dapsone‐ and nitroso dapsone‐specific activation of t
TRANSCRIPT
Allergy. 2019;1–16. | 1wileyonlinelibrary.com/journal/all
Received:26September2018 | Revised:18December2018 | Accepted:10January2019DOI:10.1111/all.13769
O R I G I N A L A R T I C L E
Skin and Eye Diseases
Dapsone‐ and nitroso dapsone‐specific activation of T cells from hypersensitive patients expressing the risk allele HLA‐B*13:01
Qing Zhao1,2,3 | Khetam Alhilali1 | Abdulaziz Alzahrani1,4,5 | Mubarak Almutairi1 | Juwaria Amjad1 | Hong Liu2,3 | Yonghu Sun2,3 | Lele Sun2,3 | Huimin Zhang2,3 | Xiaoli Meng1 | Andrew Gibson1 | Monday O. Ogese1,6 | B. Kevin Park1 | Jianjun Liu7 | David A. Ostrov8 | Furen Zhang2,3 | Dean J. Naisbitt1
1MRCCentreforDrugSafetyScience,DepartmentofMolecularandClinicalPharmacology,TheUniversityofLiverpool,Liverpool,UK2DepartmentofDermatology,ShandongProvincialHospitalforSkinDisease,ShandongUniversity,Jinan,China3ShandongProvincialInstituteofDermatologyandVenereology,ShandongAcademyofMedicalSciences,Jinan,China4AlBahaUniversity,PrinceMohammadBinSaud,AlBahah,SaudiArabia5PharmacologyDepartment,CollegeofClinicalPharmacy,AlBahaUniversity,AlBaha,SaudiArabia6PathologicalSciences,drugSafetyandMetabolism,IMEDBiotechUnit,AstraZeneca,Cambridge,UK7HumanGenetics,GenomeInstituteofSingapore,A*STAR,Singapore,Singapore8DepartmentofPathology,ImmunologyandLaboratoryMedicine,CollegeofMedicine,UniversityofFlorida,Gainesville,Florida
ThisisanopenaccessarticleunderthetermsoftheCreativeCommonsAttributionLicense,whichpermitsuse,distributionandreproductioninanymedium,providedtheoriginalworkisproperlycited.©2019TheAuthors.AllergyPublishedbyJohnWiley&SonsLtd.
Zhao,AlhilaliandAlzahranicontributedequallytothemanuscript.
Abbreviations:APC,antigen‐presentingcells;DDS,dapsone;DDS‐NO,nitrosodapsone;PBMC,peripheralbloodmononuclearcells;SI,stimulationindex.
CorrespondenceDeanJ.Naisbitt,TheUniversityofLiverpool,Liverpool,UK.Email:[email protected],ShandongProvincialHospitalforSkinDisease,ShandongUniversity,Jinan,China.Email:[email protected]
Funding informationThisworkwasfundedbygrantsfromtheMRC(grantnumberMR/R009635/1),theNationalKeyResearchandDevelopmentProgramofChina(2016YFE0201500),theNationalNaturalScienceFoundationofChinakeyprogramforinternationalexchangeandcooperation(81620108025),theNationalNaturalScienceFoundationofChina(81703110,81701579),theKeyresearchanddevelopmentprogramofShandongProvince(2016ZDJS07A06),theNaturalScienceFoundationofShandong
AbstractBackground: ResearchintodrughypersensitivityassociatedwiththeexpressionofspecificHLAalleleshas focussedon the interactionbetweenparentdrugand theHLAwithnoattentiongiventoreactivemetabolites.Forthisreason,wehavestudiedHLA‐B*13:01‐linkeddapsonehypersensitivityto(a)explorewhethertheparentdrugand/ornitrosometaboliteactivateTcellsand(b)determinewhetherHLA‐B*13:01isinvolvedintheresponse.Methods: Peripheralbloodmononuclearcells(PBMC)fromsixpatientswereculturedwith dapsone and nitroso dapsone, and proliferative responses and IFN‐γ releaseweremeasured.Dapsone‐andnitrosodapsone‐specificT‐cellclonesweregeneratedand phenotype, function, HLA allele restriction, and cross‐reactivity assessed.Dapsoneintermediateswerecharacterizedbymassspectrometry.Results: PeripheralbloodmononuclearcellsfromsixpatientsandclonedTcellspro‐liferated and secretedTh1/2/22 cytokineswhen stimulatedwithdapsone (clones:n=395; 80%CD4+ CXCR3hiCCR4hi, 20%CD8+CXCR3hiCCR4hiCCR6hiCCR9hiCCR1
2 | ZHAO et Al.
1 | INTRODUC TION
Genome‐wideassociationstudieshave identifiedstrongassoci‐ationsbetweentheexpressionofHLAallelesandsusceptibilitytodrughypersensitivity.1Thesedatasuggestthatdrugsbindse‐lectively to theHLAprotein to activate theT cells that partic‐ipate in the adverse event. Molecular docking studies supportthis concept2,3; however,modeling data have to be interpretedwith cautionas thenatureof thedrug‐HLAprotein interactionandtherequirementforaspecificpeptideinthebindinggroovehasnotbeendetermined.Themostrobustgeneticassociationsare between HLA class I alleles and abacavir hypersensitiv‐ity,4 flucloxacillin liver injury (both HLA‐B*57:01),5 carbamaze‐pine‐induced Stevens Johnson syndrome (HLA‐B*15:02),6 and allopurinol hypersensitivity (HLA‐B*58:01),7 and in each case,mechanisticstudieshaveshownthatCD8+Tcellsareactivatedwhen the drug interacts with the relevant HLA protein.7,8 Forabacavir and carbamazepine, the drugHLAbinding site is verydifferent;abacavirbindsdeepintheHLApeptidebindingpocket,whilecarbamazepinebindstoasiteclosertotheT‐cellreceptorinterface.Despitethis,bothdrugsinteractwithHLAproteinsviaareversibleinteractiontostimulateTcells.Tcellsfrompatientswithallopurinolhypersensitivityareactivatedwithastableme‐tabolite, oxypurinol, also via a direct binding interaction withHLA.Incontrast,flucloxacillin‐specificTcellsfrompatientswithliver injury are activatedwith drug‐protein adducts, via a hap‐tenmechanisminvolvingthespontaneousbindingofthedrugtoprotein and antigen processing. This brief discussion illustratesthat rapidprogresshasbeenmade inourunderstandingof therelationshipbetweendrugHLAbindingand theactivationofTcells; however, reactive drugmetabolites have been ignored inthestudyofHLAallele‐restrictedformsofdrughypersensitivity.
Thisisprimarilybecauseoftheabsenceofsyntheticreactiveme‐tabolitesforfunctionalstudieswithTcells fromhypersensitivepatients.
Areactivemetaboliteofsulfamethoxazoleisknowntoactivatepatient T cells via a hapten mechanism12,13; however, sulfame‐thoxazolereactionsarenotassociatedwithaspecificHLAallele.Thus,werecentlysynthesizedthenitrosometaboliteofdapsoneand studied the priming of naïve T cells from healthy donors.15 Dapsonecontainsasulfonegroupthat linkstwoaromaticaminemoieties.Oxidativemetabolismof theaminegroupsgeneratesahydroxylamine. The hydroxylamine undergoes spontaneous oxi‐dationtoformnitrosodapsone,whichbindscovalentlytocellularproteins.16,17NaïveCD4+ andCD8+ T cells fromhealthy donorsareactivatedwith theparentdrugandnitrosometabolitewhenTregswereremovedandthecompoundswerepresentedbyden‐driticcells.15ThesedatashowthatbothformsofthedruginteractwithmultipleHLAmoleculesandhavethecapacitytostimulateTcellswhenregulatorypathwayshavebeenmanipulated.
Dapsone is used in combinationwith other drugs for the treat‐mentofinfectiousdiseasessuchasleprosyandmalaria.0.5%‐3.6%oftreatedpatientsdevelopahypersensitivitysyndromecharacterizedbyfever,skinrash,andinternalorganinvolvement4‐6weeksaftertreat‐mentcommences.18HLA‐B*13:01isassociatedwiththedevelopmentofdapsonehypersensitivityinChineseandThaipatients,19,20 and mod‐elingdatasuggestthatdapsonemayfitinthepeptiderecognitionsiteofHLA‐B*13:01.21Dapsone‐treatedcelllinesexpressingHLA‐B*13:01havebeenshowntoactivateTcells,whileperipheralbloodmononu‐clearcells(PBMC)from2/7patientssecretehighlevelsofthecytolyticmoleculegranulysinwhenstimulatedwiththedrug.22Despitethis,adetailedanalysisof thephenotypeandfunctionofdapsone‐specificTcellshasnotbeenperformed.Moreover,theactivationofpatientTcellswithnitrosodapsonehasnotbeeninvestigated.Thus,ourstudy
Province(JQ201616),theNationalClinicalKeyProjectofDermatologyandVenereology,theShandongProvincialAdvancedTaishanScholarConstructionProject,andtheInnovationProjectofShandongAcademyofMedicalScience.CentralfundswereobtainedfromtheMRCCentreforDrugSafetyScience(G0700654).
