Tài liệu Drugs and Poisons in Humans - A Handbook of Practical Analysis (Part 51) doc

6.1
6.1
© Springer-Verlag Berlin Heidelberg 2005
II.6.1 Aconite toxins
by Michinao Mizugaki and Kitae Ito
Introduction
 e aconite plants contain Aconitum alkaloids (AAs) and other minor components, such as
chasmanine, kobusine and higenamine. AAs consist of aconitines, benzoylaconines and aco-
nines as shown in
> Figure 1.1.  e most toxic group is the aconitines, including aconitine,
mesaconitine, hypaconitine, and jesaconitine; this group is one of the most poisonous com-
pounds being contained in the plant kingdom.
Even nowadays, aconite poisoning cases take place occasionally.  ese may be due to ac-
cidental, suicidal or homicidal ingestion of the plant itself or its extracts.
In this chapter, a speci c method for GC/MS analysis of AAs in human specimens is pre-
sented.
Reagents and their preparation
• Aconitine was purchased from Sigma (St. Louis, MO, USA); mesaconitine and hypaconi-
tine from Kishida Kagaku (Osaka, Japan). Jesaconitine, benzoylaconine, benzoylmesaco-
nine, benzoylhypaconine, 14-anisoylaconine, aconine, mesaconine and hypaconine were
donated by Tsumura (Tokyo, Japan). Benzoylaconine and aconine can be obtained also
Structures of Aconitum alkaloids (AAs).
⊡ Figure 1.1
456 Aconite toxins
from Sanwa Shoyaku (Tochigi, Japan). As a trimethylsilylating reagent, N,O-bis(trime thyl-
silyl)tri uor o acet amide (BSTFA), containing 1 % trimethylsilylchlorosilane (TMCS)
a
(Pierce,
Rockford, IL, USA) is used. Pyridine to be used is of amino acid sequence analysis grade
(Wako Pure Chemical Industries, Ltd., Osaka, Japan or Sigma). Other organic solvents are
of HPLC grade; other reagents are of special grade.

• Each compound of aconitines, benzoylaconines and aconines is dissolved in acetonitrile to
prepare the 1 µg/mL standard solution separately.
Instrumental conditions
i. Instrument
A DX-303 GC/MS instrument (JEOL, Tokyo, Japan) equipped with a Van den Berg type solvent-
less injection device and a DA-5000 data processor (JEOL).
ii. GC/MS conditions
Ionization: EI; ion source temperature: 250 °C; electron energy: 70 eV; trap current: 0.3 mA;
accelerating voltage: 3 kV; GC column: a DB-5 chemical bond fused silica capillary column
(15 m × 0.25 mm i. d.,  lm thickness 0.25 µm, J&W Scienti c, Folsom, CA, USA); column
(oven) temperature: 250 °C→16 °C/min→320 °C; injection temperature: 320 °C; carrier gas:
He; its  ow rate: 25 m/min (linear velocity at 250 °C); scan range: m/z 100–800; scan speed:
1.2 s (repetition time, 2 s); resolution: 1,000.
iii. Selected ion monitoring (SIM)
 e instrumental conditions are the same as above. Each base peak of the trimethylsilyl (TMS)
derivatives of the alkaloids is used for SIM. For hypaconitine, mesaconine, aconitine and jesa-
conitine, peaks at m/z 596, 684, 698 and 728 are used, respectively. For benzoylhypaconine,
benzoylmesaconine and benzoylaconine, those at m/z 686, 774 and 788 are used, respectively;
for hypaconine, mesaconine and aconine, those at m/z 654, 742 and 756 are used as monitoring
ions, respectively.
Procedures
i. Construction of the calibration curve
i. For plotting di erent concentrations, 5, 10, 25, 50 and 100 µL aliquots of each AA standard
solution (1 µg/mL) are placed in glass vials.
ii.  ey are evaporated to dryness under a stream of nitrogen.
iii. Each residue is dissolved in 50 µL pyridine, followed by addition of 50 µL of 1 % TMCS-
BSTFA and being capped airtightly.
iv.  e derivatization reaction is made by leaving the mixture vials overnight at room temper-
ature

