Reference intervals for orotic acid in urine, plasma and dried blood spot using hydrophilic interaction liquid chromatography-tandem mass spectrometry.

JournalofChromatographyB,xxx (2011) xxx–xxx

ContentslistsavailableatSciVerseScienceDirect

Journal

of

Chromatography

B

jo u r n al h om ep a g e :w w w . e l s e v i e r . c o m / l o c a t e / c h r o m b

Reference

intervals

for

orotic

acid

in

urine,

plasma

and

dried

blood

spot

using

hydrophilic

interaction

liquid

chromatography–tandem

mass

spectrometry

夽

Oceania

D’Apolito

a,1

_,

_Daniela

_Garofalo

a,1

_,

_Giancarlo

_la

_Marca

b,c

_,

_Antonio

_Dello

_Russo

d

_,

Gaetano

Corso

a,d,∗

a_{DepartmentofBiomedicalSciences,UniversityofFoggia,Foggia,Italy} b_{DepartmentofPharmacology,UniversityofFlorence,Italy} c_{MassSpectrometryLab,MeyerChildren’sHospital,Florence,Italy}

d_{DepartmentofBiochemistryandMedicalBiotechnologies,UniversityFedericoIIofNaples,Naples,Italy}

a

r

t

i

c

l

e

i

n

f

o

Articlehistory: Received19July2011 Accepted27September2011 Available online xxx Keywords: Referenceintervals Oroticacid Driedbloodspot Tandemmassspectrometry

a

b

s

t

r

a

c

t

Oroticacid(OA),amarkerofhereditaryoroticaciduria,isusuallyusedforthedifferentialdiagnosisof somehyperammonemicinheriteddefectsofureacycleandofbasicaminoacidtransporters.Thisstudy wasaimedtoestablishagerelatedreferenceintervalsofOAinurine,andforthefirsttimeinplasma, anddriedbloodspot(DBS)from229apparentlyhealthysubjectsagedfromthreedaysto40years. ThequantificationofOAwasperformedbyapreviouslyimplementedmethod,usingastableisotope dilutionwith1,3-[15_N

2]-oroticacidandhydrophilicinteractionliquidchromatography–tandemmass

spectrometry(HILIC–MS/MS).Themethodhasprovedtobesensitiveandaccurateforaquantitative analysisofOAalsoinDBSandplasma.Accordingtopreviousstudies,urinaryOAlevels(mmol/molof creatinine)decreasesignificantlywithage.Theupperlimits(as99th%ile)wereof3.44and1.30ingroups agedfromthreedaysto1year(group1)andfrom1yearto12years(group2),respectively;inteenagers (from13to19years;group3)andadults(from20to40years;group4)urinarylevelsbecamemore stableandtheupperlimitswereof0.64and1.21,respectively.Furthermore,OAlevelsinDBS(␮M)also resultedsignificantlyhigherinsubjectsofgroup1(upperlimitof0.89)thaninsubjectsofgroups2,3and 4(upperlimitsof0.24,0.21,and0.29,respectively).OAlevelsinplasma(␮M)weresignificantlylower insubjectsofgroup3(upperlimitof0.30)thaninsubjectsofgroups1,2,and4(upperlimitsof0.59, 0.48,and0.77,respectively).ThismethodwasalsoemployedforOAquantificationinplasmaandDBS of17newbornsaffectedbyureacycledefects,resultingsensitiveandspecificenoughtoscreenthese disorders.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Orotic acid (OA; 1,2,3,6-tetrahydro-2,6-dioxo-4-pyrimidinecarboxylicacid;uracil-6-carboxylicacid)isametabolic markerof thehereditaryoroticaciduria,apyrimidinesynthesis defectcausedbyuridine-5-monophosphate(UMP)synthase defi-ciency[1](OMIM258900and258920),butthemostvaluableuse ofOAdeterminationisforthedifferentialdiagnosisofornithine transcarbamoylase (OTC, OMIM 311250) deficiency, the most commonformofaureacycledefect.Increasedexcretionofurinary

夽 Thispaperispartofthespecialissue“LC–MS/MSinClinicalChemistry”,Michael VogeserandChristophSeger(GuestEditors).

∗ Correspondingauthorat:DipartimentodiScienzeBiomediche,Facoltàdi Medic-inaeChirurgia–UniversitàdiFoggia,VialeL.Pinto1,71122Foggia,Italy. Tel.:+390881588055;fax:+390881588037.

