The neurodevelopmental regulatory role and clinical value of the multi-molecular axis in early-onset schizophrenia [circRNA]

Background: Biomarkers identified by previous epigenetic studies of schizophrenia have focused solely on diagnostics or pathology, potentially leading to a disconnection between them. This study focused on early-onset schizophrenia (EOS), aiming to identify the multi-molecular axis centered on circular RNAs (circRNAs), microRNAs (miRNAs), and messenger RNAs (mRNAs) with both neurobiological function and diagnostic value. Methods: Based on the discovery cohort and validation cohort-1, screening of candidate genes on the axis were completed via high-throughput sequencing, expressive and structural validation. Combining dual-luciferase reporter assays with overexpression/knockdown experiments, the axis with binding and regulatory relationships has been established. Subsequently, the functions of genes on the axis were explored through zebrafish embryo manipulation and neural differentiation. The clinical value of the entire axis was assessed in the validation cohort-2. Results: The regulatory effects of hsa-circ-CORO1C on hsa-miR-708-3p, hsa-miR-708-3p on target JARID2 and LNPEP were elucidated. Among them, hsa-miR-708-3p caused aberrant phenotypes including significant craniocerebral malformation and impaired neuron axon growth. JARID2 and LNPEP could facilitate neuronal differentiation and augment synaptic formation. The combined diagnostic efficacy of the whole axis, where hsa-circ-CORO1C could serve as a sponge for hsa-miR-708-3p to alleviate its suppressive effects on JARID2 and LNPEP, surpassed any individual gene we found in EOS. Conclusions: Our study demonstrated a multi-molecular axis, hsa-circ-CORO1C–hsa-miR-708-3p–JARID2+LNPEP, in EOS for the first time. By integrating evidence from genetic, neurophenotypic, and clinical perspectives, we have expanded the comprehension of the pathological mechanism and provided the reference for identifying reliable objective diagnostic biomarkers for EOS.