ABSTRACT
Methyl tert-butyl ether (MTBE) has been recognized as a groundwater contaminant due to its widespread distribution and potential threat to human health. The limited understanding of the enzymes catalyzing MTBE degradation restricts their application in MTBE bioremediation. In this study, an MTBE-degrading soluble di-iron monooxygenase that clusters phylogenetically with a known propane monooxygenase (PRM) encoded by the prmABCD gene cluster was identified and functionally characterized, revealing their role in MTBE metabolism by Mycobacterium vaccae JOB5. Transcriptome analysis demonstrated that the expression of prmABCD was upregulated when JOB5 was induced by MTBE. Escherichia coli Rosetta heterologously expressing prmABCD from JOB5 could transform MTBE, indicating that the PRM of JOB5 is capable of the initial degradation of MTBE. The loss of the gene encoding the oxygenase α-subunit or β-subunit, the coupling protein, or the reductase disrupted MTBE transformation by the recombinant E. coli Rosetta. In addition, the catalytic capacity of PRM is likely affected by residue G95 in the active site pocket and residues I84, P165, A269, and V270 in the substrate tunnel structure. Mutation of amino acids in the active site and substrate tunnel resulted in inefficiency or inactivation of MTBE degradation, and the activity in 1,4-dioxane (1,4-D) degradation was diminished less than that in MTBE degradation.
IMPORTANCE
Multicomponent monooxygenases catalyzing the initial hydroxylation of MTBE are important in MTBE biodegradation. Previous studies of MTBE degradation enzymes have focused on P450s, alkane monooxygenase and MTBE monooxygenase, but the vital role of soluble di-iron monooxygenases has rarely been reported. In this study, we deciphered the essential catalytic role of a PRM and revealed the key residues of the PRM in MTBE metabolism. Our findings provide new insight into the MTBE-degrading gene cluster and enzymes in bacteria. This characterization of the PRM associated with MTBE degradation expands our understanding of MTBE-degrading gene diversity and provides a novel candidate enzyme for the bioremediation of MTBE-contaminated sites.
KEYWORDS: Mycobacterium vaccae JOB5, methyl tert-butyl ether, propane monooxygenase, initial degradation
INTRODUCTION
The fuel oxygenate, methyl tert-butyl ether (MTBE) is one of the most commonly used petroleum additives, especially in Europe, as it can increase the octane number, reduce carbon monoxide emissions, decrease unburnt hydrocarbons levels, and enhance fuel combustion efficiency (1, 2). MTBE is extremely water soluble and thus can move rapidly through soil columns (3). Leakage of MTBE-containing gasoline causes MTBE contamination in groundwater, posing a potential threat to human health (4–7). Toxicological studies have demonstrated that MTBE can damage the kidneys and livers of rats and might also endanger humans and other organisms by causing skin and eye injuries (8–10). Biodegradation is considered a potential strategy for the bioremediation of MTBE-contaminated sites. To date, most reported MTBE-utilizing bacteria are aerobes, including Methylibium (11), Hydrogenophaga (12), Mycobacterium (13), Aquincola (14), Pseudomonas (15), Pseudonocardia (16), Rhodococcus (17, 18), and Arthrobacter (19). Most MTBE-degrading organisms degrade MTBE through cometabolism. Only a few strains, such as Aquincola tertiaricarbonis L108, have been shown to utilize MTBE as a sole carbon and energy source (1).
In the proposed MTBE degradation pathway, MTBE is initially oxidized at the methyl group near the oxygen atom to form an unstable hemiacetal intermediate, which involves a monooxygenase-catalyzed reaction, followed either by dismutation to tert-butyl alcohol (TBA) or conversion to an ester intermediate, tert-butyl formate (TBF), which is then hydrolyzed to TBA (20, 21), . TBA is then hydroxylated to 2-methyl-1,2-propandiol (MPD), followed by enzymatic oxidation to 2-hydroxyisobutyric acid (HIBA). Finally, 2-HIBA is completely mineralized to CO2 or assimilated into biomass through the tricarboxylic acid cycle (Fig. S1) (22, 23). According to previous studies of MTBE intermediate metabolites, the MTBE degradation pathway is highly similar among the different reported MTBE-degrading bacteria, such as Methylibium petroleiphilum PM1 and Hydrogenophaga flava ENV735 (24–26).
The initial oxidation of MTBE is a multicomponent monooxygenase-catalyzed reaction, and previous studies have demonstrated that various enzymes, including cytochrome P450 monooxygenase (EthB and CamC) (27, 28), MTBE monooxygenase (MdpA) (29, 30), and alkane monooxygenase (AlkB) (31), are involved in the hydroxylation of MTBE. The gene cluster ethABCD is widely distributed in Rhodococcus strains (including IFP2001 and IFP2005), A. tertiaricarbonis L108 and Mycobacterium sp. IFP2009 (24, 28, 32). The mdp gene cluster is similar to the alk gene cluster, comprising the MTBE monooxygenase-encoding gene mdpA, rubredoxin-encoding gene mdpB, rubredoxin reductase-encoding gene mdpD and translational regulator-encoding gene mdpC. The mdpA gene encoding MTBE monooxygenase is responsible for the initial degradation of MTBE in M. petroleiphilum PM1, which was the best characterized MTBE degrader that did not exhibit accumulation of the intermediate TBA during MTBE degradation (30). In addition, the presence of enzymes that are responsible for the initial oxidation of MTBE in MTBE-contaminated sites provides direct evidence for the natural depletion of MTBE in groundwater. MTBE-degrading bacteria were detected in almost all MTBE-contaminated sites by metagenomic analysis; however, known MTBE-degrading proteins, such as MdpA, EthB, and AlkB, were found in only some MTBE-contaminated sites by proteomic analysis, suggesting that other unknown proteins responsible for MTBE bioremediation might be overlooked and highlighting the necessity for further exploring novel functional proteins involved in MTBE degradation (1).
Mycobacterium vaccae JOB5 is a Gram-positive bacterium that can grow on alkanes and isoalkanes, such as propane and butane, as sole carbon sources and can cometabolize diverse organic pollutants, including MTBE, 1,4-dioxane, and tetrahydrofuran (THF), when it is fed with propane, butane, or other short-chain alkanes as the auxiliary substrate (7, 25, 33–35). JOB5 contained at least two distinct copies of an alkB gene, encoding alkane monooxygenases, which shared ~60% nucleotide identity. However, neither alkane monooxygenases were involved in MTBE degradation in JOB5, suggesting that other unknown enzymes may contribute to this process (36). SDIMOs are nonheme bacterial enzymes belonging to a multicomponent bacterial enzyme family. SDIMOs are known for their versatile degradation capabilities and high sequence diversity and can be divided into six subgroups according to their sequence similarity, genetic arrangement, and substrate preference (37, 38). Recently, group-5 THF/propane monooxygenase and group-6 propane monooxygenase were identified as being associated with the degradation of THF and 1,4-D, which share similar chemical properties with MTBE (38, 39). SDIMOs have been discovered in some MTBE-degrading strains, such as M. petroleiphilum PM1, in which a phenol hydroxylase and propane monooxygenase were identified (Table S1). In addition, the expression of SDIMOs in M. petroleiphilum PM1 was upregulated in MTBE-treated cells compared to ethanol-treated cells (26). Thus, we hypothesized that SDIMOs might play a vital role in the cometabolic degradation of MTBE in strain JOB5.
SDIMOs are composed of a minimum number of oxygenase, reductase, and regulatory units and are encoded by a gene cluster of four or six genes. The oxygenase has a dimeric structure and comprises two or three types of subunits (α2β2 or α2β2γ2) (40, 41). The active site, including a nonheme di-iron cluster coordinated by two histidine and four glutamate residues in a distinctive 4-helix bundle structure, is located in the oxygenase α-subunit (42). The protein structure and catalytic mechanism of group-1 phenol hydroxylase, group-2 toluene monooxygenase, and group-3 methane monooxygenase have been elucidated (41–43). Site-directed mutagenesis has been used to improve catalytic activity or broaden the range of nonnative substrates of SDIMOs and redesign the active site pocket of other enzymes to adapt their enzyme properties for specific needs (44, 45). Major successes have included promoting the hydroxylation activities of Δ9 desaturase by site-directed mutagenesis according to the structure of methane monooxygenase and altering the regiospecificity of toluene monooxygenase for nitrobenzene and its analogs (46, 47). However, the protein structure of PRM remains unknown, and the key amino acids responsible for its catalytic activity are uncharacterized.
In this study, a degradation gene cluster, prmABCD, encoding a group-6 PRM responsible for the initial degradation of MTBE by strain JOB5, was characterized and found to be substantially different from previously reported MTBE degradation gene clusters. Moreover, we identified the key residue of the oxygenase α-subunit PrmA via molecular docking. This research advances our understanding of the diversity of MTBE-degrading enzymes and reveals the key residue of PrmA, providing a new approach for the design of novel monooxygenases to degrade target chemicals, including MTBE and its analogs.
RESULTS
Characterization of the putative propane monooxygenase gene cluster in M. vaccae JOB5
Genome sequencing showed that the genome of M. vaccae JOB5 was 6.32 Mb in size with a 67.90% GC content, including one chromosome and two plasmids. Genome annotation of M. vaccae JOB5 revealed one alkane monooxygenase gene cluster (alkB1FG), one isolated alkane monooxygenase (alkB2), one copper-dependent membrane monooxygenase gene cluster (pmoCAB), and one propane monooxygenase gene cluster (prmABCD), in which prmABCD, encoding PRM, was speculated to be involved in MTBE degradation (Table S2). The putative MTBE monooxygenase PrmABCD has the same number and type of subunits as the reported group-6 PRM and consists of three components, including the oxygenase PrmAB, the coupling protein PrmC, and the reductase PrmD, with a total size of approximately 4.1 kb and a GC content of 63% (Fig. 1). A groEL gene that encodes a putative chaperone protein was located adjacent to and downstream of prmABCD and may play a critical role in the protein folding of SDIMO hydroxylase subunits (39). The presence of two transposase genes upstream and downstream of prmABCD and groEL, respectively, suggest that the intercellular spread of this unique gene cluster might be promoted via horizontal gene transfer in MTBE-contaminated environments.
Fig 1.
Organization of the three putative MTBE-degrading gene clusters and one putative MTBE-degrading gene in M. vaccae JOB5. Putative propane monooxygenase gene cluster prmABCD, copper-dependent membrane monooxygenase gene cluster pmoCAB, alkane monooxygenase gene cluster alkB1FG and alkane monooxygenase gene alkB2 in JOB5. The numbers on the left and right indicate the location of the gene cluster in the chromosome. The nucleotide sequence length of each gene is listed below.
NCBI BLAST results showed that the genetic arrangement of prmABCD was different from those of the representative known MTBE degradation gene clusters, such as the MTBE monooxygenase gene cluster mdpABD in M. petroleiphilum PM1, the alkane monooxygenase gene cluster alkBFG in Pseudomonas putida Gpo1, the P450 gene cluster ethABCD in A. tertiaricarbonis L108 (Fig. S2). The gene prmA encoding the oxygenase α-subunit shares low amino acid sequence identity with mdpA, ethB, camC, and alkB encoding oxygenases of the MTBE degradation gene cluster. A phylogenetic tree constructed using the oxygenase α-subunit of SDIMOs also showed that prmABCD is an SDIMO and shares a close relationship with group-6 propane monooxygenase (Fig. 2). In addition, the prmA gene shares high amino acid sequence identity with the prmA gene of Mycobacterium dioxanotrophicus PH-06 (98%), Rhodococcus wratislaviensis IFP2016 (94%), Mycobacterium chubuense NBB4 (94%), Mycobacterium sp. ENV421 (95%), and Rhodococcus rhodochrous ATCC 21198 (94%). To date, no SDIMO has been reported to be involved in MTBE degradation, which indicates that prmABCD is a completely different MTBE-degrading gene cluster. Therefore, the prmABCD gene cluster was chosen for further study to verify its involvement in MTBE degradation.
Fig 2.
Structure alignment and phylogenetic analysis of prmABCD. (A) Amino acid sequence identity and genetic arrangement analysis between the putative MTBE monooxygenase gene cluster prmABCD in M. vaccae JOB5 and other reported MTBE-degrading gene clusters. The arrows represent the gene size and direction of transcription of each gene, and genes shown with the same fill color are isoenzymes. The numbers under the genes represent the amino acid sequence identity of each component with the corresponding component in prmABCD. (B) Phylogenetic analysis of the gene encoding the oxygenase α-subunit prmA. Unrooted phylogenetic tree generated based on amino acid sequences. Six subgroup SDIMOs are represented by different colors. The strains marked with a blue triangle are able to degrade MTBE. Numbers represent the amino acid identity with the oxygenase α-subunit of strain JOB5.
The expression of PRM is upregulated during growth on MTBE and butane
To determine whether the prmABCD cluster is involved in MTBE degradation, the gene transcription levels during growth on glucose, glucose plus MTBE, butane and butane plus MTBE were determined using transcriptome sequencing. Transcriptome analysis revealed that the expression levels of 555 genes were upregulated and those of 558 genes were downregulated when strain JOB5 was grown in medium containing glucose plus MTBE relative to the expression in medium containing glucose alone (Fig. S3B). A total of 118 genes showed fold changes higher than 4.0 in cells induced with glucose plus MTBE (Fig. S4; Table S3), and the fold changes in the transcription levels of prmABCD were 15.0, 12.9, 12.2, and 11.9 compared to those in glucose-induced cells for the four subunits (Fig. 3), while the absolute transcription levels were similar (Fig. S3A). The transcription levels of the pmoCAB gene cluster were also upregulated during growth on glucose plus MTBE, with fold changes of 3.4, 5.0, and 15.1 compared to the transcription levels in glucose-induced cells. However, the alkB1FG gene cluster encoding alkane monooxygenase was downregulated, suggesting that it is not involved in MTBE degradation. In addition, the transcription levels of the entire prmABCD gene cluster except for prmA in cells grown on glucose plus MTBE were not obviously different from those in cells incubated with butane plus MTBE, and the prmA gene was also upregulated in butane-induced cells. Transcriptome analysis revealed that 1,280 genes were upregulated and 889 genes were downregulated in butane-induced cells relative to glucose-induced cells (Fig. S3C).
Fig 3.
Transcription levels of propane monooxygenase, copper-dependent membrane monooxygenase, and alkane monooxygenase-encoding genes during M. vaccae JOB5 growth on glucose plus MTBE relative to that on glucose (pink column), growth on butane relative to that on glucose (blue column), and growth on butane plus MTBE relative to that on glucose (green column). The 16S rRNA gene of strain JOB5 was used as a reference gene. At least three independent samples were collected for each substrate.
The PRM encoded by prmABCD can oxidize MTBE
To identify the role of prmABCD in MTBE degradation, the prmABCD gene cluster was cloned and expressed in E. coli Rosetta as the heterologous host. Western blot analysis indicated successful expression of the gene cluster encoding PRM in the soluble fraction (Fig. 4A), and almost all the PRM was expressed in inclusion body (Fig. S5). Purification of each His-tagged subunit of PRM and SDS‒PAGE analysis yielded the expected protein bands. All PRM subunits were expressed in soluble form, but only a small amount of PrmA was found in the supernatant (Fig. S5). E. coli Rosetta containing the recombinant plasmid pET-prmABCD expressing PRM could transform MTBE and yield TBA in comparison with the control cells with the empty vector pET-28a, which indicated that the PRM encoded by prmABCD catalyzed the initial degradation of MTBE (Fig. 4B). However, the key intermediate metabolite TBF was not detected during MTBE degradation. To identify the contribution of PRM to the metabolism of the intermediate TBF, TBF was used as a substrate in oxidation assays. The results showed that the PRM encoded by prmABCD did not catalyze the conversion of TBF to TBA, and TBF could be spontaneously transformed in liquid solution (Fig. 4C).
Fig 4.
Characterization of the function of propane monooxygenase and the essential role of each subunit in the initial degradation of MTBE. (A) Western blot analysis of prmABCD expression in the supernatant of E. coli Rosetta (DE3) pET-prm. Lanes 1, 2, 3, and 4 represent the expression of prmA, prmB, prmC, and prmD with the C-terminal 3 × FLAG tag in the supernatant, respectively. The expected protein band sizes of PrmA-3*FLAG, PrmB-3*FLAG, PrmC-3*FLAG, and PrmD-3*FLAG were 61.6, 42.9, 14.5, and 40.0 kDa, respectively. (B) Time course of MTBE transformation and intermediate TBA production by E. coli Rosetta (DE3) (pET-prm). The TBA levels of the abiotic control and the empty vector pET-28a(+) control were below the threshold of detection. (C) Time course of TBF transformation and intermediate TBA production by E. coli Rosetta (DE3) expressing plasmid pET-prm. (D) MTBE consumption by E. coli Rosetta (DE3) containing different expression plasmids. The experiments were performed in triplicate, the results shown here are the average value of three biological replicates, and error bars show standard deviations.
Different recombinant plasmids harboring all or some part of the prmABCD gene cluster were constructed to further investigate whether all four subunits were crucial for MTBE degradation activity. Recombinant strains containing the plasmids pET-prmABC, pET-prmABD, pET-prmBCD, and pET-prmAB, which lacked prmD, prmC, prmA, and prmCD, respectively, were unable to catalyze MTBE degradation (Fig. 4D). This result indicated that an intact prmABCD gene cluster was essential for MTBE degradation.
Prediction and analysis of PrmA protein structure and molecular docking with substrates
The PRM of M. vaccae JOB5 expressed in E. coli Rosetta was efficient in cyclic ether degradation, including the degradation of 1,4-D and THF (Fig. 5A). Notably, the PRM exhibited no degradation activity for MTBE analogs, including the branched ether substrates ETBE, TAME, and TBA (Fig. 5A). The oxygenase α-subunit PrmA plays a key role in the catalytic process and contains the substrate binding site of PRM (40). To investigate the mechanism of action of PrmA toward these substrates, AlphaFold2 was used to predict the native structures of oxygenase and PrmA. The oxygenase of PRM has an α2β2 dimeric structure including a di-iron cluster in each oxygenase α-subunit PrmA. The structure of PrmA obtained by AlphaFold2 was then used for molecular docking to simulate the interaction between PrmA and the substrates. As shown in Fig. 5B; Fig. S6, molecular docking revealed the potential substrate binding pocket at the active site and substrate transport tunnel in PrmA, which were similar to other reported SDIMOs, such as the group 1 phenol hydrolase PheN from Pseudomonas sp. OX1. The substrate transport tunnel consists of three cavities located at the hydrolase α-subunit leading from the active site to the protein surface, and the substrate tunnel involves mostly hydrophobic residues (41–43). THF, 1,4-D, and MTBE were well anchored in the active site pocket, while ETBE, TAME, and TBA could not fit within the pocket due to their inappropriate volume (Fig. S7). The results of molecular docking were consistent with the substrate spectrum detected, indicating the confidence of the molecular docking and the essential role of the active site in catalytic capacity.
