Survival of Vinyl Chloride Respiring Dehalococcoides mccartyi under

Subscriber access provided by Fudan University

Article

Survival of Vinyl Chloride Respiring Dehalococcoides mccartyi Under Long-Term Electron Donor Limitation Koshlan Mayer-Blackwell, Mohammad F Azizian, Jennifer K. Green, Alfred M. Spormann, and Lewis Semprini Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b05050 • Publication Date (Web): 21 Dec 2016 Downloaded from http://pubs.acs.org on January 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.

Environmental Science & Technology

Biodegradation Rate

Page 1 of 24

7-Year Electron Donor Limitation

TCE VC

3-Month Electron Donor Stimulation

TCE

TCE VC

Dehalococcoides mccartyi sp. Population Fraction

VC

D.mccartyi (tceA)

1:1 10:1 D.mccartyi (vcrA)

100:1 2009

2010

2011

2012

2013

ACS Paragon Plus Environment

2014

2015


1

Title: Survival of Vinyl Chloride Respiring Dehalococcoides mccartyi Under Long-Term

2

Electron Donor Limitation

3

Authors: Koshlan Mayer-Blackwell†, Mohammad F. Azizian‡, Jennifer K. Green‡, Alfred M.

4

Spormann †§⊥, Lewis Semprini‡*

5 6

†Civil and Environmental Engineering, §Geological and Environmental Sciences, and

7

⊥Chemical Engineering, Stanford University, Stanford, California 94305, United States

8

‡Chemical, Biological and Environmental Engineering, Oregon State University, Corvallis,

9

Oregon 97331, United States

10

*Corresponding Author

11


Page 2 of 24

Page 3 of 24

12 13


Abstract: In anoxic groundwater aquifers, the long-term survival of Dehalococcoides mccartyi

14

populations expressing the gene vcrA (or bvcA) encoding reductive vinyl chloride dehalogenases

15

are important to achieve complete dechlorination of tetrachloroethene (PCE) and trichloroethene

16

(TCE) to non-chlorinated ethene. The absence or inactivity of vcrA-containing Dehalococcoides

17

results in the accumulation of the harmful chlorinated intermediates dichloroethene (DCE) and

18

vinyl chloride (VC). While vcrA-containing Dehalococcoides subpopulations depend on

19

synergistic interaction with other organohalide-respiring populations generating their metabolic

20

electron acceptors (DCE and VC), their survival requires successful competition for electron

21

donor within the entire organohalide-respiring microbial community. To understand this dualism

22

of synergy and competition under growth conditions relevant in contaminated aquifers, we

23

investigated Dehalococcoides-level population structure when subjected to a change in ratio of

24

electron donor to chlorinated electron acceptor in continuously stirred reactors (CSTRs) operated

25

over seven years. When the electron donor formate was supplied in stoichiometric excess to

26

TCE, both tceA-containing and vcrA-containing Dehalococcoides populations persisted, and

27

near-complete dechlorination to ethene was stably maintained. When the electron donor formate

28

was supplied at sub-stoichiometric concentrations, the interactions between tceA-containing and

29

vcrA-containing populations shifted towards direct competition for the same limiting catabolic

30

electron donor substrate with subsequent niche exclusion of the vcrA-containing population.

31

After more than 2000 days of operation under electron donor limitation, increasing the electron

32

donor to TCE ratio facilitated a recovery of the vcrA-containing Dehalococoides population to its

33

original frequency. We demonstrate that electron donor scarcity alone, in the absence of

34

competing metabolic processes or inhibitory dechlorination intermediate products, is sufficient to

35

alter Dehalococcoides population structure. These results underscore the importance of electron

36

donor/chloroethene stoichiometry in maintaining balanced functional performance within

37

consortia comprised of multiple Dehalococcoides mccartyi subpopulations, even when other

38

competing electron acceptor processes are absent.

