Azaspiracid variability of Azadinium poporum (Dinophyceae) from the China Sea
Bernd Krock
a,*, Urban Tillmann
a, Matthias Witt
b, Haifeng Gu
caAlfredWegenerInstitutHelmholtz-Zentrumfu¨rPolar-undMeeresforschung,ChemischeO¨kologie,AmHandelshafen12,27570Bremerhaven,Germany
bBrukerDaltonikGmbH,Fahrenheitstr.4,28359Bremen,Germany
cThirdInstituteofOceanography,SOA,Xiamen361005,PRChina
1. Introduction
Among the known marine shellfish poisoning syndromes, azaspiracidshellfishpoisoning(AZP)isthemostrecentone,which was observed for the first time in the Netherlands in 1995.
ContaminatedmusselscultivatedinIrelandwereconsumedand intoxicatedatleasteightpeople(McMahonandSilke,1996).Three yearslatertheimplicatedtoxinwasidentified,isolated,structur- ally defined and named azaspiracid (now called azaspiracid-1 (AZA-1)(Satakeetal.,1998).Inthefollowingyearsothervariants ofAZA-1werefoundandisolatedfromshellfish(Ofujietal.,1999;
James et al., 2003). As a result,the European Unionhas set a regulatory limit for maximum levels of AZAs in shellfish (160
m
gkg1).Duetotheirstructural characteristicsAZAswere earlysuspectedtobeofdinoflagellateorigin,however,theAZA- producing organism remained unknown until the isolation of AzadiniumspinosumfromtheNorthSeain2007(Krocketal.,2009).SpeciesofthegenusAzadiniumhavesofarbeenreportedfromthe NorthSea(Tillmannetal.,2009,2010,2011),theFrenchandIrish
coastoftheeasternAtlantic(Ne´zanetal.,2012;Salasetal.,2011), theMediterraneanSea(Percopoetal.,2013),theArgentineancoast (AkselmanandNegri,2012),theKoreancoast(Potvinetal.,2012) andtheChinesecoast(Guetal.,2013).Nevertheless,thepresence ofAZAsappearedtobedistributedmuchmorewidely,reportedin Northern Africa, northern Europe, Chile, USAand China (Bran˜ a Magdalena etal., 2003; Jameset al., 2002;Klontz et al.,2009;
Lo´pez-Riveraetal.,2010;Talebetal.,2006;Yaoetal.,2010).The discrepancy between the distribution of Azadinium and AZAs suggestthatA.spinosummighthaveawiderdistribution,orother AzadiniumspeciescouldproduceAZAs.Initially,A.spinosumwas theonlyspeciesforwhichAZAswerereported.ForA. spinosum strain 3D9, thetoxin profile consistedof AZA-1, AZA-2 and an isomerofAZA-2(Krocketal.,2009),whichwaslateridentifiedas AZA-1methylesterandfoundtobeanextractionartifact(Jauffrais etal.,2012).AZA-1and-2productionwassubsequentlyconfirmed forA.spinosumstrainsfromDenmark(Tillmannetal.,2011)and Ireland(Salasetal.,2011),indicatingthatproductionandprofileof knownAZAsisastablecharacteristicofthespeciesA.spinosum.
Other related species/strains of Amphidomataceae have been reportednottocontainanyoftheknownazaspiracids.However, AZAproductionwithinAmphidomataceaeprobablyismuchmore complexand diverse;recentevidenceindicates thepresence of ARTICLE INFO
Articlehistory:
Received9October2013
Receivedinrevisedform11April2014 Accepted11April2014
Availableonline
Keywords:
Azadiniumpoporum Chinesecoastalwaters
Azaspiracidshellfishpoisoning(AZP) LC–MS/MS
Toxinprofiles AZA-40 AZA-41
ABSTRACT
AzadiniumpoporumisasmalldinoflagellatefromthefamilyAmphidomataceaewhichisknownforthe potential production of azaspiracids (AZAs) causative of azaspiracid shellfish poisoning (AZP). A.
poporumhasbeenrecordedfromEuropeanandwesternPacificwaters.Herewereportonthehigh variabilityoftoxinprofileswithinthisspeciesinChinesecoastalwaters.Outof16analyzedstrainsofA.
poporumfromdifferentgeographiclocationsalongtheChinesecoastline,threestrainsprovednotto containAZAs,whereas13strainscontaineddifferentcombinationsofAZA-2,AZA-11,AZA-36,ayet unknownisomerofAZA-1(namedAZA-40)andnewAZAwithyetunreportedmolecularmassof853Da (namedAZA-41).ThenewAZA-40,otherthanAZA-1itself,belongstotherecentlydiscovered‘‘348-type’’
group,whichintandemmassspectrometrydisplaysagroup4fragmentwithm/z348insteadofthe group4fragmentoftheclassicAZAswithm/z362,indicatingashiftofamethylgroupfromtheC24–C40
partofthemolecule(ringsF–I)totheC2–C9part(carboxylicsidechainandringA).AZA-41apparentlyis adehydrovariantofAZA-2.Inaddition,apreviouslyreportedAZAwithamolecularmass871DAcould beunambiguouslyassignedtoAZA-11,whichisknowntobeashellfishmetaboliteofAZA-2.Thisisthe firstreportofAZA-11beingalsodenovosynthetizedbydinoflagellates.
ß2014ElsevierB.V.Allrightsreserved.
* Correspondingauthor.Tel.:+4947148312055;fax:+4947148312115.
