Chronic Food Antigen-specific IgG-mediated Hypersensitivity Reaction as ARisk FactoRfoRAdolescent Depressive Disorder

Ran Tao*,Zhicheng Fu,Lijun Xiaoc

1 Department of PsychologicalMedicine,The Seventh Medical CenteRof Chinese PLAGeneral Hospital,Beijing 100700,China

2 Department of Trace Element Nutrition,National Institute foRNutrition and Health,Chinese CenteRfoRDisease Control and Prevention,Beijing 100050,China

KEYWORDS MajoRdepressive disorder;Blood-brain barrier;H istaMine;Hypersensitivity reaction;Inflammation

Abstract M ajor depressive disorder(MDD)is The most common nonfatal disease burden worldWide.SysteMic chronic low-grade inf lammation has been reported to be associated With MDD progression by affecting monoaMinergic and glutamatergic neurotransMission.However,whe The Rvarious proinflammatory cytokines are abnormally elevated before The first episode of depression is still largely unclear.Here,we evaluated 184 adolescent patientswho were experiencing The iRfirst episode of depressive disorder,and The same numbeRof healthy individuals was included as controls.We tested The seruMlevels of high-sensitivity C-reactive protein(hs-CRP),tumoRnecrosis factor-α(TNF-α),IgE,14 different types of food antigen-specific IgG,histam ine,homocysteine,S100 calcium-binding protein B,and diaMineoxidase.Wewerenot able to find any significant differences in The seruMlevels of hs-CRP oRTNF-αbetween The two groups.However,The histaMine level of The patients(12.35μM)was significantly higheRthan that of The controls(9.73μM,P＜0.001,Mann-Whitney U test).Moreover,significantly higheRseruMfood antigen-specific IgG positive rates were also found in The patient group.Fur The rmore,oveR80%of patients exhibited prolonged food intolerance With elevated levels of serum histaMine,leading to

Introduction

Majordepressivedisorder(MDD)isakind of mentaldisease that affected approximately 216Million peop le(3%of The world’s population)in 2015 and that is projected to have The secondhighest global burden of disease in all age groups and sexes by 2020[1].The present evidence suggests that many factors are involved in The pathophysiology of MDD,including heredity,neurotransMitters,immunity,oxidation,and The inflammatory system[2,3].However,The pathologicalmechanisMof MDD is not yet clear.In The field of neurocheMistry,The most Widely accepted hypo The sis of MDD is The lack of monoaMine neurotransMitters, especially 5-hydroxytryptaMine (5-HT)neurotransMitters[4].

Previous findingshaveshown thatat least some subtypesof depression are associated With chronic low-grade systeMic inflammation in adults[5,6].The levels of seruMinflammatory markersare increased in someadultMDD patients.FoRexample,The level of C-reactive protein(CRP),which reflects The state of systeMic inflammation,is elevated in The seruMof at least 30%of MDD patients[7,8].SeruMlevels of proinflammatory cytokines,such as tumoRnecrosis factor-α(TNF-α),are also higheRin patientsWith depression than in thoseWithout depression oRanxiety[9-11].Diseases related to chronic inflammation,such as cardiovasculaRdiseases,inflammatory bowel diseases,and rheumatoid diseases,are associated With Ahigh risk of MDD[12,13].

Many previous studies have indicated that proinflammatory cytokines can disrupt neurotransMittermetabolisMaffect neuronal transMission[14-16],and eventually triggeRcorresponding psychiatric symptoMs. The se studieshaveestablished Alink between depression and inflammation.However,The role of inflammation in The occurrence and development of depression is still unclear,mainly because The existing studies havebeen conducted on adult patients,andmany factors,such as chronic diseases,substance addiction,oRsocial environment,may interfere With The observed endpoints.It has been reported that adult depression of ten arises froMadolescent depression,so The study of adolescent depression can reflect The early mechanisms of MDD development[17].It is necessary to deterMine whe The R The re is Acausal relationship between systeMic inflammation and depressive disordeRby including adolescent depressive patients as The subjects.

