G72 protein regulates D-serine, a potent co-agonist of the N-methyl-D-aspartate (NMDA) neurotransmission, which is implicated in the pathophysiology of schizophrenia. This study found that the peripheral G72 protein expression is distinctively higher in patients with schizophrenia than in healthy individuals.
Evidence from the views of neurodevelopment, pathological finding, genetic vulnerability, animal models, pharmacology and clinical trials support that glutamatergic signaling, particularly the molecules of the NMDA synapse, has an important role in the pathophysiology of schizophrenia.1, 2 In clinical trials, NMDA-enhancing agents improved symptomatology in patients with schizophrenia.3
The gene encoding G72 on chromosome 13q32–q34 is primate specific. It is a strong schizophrenia susceptibility gene confirmed over several ethnic populations.1, 4 G72 may contribute to the modulation of NMDA signaling. In vitro study indicates that G72 activates D-amino acid oxidase,5 which oxidizes D-amino acids such as D-serine. Individuals with overactive G72 could exhibit attenuated NMDA activity by lowering the co-agonist, predisposing them to schizophrenia.6 Consistently, D-serine levels in serum7 and cerebrospinal fluid8 are lower in patients with schizophrenia than in controls and adjuvant D-serine treatment can improve symptoms of schizophrenia.9 There is also an increase of G72 transcripts in the dorsolateral prefrontal cortex of schizophrenia patients.10
Peripheral protein expression may be a useful surrogate for protein expression in the central nervous system if the protein is expressed in both. To examine whether G72 is overexpressed in patients with schizophrenia, we measured G72 protein levels in peripheral plasma in unrelated patients with schizophrenia and healthy controls.
There were two independent cohorts of patients with schizophrenia and healthy controls: the learning set and the testing set. For examining possible drug effects on the protein expression, both psychotropic-free (for >3 months) and medicated schizophrenia patients were recruited. The G72 protein expression levels in the plasma were examined with western blot analysis.
The participants of the learning set were 30 unrelated healthy controls (healthy controls I) and 39 medicated schizophrenia patients. We selected 30 patients from the medicated schizophrenia group to match with healthy controls I by age and gender. The expression levels of G72 in the plasma of medicated schizophrenia patients were markedly higher than that of healthy controls I (Figure 1a). The receiver operating characteristic (ROC) analysis for the matched medicated schizophrenia patients versus healthy controls I determined an optimal cutoff value, 2.017, with an excellent sensitivity (0.967) and specificity (0.933) (area under the curve (AUC)=0.986; see Supplementary Information). The participants of the testing set were another 30 unrelated healthy controls (healthy controls II) and 27 drug-free schizophrenia patients. We selected 27 individuals from the healthy control II group to match with drug-free schizophrenia patients on age and gender. The expression levels of G72 in the plasma of drug-free schizophrenia patients were higher than that of healthy controls II (Figure 1b). The ROC analysis for the matched drug-free schizophrenia patients versus healthy controls II yielded an optimal cutoff value 2.131, providing a sensitivity of 0.778 and specificity of 0.963 (AUC=0.896; see Supplementary Information).
The study results suggest that patients with schizophrenia have a remarkable elevation of G72 protein expression in plasma when compared with healthy controls. The finding that is consistent with the study showing an increase in G72 transcripts in the postmortem brains from patients with schizophrenia10 supports that G72 is a susceptibility protein. To our knowledge, the current study is the first one indicating that the peripheral expression of a single protein may have the potential to be a diagnostic biomarker for schizophrenia with favorable sensitivity and specificity.
Psychotropic-free schizophrenic patients had similar, albeit insignificantly lower, G72 levels with medicated schizophrenia patients. Whether G72 protein level is affected by psychotropic use or the higher G72 protein level of the medicated schizophrenia patients is secondary to the longer illness duration needs further elucidation by prospective study. Whether G72 protein level is affected by circadian or ultradian fluctuations as well as diet also needs further investigation by blood sampling at different time points in a single individual. Our study is limited by its cross-sectional design and relatively small sample size. If reconfirmed by further studies in larger samples with various racial populations, G72 concentration in peripheral blood, joint with clinical assessment, might be able to assist physicians in diagnosing schizophrenia.
About this article
Supplementary Information accompanies the paper on the Molecular Psychiatry website (http://www.nature.com/mp)
Frontiers in Psychiatry (2018)