Functional annotation and characterization of 3-hydroxybenzoate 6-hydroxylase from Rhodococcus jostii RHA1

Article abstract of DOI:10.1016/j.bbapap.2011.12.003

The genome of Rhodococcus jostii RHA1 contains an unusually large number of oxygenase encoding genes. Many of these genes have yet an unknown function, implying that a notable part of the biochemical and catabolic biodiversity of this Gram-positive soil actinomycete is still elusive. Here we present a multiple sequence alignment and phylogenetic analysis of putative R. jostii RHA1 flavoprotein hydroxylases. Out of 18 candidate sequences, three hydroxylases are absent in other available Rhodococcus genomes. In addition, we report the biochemical characterization of 3-hydroxybenzoate 6-hydroxylase (3HB6H), a gentisate-producing enzyme originally mis-annotated as salicylate hydroxylase. R. jostii RHA1 3HB6H expressed in Escherichia coli is a homodimer with each 47 kDa subunit containing a non-covalently bound FAD cofactor. The enzyme has a pH optimum around pH 8.3 and prefers NADH as external electron donor. 3HB6H is active with a series of 3-hydroxybenzoate analogues, bearing substituents in ortho- or meta-position of the aromatic ring. Gentisate, the physiological product, is a non-substrate effector of 3HB6H. This compound is not hydroxylated but strongly stimulates the NADH oxidase activity of the enzyme.

Full text of DOI:10.1016/j.bbapap.2011.12.003

Biochimica et Biophysica Acta 1824 (2012) 433–442
Contents lists available at SciVerse ScienceDirect
Biochimica et Biophysica Acta
journal homepage: www.elsevier.com/locate/bbapap
Functional annotation and characterization of 3-hydroxybenzoate 6-hydroxylase
from Rhodococcus jostii RHA1
⁎
Stefania Montersino, Willem J.H. van Berkel
Laboratory of Biochemistry, Wageningen University, Dreijenlaan 3, 6703 HA Wageningen, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 October 2011
Received in revised form 9 December 2011
Accepted 14 December 2011
Available online 22 December 2011
The genome of Rhodococcus jostii RHA1 contains an unusually large number of oxygenase encoding genes.
Many of these genes have yet an unknown function, implying that a notable part of the biochemical and
catabolic biodiversity of this Gram-positive soil actinomycete is still elusive. Here we present a multiple
sequence alignment and phylogenetic analysis of putative R. jostii RHA1 flavoprotein hydroxylases. Out of
18 candidate sequences, three hydroxylases are absent in other available Rhodococcus genomes. In addition,
we report the biochemical characterization of 3-hydroxybenzoate 6-hydroxylase (3HB6H), a gentisate-
producing enzyme originally mis-annotated as salicylate hydroxylase. R. jostii RHA1 3HB6H expressed in
Escherichia coli is a homodimer with each 47 kDa subunit containing a non-covalently bound FAD cofactor.
The enzyme has a pH optimum around pH 8.3 and prefers NADH as external electron donor. 3HB6H is
active with a series of 3-hydroxybenzoate analogues, bearing substituents in ortho- or meta-position of the
aromatic ring. Gentisate, the physiological product, is a non-substrate effector of 3HB6H. This compound is
not hydroxylated but strongly stimulates the NADH oxidase activity of the enzyme.
Keywords:
Fingerprint
Flavoprotein
Functional annotation
3-hydroxybenzoate 6-hydroxylase
Monooxygenase
Rhodococcus jostii RHA1
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
without a particular function, we performed a multiple sequence
alignment and phylogenetic analysis with a large set of known flavo-
Rhodococcus jostii RHA1 is a Gram-positive soil actinomycete able
to degrade a wide range of organic compounds [1–4]. It possesses one
of the largest bacterial genomes ever sequenced and encodes an
exceptional amount of oxygenases (203 putative genes), in particular
flavoprotein monooxygenases (88 putative genes) [5].
protein hydroxylases. Among the newly assigned functions, we
present the biochemical characterization of 3-hydroxybenzoate 6-
hydroxylase (3HB6H), a flavoprotein involved in the gentisate (2,5-
dihydroxybenzoate) degradation pathway [20, 21]. Notably, the
gene encoding for this flavoenzyme is mis-annotated as a salicylate
(2-hydroxybenzoate) hydroxylase. R. jostii RHA1 3HB6H expressed
in Escherichia coli is specific for 3-hydroxybenzoate derivatives and
does not interact with salicylate.
Flavoprotein monooxygenases (EC 1.14.13.x) perform a wide range
of regio- and enantioselective reactions and can be divided in six
subclasses [6]. Subclass A comprises a family of single-component
flavoprotein hydroxylases, which are crucially involved in microbial
degradation of natural and anthropogenic aromatics [7–9], polyketide
antibiotic biosynthesis [10–12], and antibiotic resistance [13–15].
Flavoprotein hydroxylases can be identified on the basis of three fin-
gerprint sequences, first defined in 4-hydroxybenzoate 3-hydroxylase
(PHBH) (Fig. 1) [16]. The GxGxxG sequence motif maps the ADP moiety
of FAD [17], while the GD consensus motif represents the residues
that interact with the riboflavin moiety of FAD [18]. Both of these FAD
fingerprints are common for many flavoproteins [19]. The third, DG
consensus motif, is specific for subclass A enzymes and serves a dual
role of recognition of both FAD and NADPH [16].
2. Materials and methods
2.1. Chemicals
DNAseI was from Boehringer Mannheim GmbH (Mannheim,
Germany). Restriction endonucleases and dNTPs were purchased
from Invitrogen (Carlsbad, CA, USA). Phusion High Fidelity DNA
polymerase was from Finnzymes (Espoo, Finland). In-Fusion PCR
cloning System was purchased from Clontech (Mountain View, CA,
USA). Oligonucleotides were synthesised by Eurogentec (Liege,
Belgium). E. coli TOP10 was from Invitrogen (Carlsbad, CA, USA).
The pBAD/Myc-His (Nde) expression vector was kindly provided by
Prof. M.W. Fraaije (University of Groningen).
Here we used the above mentioned fingerprints to detect putative
flavoprotein hydroxylase sequences in the R. jostii RHA1 genome
(Fig. 1; Table 1). Since most of the retrieved sequences are annotated
Nickel nitrilotriacetic acid (Ni-NTA) agarose was purchased from
Qiagen (Valencia, CA, USA) and Bio-Gel P-6DG was from Bio-Rad
(Hercules, CA, USA). HiLoad 26/10 Q-Sepharose HP, Superdex 200
HR10/30, low molecular weight protein marker, prestained kaleidoscope
⁎
Corresponding author. Tel.: +31 317 482861; fax: +31 317 484801.
E-mail address: willem.vanberkel@wur.nl (W.J.H. van Berkel).
1570-9639/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.bbapap.2011.12.003

434
S. Montersino, W.J.H. van Berkel / Biochimica et Biophysica Acta 1824 (2012) 433–442
Fig. 1. Sequence comparison of putative flavoprotein hydroxylases from Rhodococcus jostii RHA1. Upper panel: Schematic representation of the primary structure of PHBH from P. fluorescens
(UniProt ID: P00438) with the FAD‐binding domain in light gray and the substrate binding domain in dark gray. Middle panel: Schematic representation of the flavoprotein hydroxylase
fingerprints of PHBH from P. fluorescences used in this study. Bold characters represent residues used for fingerprint acronyms. Lower panel: Sequence alignment of flavoprotein hydroxylase
fingerprint regions among putative flavoprotein hydroxylases from R. jostii RHA1. Identical residues are shaded in black, similar residues are shaded in gray.
protein standards, and catalase (232 kDa), aldolase (158 kDa), BSA
(68 kDa) and ovalbumin (43 kDa) were obtained from Pharmacia
Biotech (Uppsala, Sweden).
