Mineralization and nitrification: Archaea dominate ammonia-oxidising communities in grassland soils

Journal Pre-proof Mineralization and nitrification: Archaea dominate ammonia-oxidising communities in grassland soils Dave R. Clark, Boyd A. McKew, Liang F. Dong, Garwai Leung, Alex J. Dumbrell, Andrew Stott, Helen Grant, David B. Nedwell, Mark Trimmer, Corinne Whitby PII:

S0038-0717(20)30022-5

DOI:

https://doi.org/10.1016/j.soilbio.2020.107725

Reference:

SBB 107725

To appear in:

Soil Biology and Biochemistry

Received Date: 8 August 2019 Revised Date:

14 January 2020

Accepted Date: 17 January 2020

Please cite this article as: Clark, D.R., McKew, B.A., Dong, L.F., Leung, G., Dumbrell, A.J., Stott, A., Grant, H., Nedwell, D.B., Trimmer, M., Whitby, C., Mineralization and nitrification: Archaea dominate ammonia-oxidising communities in grassland soils, Soil Biology and Biochemistry (2020), doi: https:// doi.org/10.1016/j.soilbio.2020.107725. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 Published by Elsevier Ltd.

1

Mineralization

and

nitrification:

2

communities in grassland soils.

Archaea

dominate

ammonia-oxidising

3 4

Authors: Dave R. Clarka, Boyd A. McKewa, Liang F. Donga, Garwai Leunga, Alex J.

5

Dumbrella, Andrew Stottb, Helen Grant b, David B. Nedwella, Mark Trimmerc, Corinne

6

Whitbya*

7 8

Affiliations:

9

a

10

b

11

Lancaster, Environment Centre, Lancaster, LA1 4AP, UK

12

c

13

Mile End Road, London E1 4NS, UK.

School of Life Sciences, University of Essex, Colchester CO4 3SQ, UK Life Sciences Mass Spectrometry Facility, Centre for Ecology & Hydrology,

School of Biological and Chemical Sciences, Queen Mary University of London,

14 15 16

*Corresponding author:

17

Corinne Whitby [email protected]

18

School of Life Sciences, University of Essex, Colchester CO4 3SQ, UK

19

Tel: +44 (0) 1206 872062

20

Fax: +44 (0) 1206 872592

21 22

Keywords:

Nitrification;

N

mineralization;

23

Nitrososphaera; amoA gene; grasslands

ammonia

oxidising

archaea;

24 25

Declarations of interest: none

1

26

Abstract

27

In grasslands, N mineralization and nitrification are important processes and

28

are controlled by several factors, including the in situ microbial community

29

composition. Nitrification involves ammonia oxidising archaea (AOA) and bacteria

30

(AOB) and although AOA and AOB co-exist in soils, they respond differently to

31

environmental

32

differentiation. Here, we investigated temporal variation in N mineralization and

33

nitrification

34

communities in grassland soils, on different geologies: clay, Greensand and Chalk.

35

Across geologies, N mineralization and nitrification rates were slower in the autumn

36

than the rest of the year. Turnover times for soil ammonium pools were <24 h, whilst

37

several days for nitrate. In clay soils, bacterial, archaeal, AOA, and AOB

38

communities were clearly distinct from those in Chalk and Greensand soils. Spatially

39

and temporally, AOA were more abundant than AOB. Notably, Nitrososphaera were

40

predominant, comprising 37.4% of archaeal communities, with the vast majority of

41

AOA found in Chalk and Greensand soils. AOA abundance positively correlated with

42

nitrate concentration, whereas AOB abundance correlated with ammonium and

43

nitrite concentrations, suggesting that these N compounds may be potential drivers

44

for AOA/AOB niche differentiation in these grassland soils.

characteristics

rates,

together

and

with

there

is

bacterial,

evidence

archaeal

of

and

AOA/AOB

niche

ammonia-oxidizer

45 46 47 48 49 50

2

51

1. Introduction

52

Grasslands cover ~40% of the Earth’s land surface, are high in organic matter

53

and represent a large reservoir of nitrogen (N) (Cambardella and Elliott, 1992). In the

54

United Kingdom, grasslands cover over a fifth of the land area (ca. 7.5 million ha)

55

and includes dry acid grasslands comprising 278,866 ha, calcareous Chalk

56

grasslands covering 788,979 ha and improved grassland with over 3 million ha

57

(Carey et al., 2007, Rodwell et al., 2007; Natural England, 2008). Grasslands were

58

also identified as a priority habitat under the UK Biodiversity Action Plan (BAP) as

59

they are important habitats for biodiversity and carbon sequestration; with UK

60

grasslands sequestering 240±200 kg of carbon per hectare per year (Natural

61

England, 2008; Ostle et al., 2009).

62

Mineralization of soil organic matter by microorganisms is an important

63

process in grassland N cycling, and is crucial for regulating the available N in soils

64

for plant growth as well as preventing a net loss of N from the environment (Yao et

65

al., 2011; Heijden et al., 2008; Zhang et al., 2012). Previous work has shown that

66

increasing mean annual temperatures may increase net N mineralization in

67

grassland soils, increasing coupled nitrification-denitrification, thus causing increased

68

greenhouse gas contribution from grasslands, and a net loss of N from these

69

ecosystems (Smith et al., 2002; Hutchinson et al., 1995; Zhang et al., 2012). Other

70

factors may also control soil N mineralization and nitrification rates, including soil

71

organic nitrogen (SON), soil water availability, total N, soil organic C (SOC), C:N

72

ratio and microbial community composition (Dalal and Meyer, 1987; Fisk and

73

Schmidt, 1995; Von Lutzow and Kögel-Knabner, 2009).

74

Ammonia oxidizing bacteria (AOB) and archaea (AOA) are the major drivers

75

of the aerobic oxidation of ammonia, which is particularly important for soil fertility

3

76

(Könneke et al., 2005; Prosser and Nicol, 2008; Prosser and Nicol, 2012). Although

77

AOA and AOB co-exist in soils, they respond differently to environmental factors and

78

there is evidence of niche differentiation between AOB and AOA (Erguder et al.,

79

2009; Prosser and Nicol, 2012; Hink et al., 2018). For example, the global

80

dominance of AOA in acidic soils (Prosser and Nicol, 2008; Gubry-Rangin et al.,

81

2011; Prosser and Nicol, 2012), and AOA rather than AOB favouring low ammonium

82

environments such as unfertilised soils (Leininger et al., 2006; Di et al., 2010; Gubry-

83

Rangin et al., 2010; Verhamme et al., 2011; Hink et al., 2017). In grazed grasslands

84

the abundance of AOA was found to be higher than that of AOB, and changes in

85

AOA community composition correlate with changes in C:N ratio, whereas changes

86

in AOB communities are independent of edaphic factors (Xie et al., 2014). Other

87

studies however, showed that nitrate concentration correlated with AOB abundance

88

(Di et al., 2009, 2010; Wertz et al., 2012). Furthermore, the recent discovery of the

89

complete oxidation of ammonia to nitrate in a single organism (commamox) within

90

the Nitrospira genus (Daims et al., 2015) and the co-occurrence with AOB has raised

91

further questions on the niche specialization of nitrifiers in terrestrial ecosystems.

92

To determine the contribution of grasslands to global N budget, particularly

93

under a changing climate, it is important to better understand grassland N dynamics

94

and the associated microorganisms involved. In the UK, nitrate runoff from

95

grasslands is one of the main sources of anthropogenically-enhanced nutrient loads

96

across all river catchments (Nedwell et al., 2002; Earl et al., 2014). This is

97

particularly important in permeable geologies were responses to land management

98

changes targeted at reducing nitrate loadings are delayed due to long water

99

residence times (Heppell et al., 2017). Given that England has more permeable

100

Chalk rivers than any other country in Europe (around 3,900 km) (Natural England

4

101

2008), it is important to understand how changes in N dynamics across catchments

102

potentially change nitrate production and delivery into rivers. Thus, examining

103

grassland N dynamics, in relation to the microbial communities driving these

104

processes, feeds directly into a landscape-scale understanding of N budgets. Here,

105

we focussed on grassland soils, adjacent to rivers in southern United Kingdom (Fig.

106

1). The aims of this study were to investigate bacterial, archaeal and ammonia-

107

oxidizer communities in grassland soils, on different geologies: clay, Greensand and

108

Chalk. We hypothesized that the less permeable clay and more permeable

109

Greensand and Chalk soils would select for different ammonia-oxidizer communities

110

spatially and temporally, which in turn would drive changes in N mineralization and

111

nitrification rates.

112 113

2. Methods and materials

114

2.1. Site Descriptions and sampling

115

The research was undertaken at the Hampshire Avon catchment (southern

116

England) which has sub-catchments of contrasting geology: clay, Greensand and

117

Chalk (Allen et al., 2014; Heppell et al., 2017). The soils used in this study are herein

118

referred to as clay (>99% Late Jurassic Kimmeridge Clay), Greensand (50% Upper

119

Greensand) and Chalk (80% Chalk geology) soils based on their respective

120

underlying geologies (Heppell et al., 2017) (Fig. 1). Soil descriptions are as follows:

121

slowly permeable, clayey soil (Clay); free-draining loamy soil (Greensand);

122

permeable, base-rich loamy soil (Chalk soil) (http://www.landis.org.uk/soilscapes).

123

Over the sampling period, soil pH ranged from the more acidic clays pH 5.5, to 7.2

124

and 7.6 for Greensand and Chalk respectively. During the sampling period the

125

catchment received a total of 824 mm rainfall (AEDA, accessed 2019).

5

126

To encompass any lateral gradients within and across sub-catchments, a

127

random 100×100m square was marked by stakes at right angles to the adjacent river

128

at each location. Spatially independent and randomised replicate (n=8) surface soils,

129

were collected from each location (0-10 cm depth), in 2013-14: spring (April/May),

130

summer (August), autumn (November), and winter (February/March) to cover the

131

range of temperatures throughout the year (autumn/winter: 7°C; spring/summer:

132

20°C). These were typical temperatures for the region, consistent with temperatures

133

we measured the previous year, which ranged between 4 to 21°C. Soil water content

134

varied between summer lows of 26% in clay, 46% and 49% in Greensand and Chalk

135

respectively; to autumn highs near 50% for clay and Greensand, and 80% for Chalk.

136

Soil samples were maintained at in situ temperatures until processing (within 24 h).

137

Sub-samples for molecular analyses were snap frozen on-site using a cryoshipper (-

138

150°C) and stored at -80°C. Gravimetric water content of soils were determined by

139

drying approximately 10 g of soil at 105°C to constant weight, (expressed as g water

140

g-1 of oven-dry soil).

141 142

2.2. Mineralization and nitrification rates. Both mineralisation and nitrification rates were quantified using

143

15

N isotope

144

microdiffusion techniques. In preliminary tests of NH3 diffusion and 15N recovery, two

145

levels of

146

of total NH4+. Ammonium chloride solution (50 ml of 100 µM NH4Cl in 2M KCl with

147

known

148

(Whatman GF/C, 25mm diameter) was suspended on a hook hanging beneath a

149

rubber stopper and 50 µl of 2.5M KHSO4 solution dropped onto the filter. The soil

150

extract was made alkaline by adding 0.2 ml 2M NaOH and the bottle stirred for 10

15

N labelling were used: natural abundance and a

15

N enrichment of 9.2%

15

N (%) enrichment was added to a serum bottle (100 ml)). Half a filter paper

6

151

min. Ammonium in the solution was converted to ammonia which diffused from

152

solution to the headspace of the serum bottle and trapped on the acidified filter. The

153

bottles were held at room temperature for 7 days until all the ammonia was absorbed

154

onto the filter. The filter was removed, placed in a vial and dried in a desiccator.

155

Recovery of NH4+ was > 99%.

