Functional expression of three rieske non-heme iron oxygenases derived from actinomycetes in Rhodococcus species for investigation of their degradation capabilities of dibenzofuran and chlorinated dioxins

Article abstract of DOI:10.1271/bbb.80680

The activity of Rieske non-heme iron oxygenases (aromatic hydrocarbon dioxygenases, AhDOs) is important for the bacterial degradation of aromatic pollutants such as polycyclic aromatic hydrocarbons and dioxins. During our analysis of the role of AhDOs in

Full text of DOI:10.1271/bbb.80680

Biosci. Biotechnol. Biochem., 73 (4), 822–827, 2009
Functional Expression of Three Rieske Non-Heme Iron Oxygenases Derived
from Actinomycetes in Rhodococcus Species for Investigation
of Their Degradation Capabilities of Dibenzofuran and Chlorinated Dioxins
1
;y
1;2
1
1
1;3
Toshiya IIDA, Yasuhiro MOTEKI, Kaoru NAKAMURA, Katsuhiko TAGUCHI, Masato OTAGIRI,
4
4
2
3;
*
Miwako ASANUMA, Naoshi DOHMAE, Ron USAMI, and Toshiaki KUDO
1
Ecomolecular Biorecycling Science Research Team, RIKEN Advanced Science Institute,
Wako, Saitama 351-0198, Japan
2
Department of Biological Applied Chemistry, Graduate School of Engineering, Toyo University,
Kawagoe, Saitama 350-8585, Japan
3
International Graduate School of Arts and Science, Yokohama City University,
Yokohama, Kanagawa 230-0045, Japan
4
Biomolecular Characterization Team, Advanced Technology Support Division, RIKEN Advanced Science Institute,
Wako, Saitama 351-0198, Japan
Received September 26, 2008; Accepted December 5, 2008; Online Publication, April 7, 2009
[doi:10.1271/bbb.80680]
The activity of Rieske non-heme iron oxygenases
aromatic hydrocarbon dioxygenases, AhDOs) is impor-
Chlorinated dioxins can be microbially degraded
(for reviews, see refs. 8–11). In general, degradation
proceeds co-metabolically, and it is relatively slow for
some congeners, especially of highly chlorinated dioxin.
This tendency may be due partially to differences in the
substrate specificities of the degrading enzymes. For
successful bioremediation of the dioxin-polluted envi-
ronment, enzymes must be found that can degrade
highly chlorinated dioxins more efficiently.
(
tant for the bacterial degradation of aromatic pollutants
such as polycyclic aromatic hydrocarbons and dioxins.
During our analysis of the role of AhDOs in dioxin
bioremediation, some enzymes derived from high
G + C Gram-positive actinomycetes were difficult to
produce in active form in the Escherichia coli protein
expression system. In this study, we constructed a
heterologous expression system for AhDOs in Rhodo-
coccus species using a constitutive expression promoter,
PdfdB, and a shuttle vector, pRK401, and analyzed the
ability of these enzymes to degrade dibenzofuran and
deplete several chlorinated dioxins. Three active AhDOs
expressed in Rhodococcus strains that were difficult to
obtain by the E. coli system showed different regiospe-
cificities for dibenzofuran bioconversion as well as
different substrate depletion specificities for chlorinated
dioxins. Moreover, AhDO derived from R. erythropolis
TA421 showed relatively diverse depletion-substrate
specificity for chlorinated dioxins.
Rieske non-heme iron oxygenases (aromatic hydro-
carbon dioxygenases, AhDOs) potentially can attack
dioxins. These multi-component enzymes consist of
subunits for electron transfer and a terminal oxygenase
that contains Rieske-type iron-sulfur clusters as redox
1
2)
centers. They catalyze the hydroxylation of aromatic
compounds using reducing equivalents of NAD(P)H and
molecular oxygen to convert substrates to dihydrodiol
derivatives. Recent findings suggest that some micro-
organisms harbor multiple copies of various AhDO
1
3)
genes, such as Rhodococcus jostii RHA1 and Bur-
1
4)
kholderia xenovorans LB400,
and these multiple
AhDOs might contribute to the degradation of aromatic
compounds, even if they are non-native substrates, since
AhDOs generally show broad substrate specifici-
Key words: Rhodococcus; Rieske non-heme iron oxy-
genase; dibenzofuran; biphenyl; dioxins
1
5–18)
ties.
To evaluate the degradation capability of the
Environmental pollution by chlorinated dibenzo-p-
dioxins and dibenzofurans (dioxins) is of increasing
concern since these compounds are unintentional by-
products of several industrial processes, such as the
combustion of garbage and the synthesis of chlorinated
organic compounds. They are carcinogenic, mutagenic,
various AhDOs, it is important to use bacterial strains
that are not a wild-type xenobiotic degraders but non-
xenobiotic degraders to express the AhDO gene hetero-
logously. Habe et al. evaluated the degradation capa-
bility of dioxins with two different AhDOs using the
E. coli protein expression system, and indicated that the
AhDOs had different spectra of degradative compe-
1
–3)
and immunotoxic to animals,
disrupting activity even at trace concentrations.
and show endocrine-
4
,5)
19)
tence. However, for some AhDOs derived from high-
Since dioxins are highly resistant to biological decom-
position, they remain in the environment for prolonged
periods.6
G+C Gram-positive actinomycetes, it has been difficult
to obtain active enzymes in the E. coli protein expres-
sion system (our unpublished results and refs. 20–23).
