Microbiology Papers in Press. Published April 29, 2014 as doi:10.1099/mic.0.079293-0

1

Production and consumption of nitrous oxide in nitrate-

2

ammonifying Wolinella succinogenes cells

3 4

Monique Luckmann1, Daniel Mania2, Melanie Kern1, Lars R. Bakken3,

5

Åsa Frostegård2 and Jörg Simon1

6 7

1

Microbial Energy Conversion and Biotechnology, Department of Biology,

8

Technische Universität Darmstadt, Schnittspahnstraße 10, 64287 Darmstadt,

9

Germany

10

2

of Life Sciences, Chr Falsens vei 1, N1432 Ås, Norway

11 12 13

Department of Chemistry, Biotechnology and Food Science, Norwegian University

3

Department of Plant and Environmental Sciences, Norwegian University of Life Sciences, PO Box 5003, N1432 Ås, Norway

14 15

For Correspondence:

16

Prof. Dr. Jörg Simon; Tel.: +49 (0)6151 165203; Fax: +49 (0)6151 162956; E-mail:

17

[email protected]

18 19

Running title: Nitrous oxide turnover in W. succinogenes

20

Contents category: Physiology and Biochemistry

21

Number of words in Summary: 290; number of words in the main text: 5564

22

Number of Tables: 0; number of figures: 7

23

Non-standard abbreviations: cNosZ, cytochrome c nitrous oxide reductase; Nrf,

24

cytochrome c nitrite reductase; pmf, proton motive force

25 1

26

SUMMARY

27

Global warming is moving more and more to the public consciousness. Besides the

28

commonly mentioned carbon dioxide and methane, nitrous oxide (N2O) is a powerful

29

greenhouse gas in addition to its contribution to depletion of stratospheric ozone. The

30

increasing concern about N2O emission has focused interest on underlying microbial

31

energy converting processes and organisms harbouring N2O reductase (NosZ) such

32

as denitrifiers and ammonifiers of nitrate and nitrite. Here, the epsilonproteobacterial

33

model organism Wolinella succinogenes is investigated towards its capacity to

34

produce and consume N2O during growth by anaerobic nitrate ammonification. This

35

organism synthesizes an unconventional cytochrome c nitrous oxide reductase

36

(cNosZ), which is encoded by the first gene of an atypical nos gene cluster. However,

37

W. succinogenes lacks a nitric oxide (NO)-producing nitrite reductase of the NirS- or

38

NirK-type as well as an NO reductase of the Nor-type. Using a robotized incubation

39

system, the wild-type strain and suitable mutants of W. succinogenes that either

40

produced or lacked cNosZ were analysed as to their production of NO, N2O and N2 in

41

both nitrate-sufficient and nitrate-limited growth medium using formate as electron

42

donor. It was found that cells growing in nitrate-sufficient medium produced small

43

amounts of N2O that derived from nitrite and, most likely, from the presence of NO.

44

Furthermore, cells employing cNosZ were able to reduce N2O to N2. This reaction,

45

which was fully inhibited by acetylene, was also observed after adding N2O to the

46

culture headspace. The results indicate that W. succinogenes cells are competent in

47

N2O and N2 production despite being correctly grouped as respiratory nitrate

48

ammonifiers. N2O production is assumed to result from NO detoxification and

49

nitrosative stress defence while N2O serves as a terminal electron acceptor in

50

anaerobic respiration. The ecological implications of these findings are discussed.

51 2

52

INTRODUCTION

53

Nitrous oxide (N2O, also known as laughing gas) is a chemically rather inert trace gas

54

in earth’s atmosphere whose level has been consistently increasing over the last

55

decades (Smith, 2010), largely driven by high input of reactive nitrogen in agriculture

56

(Reay et al. 2012). N2O is involved in stratospheric ozone depletion and, moreover, it

57

is a strong greenhouse gas. Its capacity to absorb infrared radiation is about 300

58

times greater than that of CO2 and therefore anthropogenic N2O emissions, unless

59

balanced by N2O-consuming processes, contribute significantly to climate change.

60

This is even more important in the light of the fact that an ever increasing amount of

61

synthetic nitrogen-based fertilizers is used to raise agricultural efficiency.

62

For a long time N2O has been known as a crucial intermediate in the

63

biogeochemical nitrogen cycle and the majority of atmospheric N2O is of microbial

64

origin (for recent reviews see Richardson et al., 2009 and Thomson et al., 2012). The

65

most important processes in biological N2O production and turnover are thought to be

66

denitrification (reduction of nitrate or nitrite to dinitrogen via nitric oxide (NO) and N2O

67

as intermediates) and nitrification (oxidation of ammonium or nitrite to nitrate with N2O

68

being a side product under aerobic conditions in a process referred to as nitrifier

69

denitrification) (Stein, 2011). Notably, some denitrifying organisms lack N2O reductase

70

activity and release N2O as an end product. Recently, nitrate- (or nitrite-) ammonifying

71

cells also have been reported to release N2O (Streminska et al., 2012; Giles et al.,

72

2012). In bacterial denitrification and in denitrifying nitrifiers, NO is reduced by one of

73

various membrane-bound respiratory NO reductases (predominantly cNor and qNor

74

enzymes) (Simon and Klotz, 2013). Since NO is a highly toxic compound that exerts

75

nitrosative stress on cells and organisms, it needs to be detoxified (Poole, 2005). It is

76

therefore not surprising that N2O production from NO was described for numerous 3

77

non-respiratory enzymes including flavodiiron proteins (Fdp), flavorubredoxin

78

(NorVW), cytochrome c554 (CycA; in nitrifiers), cytochrome c’-beta (CytS) and

79

cytochrome c’-alpha (CytP) (Simon and Klotz, 2013). In these cases, NO reduction is

80

thought to serve predominantly in NO detoxification, i.e. in nitrosative stress defence.

81

In many bacterial and archaeal prokaryotes, N2O is further reduced to dinitrogen

82

by a homodimeric N2O reductase (NosZ), which is a copper-containing enzyme

83

located at the outside of the cytoplasmic membrane (i.e. the periplasm in Gram-

84

negative bacteria) (Zumft, 1997, Zumft & Kroneck, 2007; Pomowski et al., 2011; Wüst

85

et al., 2012; Pauleta et al., 2013). Various NosZ enzymes carry a C-terminal extension

86

that forms a monohaem cytochrome c domain which was first described for non-

87

denitrifying Epsilonproteobacteria (Teraguchi & Hollocher, 1989; Simon et al., 2004).

88

The cytochrome c domain of such cNosZ enzymes is thought to donate electrons to

89

the CuA site of NosZ (Dell’acqua et al., 2011; Simon and Klotz, 2013). Recently, so-

90

called atypical NosZ proteins (including cNosZ enzymes) and atypical nos gene

91

clusters from non-denitrifying bacteria and archaea were found to be widespread in

92

terrestrial environments where they form an evolutionarily distinct clade of N2O

93

reductases (Sanford et al., 2012; Jones et al., 2013). Based on the abundance of

94

these genes in soils, it was suggested that the corresponding organisms could be a

95

significant sink for N2O (Jones et al. 2013). Furthermore, genome sequences suggest

96

that ammonification and N2 production from nitrate/nitrite may coexist in some bacteria

97

(Sanford et al., 2012; Mania et al., 2014).

