bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
Vitamin and amino acid auxotrophy in anaerobic consortia operating under methanogenic
condition
Valerie Hubalek*1, Moritz Buck*1,2, BoonFei Tan3,4, Julia Foght3, Annelie Wendeberg5, David
5
Berry6, Stefan Bertilsson1, Alexander Eiler*#1,7
1 Department of Ecology and Genetics, Limnology and Science for Life Laboratory, Uppsala
University, Sweden
2 NBIS, National Bioinformatics Infrastructure Sweden, Uppsala, Sweden
10
3 Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada
4 Center for Environmental Sensing and Modeling, Singapore-MIT Alliance for Research and
Technology
5 Centre for Environmental Research, Environmental Microbiology, Leipzig, Germany
6 Department of Microbiology and Ecosystem Science, Division of Microbial Ecology, University
15
of Vienna, Vienna, Austria
7 eDNA solutions AB, Mölndal, Sweden
* contributed equally
20
# Corresponding author: Correspondence and requests for materials should be addressed to
Alexander Eiler [email protected]
1
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
Abstract
Syntrophy among Archaea and Bacteria facilitates the anaerobic degradation of organic
25
compounds to CH4 and CO2. Particularly during aliphatic and aromatic hydrocarbon
mineralization, as in crude oil reservoirs and petroleum-contaminated sediments, metabolic
interactions between obligate mutualistic microbial partners are of central importance1.
Using
micro-manipulation
combined
with
shotgun
metagenomic
approaches,
we
disentangled the genomes of complex consortia inside a short chain alkane-degrading
30
cultures operating under methanogenic conditions. Metabolic reconstruction revealed that
only a small fraction of genes in the metagenome-assembled genomes of this study, encode
the capacity for fermentation of alkanes facilitated by energy conservation linked to H2
metabolism. Instead, inferred lifestyles based on scavenging anabolic products and
intermediate fermentation products derived from detrital biomass was a common feature in
35
the consortia. Additionally, inferred auxotrophy for vitamins and amino acids suggests that
the hydrocarbon-degrading microbial assemblages are structured and maintained by
multiple interactions beyond the canonical H2-producing and syntrophic alkane degrader–
methanogen partnership2. Our study uncovers the complexity of ‘interactomes’ within
microbial consortia mediating hydrocarbon transformation under anaerobic conditions.
40
Microbial consortia that degrade hydrocarbons under anaerobic conditions hold the potential to
influence hydrocarbon profiles with either adverse or beneficial effects on the recovery and
refining of oil3–5. In environmental restoration, microbial degradation is seen as an effective and
environmental-friendly treatment of oil-contaminated sites. Still, in anaerobic systems,
45
bioremediation efforts are challenged by low efficiency and poor success. Additionally there is a
severe problem with unwanted contamination of adjacent systems with pollutants such as toxic
2
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
metals, naphthenic acids and polycyclic aromatic compounds6–8 as well as accelerated release of
the potent greenhouse gas methane from the biodegradation process9.
Analyses of chemical profiles and metagenomic data from hydrocarbon-degrading consortia
50
maintained in laboratory-scale cultures10–15 have identified some of the organisms and the
metabolic pathways involved in the anaerobic degradation of various hydrocarbons, such as
alkanes, benzenes and naphthalenes14,19-21,16,17. The anaerobic degradation of hydrocarbons is
usually a two- to three-step process involving fermentation either to simple substrates for
methanogenesis (hydrogen, formate, acetate, CO2) or to other metabolites (e.g. lactate, ethanol,
55
propionate, butyrate, fatty acids) that can undergo secondary fermentation yielding
aforementioned simple methanogenic substrates18–20.
