Aerobic soil metabolism of metsulfuron-methyl

Article abstract of DOI:10.1002/(SICI)1096-9063(199904)55:4<434::AID-PS909>3.0.CO;2-Z

A laboratory study was conducted to determine the degradation rates and identify major metabolites of the herbicide metsulfuron-methyl in sterile and non-sterile aerobic soils in the dark at 20°C. Both [phenyl-U-¹⁴C]- and [triazine-2-¹⁴C]metsulfuron-methyl were used. The soil was treated with [¹⁴C]metsulfuron-methyl (0.1 mg kg^-1) and incubated in flow-through systems for one year. The degradation rate constants, DT₅₀, and DT₉₀ were obtained based on the first-order and biphasic models. The DT₅₀ (time required for 50% of applied chemical to degrade) for met sulfuron-methyl, estimated using a biphasic model, was approximately 10 days (9-11 days, 95% confidence limits) in the non-sterile soil and 20 days (12-32 days, 95% confidence limits) in the sterile soil. One-year cumulative carbon dioxide accounted for approximately 48% and 23% of the applied radioactivity in the [phenyl-U-¹⁴C] and [triazine-2-¹⁴C]metsulfuron-methyl systems, respectively. Seven metabolites were identified by HPLC or LC/MS with synthetic standards. The degradation pathways included O-demethylation, cleavage of the sulfonylurea bridge, and triazine ring opening. The triazine ring-opened products were methyl 2- [[[[[[[(acetylamino)carbonyl]amino]carbonyl]amino] carbonyl]- amino]sulfonyl]benzoate in the sterile soil and methyl 2- 1[[[[[amino](aminocarbonyl)imino]methyl] amino]carbonyl]amino]sulfonyl]benzoate in the non-sterile soil, indicating that different pathways were operable.

Full text of DOI:10.1002/(SICI)1096-9063(199904)55:4<434::AID-PS909>3.0.CO;2-Z

Pesticide Science
Pestic Sci 55:434–445 (1999)
Aerobic soil metabolism of metsulfuron-methyl
Yutai Li,1* W T Zimmerman,1 M K Gorman,2 R W Reis er,1 A J Fogiel1 and
P E Haney2
1 DuPont Agricultural Products , Experimental Station, Wilmington, DE 19880-0402, USA
2 ABC Laboratories , Inc 7200 E ABC Lane, Columbia, MO 65202, USA
Abstract: A laboratory study was conducted to determine the degradation rates and identify major
metabolites of the herbicide metsulfuron-methyl in sterile and non-sterile aerobic soils in the dark at
20ÄC. Both [phenyl-U-14C]- and [triazine-2-14C]metsulfuron-methyl were used. The soil was treated
with [14C]metsulfuron-methyl (0.1 mg kg—1) and incubated in ýow-through systems for one year. The
degradation rate constants, DT , and DT were obtained based on the ürst-order and biphasic
50
90
models. The DT (time required for 50% of applied chemical to degrade) for metsulfuron-methyl,
estimated using a biphasic model, was approximately 10 days (9–11 days, 95% conüdence limits) in the
50
non-sterile soil and 20 days (12–32 days, 95% conüdence limits) in the sterile soil. One-year cumulative
carbon dioxide accounted for approximately 48% and 23% of the applied radioactivity in the [phenyl-
U-14C] and [triazine-2-14C]metsulfuron-methyl systems, respectively. Seven metabolites were
identiüed by HPLC or LC/MS with synthetic standards. The degradation pathways
included O-demethylation, cleavage of the sulfonylurea bridge, and triazine ring opening. The triazine
ring-opened products were methyl
2-[ [ [ [ [ [ [(acetylamino)carbohyl]amino]carbonyl]amino]
carbonyl]-amino]sulfonyl]benzoate in the sterile soil and methyl 2-[ [ [ [ [amino[(amino-
carbonyl)imino]methyl] amino]carbonyl]amino]sulfonyl]benzoate in the non-sterile soil, indicating
that diþ erent pathways were operable.
( 1999 Society of Chemical Industry
Keywords: metsulfuron-methyl; degradation; soil; metabolites; DT ; synthesis
50
1
INTRODUCTION
lysis half-life at 45¡C increased from 2.1 days at pH 5
to 33 days at pH 7.11 The degradation rate of
metsulfuron-methyl in soils is aþected by soil tem-
perature, moisture, pH, and microbial viability. The
half-life of metsulfuron-methyl is shorter at higher
temperatures and moisture contents and ranges from
2.5 days (soil conditions : pH 3.1, 35¡C, 80% üeld
water holding capacity (FC)) to 36 days (soil condi-
tions : pH 5.7, 10¡C, 60% FC) depending on these
factors.10,14 The degradation rate of metsulfuron-
methyl has been positively correlated with microbial
biomass.6 Although several researchers have report-
ed the eþects of environmental conditions on the
degradation rates of metsulfuron-methyl, there have
been only a few attempts to identify the degradation
products. Some aqueous hydrolysis products, plant
Metsulfuron-methyl is a low-use-rate sulfonylurea
(SU) herbicide used for broadleaf weed control in
cereals, pasture, plantation crops and non-crop situ-
ations.1h3 It is one of the SU herbicides which were
discovered in the mid-70s by Dr George Levitt at
DuPont. It inhibits the enzyme acetolactate synthase
(ALS), also known as acetohydroxy acid synthase
(AHAS), which stops plant cell division by inhibit-
ing biosynthesis of the essential amino acids valine
and isoleucine.1h3
Pesticide transport in soils has received increasing
attention during recent years because of concern over
potential eþects on surface and ground water quality.
Soil mobility and degradation are the most important
processes that determine the fate of pesticides in
soils. Extensive research has been carried out on the
mobility of SU herbicides both in laboratory and
üeld studies.4h7
metabolites,
and
degradation
products
of
metsulfuron-methyl in soil minerals and humic acids
have been identiüed.12,15h17
The purpose of this laboratory study was to deter-
mine the degradation rate of 14C-labeled
metsulfuron-methyl in sterile and non-sterile soils, to
clarify the role of microbial metabolism in degrada-
tion, to identify major metabolites, and to propose
metabolic pathways of metsulfuron-methyl in an
aerobic soil.
Sulfonylurea herbicides degrade in soil primarily
by chemical hydrolysis and microbial metabolism
and there have been several publications which eluci-
date the signiücance of microbial degradation.3,8h11
Chemical hydrolysis of metsulfuron-methyl has been
shown to be very rapid at low pH.11h13 The hydro-
* Corres pondence to: Yutai Li, DuPont Agricultural Products ,
Experimental Station, Wilmington DE 19880-0402, USA.
E-mail: Yu-Tai.Li=us a.dupont.com
(Received 11 May 1998; revis ed vers ion received 27 Augus t 1998;
accepted 12 November 1998)
( 1999 Society of Chemical Industry. Pestic Sci 0031-613X/99/$17.50
434

