37

Chronic Effects of Waterborne PFOS Exposure on Growth,Development, Reproduction and Hepatotoxicity in Zebrafish

Yongbing DU, Xiongjie SHI, Ke YU, Chunsheng LIU and Bingsheng ZHOU

State Key Laboratory of Freshwater Ecology and Biotechnology, Institute of Hydrobiology,Chinese Academy of Sciences, Wuhan 430072, China

(Received 11 May 2008; accepted 29 July 2008)

Abstract—To evaluate the potential effects on growth and reproduction ofperfluorinated compounds (PFCs), the zebrafish fry (F0, 14 d post-fertilization,dpf) were chronically exposed to various concentrations of PFOS (0, 10, 50 and250 µg l–1) until sex mature (70 d) and effects on reproductive capacity andendocrine disrupted potential were assessed. No significant adverse effects onsurvival were observed in developing zebrafish exposed to PFOS for 70 d underall tested concentrations. Reduction of growth (weight and length) andhistological alternations most prominently with lipid droplets accumulation inthe liver were only observed in the male fish. Whole body triiodothyronine (T3)levels were not significantly changed in both male and female. Gonadalsomatic index (GSI) of the females was significantly reduced in the exposuregroups. Despite hepatic vitellogenin (VTG) gene expression was significantlyup-regulated, no intersex was found in testes as well as the sex ratio was notchanged. After 70 d exposure, remaining fish were placed to clean water for 30d to assess recovery of hepatotoxicity, but histological alternations in the malefish liver were not reversible. To determine possible developmental effects ofPFOS exposure on the F1 generation, a subset of exposed females was pairedwith unexposed males. The results showed that hatching rates were notaffected. However, embryo malformations largely with curved spines wereobserved in 50 and 250 µg l–1 of PFOS treated groups and resulted in mortality.Taken together, the study suggested that long-term maternal exposure zebrafishto lower concentration of PFOS could impair development in the F1 generation,and the risk assessment of estrogenic endocrine disruption of PFOS in theaquatic environment is likely small.

Keywords: PFOS, development, vitellogenin, hepatotoxicity, thyroid hormone,zebrafish

INTRODUCTION

The uses of perfluorinated chemicals (PFCs) as surfactants for commercial andindustry applications during several decades have resulted in a global distributionof stable precursors/metabolites, such as perfluorooctanesulfonate (PFOS),perfluorooctanoic acid (PFOA) in the aquatic environment, wildlife and human(e.g., Giesy and Kannan, 2001; Kannan et al., 2002; Hoff et al., 2003, 2005;

Interdisciplinary Studies on Environmental Chemistry—Biological Responses to Chemical Pollutants,Eds., Y. Murakami, K. Nakayama, S.-I. Kitamura, H. Iwata and S. Tanabe, pp. 37–54.© by TERRAPUB, 2008.

38 Y. DU et al.

Martin et al., 2004; Dai et al., 2006; So et al., 2007; reviewed by Lau et al., 2004,2007; Houde et al., 2006a, b). PFOS is believed to be an end-stage metabolite andenvironmental degradation product of N-alkyl-perfluorooctanesulfonamide(PFSM) chemistries used in a wide range of commercial products (3M, 2003).PFOS is resistant to biotic or abiotic degradation and thus stable in the environment,furthermore, it can be bioaccumulated through food chain, which is most abundantand detected in blood and liver samples from various aquatic mammals, birds,fish and humans and thus regarded emerging persistent organic pollutants(POPs). The persistence of PFOS in wildlife and aquatic environment has causedgreat concerns over its toxicity.

In the aquatic environment, detected PFOS concentrations in surface waterare generally low (<0.1 µg l–1). However, higher concentrations of PFOS havebeen detected in fish. For instance, the measured PFOS in the liver was up to 7,760µg kg–1 wet weight from plaice (Pleuronectes platessa) (Hoff et al., 2003), and9,031 µg kg–1 wet weight from feral gibel carp (Carassius auratus gibelio) inBelgium (Hoff et al., 2005). It is worth noting that high concentrations of PFOSwere also detected in fish eggs (145–381 µg kg–1) in lake whitefish (Coregonusclupeaformis) from Michigan waters, USA (Kannan et al., 2005), suggestingoviparous transfer of this compound. Laboratory study also demonstrated thatwater-borne exposure PFOS can be bioaccumulated in rainbow trout(Oncorhynchus mykiss) (Martin et al., 2003), and oviparous transfer in fatheadminnow (Pimephales promelas) (Ankley et al., 2005). For example, the meanconcentrations of PFOS in ovary were 64,600 µg kg–1 after fathead minnowsexposure to 30 µg l–1 PFOS for 21 d.

Based on existing data, PFOS has been shown to influence membranefunction and structure of hepatocytes, as assessed by increase in serum alanineaminotransferase (ALT) activity in carp (Cyprinus carpio) (Hoff et al., 2003) andhepatic PFOS concentration was significantly and positively related to serumALT activity in both feral carp (C. carpio) and eel (Anguilla anguilla) (Hoff etal., 2005). In reproductive and developmental toxicity, it appeared that plasmaandrogens and estrogens can be affected after fathead minnow (Pimephalespromelas) exposed to PFOS (Oakes et al., 2004, 2005; Ankley et al., 2005). PFOSexposure to zebrafish embryos resulted in developmental toxicity and alteredcertain gene expression (Shi et al., 2008). Recently, several studies showed thatestrogenic properties of PFOA (induction of vitellogenin, VTG) in rare minnow(Gobiocypris rarus) (Wei et al., 2007), induction of VTG in cultured male tilapiahepatocytes (Liu et al., 2007) and in male medaka (Oryzias latipes) treated withfluorotelomer alcohol (FTOHs) (Ishibashi et al., 2008).

Despite the widespread of PFOS detected in the environment as well asseveral studies of toxicity in fish, compared to other POPs, relatively little isknown concerning the long-term toxicity of PFOS in fish. Present availableecotoxicological information has been derived from short-term test andecotoxicological consequence in aquatic environment has been largely unknown(Oakes et al., 2004). For environmental risk assessment, a comprehensiveunderstanding on the temporal responses (i.e. the time for initial induction,

Chronic Effects of Waterborne PFOS Exposure 39

maximum induction, adaptation and recovery) of stress in fish at various biologicalhierarchies is essential (Wu et al., 2005). Therefore, the aims of the present studywere (1) to investigate the effects of PFOS on growth and reproductive toxicityin the zebrafish under chronic exposure basis for estimating ecological risk; (2)to study the reversibility of the morphological changes in fish liver after thecessation of PFOS exposure; (3) to select VTG gene expression as biomarker andevaluate whether PFOS exposure could impair fish reproduction through endocrinedisrupted activities. The developing zebrafish fry (14 dpf) were chosen, since itis well-known that early life stage is more sensitive stage to toxicant stress. Thezebrafish were exposed to various concentrations of PFOS until sex mature andthe eggs from the maternal exposure were fertilized with those unexposed male,which may be useful for evaluation of the effects of maternal derived PFOS onthe offspring.

MATERIALS AND METHODS

Chemicals

Heptadecafluorooctanesulfonic acid potassium salt (PFOS, >99%) wasobtained from Tokyo Kasei Kogyo Co. Ltd (Tokyo, Japan) and stock solution wasprepared by dissolving crystal in HPLC-grade dimethyl sulfoxide (DMSO) andstored at 4°C. MS-222 (3-aminobenzoic acid ethyl ester, methanesulfonate salt)was obtained from Sigma (St. Louis, MO, USA). All other chemicals used in thepresent study were analytical grade.

Fish

Adult zebrafish (Danio rerio) (AB strain) were maintained in charcoal-filtered recalculating aerated tap water. The light period was 12h:12 h light: darkcycle and the temperature was controlled at 27 ± 0.5°C. Fish were fed twice withfreshly hatched Artemia nauplii and once flake food daily (Tetra, Germany).Fertilized eggs were collected and examined under stereomicroscope. The frywere maintained until 14 d post-fertilization (dpf) for subsequent experiments,since this is the time when the fry are stably survival.

Experimental design

The fry were randomly distributed into 20L glass tank container for controland exposure groups. There have three replicates in each group and containedabout 140 fry. The fry were exposed in a semi-static system and the water wasrenewed half every other day. Both the control and exposure groups receivedDMSO (0.002%, v/v), respectively. From our primary experiment, nominalconcentrations of PFOS were 10, 50 and 250 µg l–1, in which fry could surviveunder these concentrations. The exposure regime included PFOS exposure period(70 d) and recovery in clean water (30 d). At 40 and 70 d exposure, about 30 fishfrom each treatment group were randomly selected for measurement of length,weight, meanwhile, fish were sampled for histological examination of the gonads

40 Y. DU et al.

and liver, VTG gene expression and T3 measurement. Remaining fish weretransferred to 20 L glass tank and reared in dechlorinated municipal tap water toallow for 30 d recovery. The fish were anesthetized in 0.03% MS-222. The bodyweight, length and gonad weight were recorded. The gonadal-somatic index (GSI= gonad weight × 100/body weight-gonad weight) and condition factor (K =weight (g) × 100/length (cm3)) were calculated. The gonads of adult fish wereexamined under a dissecting microscope to evaluate the sex. On the other hand,after 70 d exposure, the exposed female fish were removed and placed into cleanwater and were paired with unexposed male fish. The hatching rates, malformation,survival and development in the F1 embryo-larvae were assessed.

Whole fish and liver were frozen in liquid nitrogen and stored at –80°C forthyroid hormone analysis and VTG gene expression, respectively. During theexperimental period, residual food and faces in the test chamber were removedevery day and 50% of the exposure water was renewed once in 3 d. The testchambers were cleaned once every two week.

Histology

For histological examination, tissue of liver and the middle portion of testis(40, 70 d exposure and 30 d recovery) were fixed in Bouine fixative solution.Tissues were dehydrated in ethanol, embedded in paraffin wax and sectioned at4–5 µm. The sections were stained with haematoxylin and eosin (H and E) andexamined under light microscope.

RNA extraction and Quantitative real-time PCR assay

Total RNA was isolated from liver using 1 ml Trizol (Invitrogen Inc., USA)following the manufacture’s protocol. The purity of extracted total RNA wasdetermined by UV spectrophotometry (260 nm reading and 260/280 ratio). AllRNA samples were DNase treated prior to be converted to cDNA. A 2 µg totalRNA was converted to cDNA using M-MLV Reverse Transcriptase (Promega,Madison, WI, USA) with random primers, according to the manufacturer’sinstructions. Gene specific primer pairs were based on the cDNA sequences ofzebrafish VTG and β-actin (GenBank accession numbers NM001044897 andAF057040). The primers for VTG gene were 5′-AGCTGCTGAGAGGCTTGTTA-3′ and 5′-GTCCAGGATTTCCCTCAGT-3′. Forward and reverse for β-actinwere 5 ′-CGAGCAGGAGATGGGAACC-3 ′ and 5 ′-CAACGGAAACGCTCATTGC-3′. Quantitative PCR was carried out using theSYBR® Green PCR kit (Toyobo, Tokyo, Japan) and analyzed on an ABI PRISM7000 Sequence Detector System (Perkin-Elmer Applied Biosystems, Foster City,CA, USA). The thermal cycle was as follows: 10 min at 95°C, followed by 40cycles at 95°C for 30 s and one cycle at 60°C for 20 s and 72°C for 1 min. For VTGgene, PCR reactions were performed on three replicate samples. The expressionlevel of VTG was normalized to its β-actin mRNA content. Fold change was usedfor evaluating gene expression according to the methods of Ding et al. (2007).

Chronic Effects of Waterborne PFOS Exposure 41

PFO

S40

day

s ex

posu

re70

day

s ex

posu

re30

day

s re

cove

ry

(µg

L−1

)L

engt

h(c

m)

Wei

ght

(g)

Con

ditio

nfa

ctor

Len

gth

(cm

)W

eigh

t(g

)C

ondi

tion

fact

orL

engt

h(c

m)

Wei

ght

(g)

Con

ditio

nfa

ctor

02.

10 ±

0.0

70.

12 ±

0.0

11.

31 ±

0.0

42.

95 ±

0.0

50.

47 ±

0.0

21.

78 ±

0.0

43.

63 ±

0.0

41.

00 ±

0.0

52.

1 ±

0.1

102.

10 ±

0.0

30.

13 ±

0.0

11.

38 ±

0.0

32.

87 ±

0.0

40.

43 ±

0.0

21.

83 ±

0.0

43.

39 ±

0.0

8*0.

84 ±

0.0

52.

13 ±

0.0

5

502.

18 ±

0.0

40.

15 ±

0.0

11.

43 ±

0.0

32.

84 ±

0.0

30.

42 ±

0.0

21.

85 ±

0.0

53.

32 ±

0.0

3**

0.79

± 0

.04*

2.16

± 0

.09

250

2.14

± 0

.03

0.14

± 0

.01

1.49

± 0

.05*

2.95

± 0

.05

0.47

± 0

.02

1.86

± 0

.08

3.22

± 0

.06*

*0.

66 ±

0.0

2**

1.99

± 0

.06

Tab

le 1

. E

ffec

t of P

FO

S e

xpos

ure

on g

row

th o

f fem

ale

zebr

afis

h. T

he 1

4 dp

f zeb

rafi

sh w

ere

expo

sed

to 0

(con

trol

), 1

0, 5

0 an

d 25

0 µ g

PF

OS

L–1

for 4

0 an

d 70

d a

nd a

fter

reco

very

for 3

0 d.

Dif

fere

nces

in e

ach

para

met

er w

ere

asse

ssed

usi

ng o

ne-w

ay a

naly

sis

of v

aria

nce

and

Tuk

ey’s

mul

tipl

e ra

nge

test

s. V

alue

s re

pres

ent

mea

ns ±

SE

M.

*P <

0.0

5; *

* P <

0.0

1. n

= 1

5–20

.

PFO

S40

day

s ex

posu

re70

day

s ex

posu

re30

day

s re

cove

ry

(µg

L−1

)L

engt

h(c

m)

Wei

ght

(g)

Con

ditio

nfa

c tor

Len

gth

(cm

)W

eigh

t(g

)C

ondi

tion

fac t

orL

engt

h(c

m)

Wei

ght

(g)

Con

ditio

nfa

c tor

01.

96 ±

0.0

40.

10 ±

0.0

11.

31 ±

0.0

32.

91 ±

0.0

40.

34 ±

0.0

11.

35 ±

0.0

33.

07 ±

0.0

30.

43 ±

0.0

11.

48 ±

0.0

3

102.

01 ±

0.0

40.

11 ±

0.0

31.

42 ±

0.0

42.

9 ±

0.03

0.34

± 0

.01

1.38

± 0

.03

2.99

± 0

.04

0.41

± 0

.01

1.52

± 0

.03

502.

09 ±

0.0

40.

13 ±

0.0

1*1.

42 ±

0.0

42.

77 ±

0.0

40.

31 ±

0.0

11.

47 ±

0.0

2*2.

77 ±

0.0

3**

0.33

± 0

.01*

*1.

57 ±

0.0

5

250

1.98

± 0

.04

0.12

± 0

.00*

1.49

± 0

.04*

*2.

65 ±

0.0

5**

0.28

± 0

.01*

*1.

5 ±

0.03

**2.

78 ±

0.0

5**

0.36

± 0

.01*

*1.

71 ±

0.0

5**

Ta b

le 2

. E

ffe c

t of

PF

OS

exp

osur

e on

gro

wth

of

ma l

e z e

bra f

ish.

The

14

dpf

z ebr

a fis

h w

e re

e xpo

sed

to 0

(con

trol

), 1

0, 5

0 a n

d 25

0 µg

PF

OS

L–1

for 4

0 a n

d 70

d a

nd a

fte r

rec o

very

for 3

0 d.

Dif

fere

nce s

in e

a ch

para

me t

e r w

e re

a sse

sse d

usi

ng o

ne-w

a y a

naly

sis

of v

a ria

nce

a nd

Tuk

ey’s

mul

tipl

e ra

nge

test

s. V

a lue

s re

pre s

e nt

me a

ns ±

SE

M.

*P <

0.0

5; *

*P <

0.0

1. n

= 1

5–20

.

42 Y. DU et al.

Thyroid hormone extraction and radioimmunoassay

Thyroid hormone (TH) extraction procedure was followed the methoddeveloped by Crane et al. (2004). Fish were homogenized in 2 ml methanolcontaining 1 mM 6-N-propylthiouracil (PTU) and 0.1M NaOH, then thehomogenates were sonicated, vortexed and centrifuged with 3,700 g for 10 minat 4°C. The supernatants were removed to clean tube and a 2 ml methanolcontaining 1 mM PTU was added, vortexed and centrifuged. The supernatantswere removed, a 5 ml chloroform and 0.5 ml NH4OH (2 M) were added andcentrifuged. The supernatants were transferred to new tube. Finally, a 0.5 mlNH4OH (2 M) were added and repeated the above step. The supernatants thatcollected were evaporated to dryness with nitrogen and stored at –80°C. Allextracts were re-suspended in 400 µl NaOH (0.1 M) for T3 RIA analysis. Theefficiency of the thyroid hormone extraction protocol was determined by adding100 µl of 125I radiolabelled T3 to each adult fish homogenates prior to extraction.Minimum detectable levels of T3 in the RIA were 0.5 ng ml–1.

Statistical analysis

Normality of data was verified using the Kolmogorov-Smirnov test, andhomogeneity of variances checked by Levene’s test. One-way analysis of variance(ANOVA) and Tukey’s multiple range tests were used to determine significantdifference between control and exposure groups. Chi-Square test was used foranalyzing the data of sex ratio and malformation. All statistical analysis wasperformed using SPSS 13.0 soft wear (SPSS, Chicago, IL, USA) and P < 0.05 wasconsidered statistically significant.

RESULTS

Survival, growth and condition factor

There were no significant effects of exposure to PFOS on fish survival forany of the exposure regimes for all experiments. No mortality was observedduring the exposure period. PFOS did not affect body length and weight in thefemales in all PFOS-treated groups after 40 and 70 d exposure (Table 1), whilereduced body weight and length were observed after 30 d recovery in the 50 and250 µg l–1 of PFOS-treated groups compared to control fish (Table 1). Nosignificant differences were observed in the condition factor (K) between thesolvent control and treatment groups. In the male fish, a small but significantlyincreased in the weight was observed in the 50 and 250 µg l–1 PFOS-treatedgroups for 40 d (Table 2). After 70 d exposure, the body weight and length wassignificantly reduced in the 250 µg l–1 PFOS-treated group. The decreased bodyweight and length were further recorded in the 50 and 250 µg l–1 PFOS-treatedgroups after 30 d recovery (Table 2). Condition factor (K) of the males from thisexperiment in the 50 and 250 µg l–1 group was significantly higher compared tothe solvent group (Table 2).

Chronic Effects of Waterborne PFOS Exposure 43

Fig. 1. Mean (±S.E.M.) gonadosomatic index (GSI; [testes weight/body weight]∗100]) of the femalefish after exposure to 0, 10, 50 and 250 µg l–1 PFOS for 70 d and recovery for 30 d. ANOVA andTukey’s multiple range tests were used to determine significant difference between control theexposure group. *Indicates significantly different (P < 0.05) of GSI compared to the controlgroup. N = 10–20.