Still, we are far from having a detailed understanding of the molecular mechanisms that
establish and maintain such syntrophic methanogenic consortia where methanogens deplete the
partial pressure of hydrogen facilitating the fermentation process by making it thermodynamically
60
favourable. Additional members of the consortium that scavenge metabolic byproducts and
thereby participate in the biodegradation process as secondary consumers, have thus far been
overlooked21 and their additional functions in the consortia remain hidden. One possible role of
such scavengers could be as providers of essential nutrients to salvage other consortia members
from auxotrophies while they also alter the thermodynamic equilibrium. It previously has been
65
suggested that inability of individual community members to synthesize the full range of organic
nutrients required for growth may be a favourable trait for communities and consortia operating
under energy-limited conditions1. Reconstructing metabolic pathways and assigning functions
among community members using genomics is a first step towards a more holistic understanding
of microbial consortia mediating anaerobic degradation of hydrocarbons2.
70
For this study, we extracted high-quality genomes of syntrophic bacteria from public
databases. To identify taxa that have the potential to grow syntrophically, we used gene
annotations predicting enzymes involved in reverse and direct electron transfer. Many syntrophic
3
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
bacteria are known to rely on the capacity to perform reverse electron transport-driven energyconserving H2 production through electron-confurcating hydrogenase (H2ase) in combination with
75
reduced ferredoxin21. Our analysis confirmed a wide distribution of membrane-bound, iontranslocating ferredoxin:NAD+ oxidoreductase and confurcating hydrogenase that could directly
couple the oxidation of NADH and reduced ferredoxin to produce hydrogen (Suppl. Tab. 1).
Additionally, most of the putatively syntrophic bacterial isolates lacked important genes involved
in biosynthesis of one or more amino acids and vitamins, indicating that they rely on precursors
80
(or the co-factors) provided by other bacteria to produce fully functional enzymes with required
cofactors and prosthetic groups (Suppl. Fig. 1).
Next, we attempted to resolve these interactions beyond the available genomes in public
databases by isolating natural methanogenic consortia. Short-chain alkane degrading enrichment
cultures (SCADCs) from mature fine tailings of the Mildred Lake Settling Basin13 were used to
85
obtain 13 shotgun metagenomes from various subcultures of the original enrichment culture (for
details on the metagenomes see suppl. Tab. 2). The cultures have previously been described in
several metagenomic studies inferring the broader metabolic roles and identity of the community
members at the phylum level13 (for taxonomic composition see Suppl. Fig. 2). These included
eight samples where each individual sample contained approximately ten individually picked
90
filaments, pooled based on the presence or absence of visible epibionts. The physical manipulation
for sequencing was an attempt to isolate members of the consortia that exchange co-factors or
their precursors in close physical contact such as syntrophs, versus free-living community
members with peripheral roles in alkane degradation.
From the pool of ~80 selected filaments, we obtained a metagenome co-assembly of
95
805606 scaffolds totaling ~1 Gbp with an N50-length of 4902 bp using MEGAHIT. Downstream
coverage and nucleotide composition-based binning of the 150698 scaffolds of at least 1000 bp
length, resulted in 307 bins with 45 matching our criteria for being defined as high-quality
metagenome assembled genomes (MAGs). As indicated by detailed annotations of these MAGs
4
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
using the ‘MicroScope’ pipeline22, the most likely hydrocarbon degraders in the culture are
100
putative syntrophic bacteria related to Syntrophus, Pelotomaculum, Desulfobacterium and
Clostridium, all having the potential for n-alkane activation, betaoxidation and fumarate
regeneration. The genes indicative for these pathways were alkylsuccinate synthase and
benzylsuccinate synthase (C-H activation), methylmalonyl CoA mutase (carbon skeleton
rearrangement) and methylmalonyl-CoA decarboxylase (decarboxylation) for n-alkane activation;
105
enoyl-CoA hydratase, β-hydroxyacyl-CoA dehydrogenase and acyl-CoA acetyl-transferase for
beta-oxidation; propionyl-CoA carboxylase, methylmalonyl-CoA mutase, succinyl-CoA
synthetase and succinate dehydrogenase for fumarate regeneration14. Most genomes encoded
enzymes associated with fermentation and coupled pathways for acetogenesis (i.e. reductive
acetyl-CoA or Wood-Ljungdahl pathway), a process by which acetate is produced from CO2 and
110
an electron donor (Figure 1). The five methanogen genomes reconstructed from the contig bins
belonged to the genera Methanoculleus and Methanosaeta in the class Methanomicrobia and to
the class Thermoplasmata As previously described, Methanosaeta represent the acetoclastic
methanogens while Methanoculleus represent hydrogenotrophic methanogens, and
Thermoplasmata represent methylotrophic methanogens23. Besides these differences in substrates
115
used for methane production, they all encode complete sets of nitrogen fixation genes. For detailed
exploration of the genomes we direct the reader to the online sources of the ‘MicroScope’
pipeline22 where each genome’s individual KEGG and biocyc maps can be inspected.
