Previous Article | Next Article ![]()
Applied and Environmental Microbiology, February 2006, p. 1045-1054, Vol. 72, No. 2
0099-2240/06/$08.00+0 doi:10.1128/AEM.72.2.1045-1054.2006
Ohgew Kweon,1,
James P. Freeman,2
Richard C. Jones,3
Michael D. Adjei,1
Jin-Woo Jhoo,4
Ricky D. Edmondson,3 and
Carl E. Cerniglia1*
Division of Microbiology,1 Division of Biochemical Toxicology,2 Division of Systems Toxicology, National Center for Toxicological Research/U.S. FDA, Jefferson, Arkansas 72079,3 Department of Food Science and Technology in Animal Resources, Kangwon National University, Chuncheon 200-701, Republic of Korea4
Received 1 August 2005/ Accepted 7 October 2005
|
|
|---|
and ß subunits of terminal dioxygenase, were subsequently cloned and sequenced. The deduced enzyme revealed close similarities to the corresponding PAH ring-hydroxylating dioxygenases from Mycobacterium and Rhodococcus spp. but had the highest similarity, 61.9%, to the
subunit from Nocardioides sp. strain KP7. The
subunit also showed 52% sequence homology with the previously reported NidA from M. vanbaalenii PYR-1. The genes nidA3B3 were subcloned into the expression vector pET-17b, and the enzyme activity in Escherichia coli cells was reconstituted through coexpression with the ferredoxin (PhdC) and ferredoxin reductase (PhdD) genes of the phenanthrene dioxygenase from Nocardioides sp. strain KP7. The recombinant PAH dioxygenase appeared to favor the HMW PAH substrates fluoranthene, pyrene, and phenanthrene. Several other PAHs, including naphthalene, anthracene, and benz[a]anthracene, were also converted to their corresponding cis-dihydrodiols. The recombinant E. coli, however, did not show any dioxygenation activity for phthalate and biphenyl. The upregulation of nidA3B3 in M. vanbaalenii PYR-1 induced by PAHs was confirmed by reverse transcription-PCR analysis. |
|
|---|
Mycobacterium vanbaalenii PYR-1 (29) was isolated from oil-contaminated polluted sediment based on its ability to degrade pyrene (21). This bacterium can mineralize various aromatic hydrocarbons, including biphenyl, naphthalene, anthracene, fluoranthene, pyrene, 1-nitropyrene, phenanthrene, benzo[a]pyrene, benz[a]anthracene, and 7,12-dimethylbenz[a]anthracene (18-21, 25-28, 45-49, 51). Analysis of metabolites indicates that the oxidation of PAHs in the strain PYR-1 is primarily initiated by dioxygenation (21, 30, 34, 46-49) with minor monooxygenation of the aromatic ring (7). The enzyme catalyzing dioxygenation belongs to a family of aromatic-ring-hydroxylating dioxygenases that introduce two atoms of oxygen into aromatic hydrocarbons to form a cis-dihydrodiol. They are multicomponent enzyme systems consisting of an electron transport chain and a terminal dioxygenase composed of large (
) and small (ß) subunits (43). The enzyme system has been extensively studied in many microorganisms, since this initial reaction mostly determines aromatic substrate ranges that can be degraded. However, a significant amount of information comes exclusively from dioxygenases for monocyclic aromatics or low-molecular-weight PAHs. In M. vanbaalenii PYR-1, identification of the genes encoding PAH degradation indicates that this bacterium has at least three copies of the terminal dioxygenase gene (6, 30, 59). Among them, nidAB and phtAaAb were shown to transform the PAHs phenanthrene and pyrene into phenanthrene and pyrene cis-dihydrodiols (30, 59) and phthalate into phthalate cis-3,4-dihydrodiol (58), respectively. In recent years, copies of terminal dioxygenase genes have been reported from HMW-PAH-degrading bacteria (6, 10, 14, 38, 44, 56). However, detailed information in regard to the actual role and properties of each gene product is limited, and no gene product has been assigned as a dioxygenase that significantly oxidizes HMW PAHs to cis-dihydrodiols. Previously, a 20-kDa protein which showed upregulated expression in response to the exposure of M. vanbaalenii PYR-1 to several PAHs was identified by two-dimensional gel electrophoresis (2-DE) and protein mass spectrometry (MS) as a ß subunit of terminal dioxygenase (32). In this study, a gene cluster encoding the 20-kDa polypeptide was isolated and characterized. We cloned the fourth copy of the terminal dioxygenase gene with the aid of PCR and PYR-1 genome library screening. We then examined the substrate specificities and transformation rates of the enzyme that was expressed in Escherichia coli.
|
|
|---|
Bacterial strains, plasmids, reagents, and growth conditions.
