Constructing a prognostic risk model for Alzheimer’s disease based on ferroptosis

Abstract

Introduction: The aim of this study is to establish a prognostic risk model based on ferroptosis to prognosticate the severity of Alzheimer’s disease (AD) through gene expression changes. Methods: The GSE138260 dataset was initially downloaded from the Gene expression Omnibus database. The ssGSEA algorithm was used to evaluate the immune infiltration of 28 kinds of immune cells in 36 samples. The up-regulated immune cells were divided into Cluster 1 group and Cluster 2 group, and the differences were analyzed. The LASSO regression analysis was used to establish the optimal scoring model. Cell Counting Kit-8 and Real Time Quantitative PCR were used to verify the effect of different concentrations of Aβ1–42 on the expression profile of representative genes in vitro. Results: Based on the differential expression analysis, there were 14 up-regulated genes and 18 down-regulated genes between the control group and Cluster 1 group. Cluster 1 and Cluster 2 groups were differentially analyzed, and 50 up-regulated genes and 101 down-regulated genes were obtained. Finally, nine common differential genes were selected to establish the optimal scoring model. In vitro, CCK-8 experiments showed that the survival rate of cells decreased significantly with the increase of Aβ1–42 concentration compared with the control group. Moreover, RT-qPCR showed that with the increase of Aβ1–42 concentration, the expression of POR decreased first and then increased; RUFY3 was firstly increased and then decreased. Discussion: The establishment of this research model can help clinicians make decisions on the severity of AD, thus providing better guidance for the clinical treatment of Alzheimer’s disease.