http://wrap.warwick.ac.uk
Original citation:
Großkopf, Tobias and Soyer, Orkun S.. (2014) Synthetic microbial communities. Current
Opinion in Microbiology, Volume 18 . pp. 72-77. ISSN 1369-5274
Permanent WRAP url:
http://wrap.warwick.ac.uk/59943
Copyright and reuse:
The Warwick Research Archive Portal (WRAP) makes this work of researchers of the
University of Warwick available open access under the following conditions.
This article is made available under the Creative Commons Attribution 3.0 (CC BY 3.0)
license and may be reused according to the conditions of the license. For more details
see:
http://creativecommons.org/licenses/by/3.0/
A note on versions:
The version presented in WRAP is the published version, or, version of record, and may
be cited as it appears here.
Synthetic
microbial
communities
§Tobias
Großkopf
and
Orkun
S
Soyer
Whilenaturalmicrobialcommunitiesarecomposedofamixof microbeswithoftenunknownfunctions,theconstructionof syntheticmicrobialcommunitiesallowsforthegenerationof definedsystemswithreducedcomplexity.Usedinatop-down approach,syntheticcommunitiesserveasmodelsystemstoask questionsabouttheperformanceandstabilityofmicrobial communities.Inasecond,bottom-upapproach,synthetic microbialcommunitiesareusedtostudywhichconditionsare necessarytogenerateinteractionpatternslikesymbiosisor competition,andhowhigherordercommunitystructurecan emerge from these. Besides their obvious value as model systems tounderstandthestructure,functionandevolutionofmicrobial communitiesascomplexdynamicalsystems,synthetic communitiescanalsoopenupnewavenuesforbiotechnological applications.
Addresses
School of Life Sciences, University of Warwick, United Kingdom
Correspondingauthor:Soyer,OrkunS([email protected])
CurrentOpinioninMicrobiology2014,18:72–77
ThisreviewcomesfromathemedissueonCellregulation
EditedbyCecı´liaMariaArraianoandGregoryMCook
ForacompleteoverviewseetheIssueandtheEditorial
Availableonline14thMarch2014
1369-5274/$–seefrontmatter,# 2014TheAuthors.Publishedby ElsevierLtd.Allrightsreserved.
http://dx.doi.org/10.1016/j.mib.2014.02.002
The
challenges
of
understanding
natural
communities
Recentyearshaveseenasurgeintheanalysisofmicrobial
communities thanks to the development and
cost-reduction in sequencing technologies. The continuing
application of these has led to the characterisation of species diversity in different natural communities from theoceanstothehumangut[1].Despitethisrapidincrease inavailablemetagenomicdatasets,however,therearestill significantlimitationsinansweringthefundamental
eco-logical and evolutionary questions surrounding natural
microbialcommunities [2–4].In particular, we still lack aclearunderstandingofthemolecularandecologicalbases ofcommunity-levelfunctions,andthepotentialtrade-offs intheextent,stabilityandrobustnessofsuchfunctionsand othercommunityproperties(e.g.diversity,structure,size).
While even the simplest of the natural microbial
com-munitiescharacterisedtodatecontaintenstomorethan thousands of species [5], it is usually not possible to experimentallyverifywhichspeciesinsuch characteris-ationsareactivelypartofthecommunity,orare
perform-ing key functions. The assembly of natural microbial
communitiesmightalsodependondispersalofpotential seedorganisms,aswellasonenvironmentalselectionand speciessorting[6],theeffectsofwhichwouldbedifficult tocontrolorcharacterise.Further,theinfluenceofeachof suchdriversonthecompositionofacommunitycanvary accordingto theecosystemandtemporalscaleof obser-vation.Forexample,opensystemslikewastewater
treat-mentplantsarefoundto bedominatedbynear random
colonisation effects [7,8], while more enclosed systems
like the gut of higher animals seem to lead to a more
stable community over time after an initial period of
community assembly [9]. The known limitations in
detecting rare species with sequencing approaches
[5,10]createafurtherchallengeindissectingtherelation
between species composition and community function
anddynamics.
