Mapping Indicators of Toxicity for Polycyclic ... - ACS Publications

Article

Subscriber access provided by SUNY DOWNSTATE

Mapping Indicators of Toxicity for Polycyclic Aromatic Compounds in the Atmosphere of the Athabasca Oil Sands Region Narumol Jariyasopit, Tom Harner, Dongmei Wu, Andrew Williams, Sabina Halappanavar, and Ky Su Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b02058 • Publication Date (Web): 09 Sep 2016 Downloaded from http://pubs.acs.org on September 13, 2016

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 free 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 accessible to all readers and 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.

Environmental Science & Technology 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 26

Environmental Science & Technology

1

1

Mapping Indicators of Toxicity for Polycyclic Aromatic Compounds in the Atmosphere of

2

the Athabasca Oil Sands Region

3

Narumol Jariyasopit1, Tom Harner1,*, Dongmei Wu2, Andrew Williams2, Sabina Halappanavar2,

4

Ky Su1

5 6 7 8 9 10

1

Air Quality Processes Research Section, Environment and Climate Change Canada, Toronto, Ontario, M3H 5T4 Canada 2

Environmental Health Science and Research Bureau, Health Canada, Ottawa, Ontario K1A 0K9, Canada Corresponding author: Dr. Tom Harner, phone: +1-416-739-4837, e-mail address: [email protected] For submission to Environmental Science & Technology

11 12 13

ABTRACT Extracts of passive air samples collected from 15 passive sampling network sites across

14

the Athabasca Oil Sands region were used to explore the application of in-vitro assays for

15

mutagenicity (Salmonella mutation assays) and cytotoxicity (Lactate Dehydrogenase Assay) to

16

assess the toxicity of the air mixture. The air monitoring of polycyclic aromatic compounds

17

(PACs) and PAC transformation products, including nitrated polycyclic aromatic hydrocarbons

18

(NPAHs) and oxygenated polycyclic aromatic hydrocarbons (OPAHs) was then linked to the

19

potential toxicity of air. The PACs in air during April-May 2014 were elevated near mining

20

activities and declined with distance from the source region, whereas NPAHs and OPAHs

21

exhibited a more variable spatial distribution with highest levels in Fort McMurray. Overall, the

22

air samples exhibited a weak mutagenicity. The highest indirect-acting mutagenicity was

23

observed for sites closest to mining activities; however, the indirect-acting mutagenicity did not

24

decline sharply with distance from mining areas. Indirect-acting mutagenicity was strongly 1 ACS Paragon Plus Environment


2

25

correlated with levels of total PACs, benzo(a)pyrene equivalent mass, and OPAHs. Most of the

26

samples exhibited cytotoxic potential but the magnitude of the response was variable across the

27

sample region and did not correlate with levels of target analytes. This indicates that PACs and

28

PAC derivatives were not a major contributor to the cytotoxicity observed in the air samples.

29

Keywords: passive air sampling, in vitro assays, toxicity indicator map, mutagenicity,

30

cytotoxicity, polycyclic aromatic compounds, PAC, PAH, PAH transformation products, PUF

31

disk.

32

INTRODUCTION

33

Canada’s oil sands are the third largest petroleum reserves in the world. The main areas

34

are located in northern Alberta, covering approximately 142,200 km2.1 It is estimated the

35

Alberta’s reserves contain 170 billion barrels of recoverable oil.1 The Athabasca oil sands region,

36

located in northeastern Alberta along the Athabasca river valley, is the largest of three oil sands

37

deposits. Methods used to recover bitumen include open-pit surface mining and in-situ steam

38

assisted extraction. The surface mining, accounting for 20% of the oil sands production, is

39

limited to the Athabasca oil sands region where bitumen deposits are close to the surface.

40

Impacts from oil sands development and operation, which have been rapidly expanding in the

41

past decade, have been of interest from a variety of environmental aspects.