39



40 41

1. Introduction

42

In situ bioremediation using organohalide-respiring bacteria is a viable and often cost-

43

effective strategy for the removal of chlorinated solvents – such as trichloroethene (TCE) and

44

tetrachloroethene (PCE) – from anoxic contaminated groundwater.1 Complete reductive

45

dehalogenation of PCE and TCE to ethene occurs in a stepwise fashion via dichloroethene

46

(DCE) and vinyl chloride (VC) as intermediates. There is apparent metabolic specialization

47

among organohalide-respiring microorganisms and Dehalococcoides mccartyi strains to catalyze

48

specific dehalogenation steps in the overall chloroethene dehalogenation pathway. Thus, the

49

appropriate assembly of microorganisms – collectively containing the full inventory of genes

50

necessary for the complete dechlorination of PCE or TCE – is a prerequisite for preventing the in

51

situ buildup of toxic intermediates DCE or VC.

52

Complete dechlorination of PCE or TCE in situ depends not only on the initial bacterial

53

community composition, but also on the long-term survival of obligate organohalide-respiring

54

Dehalococcoides mccartyi sp. expressing the genes encoding VcrA (or BvcA) reductive

55

dehalogenases catalyzing VC reduction to ethene.2-7 Based on evidence from isolate studies,

56

vcrA-containing Dehalococcoides are niche-specialized and typically lack those reductive

57

dehalogenases required to efficiently dechlorinate PCE or TCE. Therefore, while vcrA-

58

containing Dehalococcoides populations depend on other organohalide-respiring populations for

59

a supply of metabolic electron acceptors (DCE and VC), their survival involves successful

60

competition, within the entire organohalide-respiring population, for molecular hydrogen as an

61

electron donor. Studying such competition among Dehalococcoides subpopulations in batch

62

growth experiments has limitations, as these growth conditions do not represent the persistent

63

slow flux of electron donor that may limit bacteria within subsurface plume environments.

64

We, therefore, investigated Dehalococcoides level population structure and dynamics in

65

continuously stirred reactors (CSTRs), operated as chemostats, as we varied the ratio of electron

66

donor to chlorinated electron acceptor. A chemostat is a continuous-growth laboratory model

67

system that enables direct competition among microorganism at low substrate concentrations. It

68

captures an essential feature of plume conditions where microorganisms may experience low but

69

steady substrate concentrations and prolonged electron donor limitation. However, mineral


Page 4 of 24

Page 5 of 24


70

surfaces for attached microbial growth, which may also influence population structure in

71

subsurface habitats, are missing from a chemostat. Despite this limitation, the chemostat

72

uniquely permits precise control of growth conditions and reproducible non-destructive sampling

73

of a microbial community for molecular analysis and kinetic batch testing. For this experiment,

74

we operated four CSTRs, as chemostats, over a period of seven years at a mean residence time of

75

50 days.

76

We studied the outcome of competition for electron donor in four CSTRs, catalyzing the

77

transformation of TCE to ethene (Table 1) under two growth conditions (Table 2). Formate was

78

provided as a soluble electron donor which can be converted to hydrogen via the action of

79

syntrophic bacteria in the presence of hydrogen consuming bacteria.8,9 Formate was chosen

80

based on its stoichiometric conversion to hydrogen and the production of bicarbonate, which

81

serves as a buffer for the acid released via dechlorination,10 thus providing for excellent pH

82

control. The ability to reduce the acidity produced by reductive dechlorination is especially

83

important when high concentrations of PCE and TCE, associated with DNAPL contamination,

84

are being considered.11 In a previous column study of TCE degredation – in the presence of

85

sulfate, Fe(III), and Mn(IV) – formate was found to a more effective substrate than lactate and

86

propionate.12 Thus, we selected formate as the electron donor substrate to use in our chemostat

87

studies.

88

We considered the stoichiometry of electron donor and electron acceptor on an electron

89

equivalent (eeq) basis. Under the first growth condition maintained in reactors EV5L and VS5L,

90

the concentration of the electron donor formate (45 mM formate, 90 mM eeq) was provided in

91

considerable stoichiometric excess to the concentration of available chlorinated electron

92

acceptors (9-10 mMol TCE, requiring 27-30 mM eeq for complete dehalogenation). In contrast,

93

in the second growth condition established in reactors EV2L and VS2L, the concentration of

94

formate was diminished to 25 mM (50 mM eeq), which is less electron donor required for both

95

the complete reductive dechlorination of TCE, DCE, and VC to ethene (27-39 mM eeq) and the

96

synthesis of cellular biomass (24- 25 mM eeq), given the reported Fs = 0.46 for Dehalococcoides

