TY - JOUR
AU - Zopfi,, Jakob
AB - ABSTRACT In contrast to the pervasive occurrence of denitrification in soils, anammox (anaerobic ammonium oxidation) is a spatially restricted process that depends on specific ecological conditions. To identify the factors that constrain the distribution and activity of anammox bacteria in terrestrial environments, we investigated four different soil types along a catena with opposing ecological gradients of nitrogen and water content, from an amended pasture to an ombrotrophic bog. Anammox was detected by polymerase chain reaction (PCR) and quantitative PCR (qPCR) only in the nitrophilic wet meadow and the minerotrophic fen, in soil sections remaining water-saturated for most of the year and whose interstitial water contained inorganic nitrogen. Contrastingly, aerobic ammonia oxidizing microorganisms were present in all examined samples and outnumbered anammox bacteria usually by at least one order of magnitude. 16S rRNA gene sequencing revealed a relatively high diversity of anammox bacteria with one Ca. Brocadia cluster. Three additional clusters could not be affiliated to known anammox genera, but have been previously detected in other soil systems. Soil incubations using 15N-labeled substrates revealed that anammox processes contributed about <2% to total N2 formation, leaving nitrification and denitrification as the dominant N-removal mechanism in these soils that represent important buffer zones between agricultural land and ombrotrophic peat bogs. nitrogen cycling, ammonia oxidation, wetland, ecotone, generalized linear model INTRODUCTION Microbial nitrogen (N) transformations in soils have a major influence on agricultural production (Hofstra and Bouwman 2005), water quality (Vitousek et al. 1997) and the emission of greenhouse gases such as nitrous oxide (Butterbach-Bahl et al. 2013). Our notion, however, of microbial N-cycling and its interactions with other element cycles has changed greatly in recent years. Novel processes have been discovered such as denitrification by eukaryotes (Finlay, Span and Harman 1983), denitrification coupled to methane oxidation (Raghoebarsing et al. 2006), archaeal ammonium oxidation (Könneke et al. 2005), complete nitrification by a single organism (Daims et al. 2015; van Kessel et al. 2015), as well as anaerobic ammonium oxidation with nitrite (anammox) or iron (Feammox) as oxidants (Javanaud et al. 2011; Yang, Weber and Silver 2012). The significance of anammox in natural environments was first recognized in marine sediments (Thamdrup and Dalsgaard 2002), oceanic oxygen minimum zones (Kuypers et al. 2003) and later in chemoclines of permanently stratified freshwater lakes (Schubert et al. 2006; Wenk et al. 2013). Sequences of anammox bacteria were also detected in a wide range of terrestrial ecosystems including water-saturated soils of wetlands and riparian zones, aquifers (e.g. Humbert et al. 2010a; Humbert, Zopfi and Tarnawski 2012; Kumar et al. 2017), fertilized paddies (e.g. Zhu et al. 2011) and agricultural soils (Long et al. 2013). The quantitative significance of anammox in terrestrial systems has been investigated by comparatively few studies (Humbert et al. 2010a, 2012; Zhu et al. 2011; Sato et al. 2012; Long et al. 2013; Naeher et al. 2015; Yang et al. 2015; Shan et al. 2016; Shen et al. 2016 2016; Xi et al. 2016). They revealed that in terrestrial systems, anammox bacteria are usually less active than in marine systems, despite the greater phylogenetic diversity. Anammox activity and/or abundance appear to increase (Humbert, Zopfi and Tarnawski 2012; Meng et al. 2016) or to decrease with soil depth (Sher et al. 2012; Hui et al. 2017), possibly depending on anoxic conditions (Humbert et al. 2010b; Long et al. 2013) and the availability of the required substrates. For instance, the importance of anammox seems to increase with the availability of ammonium (e.g. Sher et al. 2012; Shen, Xu and He 2014; Shen et al. 2015, 2016) or nitrate (Humbert et al. 2010b; Yang et al. 2015; Shan et al. 2016). Other factors such as soil organic matter content, C:N-ratio and soil pH have also been found to correlate with anammox activity (e.g. Yang et al. 2015; Zhou et al. 2017). To date, however, identification of general environmental factors that constrain anammox abundance and/or activity in terrestrial ecosystems is still difficult. Possible reasons include the facts that (i) studies often focus on one type of soil, (ii) they consider only few environmental variables, and (iii) some soil types are over-represented, including agricultural soil, rice paddies and wetlands (Table 1). Table 1. Literature review of anammox diversity, abundance and activity in soils where this metabolism has been detected. Open in new tab Table 1. Literature review of anammox diversity, abundance and activity in soils where this metabolism has been detected. Open in new tab In order to address this, we investigated anammox bacteria (together with other ammonia oxidizers) along a catena ranging from an amended pasture to a peat bog. This system represents an ecotone (Attrill and Rundle 2002) and acts as an important buffer zone, mitigating N-loadings from agriculture or other human activities to the environment. We investigated four different soil types that characterize this catena and related the occurrence of anammox to soil characteristics and a wide range of environmental parameters, monitored during an annual cycle. We used exploratory and predictive statistics to determine the environmental factors that explain best the occurrence and the abundance of anammox. Furthermore, anammox was compared with other N-cycling microbial metabolisms, such as aerobic ammonia oxidation (through qPCR analyses) and denitrification (through 15N incubation analyses). MATERIAL AND METHODS Study site The investigated catena is located close to Bellefontaine in the French part of the Jura Mountains (Fig. 1). Four different stations were selected based on the vegetation map of Gallandat (1982): an amended pasture (Soil 1; N 46°34’12.30'', E 006°04’50.94'', WGS84), a nitrophilic wet meadow (Soil 2; N 46°34’11.40, E 006°04’51.00''), a minerotrophic fen (Soil 3; N 46°34’8.76'', E 006°04’53.28'') and an ombrotrophic peat bog (Soil 4; N 46°34’6.36'', E 006°04’55.20''). The underlying soils were identified according to the French soil classification system (AFES, Baize and Girard 2009). Soil 1 is a saturated brunisol, a carbonate-rich soil with a clayey structure due to the parental material issued from Oxfordian marls (Guillaume and Guillaume 1963). Soil 2 is similar but was classified as a typic reductisol (i.e. gley soil) because groundwater level fluctuations and associated reduction/oxidation processes led to the redistribution of iron and the formation of visible Fe(III)-oxide patches. Soil 3, located in the minerotrophic fen, was identified as a sapric histosol (i.e. muck soil). This soil is seasonally water saturated up to the surface, leading to the accumulation of organic matter with a doughy texture. Soil 4, described as fibric histosol (i.e. peat soil) is permanently water-saturated, acidic and oligotrophic. The soil organic matter has a fibrous texture and consists essentially of dead Sphagnum biomass. Schematic drawings of the profiles of Soils 1–4 are presented in Fig. 1C. Table 2 summarizes the general bulk soils characteristics. Soils 2, 3 and 4 are water-saturated at least part of the year. Self-constructed multilevel-piezometers (Fig. S1, see online supplementary material) were installed in these locations to sample repeatedly the interstitial water from various depths and determine the physical/chemical characteristics during an annual cycle. Figure 1. Open in new tabDownload slide (A) Location of the study area in the French Jura mountains close to the border with Switzerland. Four stations representing a sequence of different soil types along a hill slope have been selected based on vegetation type (colored map modified from Gallandat 1982). Each station (St.) is indicated by a red cross. (B) Red arrows indicate the approximate positions of the four stations (St.). (C) Schematic representation of the studied catena, including the vegetation type, the soil type and the soil profile. Soils from stations 1 and 2 consist of an organo-mineral horizon ‘A’ (0–15 cm) with some living plant roots. Below, horizon ‘Sca’, is a carbonate-rich layer with a clayey structure, containing some carbonate stones, and consisting of weathered bed rock. The last horizon of Soil 1, ‘C’, consists of fragmented bed rock, whereas the last horizon of Soil 2, ‘Go’, is characterized by rusty stains, indicating fluctuating ground water level. Soils 3 and 4 are composed of a succession of holorganic horizons formed under water-saturated anoxic conditions. ‘Hf’ designates horizons with a fiber content >40%, ‘Hs’ stands for horizons with a fiber content between 10 and 40%, and ‘Hm’ for horizons with a fiber content <10%. Figure 1. Open in new tabDownload slide (A) Location of the study area in the French Jura mountains close to the border with Switzerland. Four stations representing a sequence of different soil types along a hill slope have been selected based on vegetation type (colored map modified from Gallandat 1982). Each station (St.) is indicated by a red cross. (B) Red arrows indicate the approximate positions of the four stations (St.). (C) Schematic representation of the studied catena, including the vegetation type, the soil type and the soil profile. Soils from stations 1 and 2 consist of an organo-mineral horizon ‘A’ (0–15 cm) with some living plant roots. Below, horizon ‘Sca’, is a carbonate-rich layer with a clayey structure, containing some carbonate stones, and consisting of weathered bed rock. The last horizon of Soil 1, ‘C’, consists of fragmented bed rock, whereas the last horizon of Soil 2, ‘Go’, is characterized by rusty stains, indicating fluctuating ground water level. Soils 3 and 4 are composed of a succession of holorganic horizons