42

Under the Joint Canada/Alberta Implementation Plan on Oil Sands Monitoring (JOSM),2

43

passive air sampling monitoring network was established to measure PACs, alkylated PACs

44

(Alk-PACs), dibenzothiophene (DBT), and alkylated DBTs.3 This passive air monitoring

45

network is part of a broader ‘Deposition and Effects’ component; 3-5 which is part of an even

46

more diverse and integrated monitoring effort that includes other environmental media including 2 ACS Paragon Plus Environment

Page 2 of 26

Page 3 of 26


3

47

water, vegetation, soil, sediment and wildlife to better understand the cumulative environmental

48

effects associated with mining activities.2 In the Athabasca oil sands region, the potential

49

emission sources of these chemicals consist of open mining areas, upgrader facilities, mine

50

fleets, tailing ponds, vehicular emissions, as well as forest fires. The petrogenic-derived PAC

51

composition in environmental samples has a distinct feature in which Alk-PACs, DBT and Alk-

52

DBTs dominate over unsubstituted polycyclic hydrocarbons (PAHs).6, 7 Therefore, PAH

53

diagnostic ratios, for example the pyrogenic index, FLA/PYR, etc. are commonly used to

54

distinguish environmental samples being contaminated by combustion vs petroleum sources.8-10

55

However, it should be noted that deviation from the proposed values of the PAH diagnostic

56

ratios as well as other uncertainties have been reported, and so they should be interpreted with

57

caution.3 An earlier study demonstrated that both gas- and particulate-phase PACs and Alk-

58

PACs were captured by the passive air samplers used in the network, at the same sampling rate

59

of 5 m3/d.11 Since 2013, PAC derivatives, including nitrated polycyclic aromatic hydrocarbons

60

(NPAHs) and oxygenated polycyclic aromatic hydrocarbons (OPAHs) have been included in the

61

list of target analytes for the passive sampling program. The presence of PAC derivatives in air is

62

due to secondary formation in the atmosphere and/or co-release with the parent PACs from

63

incomplete combustion processes.12, 13 A number of NPAHs and OPAHs are listed as “probably’

64

and “possibly” carcinogenic to humans 14, 15 Previous studies showed that they were significant

65

contributors to the mutagenicity of air extracts.16-18 The importance of these PAC derivatives is

66

also recognized because of their propensity to associate with particulate matter which enhances

67

their deposition to terrestrial and aquatic surfaces and introduces them to the food chain.12

68 69

Studies of environmental toxicity in the Athabasca oil sands region, have focused on contaminants present in oil sand process water 19-21 and river sediments22, 23 but toxicity

3 ACS Paragon Plus Environment


4

70

associated with contaminants in air has been rarely investigated.24 Ambient air quality is often

71

evaluated based on representative concentration limits of gas pollutants. For PAH mixtures in

72

air, air quality objectives are based on benzo(a)pyrene (BaP), which is a known carcinogen.25

73

However, these air guidelines do not include PAH transformation products, Alk-PACs, DBTs,

74

and other toxic components of ambient air that may be elevated in air due to mining activities. In

75

the current study the connection is made between chemical concentrations in air and the

76

assessment of the corresponding toxicity. Toxicity indicator responses are assessed for the air

77

mixture captured by polyurethane foam (PUF) disk type passive air samplers deployed at 15

78

sampling sites in the Athabasca oil sands region during April to May 2014. This assessment of

79

toxicity indicators takes into account the interactions among multiple chemical components,

80

which consist of known and unknown chemicals, in the air mixture. The approach has been

81

recently applied to explore the toxic potencies of air mixtures collected by PUFs in Eastern

82

Europe.26 The objectives were to 1) measure PACs, NPAHs, and OPAHs in the oil sands

83

samples, 2) assess toxicity indicator responses of the air extracts using two in-vitro bioassays

84

including salmonella mutation assay and a mammalian lactate dehydrogenase cytotoxicity assay,

85

and 3) develop toxicity indicator maps. Mapping of toxicity indicators is a pragmatic approach

86

for assessing environmental health hazard as a complement to air concentration mapping.

87

EXPERIMENTAL SECTION

88

Chemicals and Materials. The monitored parent PACs and Alk-PACs were

89

acenaphthylene (ACY), acenaphthene (ACE), fluorene (FLU), phenanthrene (PHE), anthracene

90

(ANT), dibenzothiophene (DBT), retene (RET), fluoranthene (FLA), pyrene (PYR),

91

benz(a)anthracene (BaA), chrysene (CHR), benzo(b)fluoranthene (BbF), benzo(k)fluoranthene

92

(BkF), benzo(a)pyrene (BaP), perylene (PER), triphenylene (TRI), indeno(1,2,3-c,d)pyrene 4 ACS Paragon Plus Environment

Page 4 of 26

Page 5 of 26


5

93

(IcdP), dibenzo(a,h)anthracene (DahA), benzo(ghi)perylene (BghiP), the C1-, C2-, C3-, and C4-

94

alkylated PACs for NAP, FLU, PHE/ANT, FLA/PYR, BaA/CHR/TRI, and DBT. ΣPAHs is the

95

sum of all individual unsubstituted PAH and DBT concentrations while ΣPACs is the sum of alk-

96

PAC concentrations and the ΣPAHs. The 31 NPAHs and 8 OPAHs monitored are listed in Table

97

SI.1.

