or death

Environmental Pollution 214 (2016) 806e815

Contents lists available at ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

Toxicity of perfluorooctane sulfonate and perfluorooctanoic acid to Escherichia coli: Membrane disruption, oxidative stress, and DNA damage induced cell inactivation and/or death* Gesheng Liu a, Shuai Zhang a, Kun Yang a, b, Lizhong Zhu a, b, Daohui Lin a, b, * a b

Department of Environmental Science, Zhejiang University, Hangzhou 310058, China Zhejiang Provincial Key Laboratory of Organic Pollution Process and Control, Zhejiang University, Hangzhou 310058, China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 14 February 2016 Received in revised form 19 April 2016 Accepted 25 April 2016

Perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA) are two widely used polyfluorinated compounds (PFCs) and are persistent in the environment. This study for the first time systematically investigated their toxicities and the underlying mechanisms to Escherichia coli. Much higher toxicity was observed for PFOA than PFOS, with the 3 h half growth inhibition concentrations (IC50) determined to be 10.6 ± 1.0 and 374 ± 3 mg L
1, respectively, while the bacterial accumulation of PFOS was much greater than that of PFOA. The PFC exposures disrupted cell membranes as evidenced by the dose-dependent variations of cell structures (by transmission electron microscopy observations), surface properties (electronegativity, hydrophobicity, and membrane fluidity), and membrane compositions (by gas chromatogram and Fourier transform infrared spectroscopy analyses). The increases in the contents of intracellular reactive oxygen species (ROS) and malondialdehyde and the activity of superoxide dismutase indicated the increment of oxidative stress induced by the PFCs in the bacterial cells. The fact that the cell growth inhibition was mitigated by the addition of ROS scavenger (N-acetyl cysteine) further evidenced the important role of oxidative damage in the toxicities of PFOS and PFOA. Eighteen genes involved in cell division, membrane instability, oxidative stress, and DNA damage of the exposed cells were up or down expressed, indicating the DNA damage by the PFCs. The toxicities of PFOS and PFOA to E. coli were therefore ascribed to the membrane disruption, oxidative stress, and DNA damage induced cell inactivation and/or death. The difference in the bactericidal effect between PFOS and PFOA was supposed to be related to their different dominating toxicity mechanisms, i.e., membrane disruption and oxidative damage, respectively. The outcomes will shed new light on the assessment of ecological effects of PFCs. © 2016 Published by Elsevier Ltd.

Keywords: Polyfluorinated compounds Bacterium Growth inhibition Bioaccumulation Toxicity mechanism

1. Introduction Due to their high thermal, chemical, and biological inertness (Parsons et al., 2008), polyfluorinated compounds (PFCs) have been widely applied to a variety of industrial and consumer products such as lubricants, foam in fire extinguishers, water repellents for paper, leather, carpet, and textiles (Prevedouros et al., 2006). Perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA) are the most commonly used PFCs, and have found their way into

*

This paper has been recommended for acceptance by von Hippel Frank A. * Corresponding author. Department of Environmental Science, Zhejiang University, Hangzhou 310058, China. E-mail address: [email protected] (D. Lin). http://dx.doi.org/10.1016/j.envpol.2016.04.089 0269-7491/© 2016 Published by Elsevier Ltd.

environments (Rahman et al., 2014). They have been detected in the dust (Shoeib et al., 2011), water (Boulanger et al., 2004), sediments (Lam et al., 2014), wildlife (Route et al., 2014), and humans (Zhang et al., 2013b). The determined environmental concentrations of PFOS and PFOA were generally low, such as 0.09e320.50 and 0.14e26.48 ng L
1 in the surface water, respectively (Pan et al., 2014); while their concentrations could be up to 12.566 and 6.57 mg L
1 in the waste water, respectively (Lin et al., 2009; Schultz et al., 2004). They have attracted much attention owing to their global distribution, bioaccumulation, and long half-life in the environment (PFOS, 4.8 a; PFOA, 3.5 a) (Olsen et al., 2007). Their toxicities and/or bioaccumulations to varieties of organisms have been investigated, such as bacteria (Rosal et al., 2010), microalgae (Rodea-Palomares et al., 2012), invertebrates (Sanderson et al., 2004; Xia et al., 2013, 2015), and fish (Chen et al., 2015b).

G. Liu et al. / Environmental Pollution 214 (2016) 806e815

Cell membrane disruption (Liao et al., 2010), oxidative stress (Mao et al., 2012), and/or genetic damage (Liu et al., 2014; Nobels et al., 2010) have been ascribed to be possible toxic mechanisms of PFOS and PFOA. It was indicated that cell membrane would be damaged when exposed to PFOS, since PFOS could bind with membrane phospholipids and thereby prevent the membrane synthesis process (Liao et al., 2010). Increasing evidences indicate that PFOS and PFOA could overwhelm the balance of antioxidant system, and consequently induce oxidative injury and even apoptosis (Mao et al., 2012). Genetic damage, including DNA strand break and fragmentation and chromosomal break, was another mechanism contributing to the toxicity (Liu et al., 2014). However, most previous studies addressed only one or two certain toxicity mechanisms, and no available investigation has analyzed the potential inner connection among the three mechanisms. Moreover, no experimental work has specifically examined the potential difference in the toxicity mechanism between PFOS and PFOA, though a few studies have compared their apparent toxicity effects (Zhang et al., 2013a; MacDonald et al., 2004). Bacteria, as unicellular organisms and ubiquitous in the aquatic ecosystem, are the important model organisms for evaluating the ecotoxicity and investigating the underlying mechanisms at the cellular level. The toxicities of PFOS and PFOA to bacteria have been observed in few research works. Nobels et al. (2010) picked out some genes of Escherichia coli to evaluate the toxicological endpoints of PFCs and found that PFOA expressed higher toxicity than PFOS in terms of oxidative and DNA damages. Another study used Pseudomonas putida to investigate the combined toxicity of PFOA with tetra butyl ammonium (TBA) or Cr3þ and revealed the toxicity in a descending sequence of PFOA, PFOAþCr3þ, and PFOAþTBA (Chen et al., 2013). However, the investigation on the toxicity of PFCs to bacteria is still very limited, and more systematic studies are warranted to address the toxicities and the underlying mechanisms of PFOS and PFOA to bacteria. This study was aimed at elucidating the toxicities and mechanisms of PFOS and PFOA to bacteria. Escherichia coli was selected as the target organism because of its rapid reproduction, easy observation, and clear genetic information. Results of this study will help to understand toxicity mechanisms of PFCs and assess their ecological risks. 2. Materials and methods

807

PI). The SYTO 9 stain generally labels all cells when used alone, while PI can only penetrate cells with damaged membranes, causing a reduction in the SYTO 9 stain fluorescence when both dyes are present. The stained cells were imaged with a fluorescence microscope (FM, Nikon Eclipse Ni, Japan), and the live and dead cells were distinguished. The amounts of stained live and dead cells were further counted with a flow cytometer (BD FACS Calibur, USA). The data were analyzed by using FlowJo7.6.1 to illustrate major findings. The mortality was obtained by the following equation: mortality (%) ¼ amount of dead cells/(amount of dead cells þ amount of live cells)  100. 2.3. TEM observations A drop of bacterial suspension after the exposure to PFOS or PFOA for 3 h was air-dried onto a copper grid and was observed by using a transmission electron microscope (TEM, JEM-1230, Japan). To observe potential changes of the cellular structure, the treated and untreated bacterial cells were fixed in 2.5% glutaraldehyde, dehydrated by ethanol and acetone, embedded, stained with OsO4, sectioned, counterstained, and then observed by the TEM (Lin et al., 2014). 2.4. Measurements of cell surface properties Zeta potentials of bacterial suspensions were measured with a Zetasizer (Nano ZS90, Malvern, U.K.). The cell surface hydrophobicity (CSH) was determined using the method described by Rosenberg et al. (1980) with minor adjustments described in the SI. The membrane fluidity was measured by detecting the fluorescence polarization of the probe, 1, 6-diphenyl-1, 3, 5-hexatriene (DPH) (Sigma-Aldrich Co., USA), as reported previously (Li and Zhu, 2014) with some modifications noted in the SI. The anisotropy values (rDPH) were calculated, which has a reverse correlation with the membrane fluidity. 2.5. Determination of fatty acid compositions in cell membranes Bacterial membrane fatty acids were extracted according to the standard protocol described in the MIDI system (Sasser, 1990) with a few modifications mentioned in the SI. The fatty acid analysis was performed with a gas chromatogram (GC, Agilent 7890A, USA) coupled with a flame ionization detector (FID).

2.1. Bacterial growth assays 2.6. FTIR subtraction spectroscopy PFOS (CAS number 2795-39-3, 98% purity) and PFOA (CAS number 335-67-1, 98% purity) were purchased from J&K Company (Shanghai, China). The stock solutions of PFOS (500 mg L
1) and PFOA (40 mg L
1) were prepared by dissolving the solid chemicals in ultrapure water (Milli-Q, Academic, USA). E. coli K12 was used as the test organism. The 3 h growth inhibition tests were carried out according to the method used in our previous studies (Li et al., 2011; Lin et al., 2014) with a few modifications detailed in the Supporting Information (SI). The 3 h inhibition ratios were defined as the percentage proportion of colony count of bacteria grown with PFOS or PFOA to that of the control without PFOS or PFOA. The IC50 (50% inhibitory concentration) was determined using the linear interpolation method.

Fourier transform infrared (FTIR) subtraction spectroscopy (Nicolet 6700, Thermo Fisher Scientific, USA) was used to analyze the potential changes in chemical composition of the bacterial cells according to the recorded method (Kardas et al., 2014). The exposed bacterial cells were washed by the physiological saline solution and lyophilized to remove water. FTIR spectra of cells with and without treatments were measured by the direct absorbance using the KBr pellet technique and the scanning was done within the range of 4000e400 cm
1 at a 4 cm
1 resolution at room temperature. The results were analyzed by the FTIR subtraction method using OMNIC software. 2.7. Determination of oxidative stress

2.2. Bacterial viability assays The Live/Dead Baclight Bacterial Viability Kit (Invitrogen/Molecular Probes, USA) was used to distinguish live and dead cells, which utilized mixtures of green-fluorescent nucleic acid stain (SYTO 9) and red-fluorescent nucleic acid stain (propidium iodide,

The fluorescence probe 5-(and-6)-chloromethyl-20 ,70 -dichlorodihydrofluorescein diacetate (CM-H2DCFDA) (Invitrogen/Molecular Probes, USA) was used to quantify the generation of intracellular ROS, as described in our previous paper (Zhang et al., 2015) with some modifications detailed in the SI. To measure the potential

808

G. Liu et al. / Environmental Pollution 214 (2016) 806e815

contribution of oxidative damage to the bacterial toxicity, N-acetyl cysteine (NAC) (Sigma-Aldrich Co., USA) was used as a ROS scavenger (Zhang et al., 2015) in an additional toxicity experiment. 0.05 mM NAC were added into the bacterial suspension prior to the exposure to PFOS or PFOA for 3 h, and viable cell number was measured and compared with that without NAC. The activity of superoxide dismutase (SOD) was determined using a SOD assay kit (A001-1, Nanjing JianCheng Bioengineering Institute, China) according to the provided instructions and procedures (also detailed in the SI). The malondialdehyde (MDA) content was measured according to the trichloroacetic acid method and calculated using the Hodges' equations (Long et al., 2012), as described in the SI. 2.8. Gene expression analysis Bacterial cells were exposed to PFOS (400 mg L
1) or PFOA (10 mg L
1) for 3 h and were then washed twice with the physiological saline solution. The total RNA was isolated from the washed cells using the Trizol reagent (Sangon, China) according to the manufacturer's instructions. The concentration of RNA was estimated by the spectrophotometry (the ratio of absorbances at 260 nm to 280 nm). The synthesis of first-strand cDNA was performed by reverse transcription of the total RNA using the AMV First Strand cDNA Synthesis Kit (Sangon, China). Eighteen target genes were selected, with the gene description and primer pairs tabulated in Table 1. A real-time PCR system (StepOne plus, ABI, USA) was conducted to measure the levels of mRNA for the genes.

The real-time PCR reactions were set up with the SybrGreen qPCR Master Mix (BBI, America) and 16sRNA was used as the internal reference gene. Relative mRNA levels were calculated using the 2
DDCT method and normalized to the internal reference gene (Cui et al., 2015).

2.9. Uptake of PFOS or PFOA by the bacteria The bacterial cells were exposed to 0.1 and 1 mg L
1 PFOS or PFOA solutions for 3 h, and the solutions without bacteria were taken as control. The PFC concentrations in the supernatants after the centrifugation (8000g) for 10 min were analyzed by a high performance liquid chromatogram system coupled with a triple quadrupole mass spectrometer (Agilent, USA) according to the recorded method (Giesy and Kannan, 2001) with minor adjustments noted in the SI. The bioconcentration factor (BCF, mL g
1) was calculated as the ratio of accumulated PFCs in bacteria in dry weight (mg g
1) to the equilibrium concentration of PFCs in the solution (mg mL
1).

2.10. Statistical analysis Each treatment including the control was conducted in triplicate and the results were presented as mean ± SD (standard deviation). The differences between the control and treatments were analyzed by ANOVA of SPSS 20.0, with p < 0.05 defined as significant.

Table 1 Gene descriptions and primer sequences used for real-time PCR assays. Target gene

Gene ID

Description

Primer sequences (50 -30 )

Product length (bp)

ftsZ

944786

GTP-binding tubulin-like cell division protein

142

ftsW

946322

putative lipid II flippase; integral membrane protein; FtsZ ring stabilizer

csgG

945619

curli production assembly/transport outer membrane lipoprotein

oppA

945830

oligopeptide ABC transporter periplasmic binding protein

ompA

945571

outer membrane protein A

ompF

945554

outer membrane porin 1a

plsB

948541

glycerol-3-phosphate O-acyltransferase

lpxT

946693

fadR

948652

lipid A 1-diphosphate synthase; undecaprenyl pyrophosphate:lipid A 1-phosphate phosphotransferase fatty acid metabolism regulon transcriptional regulator

fadL

946820

long-chain fatty acid outer membrane transporter

sodA

948403

superoxide dismutase, Mn

soxS

948567

superoxide response regulon transcriptional activator; autoregulator

yggE

947398

oxidative stress defense protein

dinF

948554

ahpF

947540

oxidative stress resistance protein; putative MATE family efflux pump; UV and mitomycin C inducible protein alkyl hydroperoxide reductase, F52a subunit, FAD/NAD(P)-binding

katG

948431

catalase-peroxidase HPI, heme b-containing

uvrA

948559

recA

947170

ATPase and DNA damage recognition protein of nucleotide excision repair excinuclease UvrABC DNA recombination and repair protein; ssDNA-dependent ATPase; synaptase; ssDNA and dsDNA binding protein; ATP-dependent homologous DNA strand exchanger; recombinase A; LexA autocleavage cofactor

