Staphylococcus aureus enterotoxins modulate IL-22-secreting cells in adults with atopic dermatitis

Atopic dermatitis (AD) is a chronic inflammatory immune-mediated skin disease characterized by skin colonization by Staphylococcus aureus. Interleukin (IL)-22, in cooperation with IL-17, triggers antimicrobial peptide elaboration and enhances certain immunological responses. In AD, IL-22 is related to epidermal hyperplasia, keratinocyte apoptosis, and inhibition of antimicrobial peptide (AMP) production. We aimed to evaluate the impact of staphylococcal enterotoxins on the Tc22/Th22 induction in the peripheral blood of AD patients and on CD4+/CD8+ T cells expressing IL-22 in AD skin. Our study showed inhibition of the staphylococcal enterotoxins A and B (SEA and SEB) response by Th22 (CD4+IL-22+IL-17A−IFN-γ−) cells in AD patients. In contrast, Tc22 (CD8+IL-22+IL-17A−IFN-γ−) cells were less susceptible to the inhibitory effects of staphylococcal enterotoxins and exhibited an enhanced response to the bacterial stimuli. In AD skin, we detected increased IL-22 transcript expression and T lymphocytes expressing IL-22. Together, our results provide two major findings in response to staphylococcal enterotoxins in adults with AD: dysfunctional CD4+ IL-22 secreting T cells and increased Tc22 cells. Our hypothesis reinforces the relevance of CD8 T cells modulated by staphylococcal enterotoxins as a potential source of IL-22 in adults with AD, which is relevant for the maintenance of immunological imbalance.

Furthermore, Th2 (IL-4, IL-13, and IL-31) and Th22 (IL-22) cytokines induce the inhibition of certain skin barrier protein-encoding genes, such as filaggrin, loricrin, and involucrin [19][20][21] . Th22 cells have demonstrated a capacity to infiltrate AD skin and release IL-22 and TNF-α. Likewise, significant amounts of CD3 + CD8 + T cells from inflammatory skin diseases can also stimulate IL-22 secretion 22,23 . In summary, the major function of IL-22 secreting cells in AD is related to keratinocyte effects via the STAT3 (signal transducer and activator of transcription 3) pathway, enhancing cell proliferation and migration and resulting in hyperplasia, keratinocyte apoptosis, and inhibition of AMP production 19,22 .
S. aureus colonizes skin in approximately 30-50% of healthy adults, but it is constantly detected in only 10-20% of such individuals 24 . In AD patients, more than 90% of the skin exhibits an increase in S. aureus colonization as compared with the skin of healthy subjects 25 . Chronic colonization in AD skin remains a relevant issue, considering the widespread occurrence of antibiotic-resistant strains (methicillin-resistant Staphylococcus aureus; MRSA) 26 .
In AD patients, cutaneous S. aureus is capable of inducing myeloid-derived suppressor cells, leading to in vivo immune suppression of the T cell activation in the skin; in mice exposed to wild-type S. aureus on skin, there are reduced numbers of CD4 + and CD8 + T cells in the spleen 27 . Moreover, evidence of decreased peripheral blood mononuclear cell (PBMC) proliferation response to staphylococcal enterotoxin A (SEA) and other recall antigens and mitogens (TT, CMA, and PHA) suggests a compromised immune profile in adults with AD 28 .
Effector memory T cells directed against antigens derived from skin pathogens such as S. aureus are important for providing instant protection against viral and/or bacterial infections 29 . CD4 T cell functions include cytokine production to control infections, in addition to optimizing and maintaining CD8 T cell memory 30,31 . In this study, we aimed to evaluate the potential role of staphylococcal enterotoxins (SEA and SEB) in modulating IL-22 producing CD4 + /CD8 + T cells in adults with atopic dermatitis.

Staphylococcal enterotoxins inhibit CD4 + T cells secreting IL-22.
In a previous study, our group demonstrated a decreased PBMC proliferation response to SEA compared with SEB stimulation, recall antigens, and mitogens in adults with AD 28 . Considering that IL-22 is beneficial to the host in many infectious and inflammatory disorders, with inherent pro-inflammatory properties 32 , we analyzed the effects of staphylococcal enterotoxins on IL-22 secretion by PBMCs in AD patients.
First, evaluating whether IL-22 levels changed in AD patients at the systemic level, we detected a significant increase of IL-22 serum levels in AD patients compared with the healthy control (HC) group (Fig. 1A). Next, IL-22 secretion induced by SEA and SEB stimulation in PBMC from AD patients and controls was assessed. Figure 1B shows a marked increase of IL-22 levels in AD patients with both SEA and SEB stimulation compared with the HC group (Fig. 1B).
The gating strategy for CD4 + /CD8 + T cells producing IL-22, as analyzed by flow cytometry, is shown in Fig. 1C. The frequency of these cells was evaluated after stimulation of PBMCs with SEA and SEB. The frequency of CD4 + IL-22 + T cells was decreased when compared with the controls (Fig. 1D). A different profile from that in CD4 + T cells was seen in CD8 + T cells, where IL-22 was significantly increased following SEA stimulation (Fig. 1E). Moreover, no changes in the frequencies of CD4 + or CD8 + T cells secreting IL-17 under SEA and SEB stimulation were detected in DA individuals compared with the HC group (data not shown). It is possible that IL-17-secreting cells are less refractory to the inhibitory effects of SEA/SEB. There was no relationship between disease severity and cytokine frequency (data not shown).
There was a remarkable difference between the Th22 and Tc22 cell responses in AD patients. CD4 + IL-22 + T cells showed a diminished response to SEA/SEB stimuli when comparing AD patients with HC subjects (Fig. 2B). In contrast, when evaluating CD8 + T cells, we found augmented levels of IL-22 after stimulation in AD patients compared with those in HC subjects (Fig. 2C). These findings may indicate that staphylococcal enterotoxins have opposite effects on T cell subsets, suppressing Th22 cells and enhancing Tc22 cells in AD.

Discussion
Our study showed compromised responses of Th22 cells to SEA and SEB in the peripheral blood of AD patients, with an increase in Tc22 cells, which appear to be less susceptible to the enterotoxins' effects. Moreover, we demonstrated the presence of CD4 + /CD8 + T cells secreting IL-22 within the dermis of AD skin.
The present group of atopic patients had long-lasting AD (with a mean of 19.45 years), with increased circulating IL-22 levels and high levels secreted by PBMCs after stimulation with SEA and SEB. The main source of IL-22, a cytokine often released by inflamed tissue and detected in this study in both serum and in PBMCs, may include CD4 + T cells (e.g., Th17, Th22, Th1, and smaller numbers of Th2 cells) 22,34,35 , CD8 + T cells, natural killer cells, and dendritic cells 22,37,38 .
However, when we evaluated intracellular IL-22 expression in CD4 + T cells, a diminished frequency was detected, in contrast to the increased frequency of CD8 + T cells secreting IL-22, induced by SEA and SEB stimulation in AD patients. Interestingly, the decreased IL-22 secretion by CD4 + T cells is reinforced by previous data that showed inhibited statuses in both the antigen-specific proliferative response and polyclonal activators 28 in AD, suggesting a suppressed profile of CD4 + T cells in AD patients, corroborated by reduced cytokine secretion after stimulation.
Other studies have shown enhanced IL-22 levels in the CD4 + T cells of children and adults with AD following stimulation with α-toxin from S. aureus and SEB 39,40 . Our cohort of adults may display a different profile Interestingly, our data indicated that in vitro, staphylococcal enterotoxins exert opposite effects on T cell subsets, suppressing Th22 cells and enhancing Tc22 cells after SEA and SEB stimulation in the AD group. CD4 + IL22 + T cells appeared to be more susceptible than CD8 + IL-22 + T cells to SEA/SEB effects, regardless disease severity. However, similar frequencies of circulating Th22 and Tc22 cells after stimulation with PMA/Ionomycin have been described in AD and psoriasis 23 . Moreover, in severe AD, selective expansion of circulating Th2/Tc2 and Th22/Tc22 cells has been described in CLA + T cells induced by PMA/ionomycin 33,41 . Our results suggest that circulating Tc22 cells producing IL-22 contribute to the perpetuation of inflammation in AD patients, despite the In AD dermal lesions, we detected enhanced expression of IL-22; additionally, we identified both CD4 and CD8 T cells expressing IL-22 in AD skin, in accordance with previous findings 42, 43 . These authors found dermal cellular infiltrates of mainly CD4 + and CD8 + T cells, with a CD4:CD8 ratio similar to the profile detected in the peripheral blood in AD subjects 42,43 . Furthermore, high percentages of CD8 + T cells isolated from AD skin explant cultures are related to IL-22, IFN-γ, IL-13, and IL-17 production 44 , reinforcing the important role of these cells in barrier dysfunction and as a significant source of inflammatory cytokines in AD 23 .
We also described increased IL-22 and IL-4 transcript expression, emphasizing the Th2/T22 deviation in AD patients. Epicutaneous sensitization to house dust mites induces upregulation of IL-22 in an AD-like skin mouse model, leading to pruritus, intensified dermal and systemic Th2 immunity, downregulation of epidermal differentiation complex genes, and enhanced epidermal colonization of S. aureus 45 . In AD, decreases in the skin barrier function and Th2 cytokine release favor S. aureus penetration of the skin, with diminished in vitro expression of filaggrin and human β-defensin (HBD) 3 46,47 .
In conclusion, staphylococcal enterotoxins may play a role as a modulating factor on T lymphocyte-IL-22 secreting cells in AD patients, as evidenced by the presence of CD4/CD8 T cells expressing IL-22 in skin lesions, dysfunctional circulating Th22, and Tc22 cell upregulation.

Subjects.
Thirty-eight patients with AD (aged between 19-48 years; mean age: 27.68 ± 8.45; 26 males and 12 females) and 40 healthy non-AD volunteers (aged between 19-53 years; mean age: 31.10 ± 9.05; 18 males and 22 females) were enrolled in this study. AD was diagnosed according to the Hanifin & Rajka criteria 48 . Disease severity was evaluated by the EASI (Eczema Area and Severity Index) 49 , and AD patients were classified as mild (n = 9), moderate (n = 15), or severe (n = 14). IgE serum levels ranged from 3,120 to 119,000 IU/mL (average of 29,681). None of the patients were using oral steroids or immunosuppressants. All patients were interviewed about any associations with respiratory symptoms and the age of AD onset. This study was approved by the Ethics Committee of the University of Sao Paulo School of Medicine, and informed consent was obtained from all subjects. All methods were performed in accordance with the relevant guidelines and regulations of this institution. Demographic data are shown in Tables 1 and S1 (supplementary file).   Immunohistochemistry. Immunohistochemistry was performed on 4-μm paraffin-embedded samples, as described elsewhere 50 . The primary antibody IL-22 (ab134035, Abcam, Cambridge, UK) was utilized at 1:200 dilution, and the detection system was a Novolink Max Polymer Detection System (RE7280-K, Leica Biosystems, Newcastle Upon Tine, UK), and the chromogen used was DAB (3,3′ diamibenzidine, D5637, Sigma). Total tissue distribution of IL-22 was calculated by dividing the stained area by the total area measured in the epidermis or dermis. Immunohistochemically stained specimens were scanned using an Aperio Scan-scope Cs (Aperio Technologies, Vista, CA, USA). Photographs were analyzed utilizing Image-Pro Plus version 4.5.0.29 (Media Cybernetics Inc., Bethesda, Maryland, USA) 51 .

Real-time PCR.
Six-millimeter punch samples were taken from the skin lesions of AD patients with local anesthesia. Normal skin samples as controls were obtained after plastic surgery. All of the specimens were stored in RNAlater solution (Sigma), at −20 °C. Frozen samples were cut, and the tissue was homogenized with a Tissue Ruptor (Qiagen, Valencia, CA, USA) after debris removal by centrifugation, and supernatants were used for RNA extraction. Total RNA was extracted from each skin biopsy using an RNeasy Plus Mini Kit (Qiagen), including an extra step for separation of genomic DNA (gDNA eliminator column). The samples were then treated with a DNase set (Qiagen). Reverse transcription was performed with a Reverse Transcriptase Kit (Bio-Rad, Hercules, California, USA). For Real-time PCR was performed in an Applied Biosystems 7500 system using specific primers and SYBR Green (Applied Biosystems, Carlsbad, CA, USA), as described by Pereira et al., 52 . Primers for qPCR (Life Technologies) were only accepted if their efficiency reached 100 ± 10%. Corrections were made for primer efficiency. The specificity of the reaction was examined by dissociation curve. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA levels of the samples in the same plate were analyzed to normalize the mRNA contents among the tested samples. The cycling protocol consisted of 10 minutes at 95 °C, followed by 40 cycles of 15 seconds at 95 °C and 60 seconds at 60 °C. The amplification results were analyzed using Sequence Detection System (SDS) software (Applied Biosystems). Normalized expression was calculated as previously described by Livak 53 . Statistical analysis. The Mann-Whitney test or Kruskal-Wallis with Dunn's post-test were utilized to compare 2 or 3 sets of data, respectively. Correlations were established using the Spearman non-parametric correlation test. Differences between groups were considered statistically significant when p ≤ 0.05.