Wastewater effects on Phaeodactylum tricornutum (Bohlin): Setting up a classification system

(1)

ContentslistsavailableatScienceDirect

Ecological

Indicators

jou rn 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 / e c o l i n d

Wastewater

effects

on

Phaeodactylum

tricornutum

(Bohlin):

Setting

up

a

classiﬁcation

system

G. Libralato

a,∗

_,

_E.

_Gentile

b

_,

_A.

_Volpi

_Ghirardini

a

a_Department_of_{Environmental}_Sciences,_Informatics_and_Statistics,_University_Ca’_Foscari_Venice,_Campo_della_Celestia,_2737/b,₃₀₁₂₂_Venice,_Italy b_Department_of_Biosciences,_{Biotechnology}_and_{Biopharmaceutics,}_University_of_Bari,_Via_E._Orabona,_4,_70124,_Bari,_Italy

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received5November2014 Receivedinrevisedform10June2015 Accepted13June2015 Keywords: Wastewater Toxicity Biostimulation Microalgae Effectscore

a

b

s

t

r

a

c

t

Phytoplanktonishighlyproductiveinmarinecoastalecosystemsthatgenerallypresenthighlevels ofanthropogenicpressures.Microalgaepresentsveryhighsurface-to-volumeratioandmayrespond promptlytocontaminantsshowingeitheranincreasein growthorinhibitioneffects. Thus phyto-planktonorganismsprovideinformationonthepotentialimpactsofcontaminantsonthesupported marine-coastalfoodwebs.ThediatomPhaeodactylumtricornutumBohliniscommonlyusedin toxic-itytestingaccordingtotheISO10253:2006standardisedprotocol,butacomprehensiveinventoryof testedsubstanceshasnotbeencompiledyet.Theaimofthisstudyistoestablishawastewatereffect scorebasedonP.tricornutumexposedtodomestic,municipalandindustrialwastewatersfromactivated sludgesequencingbatchreactorandultra-ﬁltrationmembranebiologicalreactors.Wastewatersamples producedstimulationandinhibitioneffectsidentiﬁedbybiostimulationunit(BU50)andtoxicityunit

(TU50)bothat50%effect,respectively.Withinthestimulationscenario,toxicitywaslowif0<BU50≤0.31,

mediumif0.31<BU50≤1.05,highif1.05<BU50≤1.64,andveryhighifBU50>1.64.Withintheinhibition

scenario,toxicitywaslowif0<TU50≤0.07,mediumif0.07<TU50≤2.67,highif2.67<TU50≤5.86,and

veryhighifTU50>5.86.Resultsevidencedthatnitrogenandphosphorusconcentrationswerenot

corre-latedtoecotoxicologicalvalues,probablyduetothepresenceofundetectedmicronutrients,conﬁrming theimportanceoftoxicity-basedhazardassessment.

1. Introduction

OneoftheobjectivesoftheWaterFrameworkDirective(WFD)

(60/2000/EC)is tointegrate national and internationalpolicies

andactionsinordertoachieveagoodenvironmental(i.e.

chem-icalandecological)statusofEuropeanwaterbodies(freshwater,

andtransitionalandcoastalwaterswithin1mile)by2015;local

assessmentstrategiesandinterventionsarestrictlylinkedtothis

(Andersenetal.,2010).Thetreatmentandmonitoringof

wastewa-terpointsourceshasbecomeincreasinglyimportant,particularly

topreventexcessdischargeoftoxicsubstancesandnutrients(e.g.

nitrogenandphosphorus)intheaquaticsystemsthatcanaffectthe

structureandfunctioningofecologicalcommunities.TheNitrates

Directive(91/676/EEC)regulateswaterqualityacrossEuropeby

preventing discharge of nitrates from non-point sources and

promotingtheuseofgoodfarming practices.The thresholdfor

∗ Correspondingauthorat:DepartmentofEnvironmentalSciences,Informatics andStatistics,UniversityCa’FoscariVenice,CampodellaCelestia2737/b,30122 Venice,Italy.Tel.:+390412347737;fax:+390415281494.

E-mailaddress:giovanni.libralato@unive.it(G.Libralato).

theidentiﬁcationofpollutedwateroratriskofpollutionwasset

at50mgN-NO3−L−1forfreshwaterbodies,estuaries,andcoastal

andmarinewaters.Also,theEuropeanMarineStrategyFramework

Directive(56/2008/EC)requiredMemberStatestoachievegood

environmentalstatus(GES)inmarinewaterstoprevent

eutrophi-cation.

AccordingtotheEuropeanInventoryof ExistingCommercial

ChemicalSubstances(2013),pointsourcescancontainmorethan

100,000substancesandtheirrelativeby-products.Oftenthe

infor-mationabouttheidentityof compoundspresentinwastewater

islimitedaswellastheirpossiblebiologicaleffectssuchas

tox-icity,genotoxicity or estrogenic potential (Newman and Unger,

2003).Thesameoccursfornutrientsconcentration,whichvaryin

accordancewiththedensityofhumanactivities(e.g.inthe

catch-mentbasinoralongthecoast),locationsandrouteofdischarge.

