Cell-state-resolved transcriptomic analysis reveals macrophage-centered pyroptosis-related inflammatory programs in
Xinxin Li1, Fu Gui2, Zheng Zheng1
1Department of Ophthalmology, Shanghai General Hospital, Shanghai Jiao Tong University, School of Medicine, Shanghai, China; National Clinical Research Center for Eye Diseases, Shanghai, China; Shanghai Key Laboratory of Fundus Diseases, Shanghai, China; Shanghai Municipal Hospital Ophthalmology Specialist Alliance, Shanghai, China.
Background:
Age-related macular degeneration (AMD) is accompanied by inflammatory changes in the retinal pigment epithelium/choroid complex, but the cellular sources of pyroptosis-related transcriptional programs in human AMD tissue remain unclear. This study profiled these programs at single-cell resolution and explored candidate regulatory molecules.
Methods:
We analyzed the human retinal pigment epithelium (RPE)/choroid single-cell RNA-sequencing dataset GSE135922 to define cell clusters, pyroptosis-related genes, and regulons. AUCell was applied to estimate pyroptosis-related signature activity in individual cell types. The macrophage cluster with the highest score was examined by pathway enrichment, subclustering, and Monocle 2 pseudo-time analysis, and SCENIC-based regulon analysis was used to infer candidate transcriptional regulators. Pyroptosis-related genes and transcription factors were also evaluated in T-cell, endothelial-cell, and fibroblast subclusters. Bulk RNA sequencing and immunofluorescence in a laser-induced choroidal neovascularization (CNV) mouse model were used for supportive evidence.
Results:
Across human RPE/choroid cell clusters, 60 cluster-specific pyroptosis-related marker genes were detected. At the cell-type ranking level, macrophages, T cells, endothelial cells, and fibroblasts showed relatively higher pyroptosis-related signature activity across clusters, with the highest signal in Macrophages-2 and lower activity in RPE cells. Marker genes of Macrophages-2 were enriched in immune and inflammatory pathways, including complement and coagulation cascades, NOD-like receptor signaling, and NF-κB signaling. Pseudo-time analysis resolved Macrophages-2 into divergent trajectories, and NLRP3 was enriched in one post-branch state, consistent with macrophage state heterogeneity rather than uniform activation. IRF1, STAT3, and NEAT1 recurred in cell-type-specific analyses, and Irf1/Stat3 protein signals were higher in CNV lesions.
Conclusion:
These findings indicate a macrophage-centered, cell-state-specific pattern of pyroptosis-related inflammatory remodeling in AMD. Branch-associated NLRP3 inflammasome signatures, together with IRF1, STAT3, and NEAT1, define candidate molecular features that warrant further mechanistic evaluation.
