Genetic Analysis of Dioxin Dioxygenase of Sphingomonas sp. Strain RW1: Catabolic Genes Dispersed on the Genome

Jean Armengaud; Birgitta Happe; Kenneth N Timmis

doi:10.1128/jb.180.15.3954-3966.1998

. 1998 Aug;180(15):3954–3966. doi: 10.1128/jb.180.15.3954-3966.1998

Genetic Analysis of Dioxin Dioxygenase of Sphingomonas sp. Strain RW1: Catabolic Genes Dispersed on the Genome

Jean Armengaud ^1,^*, Birgitta Happe ¹, Kenneth N Timmis ¹

PMCID: PMC107381 PMID: 9683494

Abstract

The dioxin dioxygenase of Sphingomonas sp. strain RW1 activates dibenzo-p-dioxin and dibenzofuran for further metabolism by introducing two atoms of oxygen at a pair of vicinal carbon atoms, one of which is involved in one of the bridges between the two aromatic rings, i.e., an angular dioxygenation. The dxnA1 and dxnA2 cistrons encoding this dioxygenase have been cloned and shown to be located just upstream of a hydrolase gene which specifies an enzyme involved in the subsequent step of the dibenzofuran biodegradative pathway. Genes encoding the electron supply system of the dioxygenase are not clustered with the dioxygenase gene but rather are located on two other distinct and separate genome segments. Moreover, whereas expression of dxnA1A2 is modulated according to the available carbon source, expression of the dbfB gene encoding the ring cleavage enzyme of the dibenzofuran pathway, which is located in the neighborhood of dxnA1A2 but oriented in the opposite direction, is constitutive. The scattering of genes for the component proteins of dioxin dioxygenase system around the genome of Sphingomonas sp. strain RW1, and the differential expression of dioxin pathway genes, is unusual and contrasts with the typical genetic organization of catabolic pathways where component cistrons tend to be clustered in multicistronic transcriptional units. The sequences of the α and β subunits of the dioxin dioxygenase exhibit only weak similarity to other three component dioxygenases, but some motifs such as the Fe(II) binding site and the [2Fe-2S] cluster ligands are conserved. Dioxin dioxygenase activity in Escherichia coli cells containing the cloned dxnA1A2 gene was achieved only through coexpression of the cognate electron supply system from RW1. Under these conditions, exclusively angular dioxygenation of dibenzofuran and dibenzo-p-dioxin was obtained. The dioxin dioxygenase was not active in E. coli cells coexpressing a class IIB electron supply system. In the course of the isolation of the dxnA1 and dxnA2 cistrons, a number of other catabolic genes dispersed over different genome segments were identified, which may indicate greater catabolic potential than was previously suspected. This finding is consistent with the catabolic versatility of members of the genus Sphingomonas, which is becoming increasingly evident, and may indicate a less well evolved and regulated but more dynamic genetic organization in this organism than is the case for better-studied pathways in organisms such as Pseudomonas species.

Polyhalogenated monocyclic and polycyclic aromatics are important and widely dispersed pollutants that are difficult to treat due to their occurrence in the environment at low absolute but toxicologically relevant concentrations. Most of these xenobiotics are chemically stable and thus biochemically recalcitrant and are of considerable concern due to their accumulation in the alimentary chain. This is particularly the case for polychlorinated dibenzo-p-dioxins and dibenzofurans, which are produced as unwanted by-products in a variety of manufacturing processes (production of herbicides, insecticides, and fungicides; paper pulp bleaching) as well during combustion of solid waste in incinerators or in accidental fires. Among the 75 and 135 different possible chlorinated congeners of dibenzo-p-dioxin and dibenzofuran, respectively, some molecules such as 2,3,7,8-tetrachloro-dibenzo-p-dioxin (the so-called Seveso dioxin) are extremely toxic (35), and their relatively high level in some milks has become a significant public concern.

Several microorganisms able to grow aerobically on dibenzo-p-dioxin or dibenzofuran as a sole carbon source, or to transform these chemicals, have been isolated in the recent years, and the corresponding degradative pathways have been elucidated (14, 20, 23, 27, 31–33, 36). Several enzymes of the converging dibenzo-p-dioxin and dibenzofuran pathways in Sphingomonas sp. strain RW1, a strain isolated from the Elbe River in northern Germany (36), have been biochemically and in some cases genetically characterized. The initial step of these pathways is the dihydroxylation of one of the aromatic rings by a three-component enzyme system, which leads to chemically unstable intermediates that spontaneously rearomatize to more stable compounds, i.e., 2,2′,3-trihydroxybiphenyl (2,2′,3-THB) and 2,2′,3-trihydroxydiphenyl-ether (2,2′,3-THD-ether) for dibenzofuran and dibenzo-p-dioxin pathways, respectively (Fig. 1).

FIG. 1 — The dioxin dioxygenase and its electron supply system. The reaction carried out by the multicomponent ring-hydroxylating dioxin dioxygenase and its electron transfer chain are shown. A flavoprotein reductase, RedA2, accepts electrons from NADH and transfers them via the ferredoxin Fdx1 to the terminal oxygenase. The reduced terminal oxygenase catalyzes the angular oxidation of dibenzo-p-dioxin and dibenzofuran. Chemical designations: (I), dibenzo-p-dioxin; (II), 4,4a-dihydroxy-dihydro-dibenzo-p-dioxin; (III), 2,2′,3-THD-ether; (IV), dibenzofuran; (V), 4,4a-dihydroxy-dihydro-dibenzofuran; (VI), 2,2′,3-THB. The two unstable compounds which spontaneously transform to other products are indicated in brackets.

The initial dioxygenases involved in the attack of aromatic compounds are key determinants of the substrate range of pathways. In concert with the other enzymes of upper pathways, they lead to the formation of catechol or salicylate or chlorinated derivatives thereof, some of which may then be processed further through lower pathways to Krebs cycle intermediates. Three-component dioxygenases like toluene dioxygenase (28, 37, 38), naphthalene dioxygenase (17), and biphenyl dioxygenases (3, 13, 16, 25, 34) have received considerable attention in recent years. Whereas most dioxygenases attack the aromatic substrates at neighboring carbon atoms not involved in bridges between rings (5), dibenzofuran-4,4a-dioxygenase of Sphingomonas sp. strain RW1 (7) and carbazole dioxygenase of Pseudomonas sp. strain CA10 (30) both attack at a bridge position (angular attack). In addition to this interesting feature, the electron supply system of the dibenzofuran-4,4a-dioxygenase/dioxin dioxygenase is atypical because it involves a putidaredoxin-type [2Fe-2S] ferredoxin and not a Rieske-type [2Fe-2S] ferredoxin (1). For this reason, it has been classified as a class IIA ring-hydroxylating oxygenase according to the classification proposed by Batie et al. (4), whereas all the other dioxygenases utilizing a ferredoxin so far genetically characterized belong to class IIB or class III.

Despite the interest in dioxin dioxygenase, and its initial biochemical characterization by Bünz and Cook (7), its genetic analysis has not been reported. We recently cloned the genes of the electron transport system of the dioxin dioxygenase of Sphingomonas sp. strain RW1, namely, the ferredoxin gene fdx1, which surprisingly was found to be clustered with genes apparently encoding two atypical decarboxylases and a glutathione-S transferase (1), and the reductase A2 gene redA2 (2), and characterized the corresponding proteins. We have now identified the open reading frames (ORFs) encoding the dioxin dioxygenase itself and obtained functional expression of this important enzyme system. In addition, we present information on a new type of degradative gene organization in Sphingomonas and discuss these results in terms of the catabolic potential of this genus.

MATERIALS AND METHODS

Materials.

Most chemicals used in this study were obtained from Sigma and Aldrich and were of the highest grade available. 2,3-Dihydroxybiphenyl (2,3-DHB) for enzymatic tests was obtained from Wako Chemicals GmbH, while 2,2′,3-THB and 2,2′,3-THD-ether used as authentic standard were kindly donated by R.-M. Wittich (GBF-Braunschweig). Oligonucleotide primers were synthesized on an Applied Biosystems model 381A DNA synthesizer, desalted, and used without further purification. [α-³²P]dCTP (3,000 Ci/mmol) and the Multiprime DNA labeling kit used for the labeling of the probes were purchased from Amersham. Qiabrane membranes from Qiagen were used for DNA blotting experiments, whereas Hybond-N⁺ gridded membranes from Amersham were used for colony lift hybridization. The peptide Ala-Lys-Arg-Asn-Ala-Val-Asp-Val-Ala-Asp-Leu-Phe-Asp-Arg-amide corresponding to the N-terminal sequence of the dioxin dioxygenase (7) was synthesized with an S-tert-butyl-protected cysteine attached to the N terminus for subsequent immobilization by standard 9-fluorenylmethoxycarbonyl/tert-butyl chemistry with O-benzo-triazolyl-N,N,N′,N′-tetramethyluronium terafluoroborate/N-methylmorpholine activation on Tentagel-SAC resin (Rapp Polymere), using a multiple synthesizer (Abimed Analysentechnik), followed by deprotection and cleavage from the resin with trifluoroacetic acid containing 3% triisobutylsilane and 2% water. The crude peptide was purified by preparative high-pressure liquid chromatography (HPLC) using a reverse-phase column eluted at a flow rate of 1 ml/min with a 30-min linear gradient of 0 to 100% acetonitrile in 0.1% aqueous trifluoroacetic acid. Peptide-containing fractions were identified by analytical reverse-phase HPLC and matrix-assisted laser desorption ionization mass spectrometry. Pure peptide fractions were concentrated and lyophilized. Aliquots were used, after coupling with bovine serum albumin by sulfo-MBS (Pierce), to immunize a rabbit whose serum was subsequently collected and tested for specific reactivity with the peptide. The polyclonal antibodies were then purified by affinity chromatography on an EAH-Sepharose 4B column (Pharmacia) to which the peptide had previously been coupled by sulfo-M-maleimidobenzoyl-sulfo-succinimide ester (Pierce) under conditions recommended by the supplier. Purified antibodies were used as previously described (29) in Western blot experiments to specifically detect the dioxin dioxygenase. The membranes were developed either by an alkaline phosphatase-based reaction or by an emission chemiluminescence-based reaction.

Escherichia coli DH5α [deoR endA1 gyrA96 hsdR17(r_K⁻ m_K⁺) recA1 relA1 supE44 thi-1 Δ(lacZYA-argF) φ80d lacZΔM15 F⁻ λ⁻] was obtained from Clontech; Sphingomonas sp. strain RW1 was kindly provided by R.-M. Wittich (36). Plasmids pCR-Script Amp SK(+), pBluescript II KS (+), and pGEM-T were obtained from Stratagene and Promega, whereas pBBR1MCS-2 was obtained from M. E. Kovach (24). Restriction enzymes and reagents for genetic procedures were purchased from New England Biolabs, Boehringer Mannheim, Promega, United States Biochemical, and Sigma.

General DNA procedures and sequence analysis.

