Differential gene expression analysis in individuals with depression in Alzheimer’s disease using RNAseq

Abstract

AbstractBackgroundDepression in individuals with Alzheimer’s disease (AD) is common, distressing, difficult to treat, may increase carer burden and is inadequately understood. It occurs more frequently in AD than in older adults without dementia. The underlying biology remains obscure.We aimed to identify patterns of differential gene expression in individuals with depression in AD compared to individuals with AD but no history of depressive illness.MethodWe used data from the Religious Orders Study and the Memory and Aging Project: the ROSMAP cohort. Individuals with other disorders which could have caused cognitive impairment e.g. Parkinson’s disease(n = 122) or clinically significant head trauma (n = 234) were excluded, and those with bipolar disorder(n = 13). Those with neuropathological or consensus clinical evidence of a dementia other than Alzheimer’s disease were also excluded (n = 116).Depression caseness was defined as a highly probable/probable clinical diagnosis of depression made by the assessing clinician at a study visit or a CESD score of >7 at a study visit. Individuals with treated depression (n = 301) who were taking antidepressants but never met depression caseness criteria and 380 individuals with subthreshold depressive symptoms on the CESD but a clinical assessment that depression was “not present” were also excluded.RNAseq data was available from the DLPFC and posterior cingulate. The posterior cingulate analysis is ongoing. Fastq files were quality controlled using fastqc and Picard following HISAT2 alignment to GRCh38. After exclusion of files which failed QC fastq files were aligned to the GRCh38 transcriptome using Salmon. Files with low transcriptome alignment (<28%) were also excluded. Differential gene expression analysis was carried out in Rstudio using DEseq2.ResultAfter exclusions and QC, data was available for 15 individuals with depression & AD and 25 individuals with AD but no depression for the DLPFC. Age, RIN, sex and Braak stage were included in the analysis model. After adjustment of the p value for multiple comparison 64 genes were differentially expressed in depression including Fbox proteins, nucleosome assembly proteins and voltage‐gated channel proteins. KEGG pathway analysis is underwayConclusionWe have identified genes differentially expressed in depression in AD which may allow better understanding of this condition.