Transcriptional signatures of the cortical morphometric similarity network gradient in first-episode, treatment-naive major depressive disorder

• DOI: 10.1038/s41386-022-01474-3
• Publication Date: 2022-10-17T07:03:09Z
• Authors: Kaizhong Xue; Lining Guo; Wenshuang Zhu; Sixiang Liang; Qiang Xu; Lin Ma; Mengge Liu; Yong Zhang; Feng Liu
• Keywords: Cerebral Cortex; Depressive Disorder, Major; Brain Mapping; 03 medical and health sciences; 0302 clinical medicine; Humans; Brain; Magnetic Resonance Imaging; 3. Good health

Recent studies have shown that major depressive disorder (MDD) is accompanied by alterations in functional and structural network gradients. However, whether changes are present in the cortical morphometric similarity (MS) network gradient, and the relationship between alterations of the gradient and gene expression remains largely unknown. In this study, the MS network was constructed, and its gradient was calculated in 71 patients with first-episode, treatment-naive MDD, and 69 demographically matched healthy controls. Between-group comparisons were performed to investigate abnormalities in the MS network gradient, and partial least squares regression analysis was conducted to explore the association between gene expression profiles and MS network gradient-based alternations in MDD. We found that the gradient was primarily significantly decreased in sensorimotor regions in patients with MDD compared with healthy controls, and increased in visual-related regions. In addition, the altered principal MS network gradient in the left postcentral cortex and right lingual cortex exhibited significant correlations with symptom severity. The abnormal gradient pattern was spatially correlated with the brain-wide expression of genes enriched for neurobiologically relevant pathways, downregulated in the MDD postmortem brain, and preferentially expressed in different cell types and cortical layers. These results demonstrated alterations of the principal MS network gradient in MDD and suggested the molecular mechanisms for structural alternations underlying MDD.