Temporal Changes in Methane Oxidizing and Denitrifying Communities and Their Activities in a Drained Peat Soil

1 23

WetlandsOfficial Scholarly Journal of the Societyof Wetland Scientists ISSN 0277-5212Volume 32Number 6 Wetlands (2012) 32:1047-1055DOI 10.1007/s13157-012-0335-3

Temporal Changes in Methane Oxidizingand Denitrifying Communities and TheirActivities in a Drained Peat Soil

Janet Andert, Gunnar Börjesson & SaraHallin

1 23

Your article is protected by copyright and

all rights are held exclusively by Society

of Wetland Scientists. This e-offprint is for

personal use only and shall not be self-

archived in electronic repositories. If you

wish to self-archive your work, please use the

accepted author’s version for posting to your

own website or your institution’s repository.

You may further deposit the accepted author’s

version on a funder’s repository at a funder’s

request, provided it is not made publicly

available until 12 months after publication.

ARTICLE

Temporal Changes in Methane Oxidizingand Denitrifying Communities and Their Activitiesin a Drained Peat Soil

Janet Andert & Gunnar Börjesson & Sara Hallin

Received: 30 June 2011 /Accepted: 30 August 2012 /Published online: 16 September 2012# Society of Wetland Scientists 2012

Abstract Northern peat soils are both sources and sinks of thegreenhouse gases nitrous oxide (N2O) and methane (CH4).These fluxes are regulated by the activity of microbial com-munities. Temporal, spatial and depth related differences inpotential activity and composition of the denitrifier and meth-anotrophic bacterial communities were compared betweenwinter and summer in relation to in situ fluxes of N2O andCH4 from a drained and forested peat land in Sweden. Thecomposition of the genetic pool of these guilds was assessedusing terminal fragment length polymorphism analysis of sig-nature genes in the denitrification and methane oxidation path-ways, respectively. The composition of the communities wassimilar throughout the soil profile, although both denitrificationand methane oxidation rates decreased with soil depth. Thepotential methane oxidation rates were significantly correlatedto soil N content,NHþ

4 andNO�3 , indicating a nitrogen induced

stimulation of methane oxidation capacity. The site was asource of N2O and a small sink for CH4, and although thefluxes did not vary significantly over time, both the methaneoxidizing and denitrifier community composition exhibitedtemporal patterns. Differences in the composition of the meth-anotrophic community reflected that the peat site shifted from asummer to a winter community. The denitrifiers exhibited asimilar trend, but also differed significantly between the sam-pling occasions during summer, as well as between locations atthe site. The latter was explained by differences in soil pH. Itwas only the differences in the methane oxidizer communitycomposition that correlated with its corresponding potentialactivity, which implies a higher degree of functional redundan-cy within the denitrifier community.

Keywords Denitrification . GHG emissions . Methaneoxidation . Nitrous oxide . Organic soil

Introduction

Peat soils are common in the Northern hemisphere andcomprise 3 % (346 million hectare; Gorham 1991) of theEarth’s terrestrial ecosystems. They play a significant role inglobal climate change due to the emissions and uptake of thegreenhouse gases carbon dioxide, nitrous oxide and meth-ane. In addition, nitrous oxide is designated as the currentdominant ozone depleting substance (Ravishankara et al.2009). Natural peat soils are a source of methane and a sinkfor carbon dioxide and can even be small sinks for nitrousoxide (Regina et al. 1996), but when drained for peat ex-traction, forestry or agriculture, decomposition of the storedorganic matter accelerates turning these soils to a majorcarbon dioxide and nitrous oxide source. Nevertheless, aer-ated soils are important global sinks for atmospheric meth-ane (LeMer and Roger 2001) and drained peat soils have the

Electronic supplementary material The online version of this article(doi:10.1007/s13157-012-0335-3) contains supplementary material,which is available to authorized users.

J. Andert : S. Hallin (*)Department of Microbiology,Swedish University of Agricultural Sciences,750 07 Uppsala, Swedene-mail: [email protected]

J. AndertMax-Planck-Institute of Colloids and Interfaces,Potsdam, Germany

G. BörjessonDepartment of Soil and Environment,Swedish University of Agricultural Sciences,750 07 Uppsala, Sweden

Present Address:J. AndertLeibniz Institute for Agricultural Engineering,Max-Eyth-Allee 100,14469 Potsdam, Germany

Wetlands (2012) 32:1047–1055DOI 10.1007/s13157-012-0335-3

Author's personal copy

potential to consume methane (Martikainen et al. 1995; Lai2009). Emissions of methane from peatlands are normallyhighest in summer (Nedwell and Watson 1995; Nykanen etal. 1995; Alm et al. 1999), but nitrous oxide emissions occurirregularly in time because of weather conditions and thelowest emissions are often reported during summer (Philippotet al. 2007 and references therein).