0hi)andnitrosodapsone(clones:n=399;78%CD4+,22%CD8+withsamechemokinereceptorprofile).CD4+andCD8+cloneswereHLAclassIIandclassIrestricted,re‐spectively, and displayed three patterns of reactivity: compound specific, weaklycross‐reactive,andstronglycross‐reactive.Nitrosodapsoneformeddimersincultureandwasreducedtodapsone,providingarationaleforthecross‐reactivity.T‐cellre‐sponsestonitrosodapsoneweredependentontheformationofacysteine‐modifiedproteinadduct,whiledapsoneinteractedinalabilemannerwithantigen‐presentingcells.CD8+clonesdisplayedanHLA‐B*13:01‐restrictedpatternofactivation.Conclusion: Thesestudiesdescribethephenotypeandfunctionofdapsone‐andni‐troso dapsone‐responsive CD4+ and CD8+ T cells from hypersensitive patients.DiscoveryofHLA‐B*13:01‐restrictedCD8+T‐cellresponsesindicatesthatdrugsandtheirreactivemetabolitesparticipateinHLAallele‐linkedformsofhypersensitivity.
K E Y W O R D S
drughypersensitivity,human,Tcells
| 3ZHAO et Al.
hadthreeprimaryobjectives:toinvestigatewhetherdapsoneand/ornitrosodapsoneactivatesCD4+andCD8+Tcellsfromhypersensitivepatients,todefinephenotypeandfunctionofdrug‐specificTcells,andtoexplorewhetherHLA‐B*13:01isdirectlyinvolvedinthedrug‐/drugmetabolite‐specificT‐cellresponse.
2 | METHODS
2.1 | Human subjects
Venousblood(50mL)wascollectedfrom6dapsonehypersensitivepatients.Table1summarizesthedemographicsofthepatients.TableS1showsresultsofHLAtyping.HLA‐B*13:01+donors(n=4)withnohistoryofdapsoneexposureandHLA‐B*13:01+dapsone‐toler‐antpatients(n=4)wereselectedasacontrolgroups.Patchtestingwas conductedon thebackof patientswith dapsone (0.1%‐25%).Thepatchwasremovedafter48hours.Resultswererecordedaftera further 24hours to exclude any false‐positive responses result‐ing from the patch tape. The study was approved by the EthicalCommittee of the Shandong Provincial Institute of Dermatologyand Venereology, and informed written consent was obtained. A
materialtransferagreementwassignedpriortoshipmentofPBMCtoLiverpool.
2.2 | Lymphocyte transformation test and PBMC ELIspot
Peripheral blood mononuclear cells (1.5×105cell/well) from hy‐persensitivepatientsandcontroldonorswereincubatedwithdap‐sone(125‐500µmol/L),nitrosodapsone(10‐40µmol/L),rifampicin(10‐100µmol/L), clofazimine (10‐100µmol/L), or tetanus toxoid(5 μg/mL,asapositivecontrol)inculturemediumfor5days.[3H]thy‐midinewasaddedforthefinal16hoursoftheexperiment. IFN‐γ‐secretingPBMCwere visualizedusingELIspotby culturingPBMC(5×105 cell/well)inmediumwith/withoutoptimalconcentrationsofdapsoneornitrosodapsonefor48hours.
2.3 | Generation and characterization of drug‐specific T‐cell clones
Dapsoneandnitrosodapsone‐responsiveT‐cellclonesweregener‐atedfrompatients5,6,and8byserialdilutionandcharacterizedin
G R A P H I C A L A B S T R A C TDapsoneandnitrosodapsoneactivateCD4+andCD8+Tcellsfromhypersensitivepatientsviadifferentmechanisms.DapsoneandnitrosodapsoneinteractwithHLA‐B*13:01toactivatecertainCD8+clones.Dapsonehypersensitivityshouldbeusedasanexampletosolvethestructureofdrugmetabolite‐modifiedpeptide‐HLAinteractionsandtherelationshipbetweenHLAbindingandT‐cellactivation.
4 | ZHAO et Al.
termsofcellularphenotype(CD,TCRVβ,andchemokinereceptors[flow cytometry]), secretion of cytokines and cytolytic molecules(ELIspot),antigenspecificity(dose‐responsestudiesandcross‐reac‐tivitywith structurally‐related compounds [3H]thymidine uptake]),pathwaysofactivation(antigen‐presentingcellpulsingexperiments,fixationofantigen‐presentingcellstoblockprocessing,andadditionof glutathione to block drugmetabolite protein binding), and theinvolvementofHLA‐B*13:01 intheT‐cell response (HLAantibodyblockingexperimentsanduseofpartlyHLAmatchedantigen‐pre‐sentingcells).DetailedmethodsareavailableinasDataS1.
2.4 | Statistics
All statistical analysis (one‐way ANOVA unless stated otherwise)wasperformedusingSigmaPlot12software(*P<0.05).
3 | RESULTS
3.1 | Patch testing and in vitro activation of hypersensitive patient PBMC
After 72‐hour drug exposure, 3 out of 6 patients displayed dap‐soneconcentration‐dependentpositivereadings(Figure1A).PBMCfromall3patchtest‐positivepatientswerestimulatedtoproliferate
stronglyinthepresenceofdapsoneandnitrosodapsone(Figure1B).Positivedapsone‐specificproliferativeresponses(stimulationindex[SI] 2 or above) and/or IFN‐γ secretion were also detected withPBMCfrom the3patch test‐negativepatients,whilenitrosodap‐soneresponsesweredetectedin2patients(Figure1B‐D).Ofnote,patient 7 displayed a negativenitrosodapsone lymphocyte trans‐formation test response on initial testing (Figure 1B) and a weakresponse when the assay was repeated (Figure 1D). PBMC werenot activated with the co‐medications rifampicin and clofazimine(Figure1D).PBMCfromdapsone‐naïve (SI<2)anddapsone‐toler‐ant(FigureS1)HLA‐B*13:01+controlsproliferatedinthepresenceofphytohemagglutinin,butnotthetestdrugs.
3.2 | Dapsone and nitroso dapsone activate CD4+ and CD8+ clones
A total of 1334 and 1374 CD4+ and CD8+ clones, respectively,wereexpandedfromdapsoneandnitrosodapsoneT‐celllinesfrompatients5,6, and8;626CD4+ clonesdisplayed reactivity againstdapsoneornitrosodapsone(FigureS2).Dapsone‐andnitrosodap‐sone‐specificCD8+ clonesweregenerated in lowernumbers;168wereactivatedwitheither thedrugordrugmetabolite.Dapsone‐and nitroso dapsone‐responsive CD4+ and CD8+ cloneswere de‐tectedinequalnumbers(Table2).
TA B L E 1 Patients’demographicsanddetailsofthehypersensitivityreaction
Patient ID Gender Age (years) Medication historyOnset of symptoms (days) Clinical presentation Skin patch test
1 Female 43 Dapsone,rifampin,andclofazimine
3 Fever,rash,andabnormalliverfunctiontests
−
3 Female 25 Dapsone,rifampin,andclofazimine
28 Fever,rash,andabnormalliverfunctiontests
−
5 Male 39 Dapsone,rifampin,andclofazimine
30 Fever +
6 Male 41 Dapsone,rifampin,andclofazimine.
48 Fever,rash,andlymphadenopathy +
7 Male 54 Dapsone,rifampin,andclofazimine
16 Fever,rash,andabnormalliverfunctiontests(AST47.7U/L;ALP132.3U/L)
−
8 Female 27 Dapsone,rifampin,andclofazimine
17 Feverandabnormalliverfunctiontests(AST56.3U/L;ALP225.7U/L;GGT184.8U/L;TBIL38.8μmol/L;DBIL21.3μmol/L;IBIL17.5)
+
F I G U R E 1 Diagnosisofdapsonehypersensitivitybyskintestingandinvitroassays.A,Skinpatchtestresultsfordapsonehypersensitivity.Skinwasexposedtodapsoneinpolyethyleneglycol200atdilutionsof0%,0.1%,0.5%,1%,5%,10%,15%,2%,and25%.Thepatchtapewaslefttotheskinfor48h,anddiagnosiswasmade24hlater.Redovalsshowareasofinflamedskin.B,Peripheralbloodmononuclearcells(PBMC)frompatientswereexposedtogradedconcentrationsofdapsoneornitrosodapsone.Proliferationwasmeasuredafter5dbytheadditionof[3H]thymidinefor16h.Resultsareexpressedasmean±SDcpmoftriplicatecultures.Adoublingofcpmindrug‐treatedculturesovervehiclecontrolisconsideredpositive.C,PBMCfrompatientswereexposedtooptimalconcentrationsofdapsoneornitrosodapsone,andIFN‐γreleasewasvisualizedbyELIspot.D,PBMCfrompatientswereexposedtogradedconcentrationsofdapsone,nitrosodapsone,rifampicin,andclofazimine.Proliferationwasmeasuredasdescribedin(B)(DDS,dapsone;DDS‐NO,nitrosodapsone;PHA,phytohaemagglutinin)
| 5ZHAO et Al.
