Metabolite cross-feeding enables concomitant catabolism of chlorinated methanes and chlorinated ethenes in synthetic microbial assemblies

Abstract Isolate studies have been a cornerstone for unraveling metabolic pathways and phenotypical (functional) features. Biogeochemical processes in natural and engineered ecosystems are generally performed by more than a single microbe and often rely on mutualistic interactions. We demonstrate the rational bottom-up design of synthetic, interdependent co-cultures to achieve concomitant utilization of chlorinated methanes as electron donors and organohalogens as electron acceptors. Specialized anaerobes conserve energy from the catabolic conversion of chloromethane or dichloromethane to formate, H2, and acetate, compounds that the organohalide-respiring bacterium Dehalogenimonas etheniformans strain GP requires to utilize cis-1,2-dichloroethenene and vinyl chloride as electron acceptors. Organism-specific qPCR enumeration matched the growth of individual dechlorinators to the respective functional (i.e. dechlorination) traits. The metabolite cross-feeding in the synthetic (co-)cultures enables concomitant utilization of chlorinated methanes (i.e. chloromethane and dichloromethane) and chlorinated ethenes (i.e. cis-1,2-dichloroethenene and vinyl chloride) without the addition of an external electron donor (i.e. formate and H2). The findings illustrate that naturally occurring chlorinated C1 compounds can sustain anaerobic food webs, an observation with implications for the development of interdependent, mutualistic communities, the sustenance of microbial life in oligotrophic and energy-deprived environments, and the fate of chloromethane/dichloromethane and chlorinated electron acceptors (e.g. chlorinated ethenes) in pristine environments and commingled contaminant plumes.


Introduction
In the environment, microorganisms usually exist within taxonomically and functionally diverse communities, where metabolically active cells constantly interact, e.g. by competing for nutrients or exchanging metabolites [1][2][3].These microbemicrobe interactions can have beneficial, neutral, or harmful effects on members of the community and are key factors determining community function, response to perturbations, and resilience.For example, cross-feeding is an interaction between microorganisms in which molecules produced and released by one microbe (e.g.vitamins, amino acids, and metabolites) are utilized by another [4,5].Cross-feeding interactions can be unidirectional or bidirectional and represent common mutualistic relationships between microbes, providing a driving force for the maintenance of diversity and functional stability in microbial communities [5].The lack of a thorough understanding of these microbial interactions impedes our ability to obtain many interesting organisms in axenic culture and develop a predictive understanding of biogeochemical processes.The knowledge-based design of synthetic communities with desirable functions by combining predefined axenic or enrichment cultures to achieve a desired function (i.e.synthetic ecology) has great promise but requires understanding of relevant microbe-microbe interactions [6][7][8].Disentangling microbial interactions in natural ecosystems, such as the gut or soil, is challenging due to the intrinsic complexity of such relationships [9].High-throughput sequencing provides insights into the taxonomic compositions of microbiomes and can reveal the metabolic potentials of community members; however, many challenges exist for microbial interaction network reconstruction and prediction of mutualistic relationships [10].Detailed laboratory studies with synthetic microbial assemblies (i.e.synthetic microbiomes) can reveal important metabolic interactions underlying the functional attributes of a microbial community [11][12][13].
Many synthetic chemicals, such as chlorinated methanes and chlorinated ethenes, are notorious groundwater pollutants.By no means are chlorinated C 1 and C 2 compounds strictly synthetic xenobiotics (gr.xenos, foreign; bios, life; foreign to life), and natural systems produce a range of organohalogens.More than 5,000 halogenated chemicals, the majority of them chlorinated, are released into the environment by biological and abiotic (geo)chemical processes [14][15][16].Chloromethane (CM) is formed during the degradation of organic matter [17][18][19], the senescence of plant material [20], and biomass burning [21], and is produced by marine macroalgae [22,23].CM is the most abundant organohalogen in the atmosphere, with annual global emissions from marine and terrestrial sources estimated to exceed 4,000 Gg [24][25][26].The majority of dichloromethane (DCM) in the environment is of anthropogenic origin, with natural emissions from oceans, wetlands, mangroves, and volcanoes contributing ∼63 Gg of DCM annually to the atmosphere [27][28][29][30][31][32].In addition to CM and DCM, a diversity of organohalogens, including chlorinated ethenes, is produced through biotic and abiotic processes in marine and terrestrial environments [15,16].For example, vinyl chloride (VC) can abiotically form in soils through ironcatalyzed oxidation of humic acids and has been detected in volcanic emissions [33,34].
To explore the poorly understood microbial interactions during the utilization of CM and DCM as sources of reducing equivalents (i.e.electron donors) and organohalogens (e.g.VC and cDCE) as electron acceptors, synthetic co-cultures were established.The co-cultures comprised an anaerobic CM-or DCM-degrading bacterium [36,37] and the organohalide-respiring bacterium D. etheniformans strain GP, which is the first non-Dehalococcoides isolate capable of metabolic reductive dechlorination of trichloroethene (TCE), all dichloroethene isomers, and VC to ethene with formate or H 2 as electron donor and acetate as carbon source [42,43].Metabolite cross-feeding supported growth of both population types and enabled concomitant utilization of the chlorinated methanes and the chlorinated ethenes as energy sources in the synthetic microbiomes without the need for an external electron donor (i.e.H 2 and formate) and a carbon source (i.e.acetate).The new findings highlight the role of the natural chlorine cycle for the development of interdependent, mutualistic microbial communities and have implications for the microbial ecology in pristine and contaminated ecosystems.

