Pentachlorophenol hydroxylase, a poorly functioning enzyme required for degradation of pentachlorophenol by sphingobium chlorophenolicum

Article abstract of DOI:10.1021/bi300261p

Several strains of Sphingobium chlorophenolicum have been isolated from soil that was heavily contaminated with pentachlorophenol (PCP), a toxic pesticide introduced in the 1930s. S. chlorophenolicum appears to have assembled a poorly functioning pathway for degradation of PCP by patching enzymes recruited via two independent horizontal gene transfer events into an existing metabolic pathway. Flux through the pathway is limited by PCP hydroxylase. PCP hydroxylase is a dimeric protein that belongs to the family of flavin-dependent phenol hydroxylases. In the presence of NADPH, PCP hydroxylase converts PCP to tetrachlorobenzoquinone (TCBQ). The k_cat for PCP (0.024 s ^-1) is very low, suggesting that the enzyme is not well evolved for turnover of this substrate. Structure-activity studies reveal that substrate binding and activity are enhanced by a low pK_a for the phenolic proton, increased hydrophobicity, and the presence of a substituent ortho to the hydroxyl group of the phenol. PCP hydroxylase exhibits substantial uncoupling; the C4a-hydroxyflavin intermediate, instead of hydroxylating the substrate, can decompose to produce H₂O₂ in a futile cycle that consumes NADPH. The extent of uncoupling varies from 0 to 100% with different substrates. The extent of uncoupling is increased by the presence of bulky substituents at position 3, 4, or 5 and decreased by the presence of a chlorine in the ortho position. The effectiveness of PCP hydroxylase is additionally hindered by its promiscuous activity with tetrachlorohydroquinone (TCHQ), a downstream metabolite in the degradation pathway. The conversion of TCHQ to TCBQ reverses flux through the pathway. Substantial uncoupling also occurs during the reaction with TCHQ.

Full text of DOI:10.1021/bi300261p

Article
pubs.acs.org/biochemistry
Pentachlorophenol Hydroxylase, a Poorly Functioning Enzyme
Required for Degradation of Pentachlorophenol by Sphingobium
chlorophenolicum
Klara Hlouchova,†,‡ Johannes Rudolph,§,‡ Jaana M. H. Pietari,∥ Linda S. Behlen,†,‡
and Shelley D. Copley*,†,‡
†Department of Molecular, Cellular and Developmental Biology, University of Colorado Boulder, Boulder, Colorado 80309, United
States
‡Cooperative Institute for Research in Environmental Sciences, University of Colorado Boulder, Boulder, Colorado 80309, United
States
§Department of Chemistry and Biochemistry, University of Colorado Boulder, Boulder, Colorado 80309, United States
∥Exponent Engineering and Scientific Consulting, Bellevue, Washington 98007, United States
S
* Supporting Information
ABSTRACT: Several strains of Sphingobium chlorophenolicum
have been isolated from soil that was heavily contaminated with
pentachlorophenol (PCP), a toxic pesticide introduced in the
1930s. S. chlorophenolicum appears to have assembled a poorly
functioning pathway for degradation of PCP by patching enzymes
recruited via two independent horizontal gene transfer events into
an existing metabolic pathway. Flux through the pathway is limited
by PCP hydroxylase. PCP hydroxylase is a dimeric protein that
belongs to the family of flavin-dependent phenol hydroxylases. In
the presence of NADPH, PCP hydroxylase converts PCP to
tetrachlorobenzoquinone (TCBQ). The kcat for PCP (0.024 s−1) is
very low, suggesting that the enzyme is not well evolved for
turnover of this substrate. Structure−activity studies reveal that
substrate binding and activity are enhanced by a low pKa for the phenolic proton, increased hydrophobicity, and the presence of a
substituent ortho to the hydroxyl group of the phenol. PCP hydroxylase exhibits substantial uncoupling; the C4a-hydroxyflavin
intermediate, instead of hydroxylating the substrate, can decompose to produce H2O2 in a futile cycle that consumes NADPH.
The extent of uncoupling varies from 0 to 100% with different substrates. The extent of uncoupling is increased by the presence
of bulky substituents at position 3, 4, or 5 and decreased by the presence of a chlorine in the ortho position. The effectiveness of
PCP hydroxylase is additionally hindered by its promiscuous activity with tetrachlorohydroquinone (TCHQ), a downstream
metabolite in the degradation pathway. The conversion of TCHQ to TCBQ reverses flux through the pathway. Substantial
uncoupling also occurs during the reaction with TCHQ. Photo used with permission from Johannes Rudolph.
entachlorophenol (PCP) is designated as a priority
pollutant by the U.S. Environmental Protection Agency.
licum with that of Sphingobium japonicum, a closely related
Sphingomonad that degrades lindane,12 suggests that the PCP
degradation pathway was assembled in a patchwork fashion
using enzymes acquired via two different horizontal gene
transfer events in addition to enzymes inherited from a
progenitor of both species.13 The pathway (see Figure 1)
begins with hydroxylation of PCP by PCP hydroxylase (PcpB)
to form TCBQ.14 TCBQ reductase (PcpD) reduces TCBQ to
TCHQ,14 which is converted to 2,6-dichlorohydroquinone
(2,6-DCHQ) via two successive dehalogenation reactions
catalyzed by TCHQ dehalogenase (PcpC).15,16 An Fe(II)-
P
PCP was introduced as a pesticide in the 1930s and was once
one of the most widely used biocides in the United States,
although current use is restricted to industrial wood treatment
applications.1 PCP and its metabolites tetrachlorobenzoqui-
none (TCBQ) and tetrachlorohydroquinone (TCHQ) are
highly toxic chemicals that damage DNA by forming adducts2
or causing strand breaks.3 PCP also uncouples oxidative
phosphorylation.4,5
Several bacteria that mineralize PCP have been isolated from
soil and groundwater contaminated with PCP;6−10 the best
a
studied of these is Sphingobium chlorophenolicum L-1. Despite
its ability to mineralize PCP, S. chlorophenolicum degrades PCP
slowly and cannot tolerate high concentrations of PCP.11
Comparison of the whole genome sequence of S. chloropheno-
Received: February 26, 2012
Revised: April 4, 2012
Published: April 6, 2012
© 2012 American Chemical Society
3848
dx.doi.org/10.1021/bi300261p | Biochemistry 2012, 51, 3848−3860

Biochemistry
Article
Figure 1. Pathway for degradation of PCP in S. chlorophenolicum. Abbreviations: PcpB, PCP hydroxylase; PcpD, TCBQ reductase; PcpC, TCHQ
dehalogenase; PcpA, 2,6-dichlorohydroquinone dioxygenase; HGT, horizontal gene transfer.
Scheme 1
dependent extradiol dioxygenase (PcpA) then cleaves 2,6-
DCHQ.17−19 Further steps leading to β-ketoadipate have been
proposed.20 Flux through the pathway appears to be limited by
the activity of PCP hydroxylase, as PCP accumulates to high
levels in cells metabolizing PCP, while the concentration of
downstream metabolites is much lower.21
PCP hydroxylase is a member of a family of flavin-dependent
phenol hydroxylases. Two members of this family, phenol
hydroxylase and p-hydroxybenzoate hydroxylase, have been
extensively characterized.22−32 The chemical events in the
catalytic cycle are intimately coupled to conformational changes
that control access of the substrate to the active site and
interactions of the flavin with NADPH and the substrate, which
occur at separate locations. The catalytic cycle begins with a
conformational change that opens a solvent channel to the
active site, allowing substrate to bind. Once substrate is bound,
deprotonation of the hydroxyl group triggers movement of the
flavin to the “out” conformation, where it can be reduced by
NADPH. The reduced flavin reacts with O2 to form C4a-
hydroperoxyflavin, which moves to the “in” conformation and
transfers a hydroxyl group to the substrate. The catalytic cycle is
completed by tautomerization of the initial hydroxylation
product and elimination of H2O from C4a-hydroxyflavin, which
returns the flavin to the oxidized state, and release of products,
which requires a conformational change to the “open” form. On
the basis of this body of work, we have proposed a comparable
mechanism for PCP hydroxylase14 (Scheme 1). Scheme 1
shows PCP binding in the phenolate form based on its
predominant form in solution (pKa = 4.5) and our studies of
the binding of chlorinated phenols that suggest that
3849
dx.doi.org/10.1021/bi300261p | Biochemistry 2012, 51, 3848−3860

