Molecular and Catalytic Properties of 2,4′-Dihydroxyacetophenone Dioxygenase from Burkholderia sp. AZ11

Article abstract of DOI:10.1271/bbb.110867

The gene dad encoding 2,4′-dihydroxyacetophenone (DHAP) dioxygenase was cloned from Burkholderia sp. AZ11. The initiation codon GTG was converted to ATG for high-level expression of the enzyme in Escherichia coli. The enzyme was moderately thermostable, and the recombinant enzyme was briefly purified. The enzyme (M_r = 90 kDa) was a homotetramer with a subunit M_r of 23 kDa. It contained 1.69 mol of non-heme iron, and had a dark gray color. On anaerobic incubation of it with DHAP, the absorption at around 400nm increased due to the formation of an enzyme-DHAP complex. Multiple sequence alignment suggested that His77, His79, His115, and Glu96 in the cupin fold were possible metal ligands. The apparent K_m for DHAP and the apparent V _max were estimated to be 1.60μM and 6.28μmol/min/mg respectively. 2-Hydroxyacetophenone was a poor substrate. CuCl₂ and HgCl₂ strongly inhibited the enzyme, while FeSO₄ weakly activated it.

Full text of DOI:10.1271/bbb.110867

Biosci. Biotechnol. Biochem., 76 (3), 567–574, 2012
Molecular and Catalytic Properties of 2,40-Dihydroxyacetophenone Dioxygenase
from Burkholderia sp. AZ11
Mayu ENYA, Keiko AOYAGI, Yoshihiro HISHIKAWA, Azusa YOSHIMURA,
y
Koichi MITSUKURA, and Kiyofumi MARUYAMA
Department of Biomolecular Science, Faculty of Engineering, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan
Received November 14, 2011; Accepted December 9, 2011; Online Publication, March 7, 2012
[doi:10.1271/bbb.110867]
The gene dad encoding 2,40-dihydroxyacetophenone
(DHAP) dioxygenase was cloned from Burkholderia sp.
AZ11. The initiation codon GTG was converted to ATG
for high-level expression of the enzyme in Escherichia
coli. The enzyme was moderately thermostable, and the
recombinant enzyme was briefly purified. The enzyme
(Mr ¼ 90 kDa) was a homotetramer with a subunit Mr
of 23 kDa. It contained 1.69 mol of non-heme iron, and
had a dark gray color. On anaerobic incubation of it
with DHAP, the absorption at around 400 nm increased
due to the formation of an enzyme-DHAP complex.
Multiple sequence alignment suggested that His77,
His79, His115, and Glu96 in the cupin fold were possible
metal ligands. The apparent Km for DHAP and the
apparent Vmax were estimated to be 1.60 ꢀM and
6.28 ꢀmol/min/mg respectively. 2-Hydroxyacetophe-
none was a poor substrate. CuCl2 and HgCl2 strongly
inhibited the enzyme, while FeSO4 weakly activated it.
dioxygenase (DAD, EC 1.13.11.41) cleaves the hydro-
xyacetyl group of DHAP to yield 4-hydroxybenzoate
and formate. DAD adds the two oxygen atoms of O2 into
two cleavage products one by one.14) Thus bisphenol A
is completely assimilated in a collaboration of bisphenol
A-degrading bacteria and DHAP-degrading bacteria.
DAD has been purified as a nonheme-iron-containing
metalloenzyme from Alcaligenes sp. 4HAP with a view
to cloning the gene dad encoding DAD.15) However,
much about the basic biochemical properties of DAD,
including substrate specificity, kinetic constants, cofac-
tor requirements, inhibitors, optimum pH, stability, and
catalytic mechanism, has not been fully determined.
Crystals of DAD from Alcaligenes sp. were prepared by
Bowyer, but its three-dimensional structure was not
determined.16) Although many genes showing high
degrees of homology to dad have been found in various
bacterial strains, no biochemical investigation of DAD
has been reported. In this paper, we explain the
molecular and catalytic properties of DAD from
DHAP-degrading bacterium Burkholderia sp. strain
AZ11, which appears to be irrelevant to the metabolism
of bisphenol A.
Key words: Burkholderia;
dihydroxyacetophenone;
dioxygenase; metalloenzyme
Microorganisms in soil degrade various aromatic
compounds. Bisphenol A (2,2-bis(4-hydroxyphenyl)-
propane), a zenobiotic that has weak endocrine disrupt-
ing activity,1,2) is cleaved to 4-hydroxybenzaldehyde and
4-hydroxyacetophenone by various bacteria.3–7) Both
compounds are further degraded in such a way that they
can enter the central metabolic pathway. The former is
oxidized to a key intermediate, 4-hydroxybenzoate.4)
The latter is converted to 4-hydroxyphenyl acetate by 4-
hydroxyacetophenone monooxygenase (EC 1.14.13.84),
and subsequently is hydrolyzed to hydroquinone.8–10)
Hydroquinone is cleaved to 4-hydroxymuconic semi-
aldehyde and metabolized further.11) Besides these
metabolic intermediates, a small amount of 2,40-dihy-
droxyacetophenone (DHAP, 2-hydroxy-1-(4-hydroxy-
phenyl)ethanone) is formed as a dead-end product,
which is not metabolized any further but accumulates
in the culture broth of bisphenol A-assimilating bacte-
rium.4) Hopper et al. found that Alcaligenes sp. 4HAP
grown on 4-hydroxyacetophenone does not degrade
hydroquinone, but oxidizes DHAP to 4-hydroxy-
benzoate.12,13) In this strain, a hydroxylase converts
4-hydroxyacetophenone to DHAP, and then DHAP
Materials and Methods
Chemicals. DHAP was synthesized from 2-bromo-40-methoxyace-
tophenone by the method of Robertson and Robinson,17) and identified
by 1H-NMR. All other chemicals were commercially available and of
analytical grade.
Microorganisms. DHAP assimilating bacterium strain AZ11 was
isolated from a soil sample in Gifu, Japan, by enrichment culture
with DHAP as carbon and energy source. Isolation and cultivation
of AZ11 was done with a liquid medium (pH 7.0) containing 4.48 g
.
of Na2HPO4 12H2O, 1.7 g of KH2PO4, 3.0 g of (NH4)2SO4, 0.15 g
of MgSO4 7H2O, 15 mg of CaCl2 2H2O, 0.25 mg of (NH4)6Mo7O24
.
