Journal of Analytical Toxicology, Vol.
30, September2006
Testingfor Atropine and Scopolaminein Hair by
LC-MS-MS after Datura inoxia Abuse
Pascal Kintz 1,*, Marion Villain 1, Yann BarguiP, Jean-Yves Chariot 2, and Vincent Cirimele 1
I Laboratoire ChemTox, 3 rue Gruninger, 67400 IIIkirch, France and 2Centre Hospitalier, Noum6a, Nouvelle Cal~donie, France
Abstract ]
Datura inoxia belongs to the family of Solanaceae. This is a
very common plant in New Caledonia that contains two
main toxic alkaloids, L-atropine and L-scopolamine. In this
study, we report the case of a 20-year-old male admitted
to an Emergency Unit after consumption of 6 dried flowers
in hot water for hallucinations, mydriasis, and agitation associated
with tachycardia and increase of systolic blood pressure to
180. Full recovery was observed after one week. Three weeks
later, a lock of about 80 hairs (200 mg) was collected from
the subject in vertex posterior with scissors to be tested for
both atropine and scopolamine. After decontamination
with dichloromethane, a strand of hair was segmented into
three parts, cut into small segments (< 1 ram), incubated
overnight in 1 mLpH 8.4 phosphate buffer in the presence of
2.5 ng atropine-d3, the internal standard, then extracted with 5 mL
dichloromethane/isopropanol/n-heptane(50:17:33). The residue
was reconstituted in 100 pL of methanol, from which 10 pL was
injected into an XTerra MS C18 column (100 • 2.1 ram, 3.5 pro)
eluted with a gradient of acetonitrile and formate buffer delivered
at a flow rate of 0.2 mL/min. A Quattro Micro triple-quadrupole
mass spectrometer (MS) was used for analyses. Ionization was
achieved using electrospray in the positive ionization mode. For
each compound, detection was related to two daughter ions
(atropine: m/z 290.2 to 124.0 and 92.9; atropine-d3: m/z 293.1 to
127.0 and 92.9; scopolamine: m/z 304.1 to 138.0 and 156.0).
Although atropine was never detected (limit of detection = 2
pg/mg), scopolamine was identified in the three segments, in the
range 14 to 48 pg/mg. The absence of atropine in hair is consistent
with its very low dosage in the flower of Datura inoxia. Hair
segmentation indicated that the subject was previously exposed on
several occasions to the plant. Liquid chromatography-tandem MS
appears to be a necessity for testing tropane alkaloids of the
Datura group, given the low concentrations to be measured.
Introduction
The Datura plants are members of the botanical family
Solanaceae, which also contains such common foods as toma-
9 Author to whom correspondence should be addressed. E-mail: pascal.kintz@wanadoo.fr.
454 Reproduction(photocopying)of editorialcontentof thisjournal is prohibitedwithoutpublisher'spermission.
toes, potatoes, eggplants, peppers, and tobacco. The daturas
are herbaceous plants with fragrant, trumpet-shaped flowers.
They grow most often in disturbed soil and waste areas such
as abandoned fields, ditches, trash heaps, and roadsides
throughout the Americas, Eurasia, and Africa.
There is archaeological evidence in the form of botanical re-
mains and petroglyphs that Datura has been used in the Amer-
ican southwest since at least 4000 years ago, often being
associated with other hallucinogenic plants, including peyote.
Datura has been also used by sorceresses to cause illness or
death or to cast a spell of love through manipulation of the
spirit world.
Though the exact taxonomy of Datura is poorly understood;
there are at least 15 different species, including Datura inoxia.
It is an upright annual (60 to 120 cm tall) plant from perennial
rootstock. The entire plant is covered with small, spreading
hairs. The dark green or purplish leaves are 20 to 30 cm long
and have an asymmetric base. The large white flowers are 20
to 30 cm long and bloom at dusk. It has a bushy growth habit
with up to 200 kidney-shaped seeds borne in pods with closely
spaced thorns.
The chemistry of Datura is primarily composed of active
tropane alkaloids including L-scopolamine (L-hyoscine), L-at-
ropine (L-hyoscyamine), tropine, meteloidine, and over 20
others (1). Tropane alkaloids are muscarinic antagonists that
block neurotransmission across muscarinic cholinergic re-
ceptors. Characteristics of muscarinic antagonism, depending
on dosage, include dry mouth, flushing or rashes, hyperten-
sion, tachycardia, bronchodilatation, blurred vision, dizziness
or vertigo, sedation, and amnesia (2). Atropine and scopo-
lamine are physiologically active at doses of 0.5-1 mg. Alka-
loids content is known to vary significantly among species,
within a species depending on season or time of day, and even
within a particular plant. It appears that the immature plants
mainly contain scopolamine. Often, roots, stems, leaves,
flowers, and seeds have differing suites of alkaloids in differing
concentrations. For this reason, preparation for recreational
versus divinatory use is often accomplished using different
species or different parts of the plant. Traditional preparations
include adding roots, leaves, or seeds to a fermented drink;
drinking an infusion of the leaves or other parts; smoking the
Journalof AnalyticalToxicology,Vol. 30, September2006