Aerobic soil metabolism of metsulfuron-methyl
2
MATERIALS AND METHODS
synthesized by DuPont Agricultural Products, E I
du Pont de Nemours and Company. The detailed
syntheses of unique metabolites 4, 8 and 9 are
described below. Other reference standards were
prepared by generally known methods or were com-
mercially available.
2.1 Chemicals and reference standards
Two separate 14C-radiolabeled metsulfuron-methyls
(methyl 2-[[[[(4-methoxy-6-methyl-1,3,5-triazin-2-
yl)amino]carbonyl]amino]sulfonyl]benzoate) (1, Fig
1) were synthesized at DuPont New England
Nuclear (NEN) Research Products (Boston, MA).
One was labeled uniformly in the phenyl ring
2.1.1 Synthesis of methyl
[phenyl-U-14C] with
a
speciüc activity of
2-[[[[[amino[(aminocarbonyl)imino]methyl]amino]-
carbonyl]amino]sulfonyl]benzoate (4)
1.42 MBq mg~1 (38.28 lCi mg~1), and the other was
labeled at the 2-carbon position in the triazine ring
[triazine-2-14C] with
Free base guanylurea was freshly prepared from
commercially available guanylurea sulfate by heating
a slurry of the material in ethanol with sodium eth-
oxide (0.75 eq) for 2 h at 60¡C. After cooling, the
mixture was ültered and the ültrate evaporated.
Traces of ethanol were removed from the solid
residue by addition and evaporation of benzene
twice, then chloroform three times, to leave a solid
residue of guanylurea, mp 104–107¡C. To a suspen-
sion of guanylurea (1.0 g, 9.8 mmol) in anhydrous
acetonitrile (75ml) was added, dropwise, a solution
of 2-(methyoxycarbonyl)benzenesulfonyl isocyanate
(2.5g; 10.3mmol) in acetonitrile (15ml) over a half-
a
speciüc activity of
1.85 MBq mg~1 (49.87 lCi mg~1). Both 14C-labeled
compounds had radiochemical purities greater than
99% as determined by high performance liquid chro-
matography (HPLC). Two solutions at concentra-
tions of 10mg litre~1 (one for each 14C-label) were
prepared by dissolving the 14C-labeled metsulfuron-
methyl in water.
All organic solvents were HPLC grade. All other
chemicals were reagent grade or better. Water was
puriüed using a Milli-Q} water puriücation system.
Unlabeled reference standards (2–9, Fig 1) were
Figure 1. Propos ed metabolic pathways of mets ulfuron-methyl in s oil (* denotes s ite of 14C). The ins et s hows 9, a pos s ible alternative to 4
(s ee Section 3.4.2).
Pestic Sci 55:434–445 (1999)
435