E-mailaddress:g.corso@unifg.it(G.Corso). 1 _{Theseauthorscontributedequallytothiswork.}

OAisalsoobservedintheotherinbornerrorsoftheureacycle,as argininosuccinatesynthetase deficiency(ASS, citrullinemiatype I,OMIM215700),argininosuccinatelyasedeficiency(ASL, argini-nosuccinicaciduria,OMIM207900),andarginasedeficiency(ARG, argininemia,OMIM207800).OAinurinealsoincreasesindefects of genes encoding somecationic amino acidtransporters such asthemitochondrialornithinetransporter1deficiency(ORNT1; HHHsyndrome,OMIM238970),andthelysinetransportery+

LAT-1 (SLC7A7; LPI, lysinuric protein intolerance, OMIM 222700), whichimpairtheintestinalabsorptionandrenalre-absorptionof basicaminoacids,respectively. Intheseconditionsthereduced bloodlevelsofornithineand/orofarginineslowdowntherateof ureacycle[2,3].OAdetermination ishelpfulforthedifferential diagnosisofhyperammonemiadisorderswhichcannotbereadily diagnosedbyaminoacidchromatography,thusreducingtheneed forenzymedeterminationintissuebiopsies.

OAexcretionincreasesalsoasaresultofdrugs,suchas allop-urinoland6-azauridine,thatappearstoinhibitthedistalpartof pyrimidinesynthesispathway.OnceOA reachestheblood, itis 1570-0232/$–seefrontmatter © 2011 Elsevier B.V. All rights reserved.

2 O.D’Apolitoetal./J.Chromatogr.Bxxx (2011) xxx–xxx

efficientlyclearedbythekidney;there isevidenceforsecretion bythetubules,aswellaslossbyfiltration[4].OAmayevenbe suc-cessfullyanalyzedinpiecesofurine-impregnatedfilterpaper[5,6] andmailedtoacentrallaboratory;today,avarietyofanalytical techniquesprovidesatisfactoryresults[2].AnalysisofurinaryOA providesanintegratedpictureofOAproductionovertime,rather thanthesnapshotthatisprovidedbyanalysisofasingleplasma sample[7].

AlthoughtheOAinurinerepresentsworldwidethereference testfordifferentialdiagnosisandforthemonitoringofdisorders abovementioned,patientsaffectedbyureacycledefectsalsoshow elevatedOAlevelsinplasmaorblood[3,4,8],andonlyfewaffected patientswiththesedisordershavebeencarefullyinvestigated ana-lyzingOAinblood orplasma.IncreasedconcentrationofOA in plasmaandbloodhasbeenfoundinASL[4],andOTCdefects[9].

We recently implemented a fast and sensitive hydrophilic interactionLC(HILIC)–MS/MSmethodforthequantitative determi-nationofOAinurine,plasma,anddriedbloodspot(DBS)requiring onlyminimalsamplepreparationandprovidingashortruntime. Usingthismethod,OAlevelsinplasmaandDBSfromapatient affectedbycitrullinemiawereatleastfiftytimeshigherthan nor-malcontrols[8].Theseraredatamakeitdifficulttodefinesafety limitsofOAinbloodanditsusefulnessasabiomarkerofdisease; moreover,referenceintervalsinbloodofhealthyhumansare com-pletelylacking.

Today,tandem MS representsa powerful tool for quantita-tiveanalysisofmetabolitesinbiologicalfluids,and,inparticular, themetabolicanalysisinDBSbyMS/MSrepresentsthereference methodfortheexpandednewbornscreeningofinbornerrorsof metabolism;therefore,theanalysisofOAinDBSmightrepresent ausefultestforthediagnosisofclassicaloroticacidurias,orasa secondtiertesttoconfirmureacycledefects.

Thepurposeofthisstudywastodeterminereferenceintervals forOAconcentrationsinurine,plasma,andDBSfromapparently healthychildrenandadults,dividedintofourgroupsagedfrom threedaysto40years,andtoassesstheeffectofage.Inaddition, wereporttheresultsofOAinDBSandplasmafromsomepatients affectedbydifferentinbornerrorsofureacycle.

2. Materialsandmethods

2.1. Chemicals

Theanalytical solvents of HPLCgrade ACN, and ammonium acetatewereobtainedfromJ.T.Baker(Deventer,TheNetherlands). OA standard was purchased from Sigma–Aldrich (Steinheim, Germany),and[1,3-15_N

2]-OA,usedasinternalstandard(IS),was

purchased from Cambridge Isotope Laboratories (Andover, MA, USA).FilterpaperforDBSpreparationwasofgrade903andwas purchasedfromWhatmanGmbH(Dassel,Germany).

