Celastrol Attenuates Psoriasiform Inflammation by Targeting the IRF1/GSTM3 Axis

Persistent inflammation and keratinocyte (KC) hyperproliferation are regarded as traits of psoriasis. KCs are considered as the source of inflammatory cytokines and to be modulating the immunophenotype of psoriasis, which are sufficient to trigger psoriasis even with normal immune cells (Armstrong and Read, 2020Armstrong A.W. Read C. Pathophysiology, clinical presentation, and treatment of psoriasis: a review.JAMA. 2020; 323: 1945-1960Crossref PubMed Scopus (779) Google Scholar; Gou et al., 2021Gou C. Ni W. Ma P. Zhao F. Wang Z. Sun R. et al.The endoribonuclease N4BP1 prevents psoriasis by controlling both keratinocytes proliferation and neutrophil infiltration.Cell Death Dis. 2021; 12: 488Crossref PubMed Scopus (9) Google Scholar; Pasquali et al., 2019Pasquali L. Srivastava A. Meisgen F. Das Mahapatra K. Xia P. Xu Landén N. et al.The keratinocyte transcriptome in psoriasis: pathways related to immune responses, cell cycle and keratinization.Acta Derm Venereol. 2019; 99: 196-205Crossref PubMed Scopus (44) Google Scholar). Celastrol (CS), an active ingredient derived from Tripterygium wilfordii Hook F, has been exploited for combating psoriasis toward inhibiting T cell differentiation, KC proliferation, and neutrophil recruitment (Meng et al., 2019Meng S. Sun L. Wang L. Lin Z. Liu Z. Xi L. et al.Loading of water-insoluble celastrol into niosome hydrogels for improved topical permeation and anti-psoriasis activity.Colloids Surf B Biointerfaces. 2019; 182: 110352Crossref PubMed Scopus (68) Google Scholar; Nguyen et al., 2020Nguyen T. Lestienne F. Cousy A. Mengeaud V. Castex-Rizzi N. Effective inhibition of Th17/Th22 pathway in 2D and 3D human models of psoriasis by celastrol enriched plant cell culture extract.J Eur Acad Dermatol Venereol. 2020; 34: 3-9Crossref PubMed Scopus (8) Google Scholar; Thouvenin et al., 2020Thouvenin M.D. Dalmon S. Theunis J. Lauze C. Coubetergues H. Mengeaud V. et al.Tolerance and efficacy of a new celastrol-containing balm as adjunct care in psoriasis.J Eur Acad Dermatol Venereol. 2020; 34: 10-16Crossref PubMed Scopus (4) Google Scholar), although its action mechanism has not been fully elucidated. To interrogate the effect of CS on KCs, we compared the affinity and specificity of CS in different KC cell lines and chose HaCaT cells for the subsequent experiments (Supplementary Figure S1a), which was consistent with a previous report (Zhou et al., 2011Zhou L.L. Lin Z.X. Fung K.P. Cheng C.H. Che C.T. Zhao M. et al.Celastrol-induced apoptosis in human HaCaT keratinocytes involves the inhibition of NF-κB activity.Eur J Pharmacol. 2011; 670: 399-408Crossref PubMed Scopus (43) Google Scholar). As verification of CS induction, a closer look was given to differentially expressed mRNAs in both unstimulated and stimulated conditions. A total of 1,036 differentially expressed mRNAs were identified (Figure 1a and Supplementary Figure S1b) (detailed in Supplementary Table S1). Glutathione metabolism pathway was highly enriched after CS stimulation (Supplementary Figure S1c and d). Studies showed that glutathione decreases in psoriasis plasma and lesions (Oršolić et al., 2014Oršolić N. Skurić J. Dikić D. Stanić G. Inhibitory effect of a propolis on di-n-propyl disulfide or n-hexyl salycilate-induced skin irritation, oxidative stress and inflammatory responses in mice.Fitoterapia. 2014; 93: 18-30Crossref PubMed Scopus (26) Google Scholar; Sorokin et al., 2018Sorokin A.V. Domenichiello A.F. Dey A.K. Yuan Z.X. Goyal A. Rose S.M. et al.Bioactive lipid mediator profiles in human psoriasis skin and blood.J Invest Dermatol. 2018; 138: 1518-1528Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar), and many identified genes in this pathway are critical in determining the pathogenetic risk of psoriasis. GSTM3 is the most remarkably changed mRNA generated from this pathway, whereas IRF1 is predicted to be the upstream target binding to the GSTM3 promoter region (Figure 1a and Supporting Documents 1–3). To ascertain the changes of corresponding genes, we found that CS significantly upregulated the expression of GSTM2, GSTM3, GSTM4, GCLC, and GCLM and downregulated IRF1 (Supplementary Figure S1e). Clinically, we showed the low-expressed GSTM3 and high-expressed IRF1 in human psoriasis specimens (Figure 1b-c and