Synthetic
microbial
communities
as
model
systems
Apromisingwayto overcomethedifficulties associated withstudying naturalcommunities is to create artificial communitiesthatretainthekeyfeaturesoftheirnatural
counterparts.These canthen act as amodel system to
assesstheroleofkeyecological,structuralandfunctional
featuresof communities in a controlled way. Here, we
define a synthetic community as one that is created
artificiallybyco-culturingofselect(twoormany)species undera(atleastinitially)well-definedmedia.Whilethe
individual species in such a community can be further
engineered,wedonotrequirepresence ofsynthetically
engineered species for a community to be defined as
synthetic.Towardsanincreasedunderstandingofnatural
communities, these synthetic communities can be
uti-lisedintwogeneralways,whichweclassifyas function-basedandinteraction-based.
Functionfirst:studyingcommunitieswithdetermined function
Thistop-downapproachfocusesondetermininga
func-tionaldefinitionforacommunityfirstandthenusingthis
definition to characterise community structure and
dynamicsindetail.Thevalueofhavingadefined ‘func-tion’foramicrobialcommunityallowsustomeasurethe extentand stability of this function and, perhaps more
importantly, gives us a way to compare different
com-munitieswiththesame function.
Within this approach, ‘natural’ microbial communities
utilised in biotechnology are of interest, as they are
operated under relatively well-defined conditions and
wheretheproductionordegradationofatargetsubstance
can act as an objective measure of performance. In
anaerobic digestion and wastewater-treatment plants,
for example, productionof biogas or reductionof toxic
organic compounds can be taken as the community
function.Thisallowslong-termmeasuringofparameters such as functionalstabilityand performance, and
corre-lationofthesewithcommunitycomposition[11].While
promising, theseapproachesrelyonourability to accu-ratelymeasurespeciesdiversityinacomplexcommunity, which isknowntobelimited[5,10].
Thislimitationcanbesurpassedbydirectlymanipulating the level of diversity in a carefully designed synthetic
community,whichallowsexploringtherelationbetween
community structure and function in a controlled way.
For example, bydefining mercury removalas the
com-munityfunctionandcreatingcommunitieswithdifferent levelsofdiversity,VonCansteinetal.foundan improve-ment in functional efficiency with increasing diversity [12].Similarly,Kassenetal.usedindirectcontrolonradial diversification of Pseudomonas fluorescens to show that productivity(i.e.therateofproductionoforganicmatter byacommunity)displaysaunimodalrelationtodiversity, peakingatintermediatelevelsofdiversity[13].
The functional approach to communities is not only
usefultostudytherelationsbetweenkeypropertiessuch asdiversity,function,andstability,butalsoallows deriv-ingamechanisticunderstandingofcommunitystructure. In the context of anaerobic digestion, the definition of
community function as conversion of complex organic
matter into methane has allowed amechanistic
under-standingoftheunderlyingcommunitiesasspecies
assem-blies interlinked through the different levels of the
degradation process [14].This mechanistic
understand-ing allows developing a sense for key species in the
system, as well as defining potential bottlenecks and
control points.
Interactionsfirst:keymetabolicinteractionpatternsas determinantsofcommunitystructureanddynamics
In this bottom-up approach the focus is on identifying
common interaction patterns and processesamong
bac-terial species, with theexpectation that thesecould be keydeterminantsoftheoverallcommunitystructureand dynamics.Speciesinteractionsinmicrobialcommunities
can be either metabolism-based or be driven by social
traits. To date, social interactions among microbes has
attracted attention and severalgood reviews have
sum-marised the study of these interactions in synthetic as wellasnaturalmicrobialcommunities[15,16–18].Here,
we emphasise the role of metabolism in driving the
speciesinteractionsin microbialcommunities.
