Data Mining-based Identification of Key Genes Mediating the Anti-Alzheimer’s Effects of Polysaccharides from Cistanche Deserticola

OBJECTIVE  To screen the key hub genes of Alzheimer’s disease(AD) through data mining, to assess their potential in AD diagnosis and treatment, and to verify the role and mechanism of the hub genes and Polysaccharides from Cistanche deserticola(CDPS) on the improvement of cognitive function in mice modeled with AD through in vivo experiments.

METHODS  This study utilized bioinformatics to obtain gene expression data for healthy control(Control) and Alzheimer’s disease(AD) groups from the GEO database. Differential expression analysis and weighted gene co-expression network analysis (WGCNA) were performed on brain tissue samples from the GSE118553 dataset. Twenty AD-associated hub genes were screened, and their potential roles in biological processes and correlation with cognitive impairment were explored via GO enrichment and GSEA analyses. Eight genes were identified as AD diagnostic biomarkers by stepwise regression and logistic regression analysis. Additionally, immune cell infiltration in the samples was further investigated. The in vivo experiments were conducted using the APP/PS1 Tg mouse model of Alzheimer’s disease. The animals were divided into the following groups: transgenic AD animal model group(Tg), CDPS low(CPL 25 mg∙kg⁻¹∙d⁻¹), medium(CPM 50 mg∙kg⁻¹∙d⁻¹) and high(CPH 100 mg∙kg⁻¹∙d⁻¹) dosage groups, and a wild-type control group(WT). The effects of CDPS on cognitive function in mice were assessed via novel object recognition and Morris water maze tests. Serum levels of inflammatory cytokines and antioxidant enzymes were measured using kits, whilst pathological alterations in brain tissue were observed.

RESULTS  Twenty AD-associated hub genes were screened and eight of them with the most significant differences between the two groups were selected to establish a diagnostic model(MICAL2, C1QTNF5, FGF9, CHGB, ABCC8, SNORD109A, TGFBR2, and CCDC184) and observed in the immune cell infiltration assay that these genes may also play an important role in AD immunity. In vivo experiments revealed that the mRNA levels of CCDC184 and ABCC8 were significantly decreased, and the mRNA levels of C1QTNF5 and TGFBR2 were significantly increased in the CPH group of mice after high-dose administration of CDPS. Behavioral experiments showed that CDPS at high doses improved the short-term memory capacity of 5-month-old AD mice and improved the number of traversed platforms, the time when the target quadrant was located, the 1–5 d escape latency, and the index of short-term memory capacity at 8 months of age. ELISA experiments showed that CDPS increased the levels of IL-10 and TGF-β in the brain tissues, and decreased the levels of IL-1β and IL-6. Serum experiments showed that CDPS significantly increased the levels of SOD, GSH-Px, CAT and decreased the level of MDA. Pathological observations showed that Aβ plaques were reduced and neuronal apoptosis was significantly reduced in the CDPS treatment group.

CONCLUSION Bioinformatics approaches validated the effects of CDPS on AD-related hub genes at the genetic level. CDPS improves the learning memory capacity of APP/PS1 mice, which may be achieved by regulating the levels of inflammatory factors and inhibiting the levels of lipid peroxidation, among other pathways.