Impact of HLA class I functional divergence on HIV control

Abstract

Heterozygosity of Human leukocyte antigen (HLA) class I genes is linked to beneficial outcomes after HIV infection, presumably through greater breadth of HIV epitope presentation and cytotoxic T cell response. Distinct allotype pairs, however, differ in the extent to which they bind shared sets of peptides. We developed a functional divergence metric that measures pairwise complementarity of allotype-associated peptide binding profiles. Greater functional divergence for pairs of HLA-A and/or HLA-B allotypes was associated with slower AIDS progression and independently with enhanced viral load control. The metric predicts immune breadth at the peptide level rather than gene level and redefines HLA heterozygosity as a continuum differentially affecting disease outcome. Functional divergence may affect response to additional infections, vaccination, immunotherapy, and other diseases where HLA heterozygote advantage occurs.

Fig. 1.. Effect of genetic homozygosity on progression to CD4<200 after HIV infection.

Progression to CD4<200 among (A) individuals homozygous at any HLA-A, -B, and/or -C locus/loci (red curve) compared with heterozygotes at all three loci (blue curve). (B) HLA-A homozygotes (red) compared to HLA-A heterozygotes (blue). (C) HLA-B homozygotes (red) compared to HLA-B heterozygotes (blue). (D) HLA-C homozygotes (red) compared to HLA-C heterozygotes (blue). All analyses were corrected for HLA-B*57, B*27, B*35Px, and racial background.

Fig. 2.. Submotif overlap and its effect on progression to CD4<200 after HIV infection.

(A) Representative motif logogram and nonmetric ordination plot as generated for each of 95 HLA allotypes studied previously (10). Submotifs corresponding to each cluster were determined by retaining only the amino acids at each position with an entropy lower then 0.1 (shown for B*44:03). (B) Determination of submotif overlap corresponding to the two HLA allotypes encoded by each genotype. Red indicates identical, whereas black indicates distinct submotifs across the two allotypes. (C) Determination of the range in functional divergence from smallest to greatest at each HLA locus across individuals. Comparisons were made between individuals falling into the top-ranked vs. bottom-ranked quantiles (shown in octiles). (D, E, F) Progression to CD4<200 was compared between individuals in the bottom octile (red curves) vs. top octile (blue curves) of functional divergence at each HLA locus (D) among all individuals, (E) after removing genetic homozygotes corresponding to each individual locus, but keeping the same octile definition as in (D) (diminishing the number of individuals in the bottom octile), or (F) after removing genetic homozygotes at HLA-A and/or HLA-B prior to selection of the bottom and top octiles. Analyses were corrected for HLA-B*57, B*27, B*35Px, and racial background.

Fig. 3.. Combined effect of HLA-A and HLA-B submotif overlap on progression to CD4<200.

(A) Progression to CD4<200 among individuals in the bottom (red curve) vs. top (blue curve) octile(s) of functional divergence at HLA-A and/or HLA-B. For individuals in whom only one of the two (HLA-A or HLA-B) was in the top octile, the other was necessarily in the intermediate range. (B) Individuals carrying intermediate levels of functional divergence at both HLA-A and HLA-B (green curve) were added to the analysis shown in (A). P = test for trend. (C) Individuals were further stratified as follows: bottom octiles at both HLA-A and HLA-B (pink curve); bottom octile at HLA-A or HLA-B and intermediate at the other (red curve); intermediate at both (green curve); bottom octile for only one and intermediate at the other (dark blue curve); top octiles at both (light blue curve). P = test for trend. The HR represents an average of the effect with each decreasing category. The analyses were performed among all individuals (A-C) or following removal of homozygotes at HLA-A and HLA-B (i.e. heterozygous at both HLA-A and HLA-B; D-F). Analyses were corrected for HLA-B*57, B*27, B*35Px, and racial background.