E-mailaddress:bernd.krock@awi.de(B.Krock).
ContentslistsavailableatScienceDirect
Harmful Algae
j our na l ho me p a ge : w ww . e l se v i e r . com / l oc a te / h a l
http://dx.doi.org/10.1016/j.hal.2014.04.012 1568-9883/ß2014ElsevierB.V.Allrightsreserved.
new AZAs with a modified substitution pattern in Azadinium poporumandAmphidomalanguida(Krocketal.,2012).NorthSea strainsandtheKoreanisolateofA.poporumwerefoundtoproduce AZAs witha characteristic m/z 348-fragment, however, in two differentvariantswithdifferentmassesintheNorthSeastrains and the Korean strain, respectively (Krock et al., 2012). Until recentlyAZPappearedtobeamerelyEuropeanproblem,butthe recentfindingsofAzadiniumspp.andAZAsatdifferentlocationsof theAsianPacific(Guetal.,2013;Potvinetal.,2012)giveincreasing evidenceAZPmayposeapotentialseafoodsafetyriskinthisto dateunaffectedworldregion.
Theaimofthisworkwastofurtherinvestigatetheoccurrence and variability of AZAs and the distribution of the producing organismsinChinesecoastalwaters.
2. Materialsandmethods 2.1. Samplecollectionandtreatment
SedimentsampleswerecollectedalongthecoastofChinausing agrabsampler(geographicalcoordinatesandsampledateswere providedinTable1).Thesedimentsampleswerestoredinthedark at48Cuntilfurthertreatment.Approximately,2gofwetsediment weremixed with 20mL of filteredseawater and sonicatedfor 2min(100W)todislodgedetritalparticles.Thewateryslurrywas incubateddirectlyinseriesofsmallcontainersinf/2-Simedium (GuillardandRyther,1962)at208C,90
m
Em2s1undera12-h light:12-hdark cycle(hereafter called‘‘standard culture condi- tions’’).Azadinium cells are characterized by swimming at low speed, interrupted by short, high-speed ‘jumps’ in various directions (Tillmann et al., 2009). Cells exhibiting such a characteristic swimming behavior were isolated by means of drawn-outPasteurpipettesandestablishedintoclonalcultures.Onlyonestrainwasestablishedfromonecontainertoguarantee theyrepresenttrueclonalstrains.16strainswereestablishedthis wayandmaintainedunderstandardcultureconditions.
2.2. PCRamplificationsandsequencing
Total algal DNA wasextracted from 10mL of exponentially growing Azadinium cultures using a plant DNA extraction kit (Sangon, Shanghai, China) according to the manufacturer’s
protocol.ThetotalITS1-5.8S-ITS2wasamplified usingITSA and ITSBprimers(Adachietal.,1996).
The PCR protocol was as follows: initial denaturation for 3.5minat 948C, followed by 35 cyclesof 50s denaturation at 948C,50sannealingat458C,and80sextensionat728C,plusa finalextensionof10minat728C.PCRproductsweresequenced directlyinbothdirectionsusingtheABIBig-Dyedye-terminator technique(AppliedBiosystems,FosterCity,CA,USA),accordingto the manufacturer’srecommendations. The BLAST programs are used for searching DNA databases for sequence similarities (Altschuletal.,1997).
2.3. Chemicalanalysisofazaspiracids
ForAZAanalysis,culturesofA.poporumweregrownin200mL Erlenmeyerflasksunderstandardcultureconditions.Around107 cellswerecollectedbycentrifugationattheexponentialphase.Cell pelletswereextractedwith400
m
Lacetonebyreciprocalshaking at6.5ms1with0.9glysingmatrixD(ThermoSavant,Illkirch, France)inaBio101FastPrepinstrument(ThermoSavant,Illkirch, France)for45s.Extractswerethencentrifuged(Eppendorf5415R, Hamburg, Germany) at 16,100g at 48C for 15min. Each supernatant was transferred toa 0.45-m
mpore-size spin-filter (MilliporeUltrafree,Eschborn,Germany)andcentrifugedfor30s at800g,andtheresultingfiltratebeingtransferredintoanLC autosamplervialforLC–MS/MSanalysis.2.3.1. Singlereactionmonitoring(SRM)measurements
Waterwasdeionizedandpurified(Milli-Q,Millipore,Eschborn, Germany)to18M
V
cm1orbetterquality.Formicacid(90%,p.a.), acetic acid(p.a.)and ammoniumformate(p.a.)werepurchased fromMerck (Darmstadt,Germany). Thesolvents,methanol and acetonitrile,werehighperformanceliquidchromatography(HPLC) grade(Merck,Darmstadt,Germany).Massspectralexperimentswereperformedtosurveyforawide arrayofAZAs.TheanalyticalsystemconsistedofanABI-SCIEX- 4000QTrap,triplequadrupolemassspectrometerequippedwitha TurboSpray1interfacecoupledtoanAgilentmodel1100LC.The LC equipment included a solvent reservoir, in-line degasser (G1379A), binary pump (G1311A), refrigerated autosampler (G1329A/G1330B), and temperature-controlled column oven (G1316A).
Table1
StrainsofAzadiniumpoporumexaminedinthepresentstudy,includingtheirorigin,collectiondates,ribotypes,AZA-cellquotas[fgcell1]andtoxinprofiles.