Methane emissions from soil are ultimately regulated by theproduction of methane by the methanogenic archaea and con-sumption by the aerobic methane oxidizing bacteria (MOB).Anaerobic oxidation of methane with nitrite as terminal elec-tron acceptor (Ettwig et al. 2010) or by an association ofmethanotrophic archaea and sulfate-reducing bacteria is alsopossible (Knittel and Boetius 2009), but most likely masked byaerobic methane oxidation in peatlands (Smemo and Yavitt2011). Nearly all MOB are found within the alpha and gammasubdivisions of Proteobacteria, although a few acidophilictypes are affiliated to the Verrucomicrobia. The majority ofthe MOB can only use methane as electron donor, but a fewfacultative methanotrophs using acetate or methanol are foundwithin the alphaproteobacteria (Dedysh et al. 2005; Im et al.2011), which has been reported as a survival strategy of peat-inhabiting Methylocystis spp. (Belova et al. 2011). Among theMOB, some are high-affinity methanotrophs and since theyoxidize methane at low concentrations these organisms areimportant atmospheric methane sinks (Bull et al. 2000). Well-drained peat soils usually have lowmethane concentrations, butboth low- and high-affinity methanotrophs can be present (Bullet al. 2000). The key enzyme in methane oxidation amongobligate methanotrophs is the membrane bound methanemonooxygenase and the pmoA gene coding for this enzymeis the most commonly used molecular marker for the detectionof MOB (Murrell et al. 1998). Nitrous oxide emissions fromsoils are mainly controlled by denitrification and the net resultis the balance between production and reduction of nitrousoxide by denitrifiers, although the microbial processes nitrifi-cation and dissimilatory nitrate reduction to ammonium alsohave the capacity to produce nitrous oxide (Robertson andTiedje 1987; Conrad 1996). Denitrification is an anaerobicrespiration pathway during which soluble NO�

3 and NO�2 are

sequentially reduced to the nitrogen gases NO, N2O and N2 viafour enzymatic complexes by a wide range of unrelated taxaacross the bacterial and archaeal domains. The genes nirK andnirS coding for the two different types of nitrite reductases arethe most common molecular markers and denitrifiers possessone or the other, whereas other genes in the pathway are eithernot unique for denitrification or are not present in all denitrifiers(Jones et al. 2008).

Methane oxidation and the N2O:N2 ratio of denitrificationend products can be influenced by several factors, includingsoil and climatic conditions (e.g. Firestone et al. 1980; LeMerand Roger 2001). Apart from the abiotic factors affectingnitrous oxide and methane emissions, our aim was to

determine if temporal dynamics of the underpinning commu-nities that are involved in emissions of these greenhouse gasesneed to be considered to establish long-term mitigation strat-egies for drained peat soils. For that purpose, we monitoredtemporal fluctuations of the denitrifier and MOB communitystructures and the potential activity of the denitrifying andhigh-affinity methane oxidizing communities, as they are thekey groups for nitrous oxide reduction and act as methanesinks, respectively. The site we sampled is a drained andforested peat land, which had previously been reported astrong net sink for methane (Weslien et al. 2009). Toconfirm that and estimate nitrous oxide emissions, in situfluxes were determined. Sampling was done at four occa-sions over a nine-month period, with two samplings eachduring the summer and winter season, and spatial differencesin the distribution of the functional groups and potentialactivities were considered by examining three different depthsin the upper 50 cm soil layer at five different locations at eachsampling occasion.

Materials and Methods

Sampling Site and Soil Properties

The sampling site is a forested peat soil situated in the South-west of Sweden (58°20′N; 13°30′ E). The site was drained inthe 1920’s for peat extraction and in the 1960’s planted withsilver birch trees (Betula pendula). The average peat depth is2.2 m with 80 % organic matter. Five replicated locations,designated with the letters A to E, with 3 m in between weresampled at four occasions distributed over a 9 months period(13May 13, August 21, and November 19, in 2008 andMarch17, 2009), with May and Augusts representing the summerseason and November and March representing the winterseason. On all occasions, soil was sampled at three differentdepths (0–15 cm, 15–30 cm, 30–45 cm). At the sampling inNovember, the soil was frozen throughout the profile.

Directly after sampling, water content (overnight drying at105 °C), organic matter content (weight loss after ignition at550 °C for 4 h) and pH(H2O) were determined. Total organicnitrogen (TON), total organic carbon (TOC), NHþ

4 � N andNO�

3 � N were determined by AgriLab (Uppsala, Sweden).The soil parameters are presented in Table S1 in Supplemen-tary Material. The soil samples were stored at 2 °C beforemeasuring of potential methane oxidation and denitrificationactivities, and soil for DNA extraction was frozen until use.

Flux Measurements and Determination of PotentialActivities

At each occasion, net fluxes of N2O and CH4 were mea-sured at each location at the site during 2 h using closed

1048 Wetlands (2012) 32:1047–1055

Author's personal copy

chambers according to Weslien et al. (2009). Partial pres-sure of N2O and CH4 was plotted against time and fittedregression lines with R2<0.5 were treated as no exchangebetween soil and atmosphere. Triplicate samples fromeach location were used to determine substrate-inducedpotential denitrification rates using the acetylene inhibitiontechnique according to Pell et al. (1996) with some mod-ifications. Soil slurries with 10 g of soil and 20 mLdistilled water in 140-mL glass flasks with the additionof 3 mM KNO3 and glucose, respectively, were incubatedat 15 °C for 2 h with agitation (180 rpm). Gas sampleswere taken every 15 min. Duplicate samples from eachlocation were used to determine substrate-induced poten-tial CH4 oxidation rates. A slurry of 10 g peat soil and20 mL distilled water in 140 mL glass flasks and 30 ppm(v) methane in the headspace was incubated at 15 °C withagitation (180 rpm). Subsamples of gas were withdrawnduring 48 h in regular time intervals. Methane and N2O wereanalyzed using a gas chromatograph (Perkin Elmer Cla-rus500, USA) using a flame ionization and electron-capturedetector, respectively.

