Decreased mitochondrial function and increased brain inflammation in bipolar disorder and other neuropsychiatric diseases.

A recent review article by Konradi and coworkers 1 summarized the results of gene expression studies based on analysis of the entire mRNA pool (transcriptome) of brains from patients with bipolar disorder. The transcriptome differs from the genome because it can be shaped by environmental influences. The authors concluded that clusters of genes (a) for energy production were downregulated, (b) for immune responses were up-regulated, and (c) for oligodendrocyte function were mostly down-regulated (Fig. 1). They compared these general outcomes with similar available data from other authors for major depressive disorder and schizophrenia and concluded that there was significant overlap, making it impossible to identify a specific profile unique for bipolar disorder. Konradi and coworkers reviewed 7 published studies between 2003 and 2009 using brain material and 4 studies using peripheral tissue from patients with bipolar disorder. These studies did not all examine the same clusters. Only 3 of the 7 studies using brain tissue investigated ‘‘markers of energy function’’; of these, 2 reported significant decrease, whereas 1 study reported an increase. Only 3 of the 7 studies using brain tissue investigated expression of genes associated with immune response, and all 3 reported significant increase. Only 1 study investigated the expression of oligodendrocyte markers and found decreased expression of some but increased level of others. Four studies using peripheral tissues were reviewed; of these, 2 studies showed decreased expression of genes regulating energy production, and only 1 study investigated genes associated with immune responses and reported significantly increased expression. No particular gene was singled out for discussion. However, the authors proposed that the transcriptome changes observed may be the early stage of an autoimmune process that is more dramatic, but similar to changes in the neuroinflammatory disease multiple sclerosis (MS). The authors correctly discussed a number of limitations in the articles reviewed. These included the postmortem nature of the tissues that could have introduced changes related to cell death, the difference in time period between death and tissue preservation, difference in the cause of death, the possible effect of any other comorbid conditions and/or treatment, lack of information on use of tobacco or recreational drugs, and the lack of ‘‘controls.’’ In addition, the mean age of most patients was 48 T 30 years, which introduces a possible contribution of advanced age. Finally, the number of patients whose brains were analyzed varied considerably from 8 to 35. The role of inflammation, especially proinflammatory cytokines, in neuropsychiatric diseases was long suspected, but these recent results certainly provide new robust evidence. The findings about decreased energy production may be quite significant. For instance, recent publications reported decreased mitochondrial function in as many as 60% of patients with autism, a neuropsychiatric developmental disorder showing many similarities with schizophrenia. In addition, the brains of autistic patients had evidence of neuroinflammation. We recently showed that serum of autistic patients has high levels of mitochondrial DNA known to stimulate autoinflammatory reactions. Neurotensin, a peptide first isolated from the brain, has also been implicated in the pathogenesis of depression and schizophrenia, is elevated in the serum of autistic children, and can stimulate mast cells. Mast cells are involved in inflammation and were recently reported to undergo mitochondrial fission and translocation during degranulation to the cell surface with subsequent release of ATP and mitochondrial DNA. ATP is considered by some a universal ‘‘alarm’’signal from cells under stress, can affect neighboring cells, and can sustain GUEST EDITORIAL