Supplementary Table S2c) (psoriatic lesion has infiltrated CD3+ staining). To validate the findings, we shifted to an in vivo mouse model and fabricated a nanometer-sized CS (nio-CS) gel, which provides CS-sufficient solubility and -sustained release (Supplementary Figure S1f–h). We then noted that nio-CS gel application resulted in GSTM3 upregulation and IRF1 downregulation (Figure 1d and e). To identify the roles of GSTM3 and IRF1 in vitro, we used small interfering RNA to perturb the expressions of GSTM3 and IRF1 (Supplementary Figure S2a and b) (detailed in Supplementary Table S3). M5 cocktail (containing IL-1α, IL-17A, IL-22, oncostatin M, and TNF-α at 2.5 ng/ml) was shown to induce cellular proliferation and inflammation recapitulating numerous features of psoriasis in vitro (Li et al., 2019Li C. Xiao L. Jia J. Li F. Wang X. Duan Q. et al.Cornulin is induced in psoriasis lesions and promotes keratinocyte proliferation via phosphoinositide 3-kinase/Akt pathways.J Invest Dermatol. 2019; 139: 71-80Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar) and exhibited obvious modifications for GSTM3 and IRF1 (Supplementary Figure S2c and d). Cell proliferation was significantly accelerated by GSTM3 knockdown and decreased by IRF1 knockdown (Figure 1f). In addition, immune factors, antimicrobial peptides, and barrier function markers were aggravated after M5 stimulation; meanwhile, worsened conditions were observed after GSTM3 knockdown, and effective mitigation was shown after IRF1 was silenced (Figure 1g and Supplementary Figure S2e and Supplementary Table S4). To characterize the interaction between IRF1 and GSTM3, we found that cell proliferation was increased after IRF1 overexpression, whereas GSTM3 overexpression reversed the hyperproliferation (Figure 1h and Supplementary Figure S2f-g). Consistent with our expectation, overexpression of GSTM3 reversed the abominable effects of IRF1 overexpression (Figure 1i and Supplementary Figure S2h). These data supported a significant negative correlation between IRF1 and GSTM3. To explore the regulatory effects of CS on the IRF1/GSTM3 axis, we conducted functional experiments in HaCaT cells. CS reduces immune and proliferation indexes and increases barrier markers, which were induced by M5 stimulation, whereas GSTM3 knockdown and IRF1 overexpression could reverse the therapeutic effect of CS (Figure 2a and b and Supplementary Figure S3a). Besides, GSTM3 overexpression abolished the attenuated CS efficacy caused by IRF1 overexpression (Figure 2c and d and Supplementary Figure S3b). From the perspective of transcription, CS directly evicts IRF1 from the promoter region of GSTM3 (Figure 2e-g, Supplementary Figure S3c, and Supplementary Table S5). Hence, CS could ameliorate the related phenotypes of HaCaT cells by weakening the binding intensity between IRF1 and GSTM3. The results mentioned earlier prompted us to evaluate the therapeutic potential of CS in the imiquimod-established psoriasis-like mice dermatitis (Figure 2h). Nio-CS application relieves dry and rough skin and prevents abnormal epidermal thickening, rete ridge elongation, hyperkeratosis, and epidermal inflammatory infiltration, with similar or even better curative effects than those of calcipotriol (Figure 2i and Supplementary Figure S4a–c). Concomitantly, we extended the relapse model of psoriasis (Supplementary Figure S4e). On secondary stimulation, mice treated with imiquimod and calcipotriol previously showed severe psoriasiform skin lesions on the last day of administration, whereas those that received previous nio-CS therapy showed mild skin lesions and inflammation (Supplementary Figure S4f–h). Notably, the expression of Gstm3 was decreased in psoriasis-like lesions in comparison with that in the controls but augmented after nio-CS treatment, whereas the opposite trend was shown in Irf1 in the two models (Supplementary Figure S4d and h). Collectively, we proposed that CS is a promising candidate for psoriasis treatment and prevention owing to its direct actions on the IRF1/GSTM3 axis. Our study enriched the understanding of the antipsoriatic mechanisms of CS. Recently, CS has been shown to target GSTM1 in mitigating obesity and metabolic imbalance (Cui et al., 2021Cui Z. Liu Y. Wan W. Xu Y. Hu Y. Ding M. et al.Ethacrynic acid targets GSTM1 to ameliorate obesity by promoting browning of white adipocytes.Protein Cell. 2021; 12: 493-501Crossref PubMed Scopus (7) Google Scholar), whereas GSTM1 is another important predictor of psoriasis (Martinez et al., 2006Martinez F.O. Gordon S. Locati M. Mantovani A. Transcriptional profiling of the human monocyte-to-macrophage differentiation and polarization: new molecules and patterns of gene expression.J Immunol. 