Substrate recycling scheme for tetrachloro-p-benzoquinone using bilirubin oxidase and NADH: Application for pentachlorophenol assay

Article abstract of DOI:10.1021/es980978k

A novel assay for tetrachloro-p-benzoquinone (TCBQ), the main oxidation product of pentachlorophenol (PCP), was developed using bilirubin oxidase (BOX) in the presence of excess NADH. TCBQ was easily and rapidly reduced by NADH to 1,4-tetrachlorohydroquinone (TCHQ), which was then recycled back to TCBQ by the enzyme. BOX exhibited no reactivity toward NADH while its catalytic activity for the oxidation of TCHQ was very high. Under an optimized condition (250 μM NADH, 0.3 U/mL BOX, and 25 mM sodium phosphate at pH 5.5), the rate of NADH consumption determined by measuring the absorbance decrease at 340 nm yielded a detection limit for TCBQ of 110 nM. Fluorescence detection of the NADH using a lower enzyme concentration (0.1 U/mL) with excitation and emission wavelengths of 345 and 450 nm, respectively, allowed for a TCBQ detection limit of 30 nM. PCP was oxidized to TCBQ with high yield using bis(trifluoroacetoxy)iodobenzene in 0.05 M trichloroacetic acid. Coupling this oxidation reaction to the BOX/NADH assay attained PCP detection limits of 170 and 50 nM using absorbance and fluorescence measurements, respectively. When tested on PCP-contaminated soil samples, the BOX assay compared very well with HPLC measurements. Chlorophenols constitute a major group of pollutants having been widely used as wood preservatives, pesticides, and herbicides. They are also formed as byproducts of many industrial activities including chlorination of potable water and paper bleaching. A novel assay for tetrachloro-p-benzoquinone (TCBQ), the main oxidation product of pentachlorophenol (PCP), was developed using bilirubin oxidase (BOX) in the presence of excess NADH. TCBQ was easily and rapidly reduced by NADH to 1,4-tetrachlorohydroquinone (TCHQ), which was then recycled back to TCBQ by the enzyme. BOX exhibited no reactivity toward NADH while its catalytic activity for the oxidation of TCHQ was very high. Under an optimized condition (250 μM NADH, 0.3 U/mL BOX, and 25 mM sodium phosphate at pH 5.5), the rate of NADH consumption determined by measuring the absorbance decrease at 340 nm yielded a detection limit for TCBQ of 110 nM. Fluorescence detection of the NADH using a lower enzyme concentration (0.1 U/mL) with excitation and emission wavelengths of 345 and 450 nm, respectively, allowed for a TCBQ detection limit of 30 nM. PCP was oxidized to TCBQ with high yield using bis(trifluoroacetoxy)iodobenzene in 0.05 M trichloroacetic acid. Coupling this oxidation reaction to the BOX/NADH assay attained PCP detection limits of 170 and 50 nM using absorbance and fluorescence measurements, respectively. When tested on PCP-contaminated soil samples, the BOX assay compared very well with HPLC measurements.

Full text of DOI:10.1021/es980978k

Environ. Sci. Technol. 1999, 33, 796-800
SCHEME 1
Substrate Recycling Scheme for
Tetrachloro-p-benzoquinone Using
Bilirubin Oxidase and NADH:
Application for Pentachlorophenol
Assay
D A V I D C YB U L S K I , † , ‡ K E I T H B . M A L E , †
J E N O M . S C H A R E R , ‡
M U R R A Y M O O - YO U N G , ‡ A N D
J O H N H . T . L U O N G * , † , ‡
ronmental samples for the investigation of animal and human
exposure.
Biotechnology Research Institute, National Research Council
Canada, Montreal, Quebec H4P 2R2, Canada, and
Department of Chemical Engineering, University of Waterloo,
Waterloo, Ontario N2L 3G1, Canada
The use of enzymes in colorimetric assays or biosensor
construction is one popular approach to achieve fast and
accurate measurement, which could be useful for field
screening during site characterization and remediation
activities. There are several examples of biosensors and
enzymatic procedures for phenolic compounds using a
variety of enzymes such as tyrosinase (3-13), horseradish
peroxidase (14-16), chloroperoxidase, and to a lesser extent,
laccase (17). The tyrosinase or laccase-based sensors are only
capable of measuring phenol, catechol, 4-chlorophenol,
2-amino-4-chlorophenol, and 2,4-dichlorophenol (3-6, 13).
