Macrophage P2Y₆R activation aggravates psoriatic inflammation through IL-27-mediated Th1 responses

Abstract

Purinergic signaling plays a causal role in the modulation of immune inflammatory response in the course of psoriasis, but its regulatory mechanism remains unclear. As a member of purinoceptors, P2Y₆R mainly distributed in macrophages was significantly up-expressed in skin lesions from patients with psoriasis in the present study. Here, the severity of psoriasis was alleviated in imiquimod-treated mice with macrophages conditional knockout of P2Y₆R, while the cell-chat algorithm showed there was a correlation between macrophage P2Y₆R and Th1 cells mediated by IL-27. Mechanistically, P2Y₆R enhanced PLC_β/p-PKC/MAPK activation to induce IL-27 release dependently, which subsequently regulated the differentiation of Th1 cells, leading to erythematous and scaly plaques of psoriasis. Interestingly, we developed a novel P2Y₆R inhibitor FS-6, which bonds with the ARG266 side chain of P2Y₆R, exhibited remarkable anti-psoriasis effects targeting P2Y₆R. Our study provides insights into the role of P2Y₆R in the pathogenesis of psoriasis and suggests its potential as a target for the development of therapeutic interventions. A novel P2Y₆R inhibitor FS-6 could be developed as an anti-psoriasis drug candidate for the clinic.

Graphical abstract

Macrophage P2Y₆R enhances release of IL-27 to mediate Th1 cell differentiation in the pathogenesis of psoriasis via PLC_β/p-PKC/MAPK signaling, which could be reversed by a novel P2Y₆R inhibitor FS-6.

1. Introduction

Psoriasis is an immune-mediated, genetic disease and is characterized by erythematous and scaly plaques of the skin. It affects approximately 1.5% of the global population¹. Histologically, psoriatic lesions are characterized by the activation of numerous immune cells infiltrating into the dermis2, 3, 4. Purine metabolism plays a crucial role in the pathogenesis of psoriasis, which is associated in the development and progression of this chronic inflammatory skin disorder^5,6. In the progression of psoriasis, UDP has to be necessarily released extracellularly by damaged cells⁷. Once in the extracellular milieu, this nucleotide may play different roles both by inducing chemotaxis of immune cells and by amplifying the inflammatory response^6,8. More importantly, modulation of purinergic signaling regulates many skin functions, including keratinocyte proliferation, skin repair, and immune response, particularly in inflammasome activation^8,9. Thus, targeting single molecular actors of the purinoceptors may develop new psoriasis treatments.

In the present study, we analyzed gene expression from psoriasis patient skin samples and found that P2Y₆R played a significant role in the progression of psoriasis. To further investigate the mechanisms of P2Y₆R, we examined the spatial localization and expression profile of P2Y₆R and generated mice with conditional knockout of P2Y₆R in macrophages. Our results demonstrated that macrophage P2Y₆R mediated PLC_β/p-PKC/MAPK signaling, enhanced skin inflammation by the release of IL-27, which led to increased Th1 cell differentiation and amplified IFN-γ-mediated skin inflammation via the IL-27RA mediated p-JAK2/p-STAT1/T-bet signaling pathway. Additionally, we sought to identify a potent P2Y₆R inhibitor that could have a direct therapeutic effect on psoriasis. These findings offer a potential treatment strategy for psoriasis through P2Y₆R-targeted therapy.

2. Materials and methods

2.1. Cells

HEK293-hP2Y₆R cells obtained from KeyGEN Biotech (Jiangsu, China, KG345) were cultured in DMEM (KGM12800N-500, KeyGEN, Jiangsu, China) supplemented with 10% FBS (FBSSA500-5, AusGeneX, Shanghai, China), 100 units/mL penicillin, 100 μg/mL streptomycin, and 2 mol/L l-glutamine in a humidified atmosphere of 5% CO₂ at 37 °C.

BMDMs were derived from bone marrow cells isolated from mice femurs and tibias and cultured in DMEM medium (11965092, Gibco, NY, USA) containing 10% FBS (FBSSA500-5, AusGeneX, Shanghai, China), 1% penicillin-streptomycin, and 20% (v/v) L-929 cell-conditioned medium to promote macrophage colony growth. The adherent cells were used for experiments after 7 days of culture at 37 °C under 5% CO₂. BMDMs were extracted and seeded in 24-well plates. After 24 h, 10 μmol/L UDP (BCCB4639, Sigma–Aldrich, Germany) was added for model stimulation; 30 min later, 4 ng/mL IMQ (HY-B0180, MedChemExpress, New Jersey, USA) was added, 12 h later, the supernatant was collected and IL-27 concentration was measured. For IL-27 reverse experiments, BMDMs were stimulated with 100 nmol/L PMA (HY-18739), 5 nmol/L Anisomycin (HY-18982), 20 μmol/L Cigilitazone (HY-W011220) and 10 μmol/L Ceramide C6 (HY-19542, all from MedChemExpress) for 12 h before UDP and IMQ treatment.

Naïve T cells were isolated from the spleens of adult mice using standard protocols. Mice were euthanized and the spleens were dissected. After washing the spleen three times in PBS containing 5% FBS, the cells were filtered through a 70 μm cell strainer to obtain a cell suspension. Following red blood cell lysis, using CD4 magnetic beads (551539, BD Biosciences, New Jersey, USA) to isolate CD4-positive T cells. Subsequently, cells were seeded in a 12-well plate pre-coated with anti-mouse CD3e (BE0001-1-1MG, Bioxcell, New Hampshire, USA). For Th1 cells polarization, 10 ng/mL IL-12 (HY-P70666) and 10 μg/mL anti-IL-4 (HY-P990121, both from MedChemExpress, New Jersey, USA) were added into IMDM complete culture medium (KGM12200-500, KeyGEN, Jiangsu, China). The cells were incubated at 37 °C in a humidified incubator with 5% CO₂ for 5 days. After 5 days of culture, Th1 cells were stimulated with 50 ng/mL PMA (TQ0198, Topscience, Shanghai, China), 500 ng/mL ionomycin (HY-13434), and 10 μg/mL brefeldin A (HY-16592, both from MedChemExpress) incubated for 4.5 h at 37 °C for subsequent flow cytometry analysis. Purified naive CD4 T-cells were cultured with the supernatants from BMDMs at a 1:1 ratio, in the presence of plate-coated α-CD3 (BE0001-1-1MG)/CD28 (BE0015-1-1MG, both from Bioxcell, New Hampshire, USA) antibodies at a concentration of 1 μg/mL, for 3 days. The IFN-γ⁺CD4⁺T-cells were detected by FACS.

2.2. Mice

The P2Y₆R deficient mice (P2Y₆R^−/−) mice in C57BL/6J background were obtained from GemPharmatech Co., Ltd. (Strain No. T052843, Jiangsu, China). P2Y₆R-floxed (P2Y₆R^fl/fl) mice were generated at GenPharmatech Co., Ltd. Lyz2^‒Cre mice (B6. 129P2-Lyz2tm1(cre)Ifo/J; stock No. 04781) were purchased from Jackson Laboratory (Bar Harbor, ME). IL-27ra^−/− mice were given by Assoc. Prof. Qian Wang, (Jinan University, Guangzhou, China) and CD4^‒cre mice were kindly donated by Prof. Xinming Jia (Shanghai Tenth People's Hospital, School of Medicine, Tongji University, Shanghai, China).

