Received: 22 February 2023 Accepted: 4 July 2023 DOI: 10.1111/1462-2920.16464 BRIEF REPORT Respiratory protein interactions in Dehalobacter sp. strain 8M revealed through genomic and native proteomic analyses Jesica M. Soder-Walz 1 | Kenneth Wasmund 2,3 | Darja Deobald 4 | Teresa Vicent 1 | Lorenz Adrian 4,5 | Ernest Marco-Urrea 1 1 Departament d’Enginyeria Química, Biològica i Ambiental, Universitat Autònoma de Abstract Barcelona (UAB), Bellaterra, Spain Dehalobacter (Firmicutes) encompass obligate organohalide-respiring bac- 2 Division of Microbial Ecology, Centre for teria used for bioremediation of groundwater contaminated with haloge- Microbiology and Environmental Systems Science, University of Vienna, Vienna, Austria nated organics. Various aspects of their biochemistry remain unknown, 3 School of Biological Sciences, University of including the identities and interactions of respiratory proteins. Here, we Portsmouth, Portsmouth, UK sequenced the genome of Dehalobacter sp. strain 8M and analysed its pro- 4 Department Environmental Biotechnology, tein expression. Strain 8M encodes 22 reductive dehalogenase homologous Helmholtz Centre for Environmental (RdhA) proteins. RdhA D8M_v2_40029 (TmrA) was among the two Research—UFZ, Leipzig, Germany 5 most abundant proteins during growth with trichloromethane and Chair of Geobiotechnology, Technische Universität Berlin, Berlin, Germany 1,1,2-trichloroethane. To examine interactions of respiratory proteins, we used blue native gel electrophoresis together with dehalogenation activity Correspondence tests and mass spectrometry. The highest activities were found in gel slices Lorenz Adrian, Department Environmental with the highest abundance of TmrA. Protein distributions across gel lanes Biotechnology, Helmholtz Centre for Environmental Research—UFZ, provided biochemical evidence that the large and small subunits of the Permoserstraße 15, 04318 Leipzig, Germany. membrane-bound [NiFe] uptake hydrogenase (HupL and HupS) interacted Email: lorenz.adrian@ufz.de strongly and that HupL/S interacted weakly with RdhA. Moreover, the inter- action of RdhB and membrane-bound b-type cytochrome HupC was Funding information European Regional Development Fund; detected. RdhC proteins, often encoded in rdh operons but without Generalitat de Catalunya, Consolidated described function, migrated in a protein complex not associated with Research Group, Grant/Award Number: HupL/S or RdhA. This study provides the first biochemical evidence of respi- 2021-SGR-01008; Spanish Ministry of Economy and Competitiveness State ratory protein interactions in Dehalobacter, discusses implications for the Research Agency, Grant/Award Number: respiratory architecture and advances the molecular comprehension of this PID2019-103989RB-I00 unique respiratory chain. INTRODUCTION Substances and Disease Registry Priority List of Haz- ardous Substances, respectively, based on a combina- Organohalides are frequently found as pollutants in the tion of the frequency of their detection, toxicity and environment. Although some organohalides are natu- potential for human exposure (ATSDR, 2022). rally generated, their presence in groundwater is due to Bioremediation using organohalide-respiring bacte- their wide use in industrial processes followed by ria (OHRB) is a feasible and low-cost strategy to treat improper management and accidental spills. Trichloro- contaminated groundwater (Bla zquez-Pallí et al., 2019; methane (TCM) and 1,1,2-trichloroethane (1,1,2-TCA) Jugder et al., 2016). OHRB are anaerobic microorgan- are among those of environmental concern and are isms characterized by their ability to gain energy for ranked 11 and 170 on the 2022 Agency for Toxic growth by using halogenated compounds as terminal This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2023 The Authors. Environmental Microbiology published by Applied Microbiology International and John Wiley & Sons Ltd. Environ Microbiol. 2023;1–17. wileyonlinelibrary.com/journal/emi 1 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2 SODER-WALZ ET AL. electron acceptors, yielding less-halogenated com- et al., 2011). PceC was believed to encode a putative pounds as products (Wang et al., 2018). Reductive transcriptional regulator in Desulfitobacterium haf- dehalogenase homologous (RdhA) are the enzymes niense strain Y51 (Jugder et al., 2015), although responsible for reductive dehalogenation and usually recently it has been tentatively proposed to form a contain two iron–sulfur clusters and a corrinoid cofactor membrane-bound protein with PceA and PceB with (Hug et al., 2013; Kräutler et al., 2003). To date, anaero- a role in the electron chain based on the predicted bic respiratory dechlorination of TCM has been found redox activity of a flavin mononucleotide (FMN)-binding only in bacteria of the genera Dehalobacter and Desulfi- domain in Des. hafniense (Buttet et al., 2018). A recent tobacterium. Characterized TCM-dechlorinating RdhA study using quantitative proteomics found a similar proteins include CtrA (Ding et al., 2014), ThmA (Zhao abundance of PceA and PceB, suggesting they could et al., 2017), TmrA (Wong et al., 2016) and CfrA (Tang & form protein complexes in Deh. restrictus strain PER- Edwards, 2013). All these RdhA proteins except for K23 and Des. hafniense strain TCE1, but PceC was ThmA are also active against 1,1,2-TCA. DcrA has not present in the complex (Cimmino et al., 2022). In dechlorination activity with both 1,1-dichloroethane and conclusion, the function of RdhC in Dehalobacter, 1,1,2-TCA (Tang & Edwards, 2013). which is homologous to PceC, is unclear. Most RdhA proteins are produced with an Previous work in our laboratory focused on the N-terminal twin-arginine leader (TAT) peptide that tar- enrichment of a Dehalobacter-containing culture from gets them for export through the cytoplasmic mem- TCM-contaminated groundwater samples (Soder-Walz brane. After cleavage of the TAT-leader peptide, they et al., 2022). The 16S rRNA gene sequence of the keep attached to the cytoplasmic membrane presum- dominant Dehalobacter strain was 99.66% similar to ably by interaction with the integral membrane protein the 16S rRNA gene sequence of Deh. restrictus strain RdhB. Other proteins involved in the flow of electrons PER-K23, and it was denominated Dehalobacter between hydrogenases and RdhA have been studied, sp. strain 8M (Soder-Walz et al., 2022). The strain 8M but their role is not fully understood. Recent studies enrichment completely transformed TCM to dichloro- provided direct biochemical evidence that quinone- methane (DCM) and 1,1,2-TCA to a mixture of vinyl independent respiration in Dehalococcoides and Deha- chloride (VC) and 1,2-dichloroethane (1,2-DCA). logenimonas depends on the presence of a large Compound-specific isotope analysis revealed signifi- membrane-bound protein complex (Kublik et al., 2016; cant differences in both carbon and chlorine isotope Seidel et al., 2018; Trueba-Santiso et al., 2021). In effects during dechlorination of TCM by our Dehalobac- addition to the RdhA and RdhB proteins, this respira- ter sp. strain 8M enrichment compared to other Dehalo- tory complex contains the three canonical subunits of a bacter-containing enrichments (Heckel et al., 2019; complex iron–sulfur molybdoenzyme with the Soder-Walz et al., 2022). This phenomenon has molybdopterin-binding subunit OmeA (potentially also recently been attributed to amino acid differences in the having an iron–sulfur cluster), a four iron–sulfur cluster vicinity of the active site of the TCM-transforming RdhA protein HupX and multiple transmembrane domains (Phillips et al., 2022). containing integral membrane protein OmeB (Rothery Here, we report the analysis of the genome of the et al., 2008) and the two subunits of hydrogen (H2) Dehalobacter sp. strain 8M and its proteome profile uptake hydrogenase (HupL and HupS; Seidel during growth with TCM and 1,1,2-TCA. We also com- et al., 2018; Trueba-Santiso et al., 2021). bined blue native polyacrylamide gel electrophoresis A tentative model for the respiratory chain in the (BNE), dehalogenation activity assays and protein Dehalobacter genus has been proposed for Dehalo- mass spectrometry to provide biochemical evidence of bacter restrictus strain PER-K23 based on biochemical, the molecular organization of the respiratory chain genomic and proteomic data and consisted of H2 of this Dehalobacter strain, together with the identifica- uptake by a cytochrome b containing Hup-type [NiFe]- tion of the RdhA involved during growth with TCM and hydrogenase (HupA [=HupS], HupB [=HupL] and 1,1,2-TCA. HupC), which was described to transfer electrons to menaquinone. From the reduced menaquinone (mena- quinol) electrons are described to be transferred to EXPERIMENTAL PROCEDURES halogenated compounds via RdhA (Maillard & Holliger, 2016). In strain PER-K23, tetrachloroethene Chemicals and cultivation (PCE) reductive dehalogenase (PceA) and its mem- brane anchor PceB are encoded in the genome as part All chemicals used were of the highest available purity of a four-gene cluster ( pceABCT), which is a unique (Supporting Information Materials and Methods S1). characteristic of the genera Dehalobacter and Desulfi- Dehalobacter sp. strain 8M was grown in 100-mL tobacterium compared to other OHRB (Duret glass serum bottles with 65 mL of an anoxic defined et al., 2012). PceT is predicted to function as a molecu- medium containing 0.2 g L 1 potassium dihydrogen lar chaperone assisting in the folding of PceA (Maillard phosphate, 0.27 g L 1 ammonium chloride, 1 g L 1 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License PROTEIN INTERACTIONS IN Dehalobacter SP. STRAIN 8M 3 sodium chloride, 0.41 g L 1 magnesium chloride hexa- Metagenome sequencing and analysis hydrate, 0.52 g L 1 potassium chloride, 0.15 g L 1 cal- cium chloride dihydrate and 0.25 mg L 1 resazurin as A volume of 700 mL fully grown culture liquid that had redox indicator. The following trace elements were transformed approximately 2100 μmol TCM (3000 μM) added: 0.7 mg L 1 zinc chloride, 0.8 mg L 1 manga- was centrifuged at 8000g for 40 min at 16 C to collect nese (II) chloride tetrahydrate, 0.06 mg L 1 boric acid, cells. Half of the supernatant was removed and dis- 0.19 mg L 1 cobalt (II) chloride hexahydrate, carded, and the remaining volume was centrifuged 0.02 mg L 1 copper (II) chloride dihydrate, 0.24 mg L 1 again. This process was repeated three times and the nickel (II) chloride hexahydrate, 0.36 mg L 1 sodium resulting cell pellet was washed once with 1 mL of ster- molybdenum oxide dihydrate, 20 mg L 1 iron ile 100 mM phosphate buffer (pH 7.2). Finally, the cells (II) chloride tetrahydrate, 128 mg L 1 nitrilotriacetic were suspended in 1 mL of sterile 100 mM phosphate acid, 0.02 mg L 1 sodium tungstate dihydrate, buffer (pH 7.2). Genomic DNA was extracted with the 0.015 mg L 1 sodium selenite and 0.68 g L 1 sodium Gentra Puregene Yeast/Bact kit (Qiagen) following acetate as carbon source. The culture medium was ali- the manufacturer’s instructions. quoted in serum bottles inside an anaerobic glovebox, DNA was sequenced in 2  150 bp paired-end closed with Teflon-coated butyl rubber stoppers and mode using a HiSeq 3000 (Illumina) at the Biomedical aluminium crimp caps and autoclaved at 121 C for Sequencing Facility, Vienna. Sequence reads used for 30 min. Once autoclaved and cooled, the following genome assembly were trimmed of adapters using components were added under anoxic and sterile con- AdapterRemoval with default settings (Schubert ditions: 0.7 g L 1 sodium bicarbonate to buffer the pH et al., 2016). Sequence reads were also trimmed based at 7.0, 24 mg L 1 Na2S9H2O and 48 mg L 1 L-cysteine on quality using a python3 script (https://github.com/ as reducing agents, 0.2 g L 1 yeast extract and a vita- kwasmund/Trim-Illumina), with the following flags -q min solution containing: 5 μg L 1 D-(+)-biotin, 15 -m 50 –b 8. Sequences were assembled using 25 μg L 1 calcium D-(+)-pantothenate, 50 μg L 1 thia- SPAdes (v. 3.10.1) with pre-assembly read-error cor- mine hydrochloride, 20 μg L 1 4-aminobenzoic acid, rection (Bankevich et al., 2012). Contigs <5000 bp were 50 μg L 1 nicotinic acid, 75 μg L 1 pyridoxine hydro- discarded. Contigs that were more or less than 60–65 chloride and 50 μg L 1 cyanocobalamin. TCM or coverage were also discarded, because BLASTX ana- 1,1,2-TCA was added with a syringe through the sep- lyses against the National Center for Biotechnology tum as electron acceptor and the cultures were gassed Information (NCBI) nr database identified that contigs with H2 to 0.4 bar overpressure, supplied as electron with this coverage all belonged to Dehalobacter. donor. All bottles were incubated statically in the dark The genome sequence was automatically anno- at 25 C. The final pH of the medium was 7.0. During tated using the RAST server using ‘classic mode’ (Aziz the exponential growth phase and when accompanied et al., 2008) and also using the MAGE MicroScope by an exponential increase of organohalide transforma- pipeline (Vallenet et al., 2020). The loci numbers used tion, active cultures were transferred into fresh medium in this manuscript are from the MicroScope annotation (4.3% v/v inoculum) prior to the amendment of H2. All and are preceded by the letters ‘D8M_v2_’. The Jspe- experiments were performed using cultures growing for ciesWS website (Richter et al., 2016) was used to cal- more than five transfers with the same electron accep- culate the average nucleotide identities (ANI) of tor. Abiotic controls without inoculum were included in genomes by BLASTn. Genome quality was evaluated triplicate to evaluate abiotic transformation. with CheckM (Parks et al., 2015). Gas chromatography Protein biochemistry Organohalide compounds were quantified using an Agi- Cells for whole-cell proteomic analysis were harvested lent 6890 N gas chromatograph with a flame ionization from biological triplicate samples of 140 mL each upon detector, equipped with a DB-624 column (30 m length, the consumption of 2500 μM of TCM or 1,1,2-TCA. The 0.32 mm inner diameter with 0.25 mm film thickness) by cultures were repeatedly centrifuged at 8000g for injecting 0.5 mL headspace samples, as described pre- 40 min at 16 C and only half of the supernatant was viously (Trueba-Santiso et al., 2017). Data acquisition removed per centrifuge cycle to avoid the loss of cells. and peak integration were performed using Chromeleon The procedure was repeated three times and 4 mL cell 6.8 Chromatography Software (Dionex Corporation). concentrates were obtained (resulting in a 35-fold con- The results are expressed as nominal concentrations, centration) for each electron acceptor. Cell concen- which correct for headspace-liquid partitioning equilib- trates were disrupted by the freeze–thaw method rium within the bottle. Calibration was based on aqueous ( 80 C/+40 C repeated five times), followed by soni- standards, with the same liquid and headspace volumes cation for 30 s. Amicon ultrafilters with an exclusion as in the microcosms. size of 10 kDa were employed to concentrate the 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 4 SODER-WALZ ET AL. samples (Amicon Ultra, 0.5 mL Centrifugal Filters, Milli- grown with TCM was cut into 20 slices; the lane pre- pore) by centrifuging at 4 C for 30 min at 14,000g. The pared from 1,1,2-TCA was cut into 18 slices. concentrated proteins were recovered by turning the fil- For dehalogenase activity assays, the dechlorinat- ters upside down and centrifuging for 2 min at 1000g ing activity was quantified in cell suspensions and gel and 4 C. Disulfide bridges in the protein samples were slices of the BNE. In both cases, the experiments were reduced with dithiothreitol and cysteine residues set up inside the anaerobic chamber using 10-mL glass were alkylated with iodoacetamide prior to digestion vials, with 2 mL (final volume) of a reaction mix contain- with 0.6 μg of porcine trypsin at 37 C overnight. The ing 200 mM potassium acetate buffer (pH 5.8), 2 mM reaction was stopped with 0.1% (final concentration) methyl viologen as electron donor and 2 mM titanium formic acid. The extracted peptides were desalted (III) citrate [2 mM in respect to titanium (III)] as reducing using C18 Zip Tip columns for further analysis by nano- agent, cell concentrate or a gel slice and finally liquid chromatography–tandem mass spectrometry amended with 300 μM of TCM or 300 μM of 1,1,2-TCA (nLC-MS/MS; Seidel et al., 2018). from 0.6 M acetonic stock solutions. The resulting ace- For analysis of protein complexes by BNE, cells tone concentration in the assays was 6.4 mM, which were harvested as described above by centrifugation of did not prevent the transformation of halogenated 210 mL of cultures that had consumed 2000 μM compounds. of TCM or 1,1,2-TCA. This resulted in a cell concen- For assays with cell suspensions, 200 μL of a cell trate with about 1010 cells mL 1, as determined by epi- concentrate (108–109 cells mL 1) harvested from the fluorescence microscopic cell counting on agarose- TCM dechlorinating culture were added to each vial coated slides (Adrian et al., 2007). The cells were dis- and triplicate samples were incubated with TCM or rupted by bead beating (Fast Prep FP120, Thermo) in 1,1,2-TCA, separately. Correspondingly, the same pro- six cycles of 40 s at 4 m s 1, alternating with 1 min on cedure was performed with the culture grown with ice. Membrane proteins were solubilized with 0.1% n- 1,1,2-TCA, although in this case, the cellular density dodecyl-D-maltoside under gentle shaking on ice for was lower (107–108 cells mL 1). 1 h. The solubilized protein extract was recovered as Dehalogenase activity assayed in gel slices from supernatant from a centrifugation step at 14,000g for the BNE of both cultures grown with TCM and 15 min at 4 C. This supernatant was filtered with a 1,1,2-TCA was performed with four non-stained lanes 0.22 μm filter (Millipore) to remove cell debris. from each gel. After cutting the slices of the BNE as Inside the anaerobic chamber, a 4–16% gradient described above, two unstained lanes were incubated Bis-Tris gel (NativePAGE Novex, Invitrogen) was with TCM and the other two with 1,1,2-TCA. The gel loaded with 25 μL per well of protein extract amended slices were smashed into small pieces with a scalpel with 0.125% w/v Coomassie G-250 additive. Native- inside the vials. The vials were sealed with Teflon- Mark Unstained Protein Standard (Invitrogen) was used coated rubber septa and aluminium crimps and as the protein ladder. Anode buffer and light blue cath- amended with the corresponding electron acceptor ode buffer required for the electrophoresis run were using a glass syringe (Hamilton). prepared according to the manufacturer’s instructions, For all dehalogenase activity assays performed, degassed and chilled to 4 C before their use. The elec- controls were added in triplicates. Abiotic controls con- trophoresis was run at 150 V for 1 h and the voltage sisted of 2 mL of the reaction mix amended with was then raised to 250 V for 30 min. The BNE system 300 μM of TCM or 1,1,2-TCA, without adding cells or was kept cold by using ice packs during the run. gel slices. In the case of the activity assays performed After the electrophoresis run, two replicate lanes with the gel slices, two additional controls were with cell extract and one lane with the protein ladder included: (i) the positive control consisted of 25 μL of were separated from the rest of the gel and these three the protein extract instead of the gel slices incubated lanes were silver-stained with a mass-spectrometry with the reaction mix and (ii) the negative control con- compatible procedure (Kublik et al., 2016). Further four sisted of the remaining pellet after membrane protein replicate lanes with cell extract were not stained and solubilization incubated with the reaction mix to assess were kept cold and wetted with anode buffer until they the performance of the solubilization protocol. were cut into slices using a scalpel that was cleaned Finally, the vials were vortexed and incubated after each cut. The cuts were guided by the bands upside down in the dark, at 25 C shaking at 150 rpm. observed in the parallel silver-stained lanes. The cut- After 24 h, gas chromatography was used to analyse out unstained gel slices were placed into individual the transformation of TCM or 1,1,2-TCA and the pro- 10-mL glass vials to perform activity assays (see duction of DCM or 1,2-DCA and VC, respectively. The below). The two stained gel lanes were used to identify activity in each experimental vial was expressed as the proteins in the visible bands by protein mass spectrom- percentage of chlorinated substrate that was trans- etry. For that, the two stained lanes were cut into slices formed into product(s) after 24 h of incubation com- of equal size for subsequent in-gel trypsin digest (see pared to the initial amount of substrate added to below). The BNE gel lane prepared from the culture each vial. 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License PROTEIN INTERACTIONS IN Dehalobacter SP. STRAIN 8M 5 For in-gel trypsin digestion and protein mass spec- Phylogenetic analysis of RdhA trometry, duplicate silver-stained slices from BNE gels were cut into slices as described above for activity Phylogenetic analysis of RdhA sequences was per- assays, washed with ddH2O and de-stained in a 1:1 formed with the Maximum Likelihood algorithm via IQ- (v/v) solution containing 30 mM K3[Fe(CN)6] and TREE by W-IQ-TREE (Trifinopoulos et al., 2016). We 100 mM Na2O3S2. Then, the proteins in the gel slices included the reference sequences set described by were reduced with dithiothreitol, their cysteine residues (Hug et al., 2013) and selected reference sequences were carbamidylated with iodoacetamide and trypsin obtained from the NCBI Genbank database from the digested. The resulting peptides were extracted from best BLASTP hits to proteins of interest, that is, Deha- the gel pieces and desalted as described previously lobacter sp. 8M proteins 40029 (TmrA) and 260094. (Vallenet et al., 2020). Extracted peptide samples Sequences were aligned using MUSCLE obtained after in-solution or in-gel digestion were ana- (Edgar, 2004), and the TAT signal peptide sequences lysed by nLC-MS/MS on a nanoUPLC system (nanoAc- were removed after the sequences were aligned. The quity, Waters) coupled to an Orbitrap Fusion mass automatic substitution model finder was used and spectrometer (Thermo Scientific) as described previ- resulted in the best-fit model being PMB + F + I + G4. ously (Seidel et al., 2018). Proteome Discoverer (v2.2, Bootstrapping was done with 1000 replications. Thermo Fisher Scientific) was used for the identification of proteins, based on a protein database from the draft genome of Dehalobacter sp. strain 8M using RESULTS SequestHT as a search engine. A false discovery rate threshold of 1% was set for peptide identification using General features of the genome and the Target Decoy PSM Validator node. The Minora proteomic analysis node in Proteome Discoverer was used to quantify the abundance of proteins using label-free quantification The sequenced genome of Dehalobacter sp. strain 8M based on intensity values in precursor scans. The rela- consists of 3,194,592 bp within 62 contigs with an aver- tive protein abundance presented in this study is the age GC content of 44.5%, and it contains 3400 pre- ratio of a protein abundance in one slice to its overall dicted protein-coding genes. No plasmids were abundance across all slices of one blue native gel lane. identified. CheckM analysis indicated that the genome The hierarchical cluster analysis was performed with was 99.4% complete with 0.8% contamination. The RStudio software using R programming language, genome of strain 8M is similar to Deh. restrictus strain applying hcluster using default settings (complete link- PER-K23, Dehalobacter sp. strain TeCB1 and Dehalo- age) and plotted with ggplot to the relative abundance bacter sp. strain E1, sharing 98.72%, 98.70% and of all proteins detected in one gel lane. 98.13% average nucleotide identities, respectively (Table S1). The average nucleotide identities between strain 8M and the two TCM-dechlorinating Dehalobac- Protein structure prediction and ter strains UNSWDHB and CF were 94.84% and visualization 94.53%, respectively. We compared the gene annotations of strain 8M for Multiple sequence alignment of RdhC homologues from selected pathways and proteins with six available different bacteria were calculated with MEGA11 using genomes of Dehalobacter sp. strains and examined an implemented MUSCLE algorithm with default set- which proteins were expressed in triplicate cultures of tings (Tamura et al., 2021). Signal peptide sequences strain 8M grown either with TCM or 1,1,2-TCA. The were identified using SignalP 6.0 (Teufel et al., 2022). shotgun proteomic analysis of strain 8M identified Protein structure prediction of RdhC 40031 of Dehalo- the expression of a total of 1014 proteins, which repre- bacter sp. strain 8M was conducted with ColabFold sented 29.8% of the annotated proteins (Table S2). (Mirdita et al., 2022). Protein homology was detected The MicroScope gene locus tag prefix ‘D8M_v2_’ will by HMM-HMM comparison using the HHpred server of be omitted in the following sections when referring to MPI Bioinformatics Toolkit (Zimmermann et al., 2018) specific genes. and with Conserved Domain Architecture Retrieval Tool The genome contained 22 putative reductive deha- (Geer et al., 2002) of the NCBI (Sayers et al., 2022). logenases homologous (rdhA) genes, genes for Cofactor binding sites were predicted with the COFAC- 20 putative membrane-anchoring proteins (rdhB; TOR server (Zhang et al., 2017) and MetalPredator Hölscher et al., 2004; Hug et al., 2013), 5 rdhC genes server (Valasatava et al., 2016). Structure modelling, (homologues to Deh. restrictus pceC; Buttet refinement and alignment of RdhC and NqrC subunit of et al., 2018) and 2 genes for the molecular chaperone the sodium (Na+)-translocating NADH:ubiquinone oxi- (rdhT; Table S3). Of the 22 rdhA genes, 18 appear to doreductase protein complex from Vibrio cholerae were encode for full-length RdhA proteins of 375–512 amino done with PyMOL 2.5.4 (Schrödinger & DeLano, 2020). acids including an N-terminal TAT-leader peptide 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 6 SODER-WALZ ET AL. F I G U R E 1 RdhA, RdhB, RdhC and HupS, HupL and HupC proteins were detected in full protein extracts of Dehalobacter sp. strain 8M grown with TCM or 1,1,2-TCA as only electron acceptors. The data were extracted from the proteomic analyses shown in Table S2. Cultures grown with TCM are represented in dark blue and cultures grown with 1,1,2-TCA in watery green. HupC, b-type cytochrome; HupS, hydrogenase small subunit; HupL, hydrogenase large subunit; RdhA, reductive dehalogenase homologous; RdhB, reductive dehalogenase anchor protein; RdhC, reductive dehalogenase probable membrane associated redox protein; TCM, trichloromethane; 1,1,2-TCA, 1,1,2-trichloroethane. (RRXFXK) and two FeS cluster binding motifs RdhC proteins were predicted to be integral membrane (CX2CX2CX3CP and CX10-12CX2CX3CP; Table S3); in proteins with 5–7 transmembrane domains and a none of the 18 sequences we detected a corrinoid- length of around 400 amino acids. The detailed analy- binding site starting in between the two FeS cluster sis of the RdhC structure is described in the last chap- binding motifs, as described for few other reductive ter of the ‘Results’ section. dehalogenases in Dehalococcoides strains DXHX2G- Shotgun proteomic analysis of strain 8M revealed X50-SXL-X42-GG and DHXG-X39-S-X32-G (Hölscher the expression of two RdhAs in the cultures grown et al., 2004; Padilla-Crespo et al., 2014). In contrast, with TCM (40029 and 260084) and three RdhAs in the we identified similar motifs downstream of the second cultures grown with 1,1,2-TCA (40029, 160012 and FeS cluster binding motifs in 17 of the 18 full RdhA 260094; Figure 1). RdhA 40029 was expressed three sequences (WXHX2G/V/A-Xn-FGYG) but the amino orders of magnitude higher than the second most acid residues of these motifs are located far away from abundant reductive dehalogenase, and it was ranked the corrinoid in the model structure; therefore they first and second relative to all proteins expressed in might fulfil other functions than corrinoid-binding. Four the cultures grown with TCM and 1,1,2-TCA, respec- rdhA genes were fragments of the N-terminal (20015) tively (Table S2). To test the substrate specificity of or C-terminal side (20010, 170062, 550001) but the the detected RdhAs, cell suspensions of cultures fragmentation of rdhA 20015 and 550001 were arte- grown on TCM or 1,1,2-TCA were used to perform facts of assembly breaks as they were the last and first independent enzymatic assays with both substrates open reading frames in a contig, respectively. Of the as putative electron acceptors. Enzymatic assays with 20 rdhB genes, 19 appear to encode functional RdhB cells grown with TCM almost completely dechlorinated proteins with a length of 100–105 amino acids (1 excep- both substrates (94.4 ± 4.8% for TCM and 100% for tion of 91 amino acids) and three transmembrane 1,1,2-TCA) and the dechlorination products generated domains each. The rdhB gene 30067 is shorter and showed almost stoichiometric conversion to the identi- would result in an RdhB protein fragment of 60 amino fied products (90.7 ± 2.8% DCM when the substrate acids and two transmembrane helices. Of the five rdhC was TCM and 91.3 ± 4.7% 1,2-DCA + VC when the genes, one (600005) was fragmented due to its position substrate was 1,1,2-TCA; Figure S1). In enzymatic at the end of a contig and was very similar to the assays with cells grown with 1,1,2-TCA, TCM and N-terminus of RdhC 20012, both containing an 1,1,2-TCA were also quantitatively converted to their N-terminal transmembrane helix. The four full-length respective metabolites but with a lower dechlorination 