In addition to classical syntrophic interactions described in hydrocarbon-degrading
communities, i.e. where one partner produces metabolites such as H2, acetate, ethanol and/or small
120
organic intermediates, which are then used by methanogens and secondary degraders, our genomic
reconstructions point to additional metabolic dependencies. Firstly, the scavenging of metabolic
products derived from detrital biomass is indicated by the presence of genes encoding transporters
for a wide diversity of proteins and fatty acids, as well as genes involved in the degradation of
intermediate fermentation products (i.e. C1 compounds and alcohols) (Figure 2). Genomes of
5
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
125
these secondary degraders belonged to the deltaproteobacterial genera Desulfovibrio,
Desulfomicrobium and Desulfobulbus previously reported to use H2, organic acids, formate, or
short-chain alcohols as electron donors for sulfate reduction24–26. In addition to their potential to
grow by using anaerobic respiration, these genera possessed genes associated with fermentation of
a wide range of alcohols and organic acids such as lactate, amino and carboxylic acids.
130
Like many of their cultivated and genome sequenced relatives, the MAGs from the putative
sulfate reducers in these consortia revealed auxotrophies for vitamins and amino acids (Figure 3).
Even though the absence of pathways, defined by less than half of the genes present in the
respective pathways, need to be assessed with caution because of the partial nature of the
individual MAGs, analogous auxotrophies in closely related MAGs and closed genomes from
135
public databases provide strong support for widespread auxotrophy. Examples of essential
metabolite pathways whose biosynthetic gene distribution is patchy in the SCADC MAGs are
those for biosynthesis of vitamins B6 (pyridoxal), B12 (adenosylcobalamin), and H (biotin).
Earlier large-scale genome sequence analyses have previously suggested that corrinoids, essential
cofactors for vitamins27, are widely shared by over 70% of the sequenced bacterial species, while
140
less than 25% of species synthesize corrinoids de novo28,29. The existence of over a dozen
structurally distinct corrinoids produced and used by various Bacteria and Archaea30 can thus hold
a central role in dictating microbial community assembly and functioning, as it is clear that the
structure of the precursor, which can be a benzimidazole, purine or a phenolic compound30,
influences the degree to which a particular corrinoid can be used by a given organism31,32.
145
Likewise, the absence of genes underlying biosynthesis of essential amino acids is apparent
in particular taxonomic groups (Figure 3B). The loss of essential biosynthesis function in various
taxonomic groups has previously been associated with energy- and nutrient-limitation selecting for
small cells with reduced genomes such as found in the open ocean and deep biosphere33,34.
Particularly costly functions are absent in most members of the SCADC community, resulting in a
150
community critically depending on essential metabolite providers. Biosynthesis of certain vitamins
6
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
such as adenosylcobalamin and specific amino acids such as phenylalanine, tyrosine and
tryptophan, is energetically costly and require a large set of genes. This loss of function and
subsequent adaptations to a cooperative lifestyle under nutrient- and energy-limitation forms the
foundation for the Black Queen hypothesis35.
155
In the context of the SCADC, the Black Queen hypothesis implies that, in addition to
fermentation products disposal by methanogens, other types of “public goods” can be exchanged.
Such "public goods" are provided by a few taxa hosting the enzymes to perform this energetically
costly biosynthesis while benefitting the other organisms in their vicinity. This enhanced cross
feeding is logical if one considers the need for metabolic processes to be optimized for reduced
160
metabolic burden at the level of an individual cell living at the thermodynamic limit.
To conclude, we show that the interactome of syntrophic alkane degrader–methanogen
partnerships surpass the standard electron- and simple carbon compound-transfer. Biosynthesis of
vitamins and other energy-expensive metabolites seem to be unequally distributed among
community members and we propose that this facilitate energy conservation at the thermodynamic
165
limit. It can be further speculated that vitamin and amino acid augmentations could possibly
enhance petroleum bioaugmentation processes in remediation of hydrocarbon-contaminated sites.
Methods
Culturing. An enriched short-chain alkane-degrading culture (SCADC) was enriched from mature
170
fine tailings of the Mildred Lake Settlings Basin as detailed by Tan et al.13. In short, the primary
enrichment cultures were established in multiple replicates by adding 25 ml oil sand tailings to 50
ml of mineral methanogenic medium36, under a headspace of O2-free 30% CO2–N2 balance gas.
After an initial 1-year incubation in the dark at room temperature, the primary cultures were
pooled and used to inoculate multiple replicates of first transfer cultures into fresh methanogenic
175
medium. After a second 6-month incubation, these cultures were pooled as an inoculum for the
experimental SCADC enrichment cultures, where 0.1% (v/v) of an alkane mixture comprising
equal volumes of C6-C9 compounds was added to the community and incubated at least 4 months
7
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
at 28 °C in the dark, prior to further subculturing13. The samples analyzed in the present study
were from several transfers of the parent SCADC culture. Three bottles containing 75-ml cultures
180
were each sampled three times using micromanipulation, originally to capture filamentous bacteria
seen to carry attached smaller cocci (Fig. 1). Fluorescence in situ hybridization (FISH) using
probes MX825b and MX825c37 revealed that most filaments represent Methanosaeta with
associated bacterial cells.
185
Filament sampling. Micromanipulation was performed using an Eppendorf TransferMan® NK 2
at the Department of Microbial Ecology, University of Vienna. Each slide used for
micromanipulation was prepared by cleaning with “DNA AWAY” surface decontaminant
(ThermoFisher Scientific) and rinsed in sterile phosphate-buffered saline. Filaments were picked
from a 12 µl culture subsample by using microscopy (Axio Observer.D1, Zeiss) at 32x
190
magnification, captured in a 4 µm diameter capillary (TransferTip®, Eppendorf) and subsequently
transferred into a sterile tube. Approximately 10 filaments were collected from each of eight
cultures and pooled according to coherent morphology.
Samples with pooled filaments (n = 8) were then lysed by subjecting them to four freezethaw cycles, freezing at -80°C followed by heat treatment at 70°C for 5 minutes. These samples
195
were subjected to whole genome amplification with multiple displacement amplification (MDA)
using the REPLI-g® Mini kit according to the manufacturer’s instructions (Qiagen).
Whole culture DNA isolation. From the SCADC culture, 100 ml were filtered through 0.2 µm
Supor® Membrane Disc Filters (Pall). Prior to bead beating, the above mentioned freeze-thaw cell
200
lysis was conducted, as preceding tests showed that the filaments needed a more vigorous
treatment. The DNA was then isolated from the filters by bead-beating using the PowerSoil®
DNA Isolation Kit (MoBio) according to the manufacturer’s instructions, and sequenced as
8
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
described below. No whole genome amplification with multiple displacement amplification
(MDA) was performed on these samples.
205
Library preparation and sequencing. The SciLifeLab SNP/SEQ sequencing facility at Uppsala
University performed the library preparation and the sequencing. First, 10 ng of genomic DNA
was sheared using a focused ultrasonicator (Covaris E220). The sequencing libraries were
prepared with the Thruplex FD Prep kit (Rubicon Genomics) according to manufacturer’s
210
protocols (R40048-08, QAM-094-002). Library size selection was made using AMPpure XP
beads (Beckman Coulter) in 1:1 ratio with the DNA sample volume. The prepared sample libraries
were quantified using KAPA Biosystems next-generation sequencing library qPCR kit and
analysing using a StepOnePlus (Life Technologies) real-time PCR instrument. The quantified
libraries were then prepared for sequencing on the Illumina HiSeq sequencing platform with a
215
TruSeq paired-end cluster kit, v3, and Illumina’s cBot instrument to generate a clustered flowcell
for sequencing. Sequencing of the flowcell contents was performed using an Illumina HiSeq2500
sequencer with Illumina TruSeq SBS sequencing kit v3, following a 2x100 indexed high-output
run recipe.
220
Assembly and binning of metagenome reads. Reads were filtered based on their quality scores
using Sickle (version 1.210)38. The read-set was digitally normalized using khmer
(arXiv:1203.4802) and subsequently assembled with MEGAHIT39. Contigs were scaffolded with
the BESST software40. Coverage was computed by mapping back the reads to the scaffolds using
bbmap (BBMap - Bushnell B. - sourceforge.net/projects/bbmap/) and for computing coverage,
225
bedtools
(version
2.18.2)41
(doi:10.7717/peerj.1165)
was
used.
Scaffolds
were
binned
using
MetaBAT
The criteria for high quality bins, so-called MAGs, are based on
CheckM(doi:10.1101/gr.186072.114) estimates of completeness (>70%) and contamination
(<5%).
9
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
230
Processing of mined data. In addition to data acquired from the newly sequenced samples, we also
accessed three datasets from the same original SCADC, downloaded from the NCBI Short Read
Archive (SRA): SRR636559 (a high depth HiSEQ dataset), and SRR634694 and SRR634695 (two
Roche 454 datasets). These archival sequences were treated bioinformatically in the same way as
the new sequences and co-assembled with them.
235
Annotation. Initial phylogenetic attribution of the MAGs was performed using PhyloPhlAn42 on
proteins predicted with Prokka (version 1.9)43, supplemented with a custom set of MAGs, single
amplified genomes (SAGs) and whole genome sequences accessed from JGI IMG/M. PFAM
annotations were performed using the Pfam database 27.044 and HMMER45 (version 3.1b1) with
240
default parameters. Detailed genome annotation analyses were performed using the ‘MicroScope’
pipeline22 with automatic annotations assisted by manual curation as described in the integrated
bioinformatics tools and the proposed annotation rules. Reconstructed MAGs are deposited on
MicroScope platform22 with identifier ‘scadc_MAG’.
245
Acknowledgements
We acknowledge the Uppsala Multidisciplinary Center for Advanced Computational Science
(UPPMAX) for access to data storage and computing resources under project b2013274. This
work was supported by the Swedish Research Council VR (Grant 2012-4592 to AE and 20123892 to SB), the Swedish Foundation for Strategic Research (Grant ICA010-015 to AE), Swedish
250
Research Council Formas (project 2012-986 to SB) and the Helmholtz Initiative and Networking
Fund.
1
0
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
255
References
1.
Morris, B. E. L., Henneberger, R., Huber, H. & Moissl-Eichinger, C. Microbial syntrophy:
interaction for the common good. FEMS Microbiol. Rev. 37, 384–406 (2013).
2.
Sieber, J. R., McInerney, M. J. & Gunsalus, R. P. Genomic Insights into Syntrophy: The
Paradigm for Anaerobic Metabolic Cooperation. Annu. Rev. Microbiol. 66, 429–452
260
(2012).
3.
Head, I. M., Jones, D. M. & Larter, S. R. Biological activity in the deep subsurface and the
origin of heavy oil. Nature 426, 344–352 (2003).
4.
Gray, N. D., Sherry, A., Hubert, C., Dolfing, J. & Head, I. M. Methanogenic degradation of
petroleum hydrocarbons in subsurface environments remediation, heavy oil formation, and
265
energy recovery. Adv. Appl. Microbiol. 72, 137–61 (2010).
5.
Berdugo-Clavijo, C., Dong, X., Soh, J., Sensen, C. W. & Gieg, L. M. Methanogenic
biodegradation of two-ringed polycyclic aromatic hydrocarbons. FEMS Microbiol. Ecol.
81, 124–133 (2012).
6.
270
Quagraine, E. K., Peterson, H. G. & Headley, J. V. In Situ Bioremediation of Naphthenic
Acids Contaminated Tailing Pond Waters in the Athabasca Oil Sands Region—
Demonstrated Field Studies and Plausible Options: A Review. J. Environ. Sci. Heal. Part A
40, 685–722 (2005).
7.
Kelly, E. N. et al. Oil sands development contributes polycyclic aromatic compounds to the
Athabasca River and its tributaries. Proc. Natl. Acad. Sci. 106, 22346–22351 (2009).
275
8.
Kelly, E. N. et al. Oil sands development contributes elements toxic at low concentrations
to the Athabasca River and its tributaries. Proc. Natl. Acad. Sci. 107, 16178–16183 (2010).
9.
Change, C. The Physical Science Basis, Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change, 2007, S. Solomon,
D. Qin, M. (2007).
280
10.
Schink, B. Degradation of unsaturated hydrocarbons by methanogenic enrichment cultures.
1
1
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
FEMS Microbiol. Ecol. 1, 69–77 (1985).
11.
Zengler, K., Richnow, H. H., Rossello-Mora, R., Michaelis, W. & Widdel, F. Methane
formation from long-chain alkanes by anaerobic microorganisms. Nature 401, 266–269
(1999).
285
12.
Wang, L.-Y. et al. Characterization of an alkane-degrading methanogenic enrichment
culture from production water of an oil reservoir after 274 days of incubation. Int.
Biodeterior. Biodegradation 65, 444–450 (2011).
13.
Tan, B., Dong, X., Sensen, C. W. & Foght, J. Metagenomic analysis of an anaerobic
alkane-degrading microbial culture: potential hydrocarbon-activating pathways and inferred
290
roles of community members. Genome 56: 599–61, (2013).
14.
Tan, B., Semple, K. & Foght, J. Anaerobic alkane biodegradation by cultures enriched from
oil sands tailings ponds involves multiple species capable of fumarate addition. 1–42
(2015).
15.
295
Tan, B. et al. Comparative analysis of metagenomes from three methanogenic hydrocarbondegrading enrichment cultures with 41 environmental samples. ISME J (2015).
16.
Wawrik, B. et al. Field and laboratory studies on the bioconversion of coal to methane in
the San Juan Basin. FEMS Microbiol. Ecol. 81, 26–42 (2012).
17.
Lyles, C. N., Le, H. M., Beasley, W. H., McInerney, M. J. & Suflita, J. M. Anaerobic
hydrocarbon and fatty acid metabolism by syntrophic bacteria and their impact on carbon
300
steel corrosion. Front. Microbiol. 5, 114 (2014).
18.
Schink, B. Synergistic interactions in the microbial world. Antonie Van Leeuwenhoek 81,
257–261 (2002).
19.
McInerney, M. J. et al. Physiology, Ecology, Phylogeny, and Genomics of Microorganisms
Capable of Syntrophic Metabolism. Ann. N. Y. Acad. Sci. 1125, 58–72 (2008).
305
20.
Struchtemeyer, C. G., Duncan, K. E. & McInerney, M. J. Evidence for syntrophic butyrate
metabolism under sulfate-reducing conditions in a hydrocarbon-contaminated aquifer.
1
2
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
FEMS Microbiol. Ecol. 76, 289–300 (2011).
21.
Narihiro, T., Nobu, M. K., Kim, N.-K., Kamagata, Y. & Liu, W.-T. The nexus of
syntrophy-associated microbiota in anaerobic digestion revealed by long-term enrichment
310
and community survey. Environ. Microbiol. 17, 1707–1720 (2015).
22.
Vallenet, D. et al. MaGe: a microbial genome annotation system supported by synteny
results. Nucleic Acids Res. 34, 53–65 (2006).
23.
Poulsen, M. et al. Methylotrophic methanogenic Thermoplasmata implicated in reduced
methane emissions from bovine rumen. Nat. Commun. 4, 1428 (2013).
315
24.
Pagani, I. et al. Complete genome sequence of Desulfobulbus propionicus type strain
(1pr3T). Stand. Genomic Sci. 4, 100–110 (2011).
25.
Jensen, A. & Finster, K. Isolation and characterization of Sulfurospirillum carboxydovorans
sp. nov., a new microaerophilic carbon monoxide oxidizing epsilon Proteobacterium.
Antonie Van Leeuwenhoek 87, 339–353 (2005).
320
26.
Gogotova, G. I. & Vainshtein, M. B. Description of a sulfate-reducing bacterium,
Desulfobacterium macestii sp-now, which is capable of autotrophic growth. Microbiology
58, 64–68 (1989).
27.
Banerjee, R. & Ragsdale, S. W. The many faces of Vitamin B12: catalysis by cobalamindependent enzymes 1. Annu. Rev. Biochem. 72, 209–247 (2003).
325
28.
Degnan, P. H., Barry, N. A., Mok, K. C., Taga, M. E. & Goodman, A. L. Human gut
microbes use multiple transporters to distinguish vitamin B 12 analogs and compete in the
gut. Cell Host Microbe 15, 47–57 (2014).
29.
Zhang, Y., Rodionov, D. A., Gelfand, M. S. & Gladyshev, V. N. Comparative genomic
analyses of nickel, cobalt and vitamin B12 utilization. BMC Genomics 10, 1 (2009).
330
30.
Renz, P. & Banerjee, R. Biosynthesis of the 5, 6-dimethylbenzimidazole moiety of
cobalamin and of the other bases found in natural corrinoids. Chem. Biochem. B12. 557–
575 (1999).
1
3
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
31.
Mok, K. C. & Taga, M. E. Growth inhibition of Sporomusa ovata by incorporation of
benzimidazole bases into cobamides. J. Bacteriol. 195, 1902–1911 (2013).
335
32.
Men, Y. et al. Identification of specific corrinoids reveals corrinoid modification in
dechlorinating microbial communities. Environ. Microbiol. 17, 4873–4884 (2015).
33.
Giovannoni, S. J., Thrash, J. C. & Temperton, B. Implications of streamlining theory for
microbial ecology. ISME J. 8, 1553–1565 (2014).
34.
340
Wu, X. et al. Microbial metagenomes from three aquifers in the Fennoscandian shield
terrestrial deep biosphere reveal metabolic partitioning among populations. ISME J in
Press, (2015).
35.
Morris, J. J., Lenski, R. E. & Zinser, E. R. The Black Queen Hypothesis: Evolution of
Dependencies through Adaptive Gene Loss. mBio 3, (2012).
36.
345
Widdel, F. & Bak, F. Gram-negative mesophilic sulfate-reducing bacteria. prokaryotes.
Springer New York 3352–3378, (1992).
37.
Crocetti, G., Murto, M. & Björnsson, L. An update and optimisation of oligonucleotide
probes targeting methanogenic Archaea for use in fluorescence in situ hybridisation (FISH).
J. Microbiol. Methods 65, 194–201 (2006).
38.
350
Joshi, N. & Fass, J. Sickle: A sliding-window, adaptive, quality-based trimming tool for
FastQ files. [Software]. Available at https://githubcom/najoshi/sickle. (2011).
39.
Li, D., Liu, C.-M., Luo, R., Sadakane, K. & Lam, T.-W. MEGAHIT: An ultra-fast singlenode solution for large and complex metagenomics assembly via succinct de Bruijn graph.
Bioinforma. (2015). doi:10.1093/bioinformatics/btv033
40.
355
Sahlin, K., Vezzi, F., Nystedt, B., Lundeberg, J. & Arvestad, L. BESST--efficient
scaffolding of large fragmented assemblies. BMC Bioinformatics 15, 281 (2014).
41.
Quinlan, A. R. & Hall, I. M. BEDTools: a flexible suite of utilities for comparing genomic
features. Bioinformatics 26, 841–842 (2010).
42.
Segata, N., Börnigen, D., Morgan, X. C. & Huttenhower, C. PhyloPhlAn is a new method
1
4
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
for improved phylogenetic and taxonomic placement of microbes. Nat. Commun. 4, 2304
360
(2013).
43.
Seemann, T. Prokka: rapid prokaryotic genome annotation. Bioinformatics 30, 2068–9
(2014).
44.
Finn, R. D. et al. Pfam: the protein families database. Nucleic Acids Res. 42, D222–D230
(2014).
365
45.
Eddy, S. R. Accelerated Profile HMM Searches. PLoS Comput Biol 7, e1002195 (2011).
Figure and Tables
Figure 1. A phylogenomic tree of recovered genomes as computed by PhyloPhlAn41. The
colorstrip delineates taxonomic affiliation. The symbols show the metabolic traits, based on
370
genome annotations. Traits with high support are indicated by closed symbols, while traits with
low support have open symbols, and categories with missing symbols indicate that no indications
of the trait were found. The two barcharts on the far right show the number of contigs assigned to
each bin and the assembly size (in Mbp) contained in the recovered genome bins.
375
Figure 2. Model of potential metabolic pathways and metabolic cross-feeding in the short-chain
alkane degrading culture (SCADC) representing (green) archaeal filaments, (blue) alkane
degraders, (red) intermediate metabolite scavengers and secondary degraders, (grey) acetogenic
bacteria. In addition, the figure shows the inferred syntrophic interactions from hydrogen (green),
carbon dioxide (dark blue) and acetate (green) to intermediate fermentation metabolites (IFMs;
380
violet) transfer as well as other forms of metabolic interdependencies such as cross-feeding of
costly metabolites such as vitamins (red) and amino acids (black).
Figure 3. Heat map of amino acid (A) and vitamin (B) biosynthetic capabilities of taxonomic bins
obtained from the short-chain alkane degrading culture (SCADC). Prototrophy predictions for
1
5
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
385
each amino acid and vitamin are based on 'pathway completion' value i.e, the number of reactions
for pathway x in a given organism/total number of reactions in the same pathway x defined in the
MetaCyc or KEGG databases. Value of 1/0 indicates all/none of the key enzymes involved in the
biosynthesis. Dendograms represent clustering based on potential biosynthesis profiles. To verify
widespread auxotrophy for vitamins and amino acids in syntrophic prokaryotes, a similar analysis
390
was performed using genomes available from genera representing our taxonomic bins (for details
see Supplementary Figure 1). As a definition of absence we used a threshold of less than 0.5
pathway completeness.
Supplementary Material
395
Supplementary Figure 1. Heat map of amino acid (A) and vitamin (B) biosynthetic capabilities of
selected genomes obtained from MaGe. Prototrophy predictions for each amino acid and vitamin
are based on 'pathway completion' value i.e, the number of reactions for pathway x in a given
organism/total number of reactions in the same pathway x defined in the MetaCyc or KEGG
databases. Value of 1/0 indicates all/none of the key enzymes involved in the biosynthesis.
400
Dendograms represent clustering based on potential biosynthesis profiles. This resembles wide
spread auxotrophy for vitamins and amino acids as inferred for the short-chain alkene degrading
culture (SCADC).
Supplementary Figure 2. Taxonomic composition of the SCADC in the different samples used for
405
shotgun metagenomic sequencing. The 16S rRNA gene Data was produced following the
procedure as outlined in Sinclair et al. (2015).
Supplementary Table 1. Distribution of membrane-bound, ion-translocating ferredoxin:NAD+
oxidoreductase and confurcating hydrogenase that could directly couple the oxidation of NADH
410
and reduced ferredoxin to produce hydrogen.
1
6
bioRxiv preprint first posted online Apr. 19, 2017; doi: http://dx.doi.org/10.1101/128660. The copyright holder for this preprint (which was
not peer-reviewed) is the author/funder. It is made available under a CC-BY-NC-ND 4.0 International license.
Supplementary Table 2. Source of metagenome and summary of sequencing data. “+Epibiont”
indicates samples including filaments with epibioints whereas “-Epibioints” without epibionts.
1
7