98

The metabolism of N2O in nitrate- or nitrite-ammonifying organisms is poorly

99

understood. The respective organisms reduce nitrate to nitrite using a membrane-

100

bound nitrate reductase (Nar) and/or a periplasmic nitrate reductase (Nap)

101

(Richardson et al., 2001; Kern & Simon, 2009; Simon and Klotz, 2013). Prominent 4

102

examples are Gamma-, Delta- and Epsilonproteobacteria such as Escherichia coli,

103

Salmonella enterica serovar Typhimurium, Shewanella oneidensis, Anaeromyxobacter

104

dehalogenans and Wolinella succinogenes. Subsequently, nitrite is reduced to

105

ammonium by cytochrome c nitrite reductase (NrfA), which obtains electrons from the

106

quinone/quinol pool through several different electron transport redox enzyme

107

systems, depending on the organism (Simon, 2002; Kern & Simon, 2009; Simon and

108

Klotz, 2013). Many NrfA-containing ammonifiers that also reduce N2O encode cNosZ,

109

for example W. succinogenes, Campylobacter fetus, Desulfitobacterium

110

dehalogenans and A. dehalogenans (Payne et al., 1982; Kern & Simon, 2009;

111

Sanford et al., 2012).

112

The Epsilonproteobacterium W. succinogenes is a versatile respiratory rumen

113

bacterium that typically uses hydrogen gas or formate as electron donor combined

114

with a multitude of electron donor substrates such as nitrate, nitrite, sulfite, polysulfide

115

and fumarate (Kröger et al., 2002; Kern & Simon, 2009; Kern et al., 2011a; Simon and

116

Kroneck, 2013). The cells stoichiometrically produce ammonium from both nitrate and

117

nitrite and the corresponding cell yields (normalized on the basis of formate

118

consumption) were found to be similar using either of these terminal electron

119

acceptors (Bokranz et al., 1983). W. succinogenes employs the Nap and Nrf systems

120

for respiratory nitrate ammonification but lacks NO-producing nitrite reductases of the

121

NirK- (copper nitrite reductase) or NirS-type (cytochrome cd1 nitrite reductase) known

122

from denitrifiers as well as a respiratory NO reductase of the cNor and qNor-families

123

(Kern & Simon, 2009). In addition to its function as terminal reductase in respiratory

124

nitrite ammonification, NrfA was recently shown to function in nitrosative stress

125

defence implying that NrfA efficiently reduces NO to ammonium (Kern & Simon, 2009;

126

Kern et al., 2011b; Simon and Klotz, 2013). 5

127

W. succinogenes cells were previously reported to grow with N2O as sole electron

128

acceptor using formate as electron donor although growth parameters have not been

129

reported (Yoshinari, 1980; Schumacher et al., 1992). The W. succinogenes cNosZ

130

enzyme has been purified from N2O-grown cells and shown to contain N2O reductase

131

activity using reduced benzyl viologen as electron donor (Teraguchi & Hollocher,

132

1989). When the genome of the W. succinogenes type strain (DSM 1740T) was

133

sequenced, an atypical nos gene cluster was found whose first gene encodes the

134

cNosZ enzyme (Baar et al., 2003; Simon et al., 2004). Interestingly, this gene was

135

found to be disrupted in the genome sequence by a copy of insertion element IS1302

136

(Simon & Kröger 1998), thus raising the question whether this strain was comparable

137

to the strains used in previous studies. Here, we report the construction of mutants

138

that either lack nosZ or contain a genetically stable nos gene cluster. The capacity of

139

these mutants and of the wild-type strain to produce and to consume N2O during

140

growth by nitrate ammonification was examined using nitrate-sufficient or nitrate-

141

limited medium containing formate as electron donor. Using a robotized cell incubation

142

system, the concentrations of nitrate, nitrite, NO, N2O and N2 were quantified with high

143

sensitivity. The applied growth conditions for nitrate respiration are regarded to be

144

more physiologically relevant than previously conducted growth experiments in the

145

presence of externally added N2O serving as sole electron acceptor.

146 147

METHODS

148

Organisms and liquid media. W. succinogenes cells used in this study are

149

described in Fig. 1. Cultures were routinely grown in liquid medium at 37°C

150

containing formate as sole electron donor and either fumarate (Kröger et al., 1994) or

151

nitrate as terminal electron acceptor. Nitrate-sufficient (80 mM sodium formate/50 6

152

mM potassium nitrate) or nitrate-limited (80 mM sodium formate/10 mM potassium

153

nitrate) medium was used. In addition to formate and nitrate, each medium (pH 7.5)

154

contained 1 mM K2HPO4, 5 mM fumaric acid, 5 mM ammonium sulfate, 1 mM MgCl2,

155

0.2 mM CaCl2, 0.68 mM glutamic acid, 0.57 mM cysteine and 0.2 ml l-1 trace element

156

solution as described by Pfennig & Trüper (1981). Media were degassed and flushed

157

with either nitrogen gas or helium several times to reduce the oxygen content. Brain-

158

heart infusion broth (BHIB; 0.5%, w/v), kanamycin (25 mg l-1) and/or chloramphenicol

159

(12.5 mg l-1) were added to the medium as appropriate.

160 161

Growth conditions, determination of cytochrome c and nitrogen compounds.

162

Standard growth curves were generated by measuring the optical density at 578 nm

163

(OD578) of cultures (50 ml) inoculated with overnight pre-cultures at an initial OD578 of

164

about 0.05.

165

Nitrate and nitrite concentrations in cell supernatants were determined using

166

colorimetric assays based on the formation of spectrophotometrically detectable

167

compounds. Nitrate was converted to 4-nitro-2,6-dimethylphenol under acidic

168

conditions (Hartley & Asai, 1963). For nitrite determination, diazotization with

169

sulfanilic acid and subsequent coupling with N-(1-naphthyl)ethylenediamine was

170

performed (Rider & Mellon, 1946).

171

Detection of cytochromes c in SDS polyacrylamide gels was performed by a

172

haem staining procedure using 3,3′-dimethoxybenzidine as described previously

173

(Kern et al., 2011).

174

For monitoring gas kinetics 120 ml serum vials (50 ml medium) equipped with a

175

triangular magnetic stirring bar were used. The vials were tightly sealed with rubber

176

septa and aluminum caps. Prior to inoculation, the air was replaced by helium gas 7

177

(six cycles of evacuation and helium-filling) using a semi-automated system (Molstad

178

et al., 2007). During this process, the medium was stirred at 950 rpm to ensure

179

sufficient gas exchange between the liquid and the gas phase. Excess pressure was

180

relieved using a water-filled syringe. The vials were incubated at a constant

181

temperature of 37°C in a robotized incubation system designed for measuring gas

182

kinetics (Molstad et al., 2007; Bergaust et al., 2008). In short, this system monitors

183

NO, N2O and N2 by frequent sampling with a syringe through the septum using a

184

peristaltic pump coupled to the inlet of a gas chromatograph and a

185

chemiluminescence NO analyzer. After each sampling, the pump is reversed and

186

pumps back helium to ensure near constant pressure (1 atm) during the entire

187

incubation period. The sampling thus dilutes the headspace with helium. This dilution

188

was 3.4% per sampling in the original system (Molstad et al., 2007), but only 0.9% in

189

the new system used for most of the experiments in the present study. There is also

190

a marginal but constant leakage of traces of N2 and O2 with each sampling (14 nmol

191

O2 and 42 nmol N2 per sampling). Dilution by sampling and the minor leakage of N2

192

into the vials are taken into account when estimating gas transformation rates for

193

each time increment between two samplings (see Molstad et al., 2007 for details),

194

and when calculating cumulative N2 production (reported N2 is invariably cumulative

195

N2 production, corrected for leakage and sampling dilution). Several batch cultivation

196

experiments were run with a variety of treatments designed to characterize the gas

197

kinetics in the various cultures. Replication was secured by repeating entire

198

experiments, rather than including replicate vials in each run.

199 200

Construction of W. succinogenes mutants. Standard genetic procedures were

201

used (Sambrook et al., 1989). Genomic DNA was isolated from W. succinogenes

202

using the DNeasy Tissue kit (Qiagen). PCR was carried out using Phusion High 8

203

Fidelity DNA polymerase (Finnzymes) for cloning procedures or Biotaq Red DNA

204

polymerase (Bioline) for mutant and plasmid screening with standard amplification

205

protocols.

206

W. succinogenes ΔnosZ was constructed through double homologous recombination

207

of the wild-type (strain DSM 1740T) genome with a deletion plasmid (pΔnosZ cat)

208

designed to replace the entire nosZ gene in the nos operon by a chloramphenicol

209

resistance gene cartridge (cat). For homologous recombination, the respective

210

deletion plasmid contained cat flanked by two DNA segments obtained by PCR that

211

were identical to appropriate regions in the W. succinogenes genome (Fig. 1). The

212

two PCR fragments were synthesized using the following primer pairs: 5'-

213

GCGAATTCCGCAGACGGATTTGAGG-3' and 5'-

214

CCGGATCCTTACCCACTCCTTTTTGTTAATTTGG-3' for amplifying the upstream

215

fragment, and 5'- CCGGATCCAACCCTCTCCTCTCC-3' and 5'-

216

CCAAGCTTGGGATACATCCCGATGTAGTGATTG-3' for the downstream fragment

217

(black bars in Fig. 1). Primers carried EcoRI, BamHI or HindIII (underlined) for

218

cloning. Both fragments and the cat fragment were consecutively inserted into

219

plasmid pPR-IBA1 (IBA Lifesciences) using E. coli XL-1 Blue as cloning host grown

220

in LB medium at 37°C. PCR analysis was used to confirm that the plasmid contained

221

cat in the same orientation as its surrounding regions. Transformation of W.

222

succinogenes DSM 1740T cells was performed by electroporation as described

223

previously (Simon et al., 1998; 2000). Transformants of W. succinogenes were

224

selected in the presence of chloramphenicol (12.5 mg l-1) and the desired gene

225

deletion was confirmed by PCR. The integrity of DNA stretches involved in

226

recombination events was confirmed by sequencing suitable PCR products.

9

227

To construct W. succinogenes kompPnosnosZ the nosZ gene lacking IS1302 was

228

restored on the genome of W. succinogenes ΔnosZ upon integration of plasmid

229

pkompPnosnosZ kan (Fig. 1) This plasmid contained an intact copy of nosZ

230

surrounded by the upstream and downstream cat-flanking fragments of pΔnosZ cat.

231

In addition, a kanamycin resistance gene cassette (kan), obtained by BamHI excision

232

from pUC4K, was inserted between nosZ and the downstream flanking fragment.

233

Finally, a nos promoter element was inserted downstream of kan to ensure

234

transcription of the residual nos operon (Fig. 1). In a first step to create

235

pkompPnosnosZ kan, the cat gene was replaced by kan in pΔnosZ cat resulting in

236

pΔnosZ kan. PCR analysis was used to confirm that the plasmid contained kan in the

237

reverse orientation relative to the nos sequence. The intact nosZ gene was amplified

238

using the primer pair 5-ATGCAAAGATTACTAAAGCAATCTTTGGTTG-3' and 5'-

239

TTAGAAGAGACCCGCGTTCTCATC-3' using genomic DNA of wild-type cells as

240

template (note that there is a small percentage of cells that transiently lack the copy

241

of IS1302 within nosZ). To avoid any adverse effects by adding additional restriction

242

sites between the native nos promotor and the nosZ gene, the amplified gene was

243

blunt-end ligated with a linear plasmid fragment obtained by PCR from pΔnosZ kan

244

with the primer pair 5'-GGGGATCCCGCTGAGGTCTGCCTCGTGAAGAAGGTG-3'

245

and 5'-TTACCCACTCCTTTTTGTTAATTTGGCATTAAC-3' resulting in pkompnosZ

246

kan. In the last step, the additional nos promotor fragment (Pnos) was amplified using

247

the primer pair 5'-GGGCGTGGGGAAGAAAATCTCC-3' and 5'-

248

TTACCCACTCCTTTTTGTTAATTTGGCATTAAC-3' and blunt-end ligated with a

249

linear plasmid fragment obtained by PCR from pkompnosZ kan with the primer pair

250

5-GGGGATCCAAAGCCACGTTGTGTCTCAAAATCTCTG-3' and 5'-

251

ATGGAAAACAGAGGAATCGCCAAAT-3' to give pkompPnosnosZ kan. The

252

corresponding mutant W. succinogenes kompPnosnosZ kan was constructed by 10

253

transforming W. succinogenes ΔnosZ cat with pkompPnosnosZ kan. Transformants

254

were selected in the presence of kanamycin (25 mg l-1). The desired double

255

homologous recombination was verified by PCR and the integrity of DNA stretches

256

involved in recombination events as well as the entire nosZ gene was confirmed by

257

sequencing suitable PCR products.

258

RESULTS

259

Construction of W. succinogenes mutants and characterization of cultures

260

As mentioned in the Introduction, the cNosZ-encoding gene from W. succinogenes

261

wild-type cells (strain DSM 1740T) was found to be disrupted by a copy of insertion

262

element IS1302. Therefore, it was assumed that the vast majority of these cells were

263

unable to produce functional cNosZ. To overcome the problem of genetic variability,

264

mutants W. succinogenes ΔnosZ and W. succinogenes kompPnosnosZ were

265

constructed through double homologous recombination using suitable plasmids as

266

described in the Methods section (Fig. 1). The ΔnosZ deletion mutant lacks the entire

267

nosZ gene, which is replaced by a chloramphenicol resistance gene cartridge (cat)

268

while mutant kompPnosnosZ possesses an intact nosZ gene, a kanamycin resistance

269

gene cartridge (kan) and an additional nos promoter to ensure transcription of

270

accessory nos genes (Fig. 1). The absence of IS1302 in the nosZ gene of mutant

271

kompPnosnosZ was routinely examined by PCR and confirmed that the mutant was

272

genetically stable. The presence of the cNosZ protein was shown in cells of mutant

273

kompPnosnosZ by haem staining whereas it proved undetectable in cells of the wild-

274

type and of mutant W. succinogenes ΔnosZ (Fig. 2).

275

All three W. succinogenes organisms were grown in anoxic batch cultures with

276

formate and nitrate as energy sources in either formate-limited/nitrate-sufficient (80

277

mM formate/50 mM nitrate) or formate-sufficient/nitrate-limited (80 mM formate/10 11

278

mM nitrate) medium, each supplemented with 0.5% (w/v) BHIB. In both media, the

279

cells showed comparable growth behaviour (Figs 3). The stationary growth phase

280

was reached after 7-8 h in nitrate-sufficient and 5-6 h in nitrate-limited medium and

281

similar doubling times of about 1.6 h were determined. However, in nitrate-sufficient

282

cultures large amounts of nitrite (30 to 40 mM) accumulated (Fig. 3A) whereas in

283

nitrate-limited medium nitrite was only transiently produced at a maximal

284

concentration of about 5 mM (Fig. 3B). Under the latter conditions, accumulating

285

nitrite is assumed to be rapidly converted to ammonium by cytochrome c nitrite

286

reductase NrfA, as described previously (Bokranz et al., 1983, Lorenzen et al., 1993;

287

Simon 2002). Production of the gaseous nitrogen compounds nitric oxide, nitrous

288

oxide and dinitrogen was determined in a series of experiments with the three strains

289

grown in either nitrate-sufficient or nitrate-limited medium. Fig 4 shows a

290

representative result for the three strains (single vial data) grown in nitrate-sufficient

291

medium. The wild-type and mutant W. succinogenes ΔnosZ accumulated low

292

amounts of nitrous oxide during the exponential growth phase (up to 1.7 µmol vial-1

293

after 15 h of incubation) whereas dinitrogen gas was very close to the detection limit.

294

The final cumulative N2 productions by these cultures were about 0.3 μmol N2. This

295

was most likely due to chemical reactions between nitrite and organic compounds in

296

the medium as discussed below. In contrast, the kompPnosnosZ mutant accumulated

297

dinitrogen but lacked nitrous oxide indicating the presence of active nitrous oxide

298

reductase. These differences between the cultures regarding N2O and N2

299

accumulation were also observed in two further replicate experiments (Fig. S2, S3).

300

Supplementary experiments were run with 10% acetylene (an inhibitor of nitrous

301

oxide reductase) in the headspace, demonstrating that the reduction of N2O to N2 by

302

mutant kompPnosnosZ was effectively inhibited by this treatment (Fig. S9, S10). For all

303

three organisms significant accumulation of nitric oxide was detected, but only during 12

304

the late exponential and stationary growth phases (Fig. 4). The NO accumulation

305

initiating during the late exponential/stationary phase was also observed in several

306

experiments shown in Fig. S9 and S10.

307

In nitrate-limited medium, the concentrations of nitric oxide invariably remained

308

below 1 nmol vial-1, nitrous oxide below 0.1 μmol vial-1 and the production of N2 was

309

below the detection limit (~0.5 μmol for cumulative N2 production over a 15 h

310

incubation period) (not shown). Nonetheless, the cNosZ and NrfA proteins were

311

detectable in cells of W. succinogenes kompPnosnosZ (Fig. 2). The very low amounts

312

of gaseous nitrogen compounds in nitrate-limited medium suggest that the

313

appearance of both NO and N2O depended on nitrite accumulation in nitrate-

314

sufficient medium. Therefore, control experiments were conducted in which varying

315

amounts of nitrite (10, 20 or 30 mM) were added to sterile nitrate-sufficient medium

316

(Fig. 5). In addition, vials with medium lacking BHIB were also analysed. N2O

317

production could not be detected in any of the treatments. On the other hand, NO

318

was found to accumulate in the medium with BHIB at a rate largely proportional to

319

the nitrite concentration but the rate declined with time (initial rate of NO formation

320

~10-4 mol NO mol-1 nitrite h-1; see Fig. S1 for a closer inspection of the kinetics). In

321

the medium without BHIB, however, NO production was marginal (~ 5% of that in the

322

presence of BHIB). The production of N2 was also significant and independent of

323

BHIB. The rate of N2 production was a linear function of nitrite concentration (~10-4

324

mol N2 mol-1 nitrite h-1, see Fig. S1). These results imply that NO is formed by

325

chemical reactions between nitrite and BHIB compounds, whereas the chemical N2

326

formation appears to be due to reactions between nitrite and compounds within the

327

basal medium. In theory, one could use the observed kinetics of chemical N2

328

formation from nitrite to correct the observed N2 formation in the cultures, thus 13

329

estimating the enzymatic production of N2. However, this turned out to be problematic

330

since the measured N2 accumulation in the wild-type and in mutant W. succinogenes

331

ΔnosZ was lower than that predicted (Fig. 4). In fact, these two organisms

332

accumulated only about 0.3 µmol N2 vial-1, which was much lower than predicted by

333

the observed reaction rate between nitrite and fresh sterile medium. Assuming that

334

the average concentration of nitrite was 35 mM for a time span of 7 hours, the

335

predicted chemical N2 formation would amount to 1.2 µmol N2 vial-1. Thus, N2-

336

formation by nitrite reacting with “spent medium” (i.e. after 7 hours of growth)

337

appears to be much lower as compared with fresh medium.

338

The fact that NO concentrations during the first 8-10 hours of the batch

339

cultivations were invariably lower than those predicted by the chemical

340

decomposition of the accumulating nitrite suggest that W. succinogenes cells are

341

able to efficiently reduce NO in order to maintain non-toxic levels. The enzymatic

342

inventory putatively serving in the underlying NO detoxification processes is

343

considered in the Discussion.

344 345

Addition of nitrous oxide and nitric oxide to W. succinogenes cultures

346

Additional experiments analogous to that shown in Fig. 4 were conducted, in which

347

1% N2O was added to the headspace of cultures grown in nitrate-sufficient medium

348

(Fig. 6). Under these conditions, only cells of W. succinogenes kompPnosnosZ

349

converted nitrous oxide stoichiometrically to dinitrogen (replicate experiments are

350

shown in Fig. S4 and S5). In nitrate-limited medium, cells of the kompPnosnosZ

351

mutant did also reduce added nitrous oxide to dinitrogen, but at a considerably lower

352

rate (about 30%) compared with nitrate-sufficient medium (Fig. S4, S5, S6).

353

Conceivably, this might be due to the apparently lower amount of cNosZ detectable 14

354

in cells grown in nitrate-limited medium (Fig. 2). Additional experiments with 10%

355

acetylene in the headspace demonstrated efficient acetylene-inhibition of N2

356

formation by strain kompPnosnosZ (Fig. S11). Finally, we conducted and experiment

357

where a bolus of NO (~1 μmol NO vial-1) was injected at the onset of the incubations.

358

All three organisms reduced NO to N2O within the first 4 hours but only cells of

359

mutant kompPnosnosZ reduced this N2O further to N2 (Fig. S7, S8).

360 361

Estimated specific cellular N2O reduction rates

362

The data for N2O reduction or N2 production rates in the various batch cultivations of

363

mutant kompPnosnosZ and the measured OD (converted to cell dry weight vial-1

364

assuming 0.4 mg cell dry weight mL-1 at OD578 = 1) were used to estimate the

365

specific N2 production rate throughout the incubations. The result is shown in Fig. 7,

366

where the rate expressed as μmol N2 g-1 cell dry weight h-1 is plotted against time for

367

the different treatments. In vials without NO or N2O injected, low and relatively

368

constant specific rates were recorded throughout the initial 8-15 hours. In vials with

369

NO injected, high rates were recorded initially, coinciding with a rapid reduction of the

370

entire injected NO which caused a transient peak in N2O during the first 4 hours (see

371

Fig. S8). In vials with excess N2O in the headspace, the specific rates increased

372

gradually during the first 10 hours, and declined rapidly after 12 to 15 hours. The N2O

373

concentration in the headspace was only reduced by about 15 %, thus the sharp

374

decline in N2 production after 15 hours was not due to substrate depletion. It is

375

currently unknown whether the presence of NO enhances the production of cNosZ in

376

nitrate-grown W. succinogenes cells.

377 378

15

379

DISCUSSION

380

The results demonstrate that nitrate-ammonifying cells of the tested W. succinogenes

381

cultures catalyse both N2O producing and N2O consuming reactions that eventually

382

result in N2 formation. Nonetheless, W. succinogenes cannot be regarded as a

383

denitrifier since the vast majority of nitrate has previously been shown to end up in

384

the respiratory ammonification pathway using the respiratory Nap and Nrf systems

385

(see Introduction). The production of N2O seems to be mainly associated with the

386

detoxification of the NO generated by chemical reactions between nitrite and medium

387

components, whereas N2O consumption is envisaged as an additional mode of

388

anaerobic respiration as discussed in the following sections.

389 390

Production of NO and N2O

391

In nitrate-sufficient cultures of W. succinogenes, the accumulation of nitrite causes

392

the emergence of small amounts of NO. This reaction is chemically obvious and

393

apparently facilitated by BHIB ingredients, presumably involving nitrosation and

394

decomposition of nitroso compounds. It has not been elucidated to date whether W.

395

succinogenes synthesizes an enzyme that produces NO from nitrite (assuming that

396

NrfA is unlikely to release NO as a by-product). The formation of NO from nitrite was

397

reported for membrane-bound nitrate reductase (NarGHI complex) from Salmonella

398

enterica serovar Typhimurium that, however, has no counterpart in W. succinogenes

399

(Gilberthorpe & Poole, 2008; Rowley et al., 2012). In contrast, there is no report that

400

a periplasmic nitrate reductase of the Nap-type is able to generate NO.

401

It is widely accepted that bacteria thriving in different habitats need to combat

402

environmental NO derived from nitrogen oxyanions (Poole, 2005). N2O and

403

ammonium are typical NO detoxification products when oxygen is absent. In case of 16

404

W. succinogenes it has been reported that at least two proteins are involved in

405

nitrosative stress defence (Kern et al., 2011b). These are periplasmic cytochrome c

406

nitrite reductase (NrfA) and a cytoplasmic flavodiiron protein (Fdp). NrfA is induced in

407

the presence of nitrate and efficiently catalyses the reduction of nitrite, NO and

408

hydroxylamine to ammonium in the periplasm (Simon, 2002; Einsle, 2011). This

409

reaction is thought to make W. succinogenes cells tolerant to high nitrite

410

concentrations, in line with the fact that nitrite is not needed to be transported into the

411

cytoplasm during nitrate ammonification. Reduction of NO to N2O by NrfA proteins

412

from Desulfovibrio desulfuricans, W. succinogenes and E. coli was also reported but

413

the physiological significance of this reaction is questionable (Costa et al., 1990;

414

Simon et al., 2011). Fdp proteins, on the other hand, are known to reduce NO to N2O

415

(Saraiva et al., 2004). However, this reaction has not been demonstrated for W.

416

succinogenes Fdp since the protein has not been purified. Further possible NO

417

reductases in W. succinogenes are the hybrid cluster protein (Hcp) and a homologue

418

(57% identity) of the HP0013 protein from Helicobacter pylori 26695 (WS1903,

419

NP_908012) (Justino et al., 2012). Fdp, Hcp and Ws1903 are all predicted to be

420

cytoplasmic proteins. Gene deletion strains lacking either nrfA, fdp or hcp are

421

available and will be examined towards their N2O production capacity in future

422

experiments.

423

The ability of nitrate-ammonifying cells to produce N2O in pure culture has been

424

reported before (Bleakley & Tiedje, 1982; Schumacher & Kroneck, 1992; Rowley et

425

al., 2012; Streminska et al., 2012). In Salmonella enterica serovar Typhimurium, 20%

426

of nitrate-N was released as N₂O under nitrate-sufficient conditions when nitrite had

427

accumulated to millimolar levels (Rowley et al., 2012). This phenotype is envisaged

428

to result from NO production from nitrite by the Nar-type nitrate reductase (see

429

above) and subsequent anoxic NO detoxification to N2O. Streminska et al. (2012) 17

430

demonstrated that soil isolates belonging to the genera Bacillus or Citrobacter formed

431

N2O under nitrate-sufficient conditions (low C-to-NO3- ratio) and found that a strain of

432

Citrobacter sp. reduced up to 5% of nitrate-N to N2O. Kaspar and Tiedje (1981)

433

reported that the nitrate-ammonifying rumen microbiota accumulated up to 0.3% of

434

the added nitrate-N as nitrous oxide and consequently stated that denitrification is

435

absent from the rumen. In the present study, exponentially growing W. succinogenes

436

cells converted about 0.15% of nitrate-N to N2O under the tested nitrate-sufficient

437

conditions (Fig. 3). These percentages, however, should be compared with caution

438

as W. succinogenes is not a dominant bacterium in cattle rumen. In nitrate-sufficient

439

medium, W. succinogenes cells continued to produce N2O after the transition from

440

the exponential to the stationary growth phase indicating that nitrite-dependent NO

441

production as well as NO detoxification continued even after cell growth had ceased.

442

This phenomenon was also reported for other nitrate ammonifiers (Bleakley & Tiedje,

443

1982; Schumacher & Kroneck, 1992).

444 445

Consumption of N2O

446

W. succinogenes cells were previously reported to grow by “N2O respiration” using

447

N2O as sole electron acceptor of anaerobic respiration (Yoshinari, 1980). In these

448

experiments the cells were gassed with pure N2O and grew to a modest optical

449

density with formate as electron donor. In contrast, our experiments describe N2O

450

turnover during respiratory nitrate ammonification. The results demonstrate that the

451

W. succinogenes cytochrome c nitrous oxide reductase (cNosZ) is able to convert

452

N2O to N2 under the applied conditions. This reaction is thought to play a minor role

453

in menaquinone replenishment since menaquinone is required for formate oxidation,

454

which is catalysed by an electrogenic, i.e. proton motive force (pmf)-generating,

18

455

membrane-bound formate dehydrogenase complex (Kern & Simon, 2009; Simon &

456

Klotz, 2013). In this way, N2O reduction is equivalent to formate-dependent nitrate

457

and nitrite reduction performed by the respiratory Nap and Nrf systems. In each case,

458

menaquinol oxidation seemingly does not contribute to the generation of a pmf in W.

459

succinogenes (Kern & Simon, 2009; Simon & Klotz, 2013). The precise architecture

460

of the electron transport chain that catalyses N2O reduction by menaquinol is not

461

known but a menaquinol dehydrogenase of the NapGH-type (named NosGH) might

462

be involved in order to direct electrons from the membrane to the periplasmic space

463

(Simon et al., 2004; Simon & Klotz, 2013). Subsequently, electrons are likely to be

464

transferred either directly or indirectly via the cytochrome c domain of cNosZ to the

465

copper-containing catalytic site of N2O reduction.

466

The specific rates of N2O reduction (Fig. 7) are very low compared to that of

467

regular denitrifying bacteria. In Paracoccus denitrificans, for instance, the maximum

468

specific rate of N2O reduction was found to be 2.5 fmol N2O cell-1 h-1 (Bergaust et al.,

469

2012), which is equivalent to 12.5 mmol N2O g-1 cell dry weight h-1 (dry weight per

470

cell = 2*10-13 g). This is two orders of magnitude higher than the maximum specific

471

rates estimated for W succinogenes kompPnosnosZ. This could be taken to suggest

472

that W. succinogenes is an insignificant sink for N2O in the environment. It remains to

473

be explored, however, to which extent the N2O reductase activity of the organism

474

could be enhanced by conditions other than those tested in the present study.

475

Conclusion

476

In recent years the view emerged that individual bacterial cells are capable to

477

simultaneously produce ammonium and dinitrogen from nitrate, albeit to different

478

extents (Sanford et al., 2012; Heylen & Keltjens, 2012, Mania et al., 2014). It remains

479

to be established, however, whether such cells indeed perform both types of 19

480

anaerobic respiration or if one reaction mainly results from detoxification. In this

481

context, the discrimination between detoxifying and respiratory processes is not

482

trivial, especially given the fact that quinol oxidation by nitrate, nitrite and N2O by the

483

respective Nap, Nrf and cNos systems is thought to be electroneutral and therefore

484

not directly involved in the generation of a pmf and hence ATP production (Simon &

485

Klotz, 2013). The results presented here suggest that N2O formation in W.

486

succinogenes resulted from NO detoxification and that N2O serves as an additional

487

electron sink during growth by respiratory nitrate ammonification.

488

Acknowledgements

489

Work in our labs is funded by the Deutsche Forschungsgemeinschaft (grant SI 848/4-

490

1 to J.S.) and the Norwegian Research Council.

491 492

REFERENCES

493

Baar, C., Eppinger, M., Raddatz, G., Simon, J., Lanz, O., Klimmek, O.,

494

Nandakumar, R., Gross, R., Rosinus, A., Keller, H., Jagtap, P., Linke, B.,

495

Meyer, F., Lederer, H. & Schuster, S.C. (2003). Complete genome sequence

496

and analysis of Wolinella succinogenes. Proc Natl Acad Sci USA 100, 11690-

497

11695.

498

Bergaust, L., Shapleigh, J., Frostegård, Ǻ. & Bakken, L.R. (2008). Transcription

499

and activities of NOx reductases in Agrobacterium tumefaciens: the influence of

500

nitrate, nitrite and oxygen availability. Environ Microbiol 10, 3070-3081.

501

Bergaust, L., van Spanning R.J.M., Frostegård Å. & Bakken L.R. (2012).

502

Expression of nitrous oxide reductase in Paracoccus denitrificans is regulated by

503

oxygen and nitric oxide through FnrP and NNR. Microbiology 158, 826-834.

20

504 505

Bleakley, B.H. & Tiedje, J.M. (1982). Nitrous oxide production by organisms other than nitrifiers or denitrifiers. Appl Environ Microbiol 44, 1342-1348.

506

Bode, C., Goebell, H. & Stähler, E. (1968). Zur Eliminierung von Trübungsfehlern

507

bei der Eiweißbestimmung mit der Biuretmethode. Z klin Chem klin Biochem 6,

508

418-422.

509

Bokranz, M., Katz, J., Schröder, I., Roberton, A.M. & Kröger, A. (1983). Energy

510

metabolism and biosynthesis of Vibrio succinogenes growing with nitrate or nitrite

511

as terminal electron acceptor. Arch Microbiol 135, 36-41.

512

Costa, C., Macedo, A., Moura, I., Moura, J.J.G., Le Gall, J., Berlier, Y., Liu, M.-Y.

513

& Payne, W.J. (1990). Regulation of the hexaheme nitrite/nitric oxide reductase

514

of Desulfovibrio desulfuricans, Wolinella succinogenes and Escherichia coli.

515

FEBS Lett 276, 67-70.

516

Dell’Acqua, S., Moura, I., Moura, J.J.G. & Pauleta, S.R. (2011). The electron

517

transfer complex between nitrous oxide reductase and its electron donors. J Biol

518

Inorg Chem 16, 1241-1254.

519 520 521

Einsle, O. (2011). Structure and function of formate-dependent cytochrome c nitrite reductase, NrfA. Meth Enzymol 496, 399-422. Gilberthorpe, N.J. & Poole, R.K. (2008). Nitric oxide homeostasis in Salmonella

522

typhimurium: roles of respiratory nitrate reductase and flavohemoglobin. J Biol

523

Chem 283, 11146-11154.

524

Giles, M., Morley, N., Baggs, E.M. & Daniell, T.J. (2012). Soil nitrate reducing

525

processes – drivers, mechanisms for spatial variation, and significance for nitrous

526

oxide production. Front Microbio 3: 407. doi:10.3389/fmicb.2012.00407.

527 528

Hartley, A.M. & Asai, R.J. (1963). Spectrophotometric determination of nitrate with 2,6-xylenol reagent. Anal Chem 35, 1207-1213.

21

529

Heylen, K. & Keltjens, J. (2012). Redundancy and modularity in membrane-

530

associated dissimilatory nitrate reduction in Bacillus. Front Microbio 3: 371.

531

doi:10.3389/fmicb.2012.00371.

532

Jones, C.M., Graf, D.R.H., Bru, D., Philippot, L. & Hallin, S. (2013). The

533

unaccounted yet abundant nitrous oxide-reducing microbial community: a

534

potential nitrous oxide sink. ISME J 7, 417-426.

535

Justino, M.C., Ecobichon, C., Fernandes, A.F., Boneca, I.G. & Saraiva, L.M.

536

(2012). Helicobacter pylori has an unprecedented nitric oxide detoxifying system.

537

Antioxid Redox Signal 17, 1190-1200.

538

Kaspar, H.F. & Tiedje, J.M. (1981). Dissimilatory reduction of nitrate and nitrite in

539

the bovine rumen: nitrous oxide production and effect of acetylene. Appl Environ

540

Microbiol 41, 705-709.

541

Kern, M. & Simon, J. (2009). Electron transport chains and bioenergetics of

542

respiratory nitrogen metabolism in Wolinella succinogenes and other

543

Epsilonproteobacteria. Biochim Biophys Acta 1787, 646-656.

544

Kern, M., Eisel, F., Scheithauer, J., Kranz, R.G. & Simon, J. (2010). Substrate

545

specificity of three cytochrome c haem lyase isoenzymes from Wolinella

546

succinogenes: unconventional haem c binding motifs are not sufficient for haem c

547

attachment by NrfI and CcsA1. Mol Microbiol 75, 122-137.

548

Kern, M., Klotz, M.G. & Simon, J. (2011a). The Wolinella succinogenes mcc gene

549

cluster encodes an unconventional respiratory sulphite reduction system. Mol

550

Microbiol 82, 1515-1530.

551

Kern, M., Volz, J. & Simon, J. (2011b). The oxidative stress defence network of

552

Wolinella succinogenes: cytochrome c nitrite reductase mediates the stress

553

response to nitrite, nitric oxide, hydroxylamine and hydrogen peroxide. Environ

554

Microbiol 13, 2478-2494. 22

555

Kröger, A., Geisler, V. & Duchêne, A. (1994). Isolation of Wolinella succinogenes

556

hydrogenase, Chromatofocusing. In A practical guide to membrane protein

557

purification. von Jagow, G. & Schägger, H. (eds). London: Academic Press, pp.

558

141-147.

559

Kröger, A., Biel, S., Simon, J., Gross, R., Unden, G. & Lancaster, C.R.D. (2002).

560

Fumarate respiration of Wolinella succinogenes: enzymology, energetics and

561

coupling mechanism. Biochim Biophys Acta 1553, 23-38.

562

Lorenzen J.P., Kröger A. & Unden G. (1993). Regulation of anaerobic respiratory

563

pathways in Wolinella succinogenes by the presence of electron acceptors. Arch

564

Microbiol 159, 477-483.

565

Mania, D., Heylen, K., van Spanning, R.J.M. & Frostegard, Ǻ (2014). The nitrate-

566

ammonifying and nosZ carrying bacterium Bacillus vireti is a potent source and

567

sink for nitric and nitrous oxides under high nitrate conditions. Environ Microbiol;

568

doi: 10.1111/1462-2920.12478.

569

Molstad, L., Dörsch, P. & Bakken, L.R. (2007). Robotized incubation system for

570

monitoring gases (O2, NO, N2O, N2) in denitrifying cultures. J Microbiol Meth 71,

571

202-211.

572 573 574

Pauleta, S.R., Dell’Acqua, S. & Moura, I. (2013). Nitrous oxide reductase. Coord Chem Rev 257, 332-349. Payne, W.J., Grant, M.A., Shapleigh, J. & Hoffman, P. (1982). Nitrogen oxide

575

reduction in Wolinella succinogenes and Campylobacter species. J Bacteriol 152,

576

915-918.

577

Pfennig, N. & Trüper, H.G. (1981). Isolation of members of the families

578

Chromatiaceae and Chlorobiaceae. In The Prokaryotes, pp. 279-289. Edited by

579

M.P. Starr, H. Stolp, H.G. Trüper, A. Balous, H.G. Schlegel. New York, Berlin,

580

Heidelberg: Springer Verlag. 23

581

Pomowski, A., Zumft, W.G., Kroneck, P.M.H. & Einsle, O. (2011). N2O binding at a

582

[4Cu:2S] copper-sulphur cluster in nitrous oxide reductase. Nature 477, 234-237.

583

Poole, R.K. (2005). Nitric oxide and nitrosative stress tolerance in bacteria. Biochem

584 585

Soc Trans 33, 176-180. Reay, D.S., Davidson, E.A., Smith, K.A., Smith, P., Melillo, J.M., Dentener, F. &

586

Crutzen, P.J. (2012). Global agriculture and nitrous oxide emissions. Nature

587

Clim. Change 2, 410-416.

588

Richardson, D.J., Berks, B.C., Russell, D.A., Spiro, S. & Taylor, C.J. (2001).

589

Functional, biochemical and genetic diversity of prokaryotic nitrate reductases.

590

Cell Mol Life Sci 58, 165-178.

591

Richardson, D., Felgate, H., Watmough, N., Thomson, A. & Baggs, E. (2009).

592

Mitigating release of the potent greenhouse gas N2O from the nitrogen cycle –

593

could enzymic regulation hold the key? Trends Biotechnol 27, 388-397.

594 595

Rider, B.F. & Mellon, M.G. (1946). Colorimetric determination of nitrite. Indust Engin Chem 18, 96-98.

596

Rowley, G., Henson, D., Felgate, H., Arkenberg A., Appia-Ayme, C., Prior, K.,

597

Harrington, C., Field, S.J., Butt, J.N., Baggs, E. & Richardson D.J. (2012).

598

Resolving the contributions of the membrane-bound and periplasmic nitrate

599

reductase systems to nitric oxide and nitrous oxide production in Salmonella

600

enterica serovar Typhimurium. Biochem J 441, 755-762.

601

Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989). Molecular cloning: a Laboratory

602

Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

603

Sanford, R.A., Wagner, D.D., Wu, Q., Chee-Sanford, J.C., Thomas, S.H., Cruz-

604

Garcia, C., Rodriguez, G., Massel-Deyá, A., Krishnani, K.K., Ritalahti, K.M.,

605

Nissen, S., Konstantinidis, K.T. & Löffler, F.E. (2012). Unexpected

24

606

nondenitrifier nitrous oxide reductase gene diversity and abundance in soils. Proc

607

Natl Acad Sci USA 109, 19709-19714.

608

Saraiva, L.M., Vicente, J.B. & Teixeira, M. (2004). The role of the flavodiiron

609

proteins in microbial nitric oxide detoxification. Adv Microb Physiol 49, 77-129.

610

Schumacher, W. & Kroneck, P.M.H. (1992). Anaerobic energy metabolism of the

611

sulfur-reducing bacterim “Spirillum“ 5175 during dissimilatory nitrate reduction to

612

ammonia. Arch Microbiol 157, 464-470.

613

Schumacher, W., Kroneck, P.M.H. & Pfennig, N. (1992). Comparative systematic

614

study on “Spirillum“ 5175, Campylobacter and Wolinella species. Arch Microbiol

615

158, 287-293.

616 617 618

Simon, J. (2002). Enzymology and bioenergetics of respiratory nitrite ammonification. FEMS Microbiol Rev 26, 285-309. Simon, J. & Kröger, A. (1998). Identification and characterization of IS1302, a novel

619

insertion element from Wolinella succinogenes belonging to the IS3 family. Arch

620

Microbiol 170, 43-49.

621

Simon, J. & Kern, M. (2008). Quinone-reactive proteins devoid of haem b form

622

widespread membrane-bound electron transport modules in bacterial anaerobic

623

respiration. Biochem Soc Trans 36, 1011-1016.

624

Simon, J. & Klotz, M.G. (2013). Diversity and evolution of bioenergetic systems

625

involved in microbial nitrogen compound transformations. Biochim Biophys Acta

626

1827, 114-135.

627 628 629

Simon, J. & Kroneck, P.M.H. (2013). Microbial sulfite respiration. Adv Microb Physiol 62, 45-117. Simon, J., Gross, R., Ringel, M., Schmidt, E. & Kröger, A. (1998). Deletion and

630

site-directed mutagenesis of the Wolinella succinogenes fumarate reductase

631

operon. Eur J Biochem 251, 418-426. 25

632

Simon, J., Gross, R., Einsle, O., Kroneck, P.M.H., Kröger, A. & Klimmek, O.

633

(2000). A NapC/NirT-type cytochrome c (NrfH) is the mediator between the

634

quinone pool and the cytochrome c nitrite reductase of Wolinella succinogenes.

635

Mol Microbiol 35, 686-696.

636

Simon, J., Pisa, R., Stein, T., Eichler, R., Klimmek, O. & Gross, R. (2001). The

637

tetraheme cytochrome c NrfH is required to anchor the cytochrome c nitrite

638

reductase (NrfA) in the membrane of Wolinella succinogenes. Eur J Biochem

639

268, 5776-5782.

640

Simon, J., Einsle, O., Kroneck, P.M.H & Zumft, W.G. (2004). The unprecedented

641

nos gene cluster of Wolinella succinogenes encodes a novel respiratory electron

642

transfer pathway to cytochrome c nitrous oxide reductase. FEBS Lett 569, 7-12.

643

Simon, J., van Spanning, R.J.M. & Richardson, D.J. (2008). The organisation of

644

proton motive and non-proton motive redox loops in prokaryotic respiratory

645

systems. Biochim Biophys Acta 1777, 1480-1490.

646

Simon, J., Kern, M., Hermann, B., Einsle, O. & Butt, J.N. (2011). Physiological

647

function and catalytic versatility of bacterial multihaem cytochromes c involved in

648

nitrogen and sulfur cycling. Biochem Soc Trans 39, 1864-1870.

649

Smith, K.A. (2010). Nitrous oxide and climate change. London, UK: Earthscan.

650

Stein, L.Y. (2011). Surveying N2O-producing pathways in bacteria. Meth Enzymol

651 652

486, 131-152. Streminska, M.A., Felgate, H., Rowley, G., Richardson, D.J. & Baggs, E.M.

653

(2012). Nitrous oxide production in soil isolates of nitrate-ammonifying bacteria.

654

Environ Microbiol Rep 4, 66-71.

655

Teraguchi, S. & Hollocher, T.C. (1989). Purification and some characteristics of a

656

cytochrome c-containing nitrous oxide reductase from Wolinella succinogenes. J

657

Biol Chem 264, 1972-1979. 26

658

Thomson, A.J., Giannopoulos, G., Pretty, J., Baggs, E.M. & Richardson, D.J.

659

(2012). Biological sources and sinks of nitrous oxide and strategies to mitigate

660

emissions. Phil Trans R Soc B 367, 1157-1168.

661

van Spanning, R.J.M. (2011). Structure, function, regulation and evolution of the

662

nitrite and nitrous oxide reductases: denitrification enzymes with a beta-propeller

663

fold. In Nitrogen Cycling in Bacteria, pp. 135-161. Edited by J.W.B. Moir. Norfolk,

664

UK: Caister Academic Press.

665

Wüst, A., Schneider, L., Pomowski, A., Zumft, W.G., Kroneck, P.M.H. & Einsle,

666

O. (2012). Nature’s way of handling a greenhouse gas: the copper-sulfur cluster

667

of purple nitrous oxide reductase. Biol Chem 393, 1067-1077.

668 669 670 671 672 673

Yoshinari, T. (1980). N2O reduction by Vibrio succinogenes. Appl Environ Microbiol 39, 81-84. Zumft, W.G. (1997). Cell biology and molecular basis of denitrification. Microbiol Mol Biol Rev 61, 533-616. Zumft, W.G. & Kroneck, P.M.H. (2007). Respiratory transformation of nitrous oxide (N2O) to dinitrogen by Bacteria and Archaea. Adv Microb Physiol 52, 107-227.

674

27

675

Figure legends

676

Fig. 1. Physical maps of nos gene loci in the genome of W. succinogenes cells

677

employed in this study. Black bars indicate DNA regions used for homologous

678

recombination during mutant construction (see Methods for details). Pnos indicates the

679

position of the nos promoter. The (putative) function of individual nos gene products

680

was discussed previously by Simon et al. (2004) and van Spanning (2011). The open

681

reading frame located between nosF and nosY encodes a hypothetical protein of 151

682

amino acid residues (WS0925) that has not yet been designated.

683

Fig. 2. Detection of cNosZ and NrfA in different W. succinogenes cells grown in

684

either formate-limited/nitrate-sufficient (80mM formate/50 mM nitrate, designated NS)

685

or formate-sufficient/nitrate-limited (80 mM formate/10 mM nitrate, designated NL)

686

medium (see text for details). Cells were harvested in the late exponential growth

687

phase and whole cell extracts were subjected to denaturing SDS polyacrylamide gel

688

electrophoresis (100 µg protein was applied to each lane). Subsequently,

689

cytochromes c were detected by haem staining. Arrows indicate cNosZ and NrfA

690

proteins (Kern et al., 2010). M, size marker proteins.

691

Fig. 3. Growth of W. succinogenes cells and concomitant nitrate and nitrite

692

concentrations in the medium. Cells were grown at 37°C in sealed 120 ml vials (50

693

ml medium). A. Formate-limited, nitrate-sufficient medium (80 mM formate and 50

694

mM nitrate). B. Formate-sufficient, nitrate-limited medium (80 mM formate and 10

695

mM nitrate). Representative experiments are shown (performed in triplicate; deviation

696

less than 5%). () optical density, () nitrate concentration, () nitrite concentration.

697

Fig. 4. Measured NO, N2O and N2 during growth of three W succinogenes organisms

698

in nitrate-sufficient medium. N2 is the cumulative N2 production (corrected for leaks 28

699

and sampling dilution). Representative experiments are shown (performed in

700

triplicate; deviation less than 5%). () nitric oxide (nmol vial-1), () nitrous oxide

701

(µmol vial -1), () dinitrogen (µmol vial -1).

702

Fig. 5. Production of NO, N2O and N2 by chemical reactions (control experiments

703

related to data shown in Fig. 3). Sterile nitrate-sufficient medium was incubated in the

704

presence (upper panel) or absence (lower panel) of 0.5% (w/v) BHIB. The indicated

705

amounts of nitrite were added. Average values from three independent experiments

706

are presented. The figure shows measured NO (), N2O () and N2 (). Error bars

707

indicate the standard error of the mean.

708

Fig. 6. Concentrations of N2O and N2 in W. succinogenes cultures with nitrous oxide

709

added to the headspace. The experiment corresponds to that shown in Fig. 4 but 1%

710

N2O was injected into the headspace prior to inoculation. Representative

711

experiments are shown (performed in triplicate). (, red) Wild-type in nitrate-

712

sufficient medium, (, red) Wild-type in nitrate-limited medium, (, green) mutant

713

ΔnosZ in nitrate-sufficient medium, (, green) mutant ΔnosZ in nitrate-limited

714

medium, (, blue) mutant kompPnosnosZ in nitrate-sufficient medium, (, blue)

715

mutant kompPnosnosZ in nitrate-limited medium.

716

Fig 7. Estimated specific cellular rates of N2 production throughout the batch

717

cultivation of W. succinogenes kompPnosnosZ in nitrate-sufficient medium with or

718

without either NO or N2O in the headspace at the onset of cultivation. (, red) and

719

(, red) NO added (two individual vials), (, blue), (, blue) and (, blue) N2O

720

added (three individual vials), (, green), no NO or N2O added (for clarity only one

721

representative vial, out of three, is shown). Details of the experiments with added NO

722

are shown in Fig. S7 and S8. dw, dry weight.

29

Fig. 1

W. succinogenes wild-type (DSM 1740T) Pnos

nssC EcoRI

nosZ‘

IS1302

nosZ‘

nosB BamHI

BamHI

nosD

nosG C1 C2

nosH

F

Y

L

HindIII

1 kbp

W. succinogenes ΔnosZ Pnos

nssC EcoRI

cat

nosB

nosD

nosG C1

C2

nosH

F

Y

L

nosG C1

C2

BamHI BamHI

HindIII

W. succinogenes kompPnosnosZ Pnos

nssC

Pnos

nosZ

kan

nosB

nosD

nosH

F

Y

L

Fig. 2

Wild-type NS

NL

kompPnosnosZ

ΔnosZ M

M

NS

NL

M

100 kDa

100 kDa

70 kDa

70 kDa

55 kDa 40 kDa 35 kDa 25 kDa

55 kDa 40 kDa 35 kDa 25 kDa

NS

NL cNosZ

(95 kDa)

NrfA

(58 kDa)

Fig. 3

A

B

Wild-type

ΔnosZ

kompPnosnosZ

Fig. 4

Wild-type

ΔnosZ

kompPnosnosZ

Fig. 5

Fig. 6

N2O (µmol/vial)

45

40

35

30 0

5

10

15

20

25

20

25

time (h)

N2 (µmol/vial)

10 8 6 4 2 0 0

5

10

15

time (h)

specific rate of N2 production (µmol/g dw ⋅ h)

Fig. 7

120

100 80

60

40

20

0 0 5 10

time (h)

15 20