The commonly considered social interactions of
competition and cooperation areonly twoof all possible interactionsamongmicrobialspecies.Theneteffectofan organismAonasecondorganismBcanbeneutral,positive or negative, resulting in thenumber of possiblemutual
interactions between two organisms to be 9. Since the
directionality ofareactionisnotof interestfor thetype of interaction (i.e. +/ = /+), there are six basal
inter-action patterns that make up the minimal interaction
motifs in microbialcommunities(Figure 1).While each
oneoftheseinteractionscanbepotentiallydriventhrough socialtraitsoftheinvolvedspecies[19]orenvironmental factors like patchiness (i.e. local differences in species abundance) [20,21],we argue that thesimplest driver
is metabolic interactions. In Figure 1, we provide the
mappingofallpossibleinteractionpatternsbetweentwo speciesintoanecologicalandacorrespondingmetabolic
representation (i.e. thecommunication between species
via their central metabolic products). A food chain for
examplecanbeseenas acommensalinteraction,where
organismBlivesoffofthewasteoforganismA,whointurn isnotaffectedbytheinteraction(0/+).Similarly, compe-titionarisesnaturallywhentwoorganismsAandButilise
the same substrate, and an intense form of cooperation
emerges,wheneachofthepartnersgainbythemetabolic reactionsof theother(i.e.syntrophy).Whilethe combi-nationoftwodifferentstrainsallowforsixpossible inter-actionstates,asystemwiththreestrainshasalready729, one withfour strains531441possiblestatesof microbial interactions. Since such combinatorial explosion of the numberofpossibleinteractionstatesquicklyreacheslarge numberswithonlyafewspecies,thechallengeistofind keymotifsthatareover-representedinnatureorthatcan havesignificantpercolatingeffectsatthecommunitylevel (e.g.stabilizingorde-stabilisinginteractionmotifs,motifs drivingoscillatoryorchaoticdynamics).
It is hypothesised for example that cooperative
inter-actions among the members of the community could
be amajor driver of community function and structure
[15]. This idea motivates recent computational and
experimentalstudiesondeterminingpairwiseinteraction patternsamongdifferentmicrobes.Thesestudies,while preliminary, indicatethat cooperationmight not bethe dominantinteractionpatternamongmicrobialspecies.A
metabolic modelling study of pairwise combinations of
118 isolated bacterial strains, for example, showed that positive interactions are unidirectional in most cases, hencefallinginthecommensalcategory[22].Likewise, an experimental study byFoster et al. found the
inter-actions between two randomly selected members of a
beechtreeholecommunitytobemostlycompetitive[19].
Thepotentialroleofthepresenceorabsenceof
coopera-tion on community dynamics can also be analysed by
synthetically engineering cooperation. Cooperation
knockingoutdifferentessentialgenesineachofthestrains
[23,24], by specific manipulation of the environment
[25,26]orbyselectingpartnerorganismsthatarebound
tocooperatebythermodynamics,forexample,whenone
organismproducesawaste product thatinhibitsitsown metabolicreactionwhenaccumulatingintheenvironment [27,28]. Besides cooperation, several other interaction motifshavealsobeentestedwithsyntheticcommunities featuringsyntheticallyengineeredspecies.Forexample, Balagadde´etal.usedengineeredstrainsofEscherichiacolito
analyze species dynamics resulting from predator-prey
interactions (corresponding to a food chain with waste
inhibitioninmetabolic-terms)[29],whileKerretal.used a similar approach to create a three-species interaction patterncorrespondingtoarock-paper-scissorsgame[30].
Whilesuchapplicationsofsyntheticbiologyin
construct-ing synthetic communities were so far driven by an
ecological view of interaction motifs, we believe that
theconstructionofinteractionsusingametabolism-based viewcouldbe highlyusefulfor several reasons. Firstly, ‘metabolicinteractionmotifs’couldbemoreeasilylinked to natural systems. Naturallyprevalent motifs couldbe
potentially identified by combining metagenomic and
Figure1
Effect Ecological Metabolic References
0/+ Commensalism Food chain Freilich et al.
[22], Xu et al.
[34]
-/- Competition Substrate competition Foster & Bell
[19]
-/+ Predation Food chain with waste product inhibition Balagadde et al.
[23]
Kerr et al. [24]
0/0 No interaction No common metabolites
+/+ Cooperation Syntrophy Shou et al. [25],
Kerner et al. [26], Hillesland et al.
