The ruminal microbiome associated with methane emissions from ruminant livestock

REVIEW Open Access The uminal mic obiome associa ed wi h me hane emissions om uminan li es ock Ilma Tapio 1 , Timo hy J. Snelling 2 , F ancesco S ozzi 3 and R. John Wallace 2* Abs ac Me hane emissions om uminan li es ock con ibu e signi ican ly o he la ge en i onmen al oo p in o ag icul u e. The umen is he p incipal sou ce o me hane, and ce ain ea u es o he mic obiome a e associa ed wi h low/high me hane pheno ypes. Despi e hei p ima y ole in me hanogenesis, he abundance o a chaea has only a weak co ela ion wi h me hane emissions om indi idual animals. The composi ion o he a chaeal communi y appea s o ha e a s onge e ec , wi h animals ha bou ing he Me hanob e ibac e go schalkii clade ending o be associa ed wi h g ea e me hane emissions. Cilia e p o ozoa p oduce abundan H 2 , he main subs a e o me hanogenesis in he umen, and hei emo al (de auna ion) esul s in an a e age 11% lowe me hane emissions in i o, bu he esul s a e no consis en .Di e en p o ozoal gene a seem o esul in g ea e me hane emissions, hough communi y ypes (A, AB, B and O) did no di e . Wi hin he bac e ia, h ee di e en ‘ umino ypes’ha e been iden i ied, wo o which p edispose animals o ha e lowe me hane emissions. The wo low-me hane umino ypes a e gene ally cha ac e ized by less abundan H 2 -p oducing bac e ia. A lowe abundance o P o eobac e ia and di e ences in ce ain Bac e oide es and anae obic ungi seem o be associa ed wi h high me hane emissions. Rumen anae obic ungi p oduce abundan H 2 and o ma e, and hei abundance gene ally co esponds o he le el o me hane emissions. Thus, mic obiome analysis is consis en wi h known pa hways o H 2 p oduc ion and me hanogenesis, bu no ye in a p edic i e manne . The p oduc ion and u ilisa ion o o ma e by he uminal mic obio a is poo ly unde s ood and may be a sou ce o a iabili y be ween animals. Keywo ds: A chaea, Me hane, Mic obiome, Rumen Backg ound Me hane is a g eenhouse gas (GHG) wi h a global wa ming po en ial 28- old ha o ca bon dioxide [1]. Ag icul u e makes a signi ican con ibu ion o o al GHG p oduc ion, wi h es ima es a ying acco ding o coun y and calcula ion me hod [2]. None heless, a global con ibu ion o be ween 7 and 18% o o al an- h opogenic GHG emissions is gene ally accep ed [2]. Ruminan p oduc ion accoun s o abou 81% o GHG om he li es ock sec o (calcula ed om H is o e al. [2]), 90% o which esul s om umen mic obial me h- anogenesis [3]. Ruminal CH 4 p oduc ion also ep esen s a loss o ene gy ( om 2 o 12% o g oss ene gy in ake [4]), which could in p inciple o he wise be a ailable o animal g ow h o milk p oduc ion. Lowe ing CH 4 emissions he e o e would bene i he en i onmen and possibly he e iciency o li es ock p oduc ion. Mo e han 87% o he CH 4 p oduced by sheep has been es ima ed o be de i ed om he umen [5], whe e a popula ion o me hanogenic a chaea con e s he H 2 and CO 2 p oduced by a complex communi y o cilia e p o ozoa, bac e ia and anae obic ungi o CH 4 [6, 7]. A massi e wo ldwide esea ch e o has in es iga ed a ious mi iga ion s a egies. Changes in managemen p ac ices can be simple and e y e ec i e [2], while eed addi i es ha migh inhibi H 2 p oduc ion, p o ide an al e na i e me abolic H sink o inhibi he a chaea hemsel es o e oppo uni ies beyond hose s aigh o - wa d managemen changes [6–11]. O he oppo uni ies include chemogenomics and immuniza ion [12–14]. One s a egy ha is o emos in se e al in es iga ions is gene ic selec ion o he li es ock. I we can demons a e ha pe sis en ly di e en CH 4 emissions in di e en animals [14–16] can be explained by hei indi idual * Co espondence: [email p o ec ed] 2 Rowe Ins i u e o Nu i ion and Heal h, Uni e si y o Abe deen, Fo es e hill, Abe deen AB16 5BD, UK Full lis o au ho in o ma ion is a ailable a he end o he a icle © The Au ho (s). 2017 Open Access 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. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed. Tapio e al. Jou nal o Animal Science and Bio echnology (2017) 8:7 DOI 10.1186/s40104-017-0141-0 uminal mic obiomes, and ha he cha ac e is ic is he i able, i should be possible o selec u u e gene a- ions o uminan s ha ha e in insically lowe CH 4 emissions. All he s a egies po en ially in ol e changing he uminal mic obiome. The aim o his sho e iew is o assess ou cu en unde s anding o he ole o di e - en membe s o he mic obiome in de e mining he ex en o me hanogenesis in he umen. The umen mic obial communi y The umen is home o a as a ay o cilia e p o ozoa, anae obic ungi, anae obic bac e ia and a chaea. The p o ozoa can comp ise up o hal he umen mic obial biomass [17, 18], he ungi we e o iginally es ima ed o be abou 8% o he biomass [19] bu may each 20% in sheep [20], he a chaea comp ise 0.3–4% [21] and he bac e ia o m he emainde , ypically he la ges compo- nen o he mic obial biomass. Ou p esen unde s and- ing o uminal mic obiology was buil ini ially upon a ew epoch-changing ad ances made many yea s ago: G uby & Dela ond’s [22] mic oscopic obse a ions o p o ozoa; Hunga e’s [23] app ecia ion o he anae obic na u e o he umen ha led o new, uly anae obic cul- u e echniques o he bac e ia; O pin’s [24] ealiza ion ha some lagella e p o ozoa we e in ac zoospo es o anae obic ungi, un il hen a con adic ion in e ms. The isola ion and s udy o pu e cul u es was and emains