DNA Extraction, PCR Amplification and T-RFLP Analysis

For community analysis, a pooled sample from each lo-cation at each of the sampling locations was used andDNA was extracted from 150 mg dry peat according toStralis-Pavese et al. (2004) without potassium acetate ad-dition. The nirK and nirS genes, coding for the two typesof nitrite reductases were used to target the denitrifiers andwere amplified using the primers and conditions describedby Throbäck et al. (2004) with the forward primers hexa-chlorofluorescein (hex) labelled (Table 1). The pmoAgene, coding for a subunit of the particulate methanemonooxygenase, was used to target the obligate MOBand amplified by a semi-nested approach using two serialprimer combinations (Stralis-Pavese et al. 2004) with theforward primer in the second PCR hex-labelled (Table 1).No nirK amplicons were obtained from the November andMarch samples, and the subsequent T-RFLP analysis wastherefore discontinued for the May and August samples.For the nirS and pmoA genes, triplicate PCR products

from each sample were pooled, purified using QiagenMinElute and digested (50 ng DNA) with HaeIII for thenirS gene and MspI for the pmoA gene. Fluorescentlylabeled TRFs were separated and detected using an ABI3730 capillary sequencer (Applied Biosystems, USA) runat 15 kV for 40 min. Before injection, 1.4 μL of thesample was denatured in the presence of 10 μL formam-ide and 0.05 μL GS-500 ROX size standard (AppliedBiosystems). The terminal restriction fragment patternswere evaluated according to Enwall and Hallin (2009)using the Peak Scanner Software (Applied Biosystems)and TRFs <50 bp or contributing with <0.5 % of the totalTRF-signal were excluded.

Statistical Analysis

Differences in soil parameters, potential denitrification andmethane oxidation rates, as well as fluxes of N2O andCH4 were compared by ANOVA at P<0.05 followed byTukey-Kramer HSD post-hoc test. Pearson correlationsbetween potential activities, fluxes and soil parameterswere determined. Data that were not normally distributedwere Box-Cox transformed prior to analysis. All analyseswere performed using the Jmp 6 software (SAS, NorthCarolina, USA).

For graphical representation of community relation-ships among samples, T-RFLP analyses of nirS andpmoA were analyzed by non-metric multidimensionalscaling (NMS). Data matrices of community fingerprintswere arc-sinus square root transformed and the Bray-Curtis distance measure was used to generate dissimilar-ity matrices. In addition, a data matrix based on pres-ence/absence was also tested and gave similar results(results not shown). The NMS was run using a randomstarting configuration, a maximum of 500 iterations andan instability criterion of 0.00001 and performed on 250runs with the real data followed by 250 runs with ran-dom data. The soil parameters were combined into asecondary matrix, relativized by column totals and incor-porated into the analysis through the use of bi-plot ordi-nations, where variables were correlated with the NMSaxes. Multi-Response Permutation Procedures (MRPP)were used to test whether the a priori groups time,location or soil depth differed within the ordinations.Pairwise comparisons were made to calculate p valuesfor post-hoc comparisons between all groups. To evaluatethe correspondence between dissimilarity matrices for thedenitrifier and methane oxidizer community structures,potential activities and emissions of N2O and CH4, respec-tively, Mantel’s test with a Monte Carlo simulation with 999randomizations was conducted. All the multivariate analyseswere performed using PC-ORD version 5.10 (MjM Software,Oregon, USA).

Table 1 Primers used for the amplification of the functional genes

Target gene Forwardprimer

Reverseprimer

References

nirK FlaCu-hex R3Cu (Throbäck et al. 2004)

nirS Cd3aF-hex R3cd (Throbäck et al. 2004)

pmoA (nested 1) A189 A682 (Holmes et al. 1995)

pmoA (nested 2) A189-hex mb661 (Costello andLidstrom 1999)

Wetlands (2012) 32:1047–1055 1049

Author's personal copy

Results and Discussion

The site was a source of N2O (0.3–40.6 μmol N2O m−2h−1)and a small sink for CH4 (−2.1–6.2 μmol CH4 m−2h−1;Figs. 1 and 2). This is in agreement with previous workshowing that drainage of peat soils can reduce methaneemissions, which in general is accompanied by an increasein nitrous oxide emissions (Martikainen et al. 1995; Weslienet al. 2009). Furthermore, the emission rates of both gaseswere in the range of what has been reported for other Nordicpeat soils (Nykanen et al. 1995; Maljanen et al. 2001;Regina et al. 2004). There were minor temporal differencesin net methane fluxes and nitrous oxide emissions at the site,but these were not significantly different between seasons or

sampling occasions. Low temperatures have been shown tolimit the activity of the nitrous oxide reductase more thanthe other enzymes in the denitrification pathway, but sincedenitrification rates can decrease when the temperaturedecreases the emissions may still be low (Holtan-Hartwiget al. 2002). However, because the substrate-induced poten-tial activities reflect past events that have determined growthof denitrifiers, temperature at the time of sampling does notnecessarily have an impact on the potential denitrificationrate measured at that occasion. In agreement the N2O emis-sions, the potential denitrification rates did not vary withsampling time, with the exception of the middle soil layerthat displayed significantly higher rates in November(Fig. 1). Nevertheless, in contrast to the rather stable in situ

