experimental insights into the importance of aquatic...
TRANSCRIPT
General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.
• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from orbit.dtu.dk on: May 24, 2018
Experimental insights into the importance of aquatic bacterial community compositionto the degradation of dissolved organic matter
Logue, J.B.; Stedmon, Colin; Kellerman, A.M.; Nielsen, N.J.; Andersson, A.F.; Laudon, H.; Lindström,E.S.; Kritzberg, E.S.Published in:ISME Journal
Link to article, DOI:10.1038/ismej.2015.131Document
Publication date:2016
Document VersionPublisher's PDF, also known as Version of record
Link back to DTU Orbit
Citation (APA):Logue, J. B., Stedmon, C., Kellerman, A. M., Nielsen, N. J., Andersson, A. F., Laudon, H., ... Kritzberg, E. S.(2016). Experimental insights into the importance of aquatic bacterial community composition to the degradationof dissolved organic matter. ISME Journal, 10(3), 533-545. DOI: 10.1038/ismej.2015.131Document
OPEN
ORIGINAL ARTICLE
Experimental insights into the importance of aquaticbacterial community composition to the degradationof dissolved organic matter
Jürg B Logue1,2, Colin A Stedmon3, Anne M Kellerman4, Nikoline J Nielsen5,Anders F Andersson6, Hjalmar Laudon7, Eva S Lindström4 and Emma S Kritzberg11Department of Biology/Aquatic Ecology, Lund University, Lund, Sweden; 2Science for Life Laboratory, Solna,Sweden; 3National Institute for Aquatic Resources, Technical University of Denmark, Charlottenlund,Denmark; 4Department of Ecology and Genetics/Limnology, Uppsala University, Uppsala, Sweden;5Department of Plant and Environmental Sciences, University of Copenhagen, Copenhagen, Denmark;6Division of Gene Technology, Science for Life Laboratory, School of Biotechnology, KTH Royal Institute ofTechnology, Solna, Sweden and 7Department of Forest Ecology and Management, Swedish University ofAgricultural Sciences, Umeå, Sweden
Bacteria play a central role in the cycling of carbon, yet our understanding of the relationshipbetween the taxonomic composition and the degradation of dissolved organic matter (DOM) is stillpoor. In this experimental study, we were able to demonstrate a direct link between communitycomposition and ecosystem functioning in that differently structured aquatic bacterial commu-nities differed in their degradation of terrestrially derived DOM. Although the same amount ofcarbon was processed, both the temporal pattern of degradation and the compounds degradeddiffered among communities. We, moreover, uncovered that low-molecular-weight carbon wasavailable to all communities for utilisation, whereas the ability to degrade carbon of greatermolecular weight was a trait less widely distributed. Finally, whereas the degradation of eitherlow- or high-molecular-weight carbon was not restricted to a single phylogenetic clade, our resultsillustrate that bacterial taxa of similar phylogenetic classification differed substantially in theirassociation with the degradation of DOM compounds. Applying techniques that capturethe diversity and complexity of both bacterial communities and DOM, our study provides newinsight into how the structure of bacterial communities may affect processes of biogeochemicalsignificance.The ISME Journal advance online publication, 21 August 2015; doi:10.1038/ismej.2015.131
Introduction
Carbon (C) cycling has received considerable atten-tion in recent years, spurred by the increase ofcarbon dioxide concentrations in the atmosphereand the therewith-associated changes in climate(Solomon et al., 2007). In the wake thereof, attemptshave been made to balance the global C budget andto develop a mechanistic understanding of itsunderlying dynamics. This has led to a revision ofthe traditional view in which inland waters wereconsidered a passive ‘pipe’ that merely transported Cfrom land to sea. It is now, however, recognised thatinland waters make up an active compartment:one that mineralises, transforms and stores C of
terrestrial origin besides transporting it to the oceans(Cole et al., 2007; Battin et al., 2009; Tranvik et al.,2009). Therefore and in view of future climaticchanges, it is of great importance to comprehendwhich factors influence the mineralisation andtransformation of terrestrially derived C in fresh-water ecosystems.
It is the bacteria that essentially decompose thisallochthonous dissolved organic matter (DOM) andintroduce it into the aquatic food web (Pomeroy1974; Azam et al., 1983; Jansson et al., 2007).Bacterial degradation of DOM is carried out byphylogenetically diverse communities, whosecomposition has been shown to be affected by thequality and quantity of DOM (for example, Logueand Lindström, 2008). Furthermore, differences inbulk bacterial processes (for example, bacterialrespiration or production) related to changes inDOM quality and quantity point towards theexistence of functionally distinct bacterial groups(for example, Kirchman et al., 2004). Yet, studies
Correspondence: JB Logue, Department of Biology/Aquatic Ecol-ogy, Lund University, Sölvegatan 37, Lund 22362, Sweden.E-mail: [email protected] 11 December 2014; revised 13 April 2015; accepted1 July 2015
The ISME Journal (2015), 1–13© 2015 International Society for Microbial Ecology All rights reserved 1751-7362/15www.nature.com/ismej
investigating how the composition of bacterialcommunities affects the cycling of C in fresh watershave to date yielded inconclusive results; whilesome argue to having observed a close relationshipbetween bacterial community composition (BCC)and C processing (Crump et al., 2003; Kirchmanet al., 2004; Judd et al., 2006; Kritzberg et al., 2006;Langenheder et al., 2006; Bertilsson et al., 2007),others found inconsistent (Comte and del Giorgio,2009, 2010; Lindström et al., 2010) or weak links(Langenheder et al., 2005). It has to be noted, thoughthat rather than actually demonstrating a directrelationship between BCC and C processing (seeLangenheder et al., 2005, 2006), most studiesillustrate that environmental parameters, suchas DOM quality and quantity, affect communitycomposition and functioning alike. Given the inter-twined nature of BCC, the environment and bacterialfunctioning, studies directly addressing the relation-ship between aquatic BCC and C processing areclearly lacking.
This lack may be partly due to former metho-dological limitations. Despite their importance inaquatic systems, DOM and microbial diversity yetremain to be characterised for the most part (Curtisand Sloan, 2005; Hertkorn et al., 2008). As DOM isone of the most complex molecular mixtures onEarth (Hedges et al., 2000) and microbial commu-nities are extremely diverse (Curtis and Sloan, 2004),studies going beyond bulk assessments of DOM aswell as the most abundant members of microbialcommunities have been rather challenging. Recenttechnological advances in the field of molecularbiology (for example, high-throughput sequencing)and adopting advanced instrumental approachesinto analytical chemistry (for example, electrosprayionisation mass spectrometry (ESI-MS)) have,however, made it possible to obtain information ofgreater resolution and depth in this respect (seeKujawinski, 2011 for an overview and Herlemannet al., 2014; Landa et al., 2014 and Shabarova et al.,2014 for studies that combine the two approaches).Such an in-depth and integrative characterisation ofboth complex DOM compounds and microbialcommunities is a prerequisite for exploring therelationship between microbial community compo-sition and the processing of DOM.
Here we studied the link between the composi-tion of aquatic bacterial communities and thedegradation of DOM of terrestrial origin. The aimwas to examine how bacterial communities differ-ent in composition differ in their processing ofDOM. We hypothesised that bacterial assemblagesof different origin differ in their ability andpotential to degrade DOM, because they vary incomposition. We tested this hypothesis by adoptinga common garden experiment in which a uniform,terrestrially derived yet artificially prepared DOMmedium was inoculated with aquatic bacterialcommunities collected from four sites of varyingenvironmental character.
Materials and methods
Study sites and sampling
Study sites. The four environmental sites that wereselected for this experiment are all situated withinthe Umeå River basin in the boreal zone of northernSweden and differed in dissolved organic carbon(DOC) characteristics (Supplementary Table S1).Two aquatic samples were taken within the Krycklancatchment at the Svartberget long-term ecologicalresearch site (Laudon et al., 2013): that is, a humicheadwater lake (EnvHL) and a groundwater (EnvGW)sample. The two remaining aquatic samples werecollected downstream of the Krycklan catchment inthe Vindelälven River (EnvVA), one of two majortributaries to the Umeå River, and its mouth in theBaltic Sea (EnvBa).