All mice were housed in a controlled environment (20 ± 2 °C, 12-h/12-h light/dark cycle), where they were maintained on a standard chow diet with free access to water. The genotypes of the mice were examined via PCR using the total DNA derived from the tail tips (Supporting Information Fig. S1). The primers used are listed in Supporting Information Table S1. The experimental animal facility has been accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC). All mouse experiments conformed to the NIH guidelines (Guide for the Care and Use of Laboratory Animals), and the ethical approvals for the animal experiments were obtained from the Animal Ethics Committee of China Pharmaceutical University (No. AECCPU201902004).

2.3. In vivo experiments

IMQ-induced psoriasis-like skin inflammation was induced by a daily topical dose of 62.5 mg of IMQ cream (5%) (Sichuan Med-Shine Pharmaceutical Co., Sichuan, China) on the shaved back for five consecutive days. One day before the initiation of experiments, mice were shaved along their upper backs using electric clippers. On the indicated days, the skin thickness was measured with a micrometer. In the IL-27 revertant experiments, a subcutaneous administration of either IL-27 recombinant protein (RP00702, Abclonal, Wuhan, China) or saline was conducted at a dosage of 400 ng per day. In the treatment of P2Y₆R inhibitor, FS-6 (at a dosage of either 10 or 20 mg/kg) MTX (at a dosage of 1 mg/kg) was administered via intraperitoneal injection on a daily basis prior to the application of IMQ. The treatment regimen was continued until the completion of the model establishment process.

2.4. Histology, IF and IHC staining

Tissue samples from the dorsal skin were harvested, 4% paraformaldehyde-fixed, and paraffin-embedded. Sections were stained with hematoxylin and eosin stain for histological analysis. For immunofluorescence, sections of mouse skin were stained with anti-mouse CD86 (ab239075, Abcam, Cambridge, UK) or anti-mouse P2Y₆R primary antibody (DF6956, Affinity, Jiangsu, China), followed by Alexa Fluor 488–conjugated secondary antibody (ab150113, Abcam). The expression of Ki67 in the skin was detected by immunohistochemistry. Antigen retrieval was carried out by boiling the sections in sodium citrate solution (pH = 6.0). The sections were washed and blocked with 3% hydrogen peroxide to inhibit endogenous peroxidase. Then, the sections were incubated with a Ki67 primary antibody (ab15580, Abcam) overnight. The following day, the sections were washed with PBS and incubated with a secondary antibody at room. Images were captured using BX53 biological microscope (OLYMPUS) and analyzed by ImageJ software.

2.5. RNA extraction and real-time quantitative PCR

Total RNA was isolated from skin biopsies using RNA isolater Total RNA Extraction Reagent (R401-01, Vazyme, Jiangsu, China). Total RNA was used to prepare cDNA using the HiScript II Q RT SuperMix for qPCR (+gDNA wiper) (R222-01, Vazyme). Real-time PCR was carried out using SYBR Green (AQ602-01, TransGen Biotech, Jiangsu, China) endpoint measurement. Expression was calculated using the ΔC_T method relative to the housekeeping gene GAPDH. Primer sequences are provided in Supporting Information Table S2.

2.6. Flow cytometry

Back skin was cut into small pieces and digested in RPMI–2% fetal calf serum (FCS) supplemented with 1 mg/mL collagenase IV (C4-28-100 MG, Sigma–Aldrich, Missouri, USA) and 100 μg/mL deoxyribonuclease (DNase) I (10104159001, Sigma–Aldrich, Missouri, USA) at 37 °C for 1.5 h. After digestion, the skin tissue was disrupted with a syringe with an 18-gauge needle, filtered through a 70-μm cell strainer, and washed with PBS. Then cells were stained with fluorophore-conjugated antibodies: CD4-FITC (11-0042-85, Invitrogen, CA, USA), CD11c-PE-cy7 (25-0114-82, Invitrogen, California, USA), CD86-FITC (11-0862-82, Invitrogen), NK1.1-PE (12-5941-82, Invitrogen), IFN-γ-PE (12-7311-82, Invitrogen), IL-17A-APC (17-7177-81, Invitrogen), CD45-PE-Cy7 (25-0451-82, Invitrogen), CD207,-Alexa Flour 488 (53-2073-82, Invitrogen), CD163-PE (12-1631-82, California, USA), MHCII-Percp-cy5.5 (562363, BD Pharmingen, New Jersey, USA). After washing, the cells were assayed with a BD FACS Celesta flow cytometer, and the data were analyzed using FlowJo software.

2.7. Immunoblotting

Cells and skin tissues were lysed for immunoblotting in radioimmunoprecipitation assay (RIPA) buffer (P0013B, Beyotime, Shanghai, China) containing PMSF (ST506, Beyotime, Shanghai, China) and phosphatase inhibitors (P1082, Beyotime) at the indicated time points. The following antibodies were used: PLC_β (bs-6976R, Bioss, Beijing, China), PKC (ab181558, Abcam, Cambridge, UK), p-PKC (EP2730Y, Abcam), p38 (AF6456, Affinity, Jiangsu, China), p-p38 (AF4001, Affinity), ERK (bs-0022R, Bioss), p-ERK (bs-3016R, Bioss), p-JAK2 (A19629, Abclonal, Wuhan, China), JAK2 (AP0531, Abclonal), p-STAT1 (bs1657R, Bioss), STAT1 (10144-2-AP, Proteintech, Wuhan, China), p-JNK (bs1640R, Bioss), JNK (AF6318, Affinity), T-bet (14-5825-82, Abclonal), β-actin (T0022, Affinity).

2.8. Enzyme-linked immunosorbent assay (ELISA)

Supernatants of cell culture, mice serum and skin were collected and stored in −70 °C freezer¹⁴. IL-27 and IFNγ were detected by mouse IL-27 ELISA kit (EMC115.96, Neobioscience Technology, Shenzhen, China) and mouse IFNγ ELISA kit (EK280HS, MULTI Sciences, Hangzhou, China) according to the manufacturer's instructions.

2.9. Transfection

Cells were seeded in 6 well plates, and grown to 80% confluence prior to infection. Plasmid (SinoBiological, Beijing, China) were added with diluted Lipofectamine™ 3000 reagent (L3000001, Invitrogen, California, USA) in Opti-MEM™ Medium (31985070, Gibco, CA, USA) according to the manufacturer's instructions. The medium was replaced with fresh medium 6 h after transfection. Cells were cultured for 48 h before treatment with the indicated agents.

2.10. Grating-coupled interferometry (GCI)

P2Y₆R was immobilized on a 4PCP WAVE chip by standard amine coupling chemistry with a flow of 10 μL/min to the desired surface density level. An empty surface was activated/deactivated and used as a reference channel. Throughout all experiments, the running buffer was the PBS-P buffer and 3% DMSO at pH 7.5 (all reagents from Sigma). The kinetic run (all at 60 μL/min) included 20 startup cycles of running buffer injections followed by a 1:2 dilution series of compound FS-6 (eight concentrations, with 100 μmol/L as the highest concentration) with one blank every injection. DMSO calibrations were performed at the end of the series. All the cycles in the kinetic run included 45 s baseline, 60 s association, and 60 s dissociation. No regeneration was required.

Data adjustments and analysis were performed with the Creoptix WAVE control software. Adjustments included DMSO calibration correction, X-offset correction, and blank subtraction using the closest blank. Global fitting with bulk correction for both association and dissociation was used to obtain the kinetic rate constants K_D and R_max. Equilibrium analysis included an offset correction.