Nutrient enrichment can lead to nuisance algal blooms, under

appropriatetemperatureandlightconditions(NewmanandUnger,

2003).Thecontrolofdischargeshasbeentraditionallycarriedout

usingmethodsthatprovidenon-speciﬁcresponsesandthrough

themeasurementofglobalparameters(dissolvedorganiccontent,

chemicaloxygendemand)providinglimitedinformationon

bio-logicaleffects(FarréandBarceló,2003).Thetraditionalapproach,

(2)

basedonchemical analyses,adopted forassessing the

environ-mentalimpactofdischargesisthereforenotappropriate,andan

integrated/extendedassessment of wastewatereffects is

essen-tialtoascertainpotentialhazard(Munawaretal.,1989).Several

authors(BurkhardandDurhan,1991;Fernándezetal.,1995;Tothill

and Turner, 1996)recommended combining differentstandard

chemicalanalysesoftargetcompoundsfortoxicityassaysfor

iden-tifyingthemaincomponentsoftoxicefﬂuentsandsludge.Atoxicity

testcan bedeﬁnedasa biological assayperformedtomeasure

theeffectsofasubstanceoranenvironmentalmatrix(leachate,

elutriate,contaminatedsediment/soil)onalivingorganism,

pro-vidingaqualitativeorquantitativeoutput(WhadhiaandThomson,

2007).Toxicityprovidesanintegratedmeasureoftheimpactof

wastewater combining differentfactors suchas pH, compound

solubilityandbioavailabilityaswellasthepotentialformixture

effect(FarréandBarceló,2003;Libralatoetal.,2010a).Duetothe

impossibilityofdeterminingthespeciﬁctoxicityofeach

compo-nentof a complex efﬂuent, the WholeEfﬂuent Toxicity(WET),

usingaquaticorganisms(bioassay),wasproposedasadirect

suit-ablemethod,relativelyinexpensive,forthetoxicitydetermination

(USEPA,2004).Bioassayscanalsomonitorpollutantsthatarenot

detectedbycommonanalytical approaches(Tothill andTurner,

1996).WhadhiaandThomson(2007)pointedoutthattheuseof

biologicalassaysisaholisticapproachthatallowstheassessment

oftoxicityandthetotaleffectofallcomponents,including

possi-bleadditive,synergisticandantagonisticeffects.Livingorganisms

integratethepositiveandnegativeeffectsofchemicalswithwhich

theycomeintocontactwiththeenvironmentalconditionsthey

areexposedtoduringtheexperimentrespondingtothe

biologi-callyactivecomponentspresent(Keddyetal.,1995).Wastewater

samplescanbetestedusinganylevelofbiologicalorganisation

fromthemolecularleveluptothewholeorganism,populationor

community(Calow,1989).Todate,theassessmentoftoxicityof

wastewaterhasbeencarriedoutusingdifferenttypesofassays

(e.g.Daphniamagna24hacutetest),butgapsstillremainabout

howto use,interpret and integratetheirresults. Sometoxicity

toolsalreadyexistforassessing(ToxicityIdentiﬁcationEvaluation

andToxicityReductionEvaluation)(USEPA,2004)andrankingof

thepotentialecotoxicityofwastewatersamplessuchasthose

pro-posedbyBulich(1982),Callejaetal.(1986),Ross(1993),Costan

etal.(1993),SwedishEPA(1997),Tonkesetal.(1999),Vindimian etal.(1999),Sarakinosetal.(2000),Phillipsetal.(2001),Persoone etal.(2003)andLibralatoetal.(2010b).Atthesametime,onlyfew

authorsmadeexplicitreferencetophytotoxicityand

phytostim-ulationdichotomyforenvironmentalqualityassessment (Sbrilli

etal.,2005).Generally,phytostimulationcanadverselyaffect

envi-ronmentalqualitysimilarlytophytotoxicity.Onlyfewsaltwater

phytoplanktonicbiologicalmodelsarereallywidespreadand

stan-dardisedfortoxicitytestingsuchasthebenthicunicellulardiatom

PhaeodactylumtricornutumBohlin(Mulleretal.,2007).Thisalga

ispresentintransitional,marine-coastalandmarinewaters(Kim

etal.,2004;Rechetal.,2005;Toepeletal.,2005)anditsusein

ecotoxicologyisstandardisedwithinISO(2006).Severalauthors

(PANPesticideDatabase,2013)consideredP.tricornutumasa

ref-erencebiologicalmodelforpureinorganicandorganicchemicals

witharecordof524testedsubstances.P.tricornutumwasalsoused

toassessemergingcontaminantssuchasengineered

nanomateri-als(Clémentetal.,2013;Morellietal.,2013)andpharmaceuticals (Claessensetal.,2013),surfaceandwatercolumnsamples(Okay etal.,1998;Macovaetal.,2010;Moreira-Santosetal.,2006a;Shaw

etal.,2009),ligninandtannin(Libralatoetal.,2011),whole

sedi-mentandothersedimentrelatedsamples(elutriate,porewater,

water fromthe sediment–water interface) (Puddu et al., 1988;

Tolunetal.,2001;Muchaetal.,2003;Moreira-Santosetal.,2006b; Moreno-Garridoetal.,2007a, 2007b; Morelliet al.,2009).

Var-ious other complex liquid environmental specimens were also

evaluatedsuchasdomestic,municipalandindustrialwastewaters

(Clarksonetal.,1998,1999;Petersetal.,2002;Moseretal.,2004;

Okayetal.,2005).ThemorphologyandgenomeofP.tricornutum

arealsowellknownthusithasfrequentlybeenusedasamodel

diatominstudiesofalgalphysiologyandecology(Bowleretal.,

2008),andofheavymetalstransportationandtransformationin

oceanecosystems(Morellietal.,2009;Dengetal.,2013).

ThispaperevaluatestheabilityofP.tricornutumin

discriminat-ingthequalityofdomestic,municipalandindustrialwastewater

samplesinordertodevelopandcalibrateaspecies-speciﬁceffect

scoreintegrating Libralatoetal.(2010b, 2010c). Therankingof

wastewatersamplesis essentialfor identifyingpotential

effect-based hazards for receiving water bodies and for undertaking

adequatemanagementactionsconsideringbothtoxicityand

bios-timulationevents.

2. Materialandmethods

2.1. Wastewatersamples

A total of93 sampleswere collectedfromthree small-scale

decentralisedwastewatertreatmentplants(WWTPs),locatedin

Venice(Italy)historicalcentre(Libralatoetal.,2009c,2012).

Sam-plesincludeddomestic(n=25),municipal(n=59)andindustrial

(n=9) wastewaters. WWTP technologies included an Activated

SludgeSequencingBatchReactor(AS-SBR)andtwoUltra-Filtration

MembraneBiologicalReactors(UF-MBR1andUF-MBR-2).Details

of theWWTPs (AS-SBR, UF-MBR1 and UF-MBR2) canbe found

inLibralatoetal. (2010c).Sampleswerelabelledin accordance

totheirorigin(A=domestic,B=municipalandC=industrial).Few

hours after their manual collection, samples were adjusted to

thesalinity ofthereceivingwater body (34‰)(Libralatoetal.,

2009a).Duringsampling,dissolvedoxygen(DO),pHand

temper-aturewererecorded,andsampleswerestoredincompletelyﬁlled

polyethyleneterephthalatecontainers.Somealiquotswereused

for physico-chemicalinvestigations and stored at4±1◦C for a

maximumof36h,whiletheremainingsampleswereusedfor

eco-toxicologicalanalysesand storedat −18◦_C_±₁◦_C ₍_OSPAR, _2000,

2005,2007;Libralatoetal.,2009b).

2.2. Physico-chemicalanalyses

ThepHwasmeasuredvia pHmeterHI 9025Microcomputer

(HANNAInstrument,Beverly,MA,USA),thesalinitywaschecked

witharefractometer(Atago,Japan)andtheDObyaWTW

multi-parametric device (Nova Analytics, Weilheim, Germany). The

determination of ionicspecies, chloride (Cl−), nitrite (N-NO2−),

nitrate(N-NO3−),ammonia (N-NH4+), phosphate(P-PO43−)and

sulphate(S-SO42−)wasperformedusing anIonChromatograph

(IC)system(column Metrohm Metrosep A Supp 5150– 4mm,

Metrohm761CompactIC,Switzerland)accordingtoAPHA(1998)

methods.ChemicalOxygenDemand(COD),TotalKjeldahl

Nitro-gen (TKN), total phosphorus (Ptot), raw wastewater Suspended

Solids (SS) were analysed according to APHA (1998) methods.

Thedetermination of metaland metallicelementssuchas

alu-minium (Al), arsenic (As), barium (Ba), cadmium (Cd), cobalt

(Co), total chromium (Cr), copper (Cu), iron (Fe), manganese

(Mn), nickel (Ni), molybdenum (Sb), selenium (Se), vanadium

(V) and zinc (Zn) was carried out according to USEPA (1992)

andAPHA(1998)methodsusinganInductivelyCoupledPlasma

Optical Emission Spectroscopy (ICP-OES Spectroﬂame Compact

E,Spectro AnalyticalInstruments, Kleve,Germany).AnICP-OES

multi-element standard solution (Merck 10580) was used for

calibrationandQualityAssurance/QualityControl(QA/QC)

proce-dures.OnlyUF-MBR2sampleswerecheckedformetalandmetallic

(3)

2.3. Toxicitytest

Toxicity tests with P. tricornutum were carried out

accord-ing to the ISO 10253 method (ISO, 2006) determining the

growthinhibition (or biostimulation)of microalgae exposed to

increasing dilutions of samples. The algal culture was kept at

20±2◦C and 6000–10,000lux, obtaining a cellular density of

more than 106_cells_mL−1_. _The _initial _algal _density _in _the _test

wasobtainedbydilutingthealgal cultureandranged between

2×103_cells_mL−1_and₁₀4_cells_mL−1_._P._tricornutum_was_exposed

toincreasingconcentrationsof samplesfor 72±2hat 20±1◦C

and6000–10,000lux.Negativeandpositive(K2Cr2O7,BDHProlabo,

CAS#7778-50-9)controlswereincludedineachexperiment.

Cel-lulardensitywasevaluatedbyopticalmicroscopyinbrightﬁeld

(40×)usingaBürkercountingchamber.Eachexperimentincluded

atleastanegativecontroland6sampledilutions(5,10,30,50,

70and90%wastewater/volume(w/v))intriplicate.Wastewater

samplesweredilutedwithISO(2006)artiﬁcialseawater.

2.4. Dataanalysis

Toxicityand stimulation effectdata normalised onnegative

controls(Abbott’s formula) were determinedas percentages of

inhibition/stimulation effect inducing a 50% growth inhibition

(inhibitionconcentration,IC50)or50%growthstimulation

(bios-timulationconcentration,BC50)intheobservedpopulation.Data

wereexpressedaspercentageofeffectatthehighestw/v(%)after

normalisationtotheaverageeffectinthenegativecontrols.

Neg-ative values ofPE indicated stimulation, whereaspositive ones

showedtoxicity.Negativeandpositivecontrolswerecomparedto

ISO(2006)thresholdforacceptability.Wheneverpossible,IC50

val-uesweretransformedintoxicityunits(TU50=100/IC50).Negative

effectdataindicatinggrowthstimulationweretreatedsimilarly

determiningthebiostimulationunits(BU50=100/BC50).IfIC50or

BC50 could not be determined,TU50 or BU50 were determined

accordingtoTU(orBU)50=50/(percentageofeffect(PE))where

PE<50%.

The hypothesis test wasveriﬁed using Analysis of Variance

(ANOVA) and Tukey’s test to check any difference among the

groups.WhenANOVArevealedsigniﬁcantdifferencesamong

treat-ments, post hoctests were carried on with Dunnett’smethod

testingthepairwisedifferencebetweeneachtreatmentandthe

control. Parametric methods wereconsidered for experimental

points’estimation.Allresultsarepresentedasmeanswiththe

rel-ative standard errorusingfor allstatistical analysisthedefault

5% rejection level. Statistical analyses were performed using

Microsoft®_Excel_2013/XLSTAT©_-Pro_(Version_7.2,_2003,_Addinsoft,

Inc.,Brooklyn,NY,USA).

