Mesocosm Studies on the Efficacy of Bioamended Activated Carbon

Subscriber access provided by UNIVERSITY OF ADELAIDE LIBRARIES

Article

Mesocosm Studies on the Efficacy of Bioamended Activated Carbon for Treating PCB-Impacted Sediment Rayford B. Payne, Upal Ghosh, Harold Douglas May, Christopher W. Marshall, and Kevin R Sowers Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b01935 • Publication Date (Web): 15 Aug 2017 Downloaded from http://pubs.acs.org on August 16, 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 37

Environmental Science & Technology

1 2 3 4

Mesocosm Studies on the Efficacy of Bioamended Activated Carbon

5

for Treating PCB-Impacted Sediment

6 7

Rayford B Payne†, Upal Ghosh‡, Harold D May§, Christopher W. Marshallǁ⊥, and Kevin R

8

Sowers†*

9 10

†

Department of Marine Biotechnology, Institute of Marine and Environmental Technology,

11

University of Maryland Baltimore County, Baltimore MD ‡

12

Department of Chemical, Biochemical, and Environmental Engineering, University of

13 14

Maryland Baltimore County, Baltimore MD §

Marine Biomedicine and Environmental Science Center, Department of Microbiology and

15

Immunology, Medical University of South Carolina, Charleston SC ǁ

16 17 18 19 20 21 22 23

Biosciences Division, Argonne National Laboratory, Argonne, Illinois

⊥

Current Address: Department of Microbiology and Molecular Genetics, University of Pittsburgh, Pittsburgh, PA

* Corresponding author: Kevin Sowers, Department of Marine Biotechnology, Institute of Marine and Environmental Technology, 701 E. Pratt St., Baltimore, Maryland 21202 Telephone: (410) 234-8878/FAX: (410) 234-8896, e-mail: [email protected]

24

1

ACS Paragon Plus Environment


25

ABSTRACT

26

This report describes results of a bench-scale treatability study to evaluate the efficacy of

27

bioaugmentation with bioamended activated carbon (AC) for in-situ treatment of polychlorinated

28

biphenyl (PCB) impacted sediments.

29

microorganisms to degrade and reduce the overall concentration of PCBs in sediment was

30

determined in 2 L recirculating mesocosms designed to simulate conditions in Abraham’s Creek

31

in Quantico, Virginia. Ten sediment mesocosms were tested for the effects of AC alone, AC

32

with slow release electron donor (cellulose) and different concentrations and combinations of

33

PCB dehalogenating and degrading microorganisms added as bioamendments. A 78% reduction

34

of total PCBs was observed using a cell titer of 5 × 105 Dehalobium chlorocoercia and

35

Paraburkholderia xenovorans cells g-1 sediment with 1.5% AC as a delivery system. Levels of

36

both higher and lower chlorinated congeners were reduced throughout the sediment column

37

indicating that both anaerobic reductive dechlorination and aerobic degradation occurred

38

concurrently.

39

bioaugmented treatments. Toxicity associated with co-planar PCBs was reduced by 90% after

40

treatment based on toxic equivalency of dioxin-like congeners. These results suggest that an in

41

situ treatment employing the simultaneous application of anaerobic and aerobic microorganisms

42

on AC could be an effective, environmentally sustainable strategy to reduce PCB levels in

43

contaminated sediment.

To this end, the ability of PCB transforming

Porewater concentrations of all PCB homologs were reduced 94-97% for

44

2


Page 2 of 37

Page 3 of 37

45


INTRODUCTION

46

Polychlorinated biphenyls (PCBs) were used widely in transformers, capacitors and other

47

electrical applications for over five decades because of their superior flame resistance and

48

insulating properties 1. Other common uses included fluorescent light ballasts, inks, paints,

49

hydraulic oils, adhesives and carbonless paper. The manufacture of PCBs was banned in the

50

United States in 1979 and worldwide in 2001.

51

environment during their widespread use, PCBs continue to pose a health risk because their

52

propensity for bioaccumulation in food chains and potential toxicity as a result of consumption

53

by animals and humans 1.

