Significant outcomes
CA, TCDCA and TDCA significantly decreased in schizophrenia patients.
A combination of CA, TCDCA and TDCA with a classification formula could predict correctly classified cases, and the accuracy of prediction was up to 95.83%.
Limitations
Limited sample size.
Different diet habits and nutriture might have an effect on the concentration of various metabolites in blood.
Introduction
Schizophrenia is a debilitating mental disorder with an overall prevalence estimate of 0.3–0.66%, but its etiology remains to be elucidated. Currently, diagnosis of schizophrenia is based on clinical interviews and careful observations (Girgis et al. Reference Girgis, Kumar and Brown2014). These descriptive methods are subjective and variable, which can lead to delay in diagnosis and/or misdiagnosis (Weickert et al. Reference Weickert, Weickert, Pillai and Buckley2013). To date, studies have particularly explored peripheral blood, due to its accessibility and ease in procurement, for the identification of biomarkers for schizophrenia (Mirnics et al. Reference Mirnics, Middleton, Marquez, Lewis and Levitt2000; Perez-Santiago et al. Reference Perez-Santiago, Diez-Alarcia, Callado, Zhang, Chana, White, Glatt, Tsuang, Everall, Meana and Woelk2012).
Bile acids (BAs) play complex roles in cell signaling and immunomodulation (Chiang Reference Chiang2013; Sipka & Bruckner Reference Sipka and Bruckner2014) and affect the function of neurotransmitter receptors, such as the γ-aminobutyric acid receptor (Schubring et al. Reference Schubring, Fleischer, Lin, Haas and Sergeeva2012; Huang et al. Reference Huang, Wang, Lan, Yang, Pan, Zhu, Lv, Wei, Shi, Wu, Zhang, Wang, Duan, Hu and Wu2015). BAs are also known to be protective against neurodegeneration and to be a putative biomarker in Alzheimer’s disease (Ackerman & Gerhard Reference Ackerman and Gerhard2016, Marksteiner et al. Reference Marksteiner, Blasko, Kemmler, Koal and Humpel2018). Moreover, it was reported that BAs could produce antidepressant-like effects in depressed mice (Lu et al. Reference Lu, Yang, Zhang, Wang, Gong, Hu, Wu, Gao and Huang2018; Cheng et al. Reference Cheng, Huang and Chen2019). Despite these findings, it remains unclear whether specific BA alters in patients with schizophrenia. Therefore, we compared BA levels between schizophrenia patients and healthy controls and analysed the data to determine if plasma BAs could be used as an auxiliary diagnostic biomarker to distinguish patients with schizophrenia from healthy controls.
Materials and methods
Clinical subjects and peripheral blood collection
Clinical healthy subjects and schizophrenia patients were recruited from Shanghai Mental Health Center, China. Patients with schizophrenia recruited in the study were first-onset patients who had been independently diagnosed by at least two experienced psychiatrists according to the Diagnosis and Statistical Manual of Mental Disorders Fourth Edition (DSM-IV) criteria for schizophrenia with the PANSS for score. Controls were drawn from unrelated volunteers. Subjects who are suffering from cirrhosis of the liver, acute viral hepatitis or chronic hepatitis, liver damage from alcohol abuse or alcoholic fatty liver, elevations of liver enzymes glutamic-pyruvic transaminase and glutamic-oxalacetic transaminase, other metabolic diseases and history of drug or alcohol addiction were excluded in this study. Subjects prepared for the test set were of Han Chinese ethnicity. The study was approved by the Ethics Committee of the Shanghai Mental Health Center. All subjects gave written informed consent.
All blood samples were collected after an overnight fast of at least 12 h between 06:00 and 07:00 a.m. Blood samples were drawn from all patients at admission for laboratory tests of metabolic parameters. Blood (5 ml) was sampled using anticoagulant tubes and kept for 1 h at 4 °C (for platelet activation) before plasma was isolated (centrifugation at 3000 rpm for 20 min at 4 °C). The separated plasma was stored at −80 °C until being analysed.