2.11. Cellular thermal shift assay (CETSA)

CETSA was performed in accordance with a previous study. Briefly, The HEK293-hP2Y₆R cells were divided into two groups. The collected HEK293-hP2Y₆R cells were incubated with FS-6 for 30 min. Each group was divided into six equal parts and heat-treated at different temperatures for 10 min. Then, the samples were quickly transferred to ice and centrifuged at 12,000 rpm with a cryogenic centrifuge for 10 min. The supernatants were obtained, added with loading buffer, and analyzed using Western blot.

2.12. GSEA, GO, and KEGG analysis

The GEO datasets were acquired from https://www.ncbi.nlm.nih.gov/geo and the analysis was conducted using Rstudio (Version 4.3.1). The identification of differentially expressed genes utilized the GO and KEGG pathway enrichment analyses, available at https://david.ncifcrf.gov. The GSEA analysis was performed using GSEA v3.0 at http://www.broadinstitute.org/gsea/.

2.13. Protein‒protein interaction network

The STRING database was used to identify the protein‒protein interaction (PPI) networks of differentially expressed genes.

2.14. Single-cell RNA seq analysis

The analysis was conducted following the standard procedure outlined in Seurat v3. In the cell quality control (QC) step, cells that expressed less than 200 genes or greater than 7500 genes were excluded, as well as those with more than 5% erythrocyte genes or more than 20% mitochondrial genes. After normalization and principal component analysis (PCA) dimension reduction, cell clustering was performed based on PCA dimensionality reduction using the first 20 principal components and a resolution value of 0.4. For cell annotation, marker genes were manually identified with the assistance of the CellMarker 2.0 database available at http://bio-bigdata.hrbmu.edu.cn/CellMarker. The marker genes for each cluster were determined using the FindAllMarkers function with default parameters.

2.15. Cibersort

CIBERSORT v2.0 was employed for cell type analysis in this study. A reference gene expression profile was obtained from publicly available gene expression datasets, which contained data from multiple known cell types. The reference gene expression profile was then compared with the sample gene expression datasets. To accurately infer the cell type composition, a linear regression model was utilized, and the sample gene expression data was fitted to the reference gene expression profile using the least squares method. In this model, consideration was given to the batch effects of the gene expression data, and necessary corrections were performed.

2.16. Cellchat

Cell‒cell communication analysis was conducted on a single-cell RNA sequencing dataset using the CellChat R package. Ligand‒receptor interactions were identified and quantified for each pair of cell types. Intercellular communication relied on the inferred ligand and receptor profiles, where the expression levels of ligands and receptors were estimated by calculating the geometric mean across individual cells of a specific cell type. These interactions indicate the strengths of interactions between all ligands and their corresponding receptors expressed in the two specified cell types.

2.17. Transcription factor enrichment analysis

ChEA3 (https://amp.pharm.mssm.edu/ChEA3) was used to find and annotate the main biological functions of the TF targets involved in sample gene expression datasets.

2.18. Chemistry

All reagents and solvents were commercially available and were used without further purification. Varian UNITY INOVA 400 and 600 MHz NMR instruments were used to determine the hydrogen and carbon spectra of NMR. TMS was the internal standard at 25 °C. The ESI-MS spectra were recorded on a Xevo G2-XS TOF mass spectrometer. The EI-MS spectra were recorded on a GCT Premier mass spectrometer. The silica gel used for the separation of compound column chromatography (CC) was a 200−300 mesh column chromatography silica gel produced by Qingdao Ocean Chemical Works. The amount of silica gel was 50–100 times the amount of separated samples. The whole elution process was tracked by thin layer chromatography (TLC). The silica gel 60 GF254 produced by Qingdao Ocean Chemical Works was used for thin layer chromatography. The silica gel 60 GF254 was detected by ultraviolet radiation at 254 nm wavelength. The purity of the compound was, analyzed by HPLC (Agilent 1100 Series; Multohyp ODS-5μ, 4.6 mm × 250 mm), and the final compounds exhibited purity greater than 95%. The applied mobile phases were 95% MeOH/5% H₂O. The flow speed was 1.2 mL/min, and injection volumes were 10 or 20 μL. Melting points were determined with Shanghai Jingke Industrial X-4A11 and are uncorrected.

2.19. Molecular dynamics (MD) simulations

The dynamic interaction patterns between compounds A6, FS-6, and P2Y₆R were investigated by the MD simulations²⁹. The general AMBER force field (gaff) and the ff99SBildn force field¹⁸ were used for the two compounds and P2Y₆R, respectively. Each compound was optimized by Gaussian 09 at the HF/6-31G∗ level, and then the atomic partial charges were obtained by fitting the electrostatic potentials using the RESP fitting technique in Amber. The system was neutralized with the counter ions of Na⁺. The whole system was immersed in a rectangular box of TIP3P water molecules¹⁷, and the water box was extended Å from any solute atom. The particle mesh Ewald (PME) method was employed for the long-range electrostatics.

Each system was relaxed by a two-stage minimization protocol: first, the protein was fixed, and the water molecules and ligand were minimized by 500 cycles of steepest descent and 500 cycles of conjugate gradient minimization; second, the whole system was minimized by 5000 cycles (1000 cycles of steepest descent and 4000 cycles of conjugate gradient minimization). Then, 10 ns MD simulation was performed under a target temperature of 300 K and a target pressure of 1 atm. The SHAKE procedure was employed to constrain all bonds involving hydrogen atoms, and the time step was set to 2 fs. Coordinate trajectories were saved every 10 ps. The MM simulations and MD optimizations were accomplished by using the sander program in AMBER11.

2.20. MM/GBSA binding free energy calculations/decompositions

The binding free energy (ΔG_bind) of each compound/P2Y₆R was calculated by the Molecular Mechanics/Generalized Born Solvent Area (MM/GBSA) methodology^29,30 as in Eq. (1):

$$\Delta G_{\text{bind}}=G_{\text{complex}}–G_{\text{protein}}–G_{\text{ligand}}$$

$$=\Delta H+\Delta G_{\text{solvation}}–T\Delta G$$

$$=\Delta E_{\text{MM}}+\Delta G_{\text{GB}}+\Delta G_{\text{SA}}–T\Delta S$$ (1)

where ΔE_MM represents the gas-phase interaction energy between the protein and ligand, which contains the electrostatic (ΔE_ele) and van der Waals (ΔE_vdw) terms, and ΔG_GB and ΔG_SA are the polar entropy (−TΔS) were not considered because of the high computational cost and low prediction accuracy. The electrostatic solvation energy (ΔG_GB) was calculated by using the GB model with the parameters developed by Onufriev et al.³¹ (igb = 2). The exterior dielectric constant was set to 80, and the solute dielectric constant value was set to 1. The nonpolar contribution of desolvation (ΔG_SA) was estimated by the solvent-accessible surface area using the LCPO method³². All energy components were calculated using 100 snapshots extracted from 2.0 to 10 ns.

The MM/GBSA free energy decomposition in the mm_pbsa program in AMBER11 was employed to analyze the interactions between each residue in P2Y₆R and compounds A6 and FS-6. The residue−ligand interaction consists of four parts: van der Waals contribution (ΔG_vdw), electrostatic contribution (ΔG_ele), the polar part of desolvation (ΔG_GB), and the nonpolar part of desolvation (ΔG_SA). The exterior dielectric constant was set to 80, and the solute dielectric constant value was set to 1. The nonpolar contribution of desolvation (ΔG_SA) was calculated by SASA using the ICOSA technique³³.

2.21. Statistical analysis

GraphPad Prism V.9 (San Diego, CA, USA) was used for statistical analysis. Data are presented as mean ± SEM. Tests used include One-way ANOVA with Tukey multiple comparison test, two-way ANOVA with Sidak's multiple comparisons test, and Unpaired T test. P < 0.05 was considered statistically significant. All data are representative of at least three independent experiments.

3. Results

3.1. P2Y₆R expression is increased in psoriatic lesional skin and its deficiency alleviates the severity of IMQ-induced psoriasis

To explore the differences between purinergic gene expressions involved in psoriasis progression, we performed a GEO2R analysis. Among all the P2X and P2Y receptor families, P2Y₆R level was dramatically up-regulated both in psoriasis skin and psoriasis mouse model (Fig. 1A). Besides, P2Y₆R expression was positively correlated with PASI scores in patients with psoriasis (GSE117468) (Fig. 1B). Further analysis comparing the two datasets revealed 110 common DEGs (Supporting Information Fig. S2A‒S2D, Table S3). Interestingly, P2Y₆R was in the top 30 DEGs in the PPI network (Fig. S2E). Thus, we speculated that P2Y₆R played an important regulatory role in psoriasis.

Figure 1.

Macrophage P2Y₆R expression is increased in psoriatic lesional skin and involved in Th1 cells mediated psoriasis processes. (A) Expression of purinergic receptors in human and mice psoriasis skin. (B) Linear regression analysis indicated a positive correlation between P2ry6 mRNA expression and PASI score in GSE117468. (C) Flow diagram. (D) Skin thickness was measured from Day 0 to Day 6 (n = 12). (E) Represent images for the psoriatic appearance on IMQ or Vaseline-treated WT mice and P2Y₆R^−/− mice skin. (F) The representative image of spleen from the mice. (G) The spleen index on Day 6 (n = 12). (H) H&E staining of IMQ-treated WT mice and P2Y₆R^−/− mice skin (n = 3). (I) Ki67 staining of skin tissues from WT or P2Y₆R^−/− mice treated with Vaseline or IMQ (n = 3). (J) Immune infiltration correlation coefficient of lymphoid cells and myeloid cells of P2ry6 in GSE181318 and GSE14905. (K) scRNA-seq data of GSE151177 were grouped into 8 clusters with UMAP. (L) Expression of P2Y₆R in different cells form scRNA-seq data. (M) Represent immunofluorescence images of Macrophages CD86 (red) and P2Y₆R (green) co-localization in mice skin treated with Vaseline or IMQ (n = 3). Scale bar = 50 μm. (N) CellChat analysis indicated the inferred interaction among macrophages in psoriasis patient. (O–P) Th1 cells (CD4⁺ IFNγ⁺) proportion was decreased in IMQ-induced P2Y₆R^−/− mice skin and spleen (n = 3). Data are expressed as mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

Accordingly, we generated P2Y₆R knockout mice (P2Y₆R^−/−) and established an IMQ-induced psoriasis mouse model (Fig. 1C)¹⁰. It was shown that the PASI scoring, spleen index, and histological analysis showed that the severity of psoriasis was alleviated in the IMQ-treated P2Y₆R^−/− mice compared to the IMQ-treated WT mice (Fig. 1D‒I).

It has been suggested that purinergic receptor signaling networks interacted with immune cell responses^2,3. To investigate whether P2Y₆R alterations were associated with immune infiltrate in psoriasis, we used CIBERSORTX to excavate its functional role in the immune system. Among the 22 immune cell fractions, psoriasis patients exhibited high infiltration levels of CD4⁺ T cells, Th cells, NK resting cells, M1 macrophages, and activated DCs (Fig. S2F). Specifically, P2ry6 was found to be highly expressed in M1 macrophages (Fig. 1J). Reanalysis of scRNA-seq of skin from healthy controls and patients with psoriasis also showed up-regulation of P2ry6 in macrophages from lesional skin in psoriasis patients (Fig. 1K, L and Supporting Information Fig. S3). This observation was validated by the co-localization of P2Y₆R with M1 macrophages in the dermis of IMQ-treated WT mice (Fig. 1M). However, despite a more comprehensive PASI analysis showing the psoriasis was alleviated in the IMQ-treated P2Y₆R^−/− mice (Supporting Information Fig. S4A), the proportion of M1 macrophages did not change significantly after P2Y₆R depletion (Fig. S4B). Given the importance of macrophages in mediating cell–cell interactions in psoriasis^2,11,12, it was suggested that macrophages had the potential connection between T cells, NK cells, or DCs in the CELLCHAT algorithm communication network (Fig. 1N). Therefore, we detected immune cell responses in the skin and spleen tissue including Th1 cells, Th17 cells, DCs, and NK cells after P2Y₆R depletion. Since the diversity of macrophage populations in psoriasis, we also detected CD163⁺ M2 macrophage cell percentage after P2Y₆R knockout. However, we found there is no significant difference in M2 cell percentage between P2Y₆R-deficient mice and WT mice, suggesting that P2Y₆R depletion only impaired inflammatory macrophage activation and did not favor the M1 or M2 phenotype. Importantly, compared with IMQ-treated WT mice, Th1 cells were the most significantly decreased (Fig. 1O, P and Fig. S4D‒S4F). Besides, to verify whether P2Y₆R on Langerhans cells is activated and affects the course of psoriasis, CD45⁺CD207⁺MHCII⁺ LCs were counted both in the epidermis and the dermis. However, we found no difference in the distribution and density of Langerhans cells and CD11c⁺ DCs between WT and P2Y₆R^−/− mice after IMQ treatment (Fig. S3G). In all, the above results indicated a correlation could exist between activated macrophage P2Y₆R and Th1 cells involved in psoriasis processes.

3.2. Macrophage P2Y₆R mediates Th1 type psoriasis-like inflammation

To further understand the mechanism of macrophage P2Y₆R driven Th1 cell differentiation, we purified Naïve CD4⁺ T-cells from WT mice and cultured these cells with culture supernatants collected from activated WT or P2Y₆R^−/− BMDMs^12,13 (Fig. 2A). UDP, which was the natural ligand enriched in purine metabolism for P2Y₆R, was involved in the inflammation formation as previously reported14, 15, 16, 17. Here, we used IMQ and UDP as the activator to stimulate macrophages. As expected, the supernatants from P2Y₆R^−/− macrophages induced a lower percentage of IFN-γ⁺ CD4⁺ T-cells compared to the supernatants from WT macrophages (Fig. 2B and C). These findings were supported by the increasing mRNA expression levels of Th1 cytokines Ifnγ, Cxcl9, Cxcl10, Cxcl11, and the concentration of IFNγ in the supernatant was attenuated after P2Y₆R^−/− macrophages supernatants intervention in CD4⁺ T cell^18,19 (Fig. 2D and E). In addition, P2ry6 was enriched among genes correlating positively with Stat1 (Fig. 2F). According to GSEA and ChEA3 analysis, Stat1 was identified as a highly ranked transcription factor and up-regulated via JAK/STAT signaling pathway in psoriasis (Fig. 2G and H). In CD4⁺ T cells, stimulation with the supernatants from P2Y₆R^−/− macrophages decreased the levels of p-JAK2, p-STAT1, and T-bet expression (Fig. 2I and J). Further, the reduction percentage of Th1 cells and mRNA level of Ifnγ induced by P2Y₆R^−/− macrophages supernatants was greatly up-regulated after the JAK/STAT pathway agonist RO8191 intervention (Fig. 2K and L). Besides, pre-treatment with RO8191 showed the same effect in T-bet protein expression (Fig. 2M and N). Overall, our findings suggested that P2Y₆R on macrophages regulated Th1 cell differentiation via p-JAK2/p-STAT1/T-bet signaling.