Sampling. Sampling was carried out on 29 May2012, towards the end of the spring flood. Twosamples were taken at each site: one for bacterialabundance and community composition and one forwater chemistry analyses. Samples for bacterialabundance and community composition werecollected in sterile 1-litre polypropylene bottles(Nalgene, Rochester, MN, USA), whereas waterchemistry samples were collected in acid-washed(p.a. quality HCl; Sigma-Aldrich, St Louis, MO, USA)and Milli-Q (ion- and nuclease-free water) -rinsedpolyethylene bottles (Mellerud Plast, Mellerud,Sweden). EnvBa, EnvHL and EnvVA were sampledtaking grab samples, whereas EnvGW was sampledfrom a shallow, perforated groundwater well.
Samples were kept cold and in the dark duringtransportation to the laboratory. In the laboratory,water chemistry samples were stored at − 20 °C untilfurther processing, while samples for bacterialabundance and community composition were firstpre-sieved (225 μm; nylon net filter) and filtered witha GF/F filter (0.7 μm, pre-combusted at 400 °C for 6 h;Whatman, Maidstone, UK) to avoid capturing largerparticles and remove grazers, respectively.
Experimental set-upThe experiment was performed as a common gardenexperiment, applying a batch culture approach inwhich a medium derived artificially from soilwas inoculated with bacterial cells from the fourenvironments (henceforth called experimental treat-ments: Ba, GW, HL, and VA). A control, consisting ofmedium only (that is, no bacterial inoculum added),was run alongside the four experimental treatments.
Batch cultures were prepared in 1-litre glassbottles with a sealing constructed as follows: apolybutylene terephthalate screw-cap with aperture,holding a silicone rubber seal pierced with twoholes; one for a metal needle connected to a sterile60-ml syringe to enable the withdrawal of samplematerial, the other one for a sterile venting filter to
Bacterial composition matters to functioningJB Logue et al
2
The ISME Journal
avoid the creation of a vacuum when withdrawingsample material. Batch cultures were filled withoutheadspace and stirred continuously throughout theexperiment. Stirring was performed using magneticstir bars in combination with magnetic stirrers.Magnetic stir bars were acid-washed (p.a. qualityHCl; Sigma-Aldrich), rinsed with Milli-Q and heat-sterilised at 120 °C before usage in batch cultures.
The medium was prepared from soil collectedfrom the topsoil layer within the riparian zone 50mdownstream of the groundwater-sampling site. Thesoil was kept cold and in the dark during transporta-tion to the laboratory, where it was stored at − 20 °Cuntil further processing. In the laboratory, ~ 200 g ofsoil were added to 0.8 l of Milli-Q and shaken on arotary table in the dark for 3 h. The soil–watermixture was then subjected to a stepwise filtration,starting from filters with a pore size of 225 down to0.2 μm. Coarse filters (225, 150, 75, and 50 μm;nylon net filters) were acid-washed (p.a. qualityHCl; Sigma-Aldrich), rinsed with Milli-Q andheat-sterilised at 120 °C, whereas filters of smallerpore size were either combusted at 400 °C for 6 h(20 and 8 μm; Cellulose Filters Ashless Grades;Whatman) or heat-sterilised at 120 °C (0.7 and0.2 μm; Supor PES Membrane Disc Filters; PallCorporation, Port Washington, WI, USA) beforeutilisation. In a final step, tangential flow filtration(50 kDa; Pellicon XL Filter; Merck Millipore,Billerica, MA, USA) was performed to obtain asterile medium with regard to bacteria. Once themedium was diluted to a final concentration of 17mg C l−1, nitrogen and phosphorus were added asNH4–NO3 (final concentration: 3.26mgN l−1) andNaH2–PO4 (final concentration: 0.97mg P l−1),respectively, thus preventing nitrogen and phos-phorus limitation.
Four inocula were prepared; one each from therespective GF/F-filtrate of the four environmentalsites. After having been stored at 4 °C and in the darkfor 6 days (that is, from sampling the bacteria in thefield to preparing the inocula in the laboratory), theinocula were concentrated to ~ 1×106 cells ml−1 viatangential flow filtration (Merck Millipore) andsubsequently added to the medium (1% of finalvolume; except to the controls). Finally, at time pointzero (that is, the beginning of the experiment), eachof the four experimental treatments and the control(5 × 1, n=5) were divided up into three batchcultures, respectively: that is, treatments and controlwere each run in three independent triplicates (5 × 3,n=15) after time point zero (see SupplementaryMethods S1 for an in-detail description of inoculapreparation, inoculation and start of the experiment).
The experiment was conducted in a constanttemperature room at 15 °C and in the dark, andterminated after 5 and a half days on the basis oflevelling off DOC concentrations. Samples weretaken for bulk DOC concentration, UV–visibleabsorbance and fluorescence, ESI-MS, bacterialabundance, and BCC. Bulk DOC concentration
and UV–visible absorbance and fluorescence weresampled on seven (that is, 0, 30, 78, 90, 102, 114, and126 h), whereas samples for bacterial abundancewere taken on nine occasions throughout theexperiment (that is, 0, 10, 30, 54, 78, 90, 102, 114,and 126 h). ESI-MS samples were taken both at thebeginning and at the end, whereas BCC was onlysampled at the end of the experiment. Samples forBCC were, moreover, also taken from the fouroriginal environments (from the respective GF/F-filtrate; Supplementary Methods S1 and SupplementaryFigure S1).
All glass- and plastic ware was acid-washedovernight in HCl (p.a. quality; Sigma-Aldrich),extensively rinsed with Milli-Q and combusted at400 °C for 6 h or heat-sterilised (120 °C), respectively.
DOM analysesSamples analysed for DOC and optical propertieswere pre-filtered with a 0.2-μm syringe filter(Puradisc PES; Whatman).
Dissolved organic carbon. The concentration ofDOC was recorded using a Sievers 900 LaboratoryTotal Organic Carbon Analyzer (UV/persulfateoxidation; GE Analytical Instruments, Manchester,UK). The manner in which the medium wasprepared ensured negligible concentrations of parti-culate organic material; hence, total organic C iscomparable to DOC.
UV–visible absorbance and fluorescence. Absorbancespectra were measured from 200 to 700 nm at 1-nmintervals, with a Lambda 35 UV–visible spectrometer(Perkin Elmer, Waltham, MA, USA). Samples weremeasured in a 1-cm quartz cuvette and distilledwater was used as a blank measurement.
Excitation–emission matrices (EEMs) were col-lected with a FluoroMax-2 spectrofluorometer (Hor-iba Scientific, Edison, NJ, USA), using a 1-cm quartzcuvette. Excitation wavelengths (λEx) spanned from250 to 445 nm in 5-nm increments, whereas emissionwavelengths (λEm) ranged from 300 to 600 nm atincrements of 4 nm. Excitation and emission slitwidths were set to 5 nm and the integration time was0.1 s. Blank subtraction, correction of EEMs andcalibration to Raman units was carried out accordingto Murphy et al. (2010). Four individual fluorescingcomponents in the EEMs were identified andvalidated with parallel factor (PARAFAC) analysis,using the MATLAB and Statistics Toolbox (R2013a;The MathWorks, Inc., Natick, MA, USA) in combina-tion with the DOMFluor toolbox (Stedmon and Bro,2008). The components were derived from the EEMsof 95 samples and their fluorescence characteristicsare depicted as insets in Figure 3b.
