| 摘要: | Background
Alzheimer's disease (AD) presents a significant global health challenge, necessitating advancements in diagnostic and therapeutic strategies. This dissertation presents findings from integrated bioinformatic approaches in transcriptomic analysis and conducted proteomic validation experiments to identify potential biomarkers for AD diagnosis and treatment.
Methods and Results
The early study focused on AD, employing a comprehensive bioinformatic strategy to integrate transcriptomic data from diverse sources, including microarray and single-cell datasets from blood and tissue samples. Nineteen candidate genes with consistent expression changes across both datasets were identified, suggesting their potential as biomarkers for AD. Single-cell sequencing highlighted their specific expression in neuronal cell types, emphasizing their relevance to AD pathology. Functional enrichment analysis revealed downregulation of three candidate genes in the synaptic signaling pathway. Proteomic validation through quick screening of serum samples showed a significant downregulation of rabphilin-3A (RPH3A) in the AD group compared to the healthy control (HC) group. In animal models, analysis of mouse brain tissue, particularly in the hippocampus, demonstrated a significant reduction in RPH3A levels. Further investigation in a larger cohort of AD patients, using a total of 121 serum samples from a case-control study at Taipei Medical University Hospital, consistently showed decreased RPH3A levels in AD patients compared to healthy controls. These findings suggest that RPH3A may serve as a reliable indicator of AD pathology. The later study explored the role of phosphatidylinositol-5-phosphate 4-kinase type 2 alpha (PIP4K2A) in the context of AD complicated with Type 2 Diabetes mellitus (T2DM). By integrating transcriptomic data from multiple cohorts, including blood and tissue samples microarray and single cell sequencing datasets, PIP4K2A was identified as predominantly expressed in oligodendrocytes. Functional enrichment analysis also indicated the potential involvement of this gene in vesicle-mediated transport. Proteomic validation, assessing PIP4K2A levels in serum samples from individuals with hyperglycemia, AD, and those afflicted by both conditions, revealed a marked elevation in PIP4K2A across these patient groups compared to healthy subjects. These observations propose PIP4K2A as a potential biomarker and a target for therapeutic intervention in AD and T2DM.
Conclusion
These studies underscore the integration of transcriptome bioinformatics and proteomic validation in elucidating the significance of identified genes and pathways in AD pathogenesis. The distinct outcomes for RPH3A and PIP4K2A provide a deeper understanding of AD as well as AD in the complicated complex with T2DM. Further investigation into the molecular mechanisms of these candidate biomarkers is essential, as it holds the potential to open new avenues for diagnostics in the future.
Background
Alzheimer's disease (AD) presents a significant global health challenge, necessitating advancements in diagnostic and therapeutic strategies. This dissertation presents findings from integrated bioinformatic approaches in transcriptomic analysis and conducted proteomic validation experiments to identify potential biomarkers for AD diagnosis and treatment.
Methods and Results
The early study focused on AD, employing a comprehensive bioinformatic strategy to integrate transcriptomic data from diverse sources, including microarray and single-cell datasets from blood and tissue samples. Nineteen candidate genes with consistent expression changes across both datasets were identified, suggesting their potential as biomarkers for AD. Single-cell sequencing highlighted their specific expression in neuronal cell types, emphasizing their relevance to AD pathology. Functional enrichment analysis revealed downregulation of three candidate genes in the synaptic signaling pathway. Proteomic validation through quick screening of serum samples showed a significant downregulation of rabphilin-3A (RPH3A) in the AD group compared to the healthy control (HC) group. In animal models, analysis of mouse brain tissue, particularly in the hippocampus, demonstrated a significant reduction in RPH3A levels. Further investigation in a larger cohort of AD patients, using a total of 121 serum samples from a case-control study at Taipei Medical University Hospital, consistently showed decreased RPH3A levels in AD patients compared to healthy controls. These findings suggest that RPH3A may serve as a reliable indicator of AD pathology. The later study explored the role of phosphatidylinositol-5-phosphate 4-kinase type 2 alpha (PIP4K2A) in the context of AD complicated with Type 2 Diabetes mellitus (T2DM). By integrating transcriptomic data from multiple cohorts, including blood and tissue samples microarray and single cell sequencing datasets, PIP4K2A was identified as predominantly expressed in oligodendrocytes. Functional enrichment analysis also indicated the potential involvement of this gene in vesicle-mediated transport. Proteomic validation, assessing PIP4K2A levels in serum samples from individuals with hyperglycemia, AD, and those afflicted by both conditions, revealed a marked elevation in PIP4K2A across these patient groups compared to healthy subjects. These observations propose PIP4K2A as a potential biomarker and a target for therapeutic intervention in AD and T2DM.
Conclusion
These studies underscore the integration of transcriptome bioinformatics and proteomic validation in elucidating the significance of identified genes and pathways in AD pathogenesis. The distinct outcomes for RPH3A and PIP4K2A provide a deeper understanding of AD as well as AD in the complicated complex with T2DM. Further investigation into the molecular mechanisms of these candidate biomarkers is essential, as it holds the potential to open new avenues for diagnostics in the future. |