2.2. Samplescollectionandpopulationstudy

Urine,plasma,andwholebloodsampleswerereferredtoour laboratory(ClinicalBiochemistryLaboratoryofUniversity Hospi-talofFoggia,Italy)forgenerallaboratorytests.Thecollectionof samples,fortheestablishmentofreferenceintervals,wasmade betweenJanuary 2010and March2011. Disease controlswere selectedfromthoseofthenewbornscreeningcenteratA.Meyer Children’s Hospital, Florence (Italy). The samples were divided accordingtopatientageintofourgroups:group1comprised51 babies(20femalesand31males;age,fromthreedaysto1year); urines were collectedfrom all, and blood for plasma and DBS preparationwerecollectedfrom24(9femalesand15males). Fur-thermore,other20DBSsampleswereselectedfromunaffected

patients(12femalesand8males;age,3–16days)whohadalready beenanalyzedformetabolicdiseases.Group2comprised40 chil-dren(17femalesand23males;age,1–12years);group3comprised 40teenagers(18femalesand22males;age,13–19years);group4 comprised78adults(55femalesand23males;age,20–40years).In addition,anothergroup(n◦5)comprised17newborns(7females and10males,fromwhich11plasmaand6DBSwereobtained) affectedbyASSdeficiency(n=11),OTCdeficiency(n=2),ASL defi-ciency (n=2), and CPS deficiency (n=2), diagnosed on clinical, biochemical,and/orgeneticgrounds.Allthesamplesanalyzedin thisstudywerereservematerialsthatwerenotneededforfurther diagnosticinvestigations,thusavoidtakingextramaterialsormore volumeofsamplefrompatients.

The selection of normal sampleswas based onthe normal-ityofthefollowinglaboratorymarkers:glucose,urea,creatinine, uricacid,aspartatetransaminase,alaninetransaminase, gamma-glutamyl-transferase,andhemoglobin.Aliquotsofurine,plasma, andwholebloodsamplesfromapparentlyhealthyindividuals,not affectedby urea-cycledisorders, were usedtoassess reference intervalsofOA.

2.3. Specimenpreparationandanalysis 2.3.1. Creatininemeasurement

Urinarycreatinineconcentrationwas determinedby a com-mercialkit,basedontheFolin’smethodwiththeJafféreaction [10],ontheautomatedinstrumentAU2700(BeckmanCoulter,Inc., Brea, CA).Theconcentrationof OAis expressedas aratio with urinarycreatinineconcentration(mmol/molofcreatinine)totake intoaccountthevariationsofurinaryvolumeamongsubjects,this procedureiscommonlyusedinclinicalbiochemistry.

2.3.2. DBS,plasma,andurinepreparation

DBSsampleswerepreparedfromEDTAbloodspottingatleast 30␮Lonfilterpaperandstored indesiccantsealedplasticbags at−20◦_{Cuntilanalysis.Plasma,obtainedfromEDTAblood,and}

urinesampleswerestoredat−20◦_{Cuntilanalysis.Adiscof6mm}

waspunchedfromDBSandtheOAwasextractedbyadding10␮L ofanaqueoussolutionofIS(5␮M),20␮Lof water,and120␮L ofACNwithammoniumacetate(5mM).Aftermixing,the solu-tionwasincubatedfor20minat30◦C,then2␮Lofsupernatant wereanalyzed.Twentymicrolitersofplasmasampleweremixed with20␮LofISsolution,andthesolutionwasdilutedwith260␮L ofACNwithammoniumacetate(4.6mM).Themixturewas vor-texed,andallowedtostandfor5minatroomtemperatureand, aftercentrifugationat3000×gfor5min,2␮Lofsupernatantwere analyzed.Theurinesampleswerefilteredtoremovebacteriaand anyparticulatematerial(membranefilterof0.22␮m;Millex-GV, Millipore,France)andthendiluted1:10withwaterbeforethe anal-ysis.Twentymicrolitersofdilutedurineweremixedwith20␮L ofIS solutionanddilutedwith260␮Lof ACNwithammonium acetate(4.6mM),mixedandallowedtostandfor5minatroom temperature;then2␮Loftheextractwereinjected.