However, as a result of release into the

54

The commonly accepted methods for treatment of sediments impacted with PCBs are

55

dredging and disposal or capping. Dredging is effective for reducing PCB levels in sediments,

56

but it is costly, disruptive to the ecosystems and increases the potential risk if PCBs are released

57

into the water 2. Capping with passive materials such as sand is an effective treatment approach

58

for containment of PCBs in sediment, but because it is subject to potential abiotic and biotic

59

disruption it does not completely eliminate the risk of later exposure and can be disruptive to the

60

existing ecosystem 3. A recently developed method that is gaining acceptance is the application

61

of amendments such as activated carbon (AC) to PCB impacted sediments to sequester PCBs 4.

62

By sequestering soluble PCBs, AC effectively decreases the bioavailability of PCBs thereby

63

minimizing the risk of exposure to the food chain 5. However, none of these methods has the

64

potential to accelerate the degradation of PCBs beyond the rates observed for natural attenuation.

65

Biological degradation of PCBs in the environment occurs by anaerobic dechlorination of

66

highly chlorinated congeners followed by the aerobic degradation of the dechlorination products.

67

The primary rate-limiting factor is the low native abundance of PCB dechlorinators in sediments

3



Page 4 of 37

68

6

69

PCB impacted sediment. For example, sequential treatment of PCB impacted sediment in an

70

anaerobic PCB halorespiring enrichment followed by transfer in an aerobic culture containing

71

Paraburkholderia xenovorans LB400 was reported to degrade weathered Aroclor 1248 and

72

Aroclor 1260 by as much a 70% and 67%, respectively

73

demonstrated the complementary role of anaerobic dechlorination and aerobic degradation in

74

PCB degradation, they were conducted in closed microcosms with controlled redox conditions.

75

In addition, delivery of microorganisms through a water column into sediments at field scale is a

76

challenge. A promising approach is to deliver the microorganisms on solid particles that could

77

integrate into the sediment environment.

78

extensive dechlorination in Aroclor 1260 enriched sediment microcosms amended with AC.

79

Payne et al. 10-11 applied granular activated carbon in the form of SediMite™ 4, a pelletized form

80

of AC, inoculated with anaerobic halorespiring and aerobic degrading bacteria in static estuarine

81

Baltimore Harbor sediment mesocosms and observed over 75% reduction in weathered Aroclor

82

1260 levels in only 120 days.

83

bioamendments, the report demonstrated that AC in principle could be used as a delivery system

84

for PCB degrading microorganisms that would form a solid substrate for active microbial

85

transformation. However, the study only tested two microorganisms at a single cell titer.

86

. Thus, enhancing this natural process with bioaugmentation is a potential treatment strategy for

7-8

. However, although both studies

Kjellerup et al.

9

reported a shift towards more

In addition to enhanced degradation of PCB levels by the

A key aspect of planning a bioremediation field application is determining the optimal

87

loading rate of treatment material into the sediment.

88

parameters for optimal efficacy of bioaugmentation using pelletized AC as a potential carrier to

89

deliver PCB degrading microorganisms to PCB impacted sediments.

90

conducted in mesocosms containing PCB contaminated sediment from a watershed drainage

4

In this study, we test and identify


Experiments were

Page 5 of 37


91

pond adjacent to Chopawamsic Creek, a tributary of the Potomac River. The study employs

92

static sediment mesocosms with recirculating overlying water to simulate in situ redox

93

conditions by maintaining a dissolved oxygen concentration in the water column similar to that

94

in the water column at the site. Parameters tested include effects of organic carbon,

95

bioamendment titer, and different combinations of microorganisms on the total and porewater

96

concentration of PCBs in sediments over the course of 375 days.

97

5



98

MATERIALS AND METHODS

99

Sediment collection and homogenization. PCB impacted sediment (25 L) was collected on 18

100

October, 2012 from Abraham’s Creek in Quantico, VA with a grab sampler at 39°16.8 N,

101

76°36.2 W and transported to the lab in four sealed 20 L polypropylene buckets. Water for the

102

treatability study was collected from the site in a 20 L carboy. Sediment collected from the site

103

was transferred to a 100 L basin in an anaerobic glove bag under a nitrogen atmosphere and the

104

pooled sediment was thoroughly mixed with a 10 cm mud mixer (TBC Inc., Long Beach, CA)

105

mounted on a power drill. Sediment was stored in the dark at 4°C prior to use.

