A Field Trial of TCE Phytoremediation by Genetically Modified Poplars

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article

A Field Trial of TCE Phytoremediation by Genetically Modified Poplars Expressing Cytochrome P450 2E1 Emily K Legault, C. Andrew James, Keith Stewart, Indulis Muiznieks, Sharon L. Doty, and Stuart E. Strand Environ. Sci. Technol., Just Accepted Manuscript • Publication Date (Web): 02 May 2017 Downloaded from http://pubs.acs.org on May 3, 2017

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 30

Environmental Science & Technology

A Field Trial of TCE Phytoremediation by Genetically Modified Poplars Expressing Cytochrome P450 2E1 Emily K. Legault1*, C. Andrew James2*, Keith Stewart1, Indulis Muiznieks3, Sharon L. Doty3, and Stuart E. Strand1,5

1

Department of Civil and Environmental Engineering, UW Box 355014, University of Washington, Seattle, WA 2

Center for Urban Waters, University of Washington Tacoma, Tacoma, WA

3

College of the Environment, School of Environmental and Forest Sciences, UW Box 352100, University of Washington, Seattle, WA

5

To whom correspondence should be addressed:

Email: [email protected]

* These authors contributed equally to the work

1 ACS Paragon Plus Environment


1

Abstract. A controlled field study was performed to evaluate the effectiveness of transgenic poplars for

2

phytoremediation. Three hydraulically contained test beds were planted with twelve transgenic poplars,

3

twelve wild type (WT) poplars, or left unplanted, and dosed with equivalent amounts of

4

trichloroethylene (TCE). Removal of TCE was enhanced in the transgenic tree bed, but not to the extent

5

of the enhanced removal observed in laboratory studies. Total chlorinated ethene removal was 87% in

6

the CYP2E1 bed, 85% in the WT bed, and 34% in the unplanted bed in 2012. Evapotranspiration of TCE

7

from transgenic leaves was reduced by 80% and diffusion of TCE from transgenic stems was reduced by

8

90% compared to WT. Cis-dichloroethene and vinyl chloride levels were reduced in the transgenic tree

9

bed. Chloride ion accumulated in the planted beds corresponding to the TCE loss, suggesting that

10

contaminant dehalogenation was the primary loss fate.

Introduction. Trichloroethylene (TCE) is one of the most common groundwater contaminants in the

11 12

United States.1 Several studies have found that plants can take up and metabolize TCE, and that the

13

presence of trees may lead to enhanced groundwater remediation.2-7 Trichloroethylene can enter trees

14

via uptake with water 8-10 and by diffusion into roots.11 Poplars metabolize TCE by an oxidative pathway,

15

producing the metabolites trichloroethanol (TCOH), dichloroacetic acid (DCAA), trichloroacetic acid

16

(TCAA), and a bound TCOH-glucoside.12 In addition to metabolism within tree tissues, TCE is subject to

17

diffusion through trunk, branches, and leaves to the atmosphere and sorption to tissues, 13-18 and

18

possible transformation by endophytes.19-21

19

Many studies of phytoremediation of TCE and other chlorinated VOCs have not attempted to obtain a

20

mass balance that accounts for all possible fates or to determine levels of the products of VOC

21

metabolism, such as chloride ion.22-24 Some studies have focused exclusively on volatile loss pathways.23,

22

24

Studies of trees growing over contaminated groundwater plumes are inherently handicapped in their


Page 2 of 30

Page 3 of 30


23

ability to draw conclusions about actual effects on groundwater TCE since TCE uptake has to be

24

estimated. Some laboratory measurements have used batch exposures of trees in small open vessels,

25

precluding the achievement of a steady exposure to TCE that is required for mass balances.22

26

Another approach is to determine mass balances of TCE and its metabolic products by exposing trees

27

in field settings to TCE in contained test beds. This technique permits direct measurement of TCE

28

exposure and allows for the measurements of all fate pathways. Newman et al. 6 applied water

29

containing TCE to the subsurface of test beds planted with hybrid poplar (Populous trichocarpa x P.

30

deltoides). The planted test bed removed approximately 99% of the TCE mass applied, while only 33%

31

was removed from the unplanted control. A chlorine mass balance indicated that the TCE loss in the

32

planted bed could be accounted for by increased soil chloride. Volatilization of the pollutants from

33

leaves and stems was insignificant in this study, and subsequent test bed mass balance studies. This

34

study and others25, 26 suggest that trees may effectively treat volatile organic compounds (VOCs) in

35

proportion to the groundwater they take up and transpire, and that the contaminant concentration in

36

the root zone water is unchanged.

37

The advantages of phytoremediation include low cost and low energy input, but uses of

38

phytoremediation have been limited, largely due to its uncertain effectiveness. Genetic transformations

39

have been undertaken to increase the activities of pollutant transformations in plants.27-30 A series of

40

laboratory experiments investigated the effects of the expression of mammalian cytochrome P450 2E1

41

(CYP2E1) in hybrid poplars (Populus tremula x P. alba) and tobacco (Nicotiana tabacum cv. Xanthii)

42

against VOCs 31-34 that are substrates for CYP2E1 in the mammal. The transgenic plants degraded a wide

43

variety of compounds such as TCE, vinyl chloride, carbon tetrachloride, chloroform, benzene, toluene,

44

and bromodichloroethane at significantly higher rates compared to wild type or vector control plants.

45

Others have engineered petunia35 and alfalfa36, 37 with CYP2E1 for phytoremediation of TCE.



46

In order to evaluate the effectiveness of the transgenic poplars in a field setting, water containing an

47

equivalent amount of TCE was added to three hydraulically contained test beds for six growing seasons.

48

One test bed was planted with genetically modified hybrid poplars expressing CYP2E1, one test bed was

49

planted with unmodified hybrid poplars, and one bed was not planted. A chlorine mass balance was

50

performed to identify the fates of chlorine. Parameters such as the VOC and chloride concentration in

51

the influent and effluent water, VOC and total organic halide (TOX) concentration in plant tissues, soil

52

chloride concentration, and VOC volatilization from the soil, leaves, and stem were measured, allowing

53

the computation of mass balances and direct comparison of the three treatments. We performed soil

54

microcosm studies to estimate microbial removal of TCE in test bed soil and to differentiate between

55

tree and microbial contributions to TCE removal. To our knowledge this is the first field-scale mass

56

balance evaluation of the effectiveness of phytoremediation of an environmentally important VOC by

57

transgenic trees.

58 59

Materials and Methods.

60

Test Facility.

61

The field site was located at the University of Washington Phytoremediation Field Facility. Three

62

adjacent, hydraulically isolated test beds were used, as described previously.6 The beds each measured 9

63

m x 5.7 m, 1.4 m deep. The interior soil was removed from the beds in September 2006 and replaced

64

with clean fill (sandy loam, 67.2% sand, 26.6% silt, 6.2% clay; TOX, 6.6%).

65

The transgenic hybrid poplar (Populus tremula x P. alba) expressed rabbit CYP2E1 under the

66

cauliflower mosaic virus (CaMV) 35S promoter.31 Thirteen transgenic poplars (designated as clone 78 in

67

the laboratory, herein referred to as r2E1), and thirteen wild type (WT) INRA 717-1B4 clones


Page 4 of 30

Page 5 of 30


68

(approximately 1m) were propagated in the laboratory, acclimated to exterior conditions, and planted

69

at the field facility on April 11, 2007. One bed was planted with twelve r2E1 poplars, one with twelve WT

70

poplars, and one left unplanted. One r2E1 and one WT poplar were planted in a separate test bed and

71

not exposed to TCE. The presence of the transgene was verified by PCR on DNA extracted from poplar

72

tissues following planting using the methods described previously.38

73

Water Supply and Chemical Dosing.

74

The three test beds received approximately equal concentrations of TCE during each growing season

75

from 2007-2012; the 2012 growing season occurred from June 19, 2012 - October 28, 2012. The target

76

influent TCE concentration was 15 mg L-1 from 2007-2011, changed to 30 mg L-1 in 2012 to increase the

77

sensitivity of metabolite detection. The dosing procedure was as described previously.6 Target water

78

level depth as measured in the effluent wells was between 20 and 30 cm; although the unplanted bed

79

water level was generally higher. A minimum of 16 L d-1 was pumped out of the test beds. Additional

80

water was pumped out of the unplanted bed to match the volume transpired in the planted beds.

81

Municipal water was used for irrigation and was added to the planted beds through the influent well or

82

by surface watering as needed to maintain tree health. All test beds were open to rainfall. Rainfall levels

83

were recorded on site.

84

Sampling and Analysis.

