# Identification of Disulfidptosis‐Associated Hub Genes in Psoriasis via Integrated Transcriptomic and Experimental Validation Approaches

## TL;DR

This study explores how disulfidptosis, a new type of cell death, is linked to psoriasis by analyzing gene activity and validating results in a mouse model.

## Contribution

The study identifies disulfidptosis-associated hub genes in psoriasis using integrated transcriptomic and experimental approaches.

## Key findings

- Seven disulfidptosis-related genes were differentially expressed in psoriatic lesions.
- Five candidate genes (GYS1, SLC3A2, FLNA, FLNB, and TLN1) showed strong discriminatory power in distinguishing psoriasis cases.
- Protein-level validation confirmed transcriptional upregulation of key genes in a mouse model of psoriasis.

## Abstract

Psoriasis is a chronic, immune‐mediated skin disease. While redox imbalance has been implicated in its pathogenesis, the involvement of newly identified cell death pathways remains unclear. Disulfidptosis, a novel form of regulated cell death driven by intracellular disulfide stress, has recently been linked to inflammatory and metabolic diseases, yet its role in psoriasis is unknown. We conducted an integrative analysis combining bulk transcriptomic data (GSE106992 and GSE11239) and single‐cell RNA sequencing data (GSE162183) to identify disulfidptosis‐related genes (DRGs) associated with psoriasis. Differentially expressed genes were intersected with a curated DRG list. Feature selection was performed using LASSO regression and random forest algorithms, followed by WGCNA to identify key modules. Immune infiltration and cell‐type‐specific expression were assessed, and findings were validated using an imiquimod‐induced psoriasis‐like mouse model through Western blotting. To investigate the role of disulfidptosis in psoriasis, we integrated bulk and single‐cell RNA sequencing datasets from psoriatic lesional and non‐lesional skin. Differential expression analysis, WGCNA, and machine learning approaches (LASSO regression and random forest) were employed to identify DRGs. Immune cell infiltration patterns were analyzed, and gene expression was validated in an imiquimod‐induced psoriasis‐like mouse model using Western blotting. Seven DRGs were differentially expressed in psoriatic lesions. Machine learning approaches converged on five candidate genes, with GYS1, SLC3A2, FLNA, FLNB, and TLN1 showing strong discriminatory power (AUC > 0.7). WGCNA identified four hub genes (GYS1, SLC3A2, TLN1, and FLNB) associated with disease‐relevant gene modules. Immune infiltration analysis revealed significant correlations between DRG expression and key immune subsets, including activated CD4+ memory T cells and M1 macrophages. Single‐cell RNA‐seq confirmed cell type–specific enrichment of DRGs in epidermal, mesenchymal, and immune cell populations. Protein‐level validation in the murine model further supported transcriptional upregulation of these genes in psoriatic lesions. Our findings suggest that disulfidptosis‐related pathways may contribute to psoriasis pathogenesis through interactions with immune infiltration. The identified hub genes, particularly GYS1 and SLC3A2, represent potential biomarkers and therapeutic targets, offering new insight into the complex molecular landscape of psoriasis.