III. Results
2. Role of IL-17E in skin inflammation
“IL-17E (IL-25) enhances innate immune responses during skin inflammation”
Luisa Senra et al. 2018 (manuscript under revision in Journal of Investigative Dermatology)
Introduction
In the previous study, our group showed that IL-17E is overexpressed by psoriatic keratinocytes and it induces inflammatory cytokines and chemokines in macrophages, suggesting a possible pathogenic role in the pathogenesis of psoriasis [137]. The aim of this study was to assess the role of IL-17E in skin inflammation using in vivo models. Here, we show that IL-17E induces skin inflammation in vivo. Similar to psoriasis, we found that IL-17E is also up-regulated in the skin of mice models of cutaneous inflammation. IL-IL-17E neutralization or genetic deletion in these models ameliorates skin inflammation and reduces neutrophil infiltration, thus supporting a novel role of IL-17E, unrelated to the promotion of type 2 immune responses. Moreover, we suggest that IL-17E may be an important denominator of chronic skin inflammation, given that it is also up-regulated in other skin diseases characterized by neutrophil-rich infiltrate.
Contribution to this work
This study constitutes the second part of my thesis project. I designed, performed, analyzed the experiments and wrote the manuscript under the supervision of Prof. Wolf-Henning Boehncke and Dr. Nicolo Brembilla. Experiment with IL17EKO mice was performed in collaboration with the group of Prof. P.G. Fallon in Dublin. For experiments with tape stripping model, the advices and help of Dr. Curdin Conrad and his group were greatly appreciated.
IL-17E (IL-25) enhances innate immune responses during skin inflammation
Luisa Senra, MSc1, Alessio Mylonas, PhD2, Ruairi D. Kavanagh3, Padraic G. Fallon, PhD3, Curdin Conrad MD2,Julia Borowczyk-Michalowska, PhD1, Ludovic J Wrobel, PhD1, Guerkan Kaya, MD4, Nikhil Yawalkar, MD5, Wolf-Henning Boehncke, MD1,4, Nicolo Costantino Brembilla, PhD1
1Department of Pathology and Immunology, Faculty of Medicine, University of Geneva, Geneva, Switzerland
2Department of Dermatology and Venereology, University Hospital of Lausanne, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland
3Trinity Biomedical Sciences Institute, School of Medicine, Trinity College Dublin, Dublin 2, Ireland
4Divison of Dermatology and Venereology, Geneva University Hospitals, Geneva, Switzerland
5Department of Dermatology, Inselspital, Bern University Hospital, University of Bern, Switzerland.
CORRESPONDENCE:
Wolf-Henning Boehncke,
Division of Dermatology and Venereology,
Geneva University Hospitals and School of Medicine, Rue Gabrielle-Perret-Gentil 4,
1211 Geneva 14, Switzerland.
wolf-henning.boehncke@hcuge.ch
Nicolò Costantino Brembilla
Department of Pathology and Immunology Centre Médicale Universitaire
Rue Michel Servet 1 CH-1211 Genève 4 Switzerland
nicolo.brembilla@unige.ch
SHORT TITLE : IL-17E enhances innate immune responses during skin inflammation
1
1 ABBREVIATIONS: AD: atopic dermatitis; AGEP: acute generalized exanthematous pustulosis; IL: interleukin; MCP: monocyte chemoattractant protein; MPO: myeloperoxidase;
TNF: tumor necrosis factor; WT: wild-type;
ABSTRACT
IL-17E (IL-25) is a member of the IL-17 cytokine family involved in the promotion of type 2 immune responses. Recently, IL-17E has been reported to be upregulated in distinct skin inflammatory diseases such as psoriasis, atopic and contact dermatitis. We assessed the role played by IL-17E in skin inflammation. Here we show that IL-17E induced skin inflammation in vivo, characterized by the expression of innate immune response genes and the recruitment of innate immune cells, particularly neutrophils. Genetic deletion or IL-17E neutralization ameliorated skin inflammation induced by imiquimod application, or tape-stripping, with reductions in neutrophil and macrophage infiltration as assessed by t-SNE-guided multiparameter flow cytometry analysis, in mice. In man, IL-17E promotes the recruitment of neutrophils via activation of macrophages in a p38-dependent mechanism. In addition, IL-17E is up-regulated in neutrophil-rich inflammatory skin diseases, such as pyoderma gangrenosum and acute generalized exanthematous pustulosis. Our data demonstrate a novel role for IL-17E in skin inflammation, which is unrelated to the development of type 2 immune reactions.
