Integrative metabolomics-genomics analysis identifies key networks in a stem cell-based model of schizophrenia.

Schizophrenia (SCZ) is a neuropsychiatric disorder, caused by a combination of genetic and environmental factors. The etiology behind the disorder remains elusive although it is hypothesized to be associated with the aberrant response to neurotransmitters, such as dopamine and glutamate. Therefore, investigating the link between dysregulated metabolites and distorted neurodevelopment holds promise to offer valuable insights into the underlying mechanism of this complex disorder. In this study, we aimed to explore a presumed correlation between the transcriptome and the metabolome in a SCZ model based on patient-derived induced pluripotent stem cells (iPSCs). For this, iPSCs were differentiated towards cortical neurons and samples were collected longitudinally at various developmental stages, reflecting neuroepithelial-like cells, radial glia, young and mature neurons. The samples were analyzed by both RNA-sequencing and targeted metabolomics and the two modalities were used to construct integrative networks in silico. This multi-omics analysis revealed significant perturbations in the polyamine and gamma-aminobutyric acid (GABA) biosynthetic pathways during rosette maturation in SCZ lines. We particularly observed the downregulation of the glutamate decarboxylase encoding genes GAD1 and GAD2, as well as their protein product GAD65/67 and their biochemical product GABA in SCZ samples. Inhibition of ornithine decarboxylase resulted in further decrease of GABA levels suggesting a compensatory activation of the ornithine/putrescine pathway as an alternative route for GABA production. These findings indicate an imbalance of cortical excitatory/inhibitory dynamics occurring during early neurodevelopmental stages in SCZ. Our study supports the hypothesis of disruption of inhibitory circuits to be causative for SCZ and establishes a novel in silico approach that enables for integrative correlation of metabolic and transcriptomic data of psychiatric disease models.
(© 2024. The Author(s).)