.
.
.
.
4H2O, 2.5 mg of FeSO4 7H2O, 50 mg of CuSO4 5H2O, 25 mg of
.
.
.
CoSO4 7H2O, 1 mg of ZnSO4 7H2O, 0.2 mg of MnSO4 4H2O,
150 mg of yeast extract, and 100 mg of DHAP in 1 L of tap water. It
was aerobic, Gram-negative, oxidase positive, motile, and rod-shaped.
The genomic DNA of strain AZ11 was prepared, and the 16S rRNA
gene was amplified by polymerase chain reaction with 27f and 1525r
primers, as described by Shinoda et al.18) The nucleotide sequence
(1,494-bp long, DDBJ/EMBL/GenBank databases accession no.
AB636680) showed high homology (99.9%) to those of Burkholderia.
Based on these results, AZ11 was identified as Burkholderia. It was
maintained by growth on 1.5% w/v solid agar slant, and stored at 4 ꢀC.
y
To whom correspondence should be addressed. Fax: +81-58-294-4633; E-mail: maruyama@gifu-u.ac.jp
Abbreviations: DAD, 2,40-dihydroxyacetophenone dioxygenase; DHAP, 2,40-dihydroxyacetophenone; IPTG, isopropyl ꢀ-thiogalactopyranoside;
PAGE, polyacrylamide gel electrophoresis; PPB, potassium phosphate buffer

568
M. ENYA et al.
It was subcultured monthly. For large-scale culture, reciprocal culture
with 500-mL Sakaguchi flasks was done at 30 ꢀC for 2 d. Escherichia
coli XL1-Blue MRA and XL1-Blue MRF0 obtained from Stratagene
(La Jolla, CA) and E. coli JM109 obtained from Takara (Ohtsu, Japan)
were used for gene cloning.19,20) For gene expression, E. coli JM109
harboring plasmid pUDP1.3m with a dad insert was cultured at 37 ꢀC
in 1.4 L of LB medium (polypeptone 20 g, yeast extract 10 g, and NaCl
20 g per 1 L distilled H2O, pH 7.0) containing 50 mg/mL ampicillin.
When A660 reached about 0.15, 1 mM isopropyl ꢀ-thiogalactopyrano-
side (IPTG) was added, and culturing was continued for a further 9 h.
Bacterial cells were collected by centrifugation at 4 ꢀC, washed once
with 50 mM potassium phosphate buffer (PPB) (pH 7.0), and stored at
ꢁ25 ꢀC until needed.
membrane (pore size 0.45 mm) (Millipore, Concord, MA) before
determination.22)
To determine the effects of various reagents on enzyme activity,
the enzyme purified from E. coli JM109 carrying pUDP1.3m was
incubated at 4 ꢀC for 1 d in the absence and the presence of 0.4% w/w
Lubrol PX. Then it was preincubated at 25 ꢀC for 5 min with various
reagents (1 mM each) in 50 mM PPB (pH 7.0), and residual activity was
measured spectrophotometrically under standard conditions.
SDS-polyacrylamide gel electrophoresis (PAGE) and amino acid
sequence. SDS–PAGE was done with
a 12.5% gel following
Laemmli.23) Coomassie Brilliant Blue R-250 or a silver staining kit
(Bexel, Union City, CA) was used for protein staining. Electroblotting
to a polyvinyridene difluoride membrane was done with semi-dry
blotting apparatus HorizBLOT AE-6675 (Atto, Tokyo). The protein
band, stained with Coomassie Brilliant Blue R-250, was excised, and
the N-terminal amino acid sequence was determined by Edman
degradation with a Procise 492 protein sequencer (Applied Biosystems,
Foster City, CA). The N-terminal amino acid residue was identified as
dancyl amino acid by the method of Gray.24)
Enzyme purification. Unless otherwise noted, all procedures were
performed at 0–4 ꢀC with 50 mM PPB (pH 7.0).
(i) Purification from Burkholderia sp. AZ11. The bacterial cells
(about 1.7 g wet weight) obtained from 2 L of liquid culture were
suspended in about 8 mL of PPB and disrupted with a Sonifier
(Branson, Danbury, CT). The solution was clarified by centrifugation.
The supernatant (crude extract) was loaded onto a column (3 ꢂ 25 cm)
of Superdex 200 (GE Healthcare, Buckinghamshire, UK) equilibrated
with PPB. The active fractions were pooled and applied to a column
(1:0 ꢂ 12 cm) of Q-Sepharose (Pharmacia, Uppsala, Sweden) previ-
ously equilibrated with PPB. After the column was washed with
200 mL of PPB, the enzyme was eluted with a linear gradient
established with 50 mL of PPB and 50 mL of PPB containing 1 M NaCl.
The active fractions were pooled. Solid (NH4)2SO4 was dissolved
in the solution to 1.5 M. The solution was loaded on a column
(2:2 ꢂ 4:5 cm) of Phenyl Sepharose (GE Healthcare) previously
equilibrated with PPB containing 1.5 M (NH4)2SO4. The column was
washed with 40 mL of equilibration buffer, followed by a linear
gradient established with 50 mL of the equilibration buffer and 50 mL
of PPB. After the column was washed with 60 mL of PPB, the enzyme
was eluted with PPB containing 0.4% Lubrol PX. The active fractions
Molecular mass. The molecular mass of the enzyme was deter-
mined by gel filtration on Superdex 200 (column size, 1:8 ꢂ 22 cm)
previously equilibrated with 50 mM PPB (pH 7.0) and calibrated with
ferritin (Mr ¼ 450 kDa), bovine liver catalase (230 kDa), yeast alcohol
dehydrogenase (150 kDa), bovine serum albumin (66 kDa), ovalbumin
(45 kDa), and bovine pancreas ꢁ-chymotrypsin (23.5 kDa).
TLC. After the enzymatic reaction, the mixture was acidified with
HCl and centrifuged to remove denatured proteins. The supernatant
was extracted with ethyl acetate and analyzed by TLC with silica gel
60 F254 (Merck, Darmstadt, Germany). Toluene-CHCl3-acetone (8:5:7,
v/v/v) was used as solvent. The spots were visualized by UV. The
following Rf-values were obtained: DHAP, 0.51; 4-hydroxybenzoate,
0.38; 2-hydroxyacetophenone, 0.69; benzoate, 0.55.
were pooled and concentrated with
¨
a collodion bag (Sartorius,
Gottingen, Germany), and stored at ꢁ25 ꢀC.
