Integration of gene expression with computational modeling reveals region-specific brain energy metabolism dysregulation in schizophrenia

Abstract

Schizophrenia is a highly heritable psychiatric disorder with suggested disturbances in multiple neurobiological pathways including brain energy metabolism, which is essential for sustaining precise neuronal signaling. The relationship between gene expression alterations and energy metabolism defects, however, is poorly understood. Here we combined differential expression (DE) analysis with computational modeling to investigate how the expression of cytosolic energy metabolism-associated genes influences the concentrations of key energy metabolites. The DE analysis used human post-mortem RNA sequencing data from the anterior cingulate cortex (ACC) and dorsolateral prefrontal cortex (DLPFC) from the CommonMind Consortium. Metabolic simulations were carried out using a biophysical model of brain energy metabolism (Winter et al. 2018). We conducted both population-average and subject-specific simulations and assessed the effects of DE genes individually and collectively. The DE analysis identified nine genes in the ACC and 24 in the DLPFC. Approximately two thirds of the DE genes were downregulated in schizophrenia in both brain areas. Decreased expression of neuronal LDHB in the ACC caused a decrease of pyruvate and an increase of NAD+/NADH ratio. Furthermore, decreased expression of neuronal PFKM in the DLPFC caused an increase of glucose and NAD+/NADH ratio and a decrease of pyruvate, lactate, and ATP. These findings provide evidence for region-specific dysregulation of brain energy metabolism in schizophrenia and suggest mechanistic links between gene expression changes and altered levels of key energy metabolites. AUTHOR SUMMARYSchizophrenia is a mental disorder that affects perception, thought, and behavior, but its disease mechanisms are still not fully understood. One important aspect of brain function is energy metabolism, which powers nearly all cellular processes and supports communication between neurons. In this study, we explored how changes in the expression of specific genes involved in energy production influence the levels of key molecules in the brain. We combined human gene expression data from two brain regions with computer simulations of neuronal function. We found that certain genes had a strong impact on molecules such as glucose, pyruvate, lactate, ATP, and NAD+/NADH balance. Importantly, the effects were specific to particular brain regions. Still, the majority of gene expression changes had non-significant effects. By connecting gene expression with brain energy metabolism simulations, this work provides new insights into the biological mechanisms of schizophrenia and may guide future research aimed at understanding and improving brain function in the disorder.