[28], Summers et al. [29],
0/- Amensalism Waste product inhibition
Current Opinion in Microbiology
speciesco-occurrenceanalyses,withthegraph-theoretical analysesofmetabolicnetworkmodelsofeachspecies.For example,theconceptof‘seedsets’,thesetofmetabolitesa microbialspeciesneedstoacquirefromtheenvironment [31],couldbecombinedwiththeknowledgeofspecies’ co-occurrence patterns [32] to predict potential metabolic
interactions. Similarly, creating combined metabolic
modelsofdifferentspeciesconnectedthroughmetabolite exchange[22,26]canallowtheidentificationofmetabolic
interaction motifs among suchspecies. Secondly,
meta-bolicinteractionmotifscanbeidentified,orexpected,from thefunctionalcharacterisationofthemicrobialcommunity understudy.Forexample,itiswellknownthat
methane-forming communities feature methanogens, which are
usually found under syntrophic metabolic interactions
(i.e. cooperation) with other species [33]. Finally, and
perhaps most importantly, metabolic interaction motifs
areeasilyconvertedintodynamicalmodels,forexample using ordinary differential equations, and allow direct
comparison of model outcomes and measurements of
metabolites and populations. This provides characteris-ation oftheresultingspeciesdynamics,asdone
success-fully in the case of syntrophic and cyclical metabolic
interactions [34,35]. Thus, the metabolism-based view
allows for the creation of both theoretical models and
experimental implementations of the simpleinteraction
motifsfoundinmicrobialcommunities.
Synthetic microbial communities can also be used to
combine the function-first and interaction-first
approaches summarised above. This can be achieved
by creating synthetic communities where both species
composition and community function are defined.The
resulting systems provide important insights into the
relation between community function, structure and
dynamics.Examplesof thiscombinedapproachinclude
thosethatfocusedoncharacterisingspeciesinteractions
and stable performance in synthetic communities for
cellulosedegradation[25,36].
Application
of
synthetic
communities
Webelievethatunderstandingmetabolicinteractionsin
microbial communities and the ability to artificially
stabilise interactions between microbes will eventually allow usto betterunderstandandharvestthefull meta-bolicpotentialofthemicrobialworld.
Severalchallenges,however,needtobeovercomeforthe rationalengineeringofmicrobialcommunities.The fore-most of theseis toachievemediacompositionsthatcan maintainmultiplespecies.Giventhatmediacomposition isadynamicparameterofthesystem,linkedtopopulation
dynamics, further development of metabolic as well as
population dynamics models [34,35,37,38] need to be
combinedwithexperimentalwork.Suchmodelscanthen
befurtherrefinedthroughexperimentaldata,but
achiev-ing this will require high-throughput data collection
methods. Population dynamics measurements through
rapid sequencing or reporter-based methods need to be
established for synthetic communities. Long-term
monitoring and culturing of communities will require
furtherdevelopmentofmini-bioreactortechnologies,with
microfluidics based approaches being most promising
[39,40].
We expect that as these challenges are overcome and
rational engineeringof synthetic communities becomes
more mainstream and high-throughput, synthetic
microbial communities will findmore applications both
in microbial ecology and in biotechnology. Especially
with respect to biotechnology, the synthetic microbial
community concept hasa greatpotentialfor the future
[41].Arationaldesignandassemblyofmicrobial special-iststoperformagivenfunctionallowsforthedifferential regulation of subfunctions bycontrolling thedensity of
the microbes that performthem [24]. This might be a
morepromisingapproachthansyntheticmanipulationof
asingle syntheticstrain,especiallywhencomplextasks arerequired[42–44],andconsequentlyitisalready find-ingitswayintofieldslikebiofuelproduction[45,46].
Conclusion
Synthetic microbial communities are abstractions of
naturalsystemsthatallowtheindetailstudyandanalysis
of the fundamental building blocks and processes that
composeamicrobialcommunity.Boththetop-downand
bottom-up approaches summarised above have yielded
importantresultsonsimplecommunitiesandarepaving
the wayfor theassembly of higher order communities.
With the aid of novel approaches such as micro bead
encapsulation of a small number of microorganisms
[39,47]microfluidicchiptechnology[40]orthe sequen-tiallayeringofmicrobesontoasyntheticbiofilm[48],we expect theconstructionofsyntheticcommunitiesto get
easierandpotentiallybecoming high-throughput.