Cultivation and synthetic microbiomes
Routine cultivation of the axenic cultures of Dehalobacterium formicoaceticum strain DMC, strain CM, and D. etheniformans strain GP, and consortium RM comprising the DCM-utilizing 'Ca.Dichloromethanomonas elyunquensis' used 160-ml glass serum bottles containing 50 ml of bicarbonate-buffered (30 mM, pH 7.3) basal salt medium reduced with 0.2 mM Na 2 S, 0.2 mM l-cysteine, and 0.5 mM dithiothreitol (DTT) and amended with resazurin (0.25 mg L −1 ) as redox indicator [44].The bottles were sealed with butyl rubber stoppers (Bellco Glass Inc., Vineland, NJ, USA) under a headspace of N 2 /CO 2 (80/20, vol/vol).Strain CM received 3 ml of CM gas (122.6 μmol), and Dehalobacterium formicoaceticum strain DMC and consortium RM were provided with 5 μl of neat DCM (78.3 μmol) as the sole energy source.D. etheniformans strain GP was grown in basal salt medium amended with 2 mM formate, 2 ml VC (81.7 μmol, 1.6 mM nominal concentration), and 1 mM acetate as electron donor, electron acceptor, and carbon source, respectively.Strain CM was isolated with CM as the sole energy source from a salt-f lat mud sample.The strict anaerobic bacterium ferments CM to formate and acetate in bicarbonate-buffered basal salt medium.The isolate's 16S rRNA gene (GenBank accession number MW682303) shares 95.4% sequence similarity with Sporobacter termitidis strain SYR T (Z49863.1)[45]; however, the classification of strain CM has to await its polyphasic characterization.
The co-culture batch experiments with strain CM and D. etheniformans strain GP were performed in 160-ml glass serum bottles containing 50 ml basal salt medium lacking reduced carbon compounds (e.g.formate and acetate).The reactor vessels were inoculated (3%, vol/vol, ∼10 7 cells ml −1 ) with strain CM and strain GP cultures.CM (∼50 μmol) was amended as the substrate for strain CM and VC (∼50 μmol) or cDCE (∼40 μmol) was amended as the electron acceptor for D. etheniformans strain GP.The vessels received additional ∼50 μmol CM when CM had been depleted.Dehalobacterium formicoaceticum strain DMC and D. etheniformans strain GP co-cultures received 3 μl DCM (47.0 μmol) and 1 ml VC (40.9 μmol).Upon DCM depletion, the cultures received additional 3or 5-μl doses of DCM.All vessels were incubated in upright position at 30 • C in the dark without agitation.The consumption of chlorinated substrates (i.e.DCM, CM, cDCE, and VC) and the formation of the reductive dechlorination product ethene were measures of activity.During growth, culture suspension samples (1 ml) were filtered onto 0.22 μm Durapore membranes (Millipore, Cork, Ireland), which were immediately stored at −80 • C for subsequent DNA extraction and qPCR analysis.Culture suspension samples (1 ml) were also collected for acetate and formate measurement using HPLC.Co-culture suspension samples were regularly collected and visually examined with a Zeiss AX10 light microscope (Jena, Germany).Axenic cultures (D. etheniformans, Dehalobacterium formicoaceticum, and strain CM) and consortium RM amended with a chlorinated methane (CM or DCM) and a chlorinated ethene (VC or cDCE) served as controls.Incubation vessels receiving autoclaved inocula served as abiotic controls.

Quantitative polymerase chain reaction (qPCR)
Organism-specific qPCR assays each targeting the 16S rRNA gene of strain CM, Dehalobacterium formicoaceticum strain DMC, and 'Ca.Dichloromethanomonas elyunquensis' strain RM, and the VC reductive dehalogenase gene (cerA) of D. etheniformans strain GP were used to enumerate cell numbers and monitor the growth of the respective populations in the co-cultures.The genomes of the bacteria included in this study harbor a single 16S rRNA gene, except Dehalobacterium formicoaceticum strain DMC, with four copies per genome [46][47][48].DNA was extracted using the DNeasy PowerLyzer PowerSoil DNA isolation kit (Qiagen, Hilden, Germany) following the manufacturer's protocol.qPCR primers and probes are listed in Table 1 or have been reported [38,42].The qPCR

Analytical methods
CM, DCM, cDCE, VC, ethene, and methane (CH 4 ) were measured by manual headspace injections (0.1 ml) into an Agilent 7890 gas chromatograph (GC) (Santa Clara, CA, USA) equipped with a DB-624 column (60 m length, 0.32 mm i.d., 1.8 μm film thickness) and a f lame ionization detector (FID).The GC inlet was maintained at 200 • C, the GC oven temperature was kept at 60 • C for 2 min, followed by an increase to 200 • C at a ramping rate of 25 • C min −1 , and the FID was operated at 300 • C as described [39].The detection limits for DCM, CM, cDCE, VC, ethene, and methane were in the range of 0.01 -0.08 μmol per 160 ml glass serum bottle.Acetate and formate were analyzed using an Agilent 1200 series high-performance liquid chromatography (HPLC) system equipped with an Aminex HPX-87H column (Bio-Rad, Hercules, CA, USA) and a UV detector set to 210 nm.The separation occurred at a column temperature of 30 • C with isocratic elution (4 mM H 2 SO 4 ) at a f low rate of 0.6 ml/min −1 for 25 min [49].Aqueous samples were acidified with 1 M H 2 SO 4 prior to HPLC analysis.The detection limit for acetate and formate was ∼0.5 μmol per bottle (equals to ∼0.01 mM).

CM fermentation supports reductive dechlorination of cDCE and VC to ethene
In co-cultures of the CM degrader strain CM and D. etheniformans strain GP, the initial amount of CM (51.0 ± 1.0 μmol) was consumed within 3 days, and three subsequent additions of CM (∼200 μmol total) were utilized over a 7-day incubation period (Fig. 1A).Concomitant with CM degradation, formate and acetate formed in the co-cultures, with maximum amounts of 51.4 ± 2.0 μmol of formate and 78.9 ± 3.1 μmol of acetate observed on Day 7 (Fig. 1B).Following CM depletion, formate decreased to 19.7 ± 2.3 μmol, whereas the amount of acetate remained unchanged over a 24-day incubation period (Fig. 1B).The initial amount of VC (55.6 ± 1.2 μmol) was completely reductively dechlorinated to 50.3 ± 0.3 μmol of ethene in the co-cultures after 20 days (Fig. 1A).The number of strain CM cells in the cocultures, determined by 16S rRNA gene-targeted qPCR, increased from (1.6 ± 0.3) × 10 8 to (2.4 ± 0.2) × 10 9 copies per ml, a 15.6fold increase, during the CM degradation phase (i.e. the initial 7 days of incubation) (Fig. 1C).The number of D. etheniformans strain GP cells, determined by qPCR enumeration of the VC reductive dehalogenase gene cerA, increased from (2.7 ± 0.7) × 10 7 to (4.3 ± 0.9) × 10 8 copies per ml, a 15.9-fold increase, during the phase of VC to ethene reductive dechlorination (Fig. 1C).CM and VC consumption occurred sequentially, with CM fermentation preceding VC reductive dechlorination (Fig. 1A), presumably because CM fermentation generates an electron donor (i.e.formate) and a carbon source (i.e.acetate) D. etheniformans strain GP requires for VC reductive dechlorination and growth.The rate of VC reductive dechlorination significantly increased following CM consumption, suggesting CM may have an inhibitory effect on VC reductive dechlorination.In support of this hypothesis, growth experiments with axenic D. etheniformans strain GP cultures that received VC, H 2 , and acetate demonstrated that CM at or above the aqueous concentration of ∼0.1 mM (∼10 μmol per vessel) completely inhibited VC to ethene reductive dechlorination (Supplementary Fig. S1).In axenic cultures of strain CM amended with both CM and VC, only CM was degraded, VC was stable, and no ethene formed (Fig. 1D).In axenic cultures of D. etheniformans strain GP amended with CM and VC but without exogenous electron donor (i.e.formate, H 2 ) and carbon source (i.e.acetate), no ethene was detected, no growth occurred, and CM and VC were not consumed (Fig. 1E).Consistent with the results of the growth experiments, microscopic analysis visualized both the characteristic curved rod-shaped strain CM and coccus-shaped Dehalogenimonas cell morphologies in co-cultures during growth with CM and VC (Fig. 1F).
CM consumption in axenic strain CM cultures resulted in acetate and formate accumulation, indicating that this bacterium cannot metabolize formate (Fig. S2).Strain CM ferments CM plus CO 2 to both formate and acetate according to (1).
Based on (1), strain CM would be expected to produce onethird mol of formate per mol of CM consumed, and the complete degradation of ∼200 μmol CM (Fig. 1A) would yield ∼66.7 μmol of formate (1), assuming no formate consumption.D. etheniformans strain GP in the co-cultures utilized formate as the electron donor to fuel the reductive dechlorination of VC to ethene according to (2).
Based on (2), the amount of formate consumed equals the amount of ethene formed, and 50.3 ± 0.3 μmol of ethene were measured in co-cultures that received ∼200 μmol CM.Therefore, a net formation of ∼16.4 μmol of formate would be expected based on (1) and (2).A total of 14.8 ± 2.4 μmol of formate was measured in the co-cultures at the end of the incubation period (Fig. 1B), an amount close to the theoretical value.CM fermentation also yields two-thirds mol of acetate according to (1) (i.e.∼133 μmol), and about 80 μmol were measured in the co-cultures that received 200 μmol of CM, consistent with some acetate being utilized as a carbon source for both strains.
According to (1), the fermentation of ∼282.0 μmol of CM would yield ∼94.0 and 188.0 μmol of formate and acetate, respectively.Based on (3), the reduction of 39.9 ± 1.0 μmol of cDCE to ethene requires about 80.0 μmol of formate as an electron donor.
Therefore, ∼14.0 μmol of formate would be expected to remain in co-cultures amended with 282 μmol of CM and 40 μmol of cDCE based on (1) and ( 3).The measured amount of 15.8 ± 1.8 μmol of formate was close to this theoretical value.CM fermentation would yield 188.0 μmol of acetate, of which 132.2 ± 9.1 μmol was measured at the end of the incubation period, consistent with acetate being utilized as a carbon source by D. etheniformans strain GP and strain CM.The experimental data demonstrate that anaerobic CM degradation by strain CM supports organohalide respiration of cDCE and VC to ethene by D. etheniformans strain GP in synthetic co-cultures without requiring formate or H 2 as an exogenous electron donor, or acetate as a carbon source.
Based on (4) and the amount of DCM consumed, a theoretical amount of ∼170 μmol of acetate would be expected (not considering that some acetate will be consumed as a carbon source).In the experimental systems, a maximum amount of 115.9 ± 8.6 μmol of acetate was measured.Of the theoretical amount of ∼340 μmol of formate, no more than 2 μmol (i.e.∼0.04 mM) were analytically captured over the course of the incubation (Fig. 3B).Dehalobacterium formicoaceticum strain DMC and D. etheniformans strain GP cell numbers increased from (3.4 ± 0.6) × 10 7 to (1.5 ± 0.1) × 10 9 copies per ml, and from (3.6 ± 0.1) × 10 7 to (2.3 ± 0.1) × 10 8 copies per ml, respectively (Fig. 3C), demonstrating that both bacterial populations grew in the co-cultures amended with DCM and VC.VC was not degraded in axenic Dehalobacterium formicoaceticum cultures during growth with DCM (Fig. 3D).Neither DCM nor VC was degraded in axenic D. etheniformans strain GP cultures without the exogenous addition of formate or H 2 as electron donor (Fig. 3E).Microscopic examination of the VC-dechlorinating co-cultures revealed rod-shaped Dehalobacterium formicoaceticum strain DMC cells and coccus-shaped D. etheniformans strain GP cells in the co-cultures during growth with DCM and VC (Fig. 3F).Like CM, DCM inhibits reductive dechlorination, and the presence of ∼0.15 mM DCM (∼10 μmol per bottle) completely prevented VC to ethene reductive dechlorination in axenic D. etheniformans strain GP cultures (Supplementary Fig. S3).
Consortium RM contains hydrogenotrophic acetogens and methanogens, which compete with D. etheniformans strain GP for H 2 as an electron donor in consortium RM-Dehalogenimonas mixtures [37][38][39].To potentially increase the electron f low toward reductive dechlorination, the methanogenesis inhibitor 2-bromoethanesulfonate (BES, 1 mM) was added.BES completely abolished methanogenesis; however, the addition of BES did not promote the reductive dechlorination of VC to ethene.No ethene was formed, and DCM degradation halted after the second dose of DCM (Supplementary Fig. S4), suggesting that 1 mM BES inhibited DCM mineralization and VC reductive dechlorination.

Metabolite cross-feeding in anaerobic dechlorinating microbial communities
Obligate OHRB, including strains of the genera Dehalococcoides and Dehalogenimonas, are metabolic specialists only capable of conserving energy for growth by utilizing organohalogens as terminal electron acceptors [41].OHRB are broadly distributed in the environment and play important roles for carbon cycling in anoxic ecosystems [51].Dehalococcoides and Dehalogenimonas have streamlined genomes with sizes ranging from 1.3 to 2.1 Mb [40,48,52], which explains their limited metabolic versatility and dependency on community members for supplying nutrients and growth factors [11,[53][54][55].Dehalococcoides are restricted to H 2 , while Dehalogenimonas can utilize both H 2 and formate as electron donors [40,42].All characterized organohalide-respiring Dehalococcoidia are heterotrophs and require external acetate as a carbon source [40,42].To date, OHRBs that utilize CM and DCM as electron acceptors have not been found, and other groups of specialized anaerobes utilize these chlorinated C 1 compounds as energy and carbon sources under anoxic conditions [35][36][37].The co-culture experiments demonstrate metabolite cross-feeding between anaerobic CM-or DCM-degrading bacteria and D. etheniformans, enabling concomitant degradation of chlorinated methanes and chlorinated ethenes in anoxic environments (Fig. 5).
Metabolite cross-feeding is a widespread phenomenon in anaerobic mixed cultures, including those with dechlorinating capabilities.For example, benzoate fermentation by a Syntrophus sp. can support complete reductive dechlorination of tetrachloroethene (PCE) to ethene by organohalide-respiring Desulfitobacterium and Dehalococcoides populations via interspecies H 2 cross-feeding [56].Acetylene (C 2 H 2 ) fermentation has been shown to support TCE and VC reductive dechlorination in synthetic co-cultures and in mixed cultures through cross-feeding between acetylene-fermenting acetylenotrophs and OHRB (i.e.Dehalococcoides mccartyi) [57,58].The complete conversion of chlorobenzene and benzene to CH 4 and CO 2 was demonstrated in microcosms bioaugmented with a chlorobenzene-dechlorinating mixed culture and a benzene-degrading methanogenic consortium [59].The fermentation of benzene generated H 2 and acetate, which supported chlorobenzene reductive dechlorination to benzene [59].Dehalococcoides mccartyi strain 195 grew more robustly in syntrophic co-cultures with a lactate-or butyratefermenting bacterium and a hydrogenotrophic methanogen through metabolite cross-feeding and the elimination of toxic by-products [11,54,60].
In the co-cultures, the CM and DCM degraders provided electron donors (i.e.formate and H 2 ) and a carbon source (i.e.acetate) to the cDCE-and VC-dechlorinating Dehalogenimonas population, which in turn consumed the metabolic products (e.g.H 2 ) generated during CM or DCM catabolism.The inhibitory impact of elevated H 2 on DCM degradation has been demonstrated, and the partnership with H 2 -consuming microorganisms, such as hydrogenotrophic methanogens, homoacetogens, or, as demonstrated in this study, OHRB can alleviate the inhibition [39,61].In scenarios like this, a syntrophic partnership benefits both populations through metabolite cross-feeding and represents a mutualistic interaction.
While the strain CM-Dehalogenimonas co-cultures completely converted cDCE and VC to stoichiometric amounts of ethene with CM as the sole source of reducing equivalents, VC reductive dechlorination to ethene in DCM-amended Dehalobacterium-Dehalogenimonas co-cultures and consortium RM-Dehalogenimonas mixtures was slow and incomplete in the batch cutivation experiments.A plausible explanation is formate dehydrogenase (Fdh) activity in Dehalobacterium formicoaceticum cells.Consistent with the genomic [47] and biochemical [36] evidence, formate was transiently produced in axenic Dehalobacterium formicoaceticum cultures grown with DCM, and acetate accumulated as the end product (Supplemenarty Fig. S5).This consumption of formate explains why repeated DCM feedings were required to provide a sufficient amount of electron donor (i.e.formate) to support VC reduction to ethene in the Dehalobacterium-Dehalogenimonas cocultures.Consortium RM harbors hydrogenotrophic methanogens and homoacetogens, indicating that the competition for electron donors (i.e.H 2 ) [37] required continuous DCM feedings to supply sufficient H 2 for VC reductive dechlorination in the RM-Dehalogenimonas mixtures.OHRB consume H 2 at lower thresholds than H 2 -consuming homoacetogens and methanogens, ref lecting the thermodynamics of these terminal electronaccepting processes [62,63].Despite the favorable energetics  2).The methanogenesis inhibitor BES prevented methane formation but did not promote VC reductive dechlorination, presumably because BES can inhibit OHRB [68].BES may also impact the DCM degrader, as the second DCM dose was not consumed in the RM-Dehalogenimonas mixtures amended with BES (Supplementary Fig. S4).Taken together, the slow and incomplete VC reductive dechlorination in DCM-grown co-cultures is largely due to the lack of a sufficient supply of formate or H 2 .

Implications for bioremediation
Anthropogenic activities have resulted in widespread groundwater contamination, and commingled plumes commonly occur [69], which can impact the efficiency of bioremediation [70][71][72].
For example, chloroform (CF) is a known potent inhibitor of Dehalococcoidia activity, leading to stalled reductive dechlorination of chlorinated ethenes [73][74][75].CF can be reductively dechlorinated to DCM by Dehalobacter and Desulfitobacterium species possessing CF reductive dehalogenases, a conversion regarded as a key step to alleviate CF inhibition of reductive dechlorination [76][77][78][79].Unexpectedly, we observed that CM and DCM inhibit VC reductive dechlorination by D. etheniformans strain GP (Supplementary Figs S1 and S3).In addition to electron donor limitation (see above), this inhibition may have also contributed to the slow and incomplete VC-to-ethene reductive dechlorination observed in the DCM-grown co-cultures, and it is possible that D. etheniformans strain GP exhibits different susceptibility to CM versus DCM.Future work should determine the CM and DCM inhibitory constants (K I values) for the reductive dechlorination process.In the co-cultures, the CM and DCM degraders not only provided electron donors (i.e.formate and H 2 ) and a carbon source (i.e.acetate) to the cDCE-and VC-dechlorinating Dehalogenimonas population but also consumed CM or DCM to alleviate their inhibitory effects on the reductive dechlorination process.Anaerobic DCM degradation has been shown to supply reducing equivalents for CF reductive dechlorination in a mixed culture, providing evidence for cross-feeding between an anaerobic DCM degrader and a CF-to-DCM respiring Dehalobacter strain [80].A recent study made a similar observation in a mixed culture but, intriguingly, assigned these activities to a single Dehalobacter population [81].Such inter-and intraspecies electron transfer processes have implications for sustainable bioremediation at sites impacted with chlorinated methanes and other organohalogens that can serve as electron acceptors for OHRB.Our coculture experiments focused on cDCE and VC as electron acceptors, but the findings extend to other organohalogens that OHRB can respire.Thus, this described interdependent process is not limited to environments where chlorinated methanes and chlorinated ethenes co-occur, and CM and DCM catabolism can support reductive dechlorination of other organohalogens.
Conventional enrichment and isolation methodologies have followed the one-contaminant, one-microbe paradigm, and yielded isolates that utilize the pollutant either as an electron donor (e.g.aerobic oxidation of VC) or as an electron acceptor (e.g.organohalide respiration with VC).The bottom-up synthetic ecology experiments using cultures with specific functional properties demonstrated interspecies cooperation and synergisms that resulted in the concomitant degradation of chlorinated C 1 and C 2 compounds.The knowledge-based construction of such assemblies (i.e.synthetic ecology) or their direct enrichment from the environment can yield novel consortia, where interspecies cooperation can substantially improve the efficiency of in situ bioremediation.

Implications for microbial ecology
Both chlorinated methanes and chlorinated ethenes are produced naturally and can coincide in pristine environments [82][83][84], and it is not surprising that microorganisms have evolved strategies to benefit from these compounds.The degradation of CM and DCM under anoxic conditions [35][36][37] resembles the catabolism of non-chlorinated C 1 compounds, such as methanol, methylamines, and methanethiol [85,86].Following demethylation, the methyl or methylene group of the C 1 compounds is channeled into the Wood-Ljungdahl pathway and metabolized via a disproportionation reaction to yield acetate, formate, and H 2 as products.These products are central intermediates and can fuel other microbial processes through metabolite cross-feeding, such as nitrate reduction, organohalide respiration (demonstrated in this study), ferric iron reduction, sulfate reduction, methanogenesis, and reductive acetogenesis (Fig. 6).The concept of CM or DCM as principle energy sources of anaerobic food webs is exemplified in consortium RM, where DCM catabolism supported methane and acetate formation, and the DCM degrader 'Ca.Dichloromethanomonas elyunquensis' methanogens, and homoacetogens co-existed over repeated transfers with DCM as the sole substrate for many years [37][38][39]87].Considering that CM and DCM have natural sources in various pristine environments [18-20, 28, 29, 31, 32, 83, 88], these compounds are overlooked, yet potentially relevant substrates for fueling microbial life in energy-deprived anoxic  environments, such as groundwater aquifers and the deep subsurface.Other examples include permanently ice-covered, oligotrophic ecosystems in Antarctica, such as Lake Vida.DCM and other chlorinated organohalogens were detected in Lake Vida brine, which presumably formed through abiotic reactions between organic matter and oxychlorines (e.g.perchlorate), and are possible substrates to sustain microbial life in this extreme, oligotrophic environment [88,89].
The synthetic co-culture experiments demonstrate that anaerobic CM-and DCM-degraders can partner with obligate OHRB.Most organohalogens are part of the natural chlorine cycle, with concentrations of individual chlorinated compounds representing the balance between production and consumption [15,17,23,33,[82][83][84]90].For example, Atlantic Ocean emissions of CM and DCM reaching 500 and 30 ppt (parts per trillion), respectively, have been reported [91].However, emission data do not inform about f luxes (i.e. the rate of formation minus the rate of consumption), and concentrations of halogenated compounds in pristine environments are compensation concentrations ref lecting balanced production and consumption, which are generally too low to be detected with contemporary analytical approaches [32,92].Thus, even in environments with low steady-state concentrations of chlorinated compounds, their f luxes may be substantial.Information about the f luxes of organohalogens in natural ecosystems is largely lacking, but the distribution of obligate OHRB and CM-/DCM-degrading bacteria clearly extends beyond contaminated sites where human activities increased the concentrations of chlorinated compounds, emphasizing the relevance of the natural chlorine cycle.Microbial ecologists rarely consider organohalogens as sources of energy and drivers of microbial activity, and our findings highlight that CM and DCM catabolism can enable various electron-accepting processes in anoxic ecosystems.Naturally produced chlorinated methanes are overlooked energy sources in anaerobic food webs and may drive carbon and electron f low in energy-deprived, anoxic environments.

Figure 1 .
Figure 1.Performance of synthetic strain CM-Dehalogenimonas co-cultures amended with CM and VC.(A) CM degradation and reductive dechlorination of VC to ethene in defined medium basal salt medium with CM as the sole source of reducing equivalents.(B) CM fermentation yields acetate and formate, with formate being consumed in cultures during VC reductive dechlorination.(C) Increase of strain CM 16S rRNA gene and D. etheniformans strain GP cerA gene copies over time.(D) No VC degradation or ethene formation (not shown) occurred in axenic strain CM cultures during growth with CM. (E) Neither VC nor CM was degraded, and no ethene was formed in axenic cultures of D. etheniformans (Dhgm) strain GP amended with CM and VC.The data represent the average values of triplicate incubations, and the error bars represent the standard deviations.Error bars are not shown when smaller than the symbol.(F) Phase contrast microscopic image of a CM-and VC-degrading, ethene-producing strain CM and D. etheniformans strain GP co-culture.

Figure 2 .
Figure 2. Performance of synthetic strain CM-Dehalogenimonas co-cultures amended with CM and cDCE.(A) CM degradation and reductive dechlorination of cDCE to VC and VC to ethene in vessels with CM as the sole source of reducing equivalents.(B) Formation of acetate and formate, the latter being consumed in the observed reductive dechlorination reactions.(C) Increase of strain CM 16S rRNA gene and D. etheniformans strain GP cerA gene copies in co-cultures amended with CM and cDCE.The data represent the averages of triplicate incubations, and the error bars represent the standard deviations.Error bars are not shown when smaller than the symbol.

Figure 3 .
Figure 3. Performance of Dehalobacterium-Dehalogenimonas co-cultures amended with DCM and VC.(A) DCM degradation and reductive dechlorination of VC to ethene in cultures that received DCM as the sole source of reducing equivalents.(B) Accumulation of acetate and formate concentrations over time.(C) Increase of Dehalobacterium formicoaceticum 16S rRNA gene and D. etheniformans strain GP cerA gene copy numbers in Dehalobacterium-Dehalogenimonas co-cultures amended with DCM and VC.(D) No VC reductive dechlorination occurred in axenic Dehalobacterium formicoaceticum cultures during growth with DCM.(E) VC or DCM were not degraded, and no ethene formation occurred in DCM-and VC-amended axenic D. etheniformans strain GP cultures.The data represent the average values of triplicate incubations, and the error bars represent the standard deviations.Error bars are not shown when smaller than the symbol.(F) Phase contrast microscopic image of the Dehalobacterium-Dehalogenimonas co-culture during growth with DCM and VC.The arrows point at Dehalobacterium formicoaceticum strain DMC and D. etheniformans (Dhgm) strain GP cells.

Figure 4 .
Figure 4. Performance of the DCM-degrading consortium RM inoculated with D. etheniformans strain GP and amended with DCM and VC.(A) DCM consumption and reductive dechlorination of VC to ethene in cultures that received DCM as the sole source of reducing equivalents.(B) Accumulation of methane and acetate over time.(C) Increase of 'Ca.Dichloromethanomonas elyunquensis' 16S rRNA gene and D. etheniformans (Dhgm) cerA gene copies in consortium RM inoculated with D. etheniformans strain GP and amended with DCM and VC.The data represent the average values of triplicate incubations, and the error bars represent the standard deviations.No error bars are shown when smaller than the symbol size.

Figure 5 .
Figure 5. Scheme of metabolite cross-feeding in the strain CM-Dehalogenimonas and Dehalobacterium-Dehalogenimonas co-cultures and consortium RM-Dehalogenimonas mixtures.Formate/H 2 and acetate formed and released during CM and DCM fermentation or mineralization were utilized by D. etheniformans strain GP as electron donors and carbon source, respectively, to carry out organohalide respiration with cDCE and VC to ethene.

Table 1 .
Primers and probes used for qPCR assays specifically targeting 16S rRNA genes of strain CM and Dehalobacterium formicoaceticum strain DMC.

Primer/probe name Target Primer/probe sequence (5 to 3 )
), and forward and reverse primers and probes at final concentrations of 300 nM each.The PCR thermal cycling protocol was as follows: 50 [38]for 2 min, then at 95 • C for 10 min, followed by 40 cycles of denaturation at 95 • C for 15 s, and annealing and extension at 60 • C for 1 min.Calibration curves used serial 10-fold dilutions of the synthetic linear DNA fragment LDFA1 (1500 bp, GeneArt Strings DNA Fragments; Invitrogen, CA, USA) comprising partial gene fragments of the 16S rRNA genes of strain CM, Dehalobacterium formicoaceticum strain DMC, and the VC reductive dehalogenase gene cerA with the respective primer and probe binding sites.For 'Ca.Dichloromethanomonas elyunquensis' strain RM, calibration curves used serial 10-fold dilutions of plasmid DNA carrying a cloned 16S rRNA gene of the DCM-degrading strain RM (pCR 2.1-TOPO vector; Invitrogen, Carlsbad, CA, USA) and spanned a concentration range from 3.44 × 10 8 to 34.4 target gene copies per assay tube[38].