156

To measure mineralization, soil cores (~9 g wet weight: 0-10cm depth) were

157

taken in 10 ml plastic syringes with the distal end removed. Representative cores

158

were extracted with 40 ml 2M KCl solution shaken for 10 s and placed on a roller for

159

1 h. Tubes were centrifuged (4000 rpm, 10 min) (Sanyo, Harrier 15/80 MSE), the

160

supernatant decanted and filtered (Whatman No 1 filter paper) and ammonium and

161

nitrate concentrations in the soil extracts were analysed colorimetrically (Johnson,

162

2002; Krom, 1980). The remaining cores were injected with 0.5 ml (15NH4)2SO4

163

(between 0.5-2.5 µg N g-1 dry soil for clay; between 1.5-3 µg N g-1 dry soil for

164

Greensand; between 2.5-5.5 µg N g-1 dry soil for Chalk) to give 30% enrichment of

165

15

166

determine the extraction efficiency of the

167

samples were incubated for 24h at in situ temperature, with their upper surfaces

168

exposed to air.

NH4+. Duplicate samples were immediately extracted for time zero (t0) controls to 15

N (which was >99%). The remaining soil

169

Nitrification rates were measured by injecting further soil cores with 0.5ml

170

K15NO3 solution (between 10-20 µg N g-1 dry soil for clay; between 17-34 µg N g-1

171

dry soil for Greensand; between 20-35 µg N g-1 dry soil for Chalk) to give 20%

172

enrichment of

173

determine the extraction efficiency of the

174

samples incubated for 24h at in situ temperatures. After incubation, samples were

175

extracted with 2M KCl. The ammonia was then removed by making alkaline with 0.2

15

NO3. Time zero (t0) controls were immediately processed to 15

N (which was >97%) and the remaining

7

176

ml 2M NaOH solution and left for 7 days to absorb onto acidified filter paper

177

(McMurray et al., 2002). Residual nitrate in the soil extract was then reduced to

178

ammonia over 8 days using 0.2 g of MgO and 0.4 g of Devarda’s alloy (Sigma-

179

Aldrich). Ammonia from the reduction of NO3- was diffused out of solution over 8

180

days and was absorbed by the acidified filter. After 8 days, the filters were removed

181

and dried in a desiccator. All

182

analysed by the NERC Life Sciences Mass Spectrometry Facility (LSMSF), Natural

183

Environmental Research Council, UK (See Supplementary Information). Rates of

184

mineralization, reassimilation,

185

according to Kirkham and Bartholomew (1954).

15

N/14N isotopic measurements for samples were

consumption

and nitrification were

calculated

186 187

2.3. Soil organic carbon (SOC) and soil organic nitrogen (SON).

188

Soil samples (0-10 cm depth) were taken with cut-off 10 ml sterile hypodermic

189

syringes. Soils were sieved and ground in a ball mill, and samples for SOC were

190

acidified with 0.5 ml 1% (v/v) HCl to remove carbonates from the soil before SOC

191

analysis. Samples were oven dried at 105°C to constant weight, cooled and sealed

192

prior to analysis. Aliquots (20 mg) of samples were enclosed in tin sheets and placed

193

in a pellet press to remove any air. SOC and SON were analysed by the Central

194

Chemistry Unit of the NERC Centre for Ecology and Hydrology, Lancaster, UK. SOC

195

was measured in a Vario EL (Elementar Analsensysteme GmbH, Hanau, Germany)

196

(See Supplementary Information).

197 198

2.4. 16S rRNA and amoA gene analysis

199

DNA was extracted from 0.25 g wet weight soil using a PowerSoil® DNA

200

Isolation Kit (MO BIO Laboratories, Inc). Gene abundance was quantified by qPCR

8

201

with a SensiFAST SYBR No-ROX Kit (Bioline) on a CFX96 Real-Time PCR

202

Detection System (BioRad). Ammonia monooxygenase (amoA) genes were targeted

203

with the primers amoA-1F and amoA-2R for AOB (Rotthauwe et al., 1997), and

204

CrenamoA-23F and CrenamoA-616R for AOA (Tourna et al., 2008). Gene

205

abundances were quantified with an absolute quantification method against an

206

internal standard calibration curve using DNA standards of each target gene from

207

102 to 107 copies in 20 µl reactions containing 200 nM of primers and 1 µl of DNA

208

template. Cycle conditions for all genes were 95 ℃ for 3 min followed by 40 cycles at

209

95 ℃ for 10s then 60 ℃ for 30 seconds. Amplification specificity was confirmed by

210

melting curve analysis.

211

Amplicon libraries were prepared by a 28-cycle (16S rRNA Bacteria) or 31-

212

cycle (16S rRNA Archaea, AOB and AOA amoA genes) PCR using the same locus-

213

specific amoA gene primers as the qPCR assays and the variable regions 3-4 of the

214

16S

215

(CCTACGGGNGGCWGCAG) and Bakt_805R (GACTACHVGGGTATCTAATCC) for

216

Bacteria (Herlemann et al., 2011), and 344F (ACGGGGYGCAGCAGGCGCGA)

217

(Raskin et al., 1994) and 915R (GTGCTCCCCCGCCAATTCCT) (Stahl et al., 1991)

218

for Archaea., but flanked with Illumina overhang sequences. PCRs were performed

219

in 25 µl reactions consisting of 12.5 µl of RedTaq ReadyMix (Sigma Aldrich) 200 nM

220

of forward and reverse primers and 1 µl of template DNA Cycling conditions for all

221

genes were 95°C for 3 min followed by 28 or 31 cycles of 95°C for 30 s; 57°C for 30

222

s (except amoA AOA which was at 55°C); 72°C for 30 s and a final extension of

223

72°C for 5 min). Amplicons were purified using Agencourt AMPure XP (Beckman

224

Coulter) beads, before adding unique combinations of Nextera XT paired-end

225

Indexes via an 8-cycle PCR (reactions contained 25 µl of RedTaq ReadyMix, 5 µl

rRNA

genes

were

targeted

using

the

primer

pairs

Bakt_341F

9

226

each of unique Nextera XT Index, 5 µl of DNA (from cleaned PCR1). Amplicons were

227

again purified using AMPure XP beads, quantified using a Quant-iT Picogreen

228

dsDNA assay kit (Life Technologies) on a Nanodrop 3300 fluorospectrometer

229

(Thermo Scientific) and pooled in equimolar concentrations. The quality of amplicon

230

libraries were verified using a DNA 1000 kit on a 2100 Bioanalyzer (Agilent) before

231

final pooled libraries were sequenced on the Illumina Miseq platform using a MiSeq

232

reagent kit V3 (2 × 300 bp) at The Earlham Institute (formally The Genome Analysis

233

Centre, Norwich, UK).

234

Sequence reads were de-multiplexed on the MiSeq platform and analysis was

235

performed on the forward reads, following protocols described by (Dumbrell et al.,

236

2016). The sequences were quality trimmed using Sickle (Joshi and Fass, 2011),

237

with a min quality threshold of q20. Sequences were then error corrected with

238

SPAdes (Nurk et al., 2013) using the BayesHammer algorithm (Nikolenko et al.,

239

2013). The sequences were then de-replicated, sorted by abundance, and

240

Operational Taxonomic Units (OTU) centroids were picked using VSEARCH

241

(Rognes et al., 2016) at 97% similarity. Singleton OTUs were removed from the

242

dataset, along with any chimeric sequences identified by both de novo and reference

243

based chimera checking with UCHIME (Edgar et al., 2011). Taxonomy assignment

244

for 16S rRNA sequences was performed with the RDP Classifier (Wang et al., 2007).

245

Non locus-specific OTUs were removed prior to statistical analyses.

246

Phylogenetic trees of translated amino acid sequences were constructed as

247

previously described (Lansdown et al., 2016). Centroid sequences of the most

248

abundant OTUs (comprising >99% of the sequencing reads for each gene) were

249

aligned by codons using MUSCLE, in the MEGA6 program (Edgar, 2004; Tamura et

250

al., 2013). Non-specific sequences detected during the alignment were discarded

10

251

from all downstream analyses, and removed from OTU tables. Sequences were

252

aligned with other known ammonia-oxidiser amoA sequences from the Fungene

253

database (Fish et al., 2013) and from BLAST analyses (Altschul et al., 1997).

254

Maximum likelihood trees were constructed on amino acid sequences, using the Le

255

and Gascuel (2008) substitution model. Evolutionary rate differences between sites

256

were modelled with a discrete Gamma distribution. The phylogeny was tested using

257

1000 bootstrap permutations.

258 259

2.5. Statistical Analyses

260

OTU tables were imported into R (version 3.4.3) for statistical analyses (R

261

Development Core Team, 2016). After discarding samples that had excessively

262

small library sizes (i.e. per sample: <675 sequences for archaeal 16S rRNA gene,

263

11,773 sequences for bacterial 16S rRNA gene, 2,928 sequences for AOA amoA

264

gene, and 1,190 sequences for AOB amoA gene), OTU tables were rarefied to an

265

even depth with the “vegan” package (Oksanen, et al., 2017). Compositional

266

differences in the soil microbial communities were quantified using the Sørensen

267

index (Baselga et al., 2017) and visualised using non-metric multidimensional scaling

268

analyses (NMDS). PERMANOVA analyses were performed to test for compositional

269

differences between different geologies, using 10,000 permutations. Differences in

270

OTU richness of AOA and AOB in relation to geology were tested with negative

271

binomial GLMs. Finally, putatively different functional groups of AOA and AOB were

272

identified by grouping OTUs with identical amoA amino acid sequences (as identified

273

through the phylogenetic analyses described previously). Differences in the

274

abundances of these putative functional groups across geologies were then tested

275

with multivariate negative binomial GLMs (Wang et al., 2007). Raw sequence data

11

276

were submitted to the European Nucleotide Archive under accession number

277

PRJEB28502.

278 279

3. Results

280

3.1. Soil physicochemical characteristics

281

There was no significant temporal variation (P > 0.05) of either SOC or SON

282

across soils. However, clay soil had significantly lower concentrations of SOC and

283

SON than Greensand and Chalk across the year (P <0.05 in both cases), and were

284

significantly more acidic than the near neutral pH Chalk (coef = 2.73, t2, 6 = 17.75, P

285

< 0.001) and Greensand (coef = 2.22, t2,

286

Turnover times of the SON pools were ca 3 years for all geologies (Table 1). The

287

C:N (g:g) ratios also differed across sites with near 10 in both clay and Greensand

288

(P > 0.05), but higher (mean 12.6, P < 0.001) in Chalk soils (Table 1).

6

= 14.43, P < 0.001) soils (Table 1).

289 290

3.2. Mineralization and nitrification rates

291

Mean mineralization rates differed across geologies with rates of 5.9 and 8.3

292

µg N g-1 soil day-1 for Greensand and Chalk respectively, compared to 4.6 µg N g-1

293

soil day-1 in the more acidic clay soils throughout the year (Table 1, Fig. 2).

294

Mineralization rates were significantly slower in the autumn compared to summer

295

(coef = 1.61, t = 3.35, P < 0.01) and Spring (coef = 1.78, t = 3.70, P < 0.001),

296

(although these differences were only statistically significant after accounting for

297

geological variation). Ammonium pool turnover times in all soils across the year were

298

<1 day (Fig. 2), indicating that the soil ammonium pools were highly dynamic

299

throughout the year. In the clay soils in the summer when soil ammonium

300

concentrations were the lowest, ammonium turnover times were the slowest. In

12

301

contrast, for the rest of the year in clay soils, when ammonium concentrations were

302

higher, ammonium turnover times were faster. In the Chalk soils (in autumn), when

303

ammonium concentrations were low, ammonium turnover times were slower, but

304

increased in the winter when ammonium concentrations were higher. With

305

Greensand soils, ammonium turnover times remained relatively slow throughout the

306

year, and ammonium concentrations ranged between 1.15 to 2.23 µg N g-1 soil day-1.