Electrospray ionisation mass spectrometry. DOMwas first isolated via solid-phase extraction (SPE) asdescribed by Dittmar et al. (2008). Note that SPE—as
Bacterial composition matters to functioningJB Logue et al
3
The ISME Journal
any presently available DOM isolation method—only retains a certain fraction of the total DOM(that is, polar compounds of low to moderatemolecular weight), yet extraction efficiency isgenerally higher compared with other isolationmethods (Green et al., 2014). In brief, experimentalsamples were filtered (0.2 μm; Puradisc PES SyringeFilter; Whatman), acidified with HCl (p.a. quality;Sigma-Aldrich) to pH 2.5 and stored at 4 °C untilSPE. SPE cartridges (Bond Elut-PPL, 1 g, 6ml;Agilent Technologies, Santa Clara, CA, USA) weresoaked overnight in methanol (LC-MS CHROMASOLV;Sigma-Aldrich), rinsed in Milli-Q, re-rinsed withmethanol, and then rinsed with acidified Milli-Q(pH=2, p.a. quality HCl; Sigma-Aldrich). Immedi-ately before SPE, samples were acidified one moretime with HCl (p.a. quality; Sigma-Aldrich) to pH 2.Acidified sample aliquots (600ml) were allowedto pass through the SPE cartridges by gravity.Cartridges were subsequently rinsed with acidifiedMilli-Q (pH=2, p.a. quality HCl; Sigma-Aldrich) anddried with gaseous N2. DOC was eluted with 4mlmethanol and stored at −20 °C.
Mass spectra were collected on a quadrupole time-of-flight mass spectrometer, operating in scan modewith negative ESI. Blank injections of the mobilephase and two selected samples (henceforth calledreference samples) were each measured four timesthroughout, with the latter to check for instrumentdrift, measurement reproducibility and analyticalprecision. A more detailed description can be foundin Supplementary Methods S2.
Data reduction of mass spectra was conducted asfollows: mass to charge ratios (m/z) were binned tointegers, which resulted in 1900m/z ranging from100 to 1999 (see Figure 5a for an example).Subsequently, sample representative mass spectrawere obtained by combining spectra across theinjection profile. An average blank measurementwas calculated and subtracted from all samples. Thefour separate measurements from each of the twoselected reference samples were thereafter used toestimate and test for analytical precision. At first, thes.d. for each m/z for both reference samples werecalculated. Next, the highest s.d. from either of thetwo was selected at each m/z and multiplied by 2.Finally, this recombined spectra was adopted todefine a threshold for indicating significant changesin DOM mass spectra during the experiment; onlychanges > 2 s.d. for each respective m/z wereconsidered significant and included in subsequentanalyses.
Bacterial abundanceBacterial cells were preserved in sterile-filtered,borax-buffered formaldehyde at a final concentra-tion of 4% w/v. Bacterial abundance was enumer-ated by flow cytometric determination (CyFlowspace; Partec, Münster, Germany) of SYTO13- (Invitrogen, Carlsbad, CA, USA) stained cells,
following the method described by del Giorgioet al. (1996).
Nucleic acid extraction, PCR and pyrosequencingBacterioplankton cells were collected onto 0.2 μmmembrane filters (Supor-200 Membrane Disc Filters;Pall Corporation), filtering 0.2 l of water. Filters wereplaced into sterile cryogenic vials (Nalgene) andfinally kept at −80 °C until further processing.
Nucleic acid extraction. Nucleic acid extractionwas performed following the protocol 3 of the Easy-DNA kit (Invitrogen) with an extra 0.2 g of 0.1mmzirconia/silica beads. Extracted nucleic acids weresized and yields quantified by means of agarose (1%)gel electrophoresis, GelRed staining (Biotium Inc.,Hayward, CA, USA) and UV transillumination beforePCR amplification.
PCR amplification and template preparation. Thebacterial hypervariable regions V3 and V4 of the 16 SrRNA gene were PCR amplified, using bacterialforward and reverse primer 341 (5′-CCTACGGGNGGCWGCAG-3′) and 805 (5′-GAC TACHVGGGTATCTAATCC-3′), respectively (Herlemann et al.,2011). The primers were modified before employ-ment according to the final configuration: AdaptorB-341F and AdaptorA-MID-805R (AdaptorA and Bare 454 Life Sciences adaptor sequences; 454 LifeSciences, Branford, CT, USA). Multiplex identifierswere seven-nucleotide long, sample specific anddeveloped following recommendations by Engel-brektson et al. (2010). PCR reactions were performedin a 20-μl reaction volume comprising 0.4 U Phusionhigh-fidelity DNA polymerase (Finnzymes, Espoo,Finland), 1 × Phusion HF reaction buffer (Finn-zymes), 200 μM of each dNTP (Invitrogen), 200 nMof each primer (Eurofins MWG, Ebersberg,Germany), 0.1 mgml− 1 T4 gene 32 protein (NewEngland Biolabs, Ipswich, UK) and finally5–10 ng of extracted nucleic acid. Thermocycling(DNA Engine (PTC-200) Peltier Thermal Cycler;Bio-Rad Laboratories, Hercules, CA, USA) wasconducted with an initial denaturation step at 95 °Cfor 5min, followed by 27 cycles of denaturation at95 °C for 40 s, annealing at 53 °C for 40 s, extension at72 °C for 1min and finalised with a 7-min extensionstep at 72 °C. Four technical replicates were run persample, pooled after PCR amplification and purifiedusing the Agencourt AMPure XP purification kit(Beckman Coulter Inc., Brea, CA, USA). Nucleic acidyields were checked on a fluorescence microplatereader (Ultra 384; Tecan Group Ltd, Männedorf,Switzerland), employing the Quant-iT PicoGreendsDNA quantification kit (Invitrogen). Finally, PCRamplicons were pooled in equimolar proportions toobtain a similar number of 454-pyrosequencingreads per sample.
Pyrosequencing. The final pooled amplicon wassequenced unidirectionally (Lib-L chemistry) on a
Bacterial composition matters to functioningJB Logue et al
4
The ISME Journal
454 GS-FLX system (454 Life Sciences) at theNorwegian High-Throughput Sequencing Centre(NSC, Oslo, Norway; http://www.sequencing.uio.no), using GS-FLX Titanium reagents.
Sequence analysesThe 454-pyrosequencing errors, PCR single baseerrors and chimeric sequences were removed fromthe 454-pyrosequencing amplicon library employingAmpliconNoise (v1.26; Quince et al., 2011) followedby Perseus (Quince et al., 2011). Pyrosequencingreads not matching multiplex identifier and/orprimer sequences were removed just as were readsshorter than 200 bp. Reads were further truncated at450 bp, eliminating additional noise (Mardis, 2008),and finally trimmed off multiplex identifier andprimer sequences.
Denoised 454-pyrosequences were clustered intooperational taxonomic units (OTUs) at a level of 97%sequence identity (AmpliconNoise, v1.29; Quinceet al., 2011) and classified based on the RDP naiveBayesian rRNA Classifier (RDP Classifier, v2.6; Wanget al., 2007). Representative sequences were alignedbased on the SILVA alignment (release 102; Quastet al., 2013) using mothur (v1.33.2; Schloss et al.,2009). Finally, pyrosequences that could neither bealigned nor assigned, or were assigned as Archaea orEukaryota (for example, chloroplasts) were furtherremoved. The 454-pyrosequencing reads of bothexperimental (Ba, GW, HL, and VA) and environ-mental (EnvBa, EnvGW, EnvHL, and EnvVA) sam-ples have been deposited at the NCBI Sequence ReadArchive under accession number SRP021096.
Data analysesPyrosequencing sampling efforts (that is, the numberof pyrosequences obtained per sample) were normal-ised for statistical data analysis. Normalisation wasdone across samples through sub-sampling andanalyses are based on 29 206 reads randomly drawnfrom each experimental sample. To analyse differ-ences in BCC among experimental treatments, amultivariate generalised linear model (Wang et al.,2012; Warton et al., 2012) was applied. The modelthat is fitted is log-linear and assumes a negativebinomial distribution of data. Relationships betweenbacterial assemblages at the end of the experimentwere visualised employing non-metric multidimen-sional scaling (Bray–Curtis distance) ordination.