Researchintosyntheticmicrobialcommunitiesis witnes-singare-incarnationafterthefirstattemptsincreating co-cultureandtri-cultureaboutseveraldecadesago[49,50]. Wefeelthatthistimetheyareheretostayasanimportant partofthetoolkitofmicrobialecologyandbiotechnology.
Acknowledgements
WethankKevinPurdyforcommentsonanearlierversionofthis manuscriptandacknowledgefundingfromBiotechnologyandBiological SciencesResearchCouncil,BBSRC(grantBB/K003240/1).
References
and
recommended
reading
Papers of particular interest, published within the period of review, havebeenhighlightedas:
ofspecialinterest ofoutstandinginterest
1. RaesJ,BorkP:Moleculareco-systemsbiology:towardsan understandingofcommunityfunction.NatRevMicrobiol2008, 6:693-699.
2. JessupCM,KassenR,FordeSE,KerrB,BucklingA,RaineyPB, BohannanBJM:Bigquestions,smallworlds:microbialmodel systemsinecology.TrendsEcolEvol2004,19:189-197.
3. ProsserJI,BohannanBJM,CurtisTP,EllisRJ,FirestoneMK, FreckletonRP,GreenJL,GreenLE,KillhamK,LennonJJetal.: Theroleofecologicaltheoryinmicrobialecology.NatRev Microbiol2007,5:384-392.
4. MoralesSE,HolbenWE:Linkingbacterialidentitiesand ecosystemprocesses:can‘‘omic’’analysesbemorethanthe sumoftheirparts? FEMSMicrobiolEcol2011,75:2-16.
5. CurtisTP,SloanWT,ScannellJW:Estimatingprokaryoticdiversity anditslimits.ProcNatlAcadSciUSA2002,99:10494-10499.
6. LeiboldM,Holyoaka,MouquetM,AmarasekareN,ChaseP, HoopesJM,HoltMF,ShurinRD,LawJB,TilmanRDetal.:The metacommunityconcept:aframeworkformulti-scale communityecology.EcolLett2004,7:601-613.
7. OfiteruID,LunnM,CurtisTP,WellsGF,CriddleCS,FrancisCA, SloanWT:Combinednicheandneutraleffectsinamicrobial wastewatertreatmentcommunity.ProcNatlAcadSciUSA 2010,107:15345-15350.
8. CurtisTPP,SloanWTT:Towardsthedesignofdiversity: stochasticmodelsforcommunityassemblyinwastewater treatmentplants.WaterSciTechnol2006,54:227.
9. CostelloEK,StagamanK,DethlefsenL,BohannanBJM, RelmanD:Theapplicationofecologicaltheorytowardan understandingofthehumanmicrobiome.Science2012, 336:1255-1262.
10. BungeJ,WillisA,WalshF:Estimatingthenumberofspeciesin microbialdiversitystudies.AnnuRevStatAppl2013,1:2.1-2.19.
11. BrionesA,RaskinL:Diversityanddynamicsofmicrobial communitiesinengineeredenvironmentsandtheir implicationsforprocessstability.CurrOpinBiotechnol2003, 14:270-276.
12. CansteinH,Von,KellyS,LiY,Wagner-Do¨blerI:Speciesdiversity improvestheefficiencyofmercury-reducingbiofilmsunder changingenvironmentalconditions.ApplEnvironMicrobiol 2002,68:2829-2837.
13. KassenR,BucklingA,BellG,RaineyPB:Diversitypeaksat intermediateproductivityinalaboratorymicrocosm.Nature 2000,406:508-512.
14. SchinkB:Energeticsofsyntrophiccooperationin methanogenicdegradation.MicrobiolMolBiolRev1997, 61:262-280.
15.
MitriS,FosterKR:Agenotypicviewofsocialinteractionsin microbialcommunities.AnnuRevGenet2013http://dx.doi.org/ 10.1146/annurev-genet-111212-133307.