,7)
y
To whom correspondence should be addressed. Present address: Japan Collection of Microorganisms, RIKEN BioResource Center, Wako,
Saitama 351-0198, Japan; Tel: +81-48-462-1111 ext. 5113; Fax: +81-48-462-4617; E-mail: tiida@jcm.riken.jp
Present address: Faculty of Fisheries, Nagasaki University, 1-14 Bunkyo-machi, Nagasaki 852-8521, Japan
*
0
Abbreviations: AhDO, aromatic hydrocarbon dioxygenase; DF, dibenzofuran; DD, dibenzo-p-dioxin; BP, biphenyl; THBP, 2,2 ,3-trihydrox-
ybiphenyl; 2-OHDF, 2-hydroxydibenzofuran; RT, retention time

Heterologous Expression of Dioxygenases in Rhodococcus
Table 1. Primer Sequences Used in This Study
823
Although similar problems not limited to the heterolo-
gous expression of AhDOs have been reported for the
0
0
a
Primer name
Sequence (from 5 to 3 )
expression of other actinomycetales genes using
4–27)
0
0
E. coli,2
PdfdB-5
PdfdB-3
CGGTCGAGGATCTGCTC
solutions have not been fully elucidated.
GGCGCCCACGGGCCCTCCCCTTTCTC
GATATCACCGTGAATGACAGTGGTC
GTTTTCCCAGTCACGAC
A functional expression system for actinomycetales
genes is essential to investigate and utilize the diverse
catabolic abilities of these organisms.
We have cloned and analyzed AhDO genes of
the dibenzofuran (DF)-utilizing strain Terrabacter sp.
YK320) and the biphenyl (BP)-utilizing strains Rhodo-
0
dfdA1-5
M13M4
0
0
bphA1TA421-5 ATGACCAACCAATTGGGTCGCAC
bphA4TA421-3 TCATATGTGGGCACGAGGCGAGC
0
bphA1K37-5 GATATCACAGTTGATTCGCGC
0
0
bphA2K37-3 TCAGAACAGAACGCTGATGT
2
8)
bphA3K37-5 TCTAGAAAGGAGGCAACACGTGAGCGGGCA-
CTCC
coccus erythropolis TA421
and R. rhodochrous
K37.2 In this study, three actinomycetales-derived
3)
bphA4K37-3 TCTAGACGTATCGCCCATCATTTC
AhDO genes for which it is difficult to produce
functionally active proteins in E. coli were functionally
expressed in a Rhodococcus strain using a constitutive
expression promoter for the genus Rhodococcus. Their
ability to degrade and deplete DF and chlorinated
dioxins was analyzed.
aRestriction endonuclease recognition sequences introduced into the primer
sequences are underlined.
Materials and Methods
Bacterial strains, media, and culture conditions. Rhodococcus sp.
strain RD229) and R. erythropolis strains TA422,30) JCM 2892, and
JCM 3201 were used as hosts in the evaluation of AhDO expression,
since they do not degrade DF or BP. Strains RD2 and TA422 have
spontaneously lost mutant genes for the catabolism of DF and BP in
Rhodococcus sp. YK2 and R. erythropolis TA421 respectively. Strain
JCM 2892 (the same as strain IAM 1399) has been used as a host
strain for the heterologous expression of the BP dioxygenase gene in
R. jostii RHA1.31) Strain JCM 3201 is the type strain of R. erythrop-
Fig. 1. Structures of AhDO-Expression Constructs Used in This
Study.
The size and direction of the open reading frames (ORFs),
promoter for dfdA (PdfdA), and heterologous promoter PdfdB are
indicated by gray, white, and black arrows respectively. Gene names
are shown in the arrows. The positions of the restriction endonu-
clease recognition sites used for construction and the positions of the
sites no longer recognized by restriction endonucleases are indicated
outside and within parenthesis respectively. The Shine-Dalgarno
sequence (SD) introduced upstream of the bphA3K37 with PCR is
shown.
ꢀ
olis. Rhodococcus strains were cultured at 30 C in Luria-Bertani
(LB) medium supplemented with 10 mM glucose (LBG) or MM3Y
supplemented with an appropriate carbon source.
2
9)
The carbon
sources used in this study were 10 mM glucose, 0.1% DF, and 0.1%
BP. Kanamycin (25 mg ml 1, except for strain TA422, at
ꢁ
ꢁ1
1
00 mg ml ), which were added to the media for the selection of
transformants. All the chemicals used in this study were of the highest
purity, from Sigma-Aldrich Chemicals (St. Louis, MO), Nacalai
Tesque (Kyoto, Japan), Kanto Chemicals (Tokyo), and Wako Pure
Chemicals (Osaka, Japan). The chemicals used in the identification of
for bphA1A2 and bphA3A4 were ligated downstream of the PdfdB, as
illustrated in Fig. 1 (pRK401-PB+bphAK37).