To investigate whether bacterial abundances orDOC concentrations were significantly differentbetween treatments at the end of the experiment, aone-way analysis of variance (ANOVA) was carriedout. Repeated-measures ANOVAs were performedfor bacterial abundances, DOC concentrations andfluorescent intensities of PARAFAC componentsbetween experimental treatments, to test for treat-ment and time effects as well as for an interaction ofthe two throughout the experiment. Permutational
ANOVAs (Euclidean distance) were computed toexamine differences between the experimental treat-ments with respect to fluorescence, fluorescent(PARAFAC) components and m/z at the end of theexperiment. The relationships between experimentaltreatments regarding m/z were, in addition, analysedby principal component analysis.
Finally, associations between bacterial OTUs andchange in m/z were examined via Mantel’s test andcorrelation analysis. Mantel’s testing was carried outbetween distance matrices derived from the finalrelative abundances of bacterial OTUs (Bray–Curtisdistance) and change inm/z from the beginning to theend of the experiment (Euclidean distance). Correla-tion analysis tested for co-variation between the finalrelative abundance of the most abundant bacterialtaxa and change inm/z. The most abundant taxa werearbitrarily defined as OTUs containing 4100 readsper OTU across all experimental samples (35 in total).To correct for multiple correlations, P-values wereadjusted according to the false discovery rate(Benjamini and Hochberg, 1995).
All statistical data analyses were conducted usingR (2015), in particular the vegan (Oksanen et al.,2008) and the mvabund (Wang et al., 2012) packages,and P-values were opposed to an α-value of 0.05.
Results
BCC analysisEnvironmental and experimental samples con-tained on average 1259 and 64 OTUs, respectively(Supplementary Table S2 and SupplementaryFigure S2). Bacterial communities from the fourenvironmental sites were distinct from one anotherin composition (Supplementary Figure S3). Thetriplicate experimental bacterial assemblages weremore similar in composition to each other than tosuch from other experimental treatments as bothnon-metric multidimensional scaling ordination(Figure 1) and multivariate generalised linear model(Wald = 35.88, P=1.00E− 03) show.
Bacterial abundance analysisBacterial abundances of the four treatmentsincreased considerably over the course of theexperiment from on average 1.6 × 104 in the begin-ning to between 4.1 ×107 and 1.8 × 108 cells ml−1 inthe end (Figure 2a). Abundances recorded in thecontrols were at, or marginally above, the detectionlimit throughout the experiment. Overall, bacterialabundances changed significantly over time anddifferently as to treatments over the course of theexperiment (Table 1). Yet, only HL differed signifi-cantly from the other treatments in terms of bacterialabundance at the end of the experiment (ANOVA;F= 21.67, P=3.39E− 04; Figure 2a). The growthcurves more or less resemble the growth of a single
Bacterial composition matters to functioningJB Logue et al
5
The ISME Journal
species batch culture (with an initial lag, anexponential and the onset of a stationary phase).
Analyses of DOMThe concentration of DOC in all treatmentsdecreased by approximately two-thirds overthe course of the experiment from on average17mg C l−1 in the beginning to 6mg C l−1 in the end(Figure 2b). The control samples also experienced adecline, albeit a considerably less pronounced one.On the whole, DOC concentrations changed signifi-cantly over time, although only VA significantlydiffered from the other three treatments over thecourse of the experiment (Table 1). DOC concentra-tions measured at the end of the experiment did notdiffer among the four treatments (ANOVA; F=0.85,P=0.51).
The changes in fluorescence in the controls wereminimal compared with the four experimentaltreatments (Figure 3a). Ba, GW and HL showed ahigh degree of similarity in qualitative (spectral)change with a distinct removal of fluorescence at~ λEm 460 and 300 nm. VA, on the other hand,exhibited a notable difference from the other experi-mental treatments in that a loss of fluorescence at~ λEm 360 nm and an increase in fluorescence at~ λEm 470 nm could be observed (Figure 3a).Furthermore, the four treatments differed signifi-cantly from each other with regard to fluorescence atthe end of the experiment (permutational ANOVA;R2 = 0.83, P=2.00E−03). PARAFAC analysis identi-fied four distinct fluorescent components for whichthe molecular structures are unknown. Componentsone and two (C1 and C2) showed locations ofmaximum peak intensities typical of what is referredto as humic like, whereas component 3 (C3) exhibitedfluorescence properties similar to that of the amino
acid tryptophan (also called protein like). Compo-nent 4 (C4) depicted intermediate characteristics.The controls, in general, showed no change influorescence intensities for all four components(Figure 3b). Compared with the other two compo-nents, C1 and C2, remained more or less unaltered influorescence throughout the experiment, with only aslight systematic removal of C1 in all treatments butVA. C3 showed a substantial decrease in intensity,whereas C4 experienced a marginal increase forall four experimental communities (Figure 3b).PARAFAC components predominantly changed sig-nificantly over time but only C3 differed significantlythroughout the experiment across all treatments(Table 1). Permutational ANOVA, furthermore, iden-tified significant differences in PARAFAC compo-nents among the four treatments at the end of theexperiment (R2 = 0.97, P=9.99E− 04).
−0.5
MDS1
MD
S2
Ba1
Ba2
Ba3
GW1GW2GW3
HL1
HL2
HL3
VA1
VA2
VA3
stress = 0.07
−0.4
0.2
−0.2
0.0
0.4
1.00.50.0
Figure 1 NMDS representation of bacterial communities from thefour experimental treatments. NMDS ordination was derived frompairwise Bray–Curtis distances. Numbers depict replicates one,two and three. Hulls were drawn to group replicates within anexperimental treatment. Abbreviations: Ba, Baltic; GW, ground-water; HL, headwater lake; VA, Vindelälven.
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
8.5
Bac
teria
l Abu
ndan
ce [l
og10
Cel
ls m
L-1]
5.0
7.5
10.0
12.5
15.0
17.5
96
Time [hrs]
Dis
solv
ed O
rgan
ic C
arbo
n C
once
ntra
tion
[mg
C L
-1]
TreatmentC
BaGWHLVA
7248240
B
A
144120
Figure 2 Trends of bacterial abundances (a) and DOC concentra-tions (b). Bacterial abundances and DOC concentrations weremeasured over the course of the experiment for the four treatmentsand the control (mean± s.e., n=3 replicates; except for time point0, where n=1). Letters A and B in a denote significant differencesbetween treatments with regard to bacterial abundances at the endof the experiment (A) or not (B), respectively, assessed by means ofTukey’s post-hoc test on an ANOVA. Abbreviations: Ba, Baltic; C,control; GW, groundwater; HL, headwater lake; VA, Vindelälven.
Bacterial composition matters to functioningJB Logue et al
6
The ISME Journal
Mass spectra (see Figure 5a for an example) of thefour treatments did not differ notably from eachother in the beginning of the experiment (Figure 4);however, they did at the end (as seen in Figure 4 andconfirmed by permutational ANOVA; R2 = 0.42,P=1.70E− 03). Some masses also decreased signifi-cantly in intensity in the controls (Figure 5b).Bacterial assemblages from Ba and HL reduced theintensity of masses from a very broad range of m/z,whereas bacteria from GW and VA tended topreferentially reduce the intensity of masses ofo600m/z (Figure 5b). Further, it is only within avery narrow mass range (~200–600m/z; correspond-ing to low-molecular-weight carbon (LMWC);Supplementary Methods S2) that bacterial assem-blages from all experimental treatments were able todegrade DOM (Figure 5c).
Associations between bacterial taxa and thedegradation of DOMExperimental communities similar in compositiondisclosed similar trends in the overall change of m/zover time (Mantel’s test: R=0.38, P=0.01). Correlat-ing the relative abundances of the most abundantbacterial taxa at the end of the experiment (classifiedpredominantly as Proteobacteria and, to a far lesserextent, Bacteroidetes) with the change in m/zthroughout the experiment showed a variety of bothpositive and negative associations, and that OTUsassociated with different range intervals of change inm/z (Figure 6). With regard to the latter, strongpositive associations between relative OTU abun-dance and the degradation of m/z of smaller or largersize could be found for β-proteobacterial taxa (that is,OTUs C1, C430 and C3), and α- (that is, OTUs C3141and C18) and β-proteobacterial OTUs (that is, C3145and C836), respectively. Furthermore, high relativeabundances of the γ- and α-proteobacteria C812 andC0, respectively, associated strongly with a decreasein m/z over almost the entire range of m/z.
Discussion
Taking advantage of technological advances inanalytical chemistry and molecular biology, weexplored how the composition of aquatic bacterialcommunities affected the degradation of DOM ofterrestrial origin. Having adopted an experimentalapproach in which model communities wereexposed to a terrestrially derived DOM substrate,our results highlight that although bacterial commu-nities that differ in composition degraded the sameamount of DOM, both the temporal pattern ofdegradation and, most importantly, the compoundsthat were degraded significantly differed. Finally, weobserved that the most abundant bacterial taxadiffered substantially in their association with thedegradation of DOM compounds.T
able
1Resultsfrom
repea
tedmeasu
resANOVA,testingfordifferencesin
bacterialab
undan
ces,
DOCco
ncentrationsan
dfluoresce
ntintensities
ofthefourco
mpon
ents
iden
tified
byPARAFACan
alysis
betw
eentheex
perim
entaltreatm
ents
andov
ertime
Bac
terial
abundan
ceDOC
df
C1
C2
C3
C4
df
FP
df
FP
FP
FP
FP
FP
Treatmen
t3
176.72
o2.00
E−16
*,a
312
.64
2.26
E−06
*,b
316
0.98
o2.00
E−16
*,c,d
71.54
o2.00
E−16
*,c,d
58.18
o2.00
E−16
*,d
19.35
1.48
E−08
*,d,e
Tim
e8
2012
.42
o2.00
E−16
*6
600.25
o2.00
E−16
*6
75.39
o2.00
E−16
*26
.83
2.66
E−14
*10
3.51
o2.00
E−16
*77
.89
o2.00
E−16
*Treatmen
t:time
2421
.54
o2.00
E−16
*18
10.06
1.78
E−11
*17
4.12
4.03
E−05
*3.48
2.73
E−04
*2.81
2.19
E−03
*5.24
1.83
E−06
*Residuals
7054
52
Abb
reviations:
ANOVA,an
alyses
ofva
rian
ce;BA,Baltic;
DOC,dissolved
organic
carbon
;GW,grou
ndwater;HL,headwater
lake
;PARAFAC,parallelfactor;VA,Vindelälve
nRiver.
*Indicatesign
ifican
tP-values.
aNotethat
experim
entaltreatm
ents
HLan
dVA
did
not
sign
ifican
tlydifferin
bacterialab
undan
cesfrom
each
other
through
outtheex
perim
ent(assessedby
linearmixed
-effects
mod
elan
dTuke
y’s
post-hoc
test).
bNotethat
only
VA
sign
ifican
tlydifferedfrom
theother
treatm
ents
through
outtheex
perim
entwithregard
toDOCco
ncentrations(assessedby
linearmixed
-effects
mod
elan
dTuke
y’spost-hoc
test).
cNotethat
experim
entaltreatm
ents
GW
andHLdid
neither
sign
ifican
tlydifferin
C1nor
C2from
each
other
through
outtheex
perim
ent(assessedby
linea
rmixed
-effects
mod
elan
dTuke
y’spost-
hoc
test).
dNotethat
linearmixed
-effects
mod
ellingan
dsu
bseq
uen
tTuke
y’spost-hoc
testingwithresp
ectto
PARAFACco
mpon
ents
C1,
C2,
C3an
dC4co
uld
only
beperform
edstartingfrom
theseco
ndtime
point,as
linearmixed
-effects
mod
elsdonot
acceptmissingdata(dataforthefirsttimepointwas
not
availableforVA).
eNotethat
BA
andGW
did
not
sign
ifican
tlydifferin
C4from
HLor
VA
andHL,resp
ective
ly,through
outtheex
perim
ent(assessedby
linearmixed
-effects
mod
elan
dTuke
y’spost-hoc
test).
Bacterial composition matters to functioningJB Logue et al
7
The ISME Journal
C
300
400
500
600
0
0.5
1
-0.2
-0.1
0
0.1
F
0
0.02
0.04
Ba
300
400
500
600
0
0.5
1
-0.2
-0.1
0
0.1
F
0
0.02
0.04G
W
300
400
500
600
0
0.5
1
-0.2
-0.1
0
0.1
F
0
0.02
0.04
HL
300
400
500
600
0
0.5
1
-0.2
-0.1
0
0.1
F
0
0.02
0.04
VA
300300 300 300
400
400 400 400
500
600
0
0.5
1
-0.2
-0.1
0
0.1
F
0
0.02
0.04
Fluorescence Spectra Mean Change in Fluorescence Standard Deviation of Change
Time [hrs]
TreatmentCBaGWHLVA
C4 (intermediate)
96 120 144
Time [hrs]
Ram
an U
nits
[R.U
.]
C3 (protein-like)
C2 (humic-like)
−0.4
−0.3
−0.2
−0.1
0.0
0.1
Ram
an U
nits
[R.U
.]
C1 (humic-like)
7248240 96 120 1447248240
−0.4
−0.3
−0.2
−0.1
0.0
0.1
300 400 5000.0
0.2
0.4
0.6
0.8
1.0
Wavelength [nm]
Nor
mal
ised
Flu
ores
cenc
e
300 400 500Wavelength [nm]
0.0
0.2
0.4
0.6
0.8
1.0
Nor
mal
ised
Flu
ores
cenc
e
0.0
0.2
0.4
0.6
0.8
1.0
Nor
mal
ised
Flu
ores
cenc
e
300 400 500Wavelength [nm]
0.0
0.2
0.4
0.6
0.8
1.0
Nor
mal
ised
Flu
ores
cenc
e
300 400 500Wavelength [nm]
Figure 3 Net changes in DOM fluorescence (a) and fluorescent intensities of components identified by PARAFAC analysis (b).(a) Excitation–emission matrices (EEMs) at the start of the experiment (n=1) together with the mean change and s.d. in fluorescence fromthe beginning to the end of the experiment across the three replicates for each treatment (n=3). Excitation (λEx) and emission (λEm)wavelengths are given on the x and y axis, respectively. (b) Pictures changes of fluorescent intensities of PARAFAC components: C1, C2, C3
and C4 (mean± s.e., n=3; except for time point 0 and time point 2 for VA only, where n=1). All components were normalised to zero fortime point zero, except the ones in VA, which were normalised to zero for the second time point, as the measurement at time point zerohad to be discarded owing to an erroneous reading. Insets visualise the respective spectral properties of the four fluorescent componentsidentified by PARAFAC analysis. Abbreviations: Ba, Baltic; C, control; GW, groundwater; HL, headwater lake; VA, Vindelälven.
Bacterial composition matters to functioningJB Logue et al
8
The ISME Journal
A major goal in ecology is to link the compositionof biological communities with processes occurringin an ecosystem. Given the entwined nature ofmicrobial community composition, the environmentand ecosystem processes, one of the greatest chal-lenges is to test for direct effects of composition onfunctioning. Common garden experiments allow forprecisely that by standardising environmental para-meters and, therefore, enabling the teasing apart ofthe effects of the environment from the compositionof microbial communities on functioning (Reed andMartiny, 2007). The downside of incubating micro-bial communities under batch growth conditions,however, is that the resulting community will differfrom the composition of its original inoculum (forexample, Christian and Capone, 2002). Indeed, ouranalyses identified a change from environmental toexperimental bacterial communities in both diver-sity and composition (Supplementary Figures S2 andS4, respectively). It has been further suggested thatsuch experiments favour micro-organisms rare innature but featuring opportunistic, copiotrophicqualities that allow for a more rapid adaptation tochanges in environmental conditions and, hence, tooutcompete others that are originally more abundant.In nature, DOM varies in quality (and quantity) overspace and time (for example, Kothawala et al., 2014),variations to which microbes need to adapt. Yet, theexposure of different bacterial communities inour experiment to a freshly prepared, terrestriallyderived and, hence, highly bioavailable DOMsubstrate as C source possibly enhanced growth ofsuch naturally rare bacteria. As the communities stilldiffer in composition at the end of the experiment, itcan, however, be assumed that the functional
differences observed are the consequence of initialcompositional differences among the bacterialcommunities.