Univariate and multivariate (principal component analysis)

analyseswereconsideredtoassessphysico-chemicaland

ecotox-icologicaldatainordertocheckforpotentialcorrelations.

3. Resultsanddiscussion

3.1. Effectsofwastewaterspecimen

Within93-wastewaterspecimen,only17samplesshowedthe

abilitytoinhibitalgaegrowthandonlyonesample(C06)showedno

toxicityatall(SupplementaryMaterialsS1andS2).Theremaining

75samplesdisplayedastimulationeffectrangingbetween−2and

−102%;thiscanbeexplainedconsideringthatalgaegrowthdoes

notreachamaximumvaluepresentingthepotentialforan

unlim-itedcarrying capacityofthesystem.BU50 rangedbetween0.06

and2.04varyingtwoordersofmagnitude(S1andS2).Thealgae

growthinhibitionrangedbetween3and100%,whileTU50varied

twoordersofmagnitudebetween0.06and5.88.Amongst

domes-tic,municipalandindustrialwastewatersamplespresentingalgae

stimulation,theaverageeffectwasof74%(A,n=25),40%(B,n=47)

and36%(C,n=3),respectively.Concerningdomestic,municipaland

industrialwastewatersamplesshowingalgaeinhibition,the

aver-ageeffectwasof66%(A,n=2),43%(B,n=10)and67%(C,n=5),

inthatorder.Themosttoxicsamplesbelongedtodomesticand

municipalwastewatersnottotheindustrialones(S1andS2).

How-ever,allindustrialwastewatersamples,exceptforC06,presented

frommediumtohighlevelsofalgaegrowthinhibitionsuggesting

thatindustrialwastewateralwaysrepresentsapotential

toxicity-basedhazardforthereceivingwaterbody.Therangeoftoxicities

presentedbydomestic,municipalandindustrialwastewater

sam-plesallowedthedeﬁnitionandcalibrationofP.tricornutumeffect

scoreforwastewaterclassiﬁcation.

3.2. Physicalandchemicaldata

The univariate and multivariate statistical analyses showed

nosigniﬁcantcorrelationsbetweenphysico-chemicaland

ecotox-icological data (p>0.05) (Supplementary Materials S3 and S4).

Table1

WastewatereffectscoreforBU50(n=75)andTU50(n=18,includingthenoeffectsampleC06) fromP.tricornutum;alldatamustbeassessedafternegativecontrolnormalisation.

Endpoint Toxicity Score Effect Score

A lg ae growt h st imulat io n BU50 > 1.64 very high 4 1.05 < BU50≤ 1.64 high 3 0.31 < BU50≤ 1.05 medium 2 0 < BU50≤ 0.31 low 1 BU50 or TU50 = 0 absent 0 A lg ae growt h in h ibi tion 0 < TU50≤ 0.07 low 1 0.07 < TU50≤ 2.67 medium 2 2.67 < TU50≤ 5.86 high 3 TU50 > 5.86 very high 4

(4)

Table2

ResultsfromtheapplicationofP.tricornutumeffectscore;datawereplottedinS5;A=domestic, B=municipalandC=industrialwastewatersamples–theintegernumberindicatedtheprogression insamplecollection.

Stimulationeffectswerenotdirectlyrelatedtonitrogen(N-NH4+,

N-NO2− and N-NO3−) and phosphorus (P-PO43− and total P)

concentrationsthatwerepresentin excessalmostinall

waste-watersamples.Thusnitrogenandphosphorousdidnotrepresent

a limiting factor for the system carrying capacity. It is likely

that undetectedstimulatingor additiveagents werepresent in

wastewater samples like as micronutrients such as silica,

cal-cium,magnesium,potassium,iron,manganese,sulphur,andcobalt

potentiallystimulatingnitrogenﬁxation(Buesseleretal.,2004).

(5)

growth for biofuel production rarely limits algal growth when

wastewaterisused(Knud-Hansenetal.,1998).

3.3. P.tricornutumeffectscoreforwastewaterclassiﬁcation

A5-classeffectscorewasdeﬁnedrankingstimulationand

tox-icitydata separately, but in a symmetricalway. The systemof

effectscoreswascentredonthe“noeffect”rankasdisplayedin

Table1basedonasimpliﬁcationofthemethodreportedinLibralato etal.(2010b)forspecies-speciﬁcscores.Preliminarily,all

stimula-tion/toxicitydatawerenormalisedtotheaverageeffectonnegative

controlsbeforetheirassessment.Then,duetotheprecautionary

principle,itwasstatedthatonlywhenBU50orTU50wereequal

tozerothesamplewasdeemedaspresenting “noeffect”.Low,

medium,highandveryhightoxicityclasseswereorganisedto

clas-sifyalleffects(Table2andS5).Thisapproachavoidedtheuseofthe

minimumsigniﬁcancedistance(MSD)foreachtest-matrixpair

eas-ingthepreliminaryphasesofdataelaboration.Threelevelsofeffect

weredeterminedconsideringmultiplesofthe90th-percentileof

effects’distributionaccountingforstimulationandtoxicity

sep-arately.Inordertoreducetoa minimumtheexpertjudgement

andconsideringsomeprevioussimilarapproaches(Burton,2002;

Libralatoet al., 2010b), the choicewas addressedto the

10th-and 50th-percentilevalues of effectdata providing information

ontheprobabilityofeffectsrarelyorlikelytooccur(Leotsinidis

andSazakli,2008).The10th-,50th-and90th-percentileof

stimu-lationeffectdata(n=75)correspondedto0.31,1.05and1.64BU50,

respectively,whilethe10th-,50th-and90th-percentileoftoxicity

effectvalues(n=18,includingthe“noeffect”sample)were0.07,

2.67and5.86TU50,respectively.Finally,theeffectscorewas

summ-arisedinTable1.Anintegerscorenumberfrom0(noeffect)to

4(veryhigheffect)andacolouraccompaniedeachclassto

sum-mariseandrepresenttheobservedeffectinasimplevisualway.

