Environmental Science and Pollution Research https://doi.org/10.1007/s11356-023-28144-1 RESEARCH ARTICLE Proteogenomics of the novel Dehalobacterium formicoaceticum strain EZ94 highlights a key role of methyltransferases during anaerobic dichloromethane degradation Kenneth Wasmund1,2 · Alba Trueba‑Santiso3,4 · Teresa Vicent3 · Lorenz Adrian5,6 · Stéphane Vuilleumier7 · Ernest Marco‑Urrea3 Received: 2 November 2022 / Accepted: 2 June 2023 © The Author(s) 2023 Abstract Dichloromethane (DCM, methylene chloride) is a toxic, high-volume industrial pollutant of long-standing. Anaerobic bio- degradation is crucial for its removal from contaminated environments, yet prevailing mechanisms remain unresolved, especially concerning dehalogenation. In this study, we obtained an assembled genome of a novel DCM-degrading strain, Dehalobacterium formicoaceticum strain EZ94, from a stable DCM-degrading consortium, and we analyzed its proteome during degradation of DCM. A gene cluster recently predicted to play a major role in anaerobic DCM catabolism (the mec cassette) was found. Methyltransferases and other proteins encoded by the mec cassette were among the most abundant pro- teins produced, suggesting their involvement in DCM catabolism. Reductive dehalogenases were not detected. Genes and corresponding proteins for a complete Wood-Ljungdahl pathway, which could enable further metabolism of DCM carbon, were also found. Unlike for the anaerobic DCM degrader “Ca. F. warabiya,” no genes for metabolism of the quaternary amines choline and glycine betaine were identified. This work provides independent and supporting evidence that mec- associated methyltransferases are key to anaerobic DCM metabolism. Keywords Dehalobacterium · Anaerobic dichloromethane degradation · Methyltransferases · Wood-Ljungdahl pathway · Shotgun proteomics Introduction Dichloromethane (DCM, methylene chloride) is produced by human activities (predominantly for industrial use as a sol- vent) and also naturally (e.g., volcanoes, oceanic emissions, and biomass burning) (Kolusu et al. 2018). It is classified Responsible Editor: Gerald Thouand as toxic and probable carcinogenic to humans (Schlosser et al. 2015), and it is also a potent ozone-depleting chemical Kenneth Wasmund and Alba Trueba-Santiso contributed equally to which is not covered by the Montreal Protocol (Hossaini this work. 4 * Ernest Marco‑Urrea Current address: Department of Chemical Engineering, ernest.marco@uab.es CRETUS Institute, Universidade de Santiago de Compostela, 15782 Santiago de Compostela, Galicia, Spain 1 Division of Microbial Ecology, Centre for Microbiology 5 Department Environmental Biotechnology, Helmholtz Centre and Environmental Systems Science, University of Vienna, for Environmental Research–UFZ, Leipzig, Germany Vienna, Austria 6 2 Chair for Geobiotechnology, Technische Universität Berlin, School of Biological Sciences, University of Portsmouth, Berlin, Germany Portsmouth, UK 7 3 Université de Strasbourg, CNRS, GMGM UMR 7156, Departament d’Enginyeria Química, Biològica i Ambiental, Génétique Moléculaire, Génomique, Microbiologie, Universitat Autònoma de Barcelona (UAB), Carrer de les Strasbourg, France Sitges s/n, 08193 Cerdanyola del Valles, Spain 13 Vol.:(0123456789) Environmental Science and Pollution Research et al. 2017). DCM ranks eleventh on the European Union play a key role in DCM metabolism (Holland et al. 2019; priority list of pollutants, and 9­ 1st on the 2022 ATSDR Chen et al. 2017). In proteogenomic studies of “Ca. D. ely- Priority List of Hazardous Substances (ATSDR 2022). It unquensis,” many methyltransferases were also found, as is frequently detected as a groundwater pollutant following well as three reductive dehalogenases, two of which were accidental or intentional spills and can migrate through the detected in proteomic analyses of cultures using DCM as unsaturated soil column to accumulate at the bottom of the growth substrate (Kleindienst et al. 2019). Distinct enzy- aquifer as dense non-aqueous phase liquid. The feasibility matic mechanisms for DCM dechlorination in the three dif- of removing DCM under oxic conditions has been widely ferent DCM-degrading bacterial systems were suggested investigated for facultative methylotrophs, which transform by the different dual C-Cl isotope fractionation patterns DCM via a glutathione S-transferase dehalogenase (Mul- observed for “Ca. D. elyunquensis” and Dehalobacterium- ler et al. 2011). Anaerobic microbial degradation of DCM containing cultures (Blazquez-Palli et al. 2019; Chen et al. has also been reported under nitrate-reducing and methano- 2018). genic conditions (Freedman and Gossett 1991; Freedman Recently, a gene cluster annotated as mec (methylene et al. 1997). chloride catabolism) and encoding various conserved meth- A strictly anaerobic bacterial strain capable of fermen- yltransferases and associated proteins was reported specifi- tation of DCM, Dehalobacterium formicoaceticum strain cally among genomes of known anaerobic DCM-degrading DMC, was reported in the 1990s (Mägli et al. 1995, 1996, bacteria, and not among other bacteria (Murdoch et al. 1998), but its genomic characterization was addressed only 2022). Several methyltransferases and associated proteins recently. To date, three anaerobic DCM-degrading bacte- of the mec cluster were among the most highly produced ria, all belonging to the Peptococcaceae family, have now proteins in D. formicoaceticum and “Ca. D. elyunquensis” been genome-sequenced: Dehalobacterium formicoaceticum cultures grown with DCM, and copy numbers of mecE and strain DMC (Chen et al. 2017); “Candidatus Dichlorometha- mecF genes in groundwater correlated with DCM contami- nomonas elyunquensis” from mixed culture RM (Chen et al. nation levels (Murdoch et al. 2022). Together, these data 2018; Kleindienst et al. 2019); and “Candidatus Formi- suggested a key role of the proteins encoded by the mec monas warabiya” (formerly referred to as strain DCMF) cluster in DCM degradation. (Holland et al. 2019, 2021). DCM is the only known carbon To date, D. formicoaceticum is the only reported DCM- and energy source for Dehalobacterium formicoaceticum. In fermenting isolate (Mägli et al. 1996), since “Ca. D. elyun- contrast, strain “Ca. F. warabiya” is also capable of grow- quensis” and “Ca. F. warabiya” are mixed cultures (Hol- ing with non-chlorinated substrates (i.e., methanol, choline, land et al. 2019, 2021; Kleindienst et al. 2019). A novel and N,N,N-trimethylglycine) (Holland et al. 2021). Simi- DCM-degrading Dehalobacterium-containing enrichment larly, the mixed culture RM containing “Ca. D. elyunquen- culture from a membrane bioreactor operating at an indus- sis” can also grow with dichloroacetate (Chen et al. 2021). trial wastewater treatment plant was recently obtained and The prevailing hypothesis is that all three known anaerobic characterized (Trueba-Santiso et al. 2017). Efforts to isolate DCM-degrading bacteria metabolize DCM via the Wood- the Dehalobacterium strain from this culture were unsuc- Ljungdahl pathway encoded by their genomes (Holland et al. cessful, and 16S rRNA gene amplicon sequencing confirmed 2019; Chen et al. 2017; Kleindienst et al. 2019). However, the presence of Dehalobacterium together with three addi- DCM metabolism may proceed through two distinct routes tional phylotypes belonging to Acetobacterium, Desulfovi- of the Wood-Ljungdahl pathway, (i) one allowing minerali- brio, and Wolinella (Trueba-Santiso et al. 2017). Failure zation of DCM to ­CO2 and ­H2, as in “Ca. D. elyunquensis”, to obtain a DCM-degrading isolate was attributed to strict and (ii) the other involving DCM fermentation, to acetate growth requirements of the Dehalobacterium strain, notably and formate in D. formicoaceticum and to acetate only in the need for a hydrogen scavenging partner, and nutrient “Ca. F. warabiya” (Holland et al. 2021; Chen et al. 2020). supply from other bacteria. This dependency also indicates Another major, distinctive, and elusive feature of the the importance of strong synergistic interactions in natural three aforementioned bacteria is the enzyme responsible attenuation of DCM at contaminated sites. Detailed physi- for transformation of DCM. For D. formicoaceticum, early ological studies of the Dehalobacterium-containing culture physiological and biochemical studies with cell-free extracts showed that DCM was transformed by three metabolic mod- as a catalyst showed that DCM reacts with tetrahydrofolate ules: (i) DCM fermentation, (ii) formate transformation to (H4F) to yield 5,10-methylene-H4F, which is then channeled ­CO2 and ­H2, and (iii) H ­ 2/CO2-fueled acetogenesis. The into the Wood-Ljungdahl pathway (Mägli et al. 1996). The Dehalobacterium strain in the culture studied in the present presence of numerous predicted methyltransferase genes in work was shown to only participate in DCM fermentation to the genomes of D. formicoaceticum and “Ca. F. warabiya,” acetate and formate (Trueba-Santiso et al. 2020). together with the lack of reductive dehalogenases in these In this study, we demonstrated the stability of this strains, led to the hypothesis that methyltransferases could Dehalobacterium-containing DCM-degrading mixed culture 13 Environmental Science and Pollution Research and obtained a draft genome assembly of its Dehalobacte- (ThermoFischer). DNA concentrations were in the range of rium strain by a hybrid Illumina/PacBio sequencing strategy, 27.4–48.8 ng·μL−1. representing the second genome of this genus and species to date. We used this genome information to enable shotgun Microbial community profiling proteomic analysis of the DCM-degrading mixed culture and obtained new evidence for the Dehalobacterium proteins The taxonomy of the microbial populations in the com- involved in growth with DCM. munity of each sample was determined by sequencing the V4-V5 region of 16S rRNA genes. Briefly, PCR amplifica- tion of the 16S rRNA gene was carried out using 5 ng of Materials and methods genomic DNA as template using proprietary 192 barcoded primers (Metabiote® MiSeq Primers, GenoScreen, Lille, Cultivation of the Dehalobacterium‑containing France) at final concentrations of 0.2 μM and an annealing culture temperature of 50°C for 30 cycles. PCR products were puri- fied with Agencourt AMPure XP-PCR Purification system The Dehalobacterium-containing mixed culture was main- (Beckman Coulter, Brea, USA), quantified according to the tained for more than two years by routine transfers in 100 manufacturer’s protocol, and mixed to equal concentra- mL glass serum bottles with 65 mL of anoxic synthetic tions. Sequencing was performed using a 250-bp paired- medium modified from Trueba-Santiso et al. 2017. Briefly, end sequencing protocol on the Illumina MiSeq platform this medium contained vitamins, trace elements, 22.8 μM (Illumina, San Diego, USA) at GenoScreen, Lille, France. tungsten, 24.2 μM selenium, and 200 mg/L of yeast extract Demultiplexing was done with CASAVA (Illumina). Num- (no acetate, pyruvate, fumarate, or formate). As a reduc- bers of reads ranged from 28,488 to 32,398 between sam- ­ a2S·9H2O and L-cysteine (0.2 mM each) were ing agent, N ples. Raw paired-end reads were quality-filtered (Schmieder included. The medium was buffered with sodium bicarbo- and Edwards 2011), read merging was done with FLASH nate (pH=7) and resazurin (10 mg/L) was added as a redox (Magoč and Salzberg 2011), and forward and reverse primer indicator. Medium was bottled inside an anoxic tent (Coy sequences were removed using CutAdapt, with no mis- Laboratories, USA) in an atmosphere composed of ­N2 and matches allowed in primer sequences. After pretreatment, 1–3% ­H2. Serum bottles were sealed with Teflon-coated chimeras were removed with a proprietary protocol based butyl rubber septa and aluminum crimp caps. Bottles were on Usearch 6.1. Sequences were grouped at 97% identity autoclaved at 121°C for 40 min and then gassed with ­N2 into operational units (OTUs) with Uclust v1.2.22q (Edgar to 0.4 bar overpressure. Neat DCM was added through the 2010). Taxonomic assignments of OTUs were obtained with septa with a μL-glass syringe (Vici) to achieve the desired the naive Bayesian classifier (v2.2) method (Wang et al. concentration in the serum bottles. Microcosms were stati- 2007) and the Greengenes database (version 13_8; www.​ cally incubated in the dark at 25°C in a thermostatized cham- green​genes.​gov) (Wang et al. 2007). ber and monitored periodically for chlorinated methanes via headspace gas sampling. Active cultures were transferred Sequencing, metagenome assembly, into fresh medium (3–7% v/v inoculum) during exponential and annotations DCM-degradation phase through the septum, maintaining anoxic sterile conditions. Cultures used in this study corre- Library preparations from DNA of the DCM-degrading con- sponded to the 5­ 0th generation of this culture line and were sortium and sequencing using Illumina and PacBio tech- sacrificed after consumption of 3.5 mM DCM. nologies were performed by Genoscreen (Lille, France). The Illumina library was prepared with the Nextera XT DNA extraction sample prep kit (Illumina) and sequenced in a paired-end 2x100 bp HiSeq 2500 rapid run. Illumina sequence reads for Six parallel cultures grown under the same conditions as genome assembly were trimmed of adapters using Adapter- above were used for genomic analyses. Cells from each cul- Removal and default settings (Schubert et al. 2016). Reads ture were pelleted by centrifugation at 7000 g and 10°C and were further trimmed for quality using a custom python3 resuspended in 1 mL of sterile PBS buffer. Bulk DNA was script (https://​github.​com/​kwasm​und/​Trim-​Illum​ina) and then extracted separately from each sample using the Gentra the command line Illumina_trim.py-f Dhb2.pair1.trun- Puregene Yeast/Bacterial DNA extraction kit (Qiagen) fol- cated.fastq-r Dhb2.pair2.truncated.fastq-o reads.filtered-q lowing the manufacturer’s instructions. DNA samples were 15-m 50-b 8. For PacBio, DNA fragmentation and ligation resuspended in 100 μL sterile Tris-HCl buffer (10 mM, pH of single-molecule real-time (SMRT) adaptors was followed 8) at 4°C under static conditions to minimize DNA frag- by BluePippin size selection (Sage Science) set at 4 kb, and mentation. DNA was quantified with a Qubit fluorimeter PacBio sequencing was performed on one SMRT cell using 13 Environmental Science and Pollution Research PacBio RSII technology with P6 chemistry. Quality control centrifuged at 7000 g at 10°C for 20 min. The supernatant was done using FastQC v0.11.5 (Andrews 2010). was discarded, and pelleted cells were washed with PBS and An initial hybrid assembly was obtained using Unicycler resuspended in 1 mL PBS. Cells were lysed by the freeze- (v 0.4.8) (Wick et al. 2017) with default settings using the thaw method (−20°C/+20°C, 6x) and then sonicated for 30 quality-controlled Illumina paired-end sequences and PacBio s. Samples were then concentrated (10 kDa cut-off Amicon sequences. The resulting contigs were binned using Meta- Ultra–0.5 mL Centrifugal Filters, Millipore) by centrifuga- BAT 2 (v 2.15) (Kang et al. 2019) with default settings for tion at 14000 g for 30 min at 10°C. Filters were inverted, and automatic binning. The Dehalobacterium genome bin was the protein extract was spun-off for 2 min at 1000 g, 10°C. identified by BLASTX of contigs against the NCBI-nr data- Protein concentrations in these samples were estimated base. Quality-controlled PacBio sequences from the original using the Bradford method. The three samples were then unbinned data were then mapped to the draft Dehalobacte- further processed in parallel. rium bin using minimap2 (v 2.17) (Li 2021) with default set- tings, and mapped sequences were collected and converted Nano‑liquid chromatography tandem mass to fastq files using SAMtools (v 1.12) (Li et al. 2009). Col- spectrometry (nLC‑MS/MS) lected PacBio sequences were then reassembled with all quality-controlled Illumina sequences using Unicycler and Protein samples were treated as described previously to automatically binned with MetaBAT 2 as described above. reduce disulfide bridges and alkylate cysteine residues Additional mapping of Illumina sequences to the initial draft (Seidel et al. 2018). Protein digestion was performed by Dehalobacterium bin, or by further iterations of PacBIO addition of 0.1 μg of porcine trypsin (Proteomics Sequenc- sequence mapping to the draft Dehalobacterium bins and/or ing Grade, Promega) and overnight incubation at 37°C on reassembly, did not result in further assembly improvement. a shaker. The obtained peptides were dried, resolubilized in The resulting Dehalobacterium genome bin was automati- 0.1% formic acid, and desalted using Zip Tip-μC18 material cally annotated using the RAST server (Aziz et al. 2008) in (Merck Millipore). Peptides were analyzed by nano-liquid “classic” mode. Manual BLASTP analysis of selected pro- chromatography tandem mass spectrometry (nLC-MS/MS) tein sequences derived from the 19 obtained contigs against on a nanoUPLC system (nanoAcquity, Waters) hyphenated the NCBI-nr database gave best hits to D. formicoaceti- via a TriVersa NanoMate (Advion, Ltd., Harlow, UK) to an cum strain DMC for all contigs, suggesting the absence of Orbitrap Fusion mass spectrometer (Thermo Scientific) as contaminating contigs. Genome bin quality was assessed described previously (Seidel et al. 2018). Glyceraldehyde- using CheckM (v1.1.1) (Parks et al. 2015). Additional auto- 3-phosphate dehydrogenase (GapDH) from Staphylococcus matic gene and protein annotations were performed on the aureus was added as internal standard before sample pro- MAGE MicroScope platform (Vallenet et al. 2020). Aver- cessing. Peptide identification was conducted by Proteome age Nucleotide Identity (ANI) analyses were conducted with Discoverer (v2.2, Thermo Fisher Scientific) using Seques- the JSpeciesWS server in BLASTN mode (“ANIb”) (Richter tHT as a search engine and a fasta file of proteins from the et al. 2016). “Methyltransferase” proteins were subject to draft genome of the Dehalobacterium strain EZ94 as the eggNOG-mapper (v2.0.0) (Cantalapiedra et al. 2021) using database. A false discovery rate threshold of 1%, defined default settings (minimum hit e-value 0.001, minimum hit versus a decoy database (Target Decoy PSM evaluator bit-score 60, minimum % of identity 40, minimum % of implemented in Proteome Discoverer) was set for peptide query coverage 20) with “-m diamond” option. They were identification. Label-free quantification of detected proteins also analyzed for conserved domains using the Conserved was performed with Minora node implemented in Proteome Domain search tool (Marchler-Bauer and Bryant 2004) Discoverer on the basis of area counts of corresponding pep- against the Conserved Domain Database (CDD) (Lu et al. tide peaks in precursor scans. 2020) using default settings and a default e-value of 0.01. A phylogenomic tree was made on the IQ-TREE web-server using automatic substitution model selection (best-fit model Results LG+F+I+G4) and ultrafast bootstrapping (1000×) (Nguyen et al. 2015). The underlying alignment consisted of concat- Sequencing of a stable DCM‑degrading consortium enated protein sequences derived from single copy marker and analysis of the dominant Dehalobacterium genes retrieved from CheckM. strain Protein extraction In a previous study, we showed that our DCM-degrading anaerobic mixed culture was composed mainly of bacte- Triplicate samples consisting of two pooled microcosms ria belonging to the genera Dehalobacterium, Desulfo- each were used for proteome profiling. Cultures were vibrio, Acetobacterium, and Wolinella (Trueba-Santiso 13 Environmental Science and Pollution Research et al. 2020). The 16S rRNA gene amplicon data con- Genomic analysis of Dehalobacterium firmed that the DCM-degrading consortium was domi- formicoaceticum strain EZ94 nated by a Dehalobacterium population (87.1%±1.9 relative abundance), with lower relative abundances of The genome of strain EZ94 harbors a gene cluster corre- the genera Acetobacterium (7.6%±1.2), Desulfovibrio sponding to the conserved methylene chloride catabolism (4.7%±1.0), and Wolinella (0.6%±0.1) (Table S1). The (mec) gene cassette encoding a multiple methyltransferase very low variability observed between biological rep- system recently shown to be associated with anaerobic licates shows the stability of the consortium under the DCM degradation (Murdoch et al. 2022). This gene clus- selective conditions used. ter (DHBM_v1_160003 to DHBM_v1_160013; in the fol- Metagenomic sequencing on DNA of the consortium lowing, the locus tag “DHBM_v1_” is omitted for clarity) culture performed using Illumina and PacBio tech- encodes 10 proteins MecA to MecJ with high sequence niques, followed by genome binning of the obtained amino acid identity (average 91.3%) to corresponding pro- contigs, allowed recovery of a draft genome of the teins from other DCM degraders (Fig. 2 and Table 1). An dominant Dehalobacterium population. It consisted open reading frame for one additional small (37 amino acid) of 3,650,560 bp in 19 contigs with a G+C content of hypothetical protein (160012) between genes for MecI and 43.2%. This is closely similar to the reported genome MecJ was also defined, unlike in other anaerobic DCM- of D. formicoaceticum strain DMC of 3,766,545 bp and degrading bacteria except for D. formicoaceticum strain has identical G+C content (Chen et al. 2017, NCBI DMC (Table 1). Notably, no reductive dehalogenase genes accession number GCA_002224645.1). The draft were detected in the genome. genome of our Dehalobacterium strain displayed an In this context, we also searched the automatically ANIb value of 98.6% to the D. formicoaceticum strain annotated genome for proteins with “methyltransferase” DMC genome, with 93% alignment. The genome was annotations and classified them further according to con- estimated to be 95.7% complete from CheckM analysis, served domains from CDD searches and by mapping to the with 2% contamination, as reported for the complete eggNOG database (eggNOG-mapper). Apart from Mec circular genome of D. formicoaceticum strain DMC. proteins, these identified 124 genes associated with C1 This suggests that CheckM missed and/or misassigned transfer, including 25 diversely annotated as mono-, di-, or marker genes from these organisms and that the genome trimethylamine:corrinoid methyltransferases (Table S2). of the Dehalobacterium strain from our DCM-degrad- The genome of strain EZ94 also encodes a complete ing consortium is near-complete. Close similarity of set of genes for the Wood-Ljungdahl pathway (Fig. 3). our strain to strain DMC was also supported by phy- This is in accordance with previous reports of anaerobic logenomic analysis of 34 single copy marker proteins DCM-degrading bacteria and indicates strain EZ94 has the (Fig. 1). Further, and as for strain DMC, six 16S rRNA capacity to metabolize DCM through this pathway. In par- gene copies assigned to Dehalobacterium were identi- ticular, all proteins in the Wood-Ljungdahl pathway were fied in an initial draft genome, which, however, were almost identical (92.83 to 100% amino acid identity) with not recovered in our improved final hybrid genome those in D. formicoaceticum strain DMC. In this context, assembly. On the basis of this high similarity of our biosynthesis pathways for essential cofactors folate, heme, strain to D. formicoaceticum DMC and of previous rec- riboflavin, cobalamin, pantothenate, and coenzyme A are ommendations that strains with ANI similarity >95% fully or almost fully encoded in the genome of strain EZ94 should be assigned to the same species (Goris et al. (Table S3). The biosynthesis pathway for biotin appears to 2007), we defined the strain in our DCM-degrading be absent in all anaerobic DCM degraders. As an intriguing consortium as D. formicoaceticum strain EZ94, a new difference among anaerobic DCM degraders, the menaqui- strain of the species D. formicoaceticum. none biosynthesis pathway is absent in strain EZ94, as in D. formicoaceticum strain DMC and Ca. F. warabiya, but Fig. 1  Phylogenomic tree based on proteins from 34 single copy marker gene genes derived from CheckM analysis (see the “Materials and methods” section). Circles on nodes indicate bootstrap values >90%. The scale bar represents 10% sequence divergence 13 Environmental Science and Pollution Research Fig. 2  Gene synteny for the mec gene cluster among DCM-degrading bacteria. Blue shading represents tBLASTx identity values between cor- responding proteins almost fully encoded in “Ca. D. elyunquensis” (Table S3). detected either. Proteomic analysis further revealed However, all DCM-degrading strains encode two genes that MecB, annotated as corrinoid methyltransferase that may catalyze the last steps of menaquinone biosyn- (160004), was the most abundant protein in the three thesis (1,4-dihydroxy-2-naphthoate octaprenyltransferase cultures analyzed (Table S4). The third most abundant (10507) and demethylmenaquinone methyltransferase protein was MecC, annotated as methylcobalamin meth- (10199), Table S3). Additionally, genes for twelve subunits yltransferase (160005) (Table S4). These two protein of the proton-pumping NADH:ubiquinone oxidoreduc- products of the mec gene cluster display protein iden- tase complex were detected in the genome of strain EZ94 tities of 88–95% and 92–96% with MecB and MecC (NuoABCDFHIJMNLK). from DCM-degrading D. formicoaceticum DMC, “Ca. D. elyunquensis” and “Ca. F. warabiya”, respectively Proteome profile of strain EZ94 (Table 1). Worthy of note, other related proteins in in the DCM‑degrading consortium databases show only low levels of identity (<54.9% for MecB and <40.4% for MecC), suggesting these methyl- Shotgun proteomic analyses were performed for three rep- transferases are specific to DCM-degrading bacteria of licate cultures of the DCM-degrading consortium grown on the Peptococcaceae. DCM for 50 consecutive transfers and resulted in the identi- The second most abundantly produced protein (50017) fication of 855 distinct proteins of D. formicoaceticum strain is a 131 amino acid long protein unlike any well-charac- EZ94 (Table S4), accounting for 23% of proteins defined terized protein in databases, automatically annotated as from the genome sequence. pyridoxamine-5′-phosphate oxidase family protein with All proteins of the mec cassette were detected with a potential flavin-mononucleotide binding domain. It was the exception of MecA, MecD and MecJ (Table 1). also the second most abundant protein found in the prote- These three proteins are predicted to be regulatory pro- ogenomic analysis of strain DMC (Murdoch et al. 2022). teins and/or to be membrane-associated. This may help Nevertheless, it is not found in the other two DCM degraders explain why they were not detected experimentally. The “Ca. D. elyunquensis” or “Ca. F. warabiya,” suggesting that small predicted ORF between MecI and MecJ was not it may not be involved or essential for DCM degradation. 13 Environmental Science and Pollution Research Table 1  Percentage identity of proteins encoded by the mec gene The MicroScope (MicS) gene locus tag identifiers were provided cassette of strain EZ94 to their homologs in other anaerobic DCM- omitting “DHBM_v1_,” “DEFO_v1_,” “LINDB_v1_,” “DCMF_,” degrading organisms. Dehalobacter sp. CF and Dehalobacter sp. “DCF50_,” and “AUUR_v1_” in Dehalobacterium spp. strains EZ94, UNSWDHB were also included for reference. Averaged protein abun- DCM, “Ca. D. elyuquensis,” and “Ca. F. warabiya,” and Dehalobac- dance and ranking are derived from Table S4. “n.d.,” not detected. ter strains CF and UNSWDHB for clarity, respectively Transcriponal MecA 160003 Autolysin sensor kinase n.d. 500003 99% 180006 82% 6030 87% p1241 84% 470027 85% regulator Dimethylamine Corrinoid MecB 160004 methyltransferase 1 500004 95% 310002 88% 6025 94% p1240 94% 470026 92% methyltransferase corrinoid protein Methylcobalamin Methylcobalamin MecC 160005 methyltransferase 3 500005 96% 310001 92% 6020 90% p1239 94% 470025 94% methyltransferase MMP0830 Stage 0 sporulaon Transcriponal regulator, 180005 88% MecD 160006 n.d. 500006 89% 6015 93% n.d. n.d. 470024 89% protein A homolog AraC family 100001 88% Uroporphyrinogen Methylcobalamin MecE 160007 decarboxylase domain- 92 500007 98% 100002 94% 6010 89% n.d. n.d. 470023 97% methyltransferase containing protein N5- Tetrahydromethanopter methyltetrahydromethan MecF 160008 in S-methyltransferase opterin:coenzyme M 89 500008 99% 100003 91% 6005 94% n.d. n.d. 470022 98% subunit H methyltransferase subunit H Conserved protein of MecG 160009 Hypothecal protein 748 500009 99% 100004 75% - - n.d. n.d. 470021 100% unknown funcon Methyltransferase Methyltransferase MecH 160010 corrinoid acvaon corrinoid acvaon 726 500010 85% 100005 79% - - n.d. n.d. 470020 92% protein protein Methylcobalamin Uroporphyrinogen MecI 160011 methyltransferase 344 500011 99% 100006 79% 6000 82% n.d. n.d. 470019 96% decarboxylase (URO-D) MMP0834 Protein of unknown Unknown 160012 Hypothecal protein n.d. 500012 100% - - - - n.d. n.d. n.d. n.d. funcon Na H Exchanger domain- MecJ 160013 Na+/H+ anporter n.d. 500013 53% 10112 51% 5995 83% n.d. n.d. 470018 99% containing protein D. formicoaceticum strain DMC is acetogenic and also identified (Table S4). For tetrahydrofolate, six of the suggested to metabolize the methyl group of DCM via seven proteins involved in the transformation of guano- the Wood-Ljungdahl pathway (Mägli et al. 1996). The sine 5′-triphosphate into 5,6,7,8-tetrahydrofolate were Wood-Ljungdahl pathway can be used in either reductive produced in at least one replicate, indicating biosynthesis (i.e., ­CO 2 fixation) or oxidative directions (i.e., acetate of this cofactor. oxidation). In strain EZ94, most of the proteins associated With regard to heme biosynthesis, four proteins were with the carbonyl and methyl branch of the Wood-Ljun- detected in at least one replicate for the reactions from L-glu- gdahl pathway were produced during growth with DCM tamyl-tRNA to coproporphyrinogen III. The genes for con- (Fig. 3). Worthy of note, however, the protein product of a version of coproporphyrinogen III to heme via the classical formate dehydrogenase subunit alpha (fdhA) gene (10437) pathway seemed to be encoded, but proteins were not detected. was not detected. Similarly, and although several genes for the cytochrome c Several Mec proteins, as well as enzymes of the Wood- assembly pathway were detected in the genome, none of Ljungdahl pathway, are associated with corrinoids or the corresponding proteins (subunits CcmA, CcmB, CcmC, tetrahydrofolate. Among the full set of genes encoding CcmE, and CcmF) were detected by LC-MS/MS. Con- for corrinoid and tetrahydrofolate biosynthesis in strain versely and despite the fact that the biosynthesis pathway for EZ94, 8 proteins associated with the upper corrinoid menaquinone is lacking in strain EZ94, three subunits of the biosynthesis pathway starting from L-glutamate to ado- proton-pumping NADH:ubiquinone oxidoreductase com- cobinamide and 3 proteins associated with nucleotide plex (NuoC, NuoD, and NuoI) were detected in two replicate loop assembly of corrinoids were identified in the genome cultures. However, genome-encoded polyprenyltransferase (Table S3). A predicted cobalt transporter (10141, CbiQ) (MenA, 10507) and methyltransferase (MenG, 10199) were and a corrinoid salvaging enzyme (20161, CbiZ) were not detected in proteomic analyses. 13 Environmental Science and Pollution Research Fig. 3  Proposed pathway for 3 CH2Cl2 metabolism of DCM by D. formicoaceticum strain EZ94. Mec proteins? Gene locus identifier numbers for corresponding enzymes are underlined, with mean protein 2 CH2=H4F 3 CH2=H4F CH2=H4F abundance rank (“#” in orange 50132 5,10-methylene-H4F 150025 boxes; “n.d.,” not detected, see #12 #n.d. Table S4). folD, methylene-H4F dehydrogenase/cyclohydrolase; folD metF fhs, formate-H4F ligase; metF, 2 CH =H4F CH3-H4F methylene-H4F reductase; acsE, 5,10-methenyl-H4F Methyl-H4F 60146 60147 60150 methyltransferase; CoFeSP, 50132 methyl-Co(I) corrinoid iron #13 #15 #19 sulfur protein; cooS, CO #12 acsE CoFeSP acsC dehydrogenase; acsB, acetyl- 2 CHO-H4F CH3 - [Co] -E CO2 CoA synthase; pta, phosphate 60152 10519 10386 10-formyl-H4F Methyl Co (I) corrinoid acetyltransferase; ackA, acetate #168 #446 #n.d. 10088 iron-sulfur protein kinase cooS cooS #n.d. [CO] s 60151 2 HCOO- Formate #49 acsB CH3CO-SCoA Acetyl-CoA 20205 20399 #n.d. #56 CH3Co-PO4 Acetyl-phosphate 60088 #75 ackA CH3COO- Acetate Discussion Early experiments with cell extracts of D. formicoaceticum DMC had already shown a requirement for methyl viologen Metagenomic analysis of our DCM-degrading consortium for dehalogenation activity and light-reversible propyl iodide confirmed that the dominant strain EZ94 belongs to the spe- inhibition of DCM degradation, suggesting the involvement of cies D. formicoaceticum, that it also encodes the mec gene corrinoids (Mägli et al. 1996). Worthy of note, the genome of cassette recently reported to be associated with metabolism strain EZ94 also harbors a full set of genes for a complete cor- of DCM by D. formicoaceticum DMC, and that the Mec pro- rinoid biosynthetic pathway, with several of the corresponding teins are highly produced during growth with DCM (Table proteins detected by mass spectrometry (Table S4). S4). Comparative genomics suggest that no close homologs Nevertheless and despite the high sequence similarity found of the mec gene cassette are found in publicly available bacte- for Mec proteins between Dehalobacterium strains, “Ca. F. rial genomes aside from D. formicoaceticum strains, “Ca. D. warabiya” and “Ca. D. elyunquensis” (Table 1), dual C and elyunquensis” and “Ca. F. warabiya”, and among Dehalobac- Cl isotope fractionation analysis previously indicated that dif- ter strains that produce DCM during organohalide respiration ferent anaerobic DCM-degrading strains from different genera with chloroform (Table 1, Deshpande et al. 2013, Tang et al. catalyze mechanistically distinct C-Cl bond cleavage reactions 2016). High amounts of Mec proteins were also detected in (Chen et al. 2018; Blazquez-Palli et al. 2019). Interestingly D. formicoaceticum and “Ca. D. elyunquensis” growing with and in contrast to the two Dehalobacterium formicoaceticum DCM (Murdoch et al. 2022). Specifically, corrinoid-associ- strains, “Ca. D. elyunquensis” also encodes reductive dehalo- ated methyltransferases MecB, MecC, MecE, and MecF were genases, two of which were produced during growth with DCM the most abundant proteins among the mec gene cluster in (Kleindienst et al. 2019). This suggests that “Ca. D. elyunquen- all three organisms, with MecB the most highly produced. sis” may be unique among known anaerobic DCM-degraders 13 Environmental Science and Pollution Research in using reductive dehalogenases for DCM dehalogenation, several transfers when amended with formate instead of DCM but this has not yet been demonstrated experimentally. Rather, as energy source (Trueba-Santiso et al. 2020). Interestingly, the Murdoch et al. (2022) proposed that for strain DMC as well as almost identical (99.8% similarity) homolog of D. formicoace- for “Ca. D. elyunquensis,” C1 carbon metabolism from DCM ticum strain DMC (accession WP_198306660) was detected first involves methyltransferase MecE that would eliminate one in a previous proteomic study (abundance rank #287/2014, chlorine substituent from DCM with subsequent chloromethyl Murdoch et al. 2022), and “Ca. F. warabiya” also features a transfer to MecB, yielding a chloromethyl corrinoid protein homolog of this gene (55% identity at the protein level). Nev- that would be further transformed to methylene-H4F by MecF. ertheless, D. formicoaceticum strain DMC and Ca. F. warabiya However, this leaves the potential role of other corrinoid-asso- cannot grow with formate either. Therefore, the metabolic func- ciated proteins MecC and MecI encoded by the mec cassette tion of this formate dehydrogenase remains unknown. that are also produced during growth with DCM unexplained. Dehalobacterium formicoaceticum strains are indeed unique In particular, MecE is produced at similar levels as MecB and among anaerobic DCM-dechlorinating bacteria in that DCM about 25x higher than MecC. Protein separation and/or het- fermentation is their only growth-supporting metabolism known erologous expression coupled with activity assessment will so far. Other bacteria with the mec cassette associated with be needed to identify which of the specific Mec proteins are DCM utilisation appear to be metabolically more versatile. For involved in the different steps of DCM dechlorination. instance, the mixed culture containing “Ca. D. elyunquensis” Pathways for biosynthesis of cobalamin, menaquinone, and also ferments dichloroacetate to acetate, carbon dioxide, and heme differ among the anaerobic DCM-degrading strains. Spe- hydrogen (Chen et al. 2021). In this case, dichloroacetate dehalo- cifically, cobalamin biosynthesis appears incomplete in “Ca. genation involves an (S)-2-haloacid dehalogenase not found in D. elyunquensis,” whereas almost all genes for both upper and D. formicoaceticum. Interestingly, re-examination of published lower pathways for de novo biosynthesis of cobalamin are found proteomic data (Chen et al. 2021) showed that MecB, MecC, in D. formicoaceticum and “Ca. F. warabiya” strains, with 11 of MecE, MecF, MecH, and MecI proteins of “Ca. D. elyunquen- the corresponding proteins of strain EZ94 detected by mass spec- sis” were 56–71 times more abundant upon growth with DCM trometry in at least one replicate culture (Table S3). Conversely, compared to growth with dichloroacetate, underlining the spe- the biosynthesis pathway for the electron carrier menaquinone is cific association of mec cassette proteins with DCM metabolism. found in “Ca. D. elyunquensis,” with two proteins detected dur- Similarly, “Ca. F. warabiya” also grows with non-chlorinated ing growth with DCM (Kleindienst et al. 2019), but it was not substrates methanol, choline, and glycine betaine (Holland et al. detected in the other anaerobic DCM-degrading strains (Table 2021). Strain EZ94 contains several genes annotated as glycine/ S3). Also, heme biosynthesis in Dehalobacterium and “Ca. sarcosine/betaine reductase complexes (10155-10158, 20275- F. warabiya” strains likely proceeds via the classical pathway, 20277, 10725, and 10726), although no choline dehydrogenase whereas “Ca. D. elyunquensis” appears to use the alternative or betaine aldehyde dehydrogenase associated with transforma- route to heme as described for Archaea and sulfate-reducing tion of choline to betaine, or for demethylation of betaine or bacteria, in which heme is synthesized from uroporphyrinogen dimethylglycine, were identified. Similarly, a large number of III via siroheme (Kühner et al. 2014). genes diversely annotated as mono-, di-, or trimethylamine or With respect to carbon processing downstream of DCM methanol corrinoid methyltransferases (Table S2) were detected dehalogenation, the genome of strain EZ94 encodes a complete in the genome of strain EZ94. It is possible that alternative com- Wood-Ljungdahl pathway, with most corresponding proteins pounds or mixtures of compounds lacking C-C bonds such as experimentally detected, indicating that metabolism of DCM those mentioned above may be found to support growth of strain proceeds via this pathway after methylene-H4F formation EZ94 in the future. (Fig. 3). Our genomic and proteomic data thus suggest that strain EZ94, as likely also strain DMC, is mixotrophic; i.e., it depends on both inorganic (­ CO2) and organic (DCM) carbon Conclusions to grow. A requirement for ­CO2 in EZ94 is further supported by physiological studies showing that omission of ­CO2/HCO3 Dehalobacterium formicoaceticum was the dominant organ- from the medium stopped DCM consumption (data not shown). ism in our DCM-degrading consortium after growth for 2 years Incidentally, this was also observed for “Ca. D. elyunquensis” with DCM. The proteogenomic characterization of strain EZ94 (Kleindienst et al. 2019), indicating that although the dechlo- reported here provides a second genome sequence for a member rinating enzymatic system of the two strains appears distinct, of the genus Dehalobacterium and for the species Dehalobac- they both metabolize tetrahydrofolate-associated DCM carbon terium formicoaceticum, after that of strain DMC (Chen et al. by the Wood-Ljungdahl pathway. 2017). Both strains grow exclusively by DCM fermentation as Although the genome encoded a formate dehydrogenase the sole energy source. As other anaerobic DCM degraders, subunit alpha (10437), the protein was not detected in the the genome of strain EZ94 contains a mec gene cluster, con- proteome. Indeed, our Dehalobacterium strain died-out after firming its potential as a DNA-level biomarker for evaluating 13 Environmental Science and Pollution Research intrinsic biodegradation potential of DCM at contaminated Open Access This article is licensed under a Creative Commons Attri- sites. Comprehensive proteome profiling of strain EZ94 in the bution 4.0 International License, which permits use, sharing, adapta- tion, distribution and reproduction in any medium or format, as long consortium culture showed that proteins of the mec gene cluster as you give appropriate credit to the original author(s) and the source, were highly produced, in agreement with recent studies with D. provide a link to the Creative Commons licence, and indicate if changes formicoaceticum DMC and “Ca. D. elyunquensis.” Our find- were made. The images or other third party material in this article are ings thus independently confirm and extend growing insights included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in into DCM fermentation and may contribute to decipher the still the article's Creative Commons licence and your intended use is not elusive mechanisms of anaerobic DCM degradation. permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a Supplementary Information The online version contains supplemen- copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. tary material available at https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 11356-0​ 23-2​ 8144-1. Author contribution All authors contributed to the study conception and design. Experimental work was performed by Alba Trueba-Santiso. Data collection and analysis were performed by Kenneth Wasmund References and Ernest Marco-Urrea. The first draft of the manuscript was written by Kenneth Wasmund and Ernest Marco-Urrea, and all authors com- Andrews S (2010) FastQC: a quality control tool for high throughput mented on previous versions of the manuscript. All authors read and sequence data. http://​www.​bioin​forma​tics.​babra​ham.​ac.​uk/​proje​ approved the final manuscript. cts/​fastqc. Accessed 19/9/2022 ATSDR (2022) Agency for toxic substances and disease registry. Funding Open Access Funding provided by Universitat Autonoma de https://​www.​atsdr.​cdc.​gov/​spl/​index.​html. Accessed 8 Jun 2023 Barcelona. This work was supported by the Spanish Ministry of Econ- Aziz RK, Bartels D, Best AA, DeJongh M, Disz T, Edwards RA, omy and Competitiveness State Research Agency (CTM2016-75587- Formsma K et al (2008) The RAST server: rapid annotations C2-1-R and PID2019-103989RB-100) cofinanced by the European using subsystems technology. BMC Genomics 9:75. https://​doi.​ Union through the European Regional Development Fund (ERDF). org/​10.​1186/​1471-​2164-9-​75 It was also partly supported by the Generalitat de Catalunya (Con- Blazquez-Palli N, Shouakar-Stash O, Palau J, Trueba-Santiso A, Varias solidate Research Group 2017-SGR-14). A.T.-S. acknowledges a pre- J, Bosch M, Soler A et al (2019) Use of dual element isotope doctoral fellowship from MINECO (BES-2014-070817). Protein mass analysis and microcosm studies to determine the origin and poten- spectrometry was performed at the Centre for Chemical Microscopy tial anaerobic biodegradation of dichloromethane in two multi- (ProVIS) at the Helmholtz Centre for Environmental Research–UFZ. contaminated aquifers. Sci Total Environ 696:134066. https://d​ oi.​ The work of S.V. was supported in part by the French Agence Nationale org/​10.​1016/j.​scito​tenv.​2019.​134066 de la Recherche (grant ANR-17-CE07-0009 to project dehalofluidX). Cantalapiedra CP, Hernández-Plaza A, Letunic I, Bork P, Huerta-Cepas J (2021) eggNOG-mapper v2: functional annotation, orthology Data availability Illumina and PacBio metagenome sequences, Illu- assignments, and domain prediction at the metagenomic scale. Mol mina 16S rRNA gene sequences, and the draft genome are available Biol Evol 38:5825–5829. https://​doi.​org/​10.​1093/​molbev/​msab2​93 under GenBank BioProject Accession PRJNA857723. The assem- Chen G, Fisch AR, Gibson CM, Mack EE, Seger ES, Campagna SR, Löf- bled genome is available at NCBI under Genbank accession number fler FE (2020) Mineralization versus fermentation: evidence for two GCA_024705665. The final assembly, as well as a preliminary draft distinct anaerobic bacterial degradation pathways for dichlorometh- genome assembly also including 16S rRNA genes, is also available ane. ISME J 14:959–970. https://​doi.​org/​10.​1038/​s41396-​019-​0579-5 on the Microscope platform at https://​mage.​genos​cope.​cns.​fr/​micro​ Chen G, Jiang N, Villalobos Solis MI, Murdoch FK, Murdoch RW, Xie scope/​mage/​viewer.​php?O_​id=​13191 and https://​mage.​genos​cope.​ Y, Swift CM, Hettich RL, Löffler FE (2021) Anaerobic metabo- cns.​fr/​micro​scope/​genom​ic/​overv​iew.​php?O_​id=​11670, respectively. lism of dichloroacetate. mBio 12:e00537–e00521. https://d​ oi.o​ rg/​ 10.​1128/​mBio.​00537-​21 Declarations Chen G, Murdoch RW, Mack EE, Seger ES, Löffler FE (2017) Com- plete genome sequence of Dehalobacterium formicoaceticum Ethics approval Not applicable. strain DMC, a strictly anaerobic dichloromethane-degrading bacterium. Genome Announc 5:e00897–e00817. https://​doi.​org/​ Consent to participate All authors that appear in the submission: 10.​1128/​genom​eA.​00897-​17 • made substantial contributions to the conception or design of the Chen G, Shouakar-Stash O, Phillips E, Justicia-Leon SD, Gilevska work T, Sherwood Lollar B, Mack EE, Seger ES, Löffler FE (2018) • drafted the work or revised it critically for important intellectual Dual carbon–chlorine isotope analysis indicates distinct anaerobic content; dichloromethane degradation pathways in two members of Pep- • approved the version to be published; and tococcaceae. Environ Sci Technol 52:8607–8616. https://​doi.​org/​ • agree to be accountable for all aspects of the work in ensuring that 10.​1021/​acs.​est.​8b015​83 questions related to the accuracy or integrity of any part of the work Deshpande NP, Wong YK, Manefield M, Wilkins MR, Lee M (2013) are appropriately investigated and resolved. Genome sequence of Dehalobacter UNSWDHB, a chloroform- dechlorinating bacterium. Genome Announc 1:e00720–e00713. Consent for publication All authors agree with the content of this https://​doi.​org/​10.​1128/​genom​eA.​00720-​13 manuscript and all give explicit consent to submit in Environmental Edgar R (2010) Search and clustering orders of magnitude faster than Science and Pollution Research. This work is new and original, and it BLAST. Bioinformatics 26:2460–2461. https://​doi.​org/​10.​1093/​ is not under consideration elsewhere. bioin​forma​tics/​btq461 Freedman DL, Gossett JM (1991) Biodegradation of dichlorometh- Competing Interests The authors declare no competing interests. ane and its utilization as a growth substrate under methanogenic 13 Environmental Science and Pollution Research conditions. Appl Environ Microbiol 57:2847–2857. https://​doi.​ Marchler-Bauer A, Bryant SH (2004) CD-Search: protein domain org/​10.​1128/​aem.​57.​10.​2847-​2857.​1991 annotations on the fly. Nucleic Acids Res 1:W327–W331. https://​ Freedman DL, Smith CR, Noguera DR (1997) Dichloromethane bio- doi.​org/​10.​1093/​nar/​gkh454 degradation under nitrate-reducing conditions. Water Environ Res Muller EEL, Bringel F, Vuilleumier S (2011) Dichloromethane-degrad- 69:115–122. https://​doi.​org/​10.​2175/​10614​3097x​125245 ing bacteria in the genomic age. Res Microbiol 162:869–876. Goris J, Konstantinidis KT, Klappenbach JA, Coenye T, Vandamme https://​doi.​org/​10.​1016/j.​resmic.​2011.​01.​008 P, Tiedje JM (2007) DNA-DNA Hybridization values and their Murdoch RW, Chen G, Murdoch FK, Mack EE, Villalobos Solis MI, relationship to whole-genome sequence similarities. Int J Syst Hettich RL, Löffler FE (2022) Identification and widespread envi- Evol Microbiol 57:81–91. https://​doi.​org/​10.​1099/​ijs.0.​64483-0 ronmental distribution of a gene cassette implicated in anaerobic Holland SI, Edwards RJ, Ertan H, Wong YK, Russell TL, Deshpande NP, dichloromethane degradation. Glob Chang Biol 28:2396–2412. Manefield MJ, Lee M (2019) Whole genome sequencing of a novel, https://​doi.​org/​10.​1111/​gcb.​16068 dichloromethane-fermenting Peptococcaceae from an enrichment Nguyen L-T, Schmidt HA, von Haeseler A, Minh BQ (2015) IQ-TREE: culture. PeerJ 7:e7775. https://​doi.​org/​10.​7717/​peerj.​7775 A fast and effective stochastic algorithm for estimating maximum- Holland SI, Ertan H, Montgomery K, Manefield MJ, Lee M (2021) likelihood phylogenies. Mol Biol Evol 32:268–274 Novel dichloromethane-fermenting bacteria in the Peptococ- Parks DH, Mc I, Skennerton CT, Hugenholtz P, Tyson GW (2015) caceae family. ISME J 15:1709–1721. https://​doi.​org/​10.​1038/​ CheckM: assessing the quality of microbial genomes recovered from s41396-​020-​00881-y isolates, single cells, and metagenomes. Genome Res 25:1043–1055 Hossaini R, Chipperfield MP, Montzka SA, Leeson AA, Dhomse SS, Richter M, Rosselló-Móra R, Glöckner FO, Peplies J (2016) JSpe- Pyle JA (2017) The increasing threat to stratospheric ozone from ciesWS: A web server for prokaryotic species circumscription dichloromethane. Nat Commun 8:15962. https://​doi.​org/​10.​1038/​ based on pairwise genome comparison. Bioinformatics 32:929– ncomm​s15962 931. https://​doi.​org/​10.​1093/​bioin​forma​tics/​btv681 Kang DD, Li F, Kirton E, Thomas A, Egan R, An H, Wang Z (2019) Schlosser PM, Bale AS, Gibbons CF, Wilkins A, Cooper GS (2015) MetaBAT 2: An adaptive binning algorithm for robust and effi- Human health effects of dichloromethane: key findings and scien- cient genome reconstruction from metagenome assemblies. PeerJ tific issues. Environ Health Perspect 123:114–119 7:e7359. https://​doi.​org/​10.​7717/​peerj.​7359 Schmieder R, Edwards R (2011) Quality control and preprocessing of Kleindienst S, Chourey K, Chen G, Murdoch RW, Higgins SA, Iyer R, metagenomic datasets. Bioinformatics 27:863–864 Campagna SR et al (2019) Proteogenomics reveals novel reduc- Schubert M, Lindgreen S, Orlando L (2016) AdapterRemoval v2: rapid tive dehalogenases and methyltransferases expressed during adapter trimming, identification, and read merging. BMC Res anaerobic dichloromethane metabolism. Appl Environ Microbiol Notes 9:88. https://​doi.​org/​10.​1186/​s13104-​016-​1900-2 85:e02768–e02718. https://​doi.​org/​10.​1128/​AEM.​02768-​18 Seidel K, Kühnert J, Adrian L (2018) The complexome of Dehalococcoides Kolusu SR, Schlünzen KH, Grawe D, Seifert R (2018) Determination of mccartyi reveals its organohalide respiration-complex is modular. chloromethane and dichloromethane in a tropical terrestrial man- Front Microbiol 9:1130. https://​doi.​org/​10.​3389/​fmicb.​2018.​01130 grove forest in Brazil by measurements and modelling. Atmos Envi- Tang S, Wang PH, Higgins SA, Löffler FE, Edwards EA (2016) Sister ron 173:185–197. https://​doi.​org/​10.​1016/j.​atmos​env.​2017.​10.​057 Dehalobacter genomes reveal specialization in organohalide respiration Kühner M, Haufschildt K, Neumann A, Storbeck S, Streif J, Layer and recent strain differentiation likely driven by chlorinated substrates. G (2014) The alternative route to heme in the methanogenic Front Microbiol 7:100. https://​doi.​org/​10.​3389/​fmicb.​2016.​00100 archaeon Methanosarcina bakeri. Archaea 2014:327637. https://​ Trueba-Santiso A, Fernández-Verdejo D, Marco-Rius I, Soder-Walz doi.​org/​10.​1155/​2014/​327637 JM, Casabella O, Vicent T, Marco-Urrea E (2020) Interspecies Li H (2021) New strategies to improve minimap2 alignment accuracy. interaction and effect of co-contaminants in an anaerobic dichlo- Bioinformatics 37:4572–4574. https://d​ oi.o​ rg/1​ 0.1​ 093/b​ ioinf​ orma​ romethane-degrading culture. Chemosphere 240:124877. https://​ tics/​btab7​05 doi.​org/​10.​1016/j.​chemo​sphere.​2019.​124877 Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abe- Trueba-Santiso A, Parladé E, Rosell M, Lliros M, Hatijah Mortan S, Mar- casis G, Durbin R, 1000 Genome Project Data Processing Subgroup tínez-Alonso M, Gaju N et al (2017) Molecular and carbon isotopic (2009) The sequence alignment/Map format and SAMtools. Bioinfor- characterization of an anaerobic stable enrichment culture contain- matics 25:2078–2079. https://​doi.​org/​10.​1093/​bioin​forma​tics/​btp352 ing Dehalobacterium sp. during dichloromethane fermentation. Sci Lu S, Wang J, Chitsaz F, Derbyshire MK, Geer RC, Gonzales NR, Total Environ 581:640–648. https://​doi.​org/​10.​1016/j.​scito​tenv.​2016.​ Gwadz M, Hurwitz DI, Marchler GH, Song JS et al (2020) CDD/ 12.​174 SPARCLE: the conserved domain database in 2020. Nucleic Vallenet D, Calteau A, Dubois M, Amours P, Bazin A, Beuvin M, Acids Res 8:D265–D268. https://​doi.​org/​10.​1093/​nar/​gkz991 Burlot L, Bussell X, Fouteau S, Gautreau G et al (2020) Micro- Mägli A, Messmer M, Leisinger T (1998) Metabolism of dichlorometh- Scope: an integrated platform for the annotation and exploration ane by the strict anaerobe Dehalobacterium formicoaceticum. of microbial gene functions through genomic, pangenomic and Appl Environ Microbiol 64:646–650. https://​doi.​org/​10.​1128/​ metabolic comparative analysis. Nucl Acids Res 48:D579–D589. aem.​64.2.​646-​650.​1998 https://​doi.​org/​10.​1093/​nar/​gkz926 Mägli A, Rainey FA, Leisinger T (1995) Acetogenesis from dichlo- Wang Q, Garrity G, Tiedje JM, Cole JR (2007) Naive bayesian classi- rmethane by a two-component mixed culture comprising a novel fier for rapid assignment of rRNA sequences into the new bacterial bacterium. Appl Environ Microbiol 61:2943e2949. https://d​ oi.o​ rg/​ taxonomy. Appl Environ Microbiol 73:5261–5267. https://d​ oi.o​ rg/​ 10.​1128/​aem.​61.8.​2943-​2949.​1995 10.​1128/​AEM.​00062-​07 Mägli A, Wendt M, Leisinger T (1996) Isolation and characterization Wick RR, Judd LM, Gorrie CL, Holt KE (2017) Unicycler: resolv- of Dehalobacterium formicoaceticum gen. nov. sp. nov., a strictly ing bacterial genome assemblies from short and long sequencing anaerobic bacterium utilizing dichloromethane as source of car- reads. PLoS Comput Biol 13:e1005595. https://​doi.​org/​10.​1371/​ bon and energy. Arch Microbiol 166:101–108. https://​doi.​org/​10.​ journ​al.​pcbi.​10055​95 1007/​s0020​30050​362 Magoč T, Salzberg SL (2011) FLASH: fast length adjustment of short Publisher’s note Springer Nature remains neutral with regard to reads to improve genome assemblies. Bioinformatics 27:2957– jurisdictional claims in published maps and institutional affiliations. 2963. https://​doi.​org/​10.​1093/​bioin​forma​tics/​btr507 13