Strain Origin Latitude Longitude Collection date
Ribo- type
AZA-40 AZA-41 AZA-2 AZA-36 AZA-11 AZA-profile Profile type
Source
AZBH01 SouthChinaSea 218230170 0 1098230170 0 24.05.2010 B 0.9 – – – – AZA-40 D Thisstudy AZBH03 SouthChinaSea 218230170 0 1098070160 0 24.05.2010 B 0.5 – – – – AZA-40 D Thisstudy
AZCH01 YellowSea 398140590 0 1228360050 0 02.05.2011 C – – – – – noAZAs A Thisstudy
AZCH10 YellowSea 398140590 0 1228360050 0 02.05.2011 B – – – 0.7 0.2 AZA-11,-36 F Thisstudy AZDH38 EastChinaSea 308000000 0 1228440480 0 19.04.2011 B – 0.9 – – – AZA-41 E Thisstudy AZDH39 EastChinaSea 308000000 0 1228440480 0 19.04.2011 C – – 3.9 0.6 – AZA-2,-36 C Thisstudy AZDH41 EastChinaSea 308000000 0 1228440480 0 19.04.2911 B – – – – 3.0 ATA-11 B Thisstudy AZDH43 EastChinaSea 308000000 0 1228440480 0 19.04.2011 B – – – – 0.8 AZA-11 B Thisstudy AZDH51 EastChinaSea 308000000 0 1228440480 0 19.04.2011 C – – – – – noAZAs A Thisstudy AZDH55 EastChinaSea 308000000 0 1228440480 0 19.04.2011 B – – – – 1.6 AZA-11 B Thisstudy AZDH56 EastChinaSea 308000000 0 1228440480 0 19.04.2011 C – – 1.8 0.3 – AZA-2,-36 C Thisstudy AZDY06 SouthChinaSea 228360080 0 1148370210 0 05.06.2010 B – – 7.6 1.2 – AZA-2,-36 C Thisstudy AZFC21 SouthChinaSea 218290580 0 1088130530 0 28.04.2011 C 0.2 – – – 0.5 AZA-11,-40 H Thisstudy AZFC22 SouthChinaSea 218290580 0 1088130530 0 28.04.2011 C – – 4.4 – – AZA-2 G Thisstudy AZLY01 YellowSea 348470490 0 1198310080 0 09.05.2011 B – – – 0.4 1.0 AZA-11,-36 F Thisstudy
AZLY02 YellowSea 348470490 0 1198310080 0 09.05.2011 B – – – – – noAZAs A Thisstudy
G25 BohaiSea 388540600 0 1178530000 0 14.08.2007 B – – – 1.4 – AZA-36 C Guetal.(2013)
G42 EastChinaSea 308000000 0 1228440480 0 19.04.2011 C – – 6 – – AZA-2 G Guetal.(2013)
G60 EastChinaSea 308000000 0 1228440480 0 19.04.2011 B – – – – – noAZAs A Guetal.(2013) G64 EastChinaSea 308000000 0 1228440480 0 19.04.2011 C – – 8–23 – – AZA-2 G Guetal.(2013) G66 EastChinaSea 308000000 0 1228440480 0 19.04.2011 B – – – – 1.5 AZA-11 G Guetal.(2013) G68 SouthChinaSea 218290580 0 1088130530 0 28.04.2011 B – – 2–5 – – AZA-2 G Guetal.(2013)
Separation of AZAs (5
m
L sample injection volume) was performedbyreverse-phasechromatographyonaC8phase.The analyticalcolumn (502mm)waspacked with3m
mHypersil BDS 120A˚ (Phenomenex, Aschaffenburg, Germany) and main- tained at 208C. The flow rate was 0.2mLmin1 and gradient elutionwasperformedwithtwoeluants,whereeluantAwaswater andBwasacetonitrile/water(95:5,v/v),bothcontaining2.0mM ammonium formate and 50mM formic acid. Initial conditions were8mincolumnequilibrationwith30%B,followedbyalinear gradientto100%Bin8minandisocraticelutionuntil18minwith 100%Bthenreturningtoinitialconditionsuntil21min(totalrun time:29min).AZAprofilesweredeterminedinoneperiod(0–18min)with curtain gas: 10psi, CAD: medium, ion spray voltage: 5500V, temperature:ambient,nebulizer gas:10psi, auxiliarygas:off, interface heater: on, declustering potential: 100V, entrance potential: 10V, exit potential: 30V). SRM experiments were carried out in positive ion mode by selecting the transitionsshownin Table 2. AZAswere calibrated againstan externalstandardsolutionofAZA-1(certifiedreferencematerial (CRM) programme of the IMB-NRC, Halifax, Canada) and expressedasAZA-1equivalents.AsolutioncontainingAZA-11, previously isolated from shellfish and purified by the Marine Institute,Oranmore,Galway,IrelandwaskindlydonatedbyJane Kilcoyne.
2.3.2. Precursorionexperiments
Precursorsofthefragmentsm/z348andm/z362werescanned inthepositiveionmodefromm/z400to950underthefollowing conditions:curtaingas:10psi,CAD:medium,ionsprayvoltage:
5500V,temperature:ambient,nebulizergas:10psi,auxiliarygas:
off,interface heater:on, declusteringpotential:100V,entrance potential:10V,collisionenergy:70V,exitpotential:12V.
2.3.3. Productionspectra
ProductionspectrawererecordedintheEnhancedProductIon (EPI) mode in the mass range from m/z 150 to 930. Positive ionizationand unit resolution mode were used. The following parameterswereapplied:curtaingas:10psi,CAD:medium,ion
spray voltage: 5500V, temperature: ambient, nebulizer gas:
10psi, auxiliary gas: off, interface heater: on, declustering potential:100V,collisionenergyspread:0,10V,collisionenergy:
70V.