For the stimulation scenario, if 0<BU50≤0.31 effect is low

(1 green), if 0.31<BU50≤1.05 effect is medium (2, yellow), if

1.05<BU50≤1.64effectishigh(3,orange),andifBU50>1.64effect

isveryhigh(4,red).WhenBU50<1,thewastewatersamplepresent

astimulationeffectthatis<50%(BU50=50/PE);valuesofBU50≥1

(BU50=100/BC50)canbedeemedasthedilutionfactorrequired

toobtaina50%algaegrowthstimulationinthetreatment.Thus,

highandveryhighstimulationratesoccurwhenitisrequiredto

dilutethesampleatleast1.05timestoobtaina50%algaegrowth

stimulationcomparedtothenegativecontrol.

For the toxicity scenario, if 0<TU50≤0.07 effect is low (1

green), if 0.07<TU50≤2.67 effect is medium (2, yellow), if

2.67<TU50≤5.86effectishigh(3,orange),andifTU50>5.86effect

isveryhigh(4,red).

IfTU50<1,thewastewatersamplepresentsatoxicityeffectthat

is<50%(TU50=50/PE);valuesofTU50≥1(TU50=100/EC50)canbe

deemedasthedilutionfactorrequiredtoobtaina50%algaegrowth

inhibitioninthetreatment.Thus,highandveryhightoxicityrates

occurwhenitisrequiredtodilutethesampleatleast2.67times

toobtaina50%algaegrowthtoxicitycomparedtothenegative

control.

The assessment of wastewater samples via the effect score

(Table2andS5)showed7,30,31and7sampleswithlow,medium,

highandveryhigheffects,respectively,amongstthe75-specimen

inducingalgaegrowthstimulation.Thus91%ofsamplespresenting

astimulationeffectbelongedtoarankfrommediumtoveryhigh.

The17samplesshowingalgaegrowthinhibitionpresented1,8,

7and1specimenwithlow,medium,highandveryhigheffects,

accordingly.Inthiscase,94%ofsamplespresentingatoxicityeffect

belongedtoarankfrommediumtoveryhigh.Onlyonesample,

C06,presentednoeffectatallcomparedtothenegativecontrol.In

general,themainpartofwastewatersamples(90%)showedfrom

mediumtoveryhighstimulationortoxicityeffects.

ThesameresultscanbediscussedinlightofTonkesetal.(1999)

andPersooneetal.(2003)toxicityscoresforwastewatersample

classiﬁcation,asreportedinLibralatoetal.(2010b).

TheoutputofTonkesetal.(1999)classiﬁcationsystem

identi-ﬁed3samplesnotacutelytoxic(i.e.>100%,w/v)and15moderately

acutelytoxic(i.e.10–100%,w/v).TheapplicationofPersooneetal.

(2003)rankingidentiﬁed5samplespresentingnoacutetoxicity,1

samplewithslightacutetoxicityand12sampleswithacute

tox-icity.Bothscoringsystemsdeemedasnotclassiﬁable75samples

duetotheirstimulationeffect.Excessinalgaegrowthstimulation

seemsnottobeconsideredbyTonkesetal.(1999)andPersoone

etal.(2003)asaclearundesirablephenomenon;thisisnotthecase

asshownbyalgalbloomsassociatedtoeutrophicationphenomena.

Actually,thesuggestedscoringsystemdoesnothavetheintent

toprovide a classiﬁcationof a wastewater sampleby anindex

oftrophic state;infact thelatter isgenerallyfocusedon

nutri-ents(nitrogenandphosphorus),chlorophyllconcentration,onsite

algalbiomassquantiﬁcationorspeciesdiversitydistribution,and

consideringthatacomplexinterplayamongfactorsthatinﬂuence

thetrophicstatecanoccur(Doddsetal.,1998;Dodds,2007).The

aimofthesuggestedscoringsystemistoclassifywastewater

sam-plesbytheintegratedapproachofferedbytoxicitytestsmainly

focusingonWholeEfﬂuentAssessment(WEA)(OSPAR,2005),thus

includingnotonlytoxicity(growthinhibition),butalsostimulation

effects.Theamountofstimulationeffectdetectedatthelaboratory

scale,andnotonaﬁeldbasisobservation,couldbeanalternative

measureonhowwastewaterinﬂuencesthetrophic stateofthe

receivingwaterbody.Thiscouldbeinterestingforretroactiveand

predictiveriskassessmentandforriskmanagementaswell.Within

thepresentcasestudy,forexample,theanalysisofNandP

concen-trationsandeffectdatadidnotevidenceanysigniﬁcantcorrelation

suggestingthateffectscanbedifﬁculttoforecastjustonthebasis

oftraditionalparameters.Thisresulthighlightstheimportanceofa

moreintegratedandholisticapproachinwastewatersample

haz-ardcharacterisationandassessmentthatshouldincludenotonly

contamination-baseddata,butalsotoxicity-basedissues.

4. Conclusions

Marinealgalbioassaysareviablemethodstoassessavariety

of environmental pollutants aswell as complex environmental

matricessuchaswastewaters.ToxicitytestswithP.tricornutum

showedtobeusefulnotonlytodetectgrowthinhibitioneffects,

butalsowastewatersinducingbiostimulationphenomena.Effect

datahavebeenassessedprovidingeffectscoresspeciﬁcfor

tox-icity and stimulation based on their percentile distribution of

effect. The P.tricornutum toxicity score is an opensource tool

that showedtoimprove theability of discriminating

wastewa-tersamplescomparedtoothertraditionalclassiﬁcationmethods,

potentially increasing theability to manage the quality of the

receivingwaterbodies andsupportingthe(near-)zero-emission

approachaswellasnon-potablewaterreuse.