98

Deuterium-labeled PAHs, NPAHs, and 13C-phenanthrene were purchased from

99

Cambridge Isotope Labs (Andover, MA), CDN Isotopes (Point-Claire, Quebec, Canada), and

100

Chiron AS (Norway). The isotopically labeled recovery PAH and NPAH surrogates included

101

2,6-dimethylnaphthalene-d12, acenaphthene -d10, 13C-phenanthrene, anthracene-d10,

102

benzo(b)naphtho(2,1-d)-thiophene-d10, chrysene-d12, benzo(b)fluoranthene-d12, benzo(e)pyrene-

103

d12, dibenz(a,h)anthracene-d14, indeno(123-cd)pyrene-d12, 1-nitronaphthalene-d7, 2-methyl-1-

104

nitronaphthalene-d9, 5-nitroacenaphthene-d9, 9-nitroanthracene-d9, 3-nitrofluoranthene-d9, 1-

105

nitropyrene-d9, and 6-nitrochrysene-d11. The labeled PAH and NPAH internal standards included

106

fluorine-d10 and benz(a)anthracene -d12, 2-nitrobiphenyl-d9, and 2-nitrofluorene-d9. PUF disks

107

(TE-1014, 14 cm diameter × 1.35 cm thick) were purchased from Tisch Environmental (Village

108

of Cleves, OH).

109

Sampling. The sampling sites and sample deployment have been described previously.3

110

Briefly, PUF disks were pre-cleaned with accelerated solvent extraction (ASE) using acetone,

111

petroleum ether, and acetonitrile before use. PUF disks were deployed from April-May 2014

112

(~60 days) at 15 passive air sampling monitoring sites (Figure 1), which have been operated by

113

the Wood Buffalo Environment Association (WBEA). A linear phase sampling rate of 5 m3/d

114

was derived previously for both gas- and particle-phase PAHs.11 Two field blanks were included

115

in the sampling. Duplicate samplers were deployed at sites L05 and L06. 5 ACS Paragon Plus Environment


6

116

Sample preparation. PUF disks were cut in half. All the samples were extracted using

117

the ASE with petroleum ether/acetone (75/25, v/v; 2 cycles). The half that was subjected to

118

toxicity indicator testing was extracted without being spiked with recovery surrogates. The

119

extracts were evaporated to dryness under a stream of N2 and the residue was dissolved in 500µl

120

of dimethyl sulfoxide (DMSO). Because the extracts were not subjected to any purification or

121

“clean-up” steps prior to conducting the toxicity assays, the responses reflect the whole air

122

mixture which consists of multiple chemical components and their interactions. The second half

123

of the PUF disk that was subjected to chemical analysis was spiked with recovery PAH and

124

NPAH surrogates prior to the extraction. The extracts were purified using 5 g silica columns

125

(Mega BE-SI, Agilent Technologies, New Castle, DE), eluted with petroleum ether/acetone

126

(50/50, v/v) and were spiked with labeled PAH and NPAH internal standards. The analysis of

127

PACs was conducted using a gas chromatography mass spectrometry (GCMS, Agilent 6890

128

coupled with an Agilent 5975 MSD), using electron impact in selected ion monitoring (SIM)

129

mode, on a DB-XLB column (30 m × 0.25 mm I.D., 0.25 µm film thickness, Agilent

130

Technologies). The analysis of NPAHs and OPAHs was conducted using a GCMS (Agilent

131

7890A coupled with and Agilent 7000 MSD), in electron capture negative ionization (ECNI), on

132

a DB-5 column (30 m × 0.25 mm I.D., 0.25 µm film thickness, Agilent Technologies).

133

Quality Assurance and Quality Control. All data were recovery and blank corrected.

134

Average surrogate recovery for PACs, PAC derivatives were 86% (±22%) and 105% (±56%),

135

respectively. The instrumental and method detection limits (IDL and MDL) are given in Table

136

SI.1. Values below MDL were replaced by 2/3 of MDL.

137

Toxicology Studies.

