Chromosome-encoded gene cluster for the metabolic pathway that converts aniline to TCA-cycle intermediates in Delftia tsuruhatensis AD9.

Abstract

Delftia tsuruhatensis AD9 was isolated as an aniline-degrading bacterium from the soil surrounding a textile dyeing plant. The gene cluster involved in aniline degradation was cloned from the total DNA of strain AD9 into Escherichia coli JM109. After shotgun cloning, two recombinant E. coli strains showing aniline oxidation activity or catechol meta-cleavage activity were obtained by simple plate assays. These strains contained 9.3 kb and 15.4 kb DNA fragments, respectively. Sequence analysis of the total 24.7 kb region revealed that this region contains a gene cluster (consisting of at least 17 genes, named tadQTA1A2BRD1C1D2C2EFGIJKL) responsible for the complete metabolism of aniline to TCA-cycle intermediates. In the gene cluster, the first five genes (tadQTA1A2B) and the subsequent gene (tadR) were predicted to encode a multi-component aniline dioxygenase and a LysR-type regulator, respectively, while the others (tadD1C1D2C2EFGIJKL) were expected to encode meta-cleavage pathway enzymes for catechol degradation. In addition, it was found that the gene cluster is surrounded by two IS1071 sequences, indicating that it has a class I transposon-like structure. PFGE and Southern hybridization analyses confirmed that the tad gene cluster is encoded on the chromosome of strain AD9 in a single copy. These results suggest that, in strain AD9, aniline is degraded via catechol through a meta-cleavage pathway by the chromosome-encoded tad gene cluster. The tad gene cluster showed significant similarity in nucleotide sequence and genetic organization to the plasmid-encoded aniline degradation gene cluster of Pseudomonas putida UCC22.

Full text

Page 1

Chromosome-encoded gene cluster for the
metabolic pathway that converts aniline to TCA-
cycle intermediates in Delftia tsuruhatensis AD9
Quanfeng Liang,1Masahiro Takeo,2Ming Chen,1Wei Zhang,1Yuquan Xu1
and Min Lin1
Correspondence
Min Lin
linmin57@vip.163.com
1Department of Microbiology, Biotechnology Research Institute, Chinese Academy of
Agricultural Sciences, 12 zhongguancun Nandajie, Beijing 100081, P. R. China
2Department of Materials Science and Chemistry, Graduate School of Engineering, University
of Hyogo, 2167 Shosha, Himeji, Hyogo 671-2201, Japan
Received 20 April 2005
Revised
18 July 2005
Accepted 18 July 2005
Delftia tsuruhatensis AD9 was isolated as an aniline-degrading bacterium from the soil
surrounding a textile dyeing plant. The gene cluster involved in aniline degradation was cloned
from the total DNA of strain AD9 into Escherichia coli JM109. After shotgun cloning, two
recombinant E. coli strains showing aniline oxidation activity or catechol meta-cleavage activity
wereobtainedbysimpleplateassays.Thesestrainscontained9?3 kband15?4 kbDNAfragments,
respectively. Sequence analysis of the total 24?7 kb region revealed that this region contains a
gene cluster (consisting of at least 17 genes, named tadQTA1A2BRD1C1D2C2EFGIJKL)
responsible for the complete metabolism of aniline to TCA-cycle intermediates. In the gene cluster,
the first five genes (tadQTA1A2B) and the subsequent gene (tadR) were predicted to encode
a multi-component aniline dioxygenase and a LysR-type regulator, respectively, while the others
(tadD1C1D2C2EFGIJKL) were expected to encode meta-cleavage pathway enzymes for
catechol degradation. In addition, it was found that the gene cluster is surrounded by two IS1071
sequences, indicating that it has a class I transposon-like structure. PFGE and Southern
hybridization analyses confirmed that the tad gene cluster is encoded on the chromosome of strain
AD9 in a single copy. These results suggest that, in strain AD9, aniline is degraded via catechol
through a meta-cleavage pathway by the chromosome-encoded tad gene cluster. The tad gene
cluster showed significant similarity in nucleotide sequence and genetic organization to the
plasmid-encoded aniline degradation gene cluster of Pseudomonas putida UCC22.
INTRODUCTION
A large quantity of anilines are used in the manufacture of
pesticides, herbicides, dyes, plastics and pharmaceuticals,
and are released into the environment through industrial
wastewaters and their direct application to soils (Meyer,
1981). They are also formed in the environment during the
microbial transformation of nitroaromatic compounds and
aniline-based pesticides (Kearney & Kaufman, 1975). Since
many of them are known to be toxic, mutagenic and
carcinogenic (Crabtree et al., 1991; Chung et al., 1997;
Bhunia et al., 2003), their fate in the environment and
biodegradation have been extensively studied (Bollag et al.,
1978; Lyons et al., 1984, 1985).
In order to determine biodegradability and mechanisms of
biodegradation of anilines, many aniline-degrading bacteria
have been isolated. Species of Alcaligenes (Rhodes, 1970),
Pseudomonas (Anson & Mackinnon, 1984; Aoki et al., 1997;
Travkin et al., 2003), Acinetobacter (Kim et al., 1997),
Rhodococcus (Aoki et al., 1983; Fuchs et al., 1991), Frateuria
(Aoki et al., 1984), Moraxella (Zeyer et al., 1985), Nocardia
(Bachofer et al., 1975) and Delftia (Kahng et al., 2000; Boon
et al., 2001; Liu et al., 2002) are able to degrade aniline and/
or its derivatives. Moreover, plasmid-dependent aniline
degradation has been reported in several bacterial strains
(Anson & Mackinnon, 1984; Latorre et al., 1984; McClure
& Venables, 1987; Fujii et al., 1997; Boon et al., 2001).
However, gene clusters responsible for the complete
conversion of aniline to TCA-cycle intermediates have
been cloned only from the aniline-degradative plasmids
Abbreviations: AD, aniline dioxygenase; Ap, ampicillin; C23O, catechol
2,3-dioxygenase.
The GenBank/EMBL/DDBJ accession numbers of the sequences of
the aniline degradation gene cluster and the 16S rDNA cloned from
strain AD9 are AY940090 and AY89912, respectively.
A supplementary table summarizing the morphological, physiological
and biochemical characteristics of strain AD9, Delftia acidovorans and
Delftia tsuruhatensis is available with the online version of this paper.
0002-8137 G 2005 SGMPrinted in Great Britain3435
Microbiology (2005), 151, 3435–3446
DOI 10.1099/mic.0.28137-0

