Anti-cytokine autoantibodies suggest pathogenetic links with autoimmune regulator deficiency in humans and mice

Abstract

Autoimmune polyendocrinopathy candidiasis ectodermal dystrophy (APECED) is a recessive disorder resulting from mutations in the autoimmune regulator (AIRE). The patients' autoantibodies recognize not only multiple organ-specific targets, but also many type I interferons (IFNs) and most T helper type 17 (Th17) cell-associated cytokines, whose biological actions they neutralize in vitro. These anti-cytokine autoantibodies are highly disease-specific: otherwise, they have been found only in patients with thymomas, tumours of thymic epithelial cells that fail to express AIRE. Moreover, autoantibodies against Th17 cell-associated cytokines correlate with chronic mucocutaneous candidiasis in both syndromes. Here, we demonstrate that the immunoglobulin (Ig)Gs but not the IgAs in APECED sera are responsible for neutralizing IFN-ω, IFN-α2a, interleukin (IL)-17A and IL-22. Their dominant subclasses proved to be IgG1 and, surprisingly, IgG4 without IgE, possibly implicating regulatory T cell responses and/or epithelia in their initiation in these AIRE-deficiency states. The epitopes on IL-22 and IFN-α2a appeared mainly conformational. We also found mainly IgG1 neutralizing autoantibodies to IL-17A in aged AIRE-deficient BALB/c mice - the first report of any target shared by these human and murine AIRE-deficiency states. We conclude that autoimmunization against cytokines in AIRE deficiency is not simply a mere side effect of chronic mucosal Candida infection, but appears to be related more closely to disease initiation.

Figure 1

Binding and neutralizing activities of immunoglobulin (Ig) isolated from autoimmune polyendocrinopathy candidiasis ectodermal dystrophy (APECED) and control sera. (a) Enzyme-linked immunosorbent assay (ELISA) was used to test for anti-interferon (IFN)-α2a autoantibodies in patient and control (ctrl) IgG fractions. The neutralizing activity of the isolated IgG and IgA fractions towards IFN-α2a, (b), IFN-ω (c) and interleukin (IL)-22 (d) in cell-based assays. Cytokine neutralizing units (NU) per μg of protein are shown. Nine patient and seven control samples were tested.

Figure 2

Immunoglobulin subclasses against interferon (IFN)-α2a and interleukin (IL)-22 in autoimmune polyendocrinopathy candidiasis ectodermal dystrophy (APECED) (a) and thymoma (b) patients. Antibody levels are shown in luminescence units (LU). Mean values with standard error of the mean are indicated. Thymoma sera were selected by prior testing for autoantibody positivity (n = 30 for anti-IFN-α2a and n = 15 for IL-22). All the sera were tested at least twice in three different experiments. Patient samples are indicated with filled circles, control samples with open triangles.

Figure 3

Epitope mapping of interferon (IFN)-α2a and interleukin (IL)-22. (a) Autoimmune polyendocrinopathy candidiasis ectodermal dystrophy (APECED) sera were tested against full-length and N- and C-terminal fragments of IFN-α2a and IL-22 using luciferase-conjugated polypeptides as antigens. (b) Control (ctrl) or APECED sera having exclusively conformation (P3)- or in addition C-terminus-specific (P1 and P2) autoantibodies to IFN-α2 were tested on Western blot against denatured human IFN-α2. Monoclonal antibody to IFN-α2 was used as a positive control.

Figure 4

Autoantibodies in autoimmune regulator (AIRE)-deficient mice against helper type 17 (Th17) cell-associated cytokines. Autoantibodies were assayed in mouse sera by enzyme-linked immunosorbent assay (ELISA) against murine interferon (IFN)-α4 (a), interleukin (IL)-17A (b), IL-17F (c) and IL-22 (d) and by cell-based neutralizing assay against IL-17A (f). Autoantibody subclasses versus IL-17A are shown in (e). KO: AIRE-deficient mice; ctrl: combined AIRE wild-type and heterozygotes. Mean values with standard error of the mean are indicated. Horizontal lines indicate positive–negative discrimination levels drawn according to the old BALB/c group (mean + 2 standard deviations).