Determinants of B-Cell Compartment Hyperactivation in European Adolescents Living With Perinatally Acquired HIV-1 After Over 10 Years of Suppressive Therapy

Abstract

Background: Despite a successful antiretroviral therapy (ART), adolescents living with perinatally acquired HIV (PHIV) experience signs of B-cell hyperactivation with expansion of 'namely' atypical B-cell phenotypes, including double negative (CD27-IgD-) and termed age associated (ABCs) B-cells (T-bet+CD11c+), which may result in reduced cell functionality, including loss of vaccine-induced immunological memory and higher risk of developing B-cells associated tumors. In this context, perinatally HIV infected children (PHIV) deserve particular attention, given their life-long exposure to chronic immune activation.

Methods: We studied 40 PHIV who started treatment by the 2^nd year of life and maintained virological suppression for 13.5 years, with 5/40 patients experiencing transient elevation of the HIV-1 load in the plasma (Spike). We applied a multi-disciplinary approach including immunological B and T cell phenotype, plasma proteomics analysis, and serum level of anti-measles antibodies as functional correlates of vaccine-induced immunity.

Results: Phenotypic signs of B cell hyperactivation were elevated in subjects starting ART later (%DN T-bet+CD11c+ p=0.03; %AM T-bet+CD11c+ p=0.02) and were associated with detectable cell-associated HIV-1 RNA (%AM T-bet+CD11c+ p=0.0003) and transient elevation of the plasma viral load (spike). Furthermore, B-cell hyperactivation appeared to be present in individuals with higher frequency of exhausted T-cells, in particular: %CD4 TIGIT+ were associated with %DN (p=0.008), %DN T-bet+CD11c+ (p=0.0002) and %AM T-bet+CD11c+ (p=0.002) and %CD4 PD-1 were associated with %DN (p=0.048), %DN T-bet+CD11c+ (p=0.039) and %AM T-bet+CD11c+ (p=0.006). The proteomic analysis revealed that subjects with expansion of these atypical B-cells and exhausted T-cells had enrichment of proteins involved in immune inflammation and complement activation pathways. Furthermore, we observed that higher levels of ABCs were associated a reduced capacity to maintain vaccine-induced antibody immunity against measles (%B-cells CD19+CD10- T-bet+, p=0.035).

Conclusion: We identified that the levels of hyperactivated B cell subsets were strongly affected by time of ART start and associated with clinical, viral, cellular and plasma soluble markers. Furthermore, the expansion of ABCs also had a direct impact on the capacity to develop antibodies response following routine vaccination.

Keywords: B-cell hyperactivation; CD11c; T-bet; caHIV-1 RNA; exhausted T-cells; late ART; perinatal HIV/AIDS; proteomic profiling immune activation.

Figure 1

Time of ART initiation and cell associated HIV-1 RNA (caRNA) are associated with phenotypic signs of B cell hyperactivation. Gating strategy is shown in (A); in (B) showed levels of DN in adolescents with perinatally aquired HIV-1 (HIV) and healthy controls (HC); (C, D) correlations between ABCs cells and age at ART start are shown; (E) correlation plot between viral correlates of recent replication and ABCs / exhausted T-cells are shown; differential analysis between levels of ABCs cells and caRNA or SPIKE being detected vs non-detected is shown in (F, G). p values are calculated using Mann Whitney test in (B, F, G). Spearman p values are shown in (B–D). Significance was set at p>0.05. DN, double negative; AM, activated memory; SW, Switched B-cells (memory B-cells); UNSW, Unswitched B-cells; MFI, mean fluorescent intensity.

Figure 2

Levels of exhausted T-cells are positively associated with hyperactivated B-cells. In (A) a cartoon showing the main findings of the figures are pictured. In (B) Heatmap plot showing Spearman correlations between exhausted T-cells and hyperactivated B-cells. Only significant correlations are shown with red indicating positive correlations and Blue the negative ones. The colored scale going between 1 and -1 indicates the rho values. DN, double negative; AM, activated memory. Significance was set at p<0.05.

Figure 3

Association between proteomic profiling and levels of hyperactivated B-cells and exhausted T-cells. (A) Heatmap plot showing Spearman correlations between the 13 unfunctional features values and the abondance of the 73 plasma proteins belong to the two clusters identified in correlation matrix with all 338 proteins. Red indicates positive correlations and Blue negative ones. Bubble plots showing the top 10 Reactome pathways (B) and GO Biological Process (C) significantly enriched (Adjusted p-value < 0.05) in proteins positively (Pos) and negatively (Neg) correlated with the 13 unfunctional features. The proteins were separated into positively and negatively correlated based on the two clusters showed in the correlation heatmap in panel (A) Colors are related at the log₁₀ adjusted p-value values and the circle diameter are related at the number of proteins for each term. Significance was set at p<0.05.

Figure 4

Association between ABCs and anti-measles humoral response. (A) Spearman correlation between CD19+CD10- B-cells T-bet+ and anti-Measle plasma IgG titers, with rho and p defining the statistical significance. (B, C) Spearman correlation between anti-Measle plasma IgG titers and Age at ART in m and years from measles vaccination, respectively, with rho and p defining the statistical significance. Color dots show the distribution of CD19+CD10- B-cells T-bet+. Significance was set at p<0.05.