Polycyclic aromatic hydrocarbon (PAH)-degrading genes nidA and nidB that encode the α and β subunits of the aromatic ring-hydroxylating dioxygenase have been cloned and sequenced from Mycobacterium vanbaalenii PYR-1 [Khan et al., Appl. Environ Microbiol. 67 (2001) 3577–3585]. In this study, the presence of nidA and nidB in 12 other Mycobacterium or Rhodococcus strains was investigated. Initially, all strains were screened for their ability to degrade PAHs by a spray plate method, and for the presence of the dioxygenase Rieske center region by polymerase chain reaction (PCR). Only Mycobacterium sp. PAH 2.135 (RJGII-135), M. flavescens PYR-GCK (ATCC 700033), M. gilvum BB1 (DSM 9487) and M. frederiksbergense FAn9T (DSM 44346), all previously known PAH degraders, were positive in both tests. From the three positive strains, complete open reading frames of the nidA and nidB genes were amplified by PCR, using primers designed according to the known nidA and nidB sequences from PYR-1, cloned in the pBAD/Thio-TOPO vector and sequenced. The sequences showed >98% identity with the M. vanbaalenii PYR-1 nidA and nidB genes. Southern DNA–DNA hybridization using nidA and nidB probes from PYR-1 revealed that there is more than one copy of nidA and nidB genes in the strains PYR-1, BB1, PYR-GCK and FAn9T. However, only one copy of each gene was observed in PAH2.135.
Polycyclic aromatic hydrocarbons (PAHs) are ubiquitous environmental pollutants. Because of their human and eco-toxicity, there is a considerable interest to determine the fate of these compounds in the environment and to consider the possible use of microorganisms for remediation of polluted sites [1,2].
Bacterial degradation of PAHs under aerobic conditions begins with oxidation of the aromatic ring, catalyzed by dioxygenases . In this reaction, both atoms of molecular oxygen are incorporated into the PAH to form cis-dihydrodiol metabolites. Aromatic ring-hydroxylating dioxygenases are multicomponent enzyme systems consisting of an electron transport chain and a terminal dioxygenase . The terminal dioxygenase is composed of large (α) and small (β) subunits [4,5]. The α subunit is the catalytic component and contains two conserved regions: the [Fe2–S2] Rieske center and the mononuclear iron binding domain, which are involved in the consecutive electron transfer to the dioxygen molecule . Both α and β subunits are necessary for function and in determining the substrate specificity of the dioxygenase . Most information about metabolic pathways, enzymes, and genes involved in PAH degradation comes from studies on Gram-negative bacteria [4,5,8,9]. Genetic analysis and biochemical mechanism data on PAH degradation by Gram-positive bacteria, including Rhodococcus, Mycobacterium, Nocardioides, and Terrabacter species, are less abundant and have been mostly reported in the past three years [10–16]. However, Gram-positive bacteria may play more important roles than Gram-negative isolates in environmental degradation of high molecular mass PAHs, especially in degradation of the four-ring compound pyrene [17–20].
Mycobacterium vanbaalenii strain PYR-1 (isolated from a petrogenic chemical-polluted site ) can mineralize pyrene, fluoranthene, naphthalene, anthracene, phenanthrene and biphenyl, and in minor amounts also benzo[a]pyrene, 1-nitropyrene and 6-nitrochrysene [21,22]. Recently, genes encoding α NidA polypeptide (nidA gene) and β NidB polypeptide (nidB gene) of the terminal dioxygenase have been cloned from M. vanbaalenii strain PYR-1, expressed and sequenced in our laboratory . They represent the first described sequences of these genes from the genus Mycobacterium. Except for the new results published in the present study, the most similar enzyme known to date is the phenanthrene dioxygenase from Nocardioides sp. KP7, with the large subunit PhdA only 57% identical to NidA . The present research goal is to elucidate the presence and diversity of nidA and nidB genes within the genus Mycobacterium.
2Materials and methods
Pyrene and phenanthrene were purchased from Chem Service (Media, PA, USA). All of the PAHs and related compounds were >99% pure. Other chemicals were of highest purity commercially available (Sigma, St. Louis, MO, USA).
