Male C57BL/6 mice, approximately 6–8 weeks old, were purchased from Chongqing Tengxin Experimental Animals Co., Ltd. (Chongqing, China). All animal procedures conform to accordance with the guidelines of the Committee on the Protection and Use of Animals in the Southwest Jiaotong University (SWJTU-2107–004(SWJTU)). After adaptive feeding for 1 week, mice were randomly assigned to four groups, and a psoriasis-like model was induced using imiquimod (IMQ), a Toll-like receptor 7 (TLR7) agonist, as previously described (Qian et al., 2024). In the IMQ group, mice were topically treated with 62.5 mg of commercial 5% IMQ cream (SICHUAN MED-SHINE PHARMACEUTICAL CO., LTD.) daily for six consecutive days (Qian et al., 2024). In the IMQ + Lico B group, mice were orally administered Lico B dissolved in saline (200 μL) 2 h prior to each IMQ application. Animals were further divided into low-dose (40 mg/kg), medium-dose (80 mg/kg), and high-dose (160 mg/kg) treatment groups. The Lico B group received the same dose of Lico B along with topical vaseline. The Control (CTRL) group administered an equivalent volume of saline and topical vaseline.

To evaluate the potential hepatic and renal toxicity of Lico B, C57BL/6 mice were orally administered Lico B dissolved in saline at different doses (200 μL per mouse; low, 40 mg/kg; medium, 80 mg/kg; high, 160 mg/kg) daily. CTRL mice received an equal volume of saline. After six consecutive days of treatment, tissues were collected for further analysis.

The keratinocyte cell line HaCaT was obtained from IMMOCELL and cultured in DMEM (Gibco) supplemented with 10% fetal bovine serum (FBS; F101-01, Vazyme) under standard conditions (37 °C, 5% CO₂). Cells were divided into four groups during the logarithmic growth phase. The IL-17 group was treated with 100 ng/mL IL-17 (Proteintech; HZ-1113); the Lico B group received 9 µM Lico B (MCE; HY-N3388); the IL-17 + Lico B group was co-treated with IL-17 and Lico B; and the control group was treated with PBS. After 24 h of treatment, cells were processed for subsequent experiments.

HaCaT cells were seeded in 96-well plates and allowed to attach for 24 h. When cell confluence reached 60%–70%, they were treated with different concentrations of Lico B for 24 h. At the end of treatment, 10 μL of MTT solution was added to each well, and cells were incubated for an additional 4 h at 37 °C in 5% CO₂. The supernatant was then carefully removed, and 100 μL of DMSO was added to dissolve the formazan crystals. Absorbance was measured at 560 nm using a microplate reader.

After fixation, the skin tissue samples from IMQ models were dehydrated through a graded alcohol series, embedded in paraffin, and sectioned into 5 µm slices. The sections were then stained with Hematoxylin-Eosin (Solarbio, G1120) reagents following the manufacturer’s protocol. Pathological alterations in the skin tissues were examined under an optical microscope.

Paraffin-embedded skin tissue sections were deparaffinized in xylene, rehydrated, and subjected to heat-induced antigen retrieval in citrate buffer at 100 °C for 15 min. To prevent nonspecific binding, sections were blocked with 10% goat serum in PBS for 1 h at room temperature. The slides were then incubated overnight at 4 °C with primary antibodies, including rabbit anti-SCD1 (Proteintech, 28678-1-AP, 1:200), rabbit anti-Ki67 (Proteintech, 12348-1-AP, 1:100), and rabbit anti-KRT17 (Proteintech, 17516-1-AP, 1:100). Following primary antibody incubation, sections were washed and incubated with Alexa Fluor 555 goat anti-rabbit IgG (Abcam, ab150078, 1:5,000), Alexa Fluor 488 goat anti-rabbit IgG (Abcam, ab150077, 1:5,000) and/or Alexa Fluor 647 goat anti-rabbit IgG (Abcam, ab150079, 1:5,000) for 2 h at room temperature. Nuclei were counterstained with DAPI staining solution. After washing three times in PBS, coverslips were mounted using 50% glycerol in PBS to prevent photobleaching. Fluorescence images were captured using a laser scanning confocal microscopy (Olympus, FV12-IXCOV).

Single-cell suspensions were prepared according to published methods for flow cytometry [PMID: 32019420]. Spleens were harvested from IMQ models and mechanically dissociated through a 70 μm cell strainer to obtain single-cell suspensions. Red blood cells were lysed using RBC lysis buffer, and the remaining cells were washed and resuspended in PBS for downstream analyses. To evaluate IL-17A expression, single-cell suspensions were stimulated for 4 h at 37 °C using a Cell Activation Cocktail (423,303, Biolegend). After stimulation, surface markers were stained with anti-mouse CD8 antibodies (eBioscience) on ice for 30 min. IL-17 was detected using an anti-mouse IL-17 antibody (eBioscience) following overnight incubation at 4 °C. Samples were analyzed using an LSR Fortessa flow cytometer (SONY, MA900).