In addition to inflammation, The blood-brain barrier(BBB)isalso closely associated With MDD and o The rmentaldiseases such as Alzheimer’s disease and Parkinson’s disease[18-20].The BBB is Ahighly selective seMi-permeable barrier,formed by close connections of special vasculaRendo The lial cells.The BBB is The main barrieRof The central nervous system(CNS)and prevents The influx of active substances froMThe peripheral circulation system[21].The high permeability of The BBB can cause The influx of some risk factors,such as toxins,pathogenic substances,and proinflammatory cytokines froMThe peripheral circulation system.The influx of The se factors has been confirmed to result in The inflammation of neurons in The brain,disturbing neurotransMittermetabolisMand function[19]. The refore,at The early stage of depressive disorders,identifying The factors that induce The high permeability of The BBB would be helpful in deterMining whe The RBBB leakage is related to The onset of depression.H istaMine has been shown to beoneof The main factors that induceshigh BBB permeability[18]. The refore,ouRpresent study focuses on histaMine and The related factors that affect histaMinemetabolism(IgG and IgE)and The iReffects on substances of cellmetabolism.

Results

In this study,we recruited adolescent depressive patient With The age-matched healthy subjectsas control.The inclusion criteriAfoRsubject enrollment were described in Materials and methods section.The adolescent depressive disordeRpatient(ADP)group consisted of 184 patientsWith moderate depression,including 114males and 70 females,With average age of 17.2±1.76 years old.The control normal adolescent student(NAS)group consisted of 184 students,including 105 males and 79 females,With average age of 17.4±1.58 years old. The re were no significant differences in sex and age between The patient and control groups(P＞0.05,Chi-square test).

H igheRlevels of seruMhistaMine were found in The ADP group

HistaMine is Akind of organic nitrogen-containing compound that participates in The immune response.H istaMine receptors,including H 1,H 2,and H 3,are allexpressed in The CNS. The se receptors are involved in The release of acetylcholine,norepinephrine,5-HT,and o The RneurotransMitters[22].Some studies have shown that high levels of histaMine are found in The cerebrospinal fluid and brain parenchymAof individuals With various neurodegenerative diseases[23-25].However, The se studies did not confirMThe general level of histaMine.We coMpared The seruMhistaMine levels between The ADP group and The NAS group.Among The 184 patients,The average concentrations of histaMine in The ADP group and The NAS group were 12.35(11.11,13.42)ng/Ml and 9.73(9.16,10.6)ng/Ml,respectively(P＜0.001,Mann-Whitney U test)(Figure 1A).To deterMine whe The RThe elevated seruMhistaMine levelwas associated With an abnormal pathway of histaMine degradation,we exaMined The level of histaMine metabolizing enzyme diaMine oxidase(DAO)in The se two groups,and no significant difference was found.The mean concentrations of DAO were 209.24(169.50,231.91)pg/Ml and 193.64(172.45,225.12)pg/Ml foRADP and NAS groups,respectively(P=0.117)(Figure 1B). The se results indicate that The elevated level of seruMhistaMine is not The result of AhistaMine-related metabolic disorder,suggesting that histaMinemay be involved in The pathogenetic process of adolescent depressive disorder.

Figure 1 Adolescent patients with depressive disordeRhave significantly-elevated seruMhistamine levelsSeruMlevels of histaMine(A)and DAO(B)were quantified in both The ADP and NAS groups.DatAare shown as median(quartile spacing)[M(P25,P75)](N=184).P values were deterMined using Mann-Whitney U test and differences With P＜0.05 are considered significant.DAO,diaMine oxidase;ADP,adolescent depressive disordeRpatient;NAS,normal adolescent student.

The ADP group exhibited higheRpositive rate of IgE and food antigen-specific IgG

H istaMine ismainly stored in labrocytes and basophils,and it is released when IgE bound to allergens interacts With IgE receptors on labrocytes and basophils.Arecent study shows that in addition to IgE receptors,IgG receptors are also expressed on The surface of labrocytes and basophils[26,27].In addition,in animal studies,antigen-bound IgG immune comp lexeshavealso been shown to inducehigh levelsof seruMhistaMine[28].To investigate whe The Rhigh levels of seruMhistaMines are caused by IgE oRIgG,we exaMined The level of IgE and 14 types of food antigen-specific IgG in seruMsamp les froMThe two groups(Table1).Among The 184 patients in The ADP group,66 (35.87%) patients had high IgE(＞100 KU/l),whereas The corresponding numbeRis 42(22.83%)in The NASgroup(P＜0.001,Chi-square test).The average IgE concentrations of The ADP and NASgroupswere 49.80(10.00,413.98)KU/l and 31.63(10.00,88.50)KU/l(P＜0.001,Mann-Whitney U test),respectively. The high positive rate of IgE in The ADP group suggests that IgE-mediated type Ihypersensitivitymay be involved in The elevated level of seruMhistaMine in adolescent patientsWith depression.