Our results, thus, show a close link between BCCand function. Going beyond a mere identification ofa link between BCC and DOM degradation, ourresults further highlight that the four experimentalcommunities degraded different components of theDOM pool. Although fluorescence analyses illustratethat certain DOM components were commonlymore bioavailable than others, both fluorescenceand ESI-MS analyses demonstrate that the fourdifferent bacterial communities differed in whichDOM components were degraded preferentially.Most importantly, ESI-MS analysis uncovered thatcommunity composition was of little importanceregarding the degradation of LMWC, whereas theutilisation of masses of greater size differed amongcommunities. This means that the ability to useLMWC is a functional property (trait) rather common
−6e−04
PC1 (30.9%)
PC
2 (1
7.1%
)
Ba
C
GW
HL
VA
Ba1Ba2
Ba3C1
C2C3
GW1GW2
GW3
HL1
HL2HL3
VA1
VA2
VA3
2e−04
1e−04
0e+00
−1e−04
−2e−04
4e−042e−040e+00−2e−04−4e−04
Figure 4 Results from principal component analysis of mass tocharge ratios (m/z) that derived from ESI-MS. The first and secondprincipal component explained 30.9% and 17.1% of the varia-bility, respectively. Samples from both the beginning (notnumbered) and end (numbered) of the experiment are visualised.Numbers depict replicates one, two and three. Hulls were drawnto group not only the replicates within a treatment but also thesamples at the beginning of the experiment. Abbreviations: Ba,Baltic; C, control; GW, groundwater; HL, headwater lake; VA,Vindelälven.
0
1
2
3
4
Mass to Charge Ratio [m/z]
100
C
Ba
GW
HL
VA
(233)
(130)
(52)
(34)
(28)
5000
4000
3000
2000
1000
19001700150013001100900700500300
Abu
ndan
ceE
xper
imen
tal T
reat
men
tsTi
mes
Red
uced
Figure 5 Results from ESI-MS analyses. (a) An example for themass spectra of DOM obtained from an experimental sample (thatis, GW) at the beginning of the experiment. Mass to charge ratios(m/z) that were significantly reduced in intensity throughout theexperiment are visualised in b. Numbers in brackets specify thetotal number of m/z significantly reduced in intensity per sampleby the end of the experiment. (c) m/z that were significantlyreduced in intensity by none of the four experimental treatmentsand their respective replicates (Times Reduced 0), all threereplicates of just one treatment (Times Reduced 1), allthree replicates of only two treatments (Times Reduced 2), allthree replicates of three and all four experimental treatments(Times Reduced 3 and 4, respectively) by the end of theexperiment. The dashed line visualises the distinction betweenLMWC (o600m/z) and high-molecular-weight carbon masses(4600m/z). Abbreviations: Ba, Baltic; C, control; GW, ground-water; HL, headwater lake; VA, Vindelälven.
Bacterial composition matters to functioningJB Logue et al
9
The ISME Journal
in all of the four bacterial communities, whereas thecapability to use C of high-molecular-weight appearsto be a trait restricted to particular bacterial commu-nities. An explanation could lie in a finding made byWeiss et al. (1991) that compounds of up to ~600 Da(that is, LMWC) can be taken up readily by micro-organisms across the cell membrane (that is, througha variety of transmembranic transport systems),whereas larger ones require extracellular cleavageby means of enzymatic hydrolysis (that is, viaindividual or interacting ectoenzymes), an abilitythat indeed not all bacterial taxa possess (forexample, Berlemont and Martiny 2013). Yet, bacter-ial members within a community vary not only withregard to the ability to produce ectoenzymes but alsoin their capability to express transmembranic trans-port systems that allow the uptake of compoundsexceeding 600 Da (for example, Teeling et al., 2012).However, a microbial community’s toolbox of traitsis more than the sum of its parts; on the one hand,some bacterial taxa may be needed to actuallyfacilitate the degradation process, allowing othermicro-organisms to either hydrolyse substratesfurther or take them up, on the other the processmay only continue when some microbes act in
concert. Pedler et al. (2014), for instance, demon-strated that the readily available fraction of a coastalDOM pool could be completely removed by a singletaxon, whereas decomposition of the less bioavail-able portion required additional members of thecommunity. Hence, it becomes apparent that notonly the chemical composition of DOM (that is,quality) but also the distribution of traits withinmicrobial communities are important when it comesto whether or not DOM evades microbial re-mineralisation and transformation. As such, bioa-vailability can be perceived as an ongoing interac-tion between the chemical composition of DOM anda microbial community’s metabolic capacity ratherthan merely an inherent property of DOM (Nelsonand Wear 2014).
Although the degradation of either low- or high-molecular-weight carbon was not restricted to asingle phylogenetic clade, our results illustrate thatbacterial taxa of similar phylogenetic classificationdiffered substantially in their association with thedegradation of DOM compounds (both at a 97% and99% sequence identity level; results for the latter arenot shown). This may be an indication for highvariation in the functional, and thus ecological,
C68
C48
C583
C783
C51
C815
C782
C3163
C23
C535
C2
C792
C4060
C37
C824
C787
C33
C470
C14
C3143
C21
C847
C3175
C18
C836
C834
C3141
C3145
C0
C812
C35
C3
C430
C1
C25
19991900180017001600150014001300120011001000900800700600500400300200100
Mass to Charge Ratio [m/z]
PhylumProteobacteriaBacteroidetes
ClassAlphaproteobacteriaBetaproteobacteriaGammaproteobacteriaSphingobacteria
Genus
Albidiferax
Alkanindiges
Asticcacaulis
BurkholderiaCollimonasDuganella
Halomonas
HerbaspirillumHerminiimonasJanthinobacteriumKerstersiaMassilia
Mucilaginibacter
Novosphingobium
Pedobacter
PolaromonasPolynucleobacter
Pseudomonas
RhizobiumSphingopyxis
UndibacteriumVariovorax
Yersinia
0.5
0
−0.5
Spearman’srho
Figure 6 Heatmap visualising the Spearman’s rank correlation coefficients (Spearman’s ρ) from the correlation analyses between therelative abundance of the 35 most abundant (that is, 4100 reads per) OTUs at the end (rows) and the change in mass to charge ratios (m/z)from the beginning to the end of the experiment (columns). A high correlation coefficient (red) stands for a strong positive correlationbetween an OTU’s relative abundance and the decrease in m/z from the beginning to the end of the experiment. The dendrogram clustersOTUs according to Spearman’s ρ, whereas the colour columns depict affiliation of OTUs in accordance with taxonomic classification(from left to right: phylum, class and genus; colours match the legend to the right of the heatmap). The size of each bubble is proportionalto the OTU’s relative abundance.
Bacterial composition matters to functioningJB Logue et al
10
The ISME Journal
potential among closely related populations withinmicrobial communities (that is, micro-diversity; seeZimmerman et al., 2013); for example, the two asHerminiimonas classified bacterial taxa C3 and C836were generally associated with the degradation oflow- and high-molecular-weight carbon, respec-tively. Hence, our results demonstrate that thecapacity of a community to degrade DOM com-pounds cannot easily be predicted from phylogeneticinformation alone, at least not from informationderived from the 16S rRNA gene (see also Covert andMoran, 2001; Fuhrman and Hagström, 2008 andMartiny et al., 2013).
Considering the associations observed between therelative abundances of the most abundant bacteriaand the degradation of DOM compounds the ques-tion arises ‘Why do the observed degradationpatterns not look more similar, given that thesebacterial taxa were generally present in all fourcommunities?’. One explanation could be that thefunctional gene repertoire of these bacteria variedbetween experimental communities as a result ofadaptation to their original environments. Anothercould be that these abundant bacteria depend onother taxa with a different set of traits fundamental tothe degradation of certain DOM compounds (seePedler et al., 2014); taxa that are rarer and may not bepresent in all communities. Such interplay will,though, not be detectable via correlation analysis. Infact, caution has to be exercised when interpretingthe results from the correlation analysis in that itdoes not allow drawing conclusions about the causeand effect, and, as such, cannot be used to unam-biguously link a specific bacterial taxon to thedegradation and utilisation of a particular DOMcompound. In addition, size (for example, m/z)represents only one property of DOM; correlatingother properties with bacterial taxa may yield morenuanced and different associations, as well as trait-specific insights. Once associations have beenestablished, they may guide researchers to conductstudies more non-generic in character, such ascontrolled experiments in which the degradationcapacities of a single bacterial population areinvestigated. Moreover, identifying functional genesinvolved in the degradation of DOM along withassigning the chemical composition to individualDOM compounds via ultrahigh-resolution MS (forexample, Fourier transform ion cyclotron resonanceMS; see Hertkorn et al., 2008) could potentiallyprovide insight into microbial traits that may or maynot be phylogenetically constrained. Combiningsuch trait-based information with knowledge of theregulation of microbial activities, the monitoring offunctional genes (metatranscriptomics or metapro-teomics; for example, Moran, 2009; Teeling et al.,2012, respectively) and/or metabolic features (singlecell genomics; for example, Rinke et al., 2013) mayoffer a way to explore the use of individual organicmatter compounds by specific microbial taxa incomplex communities to an even greater depth and
improve our understanding of how microbial com-munity composition may affect the cycling of C inthe biosphere.
Conflict of Interest
The authors declare no conflict of interest.
AcknowledgementsWe gratefully acknowledge the help of Dolly Kothawalawith UV–visible absorbance and fluorescence trouble-shooting. We thank Jan Johansson for help of absorbanceand fluorescence sample processing in the laboratory. Wethank Jérôme Comte, Silke Langenheder and Lars Tranvikfor constructive comments and suggestions on the manu-script. Computing resources were provided by theSwedish National Infrastructure for Computing (SNIC)through the Uppsala Multidisciplinary Centre forAdvanced Computational Science (UPPMAX) underProject b2010008. Financial support to this study wasprocured by the Swedish Research Council (VR; 2010-4081 granted to ESK, CAS and ESL; 2011-5689 supportingAFA), the Swiss National Science Foundation (SNSF;PA00P3_145355 granted to JBL), the strategic researcharea ‘Biodiversity and Ecosystem Services in a ChangingClimate’ (BECC) and Helge Ax:sson Johnsons Foundation(supporting JBL), the VKR Centre of Excellence onOcean Life (supporting CAS), the Swedish ResearchCouncil for Environment, Agricultural Sciences andSpatial Planning (FORMAS) via the strong researchenvironment ‘Color of Water’ (supporting AMK), ECBONUS (BLUEPRINT supporting AFA), and FutureForest, ForWater (FORMAS), the Kempe Foundation andSITES (VR) (funding the Krycklan Catchment Study).
DisclaimerThe funders had no role in study design, data collectionand analysis, decision to publish, or preparation of themanuscript.
Author contributions
ESK conceived the study with contributions fromCAS and ESL. JBL designed the study with contribu-tions from CAS, ESL and ESK. JBL collectedenvironmental samples with assistance of HL. AMKcarried out the SPE and NJN ran the ESI-MS, whileCAS performed PARAFAC and ESI-MS analyses. JBLcollected all experimental data, performed 454-pyrosequencing analyses, analysed output data inclose collaboration with AFA and wrote first draft ofthe manuscript to which all authors contributed insubsequent revisions.
ReferencesAzam F, Fenchel T, Field JG, Gray JS, Meyerreil LA,
Thingstad F. (1983). The ecological role of water-column microbes in the sea. Marine Ecol Prog Ser 10:257–263.
Bacterial composition matters to functioningJB Logue et al
11
The ISME Journal
Battin TJ, Luyssaert S, Kaplan LA, Aufdenkampe AK,Richter A, Tranvik LJ. (2009). The boundlesscarbon cycle. Nat Geosci 2: 598–600.
Benjamini Y, Hochberg Y. (1995). Controlling the falsediscovery rate - a practical and powerful approach tomultiple testing. J R Stat Soc B Met 57: 289–300.
Berlemont R, Martiny AC. (2013). Phylogenetic distribu-tion of potential cellulases in bacteria. Appl EnvironMicrobiol 79: 1545–1554.
Bertilsson S, Eiler A, Nordqvist A, Jørgensen NOG. (2007).Links between bacterial production, amino-acid utili-zation and community composition in productivelakes. ISME J 1: 532–544.
Christian RR, Capone DG. (2002). Overview of issues inaquatic microbial ecology. In: Hurst CJ, Crawford RL,Knudsen GR, McInerney MJ, Stetzenbach LD (eds).Manual of Environmental Microbiology. 2nd edn. ASMPress: Washington, pp 323–328.
Cole JJ, Prairie YT, Caraco NF, McDowell WH, Tranvik LJ,Striegl RG et al. (2007). Plumbing the global carboncycle: Integrating inland waters into the terrestrialcarbon budget. Ecosystems 10: 171–184.
Comte J, del Giorgio PA. (2009). Links between resources,C metabolism and the major components of bacter-ioplankton community structure across a rangeof freshwater ecosystems. Environ Microbiol 11:1704–1716.
Comte J, del Giorgio PA. (2010). Linking the patternsof change in composition and function in bacterio-plankton successions along environmental gradients.Ecology 91: 1466–1476.
Covert JS, Moran MA. (2001). Molecular characterizationof estuarine bacterial communities that use high- andlow-molecular weight fractions of dissolved organiccarbon. Aquat Microb Ecol 25: 127–139.
Crump BC, Kling GW, Bahr M, Hobbie JE. (2003).Bacterioplankton community shifts in an arctic lakecorrelate with seasonal changes in organic mattersource. Appl Environ Microbiol 69: 2253–2268.
Curtis TP, Sloan WT. (2004). Prokaryotic diversity and itslimits: microbial community structure in nature andimplications for microbial ecology. Curr Opin Micro-biol 7: 221–226.
Curtis TP, Sloan WT. (2005). Exploring microbial diversity- a vast below. Science 309: 1331–1333.
del Giorgio PA, Bird DF, Prairie YT, Planas D. (1996). Flowcytometric determination of bacterial abundance inlake plankton with the green nucleic acid stainSYTO 13. Limnol Oceanogr 41: 783–789.
Dittmar T, Koch B, Hertkorn N, Kattner G. (2008). A simpleand efficient method for the solid-phase extraction ofdissolved organic matter (SPE-DOM) from seawater.Limnol Oceanogr Methods 6: 230–235.
Engelbrektson A, Kunin V, Wrighton KC, ZvenigorodskyN, Chen F, Ochman H et al. (2010). Experimentalfactors affecting PCR-based estimates of microbialspecies richness and evenness. ISME J 4: 642–647.
Fuhrman JA, Hagström Å. (2008). Bacterial and archaealcommunity structure and its patterns. In: Kirchman DL(ed). Microbial Ecology of the Oceans. 2nd edn. JohnWiley & Sons: Hoboken, pp 45–90.
Green NW, Perdue EM, Aiken GR, Butler KD, Chen HM,Dittmar T et al. (2014). An intercomparison of threemethods for the large-scale isolation of oceanicdissolved organic matter. Mar Chem 161: 14–19.
Hedges JI, Eglinton G, Hatcher PG, Kirchman DL,Arnosti C, Derenne S et al. (2000). The molecularly-
uncharacterized component of nonliving organicmatter in natural environments. Organic Geochem 31:945–958.
Herlemann DPR, Labrenz M, Jürgens K, Bertilsson S,Waniek JJ, Andersson AF. (2011). Transitions inbacterial communities along the 2000 km salinitygradient of the Baltic Sea. ISME J 5: 1571–1579.
Herlemann DPR, Manecki M, Meeske C, Pollehne F,Labrenz M, Schulz-Bull D et al. (2014). Uncouplingof bacterial and terrigenous dissolved organic matterdynamics in decomposition experiments. PLoS One 9:e93945.
Hertkorn N, Frommberger M, Witt M, Koch BP,Schmitt-Kopplin P, Perdue EM. (2008). Natural organicmatter and the event horizon of mass spectrometry.Anal Chem 80: 8908–8919.
Jansson M, Persson L, De Roos AM, Jones RI, Tranvik LJ.(2007). Terrestrial carbon and intraspecific size-variation shape lake ecosystems. Trends Ecol Evol 22:316–322.
Judd KE, Crump BC, Kling GW. (2006). Variation indissolved organic matter controls bacterial productionand community composition. Ecology 87: 2068–2079.
Kirchman DL, Dittel AI, Findlay SEG, Fischer D. (2004).Changes in bacterial activity and community structurein response to dissolved organic matter in the HudsonRiver, New York. Aquat Microb Ecol 35: 243–257.
Kothawala DN, Stedmon CA, Müller RA, Weyhenmeyer GA,Köhler SJ, Tranvik LJ. (2014). Controls of dissolvedorganic matter quality: evidence from a large-scale boreallake survey. Global Change Biol 20: 1101–1114.
Kritzberg ES, Cole JJ, Pace MM, Graneli W. (2006).Bacterial growth on allochthonous carbon in humicand nutrient-enriched lakes: Results from whole-lakeC-13 addition experiments. Ecosystems 9: 489–499.
Kujawinski EB. (2011). The impact of microbial meta-bolism on marine dissolved organic matter. Annu RevMar Sci 3: 567–599.
Landa M, Cottrell MT, Kirchman DL, Kaiser K,Medeiros PM, Tremblay L et al. (2014). Phylogeneticand structural response of heterotrophic bacteria todissolved organic matter of different chemical compo-sition in a continuous culture study. Environ Microbiol16: 1668–1681.
Langenheder S, Lindström ES, Tranvik LJ. (2005).Weak coupling between community composition andfunctioning of aquatic bacteria. Limnol Oceanogr 50:957–967.
Langenheder S, Lindström ES, Tranvik LJ. (2006).Structure and function of bacterial communities emer-ging from different sources under identical conditions.Appl Environ Microbiol 72: 212–220.
Laudon H, Taberman I, Ågren A, Futter M,Ottosson-Löfvenius M, Bishop K. (2013). The KrycklanCatchment Study-a flagship infrastructure for hydrology,biogeochemistry, and climate research in the boreallandscape. Water Resour Res 49: 7154–7158.
Lindström ES, Feng XM, Graneli W, Kritzberg ES. (2010).The interplay between bacterial community compositionand the environment determining function of inlandwater bacteria. Limnol Oceanogr 55: 2052–2060.
Logue JB, Lindström ES. (2008). Biogeography of bacter-ioplankton in inland waters. Freshwater Rev 1: 99–114.
Mardis ER. (2008). Next-generation DNA sequencingmethods. Ann Rev Genomics Hum Genet 9: 387–402.
Bacterial composition matters to functioningJB Logue et al
12
The ISME Journal
Martiny AC, Treseder K, Pusch G. (2013). Phylogeneticconservatism of functional traits in microorganisms.ISME J 7: 830–838.
Moran MA. (2009). Metatranscriptomics: eavesdroppingon complex microbial communities. Microbe 4:329–335.
Murphy KR, Butler KD, Spencer RGM, Stedmon CA,Boehme JR, Aiken GR. (2010). Measurement ofdissolved organic matter fluorescence in aquaticenvironments: an interlaboratory comparison. EnvironSci Technol 44: 9405–9412.
Nelson CE, Wear EK. (2014). Microbial diversity and thelability of dissolved organic carbon. Proc Natl Acad SciUSA 111: 7166–7167.
Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR,O'Hara RB et al. (2008). vegan: Communiy EcologyPackage. R package version 2.2-1.
Pedler BE, Aluwihare LI, Azam F. (2014). Single bacterialstrain capable of significant contribution to carboncycling in the surface ocean. Proc Natl Acad Sci USA111: 7202–7207.
Pomeroy LR. (1974). The oceans food web, a changingparadigm. Bioscience 24: 499–504.
Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza Pet al. (2013). The SILVA ribosomal RNA gene databaseproject: improved data processing and web-basedtools. Nucleic Acids Res 41: D590–D596.
Quince C, Lanzén A, Davenport RJ, Turnbaugh PJ. (2011).Removing noise from pyrosequenced amplicons. BMCBioinformatics 12:38: 1–18.
R. (2015). R: A Language and Environment for StatisticalComputing. R Foundation for Statistical Computing:Vienna, Austria. http://www.R-project.org/.
Reed HE, Martiny JBH. (2007). Testing the functionalsignificance of microbial composition in natural com-munities. FEMS Microbiol Ecol 62: 161–170.
Rinke C, Schwientek P, Sczyrba A, Ivanova NN, Anderson IJ,Cheng JF et al. (2013). Insights into the phylogeny andcoding potential of microbial dark matter. Nature 499:431–437.
Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M,Hollister EB et al. (2009). Introducing mothur: open-source, platform-independent, community-supportedsoftware for describing and comparing microbialcommunities. Appl Environ Microbiol 75: 7537–7541.
Shabarova T, Villiger J, Morenkov O, Niggemann J, DittmarT, Pernthaler J. (2014). Bacterial community structureand dissolved organic matter in repeatedly floodedsubsurface karst water pools. FEMS Microbiol Ecol 89:111–126.
Solomon S, Qin D, Manning M, Chen Z, Marquis M,Averyt KB et al. (2007). Climate Change 2007: ThePhysical Science Basis. Cambridge University Press:Cambridge.
Stedmon CA, Bro R. (2008). Characterizing dissolvedorganic matter fluorescence with parallel factoranalysis: a tutorial. Limnol Oceanogr Methods 6:572–579.
Teeling H, Fuchs BM, Becher D, Klockow C,Gardebrecht A, Bennke CM et al. (2012). Substrate-controlled succession of marine bacterioplanktonpopulations induced by a phytoplankton bloom.Science 336: 608–611.
Tranvik LJ, Downing JA, Cotner JB, Loiselle SA, Striegl RG,Ballatore TJ et al. (2009). Lakes and reservoirs asregulators of carbon cycling and climate. LimnolOceanogr 54: 2298–2314.
Wang Q, Garrity GM, Tiedje JM, Cole JR. (2007). NaiveBayesian classifier for rapid assignment of rRNAsequences into the new bacterial taxonomy. ApplEnviron Microbiol 73: 5261–5267.
Wang Y, Naumann U, Wright ST, Warton DI. (2012).mvabund - an R package for model-based analysis ofmultivariate abundance data. Methods Ecol Evol 3:471–474.
Warton DI, Wright ST, Wang Y. (2012). Distance-basedmultivariate analyses confound location and disper-sion effects. Methods Ecol Evol 3: 89–101.
Weiss MS, Abele U, Weckesser J, Welte W, Schiltz E,Schulz GE. (1991). Molecular architecture and electro-static properties of a bacterial porin. Science 254:1627–1630.
Zimmerman AE, Martiny AC, Allison SD. (2013).Microdiversity of extracellular enzyme genesamong sequenced prokaryotic genomes. ISME J 7:1187–1199.
This work is licensed under a CreativeCommons Attribution-NonCommercial-
NoDerivs 4.0 International License. The images orother third party material in this article are includedin the article’s Creative Commons license, unlessindicated otherwise in the credit line; if the materialis not included under the Creative Commons license,users will need to obtain permission from the licenseholder to reproduce the material. To view a copyof this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/
Supplementary Information accompanies this paper on The ISME Journal website (http://www.nature.com/ismej)
Bacterial composition matters to functioningJB Logue et al
13
The ISME Journal