Dt

XJa

b

Ac

h

•••

ARRAA

KTASNR

1

btaca

RaS

0h

Journal of Hazardous Materials 260 (2013) 1017– 1022

Contents lists available at SciVerse ScienceDirect

Journal of Hazardous Materials

jou rn al hom epage: www.elsev ier .com/ locate / jhazmat

evelopment of aquatic life criteria for triclosan and comparison ofhe sensitivity between native and non-native species

iao-Nan Wanga,b, Zheng-Tao Liub,∗, Zhen-Guang Yanb,∗∗, Cong Zhangc, Wei-Li Wangb,un-Li Zhoub, Shu-Wei Peib

College of Water Sciences, Beijing Normal University, Beijing 100875, ChinaState Key Laboratory of Environmental Criteria and Risk Assessment, State Environmental Protection Key Laboratory of Ecological Effect and Riskssessment of Chemicals, Chinese Research Academy of Environmental Sciences, Beijing 100012, ChinaSchool of Environment, Beijing Normal University, Beijing 100875, China

i g h l i g h t s

Toxicity tests of triclosan on 9 Chinese native aquatic species were conducted.Aquatic life criteria for triclosan were derived.There is no significant difference between native and non-native species to TCS.

a r t i c l e i n f o

rticle history:eceived 24 April 2013eceived in revised form 26 June 2013ccepted 4 July 2013vailable online 12 July 2013

eyword:riclosan

a b s t r a c t

Triclosan (TCS) is an antimicrobial agent which is used as a broad-spectrum bacteriostatic and found inpersonal care products, and due to this it is widely spread in the aquatic environment. However, thereis no paper dealing with the aquatic life criteria of TCS, mainly result from the shortage of toxicity dataof different taxonomic levels. In the present study, toxicity data were obtained from 9 acute toxicitytests and 3 chronic toxicity tests using 9 Chinese native aquatic species from different taxonomic levels,and the aquatic life criteria was derived using 3 methods. Furthermore, differences of species sensitivitydistributions (SSD) between native and non-native species were compared. Among the tested species,

quatic life criteriapecies sensitivity distributionative speciesisk assessment

demersal fish Misgurnus anguillicaudatus was the most sensitive species, and the fishes were more sen-sitive than the aquatic invertebrates of Annelid and insect, and the insect was the least sensitive species.The comparison showed that there was no significant difference between SSDs constructed from nativeand non-native taxa. Finally, a criterion maximum concentration of 0.009 mg/L and a criterion continuousconcentration of 0.002 mg/L were developed based on different taxa, according to the U.S. EnvironmentalProtection Agency guidelines.

. Introduction

Personal care products (PCPs) reach the environment and cane found in all environmental medias [1,2]. The presence of PCPs inhe environment and their possible effects on nontarget organisms

re of great concern worldwide [3–5]. Among the PCPs, triclosan (5-hloro-2(2,4-dichlorophenoxy) phenol or TCS, CAS#3380-34-5) is

broad spectrum antimicrobial agent used in a variety of personal

∗ Corresponding author at: State Key Laboratory of Environmental Criteria andisk Assessment, State Environmental Protection Key Laboratory of Ecological Effectnd Risk Assessment of Chemicals, Chinese Research Academy of Environmentalciences, Beijing 100012, China. Tel.: +86 10 84915175; fax: +86 10 84915173.∗∗ Corresponding author. Tel.: +86 10 84915175; fax: +86 10 84915173.

E-mail addresses: liuzt@craes.org.cn (Z.-T. Liu), zgyan@craes.org.cn (Z.-G. Yan).

304-3894/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.jhazmat.2013.07.007

© 2013 Elsevier B.V. All rights reserved.

care products including soaps, deodorant, and toothpaste [6,7]. TCShas already been the subject of various scientific fields and has beendetected in surface water worldwide in the recent years [8–18].Toxicity tests showed that TCS is toxic to organisms [19–22], soit may impose a potential risk to the ambient water environment.However, comprehensive ecological risk assessment of TCS is hardto perform due to the absence of water quality criteria (WQC) forTCS.

WQC is an important basis for the development of water qualitystandards (WQSs), risk assessment, prediction and control, as wellas governance body of water pollution [23]. Despite there weresome toxicity data of TCS available on fish and algae, few taxonomic

levels were tested, especially for native species in China. So, nofinal WQC was derived for TCS. In China, systematic water qualitycriteria studies are getting more attention in recent years, and U.S.Environmental Protection Agency (US EPA) guidelines and other

1 ous M

sdo

wfddwMoecp(IlFn

doesota

2

2

t(r

2

S

fosaittg

2

TloArrcwnwta

018 X.-N. Wang et al. / Journal of Hazard

pecies sensitivity distribution (SSD) methods have been used toerive water quality criteria for some toxicants with an emphasisn using Chinese native species [24–26].

In the present study, 9 acute toxicity tests and 3 chronic testsere conducted for 9 Chinese native species, and the WQC of TCS

or freshwater aquatic life was derived using a battery of toxicityata according to the US EPA guidelines. The aquatic species fromifferent taxonomic levels (3 Phyla and 7 Families) were selected,hich included 4 fishes (Pseudorasbora parva, Carassius auratus,isgurnus anguillicaudatus and Tanichthys albonubes (list the sec-

nd class of protection animal of China, the wild species are rare andndangered)), a planktonic crustacean (Daphnia magna), a benthicrustacean (Neocaridina denticulata sinensis), an insect (Chironomuslumosus), an annelid (Limnodrilus hoffmeisteri) and an amphibianRana limnocharis). Moreover, the SSD method of Holland Nationalnstitute for Public Health and the Environment (RIVM) [27] and theog-logistic SSD method were both applied to validate the results.urthermore, the difference of sensitivity between native and non-ative species was compared.

