Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Denitrification Activity of a Remarkably Diverse Fen Denitrifier Community in Finnish Lapland Is N-Oxide Limited

  • Katharina Palmer,

    Current address: Water Resources and Environmental Engineering Research Group, University of Oulu, Oulu, Finland

    Affiliation Department of Ecological Microbiology, University of Bayreuth, Bayreuth, Germany

  • Marcus A. Horn

    marcus.horn@uni-bayreuth.de

    Affiliation Department of Ecological Microbiology, University of Bayreuth, Bayreuth, Germany

Denitrification Activity of a Remarkably Diverse Fen Denitrifier Community in Finnish Lapland Is N-Oxide Limited

  • Katharina Palmer, 
  • Marcus A. Horn
PLOS
x

Abstract

Peatlands cover more than 30% of the Finnish land area and impact N2O fluxes. Denitrifiers release N2O as an intermediate or end product. In situ N2O emissions of a near pH neutral pristine fen soil in Finnish Lapland were marginal during gas chamber measurements. However, nitrate and ammonium fertilization significantly stimulated in situ N2O emissions. Stimulation with nitrate was stronger than with ammonium. N2O was produced and subsequently consumed in gas chambers. In unsupplemented anoxic microcosms, fen soil produced N2O only when acetylene was added to block nitrous oxide reductase, suggesting complete denitrification. Nitrate and nitrite stimulated denitrification in fen soil, and maximal reaction velocities (vmax) of nitrate or nitrite dependent denitrification where 18 and 52 nmol N2O h-1 gDW-1, respectively. N2O was below 30% of total produced N gases in fen soil when concentrations of nitrate and nitrite were <500 μM. vmax for N2O consumption was up to 36 nmol N2O h-1 gDW-1. Denitrifier diversity was assessed by analyses of narG, nirK/nirS, and nosZ (encoding nitrate-, nitrite-, and nitrous oxide reductases, respectively) by barcoded amplicon pyrosequencing. Analyses of ~14,000 quality filtered sequences indicated up to 25 species-level operational taxonomic units (OTUs), and up to 359 OTUs at 97% sequence similarity, suggesting diverse denitrifiers. Phylogenetic analyses revealed clusters distantly related to publicly available sequences, suggesting hitherto unknown denitrifiers. Representatives of species-level OTUs were affiliated with sequences of unknown soil bacteria and Actinobacterial, Alpha-, Beta-, Gamma-, and Delta-Proteobacterial sequences. Comparison of the 4 gene markers at 97% similarity indicated a higher diversity of narG than for the other gene markers based on Shannon indices and observed number of OTUs. The collective data indicate (i) a high denitrification and N2O consumption potential, and (ii) a highly diverse, nitrate limited denitrifier community associated with potential N2O fluxes in a pH-neutral fen soil.

Introduction

Northern peatlands are important players in the global carbon and nitrogen cycles, and store more than 30% of soil carbon and nitrogen even though they cover only about 3% of the terrestrial surface [1]. Greenhouse gases such as methane (CH4) and nitrous oxide (N2O) are produced in and released from northern peatlands soils [2]. High latitude peatlands have been intensively studied with respect to their capacity to emit CH4 due to the large amount of stored carbon in peat soils (e.g., [35]). N2O has a high global warming potential (approximately 300 times higher than CO2), is a major ozone-depleting substance, and 6% of the greenhouse effect is attributed to N2O [68]. Stored organic N in certain peatlands fuels N2O release via coupling of ammonification, ammonia oxidation, and denitrification [9]. Thus, potential N2O emissions from northern peatlands are of major interest. Northern peatlands are very diverse ecosystems, including many types of pristine and managed soils. Many studies investigating N2O emissions from peatlands have focused on N2O emissions from managed peatlands, and only recently N2O fluxes from pristine northern peat soils have been investigated [1015]. Pristine northern fens include significant net sources of N2O even though emission rates are generally low [10, 13]. Negative N2O fluxes suggest that peatlands can act as temporary sinks for N2O [2, 14, 16, 17]. Understanding of the mechanisms and regulation of N2O fluxes in such systems is still incomplete.

N2O in soils is generally produced during nitrification, denitrification, or chemical processes [18, 19]. Denitrification is considered to be the main source of N2O in water-saturated soils including peatlands [19, 14]. During denitrification, nitrate or nitrite are sequentially reduced via nitric oxide (NO) and N2O to dinitrogen (N2) [20]. The reductions are catalyzed by a set of oxidoreductases, namely nitrate reductases (encoded by narG and napA), nitrite reductases (encoded by nirK and nirS), NO reductases (endoded by norBC), and N2O reductases (encoded by nosZ) [20]. N2O or N2 can be released into the atmosphere. The ratio of N2O to N2 is determined by in situ parameters such as pH, temperature, as well as nitrate/nitrite and electron donor availability [21].

High latitude peatlands are likely to be strongly affected by increasing temperatures due to climate change [3, 4, 9, 22]. Global warming might reduce the water table in northern peatlands and influence the amount of N2O released from the soil [9]. A constantly lowered water table increases N2O fluxes from nutrient rich peat, whereas fluxes from nutrient-poor peat remain largely unaffected [12]. Dissimilar denitrifier communities are related with dissimilar N2O fluxes [15]. Detailed knowledge about the microbial catalysts involved in N2O turnover in northern peatlands is scarce. Thus, the aim of this study was to assess denitrification in a pH neutral pristine fen. The main objectives were to (i) assess in situ N2O emissions of a pH-neutral fen soil, (ii) determine depth-related N2O production and consumption capacities of fen soil, and (iii) link differences in the denitrifier community composition to physiological differences of the denitrification in the peat soil over two depths.

Material and Methods

Sampling site and soil parameters

Puukkosuo fen is located in northeastern Finland (66°22'38''N, 29°18'28''E) at an elevation of 220 m above sea level. The mean annual air temperature is (-0.43±0.09)°C, and mean annual precipitation approximates (772±12) mm (average of years 1966 to 2011, measured at Oulanka research station). The fen is meso-eutrophic and water saturated. Vegetation consists mainly of mosses (Sphagnum spp.) and grasses (e.g., Carex spp.). Four replicate soil cores from layers 0 to 20 cm and 20 cm to 40 cm were taken on July 28th 2010. Soil temperatures on the day of sampling were 17.2°C in surface soil and 15.1°C in deeper soil layers (below 15 cm). Samples were transported on ice to the laboratory and stored at 4°C for microcosm analyses or at -80°C for nucleic acid extractions. Microcosm experiments were conducted within 2 weeks after sampling. Nitrate, nitrite and ammonium concentrations, soil pH, soil moisture content, total carbon (TC), dissolved organic carbon (DOC) and total nitrogen (TN) were determined from pooled soil samples as described previously [23]. Permission to access and sample Puukkosuo fen was granted by Metsähallitus (www.metsa.fi) on 12th of July 2010.