97

mccartyi 195.13 In remediation practice, other fermentable electron donors (e.g. emulsified

98

soybean oil or lactate) are commonly used as a means of slowly delivering hydrogen to

99

organohalide-respiring bacteria. Therefore, when these substrates are used instead of sodium



100

formate (the electron donor used in this study), stoichiometric analysis should also consider the

101

percentage of the primary electron donor converted to hydrogen and the concentration-dependent

102

microbiological diversion of hydrogen to methanogenesis, acetogenesis, and metal reduction.14

103

2. Materials and Methods

104

2.1 Reactor Operation

105

The design and operation of the CSTR system were described previously by Berggren et

106

al. (2013).15 Reactors were inoculated with mixed cultures enriched from aquifer material

107

obtained from the Evanite contaminated sites in Corvallis, OR, USA and a former DuPont

108

facility in Victoria, Texas, USA.16,17 Reactors EV5L and VS5L were maintained under formate-

109

excess conditions for their entire operation. Reactors EV2L and VS2L were initially maintained

110

under formate-excess for the first 180 days of operation before decreasing influent formate to

111

stoichiometric formate-limiting conditions for 2500 and 2200 days, respectively. Reactors EV2L

112

and VS2L were then switched back to formate-excess conditions. Chemical analysis included

113

frequent monitoring of the CAHs, ethene, hydrogen, and acetate concentrations in the reactors

114

and batch vial fluid. Details pertaining to the analytical methods are provided by Azizian et al.

115

(2008, 2010) and are summarized in the Supporting Information.12,18 During the final transition

116

of reactors EV2L and VS2L from formate-deficit to formate-excess conditions, reactor material

117

was sampled weekly to determine the ratio of Dehalococcoides mccartyi populations.

118

2.2 Molecular Analysis

119

DNA was isolated from centrifuged cell pellets (16,000 g and stored at -80 °C) by

120

methods described in Mayer-Blackwell et al. (2014).19 Universal bacterial primers, targeting the

121

V2 and V3 variable regions of the 16S rRNA gene, were selected to include Dehalococcoides

122

single nucleotide polymorphisms at this gene locus, which we amplified with sample-specific

123

molecular barcodes by the protocol reported by Mayer Blackwell et al. (2016).20 Amplicons were

124

sequenced en mass using an Illumina MiSeq DNA sequencing instrument, producing on average

125

32,000 unique reads per sample. DNA reads were denoised and the relative frequency of each

126

unique sequence variant [used here to define operational taxonomic units (OTUs)] in each

127

sample estimated using DADA2.21 The association of distinct OTUs with functional genes (i.e.

128

vcrA and tceA) was determined by time-series correlation with qPCR targets using primers

129

developed by Mayer-Blackwell et al. (2014).19


Page 6 of 24

Page 7 of 24

130 131


2.3. Kinetic Tests Consortium-level kinetic parameters were measured in cell suspensions derived from the

132

continuously operating reactors by methods described in detail by Breggren et al. (2013) and

133

Green (2016).22,23 Briefly, the consortium’s maximum degradation rates at a given point during

134

the reactors’ operation was determined in triplicate batch cell suspension assays. 125-mL clear

135

borosilicate glass media bottles (Wheaton, Millville, NJ) with chlorobutyl rubber septa screen

136

caps (Wheaton, Millville, NJ) were made anoxic in a glovebox atmosphere (95:5 N2:H2) before a

137

50-mL chemostat sample was transferred to each vial. The headspace of each sample was purged

138

of residual chlorinated aliphatic hydrocarbons (CAHs) and ethene from the parent CSTRs with

139

anoxic gas (75:25 N2:CO2) for approximately 15 minutes. Approximately 8 µmol of TCE and

140

100 µmol (eeq) of formate were added via anoxic stock solutions to each batch vial. Headspace

141

concentrations of chloroethenes and ethene were monitored over time using an HP 6890 Series

142

gas chromatograph (GC, Hewlett-Packard, Palo Alto, CA) fitted with a flame ionization detector

143

(FID) utilizing helium as the carrier gas (15 mL/min) and a 30 m-0.53 mm GS-Q column

144

(Agilent Technologies, Santa Clara, CA). The method for determining kinetic parameters –