Acknowledgement

AuthorsaregratefultoAndreaScandellaand Prof.Francesco

Avezzùforthesupportreceivedinthephysico-chemicalanalysis

ofwastewatersamples.

AppendixA. Supplementarydata

Supplementarydataassociatedwiththisarticlecanbefound,

intheonlineversion,athttp://dx.doi.org/10.1016/j.ecolind.2015.

(6)

References

Andersen,J.H.,Murray,C.,Kaartokallio,H.,Axe,P.,Molvær,J.,2010.Asimplemethod forconﬁdenceratingofeutrophicationstatusclassiﬁcations.Mar.Pollut.Bull. 60,919–924.

APHA,AmericanPublicHealthAssociation,1998.StandardMethodsforthe Exami-nationofWaterandWastewater,20thed.APHA,Washington,USA.

Bowler,C.,Allen,A.E.,Badger,J.H.,etal.,2008.ThePhaeodactylumgenomereveals theevolutionaryhistoryofdiatomgenomes.Nature456,239–244.

Buesseler,K.O.,Andrews,J.E.,Pike,S.M.,Charette,M.A.,2004.Theeffectsofiron fer-tilizationoncarbonsequestrationintheSouthernOcean.Science304,414–417. Bulich,A.A.,1982.Apracticalandreliablemethodformonitoringthetoxicityof

aquaticsamples.J.ProcessBiochem.17,1–3.

Burkhard,L.P.,Durhan,E.J.,1991.Identiﬁcationofnonpolartoxicantsinefﬂuents usingtoxicitybasedfractionationwithgaschromatography/massspectrometry. Anal.Chem.63,277–283.

BurtonJr.,G.A.,2002.Sedimentqualitycriteriainusearoundtheworld.Limnology 3,65–75.

Calleja,A.,Baldassano,J.M.,Mulet,A.,1986.Toxicityanalysisofleachatesfrom haz-ardouswastesviaMicrotox®_and_Daphnia_magna._Arch._Environ._Contamin.

Toxicol.1,73–83.

Calow,P.,1989.Thechoiceandimplementationofenvironmentalbioassays. Hydro-biologia188–189,61–64.

Claessens,M.,Vanhaecke,L.,Wille,K.,Janssen,C.R.,2013.Emergingcontaminants inBelgianmarinewaters:singletoxicantandmixturerisksofpharmaceuticals. Mar.Pollut.Bull.71,41–50.

Clarkson,N.,Leftley,J.W.,Meldrum,D.T.,Watson,J.W.,1998.Anassessmentof thecage-cultureturbidostatasanalternativealgalbioassay.WaterRes.32, 1162–1168.

Clarkson,N.,Redshaw,C.J.,Leftley,J.W.,Meldrum,D.T.,Watson,J.,1999.Evaluation ofanalgalbioassaycage-cultureturbidostatmethodforthetoxicityassessment ofefﬂuents.Mar.Environ.Res.47,157–173.

Clément,L.,Hurel,C.,Marmier,N.,2013.ToxicityofTiO2nanoparticlesto

clado-cerans,algae,rotifersandplants–effectsofsizeandcrystallinestructure. Chemosphere90,1083–1090.

Costan,G.,Bermingham,N.,Blaise,C.,1993.Potentialecotoxiceffectsprobe(PEEP): anovelindextoassessandcomparethetoxicpotentialofindustrialefﬂuents. Environ.Toxicol.WaterQual.8,115–140.

Deng,G.,Zhang,T.,Yang,L.,Wang,Q.,2013.Studiesofbio-uptakeandtransformation ofmercurybyatypicalunicellulardiatomPhaeodactylumtricornutum.Chin.Sci. Bull.58,256–265.

Dodds,W.K.,2007.Trophicstate,eutrophicationandnutrientcriteriainstreams. TrendsEcol.Evol.22,669–676.

Dodds,W.K.,Jones,J.R.,Welch,E.B.,1998.Suggestedclassiﬁcationofstreamtrophic state:distributionsoftemperatestreamtypesbychlorophyll,totalnitrogen, andphosphorus.WaterRes.32,1455–1462.

EINECS,2013.http://esis.jrc.ec.europa.eu/index.php?pgm=ein

Farré,M.,Barceló,D.,2003.Toxicitytestingofwastewaterandsewagesludgeby biosensors,bioassaysandchemicalanalysis.TrendsAnal.Chem.22,299–310. Fernández,A.,Tejedor,C.,Cabrera,F.,Chordi,A.,1995.Assessmentoftoxicityof

riverwaterandefﬂuentsbythebioluminescenceassayusingPhotobacterium phosphoreum.WaterRes.29,1281–1286.

Keddy,C.J.,Greene,J.C.,Bonnell,M.A.,1995.Reviewofwhole-organismbioassays: soil,freshwatersediment,andfreshwaterassessmentinCanada.Ecotoxicol. Environ.Saf.30,221–251.

Kim,K.Y.,Garbary,D.J.,McLachlan,J.L.,2004.PhytoplanktondynamicsinPomquet Harbour,NovaScotia:alagooninthesouthernGulfofSt.Lawrence.Phycologia 43,311–328.

Knud-Hansen,C.F.,McElwee,K.,Baker,J.,Clair,D.,1998.Pondfertilization:ecological approachandpracticalapplication.PondDynamics/AquacultureCollaborative ResearchSupportProgram.OregonStateUniversity,USA.