307

The soil nitrate pools were larger (8.1-29.4 µg N g-1 dry soil) than ammonium

308

pools (0.4-4.3 µg N g-1 dry soil) (Table 2); and turnover times of nitrate pools were

309

considerably longer (several days) than ammonium (fractions of a day) (Fig. 2).

310

Turnover times of the soils’ nitrate pools also showed temporal changes, with the

311

longest turnover times in the autumn (Fig. 2). Across geologies, ammonium

312

concentrations were greatest in the Chalk soils, where SON and SOC were also

313

greatest (Table 1, Table 2). Temporally, ammonium concentrations were lowest in

314

the autumn, (when nitrate was highest), yet differences in ammonium concentrations

315

across months and geologies were not statistically significant (P > 0.05 in all cases)

316

(Table 1, Table 2).

317

The

15

N reassimilation of ammonium into microbial cells (immobilisation) after

318

its mineralization to ammonium gave only extremely low, often negative, and very

319

variable rates between replicates, and were usually not significantly different (P >

320

0.05) from zero. Assimilation is generally only a small percentage (<10%) of nitrogen

321

flow during microbial growth, compared to dissimilative metabolism (e.g. Pirt, 1975).

322

Measurable remineralization of reassimilated ammonium by the soil microbiome after

323

only a week-long incubation has been reported (e.g. Bjarnason, 1988), but our

324

measurements after only one day incubation were highly unlikely to detect

325

reassimilation as was also reported in grassland soils elsewhere within 48h (Braun et

13

326

al., 2008). However, while more prolonged incubation is likely to detect

327

reassimilation, it is also likely to enhance any enclosure effect rather than reflect in

328

situ rates. Therefore, we do not consider immobilisation (reassimilation) data further

329

in the current study, apart from noting that it is a very low and variable proportion of

330

N flow in these soils compared to other processes.

331

Nitrification rates showed temporal variation (Fig. 2), with significantly slower

332

rates in the autumn compared to summer (coef = 5.16, t3, 35 = 2.32, P < 0.05) and

333

winter (coef = 5.04, t3, 35 = 2.27, P < 0.05) apart from in Greensand soils (P = 0.50),

334

and correlated with lower soil ammonium concentrations for Greensand and Chalk

335

soils, but not in the clay soils which had higher soil ammonium levels in the autumn

336

than the summer. Across sites, the proportion of SON mineralized that was

337

subsequently nitrified was (with the exception of Greensand) smallest in the autumn

338

when both ammonium concentrations and nitrification rates were lowest, but not in

339

the clay soils which had higher soil ammonium levels in the autumn than the summer

340

(Table 2). On an annual basis, the average proportion of ammonia derived from N

341

mineralization that was subsequently nitrified varied across sites from 123% in clay,

342

but only 53% in Greensand and 74% in Chalk soils (Table 2).

343 344

3.3. 16S rRNA bacterial, archaeal, AOA and AOB communities

345

Distinct bacterial, archaeal, AOA and AOB communities were found in the 16S

346

rRNA gene libraries in the less permeable clay soils, compared to the more

347

permeable Chalk and Greensand soils, which were indistinguishable from each other

348

(Fig. S1). PERMANOVA analyses revealed that communities were more similar

349

within a geology than between geologies (AOA; pseudo-F = 26.98, R2 = 0.56, P <

350

0.001; AOB; pseudo-F = 11.15, R2 = 0.35, P < 0.001). Clay soils were also distinct in

14

351

terms of ammonia oxidiser richness, being significantly lower in OTU richness than

352

Greensand and Chalk soils for both AOA, (clay-Greensand; coef = 0.99, z = 3.62, P

353

< 0.001, clay-Chalk; coef = 0.21, z = 2.66, P < 0.01), and AOB, (clay-Greensand;

354

coef = 0.28, z = 5.64, P < 0.001, clay-Chalk; coef = 0.62, z = 3.47, P < 0.001) (Fig.

355

S1E -F). However, OTU richness of AOA and AOB, did not differ significantly across

356

sampling months (P >0.05 in all cases).

357

Within archaeal 16S rRNA libraries, AOA were the most abundant with

358

Nitrososphaera accounting for 37.4% of total observed sequences, and the vast

359

majority found in Chalk (56.7%) and Greensand (63.7%) compared to clay (1.2%)

360

soils (Fig. S2). In addition to AOA, a further seven genera, which although generally

361

more abundant in clays, comprised <0.5% of the Archaeal library (Fig. S2),

362

leaving >62% of OTUs that could not be confidently assigned to a particular archaeal

363

genus. AOA 16S rRNA gene abundance did not differ significantly throughout the

364

year (P > 0.05 in all cases) (Fig. S2). Within the Nitrososphaera, 12 OTUs were

365

identified (Fig. S3). The two most abundant OTUs (OTU9, OTU6) represented

366

19.4% and 8.8% of observed archaeal 16S rRNA sequences, respectively), and

367

were 94-95% similar to Nitrososphaera viennensis.

368

Within the bacterial 16S rRNA libraries, the most abundant bacterial taxa were

369

Acidobacteria (7.1%) followed by Actinobacteria (5.8%) and Betaproteobacteria

370

(5.0%), with 11.1% of sequences corresponding to unclassified bacteria. Eight

371

genera (representing <1.4% of observed bacterial sequences) showed temporal

372

abundance shifts (Fig. S4). A total of 67 genera changed across geology,

373

(collectively representing 31.8% of the bacterial 16S rRNA library) (Fig. S5). AOB

374

represented only <0.06% of the total observed bacterial 16S rRNA gene sequences,

375

and all were Nitrosospira spp. Moreover, only two AOB OTUs were identified within

15

376

the bacterial 16S rRNA library and had 99% identity to Nitrosospira multiformis

377

(OTU1) and Nitrosospira lacus (OTU2) (Fig. S6).

378 379

3.4. AOA and AOB amoA gene analysis.

380

Generally, AOA amoA gene abundances were an order of magnitude greater

381

than AOB across geology and sampling period (Fig. 3). AOA amoA gene

382

abundances ranged from 1.7 × 106 to 1.8 ×108 copies g-1 dry soil compared to AOB

383

6.1 × 105 to 1.8 ×107 copies g-1 dry soil. AOA and AOB amoA gene abundance

384

differed significantly between geologies: clay contained significantly fewer amoA

385

genes on average than Greensand (P < 0.01 for both AOA and AOB). However, only

386

for the AOA amoA abundance did all three sites differ significantly from each other

387

(Tukey HSD test; P < 0.05 for all comparisons). Clay harboured the lowest AOA

388

abundances (1.7 × 106 to 1.5 × 107 amoA genes g-1 dry soil), compared to 1.8 ×107 -

389

1.8 ×108 amoA genes g-1 dry soil for Greensand (~15 fold higher than clay), and 1.6

390

× 107 to 1.2 × 108 amoA genes g-1 dry soil) for Chalk soils (Fig. 3A). AOB followed a

391

similar pattern to AOA, with clay having the lowest AOB abundances, followed by

392

Chalk, and Greensand soils having the highest abundances (Fig. 3B). However, only

393

clay and Greensand soils had significantly different AOB abundances over the

394

course of a year (relative to clay; coef = 1.04, z = 3.29, P < 0.01). AOA amoA

395

abundances were significantly lower (P < 0.001, for all comparisons) in winter (Fig.

396

3A), than at any other time, and were significantly higher in the autumn than summer

397

and winter (P < 0.05 for both comparisons), but not spring. AOB showed markedly

398

different temporal dynamics, with their abundance peaking in spring at all sites,

399

(relative to Spring; coef < -1.87, z < -5.06, P < 0.001, for all comparisons) (Fig. 3B,

400

Table S1).

16

401

Across geologies and sampling period, the ratio of AOA to AOB amoA usually

402

favoured AOA (1st quartile = 1.12, median = 5.71, 3rd quartile = 5.37; Fig. 3C), but

403

ranged from 0.03 (favouring AOB) to 48.2 (favouring AOA). In all sites, the ratio

404

remained close to 1 during winter and spring, indicating approximately equal

405

proportions of AOA and AOB. However, in summer and autumn, the ratio increased

406

above 1 in all geologies, particularly so in clays during autumn (between 24.7 to

407

48.2, median = 36.5), due to there being an order of magnitude fewer AOB than in

408

the same soils during the rest of the year. Conversely, in Chalk and Greensand soils

409

this increase was due to an increase in AOA populations.

410

Changes in AOA abundance were found to positively correlate with changes

411

in nitrate concentration (coef = 1.70, z = 2.22, P < 0.05), whereas AOB abundance

412

correlated with ammonium coef = 2.00, z = 4.62, P < 0.001) and nitrite (coef = 63.34,

413

z = 4.89, P < 0.001) concentrations (Fig. 3D). Statistical differences in AOA and

414

AOB abundance across soils throughout the year (based on amoA qPCR data), and

415

their correlation with overall archaeal and bacterial community size (as determined

416

by 16S rRNA qPCR), were tested for using negative binomial generalised linear

417

models (GLMs) in relation to ammonium and nitrate concentrations. We found that

418

when ammonium concentrations were low, AOA were more abundant than AOB

419

across soils. However, when nitrate concentrations were low, AOB were more

420

abundant than AOA (Fig. S7).

421

Phylogenetic analysis of AOA amoA genes showed that all the AOA OTUs

422

clustered within the family Nitrososphaeraceae (Fig. 4), specifically most closely to

423

Nitrososphaera gargensis and Nitrosocosmicus franklandus. When OTUs were

424

translated to their amino acid structure, the AOA formed three OTU groups with

425

distinct amino acid sequences (Fig. 4, Table S2). OTU Group 1 was significantly

17

426

more abundant in both Greensand (coef = 4.60, z = 12.17, P < 0.001) and Chalk

427

(coef = 4.47, z = 11.83, P < 0.001), irrespective of month, compared to clay soils,

428

whereas Group 3 showed the opposite pattern, being more abundant throughout the

429

year in clay compared to Chalk (coef = -7.90, z = -12.03, P < 0.001) or Greensand

430

(coef = -7.50, z = -12.71, P < 0.001) (Fig. S8A). Group 2 was more abundant on

431

average in clay soils throughout the year compared with Greensand (coef = -3.93, z

432

= -3.81, P < 0.001), or Chalk soils (coef = -1.99, z = -2.16, P < 0.05), except in

433

autumn where it was more abundant in Chalk soils (coef = 3.43, z = 2.66, P < 0.01).

434

Phylogenetic analysis of AOB amoA genes showed that all AOB OTUs

435

clustered within the genus Nitrosospira (Fig. 5). AOB OTUs translated into 6 unique

436

amino acid variants (Fig. 5, Table S3). Some of these groups displayed similar

437

patterns as AOA, being differentially abundant in clay versus Chalk and Greensand

438

soils. Groups 2 and 5 were both more abundant in the Greensand (Group 2; coef =

439

5.84, z = 7.30, P < 0.001, Group 5; coef = 4.05, z = 6.91, P < 0.001) and Chalk soils

440

(Group 2; coef = 4.31, z = 5.37, P < 0.001, Group 5; coef = 4.13, z = 7.04, P < 0.001)

441

than clay soils, though only Group 2 showed temporal changes, being more

442

abundant on average in spring (coef = 1.98, z = 2.41, P < 0.05). Groups 3 and 4

443

displayed the reverse pattern as they were more abundant in clay than Greensand

444

(Group 3; coef = -2.08, z = -2.58, P < 0.05, Group 4; coef = -6.66, z = -8.01, P <

445

0.001) or Chalk soils (Group 3; coef = -3.28, z = -4.03, P < 0.001, Group 4; coef = -

446

1.62, z = -2.24, P < 0.05). Group 6 was more abundant in Greensand soils compared

447

to clay or Chalk soils (coef = 1.95, z = 2.41, P < 0.05), except during the spring,

448

when their abundance in clay soils was significantly higher than in Chalk (coef = -

449

4.10, z = -3.46, P < 0.001) and Greensand soils (coef = -3.34, z = -3.26, P < 0.01).

450

18

451

4. Discussion

452

In grassland soils the rate of organic mineralization as a supply of ammonium

453

in relation to AOA and AOB communities has not been fully investigated. In this

454

study, N mineralization rates measured in grassland soils in the Hampshire Avon

455

catchment (southern United Kingdom), were similar to those reported for other

456

European grasslands (e.g. 5.3 ± 0.1 µg N g-1 day-1) (Braun et al., 2008). Here,

457

temporal differences in N mineralization were found, with lower rates in the autumn

458

than at other times of the year that may be due to changes in the physiological

459

response of the existing microbial community to some environmental or edaphic

460

factor.

461

Previous work showed that increasing mean annual temperatures may

462

increase net N mineralization in grassland soils (Smith et al., 2002; Hutchinson et al.,

463

1995; Zhang et al., 2012). In tropical soils, temperature affects mineralization rates

464

(Myers, 1975). In our study, temperatures ranged between ~20°C (spring/summer)

465

to ~7°C (autumn/winter). It is therefore possible that the lower autumn/winter

466

temperatures affected ammonifier physiology and a period of adaption to these

467

colder temperatures was required. King and Nedwell (1984) demonstrated a two

468

month lag adaption period to temperature occurred with sediment nitrate reducers.

469

Alternatively, other soil factors (e.g. pH) may have affected ammonifier physiology.

470

Indeed mean N mineralization rates differed across geologies with between 1.5 and

471

2-fold faster rates in the near neutral Greensand and Chalk soils (pH 7.2 and 7.6

472

respectively) compared to the more acidic clay (pH 5.5) soils throughout the year,

473

despite climatic conditions for the region being typical during the sampling period.

474

Other studies however, have shown that soil pH does not affect mineralization rates

475

(Dancer et al., 1972).

19

476

Changes in the size or composition of the heterotrophic or mixotrophic

477

microbial communities may also explain the differences in mineralization rates.

478

However, since the 16S rRNA archaeal community structure was found to be

479

temporally stable, and only <1.4% of bacterial abundances changed throughout the

480

year this is unlikely. Heterotrophic and mixotrophic AOA have also been reported

481

and may have contributed to the observed N mineralization rates (Hallam et al.,

482

2006; Walker et al., 2010; Prosser and Nicol, 2012; Zhalnina et al., 2012). However,

483

when mineralization rates were slowest in the autumn, AOA amoA gene abundances

484

were highest (in Greensand and Chalk) than the rest of the year. Other studies have

485

shown contrasting findings whereby AOA amoA sequences correlate both positively

486

and negatively with total soil C and N, suggesting the response of AOA to C and N is

487

very complex (Zhalnina et al., 2012; Xie et al., 2014; Liu et al., 2018). In our study,

488

AOA did not correlate with total soil C.

489

Recently, it was shown that N mineralization rates increased as organic N

490

uptake exceeded microbial growth N demand (Zhang et al., 2019). Here, we

491

determined what proportion of the SON mineralised to ammonium was subsequently

492

nitrified. In general, across geology this proportion was smallest in the autumn (for

493

clay and Chalk soils) when both ammonium concentrations and nitrification rates

494

were lowest. On an annual basis the average proportion of N mineralized that was

495

subsequently nitrified although varied, was highest in the clay soils. This might

496

suggest in clay an additional, yet unknown input of ammonium into nitrification, over

497

and above that derived from SON breakdown, whereas in Greensand and Chalk only

498

part of the mineralized nitrogen was nitrified.

499

One possible explanation for this additional N input to soil is atmospheric

500

deposition of N. However, atmospheric inputs of total nitrogen in the area of the

20

501

Hampshire Avon catchment are about 12 kg ha-1 yr-1 (based on CEH CBED model

502

data http://www.pollutant deposition.ceh.ac.uk/data), compared to 2-3 tonnes N ha-1

503

yr-1 by SON mineralization, so atmospheric deposition would seem unlikely to

504

account for the difference, and moreover would not explain the differences observed

505

between geologies. In the Greensand and Chalk soils, the proportion of N

506

mineralized but not nitrified may have also been lost by export of N by leaching or as

507

ammonia emissions across the catchment but again would not explain the

508

differences observed between geologies.

509

Within the Hampshire Avon sub-catchments, all the grassland sites were

510

grazed (albeit on a grazing rotation), however, the distribution of arable and livestock

511

farming varied across sites. Within the clay grasslands, dairy farming is supported

512

whilst arable agriculture represents a larger proportion of land use in the Greensand

513

and Chalk sites; with up to 55% of Chalk sites being arable and sheep grazing and

514

pig production as minority uses (Heppell et al., 2017). Thus, grazing by dairy cattle

515

are likely to have higher urine/faecal inputs in the clay soils compared to the sheep

516

grazed Greensand and Chalk soils, which may offer one possible explanation for the

517

imbalance between mineralization and nitrification. However, N input to soils via

518

urine/faecal addition was not measured in our study. Nitrification rates were similar to those reported elsewhere (e.g. 6-170 µg N g-

519 520

1

dry soil day-1) (Mørkved et al., 2007). In other studies on grassland soils,

521

nitrification rates showed similar trends as N mineralization rates (Zhang et al.,

522

2012), but in our study, although nitrification rates varied spatially and temporally,

523

nitrification and mineralization rates showed similar general trends. Changes in

524

temperature may also affect nitrification rates (Myers, 1975) and the proportion of

525

nitrification by AOA (Ouyang et al., 2017). In our study, in general slower nitrification

21

526

rates in the autumn (for clay and Chalk) corresponded to higher AOA amoA

527

abundances. At this time, ammonium concentrations were also lower and

528

ammonium pool turnover times were generally fastest (with the exception of clay

529

soils in the spring and Greensand between autumn to spring), and thus competition

530

for ammonium would be particularly intense. In general, ammonium pool turnover

531

times in all soils across the sampling period were <1 day whilst turnover times of

532

nitrate pools were considerably longer (several days) indicating that the soil

533

ammonium pools were highly dynamic throughout the year compared to soil nitrate.

534

In our study, AOA were generally more abundant than AOB with the exception

535

of clay soils where both AOA and AOB were low in abundance. Our findings suggest

536

that AOA were potential drivers of ammonia oxidation (particularly in the Greensand

537

and Chalk soils). Jia and Conrad (2009) showed that potential nitrification correlated

538

with AOA abundance despite high ammonium concentrations favouring AOB.

539

However, this does not explain the lower AOA (and AOB) amoA gene abundances

540

found in the clay soils when ammonium concentrations increased in the winter and

541

the significantly lower in OTU richness with the clays than the Greensand and Chalk

542

soils throughout the year.

543

Although AOA and AOB co-exist in the environment, they respond differently

544

to environmental factors and there is evidence of niche differentiation among

545

ammonia oxidisers in grasslands (Erguder et al., 2009; Prosser and Nicol, 2012; Li et

546

al., 2015; Ouyang et al., 2016; Hink et al., 2018). Soil pH is an important factor in

547

shaping ammonia oxidiser communities (Gubry-Rangin et al., 2011; Lehtovirta-

548

Morley et al. 2011; 2014 Hu et al., 2013; 2014; Hu et al., 2015). Other studies

549

showed that AOB (not AOA) positively correlate with soil pH (Liu et al., 2018).

550

However, the underlying mechanisms of soil pH in shaping the ammonia oxidiser

22

551

community are complex, with direct and indirect pH-associated influencing factors

552

(Hu et al., 2015). In the more acidic clay soils, the soil water equilibrium will favour

553

NH4+ (pKa = 9.24), whereas in Greensand and Chalk soils the equilibrium will move

554

towards ammonia gas (NH3), albeit such differences are very small. In this context,

555

Greensand and Chalk soils would likely be more susceptible to ammonia loss,

556

favouring AOA over AOB. Interactions with other soil parameters may also be

557

occurring. For example in clay soils, ammonium may adsorb to clay minerals (Hink et

558

al., 2018), reducing substrate availability. In our study, ammonium turnover time

559

across soils was highly dynamic throughout the year. Phosphate limitation may also

560

be important. In our study, phosphate concentrations were between 1.4 and 5 fold

561

lower in the clay soils than in the other soils (Table 2). However, Hertfort et al.

562

(2007) demonstrated a positive correlation between crenarchaeotal 16S rRNA gene

563

copies and phosphate concentration.

564

In our study, AOA were more abundant than AOB, spatially and temporally,

565

but we cannot attribute the nitrification rates measured in our study to either AOA or

566

AOB. Furthermore, Nitrospira performing comammox (complete ammonia-to-nitrate

567

oxidation) and their co-occurrence with AOB in the environment (Daims et al., 2015;

568

Hu et al., 2017; Palomo et al., 2018) may also be contributing to the observed

569

nitrification rates. AOA communities were entirely Nitrososphaeraceae (specifically

570

Nitrososphaera gargensis and Nitrosocosmicus franklandus), and comprised ~37%

571

of total observed archaeal 16S rRNA sequences. Phylogenetic analyses placed the

572

OTUs observed entirely within the Nitrososphaera lineage as defined by Alves et al.

573

(2018). In particular, amoA OTU group 2, which shared a common amino acid

574

structure, appeared to fit well within the α-subclade, whilst the positions of the other

575

two OTU groups are less clear without further phylogenetic analyses. The

23

576

dominance of Nitrososphaera in these grassland soils reflects that found in soils

577

elsewhere (Leininger et al., 2006; He et al., 2007; Jiang et al., 2014; Liu et al., 2018).

578

AOA and AOB niches have also been defined by ammonium concentrations,

579

with low ammonium environments selecting for AOA (Leininger et al., 2006; Martens-

580

Habbena et al., 2009; Di et al., 2010). If AOA and AOB utilise ammonia with equal

581

efficiency, AOA will only dominate activity if AOA:AOB is >10 (Prosser and Nicol,

582

2012). Here, the ratio of AOA:AOB amoA genes tended to favour AOA, particularly in

583

the summer and autumn, where the ratio rose to a maximum of 48.25. However,

584

during winter and spring, the AOA:AOB remained close to 1, indicating

585

approximately equal abundances of these two groups. Other studies on soils, found

586

AOA:AOB amoA gene ratios ranged from 3.1 to 91.0 (Liu et al., 2018), 231 in

587

unfertilised soils (Leininger et al., 2006), and 17 to >1600 in semiarid soils (Adair and

588

Schwartz, 2008). In our study, the AOA:AOB negatively correlated with ammonium

589

concentration and elsewhere it has been shown that AOA predominated in soil

590

microcosms until high ammonium concentrations were added (Verhamme et al.,

591

2011). However, other studies on grassland soils showed that addition of N does not

592

necessarily affect AOA abundance (Chen et al., 2014).

593

Differences in substrate affinity for ammonium between AOA and AOB may

594

explain these findings, as AOA have lower Ks and higher µmax values than AOB and

595

will outcompete AOB for ammonia at all concentrations (Prosser and Nicol, 2012).