(A)
(C)
(D)
(B)
6 | ZHAO et Al.
One hundred and two well‐growing clones were selectedfor dose‐titration studies and the analysis of cytokine secretion.Proliferative responses were detected with 3 well‐tolerated con‐centrations of dapsone (125‐500µmol/L) and nitroso dapsone(5‐20µmol/L)anddrugtreatmentresulted in thesecretionofTh1(IFNγ),Th2(IL‐5,IL‐13),andTh22(IL‐22)cytokines,alongsidethecy‐tolyticmoleculesperforin,granzymeB,andFasL.CD4+andCD8+ clonessecretedsimilar levelsofcytokinesandcytolyticmolecules(Figure2). Table S3 shows thepercentageof dapsone (DDS)‐ andnitrosodapsone(DDS‐NO)‐responsiveCD4+andCD8+clonesthatsecreteindividualcytokines.
3.3 | Dapsone‐ and nitroso dapsone‐responsive CD4+ and CD8+ clones display three distinct patterns of cross‐reactivity
Sixty‐three dapsone‐ and 98 nitroso dapsone‐responsive CD4+ and CD8+ clones were assayed for cross‐reactivity. When all ofthedapsone‐responsivecloneswereassessed together, low levelsof proliferation were observed with nitroso dapsone (Figure 3A).Themaximumconcentrationofnitrosodapsoneusedwas10timeslowerthanthedapsoneconcentration.Analysisofindividualclonesrevealed three distinct cross‐reactivity patterns: dapsone‐spe‐cific,andweaklyandstronglycross‐reactivewithnitrosodapsone.Approximately90%oftheclonesweredapsone‐specificorweaklycross‐reactive(Figure3B,C).
NitrosodapsoneCD4+andCD8+clonesdisplayedamuchhigherlevelofcross‐reactivity (Figure3A).However,3patternsofcross‐reactivitywereagainobservedwithindividualclones:nitrosodap‐sone‐specific,andweaklyandstronglycross‐reactivewithdapsone.Incontrasttothedapsone‐responsiveclones,approximately90%ofthenitrosodapsone‐responsive cloneswerenitrosodapsone‐spe‐cificorhighlycross‐reactive(Figure3B,C).
Dapsone‐ and nitroso dapsone‐responsive clones were alsostimulated to proliferate with dapsone hydroxylamine; however,proliferative responseswere not detectedwhen cloneswere cul‐turedwith (a)dapsoneanalogueswithsubstitutions inthesulfonegroup, (b)dapsoneanalogueswithaminegroups indifferentposi‐tionsonthearomaticrings,and(c)structurallydistinctsulfonamideantimicrobials(FigureS3).
The stability of nitrosodapsone in theproliferation assaywasassessed.Nitrosodapsonewas converted rapidly to azoxydimersandtheparentcompound.Bothcompoundsweredetectablewithin10minutes. After 2days, 50% of nitroso dapsone had been con‐vertedtodapsone(Figure3D‐E).
3.4 | Cross‐reactive clones are activated with equivalent concentrations of dapsone and nitroso dapsone
Six strong and weakly cross‐reactive clones were incubated withdapsoneandnitrosodapsoneatconcentrationsof0.1‐100µmol/Ltodefinetheminimumstimulatoryconcentrationsofthetwocom‐pounds. Clones were stimulated to proliferate with equivalentconcentrations of each compound irrespective of the extent ofcross‐reactivity(Figure4).
3.5 | Dapsone‐ and nitroso dapsone‐responsive CD4+ and CD8+ clones expressing multiple TCR Vβ chains display distinct chemokine receptor profiles
Dapsone‐andnitrosodapsone‐responsiveCD4+andCD8+clonesexpressedsingle,butvariableTCRVβchains(FigureS4A),withnoclear differences discerniblewhen CD4+ and CD8+ or dapsone‐andnitrosodapsone‐responsivecloneswerecompared.Dapsone‐and nitroso dapsone‐responsive CD4+ clones expressed high
Phenotype and drug specificity
Total number of clones
Number of drug‐spe‐cific clones
Percentage of responding clones (%)
Patient8
CD4+,DDS+ 236 176 74.58
CD8+,DDS+ 245 29 11.84
CD4+,DDS‐NO+ 384 210 54.69
CD8+,DDS‐NO+ 304 54 17.76
Patient6
CD4+,DDS+ 177 76 42.94
CD8+,DDS+ 178 14 7.87
CD4+,DDS‐NO+ 208 57 27.40
CD8+,DDS‐NO+ 285 15 5.26
Patient5
CD4+,DDS+ 180 64 35.56
CD8+,DDS+ 170 36 21.18
CD4+,DDS‐NO+ 149 43 28.86
CD8+,DDS‐NO+ 192 20 10.42
TA B L E 2 PhenotypeanddrugspecificityofT‐cellclonesgeneratedfromdapsonehypersensitivepatients
| 7ZHAO et Al.
F I G U R E 2 Proliferativeresponseandcytokinereleasebydapsone‐andnitrosodapsone‐responsiveCD4+andCD8+clones.A,OnehundredandtwoCD4+orCD8+cloneswereincubatedwithautologousantigen‐presentingcellsandeitherdapsone(125‐500µmol/L)ornitrosodapsone(5‐20µmol/L)intriplicateculturesfor48h.Proliferativeresponsesweremeasuredbytheadditionof[3H]thymidine.Resultsareexpressedasmean±SDcpmoftheindicatednumberofclones.B,DetectionofIFN‐γ,IL‐5,IL‐13,IL‐17,IL‐22,granzymeB,perforin,andFas‐ligandsecretionbydapsone‐andnitrosodapsone‐responsiveCD4+andCD8+clones.Cloneswereincubatedwithautologousantigen‐presentingcells,andeitherdapsoneornitrosodapsoneandcytokinereleasewasvisualizedbyELIspot.C,RepresentativeELIspotimagesshowingthecytokinesreleasedbydapsoneandnitrosodapsone‐responsiveCD8+clones
0
10
20
30
40
50
60
70
80
90
0 5 10 20
DDS-NO clones (n = 20)
0
100
200
300
400
500
600
IFN
-γIL
-5IL
-13
IL-1
7IL
-22
Perf
orin
Gran
zym
e B
FasL
0
100
200
300
400
500
600
IFN
-γIL
-5IL
-13
IL-1
7IL
-22
Perf
orin
Gran
zym
e B
FasL
0
10
20
30
40
50
60
70
80
90
0 125 250 500
DDS clones (n = 25)
0
10
20
30
40
50
60
70
80
90
0 5 10 20
DDS-NO clones (n = 21)
0
10
20
30
40
50
60
70
80
90
0 125 250 500
DDS clones (n = 36)
0
100
200
300
400
500
600
IFN
-γIL
-5IL
-13
IL-1
7IL
-22
Perf
orin
Gran
zym
e B
FasL
0
100
200
300
400
500
600
IFN
-γ
IL-5
IL-1
3IL
-22
Perf
orin
Gran
azym
e B
IL-1
7Fa
s L
DDS-specific CD8+ T-cell clone
0
DDS500 µmol/L
DDS-NO-specific CD8+ T-cell clone
IFN-γ IL-5 IL- 13 IL-22
0
DDS NO20 µmol/L
Granzyme-B perforinIL-17Fas-
Ligand
Vehicle Dapsone Nitroso dapsone
CD4+ clones CD8+ clones
( noitarefilorPcp
mx
10-3
)( esaeler enikotyCsf
u)
Concentration (µmol/L)
IFN-γ IL-5 IL- 13 IL-22 Granzyme-B perforinIL-17Fas-
Ligand
(A)
(C)
(B)
8 | ZHAO et Al.