106 107

Preparation of bioamended AC. Halorespiring (PCB dehalogenating) bacteria were grown

108

anaerobically in estuarine mineral medium (ECl)

109

with the following electron donors and acceptors, respectively: “Dehalobium chlorocoercia”

110

DF-1 12 with sodium formate (10 mM) and PCB 61 (2,3,4,5-tetrachlorobiphenyl; 173 µM), strain

111

o-17

112

Chloroflexus phylotypes SF1 and DEH10 14 with a fatty acid mixture containing sodium salts of

113

acetate, propionate and butyrate (2.5 mM each) and Aroclor 1260 (100 mg L-1). Cultures were

114

grown in 160-ml serum bottles containing 50 ml of medium sealed under N2-CO2 (4:1) with 20-

115

mm Teflon-coated butyl stoppers (West Pharmaceutical, Inc., Exton, PA) and incubated

116

statically at 30°C in the dark. For preparation of the bioamendment, halorespiring bacteria were

117

twice sub-cultured 1:10 in 500 mL medium with 100 µM PCE substituted for PCB as the

118

electron acceptor to remove any residual PCBs. The halorespiring cultures were harvested

119

anaerobically under N2-CO2 (4:1) by centrifugation in 250 mL Oakridge bottles 15. Cell pellets

13

12

. Halorespiring cultures were maintained

with sodium acetate (10 mM) and PCB 65 (2,3,5,6-tetrachlorobiphenyl; 173 µM) and

6


Page 6 of 37

Page 7 of 37


120

were resuspended in 100 mL of ECl medium without PCE to a final concentration of 5 × 107

121

cells mL-1. Paraburkholderia xenovorans LB400 16,17 was grown aerobically in M9 minimal medium

122 123

18

with solid biphenyl crystals (5 mM; solubility in water, 2.89×10-2 mM) as the carbon source

124

and electron donor as described previously

125

were incubated at 30 °C with shaking at 100 rpm to an O.D.600 of 1.0 (ca. 4×108 cells mL-1),

126

harvested by centrifugation, and the cell pellet was suspended in 100 mL of sterile M9 medium

127

without biphenyl.

19

. Cultures (100 ml) in 500 ml Erlenmeyer flasks

128

Immediately before addition to mesocosms the anaerobe (DF-1, ο-17 and/or SF1-

129

DEH10) and aerobe (LB400) were combined in a manual pump sprayer reservoir (1.75 L Flo-

130

Master 56HD, Root-Lowell Mfc Co., Lowell, MI). For each mesocosm the concentrated cultures

131

were sprayed onto 70 g of SediMite™ pellets (Sediment Solutions, Ellicott City MD), an

132

aggregate of activated carbon (AC), sand, clay binder and where indicated 1% cellulose, at

133

concentrations indicated below. Bioamended AC was immediately mixed into mesocosms as

134

indicated below with a Teflon spoon and mesocosms were allowed to settle for 4 hours before

135

baseline sampling. The final concentration of SediMite™ in sediment was 3% for a final

136

concentration of 1.5% (dry wt) AC.

137 138

Preparation of mesocosms. Aliquots of homogenized sediment (1.75 L) were transferred under

139

an atmosphere of N2 in an anaerobic glove bag to two-liter thin layer chromatography tanks (ID

140

in cm: 8W x 27H x 12D, sediment depth 6 cm, water depth 2 cm) modified as shown in Figure

141

S1. After treatments shown in Table 1, the tanks were sealed with glass plates aerated water

142

collected from the site was continuously circulated with a 12-channel peristaltic pump (Watson-

7



143

Marlow) at a hydraulic retention time of one hour. The aerated water entered the tanks through a

144

stainless steel manifold to create a linear flow over the surface of the sediment, thereby

145

minimizing the risk of PCB loss due to volatilization and sediment turbation into the water