(pBAD-3HB6H-His6) was verified by automated sequencing of both
strands and electroporated to E. coli TOP10 cells for recombinant
expression.
Aromatic compounds were purchased from Sigma-Aldrich (St Louis,
MO, USA) and Acros Organics (New Jersey, US). Catalase, FAD, FMN,
riboflavin and arabinose were from Sigma-Aldrich (St Louis, MO,
USA). Pefabloc SC was obtained from Roche Diagnostics GmbH
(Mannheim, Germany). All other chemicals were from commercial
sources and of the purest grade available.
The Åkta Explorer FPLC system (Pharmacia Biotech) was used for
protein chromatography. For enzyme production, E. coli TOP10 cells,
harbouring a pBAD-3HB6H plasmid, were grown in TB medium
supplemented with 100 μg mL−1 ampicillin until an optical density
(OD600 nm) of 0.8 was reached. Expression was induced by the addition
of 0.02% (w/v) arabinose and the incubation was continued for 16 h at
37 °C. Cells (36 g wet weight) were harvested by centrifugation, resus-
pended in 120 mL of 20 mM potassium phosphate, 300 mM NaCl (pH
7.4), containing 1 mM Pefabloc SC, 1 mg DNAse, 100 μM MgCl2 and sub-
sequently passed twice through a precooled French Pressure cell (SLM
Aminco, SLM Instruments, Urbana, IL, USA) at 16,000 psi. The resulting
homogenate was centrifuged at 25,000 g for 45 min at 4 °C to remove
cell debris, and the supernatant was applied onto a Ni-NTA agarose
column (13×1.6 cm) equilibrated with 20 mM sodium phosphate,
300 mM NaCl (pH 7.4). The column was washed with two volumes of
equilibration buffer. The enzyme was eluted with 300 mM imidazole in
equilibration buffer. The active pool, containing an added excess of free
FAD, was desalted using a Biogel column (14×2.6 cm), running in
50 mM BisTris–HCl, 0.1 mM EDTA (pH 7.2). The desalted protein was
loaded onto a HiLoad 26/10 Q-Sepharose HP column equilibrated with
the same buffer. After washing with two column volumes of starting
buffer, the protein was eluted with a linear gradient of NaCl (0–1 M) in
the same buffer. Active fractions were pooled and concentrated to
8 mg/mL by using Amicon filters (30 kDa cutoff) and dialysed at 4 °C
against 50 mM BisTris–HCl (pH 7.2). Purified 3HB6H was frozen in liquid
nitrogen and stored at −80 °C.
2.2. Sequence analysis
The genome of R. jostii RHA1 was analysed for the presence of
flavoprotein hydroxylases at the European Bioinformatic Institute
(www.ebi.ac.uk). FASTA analysis (www.ebi.ac.uk/Tools/sss/fasta) was per-
formed to determine protein sequence homology. Multiple sequence
alignments were made using CLUSTALW [22]. DNA cluster comparison and
database searches were carried out using Nucleotide and Protein re-
sources from the National Center for Biotechnology Information (www.
ncbi.nlm.nih.gov) and UniProt Database (www.uniprot.org). Phylogenet-
ic analysis was performed using FigTree (tree.bio.ed.ac.uk).
2.3. Cloning, expression and purification of 3-hydroxybenzoate
6-hydroxylase in E. coli
A 1.2 kb DNA fragment encoding a putative salicylate hydroxylase
(Gene ID: 4218663) was PCR amplified from R. jostii RHA1 genomic
DNA, using the oligonucleotides SM_1869fwd (5′AGGAGGAAT-
TACATATGTCGAATCTGCAGGACGCAC3′) and SM_1869rev (5′GTTCG-
GGCCCAAGCTTTGACGCGCGATCGGACG3′), introducing NdeI and
HindIII restriction sites (underlined), respectively and containing the
start codon (bold). The amplified fragment was cloned into the
pBAD/Myc-His expression vector containing a C-terminal His6-tag
by using the In-Fusion PCR Cloning System. The resulting construct
2.4. Protein analysis
SDS/PAGE was performed using 12.5% acrylamide slab gels essen-
tially as described by Laemmli [23]. Proteins were stained using

S. Montersino, W.J.H. van Berkel / Biochimica et Biophysica Acta 1824 (2012) 433–442
435
Table 1
Flavoprotein hydroxylases from R. jostii RHA1.
Locus name
TrEMBL
Annotated name
Putative function from NJ-treea (EC number)
Presence in other Rhodococcus sp
(percentage identity).
RHA1_ro00482
RHA1_ro00520
Q0SJG8
Q0SJD0
Monooxygenase/aromatic ring hydroxylase
3-(2-hydroxyphenyl) propionate monooxygenase
Acting on activated substrate
3-(2-hydroxyphenyl) propionate hydroxylase
(EC 1.14.13.x)
R. opacus B4 (74)
R. aetherivorans (90)
RHA1_ro00538
RHA1_ro01845
Q0SJB2
Aromatic ring hydroxylase
R. opacus B4 (96)
R. erythropolis TA421 (78)
R. opacus B4 (93)
Q0SFN0
Probable aromatic ring hydroxylase
R. equi ATCC 33707 (72)
R. equi 103S (72)
R. erythropolis SK121 (71)
R. erythropolis PR4 (70)
R. opacus B4 (40)
R. opacus B4 (96)
Rhodococcus NCIMB 12038 (93)
R. opacus B4 (90)
R. opacus B4 (97)
R. equi ATCC 33707 (88)
R. equi 103S (87)
RHA1_ro01860
RHA1_ro01869
Q0SFL5
Q0SFK6
Probable 3-(3-hydroxylphenyl) propionate hydroxylase
Probable salicylate monooxygenase
3-hydroxybenzoate 6-hydroxylase
(EC 1.14.13.24)
RHA1_ro01939
RHA1_ro02539
Q0SFD6
Q0SPD2
Pentachlorophenol monooxygenase
4-hydroxybenzoate 3-monooxygenase
4-hydroxybenzoate 3-hydroxylase
(EC 1.14.13.2)
R. erythropolis SK121 (89)
R. erythropolis PR4 (87)
R. opacus B4 (92)
R. opacus B4 (90)
R. opacus B4 (95)
RHA1_ro03040
RHA1_ro03540
RHA1_ro03714
Q0SC93
Q0SAU3
Q0SAC2
Probable aromatic ring hydroxylase
Pentachlorophenol monooxygenase
Possible aromatic ring hydroxylase
R. equi ATCC 33707 (60)
R. equi 103S (60)
RHA1_ro03910
RHA1_ro03981
Q0S9S6
Q0S9K8
FAD-binding monooxygenase
Probable aromatic ring monooxygenase
6-hydroxynicotinate 3-monooxygenase
(EC 1.14.13.114)
R. opacus B4 (93)
RHA1_ro04910
RHA1_ro04915
Q0S6Z5
Q0S6Z0
Monooxygenase
Monooxygenase
R. opacus B4 (94)
R. erythropolis SK121 (63)
R. erythropolis PR4 (63)
R. opacus B4 (73)
RHA1_ro07160
Q0S0L2
Probable FAD-dependent monooxygenase
Rifampicin monooxygenase (1.14.13.x)
R. equi ATCC 33707 (73)
R. equi 103S (73)
R. erythropolis SK121 (69)
R. erythropolis PR4 (69)
Rhodococcus NCIMB 9874 (80)
R. aetherivorans (59)
RHA1_ro10313
RHA1_ro11171
Q0RW34
Q0RV68
Probable 2,4-dichlorophenol 6-monooxygenase
Possible monooxygenase/hydrolase
3-(2-hydroxyphenyl) propionate hydroxylase
(EC 1.14.13.x)
a
Neighbour joining tree (Fig. 2).
