Background:
Alzheimer's disease (AD) is a neurological condition that worsens with time. It is distinguished by the buildup of improperly folded amyloid beta protein, which produces amyloid plaques and promotes tau protein aggregation into neurofibrillary tangles in the brain. Certain lipids are critically involved in the development of AD, for instance, apolipoprotein E (ApoE), which is involved in lipid metabolism and cholesterol metabolism in the body and functions as a genetic risk factor for AD. However, there is currently a lack of systematic understanding of how Lipid Metabolism-related Genes (LMGs) contribute to AD. This study aimed to identify potential biomarkers for early AD diagnosis to address this knowledge gap, specifically focusing on those associated with immune cell infiltration. Machine-learning algorithms and Gene Expression Omnibus (GEO) database were utilized to achieve this goal.
Methods:
Differentially expressed genes (DEGs) between healthy and AD patient samples were identified. The process involved using two expression profiles retrieved from the Gene Expression Omnibus(GEO)database, including the publicly available GSE5281 and GSE138260. Furthermore, functional enrichment analysis was conducted. The samples were randomly classified to either the training or the validation sets to ensure reliable results. This involved assessing Lipid Metabolism-related DEGs (LMDEGs) between AD and healthy tissues. To identify the potential gene biomarker, the machine-learning algorithm Support Vector Machine-Recursive Feature Elimination (SVM-RFE) and the Least Absolute Shrinkage and Selection Operator (LASSO) regression model were applied.The algorithm called cell-type identification by estimating relative subsets of RNA transcripts (CIBERSORT) was utilized to assess the role of infiltrating immune cells and their association with the gene biomarker. Finally, the Integrated Traditional Chinese Medicine (ITCM) database was utilized to identify associated Chinese medicines and evaluate the association of related herbal medicines with lipid metabolism and AD.
Results:
A total of 137 genes were extracted from 751 LMGs that depicted a strong correlation with autophagy regulation and immune response. LASSO and SVM-RFE were employed to determine choline acetyl transferase (CHAT), member RAS oncogene family (RAB4A), acyl-coA binding domain-containing protein 6 (ACBD6), and alpha-galactosidase A (GLA) as marker genes among the 137 AD-related genes. These genes exhibited promising treatment capabilities.Functional enrichment analysis was conducted to explore their involvement in various processes. The data indicated a link between these genes and the cell cycle, amino acid metabolism, immune response regulation, and various processes associated with AD pathogenesis.Moreover, nine Chinese Medicine drugs were identified that specifically target the four marker genes. Furthermore, the ceRNA network revealed an intricate regulatory correlation involving these marker genes. Additionally, CIBERSORT analysis implied that alterations in the immune microenvironment of individuals with AD may be associated with CHAT.
Conclusion:
These findings hold significant diagnostic potential and provide insights into the mechanism of AD. Nonetheless, additional research is essential to assess the diagnostic value of these markers for clinical application. This research sheds light on the involvement of LMGs in the mechanism involved in AD pathogenesis. The acquired data aid the treatment of the disease by providing potential therapeutic targets. This research contributes to a better comprehension of the disease and its underlying processes. Furthermore, it facilitates further research by providing a foundation for developing more effective strategies for diagnostic and therapeutic purposes.