leaves; or chewing the fruit. When the plant material is taken chloride (immersion in 5 mL for 2 min) and then segmented
orally, the effects last longer than when smoked and will also (3 segments of 3 cm). Each segment was cut into small pieces
be more narcotic and hallucinogenic. (< 1 ram). About 50 mg was incubated overnight in 1 mL of
The toxicity of Datura species is well known and has been phosphate buffer at pH 8.4, in the presence of 2.5 ng of at-
linked to poisonings and deaths, usually due to heart or respi- ropine-da used as internal standard (IS). After a liquid-liquid
ratory failure, for centuries. Datura intoxication can last from extraction with 5 mL of a mixture Ofmethylene chloride/iso-
a few hours to many days, depending on dosage and method of propanol/n-heptane (50:17:33, v/v/v) and evaporation to dry-
ingestion. Exposure manifests as a classic anticholinergic syn- ness, the residue was reconstituted in 100 l~Lof methanol.
drome comprising central and peripheral signs and symp-
toms. Central toxic effects include confusion, agitation, LC-MS-MS procedure
anxiety, hallucinations, seizures, and coma. Peripheral toxic ef- LC was performed using a Waters Alliance 2695 system.
fects include dry mucous membranes, thirst, flushed face, hy- Chromatography was achieved using an XTerra MS C18
perthermia, urinary retention, and decreased gut motility (3). column (100 • 2.1 mm, 3.5 IJm) eluted with a gradient deliv-
Datum poisoning can be observed in various situations, in- ered at a flow rate of 0.2 mL/min (TableI). An injection volume
cluding criminal activities (4), accidental cases (5-8), inten- of 10 l~L was used in all cases. A Quattro Micro triple-
tional suicide (9), and recreational abuse (10,11). Because quadrupole MS (Micromass-Waters) fitted with a Z-Spray ion
daturas, Datura stramonium and Datura inoxia in particular, interface was used for analyses. Ionization was achieved using
are used as ornamental plants, gardening practices in a com- electrospray in the positive ionization mode (ES+).
munity might provide novel opportunities for experimenting The following conditions were found to be optimal for the
with intoxicating substances. analysis of both atropine and scopolamine and the IS: capillary
We present here an original method to test for both at- voltage, 1.0 kV; source block temperature, 120~ and desol-
ropine and scopolamine in hair by liquid chromatography- vatation gas (nitrogen) heated to 350~ and delivered at a flow
tandem mass spectrometry (LC-MS-MS)and its application to rate of 550 L/h. In order to establish appropriate multiple re-
a Datura inoxia poisoning. action monitoring conditions, the cone voltage was adjusted to
maximize the intensity of the protonated molecular ion and
collision-induced dissociation of both species was performed.
Collision gas (argon) pressure was maintained at 3.0 mBarr,
Materials and Methods and the collision energy (eV) was adjusted to optimize the
signal for the two most abundant product ions (Table II). Op-
Case report timization for the atropine and scopolamine products and
The subject, a 20-year-old male and known cannabis abuser,
prepared an infusion of hot water with six dried Datura inoxia
Table i. HPLC Gradient Profiles for the Analysis of
flowers. Ten minutes after ingestion, he lapsed into a coma. Atropine and Scopolamine
Upon waking, he experienced hallucinations, tachycardia, and
an increase of systolic blood pressure up to 180. Once ad- Time (min) Acetonitrile(%) Formate Buffer (%)
mitted to the Emergency Unit, the medical staff noticed agi-
tation, mydriasis, and reduced salivary secretions. Treatment 0 5 95
included gastric decontamination and IVperfusion of 50 mg of 3 60 40
chlorazepate di-potassium. He was fully recovered one week 7 80 20
later, after long-term agitation and mydriasis. 10 80 20
Neither blood nor urine was collected during hospitaliza- 10.5 5 95
tion, but the psychiatrist sampled a strand of hair three weeks 20 5 95
later.
Drug-free hair samples were obtained from laboratory per-
sonnel. Table II. MRM Transitions and Conditions for the
Measurement of Atropine and Scopolamineand the
Chemicals and reagents Internal Standard
Acetonitrile, methanol, isopropanol, n-heptane, and methy-
lene chloride were high-performance liquid chromatography Parent Product Cone Collision
(HPLC) grade (Merck, Darmstadt, Germany).Chemicalsfor the Compound Ion (m/z) Ion (m/z) Voltage(V) Energy(eV)
saturated phosphate buffer, (NH4)2I-IP04,adjusted to pH 8.4
Atropine 290.2 124.0 60 25
with orthophosphoric acid, were purchased from Fluka (Saint-
92.9 60 30
Quentin Fallavier, France). Scopolamine, atropine, and
atropine-d3 were purchased from Sigma (Saint-Quentin Scopolamine 304.1 138.0 50 25
Fallavier, France). 156.0 50 18
Atropine-d3 293.1 127.0 60 30
Extraction 92.9 60 25
Hair strands were twice decontaminated using methylene

