The influence of oxygen and methane on nitrogen fixation in subarctic Sphagnum mosses

Kox e al. AMB Exp (2018) 8:76 h ps://doi.o g/10.1186/s13568-018-0607-2 ORIGINAL ARTICLE The in luence o oxygen andme hane onni ogen ixa ion insuba c ic Sphagnum mosses Ma ine A. R. Kox1* , Sanni L. Aal o2,3, Timo Pen ilä4, Ka ha ina F. E wig1, Mike S. M. Je en1 and Maa je A. H. J. an Kessel1 Abs ac Biological ni ogen ixa ion is an impo an sou ce o bioa ailable ni ogen in Sphagnum domina ed pea lands. Sphag- num mosses ha bo a di e se mic obiome including ni ogen- ixing and me hane (CH4) oxidizing bac e ia. The inhibi- o y e ec o oxygen on mic obial ni ogen ixa ion is documen ed o many bac e ia. Howe e , he ole o ni ogen- ixing me hano ophs in ni ogen supply o Sphagnum pea mosses is no well explo ed. He e, we in es iga ed he ole o bo h oxygen and me hane on ni ogen ixa ion in suba c ic Sphagnum pea mosses. Fi e species o Sphagnum mosses we e sampled om wo meso ophic and h ee oligo ophic si es wi hin he Lakkasuo pea land in O i esi, cen al Finland. Mosses we e incuba ed unde ei he ambien o low oxygen condi ions in he p esence o absence o me hane. S able iso ope ac i i y assays e ealed conside able ni ogen- ixing and me hane-assimila ing a es a all si es (1.4 ± 0.2 µmol 15N–N2 g−1 DW day−1 and 12.0 ± 1.1 µmol 13C–CH4 g−1 DW day−1, espec i ely). Addi ion o me hane did no s imula e inco po a ion o 15N-ni ogen in o biomass, whe eas oxygen deple ion inc eased he ac i - i y o he ni ogen- ixing communi y. Analysis o he 16S RNA genes a he bac e ial communi y le el showed a e y di e se mic obiome ha was domina ed by Alphap o eobac e ia in all si es. Bona ide me hane-oxidizing axa we e no e y abundan ( ela i e abundance less han 0.1%). Based on ou esul s we conclude ha me hano ophs did no con ibu e signi ican ly o ni ogen ixa ion in he in es iga ed pea lands. Keywo ds: Diazo ophy, Me hane oxida ion, Oxygen, Pea land, Sphagnum moss, 16S RNA amplicon sequencing © The Au ho (s) 2018. This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. In oduc ion Biological ni ogen (N2) ixa ion is o g ea impo ance o he Ea h’s biosphe e as i is he majo na u al p ocess o eplenish biologically a ailable ni ogen. In nu ien -lim- i ed ecosys ems, he be e compe i o s a e o en hose ha ind al e na i e ways o gain hei equi ed nu i- en s, i.e. by engaging in a symbiosis wi h o he o ganisms ( an de Heijden e  al. 2008). Pea lands a e nu ien - limi ed ecosys ems ha ha e been s udied in ensi ely due o hei signi ican ole in he global ca bon (C) cycle. App oxima ely 1/3 o he global e es ial ca bon is s o ed as dead o ganic ma e in pea lands (Go ham 1991). In Sphagnum-domina ed pea lands, Sphagnum mosses a e he ecosys em enginee s. They ou compe e ascula plan species in a ious ways (Malme e  al. 2003), bu mainly by c ea ing and main aining acidic (pH 3–5) and wa e logged condi ions. In addi ion, hei own biomass is di icul o deg ade, which con ibu es o he slow decomposi ion and consequen ial accumula ion o dead o ganic ma e (Clymo 1963, 1964; an B eemen 1995). Pea ens ha a e meso ophic o oligo ophic ecei e N om a mosphe ic deposi ion and g ound wa e in low. Compa ed o meso ophic ens, oligo ophic ens ecei e less nu ien s, leading o nu ien limi a ion and lowe p oduc i i y (La mola e al. 2014). In bo h sys ems, Sphagnum mosses minimize nu ien a ailabili y o as- cula plan s by apid and e icien nu ien up ake (F i z e al. 2014). Open Access *Co espondence: [email p o ec ed].nl 1 Depa men o Mic obiology, Radboud Uni e si y, Nijmegen, The Ne he lands Full lis o au ho in o ma ion is a ailable a he end o he a icle Page 2 o 9 Kox e al. AMB Exp (2018) 8:76 In hese nu ien -limi ed pea lands, Sphagnum ci cum- en s N-limi a ion by engaging in a ela ionship wi h N2 ixing mic oo ganisms (diazo ophs). Diazo ophs con- e a mosphe ic N2 o ammonia (NH3). This is a cos ly p ocess (16 ATP pe N2 molecule) ca alyzed by an oxy- gen (O2) sensi i e ni ogenase enzyme (Pos ga e 1982). In na u e, diazo ophs a e abundan , di e se and exis as ee-li ing s a e as well as in symbiosis wi h plan s. Many moss species a e known o ha bo a di e se diazo ophic communi y (Leppänen e al. 2013; Vile e al. 2014; Kno e al. 2014; Kox e al. 2016; Wes on e al. 2015). The diazo- ophic ac i i y associa ed wi h Sphagnum suppo s and explains high concen a ions o N in Sphagnum biomass (Vile e al. 2014). Al hough i is e iden ha he mosses bene i om he N supply by diazo ophs, he bene i s o he mic oo ganism a e less appa en . Especially, since i was ecen ly pos ula ed ha op imal condi ions o diazo ophic mic oo ganisms and he moss-hos a e e y di e en ( an den Elzen e al. 2017). Biological N2 ixa ion ac i i y is mos commonly de e - mined using he ace ylene educ ion assay (Ha dy e al. 1968). Ini ial s udies on he Sphagnum associa ion wi h N2 ixing pa ne s indica ed ha mainly cyanobac e- ia con ibu ed o he inco po a ion o N in Sphagnum biomass (Be g e al. 2012; Lindo e al. 2013). Al hough he ace ylene educ ion assay is a sensi i e and easy way o de e mine N2 ixing ac i i y, ace ylene i sel is an i e e sible inhibi o o he