A genetic and proteomic approach to understanding neuropsychiatric symptoms in ADRD

Abstract

AbstractBackgroundThree lines of epidemiologic evidence link several psychiatric disorders to ADRD. First, experiencing schizophrenia or a mood/anxiety disorder in early/mid‐adulthood increases the risk of developing dementia by 6‐folds. Second, approximately 80% of ADRD patients experience neuropsychiatric symptoms (NPS), including depression, apathy, anxiety, irritability, agitation, delusions, and hallucinations. Third, these NPS overlap with symptoms of schizophrenia and mood/anxiety disorders. Given these observations, this study focuses on the question of whether these psychiatric disorders and ADRD have shared pathophysiology, which may underlie the manifestation of NPS in ADRDMethodTo investigate whether there is a shared genetic and molecular basis between the psychiatric conditions (major depressive disorder [MDD], bipolar disorder [BD], neuroticism, anxiety, PTSD, schizophrenia, and alcoholism) and ADRD (AD, Lewy body dementia [LBD], ALS, Parkinson’s disease [PD]), we performed three analyses. First, we examined genetic correlations between a psychiatric disorder and an ADRD using GWAS results in each condition and LDSC regression. Second, to identify causal genes and proteins for each of these brain conditions, we integrated each GWAS results with human brain proteomes (n = 722) using multiple independent state‐of‐the‐art integration approaches (PWAS, Mendelian randomization, colocalization analysis). Third, we examined protein‐protein interactions (PPIs) between a psychiatric causal protein and an ADRD causal protein using BIOGRID database.ResultFrom LDSC regression, we found 7 pairs of a psychiatric condition and an ADRD having significant genetic correlations, suggesting a shared genetic diathesis between these psychiatric conditions (MDD, BD, neuroticism, anxiety, PTSD, alcoholism) and ADRD (AD, LBD, ALS). Furthermore, among the causal proteins we identified for each condition, 30% of the ADRD causal proteins (13 of 42) were also causal proteins in the psychiatric conditions, highlighting a shared pathophysiology. Moreover, we found 2.6‐folds more PPIs between a psychiatric causal protein and an ADRD causal protein than by chance (p = 0.003, 120 PPIs).ConclusionTogether, these 3 levels of novel evidence highlight a shared genetic and molecular basis between several psychiatric conditions and ADRD, which may underlie the pathophysiology of NPS in ADRD. The 13 shared causal proteins are promising targets for further validation in NPS in ADRD and towards the goal of developing effective therapeutics for NPS.