2006; 177: 7303-7311Crossref PubMed Scopus (1827) Google Scholar). Although GSTM3 has not been reported in psoriasis until now, it possesses similar substrate specificities and functional characteristics to those of GSTM1 (Hayes and Pulford, 1995Hayes J.D. Pulford D.J. The glutathione S-transferase supergene family: regulation of GST and the contribution of the isoenzymes to cancer chemoprotection and drug resistance.Crit Rev Biochem Mol Biol. 1995; 30: 445-600Crossref PubMed Scopus (3297) Google Scholar). IRF1 modulates the proliferation/differentiation of KCs and immune responses under both physiological and inflammatory conditions (Odanagi et al., 2004Odanagi M. Kikuchi Y. Yamazaki T. Kaneko T. Nakano H. Tamai K. et al.Transcriptional regulation of the 230-kDa bullous pemphigoid antigen gene expression by interferon regulatory factor 1 and interferon regulatory factor 2 in normal human epidermal keratinocytes [published correction appears in Exp Dermatol 2005;14:160].Exp Dermatol. 2004; 13: 773-779Crossref PubMed Scopus (10) Google Scholar). Nevertheless, it is a critical finding that there is a potential relationship between GSTM3 and IRF1 in psoriasis. The molecular mechanism of natural compounds exhibited as multitarget-, multipathway-, and multisystem-integrated actions. Although nio-CS gel achieves curative effects in the onset and maintenance state of psoriasis without marked organ toxicity (Supplementary Figure S5a and b), the standardized acute and chronic toxicity studies remain to be explored. This is the beginning of fundamental theoretical research on CS, and its application potentials abide in further study. Clinical samples were obtained from patients who provided written informed consent, with approval from the Yueyang Hospital (Shanghai, China) ethics committee (number 201929). The use of mice was approved by the Ethics Committee of Shanghai University of Traditional Chinese Medicine (Shanghai, China) (number YYLAC-2019-064-1). Datasets of RNA sequencing related to this article can be found at https://trace.ncbi.nlm.nih.gov/Traces/sra/?study=SRP289876, hosted at Sequence Read Archive database (accession number SRP289876). All the data that support the findings of this study are available from the corresponding author on reasonable request. Le Kuai: http://orcid.org/0000-0002-4747-2473 Yi Ru: http://orcid.org/0000-0001-6828-7704 Quan-gang Zhu: http://orcid.org/0000-0002-0574-4020 Shuang-yi Yin: http://orcid.org/0000-0001-8440-9580 Yong-yong Li: http://orcid.org/0000-0002-6688-5978 Mi Zhou: http://orcid.org/0000-0001-9927-8245 Ying Luo: http://orcid.org/0000-0001-5873-8794 Yue Luo: http://orcid.org/0000-0003-3498-4821 Jian-kun Song: http://orcid.org/0000-0002-2546-5694 Xiao-ya Fei: http://orcid.org/0000-0002-8346-4796 Jing-si Jiang: http://orcid.org/0000-0002-5856-5851 Cong-cong Zhu: http://orcid.org/0000-0002-8040-8924 Bin Li: http://orcid.org/0000-0002-8607-8874 The authors state no conflict of interest. This work was supported by grants from the National Key Research and Development Program of China (number 2018YFC1705305), the National Natural Science Foundation of China (numbers 81973860, 82174383, 81904214, and 82004235), Shanghai Sailing Program (number 21YF1448100), Xinglin Youth Scholar of Shanghai University of Traditional Chinese Medicine (number RY411.33.10), Clinical Transformation Incubation Program in Hospital (lczh2021-05), Three-Year Action Plan (2018‒2020) of Shanghai Municipality for further acceleration of the development of Chinese medicine (numbers ZY[2018-2020]-FWTX-4010 and ZY[2018-2020]-FWTX-1008), Shanghai clinical key specialty construction project (number shslczdzk05001), Shanghai Science and Technology Commission (number 21Y21920101), Innovative Training Program for Graduate Students in Shanghai University of Traditional Chinese Medicine (number JY611.02.03.83), and the Science and Technology Innovation Action Plan of Shanghai Science and Technology Commission (numbers 21S21900900 and 20DZ2255200). Conceptualization: LK, YR; Data Curation: YiL, XyF; Formal Analysis: LK, SyY; Funding Acquisition: BL, LK, YiL, YuL; Investigation: YuL, JsJ; Methodology: JsJ, JkS; Project Administration: BL, YyL, MZ; Resources: QgZ, MZ; Supervision: SyY, BL; Validation: LK, YyL; Visualization: YR, QgZ; Writing - Original Draft Preparation: YiL, CcZ; Writing - Review and Editing: LK, YR, SyY, XyF, BL Clinical samples were obtained from patients who provided written informed consent, with approval from the Yueyang Hospital (Shanghai, China) ethics committee (number 2019-29). Psoriasis samples were collected from the lesional skin of patients, and normal skin samples were collected from excess skin during cosmetic surgeries. Collected tissues were fixed in 70% formalin immediately after excision and processed for paraffin embedding. The clinicopathological characteristics of the patients are shown in Supplementary Table S2. HaCaT cells (300493) were obtained from CLS Cell Lines Service (Eppelheim, Germany) and were grown in DMEM with 10% fetal bovine serum, 100 U/ml of penicillin, and 100 μg/ml of streptomycin. Normal human epidermal keratinocyte (C-12006) was purchased from PromoCell (Heidelberg, Germany) and was grown in Keratinocyte Growth Medium 2 (D-39006, PromoCell). Cells were cultivated in a humidified atmosphere of 5% carbon dioxide in a 37 °C incubator. For treatments, celastrol (CS) (C0869, Sigma-Aldrich, St. Louis, MO) was dissolved in DMSO at a final concentration of 20 mM; the M5 cocktail (IL-1α, IL-17A, IL-22, oncostatin M, and TNF-α; 2.5 ng/ml) was applied to recapitulate the features of psoriasis (Li et al., 2019Li C. Xiao L. Jia J. Li F. Wang X. Duan Q. et al.Cornulin is induced in psoriasis lesions and promotes keratinocyte proliferation via phosphoinositide 3-kinase/akt pathways.J Invest Dermatol. 2019; 139: 71-80Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar). Cell samples received CS solution at different concentrations, whereas untreated samples received an equivalent volume of DMSO. Cell viability and proliferation were tested by cell counting kit-8 assay. Absorbance at 450 nm was determined, and the percent survival of HaCaT cells was measured by calculating the absorbance of wells treated with CS or DMSO. The coding region of the human IRF1 gene (RefSeq NM_002198.3) was synthesized and cloned into HindIII (ER0505, Fermentas, Waltham, MA) and EcoRI (ER0271, Fermentas) sites of pFLAG-CMV-4 (E1775, Sigma-Aldrich). Then the cloned sequence was confirmed by Sanger sequencing. The IRF1-expressing vector pFLAG-CMV-4-IRF1 or its vehicle was transfected into HaCaT cells by electroporation (BTX ECM 830, Thermo Fisher Scientific, Waltham, MA). To elucidate the role of GSTM3 correlated with IRF1, HaCaT cells were further introduced with small interfering RNA oligo targeting GSTM3 or IRF1 (GenePharma, Pallini, Greece) by electroporation. The small interfering RNA sequence is listed in Supplementary Table S3. HaCaT cells were treated with 1.332 μM of CS or DMSO for 48 hours; the cells were lysed with TRIzol for isolation of total RNA. Each sample was subjected to mRNA selection, fragmentation, cDNA synthesis, and library preparation and sequenced by TruSeq Stranded mRNA LT Sample Prep Kit (RS-122-2103, Illumina, San Diego, CA). RNA-sequencing analysis was completed by Shanghai Biochip (Shanghai, China). Differentially expressed mRNAs were selected on the basis of adjusted P < 0.05 and |log2fold change| ≥ 1. Afterward, Gene Ontology enrichment analysis was implemented by Fisher's test, and significantly enriched pathways of these differentially expressed mRNAs were then determined by Kyoto Encyclopedia of Genes and Genomes database. Nanometer-sized CS gel was prepared as follows: a Span 20, Span 60, and cholesterol at the ratio of 3:1:1 (w/w/w) mixture added with CS or not. These substances were dissolved in a chloroform:methanol (2:1) solution. The extracted fraction was dried by rotary evaporation at 45 °C. After that, the dried surfactant film was hydrated with 5 ml double distilled water at 60 °C for 30 minutes with gentle spinning. Finally, the solution was intermittently sonicated by a probe sonicator at 200 W for 4 minutes to form a nanometer-sized gel. The average particle size and polydispersity index were determined using a particle size analyzer (Microtrac S3500, Microtrac MRB, Montgomeryville, PA). The morphology of nanoformulations was evaluated by transmission electron microscopy (Hitachi HT7700, Hitachi, Tokyo, Japan). In vitro drug release study was performed by a dialysis method using PBS containing 20% ethanol (v/v) as the release media. A total of 1 ml of nanometer-sized CS gel or nanometer-sized blank gel was loaded into a dialysis bag (molecular weight cut-off = 10,000 Da) and immersed in 10 ml of the release buffer, followed by gently shaking at 37 °C at 100 r.p.m. for 24 hours. At predetermined time points, aliquots were removed, centrifuged at 10,000 r.p.m. for 5 minutes, and measured. An imiquimod-induced psoriasis-like mouse model was used for in vivo studies. For the efficacy experiment, mice were randomly assigned into three treatment groups (with 62.5 mg/mice imiquimod cream and treated with a different therapeutic strategy 6 hours later for 8 consecutive days): imiquimod group (4 mg/kg nanometer-sized blank gel), CS group (4 mg/kg nanometer-sized CS gel), and calcipotriol group (1 mg/kg calcipotriol). For the recurrence experiment, mice were randomly divided into three treatment groups and treated as mentioned earlier for 8 consecutive days, recovered for 12 days without any drugs, and rechallenged with imiquimod (20.8 mg/mice) for another 10 consecutive days (Chen et al., 2020Chen Y. Yan Y. Liu H. Qiu F. Liang C.L. Zhang Q. et al.Dihydroartemisinin ameliorates psoriatic skin inflammation and its relapse by diminishing CD8+ T-cell memory in wild-type and humanized mice.Theranostics. 2020; 10: 10466-10482Crossref PubMed Scopus (34) Google Scholar). All the animal procedures were approved by the Ethics Committee of Shanghai University of Traditional Chinese Medicine (Shanghai, China) (number YYLAC-2019-064-1). Severity of psoriasis was measured with the adapted human clinical PASI, which was scored on a scale from 0 to 4 (0, none; 1, slight; 2, moderate; 3, marked; and 4, very marked) (Feng et al., 2019Feng L. Song P. Xu F. Xu L. Shao F. Guo M. et al.cis-Khellactone inhibited the proinflammatory macrophages via promoting autophagy to ameliorate imiquimod-induced psoriasis.J Invest Dermatol. 2019; 139: 1946-1956.e3Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar). The ear thickness value has been normalized for the value on the day of treatment and was divided by the value on day 0 (Wolf et al., 2019Wolf A.R. Wesener D.A. Cheng J. Houston-Ludlam A.N. Beller Z.W. Hibberd M.C. et al.Bioremediation of a common product of food processing by a human gut bacterium.Cell Host Microbe. 2019; 26: 463-477.e8Abstract Full Text Full Text PDF PubMed Scopus (33) Google Scholar). Total mRNA was extracted from skin biopsies and HaCaT cells using standard TRIzol protocols. RNA concentration and purity were determined by measurement of absorbance 260/280 ratios with a UV spectrophotometer. Relative quantitative data were analyzed using the 2–ΔΔCT method. The primers are shown in Supplementary Table S4. To explore the transcription factors binding the promoter of GSTM3, online software USCS (http://genome.ucsc.edu/) combined with PROMO (http://alggen.lsi.upc.es/cgi-bin/promo_v3/promo/promoinit.cgi?dirDB=TF_8.3) or AliBaba (http://gene-regulation.com/pub/programs/alibaba2/index.html) were applied. We performed chromatin immunoprecipitation–qPCR on HaCaT cells. Chromatin immunoprecipitation assay was performed as described by Deng et al., 2020Deng Y. Guo W. Li G. Li S. Li H. Li X. et al.Protocatechuic aldehyde represses proliferation and migration of breast cancer cells through targeting C-terminal binding protein 1.J Breast Cancer. 2020; 23: 20-35Crossref PubMed Scopus (11) Google Scholar. The four IRF1 major sites used for chromatin immunoprecipitation–qPCR are listed in Supplementary Table S5. Protein concentrations were measured by a bicinchoninic acid protein assay kit according to the manufacturer's instructions, using anti-GSTM3 (diluted at 1:1,000; 15214-1-AP, ProteinTech, Rosemont, IL) and anti-IRF1 (diluted at 1:1,000; 8478, Cell Signaling Technology, Danvers, MA). Samples were fixed in 4% neutral buffered formaldehyde solution for 48 hours. Tissues were processed in a routine way for histological evaluation and embedded in paraffin, sectioned at 5 μm, and stained with H&E. For immunohistochemistry staining, partial sections were stained with proliferating cell nuclear antigen (diluted at 1/6,400; ab29, Abcam, Cambridge, United Kingdom), NF-κB‒p50 (diluted at 1/600; ab32360, Abcam), CD3 (diluted at 1/3,200; ab16669, Abcam), GSTM3 (diluted at 1/1,600; 15214-1-AP, ProteinTech), and IRF1 (diluted at 1/30; ab8478, Abcam) antibodies. The tissues were developed by diaminobenzidine and mounted into neutral balsam. Images were taken under a digital slice scanner. Five high-power fields (magnification, ×200) were randomly selected, and the number of positive cells was calculated to analyze the average of positive cells in each group. HaCaT cells were inoculated in 24-well plates at a density of 1.0 × 105 cells per well and allowed to grow for 24 hours before transfection. Then, plasmids were constructed with synthesized fragments containing a different number of binding motifs, and cells were cotransfected with internal control plasmids pRL-SV40 and pGL4.27, GSTM3-BS 1‒4#, GSTM3-BS 2‒4#, GSTM3-BS 3‒4#, GSTM3-BS 4#. Each of the groups mentioned earlier was treated with or without CS. Cell lysates were harvested 24 hours after transfection, and then firefly and renilla luciferase activities were detected by the Dual-Luciferase Reporter Assay System (Promega, Madison, WI) on a Berthold AutoLumat LB9507 rack luminometer (Berthold Technologies, Bad Wildbad, Germany). The value of relative luciferase activity showed the firefly luciferase activity normalized to that of renilla for each assay. Significant differences were indicated between the CS treatment groups and the empty vectors. Before statistical analysis, filtration and data normalization were performed. All values in this study are presented as median ± SD. The differences between the two groups were analyzed by the Student's t-test. For multiple comparisons, one-way ANOVA was used to compare multiple groups with each other; two-way ANOVA was applied for the variables that were repeatedly measured at different time points. Statistical analyses were performed using SPSS 22.0 (SPSS, Chicago, IL) and exhibited by GraphPad Prism 5.2 (GraphPad Software, San Diego, CA). Significant differences between or among the groups were conducted by ∗∗∗P < 0.001, ∗∗P < 0.01, and ∗P < 0.05.Supplementary Figure S2Detection of the IRF1, GSTM3, and the changes of indexes. (a, b) Western blot analysis of GSTM3 and IRF1 expression in HaCaT cells transfected with siGSTM3 and siIRF1. siGSTM3-2 and siIRF1-2 were the most efficient small interfering RNAs and were used for follow-up experiments. (c, d) qPCR quantification of GSTM3 or IRF1 in HaCaT cells stimulated with 0, 2.5, and 5 ng/ml concentrations of M5 for 24 hours. (e, h) qPCR quantification of enrichment indicators. (f, g) Western blot analysis of IRF1 and GSTM3 expression in HaCaT cells with IRF1 and GSTM3 OE plasmid. Data represent mean ± SD from three separate experiments. ∗∗∗P < 0.001, ∗∗P < 0.01, and ∗P < 0.05. M5 indicate M5 cocktail (IL-1α, IL-17A, IL-22, oncostatin M, and TNF-α; 2.5 ng/ml). OE, overexpressed; siGSTM3, GSTM3-targeted small interfering RNA; siIRF-1, IRF1-targeted small interfering RNA.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Figure S3Changes of indexes and ChIP-qPCR assays. (a, b) qPCR quantification of enrichment indicators. (c) ChIP-qPCR assays used antibodies specific for IRF1 to prove that IRF1 binds to GSTM3 promoter. Data represent mean ± SD from three separate experiments. ∗∗∗P < 0.001, ∗∗P < 0.01, and ∗P < 0.05. M5 indicates M5 cocktail (IL-1α, IL-17A, IL-22, oncostatin M, and TNF-α; 2.5 ng/ml). bp, base pair; BS, binding site; ChIP, chromatin immunoprecipitation; CS, celastrol; OE, overexpression; siGSTM3, GSTM3-targeted small interfering RNA.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Figure S4Nio-gel application ameliorates and prevents IMQ-induced psoriatic phenotypes in vivo. (a) Photos were taken after 8-day induction. Ear thickness and total PASI were qualified. (b) Images of histological staining and quantification. Bar = 200 and 500 μm. n = 5 samples per group. (c) qPCR quantification of Il17a, Il1β, Il22, and Il23 in mice skin on day 8. (d) qPCR quantification of Gstm3, Irf1, Gstm1, and Gstm4 in mice skin on day 8. (e) Different treatment plans for psoriasis recurrence mice model. (f) Photos were taken after a 30-day induction. Ear thickness and total PASI were qualified. (g) Images of histological staining and quantification. Bar = 200 μm. n = 5 samples per group. (h) qPCR quantification of Il17a, Il1β, Gstm3, and Irf1 in mice skin after a 30-day induction. Data represent mean ± SD from three separate experiments. ∗∗∗P < 0.001, ∗∗P < 0.01, and ∗P< 0.05. Cal, calcipotriol; CS, celastrol; IMQ, imiquimod; nio-CS, nanometer-sized celastrol; nio-gel, nanometer-sized blank gel.