138

Salmonella Mutation Assays. Salmonella strains TA98, Environmental Bio-detection 6 ACS Paragon Plus Environment

Page 6 of 26

Page 7 of 26


7

139

Products Inc. lyophilized preparation, were used in the study. The full procedure for the

140

Salmonella reverse mutation assays have been described in detail elsewhere by Maron and

141

Ames.27 Briefly, 40 µl samples in DMSO, 2 ml melted top agar (45ºC), 0.5 ml of 0.1 M sodium

142

phosphate buffer or rat liver S9 mix (an exogenous metabolic activation system based on rat liver

143

enzymes, Moltox Inc. and EBPI), and 0.1 ml of bacteria were mixed and added to a plate

144

containing Vogel-Bonner minimal agar. After the bacteria-containing agar was solidified for 30

145

minutes, the plates were incubated at 37ºC in inverse position for 2 days. The bacterial revertant

146

colonies (bacteria that have been mutated to grow on media deficient in histidine) were counted

147

using a National Institute of Standard and Technology (NIST) Integrated Colony Enumerator. All

148

air samples were tested in triplicate. The positive control concentrations for the assays with and

149

without the rat S9 mix were 2 µg/plate of 2-aminoanthracene (2-AA) and 10µg/plate of 2-

150

nitrofluorene (2-NF), respectively. Average revertant counts per plate of the positive controls

151

were 2608 ± 1721 and 992 ± 181 for the assays with and without rat liver S9 mix, respectively.

152

The negative control plates were plated with 40 µl of DMSO only. Average spontaneous

153

revertant counts per plate of DMSO alone were 20 ± 6 and 15 ± 6 for the assays with and without

154

rat liver S9 mix, respectively. Cell viability was verified by examining the background lawn at

155

the highest sample concentration level and no adverse effect was observed (original bacterial

156

colonies).

157

Cell Culture and Treatment: Adenocarcinomic human alveolar basal epithelial cells

158

(A549) were obtained from ATCC (Manassas, VA, USA). A549 cells are routinely used for

159

characterizing pulmonary responses to environmental chemicals. Cells were cultured in F12K

160

medium (ATCC, Manassas, VA, USA) supplemented with 10% (v/v) fetal bovine serum and 100

161

µg/ml streptomycin, 100 U/ml penicillin (Invitrogen, Burlington, ON, Canada). Cells (4x104)



8

162

were grown to ~90% confluence in 96 well plates at 37ºC in a humidified atmosphere of 5%

163

CO2. Following a 24 h incubation period, the cells were treated with 1:125, 1:250, 1:500, and

164

1:1000 dilutions of the air samples, each in triplicate. Controls were treated with DMSO (Fisher

165

Scientific, New Jersey) only (n=3).

166

LDH cytotoxicity assay: The LDH assay kit (Abcam, Toronto, ON, Canada) was used to

167

measure the membrane integrity and release of cytoplasmic LDH into the medium. The assay

168

was performed according to the manufacturer’s protocol. Cells treated with 10% cell lysis

169

solution (all cells are killed resulting in maximum LDH release) were used as positive assay

170

controls. Briefly, following 24 h exposure of cells to air sample extracts vehicle control (DMSO)

171

or 10% cell lysis solution for 24 h, the culture plates were centrifuged at 600 x g for 10 min. The

172

culture supernatants (10 µl) were transferred to wells of a new 96-well assay plate and mixed

173

with 100 µl of LDH Reaction Mix. Samples were incubated for 30 min in the dark and the

174

absorbance was measured at 450 nm. Changes in the absorbance reflect the degree of LDH

175

release. Percent cellular toxicity was calculated as average increase in LDH released from cells

176

treated with air extracts (LDH release is expected) from the matched controls treated with

177

DMSO only (where LDH release is minimal or absent).

178

RESULTS AND DISCUSSION

179

Spatial Distribution of PACs. The passive sampling site map for this study is given in

180

Figure 1. Site 11 (Lower camp) has been previously designated as the main oil sands source

181

region given its proximity to oil sands related emissions.3 Individual PAC and PAC derivative air

182

concentrations at all the sampling sites are given in Table SI.2 to Table SI.4 and expressed as

183

mass collected by the passive samplers over the common deployment period. Due to its high

184

volatility and abundance, naphthalene is not included in the reported ΣPACs values as well as the 8 ACS Paragon Plus Environment

Page 8 of 26

Page 9 of 26


9

185

sum of the 15 US Environmental Protection Agency priority PAHs (ΣPAH15priority), which is

186

consistent with our previous studies (Table 1).3, 11 Consistent with our early study, the highest

187

ΣPACs mass (76,800 ng, which represents 99.7% of the total target compounds i.e.