Analysis of metal and inorganic sulfide. Enzyme-bound iron and
copper were determined with a polarized absorption spectrophotometer
Z-5310 (Hitachi, Tokyo). Non-heme iron was also estimated colori-
metrically with 0.53–0.58 mg of the purified enzyme by the method of
Fish.25) Inorganic sulfide was determined with 1.3 mg of the purified
enzyme by the method of Fogo and Popowsky, as modified by
Lovenberg et al.26,27)
(ii) Purification from recombinant E. coli. Cells of E. coli JM109
harboring pUDP1.3m (about 3.5 g wet weight) obtained from 1.4-L
culture were suspended with 14 mL of the buffer and disrupted as
above. Cell debris was removed by centrifugation, and the volume of
the supernatant was adjusted to 140 mL with the buffer (crude extract).
The solution was maintained at 60 ꢀC for 10 min. The denatured
protein was removed by centrifugation. Then the solution was treated
with (NH4)2SO4. The precipitate obtained from 30–55% saturation was
dissolved in 6 mL of the buffer and applied to Superdex 200 column
chromatography, followed by the ion-exchange column chromatog-
raphy with Q-Sepharose, as described above. The active fractions were
pooled and stored at ꢁ25 ꢀC.
Instrumental analysis. Reversed-phase HPLC was conducted at
45 ꢀC with a LC-9A liquid chromatograph (Shimadzu) equipped with a
Cosmosil column 5C18-MS-II (4:6 ꢂ 150 mm, Nacalai Tesque, Kyoto,
Japan), a SPD-6AV UV-Vis detector, a CTO-2A column oven, and a
C-R6A data processor. The flow rate was 1 mL/min. Formate was
detected at 210 nm, and the other compounds at 254 nm. Elution was
done with MeOH-20 mM PPB (pH 7.0) (1:4, v/v). Under these
conditions, formate, 4-hydroxybenzoate, benzoate, DHAP, and 2-
hydroxyacetophenone were eluted at about 2.1, 3.2, 6.2, and 9.2 min
respectively. For quantitative determination of formate, the deprotei-
nized reaction mixture (2.5 mL) was analyzed by HPLC. For determi-
nation of aromatic carboxylates, the reaction mixture (3 mL) was
acidified with HCl and extracted 3 times with ethyl acetate. The extract
was evaporated at 25 ꢀC, the residue was dissolved in 200 mL of elution
buffer, and 10 mL was analyzed by HPLC. Absorption spectra and 1H-
NMR were measured by UV300 (Shimadzu) and by JNM-ECX400P
(JEOL, Tokyo) respectively.
Determination of enzyme activity. DHAP showed an absorption
peak at 280 nm in 50 mM PPB (pH 7.0), while the reaction products,
4-hydroxybenzoate and formate, scarcely absorbed at this wavelength.
Hence the enzyme activity was determined at 25 ꢀC by measuring the
decrease in A280 ("280 ¼ 11 mMꢁ1 cmꢁ1) with spectrophotometer UV-
300 or UV1600 (Shimadzu, Kyoto, Japan). The reaction was carried
out at 25 ꢀC in a cuvette (light path 1 cm). The standard reaction
mixture (3 mL) contained 50 mM PPB (pH 7.0), 30 mM DHAP, and the
enzyme. The activity of each fraction from column chromatography
was briefly measured with Flying Spot Scanner (Shimadzu) equipped
with a 96-well plate. The reaction mixture (200 mL) containing 50 mM
PPB (pH 7.0), 210 mM DHAP, and the enzyme was incubated at room
temperature for 10 min to measure the decrease in A280. Enzyme
activity was also determined at 25 ꢀC by measuring O2 consumption
with Biological Oxygen Monitor 5300 (Yellow Springs Instrument,
Yellow Springs, OH). The reaction mixture (4.0 mL) contained 50 mM
PPB (pH 7.0), 0.35 mM DHAP, and the enzyme. One unit of enzyme
activity was defined as the amount catalyzing the degradation of
1 mmol of substrate per min under the assay conditions. Specific
activity was expressed as units per mg of protein. Protein was
measured by the method of Lowry et al., with bovine serum albumin as
standard.21) When the diluted protein sample was treated, the protein
was precipitated with trichloroacetic acid and collected on a nylon
DNA preparation and manipulation. Genomic DNA of Burkholde-
ria sp. AZ11 was prepared as described by Smith,28) except that
proteinase K treatment (1 mg/mL, 37 ꢀC, 30 min) was also performed
after lysis of the bacterial cells. All oligonucleotides for hybridization,
DNA sequencing, and site-directed mutagenesis were obtained from
Rikakken (Nagoya, Japan). An oligonucleotide OL26 (50-GA-
CAAAGCCGTATCCGAATTCTGGCA-30), designed based on the
N-terminal amino acid sequence of DAD purified from strain AZ11
and the nucleotide sequence assumed as dad in B. cenocepacia J2315
(Genbank accession no. AM747721, locus tag BCAM0037), was
labeled at the 30-terminal with digoxigenin using a DIG-tailing label kit

2,40-Dihydroxyacetophenone Dioxygenase from Burkholderia
569
(Boehringer Mannheim, Mannheim, Germany), and used as a probe for
hybridization. Southern and colony hybridization were done at 56 ꢀC.
Hybridization signals on nylon membranes were detected with a DIG
luminescent detection kit (Boehringer Mannheim) following the
manufacturer’s protocols. All enzymes used in DNA manipulation
were obtained from Takara. Plasmids were isolated from recombinant
E. coli with a QIAprep kit (Qiagen, Hilden, Germany). Extraction of
DNA from agarose gel was done with an Easy Trap (Takara). DNA
digestion, ligation, and transformation were done as described by
Sambrook et al.29)
small amount (0.35 mmol) of DHAP in 4 mL of air-
saturated 50 mM PPB (pH 7.0), until O2 consumption,
monitored with an O2 electrode, stopped. HPLC analysis
of the reaction mixture indicated that 0.35 mmol of
DHAP and 0:34 ꢃ 0:02 mmol of O2 were consumed,
and that 0:33 ꢃ 0:01 mmol of 4-hydroxybenzoate and
0:35 ꢃ 0:04 mmol of formate were formed. Formation
of 4-hydroxybenzoate was also confirmed by silica gel
TLC. These results conformed to the stoichiometric
relationship of the DAD reaction:
Nucleotide sequence. Deletion mutants for DNA sequencing were
prepared with a Kilo-sequence deletion kit (Takara) after digestion
with appropriate restriction enzymes. Both strands were sequenced on
denatured double-stranded DNA templates with a BigDye terminator
cycle sequencing kit (Applied Biosystems, Foster City, CA) and
appropriate primers (usually 25mer). The nucleotide sequence was
determined with an ABI 3100 DNA sequencer (Perkin-Elmer,
Norwalk, CT). The nucleotide sequence data reported in this paper
are available in the DDBJ/EMBL/GenBank databases under accession
no. AB636681.