The objectives of this work are (1) a supplement to TCS toxicityatabase, (2) derivation of aquatic life criteria, and (3) comparisonf the sensitivity difference between native and non-native speciesxposed to TCS, and discussion of the possibility of using non-nativepecies toxicity data for site-specific ecological risk assessmentf TCS. This work provides valuable information for environmen-al risk assessment and pollution management imposed by TCS inmbient water environment.

. Materials and methods

.1. Collection of published ecotoxicity data for TCS

The published ecotoxicity data of TCS were collected fromhe ECOTOX database (http://cfpub.epa.gov/ecotox), the CNKIhttp://www.cnki.net) and ELSEVIER (http://www. sciencedi-ect.com).

.2. Test chemicals and organisms

TCS, C12H7Cl3O2, ≥97% purity (HPLC), was purchased fromigma–Aldrich.

According to the US EPA guidelines [28], data on at least eightamilies of aquatic animals drawn from three different phyla, andne aquatic plant are required in the derivation of WQC. In ourtudy, in addition to the published ecotoxicity data, nine residentquatic species in China were chosen for the acute and chronic tox-city tests. Prior to the toxicity tests, all test organisms were allowedo acclimatize to the dilution water during a minimum of 7 days. Alloxicity tests were conducted strictly according to ASTM standarduidelines [29–31].

.3. General test conditions

Dechlorinated tap water was used as dilution water for all tests.he measured quality parameters of dilution water were as fol-ows: pH 8.00 ± 0.20, dissolved oxygen (DO) 8.30 ± 0.30 mg/L, totalrganic carbon 0.02 mg/L, and hardness as CaCO3 190 ± 0.10 mg/L.ll tests were static-renewal whereby test solutions were totallyeplenished at 24 h intervals. The standard conditions were threeeplicate test containers each containing 10 organisms, at variousoncentrations, solvent control (DMSO) and blank control. All testsere undertaken at a light: dark photoperiod of 12:12 h. Test orga-

isms were not fed during the acute test periods. Test chambersere immersed in a water bath adjusted to maintain the water

emperature at 22 ± 2 ◦C, unless otherwise noted. Temperature, DO,nd pH were measured in test chambers daily in the acute toxicity

aterials 260 (2013) 1017– 1022

tests and at least once a week in the chronic toxicity tests. Biologicalobservations were performed at least once daily.

In the acute toxicity test, 48-h-EC50 (effective concentration in50% of the test organisms over 48 h) for D. magna and 96-h-LC50(lethal concentration in 50% of the test organisms over 96 h) forother aquatic animals were used as main endpoints. In the chronictoxicity test, 21-d-EC10 (effective concentration in 10% of the testorganisms over 21 days) for D. magna and 30-d-EC10 (effective con-centration in 10% of the test organisms over 30 days) for fishes wereused. For fry growth, the specific growth rate (SGR) was chosenbecause it is less dependent on the initial size of the fish and thetime between measurements than the other endpoint such as rel-ative growth rate (RGR) [32]. The SGR was calculated as ((ln(finalmass) − ln(initial mass)) × 100)/d of exposure [33].

2.4. Acute toxicity tests

All the organisms tested in this study were obtained from Chao-lai Market, Chaoyang, Beijing. D. magna (<24 h age) were obtainedfrom in-house cultures at our chemical laboratory of ChineseResearch Academy of Environmental Sciences.

2.4.1. P. parva acute toxicity testThe mean wet weight of the fish was 0.20 ± 0.02 g. Fish fry were

exposed to test solutions in 5 L glass aquaria for 96 h. Nominal expo-sure concentrations in the definitive test were 0.000, 0.018, 0.026,0.040, 0.059, 0.089, 0.133, and 0.200 mg/L TCS, respectively.

2.4.2. C. auratus acute toxicity testThe mean wet weight of the fish was 4.00 ± 0.80 g. Fish were

exposed to test solutions in 50 L glass aquaria for 96 h. Nominalexposure concentrations in the definitive test were 0.000, 1.350,1.490, 1.640, 1.800, 1.980, 2.180, 2.400 and 2.640 mg/L TCS, respec-tively.

2.4.3. M. anguillicaudatus acute toxicity testThe mean wet weight of the fish was 0.68 ± 0.05 g. Fish were

exposed to test solutions in 20 L glass aquaria for 96 h. Nominalexposure concentrations in the definitive test were 0.000, 0.020,0.030, 0.044, 0.067, 0.100, 0.150 and 0.225 mg/L TCS, respectively.

2.4.4. T. albonubes acute toxicity testThe mean wet weight of the fish was 0.02 ± 0.01 g. Fish fry were

exposed to test solutions in 2 L beakers for 96 h. Nominal exposureconcentrations in the definitive test were 0.000, 0.749, 0.787, 0.826,0.867, 0.910, 0.956, and 1.004 mg/L TCS, respectively.

2.4.5. D. magna acute toxicity testThe species were exposed to 100 mL test solution in 150 mL

beakers for 48 h. Nominal concentrations in the definitive test were0.000, 0.232, 0.278, 0.333, 0.400, 0.480, 0.576 and 0.691 mg/L TCS,respectively.

2.4.6. N. denticulata sinensis acute toxicity testThe mean wet weight of the crustacean was 0.02 ± 0.01 g.

Shrimp were exposed to test solutions in 2 L beakers for 96 h. Nomi-nal exposure concentrations in the definitive test were 0.000, 0.591,0.650, 0.715, 0.787, 0.865, 0.952, and 1.047 mg/L TCS, respectively.

