Partitioning and Accumulation of Perfluoroalkyl Substances in Model

Subscriber access provided by Kaohsiung Medical University

Environmental Processes

Partitioning and Accumulation of Perfluoroalkyl Substances in Model Lipid Bilayers and Bacteria Nicole J.M. Fitzgerald, Andreas Wargenau, Carlise Sorenson, Joel A. Pedersen, Nathalie Tufenkji, Paige J Novak, and Matt Francis Simcik Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b02912 • Publication Date (Web): 27 Aug 2018 Downloaded from http://pubs.acs.org on August 28, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27

Environmental Science & Technology

1

Partitioning and Accumulation of Perfluoroalkyl Substances in Model Lipid Bilayers and

2

Bacteria

3

Nicole J.M. Fitzgerald1,2*, Andreas Wargenau3, Carlise Sorenson4, Joel Pedersen5, Nathalie

4

Tufenkji3, Paige J. Novak1, Matt F. Simcik6*

5 6

1

7

Pillsbury Drive SE, Minneapolis, Minnesota 55455

Department of Civil, Environmental, and Geo- Engineering, University of Minnesota, 500

8 9 10

2

Current Address: Department of Civil and Environmental Engineering, Colorado School of

Mines 1012 14th Street, Golden, CO 80401

11 12

3

13

Quebec H3A 0C5

Department of Chemical Engineering, McGill University, 3610 University Street, Montreal,

14 15

4

16

Ave, Saint Paul, MN 55108

Department of Bioproducts and Biosystems Engineering, University of Minnesota, 1390 Eckles

17 18

5

19

Wisconsin, 1525 Observatory Drive, Madison, WI 53706

Departments of Soil Science, Civil & Environmental Engineering, and Chemistry, University of

20 21

6

22

55455

School of Public Health, University of Minnesota, 420 Delaware St. S.E. Minneapolis, MN

23

ACS Paragon Plus Environment


24

*Co-corresponding authors to whom correspondence should be addressed: NJMF: phone: 847-

25

791-2844, email: [email protected]; MFS: phone 612-626-6269, email:

26

[email protected]


Page 2 of 27

Page 3 of 27


27

ABSTRACT: Perfluoroalkyl substances (PFAS) are ubiquitous and persistent environmental

28

contaminants, yet knowledge of their biological effects and mechanisms of action is limited. The

29

highest aqueous PFAS concentrations are found in areas where bacteria are relied upon for

30

functions such as nutrient cycling and contaminant degradation, including fire-training areas,

31

wastewater treatment plants, and landfill leachates. This research sought to elucidate one of the

32

mechanisms of action of PFAS by studying their uptake by bacteria and partitioning into model

33

phospholipid bilayer membranes. PFAS partitioned into bacteria as well as model membranes

34

(phospholipid liposomes and bilayers). The extent of incorporation into model membranes and

35

bacteria was positively correlated to the number of fluorinated carbons. Furthermore,

36

incorporation was greater for perfluorinated sulfonates than for perfluorinated carboxylates.

37

Changes in zeta potential were observed in liposomes but not bacteria, consistent with PFAS

38

being incorporated into the phospholipid bilayer membrane. Complementary to these results,

39

PFAS were also found to alter the gel-to-fluid phase transition temperature of phospholipid

40

bilayers, demonstrating that PFAS affected lateral phospholipid interactions. This investigation

41

compliments other studies showing that sulfonated PFAS and PFAS with more than seven

42

fluorinated carbons have a higher potential to accumulate within biota than carboxylated and

43

shorter chain PFAS.

44 45

INTRODUCTION

46

Perfluoroalkyl substances (PFAS) are ubiquitous in the environment and are associated with a

47

variety of negative effects in organisms.1–3 In humans, PFAS serum concentrations have been

48

correlated with thyroid disease4 and decreased fecundity5. In children, perfluorooctanoic acid

49

(PFOA) and perfluorooctane sulfonate (PFOS) serum concentrations were negatively correlated



50

with serum antibody concentrations for routine vaccinations, indicating a decrease in immune

51

response.6 PFAS have also been associated with an inhibition in gap junction communication in

52

rat and dolphin cell lines.7 The effects of PFAS on bacteria have received less attention, although

53

at concentrations of 22-110 mg/L they increased floc formation of Rhodococcus jostii8 and

54

inhibited microbial degradation of trichloroethylene by Dehalococcoides mccartyi

55

enrichments.9,10 At lower concentrations (100 µg/L), PFAS increased bacterial mobility in a

56

model aquifer system,11 and at concentrations as low as 10 µg/L, PFOS increased the sensitivity

57

of the quorum sensing response in Aliivibrio fischeri.12 PFAS clearly affect a variety of biota, yet

58

their mechanisms of action, and whether they are similar in bacteria and eukaryotes, remain

59

unclear.

60 61

PFAS are amphiphilic compounds, suggesting they may partition into cell membranes and alter

62

membrane properties.13–17 PFOA and PFOS have been shown to partition to phospholipid

63

bilayers where they alter disrupt the packing of the acyl chains of phospholipids and alter

64

membrane properties.18 A recent study showed that PFAS with different head groups and

65

perfluorinated chain lengths between 4 and 8 carbons partitioned into model membranes,

66

displacing phospholipids and altering bilayer properties even after the PFAS were removed from

67

the aqueous phase.19 Partitioning of PFOS, perfluorobutane sulfonate (PFBS), and PFOA into

68

phospholipid bilayers at mg/L to g/L concentrations increases bilayer fluidity.13,15,20,21 This

69

increase in fluidity is dose-dependent for PFOS from 10 to 200 mg/L.13 Little research has been

70

performed on the partitioning and subsequent biological effect of PFAS with six or fewer

71

fluorinated carbons, so-called short-chain PFAS; nevertheless, one study showed that PFBS did


Page 4 of 27

Page 5 of 27


72

partition into phospholipid membranes, increasing fluidity at concentrations of 450 to 3,600

73

mg/L.15

74 75

While the potential for PFAS to change membrane properties in model systems is apparent,

76

PFAS can also interact with proteins. PFAS bind to several sites on serum albumin22,23 and can

77

be transported into some eukaryotic cells by fatty acid-binding liver proteins5,24 and organic

78

anion transporters5,22. The affinity of PFAS to protein was evident when PFAS accumulation was

79

studied in Daphnia magna; the presence of albumin in solution decreased PFAS partitioning to

80

Daphnia magna, showing preferential PFAS binding to albumin.25 Similarly, a study modeling

81

PFAS bioaccumulation in different organisms found that protein and phospholipid partitioning

82

descriptors were both required to accurately describe bioaccumulation.22 The interactions of

83

PFAS and protein adds a level of uncertainty with respect to how PFAS will partition in bacteria,

84

particularly given the differences in bacterial and eukaryote cell structures and the variation

85

within proteins of different organisms.

86 87

Bacteria are critical for many ecological functions and may behave differently than higher

88

organisms with respect to PFAS accumulation. Bacteria are often exposed to elevated

89

concentrations of PFAS in engineered systems where we rely on their metabolic functions to

90

treat waste. Recently, industries have been switching to the use of more short-chain PFAS.26,27

91

The environmental behavior and subsequent biological impacts of these short-chain PFAS have

92

received little study and are poorly understood, particularly in bacteria. The role of the PFAS

93

head group with respect to partitioning tendencies is also unclear. Current health standards treat

94

PFOS and PFOA as equivalent, while experimental observations show that these two PFAS



95

partition differently and cause biological effects at different concentrations.6,12,16,20,24,28,29 To

96

address some of these knowledge gaps, particularly as they pertain to bacteria, we focused on the

97

partitioning of PFAS into model membranes (liposomes and supported phospholipid bilayers)

98

and bacteria (Gram-positive and -negative). This research also investigated some of the resulting

99

effects as a function of PFAS functional group and fluorinated chain length.

100 101

METHODS

102

Chemicals. The effect of perfluorinated chain length and functional group was examined by

103

performing experiments with three perfluorinated carboxylates and three perfluorinated

104

sulfonates of varying fluorinated chain lengths. Perfluorobutanoate (PFBA, CAS 375-22-4),

105

perfluorooctane (PFOA, CAS 335-67-1), perfluorononanoate (PFNA, CAS 375-95-1),

106

perfluorobutane sulfonate (PFBS, CAS 375-73-5), perfluorohexane sulfonate (PFHxS, CAS 355-

107

46-4), and perfluorooctane sulfonate (PFOS, CAS 1763-23-1), containing 3, 7, 8, 4, 6, and 8

108

perfluorinated carbons, respectively, were tested. Mass-labeled internal standards were

109

purchased from Wellington Laboratories. Additional details on the PFAS used are found in the

110

Supporting Information. The phospholipids 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine

111

(POPC) and 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC), both dissolved in

112

chloroform at 25 mg/mL, were purchased from Avanti Polar Lipids, Inc. and were > 99 % pure.

113

Lipid solutions were stored at –20 °C. Two different phospholipids were used in experiments

114

(described below) to satisfy the fluidity and phase transition requirements of each experiment.

115 116

Culture Preparation. In this research, both Gram-positive (Staphylococcus epidermidis, ATCC

117

12228) and Gram-negative bacteria (Aliivibrio fischeri mutant strain DC43 from Dr. Eric


Page 6 of 27

Page 7 of 27


118

Stabb30) were evaluated. Source liquid cultures, started from cultures grown on solid Luria-

119

Bertani-Salt (LBS) medium,31 were grown for 24 h in liquid LBS medium and used to inoculate

120

experimental cultures. Bacteria used in experiments were inoculated into LBS with 1% source

121

culture and grown while mixed for 48 and 18 h for the partitioning and zeta potential

122

experiments, respectively (Figure SI.1). Bacteria used in the partitioning experiments were

123

grown for a longer time frame to accumulate more biomass to facilitate measurement of PFAS in

124

the biomass. S. epidermis was grown at 37 °C, and A. fischeri was grown at room temperature

125

(21-22 °C).

126 127

Liposome Preparation. To prepare liposomes, 1 mL of POPC dissolved in chloroform was

128

transferred to an ashed serum vial and chloroform was evaporated overnight under a stream of

129

pre-purified nitrogen. POPC was re-suspended in 5 mL phosphate buffer (pH 7.4, 4.9 g/L NaCl,

130

1.19 g/L K2HPO4, 0.29 g/L KH2PO4) filtered with a pre-rinsed 0.22-µm membrane and then

131

extruded 15 times through a 100-nm polycarbonate membrane using a Mini Extruder (Avanti

132

Polar Lipids Inc.). After extrusion, liposomes were diluted in filtered phosphate buffer to a lipid

133

concentration of 1 mg/mL. Liposome size distribution was verified using a ZetaPALS Potential

134

Analyzer (Brookhaven Instruments).

135 136

DMPC liposomes for the formation of supported phospholipid bilayers (SPB) in the quartz

137

crystal microbalance with dissipation monitoring (QCM-D) were prepared in a similar manner

138

with DMPC, as described in Wargenau and Tufenkji.32

139 140

Experimental Design



141

Determination of PFAS Partitioning to Bacteria. Sorption of PFBA, PFOA, PFNA, PFBS,

142

PFHxS, and PFOS to both live and dead bacteria was measured to determine if active PFAS

143

uptake occurred. After 48 h of growth, bacteria were washed twice with phosphate buffer and

144

resuspended in 80% of the original volume of phosphate buffer (800 mL), and total bacteria, as

145

mass dry solids, was measured. Details regarding mass measurements can be found in the

146

Supporting Information. Aliquots (10 mL) of washed bacteria transferred into sterile

147

polypropylene centrifuge tubes. Half of the bacterial aliquots were killed via addition of sodium

148

azide to a final concentration of 50 mM. Each PFAS was added to the washed bacteria (live and

149

dead) at a target concentration of 750 µg/L via methanol stock. PFAS-free controls were

150

included to verify the lack of PFAS contamination; control aliquots received PFAS-free

151

methanol (less than 0.25% of the total volume) so that all treatments (PFAS-containing or PFAS-

152

free) contained equivalent amounts of methanol. All treatments were performed in triplicate.

153 154

During each experiment, centrifuge tubes were shaken on an orbital rotator at 18 rpm for 20 h,

155

after which the solid and liquid fractions were separated by centrifugation at 2,500 rcf for 15

156

min. A 0.5-mL sample was taken of the supernatant, and the rest of the supernatant was

157

discarded. The supernatant and remaining solids sample were analyzed for PFAS as described

158

below. The samples containing live PFOS-exposed Gram-positive bacteria were spilled and lost

159

immediately prior to PFAS analysis.

160 161

PFAS Analytical Methods. The liquid media collected during the partitioning experiments (0.5

162

mL) were diluted with a 0.5-mL addition of methanol (Optima Grade, Fisher Scientific) for


Page 8 of 27

Page 9 of 27


163

analysis. Following dilution, a known quantity of the mass-labeled standard corresponding to the

164

analyte of interest was added to each sample.

165 166

The solid fractions were extracted three times with methanol, after which all extracts were

167

pooled and a mass-labeled internal standard was added. The pooled extracts were concentrated

168

under a stream of nitrogen and analyzed via a Hewlett Packard 1100 HPLC connected to a

169

Hewlett Packard 1100 mass spectrometer operated in negative ionization mode. Ions were

170

detected in selective ion monitoring mode. Methanol blanks were run between each sample and

171

an analytical standard was run every 6 samples. Experimental and method blanks were run every

172

8 samples. More details on the analytical approach can be found in the Supporting Information

173

and Table SI.2.

174 175

Effect of PFAS on Zeta Potential. Surface accumulation of PFAS on bacteria and liposomes

176

was determined through zeta potential measurements. Bacteria were grown, washed three times

177

with phosphate buffer, and resuspended in an equal volume of phosphate buffer. Washed

178

bacteria were aliquoted into test tubes and amended in duplicate with either methanol (control) or

179

50 mg/L PFBS, PFHxS, PFOS, PFBA, PFOA, or PFNA in a methanol stock. Equal quantities of

180

methanol were added to all treatments and represented less than 0.25% of the total volume.

181

Bacteria were incubated at room temperature with PFAS for 20 min prior to zeta potential

182

measurements. Bacteria were incubated for 20 min because partitioning to suspended lipid

183

bilayers was rapid (Figure 2). The suspended solids concentration was measured in each replicate

184

culture. Details can be found in the Supporting Information.

185



186

The interactions of PFAS on liposomes was tested in a similar manner. Briefly, liposomes in

187

phosphate buffer were aliquoted into test tubes. PFAS were added via methanol stock, and all

188

treatments, including a PFAS-free control, contained equivalent amounts of methanol. The

189

concentration of liposomes was equivalent to the mass of the phospholipid POPC present, 1

190

mg/mL.

191 192

Electrophoretic mobilities were measured using a ZetaPALS Potential Analyzer (Brookhaven)

193

and zeta potentials estimated using the Smoluchowski equation.33 Each measurement was taken

194

five times, and particle diameter and suspended solids concentration were determined before

195

analysis. Zeta potential values calculated in the five measurement cycles were evaluated for

196

increasing or decreasing trends and if such trends were detected, the measurement was repeated.

197 198

PFAS Incorporation into Supported Phospholipid Bilayers and Impact on Phase Transition.

199

A quartz crystal microbalance with dissipation monitoring (QCM-D, Q-Sense E1, Biolin

200

Scientific) was used to evaluate the extent of PFAS incorporation into phospholipid bilayers and

201

the concomitant effect on their gel-fluid phase transition behavior. QCM-D experiments were

202

performed with a Q-Sense flow module and silica-coated QCM-D crystals (Biolin Scientific,

203

QSX 303) upon which SPBs were formed. Details are provided in the Supporting Information.

204 205

PFAS incorporation into the bilayer was tested consecutively. To begin, an initial PFAS-free

206

control experiment was performed with a quantity of methanol equivalent to bilayer exposed to

207

PFAS. Tris buffer amended with methanol was injected into the QCM-D flow module for 5 min.

208

After 5 min, the pump was turned off and the bilayer was equilibrated under stagnant conditions

209

(15 min) while subjected to a linear temperature variation (from 30°C to 15°C at 0.3°C/min).


Page 10 of 27

Page 11 of 27


210

After this, the bilayer was re-equilibrated at 30°C and the bilayer was exposed to PFAS in Tris

211

buffer. Experiments were performed with PFBS (50 mg/L), PFOS (0.1 mg/L, 0.3 mg/L, 1 mg/L,

212

5 mg/L, and 50 mg/L), PFBA (50 mg/L), PFOA (1 mg/L, 50 mg/L), and PFNA (1 mg/L, 50

213

mg/L) added to Tris via methanol stock. The PFAS were injected into the QCM-D for 15 min

214

and allowed to partition into the SPB for an additional hour. After one hour of equilibration

215

under stagnant conditions, the temperature was varied as described above.

216 217

The QCM-D data was analyzed as described in the Supporting Information.

218 219

Data Analysis

220

Kbacteria Analysis. Solid-water partitioning is described by the distribution ratio, Kbacteria. The

221

equation used to determine this partitioning coefficient is given below, with Cs and Cw

222

representing the equilibrium PFAS concentration in the solid and aqueous phases, respectively.

223

The mass of the bacteria was determined through suspended solids measurements.

224

Kbacteria =

225 226

Statistical differences between Kbacteria values were determined using the two-sided Student’s t-

227

test with Welch’s correction or Spearman’s correlation, as detailed in the Supporting

228

Information. For all statistical analyses, a P

Partitioning and Accumulation of Perfluoroalkyl Substances in Model

Recommend Documents