In addition, the detection limits of such enzyme-based
sensors for the chlorophenols are not satisfactory (ca. 200
nM). 4-Chlorophenol is also the only chlorophenol detectable
by the combined horseradish peroxidase and laccase sensors.
Similarly, the substrate recycling assay using tyrosinase and
NADH reported by Brown et al. (13) was only useful for
phenol, catechol, 4-chlorophenol, and p-cresol. Although a
combined chloroperoxidase/ glucose oxidase-based sensor
is able to detect several chlorophenol congeners, the system
is not applicable for analysis of PCP and some other highly
chlorinated phenols such as 3,4,5-trichlorophenol and 2,3,5,6-
tetrachlorophenol (18, 19). In brief, depending upon the
position and number of chlorines on the aromatic ring,
various oxidation compounds can be generated in such a
reaction: quinones, coupling products, or polymers. The
oxidation product of PCP or these two chlorophenols was
reported to undergo polymerization (18, 19), i.e., it was neither
electroactive nor mediated glucose/ glucose oxidase reaction.
According to our knowledge, an enzymatic assay for PCP has
not yet been developed.
A novel assay for tetrachloro-p-benzoquinone (TCBQ), the
main oxidation product of pentachlorophenol (PCP),
was developed using bilirubin oxidase (BOX) in the presence
of excess NADH. TCBQ was easily and rapidly reduced
by NADH to 1,4-tetrachlorohydroquinone (TCHQ), which was
then recycled back to TCBQ by the enzyme. BOX exhibited
no reactivity toward NADH while its catalytic activity
for the oxidation of TCHQ was very high. Under an optimized
condition (250 µM NADH, 0.3 U/mL BOX, and 25 mM
sodium phosphate at pH 5.5), the rate of NADH consumption
determined by measuring the absorbance decrease at
340 nm yielded a detection limit for TCBQ of 110 nM.
Fluorescence detection of the NADH using a lower enzyme
concentration (0.1 U/mL) with excitation and emission
wavelengths of 345 and 450 nm, respectively, allowed for
a TCBQ detection limit of 30 nM. PCP was oxidized to
TCBQ with high yield using bis(trifluoroacetoxy)iodobenzene
in 0.05 M trichloroacetic acid. Coupling this oxidation
reaction to the BOX/NADH assay attained PCP detection
limits of 170 and 50 nM using absorbance and fluorescence
measurements, respectively. When tested on PCP-
contaminated soil samples, the BOX assay compared very
well with HPLC measurements.
This study describes a substrate recycling assay using
bilirubin oxidase together with NADH for the determination
of tetrachloro-p-benzoquinone (TCBQ), an oxidation product
of PCP and 2,3,5,6-tetrachlorophenol (Scheme 1). It was
reasoned that TCBQ could be reduced to its tetrachloro-
hydroquinone form (TCHQ) by NADH since quinones have
been reported to oxidize NADH to produce NAD+ in our
laboratory (13). This TCHQ species was then reoxidized by
bilirubin oxidase, a copper-containing enzyme, that has been
shown to be capable of oxidizing several small organic
compounds in the presence of oxygen (20, 21). As a result,
a cycle for the substrate was established where one TCBQ
molecule can consume many NADH molecules.
Introduction
Chlorophenols constitute a major group of pollutants having
been widely used as wood preservatives, pesticides, and
herbicides (1, 2). They are also formed as byproducts of many
industrial activities including chlorination of potable water
and paper bleaching. In addition to their extreme toxicity,
several chlorophenols are highly carcinogenic and considered
priority pollutants by the U.S. Environmental Protection
Agency (EPA). Pentachlorophenol (PCP), the most acutely
toxic of all chlorophenol congeners, is a widespread con-
taminant of groundwater and surface water worldwide.
Therefore, there is great interest in developing simple, fast,
and accurate methods for measuring PCP in various envi-
The substrate recycling assay could be quantitated by
measuring the consumption of NADH (µMNADH/ min) versus
the amount of TCBQ present in the sample using absorbance
or fluorescence, the latter allowing for greater sensitivity.
* Corresponding author phone: (514)496-6175; fax: (514)496-6265;
e-mail: john.luong@nrc.ca.
The resulting slopes of such calibration plots (µMNADH
/
† National Research Council Canada.
‡ University of Waterloo.
µMTCBQ‚min) serve as the referenced values for determining
the concentration of TCBQ in unknown samples. The assay
9
7 9 6 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 5, 1999
10.1021/es980978k CCC: $18.00
1999 Am erican Chem ical Society
Published on Web 01/15/1999

protocol was then demonstrated for analysis of PCP in
contaminated soil by combining this substrate recycling
scheme with a chemical oxidation technique using bis-
(trifluoroacetoxy)iodobenzene (22, 23). It should be noted
that this high-yield reaction was recently reported for the
oxidation of various chlorinated phenols including PCP under
mild conditions.
peak height expressed as the coefficient of variation (standard
deviation/ mean) was ca. 5% of all measurements. The
determination of PCP in contaminated soils was performed
in triplicate.
Results and Discussion
Substrate Cycling Assay using Bilirubin Oxidase and NADH.
The experimental data confirmed that NADH was easily and
rapidly oxidized to NAD+ by tetrachloro-p-benzoquinone
(TCBQ) by following the NADH absorbance decrease at 340
nm. A series of experiments was then conducted to search
for an enzyme with the capability of rapidly converting TCHQ
to TCBQ to complete the recycling scheme. As a prerequisite,
the enzyme must not react with NADH. Tyrosinase, laccase,
and chloroperoxidase were initially tested in the substrate
recycling assay in view of their widespread use in biosensor
construction for phenolic compounds. However, none of
these enzymes were suitable as tyrosinase and laccase
resulted in a very unsatisfactory detection limit (ca. 1000
nM) and chloroperoxidase together with hydrogen peroxide
was not applicable since this combination also oxidized
NADH to NAD+.
Bilirubin oxidase (BOX) was then tested in the recycling
assay since both catechol and hydroquinone were reported
to be substrates of BOX with o- and p-benzoquinone as their
respective reaction products (21). Notice that the primary
known function of this enzyme is the oxidation of bilirubin
to biliverdin (20). BOX turned out to be the best with a high
catalytic ability to convert TCHQ to TCBQ in the presence
of oxygen whereas it exhibited no reactivity toward NADH
in solution. When used in the recycling assay, the NADH
consumption rate (µMNADH/ min) obtained by BOX was 10-
fold higher than that of tyrosinase or laccase and twice that
of chloroperoxidase. A series of experiments was then
performed to estimate the maximal velocity (Vmax) and the
Michaelis-Menten constant (Ks) of BOX with respect to
TCHQ as a substrate, and the oxidation of TCHQ to TCBQ
was monitored at 290 nm. Notice that TCHQ was prepared
by reacting TCBQ with zinc powder using a protocol described
by Saby et al. (23). The resulting sample was filtered to remove
the powder from TCHQ. The Hanes plot (9 data points,
correlation coefficient of 0.97) provided a value of 2.02
mMTCBQ/ min and 0.057 mM for Vmax and Ks, respectively.
In view of the instability of NADH at a pH below 5, the
BOX assay was tested over a pH range of 5-8 (25 mM
phosphate buffer) using four different chlorinated p-ben-
zoquinones: monochlorobenzoquinone (MCBQ), 2,5-dichlo-
robenzoquinone (2,5-DCBQ), 2,6-DCBQ, and TCBQ. Ac-
cording to Saby et al. (22, 23), MCBQ is the oxidized product
of 4-chlorophenol (2-CP), 3-CP, 2,4-dichlorophenol (DCP),
or 3,4-DCP. The oxidation of either 2,5-DCP or 2,4,5-
trichlorophenol (TCP) will produce 2,5-DCBQ whereas 2,6-
DCBQ is the oxidized product of the following chloro-
phenols: 2,6-DCP, 3,5-DCP, 2,4,6-TCP, and 3,4,5-TCP. The
oxidized product of 2,3-DCP or 2,3,4-TCP (i.e., 2,3-DCBQ)
and trichlorobenzoquinone (the oxidized product of 2,3,5-
TCP, 2,3,6-TCP, or 2,3,4,6-tetrachlorophenol) were not com-
mercially available during the course of this study.
Experimental Section
Materials. Bilirubin oxidase (EC 1.3.3.5 from Myrothecium
verrucaria, lyophilized powder: 15-65 U/ mg of protein),
laccase (EC 1.10.3.2 from Rhus vernificera, crude powder: 50
U/ mg of solid), tyrosinase (EC 1.14.18.1 from mushroom:
2000 U/ mg of solid), chloroperoxidase (EC 1.11.1.10 from
Caldariomyces fumago, a purified suspension: 1000-2000
U/ mg of protein), and NADH were obtained from Sigma (St.
Louis, MO). Chlorophenols, m onochlorobenzoquinone
(MCBQ), 2,5-dichlorobenzoquinone (2,5-DCBQ), 2,6-DCBQ,
tetrachloro-1,4-benzoquinone (TCBQ), bis(trifluoroacetoxy)-
iodobenzene (BTFAIB), and the remaining chemicals were
purchased from Aldrich (Milwaukee, WI). All other materials
used were of the highest grade available, and deionized water
was used in all stock solutions and buffers. Certified PCP-
contaminated soil samples were obtained from Resource
Technology Corp. (Laramie, WY). Other PCP-contaminated
soil samples were obtained from several wood-preserving
plants in the Quebec region.
Bilirubin Oxidase Reaction with Chloro-p-benzoquino-
nes. The substrate recycling assay was monitored by following
the rate of NADH consumption during the recycling assay.
This was done spectrophotometrically by measuring the
absorbance decrease at 340 nm (Beckman DU 640 spectro-
photometer; Beckman, Fullerton, CA). The NADH consump-
tion rate was also followed using fluorescence measurements
with a Gilford Fluoro-IV spectrofluorometer (Gilford; Oberlin,
OH). The excitation and emission wavelengths used were
345 and 450 nm, respectively, and the detector PMT voltage
was set to +750 V. Unless otherwise indicated, all the assays
were performed in 25 mM sodium phosphate at pH 5.5.
Pentachlorophenol (PCP) Oxidation. A protocol reported
recently (22, 23) with a slight modification was used for the
oxidation of PCP to tetrachloro-p-benzoquinone. The reac-
tion was performed in 50 mM at pH 1.4 instead of in 100 mM
trichloroacetic acid (TCA) at pH 1.1, using 500 µM BTFAIB
since TCA was found later to inhibit the enzymatic reaction
using bilirubin oxidase. The reduction of TCA from 100 to
50 mM, however, did not affect the PCP oxidation yield if the
reaction was allowed to proceed for 1 h at room temperature
with agitation and light protection. Notice that at pH 3 (24),
the reaction was not complete until 2 h and the conversion
yield from PCP to TCBQ was less than 70%. The reaction was
stopped by the addition of 500 µM hydrogen peroxide, which
destroyed any remaining BTFAIB. Notice that BTFAIB (23)
displays an irreversible cyclic voltammogram with one
reduction peak at +0.14 V (glassy carbon vs Ag/ AgCl, 0.1 M
tartrate buffer, pH 3.5). In the presence of 500 µM hydrogen
peroxide, the reduction peak disappeared, an indication of
complete BTFAIB destruction.
HPLC Determ ination of TCBQ and PCP. HPLC deter-
mination of TCBQ and PCP was performed using a Waters
system (Waters, Milford, MA) consisting of a model 590 pump,
a WISP 710B autosampler, and a model 481 LC spectropho-
tometer. The column used was a Supelcosil LC-PAH column
(Supelco; Bellefonte, PA, 150 mm × 4.6 mm) containing C18
bonded silica (5 µm). The HPLC was run isocratically at a
flow rate of 1 mL/ min with a 50:50 mixture of acetonitrile
and water containing 59 µL of 85% phosphoric acid/ L as a
mobile phase. The calibration was established between 2
and 50 µM PCP. The detection limit was ca. 1 µM, and the
PCP retention time was about 20 min. Reproducibility of the
With respect to the NADH consumption rate, TCBQ assay
was optimal at pH 5.5, and for the other chlorobenzoquinones
(CBQs), it was in the range of pH 7-7.5 (Figure 1). A significant
decrease in the assay response for TCBQ at pH above 6.5 was
not completely unexpected due to the instability of this
chloroquinone (22, 23). As also shown in this figure, the assay
response toward TCBQ at pH 5.5 was at least 20-40-fold
higher than that of other chlorobenzoquinones, which
indicated a very high substrate activity of the enzyme for
TCHQ. This type of behavior showed promise for the
development of an assay specific for TCBQ at pH 5.5, an
oxidation product of PCP. On the basis of the results obtained
9
VOL. 33, NO. 5, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 7 9 7

FIGURE 3. Effect of BOX concentration on the assay response: (9)
1 µM TCBQ, (2) 5 µM TCBQ, and (]) 10 µM TCBQ. Each data point
is an average of the triplicates.
FIGURE 1. Effect of pH on the assay response for (2)10 µM MCBQ,
())10µM 2,5-DCBQ, (b)10µM 2,6-DCBQ, and (9)5µM TCBQ. Each
data point is an average of the triplicated results. 1 µM NADH )
0.0062 AU (absorption coefficient of NADH at 340 nm is 6200 M-1
cm-1).
excessive background noise. In this case, the NADH con-
centration chosen was 2 µM with a PMT voltage of +750 V.
There was an increase in the response assay with increasing
BOX concentrations up to 0.1 U/ mL. Beyond this enzyme
level, the response was no longer dependent on the enzyme
level (figure not shown). Therefore, a BOX level of 0.1 U/ mL
was chosen for fluorescence detection, and in an effort to
conserve enzyme, a BOX level of 0.3 U/ mL instead of 0.5
U/ mL was chosen for the absorbance assay.
Assays for p-Chlorobenzoquinones. The four different
chlorobenzoquinones were tested in the assay at the condi-
tions optimized for TCBQ (pH 5.5, 250 µM NADH, 0.1-0.3
U/ mL BOX) using both absorbance and fluorescence detec-
tion (Figure 4A,B). For the absorbance experiments, the slope
of the TCBQ calibration line at the optimized conditions
yielded an assay response (slope of the straight line) of 2.22
µMNADH/ µMTCBQ‚min. With respect to the absorption coef-
ficient of NADH (6200 M-1 cm-1), 1 µM NADH is equivalent
to 0.0062 AU at 340 nm. When tested in the assay, the
experimental data reconfirmed that MCBQ, 2,5-DCBQ, and
2,6-DCBQ displayed responses of 2.0%, 1.4%, and 7.1% of
the value for the TCBQ calibration line, respectively. Such a
result thus confirmed that the assay procedure was much
more sensitive and selective to TCBQ.
FIGURE 2. Effect of NADH concentration on the assay response:
(9) 1 µM TCBQ, ()) 5 µM TCBQ, and (2) 10 µM TCBQ. The
determination was performed in triplicate.
and due to the great interest in the detection and monitoring
of PCP, the subsequent study focused on the development
of an optimal assay procedure for TCBQ.
Using fluorescence, the TCBQ calibration plot gave a
response (slope of the straight line) of 1.14 × 10-2 RFU-
(relative fluorescence units)/ nMTCBQ‚min. In this case, MCBQ,
2,5-DCBQ, and 2,6-DCBQ gave responses of 12.3%, 2.9%,
and 20%, respectively, of the TCBQ value. In view of the
assay response for MCBQ and 2,5-DCBQ, the presence of
4-CP, 2-CP, 3-CP, 2,4-DCP, 3,4-DCP, 2,5-DCP, or TCP in the
sample should not significantly affect the reliability of the
recycling procedure for analysis of PCP. However, 2,6-DCP,
3,5-DCP, 2,4,6-TCP, and 3,4,5-TCP (2,6-DCBQ is an oxidized
product of these chlorophenols) when present at an equiva-
lent amount to PCP may contribute up to an error of 20%
to the final reading.
The effect of NADH concentrations was investigated using
both absorbance and fluorescence measurements. In the case
of the absorbance measurements (TCBQ ) 0.05-10 µM), the
assay response increased with increasing NADH concentra-
tion up to 200-250 µM (Figure 2). Beyond this level, the
assay response was independent of NADH concentration.
An NADH concentration higher than 250 µM could not be
used and/ or recommended because this would have given
absorbance readings beyond the limit of linear detection of
most spectrophotometers. Therefore, an NADH level of 250
µM was chosen in all subsequent experiments, allowing the
assay to progress for a sufficient amount of time (1-20 min,
depending on the TCBQ concentration) without the reaction
rate being affected by any variation in the NADH concentra-
tion. The effect of increasing the BOX concentration for the
absorbance experiment was also investigated and illustrated
in Figure 3. Above 0.5 U/ mL BOX, there was no further
increase in the assay response for the absorbance experi-
ments.
The detection limits of the assay for the absorbance and
fluorescence techniques were determined using their re-
spective pooled sample standard deviations. This was ac-
complished by setting the detection limit response to three
times the pooled standard deviation of the samples. These
standard deviations were 0.103 µMNADH/ min for the absor-
bance studies and 0.083 RFU/ min for the fluorescence
experiments. These values were substituted into the regres-
sion equations for the standard curves for TCBQ and solved
to give detection limits of 110 and 30 nM TCBQ for the
absorbance and fluorescence techniques, respectively.
Selection of the NADH concentration for the fluorescence-
based experiments was performed on the basis of the
maximum NADH concentration that would allow for the
maximum possible detector PMT voltage without giving
9
7 9 8 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 5, 1999

FIGURE 4. Assay of different CBQs using BOX. (A) Absorbance detection: (2) MCBQ, (]) 2,5-DCBQ, (b) 2,6-DCBQ, and (9) TCBQ; slope
) 2.223 ( 0.135 at 95% confidence interval; intercept ) 0.121; R 2) 0.984. (B) Fluorescence detection: (2) MCBQ, (]) 2,5-DCBQ, (b)
2,6-DCBQ, and (9) TCBQ; slope ) 1.14 × 10-2 ( 1.6 × 10-3 at 95% confidence interval; intercept ) 0.54 × 10-2; R 2) 0.989.
Oxidation of Pentachlorophenol to TCBQ Using BTFAIB.
In view of the difficulties encountered with the enzymatic/
photochemical oxidation of the chlorinated phenols (22, 23),
the possibility of using a chemical procedure was explored
as a complementary approach. Of the available methods for
PCP oxidation, the use of BTFAIB has been shown to be
rapid and reliable and to give high yields of TCBQ (22, 23).
Therefore, the oxidation of PCP (0.01-10 µM) was carried
out in 0.05 M trichloroacetic acid using 500 µM BTFAIB. After
allowing the reaction to proceed for 1 h with light protection
and agitation, the reaction was stopped by the addition of
500 µM hydrogen peroxide. HPLC analysis showed that at
least 95% of the PCP had been converted to TCBQ. It was
also observed that both trichloroacetic acid and hydrogen
peroxide exhibited extreme inhibitory effects on the activity
of the enzyme and the stability of NADH. Therefore, it was
necessary to remove these two chemicals after the oxidation
of PCP to TCBQ. This cleanup procedure was facilitated
through the use of Sep-Pak C18 cartridges (Waters; Milford,
MA). The converted samples (20 mL) were loaded onto
cartridges prewashed with acetonitrile and trichloroacetic
acid. The cartridges contain very hydrophobic octadecylsilane
packing; therefore, they will allow polar compounds such as
hydrogen peroxide and TCA in the sample to pass through.
The cartridges were then washed extensively with 25 mM
NaH2PO4, pH 1.9, to remove any residual trichloroacetic acid
and hydrogen peroxide. The bound TCBQ was then eluted
from the cartridges using 1 mL of acetonitrile and made up
to 20 mL with 25 mM NaH2PO4, pH 5.5. These eluted samples
were expected to be free from both TCA and hydrogen
peroxide and were then used directly in the BOX assay.
PCP Determ ination with the BOX Assay. The BTFAIB-
treated PCP samples behaved much like the TCBQ samples
in the BOX/ NADH assay. The only difference between the
two was that the converted PCP samples gave about 30%
lower responses in the assay than their equivalent amount
of TCBQ. The decrease in response could come from TCBQ
loss at two points: First, during the BTFAIB oxidation there
would not have been 100% conversion of PCP to TCBQ, or
some of the TCBQ might have been further oxidized to
another compound. As a matter of fact, about 5% PCP was
reported to be oxidized to 1,2-tetrachlorobenzoquinone (23).
HPLC analysis obtained in this study confirmed that about
95% of the PCP had been converted to TCBQ as mentioned
previously. Second, some of the TCBQ could have been lost
during the C18 cartridge cleanup step as verified by HPLC
for analysis of the converted samples before and after the
cleanup step. Repeated experiments confirmed the loss of
30% PCP during the combined oxidation and cleanup steps.
The assay was calibrated for PCP by measuring the NADH
consumption rate (µMNADH/ min) versus the amount of PCP
present in the sample. The resulting slopes of such calibration
plots (µMNADH/ µMPCP‚min) serve as the referenced values for
determining the PCP concentration in unknown samples.
Consequently, the detection limits for PCP using absorbance
and fluorescence detection were established in the same
manner as for TCBQ. The resulting limits were 170 and 50
nM for absorbance and fluorescence, respectively.
Analysis of Contam inated Soil Sam ples. Several PCP-
contaminated soil samples including two certified soil
samples (150-3000 ppm, µg of PCP/ g of soil) were tested
using the BOX assay by absorbance measurement. PCP was
extracted from the soil samples using 10 mM NaOH. One
gram of soil was contacted with 50 mL of the NaOH for 4 h
at room temperature with strong agitation and light protec-
tion. Once the extraction was complete, the samples were
centrifuged, and the sediment was removed. The remaining
liquid was stored at 4 °C with light protection. When the soil
9
VOL. 33, NO. 5, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 7 9 9

recycling assay, with detection limits of 170 and 50 nM for
absorbance and fluorescence detection methods, respec-
tively. The recycling assay has proven applicable for deter-
mining PCP in contaminated soils as the results obtained
agreed well with the HPLC standard procedure. Together
with a miniaturized spectrophotometer or fluorometer, this
analytical procedure may be attractive for field analysis during
site characterization and remediation activities. To improve
the detection limit as well as reproducibility, bilirubin oxidase
can be immobilized on an optical fiber to form an enzyme-
based optical fiber biosensor. Efforts toward this goal are
currently under way in our laboratory.
Acknowledgments
This work was supported by the Natural Sciences and
Engineering Research Council of Canada Strategic Research
Grant 0193178.
Literature Cited
FIGURE 5. Comparison of HPLC detection and the BOX assay for
PCP (µM). Each data point is an average of the triplicates. HPLC
vs BOX: (0,0/0,0/12.2,8.7/22.3,20.4/12.7,9.8/18,17/217,211/81,93/87,94).
(1) Ahlborg, U. J.; Thunberg, T. M. CRC Crit. Rev. Toxicol. 1980, 7,
1-35.
(2) Lindstrom, K.; Nordin, J. J. Chromatogr. 1976, 128, 13-26.
(3) Li, J.; Chia, L. S.; Goh, N. K.; Tan, S. N. Anal. Chim. Acta 1998,
362, 203-211.
(4) Deng, Q.; Dong, S. Anal. Chem. 1995, 67, 1357-1360.
(5) Makower, A.; Eremenko, A. V.; Streffer, K.; Wollenberger, U.;
Scheller, F. W. J. Chem. Technol. Biotechnol. 1996, 65, 39-44.
(6) Wang, J.; Fang, L.; Lopez, D. Analyst 1994, 119, 445-458.
(7) Kotte, H.; Grundig, B.; Vorlop, K.-D.; Strehiltz, B. Anal. Chem.
1995, 67, 65-70.
samples were used in the BTFAIB oxidation step, the various
materials extracted from the soils did not have any effect on
the conversion of PCP to TCBQ. Therefore, it was not
necessary to have an additional sample cleanup step, greatly
facilitating and reducing the amount of time required for
sample preparation.
Comparing the BOX/ NADH assay results for the soil
samples with the HPLC analysis showed that there was very
good agreement between the two methods (Figure 5), with
no significant difference between the means of the two
methods used (R ) 0.05). The PCP contamination level of 8
out of 10 extract samples tested ranged from 10 to 220 µM.
The two other soil samples (nos. 9 and 10) contained no PCP
as confirmed by both the enzymatic assay and the HPLC
analysis. The certified samples (samples 7 and 8 from
Resource Technology Corp., reference value 1109-1742 ppm
of PCP/ g of soil) gave an identical value of 93.5 µM with
recycling assay, which is equivalent to 1150 ppm of PCP/ g
of soil. The values obtained by HPLC for these two samples
were 1006 and 1078 ppm, respectively. A plot of the recycling
assay and HPLC results against one another (figure not
shown) gave a slope of 0.998 ( 0.062 at 95% confidence
interval with a correlation coefficient (R 2) of 0.994, validating
the applicability of the substrate recycling assay for measuring
PCP in contaminated soil samples.
In brief, a substrate recycling assay for tetrachloro-p-
benzoquinone was developed and optimized using the
recycling of TCBQ between reactions with NADH and
bilirubin oxidase. At pH 5.5 and with excess NADH (250 µM),
absorbance monitoring of the NADH allowed for a TCBQ
detection limit of 110 nM, while the fluorescence monitoring
method gave a detection limit of 30 nM. PCP was effectively
converted to TCBQ using BTFAIB with a high yield under
mild conditions. This allowed for detection of PCP by the
(8) Campanella, L.; Sammartino, M. P.; Tomassetti, M. Sens.
Actuators B 1992, 7, 383-388.
(9) Bondakar, M.; Vilchez, J. L.; Mottola, H. A. Electroanal. Chem.
1989, 266, 47-55.
(10) Svitel, J.; Miertus, S. Environ. Sci. Technol. 1998, 32, 828-832.
(11) Ortega, F.; Dominguez, E.; Burestedt, E.; Emnueus, J.; Gorton,
L.; Marko-Varga, G. J. Chromatogr. A 1994, 675, 65-78.
(12) Lutz, E. S. M.; Dominguez, E. Electroanalysis 1996, 8, 117-123.
(13) Brown, S. R.; Male, K B.; Luong, J. H. T. Anal. Biochem. 1994,
222, 131-139.
(14) Ruzgas, T.; Emneus, J.; Gorton, L.; Marko-Varga, G. Anal. Chim.
Acta 1995, 311, 245-253.
(15) Schubert, F.; Saini, S.; Turner, A.; Scheller, F. Sens. Actuators B
1992, 7, 408-411.
(16) Wang, J.; Wu, L.-H.; Angnes, L. Anal. Chem. 1991, 63, 2993-
2994.
(17) Papkovsky, D.; Ghindilis, A. L.; Kurochkin, I. N. Anal. Lett. 1993,
26, 1505-1518.
(18) Saby, C.; Luong, J. H. T. Electroanalysis 1998, 10, 7-11.
(19) Zhao, S.; Luong, J. H. T. Anal. Chim. Acta 1996, 327, 235-242.
(20) Murao, S.; Tanaka, N. Agric. Biol. Chem. 1982, 46, 2031-2034.
(21) Tanaka, N.; Murao, S. Agric. Biol. Chem. 1982, 46, 2499-2503.
(22) Saby, C.; Luong, J. H. T. Chem. Commun. 1997, 13, 1197-1198.
(23) Saby, C.; Male, K. B.; Luong, J. H. T. Anal. Chem. 1997, 69, 4324-
4330.
(24) Male, K. B.; Saby, C.; Luong, J. H. T. Anal. Chem. 1998, 70, 4134-
4339.
Received for review September 23, 1998. Revised manuscript
received December 3, 1998. Accepted December 3, 1998.
ES980978K
9
8 0 0 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 5, 1999