B cell depletion in HIV-1 subtype A infected Ugandan adults: relationship to CD4 T cell count, viral load and humoral immune responses

Abstract

To better understand the nature of B cell dysfunctions in subjects infected with HIV-1 subtype A, a rural cohort of 50 treatment-naïve Ugandan patients chronically infected with HIV-1 subtype A was studied, and the relationship between B cell depletion and HIV disease was assessed. B cell absolute counts were found to be significantly lower in HIV-1+ patients, when compared to community matched negative controls (p<0.0001). HIV-1-infected patients displayed variable functional and binding antibody titers that showed no correlation with viral load or CD4+ T cell count. However, B cell absolute counts were found to correlate inversely with neutralizing antibody (NAb) titers against subtype A (p = 0.05) and subtype CRF02_AG (p = 0.02) viruses. A positive correlation was observed between subtype A gp120 binding antibody titers and NAb breadth (p = 0.02) and mean titer against the 10 viruses (p = 0.0002). In addition, HIV-1 subtype A sera showed preferential neutralization of the 5 subtype A or CRF02_AG pseudoviruses, as compared with 5 pseudoviruses from subtypes B, C or D (p<0.001). These data demonstrate that in patients with chronic HIV-1 subtype A infection, significant B cell depletion can be observed, the degree of which does not appear to be associated with a decrease in functional antibodies. These findings also highlight the potential importance of subtype in the specificity of cross-clade neutralization in HIV-1 infection.

Figure 1. The absolute B lymphocyte count in HIV-1 subtype A infected patients is significantly lower than in HIV-1 negative controls.

Samples of whole blood from HIV-1 infected and uninfected individuals were analysed using flow cytometry and the MultiTEST™ IMK Kit (BD). Lymphocyte subpopulations were quantified and samples from 261 HIV-1 negative individuals (white box) were compared with HIV positives. The absolute counts of CD19+ B lymphocytes in 50 HIV-1 subtype A infected patients (grey box) and in 192 patients with all HIV-1 subtypes combined (hatched box) were significantly reduced in comparison to uninfected participants. (p<0.0001, Mann-Whitney test). The whiskers represent the range from the 10th–90th percentile.

Figure 2. Absolute CD4+ T lymphocyte counts correlate with B lymphocyte absolute counts, but not with HIV-1 viral Load.

The absolute counts of B lymphocytes correlated with the absolute number of CD4+ T lymphocytes (A). The number of plasma HIV-1 RNA copies/ml showed a significant inverse correlation with the CD4+ T lymphocyte percentage (C) and a trend towards correlation with CD4 absolute number (B).

Figure 3. B lymphocyte absolute counts in HIV-1 subtype A infected Ugandans correlate inversely with neutralizing antibody titers against different HIV-1 subtypes.

Neutralizing antibody responses in sera of 50 HIV-1 subtype A infected patients were determined against a panel of 10 pseudoviruses in the TZM-bl Assay. The mean ID₅₀ using each serum sample against all 10 pseudoviruses or against all pseudoviruses within a subtype was calculated. B lymphocyte numbers showed a significant inverse correlation with neutralizing antibody titers against the mean ID₅₀ of all 10 pseudoviruses (A) and with the mean ID₅₀ for subtype CRF02_AG (B); the same trend was observed for subtype A viruses (C). B lymphocyte numbers also showed a trend towards inverse correlation with breadth (number of panel viruses neutralized) (D).

Figure 4. Breadth of neutralization in HIV-1 subtype A infected patient sera.

Shown is the breadth of neutralization for each serum sample against a panel of 10 pseudoviruses from five different subtypes (Purple columns = subtype CRF02_AG, yellow = subtype A, green = subtype B, blue = subtype C and orange = subtype D). Red boxes indicate ID₅₀s>10 (positive neutralization). HIV-1 subtype A infected subjects showed more frequent neutralization against subtypes CRF02_AG and A, as compared to non-A subtypes (p<0.001, Fisher's exact test).

Figure 5. Patients infected with HIV-1 subtype A show greater frequency and magnitude of neutralization against subype A and CRF02_AG viruses.

Sera from 50 HIV-1 subtype A-infected patients were tested in the TZM-bl neutralization assay against 1–3 pseudoviruses of each subtype indicated. The frequency of neutralization (% of titers >10) of viruses in a given subtype (black bars), as well as the geometric mean titer (ID₅₀, white bars) against each viral subtype, is shown. For the geometric mean titers, the error bars represent the 95% confidence intervals.

Figure 6. Binding antibodies against HIV-1 gp120 correlate with mean NAb Titer and breadth of neutralization.

Subtype A gp120 binding antibody titers correlate with the mean NAb titers against all 10 pseudoviruses (A) and the mean ID₅₀s of subtype CRF02_AG (B) and subtype A (C) pseudoviruses. A significant correlation was also seen between the gp120 binding antibody titers and neutralization breadth, as measured by the number of viruses neutralized (D).