Total proteins were isolated from the cells with RIPA buffer (Beyotime, P0013B), followed by centrifugation of the lysates at 12,000×g for 30 min at 4 °C. Protein concentration was measured using a BCA assay kit (Solarbio, PC0020). Equal quantities of protein were loaded per lane and resolved on 10% SDS–PAGE gels, separated electrophoretically, and transferred onto PVDF membranes (Millipore, IPVH00010). After blocking, the membranes were probed with primary antibodies against SCD1 (Proteintech, 28678-1-AP, 1:2000), KRT17 (Proteintech, 17516-1-AP, 1:2000) and Beta Actin (Proteintech, 20536-1-AP, 1:10,000) overnight at 4 °C. Following three washes with TBST, the membranes were treated with secondary antibody (Proteintech, SA00001-2, 1:5,000) at room temperature for 2 h and imaged using an eBlot Touch (TOUCH IMAGER PRO). Band intensities were quantified with ImageJ software.

Total RNA was isolated using TRIzol reagent (Vazyme, R411-01) and converted into cDNA using HiScript II Reverse Transcriptase (Vazyme, R223-01) according to the manufacturer’s instructions. Quantitative real-time PCR (qRT-PCR) was conducted on a Real-Time PCR Instrument (Bio-Rad, CFX96 Touch PCR) with 2× TaqSYBR Green qPCR Mix (Vazyme, SQ101). Gene expression levels were quantified using the comparative 2^-ΔΔCT method, with GAPDH or β-actin serving as internal reference genes for normalization. The primer sequences of the target and control genes were as follows: il17 forward:5′CTCAGACTACCTCAACCGTTCC3′ and reverse, 5′ATG​TGG​TGG​TCC​AGC​TTT​CC3′; rorγt forward:5′GACCCACACCTCACAAATTGA3′andreverse,5′AGTAGGCCACATTACACTGCT3′; GAPDH forward:5′AGGTCGGTGTGAACGGATTTG3′ and reverse, 5′TGT​AGA​CCA​TGT​AGT​TGA​GGT​CA 3′.

Targets of Lico B were retrieved from the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP). Psoriasis-related gene expression data (GSE54456) were obtained from the NCBI Gene Expression Omnibus (GEO) database. Psoriasis-associated target genes were collected from GeneCards and the Therapeutic Target Database (TTD). The intersection of Lico B targets, psoriasis-related genes from the GSE30528 dataset, and known psoriasis-associated genes was then analyzed to identify potential key targets for subsequent investigation.

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed using the DAVID database for Lico B targets and the targets within the floralwhite module. Enrichment was assessed across biological process, cellular component, and molecular function categories. KEGG pathway enrichment analysis was performed using a significance threshold of P < 0.05, and the top 20 pathways were ranked in descending order based on the calculated p-values.

Weighted gene co-expression network analysis was performed using the ‘WGCNA’ R package on differentially expressed genes from the GSE54456 dataset, which included 174 samples. A soft-thresholding power of 19 was selected to construct a scale-free network. The adjacency matrix was then transformed into a topological overlap matrix (TOM), and gene connectivity and phase differences were assessed. Modules were identified using hierarchical clustering combined with the dynamic tree cut method. Gene significance and module membership correlations were calculated, and genes from the relevant modules were extracted for further analysis.

The crystal structure of SCD1 was obtained from the Protein Data Bank (PDB). Molecular docking of SCD1 with Lico B was performed using AutoDock Tools. The docking results were visualized in three dimensions using PyMOL, and protein–ligand interactions were further illustrated in 2D using Discovery Studio Visualizer.

All data are expressed as mean ± standard deviation (S.D.) from at least three independent experiments to ensure reliability. Statistical analysis was performed using R packages or GraphPad Prism. One-way ANOVA was applied for comparisons across multiple groups, and unpaired t-tests were used for two-group comparisons. A P value <0.05 was considered statistically significant.

The workflow of this study was shown in Figure 1. To explore the potential molecular mechanisms of Lico B, we first predicted its targets using TCMSP and subsequently performed Gene Ontology (GO) functional annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses. GO analysis revealed that these candidate targets were primarily enriched in biological processes related to estrogen response, cell cycle regulation, and nuclear receptor activity (Figure 2A). KEGG enrichment further indicated significant involvement of these targets in the cell cycle pathway and cancer-related pathways (Figure 2B). Notably, lipid and atherosclerosis pathway was enriched in Lico B′ targets, suggests the potential correlation between Lico B and lipid metabolism. Considering dysregulated lipid metabolism has been increasingly recognized as a contributing factor to chronic inflammatory skin disorders such as psoriasis, we therefore speculate that Lico B may play anti-psoriasis roles via modulation of lipid metabolic processes (Zhang X. et al., 2022; Kanno et al., 2023).