Coomassie Brilliant Blue R-250. Total protein concentrations were
estimated using the BCA protein kit from Thermo Scientific Pierce
with BSA as standard. Analytical gel filtration to investigate the
hydrodynamic properties of 3HB6H was performed on a Superdex
200 HR 10/30 column running in 50 mM potassium phosphate,
150 mM KCl (pH 7.4). Desalting or buffer exchange of small aliquots
of enzyme was performed with Bio-Gel P-6DG columns and Amicon
Ultra-0.5 filters (30 kDa cut off) (Millipore).
array detector. Reaction products were separated with a 4.0×60 mm
C18 reverse-phase column (Spherisorb, ODS 2, Pharmacia). Reaction
mixtures contained 2 mM substrate, 5 mM of NADH and 2 μM 3HB6H
in 1.5 mL air saturated 20 mM Tris–SO4 pH 8.0. At the end of the
reaction, 10 kDa spin filters were used to separate the enzyme from
the reaction mixture. HPLC analysis of the resulting supernatant was
carried out by gradient elution with 0.8% acetic acid and 20% methanol
(pH 2.9) as mobile phase (flow rate 0.8 mL/min).
2.5. Spectral analysis
2.7. Cofactor determination
Absorption spectra were recorded at 25 °C on a Hewlett Packard
(Loveland, CO, USA) 8453 diode array spectrophotometer in 50 mM
Tris-SO4 (pH 8.0). Spectra were analysed using the UV–Visible
CHEMSTATION software package (Hewlett Packard).
The molar absorption coefficient of protein-bound FAD was deter-
mined by recording the absorption spectrum of 3HB6H in the pres-
ence and absence of 0.1% (w/v) SDS, assuming a molar absorption
coefficient for free FAD of 11.3 mM−1 cm−1 at 450 nm. Purified en-
zyme concentrations were routinely determined by measuring the
absorbance at 453 nm using the molar absorption coefficient for
protein-bound FAD (10.3 mM−1 cm−1).
The flavin cofactor of 3HB6H was identified by thin layer chroma-
tography (TLC). The cofactor was released from the protein by boiling
for 5 min or acid treatment. The protein precipitate was removed by
centrifugation and the supernatant was applied together with the
reference compounds FAD, FMN and riboflavin onto a TLC plate
(Baker-flex Silica Gel 1B2; JT Baker Inc., Phillipsburg, NY, USA).
Butanol/acetic acid/water (5:3:3) served as the mobile phase.
2.8. Enzyme activity
3HB6H activity was routinely assayed by following the decrease in
absorbance of NADH at 360 nm at 25 °C on a Hewlett Packard 8453
diode array spectrophotometer. Initial velocity values were calculated
using a molar absorption coefficient (ε360) of 4.31 mM−1 cm−1. The
standard assay mixture contained 50 mM Tris–SO4 (pH 8.0), 200 μM
3-hydroxybenzoate and 250 μM NADH; the reaction was started by
2.6. HPLC product analysis
The enzymatic conversions were analysed by HPLC using an Applied
Biosystems 400 pump equipped with a Waters 996 photodiode-

436
S. Montersino, W.J.H. van Berkel / Biochimica et Biophysica Acta 1824 (2012) 433–442
addition of 45 nM enzyme. One unit of enzyme activity (U) is defined
as the amount of enzyme that consumes 1 μmol of NADH per min. The
optimal pH for activity of 3HB6H was determined using 25 mM MES,
HEPES and CHES buffers with varying pH (pH 5.5–9.5) and adjusted
to an ionic strength of 0.1 M with Na2SO4 [24].
The activity of 3HB6H with 3-hydroxybenzoate, 2,3-dihydroxy-
benzoate, 3,5-dihydroxybenzoate and 2,5-dihydroxybenzoate followed
Michaelis–Menten kinetics [25]. Kinetic parameters were calculated
from multiple measurements with various substrate concentrations
using a direct nonlinear regression fit to the data.
hydroxylases from different bacterial orders and some eukaryotic systems
(see Supporting Table S1 for complete UniProt ID list). The neighbour
joining method was used to draw the phylogenetic tree (Fig. 2).
Among the retrieved sequences, only two annotated functions
(Q0SJD0 and Q0SPD2) were confirmed by phylogenetic analysis.
Q0SJD0 encodes for 3-(2-hydroxyphenyl) propionate hydroxylase, an
enzyme characterised solely in Rhodococcocus aetherivorans [27].
Q0SPD2 encodes for 4-hydroxybenzoate 3-hydroxylase, the prototype
of the flavoprotein hydroxylase family [28]. Q0SAU3 and Q0SFD6 are
annotated as pentachlorophenol monooxygenases but neither cluster
with the characterised pentachlorophenol-converting enzymes.
Eleven sequences are annotated either as putative monooxygen-
ase, FAD binding monooxygenase or aromatic ring hydroxylase.
From our analysis, putative functions can be assigned to Q0SJG8,
Q0S9K8 and Q0S0L2. Q0SJG8 clusters with a group of flavoproteins
acting on activated substrates, including among others salicylyl-CoA
5-hydroxylase (Q7X281) and 2-aminobenzoyl-CoA monooxygenase/
reductase (Q93FB38, Q93FC6), acting on CoA-activated salicylate
and CoA-activated 2-aminobenzoate, respectively. Another member
of this group is the newly identified enzyme SibG that acts on
an activated form of sibiromycin [29]. Q0S9K8 might encode for
6-hydroxynicotinate 3-monooxygenase (39% sequence identity and
56% similarity), since it belongs to the 6-hydroxynicotinate 3-
monooxygenase clade, where two enzymes both from Pseudomonas
have been characterised [30, 31]. Q0S0L2 has high sequence homolo-
gy with rifampicin monooxygenase, an enzyme present in actinomy-
cetes such as Rhodococcus equi and Nocardia farcinica [13, 14].
Three sequences have been annotated with a probable enzymatic
function but all of them seem mis-annotated. According to the phyloge-
netic tree, Q0RW34 encodes for a 3-(2-hydroxyphenyl) propionate
hydroxylase instead of 2,4-dichlorophenol 6-monooxygenase. Three
different types of 2,4-dichlorophenol 6-monooxygenases are known
[32–34], but Q0RW34 does not cluster with any of them. Q0SFL5 does
not seem to encode for a 3-(3-hydroxyphenyl) propionate hydroxylase,
since the sequence outgroups from the clade quite early. Therefore, no
putative function for Q0SFL5 can be addressed. Finally, Q0SFK6 is anno-
tated as a putative salicylate hydroxylase, but both sequence alignment
and cluster analysis (Fig. 2) support another function (see Section 3.2).
2.9. Oxygen consumption
An OxyTherm Clark-type oxygen electrode system (Hansatech,
Norfolk, UK) was used to determine the hydroxylation efficiency of
3HB6H towards different substrates. The assay solution (final volume
1.0 mL) contained 350 μM substrate, 250 μM NADH and 1 μM 3HB6H
in 50 mM air saturated Tris–SO4 (pH 8.0) at 25 °C. At the end of the
reaction, the amount of hydrogen peroxide produced was determined
by adding 0.02 mg/mL of catalase (2 H2O2 →O2 +2 H2O).