455
 Journal of Analytical Toxicology, Vol. 30, September 2006

daughter ions was achieved by infusion through the MS of in-
dividual solutions at 10 mg/L in methanol. MassLynx4.0 soft-
ware was used for quantitation.
Method validation
A standard calibration curve was prepared in hair fortified
with the drug and obtained by preparing spiked standards con-
taining 2, 5, 10, 20, 50, and 100 pg/mg of atropine and scopo-
lamine. Within-batch precision (n = 6) was determined using
blank hair spiked with atropine and scopolamine at 50 pg/mg.
The limit of detection (LOD)was evaluated by decreasing con-
centrations of atropine and scopolamine until a response
equivalent to three times the background noise was observed.
Relative extraction recovery (n = 3) was determined by com-
paring the representative peak area of atropine and scopo-
lamine extracted from negative control hair spiked at the final
concentration of 50 pg/mg with the peak area of a methanolic
standard at the same concentration. The specificity of the
method was evaluated by analyzing hair from six non-drug
consuming subjects (from the laboratory). The matrix effect
Cat[l: O,Gng_Je 36rag_El 2.Gng
0503~8-01
Time (min)
0SO31&01
1~^
o terials present in hair. Matrix effects (ion sup-
........ 2 ~() . . . . . . . ,i.~'J . . . . . . .
Time (rain)
(n = 3) on the ES+ response was evaluated by comparing the
instrumental response for a calibrator at 50 pg/mg directly in-
jected in methanol (a) and the same amount of compound
added to pre-extracted samples (b). The data for the calibrator
in methanol provide a relative 100% value, and the matrix ef-
fect was defined as 100 x (a - b)/a.
Results and Discussion
Validation
Linearity was observed for atropine and scopolamine con-
centrations ranging from 2 to 100 pg/mg with a correlation co-
efficient of 0.992 and 0.998, respectively. Within-batch
precision (n -- 6) at 50 pg/mg was 8 and 10% for atropine and
scopolamine, respectively.Bias was 11 and 6% for atropine and
scopolamine, respectively. Relativeextraction recovery was of
82 and 67% for atropine and scopolamine, respectively. As
total digestion in strong base was not possible because of in-
stability of the drugs, it was not possible to
evaluate whether all drug content was ex-
0AA861 18-Mar-2OOG
7 59
tracted from the hair. The LOD was 2 pg/mg MRMof 6 ChannelsES~
293.08> 12704
(S/N 3) with a limit of quantitation (LOQ)at 5
pg/mg (S/N 10) for both compounds. Under
the chromatographic conditions used, there
was no interference with the analytes by M~ld of GChannels~S*
?.6g ~0 22 9 124 01
chemicals or any extractable endogenous ma- 1 ~e3
,yea
6 ~() . . . . . . .
pression) were 27.5 and 50% for atropine and
g.~ . . . . . . . 10'(~ . . . . . . . 12~ ....... X~,IO~. . . . . . . 16~(~ . . . . . . . 18'00 . . . . . . . ~]:(~
scopolamine, respectively.These are the max-