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9
0
20
40
60
80
1 2 3 4 5 6 7 8 9
0
20
40
60
80
100
120
1 2 3 4 5 6 7 8 9
0
20
40
60
80
100
1 2 3 4 5 6 7 8 90
20
40
60
80
100
120
1 2 3 4 5 6 7 8 90
10
20
30
40
50
60
1 2 3 4 5 6 7 8 90
2
4
6
8
10
1 2 3 4 5 6 7 8 9
0
20
40
60
80
100
120
1 2 3 4 5 6 7 8 90
5
10
15
20
25
1 2 3 4 5 6 7 8 90
50
100
150
1 2 3 4 5 6 7 8 9
0
20
40
60
80
100
120
1 2 3 4 5 6 7 8 90
10
20
30
40
50
1 2 3 4 5 6 7 8 90
5
10
15
1 2 3 4 5 6 7 8 9
0
10
20
30
40
1 2 3 4 5 6 7 8 90
10
20
30
40
1 2 3 4 5 6 7 8 90
10
20
30
40
50
1 2 3 4 5 6 7 8 9
Dapsone clones
CD4+ (n = 25)
CD8+ (n = 38)
Nitroso-dapsone clones
CD4+ (n = 53)
CD8+ (n = 45)
Vehicle Dapsone Nitroso dapsone( noitarefilorPcp
mx
10–3
)( noitarefilorPcp
mx
10–3
)
Weakly cross-reactiveStrongly cross-reactive
Compound-specific
0 100 250 500 0 5 10 20
0 100 250 500 0 5 10 20
0 100 250 500 0 5 10 20 0 100 250 500 0 5 10 20 0 100 250 500 0 5 10 20
0 100 250 500 0 5 10 20 0 100 250 500 0 5 10 20 0 100 250 500 0 5 10 20
0 5 10 20 0 100 250 500 0 5 10 20 0 100 250 500 0 5 10 20 0 100 250 500
0 5 10 20 0 100 250 500 0 5 10 20 0 100 250 500 0 5 10 20 0 100 250 500
Concentration (µmol/L)
Weakly crossreactive Strongly crossreactiveCompound-specific(A) (C)(B)
0
2
4
6
8
10
12
14
16
10 min 1 h 4 h 24 h 48 h
y = 3889.7x + 23480R² = 0.9994
05
1015202530354045
0 500 1000Dapsone (µmol/L)
Daps
one
(µm
ol/L
)
Time
Peak
are
a (x
100,
000)
(E)(D)
0
1000
2000
3000
4000
5000
6000
10min
1 h 4 h 24 h 48 h
Dapsone
0
5
10
15
20
25
30
35
40
10min
1 h 4 h 24 h 48 h
Azoxy dimer
)0001x( tnuoc noI
| 9ZHAO et Al.
levels of the chemokine receptors CXCR3 andCCR4. The CD8+ clonesexpressedCXCR3andCCR4alongsideCCR10,CCR9,andCCR6(FigureS4B).
3.6 | HLA‐restricted activation of dapsone‐ and nitroso dapsone‐responsive clones ensues via different mechanisms
Stimulation of dapsone‐ and nitroso dapsone‐responsive CD4+ and CD8+ clones was dependent on the presence of antigen‐present‐ing cells (Figure5A).Useofblockingantibodies revealed thatCD4+ and CD8+ proliferative responses to dapsone and nitroso dapsonewereHLAclass II and I restricted, respectively (Figure5B).Fixationofantigen‐presentingcellswithglutaraldehydehadnoeffectonthe
activationofCD4+orCD8+cloneswithdapsone.Incontrast,antigen‐presenting cell fixation reduced the extent of proliferationwith thenitrosometabolite (Figure6A).The residual responsedetectedwithnitrosodapsoneandfixedantigen‐presentingcellsrelatestothecon‐versionofnitrosodapsonetodapsoneinculture.
Nitroso dapsone‐responsive CD8+ clones were activated withantigen‐presentingcellspulsedwithnitrosodapsonefor0.5‐2hours(Figure6B).Twooutof4nitrosodapsone‐responsiveCD4+cloneswerealsostimulatedtoproliferatewithnitrosodapsone‐pulsedan‐tigen‐presentingcells.Antigen‐presentingcellspulsedwithdapsonefor 0.5‐2hours did not activate the dapsone‐responsive CD4+ or CD8+clones.
Figure S5A compares the level of nitroso dapsone glutathi‐oneadductsformedinthecellcultureassayinthepresenceand
F I G U R E 3 Cross‐reactivityofdapsone‐andnitrosodapsone‐responsiveT‐cellclones.Dapsone‐andnitrosodapsone‐responsiveCD4+ andCD8+cloneswereculturedwithirradiatedautologousEBV‐transformedBcellsandeitherdapsone(100‐500µmol/L)orthenitrosometabolite(5‐20µmol/L)intriplicateculturesfor48h.T‐cellproliferativeresponseswereassessedthroughtheadditionof[3H]‐thymidine.A,Meancross‐reactivitydatafor153clonesdividedaccordingtothedrugantigenperipheralbloodmononuclearcellswereculturedwithtogenerateclonesandCDphenotype.B,Piechartsshowingnumberofcloneswithaparticularcross‐reactivityprofile(compoundspecific,weeklycross‐reactive(cross‐reactivecompounddisplaying10%‐50%responsedetectedwithcomparator)andstronglycross‐reactive(cross‐reactivecompounddisplaying>50%responsedetectedwithcomparator).C,Representativeclonesdisplayingeachresponseprofile.D,Relativequantificationofdapsoneandazoxydimerinculturescontainingnitrosodapsone(30µmol/L),autologousEBV‐transformedBcells,andT‐cellclones.E,Absolutequantificationofdapsoneformationinculturescontainingnitrosodapsone(30µmol/L),autologousEBV‐transformedBcells,andT‐cellclones.Lefthandgraphshowsthestandardcurvefordapsone(concentrationrange:5nmol/L‐1µmol/L).Righthandsidegraphshowsthetime‐dependentformationofdapsone
F I G U R E 4 DapsoneandnitrosodapsoneactivateT‐cellclonesatthesameminimumconcentration.A,Dapsone‐and(B)nitrosodapsone‐responsiveCD4+andCD8+cloneswereculturedwithirradiatedautologousEBV‐transformedBcellsandeitherdapsone(0.1‐100µmol/L)orthenitrosometabolite(0.1‐100µmol/L)intriplicateculturesfor48h.T‐cellproliferativeresponseswereassessedthroughtheadditionof[3H]‐thymidine.Representativeweaklyandstronglycross‐reactivedapsoneandnitrosodapsone‐responsiveclonesareshown
0
20
40
60
0.05 0.5 5 500
1020304050
0.05 0.5 5 500
20
40
60
0.05 0.5 5 50
0
10
20
30
40
0.05 0.5 5 500
20
40
60
80
0.05 0.5 5 500
20406080
100
0.05 0.5 5 50
Dapsone clones
Nitroso-dapsone clones
Vehicle Dapsone Nitroso dapsone
( noitarefilorPcp
mx
10–3
)( noitarefilorPcp
mx
10–3
)
(A)
(B)
Clone 2 Clone 3 Clone 1
Clone 2 Clone 3 Clone 1
Concentration (µmol/L)
Concentration (µmol/L)
O
Nitroso dapsone
Dapsone
10 | ZHAO et Al.
absence of exogenous glutathione.23,24 Adducts were formedrapidly (within 2hours), and the level of adduct formationwasseveralordersofmagnitudehigherinthepresenceofexogenousglutathione.FigureS6containstracesshowinghowtheadductswerequantified.Theadditionofglutathione to theT‐cellassaycontainingsolublenitrosodapsonereducedthestrengthof theproliferative response, while the response to nitroso dapsone‐pulsed antigen‐presenting cellswas completely blocked (FigureS5B).
3.7 | Dapsone and nitroso dapsone bind with a degree of selectively to HLA‐B*13:01 to activate certain CD8+ T‐cell clones
EBV‐transformed B cells were generated from 9 healthy do‐nors expressing HLA alleles with >90% sequence homology toHLA‐B*13:01.HLAtypingofthehealthydonors isshowninTableS2. The B‐cell lines were used to explore the requirement forHLA‐B*13:01indapsone‐andnitrosodapsone‐specificCD8+T‐cell
F I G U R E 5 Antigen‐presentingcellsarerequiredfortheactivationofdapsone‐andnitrosodapsone‐responsiveCD4+andCD8+clones.A,Dapsone‐andnitrosodapsone‐responsiveCD4+ and CD8+cloneswereculturedwithdapsone(500µmol/L)orthenitrosometabolite(20µmol/L)intriplicateculturesfor48heitherinthepresenceorabsenceofautologousEBV‐transformedBcells.B,Dapsone‐andnitrosodapsone‐responsiveCD4+andCD8+cloneswereculturedwithantigen‐presentingcellsanddapsone(500µmol/L)orthenitrosometabolite(20µmol/L)intriplicateculturesfor48heitherinthepresenceorabsenceofanti‐HLAclassIandIIblockingantibodies.[3H]‐thymidinewasaddedfor16htomeasuredrug‐specificproliferativeresponses
0
20
40
60
80
100
020406080
100120
0
20
40
60
80
0102030405060
0
20
40
60
80
NO APC APC
Medium DDS 500uM
0
20
40
60
80
100
Medium DDS
0
10
20
30
40
50
NO APC APC
Medium DDS-NO
0
10
20
30
Medium DDS-NO
Dapsone clones
CD8+ (n = 16)
Nitroso-dapsone clones
CD4+ (n = 20)
CD8+ (n = 22)
(A) (B)
CD4+ (n = 15)
Prol
ifera
tion
(cpm
x 10
–3)
Prol
ifera
tion
(cpm
x 10
–3)
Prol
ifera
tion
(cpm
x 10
–3)
Prol
ifera
tion
(cpm
x 10
–3)
CD8+ (n = 3)
CD4+ (n = 3)
CD8+ (n = 6)
CD4+ (n = 4)
Vehicle Dapsone Nitroso dapsone
| 11ZHAO et Al.
F I G U R E 6 DapsoneandnitrosodapsoneactivateTcellsviadifferentpathways.A,Dapsone‐andnitrosodapsone‐responsiveCD4+ and CD8+cloneswereculturedwithdapsoneorthenitrosometaboliteinthepresenceofeitherirradiatedorglutaraldehyde‐fixedautologousEBV‐transformedBcells(APC,antigen‐presentingcells)intriplicateculturesfor48h.Fixationblocksantigenprocessing.B,Dapsone‐andnitrosodapsone‐responsiveCD4+andCD8+cloneswereculturedwithdrug‐ordrugmetabolite‐pulsed(0.5‐2h)irradiatedautologousEBV‐transformedBcellsintheabsenceofsolubledrugintriplicateculturesfor48h.T‐cellproliferativeresponseswereassessedthroughtheadditionof[3H]‐thymidine
05
1015202530354045
Normal APC Fixed APC
0
20
40
60
80
100
120
Normal APC Fixed APC
0
10
20
30
40
50
60
70
80
Normal APC Fixed APC
0
10
20
30
40
50
60
0
10
20
30
40
50
60
0
10
20
30
40
50
60
0
5
10
15
20
25
30
35
0102030405060708090
100
0
5
10
15
20
25
0
20
40
60
80
100
120
0
10
20
30
40
50
60
0
5
10
15
20
25
30
35
02468
1012141618
0
20
40
60
80
100
120
0
5
10
15
20
25
30
35
40
0102030405060708090
1 2 3 4 50
5
10
15
20
25
30
1 2 3 4 50
102030405060708090
1 2 3 4 50
10
20
30
40
50
60
70
80
1 2 3 4 5
Dapsone clones
CD8+ (n = 16)
Nitroso-dapsone clones
CD4+ (n = 9)
CD8+ (n = 19)
(A) (B)
CD4+ (n = 8)
( noitarefilorPcp
mx
10–3
)( noitarefilorPcp
mx
10–3
)
05
101520253035404550
Control APC Fixed APCNormal
Pulsed APC (h) Pulsed APC (h) Pulsed APC (h) Pulsed APC (h)
Pulsed APC (h) Pulsed APC (h) Pulsed APC (h) Pulsed APC (h)
Vehicle Dapsone Nitroso dapsone
Prol
ifera
tion
(cpm
x 10
–3)
Prol
ifera
tion
(cpm
x 10
–3)
Clone 1 Clone 2 Clone 3 Clone 4
Clone 1 Clone 2 Clone 3 Clone 4
Clone 1 Clone 2 Clone 3 Clone 4
Clone 1 Clone 2 Clone 3 Clone 4
12 | ZHAO et Al.
activation.Autologousantigen‐presentingcellsandantigenpresent‐ingfrom1additionalpatientswereusedascomparators.
Ashas beendescribedpreviously for other drugs,25,26 30%ofdapsoneandnitrosodapsone‐responsiveHLAclassIrestrictedCD8clonesdisplayedproliferativeresponseswiththedrugormetaboliteandantigen‐presentingcellsexpressingawidevarietyofHLAalleles(resultsnot shown).Othercloneswereactivatedwith thedrugormetabolite only in the presence of autologous antigen‐presentingcells.Despitethis,30%‐40%ofCD8+clonesdisplayedadegreeofHLA‐B*13:01allelerestriction.Figure7A,Bshows3clones(1dap‐sone‐ and 2 nitroso dapsone‐responsive) that were stimulated toproliferate exclusively in thepresenceof dapsoneor nitrosodap‐sone and antigen‐presenting cells expressingHLA‐B*13:01. Threeadditional clones (1 dapsone‐ and 2 nitroso dapsone‐responsive)areshownthatwereactivatedinthepresenceofantigen‐present‐ing cells expressingHLA‐B*13:01 andeitherB*13:02,B*58:01, orB*51:01.
The2dapsone‐responsiveCD8+cloneswereexpandedinsuffi‐cientnumbersforustoconductaproliferationassayusingantigen‐presentingcellsfromthe6hypersensitivepatientsthatallexpress
HLA‐B*13:01.Thecloneswereactivatedinthepresenceofdapsoneandantigen‐presentingcellsfromallofthepatients(Figure7C).
4 | DISCUSSION
Knowledgeof theroledrugmetabolismplays inthegenerationofantigenicdeterminantsthatactivateTcellsislimited.Circumstantialevidencesupportingarolefordrugmetabolismincludestheidenti‐ficationofprotein‐reactivemetabolites formanydrugsassociatedwithahigh incidenceofhypersensitivityandthe inductionoftox‐icity intargettissuebyreactivemetabolitesandhencethepoten‐tial to disrupt immune regulatory pathways through the provisionofdangersignals.27However,themostdirectevidencelinkingdrugmetabolismtothedevelopmentofhypersensitivityisthedetectionofhalothane‐specificantibodies inpatientswith liverfailure28 and the observation that halothane protein adducts activate T cells.29 Furthermore,halothanederivativesthatformlower levelsofreac‐tivemetaboliteareassociatedwithadecreasedriskofliverinjury.30 To explorewhethermetabolites activate hypersensitive patient T
F I G U R E 7 DapsoneandnitrosodapsonebindwithadegreeofselectivelytoHLA‐B*13:01toactivatecertainCD8+T‐cellclones.A,Nitrosodapsone‐and(B)dapsone‐responsiveCD8+cloneswereculturedwithdrugordrugmetaboliteandirradiatedEBV‐transformedBcells(APC,antigen‐presentingcells)from10donorsexpressingdifferentHLA‐Ballelesintriplicateculturesfor48h.ThecompleteHLAtypeofthedifferentdonorsisshowninTableS2.C,Dapsone‐responsiveCD8+cloneswereculturedwithdapsoneandirradiatedEBV‐transformedBcellsfrom6hypersensitivepatientsexpressingHLA‐B*13:01intriplicateculturesfor48h.T‐cellproliferativeresponseswereassessedthroughtheadditionof[3H]‐thymidine
0
5
10
15
20
25
30
35
No
APC
patie
nt A
PCB*
13:0
1
B*08
:01
B*13
:02
B*15
:01
B*35
:01
B*44
:02
B*46
:01
B*51
:01
B*58
:01
0123456789
10
No
APC
Patie
nt A
PCB*
13:0
1
B*08
:01
B*13
:02
B*15
:01
B*35
:01
B*44
:02
B*46
:01
B*51
:01
B*58
:01
0
5
10
15
20
25
30
No
APC
Patie
nt A
PCB*
13:0
1
B*08
:01
B*13
:02
B*15
:01
B*35
:01
B*35
:01
B*35
:01
B*35
:01
B*35
:01
02468
1012141618
No
APC
Patie
nt A
PCB*
13:0
1
B*08
:01
B*13
:02
B*15
:01
B*35
:01
B*44
:02
B*46
:01
B*51
:01
B*58
:01
0
2
4
6
8
10
12
14
No
APC
Patie
nt A
PCB*
13:0
1
B*08
:01
B*13
:02
B*15
:01
B*35
:01
B*44
:02
B*46
:01
B*51
:01
B*58
:01
0
5
10
15
20
25
30
No
APC
Patie
nt A
PCB*
13:0
1
B*08
:01
B*13
:02
B*15
:01
B*35
:01
B*44
:02
B*46
:01
B*51
:01
B*58
:01
0
10
20
30
40
50
60
70
0
5
10
15
20
25
Nitroso dapsone clonesVehicle Dapsone Nitroso dapsone
( noitarefilorPcp
mx
10–3
)
(A)
(C)(B)
Dapsone clones
Clone 2 Clone 3 Clone 1 Clone 4
Clone 2 Clone 1 Clone 1 Clone 2
| 13ZHAO et Al.
cellsinvitro, co‐culturestrategieshavebeendevelopedusingmicro‐somal metabolite generating systems.31,32 Enhanced drug‐specificT‐cellresponseshavebeenreportedwithdrugsinthepresenceofametabolizingsystem;however,itisverydifficulttodelineatethedisparateeffectsoftheparentdrugandmetabolites.Forthisrea‐son,wesynthesizedthereactivemetaboliteofsulfamethoxazole,33 whichcausesimmune‐mediatedskinandliverreactionsthatarenotlinked to expression of a specific HLA allele.34,35Working along‐sideotherresearchers inthefieldwehaveshownthatPBMCandinflamedtissuefromallhypersensitivepatientscontainTcellsthatareactivatedwithsulfamethoxazoleand/or itsreactivemetaboliteviadifferentpathways(ie,directHLAbindingandahaptenpathway,respectively).12,13
Inrecentyears,researchersstudyinghypersensitivityreactionsstronglylinkedtoexpressionofspecificHLAalleleshavefocussedexclusively on the interaction between parent drug and the HLAmolecule.Thisisbecausetheparentdrugis,forthemostpart,theonlyreagentavailableforinvitrocellculturestudies.Theobserva‐tion thatTcells frompatientswithhypersensitivity todrugs suchascarbamazepineareactivatedwiththeparentdrug11hasleadre‐searcherstohypothesizethatreactivemetabolitesarenotinvolvedinthedevelopmentofHLAallele‐restrictedformsofdrughypersen‐sitivity.Toinvestigatethishypothesis,wenowfocusedondapsonehypersensitivityinpatientswithleprosy.Dapsoneisamodelstudydrug as hypersensitivity reactions in patients are strongly linkedto theexpressionofHLA‐B*13:01,19 the reactivenitrosometabo‐liteofdapsonehasbeensynthesizedinastableform,15 and patient samplesareavailableformechanisticstudies.UtilizingPBMCfromHLA‐B*13:01+patients,weshowCD4+andCD8+T‐cellresponsestodapsoneandthenitrosometabolite.Furthermore,dapsoneandnitrosodapsonebindtoHLA‐B*13:01toselectivelyactivatecertainCD8+clones.
Six HLA‐B*13:01+ patients that developed hypersensitivitywhile receiving dapsone in combination with rifampicin and clo‐fazimine were recruited to the study. Three displayed a positivepatch test response to dapsone. PBMC from all six patientswerestimulatedtoproliferateand/orsecreteIFN‐γafterstimulationwithdapsoneorthenitrosometabolite.Thestrongestinvitroresponseswere seen in the patients with the positive patch test response,andthesepatientswereselectedforserialdilutionexperimentstosearchfordrug‐anddrugmetabolite‐responsiveT‐cellclones.CD4+ andCD8+ subpopulationswerepurifiedbeforeplating single cellsin culture plates for the cloning procedure.Dapsone‐ and nitrosodapsone‐responsiveTcellsweregeneratedinalmostequalnumbers(dapsone395;nitrosodapsone399).Theratioofdapsone‐ornitrosodapsone‐responsiveCD4+andCD8+T‐cellcloneswas4‐5:1 inpa‐tients6and8and2:1 inpatient5,whichdemonstrates thatdrugexposureconsistentlyresults inCD4+andCD8+responsesagainstthedrugantigens.Thisisinstarkcontrasttothefindingswithabaca‐virwhereCD8+Tcellsaretheonlycellsactivatedwiththedrug.36,37 TheactivationofCD4+andCD8+clonesthatexpressedanarrayofTCRVβreceptors,withdapsoneornitrosodapsone,wasHLAclassIIandclassIrestricted,respectively.Thisconfirmsourpreviousdata
withPBMCfromhealthydonors,15whichshowsthatbothdapsoneandnitrosodapsoneinteractwithHLAclassIandclassIImolecules.
TheavailabilityofclonedCD4+andCD8+Tcellspermittedthedetailedanalysisofcross‐reactivitybetweendapsone,nitrosodap‐soneandstructurallyrelatedcompounds,mechanismsofdapsone‐andnitrosodapsone‐specificT‐cellactivation,andtheinvolvementofthealleleHLA‐B*13:01inantigenpresentation.Cross‐reactivitystudiesrevealedthatthepositionoftheaminegroupsonthearo‐maticringsofdapsoneandtheavailabilityofthesulfonegroupwascriticalfortheactivationoftheT‐cellclones.Furthermore,asapanelofsulfonamidesthatcontainonearomaticaminegroupdidnotac‐tivatetheclones,bothofdapsones’aromaticamineswererequiredforT‐cellactivation.
25%‐50%of CD4+ andCD8+ cloneswere classified as highlyspecificastheywereonlyactivatedwithonecompound(eithertheparentdrugormetabolite [Figure3B]).This confirms thatTcellsrecognizeandrespondselectivelytothetwodifferentformsofthedapsoneantigen.Dapsonehasanunusualstructureinthatthetwoaromatic amines connected to the sulfone group are identical. Ifthestructureofanitrosodapsone‐modifiedHLAbindingpeptideiscomparedwithdapsonecomplexedtothesamepeptide,itislikelythatsimilarconformationswillbeobserved,andassuch,theinter‐actionwiththeT‐cellreceptorwillbethesame.Thus,ourworkinghypothesistoexplainthedrug‐ordrugmetabolite‐specificactiva‐tionofcertainclonesisthattheHLAbindingpeptidesalsopartici‐pateintheTCRinteractionandimpartadegreeofselectivity.
When using optimum concentrations of dapsone(100‐500µmol/L) and nitroso dapsone (5‐20µmol/L) to activateCD4+ andCD8+T cells, clonesdisplayinghigh (ie, at least50%ofthe response detectedwith the opposite compound) and low (ie,10%‐50%of the responsedetectedwith theopposite compound)levelsofcross‐reactivitywerealsodetected.Themajorityofdap‐sone‐responsive, cross‐reactive CD4+ and CD8+ clones displayedlow levels of cross‐reactivity with nitroso dapsone. Using quanti‐tative mass spectrometry, we were able to demonstrate that thenitrosometaboliteisreducedtodapsoneinthe2‐dayproliferationassay.Thus,clonesincubatedwith30µmol/Lnitrosodapsonewereexposed to 2.5‐15µmol/L of the parent drug, which stimulates asuboptimalT‐cellproliferativeresponse.Incontrast,themajorityofcross‐reactive nitroso dapsone‐responsiveCD4+ andCD8+ clonesdisplayedhighlevelsofcross‐reactivitywith100‐500µmol/Ldap‐sone.Toinvestigatetheresponseprofilesfurther,apanelofcross‐reactive dapsone‐ and nitroso dapsone‐responsive clones wereculturedwithequalconcentrationsoftheparentdrugandmetabo‐lite(0.1‐100µmol/L).Interestingly,weaklyandstronglycross‐reac‐tivecloneswereactivatedwiththesameconcentrationsofdapsoneandnitrosodapsonewithsimilardose‐responsecurvesuntilnitrosometabolite‐inducedtoxicitywasdetected.Thesedataareincontrasttoourearlierstudieswithsulfamethoxazoleanditsnitrosometab‐olite,wheretheparentdrugactivatesTcellsatsignificantlyhigherconcentrations.13ItisreasonabletoassumethatHLAbindingpep‐tides play a less important role in the triggering of cross‐reactiveTCR.
14 | ZHAO et Al.
Giventhatclonesdisplayingreactivitytowarddapsoneand/orthenitrosometaboliteweredetected,itwasimportanttodefinemechanisms of antigen presentation including the requirementforantigenprocessing.Antigen‐presentingcellswererequiredforthestrongactivationofCD4+andCD8+cloneswithdapsoneandthenitrosometabolite.Cloneswereactivatedwithdapsoneinthepresenceofirradiatedandglutaraldehyde‐fixedantigen‐present‐ingcells.Incontrast,fixationreducedthestrengthofthenitrosodapsone‐specificT‐cellproliferativeresponse.ThissuggeststhatdapsoneactivatesT cells throughadirect interactionwithHLAmolecules,whereasnitrosodapsonebinds covalently to cellularproteinformingadductsthatrequireprocessingtogenerateanti‐genicpeptides.Asdiscussedabove,thebestexplanationfortheweakresponseofnitrosodapsone‐responsivecloneswithsolublenitrosodapsoneandfixedantigen‐presentingcells is theforma‐tionof dapsone in theproliferation assay. Thedrug‐specific ac‐tivation of clones fromdapsone hypersensitive patients by twopathways has important implications for the future structuralanalysisofhowdapsoneandnitrosodapsoneinteractswithHLAmolecules. Application of drug‐ or drugmetabolite‐pulsed anti‐gen‐presentingcells insteadofsolubledrug inT‐cellassayspro‐videsatooltodifferentiatebetweentheeffectsofcovalentandnon‐covalentformsofdrug.14,39Dapsone‐responsivecloneswerenot activated with dapsone‐pulsed antigen‐presenting cells. Incontrast,6outofthe8nitrosodapsone‐responsiveclonestestedwere activated with nitroso dapsone‐pulsed antigen‐presentingcellsand thestrengthof the inducedresponsewas thesameasthat seenwith theparent drug. Twoof thenitrosodapsone‐re‐sponsivecloneswerenotactivatedwithpulsedantigen‐present‐ingcells;interestingly,bothoftheseclonescross‐reactedstronglywithdapsone.