2.2Bacterial strains, media and cultivation conditions
Mycobacterium and Rhodococcus strains used in this study are listed in Table 1. For cloning purposes, Escherichia coli TOP101 competent cells and pBAD/Thio-TOPO vector were used (pBAD/TOPO Thiofusion Expression system, Invitrogen, Carlsbad, CA, USA). Middlebrook medium from Remel (Lenexa, KS, USA) was used as a cultivation medium for Mycobacterium and Rhodococcus strains. A minimal basal salts medium with low level nutrients  was used as a base for PAH utilization experiments. To enhance growth, an amended variant of this medium (sorbitol medium) was used as well, containing 9 g l−1 sorbitol as a carbon source and 0.5 g l−1 yeast extract. To determine the PAH-degradative potential of the studied strains, both minimal and sorbitol medium agar plates were coated with phenanthrene or pyrene by the spray plate technique , using acetone as a solvent. Mycobacterium and Rhodococcus strains were cultivated on these plates at 30°C for 5 days, then kept sealed at room temperature for at least a month. The formation of clearing zones was evaluated. The same media and conditions were used to prepare 3-day-old biomass for DNA isolation to avoid problems with flocculation and to maintain selection pressure for PAH degradation genes. For cultivation of E. coli containing recombinant plasmids, Luria–Bertani medium with 100 μg ml−1 ampicillin was used.
Table 1. Bacterial strains and results of the assaysa
a+, positive result; −, negative result; n, not performed.
bPlate spraying test: positive results in the case of Phe (phenanthrene) and Pyr (pyrene) mean formation of significant clearing in the PAH layer.
Total DNA from the studied strains was purified with the Qiagen genomic DNA extraction kit (Qiagen, Valencia, CA, USA) according to the manufacturer's instructions. To detect Rieske centers, the conserved [Fe2–S2] cluster binding region of terminal dioxygenases, a mixture of degenerate primers P1.1.f, P.1.2.f, P2.1.f and P2.2.f  and, independently, primer pair DP1 and DP2 published in  were used (Table 2). The PCR reaction mix consisted of 50 mM KCl, 10 mM Tris–HCl pH 9, 0.1% Triton X-100, 1.5 mM MgCl2, 0.2 mM each dNTP, 0.025 U μl−1Taq DNA polymerase (Qiagen), 1.5 μM primers and 0.01 μg template DNA per 20 μl reaction mix. The initial hold of 3 min at 95°C was followed by 40 cycles of 1 min denaturation at 95°C, 1 min annealing at 53°C and 1 min extension at 72°C, followed by a final hold at 72°C for 7 min.
Primers for the amplification of nidA and nidB genes were designed from the published sequences of M. vanbaalenii PYR-1 nidA and nidB genes (GenBank accession numbers AF249301, AF249302) . Another set of primers (NidA1f, NidA1r, NidB1f and NidB1r) were used to amplify the full length of nidA and nidB genes, which aligned at the initiation codon and outside the open reading frame (ORF) (forward primer) and the termination codon and outside the ORF (reverse primer) (Table 2). These primer pairs were also used to detect the nidA and nidB homologues in total genomic DNA extracts of strains Mycobacterium sp. PAH 2.135 (RJGII-135), M. flavescens PYR-GCK, M. gilvum BB1 and M. frederiksbergense FAn9T, with M. vanbaalenii PYR-1 as a control. The concentration of the primers was 0.1 μg. Other PCR reaction components were present as described above. The PCR was performed as follows: 3 min at 95°C, 30 cycles of 30 s denaturation at 94°C, 1 min annealing at 58°C and 1 min extension at 72°C, and 7 min final extension at 72°C. For cloning of PCR products, the nidA and nidB genes were PCR-amplified using NidA1f, NidA1r and NidB1f, NidB1r primer pairs and the ProofStart DNA polymerase (Qiagen) according to the manufacturer's recommendations. The reaction conditions were the same as for detection of nidA and nidB, but the extension step was changed to 1 min 40 s when amplifying nidA.
2.4Pulsed-field gel electrophoresis (PFGE) and Southern hybridization
Agarose plugs were used to avoid shearing of DNA and to ensure the total genomic DNA was analyzed. Bacterial cultures were grown on sorbitol medium coated with phenanthrene at 30°C. After 72 h, a suspension of each culture with an adjusted turbidity of 0.69–0.73% in TE buffer (10 mM Tris–HCl, 1 mM EDTA, pH 8.0) was made using a turbidity meter (Dade Behring). 60 μl of 10 mg ml−1 lysozyme (Sigma) solution was added to 240 μl of cell suspension and incubated for 30 min at 37°C. Subsequently, 10 μl of 0.1 mg ml−1 mutanolysin (Sigma) was added. The next steps, including embedding the cells in the agarose plugs, lysis, washing, XbaI restriction enzyme digestion and the separation of high molecular mass restriction fragments by PFGE, were performed as described previously .
The DNA in the gels was fixed, denatured and neutralized according to the Genius system user's guide for filter hybridization (Boehringer Mannheim, Indianapolis, IN, USA). Afterwards, the DNA was transferred onto nylon membranes by capillary action as described previously . The digoxigenin (DIG)-labeled probes for nidA and nidB genes were prepared using NidAf, NidAr and NidBf, NidBr primer pairs and DIG-PCR labeling reaction kit (Roche Diagnostics Corporation, Indianapolis, IN, USA). DNA UV crosslinking, Southern hybridization with DIG-labeled probes and detection were performed according to the kit manufacturer's instructions (Roche, DIG DNA labeling and detection kit).
2.5Cloning and DNA sequencing of the nidA and nidB genes
The PCR-amplified products nidA and nidB were cloned into pBAD/Thio-TOPO vector (Invitrogen) according to the manufacturer's recommendations. The resultant plasmids were isolated with the Qiagen plasmid miniprep kit. The nucleotide sequences of the nidA and nidB genes were determined with a model 377 DNA sequencer (Applied Biosystems, Foster City, CA, USA). Both strands were sequenced by primer walking with synthetic oligonucleotide primers (Table 2). DNA sequence analysis, translation, and alignment with other related genes and proteins were done by using the computer program Lasergene (DNASTAR, Madison, WI, USA).
2.6DNA sequence analysis
DNA sequence analysis, translation, and alignment with related genes and proteins were carried out using Lasergene (DNASTAR) and Align Plus (Scientific Educational Software, State Line, PA, USA) software. The GenBank program Blast  was used to find similar genes and proteins.
3Results and discussion
3.1Screening of PAH degradation by the spray plate method
Thirteen Mycobacterium or Rhodococcus strains used in this study were screened for PAH degradation by the spray plate technique using pyrene and phenanthrene as substrates (Table 1). Only those strains known from previous studies as PAH degraders, i.e. Mycobacterium sp. PAH 2.135 (RJGII-135), M. flavescens PYR-GCK, M. gilvum strain BB1, M. frederiksbergense FAn9T and M. vanbaalenii PYR-1, showed clearing zones around colonies that had been sprayed with phenanthrene or pyrene solutions. The remaining eight strains showed no clearance zones (Table 1).
3.2PCR amplification of the Rieske center
Mycobacterium and Rhodococcus strains were screened for the presence of aromatic ring-hydroxylating dioxygenase by using degenerate PCR primers [24,25] designed from the [Fe2–S2] binding region (Rieske center) which is a conserved sequence of terminal dioxygenases (Table 2). PCR products of the expected size (78 bp) were amplified only in the known PAH-degrading bacteria. The rest of the screened strains were negative for the PCR (Fig. 1, Table 1). These results suggest that those strains, which were positive by the spray plate method, possess terminal dioxygenases.
3.3PCR amplification, cloning and nucleotide sequencing of nidA and nidB genes
The primers designed according to the known nidA and nidB sequences from strain PYR-1 amplified complete coding sequences of nidA and nidB genes from strains PYR-GCK, BB1 and FAn9T (Fig. 2). These data suggest that these three strains, although different species, have highly conserved nidA and nidB sequences. However, in strain PAH 2.135, which was positive by the spray plate method and the Rieske center PCR method, the nidA gene was not amplified (Table 1). This may be due to low homology with the PYR-1 nidA gene. The PCR-amplified nidA and nidB genes from strains PYR-GCK, BB1 and FAn9T were cloned into the pBAD/Thio-TOPO vector. Sequences of the cloned nidA and nidB genes were deposited in the GenBank/EMBL DNA database under accession numbers AF548343–AF548348.
The length of the ORF of the nidA gene for M. flavescens PYR-GCK was 1368 bp, the same as the previously published nidA sequence for M. vanbaalenii PYR-1 . However, in the case of M. frederiksbergense FAn9T and M. gilvum BB1, the ORF of the nidA gene was 1377 bp, encoding 458 amino acids. The deduced translated polypeptides of these genes showed differences of three, six and nine amino acid throughout the whole 458-amino acid alignment for M. flavescens PYR-GCK, M. frederiksbergense FAn9T and M. gilvum BB1 respectively (Jotun–Hein alignment, Lasergene software, DNASTAR). Thus, these translated polypeptides are 99.3%, 98.6% and 98.0% identical to the NidA polypeptide from M. vanbaalenii, respectively. In the case of the NidB polypeptide, the identity was 98.8%, 98.2% and 99.4%; or differences of one, three and two amino acid throughout the uniform protein size of 169 amino acids. The Rieske center iron–sulfur binding site , CXHRGX8GNX5CXZHG, was found to be conserved in all deduced NidA proteins. Also, two histidine residues and one aspartate residue, which according to Parales et al.  bind the mononuclear iron, as well as one aspartate residue proposed to play an important role in electron transfer to mononuclear iron  were conserved (alignment not shown). According to high sequence similarity to M. vanbaalenii PYR-1 and preservation of conserved residues, it can be concluded that nidA and nidB are probably functional genes.
3.4Screening for nidA and nidB genes in Mycobacterium spp. by Southern hybridization
Strain Mycobacterium sp. PAH 2.135, which was not PCR-amplified for the nidA gene but was able to produce a clear zone by the spray method and was positive in the Rieske center PCR test, was tested for the presence of nidA and nidB genes by Southern hybridization. Four other PAH-degrading Mycobacterium strains, PYR-1, PYR-GCK, BB1, FAn9T (positive control), and five strains that did not degrade PAHs (negative control) were included for Southern hybridization studies to confirm the previous results (Table 1). Fig. 3 shows the comparison between Southern blot patterns of XbaI-digested DNA blotted with the nidA and nidB gene probes respectively (negative controls listed in Table 1). One band from Mycobacterium sp. PAH 2.135 (RJGII-135) hybridized at 48 kb with both nidA and nidB probes (Fig. 3, lanes 1 and 6). These data suggest that nidA and nidB genes are present in strain PAH2.135. However, given the weak signal by Southern hybridization (and the lack of nidA PCR product), their sequence homology to PYR-1 may be lower as compared to the other three Mycobacterium strains PYR-GCK, BB1 and FAn9T. Probing of M. vanbaalenii PYR-1, M. flavescens PYR-GCK, M. gilvum BB1 and M. frederiksbergense FAn9T with nidA and nidB probes revealed the presence of multiple bands (Fig. 3). Restriction enzyme XbaI was selected because both nidA and nidB genes from these strains do not have XbaI restriction sites. These data suggest that M. vanbaalenii PYR-1, M. flavescens PYR-GCK, M. gilvum BB1 and M. frederiksbergense FAn9T contain more than one nidA and nidB gene in their genomes. We are currently investigating whether essentially identical copies or different homologous genes are present. Several aromatic compound-degrading microorganisms are known to possess different ring-hydroxylating dioxygenases within the same strain [29,30].
This study showed that four Mycobacterium spp. (Mycobacterium sp. PAH 2.135, M. flavescens PYR-GCK, M. gilvum BB1, M. frederiksbergense FAn9T and M. vanbaalenii PYR-1) possess nidA and nidB genes, although they are phylogenetically distant species based on 16S rDNA sequence comparisons [31–33]. On the other hand, M. austroafricanum ATCC 33464 does not have a nidA gene (Table 1), although it is almost identical to M. vanbaalenii at the 16S rRNA sequence level [26,34]. Bogan et al.  reported no or very limited mineralization of phenanthrene, fluoranthene or pyrene by M. austroafricanum ATCC 33464. However, another M. austroafricanum strain, GTI-23, utilizes a wide range of PAHs . To the best of our knowledge, GTI-23 has not yet been tested for the presence of the nid genes. Nevertheless, a possible explanation for these observations is that the Mycobacterium strains obtained the nid genes later in evolution, possibly by horizontal transfer. The high similarity of the nidA and nidB genes sequenced in this study to each other and to the genes in strain PYR-1 also favors this hypothesis. Whatever the common origin of nidA and nidB genes is, they are found at various geographical locations, as documented by the sites of isolation of microorganisms, USA, Germany and Denmark (Table 1). Truncated nidA-like sequences obtained by reverse transcription PCR from soil were reported in GenBank by a British research group (accession numbers AY032941, AY032938, AY032940, AY032939, AY032942). Also, pdoA1 and pdoB1 sequences recently submitted from France (accession numbers MYC494745 and MYC494744) are 98% and 100% identical to M. vanbaalenii nidA and nidB at the translated protein level. The wide distribution of PAH-degrading genes almost identical to the ones in M. vanbaalenii PYR-1 enhances the importance of previous studies performed on this strain. The consideration of M. vanbaalenii PYR-1 as a model strain among PAH-degrading mycobacteria is supported in this way.
This work was supported by the Oak Ridge Institute for Science and Education Postgraduate and Faculty Research Program at the National Center for Toxicological Research, US/FDA Jefferson, AR. We thank John B. Sutherland, Robert D. Wagner and Robin Stingley for critical review of the manuscript and Sandra Malone for graphic assistance.