Fas ligand-expressing B-1a lymphocytes mediate CD4(+)-T-cell apoptosis during schistosomal infection: induction by interleukin 4 (IL-4) and IL-10

Abstract

A previous study of the murine model of Schistosoma mansoni infection has implicated splenic CD19(+) B lymphocytes as Fas ligand (FasL)-bearing mediators of CD4(+) T-lymphocyte apoptosis. The present study shows that B-cell deficiency leads to decreased CD4(+) T-cell apoptosis during infection and compares FasL expression and killer function of B-1a- and CD5(-) B-lymphocyte subsets. B-1a cells from uninfected mice displayed constitutive expression of FasL compared with that of CD5(-) B cells. FasL expression was enhanced following worm egg deposition and antigenic stimulation on both subsets of B cells. Purified B-1a cells from uninfected mice were potent effectors of CD4(+) T-cell apoptosis, and the killing effect was enhanced during schistosome infection. FasL expression by splenic B cells required CD4(+)-T-cell help that was replaced by addition of culture supernatants from antigen-stimulated splenocytes of infected mice. The culture-supernatant-stimulated FasL expression was inhibited by anti-interleukin 10 (IL-10) and anti-IL-4 antibodies. Culture of purified B cells with recombinant IL-4 (rIL-4), rIL-10, and soluble egg antigens (SEA) led to increased expression of FasL on B-1a cells. These results suggest that FasL-expressing, splenic B-1a cells are important mediators of SEA-stimulated CD4(+)-T-cell apoptosis and that maximal FasL expression on B-1a cells is dependent on antigenic stimulation and the presence of IL-4 and IL-10.

FIG. 1.

CD4⁺-T-cell apoptosis and granuloma downmodulation is dependent on B cells. B-cell-deficient μMT (B def) and age-matched, wild-type C57BL/6J (WT) mice were infected with S. mansoni cercariae. Mice (n = 4 for each data point) were sacrificed at the indicated times postinfection, splenocytes and granuloma cells were isolated individually from each mouse, and portions of the livers were formalin fixed and prepared for hematoxylin and eosin staining. (A) Freshly isolated splenocytes and granuloma cells were analyzed for CD4⁺-T-cell apoptosis by annexin V-based three-color flow cytometry. Percentage values were normalized by arcsine transformation prior to determination of the means and standard deviations of quadruplicate samples. (B) Granuloma sizes were determined from stained liver sections of mice infected for 14 weeks by computerized morphometry. Only granulomas with a single, well-circumscribed egg were included in the analysis. Error bars indicate the standard errors of the means of 24 granulomas/mouse for 4 mice/group.

FIG. 2.

Analysis of FasL expression on B-1a- and CD5⁻-B-cell subsets by three-color flow cytometry. Freshly isolated splenocytes from uninfected and schistosome-infected mice were paraformaldehyde fixed and then stained with biotinylated anti-FasL, streptavidin-APC, anti-CD19-FITC, and anti-CD5-PE. Live cells were gated from plots of forward scatter versus side scatter and then were plotted for CD19-FITC versus CD5-PE (A). B-1a (CD5⁺/CD19⁺) and CD5⁻ B (CD5⁻/CD19⁺) cells were gated and FasL expression was compared on the subsets. Representative histograms for uninfected mice (B) and mice infected for 8 weeks (C) are shown. Because the staining appeared to be constitutive on B-1a cells, MFI was chosen for presentation of data instead of percentages of positive cells.

FIG. 3.

Surface FasL expression is increased on B-1a and CD5⁻ B cells during schistosome infection. Freshly isolated splenocytes from mice infected for the indicated number of weeks were fixed and stained as described in the legend to Fig. 2. FasL expression was analyzed on gated B-1a and CD5⁻ B cells. The ratios of MFI ± standard deviations of triplicate samples were determined in comparison to CD5⁻ B cells from uninfected mice (∗) and are plotted from a representative of three experiments.

FIG. 4.

Purified B-1a cells from uninfected and schistosome-infected mice mediate apoptosis of purified CD4⁺ T cells. Splenocytes from mice infected for 8 weeks were cultured in media for 24 h and then were enriched for CD4⁺ T cells by magnetic bead depletion of CD19⁺ B cells and CD8⁺ T cells. CD4⁺-T-cell targets were then cultured without added cells (black bar) or mixed at the indicated ratios with B-cell subsets purified from freshly isolated splenocytes of uninfected mice and mice infected for 8 weeks. After 24 h of coculture, the cells were stained with anti-CD4-PE, annexin V-FITC, and PI and were analyzed by three-color flow cytometry. The viable CD4⁺ T cells (CD4⁺/PI⁻) were gated and analyzed for binding of the early apoptosis marker, annexin V. The percentages were normalized by arcsine transformation prior to determination of means and standard deviations. Reverse transformed means ± standard deviations of quadruplicate samples from a representative of three experiments are presented.

FIG. 5.

B-1a- and CD5⁻-B-cell FasL expression is upregulated in vitro by treatment with SEA. Splenocytes from mice infected for the indicated number of weeks were cultured for 36 h in the absence or presence of 10 μg of SEA/ml. Cells were fixed, stained, and analyzed as described in the legend to Fig. 2. Ratios of MFI ± standard deviations of triplicate samples were determined against unstimulated CD5⁻ B cells from uninfected mice (∗). Data from a representative of three experiments are presented.

FIG. 6.

FasL expression on CD19⁺ B cells is dependent on soluble factors from CD4⁺ T cells. (A) Splenocytes from mice infected for 8 weeks were left unseparated or were depleted of CD4⁺ T cells prior to 36 h of culture in the presence or absence of 10 μg of SEA/ml. (B) Purified B cells were cultured for 20 h in the absence (black bar) or presence of 24-h culture supernatants from SEA-stimulated or unstimulated splenocytes of mice infected for 6 or 8 weeks. FasL expression was determined on paraformaldehyde-fixed cells by two-color flow cytometry using anti-CD19-FITC, anti-FasL-PE, and an isotype-matched, control IgG2b,κ-PE antibody. Ratios of MFI ± standard deviations are plotted for triplicate samples compared to data for indicated controls (∗). Data are from a representative of three experiments.

FIG. 7.

Supernatant-induced, B-cell FasL expression is inhibited by anti-IL-4 and anti-IL-10 antibodies. Culture supernatants from SEA-stimulated splenocytes of mice infected for 8 weeks were pretreated for 30 min with 10-μg/ml concentrations of the indicated neutralizing antibodies prior to 20 h of culture with magnetic-bead-purified CD19⁺ B cells. FasL expression was evaluated by two-color flow cytometry as for Fig. 6. Ratios of MFI ± standard deviations of triplicate samples compared to those of anti-DNP control antibody (∗) are plotted. Data are from a representative of three experiments.

FIG. 8.

Recombinant IL-4 (rIL-4) and recombinant IL-10 induce additive FasL expression on the B-1a and CD5⁻ B subsets of SEA-stimulated, purified CD19⁺ B cells. Magnetic immunobead-purified, splenic CD19⁺ B cells from mice infected for 8 weeks were cultured for 20 h in the presence of 10 μg of SEA/ml and the indicated doses of recombinant IL-4 and/or IL-10. Cells were paraformaldehyde fixed and stained as described in the legend to Fig. 2. FasL expression was analyzed on (A) gated B-1a (CD5⁺/CD19⁺) and (B) B (CD5⁻/CD19⁺) cells. Ratios of MFI of triplicate samples were determined in comparison to SEA-treated CD5⁻ B cells devoid of exogenous cytokines (∗). Data from a representative of three experiments are plotted, with error bars omitted for clarity.