188

ΣPACs+NPAHs+OPAHs) was observed at L11 (Figure 1 and Table 1) that is centrally located

189

near several mining-related sources and in an area of dense mining activity. Illustrated in Figure

190

2, spatial distribution of ΣPACs showed a declining trend with increasing distance from L11

191

(Table 1). On average, the Alk-PACs accounted for 94% of the ΣPACs for all the sampling sites,

192

except for L06 (Residential site at Fort McMurray) and L14 (Anzac) where the Alk-PACs

193

accounted for 85% and 59%, respectively (Figure SI.1). High PAHs/Alk-PACs ratios at the sites

194

L06 and L14 were previously reported by Schuster et al.3 and Hsu et al.28 reported elevated

195

PAHs in air at site L14. The elevated lower molecular weight PAH concentrations at L14 were

196

attributed to the evaporation from nearby Gregoire lake.28 The pyrogenic index for the samples is

197

given in Table 1. The pyrogenic index has been widely used in several studies to determine PAH

198

compositional features.8, 29, 30 It is defined as the ratio of the sum of other three- to six-ring PAHs

199

(ACY, ACE, ANT, FLA, PYR, BaA, BbF, BkF, BaP, PER, IcdP, DahA, BghiP, noting that

200

biphenyl and benzo(e)pyrene were not monitored in this study) to the sum of the five target Alk-

201

PACs (NAPs, PHEs, DBTs, FLUs, and CHRs). In bitumen and crude oils, unsubstituted PAHs

202

are present in smaller amounts, relative to the alkylated homologues, therefore, a higher index

203

value indicates a dominant contribution from pyrogenic sources (e.g. vehicle exhaust, wood

204

burning) versus, for instance, petrogenic sources (e.g. bitumen, crude oils).8 The pyrogenic index

205

of the oil sands samples and conventional crude oils were previously reported to range from

206

0.01 to 0.037 while diesel soot samples ranged from 0.8 to 2.0.8 The pyrogenic index values at

207

L06 (0.0706, town of Fort McMurray) and L14 (0.159, background site near Gregoire Lake)



10

208

were relatively high compared to the other sampling sites where pyrogenic index values ranged

209

from 0.01 to 0.04, indicating dominance of petrogenic sources (Table 1). The influence of

210

pyrogenic (combustion) sources at site L06 (Fort McMurray) is further evident from the marker

211

ratio FLA/PYR.9 The FLA/PYR ratio at L06 of 1.2 is substantially higher compared to a site in

212

the source region, for instance site L11 (0.38, Lower Camp) (Table SI.2). Since FLA is enriched

213

relative to PYR in combustion sources, larger values of the ratio indicate greater contribution

214

from combustion. FLA/PYR values observed in urban areas were often closer to unity, indicating

215

dominance of combustion sources of PACs.31-33 Furthermore, an analysis of raw Alberta oil

216

sands samples revealed that the PYR concentration was 3-4 times higher than FLA,7 which

217

would lead to a low value of the FLA/PYR ratio when petrogenic sources dominate.

218

The detected NPAHs and OPAHs were 1-nitronaphthalene (1-NN), 1-methyl/2methyl-5-

219

nitronaphthalene (1M5NN/2M5NN), 2-nitronaphthalene (2-NN), 9-nitroanthracene (9-NA), 2-

220

nitrodibenzothiophene (2-NDBT), 2,8-dinitrodibenzothiophene (DNDBT), 9-fluorenone (FLO),

221

9,10-anthraquinone (ANQ), 2-methyl-9,10-anthraquinone (MANQ), benzo(a)fluorenone (BaFL),

222

benzo(b)fluorenone (BbFL), benzanthrone (BENZ), and benz(a)anthracene-7,12-dione (BaAD).

223

NPAHs and OPAHs are of interest as some of them exhibit greater mutagenicity than the parent

224

PAHs34, 35; previous studies have addressed concerns over toxic potential of air extracts

225

containing atmospheric transformation products of PAHs.16, 36 However, this issue has so far not

226

been addressed in the oil sands region and levels of PAH transformation products have so far not

227

been reported for air. In the Athabasca oil sands region, the sum of ΣNPAHs and ΣOPAHs mass

228

was highest at L06 (Fort McMurray) (Figure 1 and Table 1). Unlike the PAC spatial distribution

229

which demonstrated a decline in the levels in air with increasing distance from L11, the patterns

230

of ΣNPAHs and ΣOPAHs were variable across the sampling sites (Figure 2B).