F:GTCTTTATTGCTGCGGGTATG R:ATGCCATACGCTTCTTGCC F:TTACTGTGGCTGACCTTCGG R:GATAGACACCATCACGCTTCG F:CACTGAAAGATTCTCGCTGGTT R:CATGATATTTGCCGCCGTTA F:CCTGTTTGAAGGCTTACTGGTC R:CAGATAACTGGCATACGGAGAA F:CAAAGCAACCCTGAAACCG R:ACCCAGAACAACTACGGAACC F:CGTTGCTACCTATCGTAACTCCA R:CCAAAGCCTTCGTATTCGTAG F:TTGCCAAAGCGGTAGAAGAT R:AAGCCCAGAATACGGTCAGTC F:GCCAGCCCAACATTGACTTT R:TATCTCGTGAGGCATCTTTCGT F:GCTGGGCTTCTACCACAAACT R:TGCCAAATCTCGCCACTCT F:GCGGTGCGTATCGCTTAAA R:CAATGCTTGCCCTTGCTGA F:TGCCAGCAGACAAGAAAACC R:ATCAACGGAGCCGAAGTCA F:CGATTACATTCGCCAACGC R:CGAGACATAACCCAGGTCCATT F:AGTTCAAAGTTATCGCCCTGG R:AATCGCAAGAGTGGCAATGT F:GAAACTTCCGTCGCTTGCT R:TGAGTAGCGTCATCAGAACCG F:GTGTTCGTAAACGGGAAAGAGT R:CGATATCAACGGTATCGAGGAT F:ATATCCATAACACCACAGCCACT R:TTGCGGTAGTCAAAGTCCTCA F:TGAATCCCTTTCCGCCTAC R:TGATTGTCCCCACCGTAGAA F:GTGAAAGAGGGCGAAAACG R:CCGATCTTCTCACCTTTGTAGC

The gene descriptions were referred to National Center for Biotechnology Information of USA (NCBI).

124 155 214 100 153 142 82 105 161 143 100 143 146 246 175 143 206

G. Liu et al. / Environmental Pollution 214 (2016) 806e815

809

Fig. 1A shows the dose-dependent bacterial growth inhibitions of PFOS and PFOA. The determined 3 h IC50 of PFOS and PFOA were 374 ± 3 and 10.6 ± 1.0 mg L
1, respectively. The typical FM image and flow cytometric graph for the bacterial cells treated with the highest concentration of PFOS are present in Fig. 1B. The other FM images and flow cytometric graphs for the PFOS-treated cells at the lower concentrations and PFOA-treated cells are shown in Figs. S1 and S2 in the SI, respectively. It can be seen from the FM images that the red cells apparently increased with increasing doses of PFOS or PFOA, indicating the exposure to PFOS or PFOA dosedependently damaged cell membranes. The percentage of membrane-damaged cells (mortality) was quantified, and increasing mortality was exhibited as the concentration of PFOS or PFOA gradually increased (Fig. 1C), in accordance with the FM observations. From the comparison between Fig. 1A and C, it can be told that the growth inhibition ratio was much greater than the mortality at the high exposure concentrations of PFOS or PFOA, which will be discussed later.

coli are present in Fig. 3. The untreated bacterial cells were negatively charged, with the measured zeta potential being
51.3 ± 1.4 mV. The electronegativity slightly increased at the low doses (<25 mg L
1) of PFOS and gradually and significantly decreased to
38.7 ± 0.7 mV with the concentration further increased to 500 mg L
1 (Fig. 3A). The similar variation trend in surface electronegativity, i.e. increasing at the low doses (<8 mg L
1) and decreasing at the higher doses (8e15 mg L
1), was observed for the PFOA-treated bacterial cells (Fig. 3B). Similar to the variation trend in the cell surface electronegativity, the CSH increased at the low doses of PFOS (<25 mg L
1) or PFOA (<8 mg L
1) and then decreased with the further increase in the exposure concentrations of PFOS (25e500 mg L
1) or PFOA (8e15 mg L
1). The rDPH of bacterial cells exposed to different concentrations of PFOS or PFOA for 3 h is also shown in Fig. 3. The rDPH decreased slightly but significantly from 0.30 to 0.28 (p < 0.05) with the concentration of PFOS increased from 0 to 500 mg L
1. With increasing dose of PFOA, the rDPH also displayed a slight decrease, though the variation was not significant in statistics. The decrease in rDPH indicated the exposure to PFOS or PFOA increased the membrane fluidity of E. coli.

3.2. TEM observations

3.4. Influences on the membrane fatty acid composition

Typical TEM images of the cells untreated and treated with the highest concentration of PFOS or PFOA are shown in Fig. 2. The untreated bacterial cells (Fig. 2A1, 2B1, and 2C1) exhibited normal morphology and intact structure, while obvious disruptions of the cell membranes exposed to 500 mg L
1 PFOS were noticed from the hanging-drop TEM images with the cytoplasm outflow (Fig. 2A2) and the vertically and horizontally sliced bacterial cells (Fig. 2B2 and 2C2, respectively). Invisible cell wall, wrinkled cell membrane, plasmolysis, and cytoplasm outflow were also observed for the bacterial cells exposed to 15 mg L
1 PFOA (Fig. 2A3, 2B3, and 2C3). The lower concentrations of PFOS and PFOA also disrupted cell morphology and structure to a less extent, as shown in Fig. S3.

The fatty acid compositions of E. coli treated with PFOS or PFOA are tabulated in Table S1, with the calculated ratios of saturated fatty acids (SFA) to unsaturated fatty acids (UFA) shown in Fig. 3C. The bacterial cells exhibited a decreased SFA/UFA ratio at the exposed concentrations of PFOS (50 and 500 mg L
1). No significant change in the composition of fatty acids was observed at the low dose of PFOA (2 mg L
1), while the SFA/UFA ratio significantly decreased for the cells exposed to 15 mg L
1 PFOA. The decrease in the SFA/UFA ratio was mainly caused by the decrease of C16:0 SFA and the increase of C18:1 UFA, as shown in Table S1. The numbers preceding the colon in the names of fatty acids indicate the total number of carbon atoms, and the numbers following the colon present the number of double bond.

3. Results 3.1. Bacterial toxicity assessments

3.3. Influences on the surface properties of bacteria 3.5. FTIR spectrum analysis Variations of the surface electronegativity, surface hydrophobicity, and membrane fluidity of the treated bacterial cells with the exposure doses were determined to analyze the influences of PFOS or PFOA on the surface properties of the bacteria. Changes in the electronegativity as indicated by zeta potential of the exposed E.