146

column Aerated water was pumped from a sparging flask containing a fritted glass gassing tube

147

connected to an air supply. The dissolved oxygen concentration in the water column, measured

148

with a polarographic electrode (Mettler Toledo, Columbus OH), was maintained at 6.7±0.3 mg

149

kg-1 at the mesocosm inlets and 6.5±0.3 mg kg-1 at the outlets, which is equivalent to 6.75 mg kg-

150

1

151

Amberlite XAD-2 (Rohm & Haas Co, Newark, DE) resin in the base of the sparging flask to

152

remove soluble PCBs prior to aeration. Mesocosms were operated at a temperature of 22-24oC

153

in the dark.

reported at the site in June 2012. Water flowing out of the mesocosm was passed through

154 155

Sediment analysis. Sediment TOC analysis was performed as described by Grossman and

156

Ghosh

157

5000A) and non-dispersive infrared gas analyzer as recommended by the manufacturer.

20

using a Shimadzu TOC analyzer with a solids sample module (TOC-5000A and SSM-

158 159

PCB extraction and analysis of sediment. Sediment mesocosms were sampled in triplicate

160

using a sterile 5 mL syringe barrel as a coring device. A numbered grid and random number

161

generator were used to ensure that samples were collected from random locations.

162

Approximately 5g wet weight sediment was dried with pelletized diatomaceous earth (Dionex,

163

Sunnyvale, CA) in a desiccator at room temperature. The dried sediment (1 g) was extracted

164

with an Accelerated Solvent Extractor (Dionex) following EPA method 3545 as described

165

previously 10. PCB 166 (10 µl stock of 400 µg L-1 hexane) was added as a surrogate to correct

8


Page 8 of 37

Page 9 of 37


166

for extraction efficiency. PCBs 30 and 204 (400 µg L-1 each in 10 µl acetone) were added as

167

internal standards.

168

PCB congeners were analyzed using a Hewlett-Packard 6890 series II gas chromatograph

169

(GC) with a DB-1 capillary column (60 m × 0.25 mm × 0.25 µm; JW Scientific, Folsom, CA)

170

and a

171

congeners resolved in 130 individual peaks

172

resolved and analyzed using a Hewlett-Packard 5890 series II gas chromatograph (GC) with a

173

HT8 8% phenyl polycarborane-siloxane capillary column (60 m × 0.22 mm × 0.25 µm; SGE)

174

and a 63Ni electron capture detector as described previously 21. PCB congeners 77, 81, 105, 114,

175

118, 123, 126, 169, 156, 157, 167 and 189 were quantified with a 10-point calibration curve (2-

176

800 µg L-1) using PCBs 30 and 204 as internal standards.

63

Ni electron capture detector by a modified method of EPA 8082, which detected 173 10

. Co-planar dioxin-like PCB congeners were

177 178

PCB extraction and analysis of passive samplers and XAD resin. Polyoxymethylene (POM,

179

77 µM; CS Hyde Co., Lake Villa, IL) membranes cut into 1 x 7 cm strips were pre-cleaned with

180

hexane followed by methanol, then secured in stainless steel screens with aluminum staples.

181

Three POM samplers were inserted vertically in the sediment column of each mesocosm. The

182

samplers were removed from the mesocosms 120 and 375 days after treatments, rinsed with

183

water to remove sediment, and analyzed for PCBs as described in Beckingham et al.

184

concentrations in POM were converted to estimated PCB concentrations in the porewater phase

185

based on equilibrium partitioning constants 23. Performance reference compounds were not used.

186

Amberlite XAD-2 resin (Rhom & Haas, Philadelphia, PA) was pre-cleaned with hexane

187

followed by methanol prior to use and extracted for PCB analysis after 375 days as described

188

above for the passive samplers.