Page 2

pTDN1 of P. putida UCC22 (Fukumori & Saint, 1997) and
pYA1 of Acinetobacter sp. YAA (Fujii et al., 1997). These
gene clusters are very similar in their genetic organization;
both contain genes encoding a multi-component aniline
dioxygenase (AD), a LysR-type regulator, and several meta-
cleavage pathway enzymes. The ADs encoded by these gene
clusters consist of five proteins, two of which are homo-
logous to glutamine synthetase and glutamine amidotrans-
ferase,suggesting that they areinvolved inthe transferof the
amino group of aniline. Similarities in the sequences of the
other three proteins to other aromatic compound dioxy-
genases suggest that they function as the large and small
subunits of the oxygenase component and the reductase
component in the AD enzyme systems (Fukumori & Saint,
1997; Takeo et al., 1998). Recently, AD genes of the same
type were cloned from Frateuria sp. ANA-18 (Murakami
et al., 2003) and Delftia acidovorans 7N (Urata et al., 2004).
The AD genes of the former strain are located over 1?7 kb
upstream of a catechol ortho-cleavage pathway gene cluster
(the cat1 gene cluster) encoded on the chromosome, while
those of the the latter strain are located just upstream of
a catechol 2,3-dioxygenase gene. However, other meta-
cleavage pathway enzyme genes have not been reported.
Here we report a chromosome-encoded gene cluster
responsible for the complete conversion of aniline to
TCA-cycle intermediates cloned from a Delftia strain.
Interestingly, the gene cluster, surrounded by two IS1071
sequences, is almost identical to that encoded on plasmid
pTDN1 of P. putida UCC22 (Fukumori & Saint, 1997),
which is one of the representative aniline degradation gene
clusters.
METHODS
Bacterial strains, plasmids, media and growth conditions.
The bacterial strains and plasmids used in this study are listed in
Table 1. The culture media used were Luria–Bertani (LB) medium
(Sambrook & Russell, 2001) for bacterial growth and mineral salts
(MS) medium (Saber & Crawford, 1985) for the isolation of aniline-
degrading bacteria, aniline oxygenase assays and aniline degradation
tests. Delftia spp., Acinetobacter calcoaceticus PHEA-2 and Bacillus
thuringiensis A-01 were grown at 30uC; Escherichia coli was grown at
37uC. Ampicillin (Ap) was used in selective media at a final concen-
tration of 50 mg l21.
Isolation and identification of aniline-degrading bacteria.
Soil samples were taken from the area surrounding a textile dyeing
plant in Guangzhou, China. Ten grams of each soil sample were
shaken at 180 r.p.m. in 100 ml 0?9% (w/v) sodium chloride solu-
tion at 30uC for 2 h. One millilitre of the suspension was transferred
into LB medium containing 300 mg aniline l21and the culture was
incubated at 180 r.p.m., at 30uC. After the culture became turbid,
serial dilutions were transferred onto solid MS medium containing
Table 1. Strains and plasmids
Strain or plasmid Relevant characteristicsReference or source
Delftia tsuruhatensis AD9
D. tsuruhatensis T7
Acinetobacter calcoaceticus
PHEA-2
Bacillus thuringiensis
A-01
Escherichia coli DH5a
Wild-type, aniline degradation
ATCC BAA-544T, type strain
Phenol degradation
This study
Shigematsu et al. (2003)
Xu et al. (2003)
Harbours several cryptic plasmidsAuthors’ lab. collection
supE44 DlacU169(w80lacZDM15)
hsdR17 recA1 endA1 gyrA96,
thi-1 relA1
thi D(lac–proAB) F9[traD36 proAB+
lacIqlacZDM15]
Derivative of E. coli K-12
Cloning vector, Apr, 2?7 kb
Broad-host-range vector, AprTcrKmr
Apr, 9?3 kb HindIII fragment from
genomic DNA of strain AD9 in
pUC19, tadQTA1A2BR
Apr, 15?4 kb HindIII fragment from
genomic DNA of strain AD9 in
pUC19, tadC1D2C2EFGIJKL
Apr, 8?2 kb EcoRI fragment from
genomic DNA of strain AD9 in
pUC19, tadBRD1C1D2C2
Apr, 9?3 kb HindIII fragment from
pDA1 in pVK100, tadQTA1A2BR
Sambrook & Russell (2001)
E. coli JM109
Sambrook & Russell (2001)
E. coli K-12 AS1.365
pUC19
pVK100
pDA1
Han et al. (2005)
Takara Shuzo
Sambrook & Russell (2001)
This study
pDB11
This study
pDB2
This study
pVD1
This study
3436Microbiology 151
Q. Liang and others

Page 3

300 mg aniline l21as the sole carbon and nitrogen source, and incu-
bated at 30uC for several days. Rapidly growing colonies of aniline-
utilizing micro-organisms were screened from the plates of MS
medium. Identification of the selected strains was based mainly on
their physiological and biochemical characteristics and was carried
out in the Institute of Microbiology, Chinese Academy of Sciences,
Beijing. Genomic DNA was extracted from strain AD9 and Delftia
tsuruhatensis T7 (ATCC BAA-554T) and purified according to the
method of Marmur (1961) except for the addition of proteinase K
in the SDS-treatment step. DNA base compositions were determined
by thermal denaturation (Marmur & Doty, 1962) using a spectro-
photometer (DU800, Beckman). The genomic DNA of E. coli K-12
was used as the standard for the calibration of Tmvalues. DNA–
DNA hybridizations were carried out according to the methods of
De Ley et al. (1970) and Huss et al. (1983). An almost full-length
16S rDNA sequence of strain AD9, which was obtained as described
by Rochelle et al. (1995), was used for further identification.
Construction of the genomic library of strain AD9. The geno-
mic DNA of strain AD9 was obtained by the method of Wilson
(1987) employing SDS-proteinase K lysis and selective precipitation
of cell debris and polysaccharides with cetyltrimethylammonium
bromide. The genomic DNA obtained was partially digested by
HindIII or EcoRI and separated in a 0?6% (w/v) agarose gel. DNA
fragments of 9–20 kb were purified from the gel using a QIAEXII
Gel Extraction Kit (Qiagen), ligated to pUC19 digested by HindIII
or EcoRI, and introduced into E. coli JM109 by electroporation
using a Gene pulser cell (Bio-Rad) according to the manufacturer’s
instructions.
DNA sequencing and computer sequence analysis. Cloned
DNA fragments were sequenced by TaKaRa Bio (Kyoto), and
nucleotide and amino acid sequences were analysed using DNAman
software (Lynnon Biosoft). Sequence comparisons were made
against the sequences in the GenBank using the BLAST program
(Altschul et al., 1990). Phylogenetic trees were constructed using
GENETYX-WIN Version 5.1 (Genetyx Co.).
Aniline degradation tests and aniline oxygenase assay. One
loop of strain AD9 was inoculated into LB medium and the culture
was incubated overnight at 30uC. The cells were harvested by centri-
fugation (4000 g, 4uC, 10 min), washed with 10 mM potassium
phosphate buffer (pH 7?0) twice, and suspended in MS medium to
give an OD600 of 1?0. Aniline degradation tests were started by
adding aniline to the cell suspension at various concentrations up to
5500 mg l21. The cell suspension was shaken on a rotary shaker at
180 r.p.m., at 30uC, and samples of the suspension were taken at
specific intervals. After removing the cells from the samples by cen-
trifugation (8000 g, 4uC, 10 min), aniline in the supernatant was
detected using the diazo-coupling method (Snell, 1954).
Aniline oxygenase activity was measured using a Clark-type oxygen
electrode (YSI 5100, Yellow Springs Instruments). Fresh cells, pre-
grown in 40 ml LB medium, were harvested by centrifugation (4000 g,
4uC,10 min)andwashedwiththephosphatebuffer(10 mM,pH 7?0).
The cells were suspended in 100 ml MS medium containing aniline
(600 mg l21) or succinate (10 mM), and incubated at 30uC with
rotary shaking (180 r.p.m.) for 24 h to obtain aniline-induced cells or
non-aniline-induced (succinate-grown) cells. The cells were harvested,
washed twice with the phosphate buffer, and suspended in the same
buffer. Oxygen uptake was measured polarographically at 30uC. The
reaction mixture contained 100 mg aniline l21and washed cells in
phosphatebuffer.Endogenousrespirationwasmeasuredintheabsence
of aniline, and the oxygen uptake rates obtained were corrected for the
endogenous respiration.
Expression of AD genes in E. coli. Recombinant E. coli strains
were cultivated in 6 ml LB medium containing Ap. When the
cultures reached early exponential phase, IPTG was added at a final
concentration of 1 mM. After 4 h incubation, the cells were har-
vested, washed with phosphate buffer, and suspended in 0?8% (w/v)
sodium chloride solution. Finally, cell suspensions with an OD660of
10–18 were prepared. The aniline oxygenase activity was measured
polarographically as described above.
Measurement of catechol 2,3-dioxygenase (C23O) activity.
Recombinant E. coli strains were cultivated at 37uC in 3 ml LB
medium containing Ap and IPTG (1?0 mM) until the OD660
reached approximately 1?0. The cells were harvested, washed with
phosphate buffer, and suspended in 1?6 ml 50 mM sodium chloride
solution containing 10% (v/v) ethanol. A crude cell extract was
prepared by sonication of the cell suspension using an ultrasonic
disruptor (TOMY UD-200; power 6, 2 min, three times on ice). Cell
debris was removed by centrifugation (8000 g, 4uC, 10 min). C23O
activity was measured spectrophotometrically by the increase in
absorbance at 375 nm concomitant with the formation of 2-hydro-
xymuconic semialdehyde (Nakazawa & Yokota, 1973). The reaction
mixture contained 0?1 mM catechol and the cell extract in 50 mM
sodium phosphate buffer (pH 7?5) and the reaction was carried out
at 24uC. The quantity of protein in the cell extract was determined
by the Lowry method, using bovine serum albumin as the standard.
One unit of activity was defined as the amount of enzyme required
to produce 1 mmol product min21.
Southern hybridization analysis. The genomic DNA of strain
AD9 was digested by HindIII, EcoRI or PstI. After electrophoresis in
a 1?0% (w/v) agarose gel, the digested fragments were transferred
onto a Hybond-N+nylon membrane (Amersham). A 526 bp DNA
fragment and a 580 bp DNA fragment, which correspond to parts of
the AD genes (tadQ region, see Fig. 1) and the transposase (TnpA)
gene of the IS1071 region (tnpA-L1 region, see Fig. 1), respectively,
were amplified from pDA1 by PCR using the primer sets for the
tadQgene(TadQF,59-ACGATGGTGCTGTTCCGCAA-39,
TadQR, 59-TATGAGGCAGGATGGTGACG-39) and for the TnpA
gene sequence (TnpAF, 59-GCCATTGAAGGTGTCATCCG-39, and
TnpAR, 59-AGGTATTCCACGCCATCACG-39), respectively. The
PCR mixture contained 5 ml 106 ExTaq buffer (TaKaRa), 4 ml
dNTPs (2?5 mM each), 100 pmol of each primer, 1?25 U ExTaq
(TaKaRa), 2?5 ng of the template already described, and sterile dis-
tilled water to adjust the total volume to 50 ml. PCR was carried out
using a PCR Express machine (Thermo Hybaid) and the following
temperature programme: 94uC for 3 min; 30 cycles consisting of
94uC for 1 min, 55uC for 1 min and 72uC for 1 min; and 72uC for
5 min. The PCR products were labelled with digoxigenin using a
DIG-High Prime DNA Labelling and Detection Starter Kit II
(Boehringer Mannheim Biochemicals) and used as gene probes.
Southern hybridization was carried out according to the protocol
for the labelling and detection kit.
and
PFGE. DNA was prepared by the method of Kieser et al. (1992).
PFGE was performed using CHEF DRIII PFGE systems (Bio-Rad).
Electrophoresis was performed at an electric field of 6 V cm21at
14uC for 23 h. The pulse time was increased steadily from 60 s at
the beginning to 90 s at the end or from 30 s to 60 s, and the field
angle was 120u.
RESULTS
Characterization and identification of strain
AD9
Severalmorphologicallydistinctisolateswereobtainedfrom
the enrichment culture using plates of MS medium
containing aniline as a sole carbon and nitrogen source.
http://mic.sgmjournals.org 3437
Aniline degradation gene cluster of a Delftia strain

