Autoantibodies against IL-17A, IL-17F, and IL-22 in patients with chronic mucocutaneous candidiasis and autoimmune polyendocrine syndrome type I

We searched for auto-Abs against IL-17 cytokines (IL-17A, IL-17F, IL-22, and IL-26, which is absent from mice) or against IL-17–inducing cytokines (IL-1β, IL-6, IL-21, IL-23, and TGF-β) in 33 patients diagnosed with APS-I on the basis of autoimmune polyendocrinopathy, circulating auto-Abs against IFN-α and –ω, and the presence of two mutant AIRE alleles (Table I). CMC was observed in 29 out of the 33 patients (Table I). Multiplex particle-based flow cytometry revealed a high fluorescence intensity (FI; >1,000, arbitrary definition) of IgG auto-Abs against IL-17A in the plasma of 22 patients, with a high FI of auto-Abs against IL-17F in 31 patients and of auto-Abs against IL-22 in 30 patients (Fig. 1 and Table I). All 33 patients had significant (P = 6.4 × 10⁻⁹, 4.6 × 10⁻¹², and 2.16 × 10⁻¹¹, respectively) levels of auto-Abs against at least one of these three cytokines. 5 patients had a significant reaction against a single cytokine, 6 had a reaction against two cytokines, and 22 had a reaction against all three cytokines. No auto-Abs against IL-17A, IL-17F, and IL-22 were found in plasma samples from the 37 healthy individuals tested. Similarly, no auto-Abs against any of these three cytokines were found in plasma samples from another 103 patients with various autoimmune conditions (Fig. S4, A and B). No auto-Abs against IL-1β, IL-6, IL-23, or IL-26 (whether as a monomer or a dimer) were found in the APS-I patients (Fig. S3, A–D). There were no auto-Abs against IL-12, IFN-γ, or GM-CSF (Fig. S3, E–G), consistent with the lack of mycobacterial disease and alveolar proteinosis in patients with APS-I. No auto-Abs were found against any of the other cytokines tested, including IL-10, IL-18, IFN-β, and TNF (Fig. S3, H–K). As expected, all APS-I patients had auto-Abs against IFN-α (Fig. 1 and Table I) and IFN-ω (not depicted). In addition, no IgA antibodies, a key element of mucosal immunity, directed against IL-17A, IL-17F, or IL-22 were detected in the plasma of 21 APS-I patients by classical ELISA (not depicted).

We also ran classical ELISA, with plasma samples diluted by a factor of five more than in the multiplex assays. 4 out of the 30 patients tested had high titers of IgG (OD > 1) auto-Abs against IL-17A (P = 0.056), 10 patients had high titers of auto-Abs against IL-17F (P = 1.4 × 10⁻⁷), and 12 patients had high titers of auto-Abs against IL-22 (P = 1.23 × 10⁻⁷; Fig. S1 and Table I). 17 patients had auto-Abs against at least one of these three cytokines, including 10 patients with auto-Abs against a single cytokine, 5 patients with auto-Abs against two cytokines, and 2 patients with auto-Abs against all three cytokines. No such auto-Abs were detected in the 37 healthy controls tested. This assay was not used to test patients with various autoimmune diseases. There were no detectable auto-Abs against TGF-β1, IL-6, IL-21, IL-23, or IL-26, or against IL-12 or IFN-γ in any of the APS-I patients (Fig. S2, A–G). Thus, in both ELISA and multiplex particle-based flow cytometry, the only auto-Abs detected in the 33 patients with APS-I tested were directed against IFN-α, IFN-ω, IL-17A, IL-17F, and IL-22 (Fig. 1, Fig. S1, and Table I). Moreover, all 33 patients with APS-I had auto-Abs against at least one out of three IL-17 cytokines (IL-17A, IL-17F, and IL-22). High titers of auto-Abs against all three of these cytokines were found in 22 patients.

We performed Western blotting to assess the specificity of these auto-Abs. Plasma samples from five patients with APS-I and auto-Abs against IL-17A, IL-17F, and/or IL-22, as detected by ELISA and multiplex particle-based flow cytometry, and from a healthy individual were tested for their ability to recognize rIL-17A, rIL-17F, and rIL-22 on Western blots. The three cytokines were clearly recognized by plasma samples from patients but not by the control plasma (Fig. 2). The plasma samples from patients did not recognize IL-23 in a control assay. Finally, we investigated whether the auto-Abs against IL-17A neutralized this cytokine in a simple biological assay. We incubated fibroblasts from a healthy control with up to 50 ng/ml rIL-17A in the presence of plasma from healthy individuals, or from a patient with APS-I and auto-Abs against IL-17, as detected by ELISA and Western blotting. We then measured the induction of IL-6 by ELISA on the supernatant. Strikingly, even strong dilutions of plasma samples from patients (up to 0.1%; not depicted) completely abolished fibroblastic cell responses to such high concentrations of rIL-17 (Fig. 3), whereas control plasma had no effect. Lower concentrations of IL-17A failed to induce IL-6, even in the presence of control plasma (unpublished data). The lack of robust bioassays precluded assessments of the possible neutralizing effects of auto-Abs against IL-17F and IL-22. Thus, all patients with APS-I tested had high titers of auto-Abs against IL-17A, IL-17F, and IL-22 in their plasma, and these auto-Abs were specific and neutralizing, at least in the patients tested. These data suggest that most patients with APS-I carry neutralizing auto-Abs against IL-17A, IL-17F, and IL-22.

APS-I patients may display auto-Abs against IL-17 cytokines as a result of impaired AIRE-dependent tolerance induction if the expression of the corresponding cytokine genes is normally controlled by AIRE in the thymus or in secondary lymphoid organs. Whatever the mechanism, these data suggest that impaired IL-17 immunity may account for CMC in patients with APS-I. In particular, these data are consistent with the impaired IL-17 immunity reported in other patients with known inborn errors of immunity predisposing to CMC (and other infectious diseases). Indeed, patients with low proportions (IL12B, IL12RB1) or a lack of (STAT3) IL-17–producing circulating T cells suffer from mild (IL12B, IL12RB1) or severe (STAT3) CMC, respectively. Patients with APS-I and auto-Abs against IL-17 cytokines do not suffer from severe staphylococcal disease, unlike STAT3-deficient patients (Freeman and Holland, 2009). Impaired IL-17 immunity probably contributes to staphylococcal disease in STAT3-deficient patients, as epithelial cells in the skin and lungs, the organs most frequently affected by staphylococcal infection in these patients, specifically rely on IL-17 stimulation for the induction of known antistaphylococcal target genes (Minegishi et al., 2009). The lack of overt staphylococcal disease in most APS-I patients may result from residual IL-17 immunity, as observed to an even greater extent in patients with IL-12p40 and IL-12Rβ1 deficiencies who also control staphylococcal infections normally (Filipe-Santos et al., 2006). In any case, proof that human IL-17 cytokines play an essential role in immunity to Candida will require the identification of mutations specifically impairing IL-17 immunity in other patients with inherited, isolated CMC without APS-I (Ryan et al., 2008; Hong et al., 2009). The recent observation that some patients with isolated CMC have smaller than normal proportions of IL-17–producing T cells and produce low levels of IL-17 suggests that this may be plausible (unpublished data; Eyerich et al., 2008).

Various human diseases seem to be caused by the production of auto-Abs against cytokines. Patients with pulmonary alveolar proteinosis were shown to display auto-Abs against GM-CSF in 1999 (Kitamura et al., 1999). Mutations in the GM-CSF receptor were subsequently found in other patients, establishing a causal relationship between impaired GM-CSF immunity and alveolar proteinosis (Martinez-Moczygemba et al., 2008; Suzuki et al., 2008). Patients with mycobacterial diseases and inherited IFN-γR1 deficiency were first reported in 1996 (Filipe-Santos et al., 2006). Autoimmune phenocopies, with auto-Abs against IFN-γ, were subsequently found in other patients (Döffinger et al., 2004; Kampmann et al., 2005; Patel et al., 2005). We recently reported a patient with auto-Abs against IL-6 and staphylococcal disease (Puel et al., 2008). There may be a causal relationship between auto-Abs against IL-6 and staphylococcal disease, as STAT3- and Tyk2-deficient patients suffer from staphylococcal disease and have a poor cellular response to IL-6 (Minegishi et al., 2006, 2007, 2009). However, no germline mutation in the gene encoding IL-6 or its receptor has yet been reported in similar patients. This description of patients with auto-Abs against IL-17 cytokines is the fourth reported example of pathogenic auto-Abs against human cytokines. Other auto-Abs against cytokines not tested in this study, including some that would impair the development of IL-17–producing T cells, may also contribute to the development of CMC in APS-I patients.

33 patients with APS-I (18 male and 15 female patients; Table I and Table S1), 103 patients with endocrine/autoimmune disorders, and 37 healthy controls (15 male and 22 female subjects) were enrolled in this study, with informed consent and approval obtained by the Necker Hospital and Medical School Institutional Review Board (IRB) and the Rockefeller University IRB.

Plasma and serum samples from the patients and controls were frozen at −20°C immediately after collection.

ELISA was performed as previously described (Puel et al., 2008). In brief, 96-well ELISA plates (MaxiSorp; Thermo Fisher Scientific) were coated by incubation overnight at 4°C with 2 µg/ml rIL-6, rIL-12p70, rIL-17A, rIL-17F, rIL-21, rIL-22, rIL-23, rIL-26 monomer, rIL-26 dimer, rTGF-β, and rIFN-γ (R&D Systems). Plates were then washed (PBS/Tween 0.005%), blocked by incubation with the same buffer supplemented with 5% nonfat milk powder, washed, and incubated with 1:500 dilutions of plasma samples from the patients or controls for 2 h at room temperature (or with specific mAbs as positive controls). Plates were thoroughly washed. Horseradish peroxidase (HRP)–conjugated Fc-specific IgG fractions from polyclonal goat antiserum against human IgG or IgA (Nordic Immunological Laboratories) were added to a final concentration of 2 µg/ml. Plates were incubated for 1 h at room temperature and washed. Substrate was added and OD was measured.

Recombinant human cytokines (rIL-1β, rIL-6, rIL-10, rIL-12p40, rIL-12p70, rIL-17A, rIL-17F, rIL-18, rIL-22, rIL-23, rIL-26 monomer, rIL-26 dimer, rIFN-α, rIFN-β, rIFN-γ, rTNF, and rGM-CSF; R&D Systems) were covalently coupled to carboxylated beads (Bio-Plex; Bio-Rad Laboratories). Beads were first activated with 1-ethyl-3-[3-dimethylaminopropyl]carbodiimide hydrochloride (Thermo Fisher Scientific) in the presence of N-hydroxysuccinimide (Thermo Fisher Scientific), according to the manufacturer’s instructions, to form amine-reactive intermediates. The activated beads were incubated with the corresponding cytokines at a concentration of 20 µg/ml in the reaction mixture for 3 h at room temperature on a rotator. Beads were washed and stored in blocking buffer (10 mM PBS, 1% BSA, 0.05% NaN₃). Cytokine-coupled beads were incubated with plasma or serum from patients for 1 h in 96-well filter plates (MultiScreenHTS; Millipore) at room temperature in the dark on a horizontal shaker. Fluids were aspirated with a vacuum manifold and beads were washed three times with 10 mM PBS/0.05% Tween 20. Beads were incubated for 30 min with a PE-labeled anti–human IgG-Fc antibody (Leinco/Biotrend), washed as described, and resuspended in 100 µl PBS/Tween. They were then analyzed on an analyzer (Bio-Plex) using Bio-Plex Manager 3.0 software (both from Bio-Rad Laboratories). Successful coupling of the cytokines to their respective bead sets was verified with specific mAbs.

We subjected 500 ng rIL-17A, rIL-17, rIL-22, rIL-23 (R&D Systems), or BSA to SDS-PAGE (10% acrylamide) under reducing conditions. The protein bands were electroblotted onto nitrocellulose membranes (iBlot Gel Transfer Stacks; Invitrogen). Membranes were blocked by incubation for 1 h at room temperature with PBS supplemented with 5% BSA and 0.05% Tween 20, and were washed and incubated overnight at 4°C with plasma samples from a patient or a control diluted 1:500 in PBS, 5% BSA, 0.01% Tween 20. The membranes were washed three times and incubated for 1 h at room temperature with HRP-coupled anti–human IgG and used at a final concentration of 0.67 µg/ml. The membranes were washed three times and developed for ECL (GE Healthcare).

SV-40–transformed control fibroblasts were plated in 24-well plates (60,000 fibroblasts/well) in 0.5 ml DMEM/10% FCS (Invitrogen). On the following day, cells were washed in 1× PBS (Invitrogen) and incubated with 10% plasma (from a control or a patient) in DMEM. They were left unstimulated or were stimulated with 50 ng/ml rIL-17A (R&D Systems). The supernatants were collected 24 and 48 h after stimulation, and IL-6 production was assessed by ELISA (Sanquin Mast Diagnostic) according to the manufacturer’s instructions.

Fig. S1 shows anti–IL-17A, –IL-17F, and –IL-22 IgG auto-Abs measured by ELISA in plasma from APS-I patients and controls. Fig. S2 shows anti–TGF-β1, –IL-6, –IL-21, –IL-23, –IL-12, –IFN-γ, and –IL-26 IgG auto-Abs measured by ELISA in plasma from controls and APS-I patients. Fig. S3 shows anti–IL-1β, –IL-6, –IL-23, –IL-26, –IL-12, –IFN-γ, –GM-CSF, –IFN-β, –IL-10, –IL-18, and -TNF IgG auto-Abs measured by multiplex particle-based flow cytometry in plasma from healthy controls and APS-I patients. Fig. S4 shows anti–IFN-α, –IL-17A, –IL-17F, and –IL-22 IgG auto-Abs in 100 patients suffering from autoimmune and/or endocrine disorders. Table S1 shows the autoimmune/endocrine phenotypes and treatments of each APS-I patient tested.

We would like to thank all members of the Paris and New York branches of the laboratory for helpful discussions; T. Leclerc for technical assistance; M. Courat, M. N'Guyen, B. Delaney, and Y. Nemirovskaya for secretarial assistance; and Frédéric Rieux-Laucat and B. Grimbacher for discussions. We would also like to thank the clinicians who provided us with samples and the patients who agreed to participate in our study.

The Laboratory of Human Genetics of Infectious Diseases is supported by the Rockefeller University Center for Clinical and Translational Science (grant 5UL1RR024143-03) and the Rockefeller University. We would like to thank the Rockefeller University, the Institut National de la Santé et de la Recherche Médicale, the University Paris Descartes, the Agence Nationale de la Recherche, the European Union, the March of Dimes, the Dana Foundation, the Addenbrooke's Charitable Trust, the National Institute for Health Research Cambridge Biomedical Research Center, and the Howard Hughes Medical Institute for their support. L. Marodi was supported by the Hungarian Scientific Research Fund.

The authors have no conflicting financial interests.