85

Sampling and analysis of the influent and effluent water, soil chloride, volatilization of VOC from the

86

soil and tree leaves and stem, and plant tissue analysis were performed as described previously.25, 39

87

Influent water samples were collected daily and effluent samples at least weekly using methods

88

described previously.6 A Perkin Elmer Autosystem XL Gas Chromatograph (GC) was connected to a

89

Teledyne Tekmar AQUATek 70 Vial autosampler and Tekmar 3000 Purge & Trap Concentrator. Liquid



90

samples were purged with helium for 11 min at 30°C onto the concentrator and desorbed at 225°C for 4

91

min for analysis on the GC. Concentrations of the chlorinated ethenes and ions were calculated based on

92

external standards. Analysis for vinyl chloride (VC) was performed by collecting headspace samples from

93

the effluent sample vials after they had been analyzed with the GC-ECD. All sampling was performed in

94

triplicate. Analytical detection limit for VC was approximately 20 µg L-1.

95

Chloride analysis was performed with a Dionex AS40 Automated Sampler connected to a Dionex DX-

96

120 ion chromatograph (IC) with a Dionex IonPac AS14 4, 250 mm anion exchange column; eluent was

97

3.5 mM Na2CO3/1 mM NaHCO3 in degassed, deionized water. Analytical detection limit was

98

approximately 0.1 mg L-1.

99

The dissolved oxygen (DO) concentration and temperature were measured in the groundwater of

100

each test bed approximately every 14 days. Prior to sampling, a minimum of 19 L of water was removed

101

from each effluent well and wells were allowed to recharge for one hour. DO analysis was performed

102

with a YSI model 158 or 5000 DO meter.

103

Soil samples were collected three times each growing season. In 2012, samples were collected on May

104

22, 2012, August 21, 2012, and October 16, 2012. For each bed, a total of twenty-four soil core samples

105

were collected, at six locations on a 2x3 grid, and at four depths for each location (10, 30, 60, and 100

106

cm below the soil surface). Analysis was as described previously.39 Briefly, soil samples were frozen prior

107

to analysis, dried at 100°C, extracted in water, shaken, centrifuged and analyzed using a Dionex DX-120

108

ion chromatograph. Analysis of TCE in the soil matrix was performed by collecting 10-15 g soil, preserved

109

in sealed volatile organic analysis (VOA) vials with 10mL methanol, and analyzing per EPA 8260B, and

110

results reported as mg per kg dry weight.


Page 6 of 30

Page 7 of 30

111


Two sets of soil volatilization measurements were taken for the r2E1 and the WT beds in 2012. Soil

112

volatilization from the unplanted bed was measured in 2008. Analysis was as described previously,40

113

using a soil vapor collection chamber.

114

Samples for the volatilization of VOCs from leaves were collected from both planted test beds

115

simultaneously using leaf bag sampling as described previously,6 except that several leaves were

116

enclosed in Teflon bags loosely tied around the stem and air drawn through the bag with TCE trapped in

117

activated carbon tubes for later analysis by GC-ECD. Leaf areas were measured. Six leaf bag

118

measurements were performed in the r2E1 bed and seven in the WT bed. Volatilization from the tree

119

stem was measured at 0.25, 0.4, and 0.57 m heights as described previously.39 Loss of volatiles from the

120

stems (i.e., trunks) was measured using 26 cm2 diffusion traps strapped to the trunks with air flowing

121

through carbon tubes to trap emitted TCE for later analysis by GC-ECD. Four measurements were taken

122

at each height in both beds in 2012 with the exception of three measurements at 0.57 m in the WT bed.

123

The detection limits for volatilization fluxes were about 0.8 nmol of TCE m-3 air.

124

Plant tissue samples were collected from the field and the laboratory plant growth room. Sterilization,

125

sampling, and analysis were as described previously. 31, 32 Root material was not surface sterilized to

126

avoid damaging the root tissue. Plant tissue analysis included samples of root, stem, branch, and leaf

127

tissue collected from at least three different trees in each test bed. For stem tissue, two of the three

128

samples were combined in one metabolite extraction.6, 12 Leaf and core samples were collected and

129

immediately flash frozen on-site in liquid nitrogen, then homogenized and stored on dry ice for

130

transport. Analysis for TCE and free metabolites was per Newman et al. 6 using GC-ECD as

131

described above. Analytical limit for TCE was approximately 0.05 mg g-1 tissue, respectively.

132

Trichloroethanol (TCOH) analysis was performed on selected samples as described previously.12 Leaf



133

tissue was analyzed for TCOH-glucoside per Shang et al.12 The analytical detection limits for TCOH and

134

TCOH-glucoside were approximately 1.5 mg kg-1. Analysis for di- and trichloroacetic acid (DCAA, TCAA)

135

was performed by GC-ECD after extraction in methyl tert-butyl ethyl as described previously. Analytical

136

detection limits for TCAA and DCAA were approximately 0.1 and 0.05 mg g-1 tissue, respectively.