The bacterial strains, vectors, and plasmids used in this study are listed in Table 1. E. coli strains were grown at 30°C or 37°C on Luria-Bertani (LB) medium supplemented with 100, 15, and 10 µg/ml of ampicillin, tetracycline, and gentamicin, respectively (54). M. vanbaalenii PYR-1 was grown at 30°C in Middlebrook 7H10 (Remel, Lenexa, KS) or in phosphate-based minimal medium (33) with 2% sorbitol and PAHs as described in a previous report (32). Antibiotics and all other chemicals were purchased from Sigma.
|
View this table: [in a new window] |
TABLE 1. Bacterial strains, phage, and plasmids used in this study
|
genomic library (67) was then screened using this 109-bp DNA fragment as a probe, which had been labeled with a digoxigenin oligonucleotide 3' end labeling kit (Roche Applied Science). A plaque-lift hybridization procedure was used for the screening as recommended by the digoxigenin system user's guide (Roche Applied Science). Plaques showing a positive signal were in vivo excised to produce phagemid clones using ExAssist Helper Phage with the E. coli XLOLR strain (Stratagene, La Jolla, CA). The DNA sequence was determined on both strands using a model 377 DNA sequencer (Applied Biosystems, Foster City, CA). Homology search was performed using the BLASTX program (1), and DNA sequences were processed using Lasergene (DNASTAR) software. Protein sequences were aligned with the CLUSTAL X program, version 1.81 (64), and phylogenetic trees were inferred from the alignments by the PHYLIP package version 3.63 (12, 31). The graphic for the phylogenetic tree was visualized by the TreeView program (50).
Construction of nidA3B3 dioxygenase expression system.
For gene expression in E. coli, the pET-17b expression system (Novagen, Madison, WI) was used. A DNA fragment containing nidA3B3 was amplified from M. vanbaalenii PYR-1 genomic DNA with primers EnidA3-F (5'-ACATATGGCGCCTGATGCGACGACAATG-3') and EnidB3-R (5'-CAAGCTTTTAGATCCAGAATGACAGGTT-3'). The underlined sequences are the NdeI and HindIII sites, respectively. The 1.87-kb PCR product was initially cloned into pCR2.1-TOPO (Invitrogen, Carlsbad, CA) to give pNCK32 and subjected to DNA sequencing to confirm that PCR amplification did not introduce mutations. The insert of plasmid pNCK32 was isolated by digestion with NdeI and HindIII. This fragment was ligated between the NdeI and HindIII site of pET-17b, resulting in plasmid pNCK33. Both plasmids pNCK33, containing the dioxygenase genes (nidA3B3), and pBRCD, expressing PhdCD of KP7 (38), were transformed into E. coli strain BL21(DE3).
Analysis and identification of PAH metabolites.
The transformed E. coli BL21(DE3)(pNCK33)(pBRCD) and BL21(DE3)(pNCK33) cultures and the control E. coli BL21(DE3)(pET-17b) culture were incubated overnight by inoculating a colony in 50 ml of LB medium at 30°C. A total of 5 ml of the overnight culture was transferred to 100 ml of LB medium and incubated under the same conditions. When the bacterial cells reached an optical density at 600 nm of 0.5 to 0.7, the culture was induced by adding IPTG (isopropyl-ß-D-thiogalactopyranoside) at a final concentration of 1 mM and further incubated for 2 h. The culture was harvested by centrifugation, washed with M9-glucose medium, and resuspended in 40 ml of the same solution supplemented with either ampicillin or gentamicin. PAHs were added to a final concentration of 200 µM, and the cells were further incubated overnight at 30°C. The supernatant was extracted with ethyl acetate, which was dried by a rotary evaporator, and the dried residues were dissolved in methanol for high-performance liquid chromatography (HPLC) analysis. Metabolites were resolved, isolated, and purified with an HPLC system (Hewlett-Packard 1100 series). A C18 column (Prodigy 5 µm ODS-3, 260 by 40 mm; Phenomenex, Torrance, CA) was used with a mobile phase of a 30-min linear gradient of methanol-water (from 45:55 to 95:5 [vol/vol]) at a flow rate of 1.0 ml/min.