Biochemistry
Article
deprotonation of the hydroxyl group increases the affinity for
the active site (see further below).
purified on a Ni-NTA column according to the QIAexpression-
ist protocol (Qiagen). The pooled PCP hydroxylase fractions
were dialyzed against 20 mM HEPES (pH 7.0) containing 100
mM NaCl, 1 mM TCEP, and a 1-fold molar excess of FAD
(with respect to the concentration of PCP hydroxylase).
Following concentration to 2 mL using Amicon centrifugal
filters (Millipore), the protein preparation was injected onto a
Superdex 200 column (GE Healthcare). Fractions containing
dimeric PCP hydroxylase were pooled and dialyzed as
described above. Finally, PCP hydroxylase was dialyzed against
50 mM potassium phosphate (pH 7.0) for 4 h. The final
protein sample was concentrated to approximately 20 mg/mL
(300 μM based upon monomer concentration), flash-frozen in
liquid nitrogen, and stored at −70 °C. Analysis by sodium
dodecyl sulfate−polyacrylamide gel electrophoresis (SDS−
PAGE) indicated that the enzyme was homogeneous. All
concentrations for the enzyme are given in terms of monomer
concentration.
Here we characterize the poor catalytic abilities of PCP
hydroxylase and address the ongoing debate over the identity of
the product formed from PCP by PCP hydroxylase. PCP is
converted to TCBQ, as depicted in Scheme 1. The enzyme
turns over substrate at a very low rate; the kcat is only 0.024 s−1.
Furthermore, it is unable to achieve complete coupling of the
formation of C4a-hydroperoxyflavin and the hydroxylation of
PCP. Consequently, excessive amounts of O2 and NADPH are
consumed during substrate turnover, and H2O2 is produced
because of elimination of H2O2 from C4a-hydroperoxyflavin
(see Scheme 1). The spectrum of the enzyme−PCP complex
indicates that the flavin is largely in the “out” conformation,
which may account for the poor coupling efficiency. In addition
to its poor ability to catalyze hydroxylation of PCP, the enzyme
also does not exclude structurally similar molecules from the
active site. Most detrimentally, TCHQ, a downstream
metabolite, serves as a substrate for PCP hydroxylase.
Hydroxylation of TCHQ regenerates TCBQ, an earlier
metabolite in the pathway. The reaction is largely uncoupled
in the presence of TCHQ, causing additional nonproductive
consumption of NADPH and generation of H2O2.
The concentration of flavin in purified PCP hydroxylase was
determined in 8 M urea from the absorbance at 446 nm (ε446
=
10600 M−1 cm−1).33 The protein concentration was determined
in 8 M urea by subtracting the contribution of the flavin to the
absorbance at 280 nm, which was determined experimentally to
be 1.94 (A446), from the total absorbance at 280 nm. An ε280
value of 68500 M−1 cm−1 for the apoprotein was determined
from the absorbance of a solution of a protein whose
concentration was known on the basis of amino acid analysis.
Dependence of PCP Hydroxylase Activity on Flavin
Content. Samples of PCP hydroxylase with varying FAD
contents were prepared by incubation of the enzyme (300 μM)
at room temperature for 1 h with varying concentrations of
FAD (from 0 to 300 μM) in 20 mM HEPES (pH 7.0)
supplemented with 1 mM TCEP. The protein was then
recovered by gel filtration using Illustra NAP-5 columns filled
with Sephadex G-25 (GE Healthcare). Further removal of flavin
could be accomplished by repeating the gel filtration step.
High-Performance Liquid Chromatography (HPLC)
Analysis of the Products Formed by PCP Hydroxylase.
PCP hydroxylase was incubated with substrate (PCP, TCP, or
TCHQ) and NADPH in 300 μL of 50 mM potassium
phosphate (pH 7.0) at 25 °C. Ascorbate (25 μM) was typically
added to prevent nonenzymatic oxidation of TCHQ. Control
reactions were conducted in the absence of enzyme and/or
NADPH. When the formation of TCBQ was being measured,
β-ME (6 mM) was included in the reaction mixture to trap
TCBQ as 2,3,5,6-tetrakis[(2-hydroxyethyl)thio]-1,4-hydroqui-
none (THTH) (ε350 = 2175 M−1 cm−1). The reactions were
quenched, typically after 1 min, by vortexing with 300 μL of
ethyl acetate containing 50 μM p-nitrophenol (internal
standard). Following centrifugation at 16000g for 1 min, the
ethyl acetate layer (150 μL) was recovered. The solvent was
removed by evaporation and the residue redissolved in an 80%
aqueous TFA (0.1%)/20% acetonitrile mixture for analysis of
THTH or a 40% aqueous TFA (0.1%)/60% acetonitrile
mixture for analysis of TCBQ, TCHQ, PCP, and TCP. The
reaction components were resolved on a Zorbax SB-C18 (4.6
mm × 150 mm) HPLC column (Agilent Technologies) using a
10 to 80% gradient of acetonitrile containing 0.1% aqueous
TFA over 23 min. The detector was set at 295 nm for detection
of TCBQ, PCP, and TCP, 310 nm for detection of TCHQ and
p-nitrophenol, and 350 nm for detection of THTH.
Quadruplicate aliquots were analyzed for each time point,
and each reaction was repeated at least three times.
MATERIALS AND METHODS
■
Chemicals. PCP, TCP, 2,4,6-TriCP, 3,4,5-TriCP, 2,6-DCP,
2-CP, TCBQ, and 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB)
were purchased from Sigma-Aldrich. 2,3,4,5-TCP and 2,3,4,6-
TCP were purchased from Supelco. TCHQ was purchased
from Acros Organics. pQE30 was obtained from Qiagen. All
other reagents were purchased from common commercial
sources. Chlorinated phenols except for PCP were dissolved in
DMSO prior to dilution into assay buffer. PCP was dissolved in
1−2 equiv of 100 mM NaOH and then diluted in 50 mM
potassium phosphate (pH 7.0). TCHQ solutions were
prepared either in DMSO or in 1−2 equiv of 100 mM
NaOH, followed by dilution into 50 mM potassium phosphate
(pH 7.0) containing a 1.5-fold molar excess of ascorbate to
prevent oxidation to TCBQ. TCBQ was dissolved in acetone.
Purification of PCP Hydroxylase. The gene encoding N-
terminally His6-tagged PCP hydroxylase was cloned into
pQE30 and introduced into Escherichia coli M15 (pReP4).
Cells were streaked onto a plate containing LB medium,
kanamycin (25 μg/mL), and ampicillin (100 μg/mL). All
subsequent growth media contained 25 μg/mL kanamycin and
100 μg/mL ampicillin. After overnight growth at 37 °C, 5 mL
of LB was inoculated with a single colony from the plate. The
culture was grown overnight with shaking at 37 °C and
transferred into 1 L of LB. The cells were grown with shaking at
37 °C until the OD600 reached approximately 0.6. The culture
was then chilled on ice to 25 °C. Expression of PCP
hydroxylase was induced by addition of isopropyl β-D-
thiogalactopyranoside (IPTG) to a final concentration of 0.1
mM. The cells were grown with shaking overnight at 25 °C and
were harvested by centrifugation for 10 min at 5000g and 4 °C.
Cell pellets were stored at −20 °C.
All protein purification steps were performed at 4 °C. The
harvested cells were resuspended in lysis buffer (4 mL/g of
cells) consisting of 50 mM HEPES (pH 8.0), 300 mM sodium
chloride, 10 mM imidazole, and 1 mM TCEP supplemented
with Sigma protease inhibitor cocktail for use in purification of
His-tagged proteins. The cells were lysed by two passes through
a French pressure cell at 12000 psi. PCP hydroxylase was
3850
dx.doi.org/10.1021/bi300261p | Biochemistry 2012, 51, 3848−3860

Biochemistry
Article
Determination of Dissociation Constants by Absorb-
ance Spectroscopy. PCP hydroxylase (7.1 μM) in 50 mM
potassium phosphate (pH 7.0) was titrated with various
chlorinated phenols (1 μM to 5 mM) dissolved in exactly the
same solution of protein. Absorbance spectra (350−550 nm)
were recorded at room temperature using an HP8453 diode
array spectrophotometer at more than 20 concentrations of
each ligand, and the spectrum of PCP hydroxylase alone was
subtracted from each. Dissociation constants were calculated
using eq 1 from the absorbance differences at 388 nm for PCP,
absorption at 340 nm was monitored every 5−10 s for 5−10
min in a Varioskan plate reader (Thermo Electron Corp.). The
linear part of each reaction rate was corrected for the rate of
NADPH consumption observed in the absence of the effector.
The Michaelis−Menten equation (eq 2), an equation including
substrate inhibition (eq 3), or an equation describing
cooperative substrate kinetics (eq 4) was used to perform a
weighted least-squares fit of the data.
V
max[S]
v
=
=
obs
KM + [S]
(2)
⎡
ΔAbs = KD + [E]tot + [S]
⎣
Vmax[S]
v
[S]2
obs
⎤
−
(KD + [E]tot + [S])2 − 4[E]tot[S] /(2[E]tot)
Km + [S] +
⎦
Ki
(3)
(4)
(1)
V
max[S]n
v
=
obs
n
K50 + [S]n
TCP, and 2,6-DCP and the absorbance differences at 444 nm
for 2,4,6-TriCP, 3,4,5-TriCP, and 3,5-DCP.
Spectroscopic Assays of Enzyme Activity. Formation of
the hydroxylated product and depletion of NADPH were
measured following ethyl acetate extraction of the reaction
mixture. PCP hydroxylase (0.15−5 μM) was incubated with
substrate (1−1000 μM) and NADPH (160 μM) in 1 mL of 50
mM potassium phosphate (pH 7.0) at room temperature.
Reactions with PCP and TCHQ were performed in the
presence of 6 mM β-ME to trap the TCBQ product as THTH.
Reactions with other phenols carrying a chlorine at position 4
were performed in the presence of ascorbate (5 mM) to trap
the benzoquinone product as a hydroquinone. Aliquots (0.2
mL) were removed at varying time points up to 10 min
(corresponding to <20% depletion of substrate), and the
reactions were quenched when the mixtures were vortexed with
ethyl acetate (0.2 mL). Following centrifugation at 16000g for 1
min, the ethyl acetate layer was removed and the concentration
of the product was measured by UV−vis spectroscopy. For
each substrate listed, the extinction coefficients for the
corresponding product were determined by allowing a known
amount of substrate to react to completion in the presence of
excess enzyme and NADPH: for TCP, ε310 = 7700 M−1 cm−1;
Hydrogen Peroxide Assays. H2O2 was quantified using a
colorimetric assay based upon oxidation of Fe(II) in complex
with xylenol orange.34 The assay mixture contained 250 μM
(NH4)2Fe(II)(SO4)2, 25 mM H2SO4, 100 μM xylenol orange,
and 100 mM sorbitol. The mixture was prepared immediately
before use by combining 1 volume of 25 mM (NH4)2Fe(II)-
(SO4)2 in 2.5 M H2SO4 with 100 volumes of 100 mM sorbitol
containing 100 μM xylenol orange. A 50 μL aliquot of sample
or H2O2 standard was added to 1 mL of assay mixture and the
solution incubated for 30−45 min at room temperature. The
increase in absorbance at 560 nm was measured with a Hewlett-
Packard 8453 diode array spectrophotometer. A standard curve
was determined prior to the measurements. None of the
components in standard reaction mixtures (PCP, TCBQ,
TCHQ, NADPH, or NADP) except for PCP hydroxylase at
high concentrations (>3 μM) interfered with the assay. All
reactions in the presence of substrate or effector (100 μM) and
NADPH (160 μM) were performed in triplicate, and the
average rates were determined from at least three separate
experiments. Depletion of H2O2 by reaction with PCP
hydroxylase was assessed in reaction mixtures that contained
0 or 5 μM PCP hydroxylase and 50 μM H2O2 in 50 mM
potassium phosphate (pH 7.0). Samples were removed at 1 min
intervals and analyzed for H2O2 as described above.
for 2,4,6-TriCP, ε305 = 4502 M−1 cm−1; for 2,6-DCP, ε303
=
4384 M−1 cm−1. (At these wavelengths, the corresponding
substrate had no significant absorption.) The extinction
coefficient of THTH (ε350) was determined to be 2175 M−1
cm−1 on the basis of the absorbance of product formed by
reacting a known amount of TCBQ with β-ME. The extinction
coefficient of TCHQ was confirmed to be 7700 M−1 cm−1 on
the basis of the absorbance of a solution prepared by weighing a
known amount of substrate. NADPH remaining in the aqueous
layer was quantified by UV−vis spectroscopy (ε340 = 6200 M−1
cm−1). All reactions were performed more than three times
with time points taken in triplicate. For determinations of KM,
the assay volume was increased to 5 mL, and 1 mL aliquots
were extracted with 0.22 mL of ethyl acetate to allow
concentration of the product in the ethyl acetate layer. The
concentration of each substrate or uncoupler was varied over a
range equal to or greater than 0.2−5 times the apparent KM.
For effectors, depletion of NADPH can be monitored at 340
nm because no complicating changes in the spectrum due to
transformations of the phenol and the hydroxylated products
occur. Determinations of the KM for effectors were performed
by monitoring depletion of NADPH in a reaction volume of
200 μL in a 96-well Costar plate (#3635). The decrease in
RESULTS AND DISCUSSION
■
Purification of the Dimeric Form of PCP Hydroxylase.
His6-tagged PCP hydroxylase was purified to apparent
homogeneity (based on SDS−PAGE) on a Ni-NTA column.
However, gel filtration analysis showed three prominent species
with apparent molecular masses of 129, 295, and >500 kDa.
When TCEP or DTT was included throughout the purification,
the 129 kDa species predominated, suggesting that the higher-
molecular mass species arose by covalent cross-linking between
surface-exposed cysteine residues. The molecular mass of His6-
tagged PcpB is 60 kDa. Thus, the 129 kDa species is a dimer, as
observed for other flavin monooxygenases such as phenol
hydroxylase and p-hydroxybenzoate hydroxylase.25,35,36 The
dimeric species is yellow (i.e., loaded with flavin) upon elution
from the gel filtration column, while the other species are nearly
colorless and have no observable activity.
Determination of the Level of Flavin in Purified PCP
Hydroxylase. The amount of flavin in purified samples of PCP
hydroxylase was determined after denaturation with 8 M urea
3851
dx.doi.org/10.1021/bi300261p | Biochemistry 2012, 51, 3848−3860

Biochemistry
Article
to release the flavin into solution. The absorbance of the sample
was measured at 446 nm, and an extinction coefficient of 10.6
mM−1 cm−1 was used to calculate the flavin concentration. The
sample was then diluted 5-fold, and the protein concentration
was determined as described in Materials and Methods. The
flavin content of the purified enzyme was generally 60−80%.
Flavin loading could be increased to 90−100% by incubation of
the enzyme with excess flavin at room temperature in the
presence of 1 mM TCEP (see Materials and Methods). As
shown in Figure 2, activity is proportional to the level of flavin,
Scheme 2
Figure 2. Dependence of PCP hydroxylase activity on FAD.
Comparison of activities of PCP hydroxylase samples differentially
loaded with FAD. Reaction mixtures containing 5 μM PCP
hydroxylase, 100 μM PCP, 6 mM β-ME, and 300 μM NADPH
were incubated for 3 min at room temperature, and THTH was
quantified by HPLC.
than TCHQ (see Scheme 1). In agreement with this proposed
mechanism, we previously demonstrated the formation of
TCBQ from PCP and identified an enzyme (PcpD) that
catalyzes the reduction of TCBQ to TCHQ.14 These results
have been called into question by Su et al., who maintain that
the TCBQ we detected was an artifact resulting from oxidation
of TCHQ and that PCP hydroxylase actually converts PCP to
TCHQ.42 Consequently, we have reinvestigated this issue by
comparing the products formed using both PCP and TCP as
substrates. As shown in Scheme 2, hydroxylation of PCP should
produce TCBQ and hydroxylation of TCP should produce
TCHQ.
Reaction mixtures containing PCP hydroxylase (10 μM),
PCP or TCP (100 μM), and NADPH (300 μM) were
incubated for 1 min to minimize the effects of oxidation of
TCHQ and/or nonenzymatic reduction of TCBQ by NADPH.
Reaction mixtures included β-ME (6 mM) to trap any TCBQ
formed as THTH. We also added 25 μM ascorbate to protect
any TCHQ formed from oxidation. Under these conditions,
reaction of TCBQ with β-ME is faster than reduction by
ascorbate or NADPH, so formation of THTH is fast and
essentially quantitative. Importantly, TCHQ is stable under
these conditions (Figure 3 and Table 1). Control reactions
conducted in the absence of enzyme demonstrated that TCBQ
was quantitatively converted to THTH and that TCHQ was
quantitatively recovered as TCHQ under these conditions (see
Table 1).
with a linear increase in activity up to 1 equiv of flavin/
monomer. All kinetic parameters reported in this paper were
obtained with enzyme that had been fully reconstituted with
flavin.
Development of a New Assay for PCP Hydroxylase
Activity. Measurement of PCP hydroxylase activity is difficult
because the enzyme is very inefficient. Furthermore, the
product formed from PCP (TCBQ) is unstable, reacting with
NADPH and various nucleophiles in the solution. Reduction of
TCBQ by NADPH forms TCHQ, which is itself a substrate for
PCP hydroxylase and is converted back to TCBQ (see further
below). To overcome these difficulties, we developed an assay
procedure that rapidly and quantitatively traps TCBQ as
2,3,5,6-tetrakis[(2-hydroxyethyl)thio]-1,4-hydroquinone
[THTH (Scheme 2)]. The structure of THTH was assigned by
NMR (Figure 1 of the Supporting Information). The adduct is
stable to further manipulation and can be separated from other
substrates and products by HPLC (Figure 3). Using this
procedure with p-nitrophenol as an internal standard, we can
precisely quantify both substrates and products by HPLC and
achieve a mass balance accounting for all of the substrate
initially added to the reaction mixture.
Identification of the Products Formed by Hydrox-
ylation of PCP, TCP, and TCHQ. The initial description of
PCP hydroxylase reported that both PCP and TCP were
converted to TCHQ in the presence of the enzyme and
NADPH.37−39 The reaction with PCP was reported to use two
NADPH molecules per PCP, whereas the reaction with TCP
used only one NADPH molecule per TCP. However, on the
basis of the mechanism of flavin monooxygenases,40,41 we
would predict that PCP would be converted to TCBQ, rather
Incubation of PCP with PCP hydroxylase and NADPH
results in formation of THTH; no TCHQ is detected (see
Figure 3 and Table 1). If β-ME is omitted from the reaction
mixture, some of the PCP that disappears is recovered as
TCHQ, presumably because of the nonenzymatic reduction of
the initial TCBQ product by excess NADPH in the reaction
mixture. However, 28% of the expected product is missing.
Control experiments show that PCP hydroxylase reacts rapidly
3852
dx.doi.org/10.1021/bi300261p | Biochemistry 2012, 51, 3848−3860