2.3.3. LC–MS/MSanalysis

Urine,plasma,andDBSsampleswereanalyzedbyHILIC–MS/MS performed according to the method previously described [8]. Briefly,chromatographywasperformedbyacolumnpackedwith sub-2␮m particles (Zorbax RX-SIL,1.8␮m, 50mm×2.1mm id, Agilent Technologies, USA), and OA was eluted by an isocratic mobilephaseconstitutedofACN/water(90/10)containing4mMof ammoniumacetate.TheMSdetectionwasperformedby electro-sprayionizationandcompoundsweremonitoredusingamultiple reactionmonitoring(MRM)innegativemodeofthefollowing tran-sitions:m/z155>111forOA,andm/z157>113for[1,3-15_N

O.D’Apolitoetal./J.Chromatogr.Bxxx (2011) xxx–xxx 3

Table1

Meanvalues,within-day,andbetween-dayprecision(CV%)andaccuracy(%)obtainedmeasuringOAinplasma,driedbloodspot,andurinequalitycontrols. Concentrationof

addedOA(␮M)

Expectedvalue Within-day(n=6) Between-day(n=12)

Measuredmean values

Precision(CV%) Accuracy(%) Measuredmean values

Precision(CV%) Accuracy(%) Plasma

0 N/A 0.25 13.2 N/A 0.27 18.5 N/A

2.5 2.8 3.06 3.2 9.5 3.22 8.0 15.0

15 15.2 14.7 3.8 −3.4 14.6 8.2 −4.1

DBS

0 N/A 0.12 9.6 N/A 0.12 14.1 N/A

2.5 2.6 2.55 2.8 −2.8 2.69 13.6 3.6

15 15.1 13.4 0.8 −11.1 14.0 17.3 −7.4

Urinea

0 N/A 6.1 2.2 N/A 6.4 8.8 N/A

5 11 12.3 1.9 11.6 12.8 5.8 16.5

25 30.8 33.7 2.7 9.5 35.0 9.7 13.6

150 154.3 149.1 1.9 −3.4 166.3 7.2 7.8

N/A:notapplicable.

a_{Urinesampleswerediluted1:10beforeanalysis.}

internalstandard.BothOAanditsISelutedat2.00min,andthe totalruntimewasof4min.

PlasmaandDBSconcentrationswerecalculatedoncalibration curvespreparedusingenrichedplasmaandbloodina concentra-tionrangefrom0.1to25␮M.Bloodcalibrationpointswerespotted onfilterpaperandanalyzed.Theurinaryconcentrationswere cal-culatedusingacalibrationcurvepreparedwithaqueousstandard solutionsinaconcentrationrangefrom0.78to25␮M.Thelinearity ofcalibrationcurveswasestimatedbyevaluationofcoefficientof correlation(r)thatrangedamonganalyticalbatchesfrom0.997to 0.999forwater,from0.994to0.997forplasma,andfrom0.993to 0.998forDBS.

Qualitycontrol(QC)wascarriedouttoevaluatewithin-dayand between-dayprecisionandaccuracy.IneachanalyticalbatchQC specimenswereanalyzedtogetherwithrealsamples.Analiquot ofblankpoolofblood,plasma,andurinewasusedaslowQClevel onlyforprecisionevaluation,twoaliquotswerespikedwithOA standardtoobtainQCsatconcentrationof2.5and15␮Mforplasma andblood,whileforurinethreealiquotsofthenormalpoolwere spikedwithOAtoobtainQCsatconcentrationof5,25,and150_␮M. TheQCaliquotswerestoredastheclinicalsamplesuntiluse. 2.4. Statisticalanalysis

Theacquired datawereprocessed usingtheQuanLynx soft-ware(v.MassLynx4.0,Waters)tocalculatecalibrationcurves,QCs, andsampleconcentrations.Statisticalparameterswereobtained byMicrosoftExcelsoftware(v.11.0,Microsoft)and/orwith Kalei-daGraphsoftware(version4.1.1).ANOVAanalysiswasperformed usingtheBonferronipost-testcorrection.Thedataofeachgroup weretested fordistribution and,if aGaussian distributionwas obtained,themeanvaluewasusedtosetalimitatthemean±2SD; thevaluesoutside thelimitwereconsideredtobeoutliersand rejected.Whenthedatasetsshowednon-Gaussiandistributions,a robuststatisticalmethodwasusedforthedetectionofoutliers.This robust statistical technique uses distribution-free inter-quartile range (IQR)toseta Tukeyfence [11,12]. Thedata weresorted intoascendingsequencetoobtainthelowerandupperquartile point, Q1and Q3,respectively. The differencebetweenQ3 and Q1(Q3−Q1)is named IQR. The Tukeyfenceis therange from Q1− 1.5*IQRtoQ3+1.5*IQR.ThevaluesoutsidetheTukeyfence wereconsideredtobeoutliersand removed.Afterthe elimina-tionofoutliers,thedatasetswerere-testedfordistribution,and ifaGaussiandistributionwasobtained,themeanvaluewasused toestablishalimitatthemean±2SD.Instead,ifanonGaussian

distribution was obtained the data sets were described using median,range,andpercentiles.Thevaluesoutsidethelimitof1st %ileandof99th%ilewereconsideredtobeoutliersandrejected.