The positive rate of food antigen-specific IgG was 89.67%(165/184)in The ADP group and 13.04%(24/184)in The NAS group(P＜0.001,Chi-square test;Table 1).All The 14 types of food exaMined in this study are commonly consumed by The Chinese peop le.Chronic contact With The se food antigens Might induce a higheRfood antigen-specific IgGmediated type III hypersensitivity reaction state in The ADP group.Overall,The positive rate of food antigen-specific IgG was higheRin The ADP group,suggesting that IgG-mediated type III hypersensitivity may be primarily responsible foRThe elevated seruMhistaMine level in adolescent patients With depression.

The seruMS100B levelswere higheRin The ADP group

Normally,histaMine doesnot pass through The BBB,however,long-terMhigh levelof seruMhistaMine has been confirmed to cause high permeability of The BBB[18].The finding that The seruMhistaMine levelwas higheRin The ADP group prompted us to deterMine whe The RoRnot The elevated seruMhistaMine level in ADP group is associated With BBB leakage.S100 calcium-binding protein B(S100B),which ismainly expressed in The astrocytesof The CNS,has been considered AbiomarkeRof BBB leakage.We thusmeasured The levelof seruMS100B in The se two groups.We found that The average concentration of seruMS100B in The ADP group[901.97(713.84,1039.07)ng/l]was significantly higheRthan that in The NAS group[725.17(691.17,786.37)ng/l](P＜0.001,Mann-WhitneyU test)(Figure 2). The refore,The enhanced BBB permeability in The ADP group is probably attributed to The elevated seruMhistaMine level.

Table 1 SeruMIgE and food antigen-specific IgG levels in The ADP and NAS groups

Figure 2 Adolescent patientswith depressive disordeRhave significantly-elevated seruMS100B levelsSeruMlevels of S100B were quantified in both The ADP and NAS groups.DatAare shown as median(quartile spacing)[M(P25,P75)](N=184).P valueswere deterMined using Mann-Whitney U test and differences With P＜0.05 are considered significant.S100B,S100 calcium-binding protein B.

H igheRhomocysteine levelswere detected in The ADP group

Homocysteine is Asulfur-containing aMino acid,which is mainly affected by nutritional deficiency due to loweRB vitaMin intake and worse intestinal absorption. The metabolisMof homocysteine requires The participation of vitaMin B6,vitaMin B12,folic acid,and o The Rsubstances[29].Ifan individual is deficient in The substances above,The concentration of homocysteine in The blood would increase[30].We found that The average concentration of homocysteine in The ADP group[24.00(18.5,28.45)μM]was significantly higheRthan that in The NAS group[9.55(7.45,11.53)μM](P＜0.001,Mann-Whitney U test)(Figure3). The se datAsuggest that The increase in homocysteine concentration may be due to The reduced intestinal absorption caused by hypersensitivity mediated by chronic food antigen-specific IgG and consequently The deficiency of vitaMin B.

The re was no significant difference in systemic inflammatory markers between ADP and NAS groups

Figure 3 Adolescent patients with depressive disordeRhave significantly-elevated seruMhomocysteine levelsSeruMlevels of homocysteine were quantified in both The ADP and NAS groups.DatAare shown as median(quartile spacing)[M(P25,P75)](N=184).P values were deterMined using Mann-Whitney U test and differences with P＜0.05 are considered significant.

Clinical studies have reported that systeMic inflammation plays an important role in The occurrence and development of depression[5,6].To deterMine whe The Rchronic systeMic inflammation p lays Arole in The pathogenesis of depression,we compared The seruMlevels of systeMic inflammation biomarkers,i.e.,high-sensitivity C-reactive protein(hs-CRP)and tumoRnecrosis factor-α(TNF-α),between The se two groups.We found that 37/184 (20.11%)and 34/184(18.48%)of The participants had hs-CRP at The concentration＞1mg/l,With average concentrations of 0.50(0.30,0.84)mg/l and 0.41(0.32,0.84)mg/l(P=0.321,Mann-Whitney U test)in The ADP and NAS groups,respectively.Regarding TNF-α,36/184 of The patients in The ADP group and 39/184 of The healthy controls in The NAS group had an elevated level(＞8.1 pg/Ml),With average concentrations of 6.5(5.7,7.6)pg/Ml and 6.35(5.3,8.0)pg/Ml,respectively(P=0.408,Mann-Whitney U test)(Figure 4). The refore,we did not observe significant difference in The concentrations of The two aforementioned biomarkers between ADP and NAS groups,suggesting that The pathogenesis of depression may not be attributed to systeMic inflammation.

Discussion

Figure 4 Adolescent patients with depressive disordeRand The normal adolescent students have comparable levels of seruMinflammationmarkersSeruMlevels of hs-CRP(A)and TNF-α(B)were quantified in both The ADP and NAS groups.DatAare shown as median(quartile spacing)[M(P25,P75)](N=184).P values were deterMined using Mann-Whitney’s U test and differences with P＜0.05 are considered significant.hs-CRP,high-sensitivity C-reactive protein;TNF-α,tumoRnecrosis factor-α.

Although chronic low-grade systematic inflammation hasbeen imp licated in The progression of some subtypesof depression in adults[5,6],its association With The first-episode adolescent depression is still largely unknown.In this study,we recruited Acohort of 184 adolescent patients With depression and The same numbeRof healthy controlsWith siMilaRages to investigate The possible etiology of adolescent depression.Intriguingly,we did not find any significant difference in The levels of inflammatory markers hs-CRP oRTNF-αbetween The patient and control groups.However,significantly higheRseruMlevels of histaMine,S100B,and homocysteine,as well as AhigheRpositive rate of seruMfood-specific IgG,were demonstrated in The patient group compared to The control group.AhigheRlevel of histaMine suggests Ahypersensitivity mediated by food antigen-specific IgG,which toge The Rwith The enhanced levels of S100B indicates that The BBB leakage may p lay Apotential role in The occurrence and development of adolescent depression. The refore,The concept of chronic food antigen-specific IgG-mediated hypersensitivity oRchronic food intolerance,ra The Rthan The chronic low-grade inflammation,should be acknoWledged oRemphasized in The pathogenesis of adolescent depression.

An increased seruMhistaMine levelhas been shown to lead to AhigheRBBB permeability[18],which isalso reflected by The increased level of S100B even Without brain damage[28,30].H istaMine normally can’t pass through BBB,however,The histaMine-induced increase in BBB permeability could,in turn,promote The entry of histaMine into The CNS.Such Aconcept is supported by previous reports demonstrating higheRhistaMine levels in The cerebrospinal fluid and brain parenchymAof individuals With various neurodegenerative diseases[4-6].Additionally,histaMine can increase The production of some proinflammatory cytokines including TNF-α,nitric oxide(NO),and reactive oxygen species(ROS),leading toMitochondrial membrane potential dysfunction[31].Based on The se studies and ouRfindings,it is reasonable to postulate that higheRblood histaMine levels are closely associated With The onset of adolescent depression.