Fig 5.
Substrate spectrum of propane monooxygenase and molecular docking of PrmA and MTBE in the active site pocket and substrate tunnel. (A) Degradation of THF, 1,4-D, ETBE, TAME, TBA, and butane by E. coli Rosetta (DE3) heterologously expressing propane monooxygenase. All points represent the mean values of triplicate trials with error bars denoting the standard deviations. (B) Molecular docking of the receptor PrmA and ligand MTBE (CDOCKER energy = −13.26 kcal/mol). Two black boxes with dotted lines represent the binding pockets of the substrate tunnel and active site pocket of PrmA. Yellow dotted lines represent nonbonding interaction between residues and MTBE. Cyan-colored residues represent the amino acids that differ between strains JOB5 and PH-06. The orange spheres represent iron atoms, red sticks represent oxygen atoms, green sticks and cyan sticks represent carbon atoms, and blue sticks represent nitrogen atoms.
Site-directed mutagenesis of key residues of PrmA results in inefficiency or inactivation of MTBE degradation
PrmA of M. vaccae JOB5 shares the highest identity with that of M. dioxanotrophicus PH-06, belonging to a PRM containing that is inactive in MTBE degradation (39). Here, the residues of the oxygenase α-subunit that differed between JOB5 and PH-06, which determine substrate specificity, were analyzed based on the structural model of PrmA. According to the differences in the amino acid sequence of PrmA between PH-06 and JOB5, six variants expressing the whole PRM operon containing single mutations of PrmA of JOB5, including I84V, G95A, P165A, A269S, V270T, and Q288E, were constructed and characterized by oxidation assays (Fig. 6A). The results showed that the MTBE degradation activity of the mutations I84V and G95A decreased to 36.8% and 19.2%, respectively, in comparison to PRM containing wild-type the PrmA (Fig. 6B). In addition, no activity was detected in the other four variants, namely, P165A, A269S, V270T, and Q288E. However, all six variants, I84V, G95A, P165A, A269S, V270T, and Q288E, were able to catalyze the degradation of the MTBE analog cyclic ether 1,4-D, despite the 1,4-D degradation activity decreasing to 18.3%, 82.9%, 12.9%, 55.7%, 5.9%, and 15.6%, respectively. For degradation of the native substrate butane, two variants, A269S and V270T, showed no activity and the activity of the other four variants, I84V, G95A, P165A, and Q288E, decreased to 23.0%, 23.9%, 12.5%, and 52.7%, respectively. Investigation of the PrmA structure revealed that G95 is located in the active site pocket, and I84V, A269, and V270 are located at the entrance of the substrate tunnel (Fig. 6C). Although I84V and G95A had weaker effects on the polarity of residues and electrostatic force, I84V may block the entrance of substrate, and P165A may change the secondary structure of PrmA and disrupt the alkyl hydrophobic interaction between MTBE and P165, while the A269S and V270T mutations change the polarity of the residues. G95A reduces the pocket volume of the active site to influence the activity of MTBE transformation (Fig. S8). These results suggest that P165, A269, and V270 are crucial amino acids for the catalytic capacity of PrmA. The different amino acids indicated in the sequence alignment between JOB5 and other PrmAs of group-6 PRM may be potential key sites involved in the catalytic activity of MTBE.
Fig 6.
Catalytic capacity of PrmA in TBA production during the transformation of MTBE and ligand-residue interaction between MTBE and key residues. (A) Sequence alignment of the oxygenase α-subunit of Group-6 propane monooxygenase. The deduced amino acid sequences were obtained from Mycobacterium dioxanotrophicus PH-06 (PH-06, ART74426.1), Rhodococcus wratislaviensis IFP2016 (IFP2016, AFX59912.1), Rhodococcus rhodochrous ATCC21198 (ATCC21198, ETT27060.1), Mycobacterium chubuense NBB4 (NBB4, ACZ56324.1), and Mycobacterium sp. ENV421 (ENV421, WP_102810306.1). Numbers above the sequences correspond to the numbering for PrmA from JOB5. Red boxes with lines represent the different amino acids between JOB5 and PH-06. (B) The intermediate TBA production of E. coli Rosetta (DE3) containing plasmids that express mutated PrmA in MTBE degradation. All points represent the mean values from triplicate trials, with error bars indicating the standard deviations. (C) Molecular docking of the ligand MTBE and the receptor PrmA containing the A269S and V270T mutations in the binding pocket of the substrate tunnel. Yellow dotted lines represent the nonbonding interaction between residues and MTBE, the number represents the distance between residues and MTBE, the orange spheres represent iron atoms, red sticks represent oxygen atoms, green sticks represent carbon atoms, and blue sticks represent nitro atoms.
DISCUSSION
SDIMOs are bacterial multicomponent monooxygenases that participate in the degradation of a broad variety of substrates, including aromatics, alkanes, alkenes, and chlorinated compounds, by catalyzing a series of oxygenation reactions (48–50). Previous studies showed that the widespread distribution of SDIMOs in bacterial consortia was correlated with the prevalence of contamination (51). However, the lag in the functional characterization of SDIMOs makes it easy to underestimate the actual remediation potential of microorganisms in environmental remediation. In this study, the MTBE-degrading gene cluster prmABCD, which encodes PRM in strain M. vaccae JOB5 and differs substantially from previously reported MTBE degradation gene clusters, was identified and characterized.
According to our study, the number and type of subunits and genetic arrangement of the MTBE-degrading gene cluster in strain JOB5 are the same as those of group-6 PRM ,which is widespread in THF- and 1,4-D-degrading bacteria. The group-6 PRM gene cluster is generally found in Gram-positive bacteria, including M. dioxanotrophicus PH-06, R. wratislaviensis IFP2016, M. chubuense NBB4, Mycobacterium sp. ENV421, and R. rhodochrous ATCC 21198, among which the capacity of MTBE degradation was identified in strains ENV421 and IFP2016. However, whether the group-6 PRM in strains ENV421 and IFP2016 is responsible for MTBE degradation remains unknown, and the reported group-6 PRM of M. dioxanotrophicus PH-06 was shown to be inactive in MTBE degradation (52, 53). We speculated that the key amino acids of PRM were mutated for adaptation to the complex contaminated environment and increased pollutant degradation for the following reasons: (i) the amino acids that differed only between JOB5 and PH-06 were conserved in other group-6 PRMs (Fig. 6A); and (ii) the PRM of PH-06 was inactive in MTBE degradation, while some other group-6 PRM-containing bacteria were MTBE-degrading bacteria, such as strains ENV421 and IFP2016 (52, 53). In this study, the metabolite TBA accumulated during MTBE biotransformation by E. coli Rosetta (DE3)-pET28a-prmABCD, which suggested that the initial degradation of MTBE was catalyzed by the PRM of JOB5, but the PRM was not involved in TBA degradation. However, degradation of TBA was observed in MTBE metabolism in the presence of butane as an auxiliary substrate, which suggested that another butane-induced unknown monooxygenase was involved in TBA hydroxylation. A copper-dependent membrane monooxygenase was most likely involved in TBA hydroxylation, as indicated by the upregulated expression of the pmoCAB gene cluster according to the data from the transcriptome (Fig. 3), and its function in TBA hydroxylation should be verified in further studies. Additionally, short-chain alkanes, such as propane, butane, and pentane, were auxiliary substrates and inducers of strain JOB5 in the cometabolic degradation of organic pollutants, such as MTBE, 1,4-D, THF, and 1,2,3-trichloropropane (13, 54–56). The traits of strain JOB5 suggest that the expression of genes involved in MTBE degradation, such as prmA, could be induced by butane. In addition, the widespread presence of SDIMOs in MTBE-degrading strains suggests that SDIMOs play a vital role in MTBE bioremediation and could be considered a biomarker for monitoring the natural attenuation of MTBE (Table S1).
We successfully expressed all subunits of PRM in soluble form in E. coli Rosetta (DE3) from M. vaccae JOB5. However, no MTBE-stimulated NAD(P)H-oxidizing activity or MTBE degradation was detected in the crude extracts of E. coli Rosetta (DE3) containing the plasmid pET-28a-prmABCD, despite efforts toward coexpression with the chaperonin GroEL, replacing the heterologous expression host with Mycobacterium smegmatis mc2-155, adding crude extracts of M. vaccae JOB5 as a supplement and replacing sonication with the use of a French press for cell disruption. In addition, upon heterologous expression of some kinds of SDIMOs of the other strains in E. coli, such as toluene monooxygenase from Pseudomonas stutzeri OX1, alkene monooxygenase from Xanthobacter autotrophicus Py2, and alkene monooxygenase from M. chubuense NBB4, the proteins were either inactive or formed inclusion bodies (57–60). We hypothesized that the activity of PRM was lost upon breakage of cells and that some unknown membrane-associated accessory proteins might be required for maintaining the assembly of PRM and other SDIMOs.
The predicted three-dimensional structure of the oxygenase α-subunit PrmA is similar to the structure of the previously reported methane monooxygenase of Methylococcus capsulatus Bath, sharing 36.5% sequence identity. The active site pocket and substrate tunnel were proposed as critical sites for the catalytic activity of PrmA in molecular docking, and the mutation of amino acids adjacent to the substrate tunnel resulted in the inactivation of MTBE degradation. The varied degradation efficiencies of the wild-type enzyme toward THF, 1,4-D, and MTBE and the lack of activity toward ETBE, TAME, and TBA could have partially resulted from the fit of substrates in the active site pocket (Fig. S7). Molecular docking revealed that the ethyl group of ETBE and the methyl group of TAME as hydroxylation sites of the oxidation reaction were far away from the di-iron atoms, and TBA could not be well anchored at the active site pocket due to its small volume (Fig. S7). The mutation of key residues of PrmA in the active site pocket and the substrate tunnel affected the catalytic capacity of PrmA in MTBE. I84V, P165A, A269S, and V270T influenced MTBE across the substrate tunnel to affect the catalytic ability of MTBE. I84V and P165A blocked the entrance of the substrate tunnel to hinder the substrate MTBE from entering the active site. The A269S or V270T single mutation had a weaker effect on protein structure but changed the polarity of residues that blocked the entry of MTBE. G95A affected the pocket volume of the active site to influence the binding of MTBE and PrmA. Notably, Q288 was located at the surface of PrmA but strongly influenced the activity of PrmA in MTBE transformation, while the function of Q288 in catalytic capacity remains uncertain. In addition, the activity of PRM in 1,4-D degradation was diminished less than that in MTBE degradation. This finding suggests that the substrate tunnel is as essential as the active site pocket in the catalytic activity of MTBE. In addition, the different amino acids between JOB5 and other group-6 PRMs may also be potential crucial amino acids in the catalytic activity of MTBE or affect the substrate spectrum of PRM.
In summary, we identified a novel multicomponent PRM encoded by the prmABCD gene cluster in a Gram-positive MTBE-degrading bacterium, M. vaccae JOB5, and demonstrated its activity in initiating the degradation of MTBE. The MTBE oxidation activity of PRM was verified in E. coli Rosetta heterologously expressing the prmABCD gene cluster. The key amino acids of the oxygenase α-subunit PrmA in the substrate spectrum were identified. This study expands our understanding of the diversity of MTBE degradation gene clusters and broadens the substrate spectrum of PRM. Further study on reshaping the active site pocket and substrate tunnel to broaden the substrate range of PRM for common groundwater pollutants is still ongoing.
MATERIALS AND METHODS
Bacterial strains, chemicals, and culture conditions
The bacterial strains and plasmids used in this study are listed in Table 1, and all the primers are listed in Table 2. M. vaccae JOB5 was cultivated in ammonia mineral salt (AMS) medium at 30°C on a rotary shaker (200 rpm) with 10% butane (vol/vol) injected in the headspace as the sole carbon and energy source. E. coli strains were grown in LB at 37°C. Kanamycin (50 mg/L) and chloramphenicol (25 mg/L) were added to the medium if necessary. The chemicals 1,4-D, MTBE, TBA, TBF, ethyl tert-butyl ether (ETBE), and t-amyl methyl ether (TAME) were purchased from Aladdin Industrial Corporation (Shanghai, China) at a purity of more than 99%. Phenylmethanesulfonyl fluoride (PMSF) protease inhibitor and dithiothreitol (DTT) were purchased from Beyotime Corporation (Shanghai, China). Butane was obtained from Jingong Gas Co., Ltd. (Hangzhou, China). Unless otherwise stated, all experiments with liquid culture containing MTBE were performed in 500 mL glass serum vials containing 100 mL of AMS medium. The serum vials were sealed with butyl rubber stoppers and crimped with aluminum caps to reduce the volatilization of MTBE during cultivation.
TABLE 1.
Bacterial strains and plasmids used in this study
| Strain or plasmid | Characteristicsa | Reference or source |
|---|---|---|
| Strains | ||
| M. vaccae JOB5 | Wild type, MTBE-degrading strain; G+ | DSMZ |
| M. smegmatis mc2-155 | Wild type, used for gene expression; G+ | ATCC |
| mc2-155-pJV53 | mc2-155 transformed with pJV53; Kmr | This study |
| mc2-155-pJV53-prmABCD | mc2-155 transformed with pJV53-prmABCD; Kmr | This study |
| E. coli | ||
| DH5α | supE44 lacU169(80dlacZΔM15) hsdR17 recA1 endA1 gyrA96 Δthi relA1 | Tsingke |
| BL21(DE3) | Chaperonin gene groELS integrated in genome | 37 |
| Rosetta(DE3) | F- ompT hsdSB (rB−, mB−) gal dcm(DE3) pRARE; CamR | TransGen |
| Rosetta(DE3)-pET28a | Rosetta(DE3) transformed with pET28a; Kmr | This study |
| Rosetta (DE3)-pET28a-prmABCD | Rosetta(DE3) transformed with pET28a-prmABCD; Kmr | This study |
| Rosetta(DE3)-pET28a-prm-A-3*FLAG | Rosetta(DE3) transformed with pET28a-prm-A-3*FLAG; Kmr | This study |
| Rosetta(DE3)-pET28a-prm-B-3*FLAG | Rosetta(DE3) transformed with pET28a-prm-B-3*FLAG; Kmr | This study |
| Rosetta(DE3)-pET28a-prm-C-3*FLAG | Rosetta(DE3) transformed with pET28a-prm-C-3*FLAG; Kmr | This study |
| Rosetta(DE3)-pET28a-prm-D-3*FLAG | Rosetta(DE3) transformed with pET28a-prm-D-3*FLAG; Kmr | This study |
| Rosetta(DE3)-pET28a-prmA | Rosetta(DE3) transformed with pET28a-prmA; Kmr | This study |
| Rosetta(DE3)-pET28a-prmB | Rosetta(DE3) transformed with pET28a-prmB; Kmr | This study |
| Rosetta(DE3)-pET28a-prmC | Rosetta(DE3) transformed with pET28a-prmC; Kmr | This study |
| Rosetta(DE3)-pET28a-prmD | Rosetta(DE3) transformed with pET28a-prmD; Kmr | This study |
| Rosetta(DE3)-pET28a-prmAB-His | Rosetta(DE3) transformed with pET28a-prmAB-His; Kmr | This study |
| Rosetta(DE3)-pET28a-prmC-His | Rosetta(DE3) transformed with pET28a-prmC-His; Kmr | This study |
| Rosetta(DE3)-pET28a-prmD-His | Rosetta(DE3) transformed with pET28a-prmD-His; Kmr | This study |
| Rosetta(DE3)-pET28a-prmABC | Rosetta(DE3) transformed with pET28a-prmABC-His; Kmr | This study |
| Rosetta(DE3)-pET28a-prmABD | Rosetta(DE3) transformed with pET28a-prmABD-His; Kmr | This study |
| Rosetta(DE3)-pET28a-prmBCD | Rosetta(DE3) transformed with pET28a-prmBCD-His; Kmr | This study |
| Rosetta(DE3)-pET28a-prmAB | Rosetta(DE3) transformed with pET28a-prmAB-His; Kmr | This study |
| Rosetta(DE3)-pET28a-prm-I84V | Rosetta(DE3) transformed with pET28a-prm-I84V; Kmr | This study |
| Rosetta(DE3)-pET28a-prm-G95A | Rosetta(DE3) transformed with pET28a-prm-G95A; Kmr | This study |
| Rosetta(DE3)-pET28a-prm-P165A | Rosetta(DE3) transformed with pET28a-prm-P165A; Kmr | This study |
| Rosetta(DE3)-pET28a-prm-A269S | Rosetta(DE3) transformed with pET28a-prm-A269S; Kmr | This study |
| Rosetta(DE3)-pET28a-prm-V270T | Rosetta(DE3) transformed with pET28a-prm-V270T; Kmr | This study |
| Rosetta(DE3)-pET28a-prm-Q288E | Rosetta(DE3) transformed with pET28a-prm-Q288E; Kmr | This study |
| BL21(DE3)-pET28a | BL21DE3) transformed with pET28a; Kmr | This study |
| BL21(DE3)-pET28a-prmABCD | BL21(DE3) transformed with pET28a-prm; Kmr | This study |
| Plasmids | ||
| pET-28a | Gene expression vector; Kmr | Novagen |
| pET28a-prmABCD | Vector for inducible expression of prmABCD; Kmr | This study |
| pET28a-prm-A-3*FLAG | Vector for inducible expression of prm-A-3*FLAG; Kmr | This study |
| pET28a-prm-B-3*FLAG | Vector for inducible expression of prm-B-3*FLAG; Kmr | This study |
| pET28a-prm-C-3*FLAG | Vector for inducible expression of prm-C-3*FLAG; Kmr | This study |
| pET28a-prm-D-3*FLAG | Vector for inducible expression of prm-D-3*FLAG; Kmr | This study |
| pET28a-prmA | Vector for inducible expression of prmA; Kmr | This study |
| pET28a-prmB | Vector for inducible expression of prmB; Kmr | This study |
| pET28a-prmC | Vector for inducible expression of prmC; Kmr | This study |
| pET28a-prmD | Vector for inducible expression of prmD; Kmr | This study |
| pET28a-prmAB-His | Vector for inducible expression of prmAB-His; Kmr | This study |
| pET28a-prmC-His | Vector for inducible expression of prmC-His; Kmr | This study |
| pET28a-prmD-His | Vector for inducible expression of prmD-His; Kmr | This study |
| pET28a-prmABC | Vector for inducible expression of prmABC; Kmr | This study |
| pET28a-prmABD | Vector for inducible expression of prmABD; Kmr | This study |
| pET28a-prmBCD | Vector for inducible expression of prmBCD; Kmr | This study |
| pET28a-prmAB | Vector for inducible expression of prmAB; Kmr | This study |
| pET28a-prm-I84V | Vector for inducible expression of prm-I84V; Kmr | This study |
| pET28a-prm-G95A | Vector for inducible expression of prm-G95A; Kmr | This study |
| pET28a-prm-P165A | Vector for inducible expression of prm-P165A; Kmr | This study |
| pET28a-prm-A269S | Vector for inducible expression of prm-A269S; Kmr | This study |
| pET28a-prm-V270T | Vector for inducible expression of prm-V270T; Kmr | This study |
| pET28a-prm-Q288E | Vector for inducible expression of prm-Q288E; Kmr | This study |
| pJV53 | Gene expression vector; Kmr | This study |
| pJV53-prmABCD | Vector for inducible expression of prmABCD; Kmr | This study |
Kmr, kanamycin resistance.
TABLE 2.
Primers used in this study
| Primers | Sequence (5′−3′)a | Purpose |
|---|---|---|
| pET28a-prmABCD-F | CTTTAAGAAGGAGATATACCATGTTCGCCCGCAGATCCGC | Design pET28a-prmABCD expression vector |
| pET28a-prmABCD-R | ATCTCAGTGGTGGTGGTGGTGGTGCGCGGATACCGGGGCCGACT | |
| pET28a-prmA-F | CTTTAAGAAGGAGATATACCATGTTCGCCCGCAGATCCGC | Design pET28a-prmA expression vector |
| pET28a-prmA-R | ATCTCAGTGGTGGTGGTGGTGGTGGATCTCGCTGAACGCGTCGG | |
| pET28a-prmB-F | CTTTAAGAAGGAGATATACCATGAGCGCCGAACTGATATC | Design pET28a-prmB expression vector |
| pET28a-prmB-R | ATCTCAGTGGTGGTGGTGGTGGTGCGACGTCACCGCCGCAGGAA | |
| pET28a-prmC-F | CTTTAAGAAGGAGATATACCATGAGCGACACCCAGTCCCA | Design pET28a-prmC expression vector |
| pET28a-prmC-R | ATCTCAGTGGTGGTGGTGGTGGTGGAACTTCGCCGGTTCGAGCTGA | |
| pET28a-prmD-F | CTTTAAGAAGGAGATATACCATGGCGCGCGTCACCCTGGT | Design pET28a-prmD expression vector |
| pET28a-prmD-R | ATCTCAGTGGTGGTGGTGGTGGTGCGCGGATACCGGGGCCGACT | |
| pET28a-prmAB-F | CTTTAAGAAGGAGATATACCATGTTCGCCCGCAGATCCGC | Design pET28a-prmAB expression vector |
| pET28a-prmAB-R | ATCTCAGTGGTGGTGGTGGTGGTGCGACGTCACCGCCGCAGGAA | |
| pET28a-prmABC-F | CTTTAAGAAGGAGATATACCATGTTCGCCCGCAGATCCGC | Design pET28a-prmABC expression vector |
| pET28a-prmABC-R | ATCTCAGTGGTGGTGGTGGTGGTGGAACTTCGCCGGTTCGAGCTGA | |
| pET28a-prmABD-F | CTGCGGCGGTGACGTCGTGAATGGCGCGCGTCACCCTGGT | Design pET28a-prmABD expression vector |
| pET28a-prmABD-R | ACCAGGGTGACGCGCGCCATTCACGACGTCACCGCCGCAG | |
| pET28a-prmBCD-F | CTTTAAGAAGGAGATATACCATGAGCGCCGAACTGATATC | Design pET28a-prmBCD expression vector |
| pET28a-prmBCD-R | ATCTCAGTGGTGGTGGTGGTGGTGCGCGGATACCGGGGCCGACT | |
| pET28a-prmA-FLAG-F | TGTCATGATCTTTATAATCACCGTCATGGTCTTTGTAGTCGATCTCGCTGAACGCGTCGG | Design pET28a-prmA-FLAG expression vector |
| pET28a-prmA-FLAG-R | TAAAGATCATGACATCGACTACAAGGATGACGATGACAAGTGATGAGCGCCGAACTGATA | |
| pET28a-prmB-FLAG-F | TGTCATGATCTTTATAATCACCGTCATGGTCTTTGTAGTCCGACGTCACCGCCGCAGGAA | Design pET28a-prmB-FLAG expression vector |
| pET28a-prmB-FLAG-R | TAAAGATCATGACATCGACTACAAGGATGACGATGACAAGTGAGCGACACCCAGTCCCAC | |
| pET28a-prmC-FLAG-F | TGTCATGATCTTTATAATCACCGTCATGGTCTTTGTAGTCGAACTTCGCCGGTTCGAGC | Design pET28a-prmC-FLAG expression vector |
| pET28a-prmC-FLAG-R | TAAAGATCATGACATCGACTACAAGGATGACGATGACAAGTGATTGGAGGGCAGATGGCG | |
| pET28a-prmD-FLAG-F | TGTCATGATCTTTATAATCACCGTCATGGTCTTTGTAGTCCGCGGATACCGGGGCCGACT | Design pET28a-prmD-FLAG expression vector |
| pET28a-prmD-FLAG-R | TAAAGATCATGACATCGACTACAAGGATGACGATGACAAGTGAGATCCGGCTGCTAACAA | |
| pET28a-prm-I84V-F | TTATGGAGGTCATGAAGCCG | Design pET28a-prm-I84V expression vector |
| pET28a-prm-I84V-R | CGGCTTCATGACCTCCATAA | |
| pET28a-prm-G95A-F | TTGGTCGACTTCGCCGAGTA | Design pET28a-prm-G95A expression vector |
| pET28a-prm-G95A-R | TACTCGGCGAAGTCGACCAA | |
| pET28a-prm-P165A-F | TCGAGCGGGGCGGGCGGTGTT | Design pET28a-prm-P165A expression vector |
| pET28a-prm-P165A-R | AACACCGCCCGCCCCGCTCGA | |
| pET28a-prm-A269S-F | ATTTCCAGGGCTCGGTTTAC | Design pET28a-prm-A269S expression vector |
| pET28a-prm-A269S-R | GTAAACCGAGCCCTGGAAAT | |
| pET28a-prm-V270T-F | GGGCGCGACTTACGACTACT | Design pET28a-prm-V270T expression vector |
| pET28a-prm-V270T-R | AGTAGTCGTAAGTCGCGCCC | |
| pET28a-prm-Q288E-F | TACTGGGACGAGTGGATCTG | Design pET28a-prm-Q288E expression vector |
| pET28a-prm-Q288E-R | CAGATCCACTCGTCCCAGTA | |
| pET28a-F | GAAACAAGCGCTCATGAG | Verify gene expression vector |
| pET28a-R | CCGGATATAGTTCCTCCT | |
| pJV53-prmABCD-F | TACGAGATCGGCGGCCGCATATGACTGCATCGGTCACCAC | Design pJV53-prmABCD expression vector |
| pJV53-prmABCD -R | GTCGGAATTCGCCGGGGCGCTCAGTGGTGGTGGTGGTGGTGCGCGGATACCGGGGCCGACT | |
| pJV53-F | GCCGCAGTTGTTCTCGCATA | Verify gene expression vector |
| pJV53-R | ACGTCAGGTGGCTAGCTGAT |
Underlined nucleic acid sequences represent homologous sequences derived from plasmids.
Cloning, mutant construction, and protein expression
A 3,987 bp fragment of the prmABCD cluster was amplified using the genomic DNA of JOB5 as the template, and the products were inserted into pET-28a(+) and pJV53 by using an In-Fusion HD cloning kit (Takara Bio, USA). The mutation sites were introduced by site-directed mutagenesis PCR with the appropriate primers listed in Table 2, using pET-prmABCD as a template. The program used was as follows: 98℃ for 5 min; 30 cycles, 98℃ for 20 s, 58℃ for 20 s, and 72℃ for 2 min; and 7 min at 72℃. The mutation sites were identified by sequencing. The recombinant plasmids pET-prmABCD, pJV53-prmABCD and the resulting variant plasmids were then harvested using the EZNA Plasmid Kit (Omega Bio-Tek Co., Ltd.) and transformed into E. coli Rosetta (DE3) or E. coli BL21 (DE3) via chemical transformation. Then, the sequence was checked to verify that the correct mutant was obtained. E. coli Rosetta (DE3) and E. coli BL21 (DE3) transformants with the empty vector pET-28a(+) or recombinant plasmid pET-prmABCD were grown in LB medium at 37°C to an OD600 of 0.6 and then induced with 0.5 mM isopropyl-β-D-thiogalactopyranoside (IPTG) for 4 h at 30°C for heterologous expression. M. smegmatis mc2-155 transformants with the empty vector pJV53 or recombinant plasmid pJV53-prmABCD were grown in AMS medium at 30°C and induced with 2 mM acetamide for 24 h at 30°C for heterologous expression. The method for cell harvesting and MTBE degradation was as follows: 100 mL of cultivation medium was centrifuged at 4,600 × g for 5 min for cell harvesting, and the harvested cells were then washed two times with phosphate-buffered saline (pH 7.4), resuspended in ice-cold 50 mM PB buffer (pH 7.4) and diluted to an OD600 of 10. Then, 0.3 mM MTBE was added as a substrate. The treatments were all incubated at 30°C and 200 rpm and performed in triplicate. An abiotic control without cells was prepared to evaluate the volatilization of MTBE. The samples were collected every 2 h and centrifuged to remove the cells. The supernatants were recovered to evaluate residual MTBE and intermediate metabolite (TBA) production. The assays for THF, 1,4-D, ETBE, TAME, TBA, and MTBE degradation by the variants were conducted by the same methods with a slight modification as follows: 0.2 mM THF, 1,4-D, MTBE, ETBE, TAME, and TBA were added as substrates. The samples were collected after 12 h of incubation and centrifuged to remove the cells. The supernatants were recovered to evaluate residual substrates.