Page 10 of 26

Page 11 of 26


11

231

Thus far, there has been only one study investigating NPAH levels in air using passive air

232

samplers.37 That study collected PUF disk type passive air samples (similar to those used in this

233

study) at multiple locations in north China during July to October 2011 (90 days) and reported

234

average NPAH masses of 341, 235, and 241 ng/sample for a megacity, town, and rural sites,

235

respectively. These levels are one or two orders of magnitude higher than observed in the oil

236

sands region (for 60 day deployments). Other studies have reported NPAH levels in air measured

237

using active air samplers and reported in mass per unit of air sampling volume.18, 31, 38, 39 In order

238

to put the NPAH and OPAH levels in oil sands region into a broader context, the ambient NPAH

239

and OPAH concentrations were converted to more conventional air concentration units (volume

240

based) using the sampling rate of 5 m3/d, the linear phase sampling rate that was derived

241

previously for gas- and particle-phase PAHs.11 This results in an equivalent sample air volume of

242

300 m3 for the entire sample. The resulting average concentrations in air for ΣNPAHs in the oil

243

sands region was 20.9 pg/m3, ranging from below detection limit at R03 to 47.8 pg/m3 at L05

244

(Mannix). These levels of ΣNPAHs are about an order of magnitude lower than the

245

concentrations measured in urban areas, such as Los Angeles (851 pg/m3),31 Baltimore (318

246

pg/m3),39 Marseille, France (710 pg/m3),38 and Beijing, China (792 pg/m3),18 but it is comparable

247

to a rural area in Marseille, France (28 pg/m3).38

248

Previous studies have shown that 2-NF and 2-NP, formed by gas-phase OH radical

249

reactions, were the most abundant particle-associated NPAHs found in ambient air.31 For

250

instance, Lin et al. reported an average of 2-NF value of 13.7 ng/sample for rural sites in north

251

China, using passive samplers.37 However, in the current oil sands study 2-NF and 2-NP were

252

below the detection limits in all samples. On the other hand, the compound 2-NDBT, which is

253

rarely included or detected in other air monitoring programs, was detected with the highest level



12

254

at L11 (6.9 ng/sample, Table SI.4). Overall, the NPAH concentrations in the oil sands region

255

during April-May 2014 were generally lower than the values measured in the other urban areas.

256

Although the parent PAH concentrations at sites that are closer to the source region (L11) are in

257

the range of values reported at urban sites in European and North American countries,3 the

258

relatively low abundance of the photochemically-derived NPAHs implies that atmospheric

259

nitration potential is low at this oil sands impacted site.

260

The OPAHs levels in air samples did not correspond to areas of intense mining activities

261

(L11). The average value for ΣOPAHs across all sites was 129 ng/sample, ranging from 17.9

262

ng/sample at R09 to 425 ng/sample at L06 with FLO being the dominant OPAH in all the

263

samples. FLO is volatile and found primarily in the gas-phase.38 The town of Fort McMurray

264

(L06) exhibited the highest FLO concentration of 206 ng/sample or approximately 644 pg/m3,

265

using the same sample air volume conversion as for the NPAHs above. This result for FLO is

266

higher than reported in other studies, for instance, Beijing (urban, 129 pg/m3),18 Paris (traffic

267

site, 71 pg/m3),40 and Tokyo (urban, 21 pg/m3)32 but lower than observed in Marseille, France

268

(urban, 3,577 pg/m3).38

269

OPAHs are ubiquitous in urban ambient air and previous studies have associated them

270

with fine particles released from combustion and products from reactions with atmospheric

271

oxidants.12, 41, 42 Because the OPAHs are not associated with a specific pathway their relative

272

concentrations in air cannot be used to differentiate sources. Nonetheless, the relatively high

273

levels of OPAHs at L06 was consistent with its distinct PAH marker ratios (discussed earlier)

274

which suggests that the L06 air sample was influenced by combustion sources (e.g. vehicular

275

exhausts). These combustion sources could also be associated with primary emissions of the

276

OPAHs. At the same time, OPAHs could also be formed as a transformation product in air due to 12 ACS Paragon Plus Environment

Page 12 of 26

Page 13 of 26


13

277

local photochemical atmospheric aging that might be enhanced at L06 by gaseous oxidants (e.g.