9


22

. PCB


Page 10 of 37

189 190

DNA extraction and enumeration of bacterial bioamendments. DNA was extracted by

191

adding 0.25 g sediment (wet wt.) from each sample core to a PowerBead microcentrifuge tube

192

(Power Soil DNA Isolation Kit, MOBIO Laboratories, Inc., Carlsbad, CA) as previously

193

described 10. Extracted DNA samples had an A260/280 ratio of ≥ 1.6 and an A260/230 ratio of ≥

194

2.0. Enumeration of microorganisms in each subcore was performed by real-time quantitative

195

PCR (qPCR) using iQ SYBR Green Supermix (Bio-Rad Laboratories, Hercules, CA). Primers

196

included CIOP0/CIOP1 specific for the bphA gene operon of Paraburkholderia xenovorans

197

LB400 and SKFPat9F/SKFPat9R specific for a putative reductive dehalogenase of “Dehalobium

198

chlorocoercia” DF-1 as previously described

199

enumerated by qPCR using primers Chl348F/Dehal844R specific for genes encoding 16S rRNA

200

in the halorespiring chloroflexi

201

with R2 = 0.999.

202

(CIOP0/CIOP1), 1.3 (SKFPat9F/SKFPat9R) and 2.9 (Chl348F/Dehal844R).

25

10-11, 24

.

All other halorespiring strains were

. Amplification efficiencies of standards were >89.0±9.0%

The linear range was 0.1 to 1x10-6 ng and the y-intercept was 1.2

203 204

Analysis of the microbial community. DNA was prepared for amplification using the 5PRIME

205

MasterMix (5 PRIME, Inc, Gaithersburg, MD) and 0.2 µM 515F/806R primers

206

amplified using the following conditions: 3 minutes at 94°C, 35 cycles of 94°C for 45 seconds,

207

50°C for 1 minute, and 72°C for 1.5 minutes, with a final extension for 10 min at 72°C.

208

Amplicons were then pooled to 70 ng DNA per sample, and a clean pool was then generated

209

using the QIAquick PCR Purification Kit (QIAGEN, Hilden, Germany). DNA was prepared for

210

amplification using the 5PRIME MasterMix (5 PRIME, Inc.) and 0.2 µM of 515F/806R primers

211

targeting the V4 region of the 16S rRNA gene. Clean pools were sequenced on an Illumina

10


26

. DNA was

Page 11 of 37


212

MiSeq using V3 chemistry to obtain 2x150 base pair reads. The average sequencing coverage

213

was over 17,000 sequences per sample.

214

Materials.

Community analyses are described in Supplemental

215 216

Data availability. Study: PRJNA355587 (SRP095095), sample: PCB mesocosms

217

(SRS1859570), experiment: mesocosm_seqs (SRX2422444), run: mesocosm_seqs.fastq.gz

218

(SRR5110056).

219 220

11



221

Page 12 of 37

RESULTS and DISCUSSION

222 223

Effect of treatments on reductive dechlorination and degradation of PCBs. Water content of

224

the homogenized sediment was 74% ± 6 and total organic carbon (TOC) was 6.7% ± 0.5. Total

225

PCB concentration in the pooled sediment samples was 3.40 ± 0.50 mg kg-1 with a mean of 3.5 ±

226

0.1 chlorines per biphenyl. Additional characterization of sediment is provided in Supplemental

227

Materials. The effect of treatments on PCB levels in Abraham’s Creek sediment is shown in

228

Figure 1 and Table 2. The results in non-bioamended treatments 1, 2 and 3 indicate that there

229

was no significant (ρ>0.05) biostimulation of indigenous bacterial populations or abiotic loss as

230

a result of mixing, addition of AC, or addition of cellulose as a carbon source. This observation

231

indicates that biostimulation of indigenous halorespiring and degrading PCB microorganisms

232

would not be effective for reduction of PCB levels in Abraham’s Creek sediment at rates greater

233

than those observed for natural attenuation.

234

All bioamended treatments showed significant degradation with the exception of

235

treatments 4 and 9. A titer of 103 cells g-1 showed no significant effect on PCB concentration

236

over the course of 375 days, whereas bioamended treatments 5 and 6 showed significant

237

degradation, activity, 58 and 78%, respectively. In treatment 6 most of the degradation, 55%,

238

was observed in the first 30 days and maximum degradation of 78% was observed in treatment 6

239

after 120 days. This value is similar to the maximum value of 80% total PCB reduction observed

240

365 days after bioamendment of weathered Baltimore Harbor sediment mesocosms reported by

241

Payne et al. 11. In contrast, degradation in treatment 5, 7, 8, 9 and 10 continued after 120 days.