ðpÞHO-C6H4-COCH2OH þ O2
! ðpÞHO-C6H4-COOꢁ þ HCOOꢁ þ 2Hþ
Purification of DAD from Burkholderia sp. AZ11
The results of enzyme purification from strain AZ11
grown on DHAP are summarized in Table 1A. They
showed about 39-fold purification with a yield of 10%.
The purified enzyme preparation showed a single protein
band at about 23 kDa on SDS–PAGE (Fig. 1). The Mr of
the enzyme was determined to be about 90 kDa by gel
filtration on Superdex 200 (data not shown). The N-
terminal amino acid sequence of the enzyme was
determined to be VDKAVSEFWQNIPAIANPFK by
Edman degradation. This is 55% identical to that of
DAD from Alcaligenes sp. 4HAP,15) and coincides to
the one deduced from the nucleotide sequence of
B. cenocepacia J2315 (Fig. 3). The enzyme showed
high affinity toward DHAP: the apparent Km was
estimated to be about 1.2 mM. Further investigation of
the enzyme was hampered by low yield, and preparation
of the recombinant enzyme was required.
Computational tools. The nucleotide sequence and the deduced
amino acid sequence were analyzed using the DNASIS program
(Hitachi Software Engineering, Yokohama, Japan). Multiple sequence
alignment was done using ClustalW (http://www.ddbj.nig.ac.jp/). The
domain was located with the Pfam domain data base (http://
pfam.sanger.ac.uk/), and hydrophobicity was judged with the Sosui
WWW server (http://bp.nuap.nagoya-u.ac.jp/sosui/sosui submit.html).
Cloning and expression of dad. The genomic DNA of Burkholderia
sp. AZ11 was digested with KpnI. Agarose gel electrophoresis
followed by Southern hybridization with a DIG-OL26 gave a single
hybridization signal on a 8.0-kb DNA fragment. The fragment was
extracted from agarose gel and ligated into Charomid 9–36 (Nippon
Gene, Tokyo), which was previously digested with KpnI. After in vitro
packaging with a Lambda inn (Nippon Gene) and infection of E. coli
XL1-Blue MRA, colony hybridization produced positive colonies. The
plasmid, designated pCKK8.0, was isolated from one of these, and
digested with PstI. A 5.8-kb DNA fragment with a hybridization signal
was isolated and introduced into pUC18 (Takara) to obtain plasmid
pUPP5.8 (Fig. 2). After digestion with KpnI and BamHI, deletion
mutants of pUPP5.8 were prepared to determine the nucleotide
sequence of the PstI 5.8-kb region. One of the deletion mutants,
designated pUDP1.3, contained an entire dad without any other open
reading frames (ORFs) in the 1.3-kb DNA insert, and was used to
express dad in E. coli JM109. Site-directed mutagenesis to convert
initiation codon GTG to ATG was performed with a PrimSTAR
mutagenesis kit (Takara) and PCR primers, forward primer (50-
ACAAACCATGGTCGACAAAGCCGTAT-30) and reverse primer
(50-TCGACCATGGTTTGTCTCCTTGCATG-30) (the nucleotides to
be mutagenized are underlined). Mutagenesis was confirmed by DNA
sequencing with appropriate sequencing primers.
Cloning and expression of dad in E. coli
A PstI 5.8-kb DNA fragment containing dad was
cloned from the genomic DNA into pUC18 to give
plasmid pUPP5.8 (Fig. 2). Deletion mutants of pUPP5.8
were prepared to determine the nucleotide sequence of
the PstI 5,755-bp long-insert DNA. There were four
ORFs with appropriate ribosome-binding sequences in
this region. ORF1 and ORF3 encoded a 291-amino acid
protein (calculated Mr, 32,680 Da) and a 326-amino
acid protein (calculated Mr, 35,384 Da) respectively.
These proteins had the helix-turn-helix motif for bind-
ing to DNA, and appeared to be transcriptional
regulators. ORF2 encoded a 521-amino acid protein
(calculated Mr, 55,772 Da) having a FAD-dependent
Results
Table 1. Purification of DAD from Burkholderia sp. AZ11 (A) and
E. coli JM109 Carrying pUDP1.3m (B)
Isolation and identification of a DHAP degrading
microorganism
Protein Sp. activity Activity Purification Yield
Step
(mg)
(U/mg)
(U)
(-fold)
(%)
Strain AZ11 was isolated from a soil sample by
enrichment culture with DHAP, and was identified as
Burkholderia. DAD was induced by cultivating strain
AZ11 with DHAP as carbon and energy source, but
bacterial growth was poor. The crude extract showed
high activity (0:042 ꢃ 0:009 mmol/min/mg) for DAD.
The addition of 4-hydroxybenzoate and that of glucose
to the culture medium improved bacterial growth, but
not the production of DAD. Strain AZ11 did not grow
on bisphenol A or its intermediary metabolite 4-
hydroxyacetophenone. Crude extract prepared from
DHAP-induced cells was incubated at 25 ꢀC with a
(A)
Crude extract
Superdex 200
Q-Sepharose
Phenyl Sepharose
97.2
34.3
2.92
0.25
0.0375
0.0871
0.423
1.48
3.65
2.99
1.24
0.37
1
2.3
11
39
100
82
34
10
(B)
Crude extract
Heat
371
170
38.6
0.322
0.545
1.58
119
1
100
78
92.7
61.0
42.6
32.9
1.7
4.9
10
17
(NH4)2SO4
Superdex 200
Q-Sepharose
51
36
13.1
6.0
3.25
5.49
28

570
M. ENYA et al.
A
B
Fig. 2. Construction of Recombinant Plasmids.
Fig. 1. SDS–PAGE of the Purification of DAD from Burkholderia
sp. AZ11 (A) and E. coli JM109 Carrying pUDP1.3m (B).
A, lane M, marker proteins; lane 1, crude extract (2 mg); lane 2,
Superdex 200 (1 mg); lane 3, Q Sepharose (0.2 mg); lane 4, Phenyl
Sepharose (0.07 mg). Proteins were visualized by silver staining.
B, lane M, marker proteins; lane 1, crude extract (50 mg); lane 2,
heat treatment (25 mg); lane 3, ammonium sulfate (15 mg); lane 4,
Superdex 200 (3 mg); lane 4, Q-Sepharose (3 mg). Proteins were
stained with Coomassie Brilliant Blue R-250.
The solid line shows vector pUC18, and the hollow line, the insert
DNA of pUPP5.8 (above) and its deletion mutant pUDP1.3 (below).
ORFs 1-3 are shown by white arrows, and ORF4 (dad) by black
arrows. The direction of the lac promoter of pUC18 is shown by
arrows (plac). Abbreviations: E, EcoRI; S, SacI; K, KpnI; Sm, SmaI;
B, BamHI; X, XbaI; Sa, SalI; P, PstI; Sp, SphI; H, HindIII.
Fig. 3. Multiple Sequence Alignment of DAD.
Sequences: Bsp, Burkholderia sp. AZ11(GenBank AB636681); Bce, B. cenocepacia J2315 (RefSeq YP 002232670); Asp, Alcaligenes sp.
4HAP (GenBank AJ133820);15) Pfl, Pseudomonas fluorescens Pf-5 (RefSeq YP 260468); Nar, Novoshingobium aromaticivorans DSM 12444
(RefSeq YP 497143); Rsp, Roseobacter sp. SK209-2-6 (RefSeq ZP 01753062); Hse, Herbaspirillum seropsedicae SmR1 (RefSeq
YP 003774012); Bma, B. mallei ATCC 23344 (RefSeq YP 105997); Bgl, B. glumae BGR1 (RefSeq YP 002908823); and Vsp, Vibrio sp.
Ex25 (RefSeq YP 003287777). Numbers to the right correspond to residue numbers. Cupin motifs are underlined. Positions identical in all
sequences are marked by arrows. Amino acid residues marked by both arrows may be important for metal binding. Conserved residues and
similar residues are indicated by double daggers and daggers respectively.
oxidoreductase domain. ORF4 (dad) was 534-bp long
and encoded a 177-amino acid protein (calculated Mr
20,221 Da).
One of the deletion mutants, designated pUDP1.3, had
an entire dad as sole ORF in a 1,368-bp long insert DNA
(Fig. 2). An appropriate ribosome binding sequence,
GAGA, and initiation codon GTG specify dad. The
nucleotide sequence of hybridization probe OL26
coincided perfectly with the real sequence (nucleotide
was higher in the middle region (residue nos. 74-116)
containing the cupin fold than in the N- and C-terminal
regions. Cupin is a binding site of metal ions in a variety
of proteins, in which three His residues and a Glu
residue are possible metal ligands.30,31) In DAD, they
appeared to be His77, His79, His115, and Glu96 or
Glu109 (residue number based on the amino acid
sequence of DAD from Burkholderia sp. AZ11),
because of their complete conservation in DAD from
all bacterial strains examined. Burkholderia sp. AZ11
and B. cenocepacia J2315 have GTG as the initiation
codon, whereas other bacterial strains have ATG.
Although the enzyme activity (0:086 ꢃ 0:004 mmol/
min/mg) of the crude extract from E. coli JM109
harboring pUDP1.3 was about 2 times higher than that
from Burkholderia sp. AZ11 grown on DHAP, the extent
of gene expression of dad was not sufficient. The
initiation codon of dad of Burkholderia sp. AZ11 is
GTG, which occurs less frequently than ATG in
E. coli.32) Hence the initiation codon was converted
from GTG to ATG by site-directed mutagenesis. The
resulting plasmid, designated pUDP1.3m, brought about
high-level expression of DAD in recombinant E. coli.
The specific activity (0:330 ꢃ 0:008 mmol/min/mg) of
nos. 383–408).
A
possible promoter sequence
(TTGAAT for the ꢁ35 sequence and TCCTGT for the
ꢁ10 sequence) was found in the region (nucleotide nos.
112–139) upstream of dad. The first 20 amino acid
sequence deduced from the nucleotide sequence coin-
cided with that obtained by Edman degradation of the
purified enzyme, except that the N-terminal residue was
deleted. The amino acid sequence of DAD contained
a typical ꢀ-barrel fold of the cupin superfamily at
positions 74–131 (Fig. 3).30) Multiple sequence align-
ment of DAD from 10 bacterial strains indicated that
DAD from Burkholderia sp. AZ11 was highly homol-
ogous to that from B. cenocepacia J2315 (98.9%
identity), and less homologous (84% identity) to that
from Alcaligenes sp. 4HAP. The degree of homology

2,40-Dihydroxyacetophenone Dioxygenase from Burkholderia
571
the crude extract was about 8 times higher than that of
Burkholderia sp. AZ11.
Purification of recombinant DAD
The recombinant enzyme was briefly purified from
the crude extract of E. coli JM109 harboring
pUDP1.3m. Table 1B summarizes the purification pro-
cedures. It shows 17-fold purification with 28% recovery
of the enzyme. The specific activity (5.49 mmol/min/
mg) of the purified enzyme was about 2 times higher
than that (2.85 mmol/min/mg) of the recombinant
enzyme as prepared by Hopper and Kaderbhai.15) The
activity was also higher than that of the enzyme purified
from Burkholderia sp. AZ11 (Table 1A), suggesting that
the latter enzyme preparation was partially inactivated
during the purification procedure. The purified enzyme
was stable for at least 1 month when stored at ꢁ25 ꢀC.
Repeated freezing and thawing caused a loss of activity.
On SDS–PAGE, the enzyme showed a single protein
band corresponding to Mr of 23 ꢃ 1 kDa (Fig. 1B). Its
Mr was estimated to be 90 ꢃ 3 kDa by gel filtration on
Superdex 200 (data not shown), suggesting that it is a
tetramer. The N-terminal amino acid residue of the
purified enzyme was identified as Val with dansyl
chloride, indicating that the N-terminal Met residue was
removed. Thus the recombinant enzyme was virtually
identical to the native enzyme in amino acid sequence.
The non-heme iron content of the purified enzyme was
estimated to be 1:69 ꢃ 0:04 mol/mol of enzyme by
atomic absorption spectrometry, and 1:63 ꢃ 0:04 mol/
mol of enzyme by Fish’s colorimetric method, while the
copper content was less than 0.01 mol/mol of enzyme.
The enzyme had no significant amount of inorganic
sulfide (less than 0.03 mol/mol of enzyme).
Fig. 4. Absorption Spectrum of Purified DAD.
The concentrations of the enzyme were 1.07 mg/mL (curve 1) and
6.40 mg/mL (curve 2) in 50 mM PPB (pH 7.0).
hydroxybenzoate during the reaction were monitored
(Fig. 5B and C). On mixing the enzyme with DHAP,
ꢂA400 increased rapidly, remained at the increased level,
and then decreased with exhaustion of DHAP. Most of
the dissolved O2 in the reaction mixture was rapidly
consumed immediately after the start of the reaction,
concomitantly with the consumption of DHAP and the
formation of 4-hydroxybenzoate. Thus the dissolution
rate of O2 into the mixture controlled the enzyme
reaction, and the reaction progressed slowly under the
limited O2 concentration until DHAP was exhausted.
These results suggest that the enzyme-DHAP complex
reacts with O2.
Substrate specificity
O2 consumption activity toward various compounds
(each 0.35 mM) was measured at 25 ꢀC in air-saturated
50 mM PPB (pH 7.0). Among the various acetophenone
derivatives examined, only 2-hydroxyacetophenone
(2-hydroxy-1-phenylethanone) showed weak enzyme
activity (4:4 ꢃ 0:4% as compared with DHAP). The
reaction product was extracted and identified as ben-
zoate by HPLC and silica gel TLC. No significant O2
consumption was observed for the following com-
pounds: 2-(4-hydroxyphenyl)ethanol, 4-hydroxypheny-
lacetone, 2-phenylethanol, 40-hydroxypropiophenone,
acetophenone and its derivatives including 40-hydroxy-,
30-hydroxy-, 20-hydroxy-, 40-methoxy-, 40-amino-,
40-chloro-, and 40-fluoroacetophenone.
Absorption spectrum of DAD
The absorption spectrum of the enzyme was measured
with diluted enzyme solution, since the dark-gray
enzyme protein precipitated from a concentrated en-
zyme solution. The enzyme showed a simple UV
spectrum with maximum at 280 nm and a shoulder at
290 nm (Fig. 4). The A280=A260 ratio was 1.68. It showed
a broad visible spectrum, with two maxima at 350 nm
and 560 nm, possibly due to Fe3þ bound to the enzyme
protein. The molar extinction coefficients at 280, 350,
and 560 nm were calculated to be 115, 5.8, and
1.4 mMꢁ1 cmꢁ1 respectively, based on a Mr of 90 kDa.
No significant absorption indicating the presence of
heme and flavins was detected. The spectral feature in
the visible region was similar to that of Fe3þ-containing
catechol 1,2-dioxygenase (EC 1.13.11.1), which has a
red color and shows a shoulder at about 320 nm and a
broad peak at around 440 nm.33,34) The repeated freezing
and thawing inactivated the enzyme with a decrease in
absorption at longer than 350 nm, probably owing to a
loss of iron.
Kinetic constants
Enzyme activity was measured spectrophotometri-
cally at various DHAP concentrations (1.25 mM–
14.3 mM) with an air-saturated mixture (0.25 mM O2).
The enzyme showed a hyperbolic saturation curve with
weak substrate inhibition at concentrations higher than
0.03 mM (data not shown). The apparent Km for DHAP
and the apparent Vmax were estimated by double
reciprocal plot to be 1:60 ꢃ 0:03 mM and 6:28 ꢃ
0:06 mmol/min/mg respectively.
To determine the spectral change on binding with
DHAP, the spectra of the enzyme and DHAP were
measured separately, and were subtracted from the
spectrum obtained by incubation of the enzyme with
DHAP under N2 (Fig. 5A). The absorption at around
400 nm increased, possibly due to binding of DHAP to
Fe3þ in the enzyme. A similar experiment was carried
out in air, and the concentrations of DHAP and 4-
Effects of pH, temperature, and inhibitors
The effect of pH on enzyme activity was examined
by measuring O2 consumption in various buffers.
Potassium phosphate and 2-morpholinoethanesulfonic
acid (MES) buffers were effective, and the enzyme
showed a maximum activity at pH 6–7 (data not shown).
It was moderately thermostable, and retained about 80%
of its activity after incubation at 70 ꢀC for 20 min (data

572
M. ENYA et al.
A
B
C
Fig. 5. Spectral Changes in DAD in the Enzyme Reaction.
All spectra were recorded at 600 nm/min. A, Spectral change in DAD under anaerobic conditions. The spectrum under N2 was measured with
a Thunberg-type cuvette. The enzyme (2.35 mg) was incubated at 25 ꢀC with 1 mM DHAP in 2.8 mL of 50 mM PPB (pH 7.0). The absorption
spectra of the enzyme (curve 2), DHAP (curve 3), and the enzyme plus DHAP (curve 1) were measured separately against the buffer as
reference. The inset shows a difference spectrum, which was calculated by subtracting curves 2 and 3 from curve 1. B, Spectral change in DAD
under aerobic conditions. The enzyme was incubated with DHAP under the same conditions as in A, except that an air-saturated reaction mixture
was used and mixed moderately. The difference spectra were calculated as above. Curves 1–6 were measured at 0.5, 30, 60, 90, 120, and 150 min
respectively after the start of the reaction. C, Time course of the enzyme reaction. During the reaction in B, samples (30 mL) were withdrawn at
the indicated times, and mixed with 0.27 mL of 10 mM HCl to stop the reaction. The mixture was neutralized with NaOH, deproteinized by
centrifugation, and analyzed by HPLC to determine the concentrations of DHAP ( ) and 4-hydroxybenzoate ( ). The values of ꢂA400
obtained in B are also shown.
( )
Table 2. Effects of Various Reagents on Enzyme Activity
phenanthroline, tiron (disodium 4,5-dihydroxybenzene-
1,3-disulfonate), and 8-hydroxyquinoline, the enzyme
previously treated with 0.4% Lubrol PX was inhibited
by 2,20-dipyridine and 8-hydroxyquinoline. It is prob-
able that the integrity of the protein structure was
partially destroyed by the detergent so as to allow the
chelators to react with iron in the enzyme. Diethylpy-
rocarbonate and p-chloromercuribenzoate weakly inhib-
ited the enzyme, while 2-mercaptoethanol activated it
slightly. KCN, NaN3, ascorbate (each 1 mM), and
ethanol (10%, v/v) were almost entirely ineffective
(data not shown).
Activity (%)a
Reagent (1 mM)
CuCl2
HgCl2
Ab
Bc
9
13
5
15
73
58
60
98
102
92
92
103
10
53
61
10
25
58
93
CoCl2
NiCl2
54
79
ZnCl2
MgCl2
77
97
CaCl2
FeSO4
FeCl3
98
120
104
95
EDTA
2,20-Dipyridine
1,10-Phenanthroline
Tiron
87
94
Discussion
102
98
8-Hydroxyquinoline
Diethylpyrocarbonate
p-Chloromercuribenzoate
2-Mercaptoethanol
Alcaligenes sp. 4HAP degrades 4-hydroxyacetophe-
none via DHAP, and the DAD of strain 4HAP is
involved in 4-hydroxyacetophenone degradation.13) On
the other hand, the DHAP-degrading bacterium Bur-
kholderia sp. AZ11 did not grow on 4-hydroxyaceto-
phenone. Moreover, no functional relationship to 4-
hydroxyacetophenone metabolism was found in the
genes upstream and downstream of dad of strain AZ11
and B. cenocepacea J2315, which appear to be closely
related. Hence the DAD of strain AZ11 appears to
participate in a metabolic pathway other than the 4-
hydroxyacetophenone degradation pathway. Some
plants, such as Carthamus and Rhodiola, contain
aromatic glycosides with DHAP as an aglycon.35,36) It
is likely that DAD in most soil bacteria is engaged in the
degradation of aromatic glycosides derived from plants.
Dad catalyzes oxygenative cleavage of the hydro-
xyacetyl group of DHAP. Both carbonyl and hydroxy
parts in the hydroxyacetyl group are essential for
42
92
113
aThe activity (5.1 mmol/min/mg in A, and 3.3 mmol/min/mg in B) obtained
without reagent was taken to be 100%. Values are means of duplicate
analysis.
bLubrol-untreated enzyme.
cLubrol-treated enzyme.
not shown). Although the O2 concentration in the
reaction mixture decreased with the rise in temperature,
the enzyme activity increased as the temperature
increased up to 65 ꢀC (data not shown). CuCl2 and
HgCl2 strongly inhibited the enzyme (Table 2). CoCl2,
NiCl2, and ZnCl2 weakly inhibited it, while FeSO4
weakly activated it. Although the enzyme prepared from
the recombinant E. coli was hardly affected by various
metal chelators, including EDTA, 2,20-dipyridine, 1,10-

2,40-Dihydroxyacetophenone Dioxygenase from Burkholderia
573
enzyme activity, since neither 2-phenylethanol nor
40-hydroxyacetophenone served as substrate. The 40-
hydroxy group of DHAP is also important, but not
indispensable for enzyme activity. Since the hydroxy-
acetyl group can tautomerize to the cis-1,2-ethenediol
group, the DAD reaction appears to be analogous to the
intradiol cleavage of a catechol ring catalyzed by
intradiol dioxygenases. Intradiol dioxygenases have a
characteristic color due to Fe3þ at the active site, while
Fe2þ-containing extradiol dioxygenases such as catechol
2,3-dioxygenase (EC 1.13.11.2) are colorless.37) The
recombinant DAD purified from E. coli has a dark gray
color, and shows a similar absorption spectrum to that of
catechol 1,2-dioxygenase.33,34) Therefore, although the
electronic state of the iron must be determined more
various bacterial strains revealed three His residues
(His77, 79, and 115) and a Glu residue (Glu96 or 109) as
ligands. Prediction of the protein secondary structure by
the Chou-Fasman method has suggested that Glu109 is
located in a turn structure, not in a ꢀ-strand, and that
Glu96 is adequate as a ligand for the active-site Fe ion.40)
In conclusion, we have established an easy procedure
for the purification of DAD from Burkholderia sp.
AZ11, and we found some of the molecular and catalytic
properties of DAD. The purified preparation of DAD
should be useful in future investigations of its three-
dimensional structure and catalytic mechanism.
Acknowledgment
definitely by ESR at 4K, DAD appears to be a Fe3þ
containing metalloprotein.
-
We would like to thank Ms. Keiko Takase of the
Institute of Waste Water Treatment at Gifu University
for atomic absorption analysis.
The analogy with catechol 1,2-dioxygenase suggests a
possible catalytic mechanism of DAD. The formation of
a coordination complex between the enzyme-Fe3þ and
the cis-1,2-ethenediol form of DHAP can be inferred
from the difference spectrum of the enzyme-DHAP
complex obtained under N2. Enzyme-bound DHAP
might donate 1e to Fe3þ to form a radical at C1, and
O2 might attack that radical to give a peroxy-radical
according to a mechanism similar to that of catechol
1,2-dioxygenase.38) Subsequently, migration of the 4-
hydroxybenzoyl group to form acid anhydride followed
by hydrolytic cleavage yields 4-hydroxybenzoate and
formate.
References
1) Timms BG, Howdeshell KL, Barton L, Bradley S, Richter CA,
and vom Saal FS, Proc. Natl. Acad. Sci. USA, 102, 7014–7019
(2005).
2) Gaido KW, Leonard LS, Lovell S, Gould JC, Babai D, Portier
CJ, and McDonnell DP, Toxicol. Appl. Pharmacol., 143, 205–
212 (1997).
3) Lobos JH, Leib TK, and Su T-M, Appl. Environ. Microbiol., 58,
1823–1831 (1992).
4) Spivack J, Leib TK, and Lobos JH, J. Biol. Chem., 269, 7323–
7329 (1994).
5) Sasaki M, Maki J, Oshiman K, Matsumura Y, and Tsuchido T,
Biodegradation, 16, 449–459 (2005).
The purified DAD had about 1.63–1.69 mol of non-
heme iron. The iron might be bound at the inner part
of the enzyme molecule, and does not react with
metal chelators unless the protein structure is partially
perturbed by a detergent such as Lubrol PX (Table 2).
Repeated freezing and thawing resulted in a reduction of
the specific activity of the enzyme concomitant with a
decrease in absorption in the visible region, possibly due
to loss of the bound iron. It is plausible to consider that
the enzyme loses some Fe3þ during enzyme purification
and storage. Consistently with this, the purified enzyme
was slightly activated by FeSO4 (Table 2). FeCl3 was
almost entirely ineffective, suggesting that Fe2þ binds to
the enzyme protein and is subsequently oxidized to Fe3þ
to yield an active enzyme. Based on these consider-
ations, the enzyme protein composed of a homotetramer
was assumed to have 2 mol of Fe3þ. Possibly, a Fe3þ ion
is located and functions around the interface of two
subunits, as found for catechol 1,2-dioxygenase, a dimer
containing 1 mol of Fe3þ per mol of enzyme.39) The
DAD from Alcaligenes sp. 4HAP, whose specific
activity is about half of the DAD purified from E. coli
JM109 carrying pUDP1.3m, has been reported to be
colorless and to consist of a homotetramer (Mr ¼
81:6{87 kDa) containing 1 mol of non-heme iron.15)
DAD is moderately thermostable. This made easy
purification of the recombinant DAD possible
(Table 1B). This thermostability and high affinity to
DHAP as well as narrow substrate specificity suggest
that DAD is applicable in the enzymatic determination
of DHAP. The thermostability of DAD might be due
to the cupin fold in its structure. The cupin fold is
a conserved ꢀ-barrel containing four metal-binding
ligands.30,31) Multiple sequence alignment of DAD from
6) Sasaki M, Akahira A, Oshiman K, Tsuchido T, and Matsumura
Y, Appl. Environ. Microbiol., 71, 8024–8030 (2005).
7) Sakai K, Yamanaka H, Moriyoshi K, Ohmoto T, and Ohe T,
Biosci. Biotechnol. Biochem., 71, 51–57 (2007).
8) Tanner A and Hopper DJ, J. Bacteriol., 182, 6565–6569 (2000).
9) Kamerbeek NM, Moonen MJH, van der Ven JGM, van Berkel
WJH, Fraaije MW, and Janssen DB, Eur. J. Biochem., 268,
2547–2557 (2001).
10) Moonen MJH, Kamerbeek NM, Whestphal AH, Boeren SA,
Janssen DB, Fraaije MW, and van Berkel WJH, J. Bacteriol.,
190, 5190–5198 (2008).
11) Moonen MJH, Synowsky SA, van den Berg WAM, Westphal
AH, Heck AJR, van den Hauvel RHH, Fraaije MW, and van
Berkel WJH, J. Bacteriol., 190, 5199–5209 (2008).
12) Hopper DJ and Elmorsi EA, Biochem. J., 218, 269–272
(1984).
13) Hopper DJ, Jones HG, Elmorsi EA, and Rhodes-Roberts ME,
J. Gen. Microbiol., 131, 1807–1814 (1985).
14) Hopper DJ, Biochem. J., 239, 469–472 (1986).
15) Hopper DJ and Kaderbhai MA, Biochem. J., 344, 397–402
(1999).
16) Bowyer A, University of Southampton, Faculty of Medicine,
Health and Sciences, School of Biological Science, PhD Thesis,
pp. 104–117 (2008).
17) Robertson A and Robinson R, J. Chem. Soc., 1460–1472 (1928).
18) Shinoda Y, Kato N, and Morita N, Shimadzu Hyoron (in
Japanese), 57, 121–132 (2000).
19) Jerpseth B, Greener A, Short JM, Viola J, and Kretz PL, Strat.
Mol. Biol., 5, 81–83 (1992).
20) Yanish-Perron C, Vieira J, and Messing J, Gene, 33, 103–119
(1985).
21) Lowry OH, Rosebrough NJ, Farr AL, and Randall RJ, J. Biol.
Chem., 193, 265–275 (1951).
22) Bennett TP, Nature, 18, 1131–1132 (1967).
23) Laemmli UK, Nature, 227, 680–685 (1970).
24) Gray WR, Methods Enzymol., 25, 121–138 (1972).
25) Fish WW, Methods Enzymol., 158, 357–364 (1988).
26) Fogo JK and Popowsky M, Anal. Chem., 21, 732–734 (1949).

574
M. ENYA et al.
27) Lovenberg W, Buchanan BB, and Rabinowitz JC, J. Biol.
Chem., 238, 3899–3913 (1963).
35) Zhou Y-Z, Chen H, Qiao L, Lu X, Hua H-M, and Pei Y-H, Helv.
Chim. Acta, 91, 1277–1285 (2008).
28) Smith MG, Methods Enzymol., 12A, 545–550 (1967).
29) Sambrook J, Fritsch EF, and Maniatis T, ‘‘Molecular Cloning, a
Laboratory Manual’’ 2nd ed., Cold Spring Harbor Laboratory,
Cold Spring Habor (1989).
36) Peng Y, Luo J, Lu Q, Chen X, Xie Y, Chen L, Yang W, and Du
S, J. Pharm. Biomed. Anal., 49, 828–832 (2009).
37) Kobayashi T, Ishida T, Horiike K, Takahara Y, Numao N,
Nakazawa A, Nakazawa T, and Nozaki M, J. Biochem., 117,
614–622 (1995).
30) Khuri S, Bakker FT, and Dunwell JM, Mol. Biol. Evol., 18, 593–
605 (2001).
38) Xin M and Bugg TDH, J. Am. Chem. Soc., 130, 10422–10430
(2008).
31) Dunwell JM, Purvis A, and Khuri S, Phytochem., 65, 7–17
(2004).
39) Earhart CA, Vetting MW, Gosu R, Michaud-Soret I, Que Jr L,
and Ohlendorf DH, Biochem. Biophys. Res. Commun., 338,
198–205 (2005).
32) Kozak M, Microbiol. Rev., 47, 1–45 (1983).
33) Nakai C, Nakazawa T, and Nozaki M, Arch. Biochem. Biophys.,
267, 701–713 (1988).
40) Chou PY and Fasman GD, Annu. Rev. Biochem., 47, 251–276
(1978).
34) Nakai C, Horiike K, Kuramitsu S, Kagamiyama H, and Nozaki
M, J. Biol. Chem., 265, 660–665 (1990).