:]
C503~8-G1 MRMof GChannelsES,
l~n_ 760 imum percentages of suppression of both at- 290 22 ~ 9289
O64
ropine and scopolamine peaks. Because Area

potential unexpected ion suppression can

.... ' "'" ~.~ " " " "" ";~ ~

Time (rain)
" " '" "" ~ ~ ""
occur, caution has to be taken to interpret
" ' " "" ~ ~ " ' "' "" '~Wo" "" ' ' " "~I~ " " "" h~lo6 " "'"" "i~ '~ " ' " "" "~ '~o ' " "' ' " "~oloo

C533t8.r the generated quantitative result. However, ~RM of 6 C~a~ne[~ES,

~3 3O4O; > 1::8C3

kea
within the same strand of hair, it is acceptable
o to consider that the same matrix effect will
apply; therefore, putting a quantitative inter-
Time (min)
pretation on different segments from the
050318-01 MRMof 6 Cttanttcls ES*
663 same strand of hair is possible. 3040?>13802

Figure 1 is the chromatogram obtained

after extraction of a drug-free hair spikedwith
Time (nfin) both atropine and scopolamine at a final con-
Figure1. Chromatograrn of drug-free hair spiked with atropine and scopolamine at a final con- centration of 10 pg/mg.
centration of 10 pg/mg. From the top to the bottom: the product ion of atropine-d3, the two
product ions of atropine, and the two product ions of scopolamine. Case report
A strand of hair was collected 3 weeks
after the ingestion of an infusion of D a t u r a
Table III. Results of the Hair Segmentation inoxia flowers.The results are presented in Table III. Figure 2
is the chromatogram obtained after extraction of the 0 (root)
Segment Atropine Scopolamine to 3 cm segment. No atropine was detected, and the concen-
trations of scopolamine were 14, 48, and 43 pg/mg along the
0 (root)to 3 cm < LOD* 14 pg/mg three segments that were tested. Absence of deviation in re-
3 to 6 cm < LOD 48 pg/mg tention times versus the calibrator and simultaneous pres-
6 to 9 cm < LOD 43 pg/mg ence of the two transitions in the correct range (_+20%) were
* LOD = 2 p~mg. the forensic criteria that were used to qualitatively determine
the presence of scopolamine in the subject's hair sample. These

456
Journal of Analytical Toxicology, Vol. 30, September 2006

ROM-CHA O~cm_54mg B2.Sng QAA861 184gar-2006

complement conventional blood and urine
65031844
132
MRMd 6 ~an~e~ ES*
29308 912704 analysis as it increases the window of detection
and permits differentiation, by segmentation,
1 ..... I i i

iT ....... iN ....... gN .......

A[ .
iN''"
. .
" " "1"o~ . . . . . . .
Time (rain)
. .
I?~N" ' "'" "?4N
.
....... 76~ ....... I?~N . . . . . . . N.N
" of long-term use from a single exposure. Se-
lectivity and sensitivity of MS-MS are requisite
05O318-04 MRMo( 6 ~anne~ ES~
for testing alkaloids of the Datura group be-
2902"2>12401
cause of the low concentrations to be mea-
Area
sured.
.... ' " " "~.t~"" "'" " " ~ ' " " ' "" " ~ t ~ ' " " ' " "" ~ " . . . . "' '?o~' ' "' ' ' '1'2!~"" "'" ""74~" "" "" " 7 6 ~ " " " " " h~'o~"" " ' "" ":~'~
Time (rain)

050318-04 MRMo( 6 Qlanne~sES+

16 29022>92.69
473 References
88 &'ea

.... ' . . . . .... " " "" .... ' . . . . . . . . ' . . . . . . . . ' .... ~ " " " " h~o~" " "
1. E. Miraldi, A. Masti, S. Ferri, and B. Comparini.
" "? ~" " " ' " h ' r " " " "&'~" " " " "~o~
Time (mi.) Distribution of hyoscyamine and scopolamine
050318-04 MF~Md (t ChanneL~ES~
in Datura stramonium. Fitoterapia 72" 644-648
623 304 07 > 16603 (2001).
2. J.H. Brown and P. Taylor. Muscarinic receptor
agonists and antagonists. In Goodman and
.... ' .... . .... '"" iN ' "" "'" &g'" "" "" g~''" "" "1"~I~'"" '"" "~}~ ....... I ; ~ N " " " " " " ~ ' 6 ' ~ " " " " " "1"~"" "'"" "~1~6 Gilman's The Pharmacological Basis of Thera-
'rime (rain)
peutics, J.G. Hardman and L.E. Limbird, Eds.
050318-04
623
MRMof 6 Ct~anne~sES+
304 07 9 138 02
McGraw-Hill, New York, 1990, pp 141-160.
,~r 3. M. Pekdemir, S. Yanturali, S. Akay, and
161,...,
i.N """ "" i.N ""' " s b ~ " " i~6"
_X
" hbi~ '" "~}N """ h;@~" ""'" h'~'~" "'"" "783~'" "'"" "~I~ ~
G. Alagoz. Acute anticholinergic syndrome
due to Datura innoxia Miller mixed with lime .....
Ti~e (rain) tea leaves. Vet. Hum. Toxicol. 46:176-177
Figure2. Chromatogramof a hair extractfromthe firstsegment(rootto 3 cm) fromthe subject. (2004).
4. G. P@in, M. Ch~ze, G. Duffort, and F. Vays-
Scopolamineconcentrationwas 14 pg/mg. Fromthe top to the bottom:the product ion of at-
sette. Interest of hair and tandem mass spec-
ropine-d3,the two product ions of atropine, and the two productionsof scopolamine. trometry for chemical submission: about 9
cases. Ann. Toxicol. Anal. 14:395-406 (2002).
concentrations cannot be compared with previous data, as the 5. S. Chang, M. Wu, J. Deng, C. Lee, T. Chin, and
S. Liao. Poisoning by Datura leaves used as edible wild vegeta-
international literature is lacking reports on this topic. The
bles. Vet. Hum. Toxicol. 41 9242-245 (1999).
presence of scopolamine in the three consecutive segments is 6. S.V. Raman and J. Jacob. Mydriasis due to Datura inoxia. Emerg.
inconsistent with a single exposure to Datura. This is also Med. J. 22" 310-311 (2005).
supported by the fact that three different concentrations were 7. J.P. Hanna, J.W. Schmidley, and W.E. Braselton, Jr. Datura
measured along the hair shaft, in contrast to what is obtained delirium. Clin. NeuropharmacoL 15:109-113 (1992).
P.A. Steenkamp, N.M. Harding, F.R. van Heerden, and B.E. van
when external contamination occurs. 8. Wyk. Fatal Datura poisoning: identification of atropine and scopo-
The absence of atropine in hair is consistent with its very low lamine by high performance liquid chromatography/photodiode
dosage in the flower ofDatura inoxia (0.002 pg/g versus 0.346 array/mass spectrometry. Forensic Sci. Int. 145:31-39 (2004).
pg/g for scopolamine) (12). 9. R.W. Urich, D.L. Bowerman, J.A. Lavisky, and J.L. Pflug. Datura
stramonium: a fatal poisoning. J. Forensic Sci. 27" 948-954
(1982).
lO. P. Salen, R. Shih, P. Sierzenski, and J. Reed. Effects of physostig-
mine and gastric lavage in a Datura stramonium-induced anti-
Conclusions cholinergic poisoning epidemic. Am. J. Emerg. Meal. 21 ' 316-317
(2003).
The daturas are a group of plants that have an intimate as- 11. S. Cohen, C. Berny, S. Meyran, A. Mialon, and M. Manchon. In-
sociation with man from time immemorial. They have been toxication volontaire par une tisane de feuilles de Datura. Ann.
Toxicol. Anal. 15:287-291 (2003).
used as poisons, medicines, and ritual intoxicants. However, this 12. J.P. Anger, M. Villain, A. Baert, and P. Kintz. Le datura: une plante
old plant is more and more used for recreational experience. oubli~e de la Pharmacop~e mais qui semble de plus en plus
In case of recreational abuse, hair testing should be used to pl~biscit~e par les jeunes. Ann. Toxicol. Anal. 16:166 (2004).

457