Glutathioneisatripeptideintracellularantioxidantthatprotectscells fromexposuretoaromaticnitrosocompoundsbyactingasareducingagentandthroughdirectconjugation.23,24TheadditionofglutathionetoT‐cellassayspreventsthecovalentbindingofnitrosocompounds,andhence,itispossibletoexplorewhetherTcellsareactivatedwithdrugmetabolite‐modifiedproteinadducts.40Nitrosodapsoneglutathioneadductswereformedrapidly inculturescon‐tainingthedrugmetabolite,Tcells,antigen‐presentingcells,andglu‐tathione.Moreover, glutathionedecreased the responseof clonestosolublenitrosodapsoneandblockedtheresponsetonitrosodap‐sone‐pulsedantigen‐presentingcells.Thus,theformationofacova‐lentadductbetweennitrosodapsoneandcellularproteinsisneededfortheactivationofcertainclones.
In the next section of the project, antigen‐presenting cellswere generated from healthy donors expressing HLA‐B alleleswithatleast90%sequencehomologytoHLA‐B*13:01(includinganadditionalHLA‐B*13:01+donor)toinvestigatewhethereitherdapsoneornitrosodapsone interactwithadegreeofselectivitytotheriskalleletoactivateCD8+clones.Asdescribedpreviouslywithotherformsofdrughypersensitivity,severalclonesdisplayedHLAallele‐unrestricteddrugrecognition(ie,dapsoneandnitrosodapsone stimulated a proliferative response in the presence of
multiple antigen‐presenting cells expressing differentHLA‐B al‐leles).25Other cloneswereonly activated in thepresenceofni‐trosodapsoneandautologousantigen‐presentingcells.However,a panel of clones were selectively activated with dapsone ornitroso dapsone in the presence of antigen‐presenting cells dis‐playingHLA‐B*13:01. Cross‐reactivitywas observed for severaloftheseclones;however,nopredominantTCRVβreceptorswereexpressed.This indicatesthatboththeparentdrugandreactivemetabolite interactswithHLA‐B*13:01 toactivatecertainCD8+ clones.Wearethereforesynthesizingdesignernitrosodapsone‐modified HLA‐B*13:01 binding peptides and conducting struc‐turalanalysestocomparethebindinginteractionofdapsoneandnitrosodapsonewithHLA‐B*13:01.
Different forms of drug hypersensitivity reaction are asso‐ciatedwith the secretionof a diversepanel of cytokines fromTcells.41Thus,ELIspotwasusedtoprofilecytokinesecretionfromdapsone‐andnitrosodapsone‐stimulatedclones.CD4+andCD8+ clones secreted the same panel of cytokines when stimulatedwithdapsoneornitrosodapsone,butexpresseddistinctchemo‐kine receptors. Clones secreted Th1, Th2, and Th22 cytokinesalongsidethecytolyticmoleculesperforin,granzymeB,andFasL;however,IL‐17wasnotdetected.IL‐22secretionintheabsenceofIL‐17seemstobeacommonfeatureofdrug‐specificclonesasthisprofile has now been detectedwith dapsone, sulfamethoxazole,and piperacillin.42,43 It is possible that this profile relates to thelong‐term in vitro culture of clones; thus, a future study explor‐ingthecytokineprofileinaffectedtissueswouldbeappropriate.The chemokine receptors displayedondapsone‐ or nitrosodap‐sone‐responsiveCD4+cloneswererestrictedtoCXCR3andCCR4.In contrast, CD8+ clones also expressed CCR6, 9, and 10. CCR4andCCR10havebeen implicated in themigrationofT cells intoskin44;thus,amoredetailedinvestigationofmigratorypropertiesofCD4+andCD8+Tcellsinpatientswithdapsonehypersensitivityiswarranted.
Inconclusion,ourstudyshowsthatdapsone‐andnitrosodap‐sone‐responsiveCD4+andCD8+Tcellscirculateinhypersensitivepatients.TcellswereactivatedselectivelywiththeparentdruganddrugmetaboliteviadirectHLAbindingandahaptenmechanism,re‐spectively.ThedetectionofHLA‐B*13:01‐restricteddapsone‐andnitrosodapsone‐responsiveCD8+clonesindicatesthatdapsonehy‐persensitivityshouldbeusedasanexemplartoexplorethestruc‐turalfeaturesofdrugHLAbindingandhowthisinteractionresultsinapathogenicT‐cellresponse.Forpatientswithsuspectedhyper‐sensitivity,itiscriticaltoidentifytheagentthatcausedthereaction.We are currently exploringwhether drug‐peptide antigens repre‐senteffectivereagentsforT‐cellactivation.Abetterunderstandingofdrugantigenicitywill leadtothedevelopmentof improvedevi‐dence‐baseddiagnostictools.
ACKNOWLEDG MENTS
The authors would like to thank the patients and volunteers foragreeingtodonatebloodandtissuesamples.
| 15ZHAO et Al.
CONFLIC TS OF INTERE S T
Theauthorsdeclarethattheyhavenoconflictsofinterest.
AUTHOR CONTRIBUTIONS
QZ,KA,AA,MA,JA,AG,andMOOconductedthebiologicalex‐periments. XM conducted themass spectrometric analyses.HL,LD, and HZ collected the clinical samples and conducted patchtesting.DJN, BKP,DAO, JL, and FZ designed the study and su‐pervisedtheproject.AG,MOO,XM,andDJNanalyzedthedataand drafted the manuscript. All authors critically reviewed themanuscript.
ORCID
Dean J. Naisbitt https://orcid.org/0000‐0003‐4107‐7832
R E FE R E N C E S
1. RedwoodAJ,PavlosRK,WhiteKD,PhillipsEJ.HLAs:keyregulatorsofT‐cell‐mediateddrughypersensitivity.HLA.2018;91:3‐16.
2. TehLK,SelvarajM,BannurZ,etal.Couplinggenotypingandcom‐putationalmodelinginpredictionofanti‐epilepticdrugsthatcausestevensjohnsonsyndromeandtoxicepidermalnecrolysisforcar‐rierofHLA‐B*15:02.J Pharm Pharm Sci.2016;19:147‐160.
3. Van Den Driessche G, Fourches D. Adverse drug reactions trig‐geredby the commonHLA‐B*57:01variant: amolecular dockingstudy.J Cheminform.2017;9:13.
4. MallalS,PhillipsE,CarosiG,etal.HLA‐B*5701screeningforhyper‐sensitivitytoabacavir.N Engl J Med.2008;358:568‐579.
5. DalyAK,DonaldsonPT,BhatnagarP,etal.HLA‐B*5701genotypeisamajordeterminantofdrug‐inducedliverinjuryduetoflucloxa‐cillin. Nat Genet.2009;41:816‐819.
6. ChungW‐H,HungS‐I,HongH‐S,etal.Medicalgenetics:amarkerforStevens‐Johnsonsyndrome.Nature. 2004;428:486.
7. MonshiMM,FaulknerL,GibsonA,etal.Humanleukocyteantigen(HLA)‐B*57:01‐restricted activation of drug‐specific T cells pro‐videstheimmunologicalbasisforflucloxacillin‐inducedliverinjury.Hepatology.2013;57:727‐739.
8. YunJ,MarcaidaMJ,ErikssonKK,etal.Oxypurinoldirectlyandim‐mediatelyactivatesthedrug‐specificTcellsviathepreferentialuseofHLA‐B*58:01.J Immunol.2014;192:2984‐2993.
9. Illing PT, Vivian JP, Dudek NL, et al. Immune self‐reactivitytriggered by drug‐modified HLA‐peptide repertoire. Nature. 2012;486:554‐558.
10. Ostrov DA, Grant BJ, Pompeu YA, et al. Drug hypersensitivitycausedbyalterationoftheMHC‐presentedself‐peptiderepertoire.Proc Natl Acad Sci USA.2012;109:9959‐9964.