me hane monooxygenase enzyme in me hane (CH4) oxidizing bac e ia (me hano- ophs). This p e en s me hano ophs om me aboliz- ing ca bon, ul ima ely leading o cell dea h. Due o he use o ace ylene o measu e N2 ixa ion a es, he ole o diazo ophic me hano ophs may ha e been unde - es ima ed (Leppänen e al. 2013; Vile e al. 2014). As an al e na i e me hod o measu e N2 ixa ion, 15N–ni ogen (15N–N2) s able iso ope inco po a ion can be used. Wi h his me hod, Vile e al. (2014) showed a signi ican con- ibu ion o CH4 dependen N2 ixa ion o he Sphagnum N-pool (Vile e al. 2014). Se e al Sphagnum 16S RNA gene-based mic obiome s udies indica ed ha Alphap o- eobac e ia we e he mos abundan N2 ixing bac e ia (B agina e al. 2012, 2014; Shche bako e al. 2013; Wa - en e al. 2017). In addi ion, di e si y s udies based on ni ogenase ni H gene also showed ha Alphap o eobac- e ia we e highly ep esen ed (B agina e al. 2012; Vile e al. 2014; Kox e al. 2016; Wa en e al. 2017). Some o hese N2 ixing Alphap o eobac e ia we e ound o be me hano ophs (i.e. ype II Me hylosinus spp., Me hylo- cys is spp.) as well. Me hano ophs associa ed wi h Sphagnum mosses ha e been shown o suppo moss g ow h by p oducing ca bon dioxide (CO2) which is subsequen ly aken up by Sphagnum, especially unde CO2 limi ing condi ions (Raghoeba sing e al. 2005; Kip e al. 2010). The me ha- no ophs a e hypo hesized o bene i om O2 p oduced by Sphagnum and by being p o ec ed om p eda o s inside Sphagnum’s hyaline cells (Kos ka e  al. 2016). Howe e , ac i i y measu emen s o me hano ophic diazo ophs in en i onmen al Sphagnum moss samples ha e yielded con as ing esul s (La mola e  al. 2014; Leppänen e al. 2014). Due o he high ene gy demand o N2 ixa ion, a me hano ophic diazo oph equi es high me hano ophic ac i i y in o de o sus ain N2 ixa ion. The e o e, pa ame e s con olling me hano ophy sup- posedly con ibu e o he obse ed a iabili y in he CH4 dependen N2 ixa ion (Ho and Bodelie 2015). To his poin , he composi ion o he ac i e N2 ixing communi y associa ed wi h Sphagnum and ac o s ha a ec his ac i i y ha e been in es iga ed in only a ew s udies (Ho and Bodelie 2015). The mic oo ganisms in ol ed in bio- logical N2 ixa ion ha e been iden i ied, bu hei ela i e impo ance and ac o s con olling hei ac i i y emain un esol ed (Ho and Bodelie 2015). The aim o his s udy was o elucida e he e ec o O2 and CH4 on N2 ixa ion in Sphagnum mosses o oligo- ophic and meso ophic suba c ic Sphagnum-domi- na ed pea lands. Sphagnum mosses we e sampled om wo meso ophic and h ee oligo ophic si es (Lakkasuo pea land in O i esi, Finland). Sphagnum mosses we e incuba ed unde ei he ambien o low oxygen condi- ions in he p esence o absence o 13C-labeled CH4 and 15N-labeled N2. Highe N2 ixa ion ac i i y is expec ed o se e e nu ien limi ed condi ions (oligo ophic) com- pa ed o meso ophic condi ions. On he o he hand, mo e bu e ed condi ions and highe pH p e ailing in meso ophic pea lands ha e p e iously been shown o be bene icial o N2 ixa ion ( an den Elzen e al. 2017). We hypo hesize ha CH4 will s imula e N2 ixa ion ac i i y and his ac i i y will be highes in low O2 condi ions. Ma e ials andme hods Si e desc ip ion andexpe imen al se ‑up The s udy was pe o med in he suba c ic, nu ien lim- i ed mi e complex Lakkasuo in cen al Finland (61°47′N 24°18′E; 150m.a.s.l.). Lakkasuo is a well-s udied bo eal mi e complex, wi h an annual N inpu ia ainwa e o 0.40 g N m−2 yea −1 (Laine 2004). Sphagnum mosses we e collec ed in Sep embe 2014 om oligo ophic and meso ophic ens and a ainwa e - ed bog wi hin he same pea land basin (Addi ional ile 1: Table S1). The meso ophic en si e domina ed by Sphagnum sub- secundum (si e A) was a we en sys em, whe eas meso- ophic en si e B wi h Sphagnum ob usum was d ie . The oligo ophic en si e was na u ally di ided in wo pa ches whe e Sphagnum allax (si e C) and Sphagnum papillosum (si e D), espec i ely, we e dominan . The Page 3 o 9 Kox e al. AMB Exp (2018) 8:76 ainwa e - ed bog si e was domina ed by Sphagnum majus (si e E). See Addi ional ile1: Table S2 o a ull o e iew o he si e ege a ion index. Incuba ions To es whe he a ailabili y o O2 and/o CH4 would a ec mic obial N2 ixa ion a es, Sphagnum mosses we e incuba ed unde wo di e en O2 concen a ions (ambien and deple ed) and wi h o wi hou 13C–CH4. N2 ixa ion and CH4 oxida ion ac i i y we e es ima ed by measu ing he inco po a ion o 15N–N2 and 13C–CH4 espec i ely. To co ec o he na u al p esence o 15N and 13C in he moss, con ols we e incuba ed wi hou any labelled gasses. Pe ea men , moss samples (10 plan le s pe bo - le, op 3cm) om each si e we e incuba ed in 180ml plas ic Scholl lasks and closed wi h sep um con aining sc ew-caps. Be o e incuba ion, he esh weigh (FW) o he mosses was de e mined. All bo les ecei ed 5% 15N–N2 in he headspace, excep o he backg ound con- ols. 13C–CH4 ea ed samples ecei ed addi ionally 5% 13C–CH4 in he headspace. The headspace o he ambi- en O2 condi ion consis ed o ai . Low O2 condi ions we e achie ed by eplacing ( ou cycles o acuum—helium gassing) he headspace wi h an a i icial gas mix u e con- sis ing o N2 gas (80%), CO2 (0.04%) and helium (20%). Oxygen was no added, so ha he only sou ce o oxygen was pho osyn hesis by he mosses. Samples we e incu- ba ed o 48h ou side unde p e ailing ligh and empe - a u e condi ions (Jy äskylä, Finland, Sep embe 2014). S able iso ope inco po a ion A e he incuba ion, moss samples we e s o ed a − 20°C. Nex , samples we e eeze-d ied using an Alpha 1-4LD plus (Ma in Ch is GmbH, Os e ode am Ha z, Ge many) and s o ed a − 20°C. D ied s ems om each ea men we e pooled and dis up ed using a beadbea e a 1500 pm o 3min (Mic odismemb a o U, B. B aun Bio ech In ., Meisungen, Ge many). To de e mine he ac ion o 15N and 13C inco po a ed in plan biomass, app oxima ely 5 and 0.225mg, espec i ely, we e pu in o 5 × 8mm in cups in duplo. Nex , samples we e com- bus ed by lash combus ion (1800°C) on a CNS analyze (EA 1110 Ca lo E ba, The mo Fishe Scien i ic, Wal ham, MA, USA) coupled o an Iso opic Ra io Mass Spec ome- e (Finnigan Del aPlus, The mo Elec o GmbH, B emen Ge many) ia an in e ace (Con lo III, The mo Elec o GmbH, B emen, Ge many). 15N and 13C con en o all mosses be o e incuba ion we e uni o m. The 15N–N2 ixa ion a es we e calcula ed by co ec ing he δN o he en iched plan ma e ial o he na u al 15N con en , a e which he co ec ed 15N inc ease was con- e ed o 15N–N2 ixa ion a es (µmolNg−1DWday−1). 13C–CH4 oxida ion a es we e calcula ed in a simila ash- ion as he 15N–N2 ixa ion a es. Fi s , he backg ound 13C label was sub ac ed om he measu ed inc ease in 13C o he biomass. The co ec ed 13C inc ease was sub- sequen ly con e ed in o 13C–CH4 oxida ion a es. Po ewa e composi ion A each si e pH was measu ed and po e wa e samples we e aken using hizons (po e size 2µm; 5cm leng h) a he su ace (0cm), 5, 10 and 15cm dep h. Elemen concen a ions o Al, Ca, Fe, K, Mg, Mn, Na, P, S, Si and Zn in po e wa e (10ml, acidi ied wi h 1ml HNO3) we e analyzed by induc i ely coupled plasma op ical emission spec oscopy (ICP-OES iCAP 6000, The mo Fishe Sci- en i ic, Wal ham, MA, USA). Concen a ions o NH4+, NO3−, PO43− we e analyzed colo ime ically wi h a 3 Au o Analyze sys em (B an and Luebbe GmbH, No de - s ed , Ge many) using ammonium molybda e (Hen iksen 1965), hyd azine sul a e (Kamphake e al. 1967) o salicy- la e (G assho and Johannsen 1972). Cl was de e mined wi h a Technicon Flame Pho ome e IV Con ol (B an and Luebbe, No de s ed , Ge many). To de e mine he concen a ion o dissol ed me hane, 3ml exe aine s (Labco, Ce edigion, UK) we e p epa ed wi h 1g o NaCl and closed wi h a sep um cap. Nex , 1ml o he po ewa e collec ed ia hizons was added o he closed exe aine s immedia ely a e sampling. The p essu e in he ials was measu ed and he CH4 concen- a ion in he headspace o he exe aine was measu ed using he GC as desc ibed abo e. Olsen‑P de e mina ion Diges ion o d ied and g ound Sphagnum issue was pe - o med o ob ain he Olsen P concen a ion as desc ibed ( an den Elzen e al. 2017). In b ie , samples we e hea ed o 120 °C o 45 min in a mix u e o 500 µl HNO3 (65%w/w) wi h 200µl H2O2 (30%w/w). Nex , samples we e dilu ed wi h demine alized wa e and measu ed by induc i ely-coupled plasma emission spec ome y (IRIS In epid II, The mo Elec on co po a ion, F anklin, MA, USA). DNA ex ac ion and16S RNA gene ampli ica ion Sphagnum mosses o molecula analysis we e sam- pled di ec ly om he ield and immedia ely pu in liq- uid ni ogen. In he labo a o y, samples we e s o ed a − 80°C. DNA ex ac ion was pe o med on 0.5g o sample ( esh weigh ), using he Fas DNA SPIN ki o soil (MP Biomedicals, San a Ana, CA, USA), ollow- ing manu ac u e s p o ocol. Beadbea ing was inc eased o 2 × 1.5 min a 50 Hz using a issue lyse (LT, Qia- gen, Hilden, Ge many) DNA yield was assessed using Qubi luo ome ic analysis (The mo Fishe Scien i ic, Page 4 o 9 Kox e al. AMB Exp (2018) 8:76 Wal ham, MA, USA). Ba coded 16S RNA gene amplicon lib a y was p epa ed wi h a 2 s ep PCR p o ocol (Be y e al. 2011). Used p ime s a ge ed he V3–V4 egion o he 16S RNA gene o mos bac e ia (341F-785R; Klind- wo h e al. 2013). The 25μl PCR eac ions o he i s PCR con ained 12.5μl Quan a pe ec a mix (Quan abio, Be e ly, MA, USA), 1μl o each p ime (20μM), 1μl DNA (0.5ng/μl). The PCR p og am consis ed o 25 cycles o 95°C 1min, 60°C, 1min, 72°C o 2min, a e which inal elonga ion 10min 72°C. The ob ained PCR p od- uc s we e checked o pu i y and size on 1.5% aga ose gel. PCR p oduc s o 7 pa allel eac ions we e pooled and pu i ied using QIAquick pu i ica ion ki (Qiagen, Hilden, Ge many) ollowing manu ac u e s p o ocol. The second PCR was pe o med o ba code all samples (bcPCR). Fo bcPCR he gene-speci ic The same PCR p ime s (341F and 785R) as o he i s PCR eac ion we e agged wi h adap e sequences, speci ic ba codes and key sequences a he 5′ end, compa ible wi h Ion To en sequencing echnology ( o al 60–62 nucleo ides pe p ime ). Fo each sample six nes ed bcPCR eac ions we e conduc ed in pa allel, which we e combined a e p oduc pu i y and size con ol. Subsequen ly, nes ed PCR p oduc s we e pu i ied using QIAquick pu i ica ion ki (Qiagen, Hilden, Ge many) ollowing manu ac u e s p o ocol. Amplicon sequencing P