596

However, contrasting substrate affinities does not explain that similar Ks values have

597

been reported in some AOB (e.g. Nitrosomonas europaea and Nitrosopumilus

598

maritimus) (Kits et al., 2017; Hink et al., 2018) and other AOA. Furthermore,

599

differences in substrate affinities between AOA and AOB does not explain that AOA

600

rather than AOB favour low ammonium environments such as unfertilised soils

24

601

(Leininger et al., 2006; Di et al., 2010; Gubry-Rangin et al., 2010; Verhamme et al.,

602

2011; Hink et al., 2017). In addition, whilst some AOA may be inhibited by high

603

ammonium

604

Nitrosocosmicus species can grow in up to 100 mM ammonium (Jung et al., 2016;

605

Lehtovirta-Morley et al., 2016; Sauder et al., 2017) and archaeal amoA genes have

606

been detected in reasonably high ammonium concentrations (i.e. 10-18 mM,

607

Treusch et al., 2005; Park et al., 2006). However, Ks values are poor measures of

608

affinity (Button, 1993; Nedwell, 1999), and a more robust measure of affinity is given

609

by specific affinity aA, which is equivalent to µmax/Ks. The higher the value of aA, the

610

better is the affinity for the substrate. Using the values for µmax and Ks for typical

611

growth rates of AOA and AOB given by Prosser and Nicol (2012); the aA for AOA is

612

5.28 l µmol-1 h-1 but only 0.002 l µmol-1 h-1 for AOB, showing that AOA have some

613

2600-fold higher affinity for ammonium than AOB. Indeed, the growth rate (µ) of AOA

614

would exceed that of AOB at all soil ammonium concentrations measured. It can also

615

be argued, that slow rates of ammonium supply result in low ammonium

616

concentrations in situ and therefore substrate affinity is more likely to be controlling

617

AOA and AOB competition. This suggests that in these grasslands, AOA are driving

618

ammonia oxidation when ammonium supply rate is low.

concentrations,

several

AOA

isolates

such

as

Candidatus

619 620

5. Conclusions

621

In conclusion, the 16S rRNA bacterial and archaeal communities, and the

622

AOA and AOB communities in clay were clearly distinct from those in the Chalk and

623

Greensand soils. In general, AOA (Nitrososphaera) was more abundant than AOB

624

across sites and sampling time (with the exception of clay soils in the spring, when

625

AOB were dominant). Temporal differences in mineralization and nitrification rates

25

626

across geology were also found, with lows during autumn which corresponded to

627

higher AOA amoA gene copies (in Greensand and Chalk soils) supporting the idea

628

that AOA are driving ammonia oxidation when ammonium supply rate is low. Here,

629

changes in AOA abundance positively correlated with nitrate concentration, whereas

630

AOB abundance correlated with ammonium and nitrite concentrations. These

631

findings provide a better understanding of the drivers of soil N cycling for the 7.5

632

million ha of grassland in the UK where maintenance of soil fertility, carbon stocks

633

and prevention of undesirable N loss are crucial to ecosystem function.

634 635

Acknowledgements

636

Funding: This work was supported by the Natural Environment Research Council,

637

UK, Macronutrient Program (grant numbers NE/J012106/1, NE/J011959/1 and

638

NE/P011624/1). We also thank the landowners for site access.

639 640

Supplementary Data. Supplementary data related to this article can be found on

641

online version of the paper.

642 643 644 645 646 647 648 649 650

26

651 652 653 654 655 656 657

References

658

Adair, K.L., Schwartz, E., 2008. Evidence that ammonia-oxidizing archaea are more

659

abundant than ammonia-oxidizing bacteria in semiarid soils of Northern Arizona,

660

USA. Microbial Ecology 56, 420-426.

661 662

Agricultural and Environmental Data Archive (AEDA), Freshwater biological

663

Association

664

http://www.environmentdata.org/archive/

(accessed

25th

November,

2019)

665 666

Alef, K., Beck, T.H., Zelles, L., Kleiner, D., 1988. A comparison of methods to

667

estimate microbial biomass and N-mineralization in agricultural and grassland soils.

668

Soil Biology and Biochemistry 20, 561-565.

669 670

Allen, D. J., Darling, W. G., Davies, J., Newell, A. J., Gooddy, D. C., Collins, A. L.,

671

2014. Groundwater conceptual models: implications for evaluating diffuse pollution

672

mitigation measures. Quarterly Journal of Engineering Geology and Hydrogeology

673

47, 65–80.

674

27

675

Altschul, S.F., Madden, T.L., Schäffer, A.A., Zhang, J., Zhang, Z., Miller, W., Lipman,

676

D.J., 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database

677

search programs. Nucleic Acids Research 25. 3389-3402.

678 679

Alves, R.J.E., Minh, B.Q., Urich, T., von Haeseler, A., & Schleper, C., 2018. Unifying

680

the global phylogeny and environmental distribution of ammonia-oxidising archaea

681

based on amoA genes. Nature Communications, 9,1517.

682

Barraclough, D., 1991. The use of mean pool abundances to interpret

683

experiments. Plant and Soil 131, 89-96.

15

N tracer

684 685

Baselga, A., Orme, D., Villeger, S., De Bortoli, J., Leprieur, F., 2017. Betapart:

686

partitioning beta diversity into turnover and nestedness components. R package

687

version 1.4-1.

688 689

Bjarnason, S., 1988. Calculation of gross nitrogen immobilisation and mineralization

690

in soil. European Journal of Soil Science 39, 393-406.

691 692

Braun, J., Mooshammer, M., Wanek, W., Prommer ,J., Walker, T.W.N., Rutting, T.,

693

Richter, A., 2008. Full

694

isotope pool dilution approaches for gross nitrogen mineralization. Soil Biology and

695

Biochemistry 117, 16-26.

15

N tracer accounting to revisit major assumptions of

15

N

696 697

Button, D.K., 1993. Nutrient-limited microbial growth kinetics: overview and recent

698

advances. Antonie van Leeuvenhoek 63, 225-235.

699

28

700

Cambardella, C.A., Elliott, E.T., 1992. Carbon and Nitrogen Distribution in

701

Aggregates from Cultivated and Native Grassland Soils. American Society of

702

Agronomy 57, 1071-1076.

703 704

Carey, P.D., Wallis, S.M., Emmett, B.E., Maskell, L.C., Murphy, J., Norton, L.R.,

705

Simpson, I.C., Smar, S.S. 2007. Countryside Survey: UK Headline messages from

706

2007. Centre for Ecology and Hydrology.

707

Chen, Y.L., Hu, H-W., Han, H-Y., Du, Y., Wan, S-Q., Xu, Z-W., Chen B-D., 2014.

708

Abundance and community structure of ammonia‐oxidizing archaea and bacteria in

709

response to fertilization and mowing in a temperate steppe in Inner Mongolia. FEMS

710

Microbiology Ecology 89, 67-79.

711 712

Daims, H., Lebedeva, E.V., Pjevac, P.P., Han, P., Albertsen, M., Jehmlich, N.,

713

Palatinsky, M., Vierheilig, J., Bulaev, A., Kirkegaard, R.H., von Bergen, M., Rattei, T.,

714

Bendinger, B., Nielsen, P.H., Wagner M., 2015. Complete nitrification by Nitrospira

715

bacteria. Nature 528, 24-31.

716 717

Dalal, R.C. and Meyer, R.J., 1987, Long-term trends in fertility of soils under

718

continuous cultivation and cereal cropping in Southern Queensland VII. Dynamics of

719

nitrogen mineralization potentials and microbial biomass. Australian Journal of Soil

720

Research 25, 461-472.

721 722

Dancer, W.S., Peterson, L.A., Chesters, G., 1972. Mineralization and nitrification of

723

N as influenced by soil pH and previous N treatments. Soil Science Society of

724

America Journal 37, 67-69.

29

725 726

Davidson, E.A., Hart, S.C., Shanks, C.A., Firestone, M.K., 1991. Measuring gross

727

nitrogen mineralization, and nitrification by

728

cores. European Journal of Soil Science 42, 335-349.

15

N isotopic pool dilution in intact soil

729 730

Di, H.J., Cameron, K.C., Shen, J.P., Winefield, C.S., O’Callaghan, M., Bowatte, S.,

731

He, J.Z., 2009. Nitrification driven by bacteria and not archaea in nitrogen-rich

732

grassland soils. Nature Geoscience 2, 621-624.

733 734

Di, H.J., Cameron, K.C., Shen, J-P., Winefield, C.S., O’Callaghan, M., Bowatte, S.,

735

He, J-Z., 2010. Ammonia-oxidizing bacteria and archaea grow under contrasting soil

736

nitrogen conditions. FEMS Microbiology Ecology 72, 386-394.

737 738

Dumbrell, A.J., Ferguson, R.M., Clark, D.R., 2016. Microbial community analysis by

739

single-amplicon high-throughput next generation sequencing: Data analysis–from

740

raw output to ecology. In Hydrocarbon and Lipid Microbiology Protocols. Springer,

741

Berlin, Heidelberg, p.p 155-206.

742 743

Earl, T.J., Upton, J.G., Nedwell, D.B.. 2014. UK catchments nutrient loads 1993-

744

2003, a new approach using harmonised monitoring scheme data: temporal

745

changes, geographical distribution, limiting nutrients and loads to coastal waters.

746

Environmental Science: processes and impacts 16, 1646-1658.

747 748

Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy and

749

high throughput. Nucleic Acids Research 32, 1792-1797.

30

750 751

Edgar, R.C., Haas, B.J., Clemente, J.C., Quince, C., Knight, R., 2011. UCHIME

752

improves sensitivity and speed of chimera detection. Bioinformatics 27, 2194-2200.

753 754

Erguder, T.H., Boon, N., Wittebolle, L., Marzorati, M., Verstraete, W., 2009.

755

Environmental factors shaping the ecological niches of ammonia‐oxidizing archaea.

756

FEMS Microbiology Reviews 33, 855–869.

757 758

Fish, J.A., Chai, B., Wang, Q., Sun, Y., Brown, C.T., Tiedje, J.M., Cole, J.R., 2013.

759

FunGene: the functional gene pipeline and repository. Frontiers in Microbiology 4,

760

291.

761 762

Fisk, M.C. and Schmidt, S.K., 1995. Nitrogen mineralization and microbial biomass

763

nitrogen dynamics in three alpine tundra communities. Soil Science Society of

764

America Journal 59,1036-1043.

765 766

Gubry-Rangin, C., Nicol, G.W., Prosser, J.I., 2010. Archaea rather than bacterial

767

control nitrification in two agricultural acidic soils. FEMS Microbiology Ecolology 74,

768

566-574.

769 770

Gubry-Rangin, C., Hai, B., Quince, C., Engel, M., Thomson, B.C., James, P.,

771

Schloter, M., Prosser, J.I., Nicol, G.W., 2011. Niche specialization of terrestrial

772

archaeal ammonia oxidisers Proceedings of the National Academy of Sciences 108,

773

21206-21211.

774

31

775

Hallam, S.J., Mincer, T.J., Schleper, C., Preston, C.M., Roberts, K., Richardson,

776

P.M., DeLong, E.F., 2006. Pathways of carbon assimilation and ammonia oxidation

777

suggested by environmental genomic analyses of marine crenarchaeota. PLoS

778

Biology 4, e95.

779

He, J.Z., Shen, J.P., Zhang, L.M., Zhu, Y.G., Zheng, Y.M., Xu, M.G., Di, H., 2007.

780

Quantitative analyses of the abundance and composition of ammonia-oxidizing

781

archaea of a Chinese upland red soil under long term fertilization practices.

782

Environmental Microbiology 9, 2364-2374.

783 784

Heppell, C.M., Binley, A., Trimmer, M., Darch, T., Jones, A., Malone, E, Collins, A.L.,

785

Johnes, P.J., Freer, J.E., Lloyd, C.E.M., 2017. Hydrological controls on DOC: nitrate

786

resource stoichiometry in a lowland, agricultural catchment, southern UK. Hydrology

787

and Earth System Sciences 21, 4785-4802.