Figure 2.

Macrophage P2Y₆R mediated Th1 type psoriasis-like inflammation via p-JAK2/p-STAT1/T-bet signaling. (A) Flow diagram. (B, C) FACs and the statistical analysis of IFNγ⁺ CD4⁺ T-cells (n = 3). (D) IFNγ concentration after WT or P2Y₆R^−/− mice BMDM supernatant inducing Naïve CD4⁺ T-cells (n = 4). (E) The mRNA expression of the Th1-associated cytokines, Ifnγ, Cxcl9, Cxcl10, and Cxcl11 were detected by RT-qPCR (n = 3). All mRNA levels were normalized to GAPDH. (F) Linear regression analysis indicated a positive correlation between P2ry6 and Stat1 in GSE181318 and GSE14905. (G) GSEA analysis of GSE50400 showed JNK/STAT signaling pathways were up-regulated in IMQ-induced mice. (H) Top 15 putative transcription factors based on ChEA3. (I, J) The protein expression levels of p-JAK2, p-STAT1, and T-bet in T cells. Normal protein levels were normalized to β-actin, the values for the phosphorylated forms of proteins were normalized to phosphorylation-independent levels of the same protein (n = 4). (K) CD4⁺ Naïve T cells supernatant IFNγ concentration after WT or P2Y₆R^−/− mice BMDM supernatant inducing and being treated with or without RO8191 (n = 4). (L) The proportion of IFNγ⁺ CD4⁺ T cells of CD4⁺ Naïve T cells after WT or P2Y₆R^−/− mice BMDM supernatant inducing and being treated with or without RO8191 (n = 3). (M, N) T-bet protein expression of CD4⁺ Naïve T cells after WT or P2Y₆R^−/− mice BMDM supernatant inducing and being treated with or without RO8191 (n = 4). Data are expressed as mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

3.3. Activated macrophages secrete IL-27 via P2Y₆R/PLC_β/p-PKC/MAPK axis

To explore the signaling pathway responsible for psoriasis and assess how macrophage P2Y₆R regulated Th1 cell differentiation, we selected the top 30 proteins in the PPI network based on their topological characteristics for further enrichment analysis (Fig. S2E). The top 15 KEGG pathways and GO biological process terms were shown (Fig. 3A, Supporting Information Fig. S5). The IL-27 signaling pathway was the most significantly enriched, while P2ry6 expression was correlating positively with Il27 (Fig. 3B). Based on these findings, it is necessary to clarify whether activated P2Y₆R in macrophages triggered the release of IL-27. Thus, we measured IL-27 concentration in the supernatant of BMDMs from P2Y₆R^−/− and WT mice, respectively (Fig. 3C). After UDP and IMQ intervention, the IL-27 concentration in P2Y₆R^−/− BMDMs was significantly lower than WT BMDMs (Fig. 3D).

Figure 3.

P2Y₆R enhances IL-27 released from macrophages. (A) Top15 BPs were shown through GO analysis of DEGs. (B) Linear regression analysis indicated a positive correlation between P2ry6 and Il-27 in GSE181318 and GSE14905. (C) Flow diagram. (D) The IL-27 concentration in the supernatants of P2Y₆R^−/− or WT BMDMs induced by IMQ and UDP (n = 4). (E) Flow diagram. (F–G) The proportion of Th1 cells differentiated from IL27ra^−/− Naïve T cells has no significant variability between WT BMDMs and P2Y₆R^−/− BMDMs supernatant stimulation tested by FACs (n = 3). (H) WT or IL27ra^−/− mice CD4⁺ Naïve T cells supernatant IFNγ concentration after WT or P2Y₆R^−/− mice BMDM supernatant inducing (n = 4). (I) Changes in BMDMs intracellular Ca²⁺ were measured using the Ca²⁺ detection Fluo-4AM molecular probes. (J) Quantitative analysis of fluorescence intensity. (K–L) Protein expression in different groups. Normal protein levels were normalized to β-actin, the values for the phosphorylated forms of proteins were normaliszd to phosphorylation-independent levels of the same protein (n = 4). (M) The IL-27 concentration in the supernatant. Data are expressed as mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

The previous study indicated that IL-27ra was predominantly expressed in CD4⁺ T cells²⁰. Next, to determine whether the P2Y₆R stimulated macrophages-derived Th1 cell differentiation was driven by IL-27 dependently, we isolated Naïve CD4⁺ T-cells from WT or IL-27ra^−/− mice and cultured them with supernatants collected from activated WT or P2Y₆R^−/− BMDMs, IL-27 recombinant protein was used as positive control (Fig. 3E). The supernatants of WT BMDMs significantly induced robust Th1 cell differentiation and IFNγ concentration in WT Naïve T cells compared to the supernatants from P2Y₆R^−/− BMDMs. However, the supernatants of WT or P2Y₆R^−/− macrophages showed no difference in enhancing Th1 differentiation and IFNγ secretion in the absence of IL27ra (Fig. 3F‒H).

Next, we further investigated how activated P2Y₆R in macrophages triggered the release of IL-27. It was reported that P2Y₆R stimulation could activate PLC_β/p-PKC/MAPK downstream signaling and the Ca²⁺ influx21, 22, 23. In macrophages, p-JNK mediated the IL-27 release, and increased intracellular Ca²⁺ concentration is widely recognized to contribute to inflammatory macrophage activation24, 25, 26, 27 (Fig. 3I). The intracellular Ca²⁺ fluorescence intensity in P2Y₆R^−/− BMDMs treated with IMQ and UDP was decreased compared to WT BMDMs (Fig. 3I and J). Furthermore, the expression levels of PLC_β, p-PKC, p-p38, and p-JNK were substantially decreased in IMQ and UDP-treated P2Y₆R^−/− BMDMs, while p-ERK showed relatively lower expression, but the difference was not significant (Fig. 3K and L). As we thought, pre-treatment with PMA (PKC agonist), Anisomycin (JNK agonist), and Cigilitazone (p38 agonist), but not Ceramide C6 (ERK agonist), increased IL-27 concentration in P2Y₆R^−/− BMDMs (Fig. 3M). These indicated that the mechanism by which activated P2Y₆R triggered the release of IL-27 may relied on the activation of the PLC_β/p-PKC/MAPK axis in macrophages.

To verify the findings in vitro, an assay was performed in WT and P2Y₆R^−/− mice in vivo (Supporting Information Fig. S6A). After IMQ-treated, the IL-27 and IFNγ concentrations in WT mice skin and serum were significantly increased, while attenuated in P2Y₆R^−/− mice respectively (Fig. S6B). The up-regulation of several Th1 chemokines induced by IMQ was attenuated in P2Y₆R^−/− mice (Fig. S6C). Besides, the protein expression levels of PLC_β, p-PKC, p-p38, p-JNK, p-JAK2, p-STAT1, and T-bet were also decreased in IMQ-treated P2Y₆R^−/− mice (Fig. S6D‒S6G). Taken together, we concluded that the signaling cascade of P2Y₆R activation mediated the interaction between macrophages and T lymphocytes in psoriasis.

3.4. IMQ-induced psoriasis is reduced in Lyz2^creP2Y₆R^fl/fl mice

To further directly assess the cell type-intrinsic function of P2Y₆R in macrophages, we generated a conditional macrophage-specific P2Y₆R knockout mice (Lyz2^creP2Y₆R^fl/fl). These mice with their P2Y₆R^fl/fl littermates were treated with IMQ (Fig. 4A). As for the result, loss of P2Y₆R in macrophages alleviated IMQ-induced psoriatic inflammation and the proliferation of keratinocytes (Fig. 4B‒G). ELISA result revealed that IL-27 and IFNγ level was significantly alleviated in Lyz2^creP2Y₆R^fl/fl mice treated with IMQ, but increased in IMQ-treated P2Y₆R^fl/fl mice (Fig. 4H). The alleviation of disease severity was correlated with decreased number of Th1 cells in the skin and spleen (Fig. 4I and J). The protein expression of PLC_β, p-PKC, p-p38, p-JNK, p-JAK2, p-STAT1, and T-bet were decreased in IMQ-treated Lyz2^creP2Y₆R^fl/fl (Fig. 4K‒M). Accordingly, the transcript levels of Th1-associated chemokines were also strongly down-regulated in Lyz2^creP2Y₆R^fl/fl mice treated with IMQ (Fig. 4N).

Figure 4.

IMQ-induced psoriasis mice is reduced in Lyz2^creP2Y₆R^fl/fl mice. (A) Experimental scheme. (B) Representative skin pictures of Lyz2^creP2Y₆R^fl/fl and P2Y₆R^fl/fl mice. (C) Pictures of spleen and (D) spleen index from all groups (n = 12). (E, F) H&E staining (n = 3) and skin thickness, erythema, and scaling PASI score (n = 12) of IMQ or Vaseline-treated Lyz2^cre P2Y₆R^fl/fl and P2Y₆R^fl/fl mice skin. (G) Ki67 staining of IMQ-treated skin tissues from P2Y₆R^fl/fl or Lyz2^creP2Y₆R^fl/fl mice treated with Saline or IL-27 (n = 3). (H) Serum and skin concentrations of IL-27 and IFNγ detected by ELISA assay (n = 4). (I, J) Flow analysis of Th1 cells from skin and spleens of Lyz2^creP2Y₆R^fl/fl and P2Y₆R^fl/fl mice treated with IMQ or Vaseline (n = 3). (K–M) Quantitative analysis of protein expression in mice skin. Normal protein levels were normalized to β-actin, and the values for the phosphorylated forms of proteins were normalized to phosphorylation-independent levels of the same protein. (n = 4). (N) The mRNA expression levels of Th1-related cytokine and IL-27 p28 and Ebi3 in skin tissues (n = 3). All mRNA levels were normalized to GAPDH mRNA. Data are expressed as mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

Considering the mRNA expression of P2ry6 was relatively high in CD4⁺ T cells (Fig. 1J), we further investigated the role of P2Y₆R in CD4⁺ cells by using CD4^creP2Y₆R^fl/fl mice. After 5 days of IMQ treatment, there was no significant difference in the dermal thickness, PASI score, or spleen index observed between P2Y₆R^fl/fl and CD4^creP2Y₆R^fl/fl mice (Supporting Information Fig. S7). Based on this, we inferred that the macrophage P2Y₆R was mainly responsible for increased severity of psoriasiform dermatitis, rather than P2Y₆R in CD4⁺ T cells.

3.5. IL-27 restores IMQ-induced psoriasis inflammatory phenotype in Lyz2^creP2Y₆R^fl/fl mice

To confirm the mechanism by which P2Y₆R promoted IL-27-dependent production found in mouse macrophages. IL-27 recombinant protein was administered to Lyz2^creP2Y₆R^fl/fl and P2Y₆R^fl/fl mice treated with IMQ, while saline was given as a control²⁸ (Fig. 5A). Consistent with the previous results, the skin inflammation was significantly decreased in IMQ-treated Lyz2^creP2Y₆R^fl/fl mice compared to IMQ-treated P2Y₆R^fl/fl mice. In contrast, IMQ-treated Lyz2^creP2Y₆R^fl/fl mice with IL-27 recombinant protein showed more severe psoriatic inflammation and keratinization compared to IMQ-treated Lyz2^creP2Y₆R^fl/fl mice treated with saline (Fig. 5B‒F). Additionally, in Lyz2^creP2Y₆R^fl/fl mice, there was an increase in the proportion of IFNγ⁺ CD4⁺ Th1-cells in the skin and spleen after IL-27 administration (Fig. 5G‒I). The decreased mRNA expression levels of Ifnγ, Cxcl9, Cxcl10, and Cxcl11 and the protein expression levels of JAK2, STAT1, T-bet were significantly restored in IMQ-induced Lyz2^creP2Y₆R^fl/fl mice by intradermal IL-27 injection (Fig. 5J‒L). In all, these results provided compelling evidence that the elevation of IL-27 restores the reduced psoriatic inflammation caused by macrophage P2Y₆R deficiency.

Figure 5.

Increase of IL-27 restores psoriatic inflammation in Lyz2^cre P2Y₆R^fl/fl mice. (A) Flow diagram. (B) Representative psoriatic plaque pictures of mice. (C) Pictures of the spleen. (D) Spleen index of all groups. (E, F) Skin thickness, erythema and scaling PASI score, H&E staining (n = 3), and Ki67 staining (n = 3) of IMQ-induced Lyz2^creP2Y₆R^fl/fl and P2Y₆R^fl/fl mice skin treated with Saline or IL-27. (G) The mRNA expression levels of Th1-related cytokines in skin tissues from IMQ-induced mice treated with Saline or IL-27 (n = 3). All mRNA levels were normalized to GAPDH mRNA. (H, I) Flow analysis on Th1 cells from skin and spleens of all groups. IL-27 reversed the decrease of Th1 cells (CD4⁺ IFNγ⁺) proportion in IMQ-treated Lyz2^creP2Y₆R^fl/fl mice. (J, K) Quantitative analysis of protein expression in mice skin. Normal protein levels were normalized to β-actin, and the values for the phosphorylated forms of proteins were normalized to phosphorylation-independent levels of the same protein (n = 4). (L) Serum and skin concentrations of IL-27 and IFNγ detected by ELISA assay (n = 4). Data are expressed as mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

3.6. The discovery of a novel P2Y₆R inhibitor FS-6

We next tried to find a P2Y₆R inhibitor with excellent inhibitory activity and binding affinity to further verify the role of P2Y₆R in psoriasis. In our previous study, a potent P2Y₆R inhibitor A6 with a novel scaffold was discovered^29,34. However, A6 had limitations in terms of poor aqueous solubility and bio-availability, hindering its further development. To address this, we performed docking of A6 with the homology model of P2Y₆R. We discovered that the furan ring of A6 is deeply buried in the hydrophobic pocket, while the substituent on the benzimidazole end is exposed in the solvent (Fig. 6A). Therefore, to improve the aqueous solubility of the compound while retaining its antagonistic activity, we rationally designed and synthesized a series of polar groups substituted on the benzimidazole substituent with different linkers (Fig. 6B). The optimized compound FS-6 (P2Y₆R IC₅₀ = 0.0834 nmol/L)³⁴ showed strong binding ability to P2Y₆R, and notably improved solubility (5.029 mg/mL in water) (Fig. 6C and D, Supporting Information Table S4, Scheme S1). The GCI results revealed a dose-dependent interaction between FS-6 and P2Y₆R (Fig. 6E). In the CETSA assay, the maximum difference in the stability of FS-6 and P2Y₆R is observed at 54 °C, and the stabilization effect of FS-6 on P2Y₆R showed dose-dependency (Fig. 6F‒I). The DARTs assay showed the increased stability of P2Y₆R against pronase after treating FS-6 (Fig. 6J and K). Notably, the stability of P2Y₆R was increased in a dose-dependent manner by FS-6 (Fig. 6L and M). To further investigate the key residues involved in the binding of FS-6 to P2Y₆R, we compared the docking modes of A6, FS-6, and MRS2578 with P2Y₆R. We discovered that both A6 and FS-6 bind to the same pocket, whereas MRS2578 binds at a different site due to its role as a covalent inhibitor of P2Y₆R (Supporting Information Fig. S8A and S8B). Besides the nitrogen atoms on the benzimidazole rings of A6 and FS-6 form stable hydrogen bond interactions with the ARG266 residue (Fig. S8C). Based on the result of the docking simulation, we generated point mutation at each of these residues (Y283G, R266G, Y178G, K25G) in P2Y₆R-HEK293 cells (Supporting Information Fig. S9). The mutation of ARG266 on P2Y₆R resulted in a complete loss of inhibitory activity of FS-6 (Fig. 6N). In conclusion, we discovered a novel P2Y₆R inhibitor, FS-6, which exhibits potent inhibitory activity and excellent water solubility, making it a promising candidate for further research and development.

Figure 6.

A novel P2Y₆R inhibitor FS-6 with excellent activity and binding affinity. (A) The overall structure of P2Y₆R with FS-6 bound. (B) Chemical modification from A6 to FS-6 and the possible binding sites between FS-6 and P2Y₆R. (C) Flow diagram of IC₅₀ value and binding activity of FS-6. (D) IC₅₀ curve of FS-6 was determined by ELISA (n = 6). (E) Direct binding between FS-6 and P2Y₆R. (F–I) The binding between FS-6 and P2Y₆R was evaluated by CETSA in P2Y₆R-HEK293 cells (n = 4). (J‒M) The dose-dependent binding of FS-6 to P2Y₆R was evaluated by DARTs (n = 4). (N) Arg266Gly was verified as the binding site of FS-6 and P2Y₆R to detect IP3 concentration (n = 4). Data are expressed as mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001. ns, not significant.

3.7. FS-6 attenuates Th1 cell differentiation which ameliorates IMQ-induced psoriasis

Given FS-6 has such excellent activity and binding affinity of P2Y₆R, we assessed its ability to suppress macrophage IL-27 production. The well-recognized P2Y₆R inhibitor MRS2578 was used as a positive control35, 36, 37. CCK-8 assay was used to determine the optimal drug concentration of FS-6 (Fig. 7A). After co-stimulation with IMQ and UDP, treatment with FS-6 resulted in a dose-dependent decrease in IL-27 and intracellular Ca²⁺ concentration significantly (Fig. 7B‒D). The administration of FS-6 led to a decreased PLC_β, p-PKC, p-p38, and p-JNK protein expression (Fig. 7E and F). While related agonists could reverse the IL-27 response in the FS-6 treatment group (Fig. 7G). The macrophage supernatant containing FS-6 effectively reduced Th1 cell differentiation and IFNγ concentration (Fig. 7H‒J), accompanied by the down-regulation of T cell chemokines mRNA expression (Fig. 7K). Furthermore, the expression protein levels in Naïve T cells were also significantly decreased after FS-6 intervention (Fig. 7L and M). Additionally, RO8191 markedly reversed the decreased proportion of Th1 cells, the secretion of IFNγ, and the expression of T-bet protein was induced by FS-6 treatment (Fig. 7N‒P).

Figure 7.

P2Y₆R inhibitor FS-6 regulates the secretion of IL-27 from macrophages thus inhibiting Th1 cell differentiation. (A) The effects of FS-6 on BMDMs viability (n = 6). (B) FS-6 decreased IL-27 concentration in BMDMs supernatant under the stimulation of UDP and IMQ (n = 4). (C, D) Changes in FS-6 treated BMDMs intracellular Ca²⁺ were measured and quantitative analysis of fluorescence intensity (n = 3). (E, F) FS-6 decreased protein expression in BMDMs stimulated by UDP and IMQ. Normal protein levels were normalized to β-actin, and the values for the phosphorylated forms of proteins were normalized to phosphorylation-independent levels of the same protein (n = 4). (G) IL-27 supernatant concentration of BMDMs stimulated by UDP and IMQ after PKC, JNK, p38, and ERK agonist treatment. (H, I) The proportion of Th1 cells induced by WT BMDMs supernatant was decreased under the treatment of FS-6 (n = 3). (J) FS-6 decreased CD4⁺ Naïve T-cells supernatant IFNγ concentration after BMDMs supernatant inducing (n = 4). (K) FS-6 decreased the mRNA expression levels of the Th1-associated cytokines, Ifnγ, Cxcl9, Cxcl10 and Cxcl11 (n = 3). (L, M) FS-6 decreased protein expression in T cells (n = 4). (N) The proportion of Th1 cells induced by WT BMDMs supernatant and JAK/STAT agonist was decreased under the treatment of FS-6 (n = 3). (O) FS-6 decreased supernatant IFNγ concentration of Th1 cells induced by WT BMDMs supernatant and JAK/STAT agonist (n = 4). (P, Q) FS-6 decreased the T-bet expression level of Th1 cells induced by WT BMDMs supernatant and JAK/STAT agonist (n = 4). Data are expressed as mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

To evaluate the effect of FS-6 in vivo, WT mice applied with IMQ and received daily doses of FS-6 (10 and 20 mg/kg, i.p). Methotrexate (MTX) (1 mg/kg, i.p) served as the positive control³⁸ (Fig. 8A). After 5 days of treatment, mice treated with FS-6 showed a significant reduction in epidermal thickness and dermal inflammation compared to the model group (Fig. 8B‒E). Moreover, the IL-27 and IFNγ content was decreased after administration of FS-6 in the mice serum and skin (Fig. 8F). Additionally, H&E and Ki67 staining of the skin confirmed the therapeutic effect of FS-6 in psoriasis mice, and FS-6 administration could reduce P2Y₆R⁺ CD86⁺ double-positive macrophages in mice dermis (Fig. 8G).

Figure 8.

