Serum biomarkers research of tourette’s syndrome children using proteomics techniques--a pilot study

Purpose:This study used proteomics to analyze the changes in serum proteomics between tourette’s syndrome (TS) children and healthy children in order to nd serum biomarkers that can distinguish TS children from healthy children. Experimental design: We analyzed the serum proteome of 60 TS children and 30 healthy controls children using magnetic bead-based weak cation exchange (MB-WCX) and matrix-assisted laser desorption ionization-time of ight mass spectrometry (MALDI-TOF-MS). Next, we identied candidate biomarkers using liquid chromatography-electrospray ionization-tandem mass spectrometry (LC-ESI-MS/MS). Candidate biomarkers were then validated using ELISA and western blotting. Results: 59 peaks were identied and the expression fold changes of seven peaks in the two groups were greater than 1.9. Two peaks (m/z: 6443.34 Da; m/z: 6642.05 Da;) tended to be upregulated, whileve peaks (m/z: 863.13 Da; m/z: 2175.98 Da; m/z: 2191.841 Da; m/z: 2277.19 Da;m/z: 2293.11 Da) tended to be down-regulated in TS group. The peak for a 2191.84 Da peptide was identied as FGA (Isoform 1 of the Fibrinogen alpha chain precursor, FGA);The peak for a 2175.98 Da peptide was identied as PKM2 (Isoform M2 of Pyruvate kinase isozymes,PKM2);The peak for a 2277.19 Da peptide was identied as GAPDH (Glyceraldehyde-3-phosphate dehydrogenase,GAPDH);The peak for a 863.13 Da peptide was identied as PROC (Vitamin K-dependent protein C,PROC).Enzyme-linked immunosorbent assay (ELISA) analyses revealed that the expression of FGA and PKM2 were signicantly higher in TS children than healthy controls children. Conclusion: FGA and PKM2 may be potential serum biomarkers to