We propose that IL-17E is an important common denominator of chronic skin inflammation promoting innate immune cell recruitment and activation.
KEYWORDS: IL-17E, IL-25, psoriasis, skin inflammation, neutrophils, imiquimod, tape-stripping
INTRODUCTION
The 17 cytokine family comprises 6 distinct isoforms: 17A-F. 17E, also called IL-25, is the most divergent member of the family with respect to sequence and function (Gaffen, 2009). Differently to IL-17A, IL-17E is not only produced by immune cells, but also by epithelial cells (Fort et al., 2001, Ikeda et al., 2003, Wang et al., 2007, Zaph et al., 2008).
IL-17E plays important roles in inflammation particularly occurring at gut and lung epithelial surfaces (Angkasekwinai et al., 2007, Ballantyne et al., 2007) (Caruso et al., 2009, Owyang et al., 2006, Wang et al., 2014). Unlike other members of the family, IL-17E is involved in the promotion of type 2 immune responses. In mice, IL-17E induces the production of Th2 cytokines, the recruitment of immune cells, including eosinophils, and epithelial cell hyperplasia (Fort et al., 2001, Kim M. R. et al., 2002, Pan et al., 2001). IL-17E contributes to allergic inflammation (Angkasekwinai et al., 2007, Ballantyne et al., 2007) and host protection against helminth infections (Angkasekwinai et al., 2013, Fallon et al., 2006, Owyang et al., 2006). Furthermore, IL-17E was shown to negatively regulate the development of Th17-mediated autoimmune diseases, such as experimental autoimmune encephalomyelitis (Kleinschek et al., 2007) and rheumatoid arthritis (Liu et al., 2016), by skewing the immune system towards a Th2 response.
Contrary to its relatively longstanding role in systemic inflammatory diseases, IL-17E was only recently implicated in skin inflammation. Recent reports showed an increased expression of IL-17E in distinct skin diseases: psoriasis (Senra et al., 2016), atopic dermatitis (AD) (Deleuran et al., 2012, Hvid et al., 2011) and contact dermatitis. In psoriasis, we recently showed that keratinocyte-derived IL-17E activates the production of pro-inflammatory mediators by dermal macrophages, leading to enhanced inflammation (Senra et al., 2016). In AD, IL-17E produced by dendritic cells (DC) inhibited filaggrin synthesis in keratinocytes, thereby affecting the skin barrier function (Hvid et al., 2011, Kim et al., 2013). To note,
psoriasis and AD have distinct pathogenesis that rely on Th17 and Th2 responses, respectively. In contact dermatitis, IL-17E led to enhanced Th17-, but not Th2 cell-mediated inflammation, through induction of IL-1β (Suto et al., 2018). Nevertheless, the precise role played by IL-17E in the context of skin inflammation remains unclear.
In the current study, we assessed the role of 17E in skin inflammation. We identified IL-17E as a key factor required for the recruitment of innate immune cells such as monocyte/macrophages and neutrophils to the skin, revealing a previously unknown pathogenic mechanism.
RESULTS
IL-17E promotes skin inflammation and favors the preferential recruitment of neutrophils in vivo.
