JB
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental material
Right arrow An erratum has been published
Right arrow Other Versions of this Article:
JB.01259-06v1
189/5/1931    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kane, S. R.
Right arrow Articles by Hristova, K. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kane, S. R.
Right arrow Articles by Hristova, K. R.
Journal of Bacteriology, March 2007, p. 1931-1945, Vol. 189, No. 5
0021-9193/07/$08.00+0     doi:10.1128/JB.01259-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Whole-Genome Analysis of the Methyl tert-Butyl Ether-Degrading Beta-Proteobacterium Methylibium petroleiphilum PM1{triangledown} ,{dagger}

Staci R. Kane,1,{ddagger}* Anu Y. Chakicherla,1,{ddagger} Patrick S. G. Chain,1,4 Radomir Schmidt,2 Maria W. Shin,1 Tina C. Legler,1 Kate M. Scow,2 Frank W. Larimer,3,4 Susan M. Lucas,4 Paul M. Richardson,4 and Krassimira R. Hristova2

Lawrence Livermore National Laboratory, Livermore, California,1 Department of Land Air and Water Resources, University of California, Davis, California,2 Genome Analysis Group, Oak Ridge National Laboratory, Oak Ridge, Tennessee,3 Joint Genome Institute Production Genomics Facility, Walnut Creek, California4

Received 10 August 2006/ Accepted 29 November 2006


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results and Discussion
 References
 
Methylibium petroleiphilum PM1 is a methylotroph distinguished by its ability to completely metabolize the fuel oxygenate methyl tert-butyl ether (MTBE). Strain PM1 also degrades aromatic (benzene, toluene, and xylene) and straight-chain (C5 to C12) hydrocarbons present in petroleum products. Whole-genome analysis of PM1 revealed an ~4-Mb circular chromosome and an ~600-kb megaplasmid, containing 3,831 and 646 genes, respectively. Aromatic hydrocarbon and alkane degradation, metal resistance, and methylotrophy are encoded on the chromosome. The megaplasmid contains an unusual t-RNA island, numerous insertion sequences, and large repeated elements, including a 40-kb region also present on the chromosome and a 29-kb tandem repeat encoding phosphonate transport and cobalamin biosynthesis. The megaplasmid also codes for alkane degradation and was shown to play an essential role in MTBE degradation through plasmid-curing experiments. Discrepancies between the insertion sequence element distribution patterns, the distributions of best BLASTP hits among major phylogenetic groups, and the G+C contents of the chromosome (69.2%) and plasmid (66%), together with comparative genome hybridization experiments, suggest that the plasmid was recently acquired and apparently carries the genetic information responsible for PM1's ability to degrade MTBE. Comparative genomic hybridization analysis with two PM1-like MTBE-degrading environmental isolates (~99% identical 16S rRNA gene sequences) showed that the plasmid was highly conserved (ca. 99% identical), whereas the chromosomes were too diverse to conduct resequencing analysis. PM1's genome sequence provides a foundation for investigating MTBE biodegradation and exploring the genetic regulation of multiple biodegradation pathways in M. petroleiphilum and other MTBE-degrading beta-proteobacteria.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results and Discussion
 References
 
Methylibium petroleiphilum strain PM1, which belongs to a newly described genus and species (57), is a motile bacterium belonging to the Comamonadaceae family of the Betaproteobacteria and is an important member of subsurface microbial communities in many gasoline-contaminated aquifers. Furthermore, PM1 is a methylotroph that can grow aerobically on the fuel oxygenate methyl tert-butyl ether (MTBE) and oxidize it completely to carbon dioxide (9, 34). MTBE is a suspected carcinogen that has contaminated drinking water wells throughout the United States due to the preponderance of underground leaking storage tanks, the widespread usage of MTBE, and its recalcitrance and mobility in groundwater. PM1 can also oxidize aromatic hydrocarbons (toluene, benzene, o-xylene, and phenol) (20) and n-alkanes (C5 to C12) (57; K. Hristova, unpublished data) and has been used in two bioaugmentation field trials in gasoline-contaminated aquifers in California (67) and Montana (18, 73). In contaminated sites amended with oxygen, in situ MTBE degradation was observed and corresponded to increases in native populations of Methylibium sp. (~99% similarity to PM1, based on the 16S rRNA gene) (38, 67, 82). PM1-like bacteria occur naturally in a number of MTBE-contaminated aquifers in the United States (46, 47, 82), Mexico (21), and Europe (55, 61), and their presence has been correlated with MTBE degradation activity in numerous sites (47, 67, 82), using real-time PCR analysis (37). These results suggest that PM1-like organisms may play a major role in MTBE biodegradation under aerobic conditions in contaminated aquifers. The genetic basis for MTBE metabolism is not currently understood, although there is general agreement that the initial enzymatic steps are similar to cometabolic degradation pathways (27, 68, 74), and recent reports have described genes involved in degradation of the MTBE downstream metabolites 2-methyl-1,2-propanediol (52) and 2-hydroxyisobutyrate (61). The complex regulation of the metabolism of fuel hydrocarbons and MTBE, often occurring in mixtures, is relatively unknown, while limited studies showed that MTBE degradation could be inhibited in mixtures with BTEX (benzene, toluene, ethylbenzene, xylene) compounds (20, 47).

In this paper, we present our analysis of the whole genome sequence of M. petroleiphilum PM1. We present comparative sequence analysis results between PM1 and bacteria with homologous individual genes and operons as well as comparative whole-genome hybridization analysis between PM1 and PM1-like MTBE-degrading isolates (~99% identical 16S rRNA gene sequences) from gasoline-contaminated sites. General genome features are discussed, including interesting repeated elements as well as genes and operons involved in methylotrophy, degradation of aromatic hydrocarbons, degradation of cyclic and straight-chain alkanes, cofactor biosynthesis, motility, secretion, and heavy metal resistance and transport. A noteworthy finding was the presence of a large, ~600-bp plasmid in PM1 that was highly conserved among PM1-like bacteria. Furthermore, plasmid-curing experiments showed that the plasmid was essential for MTBE and tert-butyl alcohol (TBA) biodegradation in PM1. The PM1 genome sequence has provided a foundation for understanding novel pathways and interactions in this important subsurface bacterium as well as those in phylogenetically similar MTBE-degrading bacteria.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results and Discussion
 References
 
Bacterial strains used in genome sequence and comparative hybridization analyses. Methylibium petroleiphilum strain PM1 was used for whole-genome sequencing at the Joint Genome Institute (Walnut Creek, CA). Strain PM1 was isolated from a sewage treatment plant biofilter used for treating discharge from oil refineries (9, 34). Two MTBE-degrading bacterial pure cultures (MG4 and 312) were obtained from two different gasoline-contaminated aquifers in Northern California (47) (Travis Air Force Base, Travis, CA, and San Mateo, CA, respectively). Enrichment culturing was conducted in 10 mg/liter MTBE mineral salts media (MSM) (56) with shaking at 150 rpm at room temperature. Enrichment cultures were plated onto 0.1x Trypticase soy agar (TSA), and individual colonies were picked, grown in MSM with 10 mg/liter MTBE, and analyzed for MTBE degradation activity using purge-and-trap gas chromatography-mass spectrometry, with reference to d12-MTBE as an internal standard (46). Culture purity was tested by plating (0.1x TSA) and 16S rRNA gene sequence analysis of colony genomic DNA.

Sequencing, gene prediction, and annotation. Genomic DNA was isolated and purified from M. petroleiphilum PM1, and whole-genome shotgun libraries (3-kb, 8-kb, and 40-kb DNA inserts) were constructed and sequenced as previously described (14). After quality control of the 90,327 total initial reads of draft sequence, 83,180 sequences were assembled, producing an average of 10.7-fold coverage across the genome. The whole genome sequence was assembled using the Phred/Phrap/Consed package (P. Green, University of Washington) (25, 26, 32). The reads were assembled into 24 high-quality draft sequence contigs, which were linked into three larger scaffolds by paired-end sequence information. Gaps in the sequence were closed either by walking on gap-spanning clones or with PCR products generated from genomic DNA. Physical (uncaptured) gaps were closed by combinatorial (multiplex) PCR. Sequence finishing and polishing added 308 reads, and the final assessment of the genome assembly was completed as described previously (14). The final genome assembly quality of PM1 adheres to conventional standards of less than one error per 10,000 bp. Each base is covered by at least two quality sequences, with an average of 10.7-fold coverage. Proper assembly was verified by fosmid coverage coupled with PCR data. Gene modeling and genome annotation were performed as previously described (14) to identify open reading frames likely encoding proteins (coding sequences [CDSs]).

Comparative genomics. Orthologs and CDSs unique to M. petroleiphilum PM1 were identified using the Integrated Microbial Genomes system from the Joint Genome Institute. Results were based on BLASTP analysis, with cutoff values of <10–5 (E value) and 30% identity for orthologs and <10–2 (E value) and 20% identity for unique CDSs.

Phylogenetic tree analysis. Homologs of M. petroleiphilum PM1 translated CDSs were identified using BLASTP searches against the nonredundant GenBank database from the National Center for Biotechnology Information. Sequences were aligned and alignments were refined using ClustalX, version 1.8 (41), along with manual adjustments. The Protdist program and the Neighbor program of the Phylip package (28) were used to generate the phylogenetic tree for MpeA3393. The pairwise parameters included a gap opening of 35 and a gap extension of 0.75. The multiple alignment parameters included a gap opening of 15, a gap extension of 0.3, delay divergence of 30%, and use of the Gonnet series for the protein weight matrix.

Comparative genomic hybridization and comparative genomic sequencing analyses. Comparative genomic hybridization was conducted in order to analyze the conservation of genes from MTBE-degrading isolates MG4 and 312 with those from PM1 across the entire genome. High-density arrays (~400,000 oligomers) were designed and produced by NimbleGen Systems, Inc. (Madison, WI), using 29-mer probes specific for every 26 bp for both strands of the entire genome and for every 7 bases for both strands of the plasmid. Arrays were hybridized with labeled genomic preparations of MG4, 312, and PM1. Genomic DNA was isolated (5) and digested (5 µg per array) with 0.005 U DNase I (Amersham) in 1x One-Phor-All buffer (Amersham, Piscataway, NJ) at 37°C for 5 min, with subsequent inactivation (95°C for 15 min). To the DNA digest were added 4 µl 5x terminal transferase buffer, 1 nmol biotin-N6-ddATP, and 25 U terminal transferase. The sample was incubated at 37°C for 90 min, followed by inactivation at 95°C for 15 min. Hybridization of arrays was conducted in 1x hybridization buffer for 16 h at 45°C, using a Hybriwheel device (NimbleGen). PM1 was used as a reference in the analysis and was hybridized to separate arrays. Duplicate arrays were processed per strain. Arrays were washed with nonstringent wash buffer (6x SSPE [1x SSPE is 0.18 M NaCl, 10 mM NaH2PO4, and 1 mM EDTA {pH 7.7}], 0.01% [vol/vol] Tween 20), followed by two 5-min washes with stringent buffer (100 mM morpholineethanesulfonic acid [MES], 0.1 M NaCl, 0.01% [vol/vol] Tween 20) at 47.5°C. Arrays were stained with Cy3-streptavidin conjugate (Amersham Piscataway, NJ) for 10 min, followed by washing in nonstringent buffer. Signal amplification was achieved by secondary labeling with biotinylated goat anti-streptavidin (Vector Laboratories, Burlingame, CA), washing in nonstringent buffer, and restaining with Cy3-streptavidin. Finally, arrays were washed in nonstringent wash buffer, in 0.5x SSC (1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate) two times for 30 s each, and in 70% ethanol for 15 s. Arrays were spun dry by centrifugation. Scanning was conducted at 5-µm resolution with a Genepix 4000b scanner (Axon Instruments, Union City, CA), and NimbleScan software (NimbleGen) was used to obtain pixel intensities. For higher-resolution resequencing of the MG4 and 312 plasmids, arrays were synthesized and hybridized with genomic DNA from each strain and scanned as described above. Single-nucleotide polymorphism (SNP) positions were tested for uniqueness in the genome, using custom algorithms (NimbleGen). The PM1 annotation (http://genome.ornl.gov/microbial/rgel/) was used to generate the output file in SignalMap analysis software (NimbleGen). The predicted SNPs were confirmed by producing and sequencing amplicons, using PCR primers located external to the SNP locations.

Random mutagenesis, mutant characterization, and plasmid curing of PM1. In order to label the megaplasmid with a selectable marker, random transposon mutagenesis was employed using the mini-transposon derivative pTnMod-SmO, containing the streptomycin (Sm)/spectinomycin adenylyltransferase gene (aadA) and an oriR origin of replication between the inverted repeats (22). Electrocompetent PM1 cells were prepared by being cultured in 0.5x tryptic soy broth at 27°C, with shaking, to log phase. Cells were collected by centrifugation, washed in 10% glycerol four times, and suspended in 10% glycerol to a final volume of 100 µl. A mixture of 50 µl cells and 2 µl pTnMod-SmO DNA (1 µg/µl) was electroporated in 0.1-mm-gap cuvettes, using 1.8 kV, 200 {Omega}, and 25 µF capacitance settings (22) on a Bio-Rad Gene Pulser electroporator (Bio-Rad, Hercules, CA). Following a 4-h recovery in 0.5x tryptic soy broth at 27°C with shaking, transposon mutants were selected on 0.5x TSA plates with 50 µg/ml Sm. Sm-resistant colonies were present after incubation for 5 to 6 days at 27°C, and stable transposon integration was confirmed by PCR analysis of genomic DNA, using pTnMod-SmO-specific primers.

Using the rapid cloning strategy outlined by Dennis and Zylstra (22), the <SmO> insert location was mapped in several PM1 subclones containing oriR within the transposon. Briefly genomic DNA was extracted, digested with the AvaII restriction endonuclease, self-ligated, and transformed into Escherichia coli TOP10 cells (Invitrogen, Carlsbad, CA). The resulting transformants were selected on LB agar containing 50 µg/ml Sm. Sequencing with primers against the ends of the <SmO> insert was used to determine the exact insert location. One transposon mutant, MP0005, was shown to have the <SmO> insert on the megaplasmid (in MpeB636). MP0005 was subjected to plasmid curing by heat stress as described by Trevors (78). Specifically, strain MP0005 was incubated at 37°C for 6 to 8 h before being plated on 0.5x TSA. Following replica plating on 0.5x TSA, with and without 50 µg/ml Sm, Sm-sensitive colonies were selected, and megaplasmid loss was confirmed by PCR analysis. MTBE and TBA degradation activities by strain MP0005 and a megaplasmid-free strain, MP0007, were determined by gas chromatography analysis as previously described (34).

Nucleotide sequence accession number. The nucleotide sequence annotation is available at the Joint Genome Institute website (http://genome.ornl.gov/microbial/rgel/) and has been deposited in the GenBank/EMBL database under accession numbers CP000555 for the chromosome sequence and CP000556 for the plasmid sequence.


   RESULTS AND DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results and Discussion
References
 
General genome features of chromosome and megaplasmid. The Methylibium petroleiphilum strain PM1 genome consists of a circular chromosome of 4,044,225 bp (Fig. 1A) and a megaplasmid (pPM1) of 599,444 bp (Fig. 1B) (Table 1). The genome carries 4,477 putative CDSs, of which 964 are unique to PM1 based on BLASTP searches against the nonredundant database. The pPM1-encoded proteins account for a disproportionately large number (382) of these unique genes. Of the remaining proteins, 2,801 could be assigned a putative function based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. Analysis of the top BLAST hits (against completed genomes in KEGG) revealed that the closest homologs were most often found in other beta-proteobacterial genomes (2,332), with the most hits (790) for Ralstonia solanacearum, followed by Burkholderia pseudomallei (497) and Azoarcus sp. strain EbN1 (413). This distribution appears to reflect that for the chromosome, with 2,210, 589, and 364 homologs in beta-, gamma-, and alpha-proteobacteria, respectively (Table 1). Interestingly, in contrast to the case for the chromosome, where beta- and gamma-proteobacteria account for 57.7% and 15.4% of the top BLAST hits, respectively, the distribution of top hits between beta (18.9%)- and gamma (15.6%)-proteobacteria is nearly equivalent for the megaplasmid. The smaller fraction of beta-proteobacteria-like CDSs in the megaplasmid is balanced by the large proportion of CDSs with no hits to KEGG genomes (47.7%) compared with the CDSs on the chromosome (9.9%). This surprising difference in the phylogenetic distribution of best hits, together with the discrepancy in G+C content between the plasmid (66%) and the chromosome (69.2%), points to the likelihood that the plasmid was horizontally acquired; further evidence for this statement is provided by conservation of the megaplasmid in other phylogenetically similar MTBE-degrading bacteria (discussed in detail below). Analysis of the distribution of clusters of orthologous genes (COG) (77) showed that the most abundant groups (excluding no COG or general function) were amino acid transport and metabolism (7.0%), energy production (6.4%), and transcription (6.3%) for the chromosome and replication, recombination and repair (8.0%), coenzyme transport and metabolism (7.0%), and inorganic ion transport and metabolism (5.3%) for the plasmid.


Figure 1
View larger version (25K):
[in this window]
[in a new window]

  		FIG. 1. Circular maps of the chromosome (A) and megaplasmid pPM1 (B) of M. petroleiphilum PM1. Outer rings 1 and 2 show all CDSs, colored by functional category (dark gray, hypothetical proteins; light gray, conserved hypothetical and unknown function; brown, general function prediction; red, replication and repair; green, energy metabolism; blue, carbon and carbohydrate metabolism; cyan, lipid metabolism; magenta, transcription; yellow, translation; orange, amino acid metabolism; pink, metabolism of cofactors and vitamins; light red, purine and pyrimidine metabolism; lavender, signal transduction; sky blue, cellular processes; and pale green, structural RNAs). Genes coding for major metabolic features (green, methylotrophy; red, aromatic hydrocarbon degradation; and light blue, alkane degradation) are shown on rings 3 and 4. Large repeat regions are indicated on rings 5 and 6. The larger repeated regions colored light gray (29-kb repeat) and dark gray (40-kb repeat) are described in the text, while the other colors represent repeated IS elements, ISmp1-8. Ring 7 shows the deviation from the average G+C, and the innermost ring, ring 8, shows the GC skew (G–C)/(G+C). The plasmid and chromosome are not drawn to scale.

 

View this table:
[in this window]
[in a new window]

  		TABLE 1. General features of the Methylibium petroleiphilum PM1 genomea

 
The chromosome contains a single ribosomal rrn operon (16S-tRNAAla-tRNAIle-23S-5S) and all genes coding for ribosomal proteins. Structural RNA genes for signal recognition particle RNA, rnpB, and tmRNA were present. Forty-two tRNA genes, evenly distributed on the chromosome (with the exception of a few clusters of 2 or 3 tRNAs), correspond to 40 tRNA acceptors and can recognize all possible codons. A very unusual feature of pPM1 is that it contains a single large cluster of 27 additional tRNA genes (25 are redundant with those on the chromosome, and the other 2 do not have clear anticodons). This is the first report of such a large tRNA gene island and the first report of such a cluster on a plasmid. The role of this island in translation, in genome evolution, or in positive selection of the plasmid in this or other bacterial strains is unclear.

Cell motility, secretion, and transport systems. M. petroleiphilum PM1 possesses the genes necessary for flagellar biosynthesis (for one polar flagellum), chemotactic responses, type IV pilus synthesis, and the type II secretion pathway as well as several genes related to the Agrobacterium tumefaciens type IV secretion pathway (Fig. 2; see Table S1 in the supplemental material). Type IV secretion mechanisms are often involved in pathogenesis. However, homologs to only three of the five core type IV secretion proteins (VirB9, -10, and -11, not VirB4 or -7) (6) were identified, so it is unclear at present if PM1 possesses a functional type IV secretion pathway. PM1 likely moves by both flagellum-facilitated swimming and pilus-facilitated twitching motility. Three copies of tra genes on pPM1 suggest that PM1 may be capable of conjugative transfer, a possibility currently under investigation. Thirteen chromosomally and one plasmid-encoded methyl-accepting chemotaxis protein (MCP) allow PM1 to respond to a range of environmental stimuli (see Table S1 in the supplemental material). As in other organisms (83), MCP genes in PM1 are found scattered throughout the genome. Only three MCP genes are found within taxis operons, namely, the pilG-L operon, the cheYA(MCP)W operon, and the flg/flh gene cluster. The apparent mobility of MCPs, together with the fact that six PM1 MCPs appear to be paralogs, complicates functional assignation. Nevertheless, there are five other MCP genes in addition to those already mentioned whose gene environments may offer insight into their possible functions, as follows: two paralogous MCP genes are located immediately downstream of and may be part of the same operon as the genes for the two toluene/benzene monooxygenase pathways; one of the aer-like MCPs is immediately downstream of, and in the same orientation as, the LysR-type activator gene of the ribulose 1,5-bisphophate carboxylase/oxygenase (RuBisCO) operon, which is upstream of the two regulatory genes; one MCP gene may be cotranscribed with a gene showing similarity to the direct oxygen sensor dos of E. coli; and one MCP gene may be cotranscribed with a gene showing low percent similarity to those for bacteriophytochromes and motility sensors. Neighbor-joining analysis of the putative PM1 MCPs against their homologs showed that eight MCPs cluster close to MCP1-4 of E. coli and MCPA-D of Salmonella enterica serovar Typhimurium, two appear to be related to the aerotaxis and energy sensor AER of E. coli, and one is very similar to the twitching motility protein PilJ of Pseudomonas aeruginosa.


Figure 2
View larger version (41K):
[in this window]
[in a new window]

  		FIG. 2. Schematic diagram of an M. petroleiphilum PM1 cell showing structural features and cellular processes, including predicted methylotrophy and fuel hydrocarbon degradation pathways. Two adjacent arrows imply multiple steps. Abbreviations: MTBE, methyl tert-butyl ether; HMTBE, hydroxymethyl tert-butyl ether; TBF, tert-butyl formate; TBA, tert-butyl alcohol; HIBA, hydroxyisobutyric acid. Systems involved in metal resistance and transport, secretion, motility, chemotaxis, and electron transport are shown within or associated with the cell membrane. Cofactors are labeled green, and electron donors are labeled red in the cytosol. The figure is not drawn to scale.

 
Strain PM1 has two sets of genes coding for form I RuBisCO, i.e., cbbL (MpeA1478 and MpeA2782) and cbbS (MpeA1479 and MpeA2783), and associated enzymes required for CO2 fixation via the Calvin cycle (see Table S1 in the supplemental material); however, this activity has not been demonstrated for PM1. A thorough search of the PM1 genome revealed the absence of genes encoding key enzymes from each of the three other known CO2 fixation pathways, including 2-oxoglutarate:ferredoxin oxidoreductase and ATP citrate lyase (reductive TCA cycle), the acetyl-coenzyme A (acetyl-CoA) synthase/CO dehydrogenase (reductive acetyl-CoA pathway), and malonyl-CoA reductase and propionyl-CoA synthase (3-hydroxypropionate cycle) (4, 40). This strain possesses several ABC transporters for transport of inorganic ions, such as nitrate, sulfate, magnesium, potassium, phosphate, and phosphonate, as well as amino acids, branched-chain amino acids, carbohydrates, long-chain fatty acids, dipeptides/oligopeptides, polyamines, and antibiotics (Fig. 2; see Table S2 in the supplemental material). In addition, putative regulatory/signaling proteins and cytochromes (based on CXXCH motifs) have been identified (see Table S3 in the supplemental material).

Repeated elements. The genome has a number of complex repetitive elements, including eight families of insertion sequences (ISmp1-8) (up to 12 copies) and two large genomic segments (29 and 40 kb) flanked by IS elements that have undergone what appear to be recent duplications (Fig. 1). The two replicons do not equally share the repeated insertion elements; five of the eight families are located only on the chromosome and one family is found strictly on the plasmid. The distribution patterns of the IS elements lend support to the dissimilar phylogenetic distributions of best KEGG hits among sequenced genomes and strengthen the notion of the megaplasmid's recent acquisition.

Parallel copies of ISmp8 flank two tandem copies of a 29-kb repeat, each consisting of two operons involved in phosphonate and cobalamin metabolism. The phosphonate operons (phnFDC-htxFGHIJKLMN) include genes for putative C-P lyase subunits which are 54 to 83% similar to those of Pseudomonas stutzeri WM88 (81) (see Table S2 in the supplemental material). The Htx and Phn C-P lyases support growth on methylphosphonate and additional alkylphosphonates, respectively; growth on these substrates is not yet known for PM1. Also contained in the repeat are cobalamin (vitamin B12) synthesis genes encoding the conversion of uroporphyrinogen III to cobinamide and the synthesis of dimethylbenzimidazole in the aerobic pathway for cobalamin biosynthesis (MpeB437-453 and MpeB472-488) (see Table S1 in the supplemental material). Downstream of the tandem repeat are the remaining genes (MpeB509-522) for the covalent linkage of dimethylbenzimidazole, cobinamide, and a phosphoryl group to complete the cobalamin synthesis pathway. Genes coding for the anaerobic pathway of cobalamin biosynthesis (cbi genes) are also present in the cob clusters, but a complete pathway is lacking. PM1 also lacks a cobG gene, encoding the monooxygenase that converts precorrin-3 to precorrin-4 in the aerobic pathway, but one or more of the multiple copies of cbiG (MpeB479, -480, -444, and -445) may code for a functional enzyme that performs this reaction without oxygen (60). Cobalt and cobalamin (vitamin B12) have been shown to enhance PM1's ability to grow on and degrade MTBE and its primary metabolite, TBA (Hristova, unpublished data), so it is not surprising that multiple copies of genes involved in cobalamin synthesis are present in PM1, including tandem repeats of cob and cbi genes. Recently, Rohwerder et al. (61) showed that cobalamin synthesis affected the growth rate on the MTBE metabolites TBA and 2-hydroxyisobutyrate (2-HIBA) for a beta-proteobacterial MTBE-degrading strain that was phylogenetically similar to PM1 (95.6% identical based on the 16S rRNA gene sequence). In this strain, cobalt or cobalamin was necessary for the activity of an enzyme, isobutyryl-CoA mutase, involved in metabolism of 2-HIBA (61), and 99% identical homologs to this two-component mutase are present in PM1 (MpeB538/541). As mentioned, a relatively large percentage of predicted proteins whose genes are carried on the plasmid (7.0%) belong to the COG category for coenzyme transport and metabolism. A cluster of ethanolamine utilization (eut) genes (MpeB0499-502) found between the tandem repeats and the third cobalamin cluster on the plasmid encode putative proteins which are 48 to 85% similar to EutJEMN from the cobalamin-dependent ethanolamine utilization pathway of S. enterica serovar Typhimurium (49). The latter two proteins are homologs of carboxysome shell proteins (CcmKL). While S. enterica serovar Typhimurium also contains the ethanolamine lyase subunits and regulator (EutBC and EutR) in its eut operon, in PM1 these genes (MpeA2417-8 and MpeA2415) are on the chromosome.

The largest (40 kb) repeated element is present on both the plasmid and the chromosome and encodes a putative PinR-like site-specific recombinase, a replicative DNA helicase, two putative SpoJ-like transcriptional regulators or plasmid partitioning proteins, a SpoIIIE-like DNA translocase, a tellurite resistance protein, and many hypothetical products. The presence of the repeat on both replicons suggests a recent duplicative transfer event. Although the types of genes found in this region suggest a plasmid origin, both copies interrupt similar but nonidentical copies of dcd genes (for dCTP deaminase), and the direction of duplicative transfer remains to be proven. Since this duplication, the 40-kb repeat on pPM1 has been interrupted by a transposase between genes MpeB0184 and MpeB0187.

Heavy metal tolerance and metal homeostasis. One interesting outcome of the genome analysis is evidence of PM1's potential resistance to heavy metals, suggesting promise in using the organism to treat sites containing mixed wastes. Arsenic extrusion in PM1 is probably mediated by arsRBC, present in two copies on the chromosome that are 58 to 81% similar to each other (MpeA1581-1584, arsHBCR; MpeA2343-4 and MpeA2347, arsCB and arsR, respectively). While some other bacteria have five-gene operons (arsRDABC) and use the ArsAB pump, PM1 probably extrudes arsenite by a carrier protein, ArsB, energized by the membrane potential (62), although resistance to arsenic oxyanions needs to be evaluated. ArsC is an arsenic reductase responsible for the transformation of As(V) into As(III), and ArsR is a transcriptional repressor that responds to As(III) and Sb(III) (62). The function of the fourth gene product, ArsH, found in several bacteria (Yersinia enterocolitica, Acidothiobacillus ferrooxidans, Pseudomonas aeruginosa, and Pseudomonas putida KT2440), still remains unclear (10, 63). Three chromosomal copies of chrA (MpeA2204, MpeA2205, and MpeA2526), belonging to the CHR family of transporter genes, may mediate chromate resistance in PM1. One copy of the ChrA gene (MpeA2204) is 63 and 61% similar to its homologs in Dechloromonas aromatica RCB and P. putida KT2440, respectively, although a homologous chromate reductase (ChrR) (42) was not evident in PM1.

Genome analysis revealed 15 copper resistance genes in a large cluster (~14.4 kb, at positions 1760297 to 1775675) on the PM1 chromosome with structural analogy to the plasmid-mediated (pMOL30) copper resistance cluster copOAIPRSFG in Ralstonia metallidurans (54) (see Table S2 in the supplemental material). The copOAIP-copRS cluster in PM1 is likely involved in the efflux of periplasmic copper (analogous to the cop systems of Pseudomonas syringae, R. metallidurans, and R. solanacearum and the cos system of E. coli) (13), whereas the efflux of cytoplasmic copper is mediated by a P1-ATPase, CopF1F2 (analogous to R. metallidurans CopF) (54). The genome of PM1 also encodes a putative chemiosmotic antiporter efflux system similar to CzcCBA of R. metallidurans, conferring resistance to Cd, Zn, and Co (54). In addition to copF, there are two other genes encoding putative metal-transporting P1-type ATPases, namely, MpeA2479 and MpeA3535. Additional proteins potentially involved in metal transport include NikBCDE for nickel (MpeA3117-3120), CbiOQMK for cobalt and nickel (MpeA2799-2802), and ModABC (MpeA3707, MpeA3714, and MpeA3715) for molybdenum uptake (see Table S2 in the supplemental material).

Ferric iron has also been shown to enhance PM1's ability to grow on and degrade MTBE and TBA (Hristova, unpublished data), so it is likely that active transport of iron is of particular importance. As with other gram-negative bacteria, PM1 acquires its iron supply via Fe3+ siderophores. fep genes, which function in the synthesis of polypeptides required for uptake of ferric enterobactin, were identified inside each of the four cob operons of PM1 (see Table S1 in the supplemental material). Polaromonas sp., Rhodoferax ferrireducens, and Methylobacillus flagellatus all have iron transport genes either within or in close proximity to their cob operons (data not shown). Minimally, the FepABC proteins are required for ferric enterobactin uptake. MpeA2292 and MpeA2605 have been annotated as fepA, coding for an outer membrane receptor for an iron siderophore, but it is possible that btuB genes located near the febBDC genes are also involved in iron assimilation. The TonB-dependent energy transduction complex (tonB, exbB, or tolQ and exbB) (see Table S1 in the supplemental material) encoded on the chromosome likely provides the mechanism for active transport of iron siderophores and cobalamin across the outer membrane (8, 37). The PM1 genome encodes about 39 putative proteins involved in iron transport and homeostasis, which implies the importance of iron in its physiology.

Methylotrophy. Methylotrophic metabolism of PM1 is of great interest because formaldehyde and formate are common intermediates of both methanol and MTBE or TBA oxidation by PM1 and other degraders (9, 59). M. petroleiphilum PM1 is capable of aerobic growth on methanol, formate, and succinate. Unlike other methylotrophic beta-proteobacteria, PM1 grows on MTBE, toluene, benzene, ethylbenzene, and dihydroxybenzoates as sole carbon sources (57, 59). PM1 possesses genes for the serine cycle and methylotrophy scattered in several different clusters on its chromosome (Fig. 1; Table 2). The strain does not grow on methylamine (Hristova, unpublished data), lacks a gene encoding the methylamine dehydrogenase large subunit, and likely lacks methylamine dehydrogenase activity. Despite the ability of PM1 to grow on methanol, its genome lacks true homologs of mxaF and mxaI, known genes coding for the methanol dehydrogenase (MeDH) large and small subunits, respectively, present in several methylotrophs known to date. PM1 contains an MeDH-like cluster, xoxF-J (MpeA3393-5), that is present in Methylobacterium extorquens AM1 (17), which also contains the true mxaF cluster. Comparative sequence analysis of the product of the MpeA3393 gene revealed high similarity to MxaF (74% to that of M. extorquens AM1) and the XoxF homologs present in several nonmethylotrophs (77% to that of Burkholderia fungorum). Based on phylogenetic analysis, MpeA3393 clusters with the MxaF homologs of unknown function from other methylotrophic and nonmethylotrophic Rhizobium and Burkholderia spp., while true MeDH large subunits (MxaF) cluster together and are distinct from the MxaF homologs (Fig. 3).


View this table:
[in this window]
[in a new window]

  		TABLE 2. M. petroleiphilum PM1 genes putatively involved in metabolism of methanol

 

Figure 3
View larger version (26K):
[in this window]
[in a new window]

  		FIG. 3. Phylogenetic tree of MpeA3393, which putatively codes for the large subunit of methanol dehydrogenase. The protein product from MpeA3393 is most similar to glucose dehydrogenase (GluDH) from B. fungorum (81% similar; GenBank accession no. ZP_00283396) and the methanol dehydrogenase large subunit (Mdh large) from M. capsulatus Bath (75% similar; accession no. AAU90462). Protein names are shown in square brackets.

 
A cytochrome c555 gene (MpeA3394) 56% similar to the cytochrome cH gene of Methylococcus capsulatus Bath (electron donor to the oxidase in methylotrophic bacteria) (1) is found adjacent to MpeA3394. A putative MxaJ/XoxJ protein (MpeA3395) shows 54% similarity to XoxJ from Paracoccus denitrificans and 42% similarity to MxaJ from M. capsulatus Bath. Five genes (MpeA3829 and Mpe2585-8) are involved in the biosynthesis of pyrroloquinoline quinone, a cofactor of MeDH as well as quinoprotein ethanol dehydrogenase. A cluster of genes required for MeDH synthesis, mxaLKCASR (MpeA3273-3278), is also present. To date, none of the gene clusters containing the XoxF gene homolog have been shown to be involved in methanol oxidation. Therefore, it is possible that a new enzyme is responsible for this function in the beta-proteobacterium M. petroleiphilum PM1.

Three different formate dehydrogenase genes are present in the PM1 genome, with homologs to genes in M. extorquens and M. capsulatus Bath. The functions of the tungsten-dependent formate dehydrogenase gene fdh1 (MpeA0337-339), the NAD-linked formate dehydrogenase gene fdh2 (MpeA3708-12), and the cytochrome-linked formate dehydrogenase gene fdh3 (MpeA1170-71 and -1173) for energy generation during growth on C1 substrates or for MTBE oxidation need to be explored further. MpeA3377, coding for a putative ABC-type tungstate transport system permease, links gene clusters fdh1 and fdh2 (Table 2). The fdh2 genes in PM1 have the same gene arrangement as and significant sequence identity (52 to 81%) to the NAD-dependent formate dehydrogenase gene cluster, fdsGBACD, of Ralstonia eutropha (58). Pathways involved in metabolism and detoxification of formaldehyde, a central intermediate of both methanol and MTBE degradation, may also function in MTBE metabolism. PM1 has two pathways for formaldehyde oxidation to CO2, a tetrahydromethenopterin (H4MPT)-linked metabolic module that includes an archaea-like gene cluster and an H4F-linked metabolic module. Recently, phylogenetic analysis of a subset of bacterial and archaeal H4MPT-linked C1 transfer genes placed PM1 sequences with those of other described beta-proteobacteria (44).

Fuel hydrocarbon degradation pathways. PM1 contains an operon (MpeA0814-0821) likely encoding enzymes for conversion of benzene to phenol (and catechol) and toluene to methylphenol (and methylcatechol) that is 62 to 74% similar to the benzene monooxygenase pathway in P. aeruginosa JI104 (BmoA-D1) (48) and 50 to 71% similar to the toluene para-monooxygenase (TpMO) pathway in Ralstonia pickettii PKO1 (TbuA1UBVA2CX) (76) (Table 3). A second operon (MpeA2539-2547) is 55 to 74% identical to the first operon, but it likely does not yield a functional monooxygenase since the TbuA1 homolog is interrupted by a transposon insertion and the TbuC homolog is a pseudogene. Both operons have two-component response regulator-sensor histidine kinases which are located upstream and divergently transcribed (MpeA0811-812 and MpeA2536-2537), although MpeA2537 may be truncated due to the transposon insertion. MpeA0821 encodes a putative TbuX protein (65% similar to that in PKO1), an outer membrane protein regulated by TbuT and involved in toluene uptake (43). The benzene monooxygenase pathway has been implicated in benzene and toluene degradation (48), as has the TpMO pathway (76). In addition to benzene and toluene, PM1 has been shown to degrade o-xylene (20), although the biochemical pathway used has not been elucidated to date. It is likely that m- and p-xylene can also be metabolized via the toluene monooxygenase (TMO) pathway of PM1, as described for PKO1 and other bacteria (29).


View this table:
[in this window]
[in a new window]

  		TABLE 3. M. petroleiphilum PM1 genes putatively involved in metabolism of aromatic hydrocarbons denoting the predicted role of the gene product and percent similarity with well-characterized homologs

 
M. petroleiphilum PM1 grows on phenol, and two distinct clusters of dimethylphenol (dmp)-like genes are present (MpeA2265-67 and -2272-86; MpeA3305-13 and -3321-25), although the latter lacks the key structural gene dmpP, so it may not yield a functional phenol hydroxylase (PH). Gene products from the first cluster, dmpRKLMNOPQBCDEHFGI, are 60 to 83% similar to those encoded by pVI150 in Pseudomonas sp. strain CF600 (66), including a multicomponent PH, catechol 2,3-dioxygenase, and the meta-cleavage pathway for catechol (Table 3). The second operon has transposon insertions inside dmpC and adjacent to dmpO. The PH subunit genes for the two operons are 44 to 69% similar. The DmpR homologs (encoded by MpeA3310 and MpeA2286) are similar to TbuT (69 and 65%) and may regulate TMO, PH, and the meta-cleavage genes, since TbuT was shown to regulate these genes in PKO1 via separate promoters (11). However, these operons are located together in PKO1, whereas they are quite distant in PM1. PM1 can grow on phenol, and based on the presence of a complete dmp operon, it can likely degrade alkylphenols as well, although it is not clear whether PH is essential for methylphenol degradation (as described for P. stutzeri OX1 [12]) or whether the TMO alone is capable of converting toluene to methylcatechol (as described for strain PKO1 [29]).

PM1 has nine CDSs encoding putative proteins with various similarities to cyclohexanone monooxygenases (CHMOs), sometimes referred to as Baeyer-Villiger-type MOs (MpeB0579, -B607, -B610, -A393, -A898, -A1038, -A1351, -A2885, and -A2915). Their protein products may play a role in hydroxylation of either alicyclic, aliphatic, or aryl ketones to form a corresponding ester, which can easily be hydrolyzed. Alicyclic hydrocarbons represent up to 12% (wt/wt) of total hydrocarbons in petroleum mixtures (according to the American Petroleum Institute). Aryl ketones such as acetophenone can be produced directly from atmospheric breakdown of ethylbenzene (a major petroleum component) or following abiotic conversion of ethylbenzene to ethylphenol (3) and subsequent biological conversion to the ketone. The putative CHMO genes are scattered across the genome and are not present in operons with other genes coding for subsequent metabolism after the MO reaction (i.e., esterases and alcohol and aldehyde dehydrogenases). The CHMOs have a narrow substrate range, possibly explaining the number of different flavoprotein MOs in PM1 with various levels of similarity to representatives from this class; the putative CHMOs in PM1 are more similar to phenylacetone MO (46 to 67%) (53) than to 4-hydroxyacetophenone MO (43 to 53%) (45). In PM1, the nine Baeyer-Villiger MOs have a putative NADP+-binding site that is 72 to 88% similar to the proposed site in a CHMO from Acinetobacter sp. strain NCIMB 9871 (16).

An alkane monooxygenase pathway on pPM1 may facilitate PM1's growth on n-alkanes. In addition, alkane monooxygenase (hydroxylase) has been proposed to play a role in cometabolic MTBE hydroxylation, since acetylene, an inactivator of short-chain alkane monooxygenase, was shown to inhibit MTBE degradation (68). In PM1, the hydroxylase subunit, AlkB (MpeB0606), is 69% and 66% similar to that of Alcanivorax borkumensis AP1 (72) and P. putida PGo1 (contained on the OCT plasmid) (79), respectively, and contains all eight of the conserved His residues observed in other integral membrane binuclear-iron hydrocarbon monooxygenases (33). Also present are two rubredoxin (Rd) genes (MpeB0603 and MpeB0602), whose products are 76 and 78% similar to rubredoxin 3 and 4, respectively, in Gordonia sp. strain TF6 (31). The Rd encoded by MpeB0603 is an AlkG1-type Rd, whereas that encoded by MpeB0602 is an AlkG2-type Rd, based on the CXXCG motif criterion described by van Beilen et al. (80). Only AlkG2-type Rds were shown to be functional in electron transfer from the rubredoxin reductase to alkane hydroxylase. In addition, MpeB0601 codes for an ATP-dependent transcriptional regulator which is 38% similar to AlkS from A. borkumensis SK2 (35). Separated from the putative alkS gene by three hypothetical genes is a rubredoxin reductase gene, alkT (MpeB0597), whose protein product is 49% similar to that in Gordonia sp. strain TF6. PM1 does not appear to possess any long-chain alkane (>C13) oxidation pathways, such as an alkane dioxygenase (64), P-450 monooxygenase (2), or two alkane hydroxylase complexes (AlkMa and AlkMb), similar to Acinetobacter sp. strain M-1 (75), although PM1's single AlkB protein is 54% similar to both AlkMa and AlkMb. The gene organization of the alk operon in PM1 is somewhat similar to that of Gordonia sp. strain TF6 (alkB2G1G2T), except that a transposase gene (MpeB605) and putative esterase gene (MpeB604) are between alkB and alkG1G2, and as mentioned, a putative alkS gene and three hypothetical genes (MpeB600-598) are located between alkG1G2 and alkT in PM1. Homologs to alkHJKL from P. putida GPo1, coding for aldehyde dehydrogenase, alcohol dehydrogenase, acyl-CoA synthetase, and an outer membrane protein (79), respectively, were not present on the plasmid, although the PM1 chromosome contains homologs to the AlkH (MpeA2324, 47% similar), AlkJ (MpeA3803, 58% similar), AlkK (MpeA1769, 71% similar) and AlkL (MpeA3010, 49% similar) genes.

In addition, the PM1 chromosome contains a putative propane monooxygenase pathway (MpeA950-953) whose predicted proteins are 41 to 64% identical to PrmABCD in Gordonia sp. strain TY-5, including the large hydroxylase subunit, the NADH-dependent acceptor oxidoreductase, the small hydroxylase subunit, and the coupling protein, respectively. The Prm complex in strain TY-5 was shown to catalyze the subterminal oxidation of propane, yielding 2-propanol (50), while propane oxidation in PM1 is currently under investigation. As for PrmA in Gordonia TY-5, a pair of conserved Glu-X-X-His sequences are present in the putative PrmA protein of PM1, at residues 138 to 141 and 237 to 240. The presence of these sequences is consistent with the case for other monooxygenases in the binuclear-iron oxygenase family, including soluble methane monooxygenases (23, 71), suggesting that PrmA in PM1 may catalyze the hydroxylation of propane. Like the operon in strain TY-5, a chaperone similar to GroEL is encoded adjacent to the prm cluster in PM1 (MpeA0954). Finally, PM1 has homologs to strain TY-5 alcohol dehydrogenase genes, namely, adh-1 (MpeA936) and adh-3 (MpeA599), that are 72 and 83% similar, respectively, and may facilitate 2-propanol degradation. The putative monooxygenase genes in PM1 are summarized in Table S3 in the supplemental material, including the methanesulfonate monooxygenase gene msmA and the alkanesulfonate monooxygenase gene ssuD, which are part of the msmABDCEFGHG and ssuAADCB operons, respectively. PM1 may not utilize methanesulfonate, since its msmA gene lacks the sequence encoding CXH-X26-CXXH unique to methanesulfonate utilizers (7). In general, PM1 possesses several homologous genes with other soil bacteria, including Gordonia, Alcinovorax, and Pseudomonas spp., capable of biodegradation of petroleum hydrocarbons as well as xenobiotic and recalcitrant compounds such as phthalates.

MTBE biodegradation. Although MTBE is a recent anthropogenic contaminant (released within the last 15 years), various microorganisms can utilize the compound for carbon and energy under aerobic conditions (30, 61, 65, 74). M. petroleiphilum PM1 is the best characterized of the few bacterial pure cultures reported to grow on and completely degrade MTBE and its daughter product, TBA (20, 30, 36, 65). The genetic basis for MTBE and TBA conversion is not known, although different classes of monooxygenases have been proposed to play a role in metabolism or cometabolism of these compounds (30, 36, 51, 68, 74), including P-450 monooxygenase and alkane monooxygenase (hydroxylase) systems, the latter of which was shown to play a role in cometabolic degradation of MTBE by P. putida GPo1 (69) and possibly also by Pseudomonas mendocina KR-1 (70). A known inducer of alkane hydroxylase, dicyclopropylketone, was also shown to induce MTBE conversion to TBA in GPo1 (69). As reported above, an alkane MO (AlkB) system was detected in the PM1 genome on the megaplasmid. The AlkB system in PM1 is likely involved in MTBE hydroxylation, based on its similarity to other AlkB proteins in organisms shown to be involved in MTBE degradation. Whereas the Ks values for MTBE in n-alkane-grown GPo1 were reported to be quite high (20 to 40 mM), the apparent half-saturation constant for MTBE by PM1 was 88 µM, which is in the range of Ks values for MTBE by butane-degrading bacteria (51). Unlike GPo1 and KR-1, PM1 further degrades TBA, ultimately producing CO2 and biomass. The putative AlkB protein in PM1 is proposed to oxidize only MTBE, not TBA, based on kinetic experiments with MTBE- and TBA-grown cells (19). Two separate enzyme systems were also suggested for MTBE and TBA degradation by Hydrogenophaga flava ENV735 (36). Because of its potential role in MTBE metabolism, the coding region for AlkB is the focus of current gene knockout studies. Biodegradation of a similar molecule, ethyl-tert butyl ether (ETBE), occurs via a cytochrome P-450 pathway in Rhodococcus ruber IFP2001 (15). However, homologs of protein complexes involved in ETBE degradation from R. ruber were not found in PM1; like GPo1 (69), PM1 has not been shown to degrade ETBE.

Many pollutant degradation genes are located on bacterial catabolic plasmids. Significantly, the two strains, MG4 and 312, that are capable of MTBE degradation have nearly identical plasmids to that of PM1 (ca. 99% identical), as determined by comparative genome sequencing analysis. The MG4 and 312 plasmids showed only five and four SNPs, respectively, relative to PM1 (Table 4). The MG4 and 312 plasmids also lacked transposase genes (three copies of Tra5 and transposase 8 genes and one copy of a DDE domain transposase gene) that are present on the PM1 plasmid and a 1.2-kb deletion putatively containing an esterase/lipase gene (MpeB0604) and a DDE domain transposase gene (MpeB0605) (Table 3). The promoter and coding regions for alkB (MpeB0606) did not appear to be affected by this deletion, since it is significantly upstream, although there was a SNP mapped within alkB of MG4 and 312 resulting in a putative amino acid change. As mentioned, two other PM1 plasmid-carried genes, MpeB541 and MpeB0538, code for putative large and small subunits of isobutyryl-CoA mutase, respectively. The plasmids of strains MG4 and 312 also contained identical copies of MpeB541 and MpeB0538. These gene products were shown to have 99% identical homologs in Ideonella sp. strain L108 which were predicted to allow conversion of 2-HIBA to 3-hydroxybutyrate in the presence of CoA and ATP (61). It is not known whether these mutase genes are contained on a megaplasmid in L108, although horizontal gene transfer is often evoked when such high similarities in gene sequences between bacteria are observed. In addition to alkB, the PM1 plasmid (as well as the MG4 and 312 plasmids) contains a gene coding for 3-hydroxybutyryl-CoA dehydrogenase (MpeB0547), putatively involved in conversion of 3-hydroxybutyryl-CoA, a proposed MTBE metabolite (61), to acetoacetyl-CoA.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Genomic differences between the plasmid of PM1 and those from isolates MG4 and 312

 
The role of the megaplasmid in MTBE and TBA degradation was clearly demonstrated by curing experiments. Chemical analysis of MTBE and TBA degradation by the MP0005 parent strain and the MP0007 strain lacking the megaplasmid (as evidenced by PCR analysis and loss of Sm resistance) showed that only the former was able to degrade MTBE and TBA (Fig. 4). This result is consistent with our proposal that key genes involved in both MTBE and TBA degradation are located on the PM1 megaplasmid. Since two different monooxygenases are proposed to be involved in MTBE and TBA degradation (19; Hristova, unpublished data), at least some of the required protein subunits for conversion of MTBE to TBA and conversion of TBA to the putative 2-methyl-1,2-propanediol are encoded on the megaplasmid. In addition, the possible role of selected pPM1 proteins in MTBE/TBA oxidation, based on cDNA microarray results, is currently under investigation by gene knockout methods. It is noteworthy that the PM1 plasmid does not contain genes for predicted proteins with significant homology to those proposed for degradation of 2-methyl-1,2-propanediol by Mycobacterium austroafricanum IFP 2012 (52), although the putative aldehyde dehydrogenase encoded by MpeA1909 on the chromosome was 54% similar to MpdC (hydroxyisobutyraldehyde dehydrogenase) and the alcohol dehydrogenase encoded by MpeA945 was 45% similar to MpdB (2-methyl-1,2-propanediol dehydrogenase). While the relevant alcohol and aldehyde dehydrogenases may be encoded on the chromosome, additional plasmid-encoded dehydrogenases may be more plausible and are currently being investigated for their putative role in the MTBE degradation pathway.


Figure 4
View larger version (16K):
[in this window]
[in a new window]

  		FIG. 4. MTBE and TBA degradation by M. petroleiphilum PM1. The graph shows data for the parent strain MP0005, which carries the <SmO> marker on the megaplasmid (•, MTBE; {circ}, TBA), and the megaplasmid-free mutant MP0007 ({blacksquare}, MTBE; {square}, TBA).

 
Concluding remarks. Prior to sequencing of its genome, it was not known that PM1 possessed a 600-kb megaplasmid, much less that the plasmid contained candidate CDSs coding for the MTBE and TBA monooxygenases and enzymes involved in downstream reactions. It is noteworthy that MTBE-degrading strains from diverse locations, including a biofilter treating wastewater in Southern California (PM1) and two distinct aquifers in Northern California (MG4 and 312), possess nearly identical plasmids. The presence of this highly conserved megaplasmid among PM1-like MTBE degraders, along with its different G+C content, its unique IS element complement, and the unique phylogenetic distribution of its gene products, together raise interesting questions concerning horizontal gene/plasmid transfer and evolution of pathways via plasmid-mediated mechanisms. With the whole genome sequence, putative aromatic hydrocarbon and alkane degradation pathways were also identified, providing a basis for studying the complex regulation of fuel hydrocarbon degradation in this novel subsurface bacterium; this is important since substrate interactions are expected to influence the success of bioremediation strategies for gasoline-contaminated sites. In addition to comparative genomic approaches, whole-genome microarray and two-dimensional gel electrophoresis experiments are being conducted to identify genes and proteins unique to MTBE degradation. PM1 can serve as a model for other MTBE-degrading methylotrophs such that the knowledge gained from analysis of its genome, transcriptome, and proteome can be applied to PM1-like bacteria. An understanding of the MTBE degradation pathway and its regulation will allow for optimization of MTBE bioremediation and the ability to monitor this unique process in situ using molecular tools.


   ACKNOWLEDGMENTS
 
We thank Stephanie Malfatti at Lawrence Livermore National Laboratory (LLNL) for assistance with genome finishing and Miriam Land and Loren Hauser in the Genome Analysis Group at Oak Ridge National Laboratory for their assistance with genome annotation. We thank Tom Albert (NimbleGen) for assistance with comparative genomic sequence data analysis. We also thank Hussna Wakily and Victoria Katz at University of California-Davis (UCD) for their work on random transposon mutagenesis and Janice Wu (UCD) for work on plasmid-cured PM1 strains. In addition, we thank Harry Beller (LLNL) for helpful comments on the manuscript.

We acknowledge the University of California Office of the President for funding through the Campus-Laboratory Collaboration Program as well as the LLNL Laboratory-Directed Research and Development Program. This research was also supported by grant number 5 P42 ES04699-16 from the National Institute of Environmental Health Sciences (NIEHS), NIH, with funding provided by the U.S. EPA. This work was performed under the auspices of the U.S. Department of Energy by the University of California, Lawrence Livermore National Laboratory, under contract no. W-7405-Eng-48.

The contents of this study are solely the responsibility of the authors and do not necessarily represent the official views of the NIEHS, NIH, or EPA.


   FOOTNOTES
 
* Corresponding author. Mailing address: Lawrence Livermore National Laboratory, 7000 East Avenue, L-542, Livermore, CA 94550. Phone: (925) 422-7897. Fax: (925) 422-3800. E-mail: kane11{at}llnl.gov. Back

{triangledown} Published ahead of print on 8 December 2006. Back

{dagger} Supplemental material for this article may be found at http://jb.asm.org/. Back

{ddagger} S.R.K. and A.Y.C. gave equal contributions to this study. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results and Discussion
 References
 

  1. Afolabi, P. R., F. Mohammed, K. Amaratunga, O. Majekodunmi, S. L. Dales, R. Gill, D. Thompson, J. B. Cooper, S. P. Wood, P. M. Goodwin, and C. Anthony. 2001. Site-directed mutagenesis and X-ray crystallography of the PQQ-containing quinoprotein methanol dehydrogenase and its electron acceptor, cytochrome cL. Biochemistry 40:9799-9809.[CrossRef][Medline]
  2. Asperger, O., A. Naumann, and H.-P. Kleber. 1984. Inducibility of cytochrome P-450 in Acinetobacter calcoaceticus by n-alkanes. Appl. Microbiol. Biotechnol. 19:3948-4403.
  3. Atkinson, R. 1995. Gas phase tropospheric chemistry of organic compounds. Environ. Sci. Technol. 4:65-89.[Medline]
  4. Atomi, H. 2002. Microbial enzymes involved in carbon dioxide fixation. J. Biosci. Bioeng. 94:497-505.[Medline]
  5. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Stuhl. 1987. Current protocols in molecular biology. Wiley, New York, NY.
  6. Backert, S., and T. F. Meyer. 2006. Type IV secretion systems and their effectors in bacterial pathogenesis. Curr. Opin. Microbiol. 9:207-217.[CrossRef][Medline]
  7. Baxter, N. J., J. Scanlan, J. De Marco, A. P. Wood, and J. C. Murrell. 2002. Duplicate copies of genes encoding methanesulfonate monooxygenase in Marinosulfonomonas methylotropha strain TR3 and detection of methanesulfonate utilizers in the environment. Appl. Environ. Microbiol. 68:289-296.[Abstract/Free Full Text]
  8. Braun, V., and M. Braun. 2002. Iron transport and signaling in Escherichia coli. FEBS Lett. 529:78-85.[CrossRef][Medline]
  9. Bruns, M. A., J. R. Hanson, J. Mefford, and K. M. Scow. 2001. Isolate PM1 populations are dominant and novel methyl tert-butyl ether-degrading bacteria in compost biofilter enrichments. Environ. Microbiol. 3:220-225.[CrossRef][Medline]
  10. Butcher, B. G., S. M. Deane, and D. E. Rawlings. 2000. The chromosomal arsenic resistance genes of Thiobacillus ferrooxidans have an unusual arrangement and confer increased arsenic and antimony resistance to Escherichia coli. Appl. Environ. Microbiol. 66:1826-1833.[Abstract/Free Full Text]
  11. Byrne, A. M., and R. H. Olsen. 1996. Cascade regulation of the toluene-3-monooxygenase operon (tbuA1UBVA2C) of Burkholderia pickettii PKO1: role of the tbuA1 promoter (PtbuA1) in the expression of its cognate activator, TbuT. J. Bacteriol. 178:6327-6337.[Abstract/Free Full Text]
  12. Cafaro, V., V. Izzo, R. Scognamiglio, E. Notomista, P. Capasso, A. Casbarra, P. Pucci, and A. Di Donato. 2004. Phenol hydroxylase and toluene/o-xylene monooxygenase from Pseudomonas stutzeri OX1: interplay between two enzymes. Appl. Environ. Microbiol. 70:2211-2219.[Abstract/Free Full Text]
  13. Cervantes, C., and F. Gutierrez-Corona. 1994. Copper resistance mechanisms in bacteria and fungi. FEMS Microbiol. Rev. 14:121-137.[CrossRef][Medline]
  14. Chain, P., J. Lamerdin, F. Larimer, W. Regala, V. Lao, M. Land, L. Hauser, A. Hooper, M. Klotz, J. Norton, L. Sayavedra-Soto, D. Arciero, N. Hommes, M. Whittaker, and D. Arp. 2003. Complete genome sequence of the ammonia-oxidizing bacterium and obligate chemolithoautotroph Nitrosomonas europaea. J. Bacteriol. 185:2759-2773.[Abstract/Free Full Text]
  15. Chauvaux, S., F. Chevalier, C. Le Dantec, F. Fayolle, I. Miras, F. Kunst, and P. Beguin. 2001. Cloning of a genetically unstable cytochrome P-450 gene cluster involved in degradation of the pollutant ethyl tert-butyl ether by Rhodococcus ruber. J. Bacteriol. 183:6551-6557.[Abstract/Free Full Text]
  16. Chen, Y.-C., O. P. Peoples, and C. T. Walsh. 1988. Acinetobacter cyclohexanone monooxygenase: gene cloning and sequence determination. J. Bacteriol. 170:781-789.[Abstract/Free Full Text]
  17. Chistoserdova, L., and M. E. Lidstrom. 1997. Molecular and mutational analysis of a DNA region separating two methylotrophy gene clusters in Methylobacterium extorquens AM1. Microbiology 143:1729-1736.[Abstract]
  18. Davis-Hoover, W. J., S. A. Stavnes, J. J. Fleischman, S. C. Hunt, J. Goetz, M. Kemper, M. Roulier, K. Hristova, K. Scow, K. Knutson, W. R. Mahaffey, and D. J. Slomczynski. 2003. BTEX/MTBE bioremediation: bionets containing Isolite, PM1, SOS or air, p. E-25. In V. S. Magar and M. E. Kelley (ed.), Proceedings of the Seventh International In Situ and On-Site Bioremediation Symposium. Battelle Press, Columbus, OH.
  19. Deeb, R. A., S. Nishino, J. Spain, H. Y. Hu, K. M. Scow, and L. Alvarez-Cohen. 2000. MTBE and benzene biodegradation by PM1 via two independent monooxygenase-initiated pathways. Abstr. Am. Chem. Soc. 219:ENVR 228.
  20. Deeb, R. A., H.-Y. Hu, J. R. Hanson, K. M. Scow, and L. Alvarez-Cohen. 2001. Substrate interactions in BTEX and MTBE mixtures by an MTBE-degrading isolate. Environ. Sci. Technol. 35:312-317.[Medline]
  21. De Marco, P., C. Pachecoa, A. Figueiredoa, and P. Moradas-Ferreira. 2004. Novel pollutant-resistant methylotrophic bacteria for use in bioremediation. FEMS Lett. 234:75-80.
  22. Dennis, J. J., and J. G. Zylstra. 1998. Plasposons: modular self-cloning minitransposon derivatives for rapid genetic analysis of gram-negative bacterial genomes. Appl. Environ. Microbiol. 64:2710-2715.[Abstract/Free Full Text]
  23. Elango, N., R. Radhakrishnan, W. A. Froland, B. J. Wallar, C. A. Earhart, J. D. Lipscomb, and D. H. Ohlendorf. 1997. Crystal structure of the hydroxylase component of methane monooxygenase from Methylosinus trichosporium OB3b. Protein Sci. 6:556-568.[Abstract]
  24. Reference deleted.
  25. Ewing, B., and P. Green. 1998. Basecalling of automated sequencer traces using phred. II. Error probabilities. Genome Res. 8:186-194.[Abstract/Free Full Text]
  26. Ewing, B., L. Hillier, M. Wendl, and P. Green. 1998. Basecalling of automated sequencer traces using phred. I. Accuracy assessment. Genome Res. 8:175-185.[Abstract/Free Full Text]
  27. Fayolle, F., J.-P. Vandecasteele, and F. Monot. 2001. Microbial degradation and fate in the environment of methyl tert-butyl ether and related fuel oxygenates. Appl. Microbiol. Biotechnol. 56:339-349.[CrossRef][Medline]
  28. Felsenstein, J. 2002. PHYLIP (Phylogeny Inference Package), version 3.6a3. Department of Genome Sciences, University of Washington, Seattle.
  29. Fishman, A., Y. Tao, and T. K. Wood. 2004. Physiological relevance of successive hydroxylations of toluene para-monooxygenase of Ralstonia pickettii PKO1. Biocat. Biotrans. 22:283-289.[CrossRef]
  30. François, A., H. Mathis, D. Godefroy, P. Piveteau, F. Fayolle, and F. Monot. 2002. Biodegradation of methyl tert-butyl ether and other fuel oxygenates by a new strain, Mycobacterium austroafricanum IFP 2012. Appl. Environ. Microbiol. 68:2754-2762.[Abstract/Free Full Text]
  31. Fujii, T., T. Narikawa, K. Takeda, and J. Kato. 2004. Biotransformation of various alkanes using the Escherichia coli expressing an alkane hydroxylase system from Gordonia sp. TF6. Biosci. Biotechnol. Biochem. 68:2171-2177.[CrossRef][Medline]
  32. Gordon, D., C. Abajian, and P. Green. 1998. Consed: a graphical tool for sequence finishing. Genome Res. 8:195-202.[Abstract/Free Full Text]
  33. Hamamura, N., C. M. Yeager, and D. J. Arp. 2001. Two distinct monooxygenases for alkane oxidation in Nocardioides sp. strain CF8. Appl. Environ. Microbiol. 67:4992-4998.[Abstract/Free Full Text]
  34. Hanson, J. R., C. E. Ackerman, and K. M. Scow. 1999. Biodegradation of methyl tert-butyl ether by a bacterial pure culture. Appl. Environ. Microbiol. 65:4788-4792.[Abstract/Free Full Text]
  35. Hara, A., S. H. Baik, K. Syutsubo, N. Misawa, T. H. Smits, J. B. van Beilen, and S. Harayama. 2004. Cloning and functional analysis of alkB genes in Alcanivorax borkumensis SK2. Environ. Microbiol. 6:191-197.[CrossRef][Medline]
  36. Hatzinger, P. B., K. McClay, S. Vainberg, M. Tugusheva, C. W. Condee, and R. J. Steffan. 2001. Biodegradation of methyl tert-butyl ether by a pure bacterial culture. Appl. Environ. Microbiol. 67:5601-5607.[Abstract/Free Full Text]
  37. Higgs, P. I., P. S. Myers, and K. Postle. 1998. Interaction in the TonB-dependent energy transduction complex: ExbB and ExbD form homomultimers. J. Bacteriol. 180:6031-6038.[Abstract/Free Full Text]
  38. Hristova, K., B. Gebreyesus, D. Mackay, and K. M. Scow. 2003. Naturally occurring bacteria similar to the methyl tert-butyl ether (MTBE)-degrading strain PM1 are present in MTBE-contaminated groundwater. Appl. Environ. Microbiol. 69:2616-2623.[Abstract/Free Full Text]
  39. Hristova, K. R., C. M. Lutenegger, and K. M. Scow. 2001. Detection and quantification of MTBE-degrading strain PM1 by real-time TaqMan PCR. Appl. Environ. Microbiol. 67:5154-5160.[CrossRef][Medline]
  40. Hugler, H., H. Huber, K. O. Stetter, and G. Fuchs. 2003. Autotrophic CO2 fixation pathways in archaea (Crenoarchaeota). Arch. Microbiol. 179:160-173.[Medline]
  41. Jeanmougin, F., J. D. Thompson, M. Gouy, D. G. Higgins, and T. J. Gibson. 1998. Multiple sequence alignment with Clustal X. Trends Biochem. Sci. 23:403-405.[CrossRef][Medline]
  42. Jimenez, J. I., B. Minambres, J. L. Garcia, and E. Diaz. 2002. Genomic analysis of the aromatic catabolic pathways from Pseudomonas putida KT2440. Environ. Microbiol. 4:824-841.[CrossRef][Medline]
  43. Kahng, H.-Y., A. M. Byrne, R. H. Olsen, and J. J. Kukor. 2000. Characterization and role of tbuX in utilization of toluene by Ralstonia pickettii PKO1. J. Bacteriol. 182:1232-1242.[Abstract/Free Full Text]
  44. Kalyuzhnaya, M. G., M. E. Lidstrom, and L. Chistoserdova. 2005. Analysis of gene islands involved in methanopterin-linked C1 transfer reactions reveals new functions and provides evolutionary insights. J. Bacteriol. 187:4607-4614.[Abstract/Free Full Text]
  45. Kamerbeek, N. M., M. J. H. Moonen, J. G. M. van der Ven, W. J. H. van Berkel, M. W. Fraaije, and D. B. Janssen. 2001. 4-Hydroxyacetophenone monooxygenase from Pseudomonas fluorescens ACB. A novel flavoprotein catalyzing Baeyer-Villiger oxidation of aromatic compounds. Eur. J. Biochem. 268:2547-2557.[Medline]
  46. Kane, S. R., H. R. Beller, T. C. Legler, C. J. Koester, H. C. Pinkart, R. U. Halden, and A. M. Happel. 2001. Aerobic biodegradation of methyl tert-butyl ether by aquifer bacteria from leaking underground storage tank sites. Appl. Environ. Microbiol. 67:5824-5829.[Abstract/Free Full Text]
  47. Kane, S. R., T. C. Legler, L. M. Balser, and K. T. O'Reilly. 2003. Aerobic biodegradation of MTBE by aquifer bacteria from LUFT sites, p. E-12. In V. S. Magar and M. E. Kelley (ed.), Proceedings of the Seventh International In Situ and On-Site Bioremediation Symposium. Battelle Press, Columbus, OH.
  48. Kitiyama, A., E. Suzuki, Y. Kawakami, and T. Nagamune. 1996. Gene organization and low regiospecificity in aromatic-ring hydroxylation of a benzene monooxygenase of Pseudomonas aeruginosa JI104. J. Ferment. Bioeng. 82:421-425.[CrossRef]
  49. Kofoid, E., C. Rappleye, I. Stojiljkovic, and J. Roth. 1999. The 17-gene ethanolamine (eut) operon of Salmonella typhimurium encodes five homologues of carboxysome shell proteins. J. Bacteriol. 181:5317-5329.[Abstract/Free Full Text]
  50. Kotani, T., T. Yamamoto, H. Yurimoto, Y. Sato, and N. Kato. 2003. Propane monooxygenase and NAD+-dependent secondary alcohol dehydrogenase in propane metabolism by Gordonia sp. strain TY-5. J. Bacteriol. 185:7120-7128.[Abstract/Free Full Text]
  51. Liu, C. Y., G. E. Speitel, Jr., and G. Georgiou. 2001. Kinetics of methyl t-butyl ether cometabolism at low concentrations by pure cultures of butane-degrading bacteria. Appl. Environ. Microbiol. 67:2197-2201.[Abstract/Free Full Text]
  52. Lopes Ferreira, N., D. Labbe, F. Monot, F. Fayolle-Guichard, and C. W. Greer. 2006. Genes involved in the methyl tert-butyl ether (MTBE) pathway of Mycobacterium austroafricanum IFP 2012. Microbiology 152:1361-1374.[Abstract/Free Full Text]
  53. Malito, E., A. Alfieri, M. W. Fraaije, and A. Mattevi. 2004. Crystal structure of a Baeyer-Villiger monooxygenase. Proc. Natl. Acad. Sci. USA 101:13157-13162.[Abstract/Free Full Text]
  54. Mergeay, M., S. Monchy, T. Vallaeys, V. Auquier, A. Benotmane, P. Bertin, S. Taghavi, J. Dunn, D. van der Lelie, and R. Wattiez. 2003. Ralstonia metallidurans, a bacterium specifically adapted to toxic metals: towards a catalogue of metal-responsive genes. FEMS Microbiol. Rev. 27:385-410.