Bile acids detection
The BA levels were determined by using a liquid chromatography–mass spectrometry (LC-MS) method described previously (Wang et al. Reference Wang, Xiong, Cheng, Yang, Wang and Liu2018). The 100 μl plasma from each sample was mixed with 500 μl ice cold methanol. The mixture was vortexed for 1 min and then centrifuged at 15 000 rpm for 10 min at 4 °C. The supernatant was transferred and evaporated to dryness under a nitrogen stream. The resultant dry residue was reconstituted in 100 μl methanol, and a 10 μl aliquot was injected into the LC-MS system for analysis.
Statistical analysis
BA concentrations and clinical characteristics were compared between groups (control, schizophrenia) by Student’s t-test using GraphPad Prism 7 software. In addition, correlation analysis was used to assess the association of plasma BA levels with schizophrenia. After that, liner discriminant analysis was used by MASS packages in R (3.6.1) (AT&T Bell Laboratories, USA). Selection of predictor variables was based on the correlation analysis and all variables with a p value < 0.05 were included in the model. The data are expressed as mean ± standard deviation (SD). P values less than 0.05 were regarded as statistically significant.
Results
Study subjects
Overall, 12 healthy controls and 12 patients with schizophrenia were recruited in the final qualitative analysis. All individuals in the study population were in the same age group between 19 and 54 years and did not differ between groups. All groups included more females (three males and nine females for each group), and patients had no habit of smoking. The average positive and negative syndrome scale (PANSS) score was (106.42 ± 44.24) for schizophrenia subjects. More demographic and clinical characteristics of participants can be found in Supplemental Table.
Quantification of BA levels in plasma
BA levels in plasma were detected by LC-MS. As shown in Table 1, the concentrations of cholic acid (CA), taurochenodeoxycholic acid (TCDCA) and taurodeoxycholic acid (TDCA) in schizophrenia patients were significantly less than that in healthy subjects (p = 0.006, 0.042 and 0.027, respectively). By contrast, other BA levels were either not different between two groups or out of detection limit. To determine the relevance between BAs and schizophrenia, the correlation analysis was conducted. In addition, the concentrations of CA, TCDCA and TDCA were negatively correlated with schizophrenia (healthy subjects = 0; schizophrenia = 1) with correlation coefficients of −0.590, −0.584 and −0.523, respectively (p = 0.003, 0.006 and 0.044). And other types of BAs did not show robust correlation with schizophrenia.
The concentrations of various BAs in plasma and the correlation analysis by Spearman in rcorr function of Hmisc packages (healthy subjects = 0; schizophrenia = 1)
LLOQ, lower limit of quantification.
Data are expressed as mean ± SD, N = 12 for each group. *, p < 0.05; **, p < 0.01 vs. Control group (Healthy subjects).
Diagnostic value for schizophrenia
According to the result of correlation analysis (Table 2), CA, TCDCA and TDCA were added into the linear discriminant model. After calculated by linear discriminant analysis (LDA) function in MASS packages in the R programming language, we obtained a classification formula, which is y = −3.13 × 10−2 × LCA − 3.00 × 10−3 × TCDCA − 4.33 × 10−3 × TDCA. And we instituted a rule that samples were classified as healthy subjects if y < 0, while samples were classified as schizophrenia patients if y > 0. Then, we used the train set to predict the accuracy of model. The correctly classified cases were 23, and the accuracy of prediction was up to 95.83%.
LDA: the concentrations of CA, TCDCA, TDCA were included in the model. The correctly classified cases and accuracy of prediction were validated by the train set
Classification formula: y = −3.13 × 10−2 × LCA − 3.00 × 10−3 × TCDCA − 4.33 × 10−3 × TDCA.
Rule: Classify as healthy subjects if y < 0, classify as schizophrenia patients if y > 0.
Discussion
In this study, we highlighted the association between BA levels and schizophrenia. The results of LC/MS detection showed that CA, TCDCA and TDCA were decreased in patients with schizophrenia compared with healthy controls. Additionally, correlation analysis identified that CA, TCDCA and TDCA were negatively correlated with schizophrenia. Therefore, we focused on these three BAs through LDA. The results indicated that a combination of the three BAs might be used as an auxiliary diagnostic biomarker for schizophrenia.