The development of neuroinflammation could be exp lained viAAnumbeRof mechanisMs.First,The significantly higheRpositive rate of food antigen-specific IgG found in The patient group suggests an increased ability of food antigens to enteRThe systeMic circulation and forMan immune comp lex With food antigen-specific IgG.While The large comp lexes can be easily removed by macrophages,The small ones are very difficult to remove[32]. The refore, The se small coMp lexes easily enteRThe CNS,triggering neuroinflammation,oRprecipitate and become deposited in blood vessels,joints,glomeruli,and o The Rtissues,causing tissue damage,inflammation,and metabolic abnormalities[33].Some evidence has shown that The high permeability and neurovasculaRdysfunction of The BBB are closely related to MDD[19].Second,The high intestinal wall permeability caused by food antigen-specific IgGmediated hypersensitivity reaction can cause endotoxins froMintestinalG ram-negative bacteriAto enteRThe systeMic circulation and The CNSundeRconditionsof high BBB permeability,and this process can lead to local cerebral neuroinflammation[3].Third,The higheRseruMhomocysteine levelsmay interact With The N-methyl-D-aspartate receptoRto produce free radicals that induce neurotoxicity[30].In addition,ouRprevious study revealed elevated seruMuric acid level in adolescent patients with depression[34].All of The se factors can enteRThe CNS and cause cell damage and inflammatory response through The enhanced oxidative stress undeRThe condition of high BBB permeability.Moreover,once The se factors enteRThe CNS through The highly permeable BBB, The y stimulate astrocytes and MicrogliAto produce proinflammatory cytokines. The se proinflammatory cytokines affect The metabolic processes of certain neurotransMitters,such as 5-HT and dopaMine, The reby affecting neuronal transMission.

Focusing on first-episode depression patients to explore The etiology of adolescent depression hasan advantageoveRfocusing on adult patientsWith depressive disorders becausemany factors,such as chronic diseases,substance addiction,and social environment,may interfere with clinical outcomes.The cohort that was evaluated in ouRcurrent study included only students aged 14-20 years With siMilaRsocial environments and also exclusion of any chronic diseases.All The patientswere first diagnosed Without receiving any antidepressant treatment before The clinical study. The refore,ouRfindings should accurately reflect The early changes in biological markers related to depression.More iMportantly,ouRfindings suggest that The pathogenesisof adolescent depression is different froMthat of adult depression,With systeMic inflammation p laying Arole in The disease process. The refore,ouRresultsmay suggest a stage-specific manifestation evidenced by The increased seruMuric acid level found in adolescents with depression in ouRprevious report[34].Moreover,adult depression of ten originates froMThe onset of adolescent depression[17].

Conclusion

According to ouRstudies,webelieve thatadolescentdepression may result froMimmune disorders,metabolic disorders,and nutritional imbalances. The se findings suggest that to block food antigen-specific IgG-mediated hypersensitivity may be AneWmechanisMfoRThe treatment of MDD. The refore,avoiding allergy-inducing food and using antihistaMine drugsmay be The future direction of adolescent depression treatment,and The semethods have been included in ouRfuture research p lan.Second,The detection of IgG,IgE,histaMine,and o The Rindicators would provide AneWobjective basis foRThe early diagnosis of depression,and also provide Areliable basis foRThe evaluation of The treatment of depression.Finally,in addition to MDD,o The RCNS diseases such as Alzheimer’s disease,Parkinson’s disease,and epilepsy are also associated with increased BBB permeability. The refore,we conclude that long-terMfood antigen-specific IgG-mediated hypersensitivity may also beassociated With The pathogenesisof The se CNSdiseases,and this topic remainsworthy of fur The Rinvestigations.

Materials and methods

Subject information

Patient enrollment foRThe ADP group

Patient enrollment was performed by Achiefphysician,an attending physician,and two residents froMThe Department of Addiction Medicine,The Seventh Medical CenteRof The General Hospital of The Peop le’s Liberation Army(PLAGH).

Patients first visited The Department froMFebruary 2015 to DecembeR2016 froMalloveRChina.The inclusion criteriAfoRpatient enrollment are as follows.(1)Patientsmet The diagnostic criteriAfoRdepression according to The D iagnostic and Statistical Manual of Mental D isorders,volume 5(DSM-5).(2)Patients had AHAMD-17 score ranging 17-24.(3)Patients were aged 14-20 years.(4)Patientswere not treated With any psychotropic drugs.(5)The patients were free diseases of immune system,liver,and kidney,gout,oRany physical diseases.(6)Patients had normal levels of blood ureAnitrogen and creatinine.

Control enrollment foRThe NAS group

Controlenrollmentwascompleted by an attending psychiatrist and two residents froMThe Department of Addiction Medicine,The Seventh Medical CenteRof The PLAGH.The healthy controls were Beijing high school students who came to The hospital foRamedical checkup froMFebruary 2015 to DecembeR2016.Physical exaMinations indicated that The se students were healthy.In addition, The se students did not have any mental disorders based on Apsychiatric exaMination by The attending psychiatrist.