0
the metabolites are as follows: 2,2 ,3-trihydroxybiphenyl (THBP;
Wako), 2-hydroxydibenzofuran (2-OHDF; Sigma-Aldrich), and
Resting cultures for the biotransformation of DF. Biotransformation
of DF and BP was performed in resting cell cultures prepared from
LBG-grown cells.29) For quantification of DF and its metabolites,
(1S-cis)-3-phenyl-3,5-cyclohexadiene-1,2-diol (cis-2,3-dihydro-2,3-
dihydroxybiphenyl; Sigma-Aldrich).
resting culture (2 ml adjusted to OD600 ¼ 5:0 by adding 2 mmol of DF)
ꢀ
Construction of AhDO-expression plasmid. Polymerase chain
reaction (PCR) was carried out using ExTaq DNA polymerase (Takara
Bio, Shiga, Japan) or KOD-plus DNA polymerase (Toyobo, Osaka,
Japan). The PCR primers used in this study are listed in Table 1. The
nucleotide sequences of PCR-amplified DNA were confirmed prior to
use in the experiments. We used a strong, constitutive promoter
consisting of the 629-bp promoter region of the dioxygenase gene
dfdB29) (PdfdB) with the sequence for SfoI introduced just downstream
of the start codon of dfdB. Expression constructs of AhDO were ligated
to Rhodococcus-E. coli shuttle vector pRK40120) and were used in the
transformation of Rhodococcus strains. Genetic maps of our AhDO
expression constructs are shown in Fig. 1.
was incubated for 3 h at 30 C with 180 rpm rotary shaking and
extracted twice with 4 ml of ethyl acetate after acidification to about
pH 2.0 with 6 M HCl. The dried extracts dissolved in 2 ml of
acetonitrile were analyzed by HPLC. The mobile phase was 0.05%
(v/v) trifluoroacetic acid containing 75% (v/v) acetonitrile, and the
peak areas of the residual substrates and metabolites were measured
at 250 nm in three separate resting culture samples. The lateral
dioxygenation products of DF (i.e., cis-dihydrodiol metabolites) were
measured after acid dehydration as monohydroxydibenzofuran. THBP,
2-OHDF, and DF were quantified by comparison of the areas under the
peak for each sample with those for each calibration standard solution
(0.0125–1.5 mM) at 250 nm.
A 5.2-kb PstI fragment that encoded dfdA1A2A3A4 genes derived
from Terrabacter sp. YK320) was used as a wild-type promoter control
The metabolites of the BphAK37 sample were further analyzed by
matrix-assisted laser desorption ionization time-of-flight mass spec-
trometry (MALDI-TOF MS). The sample (20 ml) was injected into a
reverse-phase Inertsil ODS-3 column (1 ꢂ 100 mm; GL Sciences,
Tokyo) connected to a model 1100 series liquid chromatography
system (Agilent Technologies, Waldbronn, Germany). The peaks were
(pRK401-dfdA). For promoter replacement (i.e., to switch from PdfdA
to PdfdB), the dfdA1-A4 genes with EcoRV sites introduced just
downstream of the start codon of dfdA1 were PCR-amplified and
ligated downstream of PdfdB (pRK401-PB+dfdA). A plasmid, pTB1,
that encoded bphA1A2A3A4 of R. erythropolis TA421 was used as the
PCR template,28) and the amplified DNA fragment extending from the
start codon of bphA1TA421 to the stop codon of bphA4TA421 was ligated
downstream of PdfdB (pRK401-PB+bphATA421). The two functionally
unidentified genes in the bph gene cluster of R. rhodochrous K3723)
separating the terminal dioxygenase genes bphA1A2 from the electron
transfer protein genes bphA3A4 were removed. The DNA fragments
ꢁ1
eluted at a flow rate of 50 ml min by isocratic elution with 0.08%
(v/v) trifluoroacetic acid in 36% (v/v) acetonitrile, and 10-ml fractions
were collected. Selected fractions were subjected to MALDI-TOF MS
(Ultraflex MALDI-TOF; Bruker-Franzen Analytik, Bremen, Germany)
in reflector mode using 2,5-dihydroxybenzoic acid as matrix. The
observed spectra were recalibrated internally using matrix related ions
(Na form 177.01638u, K form 192.99031u).

824
T. IIDA et al.
0
Assay for depletion of chlorinated dioxins. Depletion of chlorinated
Table 2. Bioconversion of Dibenzofuran to 2,2 ,3-Trihydroxybiphen-
yl with Different Expression Promoter and Host Strains
dioxins was assayed with cells prepared as described above, which
were suspended in MM3Y medium with 5 mM glucose to a density of
OD600 ¼ 15. Chlorinated dioxins dissolved in N,N-dimethylformamide
Cultivation periods
and activitya
Strains and expression
constructs
ꢁ
1
were added to 2 ml of cell suspension (final concentration, 10 mg ml
)
ꢀ
and incubated for 48 h at 30 C with 200 rpm rotary shaking. After
incubation, the remaining substrates were extracted twice with 2
volumes of ethyl acetate, and the dried extract was re-dissolved in
2
h
5 h
Rhodococcus sp. RD2
pRK401-dfdA)
2
6.4
54.4
(
1
.0 ml of toluene. Chlorinated dioxins were quantified in three
Rhodococcus sp. RD2
(pRK401-PB+dfdA)
R. erythropolis TA422
individual extracts of each substrate with GC-MS, as described
previously.32) Substrates extracted from the resting cell cultures of RD2
3
2
2
1
33
39
47
40
538
transformed with vector plasmid pRK401 were used as negative
control. The substrate depletion rates were calculated by comparing the
amount of substrate remaining in the AhDO-expressed resting cell
cultures, the mean being the amount of substrate remaining in the
negative control.
351
328
203
(
pRK401-PB+dfdA)
R. erythropolis JCM 2892
pRK401-PB+dfdA)
R. erythropolis JCM 3201
pRK401-PB+dfdA)
(
(
a
0
Results and Discussion
Activity is the amount of 2,2 ,3-trihydroxybiphenyl (mM) produced by 1
0
OD600 unit of cells. Quantification of 2,2 ,3-trihydroxybiphenyl was
performed as described in ‘‘Materials and Methods.’’
Heterologous expression of DfdA in Rhodococcus
species
There are several examples of heterologous expres-
sion of genes derived from actinomycetes in host strains
BphAK37). In this study, resting cell cultures of RD2-
transformants of both bphATA421 and bphAK37 expressed
under the control of PdfdB converted BP to cis-2,3-
dihydro-2,3-dihydroxybiphenyl (data not shown).
2
2)
other than E. coli, such as the genera Pseudomonas,
5,27,33)
Streptomyces,2
and Rhodococcus.
20,21,23,24,26)
The
genus Rhodococcus has been extensively investigated,
especially for its ability to degrade diverse organic
compounds.3 In addition, the presence of a hydro-
Biotransformation of DF with AhDO expressed in
Rhodococcus sp. RD2
4)
3
5,36)
phobic cell envelope
might provide Rhodococcus
This study compared the DF dioxygenation ability of
three AhDO enzymes using AhDO-expressing resting
cells of RD2. We dehydrated the cis-dihydrodiols of
DF to monohydroxy derivatives by acid catalysis prior to
extraction, since some of the dihydrodiols of DF
gradually dehydrated spontaneously during sample prep-
aration and HPLC analysis. The HPLC chromatograms
of the extracts at 250 nm and within UV absorption spec-
trum from 200 to 360 nm are shown in Fig. 2A and B.
DF was added at a final concentration of 1 mM to the
resting cultures, and we detected 0.88 mM of DF in the
negative control cultures (a pRK401-transformed RD2
strain). No metabolite of DF was observed in the
negative control cultures (Fig. 2A, top), suggesting that
the missing 0.12 mM of DF had been lost during the
experimental process or was incorporated into the cells.
Compared to the negative controls, the resting cultures
of the dfdA-expressing strain degraded 94% of DF
(0.83 mM) after 3 h, but two bphA-expressing strains
degraded only 19% (BphATA421) and 28% (BphAK37) of
the DF (0.17 and 0.25 mM respectively). In the extracts
of resting cultures of the dfdA-expressing strain, a major
peak and a minor peak with retention times (RT) of 3.3
and 4.6 were identified as THBP and 2-OHDF respec-
tively (Fig. 2A and B). The 2-OHDF is probably derived
from cis-1,2-dihydro-1,2-dihydroxydibenzofuran by acid
with advantages for the degradation of hydrophobic
xenobiotics. Hence we decided to use strains of the
genus Rhodococcus as hosts for the heterologous
expression of actinomycetales AhDOs that could not
be obtained in active form by means of the E. coli
protein expression system.
The activity of DF-dioxygenase (dfdA1-A4) expressed
under the control of the wild promoter (PdfdA) was
compared with that expressed under the control of a
constitutive strong promoter (PdfdB) (Fig. 1). In strain
RD2, we found that a regulatory gene involved in the
transcriptional activation of PdfdA was lacking, and we
observed low level, constitutive expression of PdfdA (our
unpublished results). Resting cells of RD2 transformed
with pRK401-dfdA (PdfdA-dfdA) or pRK401-PB+dfdA
(
PdfdB-dfdA) were incubated with DF, and the accumu-
lation of THBP (a DF metabolite) in the culture
supernatants after 2 and 5 h was determined by HPLC
(
Table 2). The switch from PdfdA to PdfdB increased
THBP accumulation about 12 times (2 h) and 10 times
5 h). Next we examined the ability of different host
(
strains with the DfdA-expression construct to degrade
DF (Table 2). The highest activity was detected in strain
RD2. Hence we decided to use strain RD2 as the host in
further analysis.
3
7)
dehydration (Fig. 2C), since the cis-1,2-dihydrodiol of
DF rather than the cis-2,3-dihydrodiol has frequently
been identified as a metabolite of DF formed by
Heterologous expression of two different BphAs in
Rhodococcus sp. RD2
1
6,33,38)
The bphA genes, bphATA421 and bphAK37, were
derived from two BP-utilizing Rhodococcus strains,
R. erythropolis TA421 and R. rhodochrous K37, re-
spectively. We have reported that both of these genes
AhDOs.
The amount of 2-OHDF accumulated in
the media was 0.014 mM (this amount was only 1.7% of
that of the THBP, 0.84 mM), suggesting that DfdA
oxygenates DF specifically at position 4, 4a.
Of the two major peaks found in the culture super-
natants of the BphA-expressing strains (Fig. 2A, two
bottom chromatograms), one (RT ¼ 4:6) was identified
as 2-OHDF. MALDI-TOF-MS analysis of the other
peak (RT ¼ 4:8) revealed the monoisotopic mass to
2
3,28)
were involved in BP utilization.
The amino acid
sequences of their respective BphA subunits had only
from 35% (BphA4) to 44% (BphA3) identity. Our
preliminary attempts to express the two BphAs in E. coli
failed (data not shown for BphATA421, see ref. 23 for

Heterologous Expression of Dioxygenases in Rhodococcus
825
A
B
C
Fig. 2. Degradation of DF with Three AhDO-Expressing RD2 Strains.
Extract of resting cell cultures of strain RD2 expressing the DfdA, BphAK37, and BphATA421 enzymes and a negative control (strain RD2
transformed with vector plasmid pRK401) were analyzed by HPLC. A, A 250-nm chromatogram (representative of three experiments) showing
separation of DF metabolites from cultures of cells transformed with the expression constructs and vector plasmid. Significant peaks
(
arrowheads) detected on HPLC analysis with retention times are indicated. The UV-absorption spectra of the peaks are shown in B. Peaks
0
identified with authentic chemicals are 2,2 ,3-trihydroxybiphenyl (RT ¼ 3:3), 2-hydroxydibenzofuran (2-OHDF; RT ¼ 4:6), and dibenzofuran
(
DF; RT ¼ 9:9). The peak with RT ¼ 4:8 was identified as mono-hydroxydibenzofuran, and was perhaps 3-hydroxydibenzofuran judging by the
spectrum features. Specific AhDO-catalyzed and spontaneous or acid-catalyzed dehydration of DF is illustrated in C.
be 184.051u, in good agreement with the calculated
monoisotopic mass for C12H8O2 (184.052u), suggesting
that this is a derivative of monohydroxyl DF. The UV
absorption maxima of the product (215, 232, 255, 298,
and 305; Fig. 2B, RT ¼ 4:8) closely resembled those of
the specificity of the DF oxygenation position between
the two relatively similar AhDOs is interesting, and
indicates that these two AhDOs can be used to
investigate enzymatic control of the oxygenation posi-
tion specificity for DF by AhDOs.
3
7,39)
3
-hydroxydibenzofuran,
leading to the speculation
that it was 3-hydroxydibenzofuran and was derived from
the cis-2,3- or cis-3,4-dihydrodiol of DF. The latter
compound (Fig. 2C) is more likely, since it is a known
AhDO-expressing Rhodococcus sp. RD2 depleted
chlorinated dioxins
To determine the depletion-substrate specificities of
the three AhDOs for chlorinated dioxins, this study
observed the ability of AhDO-expressing RD2 strains to
deplete chlorinated dioxins. The amount of residual
chlorinated dioxins after 48 h of cultivation was com-
pared among cultures of the three AhDO-expressing
RD2 strains and the vector plasmid-transformed RD2
strain (Fig. 3A–C).
1
6,38)
lateral dioxygenation product of DF by AhDOs.
A
small amount of THBP accumulation was observed in
the BphATA421 samples (2.4% of the degraded DF), but
not in the BphAK37 samples. These results suggest that
the two BphA enzymes oxygenate at the lateral rather
than angular positions of DF, and yet oxygenate at
different lateral positions.
The large subunit of the terminal dioxygenase of
The three AhDO-expressing strains almost completely
depleted 10 ppm of 2-chlorodibenzo-p-dioxin under the
culture conditions used in this study (data not shown).
The DfdA and BphATA421 strains depleted 1,3-dichlor-
odibenzo-p-dioxin (1,3-DCDD) by only 29% and 18%
respectively, while the BphAK37 strain depleted
it by 99%. The BphATA421 strain (but not the DfdA- or
BphAK37 strains) depleted 98% of 2,7-DCDD, the DfdA-
and BphATA421 strain (but not BphAK37) depleted 2,8-
dichlorodibenzofuran (2,8-DCDF) almost completely,
and the two BphA strains (but not the DfdA) depleted
2,3-DCDD. The DfdA and BphATA421 strains weakly
depleted 1,4-DCDD, but none of the strains depleted
1,6-DCDD. Only the BphATA421 strain depleted 2,3,7-
trichlorodibenzo-p-dioxin (TrCDD), by about 40%.
The dichlorinated dioxins that were almost complete-
ly depleted by AhDO-expressing resting cells included
2,8-DCDF by DfdA, 1,3-DCDD by BphAK37, and 2,3-
1
2)
AhDO contributes to substrate specificity. The large
subunits of the BP dioxygenases involved in BP-
utilization in the two Rhodococcus strains, BphA1TA421
in R. erythropolis TA421 and BphA1RHA1 in R. jostii
RHA1, showed 79.5% identity, suggesting that the two
AhDOs have similar substrate specificities. Recently,
biotransformation of several aromatic compounds by
resting cells of R. erythropolis IAM 1399 expressing
1
8)
BphARHA1 has been reported.
The results reported
and obtained in this study indicate that the AhDOs for
BP-utilization expressed in heterologous R. erythropolis
host strains are able to oxygenate DF, but that the
oxygenation-position specificities of DF are different:
the relative amount of angular DF-dioxygenation activ-
ity was greater in BphARHA1 (42% of total DF-oxygen-
ation activity) than in BphATA421 (only about 2.4% of
total DF-oxygenation activity). The clear difference in

826
T. IIDA et al.
In contrast, our results here and those of Habe et al.
A
B
C
clearly indicate that individual AhDOs can specifically
deplete chlorinated dioxins. Thus the experimental
approach used in this study can facilitate the develop-
ment of AhDOs that effectively degrade chlorinated
dioxins.
The AhDO-expressing strains used in this study had
limited ability to degrade 1,4-DCDD and 1,6-DCDD.
Schreiner et al. analyzed the influence of substitution
position on the biodegradation of all 210 congeners of
polychlorinated DDs and DFs by aromatic compound-
degrading bacteria, and found that chlorine substitution
4
9)
at position 1, 4, 6, or 9 generally retarded degradation.
Fig. 3. Depletion of Chlorinated Dioxins with Three AhDO-
Expressed RD2 Strains.
Residual chlorinated dioxins in RD2 cultures expressing genes of
Since we could not find other specific reports on the
degradation of 1,4-, 1,6- or 1,9-dichlorinated dioxin
congeners by aromatic compound-degrading bacteria,
further investigation of these congeners is needed. These
results and findings also suggest that AhDOs should be
screened for degrading activity against these dichlori-
nated dioxins to isolate unusual substrate specificities for
chlorinated dioxins.
(A) DfdA, (B) BphAK37, and (C) BphATA421, compared to the
residual substrates in negative control cultures (RD2 cells trans-
formed with vector plasmid). Bars represent the percentages of
substrates remaining in culture. The amounts in the negative controls
were arbitrarily set to 100%. The data are the means with standard
deviations from three determinations.
The BphATA421 enzyme has wider specificity for
dichlorinated dioxins and can deplete 2,3,7-TrCDD,
which closely resembles the most toxic dioxin, 2,3,7,8-
tetrachlorodibenzo-p-dioxin. These results suggest that
the BphATA421 is a better dioxin-degrading enzyme. The
BphATA421 enzyme has angular dioxygenation activity,
while the BphAK37 enzyme does not. This suggests that
enzymes acting on non-chlorinated substrates at many
different positions can catalyze the oxygenation of
dioxins chlorinated at diverse positions. Further analysis
of metabolites of chlorinated dioxins with AhDO-
expressing Rhodococcus strains is required to demon-
strate the relationships of degradative abilities and
regiospecificities.
DCDD, 2,7-DCDD, and 2,8-DCDF by BphATA421.
Clearly, the three AhDO enzymes have different
substrate depletion specificities for dichlorinated diox-
ins. Habe et al. investigated the degradation activity of
dioxins by resting cells of E. coli that heterologously
1
9)
expressed AhDO.
They used two AhDOs having
angular dioxygenation activity for DF, DbfA in Terra-
bacter sp. DBF63 (DFDO, a DF and fluorene-degrading
AhDO) and CarA in Pseudomonas sp. CA10 (CARDO,
a carbazole-degrading AhDO), and reported that DFDO-
expressing resting cells degraded over 95% of 2,3-
DCDD and 2,8-DCDF, but no 2,7-DCDD, and that
CARDO-expressing resting cells degraded 60–80% of
2
,3-DCDD and 2,8-DCDF and 20–40% of 2,7-DCDD.
In conclusion, we report the expression of active
AhDOs of actinomycetales that are difficult to obtain
using the E. coli expression system. We determined
their ability to transform and deplete DF and several
chlorinated dioxins. The experimental system used in
this study, consisting of a Rhodococcus host strain and
the constitutive expression promoter PdfdB active in
Rhodococcus strains, may provide a way to study genes
that are difficult to evaluate in the E. coli system.
These and our findings confirm that the three angular
dioxygenases DfdA, DFDO, and CARDO have strong
ability to degrade 2,8-DCDF.
We found several reports on co-metabolic degradation
of chlorinated dioxins by aromatic compound-degrading
bacteria, wild-type strains that used catabolic pathways
to degrade aromatic compounds for co-metabolism. In
one study, a Gram-negative DF- and DD-utilizing
bacterium, Sphingomonas wittichii RW1, was found
to degrade 2,3- and 2,8-DCDF efficiently and 2,3-
DCDD and 2,7-DCDF relatively slowly. Fukuda et al.
reported that DF-utilizing Sphingomonas sp. strain HL7
was similar to strain RW1 in degradation abilities as to
2
tive actinomycete, Rhodococcus opacus SAO101, uti-
Acknowledgments
4
0)
This work was partly supported by a grant from the
Ecomolecular Science Research Program of RIKEN and
a Grant-in-Aid for Young Scientists (no. 17780069)
from the Ministry of Education, Culture, Sports,
Science, and Technology (MEXT) of Japan.
4
1)
,3-DCDD and 2,8-DCDF. Moreover, a Gram-posi-
lizing DF and DD for growth, was found to deplete 2,3-,
4
2)
2
,7- and 2,8-DCDD by 16–23%.
It has been reported that S. wittichii RW1 has multiple
References
copies of the terminal dioxygenase subunit gene for
AhDO,4 and there is a possibility that these uninves-
tigated AhDOs affect dioxin degradation in RW1 cells.
Also, it has been shown that R. opacus SAO101 can
utilize diverse aromatic compounds,
might have multiple AhDO genes with different specif-
3)
1
)
Cole P, Trichopoulos D, Pastides H, Starr T, and Mandel JS,
Regul. Toxicol. Pharmacol., 38, 378–388 (2003).
2) IARC, IARC Monographs on the Evaluation of Carcinogenic
Risks to Humans, 69 (1997).
4
4)
and therefore
3
)
Van den Berg M, Birnbaum L, Bosveld AT, Brunstrom B, Cook
P, Feeley M, Giesy JP, Hanberg A, Hasegawa R, Kennedy SW,
Kubiak T, Larsen JC, van Leeuwen FX, Liem AK, Nolt C,
Peterson RE, Poellinger L, Safe S, Schrenk D, Tillitt D,
Tysklind M, Younes M, Waern F, and Zacharewski T, Environ.
Health Perspect., 106, 775–792 (1998).
4
5)
icities in addition to narAaAb genes,
for some diverse aromatics-degrading actinomycetes,
as reported
4
6)
such as Rhodococcus jostii RHA1, R. erythropolis
47,48)
TA421,2 and Mycobacterium vanbaalenii PYR-1.
8)

Heterologous Expression of Dioxygenases in Rhodococcus
827
4
)
Grassman JA, Masten SA, Walker NJ, and Lucier GW, Environ.
Health Perspect., 106 (Suppl. 2), 761–775 (1998).
Safe SH, Pharmacol. Ther., 67, 247–281 (1995).
26) Shao ZQ, Seffens W, Mulbry W, and Behki RM, J. Bacteriol.,
177, 5748–5755 (1995).
5
6
7
)
)
)
27) Smith TJ, Lloyd JS, Gallagher SC, Fosdike WL, Murrell JC, and
Dalton H, Eur. J. Biochem., 260, 446–452 (1999).
28) Arai H, Kosono S, Taguchi K, Maeda M, Song E, Fuji F, Chung
S-Y, and Kudo T, J. Ferment. Bioeng., 86, 595–599 (1998).
29) Iida T, Mukouzaka Y, Nakamura K, Yamaguchi I, and Kudo T,
Biosci. Biotechnol. Biochem., 66, 1462–1472 (2002).
30) Kosono S, Maeda M, Fuji F, Arai H, and Kudo T, Appl.
Environ. Microbiol., 63, 3282–3285 (1997).
Freeman RA and Schroy JM, Chemosphere, 14, 873–876 (1985).
Philippi M, Krasnobajew V, Zyeyer J, and H u¨ tter R, ‘‘Microbial
Degradation of Xenobiotics and Recalcitrant Compounds,’’ eds.
Leisinger T, H u¨ tter R, Cook AM, and N u¨ esch J, Academic
Press, New York, pp. 221–233 (1981).
8
9
)
)
Field JA and Sierra-Alvarez R, Chemosphere, 71, 1005–1018
(2008).
Hiraishi A, Microbes Environ., 18, 105–125 (2003).
31) Seto M, Masai E, Ida M, Hatta T, Kimbara K, Fukuda M, and
Yano K, Appl. Environ. Microbiol., 61, 4510–4513 (1995).
32) Iida T, Nakamura K, Izumi A, Mukouzaka Y, and Kudo T,
Arch. Microbiol., 184, 305–315 (2006).
1
0) Nojiri H and Omori T, Biosci. Biotechnol. Biochem., 66, 2001–
2016 (2002).
1
1) Wittich R-M, Appl. Microbiol. Biotechnol., 49, 489–499 (1998).
2) Gibson DT and Parales RE, Curr. Opin. Biotechnol., 11, 236–
1
33) Shindo K, Ohnishi Y, Chun HK, Takahashi H, Hayashi M,
Saito A, Iguchi K, Furukawa K, Harayama S, Horinouchi S,
and Misawa N, Biosci. Biotechnol. Biochem., 65, 2472–2481
(2001).
243 (2000).
1
1
1
3) McLeod MP, Warren RL, Hsiao WW, Araki N, Myhre M,
Fernandes C, Miyazawa D, Wong W, Lillquist AL, Wang D,
Dosanjh M, Hara H, Petrescu A, Morin RD, Yang G, Stott JM,
Schein JE, Shin H, Smailus D, Siddiqui AS, Marra MA, Jones
SJ, Holt R, Brinkman FS, Miyauchi K, Fukuda M, Davies JE,
Mohn WW, and Eltis LD, Proc. Natl. Acad. Sci. USA, 103,
34) Larkin MJ, Kulakov LA, and Allen CC, Curr. Opin. Biotech-
nol., 16, 282–290 (2005).
35) Sokolovska I, Rozenberg R, Riez C, Rouxhet PG, Agathos SN,
and Wattiau P, Appl. Environ. Microbiol., 69, 7019–7027
(2003).
15582–15587 (2006).
4) Chain PS, Denef VJ, Konstantinidis KT, Vergez LM, Agullo L,
Reyes VL, Hauser L, Cordova M, Gomez L, Gonzalez M, Land
M, Lao V, Larimer F, LiPuma JJ, Mahenthiralingam E, Malfatti
SA, Marx CJ, Parnell JJ, Ramette A, Richardson P, Seeger M,
Smith D, Spilker T, Sul WJ, Tsoi TV, Ulrich LE, Zhulin IB, and
Tiedje JM, Proc. Natl. Acad. Sci. USA, 103, 15280–15287
36) Sutcliffe IC, Antonie Van Leeuwenhoek, 74, 49–58 (1998).
37) Cerniglia CE, Morgan JC, and Gibson DT, Biochem. J., 180,
175–185 (1979).
38) Mohammadi M and Sylvestre M, Chem. Biol., 12, 835–846
(2005).
39) Hammer E, Krowas D, Schafer A, Specht M, Francke W, and
Schauer F, Appl. Environ. Microbiol., 64, 2215–2219 (1998).
40) Wilkes H, Wittich R, Timmis K, Fortnagel P, and Francke W,
Appl. Environ. Microbiol., 62, 367–371 (1996).
41) Fukuda K, Nagata S, and Taniguchi H, FEMS Microbiol. Lett.,
208, 179–185 (2002).
(2006).
5) Nojiri H, Nam JW, Kosaka M, Morii KI, Takemura T, Furihata
K, Yamane H, and Omori T, J. Bacteriol., 181, 3105–3113
(
1999).
6) Resnick SM and Gibson DT, Appl. Environ. Microbiol., 62,
073–4080 (1996).
1
1
1
1
2
2
2
2
2
2
4
42) Kimura N and Urushigawa Y, J. Biosci. Bioeng., 92, 138–143
(2001).
7) Iwasaki T, Miyauchi K, Masai E, and Fukuda M, Appl. Environ.
Microbiol., 72, 5396–5402 (2006).
43) Armengaud J, Happe B, and Timmis KN, J. Bacteriol., 180,
3954–3966 (1998).
8) Iwasaki T, Takeda H, Miyauchi K, Yamada T, Masai E, and
Fukuda M, Biosci. Biotechnol. Biochem., 71, 993–1002 (2007).
9) Habe H, Chung JS, Lee JH, Kasuga K, Yoshida T, Nojiri H, and
Omori T, Appl. Environ. Microbiol., 67, 3610–3617 (2001).
0) Iida T, Mukouzaka Y, Nakamura K, and Kudo T, Appl. Environ.
Microbiol., 68, 3716–3723 (2002).
44) Kitagawa W, Kimura N, and Kamagata Y, J. Bacteriol., 186,
4894–4902 (2004).
45) Kimura N, Kitagawa W, Mori T, Nakashima N, Tamura T,
and Kamagata Y, Appl. Microbiol. Biotechnol., 73, 474–484
(2006).
1) Masai E, Yamada A, Healy J, Hatta T, Kimbara K, Fukuda M,
and Yano K, Appl. Environ. Microbiol., 61, 2079–2085 (1995).
2) McKay DB, Seeger M, Zielinski M, Hofer B, and Timmis KN,
J. Bacteriol., 179, 1924–1930 (1997).
46) Kitagawa W, Suzuki A, Hoaki T, Masai E, and Fukuda M,
Biosci. Biotechnol. Biochem., 65, 1907–1911 (2001).
47) Kim SJ, Kweon O, Freeman JP, Jones RC, Adjei MD, Jhoo JW,
Edmondson RD, and Cerniglia CE, Appl. Environ. Microbiol.,
72, 1045–1054 (2006).
3) Taguchi K, Motoyama M, Iida T, and Kudo T, Biosci.
Biotechnol. Biochem., 71, 1136–1144 (2007).
48) Stingley RL, Khan AA, and Cerniglia CE, Biochem. Biophys.
Res. Commun., 322, 133–146 (2004).
4) Denome SA, Olson ES, and Young KD, Appl. Environ.
Microbiol., 59, 2837–2843 (1993).
5) Geueke B and Hummel W, Protein Expr. Purif., 28, 303–309
49) Schreiner G, Wiedmann T, Schimmel H, and Ballschmiter K,
Chemosphere, 34, 1315–1331 (1997).
(2003).