Degradation of 3-Chloro-p-toluidine Hydrochloride in Watermelon Bait. Identification and Chemical Characterization of Novel N-Glucoside and Oxopropanimine

Article abstract of DOI:10.1021/jf960044k

Stability of the avicide 3-chloro-p-toluidine hydrochloride (CPTH) in watermelon bait was assessed. When exposed to light, the presence of CPTH accelerated nonenzymic browning (Maillard) reactions and degradation of the watermelon matrix. The addition of potassium metabisulfite appeared to hinder bait degradation. These experiments resulted in the identification and chemical characterization of two novel CPTH derivatives: N-β-D-glucopyranosyl-3-chloro-4-methylaniline and N-(3-chloro4-methylphenyl)-2-oxopropanimine.

Full text of DOI:10.1021/jf960044k

J. Agric. Food Chem. 1996, 44, 3983
−3988
3983
Degr a d a tion of 3-Ch lor o-p-tolu id in e Hyd r och lor id e in Wa ter m elon
Ba it. Id en tifica tion a n d Ch em ica l Ch a r a cter iza tion of Novel
N-Glu cosid e a n d Oxop r op a n im in e
J eanne N. Tawara,*,† J ohn J . J ohnston,‡ and Margaret J . Goodall
USDA APHIS Denver Wildlife Research Center, Building 16, Denver Federal Center,
Denver, Colorado 80225
Stability of the avicide 3-chloro-p-toluidine hydrochloride (CPTH) in watermelon bait was assessed.
When exposed to light, the presence of CPTH accelerated nonenzymic browning (Maillard) reactions
and degradation of the watermelon matrix. The addition of potassium metabisulfite appeared to
hinder bait degradation. These experiments resulted in the identification and chemical characteriza-
tion of two novel CPTH derivatives: N-â-D-glucopyranosyl-3-chloro-4-methylaniline and N-(3-chloro-
4-methylphenyl)-2-oxopropanimine.
Keyw or d s: 3-Chloro-p-toluidine hydrochloride; CPTH; DRC-1339; N-â-D-glucopyranosyl-3-chloro-
4-methylaniline; N-(3-chloro-4-methylphenyl)-2-oxopropanimine; nonenzymic browning; Maillard
reaction
INTRODUCTION
Extensive damage to grapefruits and oranges are
associated with large populations of the great-tailed
grackle (Quiscalus mexicanus) in the Rio Grande Valley
of Texas. Losses of $2.2 million in citrus damage were
reported in 1987 (Glahn et al., 1995). Trapping and
frightening techniques have been used to reduce crop
damage. While trapping provides little relief, the birds
adapt to the scare tactics, and heavy damage persists.
An effective avicide, 3-chloro-p-toluidine hydrochloride
(CPTH, Figure 1), has been registered by the EPA since
1967. USDA Animal Damage Control personnel use
CPTH widely to control localized bird populations.
CPTH is effective on target species including grackles,
starlings, blackbirds, ravens, crows, gulls, and pigeons
(Schafer, 1984).
In Texas, CPTH is formulated in dry dog food baits
(Glahn et al., 1995). These baits, however, are poorly
accepted by the birds during the summer months, when
damage to citrus crops is severe. Attempts to increase
bait acceptance have included formulation of CPTH into
a watermelon matrix. Favorable results from initial
trials led to field studies on the effectiveness and safety
of 0.2% (w/w) CPTH treated watermelon.
F igu r e 1. Structures of CPTH and proposed reaction prod-
ucts.
1963). While ultraviolet (UV) and ionizing radiation
may stimulate enzymic browning of carbohydrates
(Richardson, 1976), the presence of a primary or second-
ary amine accelerates complex nonenzymic browning
(Maillard) reactions of carbohydrates (Hodge and Os-
man, 1976). Because the major saccharides in the
watermelon pulp are sucrose, fructose, and glucose
(Brown and Summers, 1985) and since CPTH is a
primary amine (Figure 1), browning reactions of the
sugars were presumably responsible for the bait deg-
radation. To determine whether enzymic or nonenzymic
processes caused the discoloration, heated and frozen
CPTH-fortified watermelon baits were monitored for
color changes when exposed to light.