b
.
v. A 1-µL aliquot of each solution is injected into GC/MS.
vi. Peak areas are plotted against compound concentrations to construct an external calibra-
tion curve.
457Aconite toxins
ii. Extraction and derivatization procedure
i. A 1.0-mL volume of a body  uid specimen (whole blood, serum or urine
c
) and 10 mL
methanol
d
are placed in a glass test tube and mixed well for deproteinization.
ii. It is centrifuged at 3,000 g for 10 min to obtain supernatant solution.
iii. To the sediment, 10 mL methanol is again added, mixed well and centrifuged to obtain
the second supernatant solution.
iv.  e supernatant solutions are combined and evaporated to dryness at 40 °C under re-
duced pressure.
v.  e residue is dissolved in 0.25 mL acetonotrile and mixed well.
vi. A Bond Elut SI cartridge (Analytical International, Harbor City, CA, USA) is equilibrated
with 5 mL n-hexane.
vii.  e above acetonitrile solution is poured into the silica gel cartridge.
viii. To the container (or a test tube) at the step v), 0.25 mL acetonitrile is again added to rinse
it well; the solution is also poured into the same cartridge.
ix.  e cartridge is washed with 10 mL chloroform.
x. It is further washed with 10 mL ethyl acetate.
xi. A 20-mL volume of diethylamine/chloroform (1:1) is passed through the cartridge to
elute AA compounds.
xii.  e eluate is evaporated to dryness under reduced pressure.
xiii.  e residue is dissolved in 50 µL pyridine, followed by the addition of 50 µL of 1 %

TMCS-BSTFA for derivatization at room temperature overnight
e
. A 1-µL aliquot of the
resulting solution is injected into GC/MS.
xiv.  e peak area of each AA obtained by SIM for a specimen is applied to the above calibra-
tion curve to calculate its concentration.
Assessment of the method
Aconite toxins can exert their toxic e ects at very low concentrations, which means that
their concentrations in human body  uids are very low.  erefore, it was di cult to identify
and quantitate them by the conventional HPLC method, because of its low sensitivity and spe-
ci city.
In this chapter, a sensitive GC/MS method for simultaneous analysis of AAs, including
aconitines [1], benzoylaconines and aconines, has been presented [2].  is method allows
accurate and simultaneous quantitation of AAs in small volumes of specimens.
i. EI mass spectra of AA-TMS derivatives
> Figures 1.2–1.4 show EI mass spectra of TMS derivatives of aconitines, benzoylaconines
and aconines.

• TMS derivatives of aconitines
By the derivatization with the above reagents, aconitine, mesaconitine and jesaconitine
give bis-TMS derivatives; while hypaconitine a mono-TMS derivative. In each mass spec-
trum, a [M–CH
3
COOH–OCH
3
]
+
ion appears as the base peak; a few small fragment peaks
also appeared (
> Figure 1.2).

458 Aconite toxins
• TMS derivatives of benzoylaconines
Benzoylaconine and benzoylmesaconine give tri-TMS derivatives; while benzoylhypaco-
nine a bis-TMS derivative. In each mass spectrum, a [M–OCH
3
]
+
ion appears as the base
peak; other fragment peaks are a few and small (
> Figure 1.3).

• TMS derivatives of aconines
Aconine and mesaconine give tetra-TMS derivatives; while hypaconine a tri-TMS deriva-
tive. In each mass spectrum, a [M–OCH
3
]
+
ion appears as the base peak (> Figure 1.4).
As shown in
> Figures. 1.2–1.4, the ratio of each base peak to the total abundance is rela-
tively high; the base peaks are very useful for trace quantitative analysis by GC/MS-SIM.
ii. Reliability of the method
> Figure 1.5 shows SIM chromatograms for an extract of 1 mL serum, into which 50 ng each
of 9 kinds of aconitines and their hydrolysis products had been spiked. For every compound,
a sharp peak
f
appeared at the same retention time as that of the authentic one; no interfering
impurity peaks were observed.
 e calibration curve consisting of peak area on the vertical axis and compound amount
(in an injected volume) on the horizontal axis showed good linearity (r

2
= 0.999) in the range
of 100 pg–7.5 ng for each alkaloid.  e detection limits by this method were about 10 pg (S/N 10)
in an injected volume, enabling highly sensitive analysis.
 e recoveries of 50 ng each of AAs, which had been spiked into 1 mL human serum, were
85.2–94.4 %.  e above pretreatment procedure can be applied to urine, whole blood and
tissue
g
specimens. In addition, the authors synthesized d
5
-aconitine for use as internal stan-
dard (IS), resulting in higher precision and sensitivity in aconitine analysis
h
[3].
Poisoning cases, and toxic and fatal concentrations
 e aconitines are neurotoxic and cardiotoxic; they exert toxic e ect by acting on the
gate mechanisms of sodium-ion channels of cell membranes and by causing hyperpolarization
of cells [4]. One of the most important poisoning symptoms is arrythmia. Arrythmia is
sometimes changed to ventricular  brillation, and cardiac and respiratory arrest resulting in
death [5].
 e history of aconite as a poison is very long, and dates back to ancient times [6, 7].
However, no reports on the concentrations of the aconite toxins (aconitine, mesaconitine,
hypaconitine and jesaconitine) were available, because of the low sensitivity of analytical
methods.
Now, the authors are undertaking the analysis of AAs in body  uids of aconite-poisoned
patients in every area of Japan, and thus accumulating the data [8–14].
> Table 1.1 shows the
concentrations of AAs in blood and urine in 6 poisoning cases.
A 50-year-old female attempted suicide by crunching and ingesting one and a half of
thumb-sized tubers of aconite, and visited a doctor 30 min later. She had had a history of de-

pression and had received psychiatric treatments.  e numbness of the tongue appeared in the
midst of her eating; but her consciousness was clear and noti ed the doctor of her ingestion of
aconite for the purpose of suicide. Her blood pressure was 100/60 mmHg; her heart beat 110/
min and arrythmic. She complained of nausea, vomiting and the numbness of her whole body;
459
EI mass spectra for TMS derivatives of aconitines. (a): aconitine; (b): mesaconitine;
(c) hypaconitine; (d): jesaconitine.
⊡ Figure 1.2
Poisoning cases, and toxic and fatal concentrations
460 Aconite toxins
EI mass spectra for TMS derivatives of benzoylaconines. (a): benzoylaconine; (b) benzoylmasaco-
nine; (c): benzoylhypaconine.
⊡ Figure 1.3
461
EI mass spectra for TMS derivatives of aconines. (a) aconine; (b): mesaconine; (c): hypaconine.
⊡ Figure 1.4
Poisoning cases, and toxic and fatal concentrations
462 Aconite toxins
SIM chromatograms for TMS derivatives of AAs. Nine kinds of AAs (50 ng each) were spiked into
1 mL serum, and extracted according to the procedure; a 5-µL aliquot of the final extract solution
(50 µL) was injected into GC/MS.
⊡ Figure 1.5
she could not stand up by herself. Her electrocardiogram showed her continuing arrythmia;
the basic rhythm of her heart beat was the sinus one at about 100/min. Multiple-sourced and
frequent ventricular arrythmia, atrial arrythmia, intraventricular aberrant conduction, QT
elongation and torsades de pointes were observed.
Transfusion, gastrolavage and administration of activated charcoal and a purgative were
performed. For the arrythmias, lidocaine and disopyramide were administered at an early
stage, but were not e ective.  us, the administration of phenytoin was started, but unsuccess-
ful.  e blood pressure was lowered about 1 h a er ingestion, and her respiration was arrested

about 2.5 h a er; arti cial respiration, a er endotracheal intubation, was started.  erea er,
although administration of lidocaine, disopyramide and phenytoin was continued, they were
not e ective. From about 7 h a er ingestion, ventricular tachycardia continuing for about 10 s
began to take place frequently.  iopental was administered, because of excitement and vigor-
ous movement of her body. A er the administrations of magnesium sulfate and propranolol,
the arrythmias markedly decreased about 11 h a er ingestion.  erea er, her conditions were
improved smoothly. She was extubated about 24 h a er and discharged on day 6 without any
sequela.  e concentration values of AAs in this case are shown as the case No. 1 in

> Table 1.1.
As shown in the above case, in the typical aconite poisoning case, the oral numbness just
a er ingestion extends to whole body, followed by hypotension and various types of arryth-
mias; in the worst case, the arrythmias are aggravated into ventricular  brillation and  nally
death.
463
⊡ Table 1.1
Concentration of Aconitum alkaloids in blood and urine in their poisoning cases
Case
No.
Age Sex Poisoning
mode
Presence
or absence
of cardio-
pulmonary
arrest
Outcome Alkaloid Blood
conc.
(ng/mL)
Sampling

time after
ingestion
(h)
Alkaloid ( ng/mL) Urine
conc.
(ng/
mL)
Sampling
time after
ingestion
(h)
1 [10] 50 F suicidal + alive jesaconitine
aconitine
2.4
1.0
(2.5) jesaconitine
aconitine
285
74.6
(12)
2 33 F homicidal + dead aconitine 22.7 (29.1)*
mesaconitine 41.4 (51.0)* at autopsy
hypaconitine 37.8 (47.3)*
3 [11] 44 M homicidal + dead jesaconitine 430 at autopsy jesaconitine 1,070 at autopsy
4 [12] 40 F
F
suicidal + + dead jesaconitine 69.1 jesaconitine 238
aconitine 1.1 at autopsy aconitine 3.3 at autopsy
aconine 3.1 mesaconitine 3.6
aconine 3.1

5 [13]
45 M suicidal + alive mesaconitine 4.0 mesaconitine 52.9
aconitine 1.7 aconitine 20.7
benzoylmesaconine 0.7 hypaconitine 3.1
benzoylaconine 0.2 (3) benzoylmesaconine 28.4 (24)
benzoylhypaconine 0.2 benzoylaconine 4.1
mesaconine 0.2 benzoylhypaconine 6.0
aconine 0.3 mesaconine 3.5
hypaconine 0.2 aconine 2.0
hypaconine 4.7
6 [14] 58 F accidental** – alive ND (5)
* Corrected value.
** Diagnosed as aconite poisoning after analyzing leaves which the patient had eaten.
ND: below detection limit.
Poisoning cases, and toxic and fatal concentrations
464 Aconite toxins
ii. Lethal doses of aconitines
 e LD
50
values in mice for AAs are shown in > Table 1.2. According to the experiments for
aconitines using mice, the toxicity is highest for jesaconitine, followed by aconitine, mesaconi-
tine and hypaconitine [15].  e oral lethal doses of aconitine for humans were reported to be
1–2 mg [16, 17].
About 500 species of aconite genus plants are growing in the world. Even in the same spe-
cies, the composition ratio of the alkaloids and their contents di er according to seasons and
growing areas.  e contents of aconitines in the plant expressed per weight is highest in the
tubers, followed by the  owers, leaves and stems [15].  e authors measured contents of acon-
itines in the tubers of Aconitum species harvested at a mountain in Nishi-Shirakawa-gun, Fu-
kushima Prefecture, according to 4 seasons; the total contents of aconitines per g were 2–4 mg
on average [18].

 e cases Nos. 2 were fatal; their concentrations of AAs in blood and urine were higher
than those in the survived Nos. 1 and 5 cases. Like in the case No. 6 of mistaken eating of aco-
nite for an edible wild plant, the detection of AAs are sometimes di cult at relatively a long
time a er ingestion.
When the blood concentrations were compared with those in urine, the latter generally gave
higher AA values. In the case No. 5, his body  uids could be sampled for analysis according to
various time intervals a er ingestion; the AAs could not be detected from blood as early as on
the 2nd day, but some of the AAs could be detected from urine even on the 7th day [13].
AAs in blood disappear in a relatively short period, but they are excreted into urine in rela-
tively large amounts continuously.  erefore, even if a relatively long time elapses a er inges-
tion in a suspicious case of aconite poisoning, it is useful to analyze urine specimens.
 e hydrolytic compounds of aconitines (benzoylaconines and aconines) are being sug-
gested to be metabolites of aconitines [13].
 ere are various poisoning symptoms observable in aconite poisoning; the relationship
between the appearance of the symptoms and AA concentrations in body  uids remains to be
explored.
 e distribution of AAs in human organs obtained at autopsy for the victim in the case
No. 4, who had died 4 h a er ingestion, is shown in
> Table 1.3.  e AA concentrations ex-
pressed as ng/g wet weight were highest in the right lobe of the liver, followed by the le lobe of
⊡ Table 1.2
LD
50
values of AAs in mice (mg/kg)
Alkaloid p. o. subcut. i. p. i. v.
aconitine 1.8 0.270 0.380 0.12
mesaconitine 1.9 0.204 0.213 0.10
hypaconitine 5.8 1.190 1.100 0.47
jesaconitine 1.0–2.0 0.2–0.25
benzoylaconine 1500 280 70 23

benzoylmesaconine 810 230 240 21
benzoylhypaconine 830 130 120 23
aconine 120
The LD
50
values were calculated by the up and down method. The value for aconine was cited from reference [7]; for
other compounds, cited from reference [15].
465
the liver, kidney, heart, right lung, le lung, psoas major, adipose tissue (around psoas major),
cerebellum and cerebrum. Especially in right and le lobes of the liver and the kidney, very high
concentrations of AAs were found.  e ratio of jesaconitine concentration in a organ to that in
serum was not less than 3 for the right and le lobes of the liver and the kidney, showing the
accumulation of AAs in the organs. On the other hand, such ratio was low for psoas major, adi-
pose tissue, cerebellum and cerebrum; it was only 0.03 for the latter two organs.  ese results
suggest that the liver and kidney are useful for analysis of AAs in fatal poisoning cases.
> Table 1.4 shows AA levels in contents of the digestive tracts obtained from a victim in
the case No. 4 (
> Table 1.1) [12].  ey were highest in the ileum contents, followed by bile,
jejunum contents, stomach contents and duodenum contents.
 e high AA levels found in the kidney is in accordance with the high levels in urine. On the
other hand, the high AA levels in bile and the contents of the ileum and jejunum show another
excretion route for AAs via the digestive tract into feces in addition to the urinary route.
⊡ Table 1.3
Distribution of AAs in human organs in Case 4
Specimen jesaconitine aconitine (ng/g) mesaconitine tissue/serum ratio
right lobe of the liver 254 4.3 3.3 3.67
left lobe of the liver 233 4.2 3.8 3.37
kidney 217 2.8 1.9 3.14
heart 67.5 1.1 ND 0.98
right lung 63.7 1.8 ND 0.92

left lung 62.0 0.9 ND 0.90
psoas major 13.1 0.2 ND 0.19
adipose tissue 3.4 ND ND 0.05
cerebellum 2.3 ND ND 0.03
cerebrum 1.9 ND ND 0.03
AAs other than the above compounds were below the detection limits. The tissue/serum ratios are the data only for
jesaconitine. ND: below the detection limit.
⊡ Table 1.4
AA concentrations in the contents of digestive tracts in Case 4
Specimen jesa-
conitine
aconitine 14-anisoyl-
aconine (ng/g)
mesaconine aconine benzoyl-
aconine
ileum
contents
471 9.0 5.5 8.5 24.9 2.9
bile 238 6.3 17.2 3.2 ND ND
jejunum
contents
130 3.6 9.6 4.7 1.5 ND
stomach
contents
87.7 2.1 ND 2.1 ND ND
duodenum
contents
72.1 2.5 4.9 2.0 ND ND
AAs other than the above compounds were below the detection limits. ND: below the detection limit.
Poisoning cases, and toxic and fatal concentrations

466 Aconite toxins
Notes
a) Only with BSTFA, a single derivative form can be obtained for each of aconitines, but not for
benzoylaconines and aconines; the derivatized forms are multiple and variable for the latters.
 erefore, it is essential to use BSTFA plus TMCS for getting reproducible derivatives.
b) It can be replaced by incubation at 60 °C for 1 h.
c) To protect AAs from their hydrolysis, the specimens should be stored in a frozen state and
the contamination by alkaline compounds (solutions) should be avoided absolutely.
d) Ethanol can be used in place of methanol without a ecting recovery rates.
e)  e  nal sample solutions gradually become brownish according to time.  is is due to an
oxidation product of diethylamine; but it does not a ect the analysis.  e sample solution,
a er use, is stable at room temperature for at least one week under airtight conditions.
f)  e TMS derivatization of aconitines gives both big and small peaks for each compound in
the chromatogram; this is due to the formation of an isomer in a small part in the injection
chamber at high temperature.  e peak areas for the two peaks should be combined for
calculation of an AA concentration.
g) For a tissue specimen, a 1-g weight of a tissue is excised from an organ. One gram of it and
0.1 mL of puri ed water are placed in a small beaker and minced into small pieces with a
clean surgical scissors. A er addition of 10 mL ethanol, the tissue sample is homogenized
using a Polytron type homogenizer; the following procedure is the same as that described
for body  uid samples.
h) By using d
5
-aconitine as IS, the recovery rates become to be 96–103 %.
References
1) Mizugaki M, Oyama Y, Kimura K et al. (1988) Analysis of Aconitum alkaloids by gas chromatography/selected
ion monitoring. Eisei Kagaku 34:359–365 (in Japanese with an English abstract)
2) Ito K, Ohyama Y, Konishi Y et al. (1997) Method for the simultaneous determination of Aconitum alkaloids and
their hydrolysis products by gas chromatography-mass spectrometry in human serum. Planta Med 63:75–79
3) Ito K, Tanaka S, Konno S et al. (1998) Report on the preparation of deuterium-labelled aconitine and mesaconi-

tine and their application to the analysis of these alkaloids from body fluids as internal standard. J Chromatogr
B 714:197–203
4) Catterall WA (1988) Structure and function of voltage-sensitive ion channels. Science 242:50–61
5) Okada Y, Sasaki M, Suzukawa M et al. (1991) Aconitine cardiotoxicity. Jpn J Toxicol 4:135–141 (in Japanese)
6) Yamazaki M (1985) Story of poisons, Chukoshinsho. Chuokoronshinsha, Tokyo, pp 159–172 (in Japanese)
7) Bisset NG (1981) Arrow poisons in China. Part II. Aconitum-botany, chemistry, and pharmacology. J Ethnophar-
macol 4:247–336
8) Mizugaki M (1996) Studies on the aconites from the viewpoints of forensic medicine and pharmacology. Jpn J
Forensic Toxicol 14:1–12 (in Japanese with an English abstract)
9) Mizugaki M, Ito K (2001) Forensic toxicological studies on Aconitum toxins. Res Pract Forensic Med 44:1–16 (in
Japanese with an English abstract)
10) Goto S, Fuke N, Ishikawa Y et al. (2000) A case of Aconitum poisoning with suicidal attempt. J Clin Anaesth
24:1513–1515 (in Japanese)
11) Mori A, Mukaida M, Ishiyama I et al. (1990) Homicidal poisoning by aconite: report of a case from the viewpoint
of clinical forensic medicine. Jpn J Legal Med 44:352–357 (in Japanese with an English abstract)
12) Ito K, Tanaka S, Funayama M et al. (2000) Distribution of Aconitum alkaloids in body fluids and tissues in a
suicidal case of aconite ingestion. J Anal Toxicol 24:348–353
13) Mizugaki M, Ito K, Ohyama Y et al. (1998) Quantitative analysis of Aconitum alkaloids in the urine and serum of
a male attempting suicide by oral intake of aconite extract. J Anal Toxicol 22:336–340
467
14) Konya M, Nagata K, Nishimura Y et al. (1999) A case of accidental poisoning due to mistaken eating of aconite
in place of an Anemone plant. Intern Med 84:1194–1196 (in Japanese)
15) Ooizumi Y, Hikino H (1981) Pharmacology of “Bushi” and its components. Part 2. Study on “Fushi”. Shuppan-
kagaku-sougo-kenkyusho, Tokyo, pp 117–154 (in Japanese)
16) Camps FE (1968) Gradwohl’s Legal Medicine. John Wright & Sons, Bristol, p 674
17) Rentoul E, Smith H (1973) Glaister’s Medical Jurisprudence and Toxicology. Churchhill Livingstone, Edinburgh,
p 520
18) Ito K, Ohyama Y, Hishinuma T et al. (1996) Determination of Aconitum alkaloids in the tubers of Aconitum japo-
nicum using gas chromatography/selected ion monitoring. Planta Med 62:57–59
Poisoning cases, and toxic and fatal concentrations