Page 4

Onesuchstrain,namedAD9,wasselectedfromtheseisolates
for further detailed analyses owing to its ability to grow
rapidly on this medium. The morphological, physiological
and biochemical characteristics of AD9 are summarized in
Supplementary Table S1 available with the online version of
this paper. Based on these characteristics, the Institute of
Microbiology (Chinese Academy of Sciences, Beijing)
tentatively identified AD9 as Delftia acidovorans. To confirm
thisidentificationatthegeneticlevel,thepartial16SrDNAof
AD9 was sequenced and compared with other sequences in
DNA databases. The result showed that the AD9 sequence
is almost identical to those of D. acidovorans MBIC1305
(AB020186, identity >99?1%) and D. tsuruhatensis T7
(AB075017, identity >99?6%). D. tsuruhatensis was
established as a bacterial species (Shigematsu et al., 2003)
after the Institute of Microbiology identified AD9 as D.
acidovorans. (Characteristics of D. acidovorans and D.
tsuruhatensis are summarized in Supplementary Table S1.)
Bathe (2004) constructed a phylogenetic tree based on the
16S rRNA sequences of bacterial strains belonging to the
family Comamonadaceae, in which D. acidovorans strains
and D. tsuruhatensis strains form different branches.
Our similar approach revealed that AD9 belongs to the
D. tsuruhatensis branch and is most closely related to
D. tsuruhatensis T7 (the type strain of D. tsuruhatensis)
(data not shown). Strain AD9 was unable to utilize
3-aminobutyrate as a carbon source; most D. acidovorans
strains can assimilate this compound (Shigematsu et al.,
2003). The G+C content of AD9 is 66?8 mol%, which is
very similar to that of D. tsuruhatensis T7 (66?2 mol%)
(Shigematsu et al., 2003). In addition, the DNA–DNA
hybridization value between AD9 and T7 was 83?8%. These
results all support the identification by the 16S rDNA
sequence. Therefore, we identified AD9 as a strain of D.
tsuruhatensis.
Aniline degradation by strain AD9
Strain AD9 grew on and degraded aniline well at 20–37uC
(the optimum temperature was around 30uC), and it grew
well in a wide pH range from 4?0 to 9?0 (the optimum pH
was around 7?0). Although it could degrade up to 5000 mg
aniline l21in MS medium, over 1000 mg aniline l21
inhibited growth. The highest concentration (5000 mg l21,
53?8 mM) at which AD9 can grow is identical to that of
Delftia sp. AN3 (Liu et al., 2002), which was reported as
the most aniline-tolerant aniline degrader. When cells pre-
grown in LB medium containing aniline were suspended at
an OD600 of 1?0 in MS medium, the culture degraded
1000 mg aniline l21completely within 18 h (data not
shown).Thisabilitytodegradeanilinewasnotlostevenafter
repeatedculturing(sixtimes)inLBmediumwithoutaniline.
Strain AD9 was also able to utilize m-toluidine and
p-toluidine as a sole source of carbon, but not o-toluidine,
4-chloroaniline, 2-chloroaniline, 2,4-xylidine, 3,4-dichloro-
aniline or 2,4-dichloroaniline.
To examine the aniline oxygenase activity of AD9, oxygen
uptake was measured using aniline-induced cells and
non-aniline-induced (succinate-grown) cells. The aniline-
induced cells showed apparent aniline oxygenase activity
of 82±6 mg O2(g dry wt)21h21, while the non-aniline-
induced cells showed only one-third of this activity
[30±2 mg O2(g dry wt)21h21]. This result indicates that
aniline oxidation in AD9 is inducible.
Cloning of aniline degradation genes from AD9
A genomic library was constructed with HindIII-digested
total DNA of AD9 and introduced into competent E. coli
JM109. When transformants were screened on LB plates
containing Ap and aniline, five colonies showed a brown
colour on the plates, indicating accumulation of catechol
resulting from aniline oxidation. A recombinant plasmid
was extracted from one of the positivecolonies and analysed
by restriction enzymes. The analysis revealed that the
recombinant plasmid, designated pDA1, had a 9?3 kb
HindIII insert in the vector pUC19. The transformants
were screened again by spraying their colonies with 0?1 M
catechol solution (in 10 mM phosphate buffer, pH 7?0).
One colony showed a brilliant yellow colour on the plate,
indicating C23O activity. The C23O-positive strain con-
tained a recombinant plasmid, designated pDB11, which
had a 15?4 kb HindIII insert fragment. Restriction analysis
showed that the insert fragment of pBD11 did not overlap
with that of pDA1. To confirm whether the two insert
fragments were situated next to each other, the transfor-
mants obtained from another library constructed with
EcoRI-digested AD9 DNA were screened by the catechol
spray. One C23O-positive strain was obtained. The strain
contained a recombinant plasmid, which had an 8?2 kb
EcoRI insert fragment and was designated pDB2. Restriction
analysis of pDB2 revealed that, as shown in Fig. 1(a), the
insert fragment of pDB2 overlaps the inserts of both pDA1
and pDB11 and that the insert of pDB11 is adjacent to that
of pDA1.
Sequence analysis of cloned fragments
The DNA fragments cloned in pDA1, pDB2 and pDB11
were sequenced, and the nucleotide sequence of the total
24?7 kb region was thereby determined. Homology searches
were performed to identify gene function. It was found that
the region contains 23 intact ORFs (ORF2–24, Table 2), at
least 17 of which (tadQTA1A2BRD1C1D2C2EFGIJKL) were
expected to be involved in the complete metabolism of
aniline to TCA-cycle intermediates as shown in Fig. 1(b).
The first five gene products (TadQTA1A2B) and the
subsequent gene product (TadR) showed significant aa
sequence identity (84–96%) to multi-component aniline
dioxygenases (ADs) and LysR-type regulators, respectively,
found in other aniline-degrading bacteria, P.putida UCC22,
D. acidovorans 7N and Acinetobacter sp. YAA. The remain-
ing 11 gene products (TadD1C1D2C2EFGIJKL) exhibited
considerable identity (73–100%) to meta-cleavage pathway
enzymes found in aromatic-compound-degrading and
aniline-degrading bacteria (Table 2).
3438Microbiology 151
Q. Liang and others

Page 5

Two ORFs, orfS and orfU, whose transcriptional directions
are opposite to each other, are located between tadC1 and
tadD2. The former was predicted to encode another LysR-
type regulator, although the latter encodes an unknown
protein. In addition, there is another ORF, orfV, between
tadC2 and tadE, which also encodes an unknown protein.
Similar ORFs to orfU and orfV have been found in the meta-
cleavage pathways of a phenol degrader, Comamonas
testosteroni TA441 (Arai et al., 2000), and a nitrobenzene
degrader, Comamonas sp. JS765 (AF190463), in addition to
P. putida UCC22 (Fukumori & Saint, 2001). However, the
functions of their gene products have not been identified.
In the region downstream of tadL, there are three ORFs,
orfXYZ, whose gene products show considerable similarity
to MarR-type regulators, b-ketoadipate enol-lactone hydro-
lases and muconate cycloisomerases, respectively (Table 2).
These ORFs are transcribed in the direction opposite to
those of the AD genes and the meta-cleavage pathway
enzyme genes.
Partial sequences encoding the TnpA of IS1071 (Nakatsu
et al., 1991) were found at both ends of this 24?7 kb region.
Inverted repeat sequences of 110 bp (nt 1744–1853 and nt
24403–24512 in AY940090), which are completely identical
to those of IS1071, were found near the TnpA-encoding
sequences,indicatingthatthistadgeneclusterissurrounded
Fig. 1. Genetic organization of the aniline degradation tad gene cluster (a) and putative aniline degradation pathway (b) of
D. tsuruhatensis AD9. The sequenced region of 24?7 kb is shown; the tad gene cluster is surrounded by two IS1071
sequences. Open arrows indicate ORFs; black and grey arrows indicate AD genes and meta-cleavage pathway genes,
respectively. Open bars and arrowheads show DNA fragments cloned in recombinant plasmids and lac promoters on the
plasmids, respectively. Hatched boxes and arrows surrounded by them indicate IS1071 regions. Black bars indicate the
regions for the gene probes used in the following Southern hybridization study. Abbreviations: AD, aniline dioxygenase; C23O,
catechol 2,3-dioxygenase; HMS, 2-hydroxymuconic semialdehyde; HMSD, 2-hydroxymuconic semialdehyde dehydrogenase;
HMSH, 2-hydroxymuconic semialdehyde hydrolase; OE, 2-oxopent-4-dienoate; OEH, 2-oxopent-4-dienoate hydratase; OC,
4-oxalocrotonate; OCD, 4-oxalocrotonate decarboxylase; OCT, 4-oxalocrotonate tautomerase; HO, 4-hydroxy-2-oxovalerate;
HOA, 4-hydroxy-2-oxovalerate aldolase; ADA, acetaldehyde dehydrogenase.
http://mic.sgmjournals.org3439
Aniline degradation gene cluster of a Delftia strain

Page 6

by two IS1071 sequences (nt 1–1853 and nt 24403–24681 in
AY940090).
Expression of AD and C23O genes
Aniline oxygenase activity was measured using cell suspen-
sions of E. coli harbouring pDA1 to examine whether the
cloned ADgenes(tadQTA1A2B)were functional. TheE.coli
cells harbouring pDA1 showed apparent aniline oxygenase
activity [32±3 mg O2(g dry wt)21h21]. The endogenous
respiration (the oxygen uptake without the addition of
aniline) subtracted was 16±2 mg O2(g dry wt)21h21.
Catecholwasdetectedinthecellsuspension astheoxidation
product by GC/MS analysis (data not shown). These results
indicate that the recombinant E. coli oxidized aniline to
catechol. The 9?3 kb HindIII fragment of pDA1 was then
subcloned into the broad-host-range plasmid pVK100 to
make the recombinant plasmid pVD1. This plasmid was
introduced by triparental mating (Winstanley et al., 1989)
into the phenol-degrading bacterium A. calcoaceticus
PHEA-2 (Xu et al., 2003), which can assimilate both
catechol and phenol, but not aniline. The resultant strain
was able to grow on aniline as a sole carbon source,
indicating that the AD genes in pVD1 allowed strain PHEA-
2 to convert aniline into catechol (data not shown).
Furthermore, when pVD1 was introduced into the parent
Table 2. Analytical data on the aniline degradation genes of D. tsuruhatensis AD9
ORF Gene namePosition
(no. of nt)
No. of
aa
Putative functionHomologous proteins* (sequence identity)
1
2
3
4
tnpA-L1D
tadQ
tadT
tadA1
1–1710 (1710)
2809–4287 (1479)
4406–5167 (762)
5204–6550 (1347)
569
492
253
448
Transposase
Amino group transfer
Amino group transfer
Large subunit of terminal
dioxygenase
Small subunit of terminal
dioxygenase
Electron-transfer protein
LysR-type regulator
Plant-type ferredoxin
Catechol 2,3-dioxygenase
Unknown product
LysR-type regulator
Plant-type ferredoxin
Catechol 2,3-dioxygenase
Unknown product
HMSd dehydrogenase
HMSd hydrolase
2-Oxopent-4-dienoate
hydratase
Acetaldehyde dehydrogenase
4-Hydroxy-2-oxovalerate aldolase
4-Oxalocrotonate decarboxylase
4-Oxalocrotonate tautomerase
MarR-type regulator
b-Ketoadipate enol-lactone
hydrolase
Muconate cycloisomerase
Transposase
TnpA (96%)1, TnpA (96%)2, TnpA (96%)3
TdnQ (94%)1, TdnQ (82%)4, TdnQ (82%)5
TdnT (84%)1, TdnT (59%)4, ORF7NB (57%)6
TdnA1 (96%)1, TdnA1 (82%)4, TdnA1 (80%)5
5 tadA26547–7191 (645)214TdnA2 (92%)1, TdnA2 (72%)4, ORF7ND (65%)6
6
7
8
9
10
11
12
13
14
15
16
17
tadB
tadR
tadD1
tadC1
orfU
orfS
tadD2
tadC2
orfV
tadE
tadF
tadG
7203–8216 (1014)
8264–9151 (888)
9163–9522 (360)
9559–10503 (945)
10536–11426 (891)
11457–12362 (906)
12633–12968 (336)
12991–13917 (927)
13931–14362 (432)
14425–15885 (1461)
15959–16819 (861)
16861–17649 (789)
337
295
119
314
296
301
111
308
143
486
286
262
TdnB (84%)1, TdnB (66%)4, ORF7NE (64%)6
TdnR (91%)1, TdnR (74%)4, AtdR (33%)7
TdnD (73%)1, CbzT (47%)8, CdoT (45%)9
TdnC (97%)1, ORF7NH (85%)6, CdoE (85%)9
CdoX1 (93%)9, ORF1 (92%)1, ORFX (81%)10
ORF2 (100%)1, AphT (69%)10, CdoR (66%)8
TdnD2 (83%)1, PhnT1 (54%)11
TdnC2 (93%)1, AphB (42%)10, PhnE2 (41%)11
ORF4 (76%)1, ORF (53%)12, ORF (52%)13
TdnE (94%)1, AphC (68%)10, CdoG (68%)9
TdnF (89%)1, DmpD (72%)14, XylF (70%)15
TdnG (87%)1, CdoH (83%)9, AphE (79%)10
18
19
20
21
22
23
tadI
tadJ
tadK
tadL
orfX
orfY
17664–18572 (909)
18594–19625 (1032)
19640–20428 (789)
20447–20638 (192)
20745–21329 (585)
21596–22405 (810)
302
343
262
63
194
269
TdnI (93%)1, CdoI (83%)9, AphF (83%)10
TdnE (93%)1, AphG (87%)10, TesG (79%)16
TdnK (93%)1, CdoK (81%)9, AphH (80%)10
TdnL (100%)1, AphI (85%)10
RSp0034 (60%)17, ORF6 (58%)18
COG0596 (45%)19
24
25
orfZ
tnpA-L2D
22402–23445 (1044)
24548–24681 (134)
347
44
CatB (51%)20, CatB1 (51%)21, CatB1 (51%)22
TnpA (100%)1, TnpA (100%)2, TnpA (100%)3
*Sequences used were obtained from: 1, BAA12804, BAA12805, BAA12806, BAA12807, BAA12808, BAA12809, BAA12810, BAB62044, BAB62045,
BAB62047, BAB62046, BAB62049, BAB62050, BAB62051, BAB62052, BAB62053, BAB62054, BAB62056, BAB62057, BAB62058, BAB62059;
2, AAT81377; 3, BAB85582; 4, BAC82524, BAC82525, BAC82526, BAC82527, BAC82528, BAC82529; 5, AAO38206, AAO38208; 6, BAD61048,
BAD61050, BAD61051, BAD61054; 7, BAA23553; 8, AAP51196; 9, AAC79917, AAC79918, AAG17128, AAC79916, AAG17134, AAG17135,
AAG17136, AAG17138; 10, BAA88498, BAA88500, BAA34176, BAA88501, BAA88502, BAA88503, BAA88504, BAA88505, BAA88507;
11, AAF02425, AAF02430; 12, AAR03452; 13, ZP_00167856.2; 14, CAA36993; 15, AAT09778; 16, BAC67696; 17, NP_521595; 18, AAL02068;
19, ZP_00167449; 20, BAB21460; 21, BAC16777; 22, NP_943040.
DIncomplete ORFs.
d2-Hydroxymuconic semialdehyde.
3440Microbiology 151
Q. Liang and others

Page 7

strain AD9, the resultant strain accumulated a large amount
of a dark brown compound, probably auto-oxidized
catechol from aniline, in liquid cultures, because of the
multi-copy dose effect of the AD genes. These results
indicate that the cloned AD genes were functional in AD9
and also in E. coli and A. calcoaceticus.
Inthecatechol spray selection,therecombinantE.colistrain
harbouring pDB2 was selected based on yellow colour
formation as an index of C23O activity. To evaluate the
C23O activity, the cell extract of the recombinant strain was
prepared and used to measure C23O activity. The activity
measured was 3?2 units (mg crude protein)21.
Copy number and location of the tad gene
cluster in AD9
Southern hybridization was carried out using a 526 bp gene
probe including a part of the AD genes (tadQ region, see
Fig. 1)inordertodeterminethecopynumberofthetadgene
cluster in strain AD9. The total DNA extracted from AD9
was digested independently by three restriction enzymes
(HindIII, EcoRI or PstI) for this study. As shown in Fig. 2(a,
b), the gene probe hybridized to one band in each lane of the
Southern blot (Fig. 2b) corresponding to the digested DNA
sample lanes of the agarose gel electrophoresis (Fig. 2a),
suggesting that AD9 has the tad gene cluster in a single copy,
if it resides on the chromosome. In order to clarify the
locationofthetadgeneclusterinAD9,PFGEwascarriedout
using the AD9 cells. Fig. 2(c) clearly demonstrates that there
are no detectable large plasmids in AD9, although several
plasmids were detected in Bacillus thuringiensis A-01 (a
plasmid-positive control strain), and the chromosome of
AD9 could be seen together with that of E. coli (negative
control and size marker). Southern hybridization with the
same gene probe revealed that the tad gene cluster is located
on the AD9 chromosome (Fig. 2d).
To determine the copy number of IS1071 on the chromo-
some of AD9, another hybridization was carried out using
a 580 bp gene probe encoding the TnpA gene of IS1071
(tnpA-L1 region, see Fig. 1). As shown in Fig. 2(e, f), two
hybridization signals were detected in the HindIII-digested
and PstI-digested AD9 DNA samples. In the case of the
EcoRI-digested AD9 DNA sample, two positive DNA
fragments with a similar size appear to give one strong
signal in the Southern blot. This result shows that there
are only two copies of IS1071 on the chromosome of AD9
and supports the conclusion that the tad gene cluster,
surrounded by the two IS1071s, resides on the chromosome
in a single copy.
Comparison of the chromosome-encoded tad
gene cluster with the plasmid-encoded tdn
gene cluster
As shown in Table 3, the putative products of the tad genes
showed striking identity to those of the plasmid-encoded
tdn genes of P. putida UCC22. Hence, the gene organization
of the tad gene cluster was compared with that of the tdn
gene cluster in detail (Fig. 3). It was found that the gene
organization of both gene clusters is quite similar and
both are surrounded by two IS1071 sequences includ-
ing TnpA-encoding regions. However, there are some
Fig. 2. Agarose gel electrophoresis profiles (a, e) and PFGE
profile (c) of D. tsuruhatensis AD9 DNA, and the corresponding
Southern blots (b, d, f) obtained in the Southern hybridization
using a 526 bp AD gene probe (b, d) or a 580 bp gene probe
for the transposase gene of IS1071 (f). Lane M1, molecular
size marker (lambda DNA EcoRI/HindIII digest); lane M2, mol-
ecular size marker (lambda concatemer); lanes 1 and 7, HindIII-
digested total DNA of AD9; lanes 2 and 8, EcoRI-digested
total DNA of AD9; lanes 3 and 9, PstI-digested total DNA of
AD9; lane 4, B. thuringiensis A-01 cells (strain harbouring
plasmids, positive control); lane 5, AD9 cells; lane 6, E. coli
K-12 cells (negative control and size marker).
http://mic.sgmjournals.org 3441
Aniline degradation gene cluster of a Delftia strain

Page 8

differencesbetweenthem.First,thetadgeneclusterlacksthe
genes corresponding to orf3 and tdnH of the tdn gene
cluster. The function of the orf3 gene product has not been
characterizedyet,andthetdnHgeneproduct,whichshowed
similarity to short-chain dehydrogenases (Fukumori &
Saint, 2001), seems to be unnecessary for catechol meta-
bolism, because many meta-cleavage pathways do not
contain thisenzyme.Secondly,intheregionbetween tadC1/
tdnC and tadD2/tdnD2 (region I in Fig. 3), the tdn gene
cluster lacks a 270 bp DNA segment of the tad gene cluster
(nt 11266–11535 in AY940090). The loss of this small
segmentinthetadgeneclustercausedthedisruptionoforfU
andorfSandresultedintheformationoflongerorf1andorf2
in the tdn gene cluster. Thirdly, in the region downstream of
tadL/tdnL (region II in Fig. 3), there is a substitution of a
2?3 kb DNA segment. In region II of the tad gene cluster,
there are three ORFs, orfXYZ, as mentioned above. In
contrast, in region II of the tdn gene cluster, there are four
ORFs, which have not been described so far (Fukumori &
Saint, 2001). Herein, we tentatively call them orf5678. The
putative gene products of orf5678 showed considerable
similarity to MarR-type regulators (e.g. AAL02068, identity
58%), the C terminal half of b-ketoadipate enol-lactone
hydrolases (e.g. CAD13826, identity 39%), KfrA-like
proteins (e.g. AAP22622, identity 27%) and integrases/
recombinases (e.g. CAI47894, identity 71%), respectively.
AsshowninFig. 3,a2?3 kbDNAsegmentcontaininghalfof
orfY and intact orfZ in the tad gene cluster (nt 22124–24402
inAY940090)issubstitutedbyasmallersegment containing
orf7 and orf8 in the tdn gene cluster.
Phylogenetic relationships of the AD and C23Os
of AD9 with other homologous proteins
To clarify the phylogenetic relationship of the AD of strain
AD9 with those of other aniline-degrading bacteria,
Fig. 3. Comparison of the chromosome-encoded tad gene cluster of D. tsuruhatensis AD9 with the plasmid-encoded tdn
gene cluster of P. putida UCC22 (a) and regions different between the two gene clusters (b, c). Black and grey arrows in (a)
indicate AD genes and meta-cleavage pathway genes, respectively. Grey regions of arrows in (b) and (c) indicate homologous
regions, and hatched boxes in (a) and (c) indicate IS1071 regions. Black triangles in (b) indicate 63 bp direct repeat
sequences.
3442Microbiology 151
Q. Liang and others

Page 9

phylogenetic trees were constructed using the amino acid
sequences of the glutamine-synthetase-like proteins and the
large subunits of the oxygenase component in the AD
enzyme systems. As shown in Fig. 4(a, b), the phylogenetic
trees constructed consist of two major branches, the Tdn-
branch and the Atd-branch, and both branches are distant
from the E. coli glutamine synthetase, GlnA, or the P. putida
benzoate dioxygenase large subunit, BenA (outer group
members). In these trees, TadQ and TadA1 are located
closest to TdnQ and TdnA1 of P. putida UCC22, respec-
tively.
Tolocate themeta-cleavage pathwayofAD9among those of
other aromatic-compound-degrading bacteria, another
phylogenetic tree was constructed using C23O amino acid
sequences (Fig. 4c). The meta-cleavage pathway of AD9
contains two C23Os, TadC1 and TadC2. In the tree, TadC1
belongs to a branch including TdnC, AlnE and ORF7NH
from other aniline-degrading bacteria, P. putida UCC22,
Pseudomonas sp. AW-2 and D. acidovorans 7N, whereas
TadC2 belongs to a branch including BupB and LapB from
alkylphenol-degradingbacteria,P.putidaMT4andP.putida
KL28. In this tree, TadC1 and TadC2 are also located closest
to TdnC and TdnC2, respectively, from UCC22.
DISCUSSION
As described in the Introduction, many bacterial strains
have been isolated as aniline-degraders. Urata et al. (2004)
Fig. 4.
based on the amino acid sequences of AD
glutamine-synthetase-like proteins (a), oxy-
genase component large subunits (b) and
C23Os (c). These trees were constructed
using the UPGMA method of GENETYX-WIN
version 5.1 (Genetyx). Sequences used were
obtained from AY940090 (AD9), D85415
(UCC22), AB089795 (ANA-18), AY168646,
AB177545(7N), M13746,
(GXA7),D86080
(PRS2000), AB107791 (MT4), AY324644
(KL28),U01825 (IC),
AJ633093(PpG7),
Z84817(HV3), AB004065
AB006479 (TA441), JC5534 (G4), U20258
(PKO1),M83673
(LB400), U53507 (JR1), U24277 (BD2),
J04996 (F1), AJ006126 (P200), X52414
(JMP222), X521415
(CTM),X67860 (FDTP-3),
Strainnames andsequence
TadQ, TadA1, TadC1 or TadC2) are shown
in parentheses.
Phylogenetictrees constructed
AY877266
AF218267 (YAA),
V01161
M33263
(mt-2),
(CF600),
(AW-2),
(KF707),X66122
(JMP222), X69504
respectively.
identity (to
http://mic.sgmjournals.org 3443
Aniline degradation gene cluster of a Delftia strain

Page 10

isolated 19 aniline-degrading bacteria from several different
places in Japan and classified them into eight groups (six
genera): Delftia acidovorans, Acinetobacter sp., Pimelobacter
simplex, Pseudomonas sp., Acinetobacter sp., Comamonas
testosteroni, Acinetobacter juni and Acidovorax sp. This
shows that aniline degradation ability is distributed among
many bacterial species. However, several research groups
have independently reported Delftia strains as aniline-
degraders (Loidl et al., 1990; Brunsbach & Reineke, 1993;
Boon et al., 2001; Liu et al., 2002; Kim et al., 2003; Urata
et al., 2004), suggesting that Delftia is one of the major
aniline-degrading bacterial groups. Despite this, gene
clusters encoding enzymes for the complete metabolism
of aniline to TCA-cycle intermediates have been cloned only
from the plasmids of P. putida UCC22 (Fukumori & Saint,
1997) and Acinetobacter sp. YAA (Fujii et al., 1997),
although AD genes have been cloned from the chromo-
somes of D. acidovorans 7N and Frateuria sp. ANA-18
(Urata et al., 2004; Murakami et al., 2003). In this study, we
isolated for the first time a chromosome-encoded aniline
degradation gene cluster from D. tsuruhatensis AD9, which
is responsible for the complete metabolism of aniline to
TCA-cycle intermediates via a meta-cleavage pathway.
Interestingly, it is quite similar to the plasmid-encoded
tdn gene cluster from UCC22, as shown in Fig. 3. Both gene
clusters (tad and tdn) are surrounded by IS1071 sequences
and their genetic organization is almost identical. These
facts suggest that these two gene clusters have evolved from
a common ancestor and they can move as a mobile element
to bacterial plasmids and chromosomes. At present, we
do not know whether the tad gene cluster can move as a
mobile element, although it has a class I transposon-like
structure (Tsuda et al., 1999). Like these gene clusters,
several catabolic gene clusters surrounded by IS1071 have
been found (reviewed by Wyndham et al., 1994; Tsuda
et al., 1999; Tan, 1999). The best-characterized of these is
Tn5271, found on the plasmid pBRC60 of Alcaligenes
sp. BR60(nowComamonassp. BR60) (Nakatsuetal.,1991),
which carries chlorobenzoate degradation genes, cbaABC.
This transposon could be integrated into the chromosomes
ofD.acidovoransandC.testosteroni(Wyndhametal.,1994).
However, this integration seems to have happened through
the transposition of a larger transposon including the com-
plete Tn5271, because the same larger DNA segments,
including Tn5271 from pBRC60, were integrated into
different sites of their chromosomes (Wyndham et al.,
1994). The detailed structureof the chromosome-integrated
Tn5271 has not been reported. In a different study, Boon
et al. (2001) reported five aniline/chloroaniline-degrading
strains belonging to the family Comamonadaceae, four of
which have a similar aniline degradative plasmid (~100 kb
insize).InaSouthernhybridizationstudyusingatdnQgene
probefromP.putidaUCC22andanIS1071geneprobefrom
Tn5271, both probes hybridized to the same 1?4 kb EcoRI/
PstI-digested DNA fragment from the plasmids (Boon et al.,
2001). This shows that AD genes are linked closely to IS1071
on the plasmids and they may form a catabolic transposon.
However, the detailed structure has not been reported and
the linkage of AD genes to IS1071 has also never previously
beenfoundonbacterialchromosomes.Moreover,exceptfor
somegenomicislandsorconjugativetransposons(Springael
& Top, 2004), only a few detailed structures of catabolic
transposons located on bacterial chromosomes have been
reported (Hoffmann et al., 2003; Shintani et al., 2003).
Therefore, it is quite rare that the same catabolic
transposon-like structure has been found on both a plasmid
of P. putida and a chromosome of D. tsuruhatensis.
Consequently, it is important to analyse their sequences
and gene arrangements to understand their history and
transfer mechanisms.
We found some differences between the tad geneclusterand
thetdngene cluster.In region IofFig. 3,thetdngenecluster
lacks a 270 bp segment of the tad gene cluster, resulting in
thedisruptionoforfSandtheformationofalongerorf2.orfS
is predicted to encode a LysR-type regulator with a normal
size (301 aa) compared to those of other LysR-type
regulators (around 300 aa), whereas orf2 is expected to
encode a larger protein with an unusual size (532 aa) for a
member of this family. The additional sequence of orf2
compared to orfS encodes no known proteins. Thus, it is
natural to think that the 270 bp segment was deleted from
the tad-type gene cluster to form the tdn-type gene cluster.
In region II, a putative transcriptional unit, orfYZ, is
disrupted by the substitution of the 2?3 kb DNA segment.
The reasons why orfY and orfZ were presumed to form a
transcriptional unit are that (i) both ORFs are thought to be
a part of a catechol ortho-cleavage pathway operon, judging
from the sequence similarity of their gene products
(Table 2), and (ii) the 59 terminus of orfY overlaps by
3 bp the 39 terminus of orfZ. Therefore, this substitution
also might have happened after the tad gene cluster had
been established. These sequence analyses suggest that
the tad gene cluster is more ancestral than the tdn gene
cluster.
Our isolate, AD9, grew on and degraded aniline at
concentrations as high as those used by the most aniline-
tolerant aniline degrader Delftia sp. AN3. In the aniline
degradation pathway of AD9, the activities of the key
enzymes, AD and C23O, were experimentally confirmed.
The AD genes of strain AD9 were expressed in E. coli and
more efficiently in A. calcoaceticus PHEA1 and the parent
strain AD9. Although the function of each subunit of ADs
remains unknown, catechol was detected as a metabolite of
aniline after the expression of the AD genes. Moreover, cell
extracts of recombinant E. coli harbouring pDB11 showed
C23O activity. However, pDB11 contains two C23O genes,
tadC1 and tadC2. The phylogenetic analysis of the gene
products TadC1 and TadC2 (Fig. 4) illustrates that these
two C23Os belong to different phylogenetic branches. This
may mean that they havecome from differentmeta-cleavage
pathways. The phylogenetic tree also revealed that TadC1
andTadC2aremostcloselyrelatedtoTdnCandTdnC2ofP.
putida strain UCC22, respectively. Fukumori & Saint (2001)
reported that TdnC and TdnC2 have distinct substrate
3444 Microbiology 151
Q. Liang and others

Page 11

specificity: the E. coli cell extract containing TdnC showed
relatively high activity on substituted catechols (catechol,
100%; 3-methylcatechol, 93%; 4-methylcatechol, 43%),
while that containing TdnC2 showed less activity on
substituted catechols (catechol, 100%; 3-methylcatechol,
4?6%;4-methylcatechol,19%).StrainUCC22canassimlate
m-toluidine (3-methylaniline) and p-toluidine (4-methyl-
aniline) via 3-methylcatechol and 4-methylcatechol, respec-
tively. Therefore, it might be necessary for cells to acquire
another C23O, TdnC, for these methylcatechols, in addi-
tion to TdnC2 for unsubstituted catechol, to expand the
assimilation range for toluidines. Strain AD9 can also
assimilate m-toluidine and p-toluidine. In our preliminary
study using recombinant E. coli cell extracts containing
TadC1orTadC2,C23Oactivitywasseenoncatechol(100%
and 100%, respectively), 3-methylcatechol (56% and 18%)
and 4-methylcatechol (28% and <1%). Therefore, we
confirmed that both tadC1 and tadC2 can produce an active
C23O. The activities of TadC2 towards methylcatechols
were lower than those of TadC1, as reported for TdnC2 and
TdnC by Fukumori & Saint (2001). However, detailed
analysis should be carried out using the purified enzymes.
ACKNOWLEDGEMENTS
This work was supported by the National Natural Science Foundation
(grant nos 30470047 and 30200007) and 863 Project (grant no.
2005AA226030) of the Ministry of Science and Technology of China.
We thank Dr Yuji Nagata (Tohoku Univ., Japan) for his helpful and
useful advice and discussion on PFGE.
REFERENCES
Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J.
(1990). Basic local alignment search tool. J Mol Biol 215, 403–410.
Anson, J. G. & Mackinnon, G. (1984). Novel Pseudomonas plasmid
involved in aniline degradation. Appl Environ Microbiol 48, 868–869.
Aoki, K., Shinke, R. & Nishira, H. (1983). Metabolism of aniline by
Rhodococcus erythropolis AN-13. Agric Biol Chem 47, 1611–1616.
Aoki, K., Ohtsuka, K., Shinke, R. & Nishira, H. (1984). Rapid
biodegradation of aniline by Frateuria species ANA-18. Agric Biol
Chem 48, 856–872.
Aoki, K., Kodama, N., Murakami, S. & Shinke, R. (1997). A high level
of accumulation of 2-hydroxymuconic 6-semialdehyde from aniline
by the transpositional mutant Y-2 of Pseudomonas species AW-2.
Microbiol Res 152, 129–135.
Arai, H., Ohishi, T., Chang, M. Y. & Kudo, T. (2000). Arrangement
and regulation of the genes for meta-cleavage enzymes required for
degradationofphenol in Comamonas
Microbiology 146, 1707–1715.
testosteroni TA441.
Bachofer, R., Lingens, F. & Schafer, W. (1975). Conversion of
aniline into pyrocatechol by a Nocardia sp. incorporation of oxygen-
18. FEBS Lett 50, 288–290.
Bathe, S. (2004). Conjugal transfer of plasmid pNB2 to activated
sludge bacteria leads to 3-chloroaniline degradation in enrichment
cultures. Lett Appl Microbiol 38, 527–531.
Bhunia, F., Saha, N. C. & Kaviraj, A. (2003). Effects of aniline – an
aromatic amine – to some freshwater organisms. Ecotoxicology 12,
397–404.
Bollag, J. M., Blattmann, P. & Laanio, T. (1978). Adsorption and
transformation of four substituted anilines in soil. J Agric Food Chem
26, 1302–1306.
Boon, N. J., Goris, P. D. V., Verstraete, W. & Top, E. M. (2001).
Genetic diversity among 3-chloroaniline- and aniline-degrading strains
of the Comamonadaceae. Appl Environ Microbiol 67, 1107–1115.
Brunsbach, F. R. & Reineke, W. (1993). Degradation of chloroani-
lines in soil slurry by specialized organisms. Appl Microbiol
Biotechnol 40, 2–3.
Chung, K. T., Kirkovsky, L., Kirkovsky, A. & Purcell, W. P. (1997).
Review of mutagenicity of monocyclic aromatic amines: quantitative
structure-activity relationships. Mutat Res 387, 1–16.
Crabtree, H. C., Hart, D., Thomas, M. C., Witham, B. H., McKenzie,
I. G. & Smith, C. P. (1991). Carcinogenic ranking of aromatic amines
and nitro compounds. Mutat Res 264, 155–162.
De Ley, J., Cattoir, H. & Reynaerts, A. (1970). The quantitative
measurement of DNA hybridization from renaturation rates. Eur
J Biochem 12, 133–142.
Fuchs, K., Schreiner, A. & Lingens, F. (1991). Degradation of
2-methylaniline andchlorinated
by Rhodococcus rhodochrous strain CTM. J Gen Microbiol 137,
2033–2039.
isomersof2-methylaniline
Fujii, T., Takeo, M. & Maeda, Y. (1997). Plasmid-encoded genes
specifying aniline oxidation from Acinetobacter sp. strain YAA.
Microbiology 143, 93–99.
Fukumori, F. & Saint, C. P. (1997). Nucleotide sequences and
regulational analysis involved in conversion of aniline to catechol in
Pseudomonas putida UCC22(pTDN1). J Bacteriol 179, 399–408.
Fukumori, F. & Saint, C. P. (2001). Complete nucleotide sequence of
the catechol metabolic region of plasmid pTDN1. J Gen Appl
Microbiol 47, 329–333.
Han, J., Sun, L., Dong, X., Zheng, Q., Sun, X., Yang, H., Wang, Y.
& Song, W. (2005). Characterization of a novel plant growth-
promoting bacteria strain Delftia tsuruhatensis HR4 both as a
diazotroph and a potential biocontrol agent against various plant
pathogens. Syst Appl Microbiol 28, 66–76.
Hoffmann, D., Kleinsteuber, S., Muller, R. H. & Babel, W. (2003). A
transposon encoding the complete 2,4-dichlorophenoxyacetic acid
degradation pathway in the alkalitolerant strain Delftia acidovorans
P4a. Microbiology 149, 2545–2556.
Huss, V. A. R., Fest, H. & Schleifer, K. H. (1983). Studies on the
spectrophotometric determination of DNA hybridization from
renaturation rates. Syst Appl Microbiol 4, 184–192.
Kahng, H. Y., Kukor, J. J. & Oh, K. H. (2000). Characterization
ofstrain HY99, a novel microorganism capable of aerobic and
anaerobic degradation of aniline. FEMS Microbiol Lett 190, 215–221.
Kearney, P. C. & Kaufman, D. D. (1975). Herbicides: Chemistry,
Degradation, and Mode of Action, 2nd edn. New York: Marcel Dekker.
Kieser, H. M., Kieser, T. & Hopwood, D. A. (1992). A combined
genetic and physical map of the Streptomyces coelicolor A3(2)
chromosome. J Bacteriol 174, 5496–5507.
Kim, S. I., Leem, S.-H., Choi, J.-S., Chung, Y. H., Kim, S., Park,
Y.-M., Park, Y. K., Lee, Y. N. & Ha, K.-S. (1997). Cloning and
characterization of two catA genes in Acinetobacter lwoffii K24.
J Bacteriol 179, 5226–5231.
Kim, H. J., Kim, S. E., Kim, J. G. & 7 other authors (2003). Cloning of
genes involved in aniline degradation from Delftia acidovorans.
Hanguk Misaengmul Saengmyong Konghakhoe Chi 31, 25–31.
Latorre, J., Reineke, W. & Knackmuss, H. J. (1984). Microbial
metabolism of chloroanilines enhanced evolution by natural genetic
exchange. Arch Microbiol 140, 159–165.
http://mic.sgmjournals.org3445
Aniline degradation gene cluster of a Delftia strain

Page 12

Liu, Z., Yang, H., Huang, Z., Zhou, P. & Liu, S. J. (2002). Degradation
of aniline by newly isolated, extremely aniline-tolerant Delftia
sp. AN3. Appl Microbiol Biotechnol 58, 679–682.
Loidl, M., Hinteregger, C., Ditzelmueller, G., Ferschl, A. &
Streichsbier, F. (1990). Degradation of aniline and monochlorinated
anilines by soil-borne Pseudomonas acidovorans strains. Arch
Microbiol 155, 56–61.
Lyons, C. D., Katz, S. & Bartha, R. (1984). Mechanisms and pathways
of aniline elimination from aquatic environment. Appl Environ
Microbiol 48, 491–496.
Lyons, C. D., Katz, S. & Bartha, R. (1985). Persistence and mutagenic
potential of herbicide-derived aniline residues in pond water. Bull
Environ Contam Toxicol 35, 696–703.
Marmur, J. (1961). A procedure for the isolation of DNA from
microorganisms. J Mol Biol 3, 208–218.
Marmur, J. & Doty, P. (1962). Determination of the base composition
of deoxyribonucleic acid from thermal denaturation temperature.
J Mol Biol 5, 109–118.
McClure, N. C. & Venables, W. A. (1987). pTND1, a catabolic
plasmid involved in aromatic amine catabolism in Pseudomonas
putida mt-2. J Gen Microbiol 133, 2073–2077.
Meyer, U. (1981). Biodegradation of synthetic organic colorants. In
Microbial Degradation of Xenobiotics and Recalcitrant Compounds,
pp. 371–385. Edited by T. Leisinger and others. London: Academic
Press.
Murakami, S., Hayashi, T., Maeda, T., Takenaka, S. & Aoki, K.
(2003). Cloning and functional analysis of aniline dioxygenase gene
cluster, from Frateuria species ANA-18, that metabolizes aniline via
an ortho-cleavage pathway of catechol. Biosci Biotechnol Biochem 67,
2351–2358.
Nakatsu, C., Ng, J., Singh, R., Straus, N. & Wyndham, C. (1991).
Chlorobenzoate catabolic transposon Tn5271 is a composite class I
element with flanking class II insertion sequences. Proc Natl Acad Sci
U S A 88, 8312–8316.
Nakazawa, T. & Yokota, T. (1973). Benzoate metabolism in
Pseudomonas putida (arvilla) mt-2, demonstration of two benzoate
pathways. J Microbiol 115, 262–267.
Rhodes, M. E. (1970). Aniline utilization by Alcaligenes faecalis: a
taxonomic reappraisal. J Appl Bacteriol 33, 714–720.
Rochelle, P. A., Will, J. A. K., Fry, J. C., Jenkins, G. J. S., Parkes, R. J.,
Turley, C.M.& Weightman,
amplification of 16S rRNA genes from deep marine sediments and
seawater to assess bacterial community diversity. In Nucleic Acids in
the Environment, pp. 219–239. Edited by J. T. Trevors & J. D. van
Elsas. New York: Springer.
A. J.(1995).
Extractionand
Saber, D. L. & Crawford, R. L. (1985). Isolation and characterization
of Flavobacterium strains that degrade pentachlorophenol. Appl
Environ Microbiol 50, 1512–1518.
Sambrook, J. & Russell, D. W. (2001). Molecular Cloning: a
Laboratory Manual, 3rd edn. Cold Spring Harbor, NY: Cold
Spring Harbor Laboratory.
Shigematsu, T., Yumihara, K., Ueda, Y., Numaguchi, M., Morimura,
S. & Kida, K. (2003). Delftia tsuruhatensis sp. nov., a terephthalate
assimilating bacterium isolated from activated sludge. Int J Syst
Bacteriol 53, 1479–1483.
Shintani, M., Nojiri, H., Yoshida, T., Habe, H. & Omori, T. (2003).
Carbazole/dioxin-degrading car gene cluster is located on the
chromosome of Pseudomonas stutzeri strain OM1 in a form differ-
ent from the simple transposition of Tn4676. Biotechnol Lett 25,
1255–1261.
Snell, F. D. (1954). Colorimetric Methods of Analysis, 3rd edn, vol. IV,
pp. 198–199. Amsterdam: Van Nostrand.
Springael, D. & Top, E. M. (2004). Horizontal gene transfer and
microbial adaptation to xenobiotics: new types of mobile genetic
elements and lessons from ecological studies. Trends Microbiol 12,
53–58.
Takeo, M., Fujii, T. & Maeda, Y. (1998). Sequence analysis of the
genes encoding a multicomponent dioxygenase involved in oxidation
of aniline and o-toluidine in Acinetobacter sp. strain YAA. J Ferment
Bioeng 85, 17–24.
Tan, H.-M. (1999). Bacterial catabolic transposons. Appl Microbiol
Biotechnol 51, 1–12.
Travkin, V. M., Solyanikova, I. P., Rietjens, I. M., Vervoort, J., van
Berkel, W. J. & Golovleva, L. A. (2003). Degradation of 3,4-dichloro-
and 3,4-difluoroaniline by Pseudomonas fluorescens 26-K. J Environ
Sci Health B 38, 121–132.
Tsuda, M., Tan, H. M., Nishi, A. & Furukawa, K. (1999). Mobile
catabolic genes in bacteria. J Biosci Bioeng 87, 401–410.
Urata, M., Uchida, E., Nojiri, H., Omori, T., Obo, R., Miyaura, N. &
Ouchiyama, N. (2004). Genes involved in aniline degradation by
Delftia acidovorans strain 7N and its distribution in the natural
environment. Biosci Biotechnol Biochem 68, 2457–2465.
Wen, A., Fegan, M., Hayward, C., Chakraborty, S. & Sly, L. I. (1999).
Phylogenetic relationships among members of the Comamonadaceae,
and description of Delftia acidovorans (den Dooren de Jong 1926 and
Tamaoka et al. 1987) gen. nov., comb. nov. Int J Syst Bacteriol 49,
567–576.
Wilson, K. (1987). Preparation of genomic DNA from bacteria. In
Current Protocols in Molecular Biology, pp. 2.10–2.12. Edited by F. M.
Ausubul and others. New York: Wiley.
Winstanley, C., Morgan, J. A., Pickup, R. W., Jones, J. G. & Saunders,
J. R. (1989). Differential regulation of lambda pL and pR promoters
by a cl repressor in a broad-host-range thermoregulated plasmid
marker system. Appl Environ Microbiol 55, 771–777.
Wyndham, R. C., Cashore, A. E., Nakatsu, C. H. & Peel, M. C. (1994).
Catabolic transposons. Biodegradation 5, 323–342.
Xu, Y., Chen, M., Zhang, W. & Lin, M. (2003). Genetic organization of
genes encoding phenol hydroxylase, benzoate 1,2-dioxygenase alpha
subunit and its regulatory proteins in Acinetobacter calcoaceticus
PHEA-2. Curr Microbiol 46, 235–240.
Zeyer, J., Wasserfallen, A. & Timmis, K. N. (1985). Microbial
mineralization of ring-substituted anilines through an ortho-cleavage
pathway. Appl Environ Microbiol 50, 447–453.
3446 Microbiology 151
Q. Liang and others