2.4.7. C. plumosus acute toxicity testThe mean wet weight of the organism was 0.03 ± 0.01 g. The

species were exposed to 25 mL test solutions in glass culture dishfor 96 h. Nominal exposure concentrations in the definitive testwere 0.000, 2.066, 2.273, 2.500, 2.750, 3.025, 3.328, 3.660, 4.026and 4.429 mg/L TCS, respectively.

X.-N. Wang et al. / Journal of Hazardous Materials 260 (2013) 1017– 1022 1019

Table 1Acute toxicity of TCS to nine resident aquatic organisms.

Species Exposure time (h) Functions R2 p LC50/EC50 (mg/L)

P. parva 96 y = 1.0416x + 3.0744 0.9417 <0.01 0.071 (0.029–0.169)C. auratus 96 y = 9.9154x + 2.3766 0.9880 <0.01 1.839 (1.649–2.050)M. anguillicaudatus 96 y = 2.6102x + 0.6783 0.9515 <0.01 0.045 (0.027–0.077)T. albonubes 96 y = 20.128x − 54.352 0.9871 <0.01 0.889 (0.845–0.934)D. magna 48 y = 6.9004x − 12.449 0.9833 <0.01 0.338 (0.278–0.410)N. denticulata sinensis 96 y = 10.952x − 26.625 0.9832 <0.01 0.772 (0.700–0.852)

30.5 3.77

8.03

2

sfw6

2

pe0t

2

2

tw0via

2

Ttwwr

2

m1gortnw0t

2

ttad

C. plumosus 96 y = 10.276x −L. hoffmeisteri 96 y = 2.6515x −R. limnocharis 96 y = 4.8031x −

.4.8. L. hoffmeisteri acute toxicity testThe mean body length of the tubifex was 3.00 ± 0.50 cm. The

pecies were exposed to 25 mL test solutions in glass culture dishor 96 h. Nominal exposure concentrations in the definitive testere 0.000, 0.364, 0.545, 0.818, 1.227, 1.841, 2.761, 4.142 and

.212 mg/L TCS, respectively.

.4.9. R. limnocharis acute toxicity testThe mean body length of the tadpoles was 1.60 ± 0.20 cm. Tad-

oles were exposed to test solutions in 2 L beakers for 96 h. Nominalxposure concentrations in the definitive test were 0.000, 0.265,.318, 0.382, 0.458, 0.550, 0.660, 0.792 and 0.950 mg/L TCS, respec-ively.

.5. Chronic toxicity tests

.5.1. M. anguillicaudatus chronic toxicity testThe fish used in the chronic toxicity test were the same as in

he acute toxicity test. Thirty-day short-term chronic toxicity testsere conducted. The fish were fed with brine shrimp at a rate of

.1% body weight twice daily. Endpoints observed included sur-ival, growth and body weight. Nominal exposure concentrationsn the definitive test were 0.000, 0.003, 0.005, 0.007, 0.010, 0.015nd 0.023 mg/L TCS, respectively.

.5.2. T. albonubes chronic toxicity testThirty-day short-term chronic toxicity tests were conducted.

he fish were fed with brine shrimp at a rate of 0.1% body weightwice daily. Endpoints observed included survival, growth and bodyeight. Nominal exposure concentrations in the definitive testere 0.000, 0.080, 0.096, 0.116, 0.139, 0.167 and 0.200 mg/L TCS,

espectively.

.5.3. D. magna chronic toxicity testThree weeks survival-reproduction tests using neonates of D.

agna (<24 h age) were conducted in 150 mL beakers filled with00 mL test solutions. The daphnia were fed once a day withreen algae (Scenedesmus obliquus) that had a cell concentrationf 1.0 × 105 cell/mL in the test solution. The survival, growth andeproduction were recorded daily. The endpoints included the timeo first brood, number to first brood, number of broods and totalumber of spawning. Nominal concentrations in the definitive testere 0.000, 0.018, 0.024, 0.031, 0.040, 0.053, 0.068, 0.089, 0.115,

.150 and 0.195 mg/L TCS, respectively, with ten replicate test con-ainers each containing one organism.

.6. Chemical analysis

The concentration of TCS in the solutions of each group at

he beginning and end of the experiments was detected. Briefly,he solutions were filtered through a 0.45-mm filter membranend followed by high performance liquid chromatography (HPLC)etection, using an LC-10AD system (SHIMADZU, Tokyo, Japan),

64 0.9572 <0.01 2.890 (2.667–3.131)9 0.9765 <0.01 2.046 (1.399–2.994)58 0.9795 <0.01 0.518 (0.436–0.615)

a ODS column (150 mm × 4.6 mm, particle size 5 �m; Shim-packCLC-ODS, SHIMADZU, Japan), mobile phase of methanol and water(90:10, v/v) with a flow rate of 1.0 mL/min. The column temper-ature was 40◦C, a loading volume of 10 �L and ultra violet (UV)detection was at 230 nm. TCS had a retention time of 5 min. Fol-lowing the establishment of a response to a known concentration,the result of measured/nominal concentration was 93.14–102.58%.The variability of TCS concentration was less than 20%, in compli-ance with the requirement of the toxicity text guidelines. Therefore,all subsequent toxicity results were expressed on the nominal con-centrations of TCS.

2.7. Statistical analysis and SSD generation

Probit methodology was employed to calculate the 48-h-EC50,96-h-LC50 values and corresponding 95% confidence intervalsdepending on the raw data distribution. Data of the chronic toxicitytests were analyzed, and the EC10 for the most sensitive biologicalendpoint in each test was estimated.

