SEB‐induced IL‐13 production in CLA+ memory T cells defines Th2 high and Th2 low responders in atopic dermatitis

Staphylococcus aureus, memory skin-homing cutaneous lymphocyte-associated antigen (CLA)+ T cells and IL-13 constitute relevant players in atopic dermatitis (AD) pathogenesis.1 Since circulating CLA+ T cells reflect cutaneous abnormalities present in human inflammatory skin diseases,2 an ex vivo coculture model made of purified circulating CLA+/− effector and central memory T cells and autologous lesional epidermal cells was established. We show a CLA-dependent production of IL-13 upon activation with staphylococcal enterotoxin B (SEB) that allows the differentiation of the Th2 high and Th2 low groups, with distinct clinical correlations between both groups, within a clinically homogeneous population of adult non-treated moderate-to-severe AD patients. Our results showed that IL-13, together with IL-4, IL-17A, IL-22, CCL17, and CCL22, was preferentially produced by circulating memory CLA+ T cells upon activation with SEB in the presence of autologous lesional epidermal cells (Figure 1A). Interestingly, SEB activation of the CLA+/Epi cocultures resulted in a predominant IL-13 production among the Th2 cytokines (IL-13, IL-4, lL-5) (Figure 1B). The amount of IL-5 and IFN-γ produced by SEB-activated CLA− T cells was higher or similar than that by CLA+ T cells, respectively, suggesting their relationship to extracutaneous sites. This model is stimulus-specific since polyclonal activators such as PMA/Ionomycin and CD3/CD28 are not CLA-specific (Figure S1A), and epidermal cells contributed to the T-cell activation (Figure S1B). Patients were stratified based on the median of the IL-13 response in the SEB-induced CLA+ T-cell AD cocultures (Figure 1C), and we found differentiated T-cell responses to SEB between the Th2 high and the Th2 low groups (Figure 1D). Although both groups were clinically homogeneous (Figure S1C), this stratification suggested differential immunological mechanisms between both groups, since they not only differed in terms of in vitro stimulation, but also in terms of severity, plasma markers, IgE levels against S. aureus and mRNA expression from cutaneous lesions. In the Th2 high group, in contrast to the Th2 low, the IL-13 response by SEB CLA+ T cells directly correlated with EASI score and plasma levels of CCL17 and sIL-2R (Figure 2A-C). This group also showed a direct correlation between anti-S. aureus IgE levels and SEB-induced CLA+ T-cell-mediated IL-13 response in vitro (Figure 2D). The mRNA expression from lesional skin biopsies was similar between both groups (Figure S2A), but the IL-13 produced by SEB-stimulated CLA+/Epi cocultures directly correlated with CCL26 (Figure 2E) and inversely correlated with LCN2 mRNA expression in the Th2 high group (Figure 2F). Additionally, the IL-13/IL-17A and IL-13/IFN-γ ratios in the SEB-stimulated CLA+ T-cell cocultures were higher in the Th2 high than the Th2 low group (Figure S2B), supporting the type 2 signature and the lowered type 17 and type 1 immunity in the Th2 high group, which may facilitate S. aureus infection.3 The study has a few limitations. We did not study the presence of S. aureus in the skin of the AD patients and the number of patients was not very high but we found consistent significant results on the relationship between the SEB-induced CLA+ T-cell IL-13 response and clinical features of the patients. The novelty of our results relies on the separation of the Th2 high and low populations, corresponding with disease activity, based on the CLA+ T-cell IL-13 response to SEB, which are key mediators in AD pathogenesis.4 Interestingly, the existence of Th2 high and low groups in non-treated moderate-to-severe AD patients has been shown by serum proteomic profiling.5 In conclusion, we consider that this new and translational approach allows obtaining readouts on cytokine production that complement current studies based on transcriptomics and flow cytometry and may help to explore the complex heterogeneity of AD pathophysiology from a more functionally point of view. The study was funded by FIS/ISCIII (Ministerio de Economía y Competitividad e Instituto de Salud Carlos III) 2021 (PI21/01179 and PI21/00335) and Fondo Europeo del Desarrollo Regional (FEDER). Additionally, Sans-De San Nicolàs L was granted by a PhD fellowship from the Agency for Management of University and Research Grants of the Generalitat de Catalunya (FI-SDUR 2020); De Jesús-Gil C was granted by a PhD fellowship from the Agency for Management of University and Research Grants of the Generalitat de Catalunya (FI-2018), co-financed with FEDER; García-Jiménez I was granted by a PhD fellowship from the Universitat de Barcelona (PREDOCS-UB 2020). We are grateful to all individuals who participated in our study. Antonio Guilabert is a consultant for Sanofi, Almirall, and AbbVie. Laia Curto-Barredo is a consultant for Sanofi, AbbVie, Leo Pharma, and Lilly. Esther Serra-Baldrich is a consultant for Sanofi, Almirall, Leo Pharma, Pfizer, Galderma, and Lilly. Michael D. Howell is an employee and shareholder of DermTech. The rest of authors declare no conflict of interests. Appendix S1 Figure 1 Figure 2 Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.