Phenotypic and Functional Comparison of Class Switch Recombination Deficiencies with a Subgroup of Common Variable Immunodeficiencies

Abstract

Primary antibody deficiencies (PADs) are the most common immunodeficiency in humans, characterized by low levels of immunoglobulins and inadequate antibody responses upon immunization. These PADs may result from an early block in B cell development with a complete absence of peripheral B cells and lack of immunoglobulins. In the presence of circulating B cells, some PADs are genetically caused by a class switch recombination (CSR) defect, but in the most common PAD, common variable immunodeficiency (CVID), very few gene defects have as yet been characterized despite various phenotypic classifications. Using a functional read-out, we previously identified a functional subgroup of CVID patients with plasmablasts (PBs) producing IgM only. We have now further characterized such CVID patients by a direct functional comparison with patients having genetically well-characterized CSR defects in CD40L, activation-induced cytidine deaminase (AID) and uracil N-glycosylase activity (UNG). The CSR-like CVID patients showed a failure in B cell activation patterns similar to the classical AID/UNG defects in three out of five CVID patients and distinct more individual defects in the two other CVID cases when tested for cellular activation and PB differentiation. Thus, functional categorization of B cell activation and differentiation pathways extends the expected variation in CVID to CSR-like defects of as yet unknown genetic etiology.

Keywords: B cells; Primary antibody deficiencies; class switch recombination defect; common variable immunodeficiency.

Fig. 1

Representative figures of the phenotype of circulating B cells from healthy adult controls, healthy cord bloods, and CD40L-, AID-, and UNG-deficient patients. B cell subsets of representative blood samples from healthy adult and cord blood samples, as well as from genotyped CD40L-, AID-, and UNG-deficient patients. Numbers indicate mean percentages of multiple experiments in the corresponding quadrant. Healthy adult controls (N = 20), healthy cord bloods (N = 15), and CD40L-, AID-, and UNG-deficient patients (N = 2–5)

Fig. 2

Proliferation and differentiation of B cells from CSR-deficient patients upon activation. The capacity of B cells from healthy adult controls, healthy cord bloods, and CD40L-, AID-, and UNG-deficient CSR patients to proliferate and differentiate in vitro were tested. CFSE-labeled PBMCs were cultured for 6 days, normalized for B cell numbers (1 × 10⁵ B cells/well). T cell-independent B cell activation was tested with CpG in the presence of IL-2. T cell-dependent B cell stimulation was mimicked by the combinations of αIgM/αCD40/IL-21. Effect of T cell stimulation was mimicked by αCD3/αCD28 stimulation, targeting T cells specifically. Representative FACS plots are shown of B cell subset distribution after 6 days of culture in the presence of the indicated stimuli. Gated on CD19⁺ lymphocytes to show CFSE dilution indicating proliferation after 6 days of culture, and to demonstrate the emergence of the subsets of Ig-producing B cells, i.e., plasmablasts and/or plasmacells (sIgD⁻/CD27⁺⁺/CD38⁺⁺)

Fig. 3

Production and release of immunoglobulins from CSR-deficient patients and CSR-like CVID patients upon B cell activation. IgM and IgG levels were measured in the supernatants by ELISA after 6 days of culture. Plotting multiple experiments per patient group (N = 2–5). Immunoglobulin production shown is normalized to that of the healthy control samples used in each separate experiment, production of Ig’s after stimulated with CpG/IL-2 is set at 100 %. Controls range from 2700 to 5000 pg/ml for IgG and 4000–9000 pg/ml for IgM. ND not done

Fig. 4

Representative diagrams of the circulating B cell phenotype at time of analysis. B cell subsets of representative blood samples from a healthy adult control, healthy cord blood, CD40L-, AID-, UNG-deficient CSR patients and five CSR-like CVID patients. Numbers indicate mean percentages of multiple experiments in the corresponding quadrant. Healthy adult controls (N = 20), healthy cord bloods (N = 15), CD40L-, AID- and UNG-deficient patients (N = 2–5), CSR-like CVID patients (N = 2–5)