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Figure S5Therapeutic efficacy and security testing of nio-CS gel application in IMQ-induced psoriatic mice. (a) ELISA quantification of renal function (urea nitrogen, creatinine, serum β2-microglobulin, and serum uric acid), liver function (alanine aminotransferase, aspartate aminotransferase, total bilirubin, direct bilirubin, and indirect bilirubin), and glycolipid metabolism indexes (total cholesterol, triglyceride, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, fasting blood glucose, and insulin) in mice blood samples from different groups on the last day of induction. n = 5 samples per group. (b) Pathological analysis of liver, kidney, and spleen sections from different groups. Bar = 100 and 250 μm. Data represent mean ± SD from three separate experiments. ∗∗∗P < 0.001, ∗∗P < 0.01, and ∗P < 0.05. IMQ, imiquimod; nio-CS, nanometer-sized celastrol; nio-gel, nanometer-sized blank gel; n.s., not significant.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Supplementary Table S1Top 20 Upregulated and Downregulated mRNAs between Blank and Celastrol-Treated HaCaT CellsIdentificationNameDefinitionLog2(FC)P-ValueTop 20 upregulated mRNAsENSG00000182459TEX19Testis expressed 193.0908521874.21E–143ENSG00000151012SLC7A11Solute carrier family 7 member 113.1797600976.48E–123ENSG00000187134AKR1C1Aldo-keto reductase family 1, member C13.6741597741.18E–113ENSG00000151632AKR1C2Aldo-keto reductase family 1, member C23.5215407844.47E–111ENSG00000204941PSG5Pregnancy specific beta-1-glycoprotein 52.4743796712.06E–90ENSG00000139211AMIGO2Adhesion molecule with Ig-like domain 22.2044209021.03E–69ENSG00000087086FTLFerritin, light polypeptide2.2324328181.06E–68ENSG00000148677ANKRD1Ankyrin repeat domain 12.5760438211.62E–65ENSG00000001084GCLCGlutamate-cysteine ligase, catalytic subunit2.0655105932.77E–63ENSG00000136235GPNMBGlycoprotein (transmembrane) NMB2.0260238483.77E–63ENSG00000151914DSTDystonin1.9922226146.79E–54ENSG00000137801THBS1Thrombospondin 11.9250412152.75E–52ENSG00000180638SLC47A2Solute carrier family 47, member 21.9365700562.37E–51ENSG00000168685IL7RIL-7 receptor2.2727169882.64E–50ENSG00000171903CYP4F11Cytochrome P450 family 4 subfamily F member 112.3692702391.41E–45ENSG00000117525F3Coagulation factor III1.7819359111.54E–43ENSG00000108846ABCC3ATP binding cassette subfamily C member 31.5924978664.52E–42ENSG00000181104F2RCoagulation factor II (thrombin) receptor1.4733115768.03E–40ENSG00000142871CYR61Cysteine-rich, angiogenic inducer, 611.6624794183.90E–39ENSG00000073150PANX2Pannexin 22.176027514.07E–38Top 20 downregulated mRNAsENSG00000149021SCGB1A1Secretoglobin, family 1A, member 1–7.515077924.61E–287ENSG00000162896PIGRPolymeric immunoglobulin receptor–6.3663455511.32E–285ENSG00000213886UBDUbiquitin D–7.3095817862.61E–262ENSG00000173432SAA1Serum amyloid A1–6.6452007383.39E–241ENSG00000163220S100A9S100 calcium binding protein A9–6.2464241663.91E–226ENSG00000148346LCN2Lipocalin 2–3.917033824.38E–193ENSG00000023445BIRC3Baculoviral IAP repeat containing 3–3.6957878423.29E–180ENSG00000134339SAA2Serum amyloid A2–6.5890265581.15E–162ENSG00000019582CD74CD74 molecule, major histocompatibility complex, class II invariant chain–6.0591865687.98E–152ENSG00000185215TNFAIP2TNF-α induced protein 2–3.5498678372.36E–146ENSG00000163739CXCL1Chemokine (C-X-C motif) ligand 1–6.6796907955.10E–130ENSG00000125730C3Complement component 3–4.6843852417.71E–124ENSG00000137501SYTL2Synaptotagmin like 2–3.0097677581.57E–115ENSG00000179776CDH5Cadherin 5, type 2–4.7299347321.64E–109ENSG00000108691CCL2Chemokine (C-C motif) ligand 2–2.9049971991.39E–106ENSG00000196167COLCA1Colorectal cancer associated 1–3.1028302991.15E–96ENSG00000129451KLK10Kallikrein related peptidase 10–2.746141811.39E–95ENSG00000181143MUC16Mucin 16, cell surface associated–3.9494438735.13E–92ENSG00000127954STEAP4STEAP4 metalloreductase–3.5474986487.88E–87ENSG00000118503TNFAIP3TNF-α induced protein 3–3.4290867691.78E–81Abbreviation: ATP, adenosine triphosphate; FC, fold