Electron Shuttles Enhance Anaerobic Ammonium Oxidation Coupled

Subscriber access provided by Northern Illinois University

Article

Electron Shuttles Enhance Anaerobic Ammonium Oxidation Coupled to Iron(III) Reduction Guowei Zhou, Xiaoru Yang, Hu Li, Christopher W Marshall, Bangxiao Zheng, Yu Yan, Jianqiang Su, and Yongguan Zhu Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b02077 • Publication Date (Web): 05 Aug 2016 Downloaded from http://pubs.acs.org on August 5, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33

Environmental Science & Technology

1

Electron Shuttles Enhance Anaerobic Ammonium Oxidation

2

Coupled to Iron(III) Reduction

3

Guo-Wei Zhou,†,‡ Xiao-Ru Yang,* , † Hu Li,†,‡ Christopher W. Marshall,¶,ǁ Bang-Xiao

4

Zheng,†,‡ Yu Yan,†,‡ Jian-Qiang Su,† and Yong-Guan Zhu†,§

5

†

6

Academy of Sciences, Xiamen 361021, People’s Republic of China

7

‡

8

China

9

§

Key Lab of Urban Environment and Health, Institute of Urban Environment, Chinese

University of Chinese Academy of Sciences, Beijing 100049, People’s Republic of

State Key Lab of Urban and Regional Ecology, Research Center for

10

Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing 100085,

11

People’s Republic of China

12

¶

13

ǁ

14

60439, U.S.A.

Department of Surgery, University of Chicago, Chicago IL 60637, U.S.A

Biosciences Division, Argonne National Laboratory, 9700, S. Cass Ave. Lemont, IL

15 16

*Correspondence: Xiaoru Yang, 1799 Jimei Road, Xiamen, 361021, China

17

E-mail: [email protected]

18

Phone: +86-592-6190560

19

Fax: +86-6190977

1

ACS Paragon Plus Environment


Page 2 of 33

20

ABSTRACT: Anaerobic ammonium oxidation coupled to iron(III) reduction, termed Feammox,

21

is a newly discovered nitrogen cycling process. However, little is known about the roles of

22

electron shuttles in the Feammox reactions. In this study, two forms of Fe(III) (oxyhydr)oxide

23

ferrihydrite (ex situ ferrihydrite and in situ ferrihydrite) were used in the dissimilatory Fe(III)

24

reduction (DIR) enrichments from paddy soil. Evidence for Feammox in DIR enrichments was

25

demonstrated using 15N-isotope tracing technique. The extent and rate of both the 30N2/29N2 and

26

Fe(II)

27

9,10-anthraquinone-2,6-disulfonate (AQDS) or biochar), and further simulated when combining

28

these two shuttling compounds. Although the Feammox-associated Fe(III) reduction was

29

accounted for only a minor proportion of total Fe(II) formation compared to DIR, it was

30

estimated that the potentially Feammox-mediated N loss (0.13-0.48 mg N L-1 d-1) was increased

31

by 17-340% in the enrichments by addition of electron shuttles. The addition of electron shuttles

32

led to increase the abundance of unclassified Pelobacteraceae, Desulfovibrio and denitrifiers but

33

decrease the Geobacter. Overall, we demonstrated a stimulatory effect of electron shuttles on

34

Feammox that led to higher N loss, suggesting that electron shuttles might play a crucial role in

35

Feammox-mediated N loss from soils.

formation

were

enhanced

when

amended

with

1


electron

shuttles

(either

Page 3 of 33

36


TOC Art

37

2



38

INTRODUCTION

39

Nitrogen (N) is one of the limiting factors for primary productivity in terrestrial ecosystems.1-3

40

N fertilizers have been applied intensively in paddy soils to promote food production for decades,

41

but the excessive application of N fertilizers has decreased the N-utilization efficiency. The

42

decrease in efficiency leads to increased N loss to the environment, causing many environmental