During these field studies, fresh watermelon bait was
prepared daily. Within hours of exposure to sunlight
and heat, the bait discolored and developed an unpleas-
ant odor. As the watermelon bait degraded, less ac-
ceptance of the bait was evident. Field personnel
suggested that formation of the brown color and odor
may have been responsible for the decreased acceptance
(Glahn, personal communication). As a result of the
field trials, preliminary laboratory studies with water-
melon baits were performed under simulated field
conditions. In these experiments, browning of the
watermelon bait and decreases in CPTH concentrations
resulted within hours of exposure to light (Goodall,
unpublished report).
Watermelon pulp consists of approximately 93%
water and 6% total carbohydrates (Watt and Merrill,
Sulfur dioxide and sulfites hinder browning due to
nonenzymic processes (McWeeney et al., 1969). Potas-
sium metabisulfite (PMB) has been used extensively as
a fruit preservative. In the presence of metabisulfite,
reducing sugars form sulfonated compounds that retard
the browning process (Burton et al., 1963; Ingles, 1966).
† Present address: Chemistry Department, Colorado
State University, Fort Collins, CO 80523.
‡ Address reprint requests to this author.
S0021-8561(96)00044-1
This article not subject to U.S. Copyright. Published 1996 by the American Chemical Society

3984 J. Agric. Food Chem., Vol. 44, No. 12, 1996
Tawara et al.
ing: gold working electrode; Ag/AgCl reference electrode; 150
mM NaOH filling solution, stainless steel counter electrode,
thin cell gasket; 10 000 nA output range. Separations were
performed using a Dionex CarboPak PA1 column, 4 mm i.d.
× 25 cm, and a Dionex CarboPak PA1 guard column, 3 mm
i.d. × 2.5 cm.
To further test if nonenzymic browning was responsible
for discoloration of the watermelon matrix, additions of
1.5% and 3% (w/w) PMB to the watermelon bait were
assessed as preservatives. In addition, extracts of light-
exposed CPTH-fortified and control watermelon baits
were analyzed by electrospray mass spectrometry (EMS),
gas chromatography-mass spectrometry (GC-MS),
high-performance liquid chromatography (HPLC), high-
performance liquid chromatography-mass spectrometry
(HPLC-MS), and nuclear magnetic resonance (NMR)
spectroscopy for CPTH degradates and Maillard reac-
tion products.
NMR Spectroscopy. NMR spectra were collected on a
1H) and 75 MHz
Bruker-ACE spectrometer operating at 300 (
(13C).
Isola tion s, Syn th eses, a n d Ch em ica l Ch a r a cter iza -
tion s of Novel Com p ou n d s. N-â-D-Glucopyranosyl-3-chloro-
4-methylaniline (Glc-CPT, Figure 1) was isolated from water-
melon extracts by vacuum liquid chromatography (VLC) on
C-18 silica gel with an acetonitrile/water gradient. The VLC
fractions were screened by HPLC. The compound of interest
eluted in the 20-30% acetonitrile fractions. These fractions
were combined and evaporated in vacuo to approximately 1-2
mL. The neat compound was unstable, so the sample was
completely evaporated just prior to collecting NMR spectra.
Glc-CPT was synthesized by the addition of CPTH (0.6
mmol) to R-D-glucose (0.6 mmol) dissolved in 10 mL of distilled
water. The spontaneous formation of Glc-CPT was monitored
by HPLC over a period of several days. The same reaction
procedures were performed with D-fructose/CPTH and sucrose/
CPTH.
N-(3-Chloro-4-methylphenyl)-2-oxopropanimine (CMPP, Fig-
ure 1) was prepared by the dropwise addition of pyruvic
aldehyde (2.9 mmol) to an aqueous solution of CPTH (2.8
mmol). The reaction mixture was stirred for approximately 2
min. PMB (1.4 mmol) dissolved in 1 mL of water was added
slowly, and a golden precipitate formed. After the solution
was stirred for 3 min, the precipitate was filtered off.