11. KoT‐M,ChungW‐H,WeiC‐Y, et al. Shared and restrictedT‐cellreceptoruseiscrucialforcarbamazepine‐inducedStevens‐Johnsonsyndrome.J Allergy Clin Immunol.2011;128:1266‐1276.
12. NassifA,BensussanA,BoumsellL,etal.Toxicepidermalnecroly‐sis:effectorcells aredrug‐specific cytotoxicTcells. J Allergy Clin Immunol.2004;114:1209‐1215.
13. CastrejonJL,BerryN,El‐GhaieshS,etal.StimulationofhumanTcellswithsulfonamidesandsulfonamidemetabolites.J Allergy Clin Immunol.2010;125:411‐418.
14. SchnyderB,BurkhartC,Schnyder‐FrutigK,etal.Recognitionofsul‐famethoxazoleanditsreactivemetabolitesbydrug‐specificCD4+Tcellsfromallergicindividuals.J Immunol.2000;164:6647‐6654.
15. AlzahraniA,OgeseM,MengX,etal.DapsoneandnitrosodapsoneactivationofnaiveT‐cells fromhealthydonors.Chem Res Toxicol. 2017;30:2174‐2186.
16. Roychowdhury S, Vyas PM, SvenssonCK. Formation and uptakeofarylhydroxylamine‐haptenatedproteinsinhumandendriticcells.Drug Metab Dispos.2007;35:676‐681.
17. VyasPM,RoychowdhuryS,KhanFD,etal.Enzyme‐mediatedpro‐teinhaptenationofdapsoneandsulfamethoxazoleinhumankerat‐inocytes‐1.ExpressionandroleofcytochromesP450.J Pharmacol Exp Ther.2006;319:488‐496.
18. RaoPN,LakshmiTS. Increase inthe incidenceofdapsonehyper‐sensitivitysyndrome–anappraisal.Lepr Rev.2001;72:57‐62.
19. ZhangFR,LiuH, IrwantoA,etal.HLA‐B*13:01andthedapsonehypersensitivitysyndrome.N Engl J Med.2013;369:1620‐1628.
20. TemparkT,SatapornpongP,RerknimitrP,etal.Dapsone‐inducedsevere cutaneous adverse drug reactions are strongly linkedwith HLA‐B*13: 01 allele in the Thai population. Pharmacogenet Genomics.2017;27:429‐437.
21. WatanabeH,WatanabeY,TashiroY,etal.Adockingmodelofdap‐soneboundtoHLA‐B*13:01explainstheriskofdapsonehypersen‐sitivitysyndrome.J Dermatol Sci.2017;88:320‐329.
22. ChenW‐T,WangC‐W,LuC‐W,etal.ThefunctionofHLA‐B*13:01involvedinthepathomechanismofdapsone‐inducedseverecuta‐neousadversereactions.J Invest Dermatol.2018;138:1546‐1554.
23. NaisbittDJ,O'Neill PM,PirmohamedM,KevinParkB. Synthesisandreactionsofnitrososulphamethoxazolewithbiologicalnucleo‐philes:implicationsforimmunemediatedtoxicity.Bioorg Med Chem Lett.1996;6:1511‐1516.
24. EllisMK,HillS,FosterPM.Reactionsofnitrosonitrobenzeneswithbiologicalthiols:identificationandreactivityofglutathion‐S‐ylcon‐jugates.Chem Biol Interact.1992;82:151‐163.
25. vonGreyerz S, BultemannG, SchnyderK, et al.Degeneracy andadditionalalloreactivityofdrug‐specifichumanalphabeta(+)Tcellclones.Int Immunol.2001;13:877‐885.
26. Zanni MP, von Greyerz S, Schnyder B,Wendland T, PichlerWJ.Allele‐unrestrictedpresentationoflidocainebyHLA‐DRmoleculestospecificàá+Tcellclones.Int Immunol.1998;10:507‐515.
27. UetrechtJ,NaisbittDJ. Idiosyncraticadversedrugreactions:cur‐rentconcepts.Pharmacol Rev.2013;65:779‐808.
28. KennaJG,KnightTL,vanPeltFN. Immunitytohalothanemetab‐olite‐modified proteins in halothane hepatitis. Ann N Y Acad Sci. 1993;685:646‐661.
29. You Q, Cheng L, Ju C. Generation of T cell responses target‐ing the reactive metabolite of halothane in mice. Toxicol Lett. 2010;194:79‐85.
30. ParkBK,PirmohamedM,KitteringhamNR.Roleofdrugdispositionindrughypersensitivity:achemical,molecular,andclinicalperspec‐tive. Chem Res Toxicol.1998;11:969‐988.
31. BergströmMA,Ott H, Carlsson A, et al. A skin‐like cytochromeP450cocktailactivatesprohaptenstocontactallergenicmetabo‐lites.J Invest Dermatol.2007;127:1145‐1153.
32. SachsB,ErdmannS,Al‐MasaoudiT,MerkHF.Invitrodrugallergydetectionsystemincorporatinghumanlivermicrosomesinchloraz‐epate‐inducedskinrash:drug‐specificproliferationassociatedwithinterleukin‐5secretion.Br J Dermatol.2001;144:316‐320.
33. NaisbittDJ,O'Neill PM, PirmohamedM, Park BK. Synthesis andreactionsofnitrososulfamethoxazolewithbiologicalnucleophiles‐implicationsforimmune‐mediatedtoxicity.Bioorg Med Chem Lett. 1996;6:1511‐1516.
34. AlfirevicA,VilarFJ,AlsbouM,etal.TNF,LTA,HSPA1LandHLA‐DRgenepolymorphismsinHIV‐positivepatientswithhypersensitivitytocotrimoxazole.Pharmacogenomics.2009;10:531‐540.
35. LonjouC,BorotN,SekulaP,etal.AEuropeanstudyofHLA‐BinStevens‐Johnsonsyndromeandtoxicepidermalnecrolysisrelatedtofivehigh‐riskdrugs.Pharmacogenet Genomics.2008;18:99‐107.
16 | ZHAO et Al.
36. BellCC,FaulknerL,MartinssonK,etal.T‐cellsfromHLA‐B*57:01+humansubjectsareactivatedwithabacavirthroughtwoindepen‐dentpathwaysandinducecelldeathbymultiplemechanisms.Chem Res Toxicol.2013;26:759‐766.
37. AdamJ,ErikssonKK,SchnyderB,FontanaS,PichlerWJ,YerlyD.AviditydeterminesT‐cellreactivityinabacavirhypersensitivity.Eur J Immunol.2012;42:1706‐1716.
38. LucasA,LucasM,StrhynA,etal.Abacavir‐reactivememoryTcellsarepresentindrugnaiveindividuals.PLoS One.2015;10:e0117160.
39. CastrejonJL,LavergneSN,El‐SheikhA,etal.Metabolicandchem‐icaloriginsofcross‐reactiveimmunologicalreactionstoarylaminebenzenesulfonamides: T‐cell responses to hydroxylamine and ni‐trosoderivatives.Chem Res Toxicol.2010;23:184‐192.
40. BurkhartC,vonGreyerzS,DeptaJ,etal.Influenceofreducedglu‐tathioneontheproliferativeresponseofsulfamethoxazole‐specificandsulfamethoxazole‐metabolite‐specifichumanCD4+T‐cells.Br J Pharmacol.2001;132:623‐630.
41. Pichler WJ. Delayed drug hypersensitivity reactions. Ann Intern Med.2003;139:683‐693.
42. Sullivan A,Wang E, Farrell J, et al. beta‐Lactam hypersensitivityinvolves expansion of circulating and skin‐resident TH22 cells. J Allergy Clin Immunol.2018;141:235‐249.
43. Gibson A, Ogese M, Sullivan A, et al. Negative regulation byPD‐L1 during drug‐specific priming of IL‐22‐secreting T cells
and the influenceofPD‐1oneffectorT cell function. J Immunol. 2014;192:2611‐2621.
44. XiaM,Hu S, FuY, et al. CCR45 regulates balancedmaintenanceand function of resident regulatory and effector T cells to pro‐mote immune homeostasis in the skin. J Allergy Clin Immunol. 2014;134:634‐644.
SUPPORTING INFORMATION
Additional supporting information may be found online in theSupportingInformationsectionattheendofthearticle.
How to cite this article:ZhaoQ,AlhilaliK,AlzahraniA,etal.Dapsone‐andnitrosodapsone‐specificactivationofTcellsfromhypersensitivepatientsexpressingtheriskalleleHLA‐B*13:01.Allergy. 2019;00:1–16. https://doi.org/10.1111/all.13769