Identification of a cluster of HIV-1 controllers infected with low replicating viruses

Abstract

Long term non-progressor patients (LTNPs) are characterized by the natural control of HIV-1 infection. This control is related to host genetic, immunological and virological factors. In this work, phylogenetic analysis of the proviral nucleotide sequences in env gene from a Spanish HIV-1 LTNPs cohort identified a cluster of 6 HIV-1 controllers infected with closely-related viruses. The patients of the cluster showed common clinical and epidemiological features: drug user practices, infection in the same city (Madrid, Spain) and at the same time (late 70's-early 80's). All cluster patients displayed distinct host alleles associated with HIV control. Analysis of the virus envelope nucleotide sequences showed ancestral characteristic, lack of evolution and presence of rare amino-acids. Biological characterization of recombinant viruses with the envelope proteins from the cluster viruses showed very low replicative capacity in TZMbl and U87-CD4/CCR5 cells. The lack of clinical progression in the viral cluster patients with distinct combinations of protective host genotypes, but infected by low replicating viruses, indicate the important role of the virus in the non-progressor phenotype in these patients.

Figure 1. Phylogenetic tree from the Bayesian MCMC (MrBayes) analysis.

The 50% majority rule consensus was constructed and posterior probabilities are indicated by asterisks in nodes (black * upper 0.85 and red * upper 0.95). MRCA including the vast of majority sequences analyzed (black ♦) and MRCA from the cluster viruses (red ♦) are marked. Branch lengths represent the mean value observed for that branch among the post-burning sampled trees. The branch colors identified nucleotide sequences origin: black correspond to ancient sequences from North-America (before 1991) and from Europe (before 1995), blue are Spanish sequences (from 1989 and 2005) and red Spanish cluster sequences (from 2004–2005, except As7 which was from 1989). Green sequences are from elite suppressors and yellow sequences from elite controller patients . Gray lines are D subtype sequences used as outgroup in the bottom of the tree.

Figure 2. Correlation between sampling time and the genetic distance with reference and viral cluster strains.

MRCA-to-tip distances were extracted from MrBayes phylogenetic tree using TreeStat v.1.2. •, values obtained for the nucleotide sequences collected at the beginning of HIV-1 epidemic (years 1981–1995) were plotted against sampling time and a linear regression analysis was performed. Red Δ, values obtained from the cluster nucleotide sequences. The red dashed line and the arrow permit the extrapolation of the year from cluster nucleotide sequences.

Figure 3. Envelope analysis of the cluster viruses.

A) Envelope length in amino acids and N-linked glycosylation sites (NxT/S). B) V1 to V5 loop and signal peptide (SP) lengths in the amino acid sequence sets analysed. Red cluster nucleotide sequences, blue Spanish nucleotide sequences, black ancient nucleotide sequences and green subtype B nucleotide sequences. P values for comparison between 2 groups shown with horizontal black bars were calculated using a 2-tailed Mann-Whitney test.

Figure 4. Biological characterization of the recombinant viruses.

A) Infectivity of env recombinant viruses in TZM-bl cells. Cells were infected with 15 units of HIV-1 p24 antigen (75 pg) of virus-supernatants. Luciferase activity was measured 48 hours post-infection and the results were normalized to the value obtained with the WT virus (89ES061). Results represented the median and SE of two independent assays with thee replicates. B) Replication kinetics of env recombinant viruses in U87-CD4/CCR5 cells. Cells were infected with 100 units of HIV- 1 p24 antigen (500 pg) of virus supernatants. Cultures were followed during 14 days, and HIV-1 production was quantified by the RT-activity in the supernatant with in-house Syber green I based real-time PCR enhanced RT assay (SGPERT). Cluster’s recombinant viruses (red) were compared with recombinant viruses from chronic progressor patients (green), with a recombinant virus obtained from laboratory strain SF-162 (black), and with the laboratory infectious clone 89ES061 (black) where the nucleotide sequences were cloned.

Figure 5. Comparison of the env gene amino acid sequences derived from cluster viruses with subtype B consensus sequence.

35 common mutated positions detected in at least 5 of the cluster viruses are shown in color amino acids. Boxes marked the unusual amino acid whose presence in the cluster is statistically significant when compared with the reference amino acid sequence sets used in the study (see Table 3).

Figure 6. C2-V5 nucleotide sequences from patient LTNP_1.

LTNP_1 samples were obtained from PBMC DNA (2005 sample), cell-free supernatant of PBMC co-culture from a 2004 sample and plasma RNA corresponding to a 2010 sample. Shaded boxes indicated amino acid mutations characteristics of the cluster viruses in this C2-V5 region. V3, V4 and V5 regions are indicated by boxes.

Figure 7. Analysis of HLA-B*57/58 and HLA-B*27 epitopes in gag nucleotide sequence from the cluster viruses.

Amino acid sequences alignment in gag gene from the cluster viruses. The most important HLA-B*57/58 (blue) and HLA-B*27 (green) epitopes described in the literature are marked. Bars indicated the start position of the different proteins in HIV-1 gag gene. X specified mixed position in the nucleotide sequences.