The FTIR subtraction spectra between 1800 and 1000 cm
1 for the bacteria exposed to PFOS (400 mg L
1) or PFOA (10 mg L
1) are present in Fig. 4. There was no obvious difference between the two subtraction spectra. The PFCs-exposed cells showed an increase in

Fig. 1. Dose-dependent effects of PFOS or PFOA on the (A) growth inhibition and (C) mortality of E. coli; (B) typical flow cytometric analysis graph and fluorescence microscopic (FM) image (the insert) of the bacterial cells after the exposure to 500 mg L
1 PFOS for 3 h. More FM images and flow cytometric graphs for the PFOS-treated cells at the lower concentrations and PFOA-treated cells are shown in Figs. S1 and S2 in the SI, respectively. The dots in R1, R2, and R3 areas in the flow cytometric graph represent the amounts of dead, live, and uncharacterized cells, respectively. Error bars represent standard deviations (n ¼ 3).

810

G. Liu et al. / Environmental Pollution 214 (2016) 806e815

Fig. 2. Typical TEM images of hanging-drop (A), vertically sliced (B) and horizontally sliced (C) bacterial cells untreated (A1, B1, and C1) or treated with 500 mg L
1 PFOS (A2, B2, and C2) or 15 mg L
1 PFOA (A3, B3, and C3). The red, blue, and black arrows point to winked cell membrane, plasmolysis, and cytoplasm outflow, respectively. More TEM images of the bacterial cells treated with PFCs at lower concentrations are shown in Fig. S3. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

absorbance compared with the control cells in the C]C IR region (1680e1500 cm
1) (Larkin, 2011), which suggests the exposure to PFOS or PFOA could produce more unsaturated products. This result was consistent with the decrease in SFA/UFA after the exposure as mentioned in the above section. Besides, the peaks at ~1450 cm
1, ~1312 cm
1, and ~1118 cm
1 were also observable in the subtraction spectra, which could be assigned to lipid (Cakmak et al., 2006), protein (Romih et al., 2015), and RNA (Romih et al., 2015), respectively. The presences of these three peaks suggested that the exposure to PFOS or PFOA induced generations of lipid, protein, and RNA in the cells. 3.6. Influences on the oxidative stress The abilities of PFOS and PFOA to induce oxidative stress in the bacterial cells were evaluated by the measured intracellular ROS, SOD, and MDA (Fig. 5). The intracellular ROS increased with increasing concentration of PFOS or PFOA, and significantly (p < 0.05) higher ROS contents compared with that in the untreated cells were observed at the high doses of PFOS (400e500 mg L
1) or PFOA (6e15 mg L
1), as shown in Fig. 5A and B, respectively. Fig. 5 shows the exposure to PFOS or PFOA also increased the SOD and MDA contents in a dose-dependent way. Compared with that in the untreated cells, significantly higher SOD activity was induced in the bacterial cells exposed to 50e500 mg L
1 PFOS or to 2e15 mg L
1 PFOA. The significantly higher contents of MDA were detected at 50e500 mg L
1 PFOS or 10e15 mg L
1 PFOA. The up-regulation of SOD and MDA levels were largely

consistent with the increase in ROS induced by the exposure to PFOS or PFOA, which demonstrated that a high level of oxidative stress occurred in the presence of PFOS or PFOA. The ROS scavenger NAC was further used to assess the potential contribution of elevated oxidative stress to the observed bacterial toxicity. As illustrated in Fig. 5C, the NAC treatment lowered the growth inhibitions of PFOS or PFOA to the bacteria, and the mitigation effect became significant at the high doses of PFOS (400 and 500 mg L
1) or PFOA (15 mg L
1). 3.7. Expression of genes To address the toxicity at the gene level, the mRNA levels of 18 genes involved in cell division, membrane instability, oxidative stress, and DNA damage were determined after the exposure to PFOS (400 mg L
1) or PFOA (10 mg L
1) (Fig. 6). PFOS and PFOA had a largely similar effect on the gene expressions. The expression levels of ftsZ and ftsW after the exposure all decreased, suggesting the presence of PFOS or PFOA inhibited the cell division of E. coli. The assembly of curli could also be hampered because its regulating gene csgG was markedly down expressed. The exposure to PFOS or PFOA affected the expressions of genes controlling the membrane stability. The genes ompA and ompF were down expressed though the expression of oppA was not affected, suggesting PFOS and PFOA inhibited the synthesis of membrane proteins. The synthesis of phospholipid might be slightly affected as indicated by the slight variation in the expression of plsB and the unchanged expression of lpxT. The gene fadR

G. Liu et al. / Environmental Pollution 214 (2016) 806e815

0.33 0.32

50

0.30 0.29 0.28 0.27

40 30 20 10

-10

zeta potential (mV)

hydrophobicity (%)

0.31

rDPH

0

60

ab a

ab

abc cd

A ef

de bcdabc

f

f

f

-20

0

-50

0.25

-10

-60

g

a ab

-40

0.26

a

a

a

-30

b

c

cd dd dd cd

0

10

b

cd

20

g b

b

a

a a a hydrophobicity rDPH zeta potential

100 200 300 400 500

PFOS concentration (mg L-1) 60

-30

0.32

50

-35

hydrophobicity (%)

0.31

rDPH

0.30 0.29 0.28

40 30 20

0.27

10

0.26

0

zeta potential (mV)

0.33

a

-45

b

a

a

b

c

0

a

a

c

-55

-2

a

a

a

-50

a

a

a

a

-40

-60

B

a

2

4

c c

c

6

8

a

a

hydrophobicity rDPH zeta potential

10 12 14 16

PFOA concentration (mg L-1)

SFA/UFA ratio

3

a

b

b

a

C b

2

1

0

control

PFOS 50

PFOS 500

PFOA 2

PFOA 15

concentration (mg L-1) Fig. 3. Changes of zeta potential, hydrophobicity, and membrane fluidity of the bacterial cells exposed to various concentrations of PFOS (A) and PFOA (B), and the SFA/ UFA ratios of the cells after the exposures to different concentrations of PFOS or PFOA (C). Values with different letters (aeg) differ significantly (p < 0.05). Error bars represent standard deviations (n ¼ 3).

was apparently up-regulated though insignificantly, and fadL was significantly up expressed by PFOS though not by PFOA. The over expressions of fadR and fadL suggested that the exposures could activate the synthesis of UFA, which might thus cause the decrease of SFA/UFA ratio as revealed in the Section 3.4. The exposure to PFOS or PFOA damaged the antioxidant system