137 138 139

Total organic halide (TOX) analysis was performed on plant tissue in 2008 by Spectra Laboratories, Tacoma, WA, per EPA Manual SW-846 Method 9076. The total mass of TCE metabolites in tissue was estimated by multiplying the mean tissue metabolite

140

concentration by the respective mass of the tree compartment. Biomass estimations were performed

141

with published allometric linear regression equations for poplar root,41 stem,42 branch,43 and leaves.44 A

142

minimum of three biomass estimations were performed for each compartment with equations from

143

different sources using the median value. Total leaf area was estimated with a published allometric

144

linear regression equation for leaf area of Populus tremuloides.45

145

Statistical analysis was performed using XLStat software (Addinsoft SARL, Paris France). Since the data

146

were not always normally distributed, nonparametric procedures were used to compare distributions of

147

VOCs and chloride (Table 1).

148

Quantitative RT-PCR

149

Leaf and root samples were collected on site, immediately frozen in liquid nitrogen, transported to

150

the laboratory in dry ice, and stored at -80°C until analysis. Samples were ground under liquid nitrogen.

151

Approximately 100 mg of tissue was used with the RNeasy Plant Mini Kit (Qiagen) to extract total RNA.

152

RNA was used to synthesize cDNA with the Bio-Rad iScript kit with SYBR Green labeling agent.

153

Quantitative RT-PCR was performed with SYBR Green-labeled probes of CYP2E1 and 18S as the

154

housekeeping gene and included negative controls. The primers for the rabbit cytochrome P450 gene


Page 8 of 30

Page 9 of 30


155

(r2E1) were (forward) AATTGCCACCGTGGAGCTTT and (reverse) TATGCATAGCAAGACCGGGTTG, and for

156

the poplar 18S rRNA gene (forward) AATTGTTGGTCTTCAACGAGGAA and (reverse)

157

AAAGGGCAGGGACGTAGTCAA.

158

Soil Microcosms.

159

Microcosm experiments were conducted to measure aerobic TCE removal in soil. Soil was collected

160

from each test bed at a depth of 30 cm and left in an open container for three days to allow TCE

161

volatilization. Roots were removed from soil by visual inspection. Forty-mL VOA vials were filled with

162

approximately 10 grams of soil and sealed with septum valve caps (Mininert) and dosed with 0.05 mg

163

TCE. The soil was at approximately field capacity moisture and had sufficient headspace to remain

164

aerobic during the experiment.

165

Biotransformation activity was evaluated under four different conditions, each in triplicate with soil

166

from each planted bed: (1) TCE only, (2) TCE and 10% v/v methane, (3) TCE, 10% v/v methane, and 1%

167

v/v ethyne, and (4) TCE and 2 mL of 200 mg L-1 sodium azide. Triplicate vials with 10 mL water and TCE

168

only (no soil) served as controls for leakage. Headspace of TCE was sampled every 24 hours for 6 days

169

and analyzed on a Perkin-Elmer Autosystem GC-ECD, as described previously.32 Vials were allowed to

170

equilibrate for two hours following dosing and prior to the time-zero measurement.

171

Root Enclosure Experiment.

172

Tree roots were isolated to estimate the efflux of chloride by roots. Portions of roots at the surface in

173

each bed were exposed while still attached to the trees, excess soil removed by gentle brushing,

174

enclosed in 8 x 14 inch FoodSaver© plastic bags with sterile sand, and sealed with Aquarium sealant or

175

Teflon tape. Three root sections were enclosed in each planted bed and also in the undosed control bed

176

containing one r2E1 and one WT poplar. The roots were not surface sterilized to avoid damaging the



177

root tissue. Prior to use, the sand was autoclaved, washed three times with de-ionized water, and dried

178

at 100°C for 24 hours. Chloride concentration was measured at the beginning and end of the

179

experimental period as described above and each sand sample was extracted in duplicate.

180 181

Results.

182

Water Use/Water Balance.

183

WT and r2E1 trees grew equally rapidly during the experimental period. The r2E1 trees were about

184

6% larger than the WT trees at planting and remained slightly larger throughout the experimental period

185

(Table S1). These differences were not significant.

186

Table S2 summarizes the water use for each test bed from June 19, 2012 through October 28, 2012.