788 789

Heijden, M.G.A.V.D., Bardgett, R.D., Straalen, N.M.V., 2008. The unseen majority:

790

soil microbes as drivers of plant diversity and productivity in terrestrial ecosystems.

791

Ecology Letters. 11, 296-310.

792 793

Herlemann, D.P.R., Labrenz, M., Jürgens, K., Bertilsson, S., Waniek, J.J.,

794

Andersson, A.F., 2011. Transitions in bacterial communities along the 2000 km

795

salinity gradient of the Baltic Sea. ISME Journal 5, 1571-1579.

796 797

Hertford, L., Schouten, S., Abba, B., 2007. Variation in spatial and temporal

798

distribution of Archaea in the North Sea in relation to environmental variables. FEMS

799

Microbiology Ecology 62, 242-257.

32

800 801

Hink, L., Lycus, P., Gubry-Rangin, C., Frostgård, Ǻ., Nicol, G.W., Prosser, J.I.,

802

Bakken, L.R., 2017. Kinetics of NH3‐oxidation, NO‐turnover, N2O‐production and

803

electron flow during oxygen depletion in model bacterial and archaeal ammonia

804

oxidisers. Environmental Microbiology 19, 4882-4896.

805 806

Hink, L., Gubry-Rangin, C., Nicol, G.W., Prosser, J.I., 2018. The consequences of

807

niche and physiological differentiation of archaeal and bacterial ammonia oxidisers

808

for nitrous oxide emissions. The ISME Journal 12, 1084-1093.

809 810

Hu, H., Zhang, L.M., Yuan, C.L., Di, H.J., He, Z.Z., 2013. pH-dependent distribution

811

of soil ammonia oxidizers across a large geographical scale as revealed by high-

812

throughput pyrosequencing. Journal of Soils and Sediments 13, 1439-1449.

813 814

Hu, B.L., Liu, S., Wang, W., Shen, L., Lou, L., Liu, W., Tian, G., Xu, X., Zheng, P.,

815

2014. pH-dominated niche segregation of ammonia-oxidising microorganisms in

816

Chinese agricultural soils. FEMS Microbiology Ecology 90, 290-299.

817 818

Hu, H.W., Zhang, L.M., Yuan, C.L., Zheng, Y., Wang, J.T., Chen, D., He, J-Z., 2015.

819

The large scale distribution of ammonia oxidisers in paddy soils is driven by pH,

820

geographic distance and climatic factors. Frontiers in Microbiology 6, 938.

821 822

Hu H.W., He J.Z., 2017. Comammox-a newly discovered nitrification process in the

823

terrestrial nitrogen cycle. Journal of Soils and Sediments 17, 2709-2717.

824

33

825

Hutchinson, G.L., 1995. Biosphere-atmosphere exchange of gaseous N oxides. In:

826

Lal r, Kimble J, Levine e, Stewart BA (eds) Soils and global change. CRC Press,

827

Boca Raton, pp 219-236.;

828

Jia, Z., Conrad, R., 2009. Bacteria rather than Archaea dominate microbial ammonia

829

oxidation in an agricultural soil. Environmental Microbiology 11, 1658-1671.

830 831

Jiang, H., Huang, L.Q., Deng, Y., Wang, S., Zhou, Y., Liu, L., Dong, H., 2014.

832

Latitudinal distribution of ammonia-oxidizing bacteria and archaea in the agricultural

833

soils of Eastern China. Applied and Environmental Microbiology 80, 5593-5602.

834 835

Johnson, K.S., Coletti, L.J., 2002. In situ ultraviolet spectrophotometry for high

836

resolution and long term monitoring of nitrate bromide and bisulfide in the oceans.

837

Deep Sea Research Part I Oceanographic Research Papers 49, 1291-1305.

838 839

Joshi, N.A., Fass, J.N., 2011. Sickle: A sliding-window, adaptive, quality-based

840

trimming tool for FastQ files, Version 1.33 [Software].

841 842

Jung, M.Y., Kim, J.G., Sinninghe Damsté, J.S., Rijpstra, W.I.C., Madsen, E.L., Kim,

843

S.J., Hong, H., Si, O.J., Kerou, M., Schleper, C., Rhee, S.K., 2016. A hydrophobic

844

ammonia oxidizing archaeon of the Nitrosocosmicus clade isolated from coal tar

845

contaminated sediment. Environmental Microbiology Reports 8, 983-992.

846 847

King, D., Nedwell, D.B., 1984. Changes in the nitrate-reducing community of an

848

anaerobic saltmarsh sediment in response to seasonal selection by temperature.

849

Journal of General Microbiology 130, 2935-2941.

34

850 851

Kirkham,

D.,

Bartholomew,

W.V.,

1954.

Equations

for

following

nutrient

852

transformations in soil, utilizing tracer data. Soil Science Society of America,

853

Proceedings 18, 33-34.

854 855

Kits, K.D., Sedlacek, C.J., Lebedeva, E.V., Han, P., Bulaev, A., Pjevac, P., Daebeler,

856

A., Romano, S., Stein, L.Y., Daims, H., Wagner, M., 2017. Kinetic analysis of a

857

complete nitrifier reveals an oligotrophic lifestyle. Nature 549, 269-272.

858 859

Könneke, M., Bernhard, A.E., de la Torre, J.R., Walker, C.B., Waterbury, J.B., Stahl,

860

D.A., 2005. Isolation of an autotrophic ammonia-oxidizing marine archaeon. Nature

861

437, 543–546.

862 863

Koper, T.E., Stark, J.M., Habteselassie, M.Y., Norton, J.M., 2010. Nitrification

864

exhibits Haldane kinetics in an agricultural soil treated with ammonium sulfate or

865

dairy waste compost. FEMS Microbiology Ecology 74, 316-322.

866 867

Krom, M.J.D., 1980. Spectrophotometric determination of ammonia: a study of a

868

modified Berthelot reaction using salicylate and dichloroisocyanurate. Analyst. 105,

869

305-317.

870 871

Lansdown, K., McKew, B.A., Whitby, C., Dumbrell, A.J., Heppell, C.M., Binley, A.,

872

Olde, L., Trimmer, M., 2016. Importance and controls of anaerobic ammonium

873

oxidation influenced by riverbed geology. Nature Geosciences 9, 357-360.

874

35

875

Le S.Q., Gascuel, O., 2008. An improved general amino acid replacement matrix.

876

Molecular Biology and Evololution 25, 1307-1320.

877 878

Lehtovirta-Morley, L.E., Stoecker, K., Vilcinskas, A., Prosser, J.I., Nicol, G.W.. 2011.

879

Cultivation of an acidophilic ammonia oxidizer from a nitrifying acid soil. Proceedings

880

of the National Academy of Sciences 108, 15892-15897.

881 882

Lehtovirta-Morley, L.E., Ge, C., Ross, J., Yao, H., Nicol, G.W., Prosser, J.I., 2014.

883

Characterisation of terrestrial acidophilic archaeal ammonia oxidisers and their

884

inhibition and stimulation by organic compounds. FEMS Microbiology Ecology 89,

885

542-552.

886 887

Lehtovirta-Morley, L.E., Ross, J., Hink, L., Weber, E.B., Gubry-Rangin, C., Thion, C.,

888

Prosser, J.I., Nicol, G.W., 2016. Isolation of ‘Candidatus Nitrosocosmicus

889

franklandus’, a novel ureolytic soil archaeal ammonia oxidiser with tolerance to high

890

ammonia concentration. FEMS Microbiology Ecology 92, fiw057.

891 892

Leininger, S., Schloter, U.T., Schwark, I., Qi, J., Nicol, G.W., Prosser, J.I., Schuster,

893

S.C., Schleper, C., 2006. Archaea predominate among ammonia-oxidizing

894

prokaryotes in soils. Nature 442, 806-809.

895 896

Li, J., Nedwell, D.B., Beddow, J., Dumbrell, A.J., McKew, B.A., Thorpe, E.L., Whitby,

897

C., 2015. amoA gene abundances and nitrification potential rates suggest that

898

benthic ammonia-oxidizing bacteria (AOB) not archaea (AOA) dominate N cycling in

899

the Colne estuary, UK. Applied and Environmental Microbiology 81, 159-165.

36

900

Liu, J., Yu, Z., Yao, Q., Sui, Y., Shi, Y., Chu, H., et al., 2018. Ammonia-oxidizing

901

archaea show more distinct biogeographic distribution patterns than ammonia-

902

oxidizing bacteria across the black soil zone of Northeast China. Frontiers in

903

Microbiology 9, 171.

904 905

Martens-Habbena, W., Berube, P.M., Urakawa, H., José, R., Stahl, D.A., 2009.

906

Ammonia oxidation kinetics determine niche separation of nitrifying Archaea and

907

Bacteria. Nature 461, 976–979.

908 909

McMurray, C., Smith, R., Crone, V., Foy, B., Gallagher, S., Gibson, C., Hale, P.,

910

Jordan, C., Ramsay, R., Nelson, G., Stevens, J., 2002. Report on the Environmental

911

Aspects of the Nitrates Directive in Northern Ireland, p.6.

912 913

Mørkved, P.T., Dörsch, P., Bakken, L.R., 2007. The N2O product ratio of nitrification

914

and its dependence on long-term changes in soil pH. Soil Biology and Biochemistry

915

39, 2048-2057.

916 917

Myers, R.J.K., 1975. Temperature effects on mineralization and nitrification in a

918

tropical soil. Soil Biology and Biochemistry 7, 83-86.

919 920

Myrold, D.D., Tiedje, J.M., 1986. Simultaneous estimation of several nitrogen cycle

921

rates using 15N: theory and application. Soil Biology and Biochemistry 18, 559-568.

922 923

Natural England Report, 2008 Biodiversity pg 1-126.

37

924

Nedwell, D.B., 1999. Effect of low temperature on microbial growth: lowered affinity

925

for substrates limits growth at low temperature. FEMS Microbiology Ecology 30, 101-

926

111.

927 928

Nedwell, D.B., Dong, L.F., Sage, A., Underwood, G.C., 2002. Variations of the

929

nutrient loads to the mainland UK estuaries: correlation with catchment areas,

930

urbanization and coastal eutrophication. Estuarine, Coastal and Shelf Science, 54,

931

951-970.

932 933

Nikolenko, S.I., Korobeynikov, A.I., Alekseyev, M.A., 2013, BayesHammer: Bayesian

934

clustering for error correction in single-cell sequencing. BMC Genomics 14, S7.

935 936

Nurk, S., Bankevich, A., Antipov, D., Gurevich, A.A., Korobeynikov, A., Lapidus, A.,

937

Prijibelski, A.D., pyshkin, A., Sirotkin, Y., Stepanauskas, R., Clingenpeel, S.R.,

938

Woyke, T., mcLean, J.S., Lasken, R., Tesler, G., Alekseyev, M.A., Pevzner, P.A.,

939

2013, Assembling single-cell genomes and mini-metagenomes from chimeric MDA

940

products. Journal of Computational Biology 20, 714-737.

941 942

Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O’Hara, R.B.,

943

Simpson, G.L., Solymos, P., Stevens, M.H.H., Wagner, H., 2017. Package ‘vegan’.

944

Community Ecology package, (R package version 2.4-5).

945 946

Ostle N.J., Levy, P.E., Evans, C.D., Smith, P., 2009. UK land use and soil carbon

947

sequestration. Land Use Policy 265, S274-S283.

948

38

949

Ouyang, Y., Norton, J.M., Stark, J.M., Reeve, J.R., Habteselassie, M.Y., 2016,

950

Ammonia-oxidizing bacteria are more responsive than archaea to nitrogen source in

951

an agricultural soil. Soil Biology and Biochemistry 96, 4-15.

952 953

Ouyang, Y., Norton, J.M., Stark, J.M., 2017. Ammonium availability and temperature

954

control contributions of ammonia oxidizing bacteria and archaea to nitrification in an

955

agricultural soil. Soil Biology and Biochemistry 113, 161-172.

956 957

Palomo A., Pedersen, A.G., Fowler, S.J., Dechesne, A., Sicheritz-Ponten, T., Smets,

958

B.F., 2018. Comparative genomics sheds light on niche differentiation and the

959

evolutionary history of comammox Nitrospira. The ISME Journal 12, 1779-1793.

960 961

Park, H.D., Wells, G.F., Bae, H., Criddle, C.S., Francis, C.A., 2006. Occurrence of

962

ammonia-oxidising archaea in wastewater treatment plant bioreactors. Applied and

963

Environmental Microbiology 72, 5643-5647.

964 965

Pirt, SJ., 1975. Principles of microbe and cell cultivation. Blackwell Scientific

966

Publications, Oxford. Pp. 224.

967 968

Prosser, J.I., Nicol, G.W., 2008. Relative contributions of archaea and bacteria to

969

aerobic ammonia oxidation in the environment. Environmental Microbiology 10,

970

2931–2941.

971

39

972

Prosser, J.I., Nicol, G.W., 2012. Archaeal and bacterial ammonia oxidisers in soil:

973

the quest for niche specialisation and differentiation. Trends in Microbiology 20, 523-

974

531.

975 976

Raskin, L., Stromley, J.M., Rittmann, B.E., Stahl, D.A., 1994. Group-specific 16S

977

rRNA hybridization probes to describe natural communities of methanogens. Applied

978

and Environmental Microbiology 60, 1232-1240.

979 980

Rodwell, J.S., Morgan, V., Jefferson, R.G., Moss, D., 2007. The European context of

981

British lowland grasslands. Joint Nature Conservation Committee JNCC Report No.

982

394. ISSN 0963 8901.

983 984

Rognes, T., Flouri, T., Nichols, B., Quince, C., Mahe, F., 2016. VSEARCH: a

985

versatile open source tool for metagenomics. PeerJ 4, e2584.

986 987

Rotthauwe, J.H., Witzel, K.P., Liesack, W., 1997. The ammonia monooxygenase

988

structural gene amoA as a functional marker: molecular fine-scale analysis of natural

989

ammonia-oxidizing populations. yApplied and Environmental Microbiology 63, 4704-

990

4712.

991 992

Sauder, L.A., Albertsen, M., Engel, K., Schwarz, J., Nielsen, P.H., Wagner, M.,

993

Neufeld, J.D., 2017. Cultivation and characterisation of Candidatus Nitrosocosmicus

994

exaquare, an ammonia-oxidizing archaeon from a municipal wastewater treatment

995

system. The ISME Journal 11, 1142-1157.

996

40

997

Smith, J.L., Halvorson, J.J., Bolton, H. Jr., 2002. Soil properties and microbial activity

998

across a 500m elevation gradient in a semi-arid environment. Soil Biology and

999

Biochemistry 34, 1749-1757.

1000 1001

Stahl, D.A., Amann, R., 1991. Development and application of nucleic acid probes,

1002

p. 205-248. In E. Stackebrandt and M. Goodfellow (ed.), Nucleic acid techniques in

1003

bacterial systematics. John Wiley & Sons Ltd., Chichester, England.

1004 1005

Tamura, K., Stecher, G., Peterson, D., Filipski, A., Kumar, S., 2013. MEGA6:

1006

molecular evolutionary genetics analysis version 6.0. Molecular Biology and

1007

Evolution 30, 2725-2729.

1008 1009

Tourna, M., Freitag, T.E., Nicol, G.W., Prosser, J.I., 2008. Growth, activity and

1010

temperature responses of ammonia‐oxidizing archaea and bacteria in soil

1011

microcosms. Environmental Microbiology 10, 1357-1364.

1012 1013

Treusch, A.H., Leininger, S., Kletzin, A., Schuster, S.C., Klenk, H.P., Schleper, C.,

1014

2005. Novel genes for nitrite reductase and AMo-related proteins indicate a role of

1015

uncultivated

1016

Microbiology 7, 1985-1995.

mesophilic

crenarchaeota

in

nitrogen

cycling.

Environmental

1017 1018

Verhamme, D.T., Prosser, J.I., Nicol, G.W., 2011. Ammonia concentration

1019

determines differential growth of ammonia oxidizing archaeal and bacteria in soil

1020

microcosms. The ISME Journal 5, 1067–1071.

1021

41

1022

Von Lutzow, M., and Kogel-KNabner, I., 2009. Temperature sensitivity of soil organic

1023

matter decomposition-what do we know? Biology and Fertility of Soils 46, 1-15.

1024 1025

Walker, C.B., de la Torre, J.R., Klotz, M.G., Urakawa, H., Pinela, N., Arp, D.J., 2010.

1026

Nitrosopumilus maritimus genome reveals unique mechnaisms for nitrification and

1027

autotrophy in globally distributed marine crenarchaea. Proceedings of the National

1028

Academy of Sciences of the United States of America 107, 8818-8823.

1029 1030

Wertz, S., Leigh, A.,K., Grayston, S.J., 2012. effects of long term fertilization of forest

1031

soils on potential nitrification and on the abundance and community structure of

1032

ammonia oxidizers and nitrite oxidizers. FEMS Microbiology Ecology 79, 142-154.

1033 1034

Xie, Z., Le Roux, X., Wang, C., Gu, Z., An, M., Nan, H., Checn, B., Li, F., Liu, Y., Du,

1035

G., Feng, H., Ma, X., 2014. Identifying response groups of soil nitrifiers and

1036

denitrifiers to grazing and associated soil environmental drivers in Tibetan alpine

1037

meadows. Soil Biology and Biochemistry 77, 89-99.

1038 1039

Yao, H., Bowman, D., Shi, W., 2011. Seasonal variations of soil microbial biomass

1040

and acticvity in warm- and cool-season turfgrass systems. Soil Biology and

1041

Biochemistry. 43, 1536-1543.

1042 1043

Yao, H., Campbell, C.D., Chapman, S.J., Freitag, T.E., Nicol, G.W., Singh, B.K.,

1044

2013. Multi-factorial drivers of ammonia oxidizer communities: evidence from a

1045

national soil survey. Environmental Microbiology 15, 2545-2556.

42

1046

Zhalnina, K., de Quadros, P.D., Camargo, F.A., Triplett, E.W., 2012. Drivers of

1047

Archaeal ammonia-oxidizing communities in soil. Frontiers in Microbiology 3, 210.

1048 1049

Zhang, S., Chen, D., Sun, D., Wang, X., Smith, J.L., Du G., 2012. Impacts of altitude

1050

and position on the rates of soil nitrogen mineralization and nitrification in alpine

1051

meadows on the eastern Qinghai-Tibetan Plateau, China. Biology and Fertility of

1052

Soils 48, 393-400.

1053 1054

Wang, Q., Garrity, G.M., Tiedje, J.M., Cole, J.R., 2007. Naive Bayesian classifier for

1055

rapid assignment of rRNA sequences into the new bacterial taxonomy. Applied and

1056

Environmental Microbiology 73, 5261-5267.

1057 1058

Table and Figure Legends

1059

Fig. 1. Map of sampling locations within the UK (A), and within the Hampshire-Avon

1060

catchment (B). The catchment area is shaded dark grey, whilst river channels are

1061

blue. Further geology and catchment details are given in (C).

1062 1063

Fig. 2. The nitrification and mineralization rates, and nitrate and ammonium pool

1064

turnover times for soils on different underlying geologies.

1065 1066

Fig. 3. The abundance of ammonia oxidiser amoA genes. Panels A and B show the

1067

abundance of ammonia oxidising Archaea and Bacteria respectively, across

1068

sampling months and underlying geologies; panel C shows the ratio of ammonia

1069

oxidising Archaea to Bacteria; panel D shows PCA correlation analysis of AOB and

1070

AOA abundance (as measured by qPCR), with specific environmental factors

43

1071

(geology, season) and vectors (soil ammonium, nitrite and nitrate concentrations)

1072

overlaid.

1073 1074

Fig. 4. A phylogenetic tree based on ammonia oxidising archaeal AmoA amino acid

1075

sequences. OTU groups consist of multiple OTUs with identical amino acid

1076

sequences, and additional AmoA sequences were obtained from BLAST and

1077

Fungene searches. The right-hand panels show the relative abundances of OTUs in

1078

OTU groups 1, group 2, and group 3 across the three geologies.

1079 1080

Fig. 5. A phylogenetic tree based on ammonia oxidising bacterial AmoA amino acid

1081

sequences. OTU groups consist of multiple OTUs with identical amino acid

1082

sequences, and additional AmoA sequences were obtained from BLAST and

1083

Fungene searches. The relative abundances of OTUs in each of the OTU groups are

1084

shown in the right-hand panels.

1085 1086

Table 1. Soil organic carbon (SOC) and soil organic nitrogen (SON) concentrations,

1087

SON mineralization rate, SON pool turnover times, at 0-10cm soil depth across

1088

geologies and sampling period (overall mean ± SE).

1089 1090 1091

Table 2. Soil ammonium, nitrate and phosphate concentrations, mineralization and

1092

nitrification rates across geologies and sampling period.

1093

.

1094 1095

44

1096 1097 1098 1099

Supplementary Table and Figure Legends

1100

Fig. S1. Non-metric multidimensional scaling analyses of (A) archaeal 16S rRNA

1101

gene sequences, (B) bacterial 16S rRNA gene sequences, (C) ammonia oxidising

1102

archaeal (AOA) amoA, and (D) ammonia oxidising bacterial (AOB) amoA gene

1103

sequences. Points close together indicate compositionally similar communities. The

1104

OTU richness of ammonia oxidising Archaea (E), and Bacteria (F) are also shown.

1105

Square points in panels E and F represent mean OTU richness, whilst error bars

1106

represent one standard error of the mean.

1107 1108

Fig. S2 Temporal and spatial changes in the abundance of archaeal genera based

1109

on the archaeal 16S rRNA gene library.

1110 1111

Fig. S3 Relative abundance of most abundant ammonia oxidising archaeal (AOA)

1112

OTUs within the archaeal 16S rRNA gene library.

1113 1114

Fig. S4. The relative abundances of bacterial genera that differed in abundance

1115

between sampling seasons (P < 0.05 in all cases).

1116 1117

Fig. S5 Changes in the most abundant bacterial 16S rRNA gene sequences with

1118

geology.

1119

45

1120

Fig. S6 Relative abundance of the two ammonia oxidising bacterial (AOB) OTUs

1121

detected within the bacterial 16S rRNA gene library.

1122 1123

Fig. S7 The ratio of AOA:AOB (as determined by amoA qPCR) in relation to

1124

ammonium and nitrate concentrations. p(AOA) indicates the probability that a

1125

random ammonia oxidiser is archaeal, higher values therefore indicate a greater

1126

ratio of AOA:AOB. Solid black lines indicate “global” fit (across all geologies) of a

1127

binomial generalised linear mixed effects model. Other lines show geology-specific

1128

relationships as indicated in the legend.

1129

Fig. S8. The relative abundances of ammonia-oxidising (A) archaeal and (B)

1130

bacterial amoA OTU groups across the sampled seasons and geologies. Each OTU

1131

group consists of OTUs with identical amoA amino acid sequences, as identified

1132

through phylogenetic analyses.

1133 1134

Supplementary Table S1. Results of post-hoc Tukey comparisons of ammonia-

1135

oxidizing archaea (AOA) and bacteria (AOB) amoA gene copy numbers between

1136

soils across sites.

1137 1138

Supplementary Table S2. The identities of OTUs forming each of the amino acid

1139

variant “OTU groups” identified by phylogenetic analyses of ammonia oxidizing

1140

archaeal (AOA) amoA genes.

