Reduced CD5(+) CD24(hi) CD38(hi) and interleukin-10(+) regulatory B cells in active anti-neutrophil cytoplasmic autoantibody-associated vasculitis permit increased circulating autoantibodies

Abstract

Pathogenesis of anti-neutrophil cytoplasmic autoantibody (ANCA)-associated vasculitis is B cell-dependent, although how particular B cell subsets modulate immunopathogenesis remains unknown. Although their phenotype remains controversial, regulatory B cells (Bregs ), play a role in immunological tolerance via interleukin (IL)-10. Putative CD19(+) CD24(hi) CD38(hi) and CD19(+) CD24(hi) CD27(+) Bregs were evaluated in addition to their CD5(+) subsets in 69 patients with ANCA-associated vasculitis (AAV). B cell IL-10 was verified by flow cytometry following culture with CD40 ligand and cytosine-phosphate-guanosine (CpG) DNA. Patients with active disease had decreased levels of CD5(+) CD24(hi) CD38(hi) B cells and IL-10(+) B cells compared to patients in remission and healthy controls (HCs). As IL-10(+) and CD5(+) CD24(hi) CD38(hi) B cells normalized in remission within an individual, ANCA titres decreased. The CD5(+) subset of CD24(hi) CD38(hi) B cells decreases in active disease and rebounds during remission similarly to IL-10-producing B cells. Moreover, CD5(+) B cells are enriched in the ability to produce IL-10 compared to CD5(neg) B cells. Together these results suggest that CD5 may identify functional IL-10-producing Bregs . The malfunction of Bregs during active disease due to reduced IL-10 expression may thus permit ANCA production.

Keywords: AAV; ANCA; CD5; interleukin-10; regulatory B cells.

Fig 1

CD5⁺CD24^hiCD38^hi and IL-10⁺ B cells decrease during active disease and rebound during remission. Population analysis of the percentage of CD24^hiCD38^hi B cells (a), the CD5⁺ subset of CD24^hiCD38^hi B cells (b) and interleukin (IL)-10⁺ B cells (c) is shown for healthy controls (HCs) (HC, n = 14–30), patients with active anti-neutrophil cytoplasmic autoantibody (ANCA)-associated vasculitis (AAV) (ACT, n = 10–29), remitting AAV (ANCA REM, n = 23–43), active myeloperoxidase (MPO)-ANCA (MPO ACT, n = 5–13), remitting MPO-ANCA (MPO REM, n = 15–25), active proteinase 3 (PR3)-ANCA (PR3 ACT, n = 5–16) and remitting PR3-ANCA (PR3 REM, n = 7–18). IL-10 producing B cells (c) were analysed after 96 h of peripheral blood mononuclear cells (PBMC) stimulation with CD40 ligand and cytosine–phosphate–guanosine (CpG) DNA. Aggregate data show increased CD24^hiCD38^hi B cells during disease remission (a) but no significant change during active disease. Percentages of both the CD5⁺ subset of CD24^hiCD38^hi B cells and IL-10⁺ B cells decrease during active disease and rebound during remission. Lines indicate median values. *P ≤ 0·005; **P ≤ 0·001 and ***P ≤ 0·0001.

Fig 2

CD24^hiCD38^hi, CD5⁺CD24^hiCD38^hi and interleukin (IL)-10⁺ B cells increase during remission in paired active and remission samples. Paired active and remission disease samples from nine patients demonstrate an increase in CD24^hiCD38^hi B cells (a) as an individual transition from active disease to remission. Paired active and remission disease samples from seven patients demonstrate that the CD5⁺ subset of CD24^hiCD38^hi B cells (b) decreased in patients with active disease and increased during remission. Paired active and remission disease samples from six patients demonstrate a similar increase in CD19⁺IL-10⁺ B cells (c) as all individuals transition from active disease to remission. The line indicates the median value for healthy controls (HCs).

Fig 3

CD5⁺ B cells are enriched in B cells capable of producing interleukin (IL)-10. Total B cells, CD5⁺ B cells and CD5^neg B cells were evaluated for IL-10-producing B cells after 72–96 h of stimulation with CD40 ligand and (CpG) DNA. Shown are representative flow histograms depicting the percentage of IL-10⁺ B cells in cultured populations of total B cells (a), CD5^neg B cells (b) and CD5⁺ B cells (c) demonstrating that CD5⁺ B cells are enriched in IL-10-producing B cells compared to CD5^neg and total B cells.

Fig 4

As CD24^hiCD38^hi, CD5⁺ CD24^hiCD38^hi or IL-10⁺ B cells increase, circulating anti-neutrophil cytoplasmic autoantibody (ANCA) titres decrease. Paired active and remission sample data were used to generate the change (Δ) in each B cell subset CD24^hiCD38^hi (a) CD5⁺CD24^hiCD38^hi (b) or interleukin (IL)-10⁺ (c) by the subtraction of the active value from the remission value (R–A = Δ). A positive value for the ΔB cell subset indicates a greater percentage of B cells present during remission than during active disease. The change in ANCA titre was generated by subtracting the ANCA titre during active disease from the ANCA titre at remission, utilizing the same time-points used for B cell subset data. A negative value for ΔANCA titre indicates a greater titre during active disease than remission. Statistical significance was determined with a signed rank test; P ≤ 0·05 is considered significant.