Multi-omics study of ABCA1 in sepsis: Integrating mendelian randomisation, single-cell transcriptomics, in silico
1Xingtai People's Hospital, Hebei Medical University, Xingtai, Hebei Province, China.
Background:
Sepsis is a life-threatening organ dysfunction caused by a dysreg- ulated host response to infection, with the immune status dynamically evolving from an early hyperinflammatory phase to a late immunosuppressive phase. ABCA1 (ATP-binding cassette transporter A1) is involved in cholesterol efflux and inflammatory regulation, but its causal relationship with sepsis, cellular ori- gin, and potential for drug targeting remain unclear. This study is the first to integrate Mendelian randomisation, single-cell transcriptomics, in silico knock- out, and virtual screening to systematically investigate ABCA1 in sepsis from genetics, cell biology, mechanisms, and translational potential.
Methods:
A two-sample Mendelian randomisation (MR) analysis was per- formed using blood eQTLs of ABCA1 as the exposure and a sepsis GWAS as the outcome. Public single-cell dataset GSE151263 was analysed using Seurat to localise ABCA1 expression. In silico knockout of ABCA1 was performed in CD16 + monocytes, followed by Gene Ontology (GO) enrichment analysis. A two-stage virtual screening strategy was employed using DrugRep for initial screening of the FDA-approved drug library, followed by refined docking of the top 10 candidates using CB-Dock2.
Results:
MR suggested a protective trend of ABCA1 against sepsis (OR = 0.809, 95% CI 0.635-1.032, P = 0.088), with no heterogeneity or horizontal pleiotropy detected. Single-cell analysis of 25,234 cells revealed that ABCA1 was predominantly expressed in CD16 + monocytes (P < 0.001 vs. other im- mune subsets). Following in silico knockout, only three genes showed significant changes (ABCA1 log2 FC ≈ __15.8, SEC14L1 log2 FC ≈ 10.2, PTGES log2 FC ≈ 9.4; all Padj < 0.001), enriched in lipid transport and fever-related path- ways. Virtual screening of approximately 2500 FDA-approved drugs identified six drugs with strong binding to ABCA1, including Dihydroergotamine (-12.2 kcal/mol) and Nilotinib (-11.0 kcal/mol).
Conclusion:
ABCA1 is specifically expressed in CD16 + monocytes, and its perturbation affects lipid-related pathways. Virtual screening has identified potential drug candidates led by Dihydroergotamine (-12.2 kcal/mol). This study offers multi-level evidence supporting ABCA1 as a protective molecule and therapeutic target in sepsis.