187

The accumulation of water in each test bed was calculated based on the change in water levels during

188

this period. The r2E1 and WT poplars transpired approximately 175 and 170 L d-1, respectively, during

189

the 2012 growing season, while loss of water from the unplanted bed was much less. The average water

190

levels over the growing season in the transgenic, WT, and unplanted beds were 28.4 ± 10.9 cm, 30.2 ±

191

8.5 cm, and 53 ± 13.9 cm, respectively.

192 193

The average DO concentrations in the groundwater were 4.1 mg L-1, 1.0 mg L-1, and 0.68 mg L-1 for the r2E1 bed, WT bed, and unplanted bed, respectively. Groundwater temperature is shown in Figure S1.

194

Influent and Effluent Water VOCs and Chloride.

195

Influent water for all beds contained an average of 31.4 ± 4.4 mg L-1 TCE during the 2012 growing

196

season. The effluent TCE concentrations are shown in Figure 1. The average effluent concentration of


Page 10 of 30

Page 11 of 30


197

TCE from June 19, 2012 through October 28, 2012 was 3.24 ± 1.3, 2.93 ± 0.96, 2.36 ± 0.97 mg L-1 for the

198

r2E1, WT, and unplanted beds, respectively. Interestingly, the distributions were highly bimodal

199

depending on season. The system appeared to reach steady-state conditions in the second half of the

200

season; a moderate downward trend was observed in both planted beds and an upward trend in the

201

unplanted bed from August 15, 2012 to October 28, 2012. Average effluent TCE concentrations during

202

this period were 3.56 ± 0.45 mg L-1, 3.55 ± 0.39 mg L-1, and 3.07 ± 0.32 mg L-1 for the r2E1, WT, and

203

unplanted beds, respectively. The effluent water cDCE and VC concentrations are shown in Figure 2.

204

The average influent TCE fluxes into the beds were 3.6, 3.7, and 0.73 g d-1 for the r2E1, WT, and

205

unplanted beds, respectively. The average effluent TCE fluxes from the beds were 0.394 ± 0.158, 0.360

206

± 0.118, and 0.057 ± 0.024 g d-1 for the r2E1, WT, and unplanted beds, respectively. Thus, the loss rates

207

of TCE from the beds were about 3.3, 3.3, and 0.67 g d-1 for the r2E1, WT, and unplanted beds,

208

respectively.

209

The average influent chloride concentration for 2012 was 2.66 ± 0.31 mg L-1. Effluent water chloride

210

concentration measurements for 2012 are shown in Figure S2. Effluent aqueous chloride concentrations

211

increased over the growing season in the unplanted bed but not in the planted beds.

212

The effluent chloride concentrations from June 2011 through June 2012 are shown in Figure S3. The

213

chloride concentration in the effluent from the r2E1 planted bed increased most during the winter

214

season, compared to the WT and unplanted beds, suggesting a higher rate of accumulation of chloride

215

in bed soil in the previous growing season. This is consistent with soil chloride measurements of the

216

unsaturated layers of the three beds (Figure S5).

217

Soil TCE and Chloride.



218 219 220

The season average concentrations of soil TCE were 0.77 ± 0.27 mg kg-1, 0.68 ± 0.25 mg kg-1, and 0.18 ± 0.03 mg kg-1 for the r2E1 bed, WT bed, and unplanted bed, respectively (Figures S7-S8). Soil chloride increased in the vadose zone of the planted beds (less than 100 cm deep) during the

221

growing season (Figures S9-S11). A similar increase was not observed in the unplanted bed. On May 22,

222

2012, chloride concentration at 30 cm was 3.19 ± 1.17, 2.64 ± 0.50, and 1.64 ± 0.33 mg kg-1 for the r2E1

223

bed, WT bed, and unplanted bed, respectively. By October 16, 2012, concentrations at the 30 cm depth

224

had increased 6-fold from the beginning of the season in the r2E1 bed and 8-fold in the WT bed to 19.8

225

± 17.2 and 21.3 ± 30.3 mg kg-1, respectively. The increase in soil chloride in the vadose zone of the

226

planted beds occurred in each growing season in this study and was also observed other

227

phytoremediation field studies with poplar and chlorinated solvents.6, 29, 39 The accumulated chloride ion

228

in the vadose zone was mobilized in the winter season due to heavy winter rain and possibly root

229

decomposition (Figure S3).

230

TCE and Metabolites in Plant Tissue.

231

The leaves, stems, and branches of the r2E1 poplars had significantly higher concentrations of TCOH

232

and TCOH-glucoside than the WT tissues and significantly lower concentrations of TCE for all tissues

233

(P

A Field Trial of TCE Phytoremediation by Genetically Modified Poplars

Recommend Documents