Gas chromatography-mass spectrometry (GC-MS) was performed on either a Finnigan TSQ 700 or a 7000 triple quadrupole mass spectrometer (Thermo Finnigan, San Jose, CA) with separation of the metabolites using a J&W DB-5ms capillary column (30 m by 0.25 mm, 0.25-µm film thickness). GC-MS analyses were performed with a capillary column temperature rate of 10°C/min and a total analysis time of 30 min or less, depending upon the PAH analyzed. Retention times between the two instrument configurations were slightly different. Electron ionization (EI) was achieved at 70 eV electron energy and 150°C ion source temperature. Derivatization prior to GC/EI-MS analysis was performed by silylation with N,O-bis(trimethylsilyl)-trifluoroacetamide with 1% trimethylchlorosilane (Regis Technologies, Morton Grove, IL). The samples were dissolved in 400 µl of acetonitrile. Two hundred microliters of dissolved sample and 100 µl of silylation reagent were mixed in an Xpertek high-recovery 1.5-ml sample vial (P.J. Cobert, St. Louis, MO) sealed with a septum cap and allowed to react for 1 h at 68°C. The injection volumes were 0.5 µl. All mass spectrometric measurements were at low resolution, and no tandem mass spectrometry methods were employed. Therefore, all fragmentation losses are reported as assumptions based on the proposed structures and available moieties.
RT-PCR analysis.
Levels of nidA3 RNA in induced M. vanbaalenii PYR-1 cultures and uninduced cultures were compared by reverse transcription (RT)-PCR analysis. The total RNA was extracted from cells grown in phosphate-based minimal medium supplemented with 100 µM naphthalene, anthracene, phenanthrene, fluoranthene, pyrene, benz[a]anthracene, or sorbitol (2%) using RNA-Bee (Tel-Test, Inc., Friendswood, TX). RNA was further extracted with phenol-chloroform. RNA was then recovered after precipitation with ethanol and treatment with RNase-free DNase I (Promega). PCR without reverse transcription was performed with the DNase-treated RNA to ensure the complete digestion of DNA. RT-PCR analyses were carried out using an Access RT-PCR system (Promega) according to the manufacturer's instructions. Two primer sets were used for RT-PCR. Primer set RnidA3-F3 (5'-GCCGAGATATTCCAGGCTATTA-3') and RnidB3-R3 (5'-CGCATCTTCAGACTCGTGTAGT-3') amplified the intergenic 469-bp sequence between nidA3 and nidB3, and the other primer set, R16S-F2 (5'-GAGAAGAAGGACCGGCCAACTACG-3') and R16S-R2 (5'-AGACCCCGATCCGAACTGAGACC-3'), was used to produce an 826-bp fragment corresponding to 16S rRNA as an amplification control. The RT-PCR conditions were as follows: reverse transcription at 48°C for 45 min followed by 95°C for 15 min, 30 cycles each consisting of 94°C for 1 min, 53°C for 1 min, and 68°C for 1 min, with a final elongation step at 68°C for 7 min. As a positive control, genomic DNA from M. vanbaalenii PYR-1 was used as a template.
Nucleotide sequence accession number.
The gene sequences were deposited in the GenBank database under accession number DQ028634.
|
|
|---|
and ß subunits of the dioxygenase components of bacterial aromatic-ring-hydroxylating dioxygenases. The translated amino acid sequence of the ß subunit was in perfect agreement with the MS-generated de novo peptide sequences (LRHYLTNVR, SGEDGDEYR, TSSDWAEDPPSR, VTSNALLVR, AETVSAER, AGGPGFSER, SMHLSDNYTSLK, and TTVPGADPP). The calculated molecular mass of NidB3 (21 kDa) is similar to the observed molecular mass of the P18 polypeptide, which was examined by 2-DE (32).