formed under water-saturated anoxic conditions. ‘Hf’ designates horizons with a fiber content >40%, ‘Hs’ stands for horizons with a fiber content between 10 and 40%, and ‘Hm’ for horizons with a fiber content <10%. Table 2. Summary of the general soil characteristics: average (±standard deviation) percentage of organic carbon content determined as weight loss on ignition, organic carbon:nitrogen ratio, pH (H2O); KCl-extractable NH4+, NO2− and NO3−, and HCl-extractable Fe(II) and Fe(III). ‘n. d.’ stands for ‘not determined’. Open in new tab Table 2. Summary of the general soil characteristics: average (±standard deviation) percentage of organic carbon content determined as weight loss on ignition, organic carbon:nitrogen ratio, pH (H2O); KCl-extractable NH4+, NO2− and NO3−, and HCl-extractable Fe(II) and Fe(III). ‘n. d.’ stands for ‘not determined’. Open in new tab Physico-chemical analysis of interstitial soil water In Soils 2, 3 and 4, in situ temperature, pH and pO2 of the free-water were measured down to 1 m below the surface every 10 cm by lowering a multi-meter (HQ40d, Hach Lange GmbH, Germany) in a piezometer tube, seven times during a complete annual cycle starting in November 2007. To determine concentrations of dissolved compounds (NH4+, NO2−, NO3−, Fe(II), colloidal Fe(III), S(-II) and SO42−), soil water was sampled down to 1 m below the surface every 6.25 or 12.5 cm using self-constructed multi-level piezometers five times between February and November 2008. More details about the design of the multi-level piezometers can be found in Fig. S1, see online supplementary material. Interstitial water was withdrawn from each piezometer level using a peristaltic pump and gas-tight tubings (PharMed®, Saint-Gobain Performance Plastics, USA). The outlet of the tube was connected directly to a 60 mL syringe, allowing sample collection without any contact with atmospheric oxygen. Then, water samples were directly filtered through 0.45 µm nylon membranes and treated as follows. For nitrite determination, a sample of 6.6 mL was fixed with 3.3 mL of sulfanilamide on site and analyzed within 24 h in the laboratory using the colorimetric method of Griess (1879), with a detection limit of 0.5 µmol L−1. Nitrate was quantified by a modified cadmium reduction protocol (Wood, Armstrong and Richards 1967): 100 µL of Na-borate buffer (pH 11) and one third of a NitraVer®6 Nitrate Reagent bag (Hach Lange GmbH, Germany) were added to 1 mL of water sample. The sample was agitated for 1 h and the produced nitrite was quantified using the Griess assay. Ammonium was determined by the hypochlorite–phenol reaction (Chaney and Marbach 1962), with a detection limit of 0.5 µmol L−1. Fe(II) and Fe(III) were measured with the ferrozine assay (Stookey 1970) on samples fixed with HCl (1 M final concentration). Dissolved S(-II), i.e. the sum of H2S, HS− and S2− species, was determined using the methylene-blue assay (Cline 1969) on Zn-acetate fixed water samples (0.1% w/v final concentration). Interstitial SO42− was quantified by ion chromatography (Dionex DX-120, USA). Pedological and geochemical characterization of the soils An Edelman auger (for Soils 1 and 2) and a Wardenaar peat profile sampler (for Soils 3 and 4; Eijkelkamp, The Netherlands) were used to collect soil material from each station along a depth profile (10 cm depth resolution) in September 2007. Soil cores sampled in triplicate from the same depth were manually homogenized in the field and sub-sampled for the different analyses. Three grams of soil were placed immediately in 30 mL of HCl 1 M for the later determination of Fe(II) and Fe(III). For determining inorganic N-forms, 15 g of soil sample were placed in 30 mL of 2 M KCl. Samples for DNA extraction were filled into two sterile 2 mL cryotubes and immediately frozen in liquid nitrogen. Soil material for all remaining analyses was stored in sealed plastic bags at 4°C. Soil water content was determined by weight loss after drying 10 g of fresh soil at 105°C for 2 days. The same samples were muffled at 450°C for 2 h (for samples of Soils 1 and 2) and at 600°C for 4 h (for samples of Soils 3 and 4). Soil organic matter content was measured by the weight loss on ignition (LOI; Ball 1964). For pH and organic matter analysis, samples were dried at 40°C and sieved at 2 mm. Soil pH (H2O) was determined with a pH-electrode (Mettler, Switzerland) in a suspension of 50 g of dried soil in 30 mL of deionized water that had been equilibrated on a rotary shaker for 1 h. Organic carbon to organic nitrogen ratios (Corg:Norg), and total organic carbon (TOC) were determined on a Carlo Erba CHN analyzer (Disnar et al. 2003). For quantification of adsorbed inorganic nitrogen compounds (NH4+, NO2−, NO3−) and Fe(II) and Fe(III), the soil samples were agitated for 1 h at 120 rpm in their sampling solution. The supernatant was then analyzed as described for interstitial water within 24 h. Adsorbed inorganic N-forms are not presented for samples from Soil 4 because of erratic measurements due to problems with the KCl extraction protocol applied to the peat samples. Biological oxygen demand after 5 days (BOD5) was used as a proxy for organic matter degradability in Soil 3 and was determined using manometric Warburg respirometers (Chase and Gray 1957). 15N incubation experiments for denitrification, anammox and Feammox Potential rates of anammox and denitrification were determined by 15N-label incubation experiments under anoxic condition (Thamdrup and Dalsgaard 2002; Naeher et al. 2015). The experimental set-up consisted of three treatments: (i) a control with addition of 15NH4+; (ii) addition of 15NH4+ and 14NO3−, where production of 29N2 is indicative for anammox; (iii) addition of 15NO3−, where production of 30N2 is caused by denitrification and 29N2 by anammox. As the formation of labeled N2 was observed in the controlled samples amended with 15NH4+ only, this control incubation was repeated with slurries from Soil 3 in order to determine if this was due to an oxygen contamination or to another microbial process, such as Feammox (Yang, Weber and Silver 2012). This time, absolute care was taken regarding oxygen contamination, and in vitro oxygen contamination was closely monitored using high-sensitivity oxygen sensors (PSt6-NAU from Presens, Germany) with a detection limit of 0.5 ppb. Soil slurries for five specific depths, as well as a soil-free extract with no amendment, amended with hydrous ferric oxides (HFO) or amended with MnO2, were used for this second incubation experiments. Methodological details concerning these incubations are presented in the supplementary material available online. Potential nitrification rates Potential aerobic nitrification rates were measured in a subset of samples from Soils 1–4 (Prosser and Nicol 2012). Ten grams of fresh soil were placed in an Erlenmeyer flask and amended with 50 mL of a solution containing 500 µM NH4Cl, 50 µM of KH2PO4 and 10 mM of NaClO3, to inhibit biological NO2− oxidation to NO3−. The flasks were continuously agitated (100 rpm) at room temperature in the dark for 24 h. The linear increase in NO2− concentrations in the supernatant was followed using the Griess method and used to calculate potential nitrification rates. Molecular analyses DNA was extracted using the FastDNA® SPIN Kit for Soil (MP Biomedicals, Santa Ana, USA) following the manufacturer's protocol. A nested-PCR approach was used to determine the diversity of anammox bacteria in the soil samples (Humbert et al. 2010a), using GoTaq DNA polymerase (Promega, Madison, Wisconsin, USA). Planctomycetes 16S rRNA genes were first amplified by PCR using Pla46f (Neef et al. 1998) and Univ1390r (Zheng et al. 1996) primers. DNA concentration in the reaction mix was between 0.04 and 2.5 ng µL−1 (0.45 ng µL−1 on average). The anammox 16S rRNA genes were then amplified in a second step using primers Amx368f and Amx890r (Schmid et al. 2005), by diluting PCR products of the previous PCR assay 100-fold in the new PCR mix. Weak or multiple bands of PCR products were considered as anammox negative. Purified PCR products of the correct size were ligated into pGEM-T vectors (Promega) and cloned in electrocompetent Escherichai coli XL1 cells. For sequencing, 27 clones were picked randomly from sample 3.4 (i.e. Soil 3, 22–28 cm depth), 50 clones from sample 3.7 (i.e. Soil 3, 41–48 cm depth), 24 clones from sample 3.10 (i.e. Soil 3, 60–75 cm depth) and 16 clones from sample 2.10 (i.e. Soil 2, 56.5–62 cm depth). Amplicons from each clone were digested by two restriction enzymes (AluI and MspI). Clones with similar restriction profiles were considered as identical and grouped in the same operational taxonomic unit (OTU). Eight OTUs from sample 3.4, 13 from sample 3.7, 3 from sample 3.10 and 4 from sample 2.10 were sequenced by Sanger sequencing (Eurofins Genomics, Germany). Sequences were uploaded to the NCBI database and can be retrieved with the accession numbers KT873861–KT873888. DNA extracts were analyzed by three qPCR assays in order to quantify anammox bacteria (based on 16S rRNA gene copy numbers), ammonia oxidizing archaea (AOA; based on archaeal amoA copy numbers) and ammonia oxidizing bacteria (AOB; based on bacterial amoA copy numbers). For all qPCR assays, each sample was run in triplicate and with three different DNA dilutions (4x, 8x and 16x) in order to limit inhibitory effects from soil contaminants co-extracted with DNA. Dilutions yielding higher copy numbers (per g of dry soil) were selected. Copy numbers of anammox 16S rRNA genes were determined using the SensiMixPlus SYBR® qPCR master mix (Quantace, Biolabo, Châtel-St-Denis, Switzerland) and the primers A438f/A684r (Humbert, Zopfi and Tarnawski 2012). Amplification efficiencies were between 71 and 73%, with R2 > 0.99. A detailed description of the qPCR amplification conditions for anammox bacteria is described elsewhere (Humbert, Zopfi and Tarnawski 2012). Genes encoding subunit A of the ammonium monooxygenase (amoA) of AOA and AOB (Prosser