Fig 1.
Inducible infectious HIV-1 reservoir viruses can downregulate HLA-C.
(A) Viral isolates were generated from ART-suppressed individuals and infected co-cultured Molt-4 cells, where HLA-C downregulation was determined by comparing HLA-C staining between uninfected and infected cells, discriminated by intracellular staining for Gag. (B) Representative staining of Molt-4 cells in cultures infected with viral isolates that preserve or downregulate HLA-C. HLA-C staining is shown for infected (red) and uninfected cells (black) compared to an isotype control stain (grey). (C) HLA-C downregulation observed for 128 isolates from 19 individuals. Statistical analyses are unpaired t tests.
Fig 2.
HLA-C downregulation in untreated HIV-1 infection.
(A) Infectious molecular clones were assessed for ability to downregulate HLA-C for 3 individuals: CH040, 162, 058. Representative staining is shown for primary CD4+ cells infected in vitro, with HLA-C staining compared between infected and uninfected cells within a culture. (B) For the 3 individuals infectious molecular clones of both transmitted/founder viruses and those 6 months after infection were tested. HLA-C staining is shown for infected (red) and uninfected cells (black) compared to an isotype control stain (grey). (C) For ART-naïve individuals multiple cloned Vpu molecules were measured for ability to downregulate HLA-C by transfection of HeLa cells, with transfected cells in a culture identified by co-transfection of a GFP expression plasmid. (D) Staining for HLA-C is shown for Vpu clones with differential downregulation, showing transfected (red) and untransfected cells (black) compared to an isotype control (grey). (E) HLA-C downregulation is shown for 15 Vpu clones from one individual sampled at 3 timepoints over 10 years. (F-H) HLA-C downregulation observed for multiple clones from a single timepoint for 3 individuals. (I,J) HLA-C downregulation observed for dominant Vpu clones from longitudinal timepoints for 2 individuals.
Fig 3.
Downregulation of HLA-C shows adaptation to host HLA genotype.
(A) For 186 individuals a single Vpu clone from chronic untreated infection was measured for ability to downregulate HLA-C by transfection of Molt-4 cells. The distribution of HLA-C downregulation is shown, with a threshold of 6-fold used to define 69 strongly downregulating (black) and 117 weakly downregulating Vpu clones (grey). (B) The proportion of Vpu clones that downregulate HLA-C strongly is shown when individuals are stratified by the presence of common alleles of each of the HLA-A,—B, and—C loci. The number of individuals in each group is shown above each plot and statistical analysis is by chi square tests. (C) HLA-C alleles can be grouped as C1 or C2 according to their differential binding of KIR2D receptors, and the proportion of Vpu clones that strongly downregulate HLA-C is shown for individuals when grouped by these genotypes. (D) HLA-C alleles differ in expression level and can be inferred for an individual from the HLA-C genotype. HLA-C expression level and the observed HLA-C downregulation by Vpu are plotted for 186 individuals. (E) For the subset of individuals where viral adaptation to host HLA-C alleles could be quantified (n = 72), the correlation between host HLA-C allele-specific viral sequence adaptation and observed downregulation of HLA-C by Vpu was determined. For 10 HLA-C alleles this correlation is plotted against the proportion of individuals with that HLA-C allele in which Vpu downregulates HLA-C. Correlations were determined using two-tailed Spearman analyses.
Fig 4.
Variation in the Vpu transmembrane region is responsible for the range of HLA-C downregulation exhibited by primary HIV-1 viruses.
(A) Sequence of representative Vpu clones that weakly (clone 1), or strongly downregulate HLA-C (clones 2–4), are shown aligned to Vpu from NL4-3. (B) A multiple linear regression model using 191 of the primary Vpu clones identifies 5 positions of Vpu at which the residues indicated associate independently with HLA-C downregulation. (C) These positions were each confirmed to influence HLA-C downregulation using single residue mutants of Vpu. HLA-C staining is shown for cells transfected with primary Vpu clones (red), single residue mutants (black), and untransfected cells (grey). (D) HLA-C staining for cells transfected with a Vpu constructed to have all 5 residues associated with HLA-C downregulation (red), or that with no residues associating with HLA-C downregulation (black), is shown compared to untransfected control (light grey). (E) HLA-C downregulation predicted by the model in panel B, compared to that observed for each of the 191 Vpu clones. (F) An NMR structure of Vpu, with N-terminal and transmembrane regions highlighted, shows the location of the 5 Vpu positions at which primary sequence variations associate with HLA-C downregulation (green).
Fig 5.
Transmembrane domain residues of HLA mediate the interaction with Vpu.
(A) 293T cells were transfected with Flag-tagged constructs of Vpu from a primary virus 2_87, Vpu from NL4-3, or GFP as a negative control. Cells were co-transfected with HA-tagged constructs of HLA from the A or C loci, or HA-tagged β-TrCP as a positive control. Flag immunoprecipitation was followed by western blotting with anti-Flag or anti-HA antibody, and blots of the lysate prior to immunoprecipitation are also shown. (B) Chimeras 1–6 were constructed with the regions exchanged between HLA-A and HLA-C shown. (C) Co-immunoprecipitations were repeated using Vpu from 2_87 and the chimeric HLA constructs. (D) Sequences of the transmembrane domain of representative HLA-A and HLA-B alleles, 3 alleles which represent the transmembrane sequences of all common HLA-C alleles, and the chimeric constructs 5 and 6. Complete sequences for all HLA constructs are detailed in S8 Fig.