2.3.4. FT-ICR-MSmeasurements
HighresolutionmassspectrawereacquiredwithasolarixXR Fouriertransformioncyclotronresonancemassspectrometer(FT- ICR-MS;BrukerDaltonikGmbH,Bremen,Germany)equippedwith a 12T refrigerated actively shielded superconducting magnet (Bruker Biospin, Wissembourg, France), a dual ion source and Paracellanalyzercell (Nikolaev et al.,2011). Thesampleswere ionizedby electrosprayionizationin positiveion mode (Bruker Daltonik GmbH, Bremen, Germany). Sample solutions were continuouslyinfusedusingasyringeataflowrateof2
m
Lmin1. 1. The detection mass range was set to m/z 150–3000. Ion accumulationtime foreach scanwassetto0.1s.Several scans wereaddedforthefinalmassspectrum.Datasetswereacquired with4MWdatapointsresultinginaresolvingpowerof450,000at m/z400.Spectrawerezero-filledtoprocesssizeof8Mdatapoints beforesineapodization.Massspectrawerecalibratedwitharginineclustersusinga linear calibration. A 10
m
gmL1 solution of arginine in 50%methanolwasused togeneratetheclusters.Ionaccumulation time was set to several seconds for MS/MS experiments for improved S/N of the fragment mass peaks. The quadrupole isolationwindowwassetto0.5Daandcollisionenergywassetto 30eV.
3. Results
3.1. Molecularcharacterization
TenstrainsofA.poporumfromtheYellowSea,EastChinaSea and SouthChinaSea shareidenticalITSsequenceswiththat of strain G25 (GenBank number: KC286572) and thus belong to ribotypeB,andtherestsixstrainsshareidenticalITSsequences withthatofstrainG42(GenBanknumber:KC286581)(Table1) andthusbelongtoribotypeC(Guetal.,2013).Theydifferfrom eachotherat11positions(98.3%similarity).
3.2. Toxinprofiles
Outofthe16A.poporumstrainstested,threestrains(AZDH51, AZCH01,AZLY02)provednottocontainanyknownAZAsatalimit of detection of 3–6agcell1 (depending on the available cell biomass)expressedasAZA-1equivalents.Inaddition,allstrains wereanalyzedintheprecursorionmodeofm/z348andm/z362in order todetectunknownAZA variants, but no additional AZAs werefound.
One strain (AZFC22) exclusively produced AZA-2 (Table 1, Fig.1)at acell quotaof4.4fgcell1.Twostrains(AZBH01and AZBH03)exclusivelyproducedanew,yetundescribedazaspiracid withamolecularmassof841Da,namedhereAZA-40atcellquotas of0.9and0.5fgcell1,respectively.AZA-40isastructuralisomer ofAZA-1,butincontrasttoAZA-1,itbelongstothe348-typeAZAs (Fig.2)asAZA-36,-37,-38,and-39,whichhavebeenshownbefore tobeproducedbyA.poporumandA.languida(Krocketal.,2012).
ThesumformulaofAZA-40wasdeterminedasC47H72NO12byhigh resolutionmassspectrometry(HRMS)(Table3).
Threestrains(AZDH43,AZDH55andAZDH41)onlyproduced AZA-11atcellquotasof0.8,1.6and3.0fgcell1(expressedasAZA- 1equivalents),respectively.TheidentityofAZA-11wasconfirmed bycomparisonofretentiontimes(notshown)andtheCIDspectra ofasolutioncontainingAZA-11andtheisobariccompoundofA.
poporum(Fig.3).
Table2
Masstransitionsm/z(Q1>Q3mass)andtheirrespectiveAZAs.
Masstransition AZA Collisionenergy
(CE)[V]
716>698 AZA-33 40
816>798 AZA-34,AZA-39 40
816>348 AZA-39 70
828>810 AZA-3 40
828>658 AZA-3 70
830>812 AZA-35,AZA-38 40
830>348 AZA-38 70
842>824 AZA-1,AZA-6,AZA-40 40
842>672 AZA-1 70
842>348 AZA-40 70
844>826 AZA-4,AZA-5 40
846>828 AZA-37 40
846>348 AZA-37 70
854>846 AZA-41 40
854>670 AZA-41 70
856>838 AZA-2 40
856>672 AZA-2 70
858>840 AZA-7,AZA-8,AZA-9,AZA-10,AZA-36 40
858>348 AZA-36 70
860>842 Undescribed 40
872>854 AZA-11,AZA-12 40
920>840 CompoundBinGuetal.(2013) 40 920>348 CompoundBinGuetal.(2013) 70 928>910 CompoundCinGuetal.(2013) 40 928>348 CompoundCinGuetal.(2013) 70
Threestrains(AZDY06,AZDH56andAZDH39)producedAZA-2 and AZA-36. Allthree strainsdisplayed a stable AZA-2/AZA-36 ratio, which consists of 6–6.5 times more AZA-2(7.6, 1.8 and 3.9fgcell1)thanAZA-36(1.2,0.3and0.6fgcell1,expressedas AZA-1equivalents).
StrainAZDH38producedanothernewAZAwithamolecular mass of 853Da at a cell quota of 0.9fgcell1, here named AZA-41.AZA-41belongstoanewgroupofAZAs,asall typical AZA group fragments of its collision induced dissociation (CID)spectrum are shiftedto 2Dalower m/z values. The sum formula of AZA-41 was determined as C48H72NO12 by HRMS (Table3).