Introduction
Tourette's syndrome (TS) is severe hierarchy of Tic disorders (TD) which is a kind of neuropsychiatric disease. TS start from childhood and take both multiple motor and one or more vocal tics have been present at some time during the illness, although not necessarily concurrently. It is a group of motor disorders with unknown causes, mainly manifested as involuntary, repeated, rapid and aimless muscle motor twitch and/or vocal twitch in one or more parts. The prevalence rate is increasing year by year, and the estimated prevalence of TD ranges from 3 to 8 per 1,000 in school-age children [ 1] . Behavioral comorbidities often associated with TS account for up to 50-90% of TS patients and may include attention de cit hyperactivity disorder(ADHD), obsessive-compulsive behavior(OCB),anxiety, and depression [ 2] and it is becoming one of the common serious chronic diseases that endanger children's health [ 3] . Because of the comorbidities, TS clinical picture interferes with normal social functioning. In recent years, a number of studies have found that TS has a highly hereditary and complex polygenic background [ 4] .The abnormal function of the cortex-striatum-thalamus-cortex circuit is the basis of TS [ 5] .
Imbalance of neurotransmitters such as dopamine, noradrenaline, and 5-hydroxytrytamine in the circuit are the main causes of TS in recent research works [ 6] , infectious and immune factors [ 7] and sociopsychological factors [ 8] are closely related to the occurrence of TS. However, so far, the exact etiology and pathogenesis of the disease remain unclear. So none was effective treatment, involving in neuroleptics, adrenergic agonists, dopamine agonists, behavioral techniques, and botulinum toxin untill now [ 9] . The diagnosis of the disease currently depends on the patient's clinical manifestations for descriptive diagnosis. Due to the lack of objective diagnostic indicators in clinical work, some TS patients cannot be diagnosed early, or even missed or misdiagnosed. Therefore, the clinically is an urgent need to nd the exactly pathogenesis and further diagnosis and treatment method.
As an important method in the post-gene era, proteomics technology has become a very important research platform in the eld of molecular medicine because of its fast, high-throughput, high sensitivity, and easy operation [ 10] . Proteomic technologies enable the large-scale and in-depth investigation of proteins, in particular, the screening of potential biomarkers in complex biological matrices in a highthroughput manner. To date, proteomics has become a fundamental component for biological study and is recognized as a key method for novel biomarker discovery and personalized therapy [ 11] , Especially after the launch of the human plasma proteome project that was proposed at Human Proteome Organisation 2010 in Sydney [ 12] . As indicators of pathogenic processes and pharmacologic responses to a therapeutic intervention, biomarkers enable us to accurately predict pathophysiology as well as understand how it is altered by treatments. The serum of blood in human physiology implies that it should be a universal re ection of an individual's state or phenotype. Pathophysiological changes in the body at any time can be re ected in the serum. Clinically, the detection of physiological indicators or parameters in the serum/plasma has always been one of the most commonly used auxiliary diagnostic and monitor therapy methods. It is more practical to take serum proteins involved in many physiological and pathological processes as research objects. Serum proteomics is currently the most commonly used method for screening disease markers. Serum proteomics studies all the proteins expressed in the serum of selected populations, and on the basis of establishing a normal protein expression pro le, looking for its differential protein spots and identi es disease-related proteins. To further study its structure and function, and open up new avenues for studying the pathophysiology of major diseases, speci c markers for early diagnosis, and drug targets [ 13] In recent years, the proteomics method has made great progress in the eld of nervous system disease research, and the study of biological targets of nervous system diseases has become a hot spot today.
Yang et al [ 14] ,Used proteomics to analyze the serum proteins of children with ASD and found that there are 8 differentially expressed proteins in the serum of children with ASD, which not only revealed the etiology of ASD from the perspective of serum proteomics, but also veri ed these 8 This differential protein may be a potential biomarker for the diagnosis of ASD.Jiang et al [ 15] ,Used SELDI-TOF-MS technology to study the serum of healthy controls and patients with schizophrenia to nd 6 marker proteins with signi cant differences in expression. This technology established an arti cial neural network model for schizophrenia. Shen et al [ 16] , Used MALDI-TOF-MS technique to compare the differential expression of serum proteins between NTD pregnant women and normal pregnant women, and found that the sensitivity and speci city of the three proteins in diagnosing NTD could reach 96% and 90% by protein ngerprint software. Bao et al [ 17] ,Conducted iTRAQ marking and analysis on plasma samples collected from patients with severe brain injury with chronic consciousness impairment in different time periods and healthy people, and identi ed 32 proteins with signi cant changes after brain injury, and these differential proteins are mainly involved in complement system pathways. The last, an iTRAQ-MS study of AD and thecontrol plasma proteome showed that Complement 4a, an in ammatoryrelated protein, is upregulated in the plasma of AD patients. This was further validated by immunoblot and ELISA [ 18] .Additionally, YKL-40 was shown to be potential biomarker of AD through LC-MS/MS-based proteomic pro ling method, and it was validated in another cohort [ 19] .
In this study, serum proteomic pro ling in children with TS and healthy controls were conducted by MB-WCX puri cation combined with MALDI-TOF-MS. ClinProTools software was used to compare the generated serum proteomic pro les. The signi cant differentially expressed peaks between TS group and healthy control group could be considered as potential serum biomarkers and were sequenced and further identi ed. The potential serum biomarkers were then analyzed by bioinformatics and validated using ELISA. The validated serum biomarkers may supply a more useful and reliable method assisting the diagnosis of TS.

Participants
A total of 90 Chinese Han children were used in this study, including 60 in the TS group and 30 in the healthy control group. Children in the TS group were recruited from Xi'an Children's Hospital in Shaanxi Province, China, and all children with TS were examined by a pediatric psychiatrist. Inclusion criteria: (1)The TS group all met the diagnostic criteria for Diagnostic and statistical manual of mental disorders(DSM5) 1 tics in the United States;(2) Children under 18 years old, not limited to male and female;(3)No other neuropsychiatric diseases;(4) Have not taken any drugs that affect neurotransmitters. There was no signi cant difference in age and gender among all recruiters.
The study was approved by the Ethics Committee of Xi'an Children's Hospital. After understanding the potential risks of the study, the guardians of all participants have signed written informed consent. All experiments were performed in accordance with the approved guidelines.
A fasting blood draw was performed on all TS and control subjects between 8 and 10 a.m. Blood samples were collected in vacuum blood collection tubes lacking heparin and centrifuged at 3000 g for 10 min at 4°C to separate the serum. The supernatant was divided into 100 μL aliquots and stored at -80°C until processing.

