts were monitored monthly. All four patients experienced a significant decrease in viral load (> 1 log) within the first month of therapy with T-20 (patient 1: −1.89 logs; patient 2: −1.66 logs; patient 3: −1.17 logs; and patient 4: −1.75 logs). However, between months 2 and 3, the viral loads rebounded up to baseline values in all cases. At the initiation of treatment with T-20, the mean CD4 cell count was 217 ± 86.5 cells/μl, and remained without significant change during the whole study period in all but one individual (no. 4), in whom a net gain of 199 cells/μl was recorded (Fig. 1).Fig. 1.: Evolution of plasma HIV-1 RNA and CD4 T cells in patients on T-20. ––○–– Plasma HIV-1 RNA; – —▪— – CD4 T cells.DNA sequence analyses and the encoded protein sequences of gp41 revealed that all four patients carried a consensus GIV motif at baseline. The intrapatient variability in gp41 sequences during treatment with T-20 was low, ranging from 0 to 0.5%, and resulting almost always in synonymous substitutions. None of the sequences analysed harboured amino acid changes previously associated with resistance to T-20 in vitro, not even after 9 months of virological failure. However, one subject (no. 1) developed a change G←V at position 36 at month 7, although being on T-20. The viral loads and CD4 cell counts during follow-up did not differ in this patient in respect to the others. Another subject (no. 3) presented with three changes (Q110Q/E, E119E/Q and R122R/K) within the HR2 domain after 9 months on T-20. Sequences generated in this study have received the following GenBank accession numbers: AF500084–AF500093. A GenBank database search has so far not recorded the G36V change in the GIV motif of the HR1 region in any of the reported sequences. Moreover, to our knowledge no gp41 sequences from patients receiving T-20 have shown amino acid substitutions in the GIV motif. The recognition of non-conservative amino acid substitutions in the HR2 genomic region may equally be involved in the loss of susceptibility to T-20, affecting the molecular interactions between the heptad repeats [3]. Phenotypic studies are currently underway, using individual clones with genotypic differences found in HR1 and HR2, to characterize these mutations and determine their possible association with resistance to T-20. All four patients showed a rapid and significant decline in plasma viraemia within one month of therapy with T-20. This dramatic drop may not be attributed to other antiretroviral drugs in the regimen taken along with T-20, because large and broad cross-resistance to the agents in use was present (data not shown). Nevertheless, all patients experienced a rebound in their viral load 2–3 months later. All but one subject have continued on T-20 despite virological failure for longer than 9 months. The possible immunological benefit provided by T-20, despite virological failure in these patients, may be of interest. The mechanism involved is unknown. Although a net gain of CD4 T lymphocytes was only recorded in one patient (no. 4), a clear disconnection with the viral load was noticed in the remaining cases. So far, all patients are clinically asymptomatic and have not developed opportunistic infections. Although virus susceptibility to T-20 seems to be greatly dependent on the GIV motif, co-receptor specificity defined by the V3 loop of gp120 modulates it [4]. Recent studies support the finding that gp120 interactions with CCR5 are more vigorous than interactions with CXCR4, suggesting that co-receptor influence on HR1-targeted peptide inhibitors is a general phenomenon [5]. This is in contrast to previous reports [6], which found no differences in T-20 susceptibility in viruses as a function of the distinct co-receptor use for cell entry. Amino acid changes in other regions of the env gene, such as V2V3, might be involved in the loss of susceptibility to T-20. Primary CXCR4 syncytium-inducing (SI) isolates seem to be more sensitive to inhibition by T-20 than are CCR5 non-SI isolates [4]. The characteristics of HIV infection in our four patients (long duration of infection, multiple treatment failures, etc.) might have favoured the presence of CXCR4 SI viruses. Pharmacological pressure with T-20 could have resulted in a shift in virus co-receptor use, favouring the replication of CCR5 over CXCR4 isolates. Accordingly, V2V3 sequences from our patients after several months under T-20 showed critical domains associated with CCR5 co-receptor use [7]. Hypothetically, these changes in cell tropism might have conferred a biological advantage to viruses in the face of T-20 therapy. Besides, this fact might have resulted in an amelioration of the CD4 cell destruction rate in our patients. In conclusion, this study shows that the HR1 and HR2 regions of the HIV-1 genome are highly conserved throughout long-term therapy with T-20, and that other env regions may be involved in the development of T-20 resistance. A longer follow-up of these patients may provide additional insights to give a better understanding of the mechanisms of escape of HIV in response to T-20 inhibition. Acknowledgements Sponsorship: This work was partly funded by grants from the Asociación Investigación y Educación en SIDA (AIES), Comunidad Autónoma de Madrid (CAM), and FIPSE.