![]() View larger version (72K): [in a new window] |
FIG. 1. (A) 2-DE protein expression profile of M. vanbaalenii PYR-1 induced by PAH fluoranthene. (B) Relative synthesis rates of the 20-kDa polypeptide from 2-DE analysis of M. vanbaalenii PYR-1 during growth with glucose (Con), pyrene (Pyr), phenanthrene (Phe), anthracene (Ant), and fluoranthene (Flu). The inset below each bar shows a 2-DE spot image of the corresponding P18 polypeptide from each condition.
|
![]() View larger version (32K): [in a new window] |
FIG. 2. Physical map of the phagemid clone pNCK31 carrying nidA3B3 genes and conserved sequence alignment of the four subunits of ring-hydroxylating dioxygenase discovered in M. vanbaalenii PYR-1. The amino acid residues involved in binding to the Rieske-type [2Fe-2S] cluster and to the mononuclear Fe(II) atom are indicated by highlighted characters.
|
subunits recently identified from M. vanbaalenii PYR-1. Although NidA3 and NidB3 share common sequence features with other terminal dioxygenases, the overall identity of the NidA3B3 sequence with those of other dioxygenases was moderate. The
subunit of a phenanthrene dioxygenase from Nocardioides sp. strain KP7 showed the highest degree of sequence similarity to NidA3 (61.7%). It also exhibited 52 to 60% identity with that of several Mycobacterium spp. (6, 14, 38, 44, 56) and 37 to 48% with Rhodococcus spp. (36, 40, 65), whereas other well-known
subunits from Pseudomonas and Sphingomonas spp. revealed levels of sequence identity lower than 32%. The enzyme showed 52% sequence homology with the previously reported NidA from M. vanbaalenii PYR-1. The NidB3 protein exhibited 46% identity at most to the ß subunit from Mycobacterium spp. and about 42% to those from strain KP7 and other Rhodococcus spp. A sequence comparison of NidA3 with three other
subunits from M. vanbaalenii PYR-1 and other known
subunits of ring-hydroxylating dioxygenases is presented in Fig. 3. In this phylogenetic analysis, NidA3 clustered with groups of the
oxygenase subunits that catalyzed the hydroxylation of PAHs.
![]() View larger version (48K): [in a new window] |
FIG. 3. Phylogenetic tree of NidA3 obtained from alignment with related proteins. The protein sequences of the 26 subunits of ring-hydroxylating dioxygenases, including NidA3, are classified. The multiple-alignment analysis was performed with the PHYLIP software package, and the phylogenetic unrooted tree was drawn by using TreeView. The numbers on branches refer to the percentage confidence, estimated by a bootstrap analysis with 1,000 replications. Scale bar, percentage divergence. GenBank accession numbers for the sequences are as follows: DbfA1, AB054975; NdoB, M23914; PahA3, D84146; DxnA1, X72850; TodC1, J04996; BphA1, D17319; BphA, M86348; ORF25, AY365117; CmtAb, U24215; TftA, U11420; AntA, AF071556; XylX, M64747; BenA, AF009224; TdnA1, D85415; AtdA3, D86080; DitA1, AF119621; PhdA, AB017794; PdoA2, AJ494743; NidA, AY365117; PdoA1, AJ494745; NidA, AF121905; NarAa, AF082663; PhtAa, AY365117; PhtA1, AB084235; PhtA1, AF331043.
|
and ß subunits, respectively, was different from the previously identified dioxygenase nidAB of M. vanbaalenii PYR-1, where the gene order was found to be the reverse (30). The presence of the putative MarR-family regulator (nidR) has not been reported for other Mycobacterium spp., such as strain 6PY1 (38) or S65 (56), or for Nocardioides sp. KP7 (53) or Rhodococcus sp. I24 (65).
Functional analysis of nidA3B3.
To test whether the NidA3B3 proteins conferred any aromatic-ring-hydroxylating abilities to E. coli, the genes were cloned into pET-17b to construct the expression vector pNCK33. The E. coli strain transformed with pNCK33 seemed to actively express the NidA3B3 enzyme; however, no PAH degradation activity was detected. It appeared that the enzyme system was inactive, probably due to the lack of an electron carrier system, which often consists of reductase and ferredoxin, two components normally needed for electron transfer to the terminal dioxygenase. Since none of these genes were found in the 3.7-kb clone, the second plasmid, providing the electron transfer components from another system, was introduced into E. coli BL21(DE3)(pNCK33). For this purpose, the plasmid pBRCD, containing the cistrons encoding the ferredoxin (phdC) and ferredoxin reductase (phdD) of the phenanthrene dioxygenase from Nocardioides sp. strain KP7 (38), was used. The plasmid pBRCD is a derivative of pBBR11MCS-5 (37) and compatible with pET-17b. The enzyme activity was then achieved by coexpressing these two plasmids, which resulted in the hydroxylation of aromatic substrates.
To determine the specificity of the NidA3B3 enzyme, phthalate, biphenyl, naphthalene, phenanthrene, anthracene, pyrene, fluoranthene, and benz[a]anthracene were tested (Table 2). The accumulated products formed from aromatic substrates were analyzed by HPLC and GC-MS (Fig. 4). The observed PAH intermediates in this study were compared to authentic compounds and to the previously identified initial PAH catabolic intermediates proposed from the biotransformation experiments using the wild-type M. vanbaalenii PYR-1 whole cells. The ethyl acetate-extractable metabolites were analyzed by GC-MS as trimethylsilylated derivatives. In many cases, since cis-PAH dihydrodiol metabolites easily undergo dehydration during analysis, the phenolic compounds detected are interpreted as the thermal dehydration products. In addition, dihydroxylated PAH (diols) compounds were also detected as PAH metabolic products, which are considered to be catalyzed by a nonspecific dehydrogenase(s) from the host E. coli. As shown in Table 2, the relative conversion percentages for initial oxidation of PAHs showed differences in substrate preference. Dihydroxylated metabolites of phthalate and biphenyl were not detected in the GC-MS and HPLC assays. Figure 4 shows GC-extracted ion chromatograms of naphthalene, phenanthrene, anthracene, fluoranthene, pyrene, and benz[a]anthracene degradation products from E. coli (DE3)(pNCK33)(pBRCD). In the analysis for naphthalene biotransformation (Fig. 4A), the detected metabolite had the same GC retention time (10.58 min), GC-MS properties (molecular ion at m/z 306) (Fig. 5A), and UV characteristics (
max, 262 nm) as authentic cis-1,2-dihydroxy-1,2-dihydronaphthalene (26). Small amounts of 1- and 2-naphthol (molecular ion at m/z 216) gave an identical GC retention time and mass spectral fragment ions identical to authentic compounds (Table 1). GC-MS analysis of the neutral extracts of phenanthrene metabolites produced by the recombinant E. coli cells gave two major (14.56 min and 15.61 min) and one minor (16.29 min) compound (Fig. 4B) that give a molecular ion at m/z 356 [M+·] (Fig. 5B) that is consistent with trimethylsilylated derivative of phenanthrene cis-dihydrodiol (34). These results are consistent with the previous reports in which M. vanbaalenii PYR-1 was proposed to generate three cis-dihydrodiols bearing hydroxyl groups at the 1,2-, 3,4-, and 9,10- positions (34, 35, 47). The UV absorption spectra of the two major compounds (44% and 21%) corresponded to those of phenanthrene 3,4- and 9,10-dihydrodiol, respectively (3, 47, 61). The minor component (17%) was tentatively identified as phenanthrene cis-1,2-dihydrodiol. Small amounts of 2- and 3-hydroxyphenanthrene and two phenanthrenediols were also detected and considered to be formed by the dehydration and dehydrogenation, respectively, of the dihydrodiols (Table 1). Approximately 85% of the phenanthrene added was converted to hydroxylated metabolites (Table 1). For anthracene bioconversion studies (Fig. 4C), GC-MS analysis of the derivatized ethyl acetate neutral extract showed a major metabolite (22%) eluting at 15.5 min with a molecular ion at m/z 356 and fragment ions (Fig. 5C) identical to authentic anthracene cis-1,2-dihydrodiol. HPLC analysis indicated that the metabolite has the same UV absorption properties as those reported for anthracene cis-1,2-dihydrodiol (47). Two other minor metabolites with molecular ions at m/z 356 were also detected at 14.02 min and 15.05 min in the GC-MS analysis (Fig. 4C). These compounds could be other dihydrodiol isomers; however, the position of substitution could not be determined due to limited material and the lack of authentic standards. Two additional metabolites with apparent molecular ions at m/z 354 in the GC-MS analysis were presumed to be anthracenediols formed from dehydrogenation of anthracene cis-dihydrodiols. Anthraquinone ([M+·] at m/z 208) was also detected as an autoxidation product. In the fluoranthene experiment, the GC-MS analysis of the trimethylsilylated product showed a major compound (61%) at 18.27 min (Fig. 4D), which is consistent with fluoranthene cis-dihydrodiol (molecular ion at m/z 380 in Fig. 5D). HPLC and UV spectral analysis indicated that this metabolite had a UV absorption spectrum similar to that reported for fluoranthene cis-2,3-dihydrodiol (52). As shown in Table 2, we also detected compounds whose mass spectra were similar to those of monohydroxyfluoranthene ([M+·] at m/z 290) and fluoranthenediols ([M+·] at m/z 378) which we considered to be derived from dehydration of fluoranthene cis-dihydrodiol. Thus, with fluoranthene as the substrate, the sum (>99) of the areas of all these metabolite peaks was the highest of the studied aromatic compounds. Fluoranthene was not detected in the culture medium after overnight incubation. In the pyrene analysis, two metabolites (17.49 min and 18.27 min) were determined by GC-MS to have the same molecular ion at m/z 380 (Fig. 4E and 5E), which is consistent with a pyrene cis-dihydrodiol (34). The compound eluting at 17.49 min had the same retention time as authentic pyrene cis-4,5-dihydrodiol. The predominant metabolite peak (53%) further detected by HPLC and had a UV spectrum identical to that of pyrene cis-4,5-dihydrodiol (2, 21). The minor metabolite at 18.27 min was tentatively identified as pyrene cis-1,2-dihydrodiol based on the symmetry of the pyrene structure and previous biotransformation studies in M. vanbaalenii PYR-1 (34). Two pyrenols ([M+·] at m/z 290) and two pyrenediols ([M+·] at m/z 378) were also detected, and the former compounds are believed to be dehydrated products from cis-pyrene dihydrodiols. GC-MS analysis of the benz[a]anthracene extract revealed two trimethylsilylated metabolites at 18.45 min and 18.70 min that gave molecular ions at m/z 406 (Fig. 4F and 5F). The mass spectra are consistent with benz[a]anthracene dihydrodiols (46). HPLC analysis indicated that the major compound eluting at 18.4 min had a UV absorption spectrum identical to that of benz[a]anthracene cis-10,11-dihydrodiol (46). The compound at 18.7 min (Fig. 4F) had a GC retention time identical to that of authentic benz[a]anthracene cis-5,6-dihydrodiol. In addition, we found benz[a]anthracene dione ([M+·] at m/z 258), which was due to autoxidation of benz[a]anthracene.
|
View this table: [in a new window] |
TABLE 2. GC-MS and HPLC data for the oxidation products showing PAH selectivity of NidA3B3 as expressed in E. coli
|
![]() View larger version (15K): [in a new window] |
FIG. 4. Extracted ion chromatograms of the trimethylsilylated metabolites formed from transformed E. coli BL21(DE3)(pNCK33)(pBRCD) cells incubated with naphthalene (A), phenanthrene (B), anthracene (C), fluoranthene (D), pyrene (E), and benz[a]anthracene (F). Metabolite abbreviations are as follows: naphthalene cis-1,2-dihydrodiol, N1,2D; phenanthrene cis-1,2-dihydrodiol, P1,2D; phenanthrene cis-3,4-dihydrodiol, P3,4D; phenanthrene cis-9,10-dihydrodiol, P9,10D; anthracene cis-1,2-dihydrodiol, A1,2D; fluoranthene cis-2,3-dihydrodiol, F2,3D; pyrene cis-1,2-dihydrodiol, PY1,2D; pyrene cis-4,5-dihydrodiol, PY4,5D; benz[a]anthracene cis-5,6-dihydrodiol, BaA5,6D; benz[a]anthracene cis-10,11-dihydrodiol, BaA10,11D.
|
![]() View larger version (20K): [in a new window] |
FIG. 5. EI mass spectra of the dihydrodiol trimethylsilyl ethers: naphthalene cis-dihydrodiol (A), phenanthrene cis-dihydrodiol (B), anthracene cis-dihydrodiol (C), fluoranthene cis-dihydrodiol (D), pyrene cis-dihydrodiol (E), and benz[a]anthracene cis-dihydrodiol (F).
|
![]() View larger version (16K): [in a new window] |
FIG. 6. Transcriptional analysis to compare levels of nidA3B3 mRNA in M. vanbaalenii PYR-1 cells grown with sorbitol (uninduced) or with PAHs (induced). For reference purposes, the PCR signal obtained with M. vanbaalenii PYR-1 genomic DNA is also shown. The length of DNA size standards (1,000 and 500 bp) in lane M and the sizes of transcript (826 and 469 bp) are indicated. Lanes 1 and 2, RT-PCR products from total RNA from PYR-1 cells grown with sorbitol; lanes 3 and 4, RT-PCR products from total RNA from cells grown with naphthalene; lanes 5 and 6, RT-PCR products from cells grown with anthracene; lanes 7 and 8, RT-PCR products from cells grown with phenanthrene; lanes 9 and 10, RT-PCR products from cells grown with fluoranthene; lanes 11 and 12, RT-PCR products from cells grown with pyrene; lanes 13 and 14, RT-PCR products from cells grown with benz[a]anthracene; lanes 15 and 16, PCR products from M. vanbaalenii PYR-1 genomic DNA. Samples in lanes 1, 3, 5, 7, 9, 11, 13, and 15 were with the primer set for nidA3B3 and lanes 2, 4, 6, 8, 10, 12, 14, and 16 were with the primer set for 16 rRNA.
|
|
|
|---|
subunit, which is typically coupled with the ß subunit, would be discovered by gene walking methodologies. As revealed in the sequence analysis, transcription of these genes was thought to be coupled, since the start codon of nidB3 overlaps with the stop codon of nidA3. This was further confirmed by an RT-PCR experiment, which showed that RNA extracts derived from cells grown with naphthalene, anthracene, phenanthrene, pyrene, and fluoranthene gave amplified products of the anticipated size, whereas RNA from cells grown with sorbitol barely amplified the gene product. From these results, we concluded that nidA3B3 is induced in response to PAHs other than benz[a]anthracene but not sorbitol, and these genes are therefore regulated at the transcriptional level. The results were consistent with the previous 2-DE investigation (32), in which the upregulation of NidB3 after exposure to PAHs in M. vanbaalenii PYR-1 was demonstrated at the protein level.
In the segment of cloned DNA flanking nidA3B3, we found two additional ORFs, designated orf1 and orf4, which are oriented in the same direction. The product of orf1 (NidR) showed similarity to various members of the putative MarR-family transcriptional regulators. Because many proteins belonging to this family have been known to control a variety of microbial functions, including antibiotic resistance and catabolism of various substrates (11, 22, 55), the gene product might be involved in the transcriptional control of nidA3B3. The deduced amino acid sequence of orf4 immediately downstream of nidA3B3 was similar to those of short-chain alcohol dehydrogenases, to which most cis-dihydrodiol dehydrogenase genes involved in the initial dehydrogenation of PAH cis-dihydrodiols normally belong. Since the dehydrogenase genes are often located in the neighboring region of dioxygenase genes (15, 40, 42, 56, 63, 65), the ORF4 protein may be the cis-dihydrodiol dehydrogenase enzyme that converts the products of NidA3B3 to their dihydroxy compounds.
Recombinant dioxygenase NidA3B3 produced in E. coli required coexpression of a [3Fe-4S] type ferredoxin and reductase gene for activity. Previously, the nidAB dioxygenase genes of M. vanbaalenii PYR-1 were reported to be functional in an E. coli system, which was not provided with ferredoxin and reductase components (30). At the time, as has been demonstrated by other investigations (39, 41, 57), it was speculated that the NidAB dioxygenase subunits might borrow the ferredoxin and reductase components of another electron transport system in the recombinant E. coli cells. In the current study, despite the resting cells of recombinant E. coli BL21(DE3)(pNCK33) containing nidA3B3 being assumed to actively express the nidA3B3 genes, no enzyme activity was detected with PAH substrates, as determined by HPLC analysis, which indicates that an additional supplemental electron transport system must be required for enzyme activity. As revealed in the sequence comparison, NidA3B3 is similar to those Rieske nonheme iron dioxygenases from strains of Nocardioides, Mycobacterium, Rhodococcus, Terrabacter, and Arthrobacter spp. Interestingly, without exception, these dioxygenase genes are always found with genes for [3Fe-4S] type iron-sulfur ferredoxin, if there are any genes for ferredoxin components in the nearby region (GenBank accession numbers AB031319, AB048709, AY502076, AF331043, and AB084235 for Nocardioides sp. KP7, Rhodococcus sp. RHA1, Rhodococcus sp. DK17, Arthrobacter keyseri 12B, and Terrabacter sp. DBF63, respectively). Among them, phdCD, encoding a [3Fe-4S] type ferredoxin and ferredoxin reductase component from Nocardioides sp. KP7 (53), was previously used with the PdoA2B2 dioxygenase from Mycobacterium sp. 6PY1 (38) as well as with the respective dioxygenase PhdAB from Nocardioides sp. KP7 for the recovery of enzyme activity. It is also worth noting that surrogate [2Fe-2S] type ferredoxins did not result in functional enzymes when coupled to dioxygenases requiring [3Fe-4S] ferredoxin as an electron carrier protein (42, 62). NidA3B3 dioxygenase activity was reconstituted when coexpressed with pBRCD carrying phdCD from Nocardioides sp. KP7.
In the biotransformation test using E. coli, a range of PAHs was converted by the action of NidA3B3 dioxygenase to their corresponding cis-dihydrodiol compounds (Fig. 4). Monohydroxylated compounds were also identified as oxidized metabolites produced during biotransformation analysis. The compounds were presumed to be derived by nonenzymatic loss of water of the corresponding cis-dihydrodiols produced by NidA3B3. Small amounts of dihydroxy compounds were also detected as PAH metabolic products. In fact, we previously reported a similar experimental result with M. vanbaalenii PYR-1's first dioxygenase, NidAB, which showed the biotransformation of pyrene into 4,5-dihydroxypyrene (pyrene-4,5-diol) (30). Thus, we speculated that these dihydroxy compounds in the present study might be the product of a nonspecific dehydrogenation reaction catalyzed by an E. coli enzyme. As shown in Table 2, fluoranthene was the most preferred substrate followed by phenanthrene, but pyrene, anthracene, and naphthalene were also oxidized significantly. The results showing the NidA3B3 enzyme's highest substrate preference for fluoranthene support our previous 2-DE study (32), in which the NidB (P18) polypeptide showed the highest upregulation when PYR-1 was incubated with fluoranthene compared to other PAHs. It was sevenfold higher than the next highest inducer, pyrene.
Based on the identification of isomeric cis-dihydrodiols formed during the degradation of phenanthrene, anthracene, benz[a]anthracene, pyrene, and benzo[a]pyrene by M. vanbaalenii PYR-1, it was proposed that this strain might have a dioxygenase with a relaxed substrate specificity for degrading a broad range of PAHs (34, 46-48). The results in this study indicate that this hypothesis has merit because the fourth dioxygenase, NidA3B3, extensively metabolized those representative (HMW) PAHs with positional hydroxylation, which resembled the pattern of PAH metabolism found in wild-type M. vanbaalenii PYR-1 (Fig. 7). Although the occurrence of bacterial species degrading HMW PAHs has been reported (6, 8, 10, 14, 38, 44, 56, 66), relatively little is known about HMW PAH dioxygenases. M. vanbaalenii PYR-1 was one of the first microorganisms found to utilize HMW PAHs (21). The nidA3B3 gene discussed in this investigation is the fourth copy of a terminal dioxygenase gene from M. vanbaalenii PYR-1, and the gene product NidA3B3 can oxidize PAHs, including HMW PAHs. To our knowledge, this is the first report to describe the dioxygenase which shows high enzyme activity for the four-ring HMW PAHs fluoranthene and pyrene.
![]() View larger version (29K): [in a new window] |
FIG. 7. Structures of products converted from aromatic substrates by recombinant E. coli cells.
|
This work was supported in part by an appointment to the Postgraduate Research Program at the National Center for Toxicological Research administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and the U.S. Food and Drug Administration.
Supplemental material for this article may be found at http://aem.asm.org/. ![]()
These two authors contributed equally to this work. ![]()
|
|
|---|
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»