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License PROTEIN INTERACTIONS IN Dehalobacter SP. STRAIN 8M 7 activity (5.6 ± 2.7% for TCM and 9.8 ± 3.0% for Interaction of proteins involved in 1,1,2-TCA; Figure S1). organohalide respiration Phylogenetic analysis of RdhA protein sequences of Dehalobacter sp. strain 8M showed that the RdhA In the cultures grown with TCM, the gel lanes were cut 40029 affiliated directly with a cluster of known TCM into 20 slices, and dechlorinating activity of TCM and and 1,1,2-TCA dechlorinating RdhA enzymes 1,1,2-TCA was observed in slices 5–7 and 4–7, respec- (Figure S2). Together these form ‘orthologue group 46’ tively (Figures 2A and S3A for the duplicate lanes). (all with >90% identity to each other; Hug et al., 2013; Seven different RdhA proteins were detected in the gel Yang et al., 2020). Based on all these characteristics slices, but only three of them (40029, 260084 and we named RdhA 40029 ‘TmrA’, according to the first 260094) were found in both duplicates. The most abun- RdhA protein in orthologue group 46 described to trans- dant RdhA was by far TmrA (40029), with the second form TCM and 1,1,2-TCA (Wong et al., 2016). RdhA (260084) and third (260094) RdhA proteins at one and 260094 clustered in the vicinity of the orthologue group two orders of magnitudes lower, respectively 46, but is distinct and most similar to several uncharac- (Table S6). Interestingly, the three most abundant terized RdhA proteins. The RdhA 260084 is not related RdhA proteins (TmrA, 260084 and 260094), the to any TCM or 1,1,2-TCA dechlorinating RdhA. hydrogenase subunits HupL (230004) and HupS RdhB 40030 and RdhC 40031 that are co-located in (230003) and RdhB (260083) showed a similar migra- one operon with RdhA 40029 were detected in all cul- tion pattern and shared their maximum of relative pro- tures grown either with TCM or 1,1,2-TCA. In addition, tein abundance in the slice that showed the highest the RdhB 140195 was detected in one replicate of the dechlorinating activity (slice 6, 155 kDa; Figures 2 and cultures grown with TCM, and the RdhC 260096 was S4). The two abundant RdhC proteins (40031 detected in one replicate in cultures grown with TCM and 260096) showed identical migration patterns and and 1,1,2-TCA (Tables S2 and S3 and Figure 1). shared their maximum relative protein abundance in Other genes encoding key metabolic functions, and slice 7, at 177 kDa. The two membrane subunits HupC their protein expression, are described in detail in the (230005) and RdhB (40030) were more broadly distrib- Supporting Information and summarized here: For car- uted between slices 1 to 8 but with a similar migration bon metabolism, the genome of strain 8M encodes and pattern to each other (Figures 2B and S4B). The RdhAs expresses a complete Wood–Ljungdahl pathway 20004, 160016 and 140196 had clear maxima at slices and pyruvate: ferredoxin oxidoreductase. Genes for 18, 7 and 1, respectively, but they were expressed at a biosynthesis pathways for key co-factors including very low abundance and only detected in one single menaquinone, riboflavin, tetrahydrofolate, pantothe- slice which distorted the analysis of relative expression nate, CoA, cobalamin, NAD, pyridoxal phosphate and (Figures 2A and S4A and Table S6). All ATP synthase haeme are fully or mostly encoded. For haeme biosyn- subunits (170067–170072, 170074) shared the same thesis, the classical bacterial pathway is absent, but migration pattern and showed a maximum of relative the sirohaeme-dependent route described in some protein abundance in slice 18. The results of protein sulfate-reducing bacteria and archaea is encoded migration in the duplicate lane are consistent with the (Layer, 2021). Genes for biotin biosynthesis are not described results (Figure S4). encoded but the genes for a biotin transporter are pre- In the cultures grown with 1,1,2-TCA, the gel lanes sent. Genes for ubiquinone biosynthesis are not were cut into 18 slices and the dechlorinating activity of encoded in the genome (Table S4). Corrinoids are TCM and 1,1,2-TCA was identified in slices 4, 5 and essential cofactors for reductive dehalogenases. The 6 (Figures 2C and S3B). Seven different RdhA proteins genome encodes a complete set of corrinoid biosynthe- were detected in the gels, but the total abundance of sis enzymes in both, the upper pathway (from glutamyl- RdhA 40029 was between one and four orders of mag- tRNA to cobyrinate) and the lower pathway. In contrast nitude higher than the second and third most abundant to Deh. restrictus strain PER-K23, the genome of strain RdhAs 260094 and 160016, respectively (Table S6). 8M contains an intact cbiH gene (170131) that could The RdhA 40029 had the maximum relative protein confer the capacity of growth without the external addi- abundance in the slices with the highest dechlorination tion of cobalamin. In strain 8M, 22 of the 24 proteins activity of TCM and 1,1,2-TCA in both duplicates (slice involved in corrinoid biosynthesis were detected by 5, 149 kDa), whereas RdhA 260094 had its maximum mass spectrometry in at least one replicate culture, and in a slice that showed high dechlorination activity most of these proteins were detected in all replicates against 1,1,2-TCA (slice 4, 117 kDa) but only in one of (Table S4). For hydrogen metabolism strain 8M the duplicate samples (Figure S4C). The RdhAs encodes and expresses an outwards-facing 160016 and 160006 had relative maxima in slices with membrane-bound three-subunit Hup-type [NiFe] no dechlorinating activity. As observed in the BNE gels hydrogenase and in addition hydrogenases of the Hyc, with cell suspensions grown with TCM, HupL (230004) Ech and Fe-only type (Table S5). and HupS (230003) co-migrated, with a maximum of 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 8 SODER-WALZ ET AL. F I G U R E 2 TCM and 1,1,2-TCA dechlorination percentages in slices of a BNE gel lane relative to the activity in the whole gel lane (bars), as well as the relative abundance of respiratory proteins throughout the gel lane (lines). (A) Culture grown with TCM as electron acceptor illustrates the reductive dehalogenases and the dechlorinating activity and (B) illustrates the distribution of the remaining respiratory proteins. (C) Culture grown with 1,1,2-TCA as electron acceptor illustrates the reductive dehalogenases and the dechlorinating activity and (D) illustrates the distribution of the remaining respiratory proteins. ATP synthase proteins are shown in the figure in (B) and (D) as a positive control for the extraction and analysis of strongly interacting subunits of a membrane-integrated protein complex. Replicate results are available in Figure S4. BNE, blue native polyacrylamide gel electrophoresis; TCM, trichloromethane; 1,1,2-TCA, 1,1,2-trichloroethane. relative protein abundance in slice 4, at 117 kDa. The slice 16 (610 kDa). The results of protein migration two RdhC 40031 and 260096 also co-migrated and in the duplicate lanes are very similar (Figure S4). showed their maximum relative protein abundance in To identify protein interactions potentially important slice 6, at 175 kDa, and HupC (230005) and RdhB in organohalide respiration, we performed a hierarchi- (40030) were distributed throughout the gel (slices 1– cal cluster analysis of the migration patterns and rela- 8). All ATP synthase subunits (170067–170072, tive abundances of different proteins from all four gel 170074) shared the same migration pattern and lanes. We also integrated the distribution of the values showed a maximum of relative protein abundance in for the relative activities against TCM and 1,1,2-TCA in 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License PROTEIN INTERACTIONS IN Dehalobacter SP. STRAIN 8M 9 F I G U R E 3 Hierarchical cluster analysis of BNE gel migration patterns. Note, this is not a phylogenetic analysis of protein sequences. The graph shows similarities in the migration patterns of the dehalogenating activity (black, bold) and the 60 most abundant proteins plus all RdhA (red), RdhB (green), RdhC (violet), HupS (cyan) and HupC (cyan). The ATPase subunits (ocher) function in the analysis as a positive control to confirm that the cluster analysis of the BNE data shows co-migration of the protein subunits of a larger cluster. (A) Proteins from cultures grown with TCM, (B) proteins from cultures grown with 1,1,2-TCA. Replicates of hierarchical cluster analysis with 60 and 125 most abundant proteins are available in Figures S5 and S6, respectively. BNE, blue native polyacrylamide gel electrophoresis; TCM, trichloromethane; 1,1,2-TCA, 1,1,2-trichloroethane. 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 10 SODER-WALZ ET AL. F I G U R E 4 AlphaFold2 model of RdhC 40031 from Dehalobacter sp. strain 8M. (A) Side view highlighting the seven transmembrane helices. The N-terminal domain of RdhC (in green) is homologous to NqrC and RnfG, has one transmembrane anchor and contains 150 exoplasmic amino acid residues forming a flavin mononucleotide (FMN)-binding motif. The C-terminal domain of RdhC (in purple) is similar to NapH and NosR and is composed of two transmembrane helix repeats, each bearing two transmembrane helices and a conserved CX3CP motif. The cysteine residues of the two motifs are highlighted in orange and form a structure potentially complexing a [2Fe-2S] cluster (red and yellow dots) in the middle of the membrane. The glutamic acid residue (E332, cyan stick representation) located between the putative [2Fe-2S] binding site and the FMN (yellow stick representation) might be involved in electron and proton transfer from the cytoplasm through the membrane. Distances are given in Ångstrom (Å). (B) Close-up view from above on the closely aligned sulfhydryl groups of the two CX3CP motifs oriented towards each other, which putatively coordinate a [2Fe-2S] cluster (red and yellow dots). this cluster analysis. The dendrograms support that dis- multiple transmembrane domains at the C-terminus. tinct groups of proteins migrate together, regardless of The N-terminal transmembrane helix and the exoplas- the electron acceptor used: HupL (230004)/HupS mic domain of RdhC 40031 harbour a conserved flavin (230003), RdhB (40030)/HupC (230005) and RdhC mononucleotide (FMN)-binding domain that is also (40031)/RdhC (260096) (Figure 3; Figure S5 shows the found in the redox-coupled sodium transporter NqrC of distribution in the replica lanes). The clustering of ATP Vibrio cholerae (RCSB Protein data bank [PDB] synthase proteins (170067–170072, 170074) indicates no. 4U9S and 7XK7_C) and RnfG of Clostridium teta- they interact with each other forming an independent nomorphum (PDB no. 7ZC6_6) (Buttet et al., 2018). complex. The two measured activities are tightly co- The C-terminal domain of RdhC has no overall strong associated with the most abundant RdhA in each cul- sequence similarity to known proteins. It includes a ture. In the cluster analysis of cultures grown with TCM, duplicated motif with two transmembrane helices each the TmrA (RdhAs 40029), 260094 and 260084 clusters (running through the membrane from the inside to the with both dehalogenating activities in both replicates. In outside and back), each carrying a highly conserved cultures grown with 1,1,2-TCA the activity with TCM CX3CP motif in the cytosolic loop at the end of the two tightly co-migrated with TmrA (RdhA 40029) in both transmembrane helices. replicates but not with other RdhA proteins. The same In the AlphaFold2 structural model of RdhC 40031, was observed with the activity against 1,1,2-TCA, the N-terminal FMN-binding domain is attached to the except for one replicate that clustered with HupL, membrane via an N-terminal transmembrane helix HupS, HupC and RdhB. The same three interactions while the rest of the domain is located outside the cell. as described above were observed when we performed In the model, the FMN binding site is located close to the hierarchical cluster analysis on the 125 most abun- the membrane and oriented towards it so that the FMN dant proteins in a gel lane (Figure S6) or when calcu- aromatic rings would be immersed into the membrane lated for all proteins detected (more than 1000 proteins, (Figure 4A). The two CX3CP motifs are predicted to be data not shown). located in the membrane-integrated C-terminal part of RdhC and are arranged three-dimensionally in such a way that they point towards each other giving space to Structural features of RdhC chelate a metal cluster in the middle of the membrane plane (Figure 4B). In the structural model, this metal The principal architecture of the four full RdhC copies in chelating site is water accessible from the cytoplasmic strain 8M appears to include a single long transmem- side and 16.4 Å distant from the FMN molecule. In the brane helix at the N-terminus, followed by a long stretch middle between the metal chelating site and the FMN of approximately 150 exoplasmic amino acids and molecule, a highly conserved glutamate residue is 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License PROTEIN INTERACTIONS IN Dehalobacter SP. STRAIN 8M 11 located, possibly facilitating electron and/or proton con- Identification of the RdhA involved in TCM duction (Figure 4A). For metal cluster prediction, the and 1,1,2-TCA dechlorination COFACTOR server predicted no cluster formation with a probability above 1%. In contrast, MetalPredator sug- TmrA (RdhA 40029) was the first and second most gested the formation of a [4Fe-4S] cluster with an abundant protein in cultures grown with 1,1,2-TCA and e-value of 3.9  10 15. TCM, respectively. Interestingly, TmrA was the only reductive dehalogenase expressed in all duplicate cul- tures grown with either TCM or 1,1,2-TCA in the prote- DISCUSSION ome profile (Figure 1). This protein was also the most abundant RdhA in all BNE gel lanes (Table S6) and Comparison of the genome of had its maximum relative protein abundance in the Dehalobacter sp. 8M with other BNE slices with the highest dechlorinating activity Dehalobacter genomes towards TCM and 1,1,2-TCA (Figures 2 and S4). Clus- ter analyses supported this close association of TmrA Overall, the genome of strain 8M shares similar func- with reductive dehalogenase activity. Therefore, our tional gene contents with the other six available Deha- data suggest that TmrA was responsible for the reduc- lobacter genomes. Strain 8M possesses a complete tive dechlorination of both TCM and 1,1,2-TCA. TmrA set of genes encoding the Wood-Ljungdahl pathway, displays high amino acid identity with TmrA (97%, which is encoded in all Dehalobacter strains except for EQB22800.1), CfrA (95%, WP_242827270.1), ThmA strain EB1. It has been hypothesized that the Wood- (95%, ANI21407.1), DcrA (94%, WP_242824980.1) Ljungdahl pathway might have another role than CO2- and CtrA (93%, AGO27983.1) from Dehalobacter fixation in Dehalobacter strains because strain TeCB1 sp. strains UNSWDHB, CF, THM1 and DCA and Desul- was isolated and grown in a bicarbonate-free medium fitobacterium sp. strain PR, respectively (Table S3). (Alfan-Guzman et al., 2017b) and no known Dehalo- Collectively, all these reductive dehalogenases share bacter isolate was described to grow autotrophically their ability to dechlorinate TCM, but the RdhA proteins with H2 and CO2. The Wood-Ljungdahl pathway in of strains 8M and UNSWDHB differ from the others by Dehalobacter species might operate in the oxidative their inability to completely (100% of the substrate) direction. In this direction, acetate is enzymatically transform 1,1-dichloroethane and 1,1,1-trichloroethane cleaved, resulting in the production of a methyl group (Ding et al., 2014; Soder-Walz et al., 2022; Tang & and CO. The methyl group is subsequently transferred Edwards, 2013; Wong et al., 2016). Testing of TCM to tetrahydrofolate, while the electrons generated from and 1,1,2-TCA transformation in enzymatic assays with the oxidation of CO to CO2 by carbon monoxide dehy- cell suspensions of strain 8M showed similar dechlori- drogenase can be channelled into a reduction pool nating activity for both compounds, regardless of the (Alfan-Guzman et al., 2017b). The corrinoid electron acceptor used to grow strain 8M (Figure S1), biosynthesis-associated genes are fully encoded, which is in agreement with the dechlorination activity of which is a common feature among Dehalobacter the TCM-dechlorinating RdhA TmrA towards both TCM strains except for strains PER-K23 and E1 (Rupakula and 1,1,2-TCA (Jugder et al., 2017). et al., 2013). However, the ability of strain 8M to grow Although TCM-dechlorinating RdhA proteins share without exogenously supplied cobalamin needs experi- high amino acid sequence similarity, previous studies mental confirmation because the culture medium used have shown differences in both carbon and chlorine contained cyanocobalamin, and the accompanying isotope fractionation during TCM dechlorination by dif- bacteria might also provide corrinoids. Menaquinone ferent Dehalobacter strains and their associated RdhA, seems to be the only quinone-type electron carrier that including strain 8M, strain CF (CfrA), UNSWDHB could be synthesized by strain 8M, and it could shuttle (TmrA) and a RdhA identified in the commercial enrich- electrons between [NiFe]-hydrogenase and RdhA as ment culture KB-1 Plus CF (Heckel et al., 2019; Phillips proposed previously for Deh. restrictus PER-K23 et al., 2022; Soder-Walz et al., 2022; Table S7). This (Schumacher & Holliger, 1996). Genome analyses fur- difference in the isotopic pattern has led to questions ther suggest that Dehalobacter synthesize haeme via about the reaction mechanisms of very similar RdhA the sirohaeme-dependent pathway, as described for proteins. In a recent study, Phillips et al. (2022) hypoth- several sulfate-reducing bacteria and archaea (Bali esized that these differences in carbon and chlorine et al., 2011; Kühner et al., 2014; Layer, 2021). This cor- isotope effects are related to amino acid differences in relates with the presence of cytochrome-encoding the active sites of CfrA compared to TmrA and the genes, for example, hupC. In line with other available KB-1 Plus CF RdhA. They proposed that the low C and Dehalobacter genomes, the biosynthesis pathway for Cl isotope effects observed in cultures containing TmrA biotin is missing in strain 8M, suggesting that biotin is and the KB-1 Plus CF RdhA are associated with the required in the medium for growth. similarity in their amino acid residues Phe80, Phe388 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 12 SODER-WALZ ET AL. F I G U R E 5 Alignment of the RdhA region has been suggested to determine isotopic fractionation of TCM in different OHRBs (Phillips et al., 2022). Highlighted residues indicate differences from the consensus sequence. Alignments were prepared using COBALT software (Papadopoulos & Agarwala, 2007). The enzymes responsible for the TCM transformation are indicated in parenthesis. Dfb, Desulfitobacterium; Dhb, Dehalobacter; OHRB, obligate organohalide-respiring bacteria; TCM, trichloromethane. and Trp391. The influence of these three amino acids activity was in the same slice showing that reductive on the isotopic fractionation pattern observed on C and dehalogenases are not occurring as monomers. A Cl isotope effects is further supported by the identifica- dimerization of TmrA would not account for the tion of TmrA and its amino acid sequence in this study observed mass of 150 kDa. A heterotetramer with since TmrA displayed the same amino acid residues at Rdh(AB)2 stoichiometry would have a mass of 112 kDa positions 80, 388 and 391 as the RdhA proteins TmrA which might match the observed size in the gel, espe- and KB-1 Plus CF (Figure 5) and had the same low car- cially when considering that the gel positions in native bon and chloride enrichment factors. gels can be shifted due to structural characteristics. Besides TmrA, strain 8M expressed other RdhA However, the migration analyses in the native gels proteins when grown with a single-electron acceptor. reproducibly show that RdhB co-migrates with HupC, RdhA 260084 was not detected in cultures grown with that is, another membrane-integrated protein, indicating 1,1,2-TCA, but it was produced abundantly with TCM that RdhB was not associated anymore with RdhA pro- and reached its maximum relative abundance in slices teins in the gel and therefore a heterotetramer of with the highest dechlorinating activity against TCM Rdh(AB)2 can be ruled out as the protein complex in and 1,1,2-TCA (Figures 2A and S4A). In the cultures the observed active band. Another explanation for the grown with TCM, RdhA 260084 clustered with the occurrence of the RdhA protein in the 150 kDa region dehalogenating activity of TCM in both replicates and is that RdhA is part of a larger respiratory complex. this interaction is even more evident when we included Such a respiratory multiprotein complex has been more proteins in the analysis (Figure S6). Despite these found in Dehalococcoides mccartyi strain CBDB1 consistent co-migration patterns, the involvement of (Kublik et al., 2016) and Dehalogenimonas alkeni- RdhA 260084 in TCM dechlorination is unlikely gignens strain BRE15M (Trueba-Santiso et al., 2021). because the abundance of RdhA 260084 was two In these complexes, however, a complex iron–sulfur– orders of magnitude lower than that of TmrA, and RdhA molybdoprotein (CISM) acts as a scaffold or relay for 260084 is almost identical (99.78% amino acid electron transport between hydrogenase and reductive sequence identity) to the RdhA encoded in the genome dehalogenase, obviating quinone involvement. In con- of Deh. restrictus strain PER-K23 strain that cannot trast to Dehalobacter strains, neither Dehalococcoides grow with TCM. Interestingly, the membrane anchor nor Dehalogenimonas encodes quinone biosynthesis. RdhB 260083 was solely detected in cells cultured with The main question at this point is therefore, if the Deha- TCM as observed for its syntenic RdhA 260084. This lobacter RdhA proteins are part of a larger protein also suggests an interaction between the RdhA and its complex and, if so, which specific proteins they associated RdhB protein. interact with. Cluster analyses of the distribution of relative pro- tein abundances across replicate gel lanes of cultures The architecture of the respiratory chain in grown with TCM and 1,1,2-TCA suggest the existence Dehalobacter sp. strain 8M of three protein–protein interactions of respiratory pro- teins (Figure 6A). The predicted amino acid sequence of TmrA without the TAT-leader peptide has a calculated molecular mass of 44.4 kDa and the putative membrane anchor Interaction I RdhB 40030 of 11.4 kDa. However, gel slices with TmrA were located between 100 and 200 kDa with the HupL and HupS had identical migration patterns, indi- highest abundance at around 150 kDa. The maximum cating firm interaction (Figures 3, S5 and S6), as has 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License PROTEIN INTERACTIONS IN Dehalobacter SP. STRAIN 8M 13 F I G U R E 6 Scheme of protein interactions in Dehalobacter sp. strain 8M. (A) Overview of proteins and detected protein–protein interactions related to organohalide respiration. Cluster analysis indicated that RdhA 40029 (TmrA) weakly interacts with HupLS. The colour code is maintained through the figures and graphs. (B) Proposed model of an organohalide respiration complex combining the detected protein–protein interactions with the assumed interactions of HupC with HupS and of RdhA with RdhB, putatively disrupted by the detergent treatment. The roles of RdhC and quinones are elusive. HupC, b-type cytochrome; HupS, hydrogenase small subunit; HupL, hydrogenase large subunit; RdhA, reductive dehalogenase; RdhB, reductive dehalogenase anchor protein; RdhC, possibly a mixture between RdhC 40031 and RdhC 260096. also been described for Deh. mccartyi strain CBDB1 size of 115 kDa, almost in the range observed in the and Deh. alkenigignens strain BRE15M (Seidel BNE gels, but as with the RdhA, cluster analysis clearly et al., 2018; Trueba-Santiso et al., 2021). In our study, shows the localization of HupC in other regions of both proteins had the maximum relative protein abun- the gel. dance in slices 5 and 6 (127–155 kDa for TCM) and 4 and 5 (117–149 kDa for 1,1,2-TCA). The calculated size of HupL (57.1 kDa) and HupS (34.1 kDa with Interaction II removed TAT-leader peptide) together is 91.2 kDa, which is lower than observed in the gel. Together with The relative abundance of the two membrane proteins, the b-type cytochrome HupC (23.6 kDa, four trans- RdhB 40030 (11.4 kDa) and HupC 230005 (23.6 kDa), membrane helices), the Hup-complex would have a did not exhibit a distinct maximum of relative 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 14 SODER-WALZ ET AL. abundance but were distributed across the protein as the use of detergents can weaken protein associa- slices up to approximately 200 kDa. Only cluster analy- tions and negatively charged Coomassie dye can dis- sis revealed that the migration pattern of HupC and rupt fragile protein complexes. Additional techniques RdhB was very similar in all the different analyses (dif- like immunoprecipitations or co-migration under differ- ferent cultures used or different numbers of proteins ent conditions, such as non-sized-based chromatogra- included in the analyses), suggesting that they physi- phy, can confirm protein–protein interactions (Eubel cally interact with each other in the membrane through et al., 2005). their hydrophobic regions. The detected protein–protein interactions constrain a model for the respiratory architecture in Dehalobacter strains. All detected proteins together would give a Interaction III complex of 221.1 kDa (HupLS: 91.2 kDa; HupC and RdhB: 34.9 kDa, RdhC: 50.5 kDa; RdhA: 44.4 kDa), The RdhC proteins 40031 (50.5 kDa) and 260096 which is larger than the observed molecular mass of (46.6 kDa) showed identical migration patterns with a 100–200 kDa, where the highest dehalogenation activ- maximum of relative protein abundance at a molecular ity was determined. Again, it must be kept in mind that mass of approximately 175 kDa, which is much more the size estimation is not precise in BNE gels (Seidel than the predicted molecular mass of the single pro- et al., 2018). A possible explanation is that the teins. The dechlorinating activity was also observed in membrane-integrated proteins HupC and RdhB were the slice at 175 kDa, and in addition to both RdhC, the removed by detergent treatment resulting in a theoreti- same slices showed the presence of RdhA, RdhB, cal mass of 186.1 kDa (HupLS, RdhA, RdhC) or HupL, HupS and HupC. This suggests that RdhC is not 135.6 kDa (HupLS, RdhA). A removal of HupC and completely independent from the other proteins of the RdhB from the protein complex would probably also respiratory chain, but we cannot conclude on specific weaken the overall integrity of a complex assuming interactions. We can also not differentiate if the two in vivo interactions of HupLS with HupC and RdhA with RdhC proteins interact with each other or if they interact RdhB (Bommer et al., 2014). separately with other proteins. RdhB 40030 and RdhC 40031 were both detected by mass spectrometry in our studies. This is remarkable because these subunits fre- The role of RdhC quently evade identification due to their small size, their low number of trypsin digestion sites, the high m/z The overall structure formed by the single protein values of tryptic peptides and the hydrophobic charac- RdhC is similar to structures formed by three protein teristics of the peptides (Alfan-Guzman et al., 2017a; subunits in other organisms, that is, Nqr (NADH: qui- Cimmino et al., 2022). none oxidoreductase) of Vibrio cholerae (Steuber Our data also suggest weak interaction of RdhA et al., 2014) and Rnf (ferredoxin: NAD oxidoreductase) with the HupLS dimer. This can be tentatively judged of Clostridium tetanomorphum (Vitt et al., 2022). Both by evaluating the abundance distribution of RdhA and complexes participate in membrane potential coupled HupLS along the gel lanes in Figures 2 and S4 where redox reactions (Reyes-Prieto et al., 2014). The exo- the maxima consistently appear in the same or adja- plasmic FMN-binding domain of RdhC is homologous cent gel slices. Cluster analysis with the 60 most abun- to NqrC and RnfG, whereas the architecture of the dant proteins does not represent this well (Figure 3). transmembrane helices in the C-terminal membrane- Only when more proteins were included in the cluster integrated half of RdhC is strikingly similar to the analysis (Figure S6, including the 125 most abundant membrane-integrated subunit twins NqrDE and RnfEA. proteins) clustering of the RdhA proteins 40029 (TmrA) These subunit twins of NqrDE and RnfEA form sym- and 260094 with the HupLS dimer was evident in all metrical dimers within the membrane, where the mono- four analyses (duplicate analysis with cells grown on mers are rotated by 180 . Four central helices of 1,1,2-TCA or TCM). Indeed, the interaction of RdhA NqrDE or RnfEA form a core structure, where a [2Fe- with HupLS (together 135.6 kDa) correlates well with 2S] cluster is coordinated by four cysteine residues the observed sizes in native gels. The HupLS dimer (Fritz, 2023; Kishikawa et al., 2022; L. Zhang & might partially dissociate from this interaction or interact Einsle, 2022), and not a mononuclear iron atom as has with other proteins explaining the loose clustering. BNE been previously described (Steuber et al., 2014; Vitt is a valuable approach for analysing protein–protein et al., 2022). Two cysteine residues originate from the interactions, but it has limitations that can lead to misin- core helices of one subunit each of NqrDE or RnfEA. terpretation of data. Co-migration alone does not imply The [2Fe-2S] cluster is located in the middle of the physical interaction, as similar-sized complexes can membrane plane in the Nqr and Rnf complexes, possi- appear in the same protein band. However, physical bly enabling the conduction of electrons through the interaction can explain co-migration. Weak or transient membrane (Kishikawa et al., 2022; Steuber protein–protein interactions pose challenges in BNE, et al., 2014; Vitt et al., 2022). The C-terminal part of 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License PROTEIN INTERACTIONS IN Dehalobacter SP. STRAIN 8M 15 RdhC has the same architecture and four cysteine resi- (equal); writing—original draft (supporting); writing— dues within one polypeptide chain forming a cavity for review and editing (equal). Teresa Vicent: Resources putative metal binding. The strong structural similarity (equal); validation (equal); writing—review and editing suggests that indeed a [2Fe-2S] cluster is bound to (equal). Lorenz Adrian: Conceptualization (support- RdhC and not a [4Fe-4S] cluster as predicted by Metal- ing); data curation (supporting); resources (equal); Predator. Experimental evidence is necessary to test supervision (supporting); writing—original draft (sup- this hypothesis. Together, the observed similarities to porting); writing—review and editing (equal). Ernesto Nqr and Rnf complexes suggest that RdhC could form Marco-Urrea: Conceptualization (lead); data curation an electron path through the membrane with a central (equal); formal analysis (equal); funding acquisition [2Fe-2S] cluster in the membrane plane, electrically (lead); project administration (lead); supervision (lead); connected with an exoplasmic FMN unit. The participa- writing—original draft (supporting); writing–review and tion of RdhC in organohalide respiration has been editing (equal). hypothesized before based on its co-location in the reductive dehalogenase operon and the presence of an AC KN OW LED GE M EN T FMN-binding motif (Buttet et al., 2018; Cimmino Protein mass spectrometry was performed at the Centre et al., 2022). The integration of this protein in the orga- for Chemical Microscopy (ProVIS) at the Helmholtz Cen- nohalide respiration chain, however, is not clear. One tre for Environmental Research—UFZ. Open Access possibility for a biochemical role of RdhC is the cou- funding enabled and organized by Projekt DEAL. pling of electron flow from haeme groups of HupC to the reductive dehalogenase to proton or sodium trans- F UN DI NG IN F O R MA T I ON port across the membrane. Then RdhC would have a This work was funded by MCIN/AEI/10. similar role as OmeAB might have in the dehalogenase 13039/501100011033 (grant ref. PID2019-103989RB- complex of Dehalococcoidia (Kublik et al., 2016). Such I00) and partially supported by the Generalitat de Catalunya involvement of RdhC would obviate the involvement of (Consolidated Research Group 2021-SGR-01008). quinones in electron transport in Dehalobacter and resolve the paradox of how electrons can be CONF LIC T OF IN TE RES T ST ATE M EN T transferred to the low-potential corrinoid cluster via a The authors declare no conflicts of interest. high-potential quinone pool. Also, this may give an explanation of why some Dehalobacter strains are DA T A A V A IL A B IL I T Y S T AT E M E N T capable of reductive dehalogenation of chloroben- Data supporting the findings are available in the supple- zenes, chloroanilines, chlorophenols and chlorodioxins mentary material. The genome sequence and annota- otherwise only described for members of the Dehalo- tions are publicly available on the MicroScope platform: coccoidia (Alfan-Guzman et al., 2017a; Nelson https://mage.genoscope.cns.fr/microscope/mage/viewer. et al., 2014; Wang et al., 2014; S. Zhang, Wondrousch, php?O_id=14849. The genome sequence was also et al., 2017). However, both our complexome data and submitted to NCBI under the BioProject PRJNA922222: stoichiometric analysis of RdhC versus RdhAB https://www.ncbi.nlm.nih.gov/bioproject/PRJNA922222. (Cimmino et al., 2022) suggest that RdhC is not directly involved, although we cannot rule out that the complex- OR CI D ome approach misses important interactions. More- Jesica M. Soder-Walz https://orcid.org/0000-0002- over, previous biochemical experiments have indicated 5253-9898 the involvement of quinones in the reductive dehalo- Kenneth Wasmund https://orcid.org/0000-0001- genation of perchloroethene by Deh. restrictus 6706-7291 (Schumacher & Holliger, 1996). A resolution to this Darja Deobald https://orcid.org/0000-0003-0882- conundrum could be the existence of different electron 5499 paths for different halogenated electron acceptors in Teresa Vicent https://orcid.org/0000-0002-6547-7358 Dehalobacter strains. At present, the specific role of Lorenz Adrian https://orcid.org/0000-0001-8205- RdhC in this process remains elusive. 0842 Ernest Marco-Urrea https://orcid.org/0000-0002- A UT H O R C ON T R IB UT IO NS 8033-6553 Jesica M. Soder-Walz: Conceptualization (support- ing); data curation (equal); formal analysis (equal); REF ER ENC ES investigation (lead); methodology (lead); software Adrian, L., Hansen, S.K., Fung, J.M., Görisch, H. & Zinder, S.H. (equal); writing—original draft (lead); writing—review (2007) Growth of Dehalococcoides strains with chlorophenols as electron acceptors. Environmental Science & Technology, 41, and editing (equal). Kenneth Wasmund: Data curation 2318–2323. (equal); formal analysis (equal); methodology (equal); Agency for Toxic Substances and Disease Registry (ATSDR). (2022) software (equal); writing—review and editing (equal). Priority list of hazardous substances. Available from: https:// Darja Deobald: Formal analysis (equal); software www.atsdr.cdc.gov/spl/#2022spl [Accessed 18 January 2023]. 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 16 SODER-WALZ ET AL. Alfan-Guzman, R., Ertan, H., Manefield, M. & Lee, M. (2017a) Isola- common in Dehalococcoides. Applied and Environmental Micro- tion and characterization of Dehalobacter sp. strain TeCB1 biology, 70, 5290–5297. including identification of tcbA: a novel tetra- and trichloroben- Jugder, B.-E., Bohl, S., Lebhar, H., Healey, R.D., Manefield, M., zene reductive dehalogenase. Frontiers in Microbiology, 8, 558. Marquis, C.P. et al. (2017) A bacterial chloroform reductive Alfan-Guzman, R., Ertan, H., Manefield, M. & Lee, M. (2017b) dehalogenase: purification and biochemical characterization. Genome sequence of Dehalobacter sp. strain TeCB1, able to Microbial Biotechnology, 10, 1640–1648. respire chlorinated benzenes. Genome Announcements, 5. Jugder, B.-E., Ertan, H., Bohl, S., Lee, M., Marquis, C.P. & https://doi.org/10.1128/genomea.01681-16 Manefield, M. (2016) Organohalide respiring bacteria and reduc- Aziz, R.K., Bartels, D., Best, A.A., DeJongh, M., Disz, T., tive dehalogenases: key tools in organohalide bioremediation. Edwards, R.A. et al. (2008) The RAST server: rapid annotations Frontiers in Microbiology, 7, 249. using subsystems technology. BMC Genomics, 9, 75. Jugder, B.-E., Ertan, H., Lee, M., Manefield, M. & Marquis, C.P. Bali, S., Lawrence, A.D., Lobo, S.A., Saraiva, L.M., Golding, B.T., (2015) Reductive dehalogenases come of age in biological Palmer, D.J. et al. (2011) Molecular hijacking of siroheme for the destruction of organohalides. Trends in Biotechnology, 33, synthesis of heme and d1 heme. Proceedings of the National 595–610. Academy of Sciences, 108, 18260–18265. Kishikawa, J.-I., Ishikawa, M., Masuya, T., Murai, M., Kitazumi, Y., Bankevich, A., Nurk, S., Antipov, D., Gurevich, A.A., Dvorkin, M., Butler, N.L. et al. (2022) Cryo-EM structures of Na+-pumping Kulikov, A.S. et al. (2012) SPAdes: a new genome assembly NADH-ubiquinone oxidoreductase from Vibrio cholerae. Nature algorithm and its applications to single-cell sequencing. Journal Communications, 13, 4082. of Computational Biology, 19, 455–477. Kräutler, B., Fieber, W., Ostermann, S., Fasching, M., Ongania, K.-H.,  zquez-Pallí, N., Rosell, M., Varias, J., Bosch, M., Soler, A., Bla Gruber, K. et al. (2003) The cofactor of tetrachloroethene reduc- Vicent, T. et al. (2019) Integrative isotopic and molecular tive dehalogenase of Dehalospirillum multivorans is norpseudo- approach for the diagnosis and implementation of an efficient in- B12, a new type of a natural corrinoid. Helvetica Chimica Acta, situ enhanced biological reductive dechlorination of chlorinated 86, 3698–3716. ethenes. Water Research, 167, 115106. Kublik, A., Deobald, D., Hartwig, S., Schiffmann, C.L., Andrades, A., Bommer, M., Kunze, C., Fesseler, J., Schubert, T., Diekert, G. & von Bergen, M. et al. (2016) Identification of a multi-protein Dobbek, H. (2014) Structural basis for organohalide respiration. reductive dehalogenase complex in Dehalococcoides mccartyi Science, 346, 455–458. strain CBDB1 suggests a protein-dependent respiratory electron Buttet, G.F., Willemin, M.S., Hamelin, R., Rupakula, A. & Maillard, J. transport chain obviating quinone involvement. Environmental (2018) The membrane-bound c subunit of reductive dehalo- Microbiology, 18, 3044–3056. genases: topology analysis and reconstitution of the FMN- Kühner, M., Haufschildt, K., Neumann, A., Storbeck, S., Streif, J. & binding domain of PceC. Frontiers in Microbiology, 9, 755. Layer, G. (2014) The alternative route to heme in the methano- Cimmino, L., Schmid, A.W., Holliger, C. & Maillard, J. (2022) Stoichi- genic archaeon Methanosarcina barkeri. Archaea, 2014, ometry of the gene products from the tetrachloroethene reduc- 327637. tive dehalogenase operon pceABCT. Frontiers in Microbiology, Layer, G. (2021) Heme biosynthesis in prokaryotes. Biochimica et Bio- 13, 838026. physica Acta (BBA)—Molecular Cell Research, 1868, 118861. Ding, C., Zhao, S. & He, J. (2014) A Desulfitobacterium sp. strain PR Maillard, J. & Holliger, C. (2016) The genus Dehalobacter. In: reductively dechlorinates both 1,1,1-trichloroethane and chloro- Adrian, L. & Löffler, F.E. (Eds.) Organohalide-respiring bacteria. form. Environmental Microbiology, 16, 3387–3397. Berlin, Heidelberg: Springer Berlin Heidelberg, pp. 153–171. Duret, A., Holliger, C. & Maillard, J. (2012) The physiological oppor- Maillard, J., Genevaux, P. & Holliger, C. (2011) Redundancy and tunism of Desulfitobacterium hafniense strain TCE1 towards specificity of multiple trigger factor chaperones in Desulfitobac- organohalide respiration with tetrachloroethene. Applied and teria. Microbiology, 157, 2410–2421. Environmental Microbiology, 78, 6121–6127. Mirdita, M., Schütze, K., Moriwaki, Y., Heo, L., Ovchinnikov, S. & Edgar, R. (2004) MUSCLE: multiple sequence alignment with high Steinegger, M. (2022) ColabFold: making protein folding acces- accuracy and high throughput. Nucleic Acids Research, 32, sible to all. Nature Methods, 19, 679–682. 