Alterations in redox modulation are consistently reported in bipolar disorder (BD). MicroRNAs are targeted regulators of gene expression.

We aimed to examine if microRNAs that target redox modulators can discriminate between BD and healthy controls.

Data from brains of individuals with and without BD were obtained from Array Collection datasets. MicroRNAs targeting redox modulators were assessed for their ability to discriminate BD from the control group using machine-learning algorithms. Methylation of microRNAs, expression of their transcription factors and redox targets were assessed with ANCOVA with FDR correction. For validation, we acquired plasma samples belonging to 2 families of individuals with and without BD (n=9). Plasma microRNAs were sequenced using the Ion Total RNA Sequencing Kit (Thermo Fisher Scientific), and microRNAs identified from the in silico analysis were examined in the validation dataset.

We identified 5 miRNAs (hsa-miR-299, hsa-miR-125a, hsa-miR-145, hsa-miR-30b, hsa-miR424) that were common in two of the four in silico datasets. Target genes glutathione peroxidase 4, ATP5A1, ATP5G1, NDUFS1, NDUFC2, and catalase were expressed at different levels between BD and the control group. Furthermore, our results showed that transcription factors CTCF and USF1 might control the expression of hsa-miR-145, while methylation differences were not found. Finally, hsa-miR-30b was significantly increased in the plasma of patients with BD compared to controls in the validation experiment.

Our study demonstrates that microRNAs may have an important role in the initiation of redox changes in BD.