278

O3, NOx) which are related to the combustion sources. Yet a third possibility is that the OPAHs

279

at site L06 are associated with regional transport and transformation (photochemical aging) of

280

primary emissions from site 11 which is located upwind of L06.

281

Mapping of Toxic Indicators. Air samples extracts were subjected to in vitro assays to

282

assess potential toxicity endpoints including mutagenicity (both direct and indirect) and

283

cytotoxicity. Approximately 4% of the sample was used for this purpose representing an

284

equivalent sample air volume of 12 m3 air.

285

Mutagenicity. Mapping of indirect-acting and direct-acting mutagenicity are shown in

286

Figure 3 and Figure SI.2. Parent PACs as well as some PAH derivatives are known to be

287

indirect-acting mutagens, requiring metabolic activation system, in the Salmonella assay (and in

288

humans), to convert them into the active form.43, 44 On the other hand, some NPAHs and OPAHs

289

exhibit direct-acting mutagenicity in the Salmonella assay without metabolic activation.45 The

290

results of the mutagenicity tests are expressed as relative mutagenicity, which is the mean

291

mutagenic response of a sample divided by the background response (assay blank tests) (Table 1)

292

in which all the samples were tested in triplicate. Mutagenic responses (number of revertants

293

(mutated bacteria colonies) per plate) are given in Table SI.6. In this case a relative mutagenicity

294

of 1.0 indicates the same response as for the assay blank. The mutagenicity of air sample field

295

blanks was close to unity for both assays (with and without metabolic activation) indicating that

296

the field blanks and sample matrix contained insignificant amount of mutagens. Overall, the

297

indirect-acting mutagenicity of the samples rarely exceeded two times of the background,

298

indicating a weak mutagenicity. Although the two-fold rule has been widely applied as an

299

indicator of a positive response, there have been published studies questioning validity of the 13 ACS Paragon Plus Environment


14

300

rule.46-48 As such, the evidence calls for future study in which multiple doses are tested.

301

However, results based on the single-dose testing in this study are informative in the aspect of

302

spatial comparison of mutagenicity. The highest and statistically significant relative indirect-

303

acting mutagenicity of 1.8 was observed at site L11 corresponding to the site having the highest

304

ΣPACs in air; however, the indirect-acting mutagenicity did not decline with increasing distance

305

from L11 in the same way that the PAC concentrations in air declined.

306

The PACs concentrations values in air that were used for determining correlations with

307

toxicity indicators were not blank corrected in order to reflect the actual amounts of chemicals in

308

the samples which could potentially induce the toxic responses. PACs were shown to be a major

309

contributor to the indirect-acting mutagenicity of the samples given a strong and statistically

310

significant correlations between the indirect-acting mutagenicity and ΣPACs (R = 0.701, p-value

311

< 0.05) (Table SI.5). The correlation with the indirect-acting mutagenicity was slightly better for

312

the sum of unsubstituted PAHs (R = 0.758, p-value < 0.05) and was slightly lower for the ΣAlk-

313

PACs (R = 0.698, p-value < 0.05) (Table SI.5). To a lesser extent, the ΣOPAHs in air was

314

significantly correlated with the indirect-acting mutagenicity (R = 0.51, p-value < 0.05),

315

however, the ΣNPAHs was not significantly correlated with the indirect-acting mutagenicity.

316

Mutagenicity in the air samples can also arise from direct-acting mutagens, including

317

some NPAHs and OPAHs,12, 16, 49 that do not require metabolic activation (rat S9 mix) to be

318

mutagenic. Similar to the indirect-acting mutagenicity, L11 had the greatest direct-acting

319

mutagenic potential; however, all the samples generally induced low responses (Figure SI.2). It

320

is interesting that the L06 (Fort McMurray) sample that had the highest levels of sum of NPAH

321

and OPAH yielded considerably low direct-mutagenic response (i.e. relative mutagenicity of

322

1.2). The L06 sample was dominated by FLO and ANQ which accounted for 48% and 20%, 14 ACS Paragon Plus Environment

Page 14 of 26

Page 15 of 26


15

323

respectively, of the sum of NPAHs and OPAHs. Despite the abundance of FLO in ambient air

324

from studies discussed earlier, data on the mutagenicity of FLO are scarce. Although data on in-

325

vitro mutagenic potential of ANQ are available, the results are so far inconclusive due to issues

