The University of Manchester Research

IL-36 Promotes Systemic IFN-I Responses in Severe Forms of

DOI: 10.1016/j.jid.2019.08.444

Document Version Accepted author manuscript

Link to publication record in Manchester Research Explorer

Citation for published version (APA): Catapano, M., Vergnano, M., Romano, M., Mahil, S. K., Choon, S., Burden, A. D., Young, H. S., Carr, I. M., Lachmann, H. J., Lombardi, G., Smith, C. H., Ciccarelli, F. D., Barker, J. N., & Capon, F. (2019). IL-36 Promotes Systemic IFN-I Responses in Severe Forms of Psoriasis. Journal of Investigative Dermatology. https://doi.org/10.1016/j.jid.2019.08.444 Published in: Journal of Investigative Dermatology

Citing this paper Please note that where the full-text provided on Manchester Research Explorer is the Author Accepted Manuscript or Proof version this may differ from the final Published version. If citing, it is advised that you check and use the publisher's definitive version.

General rights Copyright and moral rights for the publications made accessible in the Research Explorer are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

Takedown policy If you believe that this document breaches copyright please refer to the University of Manchester’s Takedown Procedures [http://man.ac.uk/04Y6Bo] or contact [email protected] providing relevant details, so we can investigate your claim.

Download date:03. Oct. 2021 Journal of Investigative Dermatology

Interleukin-36 promotes systemic Type-I IFN responses in severe forms of psoriasis

Journal: Journal of Investigative Dermatology

Manuscript ID JID-2019-0205.R1 Article Type:ForOriginal Review Article Only Date Submitted by the 14-Jun-2019 Author:

Complete List of Authors: Catapano, Marika; King's College London, Division of Genetics and Molecular Medicine Vergnano, Marta; King's College London, Division of Genetics and Molecular Medicine Romano, Marco; King's College London, Department of Immunobiology, School of Immunology & Microbial Sciences Mahil, Satveer; King’s College London, Department of Medical and Molecular Genetics Choon, Siew-Eng; Hospital Sultanah Aminah, Department of Dermatology Burden, David; University of Glasgow, Department of Dermatology Young, Helen; The University of Manchester, Dermatology Research Centre Carr, Ian; University of Leeds, School of Medicine Lachmann, Helen; University College London, National Amyloidosis Centre and Centre for Acute Phase , Division of Medicine Lombardi, Giovanna; King's College London, MRC Centre for Transplantation Smith, Catherine; Kings College London, St John's Institute of Dermatology Ciccarelli, Francesca; Francis Crick Institute, Cancer Systems Biology Laboratory Barker, Jonathan; St John's Institute of Dermatology, St Thomas' Hospital, Division of Genetics and Moilecular Medicine Capon, Francesca; King's College London, Division of Genetics and Molecular Medicine

Autoinflammatory Diseases, , Inflammatory Skin Diseases, Keywords: Neutrophils, Psoriasis

Page 1 of 62 Journal of Investigative Dermatology

Head of School Department of 1 King’s College London Faculty of Life Sciences Professor Mathias Gautel Medical & 2 Molecular Genetics 3 & Medicine Head of Department Floor 8 Tower Wing 4 Professor Tim J P Hubbard Guy’s Hospital 5 School of Basic and London SE1 9RT 6 Medical Biosciences 7 8 http://www.kcl.ac.uk/gm 9 Please reply to: Francesca Capon 10 mReader in Genetics 11 e-mail: [email protected] 12 13 14 15 16 17 14th June 2019 18 19 20 21 Prof Mark C Udey, For Review Only 22 Editor, Journal of Investigative Dermatology 23 24 25 Dear Prof Udey, 26 Re: MS# JID-2019-0205 27 28 We are submitting revised version of the above manuscript entitled “-36 promotes 29 30 systemic Type-I IFN responses in severe psoriasis “. 31 32 To address the reviewers’ comments, we have undertaken the analysis of additional datasets, 33 provided more information on our patient resources and included further experimental data on IL- 34 36R expression. We believe that the manuscript, which has also been edited for clarity, is much 35 36 improved and will now meet the criteria for publication on the Journal of Investigative Dermatology. 37 38 39 I look forward to hearing from you and remain 40 41 Yours sincerely 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

www.kcl.ac.uk Journal of Investigative Dermatology Page 2 of 62

1 2 3 RESPONSE TO REVIEWERS 4 5 Reviewer 1 6 The data is reliable… however, there are too many miss spelling in the text. 7 8 We have carefully proofread the text and eliminated typos as well as instances of inconsistent spelling 9 One concern is as if the author checked IL-36RN mutation or IL-1RN mutation in GPP patients or not. 10 Based on the diagnosis of GPP, we have screened all patients for IL36RN mutations. This has now been 11 st 12 clarified in the Methods (p.11, 1 paragraph). 13 Add the comment that treatment with IL-17 antibody or IL-17R inhibitor will decrease skin 14 15 inflammation and production of IL-36 from , leading to the down regulation of Type-I IFN. 16 The comment has been added to the Discussion (p.10, 3rd paragraph). 17 18 Reviewer 2 19 1) It is questionable if any of the findings described is specific for IL-36 as compared to other pro- 20 21 inflammatory cytokines andFor in particular Review members of the Only IL-1 family. Importantly Swindell et al (Front 22 Immunol. 2018 Jan 29; 9:80) have demonstrated that IL1 and IL-36 signalling is essentially the same. 23 It is likely that all effects shown here would be equally inducible by other IL-1 family members some of 24 which are also upregulated in psoriatic inflammation. 25 We have revised the Discussion (p.8, 2nd paragraph), to quote the important work of Swindell et al and 26 27 acknowledge the similarities between IL-1 and IL-36 signalling. At the same time, we note that Swindell 28 did not identify any of our signature among the transcripts induced by IL-1. In keeping 29 with this observation, we find that no IFN signature genes are up-regulated in the whole-blood of 30 patients with CAPS, a condition caused by excessive IL-1 signalling. This pattern was observed in two 31 independent datasets, the second of which was analysed as part of the revision process. The result of 32 33 this new analysis are reported in Supplementary Figure S1. 34 2) A significant body of literature highlights the role of LL-37 in psoriatic inflammation being an 35 36 important contributor to type I IFN responses also in pustular phenotype development. LL-37 is neither 37 mentioned nor included in experimental setups in this paper and should be considered to gain a more 38 balanced information on which factors contribute to clinically relevant IFN expression. Conrad et 39 al (Nat Commun. 2018 Jan 2;9(1):25) and other publications by Gilliet group. 40 The reviewer makes a very interesting point, which we have initially addressed by querying the 41 42 expression of CAMP (the encoding LL-37) in our transcriptomic datasets. This revealed that CAMP 43 is up-regulated in the skin of psoriatic patients (fold change >2.0, FDR <0.05), but not in their blood. 44 In keeping with the latter observation, real-time PCR showed that IL-36 treatment does not increase 45 CAMP expression in PBMCs. 46 47 LL37 48 PBMCs were stimulated with IL-36 49 3 50 for the indicated time points. Data 51 2 represent the mean (+/- SEM) of 52 triplicate stimulations 53 1 54 0 55 0h 3h 6h 8h 24h 56 57 Finally, we observed that CAMP whole-blood expression does not correlate with the up-regulation of 58 Type-I IFN genes in either GPP or PsV samples (r<0.1 in both datasets). This suggests that LL-37 is 59 unlikely to be driving the Type-I IFN responses we observed at the systemic level. 60

1 Page 3 of 62 Journal of Investigative Dermatology

1 2 3 We have now edited the Discussion (p.9, final paragraph) to mention these negative findings and 4 quote the work of the Gilliet group, which provides the underlying context. 5 6 3) The clinical relevance of an increase IFN signature (which clearly overlaps with IFNg signature – the 7 latter is known to be highly expressed in psoriasis) remains to be demonstrated. To further explore the 8 9 significance of the induced IFN response a comparison to a disease recognised for its type I IFN pathway 10 activation such as systemic should be included. Some comparison with other diseases is based 11 only on literature data. Figure 1F as it stands gives too little information to demonstrated convincing 12 evidence. Has the same cell population been analysed, was treatment given to the patients, are patient 13 groups regarding to basic demographics comparable etc. 14 15 First of all, we have edited the Methods (RNA-seq data analysis, p.11) to clarify that we did not base 16 the comparisons with other diseases on gene lists published in the literature. On the contrary, we 17 retrieved raw sequence files for every dataset and processed them with the same computational 18 pipeline that was used for the QC and differential expression analysis of our own data. Thus, the up- 19 regulated genes reported in Figures 1f and S1 were identified through a robust, standardised process. 20 21 To fully document these analyses,For we Review have also added a Onlysupplementary Table S4, which summarises 22 the key features of the publicly available datasets and shows that they were all generated in whole- 23 blood samples obtained from European individuals. 24 Having clarified in the text (p.4, final paragraph) that IFN-pathies are caused by abnormal activation 25 of Type-I IFN responses, we have validated the findings presented in Figure 1f through the analysis of 26 27 a second dataset (the results are shown in the revised Figure). Finally, we note in the Discussion (p.8 28 final paragraph) that IFN signature genes such as IFI6 and OASL are also up-regulated in SLE. 29 30 4) Disease controls: It would be really important to include information on disease activity and 31 potential immunoregulatory systemic treatment regarding the patients included in the study. In 32 particular: CAPS blood during flare? / on aIL-1 therapy; APP: active disease when blood was taken? 33 GPP: active disease when blood was taken?; Obviously psoriasis patient had active disease when 34 analysed and this MUST be stated for the disease controls as well! 35 36 We agree these are important details and have included them in a new supplementary Table (Table 37 S3). This shows that most patients had active disease when the blood was taken. 38 39 5) How did author select these 5 IFN regulated genes for their score? 40 We have now clarified (p.5, 1st paragraph) that the five genes were selected based on two criteria: i) 41 evidence for over-expression in our GPP resource and ii) annotation from the interferome database, 42 indicating that the gene is up-regulated by Type-I IFN. Of note all five genes have been previously used 43 to calculate IFN scores or establish Type-I IFN gene signatures (e.g. Pfeffer et al PLoS ONE 2014; Der 44 45 et al Nat Immunology 2019). 46 6) GPP cases: could the authors comment why – when having access to neutrophils/whole blood 47 48 samples from 17 GPP cases (page 8) only 8 or 9 are included in Figure 1g and Fig 2e analysis? 49 We have clarified (p.6, 1st paragraph) that the additional GPP cases were recruited after the RNA- 50 sequencing experiment was completed. 51 52 7) Fig 2: it would be meaningful to show IL-36R expression as fluorescence intensity compared to 53 isotype and not as percent positive cells. Receptor expression in skin infiltrates would be of interest. 54 To avoid any ambiguity, we have edited the text of p.6 to clarify that the data reported in Figure 4 55 + 56 relates to the number of IL36R cells rather than the expression of the receptor. To complement these 57 findings, we have added an additional Figure (Supplementary Figure S2), showing IL36R mean 58 fluorescence intensity in the various leukocyte populations. This confirms that the receptor expression 59 is highest in mDC and pDC. 60

2 Journal of Investigative Dermatology Page 4 of 62

1 2 3 The presence of the receptor on skin resident immune cells has been thoroughly investigated by the 4 group of Cem Gabay (Dietrich et al, 2016), who reported robust IL36R expression in dendritic cells. We 5 6 have amended the text of the Discussion (p.9) to mention this work and note that the results are 7 consistent with our own findings. 8 9 8) Figure 5 should include IL-1a/b, TNF, IL-18 10 As requested by the reviewer, we have examined the expression of IL-1b, TNF and IL-18 in the PBMCs 11 12 we had treated with CpG and/or IL-36. As shown in the plots below, we found that TNF and IL-18 are 13 not induced by either stimulus (left panel). While IL1b is up-regulated by CpG and to a greater extent 14 by IL-36, we did not observe a synergistic effect when the two agents were combined (right). We 15 concluded that IL-36 does not potentiate CpG induced production of IL1b, TNF or IL-18. We have 16 17 therefore not included this data in the revised manuscript. 18 19 20 21 For Review Only 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Minor points 36 37 1) pDC have been discussed to be of relevance in diseases but potentially less so in established disease. 38 Haensel et al (J Allergy Clin Immunol. 2011; 127:787-94 e1-9) have highlighted slanDCs to be increased 39 in the skin compartment. Authors could include some comments on APC subtypes in psoriatic skin 40 We have edited the Discussion (p.9, 2nd paragraph) to mention the accumulation of slanDC and pDCs 41 within psoriatic skin lesions. We have quoted the work of Hansel et al, alongside that of Nestle and 42 43 Gilliet. 44 2) Page 5 abnormal systemic IFN responses – Gilliet et al have proposed this in previous papers which 45 46 should be acknowledged. 47 We have acknowledged an important study of the Gilliet group, which showed that Type-I IFN 48 production by pDCs contributes to systemic autoimmunity (p.9, final paragraph). 49 50 3) Results /Figure 1 – what is described as IL-36 signature genes seems not specific for IL-36; this 51 induction can be seen with other cytokines including IL-1a/b, TNF, IL-18. 52 We have edited the first paragraph of the results (p.4) and replaced the expression “IL-36 signature 53 54 genes” with “genes that can be up-regulated by IL-36”, to reflect the fact that some of these transcripts 55 can also be induced by other cytokines. 56 57 58 59 60

3 Page 5 of 62 Journal of Investigative Dermatology

1 2 3 Interleukin-36 promotes systemic Type-I IFN responses in severe forms of psoriasis 4 5 Marika Catapano1,11, Marta Vergnano1,11, Marco Romano2, Satveer K Mahil3, Siew-Eng Choon4, A 6 7 David Burden5, Helen S Young6, Ian M Carr7, Helen J Lachmann8, Giovanna Lombardi2, Catherine H 8

9 3 9,10 3 1 10 Smith , Francesca D. Ciccarelli , Jonathan N Barker , Francesca Capon *. 11 12 13 14 M. Catapano: 0000-0003-2344-6067, M. Vergnano: 0000-0003-4654-5519, M. Romano: 0000-0001- 15 16 6089-5828, S. K. Mahil: 0000-0003-4692-3794, S.E. Choon: 0000-0002-7796-5746, A. D. Burden: 17 18 0000-0001-7395-9931, H. S. Young: 0000-0003-1538-445X, I. M. Carr: 0000-0001-9544-1068, H. J. 19 20 Lachmann: 0000-0001-8378-2498, G. Lombardi: 0000-0003-4496-3215, C. H. Smith: 0000-0001- 21 For Review Only 22 9918-1144, F. D. Ciccarelli: 0000-0002-9325-0900, J. N. Barker: 0000-0002-9030-183X, F. Capon: 23 24 0000-0003-2432-5793 25 26 27 28 1Department of Medical and Molecular Genetics, School of Basic & Medical Biosciences, King’s 29 30 2 31 College London, London, UK; Department of Immunobiology, School of Immunology & Microbial 32 3 33 Sciences, King’s College London, London, UK; St John’s Institute of Dermatology; School of Basic 34 35 & Medical Biosciences, King’s College London, London, UK; 4Department of Dermatology, Sultanah 36 37 Aminah Hospital, Johor Bahru, Malaysia; 5Department of Dermatology, University of Glasgow, 38 39 Glasgow, UK; 6Division of Musculoskeletal and Dermatological sciences, University of Manchester, 40 41 Manchester, UK; 7School of Medicine, University of Leeds, Leeds, UK; 8National Amyloidosis Centre 42 43 and Centre for Acute Phase Proteins, Division of Medicine, University College London, London, UK; 44 45 9Cancer Systems Biology Laboratory, The Francis Crick Institute, London, UK; 10School of Cancer & 46 47 Pharmaceutical Sciences, King’s College London, London, UK. 48 49 50 51 11 52 These authors contributed equally. *Corresponding author: Francesca Capon, Department of Medical 53 54 and Molecular Genetics, School of Basic and Medical Biosciences, King’s College London, 9th floor 55 56 Tower Wing Guy’s Hospital, London SE1 9RT, United Kingdom. E-mail: [email protected] 57 58 59 60 Short title: A new IL-36/Type I IFN axis in psoriasis

