Search

Bipolar disorder is an often-severe mental health condition characterized by alternation between extreme mood states of mania and depression. Despite strong heritability and the recent identification of 64 common variant risk loci of small effect, pathophysiological mechanisms remain unknown. Here, we analyzed genome sequences from 41 multiply-affected pedigrees and identified variants in 741 genes with nominally significant linkage or association with bipolar disorder. These 741 genes overlapped known risk genes for neurodevelopmental disorders and clustered within gene networks enriched for synaptic and nuclear functions. The top variant in this analysis - prioritized by statistical association, predicted deleteriousness, and network centrality - was a missense variant in the gene encoding D-amino acid oxidase (DAOG131V). Heterologous expression of DAOG131V in human cells resulted in decreased DAO protein abundance and enzymatic activity. In a knock-in mouse model of DAOG131, DaoG130V/+, we similarly found decreased DAO protein abundance in hindbrain regions, as well as enhanced stress susceptibility and blunted behavioral responses to pharmacological inhibition of N-methyl-D-aspartate receptors (NMDARs). RNA sequencing of cerebellar tissue revealed that DaoG130V resulted in decreased expression of two gene networks that are enriched for synaptic functions and for genes expressed, respectively, in Purkinje neurons or granule neurons. These gene networks were also down-regulated in the cerebellum of patients with bipolar disorder compared to healthy controls and were enriched for additional rare variants associated with bipolar disorder risk. These findings implicate dysregulation of NMDAR signaling and of gene expression in cerebellar neurons in bipolar disorder pathophysiology and provide insight into its genetic architecture.

Lithium (Li) is the mainstay mood stabilizer for the treatment of bipolar disorder (BD), although its mode of action is not yet fully understood nor is it effective in every patient. We sought to elucidate the mechanism of action of Li and to identify surrogate outcome markers that can be used to better understand its therapeutic effects in BD patients classified as good (responders) and poor responders (nonresponders) to Li treatment. To accomplish these goals, RNA-sequencing gene expression profiles of lymphoblastoid cell lines (LCLs) were compared between BD Li responders and nonresponders with healthy controls before and after treatment. Several Li-responsive gene coexpression networks were discovered indicating widespread effects of Li on diverse cellular signaling systems including apoptosis and defense response pathways, protein processing and response to endoplasmic reticulum stress. Individual gene markers were also identified, differing in response to Li between BD responders and nonresponders, involved in processes of cell cycle and nucleotide excision repair that may explain part of the heterogeneity in clinical response to treatment. Results further indicated a Li gene expression signature similar to that observed with clonidine treatment, an α2-adrenoceptor agonist. These findings provide a detailed mechanism of Li in LCLs and highlight putative surrogate outcome markers that may permit for advanced treatment decisions to be made and for facilitating recovery in BD patients.

Bipolar (BP) disorder or manic depressive illness is a major psychiatric disorder for which numerous family, twin and adoption studies support a substantial genetic contribution. Recently, we reported the results of a genome-wide search for BP disorder susceptibility loci in 20 pedigrees. Suggestive evidence for linkage was found in this study at three markers on 13q, representing possibly two peaks separated by 18 cM. We have now collected a second set of 32 pedigrees segregating BP disorder and have tested for evidence of linkage to markers on human chromosome 13q. In this sample, we have replicated the linkage result in 13q32 at D13S154 (lod ¼ 2.29), the more proximal of the two original peaks. When all 52 pedigrees were combined, the multipoint maximum lod score peaked approximately 7 cM proximal to D13S154 (lod ¼ 3.40), with a second peak occurring between D13S225 and D13S796 (lod ¼ 2.58). There have been several other reports of significant linkage to both BP disorder and schizophrenia in this region of chromosome 13. These pedigrees provide additional evidence for at least one locus for BP disorder in 13q32, and are consistent with other reports of a possible genetic overlap between these disorders. Molecular Psychiatry (2003) 8, 558–564. doi:10.1038/ sj.mp.4001267