326

related to impurity in the tested standards.15 In this study the observation of low direct-mutagenic

327

response at L06 may suggest that the measured NPAHs and OPAHs and other unknown

328

chemicals are possibly weak direct-acting mutagens. The known potent direct-acting mutagens

329

among the NPAHs and OPAHs are 1-nitropyrene and dinitropyrenes, which were below the

330

detection limits for all samples in this study. In addition to the absence of potent direct-acting

331

mutagens, such low response could be attributed to suppression of the direct-acting mutagenicity

332

which may have been caused by the presence of PAHs. This was believed to promote the

333

interactions between aromatic compounds and suppress nitroreductase activity.17, 50 Although

334

direct-acting mutagenicity was also correlated with PACs (R = 0.600, p-value < 0.05), this

335

response is probably attributed to unknown chemicals that were co-emitted with or degraded

336

from the parent PACs, since PACs are indirect-acting mutagens that require exogeneous

337

bioactivation.

338

Cytotoxicity. Cytotoxicity is expressed as percentage increase in LDH release in cells

339

treated with air extracts relative to the control samples (DMSO). Field blank extracts did not

340

result in increased LDH release. Cells were exposed to different doses in triplicate and dose-

341

response curves for all the sites were calculated, which exhibited positive dose-response

342

relationships (Figure SI.3). For spatial comparison, the percent increases in LDH release

343

associated with the lowest dilution level (1:125) are shown in Figure 3B. The average percent

344

increase in LDH release (cytotoxicity) for samples was 120%, ranging from 3% at L13 to 223%

345

at R01 (Table 1). The increase for L11 and L06 (Fort McMurray) samples was 160% and 179%,



16

346

respectively (Table 1). In general, most of the air samples exhibited cytotoxic potential, except

347

for two sites R05 and L13 in which the increase did not exceed 20% (Figure 3). Interestingly, the

348

samples that expressed higher cytotoxicity (>200%) were from remote sites including R08, R03,

349

and R01. The prevailing winds at R08 during the sampling period were from the oil sands area,

350

whereas they were typically from the southeast at site R03 and southwest and northeast at site

351

R01, representing remote forested areas.51 This result coupled with the lack of correlations

352

between the cytotoxicity and the PACs, NPAHs, and OPAHs (Table SI.5) concentrations

353

indicate that the PACs and PAH derivatives were not a major contributor to the cytotoxicity of

354

these samples. We also conducted a statistical analysis on the multiple-dose cytotoxicity results

355

to explore dose and site relationships among the sampling sites. The results are summarized in

356

Figures SI.4 (hierarchical cluster analysis represented as a tree diagram (Dendogram)) and SI.5

357

(a two-dimensional scaling plot). The dendogram reflects similarities and differences among

358

sites based on their dose-response curves (Figure SI.3). For instance the L06 duplicate samplers

359

and site R03 occur in the same “cluster” due to their very similar dose-response curves; whereas

360

the field blank is separate from all the other sites/clusters reflecting significantly different

361

cytotoxicity response. A similar clustering or grouping of sites is also seen in two-dimensional

362

scaling plot (SI.5). The results of the statistical analyses suggested that there were statistically

363

significant site-specific differences in the cytotoxic response induced by air extracts; however,

364

the responses did not correlate with the PACs, NPAHs or OPAHs content. Future studies are

365

needed to investigate contributions of other air contaminants (not measured in the current study)

366

to the cytotoxicity responses among sampling sites.

367 368

BaP equivalent (BaPeq) mass represents the contribution of carcinogenic PAHs in a mixture expressed in terms of an equivalent amount of BaP. In this study, individual BaP


Page 16 of 26

Page 17 of 26


17

369

equivalent mass is determined by multiplying individual PAH concentration by its relative

370

potency factor (RPF).52 Derivation of RPF values were mainly based on carcinogenicity bioassay

371

data. BaPeq of each sample is the sum of the individual BaPeq masses. RPF values for the PAHs

372

measured in this study are presented in Table SI.6. Despite the exclusion of Alk-PACs, the

373

spatial distribution of BaPeq was still consistent with that of the PACs showing the highest BaPeq

374

value at L11 (Table 1). BaPeq had stronger correlation with the indirect-acting mutagenicity (R =

375

0.72, p-value < 0.05) in comparison with the direct-acting mutagenicity (R = 0.59, p-value

Mapping Indicators of Toxicity for Polycyclic ... - ACS Publications

Recommend Documents