Analytical methods
Butane, MTBE, and the MTBE analogs ETBE, TAME, THF, 1,4-D, and TBA were detected by GC-2014C gas chromatography equipped with a flame ionization detector, an AOC-20i autoinjector and an SH-Stabilwax-DA column (Shimadzu, Shanghai, China). The MTBE, ETBE, TAME, and TBA concentrations were determined as follows: the column temperature was held at 60℃ for 1 min, followed by a stepwise gradient to 160℃ over 5 min. The injector and detector temperatures were set to 220°C and 250°C, respectively. The THF and 1,4-D concentrations were measured by the same program with a slight modification as follows: the column temperature was raised by a stepwise gradient from 60℃ to 160℃ over 5 min. The program for determining butane concentration was modified as follows: the column temperature was held at 40℃ for 3 min, and the injector and detector temperatures were set to 220°C and 250°C, respectively. Standard curves of butane, MTBE, ETBE, TAME, TBA, THF, and 1,4-D were appropriately prepared using the authentic standards of substrates, and the substrate concentrations were calculated from the standard curves. The limit of detection for THF, 1,4-D, MTBE, ETBE, TAME, and TBA was 0.01 mM, and that for butane was 1 µM, determined using the methods described above.
Preparation of cell extracts and crude enzyme assays
E. coli BL21-groELS and E. coli Rosetta carrying the recombinant plasmid pET-prmABCD or empty vector pET-28a(+) were grown in LB medium for heterologous expression as previously described. About 100 mL of cultivation liquid was centrifuged at 12,000 × g for 10 min, and the cells were harvested and then resuspended in 10 mL of ice-cold 50 mM Tris-HCl buffer (pH 7.4) (37, 38). The E. coli suspension was disrupted by sonication or in a French press cell in an ice-cold bath for 30 min (sonication for 3 s with 4 s intervals and pressurization in a French press cell for 12,000 psi), in which 1 mM PMSF was added as a protease inhibitor, and the cell debris was removed by ultracentrifugation at 100,000 × g for 1 h at 4°C. M. vaccae JOB5 was cultivated in AMS medium with 10% butane (vol/vol) injected in the headspace at 30°C to an OD600 of 1.0, and M. smegmatis mc2-155 transformant cells were cultivated and induced as previously described. Cells were harvested and disrupted by sonication, as described for E. coli. Protein concentrations were estimated by using a BCA kit (Beyotime, Shanghai, China). The reaction mixtures contained E. coli cell extract (50 mg/mL) or M. smegmatis mc2-155 transformant cell extract (50 mg/mL), 0.2 mM MTBE, 0.15 mM NADH/NADPH, 5 mM DTT, and JOB5 cell extract (50 mg/mL), and the final volume was adjusted to 1 mL with 50 mM Tris-HCl buffer (pH 7.4). The assay was initiated by the addition of MTBE, and the activity of the crude enzyme was estimated by measuring the decrease in absorbance at 340 nm due to the consumption of NADH/NADPH in the presence of MTBE. One unit of enzyme activity was defined as the amount of enzyme required to reduce 1 mM electron donor per min at 30°C.
Genome and transcriptome sequencing
In this study, genomic DNA extraction, sequencing library construction, and DNA and RNA sequencing were outsourced to Novogene Biotech Company (Beijing, China). Whole-genome sequencing of strain JOB5 was performed using the PacBio platform, and genes were predicted by Rapid Annotations using Subsystems Technology (RAST; http://rast.nmpdr.org/). The Kyoto Encyclopedia of Genes and Genomes (KEGG), Clusters of Orthologous Groups (COG), Non-Redundant (NR) Protein Database, and Gene Ontology (GO) databases were used for general functional annotation. For transcriptome sequencing, JOB5 cells were collected after growth to the mid-logarithmic phase on 5 mM glucose, glucose plus 5 mM MTBE, 20% butane (vol/vol), or butane plus MTBE. Three independent samples were collected from three liquid cultures in 250 mL butyl rubber-sealed shake flasks. The total RNA of strain JOB5 was extracted using an EZNA Bacterial RNA Kit (Omega Bio-Tek Co., Ltd.) and stored at −80°C, according to the manufacturer’s instructions.
Protein purification and western blot analysis
Three ORFs encoding each subunit of PRM were amplified and inserted into pET-28a(+), yielding the plasmids pET-prmAB, pET-prmC, and pET-prmD. Recombinant E. coli Rosetta (DE3) expressing the subunits PrmAB, PrmC, or PrmD was prepared as described above. His-tagged proteins were purified from E. coli Rosetta (DE3) with a tagged-protein purification kit (Beijing ComWin Biotech Co., Ltd.). All purification procedures were carried out at 4°C. The prmA, prmB, prmC, and prmD genes were tagged with 3 × FLAG, resulting in the plasmids pET-prm-A-FLAG, pET-prm-B-FLAG, pET-prm-C-FLAG, and pET-prm-D-FLAG, respectively. Western blotting was performed as previously described as follows (61): proteins separated by SDS‒PAGE were electrophoretically transferred to a polyvinylidene difluoride membrane using a Mini-transblot module (Bio-Rad). The blotted membrane was probed with a mouse monoclonal antibody (1:2,000) against the 3 × FLAG tag. The rabbit anti-mouse IgG horseradish peroxidase antibody (Abcam Co., Ltd.) was used as the secondary antibody (1:2,000), and the blot was visualized using a chemiluminescence Western blotting kit (Fdbio Co., Ltd.) following the manufacturer’s instructions.
Structure prediction and molecular docking
The structure of PrmA was predicted by AlphaFold2, and the substrate MTBE was docked into the catalytic pocket of PrmA using Discover Studio 2019. The ColabFold Google Colabs notebook available at https://colab.research.google.com/github/sokrypton/ColabFold/blob/main/AlphaFold2.ipynb was used, and an AlphaFold2 model was obtained with an average plDDT of 95 (a high confidence level). Molecular docking (MD) simulation was carried out using CDOCKER with the force field CHARMm. The MD simulation involved setting up the box, adding water and ions, energy minimization, and MD calculation. The substrate tunnel was identified by Caver 3.0.3 in PyMOL, and the steps involved included inputting the model, setting the starting point, and computing the tunnel.
ACKNOWLEDGMENTS
This work was financially supported by grants from the National Natural Science Foundation of China (Nos. 32170107 and 41721001) and the Key Research and Development Program of Zhejiang Province (grant number 2021C03168).
Contributor Information
Zhenmei Lu, Email: lzhenmei@zju.edu.cn.
Jeremy D. Semrau, University of Michigan-Ann Arbor, Ann Arbor, Michigan, USA
DATA AVAILABILITY
The genome sequence of M. vaccae JOB5 has been deposited in the SRA database and GenBank under accession numbers SRR23703833 and CP126901 to CP126903, respectively. The transcriptomic data for strain JOB5 growing on different substrates have been deposited in the SRA database under accession numbers SRR23725360 to SRR23725371.
SUPPLEMENTAL MATERIAL
The following material is available online at https://doi.org/10.1128/aem.01187-23.
The file "Supporting information" includes 8 figures and 3 tables as Supplemental figures and tables.
ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.
REFERENCES
- 1. Kucharzyk KH, Rectanus HV, Bartling CM, Rosansky S, Minard-Smith A, Mullins LA, Neil K. 2019. Use of omic tools to assess methyl tert-butyl ether (MTBE) degradation in groundwater. J Hazard Mater 378:120618. doi: 10.1016/j.jhazmat.2019.05.011 [DOI] [PubMed] [Google Scholar]
- 2. Liu T, Ahn H, Sun W, McGuinness LR, Kerkhof LJ, Häggblom MM. 2016. Identification of a Ruminococcaceae species as the methyl tert-butyl ether (MTBE) degrading bacterium in a methanogenic consortium. Environ Sci Technol 50:1455–1464. doi: 10.1021/acs.est.5b04731 [DOI] [PubMed] [Google Scholar]
- 3. Roy JW, Bickerton G. 2010. Proactive screening approach for detecting groundwater contaminants along urban streams at the reach-scale. Environ Sci Technol 44:6088–6094. doi: 10.1021/es101492x [DOI] [PubMed] [Google Scholar]
- 4. Ayotte JD, Argue DM, McGarry FJ. 2005. Methyl tert-butyl ether occurrence and related factors in public and private wells in southeast New Hampshire. Environ Sci Technol 39:9–16. doi: 10.1021/es049549e [DOI] [PubMed] [Google Scholar]
- 5. Connor JA, Kamath R, Walker KL, McHugh TE. 2015. Review of quantitative surveys of the length and stability of MTBE, TBA, and benzene plumes in groundwater at UST sites. Ground Water 53:195–206. doi: 10.1111/gwat.12233 [DOI] [PubMed] [Google Scholar]
- 6. Schmidt TC, Haderlein SB, Pfister R, Forster R. 2004. Occurrence and fate modeling of MTBE and BTEX compounds in a Swiss Lake used as drinking water supply. Water Res 38:1520–1529. doi: 10.1016/j.watres.2003.12.027 [DOI] [PubMed] [Google Scholar]
- 7. Flanagan SM, Levitt JP, Ayotte JD. 2017. Trends in methyl tert-butyl ether concentrations in private wells in southeast New Hampshire: 2005 to 2015. Environ Sci Technol 51:1168–1175. doi: 10.1021/acs.est.6b04149 [DOI] [PubMed] [Google Scholar]
- 8. de Peyster A, MacLean KJ, Stephens BA, Ahern LD, Westover CM, Rozenshteyn D. 2003. Subchronic studies in Sprague-Dawley rats to investigate mechanisms of MTBE-induced Leydig cell cancer. Toxicol Sci 72:31–42. doi: 10.1093/toxsci/kfg011 [DOI] [PubMed] [Google Scholar]
- 9. Phillips S, Palmer RB, Brody A. 2008. Epidemiology, toxicokinetics, and health effects of methyl tert-butyl ether (MTBE). J Med Toxicol 4:115–126. doi: 10.1007/BF03160966 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Zhang L, Qin J, Zhang Z, Li Q, Huang J, Peng X, Qing L, Liang G, Liang L, Huang Y, Yang X, Zou Y. 2016. Concentrations and potential health risks of methyl tertiary-butyl ether (MTBE) in air and drinking water from Nanning, South China. Sci Total Environ 541:1348–1354. doi: 10.1016/j.scitotenv.2015.10.038 [DOI] [PubMed] [Google Scholar]
- 11. Hanson JR, Ackerman CE, Scow KM. 1999. Biodegradation of methyl tert-butyl ether by a bacterial pure culture. Appl Environ Microbiol 65:4788–4792. doi: 10.1128/AEM.65.11.4788-4792.1999 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Streger SH, Vainberg S, Dong H, Hatzinger PB. 2002. Enhancing transport of Hydrogenophaga flava ENV735 for bioaugmentation of aquifers contaminated with methyl tert-butyl ether. Appl Environ Microbiol 68:5571–5579. doi: 10.1128/AEM.68.11.5571-5579.2002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Smith CA, O’Reilly KT, Hyman MR. 2003. Characterization of the initial reactions during the cometabolic oxidation of methyl tert-butyl ether by propane-grown Mycobacterium vaccae JOB5. Appl Environ Microbiol 69:796–804. doi: 10.1128/AEM.69.2.796-804.2003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Müller RH, Rohwerder T, Harms H. 2008. Degradation of fuel oxygenates and their main intermediates by Aquincola tertiaricarbonis L108. Microbiology 154:1414–1421. doi: 10.1099/mic.0.2007/014159-0 [DOI] [PubMed] [Google Scholar]
- 15. Smith CA, Hyman MR. 2004. Oxidation of methyl tert-butyl ether by alkane hydroxylase in dicyclopropylketone-induced and n-octane-grown Pseudomonas putida Gpo1. Appl Environ Microbiol 70:4544–4550. doi: 10.1128/AEM.70.8.4544-4550.2004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Vainberg S, McClay K, Masuda H, Root D, Condee C, Zylstra GJ, Steffan RJ. 2006. Biodegradation of ether pollutants by Pseudonocardia sp. strain ENV478. Appl Environ Microbiol 72:5218–5224. doi: 10.1128/AEM.00160-06 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Lee EH, Cho KS. 2009. Effect of substrate interaction on the degradation of methyl tert-butyl ether, benzene, toluene, ethylbenzene, and xylene by Rhodococcus sp. J Hazard Mater 167:669–674. doi: 10.1016/j.jhazmat.2009.01.035 [DOI] [PubMed] [Google Scholar]
- 18. Tupa PR, Masuda H. 2018. Draft genome sequence of a propanotroph, Rhodococcus sp. strain ENV425, capable of degrading methyl tert-butyl ether and N-nitrosodimethylamine. Genome Announc 6:e00051-18. doi: 10.1128/genomeA.00051-18 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Purswani J, Pozo C, Rodríguez-Díaz M, González-López J. 2008. Selection and identification of bacterial strains with methyl-tert-butyl ether, ethyl-tert-butyl ether, and tert-amyl methyl ether degrading capacities. Environ Toxicol Chem 27:2296–2303. doi: 10.1897/08-096.1 [DOI] [PubMed] [Google Scholar]
- 20. Müller RH, Rohwerder T, Harms H. 2007. Carbon conversion efficiency and limits of productive bacterial degradation of methyl tert-butyl ether and related compounds. Appl Environ Microbiol 73:1783–1791. doi: 10.1128/AEM.01899-06 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Thornton SF, Nicholls HCG, Rolfe SA, Mallinson HEH, Spence MJ. 2020. Biodegradation and fate of ethyl tert-butyl ether (ETBE) in soil and groundwater: a review. J Hazard Mater 391:122046. doi: 10.1016/j.jhazmat.2020.122046 [DOI] [PubMed] [Google Scholar]
- 22. Ferreira NL, Labbé D, Monot F, Fayolle-Guichard F, Greer CW. 2006. Genes involved in the methyl tert-butyl ether (MTBE) metabolic pathway of Mycobacterium austroafricanum IFP 2012. Microbiology 152:1361–1374. doi: 10.1099/mic.0.28585-0 [DOI] [PubMed] [Google Scholar]
- 23. Hyman M. 2013. Biodegradation of gasoline ether oxygenates. Curr Opin Biotechnol 24:443–450. doi: 10.1016/j.copbio.2012.10.005 [DOI] [PubMed] [Google Scholar]
- 24. Lopes Ferreira N, Malandain C, Fayolle-Guichard F. 2006. Enzymes and genes involved in the aerobic biodegradation of methyl tert-butyl ether (MTBE). Appl Microbiol Biotechnol 72:252–262. doi: 10.1007/s00253-006-0494-3 [DOI] [PubMed] [Google Scholar]
- 25. House AJ, Hyman MR. 2010. Effects of gasoline components on MTBE and TBA cometabolism by Mycobacterium austroafricanum JOB5. Biodegradation 21:525–541. doi: 10.1007/s10532-009-9321-8 [DOI] [PubMed] [Google Scholar]
- 26. Hristova KR, Schmidt R, Chakicherla AY, Legler TC, Wu J, Chain PS, Scow KM, Kane SR. 2007. Comparative transcriptome analysis of Methylibium petroleiphilum PM1 exposed to the fuel oxygenates methyl tert-butyl ether and ethanol. Appl Environ Microbiol 73:7347–7357. doi: 10.1128/AEM.01604-07 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Jechalke S, Rosell M, Martínez-Lavanchy PM, Pérez-Leiva P, Rohwerder T, Vogt C, Richnow HH. 2011. Linking low-level stable isotope fractionation to expression of the cytochrome P450 monooxygenase-encoding ethB gene for elucidation of methyl tert-butyl ether biodegradation in aerated treatment pond systems. Appl Environ Microbiol 77:1086–1096. doi: 10.1128/AEM.01698-10 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Schuster J, Purswani J, Breuer U, Pozo C, Harms H, Müller RH, Rohwerder T. 2013. Constitutive expression of the cytochrome P450 EthABCD monooxygenase system enables degradation of synthetic dialkyl ethers in Aquincola tertiaricarbonis L108. Appl Environ Microbiol 79:2321–2327. doi: 10.1128/AEM.03348-12 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Kane SR, Chakicherla AY, Chain PSG, Schmidt R, Shin MW, Legler TC, Scow KM, Larimer FW, Lucas SM, Richardson PM, Hristova KR. 2007. Whole-genome analysis of the methyl tert-butyl ether-degrading beta-proteobacterium Methylibium petroleiphilum PM1 . J Bacteriol 189:1931–1945. doi: 10.1128/JB.01259-06 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Schmidt R, Battaglia V, Scow K, Kane S, Hristova KR. 2008. Involvement of a novel enzyme, MdpA, in methyl tert-butyl ether degradation in Methylibium petroleiphilum PM1. Appl Environ Microbiol 74:6631–6638. doi: 10.1128/AEM.01192-08 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Bravo AL, Sigala JC, Le Borgne S, Morales M. 2015. Expression of an alkane monooxygenase (alkB) gene and methyl tert-butyl ether co-metabolic oxidation in Pseudomonas citronellolis. Biotechnol Lett 37:807–814. doi: 10.1007/s10529-014-1737-5 [DOI] [PubMed] [Google Scholar]
- 32. Hannemann F, Bichet A, Ewen KM, Bernhardt R. 2007. Cytochrome P450 systems--biological variations of electron transport chains. Biochim Biophys Acta 1770:330–344. doi: 10.1016/j.bbagen.2006.07.017 [DOI] [PubMed] [Google Scholar]
- 33. Johnson EL, Smith CA, O’Reilly KT, Hyman MR. 2004. Induction of methyl tertiary butyl ether (MTBE)-oxidizing activity in Mycobacterium vaccae JOB5 by MTBE. Appl Environ Microbiol 70:1023–1030. doi: 10.1128/AEM.70.2.1023-1030.2004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Johnson NW, Gedalanga PB, Zhao L, Gu B, Mahendra S. 2020. Cometabolic biotransformation of 1,4-dioxane in mixtures with hexavalent chromium using attached and planktonic bacteria. Sci Total Environ 706:135734. doi: 10.1016/j.scitotenv.2019.135734 [DOI] [PubMed] [Google Scholar]
- 35. Myers MA, Johnson NW, Marin EZ, Pornwongthong P, Liu Y, Gedalanga PB, Mahendra S. 2018. Abiotic and bioaugmented granular activated carbon for the treatment of 1,4-dioxane-contaminated water. Environ Pollut 240:916–924. doi: 10.1016/j.envpol.2018.04.011 [DOI] [PubMed] [Google Scholar]
- 36. Sharp JO, Sales CM, Alvarez-Cohen L. 2010. Functional characterization of propane-enhanced N-nitrosodimethylamine degradation by two actinomycetales. Biotechnol Bioeng 107:924–932. doi: 10.1002/bit.22899 [DOI] [PubMed] [Google Scholar]
- 37. Deng D, Pham DN, Li F, Li M, Kelly RM. 2020. Discovery of an inducible toluene monooxygenase that cooxidizes 1,4-dioxane and 1,1-dichloroethylene in propanotrophic Azoarcus sp. strain DD4. Appl Environ Microbiol 86:e01163-20. doi: 10.1128/AEM.01163-20 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Ren H, Wang H, Wang Y, Chen Y, Lu Z, Semrau JD. 2022. A multicomponent THF hydroxylase initiates tetrahydrofuran degradation in Cupriavidus metallidurans ZM02. Appl Environ Microbiol 88:e0188021. doi: 10.1128/AEM.01880-21 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Li F, Deng D, Li M. 2020. Distinct catalytic behaviors between two 1,4-dioxane-degrading monooxygenases: kinetics, inhibition, and substrate range. Environ Sci Technol 54:1898–1908. doi: 10.1021/acs.est.9b05671 [DOI] [PubMed] [Google Scholar]
- 40. Osborne CD, Haritos VS. 2019. Beneath the surface: evolution of methane activity in the bacterial multicomponent monooxygenases. Mol Phylogenet Evol 139:106527. doi: 10.1016/j.ympev.2019.106527 [DOI] [PubMed] [Google Scholar]
- 41. McCormick MS, Lippard SJ. 2011. Analysis of substrate access to active sites in bacterial multicomponent monooxygenase hydroxylases: X-ray crystal structure of xenon-pressurized phenol hydroxylase from Pseudomonas sp. OX1. Biochemistry 50:11058–11069. doi: 10.1021/bi201248b [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Banerjee R, Jones JC, Lipscomb JD. 2019. Soluble methane monooxygenase. Annu Rev Biochem 88:409–431. doi: 10.1146/annurev-biochem-013118-111529 [DOI] [PubMed] [Google Scholar]
- 43. Sazinsky MH, Bard J, Di Donato A, Lippard SJ. 2004. Crystal structure of the toluene/o-xylene monooxygenase hydroxylase from Pseudomonas stutzeri OX1. J Biol Chem 279:30600–30610. doi: 10.1074/jbc.M400710200 [DOI] [PubMed] [Google Scholar]
- 44. Vardar G, Wood TK. 2004. Protein engineering of toluene-o-xylene monooxygenase from Pseudomonas stutzeri OX1 for synthesizing 4-methylresorcinol, methylhydroquinone, and pyrogallol. Appl Environ Microbiol 70:3253–3262. doi: 10.1128/AEM.70.6.3253-3262.2004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Lock M, Nichol T, Murrell JC, Smith TJ. 2017. Mutagenesis and expression of methane monooxygenase to alter regioselectivity with aromatic substrates. FEMS Microbiol Lett 364:fnx137. doi: 10.1093/femsle/fnx137 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Tupec M, Culka M, Machara A, Macháček S, Bím D, Svatoš A, Rulíšek L, Pichová I. 2022. Understanding desaturation/hydroxylation activity of castor stearoyl delta(9)-desaturase through rational mutagenesis. Comput Struct Biotechnol J 20:1378–1388. doi: 10.1016/j.csbj.2022.03.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Kurt C, Sönmez B, Vardar N, Yanık-Yıldırım KC, Vardar-Schara G. 2016. Cavity residue leucine 95 and channel residues glutamine 204, aspartic acid 211, and phenylalanine 269 of toluene o-xylene monooxygenase influence catalysis. Appl Microbiol Biotechnol 100:7599–7609. doi: 10.1007/s00253-016-7658-6 [DOI] [PubMed] [Google Scholar]
- 48. Li F, Deng D, Zeng L, Abrams S, Li M. 2021. Sequential anaerobic and aerobic bioaugmentation for commingled groundwater contamination of trichloroethene and 1,4-dioxane. Sci Total Environ 774:145118. doi: 10.1016/j.scitotenv.2021.145118 [DOI] [PubMed] [Google Scholar]
- 49. Li M, Mathieu J, Yang Y, Fiorenza S, Deng Y, He Z, Zhou J, Alvarez PJJ. 2013. Widespread distribution of soluble di-iron monooxygenase (SDIMO) genes in arctic groundwater impacted by 1,4-dioxane. Environ Sci Technol 47:9950–9958. doi: 10.1021/es402228x [DOI] [PubMed] [Google Scholar]
- 50. Martin KE, Ozsvar J, Coleman NV. 2014. SmoXYB1C1Z of Mycobacterium sp. strain NBB4: a soluble methane monooxygenase (sMMO)-like enzyme, active on C2 to C4 alkanes and alkenes. Appl Environ Microbiol 80:5801–5806. doi: 10.1128/AEM.01338-14 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. He Y, Mathieu J, da Silva MLB, Li M, Alvarez PJJ. 2018. 1,4-dioxane-degrading consortia can be enriched from uncontaminated soils: prevalence of Mycobacterium and soluble di-iron monooxygenase genes. Microb Biotechnol 11:189–198. doi: 10.1111/1751-7915.12850 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52. Auffret M, Labbé D, Thouand G, Greer CW, Fayolle-Guichard F. 2009. Degradation of a mixture of hydrocarbons, gasoline, and diesel oil additives by Rhodococcus aetherivorans and Rhodococcus wratislaviensis. Appl Environ Microbiol 75:7774–7782. doi: 10.1128/AEM.01117-09 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53. Tupa PR, Masuda H. 2018. Genomic analysis of propane metabolism in methyl tert-butyl ether-degrading Mycobacterium sp. strain ENV421. J Genomics 6:24–29. doi: 10.7150/jgen.24929 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54. Wang B, Chu KH. 2017. Cometabolic biodegradation of 1,2,3-trichloropropane by propane-oxidizing bacteria. Chemosphere 168:1494–1497. doi: 10.1016/j.chemosphere.2016.12.007 [DOI] [PubMed] [Google Scholar]
- 55. Hand S, Wang B, Chu KH. 2015. Biodegradation of 1,4-dioxane: effects of enzyme inducers and trichloroethylene. Sci Total Environ 520:154–159. doi: 10.1016/j.scitotenv.2015.03.031 [DOI] [PubMed] [Google Scholar]
- 56. Mahendra S, Alvarez-Cohen L. 2006. Kinetics of 1,4-dioxane biodegradation by monooxygenase-expressing bacteria. Environ Sci Technol 40:5435–5442. doi: 10.1021/es060714v [DOI] [PubMed] [Google Scholar]
- 57. Champreda V, Zhou NY, Leak DJ. 2004. Heterologous expression of alkene monooxygenase components from Xanthobacter autotrophicus Py2 and reconstitution of the active complex. FEMS Microbiol Lett 239:309–318. doi: 10.1016/j.femsle.2004.09.002 [DOI] [PubMed] [Google Scholar]
- 58. Sims LP, Lockwood CWJ, Crombie AT, Bradley JM, Le Brun NE, Murrell JC. 2022. Purification and characterization of the isoprene monooxygenase from Rhodococcus sp. strain AD45. Appl Environ Microbiol 88:e0002922. doi: 10.1128/aem.00029-22 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59. McCarl V, Somerville MV, Ly M-A, Henry R, Liew EF, Wilson NL, Holmes AJ, Coleman NV. 2018. Heterologous expression of Mycobacterium alkene monooxygenases in gram-positive and gram-negative bacterial hosts. Appl Environ Microbiol 84:e00397-18. doi: 10.1128/AEM.00397-18 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60. Cafaro V, Scognamiglio R, Viggiani A, Izzo V, Passaro I, Notomista E, Piaz FD, Amoresano A, Casbarra A, Pucci P, Di Donato A. 2002. Expression and purification of the recombinant subunits of toluene/o-xylene monooxygenase and reconstitution of the active complex. Eur J Biochem 269:5689–5699. doi: 10.1046/j.1432-1033.2002.03281.x [DOI] [PubMed] [Google Scholar]
- 61. Wang W, Liang Y, Liu L, Han S, Wu S, Gao H. 2022. Pleiotropic effects of Hfq on the cytochrome c content and pyomelanin production in Shewanella oneidensis. Appl Environ Microbiol 88:e0128922. doi: 10.1128/aem.01289-22 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
The file "Supporting information" includes 8 figures and 3 tables as Supplemental figures and tables.
Data Availability Statement
The genome sequence of M. vaccae JOB5 has been deposited in the SRA database and GenBank under accession numbers SRR23703833 and CP126901 to CP126903, respectively. The transcriptomic data for strain JOB5 growing on different substrates have been deposited in the SRA database under accession numbers SRR23725360 to SRR23725371.






