Molecular explanation for the contradiction between systemic Th17 defect and localized bacterial infection in hyper-IgE syndrome

Abstract

Hyper-IgE syndrome (HIES) is a primary immunodeficiency characterized by atopic manifestations and susceptibility to infections with extracellular pathogens, typically Staphylococcus aureus, which preferentially affect the skin and lung. Previous studies reported the defective differentiation of T helper 17 (Th17) cells in HIES patients caused by hypomorphic STAT3 mutations. However, the apparent contradiction between the systemic Th17 deficiency and the skin/lung-restricted susceptibility to staphylococcal infections remains puzzling. We present a possible molecular explanation for this enigmatic contradiction. HIES T cells showed impaired production of Th17 cytokines but normal production of classical proinflammatory cytokines including interleukin 1beta. Normal human keratinocytes and bronchial epithelial cells were deeply dependent on the synergistic action of Th17 cytokines and classical proinflammatory cytokines for their production of antistaphylococcal factors, including neutrophil-recruiting chemokines and antimicrobial peptides. In contrast, other cell types were efficiently stimulated with the classical proinflammatory cytokines alone to produce such factors. Accordingly, keratinocytes and bronchial epithelial cells, unlike other cell types, failed to produce antistaphylococcal factors in response to HIES T cell-derived cytokines. These results appear to explain, at least in part, why HIES patients suffer from recurrent staphylococcal infections confined to the skin and lung in contrast to more systemic infections in neutrophil-deficient patients.

Figure 1.

HIES T cells produce greatly reduced amounts of Th17 cytokines and normal amounts of classical proinflammatory cytokines upon activation. PBMCs from HIES patients (Pt.) and control subjects (Cont.; n = 8 each, indicated by dots) were stimulated (+) or not (−) with anti-CD3 and anti-CD28 for 72 h, and the concentration of the indicated cytokines in their culture supernatants was determined by ELISA. The results shown are representative of three independent experiments. **, P < 0.01.

Figure 2.

Supernatants of activated HIES T cells fail to stimulate keratinocytes to secrete significant amounts of antibacterial factors. In the presence or absence of anti–IL-17A plus anti–IL-22 (A and B), anti-BD3 (C), or isotype-matched control antibodies, primary human keratinocytes were incubated for 48 h with the supernatants of HIES (Pt.) or control (Cont.) T cells that had been stimulated (+) or not (−) with anti-CD3 and anti-CD28 for 72 h as in Fig. 1. (A) The concentration of CXCL8, BD2, and BD3 in keratinocytes supernatants was determined by ELISA. Representative data from one patient and one control are shown (mean ± SD; n = 3), and similar results were obtained from the other patients and controls. (B and C) The culture supernatants of keratinocytes were analyzed for their antistaphylococcal activity by the colony assay (mean ± SD; n = 3). The results shown in are representative of at least three independent experiments. **, P < 0.01.

Figure 3.

Keratinocytes and bronchial epithelial cells show greater dependence on Th17 cytokines for the production of chemokines and BDs than other cell types. Primary human keratinocytes, bronchial epithelial cells, dermal fibroblasts, endothelial cells (HUVEC and HMVEC-L), and macrophages were incubated for 48 h with T cell supernatants that were prepared as described in Fig. 1 A or with the Th17 cytokine cocktail (Th17 mix: IL-17A + IL-17F + IL-22), the classical proinflammatory cytokine cocktail (classical mix: TNF-α + IL-1β + IFN-γ), or the combination of both (both mix; B). The concentration of CXCL8, BD2, and BD3 in their culture supernatants was determined by ELISA. Representative data from one patient and one control are shown in A (mean ± SD; n = 3), and similar results were obtained from the other patients and controls. The results shown are representative of at least three independent experiments. **, P < 0.01.

Figure 4.

Distinct expression and regulation of the cytokine receptors in keratinocytes and fibroblasts compared with those in other cell types. (A) The expression of IL-1R1, IL-1Ra, and IL-1R2 in the indicated cells was determined by quantitative RT-PCR. Data shown were normalized to HPRT levels, and the level of expression in keratinocytes was defined as 1.0. (B) IL-1R1, IL-1R2, and IL-1Ra proteins in keratinocytes and fibroblasts were detected by immunoblotting. (C) Keratinocytes and fibroblasts were cultured for 15 min with the indicated concentration of IL-1β and analyzed by quantitative RT-PCR for the expression of c-Fos and IL-6. Data shown were normalized to HPRT levels, and the level of expression in cells cultured without IL-1β was defined as 1.0 for each cell type. (D) Keratinocytes cultured as in Fig. 3 B were analyzed by quantitative RT-PCR for the expression of IL-17RA, IL-17RC, and IL-22R. The data shown were normalized to the HPRT levels, and the level of expression in cells cultured without any added cytokine was defined as 1.0. The results shown are representative of three independent experiments. Error bars show mean ± SD (n = 3). *, P < 0.05; **, P < 0.01.

Figure 5.

HIES T cells show poor ability of stimulating keratinocytes in response to staphylococcal superantigens and candida antigens. (A) PBMCs from HIES patients (Pt.) and control subjects (Cont.; n = 6 each, indicated by dots) were stimulated or not (−) with SEB (S, 100 ng/ml) or CAA (C, 1/20,000 vol/vol) for 5 d, and the concentration of the indicated cytokines in their culture supernatant was determined by ELISA. (B and C) Fibroblasts and keratinocytes were cultured for 48 h in the absence (−) or presence (S) of SEB or with the supernatants of patients (Pt.) or control (Cont.) PBMCs that had been unstimulated (−) or stimulated with SEB (S) as in A, in the presence or absence of anti–IL-17A + anti–IL-22 or isotype-matched control antibodies. Their culture supernatants were analyzed by ELISA for the secretion of CXCL8 and BD2 (B) and evaluated for their neutrophil chemotactic activity (C). The results shown are representative of two independent experiments. Error bars show mean ± SD (n = 3). **, P < 0.01.