MALDI-TOF MS Analysis
MB-WCX Pro ling Kit (Bruker Daltonics), based on magnetic bead technology utilizing the interaction mechanisms of cation exchange, is designed for sophisticated biomarker pro ling studies and was used for proteomics or peptidome separation of serum samples according to the standard protocol by Wang et al [ 20] .A total of 90 serum samples were fractionated. Magnetic beads were then thoroughly mixed on a vortex device for at least 1 min, before 10μL of magnetic beads with 10μL binding buffer was added to a standard thin wall PCR-tube and mixed by pipetting up and down. Next, 5μL serum was added to the solution and mixed by pipetting up and down ve times. After 5 min of incubation, the tube was placed into the magnetic separator and the beads were automatedly collected at the wall of the tube for 1 min.
The supernatant was collected carefully until it was clear from magnetic beads, and then washed three times with 100μL washing buffer. The peptide fraction was eluted from the magnetic beads with 5μL of elution solution and 5μL of stabilization buffer. The eluted peptides were spotted onto cleaned MALDI-TOF-MS targets with 1μL of matrix consisting of 4 mg mL -1 a-cyano-4-hydroxy-cinnamic acid in 50% acetonitrile. Then, 0.5% tri uoroacetic acid was added twice to the MALDI AnchorChip surface. To ensure reliability of the method,all samples were spotted in triplicate. Instrument calibration parameters were determined using standard peptide and protein mixtures (Bruker).

ClinProTools Analysis
Immediately after, all targets were analyzed by calibrated Auto ex III MALDI-TOF MS (Bruker) using an optimized protocol of FlexControl version 3.0 software (Bruker). The ClinProTools software (Bruker Daltonik) has been used for all data interpretation steps, which include baseline correction, smoothing, peak detection and alignment, as well as further analysis.The whole data pretreatment has been completed using default settings and was performed automatically. The pretreated data have been used for visualization and statistical analysis in ClinProTools. Peak statistics has been performed by means of a Welch's t-test.Use FlexAnalysis3.0 peak software and ClinproTools statistical software to analyze protein peaks.Genetic algorithms (GA), based on evolutionary survival, were used for model analysis and were applied to select the combinations of peaks that perform best in separating the classes. It is suited for a population with a large variety of peak combinations. The ttest peak combinations were selected and the less capable abandoned. This procedure aims for the most favorable class separation with high variance between classes to optimize a cost function.The difference peak standard P <0.05 was selected, and the multiple of the difference was> 1.9. The area under the working characteristic curve (AUC) of the subjects was> 0.7.

Peptide identi cation
After completing the statistical analysis, selected peptides or proteins biomarkers were puri ed and separated using Nano Aquity UPLC C18 beads and serially eluted with 5% and 95% acetonitrile.Use Q-Exactive HF mass spectrometer (Thermo Sceienti c) for mass spectrometry analysis, the raw data of mass spectrometry analysis is RAW le, use software software Mascot 2.2 and Proteome Discover2.0 to search the database uniprot_human_fasta, identify the selected peptide or protein biomarker Thing.

ELISA
In the ELISA analysis, all serum samples were tested in a blinded fashion, standards and samples were run in triplicate.The concentrations of Isoform 1 of Fibrinogen alpha chain(FGA),Isoform M2 of Pyruvate kinase isozymes(PKM2), Tubulin beta chain(TUBB), 2 Glyceraldehyde-3-phosphate dehydrogenase(GAPDH), and Vitamin K-dependent protein C(PROC), quanti ed using a Human FGA ELISA Kit ,Human PKM2 ELISA Kit,Human TUBB ELISA Kit,Human GAPDH ELISA Kit,and HumanPROC ELISA Kit.A standard curve was generated and used to determine the concentrations of FGA, PKM2,TUBB,GAPDHand PROC in the samples analyzed.