CMPP was isolated as a binary mixture of CMPP and CPTH
from the synthetic reaction products. All attempts to isolate
CMPP were hindered by its unstable nature. By GC-MS
analyses, the synthetic CMPP and that present in the water-
melon extracts eluted at the same retention time with nearly
identical fragmentation patterns. The NMR assignments of
CMPP were deduced by subtraction of the resonances due to
CPTH.
METHODS AND MATERIALS
Ch em ica ls a n d Ma ter ia ls. All solvents and chemicals
were reagent grade unless otherwise noted: CPTH, P. M.
Resources, Inc.; D-fructose and R-D-glucose, Aldrich Chemical
Co.; 5-(hydroxymethyl)-2-furfural (HMF), Aldrich; 5-nitro-o-
cresol, TCI America; PMB, Aldrich; pyruvic aldehyde, 40 wt
% solution in water, Aldrich; iron filings, 40 mesh, Fisher
Scientific; C-18 silica gel, Alltech, 30-70 µm; NMR solvents,
Aldrich.
In str u m en ta tion . EMS. EMS analyses were performed
on a Fisons VG Quattro-SQ mass spectrometer in the negative
(0.1% NaOH) and positive (0.1% formic acid) ion modes with
a cone voltage of 45 V. The samples were dissolved in 50:50
acetonitrile/water.
Environmental Chamber. A Revco Model PG-8-1045-A
environmental chamber was used to expose bait samples to
light. UV light, as well as incandescent light bulbs in the
chamber, was engaged during the experiments. When the
chamber was unavailabe for use, a sun lamp (General Electric,
UV bulb, 275 W) and an incandescent bulb (General Electric,
200 W) were suspended above the samples contained in a box
with reflective (aluminum foil lined) sides. The bulbs were
situated approximately 40 cm above the samples. Since all
the samples were sealed in quartz test tubes, the humidity
was not monitored.
The synthesis of 3-hydroxy-4-methylaniline (HMA, Figure
1) was performed following the procedure of Hazlet and
Dornfeld (1944) for the reduction of 5-nitro-o-cresol with the
following modifications. After the mixture was refluxed for 6
h, the reaction was stopped by filtering off the Fe filings and
extracting the benzene solution with 1 N NaOH (4 × 50 mL).
The NaOH extracts were combined and back-extracted with
benzene (2 × 40 mL). Then the pH of the aqueous extracts
was adjusted to approximately 8.5, and the mixture was
extracted with ethyl acetate (3 × 75 mL). The ethyl acetate
extracts were combined, dried over sodium sulfate, and
evaporated, leaving a white solid. No other purification
procedures were necessary.
Sa m p le P r ep a r a tion . The watermelon samples were
prepared by blending the pulp to a watery consistency (60 s
in a Waring blender) prior to treatment by heating, freezing,
or addition of PMB. Heat-treated watermelon samples were
prepared from blended watermelon pulp boiled at 99 °C for 5
min, while frozen watermelon samples were prepared from the
blended pulp frozen for 24 h at -26 °C and thawed at room
temperature. PMB (1.5% and 3% w/w) was added to the
blended pulp and mixed for approximately 2 min with a
spatula. The CPTH (0.2% w/w) was mixed similarly into the
blended pulp immediately prior to commencement of each
experiment. For each of the three treatments, a series of
control samples (without CPTH) were prepared and treated
identically. Triplicate 3.0 g watermelon samples (control and
CPTH fortified) for each time interval were accurately weighed
into quartz test tubes. While exposed to light, the tubes were
aligned horizontally and turned periodically to allow for
uniform exposure of each sample. The measurement of time
intervals was based on the initial addition of CPTH (as opposed
to when samples were placed under light) and included 0, 4,
8, and 24 h. For time zero analyses, the sample preparation
time precluded an analysis reflecting an exact measurement
of the compounds present in the watermelon extracts.
GC-MS. The GC-MS analyses were performed on an
Hewlett-Packard (HP) 5890 gas chromatograph coupled to an
HP 5970 mass selective detector. The GC was equipped with
a DB-1 (J & W Scientific) 30 m × 0.25 mm i.d., 0.25 µm film
thickness capillary column. Operating conditions included the
following GC oven temperature program: 50 °C initial tem-
perature, ramped at 10 °C/min to 160 °C (0.10 min) followed
by a second oven program of 30 °C/min to a final temperature
of 290 °C (2.5 min). Ionization was achieved by electron impact
(70 eV) with an ion source temperature of 280 °C. Analyses
were conducted in the scan mode (m/ z 70-400).