io o Ion o en lib a y cons uc ion, concen a ion and agmen leng h o samples was de e mined wi h a Bioanalyze 2100 and he High Sensi i i y DNA ki (Agi- len Technologies, San a Cla a, CA, USA). The lib a ies we e dilu ed o a inal concen a ion o 26pM. Acco d- ing o manu ac u e ’s p o ocol he lib a y agmen s we e a ached o Ion Sphe e pa icles using he One Touch ins umen and Ion PGM Templa e OT2 400 ki (Li e Technologies, Ca lsbad, CA, USA). Subsequen ly, Ion Sphe e pa icles we e loaded on an Ion 318 2 Chip o he i s un and an Ion 314 2 chip o he second un, a e which he amplicon lib a ies we e sequenced acco ding o manu ac u e ’s p o ocol using he ION PGM Sequencing 400 Ki , using 850 nucleo ide lows. Run 1 (318 chip) esul ed in 35,790 eads wi h an a e age ead leng h o 132 base pai s. Run 2 esul ed in 180,870 eads wi h an a e age ead leng h o 275 base pai s. Sequence analysis Sequences om bo h uns we e me ged esul ing in a o al o 216,660 eads wi h an a e age leng h o 252 nucleo ides. The eads we e analysed using Mo hu ( 1.38; Schloss e  al. 2009) and he Mo hu 454 SOP (Schloss e al. 2011). Fi s , eads we e quali y il e ed on ead leng h (200–450 bp) allowing o homopolyme s (maximum 8), a maximum o 2 di e ences wi h he p ime sequence and minimum a e age quali y sco e o 20 o e a window o 50 base pai s. Imp o ed eads we e aligned o he Sil a n da abase ( elease 123) (Quas e al. 2013). Chime a’s we e emo ed using Uchime. Nex , eads we e classi ied a boo s ap alue o 80%, a e which he unwan ed and non- a ge lineages A chaea, Euka yo a, chlo oplas , mi ochond ia and unknown we e emo ed. The inal da ase om which OTUs we e clus- e ed consis ed o 49,975 sequences o which 16,493 we e unique, wi h an a e age ead leng h o 221 nucleo ides ( o a de ailed o e iew o quali y il e ing see Addi ional ile1: Table S3). OTU’s we e clus e ed using he a e age clus e ing algo i hm a a cu o o 0.03, ollowed by sin- gle on emo al and esul ed in 2001 OTUs. Sequences we e deposi ed in NCBI SRA unde p ojec numbe PRJNA432031. S a is ical analysis Da a was analyzed using R e sion 3.4.0 by he R De el- opmen Co e Team (2017). No mali y o he esiduals was es ed using Shapi o–Wilk’s no mali y es (s a s- package). Homogenei y o a iance was es ed using Le ene’s es (ca -package). In he analysis o N2 ixa ion, CH4 oxida ion ac i i y and po e wa e composi ion, he si es we e g ouped based on hei nu i ional ophic s a e, esul ing in he le els: meso ophic and oligo- ophic si es. G ouping was only allowed i he di e ence be ween species was no signi ican (p > 0.05), which was es ed i s using a one-way ANOVA. N2 ixa ion ac i - i y da a was ans o med by squa e oo ans o ma ion o pe mi pa ame ic es s. Nex , di e ences in N2 ixa- ion ac i i y unde he di e en O2 condi ions, p esence o 13C–CH4 and si es (oligo ophic o meso ophic), we e es ed using a 3-way ANOVA (s a s-package; o sample size (n) see Addi ional ile1: Table S5). 13C–CH4 oxida- ion ac i i y da a was no mally dis ibu ed. Di e ences in ac i i y be ween he wo oxygen ea men s and si es we e es ed using a 2-way ANOVA (s a s-package). All g aphs we e cons uc ed in R using ggplo 2. Fo downs eam analysis o he communi y compo- si ion he OTU able and axonomy iles gene a ed by Mo hu , we e impo ed in o R and analyzed using phy- loseq (McMu die and Holmes 2013). Resul s Po ewa e composi ion The oligo ophic si es we e mo e acidic compa ed o he meso ophic si es and con ained less dissol ed CH4 (see Addi ional ile 1: Figu e S1). Concen a- ions o NH4+, NO3− and PO4− we e lowes in he ainwa e ed bog (no shown) and we e p esen in sligh ly highe concen a ions in bo h oligo ophic and meso ophic ens. Mo eo e , bo h oligo ophic Page 5 o 9 Kox e al. AMB Exp (2018) 8:76 and meso ophic ens showed o e all highe elemen- al concen a ions o S, Fe Mg, K and Ca, compa ed o he bog. Olsen P was highe in mosses o igina ing om meso ophic si es (0.65 ± 0.02 μmol P g−1 DW, see Addi ional ile 1: Figu e S2) han in oligo ophic si es (0.54 ± 0.01μmolPg−1DW). N2 ixa ion and CH4 oxida ion ac i i y N2 ixa ion a es as measu ed by 15N–N2 inco po a ion we e simila in all si es (meso ophic 1.6 ± 0.4 and oli- go ophic 1.5 ± 0.4 μmol 15N–N2g −1 DW day−1). The N2 ixa ion a es we e no a ec ed by 13C–CH4 addi ion (p > 0.05), bu we e a ec ed by absence o O2 (F1,16 = 6.40 p = 0.022; see Fig.1). Incuba ion unde low O2 condi ions yielded highe N2 ixa ion ac i i y (2.0 ± 0.5 μmol 15N– N2g−1DWday−1) compa ed o incuba ions a ambien O2 condi ions (0.8 ± 0.2 μmol 15N–N2g−1DWday−1). CH4 oxida ion occu ed in all incuba ions, and 13C– CH4 inco po a ion was simila (p > 0.05) in all incuba ions (o e all a e 12.0 ± 1.1 µmol 13C–CH4g −1 DW day−1) (Fig.2). Mic obial communi y analysis The bac e ial communi y analysis showed ha he P o eo- bac e ia we e he mos dominan phylum in all samples (Fig.3). In he oligo ophic en samples Alphap o eobac- e ia we e he mos abundan class p esen (> 50% wi hin phylum P o eobac e ia; see Fig.3 and Addi ional ile1: Figu e S4). The mos abundan amilies wi hin he Alp- hap o eobac e ia a e: Ace obac e aceae and he Caulo- bac e aceae. The o he phyla ha