Assessment of in situ gas emissions

In situ gas emissions of unfertilized soil and soil supplemented with either nitrate or ammonium were determined in closed poly(methyl methacrylate) (PMMA) chambers. Chambers were placed onto the soil surface and surrounded by metal collars, which had been inserted into the soil for a few centimeters to ensure that the chambers were gas tight. The transition between the plexiglas chamber and the metal ring was sealed with a rubber band to avoid exchange of gases from the chamber with the surrounding air. Before the installation of the gas chambers, 2l of fen pore water with 20 mM of added nitrate or ammonium was applied homogeniously onto the soil surface in 4 replicate treatments each and unsupplemented controls received pure fen pore water. Gas samples (5 ml per sampling timepoint) were taken from gas outlets and injected into gas tight evacuated containers (Exetainer, Labco Limited, High Wycombe, UK) at the start of the experiment, after 0.5, 1 and 3 hours.

Assessment of denitrification potentials in soil microcosms

Denitrification potentials of pH-neutral fen soil (0 to 20 cm and 20 to 40 cm) were assessed in unsupplemented and nitrate-, nitrite- or N2O-supplemented anoxic microcosms as described earlier [14, 15, 23]. Supplemental nitrate and nitrite ranged from 0 to 1000 μM, while supplemental N2O ranged from 0 to 4 μM. Acetylene blockage was used to distinguish between total N2O production and total denitrification as described earlier [15, 24]. Incubations were conducted at 20°C in the dark. N2O production rates and apparent kinetic parameters [Michaelis-Menten constants (KM) and maximum reaction velocitites (vmax)] were determined as described [14]. Michaelis-Menten regressions obtained for different incubation conditions were compared using the “extra sum of squares” principle to test for significant differences between the regressions [25]. Obtained values for KM and vmax were compared by t-tests.

Molecular characterisation of fen denitrifier communities

Nucleic acids were extracted from homogenized pooled fen soil of both soil layers as previously described using a bead-beating protocol [15, 26]. DNA yields were 4 to 12 μg DNA per gram (fresh weight) of soil. A260/A230 values approximating 0.94 to 1.56 indicated DNA with moderate to low humic acid content. The structural genes narG, nirK, nirS, and nosZ were amplified using the primer pairs narG1960f (TAY GTS GGS CAR GAR AA)/narG2650r (TTY TCR TAC CAB GTB GC) [27], F1aCu (ATC ATG GTS CTG CCG CG)/R3Cu (GCC TCG ATC AGR TTG TGG TT) [28], cd3aF (GTS AAC GTS AAG GAR ACS GG)/R3cd (GAS TTC GGR TGS GTC TTG A) [28], and nosZF (CGC TGT TCI TCG ACA GYC AG)/nosZR (ATG TGC AKI GCR TGG CAG AA) [29], respectively, and subjected to barcoded pyrosequencing as previously described [15, 23]. Barcodes used to identify sequences after pyrosequencing were ACTGCG and AGTATG for 0 to 20 cm and 20 to 40 cm fen soil, respectively. Pyrosequencing was performed at the Göttingen Genomics Laboratory using the Roche GS-FLX 454/Titanium technology as previously described [15, 23]. Pyrosequencing and PCR errors of the obtained reads were corrected using the AmpliconNoise pipeline [30] and sequences were clustered at species-level (i.e., for narG, nirK, nirS, and nosZ, respectively), and 97% threshold similarities using Qiime as previously described [23, 31]. Species-level threshold similarities were determined from pairwise comparisons of 16S rRNA gene similarities and structural gene similiarities of cultured denitrifiers [32]. Such OTUs indicate a minimal estimate of species-level diversity, i.e., is likely to underestimate “real” species-level diversity. Phylogenetic trees with cluster representatives were constructed in MEGA 5.0 [33]. Alpha- and beta-diversity measures were calculated in Qiime from rarified OTU tables as described [23, 34] to allow statistical comparison of the structural gene diversity from both soil layers. Rarified OTU tables were generated in Qiime by randomly subsampling original OTU tables 100 times at depth of 1000, 1500, 2500, and 500 sequences for narG, nirK, nirS, and nosZ, respectively. OTU representative sequences of narG, nirK, nirS, and nosZ were deposited at EMBL under accession numbers HE995549 to HE995577. Complete sequence data sets were deposited deposited in the European Nucleotide Archive (ENA) under the study accession number ERP008864.

Quantitative kinetic real-time PCRs (qPCRs) were performed in 6 technical replicates as described [15]. Obtained gene copy numbers were corrected for inhibition with inhibition factors ranging from 0.5–0.6, 0.3–0.4, 0.5–1.0, 0.9–1.0, and 0.9–1.0 for narG, nirK, nirS, nosZ, and 16S rRNA genes, respectively [35]. Normal distribution of the data was verified by Kolmogorov-Smirnov as well as Shapiro-Wilk tests. Copy numbers of narG, nirK, nirS, and nosZ in 0 to 20 cm and 20 to 40 cm soil were statistically evaluated using Student’s t-test (based on the 6 replicates for each gene).

Results

Soil parameters

Soil moisture content of Puukkosuo fen soil was 90% in both soil layers (Table 1). Soil pH in water was 6.8 and 6.9 in 0 to 20 cm and 20 to 40 cm fen soil, respectively. Nitrate was below the detection limit of 5.8 μg gDW-1 (Table 1). Values for carbon and nitrogen contents appeared to be marginally higher in 20 to 40 cm than in 0 to 20 cm fen soil, but C/N ratios and DOC concentrations were similar in both soil layers (Table 1).

In situ gas emissions of fen soil

During gas chamber measurements, only minor amounts of N2O accumulated in gas chambers placed on unsupplemented fen soil on average (Fig 1). Increases of about 10 ppb in N2O mixing ratio were observed in two of the four replicate gas chambers, while decreases in N2O mixing ratio were observed in the other two replicate gas chambers (-1 to -17 ppb decrease in mixing ratio). Nitrate-addition initially lead to accumulation of N2O in the gas chambers. However, this accumulation of N2O was restricted to the first 30 minutes after nitrate-addition, and initially accumulated N2O was subsequently consumed after 30 minutes (Fig 1). Ammonium likewise led to accumulation of N2O in the gas chambers, however this initial accumulation of N2O was slower than after nitrate-addition (Fig 1). Moreover, initially accumulated N2O was subsequently consumed after the first hour.

thumbnail
Fig 1. Effect of in situ nitrate and ammonium fertilization on N2O accumulation in closed chambers from pH-neutral fen soil.

Mean values and standard errors of 4 replicates are displayed. Closed circles represent unfertilized controls, closed and open squares represent soil fertilized with 20 mM NaNO3 and 20 mM NH4Cl, respectively.

https://doi.org/10.1371/journal.pone.0123123.g001

Denitrification potentials in fen soil microcosms

In anoxic microcosms, unsupplemented fen soil from both soil layers produced only minor amounts of N2O in the absence of acetylene, and initially produced N2O was subsequently consumed (Fig 2). However, N2O production was significantly higher in anoxic microcosms when N2O-reductase was blocked by acetylene (Fig 2). N2O mixing ratios increased from 0.04 ± 0.004 to about 35 ± 3.5 ppm within the first 94 hours in acetylene-amended microcosms with 0 to 20 cm fen soil, and the concentration of N2O plateaued out after the first 94 hours (Fig 2), indicating that endogenous nitrate had been consumed. The increase in N2O mixing ratio was significantly lower in microcosms with 20 to 40 cm fen soil than in those with 0 to 20 cm fen soil (Fig 2). N2O mixing ratios increased from 0.03 ± 0.001 to 1.4 ± 0.5 ppm within the first 94 hours in acetylene-amended microcosms with 20 to 40 cm fen soil. In 20 to 40 cm fen soil microcosms without acetylene, mixing ratios increased only to 91 ± 21 bbp N2O within the first 94 hours. The initially accumulated N2O was subsequently consumed within the next 74 hours. Both soil layers displayed the capability to consume subatmospheric concentrations of N2O in microcosms without acetylene. 0 to 20 cm fen soil reduced N2O from 290 ppb to 55 ppb, while 20 to 40 cm fen soil reduced N2O from 91 ppb to 39 ppb (Fig 2).

thumbnail
Fig 2. N2O production and consumption in unsupplemented anoxic fen soil microcosms.

Squares and circles represent 0 to 20 cm and 20 to 40 cm fen soil, respectively. Open and closed symbols represent microcosms without and with acetylene addition, respectively. Mean values and standard errors of three replicates are displayed. The inset represents an enlargement of the lower N2O mixing ratios to allow better visualization of N2O production and consumption in 20 to 40 cm fen soil and microcosms without acetylene addition. The horizontal line indicates the atmospheric N2O mixing ratio (319 ppb).

https://doi.org/10.1371/journal.pone.0123123.g002

Supplemental nitrate and nitrite stimulated the production of N2O without apparent delay in microcosms with fen soil from both soil layers (Fig 3 A and S1 Fig), while N2O consumption was stimulated in N2O supplemented microcosms (Fig 3 B). Stimulation of N2O production with nitrate was smaller than with nitrite, and N2O production in nitrate-supplemented microcosms was less than 25% of that in nitrite-supplemented microcosms (Fig 3 A and S1 Fig). In microcosms with fen soil from 0 to 20 cm, N2O production in acetylene-amended microcosms was in a similar magnitude for all supplemented nitrate concentrations ≥100 μM. N2O production in microcosms with fen soil from 20 to 40 cm was highest when 50 μM nitrate were supplied, and decreased with increasing nitrate concentrations, indicating that denitrifiers in fen soil were saturated at low nitrate concentrations, and were inhibited by higher nitrate concentrations (Fig 3 A). In nitrite-supplemented microcosms, N2O production rates increased with increasing nitrite concentrations in both soil layers (Fig 3 A). N2O consumption was likewise stimulated by increasing N2O concentrations (Fig 3 B). N2O production and consumption capacities were higher in 0 to 20 cm fen soil than in 20 to 40 cm fen soil.

thumbnail
Fig 3. Effect of added nitrate, nitrite (A) and N2O (B) on N2O production and consumption in anoxic fen soil microcosms.

Closed and open symbols represent 0 to 20 cm and 20 to 40 cm fen soil microcosms, respectively. Circles, squares and triangles represent microcosms supplemented with nitrate, nitrite or N2O, respectively. Mean values and standard errors of three replicates are displayed. The x-axis displays the amount of supplemented (i.e., additional) nitrate, nitrite or N2O. Solid lines indicate Michaelis-Menten curves fitted to the data.

https://doi.org/10.1371/journal.pone.0123123.g003

The ratio of N2O to (N2 + N2O) was below 30% and 40% for all supplied nitrate concentrations in microcosms with 0 to 20 cm and 20 to 40 cm fen soil, respectively (S2 Fig), indicating that more than half of the N2O produced from nitrate was further reduced to N2 in fen soil. The ratio of N2O to (N2 + N2O) was below 30% in microcosms with fen soil from 0 to 20 cm when nitrite concentrations were 100 μM or smaller and increased to about 75% for higher nitrite concentrations. In microcosms with 20 to 40 cm fen soil, the ratio of N2O to (N2 + N2O) was between 50% and 100% for all supplied nitrite concentrations (S2 Fig), indicating that N2O was a major product of denitrification in that soil layer when nitrite was provided as electron acceptor.

Initial nitrite-dependent N2O production rates of fen soil microcosms amended with acetylene followed apparent Michaelis-Menten kinetics, as did nitrate-dependent N2O production rates of fen soil microcosms from 0 to 20 cm depth and N2O-dependent N2O consumption rates in both layers (Fig 3). The Michaelis-Menten kinetics differed significantly between the different treatments and soil layers (p ≤ 0.03 for all comparisons). Apparent maximal reaction velocities (vmax) were highest for nitrite-dependent N2O production, followed by N2O-dependent N2O consumption rates. vmax was lowest for nitrate-dependent N2O production (Table 2). vmax values for nitrate and nitrite dependent N2O production, as well as N2O-dependent N2O consumption were significantly higher in 0–20 cm than 20–40 cm fen soil (p < 0.001, and p < 0.001, as well as p = 0.02, respectively). Apparent Michaelis-Menten constants KM were about 60 to 140 times lower for N2O consumption than for nitrite dependent N2O production in 0 to 20 cm fen soil (p = 0.003), indicating a high affinity of fen denitrifiers for N2O (Table 2).

Phylogenetic analysis of denitrifiers in high latitude peatlands

Approximately 14 000 denoised quality-filtered sequences of the structural gene markers narG, nirK, nirS, and nosZ were utilized in total for further analyses. Forward and reverse reads for nirK and nirS showed a sufficiently long overlap (amplicon lengths of approximately 470 and 410 bp, respectively) to allow combined assessment of forward and reverse reads per gene for further analyses. Only forward reads of narG and nosZ were analyzed, as the overlap of forward and reverse reads was not sufficient to allow a combined analysis of forward and reverse reads (amplicon lengths approximately 670 and 700 bp for narG and nosZ, respectively), and previous studies indicate that results obtained from forward and reverse reads of narG and nosZ are similar [15, 23]. More than 99% of sequences generated from amplicons of a certain gene specific (i.e., narG, nirK, nirS, nosZ) primer set were specific amplicons of the target gene. All library coverages were greater than 99% at species-level DNA sequence dissimilarities of 33%, 17%, 18%, and 20% for narG, nirK, nirS, and nosZ, respectively, and varied from 80% to 97% at 3% sequence dissimilarity (Table 3), indicating that the number of sequences generated was sufficient.

thumbnail
Table 3. Diversity measures of amplicon pyrosequencing libraries of fen soil from rarified and non-rarified OTU tables of narG, nirK, nirS and nosZ.

https://doi.org/10.1371/journal.pone.0123123.t003

narG sequences were assigned to 7 species-level OTUs in total. 7 and 4 OTUs were detected in 0 to 20 cm and 20 to 40 cm of fen soil, respectively (Table 3). narG community composition was similar in both sampled soil layers (Fig 4 A). Three OTUs had a relative abundance greater than 1%. Of those OTUs, OTU 1 dominated narG in fen soil (about 60% in both soil layers). About 40% of narG belonged to OTUs 2 and 3 (Fig 4 A). OTU 2 was more abundant in 0 to 20 cm than in 20 to 40 cm fen soil (relative abundances of 33% and 9%, respectively), whereas OTU 3 was more abundant in 20 to 40 cm fen soil (23% vs. 6% in 0 to 20 cm fen soil; Fig 4 A). Most of the OTUs were only distantly related to narG of cultured organisms or environmental sequences (i.e., sequence dissimilarities of OTU representatives were 10–23%) (Table 4 and S3 Fig). Sequences of OTUs 1, 2, and 3 affiliated with narG of Alphaproteobacteria, Actinobaceria, and Deinococci, respectively, more specifically they were related to narG of uncultured bacteria and to those of Oligotropho carboxidovorans, Salinispora arenicola, and Marinithermus hydrothermalis, respectively (Table 4 and S3 Fig). Observed narG diversity was higher at 97% threshold similarity than at species-level threshold similarity (Table 3). At 97% threshold similarity, 359 and 230 OTUs were detected in 0 to 20 cm and 20 to 40 cm fen soil, respectively (Table 3). Shannon diversity, species evenness indices, and the observed number of OTUs calculated from rarified OTU tables indicated significantly higher diversity in 0–20 cm than 20–40 cm fen soil at 97% and species-level threshold similarity (Table 3). Beta-diversity measures indicated greater differences in community composition at 97% than at 67% threshold similarity (Table 3).

thumbnail
Fig 4. Relative abundances of denitrification associated genes in amplicon libraries of pH-neutral fen soil.

OTUs of narG forward reads (A), nirK (B), nirS (C), and nosZ forward reads (D) were derived at species-level thresholds of 33%, 17%, 18%, and 20%, respectively.

https://doi.org/10.1371/journal.pone.0123123.g004

nirK were assigned to 24 species-level OTUs in total. 23 and 17 OTUs were detected in fen soil from 0 to 20 cm and from 20 to 40 cm, respectively (Table 3). Community composition differed significantly between the soil layers (Fig 4 B). OTU 2 dominated nirK in fen soil from 0 to 20 cm (about 60%), while OTU 1 dominated nirK in fen soil from 20 to 40 cm, respectively (about 70%; Fig 4 B). Similarities of OTU representative sequences to nirK of cultured organisms ranged from 75–100% (Table 4). Most OTUs were related to Alphaproteobacterial nirK. OTUs 1, 2, and 3 were related to nirK of Brucella canis, Rhizobium etli, and Castellaniella sp., respectively (Table 4 and S4 Fig). Further OTUs were related to nirK of Bosea sp., Afipia sp., or uncultured bacteria (Table 4 and S4 Fig). nirS were assigned to 25 species-level OTUs in total. 22 and 23 OTUs were detected in fen soil from 0 to 20 cm and from 20 to 40 cm, respectively (Table 3). Differences in community composition of nirS from the soil layers were more pronounced than those of nirK (Fig 4 B and 4 C). nirS of fen soil was dominated by OTUs affiliated to Beta- and Gammaproteobacterial nirS. However, about 26% of detected nirS from 20 to 40 cm affiliated with Alphaproteobacterial nirS (S5 Fig). nirS of OTU representatives were only distantly related to nirS of cultured organisms (i.e., similarities ranged from 74–84%, Table 4). Many OTUs of both soil layers were related to nirS of uncultured wetland or marine sediment bacteria, and distantly related to nirS of e.g., Thiobacillus denitrificans, Dechloromonas sp., and Arthrobacter sp. (Table 4 and S5 Fig). Diversity estimates calculated from rarified OTU tables of nirK based on species-level threshold similarities differed significantly between 0–20 and 20–40 cm fen soil (Table 3). Chao1 richness estimates of nirS did not differ significantly at species-level similarity thresholds, amounting to about 24 in both soil layers, while Shannon diversity as well as species evenness were significantly higher in the lower soil layer (Table 3). On the contrary, Shannon diversity, species Evenness, and Chao1 richness estimates of nirK and nirS calculated from rarified OTU tables based on 97% threshold similarity were consistently higher in 0 to 20 cm than 20–40 cm fen soil (Table 3).

nosZ forward reads were assigned to 10 species-level OTUs in total. 8 OTUs were detected in each soil layer (Table 3). OTU 1 dominated nosZ of fen soil from both soil layers (Fig 4 D). Essentially all nosZ from both soil layers affiliated with Alpha- and Betaproteobacterial nosZ (S6 Fig). Most nosZ sequences from fen soil were distantly related to nosZ of cultured organisms with sequence dissimilarities ranging from 11–27% (Table 4), indicating hitherto uncultured denitrifiers capable of N2O reduction in fen soil. nosZ sequences clustered with nosZ of wetland and upland soils, as well as Achromobacter sp., Herbaspirillum sp., and Ralstonia sp. (Table 4 and S6 Fig). Shannon diversity and species evenness calculated from rarified OTU tables at species-level threshold similarity were significantly higher in 0 to 20 cm than in 20 to 40 cm soil, while there was no significant difference in Chao1 richness estimates (Table 3). At 97% threshold similarity, all diversity estimates calculated from rarified OTU tables were significantly higher in the upper soil layer (Table 3). The difference in threshold similarity most strongly affected on the number of observed and estimated OTUs, which were similar at species-level threshold similarity (around 8 in both soil layers), but were about 3 times higher in 0 to 20 cm soil at 97% similarity threshold (Table 3). Beta-diversity was higher at 97% than at species-level threshold similarity (Table 3).

Quantification of narG, nirK, nirS, and nosZ relative to 16S rRNA genes

Copy numbers of all genes investigated in this study were corrected by inhibition factors that were experimentally determined for every DNA extract and gene analyzed (see Material and Methods). 16S rRNA gene copy numbers of 0–20 and 20–40 cm fen soil were (5.3 ± 0.3) x 105 and (8.6 ± 0.3) x 105 per ng DNA. Copy numbers of narG accounted for 7 and 3% of bacterial 16S rRNA gene copy numbers in 0 to 20 cm and 20 to 40 cm fen soil, respectively (Fig 5). Copy numbers of nirK, nirS, and nosZ were lower than narG copy numbers (Fig 5). Copy numbers of nirS were app. 100x and 10x higher than copy numbers of nirK and nosZ, respectively, in both soil layers (Fig 5). Copy numbers of narG and nosZ were 3 x higher (P < 0.01), and those of nirK were slightly lower in 0 to 20 cm than 20 to 40 cm fen soil (P = 0.1). Ratios of nosZ to narG were similar in both soil layers. Those of nosZ to nirK and nirS were 30 and 3 x higher, respectively, in 0 to 20 cm than 20 to 40 cm fen soil (Fig 5).

thumbnail
Fig 5. Abundance of denitrification associated genes normalized by 16S rRNA gene copy numbers in pH-neutral fen soil.

The inset shows ratios of normalized nosZ gene abundances and narG, nirK, as well as nirS. Mean values and standard errors of six replicates are displayed. Black and white bars indicate fen soil from 0–20 and 20–40 cm depth, respectively.

https://doi.org/10.1371/journal.pone.0123123.g005

Discussion

pH neutral fen soil as N2O sink

Peatlands are important ecosystems in the northern hemisphere and cover more than 30% of the Finnish land surface [36]. The potential of those peatlands to produce or consume greenhouse gases is of great interest, especially in respect to climate warming which is predicted to have a strong impact on peatlands [37]. N2O emissions from natural wetlands are highly variable, and many water-saturated soils are also sinks for N2O [2, 16, 17]. Many studies demonstrate that undrained, pH-neutral and acidic fens are sources of molecular nitrogen, and act as sinks for N2O depending on environmental conditions [10, 14, 16, 3840]. N2O accumulation in gas chamber experiments from Puukkosuo fen were also variable, ranging from 10 ppb to -17 ppb at the time of soil sampling. Fen soil in situ consumed initially produced N2O during nitrate or ammonium fertilization experiments (Fig 1). Previous studies show that mainly complete denitrification to N2 occurs in pristine pH-neutral fens at in situ nitrate concentrations [10, 40]. Thus, the absence of in situ N2O emission from Puukkosuo fen soil is likely due to complete denitrification to N2 as the major end product (Fig 1). Even though the amount of stored nitrogen in the soil is high, low concentrations of available nitrate are observed in a northern boreal fen, where denitrification is thus N-limited [10]. Nitrate concentrations in Puukkosuo fen soil were likewise low, nitrite was not detected (Table 1), and nitrate as well as nitrite stimulated denitrification (Fig 3), indicating nitrate- and nitrite-limitation of fen denitrifiers. Microcosms and in situ fertilization with nitrate resulting in temporary in situ emission of N2O with subsequent consumption indicated ongoing complete denitrification (Fig 1).

Peatland soils are temporarily or permanently water-logged, and oxygen generally penetrates only the uppermost centimeters, leading to oxygen-limitation in lower soil layers. In the absence of oxygen and nitrate, nitrous oxide is a potent sink for electrons released during the oxidation of organic carbon compounds, as the reduction of N2O by H2 is even more exergonic than O2 reduction by H2 (N2O half-cell potential of E0 (pH 7.0) = 1.35 V; ΔG0 = -339.5 kJ*mol-1; reviewed in [41]). In situ relevant concentrations of dissolved organic carbon (app. 5 μM glucose equivalents; e.g., [42]) and atmospheric concentrations of N2O result in -360 kJ*molN2O-1 (http://cms.uni-konstanz.de/schink/dg-calculator/), allowing for high energy conservation of organisms capable of N2O reduction. Lower below-surface N2O concentrations than at atmospheric equilibrium are observed in fens and suggest ongoing N2O consumption [10, 16, 38]. Indeed, apparent KM values of Puukkosuo fen soil were approximately 60 times lower for N2O than for nitrate (Table 2), indicating a higher affinity of fen denitrifiers for N2O than for nitrate. The assumed absence of oxygen, the observed nitrate-limitation and high N2O affinity indicate a strong in situ sink potential of Puukkosuo fen for N2O.

Diverse denitrifier communities are associated with denitrification activities in pH-neutral fen soil

Unsupplemented fen soil from both sampled soil layers produced N2O in acetylene-amended microcosms, demonstrating the denitrification potential of the fen soil. However, nearly no N2O was produced in the absence of acetylene and initially produced N2O was subsequently consumed (Fig 2). Nitrate- and oxygen-limitation might select for denitrifiers capable of complete denitrification, and hitherto unknown denitrifiers as well as N2O reducers might occur in Puukkosuo fen soil. Indeed, nosZ copy numbers in 0 to 20 cm fen soil were of a similar magnitude as nitrite reductase copy numbers (Fig 5), and newly-discovered nirK/S and nosZ (Table 4) indicate that a high percentage of uncharacterized denitrifiers in that soil layer possessed a complete denitrification pathway.

Supplemental nitrate and nitrite resulted in immediate N2O production in fen soil after internal nitrate and nitrite were consumed. Stimulation was greater with nitrite than with nitrate in both soil layers (Fig 3 A). This reflects the fact that all denitrifiers “sensu stricto” use nitrite as electron acceptor, while many cultured denitrifiers lack the ability to use nitrate as electron acceptor [20, 43]. Stimulation was also greater in the top soil layer (Fig 3 A), reflecting a greater denitrification potential of the top soil. In other wetland and also agricultural soils, denitrification potentials are also highest in the top soil layers (e.g., [14, 44]). In 20 to 40 cm fen soil, N2O production and total denitrification decreased with increasing nitrate concentrations, indicating substrate inhibition of denitrification at high nitrate concentrations. This finding is in contrast to denitrification potentials reported for deeper horiozonts of agricultural soils, suggesting that the fen denitrifier community of 20–40 cm depth is well adapted to low nitrate concentrations (e.g., [44]). Nitrate and nitrite reduction compete for electrons at high nitrate concentrations, and nitrate reduction is favored over the rest of the denitrification pathway, causing eventually accumulation of nitrite when electron donors are limiting [45].

The ratio of N2O to (N2+N2O) was lower in nitrate- and nitrite-amended microcosms with 0 to 20 cm fen soil than in 20 to 40 cm fen soil when nitrate or nitrite were supplied (S2 Fig), and consumption of supplied N2O was about 2-fold higher in 0 to 20 cm than in 20 to 40 cm fen soil (Fig 3 B). Indeed, the ratio of nitrite to nitrous oxide reductases was higher in 20 to 40 cm fen soil than in 0 to 20 cm fen soil (Fig 5), indicating an increased amount of denitrifiers lacking nitrous oxide reductase in the lower soil layer. The ratio of nitrite reductase genes to N2O reductase genes is highly variable in soils, and often nitrite reductase copy numbers largely exceed N2O reductase copy numbers [15, 23, 46]. However, non-denitrifying N2O consumers were recently shown to be quantitatively important in certain soils [47, 48]. Relative abundances of both atypical and typical nosZ assigned to non-denitrifiers and denitrifiers, respectively, are variable in soil metagenomes. Hence, further analyses including both groups are demanded for better understanding of N2O reducers in fens [49]. Nevertheless, N2O produced in lower layers of fen soil can diffuse upwards and be further reduced to N2 in upper soil layers, and thus emission of N2O into the atmosphere can be reduced [14, 16, 38]. It is thus hypothesized that also in Puukkosuo fen soil lower soil layers are N2O sources while upper soil layers are N2O sinks.

The analysis of denitrification-specific gene markers indicated a higher diversity of these genes in 0 to 20 cm than in 20 to 40 cm fen soil (Table 3). Detected narG and nosZ were more similar in 0 to 20 cm and 20 to 40 cm fen soil than nirK and nirS (Fig 4), indicating that nitrite reductases show a higher variability in fen soil than nitrate and N2O reductases. Nitrite reductase community composition is highly variable in other types of peatland soils, including permafrost affected systems, while variations in nitrate and N2O reductase community composition are much less pronounced [15, 23]. Indeed, the distribution of nitrite reductases is more heavily impacted by changes in environmental conditions than those of nitrate or N2O reductases [5052]. Nitrite reductase genes from fen soil were affiliated with Proteobacterial nirK/S (Table 4 and S4 Fig and S5 Fig). For nirS, sequences related to Rhodanobacter/Bradyrhizobium were detected (S5 Fig). Such sequences are also detected in other peatland soils such as permafrost affected tundra and palsa peat soils [15, 23]. Proteobacteria-affiliated sequences of narG and nosZ (Table 4 and S3 Fig and S6 Fig) further support that Proteobacteria play an important role for denitrification in this pH-neutral fen soil. Denitrification-associated genes related to Proteobacteria are also found in acidic fen soils or permafrost-affected peatlands [14, 15, 23], indicating that Proteobacteria represent general peatland denitrifiers. Sequences of narG were also affiliated with Actinobacterial narG (10–30%; S3 Fig). Actinobacteria are common in soils, include many genera capable of nitrate reduction, and are in general considered to be more tolerant to extreme environmental conditions such as low pH or low temperature [5355]. Actinobacteria and Actinobacteria-affiliated gene markers are frequently detected in a variety of peatlands including acidic fen soils, permafrost-affected tundra and palsa peat soils [14, 15, 23]. However, in those more extreme environments, Actinobacteria often dominate the narG communities, indicating that Actinobacteria are further important players involved in nitrate reduction and potentially denitrification in pH-neutral fen soil [15, 23]. The nitrate reducer community in pH-neutral fen soil also contained a substantial portion of Deinococci-affiliated narG (Fig 4A and S3 Fig.), which are not detected in the above mentioned more extreme habitats such as acidic fens, frost-affected tundra and palsa peat soils [15, 23]. Soil pH is a driver of the general microbial community structure [56]. Denitrifier diversity in pH-neutral fen soil is high when compared to more acidic pristine peatland soils [14, 15, 23], suggesting that soil pH likewise plays an important role in shaping denitrifier communities.

Denitrifier diversity and quantity is routinely underestimated due to choice of primer sets, e.g., gram-positive denitrifiers escaped detection in many studies [46, 50, 57]. Soil metagenomes might represent an alternative strategy to obtain a more complete picture of denitrifier diversity in soils. However, the low abundance of denitrification associated genes on denitrifier genomes (i.e., app. 1%; most of the genes on denitrifier genomes are associated with other functions than denitrification like anabolism, motility, etc.) in combination with a low number of denitrifiers compared to total number of prokaryotes in soil (app. 1%) limits their detection by metagenomics [49, 58, 59]. However, metagenomes are extremely useful for the design of denitrification gene specific primers. Although amplicon based approaches combined with next generation sequencing depend on the choice of primers, such approaches currently provide a cost-effective way for the detection of a large denitrifier diversity.

The collective data indicate that (i) a core nitrate reducer/denitrifier community might be common to all kinds of (northern) peatlands, (ii) some nitrate reducers/denitrifiers are unique in pH-neutral fen soil, possible due to the lack of environmental stress that might be induced by acidic pH, (iii) denitrifier communities are from upper and lower layers are dissimilar as indicated by apparent Michaelis-Menten kinetics and structural gene marker analyses, and (iv) pH-neutral fens are a strong potential sink for atmospheric N2O.

Supporting Information

S1 Fig. Effect of supplemental nitrate (1) and nitrite (2) on N2O production and consumption in microcosms with fen soil.

Squares and circles represent fen soil from 0 to 20 cm and 20 to 40 cm depth, respectively. Microcosms with and without acetylene are represented by closed and open symbols, respectively. Supplied concentrations of nitrate or nitrite were 0 μM (A), 10 μM (B), 20 μM (C), 50 μM (D), 100 μM (E), 500 μM (F), and 1000 μM (G). Mean values and standard errors of three replicate microcosms are shown.

https://doi.org/10.1371/journal.pone.0123123.s001

(TIF)

S2 Fig. Effect of supplemental nitrate (black) or nitrite (white) on the ratio of N2O to (N2+N2O) in anoxic microcosms with fen soil from 0 to 20 cm (A) and 20 to 40 cm (B) depth.

Mean values and standard errors of three replicates are shown.

https://doi.org/10.1371/journal.pone.0123123.s002

(TIF)

S3 Fig. Phylogenetic tree of narG OTU representatives detected in 0 to 20 cm and 20 to 40 cm fen soil.

The trees was calculated based in translated amino acid sequences of narG forward reads. OTUs were grouped at a species-level threshold dissimilarity of 33%. Numbers preceeding sequence names refer to sequence accession numbers of reference sequences from public databases. Values given in parentheses show the relative abundances of each OTU in 0 to 20 cm (left) and 20 to 40 cm (right) fen soil. Grey boxes indicate reference sequences belonging to the same phylogenetic group. The percentage of replicate trees that produced the observed clustering of taxa in the bootstrap test (10 000 replications) are shown next to the branches. Bootstrap supports below 50% are not displayed. narG of Haloarcula marismortui ATCC 43049 was used as outgroup to root the tree.

https://doi.org/10.1371/journal.pone.0123123.s003

(TIF)

S4 Fig. Phylogenetic tree of nirK OTU representatives detected in 0 to 20 cm and 20 to 40 cm fen soil.

The tree was calculated based on translated amino acid sequences. OTUs were grouped at a species-level threshold dissimilarity of 17%. Numbers preceeding sequence names refer to sequence accession numbers of reference sequences from public databases. Values given in parentheses show the relative abundances of each OTU in 0 to 20 cm (left) and 20 to 40 cm (right) fen soil. Grey boxes indicate reference sequences belonging to the same phylogenetic group. The percentages of replicate trees that produced the observed clustering of taxa in the bootstrap test (10 000 replications) are shown next to the branches. Bootstrap supports below 50% are not displayed. nirK of Nitrosomonas sp. C-56 was used as outgroup to root the tree.

https://doi.org/10.1371/journal.pone.0123123.s004

(TIF)

S5 Fig. Phylogenetic tree of nirS OTU representatives detected in 0 to 20 cm and 20 to 40 cm fen soil.

The tree was calculated based on translated amino acid sequences. OTUs were grouped at a species-level threshold dissimilarity of 18%. Numbers preceeding sequence names refer to sequence accession numbers of reference sequences from public databases. Values given in parentheses show the relative abundances of each OTU in 0 to 20 cm (left) and 20 to 40 cm (right) fen soil. Grey boxes indicate reference sequences belonging to the same phylogenetic group, white boxes indicate single taxa not belonging to the major phylogenetic group. The percentages of replicate trees that produced the observed clustering of taxa in the bootstrap test (10 000 replications) are shown next to the branches. Bootstrap supports below 50% are not displayed. nirS of Rhodothermus marinus DSM 4252 was used as outgroup to root the tree.

https://doi.org/10.1371/journal.pone.0123123.s005

(TIF)

S6 Fig. Phylogenetic tree of nosZ OTU representatives detected in 0 to 20 cm and 20 to 40 cm fen soil.

The tree was calculated based on translated amino acid sequences of nosZ forward reads. OTUs were grouped at a species-level threshold dissimilarity of 20%. Numbers preceeding sequence names refer to sequence accession numbers of reference sequences from public databases. Values given in parentheses show the relative abundances of each OTU in 0 to 20 cm (left) and 20 to 40 cm (right) fen soil. Grey boxes indicate reference sequences belonging to the same phylogenetic group, white boxes indicate single taxa not belonging to the major phylogenetic group. The percentages of replicate trees that produced the observed clustering of taxa in the bootstrap test (10 000 replications) are shown next to the branches. Bootstrap supports below 50% are not displayed. nosZ of Haloarcula marismortui ATCC 43049 was used as outgroup to root the tree.

https://doi.org/10.1371/journal.pone.0123123.s006

(TIF)

Acknowledgments

We are thankful to Jyrki Manninen for organisational help, the team at Oulanka research station for excellent on-site support, Christian Hofmann for assistance with gas measurements, the Central Analytics Department of BayCEER for analyses of nitrate, nitrite, ammonium, TC, TN, and TOC, Rolf Daniel and Andrea Thürmer for pyrosequencing, and Steffen Kolb, Markus Nebel, Sebastian Wild, Justin Kuczynski and Christopher Quince for help with sequence analyses.

Author Contributions

Conceived and designed the experiments: KP MAH. Performed the experiments: KP. Analyzed the data: KP MAH. Contributed reagents/materials/analysis tools: MAH. Wrote the paper: KP MAH.

References

  1. 1. Gorham E. Northern peatlands—role in the carbon-cycle and probable responses to climatic warming. Ecol Appl 1991;1: 182–195.
  2. 2. Takakai F, Desyatkin AR, Lopez CML, Fedorov AN, Desyatkin RV, Hatano R. CH4 and N2O emissions from a forest-alas ecosystem in the permafrost taiga forest region, eastern Siberia, Russia. J Geophys Res—Biogeo 2008;113: G02002.
  3. 3. Christensen TR, Ekberg A, Strom L, Mastepanov M, Panikov N, Oquist M, et al. Factors controlling large scale variations in methane emissions from wetlands. Geophys Res Lett 2003;30: 1414.
  4. 4. Roulet N, Moore T, Bubier J, Lafleur P. Northern fens—methane flux and climatic change. Tellus B 1992;44: 100–105.
  5. 5. Whalen SC. Biogeochemistry of methane exchange between natural wetlands and the atmosphere. Environ Eng Sci 2005;22: 73–94.
  6. 6. Forster P, Ramaswamy V, Artaxo P, Berntsen T, Betts R, Fahey D, et al. Changes in atmospheric constituents and in radiative forcing. In: Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB, et al., editors. Climate Change 2007: The physical science basis. Contribution of working group I to the fourth assessment report of the intergovernmental panel on climate change. Cambridge: Cambridge University Press; 2007. pp. 129–234.
  7. 7. Schulze ED, Luyssaert S, Ciais P, Freibauer A, Janssens IA, Soussana JF, et al. Importance of methane and nitrous oxide for Europe's terrestrial greenhouse-gas balance. Nature Geosci 2009;2: 842–850.
  8. 8. Ravishankara AR, Daniel JS, Portmann RW. Nitrous oxide (N2O): The dominant ozone-depleting substance emitted in the 21st century. Science 2009;326: 123–125. pmid:19713491
  9. 9. Smith KA. The potential for feedback effects induced by global warming on emissions of nitrous oxide by soils. Glob Change Biol 1997;3: 327–338.
  10. 10. Lohila A, Aurela M, Hatakka J, Pihlatie M, Minkkinen K, Penttilä T, et al. Responses of N2O fluxes to temperature, water table and N deposition in a northern boreal fen. Eur J Soil Sci 2010;61: 651–661.
  11. 11. Maljanen M, Sigurdsson BD, Guomundsson J, Oskarsson H, Huttunen JT, Martikainen PJ. Greenhouse gas balances of managed peatlands in the nordic countries—present knowledge and gaps. Biogeosciences 2010;7: 2711–2738.
  12. 12. Martikainen PJ, Nykänen H, Crill P, Silvola J. Effect of a lowered water-table on nitrous-oxide fluxes from northern peatlands. Nature 1993;366: 51–53.
  13. 13. Marushchak M, Pitkämäki A, Koponen H, Biasi C, Seppälä M, Martikainen P. Hot-spots for nitrous oxide emissions found in different types of permafrost peatlands. Glob Change Biol 2011;17: 2601–2614.
  14. 14. Palmer K, Drake HL, Horn MA. Association of novel and highly diverse acid-tolerant denitrifiers with N2O fluxes of an acidic fen. Appl Environ Microb 2010;76: 1125–1134. pmid:20023077
  15. 15. Palmer K, Biasi C, Horn MA. Contrasting denitrifier communities relate to contrasting N2O emission patterns from acidic peat soils in arctic tundra. ISME J 2012;6: 1058–1077. pmid:22134649
  16. 16. Goldberg SD, Knorr KH, Gebauer G. N2O concentration and isotope signature along profiles provide deeper insight into the fate of N2O in soils. Isot Environ Healt S 2008;44: 377–391. pmid:19061068
  17. 17. Kolb S, Horn MA. Microbial CH4 and N2O consumption in acidic wetlands. Front Microbiol 2012;3: 78. pmid:22403579
  18. 18. Bremner JM. Sources of nitrous oxide in soils. Nutr Cycl Agroecosys 1997;49: 7–16.
  19. 19. Conrad R. Soil microorganisms as controllers of atmospheric trace gases (H2, CO, CH4, OCS, N2O, and NO). Microbiol Rev 1996;60: 609–640. pmid:8987358
  20. 20. Zumft WG. Cell biology and molecular basis of denitrification. Microbiol Mol Biol R 1997;61: 533–615. pmid:9409151
  21. 21. van Cleemput O. Subsoils: Chemo- and biological denitrification, N2O and N2 emissions. Nutr Cycl Agroecosyst 1998;52: 187–194.
  22. 22. Moore TR, Roulet NT, Waddington JM. Uncertainty in predicting the effect of climatic change on the carbon cycling of Canadian peatlands. Clim Change 1998;40: 229–245.
  23. 23. Palmer K, Horn MA. Actinobacterial nitrate reducers and Proteobacterial denitrifiers are abundant in N2O-metabolizing palsa peat. Appl Environ Microb 2012;78: 5584–5596. pmid:22660709
  24. 24. Yoshinari T, Knowles R. Acetylene inhibition of nitrous oxide reduction by denitrifying bacteria. Biochem Bioph Res Co 1976;69: 705–710.
  25. 25. Ratkowsky D. Nonlinear regression modeling: a unified and practical approach. 1st ed. New York: Marcel Dekker Inc.; 1983.
  26. 26. Persoh D, Theuerl S, Buscot F, Rambold G. Towards a universally adaptable method for quantitative extraction of high-purity nucleic acids from soil. J Microbiol Meth 2008;75: 19–24.
  27. 27. Philippot L, Piutti S, Martin-Laurent F, Hallet S, Germon JC. Molecular analysis of the nitrate-reducing community from unplanted and maize-planted soils. Appl Environ Microb 2002;68: 6121–6128. pmid:12450836
  28. 28. Throbäck IN, Enwall K, Jarvis A, Hallin S. Reassessing PCR primers targeting nirS, nirK and nosZ genes for community surveys of denitrifying bacteria with DGGE. FEMS Microbiol Ecol 2004;49: 401–417. pmid:19712290
  29. 29. Rich JJ, Heichen RS, Bottomley PJ, Cromack K, Myrold DD. Community composition and functioning of denitrifying bacteria from adjacent meadow and forest soils. Appl Environ Microb 2003;69: 5974–5982. pmid:14532052
  30. 30. Quince C, Lanzen A, Davenport RJ, Turnbaugh PJ. Removing noise from pyrosequenced amplicons. BMC Bioinformatics 2011;12: 38. pmid:21276213
  31. 31. Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, et al. Qiime allows analysis of high-throughput community sequencing data. Nat Methods 2010;7: 335–336. pmid:20383131
  32. 32. Palmer K, Drake HL, Horn MA. Genome-derived criteria for assigning environmental narG and nosZ sequences to operational taxonomic units of nitrate reducers. Appl Environ Microb 2009;75: 5170–5174. pmid:19502444
  33. 33. Kumar S, Nei M, Dudley J, Tamura K. MEGA: A biologistcentric software for evolutionary analysis of DNA and protein sequences. Brief Bioinform 2008;9: 299–306. pmid:18417537
  34. 34. Hughes JB, Hellmann JJ, Ricketts TH, Bohannan BJM. Counting the uncountable: statistical approaches to estimating microbial diversity. Appl Environ Microbiol 2001;67: 4399–4406. pmid:11571135
  35. 35. Zaprasis A, Liu YJ, Liu SJ, Drake HL, Horn MA. Abundance of novel and diverse tfdA-like genes, encoding putative phenoxyalkanoic acid herbicide-degrading dioxygenases, in soil. Appl Environ Microbiol 2010;76: 119–128. pmid:19880651
  36. 36. Eurola S, Hicks S, Kaakinen E. Key to Finnish mire types. In: Moore PD, editor. European mires. London: Academic Press Inc; 1984. pp. 11–118.
  37. 37. Lemke P, Ren J, Alley R, Allison I, Carrasco J, Flato G, et al. Observations: Changes in snow, ice and frozen ground. In: Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB, et al., editors. Climate Change 2007: The physical science basis. Contribution of working group I to the fourth assessment report of the intergovernmental panel on climate change. Cambridge: Cambridge University Press; 2007. pp. 337–283.
  38. 38. Blicher-Mathiesen G, Hoffmann C. Denitrification as a sink for dissolved nitrous oxide in a freshwater riparian fen. J Environ Qual 1999;28: 257–262.
  39. 39. Chapuis-Lardy L, Wrage N, Metay A, Chotte JL, Bernoux M. Soils, a sink for N2O? A review. Glob Change Biol 2007;13: 1–17.
  40. 40. Roobroeck D, Butterbach-Bahl K, Brueggemann N, Boeckx P. Dinitrogen and nitrous oxide exchanges from an undrained monolith fen: shortterm responses following nitrate addition. Eur J Soil Sci 2010;61: 662–670.
  41. 41. Zumft WG, Kroneck PMH. Respiratory transformation of nitrous oxide (N2O) to dinitrogen by Bacteria and Archaea. Adv Microb Physiol 2007;52: 107–227. pmid:17027372
  42. 42. Kirchman DL, Meon B, Ducklow HW, Carlson CA, Hansell DA, Steward GF. Glucose fluxes and concentrations of dissolved combined neutral sugars (polysaccharides) in the Ross Sea and Polar Front Zone, Antarctica. Deep Sea Res. Part II Top Stud Oceanogr 2001; 48: 4179–4197.
  43. 43. Mahne I, Tiedje JM. Criteria and methodology for identifying respiratory denitrifiers. Appl Environ Microbiol 1995;61: 1110–1115. pmid:16534960
  44. 44. Jahangir MMR, Khalil MI, Johnston P, Cardenas LM, Hatch DJ, Butler M, et al. Denitrification potential in subsoils: A mechanism to reduce nitrate leaching to groundwater. Agric Ecosyst Environ 2012;147: 13–23.
  45. 45. Almeida JS, Reis MAM, Carrondo MJT. Competition between nitrate and nitrite reduction in denitrification by Pseudomonas fluorescens. Biotechnol Bioeng 1995;46: 476–484. pmid:18623340
  46. 46. Henry S, Bru D, Stres B, Hallet S, Philippot L. Quantitative detection of the nosZ gene, encoding nitrous oxide reductase, and comparison of the abundances of 16S rRNA, narG, nirK, and nosZ genes in soils. Appl Environ Microbiol 2006;72: 5181–5189. pmid:16885263
  47. 47. Sanford RA, Wagner DD, Wu Q, Chee-Sanford JC, Thomas SH, Cruz-García C, et al. Unexpected nondenitrifier nitrous oxide reductase gene diversity and abundance in soils. Proc Nat Acad Sci USA 2012;109: 19709–19714. pmid:23150571
  48. 48. Jones CM, Graf DRH, Bru D, Philippot L, Hallin S. The unaccounted yet abundant nitrous oxide-reducing microbial community: a potential nitrous oxide sink. ISMEJ 2013;7: 417–426. pmid:23151640
  49. 49. Orellana LH, Rodriguez-R LM, Higgins S, Chee-Sanford JC, Sanford RA, Ritalahti KM, et al. Detecting nitrous oxide reductase (nosZ) genes in soil metagenomes: Method development and implications for the nitrogen cycle. mBio 2014;5: e01193–14. pmid:24895307
  50. 50. Braker G, Conrad R. Diversity, structure, and size of N2O-producing microbial communities in soils-what matters for their functioning? Adv Appl Microbiol 2011;75: 33–70. pmid:21807245
  51. 51. Bru D, Ramette A, Saby NPA, Dequiet S, Ranjard L, Jolivet C, et al. Determinants of the distribution of nitrogen-cycling microbial communities at the landscape scale. ISME J 2011;5: 532–542. pmid:20703315
  52. 52. Cuhel J, Simek M, Laughlin RJ, Bru D, Cheneby D, Watson CJ, et al. Insights into the effect of soil pH on N2O and N2 emissions and denitrifier community size and activity. Appl Environ Microbiol 2010;76: 1870–1878. pmid:20118356
  53. 53. Männistö MK, Häggblom MM. Characterization of psychrotolerant heterotrophic bacteria from Finnish Lapland. Syst Appl Microbiol 2006;29: 229–243. pmid:16564959
  54. 54. Zenova GM, Manucharova NA, Zvyagintsev DG. Extremophilic and extremotolerant actinomycetes in different soil types. Eurasian Soil Sci 2011;44: 417–436.
  55. 55. Whitman WB, Goodfellow M, Kämpfer P, Busse HJ, Trujillo ME, Suzuki KI, et al. Bergey's manual of systematic bacteriology: Volume 5—The Actinobacteria. 2 nd ed. New York: Springer; 2012.
  56. 56. Fierer N, Jackson RB. The diversity and biogeography of soil bacterial communities. Proc Nat Acad Sci USA 2006;103: 626–631. pmid:16407148
  57. 57. Jones CM, Welsh A, Throback IN, Dorsch P, Bakken LR, Hallin S. Phenotypic and genotypic heterogeneity among closely related soil-borne N2- and N2O-producing Bacillus isolates harboring the nosZ gene. FEMS Microbiol Ecol 2011;76: 541–552. pmid:21348884
  58. 58. Demanèche S, Philippot L, David MM, Navarro E, Vogel TM, Simonet P. Characterization of denitrification gene clusters of soil bacteria via a metagenomic approach. Appl Environ Microbiol 2009;75: 534–537. pmid:19011059
  59. 59. Singh BK, Campbell CD, Sorenson SJ, Zhou J. Soil genomics. Nat Rev Micro 2009;7: 756–756.