Acomprehensivereviewaboutmicrobialinteractionsthatencompasses theevolutionaryoriginofsocialtraitsinmicrobes.
16. XavierJB:Socialinteractioninsyntheticandnaturalmicrobial communities.MolSystBiol2011,7:483.
17. TanouchiY,SmithRP,YouL:Engineeringmicrobialsystemsto exploreecologicalandevolutionarydynamics.CurrOpin Biotechnol2012,23:791-797.
18. WintermuteEH,SilverPA:Dynamicsinthemixedmicrobial concourse.GenesDev2010,24:2603-2614.
19. FosterKR,BellT:Competition,notcooperation,dominates interactionsamongculturablemicrobialspecies.CurrBiol 2012,22:1845-1850.
20.
BachmannH,FischlechnerM,RabbersI,BarfaN,Brancodos SantosF,MolenaarD,TeusinkB:Availabilityofpublicgoods shapestheevolutionofcompetingmetabolicstrategies.Proc NatlAcadSciUSA2013,110:14302-14307.
Thisstudyexperimentallydemonstrates‘Simpson’sparadox’,i.e.how environmentalsegregationcanbedesignedtobenefitspecieswitheither fastgrowthorhighyield.
21. NadellCD,FosterKR,XavierJB:Emergenceofspatialstructure incellgroupsandtheevolutionofcooperation.PLoSComput Biol2010,6:e1000716.
22.
FreilichS,ZareckiR,EilamO,SegalES,HenryCS,KupiecM, GophnaU,SharanR,RuppinE:Competitiveandcooperative metabolicinteractionsinbacterialcommunities.NatCommun 2011,2:589.
Inthiscomputationalhighthroughputstudytheauthorsusemetabolic modelsof118speciesandtheirpairwise-couplingtoscreenfor inter-action patterns.
23. ShouW,RamS,VilarJMG:Syntheticcooperationinengineered yeastpopulations.ProcNatlAcadSciUSA2007,104:1877-1882.
24. KernerA,ParkJ,WilliamsA,LinXN:Aprogrammable
Escherichiacoliconsortiumviatunablesymbiosis.PLoSOne 2012,7:e34032.
25.
ZuroffTR,XiquesSB,CurtisWR:Consortia-mediated bioprocessingofcellulosetoethanolwithasymbiotic Clostridiumphytofermentans/yeastco-culture.Biotechnol Biofuels2013,6:59.
Theauthorsestablishaco-cultureofaClostridiumandaYeaststrain whichisinitiallyofcompetitivenature.Byfine-tuningtheoxygen con-centrationtheythencreateastabilisingsymbiosis,showingthe impor-tanceofmediaindeterminingthenatureofspeciesinteractions.
26. KlitgordN,SegreD,Segre` D:Environmentsthatinduce syntheticmicrobialecosystems.PLoSComputBiol2010, 6:e1001002.
27. HilleslandKL,StahlDA:Rapidevolutionofstabilityand productivityattheoriginofamicrobialmutualism.ProcNatl AcadSciUSA2010,107:2124-2129.
28. SummersZM,FogartyHE,LeangC,FranksAE,MalvankarNS, LovleyDR:Directexchangeofelectronswithinaggregatesof anevolvedsyntrophiccocultureofanaerobicbacteria. Science2010,330:1413-1415.
29. Balagadde´ FKF,SongH,OzakiJ,CollinsCH,BarnetM,ArnoldFH, QuakeSR,YouL:AsyntheticEscherichiacolipredator–prey ecosystem.MolSystBiol2008,4:187.
30. KerrB,RileyMA,FeldmanMW,BohannanBJM:Localdispersal promotesbiodiversityinareal-lifegameof rock-paper-scissors.Nature2002,418:171-174.
31. BorensteinE,KupiecM,FeldmanMW,RuppinE:Large-scale reconstructionandphylogeneticanalysisofmetabolic environments.ProcNatlAcadSciUSA2008,105:14482-14487.
32. FreilichS,KreimerA,MeilijsonI,GophnaU,SharanR,RuppinE: Thelarge-scaleorganizationofthebacterialnetworkof ecologicalco-occurrenceinteractions.NucleicAcidsRes2010, 38:3857-3868.