Fig. 2. Sex ratio of zebrafish (Danio rerio) treated with PFOS from 14 dpf until sexual mature withPFOS concentrations of 0, 10, 50 and 250 µg l–1, respectively. Black (�) represents percentageof males, white (�) fish percentage of females. A χ2-test indicated no statistically significanteffects of the PFOS treatment compared with the unexposed control group.

44 Y. DU et al.

GSI, intersex and Sex Ratio

After 70 d exposure, the gonads of the female fish were removed and the GSIwas calculated. The average GSI in the 50 and 250 µg l–1 of PFOS treated groupswas significantly (P < 0.05) lower than the control fish (Fig. 1). Lower GSI wasfurther observed in the 50 and 250 µg l–1 exposure groups after 30 d recovery (Fig.1). In the present study, testes were also examined and morphological inter-sexwas not observed in all the exposure groups in the male fish. A male to femalegonadal sex ratio of 72:28 was present in the solvent-control group. The recordedof females was 34.1%, 36.4% and 40.1% in the 10, 50 and 250 µg l–1 PFOS treatedgroups, respectively (Fig. 2). The results of Chi-square analysis for trends (P >0.05) and for sex ratio was not show significantly changed in the exposure groupscompared with the control (P > 0.05).

Liver histology

For histological examination, 4–5 male and female from each group weresampled. In the control fish, the liver of zebrafish appeared soft and the color issanguine. In the male fish, after 40 d exposure, there were no obvious morphologicalalternations in the lower concentration (10, 50 µg l–1) exposure groups comparedwith the control. However, exposure of 250 µg l–1 of PFOS for 40 d, the liver

Fig. 3. Light micrographs of liver from zebrafish. (A) Male fish of control group, X400; (B) malefish exposed to PFOS for 40 d at 250 µg l–1; (C) Control female; (D) exposed female for 40 dat 250 µg l–1 PFOS. Accumulation of lipid droplets (arrow B) can be observed. X400.

Chronic Effects of Waterborne PFOS Exposure 45

seemed brittle and pale in all the male fish. By histological examination of liverstructure, the homogeneous parenchyma of zebrafish liver from control fish wascomposed of hepatocytes arranged in a typically three-dimensional architecture(Fig. 3A). After 40 d exposure to 250 µg l–1 of PFOS, the most marked change wasvacuolization mainly due to accumulation of lipid droplets (Fig. 3B). Likewise,similar alternation could be observed in the 250 µg l–1 PFOS treated male fish andin the group after 40 d exposure. Furthermore, the same symptoms were stillobserved in the 250 µg l–1 PFOS-treated males after recovery by 30 d. In thefemale fish, the control structure is shown in Fig. 3C, no obvious treatment relatedmorphological differences were observed in all the female fish during theexposure and recovery period (Fig. 3D).

Fig. 4. Light micrographs of testis from zebrafish. (A) Testis of control male, showing all stages ofspermatogenesis, including spermatocyte and spermatid (X200); (B) Testis of male exposed toPFOS of 250 µg l–1 for 70 d, showing a testis lacking the later stages of spermatogenesis, suchas spermatid, 200×. SG, spermatogonia; SC, spermatocyte; ST, spermatid; n = 6.

46 Y. DU et al.

Testis histology

After 40 days exposure, all sampled fish were sexually differentiated. Thetestis of the mature zebrafish contained all spermatogenetic cells, such asspermatogonia, spermatocyte and spermatid. In the solvent control group, fewerspermatogonia could be observed and the structure of cysts major containingspematocytes and spermatid (Fig. 4A). However, the testis contained mainlyearly spermatogenetic stages including spermatogonia and primary spermatocytein 250 µg l–1 treated group after 70 d exposure, suggesting developmental delay

Fig. 5. The effects of PFOS on mRNA expression in zebrafish after treated with 0.002% DMSO, 10,50 and 250 µg l–1 PFOS for 40 and 70 d and recovery for 30 d. The mRNA levels were measuredby real-time PCR and normalized using β-actin as internal standard, results are means from fourindividual fish (mean ± SEM) and expressed as fold relative to DMSO control. (A) Male fish;(B) Female fish. Significant differences between control and exposure groups is indicated by *P< 0.05; **P < 0.01.

Chronic Effects of Waterborne PFOS Exposure 47

(Fig. 4B). There were no obvious histological alternations in the exposure groups,suggesting the lack of fundamental changes in testicular morphology in this studysupports the absence of reproductive impairment in response to PFOS exposureunder these concentrations.

VTG gene expression

The mRNA expression of VTG was detected at 40, 70 d exposure and 30 drecovery in both male and female fish. In male fish, VTG was significant up-regulated by 1.84-fold and 3.73-fold in 50 and 250 µg l–1 PFOS treated group after40 d exposure; A significant induction of VTG could further be observed in 10 µgl–1 PFOS (3.13-fold) and 50 (5.09-fold) µg l–1 after 70 d PFOS exposure; After 30d recovery, A strong up-regulation (p < 0.01) of VTG gene expression wasdetected in 50 µg l–1 (16.08-fold) PFOS exposure group (Fig. 5A). In female fish,similar induction (p < 0.01) of VTG transcript levels was also observed in 250 µgl–1 treated group. No significant differences were found between control andtreated groups after 70 d exposure. After recovery, VTG transcript level weresignificantly decreased (P < 0.01) in the 250 µg l–1 PFOS treated group (Fig. 5B).

Thyroid hormone (T3) concentration

The total T3 level (TT3) from the whole body homogenate showed nosignificant difference between the control and treatment group after 40 and 70 dexposure and after 30 d recovery (Table 3).

Hatching, fecundity, deformation and survival in the F1 generation

After 70 d exposure, the exposed females were removed into clean water andwere paired with unexposed males. No significant differences in the total numberof eggs per breeding trial were observed between the solvent control group andthe treatment groups (data not shown). The hatching rates were 74.3%, 78.3%,81.25% and 82.2% in the control, 10, 50 and 250 µg l–1 of PFOS-treated groups,respectively, and showed no significant difference. The malformation was not

Table 3. Effect of PFOS exposure on T3 concentration of the zebrafish (ng g–1). The 14 dpf zebrafishwere exposed to 0 (control), 10, 50 and 250 µg PFOS L–1 for 40 and 70 d and after recovery for30 d. Differences were assessed using one-way analysis of variance. Values represent means ±SEM. n = 5–6.

PFOS Female Male

(µg L–1)40 days

exposure70 daysexposure

30 daysrecovery

40 daysexposure

70 daysexposure

30 daysrecovery

0 6.83 ± 2.58 4.17 ± 0.31 5.61 ± 0.53 4.20 ± 1.19 3.14 ± 0.12 4.29 ± 0.1910 2.74 ± 0.26 4.45 ± 0.39 4.81 ± 0.65 6.55 ± 1.89 3.12 ± 0.31 4.65 ± 0.4650 3.76 ± 0.79 4.38 ± 0.31 5.97 ± 0.61 7.62 ± 1.02 2.64 ± 0.18 4.03 ± 0.19

250 6.34 ± 0.95 5.49 ± 0.57 4.92 ± 0.98 6.04 ± 1.97 3.99 ± 0.22 4.80 ± 0.26

48 Y. DU et al.

observed in the control and 0.01 µg l–1 PFOS-treated fish. However, the observedpercentage of malformations was significantly higher in the 50 and 250 µg l–1 ofPFOS-treated groups with 37.5% and 100% and all the deformed larvae died after96h hatching, while those larvae appeared normal could survive. The mostpronounced malformation was skeletal deformities with curved spines.

DISCUSSION

Zebrafish fry (14 dpf) were selected for exposure, since early life stage ismore sensitive to toxicants stress. In addition, we conducted partial life-cycle testand studied the effects of PFOS on the deformation and survival of F1 generation.Compared with previous studies in fish, the present study employed lowerconcentration of PFOS, which detected in fish. In the present study, we wereunable to determine actual test PFOS concentrations during exposures, and thuswe reported nominal concentrations. However, we employed a static-renewalexposure regime with 50% test chemical replacement every 3 d and thus consistentPFOS concentrations would be expected. There was no mortality in all the treatedgroups (10, 50, 250 µg l–1) in all the exposure regimes. The results are consistentwith a previous study using adult fathead minnow, which showed no mortalityafter exposure to 300 µg l–1 PFOS, but exposure to 1,000 µg l–1 for 21 d reducedsurvival of the fathead minnow (Ankley et al., 2005). Significant reduction oftotal body length and weight was only observed for male fish raised undercontinuous exposure to 250 µg l–1 PFOS, whereas no effects were observed infemale fish, suggesting the inhibition of growth may be gender-specific. Ankelyet al. (2005) reported that no significant adverse effects on growth were observedin developing fathead minnows held for 24 d at PFOS concentrations up to 300µg l–1. In mammals, several sub-chronic exposures of rats, mice and monkeys toPFOS have resulted in effects on body weight gain in females and in males(Seacat et al., 2002, 2003; Thibodeaux et al., 2003; Luebker et al., 2005a, b).Overall, the results from the mammalian and our study indicate that chronic PFOSexposure has similar effects on growth. However, PFOS exposure does notappear to affect condition factor (K) in the male and female fish. In contrast,Oakes et al. (2004) showed that the condition factor of fathead minnow in thefemale was significantly decreased, while no difference was observed in the malefish after exposure to 300 µg l–1 for 39 d. Condition factor is an easily measuredparameter and has been shown to be a useful indicator of toxic stress to organicpollutants in fish (Anderson et al., 2003). Furthermore, it is reflective of changesin food intake, lipid deposition and protein budgets (Smolders et al., 2003). Theunchanged condition factors in the female fish in the present study may suggestthat the overall health and somatic fitness of zebrafish was not compromised. Theincreased K in the male fish could be related to the accumulation of lipid dropletsin the liver, since study has shown that PFOS exposure caused increasedintracellular free fatty acids and free cholesterol in the liver (Seacat et al., 2002).

In order to evaluate the effects of PFOS exposure on the F1 generation, theexposed female fish were paired with unexposed males. A significant malformationas well as mortality was observed in 50 and 250 µg l–1 PFOS treated-groups.

Chronic Effects of Waterborne PFOS Exposure 49

Unfortunately, we were unable to measure PFOS concentrations in the eggs. Theresults may suggest that oviparous transfer of the compound from adult tooffspring and adult exposure (F0) could significant affect F1 generation survival.In contrast, a previous study showed that PFOS could be transferred to gonads infathead minnows (Ankley et al., 2005), however, Ankley and co-authors did notshow any abnormal change and survival of the embryos and larvae derived fromadults exposed to 300 µg l–1 PFOS. In mammals, PFOS has been shown in rats andmice throughout pregnancy, to severely affect postnatal survivor of neonatal ratsand mice (Thibodeaux et al., 2003; Luebker et al., 2005a, b; Fuentes et al., 2006,2007).

Relative gonad size (GSI) was significantly smaller in the female fish afterexposure to 250 µg l–1, suggesting PFOS could reduce fish reproductive potential.The quality of eggs as well as the offspring viability would depend on GSI andthe levels of sex hormones produced during gametogenesis (Kime, 1998). Ankelyet al. (2005) reported that adult fathead minnow exposed to 300 µg l–1 PFOS for21 d exhibited decreased aromatase activity and elevated concentrations ofplasma 11-ketotestosterone and testosterone. Oakes et al. (2004) reported thatexposure of fathead minnows to PFOA, did not find any significant changes infemale GSI, but altered plasma concentrations of both steroidal androgens andestrogens and they suggested that as reduced circulation steroid hormones areassociated with reduced GSI, reduction of GSI might be expected given theobserved reduction of 17β-estradiol.

The results presented confirm the estrogenic activities of PFOS in zebrafish.VTG is synthesized in the liver of female fish in response to estradiol-17β (E2)-induced estrogen receptor (ER) activation. Measurement of vitellogenin (VTG)mRNA or VTG levels in male fish has been one of the most commonly usedbiomarkers for exposure to estrogenic activities of chemicals in the aquaticenvironment (Denslow et al., 1999; Jobling et al., 2003). No sex chromosomeshave been identified in zebrafish, thus environmental factors (e.g., sex steroidsand xeno-estrogens) could affect on the processes of sex differentiation. Ingeneral, the sex differentiation fish around 45 dpf in zebrafish. For example,changes in the proportions of males and females were observed in populations ofjuvenile zebrafish exposed, during the critical period of gonadal development, toxenoestrogens (Örn et al., 2003, 2006; Drastichova et al., 2005). Circulation oftestosterone was reduced while E2 was evaluated after exposure to 1000 µg l–1

PFOA in fathead minnow (Oakes et al., 2004). Although several studies as wellas our study showed estrogenic activities of PFOS, PFOA in fish or culturedtilapia hepatocytes (Oakes et al., 2005; Wei et al., 2007; Liu et al., 2007), sex ratiowas not changed as well as intersex was not observed. This result could beexplained as the weak estrogenic activity of PFOS. Our results suggest that geneexpression may be more sensitive than gross morphological endpoints (e.g., GSI,intersex, sex ratio) and different endpoints should be employed when evaluatingenvironmental estrogenic activities of chemicals in the aquatic environment.

Histological examination revealed that the most pronounced morphologicalalternation is accumulation of lipid in the liver of male fish, suggesting the

50 Y. DU et al.

hepatic toxicity is gender specific. Moreover, such a “recovery/adaptive” responsewas not observed after recovery, as indicated by lipid droplets accumulation inthe hepatocytes. Liver is a major target organ for PFOS accumulation. PFOS hasalso been shown to accumulate in wildlife by binding to proteins in blood andliver tissues (Giesy and Kannan, 2001). Seacat et al. (2002) showed that PFOSaccumulates in liver and induces hepatocellular hypertrophy, centrilobularvacuolation, and mild bile stasis, lipid-droplet accumulation in both male andfemale rats. It is possible that these affects may be directly attributed to PFOSbinding in the hepatocytes. The gender-specific toxicity may be due to theelimination of PFOS and interaction between PFOS and endogenous steroidhormones. Study has shown that the elimination of PFOA is related to testosteroneand cortisol. Since PFOS is metabolically inert, sex-related difference observedin the biological effects of PFOS is not due to a difference in metabolism of PFOS.By using treatment of estradiol or testosterone in rats, Kudo et al. (2002) showedthat PFOA is transported into urine by a several organic anion transporters andsex hormones regulated the transporters gene expression differentially in kidneyof male and female adult rats. So far the gender difference in elimination of PFOSin fish is not well known. Future research on pharmacokinetics of PFOS in fishmay be warranty.

Thyroid hormones are known to regulate growth and development in fish. Infish, thyroid hormones have a major role on regulation of growth and developmentthat can be affected as a result of xenobiotic exposure (Arcand-Hoy and Benson,1998; Power et al., 2001). In zebrafish, thyroid hormones are necessary forembryonic to larval transition and larval to juvenile transition (Liu and Chan,2002). Hence, if thyroid disruption occurs during either transition, growthretardation might occur (Brown, 1997). In the present study, total T3 levelsremained unchanged after PFOS exposure for 40 and 70 d as well as afterrecovery. The results suggested that chronic PFOS exposure did not arrestedthyroid hormone production. In contrast, significantly reduced in the levels ofserum thyroxine (T4) and triiodothyronine (T3) have been measured in rats andmice after exposure to PFOS (Seacat et al., 2002; Thibodeaux et al., 2003; Lauet al., 2003; Luebker et al., 2005a, b; Chang et al., 2008), whereas changes inthyroid gland histology or clinically significant elevations of thyroid stimulatinghormone (TSH) did not occur. Recently, Chang et al. (2008) studied PFOSexposure to rats and hypothesized that exposure to PFOS may increase freethyroxine (FT4) in the rat serum due to the ability of PFOS to compete withthyroxine for binding proteins and increase in FT4 would increase the availabilityof the thyroid hormone to peripheral tissues for utilization, metabolic conversation,and therefore increase T3. Since thyroid hormones play important roles indevelopment, further investigations are necessary to elucidate the changes onthyroid hormone due to PFOS exposure.

In summary, results from the present study indicated that long-term PFOSexposure could result in significant mortality and impair fish reproduction in theoffspring. Although hepatic VTG gene expression was significantly up-regulatedand showed estrogenic activities in zebrafish, sex ratio was not changed, suggesting

Chronic Effects of Waterborne PFOS Exposure 51

that lower concentrations of PFOS does not affect fish due to endocrine disruption,but may be through bioaccumulation through food web and impairment on theoffspring in the aquatic environment. Although the concentration of PFOS wasgeneral low (0.1 µg l–1) in surface water measured from North America and Asia(Hansen et al., 2002; Taniyasu et al., 2003; Boulanger et al., 2004; Tseng et al.,2006; So et al., 2007; Zhao et al., 2007; Becker et al., 2008) and the exposureconcentration of PFOS was relatively higher than those detected in surface waterin the present study, it should be noted that PFOS can be bioaccumulated throughfood chain and can be oviparous transferred to offspring. Therefore, in the aquaticenvironment, PFOS can be magnified and bioaccumulated in fish and exhibitedreproductive toxicity in the offspring and may lead to ecological consequence.

Acknowledgments—The present study was supported by the National Nature ScienceFoundation of China (20577066) and a Science and Technology Development Projectfrom Hubei Province (2006AA301B47).

REFERENCES

Anderson, M. J., D. Cacela, D. Beltman, S. J. Teh, M. S. Okihiro, D. E. Hinton, N. Denslow and J.T. Zelikoff (2003): Biochemical and toxicopathic biomarkers assessed in smallmouth bassrecovered from a polychlorinated biphenyl-contaminated river. Biomarkers, 8, 371–393.

Ankley, G. T., D. W. Kuehl, M. D. Jensen, K. M. Linnum, A. Linnum, R. L. Leino and D. A.Villeneuve (2005): Reproductive and developmental toxicity and bioconcentration ofperfluorooctanesulfonate in a partial life-cycle test with the fathead minnow (Pimephalespromelas). Environ. Toxicol. Chem., 24, 2316–2324.

Arcand-Hoy, L. D. and W. H. Benson (1998): Fish reproduction: an ecologically relevant indicatorof endocrine disruption. Environ. Toxicol. Chem., 17, 49–57.

Becker, A. M., S. Gerstmann and H. Frank (2008): Perfluorooctane surfactants in waste waters, themajor source of river pollution. Chemosphere, 72, 115–121.

Boulanger, B., J. Vargo, G. Schnoor and K. Hornbuckle (2004): Detection of perfluorooctanesurfactants in Great Lakes water. Environ. Sci. Technol., 38, 4064–4070.

Brown, D. D. (1997): The role of thyroid hormone in zebrafish and axolotl development. Proc. Natl.Acad. Sci. USA, 94, 13011–13016.

Chang, S. C., J. R. Thibodeaux, M. L. Eastvold, D. J. Bjork, J. W. Froehlich, C. Lau, R. J. Singh, K.B. Wallace and J. L. Butenhoff (2008): Thyroid hormone status and pituitary function in adultrats given oral doses of perfluorooctanesulfonate (PFOS). Toxicology, 243, 330–339.

Crane, H. M., D. B. Pickford, T. H. Hutchinson and J. A. Brown (2004): Developmental changes ofthyroid hormones in the fathead minnow, Pimephales promelas. Gen. Comp. Endocrinol., 139,55–60

Dai, J. Y., M. Li, Y. H. Jin, N. Saito, M. Q. Xu and F. W. Wei (2006): Perfluorooctanesulfonate andperfluorooctanoate in Red Panda and Giant Panda from China. Environ. Sci. Technol., 40, 5647–5652.

Denslow, N. D., M. C. Chow, K. J. Kroll and L. Green (1999): Vitellogenin as a biomarker ofexposure for estrogen or estrogen mimics. Ecotoxicology, 8, 385–398.

Ding, L., M. B. Murphy, Y. He, Y. Xu, L. W. Y. Yeung, J. Wang, B. Zhou, P. K. S. Lam, R. S. S.Wu and J. P. Giesy (2007): Effects of brominated flame retardants and brominated dioxins onsteroidogenesis in H295R human adrenocortical carcinoma cell line. Environ. Toxicol. Chem.,26, 764–772.

Drastichová, J., Z. Svobodová, M. Groenland, R. Dobs̆íková, V. Z̆lábek, D. Weissová and M.Szotkowská (2005): Effect of exposure to bisphenol A and 17β-estradiol on the sex differentiationin zebrafish (Danio rerio). Acta Vet. Brno, 74, 287–291.

Fuentes, S., M. T. Colomina, J. Rodriguez, P. Vicens and J. L. Domingo (2006): Interactions in

52 Y. DU et al.

developmental toxicology: Concurrent exposure to perfluorooctane sulfonate (PFOS) and stressin pregnant mice. Toxicol. Lett., 164, 81–89.

Fuentes, S., M. T. Colomina, P. Vicens, N. Franco-Pons and J. L. Domingo (2007): Concurrentexposure to perfluorooctane sulfonate and restraint stress during pregnancy in mice: effects onpostnatal development and behavior of the offspring. Toxicol. Sci., 98, 589–598.

Giesy, J. P. and K. Kannan (2001): Global distribution of perfluorooctane sulfonate in wildlife.Environ. Sci. Technol., 35, 1339–1342.

Hansen, K. J., H. O. Johnson, J. S. Eldridge, J. L. Butenhoff and L. A. Dick (2002): Quantitativecharacterization of trace levels of PFOS and PFOA in the Tennessee River. Environ. Sci.Technol., 36, 1681–1685.

Hoff, P. T., K. V. Vijver, W. V. Dongen, E. L. Esmans, R. Blust and W. M. De Coen (2003):Perfluorooctane sulfonic acid in bib (Trisopterus luscus) and plaice (Pleuronectes platessa)from the Western Scheldt and the Belgian North Sea: distribution and biochemical effects.Environ. Toxicol. Chem., 22, 608–614.

Hoff, P. T., K. Van Campenhout, K. Van de Vijver, A. Covaci, L. Bervoets, L. Moens, G. Huyskens,G. Goemans, C. Belpaire, R. Blust and W. De Coen (2005): Perfluorooctane sulfonic acid andorganohalogen pollutants in liver of three freshwater fish species in Flanders (Belgium):relationship with biochemical and organismal effects. Environ. Pollut., 137, 324–333.

Houde, M., J. W. Martin, R. J. Letcher and K. R. Solomon (2006a): Biological monitoring ofpolyfluoroalkyl substances: a review. Environ. Sci. Technol., 40, 3463–3473.

Houde, M., T. A. D. Bujas, J. Small, R. S. Wells, P. A. Fair, G. D. Bossart, K. R. Solomon and D.C. G. Muir (2006b): Biomagnification of perfluoroalkyl compounds in the bottlenose dolphin(Tursiops truncatus) food web. Environ. Sci. Technol., 40, 4138–4144.

Ishibashi, H., R. Yamauchi, M. Matsuoka, J. W. Kim, M. Hirano, A. Yamaguchi, N. Tominaga andK. Arizono (2008): Fluorotelomer alcohols induce hepatic vitellogenin through activation of theestrogen receptor in male medaka (Oryzias latipes). Chemosphere, 71, 1853–1859.

Jobling, S., D. Casey, T. Rodgers-Gray, J. Oehlmann, U. Schulte-Oehlmann, S. Pawlowski, T.Baunbeck, A. P. Turner and C. R. Tyler (2003): Comparative responses of molluscs and fish toenvironmental estrogens and an estrogenic effluent. Aquat. Toxicol., 65, 205–220.

Kannan, K., J. Choi, N. Iseki, K. Senthilkumar, D. H. Kim, S. Masunaga and J. P. Giesy (2002):Concentrations of perfluorinated acids in livers of birds from Japan and Korea. Chemosphere,41, 225–231.

Kannan, K., L. Tao, E. Sinclair, S. D. Pastva, D. J. Jude and J. P. Giesy (2005): Perfluorinatedcompounds in aquatic organisms at various trophic levels in a great lakes food chain. Arch.Environ. Contam. Toxicol., 48, 559–566.

Kime, D. E. (1998): Endocrine Discruption in Fish. Kluwer, Boston, 416 pp.Kudo, N., M. Katakura, Y. Sato and Y. Kawashima (2002): Sex hormone-regulated renal transport

of perfluorooctanoic acid. Chem.-Biol. Interact., 139, 301–316.Lau, C., J. R. Thibodeaux, R. G. Hanson, J. M. Rogers, B. E. Grey, M. E. Stanton, J. L. Butenhoff

and L. A. Stevenson (2003): Exposure to perfluorooctane sulfonate during pregnancy in rat andmouse. II. Postnatal evaluation. Toxicol. Sci., 74, 382–392.

Lau, C., J. L. Butenhoff and J. M. Rogers (2004): The developmental toxicity of perfluoroalkyl acidsand their derivatives. Toxicol. Appl. Pharmacol., 198, 231–241.

Lau, C., K. Anitole, C. Hodes, D. Lai, A. Pfahles-Hutchens and J. Seed (2007): Perfluoroalkyl acids:A review of monitoring and toxicological findings. Toxicol. Sci., 99, 366–394.

Liu, C., Y. Du and B. Zhou (2007): Evaluation of estrogenic activities and mechanism of action ofperfluorinated chemicals determined by vitellogenin induction in primary cultured tilapiahepatocytes. Aquat. Toxicol., 85, 267–277.

Liu, W. Y. and W. K. Chan (2002): Thyroid hormones are important for embryonic to larvaltransitory phase in zebrafish. Differentiation, 70, 36–45.

Luebker, D. J., M. T. Case, R. G. York, J. A. Moore, K. J. Hansen and J. L. Butenhoff (2005a): Two-generation reproduction and cross-foster studies of perfluorooctanesulfonate (PFOS) in rats.Toxicology, 215, 126–148.

Luebker, D. J., R. G. York, K. J. Hansen, J. A. Moore and J. L. Butenhoff (2005b): Neonatal mortality

Chronic Effects of Waterborne PFOS Exposure 53

from in utero exposure to perfluorooctanesulfonate (PFOS) in Sprague-Dawley rats: Dose-response, and biochemical and pharamacokinetic parameters. Toxicology, 215, 149–169.

Martin, J. W., S. A. Mabury, S. K. Solomon and D. C. G. Muir (2003): Bioconcentration and tissuedistribution of perfluorinated acid in rainbow trout (Oncorhynchus mykiss). Environ. Toxicol.Chem., 22, 196–204.

Martin, J. W., M. M. Smithwick, B. M. Braun, P. F. Hoekstra, D. C. Muir and S. A. Mabry (2004):Identification of long-chain perfluorinated acids in biota from the Canadian Arctic. Environ. Sci.Technol., 38, 373–380.

Oakes, K. D., P. K. Sibley, J. W. Martin, K. R. Solomon, S. A. Mabury and G. J. Van Der Kraak(2004): Impact of perfluorooctanoic acid on fathead minnow (Pimephales promelas) fatty acyl-CoA oxidase activity, circulating steroids, and reproduction in outdoor microcosms. Environ.Toxicol. Chem., 23, 1912–1919.

Oakes, K. D., P. K. Sibley, J. W. Martin, K. R. Solomon, S. A. Mabury and G. J. Van Der Kraak(2005): Short-term exposures of fish to perfluorooctane sulfonate: Acute effects on fatty acyl-CoA oxidase activity, oxidative stress, and circulating sex steroids. Environ. Toxicol. Chem., 24,1172–1181.

Örn, S., H. Holbech, T. H. Madsen, L. Norrgren and G. I. Petersen (2003): Gonad development andvitellogenin production in zebrafish (Danio rerio) exposed to ethinylestradiol andmethyltestosterone. Aquat. Toxocol., 65, 397–411.

Örn, S., S. Yamani and L. Norrgren (2006): Comparison of vitellogenin induction, sex ratio, andgonad morphology between zebrafish and Japanese medaka after exposure to 17α-ethinylestradioland 17β-trenbolone. Arch. Environ. Contam. Toxicol., 51, 237–243.

Power, D. M., L. Llewellyn, M. Faustino, M. A. Nowell, B. T. H. Björnsson, I. E. Einarsdottir, A.V. M. Canario and G. E. Sweeney (2001): Thyroid hormones in growth and development of fish.Comp. Biochem. Physiol., 130C, 447–459.

Seacat, A. M., P. J. Thomford, K. J. Hansen, G. W. Olsen, M. T. Case and J. L. Butenhoff (2002):Subchronic toxicity studies on perfluorooctanesulfonate potassium salt in cynomolgus monkeys.Toxicol. Sci., 68, 249–264.

Seacat, A. M., P. J. Thomford, K. J. Hansen, L. A. Clemen and S. R. Eldridge (2003): Sub-chronicdietary toxicity of potassium perfluorooctanesulfonate in rats. Toxicology, 183, 117–131.

Shi, X., Y. Du, P. K. S. Lam, R. S. S. Wu and B. Zhou (2008): Developmental toxicity and alterationof gene expression in zebrafish embryos exposed to PFOS. Toxicol. Appl. Pharm., 230, 23–32.

Shi, Z., H. Zhang, Y. Liu, M. Xu and J. Dai (2007): Alterations in gene expression and testosteronesynthesis in the testes of male rats exposed to perfluorododecanoic acid. Toxicol. Sci., 98, 206–215.

Smolders, R., G. D. Boeck and R. Blust (2003): Changes in cellular energy budget as a measure ofwhole effluent toxicity in zebrafish (Danio rerio). Environ. Toxicol. Chem., 22, 890–899.

So, M. K., Y. Miyake, W. Y. Yeung, Y. M. Ho, S. Taniyasu, P. Rostkowski, N. Yamashita, B. S.Zhou, X. J. Shi, J. X. Wang, J. P. Giesy, H. Yu and P. K. S. Lam (2007): Perfluorinatedcompounds in the Pearl River and Yangtze River of China. Chemosphere, 68, 2085–2095.

Taniyasu, S., K. Kannan, Y. Horii, N. Hanari and N. Yamashita (2003): A survey of perfluorooctanesulfonate and related perfluorinated organic compounds in water, fish, birds, and humans fromJapan. Environ. Sci. Technol., 37, 2634–2639.

Thibodeaux, J. R., R. G. Hanson, J. M. Rogers, B. E. Grey, B. D. Barbee, J. H. Richards, J. L.Butenhoff, L. A. Stevenson and C. Lan (2003): Exposure to perfluorooctane sulfonate duringpregnancy in rat and mouse. I: Maternal and prenatal evaluations. Toxicol. Sci., 74, 369–381.

Tseng, C., L. Liu, C. Chen and W. Ding (2006): Analysis of perfluorooctanesulfonate and relatedfluorochemicals in water and biological tissue samples by liquid chromatography-ion trap massspectrometry. J. Chromatog A, 10, 119–126.

Wei, Y., J. Dai, M. Liu, J. Wang. M. Xu, J. Zha and Z. Wang (2007): Estrogen-like properties ofperfluorooctanoic acid as revealed by expressing hepatic estrogen-responsive genes in rareminnows (Gobiocypris rarua). Environ. Toxicol. Chem., 26, 2440–2447.

Wu, R. S. S., W. H. L. Siu and P. K. S. Shin (2005): Induction, adaptation and recovery of biologicalresponses: implications for environmental monitoring. Mar. Pollut. Bull., 51, 623–634.

54 Y. DU et al.

Zhao, X., J. Li, Y. Shi, Y. Cai, S. Mou and G. Jiang (2007): Determination of perfluorinatedcompounds in wastewater and river water samples by mixed himimicelle-based soli-phaseextraction before liquid chromatography-electrospray tandem mass spectrometry detection. J.Chromatogr. A, 1154, 52–59.

3M (2003): Environmental and Health Assessment of Perfluorooctanesulfonate and its Salts.Available on US EPA Administrative Record AR-226.

Y. Du, X. Shi, K. Yu, C. Liu and B. Zhou (e-mail: bszhou@ihb.ac.cn)