811

in the bacterial cells, as indicated by the remarkable up-expressions of genes involved in superoxide response (sodA and soxS), oxidative stress resistance (yggE and dinF), and ROS scavenger (ahpF and katG). The genes involved in the recognition of DNA damage (uvrA) and repair (recA) were also remarkably over expressed. 3.8. Uptake of PFOS or PFOA The uptakes of PFOS and PFOA by bacteria are shown in Table 2. With the concentration increased from 0.1 to 1 mg L
1, the uptakes of PFOS and PFOA increased from 7.50 ± 0.48 to 40.7 ± 0.9 mg g
1 and from 0.667 ± 0.291 to 6.87 ± 2.71 mg g
1, respectively. The BCFs of PFOS and PFOA were low (11.8e128 mL g
1), indicating the weak bioaccumulation to the bacteria. PFOS had 4.4 to 7.9 times higher BCFs than PFOA. Higher BCFs have been reported for PFOS than PFOA, e.g., it was reported that BCFs of PFOS and PFOA in fish, crab, and bivalve were 2691 and 11.2 mL g
1, 257 and 29.5 mL g
1, and 77.6 and 44.7 mL g
1, respectively (Naile et al., 2013). 4. Discussions 4.1. Cell membrane damage of the bacteria Cell membrane is an important barrier between cell and the outer environment, sustaining the cell basic metabolism and protecting the cell from external disturbance. E. coli cell possesses an outer membrane, which is mainly made up of porins and lipopolysaccharide molecules; in addition, there is a peptidoglycan layer between the outer membrane and the cytoplasmic membrane. Hence, it is necessary to address the effects of PFOS and PFOA on the cell membrane prior to the discussion on the cell inactivation and/or death. As described in the Section 3.3, the exposure to PFOS or PFOA increased cell surface hydrophobicity at the low doses but decreased the hydrophobicity at the higher concentrations. It is supposed that when the bacterial cells exposed to PFOS or PFOA of low concentrations, porins and lipopolysaccharide would be produced to reduce the damage of cell. Thus, the increasing hydrophobicity could be owing to the release of lipopolysaccharide. At the high doses, a fraction of PFOS/PFOA was supposed to accumulate on the cell surfaces and occupy the hydrophobic sites, with the hydrophilic parts exposed to the aqueous phase and thereby lowering the hydrophobicity of cell surfaces. The combination of hydrophobic parts of PFOS or PFOA with the lipid bilayer could also cause the swelling of cell membrane, along with the increase in membrane fluidity. Similar cell membrane fluidity increases caused by PFOS (Xie et al., 2010) and other organic compounds, such as pinene and tetralin (Sikkema et al., 1994), have been reported. The increase in membrane fluidity could destroy the structure of bacterial cells at the high doses, as indicated by the disintegration and plasmolysis of bacterial cells in the TEM images (Fig. 2). Thus, the cellular inclusions which have electric neutrality would be exposed to the outer environment, which could result in the decreasing trend of electronegativity. Hence, PFOS or PFOA of high concentrations could bond to the lipid in the membrane, disrupt the membrane structure, and give rise to the cell inactivation and/or death. Li and Zhu (2012) demonstrated that surfactants could enhance lipopolysaccharide release at low concentration and induce bacterial cell death at high concentration, which agreed with our speculation. 4.2. Growth inhibition and mortality of the bacteria The bacterial toxicity was assessed with two methods. The bacterial growth inhibition is a definition different from the

812

G. Liu et al. / Environmental Pollution 214 (2016) 806e815

-C=C1612

lipid 1450

1556

RNA 1118

protein 1312

PFOS PFOA

1800

1600

1400

1200

1000

wavelength (cm-1)

A a

0.6

ab

b

a

b bc

0.3 c e

0.0

c

e

d

bc

d

cd

0

100

b

bc

bc

b

bc

200

a

a

ROS SOD MDA

300

400

B

0.9

b

0.6

b

d

0.3

d

cd

-2

d

d

e

0.0

500

cd

a

c

b

cd

bc

b

a

b cd

d

0

c

2

4

6

8

ROS SOD MDA

10

12

14

80 d

60

b

c

c

b

b

40 a

20 a

a a

0

16

PFOA PFOA PFOA 10 15 2

PFOS PFOS PFOS 400 500 50

PFOA concentration (mg L-1)

PFOS concentration (mg L-1)

without NAC with NAC d c

C

100 a

growth inhibition (%)

0.9

content of ROS,SOD,MDA

content of ROS,SOD,MDA

Fig. 4. FTIR subtraction spectra of bacterial cells treated with 400 mg L
1 PFOS and 10 mg L
1 PFOA.

concentration (mg L-1)

Fig. 5. Intracellular ROS (ROS signal per 106 cells), SOD (U per 109 cells), and MDA (nMol per 109 cells) contents of bacterial cells induced by PFOS (A) and PFOA (B), and the bacterial growth inhibition of PFOS and PFOA with and without NAC (C). Values with different letters (aee) differ significantly (p < 0.05). Error bars represent standard deviations (n ¼ 3).

2.5

control

a

relative of leves of mRNA

PFOS 2.0

PFOA

1.5

a 1.0

a

aaa a

a

a a

bb

ftsZ

ftsW

a

aa

bb

a bac aaa a

a a b

a

a

aa

a

aa a

a

a

b

a

a

b

b b b

b

a

b

b

c

b

c

fadL

soxS

yggE

dinF

ahpF

katG

uvrA

recA

bb

0.5

0.0 csgG

oppA ompA ompF

plsB

lpxT

fadR

sodA

Fig. 6. Relative mRNA levels of 18 genes in E. coli exposed to PFOS (400 mg L
1) or PFOA (10 mg L
1) for 3 h mRNA levels of genes were normalized to 16sRNA. Values with different letters (aec) differ significantly (p < 0.05). Error bars represent standard deviations (n ¼ 3).

Table 2 Bioaccumulations of PFOS or PFOA to Escherichia coli after the exposure for 3 h. PFCs

Initial concentration (mg L
1)

Equilibrium concentration (mg L
1)

PFOS

0.1 1 0.1 1

0.058 0.781 0.041 0.583

PFOA

± ± ± ±

0.002 0.012 0.001 0.008

mortality, yet both of them are the fundamental experiments to evaluate the toxicity. The growth inhibition ratio which was

Bioaccumulated concentration (mg g
1) 7.50 40.7 0.667 6.87

± ± ± ±

0.48 0.9 0.291 2.71

BCF (mL g
1) 128 52.1 16.2 11.8

± ± ± ±

8 1.2 7.0 4.7

measured through the plate counting referred to the inactivation of bacteria that stopped reproducing and/or dead, while the mortality

G. Liu et al. / Environmental Pollution 214 (2016) 806e815

measured by using the Live/Dead Baclight Bacterial Viability Kit reflected the induced cell death with membrane damaged. Concentration-dependent growth inhibition and/or mortality of various organisms by PFOS or PFOA have been well reported (Yang et al., 2014; Zhang et al., 2013a; Mhadhbi et al., 2012). The recorded IC50 or EC50 (the concentration caused 50% of the maximum effect) of PFOS and PFOA for Isochrysis galbana, Brachionus calyciflorus, and Daphnia magna were 37.5 and 163.6 mg L
1 (Mhadhbi et al., 2012), 61.8 and 150.0 mg L
1 (Zhang et al., 2013a), and 78.09 and 201.85 mg L
1 (Yang et al., 2014), respectively. It was shown that PFOS was more toxic than PFOA. However, much higher toxicity was observed for PFOA than PFOS in our study, which could be owing to the different test organisms used. Higher toxicity of PFOA than PFOS to E. coli was also observed in a study through measuring the effect on gene expressions (Nobels et al., 2010). It can be seen from Fig. 1 that mortalities were lower than growth inhibition ratios at the high doses of PFOS or PFOA. This suggests that there were some cells that lose the ability of multiplication but still remain the intact membrane structure. The exposure-induced genetic damage and/or oxidative stress could cause the growth inhibition of bacterial cells without breaking cell membranes, as discussed below. 4.3. Genetic damage of the bacteria The exposure to PFOS or PFOA substantially affected gene expressions of the bacteria (Fig. 6). The genetic damage further evidenced by the up-expressions of uvrA and recA could regulate the cell growth inhibition, membrane instability, and oxidative stress. The down-expressions of ftsZ, ftsW, and csgG indicate that the exposure to PFOS or PFOA could restrain the cell division and curli fiber assembly, and thus inhibited the growth and multiplication of bacterial cells. The curli fiber is taken as a protective structure that can adhere and immobilize host protein to adopt biofilm lifestyles (Taylor et al., 2011). In view of the expressions of ompA, ompF, fadL, and fadR, we could infer that the synthesis of membrane proteins and fatty acids were influenced after the treatment with PFOS or PFOA, resulting in membrane damage. The up-regulations of oxidative-stress-related genes indicate that the antioxidant system of cells was destroyed, which gave rise to more ROS and SOD productions and contributed to the cell inactivation and/or death. Similarly, the up-regulation of katG was observed in E. coli exposed to PFOS or PFOA (Nobels et al., 2010) and sod1 gene was significantly up expressed in zebrafish by the treatment of PFOS (Kim et al., 2011), which demonstrated the damage of antioxidant system. Some studies investigated the expressions of Bcl-2, Bax, and Caspase-3 to illustrate toxicities of PFOS and/or PFOA to zebrafish and A549 cells by inducing apoptosis (Cui et al., 2015; Mao et al., 2012). 4.4. The effects of oxidative stress on the bacteria ROS are known to cause the modification of cellular macromolecules, even cell apoptosis through the interaction with proteins and have been proposed to play a critical role in cell death. SOD is an important antioxidant enzyme which is the first position of defense against oxidation. The exposure to PFOS or PFOA damaged the antioxidative system and increased the oxidative stress on the bacteria, as revealed by the dose-dependent rising of intracellular ROS, MDA, and SOD contents (Fig. 5). Similar concentration-dependent increases in ROS and SOD contents have been reported for cerebellar granule cells and earthworms exposed to PFOS and/or PFOA (Reistad et al., 2013; Xu et al., 2013). The accumulated ROS could induce the membrane lipid peroxidation. Lipid peroxidation as a result of the oxidative stress in the

813

presence of PFOS has been described elsewhere (Xu et al., 2013). The increase in MDA content in the exposed cells indicated an increased lipid peroxidation which was indicative of the damage of cell membrane integrity. The enhancement of oxidative stress could also affect the protein expression and induce the oxidative DNA damage, triggering the inactivation and/or death of bacterial cells. The fact that the bacteria growth inhibition was mitigated by the ROS scavenger NAC further evidenced the important role of induced oxidative stress in the antibacterial effect of PFOS or PFOA. However, the presence of NAC could only alleviate part of the toxicity as shown in Fig. 5C, which suggests other toxicity mechanisms besides the oxidative damage contributed to the cell inactivation and/or death. 4.5. The effect of interaction with biological molecules on the bacteria The cell-accumulated PFOS or PFOA could interact with biological macromolecules, such as phospholipid, protein, and DNA/ RNA, triggering or contributing to the toxicity to the bacteria. The changes in SFA/UFA ratio demonstrated that PFOS or PFOA changed the composition of fatty acids possibly through the interaction with membrane phospholipid molecules. The decrease in SFA/UFA ratio of cell membranes was also observed for Stenotrophomonas maltophilia exposed to diesel oil or Triton X-100 (Kaczorek et al., 2013). Because of the linear chain of SFA and bending chain of UFA, the decrease in SFA/UFA ratio indicated an increase in the membrane fluidity. The membrane fluidity increase was evidenced by the decrease in rDPH of bacteria (Fig. 3). The lipophilic fraction of PFOS or PFOA could combine with the membrane phospholipids, interfere with the assembly process of cell membrane, and disrupt the membrane. The FTIR analysis revealed the exposure to PFOS or PFOA altered contents and/or compositions of lipid, protein, and RNA in the bacterial cells. The emerged IR peaks of protein in the presence of PFOS or PFOA could originate from the combination of PFCs with other biological molecules, such as membrane phospholipids. It is known that the membrane is mainly composed of phospholipids and proteins. When phospholipids combined with PFOS or PFOA, more proteins would be exposed, thus, resulting in the enhancement of IR signal of protein (Liao et al., 2010). The slight increase in the content of cellular RNA as indicated by the FTIR analysis suggested that PFOS and PFOA could have access to cellular inclusion and RNA, resulting in the genetic damage. To date, a few studies have investigated interactions between PFOS/PFOA and biological macromolecules by molecular docking, such as human estrogen receptor (Gao et al., 2013), bovine serum albumin (Chen et al., 2015a), and liver fatty acid binding proteins (Ng and Hungerbuehler, 2015). It was demonstrated that the structure of protein could be changed in the presence of PFCs and the bind sites for PFOS were different from those for PFOA (Chen et al., 2015a). It was also reported that PFOS could combine with and accumulate in the lipid bilayer of membrane (Liao et al., 2010). However, very few studies have specifically investigated the process how PFOS or PFOA combining with varieties of macromolecules in cells. More specific researches on the interaction of PFCs with biological macromolecules are warranted. 4.6. Toxicity mechanisms of PFOS and PFOA Based on the above result analyses and discussions, toxicity mechanisms of PFOS and PFOA to the bacteria are proposed in Fig. 7. The PFCs accumulated in the bacterial cells and interacted with biomolecules, which induced the oxidative stress, DNA damage, and membrane disruption. The elevated oxidative stress could increase the DNA damage and membrane disruption. The membrane

814

G. Liu et al. / Environmental Pollution 214 (2016) 806e815

PFOS/PFOA

bioaccumulation and interaction with biomolecules

damage induced cell inactivation and/or death. But, the dominating toxicity mechanisms of PFOS and PFOA could be different and related to the membrane disruption and the oxidative stress, respectively. The results obtained in this study improved our understanding of the ecological effects of PFCs. Acknowledgement

oxidative stress

DNA damage

This work was supported by the 973 program of China (2014CB441104) and the National Natural Science Foundation of China (21525728, 21337004, and 21477107).

membrane disruption

cell inactivation/death

Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.envpol.2016.04.089. References

Fig. 7. Schematic summary of toxicity mechanisms of PFOS/PFOA to bacterial cells. The bioaccumulated PFCs interacted with biomolecules and induced oxidative stress, DNA damage, and membrane disruption. The elevated oxidative stress increased the DNA damage and membrane disruption. The membrane disruption and breakage resulted in the cytoplasm outflow and cell death. The DNA damage interfered with the membrane synthesis and even triggered the cell apoptosis. The increase in oxidative stress together with the DNA damage and membrane disruption caused the cell growth inhibition and/or death.

disruption and breakage could result in the cytoplasm outflow and cell death. The DNA damage could interfere with the membrane synthesis and even trigger the cell apoptosis. The increase in oxidative stress together with the DNA damage and membrane disruption caused the cell inactivation and/or death. Though the toxicity mechanisms of PFOS and PFOA were largely similar, the difference could also be noticed. The higher BCF but much lower toxicity of PFOS than PFOA was indicative of the potential difference in the toxicity mechanism. It can be seen from Fig. 3 that the exposure to PFOS changed the cell surface properties more severe than PFOA. This suggests PFOS could be more likely to disrupt the cell membrane. The up-expressions of plsB and fadL by PFOS but not PFOA also indicated the more severe effect of PFOS than PFOA on cell membranes. The exposure to PFOA induced higher oxidative stress than PFOS, as indicated by the higher production of intracellular ROS (Fig. 5) and the significantly higher upexpression of katG (Fig. 6) by PFOA. The elevated oxidative stress by PFOA could induce the cell inactivation and/or death mainly through the genetic interruption pathway rather than the direct membrane lipid peroxidation as indicated by the lower production of MDA under the treatment of PFOA than PFOS (Fig. 5). The much higher MDA production by PFOS further suggested the more severe membrane disruption of PFOS than PFOA. The different toxicity mechanisms of PFOS and PFOA could be mainly ascribed to the difference of hydrophilic functional groups in the molecules, which might result in different binding affinity and sites when they interacted with biomacromolecules (Ng and Hungerbuehler, 2015; Gao et al., 2013). The different effect and its underlying molecular mechanism of carboxyl and sulfonyl functional groups on the toxicity of PFCs merit more specific studies. 5. Conclusions This study systematically investigated toxicities of PFOS and PFOA to E. coli and addressed the toxicity mechanisms. PFOS exhibited much lower toxicity but greater BCFs to the bacteria. The toxicity mechanisms of PFOS and PFOA were largely similar and ascribed to the membrane disruption, oxidative stress, and DNA

Boulanger, B., Vargo, J., Schnoor, J.L., Hornbuckle, K.C., 2004. Detection of perfluorooctane surfactants in Great Lakes Water. Environ. Sci. Technol. 38, 4064e4070. Cakmak, G., Togan, I., Severcan, F., 2006. 17b-Estradiol induced compositional, structural and functional changes in rainbow trout liver, revealed by FT-IR spectroscopy: a comparative study with nonylphenol. Aquat. Toxicol. 77, 53e63. Chen, H., He, P., Rao, H., Wang, F., Liu, H., Yao, J., 2015a. Systematic investigation of the toxic mechanism of PFOA and PFOS on bovine serum albumin by spectroscopic and molecular modeling. Chemosphere 129, 217e224. Chen, H., Yao, J., Wang, F., Cai, M., Liu, H., 2013. Toxicity of perfluorooctanoic acid to Pseudomonas putida in the aquatic environment. J. Hazard. Mater 262, 726e731. Chen, M., Qiang, L., Pan, X., Fang, S., Han, Y., Zhu, L., 2015b. In Vivo and in vitro isomer-Specific biotransformation of perfluorooctane sulfonamide in common carp (Cyprinus carpio). Environ. Sci. Technol. 49, 13817e13824. Cui, Y., Liu, W., Xie, W., Yu, W., Wang, C., Chen, H., 2015. Investigation of the effects of perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) on apoptosis and cell cycle in a zebrafish (Danio rerio) liver cell line. Int. J. Environ. Res. Public Health 12, 15673e15682. Gao, Y., Li, X., Guo, L., 2013. Assessment of estrogenic activity of perfluoroalkyl acids based on ligand-induced conformation state of human estrogen receptor. Environ. Sci. Technol. 47, 634e641. Giesy, J.P., Kannan, K., 2001. Global distribution of perfluorooctane sulfonate in wildlife. Environ. Sci. Technol. 35, 1339e1342.
ska-Bajorek, B., 2013. Cell surface properties Kaczorek, E., Sałek, K., Guzik, U., Dudzin and fatty acids composition of Stenotrophomonas maltophilia under the influence of hydrophobic compounds and surfactants. New Biotechnol. 30, 173e182. Kardas, M., Gozen, A.G., Severcan, F., 2014. FTIR spectroscopy offers hints towards widespread molecular changes in cobalt-acclimated freshwater bacteria. Aquat. Toxicol. 155, 15e23. Kim, S., Ji, K., Lee, S., Lee, J., Kim, J., Kim, S., Kho, Y., Choi, K., 2011. Perfluorooctane sulfonic acid exposure increases cadmium toxicity in early life stage of zebrafish, Danio rerio. Environ. Toxicol. Chem. 30, 870e877. Lam, N., Cho, C., Lee, J., Soh, H., Lee, B., Lee, J., Tatarozako, N., Sasaki, K., Saito, N., Iwabuchi, K., Kannan, K., Cho, H., 2014. Perfluorinated alkyl substances in water, sediment, plankton and fish from Korean rivers and lakes: a nationwide survey. Sci. Total Environ. 491e492, 154e162. Larkin, P., 2011. IR and Raman Spectroscopy. Elsevier. Liao, T.T., Shi, Y.L., Jia, J.W., Jia, R.W., Wang, L., 2010. Sensitivity of morphological change of Vero cells exposed to lipophilic compounds and its mechanism. J. Hazard. Mater 179, 1055e1064. Li, F., Zhu, L.Z., 2012. Effect of surfactant-induced cell surface modifications on electron transport system and catechol 1,2-dioxygenase activities and phenanthrene biodegradation by Citrobacter sp. SA01. Bioresour. Technol. 123, 42e48. Li, F., Zhu, L.Z., 2014. Surfactant-modified fatty acid composition of Citrobacter sp. SA01 and its effect on phenanthrene transmembrane transport. Chemosphere 107, 58e64. Li, M., Zhu, L.Z., Lin, D.H., 2011. Toxicity of ZnO Nanoparticles to Escherichia coli: mechanism and the influence of medium components. Environ. Sci. Technol. 45, 1977e1983. Lin, A.Y., Panchangam, S.C., Lo, C., 2009. The impact of semiconductor, electronics and optoelectronic industries on downstream perfluorinated chemical contamination in Taiwanese rivers. Environ. Pollut. 157, 1365e1372. Lin, X.C., Li, J.Y., Ma, S., Liu, G.S., Yang, K., Tong, M.P., Lin, D.H., 2014. Toxicity of TiO2 nanoparticles to Escherichia coli: effects of particle size, crystal phase and water chemistry. PLoS One 9, e110247. Liu, C., Chang, V.W.C., Gin, K.Y.H., Nguyen, V.T., 2014. Genotoxicity of perfluorinated chemicals (PFCs) to the green mussel (Perna viridis). Sci. Total Environ. 487, 117e122. Long, Z.F., Ji, J., Yang, K., Lin, D.H., Wu, F.C., 2012. Systematic and quantitative investigation of the mechanism of carbon nanotubes' toxicity toward algae.