Y Li et al
hour period at room temperature (c. 20¡C) under
nitrogen. After 24h the suspension was ültered and
the solids washed successively with acetonitrile,
ether, then dichloromethane (]3). Trituration of
these solids with warm acetonitrile (50¡C) and ül-
tration with dichloromethane rinse aþorded 1.42g
0.1 mg kg~1. Two ýasks treated with each of the two
14C-labeled metsulfuron-methyl solutions were con-
sidered as replicates. Metsulfuron-methyl and the
soil in each ýask were mixed by hand shaking. Suffi-
cient Milli-Q} water was added to each ýask to
adjust the moisture content of the soil to approx-
imately 75% of FC (FC \ 17.4%). The soil moisture
content in each ýask was maintained at 75% of FC
by addition of water at least monthly throughout the
study. Additionally, four ýasks were treated with
[14C]metsulfuron-methyl (two ýasks per 14C-label)
at a concentration of 1 mg kg~1 (high dose) in order
to generate sufficient quantity for metabolite identi-
ücation. The high-dose ýasks were treated and incu-
bated in the same manner as the ýasks containing
0.1 mg kg~1 metsulfuron-methyl. Flasks with treated
soil were sampled for soil extraction at days 0, 1, 3,
7, 14, and 21, and at months 1, 2, 3, 4, 6, 9, and 12.
(42%) of
4
as a white powder, mp [ 260¡C:
[1H]NMR [dimethylsulfoxide(DMSO)-d , 360
MHz] d 3.76 (s, 3H), 7.42 (m, 2] NH), 7.56 (m,
6
3H), 7.97 (d, 1H), 8.75 (br, NH), 10.40 (br, NH);
LC/MS (electrospray, negative mode) m/z 342 (40%,
[M [ H]~, 299 (100%), 214 (20%), 182 (50%),
(positive mode) m/z 344 (100%, MH`).
2.1.2 Synthesis of methyl 2-[[[[[[[(acetylamino)-
carbonyl] amino] carbonyl]amino]carbonyl]amino]-
sulfonyl]benzoate (8)
To a sample of acetylbiuret (0.24g; 1.65mmol)18
was added a solution of 2-(methoxycarbonyl)benze-
nesulfonyl isocyanate in xylenes (340g litre~1;
2.5 ml). The mixture was heated on a steam bath
under a stream of nitrogen to evaporate the solvent.
After 5h, the mixture was cooled and triturated with
ethyl acetate and the solids ültered. This material
was further puriüed by trituration with warm (40¡C)
acetonitrile, ültered and dried in vacuum to aþord
0.30 g (47%) of 8 as a white crystalline powder, mp
2.3.2 Sterile system
The ýasks were prepared as described above for the
non-sterile system except that each ýask contained
the sterile soil (50g dry weight equivalent) auto-
claved (at 15psi and 120¡C) for 1h on four consecu-
tive days. The moisture content of the sterile soil was
adjusted to 75% of FC with sterilized water (0.2lm
ülter). The sterile system was connected as described
in Section 2.3.4 with the addition of a 0.2lm ülter
disk (Gelman ACRO 50 PTFE) inserted before the
224–226¡C: [1H]NMR (DMSO-d , 300MHz) d
6
2.09 (s, 3H), 3.85 (s. 3H), 7.80 (m, 3H), 8.11 (d, 1H),
ýask
to
maintain
sterility.
For
each
10.26 (s br, NH), 10.97 (m br, 3 ] NH); LC/MS
(electrospray, negative mode) m/z 385 (100%,
[M [ H]~).
[14C]metsulfuron-methyl treatment, two ýasks were
sampled, one for extraction to analyze for parent and
metabolites, and the other for plate counts to deter-
mine sterility, at day 0 and at months 1, 2, 4, 5, 6, 9,
and 12.
2.1.3 Synthesis of methyl 2-[[[[[(acetylamino)-
carbonyl]amino] carbonyl]amino]sulfonyl]benzoate (9)
This substance (see Fig 1) was prepared from acety-
lurea by an analogous procedure to that described
above for 8. Data for 9: mp 184–189¡C; [1H]NMR
2.3.3 Control and biomass system
The ýasks were dosed with non-radiolabeled
metsulfuron-methyl solution at the same concentra-
tion (0.1mg kg~1) and maintained under identical
conditions to the non-sterile soil ýasks. The soils in
the ýasks were only sampled at the termination of
the study to determine the biomass and microbial
plate counts. Soil biomass and microbial plate counts
were also conducted at the initiation of the study
prior to application of metsulfuron-methyl. Biomass
was determined by Anderson-Domsch’s glucose-
substrate induced respiration method.19 Bacteria and
fungi were enumerated using plate-count agar and
Sabaroud agar (Difco laboratories), respectively.
(DMSO-d , 300MHz) d 2.08 (s, 3H), 3.85 (s, 3H),
6
7.80 (m, 3H), 8.12 (d, 1H), 10.58 (s, NH), 11.06 (s,
NH); LC/MS (electrospray, negative mode) m/z 342
(100%, [M [ H]~).
2.2 Soil
Matapeake silt loam (pH 5.2; 1.8% organic matter ;
16.8% clay, 55.6% silt, 27.6% sand; cation exchange
capacity 7.0meq 100g~1 soil) was collected from the
top 15cm of a üeld located in Middletown, Dela-
ware. The fresh soil was passed through a 2-mm
sieve and thoroughly homogenized before the appli-
cation of metsulfuron-methyl.
2.3.4 Flow-through systems
In both the non-sterile and the sterile systems
(Sections 2.3.1 and 2.3.2), each ýask was ütted with
a glass impinger with an air inlet and outlet tubing
connected to a series of six traps. The soil ýask inlet
was connected to an outlet of a water bottle to
provide humidiüed air (an inlet of the water bottle
was open to the air). The traps used in order of prox-
imity to the ýask were a blank overýow trap, two
2.3 Flow-through systems
2.3.1 Non-sterile system
The soil (equivalent to 50g, oven-dry weight) was
weighed into Pyrex} glass ýasks (250ml). The
respective 14C-labeled metsulfuron-methyl solutions
(10mg litre~1) were applied to the soil in each of the
designated ýasks to produce a soil concentration of
436
Pestic Sci 55:434–445 (1999)

Aerobic soil metabolism of metsulfuron-methyl
ethylene glycol traps (20ml), a polyurethane foam
plug trap to collect any organic volatiles, and two
potassium hydroxide traps (1N; 20ml) containing
phenolphthalein indicator (to indicate carbon dioxide
saturation of traps) to collect [14C]carbon dioxide
evolved from the [14C]metsulfuron-methyl. The
outlet of the last potassium hydroxide trap was con-
nected to a vacuum manifold which was used to
regulate the air ýow in the system.
In the control and biomass system (Section 2.3.3),
all the ýasks were connected in series and the last
ýask was connected to three traps (ethylene glycol,
foam plug, and potassium hydroxide) to permit mea-
surement of background radioactivity. The outlet of
the potassium hydroxide trap was connected to a
vacuum manifold which was used to regulate the air
ýow in the system. The water bottle was placed
before the soil ýasks to provide humidiüed air as in
the sterile and non-sterile systems.
In order to monitor the formation of [14C]carbon
dioxide and 14C-organic volatiles resulting from the
degradation of [14C]metsulfuron-methyl, each trap
(foam plug, potassium hydroxide, and ethylene
glycol traps) was sampled and replaced with fresh
solutions or new foam plug monthly and at each soil
sampling date. Although no soil sample was taken
from the control and biomass system, the trapping
solutions were collected in the same manner as for
the non-sterile system for background radioactivity.
The connected ýow-through systems were main-
tained in a temperature-controlled chamber at 20¡C
(^2¡C) in the dark under aerobic conditions for one
year.
extracts were pooled. Triplicate aliquots (3] 1ml)
were analyzed by LSC. If the estimated non-
extractable residues (bound residues) were greater
than 10% of the applied radioactivity (AR) (bound
residues %AR \ total applied (100% AR)-%AR in
extracts-%AR in traps), extraction step 2 was con-
ducted.
Step 2: After step 1 extraction, the soil sample was
combined with methylene chloride ] methanol ] 2M
ammonium carbonate (100ml, 3] 4] 1 by volume)
and shaken for 1 h at room temperature. The extrac-
tion was conducted three times and the extracts were
pooled. Triplicate aliquots (3] 1ml) were analyzed
by LSC. Step 1 and step 2 extracts were pooled and
concentrated by vacuum using rotary evaporation.
The residues were dissolved in water. Triplicate ali-
quots were analyzed by LSC to determine the total
radioactivity in the extract. The extract was also
analyzed by HPLC to determine concentrations of
metsulfuron-methyl and its metabolites.
2.4.3 Analysis of soil extracts
Soil extracts were analyzed using a HPLC equipped
with both a UV-Vis detector and an on-line radio-
chemical detector (Ramona, Raytest Inc). HPLC
method
1
used
a
Zorbax Rx-C8 column
(250 ] 4.6mm, 5lm) and a gradient with mobile
phase of A, water (pH \ 2.3 with 1g litre~1 phos-
phoric acid), and B, acetonitrile (from 0 to 5min, 0%
B; at 15 min, 15% B; at 25 min, 40% B; at 30min,
100% B; from 30 to 35 min, 100% B, at 38min, 0%
B). HPLC method
2 used a PRP-1 column
(305 ] 7.0 mm, 10 lm) with the same mobile phase
as HPLC method 1 but a diþerent gradient (from 0
to 3min, 10% B; at 10 min, 20% B; at 20 min, 40%
B; at 30 min, 90% B; at 35 min, 10% B). The
mobile phase ýow rate was 1.5ml min~1 in both
methods. HPLC method 1 was used for all the
sample analyses and HPLC method 2 was used
for conürmatory analyses. Selected soil extracts were
spiked with unlabeled reference standards of
2.4 Analytical methods
2.4.1 Analysis of radiolabeled volatiles
Triplicate aliquots (3] 1 ml) of the potassium
hydroxide and ethylene glycol trap solutions used to
trap [14C]carbon dioxide and organic volatiles were
combined with scintillation ýuid (5ml) and analyzed
for total radioactivity by LSC (liquid scintillation
counting, Beckman Instrument Inc). The poly-
urethane foam plugs were extracted with
acetonitrile ] 2M ammonium carbonate (3] 10 ml,
90 ] 10 by volume). Triplicate aliquots (3] 1ml) of
the extract were analyzed for total radioactivity by
LSC.
2
(2-[[[[(4-methoxy-6-methyl-1,3,5-triazin-2-yl)-
amino]carbonyl]amino]sulfonyl]benzoic acid, com-
mon name: metsulfuron), (methyl 2-[[[[(4-
3
hydroxy - 6 - methyl - 1,3,5 - triazine - 2 - yl)amino] -
carbonyl]amino]sulfonyl]benzoate, trivial name: O-
desmethyl metsulfuron-methyl), and 7 (methyl 2-
(aminosulfonyl)benzoate, trivial name: benzene
sulfonamide) and analyzed using both HPLC
methods described above to conürm the identities of
metabolites. Metabolite 6 (4-methoxy-6-methyl-1,3,
5-triazine-2-amine, trivial name: triazine amine) was
isolated from the two-month samples (high dose) and
analyzed using both HPLC methods to conürm its
presence by retention time matches between the
sample and 14C-labeled reference standard. Metabo-
lites 4,5, and 8 were analyzed by LC/MS for identiü-
cation under the conditions described below.
2.4.2 Soil extraction
At each soil sampling time as described above, two
ýasks (one from each of the two [14C]metsulfuron-
methyl treatments) were taken. A two-step extraction
method was used.
Step 1: The soil in each ýask was combined with
acetonitrile ] 2M ammonium carbonate (100ml,
9] 1 by volume) and shaken for 1 h on a platform
shaker at room temperature (c. 20¡C). The solution
was centrifuged (approximately 2500 rev min~1) for
15min and the supernatant was decanted. This
extraction was conducted three times and the
An LC (equipped with Micro-Tech Scientiüc
ultra-plus gradient pump, a Zorbax SB-C18 column
Pestic Sci 55:434–445 (1999)
437