3. Results

3.1. Qualitycontrolresults

QCsampleswereanalyzedduringtheassaysofpatientsamples. Within-dayprecision(CV%)andaccuracy(%)werecalculatedby analyzingeachlevelofQCsamples6timesinthesameanalytical run.TheexpectedvaluewascalculatedaddingOAvalueinblank sampletotheaddedamountofOAstandard,correctedbysample volumesusedtopreparethemixture(Table1).TheOAprecision (CV%)onthreeQClevelsrangedfrom3.2to13.2forplasma,from 0.8to9.6forDBS,andfrom1.9to2.7forurine,whiletheaccuracy (%)rangedfrom−3.4to9.5forplasma,from−11.1to−2.8forDBS, andfrom−3.4to11.6forurine.Thebetween-dayprecisionand accuracywerecalculatedbyanalyzingeach levelofQC samples onceadayfor12workingdays.Theprecision(CV%)wasestimated from8.0to18.5forplasma,from13.6to17.3forDBS,andfrom 5.8to9.7forurine.Theaccuracy(%)rangedfrom−4.1to15.0for plasma,from−7.4to3.6forDBS,andfrom7.8to16.5forurine (Table1).

3.2. Populationstudyresults

ThedescriptivestatisticofOAinplasma,DBSandurineof ref-erence populationis reported in Table2.Urinary OA rawdata ofeachgroupshowedanon-Gaussiandistribution;therefore,all datawereelaborated bytherobuststatisticmethodtoidentify and torejecttheoutliersabove thehighTukeyfence.The val-uesexcluded were 2/51(3.9%) from group1, 3/40 (7.5%) from group2,5/40(12.5%)fromgroup3,and3/78(3.8%)fromgroup 4.Aftertheeliminationof outliers,data setswerere-testedfor distributionandaGaussiandistributionwasobtainedforgroups 1, 2, and 3; instead, thegroup 4 still showed a non Gaussian distribution.However,parametricandnon-parametricresultsin each group are very close; instead, comparing urinary excre-tion ofOA among thegroups,it wasevident that thelevelsin group 1 were at least two times higher than in groups 2, 3, and4.Theanalysisofvarianceamonggroupsshowedsignificant differencesbetweengroup1vs2,group1vs3,andgroup1vs4 (p<0.0001).Wealsocorrelatedtheageofpatientsandtheurinary

4 O.D’Apolitoetal./J.Chromatogr.Bxxx (2011) xxx–xxx

Table2

Descriptivestatisticofage-relatedoroticacidconcentrationsinplasma,driedbloodspot,andurinefromunaffectedsubjects. Populationstudy(ageintervals,years)

Group1(<1) Group2(1–12) Group3(13–19) Group4(20–40)

Plasma(␮M) (n) (23) (40) (39) (73) Mean(SD) 0.30(0.12) 0.30(0.09) 0.15(0.07)a _0.28(0.16) Mean±2SD 0.06–0.54 0.12–0.48 0.01–0.29 0.0–0.60 Median 0.27 0.27 0.13 0.22 Min–max 0.12–0.60 0.15–0.48 0.03–0.31 0.06–0.82 1st–99th%ile 0.13–0.59 0.16–0.48 0.04–0.30 0.09–0.77 DBS(␮M) (n) (43) (40) (39) (77) Mean(SD) 0.38(0.24)b _{0.12(0.06) 0.11(0.04) 0.14(0.05)} Mean±2SD 0.0–0.86 0.0–0.24 0.03–0.19 0.04–0.24 Median 0.35 0.10 0.10 0.13 Min–max 0.07–0.91 0.04–0.25 0.04–0.22 0.05–0.31 1st–99th%ile 0.07–0.89 0.04–0.24 0.04–0.21 0.07–0.29

Urine(mmol/molofcreatinine)

(n) (49) (37) (35) (75) Mean(SD) 1.65(0.80)c _{0.63(0.26) 0.39(0.12) 0.50(0.25)} Mean±2SD 0.05–3.25 0.11–1.15 0.15–0.63 0.0–1.00 Median 1.49 0.60 0.39 0.45 Min–max 0.40–3.47 0.22–1.39 0.17–0.65 0.01–1.27 1st99th%ile 0.49–3.44 0.23–1.30 0.18–0.64 0.13–1.21

ANOVAwithBonferroni’sposttest.

a_{p<0.0001group3vsgroup1,group2,andgroup4.} b _{p<0.0001group1vsgroup2,group3,andgroup4.} c _{p<0.0001group1vsgroup2,group3,andgroup4.}

OAexcretionand,asreportedinFig.1,thereisasignificantnegative polynomialcorrelation(p<0.0001).

PlasmaOAraw datashowedanon-Gaussiandistributionfor group1,3,and4;therefore,aftertherejectionofoutliersabove thehighTukeyfence,thevaluesexcludedwere1/24(4.2%)from group1,nonefromgroup2,1/40(2.5%)fromgroup3,and5/78 (6.4%)fromgroup4.Aftertheeliminationofoutliers,datasetswere re-testedforGaussiandistribution,thegroup4stillshowedanon Gaussiandistribution.Alsofor plasma,theparametricand non-parametricresultsareverycloseineachgroup(Table2).Thegroup 3showedOAplasmalevelslowerthangroup1,group2,andgroup 4.Fig.2reportsthedot-plotofplasmaOAlevelsinallgroups,and

Fig.1.Scatter-plotofreferenceurinaryoroticacidvaluesingroups1–4vsage (months).

theanalysisofvarianceamonggroupsshowedsignificant differ-encesbetweengroup3vsgroup1,group3vs2,andgroup3vs4 (p<0.0001).

OAraw datainDBSshoweda non-Gaussiandistributionfor group1,3,and4;therefore,aftertherejectionofoutliersabove thehighTukeyfence,thevaluesexcludedwere1/44(2.3%)from group1,nonefromgroup2,1/40(2.5%)fromgroup3,and1/78 (1.3%)fromgroup4.Aftertheeliminationofoutliers,datasetswere re-testedforGaussiandistribution,whichstillwasabnormalfor group4.AlsoforDBS,theparametricandnon-parametricresults areverycloseineachgroup(Table2).Group1showedOADBS lev-elshigherthangroups2–4.Fig.3reportsthedot-plotofOAinDBS ofallgroups,andtheanalysisofvarianceamonggroupsshowed

O.D’Apolitoetal./J.Chromatogr.Bxxx (2011) xxx–xxx 5

Fig.3. Dot-plotofDBSoroticacidvaluesinallreferencegroups.

significantdifferencesbetweengroup1vsgroup2,group1vs3, andgroup1vs4(p<0.0001).

Resultsfromaffectedpatients(group5)arereportedinTable3. OAlevels(_␮M)rangedfrom0.12to117inplasmaandfrom12.8to 46.6inDBSofpatientswithASSdeficiency,from4.8to86.3inDBS ofpatientswithOTCdeficiency,andfrom0.02to0.03inplasma ofpatientswithASLdeficiencyduringtherapeutictreatment.Two patientswithCPSdeficiencyshowedOAlevelsof0.28inplasmaand of0.35inDBS.Inaddition,theurinaryexcretionofOA(mmol/mol ofcreatinine)measuredatmomentofdiagnosisinASSpatients# 1,and#2wasof1.34,and1.63,respectively.

4. Discussion

Themain purposeof this studywastoestablishagerelated referenceintervalsofOAinDBS,plasma,andurineinunaffected

Table3

OroticacidlevelsinplasmaandDBSsamplesofaffectedpatients. Patient

samples

Age (days)

Defects Oroticacid(␮M)

Plasma DBS mf#1 4 ASS-I 0.17 mf#2 4 ASS-I 0.12 mf#3 3 CPS 0.35 mf#5 7 CPS 0.28 mf#6 4 ASS-I 46.5 mf#7 4 ASS-I 62.0 mf#10 4 ASS-I 6.1 mf#11 4 ASL* 0.03 mf#12 3 OTC 86.3 mf#13 4 ASLa _0.02 mf#14 3 OTC 4.83 mf#15 4 ASS-I 46.6 mf#16 4 ASS-I 12.8 mf#17 4 ASS-I 15.0 mf#18 4 ASS-I 1.0 mf#19 4 ASS-I 60.1 fg#39 5 ASS-I 116.8 Reference intervals Group1 <1year 0.13–0.59 0.07–0.89