1141

46

1142

Supplementary Table S3. The identities of OTUs forming each of the amino acid

1143

variant “OTU groups” identified by phylogenetic analyses of ammonia oxidizing

1144

bacterial (AOB) amoA genes.

1145 1146

47

1147

Table 1. Soil organic carbon (SOC) and soil organic nitrogen (SON) concentrations, SON mineralization rate, SON pool turnover

1148

times, at 0-10cm soil depth across geologies and sampling period (overall mean ± SE).

1149 Site

pH

SOC

SON

C:N g/g

(mg C g-1soil)

(mg N g-1soil)

ratio

Mean SON

SON pool

mineralization rate

turnover time

(µg N g-1 soil day-1)

(years)

Clay

4.9 ± 0.2

52.4 ± 6.62

5.38 ± 1.56

9.74 ± 0.3

4.6

3.2

Greensand

7.1 ± 0.3

74.64 ± 16.69

7.62 ± 1.56

9.8 ± 0.4

5.9

3.5

Chalk

7.6 ± 0.2

127.50 ± 1.41

10.11 ± 3.75

12.6 ± 4.5

8.3

3.3

1150 1151 1152 1153 1154 1155

48

1156

Table 2. Soil ammonium, nitrate and phosphate concentrations, mineralization and nitrification rates across geologies and sampling

1157

period.

1158 Site/Season

Substrate/Process (± SE)

Summer

Autumn

Winter

Spring

Mean

Clay

Nitrate (µg N g-1 dry soil)

13.90 ±2.16

16.63 ±0.98

10.36 ±0.44

8.10 ±1.17

12.25 ±1.12

0.07 ±0.01 0.36 ±0.33 4.19 ±1.02 6.82 ±2.08

0 ±0.0 1.04 ±0.05 2.88 ±0.17 1.65 ±0.51

0.03 ±0.02 1.94 ±0.41 3.86 ±0.68 6.69 ±0.89

0.86 ±0.01 1.52 ±0.06 4.67 ±0.25 4.64 ±0.71

0.24 ±0.01 1.22 ±0.21 3.9 ±0.53 4.95 ±1.05

162 25.25 ±1.77 0.06 ±0.01

57 28.02 ±6.27 0 ±0.0

173 14.39 ±2.03 0.6 ±0.02

100 15.30 ±3.63 0.71 ±0.09

123 20.74 ±3.43 0.34 ±0.12

2.23 ±2.18 6.33 ±0.40 3.95 ±2.28

1.15 ±0.13 4.90 ±0.34 2.62 ±1.22

1.55 ±0.34 5.42 ±0.27 2.21 ±0.44

1.82 ±0.27 6.96 ±0.22 4.03 ±0.62

1.69 ±0.73 5.90 ±1.23 12.81 ±1.14

62 29.37 ±13.29 0.09 ±0.01 3.58 ± 3.40 9.33 ±0.59 6.00 ±1.23

53 24.72 ±0.80 0 ±0.0 1.91 ±0.19 7.23 ±0.40 2.61 ±0.75

40 17.31 ±1.32 0.77 ±0.05 4.29 ±1.26 7.74 ±1.18 7.82 ±3.17

58 19.77 ±4.96 3.88 ±0.09 3.09 ±0.21 8.72 ±0.86 8.37 ±1.93

53 22.79 ±5.09 1.19 ±0.04 3.22 ±1.27 8.23 ±0.76 24.80 ±1.77

64

36

101

96

74

-1

Greensand

Chalk

Phosphate (µmol g dry soil) Ammonium (µg N g-1 dry soil) Mineralization (µg N g-1 dry soil day-1) Nitrification (µg N g-1 dry soil day-1) Proportion of N mineralized then nitrified (%) Nitrate (µg N g-1 dry soil) Phosphate (µmol g-1 dry soil) Ammonium (µg N g-1 dry soil) Mineralization (µg N g-1 dry soil day-1) Nitrification (µg N g-1 dry soil day-1) Proportion of N mineralized then nitrified (%) Nitrate (µg N g-1 dry soil) Phosphate (µmol g-1 dry soil) Ammonium (µg N g-1 dry soil) Mineralization (µg N g-1 dry soil day-1) Nitrification (µg N g-1 dry soil day-1) Proportion of N mineralized then nitrified (%)

1159

49

A

B 51.4°N

Greensand

Latitude (decimal degrees)

N

Chalk

51.2°N

51°N

Clay

50.8°N

50.6°N 20 km

2.6°W 2.4°W 2.2°W 2°W 1.8°W 1.6°W 1.4°W Longitude (decimal degrees)

C

Clay site

Greensand site

Chalk site

Underlying geologya

>99% Kimmeridge Clay 50% Upper Greensand

80% Chalk

Adjacent rivera

Sem

Avon

Wylye

Major land usea

Grassland (90−95%), Arable (5−10%)

Grassland (50%), Arable (25%)

Grassland (35%), Arable (50%)

Catchment sizea

4.9 km2

59.2 km2

53.5 km2

Soil pH

4.9 ± 0.2

7.1 ± 0.3

7.6 ± 0.2

a

Data modified from Heppell et al ., 2017

B

AOB amoA gene copies g−1

AOA amoA gene copies g−1

A 1e+08

1e+07

1e+06

Summer Autumn Winter

1e+08

1e+07

1e+06 Geology

Spring

Summer Autumn Winter

Season C

Spring

Season

50

D

Month

4

Summer Autumn Winter Spring

NH+ 4

2 30

PC2

AOA:AOB ratio

40

20

AOB NO− 2

0

10 NO− 3

−2

AOA

0 Summer Autumn Winter

Season

Spring

Clay Greensand Chalk

−1

0

1

PC1

2

3

OTU group 1 OTU group 1

1.00

KU290366 C and i d at us N i t rosocosmi cus f rank l and us C13

0.75

CP002408 C and i d at us N i t rososph aera gargensi s Ga9.2

0.50

EU281318 C and i d at us N i t rososph aera gargensi s

0.25

OTU group 2

0.00

85

FR773159 C and i d at us N i t rososph aera vi ennensi s EN76 CP007174 C and i d at us N i t rososph aera evergl ad ensi s SR1

OTU group 3 KF957666 C and i d at us N i t rosopumi l us sp. PS0 KF957665 C and i d at us N i t rosopumi l us sp. HCA1 CP021324 C and i d at us N i t rosomari nus cat al i na SPOT01 66 AEXL02000009 C and i d at us N i t rosopumi l us sal ari a BD31 50 57

CP010868 C and i d at us N i t rosopumi l us pi ranensi s D3C CP011070 C and i d at us N i t rosopumi l us ad ri at i cus NF5

HQ331117 C and i d at us N i t rosoarch aeum k oreensi s MY1 57 AEGP01000066 C and i d at us N i t rosoarch aeum l i mni a SFB1 KX034182 C and i d at us N i t rosot enui s sp. AQ6f

0.20 0.15 0.10 0.05 0.00 OTU group 3

1.00 0.75 0.50 0.25 0.00

CP011097 C and i d at us N i t rosot enui s cl oacae SAT1 LN890280 C and i d at us N i t rosot al ea d evanat erra

G

C

85

OTU group 2

re lay en sa nd C ha lk

94

Relative abundance

0.01

Geology

OTU group 1

0.10

EF175097 N i t rosospi ra sp. En13

0.75

OTU group 1

0.50

0.05

AJ298720 N i t rosospi ra t enui s

0.25

DQ228465 N i t rosovi b ri o sp. FJI82

95

EF175099 N i t rosospi ra sp. Wyke8

OTU group 4 OTU group 5 OTU group 6

63

60 88 77

0.09

HM345612 N i t rosomonas mari na AF272403 N i t rosomonas ureae

AF272400 N i t rosomonas aest uari i AF272406 N i t rosomonas ol i got roph a

100

OTU group 3

1.00

0.4 0.3

0.75

0.2

0.50

0.1

0.25

0.0

0.00 OTU group 5

0.08

0.75

0.06

0.50

0.04

0.25

0.02

0.00

0.00

AB900134 N i t rosomonas st ercori s AF272404 N i t rosomonas ni t rosa

100 CP002086 N i t rosococcus wat soni i C − 113 U96611 N i t rosococcus oceani ATCC 19707

re lay en sa nd C ha lk

AL954747 N i t rosomonas europaea ATCC 19718 64

C

AF272398 N i t rosomonas h al oph i l a

G

63

OTU group 6

1.00

AF314753 N i t rosomonas cryot ol erans ATCC 49181 51 CP000450 N i t rosomonas eut roph a C91 AF272399 N i t rosomonas communi s

OTU group 4

re lay en sa nd C ha lk

CP021106 N i t rosospi ra l acus APG3

0.00

C

OTU group 3

Relative abundance

EF175098 N i t rosospi ra sp. Wyke2

0.00

G

OTU group 2 CP012371 N i t rosospi ra b ri ensi s C − 128

OTU group 2

1.00

78 DQ228466 N i t rosovi b ri o sp. RY3C KU747133 N i t rosospi ra mul t i f ormi s NI13

Geology

Clay

Clay

10.0

7.5

7.5

5.0

5.0

) −1

dry soil day

2.5 0.0 Greensand

−1

10.0

Nitrification rate (μg N g

7.5 5.0 2.5 0.0 Chalk

10.0 7.5 5.0 2.5

2.5 0.0 Greensand

10.0 7.5 5.0 2.5 0.0 Chalk

10.0 7.5 5.0 2.5

0.0 Clay Greensand Chalk

9 6 3 Summer Autumn 20°C 7°C

Winter 7°C

Season

Spring 19°C

+ NH4 pool turnover (days)

NO 3 pool turnover (days)

Mineralisation rate (μg N g

−1

dry soil day

−1

)

10.0

0.0 0.8

Clay Greensand Chalk

0.6 0.4 0.2

Summer Autumn 20°C 7°C

Winter 7°C

Season

Spring 19°C

1

Mineralization and nitrification: Archaea dominate ammonia-oxidising communities

2

in grassland soils.

3 4

Authors: Dave R. Clarka, Boyd A. McKewa, Liang F. Donga, Garwai Leunga, Alex J.

5

Dumbrella, Andrew Stottb, Helen Grant b, David B. Nedwella, Mark Trimmerc, Corinne

6

Whitbya*

7 8

Affiliations:

9

a

10

b

11

Environment Centre, Lancaster, LA1 4AP, UK

12

c

13

Road, London E1 4NS, UK.

School of Life Sciences, University of Essex, Colchester CO4 3SQ, UK Life Sciences Mass Spectrometry Facility, Centre for Ecology & Hydrology, Lancaster,

School of Biological and Chemical Sciences, Queen Mary University of London, Mile End

14 15 16

*Corresponding author:

17

Corinne Whitby [email protected]

18

School of Life Sciences, University of Essex, Colchester CO4 3SQ, UK

19

Tel: +44 (0) 1206 872062

20

Fax: +44 (0) 1206 872592

21 22

Keywords: Nitrification, N mineralization, ammonia oxidising archaea, Nitrososphaera,

23

amoA gene, grasslands

24 25

Declarations of interest: none

26

1

27

Highlights

28

•

Turnover times for soil ammonium were <24 h, whilst several days for nitrate.

29

•

AOA and AOB communities in clay were clearly distinct from those in chalk and greensand soils.

30 31

•

oxidising bacteria.

32 33

•

36

AOA abundance positively correlated with nitrate, whereas AOB abundance correlated with ammonium and nitrite concentrations

34 35

Generally, ammonia oxidising archaea were more abundant than ammonia

•

This study showed additional evidence for niche differentiation among ammoniaoxidisers in grasslands.

37

2

Declaration of interests: None ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. ☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: