Forensic Science International 189 (2009) e13–e20
Contents lists available at ScienceDirect
Forensic Science International journal homepage: www.elsevier.com/locate/forsciint
Case report
Suicidal death after injection of a castor bean extract (Ricinus communis L.) Vera Coopman a, Marc De Leeuw b,c, Jan Cordonnier a,*, Werner Jacobs c a
Department of Analytical Toxicology, Chemiphar N.V., Lieven Bauwensstraat 4, B-8200 Brugge, Belgium Emergency Department, Algemeen Stedelijk Ziekenhuis, Merestraat 80, B-9300 Aalst, Belgium c Centre for Forensic Medicine, Antwerp University Hospital, Wilrijkstraat 10, B-2650 Edegem, Belgium b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 20 March 2008 Received in revised form 19 March 2009 Accepted 1 April 2009 Available online 23 May 2009
The castor bean plant (Ricinus communis L.) or wonder tree is cultivated in many countries as an ornamental annual plant in gardens. The highest concentration of the lectin ricin is present in the seeds and pods. Ricin is considered as one of the most toxic natural poisons. Ricinine is a piperidine alkaloidal toxin present in castor bean and is described as a biomarker for the exposure to ricin. A case report is presented of a 49-year-old man who committed suicide by intravenous and subcutaneous injection of a castor bean extract. He was brought to the emergency department 24 h after injecting himself. On admission, the patient was conscious and he presented with a history of nausea, vomiting, diarrhoea, dyspnoea, vertigo and muscular pain. Despite symptomatic and supportive intensive care, the man died 9 h after admission due to multiorgan failure. A body external examination was performed. Blood, urine, vitreous humour and the castor bean extract were submitted to the laboratory for toxicological analysis. The identification of ricinine in the extract was performed by solid phase extraction in combination with full-scan gas chromatography/ mass spectrometry, high-performance liquid chromatography with photodiode array detection and liquid chromatography/mass spectrometry operated in the full-scan mode, respectively. An extraction procedure with Oasis HLB solid phase extraction cartridges was applied. Chromatography was achieved using a Symmetry1 C18 column using a gradient mode with 0.15% formic acid and 0.15% formic acid in acetonitrile as mobile phase. Exposure to the castor bean extract was confirmed by identification of the biomarker ricinine in blood, urine and vitreous humour using solid phase extraction and liquid chromatography tandem mass spectrometry with electro spray source in positive ionization mode. Multiple reaction monitoring was used for specific detection. To the authors’ knowledge, it is the first time that ricinine has been identified in vitreous humour in a case of castor bean poisoning. Based on the clinical symptoms and the results of the toxicological analysis, we concluded that death was caused by intoxication with plant toxins originated from R. communis L. ß 2009 Elsevier Ireland Ltd. All rights reserved.
Keywords: Ricinus communis L. Ricinine Lethal intoxication Injection LC–MS/MS
1. Introduction The castor bean plant (Ricinus communis L.) or wonder tree belongs to the family of Euphorbiaceae and is cultivated in many countries, including Belgium, as an ornamental annual plant in gardens. The plant is native to tropical Africa. The plant is a shrub, 1–4 m high, branched with green or reddish leaves. It has clusters of seed pods covered with fleshly spines. The pods contain three seeds per capsule. The seeds have a smooth and glossy coat and are usually mottled with black, brown, grey and white spots (Fig. 1). All parts of the plant are poisonous [1–3]. The highest concentration of the lectin ricin is present in the seeds and pods. Ricin is considered
* Corresponding author. Tel.: +32 50 31 02 52; fax: +32 50 31 02 54. E-mail address:
[email protected] (J. Cordonnier). 0379-0738/$ – see front matter ß 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.forsciint.2009.04.019
as one of the most toxic natural poisons. It is a glycoprotein composed of two polypeptide chains, the A-chain (30 kDa) and Bchain (32 kDa), linked with a disulfide bond and with a molecular weight of about 63,000. Ricin inhibits protein synthesis by inactivating the ribosomes [1–4]. Reported ricin content in the castor bean varies between 1% and 5%. Castor oil is industrially produced from castor beans and is used in lubricating oils, paints, varnishes and is orally administrated as a purgative in medicine. After isolation of the oil, ricin remains in the bean pulp but is inactivated if done under heated conditions. The castor oil is not considered to contain ricin [3,5]. The monographs of castor oil in the European Pharmacopoeia do not contain a test for the detection of the plant toxins [5]. Since it is available as a by-product of castor oil production, is present in large quantities in nature, and is cheap and easy to extract, ricin is a potential biological warfare agent [1,3]. Exposure can occur by ingestion, injection or inhalation. The
e14
V. Coopman et al. / Forensic Science International 189 (2009) e13–e20
Fig. 1. Castor bean plant (Ricinus communis L.) and castor beans.
toxicity of ricin varies with route of administration. The most potent routes of introducing ricin are inhalation and injection. When ingested, the castor bean is non-toxic as long as the hard, water-impermeable seed coat remains intact. The toxicity depends on the level of mastication or maceration of the seeds [1,3]. Initial clinical findings of ricin poisoning by injection can include nonspecific signs and symptoms such as generalized weakness and myalgias. After 24–36 h the clinical symptoms progress with possible vomiting, fever, dehydration, hypotension, and/or multiorgan failure and death [3]. At present, no antidote or effective therapy is available for the treatment of ricin intoxication. Only symptomatic and supportive measures can be taken [1,3]. Ricinine (1,2-dihydro-4-methoxy-1-methyl-2-oxo-3-pyridinecarbonitrile) (Fig. 2) is a piperidine alkaloid toxin with low molecular weight (164.2) and a molecular formula of C8H8N2O2 [6]. Ricinine contributes to the toxicity of castor bean and is present in small amounts in the castor bean and leaves. Reported clinical manifestations of ricinine ingestion are nausea, vomiting, hemorrhagic gastroenteritis, hepatic and renal damage, convulsions, coma, hypotension, respiratory depression and death [7,8]. No reports of intoxication cases by ricinine only were found in literature. Ricinine is co-extracted with ricin and is described as a biomarker for ricin exposure [2,9–11]. Analytical methods based on paper chromatography and UV detection, electron impact mass spectrometry and liquid chromatography-(photodiode array detection)–mass spectrometry were described for the determination of ricinine in plant material [2,11–13]. Case reports of intoxications by injection of a castor bean extract are rare [8,14– 18]. The most cited case is the assassination in 1978 of the exiled
Fig. 2. Chemical structure of ricinine.
Bulgarian journalist Georgi Markov who was most likely injected with ricin present in a platinum pellet injected into him when he was prodded with an umbrella [2,9,11]. A solid phase extraction and liquid chromatography–mass spectrometry method for the quantification of ricinine in urine was published by Johnson et al. and was applied in a forensic case of a man who committed suicide by ricin injection [9]. A case report is presented of a 49-year-old man who committed suicide by intravenous and subcutaneous injection of a castor bean extract. Ricinine was identified in the extract using full-scan gas chromatography/mass spectrometry (GC/MS), high-performance liquid chromatography with photodiode array detection (HPLC/DAD) and liquid chromatography/mass spectrometry (LC–MS) operated in the full-scan mode, respectively. Solid phase extraction (SPE) and liquid chromatography–tandem mass spectrometry with electro spray source in positive ionization mode (LC–MS/MS) were applied for the determination of ricinine in blood, urine and vitreous humour. 2. Case history A 49-year-old man (83 kg, 175 cm tall) was brought to the emergency department 24 h after injecting himself intravenously and subcutaneously with approximately 10 mL of a ‘self-made’ acetone extract of castor beans. He presented with a history of nausea, vomiting, diarrhoea, dyspnoea, vertigo and muscular pain. On admission, the patient was conscious with a Glasgow Coma Score 14. He stated that he intended to commit suicide by ricin poisoning. Clinical findings were consistent with hypovolemic shock and severe dehydration and fluid depletion. Blood pressure was 70/50 mmHg, pulse rate 120 min 1 and respiratory rate 30 min 1. His temperature was 37.1 8C. Arterial blood gas analysis showed respiratory compensated metabolic acidosis (pH 7.36 – pCO2 25.1 mmHg – pO2 85 mmHg – bicarbonate 14 mequiv./L). Initial laboratory test results were as follows: blood urea nitrogen 29.7 mg/dL and creatinine 2.56 mg/dL, glucose 257 mg/dL, LDH 2959 U/L, SGOT 229 U/L, SGPT 55 U/L. D-dimer level was 9930 ng/mL and APTT 66.6 s. Initial treatment consisted of infusion of saline, 4.5% human plasma protein solution and Fresh Frozen Plasma and intravenous administration of bicarbonate, ranitidine,
V. Coopman et al. / Forensic Science International 189 (2009) e13–e20
alizapride and phytomenadione. Oxygen was administered with a non-rebreathing mask at a flow rate of 10 L/min. He was admitted to the intensive care unit, where his respiratory and hemodynamic function were closely monitored. One and a half hour after admission, a control arterial blood gas analysis showed no improvement of the metabolic acidosis. Laboratory tests confirmed liver failure (LDH 4875 U/L), renal dysfunction (creatinine 2.93 mg/dL) and haemolysis. Despite intubation, mechanical ventilation, fluid administration and inotropic support, the patient died 9 h after admission due to multiorgan failure. Only a body external examination was performed 32 h after death. A puncture mark with a small scar was present in the right antecubital fossa. On the left side of the umbilicus, a dark purple ecchymotic zone (2 cm) with a central puncture mark was observed. The following samples were taken and transferred to the laboratory for toxicological analysis: blood (EDTA tubes), urine and vitreous humour. A drip with empty syringe (20 mL) and a bottle with the ‘self-made’ extract were submitted to the laboratory for analysis. When the bottle was opened a scent of acetone emanated. The content consisted of liquid and a sediment of what appeared to be macerated seeds. A few pieces of coating and two peeled seeds were present. On arrival, the biological samples were analyzed according to the standard systematic toxicological analysis procedure of the laboratory. The presence of plant toxins from castor bean in the content from the bottle was confirmed through the detection of the biomarker ricinine using full-scan GC/MS, HPLC/DAD and LC–MS operated in the full-scan mode, respectively. Seeds of R. communis L. were analyzed. SPE and LC–MS/MS were applied for the determination of ricinine in blood, urine and vitreous humour. 3. Materials and methods 3.1. Systematic toxicological analysis A comprehensive systematic toxicological analysis (STA) was performed on the post-mortem samples to investigate for illegal drugs, medical drugs, alcohol and volatile substances. Screening for the presence of drugs of abuse in urine and medication in serum was carried out using fluorescence polarization immunoassay on the Abbott AXSYMTM analyzer. Enzyme linked immunosorbent assay was used to screen for LSD in urine and opiates in serum. Presumptive colour tests on urine were used to detect salicylates, acetaminophen, phenothiazines and imipramines. Screening for the presence of basic drugs after liquid–liquid extraction was performed by HPLC/DAD in blood. Not enough urine was available for the screening on the presence of basic drugs by liquid–liquid extraction and GC/MS. Quantification of alcohol and acetone in the samples and identification of other volatile substances in blood and liquid was performed by gas chromatography and static headspace gas chromatography with flame ionization detector, respectively. After addition of deuterated internal standards, blood was extracted at alkaline pH by liquid–liquid extraction and analyzed with liquid chromatography–mass spectrometry (LC–MS/MS) in multiple reaction monitoring mode to identify and quantify benzodiazepines and metabolites, zolpidem, zopiclone, methadone, fentanyl and norfentanyl. 3.2. Chemicals and reagents Ricinine (purity min. 98%) was purchased from Latoxan (Valence, France) and the external standards fentanyl-d5 (100 mg/mL in methanol) was from Promochem (Hertfordshire, England). Standard compound ricinine was dissolved in water (4.6 mg/5.0 mL). The stock solution was stored at 18 8C. Working standards were prepared by diluting the stock solution with methanol (0.092 mg/mL) and were kept at 4 8C until used. Calibrators and controls were preparations of different working solutions from the same ricinine stock solution. The external standard fentanyl-d5 was diluted in methanol (1.0 mg/mL) and refrigerated until use. Acetonitrile and methanol were obtained from Sharlau Chemie S.A. (TMC, Belgium). Formic acid (98–100%) was purchased from Merck-Eurolab (Leuven, Belgium). Water was purified by a Sartorius AG system obtained from VWR (Leuven, Belgium). All solvents and inorganic chemicals were of analytical grade. The phosphate buffer saline (PBS–HCl pH 5.0) was prepared by dissolving the anhydrous salts of sodium chloride 8.00 g, potassium chloride 0.20 g, disodium hydrogen phosphate 1.15 g and potassium dihydrogen phosphate 0.20 g into a 1000 mL volumetric flask and diluting to volume with water. The pH was adjusted to 5.0 with 10% hydrochloric acid, and the solution was stored refrigerated. Oasis HLB solid phase extraction cartridges (60 mg) were obtained from Waters Corporation (Milford, USA). A Vac Elut manifold was used for SPE analysis (Varian, USA).
e15
3.3. Sample preparation and extraction 3.3.1. Seeds of Ricinus communis, extract from the bottle (sediment and liquid) Seeds of R. communis L. were purchased from a local garden centre and stored at room temperature. Extraction of the seeds of R. communis L. and the sediment from the bottle was performed with PBS–HCl pH 5.0 and Oasis HLB SPE columns (60 mg) as previously described by Steenkamp [2] with the following modifications: approximately 1.5 g of peeled seeds were chopped up. The sediment was filtered over a paper filter and was left overnight at room temperature to allow the acetone to evaporate. The pulp from the seeds or sediment (approximately 2 g) was transferred to a polypropylene centrifuge tube with screw cap and was homogenized in 15 mL PBS–HCl pH 5.0 by means of a Ultra Turrax1 (IKA T18 basis). The tube was centrifuged for 10 min at 3000 rpm. The aqueous phase contained a white slurry on the top. To avoid clogging, the aqueous phase was filtered over a paper filter before loading the Oasis HLB SPE column. Ricinine was eluted from the column with 6 mL methanol. The collected eluate was clear and colourless. The eluate was evaporated to dryness using a Zymark (Hopkington, MA) Turbovap LV Evaporator (at 40 8C using nitrogen) and reconstituted in 0.5 mL of methanol. Quantification of ricinine in the extract was performed using the LC–MS/MS method. The liquid was diluted with LC–MS/MS mobile phase (dilution factor 1000, 1500 and 2000). To 1.0 mL of the diluted extract, 25 mL external standard (fentanyl-d5 1 mg/mL) was added. Calibrators (range 2–30 ng/mL) were prepared in LC–MS/MS mobile phase and 25 mg external standard was added. A five point calibration curve was achieved by plotting the peak-area ratios for the product ions of ricinine and the external standard fentanyl-d5 as a function of the analyte concentration. Linear curve fit with no weighting was used. 3.3.2. Blood, urine and vitreous humour To 1.0 mL of blood or urine or an aliquot of the vitreous humour in a glass tube, 1.0 mL of water was added. As the samples were homogenized by vortex mixing, 2.0 mL of acetonitrile was added drop wise. The samples were allowed to stand for 15 min and were centrifuged for 10 min at 3000 rpm. The supernatant was mixed with 5.0 mL PBS–HCl (pH 5.0) and loaded on the SPE cartridge. The Oasis HLB column was conditioned with 2 mL of methanol and 2 mL of PBS– HCL (pH 5.0). The sample was slowly passed through the column. The column was rinsed with 6.0 mL of water and dried under vacuum for 5 min. The retained ricinine was eluted with 6.0 mL of methanol and 5 mL of external standard solution (1 mg/ mL) was added. The eluates were dried under nitrogen (40 8C). The residues were reconstituted in 200 mL of LC–MS/MS mobile phase and centrifuged for 5 min at 14 000 g. 3.4. Liquid chromatography The HPLC/DAD analysis was performed using a Varian Prostar solvent delivery module in combination with a Varian Prostar 410 auto sampler and Varian Prostar photodiode array detector (Varian). Data acquisition and analysis were performed with the Varian Star and Polyview software. The separation of analyte was performed in the isocratic mode with the use of a ChromSpher C18 column (150mm length 4.6 mm i.d., 5 mm particle size, Chrompack) with a C18 guard column (4-mm length 3.0 mm i.d., 3.5 mm particle size). The oven temperature was set at 35 8C. The mobile phase consisted of 90% water and 10% acetonitrile with a flow rate of 1.0 mL/min. The injection volume was 50 mL. The chromatogram was monitored at 309 nm for 10 min. A 50 mL aliquot of reconstituted SPE extract was evaporated to dryness (at 40 8C, nitrogen) and was redissolved in 100 mL of mobile phase. 3.5. Mass spectrometry GC/MS analysis was performed using a Varian Star 3400 gas chromatograph in combination with a Varian 8200 auto sampler and Varian Saturn 2000 GC/MS. A Varian CP-SIL 8 CB low bleed capillary column (30 m 0.25 mm i.d., 0.25-mm film thickness) was used. Carrier gas was helium at a constant pressure of 12 psi. The GC oven was programmed to an initial temperature of 70 8C and hold time of 2 min, followed by a 8 8C/min ramp to a final temperature of 310 8C and hold time of 7 min. The total run time was 39.0 min. The temperatures of the injection port and the detector were set at 300 8C and 230 8C, respectively. The injection volume was 2 mL using the splitless injection mode. The mass spectrometer was operated in the electron impact (EI) mode at 70 eV of electron energy. Mass spectra were recorded in the range m/z 40–500. A 100 mL aliquot of reconstituted SPE extract was evaporated to dryness (at 40 8C, nitrogen) and was redissolved in 50 mL of methanol. 3.6. Liquid chromatography/mass spectrometry The LC–MS/MS analysis was performed using a Alliance1 2695 system and a Quatro MicroTM mass spectrometer (Waters, Milford, USA). Chromatography was achieved using a Symmetry1 C18 column (100-mm length 2.1 mm i.d., 3.5 mm particle size) (Waters) with the oven temperature set at 35 8C. The mobile phases consisted of 0.15% (v/v) formic acid in water (A) and 0.15% (v/v) formic acid in acetonitrile (B). The following gradient elution was used: at time 0 min 90% A held
V. Coopman et al. / Forensic Science International 189 (2009) e13–e20
e16
Table 1 MRM transitions and conditions for the measurement of ricinine and external standard fentanyl-d5. Compound
Parent ion (m/z)
Production (m/z)
Cone volts (V)
Collision energy (eV)
Ricinine
164.80
137.60 83.30
35.00 35.00
16.00 22.00
Fentanyl-d5
341.80
188.00 104.50
35.00 35.00
25.00 43.00
for 5 min, changed to 20% A in 10 min, changed back to 90% A in 2 min, and held for 3 min. The flow rate and injection volume were 0.20 mL/min and 40 mL, respectively. Mass spectrometry detection was carried out by a Quatro MicroTM mass spectrometer (Waters, Milford, USA) with electro spray source in positive ionization mode (ESI+). Full-scan spectra were recorded from m/z 50 to 300, at a scan time of 10 s and an interscan delay of 0.1 s. Other instrument settings were: capillary voltage 3.50 kV, cone voltage 20 V and 45 V, extractor 2 V, ion energy 1.0 V, source temperature 120 8C, desolvatation temperature 350 8C, cone gas (nitrogen) flow rate 500 L/h and desolvatation gas (nitrogen) flow rate 50 L/h. A 50 mL aliquot of reconstituted SPE extract was evaporated to dryness (at 40 8C, nitrogen) and was redissolved in 500 mL of mobile phase. For the quantification of ricinine, product ions were obtained by collisioninduced dissociation, and the MS/MS was operated in the multiple reaction monitoring (MRM) mode. One quantifier and two qualifiers (ion ratio) were monitored for each analyte to provide sufficient identification. The MRM transitions and conditions for the measurement of ricinine and external standard fentanyl-d5 are shown in Table 1. The dwell time was set at 0.50 s. The desolvatation gas (nitrogen) temperature was set at 350 8C and delivered at a flow rate of 500 L/h. The capillary voltage was 3.50 kV. The collision gas (argon) pressure was maintained at 3.0 10 3 mbar. Waters Mass-lynx system software Version 3.5 was used for instrument control and quantification.