Leotsinidis,M.,Sazakli,E.,2008.Evaluatingcontaminationofdredgesanddisposal criteriainGreekcostalareas.Chemosphere72,811–818.

Libralato,G.,Avezzù,F.,Losso,C.,VolpiGhirardini,A.,2009a.Inﬂuenceofstorage methods,refrigerationorfreezing,onthetoxicityofwastewatersamplesto oysterembryos.Environ.Technol.30,535–541.

Libralato,G.,Losso,C.,Avezzù,F.,VolpiGhirardini,A.,2009b.Inﬂuenceofsalinity adjustmentmethods,saltsandbrine,onthetoxicityofwastewatersamplesto musselsembryos.Environ.Technol.30,85–91.

Libralato,G.,VolpiGhirardini,A.,Avezzù,F.,2009c.Performanceassessmentof AS-SBRandUF-MBRforhotelwastewatertreatment.WaterSci.Technol.60, 1701–1709.

Libralato,G., Volpi Ghirardini, A., Avezzù, F., 2010a. Seawater ecotoxicity of monoethanolamine,diethanolamineandtriethanolamine.J.Hazard.Mater.176, 535–539.

Libralato,G.,VolpiGhirardini,A.,Avezzù,F.,2010b.Howtoxicistoxic?Aproposalfor wastewatertoxicityhazardassessment.Ecotoxicol.Environ.Saf.73,1602–1611. Libralato,G.,VolpiGhirardini,A.,Avezzù,F.,2010c.Toxicityremovalefﬁciencyof decentralisedsequencingbatchreactorandultra-ﬁltrationmembrane bioreac-tors.WaterRes.44,4437–4450.

Libralato,G.,Avezzù,F.,VolpiGhirardini,A.,2011.Ligninandtannintoxicityto Phaeodactylumtricornutum(Bohlin).J.Hazard.Mater.194,435–439. Libralato,G.,VolpiGhirardini,A.,Avezzù,F.,2012.Tocentraliseortodecentralise:

anoverviewofthemostrecenttrendsinwastewatertreatmentmanagement. J.Environ.Manage.94,61–68.

Macova,M.,Escher,B.I.,Reungoat,J.,Carswell,S.,Chue,L.K.,Keller,J.,Mueller,J.F., 2010.Monitoringthebiologicalactivityofmicropollutantsduringadvanced wastewatertreatmentwithozonationandactivatedcarbonﬁltration.Water Res.44,477–492.

Moreira-Santos,M.,Guilhermino,L.,Ribeiro,R.,2006a.Immobilizationofthemarine microalgaPhaeodactylumtricornutuminalginateforinsituexperiments:bead stabilityandsuitability.EnzymeMicrob.Technol.38,135–141.

Moreira-Santos,M.,Guilhermino,L.,Ribeiro,R.,2006b.Aninsituassaywiththe microalgaPhaeodactylumtricornutumforsediment-overlyingwatertoxicity evaluationsinestuaries.Environ.Toxicol.Chem.25,2272–2279.

Morelli,E.,Marangi,M.L.,Fantozzi,L.,2009.Aphytochelatin-basedbioassayin marinediatomsusefulfortheassessmentofbioavailabilityofheavymetals releasedbypollutedsediments.Environ.Int.35,532–538.

Morelli,E.,Salvadori,E.,Bizzarri,R.,Cioni,P.,Gabellieri,E.,2013.Interactionof CdSe/ZnSquantumdotswiththemarinediatomPhaeodactylumtricornutumand thegreenalgaDunaliellatertiolecta:abiophysicalapproach.Biophys.Chem.182, 4–10.

Moreno-Garrido,I.,Lubián,L.M.,Blasco,J.,2007a.Sedimenttoxicitytestsinvolving immobilizedmicroalgae(PhaeodactylumtricornutumBohlin).Environ.Int.33, 481–485.

Moreno-Garrido,I.,Lubián,L.M.,Jimenez,B.,Soares,A.M.V.M.,Blasco,J.,2007b. Estuarinesedimenttoxicitytestsondiatoms:sensitivitycomparisonforthree species.Estuar.Coast.ShelfSci.71,278–286.

Moser,G.A.O.,Sigaud-Kutner,T.C.S.,Cattena,C.O.,Gianesella,S.M.F.,Braga,E.S., Schinke,K.P.,Aidar,E.,2004.Algalgrowthpotentialasanindicatorof eutroph-cationdegreeincoastalareasundersewagedisposalinﬂuence.Aquat.Ecosyst. HealthManage.7,115–126.

Mucha,A.P.,Leal,M.F.C.,Bordalo,A.A.,Vasconcelos,M.T.S.D.,2003.Comparisonof theresponseofthreemicroalgaespeciesexposedtoelutriatesofestuarine sed-imentsbasedongrowthandchemicalspeciation.Environ.Toxicol.Chem.22, 576–585.

Muller,R.,Tang,J.,Thier,R.,Mueller,J.,2007.Combiningpassivesamplingand tox-icitytestingforevaluationofmixturesofpolarorganicchemicalsinsewage treatmentplantefﬂuents.J.Environ.Monit.9,104–109.

Munawar,M.,Munawar,I.F.,Mayﬁeld,C.I.,McCarthy,L.H.,1989.Probing ecosys-temhealth:amulti-disciplinaryandmulti-trophicassaystrategy.Hydrobiologia 188–189,93–115.

Newman,M.C.,Unger,M.A.,2003.FundamentalsofEcotoxicology.CRCPressLLC, BocaRaton.

Okay,O.S.,Egesel,L.,Tüfekc¸i,V.,Morkoc¸,E.,Gaines,A.,1998.Investigationofthree wastewatersenteringIzmitBay(Turkey)bymeansofbatchandchemostat culturealgalbioassays.Mar.Environ.Res.46,283–288.