242

The addition of cellulose as a slow release carbon source in combination with

243

bioamendments appeared to increase the extent of PCB reduction. However, the difference in

12


Page 13 of 37


244

PCB levels after 376 days in treatment 6, which contained 105 cells g-1 of DF-1 and LB400 with

245

cellulose, and treatment 7, which contained the same bioamendment titer without cellulose was

246

not significant (ρ>0.05). The results suggest that digestible organic carbon was sufficiently

247

available in the sediments.

248

Nona- through hexa-chlorobiphenyls were reduced by 80% in treatment 6 to a final total

249

concentration of 0.12 mg kg-1 after 375 days (Figure S2). Since these homolog groups are not

250

attacked or poorly attacked by LB400 27, most of the reduction in concentration is attributable to

251

reductive dechlorination by DF-1 and possibly by indigenous halorespirers

252

homologs accounted for only 22% of the total weathered PCBs in Abrahams Creek sediment,

253

bioaugmentation with a halorespirer such as DF-1 would have a critical role in less weathered

254

sediment where nona- to hexa-chlorobiphenyls can account for an average 88% of the total

255

congener mass in Aroclor 1260 28. Pentachlorobiphenyls 146, 151 and 153, which accounted for

256

5.2% of the total mean PCB concentration in the sediment, were degraded by 86, 87 and 89%,

257

respectively, in treatment 6. However, these congeners are not dechlorinated by DF-1

258

poorly degraded by LB400

259

bioamendment with DF-1, a phenomenon that has been reported previously, 10,11,12 indicates that

260

indigenous halorespiring or degrading microorganisms were active in the mesocosm, possibly

261

stimulated by dechlorination products of DF-1.

27

.

12

. Although these

12

and

The transformation of non-substrate congeners after

262

Halorespiring bacteria with different substrate specificities were added to mesocosms 8

263

through 10 to test their effectiveness for reducing total PCB concentrations as co-amendments

264

with LB400. The extent of degradation with addition of o-17 in treatment 8, which preferentially

265

attacks congeners in single flanked ortho substituted and double flanked meta chlorine

266

substituted positions, and with DEH10 and SF-1 in treatment 9, which preferentially attack

13



Page 14 of 37

267

congeners in single and double flanked meta chlorine substituted positions, was less than that

268

observed with a similar cell concentration of DF-1 in treatment 6. Although DF-1 is limited to

269

reduction of doubly flanked chlorines, the dechlorination patterns were similar between the

270

amendments.

271

chlorobiphenyls contained one or more doubly flanked chlorines

272

attacked prior to singly flanked chlorines by both SF1/DEH10 and o-17

273

in few observed differences between dechlorination patterns of higher chlorinated congeners by

274

DF-1 and the other halorespiring bacteria. Differences in the dehalogenation patterns by the

275

different halorespiring strains would also be masked by the relatively rapid degradation rates of

276

aerobic degradation

277

differences would likely have been observed as distinct singly flanked chlorines would have

278

been available and targeted as PCBs were dechlorinated to congeners with six or less chlorines.

279

Interestingly, total PCB degradation by the combination of all three halorespiring

280

bioamendments appeared to be slightly less than bioaugmentation with only DF-1 and LB400 in

281

treatments 5 and 6, although this difference was not significant (ρ>0.05). The results indicate

282

that addition of 105 cells g-1 each of DF-1 and LB400 was the most robust treatment for reducing

283

total PCB levels. This observation along with the fact that expected dechlorinated intermediates

284

did not accumulate confirm that the total reduction in PCB concentration in the mesocosms was

285

the result of anaerobic halorespiration and aerobic degradation occurring concurrently.

One explanation for this observation is that most of the hexa- to nona-

27, 30

, compared with halorespiration

14,10

.