made up a leas 1% o Meso ophic Oligo ophic Ambien Low Ambien Low 0 1 2 3 4 5 O condi ion 15N-N2 inco po a ion µmol g−1DW day−1 Labelling 15 15 2 N-N N-N + 4 13 2C-CH 2 ** Fig. 1 Inco po a ion o 15N–N2 (μmol g−1 DW day−1) in he meso ophic ens (n = 2) and oligo ophic ens and bog (n = 3) Sphagnum mosses incuba ed wi h ambien O2 condi ions (da k g ey ba s) o low O2 condi ions (ligh g ey ba s), supplemen ed wi h 5% 15N–N2 (da k g ey ba s) o 5% 15N–N2 + 5% 13C–CH4 (ligh g ey ba s) 0 5 10 15 20 25 Meso ophic Oligo ophic Si e 13C-CH4inco po a ion µmolg −1DW day−1 O condi ion Ambien Low 2 Fig. 2 Inco po a ion o 13C–CH4 (μmol g−1 DW day−1) in Sphagnum mosses om meso ophic ens and oligo ophic ens and bog si es, incuba ed wi h ambien O2 condi ions (da k g ey ba s) o low O2 condi ions (ligh g ey ba s) Page 6 o 9 Kox e al. AMB Exp (2018) 8:76 he mic obial communi y and ha appea ed o be p esen in all si es we e he Acidobac e ia, Ve ucomic obia and he Planc omyce es (Fig.3). The phylum o Acidobac e ia had a compa able ela i e abundance ac oss all si es. In con as , he Bac e oide es and Cyanobac e ia we e only p esen in some si es. Genuine me hano ophic p o eo- bac e ial 16S RNA sequences ( ype I and ype II) we e less han 0.1% in ela i e abundance in all samples. In he meso ophic si es, highe mic obial di e si y was obse ed (see Addi ional ile1: Figu e S3) compa ed o he oligo ophic si es. Ve ucomic obia we e ela i ely less abundan in he meso ophic si es compa ed o he oligo ophic si es, whe eas he Bac e oide es (4% o o al eads) and Cyanobac e ia (3% o o al eads) we e mo e abundan ly ound in he meso ophic si es. Discussion This s udy aimed o elucida e he e ec o CH4 and O2 a ailabili y on N2 ixa ion ac i i y in oligo ophic and meso ophic Sphagnum-domina ed pea lands. Biologi- cal N2 ixa ion ac i i y was expec ed o be s imula ed by he p esence o CH4 and he absence o O2. Fo he si e e ec (oligo ophic s. meso ophic), expec a ions we e no so clea -cu , as di e en con ounding ac o s play a ole. Highe a ailabili y o N in meso ophic si es may dec ease he demand o biological N2 ixa ion, while highe P con en may lead o ela i e N sca ci y. Also, he be e bu e ing and highe pH o meso ophic si es may be a o able o N2 ixa ion. In he oligo ophic s udy si es, howe e , e y low con- cen a ions o he N compounds ni a e and ammonium, ypical o oligo ophic condi ions, we e no ound (Addi- ional ile1: Figu e S1); hey we e compa able o concen- a ions a he meso ophic si es. Also he N:P ela ion in po ewa e was simila , so ha no bigge ela i e N sca - ci y could be diagnosed o he oligo ophic si e. Only in he biomass, sligh ly highe Olsen P e lec ed a gene ally highe a ailabili y o P in he meso ophic si e. Taken oge he , his may explain why N2 ixa ion a es as meas- u ed by 15N–N2 inco po a ion did no di e be ween nu i ional meso ophic and oligo ophic ophic s a es. This is in con as wi h ecen obse a ions by a s udy o Van den Elzen e al. (2017), whe e i was ound ha N2 ixa ion is s imula ed by mo e bu e ed condi ions and highe phospho us a ailabili y. I is possible ha N2 ixa- ion a es in ou esea ch we e a ec ed by o he ac o s such as he ela i e d y pe iod in which samples we e aken o lack o ce ain ace elemen s (Vi ousek e al. 2013; Wa en e al. 2017). Me hane addi ion lead o signi ican up ake o 13C– CH4 de i ed ca bon, bu did no a ec N2 ixa ion a es compa ed o con ols, indica ing ha CH4 dependen N2 ixa ion is p obably no a majo con ibu o o N2 ixa- ion in he s udied ecosys em. These esul s a e simila o indings o bo h Leppänen e al. (2014) and Wa en 0.00 25.0 50.0 75.0 100 Si e A Meso ophic en S. subsecundum Si e B Meso ophic en S. ob usum Si e C Oligo ophic en S. allax Si e D Oligo ophic en S. papillosum Si e E Oligo ophic bog S. majus Rela i e Abundance (Phyla >1%) Phylum Acidobac e ia Ac inobac e ia Bac e oide es Chlo o lexi Cyanobac e ia Gemma imonade e s Pa cubac e ia Planc omyce es P o eobac e ia Ve ucomic obia WD272 unclassi ied Fig. 3 Taxonomic composi ion (16S RNA) o he mic obial communi y associa ed wi h Sphagnum moss om si e A–E. Ba cha s ep esen he ela- i e abundance o he di e en phyla p esen in each si e. Only phyla wi h a RA > 1% a e shown Page 7 o 9 Kox e al. AMB Exp (2018) 8:76 e al. (2017). In bo h s udies he addi ion o me hane o ba ch incuba ions wi h Sphagnum did no a ec N2 ixa- ion ac i i y ei he . Fu he mo e, bo h s udies showed much lowe CH4 oxida ion ac i i y han obse ed in he p esen s udy, ques ioning he possibili y o an N2- ixing li e s yle o he me hano oph (Ho and Bodelie 2015). Al hough me hano ophs comp ised less han 0.1% o he 16S RNA, his alone does no exclude hem om being a majo con ibu o o N2 ixa ion. Recen ly Bae e al. (2018) showed ha a chaea cons i u ing only 0.27% o he mic obial communi y could accoun o 44% o he N ixed in an Flo ida pea sys em. In addi ion, N2 ixa ion a es may di e o e seasons (Le and Michelsen 2014) and hus he mic obial guilds ac i ely pe o ming N2 ixa- ion may a y wi h season. To u he con i m which N2 ixing mic oo ganisms a e ac i e, ansc ip omic s udies combined wi h ac i i y assays o e di e en seasons a e equi ed in u u e s udies. In con as o CH4 addi ion, O2 deple ion did s imula e N2 ixa ion. While he mosses in he incuba ion bo les did p oduce O2 as hey pe o med oxygenic pho osyn- hesis du ing he day, he ne e heless lowe O2 le el esul ed in highe N2 ixa ion a es compa ed o he ambien O2 le els, which is in line wi h ou hypo hesis. The ni ogenase enzyme ha is esponsible o N2 ixa- ion is i e e sibly inhibi ed by O2 in mos mic oo gan- isms (Vi ousek e al. 2002). Ou inding indica es ha ac i i y o diazo ophs migh be s ongly con olled by he oxygen concen a ion, and he e o e by dep h, posi- ion unde o abo e he wa e le el and he cu en a e o pho osyn hesis o Sphagnum. N2 ixa ion migh be highe in he da k han in he ligh , due o O2 elease by Sphagnum du ing he ligh pe iod. The CH4 oxida ion a es obse ed in his s udy a e compa able o CH4 oxida ion a es measu ed in o he Sphagnum mosses (Kip e al. 2010). The CH4 oxida ion a es we e sligh ly highe in he meso ophic si es p ob- ably due o highe pH. High CH4 oxida ion ac i i y is essen ial o me hano ophs in o de o ul il he high ene gy equi ed o N2 ixa ion (Ho and Bodelie 2015). Wi h hese high CH4 oxida ion a es, ene gy seems no he limi ing ac o o CH4 dependen N2 ixa ion. Taken oge he i seems highly unlikely ha CH4 dependen N2 ixa ion is a majo N-supplie o he N-pool o he Sphag- num mosses in es iga ed in his s udy. Mic obial communi y composi ion The obse ed mic obial communi y composi ion o he Sphagnum associa ed mic obial communi y is compa a- ble o he composi ion ound in o he s udies (B agina e al. 2012, 2014; Kox e al. 2016) wi h he P o eobac e- ia as dominan phylum (B agina e al. 2011, 2014; Pu - kinen e al. 2014). The po en ial o pe o m N2 ixa ion has been epo ed o many phylogene ic g oups (Zeh e al. 2003; Dixon and Kahn 2004; Khadem e al. 2010), o which he Alphap o eobac e ia, Cyanobac e ia, Bac e- oide es and Ve ucomic obia we e p esen in ou s udy. The Alphap o eobac e ia which we e p esen in all si es, ha e p e iously been iden i ied as po en ially impo an N2 ixing pa ne s o Sphagnum mosses (B agina e al. 2012; Vile e al. 2014). RT-qPCR and/o ansc ip omic sequencing we e no pe o med in his s udy, he e o e we can only specula e which N2 ixe s we e ac i e. When we compa e he di e en si es, i is appa en ha he oligo ophic ens only con ain a limi ed numbe o cyanobac e ial species, whe eas he meso ophic ens do con ain mo e Cyanobac e ia. Po en ially he Cyanobac e- ia may ha e been a ec ed by he nu ien a ailabili y in he pea land, wi h hem becoming mo e abundan wi h highe nu ien and p oduc i i y le els. The Ve ucomi- c obial OTU coun inc eased om meso ophic o he oligo ophic si es, his sugges s ha species o Ve ucomi- c obia migh h i e be e unde mo e oligo ophic con- di ions (B agina e al. 2015). The high ela i e abundance o Ve ucomic obia p esen in he pea mosses ha e been ound in ea lie s udies as well (Pu kinen e al. 2014; B a- gina e al. 2014). Fu u e isola ion o Ve ucomic obial spe- cies and physiological s udies should e eal hei me abolic po en ial wi h espec o N2 ixa ion and CH4 oxida ion. Up o his poin , ac o s con olling N2 ixa ion in Sphagnum mosses ha e yielded con as ing esul s. This s udy has ocused on he e ec o O2 and CH4 on N2 ixa- ion ac i i y in Sphagnum mosses in oligo ophic and meso ophic suba c ic pea lands. Based upon he esul s we conclude ha CH4 dependen N2 ixa ion was no a majo sou ce o ni ogen o Sphagnum mosses a he s udied si es. The mic obial communi y associa ed wi h Sphagnum was domina ed by P o eobac e ia (mainly Alphap o eobac e ia), which is compa able o o he s ud- ies. Fu u e s udies should combine ield and mesocosm s udies, wi h ac i i y assays, communi y analysis and ansc ip omic da a o unco e con ols o biological ni ogen ixa ion in Sphagnum mosses. Au ho s’ con ibu ions MARK, KFE, MSMJ and SLA designed he expe imen s; MARK, KFE, SLA, and TP sampled he pea mosses and pe o med he expe imen s; MARK and MAHJ K analysed he da a. The manusc ip was w i en by MARK, MAHJ K and MSMJ. All co-au ho s i e a i ely e iewed he manusc ip p io o submission. All au ho s ead and app o ed he inal manusc ip . Au ho de ails 1 Depa men o Mic obiology, Radboud Uni e si y, Nijmegen, The Ne h- e lands. 2 Depa men o Biological and En i onmen al Science, Uni e si y Addi ional ile Addi ional ile1. Supplemen a y ables and igu es. Page 8 o 9 Kox e al. AMB Exp (2018) 8:76 o Jy äskylä, PO Box 35, 40014 Jy äskylä, Finland. 3 Depa men o En i on- men al and Biological Sciences, Uni e si y o Eas e n Finland, PO Box 1627, 70211 Kuopio, Finland. 4 Na u al Resou ces Ins i u e Finland, PO Box 2, 00791 Helsinki, Finland. Acknowledgemen s Ma ja Tii ola is hanked o eedback on he expe imen s and manusc ip . Je oen F ank is hanked o se ing up bioin o ma ics pipelines and o use ul discussions on 16S RNA gene di e si y analysis. Compe ing in e es s The au ho s decla e ha hey ha e no compe ing in e es s. A ailabili y o da a and ma e ials All sequencing da a has been deposi ed in he NCBI SRA da abase, P ojec Numbe PRJNA432031. Consen o publica ion No applicable. E hics app o al and consen o pa icipa e No applicable. Funding M.A.R.K. was suppo ed by Eu opean Resea ch Council Ecomom 339880 o M.S.M.J., who was u he suppo ed by he Ne he lands O ganisa ion o Scien i ic Resea ch (SIAM G a i a ion G an 024 002 002 and Spinoza Awa d). K.F.E. was suppo ed by he Ne he lands O ganisa ion o Scien i ic Resea ch (Veni G an 863.13.007). M.A.H.J. .K. was suppo ed by Technology Founda- ion STW (G an 13146). S.L.A. was suppo ed by Academy o Finland p ojec 260797 and he Eu opean Resea ch Council Consolida o p ojec Mic o-RIP (615146). Publishe ’s No e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in pub- lished maps and ins i u ional a ilia ions. Recei ed: 24 Ma ch 2018 Accep ed: 28 Ap il 2018 Re e ences Bae HS, Mo ison E, Chan on JP, Og am A (2018) Me hanogens a e majo con ibu o s o ni ogen ixa ion in soils o he Flo ida E e glades. Appl En i on Mic obiol. h ps://doi.o g/10.1128/aem.02222-17 Be g A, Danielsson Å, S ensson BH (2012) T ans e o ixed-N om N2- ixing cyanobac e ia associa ed wi h he moss Sphagnum ipa ium esul s in enhanced g ow h o he moss. Plan Soil 362:271–278. h ps://doi. o g/10.1007/s11104-012-1278-4 Be y D, Mah oudh KB, Wagne M, Loy A (2011) Ba coded p ime s used in mul iplex amplicon py osequencing bias ampli ica ion. Appl En i on Mic obiol 77:7846–7849. h ps://doi.o g/10.1128/AEM.05220-11 B agina A, Be g C, Ca dinale M, Shche bako A, Chebo a V, Be g G (2011) Sphagnum mosses ha bou highly speci ic bac e ial di e si y du ing hei whole li ecycle. ISME J 6:1–12. h ps://doi.o g/10.1038/ismej.2011.151 B agina A, Maie S, Be g C, Mülle H, Chobo V, Hadacek F, Be g G (2012) Simila di e si y o Alphap o eobac e ia and ni ogenase gene amplicons on wo ela ed Sphagnum mosses. F on Mic obiol 2:1–10. h ps://doi. o g/10.3389/ micb.2011.00275 B agina A, Obe aune -Wappis L, Zachow C, Halwachs B, Thallinge GG, Mülle H, Be g G (2014) The Sphagnum mic obiome suppo s bog ecosys em unc ioning unde ex eme condi ions. Mol Ecol. h ps://doi.o g/10.1111/ mec.12885 B agina A, Be g C, Be g G (2015) The co e mic obiome bonds he Alpine bog ege a ion o a anskingdom me acommuni y. Mol Ecol 24:4795–4807. h ps://doi.o g/10.1111/mec.13342 Clymo RS (1963) Ion exchange in Sphagnum and i s ela ion o bog ecology. Ann Bo 27:309–324. h ps://doi.o g/10.1093/ox o djou nals.aob.a083847 Clymo RS (1964) The o igin o acidi y in Sphagnum bogs. B yologis 67:427. h ps://doi.o g/10.2307/3240768 Dixon R, Kahn D (2004) Gene ic egula ion o biological ni ogen ixa ion. Na Re Mic obiol 2:621–631. h ps://doi.o g/10.1038/n mic o954 F i z C, Lame s LPM, Riaz M, an den Be g LJL, Elzenga TJTM (2014) Sphagnum mosses—mas e s o e icien N-up ake while a oiding in oxica ion. PLoS ONE 9:1–11. h ps://doi.o g/10.1371/jou nal.pone.0079991 Go ham E (1991) No he n pea lands: ole in he ca bon cycle and p obable esponses o clima ic wa ming. Ecol Appl 1:182–195 G assho K, Johannse H (1972) New sensi i e and di ec me hod o au oma ic de e mina ion o ammonia in seawa e . J Cons 34:516–521 Ha dy RW, Hols en RD, Jackson EK, Bu ns RC (1968) The ace ylene-e hylene assay o N2 ixa ion: labo a o y and ield e alua ion. Plan Physiol 43:1185–1207 Hen iksen A (1965) An au oma ed me hod o de e mining low-le el concen- a ions o phospha e in esh and saline wa e s. Analys 90:29–34 Ho A, Bodelie PL (2015) Diazo ophic me hano ophs in pea lands: he missing link? Plan Soil 389:185–196. h ps://doi.o g/10.1007/ s11104-015-2393-9 Kamphake LJ, Hannah SA, Cohen JM (1967) Au oma ed analysis o ni a e by hyd azine educ ion. Wa e Res 1:205–216 Khadem AF, Pol A, Je en MSM, den Camp HJMO (2010) Ni ogen ixa ion by he e ucomic obial me hano oph “Me hylacidiphilum uma iolicum”SolV. Mic obiology 156:1052–1059 Kip N, an Winden JF, Pan Y, Bod ossy L, Reicha GJ, Smolde s AJP, Je en MSM, Dams é JSS, den Camp HJMO (2010) Global p e alence o me hane oxida ion by symbio ic bac e ia in pea -moss ecosys ems. Na Geosci 3:617–621 Klindwo h A, P uesse E, Schwee T, Peplies J, Quas C, Ho n M, Glöckne FO (2013) E alua ion o gene al 16S ibosomal RNA gene PCR p ime s o classical and nex -gene a ion sequencing-based di e si y s udies. Nucleic Acids Res 41:1–11. h ps://doi.o g/10.1093/na /gks808 Kno KH, Ho n MA, Bo ken W (2014) Signi ican non-symbio ic ni ogen ixa ion in Pa agonian omb o ophic bogs. Glob Change Biol. h ps://doi. o g/10.1111/gcb.12849 Kos ka JE, Wes on DJ, Glass JB, Lillesko EA, Shaw AJ, Tu e sky MR (2016) The Sphagnum mic obiome: new insigh s om an ancien plan lineage. New Phy ol 211:57–64. h ps://doi.o g/10.1111/nph.13993 Kox MAR, Lüke C, F i z C, an den Elzen E, Alen T, Op den Camp HJM, Lam- e s LPM, Je en MSM, E wig KF (2016) E ec s o ni ogen e iliza ion on diazo ophic ac i i y o mic oo ganisms associa ed wi h Sphag- num magellanicum. Plan Soil 406:83–100. h ps://doi.o g/10.1007/ s11104-016-2851-z Laine J (2004) Lakkasuo: a guide o mi e ecosys em. Me säekologian Lai os, Helsinki La mola T, Leppänen SM, Tui ila ES, Aa a M, Me ilä P, F i ze H, Tii ola M (2014) Me hano ophy induces ni ogen ixa ion du ing pea land de elop- men . P oc Na l Acad Sci USA 111:734–739. h ps://doi.o g/10.1073/ pnas.1314284111 Leppänen SM, Salemaa M, Smolande A, Mäkipää R, Tii ola M (2013) Ni ogen ixa ion and me hano ophy in o es mosses along a N deposi- ion g adien . En i on Exp Bo 90:62–69. h ps://doi.o g/10.1016/j. en expbo .2012.12.006 Leppänen SM, Rissanen AJ, Tii ola M (2014) Ni ogen ixa ion in Sphagnum mosses is a ec ed by moss species and wa e able le el. Plan Soil. h ps://doi.o g/10.1007/s11104-014-2356-6 Le S, Michelsen A (2014) Seasonal a ia ion in ni ogen ixa ion and e ec s o clima e change in a suba c ic hea h. Plan Soil 379:193–204. h ps://doi. o g/10.1007/s11104-014-2031-y Lindo Z, Nilsson MC, Gundale MJ (2013) B yophy e-cyanobac e ia associa ions as egula o s o he no he n la i ude ca bon balance in esponse o global change. Glob Change Biol 19:2022–2035. h ps://doi.o g/10.1111/ gcb.12175 Malme N, Albinsson C, S ensson BM, Wallén B (2003) In e e ences be ween Sphagnum and ascula plan s: e ec s on plan communi y s uc u e and pea o ma ion. Oikos 100:469–482 McMu die PJ, Holmes S (2013) Phyloseq: an R package o ep oducible in e ac i e analysis and g aphics o mic obiome census da a. PLoS ONE. h ps://doi.o g/10.1371/jou nal.pone.0061217 Pos ga e JR (1982) Biology ni ogen ixa ion: undamen als. Philos T ans R Soc B Biol Sci 296:375–385 Page 9 o 9 Kox e al. AMB Exp (2018) 8:76 Pu kinen A, La mola T, Tuomi i a T, Siljanen HMP, Bod ossy L, Tui ila ES, F i ze H (2014) Pea land succession induces a shi in he communi y composi- ion o Sphagnum-associa ed ac i e me hano ophs. FEMS Mic obiol Ecol 88:596–611. h ps://doi.o g/10.1111/1574-6941.12327 Quas C, P uesse E, Yilmaz P, Ge ken J, Schwee T, Ya za P, Peplies J, Glöckne FO (2013) The SILVA ibosomal RNA gene da abase p ojec : imp o ed da a p ocessing and web-based ools. Nucleic Acids Res 41:590–596. h ps:// doi.o g/10.1093/na /gks1219 R De elopmen Co e Team (2017) R: a language and en i onmen o s a is i- cal compu ing. R Founda ion o S a is ical Compu ing, Vienna Raghoeba sing AA, Smolde s AJP, Schmid MC, Rijps a WIC, Wol e s-A s M, De ksen J, Je en MSM, Schou en S, Dams é JSS, Lame s LPM, Roelo s JGM, den Camp HJMO, S ous M (2005) Me hano ophic symbion s p o ide ca bon o pho osyn hesis in pea bogs. Na u e 436:1153–1156. h ps://doi.o g/10.1038/na u e03802 Schloss PD, Wes co SL, Ryabin T, Hall JR, Ha mann M, Hollis e EB, Lesniewski RA, Oakley BB, Pa ks DH, Robinson CJ, Sahl JW, S es B, Thallinge GG, Van Ho n DJ, Webe CF (2009) In oducing mo hu : open-sou ce, pla o m- independen , communi y-suppo ed so wa e o desc ibing and com- pa ing mic obial communi ies. Appl En i on Mic obiol 75:7537–7541. h ps://doi.o g/10.1128/AEM.01541-09 Schloss PD, Ge e s D, Wes co SL (2011) Reducing he e ec s o PCR ampli- ica ion and sequencing a i ac s on 16s RNA-based s udies. PLoS ONE. h ps://doi.o g/10.1371/jou nal.pone.0027310 Shche bako AV, B agina AV, Kuzmina EY, Be g C, Mun yan AN, Maka o a NM, Mal ano a NV, Ca dinale M, Be g G, Chebo a VK, Tikhono ich IA (2013) Endophy ic bac e ia o Sphagnum mosses as p omising objec s o ag i- cul u al mic obiology. Mic obiology 82:306–315. h ps://doi.o g/10.1134/ S0026261713030107 an B eemen N (1995) How Sphagnum bogs down o he plan s. T ends Ecol E ol 10:270–275. h ps://doi.o g/10.1016/0169-5347(95)90007-1 an den Elzen E, Kox MAR, Ha penslage SF, Hensgens G, F i z C, Je en MSM, E wig KF, Lame s LPM (2017) Symbiosis e isi ed: phospho us and acid bu e ing s imula e N2 ixa ion bu no Sphagnum g ow h. Biogeo- sciences 14:1111–1122. h ps://doi.o g/10.5194/bg-14-1111-2017 an de Heijden MG, Ba dge RD, an S aalen NM (2008) The unseen majo i y: soil mic obes as d i e s o plan di e si y and p oduc i - i y in e es ial ecosys ems. Ecol Le 11:296–310. h ps://doi. o g/10.1111/j.1461-0248.2007.01139.x Vile MA, Wiede RK, Ži ko ić T, Sco KD, Vi DH, Ha sock JA, Iosue CL, Quinn JC, Pe ix M, Fillingim HM, Popma JM (2014) N2- ixa ion by me hano ophs sus ains ca bon and ni ogen accumula ion in p is ine pea lands. Biogeo- chemis y 121:317–328. h ps://doi.o g/10.1007/s10533-014-0019-6 Vi ousek P, Cassman K, Cle eland C (2002) Towa ds an ecological unde s and- ing o biological ni ogen ixa ion. Biogeochemis y 57(58):1–45 Vi ousek PM, Menge DNL, Reed SC, Cle eland CC (2013) Biological ni ogen ixa ion: a es, pa e ns and ecological con ols in e es ial ecosys ems. Philos T ans R Soc Lond B Biol Sci 368:20130119. h ps://doi.o g/10.1098/ s b.2013.0119 Wa en MJ, Lin X, Gaby JC, K e z CB, Kol on M, Mo on PL, Pe -Ridge J, Wes on DJ, Schad CW, Kos ka JE, Glass JB (2017) Molybdenum-based diazo- ophy in a Sphagnum pea land in no he n Minneso a. Appl En i on Mic obiol. h ps://doi.o g/10.1128/aem.01174-17 Wes on DJ, Timm CM, Walke AP, Gu L, Muche o W, Schmu z J, Shaw AJ, Tuskan GA, Wa en JM, Wullschlege SD (2015) Sphagnum physiology in he con- ex o changing clima e: eme gen in luences o genomics, modelling and hos -mic obiome in e ac ions on unde s anding ecosys em unc- ion. Plan Cell En i on 38:1737–1751. h ps://doi.o g/10.1111/pce.12458 Zeh JP, Jenkins BD, Sho SM, S ewa d GF (2003) Ni ogenase gene di e si y and mic obial communi y s uc u e: a c oss-sys em compa ison. En i on Mic obiol 5:539–554