Okay,O.S.,Gibson,M.,Gaines,M.,Davie,A.,2005.Introductiontocompetition betweencontinuousculturesofPhaeodactylumtricornutumandDunaliella ter-tiolecta.J.Environ.Sci.HealthA40,2117–2134.

OSPAR,2000.OSPARBackgroundDocumentconcerningtheElaborationof Pro-grammesandMeasuresrelatingtoWholeEfﬂuentAssessment.

OSPAR,2005.Wholeefﬂuentassessment,Publication#2005/219.

OSPAR, 2007. Practical Guidance Document on Whole Efﬂuent Assessment, 316/2007.

PANPesticide,2013.Kegley,S.E.,Hill,B.R.,OrmeS.,ChoiA.H.,PANPesticideDatabase, PesticideActionNetwork,NorthAmerica(SanFrancisco,CA,2011),http://www. pesticideinfo.org

Persoone,G.,Marsalek,B.,Blinova,I.,Torokne,A.,Zarina,D.,Manusadzianas,L., Nalecz-Jawecki,G.,Tofan,L.,Stepanova,N.,Tothova,L.,Kolar,B.,2003.A practi-calanduser-friendlytoxicityclassiﬁcationsystemwithmicrobiotestsfornatural watersandwastewaters.Environ.Toxicol.18,395–402.

Peters,C.,Becker,S.,Noack,U.,Pﬁtzner,S.,Bülow,W.,Barz,K.,Ahlf,W.,Berghahn, R.,2002.Amarinebioassaytestsettoassessmarinewaterandsedimentquality –itsneed,theapproachandﬁrstresults.Ecotoxicology11,379–383. Phillips, B., Hunt, J.W., Anderson, B.S., Puckett, H.M., Fairey, R., Wilson, C.J.,

Tjeerdema,R.,2001.Statisticalsigniﬁcanceofsedimenttoxicitytestresults: thresholdvaluesderivedbythedetectablesigniﬁcanceapproach.Environ. Tox-icol.Chem.20,371–373.

Puddu,A.,Pettine,M.,LaNoce,T.,Pagnotta,R.,Bacciu,F.,1988.Factorsaffecting thalliumandchromiumtoxicitytomarinealgae.Sci.TotalEnviron.71,572. Rech,M.,Mouget,J.-L.,Morant-Manceau,A.,Rosa,P.,Tremblin,G.,2005.Long-term

acclimationtoUVradiation:effectsongrowth,photosynthesisandcarbonic anhydraseactivityinmarinediatoms.Bot.Mar.48,407–420.

Ross,P.,1993.Theuseofbacterialluminescencesysteminaquatictoxicitytesting. In:Richardson,M.(Ed.),EcotoxicologyMonitoring.,1sted.VCH,Weinheim/New York/Basel/Cambridge/Tokyo,p.384.

Sarakinos,H.C.,Bermingham,N.,White,P.A.,Rasmussen,J.B.,2000.Correspondence betweenwholeefﬂuenttoxicityandthepresenceofprioritysubstancesin com-plexindustrialefﬂuents.Environ.Toxicol.Chem.19,63–71.

Sbrilli,G.,Bimbi,B.,Cioni,F.,Pagliai,L.,Luchi,F.,Lanciotti,E.,2005.Surfaceand groundwaterscharacterizationinTuscany(Italy)byusingalgalbioassayand pesticidedeterminations:comparativeevaluationoftheresultsandhazard assessmentofthepesticidesimpactonprimaryproductivity.Chemosphere58, 571–578.

Shaw,M.,Negri,A.,Fabricius,K.,Mueller,J.F.,2009.Predictingwatertoxicity:pairing passivesamplingwithbioassaysontheGreatBarrierReef.Aquat.Toxicol.95, 108–116.

SwedishEPA,1997.CharacterisationofDischargesfromChemicalIndustry–The StorkProject.SwedishEnvironmentalProtectionAgency,Stockholm,Reportn. 4766.

(7)

Toepel,J.,Langner,U.,Wilhelm,C.,2005.Combinationofflowcytometryand sin-glecellabsorptionspectroscopytostudythephytoplanktonstructureandto calculatethechiaspecificabsorptioncoefficientsatthetaxonlevel.J.Phycol., 1099–1109.

Tolun,L.G.,Okay,O.S.,Gaines,A.F.,Tolay,M.,Kiratli,N.,2001.Thepollutionstatusand thetoxicityofsurfacesedimentsinIzmitBay(MarmaraSea),Turkey.Environ. Int.26,163–168.

Tonkes,M.,deGraaf,P.J.F.,Graansma,J.,1999.Assessmentofcomplexindustrial efﬂuentsintheNetherlandsusingaWholeEfﬂuentToxicity(orWET)approach. WaterSci.Technol.39,55–61.

Tothill,I.E.,Turner,A.P.F.,1996.Developmentsinbioassaymethodsfortoxicity testinginwatertreatment.TrendsAnal.Chem.15,178–188.

USEPA,1992.MethodsforDeterminationofMetalsinEnvironmentalSample.CRC PressInc,BocaRaton,USA.

USEPA,2004.NPDESComplianceInspectionManual,EPA305-X-03-004. Vindimian,E.,Garric,J.,Flammarion,P.,Thybaud,E.,Babut,M.,1999.Anindex

ofefﬂuentaquatictoxicitydesignedbypartialleastsquaresregression,using acuteandchronictestsandexpertjudgements.Environ.Toxicol.Chem.18, 2386–2391.

Whadhia,K.,Thomson,C.K.,2007.Lowcostecotoxicitytestingofenvironmental samplesusingmicrobiotestsforpotentialimplementationoftheWater Frame-workDirective.TrendsAnal.Chem.26,300–307.