and Nicol 2012) were quantified using the Power SYBR® Green PCR Master Mix (Applied BiosystemsTM, Darmstadt, Germany), as well as the primer set amoA-1f/amoA-2r (Rotthauwe et al., 1997) and 19f/CrenamoA616r48x (Nicol et al. 2008), respectively. qPCR efficiencies ranged between 64 and 71% with R2 > 0.98 for the AOB qPCR assay, and between 72 and 78% with R2 > 0.99 for the AOA qPCR assay. Detailed amplification conditions for AOA/AOB are described in Hai et al. (2009). Bioinformatic analyses A reference alignment was constructed by aligning reference 16S rRNA genes from known anammox organisms (see Fig. 6 for Silva accessions numbers) and two outliers: Gemmata obscuriglobus (ABGO01000192) and Pseudomonas aeruginosa (AAQW01000001) with default parameters of MAFFT v.7 (http://mafft.cbrc.jp/alignment/server/; Katoh and Standley 2013). Shorter 16S rRNA anammox sequences retrieved in this study were added to the reference alignment using the ‘–addfragment’ option of MAFFT. The alignment was cleaned using the default parameters of BMGE v.1.12 (https://galaxy.pasteur.fr/; Criscuolo and Gribaldo 2010) and the alignment edges were manually trimmed off using AliView v.1.18 (Larsson 2014). Finally, a maximum-likelihood tree was constructed using the default parameters of IQ-TREE v.1.5.3 (http://iqtree.cibiv.univie.ac.at/; Minh, Nguyen and von Haeseler 2013; Nguyen et al. 2015). The tree was then annotated with iTOL v.3 (http://itol.embl.de/; Letunic and Bork 2016) and Inkscape v.0.91 (https://inkscape.org). To compare 16S rRNA genes from anammox-related microorganisms detected in different soil surveys, sequences were downloaded via NCBI Batch Entrez (https://www.ncbi.nlm.nih.gov/sites/batchentrez) using accession numbers. To obtain the same directionality, sequences were added to the anammox reference alignment using the ‘–addfragments’ and ‘–adjustdirection’ in MAFFT. UPARSE implemented in USEARCH v10.0.240 (Edgar 2013) was then used for sequence dereplicating, OTU picking (using a similarity threshold of 97% and a minimum size of 1) and for building an OTU table. OTUs were taxonomically annotated using UCLUST (Edgar 2010) implemented in QIIME v1.9.1 (Caporaso et al. 2010) and the SILVA v.128 as database (Pruesse et al. 2007). OTUs annotated as Brocadiaceae were placed on a reference tree (generated as described above but only containing reference sequences) by RAxML v.8.29 (Stamatakis 2014). The detailed bioinformatic analyses can be found at https://github.com/alex-bagnoud/anammox-soil-tree Statistical analyses In order to integrate all measurements (i.e. those from soil samples, the piezometer tubes and from multi-level piezometers) into a single data matrix, the depths of soil samples were defined as the standard depths. Where necessary, values and depths of other measurements (e.g. those measured on water samples) were adjusted by simple interpolation in order to match their depths with the standard depths. Statistical analyses were performed in R version 3.3.1 (R Core Team 2011). Pairwise correlations between variables such as: depth, anammox presence (as a binary variable), abundance of anammox bacteria, AOB and AOA, the number of days below the water table, organic carbon (measured as the loss on ignition), carbon and nitrogen content in soil samples, the extractable concentrations of NH4+ and NO3− from soil samples (only for Soils 1, 2 and 3), the concentrations of extractable Fe(II), Fe(III), and S(-II) from soil samples, and the average pH and concentrations of NH4+, NO2−, NO3−, O2, Fe(II), Fe(III), S(-II) and SO42− in the interstitial water (for Soils 2, 3 and 4), were computed using Pearson's correlation coefficient and the Hmisc package (https://CRAN.R-project.org/package=Hmisc). Presence of anammox bacteria and their abundance was modeled using generalized linear models (GLM), based on binomial and Poisson distribution, respectively, and significantly correlated variables as predictors, also including interactions. When considering interaction between variables as predictors, the Bayesian version of GLM was used to address a ‘quasi complete separation problem’, using the arm package (https://CRAN.R-project.org/package=arm). In order to account for putative dependency of samples, generalized linear mixed models (GLMM), were computed using the glmm package (https://CRAN.R-project.org/package=glmm). The detailed statistical analysis, including the data set and the script, can be found at https://github.com/alex-bagnoud/AnammoxBellefontaine RESULTS Interstitial water chemistry Free interstitial water could only be sampled in Soils 2, 3 and 4. Data from August and October are presented in Fig. 2 to illustrate the temporal variability of porewater chemistry in the different soils. August concentration profiles mirror the summer situation with dry weather and active vegetation. In October, vegetation dormancy commenced and fall precipitation led to increased water levels (Fig. 3A). Generally, we observed a decrease in the concentrations of soluble N-compounds from Soil 2 towards Soil 4. Ammonium concentrations in interstitial waters were highest in the deepest soil sections (i.e. below 60 cm) and decrease gradually towards the upper, oxic soil layers. Concentrations of nitrate, when present, were higher in the oxic soil layers and decreased with depth together with decreasing oxygen concentrations. Particularly in Soils 2 and 3, overlapping zones of ammonium and nitrate were observed, albeit mostly at times of vegetation dormancy (i.e. October, February, April). Highest nitrate concentrations were detected in February (not shown) and reached 30–55 µmol L−1 in Soil 2 and 4–8 µmol L−1 in Soil 3. No inorganic nitrogen was detected in the interstitial water of Soil 4, with the exception of ammonium below a depth of 62 cm. Nitrite concentrations in all soils were mostly below the detection limit of 0.5 µmol L−1. Figure 2. Open in new tabDownload slide Depth profiles of NH4+, NO3− and O2 concentrations (A and C), colloidal Fe(III), dissolved Fe(II), S(-II) and SO42− concentrations (B and D) in the interstitial water of Soils 2, 3 and 4, measured in fall (A and B) and in summer (C and D). No interstitial water could be extracted from Soil 1. NO2− concentrations are not shown as they were usually below the detection limit (0.5 μM). Figure 2. Open in new tabDownload slide Depth profiles of NH4+, NO3− and O2 concentrations (A and C), colloidal Fe(III), dissolved Fe(II), S(-II) and SO42− concentrations (B and D) in the interstitial water of Soils 2, 3 and 4, measured in fall (A and B) and in summer (C and D). No interstitial water could be extracted from Soil 1. NO2− concentrations are not shown as they were usually below the detection limit (0.5 μM). Figure 3. Open in new tabDownload slide (A) Seasonal groundwater level variations in the nitrophilic wet meadow (Soil 2), the minerotrophic fen (Soil 3) and the peat bog (Soil 4). Seasonal changes of the mean values of the physical and chemical parameters of the interstitial water in Soil 3: (B) depth-weighted averages of temperature, pH and dissolved O2; (C) depth-weighted average concentrations of ammonium, nitrite and nitrate. In B) and C), error bars represent standard deviations along the soil profile. Figure 3. Open in new tabDownload slide (A) Seasonal groundwater level variations in the nitrophilic wet meadow (Soil 2), the minerotrophic fen (Soil 3) and the peat bog (Soil 4). Seasonal changes of the mean values of the physical and chemical parameters of the interstitial water in Soil 3: (B) depth-weighted averages of temperature, pH and dissolved O2; (C) depth-weighted average concentrations of ammonium, nitrite and nitrate. In B) and C), error bars represent standard deviations along the soil profile. Dissolved Fe(II) and Fe(III) as well as SO42− (sulfate) and S(-II) (sulfide) provide additional information on the redox conditions in the soils (Fig. 2C and D). For all soils and both seasons (except Soil 3 during summer), Fe(II) concentration increased and/or SO42− concentration decreased with depth, mirroring the development of anoxic conditions in the deepest part of the soils. This is consistent with generally decreasing O2 concentrations with depth (Fig. 2A and C). The annual variation of the water level in Soils 2, 3 and 4 is presented in Fig. 3A. Even though the water table of Soil 3 was stable throughout the year (Fig. 3A), this soil showed a pronounced seasonal dynamic of the physical and chemical parameters (Fig. 3B and C). Water level, NO3− and O2 concentrations, and pH are higher during winter, while NH4+ concentrations and temperature are higher during summer. For Soil 4, no significant seasonal variations occurred in the interstitial water chemistry or the water level (Figs 2 and 3A). Abiotic soil properties and potential biotic activities Table 2 summarizes the pedological characterization of Soils 1–4. Average NH4+ concentrations in soil extracts increase slightly towards the minerotrophic fen from 0.1 to 0.5 µmol g−1 dry soil. Nitrate concentrations of soil extracts were low for all samples (between 0.01 and 0.03 µmol g−1 dry soil) and nitrite was always undetectable. Average total organic carbon contents and Corg:Norg ratios increase towards the peat bog. Biological oxygen demand (i.e. soil respiration) of Soil 3 was decreasing with depth, reaching values close to zero below 35 cm (Fig. 4). In all soils, except Soil 4, potential nitrification rates decreased with depth (results not shown), from 0.1 µmol NH4+ g−1 dry soil h−1 for Soils 1 and 3, and from 0.2 µmol NH4+ g−1 dry soil h−1 for Soil 2, and reached rates <0.01 µmol NH4+ g−1 dry soil h−1 below a depth of 60 cm. Potential nitrification rates remained low throughout the Soil 4 profile (0 to 0.02 µmol NH4+ g−1 dry soil h−1). Figure 4. Open in new tabDownload slide (A) Soil organic matter content in Soils 1–4 expressed as loss on ignition. (B) Depth profile of the biological oxygen demand (determined as BOD5) in the minerotrophic fen (Soil 3). Error bars