HIV-dependent depletion of influenza-specific memory B cells impacts B cell responsiveness to seasonal influenza immunisation

Abstract

Infection with HIV drives significant alterations in B cell phenotype and function that can markedly influence antibody responses to immunisation. Anti-retroviral therapy (ART) can partially reverse many aspects of B cell dysregulation, however complete normalisation of vaccine responsiveness is not always observed. Here we examine the effects of underlying HIV infection upon humoral immunity to seasonal influenza vaccines. Serological and memory B cell responses were assessed in 26 HIV+ subjects receiving ART and 30 healthy controls immunised with the 2015 Southern Hemisphere trivalent inactivated influenza vaccine (IIV3). Frequencies and phenotypes of influenza hemagglutinin (HA)-specific B cells were assessed by flow cytometry using recombinant HA probes. Serum antibody was measured using hemagglutination inhibition assays. Serological responses to IIV3 were comparable between HIV+ and HIV- subjects. Likewise, the activation and expansion of memory B cell populations specific for vaccine-component influenza strains was observed in both cohorts, however peak frequencies were diminished in HIV+ subjects compared to uninfected controls. Lower circulating frequencies of memory B cells recognising vaccine-component and historical influenza strains were observed in HIV+ subjects at baseline, that were generally restored to levels comparable with HIV- controls post-vaccination. HIV infection is therefore associated with depletion of selected HA-specific memory B cell pools.

Figure 1. Baseline characteristics of B cells within PBMCs from HIV+ and HIV− subjects.

(A) The frequencies of CD19+ lymphocytes and proportion of class-switched IgD- B cells was assessed by flow cytometry in HIV− (N = 30) and HIV+ subjects (N = 26). (B) Distribution of IgG, IgA and IgM expression memory B cells in both cohorts. (C) Distribution of CD19+ B cell subsets divided by CD27 and CD21 staining: CD27− CD21+ naïve, CD27+ CD21+ resting memory, CD27+ CD21− activated memory and CD27− CD21− tissue-like populations.

Figure 2. Memory B cell responses to H3N2 after IIV3 immunisation.

(A) Cryopreserved PBMC samples were stained with a panel of monoclonal antibodies allowing identification of live CD19+IgD− B cells that did not bind an SA-BB515 mock probe control. Subsequent binding to A/Switzerland/9715293/2013 (SW13) and A/Hong Kong/1/1968 (HK68) probes allows the identification of three distinct populations of IgD− memory B cells; SW13+, HK68+ and a cross-reactive SW13+HK68+ population. Shown for a representative HIV− individual. (B) Frequencies of SW13+ memory B cells prior to and following IIV3 vaccination in HIV+ (N = 26) and HIV− subjects (N = 30). (C) Degree of SW13+ memory B cell expansion (%) in HIV+ and HIV− subjects segregated by baseline seropositivity (HI ≥ 40) to a SW13-like virus as measured by HI assay. (D) Correlation between SW13+ memory B cell frequencies at baseline and post-IIV3. (E) HK68+ and (F) cross-reactive SW13+HK68+ memory B cell frequencies before and following IIV3. Data is shown as median+/−interquartile range. Populations were compared using Mann-Whitney U tests and p values denoted. Correlations were assessed using Spearman’s rank-order test and r and p values are denoted. ns; not significant.

Figure 3. Phenotype of SW13+ memory B cells.

(A) Staining of surface markers CD21 and CD27 is shown for back-gated SW13+ memory B cells relative to the parental CD19+ population before or after vaccination. CD27− CD21+ naïve (N), CD27+ CD21+ resting memory (RM), CD27+ CD21− activated memory (AM) and CD27− CD21− tissue-like populations (TL) are denoted. (B) The proportion of SW13+ memory B cells with an activated memory (CD27+CD21−) phenotype before and after IIV3 immunisation in HIV+ (N = 26) and HIV− subjects (N = 30). (C) Proportions of IgG, IgA and IgM immunoglobulin subclass expression upon SW13+ memory B cells before and after immunisation.

Figure 4

(A) Memory B cell responses to H1N1. PBMC samples were co-stained as before with HA probes for H1N1 strains A/California/4/2009 (CA09) and A/New Caledonia/20/1999 (NC99). (B) Frequencies CA09+ memory B cells before and following IIV3 vaccination in HIV+ (N = 10) and HIV− subjects (N = 10). (C) Degree of CA09+ memory B cell expansion in HIV+ and HIV− subjects segregated by baseline HI activity seropositivity (HI ≥ 40) to CA09. (D) NC99+ and (E) cross-reactive CA09+NC99+ memory B cell frequencies before and following IIV3. Data is shown as median +/− interquartile range. Populations were compared using Mann-Whitney U tests and p values denoted.