Y Li et al
(150 ] 1mm, 3.5lm), acetonitrile with 0.5ml litre~1
formic acid ] water with 0.5ml litre~1 formic acid as
mobile phase (3] 97 by volume as initial condition))
was connected to both a Ramona radiochemical
detector and a mass spectrometer (MAT 900,
Finnigan) interfaced with a MAT 900 series atmo-
spheric pressure ionization (API) system, electro-
spray mode (Finnigan). The analyses were made in
the negative ion and/or positive ion modes. The MS
parameters were as follows : scan range, 45–1000 amu
(atomic mass units); ion source temperature, 40¡C;
electron multiplier voltage, 1.80kV positive ion,
1.90 kV negative ion, 2.0 kV for accurate mass
experiment. CID (collision induced dissociation)
capillary and tube lens both oþset at 30 volts.
indicated that the soil micro-organisms remained
viable throughout the study in non-sterile ýasks
(Table 1). The dynamics of microbial biomass and
population after application of metsulfuron-methyl
were not monitored in this study. An earlier study21
demonstrated that application of metsulfuron-methyl
would not adversely aþect soil microbial biomass and
population once soil microbes adapted to the pres-
ence of metsulfuron-methyl.
3.1.2 Soil sterility in the sterile system
Results from plate counts indicated that soil
remained sterile for six months. After six months,
the system became contaminated, since microbial
colonies were seen in the bacteria and fungi plate
media for both 14C-labeled metsulfuron-methyl
treatments (data not shown). Therefore, only data up
to six months were used to interpret results in the
sterile system.
2.4.4 Determination of soil bound residues
All the post-extracted soil samples were air-dried in
a laboratory hood. When dry, the samples were
homogenized and weighed. Triplicate aliquots were
combusted using
a Harvey biological oxidizer
(Harvey Instrument Inc, model OX 500) and the
[14C]carbon dioxide released from the combustion
process was trapped in 14C-cocktail scintillation ýuid
(15 ml) and radioactivity was measured by LSC.
3.2 Recovery and distribution of 14C-radioactivity
Data on the recovery and distribution of radioac-
tivity from the non-sterile soil treated with the [14C-
phenyl]- or [14C-triazine]-labeled metsulfuron-
methyl are presented in Fig 2. The total 14C
recovery was 96.7% AR (^4.4) for the [14C-phenyl]
label and 98.2% AR (^2.9) for the [14C-triazine]
label. In the sterile soil, the 14C recovery ranged
from 91% AR to 110% AR indicating that mass
balance was maintained.
2.4.5 Fractionation of soil bound residues
Soil bound residues were fractionated with strong
base and acid into three fractions: humin, humic acid
(HA), and fulvic acid (FA).20 Radioactivity in the
three fractions was determined by LSC. This frac-
tionation procedure was conducted only on four-
month samples. The post-extracted soil was mixed
with aqueous sodium hydroxide (1M; 100 ml) and
shaken overnight on a platform shaker at room tem-
perature. The solid phase was separated from the
aqueous phase by centrifugation, and washed with
aqueous sodium hydroxide (1 M; 2] 20 ml). The
combined aqueous sodium hydroxide extract was
adjusted to 150ml and three aliquots (3] 2ml) were
analyzed by LSC. The solid (humin fraction) was
combusted in a Harvey biological oxidizer and the
[14C]carbon dioxide released from the combustion
process was trapped in 14C-cocktail scintillation ýuid
(15 ml) and radioactivity in the humin fraction was
measured by LSC. The sodium hydroxide extract
was acidiüed to pH 2 with concentrated hydrochloric
acid to precipitate the humic acid fraction (HA). The
radioactivity of the remaining supernatant (FA) was
determined by LSC. The HA precipitate was
redissolved in aqueous sodium hydroxide (1M;
25 ml) and the solution radioactivity (HA fraction)
was determined by LSC.
3.2.1 Mineralization
In the non-sterile system, the rate of [14C]carbon
dioxide evolution was faster in the soil treated with
[phenyl-U-14C]metsulfuron-methyl (48.2% AR as
carbon dioxide at 12 months) than the soil treated
with [triazine-2-14C]metsulfuron-methyl (22.8% AR
as carbon dioxide at 12 months) (Fig 2). The diþer-
ent rates of mineralization indicate that the phenyl
moiety is more susceptible to microbial attack than
the triazine moiety of the molecule. In the sterile
system, there was no detectable carbon dioxide
(detection limit of 0.3% AR) evolved for six months
(data not shown).
Table 1. Soil viability tes t
Parameters
Microbes (cfu gÉ1) ]105
Study
Study
termination
(1 yr)
initiation
3
RESULTS AND DISCUSSION
3.1 Soil viability and sterility
Bacteria
8.19 (^6.21)
0.25 (^0.06)
0.104
172 (^23)
6.83 (^0.85)
0.116
3.1.1 Soil viability in the control and biomass system
Measurements of biomass and microbial plate counts
conducted at initiation and termination of the study
Fungi
Biomas s (mg C gÉ1)
438
Pestic Sci 55:434–445 (1999)