Three procedures, the US EPA guidelines for aquatic life crite-ria [28], the software ETX2.0 [19] exploited by the RIVM andthe log-logistic SSD method [34–36] were used to derive theWQC for TCS. The data analysis softwares are SPSS 20.0 andOriginLab 8.0.

3. Results and discussion

3.1. Results of toxicity tests of nine native aquatic organisms andpublished toxicity data

Toxicity values of TCS to 8 aquatic species and an endangeredspecies are shown in Tables 1 and 2. No mortality was observed inthe control groups and the solvent control groups. Results of acutetoxicity tests showed that M. anguillicaudatus with a 96-h-LC50 of0.045 mg/L was the most sensitive species to TCS followed by P.parva, D. magna, R. limnocharis, N. denticulata sinensis, T. albonubes,C. auratus, L. hoffmeisteri, while the least sensitive species was C.plumosus with a 96-h-LC50 of 2.890 mg/L. Results of the presentstudy indicate that TCS is highly toxic to native freshwater aquaticorganisms. Liang et al. [37] reported that the 96-h-LC50 of TCSon fish X. helleri was 1.47 mg/L. This is consistent with our study(0.889–1.839 mg/L for fishes except 0.045 mg/L for M. anguillicau-datus and 0.071 mg/L for P. parva). Previous study also reportedthat P. parva is sensitive to brominated flame retardant TBBPA(Tetrabromobisphenol A) [25], and Wang et al. [38] reported thatP. parva is sensitive to organic pollutants, especially to pesticides.So it might be a sensitive species to antimicrobial agent, too. Inaddition, our previous study found that demersal fish M. anguil-licaudatus is sensitive to some organochlorine pesticide. Brausch

and Rand also indicated that TCS could affect benthic animals dueto its transfer characteristics [39]. Therefore, TCS might have specialtoxic modes of action on M. anguillicaudatus and P. parva. Amongdifferent species, the fishes were more sensitive than the aquatic

1020 X.-N. Wang et al. / Journal of Hazardous Materials 260 (2013) 1017– 1022

Table 2Chronic toxicity of TCS to three resident aquatic organisms.

Species Endpoints Functions R2 p EC10 (mg/L)

M. anguillicaudatus Growth y = 1.5822x + 2.221 0.9394 <0.01 0.009T. albonubes Growth y = 5.2638x − 6.4808 0.9839 <0.01 0.087D. magna Time to first brood (days) y = 1.5045x − 2.3882 0.9730 <0.01 0.045

y = 4.6941x − 3.9019 0.9271 <0.01 0.042y = 1.6896x + 1.2411 0.9571 <0.01 0.029y = 2.5486x − 1.1721 0.9086 <0.01 0.083

isct

s9O[s0iw04wsgwiaof2obapTpeLSat

swTtc

TA

Table 4Ranked GMAVs with SACRs according to USEPA guidelines.

Rank Species SMAVs(mg/L)

GMAVs(mg/L)

SACRs Reference

1 M. anguillicaudatus 0.045 0.045 5.00 In this study2 P. parva 0.071 0.071 – In this study3 D. magna 0.338 0.338 11.66 In this study4 R. limnocharis 0.518 0.518 – In this study5 N. denticulata sinensis 0.772 0.772 – In this study6 T. albonubes 0.889 0.889 10.22 In this study7 X. helleri 1.470 1.470 – [37]8 C. auratus 1.839 1.839 – In this study9 L. hoffmeisteri 2.046 2.046 – In this study

native species, only SSD constructed from acute toxicity data ofnative (Table 4) and non-native (Table 3) species were compared.

Based on the comparison in this study (Fig. 1), the SSD curveof native species below 0.15 (HC15, hazardous concentration for

Number to first brood (n)

Total number of spawning (n)

Number of broods (n)

nvertebrates of Annelid and insect, and the insect was the least sen-itive species. The chronic data and the other acute toxicity valueannot be compared with previous studies due to lack of residentoxicity data in China.

Results of our study and previous studies using non-nativepecies (Table 3) were compared. Previous studies reported that6-h-LC50 for fish Pimephales promelas, Lepomis macrochirus andryzias latipes were 0.260, 0.370 and 0.602 mg/L, respectively

19,20], and this was in accordance with our study. In the presenttudy we found that 96-h-LC50 for amphibian R. limnocharis was.518 mg/L. It is in agreement with previous studies that tox-

city values for amphibian Xenopus laevis, Acris blanchardii, Bufooodhousii and Rana sphenocephala were 0.259, 0.367, 0.152 and.562 mg/L, respectively [21,22]. Orvos et al. [19] reported that the8-h-EC50 of TCS on planktonic crustacean D. Magna and C. dubiaere 0.390 and 0.168 mg/L, and this was in accordance with our

tudy that the 48-h-EC50 of TCS on D. magna was 0.338 mg/L. Ineneral the sensitivities of the native species tested in this studyere similar to those reported in previous studies. The acute tox-

city of the shrimp N. denticulata sinensis, the insect C. plumosusnd the annelid L. hoffmeisteri cannot be compared with previ-us studies due to lack of toxicity data for resident and non-nativereshwater species. For chronic test, Orvos et al. [19] reported that1 days NOEC (no observed effect concentration) for the survivalf D. Magna was 0.2 mg/L, whereas the 21-d-EC10 for total num-er of spawning was 0.029 mg/L and no mortality was observed inll concentration groups in this study, which indicated that end-oint of total number of spawning is more sensitive than survival.he 30-d-EC10 for growth of fish T. albonubes was 0.087 mg/L, andrevious studies reported that the 21 days LOEC (lowest observedffect concentration) for growth of fish O. latipes [40] and 96 daysOEC for survival of O. mykiss [19] were 0.2 and 0.071 mg/L TCS.imilarly, as mentioned in the above paragraph, demersal fish M.nguillicaudatus is also the most sensitive species in the chronicests in this study.

Previous studies reported that the 96-h-EC50s for algae Chlorellapp., Selenastrum capricornutum, S. subspicatus and N. pelliculosa

ere 0.065, 0.092, 0.001 and 0.019 mg/L, respectively [19,42,43].

herefore, compared with other taxonomic levels the algal washe most sensitive taxa and the algal growth was affected at con-entrations less than 1 �g/L. High TCS sensitivity in algae is likely

able 3cute toxicity data of TCS to non-native aquatic organisms.

Rank Species EC50/LC50 (mg/L) Time (days) Reference

1 B. woodhousii 0.152 4 [21]2 C. dubia 0.168 2 [19]3 X. laevis 0.259 4 [22]4 P. promelas 0.260 4 [19]5 O. mykiss 0.288 4 [41]6 A. blanchardii 0.367 4 [21]7 L. macrochirus 0.370 4 [19]8 T. platyurus 0.470 1 [40]9 R. sphenocephala 0.562 4 [21]10 O. latipes 0.602 4 [20]Algae S. subspicatus 0.001 4 [19]Algae N. pelliculosa 0.019 4 [19]

10 C. plumosus 2.890 2.890 – In this studyAlgae Chlorella spp 0.065 [43]Algae S. capricornutum 0.092 [42]

due to TCS antibacterial characteristics, by uncoupling of oxidativephosphorylation [44], membrane destabilization [45], or disrup-tion of lipid synthesis through the FabI (fatty acid synthesis) andFASII (enoyl acyl carrier protein reductase) pathways [46] whichare similar between algae and bacteria [47].

3.2. Comparison of SSD between native and non-native taxa

Davies et al. [48] reported that using toxicity data of speciesfrom one geographical region to assess the hazard posed to speciesin a different region has been questioned. Moreover, differences inthe sensitivity of cold-water, temperate, and tropical fish specieshave been reported [49]. In the present study, the difference of SSDbased on native and non-native species was compared. Since therehas been little information on the toxicity of TCS to native and non-

Fig. 1. Species sensitivity distribution of native, non-native and total species toxicitydata for TCS.

ous M

tltwHfmntp

bs(CnnSwh[nJdennpath

3

to(TR

m(wGc(r(0(cbC

ttmwc

etr([

X.-N. Wang et al. / Journal of Hazard

he 15% of species) was shifted to the left of non-native species. Aow HC5 (hazardous concentration for the 5% of species) indicatedhat Chinese freshwater organisms with acute exposures to TCSere more sensitive than the non-native species at the tail of SSD.owever, the HC50 (hazardous concentration for the 50% of species)

or native taxa was higher than that for non-native taxa. The reasonight be the lack of toxicity data for Annelid and insect in non-

ative taxa, while the two taxa were demonstrated to be toleranceo TCS in our study (2.046 mg/L for L. hoffmeisteri, 2.890 mg/L for C.lumosus).

Although there were differences for HC5 and HC50etween native and non-native taxa, the comparison of SSDhowed that there was no statistically significant differenceKolmogorov–Smirnov test: ks = 1.342, n1 = 10, n2 = 10, p = 0.06).onsidering the lack of toxicity data for Annelid and insect inon-native taxa, the difference was also not significant (ks = 1.054,1 = 8, n2 = 10, p = 0.22) when removing the two taxa toxicity values.imilarly, sensitivities among North American and European taxaith different geographic distributions to a series of pollutantsave been shown to be no statistically significant difference49,50]. Moreover, the difference between SSDs of Australian andon-Australian organisms to endosulfan was not significant [51].

in et al. [26] also reported that there was no statistically significantifference between SSDs of native and non-native species whenxposed to 2,4-dichlorophenol. In addition, in our study, there waso statistically significant difference between the sensitivity ofative and total species (Fig. 1) as well (ks = 0.78, n1 = 10, n2 = 20,

= 0.59). Previous study found that natural history, habitat typend geographical distribution of the species used to constructhe SSD did not have a significant influence on the assessment ofazard [51,52], and this was in accordance with our study.

.3. Aquatic life criteria derivation

The US EPA guidelines of aquatic life criteria prescribed that onlyhe toxicity data of resident species could be used in the calculationf criteria [28]. In the present study, using 10 species toxicity valuesTable 4), 3 methods were used to derive the aquatic life WQC ofCS including US EPA guidelines, software ETX 2.0 (exploited byIVM) and log-logistic (SSD) method.

Using the methods described in US EPA guidelines, the speciesean acute values (SMAVs) and the genus mean acute values

GMAVs) were calculated. The species acute chronic ratios (SACRs)ere calculated as a ratio of acute and chronic values. RankedMAVs with SACRs are listed in Table 4. A criterion maximum con-entration (CMC) of 0.009 mg/L TCS was obtained by dividing FAVfinal acute value, 0.017 mg/L TCS) by two. The final acute-chronicatio (FACR) was calculated as geometric means of the three SACRs10.22, 11.66 and 5.00) to be 8.41. The FCV (final chronic value,.002 mg/L) was obtained by dividing the FAV by the FACR. FPVfinal plant value) was the toxicity data of Chlorella spp. and S. capri-ornutum, calculated to be 0.065 and 0.092 mg/L. The lower valueetween FCV and FPV was selected as the Criterion Continuousoncentration (CCC) 0.002 mg/L.

In addition, the values of HC5 derived based on the ETX 2.0 andhe constructed log-logistic SSD were 0.054 and 0.060 mg/L, respec-ively. Therefore, the CMCs of TCS developed with the two SSD

ethods were 0.027 and 0.030 mg/L when the factor is 2, whichere in the same order of magnitude with the CMC (0.009 mg/L)

alculated according to the US EPA guidelines.Due to the wide use of the TCS, the levels of TCS in aquatic

nvironments ranged from ng/L to �g/L [8,16,53], therefore its risk

o the aquatic organisms cannot be ignored. The potential envi-onmental risk of TCS could be assessed by using risk quotientRQ) according to the European technical guidance document (TGD)54]. In China, the highest concentration for TCS reported so far in

[

aterials 260 (2013) 1017– 1022 1021

surface water was 478 ng/L in Shijing River [9], and it is lower thanthe CCC of 0.002 mg/L in this study. The CCC in this study mightprovide useful information in site-specific ecological risk assess-ment because our study and previous studies both indicated thatthere was no significant difference between the sensitivity of nativeand non-native species. So it is possible to use non-native speciestoxicity data for site-specific ecological risk assessment of TCS. Tak-ing into account that the higher concentrations detected so far insurface waters for TCS was 5160 ng/L in India [8,53] and 2300 ng/Lin U.S. streams [53], and the CCC was 0.002 mg/L, the RQ valueswere calculated to be 2.58 and 1.15 which indicated that the TCSmight pose risk to the aquatic species in these places.

4. Conclusions

This study is a contribution in the assessment of the effect ofTCS in the aquatic environment. Toxicity values of 9 acute and 3chronic tests for 9 native species from 3 Phyla and 7 Families wereobtained in this study, in which demersal fish M. anguillicaudatuswas the most sensitive species, and the fishes were more sensitivethan the aquatic invertebrates of Annelid and insect, and the insectwas the least sensitive species. Comparing the species sensitivitydistributions, there was no significant difference between nativespecies and non-native species. It indicates that toxicity data fromdifferent geographic region can be used in site-specific ecologicalrisk assessment of TCS. Furthermore, the CMC derived using US EPAguidelines is similar with the results of RIVM ETX2.0 and the log-logistic SSD method, the CMC and CCC of aquatic life criteria for TCSare 0.009 and 0.002 mg/L, respectively.

Acknowledgments

This work was financially supported by the National Scienceand Technology Project of Water Pollution Control and Abatementof China (Grant no. 2012ZX07501-003-06), the Revolution StartupSpecial Project of Chinese Research Academy of EnvironmentalSciences (Grant nos. 2011GQ-02, 2012YSKY18) and the Programof Environmental Protection Commonweal Research (Grant no.2011467054).

References

[1] M.J. Amorim, E. Oliveira, A.M. Soares, J.J. Scott-Fordsmand, Predicted No EffectConcentration (PNEC) for triclosan to terrestrial species (invertebrates andplants), Environ. Int. 36 (2010) 338–343.

[2] T.A. Ternes, A. Joss, H. Siegrist, Peer reviewed: scrutinizing pharmaceuticalsand personal care products in wastewater treatment, Environ. Sci. Technol. 38(2004) 392–399.

[3] C.G. Daughton, T.A. Ternes, Pharmaceuticals and personal care products in theenvironment: agents of subtle change? Environ. Health Perspect. 107 (1999)907.

[4] N. Veldhoen, R.C. Skirrow, H. Osachoff, H. Wigmore, D.J. Clapson, M.P. Gunder-son, G. Van Aggelen, C.C. Helbing, The bactericidal agent triclosan modulatesthyroid hormone-associated gene expression and disrupts postembryonic anu-ran development, Aquat. Toxicol. 80 (2006) 217–227.

[5] K. Kümmerer, Pharmaceuticals in the environment: sources, fate, effects andrisks, Springer Verlag, Berlin, Germany, 2004.

[6] O.R. Price, R.J. Williams, R. van Egmond, M.J. Wilkinson, M.J. Whelan, Predictingaccurate and ecologically relevant regional scale concentrations of triclosan inrivers for use in higher-tier aquatic risk assessments, Environ. Int. 36 (2010)521–526.

[7] D.C. McAvoy, B. Schatowitz, M. Jacob, A. Hauk, W.S. Eckhoff, Measurement oftriclosan in wastewater treatment systems, Environ. Toxicol. Chem. 21 (2009)1323–1329.

[8] B.R. Ramaswamy, G. Shanmugam, G. Velu, B. Rengarajan, D. Larsson, GC–MSanalysis and ecotoxicological risk assessment of triclosan, carbamazepine andparabens in Indian rivers, J. Hazard. Mater. 186 (2011) 1586–1593.

[9] J. Zhao, G. Ying, Y. Liu, F. Chen, J. Yang, L. Wang, Occurrence and risks of triclosanand triclocarban in the Pearl River system, South China: from source to thereceiving environment, J. Hazard. Mater. 179 (2010) 215–222.

10] J.A. Dougherty, P.W. Swarzenski, R.S. Dinicola, M. Reinhard, Occurrence ofherbicides and pharmaceutical and personal care products in surface water and

1 ous M

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

022 X.-N. Wang et al. / Journal of Hazard

groundwater around Liberty Bay, Puget Sound, Washington, J. Environ. Qual.39 (2010) 1173–1180.

11] M.J. Benotti, R.A. Trenholm, B.J. Vanderford, J.C. Holady, B.D. Stanford, S.A.Snyder, Pharmaceuticals and endocrine disrupting compounds in US drinkingwater, Environ. Sci. Technol. 43 (2008) 597–603.

12] B. Kasprzyk-Hordern, R.M. Dinsdale, A.J. Guwy, The occurrence of pharmaceut-icals, personal care products, endocrine disruptors and illicit drugs in surfacewater in South Wales, UK, Water Res. 42 (2008) 3498–3518.

13] J.W. Kim, H.S. Jang, J.G. Kim, H. Ishibashi, M. Hirano, K. Nasu, N. Ichikawa, Y.Takao, R. Shinohara, K. Arizono, Occurrence of pharmaceutical and personalcare products (PPCPs) in surface water from Mankyung River, South Korea, J.Health Sci. 55 (2009) 249–258.

14] J. Zhao, G. Ying, L. Wang, J. Yang, X. Yang, L. Yang, X. Li, Determination of phenolicendocrine disrupting chemicals and acidic pharmaceuticals in surface waterof the Pearl Rivers in South China by gas chromatography–negative chemicalionization-mass spectrometry, Sci. Total Environ. 407 (2009) 962–974.

15] L. Wang, G. Ying, J. Zhao, S. Liu, B. Yang, L. Zhou, R. Tao, H. Su, Assessingestrogenic activity in surface water and sediment of the Liao River system innortheast China using combined chemical and biological tools, Environ. Pollut.159 (2011) 148–156.

16] C. Ciniglia, C. Cascone, R.L. Giudice, G. Pinto, A. Pollio, Application of methodsfor assessing the geno-and cytotoxicity of Triclosan to C. ehrenbergii, J. Hazard.Mater. 122 (2005) 227–232.

17] A.K. Venkatesan, B.F. Pycke, L.B. Barber, K.E. Lee, R.U. Halden, Occurrence oftriclosan, triclocarban, and its lesser chlorinated congeners in Minnesota fresh-water sediments collected near wastewater treatment plants, J. Hazard. Mater.229–230 (2012) 29–35.

18] A. Dayan, Risk assessment of triclosan [Irgasan®] in human breast milk, FoodChem. Toxicol. 45 (2007) 125–129.

19] D.R. Orvos, D.J. Versteeg, J. Inauen, M. Capdevielle, A. Rothenstein, V. Cun-ningham, Aquatic toxicity of triclosan, Environ. Toxicol. Chem. 21 (2009)1338–1349.

20] H. Ishibashi, N. Matsumura, M. Hirano, M. Matsuoka, H. Shiratsuchi, Y. Ishibashi,Y. Takao, K. Arizono, Effects of triclosan on the early life stages and reproductionof medaka Oryzias latipes and induction of hepatic vitellogenin, Aquat. Toxicol.67 (2004) 167–179.

21] N.M. Palenske, G.C. Nallani, E.M. Dzialowski, Physiological effects and biocon-centration of triclosan on amphibian larvae, Comp. Biochem. Physiol. C 152(2010) 232–240.

22] L. Yang, G. Ying, H. Su, J.L. Stauber, M.S. Adams, M.T. Binet, Growth-inhibitingeffects of 12 antibacterial agents and their mixtures on the freshwatermicroalga Pseudokirchneriella subcapitata, Environ. Toxicol. Chem. 27 (2009)1201–1208.

23] US EPA, Developing Water Quality Criteria for Suspended and Bedded Sed-iments (SABS). http://water.epa.gov/scitech/swguidance/standards/criteria/aqlife/sediment/upload/2004 08 17 criteria sediment sab-discussion-paper.pdf (assessed on August 2003).

24] Z. Yan, Z. Zhang, H. Wang, F. Liang, J. Li, H. Liu, C. Sun, L. Liang, Z. Liu, Devel-opment of aquatic life criteria for nitrobenzene in China, Environ. Pollut. 162(2012) 86–90.

25] S. Yang, Z. Yan, F. Xu, S. Wang, F. Wu, Development of freshwater aquatic lifecriteria for Tetrabromobisphenol A in China, Environ. Pollut. 169 (2012) 59–63.

26] X. Jin, J. Zha, Y. Xu, Z. Wang, S.S. Kumaran, Derivation of aquatic predicted no-effect concentration (PNEC) for 2,4-dichlorophenol: comparing native speciesdata with non-native species data, Chemosphere 84 (2011) 1506–1511.

27] P.L.A. Van Vlaardingen, T.P. Traas, A.M. Wintersen, ETX2.0-A Program to Cal-culate Hazardous Concentration and Fraction Affected. Based on NormallyDistributed Toxicity Data. Report 601501028. National Institute for PublicHealth and the Environment (RIVM), Bilthoven, the Netherlands, 2004.

28] US EPA, Guidelines for Deriving Numerical National Water Quality Criteria forthe Protection of Aquatic Organisms and Their Uses (PB 85-227049), Office ofResearch and Development, Environmental Research Laboratories, U.S. Envi-ronmental Protection Agency, Duluth, Minnesota, Narragansett, Rhode Island;Corvallis, Oregon, USA, 1985.

29] ASTM, Conducting static acute toxicity test on wastewaters with Daphniamagna, in: Annual Book of ASTM Standards, American Society of Testing andMaterials, Philadelphia, PA, USA, 1993.

30] ASTM, Standard guide for conducting acute toxicity tests with fishes, macroin-

vertebrates and amphibians, in: Annual Book of ASTM Standards, AmericanSociety of Testing and Materials, Philadelphia, PA, USA, 1993.

31] ASTM, Standard guide for conducting renewal life-cycle toxicity tests withDaphnia magna, in: Annual Book of ASTM Standards, American Society of Test-ing and Materials, Philadelphia, PA, USA, 1993.

aterials 260 (2013) 1017– 1022

32] M. Mallett, N. Grandy, R. Lacey, Interlaboratory comparison of a method toevaluate the effects of chemicals on fish growth, Environ. Toxicol. Chem. 16(1997) 528–533.

33] N. Crossland, A method to evaluate effects of toxic chemicals on fish growth,Chemosphere 14 (1985) 1855–1870.

34] C. Feng, F. Wu, S. Dyer, H. Chang, X. Zhao, Derivation of freshwater qualitycriteria for zinc using interspecies correlation estimation models to protectaquatic life in China, Chemosphere 90 (2013) 1177–1183.

35] J. Wheeler, E. Grist, K. Leung, D. Morritt, M. Crane, Species sensitivitydistributions: data and model choice, Mar. Pollut. Bull. 45 (2002) 192–202.

36] M.C. Newman, D.R. Ownby, L.C. Mezin, D.C. Powell, T.R. Christensen, S.B. Ler-berg, B.A. Anderson, Applying species sensitivity distributions in ecologicalrisk assessment: assumptions of distribution type and sufficient numbers ofspecies, Environ. Toxicol. Chem. 19 (2000) 508–515.

37] X. Liang, X. Nie, G. Ying, T. An, K. Li, Assessment of toxic effects of triclosan onthe swordtail fish (Xiphophorus helleri) by a multi-biomarker approach, Chemo-sphere 90 (2013) 1281–1288.

38] X. Wang, Z. Liu, Z. Yan, C. Zhang, L. He, S. Meng, Species sensitivity evaluationof Pseudorasbora parva, Environ. Sci. 34 (2013) 265–270 (in Chinese).

39] J.M. Brausch, G.M. Rand, A review of personal care products in the aquatic envi-ronment: environmental concentrations and toxicity, Chemosphere 82 (2011)1518–1532.

40] J.W. Kim, H. Ishibashi, R. Yamauchi, N. Ichikawa, Y. Takao, M. Hirano, M. Koga,K. Arizono, Acute toxicity of pharmaceutical and personal care products onfreshwater crustacean (Thamnocephalus platyurus) and fish (Oryzias latipes), J.Toxicol. Sci. 34 (2009) 227–232.

41] Office of pesticide programs, Pesticide ecotoxicity database (formerly: envi-ronmental effects database (EEDB)), Environmental Fate and Effects Division,U.S. Environmental Protection Agency, Washington, DC, USA, 2000.

42] W. Liao, Photolysis of Lamivudine and Triclosan in Aqueous Solution and ToxicAssessment of Their Photolytical Products to Hydrobiose, Ji Nan University,2012 (in Chinese).

43] X. Wu, R. Liu, H. Li, G. Na, Z. Yao, D. Guan, Effects of Triclosan on the growth ofChlorella spp., Mar. Sci. Bull. 28 (2009) 117–120 (in Chinese).

44] A.P.N. Newton, S.M.S. Cadena, M.E.M. Rocha, E.G.S. Carnieri, M.B. Martinellide Oliveira, Effect of triclosan (TRN) on energy-linked functions of rat livermitochondria, Toxicol. Lett. 160 (2005) 49–59.

45] S. Franz, R. Altenburger, H. Heilmeier, M. Schmitt-Jansen, What contributes tothe sensitivity of microalgae to triclosan? Aquat. Toxicol. 90 (2008) 102–108.

46] S. Lu, M.C. Archer, Fatty acid synthase is a potential molecular target for thechemoprevention of breast cancer, Carcinogenesis 26 (2005) 153–157.

47] M.A. Coogan, R.E. Edziyie, T.W. La Point, B.J. Venables, Algal bioaccumu-lation of triclocarban, triclosan, and methyl-triclosan in a North Texaswastewater treatment plant receiving stream, Chemosphere 67 (2007) 1911–1918.

48] P. Davies, L. Cook, D. Goenarso, Sublethal responses to pesticides of severalspecies of Australian freshwater fish and crustaceans and rainbow trout, Envi-ron. Toxicol. Chem. 13 (1994) 1341–1354.

49] S. Dyer, S. Belanger, G. Carr, An initial evaluation of the use of Euro/NorthAmerican fish species for tropical effects assessments, Chemosphere 35 (1997)2767–2781.

50] L. Maltby, N. Blake, T. Brock, P. Van den Brink, Addressing Interspecific Vari-ation in Sensitivity and the Potential to Reduce this Source of Uncertainty inEcotoxicological Assessments, DEFRA Project Code PN0932, Department forEnvironment, Food and Rural Affairs, London, UK, 2002.

51] G. Hose, P. Van den Brink, Confirming the species-sensitivity distribution con-cept for endosulfan using laboratory, mesocosm, and field data, Arch. Environ.Contam. Toxicol. 47 (2004) 511–520.

52] L. Maltby, N. Blake, T. Brock, P. Van den Brink, Insecticide species sensitivitydistributions: importance of test species selection and relevance to aquaticecosystems, Environ. Toxicol. Chem. 24 (2005) 379–388.

53] D. Kolpin, E. Furlong, M. Meyer, E.M. Thurman, S. Zaugg, L. Barber, H. Buxton,Pharmaceuticals, hormones, and other organic wastewater contaminants inUS streams, 1999-2000: a national reconnaissance, Environ. Sci. Technol. 36(2002) 1202–1211.

54] EC. (European Commission), Technical Guidance Document (TGD) on riskassessment in support of Commission Directive 93/67/EEC on risk assessment

for new notified substances and Commission Regulation (EC) No 1488/94 onrisk assessment for existing substances and Directive 98/8/EC of the Europeanparliament and of the council concerning the placing of biocidal products onthe market, Part II, Technical Report, European Commission, Brussels, Belgium,2003.