33. StamsAJM,PluggeCM:Electrontransferinsyntrophic communitiesofanaerobicbacteriaandarchaea.NatRev Microbiol2009,7:568-577.
34. XuA,DolfingJ,CurtisTP,MontagueG,MartinE:Maintenance affectsthestabilityofatwo-tieredmicrobial‘‘foodchain’’?J TheorBiol2011,276:35-41.
35. KhatriBBS,FreeA,AllenRRJ:Oscillatingmicrobialdynamics drivenbysmallpopulations,limitednutrientsupplyandhigh deathrates.JTheorBiol2012,314:120-129.
36. KatoS,HarutaS,CuiZJ,IshiiM,IgarashiY:Network
relationshipsofbacteriainastablemixedculture.MicrobEcol 2008,56:403-411.
37. KhandelwalR,Oliviera,Ro¨lingBG,TeusinkWFM,BruggemanFJ: Communityfluxbalanceanalysisformicrobialconsortiaat balancedgrowth.PLoSOne2013,8:e64567.
38. StolyarS,VanDienS,HilleslandKL,PinelN,LieTJ,LeighJA, StahlDA:Metabolicmodelingofamutualisticmicrobial community.MolSystBiol2007,3:92.
39.
ParkJ,KernerA,BurnsMa,LinXN:Microdroplet-enabledhighly parallelco-cultivationofmicrobialcommunities.PLoSONE 2011,6:e17019.
40. LeungK,ZahnH,LeaverT,KonwarKM,HansonNW,Page´ AP, LoC-C,ChainPS,HallamSJ,HansenCL:Aprogrammable droplet-basedmicrofluidicdeviceappliedtomultiparameter analysisofsinglemicrobesandmicrobialcommunities.Proc NatlAcadSciUSA2012,109:7665-7670.
41. ShongJ,JimenezDiazMR,CollinsCH:Towardssynthetic microbialconsortiaforbioprocessing.CurrOpinBiotechnol 2012,23:798-802.
42. JohnsonDR,GoldschmidtF,LiljaEE,AckermannM:Metabolic specializationandtheassemblyofmicrobialcommunities. ISMEJ2012,6:1985-1991.
43. BrennerK,YouL,ArnoldFH:Engineeringmicrobialconsortia:a newfrontierinsyntheticbiology.TrendsBiotechnol2008, 26:483-489.
44. MeeMT,WangHH:Engineeringecosystemsandsynthetic ecologies.MolBiosyst2012,8:2470-2483.
45. Ortiz-MarquezJCF,DoNascimentoM,ZehrJP,CurattiL:Genetic engineeringofmultispeciesmicrobialcellfactoriesasan alternativeforbioenergyproduction.TrendsBiotechnol2013, 31:521-529.
46. MintyJJ,SingerME,ScholzSA,BaeC-H,AhnJ-H,FosterCE, LiaoJC,LinXN:Designandcharacterizationofsynthetic fungal-bacterialconsortiafordirectproductionofisobutanol fromcellulosicbiomass.ProcNatlAcadSciUSA2013, 110:14592-14597.
47. JakielaS,KaminskiTS,CybulskiO,WeibelDB,GarsteckiP: Bacterialgrowthandadaptationinmicrodropletchemostats. AngewChemIntEdEngl2013,52:8908-8911.
48. StubblefieldBA,HoweryKE,IslamBN,SantiagoAJ,
CardenasWE,GilbertES:Constructingmultispeciesbiofilms withdefinedcompositionsbysequentialdepositionof bacteria.ApplMicrobiolBiotechnol2010,86:1941-1946.
49. WinterJ,WolfeRS:Completedegradationofcarbohydrateto carbondioxideandmethanebysyntrophicculturesof
AcetobacteriumwoodiiandMethanosarcinabarkeri.Arch Microbiol1979,121:97-102.
50. BryantM,CampbellL:Growthofdesulfovibrioinlactateor ethanolmedialowinsulfateinassociationwithH2-utilizing methanogenicbacteria.ApplEnvironMicrobiol1977, 33:1162-1169.