To further identify potential molecular targets associated with psoriasis, we analyzed the GSE54456 dataset from the GEO database, which includes 92 psoriatic lesions and 82 healthy skin biopsy samples. Differential expression genes (DEGs) analysis was performed, and the volcano plot and heatmap demonstrated distinct sets of upregulated and downregulated genes between psoriatic and healthy tissues (Figures 3A,B). To further characterize disease-related transcriptional patterns, WGCNA was performed. According to the WGCNA principle, an appropriate soft-thresholding power (β) is required to construct an adjacency matrix that satisfies the scale-free topology criterion. Candidate power values ranging from 1 to 30 were evaluated, and for each value, the scale-free topology fit index and mean connectivity were calculated (Figure 4A). The left panel illustrates the relationship between different power values and the scale-free topology fit, whereas the right panel depicts the corresponding mean connectivity of the constructed networks. These analyses indicated that a soft-thresholding power of β = 19 achieved an optimal balance between high scale-free topology fit and adequate mean connectivity. Therefore, β = 19 was selected for subsequent network construction. Using this selected power, a WGCNA was established, resulting in the identification of 9 distinct gene modules. These modules were visualized using a hierarchical clustering dendrogram (Figure 4B), in which the upper panel represents gene clustering based on topological overlap, and the lower panel indicates module assignment. Genes sharing the same color were classified into the same module, while the colors labeled as “Dynamic Tree Cut” indicate modules initially identified using the dynamicTreeCut algorithm. Given the high similarity among certain modules, closely related modules were further merged, as shown in the “Merged dynamic” annotation. A heatmap depicting the clustering of all genes was also generated (Figure 4C). Module–trait associations were subsequently assessed by calculating Pearson correlation coefficients and corresponding p values between module eigengenes and clinical traits. As shown in Figure 4D, the floralwhite module, comprising 2,617 genes, exhibited the strongest positive correlation with psoriasis (cor = 0.98, p = 4 × 10⁻¹¹⁶). Consequently, the floralwhite module was selected for further analysis. A scatter plot revealed a strong positive correlation between gene significance (GS) and module membership (MM) within the floralwhite module (cor = 0.99, p = 1 × 10⁻²⁰⁰), indicating that genes highly associated with psoriasis were also central components of this module. Next, we conducted an in-depth study of gene functions in the floralwhite module. GO functional enrichment analyses indicated that the enriched biological processes were mainly associated with positive regulation of cytokine production, regulation of innate immune response and regulation of immune effector process. In terms of cellular components, the targets were related to predominantly enriched external side of plasma membrane, cornified envelope and endocytic vesicle. Regarding molecular function, the targets were predominantly enriched in cytokine and chemokine-related activities (Figure 4E). Subsequent KEGG pathway analysis of the floralwhite module indicated that the potential pathways including cytokine–cytokine receptor interaction, cornified envelope formation, chemokine signaling pathway, lipid and atherosclerosis and Th17 cell differentiation (Figure 4F). These findings suggested that psoriasis involves immune dysregulation, imbalanced proliferation and differentiation of keratinocytes, and accompanying lipid metabolism disorders.

To identify key targets of Lico B in psoriasis, we integrated multiple datasets for intersection analysis. Specifically, we compared the predicted targets of Lico B, the psoriasis-related gene set from the floralwhite module, and psoriasis-associated genes from the Therapeutic Target Database and GeneCards Databases were analyzed. The integrated analysis yielded a single overlapping gene SCD1. This indicates that SCD1 is both a core regulatory gene in psoriatic pathology and a potential direct target of Lico B intervention (Figure 5A). Molecular docking revealed favorable binding of Lico B within the SCD1, involving tryptophan, asparagine, leucine and phenylalaninem, with a calculated binding energy of −11.7 kcal/mol (Figure 5B). Surface plasmon resonance assays (SPR) further confirmed that Lico B directly interacted with SCD1 protein in a positive dose-dependent manner. The determined equilibrium dissociation constant (Kd) for Lico B binding to the SCD1 protein was 7.1 μmol/L (Figure 5C). Given that keratinocytes are key effector cells in psoriasis, we next examined the effect of Lico B in IL-17A treated HaCaT cells. Cell viability assays confirmed that Lico B exhibited no detectable cytotoxicity at concentrations ≤12 μM after 24 h (Figure 5D). 9 μM was therefore selected for subsequent experiments. Western blot analysis showed that IL-17A stimulation markedly increased SCD1 expression in keratinocytes, whereas Lico B treatment effectively suppressed this upregulation (Figures 5E,F). Consistently, immunofluorescence staining revealed significantly reduced SCD1 expression in the Lico B+ IL17 group compared to the IL17 group in vivo (Figures 5G,H). Because SCD1 regulates the synthesis of monounsaturated fatty acids and thereby influences neutral lipid storage (Su et al., 2023; Sarandi et al., 2023; Kanno et al., 2023). We further examined intracellular lipid droplets as a downstream readout of SCD1 linked lipid metabolism in HaCaT cells. A downstream production of SCD1, intracellular lipid droplet was assessed in HaCaT cells. Immunofluorescence staining of lipid droplet showed that IL-17A stimulation increased lipid droplet accumulation, whereas Lico B treatment markedly reduced this accumulation (Figure 5I), indicating that Lico B may modulate lipid metabolism by targeting SCD1.