Both conjugated and unconjugated BAs can be detected in the brain of humans and rodents (Mano et al. Reference Mano, Goto, Uchida, Nishimura, Ando, Kobayashi and Goto2004). Twenty BAs have been identified in the rat brain, consisting of nine unconjugated BAs and eleven conjugated BAs (Zheng et al. Reference Zheng, Chen, Zhao, Wang, Xie, Huang, Liu, Zhao, Wang, Wang, Zhou, Panee, He and Jia2016). Most of these 20 BAs are also found in the blood of rats (Xie et al. Reference Xie, Zhong, Li, Li, Qiu, Zheng, Chen, Zhao, Zhang, Zhou, Zeisel and Jia2013). CA is a kind of primary bile acid that mainly contains in the total bile acid in human. It is synthesized from cholesterol in the liver and can be conjugated to either glycine (glycocholic acid) or taurine (taurocholic acid) in the bile. TCDCA is synthesized with taurocholic acid and chenodeoxychlolic acid as TDCA is synthesized with taurocholic acid and deoxycholic acid in the liver. Although the origin of brain BAs remains unclear, they can be synthesized in the brain or transported into the brain from the peripheral circulation by BA transporters and/or diffuse across the blood-brain barrier (BBB). Some studies had suggested that the dysfunction of BBB representing a novel neural dysfunction was closely related with schizophrenia (Kealy et al. 2018; Pollak et al. Reference Pollak, Drndarski, Stone, David, McGuire and Abbott2018). The dysfunctional BBB loses its function in the control of the inflow or outflow of BAs from the blood or brain, respectively, resulting in the disturbance of the BAs homeostasis. Patients with schizophrenia were at greater risk of dyslipidemia, and the concentration of cholesterol was extremely bounded up with the synthesis of BAs and various neurotransmitters (Paul et al. Reference Paul, Choudhury, Kumar, Giri, Sandhir and Borah2017; Yang et al. Reference Yang, Shen, Wen, Wang and Hu2018). So far, the neuropsychiatric function of BAs in the central nervous system has not been clarified yet. Therefore, despite in the current study, we observed the decrease of BAs in schizophrenia patients, we do not know if there is a causality between the alteration of BAs and the occurrence of the disorder as well as the possible underlying mechanism. Further studies to link BAs with schizophrenia as well as other neuropathic diseases are necessary.
Increasing evidence suggests interactions between the gut microbial community, BA biotransformation (including deconjugation, dehydroxylation and reconjugation) and disease states (Ridlon et al. Reference Ridlon, Kang and Hylemon2006; Long et al. Reference Long, Gahan and Joyce2017). Recently, Nguyen et al. found that differentially abundant genera were revealed between schizophrenia and healthy subjects (Nguyen et al. Reference Nguyen, Kosciolek, Maldonado, Daly, Martin, McDonald, Knight and Jeste2019). Since gut bacteria metabolize majority of taurine into inorganic sulfate, which may result in decreasing taurine available for bile acid conjugation in the liver and alter the systematic BA profile (Manley & Ding Reference Manley and Ding2015). However, the specifically altered BAs in schizophrenia needs to be further investigated along with the emerging links of gut microbiome.
This study could be considered as a preliminary investigation and had several limitations: a main point of the present study is the limited sample size and subtypes of schizophrenia were not considered. This study should be followed up by more large-scale samples that need to include correctly classified subtypes in schizophrenia. In addition, it has been recognized that different diet habits and nutriture might have an effect on the concentration of various metabolites in blood. It is needed to follow more details about various indicators of healthy subjects and schizophrenia patients.
In conclusion, our data showed that CA, TCDCA and TDCA significantly decreased in schizophrenia patients. And a combination of these three BAs may be used as a potential biomarker for diagnosis specification and therapeutics monitoring.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/neu.2020.42
Author contributions
Conceptualization, X.W. and D.C.; methodology, F.Z. and Y.T.; software, Y.T.; formal analysis, Y.T.; investigation, F.Z.; resources, D.C.; data curation, Y.T.; writing – original draft preparation, Y.T.; writing – review and editing, F.H. and X.W.; supervision, X.W. and F.H.; funding acquisition, F.H.
Financial support
This work was supported by National Natural Science Foundation of China [F.H., grant number 81703734] and Program from Shanghai Municipal Commission of Health and Family Planning [F.H., grant number 20184Y0177].
Conflict of interest
The authors declare no conflict of interest.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.
