Alterations in B Cell Compartment Correlate with Poor Neutralization Response and Disease Progression in HIV-1 Infected Children

Abstract

Several B cell defects are reported in HIV-1 infected individuals including variation in B cell subsets, polyclonal B cell activation and exhaustion, with broadly neutralizing antibodies elicited in less than 10-20% of the infected population. HIV-1 disease progression is faster in children than adults. B Lymphocyte Stimulator (BLyS), expressed on dendritic cells (DCs), is a key regulator of B cell homeostasis. Understanding how DCs influence B cell phenotype and functionality (viral neutralization), thereby HIV-1 disease outcome in infected children, is important to develop interventional strategies for restoration of B cell function. In this study, a total of 38 vertically transmitted HIV-1 infected antiretroviral therapy (ART) naïve children and 25 seronegative controls were recruited. Based on the CD4 counts and years post-infection, infected children were categorized as long-term non-progressors (LTNPs) (n = 20) and progressors (n = 18). Eight of these progressors were followed up at 6-12 months post-ART. Percentages (%) of DCs, B cell subsets, and expression of BLyS on DCs were analyzed by flow-cytometry. Plasma levels of B cell growth factors were measured by ELISA and viral neutralization activity was determined using TZM-bl assay. Lower (%) of myeloid DCs (mDCs), plasmacytoid DCs, and high expression of BLyS on mDCs were observed in HIV-1 infected progressors than seronegative controls. Progressors showed lower % of naive B cells, resting memory B cells and higher % of mature activated, tissue-like memory B cells as compared to seronegative controls. Higher plasma levels of IL-4, IL-6, IL-10, and IgA were observed in progressors vs. seronegative controls. Plasma levels of IgG were high in progressors and in LTNPs than seronegative controls, suggesting persistence of hypergammaglobulinemia at all stages of disease. High plasma levels of BLyS in progressors positively correlated with poor viral neutralizing activity. Interestingly on follow up, treatment naïve progressors, post-ART showed increase in resting memory B cells along with reduction in plasma BLyS levels that correlated with improvement in viral neutralization. This is the first study to demonstrate that reduction in plasma BLyS levels correlates with restoration of B cell function, in terms of viral neutralization in HIV-1-infected children.

Keywords: B cells; B lymphocyte stimulator; HIV-1; dendritic cells; long-term non-progressors; neutralizing activity; progressors.

Figure 1

Flow-cytometric analysis of dendritic cell (DC) subsets in whole blood and surface expression of B lymphocyte stimulator (BLyS) on myeloid DCs (mDCs) and plasmacytoid DCs (pDCs). Representative gating strategy for DC subsets. Cells were gated on whole blood, mDCs were defined as Lin1 (CD3, CD14, CD16, CD19, CD20, and CD56)⁻ HLADR⁺ CD11c⁺, pDCs were defined as Lin1⁻ HLADR⁺ CD123⁺. Percentages of mDCs and pDCs are expressed as % of whole blood (leukocytes), which includes lymphocytes, monocytes, and granulocytes.

Figure 2

Monitoring of percentage of mDCs, pDCs, and B lymphocyte stimulator (BLyS) expression on surface of mDCs and pDCs in HIV-1 infected children at different stages of disease. (A–D) show comparison of % myeloid DCs (mDCs), % pDCs, MFI BLyS on mDCs, MFI BLyS on pDCs among 25 HIV-1 negative seronegative controls, 20 HIV-1-infected ART naive long-term non-progressors (LTNPs), 18 progressors pre-ART and 8 progressors post 6–12 months of ART, respectively. The p-values are calculated using Mann–Whitney U test for seronegative controls, LTNPs, progressors pre-ART. Wilcoxon signed rank test was used for paired analysis of progressors pre-ART and post 6–12 months of ART. The error bars show the median with the interquartile range. *p ≤ 0.05, ****p ≤ 0.0001.