Both patients and control individuals answered The HAMD-17 questionnaire and read and signed The informed consent forMthat was reviewed and approved by The Ethical RevieWComMittee of The Seventh Medical Center,PLAGH.

Blood sample collection

Blood saMples froMThe patient group were collected by The Department of Addiction Medicine,The Seventh MedicalCenteRof The PLAGH Within 3 days of patient adMission,while blood saMp les froMThe control group were collected on The day of medical check-up.All subjects were forbidden to smoke,drink oRuse various drugs three days before testing and consumed Alight diet.FoReach subject,4Ml of fasting blood samp leswere collected in The morning into blood collection tubes and agglutinated at The rooMtemperature.SeruMwas centrifuged at 3000 rpMfoR10Min at 4°C and The resulting supernatantwas stored at-20°C or-80°C prioRto use.

BiocheMical analyses

All blood samp les were sent to D i’an Medical ExaMination Center(Beijing,China)foRtesting.

Measurementof CRP(hs-CRP),IgE,and TNF-αconcentrations

An emulsion technique and electrocheMiluMinescence were adopted foRThe measurement of CRP and IgE using The c701 and e602 modules,respectively,on an automatic biocheMicalanalyzer(Catalog No Cobas8000;Roche,Germany),With The matching original reagents(Catalog Nos05950864190[35]and 04827031190[36]),respectively.The concentration of TNF-αwasmeasured cheMiluMinescently on an immunoassay system(Immulite1000,Catalog No.04827031190;Siemens)[35].

Measurement of food antigen-specific IgG concentrations

An ELISAassay was used to simultaneously deterMine The levels of 14 food-specific IgG antibodies in The seruMsamp les obtained from all subjects,folloWing The manufacturer’s instructions(Catalog No.7194;Biomerica;Irvine,CA)[36].The food-specific IgG antibodies tested include Milk,beef,chicken,pork,cod fish,corn,rice,crab,eggwhite/yolk,shrimp,soybean,tomato,mushroom,and wheat.

Measurement of homocysteine,histamine,S100B,and DAO concentrations

Enzymatic cycling assay wasused tomeasure homocysteine on an automatic biocheMical analyzer(Catalog No Cobas8000;Roche,Germany)using The c701 module and The matching original reagent(Catalog No.C0805)[37].Concentration of histaMine,S100B,and DAO was measured by ELISA,double-antibody sandWich ELISA, and Micromethod using The reagent kit froMNovus Biologicals(Catalog No.NBP2-62860),R&D Systems(Catalog No.DY 1820-05),and Shanghai Yubo Biotechnology(Catalog No.YB-DAO-Hu),respectively.All The assays were measured on an A0018T Microp late reader( The rmo-FisheRScientific,USA)folloWing The instructions of reagent supp liers[38,39].

Statistical analysis

DatAwere presented as numbeRand percentages foRcategorical variables,while continuous datAof normal distribution wasexpressed asmean±standard deviation(SD)and continuous datAof non-normal distribution were expressed asmedian(quartile spacing)[M(P25,P75)],and Mann-Whitney’s U test was used foRcomparison between groups.Inter-group difference was coMpared using Chi-square test oRFisher’s exact test foRcategorical variables(gender,distribution of patients positive foRTNF-α,hs-CRP,IgE,and food antigenspecific IgG).All statistical analyses were performed using G raphPad PrisM6.0(G raphPad Sof tware,San D iego,CA),where P＜0.05 was considered statistically significant.

Authors’contributions

RT conceived The ideAand designed The project.Zfand LX analyzed The data.Zfand LX drafted The manuscript.All authors edited The manuscript,read and approved The final manuscript.

Competing interests

The authors declare no conflicts of interest.

AcknoWledgments

This study was supported by The Youth Psychological Development Base in China.We are very grateful foRThe assistance of various parties.Wewould like to thank Jie Liu and Shuyan Zhang foRperforMing The analysesof biologicalspecimens.We would also like to thank Prof essoRHong An and X injie Yang foR The iRhelp With English language editing.