change. Open table in a new tab Supplementary Table S2Characteristics of Patients with or without PsoriasisCaseGenderAge (y)OriginSite001Male51Normal skinNeck002Female34Normal skinBack003Male46Normal skinFront shoulder004Male48Normal skinFace005Female35PsoriasisBack006Male46PsoriasisLeg (left)007Female26PsoriasisArm (left)008Male52PsoriasisBack Open table in a new tab Supplementary Table S3The Sequence of siRNANamePrimer SequencesHuman siGSTM3F: 5'-CCUGGACAACUGAAACAAUTT-3'R: 5'-AUUGUUUCAGUUGUCCAGGTT-3'Human siIRF1F: 5'-CCAACUUUCGCUGUGCCAUTT-3'R: 5'-AUGGCACAGCGAAAGUUGGTT-3'Abbreviations: F, forward; R, reverse; siGSTM3, GSTM3-targeting small interfering RNA; siIRF1, IRF1-targeting small interfering RNA; siRNA, small interfering RNA. Open table in a new tab Supplementary Table S4Primers Used for qPCRNamePrimer SequencesHuman PCNAF: 5'-TGGAGAACTTGGAAATGGAAAC-3'R: 5'-CCGTTGAAGAGAGTGGAGTG-3'Human IL1βF: 5'-AGCCATGGCAGAAGTACCTG-3'R: 5'-CCTGGAAGGAGCACTCATCT-3'Human IL6F: 5'-CTGCTCCTGGTGTTGCCTG-3'R: 5'-TCACCAGGCAAGTCTCCT CA-3'Human TNF-αF: 5'-CCTGTGAGGAGGACGAACAT-3'R: 5'-TCGAAGTGGTGGTCTTGTTG-3'Human NF-κBF: 5'-GCCACCCGGCTTCAGAAT-3'R: 5'-TATGGGCCATCTGTTGGCAG-3'Human GSTM2F: 5'-GAGAAACCAAGTATTTGAGCCCAGC-3'R: 5'-GGGCAGCAGGCTGAGTATG-3'Human GSTM3F: 5'-TCCTGGAGTTCACGGATACCT-3'R: 5'-CCACACATGTTGTGCTTGCG-3'Human GSTM4F: 5'-CGCTTCCTCCCAAAACCTCT-3'R: 5'-TAGCTGAGGCTTCAAAGGGC-3'Human GCLCF: 5'-CGGAGGAACAATGTCCGAGT-3'R: 5'-CCGGCTTAGAAGCCCTTGAA-3'Human GCLMF: 5'-GGGGAACCTGCTGAACTGG-3'R: 5'-TCCAGCTGTGCAACTCCAAG-3'Human IRF1F: 5'-TGGGGATTCCAGCCCTGATA-3'R: 5'-CCATCCACGTTTGTTGGCTG-3'Human CCL2F: 5'-TCTCAAACTGAAGCTCGCACT-3'R: 5'-TGGGGCATTGATTGCATCTGG-3'Human CXCL8F: 5'-AGCTCTGTGTGAAGGTGCAG-3'R: 5'-TCTCAGCCCTCTTCAAAAACTTC-3'Human CXCL10F: 5'-CCACGTGTTGAGATCATTGCT-3'R: 5'-TGCATCGATTTTGCTCCCCT-3'Human S100A7F: 5'-GCTTCCCAGCTCTGGCTTTT-3'R: 5'-ATCGGCGAGGTAATTTGTGC-3'Human S100A8F: 5'-AGCTGTCTTTCAGAAGACCTGA-3'R: 5'-CGTCTGCACCCTTTTTCCTG-3'Human S100A9F: 5'-TCTGCAAAATTTTCTCAAGAAGGA-3'R: 5'-GGCCTCCTGATTAGTGGCTG-3'Human FLGF: 5'-CAGGCTCCTTCAGGCTACATT-3'R: 5'-GGGTCATCTGGATTCTTCAGGA-3'Human FLG2F: 5'-TCATCCAGTTCTGAAGAACCCA-3'R: 5'-GTGACCACGCCTATGCTTCT-3'Human β-ACTINF: 5'-CTGGGACGACATGGAGAAAATC-3'R: 5'-CCCCTCGTAGATGGGCACA-3'Mouse Il17aF: 5'-CTCATCCATCCCCAGTTGATT-3'R: 5'-TCCAGGCTCAGCAGTA-3'Mouse Il22F: 5'-TTGTGCGATCTCTGATGGCT-3'R: 5'-TATCTGCAAGGCTGGCCTCC-3'Mouse Il23F: 5'-ACCAGCGGGACATATGAATCT-3'R: 5'-AGACCTTGGCGGATCCTTTG-3'Mouse PcnaF: 5'-CGAGACCTTAGCCACATTGGA-3'R: 5'-ACAGTGGAGTGGCTTTTGTG-3'Mouse Il1βF: 5'-TGCCACCTTTTGACAGTGATG-3'R: 5'-AAGGTCCACGGGAAAGACAC-3'Mouse Il6F: 5'-CAACGATGATGCACTTGCAGA-3'R: 5'-TGTGACTCCAGCTTATCTCTTGG-3'Mouse Tnf-αF: 5'-CCACCACGCTCTTCTGTCTAC-3'R: 5'-GGGCTACAGGCTTGTCACT-3'Mouse Nf-κbF: 5'-CCCTACGGAACTGGGCAAAT-3'R: 5'-AGCGGAATCGAAATCCCCTC-3'Mouse Gstm1F: 5'-GCTGTTCCATTGCCAAACCC-3'R: 5'-AGGCACTTGGGCTCAAACAT-3'Mouse Gstm3F: 5'-GTTTGCAGGGGACAAGGTCA-3'R: 5'-GCCCCACTGGGCTATCTTAG-3'Mouse Gstm4F: 5'-GTGGAGCTCAAAGGGAGCTT-3'R: 5'-CCATCCCTGGTGACCAAGAC-3'Mouse Irf1F: 5'-CGACACACATCGATGGCAAG-3'R: 5'-TCGTGATCGACGCATGTCAA-3'Mouse Ccl2F: 5'-CACTCACCTGCTGCTACTCA-3'R: 5'-GCTTGGTGACAAAAACTACAGC-3'Mouse Cxcl10F: 5'-CACGTGTTGAGATCATTGCC-3'R: 5'-GAGGCTCTCTGCTGTCCATC-3'Mouse S100a7F: 5'-CACCAAGAGCAACAGACTC-3'R: 5'-TTGTTTTTGTCAGCTGCCCG-3'Mouse S100a8F: 5'-TTCGTGACAATGCCGTCTGA-3'R: 5'-GGCCAGAAGCTCTGCTACTC-3'Mouse S100a9F: 5'-GCCAACAAAGCACCTTCTCAG-3'R: 5'-AAGGTTGCCAACTGTGCTTC-3'Mouse FlgF: 5'-GGCCGCAACTCAACCAAG-3'R: 5'-GATGGATCCGGCCTTTCCAG-3'Mouse Flg2F: 5'-CGGCCAATTCTGAAGAATCCAG-3'R: 5'-ATGTCCATGCTCCTCTCCCT-3'Mouse β-actinF: 5'-CAATTCCATCATGAAGTGTGAC-3'R: 5'-CCACACAGAGTACTTGCGCTC-3'Abbreviations: F, forward; R, reverse. Open table in a new tab Supplementary Table S5The Four Primers for Detecting IRF1 major sitesNameForward (5'‒3')Reverse (5'‒3')Amplicon (bp)pGSTM3-site-1AATTTATAGCACTAAATGCCCACATTGCGTCTATTTGATTCTTCTCTC286pGSTM3-site-2GCTCTCATTTTTCCCTAACCTTTAGCCTTCTCAGAATCACCCTCTA251pGSTM3-site-3GGAGGCACTGTCTACAAAAAAATCCTGAAAGGAAAATACACCAAAT125pGSTM3-site-4CTTGTTTTGTTTCATCCTGTCCCCTTTTTCCCTTTTAGCCTTTA126Abbreviation: bp, base pair. Open table in a new tab Abbreviation: ATP, adenosine triphosphate; FC, fold change. Abbreviations: F, forward; R, reverse; siGSTM3, GSTM3-targeting small interfering RNA; siIRF1, IRF1-targeting small interfering RNA; siRNA, small interfering RNA. Abbreviations: F, forward; R, reverse. Abbreviation: bp, base pair.