Dysregulated keratinocyte proliferation is a fundamental driver of epidermal hyperplasia in psoriasis. Pathway enrichment analysis indicated that Lico B associated targets are enriched in cornified envelope formation and cell cycle regulation in Figure 2, suggesting Lico B in the control of keratinocyte differentiation and proliferation. We therefore examined whether Lico B could suppress IL-17–induced proliferative responses in HaCaT cells. Western blot analysis showed that co-treatment with IL-17 and Lico B reduced the expression of the hyperproliferation marker KRT17 compared with IL-17 alone (Figures 6A,B). Consistently, immunofluorescence staining demonstrat significantly decreased expression of both KRT17 and the proliferation marker Ki67 in the Lico B + IL17 group compared with the IL-17–stimulated group (Figures 6C–F). These findings indicate that Lico B effectively attenuates IL-17-driven keratinocyte hyperproliferation, suggesting its potential to ameliorate psoriatic epidermal pathology.

To assess the therapeutic effects of Lico B, we established a psoriasis-like mouse model induced by the TLR7 agonist IMQ (Figure 7A). On the day of sacrifice, we observed that mice in the IMQ group exhibited pronounced erythema, increased scaling, and marked epidermal thickening. Treatment with Lico B alleviated these symptoms, leading to reduced erythema, scaling, and skin thickness. The PASI scores were significantly reduced in the Lico B-treated IMQ model group compared with the IMQ group (Figures 7C–F). To further assess the histopathological effects of Lico B, hematoxylin and eosin (H&E) staining of back skin sections was performed. Results showed that the IMQ group displayed pronounced epidermal hyperkeratosis, acanthosis, elongation of rete ridges, and increased infiltration of immune cells in the dermis. In contrast, these psoriatic pathological features were markedly alleviated in the IMQ + Lico B group (Figure 7B). Moreover, we found that Lico B treatment significantly reduced epidermal thickness and decreased spleen size compared with the IMQ group (Figures 6G,H). Collectively, these findings indicate that Lico B effectively mitigates IMQ-induced psoriasis-like skin inflammation and epidermal hyperplasia.

Th17 cell differentiation and its effector cytokine IL-17 are central drivers of psoriatic phenotypes (Blauvelt and Chiricozzi, 2018; Li et al., 2020; Shen et al., 2024). Therefore, we next examined whether Lico B could modulate Th17 responses in IMQ-induced psoriasis-like mice. Flow cytometry analysis revealed that the proportion of splenic Th17 cells in the spleen was significantly reduced in the IMQ + Lico B group compared with the IMQ model group (Figures 8A,B). Consistently, qRT-PCR analysis of dorsal skin demonstrated that Lico B treatment markedly decreased the expression of IL-17 and its key transcriptional factor RORγt mRNA expression (Figures 8C,D), suggesting that Lico B suppresses Th17 cell differentiation and IL-17 production. To evaluate the impact of Lico B on keratinocyte proliferation in vivo, we next assessed the expression of hyperproliferation markers KRT17 and Ki67 in skin tissues. Western blot results showed that KRT17 protein levels were significantly reduced in the Lico B + IMQ group compared with the IMQ group (Figures 8E,F). Immunofluorescence staining further confirmed that Lico B treatment significantly decreased KRT17 and Ki67 expression in psoriatic mouse skin tissue (Figures 8G,H,J,K). Because SCD1 is a key lipid metabolic enzyme upregulated in our psoriatic model, we also examined its expressions in skin lesions. Immunofluorescence staining revealed that SCD1 level was also reduced in the epidermis of IMO-induced mice following Lico B administration (Figures 8I,L). Given that IL-17A promotes keratinocyte hyperproliferation through induction of IL-19 and IL-36γ (Nograles et al., 2008; Sa et al., 2007), the coordinated reduction in Th17 differentiation, IL-17 expression and epidermal hyperproliferation suggests that Lico B alleviates epidermal hyperproliferation in psoriasis-like mice by inhibiting Th17 cell differentiation and attenuating IL-17 signaling.