Gene mdpC plays a regulatory role in the methyl-tert-butyl ether degradation pathway of Methylibium petroleiphilum strain PM1

Abstract

Among the few bacteria known to utilize methyl tert-butyl ether (MTBE) as a sole carbon source, Methylibium petroleiphilum PM1 is a well-characterized organism with a sequenced genome; however, knowledge of the genetic regulation of its MTBE degradation pathway is limited. We investigated the role of a putative transcriptional activator gene, mdpC, in the induction of MTBE-degradation genes mdpA (encoding MTBE monooxygenase) and mdpJ (encoding tert-butyl alcohol hydroxylase) of strain PM1 in a gene-knockout mutant mdpC⁻. We also utilized quantitative reverse transcriptase PCR assays targeting genes mdpA, mdpJ and mdpC to determine the effects of the mutation on transcription of these genes. Our results indicate that gene mdpC is involved in the induction of both mdpA and mdpJ in response to MTBE and tert-butyl alcohol (TBA) exposure in PM1. An additional independent mechanism may be involved in the induction of mdpJ in the presence of TBA.

This study utilizes mutant analysis to underline the importance of a regulatory gene in a bacterial pathway involved in the biodegradation of methyl-tert-butyl ether, a groundwater pollutant in the United States.

Graphical Abstract Figure.

This study utilizes mutant analysis to underline the importance of a regulatory gene in a bacterial pathway involved in the biodegradation of methyl-tert-butyl ether, a groundwater pollutant in the United States.

INTRODUCTION

Methyl-tert-butyl ether (MTBE) was used as a fuel oxygenate in the United States and Europe for almost two decades. In California, it was banned from use in 2003 but has left a legacy of groundwater contamination in all parts of the state (Richardson and Ternes 2005) and throughout the United States (Hatzinger et al. 2001). One of the biodegradation products of MTBE is tertiary butyl alcohol (TBA), a potential carcinogen.

Several bacterial species and strains can mineralize MTBE and TBA including Methylibium petroleiphilum PM1 (Hanson, Ackerman and Scow 1999; Nakatsu et al. 2006), Methylibium sp. R8 (Rosell et al. 2007), Hydrogenophaga flava ENV735 (Hatzinger et al. 2001), Mycobacterium austroafricanum IFP2012 and IFP2015 (Francois et al. 2002, 2003; Lopes Ferreira et al. 2006), Variovorax paradoxus CL-8 (Zaitsev, Uotila and Häggblom 2007) and Aquincola tertiaricarbonis L108 (Lechner et al. 2007).

The involvement and efficacy of strain PM1 in aerobic biodegradation of MTBE has been demonstrated in several field studies (Wilson, Mackay and Scow 2002; Smith et al. 2005; Hicks et al. 2014). PM1-like bacteria have also been found in several MTBE-contaminated groundwater aquifers in California and play a significant role in the biodegradation process (Kane et al. 2001; Hristova et al. 2003; North et al. 2012a,b). The MTBE-degradation pathway of strain PM1 consists of genes present on an extrachromosomal megaplasmid that are likely controlled as a regulon; genes mdpA and mdpJ have been proposed to play a role in the degradation of MTBE and TBA, respectively (Hristova et al. 2007). Mutagenesis studies indicate that the gene mdpA, encoding MTBE monoxygenase, is involved in the first step of the degradation pathway, converting MTBE into hydroxymethyl-tert-butyl ether (Schmidt et al. 2008). Real-time qPCR assays targeting genes mdpA and mdpJ (encoding TBA hydroxylase) indicated the induction of transcription of these genes in response to MTBE and TBA (G. Joshi, unpublished data).

A putative ATP-dependent transcriptional regulator has been identified within the MTBE-degradation gene cluster, designated Mpe_B0601 or mdpC (Kane et al. 2007). To examine the role of mdpC as a regulator of the MTBE-degradation pathway of strain PM1, we developed an mdpC insertion-mutant strain of PM1 and designed an RT-qPCR assay to quantify mdpC transcription in the presence or absence of MTBE and TBA.