a_{Bothpatientswereundertherapeutictreatment.}

children, teenagers, and adults. Both parametric and non-parametric results have been reported. The urinary values for unaffectedpatientssuggestedasignificantbutcontinuousnegative polynomialcorrelationofOAconcentrations(mmol/molof creati-nine)withageinchildrenfromthree daysuptoadults(Fig.1). Inagreementwithpreviousreportsthistrendislikelyduetothe increaseofcreatinineexcretionwithage[5,13–16].Wedefinedthe limitsofOAreferenceintervalsbetweenthe1st%ileand99th%ile, althoughinclinicalpracticethe99th%ilelimitismostlyof inter-est.UrinaryOAlimitsinunaffectedchildrenbetweenthreedaysof birthand1year(group1)wouldbefrom0.49to3.44mmol/molof creatinine,andingroup2,between1yearand12years,thelimits wouldbefrom0.23to1.30mmol/molofcreatinine.Inunaffected teenagers,between13and19years(group3),thelimitswouldbe from0.18to0.64mmol/molofcreatinine,andinadults,between20 and40years(group4),wouldbefrom0.13to1.21mmol/molof cre-atinine.ThelevelsofurinaryOAingroup1aresignificantlyhigher thangroup2,3,and4.Theageperiodfrombirthto12 months (group1)ismorecriticalforthedetectionofmanyinborn patholo-gies.Instead,thevariationsofOAlevelsamonggroups2,3,and4 appearrelativelysmallandnotstatisticallysignificant.

The1stand99th%ileofOAinplasmaofunaffectedchildrenof group1wouldbefrom0.13to0.59␮M,andinthegroup2would befrom0.16to0.48␮M.Inunaffectedteenagers,thelimitswould befrom0.04to0.30␮M,andinadults,from0.09to0.77␮M.The levelsofplasmaOAingroup3aresignificantlylowerthanother groups(1,2,and4),andthevariationsofOAlevelsamonggroups 1,2,and4werenotstatisticallysignificant.

The1stand99th%ileofOAinDBSofunaffectedchildrenofgroup 1wouldbefrom0.07to0.89␮M,andinthegroup2wouldbefrom 0.04to0.24␮M.Limitsofgroup3wouldbefrom0.04to0.21␮M, andinadultsfrom0.07to0.29␮M.ThelevelsofplasmaOAingroup 1aresignificantlyhigherthanothergroups(2–4).Instead,the vari-ationsofOAlevelsamonggroups2,3,and4appearverysmalland notstatisticallysignificant.DBSandnormalizedurinaryOAlevelsof group1bothshowsignificantlyhigherOAlevelsthanothergroups. AssessmentofurinaryOAisfacilitatedbyitsremarkable stabil-ity.NormalhumanurinecontainstraceamountsofOA[17]and, aspreviouslyreportedbyus[8]andconfirmedinthiswork,the compoundisquantifiableatsub-micromolarlevelalsoinplasma andbloodofunaffectedpatients.

TheearlycolorimetricmethodsforOAdeterminationinurine weresimpleandrapid,yetnon-selective,whichmaygiveriseto false-positiveresults.OAreferenceintervalsobtainedby colorimet-ricmethods[18,19]arewiderthanthoseobtainedbyourmethod, and also by gas chromatography–mass spectrometry (GC–MS) methods[5,16,20],andbyion-exchangechromatographymethods withUVdetection[13,14].Unfortunately,eachoftheselasttwo methodshavedifficultiesandlimitations,aretimeconsumingand notadaptableasscreeningmethods.

Todate,sinceItoetal.[21]describedaLC–ESI-MS/MSmethod for a large number of urinary metabolites, including OA, oth-ers have proposed a more specific, high-throughput, sensitive stable-isotope-dilutionLC–ESI-MS/MSmethodforthe determina-tionofurinaryOAandintermediatemetabolitesinurinespecimens [6,15,22].EventhoughreferenceintervalsofurinaryOAobtained byLC–ESI-MS/MSarecomparabletothosereportedinthisstudy byHILIC–ESI-MS/MS,ourreferencelimitsarenarrowerandwell overlappedwiththoseobtainedbystableisotopeGC–MS meth-ods[5,16].Inthemethodreportedhere,theOAfromDBS,plasma, andurinewaselutedfromcolumnpumpinganisocraticmixtureof solvents,whichimprovesthereproducibilityofretentiontimeand thesensitivity;moreover,nointerferenceswereobservedduring ananalyticalbatchof100samples.

