Objective To investigate the potential therapeutic effects of Liangxue Xiaofeng Powder (凉血消风散, LXXFP) on atopic dermatitis (AD) rats with wind-dampness-heat syndrome, and to explore the potential mechanisms associated with inflammatory responses and gut microbiota alterations.

Methods Active components of LXXFP and their corresponding targets were identified using the Bioinformatics Analysis Tool for Molecular mechANism of Traditional Chinese Medicine (BATMAN-TCM) database and HTDocking platform. AD-associated targets were retrieved from the DisGeNET and GeneCards databases. The overlapping targets between LXXFP and AD were identified using Perl software and visualized via a Venn diagram. Cytoscape 3.7.2 was employed to construct an LXXFP active component-AD overlapping target network. The overlapping targets were subsequently imported into the Search Tool for the Retrieval of Interaction Gene/Proteins (STRING) platform to build a protein-protein interaction (PPI) network, which was further visualized and analyzed using Cytoscape 3.7.2 to identify the core targets based on degree values. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed to preliminarily identify biological processes and signaling pathways potentially involved in the effects of LXXFP on AD. Molecular docking between the top five active components of LXXFP and the five core targets was conducted using AutoDock Vina 1.5.6. In the animal experiment, 36 female Sprague-Dawley (SD) rats were randomly divided into six groups (n = 6 per group): blank, disease, syndrome, and low-, medium-, and high-dose LXXFP (LXXFP-L, LXXFP-M, and LXXFP-H, respectively) groups. The disease group was induced by 1-chloro-2,4-dinitrobenzene (DNCB) alone; the syndrome group and the three LXXFP groups were established as AD rats with wind-dampness-heat syndrome through combined climate simulation and DNCB induction. The blank, disease, and syndrome groups received an equal volume of double-distilled water by gavage daily. The LXXFP-L, LXXFP-M, and LXXFP-H groups were administered LXXFP at 1.395, 2.79, and 5.58 g/(kg·d), respectively, by gavage once daily for two consecutive weeks. Skin lesion scores were evaluated based on erythema, scales, edema, and epidermal exfoliation, and scratching frequency was recorded. Skin lesion morphology was observed using hematoxylin and eosin (HE) staining and toluidine blue (TB) staining. Serum levels of high-mobility group box 1 (HMGB-1), immunoglobulin E (IgE), interleukin (IL)-17, tumor necrosis factor (TNF)-α, IL-4, IL-1β, IL-6, and interferon (IFN)-γ were measured by enzyme-linked immunosorbent assay (ELISA). Protein and mRNA expression levels of TNF-α, IL-4, IL-1β, IL-6, and IFN-γ in colonic tissues were determined by Western blot and reverse transcription-quantitative polymerase chain reaction (RT-qPCR), respectively. 16S ribosomal RNA (16S rRNA) sequencing was performed on fecal samples from the blank, syndrome, and LXXFP-H groups.

Results A total of 376 active components in LXXFP, 2 202 LXXFP-related drug targets, 220 AD-associated targets, and 62 overlapping targets between LXXFP and AD were identified through network pharmacology. Based on degree values in the PPI network, the top five core targets were TNF-α, IL-4, IL-1β, IL-6, and IFN-γ. GO enrichment analysis showed that the effects of LXXFP on AD may involve leukocyte migration and regulation of immune effector processes. KEGG enrichment analysis indicated potential involvement of inflammatory bowel disease, malaria, and cytokine-cytokine receptor interactions. Molecular docking indicated potential binding affinities between the selected active components and core target proteins, with the IL-4–dihydropinosylvin combination showing the lowest binding energy of − 7.5 kcal/mol, followed by IL-4–carvacrol at − 6.1 kcal/mol and IFN-γ–carvacrol at − 5.6 kcal/mol. In the animal experiment, compared with syndrome group, the three LXXFP groups showed a significant reduction in skin lesion scores and scratching frequency (P < 0.01). LXXFP promoted epidermal repair and crust disappearance, and alleviated acanthosis, inflammatory infiltration, and mast cell degranulation. Serum levels of HMGB-1, IgE, IL-17, TNF-α, IL-4, IL-1β, IL-6, and IFN-γ were significantly decreased in the three LXXFP groups compared with syndrome group (P < 0.05 or P < 0.01). Protein and mRNA expression levels of TNF-α, IL-4, IL-1β, IL-6, and IFN-γ in colonic tissues were also significantly decreased (P < 0.01). For α-diversity indices, the Simpson index exhibited a significant difference between the syndrome and LXXFP-H groups (P = 0.044), whereas the Chao1 index showed no statistical difference among the blank, syndrome, and LXXFP-H groups (P = 0.93). Principal coordinate analysis (PCoA) of Bray-Curtis distances demonstrated obvious separation of microbial communities among the three groups, and analysis of similarities (ANOSIM) confirmed significant differences in community structure (R = 0.71, P < 0.01). After LXXFP-H intervention, the abundances of Allobaculum, Lactobacillus, and Bifidobacterium showed a marked increase, whereas those of Adlercreutzia and Ruminococcaceae_Ruminococcus decreased compared with syndrome group (P < 0.05 or P < 0.01).

Conclusion LXXFP down-regulated the expression levels of the five inflammatory factors predicted by network pharmacology and modulated the gut microenvironment. Its therapeutic effects against AD with wind-dampness-heat syndrome may be associated with the reduction of inflammatory mediators and alterations in gut microbiota.