G. Liu et al. / Environmental Pollution 214 (2016) 806e815 Environ. Sci. Technol. 46, 8458e8466. MacDonald, M.M., Warne, A.L., Stock, N.L., Mabury, S.A., Solomon, K.R., Sibley, P.K., 2004. Toxicity of perfluorooctane sulfonic acid and perfluorooctanoic acid to Chironomus tentans. Environ. Toxicol. Chem. 23, 2116e2123. Mao, Z., Xia, W., Wang, J., Chen, T., Zeng, Q., Xu, B., Li, W., Chen, X., Xu, S., 2012. Perfluorooctane sulfonate induces apoptosis in lung cancer A549 cells through reactive oxygen species-mediated mitochondrion-dependent pathway. J. Appl. Toxicol. 33, 1268e1276.
rez, S., Beiras, R., 2012. Ecological risk assessment of perMhadhbi, L., Rial, D., Pe fluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) in marine environment using Isochrysis galbana, Paracentrotus lividus, Siriella armata and Psetta maxima. J. Environ. Monit. 14, 1375. Naile, J.E., Khim, J.S., Hong, S., Park, J., Kwon, B., Ryu, J.S., Hwang, J.H., Jones, P.D., Giesy, J.P., 2013. Distributions and bioconcentration characteristics of perfluorinated compounds in environmental samples collected from the west coast of Korea. Chemosphere 90, 387e394. Ng, C.A., Hungerbuehler, K., 2015. Exploring the use of molecular docking to identify bioaccumulative perfluorinated alkyl acids (PFAAs). Environ. Sci. Technol. 49, 12306e12314. Nobels, I., Dardenne, F., Coen, W.D., Blust, R., 2010. Application of a multiple endpoint bacterial reporter assay to evaluate toxicological relevant endpoints of perfluorinated compounds with different functional groups and varying chain length. Toxicol. in Vitro 24, 1768e1774. Olsen, G.W., Burris, J.M., Ehresman, D.J., Froehlich, J.W., Seacat, A.M., Butenhoff, J.L., Zobel, L.R., 2007. Half-life of serum elimination of perfluorooctanesulfonate, perfluorohexanesulfonate, and perfluorooctanoate in retired fluorochemical production workers. Environ. Health Perspect. 115, 1298e1305. Pan, C., Ying, G., Liu, Y., Zhang, Q., Chen, Z., Peng, F., Huang, G., 2014. Contamination profiles of perfluoroalkyl substances in five typical rivers of the Pearl River Delta region, South China. Chemosphere 114, 16e25. Parsons, J.R., S
aez, M., Dolfing, J., de Voogt, P., 2008. Biodegradation of perfluorinated compounds. Rev. Environ. Contam. Toxicol. 196, 53e71. Prevedouros, K., Cousins, I.T., Buck, R.C., Korzeniowski, S.H., 2006. Sources, fate and transport of perfluorocarboxylates. Environ. Sci. Technol. 40, 32e44. Rahman, M.F., Peldszus, S., Anderson, W.B., 2014. Behaviour and fate of perfluoroalkyl and polyfluoroalkyl substances (PFASs) in drinking water treatment: a review. Water Res. 50, 318e340. Reistad, T., Fonnum, F., Mariussen, E., 2013. Perfluoroalkylated compounds induce cell death and formation of reactive oxygen species in cultured cerebellar granule cells. Toxicol. Lett. 218, 56e60.
s, F., Rosal, R., Ferna
ndez-Pin ~ as, F., 2012. Toxicological Rodea-Palomares, I., Legane interactions of perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA) with selected pollutants. J. Hazard. Mater 201e202, 209e218. Romih, T., Jemec, A., Novak, S., Vaccari, L., Ferraris, P., Simon, M., Kos, M., Susic, R., Kogej, K., Zupanc, J., Drobne, D., 2015. FTIR microscopy reveals distinct biomolecular profile of crustacean digestive glands upon subtoxic exposure to ZnO nanoparticles. Nanotoxicology 1e9. ~ as, F., Legane
s, F., Petre, A., Rosal, R., Rodea-Palomares, I., Boltes, K., Fern
andez-Pin 2010. Ecotoxicological assessment of surfactants in the aquatic environment:

815

combined toxicity of docusate sodium with chlorinated pollutants. Chemosphere 81, 288e293. Rosenberg, M., Rosenberg, E., Gutnick, D., 1980. Adherence of bacteria to hydrocarbons: a simple method for measuring cell-surface hydrophobicity. FEMS Microbiol. Lett. 9, 29e33. Route, W.T., Russell, R.E., Lindstrom, A.B., Strynar, M.J., Key, R.L., 2014. Spatial and temporal patterns in concentrations of perfluorinated compounds in bald eagle nestlings in the upper midwestern United States. Environ. Sci. Technol. 48, 6653e6660. Sanderson, H., Boudreau, T.M., Mabury, S.A., Solomon, K.R., 2004. Effects of perfluorooctane sulfonate and perfluorooctanoic acid on the zooplanktonic community. Ecotoxicol. Environ. Safe. 58, 68e76. Sasser, M., 1990. Identification of Bacteria by Gas Chromatography of Cellular Fatty Acids. MIDI Technical Note 101. MIDI Inc., Newark, DE. Schultz, M.M., Barofsky, D.F., Field, J.A., 2004. Quantitative determination of fluorotelomer sulfonates in groundwater by LC MS/MS. Environ. Sci. Technol. 38, 1828e1835. Shoeib, M., Harner, T., Webster G., M., Lee, S.C., 2011. Indoor sources of poly- and perfluorinated compounds (PFCS) in Vancouver, Canada: implications for human exposure. Environ. Sci. Technol. 45, 7999e8005. Sikkema, J., de Bont, J.A., Poolman, B., 1994. Interactions of cyclic hydrocarbons with biological membranes. J. Biol. Chem. 269, 8022e8028. Taylor, J.D., Zhou, Y., Salgado, P.S., Patwardhan, A., McGuffie, M., Pape, T., Grabe, G., Ashman, E., Constable, S.C., Simpson, P.J., Lee, W., Cota, E., Chapman, M.R., Matthews, S.J., 2011. Atomic resolution insights into curli fiber biogenesis. Structure 19, 1307e1316. Xia, X., Rabearisoa, A.H., Dai, Z., Jiang, X., Zhao, P., Wang, H., 2015. Inhibition effect of Naþ and Ca2þ on the bioaccumulation of perfluoroalkyl substances by Daphnia magna in the presence of protein. Environ. Toxicol. Chem. 34, 429e436. Xia, X., Rabearisoa, A.H., Jiang, X., Dai, Z., 2013. Bioaccumulation of perfluoroalkyl substances by Daphnia magna in water with different types and concentrations of protein. Environ. Sci. Technol. 47, 10955e10963. Xie, W., Ludewig, G., Wang, K., Lehmler, H., 2010. Model and cell membrane partitioning of perfluorooctanesulfonate is independent of the lipid chain length. Colloid Surf. B 76, 128e136. Xu, D., Li, C., Wen, Y., Liu, W., 2013. Antioxidant defense system responses and DNA damage of earthworms exposed to perfluorooctane sulfonate (PFOS). Environ. Pollut. 174, 121e127. Yang, S., Xu, F., Wu, F., Wang, S., Zheng, B., 2014. Development of PFOS and PFOA criteria for the protection of freshwater aquatic life in China. Sci. Total Environ. 470e471, 677e683. Zhang, L.Q., Lei, C., Chen, J.J., Yang, K., Zhu, L.Z., Lin, D.H., 2015. Effect of natural and synthetic surface coatings on the toxicity of multiwalled carbon nanotubes toward green algae. Carbon 83, 198e207. Zhang, L., Niu, J., Li, Y., Wang, Y., Sun, D., 2013a. Evaluating the sub-lethal toxicity of PFOS and PFOA using rotifer Brachionus calyciflorus. Environ. Pollut. 180, 34e40. Zhang, Y., Beesoon, S., Zhu, L., Martin, J.W., 2013b. Biomonitoring of perfluoroalkyl acids in human urine and estimates of biological half-Life. Environ. Sci. Technol. 47, 10619e10627.