Tài liệu Drugs and Poisons in Humans - A Handbook of Practical Analysis (Part 57) docx

6.7
6.7
© Springer-Verlag Berlin Heidelberg 2005
II.6.7 Oleander toxins
by Chiaki Fuke and Tomonori Arao
Introduction
Oleander ( Nerium oleander and Nerium indicum) is a relatively small evergreen tree of an
Indian origin, and growing in Honshu, Shikoku, Kyushu and Okinawa islands in Japan.  e
plant contains cardiac glycosides in its leaves, stems and  owers and is known as one of poison-
ous plants; poisoning and fatal cases for domestic animals and humans due to ingestion of this
plant were reported [1–6].  e main toxin of oleander is oleandrin.
Oleandrin can be measured using cross-reaction of an immunoassay kit for digoxin [1],
TLC [2], HPLC [7, 8] and LC/MS [3, 6]. Oleandrin is thermolabile; it is di cult to analyze it by
GC or GC/MS, because it gives 4 peaks due to decomposition.
In this chapter, a method for LC/MS analysis of oleandrin and its metabolite desacetylole-
andrin [9] together with their related compounds, such as oleandrigenin and gitoxigenin, con-
tained in human specimens, is presented.
 e structures and their molecular weights of oleandrin and its related compounds are
shown in
> Figure 7.1.
⊡ Figure 7.1
Structures and molecular weights of oleandrin and its related compounds.
Reagents and their preparation
• A 1-mg aliquot each of oleandrin, oleandrigenin, desacetyloleandrin
a
, gitoxigenin and digi-
toxigenin (Sigma, St. Louis, MO, USA) is dissolved in 10 mL acetonitrile (100 µg/mL)
separately.
• A 0.1-mL volume of the above digitoxigenin solution is diluted with acetonitrile to 10 mL
(1 µg/mL; internal standard, IS).
520 Oleander toxins

HPLC conditions
Instrument: a Hitachi M-8000 type LC/3DQMS system; column: GH-C18 (III) (150 × 2.1 mm
i.d., particle size 5 µm, Hitachi Ltd., Tokyo, Japan); column temperature: 40 °C; mobile phase:
methanol/water (6:4, v/v); its  ow rate: 0.2 mL/min.
MS conditions
Ionization: sonic spray ionization ( SSI)
b
; shield temperature: 250 °C; aperture-1 temperature:
150 °C; aperture-2 temperature: 120 °C; dri voltage: 70 V; ion detection mode: positive;
microscan: 5 s; mass defect: 55/100 amu; scan range: m/z 350–650; low mass cuto : m/z 120;
accumulation time: 500 ms.
MS/MS conditions
c
Ion accumulation step: Ion accumulation mass range: m/z 350–650; low mass cuto :
m/z 120; ion accumulation time: 300 ms; ion accumulation voltage: 0 V.
Ion isolation step (MS-1): Isolation mass range: m/z 595.48–602.77; low mass cuto :
m/z 569.06; isolation time: 10 ms; isolation voltage: 0.175 V.
CID step (MS-2): CID mass range: m/z 584.15–614.64; low mass cuto : m/z 190; CID time:
50 ms; CID voltage: 0.188 V.
Procedure
i. A 1-mL volume of a specimen
d
is mixed with 4 mL distilled water and 100 µL IS
solution.
ii.  e above mixture is extracted with 2 mL of 1-chlorobutane by shaking for 15 min.
iii. It is centrifuged at 2,000 g for 5 min; the organic phase is transferred to a test tube.
iv.  e steps ii and iii are repeated twice.
v.  e organic phases are combined and evaporated to dryness under a stream of nitrogen
with warming at 40 °C.
vi.  e residue is dissolved in 0.5 mL of 80 % methanol aqueous solution, and washed with

1 mL hexane twice
e
.
vii.  e 80 % methanol layer is evaporated to dryness under a stream of nitrogen with warm-
ing at 40 °C in a water bath.
viii.  e residue is dissolved in 100 µL mobile phase and centrifuged at 12,000 g for 5 min; a
5-µL aliquot of the supernatant solution is injected into LC/MS.
ix. Each calibration curve is constructed using spiked specimens with digitoxigenin as IS.
 e concentration of an oleander toxin in a specimen is calculated using the calibration
curve.
521
Assessment of the method
Oleandrin is one of cardiac glycosides and exerts its e ect at low concentrations. To detect its
therapeutic concentrations, the detection limit by an analytical method should be in the nano-
grams/mL order. When the present method was used in an oleander poisoning case, oleandrin
could be detected from blood and cerebrospined  uid (CSF), showing the applicability of the
method.
> Figure 7.2 shows a TIC and mass spectra for the authentic standards of the  ve com-
pounds.  e spectra showed intense [M + Na]
+
adduct ions at m/z 599 for oleandrin, m/z 557
for desacetyloleandrin, m/z 455 for oleandrigenin and m/z 397 for digitoxigenin used as IS; for
gitoxigenin, both [M + Na]
+
and [M + Na – H
2
O]
+
ions appeared at m/z 413 and 395, respec-
tively.

⊡ Figure 7.2
TIC and mass spectra for each peak obtained by LC/MS for the authentic compounds (1 µg/mL
each) of oleandrin and its related compounds.
Oleander toxins
522 Oleander toxins
> Figure 7.3 shows a TIC and mass chromatograms (MCs) for the above 5 compounds,
which had been spiked into blood (0.1 µg/mL) and extracted from it.  ere were no interfering
impurity peaks for each test compound in the chromatogram of blank blood.
 e recovery rates for oleandrin, desacetyloleandrin and oleandrigenin were not lower
than 70 %; but that for gitoxigenin was as low as about 20 %.  ere was good linearity in the
TIC and MCs obtained by LC/MS for an extract of blood, into which oleandrin and its related
compounds had been spiked (0.1 µg/mL each).
⊡ Figure 7.3
523
range of 5–100 ng/mL for oleandrin, desacetyloleandrin and oleandrigenin.  e detection
limits from blood were 3 ng/mL for oleandrin, 2 ng/mL for desacetyloleandrin and oleandri-
genin, and 30 ng/mL for gitoxigenin.
Poisoning case, and toxic and fatal concentrations
A 49-year-old female boiled an oleander branch with leaves in water, and took a large amount
of the extract solution; she underwent therapy, but died one day later.  e blood and CSF
specimens obtained at the postmortem inspection of the above victim were analyzed by the
present method.  e TIC and MCs obtained for the victim by LC/MS are shown in
> Fig-
ure 7.4. Oleandrin could be detected; but desacetyloleandrin, oleandrigenin and gitoxigenin
could not.  e peak at m/z 599 observed in the MC was con rmed to be due to oleandrin by
MS/MS analysis as shown in
> Figure 7.5.  e concentration of oleandrin was 10 ng/mL for
both blood and CSF.
 e blood or plasma concentrations in cases of poisoning by oleandrin, digoxin and digi-
toxin are shown in

> Table 7.1.  ere is another report dealing with LC/MS detection of
oleandrin in an oleander poisoning case [6] except our case; there are also 2 reports dealing
with the immunoassay detection of oleandrin using its cross-reaction [1, 4]; the immunoassay
kit had been developed for measurements of digoxin, and thus the values of oleandrin in blood
were expressed as the concentrations of digoxin (5.8 and 4.2 ng/mL). However, the digoxin
immunoassay method does not give quantitative results for oleandrin; it seems useful only for
tentative qualitative analysis, but is not reliable for its quantitation.
⊡ Table 7.1
Concentrations of oleandrin, digoxin and digitoxin in blood or plasma of cardiac glycoside
poisoning cases
Compound Concentration
(ng/mL)
Specimen Outcome Ref.
oleandrin 1.1
10
blood
blood
survived
dead
[6]
the present case
digoxin 7–24
22
30
plasma
blood
blood
dead
dead
dead

[10]
[10]
[10]
digitoxin 260
320
plasma
plasma
survived
dead
[10]
[10]
Oleander toxins
524 Oleander toxins
TIC and MCs obtained by LC/MS for an extract of blood in a case of oleander poisoning.
⊡ Figure 7.4
525
Authentic oleandrin
blood extract
MS/MS mass spectra of an extract of blood in a case of oleander poisoning and of the authentic
oleandrin. Product ions were obtained from peaks detected by mass chromatography at m/z 599.
⊡ Figure 7.5
Notes
a) Desacetyloleandrin was synthesized by deacetylation in anhydrous methanol with sodium
methoxide as catalyst.
b) Sonic spray ionization ( SSI) is relatively similar to atmospheric pressure chemical ionization
( APCI).  e mobile phase is electrically neutral; but in a small region, especially around
the surface layer of the solution, charge separation can occur. In SSI, nebulization is done
so that the surface layer of the solution, in the region of charge separation, is stripped by
fast nitrogen gas  ow and electrically charged airborne droplets are created.  e diameters
of these electrically charged droplets shrinks by vaporization of solvent molecules from the

surface, and protonated molecular ions are formed in the gas phase.  e interface does not
require heating upon nebulizing; thus it is suitable for sensitive analysis of thermolabile
compounds.
c)  e MS/MS conditions with 3-dimensional QMS for oleandrin are described here; the con-
ditions are highly dependent on a compound to be analyzed. It is essential to optimize
conditions for each compound.
d) As specimens, blood, plasma and urine can be used.
e) Such washing to remove compounds of low polarity is useful, especially when repeated
analyses are required.
References
1) Osterloh J, Herold S, Pond S (1982) Oleander interference in the digoxin radioimmunoassay in a fatal ingestion.
JAMA 247:1596–1597
2) Blum LM, Rieders F (1987) Oleandrin distribution in a fatality from rectal and oral Nerium oleander extract
administration. J Anal Toxicol 11:219–221
3) Rule G, McLaughlin LG, Henion J (1993) A quest for oleandrin in decayed human tissue. Anal Chem 65:857–863
4) Nakata M, Miyata S, Endo K et al. (1995) A fatal case of oleander poisoning. J Okinawa Med Assoc 34:97 (in
Japanese)
Oleander toxins
526 Oleander toxins
5) Gupta A, Joshi P, Jortani SA et al. (1997) A case of nondigitalis cardiac glycoside toxicity. Ther Drug Monit
19:711–714
6) Tracqui A, Kintz P, Branche F et al. (1998) Confirmation of oleander poisoning by HPLC/ MS. Int J Legal Med
111:32–34
7) Tor ER, Holstege DM, Galey FD (1996) Determination of oleander glycosides in biological matrices by high-per-
formance liquid chromatography. J Agric Food Chem 44:2716–2719
8) Namera A, Yashiki M, Okada K et al. (1997) Rapid quantitative analysis of oleandrin in human blood by high-
performance liquid chromatography. Jpn J Legal Med 51:315–318
9) Takaesu H, Fuke C, Arao T et al. (1998) A study on the methods of verifying oleander poisoning through the
analysis of biological materials. Jpn J Forensic Toxicol 16:136–137 (in Japanese with an English abstract)
10) Moffat AC, Jackson JV, Moss MS et al. (eds) (1986) Clarke’s Isolation and Identification of Drugs, 2nd edn. The

pharmaceutical Press, London, pp 541–544