MATERIALS AND METHODS

Culture growth and preparation

Methylibium petroleiphilum PM1 cultures were routinely grown in 0.33X tryptic soy broth (TSB) at 28°C with rotary shaking at 150 rpm or on 0.33X TSB agar at 28°C. When required, antibiotics were used in the following final concentrations: kanamycin (Km), 50 μg mL⁻¹; streptomycin (Sm), 50 μg mL⁻¹; spectinomycin (Spm) 50 μg mL⁻¹; ampicillin (Ap) 100 μg mL⁻¹. For experiments examining gene transcription, cultures were grown in mineral salts medium (MSM) (Schmidt et al. 2008) supplemented with 500 mg L⁻¹ pyruvate and harvested at mid-log phase (OD_595nm = 0.25 – 0.4). Escherichia coli DH5α cells (Life Technologies, Grand Island, NY, USA) were used for all transformations that involved vectors carrying Sm resistance. For all other transformations, E. coli TOP10 (Life Technologies) cells were used. All E. coli cultures were grown on Luria–Bertani (LB) agar at 37°C.

Construction of mdpC knockout strain of M. petroleiphilum PM1

A primer set (mdpC-F 5^′-GCAGGTGAGCAACAACCTCT-3^′ and mdpC-R 5^′-GACTTGTCCCGACTGCTCAT-3^′) was designed to amplify mdpC from PM1 genomic DNA. The PCR product was cloned into pCR-XL-TOPO (Life Technologies,) using the manufacturer's protocol. Briefly, clones were screened by PCR using kit-supplied primers, confirmed by sequencing, and subsequently used for in vitro mutagenesis. Equimolar amounts of a correct clone (pGJ001) and EZ-Tn5<SmQ> (Schmidt et al. 2008) were mixed together with transposase and incubated to disrupt mdpC. The reaction was stopped and DNA was transformed into E. coli DH5α cells. Transformants were selected on LB agar containing 50 μg mL⁻¹ Km and Sm each. Transposon inserts were screened by PCR using mdpC-specific primers. The exact site of insertion of likely candidates was determined by sequencing. Selected plasmid (pGJ001-SmQ) was extracted and the mdpC-SmQ fragment amplified using primers supplied by manufacturer. The purified amplicon (0.6 pmol; Qiagen PCR purification kit; Qiagen) was transformed into PM1 cells washed in 10% glycerol by electroporation in MicroPulser Electroporator (Bio-Rad Laboratories, Hercules, CA, USA) at 1.8 kV for 5 ms. Transformants were selected on 0.33X TSB agar containing 50 μg mL⁻¹ Sm and Spm each. Potential mutants were screened by PCR and confirmed by sequencing of the PCR product.

Resting-cell experiments

One liter of pyruvate grown PM1 cultures were harvested by centrifugation at 6000 rpm for 10 min, washed twice in MSM with no carbon source and finally resuspended in 10 mL MSM supplemented with 50 mg L⁻¹ MTBE, or TBA providing an equivalent amount of carbon. Experiments were conducted in 50 mL sterile glass bottles fitted with Teflon-lined mininert valve caps (Restek Corporation, Bellefonte, PA, USA). Microcosm bottles with MSM and appropriate amounts of carbon source were incubated in dark at 28°C, 150 rpm for 3 h prior to addition of washed PM1 cells in order to facilitate equilibration of aqueous partitioning of organic compounds. Final volume in microcosm bottles was 50 mL and OD_595nm was 0.9–1.0. Microcosms were incubated for up to 48 h. Samples for RNA extraction (1 mL) and for analysis by gas chromatography and total protein analysis were collected aseptically using Micro-Mate Glass Syringes fitted with 20G, 6-inch stainless steel deflected point septum penetration needles with luer hub (Cadence, Inc., Staunton, VA, USA). Samples for RNA extraction were preserved with RNAProtect Bacteria Reagent (Qiagen) according to manufacturer's instructions and stored at −70°C no longer than 2 weeks prior to extraction. Samples for gas chromatography were stored in 10-mL headspace vials preserved with sodium phosphate tribasic dodecahydrate at a concentration of 1% by weight and sealed with 20-mm Teflon-lined septa and aluminum crimp caps. Samples were stored no longer than a week at 4°C prior to analysis. Abiotic controls for each microcosm were also set up, with no PM1 cells added and were sampled for analysis by GC.

RNA extraction, reverse transcription and real-time quantitative PCR

RNA was extracted from preserved cell pellets using the RNeasy Mini Kit (Qiagen), and DNA was removed using the Ambion TURBO DNA-free Kit (Life Technologies). RNA was quantified using the Qubit RNA Assay Kit (Life Technologies) and converted to single-stranded cDNA using the SuperScript III First-Strand Synthesis SuperMix for qRT-PCR (Life Technologies). cDNA (8 ng per reaction) was used as template for the qPCR assays designed to detect PM1 mdpA, mdpJ and mdpC transcripts. In addition, eight housekeeping genes were also quantified to serve as internal standards using the geNorm approach (Vandesompele et al. 2002). Primers were designed using Primer3 (Koressaar and Remm 2007; Untergrasser et al. 2012) and are described in Table 1. Briefly, 0.5 mM of each primer was used in 25 μL PCR reactions prepared with SYBR GreenER™ qPCR SuperMix for ABI PRISM (Life Technologies, Grand Island, NY, USA) in MicroAmp optical 96-well reaction plates and run on a 7300 Real-Time PCR System (Applied Biosystems, Carlsbad, CA, USA).

Table 1.

RT-qPCR assays and primers used in this study. All primers are specific to genes in M. petroleiphilum PM1.

Primer name	Targeted gene	Function in assay	Sequence (5′ – 3′)	PCR conditions
				Initial denaturation	Denaturation	Annealing	Cycles
mdpA 211F	MTBE monooxygenase	Gene of interest	TCATCATCGGGATGTTGCTA	95°C, 10 min	95°C, 15 s	58°C, 1 min	40
mdpA 211R			ACCGTAGAAGACCAGCGAGA
mdpJ 245F	TBA hydroxylase	Gene of interest	TCTCCAATGTCTTCGACTGC	95°C, 10 min	95°C, 15 s	58°C, 1 min	40
mdpJ 245R			GATTCGGATCCAGACTTCGT
qmdpC F	Putative transcriptional activator	Gene of interest	AGCCCTCAATCAATCGGTAA	95°C, 10 min	95°C, 15 s	63°C, 1 min	40
qmdpC R			AAAAGAGCGATCCAAAGACG
pykF F	Pyruvate kinase	Internal standard	GAGCTTCCAGTGCGAGTACC	95°C, 10 min	95°C, 15 s	58°C, 1 min	40
pykF R			TCGCTAGCCTTGAGGATCTG
gyrA F	DNA gyrase subunit A	Internal standard	GACAAGAGCTGGGATTCGTC	95°C, 10 min	95°C, 15 s	58°C, 1 min	40
gyrA R			CGCCATCACCTCCTTGTATT
gyrB F	DNA gyrase subunit B	Internal standard	AATACAACCCCGACAAGCTG	95°C, 10 min	95°C, 15 s	56°C, 1 min	40
gyrB R			CCGTCCTTCAGGTACTGCTC
glyA F	Serine hydroxymethyl transferase	Internal standard	CAGCAACAGGTGGTGAAGAA	95°C, 10 min	95°C, 15 s	62°C, 1 min	40
glyA R			TCTTGTTGCAGGTCATGTGC
rho F	RNA polymerase rho subunit	Internal standard	GGGCTGAGGTAGATGTCGTC	95°C, 10 min	95°C, 15 s	58°C, 1 min	40
rho R			CCTTGGAGATCGAGAACAGC
gap1 F	glyceraldehyde-3-phosphate dehydrogenase	Internal standard	CAAGAAGCACGACATCCAGA	95°C, 10 min	95°C, 15 s	62°C, 1 min	40
gap1 R			TTCCTTGGTCGTGAAGAAGC
gap2 F	- ′′ -	Internal standard	TGAGTTCGGGGTAGATCAGC	95°C, 10 min	95°C, 15 s	58°C, 1 min	40
gap2 R			TGAGTTCGGGGTAGATCAGC
gap3 F	- ′′ -	Internal standard	TGAGTTCGGGGTAGATCAGC	95°C, 10 min	95°C, 15 s	58°C, 1 min	40
gap3 R			GGCAGTTGGTGGTACAGGAG

