Aim: Depression is a psychiatric disease which is accompanied by metabolic disorder. Though depression has been widely studied, its metabolism is yet to be illustrated. We aimed to manifest the underlying mechanisms to diagnose depression.
Methods: One hundred thirty serum samples, including 65 patients and 65 healthy controls from different hospitals (training and validation cohorts), were recruited into the research. Sensitive Profiling for ChemoSelective Derivatization Carboxylomics (SPCSDCarboxyl) was applied to deeply hunt for the differential metabolites. Then, the serum, CSF, and hippocampus from depression rat models (CUMS group) were used to further confirm the results. Additionally, the co-occurrence between enzymes and biomarkers, as well as the combinatorial marker panel and the correlation of biomarkers among serum, CSF, or hippocampus were elucidated.
Results: Two hundred eight metabolites were identified from the sera of patients. Proline, 1-pyrroline-5-carboxylate (P5C), and glutamic acid could discriminate patients from healthy humans and were confirmed to be the potential biomarkers. After further validation through CUMS rats, proline, and P5C were enriched, while glutamic acid was depleted in the CUMS group. The co-occurrence analysis of enzymes and biomarkers indicated that they could be used for the diagnosis of depression. Moreover, the combinatorial marker panel and the correlation analysis of biomarkers between serum and CSF or between serum and hippocampus revealed that serum could be an alternative approach to directly reflect the potential physiological mechanisms and diagnose depression instead of brain samples.
Conclusion: These integrated methods may facilitate the identification of biomarkers and help manifest the underlying mechanisms of depression.
Keywords: SPCSDCarboxyl; biomarker identification; co-occurrence analysis; depression; iPath.
© 2022 The Authors. Psychiatry and Clinical Neurosciences © 2022 Japanese Society of Psychiatry and Neurology.