2.10. Substrate binding studies
The interaction of 3HB6H (25 μM) with substrate analogues was
studied in 50 mM Tris–SO4 (pH 8.0). Dissociation constants (Kd) of
enzyme/substrate complexes were determined from flavin absorp-
tion difference spectra. The Kd was calculated from the changes in
absorbance at 490 nm using a direct nonlinear regression fit to the
data with IGOR (Wavemetrics, Lake Oswego, OR, USA). Apparent
dissociation constants were calculated by fitting the relative absor-
bencies to Eq. (1), a modified equation compared with the one
described elsewhere [26]:
A490 ¼ εES ꢀ ES þ εE ꢀ ðEt−ESÞ
ð1Þ
where
ꢀ
ꢁ
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
ES ¼ ðEt þ St þ KdÞ− ðEt þ St þ KdÞ −4ES =2
3.2. Functional annotation of 3HB6H
and A490, Et, St, ES, Kd, εES, εE are the absorbances observed at 490 nm,
the total enzyme concentration, the total substrate concentration, the
enzyme–substrate complex concentration, the dissociation constant
of the enzyme–substrate complex, the molar absorption coefficient
of the enzyme–substrate complex and the molar absorption coeffi-
cient of the free enzyme, respectively.
3.2.1. Gene order conservation
Putative salicylate monooxygenase (RHA1_ro01869, GENE ID:
4218663) DNA locus region contains genes belonging to the gentisate
degradation pathway (Fig. 3): RHA1_ro01866 (GENE ID: 4218660)
encodes for a ring-fission dioxygenase that converts gentisate to
maleylpyruvate. This product is further converted to fumarylpyruvate
by a maleylpyruvate isomerase encoded by RHA1_ro01865 (GENE ID:
4218659). DNA regions encoding for known 3HB6Hs possess the
same gene order organisation as the RHA1_ro01869 DNA region,
especially the 3HB6H degradation operon from Corynebacterium
glutamicum (Fig. 3) [33, 34]. Q0SFK6 shares 93% amino acid sequence
identity with 3HB6H from Rhodococcus NCIMB 12038 (narX, GenBank:
HM852512.1) [35] and 96% with putative 3HB6H from R. opacus
B4 (ROP_15470, GENE ID: 7741586). Thus, based on protein sequence
similarity, phylogenetic tree clustering and gene order conservation,
we hypothesised that Q0SFK6 encodes for 3HB6H instead of salicylate
hydroxylase. Very recently, a homologue 3HB6H gene was identified
in Candida parapsilosis [36]. The C. parapsilosis 3HB6H sequence does
not cluster with bacterial 3HB6Hs (Fig. 2). Instead, it forms a separate
clade together with 4-hydroxybenzoate 1-hydroxylase from the same
fungus [36, 37], probably due to the common eukaryotic origin.
3. Results
3.1. Analysis of R. jostii RHA1 genome
3.1.1. Identification of R. jostii RHA1 flavoprotein hydroxylases
By searching the R. jostii RHA1 genome with the three flavoprotein
hydroxylase fingerprints of PHBH from Pseudomonas fluorescens
(Fig. 1) [16], we retrieved eighteen candidate sequences (Fig. 1 and
Table 1). Out of the candidate sequences, fifteen are present in
other sequenced Rhodococcus strains with highest similarities found
in Rhodococcus opacus B4. Q0S9S6, Q0S6Z5 and Q0RV68 are only
present in R. jostii RHA1. From the alignments it is clear that the
consensus sequences of the selected proteins are well conserved,
even though small deviations occur.
3.1.2. Multiple sequence alignment and phylogenetic tree analysis
Most of the retrieved sequences are not linked to a clear annotated
function. To predict protein functions, we performed a multiple sequence
alignment and phylogenetic analysis with characterised flavoprotein
3.2.2. Cloning, expression and purification of Q0SFK6
To assess the activity of Q0SFK6, the gene was cloned into vector
pBAD/Myc-His under control of the inducible ara promoter. The

S. Montersino, W.J.H. van Berkel / Biochimica et Biophysica Acta 1824 (2012) 433–442
437
Fig. 2. Neighbour‐joining tree of flavoprotein hydroxylases. Underlined UniProt IDs represent hydroxylases from R. jostii RHA1. Italics UniProt IDs are sequences of hydroxylases with known
crystal structures. On the right, enzymatic function has been reported. For the complete list of sequences used see supporting information (Supporting Table S1).
expression vector was transformed in E. coli TOP10 cells and overex-
pression of the His-tagged protein was induced by adding 0.02%
(w/v) arabinose to the Terrific Broth medium. High level of expres-
sion was found after 16 h of induction at 37 °C. No expression was
achieved in Luria Bertani medium. The recombinant protein was pu-
rified to apparent homogeneity by two successive chromatographic
steps (Table 2). Approximately 420 mg of recombinant Q0SFK6 pro-
tein could be purified from 6 L batch culture containing 36 g (wet
weight) of cells. SDS-PAGE showed a single band with an apparent
molecular mass of 47 kDa (Fig. 4A), in agreement with the Q0SFK6
amino acid sequence. The relative molecular mass of the native
recombinant protein was estimated to be ~108 kDa by analytical gel
filtration (Fig. 4B), which indicates a dimer conformation in solution.
3.2.3. Spectral properties
Recombinant Q0SKF6 showed a typical flavoprotein absorption
spectrum with maxima at 274 nm, 383 nm and 453 nm and a shoul-
der at 480 nm (Fig. 5A). The molar absorption coefficient of protein-

438
S. Montersino, W.J.H. van Berkel / Biochimica et Biophysica Acta 1824 (2012) 433–442
Fig. 3. Physical maps of gene clusters encoding for 3‐hydroxybenzoate and gentisate degradation. ORFs annotated as unknown function are depicted in white. The sequences were taken
from (GenBank ID) R. jostii RHA1 (CP000431), Rhodococcus. sp. NCIMB 12083 (HM852512), R. opacus B4 (NC_012522), C. glutamicum ATCC13032 (BA000036), K. pneumoniae M5a1
(AY648560), P. alcaligenes NCIMB 9867 (AF173167, DQ394580) and P. naphthalenivorans CJ2 (DQ167474).
bound flavin was determined to be 10.3 mM−1 cm−1 at 453 nm. The
274/A453 ratio of the FAD-saturated protein preparation was 11.9.
The A383/A453 ratio (Fig. 5A) is 1.2, a rather high value for flavopro-
teins. The flavin cofactor could be released from the protein by boiling
or acid treatment and was identified as FAD by TLC.
was strongly stimulated by 3-hydroxybenzoate. The aromatic product
was identified as gentisate (2,5-dihydroxybenzoate) by absorption
spectral analysis, and HPLC (Table 4). Thus, enzyme activity and
product analysis confirms the new annotation of Q0SKF6 as 3-
hydroxybenzoate 6-hydroxylase (3HB6H).
A
3.3.2. Kinetic parameters
3.3. Catalytic properties of 3HB6H
3HB6H from R. jostii RHA1 catalyses the para-hydroxylation of
3-hydroxybenzoate with the consumption of NAD(P)H and oxygen.
The enzyme displayed a maximum activity around pH 8.0 in Tris–SO4
buffer and around pH 8.6 in HEPES buffer (data not shown). Like other
flavoprotein hydroxylases [38–40], the enzyme was strongly inhibited
by chloride ions. The steady-state kinetic parameters of 3HB6H were de-
termined at 25 °C in 50 mM Tris–SO4 (pH 8.0) (Table 3). This analysis
revealed that the enzyme clearly prefers NADH (Km=48 4 μM) over
NADPH (Km=390 80 μM) as external electron donor.
3.3.1. Enzyme activity and product analysis
Enzyme activity was assayed by measuring the consumption of
NADH spectrophotometrically. Q0SFK6 showed a very low NADH
oxidase activity (b1 Umg−1). No increase in enzyme activity was
detected in the presence of salicylate, while NADH consumption
Table 2
Purification of R. jostii RHA1 Q0SFK6 expressed in E. coli.
Step
Protein
(mg)
Activity
(U)
Specific activity
Yield
%
3.3.3. Substrate and effector specificity
(U mg−1
)
3HB6H from R. jostii RHA1 catalyses the regioselective para-
hydroxylation of 3-hydroxybenzoate forming gentisate (Fig. 6).
Besides the parent substrate, 3HB6H was most active with 2,3-
dihydroxybenzoate and 3,5-dihydroxybenzoate (Table 3; Fig. 6).
Cell extract
Ni-NTA agarose
Q-Sepharose
3360
780
450
18,718
14,997
10,738
6
19
24
100
80
57

S. Montersino, W.J.H. van Berkel / Biochimica et Biophysica Acta 1824 (2012) 433–442
439
Fig. 5. Spectral properties of 3HB6H. A) Absorption spectra of free 3HB6H (solid line)
and 3HB6H in complex with 890 μM 3‐hydroxybenzoate (dashed line). The enzyme
concentration was 25 μM. B) Absorption difference spectra observed upon binding of
3‐hydroxybenzoate to 25 μM 3HB6H. The curves shown are the difference spectra,
the corrected enzyme spectra in the presence of 17, 50, 82, 177, 455 and 890 μM 3‐
hydroxybenzoate, minus the enzyme spectrum in the absence of 3‐hydroxybenzoate. All
experiments were performed in 50 mM Tris—SO4 (pH 8.0).
Fig. 4. Hydrodynamic properties of 3HB6H. A) SDS/PAGE analysis of the purification of
3HB6H from R. jostii RHA1 expressed in E. coli. Lane A, cell extract; Lane B, Ni‐NTA pool;
Lane C, Q‐sepharose pool; Lane D, low molecular weight marker. B) Analytical gel filtra-
tion. Reference proteins used to calibrate the column and calculate the 3HB6H apparent
molecular mass: cytochrome
c (12.3 kDa), myoglobin (17.8 kDa), α‐chymotrypsin
(25 kDa), ovalbumin (43 kDa), bovine serum albumin (68 and 136 kDa), 4‐hydroxy-
benzoate 3‐hydroxylase (90 kDa), lipoamide dehydrogenase (102 kDa), phenol 2‐hy-
droxylase (152 kDa), catalase (232 kDa), ferritin (440 kDa) and vanillyl‐alcohol oxidase
(510 kDa); ■ 3HB6H.
4. Discussion
R. jostii RHA1 is a versatile Gram-positive soil bacterium degrading
a wide range of organic compounds, including lignin [1]. It contains
an impressive amount of oxygenases, especially flavoprotein mono-
oxygenases. Recently, Szolkowy and colleagues characterised the
Baeyer–Villiger monooxygenase (BMVO) subclass of R. jostii RHA1
pointing out an exceptional diversity in regio- and enantioselectivity
[41].
In the present study, we searched for new flavoprotein hydroxy-
lases in the R. jostii RHA1 genome. Most of these enzymes are
single-component proteins that form a specific class of flavoprotein
monooxygenases (subclass A). By using three different fingerprints
we were able to retrieve eighteen candidate flavoprotein hydroxylase
sequences. Three of these sequences turned out to be R. jostii
RHA1 specific, while the others are present also in other available
Next to that, the enzyme slowly converted (kcatb5 s−1) a number of 3-
hydroxybenzoate derivatives with substituents at the 4-position
(Table 4). Among these compounds, 3,4-dihydroxybenzoate is
a
very poor substrate (Table 4). In analogy with the conversion of 2,3,4-
trihydroxybenzoate to 2,3,4,6-tetrahydroxybenzoate, the enzyme trans-
formed the initial product of the reaction with 3,5-dihydroxybenzoate
(2,3,5-trihydroxybenzoate) to 2,3,5,6-tetrahydroxybenzoate (Table 4).
To discriminate between true substrates and non-substrate effec-
tors, oxygen consumption experiments were performed in the ab-
sence and presence of catalase. With all substrates, a certain degree
of uncoupling of hydroxylation was observed, as evidenced by the
production of hydrogen peroxide (Table 3). Gentisate, the physiolog-
ical product, is a good effector of 3HB6H. This compound is not
converted but strongly stimulates the NADH oxidase activity of the
enzyme (Fig. 6).
Table 3
Substrate and effector specificity of 3HB6H.
Substrate
Kd
(μM)
Km
(μM)
kcat
kcat/Km
Uncoupling
(%)
(s−1
)
(M−1 s−1
)
8×105
5×104
2×105
5×104
5
10
20
99
3.3.4. Substrate binding
3-hydroxybenzoate
3,5-dihydroxybenzoate 310 60 250 10 13
2,3-dihydroxybenzoate
2,5-dihydroxybenzoate 190 10 140 30
48
2
46
3
35
1
1
1
1
Upon titration of 3HB6H with 3-hydroxybenzoate derivatives,
characteristic perturbations in the absorption properties of the FAD
cofactor were observed (Fig. 5B). From the absorption differences
around 490 nm, dissociation constants (Kd values) were estimated
for the binary enzyme–substrate complexes (Table 3). Similar titra-
tion experiments with salicylate established that this compound
does not induce any flavin absorption change (data not shown).
63
9
51
7
10
7
Apparent kinetic constants were determined at 25 °C in 50 mM Tris–SO4 (pH 8.0). Values
are presented as the mean of three experiments. The percentage of uncoupling of
hydroxylation was determined from polarographic measurements using a Clarke-type
electrode. The percentage of uncoupling equals twice the amount of oxygen, produced
after addition of catalase at the end of the reaction (2 H2O2→O2+2 H2O).

440
S. Montersino, W.J.H. van Berkel / Biochimica et Biophysica Acta 1824 (2012) 433–442
flavoprotein hydroxylases, usually performed by subclass B mono-
oxygenases [42–47]. However, a hybrid kinetic mechanism be-
tween flavoprotein hydroxylases and subclass B monooxygenases
has been reported for L-ornithine N5-oxygenase [46, 48]
Q0RW34 and Q0S9K8 are predicted to be a 3-(2-hydroxyphenyl)
propionate hydroxylase and 6-hydroxynicotinate 3-monooxygenase,
respectively. The first enzyme plays a role in lignin degradation
[27], whereas the latter is involved in the metabolism of nicotinic
acid in P. fluorescence T5 and Pseudomonas putida KT2400 [30, 31],
performing decarboxylative hydroxylation of 6-hydroxynicotin acid
into 2,5-dihydroxypyridine.
Q0SJG8 most likely acts on an activated substrate, similarly to
SibG [29, 49, 50], 2-aminobenzoylCoA monooxygenase/reductase
(ACMR) and salicylyl-CoA 5-hydroxylase. Interestingly, all the latter
enzymes cluster in the same part of the tree (Fig. 2), suggesting that
they might have a conserved mode of CoA-substrate recognition.
Q0SFK6 represents a clear mis-annotation in the R. jostii genome.
From comparing several known 3HB6H amino acid sequences and
gene clusters with the Q0SFK6 Rhodococcus DNA locus, we conclude
that Q0SFK6 encodes for a 3HB6H which is involved in the gentisate
degradation pathway. The organisation of the gene cluster is con-
served in C. glutamicum and in R. opacus B4, and to a lesser extent in
Rhodococcus NCIMB 12038 (Fig. 3). The presence of many regulatory
elements or transposase-like elements surrounding most of the
3HB6H clusters could be viewed as an element of plasticity of
this cluster, since gentisate formation could involve also salicylate
5-hydroxylase.
The enzymatic activity of Q0SKF6 was addressed with the recom-
binant protein. In agreement with the above proposition derived from
sequence data, activity was found with 3-hydroxybenzoate and not
with salicylate. Product analysis confirmed that Q0SKF6 encodes for
3HB6H and thus is mis-annotated as a salicylate hydroxylase.
Analytical gel filtration indicated that 3HB6H from R. jostii RHA1
is a homodimer. So far, a monomeric [51] or trimeric [34, 35, 52, 53]
nature has been reported for 3HB6H enzymes. 3HB6H from R. jostii
RHA1 has narrow substrate specificity. Besides the parent substrate,
the enzyme accepts 3-hydroxybenzoate compounds with substitu-
ents in ortho- and meta-position. Importantly, it was established
here for the first time that gentisate is a non-substrate effector of
3HB6H, stimulating FAD reduction and NADH consumption with
production of hydrogen peroxide. Such behaviour of the aromatic
product is also seen with the prototype of this flavoprotein subfamily,
PHBH [28]. In case of 3HB6H, waste of NADH consumption will be
negligible under physiological conditions, since gentisate dioxy-
genases convert gentisate very efficiently [54], preventing hydrogen
peroxide production in the cell.
Three different microbial enzymes that produce gentisate have
been reported: flavin-dependent salicylyl-CoA 5-hydroxylase, Rieske
(2Fe–2S) dependent salicylate 5-hydroxylase, and flavin-dependent
3HB6H. Salicylyl-CoA 5-hydroxylase has been characterised in
Streptomyces WA46 [49], while two-component Rieske type salicylate
5-hydroxylase has been studied in Ralstonia sp. U2 and Polaromonas
naphthalenivorans CJ2 [21, 55]. Different 3HB6Hs have been described
in Gram-positive [34, 35] and Gram-negative bacteria [51–53], and
recently in C. parapsilosis [36].
Fig. 6. Reactions catalyzed by 3HB6H. From top to bottom: reactions with the best sub-
strates (3‐hydroxybenzoate, 2,3‐dihydroxybenzoate and 3,5‐dihydroxybenzoate) and
the best non‐substrate effector (2,5‐dihydroxybenzoate).
Rhodococcus genomes, especially in R. opacus strain B4. The highest
sequence similarity with flavoprotein hydroxylases from R. opacus
strain B4 is in agreement with the overall synteny between these
Rhodococcus strains.
Most of the candidate flavoprotein hydroxylase sequences have
been annotated without a specific function or are mis-annotated.
From multiple sequence alignment and phylogenetic analysis we
could assess some new putative functions.
Q0S0L2 encodes for a rifampicin monooxygenase, an enzyme
able to decompose the antibiotic via N-hydroxylation. This type of
decomposition mechanism is driven by flavoprotein monooxy-
genases in N. farcinica and R. equi, both pathogen strains [13, 14].
Rifampicin is used for tuberculosis treatment and drug inactivation
in Mycobacterium is driven by decomposition as in Rhodococcus and
Nocardia species [14]. N-hydroxylation is an unusual reaction for
Table 4
HPLC product analysis of the 3HB6H reaction (see Section 3.3).
Substrate
Retention Product(s)
time
Retention
time
Among Rhodococcus species, gentisate formation has been mostly
related to salicylate 5-hydroxylase activity. Grund and colleagues
measured salicylyl-CoA 5-hydroxylase activity in R. opacus B4 cell
extracts [56] and Di Gennaro and colleagues [57] described a gene
cluster in Rhodococcus R7 containing a salicylyl-CoA 5-hydroxylase
gene. However, in both cases, no sequence or evidence at the enzyme
level was reported. By using the R. jostii 3HB6H sequence as bait
among the Rhodococcus genomes available, we found a putative
3HB6H in R. opacus B4 contained in a conserved 3-hydroxybenzoate/
gentisate-degradation operon (Fig. 3). In contrast, with salicylyl-CoA
5-hydroxylase (Q7X281) and salicylate 5-hydroxylase (Q3S4D3) as
3-hydroxybenzoate
2,5-dihydroxybenzoate
3,5-dihydroxybenzoate
19
11.7
7.2
2,5-dihydroxybenzoate
–
12.3
12.3
4
2.7
4.1
7.8
2.9
2,3,5-trihydroxybenzoate
2,3,5,6-tetrahydroxybenzoate
2,3,6-trihydroxybenzoate
–
2,3-dihydroxybenzoate
3,4-dihydroxybenzoate
2,3,4-trihydroxybenzote
3-hydroxy-4-aminobenzoate
19.5
8.3
8.6
2,3,4,6-tetrahydroxybenzoate
4-amino-2,5-dihydroxybenzoate 5.8
4-methyl-2,5-dihydroxybenzoate 17.5
5.1
3-hydroxy-4-methylbenzoate 22.8
3-hydroxy-4-
21.5
4-methoxy-2,5-
4.7
methoxybenzoate
dihydroxybenzoate

S. Montersino, W.J.H. van Berkel / Biochimica et Biophysica Acta 1824 (2012) 433–442
441
[12] K.S. Ryan, A.R. Howard-Jones, M.J. Hamill, S.J. Elliott, C.T. Walsh, C.L. Drennan,
Crystallographic trapping in the rebeccamycin biosynthetic enzyme RebC, Proc.
Natl. Acad. Sci. U. S. A. 104 (2007) 15311–15316.
[13] S.J. Andersen, S. Quan, B. Gowan, E.R. Dabbs, Monooxygenase-like sequence of a
Rhodococcus equi gene conferring increased resistance to rifampin by inactivating
this antibiotic, Antimicrob. Agents Chemother. 41 (1997) 218–221.
[14] Y. Hoshino, S. Fujii, H. Shinonaga, K. Arai, F. Saito, T. Fukai, H. Satoh, Y. Miyazaki, J.
Ishikawa, Monooxygenation of rifampicin catalyzed by the rox gene product of
Nocardia farcinica: structure elucidation, gene identification and role in drug
resistance, J. Antibiot. (Tokyo) 63 (2010) 23–28.
baits, no Rhodococcus homologues were found. Recently, a 3HB6H was
described in Rhodococcus sp. NCIMB 12038 [35]. 3HB6H from R. jostii
RHA1 shares 93% amino acid sequence with 3HB6H from Rhodococcus
sp. NCIMB 12038. To our best knowledge, these are the only Rhodococcus
gentisate producing enzymes characterised thus far.
5. Conclusion
[15] G. Volkers, G.J. Palm, M.S. Weiss, G.D. Wright, W. Hinrichs, Structural basis for a
new tetracycline resistance mechanism relying on the TetX monooxygenase,
FEBS Lett. 585 (2011) 1061–1066.
[16] M.H. Eppink, H.A. Schreuder, W.J. van Berkel, Identification of a novel conserved
sequence motif in flavoprotein hydroxylases with a putative dual function in
FAD/NAD(P)H binding, Protein Sci. 6 (1997) 2454–2458.
[17] R.K. Wierenga, P. Terpstra, W.G. Hol, Prediction of the occurrence of the ADP-
binding βαβ-fold in proteins, using an amino acid sequence fingerprint, J. Mol.
Biol. 187 (1986) 101–107.
[18] G. Eggink, H. Engel, G. Vriend, P. Terpstra, B. Witholt, Rubredoxin reductase of
Pseudomonas oleovorans. Structural relationship to other flavoprotein oxidore-
ductases based on one NAD and two FAD fingerprints, J. Mol. Biol. 212 (1990)
135–142.
[19] P. Macheroux, B. Kappes, S.E. Ealick, Flavogenomics — a genomic and structural
view on flavin-dependent proteins, FEBS J. 278 (2011) 2625–2634.
[20] S.L. Fuenmayor, M. Wild, A.L. Boyes, P.A. Williams, A gene cluster encoding steps in
conversion of naphthalene to gentisate in Pseudomonas sp. strain U2, J. Bacteriol.
180 (1998) 2522–2530.
[21] C.O. Jeon, M. Park, H.-S. Ro, W. Park, E.L. Madsen, The naphthalene catabolic (nag)
genes of Polaromonas naphthalenivorans CJ2: evolutionary implications for two gene
clusters and novel regulatory control, Appl. Environ. Microbiol. 72 (2006) 1086–1095.
[22] J.D. Thompson, D.G. Higgins, T.J. Gibson, CLUSTAL W: improving the sensitivity of pro-
gressive multiple sequence alignment through sequence weighting, position-specific
gap penalties and weight matrix choice, Nucleic Acids Res. 22 (1994) 4673–4680.
[23] U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of
bacteriophage T4, Nature 227 (1970) 680–685.
In conclusion, we have retrieved a pool of flavoprotein hydroxy-
lases in R. jostii RHA1 that share conserved flavoprotein fingerprints
but can act on diverse aromatic substrates. Furthermore, from multi-
ple sequence alignments and phylogenetic analysis, a number of
functional annotations have been resolved.
Q0SFK6 is an example of a successful functional annotation.
From biochemical analysis we obtained clear evidence that Q0SFK6
was mis-annotated as a salicylate hydroxylase, an enzyme related
to 3HB6H but with a different substrate specificity. 3HB6H is one
of the few para-hydroxylating flavoprotein monooxygenases charac-
terised. Recently, Hiromoto and coworkers [58] obtained the crystal
structure of 3-hydroxybenzoate 4-hydroxylase (3HB4H) from
Comamonas testosteroni. 3HB4H is using the same substrate as
3HB6H but performs an ortho-hydroxylation reaction. Comparison
between structure and mechanism of 3HB4H and 3HB6H might
help in understanding the striking regioselectivity of flavoprotein
aromatic hydroxylases.
Supplementary data to this article can be found online at doi:10.
1016/j.bbapap.2011.12.003.
[24] R.A. Wijnands, J. van der Zee, J.W. Van Leeuwen, W.J. van Berkel, F. Müller, The
importance of monopole–monopole and monopole–dipole interactions on the
binding of NADPH and NADPH analogues to p-hydroxybenzoate hydroxylase
from Pseudomonas fluorescens. Effects of pH and ionic strength, Eur. J. Biochem.
139 (1984) 637–644.
Acknowledgements
[25] K.A. Johnson, R.S. Goody, The original Michaelis constant: translation of the 1913
Michaelis–Menten paper, Biochemistry 50 (2011) 8264–8269.
[26] Y.J. Bollen, S.M. Nabuurs, W.J. van Berkel, C.P. van Mierlo, Last in, first out: the role
of cofactor binding in flavodoxin folding, J. Biol. Chem. 280 (2005) 7836–7844.
[27] J.A. Powell, J.A. Archer, Molecular characterisation of a Rhodococcus ohp operon,
Antonie Van Leeuwenhoek 74 (1998) 175–188.
We are grateful to Anette Riebel and Daniel Torres Pazmiño for
In-Fusion PCR Cloning System optimisation, Laura de Hoon for exper-
imental contributions and Adrie Westphal for technical assistance.
This study was supported by the Integrated Biosynthesis Organic Syn-
thesis (IBOS) project of the Netherlands Organization for Scientific
Research (NWO).
[28] B. Entsch, W.J. van Berkel, Structure and mechanism of para-hydroxybenzoate
hydroxylase, FASEB J. 9 (1995) 476–483.
[29] T.W. Giessen, F.I. Kraas, M.A. Marahiel, A four-enzyme pathway for 3,5-dihydroxy-4-
methylanthranilic acid formation and incorporation into the antitumor antibiotic
sibiromycin, Biochemistry 50 (2011) 5680–5692.
References
[30] J.I. Jimenez, A. Canales, J. Jimenez-Barbero, K. Ginalski, L. Rychlewski, J.L. Garcia, E. Diaz,
Deciphering the genetic determinants for aerobic nicotinic acid degradation: the nic
cluster from Pseudomonas putida KT2440, Proc. Natl. Acad. Sci. U. S. A. 105 (2008)
11329–11334.
[31] H. Nakano, M. Wieser, B. Hurh, T. Kawai, T. Yoshida, T. Yamane, T. Nagasawa, Purifica-
tion, characterization and gene cloning of 6-hydroxynicotinate 3-monooxygenase
from Pseudomonas fluorescens TN5, Eur. J. Biochem. 260 (1999) 120–126.
[32] N.L. Huong, K. Itoh, M. Miyamoto, K. Suyama, H. Yamamoto, Chlorophenol
hydroxylase activity encoded by TfdB from 2,4-dichlorophenoxyacetic acid
(2,4-D)-degrading Bradyrhizobium sp. strain RD5-C2, Biosci. Biotechnol. Biochem.
71 (2007) 1691–1696.
[1] M. Ahmad, J.N. Roberts, E.M. Hardiman, R. Singh, L.D. Eltis, T.D. Bugg, Identifica-
tion of DypB from Rhodococcus jostii RHA1 as a lignin peroxidase, Biochemistry
50 (2011) 5096–5107.
[2] M.J. Larkin, L.A. Kulakov, C.C.R. Allen, Biodegradation and Rhodococcus-masters of
catabolic versatility, Curr. Opin. Biotechnol. 16 (2005) 282–290.
[3] M.J. Larkin, L.A. Kulakov, C.C.R. Allen, Biodegradation by members of the genus
Rhodococcus: biochemistry, physiology, and genetic adaptation, Adv. Appl. Microbiol.
59 (2006) 1–29.
[4] L. Martínková, B. Uhnáková, M. Pátek, J. Nesvera, V. Kren, Biodegradation potential of
the genus Rhodococcus, Environ. Int. 35 (2009) 162–177.
[5] M.P. McLeod, R.L. Warren, W.W.L. Hsiao, N. Araki, M. Myhre, C. Fernandes, D.
Miyazawa, W. Wong, A.L. Lillquist, D. Wang, M. Dosanjh, H. Hara, A. Petrescu, R.D.
Morin, G. Yang, J.M. Stott, J.E. Schein, H. Shin, D. Smailus, A.S. Siddiqui, M.A. Marra,
S.J.M. Jones, R. Holt, F.S.L. Brinkman, K. Miyauchi, M. Fukuda, J.E. Davies, W.W.
Mohn, L.D. Eltis, The complete genome of Rhodococcus sp. RHA1 provides insights
into a catabolic powerhouse, Proc. Natl. Acad. Sci. U. S. A. 103 (2006) 15582–15587.
[6] W.J.H. van Berkel, N.M. Kamerbeek, M.W. Fraaije, Flavoprotein monooxygenases,
a diverse class of oxidative biocatalysts, J. Biotechnol. 124 (2006) 670–689.
[7] R. Hammann, H.J. Kutzner, Key enzymes for the degradation of benzoate, m-and
p-hydroxybenzoate by some members of the order Actinomycetales, J. Basic
Microbiol. 38 (1998) 207–220.
[8] C.S. Harwood, R.E. Parales, The β-ketoadipate pathway and the biology of self-
identity, Annu. Rev. Microbiol. 50 (1996) 553–590.
[9] P.S. Phale, A. Basu, P.D. Majhi, J. Deveryshetty, C. Vamsee-Krishna, R. Shrivastava,
Metabolic diversity in bacterial degradation of aromatic compounds, OMICS 11
(2007) 252–279.
[10] P. Kallio, Z. Liu, P. Mantsala, J. Niemi, M. Metsa-Ketela, Sequential action of two
flavoenzymes, PgaE and PgaM, in angucycline biosynthesis: chemoenzymatic
synthesis of gaudimycin C, Chem. Biol. 15 (2008) 157–166.
[11] Y. Lindqvist, H. Koskiniemi, A. Jansson, T. Sandalova, R. Schnell, Z. Liu, P. Mäntsälä,
J. Niemi, G. Schneider, Structural basis for substrate recognition and specificity
in aklavinone-11-hydroxylase from rhodomycin biosynthesis, J. Mol. Biol. 393
(2009) 966–977.
[33] X.-H. Shen, C.-Y. Jiang, Y. Huang, Z.-P. Liu, S.-J. Liu, Functional identification
of novel genes involved in the glutathione-independent gentisate pathway in
Corynebacterium glutamicum, Appl. Environ. Microbiol. 71 (2005) 3442–3452.
[34] Y.F. Yang, J.J. Zhang, S.H. Wang, N.Y. Zhou, Purification and characterization of
the ncgl2923-encoded 3-hydroxybenzoate 6-hydroxylase from Corynebacterium
glutamicum, J. Basic Microbiol. 50 (2010) 599–604.
[35] T.T. Liu, Y. Xu, H. Liu, S. Luo, Y.J. Yin, S.J. Liu, N.Y. Zhou, Functional characterization
of a gene cluster involved in gentisate catabolism in Rhodococcus sp. strain NCIMB
12038, Appl. Microbiol. Biotechnol. 90 (2010) 671–678.
[36] Z. Holesova, M. Jakubkova, I. Zavadiakova, I. Zeman, L. Tomaska, J. Nosek, Gentisate
and 3-oxoadipate pathways in the yeast Candida parapsilosis: Identification and
functional analysis of the genes coding for 3-hydroxybenzoate 6-hydroxylase and
4-hydroxybenzoate 1-hydroxylase, Microbiology 157 (2011) 2152–2163.
[37] M.H. Eppink, S.A. Boeren, J. Vervoort, W.J. van Berkel, Purification and properties
of 4-hydroxybenzoate 1-hydroxylase (decarboxylating), a novel flavin adenine
dinucleotide-dependent monooxygenase from Candida parapsilosis CBS604,
J. Bacteriol. 179 (1997) 6680–6687.
[38] P.J. Steennis, M.M. Cordes, J.H. Hilkens, F. Muller, On the interaction of para-
hydroxybenzoate hydroxylase from Pseudomonas fluorescens with halogen ions,
FEBS Lett. 36 (1973) 177–180.
[39] W.J. van Berkel, W.J. van den Tweel, Purification and characterisation of 3-
hydroxyphenylacetate 6-hydroxylase: a novel FAD-dependent monooxygenase
from a Flavobacterium species, Eur. J. Biochem. 201 (1991) 585–592.

442
S. Montersino, W.J.H. van Berkel / Biochimica et Biophysica Acta 1824 (2012) 433–442
[40] W.J. van Berkel, F. Müller, Flavin-dependent monooxygenase with special references
to p-hydroxybenzoate hydroxylase, in: B. Raton (Ed.), Chemistry and biochemistry
of flavoenzymes, 2, CRC Press, 1991, pp. 1–29.
[41] C. Szolkowy, L.D. Eltis, N.C. Bruce, G. Grogan, Insights into sequence–activity
relationships amongst Baeyer–Villiger monooxygenases as revealed by the intra-
genomic complement of enzymes from Rhodococcus jostii RHA1, Chembiochem
10 (2009) 1208–1217.
[42] D.E. Torres Pazmiño, H.M. Dudek, M.W. Fraaije, Baeyer–Villiger monooxygenases:
recent advances and future challenges, Curr. Opin. Chem. Biol. 14 (2010) 138–144.
[43] S.W. Chocklett, P. Sobrado, Aspergillus fumigatus SidA is a highly specific ornithine
hydroxylase with bound flavin cofactor, Biochemistry 49 (2010) 6777–6783.
[44] L. Ge, S.Y. Seah, Heterologous expression, purification, and characterization of an
L-ornithine N(5)-hydroxylase involved in pyoverdine siderophore biosynthesis in
Pseudomonas aeruginosa, J. Bacteriol. 188 (2006) 7205–7210.
[45] J.A. Mayfield, R.E. Frederick, B.R. Streit, T.A. Wencewicz, D.P. Ballou, J.L. DuBois,
Comprehensive spectroscopic, steady state, and transient kinetic studies of a
representative siderophore-associated flavin monooxygenase, J. Biol. Chem. 285
(2010) 30375–30388.
[46] K.M. Meneely, E.W. Barr, J.M. Bollinger, A.L. Lamb, Kinetic mechanism of ornithine
hydroxylase (PvdA) from Pseudomonas aeruginosa: substrate triggering of O2
addition but not flavin reduction, Biochemistry 48 (2009) 4371–4376.
[47] K.M. Meneely, A.L. Lamb, Biochemical characterization of a flavin adenine dinucleotide-
dependent monooxygenase, ornithine hydroxylase from Pseudomonas aeruginosa,
suggests a novel reaction mechanism, Biochemistry 46 (2007) 11930–11937.
[48] J. Olucha, A.L. Lamb, Mechanistic and structural studies of the N-hydroxylating
flavoprotein monooxygenases, Bioorg. Chem. 39 (2011) 171–177.
[50] K. Schuhle, M. Jahn, S. Ghisla, G. Fuchs, Two similar gene clusters coding for
enzymes of a new type of aerobic 2-aminobenzoate (anthranilate) metabolism
in the bacterium Azoarcus evansii, J. Bacteriol. 183 (2001) 5268–5278.
[51] M. Suarez, E. Ferrer, A. Garrido-Pertierra, M. Martin, Purification and characteriza-
tion of the 3-hydroxybenzoate-6-hydroxylase from Klebsiella pneumoniae, FEMS
Microbiol. Lett. 126 (1995) 283–290.
[52] X. Gao, C.L. Tan, C.C. Yeo, C.L. Poh, Molecular and biochemical characterization of
the xlnD-encoded 3-hydroxybenzoate 6-hydroxylase involved in the degradation
of 2,5-xylenol via the gentisate pathway in Pseudomonas alcaligenes NCIMB 9867,
J. Bacteriol. 187 (2005) 7696–7702.
[53] M. Park, Y. Jeon, H.H. Jang, H.-S. Ro, W. Park, E.L. Madsen, C.O. Jeon, Molecular and bio-
chemical characterization of 3-hydroxybenzoate 6-hydroxylase from Polaromonas
naphthalenivorans CJ2, Appl. Environ. Microbiol. 73 (2007) 5146–5152.
[54] M.A. Adams, V.K. Singh, B.O. Keller, Z. Jia, Structural and biochemical characteriza-
tion of gentisate 1,2-dioxygenase from Escherichia coli O157:H7, Mol. Microbiol.
61 (2006) 1469–1484.
[55] N.-Y. Zhou, J. Al-Dulayymi, M.S. Baird, P.A. Williams, Salicylate 5-hydroxylase
from Ralstonia sp. strain U2: a monooxygenase with close relationships to and
shared electron transport proteins with naphthalene dioxygenase, J. Bacteriol.
184 (2002) 1547–1555.
[56] E. Grund, B. Denecke, R. Eichenlaub, Naphthalene degradation via salicylate and gen-
tisate by Rhodococcus sp. strain B4, Appl. Environ. Microbiol. 58 (1992) 1874–1877.
[57] P. Di Gennaro, P. Terreni, G. Masi, S. Botti, F. De Ferra, G. Bestetti, Identification and
characterization of genes involved in naphthalene degradation in Rhodococcus
opacus R7, Appl. Microbiol. Biotechnol. 87 (2010) 297–308.
[58] T. Hiromoto, S. Fujiwara, K. Hosokawa, H. Yamaguchi, Crystal structure of 3-
hydroxybenzoate hydroxylase from Comamonas testosteroni has a large tunnel for
substrate and oxygen access to the active site, J. Mol. Biol. 364 (2006) 878–896.
[49] D. Ishiyama, D. Vujaklija, J. Davies, Novel pathway of salicylate degradation by
Streptomyces sp. strain WA46, Appl. Environ. Microbiol. 70 (2004) 1297–1306.