FS-6 ameliorates IMQ-induced psoriasis in mice. (A) Experimental scheme for assessment of psoriasiform dermatitis (n = 12). (B) Representative skin pictures. (C) Picture of spleen and (D) spleen index from all groups (n = 12). (E) PASI score (n = 12). (F) Serum and skin concentrations of IL-27 and IFNγ (n = 4). (G) FS-6 reduced epidermal thickening, keratinocyte proliferation, and co-localization of macrophages and P2Y₆R (n = 3). (H, I) Flow analysis on Th1 cells from skin and spleens of mice (n = 3). Data are expressed as mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

We then explored whether administration of FS-6 could suppress the differentiation of Th1 cells. Compared to the model group, the Th1 cell differentiation rate in mice spleen and skin lesions was significantly decreased after FS-6 was administered (Fig. 8H and I). Given the role of P2Y₆R in vitro, we next confirmed whether FS-6 inhibits P2Y₆R-meditated downstream signaling pathway in vivo (Supporting Information Fig. S10A and S10B). Notably, FS-6 significantly inhibited the mRNA expression levels of Ifnγ, Cxcl9, Cxcl10, Cxcl11 and two submits of IL-27, p28 and Ebi3 (Fig. S10C). Collectively, these data confirmed that targeted P2Y₆R was a potential therapeutic strategy in psoriasis.

4. Discussion

P2Y₆R has been implicated in the pathogenesis of inflammatory bowel disease, atherosclerosis, Parkinson's disease, and remodeling in experimental allergic airway inflammation^14,16,17,34, 35, 36^,39, 40, 41. Due to its pro-inflammatory activity, P2Y₆R is highly abundant in a wide range of immune cells, including macrophages and T cells. M1 macrophages significantly promote dermatitis pathogenesis in a psoriasis-like mouse model⁴². Besides, treatments that target M1 macrophage activation may serve as a general therapeutic candidate for psoriasis^43,44. Here, we demonstrated that P2Y₆R was up-regulated in psoriasis and was co-localized with CD86⁺ M1 macrophages. The deficiency of P2Y₆R hindered the progression of psoriasis.

Extracellular UDP, released as a danger signal by stressed or apoptotic cells, played an important role in a series of physiological processes⁴⁵. In innate immune responses, macrophages release UDP as damage-associated molecular pattern molecules cause inflammation^46,47. Besides, our previous studies have shown that macrophage P2Y₆R was activated by UDP and played a role in immune response and inflammation modulation²². In the present study, we found that UDP, along with IMQ, mediated the activation of macrophagic P2Y₆R, leading to the activation of PLC_β/p-PKC/MAPK signaling pathways, which ultimately triggered the release of IL-27. In the MAPK pathway, JNK signaling is a commonly dysregulated signaling pathway in various human diseases, including psoriasis⁴⁸. JNK activation is known to be mediated by IP3 Ca²⁺ stores, while P2Y₆R couples to Gq protein, activating PLC_β, resulting in increased intracellular Ca²⁺ level^22,49,50. Additionally, P2Y₆R-mediated JNK activation in metabolic dysfunction has been reported previously⁴¹. Therefore, we hypothesized that the activation of the UDP positive feedback loop may result in the persistent activation of macrophagic P2Y₆R in psoriasis, contributing to its pathogenesis.

IL-27 level was observed to be higher in both the serum and the diseased tissue in psoriatic patients, and this increase correlated with the severity of the disease⁵¹. Moreover, IL-27 injected locally aggravated the progression in the IMQ-induced mouse psoriasis model²⁸. Therefore, inhibition of IL-27 was considered a potential strategy for preventing of psoriasis⁵². Mechanically, IL-27 is primarily produced by activated antigen-presenting cells and plays an important role in the regulation of CD4⁺ Th1 cell differentiation and immune response53, 54, 55, 56. We observed the same phenomenon in macrophages stimulated with IMQ and UDP, while IL-27 production was significantly decreased following P2Y₆R depletion. In accordance with the GO enrichment and GSEA analysis results, P2Y₆R induced the relevant interaction between macrophages and Th1 cells, and this interaction was mediated by IL-27.

Previous literature reported that P2Y₆R can promote the differentiation of Th1 cells in intestinal inflammation³⁹. Th1 cells produce Ifnγ and T-bet and play a pathogenic role in psoriasis⁵⁷. It has been reported that IL-27 induces STAT-1 phosphorylation and increases T-bet expression⁵⁸. The dysregulated p-JAK2-p-STAT1 signaling axis in psoriasis has been implicated in the increased expression of cytokines, such as IFN-γ, which contributed to the pro-inflammatory environment in psoriatic lesions59, 60, 61. Collectively, our data showed that P2Y₆R had critical functions in macrophages, including the promotion of pro-inflammatory cytokine IL-27, while IL-27 induced Th1 cell differentiation via p-JAK2/p-STAT1/T-bet axis, ultimately exacerbating the psoriasis progression.

Developing small-molecule inhibitors targeting P2Y₆R has been challenging due to the lack of a resolved crystal structure and the absence of key residues on the binding sites. In our previous study, a potent P2Y₆R inhibitor A6 with a novel scaffold was discovered^29,34. However, the poor aqueous solubility and bioavailability of A6 have hindered further pharmacological research and development. The chemical optimization on the benzimidazole ring of A6 may be a promising strategy for P2Y₆R inhibitors with improved aqueous solubility and bioavailability. In this study, we utilized docking modes and binding free energy calculations from prior research to design mutagenesis experiments on the P2Y₆R, ARG266, which demonstrated a complete loss of inhibitory activity by FS-6, indicating a promising small-molecule binding site. Besides, the FS-6 therapeutic mechanism may avoid the currently observed lack of selectivity in psoriasis small molecule therapeutics such as JAK inhibitors and the prevalent gastrointestinal adverse effects associated with PDE4 inhibitors³. Additionally, monoclonal antibodies, which are commonly employed in the treatment of psoriasis, are expensive and require administration via injection¹. FS-6 offers a more accessible oral dosing option. This potential for oral delivery not only enhances patient convenience but also promises a more cost-effective solution. In conclusion, FS-6 may serve as a framework to guide the discovery of inhibitors for P2Y₆R.

However, our study contains several limitations. Although the mechanism underlying cross-talk between IL-27RA in Th1 cells and P2Y₆R in macrophages has been uncovered. However, a more complex mechanism, involving the resolution of P2Y₆R crystal structure would be required. Targeted drugs for macrophage P2Y₆R in treating psoriasis remain to be developed and explored. In summary, our present study provided evidence of mechanism underlying P2Y₆R in the pathogenesis of psoriasis and highlighted the potential of P2Y₆R inhibitors as drug candidates for psoriasis therapy. On the other hand, FS-6 displayed inhibitory effects on psoriasis progression as a novel P2Y₆R inhibitor, which confirmed the feasibility of P2Y₆R targeted therapy on psoriasis.

Author contributions

Li Yin: Data curation, Formal analysis, Investigation, Methodology, Validation, Writing – original draft, Writing – review & editing. Enming Zhang: Data curation, Formal analysis, Software, Supervision, Validation, Visualization. Tianqi Mao: Data curation, Methodology, Resources, Validation, Visualization. Yifan Zhu: Data curation, Formal analysis, Methodology. Shurui Ni: Data curation, Methodology, Resources. Yehong Li: Writing – review & editing. Chunxiao Liu: Writing – review & editing. Yafei Fang: Writing – review & editing. Kexin Ni: Data curation, Formal analysis. Yuhe Lu: Data curation, Methodology. Huanqiu Li: Conceptualization, Data curation, Funding acquisition, Project administration, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing. Mengze Zhou: Conceptualization, Data curation, Funding acquisition, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing. Qinghua Hu: Conceptualization, Data curation, Funding acquisition, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing.

Conflicts of interest

The authors declare that they have no competing interests.

Appendix A. Supporting information

The following is the Supporting Information to this article: