Tài liệu Drugs and Poisons in Humans - A Handbook of Practical Analysis (Part 60) ppt

7. 3
7. 3
© Springer-Verlag Berlin Heidelberg 2005
II.7.3 Glufosinate and
glyphosate
by Yasushi Hori and Manami Fujisawa
Introduction
Non-selective phosphorus-containing amino acid-type herbicides (PAAHs) to be used for foli-
age exhibit lower toxicities than paraquat and are easily obtainable; they, thus, have come into
wide use since 1980.  e PAAHs include glufosinate (GLUF), glyphosate (GLYP) and biala-
phos (BIAL). In Japan, there are many kinds of products containing GLUF and GLYP com-
mercially available, and the number of suicidal cases using them is increasing [1].
In acute poisoning by GLUF, there is a latent period for 4–60 h before appearance of poi-
soning symptoms, such as lowered consciousness levels, respiratory arrest and generalized
convulsion; when more than 100 mL of BASTA Fluid
®
(GLUF, 18.5 %; anion surfactant; blue-
green in color) is ingested, the physical conditions of the victim are seriously aggravated with
high incidence [2]. Respiratory controls, such as securance of the respiratory tract and arti cial
respiration, are very important for rescuing such victims. Since it is possible to predict the
aggravation of the GLUF poisoning for a victim from the time a er its ingestion and from a
blood GLUF concentration [3], the rapid analysis of blood GLUF becomes very meaningful
not to miss the timing for starting the respiratory control; it is critical to prevent a victim from
falling into the unfortunate turning point.
For analysis of GLUF and GLYP in biomedical specimens, various methods by a modi ed
technique of the standard GC-NPD with N-acetyl and O-methyl derivatizations [4], GC/MS
using tert-butyldimethylsilyl (t-BDMS) derivatization [5–7], TLC [8], HPLC with  uorescence
detection a er post-column derivatization using o-phthalaldehyde [9], HPLC with  uores-
cence detection a er pre-column derivatization using 9- uorenylmethyl chloroformate
(FMOC-Cl) [10], HPLC with UV detection a er pre-column derivatization using phenyl iso-
thiocyanate [11], ion chromatography with electrochemical detection without any derivatiza-

tion [12], LC/MS with N-acetyl and O-methyl derivatizations [13] and HPLC with UV detec-
tion a er pre-column derivatization using p-nitrobenzoyl chloride [14] were reported.
In this chapter, some details on GC/MS [7], HPLC with  uorescence detection [10] and
HPLC-UV [14], a er each derivatization for analysis of PAAHs, are described.
GC/MS analysis [7]
Reagents and their preparation
• GLUF (DL-homoalanin-4-yl(methyl)phosphinate monoammonium salt) and its metaboli-
te 3-methylphosphinicopropionic acid (MPPA) can be purchased from Wako Pure Chemi-
cal Industries, Ltd., Osaka, Japan; GLYP (N-(phosphonomethyl)glycine) and its metabolite
546 Glufosinate and glyphosate
aminomethyl phosphonic acid (AMPA) from Sigma (St. Louis, MO, USA).  eir chemical
structures are shown in
> Figure 3.1. Each compound was dissolved in 10 % methanol
aqueous solution; they are stable at least for 6 months under refrigeration.  e trace
amounts of these compounds can adsorb to glassware; when low levels of the compounds
are dealt with, the tools made of Te on should be used [13].
• DL-2-Amino-3-phosphonopropionic acid (APPA) purchased from Aldrich (Milwaukee,
WI, USA) is used as internal standard (IS)
a
and dissolved in 10 % methanol aqueous solu-
tion to prepare its 100 µg/mL solution.
• N-Methyl-N-(tert-butyldimethylsilyl)tri uoroacetamide (MTBSTFA) and N,N-dimethyl-
formamide can be purchased from Aldrich and should be stored in a dry state (not to be
contaminated by water).
• 0.1 M NaOH solution: the 1 M solution of reagent grade is diluted 10-fold with distilled
water.
• To construct calibration curves, various amounts of GLUF, MPPA, GLYP or AMPA to-
gether with a  xed amount of APPA (IS) are spiked into the extracts of the standard human
serum, evaporated to dryness and derivatized before analysis
b

.
GC/MS conditions
Instrument: a GC 17A gas chromatograph/a QP5050 mass spectrometer (Shimadzu Corp.,
Kyoto, Japan); column: DB-5MS (15 m × 0.25 mm i.d.,  lm thickness 0.25 µm, J&W Scienti c,
Folsom, CA, USA); column temperature: 80 °C (2 min) → 15 °C/min → 300 °C (5 min); carrier
gas: He; its  ow rate; 1.0 mL/min; injection: split/splitless mode (splitless for 2 min); split ratio:
10; injection amount: 1 µL; injection temperature: 300 °C; interface temperature: 280 °C; ion-
ization mode: EI; scan range: m/z 70–650.
Structures of glufosinate (GLUF) and glyphosate (GLYP) and their metabolites.
⊡ Figure 3.1
547
Procedure
i. As a specimen, serum, urine or stomach contents are used; 500 µL of undiluted serum or
500 µL of urine diluted 10-fold with distilled water is subjected to the following procedure.
ii.  e above specimen is mixed with 500 µL acetone, vortex-mixed and centrifuged (3,000 rpm,
5 min) for deproteinization
c
; 100 µL of the supernatant solution is subjected to the next step.
For stomach contents, they are diluted with distilled water appropriately and pass through a
membrane  lter (0.45 µm); 100 µL of the  ltrate is subjected to the next step.
iii. Isolute
®
HAX 100 mg cartridges (International Solvent Technology, Mid Glamorgan, UK)
d
,
which have anion exchanging and hydrophobic interaction properties, are used for
extraction. One of the cartridges is activated by passing 1 mL methanol, 1 mL of 0.1 M
NaOH solution and 1 mL distilled water through it at a  ow rate of 1 mL/min.
iv.  e above 100 µL specimen solution is mixed well with 10 µL of IS solution and 1 mL dis-
tilled water, and poured into the cartridge.  e pH of the solution should be 6.4–8.5; there-

fore, it is generally not necessary to adjust pH for the serum or urine specimens.
v.  e cartridge is washed with 1 mL distilled water, and a target compound and IS are eluted
with 500 µL of 1 M HCl solution/methanol (4:1) at a  ow rate of 500 µL/min.  e eluate is
evaporated to dryness under reduced pressure with warming at 50 °C.
vi.  e residue is mixed with 50 µL MTBSTFA and 50 µL N,N-dimethyl-formamide, son icated
for 2 min and heated at 80 °C for 30 min
e
for t-BDMS derivatization.
vii. A er cooling to room temperature, 1 µL of the  nal solution is injected into GC/MS.
Assessment of the method
In this method, solid-phase extraction was used to extract a PAAH and its metabolite in a bio-
medical specimen for their GC/MS analysis [15] a er t-BDMS derivatization.
> Figure 3.2
shows mass spectra of t-BDMS derivatives of GLUF, MPPA, GLYP, AMPA and APPA.  e base
peaks at m/z M–57 appear for all compounds.  e quantitation using the selected ion monitor-
ing (SIM) is made with each base peak (GLUF: m/z 466; MPPA: m/z 323; GLYP: m/z 454;
AMPA: m/z 396; APPA: m/z 568).
Various derivatization methods were reported for PAAHs [16]; the advantage of the use of
t-BDMS derivatization is the one-step reaction
f
, which completes in only 30 min. When pg
levels of PAAHs are derivatized with high e ciency, the N-acetyl and O-methyl derivatizations
using acetic acid and trimethyl orthoacetate are useful [17].
 e detection limit for both GLUF and GLYP in the scan mode is about 100 pg on-column
(about 0.1 µg/mL in bood); that of both MPPA and AMPA is about 10 pg on-column. In the
SIM mode, the CV values re ecting reproducibility for the 4 compounds (100 ng each for de-
rivatization) using APPA as IS are not larger than 3 % (n = 5); GLUF and GLYP show linearity
in the range of 100 pg–100 ng on-column.  e detection limit (S/N ratio = 5) of GLUF and
GLYP in the SIM mode is about 10 pg on-column; that of MPPA and AMPA is even lower.
 e recovery rates for GLUF and GLYP, which had been spiked into sera at a concentration

of 1 µg/mL, a er extraction with the Isolute
®
HAX cartridge, were as good as 93.3 ± 6.7 %
(n = 5) and 92.6 ± 7.2 % (n = 5), respectively. Upon extraction with the cartridge, a urine spec-
imen should be diluted su ciently, because in the presence of strong anions in a specimen, the
recovery rate becomes low. Since unchanged forms of GLUF and GLYP are rapidly excreted
GC/MS analysis
548 Glufosinate and glyphosate
EI mass spectra of t-BDMS derivatives of GLUF, MPPA, GLYP, AMPA and APPA (IS).
⊡ Figure 3.2
549
TIC (upper panel) and SIM chromatograms (lower panel) obtained by GC/MS for t-BDMS
derivatives of GLUF, MPPA, GLYP, AMPA and APPA (IS) (10 ng each on-column).
⊡ Figure 3.3
TIC (upper panel) and SIM chromatograms (lower panel) obtained by GC/MS for an extract of
serum of a patient, who had ingested a GLUF-containing herbicide.
⊡ Figure 3.4
GC/MS analysis
550 Glufosinate and glyphosate
into urine, there are many cases, in which they are su ciently detectable even from 100-fold
diluted urine.
> Figure 3.3 shows a TIC and SIM chromatograms for t-BDMS derivatives of the authen-
tic GLUF, MPPA, GLYP, AMPA and APPA (IS);
> Figure 3.4 shows comparable chromato-
grams for the extract of serum, which had been obtained from a poisoned victim 7 h a er in-
gestion of 80 mL of a GLUF product (BASTA Fluid
®
). Using the base peaks at m/z M–57,
GLUF and MPPA could be speci cally detected by SIM from the extract of the crude matrix
obtained from the actual case; the concentrations of GLUF and MPPA were 74.3 and 0.32 µg/mL,

respectively.
HPLC analysis with fluorescence detection [10]
Reagents and their preparation
• FMOC-CL is purchased from Sigma, and dissolved in acetone to prepare its 0.1 % solution
just before use.
• Borate bu er solution (0.1 M, pH 8.5): 2 g of sodium tetraborate is dissolved in 100 mL
distilled water and the pH is adjusted to 8.5 with 2 M HCl solution.
• Phosphate bu er solution (10 mM, pH 2.5): 240 mg of sodium dihydrogenphosphate is
dissolved in 200 mL distilled water and the pH is adjusted to 2.5 with phosphoric acid.
HPLC conditions
Instruments: an LC-10ADVP pump, a CTO-10ACVP column oven, an RF10AXL  uoro-
photometer, an SIL-10ADVP autosampler, a CLASS-VP analysis so ware (all from Shimadzu
Corp.); column: Inertsil
®
ODS-2 (150 × 4.6 mm i.d., particle size 5 µm, GL Sciences, Tokyo,
Japan); column temperature: 40 °C; mobile phase: acetonitrile/10 mM phosphate bu er solu-
tion (ph 2.5); gradient elution: the ratio of the above acetonitrile and 10 mM phosphate bu er
solution of the mobile phase is held at 3:7 (v/v) for 7 min, and changed to 1:1 a er 13 min and
to 8:2 a er 15 min (1 min-hold) (the gradient elution a er 13 min is conducted for washing the
column);  ow rate of the mobile phase: 1 mL/min; detector: a  uorophotometer; excitation
wavelength: 265 nm; emission wavelength: 315 nm.
Procedure
i. A 100-µL volume of undiluted serum or urine diluted 10-fold with 0.1 M borate bu er
solution (pH 8.5) is mixed with 400 µL of 0.1 M borate bu er solution and 1 mL acetone.
ii.  e above solution is vortex-mixed and centrifuged at 3,000 rpm for 5 min for depro-
teinization; the resulting supernatant solution is subjected to the below derivatization.
iii. A 50-µL volume of the above solution is mixed with 200 µL of 0.1 M borate bu er solution
(pH 8.5) and 200 µL of 0.1 % FMOC-Cl acetone solution, capped and mixed well
g
.  e

mixture is incubated at 40 °C for 10 min.
551
iv. A 500-µL volume of ethyl acetate is added to the mixture and shaken to remove excessive
FMOC-Cl. A 100-µL aliquot of the aqueous layer is mixed with 400 µL of 0.1 M borate
bu er solution (pH 8.5); 10 µL of it is injected into HPLC.
Assessment of the method
> Figure 3.5 shows HPLC chromatograms for the authentic standard solutions of GLUF and
GLYP (10 µg/mL) and for an extract of serum sampled 8 h a er ingestion from an actual poi-
soned patient, who had ingested 25 mL of BASTA Fluid
®
.
GLUF and GLYP are highly polar in their unchanged forms, and thus suitable for separa-
tion by ion-exchange chromatography. A er derivatization with FMOC-Cl, the compounds
become separable by reversed phase HPLC using an acetonitrile-phosphate mobile phase.
 e derivatization can be completed under mild conditions at 40 °C for 10 min; the deriva-
tives are stable for at least 19 h.  e detection limit is as low as about 1 ng/mL in a specimen;
the whole procedure is accomplished in about 40 min.
⊡ Figure 3.5
Chromatograms obtained by HPLC with fluorescence detection for the authentic GLUF and GLYP
(upper panel) and for an extract of serum of a patient, who had ingested a GLUF-containing
herbicide (lower panel), after derivatization with FMOC-Cl.
HPLC analysis with fl uorescence detection
552 Glufosinate and glyphosate
 e method with  uorescence detection is higher in both speci city and sensitivity than
that with UV detection.  ere is another report [9] dealing with HPLC with  uorescence de-
tection of PAAHs, in which post-column derivatization with o-phthalaldehyde is employed.
However, it requires a special device for post-column derivatization. By the method described
in this section, PAAHs can be simply measured only by combining usual reversed phase HPLC
with a  uorescence detector.
HPLC analysis with UV absorption detection

Reagents and their preparation
• p-Nitrobenzoyl chloride (PNBC) can be obtained from Aldrich (Milwaukee, WI, USA)
and other manufacturers; it is dissolved in acetonitrile (of the highest purity)
h
to prepare
1 % solution just before use.
• Borate bu er solution (0.1 M, pH 8.5): 2 g of sodium tetraborate is dissolved in 100 mL
distilled water and the pH is adjusted to 8.5 with 2 M HCl solution.
• Ammonium acetate solution (10 mM, pH 5): 154 mg of ammonium acetate (of the highest
purity) is dissolved in 200 mL of ultra-pure distilled water and its pH is adjusted to 5 with
acetic acid.
HPLC conditions
Instruments: the same pump, column oven, autosampler and so ware as described in the sec-
tion of HPLC analysis with  uorescence detection, and an SPD-M10AVP diode array detector
(all from Shimadzu Corp.); column: Inertsil
®
Ph-3 (150 × 4.6 mm i. d., particle size 5 µm, GL
Sciences); column temperature: 40 °C; mobile phase: acetonitrile/10 mM ammonium acetate
solution (pH 5.0) (1:9, v/v);  ow rate: 0.8 mL/min; detection wavelength: 272 nm.
Procedure
i. A 500-µL volume of undiluted serum or urine diluted 10-fold with distilled water is mixed
with 500 µL acetone, vortex-mixed and centrifuged at 3,000 rpm for 5 min for deprotein-
ization.
ii. A 100-µL aliquot of the above supernatant solution is mixed with 200 µL of 0.1 M borate
bu er solution (pH 8.5) and 100 µL of 1 % PNBC acetonitrile solution, and le at 22–25 °C
for 10 min; 10 µL of the solution is injected into HPLC.
Assessment of the method
 e advantages of this method are that the derivatization reaction is completed in 10 min at
room temperature and that a usual reversed phase HPLC-UV detection can be used; with the
553

minimum instruments and time, the screening and quantitation of GLUF and GLYP can be
achieved.
> Figure 3.6 shows HPLC chromatograms for the authentic GLUF and GLYP and for the
extract of serum, into which GLUF and GLYP had been spiked. Since the polarity of the com-
pounds is high even a er derivatization, their peaks appear at early retention times by the re-
versed phase HPLC.  e peak 3 shown in the upper chromatogram of
> Figure 3.6 corre-
sponds to the unreacted reagent (PNBC); the peak 4 to p-nitrobenzoic acid formed from PNBC
by its reaction with water.  e λ
max
wavelengths for the derivatives of GLUF and GLYP are
272.8 and 273.1 nm, respectively (
> Figure 3.7).
When a usual ODS column is used, the GLYP peak appears without any interference, but
the GLUF peak may be interfered with by impurity peaks derived from the crude matrix. By
using the Inertsil
®
Ph-3 column, which includes phenyl groups for their interaction with the
target compounds, the GLUF peak can be better separated from impurities.
 e detection limit for both authentic GLUF and GLYP in clean solution is 0.01 µg/mL;
while that for GLUF and GLYP in serum or urine is 0.1 µg/mL.  e average recovery of GLUF
from sera at the concentration of 1.0 µg/mL was as good as 95.3 % (n = 5); that from urine at
10.0 µg/mL 97.3 % (n = 5).
Chromatograms obtained by HPLC with UV absorption detection for the authentic GLUF and
GLYP (a) and for an extract of serum (b), into which GLUF and GLYP had been spiked, after
derivatization with PNBC. Peaks 3 and 4 are due to the unreacted reagent (PNBC) and a by-
product (p-nitrobenzoic acid), respectively.
⊡ Figure 3.6
HPLC analysis with UV absorption detection
554 Glufosinate and glyphosate

Toxic and fatal concentrations
GLUF [18]
 e number of poisoning cases by GLUF counts 100–200 per year; most cases are due to suicidal
ingestion of GLUF products. Its main products are BASTA Fluid
®
(GLUF, 18.5 % anion surfactant,
30 %; blue-green color) and Hayabusa
®
(GLUF, 8.5 %; anion surfactant, 50 %; blue color).
 e oral LD
50
values (mg/kg) for GLUF are 1,660/1,510 (male/female) in rats and 436/464
in mice. In humans, the poisoning symptoms become severe, when more than 100 mL of the
18.5 % solution of GLUF is ingested.  e oral LD
50
value (mg/kg) for the anion surfactant
being contained in the BASTA Fluid
®
is 4,500 in rats.
In GLUF poisoning, there is a characteristic latent period without any symptom lasting for
not less than 6 h; a er this period, poisoning symptoms, such as lowering of the consciousness
level, respiratory suppression and generalized convulsion, suddenly appear.  e severity of the
GLUF poisoning can be predicted by plotting the time a er ingestion on the horizontal axis
and the logarithm of serum GLUF concentration on the vertical axis. Koyama et al. [2] mea-
sured serum GLUF levels in 99 patients with GLUF poisoning, and drew two linear lines A and
B by connecting a point of 70 µg/mL at 2 h a er GLUF ingestion with a point of 5 µg/mL at 8 h
for A and by connecting a point of 200 µg/mL at 2 h with that of 15 µg/mL at 8 h for B.  ey
reported that any plot below line A indicated a mild case, and one above line B a severe case; in
the area between lines A and B mild and severe cases were mixed. Both GLUF and coexisting
surfactant seem exerting toxic e ects in many GLUF poisoning cases.

GLUF shows contradictory e ects on the central nervous system, viz., its excitation
and suppression; it may act on glutamate synthase, glutamate decarboxylase and inhibitory
glutamic acid receptors.  e surfactant contained in the GLUF product is being considered
responsible for vomiting, erosion of the upper digestive tracts, edema appearing from the oral
mucosa to the larynx and shock accompanied by the peripheral resistance.
UV absorption spectra for GLUF and GLYP after derivatization with PNBC.
⊡ Figure 3.7
555
About 20 % of total GLUF, which had been orally administered to rats, is rapidly absorbed
into the animal bodies; about 90 % of the absorbed GLUF is excreted into urine also rapidly. In
humans, a peak serum GLUF concentration is observed 40–50 min a er ingestion; more than
95 % of an absorbed amount of the compound is excreted into urine within 24 h; a major part
of the excreted compounds is in the unchanged form.  e author con rmed that the concen-
tration ratios of MPPA to GLUF in urine were only 0.005–0.01.
Toxicokinetic parameters were reported in 2 patients, who had ingested the BASTA Fluid
®
;
the results in one case [19] were: distribution half-life, 1.8 h; elimination half-life, 9.6 h; and
distribution volume in the body, 1.4 L/kg. In another case [20], they were: distribution half-life
1.2 h; elimination half-life, 9.2 h; and distribution volume in the body, 1.9 L/kg.
It is considered that 99 % of GLUF in human serum is not bound with proteins, but exists
in its free form [21]. It is easily expected that GLUF hardly passes through the blood-brain
barrier, because of its high polarity. However, there is a case report describing a serum GLUF
concentration at 31.7 µg/mL and a cerebrospinal  uid GLUF concentration at 0.4 µg/mL 4 h
a er ingestion; these values suggest that GLUF can be incorporated into the brain.
GLYP [22]
 e main products of GLYP are Roundup
®
 uid (GLYP, 41 %; anion surfactant, 15 %; yellow-
brown color; odorless  uid at pH 4.8), Touchdown

®
and Impulse
®
 uid.
GLYP exerts its herbicidal action by inhibiting the biosynthesis of chlorophylls and carot-
enoids and is said not to be active on mammals.  e oral LD
50
values (mg/kg) of GLYP are
6,250 and 7,810 for male and female rats, respectively; the percutaneous LD
50
value in rabbits
is as high as 5,000.  e oral LD
50
value of the Roundup
®
 uid is 2 mL/kg in humans. Masui et
al. [23] reported that either oral administration of 15 % surfactant or 41 % GLYP did not cause
fatalities of animals, but the mixture of them caused fatalities for all animals, and thus pointed
out their synergistic e ect. Nowadays, the acute toxicity of a GLYP product is said to be mainly
due to the surfactant; the poisoning symptoms are stimulation of the digestive tract, vomiting
due to its erosive e ect, diarrhea, bleeding of the digestive tract, edema of the intestines,
enhanced permeability of the vessels, generalized edema due to swelling of the cells and  nally
a shock state due to reduced total blood volume.
In animal experiments, about 30 % of an orally administered amount GLYP is absorbed
into bodies through the digestive tract, and a peak blood GLYP concentration can be attained
in 3–4 h.  e main excretion route is the urinary system; but a part is excreted into feces.
Notes
a) As an IS except APPA, N-(phosphonomethyl)-β-alanine [13] is usable, because its proper-
ties for the extraction and the t-BDMS derivatization are almost the same as those of
PAAHs; the compound is not commercially available and thus should be synthesized.  ere

are reports using n-docosane (Aldrich) as IS [15, 17], but the authors have no experience
of using them.
b)  e reproducibility of GC/MS analysis of trace amounts (not larger than 10 ng on-column)
of the authentic compound only a er its evaporation and derivatization with MTBSTFA is
Toxic and fatal concentrations
556 Glufosinate and glyphosate
bad.  e cause of this variation is not clear, but the addition of the blank serum extract
markedly improves the reproducibility.
c)  e ratios of GULF bound with serum proteins to total GLUF are not larger than 1 % [21],
but the deproteinization process is required to enhance the derivatization e ciency.
d)  e authors have introduced Isolute
®
HAX cartridges in this chapter. Other cartridges or
columns with similar properties, such as Bond Elut Certify
®
II (Varian, Harbor City, CA,
USA) and Oasis
®
MAX (Waters, Milford, MA, USA), can be also used; they have both
hydrophobic interaction and anion exchanging properties for extraction.  ese mixed
mode cartridges (columns) are less in uenced by variation of ion intensities of specimens,
and thus give more reproducible results.
e) In Tsunoda’s report [15], the derivatization was completed at 80 °C in 30–80 min. In authors’
experiments, su cient quantitativeness and reproducibility were secured by derivatization
at 80 °C for 30 min.
f)  e acylation with halogenated acid anhydride-halogenated alcohol also gives one-step
derivatization reaction (100 °C, 1 h) for PAAHs; this kind of derivatization is more suitable
for trace level ranges, because of better derivatization e ciency [16]. When the resulting
derivatives are analyzed with a DB-5MS column, enantiomers are separated for GLUF and
APPA, giving 2 peaks for each compound.

g) When 50 µL of a specimen is mixed with 200 µL of 0.1 M borate bu er solution (pH 8.5)
and 200 µL of 0.1 % FMOC-Cl acetone solution, boric acid occasionally precipitates; this
may cause clogging of the autoinjector. Such precipitates should be removed by passing the
solution through a membrane  lter (0.45 µm) before injection into HPLC.
h) PNBC easily reacts with water to produce p-nitrobenzoic acid. Although the latter com-
pound does not interfere with the assays, ultra-pure acetonitrile without any water should
be used.
References
1) Tsunoda N (1990) Phosphorus-containing amino acid type herbicides. Jpn J Forensic Toxicol 8:100–111 (in
Japanese)
2) Koyama K, Hirose Y, Taze C et al. (2000) An indicator of aggravation in poisoning by intake of BASTA Fluid
®
;
comparison of estimated intake amounts with serum glufosinate concentrations ([GLF]s). Jpn J Toxicol 13:469
(in Japanese)
3) Koyama K, Hirose Y, Okuda T et al. (1997) Relationship between aggravation of poisoning by intake of a glufos-
inate-containing herbicide (BASTA Fluid
®
) and serum glufosinate concentrations. J Jpn Assoc Acute Med
8:617–618 (in Japanese)
4) Goto M, Kato M (1987) Method for Analysis of Residual Pesticides, Enlarged edn. Soft Science, Tokyo, p 236 (in
Japanese)
5) Kageura M, Hieda Y, Hara K et al. (1988) Analysis of glyphosate and (aminomethyl) phosphonic acid in a sus-
pected poisoning case. Jpn J Legal Med 42:128–132
6) Tsunoda N (1994) Analysis of phosphorus-containing amino acid-type herbicides and their problems in foren-
sic examination. Jpn J Forensic Toxicol 12:104–107 (in Japanese with an English abstract)
7) Hori Y, Fujisawa M, Shimada K et al. (2001) Determination of glufosinate ammonium and its metabolite 3-meth-
ylphosphinicopropionic acid in human serum by gas chromatography-mass spectrometry following mixed-
mode solid phase extraction and t-BDMS derivatization. J Anal Toxicol 25:680–684
8) Suzuki A, Kawana M (1989) Rapid and simple method for identification of glufosinate- ammonium using paper

chromatography. Bull Environ Contam Toxicol 43:17–21
9) Okuda T, Naotsuka K, Sameshima I et al. (1993) A new HPLC analysis of glufosinate caused acute poisoning. Jpn
J TDM 9:39–44
557
10) Akuzawa N, Akaiwa H (1997) Rapid determination of DL-homoalanin-4-yl-(methyl)phosphinic acid by high-
performance liquid chromatography with fluorescence detection. Bunseki Kagaku 46:69–74 (in Japanese with
an English abstract)
11) Nishida K, Narihara M, Tsutsumi K et al. (1998) A study on a screening method for water-soluble herbicides.
Abstracts of Annual Meeting of Japanese Association of Science and Technology for Identification, p 69 (in
Japanese)
12) Sato K, Suetsugu K, Takegoshi Y et al. (2000) Ion chromatography with electrochemical detection for phospho-
rus-containing amino acid type herbicides. Abstracts of Annual Meeting of Japanese Association of Science
and Technology for Identification, p 84 (in Japanese)
13) Honda M, Sato M, Kikuchi M et al. (2001) Simultaneous analysis of phosphorus-containing amino acid type
herbicides by LC/MS. Jpn J Forensic Toxicol 19:176–177 (in Japanese with an English abstract)
14) Hori Y, Fujisawa M, Shimada K et al. (2002) Quantitative determination of glufosinate in samples by HPLC with
UV detector after p-nitrobenzoyl derivatization. J Chromatogr B 767:255–262
15) Tsunoda N (1993) Simultaneous determination of the herbicides glyphosate, glufosinate and bialaphos and
their metabolites by capillary gas chromatography-ion-trap mass spectrometry. J Chromatogr 637:167–173
16) Stalikas CD, Konidari CN (2001) Analytical methods to determine phosphonic and amino acid group-contain-
ing pesticides. J Chromatogr A 907:1–19
17) Stalikas CD, Pilidis GA (2000) Development of a method for the simultaneous determination of phosphoric and
amino acid group containing pesticides by gas chromatography with massspective detection. Optimization of
the derivatization procedure using an experimental design approach. J Chromatogr A 872:215–225
18) Koyama K (2001) Glufosinate (BASTA Fluid
®
). Jpn J Acute Med 25:141–143 (in Japanese)
19) Hirose Y, Kobayashi M, Koyama K et al. (1999) A toxicokinetic analysis in a patient with acute glufosinate poison-
ing. Hum Exp Toxicol 18:305–308
20) Honda T, Nakashima K, Higashi H et al. (1999) Analysis of toxicokinetics of glufosinate (GLF) in a case of poison-

ing by oral intake of BASTA Fluid
®
. Jpn J Toxicol 12:468 (in Japanese)
21) Hori Y, Koyama K, Fujisawa M et al. (2001) Protein binding of glufosinate and factors affecting it revealed by an
equilibrium dialysis technique. J Anal Toxicol 25:439–442
22) Japan Poison Information Center (ed) (2000) Glyphosate. In:Poisoning Accidents and Their Countermeasures
with Special Reference to Actual Cases, revised edn. Jiho Inc., Tokyo, pp 220–223 (in Japanese)
23) Masui M, Ikeda H, Takagi K et al. (1988) Forensic toxicological studies on the herbicide Roundup
®
(II). Jpn J Legal
Med 42 (Suppl. 116) (in Japanese)
HPLC analysis with UV absorption detection