Standard DNA manipulations, as well as transfer and DNA hybridizations, were carried out essentially as specified by Sambrook et al. (29), whereas plasmid extraction was achieved by means of a Qiawell-8 kit as recommended by the supplier (Qiagen). Nucleotide sequencing of both DNA strands was carried out by using a PRISM Ready Reaction DyeDeoxy Terminator Cycle Sequencing kit (Perkin-Elmer) with double-stranded templates in the presence of 5% dimethyl sulfoxide. Samples containing fluorescence-labeled dideoxynucleotide terminators were processed on a 373 Stretch Applied Biosystem automated sequencer. Sequence analysis was performed as described previously (1).

Generation of two specific probes and library screening.

A set of degenerate primers was designed from the N-terminal sequence of the α subunit of the dioxin dioxygenase (7) and the two hydrolases, H1 and H2 (8), as well as from consensus sequences of Rieske-type [2Fe-2S] ferredoxins, hydrolases, and class IIB dioxygenases, as indicated in Fig. 2. Sphingomonas sp. strain RW1 genomic DNA extracted by means of a Qiagen genomic DNA extraction kit from cells grown on dibenzofuran as the sole carbon source was used as template for PCR (Hoffman-La Roche) amplification. Conditions for PCR as well as analysis and cloning of the PCR products were as described by Armengaud and Timmis (1). Alternatively, PCR of long fragments was performed by using an Expand Long Template PCR system kit from Boehringer under conditions recommended by the supplier. PCR amplification of segments of the dioxin dioxygenase and hydrolase H1 genes was carried out with primers AJ118 (ATGGCIAARMGIAAYGCIGT) and AJ124 (CATYTCDATRTARTGIGT) and primers AJ121 (ACICAYTAYATHGARATG) and AJ127 (TGYTCDATYTGIAYCCARTG), respectively. These two fragments, 2,179 and 720 bp, respectively, in length, were cloned into plasmid pGEM-T (Promega) to produce plasmids pAJ112 and pAJ113. The two inserts, designated AR22.4 and AR31.2, were excised by digestion with NotI and PstI, purified (after analysis on a 1.5% agarose gel) with a QiaexII gel extraction kit (Qiagen), and radiolabeled by random oligonucleotide priming. They were then used as specific probes to screen a previously constructed Sphingomonas sp. strain RW1 pLAFR3-based cosmid library (18). Four positive cosmids, designated pAJ114 to pAJ117, were isolated and shown to contain a common 5.5-kb EcoRI fragment hybridizing strongly with the two probes. This fragment was subcloned from cosmid pAJ114 into pBluescript to produce plasmid pAJ118 and entirely sequenced on both strands, using specific 21-nucleotide primers designed on the basis of the known sequence as well as universal primers.

FIG. 2 — PCR-based strategy developed to clone the *dxnA1* and *dxnA2* genes. From different alignments of related proteins, C-terminal (Cter) and N-terminal (Nter) consensus sequences were defined as indicated by the central boxes. Degenerate forward and reverse primers, indicated by small arrows, were then designed from these consensus sequences and from the N-terminal sequence of each purified polypeptide determined by Edman degradation. The two PCR products obtained from the multiple combinations of primers (indicated with dashed lines) resulted in determination of the gene organization of the *dxn* locus.

Construction of a dioxin dioxygenase expression vector.

For expression purposes, a 2.1-kb fragment containing the dxnA1 and dxnA2 (collectively referred to as dxnA1A2) cistrons and a 174-nucleotide upstream region which includes the putative ribosome binding site of dxnA1 was introduced into the broad-host-range vector pBBR1MCS-2 (24). A 1,433-bp EagI fragment from plasmid pAJ118 was subcloned into pBluescript II KS (+) digested with EagI, to produce plasmid pAJ126. Plasmid pAJ118 was digested with KpnI and HindIII, and the resulting 1.4-kb fragment was purified. Plasmid pAJ126 was digested with KpnI, and the resulting 0.7-kb fragment was purified and ligated with the insert from pAJ118 and plasmid pBBR1MCS-2, previously digested with KpnI and HindIII and dephosphorylated, to produce plasmid pAJ127 carrying the dxnA1A2 genes under the control of the P_lac promoter.

Resting cell assays.

Dioxin dioxygenase activity in resting cells of E. coli DH5α(pAJ127)(pAJ130) grown in LB medium containing kanamycin (30 μg/ml) and ampicillin (100 μg/ml) was measured under conditions identical to those described by Beil et al. (5). Briefly, cells were washed two times with the assay buffer (10 mM glucose in 0.1× M9 mineral medium) and resuspended in 10 ml of prewarmed assay buffer containing 0.5 mM substrate to an absorbance at 600 nm of 2.0. Samples were taken at regular intervals, immediately shock frozen in liquid nitrogen, thawed, and centrifuged, and the resulting supernatant fluids and controls were analyzed by reverse-phase HPLC with a Shimadzu LC-10AD liquid chromatograph system equipped with a DGU-3A degasser and a 3PD-M10A photodiode array detector. Dibenzofuran, dibenzo-p-dioxin, and their metabolites were separated on an analytical SC 125- by 4.6-mm Lichrospher 100 RP8 5.0-μm column eluted with 0.1% ortho-phosphoric acid in water containing 54% methanol at a flow rate of 1 ml/min. The column effluent was monitored by measuring the absorption spectrum between 200 and 400 nm.

Identification of additional α-subunit-encoding cistrons.

A clone designated pHP133, able to convert 2,3-DHB to the bright yellow product 2-hydroxy-6-oxophenylhexa-2,4-dienoate (HOPDA), was selected from the Sphingomonas sp. strain RW1 pLAFR3-based cosmid library, using conditions previously described (18). Different overlapping subfragments of the cloned Sphingomonas DNA, including 2.4-kb Sau3A, 3.3-kb Sau3A, 6.4-kb Sau3A-PstI, 4.5-kb PstI, 2.4-kb XhoI, and 6.3-kb PstI fragments of cosmid pHP133 were inserted into pBluescript II KS (+) to produce pRW0, pRW1, pRW3, pRW11, pRW34, and pRW35, respectively. A 10-kb nucleotide sequence was determined by a walking strategy with multiple initiation points, using these subclones as templates. E. coli strains carrying cloned DNA fragments encoding indole oxidation activity were identified by spreading transformants onto LB plates supplemented with antibiotics and 1 mM indole. Positive clones such as E. coli(pRW0) were detected as blue colonies, indicative of conversion of indole to indigo (12).

The PCR amplification of a segment of a ring-hydroxylating α-subunit-encoding gene was obtained by using primers AJ025 (TAYATGGGBGARGAYCCVGT) and AJ026 (GCRAAYTTCCARTTRCABGG) designed from conserved motifs identified in class IIB dioxygenases. This 434-bp fragment was cloned into pGEM-T plasmid (Promega). The insert of the resulting plasmid, pAJ132, was excised by NotI and PstI digestion, purified, and radiolabeled. The cosmid library was then screened using this fragment as a specific probe. Several cosmids were isolated and shown to contain a common 12-kb HindIII fragment which hybridized with the probe. This fragment was subcloned from one of them (pAJ133) into pBluescript to produce plasmid pAJ139 and then partially sequenced.

Nucleotide sequence accession number.

The nucleotide sequences described in this publication have been deposited in the DDBJ/EMBL/GenBank databases under accession no. AJ223219 and AJ223220, and the nucleotide sequence X72850 has been updated.

RESULTS

The cistrons specifying the dioxin dioxygenase are linked to a hydrolase gene.

Several proteins involved in the degradation of dioxin or dibenzofuran by Sphingomonas sp. strain RW1 have been purified, and their N-terminal sequences have been determined (7, 8). The sequence of the N terminus of the α subunit of the dioxin dioxygenase [Ala-Lys-Arg-Asn-Ala-Val-Asp-Val-Ala-Asp-Leu-Phe-Asp-Arg-(Asp/Ser)-Thr-(Gly/Ser)-Val-Leu-Lys] does not exhibit any significant similarities with other known dioxygenases. Nevertheless, comparison of the known multicomponent α-subunit dioxygenases of the same size (45 ± 6 kDa) revealed that their sequences are quite conserved except in their N termini. We therefore attempted to produce a specific probe for the structural gene of the α subunit of the dioxin dioxygenase by means of a PCR involving a primer based on the N-terminal sequence and a reverse primer based on a consensus sequence from the C-terminal part of this type of dioxygenase (Fig. 2).

The cistrons specifying the different proteins of known three-component dioxygenase systems (class IIB and class III) are usually clustered. Assuming that the cistrons encoding the α and β subunits and those of the ferredoxin and reductase which are associated with the dioxygenase are also clustered, it is also possible to design a PCR strategy using primers from consensus sequences of the ferredoxin and reductase. We therefore devised degenerate reverse primers for Rieske-type [2Fe-2S] ferredoxins and class IIB reductases, as well as for putidaredoxin-type [2Fe-2S] ferredoxin and class I cytochrome P-450 reductases. The former electron supply system is associated with class IIB dioxygenases, while the latter is thought to be associated with class IIA dioxygenases (4). Genes encoding the different enzymes of the same degradative pathway are also usually clustered (15, 19). Bünz et al. (8) have determined the N-terminal sequences of two hydrolases possibly involved in the third step of the dibenzofuran degradative pathway in Sphingomonas sp. strain RW1. We therefore designed several sets of degenerate primers and reverse primers from the N-terminal sequences of these two hydrolases and from well-conserved motifs assumed to be located at their C termini.

Figure 2 presents the 20 different primers used in 39 combinations in both long-template and short-template PCRs to amplify specific sequences from total genomic DNA from Sphingomonas sp. strain RW1. Two pairs of primers, AJ118-AJ124 and AJ121-AJ127, yielded major products 2.2 and 0.7 kb, respectively, in length (Fig. 2), while either no product or unspecific multiple fragments were obtained with other primer combinations. However, the pair AJ025-AJ026 also gave a product 434 bp in length, which is further commented on below. The first fragment, which had been assumed to contain cistrons specifying the α and β subunits, the ferredoxin, and the reductase, was unexpectedly short, whereas the size of the second fragment was exactly that expected for the internal segment of the hydrolase gene. The fragments were purified and cloned into a T-cloning vector to produce plasmids pAJ107 and pAJ108, respectively, and both strands of the cloned fragments were sequenced. The translated nucleotide sequence of the 2.2-kb fragment corresponds to two polypeptides sharing similarities with α and β subunits of class IIB dioxygenases. In addition, the N-terminal sequence of the first encoded polypeptide is in perfect agreement with the experimentally determined N-terminal sequence of the α subunit of dioxin dioxygenase. The other extremity of the amplified sequence encodes the first 21 codons of hydrolase H1. The second amplified fragment specifies part of a protein showing marked similarities to known hydrolases, and its N-terminal amino acid sequence is in close agreement with that of the H1 protein purified by Bünz et al. (8). Thus, these two PCR fragments constituted specific probes for the dioxin dioxygenase and hydrolase genes. Moreover, as indicated in Fig. 2, a preliminary genetic organization of the dxn locus could be deduced from these results: the dioxin dioxygenase structural gene is linked to that of a hydrolase rather than to the genes of its electron supply system.

Cloning and sequence analysis of the dxn cluster.

The inserts of plasmids pAJ112 and pAJ113, namely, AR22.4 and AR31.2 were used as α-³²P-labeled probes to screen by hybridization colony lifts of a Sphingomonas sp. strain RW1 pLAFR3-based cosmid library. Four different cosmids were isolated and shown to contain a common 5.5-kb EcoRI fragment which hybridized with both probes. This 5.5-kb fragment was subcloned from one cosmid, pAJ114, into pBluescript II KS (+) to give plasmid pAJ118 and subsequently sequenced on both strands, using 21-mer synthetic oligonucleotide primers for the known sequences. The four possible ORFs present within the fragment (Fig. 3, locus A) all have the same orientation and may be cotranscribed since no G/C-rich inverted repeat capable of forming a stable stem-loop structure which might serve as a transcriptional terminator was detected within the fragment. These ORFs were therefore all designated dxn. Their derived amino acid sequences were compared with those of other proteins in the SwissProt, GenBank, and EMBL databases (Table 1).

FIG. 3 — Genetic organizations of loci encompassing the genes described in this study. The positions and orientations of the different ORFs detected within the loci described in the text are shown by full arrows. The scale bar at the bottom indicates gene sizes in kilobase pairs.

TABLE 1.

Genes and gene products of the dxnA1A2 locus

Gene	Position in accession no. X72850	Probable function	Calculated molecular mass (Da)	% Identity^b	Homology		Reference^c
Gene	Position in accession no. X72850	Probable function	Calculated molecular mass (Da)	% Identity^b	Name of protein	Source	Reference^c
dbfB	4399–3524	2,2′,3-THB dioxygenase	32,259	34	BphC-LB400	Burkholderia sp. strain LB400	BPHC_BURCE
				34	BphC-KF707	Pseudomonas pseudoalcaligenes KF707	BPHC_PSEPS
				30	BphC-KKS102	Pseudomonas strain KKS102	BPHC_PSES1
				27	TodE-F1	P. putida F1	TODE_PSEPU
dxnA1	8930–10237	Dioxin dioxygenase (α subunit)	48,293^a	40	TodC1-F1	P. putida F1	TOD1_PSEPU
				40	BedC1-ML2	P. putida ML2	BED1_PSEPU
				38	BnzA-K12	E. coli K-12	BNZA_ECOLI
				37	BphA-LB400	Burkholderia sp. strain LB400	BPHA_BURCE
				36	BnzA_BE81	P. putida BE-81	BNZA_PSEPU
				33	ORF G5-RW1	Sphingomonas sp. strain RW1	(this study)
dxnA2	10234–10773	Dioxin dioxygenase (β subunit)	20,912	39	BedC2-ML2	P. putida ML2	BED2_PSEPU
				39	BphE-LB400	Burkholderia sp. strain LB400	BPHE_BURCE
				37	BnzB-BE81	P. putida BE-81	BNZB_PSEPU
				37	TodC2-F1	P. putida F1	TOD2_PSEPU
				16	NdoC-C18	Pseudomonas sp. strain C18	NDOC_PSEPU
dxnB	11046–11882	Hydrolase	30,165^a	33	DmpD-CF600	P. putida CF600	DMPD_PSEPU
				32	TodF-F1	P. putida F1	TODF_PSEPU
				27	XylF-mt2	P. putida mt-2(pWW0)	XYLF_PSEPU
				19	BphD-LB400	Burkholderia sp. strain LB400	BPHD_BURCE
				14	BphD-KKS102	Pseudomonas strain KKS102	BPHD_PSES1
dxnC	12028–12372	Bacterial receptor	>11,307	27	ViuA	Vibrio cholerae	VIUA_VIBCH
				18	IroA-BNVC	Neisseria meningitidis BNVC	IROA_NEIME
				18	FyuA-6/69	Yersinia pestis 6/69	FYUA_YERPE
				17	FyuA-WAC	Yersinia enterolitica WA-C	FYUA_YEREN

Open in a new tab

The molecular mass of the polypeptide was determined after removal of the initial methionine (7, 8).

Percentage of amino acids that are identical when sequences are aligned with sequences listed in the GenBank database by using the algorithm of Needleman and Wunsch via the Blitz program of the European BioInformatic Institute facilities.

Citation in Netserv format.

(i) Cistrons dxnA1A2 encode dioxin dioxygenase.

ORFs dxnA1 and dxnA2, spanning nucleotides 8930 to 10237 and nucleotides 10234 to 10773, respectively, were identified as the cistrons for the α and β subunits of the dioxin dioxygenase because they exhibit significant similarities with genes encoding aromatic ring-activating multicomponent dioxygenases, and because the N terminus of the polypeptide specified by dxnA1 corresponds exactly to the partial sequence reported by Bünz and Cook (7) except for the initial methionine of the ORF, which is lacking in the mature enzyme. The overall identity of the DxnA1 sequence with its counterparts in class IIB and class III dioxygenases is relatively low (40%) but clearly shows a phylogenetic relationship with three-component dioxygenases. Sequence comparison of known large subunits of ring-hydroxylating dioxygenases resulted in the unrooted tree shown in Fig. 4, which presents our current understanding of their phylogenetic relationships. The consensus sequence Cys-X₁-His-X₁₆-Cys-X₂-His of a Rieske-type [2Fe-2S] cluster binding site can be identified in the sequence of DxnA1 (residues 84 to 107), as well as the four ligands of an Fe(II) prosthetic group (residues Glu₂₀₀, Asp₂₀₅, His₂₀₈, and His₂₁₃) which are well conserved among three-component dioxygenases (21) (Fig. 5). Comparison of sequences of the different α subunits and the Sphingomonas sp. strain RW1 DxnA1 reveals highly conserved stretches of amino acids around the first motif, highlighting the importance of this binding site and its immediate environment. However, detailed comparison of the environment of the Fe(II) ligands reveals some differences between DxnA1 and the class IIB dioxygenases (Fig. 5); for example, the last two histidines of the motif of DxnA1 are separated by four amino acids, whereas they are separated by five in other class IIB dioxygenases. Such an organization is also found in class I and class III dioxygenases. Thus, the geometry of the ligands of the active site where oxygen activation is thought to occur may differ slightly between the dioxin dioxygenase and the class IIB dioxygenases. Moreover, the residues in the close environment of the active site are not as conserved as in the [2Fe-2S] cluster binding region.

FIG. 4 — Phylogenetic tree obtained from alignment of the four α-subunit sequences with related proteins. The sequences of 16 α subunits of ring-hydroxylating oxygenases and the four sequences reported in this study were compiled by using the GeneWorks software (version 2.5N) from IntelliGenetics; the multiple alignment analysis was performed with the Phylip package programs. The phylogenetic unrooted tree was drawn by using TreeView. The horizontal bar indicates the percent divergence (distance). The numbers on some of the branches refer to the confidence (percent) estimated by bootstrap analysis (100 replications). The proteins are labeled by trivial abbreviations. Their accession codes in the SwissProt or GenBank databases and their origins are D86080 for aniline dioxygenase from *Acinetobacter* sp. strain YAA (AtdA-YAA), D90884 for dioxygenase from *E. coli* K-12 (BedC1-K12), L046642 for benzene dioxygenase from *Pseudomonas putida* ML2 (BedC1-ML2), M76990 for benzoate oxygenase from *Acinetobacter calcoaceticus* ADP1 (BenA-ADP1), X80041 for biphenyl dioxygenase from *Rhodococcus globerulus* P6 (BphA1-P6), P37333 for biphenyl dioxygenase from *Burkholderia* sp. strain LB400 (BphA-LB400), U47637 for biphenyl dioxygenase from *Comamonas testosteroni* B-356 (BphA-B356), M83949 for naphthalene 1,2-dioxygenase from *P. putida* G7 (NahAc-G7), U49496 for naphthalene dioxygenase from *Pseudomonas* sp. strain 9816 (NahAc-9816), U49504 for 2-nitrotoluene dioxygenase from *Pseudomonas* sp. strain JS42 (NtdAc-JS42), U62430 for ORF 2-like dioxygenase from *Burkholderia* sp. strain DNT (ORF 2-DNT), AB004059 for polycyclic aromatic hydrocarbon dioxygenase from *P. putida* OUS82 (PahAc-OUS82), U78099 for chlorobenzene dioxygenase from *Burkholderia* sp. strain PS12 (TecA1-PS12), U11420 for 2,4,5-trichlorophenoxyacetic acid oxygenase from *Pseudomonas cepacia* ACC1100 (TftA-ACC1100), J04996 for toluene dioxygenase from *P. putida* F1 (TodC1-F1), U51165 for dioxygenase from *Cycloclasticus oligotrophus* RB1 (XylC1-RB1), and P23099 for toluate 1,2-dioxygenase from *P. putida* ML2 (XylX-mt2).

FIG. 5 — Conserved sequences that characterize the Rieske-type [2Fe-2S] cluster and Fe(II) binding sites in α subunits. Well-conserved fingerprint sequence regions in the N-terminal part of the α subunits of ring-hydroxylating dioxygenases are aligned. Shaded characters represent conserved residues; arrows indicate the amino acids involved in binding the Rieske-type [2Fe-2S] cluster (A) and the mononuclear iron atom (B). The corresponding consensus sequences for the entire family of this type of protein are shown. Accession numbers in the GenBank database and origins of the large α subunits are P37333 for biphenyl dioxygenase from *Burkholderia* sp. strain LB400 (BphA-LB400), D17319 for biphenyl dioxygenase from *Pseudomonas* sp. strain KKS102 (BphA-KKS102), U78099 for chlorobenzene dioxygenase from *Burkholderia* sp. strain PS12 (TecA1-PS12), and J04996 for toluene dioxygenase from *Pseudomonas putida* F1 (TodC1-F1).

The dxnA2 ORF codes for a 21,086-Da polypeptide exhibiting 39% identity to the β subunits of three-component dioxygenases. Therefore, DxnA2 and DxnA1 exhibit the same phylogenetic distance from their class IIB counterparts.

(ii) Two other genes clustered with the dxnA1A2 locus.

At nt 11046, 270 nt downstream from dxnA2 (Fig. 3), begins an ORF, designated dxnB, which encodes a polypeptide 277 amino acids long (calculated molecular weight, 30,165) that corresponds to the hydrolase H1 previously purified from Sphingomonas sp. strain RW1 (8). The deduced polypeptide shows significant amino acid identity with several 2-hydroxymuconic semialdehyde hydrolases involved in the metabolism of chlorinated or nonchlorinated aromatics (Table 1). Consensus sequences containing the amino acids Ser₁₀₈, Asp₂₂₆, and His₂₅₅ thought to be involved in catalysis (6, 9) are present in the DxnB protein.

Further downstream of the dxnB gene (Fig. 3) is the 5′ end of an ORF, designated dxnC, only 111 codons of which were sequenced. The translation product of dxnC exhibits some similarities with certain bacterial receptors (Table 1) but not with previously described membrane proteins or transporters associated with bacterial aromatic degradative pathways.

No obvious sequence resembling a promoter and no other ORF were found upstream of dxnA1A2BC in the cloned 5.5-kb EcoRI fragment.

The dxnA1A2BC is located 4.5 kb downstream of the dbfB gene.

The spraying of colonies of E. coli HB101 containing the dxn locus-carrying cosmid pAJ114 with 2,3-DHB revealed the production of a ring cleavage enzyme. Appearance of the yellowish product of the reaction was rapidly followed by bleaching of the colonies, which suggests that they produced both a meta-cleavage enzyme and an active hydrolase which further transformed the meta-cleavage product. Subclones of cosmid pAJ114 encoding the ability to convert 2,3-DHB were sequenced, leading to the characterization of a 6-kb DNA segment upstream of the 5.5-kb EcoRI fragment encompassing the dxnA1A2 locus (Fig. 3). Only one ORF encoding a polypeptide exhibiting relatively high similarities to described proteins was detected and shown to encode a meta-cleavage dioxygenase. This gene is located 4.5 kb upstream of and oriented in the direction opposite that of the dxnA1A2 cistrons; it has been previously designated dbfB and characterized as encoding a 2,2′,3-THB ring-cleaving dioxygenase (18). The possibility of multiple copies of the dbfB gene was considered because the nucleotide sequence of dbfB determined in this study is slightly different from that reported previously. However, only one copy of this gene in Sphingomonas sp. strain RW1 was detected by hybridization carried out under stringent conditions to detect genes closely related to dbfB (data not shown). The previously published sequence contained two single-base errors, and the current sequence corrects an earlier 33-residue frameshift in the enzyme sequence. The corrected sequence of DbfB exhibits higher similarities to other meta-cleavage dioxygenases than did the previously published sequence.

Expression of dxnA1 is repressed in RW1 grown in rich medium.

As several genes potentially encoding subunits of class IIA or class IIB dioxygenases have been identified in RW1 (see below), we decided to analyze the expression of the dxnA1 gene in RW1 by means of polyclonal antibodies directed against a peptide corresponding to a region of the N-terminal sequence of the α subunit of the dioxin dioxygenase, which is the most variable region of these polypeptides. Peptide affinity-purified polyclonal antibodies gave a specific signal in Western blots of extracts of E. coli DH5α containing pAJ127 but no signal in negative controls (Fig. 6). They also reacted with a 48-kDa polypeptide in extracts of Sphingomonas sp. strain RW1 cells grown on dibenzofuran, salicylate, or acetate as the sole carbon source but gave only a faint signal with extracts of cells grown in LB medium (Fig. 6). The difference in amounts of the dioxin dioxygenase α subunit in cells grown with LB or dibenzofuran correlated with the ability of dibenzofuran-, salicylate-, or acetate-grown Sphingomonas sp. strain RW1 cells to metabolize dibenzofuran and the inability of LB-grown cells to convert this substrate.

FIG. 6 — Immunodetection of DxnA1 in *Sphingomonas* sp. strain RW1. Cells harvested from liquid cultures of *Sphingomonas* sp. strain RW1 and *E. coli*(pAJ127) grown under different conditions were lysed in sodium dodecyl sulfate-containing buffer, and their proteins were subjected to electrophoresis on 10% sodium dodecyl sulfate–glycine polyacrylamide gels and immunostained after transfer to nitrocellulose membranes. One membrane (A) was stained by an alkaline phosphatase-based reaction, whereas the other (B) was stained by a chemiluminescence-based reaction. (A) Samples of *Sphingomonas* sp. strain RW1 grown in M9 medium with 2 mM dibenzofuran as the sole carbon source (lane 1), LB medium (lane 2), M9 medium with 2 mM salicylate (lane 3), and M9 medium with 2 mM acetate (lane 4); (B) samples of *E. coli* DH5α(pAJ127) grown in LB medium (lane 5) and *Sphingomonas* sp. strain RW1 grown in M9 medium with 2 mM dibenzofuran as the sole carbon source (lane 1). The arrowhead shows the position of the 48-kDa large subunit of the dioxin dioxygenase.

Recombinant dioxin dioxygenase produced in E. coli requires coexpression of a putidaredoxin-type [2Fe-2S] ferredoxin gene for full activity.

Despite the fact that resting cells of E. coli DH5α(pAJ114) contain a strong hydrolase (presumably DxnB) activity and hence are assumed to actively express the dxnA1A2BC cluster and contain significant amounts of dioxin dioxygenase, no enzymatic activity was detected with dibenzofuran or dibenzo-p-dioxin as the substrate, as determined by HPLC analysis. It seemed probable, therefore, that the enzyme system is inactive due to the lack of a specific electron supply system encoded elsewhere on the RW1 chromosome (1). If this were the case, expression of active dioxin dioxygenase in E. coli would require coexpression of the α and β subunits with a suitable electron donor system. Converging arguments have been presented concerning the nature of the two proteins involved in this electron chain (7). Dioxin dioxygenase has been shown to function efficiently with Fdx1, a putidaredoxin-type [2Fe-2S] ferredoxin (1), and RedA2, a class I cytochrome P-450-type reductase (2).

We designed a two-plasmid expression system with one plasmid (pAJ127) containing the cistrons of the α and β subunits and the other (pAJ130) containing the cistrons of the Fdx1 ferredoxin and the RedA2 reductase from Sphingomonas sp. strain RW1. These two plasmids are derivatives of pBBR1-MCS2 and pVLT35, respectively, and compatible with one another. To assess the specificity of dependency of the dioxin dioxygenase on the electron supply system, we measured the enzymatic activity in bacteria of E. coli DH5α(pAJ127)(pAJ130) expressing the structural genes of the dioxin dioxygenase with the class IIA-type electron transfer system and of E. coli DH5α(pAJ127)(pFY31) expressing the class IIB-type electron chain. E. coli DH5α(pAJ127)(pAJ130) bacteria exhibited substantial dioxygenase activity (Fig. 7) toward dibenzofuran (18 ± 4 nmol of 2,2′,3-THB/min/mg of protein) and dibenzo-p-dioxin (17 ± 4 nmol of 2,2′,3-THD-ether/min/mg protein), whereas E. coli DH5α(pAJ127)(pFY31) and the control strains E. coli DH5α, E. coli DH5α(pAJ127), E. coli DH5α(pAJ130), and E. coli DH5α(pFY31) exhibited no detectable activity toward these substrates. It should be emphasized that the class IIB electron supply system encoded by pFY31 has been shown to be correctly produced and gives full activity when coexpressed with a class IIB dioxygenase by means of an analogous two-plasmid system (20a).

Only one product was formed by the dioxygenase from each substrate tested (Fig. 7); 2,2′,3-THB from dibenzofuran, with UV-visible maxima at 206, 244, and 283 nm; and 2,2′,3-THD-ether from dibenzo-p-dioxin, with maxima at 202 and 275 nm (Fig. 7). These compounds are produced via an angular attack at carbons involved in the bridges of the multiring aromatic molecules and not via a lateral attack which typically will give 2,3-dihydroxy-dihydro-dibenzo-p-dioxin or 1,2-dihydroxy-dihydro-dibenzo-p-dioxin from dibenzo-p-dioxin (5). As no other products were detected, there seems to be a high regiospecificity of this enzyme, resulting in an angular attack of these substrates. The amount of product detected by HPLC in the assay increased linearly over a period of 3 h (data not shown), indicating that the enzyme system is not inhibited by the reaction products and that these products are not toxic for the cells during the period of the assay. No problem of instability of the two-plasmid expression system was noted during this study.

Identification of three additional α-subunit-encoding cistrons in Sphingomonas sp. strain RW1. (i) Cistrons of two α subunits are clustered with an extradiol dioxygenase gene.

A clone exhibiting a meta-cleavage activity was identified in the Sphingomonas sp. strain RW1 pLAFR3-based cosmid library by spraying with 2,3-DHB. The corresponding cosmid, pHP133, was shown not to encompass the dbfB gene previously identified as specifying the 2,2′,3-THB dioxygenase (18). One subclone derived from pHP133, pRW3, containing a 6.4-kb PstI fragment was shown to convert 2,3-DHB, whereas another subclone, pRW0, containing a 2.4-kb Sau3A fragment was able to convert indole to indigo. Indole oxidation together with the formation of indigo by ring-hydroxylating dioxygenases, such as naphthalene and isopropylbenzene dioxygenases (11, 12), and monooxygenases, such as xylene monooxygenase (26), expressed in E. coli has proven to be a useful indicator of the existence of the genes of such enzymes on cloned fragments. We therefore determined the nucleotide sequence of part of cosmid pHP133 to investigate further this ring-hydroxylating enzyme. The nucleotide sequence of a 10-kb fragment, designated locus E (Fig. 3), from pHP133 revealed the presence of several degradative genes, including that of a monooxygenase specified by ORF G4 (Table 2). Indole oxidative activity was measured from a subclone derived from pRW0 and containing only ORF G4. The deduced polypeptide sequence of ORF G4 exhibits significant amino acid identity with several flavin-containing monooxygenases from diverse sources. Upstream of this gene, ORF G2 (EDO 2) specifies a meta-cleavage dioxygenase, which is able to cleave 2,3-DHB and which exhibits similarities with several extradiol dioxygenases (Table 2). Between the genes encoding this meta-cleavage dioxygenase and the monooxygenase lies a short ORF encoding a 106-amino-acid polypeptide with similarities to Rieske-type [2Fe-2S] ferredoxins associated with class IIB and class III ring-hydroxylating dioxygenases and containing the two cysteine and two histidine residues organized in the typical motif Cys-X₁-His-X₁₇-Cys-X₂-His containing the ligands of the [2Fe-2S] cluster. Further upstream of the EDO 2 gene is located ORF G1, which encodes a 45,295-Da polypeptide (Fig. 3) and exhibits similarities with the large subunit of ring-hydroxylating dioxygenases (Table 2). Most of the conserved amino acids are located in the N-terminal half of the protein containing the important residues that act as ligands of either a Rieske-type [2Fe-2S] cluster or the mononuclear ferrous iron atom, whereas few conserved residues were found in the C-terminal part of the protein. Downstream of ORF G4 encoding the monooxygenase lie four contiguous ORFs specifying polypeptides exhibiting marked similarities with the large and small subunits of ring-hydroxylating dioxygenases, HOPDA hydrolases, and indole-acetamide hydrolases, respectively (Table 2). The cistron order ORF G5 ORF G6 ORF G7 is similar to that of the dxnA1A2B cistrons which encode related polypeptides, as described above.

TABLE 2.

Genes and gene products of the EDO2 locus

Gene	Position in accession no. AJ223220	Probable function	Calculated molecular mass (Da)	% Identity^a	Homology		Reference^b
Gene	Position in accession no. AJ223220	Probable function	Calculated molecular mass (Da)	% Identity^a	Protein	Source	Reference^b
ORF G1	245–1447	Ring-hydroxylating dioxygenase (α subunit)	45,295	22	ORF G5-RW1	Sphingomonas sp. strain RW1	This study
				20	DxnA1-RW1	Sphingomonas sp. strain RW1	This study
				19	XylX-mt2	P. putida mt-2	XYLX_PSEPU
				18	BenA-BD143	A. calcoaceticus BD143	BENA_ACICA
				13	NdoB-C18	Pseudomonas sp. strain C18	NDOB_PSEPU
EDO 2	1966–2883	Extradiol dioxygenase	34,409	40	BphC-LB400	Burkholderia sp. strain LB400	BPHC_BURCE
				39	BphC-KF707	P. pseudoalcaligenes KF707	BPHC_PSEPS
				38	BphC-KKS102	Pseudomonas strain KKS102	BPHC_PSES1
				35	TodE-F1	P. putida F1	TODE_PSEPU
ORF G3	2886–3206	Ferredoxin	11,765	36	BphF-LB400	Burkholderia sp. strain LB400	BPHF_BURCE
				34	BedB-ML2	P. putida ML2	BEDB_PSEPU
				33	TodB-F1	P. putida F1	TODB_PSEPU
				32	BnzC-BE81	P. putida BE-81	BNZC_PSEPU
ORF G4	3405–4781	Monooxygenase	51,627	22	Yz20-H37RV	M. tuberculosis H37RV	YZ20_MYCTU
				20	CymO-NCIB9871	Acinetobacter strain NCIB 9871	CYMO_ACISP
				13	FmO2	Cavia porcellus	FMO2_CAVPO
ORF G5	5773–7131	Ring-hydroxylating dioxygenase (α subunit)	50,748	41	TodC1-F1	P. putida F1	TOD1_PSEPU
				41	BedC1-ML2	P. putida ML2	BED1_PSEPU
				41	BnzA-K12	E. coli K-12	BNZA_ECOLI
				39	BphA-LB400	Burkholderia sp. strain LB400	BPHA_BURCE
				33	DxnA1-RW1	Sphingomonas sp. strain RW1	This study
ORF G6	7133–7663	Ring-hydroxylating dioxygenase (β subunit)	21,086	42	BphE-LB400	Burkholderia sp. strain LB400	BPHE_BURCE
				41	TodC2-F1	P. putida F1	TOD2_PSEPU
				41	BnzB-BE81	P. putida BE-81	BNZB_PSEPU
				39	BedC2-ML2	P. putida ML2	BED2_PSEPU
				35	DxnA2-RW1	Sphingomonas sp. strain RW1	This study
ORF G7	7673–8518	Hydrolase	31,114	27	DmpD-CF600	P. putida CF600	DMPD_PSEPU
				26	TodF-F1	P. putida F1	TODF_PSEPU
				25	BphD-KKS102	Pseudomonas KKS102	BPHD_PSES1
				21	BphD-LB400	Burkholderia sp. strain LB400	BPHD_BURCE
				21	DxnB-RW1	Sphingomonas sp. strain RW1	This study
ORF G8	8524–9882	Indole-acetamide hydrolase	46,572	22	HyiN-Y30	P. syringae Y30	HYIN_PSESY
				21	HyiN-S4	Agrobacterium vitis S4	HYIN_AGRVI

Open in a new tab

Citation in Netserv format.

(ii) A fourth α-subunit ring-hydroxylating dioxygenase gene is present in the genome of Sphingomonas sp. strain RW1.

By means of the PCR strategy described above, a 434-bp fragment encoding part of a ring-hydroxylating α-subunit was obtained by using two primers designed from conserved motifs in class IIB dioxygenases. As the corresponding amino acid sequence differed from those of DxnA1 and the two other described α subunits, the cosmid library was screened with this PCR fragment as a probe. A 12-kb HindIII fragment was cloned from one of the positive cosmids thereby identified and partially sequenced. This fragment, designated locus D in Fig. 3, contains ORF H1 encoding a putative 47,541-Da polypeptide showing marked identities to α-subunit ring-hydroxylating dioxygenases.

Comparison of the four putative α subunits of ring-hydroxylating dioxygenases identified in RW1.

A sequence comparison of the four α subunits identified in this study and all known large subunits of ring-hydroxylating dioxygenases is presented in Fig. 4 in the form of an unrooted tree. The four polypeptide sequences, while conserving the main consensus traits of this family of proteins, do not cluster with other groups of dioxygenases, or with one another, and seem only distantly related to known dioxygenases.

A detailed comparison of the environment of the Fe(II) ligands and of the [2Fe-2S] cluster binding site of the four α subunits is presented in Fig. 5. The Cys-X₁-His-X₁₇-Cys-X₂-His motif is conserved in all four putative α subunits, indicating that they could in principle accommodate a Rieske-type [2Fe-2S] cluster. However, only the polypeptide encoded by ORF G5 and dxnA1 contains the four correctly spaced residues Glu-X₄-Asp-X₂-His-X_4/5-His of Fe(II) ligands as defined by Jiang et al. (21). The polypeptide specified by ORF H1 does contain the four key residues but with a different spacing between the two histidines. The polypeptide encoded by ORF G1, despite a good conservation in this part of the polypeptide and the presence of correctly spaced last three amino acids of the motif, lacks the initial residue. Therefore, whether ORF H1 and ORF G1 specify nonfunctional α subunits, or α subunits accommodating an Fe(II) with unusual ligands, is unclear.

DISCUSSION

Chlorinated dibenzofuran and dibenzo-p-dioxin are important environmental pollutants. The recent isolation of bacteria able to degrade unchlorinated dibenzofuran and dibenzo-p-dioxin and to transform some of their lower chlorinated congeners, and the purification and characterization of the initial dioxygenase of one such pathway, that of Sphingomonas sp. strain RW1, constituted major steps toward characterization of microbial interactions with these compounds.

The considerable efforts invested to genetically analyze the catabolic pathway of Sphingomonas sp. strain RW1 have, however, until now met with only modest success, restricted to the 2,2′,3-THB ring cleavage dioxygenase (18). In this report, we describe the use of degenerate primers based on the determined polypeptide sequences of the dioxin dioxygenase and hydrolase, and on consensus regions of related enzymes, to generate specific probes for the identification of the dioxin dioxygenase cistrons in a gene bank of Sphingomonas sp. strain RW1. The dxnA1 and dxnA2 cistrons, encoding the large and small subunits of the dioxin dioxygenase, respectively, were thereby identified and sequenced. Hyperexpression of dxnA1A2 in E. coli DH5α did not result in active dioxin dioxygenase, whereas coexpression of these cistrons with the cognate electron supply system consisting of the Fdx1 ferredoxin and the RedA2 reductase from Sphingomonas sp. strain RW1 did. Resting cells producing these four polypeptides converted dibenzo-p-dioxin and dibenzofuran to 2,2′,3-THD-ether and 2,2′,3-THB, respectively. Only one product was detected for each substrate tested, showing that the angular attack carried out by the dioxin dioxygenase is highly specific. This is not the case for another dioxygenase, tetrachlorobenzene dioxygenase from Burkholderia sp. strain PS12, which has a broad substrate spectrum that includes dibenzo-p-dioxin and dibenzofuran (5). In this case, attack is mainly lateral, yielding 1,2-dihydroxy-dihydro-dibenzo-p-dioxin and 1,2-dihydroxy-dihydro-dibenzofuran, respectively, as major products, although angular attack also occurred to a minor extent (15%).

As can be seen in Fig. 4, which shows the phylogenetic relationships of the major subunits of aromatic dioxygenases, the sequence of the DxnA1 polypeptide shows some divergence from those of ring-hydroxylating dioxygenases already described such that it falls outside the major clusters. However, the overall structures of the dioxin dioxygenase subunits, as well as the major functional elements, such as the environment of the Rieske-type [2Fe-2S] cluster and that of the mononuclear Fe(II) atom, are probably conserved. Interestingly, two residues in the Fe(II) binding site are highly conserved except in the large subunit of the dioxin dioxygenase. Here, a histidine residue at position 202 is found in place of the usual phenylalanine, and there is a leucine residue at position 207 instead of the usual tyrosine (Fig. 5). Given the possible importance of these two residues near the catalytic site of ring-hydroxylating dioxygenases, it will be interesting to assess their potential role in the peculiar regiospecificity of the dioxin dioxygenase, namely, to ring hydroxylate substrates via an angular attack.

We first reported the identification of the ferredoxin gene fdx1 on a 4.6-kb DNA genome segment presented in the Fig. 3 as locus B (1). We also identified the reductase gene redA2 at another unlinked locus, designated locus C (2). We now clearly establish in this work that the genes dxnA1 and dxnA2 are not clustered with the genes specifying the cognate electron donor system but rather are on a different genome segment designated locus A (Fig. 3). The A, B, and C loci are physically distinct, as they are located on cosmids containing around 40-kb fragments of total DNA from Sphingomonas sp. strain RW1 which do not cross-hybridize. As shown in Fig. 8, the genetic organization of hitherto investigated ring-hydroxylating dioxygenases involves cistrons encoding the α and β subunits which are contiguous with the genes of the specific electron carrier, or at least are clustered within the same transcriptional unit, as is the case for the carbazole dioxygenase of Pseudomonas sp. strain CA10 (30) and the p-cumate dioxygenase from Pseudomonas putida F1 (10). Moreover, the gene of the reductase associated with the electron carrier is also generally, though not always, present in such dioxygenase gene clusters (Fig. 8). The genetic organization of dioxin dioxygenase system of RW1 involves unlinked loci for the cistrons of the dioxygenase, the electron transfer protein, and the reductase. This, combined with the instability of the dioxin/dibenzofuran degradation phenotype in RW1 and the difficulties of carrying out genetic studies with this organism, accounts for previous failures to carry out a genetic analysis of the dioxin dioxygenase and to clone its genetic elements by constructing expression libraries and screening for ring-hydroxylating activity. As this enzyme is the first genetically characterized ring-hydroxylating dioxygenase belonging to class IIA, it remains to be established whether such a genetic organization is common for this subclass of enzymes.

FIG. 8 — Comparative molecular organization of three-component dioxygenases. The organization of each of the genes encoding the different polypeptides of representative hydroxylating enzymes is shown. The general name of each cluster, the name of the microorganism containing the enzymes, and the class (according to Batie et al., [4]) to which the enzymes belong are indicated at the left. Gene names are given above the ORFs, and arrows indicate the direction of transcription. Each cluster was aligned with the others, taking the gene specifying the α subunit dioxygenase as a reference. Genes are drawn according to size and relative position. Accession numbers in the DDBJ/EMBL/GenBank database of the corresponding sequences are X79076 for *cbdA* (dibenzothiophene dioxygenase) from *Pseudomonas* sp., M64747 for *xylX* (toluate dioxygenase) from *Pseudomonas putida*, J04996 for *todC1* (toluene dioxygenase) from *P. putida* F1, L04642 for *bedC1* (benzene dioxygenase) from *P. putida* ML2, M17904 for *bnzA* (benzene dioxygenase) from *P. putida*, U24277 for *ipbA1* (isopropylbenzene dioxygenase) from *Rhodococcus erythropolis* BD2, X80041 for *bphA1* (biphenyl dioxygenase) from *Rhodococcus globerulus* P6, D17319 for *bphA1* (biphenyl dioxygenase) from *Pseudomonas* sp. strain KKS102, U47637 for *bphA* (biphenyl dioxygenase) from *Comamonas testosteroni* B-356, M86348 for *bphA* (biphenyl dioxygenase) from *Burkholderia* sp. strain LB400, D89064 for *carAa* (carbazole dioxygenase) from *Pseudomonas* sp. strain CA10, U24215 for *cmtAa* (p-cumate dioxygenase) from *P. putida* F1, M60405 for *doxB* (dibenzothiophene dioxygenase) from *Pseudomonas* sp., D84146 for *pahA3* (naphthalene dioxygenase) from *Pseudomonas aeruginosa* PaK1, M83949 for *nahAc* (naphthalene dioxygenase) from *P. putida* NCIB 9816-4, and U49504 for *ntdAc* (2-nitrotoluene dioxygenase) from *Pseudomonas* sp. strain JS42.

Most catabolic pathways in bacteria that have been studied so far have a genetic organization characterized by clustering of the genes of entire pathways, or independently functioning pathway segments, in single transcriptional units transcribed from highly regulated promoters (15), typically located on transposable elements which facilitate the horizontal transfer of catabolic functions among bacteria and the evolution of new catabolic phenotypes (19). We have now identified all of the genetic elements of the upper pathway for dibenzo-p-dioxin and dibenzofuran degradation by Sphingomonas sp. strain RW1. The gene of the H1 hydrolase, which carries out the third step in the degradation of dibenzofuran, is directly linked to the dxnA1A2 cluster, whereas the gene dbfB, encoding the second enzyme of the pathway, which mediates cleavage of the first aromatic ring, is located 4.5 kb upstream of dxnA1 but oriented in the opposite direction. Therefore, it is probable that the dxnA1 and dxnA2B genes together form one transcriptional unit, while the dbfB, fdx1 and redA2 genes form separate transcriptional units. These transcriptional units seem not to be coregulated: the use of polyclonal antibodies directed toward a peptide derived from the N-terminal sequence of the α subunit of the dioxin dioxygenase revealed that expression of the dxnA1 gene varies according to carbon source and/or growth conditions, whereas the dbfB gene has been shown to be expressed constitutively (17a). The organization of the dibenzo-p-dioxin and dibenzofuran degradation upper pathways in Sphingomonas sp. strain RW1 is thus clearly different from that of other catabolic pathways thus far characterized in bacteria. Whether this atypical organization has a physiological or evolutionary significance remains unclear. One of the other α-subunit genes, ORF G5, is also linked to the genes encoding a β subunit and a hydrolase but not to any gene encoding a component of an electron supply system, thus exhibiting precisely the same organization as found for dxnA1A2B (Fig. 8).

It appears, therefore, that the dioxin dioxygenase gene organization is not restricted to the dioxin catabolic pathway and may be a general feature of catabolic gene organization in Sphingomonas. Moreover, our study has revealed an unexpected variety of ORFs whose products would resemble catabolic enzymes such as α subunits (4) and β subunits (2) of ring-hydroxylating dioxygenases, monooxygenases (1), extradiol dioxygenases (2), ferredoxins (2), HOPDA hydrolases (2), an indole-acetamide hydrolase, and a putative transport protein. This wealth of catabolic genes in Sphingomonas is consistent with our growing awareness of the catabolic potential of organisms of this genus and hints that RW1 may have much greater catabolic versatility—either in terms of its current potential or in terms of its evolutionary potential—than currently appreciated. The finding that these genes are scattered over unlinked DNA segments and appear to lack coordinated regulation suggests that RW1 may contain numerous primitive genetic elements of different evolving catabolic functions and that the collection and loss of such determinants may be a rather dynamic process. The fact that the catabolic genes of RW1 are not clustered renders their characterization and identification by functional cloning difficult. Another difficulty arises from the relatively high sequence divergence of related polypeptides, which renders difficult or impossible the screening by hybridization with consensus sequence probes. Several communications have reported the existence of cistrons potentially encoding α and β subunits of ring-hydroxylating dioxygenases-related proteins (22, 39), but the functional expression of these cistrons and therefore confirmation of their function has remained elusive. Whether the new dioxygenase specified by ORF G5 and ORF G6 is functional and whether it functions with the electron supply system associated with the dioxin dioxygenase or with another type of electron supply system remain to be investigated. Another α subunit which is not directly linked to any other genes appears to be encoded by ORF G1, although genes for a meta-cleavage dioxygenase and a ferredoxin are found further downstream. Recently, Sato et al. (30) reported the genetic characterization of carbazole dioxygenase, a class III-related ring-hydroxylating dioxygenase, which has been found to be fully active even if comprising only a large subunit. In this case, the ferredoxin and reductase components of the electron supply system are encoded by two contiguous genes present 3 kb downstream of the dioxygenase genes (Fig. 8). Perhaps ORF G1 also specifies a functional dioxygenase with only one large subunit or perhaps it recruits the polypeptide encoded by dxnB or ORF G6 as the small subunit. Whether ORF G3 encodes the electron carrier component of the electron supply system associated with ORF G1 remains to be studied. The fact that ORF G1 and ORF G5 encoding two putative large subunits are clustered in the same locus may indicate that these two enzymatic systems function in the same degradative pathway. A fourth cistron encoding an α-subunit-like protein is located on a separate locus (locus D in Fig. 3), and the same questions arise for this fourth element.

This study highlights the general power of the genetic strategy adopted here to explore the potential of catabolic determinants that bacteria contain and has finally opened up to further genetic, biochemical, and physiological investigations the dioxin/dibenzofuran pathway of Sphingomonas sp. strain RW1. This will in turn facilitate efforts to develop biodegradative strategies for this class of serious environmental pollutants. As all of the genetic elements of the upper pathway for dibenzo-p-dioxin and dibenzofuran conversion to catechol and salicylate, respectively, have now been identified, the de novo construction of an expression cassette for this upper pathway, and its introduction into microorganisms able to handle the chlorinated derivatives of catechol and salicylate not degraded by Sphingomonas sp. strain RW1, can be envisaged.

ACKNOWLEDGMENTS

We gratefully acknowledge M. E. Kovach (Louisiana State University Medical Center) for his generous gift of plasmid pBBR1MCS-2, H. Overwin (Molecular Recognition Group, GBF) for expert technical assistance in peptide synthesis, and the following from GBF: S. Backhaus for expert assistance with sequencing gel runs, E. Moore for use the sequencing facilities available in his group, B. Hofer for providing plasmid pFY31, M. d’Enza for help in final checking of the dbfB gene sequence, B. Averhof, H. Poth, and M. Strätz for their earlier work on the genetics of Sphingomonas sp. strain RW1, S. Beil and D. B. McKay for helpful discussions, and R.-M. Wittich and D. Pieper for their continuous interest in our work.

J. Armengaud was initially supported by a long-term grant from the FEBS and thanks the persons acting generously in this federation. This research was funded in part by the German Ministry for Education, Research and Technology (BMBF grant 0318896C). K. N. Timmis gratefully acknowledges the generous support of the Fonds der Chemischen Industrie.

REFERENCES

1.Armengaud J, Timmis K N. Molecular characterization of Fdx1, a putidaredoxin-type [2Fe-2S] ferredoxin able to transfer electrons to the dioxin dioxygenase of Sphingomonas sp. RW1. Eur J Biochem. 1997;247:833–842. doi: 10.1111/j.1432-1033.1997.00833.x. [DOI] [PubMed] [Google Scholar]
2.Armengaud J, Timmis K N. The reductase RedA2 of the multicomponent dioxin dioxygenase system of Sphingomonas sp. RW1 is related to class-I cytochrome P450-type reductases. Eur J Biochem. 1998;253:437–444. doi: 10.1046/j.1432-1327.1998.2530437.x. [DOI] [PubMed] [Google Scholar]
3.Asturias J A, Diaz E, Timmis K N. The evolutionary relationship of biphenyl dioxygenase from gram-positive Rhodococcus globerulus P6 to multicomponent dioxygenases from gram-negative bacteria. Gene. 1995;156:11–18. doi: 10.1016/0378-1119(94)00530-6. [DOI] [PubMed] [Google Scholar]
4.Batie C J, Ballou D P, Correll C C. Phthalate dioxygenase reductase and related flavin-iron-sulfur containing electron transferases. In: Müller F, editor. Chemistry and biochemistry of flavoenzymes. Boca Raton, Fla: CRC Press; 1992. pp. 543–566. [Google Scholar]
5.Beil S, Happe B, Timmis K N, Pieper D H. Genetic and biochemical characterization of the broad spectrum chlorobenzene dioxygenase from Burkholderia sp. strain PS12: dechlorination of 1,2,4,5-tetrachlorobenzene. Eur J Biochem. 1997;247:190–199. doi: 10.1111/j.1432-1033.1997.00190.x. [DOI] [PubMed] [Google Scholar]
6.Brady L, Brzozowski A M, Derewenda Z S, Dodson E, Dodson G, Tolley S, Turkenburg J P, Christiansen L, Huge-Jensen B, Norskov L, et al. A serine protease triad forms the catalytic centre of a triacylglycerol lipase. Nature. 1990;343:767–770. doi: 10.1038/343767a0. [DOI] [PubMed] [Google Scholar]
7.Bünz P V, Cook A M. Dibenzofuran 4,4a-dioxygenase from Sphingomonas sp. strain RW1: angular dioxygenation by a three-component enzyme system. J Bacteriol. 1993;175:6467–6475. doi: 10.1128/jb.175.20.6467-6475.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Bünz P V, Falchetto R, Cook A M. Purification of two isofunctional hydrolases (EC 3.7.1.8) in the degradative pathway for dibenzofuran in Sphingomonas sp. strain RW1. Biodegradation. 1993;4:171–178. doi: 10.1007/BF00695119. [DOI] [PubMed] [Google Scholar]
9.Diaz E, Timmis K N. Identification of functional residues in a 2-hydroxymuconic semialdehyde hydrolase. A new member of the alpha/beta hydrolase-fold family of enzymes which cleaves carbon-carbon bonds. J Biol Chem. 1995;270:6403–6411. doi: 10.1074/jbc.270.11.6403. [DOI] [PubMed] [Google Scholar]
10.Eaton R W. p-Cumate catabolic pathway in Pseudomonas putida F1: cloning and characterization of DNA carrying the cmt operon. J Bacteriol. 1996;178:1351–1362. doi: 10.1128/jb.178.5.1351-1362.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Eaton R W, Timmis K N. Characterization of a plasmid-specified pathway for catabolism of isopropylbenzene in Pseudomonas putida RE204. J Bacteriol. 1986;168:123–131. doi: 10.1128/jb.168.1.123-131.1986. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Ensley B D, Ratzkin B J, Osslund T D, Simon M J, Wackett L P, Gibson D T. Expression of naphthalene oxidation genes in Escherichia coli results in the biosynthesis of indigo. Science. 1983;222:167–169. doi: 10.1126/science.6353574. [DOI] [PubMed] [Google Scholar]
13.Erickson B D, Mondello F J. Nucleotide sequencing and transcriptional mapping of the genes encoding biphenyl dioxygenase, a multicomponent polychlorinated-biphenyl-degrading enzyme in Pseudomonas strain LB400. J Bacteriol. 1992;174:2903–2912. doi: 10.1128/jb.174.9.2903-2912.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Fortnagel P, Harms H, Wittich R-M, Krohn S, Meyer H, Sinnwell V, Wilkes H, Francke W. Metabolism of dibenzofuran by Pseudomonas sp. strain HH69 and the mixed culture HH27. Appl Environ Microbiol. 1990;56:1148–1156. doi: 10.1128/aem.56.4.1148-1156.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Franklin F C, Bagdasarian M, Bagdasarian M M, Timmis K N. Molecular and functional analysis of the TOL plasmid pWWO from Pseudomonas putida and cloning of genes for the entire regulated aromatic ring meta cleavage pathway. Proc Natl Acad Sci USA. 1981;78:7458–7462. doi: 10.1073/pnas.78.12.7458. [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Furukawa K, Miyazaki T. Cloning of a gene cluster encoding biphenyl and chlorobiphenyl degradation in Pseudomonas pseudoalcaligenes. J Bacteriol. 1986;166:392–398. doi: 10.1128/jb.166.2.392-398.1986. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Gibson D T, Resnick S M, Lee K, Brand J M, Torok D S, Wackett L P, Schocken M J, Haigler B E. Desaturation, dioxygenation, and monooxygenation reactions catalyzed by naphthalene dioxygenase from Pseudomonas sp. strain 9816-4. J Bacteriol. 1995;177:2615–2621. doi: 10.1128/jb.177.10.2615-2621.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
17a.Happe, B. Unpublished results.
18.Happe B, Eltis L D, Poth H, Hedderich R, Timmis K N. Characterization of 2,2′,3-trihydroxybiphenyl dioxygenase, an extradiol dioxygenase from the dibenzofuran- and dibenzo-p-dioxin-degrading bacterium Sphingomonas sp. strain RW1. J Bacteriol. 1993;175:7313–7320. doi: 10.1128/jb.175.22.7313-7320.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Harayama S, Timmis K N. Aerobic biodegradation of aromatic hydrocarbons by bacteria. In: Sigel H S, Sigel A, editors. Metal ions in biological systems. New York, N.Y: Marcel Dekker; 1992. pp. 99–156. [Google Scholar]
20.Harms H, Wittich R-M, Sinnwell V, Meyer H, Fortnagel P, Francke W. Transformation of dibenzo-p-dioxin by Pseudomonas sp. strain HH69. Appl Environ Microbiol. 1990;56:1157–1159. doi: 10.1128/aem.56.4.1157-1159.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
20a.Hofer, B. Unpublished results.
21.Jiang H, Parales R E, Lynch N A, Gibson D T. Site-directed mutagenesis of conserved amino acids in the alpha subunit of toluene dioxygenase: potential mononuclear non-heme iron coordination sites. J Bacteriol. 1996;178:3133–3139. doi: 10.1128/jb.178.11.3133-3139.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Kim Y-C, Kim S-J, Kim C-K, Kim S-J. A novel organization of phn operons responsible for biodegradation of polyaromatic hydrocarbons (PAHs) by Pseudomonas sp. strain DJ77. In: Timmis K N, Lalucat J, editors. Biodegradation of organic pollutants. Abstract book, UIB-GBF-CSIC-TUB symposium. Braunschweig, Germany: GBF; 1996. pp. 189–190. [Google Scholar]
23.Klecka G M, Gibson D T. Metabolism of dibenzo[1,4]dioxan by a Pseudomonas species. Biochem J. 1979;180:639–645. doi: 10.1042/bj1800639. [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Kovach M E, Phillips R W, Elzer P H, Roop R M N, Peterson K M. pBBR1MCS: a broad-host-range cloning vector. BioTechniques. 1994;16:800–802. [PubMed] [Google Scholar]
25.McKay D B, Seeger M, Zielinski M, Hofer B, Timmis K N. Heterologous expression of biphenyl dioxygenase-encoding genes from a gram-positive broad-spectrum polychlorinated biphenyl degrader and characterization of chlorobiphenyl oxidation by the gene products. J Bacteriol. 1997;179:1924–1930. doi: 10.1128/jb.179.6.1924-1930.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Mermod N, Harayama S, Timmis K N. New route to bacterial production of indigo. Bio/Technology. 1986;4:321–324. [Google Scholar]
27.Monna L, Omori T, Kodama T. Microbial degradation of dibenzofuran, fluorene, and dibenzo-p-dioxin by Staphylococcus auriculans DBF63. Appl Environ Microbiol. 1993;59:285–289. doi: 10.1128/aem.59.1.285-289.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Robertson J B, Spain J C, Haddock J D, Gibson D T. Oxidation of nitrotoluenes by toluene dioxygenase: evidence for a monooxygenase reaction. Appl Environ Microbiol. 1992;58:2643–2648. doi: 10.1128/aem.58.8.2643-2648.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Sambrook J, Fritsch E F, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory; 1989. [Google Scholar]
30.Sato S I, Nam J W, Kasuga K, Nojiri H, Yamane H, Omori T. Identification and characterization of genes encoding carbazole 1,9a-dioxygenase in Pseudomonas sp. strain CA10. J Bacteriol. 1997;179:4850–4858. doi: 10.1128/jb.179.15.4850-4858.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Schmid A, Rothe B, Altenbuchner J, Ludwig W, Engesser K H. Characterization of three distinct extradiol dioxygenases involved in mineralization of dibenzofuran by Terrabacter sp. strain DPO360. J Bacteriol. 1997;179:53–62. doi: 10.1128/jb.179.1.53-62.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Strubel V, Engesser K H, Fischer P, Knackmuss H J. 3-(2-Hydroxyphenyl)catechol as substrate for proximal meta ring cleavage in dibenzofuran degradation by Brevibacterium sp. strain DPO 1361. J Bacteriol. 1991;173:1932–1937. doi: 10.1128/jb.173.6.1932-1937.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Strubel V, Rast H G, Fietz W, Knackmuss H J, Engesser K H. Enrichment of dibenzofuran utilizing bacteria with high co-metabolic potential towards dibenzodioxin and other anellated aromatics. FEMS Microbiol Lett. 1989;49:233–238. doi: 10.1016/0378-1097(89)90044-x. [DOI] [PubMed] [Google Scholar]
34.Taira K, Hirose J, Hayashida S, Furukawa K. Analysis of bph operon from the polychlorinated biphenyl-degrading strain of Pseudomonas pseudoalcaligenes KF707. J Biol Chem. 1992;267:4844–4853. [PubMed] [Google Scholar]
35.Van den Berg M, De Jongh J, Poiger H, Olson J R. The toxicokinetics and metabolism of polychlorinated dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) and their relevance for toxicity. Crit Rev Toxicol. 1994;24:1–74. doi: 10.3109/10408449409017919. [DOI] [PubMed] [Google Scholar]
36.Wittich R M, Wilkes H, Sinnwell V, Francke W, Fortnagel P. Metabolism of dibenzo-p-dioxin by Sphingomonas sp. strain RW1. Appl Environ Microbiol. 1992;58:1005–1010. doi: 10.1128/aem.58.3.1005-1010.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Yeh W K, Gibson D T, Liu T N. Toluene dioxygenase: a multicomponent enzyme system. Biochem Biophys Res Commun. 1977;78:401–410. doi: 10.1016/0006-291x(77)91268-2. [DOI] [PubMed] [Google Scholar]
38.Zylstra G J, Gibson D T. Toluene degradation by Pseudomonas putida F1. Nucleotide sequence of the todC1C2BADE genes and their expression in Escherichia coli. J Biol Chem. 1989;264:14940–14946. [PubMed] [Google Scholar]
39.Zylstra G J, Goyal A K, Kim E. Diversity of genes for polycyclic aromatic hydrocarbon degradation. In: Ramos J L, editor. Environmental biotechnology. Abstract book. Granada, Spain: EC-US Task Force on Biotechnology Research symposium. CSIC; 1996. pp. 38–41. [Google Scholar]

[B1] 1.Armengaud J, Timmis K N. Molecular characterization of Fdx1, a putidaredoxin-type [2Fe-2S] ferredoxin able to transfer electrons to the dioxin dioxygenase of Sphingomonas sp. RW1. Eur J Biochem. 1997;247:833–842. doi: 10.1111/j.1432-1033.1997.00833.x. [DOI] [PubMed] [Google Scholar]

[B2] 2.Armengaud J, Timmis K N. The reductase RedA2 of the multicomponent dioxin dioxygenase system of Sphingomonas sp. RW1 is related to class-I cytochrome P450-type reductases. Eur J Biochem. 1998;253:437–444. doi: 10.1046/j.1432-1327.1998.2530437.x. [DOI] [PubMed] [Google Scholar]

[B3] 3.Asturias J A, Diaz E, Timmis K N. The evolutionary relationship of biphenyl dioxygenase from gram-positive Rhodococcus globerulus P6 to multicomponent dioxygenases from gram-negative bacteria. Gene. 1995;156:11–18. doi: 10.1016/0378-1119(94)00530-6. [DOI] [PubMed] [Google Scholar]

[B4] 4.Batie C J, Ballou D P, Correll C C. Phthalate dioxygenase reductase and related flavin-iron-sulfur containing electron transferases. In: Müller F, editor. Chemistry and biochemistry of flavoenzymes. Boca Raton, Fla: CRC Press; 1992. pp. 543–566. [Google Scholar]

[B5] 5.Beil S, Happe B, Timmis K N, Pieper D H. Genetic and biochemical characterization of the broad spectrum chlorobenzene dioxygenase from Burkholderia sp. strain PS12: dechlorination of 1,2,4,5-tetrachlorobenzene. Eur J Biochem. 1997;247:190–199. doi: 10.1111/j.1432-1033.1997.00190.x. [DOI] [PubMed] [Google Scholar]

[B6] 6.Brady L, Brzozowski A M, Derewenda Z S, Dodson E, Dodson G, Tolley S, Turkenburg J P, Christiansen L, Huge-Jensen B, Norskov L, et al. A serine protease triad forms the catalytic centre of a triacylglycerol lipase. Nature. 1990;343:767–770. doi: 10.1038/343767a0. [DOI] [PubMed] [Google Scholar]

[B7] 7.Bünz P V, Cook A M. Dibenzofuran 4,4a-dioxygenase from Sphingomonas sp. strain RW1: angular dioxygenation by a three-component enzyme system. J Bacteriol. 1993;175:6467–6475. doi: 10.1128/jb.175.20.6467-6475.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B8] 8.Bünz P V, Falchetto R, Cook A M. Purification of two isofunctional hydrolases (EC 3.7.1.8) in the degradative pathway for dibenzofuran in Sphingomonas sp. strain RW1. Biodegradation. 1993;4:171–178. doi: 10.1007/BF00695119. [DOI] [PubMed] [Google Scholar]

[B9] 9.Diaz E, Timmis K N. Identification of functional residues in a 2-hydroxymuconic semialdehyde hydrolase. A new member of the alpha/beta hydrolase-fold family of enzymes which cleaves carbon-carbon bonds. J Biol Chem. 1995;270:6403–6411. doi: 10.1074/jbc.270.11.6403. [DOI] [PubMed] [Google Scholar]

[B10] 10.Eaton R W. p-Cumate catabolic pathway in Pseudomonas putida F1: cloning and characterization of DNA carrying the cmt operon. J Bacteriol. 1996;178:1351–1362. doi: 10.1128/jb.178.5.1351-1362.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B11] 11.Eaton R W, Timmis K N. Characterization of a plasmid-specified pathway for catabolism of isopropylbenzene in Pseudomonas putida RE204. J Bacteriol. 1986;168:123–131. doi: 10.1128/jb.168.1.123-131.1986. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B12] 12.Ensley B D, Ratzkin B J, Osslund T D, Simon M J, Wackett L P, Gibson D T. Expression of naphthalene oxidation genes in Escherichia coli results in the biosynthesis of indigo. Science. 1983;222:167–169. doi: 10.1126/science.6353574. [DOI] [PubMed] [Google Scholar]

[B13] 13.Erickson B D, Mondello F J. Nucleotide sequencing and transcriptional mapping of the genes encoding biphenyl dioxygenase, a multicomponent polychlorinated-biphenyl-degrading enzyme in Pseudomonas strain LB400. J Bacteriol. 1992;174:2903–2912. doi: 10.1128/jb.174.9.2903-2912.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B14] 14.Fortnagel P, Harms H, Wittich R-M, Krohn S, Meyer H, Sinnwell V, Wilkes H, Francke W. Metabolism of dibenzofuran by Pseudomonas sp. strain HH69 and the mixed culture HH27. Appl Environ Microbiol. 1990;56:1148–1156. doi: 10.1128/aem.56.4.1148-1156.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B15] 15.Franklin F C, Bagdasarian M, Bagdasarian M M, Timmis K N. Molecular and functional analysis of the TOL plasmid pWWO from Pseudomonas putida and cloning of genes for the entire regulated aromatic ring meta cleavage pathway. Proc Natl Acad Sci USA. 1981;78:7458–7462. doi: 10.1073/pnas.78.12.7458. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B16] 16.Furukawa K, Miyazaki T. Cloning of a gene cluster encoding biphenyl and chlorobiphenyl degradation in Pseudomonas pseudoalcaligenes. J Bacteriol. 1986;166:392–398. doi: 10.1128/jb.166.2.392-398.1986. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B17] 17.Gibson D T, Resnick S M, Lee K, Brand J M, Torok D S, Wackett L P, Schocken M J, Haigler B E. Desaturation, dioxygenation, and monooxygenation reactions catalyzed by naphthalene dioxygenase from Pseudomonas sp. strain 9816-4. J Bacteriol. 1995;177:2615–2621. doi: 10.1128/jb.177.10.2615-2621.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B17a] 17a.Happe, B. Unpublished results.

[B18] 18.Happe B, Eltis L D, Poth H, Hedderich R, Timmis K N. Characterization of 2,2′,3-trihydroxybiphenyl dioxygenase, an extradiol dioxygenase from the dibenzofuran- and dibenzo-p-dioxin-degrading bacterium Sphingomonas sp. strain RW1. J Bacteriol. 1993;175:7313–7320. doi: 10.1128/jb.175.22.7313-7320.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B19] 19.Harayama S, Timmis K N. Aerobic biodegradation of aromatic hydrocarbons by bacteria. In: Sigel H S, Sigel A, editors. Metal ions in biological systems. New York, N.Y: Marcel Dekker; 1992. pp. 99–156. [Google Scholar]

[B20] 20.Harms H, Wittich R-M, Sinnwell V, Meyer H, Fortnagel P, Francke W. Transformation of dibenzo-p-dioxin by Pseudomonas sp. strain HH69. Appl Environ Microbiol. 1990;56:1157–1159. doi: 10.1128/aem.56.4.1157-1159.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B20a] 20a.Hofer, B. Unpublished results.

[B21] 21.Jiang H, Parales R E, Lynch N A, Gibson D T. Site-directed mutagenesis of conserved amino acids in the alpha subunit of toluene dioxygenase: potential mononuclear non-heme iron coordination sites. J Bacteriol. 1996;178:3133–3139. doi: 10.1128/jb.178.11.3133-3139.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B22] 22.Kim Y-C, Kim S-J, Kim C-K, Kim S-J. A novel organization of phn operons responsible for biodegradation of polyaromatic hydrocarbons (PAHs) by Pseudomonas sp. strain DJ77. In: Timmis K N, Lalucat J, editors. Biodegradation of organic pollutants. Abstract book, UIB-GBF-CSIC-TUB symposium. Braunschweig, Germany: GBF; 1996. pp. 189–190. [Google Scholar]

[B23] 23.Klecka G M, Gibson D T. Metabolism of dibenzo[1,4]dioxan by a Pseudomonas species. Biochem J. 1979;180:639–645. doi: 10.1042/bj1800639. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B24] 24.Kovach M E, Phillips R W, Elzer P H, Roop R M N, Peterson K M. pBBR1MCS: a broad-host-range cloning vector. BioTechniques. 1994;16:800–802. [PubMed] [Google Scholar]

[B25] 25.McKay D B, Seeger M, Zielinski M, Hofer B, Timmis K N. Heterologous expression of biphenyl dioxygenase-encoding genes from a gram-positive broad-spectrum polychlorinated biphenyl degrader and characterization of chlorobiphenyl oxidation by the gene products. J Bacteriol. 1997;179:1924–1930. doi: 10.1128/jb.179.6.1924-1930.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B26] 26.Mermod N, Harayama S, Timmis K N. New route to bacterial production of indigo. Bio/Technology. 1986;4:321–324. [Google Scholar]

[B27] 27.Monna L, Omori T, Kodama T. Microbial degradation of dibenzofuran, fluorene, and dibenzo-p-dioxin by Staphylococcus auriculans DBF63. Appl Environ Microbiol. 1993;59:285–289. doi: 10.1128/aem.59.1.285-289.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B28] 28.Robertson J B, Spain J C, Haddock J D, Gibson D T. Oxidation of nitrotoluenes by toluene dioxygenase: evidence for a monooxygenase reaction. Appl Environ Microbiol. 1992;58:2643–2648. doi: 10.1128/aem.58.8.2643-2648.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B29] 29.Sambrook J, Fritsch E F, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory; 1989. [Google Scholar]

[B30] 30.Sato S I, Nam J W, Kasuga K, Nojiri H, Yamane H, Omori T. Identification and characterization of genes encoding carbazole 1,9a-dioxygenase in Pseudomonas sp. strain CA10. J Bacteriol. 1997;179:4850–4858. doi: 10.1128/jb.179.15.4850-4858.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B31] 31.Schmid A, Rothe B, Altenbuchner J, Ludwig W, Engesser K H. Characterization of three distinct extradiol dioxygenases involved in mineralization of dibenzofuran by Terrabacter sp. strain DPO360. J Bacteriol. 1997;179:53–62. doi: 10.1128/jb.179.1.53-62.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B32] 32.Strubel V, Engesser K H, Fischer P, Knackmuss H J. 3-(2-Hydroxyphenyl)catechol as substrate for proximal meta ring cleavage in dibenzofuran degradation by Brevibacterium sp. strain DPO 1361. J Bacteriol. 1991;173:1932–1937. doi: 10.1128/jb.173.6.1932-1937.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B33] 33.Strubel V, Rast H G, Fietz W, Knackmuss H J, Engesser K H. Enrichment of dibenzofuran utilizing bacteria with high co-metabolic potential towards dibenzodioxin and other anellated aromatics. FEMS Microbiol Lett. 1989;49:233–238. doi: 10.1016/0378-1097(89)90044-x. [DOI] [PubMed] [Google Scholar]

[B34] 34.Taira K, Hirose J, Hayashida S, Furukawa K. Analysis of bph operon from the polychlorinated biphenyl-degrading strain of Pseudomonas pseudoalcaligenes KF707. J Biol Chem. 1992;267:4844–4853. [PubMed] [Google Scholar]

[B35] 35.Van den Berg M, De Jongh J, Poiger H, Olson J R. The toxicokinetics and metabolism of polychlorinated dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) and their relevance for toxicity. Crit Rev Toxicol. 1994;24:1–74. doi: 10.3109/10408449409017919. [DOI] [PubMed] [Google Scholar]

[B36] 36.Wittich R M, Wilkes H, Sinnwell V, Francke W, Fortnagel P. Metabolism of dibenzo-p-dioxin by Sphingomonas sp. strain RW1. Appl Environ Microbiol. 1992;58:1005–1010. doi: 10.1128/aem.58.3.1005-1010.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B37] 37.Yeh W K, Gibson D T, Liu T N. Toluene dioxygenase: a multicomponent enzyme system. Biochem Biophys Res Commun. 1977;78:401–410. doi: 10.1016/0006-291x(77)91268-2. [DOI] [PubMed] [Google Scholar]

[B38] 38.Zylstra G J, Gibson D T. Toluene degradation by Pseudomonas putida F1. Nucleotide sequence of the todC1C2BADE genes and their expression in Escherichia coli. J Biol Chem. 1989;264:14940–14946. [PubMed] [Google Scholar]

[B39] 39.Zylstra G J, Goyal A K, Kim E. Diversity of genes for polycyclic aromatic hydrocarbon degradation. In: Ramos J L, editor. Environmental biotechnology. Abstract book. Granada, Spain: EC-US Task Force on Biotechnology Research symposium. CSIC; 1996. pp. 38–41. [Google Scholar]

PERMALINK

Genetic Analysis of Dioxin Dioxygenase of Sphingomonas sp. Strain RW1: Catabolic Genes Dispersed on the Genome

Jean Armengaud

Birgitta Happe

Kenneth N Timmis

Abstract

FIG. 1.

MATERIALS AND METHODS

Materials.

General DNA procedures and sequence analysis.

Generation of two specific probes and library screening.

FIG. 2.

Construction of a dioxin dioxygenase expression vector.

Resting cell assays.

Identification of additional α-subunit-encoding cistrons.

Nucleotide sequence accession number.

RESULTS

The cistrons specifying the dioxin dioxygenase are linked to a hydrolase gene.

Cloning and sequence analysis of the dxn cluster.

FIG. 3.

TABLE 1.

(i) Cistrons dxnA1A2 encode dioxin dioxygenase.

FIG. 4.

FIG. 5.

(ii) Two other genes clustered with the dxnA1A2 locus.

The dxnA1A2BC is located 4.5 kb downstream of the dbfB gene.

Expression of dxnA1 is repressed in RW1 grown in rich medium.

FIG. 6.

Recombinant dioxin dioxygenase produced in E. coli requires coexpression of a putidaredoxin-type [2Fe-2S] ferredoxin gene for full activity.

FIG. 7.

Identification of three additional α-subunit-encoding cistrons in Sphingomonas sp. strain RW1. (i) Cistrons of two α subunits are clustered with an extradiol dioxygenase gene.

TABLE 2.

(ii) A fourth α-subunit ring-hydroxylating dioxygenase gene is present in the genome of Sphingomonas sp. strain RW1.

Comparison of the four putative α subunits of ring-hydroxylating dioxygenases identified in RW1.

DISCUSSION

FIG. 8.

ACKNOWLEDGMENTS

REFERENCES

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases