Psychiatric disorders including depression, anxiety disorder, schizophrenia affect approximately one in eight people globally. Schizophrenia influences about 24 million individuals worldwide and 0.45 % adults were diagnostic schizophrenia according to the report of World Health Organization (WHO). Patients with schizophrenia often exhibit persistent delusions and hallucinations, disorganized thinking and highly disorganized behavior. The life expectancy of schizophrenia patients is significantly lower, with a mortality rate two to three times higher than the general population due to suicide, substance abuse, depression, and other health complications. The exact mechanisms of schizophrenia remain unclear, but the gene and environment factors would induce the disease. Additionally, a lack of professional diagnoses and treatments can delay the effective management and cure of schizophrenia. Currently, the diagnosis of schizophrenia primarily relies on the assessment of patient behavior and symptoms, highlighting the need for more precise diagnostic tools and treatments.

The study of schizophrenia has mainly focused on brain pathological changes by neuroimaging and genetic research. Magnetic resonance imaging (MRI) studies have revealed thinner cortices and smaller subcortical volumes in patients with schizophrenia [1]. Okada et al [2] found that leftward asymmetry for thalamus, lateral ventricle, caudate and putamen volumes, and rightward asymmetry for amygdala and hippocampal volumes. Schizophrenia is a high heritability psychiatric disorder and genome-wide association studies (GWAS) has identified 270 distinct common genetic loci associated with schizophrenia [3]. Proteomics is beneficial for drug targets and putative biomarker discovery. In qualitative and quantitative proteomics research on prefrontal cortex of schizophrenia brain, GNA13-Erk pathway related proteins were found to be down-regulated, which might adversely affect synaptic plasticity [4]. In 4D-DIA MS analysis of 40 schizophrenia serum compared with 40 healthy people, 79 differentially expressed proteins were observed, where profilin-1, glyceraldehyde-3-phosphate dehydrogen and beta-actin-like protein 2 were found to have the potential ability to diagnose schizophrenia by machine learning with an AUC value of 0.972, accuracy of 0.88, specificity of 0.92 and sensitivity of 0.83 [5].

N-glycosylation in schizophrenia has not been deeply examined although it has been demonstrated that the brains of schizophrenia patients exhibit abnormal glycosylation [6]. The SLC39A8 gene, encoding ZIP8 protein, is a schizophrenia risk locus. The missense variant rs13107325 (C/T, p.Ala391Thr) in this gene induced abnormal regulation of metal ions concentration and cortical dendritic spine density decreased related with schizophrenia [[7], [8], [9]]. Mutated SLC39A8 impaired glycosylation resulting in reduced complexity and branching of N-glycan attached to plasma proteins. While in cortex, high-mannose N-glycan reduced and complex-type tri- and tetra-antennary N-glycans increased but the N-glycan in cerebellum behaved opposite [10,11]. In the brains of schizophrenia patients, several proteins such as excitatory amino acid transporters (EAATs) [12], gamma-aminobutyric type A receptor (GABAAR) subunits [13] and N-methyl-d-aspartate and kainate (KA) receptor subunits [14] were proved abnormally glycosylated suggesting putative signal transporting mistaken or abnormal neurotransmission. Glycosylation plays a critical role in neurodevelopmental, neurodegenerative, and neuropsychiatric disorders [15].

Site- and structure-specific N-glycoproteomics at the intact N-glycopeptide level has been the state of the art analytical pipeline for qualitative and quantitative characterization of differential N-glycosylation to provide comprehensive information of N-glycoproteins including N-glycosites, N-glycan sequence and linkage structures, as well as fold change; the pipeline has been successfully applied to various N-glycoproteome systems including cancer cell models, body fluids and tissues [[16], [17], [18], [19], [20], [21], [22]]. Here we report site- and structure-specific quantitative N-glycoproteomics study of differential N-glycosylation in the sera of schizophrenia patients. 7855 intact N-glycopeptides were identified; 1088 were differentially expressed (fold-change≥1.5, p-value≤0.05), and the corresponding N-glycoproteins were annotated to be primarily related with neurodegeneration as well as complement and coagulation cascades.