represent the standard errors of the mean (SEM). Figure 4. Open in new tabDownload slide (A) Soil organic matter content in Soils 1–4 expressed as loss on ignition. (B) Depth profile of the biological oxygen demand (determined as BOD5) in the minerotrophic fen (Soil 3). Error bars represent the standard errors of the mean (SEM). Abundance of aerobic and anaerobic ammonia oxidizers The nested-PCR approach for anaerobic ammonia oxidizers (anammox) yielded positive amplification of a 477 bp long fragment for all samples from Soil 2 below 55 cm depth and for all samples of Soil 3 below 20 cm depth. Results of direct qPCR amplification of anaerobic ammonia oxidizers (Fig. 5) were consistent with the nested-PCR data. Anammox bacteria were absent from the upper part of Soils 2 and 3 and increase with depth to reach 7 x 105 and 2.7 x 107 copies g−1 dry soil, respectively. Figure 5. Open in new tabDownload slide (A) Number of amoA gene copies g−1 dry soil of ammonia oxidizing bacteria (AOB; black triangles) and archaea (AOA; grey triangles), and number of 16S rRNA gene copies g−1 dry of anammox bacteria (AMX; open triangles). Error bars indicate standard deviations of triplicate runs. The roman numbers in parenthesis indicate the samples for which the anammox community composition has been determined by clone sequencing. (B) Pie charts showing the phylogenetic repartition of the clones retrieved from one sample of Soil 2 (59 cm, 16 clones) and three samples of Soil 3 (25 cm, 20 clones; 44.5 cm, 50 clones; 67.5 cm, 21 clones). The four anammox clusters and their colors correspond to the ones shown in Fig. 6. Light grey represents non-anammox sequences. Figure 5. Open in new tabDownload slide (A) Number of amoA gene copies g−1 dry soil of ammonia oxidizing bacteria (AOB; black triangles) and archaea (AOA; grey triangles), and number of 16S rRNA gene copies g−1 dry of anammox bacteria (AMX; open triangles). Error bars indicate standard deviations of triplicate runs. The roman numbers in parenthesis indicate the samples for which the anammox community composition has been determined by clone sequencing. (B) Pie charts showing the phylogenetic repartition of the clones retrieved from one sample of Soil 2 (59 cm, 16 clones) and three samples of Soil 3 (25 cm, 20 clones; 44.5 cm, 50 clones; 67.5 cm, 21 clones). The four anammox clusters and their colors correspond to the ones shown in Fig. 6. Light grey represents non-anammox sequences. In contrast to anammox bacteria, both AOA and AOB were detected in all four soils and all depths, with amoA gene abundances ranging between 4.3 x 106 and 3.1 x 109 copies g−1 dry soil (Fig. 5). Decreasing AOA and AOB abundances with soil depth were only observed in Soil 2. In all other soil profiles, the copy numbers remained relatively constant. In Soil 3, an increase in the AOA:AOB ratio with depth was observed; archaeal amoA was up to 70 times more abundant than bacterial amoA. There was a good correlation between the log of the number of AOB copies g−1 dry soil and the log of potential nitrification rates (linear regression, R2 = 0.86) but not between the log of the number of AOA copies g−1 dry soil and the log of potential nitrification rates (linear regression, R2 = 0.12; Fig. S2, see online supplementary material). Diversity of anammox bacteria All anammox-like sequences from three samples from Soil 3 (22–28 cm, 41–48 cm, 60–75 cm) and one sample from Soil 2 (56–62 cm) fall into the Brocadiaceae and form four distinct clusters with high ultrafast bootstrap support of ≥88% (Figs 5 and 6). A fifth group of retrieved sequences was not related to anammox bacteria and hence was not examined any further. Ten clones out of a total of 107 (9.3%) represented non-anammox sequences, indicating that this primer pairs can yield false positive amplification in samples with low anammox abundance as reported earlier (Sonthiphand and Neufeld 2013). The relative abundance of these anammox clusters for each of the sequenced samples is summarized in Fig. 5B. With the phylogenetic analysis presented in Fig. 6, Cluster 2 could be assigned to Ca. Brocadia and Cluster 4 to Ca. Anammoxoglobus, while Clusters 1 and 3 could not be assigned to any known anammox candidate genus. However, when using UCLUST annotation based on the SILVA 128 database, Cluster 4 was not assigned to any known anammox genus (Cluster 4 falls within OTU67 in Fig. S5, see online supplementary material). Figure 6. Open in new tabDownload slide Maximum-likelihood tree showing the phylogenetic relationships between known anammox bacteria and closely related 16S rRNA gene sequences retrieved from the nitrophilic wet meadow (Soil 2) and the minerotrophic fen (Soil 3). Clone names are composed as follows: soil number (2 or 3), sampling depth (a = 22–28 cm, b = 41–48 cm, c = 56–62 cm and d = 60–75 cm), clone number with the number of identical clones retrieved in parenthesis. Clones 2c are labeled with grey triangles, 3a with black squares, 3b with black circles and 3d with black diamonds. The colored boxes highlight the four anammox clusters shown in Fig. 5. Ultrafast bootstrap values >80% are indicated for each node. Sequences from Gemmata obscuriglobus (ABGO01000192) and Pseudomonas aeruginosa (AAQW01000001) were used as outgroups for rooting the tree. Figure 6. Open in new tabDownload slide Maximum-likelihood tree showing the phylogenetic relationships between known anammox bacteria and closely related 16S rRNA gene sequences retrieved from the nitrophilic wet meadow (Soil 2) and the minerotrophic fen (Soil 3). Clone names are composed as follows: soil number (2 or 3), sampling depth (a = 22–28 cm, b = 41–48 cm, c = 56–62 cm and d = 60–75 cm), clone number with the number of identical clones retrieved in parenthesis. Clones 2c are labeled with grey triangles, 3a with black squares, 3b with black circles and 3d with black diamonds. The colored boxes highlight the four anammox clusters shown in Fig. 5. Ultrafast bootstrap values >80% are indicated for each node. Sequences from Gemmata obscuriglobus (ABGO01000192) and Pseudomonas aeruginosa (AAQW01000001) were used as outgroups for rooting the tree. BLASTN analyses revealed that the anammox bacteria from this study are highly similar (≥99% sequence identity) to representatives from other soil environments (Hu et al. 2011, 2013; Zhu et al. 2011; Humbert, Zopfi and Tarnawski 2012; Sato et al. 2012; Yang et al. 2015; Shen et al. 2016). Sequences from Cluster 1 are highly similar to anammox sequences detected in paddy soils (e.g. AB602695.1, GU083873.1, JN176720.1 and KJ508609.1), in a peat soil (e.g. HQ637487.1), in a wetland soil (e.g. KT162137.1) and in a lake-shore soil (e.g. FM174261.1). Members of Clusters 2 and 4 share high similarity with paddy soil anammox sequences (e.g. AB602629.1 for Cluster 2, and AB602743.1, GU083864.1 and KJ508632.1 for Cluster 4). There are no highly similar hits for sequences from Cluster 3. USEARCH was also used for defining OTUs at 97% identities in order to compare sequences from various terrestrial surveys, including those from the present study. This clustering analysis allowed 113 OTUs to be defined, but only 26 were annotated as Brocadiaceae. Among them, all sequences from Cluster 1 (except the one from clone 3b-1 that was considered as a noisy chimeric by the pipeline) were assigned to OTU4, which was also detected in 8 other studies conducted in wetland soils (Humbert, Zopfi and Tarnawski 2012; Shen et al. 2016), paddy soils (Hu et al. 2011, 2013; Zhu et al. 2011; Sato et al. 2012; Yang et al. 2015) and lake shore soils (Humbert et al. 2010a; Humbert, Zopfi and Tarnawski 2012). All sequences from Cluster 2 were assigned to OTU1, also detected in 10 other studies (Humbert et al. 2010a; Humbert, Zopfi and Tarnawski 2012; Sato et al. 2012; Han, Li and Gu 2013; Hu et al. 2013; Shen et al. 2013; Shen, Xu and He 2014; Naeher et al. 2015; Yang et al. 2015; Shen et al. 2016). The two sequences from Cluster 3 were assigned to OTU67, only detected in 1 other study in a paddy soil (Sato et al. 2012). Finally, all sequences from Cluster 4 were assigned to OTU5, detected in 6 other studies from wetland soils (Shen et al. 2016), paddy soils (Zhu et al. 2011; Sato et al. 2012; Hu et al. 2013; Yang et al. 2015) and agricultural soils (Shen et al. 2015). Brocadiaceae OTUs were placed on the reference tree using an evolutionary placement algorithm (Fig. S5, see online supplementary material). This Figure also summarizes the repartition of OTUs among the studies and among the different soil types. Anaerobic ammonium oxidation in selected soil layers Anoxic incubations of soil slurries with 15N-labeled NH4+ and/or NO3− were carried out to confirm anammox and denitrification activities in Soil 2 (50–60 cm) and Soil 3 (40–50 cm) where molecular data indicated the presence of anammox cells. Incubations amended only with 15NH4+ produced 29N2 and 30N2, indicating that NH4+ was oxidized during the incubations (Fig. S3, see online supplemenatry material). This made the assessments of denitrification and anammox rates more complicated. Therefore, it was decided to use data from the 15NO3− incubation only, assuming the intrinsic presence of 14NH4+, and the possibility of its oxidation into 14NO3−. In these incubations, rates of 29N2 and 30N2 production over 12 h are steady (Fig. S3, see online supplemenatry material). Denitrification was the most active process in both soils with NO3− reduction rates of 0.99 µmol g−1 dry soil d−1 and 6.63 µmol g−1 dry soil d−1 in Soils 2 and 3, respectively. Anammox activity was low, accounting for about 2.1% in Soil 2 and 1.4% in Soil 3 of the N2 produced, corresponding to anammox potential rates of 0.02 µmol g−1 dry soil d−1 in Soil 2, and 0.09 µmol g−1 dry soil d−1 in Soil 3. Additional incubations were performed with fresh soil material for determining whether alternative anaerobic ammonium oxidation pathways with Fe(III) or Mn(IV) as oxidant were taking place, as suggested by the production of labeled N2 in the control incubations amended only with 15NH4+ (Fig. S3,A, see online supplementary material). During the entire incubation time that lasted 3 months, the concentration of dissolved oxygen in the incubation bottles stayed below the detection limit of the high sensitivity O2-sensor spot (<0.5 ppb). After subsampling, 1 ppb O2 was detected transiently in a handful of bottles. After 86 days of incubation, 15N2 reached at maximum 0.003 µmol g−1 soil for the soil slurry incubations, 0.0004 µmol L−1 for the soil-less incubations with no amendment, 0.0005 µmol L−1 for the soil-less incubations amended with HFO, and 0.0007 µmol L−1 for the soil-less incubations amended with MnO2 (Fig. S4). This insignificant NH4+ oxidation was likely due to very small oxygen contamination through the bottle septa and did not match what was observed in the control experiments of the first set of 15N-incubations, where labeled N2 reached more than 10 µmol L−1 in <10 h when only 15NH4+ was amended (Fig. S3, see online supplemenatry material). There was no significant stimulation of anaerobic NH4+ oxidation in the presence of Mn(IV)- or Fe(III)-oxides. Statistical model for explaining the presence of anammox bacteria The presence of anammox bacteria was significantly correlated with the number of days a soil sample was below the water table per year (r2 = 0.50, P-value = 0.0009), but also with the organic N content of soil samples (r2 = 0.45, P-value = 0.0035), with the pH of soil samples (r2 = 0.50, P-value = 0.0008), the concentration of NH4+ extracted from soil samples (r2 = 0.51, P-value = 0.0042), with the average NH4+ concentration in the interstitial water (r2 = 0.46, P-value = 0.0087) and with the average Fe(II) concentration in the interstitial water (r2 = 0.41, P-value = 0.0212). Fig. 7 summarizes the significant correlations between environmental parameters and the presence/absence of anammox, as well as the abundances of anammox, AOA and AOB. Based on these results, GLMs were fitted to predict the presence/absence of anammox bacteria (using binomial distribution). Three models could accurately describe the presence of anammox bacteria in the soil system. The models use as predictor variables (i) the number of days that soil samples were submerged per year (P-value = 0.0044), (ii) the pH of soil samples (P-value = 0.0273) and (iii) the average NH4+ concentration in interstitial water (P-value = 0.0131). The two last-mentioned models were considered redundant because their respective predicting variables, the pH of soil samples and the average NH4+ concentration, are correlated (r2 = 0.59, P-value = 0.0004). Also, the pH of soil samples correlates with the presence of anammox bacteria because the low-pH values were all measured in Soil 4, where NH4+ and NO3− concentrations are negligible and thus anammox is absent. All three predictor variables interact significantly with a fourth variable, the average NO3− concentration in the interstitial water (P-values of interactions = 0.0079, 0.04 and 0.0166, respectively). Akaike information criterion (AIC) scores of GLMs are 13.95, 10.65 and 22.73, respectively. The same set of variables and variable interactions can also significantly model the abundance of anammox bacteria (in a log scale), based on a Poisson distribution. Figure 7. Open in new tabDownload slide Significant correlations between selected environmental parameters and anammox abundance (16S rRNA gene copies g−1 dry soil) and presence (binary variable), and between AOB and AOA abundance (amoA gene copies g−1 dry soil). Three groups of samples are considered here: Soils 1–4 (n = 41), soils 2–4 (n = 32) and soils 1–3 (n = 30). Environmental parameters are the following: depth (cm), number of days a soil sample spent under the water table (over a year), AOA and AOB abundance (amoA copies g−1 dry soil), loss of ignition, content in organic carbon and nitrogen, soil pH, soil content in NH4+, Fe(II), Fe(III) (µg g−1 dry soil), and average concentration of NH4+, Fe(II), Fe(III) and S(-II) in the interstitial water (µM). The size of the dots and their color represent the Pearson's correlation coefficient (see scale for more details). Crosses represent non-significant correlations (P-value > 0.05) and empty boxes represent incomplete datasets. Figure 7. Open in new tabDownload slide Significant correlations between selected environmental parameters and anammox abundance (16S rRNA gene copies g−1 dry soil) and presence (binary variable), and between AOB and AOA abundance (amoA gene copies g−1 dry soil). Three groups of samples are considered here: Soils 1–4 (n = 41), soils 2–4 (n = 32) and soils 1–3 (n = 30). Environmental parameters are the following: depth (cm), number of days a soil sample spent under the water table (over a year), AOA and AOB abundance (amoA copies g−1 dry soil), loss of ignition, content in organic carbon and nitrogen, soil pH, soil content in NH4+, Fe(II), Fe(III) (µg g−1 dry soil), and average concentration of NH4+, Fe(II), Fe(III) and S(-II) in the interstitial water (µM). The size of the dots and their color represent the Pearson's correlation coefficient (see scale for more details). Crosses represent non-significant correlations (P-value > 0.05) and empty boxes represent incomplete datasets. Additionally, GLMMs were tested in order to account for the fact that the soil samples were not independent, as they originate from four soil depth profiles. However, the random effects of these models (that express how much variability there is between the different soils) were not significant and hence the GLM was sufficient for modeling the presence and abundance of anammox in these soils. The detailed results of the statistical analyses can be found at https://github.com/alex-bagnoud/AnammoxBellefontaine DISCUSSION Chemical conditions in the soils and sources of inorganic nitrogen compounds The physical and chemical conditions vary strongly in the different soils during the annual cycle, particularly in Soils 2 and 3. During the cold season, water levels are higher and the concentrations of oxygen and NO3− are elevated (Fig. 2A). This period of low vegetation activity allows the source of inorganic nitrogen compounds in the different soils to be identified. For example, in fall, interstitial water concentrations of inorganic nitrogen compounds were highest in the nitrophilic wet meadow (Soil 2) at the foot of a gentle hill slope adjacent to an amended pasture (Figs 1 and 2A.). Nitrogen brought out by manuring likely reaches this soil by lixiviation (Barakat, Cheviron and Angulo-Jaramillo 2016). The nitrophilic wet meadow (Soil 2) serves thus as a buffer zone between the amended pasture (Soil 1), the fen (Soil 3) and the peat bog (Soil 4). There, concentrations of dissolved nitrogen compounds never exceeded 1 µmol L−1 (except for ammonium in the deeper layer; Fig. 2), because of the ombrotrophic nature of the peat bog and the presence of Sphagnum sp. moss, that efficiently scavenge atmospherically deposited nutrients (Fritz et al. 2014). In spring, the induction of biological activity happens within a remarkably short time. Evapotranspiration lowered the groundwater table in Soils 2 and 3 (Fig. 3A), as was also observed in another bog system (Lafleur et al. 2005). Furthermore, respiration of plant roots and microorganisms, stimulated by rising temperature and rhizodeposition, drastically reduced pO2 in the waterlogged soil zones (Fig. 3B; Revsbech et al. 1999; Kuzyakov and Cheng 2001). Nitrate is assimilated by the growing plants (Miller and Cramer 2005) or reduced by microorganisms, leading to the production of nitrogen gas or ammonium (Tiedje 1988; Fig. 3C). Ammonium is also liberated continuously by mineralization of organic matter (Landi et al. 2006). The Corg:Norg ratios in Soils 1, 2 and 3 indicate that organic matter still contains significant amounts of nitrogen, as opposed to Soil 4, which is particularly poor in nitrogen (Table 2). Soil biological activity and water level are key parameters for the distribution of nitrate and ammonium along the soil profiles. Waterlogged conditions reduce the transport of oxygen into deeper soil sections, facilitating anoxic conditions and anaerobic metabolism. Hence, Soils 2 and 3 are characterized by periods of oxic and anoxic conditions that alternate predictably on a seasonal scale, but may also change on a short-term scale according to meteorological conditions and groundwater table fluctuation. Environmental factors controlling the distribution of anaerobic ammonia oxidizers Abundances of anammox bacteria in Soils 2 and 3 are in the same range as found in other soil surveys (Humbert, Zopfi and Tarnawski 2012; Hu et al. 2013; Long et al. 2013; Naeher et al. 2015; Shen et al. 2015, 2016; Yang et al. 2015; Meng et al. 2016; Shan et al. 2016; Hui et al. 2017; Zhou et al. 2017). We observe, however, a distinct spatial distribution, where anammox bacteria are only present below a certain depth, i.e. 50 cm below the surface in Soil 2 and 20 cm below the surface in Soil 3, respectively. Statistical modeling clearly shows that in this catena anammox bacteria are found in soil sections that remain waterlogged for an extended period of time and contain NO3− and NH4+ in the interstitial water. Nitrite, however, which is one of the two primary substrates of anammox bacteria, does not explain anammox distribution. Nitrite concentrations are close to or below the detection limit in all soils and at all times. Hence, anammox bacteria in this catena depend on the continuous production of nitrite by nitrate-reducing and/or ammonia-oxidizing microorganisms. Which one of the two processes provides nitrite to anammox bacteria is likely changing during the year along with the water level and degree of oxygenation. The fact that anammox bacteria are significantly correlated with AOA (Fig. 7) suggests that anammox utilize the NO2− produced from AOA. Besides availability of inorganic nitrogen, extended water-logged conditions seem to be key for the establishment of anammox bacteria. Indeed, anammox was not detected in the part of the soils that are above the groundwater table for longer periods. This is the case for the shallow, well-drained Soil 1 as well as for the upper parts of Soils 2 and 3. In these soils, there is a good correspondence between the groundwater level measured during the summer (Fig. 3A) and the depths where anammox can be detected (Fig. 5). This observation is consistent with Humbert et al. (2012) who detected anammox bacteria in the predominantly water-saturated zones of a wetland. Against initial expectations, soil organic matter content did not have any statistical relation with the distribution of anammox bacteria. High contents of organic matter may favor heterotrophic nitrate-reducing bacteria but do not hinder the presence of anammox bacteria. Anammox bacteria were detected in Soil 3, which consists of more than 60% organic matter (Fig. 4A). The quality, i.e. degradability, of organic matter, rather than the absolute quantity, is likely the determining factor for the distribution and activity of anammox and denitrifying bacteria. For example, in Soil 3, the abundance of anammox bacteria (Fig. 5) is anti-correlated to soil respiration (Fig. 4B). The fact that the organic matter of Soil 3 below 25 cm is poorly degradable by microorganisms may explain the presence of anammox bacteria in this part of the profile, because they do not rely on an organic source of carbon. Conversely, the N content of organic matter shows a positive correlation with anammox and particularly AOA, suggesting that the N (as ammonium) liberated during organic matter mineralization serves as an important substrate for AOA and anammox, particularly in Soil 3 where free inorganic N pools and external N inputs are low. Distribution and abundance of aerobic ammonia oxidizers Copy numbers of AOA and AOB amoA were generally 1–2 orders of magnitude higher than maximum values of anammox bacterial 16S rRNA gene copy numbers. Assuming a single rrn operon per anammox bacterial genome, it means that aerobic ammonia oxidizing microorganisms are clearly more abundant than the anaerobic ones, and thus remain the key players in microbial ammonia oxidation. Only in Soil 3 were AOB distinctly outnumbered by their archaeal counterpart. The observed differences were less pronounced than reported by Leininger et al. (2006) who observed up to 3000 times more AOA than AOB. The abundance of AOA in this soil, in the range of 109 copies g−1 dry soil is higher than what is typically observed in soil (Leininger et al. 2006; Erguder et al. 2009). Because Soil 3 is holorganic, this observation could signify that AOA can exhibit a mixotrophic mode of growth, and preferably oxidize NH4+ derived from organic matter mineralization (Hatzenpichler 2012). Indeed, a positive correlation was observed between soil organic N content and AOA amoA copy numbers (Fig. 7). Also, the high density of AOA in Soil 3 is consistent with their ability to grow in environments with low O2 content (Hatzenpichler 2012), because this soil is less oxygenated than Soil 2 (Fig. 2B). In Soil 1, AOA abundances remain stable with depth, while AOB decrease, which corresponds to findings in other agricultural soils (Erguder et al. 2009). However, it is unclear as to why AOA abundances are decreasing with depth in Soil 2, because this latter is similar to Soil 1 in terms of structure, composition and nitrogen content. In Soil 4, AOA and AOB are less numerous than in the other soils of the catena, which is consistent with the very low level of NH4+ in the interstitial water and the low N content of the organic matter (Fig. 2; Table 2). We found a good correlation between potential nitrification rates and the bacterial amoA abundances but not the archaeal amoA abundances (Fig. S2, see online supplementary material). This lack of correlation has been observed before (Bernhard et al. 2010) and may indicate that the classical potential nitrification assay is unsuitable for AOA, which are adapted to low O2 and low NH4+ concentrations. AOA preferably oxidize ammonia at low concentrations derived from organic matter mineralization (Hatzenpichler 2012; Prosser and Nicol 2012). Diversity of anammox bacteria The observation of four distinct anammox clusters in Soil 3 is consistent with earlier findings of an increased phylogenetic diversity of anammox bacteria in terrestrial environments (Table 1) as opposed to aquatic ecosystems (Schmid et al. 2007). Similar observations were made in river estuary sediments, where the highest diversity was found in the least saline part of the estuary (Dale, Tobias and Song 2009). A reduced ecological pressure and the heterogeneity of the soil (and sediments), as opposed to the more homogenous water column habitats, may explain this higher diversity. The change in composition of the anammox community along the profile of Soil 3 suggests niche segregation between the different anammox organisms. In Soil 2, however, the anammox diversity is minimal, with only the presence of Cluster 4 representatives. The anammox sequences detected in this study are also present in other terrestrial environments (with similarities >99%), suggesting that they may have a widespread distribution and are not tied to a specific location or soil type (Fig. S5, see online supplementary material). Activity of anammox Anammox and denitrification activities were assessed in anoxic soil incubation experiments, where the 15N-labeled substrates were added to levels above natural concentrations. Such incubation experiments do not provide actual in situ rates, but provide proof of activity and serve as proxies for the potential metabolic activity of denitrifying and anammox bacteria, respectively, under substrate replete conditions. The actual in situ rates will be lower due to substrate limitation, and variable in time because of e.g. changing redox conditions. The incubation bottles had been flushed with O2-free gas and pre-incubated for 3 days to ensure anoxic conditions and to deplete the intrinsic NO3− pool. The production of 30N2 in the 15NH4+ incubation (control) can be explained by oxygen contamination during tracer addition, incubation and/or subsampling. When repeating these control incubations, no consistent oxygen contamination or significant labeled N2 could be detected. Still, this illustrates that O2 contamination can occur during manipulations and can significantly influence the outcome of such experiments. Careful experimental procedures (e.g. subsampling under an N2 atmosphere in an anaerobic chamber) and oxygen monitoring is indispensable to highlight anoxic ammonium oxidation pathways. Despite this, the 15NO3− incubations can be used to quantify denitrification and to estimate the maximum contribution of anammox to N2 formation. Production of 30N2 is due to denitrification, whereas 29N2 can be formed by denitrifying bacteria from 15NO3− and 14NO3− (from re-oxidized 14NH4+) or by anammox bacteria from 15NO3− and 14NH4+. Our data show that denitrification is dominating anammox in both soil samples and that anammox accounts for <2.1% in Soil 2 and <1.4% in Soil 3 of total N2 produced. This strong dominance of denitrification may be due to the higher organic matter content in soils than in aquatic environments. Similar observations have been made in marine sediments where anammox was hardly detectable in organic-rich bay and shelf sediments, but became progressively more important with distance from the shore, in concert with decreasing quantity and reactivity of organic matter (Dalsgaard, Thamdrup and Canfield 2005). The potential rates of anammox and its contribution to N2 production reported here fall within the lower range of what was observed in soils (Table 1). Higher anammox activity in certain soils can be explained by greater availability of NH4+ and NO3−. For instance, in fertilized paddy soils, where anammox contributes up to 37% of total N2 formation (Zhu et al. 2011), the concentration of inorganic N is one or two orders of magnitude higher than in the nearly pristine wetland soils investigated here. CONCLUSION In this soil system, anammox bacteria can be found in deeper soil compartments where the conditions are favorable for their growth. They depend on long-term water-saturated conditions, and nitrate concentrations at least in the μmol L−1 range, which is consistent with the GLM analysis. They can be present even if organic carbon contents are elevated. Anammox bacteria are less abundant and active than aerobic ammonia oxidizing microorganisms or denitrifying bacteria. Their activity in this system falls into the lower range of what has been determined for soil environments, which is probably related to the low concentrations of ammonium in this natural wetland system. Nevertheless, the studied soil transect revealed a remarkably high diversity of anammox bacteria, which greatly contrasts with results from marine and lacustrine water columns where typically a single candidate genus dominates. Four distinct anammox clusters could be identified, which are also present in various other soil systems, as shown by pairwise comparison of 16S rRNA sequences. Along the depth profile of Soil 3, the structure of the anammox community changes, which suggests niche segregation between the different species. Even though anammox bacteria are not the main actors in the biological N cycle in the investigated soil system, their distinct distribution along environmental gradients provides important insights about the key environmental factors that shape anammox communities in terrestrial systems. Authors' Contributions J.Z and A.B. designed the study, A.B., S.G.-H. and J.Z. performed the field work, B.S.-H. and M.S. supported the amoA qPCR analysis, A.B. and S.G.-H. performed laboratory analyses, all authors analyzed the data, A.B. and J.Z. wrote the manuscript, all authors discussed and revised it. ACKNOWLEDGEMENTS Franz Conen, Barbara Seth and Thomas Kuhn (Department of Environmental Sciences, University of Basel) are thanked for their help with the isotopic measurements of N2. Daniel Hunkeler and Pascale Ducommun (CHYN, University of Neuchâtel, Switzerland) are acknowledged for their help with constructing the multilevel piezometers. Alix Zollinger (Swiss Institute of Bioinformatics, Lausanne) is thanked for advice on the statistical analyses. All colleagues, friends and family members who participated in the field campaigns are warmly acknowledged for their support and good spirits. FUNDING This work was supported by the Swiss State Secretariat for Science and Education through COST Action 856 [Project SER No C04.0269] and by the Swiss National Science Foundation through an Early Postdoc Mobility Grant [P2ELP2_172 313]. Conflict of Interest. None declared. REFERENCES AFES , Baize D , Girard M-C . Référentiel pédologique 2008 . Versailles, France : Editions Quae , 2009 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Attrill MJ , Rundle SD . Ecotone or Ecocline: Ecological boundaries in estuaries . Estuar Coast Shelf Sci . 2002 ; 55 : 929 – 36 . Google Scholar Crossref Search ADS WorldCat Ball DF. Loss-on-Ignition as an estimate of organic matter and organic carbon in Non-Calcareous soils . J Soil Sci . 1964 ; 15 : 84 – 92 . Google Scholar Crossref Search ADS WorldCat Barakat M , Cheviron B , Angulo-Jaramillo R . Influence of the irrigation technique and strategies on the nitrogen cycle and budget: A review . Agric Water Manag . 2016 ; 178 : 225 – 38 . Google Scholar Crossref Search ADS WorldCat Bernhard AE , Landry ZC , Blevins A et al. . Abundance of Ammonia-Oxidizing archaea and bacteria along an estuarine salinity gradient in relation to potential nitrification rates . Appl Environ Microbiol . 2010 ; 76 : 1285 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Butterbach-Bahl K , Baggs EM , Dannenmann M et al. . Nitrous oxide emissions from soils: how well do we understand the processes and their controls? . Philos Trans R Soc B Biol Sci . 2013 ; 368 : 20130122 . Google Scholar Crossref Search ADS WorldCat Caporaso JG , Kuczynski J , Stombaugh J et al. . QIIME allows analysis of high-throughput community sequencing data . Nat Methods . 2010 ; 7 : 335 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat Chaney AL , Marbach EP . Modified reagents for determination of urea and ammonia . Clin Chem . 1962 ; 8 : 130 – 2 . Google Scholar Crossref Search ADS PubMed WorldCat Chase FE , Gray PHH . Application of the warburg respirometer in studying respiratory activity in soil . Can J Microbiol . 1957 ; 3 : 335 – 49 . Google Scholar Crossref Search ADS WorldCat Chen J , Zhou HC , Pan Y et al. . Effects of polybrominated diphenyl ethers and plant species on nitrification, denitrification and anammox in mangrove soils . Sci Total Environ . 2016 ; 553 : 60 – 70 . Google Scholar Crossref Search ADS PubMed WorldCat Cline JD. Spectrophotometric determination of hydrogen sulfide in natural waters . Limnol Oceanogr . 1969 ; 14 : 454 – 8 . Google Scholar Crossref Search ADS WorldCat Criscuolo A , Gribaldo S . BMGE (Block Mapping and Gathering with Entropy): a new software for selection of phylogenetic informative regions from multiple sequence alignments . BMC Evol Biol . 2010 ; 10 : 210 . Google Scholar Crossref Search ADS PubMed WorldCat Daims H , Lebedeva EV , Pjevac P et al. . Complete nitrification by Nitrospira bacteria . Nature . 2015 ; 528 : 504 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Dale OR , Tobias CR , Song B . Biogeographical distribution of diverse anaerobic ammonium oxidizing (anammox) bacteria in Cape Fear River Estuary . Environ Microbiol . 2009 ; 11 : 1194 – 207 . Google Scholar Crossref Search ADS PubMed WorldCat Dalsgaard T , Thamdrup B , Canfield DE . Anaerobic ammonium oxidation (anammox) in the marine environment . Res Microbiol . 2005 ; 156 : 457 – 64 . Google Scholar Crossref Search ADS PubMed WorldCat Disnar JR , Guillet B , Keravis D et al. . Soil organic matter (SOM) characterization by Rock-Eval pyrolysis: scope and limitations . Org Geochem . 2003 ; 34 : 327 – 43 . Google Scholar Crossref Search ADS WorldCat Dong X , Reddy G . Soil bacterial communities in constructed wetlands treated with swine wastewater using PCR-DGGE technique . Bioresour Technol . 2010 ; 101 : 1175 – 82 . Google Scholar Crossref Search ADS PubMed WorldCat Edgar RC . Search and clustering orders of magnitude faster than BLAST . Bioinformatics . 2010 ; 26 : 2460 – 1 . Google Scholar Crossref Search ADS PubMed WorldCat Edgar RC . UPARSE: highly accurate OTU sequences from microbial amplicon reads . Nat Methods . 2013 ; 10 : 996 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Erguder TH , Boon N , Wittebolle L et al. . Environmental factors shaping the ecological niches of ammonia-oxidizing archaea . FEMS Microbiol Rev . 2009 ; 33 : 855 – 69 . Google Scholar Crossref Search ADS PubMed WorldCat Finlay BJ , Span ASW , Harman JMP . Nitrate respiration in primitive eukaryotes . Nature . 1983 ; 303 : 333 – 6 . Google Scholar Crossref Search ADS WorldCat Fritz C , Lamers LPM , Riaz M et al. . Sphagnum Mosses - Masters of efficient N-Uptake while avoiding intoxication . PLoS ONE . 2014 ; 9 , DOI: 10.1371/journal.pone.0079991 . OpenURL Placeholder Text WorldCat Crossref Gallandat JD , Prairies marécageuses du Haut-Jura . Beiträge zur geobotanischen Landesaufnahme des Schweiz . 1982 ; 85 : 1 – 327 . OpenURL Placeholder Text WorldCat Griess P . Bemerkungen zu der Abhandlung der HH. Weselsky und Benedikt „Ueber einige Azoverbindungen” . Berichte Dtsch Chem Ges . 1879 ; 12 : 426 – 8 . Google Scholar Crossref Search ADS WorldCat Guillaume S , Guillaume M . Morez - Bois d'Amont, feuille XXXIII-XXXIV-27 . 1963 . Hai B , Diallo NH , Sall S et al. . Quantification of key genes steering the microbial nitrogen cycle in the rhizosphere of sorghum cultivars in tropical agroecosystems . Appl Environ Microbiol . 2009 ; 75 : 4993 – 5000 . Google Scholar Crossref Search ADS PubMed WorldCat Han P , Li M and Gu JD Biases in community structures of ammonia/ammonium-oxidizing microorganisms caused by insufficient DNA extractions from Baijiang soil revealed by comparative analysis of coastal wetland sediment and rice paddy soil . Appl. Microbiol. Biotechnol. 2013 ; 97 : 8741 – 56 . Google Scholar Crossref Search ADS PubMed WorldCat Hatzenpichler R . Diversity, physiology, and niche differentiation of Ammonia-Oxidizing archaea . Appl Environ Microbiol . 2012 ; 78 : 7501 – 10 . Google Scholar Crossref Search ADS PubMed WorldCat Hofstra N , Bouwman AF . Denitrification in agricultural soils: Summarizing published data and estimating global annual rates . Nutr Cycl Agroecosystems . 2005 ; 72 : 267 – 78 . Google Scholar Crossref Search ADS WorldCat Hu B , Rush D , van der Biezen E et al. . New anaerobic, Ammonium-Oxidizing community enriched from peat soil . Appl Environ Microbiol . 2011 ; 77 : 966 – 71 . Google Scholar Crossref Search ADS PubMed WorldCat Hu B , Shen L , Liu S et al. . Enrichment of an anammox bacterial community from a flooded paddy soil . Environ Microbiol Rep . 2013 ; 5 : 483 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Hui C , Guo X , Sun P et al. . Depth-specific distribution and diversity of nitrite-dependent anaerobic ammonium and methane-oxidizing bacteria in upland-cropping soil under different fertilizer treatments . Appl Soil Ecol . 2017 ; 113 : 117 – 26 . Google Scholar Crossref Search ADS WorldCat Humbert S , Tarnawski S , Fromin N et al. . Molecular detection of anammox bacteria in terrestrial ecosystems: distribution and diversity . ISME J . 2010a ; 4 : 450 – 4 . Google Scholar Crossref Search ADS WorldCat Humbert S , Tarnawski S-E , Conen F et al. . Bactéries anammox terrestres: diversité et activité . Bull BGS . 2010b , 30 : 104 – 8 . OpenURL Placeholder Text WorldCat Humbert S , Zopfi J , Tarnawski S-E . Abundance of anammox bacteria in different wetland soils . Environ Microbiol Rep . 2012 ; 4 : 484 – 90 . Google Scholar Crossref Search ADS WorldCat Javanaud C , Michotey V , Guasco S et al. . Anaerobic ammonium oxidation mediated by Mn-oxides: from sediment to strain level . Res Microbiol . 2011 ; 162 : 848 – 57 . Google Scholar Crossref Search ADS PubMed WorldCat Katoh K , Standley DM . MAFFT multiple sequence alignment software version 7: Improvements in performance and usability . Mol Biol Evol . 2013 ; 30 : 772 – 80 . Google Scholar Crossref Search ADS PubMed WorldCat Kumar S , Herrmann M , Thamdrup B et al. . Nitrogen loss from pristine Carbonate-Rock aquifers of the hainich critical zone exploratory (Germany) is primarily driven by chemolithoautotrophic anammox processes . Front Microbiol . 2017 ; 8 : 1951 . Google Scholar Crossref Search ADS PubMed WorldCat Kuypers MMM , Sliekers AO , Lavik G et al. . Anaerobic ammonium oxidation by anammox bacteria in the Black Sea . Nature . 2003 ; 422 : 608 – 11 . Google Scholar Crossref Search ADS PubMed WorldCat Kuzyakov Y , Cheng W . Photosynthesis controls of rhizosphere respiration and organic matter decomposition . Soil Biol Biochem . 2001 ; 33 : 1915 – 25 . Google Scholar Crossref Search ADS WorldCat Könneke M , Bernhard AE , de la Torre JR et al. . Isolation of an autotrophic ammonia-oxidizing marine archaeon . Nature . 2005 ; 437 : 543 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat Lafleur PM , Hember RA , Admiral SW et al. . Annual and seasonal variability in evapotranspiration and water table at a shrub-covered bog in southern Ontario, Canada . Hydrol Process . 2005 ; 19 : 3533 – 50 . Google Scholar Crossref Search ADS WorldCat Landi L , Valori F , Ascher J et al. . Root exudate effects on the bacterial communities, CO2 evolution, nitrogen transformations and ATP content of rhizosphere and bulk soils . Soil Biol Biochem . 2006 ; 38 : 509 – 16 . Google Scholar Crossref Search ADS WorldCat Larsson A . AliView: a fast and lightweight alignment viewer and editor for large datasets . Bioinformatics . 2014 ; 30 : 3276 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Leininger S , Urich T , Schloter M et al. . Archaea predominate among ammonia-oxidizing prokaryotes in soils . Nature . 2006 ; 442 : 806 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Letunic I , Bork P . Interactive tree of life (iTOL) v3: an online tool for the display and annotation of phylogenetic and other trees . Nucleic Acids Res . 2016 ; 44 : W242 – 5 . Google Scholar Crossref Search ADS PubMed WorldCat Long A , Heitman J , Tobias C et al. . Co-Occurring Anammox, Denitrification, and Codenitrification in Agricultural Soils . Appl Environ Microbiol . 2013 ; 79 : 168 – 76 . Google Scholar Crossref Search ADS PubMed WorldCat Meng H , Wang Y-F , Chan H-W et al. . Co-occurrence of nitrite-dependent anaerobic ammonium and methane oxidation processes in subtropical acidic forest soils . Appl Microbiol Biotechnol . 2016 ; 100 : 7727 – 39 . Google Scholar Crossref Search ADS PubMed WorldCat Miller AJ , Cramer MD . Root nitrogen acquisition and assimilation . In: Lambers H , Colmer TD (eds). Root Physiology: From Gene to Function . Dordrecht : Springer Netherlands , 2005 , 1 – 36 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Minh BQ , Nguyen MAT , von Haeseler A . Ultrafast approximation for phylogenetic bootstrap . Mol Biol Evol . 2013 ; 30 : 1188 – 95 . Google Scholar Crossref Search ADS PubMed WorldCat Naeher S , Huguet A , Roose-Amsaleg CL et al. . Molecular and geochemical constraints on anaerobic ammonium oxidation (anammox) in a riparian zone of the Seine Estuary (France) . Biogeochemistry . 2015 ; 123 : 237 – 50 . Google Scholar Crossref Search ADS WorldCat Neef A , Amann R , Schlesner H et al. . Monitoring a widespread bacterial group: in situ detection of planctomycetes with 16S rRNA-targeted probes . Microbiol Read Engl . 1998 ; 144 : 3257 – 66 . Google Scholar Crossref Search ADS WorldCat Nguyen L-T , Schmidt HA , von Haeseler A et al. . IQ-TREE: A Fast and Effective Stochastic Algorithm for Estimating Maximum-Likelihood Phylogenies . Mol Biol Evol . 2015 ; 32 : 268 – 74 . Google Scholar Crossref Search ADS PubMed WorldCat Nicol GW , Leininger S , Schleper C et al. . The influence of soil pH on the diversity, abundance and transcriptional activity of ammonia oxidizing archaea and bacteria . Environ Microbiol . 2008 ; 10 : 2966 – 78 . Google Scholar Crossref Search ADS PubMed WorldCat Prosser JI , Nicol GW. Archaeal and bacterial ammonia-oxidisers in soil: the quest for niche specialisation and differentiation . Trends Microbiol . 2012 ; 20 : 523 – 31 . Google Scholar Crossref Search ADS PubMed WorldCat Pruesse E , Quast C , Knittel K et al. . SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB . Nucleic Acids Res . 2007 ; 35 : 7188 – 96 . Google Scholar Crossref Search ADS PubMed WorldCat Raghoebarsing AA , Pol A , van de Pas-Schoonen KT et al. . A microbial consortium couples anaerobic methane oxidation to denitrification . Nature . 2006 ; 440 : 918 – 21 . Google Scholar Crossref Search ADS PubMed WorldCat R Core Team . R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing . Vienna, Austria : the R Foundation for Statistical Computing , 2011 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Revsbech NP , Pedersen O , Reichardt W et al. . Microsensor analysis of oxygen and pH in the rice rhizosphere under field and laboratory conditions . Biol Fertil Soils . 1999 ; 29 : 379 – 85 . Google Scholar Crossref Search ADS WorldCat Rotthauwe JH , Witzel KP and Liesack W The ammonia monooxygenase structural gene amoA as a functional marker: molecular fine-scale analysis of natural ammonia-oxidizing populations . Appl. Environ. Microbiol. 1997 ; 63 : 4704 – 12 . Google Scholar Crossref Search ADS PubMed WorldCat Sato Y , Ohta H , Yamagishi T et al. . Detection of anammox activity and 16S rRNA genes in ravine paddy field soil . Microbes Environ . 2012 ; 27 : 316 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Schmid MC , Maas B , Dapena A et al. . Biomarkers for in situ detection of anaerobic Ammonium-Oxidizing (Anammox) bacteria . Appl Environ Microbiol . 2005 ; 71 : 1677 – 84 . Google Scholar Crossref Search ADS PubMed WorldCat Schmid MC , Risgaard-Petersen N , Van De Vossenberg J et al. . Anaerobic ammonium-oxidizing bacteria in marine environments: widespread occurrence but low diversity . Environ Microbiol . 2007 ; 9 : 1476 – 84 . Google Scholar Crossref Search ADS PubMed WorldCat Schubert CJ , Durisch-Kaiser E , Wehrli B et al. . Anaerobic ammonium oxidation in a tropical freshwater system (Lake Tanganyika) . Environ Microbiol . 2006 ; 8 : 1857 – 63 . Google Scholar Crossref Search ADS PubMed WorldCat Shan J , Zhao X , Sheng R et al. . Dissimilatory nitrate reduction processes in typical chinese paddy soils: Rates, relative contributions, and influencing factors . Environ Sci Technol . 2016 ; 50 : 9972 – 80 . Google Scholar Crossref Search ADS PubMed WorldCat Shen LD , Liu S , Lou LP et al. . Broad distribution of diverse anaerobic ammonium-oxidizing bacteria in chinese agricultural soils . Appl. Environ. Microbiol. 2013 ; 79 : 6167 – 72 . Google Scholar Crossref Search ADS PubMed WorldCat Shen J-P , Xu Z , He J-Z . Frontiers in the microbial processes of ammonia oxidation in soils and sediments . J Soils Sediments . 2014 ; 14 : 1023 – 9 . Google Scholar Crossref Search ADS WorldCat Shen L , Wu H , Gao Z et al. . Distribution and activity of anaerobic ammonium-oxidising bacteria in natural freshwater wetland soils . Appl Microbiol Biotechnol . 2016 ; 100 : 3291 – 300 . Google Scholar Crossref Search ADS PubMed WorldCat Shen L , Wu H , Gao Z et al. . Occurrence and importance of anaerobic ammonium-oxidising bacteria in vegetable soils . Appl Microbiol Biotechnol . 2015 ; 99 : 5709 – 18 . Google Scholar Crossref Search ADS PubMed WorldCat Sher Y , Baram S , Dahan O et al. . Ammonia transformations and abundance of ammonia oxidizers in a clay soil underlying a manure pond . FEMS Microbiol Ecol . 2012 ; 81 : 145 – 55 . Google Scholar Crossref Search ADS PubMed WorldCat Sonthiphand P , Neufeld JD . Evaluating primers for profiling anaerobic ammonia oxidizing bacteria within freshwater environments . PLOS ONE . 2013 ; 8 : e57242 . Google Scholar Crossref Search ADS PubMed WorldCat Stamatakis A . RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies . Bioinformatics . 2014 ; 30 : 1312 – 3 . Google Scholar Crossref Search ADS PubMed WorldCat Stookey LL . Ferrozine - a new spectrophotometric reagent for iron . Anal Chem . 1970 ; 42 : 779 – 81 . Google Scholar Crossref Search ADS WorldCat Thamdrup B , Dalsgaard T . Production of N2 through anaerobic ammonium oxidation coupled to nitrate reduction in marine sediments . Appl Environ Microbiol . 2002 ; 68 : 1312 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Tiedje JM . Ecology of denitrification and dissimilatory nitrate reduction to ammonium . Biol Anaerob Microorg . 1988 ; 717 : 179 – 244 . OpenURL Placeholder Text WorldCat van Kessel MAHJ , Speth DR , Albertsen M et al. . Complete nitrification by a single microorganism . Nature . 2015 ; 528 : 555 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Vitousek PM , Aber JD , Howarth RW et al. . Human alteration of the global nitrogen cycles: sources and consequences . Ecol Appl . 1997 ; 7 : 737 – 50 . OpenURL Placeholder Text WorldCat Wang J , Gu J-D . Dominance of Candidatus Scalindua species in anammox community revealed in soils with different duration of rice paddy cultivation in Northeast China . Appl Microbiol Biotechnol . 2013 ; 97 : 1785 – 98 . Google Scholar Crossref Search ADS PubMed WorldCat Wang Y , Zhu G , Harhangi HR et al. . Co-occurrence and distribution of nitrite-dependent anaerobic ammonium and methane-oxidizing bacteria in a paddy soil . FEMS Microbiol Lett . 2012 ; 336 : 79 – 88 . Google Scholar Crossref Search ADS PubMed WorldCat Wenk CB , Blees J , Zopfi J et al. . Anaerobic ammonium oxidation (anammox) bacteria and sulfide-dependent denitrifiers coexist in the water column of a meromictic south-alpine lake . Limnol Oceanogr . 2013 ; 58 : 1 – 12 . Google Scholar Crossref Search ADS WorldCat Wood ED , Armstrong FAJ , Richards FA . Determination of nitrate in sea water by cadmium-copper reduction to nitrite . J Mar Biol Assoc U K . 1967 ; 47 : 23 – 31 . Google Scholar Crossref Search ADS WorldCat Xi D , Bai R , Zhang L et al. . Contribution of anammox to nitrogen removal in two temperate forest soils . Appl Environ Microbiol . 2016 ; 82 : 4602 – 12 . Google Scholar Crossref Search ADS PubMed WorldCat Yang WH , Weber KA , Silver WL . Nitrogen loss from soil through anaerobic ammonium oxidation coupled to iron reduction . Nat Geosci . 2012 ; 5 : 538 – 41 . Google Scholar Crossref Search ADS WorldCat Yang X-R , Li H , Nie S-A et al. . Potential contribution of anammox to nitrogen loss from paddy soils in Southern China . Appl Environ Microbiol . 2015 ; 81 : 938 – 47 . Google Scholar Crossref Search ADS PubMed WorldCat Zheng D , Alm EW , Stahl DA et al. . Characterization of universal small-subunit rRNA hybridization probes for quantitative molecular microbial ecology studies . Appl Env Microbiol . 1996 ; 62 : 4504 – 13 . Google Scholar Crossref Search ADS WorldCat Zhou H , Zhang Q , Gu C et al. . Soil anammox community structure in different land use soils treatment with (13)C urea as determined by analysis of phospholipid fatty acids . Appl Microbiol Biotechnol . 2017 ; 101 : 6659 – 69 . Google Scholar Crossref Search ADS PubMed WorldCat Zhu G , Wang S , Wang Y et al. . Anaerobic ammonia oxidation in a fertilized paddy soil . ISME J . 2011 ; 5 : 1905 – 12 . Google Scholar Crossref Search ADS PubMed WorldCat © FEMS 2019. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Environmental factors determining distribution and activity of anammox bacteria in minerotrophic fen soils
JF - FEMS Microbiology Ecology
DO - 10.1093/femsec/fiz191
DA - 2020-02-01
UR - https://www.deepdyve.com/lp/oxford-university-press/environmental-factors-determining-distribution-and-activity-of-anammox-JE10fe0Dj1
VL - 96
IS - 2
DP - DeepDyve
ER -