1 Journal of Investigative Dermatology Page 6 of 62

1 2 3 ABSTRACT 4 5 Psoriasis is an immune-mediated skin disorder associated with severe systemic co-morbidities. While 6 7 IL-36 is a key disease driver, the pathogenic role of this has mainly been investigated in skin. 8 9 10 Thus, its effects on systemic immunity and extra-cutaneous disease manifestations remain poorly 11 12 understood. 13 14 To address this issue, we investigated the consequences of excessive IL-36 activity in circulating 15 16 immune cells. We initially focused our attention on generalised pustular psoriasis (GPP), a clinical 17 18 variant associated with pervasive up-regulation of IL-36 signalling. By undertaking blood and 19 20 neutrophil RNA-sequencing, we demonstrated that affected individuals display a prominent Type-I IFN 21 For Review Only 22 signature, which correlates with abnormal IL-36 activity. We then validated the association between 23 24 IL-36 de-regulation and Type-I IFN over-expression in patients with severe psoriasis vulgaris (PsV). 25 26 We also found that the activation of Type-I IFN genes was associated with extra-cutaneous morbidity, 27 28 in both GPP and PsV. Finally, we undertook mechanistic experiments, demonstrating that IL-36 acts 29 30 31 directly on plasmacytoid dendritic cells (pDCs), where it potentiates Toll-like Receptor (TLR)-9 32 33 activation and IFN-α production. This effect was mediated by the up-regulation of PLSCR1, a 34 35 phospholipid scramblase mediating endosomal TLR-9 translocation. 36 37 These findings identify a new IL-36/Type-I IFN axis contributing to extra-cutaneous inflammation in 38 39 psoriasis. 40 41 42 43 Key words: 44 45 Generalized pustular psoriasis, psoriasis vulgaris, systemic inflammation, IL-36, Type-I IFN, PLSCR1 46 47 48 49 Abbreviations: CAPS, cryopyrin associated periodic syndrome; DEG, differentially expressed genes; 50 51 FC, fold change; GPP, generalised pustular psoriasis; IFN, interferon; IL-36, interleukin-36; IL36R, IL- 52 53 36 receptor; IL36RN: IL-36 receptor antagonist; MAPK, mitogen-activated kinases; pDCs, 54 55 plasmacytoid dendritic cells; PLSCR1, Phospholipid Scramblase 1; PsV: psoriasis vulgaris; TLR- 9: 56 57 Toll-like receptor- 9 58 59 60

2 Page 7 of 62 Journal of Investigative Dermatology

1 2 3 INTRODUCTION 4 5 Interleukin-36α, -β and -γ (hence IL-36) are group of IL-1 family cytokines that are mainly produced 6 7 by keratinocytes, monocytes and dendritic cells (Bassoy et al., 2018). IL-36 signalling plays an 8 9 10 important role in epithelial immune homeostasis and its de-regulation has been repeatedly implicated 11 12 in the pathogenesis of psoriasis vulgaris (PsV), a common and chronic, immune-mediated skin disorder 13 14 (Bassoy et al., 2018). 15 16 Numerous studies have shown that IL-36 responses are elevated in PsV skin (Mahil et al., 2017; 17 18 Quaranta et al., 2014; Swindell et al., 2015) where they stimulate production and amplify 19 20 the effects of IL-17 signalling (Mahil et al., 2017). Animal studies have also demonstrated that IL-36 21 For Review Only 22 promotes the activation of dendritic cells and the polarization of T lymphocytes into Th17 cells (Tortola 23 24 et al., 2012). Thus, the mechanisms whereby IL-36 contributes to cutaneous inflammation have been 25 26 extensively investigated. Its effects on circulating leukocytes, however, remain poorly understood. 27 28 We and others have shown that recessive mutations of the IL-36 receptor antagonist (IL36RN) are 29 30 31 associated with generalised pustular psoriasis (GPP), a disease variant characterized by severe extra- 32 33 cutaneous symptoms (Marrakchi et al., 2011; Onoufriadis et al., 2011). In fact, GPP patients suffer 34 35 from flares of skin pustulation that are often accompanied by systemic upset (fever, elevation of acute 36 37 phase reactants and neutrophilia) (Burden and Kirby, 2016). This suggests that IL-36 signalling is likely 38 39 to influence immune responses beyond skin. 40 41 Extra-cutaneous co-morbidities are also well documented in PsV, as individuals suffering from severe 42 43 disease are at high risk of , metabolic syndrome and atherosclerosis (Burden and Kirby, 44 45 2016; Fang et al., 2016; Shah et al., 2017). It has therefore been proposed that PsV is a systemic disease, 46 47 manifesting with skin, joint and vascular inflammation (Davidovici et al., 2010; Reich, 2012). 48 49 50 In this context, we hypothesise that abnormal IL-36 signalling has extra-cutaneous effects in both GPP 51 52 and PsV, driving acute systemic flares in the former and contributing to a state of chronic systemic 53 54 inflammation in the latter. To explore this model, we integrated the transcription profiling of patient 55 56 leukocytes with ex-vivo IL-36 stimulations. We show that IL-36 potentiates Toll-like receptor (TLR)- 57 58 9 activation and enhances the production of Type-I IFN, a cytokine that contributes to systemic 59 60 immunity, arthritis and atherosclerosis.

3 Journal of Investigative Dermatology Page 8 of 62

1 2 3 4 5 RESULTS 6 7 Expression profiling identifies a Type-I IFN signature in GPP and PsV whole-blood samples 8 9 10 We reasoned that GPP would represent an ideal model in which to investigate the systemic effects of 11 12 IL-36, given the well-established link with IL36RN mutations (Marrakchi et al., 2011; Onoufriadis et 13 14 al., 2011) and enhanced IL-36 activity (Johnston et al., 2016). We therefore undertook whole-blood 15 16 RNA-sequencing in 9 affected individuals and 7 healthy controls (Supplementary Table S1a). While 17 18 the deconvolution of transcription profiles showed that leukocyte frequencies were comparable in cases 19 20 vs. controls (Supplementary Table S1b), differential expression analysis identified 111 genes that were 21 For Review Only 22 over-expressed (fold change≥1.5; FDR< 0.05) in patients (Figure 1a, Supplementary Table S2a). As 23 24 expected, genes that can be induced by IL-36 (IL1B, PI3, VNN2, TNFAIP6, SERPINB1) were 25 26 collectively up-regulated in cases vs. controls (P=0.019) (Figure 1b). Of note, the analysis of a publicly 27 28 available PsV dataset (Wang et al., 2014) identified a moderate, but statistically significant, over- 29 30 31 expression of the same genes in patient whole-blood (P=0.001) (Figure 1b), providing the first 32 33 indication that IL-36 may have systemic effects in PsV. 34 35 To further explore the biological significance of our findings, we mapped the genes up-regulated in 36 37 GPP to the blood co-expression modules described by Li et al (Li et al., 2014). We found that the over- 38 39 expressed genes were significantly enriched among modules related to innate immune activation (e.g. 40 41 enriched in activated dendritic cells, FDR<0.005) and antiviral responses (e.g. type I IFN response; 42 43 FDR<0.05) (Figure 1c). These findings were validated by Ingenuity Pathway Analysis (IPA), which 44 45 identified interferon signalling as the most significantly enriched pathway (FDR<5x10-6) (Figure 1d). 46 47 An upstream regulator analysis also highlighted IRF7, STAT1 and STAT3 as the transcriptional 48 49 -10 50 activators that are most strongly associated with gene over-expression (FDR<10 for all) (Figure 1e, 51 52 Supplementary Table S2c). This is of interest since proteins are critical mediators of IFN signal 53 54 transduction and IFN-α production by pDCs (Honda et al., 2005). 55 56 Finally, the analysis of two publicly available datasets (Liu et al., 2012; Rodero et al., 2017) 57 58 demonstrated a significant overlap (P<10-10) between the genes that are up-regulated in GPP and those 59 60 that are over-expressed in autoinflammatory syndromes caused by abnormal activation of Type-I IFN

4 Page 9 of 62 Journal of Investigative Dermatology

1 2 3 responses (Figure 1f). Of note, no overlap was found with the up-regulated genes detected in cryopyrin 4 5 associated periodic syndrome (CAPS), a disease caused by excessive IL-1 activity, which was analysed 6 7 as a negative control (Supplementary Figure S1). Thus, the presence of a Type-I IFN signature in GPP 8 9 10 leukocytes is supported by several lines of evidence. 11 12 To further investigate the relevance of these observations we built an interferon score by measuring the 13 14 aggregate expression of 5 genes (IFI6, IFIT3, IFITM3, OASL, PLSCR1), which were up-regulated in 15 16 the GPP dataset and annotated as Type-I IFN dependent in the interferome database (Rusinova et al., 17 18 2013). As expected, the score was elevated in GPP cases, compared to controls. A similar increase was 19 20 observed in the publicly available PsV dataset (Figure 1g). Importantly, we found that the interferon 21 For Review Only 22 score documented in GPP and PsV significantly correlated with the up-regulation of IL-36 related genes 23 24 (P<0.01) (Figure 1h). Thus, we have shown that systemic Type-I IFN responses are abnormally active 25 26 in psoriasis, which may be linked to increased IL-36 production. 27 28 29 30 31 The Type-I IFN signature is driven by gene up-regulation in neutrophils 32 33 The presence of heterogeneous cell populations in whole-blood can complicate the interpretation of 34 35 transcription profiling experiments. We therefore sought to validate our results through an independent 36 37 analysis of a single cell type. We focused our attention on neutrophils, as they play a critical role in 38 39 systemic inflammation and can be activated by Type-I IFN (Zimmermann et al., 2016). 40 41 We obtained fresh blood samples from 8 GPP cases and 11 controls (Supplementary Table S1a). 42 43 Following neutrophil isolation and RNA-sequencing, we detected 200 up-regulated genes (Figure 2a, 44 45 Supplementary Table S2b). The analysis of transcriptional networks identified Type-I interferon 46 47 response as the most significantly enriched module (FDR<10-12), followed by innate antiviral response 48 49 -10 50 and antiviral interferon signature (FDR<10 ) (Figure 2b). IPA also demonstrated a marked enrichment 51 -11 52 of pathways related to interferon signalling (FDR<10 ) (Figure 2c) and highlighted IRF7 and STAT1 53 54 as the most likely drivers of gene up-regulation (FDR<10-30) (Figure 2d, Supplementary Table S2d). In 55 56 keeping with these findings, interferon scores were elevated in GPP cases compared to controls 57 58 (P=0.02) (Figure 2e). These observations validate the results obtained in whole-blood and suggest that 59 60 the Type-I IFN signature is driven at least in part, by gene up-regulation in neutrophils.

5 Journal of Investigative Dermatology Page 10 of 62

1 2 3 4 5 The Type-I IFN signature can be validated in extended PsV and GPP datasets 6 7 We next sought to validate the type I IFN signature through the analysis of further affected individuals. 8 9 10 We examined neutrophils obtained from 17 GPP cases (including 8 newly recruited cases) and 16 PsV 11 12 patients suffering from severe disease (average Psoriasis Area and Severity Index: 17.9). We also 13 14 analysed two control groups including 9 individuals affected by CAPS and 26 healthy volunteers. Real- 15 16 time PCR demonstrated that the interferon score was significantly increased in GPP and PsV cases 17 18 compared to healthy controls (P<0.005). Conversely, and in keeping with the specificity of our 19 20 observations, the scores of CAPS patients were within the normal range defined in unaffected 21 For Review Only 22 individuals (Figure 3a). 23 24 Of note, medical records showed that GPP patients with high IFN scores were more likely to experience 25 26 systemic flares than those with low scores (88% vs 33%; P=0.049). Likewise, the prevalence of 27 28 psoriatic arthritis was higher among PsV subjects with high IFN scores (80% vs 17%; P=0.03) (Figure 29 30 31 3b). 32 33 Thus, the Type-I IFN signature detected by RNA-sequencing can be validated in independent PsV and 34 35 GPP samples, where it is associated with extra-cutaneous morbidity. 36 37 38 39 The IL-36 receptor is expressed on the surface of plasmacytoid dendritic cells 40 41 We next hypothesised that IL-36 has a direct effect on Type-I IFN producing cells. To investigate this 42 43 possibility, we systematically examined the surface expression of the IL-36 receptor (IL36R) in innate 44 45 immune cells. In keeping with published findings (Foster et al., 2014), we found that IL36R was barely 46 47 detectable on the surface of healthy neutrophils (Figure 4a), suggesting that the effects of IL-36 on these 48 49 50 cells are mediated by the activation of different immune population(s). 51 + 52 We also showed that IL36R cell numbers were low among innate lymphoid cells (Figure 4b) and in 53 54 monocytes (Figure 4c). Higher IL36R levels were observed in myeloid (mDC) and plasmacytoid 55 56 dendritic cells (pDCs) (Figure 4d, Supplementary Figure S2), with the largest percentage of IL36R+ 57 58 cells detected in the pDCs of GPP patients (Figure 4e). Thus, we have shown that IL-36R is robustly 59 60

6 Page 11 of 62 Journal of Investigative Dermatology

1 2 3 expressed in pDCs, which are the main producers of IFN-α (a member of the Type-I IFN family) in the 4 5 immune system. 6 7 8 9 10 IL-36 potentiates IFN-α production in response to Toll-like receptor 9 stimulation 11 12 Based on the results obtained in the above experiments, we hypothesised that IL-36 potentiates Type-I 13 14 IFN production by pDCs. To investigate this possibility, we pre-treated PBMCs obtained from healthy 15 16 donors with IL-36 or vehicle. We then stimulated the cells with CpG-containing DNA (hence CpG), a 17 18 TLR-9 ligand which induces IFN-α release by pDCs. Finally, we measured the up-regulation of the IFN 19 20 signature genes as a readout of Type-I IFN production. While CpG increased the expression of most 21 For Review Only 22 signature genes, its effect was more pronounced in cells that had been pre-incubated with IL-36 (P<0.05 23 24 for IFIT3, OASL and PLSCR1) (Figure. 5a). This observation was validated by direct measurements of 25 26 IFN-α production, showing increased cytokine release following IL-36 pre-treatment (Figure. 5b). 27 28 Finally, flow cytometry documented an increased proportion of IFNα+ pDCs among the cells that had 29 30 31 been stimulated with IL-36 and CpG, compared to those that had been exposed to CpG alone (Figure. 32 33 5c). Thus, multiple experimental readouts support the notion that IL-36 up-regulates TLR-9 dependent 34 35 IFN-α release. 36 37 38 39 IL-36 up-regulates PLSCR1, a known TLR-9 transporter 40 41 We next sought to define the mechanisms whereby IL-36 enhances cytokine production downstream of 42 43 TLR-9. A closer inspection of the PBMC stimulation results showed that IL-36 treatment up-regulates 44 45 PLSCR1, even in the absence of CpG. This is of interest, as the gene encodes phospholipid scramblase 46 47 1, a protein which regulates TLR-9 trafficking to the endosomal compartment (Talukder et al., 2012). 48 49 50 To further explore the link between IL-36 and PLSCR1, we first validated our initial observation in 51 52 additional donors (Figure 6a). Next, we demonstrated that IL-36 treatment increases PLSCR1 protein 53 54 levels in isolated pDCs, showing a direct effect of the cytokine on these cells (P<0.05) (Figure 6b). 55 56 Finally, we investigated the mechanism whereby IL-36 up-regulates PLSCR1. As expected for an IFN 57 58 signature gene, an analysis of the PLSCR1 promoter uncovered a STAT1 binding site. Given that IL- 59 60 36 can signal through mitogen-activated protein kinases (MAPK) (Bassoy et al., 2018), and that there