145

based on error-minimizing fit to a multi-Monod model considering each dechlorination step – is

146

described in detail by Bergren et al. (2013) and Green (2016).22,23

147

3 Results and Discussion

148

3.1 Long-Term Dehalococcoides-Population Structure Under Formate Excess and Formate

149

Limiting Conditions

150

We studied four TCE-dehalogenating CSTRs dominated by Dehalococcoides mccartyi

151

bacteria under two growth conditions (Table 2). One set of reactors (EV5L and VS5L) were

152

maintained under the constant electron donor excess condition (45 mM influent formate) for the

153

entire duration of the seven-year experiment, while a second set of reactors (EV2L and VS2L)

154

were exposed to prolonged electron donor limitation (25 mM influent formate). Population

155

structure, as determined and quantified at the 16S rRNA gene level, showed that two distinct

156

Dehalococcoides ribosomal sequence variants (OTUs) accounted for between 90-99% of the

157

entire bacterial population in each reactor under all conditions studied (Figure 1, Figure S1,

158

Supporting Information). Dehalococcoides mccartyi’s small (0.5 µm diameter) cell size implies

159

that its fractional representation at the DNA level is likely greater than in the corresponding



160

fraction of the total consortium’s biomass. A Desulfuromonadales OTU with high nucleotide

161

similarity to isolated members of the Geobacter genus was the third most abundant ribosomal

162

sequence variant, accounting for an overall population frequency ranging from 1 to 10 % across

163

the duration of the experiment. (The population fraction represented by Desulfuromonadales

164

OTU is shown in red in Figure S1, Supporting Information). Other OTUs including ribosomal

165

sequences associated with the bacterial classes Selenomondales and Spirochaetes were

166

consistently detected in all four reactors under both growth conditions, usually at less than 1% of

167

the overall population (Figure S1, Supporting Information). Our molecular approach, based on

168

the amplification and enumeration of bacterial 16S gene variants, does not measure the possible

169

fraction of archaea in the reactors. However, methane – the primary expected metabolic end

170

product if methanogenic archaea were abundant – was not detected in the reactors’ headspace.

171

Figure 1 presents key molecular and chemical data obtained from the four reactors

172

studied. Two reactors (EV2L and VS2L) were exposed to electron donor limiting conditions (25

173

mM influent formate) at the onset of the experiment, while the parent reactors (EV5L and VS5L)

174

were fed influent containing 45 mM formate. Figure 1A and Figure 1B show the population

175

structure response in reactors (EV2L and VS2L) exposed to electron donor limitation, which can

176

be directly compared to Figure 1E and Figure 1F showing the relatively stable Dehalococcoides

177

population structure present in the parent reactors (EV5L and VS5L) maintained under electron

178

donor excess. Figure 1C and Figure 1D, which can be compared to Figure 1G and Figure 1H,

179

show the major differences in the steady state concentrations of the hydrogen as well acetate

180

resulting from the different influent electron donor concentrations supplied to each reactor set.

181

The dashed lines in Figure 1 – in panels A, B, C, and D – show the point in the experiments at

182

which the concentration of formate in EV2L and VS2L influent was increased from 25mM to 45

183

mM. For continuity, the same dashed line is presented in Figure 2, which presents an expanded

184

view of this transition period. Figure 2 also displays the perturbation’s influence on

185

concentrations of TCE, DCE, VC and ethene in the reactors.

186

Under the formate-excess growth condition, where the flux of electron donor (45 mM, 90

187

mM eeq) exceeded the flux of chloroethene electron accepting equivalents (27-30 mM eeq), we

188

observed: (i) steady state aqueous hydrogen concentration between 20-40 nM (Figure 1G and

189

Figure 1H), (ii) persistent co-existence of OTUs representing the vcrA-containing and tceA-


Page 8 of 24

Page 9 of 24


190

containing Dehalococcoides mccartyi populations (Figure 1E and Figure 1F), and (iii) 20-25 %

191

of total influent reducing equivalents consumed by homoacetogenesis. In both the EV5L and

192

VS5L chemostats, TCE was transformed to ethene (~99%) and VC (~1%). Under the formate-

193

limiting conditions, imposed on EV2L and VS2L, the steady-state hydrogen concentration