28

, which are preferentially 14,29

. This would result

In the absence of LB400,

286 287

PCB degradation throughout depth profile. The PCB concentrations were not significantly

288

different (ρ>0.05) for the upper and lower 3 cm of the sediment column in any of the mesocosms

289

(Figure S3). The results indicate that bioaugmentation was effective throughout the 6 cm

14


Page 15 of 37


290

sediment column. PCB degradation by LB400 requires oxygen for dioxygenase mediated ring

291

cleavage

292

water column, discoloration of the sediment by oxidation was only obvious in the top 0.3-0.5 cm.

293

Below the oxidized layer the sediments remained black, likely due to reduced conditions

294

maintained by anaerobic decay of native and added organic matter. Benthic activity due to

295

worms was observed throughout the entire sediment depth in treatments 5 and 6 for the first four

296

months, which also had the greatest extent of degradation. Based on these results it is not

297

possible to draw any conclusion on the role of bioturbation for oxygenation of the lower

298

sediment profile. However, the results indicate that even in mesocosms where benthic activity

299

was not observed there was sufficient diffusion of oxygen through the porewater to support

300

aerobic degradation. The similarity of the homolog patterns between the top and bottom cores

301

after 120 days further confirms that both anaerobic dechlorination and aerobic degradation

302

occurred throughout the sediment column in treatments 5 to 10.

31

. Although dissolved oxygen levels were maintained at 6.7 mg L-1 throughout the

303 304

Effects of treatments on toxicity of PCBs. Most of the toxic effects of PCBs for humans are

305

mediated through the aryl hydrocarbon receptor (AhR), a cytosolic receptor protein present in

306

most vertebrate tissues with high affinity for 2,3,7,8-substituted PCDD/Fs and some coplanar

307

PCB congeners

308

2,3,4,4’5-pentachlorobiphenyl (PCB 114), 2,3,3’,4,4’,5-hexachlorobiphenyl (PCB156) and

309

2,3,3’,4,4’,5’-hexachlorobiphenyl (PCB 157) (Figure 2). The total concentration of these three

310

congeners on day 0, 7.40 ng g-1 dw, was reduced 90% to 0.75 ng g-1 dw 375 days after treatment.

311

Factoring in the toxic equivalency factor (TEF) for each coplanar congener relative to 2,3,7,8-

312

tetra dibenzo-p-dioxin 33, the total toxic equivalency (TEQ) was reduced from 0.22 to 0.02 ρg g-

32

. Three coplanar congeners were detected in Abraham’s Creek sediment:

15



Page 16 of 37

313

1

314

estimating relative changes in potential exposure to dioxin-like chemicals from consumption of

315

aquatic food products as this approach does not take into account a number of factors such as

316

inhibition of TCDD toxic effects by other congeners in PCB mixtures

317

bioaccumulation within the food chain such as partitioning coefficient 35.

. TEQ methodology in human risk assessment in the context of this study is only intended for

34

and factors affecting

318 319

Effect of treatments on PCBs in porewater. Changes in PCB concentrations in porewater

320

were measured by passive equilibrium sampling 375 days after treatments Figure 3. There was

321

a 30-35% reduction in PCB porewater concentrations after treatment with non-bioamended AC

322

in treatments 2 and 3 compared with untreated sediment.

323

concentration of PCBs with abiotic AC is less compared with previous reports of >95%

324

reduction of porewater PCBs with AC amendments 36. The difference can be largely attributed

325

to the low AC dose in the present study (1.5%) compared to much higher doses of AC in

326

previous work (>2.5%). The AC dose was kept low in the current study to focus primarily on

327

the effect of bioamendment.

328

bioamended with 103 (treatment 4), but the difference was not significant (ρ> 0.05) from the

329

abiotic AC treatments 2 and 3. All of the remaining bioamended treatments showed a significant

330

reduction of PCBs in the porewater compared with untreated sediment (ρ< 0.05) ranging from

331

94% in treatment 10 to the greatest reduction, 97%, in treatment 6. Thus, the overall reductions

332

in the freely dissolved porewater concentrations were larger than the corresponding reductions in

333

the total PCBs in sediments indicating that the more soluble dechlorination products were not

334

accumulating, but were further degraded by the aerobes. Presence of 1.5% AC was also

The reductions in porewater

Porewater PCB was reduced 40% after treatment with AC

16


Page 17 of 37


335

contributing to some of the reductions in porewater concentration beyond mass reduction by

336

degradation.