1792–1797. Nelson, J.L., Jiang, J. & Zinder, S.H. (2014) Dehalogenation of chlo- Eubel, H., Braun, H.-P. & Millar, A. (2005) Blue-native PAGE in robenzenes, dichlorotoluenes, and tetrachloroethene by three plants: a tool in analysis of protein-protein interactions. Plant Dehalobacter spp. Environmental Science & Technology, 48, Methods, 1, 11. 3776–3782. Fritz, G. (2023) A [2Fe-2S] cluster is bound between the NqrD and Padilla-Crespo, E., Yan, J., Swift, C., Wagner, D.D., Chourey, K., NqrE subunits of the NQR complex of Vibrio cholerae. SPP1927 Hettich, R.L. et al. (2014) Identification and environmental distri- Meeting, personal communication on 04.05.2023. Universität bution of dcpA, which encodes the reductive dehalogenase cat- Hohenheim. alyzing the dichloroelimination of 1,2-dichloropropane to Geer, L.Y., Domrachev, M., Lipman, D.J. & Bryant, S.H. (2002) propene in organohalide-respiring Chloroflexi. Applied and Envi- CDART: protein homology by domain architecture. Genome ronmental Microbiology, 80, 808–818. Research, 12, 1619–1623. Papadopoulos, J.S. & Agarwala, R. (2007) COBALT: constraint- Heckel, B., Phillips, E., Edwards, E., Sherwood Lollar, B., Elsner, M., based alignment tool for multiple protein sequences. Bioinfor- Manefield, M.J. et al. (2019) Reductive dehalogenation of tri- matics, 23, 1073–1079. chloromethane by two different Dehalobacter restrictus strains Parks, D.H., Imelfort, M., Skennerton, C.T., Hugenholtz, P. & reveal opposing dual element isotope effects. Environmental Tyson, G.W. (2015) CheckM: assessing the quality of microbial Science & Technology, 53, 2332–2343. genomes recovered from isolates, single cells, and metagen- Hug, L.A., Maphosa, F., Leys, D., Löffler, F.E., Smidt, H., omes. Genome Research, 25, 1043–1055. Edwards, E.A. et al. (2013) Overview of organohalide-respiring Phillips, E., Bulka, O., Picott, K., Kümmel, S., Edwards, E.A., bacteria and a proposal for a classification system for reductive Nijenhuis, I. et al. (2022) Investigation of active site amino acid dehalogenases. Philosophical Transactions of the Royal Society influence on carbon and chlorine isotope fractionation during reduc- B: Biological Sciences, 368, 20120322. tive dechlorination. FEMS Microbiology Ecology, 98, fiac072. Hölscher, T., Krajmalnik-Brown, R., Ritalahti, K., von Reyes-Prieto, A., Barquera, B. & Jua  rez, O. (2014) Origin and evolu- Wintzingerode, F., Görisch, H., Löffler, F. et al. (2004) Multiple tion of the sodium-pumping NADH: ubiquinone oxidoreductase. nonidentical reductive-dehalogenase-homologous genes are PLoS One, 9, e96696. 14622920, 0, Downloaded from https://ami-journals.onlinelibrary.wiley.com/doi/10.1111/1462-2920.16464 by Readcube (Labtiva Inc.), Wiley Online Library on [16/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License PROTEIN INTERACTIONS IN Dehalobacter SP. STRAIN 8M 17 Richter, M., Rossello  -Mo  ra, R., Oliver Glöckner, F. & Peplies, J. Vallenet, D., Calteau, A., Dubois, M., Amours, P., Bazin, A., (2016) JSpeciesWS: a web server for prokaryotic species cir- Beuvin, M. et al. (2020) MicroScope: an integrated platform for cumscription based on pairwise genome comparison. Bioinfor- the annotation and exploration of microbial gene functions matics, 32, 929–931. through genomic, pangenomic and metabolic comparative anal- Rothery, R.A., Workun, G.J. & Weiner, J.H. (2008) The prokaryotic ysis. Nucleic Acids Research, 48, D579–D589. complex iron–sulfur molybdoenzyme family. Biochimica et Bio- Vitt, S., Prinz, S., Eisinger, M., Ermler, U. & Buckel, W. (2022) Purifi- physica Acta—Biomembranes, 1778, 1897–1929. cation and structural characterization of the Na+-translocating Rupakula, A., Kruse, T., Boeren, S., Holliger, C., Smidt, H. & ferredoxin: NAD+ reductase (Rnf) complex of Clostridium teta- Maillard, J. (2013) The restricted metabolism of the obligate nomorphum. Nature Communications, 13, 6315. organohalide respiring bacterium Dehalobacter restrictus: les- Wang, S., Qiu, L., Liu, X., Xu, G., Siegert, M., Lu, Q. et al. (2018) sons from tiered functional genomics. Philosophical Transac- Electron transport chains in organohalide-respiring bacteria and tions of the Royal Society B: Biological Sciences, 368, bioremediation implications. Biotechnology Advances, 36, 20120325. 1194–1206. Sayers, E.W., Bolton, E.E., Brister, J.R., Canese, K., Chan, J., Wang, S., Zhang, W., Yang, K.-L. & He, J. (2014) Isolation and char- Comeau, D.C. et al. (2022) Database resources of the National acterization of a novel Dehalobacter species strain TCP1 that Center for Biotechnology Information. Nucleic Acids Research, reductively dechlorinates 2,4,6-trichlorophenol. Biodegradation, 50, D20–D26. 25, 313–323. Schrödinger, L. & DeLano, W. (2020) PyMOL. Available from: http:// Wong, Y.K., Holland, S.I., Ertan, H., Manefield, M. & Lee, M. (2016) wwwpymolorg/pymol Isolation and characterization of Dehalobacter sp. strain Schubert, M., Lindgreen, S. & Orlando, L. (2016) AdapterRemoval v2: UNSWDHB capable of chloroform and chlorinated ethane respi- rapid adapter trimming, identification, and read merging. BioMed ration. Environmental Microbiology, 18, 3092–3105. Central Research Notes, 9, 1–7. Yang, M.I., Previdsa, M., Edwards, E.A. & Sleep, B.E. (2020) Two Schumacher, W. & Holliger, C. (1996) The proton/electron ratio of the distinct Dehalobacter strains sequentially dechlorinate menaquinone-dependent electron tansport from dihydrogen to 1,1,1-trichloroethane and 1,1-dichloroethane at a field site trea- tetrachloroethene in ‘Dehalobacter restrictus’. Journal of Bacte- ted with granular zero valent iron and guar gum. Water riology, 178, 2328–2333. Research, 186, 116310. Seidel, K., Kühne, J. & Adrian, L. (2018) The complexome of Dehalo- Zhang, C., Freddolino, P.L. & Zhang, Y. (2017) COFACTOR: coccoides mccartyi reveals its organohalide respiration-complex improved protein function prediction by combining structure, is modular. Frontiers in Microbiology, 9, 1130. sequence and protein–protein interaction information. Nucleic Soder-Walz, J.M., Torrento , C., Algora, C., Wasmund, K., Cortés, P., Acids Research, 45, W291–W299. Soler, A. et al. (2022) Trichloromethane dechlorination by a Zhang, L. & Einsle, O. (2022) Architecture of the NADH:ferredoxin novel Dehalobacter sp. strain 8M reveals a third contrasting C oxidoreductase RNF that drives biological nitrogen fixation. and Cl isotope fractionation pattern within this genus. Science of bioRxiv 2022.2007.2008.499327. the Total Environment, 813, 152659. Zhang, S., Wondrousch, D., Cooper, M., Zinder, S.H., Steuber, J., Vohl, G., Casutt, M.S., Vorburger, T., Diederichs, K. & Schüürmann, G. & Adrian, L. (2017) Anaerobic dehalogenation Fritz, G. (2014) Structure of the V. cholerae Na+-pumping of chloroanilines by Dehalococcoides mccartyi strain CBDB1 NADH:quinone oxidoreductase. Nature, 516, 62–67. and Dehalobacter strain 14DCB1 via different pathways as Tamura, K., Stecher, G. & Kumar, S. (2021) MEGA11: molecular evo- related to molecular electronic structure. Environmental Sci- lutionary genetics analysis version 11. Molecular Biology and ence & Technology, 51, 3714–3724. Evolution, 38, 3022–3027. Zhao, S., Rogers, M.J. & He, J. (2017) Microbial reductive dehalogena- Tang, S. & Edwards, E.A. (2013) Identification of Dehalobacter reduc- tion of trihalomethanes by a Dehalobacter-containing co-culture. tive dehalogenases that catalyse dechlorination of chloroform, Applied Microbiology and Biotechnology, 101, 5481–5492. 1,1,1-trichloroethane and 1,1-dichloroethane. Philosophical Zimmermann, L., Stephens, A., Nam, S.-Z., Rau, D., Kübler, J., Transactions of the Royal Society B: Biological Sciences, 368, Lozajic, M. et al. (2018) A completely reimplemented MPI bioin- 20120318. formatics toolkit with a new HHpred server at its core. Journal of Teufel, F., Almagro Armenteros, J.J., Johansen, A.R., Gíslason, M.H., Molecular Biology, 430, 2237–2243. Pihl, S.I., Tsirigos, K.D. et al. (2022) SignalP 6.0 predicts all five types of signal peptides using protein language models. Nature S UP P O RT IN G I NF O R M AT IO N Biotechnology, 40, 1023–1025. Trifinopoulos, J., Nguyen, L.-T., von Haeseler, A. & Minh, B.Q. (2016) Additional supporting information can be found online W-IQ-TREE: a fast online phylogenetic tool for maximum likeli- in the Supporting Information section at the end of this hood analysis. Nucleic Acids Research, 44, W232–W235. article. Trueba-Santiso, A., Parladé, E., Rosell, M., Lliros, M., Mortan, S.H., Martínez-Alonso, M. et al. (2017) Molecular and carbon isotopic characterization of an anaerobic stable enrichment culture con- How to cite this article: Soder-Walz, J.M., taining Dehalobacterium sp. during dichloromethane fermenta- Wasmund, K., Deobald, D., Vicent, T., Adrian, L. tion. Science of the Total Environment, 581, 640–648. Trueba-Santiso, A., Wasmund, K., Soder-Walz, J.M., Marco- & Marco-Urrea, E. (2023) Respiratory protein Urrea, E. & Adrian, L. (2021) Genome sequence, proteome pro- interactions in Dehalobacter sp. strain 8M file, and identification of a multiprotein reductive dehalogenase revealed through genomic and native proteomic complex in Dehalogenimonas alkenigignens strain BRE15M. analyses. Environmental Microbiology, 1–17. Journal of Proteome Research, 20, 613–623. Available from: https://doi.org/10.1111/1462- Valasatava, Y., Rosato, A., Banci, L. & Andreini, C. (2016) MetalPre- dator: a web server to predict iron–sulfur cluster binding pro- 2920.16464 teomes. Bioinformatics, 32, 2850–2852.