43

problems such as air and water pollution.2,4-6 Several anaerobic pathways contribute to N loss in

44

terrestrial ecosystems, including denitrification, codenitrification and anaerobic ammonium

45

oxidation (anammox).2,3 It was recently reported that anammox could be coupled to iron(III)

46

reduction (termed Feammox) with N2, NO2-, or NO3- as the end-products.4-7 The following three

47

equations and accompanying free energies summarize these three end routes of Feammox.4-7

48

3Fe(OH)3 + 5 H+ + NH4+ → 3Fe2+ + 9H2O + 0.5N2 ∆rGm = -245 kJ mol-1 (1)

49

6Fe(OH)3 + 10 H+ + NH4+ → 6Fe2+ + 16H2O + NO2- ∆rGm = -164 kJ mol-1 (2)

50

8Fe(OH)3 + 14 H+ + NH4+ → 8Fe2+ + 21H2O + NO3- ∆rGm = -207 kJ mol-1 (3)

51

The first equation could occur energetically in a wide pH range while the second and third

52

reactions consume more Fe(III) but conserve less energy and exist only below pH 6.5.5

53

Feammox has been detected in freshwater, marine, wetland, tropical forest, paddy soils, and

54

even wastewater ecosystems.7-9 It has been demonstrated that Feammox coupled to direct

55

reduction to N2 is the predominant mechanism of the three possible microbial pathways

56

responsible for gaseous N loss from tropical forests, paddy soils, and intertidal wetlands.7-9

57

Conservative estimation indicated that Feammox metabolizes 7.8-61 (3.9-31% of N fertilizer

58

loss) and 1-4 kg NH4+-N ha-1 y-1 in paddy soil and upland surface soil, respectively.4,5 Thus, due

59

to the abundance of Fe(III) (oxyhydr)oxides in most soils and the detrimental effects of N2 loss 3


Page 4 of 33

Page 5 of 33


60

through Feammox, it is important to study the mechanism of iron(III) reduction coupled to

61

ammonium oxidation by Feammox bacteria.

62

In addition to direct contact, bacterial nanowires and Fe(III) complexation, electron shuttling

63

also has been suggested to contribute to Fe(III) reduction.10-14 Electron shuttles, including humic

64

substances and biochar, have been reported to facilitate electron transfer between a wide range

65

of microbes and Fe(III) (oxyhydr)oxides, enhancing microbial iron(III) reduction and Fe cycling

66

in soils.10-14 In many electron shuttles, quinone moieties are the redox-active components for

67

electron transfer.10-14 Humic substances are quinone-rich and ubiquitous in the environment,

68

accounting for up to 10% in soils.15 Biochar also possesses quinone structures. It has been

69

demonstrated to enhance soil fertility and to be an effective electron shuttle for iron(III)

70

reduction.16 We thus hypothesize that humic substances and biochar act as electron shuttles to

71

facilitate the transfer of electrons from NH4+ toward solid Fe(III) (oxyhydr)oxides and that this is

72

mediated by Feammox bacteria, ultimately leading to an increase in N loss from the system.

73

Published methods for 15N/14N analysis of N2 have been reported as a powerful tool to track the

74

N cycle and pathways in the study of Feammox.4-6 Culture studies have shown that

75

be produced through four different pathways in the anoxic environment.4,5 Under these

76

conditions,

77

anammox, denitrification or chemodenitrification-based denitrification (Table 1).4,5,17,18 Also,

78

several pathways detailed in the following table make contribution to 29N2 production, but all of

79

the pathways first required the anaerobic oxidation of ammonium.

30

N2 are directly from Feammox or Feammox-generated

4


15

30

N2 could

NO2-/15NO3- followed by


Table 1 Possible pathways for

30

N2 and

29

N2 production from

Page 6 of 33

15

NH4+ under anoxic conditions

(Modified from Ding et al. and Yang et al.,)4,5 Product

30

29

Nitrogen 1 15

NH4+ (added)

15

NH4+ (added)

15

15

Nitrogen 2 15

Process

NH4+ (added)

Feammox to N2

15

NO2-/15NO3- (Feammox)

Anammox


15


Denitrification17,19


15


Chemodenitrification18,20 and Denitrification

N2

N2

15

NH4+ (added)

15

NH4+ (added)

15


15


15


14

NH4+ (indigenous)

Feammox to N2

14

NO2-/14NO3- (indigenous)

Anammox

14


Denitrification Codenitrification17,19

amino compounds (indigenous) 14


Chemodenitrification18,20 and Denitrification

80 81

To address this hypothesis, enrichments for dissimilatory iron(III)-reducing bacteria (DIRB)

82

with poorly crystalline amorphous ferrihydrite as the electron acceptor were established. Two

83

forms of ferrihydrite (ex situ ferrihydrite and in situ ferrihydrite) produced by different

84

procedures were used. Quinone compound 9,10-anthraquinone-2,6-disulfonate (AQDS), which

85

is a model compound for quinone moieties in humic substances, and biochar were added to the

86

enrichment cultures to investigate the impact of electron shuttles on Feammox. By using

87

15

88

main objectives of this study were to: (i) demonstrate the existence of microbially-mediated 5

NH4+-based isotopic tracing technique and 16S rRNA gene based amplicon sequencing, the


Page 7 of 33


89

Feammox, (ii) quantitatively determine the impact of electron shuttles on the extent and rate of

90

Feammox in different iron(III)-reducing enrichments, and (iii) characterize the shift in

91

Feammox-related bacterial community and confirm the microbial role in Feammox.

92 93

MATERIALS AND METHODS

94

Characterization of Biochar. Biochar used in experiments was produced from air-dried rice

95

stalks via slow pyrolysis at 500°C for 4 h in a muffle furnace (Isotemp, Fisher Scientific, USA)

96

purged with N2. Biochar was sieved (pore size 0.15 mm) and washed for three times with

97

deionized water before application. The basic properties of biochar are shown in Table S1 of the

98

supporting information.

99

DIRB Enrichments Cultivation. Paddy soil was collected from Yingtan (116°82′ N, 28°2 ′

100

E), Jiangxi Province, China. It is a typical acid red soil containing high level of Fe(III)

101

(oxyhydr)oxide but low level of organic carbon in South China, which may be readily for

102

iron(III)-reducing bacteria enrichment. The basic properties of soil were detailed in

103

Supplementary Information (Table S2). Anaerobic incubations were prepared by mixing 3 g

104

fresh soil into 50 mL anoxic distilled water with shaking at 120 rpm for 2 h at 25°C. Aliquots (2

105

mL) of the soil slurry were transferred into 50 mL serum vials with 20 mL sterilized

106

(autoclaved, 120 °C for 20 min) anoxic medium and incubated at 25°C in the dark. The basal

107

medium (pH 6.8-7.2) contains MgCl2·6H2O (0.4 g L−1), CaCl2·H2O (0.1 g L−1), NH4Cl (0.027 g

108

L−1), and KH2PO4 (0.6 g L−1), 1 ml L-1 vitamin solution,21 1 ml L-1 trace element solutions,21 and

109

30 mmol L-1 bicarbonate buffer. Ferrihydrite and acetate were added at final concentrations of

110

10 mmol L-1 and 2 mmol L-1, respectively. The headspace of the medium was flushed with 6



Page 8 of 33

111

N2/CO2 (80/20%). In our study, two forms of ferrihydrite, ex situ ferrihydrite and in situ

112

ferrihydrite, were used. The amount of ex situ ferrihydrite added to cultures was calculated using

113

the formula of Fe5HO8·4H2O which was synthesized according to Schwertmann and Cornell.22,23

114

The in situ ferrihydrite was formed by adding iron(III) chloride to the medium and adjusting the

115

pH of the medium to 6.8-7.2. The basal media, in situ ferrihydrite and ex situ ferrihydrite were

116

autoclaved (120°C for 20 min) before inoculation. The vitamin solution, trace element solution

117

and acetate from stock solutions were filtrated with 0.22 µm filter and added into the sterilized

118

media, respectively.

119

Experimental Setup. To investigate the effect of electron shuttles on Feammox, six

120

treatments (n=3, each) were set up: (1) abiotic treatment inoculated with sterilized (autoclaved,

121

120 °C for 20 min) DIRB inoculum (named as CK abiotic ); (2) biotic treatment inoculated with

122

alive DIRB inoculum (CK biotic); (3) biotic treatment amended with 500 µmol L-1AQDS

123

(AQDS biotic); (4) biotic treatment amended with 2.5 g L-1 biochar (biochar biotic); (5) biotic

124

treatment amended with 500 µmol L-1AQDS and 2.5 g L-1 biochar simultaneously (AQDS +

125

biochar biotic); (6) abiotic treatment amended with 500 µmol L-1AQDS and 2.5 g L-1 biochar

126

simultaneously (AQDS + biochar abiotic ). The DIRB enrichments were transferred (10%, v/v)

127

to fresh medium monthly for four generations before the start of the 6 treatments. The isotope

128

tracing experiments were initiated by inoculating 2% (v/v) well-mixed DIRB enrichment

129

cultures into 20 mL fresh medium, in which the

130

L-1;

131

solution was omitted to avoid unwanted electron transfer reactions. The headspace of the labeled

132

medium was flushed with ultra-pure helium instead of nitrogen gas. The biochar was washed

15

14

NH4Cl was replaced by

15

NH4Cl (0.5 mmol

N, 99 atom%; Cambridge Isotope Laboratories, Andover, MA, USA), and the vitamin

7


Page 9 of 33


133

before use, and AQDS solution was filtrated with 0.22 µm filter from stock solution prior to the

134

addition.

135

Chemical Analyses. Ferrous iron [Fe(II)] and total Fe were measured as described by

136

Klueglein & Kappler.24 Briefly, for Fe(II) determination, 100 µL of culture suspension was

137

transferred anaerobically with a syringe into 900 µL of 40 mmol L-1 sulfamic acid for 1 h

138

incubation at room temperature. Total Fe was extracted with 20 mmol L-1 hydroxylamine

139

hydrochloride in 20 mmol L-1 sulfamic acid. A 100 µL of extract was transferred to 10 mL

140

ferrozine solution (1 g ferrozine in 50 mmol L-1 HEPES buffer, pH=7) to form the ferrous

141

complex. The complex was quantified at 562 nm UV/Vis spectrometer. Iron(III) reduction rates

142

were calculated from the linear change in Fe(II) concentrations between two given time points.

143

The concentrations of acetate, NH4+, NO3- and NO2- were measured by ion chromatography

144

(Dionex ICS-3000 system, Diones, Sunnyvales, CA, USA). Liquid samples were taken in the

145

anoxic glovebox and filtered through 0.22 µm filters. pH was determined with a Dual Channel

146

pH/Ion/Conductivity/DO Meter (X60, Fisher Scientific). Total organic carbon (TOC) was

147

determined with a TOC analyzer (Shimadzu TOC-Vcph, Japan) with a solid sample module

148

(SSM-5000A), and C, N and S with an elemental analyzer (Vario MAX CNS, Germany).

149

Extractable major and trace metal concentrations were digested using a strong acid digestion

150

method,25 and measured with ICP-OES (Optima, 7000DV; Perkin Elmer) and ICP-MS (Agilent

151

7500 cx; Agilent Technologies, Inc., Tokyo, Japan).

152

Headspace N2 concentration was analyzed using a robotized sampling and analyzing system

153

(GC-TCD 7890, Aglient Technologies. USA).26 For analysis of 15N-N2, 1 mL gas samples were

154

collected every three days by gastight syringes and then injected into 12 mL glass vials 8



155

(Exetainer, Labco, U.K) full with ultrapure helium. The ratio of 15N in total N2 was measured by

156

GC-isotope ratio mass spectrometry (Thermo Finnigan Delta V Advantage, Bremen, Germany)

157

as described previously.27 15N-N2 concentration was calculated as the product of N2 and 15N-N2

158

atom % excess above its natural abundance. Gas production rate was calculated from the linear

159

change of gas concentrations in the headspace between two given time points.

160

Bacterial 16S rRNA Gene Amplification, Illumina Sequencing and Data Processing.

161

After a 30-day incubation, FastDNA Spin Kit (MP Biomedical, France) was used to extract

162

DNA from the samples (harvested by centrifugation; 14000 g, 15 min) according to the

163

manufacturer’s protocol. The V4-V5 regions of bacterial 16S rRNA genes were amplified using

164

the DNA extracted from the samples as template. The forward primer was 515F

165

(5’-GTGCCAGCMGCCGCGG-3’), and the reverse primer consisted of a 6-bp barcode and

166

907R (5’- CCGTCAATTCMTTTRAGTTT -3’).28 The amplicons were purified, quantified,

167

pooled and then sequenced on an Illumina Miseq PE 250 platform at Novogene, Beijing,

168

China.29 Sequences were processed and analyzed using Quantitative Insights Into Microbial

169

Ecology (QIIME).30 The open-reference operational taxonomic unit (OTU) picking was

170

performed after removing any low quality or ambiguous reads according to the online

171

instruction of QIIME.31 The OTU was defined at 97% similarity level using UCLUST

172

clustering.31 The most abundant sequence from each OTU was selected as the representative

173

sequence, which was assigned to taxonomy using an RDP classifier.31 The differences between

174

microbial communities were investigated by non-metric multidimensional scaling (NMDS),

175

which was based on weighted UniFrac dissimilarity among samples.32,33 The ordination axes

176

explain variance in the dissimilarities.33 9


Page 10 of 33

Page 11 of 33


177

Statistical Analyses. Statistical tests, which included analysis of variance (ANOVA) and

178

Pearson correlation analysis, were performed using SPSS 18.0 (SPSS Inc, Chicago, IL, USA)

179

and Origin 9.0 (Inc., OringinLab, USA). Statistical significance was determined by Duncan’s

180

multiple range test and denoted at P < 0.05.

181 182

Data Accessibility. The 16S rRNA gene sequences have been deposited in GenBank with accession number SRX1618418.

183 184

RESULTS

185

30

N2/29N2 Production and Total NH4+ Removal, NO3-/NO2- Formation from DIRB 30

186

Enrichments. The concentrations of

187

15.48 µmol L-1 and 5.32 µmol L-1, which represented a 319% and 48% increase compared to the

188

non-amended controls in the ex situ ferrihydrite (3.69 µmol L-1 in the CK biotic) and in situ

189

ferrihydrite enrichments (3.59 µmol L-1 in the CK biotic), respectively (Figure 1A, 1D and Table

190

S3). After biochar addition,

191

non-amended controls after a 30-day incubation in ex situ ferrihydrite and in situ ferrihydrite

192

enrichments, respectively (Figure 1A, 1D and Table S3). AQDS with biochar also significantly

193

(P < 0.05) stimulated the rate of 30N2 production extent in both enrichments (Figure 1C and 1F).

194

Moreover, the extent and rate of

195

and biochar was significantly (P < 0.05) higher than those amended with AQDS or biochar alone

196

in the in situ ferrihydrite enrichment, but not in the ex situ ferrihydrite enrichment (Figure 1A,

197

1C, 1D, 1F and Table S3). In comparison with

198

formation was observed in all biotic treatments (Figure 1B and 1E). The ratios of 30N2 and 29N2

30

N2 in the treatment of AQDS reached a maximum of

N2 production extent increased 17% and 94% compared to

30

N2 production in the enrichment amended with both AQDS

30

N2 accumulation, greater

10


29

N2 (14N15N)


199

ranged from 2.9-7.8 in the enrichments (Table S4). The kinetics of 29N2 production extents and

200

rates shared similar trends with that of 30N2 production in the both enrichments (Figure 1B, 1C,

201

1D, 1E and Table S3). In addition, amendment with AQDS had a much greater impact on the

202

extent and rate of N2 production than that of biochar amendment in the ex situ ferrihydrite

203

enrichment, while biochar showed greater enhancement than AQDS in the in situ ferrihydrite

204

enrichment (Figure 1 and Table S3).

205

The concentration of total NH4+ decreased, resulting in larger NH4+ consumption when

206

amended with electron shuttles (210-300 µmol L-1) compared with the CK setups (150-160 µmol

207

L-1) in both the ferrihydrite enrichments after 30 days of incubation (Figure S1). The NO3-/NO2-

208

concentrations also decreased (1-13 µmol L-1 for NO3-; 14-22 µmol L-1 for NO2-) in all the biotic

209

setups in ex situ ferrihydrite enrichment (Figure S1). However, for the in situ ferrihydrite

210

enrichments, the NO3-/NO2- concentrations was decreased (1.5-6.7 µmol L-1 for NO3-; 30-32

211

µmol L-1 for NO2-) in the AQDS amendment, but increased (1-15 µmol L-1 for NO3-; 15-22 µmol

212

L-1 for NO2-) in the biochar/biochar and AQDS amendments (Figure S1).

Page 12 of 33

213

Iron(III) Reduction in DIRB Enrichments. The final concentration of ferrous iron(II) was

214

increased by 16% and 54% in the AQDS amendments compared to the CK biotic after a 30-day

215

incubation in the ex situ ferrihydrite and in situ ferrihydrite enrichments, respectively (Figure 2A,

216

2C and Table S3). Biochar also significantly (P < 0.05) increased (by 130% and 100% in the ex

217

situ ferrihydrite and in situ ferrihydrite enrichments, respectively) iron(III) reduction in both

218

enrichments (Figure 2A, 2C and Table S3). Moreover, a further increase in ferrous iron

219

concentration was found in the treatment of AQDS and biochar addition in both DIRB

220

enrichments (Figure 2A, 2C and Table S3). The average rate of iron(III) reduction was also 11


Page 13 of 33


221

facilitated by amending with AQDS and/or biochar, showing a similar trend in both enrichments

222

(Figure 2B and 2D). Additionally, the extent and rate of iron(III) reduction were increased more

223

significantly by biochar addition than those by AQDS addition in both enrichments (Figure 2A,

224

2B, 2C and 2D). Total extractable Fe in all the treatments remained constant throughout the

225

experiments (Figure S2). Sterilized controls showed low levels of abiotic Fe(III) reduction,

226

which accounted for 6-17.5% and 12.5-30.4% of total Fe(III) in the treatments of CK abiotic and

227

(AQDS + biochar) abiotic, respectively (Figure 2A and 2C).

228

A linear and positive correlation (P < 0.0001) was found between the production rate of 30N2 29

229

and

N2 and the rate of iron(III) reduction among all biotic treatments in the two enrichments

230

(Figure 3). The slope of lines related to

231

production in the both the ferrihydrite enrichments (Figure 3). The slope of all treatments in the

232

in situ ferrihydrite enrichment (AQDS biotic > AQDS + biochar biotic > CK biotic > biochar

233

biotic) exhibited a distinct trend from the ex situ ferrihydrite enrichment (AQDS + biochar

234

biotic > biochar biotic > CK biotic > AQDS biotic) (Figure 3C and 3D).

30

N2 production shared a similar trend with

29

N2

235

Proportion of Iron(III) Reduction Associated with Feammox and or dissimilatory

236

iron(III) reduction. According to the three equations 1-3, oxidation of one mole of NH4+ could

237

be coupled with reduction of 3-8 moles of Fe(III) by microbes. On the basis of this ratio, it was

238

estimated that 0.4-6.1% of Fe(III) reduction was associated with Feammox in all biotic

239

ferrihydrite enrichments (Table S5). The major proportion (57.5-85.8%) of produced Fe(II) was

240

coupled to organic matter (acetate et al.,) oxidation (Table S5).

241

Microbial Community in DIRB Enrichments. After 30 days, DNA only obtained from

242

biotic treatments. The number of OTU varied from 965 ± 76 to 6280 ± 408 per sample, while the 12



243

sequence number ranged from 59784 to 146120 (Table S6). Non-metric multidimensional

244

scaling (NMDS) analysis showed that microbial community compositions in the CK biotic,

245

electron shuttle amendments and YT (intact soil) were distinctly clustered (stress value = 0.18),

246

respectively (Figure 4A), suggesting that the bacterial community varied with the process of the

247

DIRB enrichment and electron shuttle addition. In order to understand the observed clustering of

248

the microbial communities, the 28 most abundant genera were selected from each treatment in

249

both enrichments. In the presence of Fe(III) (oxyhydr)oxides, the proportion of Geobacter was

250

significantly (P < 0.05) increased compared to the intact soil, and accounted for 69%-88% of

251

total bacterial community in both shuttles-free enrichments (CK) (Figure 4B). Whereas, after

252

amendment with AQDS and/or biochar, a significant (P < 0.05) decrease in Geobacter but

253

increase in the unclassified Pelobacteraceae was observed in both enrichments (Figure 4B).

254

Furthermore, amendment with biochar significantly (P < 0.05) enhanced the abundance of

255

Desulfovibrio (13% and 20% in the ex situ ferrihydrite and in situ ferrihydrite enrichments,

256

respectively) (Figure 4B). Additionally, only a minor proportion of other genera, including

257

Clostridium, Desulfosporosinus and Pseudomonas, were detected out of total bacterial

258

community in all treatments (Figure 4B).

259 260

DISCUSSION

261

Evidence for the Occurrence of Feammox in the Ferrihydrite Enrichment. After adding

262

labeled 15NH4+ into the DIRB enrichments, significant 30N2 accumulation in both the ferrihydrite

263

enrichments provided evidence for the occurrence of Feammox (Figure 1). To avoid aerobic

264

nitrification during incubations, all operations were performed under strict anoxic conditions 13


Page 14 of 33

Page 15 of 33


265

throughout the experiments. In addition, codenitrification, which is another potential source of

266

30

N2 and co-occurs with denitrification, can be ruled out in this study because no other

267

15

N-labeled nitrogen compounds (including hydrazine and amino compounds) that could reduce

268

15

NO2- and

269

medium (about 20:1) was sufficient to drive the reaction of Feammox that produced N2, NO2-, or

270

NO3- according to the equations 1-3. The positive correlation between the rate of 30N2 production

271

and Fe(III) reduction provided further proof for the existence of Feammox in both DIRB

272

enrichments (Figure 3). Hence, the accumulation of

273

occurrence of Feammox, since Feammox played an indispensable role in 30N2 production based

274

on the potential pathways introduced in Table 1 in our incubation system.

275

accumulated in these two DIRB enrichments, indicating that the medium was well-mixed

276

(Figure 1).5 The 15N atom of 29N2 was undoubtedly from the labeled 15NH4+ while the 14N was

277

potentially derived from the DIRB inoculum or the 14NH4+ contained by the 99 atom%-15NH4Cl.

278

29

279

pathways first required the anaerobic oxidation of ammonium. Moreover, the

280

rate was significantly correlated with the iron(III) reduction rate (P < 0.0001), which further

281

demonstrated the existence of Feammox in DIRB enrichments leading to more NH4+

282

consumption (Figure 3). The previous studies have shown that substrate molecules containing

283

the lighter 14N atom are typically consumed at higher rates than the heavier 15N atom due to the

284

isotopic fraction effect.34-36 The higher concentration of 29N2 production than 30N2 suggested that

285

preferential conversion of 14N-NH4Cl and 14N-NOx to N2 in both the ferrihydrite enrichments.

15

NO3- were available in the enrichments.2,4 The ratio of Fe(III) to NH4+ in the

30

N2 provided a solid evidence for the

29

N2 was also

N2 could be produced through a variety of combinations detailed in Table 1, but all of the

14


29

N2 production


286

Potential Feammox rates were conservative estimates calculated from

Page 16 of 33

30

N2 production rate

287

alone.5 It was estimated that Feammox rates were 0.11 mg N L-1 d-1 in both DIRB enrichments,

288

which were comparable to those calculated in tropical soils (approximately 0.32 mg N L-1 d-1),

289

intertidal wetlands (0.24-0.36 mg N L-1 d-1), and paddy soils (0.17-0.59 mg N L-1 d-1).4-6

290

Effect of Electron Shuttles on Feammox. Amendment of electron shuttles (AQDS and

291

biochar) in DIRB enrichments significantly stimulated the rates and extent of Feammox

292

(Feammox rate were 0.13-0.46 mg N L-1 d-1) and Fe(II) production (Figure 1 and 2). Moreover,

293

the effect on Feammox (Feammox rate were 0.28-0.48 mg N L-1 d-1) and Fe(II) production rate

294

and extent were further enhanced when both AQDS and biochar were added (Figure 1 and 2). It

295

was likely ascribed to the accumulation of quinone compounds contained by AQDS and biochar,

296

also, the capacity of biochar to adsorb the Fe(II) might lead to a faster iron(III) reduction rate in

297

the ferrihydrite enrichments.16,51 This further indicated the impact of electron shuttles on

298

Feammox-based N loss from the enrichments. Taking both 29N2 and 30N2 into consideration, the

299

utilization of

300

shuttles were added. These results suggested that electron shuttles played a pivotal role in

301

anaerobic NH4+ oxidation, a fact that is often overlooked in the estimation of N loss in

302

humic-rich environments. The trend of total NH4+ consumption further supported this conclusion

303

(Figure S1). In addition, although biochar was reported to be a useful additive to agricultural

304

land to modify soil quality and increase crop yields,25 it should be carefully considered given

305

that biochar application may cause higher Feammox-mediated N loss. The properties of biochar

306

are affected by the feedstock, treatment temperature and even particle sizes.37-40 Interestingly, no

307

significant enhancement in rate and extent of Feammox and iron(III) reduction was observed in

15

NH4+ was increased 4.1-11.5% in the two DIRB enrichments when electron

15


Page 17 of 33


308

the ex situ ferrihydrite enrichment amended with larger particle size (2 mm) of biochar (data not

309

shown). Therefore, more studies will still be needed to better understand the effect of different

310

properties of biochar on Feammox.

311

The pH was ranged from 6.8–7.2 during the experiment (Figure S3), suggesting that the direct

312

oxidation of ammonium to N2 is the only feasible pathway according to the theoretical

313

equations.

314

However, the change in the concentrations of NO3- and NO2- and headspace of N2O production

315

indicated that Feammox-dependent denitrification and chemodenitrification may make a

316

contribution to the N2 production (Figure S1 and S4). This may be due to the difference between

317

the theoretical reactions and actual reactions. The amount of total N2O was significantly (P

Electron Shuttles Enhance Anaerobic Ammonium Oxidation Coupled

Recommend Documents