Diagnostic And Therapeutic Biomarkers For Bipolar Disorder Based On Targeted Metabolomics Of Plasma Neurotransmitters

Background:Bipolar disorder(BD)is a kind of mental disease with complex clinical manifestations,high recurrence rate and high disability rate.Neurotransmitter dysfunction may be the main pathogenesis of BD.Currently,there are no effective biomarkers for early diagnosis and treatment monitoring of BD.As an analytical technique that can evaluate metabolic characteristics such as disease process and drug response,metabolomics provides a methodological basis for finding biomarkers of BD.The overall purpose of this study is to explore the potential and application of peripheral blood neurotransmitters and related metabolites as BD biomarkers.The study is divided into four parts:1.Analyze the differences in peripheral blood neurotransmitter metaboliteomics between healthy individuals and untreated BD patients,and investigate their correlations with clinical symptoms.2.Establish a BD diagnostic model based on peripheral blood neurotransmitter metaboliteomics,and evaluate its diagnostic efficacy.3.Explore the impact and clinical significance of psychotropic drug therapy on the peripheral blood neurotransmitter metaboliteomics of BD.4.Investigate the impact of probiotic assisted therapy on the peripheral blood neurotransmitter metabolome of BD and its clinical efficacy.Methods:This study is mainly divided into two major parts and four main sections.The cross-sectional study will explore the potential and classification effect of peripheral blood neurotransmitter metabolites in distinguishing BD patients and healthy individuals;The longitudinal study will explore the application of peripheral blood neurotransmitter metabolites as biomarkers related to BD treatment.1.Recruit drug-naive BD patients and healthy individuals,and conduct clinical symptom assessment(using HAMD to assess depressive symptoms,HAMA to assess anxiety symptoms,YMRS to assess manic symptoms)and cognitive testing(using RBANS and STOP color word tests)on BD subjects.14 plasma neurotransmitters and related metabolites were quantitatively detected by liquid chromatography electrospray ionization mass spectrometry tandem mass spectrometry(UPLS-MS/MS).Explored the differential metabolites and differential metabolic ratios between BD and HC respectively;Differential metabolites and differential metabolic ratios among subjects with different demography characteristics;The differential metabolites and metabolic ratios among BD subjects with different clinical characteristics were analyzed,and the correlation between the plasma neurotransmitter levels and metabolic ratios of BD subjects and clinical symptoms was analyzed.2.The peripheral blood neurotransmitter metabolites of participants in BD and HC groups were analyzed using principal component analysis(PCA)and fitted using orthogonal partial least squares discriminant analysis(OPLS-DA)model based on orthogonal signal correction.Use cross validation,permutation testing,and external test set validation to test the fitting effect and discriminant ability of the model.The receiver operating characteristic curve(ROC)and area under the ROC curve(AUC)are used to determine classification performance.Finally,identify key metabolites to optimize the diagnostic model and conduct testing.3.Subjects entering the longitudinal study will collect blood samples for the measurement of plasma neurotransmitters and related metabolites during pre treatment(baseline period),1 month after treatment,3 months after treatment,and 6 months after treatment follow-up.Linear mixed effect model(LMM)was used to compare the repeated measurement of symptomatology indicators and the content of neurotransmitter related metabolites in peripheral blood before and after treatment.Establish PCA and OPLS-DA models for metabolic profile analysis and identify differential biomarkers before and after treatment,as well as for metabolites related to neurotransmitters in healthy individuals’peripheral blood.4.Based on a randomized clinical trial of probiotic adjuvant therapy for BD,peripheral blood neurotransmitter metabolomics were targeted measured in BD subjects receiving placebo or probiotic adjuvant therapy,and the intervention lasted for 6 months to evaluate the clinical symptoms and differences in plasma neurotransmitter metabolites between the two groups at different intervention time points.Results:1.In the first part,138 BD subjects and 59 HC subjects who did not take the first treatment were included.A total of 14 neurotransmitters and related metabolites were detected,and half of the substances showed significant differences in content between BD and HC.The plasma levels of GLN,VMA,MHPG,DA,HVA,TRP and 5-HIAA in the BD group were significantly lower than those in the HC group(p<0.01).The metabolic rates of the NE system(VMA+MHPG)/NE and MHPG/NE in the BD group were significantly lower than those in the HC group,while the metabolic rates of the 5-HT system,5-HIAA/5-HT,were significantly lower than those in the HC group(z=5.77,p<0.001);The conversion rate of TRP canine uric acid pathway KYN/TRP was significantly higher than that of HC group;The metabolic rate(GLU+GABA)/GLN of the GLN system was significantly higher than that of the HC group(p<0.05).No significant differences in plasma neurotransmitter and related metabolites levels were found between the BD participants who were currently experiencing depressive episodes,manic episodes,and mixed episodes(p>0.05),and no significant differences in neurotransmitter