7 Journal of Investigative Dermatology Page 12 of 62

1 2 3 have been reports of cross-talk between STAT1 and MAPK signalling (Zhang et al., 2004), we reasoned 4 5 that the latter pathway was likely to be involved. Real-time PCR experiments confirmed this hypothesis, 6 7 as the SB-203580 MAPK inhibitor abolished the effect of IL-36 on PLSCR1 expression (Figure 6c). 8 9 10 Thus, we have demonstrated that IL-36 can act directly on pDCs, where it up-regulates PLSCR1, in a 11 12 MAPK-dependent fashion. 13 14 15 16 DISCUSSION 17 18 While PsV has been historically described as a dermatological condition, the importance of extra- 19 20 cutaneous co-morbidities is increasingly recognised (Armstrong et al., 2013). Of note, the prevalence 21 For Review Only 22 of most co-morbid conditions increases with the severity and the duration of the disease (Burden and 23 24 Kirby, 2016; Egeberg et al., 2017). There is therefore a dose-dependent association between cutaneous 25 26 and extra-cutaneous inflammation, which suggests a shared systemic pathogenesis. The underlying 27 28 pathways, however, remain poorly understood. 29 30 Here, we demonstrated that IL-36 signalling is enhanced in the leukocytes of PsV patients, where 31 32 abnormal IL-36 activity correlates with Type-I IFN over-expression. While many of the genes that are 33 34 35 induced by IL-36 are also up-regulated by IL-1, this set of shared targets does not include mediators of 36 37 Type-I IFN production (Swindell et al., 2018). Accordingly, we found that IFN signature genes are not 38 39 over-expressed in CAPS, a condition caused by excessive IL-1 signalling. Thus, IL-1 is unlikely to play 40 41 a significant role in promoting Type-I IFN responses in psoriasis. 42 43 Several studies have found that Type-I IFN is a mediator of vascular inflammation, which promotes the 44 45 recruitment of leukocytes to atherosclerotic plaques (Goossens et al., 2010; Niessner et al., 2007). 46 47 Experiments carried out in animal models have also shown that TLR-9 dependent Type-I IFN 48 49 production is a key driver of systemic autoimmunity (Di Domizio et al., 2012). 50 51 In keeping with these observations, signatures of excessive Type-I IFN activity have been documented 52 53 in various diseases presenting with prominent systemic involvement. One notable example is systemic 54 55 56 lupus erythematosus (SLE), a disorder associated with skin and joint inflammation, accelerated 57 58 atherosclerosis and up-regulation of genes such as IFI6 and OASL (El-Sherbiny et al., 2018). Of interest, 59 60 three independent studies have reported that IL-36 serum levels correlate with disease activity in SLE

8 Page 13 of 62 Journal of Investigative Dermatology

1 2 3 (Chu et al., 2015; Ismail et al., 2018; Mai et al., 2018), which further reinforces the link between IL-36 4 5 and Type-I IFN. Our work adds to these observations and provides mechanistic insights into the 6 7 underlying pathways. 8 9 10 Our computational and experimental results implicate pDCs as the most likely mediators of IL-36 11 12 activity. First, we identified IRF7 as one of the most significant drivers of differential 13 14 in GPP. Second, we demonstrated that IL-36R levels are highest in pDCs, especially among GPP 15 16 patients. Of note, it has long been established that pDCs accumulate within psoriatic skin lesions, where 17 18 they contribute to early disease processes alongside slanDC (Hansel et al., 2011; Nestle et al., 2005). It 19 20 has also been reported that IL36R is abundantly expressed in various classes of skin-resident DC 21 For Review Only 22 (Dietrich et al., 2016). Thus, it is tempting to speculate that IL-36 mediated pDC activation may also 23 24 have a pathogenic role in skin. 25 26 Our results show that the effects of IL-36 on pDCs are mediated at least in part by PLSCR1 up- 27 28 regulation. Interestingly, PLSCR1 siRNA knockout inhibits Type I IFN production by human pDCs 29 30 31 (Talukder et al., 2012), so it is reasonable to hypothesise that an increase in gene expression would have 32 33 the opposite effect. While the PLSCR1 induction observed in our IL-36 stimulation experiments was 34 35 modest (1.5-2.0 fold), it might be sufficient to activate a feed-forward loop whereby up-regulated 36 37 PLSCR1 promotes the production of Type-I IFN, which in turn induces further PLSCR1 transcription. 38 39 In fact, self-amplifying loops are a key feature of Type-I IFN signalling, as they are required for robust 40 41 antiviral responses (Hall and Rosen, 2010). 42 43 We cannot exclude the possibility that additional IL-36 responsive genes or cell types may also 44 45 contribute to the up-regulation of Type-I IFN. However, we have found that IL-36 does not affect the 46 47 expression of TLR9 or that of key downstream genes (IRF1, IRF3, IRF7; data not shown). We have also 48 49 50 observed that genes driving other antiviral pathways (DDX58/RIG-I, IFIH1/MDA5, 51 52 TMEM173/STING) are not up-regulated in PsV or GPP whole-blood. 53 54 While our pDC stimulations were carried out with a synthetic TLR-9 agonist, the identity of the agents 55 56 that cause IFN-α production in patients remains to be determined. In lesional skin, pDCs are activated 57 58 by self-nucleic acids released by apoptotic keratinocytes and bound to the LL-37 antimicrobial peptide 59 60 (Lande et al., 2007). Our transcriptomic data, however, suggests that this mechanism is unlikely to be

9 Journal of Investigative Dermatology Page 14 of 62

1 2 3 relevant at the systemic level. While CAMP (the gene encoding LL-37) was up-regulated in psoriatic 4 5 skin, it was not over-expressed in GPP or PsV whole-blood. Moreover, there was no correlation between 6 7 CAMP whole-blood expression and the up-regulation of Type-I IFN genes (r<0.1). Thus, the agents that 8 9 10 activate the circulating pDC of psoriatic patients may be different from those that are present in skin. 11 12 In conclusion, we have identified an IL-36/TLR-9 axis which up-regulates systemic Type-I IFN 13 14 production in psoriasis (Figure 6d). In GPP patients, the effects of IL-36 signalling are amplified by 15 16 inherited IL36RN mutations, a phenomenon which is likely to account for the severe nature of systemic 17 18 flares. In PsV, the Th17-dependent up-regulation of IL-36 cytokines is associated with a less 19 20 pronounced transcriptional signature and with signs of chronic systemic inflammation. 21 For Review Only 22 Given that IL-36 is down-regulated by IL-17 inhibitors such as (Kolbinger et al., 2017), 23 24 it is possible that treatment of psoriasis with IL-17 antagonists might also modulate Type-I IFN 25 26 production. Of note, the effects of direct IL-36 inhibition are currently being investigated in clinical 27 28 trials, with promising results obtained in a Phase I study (Bachelez, 2018). In this context, our work 29 30 31 suggests that IL-36 antagonists have the potential to improve systemic Type I IFN up-regulation and 32 33 extra-cutaneous manifestations of psoriasis. 34 35 36 37 METHODS 38 39 Human subjects 40 41 The study was performed according to the principles of the Declaration of Helsinki. Patients were 42 43 ascertained at St John’s Institute of Dermatology and Royal Free Hospital (London, UK), Glasgow 44 45 Western Infirmary (Glasgow, UK), Salford Royal Foundation Trust (Manchester, UK) and Hospital 46 47 Sultanah Aminah (Johor Bahru, Malaysia). The study was approved by the ethics committees of 48 49 50 participating institutions and written informed consent was obtained from all participants. 51 52 Nine unrelated GPP patients and 7 healthy controls were recruited for whole-blood RNA-sequencing, 53 54 while neutrophil RNA-sequencing was carried out in 8 GPP patients and 11 healthy controls. Five 55 56 controls and six cases were common to both studies (Supplementary Table S1a). For the validation of 57 58 neutrophil RNA-sequencing results, fresh blood was obtained from 17 GPP, 26 control, 9 CAPS and 59 60 17 PsV individuals (Supplementary Table S3). All PsV patients suffered from moderate-to-severe

10 Page 15 of 62 Journal of Investigative Dermatology

1 2 3 disease (Psoriasis Area Severity Index >10). None had received anti-TNF treatment. The IL36RN gene 4 5 was screened in all GPP cases and mutations were identified in 4 individuals (Supplementary Table 6 7 S1a). 8 9 10 11 12 RNA sequencing data analysis 13 14 The raw sequence data generated in house and that retrieved from public repositories (Supplementary 15 16 Table S4) were processed with the same computational pipeline (described in supplementary methods), 17 18 in order to standardise the data analysis process. Genes were considered up-regulated if the fold change 19 20 exceeded 1.5 (FDR<0.05). When RNA-sequencing and microarray data were compared, the analysis 21 For Review Only 22 focused on the 100 genes that were most significantly up-regulated in each sample, in order to account 23 24 for the different sensitivity of the two platforms. 25 26 Genes up-regulated in GPP were used as input for pathway and upstream regulator enrichment analyses 27 28 (IPA, Qiagen). STAT1- STAT3- and IRF7-centered networks were visualised with the igraph v1.0.1 R 29 30 31 Package. 32 33 The transcriptional modules that were active in our datasets were selected from the library published by 34 35 Li et al (Li et al., 2014). The enrichment_test function was then applied to the lists of up-regulated 36 37 genes. 38 39 The interferon score was built using the five Type-I IFN dependent genes that were most up-regulated 40 41 in GPP whole-blood (PLSCR1, OALS, IFI6, IFIT3, IFITM3). As IL-36 dependent genes have not been 42 43 systematically characterised in leukocytes, the IL-36 score was based on the analysis of five genes 44 45 which were strongly induced by IL-36 in keratinocytes (Mahil et al., 2017) and robustly expressed in 46 47 whole-blood (IL1B, PI3, VNN2, TNFAIP6, SERPINB1). Both scores were derived by normalising 48 49 50 RPKM values to a calibrator sample and then computing the median expression of the five genes. 51 52 53 54 Statistics 55 56 Differences between patient and control cytokine scores were assessed using an unpaired t-test or one- 57 58 way ANOVA, as appropriate. To account for donor variability in cytokine responses, IL-36/CpG 59 60 stimulations were analysed with non-parametric methods (Wilcoxon signed-rank test for comparisons

11 Journal of Investigative Dermatology Page 16 of 62

1 2 3 between two groups and Friedman’s test for comparison between three groups), as these do not assume 4 5 equal variance among samples. The correlation between cytokine scores was calculated using Spearman 6 7 method. The significance of overlaps observed in Venn diagrams was computed with a hyper-geometric 8 9 10 test and confirmed by bootstrap analysis. Fisher’s exact test was used to compare the clinical features 11 12 of patients with high and low IFN scores. 13 14 15 16 CREDIT STATEMENT: MC: data curation, formal analysis, investigation, visualization; MV: formal 17 18 analysis, investigation, validation, visualization; MR: investigation; SKM, SEC, ADB, HSY, IMC, 19 20 HJL: resources; GL: supervision; CHS: resources, writing-review & editing; FDC: supervision, writing- 21 For Review Only 22 review & editing; JNB: funding acquisition, resources, writing-review & editing; FC: conceptualization, 23 24 funding acquisition, project administration, supervision, writing-original draft. 25 26 27 28 ACKNOWLEDGEMENTS 29 30 31 We are grateful to Paola Di Meglio for her comments and technical advice. We acknowledge support 32 33 from the Department of Health via the NIHR BioResource Clinical Research Facility and 34 35 comprehensive Biomedical Research Centre award to Guy’s and St Thomas’ NHS Foundation Trust in 36 37 partnership with King’s College London and King’s College Hospital NHS Foundation Trust (guysbrc- 38 39 2012-1). The APRICOT clinical trial is funded by the Efficacy and Mechanism Evaluation Programme, 40 41 an MRC and NIHR partnership (grant EME 13/50/17 to CHS, FC and JNB). MC is supported by the 42 43 Psoriasis Association, MV by the UK Medical Research Council and SKM by a NIHR Clinical 44 45 Lectureship. 46 47 The views expressed in this publication are those of the author(s) and not necessarily those of the MRC, 48 49 50 NHS, NIHR or the Department of Health. 51 52 53 54 CONFLICT OF INTERESTS 55 56 The authors have received funding or fees from Abbvie and Novartis (CHS, HSY, JNB); Almirall, 57 58 Jansen, Leo Pharma and UCB (HSY, JNB); AnaptysBio (FCa); Aspire Pharma, Johnson and Johnson, 59 60 MEDA Pharmaceuticals (HSY); Boehringer Ingelheim (FCa, JNB, ADB); Bristol Myers Squibb,

12 Page 17 of 62 Journal of Investigative Dermatology

1 2 3 Celegene, Ely Lily, Pfizer, Samsung, Sienna, Sun Pharma (JNB); GSK, Roche, Regeneron, Sanofi 4 5 (CHS). 6 7 8 9 10 DATA AVAILABILITY 11 12 The RNA-sequencing data generated in this study are available through the Gene Expression Omnibus 13 14 (identifier: GSE123787). 15 16 17 18 REFERENCES 19 20 Armstrong EJ, Harskamp CT, Armstrong AW (2013) Psoriasis and major adverse 21 cardiovascular events: a systematicFor Review review and meta-analysis Only of observational studies. Journal 22 of the American Heart Association 2:e000062. 23 24 25 Bachelez H. Efficacy and safety of BI 655130, an anti-interleukin-36 receptor antibody, in 26 27 patients with acute generalised pustular psoriasis. In: Proceedings of the Conference Efficacy 28 and safety of BI 655130, an anti-interleukin-36 receptor antibody, in patients with acute 29 generalised pustular psoriasis; 2018; Paris. 30 31 32 Bassoy EY, Towne JE, Gabay C (2018) Regulation and function of interleukin-36 cytokines. 33 Immunological reviews 281:169-78. 34 35 36 Burden AD, Kirby B (2016) Psoriasis and related disorders. In: Rook’s Textbook of 37 Dermatology (Griffiths CEM, Barker JN, Bleiker T, Chalmers RJ, Creamer D, eds), 38 Chichester: Wiley-Blackwell. 39 40 41 Chu M, Wong CK, Cai Z, et al. (2015) Elevated Expression and Pro-Inflammatory Activity of 42 IL-36 in Patients with Systemic Lupus Erythematosus. Molecules 20:19588-604. 43 44 45 Davidovici BB, Sattar N, Prinz J, et al. (2010) Psoriasis and systemic inflammatory diseases: 46 potential mechanistic links between skin disease and co-morbid conditions. J Invest Dermatol 47 48 130:1785-96. 49 50 51 Di Domizio J, Dorta-Estremera S, Gagea M, et al. (2012) Nucleic acid-containing amyloid 52 fibrils potently induce type I interferon and stimulate systemic autoimmunity. Proc Natl Acad 53 Sci U S A 109:14550-5. 54 55 56 Dietrich D, Martin P, Flacher V, et al. (2016) Interleukin-36 potently stimulates human M2 57 macrophages, Langerhans cells and keratinocytes to produce pro-inflammatory cytokines. 58 Cytokine 84:88-98. 59 60

13 Journal of Investigative Dermatology Page 18 of 62

1 2 3 Egeberg A, Skov L, Joshi AA, et al. (2017) The relationship between duration of psoriasis, 4 vascular inflammation, and cardiovascular events. J Am Acad Dermatol 77:650-6 e3. 5 6 7 El-Sherbiny YM, Psarras A, Md Yusof MY, et al. (2018) A novel two-score system for 8 9 interferon status segregates autoimmune diseases and correlates with clinical features. 10 Scientific reports 8:5793. 11 12 13 Fang N, Jiang M, Fan Y (2016) Association Between Psoriasis and Subclinical Atherosclerosis: 14 A Meta-Analysis. Medicine 95:e3576. 15 16 17 Foster AM, Baliwag J, Chen CS, et al. (2014) IL-36 promotes myeloid cell infiltration, 18 activation, and inflammatory activity in skin. J Immunol 192:6053-61. 19 20 21 Goossens P, Gijbels MJ, ForZernecke Review A, et al. (2010) Myeloid Only type I interferon signaling promotes 22 atherosclerosis by stimulating macrophage recruitment to lesions. Cell Metab 12:142-53. 23 24 25 Hall JC, Rosen A (2010) Type I : crucial participants in disease amplification in 26 autoimmunity. Nat Rev Rheumatol 6:40-9. 27 28 29 Hansel A, Gunther C, Ingwersen J, et al. (2011) Human slan (6-sulfo LacNAc) dendritic cells 30 are inflammatory dermal dendritic cells in psoriasis and drive strong TH17/TH1 T-cell 31 responses. The Journal of allergy and clinical immunology 127:787-94 e1-9. 32 33 34 Honda K, Yanai H, Negishi H, et al. (2005) IRF-7 is the master regulator of type-I interferon- 35 36 dependent immune responses. Nature 434:772-7. 37 38 39 Ismail SM, Abd EMK, Mohamed MS (2018) Serum Levels of Pentraxin3 and Interlukin36 in 40 Patients with Systemic Lupus and their Relation to Disease Activity. The Egyptian journal of 41 immunology 25:81-91. 42 43 44 Johnston A, Xing X, Wolterink L, et al. (2016) IL-1 and IL-36 are dominant cytokines in 45 generalized pustular psoriasis. The Journal of allergy and clinical immunology. 46 47 48 Kolbinger F, Loesche C, Valentin MA, et al. (2017) beta-Defensin 2 is a responsive biomarker 49 of IL-17A-driven skin pathology in patients with psoriasis. The Journal of allergy and clinical 50 immunology 139:923-32 e8. 51 52 53 Lande R, Gregorio J, Facchinetti V, et al. (2007) Plasmacytoid dendritic cells sense self-DNA 54 coupled with antimicrobial peptide. Nature 449:564-9. 55 56 57 Li S, Rouphael N, Duraisingham S, et al. (2014) Molecular signatures of antibody responses 58 derived from a systems biology study of five human vaccines. Nat Immunol 15:195-204. 59 60

14 Page 19 of 62 Journal of Investigative Dermatology

1 2 3 Liu Y, Ramot Y, Torrelo A, et al. (2012) Mutations in proteasome subunit beta type 8 cause 4 chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature with 5 6 evidence of genetic and phenotypic heterogeneity. Arthritis Rheum 64:895-907. 7 8 9 Mahil SK, Catapano M, Di Meglio P, et al. (2017) An analysis of IL-36 signature genes and 10 individuals with IL1RL2 knockout mutations validates IL-36 as a psoriasis therapeutic target 11 Science translational medicine 9:eaan2514. 12 13 14 Mai SZ, Li CJ, Xie XY, et al. (2018) Increased serum IL-36alpha and IL-36gamma levels in 15 patients with systemic lupus erythematosus: Association with disease activity and arthritis. 16 International immunopharmacology 58:103-8. 17 18 19 Marrakchi S, Guigue P, Renshaw BR, et al. (2011) Interleukin-36-receptor antagonist 20 deficiency and generalized pustular psoriasis. N Engl J Med 365:620-8. 21 For Review Only 22 23 Nestle FO, Conrad C, Tun-Kyi A, et al. (2005) Plasmacytoid predendritic cells initiate psoriasis 24 through interferon-alpha production. The Journal of experimental medicine 202:135-43. 25 26 27 Niessner A, Shin MS, Pryshchep O, et al. (2007) Synergistic proinflammatory effects of the 28 antiviral cytokine interferon-alpha and Toll-like receptor 4 ligands in the atherosclerotic 29 30 plaque. Circulation 116:2043-52. 31 32 33 Onoufriadis A, Simpson MA, Pink AE, et al. (2011) Mutations in IL36RN/IL1F5 are 34 associated with the severe episodic inflammatory skin disease known as generalized pustular 35 psoriasis. Am J Hum Genet 89:432-7. 36 37 38 Quaranta M, Knapp B, Garzorz N, et al. (2014) Intraindividual genome expression analysis 39 reveals a specific molecular signature of psoriasis and eczema. Science translational medicine 40 6:244ra90. 41 42 43 Reich K (2012) The concept of psoriasis as a systemic inflammation: implications for disease 44 management. Journal of the European Academy of Dermatology and Venereology : JEADV 26 45 Suppl 2:3-11. 46 47 48 Rodero MP, Tesser A, Bartok E, et al. (2017) Type I interferon-mediated autoinflammation 49 50 due to DNase II deficiency. Nature communications 8:2176. 51 52 53 Rusinova I, Forster S, Yu S, et al. (2013) Interferome v2.0: an updated database of annotated 54 interferon-regulated genes. Nucleic acids research 41:D1040-6. 55 56 57 Shah K, Paris M, Mellars L, et al. (2017) Real-world burden of comorbidities in US patients 58 with psoriatic arthritis. RMD open 3:e000588. 59 60

15 Journal of Investigative Dermatology Page 20 of 62

1 2 3 Swindell WR, Beamer MA, Sarkar MK, et al. (2018) RNA-Seq Analysis of IL-1B and IL-36 4 Responses in Epidermal Keratinocytes Identifies a Shared MyD88-Dependent Gene Signature. 5 6 Frontiers in immunology 9:80. 7 8 9 Swindell WR, Remmer HA, Sarkar MK, et al. (2015) Proteogenomic analysis of psoriasis 10 reveals discordant and concordant changes in mRNA and protein abundance. Genome medicine 11 7:86. 12 13 14 Talukder AH, Bao M, Kim TW, et al. (2012) Phospholipid scramblase 1 regulates Toll-like 15 receptor 9-mediated type I interferon production in plasmacytoid dendritic cells. Cell research 16 22:1129-39. 17 18 19 Tortola L, Rosenwald E, Abel B, et al. (2012) Psoriasiform dermatitis is driven by IL-36- 20 mediated DC- crosstalk. The Journal of clinical investigation 122:3965-76. 21 For Review Only 22 23 Wang CQF, Suarez-Farinas M, Nograles KE, et al. (2014) IL-17 induces inflammation- 24 associated gene products in blood monocytes, and treatment with reduces their 25 26 expression in psoriasis patient blood. J Invest Dermatol 134:2990-3. 27 28 Zhang Y, Cho YY, Petersen BL, et al. (2004) Evidence of STAT1 phosphorylation modulated 29 30 by MAPKs, MEK1 and MSK1. Carcinogenesis 25:1165-75. 31 32 33 Zimmermann M, Arruda-Silva F, Bianchetto-Aguilera F, et al. (2016) IFNalpha enhances the 34 production of IL-6 by human neutrophils activated via TLR8. Scientific reports 6:19674. 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

16 Page 21 of 62 Journal of Investigative Dermatology

1 2 3 FIGURE LEGENDS 4 5 Figure 1. Transcription profiling of GPP and PsV whole-blood uncovers a Type-I IFN signature 6 7 that correlates with IL-36 activity. (A) Identification of genes that are differentially expressed in GPP. 8 9 10 Horizontal and vertical lines represent significance and fold change thresholds, respectively. The genes 11 12 underlying the IFN score are in red. (B) Higher expression of IL-36 dependent genes in whole-blood of 13 14 GPP and PsV patients, compared to controls (CTR). (C) Transcriptional modules enriched among genes 15 16 up-regulated in GPP. The FDR for each module is reported, with the underlying up-regulated genes 17 18 shown as grey cells. (D) Enriched pathways detected among genes over-expressed in GPP. (E) Key 19 20 transcriptional factors driving gene over-expression in GPP. (F) Overlap between the genes that are up- 21 For Review Only 22 regulated in GPP and IFNpathies. (G) Elevated IFN score in whole-blood samples of GPP and PsV 23 24 patients, compared to controls (CTR). (H) IL-36 and IFN scores are significantly correlated, in both 25 26 GPP and PsV patients. Dashed regression lines are plotted with 95% confidence intervals (grey areas). 27 28 The data in (B) and (G) are presented as mean +/- SD; *P<0.05, **P<0.01 (unpaired t-test). 29 30 31 32 33 Figure 2. Transcription profiling of GPP neutrophils confirms the presence of a Type-I IFN 34 35 signature (A) Identification of genes that are differentially expressed in GPP. Horizontal and vertical 36 37 lines represent significance and fold change thresholds, respectively. (B) Transcriptional modules 38 39 enriched among the genes that are up-regulated in GPP. The FDR for each module is reported, with the 40 41 underlying up-regulated genes shown as grey cells. (C) Enriched pathways detected among the genes 42 43 that are up-regulated in GPP. IFN-related pathways are highlighted in bold (D) Upstream regulator 44 45 analysis showing that IRF7 and STAT1 drive the up-regulation of numerous genes that are over- 46 47 expressed in GPP. (E) Elevated IFN score in the neutrophils of GPP patients, compared to controls. The 48 49 50 data are presented as mean +/- SD; *P<0.05 (unpaired t-test). 51 52 53 54 Figure 3. Validation of the Type-I IFN signature in extended datasets (A) Elevated IFN score in 55 56 the neutrophils of GPP and PsV patients, compared to healthy individuals. CAPS cases were analysed 57 58 as negative controls. The data are presented as mean +/- standard deviation; **P<0.01 and ***P<0.001 59 60 (one-way ANOVA followed by Dunnett’s post-test). (B) Left: systemic flares are more prevalent in

17 Journal of Investigative Dermatology Page 22 of 62

1 2 3 GPP patients with high interferon scores (n=8) compared to those with low interferon scores (n=9). 4 5 Right: psoriatic arthritis (PsA) is more prevalent in PsV patients with high interferon scores (n=6) 6 7 compared to those with low interferon scores (n=11). In both groups, the cut-off between high and low 8 9 10 scores was defined as the median +2SD of the values observed in healthy controls. *P<0.05 (Fisher’s 11 12 exact test). 13 14 15 16 Figure 4. The IL-36 receptor is preferentially expressed by pDCs. (A-E) Representative flow 17 18 cytometry plots showing IL36R surface expression, compared to fluorescence minus one (FMO) 19 20 controls. (A) neutrophils (gated as CD14+, CD15+, CD16+ cells); (B) innate lymphoid cells (lineage- 21 For Review Only 22 (CD3-, CD4-, CD19-, CD20-, CD56-), CD127+); (C) monocytes (CD3-, CD20-, CD19-, CD56-) 23 24 - high + + 25 separated into classical (CD16 , CD14 ), intermediate (CD16 , CD14 ) and pro-inflammatory 26 27 (CD16high, CD14-) populations; (D) pDCs (lineage-, HLADR+, CD123+, CD11c-) and mDCs (lineage-, 28 29 HLADR+, CD123-, CD11c+). (E) Histogram showing the percentage IL36R+ cells in each leukocyte 30 31 32 population. Data were obtained in at least 3 GPP cases and 3 sex-matched controls. Results are 33 34 presented as mean +/- SEM. No significant differences were observed between GPP cases and healthy 35 36 donors. 37 38 39 40 Figure 5. IL-36 enhances the production of IFN-α downstream of Toll-like receptor 9. (A) PBMCs 41 42 were stimulated with CpG for 6h, in the presence or absence of IL-36 pre-treatment (6h). The expression 43 44 of interferon signature genes was measured by real-time PCR. Data represent the mean +/- SEM of 45 46 results obtained in three independent donors, each stimulated in triplicate. (B) Following PBMC 47 48 stimulation, IFN-α production was measured by ELISA. Data represent the mean +/- SEM of results 49 50 51 obtained in two independent donors, each stimulated in triplicate. (C) Following PBMC stimulation, 52 + 53 the percentage of IFNα pDCs was determined by flow cytometry. A representative set of plots is 54 55 shown (left), together with the data obtained in 3 independent healthy donors (right). *P<0.05; 56 57 **P<0.01 (Friedman’s test, with Dunn’s post-test). 58 59 60

18 Page 23 of 62 Journal of Investigative Dermatology

1 2 3 Figure 6. IL-36 up-regulates PLSCR1 (A) Following treatment of PBMCs with IL-36, PLSCR1 4 5 expression was measured by real-time PCR. (B) Following IL-36 treatment of pDCs, PLSCR1 mean 6 7 fluorescence intensity (MFI) was measured by flow-cytometry, in gated PSLCR1+ pDCs. A 8 9 10 representative histogram is shown on the left. (C) Following pre-treatment with SB203580 (MAPKi), 11 12 PBMCs where stimulated with IL-36. PLSCR1 expression was then determined by real-time PCR. (D) 13 14 Proposed pathogenic model. Interleukin-36 produced by mDC up-regulates PLSCR1 in pDCs, 15 16 potentiating TLR-9 dependent IFN-α release. IFN-α induces further PLSCR1 transcription, thus 17 18 propagating an inflammatory feed-forward loop. All data are shown as mean +/- SEM of results 19 20 obtained in at least 3 donors, each stimulated in triplicate. *P<0.05 (Wilcoxon signed-rank test (A, B) 21 For Review Only 22 and Friedman’s test with Dunn’s post-test (C)) 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

19 Journal of Investigative Dermatology Page 24 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Review Only 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Figure 1 46 47 209x297mm (300 x 300 DPI) 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 25 of 62 Journal of Investigative Dermatology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Review Only 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Figure 2 36 37 209x185mm (300 x 300 DPI) 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Investigative Dermatology Page 26 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Review Only 20 21 22 23 24 25 26 27 28 29 Figure 3 30 31 192x133mm (300 x 300 DPI) 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 27 of 62 Journal of Investigative Dermatology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Review Only 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Figure 4 46 47 209x297mm (300 x 300 DPI) 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Investigative Dermatology Page 28 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Review Only 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 Figure 5 39 40 209x207mm (300 x 300 DPI) 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 29 of 62 Journal of Investigative Dermatology

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 For Review Only 20 21 22 23 24 25 26 27 28 29 Figure 6 30 209x143mm (300 x 300 DPI) 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Investigative Dermatology Page 30 of 62

1 2 3 SUPPLEMENTAL METHODS 4 5 RNA-sequencing 6 7 Total RNA was isolated from whole blood collected in Tempus™ Blood RNA Tube using a Tempus™ 8 Spin RNA Isolation Kit (Thermo). Samples were subjected to globin depletion using a GLOBINclear™ 9 10 Kit (Thermo). Neutrophil RNA was isolated with GeneJET RNA purification kit (Thermo). 11 12 Whole-blood RNA was sequenced on a HiSeq 3000 Illumina platform, obtaining 150bp paired-end 13 14 reads. Neutrophil RNA was sequenced on a NextSeq 500 Illumina platform obtaining 75bp single-end 15 reads. The quality of the sequence data was assessed using FastQC. Alignment against the HG38 16 17 was implemented in TopHat (Kim et al., 2013) with indexes generated by Botwie2. 18 19 Read counts produced by HTseq-count were used as input for the differential expression analysis, 20 which was performed with DESeq2 (Love et al., 2014) (R package, v 16.2), using sex as a co-variate. 21 For Review Only 22 23 24 Cell isolation and culture 25 Neutrophils were purified using the MACSxpress Whole Blood Neutrophil Isolation Kit (Miltenyi 26 27 Biotec). PBMCs were isolated using Ficoll-Paque PLUS (GE Healthcare). pDCs were purified from 28 29 PBMCs using a Plasmacytoid Dendritic Cell Isolation Kit (Miltenyi Biotec). PBMCs and pDCs were 30 cultured at a density of 2.5x106 cells/ml and 2.5x105 cells/ml, respectively, in RPMI Glutamax (Gibco) 31 32 supplemented with 10% FBS and 1% penicillin-streptomycin. Cells were stimulated with 50 ng/ml IL- 33 34 36α (Bio-Techne) for 6 hours and with 1.6 ng/ml ODN-A CpG (Invivogen) for a further 6 hours. For IFN- 35 α and PLSCR1 flow cytometry analysis, Brefeldin A (BioLegend) was added to the stimulated cells at a 36 37 1: 1,000 dilution after 9 hours. Response to stimulation was measured by real-time PCR, ELISA or flow- 38 39 cytometry. 40 41 42 Real-time PCR and ELISA 43 44 RNA samples were isolated with the GeneJET RNA purification kit (Thermo). Following reverse 45 transcription with the nanoScript2 kit (Primerdesign), gene expression was assessed by real-time PCR 46 47 using PrecisionPLUS Master Mix with SYBR and ROX (Primerdesign) in conjunction with the following 48 49 primer pairs: 50 IFI6: 5’-TTTCTTACCTGCCTCCACCC-3’; 5’-CCATCTATCAGCAGGCTCCG-3’ 51 52 IFIT3: 5’-TTGGTGACCTCACTCATGATGG-3’; 5’-GCACAGACCTAACAGCACCC-3’ 53 54 IFITM3: 5’-CACTGGGATGACGATGAGCA-3’; 5’-TCGCCTACTCCGTGAAGTCTA-3’; 55 OASL: 5’-GGAACCTGGAAGGACAGACG-3’; 5’-GTACCAGCAGAGGGCACG-3’ 56 57 PLSCR1: 5’-AGGAGGATACCCAACTGGCA-3’; 5’-CGGCAGCCAGAGAACTGTTTTA-3’ 58 59 IL1B: 5’-GCCCTAAACAGATGAAGTGCTC-3’; 5’-GAACCAGCATCTTCCTCAG-3’. 60

1 Page 31 of 62 Journal of Investigative Dermatology

1 2 3 The IFN score was derived by computing the median RQ of the five signature genes, (PLSCR1, OALS, 4 5 IFI6, IFIT3, IFITM3) according to the method described by Rice et al(Rice et al., 2013). The production 6 7 of IFN-α was measured using the Human IFN-alpha ELISA kit (Bio-Techne). For whole blood and PBMC 8 samples, transcript levels were normalised to B2M expression, while RPL13A was used for neutrophils. 9 10 11 12 Flow cytometry 13 The purity of neutrophil isolated for RNA-sequencing was measured by staining cells with anti-CD45, 14 15 anti-CD15, anti-CD16, anti-CD3, anti-CD24 and anti-CD19 antibodies. IL-36 receptor (IL36R) surface 16 17 expression and IFN-α levels were measured by staining PBMCs with LIVE/DEAD™ Fixable Near-IR 18 (Invitrogen), Fc and monocyte blocker (Biolegend), antibody against the protein of interest and an 19 20 antibody cocktail for monocytes (anti-CD3, anti-CD20, anti-CD19, anti-CD16, anti-CD14, anti-CD56) or 21 For Review Only 22 DCs and ILCs (Lineage, HLA-DR, CD123, CD11c, CD127). IL36R expression on neutrophils was 23 determined by staining for the receptor as well as CD15, CD16 and CD14. PLSCR1 expression was 24 25 measured by staining purified pDCs with anti-CD123, anti-HLA-DR, anti-CD11c and anti-PLSCR1. Cells 26 27 were acquired on a BD Fortessa LSR or a BD FACSCanto II instrument. All data was analysed using 28 FlowJo v10 software. The details of all antibodies are reported in Supplementary Table S5. 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

2 Journal of Investigative Dermatology Page 32 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Figure S1: Lack of evidence for an IFN signature in the leukocytes of CAPS patients 18 19 Venn diagrams showing the overlap between the genes that are up-regulated in GPP and in the CAPS 20 datasets analysed by Canna et al (CAPS-1) and Balow et al (CAPS-2). Neither overlap was statistically 21 significant. For Review Only 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 Figure S2: IL-36R surface expression measured as mean fluorescence intensity 52 53 Data are presented as the mean (+/- SEM) of measurements obtained in three unrelated healthy 54 donors. ILC, innate lymphoid cells; Mo, monocyte. 55 56 57 58 59 60

3 Page 33 of 62 Journal of Investigative Dermatology

1 2 3 Table S3: Details of individuals analysed for IFN score validation 4 5 Group Individuals with Sex Average age Treatment 6 active disease 7 CAPS 4/9 3M, 6F 44.6 Cankinumab (n=7), (n=2) 8 9 GPP 11/17 2M, 15F 56.5 Adalimumab (n=3), Certolizumab (n=1), 10 Infliximab (n=2), (n=1), 11 Acitretin (n=1), Ciclosporin (n=2), MTX 12 (n=1), Prednisolone (n=1), topical 13 treatment (n=4), no treatment (n=1), 14 PsV 18/18 13M, 4F 42.5 Apremilast (n=1), Etanercept (n=1), 15 Infliximab (n=1), Ixekizumab (n=1), 16 Secukinumab (n=1), Ustekinumab (n=4), 17 18 MTX (n=2), topical treatment (n=6), PUVA 19 (n=1) 20 Control n/a 3M, 23F 46.4 n/a 21 CAPS, cryopyrin associated periodicFor syndrome; Review GPP, generalised Only pustular psoriasis; PsV, psoriasis vulgaris; MTX, 22 Methotrexate 23 24 25 26 27 Table S4: Details of publicly available datasets 28 29 Dataset Disease Ethnicity Tissue Platform Identifier 30 CAPS-1 NOMID European Whole blood RNAseq GSE57253 31 CAPS-2 CAPS European Whole blood microarray Suppl. Material, 32 PMID: 23223423 33 IFNpathy-1 Type-I IFN mediated European Whole blood RNAseq E-MTAB-5735 34 autoinflammation 35 36 IFNpathy-2 CANDLE European Whole blood microarray data provided by 37 authors 38 PsV Psoriasis Vulgaris European Whole blood RNAseq GSE67785 39 40 All patients had active disease. Raw data were downloaded from the Gene Expression Omnibus as CEL (CAPS-2 41 and IFN-pathy2) or FASTQ files. The former were analysed with the limma package, while the latter were 42 processed with the standardised pipeline described above (supplemental methods). 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

4 Journal of Investigative Dermatology Page 34 of 62

1 2 3 Table S5: Flow cytometry antibodies 4 5 Target Dilution Cat number Fluorochrome Supplier 6 CD16 1:20 48-0168-42 Efluor450 ThermoFisher 7 1:22 130-098-101 APC Miltenyi Biotec 8 CD56 1:33 318316 Alexa Fluor 700 BioLegend 9 10 CD19 1:20 302242 BV510 BioLegend 11 1:22 555412 FITC BD 12 CD20 1:33 130-096-649 PE-Cy7 Miltenyi Biotec 13 CD14 1:20 555398 PE BD 14 15 CD3 1:33 317306 FITC BioLegend 16 1:22 130-098-162 FITC Miltenyi Biotec 17 CD127 1:20 351320 PE-Cy7 BioLegend 18 HLA-DR 1:33 307636 BV421 BioLegend 19 20 CD11c 1:20 301638 BV650 BioLegend 21 CD123 For Review1:30 Only306030 BV711 BioLegend 22 Lineage marker1 1:10 B29559 PE Beckman Coulter 23 24 CD15 1:33 301904 FITC BioLegend 25 1:22 130-098-010 PE Miltenyi Biotec 26 IL36R 1:10 BAF Streptavidin BD 27 Streptavidin2 1:100 405207 APC BioLegend 28 29 CD45 1:22 130-098-139 VioBlue Miltenyi Biotec 30 CD24 1:22 130-099-935 APC Vio770 Miltenyi Biotec 31 PLSCR1 1:50 ab180518 Rabbit-IgG Abcam 32 Rabbit IgG3 1:100 406416 Alexa Fluor 488 BioLegend 33 34 IFN-α 1:10 130-092-602 APC Miltenyi Biotec 35 1CD3/CD14/CD19/CD20/CD56; 2Target of secondary biotinylated antibody used for IL36R detection; 3Target of 36 secondary antibody used for PLSCR1 detection 37 38 39 40 41 Supplemental References 42 43 Kim D, Pertea G, Trapnell C, et al. (2013) TopHat2: accurate alignment of transcriptomes in the 44 presence of insertions, deletions and gene fusions. Genome biology 14:R36. 45 46 Love MI, Huber W, Anders S (2014) Moderated estimation of fold change and dispersion for RNA-seq 47 data with DESeq2. Genome biology 15:550. 48 Rice GI, Forte GM, Szynkiewicz M, et al. (2013) Assessment of interferon-related biomarkers in Aicardi- 49 50 Goutieres syndrome associated with mutations in TREX1, RNASEH2A, RNASEH2B, RNASEH2C, 51 SAMHD1, and ADAR: a case-control study. Lancet Neurol 12:1159-69. 52 53 Ritchie ME, Phipson B, Wu D, et al. (2015) limma powers differential expression analyses for RNA- 54 sequencing and microarray studies. Nucleic acids research 43:e47. 55 56 57 58 59 60

5 Page 35 of 62 Journal of Investigative Dermatology

1 2 Table S1 A. Clinical and demographic features of GPP patients 3 4 Patient ID Ethnicity Sex1 Age of onset Concurrent PsV 5 6 GBR0006 European F 51 No 7 GBR0010 European F 13 No 8 GYFAP0011 European F 42 Yes 9 GYFAP0014 European F 10 No 10 GYFAP0016 European M 5 No 11 12 GYFAP0029 European F 7 No 13 GYFAP0032 European F 5 No 14 GYFAP0041 Asian F 31 Yes 15 GYFAP0089 European F 29 No 16 GYFAP0096 European F 19 Yes 17 18 GYPLM0001 European F 17 No 19 20 Two patients were experiencing an acute systemic GPP flare when samples were taken for whole-blood (GYFAP0011) and neutrophil (GYFAP0096) RNA-sequencing. CRP=C-reactive protein 21 For Review Only 1 The control sample included 2 males and 10 females 22 2 23 Fever is reported if >38C; elevated CRP= CRP>100mg/L 24 3The average age at recruitment was 56 for cases and 51 for controls 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Investigative Dermatology Page 36 of 62

1 2 3 4 History of systemic Inflammation2 Treatment IL36RN status 5 6 Yes (Fever, Neutrophilia, Elevated CRP) Unknown S113L/S113L 7 Yes (Elevated CRP) Unknown wild-type 8 Yes (Neutrophilia, Elevated CRP) No treatment S113L/- 9 Yes (Fever, Neutrophilia) Infliximab wild-type 10 Yes (Fever, Neutrophilia, Elevated CRP) Methotrexate S113L/S113L 11 12 Yes (Fever, Elevated CRP) Infliximab/ Methotrexate R48W/S113L 13 Yes (Fever, Elevated CRP) Acitretin wild-type 14 Unknown Ustekinumab/ Methotrexate wild-type 15 Yes (Neutrophilia, Elevated CRP) Adalimumab/ Methotrexate wild-type 16 Yes (Neutrophilia) Topical wild-type 17 18 Yes (Fever) Topical wild-type 19 Two patients were experiencing an acute systemic GPP flare when20 samples were taken for whole-blood (GYFAP0011) and neutrophil (GYFAP0096) RNA-sequencing. CRP=C-reactive protein 21 For Review Only 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 37 of 62 Journal of Investigative Dermatology

1 2 3 4 Age at recruitment3 RNA-sequencing dataset 5 6 65 Whole-blood 7 90 Whole-blood 8 47 Whole-blood/ Neutrophil 9 48 Whole-blood/ Neutrophil 10 43 Whole-blood/ Neutrophil 11 12 46 Whole-blood/ Neutrophil 13 88 Whole-blood/ Neutrophil 14 47 Whole-blood 15 39 Whole-blood/ Neutrophil 16 71 Neutrophil 17 18 30 Neutrophil 19 Two patients were experiencing an acute systemic GPP flare when samples were taken for whole-blood (GYFAP0011) and neutrophil (GYFAP0096) RNA-sequencing.20 CRP=C-reactive protein 21 For Review Only 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Investigative Dermatology Page 38 of 62

1 2 Table S1 B. Frequency of immune populations derived by deconvolution of RNA-sequencing profiles1 3 Table S1 B. Frequency of immune populations derived by deconvolution of RNA-sequencing profiles1 4 5 Individual Status naïve B cells Memory B cells Plasma cells 6 GBR0006 GPP case 0.0182 0.0210 0.0000 7 GBR0010 GPP case 0.0403 0.0000 0.0000 8 GYFAP0011 GPP case 0.0285 0.0000 0.0001 9 GYFAP0014 GPP case 0.0579 0.0003 0.0000 10 11 GYFAP0016 GPP case 0.0097 0.0041 0.0000 12 GYFAP0029 GPP case 0.0239 0.0000 0.0000 13 GYFAP0032 GPP case 0.0254 0.0000 0.0000 14 GYFAP0041 GPP case 0.1237 0.0000 0.0002 15 GYFAP0089 GPP case 0.0592 0.0031 0.0000 16 17 CTR1 Healthy control 0.1272 0.0000 0.0000 18 CTR2 Healthy control 0.0616 0.0000 0.0000 19 CTR3 Healthy control 0.0387 0.0140 0.0000 20 CTR4 Healthy control 0.0525 0.0045 0.0000 21 For Review Only CTR5 Healthy control 0.0637 0.0133 0.0000 22 23 CTR6 Healthy control 0.0542 0.0458 0.0000 24 CTR7 Healthy control 0.0716 0.0000 0.0000 25 26 Pvalue t.test GPP vs. CTR 0.16 0.21 0.24 27 28 29 1 Analysis carried out with CIBERSORT (Newman et al. Nature Methods, 12: 453-7, 2015) 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 39 of 62 Journal of Investigative Dermatology

1 2 3 4 5 CD8 T cells CD4 naïve T cells CD4 memory resting T cells CD4 memory activated T cells 6 0.0573 0.0615 0.1265 0.0112 7 0.0000 0.0000 0.0691 0.0000 8 0.0155 0.0367 0.0558 0.0188 9 0.0658 0.0780 0.2006 0.0000 10 11 0.0289 0.0159 0.1340 0.0000 12 0.0533 0.0155 0.0000 0.0253 13 0.1506 0.0645 0.0760 0.0058 14 0.1174 0.0591 0.1155 0.0002 15 0.1992 0.0990 0.0904 0.0000 16 17 0.1089 0.1034 0.0000 0.0000 18 0.1159 0.0207 0.0230 0.0000 19 0.0355 0.1469 0.1492 0.0000 20 0.1383 0.0772 0.0680 0.0000 21 For Review Only 0.1204 0.0836 0.1375 0.0000 22 23 0.2116 0.0841 0.1127 0.0000 24 0.1447 0.0040 0.2038 0.0000 25 26 0.13 0.21 0.93 0.08 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Investigative Dermatology Page 40 of 62

1 2 3 4 5 regulatory T cells resting NK cells activated NK cells Monocytes 6 0.0354 0.0749 0.0000 0.1899 7 0.0112 0.0555 0.0000 0.1279 8 0.0092 0.0641 0.0000 0.2552 9 0.0363 0.0924 0.0040 0.2058 10 11 0.0391 0.1359 0.0000 0.2256 12 0.0130 0.1023 0.0000 0.1693 13 0.0042 0.1387 0.0000 0.1841 14 0.0441 0.0486 0.0016 0.1886 15 0.0107 0.1342 0.0000 0.2079 16 17 0.0464 0.1629 0.0000 0.0949 18 0.0261 0.1074 0.0000 0.1716 19 0.0564 0.0941 0.0000 0.2309 20 0.0251 0.1338 0.0000 0.1843 21 For Review Only 0.0283 0.0970 0.0000 0.2281 22 23 0.0087 0.1606 0.0000 0.2285 24 0.0261 0.1446 0.0000 0.1644 25 26 0.30 0.06 0.25 0.68 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 41 of 62 Journal of Investigative Dermatology

1 2 3 4 5 activated dendritic cells resting mast cells neutrophils 6 0.0000 0.0190 0.3852 7 0.0031 0.0108 0.6675 8 0.0001 0.0213 0.4947 9 0.0000 0.0198 0.2386 10 11 0.0009 0.0245 0.3814 12 0.0000 0.0271 0.5240 13 0.0000 0.0202 0.3303 14 0.0023 0.0171 0.2816 15 0.0000 0.0042 0.1921 16 17 0.0000 0.0091 0.3411 18 0.0000 0.0229 0.4507 19 0.0103 0.0058 0.2053 20 0.0000 0.0116 0.3048 21 For Review Only 0.0001 0.0215 0.2065 22 23 0.0000 0.0096 0.0843 24 0.0000 0.0208 0.2171 25 26 0.58 0.31 0.08 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Investigative Dermatology Page 42 of 62

1 2 Table S2 A. Genes differentially expressed in GPP whole-blood Table S2 B. Genes differentially expressed in GPP neutrophils 3 4 Symbol log2FoldChange P value FDR Symbol 5 CD177 3.04295775 0.00000528 0.008398714 CYP26B1 6 7 MMP9 2.397371799 0.000266966 0.031067237 OTOF 8 IFITM3 1.958710941 0.000172415 0.026846582 OAS2 9 FCGR1A 1.950358009 0.000325286 0.033853443 SIGLEC1 10 TNFAIP6 1.841118768 0.00000508 0.008398714 USP18 11 LINC01506 1.666671765 0.0000551 0.022249039 RSAD2 12 13 FCGR1B 1.652856909 0.0000717 0.024241989 IFI44L 14 IL1R2 1.634533944 0.000112311 0.024241989 OAS1 15 RAB20 1.585193872 0.0000879 0.024241989 OAS3 16 LINC00694 1.58370161 0.00022307 0.028142172 GBP1P1 17 SERPING1 1.563847141 4.11E-08 0.000326472 LY6E 18 19 BTNL8 1.526139204 0.00000512 0.008398714 SPATS2L 20 VNN2 1.467233465 0.000184872 0.026846582 IFI44 21 PLSCR1 1.452924592For0.0000129 Review0.01022295 Only RTP4 22 LY96 1.440496198 0.000167074 0.026846582 CMPK2 23 GRINA 1.404850678 0.000178734 0.026846582 ISG15 24 25 CEBPD 1.39126541 0.0000206 0.011709421 EPSTI1 26 DSC2 1.357701906 0.000506189 0.042349396 SEPT4 27 FCAR 1.351216375 0.000404978 0.036730403 FSTL4 28 NAMPT 1.341116229 0.000288296 0.031388683 AGRN 29 30 DUSP1 1.321742696 0.0000154 0.01022295 DDX60 31 TREM1 1.308375777 0.0000928 0.024241989 MT2A 32 FUT7 1.30779229 0.000406678 0.036730403 IFI6 33 KREMEN1 1.293351423 0.0000153 0.01022295 NRIR 34 PLAUR 1.291985941 0.000113972 0.024241989 RMI2 35 36 SCO2 1.283351297 0.0000088 0.009993647 IFIT1 37 SAT1 1.275800594 0.0000823 0.024241989 LAMP3 38 IFI6 1.268876241 0.000556746 0.0444672 HERC5 39 OASL 1.262042923 0.0000116 0.01022295 XAF1 40 SECTM1 1.252936596 0.000166672 0.026846582 MTRNR2L8 41 42 SH2B2 1.247794605 0.000122003 0.024241989 SERPING1 43 IL1R1 1.241220836 0.0006624 0.047861396 MX1 44 MIR223 1.229072695 0.000108435 0.024241989 BATF2 45 BCL3 1.213916816 0.000396979 0.036688277 OASL 46 CEBPB 1.211972891 0.000119267 0.024241989 CARD17 47 48 TGFA 1.210356329 0.0000502 0.022249039 ATF3 49 BRI3 1.210073601 0.000247737 0.029833516 TRIM22 50 NINJ1 1.20536251 0.000173177 0.026846582 PNPT1 51 IFIT3 1.201503001 0.000353181 0.034655383 ZNF496 52 TXN 1.188830189 0.000112743 0.024241989 SOCS1 53 54 NFIL3 1.184005814 0.000605531 0.044979105 DHX58 55 FPR2 1.178124673 0.000742685 0.04981533 TIMM10 56 CCDC71L 1.163316062 0.000481707 0.04072986 IFIT3 57 PILRA 1.151790766 0.0000889 0.024241989 FRMD3 58 TMEM120A 1.15016736 0.0000997 0.024241989 GBP5 59 60 LST1 1.147247274 0.000222423 0.028142172 BST2 CD14 1.136557808 0.000269708 0.031067237 GBP1 Page 43 of 62 Journal of Investigative Dermatology

1 2 CSRNP1 1.135025792 0.000195911 0.026846582 HERC6 3 SCARF1 1.117577421 0.000565146 0.044473082 PLSCR1 4 KLHDC8B 1.107370736 0.000189244 0.026846582 CASP7 5 CYP27A1 1.105909791 0.00000773 0.009993647 LHFPL2 6 7 SMARCD3 1.100786838 0.000464089 0.040093223 TFEC 8 NCF1 1.094126316 0.000320547 0.033853443 MOV10 9 ASPRV1 1.088922741 0.000282991 0.031239016 FGF13 10 CFP 1.084963024 0.0000923 0.024241989 PDCD1LG2 11 ZNF467 1.084675182 0.000280167 0.031239016 KCNH3 12 13 STX11 1.081622222 0.0000323 0.016035698 IFIT5 14 TMEM140 1.071770225 0.000116827 0.024241989 PARP12 15 TSC22D3 1.070545102 0.0000254 0.013437826 RSPH9 16 VNN3 1.062781443 0.000672063 0.048122121 GADD45A 17 CCR1 1.046325469 0.000346339 0.034408778 EXOC3L1 18 19 SLC16A3 1.043291184 0.000413073 0.036888854 IFI35 20 CEP19 1.041133122 0.000165965 0.026846582 GPR84 21 GLIPR2 1.040078347For0.000602893 Review0.044979105 Only FBXO6 22 DDX60L 0.971244183 0.000168495 0.026846582 PLVAP 23 LAT2 0.969533048 0.000434617 0.03827743 GYG1 24 25 F2RL1 0.959517707 0.0000949 0.024241989 ANKRD22 26 MXD1 0.948687245 0.000631177 0.046023784 PPM1K 27 HIP1 0.940892407 0.0000877 0.024241989 OLAH 28 WAS 0.939083067 0.000602265 0.044979105 LINC00968 29 30 B4GALT5 0.938956141 0.000474198 0.040526084 CD2AP 31 IL10RB 0.930192661 0.000684041 0.048542472 APOL6 32 NFAM1 0.929966061 0.000517666 0.042491753 CD274 33 PARP9 0.921613008 0.0000152 0.01022295 GBP6 34 TMEM164 0.916479563 0.000277091 0.031239016 SPATC1 35 36 IFIT2 0.911671135 0.000327971 0.033853443 IFIT2 37 FAM214B 0.900344638 0.000594281 0.044979105 P2RY14 38 RNF130 0.899068664 0.00026931 0.031067237 PARP10 39 CTSL 0.896156972 0.000200966 0.027072497 ZBP1 40 HSD17B11 0.896066515 0.000179688 0.026846582 FCGR1A 41 42 SRA1 0.891298978 0.000577688 0.044934475 ANXA2R 43 SHISA5 0.890487969 0.0000652 0.024241989 SAMD9L 44 IL13RA1 0.87127311 0.000338905 0.034096457 KIAA0895L 45 CTSB 0.864037569 0.000374024 0.035816151 CMAHP 46 CHMP5 0.862681239 0.0000897 0.024241989 STAT2 47 48 PSMB9 0.855259523 0.000068 0.024241989 TMEM62 49 STAB1 0.854836857 0.000229157 0.028458478 AIM2 50 IRF1 0.845977325 0.0000187 0.011418132 GCH1 51 RHBDD2 0.834192928 0.000187455 0.026846582 UBE2L6 52 ODF3B 0.826143436 0.000690854 0.048592116 AANAT 53 54 ERGIC1 0.818750589 0.000543189 0.044053704 KPTN 55 DGAT1 0.786713662 0.000157624 0.026846582 PML 56 MX2 0.785979868 0.000338657 0.034096457 GRAMD1B 57 SOCS3 0.778990735 0.000379164 0.035820532 EIF2AK2 58 SOWAHD 0.775922159 0.00020487 0.027138445 RNF144A 59 60 EEPD1 0.769258755 0.000194695 0.026846582 SP140 NMI 0.746899529 0.000438254 0.03827743 LILRA5 Journal of Investigative Dermatology Page 44 of 62

1 2 NOD2 0.74400574 0.0000512 0.022249039 TLDC2 3 NEU1 0.743779072 0.000710499 0.049104774 CMTR1 4 MYADM 0.732701032 0.000241614 0.029543839 EFCAB2 5 TAP1 0.728549188 0.00018559 0.026846582 FRMD4B 6 7 GBP5 0.706955725 0.000582316 0.044934475 PCGF5 8 GALNT3 0.683403165 0.000518583 0.042491753 C5 9 PSMB8 0.675066415 0.000139079 0.026318992 SYNPO2 10 APOL2 0.668305203 0.000294704 0.031652779 CARD14 11 APOL1 0.666677174 0.000056 0.022249039 RNF213 12 13 STK19 0.646144047 0.000113589 0.024241989 EXT1 14 CMTM7 0.639456007 0.000126933 0.024606455 SAMHD1 15 H2AFY 0.606824543 0.000733701 0.04981533 ASPHD2 16 SEC61B 0.567143165 0.000745851 0.04981533 ZCCHC2 17 SF3B5 0.407198213 0.000215272 0.028048941 SSPN 18 19 TMPO -0.42712245 0.000559477 0.0444672 ANXA5 20 CARNS1 -0.634539261 0.000358644 0.034762247 DDX58 21 ZNF37BP -0.664722004For0.00070223 Review0.048958961 Only STOML1 22 PLEKHA1 -0.740903036 0.000741473 0.04981533 TNFAIP6 23 CLUHP3 -0.901032271 0.000612524 0.045077264 SPTSSA 24 25 NELL2 -1.023445283 0.000178216 0.026846582 VAMP5 26 IRF7 27 NT5C3A 28 KLF4 29 30 RARRES3 31 CACNA1A 32 TNFSF13B 33 SCO2 34 RHBDF2 35 36 FCGR1B 37 TOR1B 38 UBE2F 39 LINC01270 40 GIMAP2 41 42 NEK11 43 ODF3B 44 ZC3HAV1 45 FANCL 46 FANCA 47 48 MDK 49 VPS9D1 50 CNP 51 DRAP1 52 KLHDC7B 53 54 STAT1 55 CNIH4 56 VPS9D1-AS1 57 GPR160 58 TMSB10 59 60 PARP9 NASP Page 45 of 62 Journal of Investigative Dermatology

1 2 ISG20 3 COPG2 4 NLRC5 5 KIAA1958 6 7 IFI16 8 ARHGAP24 9 TDRD7 10 RBM43 11 CARD16 12 13 MSL3 14 IL18R1 15 EPAS1 16 SESTD1 17 DUSP3 18 19 NOD1 20 SNX20 21 For Review Only EPB41L5 22 PHF11 23 OBFC1 24 25 STX11 26 GIMAP4 27 PSMB9 28 TRIM69 29 30 GBP2 31 TAP1 32 LGALS3BP 33 CASP1 34 TRIM38 35 36 TMEM170B 37 C18orf25 38 PHACTR2 39 SRBD1 40 ACOT9 41 42 LRRC46 43 GSTK1 44 TXN 45 AFF1 46 SIAH2 47 48 C4orf3 49 SPATA13 50 JAG1 51 DAPP1 52 TYMP 53 54 HEXDC 55 RAB12 56 TCAIM 57 FAM46C 58 ERLIN1 59 60 CLEC2B CARS2 Journal of Investigative Dermatology Page 46 of 62

1 2 TIPARP 3 DPYD 4 LY96 5 PDCD11 6 7 SYNGAP1 8 SLBP 9 C2CD2L 10 CEP104 11 GNPDA1 12 13 MFNG 14 DDX28 15 SEMA6C 16 SSBP3 17 AGAP3 18 19 NFATC3 20 BOD1 21 For Review Only OBSCN 22 FBXL16 23 ZAP70 24 25 P2RY10 26 HSPG2 27 MYCL 28 CD5 29 30 RHPN2 31 ESRP2 32 CCDC163P 33 FADS1 34 CD160 35 36 CLIC3 37 GSTM4 38 GSTM2 39 MYOM2 40 FAM3B 41 42 FADS2 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 47 of 62 Journal of Investigative Dermatology

1 Table2 S2 B. Genes differentially expressed in GPP neutrophils 3 4 log2FoldChange P value FDR 5 9.172412148 6.27E-06 0.002025496 6 7 5.119228745 4.50E-08 0.000137377 8 3.971808127 3.38E-07 0.000375607 9 3.947811981 2.34E-05 0.004503355 10 3.841053847 5.04E-06 0.001709095 11 3.663108849 2.90E-07 0.000354599 12 13 3.563097345 6.66E-07 0.000488218 14 3.523123021 3.76E-06 0.001444593 15 3.41407656 2.19E-06 0.001028476 16 3.328636236 2.35E-07 0.000318987 17 3.308108285 3.40E-06 0.001434257 18 19 3.271775597 9.52E-07 0.000567181 20 3.21792139 7.32E-07 0.000488218 21 3.105936328 4.81E-07For0.000419908 Review Only 22 3.055439406 2.77E-06 0.001208097 23 3.028183431 7.59E-07 0.000488218 24 25 3.003631939 9.98E-06 0.002836422 26 2.944237993 2.08E-08 8.48E-05 27 2.764056435 0.000400287 0.026431925 28 2.755875361 4.16E-05 0.006279419 29 30 2.538926661 4.51E-06 0.001666504 31 2.514422494 5.75E-08 0.000140394 32 2.507376228 1.99E-05 0.004267589 33 2.458692631 5.80E-05 0.00762578 34 2.412486832 2.11E-07 0.000318987 35 36 2.341066217 8.61E-05 0.009744028 37 2.310196414 0.000298783 0.02212082 38 2.243206042 3.41E-05 0.005449044 39 2.21414774 3.34E-05 0.005445064 40 2.074454003 0.00032389 0.023274373 41 42 2.054662518 1.38E-06 0.000702275 43 2.008222687 0.00063619 0.035649963 44 2.002474113 8.48E-06 0.002590667 45 1.955295475 6.84E-07 0.000488218 46 1.89526733 2.60E-05 0.00472263 47 48 1.87642956 1.84E-05 0.004008361 49 1.837133962 1.09E-07 0.000222672 50 1.827403125 0.000146834 0.014235937 51 1.767046002 6.03E-05 0.007840177 52 1.762221129 6.50E-06 0.002035366 53 54 1.745220861 8.79E-05 0.009765761 55 1.737338305 0.00038316 0.026043408 56 1.73483868 6.51E-05 0.008112721 57 1.711113474 6.39E-05 0.008047376 58 1.680531557 2.37E-05 0.004503355 59 60 1.619525415 2.19E-05 0.004369047 1.611339308 1.15E-05 0.003096475 Journal of Investigative Dermatology Page 48 of 62

1 2 1.597937324 0.000179746 0.016265027 3 1.544410476 2.40E-05 0.004503355 4 1.534666214 8.90E-05 0.009800224 5 1.523816094 4.64E-06 0.001666504 6 7 1.510789858 0.000357379 0.02500654 8 1.502862096 1.17E-05 0.003096475 9 1.502339718 0.000287129 0.02178617 10 1.498961218 0.000514936 0.030987477 11 1.49789933 0.00051133 0.030922786 12 13 1.492253759 0.000198352 0.017278784 14 1.476740184 0.0001245 0.012569356 15 1.470266601 0.000132769 0.013079894 16 1.458655659 0.000609295 0.034840457 17 1.458276278 0.000362168 0.025137738 18 19 1.443433488 2.67E-05 0.00472263 20 1.441671885 0.000999832 0.04644086 21 1.440837799 1.31E-06For0.000698126 Review Only 22 1.439279424 0.00065625 0.036150616 23 1.433990899 0.000893309 0.043625102 24 25 1.433983216 4.50E-07 0.000419908 26 1.427949724 0.000732806 0.039091511 27 1.405226179 0.000774575 0.040675997 28 1.399872758 8.08E-05 0.0093955 29 30 1.397288889 2.15E-05 0.004369047 31 1.396735209 4.14E-05 0.006279419 32 1.372119138 2.50E-05 0.004621926 33 1.365253514 2.06E-05 0.004337684 34 1.354692662 0.000868531 0.043120296 35 36 1.344909309 0.000275355 0.021289443 37 1.326831635 0.000113672 0.012041279 38 1.325405223 3.78E-06 0.001444593 39 1.32482501 9.29E-06 0.002766958 40 1.317629758 4.49E-05 0.006399585 41 42 1.311304513 1.02E-06 0.000567181 43 1.309580254 2.98E-05 0.00505346 44 1.302366152 0.00050729 0.030846181 45 1.280619965 0.000170584 0.015551166 46 1.279120204 7.63E-05 0.009067873 47 48 1.278234659 1.39E-05 0.003276561 49 1.272582865 0.000126204 0.012636938 50 1.268234096 3.52E-05 0.005449044 51 1.267529814 1.60E-05 0.003564108 52 1.266669793 3.50E-05 0.005449044 53 54 1.252416073 7.44E-05 0.009067873 55 1.220691205 0.000280986 0.021588219 56 1.213561027 4.43E-05 0.006399585 57 1.212456821 0.000120418 0.012466353 58 1.2106128 0.000284576 0.021727356 59 60 1.20790437 0.00010958 0.011809536 1.200638097 0.000217577 0.018330454 Page 49 of 62 Journal of Investigative Dermatology

1 2 1.192035868 0.000313643 0.022942887 3 1.185436073 0.000486149 0.03050535 4 1.18052117 2.88E-05 0.004960046 5 1.167068583 0.000823488 0.041840852 6 7 1.165649202 5.81E-05 0.00762578 8 1.162588266 1.31E-05 0.003248412 9 1.130480611 0.000557197 0.032882701 10 1.127423802 0.000397318 0.02637844 11 1.120286036 0.000210641 0.018090676 12 13 1.092510749 1.12E-05 0.003096475 14 1.088511452 0.000496936 0.03050535 15 1.077181925 8.54E-05 0.009744028 16 1.071297761 0.00111444 0.049868115 17 1.052259363 8.45E-05 0.009739882 18 19 1.051622007 0.000245605 0.019609877 20 1.047786872 0.000229388 0.018806751 21 1.047490268 0.000548996For0.032555971 Review Only 22 1.041869796 0.000495858 0.03050535 23 1.038312831 0.000870785 0.043120296 24 25 1.038295865 0.000143683 0.014041867 26 1.037956611 0.000224668 0.018544223 27 1.037285357 1.78E-07 0.000310716 28 1.023918066 0.000829156 0.041840852 29 30 1.012704298 0.000486544 0.03050535 31 1.008040885 2.20E-05 0.004369047 32 1.003385229 0.000532886 0.0317548 33 0.998373457 0.000198592 0.017278784 34 0.991076103 1.31E-05 0.003248412 35 36 0.987986558 0.000471187 0.029979297 37 0.983658892 0.000488411 0.03050535 38 0.982310352 0.000586514 0.033796465 39 0.970905226 0.000997495 0.04644086 40 0.965919995 8.00E-05 0.009395055 41 42 0.940478444 0.000318401 0.023152279 43 0.928607387 0.000123543 0.012569356 44 0.925168614 0.000167536 0.015504698 45 0.913034414 0.000254294 0.020041653 46 0.912148836 0.000812271 0.041517588 47 48 0.91128109 0.001104422 0.049784564 49 0.905923093 6.60E-05 0.008147868 50 0.904153596 0.000993786 0.04644086 51 0.878907907 0.000391563 0.02637844 52 0.877180454 1.36E-05 0.003258484 53 54 0.869897453 0.000628592 0.035649963 55 0.869149679 0.000117527 0.01227099 56 0.868483281 8.73E-05 0.009765761 57 0.86503306 1.60E-05 0.003564108 58 0.860653865 1.33E-05 0.003248412 59 60 0.858895586 0.000103644 0.01130458 0.857743411 2.22E-05 0.004369047 Journal of Investigative Dermatology Page 50 of 62

1 2 0.840318544 2.65E-05 0.00472263 3 0.838048423 0.000967808 0.045647637 4 0.836908327 1.25E-05 0.003248412 5 0.833798743 0.000924866 0.044395601 6 7 0.832350065 0.000373172 0.025755225 8 0.830986669 0.000153733 0.014446194 9 0.823932363 0.000829064 0.041840852 10 0.811593154 0.000571785 0.033268441 11 0.802449754 0.000528426 0.031643362 12 13 0.795148856 0.000455724 0.029147276 14 0.784017801 0.000752002 0.039941094 15 0.778843151 0.000779788 0.040708915 16 0.773683913 0.000231977 0.018892219 17 0.769938634 5.05E-05 0.006803082 18 19 0.751345287 0.000448603 0.028861116 20 0.738685074 7.65E-05 0.009067873 21 0.737614314 0.000122092For0.012533433 Review Only 22 0.736538584 0.000189075 0.016737288 23 0.729353987 0.000656961 0.036150616 24 25 0.725850843 2.02E-06 0.000989427 26 0.709501528 0.000492613 0.03050535 27 0.705399798 0.00018206 0.016353292 28 0.702990532 6.18E-05 0.007946945 29 30 0.697778501 4.87E-05 0.006686955 31 0.694168739 0.000242001 0.019449252 32 0.68711756 0.000801578 0.041425218 33 0.668810281 6.25E-05 0.007946945 34 0.665362102 0.000114341 0.012041279 35 36 0.661466083 0.000188423 0.016737288 37 0.655366038 3.75E-07 0.000381997 38 0.650048481 0.000997671 0.04644086 39 0.643872457 0.001078789 0.049173475 40 0.635180714 0.000322153 0.023274373 41 42 0.634002664 3.27E-05 0.005398871 43 0.626735822 0.000507538 0.030846181 44 0.625004267 0.000110207 0.011809536 45 0.615612791 4.85E-05 0.006686955 46 0.596116122 9.53E-06 0.002770752 47 48 0.584110058 0.001004416 0.046477048 49 0.583286873 2.88E-05 0.004960046 50 0.580393657 0.000394962 0.02637844 51 0.55931579 0.000337492 0.024073345 52 0.558459135 0.000808486 0.041497774 53 54 0.553808905 0.000291449 0.021895219 55 0.547963099 0.000787092 0.040915392 56 0.543945782 0.000896357 0.043625102 57 0.543479462 0.000293581 0.021895219 58 0.542717998 0.00043665 0.028372938 59 60 0.542113635 0.000382774 0.026043408 0.536302284 4.26E-05 0.006342136 Page 51 of 62 Journal of Investigative Dermatology

1 2 0.520407717 0.000922971 0.044395601 3 0.513882555 4.51E-05 0.006399585 4 0.501502866 0.000855325 0.042822351 5 -0.5037905 0.000666236 0.036496611 6 7 -0.506270024 0.000887589 0.04354535 8 -0.515947183 0.000803682 0.041425218 9 -0.547227699 4.31E-05 0.006342136 10 -0.575412769 0.000358231 0.02500654 11 -0.593530531 0.000293944 0.021895219 12 13 -0.59705505 0.000340921 0.024073345 14 -0.613373428 0.000267114 0.020917076 15 -0.630243096 0.000571903 0.033268441 16 -0.634790609 0.000922404 0.044395601 17 -0.675683931 0.000775827 0.040675997 18 19 -0.701629022 1.39E-09 8.50E-06 20 -0.70236116 1.60E-05 0.003564108 21 -0.807600057 7.63E-05For0.009067873 Review Only 22 -0.954345541 0.000768664 0.040649346 23 -1.043941491 0.000688563 0.03738438 24 25 -1.103321241 0.000634441 0.035649963 26 -1.105324054 0.000491941 0.03050535 27 -1.148689702 0.000150584 0.014371374 28 -1.299678648 4.99E-06 0.001709095 29 30 -1.301153916 0.000221208 0.018508707 31 -1.331327558 3.45E-05 0.005449044 32 -1.377483528 5.07E-05 0.006803082 33 -1.438401712 3.56E-06 0.001444593 34 -1.452242814 0.000698563 0.037759519 35 36 -1.465532635 0.00093954 0.044486117 37 -1.475580157 0.000718722 0.03867799 38 -1.717932525 3.15E-05 0.005268772 39 -2.019043881 0.000418001 0.027306408 40 -3.637480697 0.000678806 0.037019155 41 42 -3.768072503 4.43E-12 5.42E-08 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Investigative Dermatology Page 52 of 62

1 2 Table S2 C. Upstream regulator enrichment among genes over-expressed in GPP whole-blood 3 4 Upstream regulator Predicted activation state FDR 5 STAT3 8.0238E-15 6 7 STAT1 Activated 1.1303E-13 8 IRF7 Activated 7.3474E-13 9 CEBPA 2.4525E-08 10 TRIM24 Inhibited 6.1363E-08 11 RELA Activated 1.1936E-07 12 13 IRF3 Activated 1.6502E-07 14 IRF1 Activated 3.1462E-06 15 IRF5 Activated 2.3399E-05 16 NKX2-3 Inhibited 2.6842E-05 17 NFKBIA Activated 3.4256E-05 18 19 STAT2 7.3124E-05 20 TCL1A 7.4640E-05 21 CEBPB ActivatedFor Review8.9413E-05 Only 22 CNOT7 1.8362E-04 23 SPI1 1.8362E-04 24 25 TP53 2.4514E-04 26 IRF9 2.6835E-04 27 SMARCA4 Activated 4.9188E-04 28 FOS 7.6835E-04 29 30 NFATC2 Activated 8.0454E-04 31 ETS1 8.5360E-04 32 STAT6 8.8044E-04 33 CEBPE 9.7654E-04 34 NLRC5 1.2382E-03 35 36 BRCA1 Activated 1.6286E-03 37 IRF2 1.7947E-03 38 JUND 1.9732E-03 39 NFKB1 2.3325E-03 40 JUN 2.4945E-03 41 42 IRF8 3.0380E-03 43 ZBTB16 3.3789E-03 44 CREBBP 3.7731E-03 45 HIF1A 4.0893E-03 46 STAT4 Activated 4.1511E-03 47 48 CREM 4.3906E-03 49 NUPR1 4.4706E-03 50 ATF1 4.9305E-03 51 CEBPD 5.6556E-03 52 ID3 7.3742E-03 53 54 PHB2 8.9400E-03 55 JUNB 1.0763E-02 56 MYC 1.0843E-02 57 EBF1 1.2467E-02 58 PML 1.2637E-02 59 60 PMF1/PMF1-BGLAP 1.2637E-02 NCOA3 1.3057E-02 Page 53 of 62 Journal of Investigative Dermatology

1 2 ZFPM1 1.3805E-02 3 TRPS1 1.4479E-02 4 SMAD4 1.5193E-02 5 NCOA2 1.5573E-02 6 7 IRF4 1.5971E-02 8 SP1 Activated 1.7128E-02 9 PPARGC1A 1.8204E-02 10 ECSIT 1.8204E-02 11 CREB1 Activated 1.8703E-02 12 13 NOTCH1 1.9278E-02 14 KLF4 1.9692E-02 15 ZFP36 1.9692E-02 16 ZNF366 1.9692E-02 17 MED6 1.9692E-02 18 19 SATB1 1.9692E-02 20 NCOR1 2.0034E-02 21 ETV4 For Review2.0567E-02 Only 22 HOXC8 2.0567E-02 23 ID2 2.1702E-02 24 25 FOXO3 2.3053E-02 26 FOSB 2.4312E-02 27 GFI1 2.4312E-02 28 GLI2 2.4312E-02 29 30 TRIM66 2.4312E-02 31 TRIM32 2.4312E-02 32 ZNF350 2.4312E-02 33 CIITA 2.4312E-02 34 HNF4A 2.5951E-02 35 36 TFEB 2.6044E-02 37 CDKN2A 2.6852E-02 38 ERG 2.7194E-02 39 IRF6 2.7782E-02 40 BCL3 2.8059E-02 41 42 EPAS1 2.8059E-02 43 ZXDC 2.8059E-02 44 TAF12 2.8059E-02 45 BCL6 2.9096E-02 46 SREBF1 2.9829E-02 47 48 GATA3 3.2034E-02 49 FOXO4 3.2552E-02 50 GPS2 3.2552E-02 51 TAF9 3.2552E-02 52 CITED1 3.2552E-02 53 54 GTF3A 3.2552E-02 55 HNF1A 3.4159E-02 56 VHL 3.6151E-02 57 MAF 3.6151E-02 58 KLF3 3.6151E-02 59 60 SMAD7 3.6151E-02 NKX3-1 3.6151E-02 Journal of Investigative Dermatology Page 54 of 62

1 2 CCAR1 3.6151E-02 3 TRIM25 3.6151E-02 4 YWHAB 3.6151E-02 5 MKL1 3.6939E-02 6 7 KLF5 3.7107E-02 8 SMAD3 3.8168E-02 9 HDAC10 3.9873E-02 10 SNW1 3.9873E-02 11 LMO1 3.9873E-02 12 13 FLI1 4.1317E-02 14 MED14 4.3070E-02 15 HBP1 4.3070E-02 16 ATF7 4.3070E-02 17 E2F1 4.3768E-02 18 19 ATF4 4.5493E-02 20 21 For Review Only 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 55 of 62 Journal of Investigative Dermatology

1 Table S2 C. Upstream regulator enrichment among genes 2over-expressed in GPP whole-blood 3 4 Target molecules 5 BCL3,CCR1,CEBPB,CEBPD,CTSB,CTSL,FCGR1A,GBP5,IFI6,IFIT2,IFIT3,IFITM3,IL1R1,IRF1,MMP9,MX2,NAMPT,OASL,PLAUR,PLSCR1,PSMB8,PSMB9,SOCS3,TAP1 6 7 CD14,CEBPD,FCGR1A,GBP5,IFI6,IFIT2,IFIT3,IFITM3,IL1R1,IRF1,LY96,MMP9,OASL,PARP9,PSMB8,PSMB9,SERPING1,SOCS3,TAP1 8 FCGR1A,GBP5,IFI6,IFIT2,IFIT3,IFITM3,IRF1,MX2,NAMPT,NMI,OASL,PLSCR1,PSMB8,PSMB9,TAP1 9 ASPRV1,CCR1,CD14,CEBPB,CEBPD,FCAR,FCGR1A,IFI6,LST1,mir-223,NFIL3,SECTM1,SOCS3,TNFAIP6,TSC22D3 10 IFIT2,IFIT3,IRF1,NMI,OASL,PSMB8,PSMB9,SHISA5,TAP1 11 BCL3,CD14,CEBPB,CTSB,DUSP1,IRF1,MMP9,NAMPT,NMI,NOD2,PSMB9,SCO2,TAP1,TREM1 12 13 B4GALT5,FCGR1A,GBP5,IFI6,IFIT2,IFIT3,IFITM3,IRF1,MMP9,OASL,TAP1 14 FPR2,IFIT2,IFIT3,IFITM3,IRF1,MMP9,PSMB8,PSMB9,TAP1 15 IFIT2,IFIT3,IFITM3,NAMPT,OASL,PLSCR1 16 CSRNP1,DDX60L,F2RL1,PARP9,PLSCR1,PSMB8,PSMB9,TAP1 17 BCL3,CCR1,CEBPB,CEBPD,CTSB,IFI6,IRF1,MMP9,NINJ1,NOD2,SAT1,SOCS3 18 19 IFI6,IFIT2,IFIT3,IRF1,PSMB8 20 CCR1,CFP,CTSB,IL1R2,MMP9,NEU1 21 CD14,CEBPB,CEBPD,FCAR,GLIPR2,IFITM3,mir-223,PLAUR,SAT1,SOCS3,TNFAIP6For Review Only 22 IFI6,PLSCR1,PSMB8,TAP1 23 CD14,FUT7,IL1R2,mir-223,NCF1,PSMB8,PSMB9,TREM1 24 25 APOL1,BCL3,CEBPB,CTSB,DUSP1,H2AFY,MMP9,NAMPT,NCF1,NFAM1,NINJ1,PLAUR,SAT1,SCO2,SEC61B,SERPING1,SHISA5,TAP1,TGFA,TSC22D3 26 IFIT2,IFIT3,IRF1,SOCS3 27 CCR1,CEBPB,CFP,CTSB,IFITM3,IRF1,MXD1,PLAUR,PSMB9,TAP1,TREM1 28 CD14,CTSB,MMP9,MXD1,NFIL3,PLAUR,SOCS3,STX11,TREM1,TXN 29 30 IFIT2,IFIT3,IRF1,NMI,OASL,SOCS3 31 B4GALT5,CD14,MMP9,NCF1,NFIL3,TGFA,WAS 32 BCL3,CTSB,FCGR1A,IFIT3,IRF1,MMP9,NFIL3,PLSCR1 33 CD14,CEBPB,MMP9,NCF1 34 PSMB9,TAP1 35 36 IFI6,IFIT2,IFIT3,PLSCR1,SAT1,TAP1 37 IRF1,PSMB8,PSMB9,TAP1 38 BCL3,MMP9,PLAUR,SAT1 39 BCL3,CTSB,DUSP1,IRF1,MMP9,NMI,NOD2 40 BCL3,CD14,CTSL,DUSP1,MMP9,PLAUR,SOCS3,TXN 41 42 CD14,IFI6,IFIT2,IFIT3,MMP9 43 CD14,F2RL1,LY96,MMP9,TSC22D3 44 DUSP1,FUT7,IRF1,mir-223,MMP9,PSMB9,SOCS3 45 ERGIC1,FUT7,MMP9,PLAUR,RAB20,SLC16A3,SOCS3,TGFA 46 BCL3,IFIT2,IRF1,RNF130,SAT1,SOCS3 47 48 CEBPB,CSRNP1,DUSP1,FUT7,NFIL3 49 CEBPB,IFIT2,IL13RA1,MX2,MXD1,NFIL3,PARP9,RAB20,SAT1 50 CEBPB,FUT7 51 CD14,CEBPB,CEBPD,TNFAIP6 52 BCL3,CEBPB,DUSP1,MMP9,SOCS3 53 54 SOCS3,TXN 55 BCL3,DUSP1,MMP9,PLAUR 56 ASPRV1,CEBPD,CTSB,DUSP1,MMP9,NMI,PLAUR,PLSCR1,PSMB8,SAT1,SLC16A3,TXN 57 CEBPB,IRF1,NFIL3,SOCS3 58 PSMB8,PSMB9,TAP1,TXN 59 60 SAT1 IRF1,MMP9,SLC16A3 Journal of Investigative Dermatology Page 56 of 62

1 2 CEBPB,IL10RB,IL1R1 3 GALNT3,SERPING1 4 FUT7,NAMPT,PILRA,SHISA5,TNFAIP6 5 IFIT2,OASL,TSC22D3 6 7 IL1R1,IRF1,PLAUR,PLSCR1 8 B4GALT5,CEBPB,CEBPD,FPR2,IFITM3,IRF1,MMP9,PLAUR 9 CEBPB,CEBPD,DGAT1,SCO2,SOCS3 10 BCL3,PLAUR 11 CEBPB,CEBPD,CSRNP1,DUSP1,FUT7,NFIL3,SAT1,TXN 12 13 CD14,CEBPB,CEBPD,IRF1,NCF1 14 ASPRV1,CD14,DUSP1,IFITM3,PLAUR 15 IRF1,MXD1,PLAUR 16 TSC22D3 17 IFIT2 18 19 FCGR1A,IL1R2,TGFA,TSC22D3 20 BCL3,DUSP1,MMP9 21 B4GALT5,MMP9 For Review Only 22 IL1R2,SLC16A3 23 BCL3,CEBPB,DUSP1,SOCS3 24 25 CTSB,MMP9,MXD1,NAMPT,SCO2 26 CEBPB,MMP9 27 BCL3,IL1R1,IRF1 28 CCR1,MMP9,TGFA 29 30 IFIT2 31 IFIT2 32 MMP9 33 MMP9 34 ASPRV1,CEBPB,CEBPD,CEP19,CTSB,CYP27A1,DSC2,NAMPT,NOD2,PARP9,PLSCR1,SEC61B,SHISA5,STK19,TMEM140,TNFAIP6,TXN 35 36 CTSB,NEU1 37 CEBPB,CEBPD,DUSP1,IL1R2,PSMB9 38 MMP9,NFIL3,PLAUR,TSC22D3 39 CD14,FPR2 40 IRF1,PLAUR 41 42 CEBPB,NAMPT,NFIL3,TGFA 43 CD14 44 IRF1 45 BCL3,FUT7,IL13RA1,SOCS3 46 CEBPB,CEBPD,IL1R2,NFIL3 47 48 ASPRV1,CCR1,mir-223,NFIL3 49 MMP9,NAMPT 50 RAB20,SHISA5 51 IRF1 52 TGFA 53 54 MMP9 55 CYP27A1,FUT7,SERPING1,SOCS3,TMEM140,VNN3 56 IFITM3,SLC16A3,TGFA 57 IL1R1,TXN 58 BCL3,CFP,GLIPR2,LY96,MYADM 59 60 MMP9,TGFA,TXN CTSB,MMP9 Page 57 of 62 Journal of Investigative Dermatology

1 2 CEBPD 3 IFIT2 4 DUSP1 5 CD177,MMP9,WAS 6 7 DUSP1,MMP9 8 FPR2,MMP9,MXD1,TGFA 9 MMP9 10 MMP9 11 mir-223 12 13 FCAR,NEU1 14 DUSP1 15 CTSB 16 DUSP1 17 CTSB,DUSP1,mir-223,NMI,PSMB9,TAP1 18 19 CEBPB,CYP27A1,FUT7 20 21 For Review Only 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Investigative Dermatology Page 58 of 62

1 2 Table S2 D. Upstream regulator enrichment among genes over-expressed in GPP neutrophils 3 4 Upstream regulator Predicted activation state FDR 5 IRF7 Activated 1.33E-56 6 7 STAT1 Activated 2.08E-39 8 IRF3 Activated 4.28E-37 9 TRIM24 Inhibited 3.23E-35 10 CNOT7 Inhibited 2.00E-28 11 STAT3 2.91E-28 12 13 NKX2-3 Inhibited 2.55E-27 14 IRF1 Activated 9.85E-26 15 IRF5 Activated 4.61E-24 16 STAT2 Activated 1.10E-16 17 IRF2 2.66E-12 18 19 IRF9 2.66E-12 20 IRF8 1.26E-09 21 STAT6 For Review 5.19E-09Only 22 NFATC2 Activated 4.29E-08 23 BRCA1 1.87E-06 24 25 MSC Activated 2.69E-06 26 IRF4 Inhibited 2.32E-05 27 SATB1 5.48E-05 28 CREBBP 0.000553391 29 30 IFI16 Activated 0.001453882 31 NLRC5 0.001880637 32 PRDM1 Inhibited 0.002046261 33 BCOR 0.002100147 34 IKZF3 0.002100147 35 36 RELA Activated 0.0024272 37 ATF3 0.002880714 38 POU2AF1 0.004090787 39 TCF4 0.004315978 40 TCF3 0.005973402 41 42 STAT4 Activated 0.00788597 43 SMARCB1 Activated 0.008132673 44 EP300 0.009566182 45 NUPR1 Inhibited 0.009663529 46 TFAP2B 0.012428122 47 48 IKZF1 Inhibited 0.013090345 49 PHB2 0.014760458 50 TFAP2D 0.014760458 51 ATF4 0.017631769 52 CEBPA 0.019560839 53 54 SPI1 0.020125952 55 MYC 0.02235102 56 ZFP91 0.023780805 57 MED6 0.023780805 58 SMARCA4 Activated 0.024255556 59 60 EZH2 0.02862806 POU2F1 0.02862806 Page 59 of 62 Journal of Investigative Dermatology

1 2 TP53 Activated 0.031419672 3 TRIM66 0.031419672 4 TRIM32 0.031419672 5 VSX2 0.031419672 6 7 ELOA 0.031419672 8 TP73 0.032969106 9 NCOA2 0.033180645 10 KLF3 0.033180645 11 KLF2 0.046343779 12 13 14 15 16 17 18 19 20 21 For Review Only 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Investigative Dermatology Page 60 of 62

1 Table S2 D. Upstream regulator enrichment among genes over-expressed2 in GPP neutrophils 3 4 Target molecules 5 C5,CARD16,CMPK2,DDX58,DHX58,FCGR1A,GBP1,GBP5,HERC5,IFI16,IFI35,IFI44,IFI44L,IFI6,IFIT1,IFIT2,IFIT3,IRF7,ISG15,ISG20,LILRA5,MX1,NT5C3A,OAS1,OAS2,OAS3,OASL,PARP12,PHF11,PLSCR1,PSMB9,RSAD2,RTP4,SAMD9L,SOCS1,STAT1,STAT2,TAP1,TDRD7,TNFSF13B,TOR1B,TRIM22,UBE2L6,USP18,XAF1,ZBP1,ZC3HAV1 6 7 APOL6,BATF2,CASP1,CD274,CMPK2,DDX60,EIF2AK2,FCGR1A,GBP1,GBP1P1,GBP2,GBP5,GBP6,HERC6,IFI16,IFI35,IFI6,IFIT1,IFIT2,IFIT3,IRF7,ISG15,LY6E,MX1,NLRC5,OAS1,OAS2,OASL,PARP9,PDCD1LG2,PSMB9,RNF213,RSAD2,RTP4,SAMD9L,SAMHD1,SERPING1,SOCS1,STAT1,STAT2,TAP1,TNFSF13B,TRIM22,TYMP,USP18,XAF1 8 CD274,CMPK2,DDX58,DDX60,DHX58,EIF2AK2,FCGR1A,GBP1,GBP5,IFI16,IFI44,IFI6,IFIT1,IFIT2,IFIT3,IRF7,ISG15,ISG20,NLRC5,NT5C3A,OAS1,OAS2,OAS3,OASL,PARP12,PHF11,PML,RSAD2,SAMD9L,STAT1,STAT2,TAP1,TDRD7,UBE2L6,USP18,ZBP1,ZC3HAV1 9 AGRN,CMPK2,DDX58,DDX60,DHX58,EPSTI1,GBP2,HERC6,IFI35,IFI44,IFIT2,IFIT3,IRF7,ISG15,LGALS3BP,MOV10,OAS1,OASL,PARP12,PHF11,PSMB9,RTP4,SAMD9L,SAMHD1,SOCS1,STAT1,STAT2,TAP1,USP18 10 BST2,CMPK2,HERC6,IFI35,IFI44L,IFI6,IFIT5,ISG15,LGALS3BP,OAS1,OAS2,OAS3,PARP12,PLSCR1,PPM1K,STAT1,TAP1,UBE2L6 11 BST2,C5,CASP1,CASP7,CD274,CMPK2,EIF2AK2,EPAS1,FCGR1A,GBP2,GBP5,GBP6,HERC5,HERC6,IFI16,IFI35,IFI44,IFI6,IFIT1,IFIT2,IFIT3,IFIT5,IL18R1,IRF7,ISG15,ISG20,JAG1,MX1,OAS1,OAS2,OAS3,OASL,PDCD1LG2,PLSCR1,PML,PSMB9,RSAD2,SOCS1,STAT1,STAT2,TAP1,TRIM22,USP18,XAF1 12 13 BATF2,CASP1,CMPK2,DDX58,DDX60,DHX58,EIF2AK2,FBXO6,GBP1,GBP2,GCH1,LY6E,NT5C3A,PARP10,PARP12,PARP9,PLSCR1,PNPT1,PSMB9,RNF213,RTP4,SAMD9L,STAT1,STAT2,TAP1,TRIM22,TYMP,UBE2L6,USP18,XAF1,ZC3HAV1 14 CASP1,CASP7,CD274,EIF2AK2,GBP2,IFI35,IFI44L,IFIT1,IFIT2,IFIT3,IFIT5,IRF7,ISG15,MX1,OAS1,OAS2,PDCD1LG2,PML,PSMB9,RARRES3,RSAD2,SOCS1,STAT1,STAT2,TAP1,TNFSF13B,TRIM22 15 CMPK2,DDX58,DHX58,IFI44,IFIT1,IFIT2,IFIT3,IRF7,ISG15,ISG20,NT5C3A,OAS1,OAS2,OASL,PARP12,PLSCR1,RSAD2,STAT1,STAT2,UBE2L6 16 GBP1,IFI35,IFI6,IFIT1,IFIT2,IFIT3,IRF7,ISG15,MX1,OAS1,OAS2,RSAD2,SOCS1,USP18 17 CASP1,EIF2AK2,GBP1,IFI35,IRF7,ISG15,OAS1,PSMB9,SOCS1,TAP1,TNFSF13B,USP18 18 19 GBP1,IFIT2,IFIT3,IRF7,ISG15,MX1,OAS2,SOCS1,STAT1,STAT2 20 ATF3,DDX58,GBP1,IFI44L,IFI6,IFIT2,IFIT3,ISG15,OAS1,PML,STAT1,STAT2,TNFSF13B 21 EXT1,FCGR1A,GBP2,IFI16,IFIT3,IL18R1,IRF7,ISG15,ISG20,JAG1,LGALS3BP,MOV10,PDCD1LG2,PLSCR1,SOCS1,STAT2,TFECFor Review Only 22 CMPK2,IFIT2,IFIT3,IRF7,ISG15,ISG20,OASL,PML,RSAD2,SOCS1,STAT1,STAT2 23 DDX58,DUSP3,IFI6,IFIT1,IFIT2,IFIT3,IRF7,MX1,PLSCR1,STAT1,TAP1 24 25 DDX60,EPSTI1,IFI44,IFI44L,IFIT1,IRF7,XAF1 26 GBP1,IRF7,ISG15,OAS1,PHF11,PLSCR1,STAT1,STAT2,TNFSF13B 27 CLEC2B,EPAS1,EPSTI1,FCGR1A,IRF7,TNFSF13B,TRIM22,UBE2L6,XAF1 28 ARHGAP24,EPSTI1,FRMD4B,GIMAP4,IL18R1,ISG15,LGALS3BP,OAS3,PSMB9,RSAD2,RTP4,USP18 29 30 DDX58,IFI16,ISG15,OAS1,STAT2 31 DDX58,PSMB9,TAP1 32 AIM2,GPR84,MT2A,RARRES3,RSAD2,TNFAIP6,TNFSF13B,TRIM38 33 DDX58,ISG15,STAT2 34 DDX58,IFIT3,IRF7 35 36 CD274,EPAS1,GBP1,GCH1,IRF7,ISG15,NOD2,OAS2,PSMB9,SCO2,TAP1 37 ATF3,GBP2,GBP6,MT2A,RSAD2 38 IFI44,IFI44L,IFIT3,MX1 39 DDX58,FAM46C,GIMAP4,IFI16,PML,SAMD9L,ZBP1 40 ATF3,DDX58,FAM46C,GIMAP4,IFI16,PML,SAMD9L,SOCS1,ZBP1 41 42 DDX58,IFIT2,IL18R1,ISG15,ISG20,PCGF5,STAT1 43 EIF2AK2,FCGR1A,IFI16,MX1,OAS1,OAS3 44 ARHGAP24,EPSTI1,FRMD4B,GIMAP4,IL18R1,LGALS3BP,OAS3,RSAD2,RTP4,USP18 45 AGRN,ATF3,GBP2,GCH1,IFIT2,LHFPL2,PARP9,SAMHD1,SPATS2L,SPTSSA,ZC3HAV1 46 EPAS1,MT2A 47 48 EXOC3L1,FGF13,IFI16,IL18R1,LHFPL2 49 SOCS1,TXN 50 MT2A 51 ARHGAP24,ATF3,GCH1,JAG1,MT2A 52 FCGR1A,GBP1,GCH1,GPR84,IFI6,ISG15,MT2A,OAS2,TNFAIP6 53 54 GPR84,ISG15,MT2A,PML,PSMB9,TFEC 55 ANXA5,CASP1,CD274,CNP,GBP2,HERC5,IFI16,IFI35,IFIT1,IRF7,OAS1,PLSCR1,PML,RSAD2,TMSB10/TMSB4X,TXN,USP18 56 TNFSF13B 57 IFIT2 58 AIM2,CARD16,CARD17,CASP1,CNP,GBP1,GIMAP2,IFI16,IFIT1,PSMB9,TAP1 59 60 ATF3,EPAS1,IL18R1,PARP12,RARRES3,SOCS1,TRIM38 GBP2,GSTK1,LY6E,SERPING1 Page 61 of 62 Journal of Investigative Dermatology

1 2 ATF3,CASP1,FGF13,GBP1,HERC5,IFI16,IFI35,IRF7,ISG15,MX1,OAS1,PML,RARRES3,SCO2,SERPING1,SSPN,STAT1,TAP1,TMSB10/TMSB4X,TRIM22,XAF1 3 IFIT2 4 IFIT2 5 TFEC 6 7 ATF3 8 ATF3,CASP1,JAG1,PML,SERPING1,TAP1,TRIM22 9 IFIT2,ISG15,OASL 10 ACOT9,ANXA5,CMPK2,GSTK1,NOD1,SP140,TDRD7 11 CNP,EPAS1,MT2A,STAT1 12 13 14 15 16 17 18 19 20 21 For Review Only 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Journal of Investigative Dermatology Page 62 of 62

1 2 3 4 5 NOTICE TO EDITORS AND REVIEWERS: 6 7 8 9 10 11 12 13 The authors have provided large excel 14 15 16 tables (1-2) which are too large to include in 17 18 the .pdf of the manuscript. 19 For Review Only 20 21 22 Please contact the editorial office to 23 24 25 26 request the files to be sent to you - 27 28 29 [email protected]. 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60