194

dropped to 1-2 nM and acetate dropped below its detection limit (20 uM) (Figure 1C and 1D). A

195

reduction in the frequency of Selenomondales and Spirochaetes OTUs occurred in the 1000 days

196

following the onset of electron donor limitation (Figure S1, Supporting Information). Most

197

notably, the frequency of OTU representing the tceA-containing Dehalococcoides mccartyi

198

population increased, numerically dominating the vcrA-Dehalococcoides population. After 2000

199

days of reactor operation, greater than 99 % of total Dehalococcoides were the tceA-containing

200

population in both reactors EV2L and VS2L (Figure 1A and Figure 1B). The change in non-

201

Dehalococcoides population members is shown in the Supporting Information, Figure S1. Under

202

the formate-limiting conditions in EV2L and VS2L, TCE was transformed to ethene (~50%) and

203

VC (~50%) as shown in Figure 2E and Figure 2F.

204

After approximately 2500 days of operation of the EV2L and VS2L reactors under

205

formate-limiting condition, the OTU frequencies associated with the tceA-containing

206

Dehalococcoides population exceeded that of vcrA-associated OTU by more than three orders of

207

magnitude (Figure 1). After 2500 days, the vcrA-containing population had decreased in the

208

EV2L reactor to non-detectable levels, equivalent to a frequency of less than ~ 0.0001 of the

209

population at the DNA level (Figure 1A). After 2200 days, the vcrA-containing population

210

decreased in the VS2L reactor to less than 0.001 of the population at the DNA level (Figure 1B).

211

By contrast, in control reactors EV5L and VS5L, the long-running parent reactors of EV2L and

212

VS2L that had been maintained under constant formate-excess conditions, the vcrA-containing

213

Dehalococcoides population frequency remained above 0.1 of the total community, at the DNA

214

level, for the entire period of the seven-year experiment (Figure 1E and Figure 1F).

215

We attribute the decrease in vcrA-containing Dehalococcoides under formate limiting

216

conditions to the ecological process of competitive exclusion. When electron donor becomes the

217

environment’s limiting catabolic resource, the non-vcrA Dehalococcoides subpopulation exhibits

218

higher relative fitness than in the non-limited environment. The gradual competitive exclusion

219

observed suggests a possible mechanism underlying the fitness difference between



220

Dehalococcoides subpopulations other than a strict hydrogen threshold difference between

221

subpopulations, which is not well supported by the data. Its stands to reason that if the tceA-

222

population had a lower hydrogen threshold than the vcrA-population, steady-state hydrogen

223

levels would have been depressed below the higher threshold concentration, causing the abrupt

224

loss of one but not the other population. Instead, the fitness advantage manifest by the tceA

225

population is more likely explained by differences in effective dechlorination rates, since these

226

effectively govern the rate at which each Dehalococcoides subpopulation can consume electron

227

donor and thus control the relative rate of ATP synthesis in each population. If the Kmax,TCE,tceA

228

(the maximum rate of TCE dechlorination by the tceA population) is greater than the Kmax,VC,vcrA

229

and the Kmax,DCE,tceA is greater or equal to the Kmax,DCE,vcrA, then the tceA strain would be predicted

230

to consume a larger fraction of environment’s limiting resource, resulting in the decline of the

231

vcrA-containing population under stoichiometric electron donor deficit conditions. Based on this

232

differential kinetic hypothesis, ecological exclusion need not be rapid.

233

After seven years of operation, we increased the electron donor concentration in the

234

EV2L and VS2L reactors from 50 mM eeq formate back to the conditions of the parent reactors

235

– 90 mM eeq formate. Figure 2 shows the relative abundance of the vcrA and tceA-containing

236

Dehalococcoides populations shifting dramatically. The frequency of the vcrA-population

237

increased from less than 0.001 in both reactors to 0.38 in reactor EV2L and 0.57 in reactor

238

VS2L. The near convergence in the frequency of the vcrA and tceA populations occurred within

239

two CSTR residence times (100 days). In both reactors, the increase in influent formate

240

concentration did not have a major impact on the frequency of the Desulfuromonadales OTU.