Forthefirsttimewereportthereferenceintervalsinplasmaand DBSfromapparentlyhealthysubjectsagedfromthreedaysto40

6 O.D’Apolitoetal./J.Chromatogr.Bxxx (2011) xxx–xxx

years,demonstratingthatthismethodallowsaquantitative anal-ysisofOAalsointhesebiologicalfluids.Inaddition,thismethod hasbeenemployedtoanalyzesamplesfrompatientsaffectedby fourdifferentdefectsofureacycle.LevelsofOAinplasmaandDBS frompatientsaffectedbyASS(n=9/11)andOTC(n=2/2) deficien-ciesweresignificantlyoverthe99th%ileofourreferenceintervals; tonotethatonlytwopatientsaffectedbyASSdeficiency(plasma samples#1and#2)showednormalvalues,anexpectedresult becauseatmomentofsamplecollectiontheyalsoshowednormal excretionoforoticacidinurine(1.34,and1.63mmol/molof creati-nine,respectively).Whereas,twopatientsaffectedbyCPS,adefect withoutoroticaciduria,andtwopatientsaffectedbyASLunder therapeutictreatmentshowednormalandlowOAlevels, respec-tively.Therefore,thesefindingssuggestthatdeterminationofOA inplasmaandDBScouldbeusedasanadditionaltesttoscreen severalmetabolicdiseases,eventhoughfurtherevaluationshould beperformedonahighernumberofpositivesamples.

5. Conclusions

Althoughthefrequencyofdisorderspresentingoroticaciduria islow,afastandaccurateanalysisofOAusingasmallandeasily transportablesample,suchasDBS,shouldbeusefulandclinically relevant,inparticularfornewbornscreening.

Acknowledgements

We are grateful tocolleagues Maria Assunta Schiavone and Clelia Scarano (Clinical Biochemistry Laboratory, University-HospitalofFoggia,Italy),andMonicaGelzo(UniversityFedericoII

ofNaples)fortheirvaluablecollaboration.Theauthorsalsothank theUniversityofFoggiaforfunding(PremialitàPRIN2008).

References

[1]L.H.Smith,M.Sullivan,C.M.Huguley,J.Clin.Invest.40(1961)656. [2]C.Salerno,C.Crifò,J.Chromatogr.B781(2002)57.

[3]M.E.Brosnan,J.T.Brosnan,J.Nutr.137(2007)1656S. [4]J.O.Sass,D.Skladal,Pediatr.Nephrol.13(1999)912.

[5] M.T.McCann,M.M.Thompson,I.C.Gueron,M.Tuchman,Clin.Chem.41(1995) 739.

[6] A.B.P.vanKuilenburg,H.vanLenthe,M.Loffler,A.H.vanGennip,Clin.Chem. 50(2004)2117.

[7] J.Rajantie,Pediatr.Res.15(1981)115.

[8]O.D’Apolito,D.Garofalo,G.Paglia,A.Zuppaldi,G.Corso,J.Sep.Sci.33(2010) 966.

[9]H.R.Yoon,Arch.Pharm.Res.30(2007)387. [10] O.Folin,J.Biol.Chem.17(1914)475.

[11] J.W.Tukey,ExploratoryDataAnalysis,Addison-Wesley,Reading,MA,1977,p. 506.

[12] P.S.Horn,L.Feng,Y.Li,A.J.Pesce,Clin.Chem.47(2001)2137. [13]C.Bachaman,J.P.Colombo,J.Clin.Chem.Clin.Biochem.18(1980)293. [14]M.Asai,S.Sumi,K.Kidouchi,H.Imaeda,H.Togari,Y.Wada,Pediatr.Int.42

(2000)499.

[15] M.S.Rashed,M.Jacob,M.Al-Amoudi,Z.Rahbeeni,M.A.D.Al-Sayed,L. Al-Ahaidib,etal.,Clin.Chem.49(2003)499.

[16]J.Chen,M.J.Bennett,MethodsMol.Biol.603(2010)445. [17]M.Lotz,H.J.Fallon,L.H.Smith,Nature197(1963)194. [18] M.L.Harris,V.G.Oberhoizer,Chin.Chem.26(1980)473.

[19]D.ElKhiari,N.Tebib,N.Kaabachi,M.Feki,Y.BenHamouda,A.Mebazaa,Arch. Pediatr.5(1998)15.

[20]M.Rimoldi,P.Bergomi,A.Romeo,S.Diflinato,J.Inherit.Metab.Dis.17(1994) 243.

[21]T.Ito,A.B.P.vanKuilenburg,A.H.Bootsma,A.J.Haasnoot,A.vanCruchten,Y. Wada,etal.,Clin.Chem.46(2000)445.

[22]S.Hartmann,J.G.Okun,C.Schmidt,C.D.Langhans,S.F.Garbade,P.Burgard, etal.,Clin.Chem.52(2006)1127.