Gas chromatography

MTBE and TBA were quantified on an Agilent 6890N gas chromatograph equipped with a flame ionization detector and an HP 7694 headspace autosampler as described previously (Schmidt et al. 2008). Organic compounds were separated using an Agilent HP1 capillary column (60 m by 1 μm by 0.320 μm).

Total protein analysis

Total protein content was determined using 1 mL samples collected at the beginning and end of incubations. Total protein was extracted with BugBuster Protein Extraction Reagent containing Benzonase Nuclease (EMD Millipore, Merck KGaA, Darmstadt, Germany) and was quantified using the Bio-Rad Protein Assay Kit (Bio-Rad Laboratories) according to manufacturer's protocols.

Data analysis

Bacterial gene expression data were normalized by dividing gene quantity values determined by RT-qPCR with the geometric mean of gene quantities of the two (or three) most stable housekeeping genes as determined by the geNorm package for Microsoft Excel (Vandesompele et al. 2002). The output from the gas chromatograph was analyzed using ChemStation revision A.10.02 software (Agilent, Santa Clara, CA, USA).

Sequence analysis and generation of phylogenetic trees

Functional homologs of translated M. petroleiphilum PM1 coding sequence were identified using BLASTP searches against the Swiss-prot database. Multiple alignments were performed using ClustalW. The evolutionary history was inferred using the neighbor-joining method (Saitou and Nei 1987) with bootstrap analysis (N = 2000) (Felsenstein 1985). The evolutionary distances were computed using the Poisson correction method (Zuckerkandl and Pauling 1965) in MEGA5 (Tamura et al. 2011).

RESULTS AND DISCUSSION

MdpC protein

MdpC is a 901 amino-acid long, predicted ATP-dependent transcriptional activator in the MalT family. Two conserved domains were identified: an AAA_16 ATPase domain between residues 28 and 149, and a LuxR DNA-binding region between residues 839 and 891. Among the top BLAST identity matches for MdpC were protein kinase PknK of M. bovis AF2122/97 (26%), maltose-binding regulatory protein MalT of E. coli K-12 (25%), alkane-responsive regulatory protein AlkS of Alcanivorax borkumensis SK2 (38%) and AlkS of Pseudomonas putida Gpo1 (20%). These proteins belong to the MalT family of ATP-dependent transcriptional activators. A phylogenetic tree of the top BLAST hits (Fig. 1a) shows deep branching of MdpC.

Figure 1.

Phylogenetic tree of M. petroleiphilum strain PM1 MdpC and related ATP-dependent transcriptional activators obtained through the Swiss-prot database (a). The predicted transcriptional activator protein MdpC is distinct from the MalT of E. coli and other enteric bacteria and AlkS of Pseudomonas and Alcanivorax, and Mycobacterium PknK. Phylogenetic tree of luxR DNA-binding domains (b) and AAA_16 ATPase domains (c) of M. petroleiphilum MdpC and related proteins obtained through the Swiss-prot database. The predicted DNA-binding and ATPase domains of MdpC form a distinct lineage separate from related proteins. Evolutionary distances represented by branch lengths are in the units of the number of amino-acid substitutions per site.

Regions of MdpC have 23% (amino-acid positions 182–430) and 44% (amino-acid positions 838–899) identity with corresponding regions (positions 187–421 and 811–872, respectively) of AlkS from A. borkumensis SK2 (Hara et al. 2004; Reva et al. 2008). The AlkS of A. borkumensis SK2 has 48% similar amino-acid identity with AlkS of P. putida, which has been grouped within the MalT subfamily of LuxR transcriptional activators (Danot 2001; Reva et al. 2008). AlkS is known to be the transcriptional activator involved in the alkB1GHJ cluster for alkane degradation in A. borkumensis AP1 (van Beilen et al. 2004). mdpA of PM1 (encoding MTBE monooxygenase) is 69 and 66% identical to the alkane monooxygenase AlkB of A. borkumensis AP1 and P. putida GPo1 (Hristova et al. 2007). Phylogenetic trees generated from the alignments of the two predicted functional regions in MdpC with those in related proteins also show distinct branching (Fig. 1b and c).

The phosphate-binding Walker A type GYGKS motif typical of AAA_16 type ATPases at position 39–43 of MdpC is also present (as G[Y/F]GK[S/T]) in AAA_16 regions of P. putida AlkS (positions 54–58), E. coli MalT (positions 42–46) and M. tuberculosis PknK (positions 371–375). MpdC also has a DNA-binding, helix-turn-helix LuxR-type domain at position 839–891. Most luxR-type regulators act as transcription activators, but some can be repressors or have a dual role for different functional domains. The presence of this domain indicates potential role of MdpC in transcription of MTBE-degradation genes.

Effect of chloramphenicol on transcription of PM1 MTBE-degradation genes and mdpC

Chloramphenicol exposure was employed to test if inhibition of de novo protein synthesis affected the induction of genes mdpA and mdpJ. In the presence of MTBE, mdpA and mdpJ were induced to 7- and 42-fold of their initial levels, while mdpC was expressed at a constant level throughout the course of the incubation. In the presence of chloramphenicol, transcription of mdpA, mdpJ and mdpC stayed constant throughout the incubation, up to 12 h (Fig. 2a–c). MTBE was degraded by 84.3% (±10.5%) by untreated cells, whereas 38.4% (±20.9%) MTBE was degraded in chloramphenicol-treated PM1 cells after 24 h. There were no significant differences in the total protein content among the different treatments. In the presence of pyruvate alone, none of the genes under study were induced (data not shown).

Figure 2.

Effect of chloramphenicol on the transcription of PM1 MTBE-degradation genes. Induction of mdpA (a) and mdpJ (b) transcription occurred in wild-type PM1 treated with MTBE (⧫, ▪ and •) but not in cells treated with 15 μg mL^–1 chloramphenicol (◊, □ and ◯) following exposure to 50 mg L^–1 MTBE. There was slightly elevated transcription of mdpC (c) in the absence of chloramphenicol (MTBE alone) but was not statistically significant (P = 0.33).

Non-induction of mdpC in response to MTBE in strain PM1 is in contrast with AlkS of P. putida Gpo1, in which the mRNA associated with alkS increases in response to alkanes (Canosa, Yuste and Rojo 1999). In the absence of alkanes, AlkS represses P_alkS1. In the presence of alkanes, AlkS activates P_alkS2, which is a strong promoter that induces more AlkS formation, and also, P_alkB, which turns on a cascade of alkane-degradation genes (Canosa et al. 2000). Unlike AlkS, MdpC does not appear to be a self-inducer in response to the substrate MTBE.

In the presence of chloramphenicol, which is an inhibitor of prokaryotic protein synthesis, the number of transcripts of mdpA and mdpJ were reduced in wild-type PM1 cells. There were low levels of mdpA, mdpJ and mdpC transcripts in the presence of chloramphenicol, and these levels were similar to those measured in the presence of pyruvate (non-inducer) at time 0. A previous study has shown a low rate of MTBE degradation in PM1 cells treated with chloramphenicol (Schmidt et al. 2008). These results indicate that there is a low background level of mRNA of these genes in PM1 cells, and post-MTBE exposure, elevated transcription (induction) of mdpA and mdpJ.

Impact of mdpC knockout on the transcription of MTBE-degradation genes

There was no significant change (or induction) in the transcription of mdpA, mdpJ and mdpC from the initial levels after 12 h of MTBE exposure in the mutant strain, while in the wild-type, transcription of mdpA was induced 13-fold and mdpJ was induced ∼66-fold after 12 h of exposure to MTBE. A similar trend was also observed upon exposing cells to TBA (Fig. 3a–c).

Figure 3.

mdpC transcription in wild type and mdpC-knockout PM1 strains. Induction of mdpA (a) and mdpJ (b) transcription occurred in wild-type PM1 (⧫, ▪ and •) but not in the mdpC-knockout mutant (◊, □ and ◯) following exposure to 50 mg L^–1 MTBE (top row) and 52.4 mg L^–1 TBA (bottom row). Differences in mdpC transcription (c) observed in wild-type versus mutant PM1, or over 0–12h for each strain were not statistically significant (P > 0.05).

mdpJ transcription increased 61(±27)-fold (P = 0.0176) in 12 h of TBA exposure in mdpC⁻ cells, and 27.4 (±2.3)-fold (P = 0.0001) in wild-type cells. However, total number of mdpJ transcripts in mdpC⁻ cells was still 45 (±11.3)-fold lower (P = 0.0024) than in wild-type cells at the beginning of the incubation, and 19.9 (±0.97)-fold lower (P = 0.0000) after 12 h (Fig. 3b), indicating that any level of induction in mutant was not as high as that in wild type. The rate of MTBE degradation was greater in wild-type cells than mdpC⁻ mutant after 48 h, whereas the rates of TBA degradation among the two strains were comparable (Table 2). There were no significant differences in the total protein content of the cells among the different treatments.

Table 2.

Degradation rates of carbon source by different PM1 strains. Concentrations were measured for 48 h.

	MTBE	TBA
Wild-type	0.036 h−1	0.01 h−1
mdpC− mutant	0.013 h−1*	0.017 h−1

*Value was not statistically different from abiotic control (P = 0.52).

The induction of both mdpA and mdpJ in the presence of MTBE in the wild type but not in the mutant is suggesting a direct mdpC role in transcriptional activation of both genes. There is a large gap of ∼54 kb between these two genes in the megaplasmid and they are not part of a single operon. The induction of mdpJ in mutant cells from its initial level in the presence of TBA was lower than that observed in wild type. Therefore, a second, though less efficient, regulatory mechanism independent of mdpC is possibly involved in the induction of mdpJ. Alternatively, the limited mdpJ induction in the presence of chloramphenicol could be due to a broad regulatory response such as cellular shock response.

The role of a regulatory protein in bacterial MTBE degradation has not been studied prior to our work. Our results provide direct evidence for the role of mdpC in the regulation of the MTBE-degradation pathway in strain PM1. This regulatory role is via MdpC, a transcriptional activator with MalT-like ATPase and LuxR DNA-binding domains, but overall limited homology with similar proteins in other bacteria. Future studies to address questions such as interaction of overexpressed MdpC with specific promoter regions of the MTBE-gene cluster in PM1 megaplasmid and effect of known repressors of MTBE-degradation in PM1, such as ethylbenzene (G. Joshi, unpublished data), on expression of mdpA and mdpJ in the mdpC⁻ mutant, will provide a detailed understanding of the mechanism of regulation of the pathway via MdpC. Additionally, it will be of interest to determine if MdpC binds directly to MTBE, or to other molecules, or a protein intermediate, in order to regulate MTBE degradation in strain PM1, through substrate binding studies with overexpressed MdpC. A deeper understanding of regulatory mechanism of the MTBE-degradation pathway of strain PM1 could aid the development of novel approaches for detection and bioremediation of these chemicals.

FUNDING

This project was supported in part by (i) grant number T32-GM008799 from NIGMS-NIH, (ii) National Institute of Environmental Health Sciences of the National Institutes of Health under award number P42ES004699 and (iii) by an industry/campus supported fellowship under the Training Program in Biomolecular Technology (T32-GM008799) at the University of California, Davis. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIGMS and NIH.

Conflict of interest statement. None declared.