HPLC. The watermelon extracts were analyzed on an HP
1090M HPLC equipped with an UV diode array detector (240
nm), a 250 × 4.6 mm Keystone Deltabond 5 µm C-8 analytical
column, and a Keystone Deltabond C-8 guard column (15 × 3
mm). Injection volumes were 10 µL, the mobile phase was
70:30 acetonitrile/water with a flow rate of 1 mL/min, and the
column was maintained at 40 °C. Chromatogram run times
varied from 12 to 35 min depending on the samples analyzed.
HPLC-MS. HPLC-MS analyses were performed on an HP
1090 liquid chromatograph and an HP 5989 mass spectrom-
eter. The system was equipped with an HP 5990A particle
beam interface and operated in the electron impact (EI)
ionization mode. Mass spectrometry conditions were as fol-
lows: helium nebulizer pressure, 40 psi; ionization potential,
70 eV; column temperature, 40 °C; desolvation chamber
temperature, 40 °C; source temperature, 250 °C; transfer line
temperature, 250 °C. Analyses were conducted in the scan
mode (m/ z 100-400). Chromatographic separation was per-
formed using a 250 × 3 mm Keystone ODS/H column along
with a Keystone Deltabond C-8 guard column (15 × 3 mm).
The mobile phase consisted of methanol/water (65:35), and the
flow rate was 0.5 mL/min.
Ion Chromatography (IC). The IC analyses were performed
on a Dionex 4000i ion chromatograph equipped with a pulsed
amperometric detector. The detector consisted of the follow-

Degradation of CPTH in Watermelon Bait
J. Agric. Food Chem., Vol. 44, No. 12, 1996 3985
F igu r e 2. Amadori rearrangement products of Glc-CPT and proposed Maillard reaction schemes for the formation of HMF and
DDMP in watermelon bait.
F igu r e 3. Glc-CPT and CPTH in untreated watermelon bait
exposed to UV and incandescent light.
F igu r e 4. Glc-CPT and CPTH in heated watermelon bait
exposed to UV and incandescent light.
Watermelon samples were extracted with 15 mL of mobile
phase (70:30 acetonitrile/water). The samples were vortexed
and shaken mechanically for 10 min (Eberbach shaker). The
samples were centrifuged for 5 min, and the supernatants were
filtered (0.45 µm nylon filters) into sample vials for HPLC
analyses.
Gr a p h ica l An a lysis. The unstable nature of Glc-CPT
prevented purification for use as a calibration standard.
Therefore the peak areas for Glc-CPT and CPTH were plotted
to compare the relative variations of each compound in the
samples analyzed (Figures 3-6). Each line represents the
mean of the triplicate values for each time interval plotted.
Quantitation of the loss of CPTH in each sample (Figure 7)
was possible by using a single-point calibration, since the
HPLC UV diode array detector response for CPTH is propor-
tionally linear in the range from 20 to 200 ppm.
F igu r e 5. Glc-CPT and CPTH in frozen watermelon bait
exposed to UV and incandescent light.
RESULTS
Ch em ica l Ch a r a cter iza tion s of Glc-CP T: HPLC-
MS m/ z 303 (M+, <1), 185 (6), 183 (17), 156 (30), 154
(100), 143 (7), 141 (22), 140 (21), 125 (13), 106 (19); EMS
4), 3.72 (dd, 1, J ) 12, 5.4 Hz), 3.88 (dd, 1, J ) 12, 2.1
Hz), 4.70 (d, 1, J ) 8.8 Hz, anomeric H), 6.73 (dd, 1, J
) 8.2, 2.4 Hz, H-6), 6.92 (d, 1, J ) 2.3 Hz, H-2), 7.17 (d,
1, J ) 8.3 Hz, H-5). Structurally, 1H NMR data suggest
the presence of the cyclic form of glucose in Glc-CPT,
1
(ES+) M + H ) 304, 306; (ES-) M - H ) 302, 304; H
NMR (D2O, 4.8 ppm) δ 2.26 (s, 3, CH3), 3.39-3.62 (m,

3986 J. Agric. Food Chem., Vol. 44, No. 12, 1996
Tawara et al.
(HMF, Figure 1) and 2,3-d ih yd r o-3,5-d ih yd r oxy-6-
m eth yl-4H-p yr a n -4-on e (DDMP, Figure 1). A stan-
dard reference solution of HMF and one compound of
interest from the watermelon extracts had virtually
identical retention times and fragmentation patterns.
The fragmentation pattern for the other compound was
nearly identical to that for DDMP (Mills et al., 1970).
DISCUSSION
Dramatic changes in the physical appearances of the
untreated, heated, and frozen CPTH-fortified water-
melon baits occurred during the course of the experi-
ment. After exposure to light for 24 h, the bait samples
were dark brown and the extracts were golden yellow.
Also, unpleasant odors were detected in all these
samples during the extraction procedure. The control
samples which contained no CPTH appeared faded but
not brown. No odor was apparent in the control
samples, and the extracts were clear to slightly yellow
in color. Because the heating and freezing treatments
should have denatured the enzymes responsible for
enzymic processes and the browning occurred only in
CPTH-fortified samples, it appears the discoloration was
due to nonenzymic processes. The rapid discoloration
of the watermelon samples in the presence of CPTH was
consistent with Maillard reactions for monosaccharides
(Ledl and Schleicher, 1990).
F igu r e 6. Glc-CPT and CPTH in 1.5% PMB treated water-
melon bait exposed to UV and incandescent light.
Two major compounds present in the untreated,
heated, and frozen watermelon baits were HMF and
DDMP, reaction intermediates of numerous browning
products from the degradation of glucose and fructose
(Ledl and Schleicher, 1990). In general, the presence
of DDMP serves as an indicator for the occurrence of
the Maillard reaction. This supports the hypothesis
that nonenzymic browning was primarily responsible
for bait degradation.
The watermelon extracts also contained several un-
determined compounds, one of which was a major
component of the extracts and was identified as Glc-
CPT, a novel CPTH derivative (Figure 1). Glc-CPT
formation is typical of the reaction between a primary
amine and glucose (Hodge and Osman, 1976). Although
IC analyses of the watermelon samples revealed the
presence of glucose, fructose, and sucrose as the major
saccharides, only the formation of Glc-CPT was ob-
served. One minor component of the extracts, another
novel CPTH derivative, was identified as CMPP (Figure
1). Presumably CMPP was formed by the reaction
between CPTH and pyruvic aldehyde, a decomposition
product of glucose (Kort, 1990).
Although other studies revealed the formation of
CPTH derivatives in biological or environmental samples,
comparisons between the watermelon extracts and
standard samples of these compounds were unsuccessful
in identification of other undetermined compounds. The
standard samples included the cis and trans isomers of
3,3′-chloro-4,4′-methylazobenzene (azo-CPT, Figure 1,
Primus et al., 1996); N-(3-chloro-4-methylphenyl)aceta-
mide (N-acetyl-CPT, Figure 1; Tweedy et al., 1970; Lee
and Kim, 1978; Engelhardt et al., 1979) and HMA
(Figure 1; Yao and Mill, 1993).
F igu r e 7. Comparison of CPTH recovered from untreated,
heated, and frozen watermelon bait extracts versus the 1.5%
PMB treated extract.
and the large coupling constant of the anomeric proton
(J ) 8.8 Hz) indicates a â-coupling (Collins and Ferrier,
1995) between the glucose and CPT moieties, as indi-
1
cated in Figure 1. The H NMR data are very similar
to that reported for N-(3,4-dichlorophenyl)glucosylamine
(Gareis et al., 1992): 13C NMR (D2O) δ 20.9, 63.5, 72.5,
75.4, 79.3, 79.6, 87.7, 116.1, 117.3, 129.5, 134.4, 137.1,
147.5. Glc-CPT degraded in D2O during NMR analyses.
After the NMR spectra were collected, HPLC analysis
of the sample revealed that Glc-CPT had decomposed
into CPTH and several undetermined compounds. The
HPLC retention times and UV absorption spectra of the
synthesized and the isolated Glc-CPT compounds were
nearly identical.
Ch em ica l Ch a r a cter iza tion s of CMP P : GC-MS
m/ z 195 (M+, 11), 197 (M + 2, 3), 154 (33), 152 (100, M
- COCH3), 127 (24), 125 (78, M - NCHCOCH3), 89 (53);
1H NMR (CDCl3, 7.24 ppm) δ 1.98 (s, 3, acetyl-CH3),
2.39 (s, 3, CH3), 5.14 (s, 1, NdCH), 6.42 (dd, 1, J ) 8.3,
2.3 Hz, H-6), 6.57 (d, 1, J ) 2.3 Hz, H-2), 6.76 (d, 1, J )
8.1 Hz, H-5); 13C NMR (CDCl3, 77.0 ppm) δ 19.4 (CH3),
29.3 (CH3), 78.6 (CH), 118.8 (CH), 120.8 (CH), 131.1,
131.4 (CH), 134.4, 134.7, 201.9 (CO). Instability of
CMPP prevented complete NMR characterization. Since
CMPP was a minor component of the watermelon
extracts, its relative concentrations were not monitored
during the experimental procedures.
Ch em ica l Ch a r a cter iza tion s of HMA: GC-MS
m/ z 123 (M+, 100), 122 (93), 106 (13), 94 (39), 77 (28);
1H NMR (CD3OD, 3.30 ppm) δ 2.04 (s, 3, CH3), 6.17 (dd,
1, J ) 7.8, 2.4 Hz, H-6), 6.21 (d, 1, J ) 2.4 Hz, H-2),
6.77 (d, 1, J ) 7.8 Hz, H-5).
Figures 3-5 depict relative amounts of CPTH and
Glc-CPT in untreated, heated, and frozen CPTH wa-
termelon baits, respectively. The general trend in
Figures 3-5 reveals that within the first 4 h CPTH
levels decreased rapidly, while the Glc-CPT levels
increased. The early appearance of Glc-CPT in the
watermelon extracts implicates the spontaneous forma-
GC-MS analyses of the watermelon extracts con-
firmed the presence of 5-(h yd r oxym eth yl)-2-fu r fu r a l

Degradation of CPTH in Watermelon Bait
J. Agric. Food Chem., Vol. 44, No. 12, 1996 3987
tion of the amino sugar as suggested by Winkler and
Sandermann (1992). The CPTH levels decreased fur-
ther with time, as did the Glc-CPT. Mutarotation at
the anomeric carbon of N-glucosides (Collins and Fer-
rier, 1995) occurs commonly, and the glucose-amine
bond cleaves easily in neutral and slightly acidic aque-
ous media (Pigman, 1957) which may account for the
decrease in Glc-CPT levels. Presumably Glc-CPT de-
graded to HMF, DDMP (Figure 2), and other browning
products.
Additional losses of CPTH depicted in Figures 3-5
may have resulted from CPTH complexation with the
matrix, including lignins and other complex carbohy-
drates (Marco and Novak, 1991). In previous studies
involving aniline-based pesticides, N-glucosides were
found commonly in plants exposed to the aniline (Still,
1968; Yih et al., 1968; Schmidt et al., 1995). As a result
of metabolism in the plants, the pesticides were found
associated with polymeric lignin components of the
plant matrices (Yih et al., 1968; Bartha and Pramer,
1970). The aniline lignin complexes have been referred
to as bound nonextractable residues (Schmidt et al.,
1995), as analyses of these plant materials precluded
recovery of the analyte of interest.
In contrast to the untreated, heated, and frozen baits,
the PMB-treated CPTH watermelon samples, as well
as the respective controls, were salmon pink in color
compared to the original reddish-pink pulp. Further-
more an odor was not detected, and the extracts ranged
from clear to slightly yellow. It appears that the
addition of PMB to the CPTH watermelon bait mini-
mized color changes and production of undesirable
odors.
In the Maillard reaction scheme (Figure 2), Glc-CPT
and the Schiff base appear to be the key intermediates
to the integrity of the watermelon bait, since Amadori
rearrangement of the Schiff base leads to sugar degra-
dation (Ledl and Schleicher, 1990). The observed
preservation of PMB-CPTH watermelon bait may
correspond to hindered formation of the N-glucoside by
PMB through sulfonation of the aldehyde functionality
of glucose (Burton et al., 1963).
browning in fruit (Friedman, 1996) and may be consid-
ered in future studies.
ACKNOWLEDGMENT
We express our gratitude to Sherry Turnipseed, FDA
Animal Drugs Research Center, for HPLC-MS analyses;
Rori Craver, USDA Denver Wildlife Research Center,
for the IC analyses; and Doreen Griffin, USDA Denver
Wildlife Research Center, for technical assistance. We
convey a special acknowledgment to the memory of
J erry Roberts, formerly of USDA Denver Wildlife Re-
search Center, for his assistance with bait preparation.
LITERATURE CITED
Bartha, R.; Pramer, D. Metabolism of acylanilide herbicides.
Adv. Appl. Microbiol. 1970, 13, 317-341.
Brown, A. C., J r.; Summers, W. L. Changes in carbohydrate
concentration during watermelon juice storage. Hortic. Sci.
1985, 20, 896-897.
Burton, H. S.; McWeeny, D. J .; Biltcliffe, D. O. Non-enzymic
browning: the role of unsaturated carbonyl compounds as
intermediates and of SO2 as an inhibitor of browning. J .
Sci. Food Agric. 1963, 14, 911-920.
Collins, P.; Ferrier, R. Monosaccharides, Their Chemistry and
Their Roles in Natural Products; Wiley: West Sussex,
England, 1995.
Engelhardt, G.; Wallnofer, P. R.; Ziegler, W. Degradation of
the phenylamide herbicides Monolinuron and Solan by
Fusarium oxysporum Schlecht. Chemosphere 1979, 10, 725-
730.
Friedman, M. Food browning and its prevention: an overview.
J . Agric. Food Chem. 1996, 44, 631-653.
Gareis, C.; Rivero, C.; Schuphan, I.; Schmidt, B. Plant me-
tabolism of xenobiotics. Comparison of the metabolism of
3,4-dichloroaniline in soybean excised leaves and soybean
cell suspension cultures. Z. Naturforsch., A 1992, 47, 823-
829.
Glahn, J . F. USDA APHIS Denver Wildlife Research Center,
Denver, CO, personal communication, 1995.
Glahn, J . F.; Palacios, J . D.; Garrison, M. V. Evaluation of
methods for reducing great-tailed grackle damage to citrus
in the lower Rio Grande Valley. Unpublished report. Denver
Wildlife Research Center, Denver, CO, 1995.
Goodall, M. J . USDA APHIS Denver Wildlife Research Center,
Denver, CO, unpublished report, 1995.
Hazlet, S. E.; Dornfeld, C. A. The reduction of aromatic nitro
compounds with activated iron. J . Am. Chem. Soc. 1944,
66, 1781-1782.
Hodge, J . E.; Osman, E. M. Carbohydrates. In Principles of
Food Science, Part I, Food Chemistry; Fennema, O. R., Ed.;
Dekker: New York, 1976.
Ingles, D. L. Prevention of non-enzymic browning. Food
Preserv. Q. 1966, 26, 39-44.
Kort, M. J . Reactions of free sugars with aqueous ammonia.
Adv. Carbohydr. Chem. Biochem. 1990, 25, 311-349.
Ledl, F.; Schleicher, E. New aspects of the Maillard reaction
in foods and in the human body. Angew. Chem. 1990, 29,
565-594.
Lee, J . K.; Kim, K. C. A study on the degradation of 3,4-
dichloroaniline by a soil fungus, Chaetomium globosum (part
I) with special emphasis on acetylation. J . Korean Chem.
Soc. 1978, 21, 197-203.
The 1.5% PMB-CPTH watermelon samples showed
a marked difference in CPTH and Glc-CPT levels
(Figure 6) from those observed in untreated, heated, and
frozen CPTH samples (Figures 3-5). Data from the 3%
PMB-CPTH watermelon samples reflected the results
of the 1.5% samples; therefore only the 1.5% PMB-
CPTH results are depicted (Figure 6). The amount of
CPTH remained relatively constant, while the formation
of Glc-CPT was maintained at a lower level. The initial
loss of CPTH noted in the PMB-treated samples may
be accounted for in the formation of Glc-CPT, as well
as in nonextractable residues, while the decomposition
of Glc-CPT could account for the gradual decrease in
Glc-CPT and the increase in CPTH during the 4-24 h
range. The amount of recoverable CPTH in the PMB-
treated extracts was greater than that observed in the
CPTH samples without PMB (Figure 7).
Marco, G. J .; Novak, R. A. Natural product interactions during
aniline metabolism including their incorporation in biopoly-
mers. J . Agric. Food Chem. 1991, 39, 2101-2111.
McWeeney, D. J .; Biltcliffe, D. O.; Powell, R. C. T.; Spark, A.
A. The Maillard reaction and its inhibition by sulfite. J . Food
Sci. 1969, 34, 641-643.
Mills, F. D.; Weisleder, D.; Hodge, J . E. 2,3-Dihydro-3,5-
dihydroxy-6-methyl-4H-pyran-4-one, a novel nonenzymatic
browning product. Tetrahedron Lett. 1970, 15, 1243-1246.
Although light exposures differ from laboratory to
field conditions, these data suggest that degradation of
CPTH watermelon bait occurs primarily through non-
enzymic browning reactions. Further field studies are
necessary to assess bait acceptance and efficacy, as well
as the effectiveness of PMB as a preservative. Other
sulfhydryl compounds, such as cysteine, N-acetylcys-
teine, and reduced glutathione, also prevent nonenzymic

3988 J. Agric. Food Chem., Vol. 44, No. 12, 1996
Tawara et al.
Pigman, W. Nitrogenous derivatives. In The Carbohydates,
Chemistry, Biochemistry, Physiology; Pigman, W., Ed.;
Academic Press: New York, 1957.
Primus, T.; Tawara, J .; J ohnston, J .; Cummings, J .; Volz, S.;
Goodall, M.; Hurlbut, D.; Griffin, D.; Turnipseed, S. Deg-
radation of the avicide 3-chloro-p-toluidine hydrochloride
(CPTH) in Louisiana rice fields. Environ. Sci. Technol. 1996,
in press.
Richardson, T. M. Enzymes In Principles of Food Science, Part
I, Food Chemistry; Fennema, O. R., Ed.; Dekker: New York,
1976.
Schafer, E. W. Proc. Vertebr. Pest Conf. 1984, 11, 217-222.
Schmidt, B.; Rivero, C.; Thiede, B. 3,4-Dichloroaniline N-
glucosyl- and N-malonyltransferase activities in cell cultures
and plants of soybean and wheat. Phytochemistry 1995, 39,
81-84.
Watt, B. K.; Merrill, A. L. Composition of Foods, Agriculture
Handbook 8; USDA: Washington, DC, 1963.
Winkler, R.; Sandermann, H., J r. N-Glucosyl conjugates of
chlorinated anilines: spontaneous formation and cleavage.
J . Agric. Food Chem. 1992, 40, 2008-2012.
Yao, D.; Mill, T. Aquatic photolysis study of 3-chloro-p-
toluidine hydrochloride. Unpublished Report-QA-268. SRI,
Menlo Park, CA/Denver Wildlife Research Center, Denver,
CO, 1993.
Yih, R. Y.; McRae, D. H.; Wilson, H. F. Metabolism of 3′,4′-
dichloropropionanilide: 3,4-dichloroaniline-lignin complex
in rice plants. Science 1968, 161, 376-377.
Received for review J anuary 23, 1996. Revised manuscript
received September 16, 1996. Accepted September 23, 1996.X
Still, G. G. Metabolism of 3,4-dichloropropionanilide in
plants: the metabolic fate of the 3,4-dichloroaniline moiety.
Science 1968, 159, 992-993.
J F960044K
Tweedy, B. G.; Loeppky, C.; Ross, J . A. Metobromuron:
acetylation of the aniline moiety as a detoxification mech-
anism. Science 1970, 168, 482-483.
X Abstract published in Advance ACS Abstracts, No-
vember 1, 1996.