HIV-Exposed Uninfected Infants Show Robust Memory B-Cell Responses in Spite of a Delayed Accumulation of Memory B Cells: an Observational Study in the First 2 Years of Life

Abstract

Improved HIV care has led to an increase in the number of HIV-exposed uninfected (HEU) infants born to HIV-infected women. Although they are uninfected, these infants experience increased morbidity and mortality. One explanation may be that their developing immune system is altered by HIV exposure, predisposing them to increased postnatal infections. We explored the impact of HIV exposure on the B-cell compartment by determining the B-cell subset distribution, the frequency of common vaccine antigen-specific memory B cells (MBCs), and the levels of antibodies to the respective antigens in HEU and HIV-unexposed uninfected (HUU) infants born to uninfected mothers, using flow cytometry, a B-cell enzyme-linked immunosorbent spot assay, and an enzyme-linked immunosorbent assay, respectively, during the first 2 years of life. For the majority of the B-cell subsets, there were no differences between HEU and HUU infants. However, HIV exposure was associated with a lower proportion of B cells in general and MBCs in particular, largely due to a lower proportion of unswitched memory B cells. This reduction was maintained even after correcting for age. These phenotypic differences in the MBC compartment did not affect the ability of HEU infants to generate recall responses to previously encountered antigens or reduce the antigen-specific antibody levels at 18 months of life. Although HIV exposure was associated with a transient reduction in the proportion of MBCs, we found that the ability of HEU infants to mount robust MBC and serological responses was unaffected.

FIG 1

Distribution of memory B cells (CD19⁺ CD10⁻ CD27⁺) (a), unswitched memory B cells (CD19⁺ CD10⁻ CD27⁺ IgM⁺) (b), and switched memory B cells (CD19⁺ CD10⁻ CD27⁺ IgM⁻) (c) in HUU infants (red dots) and HEU infants (black dots). Straight lines show the best-fit prediction of the increment in subset proportions over the 2 years of life. The percentage of memory B cell subsets is presented on the y axis as the natural log of the MBC percentages.

FIG 2

Levels of IgG antibodies to total IgG (a) and selected vaccine antigens, TT protein (b), diphtheria toxin (c), Hib (d), PCPs (e), measles virus antigen (f), and RSV (g), at 18 months of life. Antibody concentrations were compared between HUU infants (open circles) and HEU infants (closed circles). The Wilcoxon rank sum test was used, and medians are presented. P values of <0.05 were considered significant. Arrows in panels d and f, cutoff for protective antibody concentration. Since antibody concentrations were not normally distributed, natural log-transformed values of arbitrary antibody concentrations (TT, diphtheria toxin, PCPs, total IgG) and absolute concentrations (measles virus antigen [in mIU/ml], Hib [in mg/liter], and RSV [in U/ml]) are presented.

FIG 3

Correlation of circulating antigen-specific memory B cells with the percentage of switched memory B cells (first row) or levels of IgG antibodies (second row) against measles virus antigen, TT protein, PCPs, diphtheria toxin, and total IgG. Spearman correlation coefficients were determined, and the Spearman rho values are presented. P values of <0.05 were considered significant.