Identification of potential short-chain fatty acid biomarkers in Alzheimer's disease through bioinformatics analysis

Alzheimer's disease (AD) is a highly prevalent neurodegenerative disorder. Accumulating evidence suggests that short-chain fatty acids (SCFAs) can regulate the central nervous system, thereby affecting cognitive and behavior function.

This study aimed to investigate the association between the AD development and SCFA metabolism via bioinformatic analysis.

Gene expression profiles were obtained from the GEO database. 1243 genes related to SCFA were screened from Genecards database. Through weighted gene co-expression network analysis (WGCNA) and differential analysis, 10 SCFA hub genes were screened. Machine learning algorithms, including support vector machine recursive feature elimination (SVM-RFE) and least absolute shrinkage and selection operator (LASSO) regression models, were used to identify candidate biomarkers. The CIBERSORT algorithm was utilized to evaluate the infiltration of immune cells and its relationship with the potential biomarkers. The candidate biomarker chemicals were identified in the Comparative Toxicogenomics Database as underlying targeted drugs for treating AD.

Five genes-EZR, SNCA, GFAP, NFKBIA, and SST-were identified as potential biomarkers for AD through LASSO and SVM-RFE analyses. These genes can also be used to predict the risk of AD and have good diagnostic effects. The candidate biomarkers are associated with plasma cells, activated dendritic cells, M1 macrophages and resting natural killer cells. Notably, valproic acid and tretinoin were found to target these candidate genes, suggesting a new treatment approach for AD.

This study identified EZR, SNCA, GFAP, NFKBIA, and SST as potential key SCFA-related genes associated with the progression of AD, providing new insights into the prevention and treatment of AD.