To investigate the effects of IL-17E in skin inflammation, mice were intra-dermally injected with IL-17E in the back for up to 3 consecutive days. IL-17E injection induced immune cell infiltration already at day 1, that progressively increased at day 2 and 3 (Supplemental Figure 1A and Figure 1A). IL-17E induced a rather acute effect since interruption of injections quickly led to restoration of skin homeostasis at day 5 (Supplemental Figure 1A). We next profiled the IL-17E-induced skin inflammatory infiltrate by using 17-parameter flow cytometry analysis (gating strategy shown in Supplemental Figure 2). Consistent with the histological analysis, IL-17E injection induced a significant increase in the absolute number of CD45+ leucocytes infiltrating the skin (Figure 1E, left panel). This was mainly due to a skewing of the cell infiltrate towards the preferential recruitment of neutrophils to the detriment of T lymphocytes, particularly CD8 T cells (Figure 1E, right panel). No significant difference in the relative frequency of Ly6Chi and Ly6Clow monocyte/macrophages, dendritic cells, B cells, gd T cells, NK cells and NK T cells infiltrating the skin was found.
Consistently, neutrophils (Ly6G+) were found increased, and T-cells (CD3+) decreased by immunofluorescence analysis when normalized to either area of the region analysed or total number of nuclei (Supplemental Figure 1B and C).
In addition, IL-17E injection induced keratinocyte hyperproliferation, as determined by Ki-67 staining (Figure 1B and 1C), and significantly increased the epidermal thickness as compared to saline (Figure 1D).
Overall, these results indicate that IL-17E participates in skin inflammation in mice, particularly by promoting a preferential recruitment of neutrophils.
IL-17E induces the expression of neutrophil chemoattractant and innate immune response genes in vivo.
We next determined the gene levels of inflammatory mediators expressed in the skin of IL-17E treated mice. The mRNA levels of Tnf and Cxcl1 (murine homolog of human IL-8) were transiently increased in IL-17E- compared to saline-treated mice with a peak at 3h (Figure 2A). A wider panel of genes was profiled at this time-point by using a Nanostring assay testing 254 genes related to inflammation (full list of genes provided as Supplemental Table 1). We found a total of 8 upregulated and 1 downregulated gene transcripts (Figure 2B and Supplemental Table 2). Among the up-regulated genes, several are involved in neutrophil chemoattraction (Cxcl1, Cxcl10, Ccl20) and type I interferon signaling (Nod2, Ifit2 and Irf7).
Additional upregulated genes included Prostaglandin-Endoperoxide Synthase 2 (Ptgs2), a downstream target of NF-kB, and amphiregulin (Areg), which has been reported to induce keratinocyte proliferation and psoriasiform skin lesions in mice (Cook et al., 2004, Cook et al., 1997, Piepkorn et al., 1994). The expression of Masp2, a serine protease involved in complement system activation (Dobo et al., 2016) was instead found downregulated.
These data provide further evidence that IL-17E participates in skin inflammation in mice through induction of innate immune responses.
IL-17E contributes to the development of imiquimod-induced skin inflammation
We next investigated whether IL-17E plays a role in a model of skin inflammation. Kinetic analysis of imiquimod-treated skin revealed that IL-17E mRNA levels were promptly upregulated, with a peak at 1h post-treatment, and rapidly declined (Figure 3A). To further evaluate the role of IL-17E, we applied imiquimod on the ears of Il17e-/- mice for 5 consecutive days. Il17e-/- mice displayed markedly reduced ear swelling, erythema and scaling compared to wilde-type (WT) mice. In addition, pustules were smaller and body weight
variations were inferior in Il17e-/- mice compared to controls (Figure 3B, C and D). The reduced ear swelling corresponded to a decrease in both epidermal thickness and dermal swelling as assessed by blinded measurements of haematoxylin-eosin sections (Figure 3B and D). Overall, these results demonstrate that Il17e-/- mice are protected from imiquimod-induced inflammation.
IL-17E neutralization through blocking antibodies ameliorates imiquimod-induced skin inflammation and reduces innate immune cell infiltration.
Next, we examined whether IL-17E neutralization could prevent the induction of imiquimod-skin inflammation. Thus, we applied imiquimod on the shaved back skin and an ear of WT mice intra-peritoneally injected with neutralizing α-IL17E or irrelevant isotype antibodies.
Mice injected with α-IL17E antibody showed reduced weight loss, decreased ear swelling (Figure 4A) and amelioration of skin inflammation. The epidermal thickness was also significantly reduced by the treatment, although it could not completely prevent the imiquimod-induced acanthosis (Figure 4B).
We further analysed the infiltrate by flow cytometry, as described earlier. The total numbers of CD45+ leucocytes were significantly reduced in the α-IL17E treated group (Figure 4C).
This reflected a general reduction in the absolute numbers of multiple innate immune cells, namely neutrophils, eosinophils, macrophages (ly6Chi) and NK cells (Figure 4D). To further characterize the immune cell infiltrate, we performed population clustering based on t-Distribution Stochastic Neighbor Embedding (t-SNE) analysis. We used t-SNE to reduce 15 parameters into two dimensions (t-SNE1 and t-SNE2). Based on this unbiased projection, we manually identified 13 distinct cell clusters (Figure 5A). For each cluster, the median intensity of each marker was quantified and summarized by using a heat-map illustration (Figure 5B).
Based on reported phenotypes, the 13 clusters were segregated into 5 distinct groups,
corresponding to i) NK cells (cluster 1 to 3), ii) lymphocytes (cluster 4 to 6), iii) eosinophils (cluster 7), iv) monocyte/macrophages (cluster 8) and v) neutrophils (clusters 10 to 13). A cell population (cluster 9) expressing characteristic markers of both neutrophils and eosinophils (Ly6G+, SiglecF+) was also identified. The NK cell group of clusters included mature CD49b+CD11bhiCD45hi NK cells (cluster 1), immature CD49b+CD11blowCD45low NK cells (cluster 2) and CD49b+CD3+ NK T cells (cluster 3). Treatment with α-IL17E elicited the reduction of immature NK cells and the augmentation of NK T cells (Figure 5B and 6C). The lymphocyte group of clusters was not modified by IL-17E neutralization. This included CD3+CD4+ T cells (cluster 4), CD3+CD8+ T cells (cluster 5) and CD3-CD19+ B cells (cluster 6). Treatment with α-IL17E antibody resulted instead in the disappearance of CD11b+F4/80+Ly6C+SiglecF+ eosinophils (cluster 7) and CD11b+F4/80+Ly6C+ monocyte/macrophages (cluster 8), consistent with a preferential role of IL-17E in the recruitment of these subsets (Figure 5A and B). In addition, neutralization of IL-17E induced a skewing of neutrophils from a “classical” phenotype (cluster 11 and 12) to cells characterized by higher expression levels of F4/80 (cluster 13) (Figure 5B and D).
To expand our results, we assessed the role of IL-17E in a distinct model of skin inflammation relying on tape-stripping. As observed in the case of imiquimod, IL-17E mRNA levels were induced by tape-stripping and peaked as early as 1h after treatment (Supplemental Figure 3A). Preventive IL-17E neutralization resulted in a specific and significant decrease in the recruitment of CD45+ leucocytes, neutrophils and Ly6Chi monocytes/macrophages, but not T and B lymphocytes (Supplemental Figure 3B).
Together, these data highlight that IL-17E participates in innate cell recruitment in murine skin inflammation.
IL-17E promotes the recruitment of neutrophils via activation of macrophages in a p38- dependent mechanism in man.
We have recently shown that the number of IL-17E+ cells correlates with the number of neutrophils in psoriasis (Senra et al., 2016, Xu et al., 2018). We therefore investigated the mechanisms by which IL-17E may promote the recruitment of neutrophils using human primary cells. IL-17E was reported to target skin macrophages, particularly the M2 phenotype, and keratinocytes (Senra et al., 2016, Xu et al., 2018). M2 macrophages responded to IL-17E by producing inflammatory mediators, including IL-8, TNF-α, IL-6 and MCP-1 (Figure 6B). Conversely, human keratinocytes were not capable of producing 8 upon IL-17E treatment (Figure 6E). In macrophages, IL-IL-17E stimulation resulted in NF-kB and p38 activation, as indicated by their time-dependent phosphorylation, whereas it failed to induce ERK1/2 activation (Figure 6A). The levels of IL-8 and IL-6 mRNA induced by IL-17E stimulation were decreased in presence of a p38 MAPK inhibitor (SB202190), but not in the presence of NF-kB inhibitor (JSH23), whereas simultaneous addition of both inhibitors had similar effects as the p38 inhibitor alone (Figure 6B). Moreover, IL-17E-mediated TNF-α production decreased in presence of p38 and NF-kB inhibitors, with similar effects observed when both pathways were inhibited (Figure 6B). Of interest, MCP-1 levels were not affected by the use of either p38 or NF-kB inhibitors. These results indicate that MAPK p38 pathway is required for the induction of neutrophil chemokines (IL-8) and pro-inflammatory molecules (TNF-α and IL-6) in M2 macrophages.
We next assessed whether IL-17E may be implicated in neutrophil recruitment via macrophage-dependent IL-8. Contrary to IL-8, IL-17E was not able to promote neutrophil recruitment on its own (Figure 6C). However, supernatants of IL-17E-stimulated macrophages increased neutrophil migration compared to supernatants of control macrophages. This effect was blocked in the presence of a saturating dose of neutralizing
anti-IL-8 antibody or when supernatants of IL-17E-stimulated macrophages cultured in the presence of a p38 inhibior were used (Figure 6C). Consistently, supernatants of IL-17E-stimulated macrophages contained higher levels of IL-8 compared to non-IL-17E-stimulated supernatants, and levels were reduced in presence of p38, but not NF-kB inhibitor (Figure 6D).
These findings suggest that IL-17E favours the recruitment of neutrophils in a mechanism dependent on macrophage-derived IL-8.
IL-17E+ cell number predicts neutrophil infiltration in human skin inflammatory conditions.
We next investigated whether the ability of IL-17E to recruit neutrophils, which we initially observed in psoriasis (Senra et al., 2016) and confirmed here in mouse models for psoriasis, applies to human inflammatory skin diseases at large. We therefore analysed IL-17E expression in two pathologies characterized by an inflammatory infiltrate rich in neutrophils, namely pyoderma gangrenosum and acute generalized exanthematous pustulosis (AGEP). In both diseases, but not in healthy skin, an increased number of IL-17E+ cells was observed in the dermis, concomitantly with augmented MPO+ neutrophils (Figure 7A and B). To note, the expression of IL-17E was also significantly upregulated in the epidermis of both neutrophilic diseases (Figure 7C). These data strongly suggest that IL-17E plays a role in neutrophil recruitment in skin inflammation.
DISCUSSION
IL-17E is considered a type 2 cytokine, as revealed by works performed in allergy and gut inflammation (Angkasekwinai et al., 2007, Ballantyne et al., 2007, Caruso et al., 2009, Reynolds et al., 2015). Recent studies have however challenged this assumption when referring to the skin. IL-17E was found similarly overexpressed in skin diseases characterized by a Th2 signature, i.e. AD, or Th17 signature, i.e. psoriasis (Hvid et al., 2011, Senra et al., 2016, Suto et al., 2018). These observations suggest a role for IL-17E in skin inflammation unrelated to the promotion of type 2 responses. Here, we show that IL-17E is critical for the recruitment of innate immune cells during skin inflammation, particularly neutrophils and macrophages. In addition, we identified that activation of p38 and consequent release of IL-8 by IL-17E-stimulated macrophages is a critical step for neutrophil recruitment.
Our data suggest that the ability of IL-17E to recruit innate immune cells applies broadly to skin inflammation. IL-17E neutralization led to inhibition of neutrophil and macrophage recruitment in two distinct models of skin inflammation: one dependent on immune activation via TLR7 stimulation (imiquimod) (van der Fits et al., 2009); the other characterized by the induction of an inflammatory response secondary to an epidermal challenge (tape-stripping) (Gregorio et al., 2010, Jin et al., 2009). Consistently, a recent report identified IL-17E as a key player in the recruitment of skin cells upon imiquimod application (Xu et al., 2018). To note, imiquimod and tape stripping-induced inflammation do not match every aspect of any particular disease, bur rather model several skin conditions (Conrad et al., 2018, Jin et al., 2009, Sano et al., 2005, Swindell et al., 2017). Consistent with our results, IL-17E expression in psoriasis positively correlated with the number of infiltrating neutrophils (Senra et al., 2016). In addition, though being clearly less pronounced than in psoriasis, AD and contact dermatitis lesional skin also present increased neutrophils (Dhingra et al., 2013). Here we provide new evidence that other skin diseases characterized by a neutrophil-rich inflammatory
infiltrate, such as AGEP and pyoderma gangrenosum, share upregulated IL-17E expression.
The cells and mechanisms responsible for IL-17E production remain partially unknown. We reported increased IL-17E mRNA levels in the epidermis of psoriatic skin (Senra et al., 2016).
Consistently, keratinocytes were shown to produce IL-17E when cultured with IL-17A, at least in mice (Xu et al., 2018), and upon stimulation with ragweed (Wang et al., 2007). The induction of IL-17E following tape-stripping is another indication that the source of IL-17E during skin inflammation is the epidermis. Furthermore, we show a rapid up-regulation of IL-17E upon imiquimod application. Imiquimod (Aldara) has been reported to directly affect keratinocytes, via inflammasome activation or through adenosin receptors (Li et al., 2013, Walter et al., 2013). Alternatively, imiquimod activates immune cells, as DC, via TLR7.
Along this line, IL-17E-producing DC were recently described in the skin of AD patients (Hvid et al., 2011).
IL-17E administration results in the up-regulation of several genes involved in chemotaxis, including Cxcl1, Cxcl10 and Ccl20. This finding was paralleled by the preferential recruitment of innate cells. To note, CXCL1 is a well-described chemokine for neutrophils and CCL20, although implicated in T cell recruitment, has also been shown to function as neutrophil-recruiting chemokine (Schutyser et al., 2003). CXCL10 favors the recruitment of monocyte/macrophages and T cells, although to a lesser extent (Luster and Leder, 1993). In addition, an up-regulation of genes involved in type-I interferon response (Nod2, Ifit2 and Irf7) and NF-kB signaling (Ptsg2) was identified, underlying that IL-17E principally activates an innate immune response. Innate responses and type-I interferon signature are both implicated during early phases of psoriasis and other skin diseases (Cook et al., 1997, Nestle et al., 2005, Piepkorn, 1996), pointing to a possible role of IL-17E in the onset rather than the disease maintenance.
We used high-dimensional mapping of the immune cell infiltrate coupled with t-SNE-guided analysis to determine cell types strictly relying on IL-17E for recruitment to the skin in an unbiased manner (Simoni et al., 2017). We identified a previously unknown population (unchanged upon IL-17E neutralization) displaying phenotypic characteristics of neutrophils (CD11bhiLy6G+) and eosinophils (SiglecF+). The function of such population remains to be determined. Neutralization of IL-17E resulted in reduction of CD45lowCD11blow NK cells, a phenotype consistent with more immature cells (Kim S. et al., 2002, Krzywinska et al., 2016), whereas induced NK T cells. Whether IL-17E contributes to the recruitment of NK cells is unknown, although a similar function has been reported for other IL-17 family members (O'Sullivan et al., 2014). A remarkable skew in the phenotype of neutrophils has also been observed. Our analysis revealed four different neutrophil (CD11b+Ly6G+) phenotypes:
CD11bbrightCD49bhiSiglecFhi (cluster 10); Ly6C-F4/80- cells expressing higher levels of CD11c (cluster 11); and two subsets differing for the expression of F4/80 (cluster 12 and 13).
An increased expression of CD11c linked to inhibition of T-cell responses has been recently
An increased expression of CD11c linked to inhibition of T-cell responses has been recently