Twostrains(AZLY01andAZCH10)producedAZA-11(1.0and 0.2fgcell1, respectively) and AZA-36 (0.4 and 0.7fgcell1, respectively)andstrainAZFC21producedAZA-11andAZA-40at cellquotasof0.5and0.2fgcell1,respectively.Insummary,these 16strainsdisplayedeightdifferentAZAprofilesincludingthenon- toxigenicvariant.
Inaddition,someof thestrains(AZCH10,AZDH39,AZDH56, AZDY06) produced minor amounts of an AZA variant with a molecular mass of 919Da previously observed in this species (compoundBinGuetal.,2013).However,innoneofthestrains thiscompoundexceeded1%ofthemajorAZA.
4. Discussion 4.1. Toxinprofiles
Guetal.(2013)reportedthreedifferentAZAprofilesoutofsix strains of A. poporum that previously had been isolated from Chinese coastalwaters.This is noteworthybecause it wasfirst indicationthatvariabilityofAZAsinthisspeciesseemedtobehigh comparedtothecloselyrelatedspeciesA.spinosum,whichtodate among allfouravailable strainsdisplayed a uniqueAZA profile Fig.1.RelativeAZA-profilesoftheChineseA.poporumstrains:AZA-41:blackbars,AZA-40:whitebars,AZA-36:darkgreybars,AZA-11:lightgreybars,AZA-2:stripedbars
( ).
Fig.2.Collisioninduceddissociation(CID)spectrumofAZA-40.
Fig.3. Collisioninduceddissociation(CID)spectraofcompound871Da ofA.
poporum(top)andAZA-11(bottom).
consistingonlyofAZA-1,-2andAZA-33(AZAwithmolecularmass of 715Da,Tillmann et al., 2012; Kilcoyne, pers. comm.). With respecttotoxinprofileinA.poporum,strainsfromtheNorthSea producedAZA-37(compound2inKrocketal.,2012),A.poporum fromthewest coastofKoreaproducedAZA-36(compound1in Krocketal. (2012)), and, asmentioned before, sixstrainsof A.
poporumfromChinesecoastalwaters(Guetal.,2013)produced AZA-2,AZA-36(compound1)andanAZAwithamolecularmassof 871Da(compoundAinGuetal.,2013),whichwenowidentified asAZA-11(Table4).Both,retentiontimes(datanotshown)and CIDspectraofAZA-11andthecompoundwiththemolecularmass of 871Da in A. poporum are identical (Fig. 3). Although this structure has already been proposed based on mass spectral interpretation,theunambiguousidentificationasAZA-11wasstill pending.
TheoccurrenceofAZA-11isparticularlyinteresting,because untilnowithasonlybeenregardedasa shellfishmetaboliteof AZA-2,whichis formedbyenzymatic hydroxylationinbivalves (Jamesetal.,2003).HereweclearlyshowthatAZA-11,apartfrom beingshellfishmetabolite,alsoisdenovosynthesizedbyatleast onespeciesofdinoflagellates.
Amongthe16strainsofA.poporuminvestigatedinthiswork, wefoundeightdifferentprofileswithdifferentcombinationsofthe alreadyobservedAZAs,butalsonewones.
4.2. AZA-40
Threestrains(AZBH01,AZBH03andAZFC21)gaveapeakinthe SRMmeasurementsintheiontransitionforAZA-1m/z842>824 almostattheretention timeof AZA-1.However, thequalifying transitionm/z842>672wasmissing,questioningtheidentityof AZA-1. For this reason we recorded a CID spectrum of this compoundandinfactitturnedoutnottobeAZA-1,butanisomer ofAZA-1belongingtothe348-type,which previouslyhasbeen described in strains of Azadinium poporum and Amphidoma languida.The348-typeAZAshaveamethyl(ormethylene)group lessinthepartofthemoleculeconsistingofC33-C39(ringsH,I)in comparisontotheclassical362-typeAZAs.Thisisalsoconfirmed byHRMS,astheexactmassesoftheAZAgroupfragmentsmatch the theoretical values (Table 3). Even though the chemical structure of AZA-40 cannot be assigned by mass spectrometry alone but needs confirmation by nuclear magnetic resonance (NMR)spectroscopy,this findingisimportantin ordertoavoid misinterpretation of AZA-1 in the presence of AZA-40. Both
compounds have identical molecular masses and, in addition, almost identical retention times. Under our chromatographic conditionsAZA-1waseluting at12.09minand AZA-40slightly earlierat12.03min.Thisisespecially critical,becauseAZA-1is regulatedinmanycountriesincludingtheEuropeanUnionbyfood safetylegislation,butAZA-40,asanewcompound,obviouslyis not.
In2010,Yaoetal. (2010)reportedthepresenceofAZA-1in Chineseshellfish.Themodeofdetectiontheyusedincludedfast chromatographyandonlyoneiontrace.Theseconditionsarenot sufficientforthedistinctionofAZA-1andAZA-40.Giventhefact that todate there is no known producer of AZA-1 in Chinese waters, but we could show the presence of AZA-40 by its progenitorA.poporuminthisregion,itmayverywellbethecase thatAZA-1hasbeenmisidentifiedbyYao etal.However, more work is needed to fully understand the global geographic distributionofAZAs,butespeciallyinthewesternPacificregion.