Statistical Analysis
Statistical analyses were performed using GraphPad Prism version 5.0 (GraphPad Software, La Jolla, CA, USA). All data are shown as the mean ± standard deviation. A p-value below 0.05 was considered statistically signi cant. Students t-test.were used to compare differences between groups.

Clinical Characteristics Analysis
Demographic features between TS group and healthy control group were similar.In the TS group, there were 52 male and 8 female subjects with an average age of 89.52±26.76 months.In the healthy control group, there were 26 male and 4 female subjects with an average age of 88.32±25.32 months.There were no statistically signi cant difffferences between the groups with respect to the distribution of age and sex (p > 0.05).

Serum Proteomic Pro les of Difffferent Groups
To ensure reproducibility and stability of mass spectra, we analyzed all samples in triplicate and found closely reproducible peaks (Figure1A,B). MB-WCX and MALDI-TOF-MS were used to compare the proteomic pro les of the TS group and the control group. Fractionation of serum samples showed that TS (red) and control subjects (green) had proteomic pro les from 0.8-10kDa (Figure 1).The differentially expressed peaks between the two groups were detected in this mass range. According to the serum proteomics analysis, only a few overlapping areas between TS group and healthy control group were found, indicating that the groups were accurately distinguished (Figure 2A,B).

Peaks Selection
TS group and healthy control group have 59 distinguishable differentially expressed protein peaks. Taking the multiples of peak intensity difference between groups (average ratio of peak intensity between groups)>1.9 as the signi cant difference, 7 protein peaks have signi cant differences (P <0.001) and their AUC values are greater than 0.70;In these 7 protein peaks, the relative molecular masses were 6443.34 Da, and 6642.05 Da expression was up-regulated, the masses were 863. 13 Figure 3.   Table 3. All data are represented as mean ± SD. The statistical analysis was performed with Prism 5.0 software. The serum levels of FGA and PKM2 protein in children with TS group were signi cantly lower than those in healthy controls (P <0.05, Figures 8), which were basically consistent with the results of mass spectrometry analysis.

Discussion
TS is a neuropsychiatric disorder that starts in childhood and is mainly manifested by tics [21] .Strong evidence supports involvement of cortical-basal ganglia-thalamocortical circuits in the pathophysiology of tic disorders. Several neurotransmitters,including dopamine, glutamate,GABA, serotonin, acetylcholine, norepinephrine,endogenous cannabinoids, opioids, histamine, and adenosine, are active participants within these circuits and may be dynamic factors in the pathophysiology of tics. However, the etiology and pathogenesis of the disease have not been clari ed so far, the clinical diagnosis lacks speci c indicators, so that the disease cannot be diagnosed in a timely and accurate manner, which affects the prognosis; This study used proteomics techniques to study serum proteins in children with tic disorder and found that two proteins named FGA and PKM2 may be serum biomarkers that distinguish TS group from healthy children.

Isoform 1 of Fibrinogen alpha chain (FGA) is a subtype of brinogen. Fibrinogen is a glycoprotein found
in the plasma of all vertebrates [ 22] , which has the functions of regulating immune in ammatory response, enhancing blood clotting function and inhibiting tissue wound repair.As an acute in ammatory response protein, brinogen has a unique molecular structure that binds to integrins secreted by various immune cells and participates in the in ammatory response of the nervous system [ 23] .Fibrinogen binds to CD11b/CD18 integrins, which are synthesized and secreted by various immune cells, widely stimulates in ammatory response signal of cells, promotes local production of in ammatory cytokines such as TNF-and IL-1β, and participates in regulating immune function, inducing adhesion, migration, chemotaxis and phagocytosis of immune cells, and strengthening host active immune defense function [ 24] .In recent years, many studies have shown that tic disorder is related to immune response.

Some researchers[ 25][26]
[27] [28] at home and abroad have studied the cytokines in the serum of children with TS and found that some cytokines in the serum of children with TS are abnormal, and the concentration level is related to the severity of the disease. The level is positively correlated, suggesting that the abnormality of cytokines is closely related to the occurrence of tics.The mechanism may be that the neuroendocrine -immune network is involved in the pathogenesis of TS.Children with TS are in the state of cellular immune activation, and cytokines may promote the occurrence of TS. Kawikova[ 29] , Bos-Veneman [ 30] , Landau[ 31] and other teams studied the expression of some immunoglobulins in children with TS and found that the occurrence of TS is closely related to the function of humoral immunity. The pathogenesis may be that the antigen and antibody combine to form an antigen-antibody complex, which activates cytokines, thereby triggering neurological damage, leading to the occurrence of TS.Our research found that FGA1 was signi cantly underexpressed in the serum of TS children. That is to say it is possible that FGA take role in immune response related to immunologic mechanism of TS. It is worth to do further research deeply about the role in TS immunologic mechanism. The result also indicates FGAcould be a potential serum marker for children with TS.
Isoform M2 of Pyruvate kinase isozymes (PKM2) is a subtype of Pyuvate kinase (PK) and the ratelimiting enzyme in the last step of the glycolysis process. It can decarboxylate phosphoenolpyruvate to pyruvate and produce ATP [ 32] .PKM2 is mainly expressed in embryonic cells, tumor cells and other rapidly dividing and proliferating cells [ 33] [].The Warburg effect in which PKM2 is involved is a peculiar glycolysis phenomenon. As a major metabolic pathway in the early development of brain nerves after birth, the Warburg effect is involved in axonal elongation, synapse generation and myelin process [ 34] .The Warburg effect in which PKM2 participates is closely related to the occurrence of neurological diseases,The Afaf [ 35] study found that the content of pyruvate kinase in the plasma of autistic patients was signi cantly reduced compared with the control group, which re ected the decreased adaptability of the patients to impaired energy metabolism. Newington et al found that the Warburg effect can alleviate the process of Alzheimer's diseases(AD) by inhibiting the cell death mediated by amyloid β-mediated(Aβ)cell death in the early stage of AD.Some researchers have found that the activation of PI3K/Akt pathway in HD patients increases the lactate/pyruvate ratio in YAC128 mice and HD patients, and up-regulates the Warburg effect. This energy metabolism disorder accelerated apoptosis and accelerated the course of HD disease[ 36] [37] .Because the developmental brain tissue is dependent on energy supply, especially glucose metabolism, a slight lack of energy production may have a signi cant impact on neurons and glial [ 38] .TS is a multi-causal disease of brain development disorders, impaired energy metabolism may be one of the causes of this multi-factor disease TS symptoms may worsen when tired, whether or not impaired PKM2 metabolism is related to this phenomenon requires further study. The result also indicates PKM2 could be a potential serum marker for children with TS.

Limitations
The limitation this research relates to the small sample size. Future research should expand the sample size so that there are enough samples to test the role of the FGA and PKM2 in TS children, and verify the quantity in TS serum .
This study uses proteomics technology to study the differential proteins between TS children and healthy children. Although two speci c serum biomarkers (FGA and PKM2) were identi ed that might distinguish TS group from healthy control group. However, the relationship between these markers and pathogenesis needs further deeper research. We need to further clarify the relationship between these serum markers and the pathogenesis of TS. Ethical approval

Abbreviations
This study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving human subjects were approved by the Ethics Committee of Children's Hospital a liated to Xi 'an Jiaotong University. Written informed consent was obtained from all subjects.

Con ict of interest
The authors have no con icts of interest to disclose.

Contributions
HWX participated in the data collection, statistical analysis and data interpretation, manuscript drafting and critical revision of the manuscript.. WD,ZH, and MYF participated in the data collection.WD and ZHQ contributed to statistical analysis and data interpretation. WD participated in the design and execution of the study. CYN supervised the design and execution of the study, performed the nal data analyses and Figure 1 Comparative analysis of serum proteomic pro ling between difffferent groups. A) Representative masss pectra of a healthy control group (green); B) representative mass spectra of an TS group individual (red).  The results of ELISA. ***indicates P<0.0001