STAT2 is involved in the pathogenesis of psoriasis by promoting CXCL11 and CCL5 production by keratinocytes

Abstract

The JAK/STAT signaling pathway is suggested to play an important role in the pathogenesis of psoriasis, and recently JAK/STAT inhibitors have shown promising results in psoriasis treatment. The present study aimed to characterize the role of STAT2 in psoriasis. We demonstrated an increased expression of STAT2 and an increased level of phosphorylated/activated STAT2 in lesional compared with nonlesional psoriatic skin. Gene silencing of STAT2 by siRNA in human keratinocytes revealed that upon IFNα stimulation CXCL11 and CCL5 were the only two cytokines, among 102 analyzed, found to be regulated through a STAT2-dependent mechanism. Moreover, the regulation of CXCL11 and CCL5 depended on IRF9, but not on STAT1 and STAT6. The CXCL11 and CCL5 expression was increased in lesional compared with nonlesional psoriatic skin, and analysis demonstrated positive correlation between the expression of CXCL11 and IFNγ and between the expression of CCL5 and IFNγ in lesional psoriatic skin. In contrast, no correlation between the expression of CXCL11 and IL-17A and the expression of CCL5 and IL-17A in lesional psoriatic skin was found. Our data suggest that STAT2 plays a role in the psoriasis pathogenesis by regulating the expression of CXCL11 and CCL5, and thereby attracting IFNγ-producing immune cells to the skin.

Fig 1. STAT2 expression is elevated in psoriatic skin.

(A) STAT2 mRNA expression was examined by qPCR in biopsies obtained from normal healthy volunteers as well as from paired punch biopsies obtained from nonlesional (NLS) and lesional (LS) skin from patients suffering from psoriasis or atopic dermatitis (AD). The mRNA expression of RPLP0 was used for normalization. Biopsies from 6 healthy volunteers, 16 psoriatic patients and 6 atopic dermatitis patients were examined. The results are presented as dot plots with the horizontal line expressing the mean value. All measurements were performed in triplicates. (B) Whole cell protein extracts were prepared from paired keratome biopsies taken from nonlesional and lesional psoriatic skin from five psoriatic patients. Phosphorylated STAT2 as well as total STAT2 protein was analyzed by western blotting. Equal protein loading was assessed by detecting the protein level of β-actin. Protein extract from keratinocytes stimulated with IFNα for 1 hour was included as a positive control (PC). MW; molecular weight marker. Densitometic analysis of the band intensity was carried out and values were normalized to β-actin. *P < 0.01. (C-E) Immunofluorescence analysis was performed on paraffin-embedded punch biopsies from (C) normal skin as well as (D) nonlesional and (E) lesional psoriatic skin. Nuclear staining was performed using 4’, 6-diamidine-2’-phenylindole dihydrochloride (DAPI). Green color (Alexa Fluor 488) represents STAT2 protein. Three sets of biopsies from three different patients were investigated. Arrows show nuclear staining. Scale bar = 100 μm.

Fig 2. STAT2-mediated regulation of CXCL11 and CCL5 expression.

(A) Cultured human keratinocytes were stimulated with IFNα (1000 U/ml) for the indicated time points, and the phosphorylation level of STAT2 analyzed by western blotting (n = 3). (B) Keratinocytes were transfected with STAT2 siRNA (siSTAT2), control siRNA (siCon) or transfection reagent alone (Mock) before being stimulated with IFNα (1000 U/ml) for 24 hours. Protein extracts were isolated from the cells and STAT2 and P-STAT2 expression analyzed by western blotting (n = 3). β-actin was used as a loading control. (C) Human keratinocytes were transfected with STAT2 siRNA (siSTAT2) or control siRNA (siCon) before stimulation with IFNα (1000 U/ml) for 24 hours. Then the cell culture medium was screened for 102 proinflammatory cytokines and chemokines using a proteome profile array. Numbers on the membrane marks the following targets: 1 (CXCL11), 2 (IL-1ra), 3 (IL-1α), 4 (CCL5), 5 (MCP-1), 6 (IL-8), 7 (CXCL1) and 8 (ENA-78). (D, E) Human keratinocytes were stimulated as in (C) and the expression level of CXCL11 and CCL5 analyzed by ELISA (n = 3). Results are expressed as mean ± standard deviation. *P < 0.05.

Fig 3. IFNα-induced CXCL11 and CCL5 expression involves STAT2 and IRF9, but not STAT1.

Cultured human keratinocytes were transfected with siRNA directed against STAT2 (siSTAT2), STAT1 (siSTAT1), IRF9 (siIRF9), or control siRNA (siCon) before IFNα (1000 U/ml) stimulation for 24 hours. (A) Protein extracts were isolated from the cells and the protein level of STAT2, STAT1, and IRF9 examined by western blotting. β-actin was used as a loading control. One representative gel out of three is shown. (B, C) The protein level of (B) CXCL11 and (C) CCL5 in the cell culture medium was analyzed by ELISA (n = 3). Results are expressed as mean ± standard deviation. *P < 0.05.

Fig 4. Increased mRNA expression of CXCL11 and CCL5 in lesional psoriatic skin.

Total RNA was isolated from biopsies obtained from normal healthy volunteers as well as lesional and nonlesional psoriatic skin. mRNA expression of (A) CXCL11 and (B) CCL5 was analyzed by qPCR. RPLP0 mRNA expression was used for normalization. Scatterplot shows the result from 6 healthy volunteers and 16 psoriatic patients. *P < 0.01.

Fig 5. The expression of CXCL11 and CCL5 is positively correlated with IFNγ expression in lesional psoriatic skin.

(A) Chemotaxis assays were performed using a 24-well plate-based assay. PBMCs were isolated from blood taken from psoriatic patients and loaded on the upper chamber containing a 5-μm pore-size filter. Wells (lower chamber) were added vehicle or 10 nM recombinant CXCL11 or CCL20 and incubated for 20 hours. Cells migrating to the lower chamber were analyzed by flow cytometry (n = 9). Results are expressed as mean ± standard deviation. *P < 0.01. (B-E) RNA from punch biopsies taken from 16 psoriatic patients was isolated and the mRNA expression of CXCL11, CCL5, IFNγ, and IL-17A analyzed by qPCR. The correlation between the expression levels of these cytokines was analyzed as indicated in the figure.