in- aluable in unde s anding he likely ole o di e en species o bac e ia, p o ozoa o ungi in he o e all e men a ion. D awbacks o cul i a ion echniques a e ha only a e y small numbe o samples can be es ed, and ha hey su e om bias, whe eby he composi ion o he g ow h medium, gene ally oo ich, de e mines which species can g ow [25]. De elopmen o molecula echniques, based mainly on ssu RNA gene and in e - genic space sequence ( o he ungi) analyses, opened new oppo uni ies in umen esea ch. Cloning and sequencing p o ided communi y analyses ha we e no p one o he biases imposed by cul i a ion echniques, al hough di e en bias was in oduced by o he ac o s, like s o age condi ions [26], he di e en ial e iciency o DNA ex ac ion om di e en species and ampli ica ion bias [27–29]. Rela ed echniques o mic obiome analysis quickly ollowed (DGGE, TGGE, T-RFLP, ARISA). Quan i a i e PCR and FISH enabled mic obial g oups o species o be quan i ied [30]. Now, me agenomic se- quencing enables apid communi y analysis o be ca ied ou , wi hou he cul i a ion bias o a ia ion associa ed wi h p ime selec ion o PCR ampli ica ion i egula i ies [25, 31]. The p oblem o DNA ex ac ion emains, how- e e , and da abases a e ela i ely weak whe e uminal o - ganisms a e conce ned [32]. Ne e heless, i we can use his app oach o de e mine how he unc ional ac i i y o he umen mic obial communi y in luences me hane emissions, he knowledge should enable s a egies o dec ease he en i onmen al impac o li es ock ag icul- u e. Fu he mo e, i migh be expec ed o imp o e animal p oduc ion e iciency. Ruminal communi y analysis ela ing o me hane emissions A chaea The e a e wo main ou es o me hanogenesis in he umen, bo h ca ied ou by a chaea. The hyd ogeno- ophic pa hway con e s H 2 and CO 2 p oduced by he p o ozoa, bac e ia and ungi o CH 4 [3, 6]. I is usually assumed ha o ma e, which can be used by all he mos abundan uminal a chaea, is equi alen o H 2 +CO 2 ,so o ma e is included in he hyd ogeno ophic ca ego y [21, 33]. A second ca ego y o subs a e o me hanogen- esis is me hyl g oups, such as hose p esen in me hyl- amines and me hanol [34, 35]. Me hylamines a e de i ed om glycine be aine ( om bee ) and choline ( om plan memb anes), while me hanol is de i ed om he hyd olysis o me hanolic side-g oups in plan polysac- cha ides. The mos common hyd ogeno ophic a chaea a e om he genus Me hanob e ibac e , which has been di ided in o wo subg oups, one known as he SGMT clade (Mbb. smi hii,Mbb. go schalkii,Mbb. mille ae and Mbb. haue i), he o he (RO) clade comp ising p incipally Mbb. uminan ium and Mbb. olleyae [21, 36]. O he signi ican hyd ogeno ophic gene a include Me hanosphae a, Me hanimic ococcus and Me hanobac- e ium. The less abundan me hylo ophs (Me hanosa ci- nales, Me hanosphae a, Me hanomassiliicoccaceae) can use me hylamines and me hanol, and he e a e a chaea (Me hanosa cinales) ha p oduce me hane ia he ace i- clas ic pa hway ( e iewed in Mo ga i e al. [7]). Rumen me hanogenic a chaeal di e si y is es ic ed o ou o de s [21] and is highly conse ed ac oss 32 uminan species collec ed wo ldwide [32]. In ui i ely, a chaea should be he mic obial g oup mos closely co ela ed wi h me hane emissions. How- e e , some s udies ha e shown no such co ela ion wi h hei o e all abundance while in o he s he co ela ion has been weak. Mo ga i e al. [37], Zhou e al. [38], Danielsson e al. [39] and Danielsson [40] ound no co espondence be ween he numbe s o me hanogens and me hane emissions om dai y cows when measu ed using me agenomics and qPCR echniques. Ki elmann e al. [41] and Shi e al. [42] o med a simila conclusion in sheep. A weak co ela ion be ween a chaeal abun- dance ela i e o bac e ia was ound in bee s ee s [43] bu none was ound wi h dai y cows in he RuminOmics p ojec [h p://www. uminomics.eu/] when exp essed as he a chaea:bac e ia a io (Fig. 1). Shi e al. [42] also obse ed ha a chaeal gene exp ession a he han gene abundance was co ela ed o me hane emissions om Tapio e al. Jou nal o Animal Science and Bio echnology (2017) 8:7 Page 2 o 11 indi idual sheep. I is easy o see why gene exp ession migh be a use ul p oxy o me hanogenesis in a s a ic sys em like soil [44], bu less so in a lowing sys em like he umen, whe e o physiological easons biomass mus be di ec ly co ela ed o gene abundance unless o he p ocesses, such as uncoupled CH 4 p oduc ion occu [45]. Gi en he high a iabili y o he ela ionship wi h o e - all a chaeal abundance, i may be ha he composi ion o he a chaeal communi y a he han jus i s size may ha e g ea e signi icance wi h ega d o me hane emis- sions. Zhou e al. [38], Danielsson e al. [39], Shi e al. [42] and Danielsson [40] all ound a posi i e co ela ion be ween he ela i e abundance o Me hanob e ibac e SGMT clade and me hane emissions. Danielsson [40] in e p e ed his co ela ion in e ms o di e en a ini ies o H 2 in he wo g oups, wi h he SGMT clade posses- sing me hyl coenzyme M educ ase isozymes Mc I and Mc II [12], which enables he a chaea o u ilise H 2 a highe concen a ions, agains he RO clade ha possess only Mc I [3, 12]. The dynamics o he o he a chaeal communi y composi ion and hus he e iciency o H 2 u iliza ion would in u n would be a consequence o di e ing H 2 p oduc ion by di e en bac e ia [33, 41] and p esumably also p o ozoal and ungal communi ies. Fu he mo e, he p opo ion o Me hanosphae a spp. in o al a chaea was nega i ely associa ed wi h me hane p oduc ion in sheep [41], al hough no in bee ca le [46]. Thus, di e ing me hane emissions a e a leas pa ly due o a ying ela i e abundances wi hin he communi y o me hanogenic a chaea. O he obse a ions ega ding he a chaeal communi y, some imes called he a chaeome, include hose o Pi a e al. [47], who ound ha a chaeal abundance inc eased in s ee s su e ing o hy bloa , and Pei e al. [48], who disco e ed a chaea associa ed wi h he umen epi he- lium. In he o me case, he CH 4 con en o he gas was no measu ed, so i is unclea he impac he bloa would ha e on me hanogenesis. In he la e , he inding was su p ising because he umen wall is conside ed o be an ae obic/anae obic in e ace, and he ela i e abun- dance o O 2 migh be conside ed o supp ess he g ow h o he ex emely O 2 -sensi i e me hanogens. In ac , one migh ha e possibly expec ed CH 4 oxidise s o be p esen , in spi e o hei absence om he deep uminal diges a [49]. Cilia e p o ozoa Ruminal cilia es a e in ima ely in ol ed in me hanogen- esis, pa ly ia hei abundan H 2 p oduc ion [50] and, aking ad an age o his, hei associa ed me hanogens, which a e ound bo h as in acy oplasmic commensals and on he ex e io su ace o he p o ozoa [3, 18, 51–53]. Se e al s udies sugges ed a co ela ion be ween he abundance o p o ozoa and me hane emissions (colla ed in [18, 54, 55]), while o he s do no [37, 43]. Guyade e al. [56] conduc ed a me a-analysis con aining 28 expe imen s and 91 ea men s. This me a-analysis showed a linea posi i e ela ionship be ween log 10 p o ozoal numbe s and me hane emissions exp essed pe uni DMI. An = 0.96 showed ha he e is indeed a easonably s ong ela ionship (Fig. 2). Fig. 1 A chaea:bac e ia ela i e abundance in ela ion o me hane emissions, p elimina y da a om he 1000-cow RuminOmics p ojec . Dai y cows on di e en a ms h oughou Eu ope ecei ed g ass o maize silage:concen a e die s o simila nu ien composi ion. Feed in ake was measu ed ei he di ec ly o calcula ed om aecal long-chain hyd oca bons. Samples o umen con en s we e emo ed by s omach ube and DNA was ex ac ed by he Yu & Mo ison me hod [110]. Abundances we e calcula ed om qPCR o 16S RNA genes using uni e sal p ime s o a chaea and bac e ia Tapio e al. Jou nal o Animal Science and Bio echnology (2017) 8:7 Page 3 o 11 De auna ion ( he emo al o he cilia es om he umen) has he e o e been in es iga ed in ela ion o me hane p oduc ion. Al hough in some cases he esul s o de auna ion on CH 4 emissions ha e no been encou - aging [57–60], Newbold e al. [18] ca ied ou a me a- analysis o de auna ion s udies and concluded ha CH 4 was dec eased on a e age by 11%. Despi e he lowe CH 4 p oduc ion, he o al a chaeal abundance was no signi i- can ly dec eased in he Newbold e al. me a-analysis, sugges ing ha he a chaeal communi y in de auna ed animals may ha e a lowe CH 4 -emi ing speci ic ac i i y han ha o he p o ozoa-associa ed communi y. As wi h he a chaea, he ques ions hen e e o whe he some indi idual p o ozoal gene a o species, and hei associa ed a chaea, a e mo e linked wi h me h- anogenesis han o he s. In gene al, he p o ozoa ha bou an a chaeal popula ion ha , like he gene al a chaeal communi y, is domina ed by Me hanob e ibac e spp. [61–64], al hough di e ences we e obse ed in he abundance o di e en a chaea ound in he p o ozoa and in he non-associa ed a chaea [18, 61, 65] ha migh lead o di e en me hanogenic speci ic ac i i ies in he wo popula ions. Fu he mo e, a chaeal colonisa ion abundance may di e be ween di e en p o ozoal spe- cies [51] and each may be associa ed wi h di e en p edominan a chaeal gene a/species. Holo ichs in pa - icula had an a chaeal communi y ha di e ed om en odiniomo phid p o ozoa [53]. La ge cilia es appea o be mo e hea ily colonized by me hanogens han smalle cilia es [53, 66], and also by bac e ia, sugges ing ha he e is no a selec i e colonisa ion by a chaea [53]. The lowe me abolic ac i i y in e ms o H 2 p oduc ion o he la ge p o ozoal species pe uni biomass [50, 54, 58] p esumably explains ha smalle p o ozoa, and hei associa ed a chaea, will be ela i ely mo e ac i e in me hanogenesis han la ge species. Indeed, in i o s udies indica ed ha he smalle En odinium spp. we e mo e associa ed wi h me hane p oduc ion han la ge species like Polyplas on mul i esicula um [50, 58]. In i o s udies a e inconsis en , howe e . Re auna ion expe imen s indica ed ha he abundance o En odinium spp. [67, 68] o holo ichs [68] co ela ed wi h highe me hane emissions. A la ge amplicon sequencing s udy in sheep ne e heless ound no ela ionship be ween he ela i e abundance o di e en cilia es and me hane emissions [41]. Fu he mo e, cilia e communi ies all in o a small numbe o ypes (A, AB, B and O [69]) depending on in e ac ions, p incipally in e -species p eda ion. Despi e he la ge di e ences in ela i e abundance o di e en p o ozoa ypes in he di e en communi y ypes, me hane emissions could no be co ela ed wi h p o ozoal communi y s uc u e [70]. The a ying colonisa ion by a chaea depending on he ime a e eeding [71] is ano he con ounding ac o in ying o e alua e he ole o p o ozoa in me hanogenesis. Bac e ia Ruminal bac e ia o m he mos di e se g oup wi hin he umen, capable o u ilizing ib e, s a ch, p o ein and suga s [72]. Among nume ous bac e ial phyla ound in di e en s udies, Fi micu es, Bac e oide es and P o eo- bac e ia a e he mos abundan [32]. Fib oly ic bac e ia, especially celluloly ic Ruminococcus and se e al Eubac- e ium spp (Fi micu es), a e well s udied H 2 p oduce s. On he o he hand, he p ominen celluloly ic genus, Fib obac e , does no p oduce H 2 , while Bac e oide es a e ne H 2 u ilize s [72]. Mic obiome analysis has iden i- ied h ee di e en ‘ umino ypes’ ha seemed o be associa ed wi h a ia ions in me hane p oduc ion by sheep [41]. The low-CH 4 p oduc ion umino ype Q was cha ac- e ised by high ela i e abundances o he p opiona e- p oducing Quinella o alis.Low-CH 4 umino ype S had Fig. 2 Rela ionship be ween me hane emission and umen p o ozoa concen a ion in a me a-analysis o 28 di e en expe imen s. The black dashed line ep esen s he a e age wi hin-expe imen ela ionship. Rep oduced om [56] wi h pe mission Tapio e al. Jou nal o Animal Science and Bio echnology (2017) 8:7 Page 4 o 11 highe abundances o lac a e- and succina e-p oducing Fib obac e spp., Kandle ia i ulina,Olsenella spp., P e o- ella b yan ii,andSha pea azabuensis. The high-CH 4 p oduc ion umino ype H had highe ela i e abundances o species belonging o Ruminococcus, o he Ruminococ- caceae, Lachnospi aceae, Ca abac e iaceae, Cop ococcus, o he Clos idiales, P e o ella, o he Bac e oidales, and Alphap o eobac e ia. The o e all in e p e a ion would be ha me hane emissions depend on he abundance o he H 2 -p oducing bac e ia p esen ; a co olla y o his is he obse a ion ha chemical inhibi ion o me hanogenesis in goa s led o inc eases in he abundance o H 2 -consuming P e o ella and Selenomonas spp.[73]. P o eobac e ia we e 4- old less abundan (2.7 s. 11.2% o bac e ia) in high emi ing bee ca le [46] and a simila inding was made in dai y cows [40]. The dominan amily among P o eobac- e ia was Succini ib ionaceae. This inding seems o pa allel he high numbe s o Succini ib ionaceae in he Tamma wallaby [74], which, like he uminan , is a he bi o ous o egu e men e . I p oduces only abou one- i h o he me hane pe uni o eed in ake o umi- nan s, which is a ibu ed o he la ge communi y o Succino ib ionaceae. An in iguing addi ional obse a ion common o hese s udies [40, 41] was ha wi hin di e en P e o ella OTUs, some we e co ela ed wi h a high CH 4 pheno ype, while o he s we e associa ed wi h low emis- sions. The di e en OTUs seem o clus e oge he (Fig. 3), sugges ing unc ional e sa ili y wi hin he P e o ella genus. Fu he in es iga ion o he pheno- ypes o hese dominan uminal bac e ia is needed, which may well p o ide clues o u u e exploi a ion, pa icula ly as some P e o ella a e epo ed o p oduce o ma e [72]. In a dai y ca le s udy [75] wi h wo CH 4 -mi iga ing eed addi i es, g apema c and a combina ion o lipids and annins, i was ound ha he mic obiome di e ed om he con ol die in a simila way. Faecalibac e ium p ausni zii was o e - ep esen ed in he low- CH 4 die s, and o he mic obiome ma ke s ha could be p edic i e o low-CH 4 pheno ypes we e iden i ied. F. p ausni zii is a bac e ial species ha is abundan in he human colon [76] bu is seldom men ioned in he con ex o he umen. I may p o e a use ul ma ke , bu i is no ob ious how i s p ope ies could be mechanis ically connec ed o he low-CH 4 pheno ype. Anae obic ungi The anae obic ungi, like he p o ozoa, p oduce abun- dan amoun s o H 2 , along wi h CO 2 , o ma e and ace a e as me abolic end p oduc s [77]. Six ungal gene a ha e been de ec ed in he umen bu ecen molecula esea ch sugges s exis ence o se e al new axa [78], wi h unc ions s ill o be unde s ood. Me hanogens a e ound in close associa ion wi h ungal hyphae [79]. Al hough he e is eason o suppose ha ungal abundance migh be ela ed o me hane emissions, epo s a e ew. Ki elmann e al. [41] no ed no di e ence in ungal com- muni y s uc u e in ela ion o me hane emissions om sheep. In he RuminOmics p ojec , howe e , p elimina y esul s sugges ha wo ungal species, Caecomyces communis and Neocallimas ix on alis, a e nega i ely ela ed o me hanogenesis ( = -0.50 and -0.45, P< 0.001; R.J. Wallace e al., unpublished]. The me a-analysis o Newbold e al. [18] no ed ha one o la ges e ec s o de auna ion, which leads o lowe CH 4 p oduc ion, was a dec ease in ungal abundance. Whe he his dec ease is a majo o di ec cause o lowe CH 4 p oduc ion in de auna ed animals is unclea . Gene al conside a ions on a ia ions in me hanogenesis and he mic obiome Con ibu ion o non-hyd ogeno ophic me hanogenesis The main subs a es o me hanogenesis in he umen a e known o be H 2 +CO 2 , o ma e and compounds con aining me hyl g oups like he me hylamines and me hanol [21]. In he e iews al eady men ioned he e, o ma e and H 2 +CO 2 a e usually conside ed o be equi alen as subs a es o me hanogenesis and o ma e is no ea ed sepa a ely. Fo ma e eeds di ec ly in o he me hanogenesis pa hway a he e y beginning ia o - ma e dehyd ogenase [80]. Hunga e e al. [81] es ima ed ha 18% o me hane was o med ia o ma e a he han H 2 +CO 2 . Ye he e a e some impo an aspec s o o ma e me abolism abou which ou unde s anding is incomple e. The ela ionship be ween bac e ial abun- dances om mic obiome es ima es, abo e, was discussed in ela ion o whe he bac e ia o m H 2 , as in o he ana- lyses [33, 41, 43, 46], wi h li le indica ion abou o ma e p oduce s. The e is a la ge unce ain y abou bac e ial o ma e p oduc ion, e lec ed in he summa y ables o S ewa e al. [72]. Al hough many species p oduce some o ma e, p ecise amoun s a e no known and he e o e he impo ance o his p oduc ion is di icul o es ima e. Pe haps he Hunga e 1000 collec ion (www. mgne wo k.o g/ hunga e1000.h ml) could be used as a esou ce o make such measu emen s. A p esen , he Hunga e 1000 p ojec has i s emphasis on s eng hening gene ic da abases [3], bu much pheno ypic in o ma ion is being collec ed alongside he main h us o he p ojec . Assessing bac e ial o ma e p oduc ion is u he complica ed by he knowledge ha co-cul u e expe imen s demons a e ha he me abolism o some bac e ia and ungi g own in he p esence o me hanogens can be pulled in he di ec ion o H 2 o o ma e p oduc ion [82–85], so i is e y di icul o be su e wha he ole o di e en species migh be in he mixed umen communi y. And pe haps mos c ucially, me hanogenesis is no he sole a e o o ma e in he umen. Hunga e e al. [81] no ed o ma e u ilisa ion in he absence o Tapio e al. Jou nal o Animal Science and Bio echnology (2017) 8:7 Page 5 o 11 Fig. 3 Neighbo Joining ee o P e o ella-like OTUs ha had a nega i e (blue do s) o posi i e ( ed do s) ela ion o me hane (exp essed in e ms o g me hane/kg DMI) in he 1,000-cow RuminOmics p ojec . Mul iple alignmen was done using MUSCLE [111]. The Neighbo Joining ee was cons uc ed using p-dis ance and pai wise-dele ion pa ame e s. The ee was esampled1,000 imesandboo s ap aluesa eindica ed.The linea ized ee was compu ed using MEGA 5.1 [112] by using mos abundan Bac e oidales OTUs o c ea e an “ou g oup” Tapio e al. Jou nal o Animal Science and Bio echnology (2017) 8:7 Page 6 o 11 me hanogenesis, p esumably by bac e ia. Species like Wolinella succinogenes use o ma e as an ene gy sou ce [72]. So, al hough i is usually s a ed ha u- minal a chaea u ilise ei he H 2 +CO 2 o o ma e [3], i is unclea whe he hey a e indeed equi alen o di e en a chaea. Fo example, in co-cul u es be ween umen anae obic ungi and h ee me hanogens, all he me hanogens used H 2 bu o ma e was only u ilised simul aneously by M. smi hii [86]. The di e en ial exp es- sion o o ma e dehyd ogenase was one o he la ges di e ences be ween high- and low-emi ing sheep [42]. The o ma e dehyd ogenase o M. uminan ium M1 was induced by co-cul u e wi h he o ma e-p oducing Bu y i- ib io p o eoclas icus [12]. Thus he e a e se e al easons o conclude ha hinking abou o ma e as a subs a e in he con ex o mic obiomes di e ing in hei me hano- genic ac i i y migh p o e ui ul. Fu he mo e, despi e he emphasis on H 2 p oduced by cilia e p o ozoa, he quan i y o o ma e p oduced seems o be many imes g ea e han H 2 [65]. The me hylamines and me hanol a e me hyl dono s o me hanogenesis by me hylo ophic a chaea, as desc ibed abo e. Thei con ibu ion o me hanogenesis will depend o some ex en on he concen a ion o me hylamines in he die [34, 35]. Bu how e icien is he p ocess? A e me hylamines con e ed quan i a i ely o CH 4 , and a e me hylamine, dime hylamine and ime hylamine equi alen in ha espec ? I is possible ha a ia ion in CH 4 emissions be ween indi idual animals on some die s may be due o di e en e icien- cies whe eby me hylamines a e eleased om eed ma e- ials and con e ed o CH 4 . One o he mo e su p ising indings in he Mbb. uminan ium M1 genome was he p esence o h ee genes encoding alcohol dehyd ogenase [12]. I has been demon- s a ed ha e hanol can be used as a C sou ce, bu no as sole C sou ce [3]. Thus, he a ailabili y o e hanol om bac e ial e men a ion may in luence he dependence o a chaea on me hanogenesis o ATP p oduc ion, and he e o e a ec he quan i y o CH 4 p oduced. In luence o die and mi iga ion measu es An impo an p inciple unde lying his e iew is ha some mic obiomes lead o di e en CH 4 emissions when o he ac o s emain cons an . Thus, key membe s o he mic obiome leading o high o low emissions should be able o be iden i ied. In he RuminOmics p ojec , all dai y cows ecei ed die s ha we e as nu i ionally simila as was possible gi en he di e en loca ions. Only by keeping as many o he ac o s as possible unchanged will i be possible o dissec he ole o di e en membe s o he mic obial communi y in de e mining low- and high- emi ing indi iduals. I should be no ed he e ha we ha e chosen o exp ess CH 4 p oduc ion in e ms o DMI, o he simple eason ha i makes i easie o iden i y a low-CH 4 mic obiome a he han a mic obiome ha o ms less CH 4 only because he hos animal ea s less. The esul s o mic obiome analysis so a we e expec ed in some espec s, in he sense ha die s high in s a ch con en a e known o lead o lowe me hane emissions, because s a ch u ilising bac e ia end o p oduce less H 2 han o he s, o example [33, 72]. In a simila way, he changed e men a ion s oichiome y linked wi h me hane emissions is a e y long es ablished obse a ion [87, 88]. New ques ions ha e been highligh ed ega ding di e en species associa ed wi h high and low CH 4 emissions unde simila condi ions. Unexpec ed co ela ions ha e been ound. Bu many ques ions emain. I is also wo h no ing ha widely di e en axa may ha e simila me abolic ac i i ies [89], so he e a e se e al di e en mic obio a ha could lead o simila me abolic p ope ies. Mi iga ion measu es ha e been desc ibed comp ehen- si ely elsewhe e [2, 3, 6–10]. Pe haps he mos p omising o hese is 3-ni ooxyp opanol, a molecule ob ained a ionally by i s s uc u al simila i y o me hyl-CoM [90–92]. As ye we do no know he ull implica ions o 3-ni ooxyp opanol, bu encou agemen can be ob ained ha he conce n ha H 2 accumula ion migh inhibi o e all e men a ion does no seem o be such a p oblem as was sugges ed by some in i o expe imen s [33, 93]. I is also wo h no ing ha a 50% educ ion in he g ow h a e o me hanogens would be su icien o cause hei washou om he umen [3, 33]. Comple e inhibi ion o g ow h is he e o e no necessa y. Me hane and eed e iciency CH 4 p oduc ion and eed e iciency a e linked, in he sense ha a low eed e iciency, exp essed as esidual eed in ake (RFI), is accompanied by lowe CH 4 p oduc ion [94–96]. The e e se does no apply, howe e , as has been ound in dai y cows in he RuminOmics p ojec . The indings ha he abundance o ce ain P e o ella changes acco ding o eed e iciency in bee ca le [97, 98] and many o he axa change in abundance [98] u he empha- sises ou need o unde s and he ole o P e o ella and i s di e en bio ypes on uminal e men a ion and me hano- genesis. Shaba e al. [99] disco e ed ha Megasphae a elsdenii was mo e abundan in low-e iciency cows, as we e genes o he ac yla e pa hway, used by M. elsdenii in p opiona e o ma ion. The explana ion o lowe e iciency was ha M. elsdenii in oduced a ype o u ile cycle in he p oduc ion and subsequen u ilisa ion o lac a e, an ene ge ically ine icien p ocess. The in luence o he hos animal Many esea che s belie e, and some s udies a e begin- ning o show, ha he hos animal exe s a con olling Tapio e al. Jou nal o Animal Science and Bio echnology (2017) 8:7 Page 7 o 11 e ec on i s own gu mic obio a [100–102]. The mech- anism could concei ably be a a molecula le el, pe haps ia complex in e ac ions wi h ecep o s in he umen wall [103, 104] o an ibodies in sali a [3, 105, 106]. Mo e likely, howe e , is ha he physical s uc u e and dynam- ics o gu diges a a e di e en in di e en animals. Goopy e al [15] ound ha lowe me hanogenesis in sheep was he i able and accompanied by he animals’ ha ing smalle umen olumes and he e o e al e ed luxes o nu ien s h ough he ac . This would ha e he e ec ha less eed would be e men ed in he umen, leading o lowe me hanogenesis. Va ia ions in sali a p oduc ion could lead o a simila esul [107]. Bo h would likely in luence he uminal mic obiome. The e o e, cau ion should be exe cised in in e p e ing mic obiome analyses – he changed mic obiome may be associa ed wi h, bu no cause, a dec ease in me hanogenesis. Ross e al. [108] ound good co ela ions be ween CH 4 emissions and he b oad cha ac e is ics o he mic o- biome. Now, me agenomics has shown ha he abundance o ce ain g oups o mic obial genes can be highly p edic i e o CH 4 emissions [46, 109] and eed e iciency [99]. Fo example, 20 mic obial genes ex- plained 81% o a ia ion in CH 4 emissions om bee ca le, while 49 genes explained 86% o a ia ion in RFI [109]. Fu he mo e, he animal’s gene ic backg ound was a ac o in de e mining hese gene abundances [109]. This is he ea ly phase o wha is su e o be a e ile a ea in which animal-mic obiome-emissions can be delin- ea ed by me agenomics p o iling, and animal b eeding based on hese gene abundances may lead o animals wi h lowe CH 4 emissions. Conclusions Recen la ge scale p ojec s such as he Global Rumen Census, he Hunga e 1000 and RuminOmics, om which some p elimina y esul s a e p esen ed he e, ha e p o ided new dep h o insigh in o he composi ion and unc ion o he umen mic obial communi y. By e eal- ing he some o he ela ionships be ween he mic o- biome and he animal pheno ype, hey ha e shown how unde s anding he ole o he umen mic obio a can help in he e o s o educe he en i onmen al impac o li es ock ag icul u e, in pa icula wi h he amelio - a ion o g eenhouse gas emissions. The a chaea ha e been he main a ge o esea ch, being di ec ly associa ed wi h me hane p oduc ion in he umen. How- e e , o he majo mic obial g oups such as he cilia e p o ozoa, he anae obic ungi, Succino ib ionaceae and P e o ella, among o he s, ha e shown o be associa ed wi h bo h high and low me hane p oduc ion. The esul s illus a e ha he e a e basic pheno ypic cha ac e is ics, such as o ma e me abolism, ha a e insu icien ly unde s ood. When placed in he con ex o he many as ye uncul i a ed mic obial species o he umen, i becomes clea ha he powe ul ool o molecula analysis mus be accompanied by cul u al and me abolic/pheno- ypic analysis i we a e o uly unde s and he ela ion be ween he uminal mic obiome and me hanogenesis. Abb e ia ions ARISA: Au oma ed ibosomal in e genic space analysis; DGGE: Densi y g adien gel elec opho esis; FISH: Fluo escence in i o hyb idiza ion; qPCR: Quan i a i e polyme ase chain eac ion; TGGE: Tempe a u e g adien gel elec opho esis; T-RFLP: Res ic ion agmen leng h polymo phism Acknowledgemen s The au ho s g a e ully acknowledge he wo k o , and discussions wi h, o he membe s o he RuminOmics conso ium and hank hem o hei pe mission o use p elimina y da a o illus a e his a icle. Funding The Rowe Ins i u e is unded by he Ru al and En i onmen Science and Analy ical Se ices Di ision (RESAS) o he Sco ish Go e nmen . This s udy was inancially suppo ed by Ruminomics (p ojec no. 289319 o EC 7 h F amewo k P og amme: Food, Ag icul u e, Fishe ies and Bio echnology). A ailabili y o da a and ma e ials The au ho s a e bound by he Collabo a ion Ag eemen eached by he RuminOmics conso ium, in which all da a gene a ed in he p ojec become eely a ailable in Janua y 2018, bu canno be eleased wi hou consen o he conso ium be o e hen. The o iginal da a unde pinning Fig. 1 and Fig. 3 a e subjec o hese condi ions. Indi idual applica ions o elease he da a be o e 2018 should be di ec ed o he co esponding au ho . Au ho s’con ibu ions The au ho s w o e he manusc ip oge he , RJW ha ing ini ia ed he p ojec . All au ho s ead and app o ed he inal manusc ip . Compe ing in e es s The au ho s decla e ha hey ha e no compe ing in e es s. Consen o publica ion All au ho s ha e gi en hei consen o submission o his a icle. The submission has also been app o ed by he publica ions commi ee o he RuminOmics conso ium. E hics app o al Animal expe imen a ion was conduc ed using p o ocols complying wi h na ional legisla ion and app o ed by local e hics commi ees. Au ho de ails 1 G een Technology, Na u al Resou ces Ins i u e Finland, Jokioinen, Finland. 2 Rowe Ins i u e o Nu i ion and Heal h, Uni e si y o Abe deen, Fo es e hill, Abe deen AB16 5BD, UK. 3 PTP, Via Eins ein - Loc. Cascina Codazza, 26900 Lodi, I aly. Recei ed: 31 July 2016 Accep ed: 3 Janua y 2017 Re e ences 1. IPCC. Clima e change 2014: syn hesis epo . In: Pachau i RK, Meye LA, edi o s. Con ibu ion o wo king g oups I, II and III o he i h assessmen epo o he in e go e nmen al panel on clima e change. Gene a: IPCC; 2014. p. 151. 2. H is o AN, Oh J, Lee C, Meinen R, Mon es F, O F, e al. Mi iga ion o g eenhouse gas emissions in li es ock p oduc ion. In: Ge be PJ, Hende son B, Makka HPS, edi o s. A e iew o op ions o non-CO 2 emissions. Rome: FAO; 2013. p. 226. 3. McAllis e TA, Meale SJ, Valle E, Guan LL, Zhou M, Kelly WJ, e al. Use o genomics and ansc ip omics o iden i y s a egies o lowe uminal me hanogenesis. J Anim Sci. 2015;93:1431–49. 4. Johnson KA, Johnson DE. Me hane emissions om ca le. J Anim Sci. 1995;73:2483–92. Tapio e al. Jou nal o Animal Science and Bio echnology (2017) 8:7 Page 8 o 11 5. Mu ay RM, B yan AM, Leng RA. Ra es o p oduc ion o me hane in he umen and la ge in es ine o sheep. B J Nu . 1976;36:1–14. 6. Ma in C, Mo ga i DP, Do eau M. Me hane mi iga ion in uminan s: om mic obe o he a m scale. Animal. 2010;4:351–65. 7. Mo ga i DP, Fo ano E, Ma in C, Newbold CJ. Mic obial ecosys em and me hanogenesis in uminan s. Animal. 2010;4:1024–36. 8. Knapp JR, Lau GL, Vadas PA, Weiss WP, T ica ico JM. In i ed e iew: en e ic me hane in dai y ca le p oduc ion: quan i ying he oppo uni ies and impac o educing emissions. J Dai y Sci. 2014;97:3231–61. 9. Kuma S, Choudhu y PK, Ca o MD, G i i h GW, Daga SS, Puniya M, e al. New aspec s and s a egies o me hane mi iga ion om uminan s. Appl Mic obiol Bio echnol. 2014;98:31–44. 10. Beauchemin KA, K euze M, O’Ma a F, McAllis e TA. Nu i ional managemen o en e ic me hane aba emen : a e iew. Aus J Exp Ag ic. 2008;48:21–7. 11. A wood GT, Al e mann E, Kelly WJ, Leahy SC, Zhang L, Mo ison M. Explo ing umen me hanogen genomes o iden i y a ge s o me hane mi iga ion s a egies. Anim Feed Sci Technol. 2011;166-67:65–75. 12. Leahy SC, Kelly WJ, Al e mann E, Ronimus RS, Yeoman CJ, Pacheco DM, e al. The genome sequence o he umen me hanogen Me hanob e ibac e uminan ium e eals new possibili ies o con olling uminan me hane emissions. Plos One. 2010;5(1):e8926. 13. W igh ADG, Kennedy P, O’Neill CJ, Too ey AF, Popo ski S, Rea SM, e al. Reducing me hane emissions in sheep by immuniza ion agains umen me hanogens. Vaccine. 2004;22:3976–85. 14. Pina es-Pa ino CS, Hickey SM, Young EA, Dodds KG, MacLean S, Molano G, e al. He i abili y es ima es o me hane emissions om sheep. Animal. 2013;7:316–21. 15. Goopy JP, Robinson DL, Woodga e RT, Donaldson AJ, Oddy VH, Ve coe PE, e al. Es ima es o epea abili y and he i abili y o me hane p oduc ion in sheep using po able accumula ion chambe s. Anim P od Sci. 2015;56:116-22. h p://dx.doi.o g/10.1071/AN13370. 16. de Haas Y, Windig JJ, Calus MPL, Dijks a J, de Haan M, Bannink A, e al. Gene ic pa ame e s o p edic ed me hane p oduc ion and po en ial o educing en e ic emissions h ough genomic selec ion. J Dai y Sci. 2011;94:6122–34. 17. Hunga e RE. The umen and i s mic obes. New Yo k: Academic; 1966. 18. Newbold CJ, de la Fuen e G, Belanche A, Ramos-Mo ales E, McEwan NR. The ole o cilia e p o ozoa in he umen. F on Mic obiol. 2015;6:1313. 19. O pin CG. Fungi in uminan deg ada ion. In: Ag icul u al science semina : deg ada ion o plan cell wall ma e ial. London: Ag icul u al Resea ch Council; 1981. p. 129–50. 20. Rezaeian M, Beakes GW, Pa ke DS. Dis ibu ion and es ima ion o anae obic zoospo ic ungi along he diges i e ac s o sheep. Mycol Res. 2004;108:1227–33. 21. Janssen PH, Ki s M. S uc u e o he a chaeal communi y o he umen. Appl En i on Mic obiol. 2008;74:3619–25. 22. G uby D, Dela ond HMO. Reche ches su des animalcules se de élopan en g and nomb e dans l'es omac e dans les in es ins pemdan la diges ion des animaux he bi o es e ca ni o es. C R Hebd Seances Acad Sci. 1843;17:1304–8. 23. Hunga e RE. S udies on cellulose e men a ion. III he cul u e and isola ion o cellulose-decomposing bac e ia om he umen o ca le. J Bac e iol. 1947;53:631–45. 24. O pin CG. S udies on he umen lagella e Neocallimas ix on alis. J Gen Mic obiol. 1975;91:249–62. 25. Ken e s N, Hende son G, Jeyana han J, Ki elmann S, Janssen PH. Isola ion o p e iously uncul u ed umen bac e ia by dilu ion o ex inc ion using a new liquid cul u e medium. J Mic obiol Me h. 2011;84:52–60. 26. McKain N, Genc B, Snelling TJ, Wallace RJ. Di e en ial eco e y o bac e ial and a chaeal 16S RNA genes om uminal diges a in esponse o glyce ol as c yop o ec an . J Mic obiol Me h. 2013;95:381–3. 27. Hende son G, Cox F, Ki elmann S, Mi i VH, Ze ho M, Noel SJ, e al. E ec o DNA ex ac ion me hods and sampling echniques on he appa en s uc u e o cow and sheep umen mic obial communi ies. Plos One. 2013;8:e74787. 28. Ki elmann S, Seedo H, Wal e s WA, Clemen e JC, Knigh R, Go don JI, e al. Simul aneous amplicon sequencing o explo e co-occu ence pa e ns o bac e ial, a chaeal and euka yo ic mic oo ganisms in umen mic obial communi ies. Plos One. 2013;8:e47879. 29. Tymensen LD, McAllis e TA. Communi y s uc u e analysis o me hanogens associa ed wi h umen p o ozoa e eals bias in uni e sal a chaeal p ime s. Appl En i on Mic obiol. 2012;78:4051–6. 30. Zhou M, McAllis e TA, Guan LL. Molecula iden i ica ion o umen me hanogens: echnologies, ad ances and p ospec s. Anim Feed Sci Technol. 2011;166-67:76–86. 31. C ee ey CJ, Kelly WJ, Hende son G, Leahy SC. De e mining he cul u abili y o he umen bac e ial mic obiome. Mic ob Bio echnol. 2014;7:467–79. 32. Hende son G, Cox F, Ganesh S, Jonke A, Young W, Janssen PH. Rumen mic obial communi y composi ion a ies wi h die and hos , bu a co e mic obiome is ound ac oss a wide geog aphical ange. Sci Rep. 2015;5:14567. 33. Janssen PH. In luence o hyd ogen on umen me hane o ma ion and e men a ion balances h ough mic obial g ow h kine ics and e men a ion he modynamics. Anim Feed Sci Technol. 2010;160:1–22. 34. Neill AR, G ime DW, Dawson RMC. Con e sion o choline me hyl g oups h ough ime hylamine o me hane in he umen. Biochem J. 1978;170:529–35. 35. Poulsen M, Schwab C, Bo g JB, Engbe g RM, Spang A, Canibe N, e al. Me hylo ophic me hanogenic The moplasma a implica ed in educed me hane emissions om bo ine umen. Na Commun. 2013;4:1428. 36. Ki elmann S, Seedo H, Wal e s WA, Clemen e JC, Knigh R, Go don JI, e al. Simul aneous amplicon sequencing o explo e co-occu ence pa e ns o bac e ial, a chaeal and euka yo ic mic oo ganisms in umen mic obial communi ies. Plos One. 2013;8:e103171. 37. Mo ga i DP, Ma in C, Jouany JP, Ranilla MJ. Rumen p o ozoa and me hanogenesis: no a simple cause-e ec ela ionship. B J Nu . 2012;107:388–97. 38. Zhou M, Chung YH, Beauchemin KA, Hol shausen L, Oba M, McAllis e TA, e al. Rela ionship be ween umen me hanogens and me hane p oduc ion in dai y cows ed die s supplemen ed wi h a eed enzyme addi i e. J Appl Mic obiol. 2011;111:1148–58. 39. Danielsson R, Schnu e A, A hu son V, Be ilsson J. Me hanogenic popula ion and CH 4 p oduc ion in Swedish dai y cows ed di e en le els o o age. Appl En i on Mic obiol. 2012;78:6172–9. 40. Danielsson R. Me hane p oduc ion in dai y cows, Impac o eed and umen mic obio a; Ac a uni e si a is ag icul u ae sueciae, 2016, doc o al hesis no. 2016. p. 45. A ailable a h p://pub.epsilon.slu.se/13308/1/danielsson_ _160427.pd . 41. Ki elmann S, Pina es-Pa ino CS, Seedo H, Ki k MR, Ganesh S, McEwan JC, e al. Two di e en bac e ial communi y ypes a e linked wi h he low-me hane emission ai in sheep. Plos One. 2014;9(7):e103171. 42. Shi WB, Moon CD, Leahy SC, Kang DW, F oula J, Ki elmann S, e al. Me hane yield pheno ypes linked o di e en ial gene exp ession in he sheep umen mic obiome. Genome Res. 2014;24:1517–25. 43. Wallace RJ, Rooke JA, Du hie CA, Hyslop JJ, Ross DW, McKain N, e al. A chaeal abundance in pos -mo em uminal diges a may help p edic me hane emissions om bee ca le. Sci Rep. 2014;4:5892. 44. F ei ag TE, Toe S, Ineson P, P osse JI. Links be ween me hane lux and ansc ip ional ac i i ies o me hanogens and me hane oxidize s in a blanke pea bog. FEMS Mic obiol Ecol. 2010;73:157–65. 45. Cos a KC, Yoon SH, Pan M, Bu n JA, Baliga NS, Leigh JA. E ec s o H 2 and o ma e on g ow h yield and egula ion o me hanogenesis in Me hanococcus ma ipaludis. J Bac e iol. 2013;195:1456–62. 46. Wallace RJ, Rooke JA, McKain N, Du hie CA, Hyslop JJ, Ross DW, e al. The umen mic obial me agenome associa ed wi h high me hane p oduc ion in ca le. BMC Genomics. 2015;16:839. 47. Pi a DW, Pinchak WE, Indugu N, Vecchia elli B, Sinha R, Ful o d JD. Me agenomic analysis o he umen mic obiome o s ee s wi h whea -induced o hy bloa . F on Mic obiol. 2016;7:689. 48. Pei CX, Mao SY, Cheng YF, Zhu WY. Di e si y, abundance and no el 16S RNA gene sequences o me hanogens in umen liquid, solid and epi helium ac ions o Jinnan ca le. Animal. 2010;4:20–9. 49. Mi sumo i M, Sun W. Con ol o umen mic obial e men a ion o mi iga ing me hane emissions om he umen. Asian-Aus al J Anim Sci. 2008;21:144–54. 50. Newbold CJ, Lassalas B, Jouany J-P. The impo ance o me hanogens associa ed wi h cilia e p o ozoa in uminal me hane p oduc ion in i o. Le Appl Mic obiol. 1995;21:230–4. 51. Vogels GD, Hoppe WF, S umm CK. Associa ion o me hanogenic bac e ia wi h umen cilia es. Appl En i on Mic obiol. 1980;40:608–12. 52. K umholz LR, Fo sbe g CW, Vei a DM. Associa ion o me hanogenic bac e ia wi h umen p o ozoa. Can J Mic obiol. 1983;29:676–80. 53. Belanche A, de la Fuen e G, Newbold CJ. S udy o me hanogen communi ies associa ed wi h di e en umen p o ozoal popula ions. FEMS Mic obiol Ecol. 2014;90:663–77. 54. Coleman GS. The me abolism o umen cilia e p o ozoa. FEMS Mic obiol Re . 1986;39:321–44. Tapio e al. Jou nal o Animal Science and Bio echnology (2017) 8:7 Page 9 o 11