4.3. AZA-41
StrainAZDH38displayedapeakintheionchromatogramsof transitionsm/z856>838andm/z856>672,whicharecharac- teristicforAZA-2.However,thispeakelutedat12.23min,slightly beforeAZA-2(Fig.5),whichelutedat12.48min.ACIDspectrumof m/z856showedtheidenticalfragmentsasAZA-2,butallgroup fragments consisted of three subsequent nominal masses, for example, the group 2 fragment m/z 672 consisted of three subsequent mass peaks with m/z 670, 671 and 672 (data not shown).Thisledtotheconclusionthatm/z856mightnotbethe pseudo-molecularionofthiscompound,butofanionwithtwo13C atomsincorporatedintoacompoundwithalowermolecularmass.
Accordingly,werecordedaCIDspectrumwithm/z854instead.The CIDspectrumofm/z854inturnlookedlikea‘‘true’’AZAspectrum (Fig.4)anditshowsthatallgroupfragmentions(downtogroup6 fragmentm/z168for362-typeAZAs)areshiftedto2Dasmaller masses(Table3).HRMSrevealedthatthesemassshiftscorrespond
Table4
Retentiontimes(formethoddetailsseeSection2.3.1)andpseudomolecularions ([M+H]+]ofAZAspresentintheChinesestrainsofA.poporumandAZA-1.
Azaspiracid Retentiontime[min] m/z([M+H]+)
AZA-36 10.84 858
AZA-11 11.13 872
AZA-40 12.03 842
AZA-1 12.09 842
AZA-41 12.23 854
AZA-2 12.48 856
Table3
exactmassesofAZA-1,AZA-40andAZA-41andtheircharacteristicfragments.
AZA-1 AZA-40 AZA-41
m/z Composition Observed Composition ppm Observed Composition ppm
Group1 842 C47H72NO12 842,5049 C47H72NO12 0.0 854,5050 C48H72NO12 0.2
Group2 672 C38H58NO9 658,3950 C37H56NO9 0.0 670,4262 C38H56NO9 0.0
Group3 462 C27H44NO5 448,3057 C26H42NO5 0.1 460,3057 C27H42NO5 0.0
Group4 362 C22H36NO3 348,2533 C21H34NO3 0.1 360,2533 C22H34NO3 0.1
Group5 262 C16H24NO2 248,1645 C15H22NO2 0.0 260,1645 C16H22NO2 0.1
Group6 168 C10H18NO 154,1227 C9H16NO 0.2 166,1227 C10H16NO 0.3
Fig.4.CIDspectrumofAZA-41.
tosumformulascontaining2Hatomslessthanthecorresponding 362-typefragments.Thisfindingisastrongindicationforadouble bondinthepartofthemoleculeconsistingofC33-C39(ringsH,I) incomparisontothe362-typeAZA-2.Thismeansthatbesidesthe 348-typeand362-type,withAZA-41thereisthirdclass,the360- typeclass.However,asinthecaseofAZA-40,theexactstructureof AZA-41needstobeelucidatedbyNMRspectroscopyandwillbe subjectoffuturework.
As in the case of AZA-40, the occurrence of AZA-41 is noteworthy, because there is the potential of misidentifying
AZA-41asAZA-2,duetothefactthatbothcompoundseluteina narrowretentiontimewindowandgivesignalsinthesameMS/MS transitions, because of the naturally occurring stable carbon isotope13C,eventhoughbothcompoundsdifferintheirmolecular masses in two nominal mass units. AZA-41 with two 13C incorporatedhasthesamemolecularionandfragmentsasonly
12CcontainingAZA-2.LikeinthecaseofAZA-1andAZA-40,alsoin thecaseofAZA-2andAZA-41therearetwostructurallydifferent AZAs easy to beconfused, but one is regulated by food safety legislation(AZA-2)andtheother,obviousforanewcompound,is not(AZA-41).
4.4. Geographicdistribution
Combiningthepresentdatasetwithpreviouslypublisheddata (Gu etal., 2013), wenow haveAZA-profiles of 22 strainsofA.
poporumfromChinesecoastalwaters(Table1),whichallowan evaluation ofpotentialdistributionpattern(Fig.6). AZA-11and AZA-36 werethemost widelydistributed AZAs. They occurred fromtheBohai/YellowSeainthenorth,throughtheEastChinaSea totheSouthChinaSea.AZA-2wasonlydetectedinstrainsfromthe EastChinaSeaandtheSouthChinaSea,whereasAZA-40wasonly detectedintwostrainsbothfromtheSouthChinaSea,andAZA-41 onlyintheEastChinaSea,albeitinjustonestrain.Strainswithout anyAZAswereonlyfoundintheYellowSeaandtheEastChinaSea.
In general, there seemsto beno clear trend in thegeographic distributionof neitherthetwo ribotypesofA. poporumnorthe differentAZAvariantsinthedifferentstrainsofA.poporum.Both ribotypeswerefoundalongtheentireChinesecoastlinesuchas AZA-11andAZA-36.Theseeminglymorerestrictedoccurrenceof Fig.5.LC–MS/MS selected ionchromatograms(m/z856>838)of AZA-2and
AZA-41.
Fig.6.Ribotypes,relativeAZAprofilesandgeographicoriginoftheA.poporumstrainsusedinthisstudy.
non-toxigenic strains, AZA-2, AZA-40 and AZA-41 along the Chinesecoastmight simplybe due tothestill limited number ofisolates. In any case, theAZA profilesof thewestern Pacific isolatesofA.poporumhaveaninterestingfeature:Whereas the other known AZA-producing strains of other species either exclusivelyproduce39-methyl-AZAs (362-type AZAs) as in the caseoffourstrainsofA.spinosum (Tillmann etal.,2012;Krock etal.,2013)orexclusively39-desmethyl-AZAs(348-typeAZAs)as inonestrainofAmphidomalanguida(Krocketal.,2012),bothtypes, namelythe348-typeAZAsAZA-36,AZA-40andAZA-41andthe 362-typeAZAsAZA-2andAZA-11,whichwerefoundinA.poporum fromthenorthwesternPacific,wereevensimultaneouslypresent insomeisolates.
5. Conclusions
WhereasthefirstspeciesidentifiedasprogenitorofAZAsand AZP,A.spinosum,seemstohaveastabletoxinprofileconsistingof AZA-1,-2,and-33,themorerecentlyidentifiedsourceorganism forAZAs,A.poporum,displaysa muchhighervariability ofAZA profiles. Toxin profiles do not only differ among different geographicregionssuchastheNorthEastAtlanticandtheNorth WestPacific,butalsoamongpopulationsofthesamegeographic origin.In 16 isolates of A. poporum,we found8 different AZA profiles including non-toxigenic strains without any AZAs. In additionwefoundtwonewAZAs,AZA-40andAZA-41.AZA-40is anisomerofAZA-1andhavinganalmostidenticalretentiontime asAZA-1,it maybeeasilymisidentifiedasAZA-1.Eventhough AZA-41differsfromAZA-2ina2Dalowermolecularmass,AZA-41 alsocanbeeasilymisinterpretedasAZA-2,duetothesignificant proportionofthedouble13Csignal.AsAZA-1and-2areregulated, but not AZA-40 and AZA-41, it is particularly important to distinguishbetweenthesetwopairsofAZAs.Forafullevaluation of potential risks of such a misidentification for shellfish consumers,toxicitydataofthenewcompoundsareneeded.
Acknowledgements
WethankJaneKilcoyne,MarineInstitute,Oranmore,Galway, Irelandfor thekinddonationofan AZA-11containingsolution.
FinancialsupportwasprovidedbythePACESresearchprogramof theAWIaspartoftheHelmholtzFoundationinitiativeinEarthand Environment.This study(Grant-Aid Agreement No.PBA/AF/08/
001(01))waspartiallycarriedoutundertheSeaChangestrategy withthesupportoftheMarineInstituteandtheMarineResearch Sub-ProgrammeoftheNationalDevelopmentPlan2007–2013,co- financedundertheEuropeanRegionalDevelopmentFund.[SS].
References
Adachi,M.,Sako,Y.,Ishida,Y.,1996.Cross-reactivityoffluorescentDNAprobesto isolatesofthegenusAlexandriumbyinsituhybridization.In:Yasumoto,T., Oshima,Y.,Fukuyo,Y.(Eds.),HarmfulandToxicAlgalBlooms.IOC-UNESCO, Paris, pp.455–458.
Akselman,R.,Negri,R.M.,2012.BloomsofAzadiniumcf.spinosumElbra¨chteret Tillmann(Dinophyceae)innorthernshelfwatersofArgentina,Southwestern Atlantic.HarmfulAlgae19,30–38.
Altschul,S.F.,Madden,T.L.,Schaffer,A.A.,Zhang,J.,Zhang,Z.,Miller,W.,Lipman,D.J., 1997.GappedBLASTandPSI-BLAST:anewgenerationofprotein database searchprograms.NucleicAcidsRes.25,3389–3402.
Bran˜ aMagdalena,A.,Lehane,M.,Krys,S.,Ferna´ndez,M.L.,Furey,A.,James,K.J., 2003.ThefirstidentificationofazaspiracidsinshellfishfromFranceandSpain.
Toxicon42,105–108.
Gu,H.,Luo,Z.,Krock,B.,Witt,M.,Tillmann,U.,2013.Morphology,phylogenyand azaspiracidprofileofAzadiniumpoporum(Dinophyceae)fromtheChinaSea.
HarmfulAlgae21-22,64–75.
Guillard,R.R.L.,Ryther,J.H.,1962.Studiesofmarineplanktonicdiatoms,I.Cyclotella nanaHustedtandDetonulaconfervaceaCleve.Can.J.Microbiol.8,229–239.
James,K.J.,Furey,A.,Lehane,M.,Ramstad,H.,Aune,T.,Hovgaard,P.,Morris,S., Higman,W.,Satake,M.,Yasumoto,T.,2002.Firstevidenceofanextensive northernEuropeandistributionofazaspiracidpoisoning(AZP)toxinsinshell- fish.Toxicon40,909–915.
James,K.J.,Sierra,M.D.,Lehane,M.,Bran˜ aMagdalena,A.,Furey,A.,2003.Detection offivenewhydroxylanaloguesofazaspiracidsinshellfish usingmultiple tandemmassspectrometry.Toxicon41,277–283.
Jauffrais,T.,Herrenknecht,C.,Se´chet,V.,Sibat,M.,Tillmann,U.,Krock,B.,Kilcoyne,J., Miles,C.O.,McCarron,P.,Amzil,Z., etal.,2012.Quantitativeanalysisofazaspir- acidsinAzadiniumspinosumcultures.Anal.Bioanal.Chem.403,833–846.
Klontz, K.C., Abraham, A., Plakas,S.M., Dickey, R.W., 2009. Mussel-associated azaspiracidintoxicationintheUnitedStates.Ann.InternalMed.150,361.
Krock,B.,Tillmann,U.,John,U.,Cembella,A.D.,2009.Characterizationofazaspir- acids in planktonsize-fractionsand isolationof anazaspiracid-producing dinoflagellatefromtheNorthSea.HarmfulAlgae8,254–263.
Krock,B.,Tillmann,U.,Voß,D.,Koch,B.P.,Salas,R.,Witt,M.,Potvin,E´.,Jeong,H.J., 2012.NewazaspiracidsinAmphidomataceae(Dinophyceae).Toxicon60,830–
839.
Krock,B.,Tillmann,U.,Alpermann,T.J.,Voß,D.,Zielinski,O.,Cembella,A.D.,2013.
PhycotoxincompositionanddistributioninplanktonfractionsfromtheGer- manBightandwesternDanishcoast.J.PlanktonRes.35,1093–1108.
Lo´pez-Rivera,A.,O’Callaghan,K.,Moriarty,M.,O’Driscoll,D.,Hamilton,B.,Lehane,M., James,K.J.,Furey,A.,2010.Firstevidenceofazaspiracids(AZAs):Afamilyof lipophilicpolyethermarinetoxinsinscallops(Argopectenpurpuratus)andmus- sels(Mytiluschilensis)collectedintworegionsofChile.Toxicon55,692–701.
McMahon,T.,Silke,J.,1996.WestcoastofIreland;wintertoxicityofunknown aetiologyinmussels.HarmfulAlgaeNews14,2.
Ne´zan,E.,Tillmann,U.,Bilien,G.,Boulben,S.,Che`ze,K.,Zentz,F.,Salas,R.,Chomerat, N.,2012.TaxonomicrevisionofthedinoflagellateAmphidomacaudata:transfer tothegenusAzadinium(Dinophyceae)andproposaloftwovarietiesbasedon morphologicalandmolecularphylogeneticanalyses.J.Phycol.48,925–939.
Nikolaev,E.N.,Boldin,I.A.,Jertz,R.,Baykut,G.,2011.Initialexperimentalcharac- terizationofanewultra-highresolutionFTICRcellwithdynamicharmoniza- tion.J.Am.Soc.MassSpectrom.22,1125–1133.
Ofuji,K.,Satake, M.,McMahon, T., Silke,J.,James, K.J.,Naoki,H.,Oshima, Y., Yasumoto,T.,1999.Twoanalogsofazaspiracidisolatedfrommussels,Mytilus edulis,involvedinhumanintoxicationinIreland.Nat.Toxins7,99–102.
Percopo,I.,Siano,R.,Rossi,R.,Soprano,V.,Sarno,D.,Zingone,A.,2013.Anew potentiallytoxicAzadiniumspecies(Dinophyceae)fromtheMediterraneanSea, A.dexteroporumsp.nov.J.Phycol.,http://dx.doi.org/10.1111/jpy.12104.
Potvin,E´.,Jeong,H.J.,Kang,N.S.,Tillmann,U.,Krock,B.,2012.FirstReportofthe photosyntheticdinoflagellategenusAzadiniuminthePacificOcean:morphol- ogyandmolecularcharacterizationofAzadiniumcf.poporum.J.Eukaryotic Microbiol.59,145–156.
Salas,R.,Tillmann,U.,John,U.,Kilcoyne,J.,Burson,A.,Cantwell,C.,Hess,P.,Jauffrais, T.,Silke,J.,2011.TheroleofAzadiniumspinosum(Dinophyceae)intheproduc- tionofazaspiracidshellfishpoisoninginmussels.HarmfulAlgae10,774–783.
Satake,M.,Ofuji,K.,Naoki,H.,James,K.J.,Furey,A.,McMahon,T.,Silke,J.,Yasumoto, T.,1998.Azaspiracid,anewmarinetoxinhavinguniquespiroringassemblies, isolatedfromIrishmussels,Mytilusedulis.J.Am.Soc.MassSpectrom.120, 9967–9968.
Taleb,H.,Vale,P.,Amanhir,R.,Benhadouch,A.,Sagou,R.,Chafik,A.,2006.First detectionofazaspiracidsinmusselsinnorthwestAfrica.J.ShellfishRes.25, 1067–1070.
Tillmann,U.,Elbra¨chter,M.,Krock,B.,John,U.,Cembella,A.D.,2009.Azadinium spinosumgen.etspnov(Dinophyceae)identifiedasaprimaryproducerof azaspiracidtoxins.Eur.J.Phycol.44,63–79.
Tillmann,U.,Elbra¨chter,M.,John,U.,Krock,B.,Cembella,A.D.,2010.Azadinium obesum(Dinophyceae),anewnontoxicspeciesinthegenusthatcanproduce azaspiracidtoxins.Phycologia49,169–182.
Tillmann,U.,Elbra¨chter,M.,John,U.,Krock,B.,2011.Anewnon-toxicspeciesinthe dinoflagellategenusAzadinium:A.poporumsp.nov.Eur.J.Phycol.46,74–87.
Tillmann,U.,Soehner, S.,Ne´zan,E.,Krock,B.,2012. Firstrecordofthegenus Azadinium(Dinophyceae)fromtheShetlandIslands,includingthedescription ofAzadiniumpolongumsp.nov.HarmfulAlgae20,142–155.
Yao, J.,Tan, Z., Zhou, D.,Guo, M., Xing, L.,Yang, S.,2010. Determination of azaspiracid-1inshellfishbyliquidchromatographywithtandemmassspec- trometry.Chin.J.Chromatogr.28,363–367.