0

5

10

15

20

25

0

20

40

60

80

100

120

140

160

180

A B C D E A - E

Pot

entia

l den

itrifi

catio

n (µ

mol

N2O

-N m

-3dr

y so

il h-

1 )E

mis

sion

(µm

ol N

2O-N

m-2

h-1)

0-15 cm

15-30 cm

30-45 cm

Emission

0

5

10

15

20

25

0

20

40

60

80

100

120

140

160

180

A B C D E A - E

Sampling location

0

5

10

15

20

25

0

20

40

60

80

100

120

140

160

180

A B C D E A - E

Pot

entia

l den

itrifi

catio

n (µ

mol

N2O

-N m

-3dr

y so

il h-

1 )E

mis

sion

(µm

ol N

2O-N

m-2

h-1 )

Sampling location

0

20

40

60

80

100

120

140

160

180

A B C D E A - E

0

5

10

15

20

25

a

b

a

ab

a

a

a

b

c) November

b) Augusta) May

b

a

ab

d) March

b

Fig. 1 Nitrous oxide emission(▲) and potential denitrificationrates (bars) from samplinglocations A–E in May (a),August (b), November (c) andMarch (d). The black, grey andwhite bars indicate averagepotential rates at sampling depths0–15, 15–30, and 30–45 cm,respectively (±SD; n03). Theoverall emissions and potentialrates at each sampling occasionrepresented as the mean of thefive locations A–E are shown inthe shaded areas on the left-handside in each graph. Different let-ters above the bars refer to sig-nificant differences in potentialactivity for each soil layer amongsampling dates (P<0.05). Therewere no temporal differences inthe N2O emissions

1050 Wetlands (2012) 32:1047–1055

Author's personal copy

net uptake of methane at the site, the potential methaneoxidation activity varied with time (Fig. 2). The rates inMay and August were similar and significantly differentfrom those in November and March. These in turn differedsignificantly from each other with higher rates observed inMarch and the lowest rates in November.

Emissions of nitrous oxide varied across the site on allsampling occasions and correlated significantly with pH(Fig. 1, Table 2). Site A was the most acidic soil with a pHbelow 4.8 at all occasions (Table S1 in Supplementary Mate-rial) and the highest emissions were observed at this locationat all sampling occasions. This was also the only site emittingCH4, but only in winter (Fig. 2). Although denitrification

generally decreases with decreasing pH, it has been estab-lished that pH is negatively correlated with N2O:N2 productratios during denitrification (e.g. Simek and Cooper 2002;Dannenmann et al. 2008; Jones et al. 2011). This is mostlikely due to inhibition of the nitrous oxide reductase(Thomsen et al. 1994; Kesik et al. 2006). Interestingly, most“outliers” in the nirS denitrifier ordination corresponded tosamples from the acidic location (A) and the MRPP andpairwise comparisons of the distance matrices of the locationsverified that the denitrifier community was significantly dif-ferent from the nirS communities at the other locations. ThepH vector in the NMS bi-plot was directed away from thesesamples indicating that pHwas the overriding factor imposing

0

1

2

3

4

5

6

7

A B C D E A-E

-3

-2

-1

0

1

2

3

4

5

6

7

b

ab

b

b) August

0

1

2

3

4

5

6

7

A B C D E A-E

Sampling locations

-3

-2

-1

0

1

2

3

4

5

6

7

a

a

a

d) March

0

1

2

3

4

5

6

7

A B C D E A-E

Pot

entia

l met

hane

oxi

datio

n (n

mol

CH

4 g-1

dry

soi

l h-1

)E

mis

sion

(µm

ol C

H4 m

-2 h

-1)

-3

-2

-1

0

1

2

3

4

5

6

7

0-15 cm

15-30 cm

30-45 cm

Emission

a) May

bcb

b

0

1

2

3

4

5

6

7

A B C D E A-E

Sampling locations

Pot

entia

l met

hane

oxi

datio

n (n

mol

CH

4 g-1

dry

soi

l h

-1)

Em

issi

on (

µmol

CH

4 m-2

h-1

)

-3

-2

-1

0

1

2

3

4

5

6

7c) November

c

cb

Fig. 2 Methane emission (▲)and potential methane oxidationrates (bars) from samplinglocations A–E in May (a),August (b), November (c) andMarch (d). The black, grey andwhite bars indicate averagepotential rates at samplingdepths 0–15, 15–30, and 30–45 cm, respectively (n02) . Theoverall emissions and potentialrates at each sampling occasionrepresented as the mean of thefive locations A–E are shown inthe shaded areas on the left-hand side in each graph. Dif-ferent letters above the barsrefer to significant differencesin potential activity amongsampling dates (P<0 05). Therewere no temporal differences inthe methane emissions

Wetlands (2012) 32:1047–1055 1051

Author's personal copy

differences in community composition (Fig. 3a). No other soilfactor correlated with the NMS axes. Soil pH has previouslybeen shown to have a strong effect on the denitrifier commu-nity composition in arable soil (Enwall et al. 2005) and wassuggested to predict bacterial diversity at the continental scale(Lauber et al. 2009). In contrast to the denitrifiers, the MOBcommunities did not display location related differences with-in the site and none of the measured soil factors correlatedwith the NMS axes in the MOB ordination.

In contrast to the nitrous oxide emissions and potentialdenitrification activity, temporal patterns were observed forthe nirS denitrifier community composition, which wasverified by the MRPP (P<0.001). The pairwise comparisonsof the distance matrices of the seasons showed that thecomposition was significantly different in August compared

to the other sampling occasions (P<0.001; Fig. 3a). Withinthe mixed cluster in the ordination, the May samples formeda sub-cluster that was significantly different from the March(P<0.001) samples, but not from those sampled in Novem-ber. November and March formed separate clusters thatdiffered significantly from each other (P<0.05). Temporalvariations in the denitrifier community are further under-lined by the fact that the nirK-type denitrifiers could not bedetected in the two winter samples, but were readily ampli-fied in the May and August samples. Also the compositionof the methane oxidizing community exhibited temporalpatterns, as verified by the MRPP test (P<0.001) and pair-wise post-hoc comparisons. The pmoA communities at thesite were significantly different in May and August com-pared to those detected in November and March (P<0.001;

Table 2 Pearson correlations between soil parameters, potential denitrification rates (PDA), potential methane oxidation rates (PMO), and in situ emissionsof methane and nitrous oxide. Means of each locations and soil depth from each sampling occasion were included. Significant correlations in bold

pH Water TOCa TONb NO�3 NHþ

4 C:N CH4 PDA PMO N2O

Correlation coeeficient (r) pH NS NS NS NS ** NS NS ** ** *** Significance (P)Water 0.21 *** ** NS NS *** NS *** NS NS

TOC 0.11 0.75 NS NS ** *** NS *** * NS

TON 0.01 0.27 0.16 NS NS *** ** NS *** NS

NO�3 0.23 −0.18 −0.04 0.04 NS NS NS NS ** **

NHþ4 0.41 −0.13 −0.28 0.24 0.05 ** NS * *** *

CN 0.09 0.52 0.81 −0.45 −0.08 −0.38 ** *** *** NS

CH4 −0.02 0.13 −0.12 0.35 −0.25 0.04 −0.33 NS NS *

PDA −0.28 −0.73 −0.82 −0.10 −0.03 0.25 −0.69 0.068 NS NS

PMO 0.36 −0.19 −0.30 0.46 0.33 0.65 −0.54 −0.05 0.18 NS

N2O −0.58 0.05 −0.05 0.23 −0.37 −0.32 −0.18 0.27 0.08 −0.25

a Total organic carbon, b Total organic nitrogen, *** P≤0.001, ** 0.001<P≤0.01, * 0.01<P≤0.05, NS not significant at P>0.05

NMS axis 1

B

B

C CCDD DE

E

A

AA

B

B

BC

C

C

D

DD

E

EEA

A

B B

B

C

CC

D

D

D

E

E

E

A

B

B

C

C

C

DDEEE

pH

NM

S a

xis

2

S = 12.6R2 = 0.82

a) nirS b) pmoA

A

A A

BB

B

CC

CD

DD EE E

A

A

B

B

B

C

C

C

D

D

D

EE

E

A

B

BB

C

C

D

D

D

E

D

B

B

B

C CC

D D

D

E

E

E

S = 12.9R2 = 0.90

Fig. 3 Non-metric multidimensional scaling ordination plots from T-RFLP fingerprints of the nirS gene of the denitrifiers (a) and the pmoAgene of the methane oxidizers (b) from sampling locations A–E anddifferent depths in May 08 (▲), August 08 (●), November 08 (Δ) and

March 09 (○). The ellipses indicate the most significant temporaldifferences among the communities (P<0.001). Stress values (S) andcoefficients of determination for the correlations between ordinationdistances and distances in the original space are indicated (P<0 001)

1052 Wetlands (2012) 32:1047–1055

Author's personal copy

Fig. 3b), but those in May and August or November andMarch were not significantly different from each other. Thissuggests that a shift from a summer to a winter MOBcommunity occurred at the site. These temporal differenceswere weakly but significantly correlated with differences inpotential methane oxidation capacity observed across seasons(P<0.05; Fig. 2). These results suggest that the MOB com-munity composition matters for potential functioning, whichis in agreement with the findings by Mohanty et al. (2006).The only temporal changes in soil properties were the NHþ

4

content that was significantly higher at the sampling occasionin March at all depths and the NO�

3 content that was signif-icantly lower at this occasion throughout the soil profile thanat the other sampling times (P<0.05), but these factors werenot related to the shifts in community composition of neitherthe nirS denitrifiers nor the MOB. That temporal changes inthe methane oxidizing community, but not in the denitrifiercommunity correlated with changes in the corresponding po-tential rates may indicate a higher degree of functional redun-dancy within the denitrifier community as compared to themethane oxidizers. The hypothesis of functional redundancyclaims that when many species have similar features, biodi-versity loss or community shifts does not have a major impacton ecosystem functioning (Naeem 1998; Walker 1992). Func-tional redundancy within the denitrifier community could beexplained by the fact that denitrification is a trait found amongmany bacterial genera spread over distantly related phyla(Philippot et al. 2007; Jones et al. 2008) covering variouslifestyle features, whereas aerobic oxidation of methane bynon-thermophilic methanotrophs is mainly occurring withinmonophyletic lineages of the alpha and gamma subdivisionsof the Proteobacteria (Bowman 2006). By only looking atcommunity structure based on T-RFLP, the identity of thedenitrifier and methane oxidizer community members is notknown. For the MOB when comparing with other studies,where peaks were verified by clone libraries and generatedwith the same restriction enzymes as in our study, the threemajor TRFs of the pmoA gene profiles in the present study areshared (Horz et al. 2001; Hoffmann et al. 2002; Horz et al.2005). One of the major pmoA peaks (447) can potentially beassigned to Type I methanotrophs and the others (245, 350) toType II, but care must be taken when interpreting TRFsby comparisons with in silico digestions from sequencesretrieved elsewhere.

There were no differences in community composition forany of the two functional groups according to soil depth,although the potential denitrification rates decreased rapidlywith depth (Fig. 1). The potential methane oxidation rateswere also significantly lower in the deepest layer comparedto the two upper layers (P<0.05; Fig. 2). These resultsindicate that environmental parameters and/or the commu-nity abundance, rather than the community compositionregulated the potential denitrification and methane oxidation

activities throughout the soil profile. However, the lack ofcorrelation can also be due to that a DNA-based approachwas used to estimate community composition as the compo-sition is the result of events occurring earlier than eventsaffecting increases or decreases in substrate-induced potentialactivities.Water content, pH, TOC, organic matter content andC:N ratio increased with depth at all sampling occasions(Table S1). The NHþ

4 concentration decreased with depth,with the 0–15 cm layer being significantly different from the30–45 cm layer (P<0.05). By contrast, the TON contentremained the same at all occasions and depths, and the NO�

3

did not show any consistent pattern with depth. Effects of thevarious soil parameters on denitrification activity were diffi-cult to interpret, e.g. the negative correlations with TOC andsoil moisture (Table 2), due to the strong autocorrelations withdepth, and other soil factors not determined here were likelylimiting denitrification potential in the deeper soil layers. Thedecrease in denitrification rates with soil depth could bebecause of a decrease in microbial biomass, which generallyoccurs in soil by depth (Blume et al. 2002; Fierer and Jackson2006). However, Palmer et al. (2010) did not find a decrease incell numbers of denitrifiers in an acidic fen even thoughactivity decreased with increasing depth, as in our study.

Interestingly, the potential methane oxidation rates of pre-sumably high-affinity methanotrophs (as suggested by ourexperimental conditions) correlated significantly to the totalorganic soil N content, C:N ratio and NHþ

4 concentration(Table 2). Further, the high oxidation potential in the Marchsamples coincided with high soil NHþ

4 concentrations (Fig. 2;Table S1). These findings suggest that nitrogen, especially NHþ

4 content plays a role for high-affinity methane oxidation inthis peat soil. By contrast, others have reported a stimulation ofmethane oxidation by low-affinity methane oxidizers in thepresence of NHþ

4 (Bodelier et al. 2000; Bodelier andLaanbroek 2004) and elevated NHþ

4 concentrations haveshown to suppress the oxidation of atmospheric methane(Gulledge and Schimel 1998). However, for atmosphericmethane oxidation in combination with other energy sources,like methanol or acetate, NHþ

4 might be beneficial and en-hance methane consumption (Bodelier and Laanbroek 2004).Organic acids in peat soils could potentially play a role here.Ammonium has been shown to have differential effects onMOB, with Type I methanotrophs favored by NHþ

4 , both inforest and rice paddy soils, whereas the Type II methanotrophswere suppressed (Mohanty et al. 2006).

To conclude, although the net fluxes of nitrous oxide andmethane did not vary significantly over time, the microbialcommunities that have the capacity to control and reducethese emissions demonstrated temporal patterns. That theseshifts over time were not reflected in the emission ratescould be due to the sampling frequency as significant peakemissions could have been overlooked. It was only the

Wetlands (2012) 32:1047–1055 1053

Author's personal copy

differences in the composition of the methanotrophic commu-nity that could be linked to differences of the correspondingpotential activities, which indicates a higher degree of func-tional redundancy in the denitrifier community.

Acknowledgements The project was supported by the Swedish Uni-versity of Agricultural Sciences (SLU), The Research Council Formas(No. 2005-246), and the European Science Foundation (METHECO).

References

Alm J, Saarnio S, Nykänen H, Silvola J, Martikainen PJ (1999) WinterCO2, CH4 and N2O fluxes on some natural and drained borealpeatlands. Biogeochemistry 44:163–186

Belova SE, Baani M, Suzina NE, Bodelier PLE, Liesack W, DedyshSN (2011) Acetate utilization as a survival strategy of peat-inhabiting Methylocystis spp. Environ Microbiol 3:36–46

Blume E, Bischoff M, Reichert JM, Moorman T, Konopka A, TurcoRF (2002) Surface and subsurface microbial biomass, communitystructure and metabolic activity as a function of soil depth andseason. Appl Soil Ecol 20:171–181

Bodelier PLE, Laanbroek HJ (2004) Nitrogen as a regulatory factor ofmethane oxidation in soils and sediments. FEMS Microbiol Ecol47:265–277

Bodelier PLE, Roslev P, Henckel T, Frenzel P (2000) Stimulation byammonium-based fertilizers of methane oxidation in soil aroundrice roots. Nature 403:421–424

Bowman J (2006) The methylotrophs — The families Methylococca-ceae and Methylocystaceae. In: Dworkin M et al (eds) The pro-karyotes. Springer, New York, pp 266–289

Bull ID, Parekh NR, Hall GH, Ineson P, Evershed RP (2000) Detectionand classification of atmospheric methane oxidizing bacteria insoil. Nature 405:175–178

Conrad R (1996) Soil microorganisms as controllers of atmospherictrace gases (H2, CO, CH4, OCS, N2O, and NO). Microbiol Rev60:609–640

Costello AM, Lidstrom ME (1999) Molecular characterization of func-tional and phylogenetic genes from natural populations of meth-anotrophs in lake sediments. Appl Environ Microbiol 65:5066–5074

Dannenmann M, Butterbach-Bahl K, Gasche R, Willibald G, Papen H(2008) Dinitrogen emissions and the N2:N2O emission ratio of aRendzic Leptosol as influenced by pH and forest thinning. SoilBiol Biochem 40:2317–2323

Dedysh SN, Knief C, Dunfield PF (2005) Methylocella species arefacultatively methanotrophic. J Bacteriol 187:4665–4670

Enwall K, Hallin S (2009) Comparison of T-RFLP and DGGE techni-ques to assess denitrifier community composition in soil. LettAppl Microbiol 48:145–148

Enwall K, Philippot L, Hallin S (2005) Activity and composition of thedenitrifying community respond differently to long-term fertiliza-tion. Appl Environ Microbiol 71:8335–8343

Ettwig KF, Butler MK et al (2010) Nitrite-driven anaerobic methaneoxidation by oxygenic bacteria. Nature 464:543–548

Fierer N, Jackson RB (2006) The diversity and biogeography of soilbacterial communities. Proc Natl Acad Sci U S A 103:626–631

Firestone MK, Firestone RB, Tiedje JM (1980) Nitrous oxide from soildenitrification: factors controlling its biological production. Sci-ence 208:749–751

Gorham E (1991) Northern Peatlands: role in the carbon cycle andprobable responses to climatic warming. Ecol Appl 1:182–195

Gulledge J, Schimel JP (1998) Low-concentration kinetics of atmo-spheric CH4 oxidation in soil and mechanism of NH4

+ inhibition.Appl Environ Microbiol 64:4291–4298

Hoffmann T, Horz H-P, Kemnitz D, Conrad R (2002) Diversity of theparticulate methane monooxygenase gene in methanotrophic sam-ples from different rice field soils in China and the Philippines.Syst Appl Microbiol 25:267–274

Holmes AJ, Costello A, Lidstrom ME, Murrell CJ (1995) Evidencethat participate methane monooxygenase and ammonia monoox-ygenase may be evolutionarily related. FEMS Microbiol Lett132:203–208

Holtan-Hartwig L, Dörsch P, Bakken LR (2002) Low temperaturecontrol of soil denitrifying communities: kinetics of N2O produc-tion and reduction. Soil Biol Biochem 34:1797–1806

Horz H-P, Yimga MT, Liesack W (2001) Detection of methanotrophdiversity on roots of submerged rice plants by molecular retrievalof pmoA, mmoX, mxaF and 16S rRNA and ribosomal DNA,including pmoA-based Terminal Restriction Length Polymor-phism profiling. Appl Environ Microbiol 67:4177–4185

Horz H-P, Rich V, Avrahami S, Bohannan BJM (2005) Methane-oxidizing bacteria in a California upland soil: diversity andresponse to simulated global change. Appl Environ Microbiol71:2642–2652

Im J, Lee S-W, Yoon S, DiSpirito AA, Semrau JD (2011) Character-ization of a novel facultative Methylocystis species capable ofgrowth on methane, acetate and ethanol. Environ Microbiol Rep3:174–181

Jones CM, Stres B, Rosenquist M, Hallin S (2008) Phylogeneticanalysis of nitrite, nitric oxide, and nitrous oxide respiratoryenzymes reveal a complex evolutionary history for denitrification.Mol Biol Evol 25:1955–1966

Jones CM, Welsh A, Throbäck IN, Dörsch P, Bakken LR, Hallin S(2011) Phenotypic and genotypic heterogeneity among closelyrelated soil-borne N2- and N2O-producing Bacillus isolates har-boring the nosZ gene. FEMS Microbiol Ecol 76:541–552

Kesik M, Blagodatsky S, Papen H, Butterbach-Bahl K (2006) Effect ofpH, temperature and substrate on N2O, NO and CO2 productionby Alcaligenes faecalis. J Appl Microbiol 101:655–667

Knittel K, Boetius A (2009) Anaerobic oxidation of methane: progresswith an unknown process. Annu Rev Microbiol 63:311–334

Lai DYF (2009) Methane dynamics in northern peatlands: a review.Pedosphere 19:409–421

Lauber CL, Hamady M, Knight R, Fierer N (2009) Pyrosequencing-based assessment of soil pH as a predictor of soil bacterialcommunity structure at the continental scale. Appl EnvironMicrobiol 75:5111–5120

LeMer J, Roger P (2001) Production, oxidation and consumption ofmethane by soils: a review. Eur J Soil Biol 37:25–50

Maljanen M, Hytönen J, Martikainen PJ (2001) Fluxes of N2O, CH4

and CO2 on afforested boreal agricultural soils. Plant Soil231:113–121

Martikainen PJ, Nykänen H, Alm J, Silvola J (1995) Change in fluxesof carbon dioxide, methane and nitrous oxide due to forest drain-age of mire sites of different trophy. Plant Soil 168:571–577

Mohanty SR, Bodelier PLE, Floris V, Conrad R (2006) Differentialeffects of nitrogenous fertilizers on methane-consuming microbesin rice field and forest soils. Appl Environ Microbiol 72:1346–1354

Murrell JC, McDonald IR, Bourne DG (1998) Molecular methods forthe study of methanotroph ecology. FEMS Microbiol Ecol27:103–114

Naeem S (1998) Species redundancy and ecosystem reliability. Con-serv Biol 12:39–45

Nedwell DB, Watson A (1995) CH4 production, oxidation and emis-sion in a UK ombrotrophic peat bog: influence of SO4

2– from acidrain. Soil Biol Biochem 27:893–903

1054 Wetlands (2012) 32:1047–1055

Author's personal copy

Nykanen H, Alm J, Lang K, Silvola J, Martikainen PJ (1995) Emis-sions of CH4, N2O and CO2 from a virgin fen and a fen drainedfor grassland in Finland. J Biogeogr 22:351–357

Palmer K, Drake HL, Horn MA (2010) Association of novel andhighly diverse acid-tolerant denitrifiers with N2O fluxes of anacidic fen. Appl Environ Microbiol 76:1125–1134

Pell M, Stenberg B, Stenström J, Torstensson L (1996) Potentialdenitrification activity assay in soil — with or without chloram-phenicol? Soil Biol Biochem 28:393–398

Philippot L, Hallin S, Schloter M (2007) Ecology of denitrifyingprokaryotes in agricultural soil. Adv Agron 96:249–305

Ravishankara AR, Daniel JS, Portmann RW (2009) Nitrous oxide(N2O): the dominant ozone-depleting substance emitted in the21st century. Science 326:123–125

Regina K, Nykänen H, Silvola J, Martikainen P (1996) Fluxes ofnitrous oxide from boreal peatlands as affected by peatland type,water table level and nitrification capacity. Biogeochemistry35:401–418

Regina K, Syväsalo E, Hannukkala A, Esala M (2004) Fluxes of N2Ofrom farmed peat soils in Finland. Eur J Soil Sci 55:591–599

Robertson GP, Tiedje JM (1987) Nitrous oxide sources in aerobic soils:nitrification, denitrification and other biological processes. SoilBiol Biochem 19:187–193

Simek M, Cooper JE (2002) The influence of soil pH on denitrifica-tion: progress towards the understanding of this interaction overthe last 50 years. Eur J Soil Sci 53:345–354

Smemo KA, Yavitt JB (2011) Anaerobic oxidation of methane: anunderappreciated aspect of methane cycling in peatland ecosys-tems? Biogeosciences 8:779–793

Stralis-Pavese N, Sessitsch A, Weilharter A, Reichenauer T, Riesing J,Csontos J, Murrell CJ, Bodrossy L (2004) Optimization ofdiagnostic microarray for application in analyzing landfill methano-troph communities under different plant covers. Environ Microbiol6:401–417

Thomsen J, Geest T, Cox R (1994) Mass-spectrometric studies of theeffect of pH on the accumulation of intermediates in denitrificationby Paracoccus denitrificans. Appl Environ Microbiol 60:536–541

Throbäck IN, Enwall K, Jarvis Å, Hallin S (2004) Reassessing PCRprimers targeting nirS, nirK and nosZ genes for community sur-veys of denitrifying bacteria with DGGE. FEMS Microbiol Ecol49:401–417

Walker BH (1992) Biodiversity and ecological redundancy. ConservBiol 6:18–23

Weslien P, Klemedtsson ÅK, Börjesson G, Klemedtsson L (2009)Strong pH influence on N2O and CH4 fluxes from forested or-ganic soils. Eur J Soil Sci 60:311–320

Wetlands (2012) 32:1047–1055 1055

Author's personal copy