Aerobic soil metabolism of metsulfuron-methyl
Figure 2. Mas s balance of
(a) [phenyl-U-14C] and
(b) [triazine-2-14C] labeled
mets ulfuron-methyl in the
non-s terile s oil.
3.2.2 Soil bound residues
estimation of DT and DT values. DT was esti-
50
90
50
In this study, we found approximately two-fold more
14C-soil bound residues in the non-sterile than in the
sterile soils. Therefore, we believe that microbial
activity contributed to soil bound residues. Of the
radioactivity remaining in the soil (four-month
samples), 14.3%, 0.92%, and 9.14% of AR (the
phenyl label) were found in humic acid, fulvic acid,
and humin fractions, respectively. For the triazine
label, 12.5%, 0.43%, and 6.81% of AR were found in
humic acid, fulvic acid, and humin fractions, respec-
tively. The results indicated that bound radioactivity
was associated mainly with humin and humic acid
fractions.
mated by using the equation: DT \ ln(2)/k. The
50
degradation rate constant (k) was determined using
linear regression of ln(c/c ) (c: concentration) over
0
time and was the slope of the linear regression line.
Based on the ürst-order model, the estimated values
of DT
(the time required for 50% of applied
50
chemical to degrade) were 30 days (26–35 days, 95%
conüdence limits) and 43 days (36–51 days, 95%
conüdence limits) in the non-sterile and sterile soils,
respectively (Table 2). These results are similar to
those of James et al14 who found half-lives of eight
to 36 days (based on the ürst-order model) in an
acidic soil (pH 5.7 and 7.3% OC) at various soil
moistures and temperatures.
The
result
of
the
interaction
eþect
(treatment ] time) using the JMP} Program (JMP},
V3.2.2, SAS Institute Inc, Cary, NC) indicates that
the rate constants (k, linear regression slope) were
signiücantly diþerent between the two treatments,
3.3 Rate of degradation of metsulfuron methyl
in aerobic soil
Since the sterile soil was contaminated after six
months, only data up to six months were used in the
sterile and non-sterile soils. Therefore, the DT
values (\ln 2/k) of 30 and 43 days were signiücantly
50
calculation of degradation rate constant (k), DT
,
50
and DT of metsulfuron-methyl for both the sterile
diþerent between the non-sterile and sterile soils
based on the ürst-order model.
90
and non-sterile soils in order to make a direct com-
parison between the two.
The lack-of-üt test (using JMP} program) indi-
cates that the ürst-order model is not a good üt.
Both the ürst-order and biphasic models were
used to üt the data (Table 2 and Fig 3) even though
the ürst-order model has been widely used in the
Based on the biphasic model, the estimated DT
50
was more close to the measured data (Fig 3). The
Pestic Sci 55:434–445 (1999)
439

Y Li et al
Figure 3. Dis s ipation of
0.1 mg kgÉ1 of [14C]mets ulfuron-
methyl over time in s oil (pH 5.2),
1.8% OM, 75% FC, and 20¡C in
the dark, (a) fitted lines bas ed
on the firs t-order model, (b)
fitted line bas ed on the biphas ic
model, (>) non-s terile s oil (=)
s terile s oil.
DT values of metsulfuron-methyl were 10 days
(9–11 days, 95% conüdence limits) and 20 days
and dissipation of metabolites in the non-sterile soil
are presented in Fig 6.
50
(12–32 days, 95% conüdence limits) in the non-
sterile and sterile soils, respectively. The 95% con-
3.4.1 Identiücation of 2, 3, 6, and 7 by
co-chromatography
üdence limits of DT did not overlap between the
50
Metabolites 2, 3, and 7 were conürmed by compari-
son of the retention times of unlabeled reference
standards with the corresponding radiochemical
peaks. Metabolite 6 was conürmed using the radiola-
beled standard. Metabolites 3, 6, and 7 were found in
both sterile and non-sterile soils, while metabolite 2
was found only in the non-sterile soil.
sterile and non-sterile soils, which suggests that bio-
logical degradation (by soil micro-organisms) signiü-
cantly contributes to the dissipation of this
compound. The dissipation of metsulfuron-methyl
in soil involves both chemical and microbial pro-
cesses, and it has been found in other SUs as
well.11,22,23 Degradation of metsulfuron-methyl in
an acidic soil environment is relatively rapid because
of SU bridge hydrolysis which diminishes in alkaline
soils.
3.4.2 Identiücation of metabolite 4 by LC/MS
Carbamoyl guanidine (4) was found only in the non-
sterile soil treated with either [phenyl-U-14C]- or
[triazine-2-14C]metsulfuron-methyl. It was isolated
and identiüed using HPLC method 1 (C8 column)
and conürmed by LC/MS.
3.4 Soil metabolites of metsulfuron-methyl
Six major metabolites (2–7) were detected by HPLC
in the non-sterile soil (Fig 4) and üve major metabo-
lites (3, 5–8) in the sterile soil (Fig 5). The formation
LC/MS–ESI negative ion analysis shows a spec-
trum with a [M [ H]~ m/z 342 corresponding to the
440
Pestic Sci 55:434–445 (1999)

Aerobic soil metabolism of metsulfuron-methyl
Table 2. Degradation model parameters for mets ulfuron-methyl in s oil
Sys tem
Model parameters
DT
DT
50
(days )
90
(days )
Firs t-order parameters a
Intercept
4.057
k(s lope)
r2
0.90
0.95
Non-s terile s ys tem
95% confidence limits b
Sterile s ys tem
0.0233
0.0163
30
99
86–116
141
26–35
43
4.333
95% confidence limits
36–51
121–170
Biphas ic model parameters
DT
DT
50
(days )
90
(days )
a
13.2
44.6
k1
k2
r2
Non-s terile s ys tem
95% confidence limits c
Sterile s ys tem
0.0118
0.0126
0.0837
0.0622
0.990
0.977
10
9–11
20
44
40–50
119
95% confidence limits
12–32
99–134
a Calculated us ing data through s ix months for the non-s terile and s terile s oils . DT \ ln2/k, DT \ ln(10)/k where k was firs t-order rate
50 90
cons tant and was the s lope determined bas ed on linear regres s ion of ln(c/c ) vers us time. The firs t-order model: C \C · eÉkÃt where,
0
0
C
: initial concentration (mg kgÉ1); C: concentration at time t (mg kgÉ1); k: degradation rate cons tant.
0
b 95% confidence levels bas ed on linear regres s ion us ing ln(c/c ) vers us time.
c Biphas ic model:
0
C/C \a · eÉk1Ãt ] (1 [a) · eÉk2Ãt
Where, C : initial concentration (mg kgÉ1); C: concentration at time t (mg kgÉ1); k and k : s low and fas t degradation rate cons tants . a:
0
0
1
2
cons tant. Cons tants and confidence limits were determined us ing JMP} program (SAS Ins titute, Inc. Cary, NC).
Note: r2 was reported bas ed on ln s cale fitting of both models .
Figure 5. 14C-radiochromatogram of one-month s amples in the
Figure 4. 14C-radiochromatogram of one-month s amples in the
non-s terile s oil ((a) [phenyl-U-14C] and (b) [triazine-2-14C]
labeled mets ulfuron-methyl, HPLC conditions : Rx-C8 column,
250 ] 4.6 mm, flow rate of 1.5 ml minÉ1, mobile phas e:
s terile s oil ((a) [phenyl-U-14C] and (b) [triazine-2-14C] labeled
mets ulfuron-methyl, HPLC conditions : Rx-C8 column,
250 ] 4.6 mm, flow rate of 1.5 ml minÉ1, mobile phas e:
acetonitrile and acidified water (0.1% H PO ), s ee Section 2.4.3
acetonitrile and acidified water (0.1% H PO ), s ee Section 2.4.3
3
4
3
4
for the mobile phas e gradient).
for the mobile phas e gradient).
Pestic Sci 55:434–445 (1999)
441

Y Li et al
[M [ H]~ m/z 182 and no fragmentation by CID.
Therefore, this spectrum is consistent with the struc-
ture for 5 and conürmed as saccharin.
3.4.4 Identiücation of metabolite 8 by LC/MS and
HPLC
Acetyl triuret (8) was found only in the sterile soil
treated with either [phenyl-U-14C]- or [triazine-2-
14C]metsulfuron-methyl. It reached approximately
18% AR in the six-month samples (average of the
two labels). The radiochemical peak was isolated
using HPLC method 1 by fractionation of the
column effluent. The retention time of the reference
standard matched with the isolated metabolite using
the HPLC method 1. The isolated metabolite was
further conürmed by LC/MS.
LC/MS–ESI negative ion analysis shows a spec-
trum with a [M [ H]~ m/z 385 with a signiücant
m/z 387 ion in the triazine label corresponding to the
on-line 14C peak. This spectrum is consistent with
the structure for 8 and the CID spectrum shows a
match for the CID mass spectrum of the reference
standard (Fig 8) except for the lack of 14C ion (m/z
387) in the reference standard. An aliquot of the iso-
lated sample spiked with the reference standard has
coelution of the [M [ H]~ m/z 385 mass peak (data
not shown).
Figure 6. Metabolite profiles of (a) [phenyl-14C] and (b) [triazine-
2-14C] labeled mets ulfuron-methyl in the non-s terile s oil (pH 5.2,
20¡C in the dark; for s tructure of metabolites , s ee Fig 1).
on-line 14C peak. The CID mass spectrum (Fig 7)
shows fragmentation consistent with the proposed
structure (4, Fig 1). The 14C ion appears in the spec-
trum of triazine-labeled metabolite 4 (Fig 7), since
this is labeled at one speciüc carbon atom. The ratio
is close to that calculated from speciüc activity plus
small contributions from 34S and 18O.
LC/MS–ESI positive ion analysis shows a spec-
trum with a MH` m/z 344 (100%) and fragmenta-
tion m/z 129 (10%) consistent with the proposed
structure. LC/MS–ESI(]) high resolution/accurate
mass/elemental composition gives an exact mass of
3.5 Metabolic pathway of radiolabeled
metsulfuron-methyl in the soil
Degradation pathways of metsulfuron-methyl in soil
under aerobic conditions are proposed in Fig 1. The
main metabolites that contained both the triazine and
phenyl moieties were 2, 3, and 4 in the non-sterile
soil, 3 and 8 in the sterile soil. Metabolite 6 con-
tained only the triazine moiety, and metabolites 5
and 7 contained only the phenyl moiety (Fig 1). All
the metabolites increased early in the study then dis-
sipated to less than 10% AR by the end of the study
(Fig 6), except metabolite 6, which did not decline
rapidly. Both the [phenyl-U-14C]- and [triazine-2-
14C]-metsulfuron-methyl were mineralized to
[14C]carbon dioxide extensively in the non-sterile
soil.
Cleavage of the sulfonylurea linkage, which has
been commonly found in other SU herbi-
cides,2,8,24,25 results in the formation of metabolites
6 and 7. Urea bridge cleavage occurs rapidly in acidic
aqueous solution in the absence of micro-organisms,
but is slow at neutral pH. In the non-sterile soil, this
cleavage proceeds through both abiotic bridge
hydrolysis as well as microbially mediated hydro-
lysis.
344.0650
with
a
molecular
formula
of
C H N O S. Mass spectral features of the syn-
11 14 5 6
thetic reference standard of 4 are in good agreement
with the samples which contained metabolite 4. An
alternative structure (9, see Section 2.1.3) with the
same mass but a diþerent elemental composition was
synthesized initially but did not match metabolite 4
by CID mass spectrum and accurate mass/elemental
composition.
3.4.3 Identiücation of metabolite 5 by LC/MS
Metabolite
5
(1,2-benzisothiazol-3(2H)-one 1,1-
dioxide, common name: saccharin) was isolated from
soil extracts based on HPLC method 1 and con-
ürmed by LC/MS. LC/MS–ESI negative ion
analysis shows a spectrum with [M [ H]~ m/z 182
corresponding to the on-line 14C peak. A CID spec-
trum produced no fragmentation of the compound.
Analysis of the reference standard of 5 also shows a
O-Demethylation of the triazine moiety of
metsulfuron-methyl may be a chemical or microbial
process (or both) since metabolite 3 was found in
both sterile and non-sterile soils. O-Demethylation
of the ester on the phenyl moiety of metsulfuron-
methyl is probably a microbial degradation process
442
Pestic Sci 55:434–445 (1999)

Aerobic soil metabolism of metsulfuron-methyl
Figure 7. Mas s s pectrum of
metabolite 4 and the reference
s tandard (* denotes s ite of 14C).
(a) Negative ion CID ([30 volts )
mas s s pectrum of metabolite 4
(phenyl label). (b) Negative ion
CID ([30 volts ) mas s s pectrum
of metabolite 4 (triazine label).
(c) Negative ion CID ([30 volts )
mas s s pectrum of the reference
s tandard.
since metabolite 2 was found only in the non-sterile
soil.
with an elemental composition of C H N O S
13 14 4 8
(that is equivalent to metabolite 8 in our study), and
Metabolite 8 was unique to the sterile soil, sug-
gesting that it is an abiotic hydrolysis product. Tri-
azine ring-opened degradation products analogous to
8 have been reported for other SU herbicides
two structures were proposed.28 Neither of these
proposed structures has proven to be correct in light
of our unequivocal synthesis of the reference stan-
dard 8 and its comparison with the observed metabo-
lite in the sterile soil by LC/MS and HPLC.
(prosulfuron,24
thifensulfuron-methyl26
and
chlorsulfuron27) in hydrolysis studies. A triazine
ring-opened degradation product of metsulfuron-
methyl was reported in an aqueous hydrolysis study
at pH 5 and in the sterile soil in a soil degradation
study.9,12 Both hydrolysis and soil studies reported
the same triazine ring-opened degradation product
Metabolite 8 was not observed in the non-sterile
soil. There are two possible explanations. First,
metabolite 3 may be further degraded preferentially
into metabolite 4 by microbial transformation instead
of metabolite 8 by chemical hydrolysis (Fig 1). The
microbial enzymatic reaction from metabolite 3 to 4
Pestic Sci 55:434–445 (1999)
443

Y Li et al
Figure 8. Mas s s pectrum of metabolite 8 and the reference s tandard (* denotes s ite of 14C). (a) Negative ion CID ([30 volts ) mas s
s pectrum of metabolite 8 (triazine label). (b) Negative ion CID ([30 volts ) mas s s pectrum of the reference s tandard.
may be faster than the hydrolysis reaction to 8.
Second, metabolite 8 may have formed but degraded
microbially to undetectable levels by the time the
samples were taken.
hydrolytic bond cleavage of two of the triazine ring
C–N bonds is necessary to arrive at the carbamoyl
guanidine structure of 4.
Metabolite 4 appears to be a biotransformation
product since it was not detected in the sterile soil.
We believe that this is the ürst report of such a
metabolite for metsulfuron-methyl that has been
identiüed and conürmed by comparison with a syn-
thesized authentic reference standard. An analogous
metabolite has been observed in a chlorsulfuron soil
dissipation study.29 A mechanism for the formation
of 4 can be proposed as follows: The ürst step
appears to be O-demethylation of the triazine ring to
form metabolite 3, since this metabolite was
observed initially, then declined while the presence
of 4 gradually increased (Fig 6). Enzyme-mediated
4
CONCLUSIONS
Metsulfuron-methyl degraded rapidly in an acidic
soil in the dark at 20¡C (^2)¡C under aerobic condi-
tions. The estimated DT
and DT
values of
50
90
metsulfuron-methyl in the non-sterile soil using a
biphasic model were approximately 10 days (9–11
days, 95% conüdence limits) and 44 days (40–50
days, 95% conüdence limits), respectively. This
result is consistent with those previous reported.14,30
The principal degradation product after 12 months
is carbon dioxide, which accounted for approxim-
ately 48% AR and 23% AR in the [phenyl-U-14C]-
444
Pestic Sci 55:434–445 (1999)

Aerobic soil metabolism of metsulfuron-methyl
metsulfuron-methyle. (Chemical or microbial degradation of
sulfonylureas in the soil. II. Metsulfuron-methyl. Weed Res
31:149–155 (1992).
and
[triazine-2-14C]-metsulfuron-methyl
treat-
ments, respectively. The major routes of degradation
are O-demethylation, sulfonylurea bridge cleavage
and triazine ring opening. Two triazine ring-opened
products, metabolite 8 in the sterile soil, and metab-
olite 4 in the non-sterile soil, were found in this
study. The acetyl triuret 8 and analogous compounds
have been observed in the hydrolysis of metsulfuron-
methyl and related SU herbicides, as discussed pre-
viously, while the carbamoyl guanidine 4 has not
been previously reported for any metsulfuron-methyl
study. Microbial metabolism best explains the for-
mation of 4 while chemical hydrolysis leads to the
formation of 8. The potential for soil persistence of
sulfonylurea herbicides containing the triazine
moiety breakdown products has been previously the
subject of speculation;31 however, using sensitive
LC/MS technology, the structures of major triazine
ring breakdown products have been deünitively elu-
cidated. The results indicate that triazine ring
opening did occur and that hydrolytic and microbial
mechanisms were operable.
10 Ismail BS and Lee HJ, Persistence of metsulfuron-methyl in
two soils. J Environ Sci Health B30:485–497 (1995).
11 Brown HM and Kearney PC, ] Plant biochemistry, environ-
mental properties, and global impact of the sulfonylurea her-
bicides, in Synthesis and Chemistry of Agrochemicals II, ed
by Baker DR, Fenyes JG and Moberg WK. ACS Sympo-
sium Series 443. Ch 3, pp 32–49 (1991).
12 Sabadie J, Chemical acidic hydrolysis of metsulfuron-methyl.
Weed Res 30:413–419 (1990).
13 Hemmamda S, Calmon M and Calmon JP, Kinetics and
hydrolysis mechanism of chlorsulfuron and metsulfuron-
methyl. Pestic Sci 40:71–76 (1994).
14 James TK, Klaþenbach P, Holldand PT and Rahman A, Deg-
radation of primisulfuron-methyl and metsulfuron-methyl in
soil. Weed Res 35:113–120 (1995).
15 Sabadie J and Bastide J, Degradation of metsulfuron-methyl
on various minerals. Weed Res 30:1–8 (1990).
16 Sabadie J, Degradation of chlorsulfuron and metsulfuron-
methyl in the presence of humic acids. Weed Res 33:397–407
(1993).
17 Anderson JJ, Priester TM and Shalaby LM, Metabolism of
metsulfuron-methyl in wheat and barley. J Agric Food Chem
37:1429–1434 (1989).
18 Bestian H and Grimm D, Procedure for preparation of urea
derivatives, French Patent 1 577 123 (1969); Chem Abstr
72:89756 (1970).
19 Anderson JPE and Domsch KH, A physiological method for
the quantitative measurement of microbial biomass in soils.
Soil Biol Biochem 10:215–221 (1978).
20 Schnitzer M and Kahn SU, Humic substances in the environ-
ment. Dekker, New York. pp 17 (1972).
21 Ismail BS, Goh KM and Kader J, Eþects of metsulfuron-
methyl on microbial biomass and populations in soils. J
Environ Sci Health B31:987–999 (1996).
22 Beyer EM, Brown HM and Duþy MJ, Sulfonylurea herbicide
soil relations. Prog Brighton Crop Prot Conf – Weeds. pp
531–540 (1987).
23 Taylor JA, Skjemstad JO and Ladd JN, Factors inýuencing
the breakdown of sulfonylurea herbicides in solution and in
soil. CSIRO Australia, Division of Soil, Divisional Report
No 127, ISSN 0725-8526 (1995).
ACKNOWLEDGEMENTS
The authors gratefully acknowledge the assistance of
RF Dietrich for helpful discussions on possible tri-
azine ring-opened metabolites; the assistance of DR
Tabibian for the synthesis of metabolites; HJ Strek,
AC Barefoot, CA Bellin, HM Brown, LD Butler
from DuPont and Dr R Velagaleti from ABC labor-
atories, Inc for helpful discussions and review of this
manuscript; and D Berengut, consultant (DuPont
Engineering) for running JMP} program and inter-
preting the statistical results.
REFERENCES
24 Bray LD, Heard NE, Overman MC, Vargo JD, King DL,
Lawrence LJ and Phelps AW, Hydrolysis of prosulfuron at
pH 5: Evidence for a resonance-stabilized triazine cleavage
product. Pestic Sci 51:56–64 (1997).
25 Sabadie J, Chemical acidic hydrolysis of chlorimuron-ethyl.
Weed Res 35:33–40 (1995).
26 Cambon J and Bastide J, Hydrolysis kinetics of thifensulfuron-
methyl in aqueous buþer solutions. J Agric Food Chem
44:333–337 (1996).
27 Reiser RW, Barefoot AC, Dietrich RF, Fogiel AJ, Johnson
WR and Scott MT, Application of microcolumn liquid
chromatography-continuous-ýow fast atom bombardment
mass spectrometry in environmental studies of sulfonylurea
herbicides. J Chrom 554:91–101 (1991).
28 Badon R, Bastide J and Sabadie J, Reactive de l’herbicide
metsulfuron-methyle, synthese des products de degradation.
Chemosphere 21:289–293 (1990).
29 Strek HJ, Fate of chlorsulfuron in the environment, 1. Labor-
atory evaluations. Pestic Sci 53:29–51 (1998).
30 Pons N and Barriuso E, Fate of metsulfuron-methyl in soils in
relation to pedo-climatic conditions. Pestic Sci 53:311–323
(1998).
31 Barrett MR, The environmental impact of pesticide degradates
in groundwater, in Herbicide Metabolites in Surface Water
and Ground Water, ACS Symposium Series 630, eds by
Meyer MT and Thurman EM. Ch 16, pp 200–225 (1995).
1 Ray TB, The site of action of the sulfonylurea herbicides. Proc
Brighton Crop Prot Conf – Weeds. pp 131–138 (1985).
2 Brown HM, Mode of action, crop selectivity, and soil relations
of the sulfonylurea herbicides. Pestic Sci 29:263–281 (1990).
3 Tomlin C (ed), The Pesticide Manual, British Crop Protection
Council, Farnham, Surrey, UK. Eleventh edn. pp 842–844
(1994).
4 Nordh-Christerson M and Bergstrom L, Field observation of
soil movement and residues of sulfonylureas in Sweden. Proc
Brighton Crop Prot Conf – Weeds. pp 1127–1132 (1989).
5 Wadd DJ and Drennan DSH, A üeld study of the persistence
and leaching of chlorsulfuron and metsulfuron-methyl. Proc
Brighton Crop Prot Conf. – Weeds, 1133–1138 (1989).
6 Walker A, Cotterill EG and Welch SJ, Adsorption and degra-
dation of chlorsulfuron and metsulfuron-methyl in soils
from diþerent depths. Weed Res 29:281–287 (1989).
7 Bergstrom L, Leaching of chlorsulfuron and metsulfuron-
methyl in three Swedish soils measured in üeld lysimeters. J
Environ Qual 19:701–706 (1990).
8 Beyer EM Jr, Duþy MJ, Hay JV and Schlueter DD, Sulfonyl-
ureas, in Herbicides: Chemistry, Degradation and Mode of
Action, ed by Kearney PC and Kaufman D, Dekker, New
York. Vol 3 pp 117–190 (1988).
9 Vega D, Bastide J and Poulain C, Degradation chimique ou
microbiologique des sulfonylurees dans le sol II. Cas de
Pestic Sci 55:434–445 (1999)
445