Biochemistry
Article
Figure 3. Representative HPLC traces showing the products formed by PCP hydroxylase from PCP, TCP, and TCHQ. Each substrate (100 μM)
was incubated in 50 mM potassium phosphate (pH 7.5) containing 6 mM β-ME and 25 μM ascorbate for 1 min at room temperature. PCP
hydroxylase (10 μM) and NADPH (300 μM) were added as indicated. Because the various compounds absorb optimally at different wavelengths,
signals detected at different wavelengths are overlaid in the chromatograms: green for 295 nm, blue for 310 nm, and red for 350 nm. p-Nitrophenol
was used as an internal standard.
a
Table 1. Analysis of the Products Formed by PCP Hydroxylase from PCP, TCP, and TCHQ
reaction components after incubation for 1 min at room
temperature
reactant
[reactant] (μM)
[PCP hydroxylase] (μM)
[NADPH] (μM)
[β-ME] (mM)
THTH
TCHQ
PCP
TCP
Enzymatic Reactions
PCP
100
100
100
100
100
10
10
10
10
10
300
6
6
6
28
9
2
1
2
70
75
2
1
300
18
75
1
8
TCP
300
10
10
1
1
300
78 11
TCHQ
TCBQ
300
23
77
2
Control Reactions
20
20
20
20
20
6
6
6
6
6
21
17
21
4
3
3
10
10
300
300
TCHQ
20
21
1
1
a
All reactions were performed in the presence of 25 μM ascorbate.
with TCBQ, depleting an up to 5-fold molar excess of TCBQ
within 1 min in the absence of NADPH (see Figure 2 of the
Supporting Information). Thus, we expect that TCBQ
partitions between rapid reactions with PCP hydroxylase and
NADPH in the absence of β-ME.
Incubation of TCP with PCP hydroxylase and NADPH
results in formation of TCHQ as the major product (see Figure
3 and Table 1), demonstrating that TCHQ can be detected if it
is formed during the reaction. However, a small amount of
THTH is also detected. The THTH might have been formed
from TCBQ generated by oxidation of TCHQ during the
incubation. However, control experiments demonstrated that
TCHQ was entirely stable when the reaction was conducted in
the absence of the enzyme. Consequently, we considered the
possibility that TCHQ, which resembles PCP, might serve as a
substrate for PCP hydroxylase. Hydroxylation of TCHQ would
be predicted to form TCBQ (see Scheme 2). Indeed,
incubation of TCHQ with the enzyme in the presence of
NADPH and β-ME for 1 min resulted in 23% conversion to
THTH.
These results conclusively demonstrate that PCP is
converted to TCBQ by PCP hydroxylase. However, we must
3853
dx.doi.org/10.1021/bi300261p | Biochemistry 2012, 51, 3848−3860

Biochemistry
Article
Figure 4. Relationship between reaction velocity and substrate concentration for (A) PCP, (B) TCP, (C) 2,6-DCP, and (D) TCHQ. Reactions were
conducted in 50 mM potassium phosphate (pH 7.0) containing 160 μM NADPH, 6 mM β-ME, 25 μM ascorbate, and PCP hydroxylase.
Table 2. Steady-State Kinetic Parameters for PCP Hydroxylase in the Presence of 160 μM NADPH
substrate
PCP
KM (μM)
kcat (s−1
)
kcat/KM (M−1 s−1
>2.4 × 104
)
a
a
,b
<1
0.024 0.002
c
e
b
TCP
5.6 3.6 (nH = 2)
0.32 0.03
(5.7 1.2) × 104
b
2,6-DCP
TCHQ
12.6 3.1
0.046 0.006
3809 970
f
ab
,
g
nd
>0.015
0.010 0.001
na
d
3,4,5-TriCP
3,5-DCP
59 12
98 13
169 37
102 14
d
0.010 0.0012
a
b
c
d
In the presence of 6 mM β-ME. kcat for the hydroxylation reaction. This is actually a K50 (see eq 4). kcat for the disappearance of NADPH during
e
f
g
the completely uncoupled reaction. nH, Hill coefficient. nd, not determined. na, not applicable.
address the reasons for the discrepancies between our results
and those of Su et al.42 Su et al. did not report a mass balance
for their experiments and did not compare the amount of
TCHQ formed to the amount of PCP consumed. We suspect
that the TCHQ detected in their experiments was due to a
small amount of nonenzymatic reduction of TCBQ to TCHQ
by NADPH in the reaction mixture. They also did not
demonstrate that TCBQ could be recovered after control
reactions conducted in the absence of either the enzyme or
NADPH. Thus, we suspect that TCBQ was not detected
because of reaction with the enzyme and NADPH. Su et al.
argue that our observation of TCBQ as the reaction product
was due to oxidation of TCHQ upon ethyl acetate extraction.
This argument is based upon inspection of the UV spectra of
TCHQ and TCBQ in phosphate buffer and ethyl acetate.
Figure 3 of the Supporting Information shows that TCHQ is
not affected by extraction into ethyl acetate based upon its
chromatographic behavior. The shift in the UV spectrum of
TCHQ observed in ethyl acetate is caused by solvatochromism
(i.e., the change in the absorption band accompanying the
change in solvent polarity), rather than by oxidation of TCHQ
to TCBQ.
Turnover of PCP by PCP Hydroxylase Is Extremely
Slow. We have assayed PCP hydroxylase activity by following
both consumption of PCP and formation of TCBQ (monitored
by capturing TCBQ as THTH). All experiments were
conducted at ambient O2 concentrations, and thus, the derived
kinetic parameters are apparent parameters. The apparent
3854
dx.doi.org/10.1021/bi300261p | Biochemistry 2012, 51, 3848−3860

Biochemistry
Article
kcat,PCP determined by measuring the initial rate of THTH
formation is 0.024 s−1 (see Figure 4A and Table 2). The KM for
PCP is <1 μM and could not be determined accurately because
of the lack of sensitivity in the assay procedure (ε350 = 2175
M−1 cm−1 for THTH) and the high concentration of protein
needed to detect activity (∼1 μM).
The low values of KM,PCP suggest that the enzyme is likely to
be saturated during degradation of PCP. Previous studies21
have shown that cells harvested from medium containing 670
μM PCP contained 4.6 mM PCP, most of which would have
been concentrated in the membrane because of the hydro-
phobicity of PCP. If conversion of PCP to TCBQ is much
slower than its release from the membrane, then PCP would
equilibrate between the membrane and cytoplasm, and its
concentration should be ∼120 μM. Thus, the enzyme would
have been saturated, and flux through the pathway would have
since the 1930s, and thus, PCP hydroxylase likely evolved to
hydroxylate some other substrate. In addition, the reaction
might be slow because of the presence of five electron-
withdrawing substituents on the aromatic ring. The presence of
five fluorine substituents slows the kcat for phenol hydroxylase
by 10-fold from 7.8 s−1 for phenol to 0.7 s−1 for
pentafluorophenol.46 Chlorine substituents are more electron-
withdrawing than fluorine substituents (for Cl, σmeta = 0.37 and
σpara = 0.23; for F, σmeta = 0.34 and σpara = 0.06), so chlorine
substituents would be expected to further diminish the
reactivity of a phenolic substrate.
Formation of the C4a-Hydroperoxyflavin Intermedi-
ate and Hydroxylation of PCP Are Uncoupled. The rates
of formation of TCBQ and disappearance of NADPH during
turnover of PCP should be equal. However, in the presence of
100 μM PCP and 160 μM NADPH, the rate of disappearance
of NADPH significantly exceeds the rate of formation of TCBQ
(see Table 3). These data suggest that formation of the C4a-
hydroperoxyflavin is not tightly coupled to hydroxylation of
PCP (see Scheme 1). Such uncoupling is common in flavin
monooxygenases that hydroxylate phenolic compounds,
particularly when nonphysiological substrates are used.
Uncoupling has been observed with phenol hydroxylase when
p-fluorophenol, p-chlorophenol, resorcinol, p-aminophenol, o-
methylphenol, and even phenol itself are used as substrates.22
In extreme cases, compounds that cannot be hydroxylated but
nevertheless bind to the active site act as “effectors”, facilitating
reduction of the flavin by NADPH, reaction with O2, and
stoichiometric production of H2O2. For example, 5-hydrox-
ypicolinate, 6-hydroxynicotinate, and 3,4-dihydroxybenzoate
serve as effectors for p-hydroxybenzoate hydroxylase.26
To confirm the uncoupling of the PCP hydroxylase reaction,
we measured the rate of H2O2 formation. In the presence of
PCP, the rate of H2O2 formation accounts for the difference
between the rates of NADPH consumption and formation of
TCBQ (see Table 3). Notably, one in five catalytic cycles fails
to hydroxylate PCP and generates H2O2 instead. Even in the
absence of PCP, PCP hydroxylase slowly consumes NADPH
and forms a small amount of H2O2 (see Table 3). The
discrepancy between the rates of NADPH consumption
(0.0034 s−1) and H2O2 formation (0.0014 s−1) in the absence
of substrate or effector can be attributed to depletion of H2O2
by reaction with components of the reaction such as PCP
hydroxylase. [When PCP hydroxylase (5 μM) was incubated
with 50 μM H2O2, more than 40% of the H2O2 disappeared
within 5 min (data not shown).]
Uncoupling at the active site of PCP hydroxylase has not
been previously recognized, although it is consistent with
previously reported findings. Xun et al.37 reported that 1.4
equiv of O2 and 2 equiv of NADPH were consumed for each
PCP consumed. These data were interpreted as evidence of the
conversion of PCP to TCHQ, which was predicted to require 2
equiv of NADPH. Because we know now that PCP hydroxylase
converts PCP to TCBQ, consuming one NADPH, and that
there is significant uncoupling during the reaction, these data
can be explained in a different way. The excess consumption of
O2 was due to the uncoupled reaction, and the excess
consumption of NADPH was due to both the uncoupled
reaction and the nonenzymatic reaction of NADPH with
TCBQ.
been limited by kcat,PCP
.
Previous studies of the kinetic parameters of PCP
hydroxylase have reported widely varying values for the PCP
reaction. Xun and Orser reported a kcat,PCP of 17 5 s−1 and a
KM of 30
7 μM for the native enzyme purified from S.
chlorophenolicum by measuring the initial rate of formation of
TCHQ after reaction for 2 min.37 Given these kinetic
parameters, the substrate would have been depleted within 2
s under the reported experimental conditions (5 μM PCP
hydroxylase, 20−50 μM PCP, and 100 μM NADPH). Thus,
these kinetic parameters may be erroneous.
Wang et al. measured a kcat,PCP of 0.033 s−1 and a KM,PCP of
50 μM by following consumption of PCP using initial PCP
concentrations between 20 and 80 μM.43 This is not a sufficient
range of substrate concentrations to allow an accurate
determination of KM. Furthermore, the GST-tagged enzyme
was isolated as an insoluble apoenzyme and reconstituted with
flavin after solubilization with detergent, purification by
glutathione affinity chromatography, and cleavage of the fusion
protein. Given the uncertainties about the proper refolding of
the protein and the flavin content achieved by this procedure,
we cannot be sure that these parameters accurately reflect those
for the native enzyme.
Nakamura et al. measured kinetic parameters for PCP
hydroxylase using a gas chromatography−mass spectrometry
assay for TCHQ and PCP concentrations ranging between 0.4
and 20 μM.44 They reported a KM,PCP of 0.5 μM and a kcat of
0.009 s−1. The relatively small 3-fold difference in kcat,PCP
between this work and ours could be due to the fact that
activity was assessed by following TCHQ formation; as
demonstrated above, TCHQ is actually a substrate for the
enzyme. Because some of the TCBQ formed by hydroxylation
of TCHQ would have reacted with the enzyme, the rate of
appearance of TCHQ would have been lower than the actual
rate of depletion of PCP.
The kcat,PCP we have measured for PCP hydroxylase is much
lower than those reported for most other flavin monoox-
ygenases. The kcat for phenol hydroxylase with phenol as the
substrate is 9 s−1.22 The kcat for p-hydroxybenzoate hydroxylase
with p-hydroxybenzoate as the substrate is 57 s−1.45 However,
many phenol hydroxylases are promiscuous and hydroxylate
substrate analogues with lower efficiencies. For example, p-
hydroxybenzoate hydroxylase has a kcat of 1.4 s−1 for turnover
of 2,3,5,6-tetrafluoro-p-hydroxybenzoate.22 The slow rate of
substrate turnover by PCP hydroxylase suggests that PCP may
not be the optimal substrate for the enzyme. This is not
unexpected, as PCP has been present in the environment only
Spectroscopic Analysis Suggests That the Flavin Is
Preferentially in the “Out” Conformation in the
Presence of PCP. Extensive structural and spectroscopic
3855
dx.doi.org/10.1021/bi300261p | Biochemistry 2012, 51, 3848−3860

Biochemistry
Article
Figure 5. Overlay of representative difference spectra for the flavin of PCP hydroxylase (7.1 μM) in the presence vs absence of various ligands: PCP
(15 μM, red), TCP (18 μM, black), 2,4,6-TriCP (80 μM, blue), and 3,5-DCP (2500 μM, green).
Figure 6. Perturbation of the flavin spectrum at 388 nm by binding of PCP (A), TCP (B), and 2,6-DCP (C).
characterization of phenol hydroxylase and p-hydroxybenzoate
hydroxylase has shown that the flavin can occupy three different
conformations in the active site.30 When the flavin is in the
“open” conformation, substrate can access the active site. In the
“out” conformation, the flavin can interact with NADPH. In the
“in” conformation, the flavin is positioned near the substrate
and, when the flavin-C4a-hydroperoxide is formed, is optimally
aligned to transfer a hydroxyl group to the substrate. The
differences in solvent exposure in these three positions result in
changes in the absorbance spectrum of the flavin. Typically,
movement of the flavin to the “out” position is correlated with
an increase in absorbance at 444 nm.
PCP hydroxylase, like other flavin monooxygenases, under-
goes ligand-dependent conformational changes as detected by
changes in the spectrum of the enzyme-bound flavin (see
Figure 5). The spectral changes are similar to those observed
previously for phenol hydroxylase36 and p-hydroxybenzoate
hydroxylase,27 particularly for substrate analogues for which the
reaction is partially or fully uncoupled. The tendency for the
flavin to prefer the “out” conformation may be responsible for
the uncoupling that occurs during turnover in the presence of
PCP. After reduction of the flavin and reaction with O2, the
C4a-hydroperoxyflavin must adopt the “in” conformation to
hydroxylate the substrate. If the C4a-hydroperoxyflavin lingers
in the “out” position in the presence of PCP, elimination of
H2O2 will occur more often.
The changes in the flavin spectrum can be used to measure
the dissociation constants for substrates and substrate
analogues. PCP binds extremely tightly, with a dissociation
constant of <50 nM (see Figure 6). Such tight binding of a
substrate is unusual for metabolic enzymes in general and for
this class of enzymes in particular. For p-hydroxybenzoate
hydroxylase, the KD for p-hydroxybenzoate is 43 μM.27 Phenol
hydroxylase shows cooperative substrate binding, with KD
values ranging between 1 and 19 μM.36
Insights into Substrate Binding from Structure−
Activity Relationships. We examined the binding specificity
of PCP hydroxylase by titrating the enzyme with various
chlorinated phenols. Binding of all of the tested compounds
results in increased absorption at 444 nm, consistent with a
3856
dx.doi.org/10.1021/bi300261p | Biochemistry 2012, 51, 3848−3860

Biochemistry
Article
Figure 7. Summary of binding constants, pKa values, and logP values for various chlorinated phenols.
a
Table 3. Rates of the Uncoupled Reaction in the Presence of Substrates and Effectors of PCP Hydroxylase
substrate
PCP
product appearance (s−1
)
NADPH disappearance (s−1
)
NADPH/product
H2O2 appearance (s−1
)
b
b
b
0.020 0.003
0.025 0.001
1.3
0.0070 0.0009
0.0132 0.008
0.035 0.005
TCP
0.076 0.008
0.111 0.018
1.5
c
c
c
TCHQ
0.0048 0.0004
0.030 0.003
6.3
b
b
b
2,4,6-TriCP
3,4,5-TriCP
2,6-DCP
3,5-DCP
3,4-DCP
none
0.0033 0.0002
0.0048 0.003
1.5
0.0012 0.0002
d
e
none
0.012 0.001
0.051 0.004
na
nd
0.060 0.002
none
0.9
0.0011 0.0002
0.0029 0.0006
d
0.0079 0.0016
0.0045 0.0010
0.0034 0.0008
na
d
e
none
na
nd
d
none
na
0.0014 0.0002
a
b
c
All reactions were performed in the presence of 100 μM substrate and 160 μM NADPH. In the presence of 5 mM ascorbate. In the presence of 6
d
e
mM β-ME. na, not applicable. nd, not determined.
shift of the flavin to the “out” position. (See Figure 5 for
examples of difference spectra). However, the difference spectra
vary depending upon the ligand. In particular, a change in
absorbance at 388 nm is observed for only some ligands. The
different spectroscopic signatures do not correlate with any
obvious property of the ligand such as the number or pattern of
substituents, pKa, logP, or solvent accessible surface area. The
observed differences may reflect not only differences in the
equilibrium among the three positions of the flavin but also
smaller differences in the positioning of the flavin that influence
hydrogen bonding and/or hydrophobic interactions between
the flavin and the protein.
The affinities of the various chlorinated phenols for PCP
hydroxylase were determined by titration of the ligands in the
presence of a constant concentration of enzyme (see Figures 6
and 7). Binding affinity appears to depend on three factors: (1)
pKa, (2) hydrophobicity, and (3) the presence of a substituent
ortho to the hydroxyl group. PCP, which has the lowest pKa
and the highest logP, was by far the tightest binding ligand.
Dissociation constants for the tetrachloro compounds TCP and
TCHQ were in the low micromolar range; these compounds
have pKa values similar to that of PCP, but they are less
hydrophobic. Trichloro- and dichlorophenols, which have
higher pKa values and are less hydrophobic, bind even less
tightly. Notably, tri- and dichlorophenols with an ortho
chlorine substituent bound more tightly than isomers with an
ortho hydrogen. These results suggest that the substrate
binding pocket of PCP hydroxylase prefers a hydrophobic
compound with a deprotonated hydroxyl group and a bulky
substituent at the ortho position. Further, because PCP is the
tightest binding substrate, we suspect that the pocket is very
hydrophobic and that substrates that are smaller than PCP
cannot interact optimally with the surrounding protein.
The preference for binding the phenolate form of the
substrate is interesting from a mechanistic standpoint. A proton
must be removed from the hydroxyl group for the aromatic ring
to attack the C4a-hydroperoxyflavin intermediate (see Scheme
1). In phenol hydroxylase and p-hydroxybenzoate hydroxylase,
the substrates have relatively high pKa values, and the proton is
removed by an internal proton transport network.47 In contrast,
PCP is nearly completely ionized at neutral pH; thus, there is
no need for deprotonation to occur at the active site.
Insights into Substrate Reactivity from Structure−
Activity Relationships. We measured steady-state kinetic
parameters for turnover of PCP, TCP, 2,6-DCP, and TCHQ
(see Table 2 and Figure 4), as well as consumption of NADPH
in the presence of two effectors [3,5-DCP and 3,4,5-TriCP (see
further below)]. Surprisingly, the plots for various substrates
are different. The plot for 2,6-DCP is hyperbolic, the plot for
TCP is sigmoidal, and the plot for TCHQ shows substrate
inhibition above a concentration of 40 μM. The physiological
concentration of TCHQ is on the order of 5 μM,21 so the
substrate inhibition observed at high conentrations is likely not
physiologically relevant. However, the substrate inhibition does
complicate the determination of the catalytic constants. The
line shown in Figure 4D applies when the KM is set to be equal
to the Ki and is shown for illustrative purposes only. Reliable
values for KM, Ki, and kcat cannot be determined from these
3857
dx.doi.org/10.1021/bi300261p | Biochemistry 2012, 51, 3848−3860

Biochemistry
Article
Figure 8. Stereo view of the active site of p-hydroxybenzoate hydroxylase showing the substrate, p-hydroxybenzoate (grey), and a C4a-
hydroperoxyflavin (green) modeled in place of the FAD in the “in” position (PDB 1PBE).
data. We can conclude only that the value of kcat must be higher
than the highest observed rate constant (0.015 s−1) and the KM
must be >7 μM.
hydroxyl substituents and the protein. The peroxy substituent
on the flavin approaches the substrate from above the ring. The
data presented above for PCP hydroxylase demonstrate that a
chlorine substituent at the ortho position increases binding
affinity, suggesting that there is a pocket that accommodates the
chlorine at this position. A two-point attachment of a planar
substrate in the active site mediated by interactions with the
negatively charged oxygen and the ortho chlorine substituent
would suffice to fix the orientation of the planar substrate in the
active site. In the absence of a chlorine substituent at the ortho
position, the substrate might twist in the active site, impeding
movement of the C4a-hydroperoxyflavin over the face of the
ring and thus leading to uncoupling. Similarly, bulky
substituents around position 4, to which the hydroxyl group
is transferred, increase the extent of uncoupling, presumably by
impeding the approach of the C4a-hydroperoxyflavin. The
predicted effects of substituents are consistent throughout the
set of chlorinated phenols. 2,6-DCP shows no uncoupling; it
has a chlorine at the ortho position to enforce orientation and
no bulky substituents to hinder hydroxylation. At the other
extreme, 3,5-DCP, 3,4-DCP, and 3,4,5-TriCP are all effectors,
causing complete uncoupling. These compounds have no
chlorine at the ortho position to fix the orientation of the ring
and have steric bulk around position 4. Interestingly, the
turnover of TCHQ is more highly uncoupled than that of PCP
itself, suggesting that the presence of a hydroxyl group is more
problematic than the presence of a chlorine at position 4.
The results of these kinetic studies suggest that TCP is more
reactive than PCP, in agreement with previous studies.38 The
presence of a halogen substituent at C-4 should decrease the
electron density at position 4, which would correspondingly
weaken the HOMO−LUMO interaction required for transfer
of the hydroxyl group. The increased steric bulk of a chlorine
substituent at C-4 should further weaken the HOMO−LUMO
interaction by impeding close approach of the terminal oxygen
of the peroxyl group to the substrate. These results also suggest
that the reason for the poor activity of PCP hydroxylase with
PCP cannot be attributed to the electron-withdrawing effects of
the chlorine substituents, as the kcat for TCP is higher than that
for 2,6-DCP. Indeed, the chlorine substituents may actually
increase the reactivity of the ring by decreasing the pKa of the
hydroxyl group and obviating the need for proton removal.
We examined the extent of uncoupling in the presence of a
larger set of chlorinated phenols using a fixed concentration of
substrate (100 μM) and NADPH (160 μM) (see Table 3).
(Note that only some of the substrates are at saturating
concentrations, so these rate constants are not necessarily
equivalent to kcat.) More complete kinetic analyses to obtain
values for kcat and KM were conducted for depletion of NADPH
in the presence of the effectors 3,4,5-TriCP and 3,5-DCP (see
Table 2). Values for kcat are the same for both effectors and may
reflect the intrinsic rate of elimination of H2O2 when the C4a-
hydroperoxyflavin is unable to approach the phenolic substrate.
Two features of the substrate strongly affect the extent of
uncoupling. (1) A chlorine atom at the ortho position decreases
the extent of uncoupling, and (2) steric bulk at positions 3−5
increases the extent of uncoupling. These factors can be
understood on the basis of the orientation of the substrate and
flavin that must be achieved for hydroxyl transfer. Figure 8
shows the orientation of the flavin, with the peroxy group
modeled in, with respect to the substrate in p-hydroxybenzoate
hydroxylase. (Note that the substrate is hydroxylated at the
position ortho to the hydroxyl group by p-hydroxybenzoate
hydroxylase, rather than at the para position, as occurs at the
active site of PCP hydroxylase.) The substrate is fixed firmly in
the active site by interactions between the carboxylate and
CONCLUSIONS
■
The findings reported here suggest that PCP hydroxylase has
not yet evolved to be an optimal catalyst for hydroxylation of
PCP. The enzyme is rather nonspecific, which is consistent with
its poor catalytic efficiency, as precise orientation of substrates
in the proximity of catalytic groups is critical for catalytic
efficiency. The enzyme is unable to couple formation of C4a-
hydroperoxyflavin and hydroxylation of the substrate efficiently,
leading to excess consumption of NADPH and production of
H2O2. Finally, the enzyme is unable to discriminate between
PCP and TCHQ, a downstream metabolite. When TCHQ is
bound, the uncoupled reaction causes substantial production of
H2O2 at the active site. Furthermore, hydroxylation of TCHQ
forms TCBQ. This reaction generates a futile cycle in which
3858
dx.doi.org/10.1021/bi300261p | Biochemistry 2012, 51, 3848−3860

Biochemistry
Article
(4) Weinbach, E. C. (1954) The effect of pentachlorophenol on
oxidative phosphorylation. J. Biol. Chem. 210, 545−550.
(5) Steiert, J. G., Thoma, W. J., Ugurbil, K., and Crawford, R. L.
(1988) 31P nuclear magnetic resonance studies of effects of some
chlorophenols on Escherichia coli and a pentachlorophenol-degrading
bacterium. J. Bacteriol. 170, 4954−4957.
(6) Saber, D. L., and Crawford, R. L. (1985) Isolation and
characterization of Flavobacterium strains that degrade pentachlor-
ophenol. Appl. Environ. Microbiol. 50, 1512−1518.
(7) Radehaus, P. M., and Schmidt, S. K. (1992) Characterization of a
novel Pseudomonas sp. that mineralizes high concentrations of
pentachlorophenol. Appl. Environ. Microbiol. 58, 2879−2885.
(8) Resnick, S. M., and Chapman, P. J. (1994) Physiological
properties and substrate specificity of a pentachlorophenol-degrading
Pseudomonas species. Biodegradation 5, 47−54.
NADPH is consumed unproductively. Because the KM for PCP
is lower than that for TCHQ, and PCP is present at a higher
concentration in vivo,21 the active site of PCP hydroxylase will
be primarily engaged in converting PCP to TCBQ. The
uncoupling that occurs with PCP, and to a lesser extent with
TCHQ, will deplete NADPH and generate substantial
quantities of H2O2, causing oxidative stress. TCBQ generated
by PCP hydroxylase may damage cytoplasmic proteins if it is
not efficiently captured and converted to TCHQ by TCBQ
reductase. The multiple shortcomings of PCP hydroxylase and
the intrinsic toxicity of TCBQ may explain the inability of S.
chlorophenolicum to degrade high levels of PCP.11
ASSOCIATED CONTENT
* Supporting Information
Methods and supplementary Figures 1−3. This material is
■
(9) Leung, K. T., Cassidy, M. B., Shaw, K. W., Lee, H., Trevors, J. T.,
Lohmeier-Vogel, E. M., and Vogel, H. J. (1997) Pentachlorophenol
biodegradation by Pseudomonas spp. UG25 and UG30. World J.
Microbiol. Biotechnol. 13, 305−313.
S
available free of charge via the Internet at http://pubs.acs.org.
(10) Tiirola, M. A., Mannisto, M. K., Puhakka, J. A., and Kulomaa, M.
S. (2002) Isolation and characterization of Novosphingobium sp. strain
MT1, a dominant polychlorophenol-degrading strain in a groundwater
bioremediation system. Appl. Environ. Microbiol. 68, 173−180.
(11) Dai, M., and Copley, S. D. (2004) Genome shuffling improves
degradation of the anthropogenic pesticide pentachlorophenol by
Sphingobium chlorophenolicum ATCC 39723. Appl. Environ. Microbiol.
70, 2391−2397.
(12) Nagata, Y., Ohtsubo, Y., Endo, R., Ichikawa, N., Ankai, A.,
Oguchi, A., Fukui, S., Fujita, N., and Tsuda, M. (2010) Complete
genome sequence of the representative γ-hexachlorocyclohexane-
degrading bacterium Sphingobium japonicum UT26. J. Bacteriol. 192,
5852−5853.
AUTHOR INFORMATION
Corresponding Author
*Phone: (303) 492-6328. E-mail: shelley.copley@colorado.edu.
Funding
This work was supported by National Institutes of Health
Grant GM078554 to S.D.C. and a CIRES Visiting Fellowship
to K.H.
■
Notes
The authors declare no competing financial interest.
K.H. and J.R. are co-first authors.
(13) Copley, S. D., Rokicki, J., Turner, P., Daligault, H., Nolan, M.,
and Land, M. (2012) The whole genome sequence of Sphingobium
chlorophenolicum L-1: Insights into the evolution of the pentachlor-
ophenol degradation pathway. Genome Biol. Evol. 4, 184−198.
(14) Dai, M., Rogers, J. B., Warner, J. R., and Copley, S. D. (2003) A
previously unrecognized step in pentachlorophenol degradation in
Sphingobium chlorophenolicum is catalyzed by tetrachlorobenzoquinone
reductase (PcpD). J. Bacteriol. 185, 302−310.
(15) Xun, L., Topp, E., and Orser, C. S. (1992) Purification and
characterization of a tetrachloro-p-hydroquinone reductive dehaloge-
nase from a Flavobacterium sp. J. Bacteriol. 174, 8003−8007.
(16) Orser, C. S., Dutton, J., Lange, C., Jablonski, P., Xun, L., and
Hargis, M. (1993) Characterization of a Flavobacterium glutathione-S-
transferase gene involved in reductive dehalogenation. J. Bacteriol. 175,
2640−2644.
ACKNOWLEDGMENTS
We thank Dr. Richard Shoemaker and Dr. Natasha
Chumachenko for assistance with NMR spectroscopy.
■
ABBREVIATIONS
■
β-ME, β-mercaptoethanol; IPTG, isopropyl β-D-thiogalactopyr-
anoside; PCP, pentachlorophenol; TCBQ, tetrachlorobenzo-
quinone; TCHQ, tetrachlorohydroquinone; TCP, 2,3,5,6-
tetrachlorophenol; THTH, 2,3,5,6-tetrakis[(2-hydroxyethyl)-
thio]-1,4-hydroquinone; 2,4,6-TriCP, 2,4,6-trichlorophenol;
3,4,5-TriCP, 3,4,5-trichlorophenol; 2,6-DCP, 2,6-dichlorophe-
nol; 3,5-DCP, 3,5-dichlorophenol; TCEP, tris(2-carboxyethyl)-
phosphine.
(17) Xun, L., Bohuslavek, J., and Cai, M. (1999) Characterization of
2,6-dichloro-p-hydroquinone 1,2-Dioxygenase (PcpA) of Sphingomo-
nas chlorophenolica ATCC 39723. Biochem. Biophys. Res. Commun. 266,
322−325.
(18) Xu, L., Lawson, S. L., Resing, K., Babbitt, P. C., and Copley, S.
D. (1999) Evidence that pcpA encodes 2,6- dichlorohydroquinone
dioxygenase, the ring-cleavage enzyme required for pentachlorophenol
degradation in Sphingomonas chlorophenolica strain ATCC 39723.
Biochemistry 38, 7659−7669.
(19) Ohtsubo, Y., Miyauchi, K., Kanda, K., Hatta, T., Kiyohara, H.,
Senda, T., Nagata, Y., Mitsui, Y., and Takagi, M. (1999) PcpA, which is
involved in the degradation of pentachlorophenol in Sphingomonas
chlorophenolica ATCC39723, is a novel type of ring-cleavage
dioxygenase. FEBS Lett. 459, 395−398.
(20) Cai, M., and Xun, L. (2002) Organization and regulation of
pentachlorophenol-degrading genes in Sphingobium chlorophenolicum
ATCC 39723. J. Bacteriol. 184, 4672−4680.
ADDITIONAL NOTE
The S. chlorophenolicum L-1 strain originally deposited as
ATCC 39723 has lost the ability to degrade PCP. It has been
replaced by ATCC 53874.
■
a
REFERENCES
■
(1) Miller, L. L., Ingerman, L. D., and Singh, M. (2001) Toxicological
profile for pentachlorophenol. Agency for Toxic Substances and
Disease Registry, U.S. Department of Health and Human Services,
Washington, DC.
(2) Lin, P.-H., Nakamura, J., Yamaguchi, S., Upton, P. B., La, D. K.,
and Swenberg, J. A. (2001) Oxidative damage and direct adducts in
calf thymus DNA induced by the pentachlorophenol metabolites,
tetrachlorohydroquinone and tetrachloro-1,4-benzoquinone. Carcino-
genesis 22, 627−634.
(3) Wispriyono, B., Matsuoka, M., and Igisu, H. (2002) Effects of
pentachlorophenol and tetrachlorohydroquinone on mitogen-activated
protein kinase pathways in Jurkat T cells. Environ. Health Perspect. 110,
139−143.
(21) McCarthy, D. L., Claude, A., and Copley, S. D. (1997) In vivo
levels of chlorinated hydroquinones in a pentachlorophenol-degrading
bacterium. Appl. Environ. Microbiol. 63, 1883−1888.
3859
dx.doi.org/10.1021/bi300261p | Biochemistry 2012, 51, 3848−3860

Biochemistry
Article
(22) Neujahr, H. Y., and Kjellen, K. G. (1978) Phenol hydroxylase
from yeast. Reaction with phenol derivatives. J. Biol. Chem. 253, 8835−
8841.
(41) Detmer, K., and Massey, V. (1985) Effect of substrate and pH
on the oxidative half-reaction of phenol hydroxylase. J. Biol. Chem. 260,
5998−6005.
(42) Su, Y., Chen, L., Bandy, B., and Yang, J. (2008) The Catalytic
Product of Pentachlorophenol 4-Monooxygenase is Tetra-chlorohy-
droquinone rather than Tetrachlorobenzoquinone. Open Microbiol. J.
2, 100−106.
(43) Wang, H., Tiirola, M. A., Puhakka, J. A., and Kulomaa, M. S.
(2001) Production and characterization of the recombinant
Sphingomonas chlorophenolica pentachlorophenol 4-monooxygenase.
Biochem. Biophys. Res. Commun. 289, 161−166.
(44) Nakamura, T., Motoyama, T., Hirono, S., and Yamaguchi, I.
(2004) Identification, characterization, and site-directed mutagenesis
of recombinant pentachlorophenol 4-monooxygenase. Biochim.
Biophys. Acta 1700, 151−159.
(45) Entsch, B., and Ballou, D. P. (1989) Purification, properties and
oxygen reactivity of p-hydroxybenzoate hydroxylase from Pseudomonas
aeruginosa. Biochim. Biophys. Acta 999, 313−322.
(46) Peelen, S., Rietjens, I. M. C. M., Boersma, M. G., and Vervoort,
J. (1995) Conversion of phenol derivatives to hydroxylated products
by phenol hydoxylase from Trichosporon cutaneum: A comparison of
regioselectivity and rate of conversion with calculated molecular orbital
substrate characteristics. Eur. J. Biochem. 227, 284−291.
(47) Palfey, B. A., Moran, G. R., Entsch, B., Ballou, D. P., and Massey,
V. (1999) Substrate recognition by “password” in p-hydroxybenzoate
hydroxylase. Biochemistry 38, 1153−1158.
(23) Kalin, M., Neujahr, H. Y., Weissmahr, R. N., Sejlitz, T., Johl, R.,
Fiechter, A., and Reiser, J. (1992) Phenol hydroxylase from
Trichosporon cutaneum: Gene cloning, sequence analysis, and func-
tional expression in Escherichia coli. J. Bacteriol. 174, 7112−7120.
(24) Maeda-Yorita, K., and Massey, V. (1993) On the reaction
mechanism of phenol hydroxylase: New information obtained by
correlation of fluorescence and absorbance stopped flow studies. J.
Biol. Chem. 268, 4134−4144.
(25) Enroth, C., Neujahr, H., Schneider, G., and Lindqvist, Y. (1998)
The crystal structure of phenol hydroxylase in complex with FAD and
phenol provides evidence for a concerted conformational change in
the enzyme and its cofactor during catalysis. Structure 6, 605−617.
(26) Entsch, B., Ballou, D. P., and Massey, V. (1976) Flavin-oxygen
derivatives involved in hydroxylation by p-hydroxybenzoate hydrox-
ylase. J. Biol. Chem. 251, 2550−2563.
(27) Husain, M., and Massey, V. (1979) Kinetic studies on the
reaction of p-hydroxybenzoate hydroxylase. J. Biol. Chem. 254, 6657−
6666.
(28) Husain, M., Entsch, B., Ballou, D. P., Massey, V., and Chapman,
P. J. (1980) Fluoride elimination from substrates in hydroxylation
reactions catalyzed by p-hydroxybenzoate hydroxylase. J. Biol. Chem.
255, 4189−4197.
(29) Wang, J., Ortiz-Maldonado, M., Entsch, B., Massey, V., Ballou,
D., and Gatti, D. L. (2002) Protein and ligand dynamics in 4-
hydroxybenzoate hydroxylase. Proc. Natl. Acad. Sci. U.S.A. 99, 608−
613.
(30) Entsch, B., Cole, L. J., and Ballou, D. P. (2005) Protein
dynamics and electrostatics in the function of p-hydroxybenzoate
hydroxylase. Arch. Biochem. Biophys. 433, 297−311.
(31) Moran, G. R., Entsch, B., Palfey, B. A., and Ballou, D. P. (1996)
Evidence for flavin movement in the function of p-hydroxybenzoate
hydroxylase from studies of the mutant Arg220Lys. Biochemistry 35,
9278−9285.
(32) Ortiz-Maldonado, M., Ballou, D. P., and Massey, V. (2001) A
rate-limiting conformational change of the flavin in p-hydroxybenzoate
hydroxylase is necessary for ligand exchange and catalysis: Studies with
8-mercapto- and 8-hydroxy-flavins. Biochemistry 40, 1091−1101.
(33) Entsch, B., Palfey, B. A., Ballou, D. P., and Massey, V. (1991)
Catalytic function of tyrosine residues in para-hydroxybenzoate
hydroxylase as determined by the study of site-directed mutants. J.
Biol. Chem. 266, 17341−17349.
(34) Jiang, Z. Y., Woollard, A. C., and Wolff, S. P. (1990) Hydrogen
peroxide production during experimental protein glycation. FEBS Lett.
268, 69−71.
(35) Muller, F., Voordouw, G., Van Berkel, W. J. H., Steenis, P. J.,
̈
Visser, S., and Van Rooyen, P. J. (1979) A study on p-hydroxybenzoate
hydroxylase from Pseudomonas fluorescens: Improved purification,
relative molecular mass and amino acid composition. Eur. J. Biochem.
101, 235−244.
(36) Sejlitz, T., and Neujahr, H. Y. (1987) Phenol hydroxylase from
yeast. A model for phenol binding and an improved purification
procedure. Eur. J. Biochem. 170, 343−349.
(37) Xun, L., and Orser, C. S. (1991) Purification and properties of
pentachlorophenol hydroxylase, a flavoprotein from Flavobacterium sp.
strain ATCC 39723. J. Bacteriol. 173, 4447−4453.
(38) Xun, L., Topp, E., and Orser, C. (1992) Diverse substrate range
of a Flavobacterium pentachlorophenol hydroxylase and reaction
stoichiometries. J. Bacteriol. 174, 2898−2902.
(39) Xun, L., Topp, E., and Orser, C. S. (1992) Confirmation of
oxidative dehalogenation of pentachlorophenol by a Flavobacterium
pentachlorophenol hydroxylase. J. Bacteriol. 174, 5745−5747.
(40) Entsch, B., and van Berkel, W. J. (1995) Structure and
mechanism of para-hydroxybenzoate hydroxylase. FASEB J. 9, 476−
483.
3860
dx.doi.org/10.1021/bi300261p | Biochemistry 2012, 51, 3848−3860