337

PCBs were not detected in the XAD resin indicating the reduction in PCB levels was not

338

attributable to abiotic loss in the water column. Coplanar PCB congeners levels in the overlying

339

water phase were below the detection limit of 0.01 ng L-1 in all of the mesocosms, including the

340

untreated mesocosm. The results indicate that the addition of bioamendment was effective at

341

reducing freely dissolved PCBs levels in sediment porewater, thereby reducing the potential for

342

PCB flux into the overlying water and bioaccumulation in the aquatic food web including fish.

343 344

Sustainability of the bioamendment. DF-1 and LB400 in treatment 6 were detected throughout

345

the experiment, but their titer decreased approximately 2-3 orders of magnitude after 375 days

346

(Figure 4), whereas the number of DF-1 and LB400 gene copies g-1 in treatments in non-

347

bioaugmented treatments was below the theoretical detection limit of 102 gene copies g-1

348

sediment. During the first sixty days when the cell titer was highest, there was a decrease in

349

PCBs, but no significant increase in intermediate dechlorination products. Some dechlorination

350

products were observed to accumulate on days 120 and 375, which coincides with two orders of

351

magnitude decrease in the titer of LB400 (Figure S4). Overall, the results suggest that the

352

aerobic degradation rate by LB400 was greater than the halorespiration rate of DF-1 in the first

353

two months, but as the titer of LB400 decreased the net rate of degradation no longer exceeded

354

that of anaerobic dechlorination, resulting in accumulation of some dechlorinated congeners.

355

This observation is not surprising as degradation of halogenated biphenyls is co-metabolic and

356

will not support growth 16. The titer of DF-1 also decreased over the course of 375 days, but at a

357

slower rate than LB400. Although halorespiration of PCBs supports growth,14, 12 Lombard et al.

17



358

37

359

relevant soluble PCB concentrations is too low to maintain a large population of bacteria. In the

360

present study this limitation was overcome by introducing a high cell titer sufficient to reduce the

361

bioavailable portion of PCBs without growth.

Page 18 of 37

calculated that the thermodynamic cell yield of halorespiring bacteria at environmentally

362

In a prior study, activity in Baltimore Harbor sediment mesocosms bioaugmented with

363

DF-1 and LB400 degradation was not detected after 80% of total PCBs were removed, although

364

the remaining congeners were potential substrates for PCB dehalogenation or degradation. The

365

report suggested that activity stopped because PCBs were no longer bioavailable and/or oxygen

366

became limiting in the static mesocosms. However, in this study the dissolved oxygen in the

367

water column was maintained at above 6% throughout the 375 days incubation period and there

368

were no significant difference in the degradation patterns in the sediment column, which

369

suggests that oxygen was likely not the limiting factor. Another possibility for the observed limit

370

of degradation at 78% of the original PCB concentration in treatment 6 was the reduction in cell

371

titer over time. However, repeating the treatment of mesocosm 6 after 375 days (105 cells g-1

372

each of DF-1 and LB400 applied with AC and 0.03% cellulose) did not stimulate further

373

degradation of total PCBs after 79 days, which indicates that the decline in activity was not due

374

to a decrease in the titer of an active PCB degrading population or TOC. This observation

375

combined with the low concentration of soluble PCBs (97% reduction) suggests that activity was

376

inhibited by low bioavailability of the remaining PCBs. The total PCBs remaining are likely

377

tightly bound to the AC and organic fraction in the sediment, and are below the threshold for

378

uptake by the bioamendments.

379

18


Page 19 of 37


380

Effect of treatments on the indigenous microbial population. Overall, the microbial diversity

381

significantly decreased over the course of the experiment (paired t-test, ρ

Mesocosm Studies on the Efficacy of Bioamended Activated Carbon

Recommend Documents