Figure 3

Flow-cytometric analysis of B cell subsets (gated on CD19⁺ B cells) in HIV-1-infected children at different stages of disease. Representative gating strategy for naïve B cells (CD19⁺ CD27⁻ CD10⁻ CD21⁺), tissue-like memory (TLM) B cells (CD19⁺ CD27⁻ CD10⁻ CD21⁻), transitional immature (TI) B cells (CD19⁺ CD27⁻ CD10⁺ CD21⁻) resting memory (RM) B cells (CD19⁺ CD27⁺ CD10⁻ CD21⁺) and mature activated (MA) B cells as (CD19⁺ CD27⁺ CD10⁻ CD21⁻). Lower left panel shows naïve, TI and TLM B cells gated on CD19⁺ CD27⁻ cells, lower right panel shows mature activated and resting memory B cells gated on CD19⁺ CD27⁺ cells. Percentages of B cell subsets are expressed as % of CD19⁺ B cells.

Figure 4

Monitoring of percentage of B cell subsets in HIV-1 infected children at different stages of disease. (A–D) show comparison of naïve B cells, tissue-like memory B cells, resting memory B cells, mature activated B cells among 25 HIV-1 negative seronegative controls, 20 HIV-1-infected ART naive long-term non-progressors (LTNPs), 18 progressors pre-ART and 8 progressors post 6–12 months of ART, respectively. The p-values are calculated using Mann–Whitney U test for seronegative controls, LTNPs, progressors pre-ART, Wilcoxon signed rank test was used for paired analysis of progressors pre-ART and post 6–12 months ART. The error bars show the median with the interquartile range. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.

Figure 5

Plasma concentration of B-cell growth factors and immunoglobulins in HIV-1 infected children at different disease stages of disease. (A–F) show levels of B lymphocyte stimulator (BLyS), IL-6, IL-4, IL-10, IgA, and IgG among 25 HIV-1 seronegative controls, 20 HIV-1-infected ART naive long-term non-progressors (LTNPs), 18 progressors pre-ART and 8 progressors post 6–12 months of ART. The p-values are based on Mann–Whitney U test among seronegative donors, LTNPs, progressors pre-ART. Wilcoxon signed rank test was used for paired analysis of progressors pre-ART and post 6–12 months of ART. The error bars indicate median values with interquartile range. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.

Figure 6

Heat-map analysis of ID₅₀ titers of plasma samples from HIV-1-infected children based on hierarchical clustering and transformed natural log data of ID₅₀ values. (A) shows comparison of ID₅₀ titers of 10 ART naive long-term non-progressors (LTNPs) and 18 progressors, and (B) shows comparison of ID₅₀ titers of 8 progressors pre-ART and post 6–12 months of ART against a panel of 11 virus strains. In the heat-map, each row shows the natural log ID₅₀ values for each plasma sample, while the columns display the viruses. Darker color represents better neutralizing activity. Alpha numerals given after the underscore along with the viruses indicate the tier of the virus.

Figure 7

Evaluation of neutralization potency and % of viruses neutralized. (A) shows the comparison of geometric mean titers (GMTs), and (B) shows the comparison of % of viruses neutralized with ID₅₀ ≥ 300 achieved amongst 10 ART naïve long-term non-progressors (LTNPs), 18 progressors pre-ART, and 8 progressors post 6–12 months of ART against a panel of 11 pseudoviruses. p-Values are based on Mann–Whitney U test among seronegative donors, LTNPs, progressors pre-ART, Wilcoxon signed rank test was used for paired analysis of progressors pre-ART and post 6–12 months ART. The error bars show the median with the interquartile range. *p ≤ 0.05, **p ≤ 0.01.

Figure 8

Association of HIV-1-specific neutralization response with B lymphocyte stimulator (BLyS). (A) shows correlation of plasma levels of BLyS with geometric mean titers (GMTs), and (B) shows correlation of plasma BLyS with % of viruses neutralized with ID₅₀ ≥ 300 against a panel of 11 pesudoviruses of different clades and tiers. Correlation coefficient (r) was calculated using spearman rank correlation, for graphical representation x-axis, y-axis was converted to log scales.