241

With respect to other potential homoacetogenic populations in reactor EV2L, we observed a

242

transient increase in the frequency in Selenomondales OTU 3, peaking 77 days following the

243

onset of formate stimulation (EV2L, day 2684). In reactor VS2L, Spirochaetes OTU 9

244

temporarily increased between the samples taken at operational days 2376 and 2406 following

245

the onset of formate stimulation (VS2L,day 2380), but the aggregate frequency of

246

Selenomondales and Spirochaetes OTUs did not return to the level observed at the onset of the

247

formate-deficit condition (Figure S1, Supporting Information).

248 249

In both reactor EV2L and VS2L, hydrogen rapidly increased from < 2 nM to steady state levels between 20-40 nM within four days after formate-excess conditions were restored.


Page 10 of 24

Page 11 of 24


250

Acetate, which previously had not been observed above the limit of detection (20 µM), gradually

251

increased to more than 3 mM within 100 days after formate-excess conditions were restored

252

(Figure 2C and Figure 2D). The acetate concentration of 3 mM is in the range of that produced in

253

the EV5L and VS5L chemostats (Figure 1G and Figure 1H). Within 100 days following the

254

increase in the influent formate concentration fed to the EV2L and VS2L reactors, the fraction of

255

non-Dehalococcoides microbes as well as their total biomass had not recovered to those levels

256

observed at the beginning of the study (Figure S1, Supporting Information). The residual VC

257

concentration (4 mM) in the reactor remaining at the onset of increase in influent formate

258

concentrations may have biased the assemblage of the re-stimulated consortium towards

259

organohalide-respiring organisms. The production of methane was not observed after restoring

260

the reactors back to the 45 mM influent formate condition.

261

3.2 Vinyl Chloride Dechlorination Kinetics After Long-Term Electron Donor Limitation

262

According to the consensus view of organohalide-respiration, specific VC reductases, not

263

present in all strains of Dehalococcoides mccartyi, are necessary to efficiently catalyze complete

264

chloroethene degradation to ethene. Other strains of Dehalococcoides mccartyi may contain

265

reductive dehalogenases with non-specific, and perhaps cometabolic, VC activity.24,25 In our

266

reactor study, constant VC dechlorination did indeed proceed even as the vcrA-containing

267

Dehalococcoides population was greatly depleted from the EV2L and VS2L reactor consortia

268

during formate-limiting conditions. In the reactors during formate-limitation, the steady-state rate

269

of VC dechlorination was approximately 80 µmol L-1d-1. Notably, this VC dechlorination rate

270

was sustained under high steady-state VC concentrations in excess of 4 mM, favorable for co-

271

metabolic or non-specific VC dechlorination by reductive dehalogenases with low affinity for

272

VC as a halogenated substrate. VC transformation Ks values of 0.29 mM and 0.602 mM have

273

been reported for Dehalococcoides cultures when the cometabolic transformation of VC was

274

implied.14,26 Thus, 4 mM VC in the chemostat is in the range where significant transformation

275

via cometabolism is expected. At these high concentrations, and given reported KS,VC values,

276

the maximum rate of VC dechlorination would be expected; however, other factors such as the

277

low concentration of hydrogen contribute to the overall rates observed. To elucidate the

278

functional consequences of depleting the vcrA-subpopulation from the dechlorinating

279

community, we measured consortium level dechlorination kinetics after exposure to the long

280

period of electron donor deficit. To enable kinetic studies over a range of substrate



281

concentrations, we transferring reactor biomass to batch reaction vials which we spiked with

282

TCE and formate. Although we had observed an appreciable steady-state rate of VC

283

dechlorination (80 µmol L-1d-1) at millimolar VC concentrations (4 mM, 250 ppm) in the EV2L

284

and VS2L, batch kinetic tests revealed that the depletion of the vcrA-containing Dehalococcoides

285

population greatly reduced these consortia’s capacity for VC reduction at sub-millimolar

286

concentrations. This was indicated in batch depletion studies where VC was present at an

287

aqueous concentration of 50 µM – 1 µM (3.2 ppm - 62.5 ppb). Here, the maximum observed rate

288

of VC dechlorination by the vcrA-depleted consortium, after TCE and DCE had been fully

289

consumed, was 5 µmol L-1 d-1 (95 % CI [2,8] µmol L-1 d-1). By comparison – at similar low µM

290

concentrations – the maximum observed rates of TCE and DCE dechlorination in these batch

291

depletion assays were more than two orders of magnitude greater than those observed for VC.

292

The observed dechlorination rates are presented in Table S1, Supporting Information. Notably,

293

the slow rate of VC dechlorination manifest in the vcrA-depleted consortia at micromolar VC

294

concentrations would alone be kinetically insufficient to support a tceA-Dehalococcoides

295

population’s net growth given natural decay rates.27

296

3.3 Vinyl Chloride Dechlorination Kinetics after Electron Donor Stimulation

297

When the EV2L and VS2L reactors were returned to formate-excess conditions, the rates

298

of VC dechlorination in subsequent batch depletion assays increased after a lag period of 30

299

days. A strong positive correlation was found between the consortium’s maximum VC

300

dechlorination rate in batch assays and the abundance of the vcrA-containing Dehalococcoides

301

mccartyi population in the reactor at the time the batch assay was conducted (Figure 3).

302

Additionally, the consortium’s affinity for VC dechlorination at low concentrations increased

303

following the return to formate excess-conditions. Figure 4 shows a direct comparison of batch

304

kinetics observed with reactor material sampled before (Figure 4A) and after (Figure 4B) the

305

reactors’ return to growth under elevated concentrations of electron donor. Once the vcrA-

306

containing population exceeded 25 % of the total bacteria community, on the DNA level, high

307

rates of VC dechlorination (> 200 - 600 µmol L-1 d-1) were observed even at low (< 25 µM, 1.5

308

ppm) VC concentrations (Figure 4, Table S1, Supporting Information).

309

Close examination of the temporal changes in the VC concentrations in the chemostats

310

reveals two time periods after the influent formate concentration was increased where different


Page 12 of 24

Page 13 of 24


311

VC transformation processes occurred (Figure 2). First, an immediate decrease in effluent VC

312

was observed after increasing the influent formate concentration, followed by a plateau before a

313

second acceleration in rate was observed 30 days later (Figure 2). This pattern is mirrored by a

314

step-wise increase in ethene concentrations. A log plot of VC concentrations versus time well

315

illustrates these discrete shifts in the CSTRs’ rate of VC dechlorination (Figure S3, Supporting

316

Information). The first decrease in VC concentration likely resulted from enhanced VC

317

transformation by the tceA-enriched population dominant in the chemostat at the time of the

318

formate shift. This response occurred with the addition of higher influent formate and the rapid

319

increase in hydrogen concentrations from ~ 1 nM to 30 nM. A previous study estimated the Ks

320

values for hydrogen in Dehalococcoides mccartyi strain VS to be 5 nM.28 Thus if the

321

Dehalococcoides mccartyi in the CSTRs under investigation here had similar affinity for

322

hydrogen, an immediate increase in overall dechlorination rate would be expected following the

323

observed increase in hydrogen concentration after relieving formate-limitation. The second

324

decrease in VC concentrations starting at 2740 and 2440 days in EV2L and VS2L chemostats,

325

respectively, coincides with the increased frequency in vcrA-population.

326

In the reactor experiments, the recovery of the vcrA-population and dechlorination of VC

327

to low micromolar concentrations occurred in less than two reactor residence times. The resilient

328

population response was likely hastened by the 4 mM residual VC remaining in the reactor when

329

the influent formate concentration was increased from 25 mM to 45 mM. In situ recovery times

330

of the vcrA-containing populations in a groundwater contaminant plume, without a large

331

reservoir of VC from stalled TCE dechlorination, would likely be slower. Nevertheless, the

332

apparent difference in affinity for VC at low concentrations observed in the balanced tceA/vcrA

333

consortium versus the tceA-dominated consortium (Figure 4 and Table S1, Supporting

334

Information) indicates that the vcrA-population would rebound in the presence of excess electron

335

donor even in cases where VC is present at lower steady state concentrations.

336

3.4 Implications for Contaminated Sites

337

Our results underscore the importance of electron donor/chloroethene stoichiometry

338

whether formate or more commonly used substrates (e.g., lactate or emulsified vegetable oil) are

339

used. Electron donor competition is a well-recognized determinant of dechlorination efficiency

340

in the environment. A number of previous investigations have shown that electron donor



341

limitation, often exacerbated by the presence of competing terminal electron accepting processes

342

(e.g. metal reduction, methanogenesis, or sulfate reduction), can diminish rates of dechlorination

343

and survival of Dehalococcoides.14,17,22,29-31 In this study, we demonstrated that electron donor

344

scarcity alone, in the absence of competing metabolic processes or inhibitory concentrations of

345

intermediate dechlorination products, is sufficient to greatly alter intra-Dehalococcoides

346

population structure. Where the flux of available hydrogen is in stoichiometric deficit to TCE

347

input, our results suggest that tceA-containing Dehalococcoides, if present, will likely

348

outcompete vcrA-containing populations, increasing the risk of incomplete chloroethene

349

removal.

350

As greater attention is being paid to catabolic food webs, cross-feeding, and competition

351

within organohalide respiring consortia,32,33 the sensitive ecological balance between closely

352

related Dehalococcoides mccartyi populations should not be overlooked. In this study, we

353

showed directly through time-series community characterization paired with consortium level

354

kinetic tests that these intra-species population shifts have significant functional consequences.

355

Even though our results suggest appreciable VC dechlorination may still proceed after the

356

depletion of the VC-specialized Dehalococcoides population, the survival of a vcrA-population is

357

likely crucial for low-concentration VC removal to stringent cleanup levels (MCLVC: 2 ppb, 30

358

nM in the United States) mandated when groundwater is to be restored to drinking water

359

standards. Our results provide a cautionary perspective on remediation forecasts that assume

360

dispersal and growth of multiple Dehalococcoides respiring organisms following a discrete

361

biostimulation event. Therefore, asymmetrical growth rates among different Dehalococcoides

362

subpopulations during electron donor limitation should be included in remediation models.

363

However, this study is encouraging for the design of multi-phase biostimulation efforts given the

364

observed long-survival times and robust recovery of the depleted vcrA-containing

365

Dehalococcoides populations after more than six years of electron donor-limitation.

366

ACKNOWLEDGMENTS

367

This work was funded by an NSF Grant (MCB-1330832). KMB was additionally

368

supported by an NSF Graduate Research Fellowship (2011103493). We wish to thank Benjamin

369

Callahan for assistance in denoising and variant detection performed on the DNA amplicon

370

sequences data.


Page 14 of 24

Page 15 of 24


371 372 373

Supporting Material

374

assays, after increasing the influent concentration formate from 25 mM to 45 mM in reactors

375

EV2L and VS2L. Supporting Figure S1 displays the bacterial population structure during

376

prolonged exposure to electron donor limitation and electron donor stimulation in reactor EV2L.

377

Supporting Figure S2 displays the bacterial population structure, hydrogen levels, and

378

dechlorination products in EV2L and VS2L reactors upon electron donor stimulation.

379

Supporting Figure S3 shows the log-concentration of VC and ethene in reactors EV2L and

380

VS2L, emphasizing two discrete changes in VC rate following the increase in influent electron

381

donor.

Supporting Table S1 reports the dechlorination kinetics measured in batch cell suspension

382 383 384



385 386 387

Page 16 of 24

Tables and Figures Table 1: CSTRs in this study Reactor

Electron Acceptor

Electron Donor

Formate-Excess Conc.

EV5L

Conc.

Formate-Limiting Conc.

Compound

(mM)

Compound

(mM)

Days

(mM)

Days

TCE

8.5

Formate

45

1-3000

-

-

1-180; 2684EV2L

TCE

8.5

Formate

45

2830

25

181-2684

VS5L

TCE

8.5

Formate

45

1-2500

-

-

25

181-2380

1-180; 2380VS2L

TCE

8.5

Formate

45

2600

388 389


Page 17 of 24

390


Table 2: CSTR growth conditions

Formate-Excess

Formate-Deficit

45 mM

25 mM

(90 eeq)

(50 eeq)

8-9 mM

8-9 mM

(48-54 eeq)

(48-54 eeq)

Residence time

50 days

50 days

Equilibrium hydrogen (aq)

~ 40 nM

~ 2 nM

30%

Survival of Vinyl Chloride Respiring Dehalococcoides mccartyi under

Recommend Documents