metabolism rate or metabolic ratio were found between the groups(p>0.05).The plasma HVA and NE levels of BD participants with anxiety characteristics were significantly lower than those without anxiety characteristics(p<0.05),while KA levels were significantly higher than those without anxiety characteristics(p=0.024).The total scores of HAMD and HAMA were not significantly correlated with the concentration of metabolites in various neurotransmitter systems,the metabolic rate of neurotransmitters within each system,and the metabolic rate between monoamine neurotransmitter systems(p>0.05),while the total scores of YMRS were significantly negatively correlated with the metabolic rate of glutamate system(GABA/GLU)(r=-0.183,p=0.036).2.In Part 2,a total of 138 BD and 59 HC(Cohort 1)were included for model construction and cross-validation,and 14 BD and 6 HC(Cohort 2)were included for external test of the model.PCA and OPLS-DA models of BD and HC were successfully constructed.In the OPLS-DA model,the two groups were completely separated,R~2Y(cum)=0.582,Q~2(cum)=0.527 by the 70%cross validation method,and the regression line slope of the replacement test chart was positive,which strongly indicated the validity of the original model.The AUC of the original model(Cohort 1)was 0.976,with a sensitivity of 94.93%and a specificity of 84.75%.The AUC of external validation(Cohort 2)was0.976,with a sensitivity of 100%and a specificity of 83.33%.Five key biomarkers including 5-HIAA,VMA,TRP,GLN and DA were screened by VIP combined with P-value method.The model established by the above key markers could still efficiently classify BD and HC(AUC=95.96%,sensitivity=94.2%,specificity=79.66%).3.72 BD subjects received longitudinal follow-up with mood stabilizers as the main treatment,and their HAMD,HAMA,and YRMS scores showed significant differences between the four time points before treatment,1 month,3 months,and 6 months of treatment,showing an overall downward trend;14 neurotransmitters or related metabolites were successfully detected in all subjects.Except for GABA levels,there were significant differences in the content of the remaining 13 metabolites at the time points before treatment and 1 month,3 months,and 6 months after treatment(P<0.05).After correlation analysis and correction of demography and clinical characteristics,the score reduction of HAMA scale was significantly negatively correlated with the increase of KA content at 1 and 3 months of treatment(β=-0.224,p=0.047;β=-0.35,p=0.006),the clinical symptoms at other clinical time points showed no significant correlation with changes in the content of any metabolites compared to baseline.Further multivariate statistical analysis showed that the five groups of samples showed a trend of separation on the OPLS-DA score map.The spatial positions of the four subgroup sample points before and after BD treatment showed continuity and changes in the same direction with the change of treatment time.They firstly approached the healthy group sample points,and then gradually moved away from the healthy group sample points.There are obvious distinctions between HC and HC at 6 months after BD treatment,between 3 months after treatment and before treatment,and between 6 months after treatment and before treatment.However,the metabolic differences between short-term BD patients and HC cannot be well distinguished,indicating that their metabolic profiles are relatively similar.4.138 BD subjects were included in a double-blind,randomized,placebo-controlled clinical study of Bifidobacterium adjuvant therapy for BD(70 in the probiotic group and 68 in the placebo group).Intragroup comparison showed that the HAMD,HDMA,and YMRS scores of the two groups significantly decreased compared to baseline after 1 month,3months,and 6 months of treatment.Inter group comparison found that the HAMD scores of the probiotic group were significantly lower than those of the placebo group after 6 months of treatment(d=-3.94,p=0.002).In terms of metabolite levels,there was a significant difference in the content of 13 other metabolites at different follow-up time points(p<0.001)among the placebo group subjects,except for GABA levels;In the probiotic group,no significant differences were found in the levels of GABA and KA at baseline,1 month,3 months,and 6 months after treatment,while the content of the remaining 12 metabolites showed significant differences.Conclusions:Summarize the content of each chapter and summarize the following main conclusions:(1)There are significant differences in the content,metabolic rate,and metabolic ratio of peripheral blood neurotransmitters and related metabolites between BD and HC,and there is a correlation with the cognitive performance of BD,indicating that peripheral blood neurotransmitters have the potential to serve as biological biomarkers of BD;(2)The optimized combination of peripheral blood neurotransmitters and related metabolites can serve as biomarkers for the identification of BD and HC,and the constructed OPLS-DA has good classification performance;(3)Psychotropic drugs can cause changes in the content of neurotransmitter metabolites and overall metabolic profile in peripheral blood;(4)Probiotic intervention can affect the metabolic changes of peripheral blood neurotransmitters during BD treatment.Overall,peripheral blood neurotransmitters and related metabolites have the potential and application prospects as biomarkers for the diagnosis and treatment of BD.Metabolomics analysis methods are helpful in the screening and validation of biomarkers.