Rare and new V3 loop variants in HIV-1-positive long-term non-progressors from Rwanda

We identified uncommon amino acid substitutions in the V3 loop regions of HIV-1 strains infecting patients from Rwanda. Their frequency was greater in long-term non-progressors (LTNP) compared with late-stage patients (P= 0.006), particularly in a sequence region that has crucial interactions with the cell surface, and is highly relevant for the host's immune response. These variants might reflect a viral response to a strong immune pressure, or represent attenuated HIV-1 strains infecting LTNP in Rwanda. The timespan from initial infection to the advanced stages of the disease is highly variable in HIV-1-infected patients. Both viral and host factors are thought to interact in a very complex interplay system, which modulates the natural history of HIV-1 infection [1]. The V3 region of the HIV-1 envelope glycoprotein gp120 probably contributes to this variability, carrying determinants for neutralizing antibodies, cytotoxic T lymphocytes and viral tropism [2]. We characterized the V3 loop sequences of HIV-1 strains infecting patients with different HIV-1 disease histories. The study population included 15 long-term non-progressor (LTNP) women and 14 patients with severe HIV-1 disease (late-stage patients; LSP). None of the patients had received any antiretroviral agents. All plasma samples were collected at the Retrovirology Laboratory of Kigali, Rwanda. LTNP were first screened in the antenatal clinic in 1986–1987 [3]. Enrolment in prospective studies followed in 1988 [4]. Of the 15, seven LTNP had been HIV positive since their first test in 1986–1987, the remainder seroconverted thereafter, either between screening and enrolment (N = 3), or between 6 and 42 months after enrolment [5]. The mean age at enrolment in 1988 was 28.6 years (range 19–35). At the time of sampling, the mean number of months of follow-up since the first positive test was 128 (range 83–145); all LTNP had CD4 T cell counts greater than 500 × 106 cells/l (mean 818 × 106 cells/l, range 510–1390 × 106) and were WHO clinical stage I (N = 14) or II [6]. The participants continue to be followed at 3 monthly intervals, and all 15 are actively enrolled and in good health as of March 2002. LSP were hospitalized patients with WHO clinical stage IV. Viral RNA was purified from plasma using Qiagen methodology (Leusden, the Netherlands). V3 loop coding regions for each patient (LTNP1–15, GenBank accession numbers AY123254–AY123268 and LSP1–14, GenBank accession numbers AY123269–AY123282, see Fig. 1) were directly sequenced using a prototype kit provided by Visible Genetics (Toronto, Canada), and the subtype was determined by submitting the sequence via internet to the National Centre for Biotechnology Information USA (http://www.ncbi.nlm.nih.gov/retroviruses/subtype/subtype.html). To evaluate the subtype of LTNP more accurately, reverse transcriptase and protease were also sequenced and subtyped using the same methodology. After alignment, each deduced 35 amino acid sequence situated between the two cysteines of the V3 loop was compared with the corresponding V3 loop amino acid subtype consensus sequences of the Los Alamos HIV Database [7]. Subsequent comparisons between LTNP and LSP concerned residue positions for which a predominant amino acid could be deduced from the direct sequencing of the viral populations infecting each patient.Fig. 1.: Deduced amino-acid sequences of HIV-1 V3 loop from 15 long-term non-progressors (LTNP1–LTNP15, GenBank accession numbers AY123254–AY123268) and 14 late-stage patients (LSP1–LSP14, GenBank accession numbers AY123269–AY123282). The clade A and C amino acid consensus sequences (consA and consC) [6] are shown at the top and the bottom of the figure, respectively. Sequences in italics represent clade C sequences. Boxed and shaded residues indicate unique variants, and boxed residues only indicate very rare variants (less than 0.5% of clade A or C sequences). Residue positions in which a predominant amino acid was not present are described in lower case.In both groups, subtype A was predominant (14/15 in LTNP and 12/14 in LSP). Subtype C was also found (1/15 in LTNP and 2/14 in LSP). These findings are consistent with the normal subtype distribution in Rwanda [8]. All sequences were compatible with a non-syncytium-inducing, CCR5-using phenotype, on the basis of their amino acid content [9]. The frequency of very rare (previously reported for less than 0.5% in the HIV Database) or new mutations compared with the corresponding V3 loop amino acid consensus sequences was higher in LTNP (2.3%) than in LSP (0.6%) (P = 0.053). The number of strains bearing at least one very rare or new mutation was higher in LTNP (46.7%) than in LSP (27.2%); this difference was statistically significant when the sequences between residues 1–28 were compared (46.7% in LTNP versus 0% in LSP, P = 0.006). Among rare mutations, a deletion at position 24 and a His34Ser substitution were highly prevalent (20 and 13%, respectively) in LNTP. Asp25Thr, which was not reported in the HIV Database, was identified twice, in LTNP2 and LTNP11, and was not found in LSP. LTNP7 and LTNP10 associated several rare mutations. In contrast, no more than one rare mutation was identified per LSP. A neighbour-joining tree analysis of reverse transcriptase, protease and V3 loop sequences from all 15 LTNP was generated with F84 model distances and a transition to transversion ratio of 1.3 using PHYLIP. No clustering was identified, suggesting the absence of an epidemiological link between the patients (data not shown). In conclusion, our results show a high frequency of rare or new V3 loop variants in a cohort of LTNP, compared with LSP. Whether these rare mutations play a role in, or are subsequent to, the HIV pathogenesis of LTNP remains to be determined. However, these mutations were significantly more prevalent in the first 28 residues of the V3 loop, which include most of the known cytotoxic T lymphocytes, antibodies, and helper T cell epitopes of this region [10], possibly suggesting a specific viral response to a strong immune pressure in LTNP. Alternatively, these variants might represent specific patterns of attenuated HIV-1 strains infecting LTNP in Rwanda. Interestingly, Asp25Thr, which has not been described previously in subtype A, was represented twice in LTNP, at a critical residue position for co-receptor usage [9], and might account for impaired virus entry functions. François Romana Etienne Karitab Alexia Monnetc Christine Lamberta Elodie Fontainea Susan Allend François Schneidera Robert Hemmera Jean-Claude Schmita Vic Arendta

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.