3.7. Calibration and quality controls Calibrators and quality controls were prepared by addition of ricinine working standard solutions to ricinine-free matrix prior to SPE extraction. Ricinine-free blood and urine were collected from deceased with known cause of death, other than intoxication. Five point calibration curves were achieved by plotting the peakarea ratios for the product ions of ricinine and the external standard fentanyl-d5 as a function of the analyte concentration. The concentrations of the calibrators ranged from 0.2 ng/mL to 5.0 ng/mL. Linear curve fits with no weighting were used. Specificity was defined as the ability of the assay to distinguish ricinine from other possibly interfering substances in ricinine-free matrices. Six different blank samples of blood and urine were analyzed. To evaluate the possibility of ion suppression, peak areas of ricinine (0.5 ng/mL and 3.0 ng/mL) and external standard fentanyl-d5 spiked after extraction of drug-free blood and urine samples were compared to those obtained for standard solutions at the same concentrations in LC–MS/MS mobile phase. The within-day precision, within-day accuracy and extraction recovery were evaluated by analyzing six quality control samples in a single batch at low (0.5 ng/ mL) and high (3.0 ng/mL) concentration levels in urine. The precision was assessed as the percent coefficient of variation from the concentration. The quantitative accuracy (%) was determined by expression of the mean measured concentration as a percentage of the nominal concentration. The extraction recovery (%) was determined by comparing the peak areas of extracted quality controls with the peak areas of standard solutions with the same concentration.
4. Results The STA revealed the presence of laudanosine (atracurium metabolite), lidocaine, clonazepam, 7-aminoclonazepam, acetaminophen and metoclopramide in the blood. Acetone was determined in blood (30 mg/L), vitreous humour (34 mg/L) and urine (23 mg/L) by gas chromatography with flame ionization detector.
Fig. 3. Chromatogram of the HPLC/DAD analysis of the SPE extract of the seeds of Ricinus communis L. (A) with the ultraviolet spectrum of ricinine (B).
V. Coopman et al. / Forensic Science International 189 (2009) e13–e20
e17
Fig. 4. EI chromatogram of the SPE extract of the seeds of Ricinus communis L. with the mass spectrum of ricinine (Rt 22.5 min).
The morphological features of the peeled seeds and colour of the coat present in the bottle were consistent with seeds of R. communis L. Ricinine was identified in the sediment by comparison of retention times and spectra obtained by full-scan GC/MS, HPLC/DAD and LC/MS operated in the full-scan mode at cone voltage 20 V and 45 V with the retention times and spectra obtained from the analysis of a native ricinine standard. Seeds of R. communis L. were purchased from a local garden centre and were analyzed as reference. Fig. 3 shows the chromatogram of the HPLC/DAD analysis of the SPE extract of the seeds of R. communis L. with the ultraviolet spectrum of ricinine. The EI chromatogram of the SPE extract of the seeds of R. communis L. and the mass spectrum of ricinine are given in Fig. 4. The peak at 25.53 min in the EI chromatogram was identified as ricinoleic acid by NIST computer library search. In Fig. 5 the ESI+ chromatogram and spectra of the SPE extract of the seeds of R. communis L. at cone voltage 20 V and 45 V are shown. A five point calibration curve was constructed for the quantification of ricinine in the extract (y = 0.0360x + 0.0051, r2 = 0.9997). A concentration of 21.66 mg/mL ricinine (CV 0.96%) in the liquid was found. The liquid
was pH 7. Quantification of acetone in the extract was performed using gas chromatography with flame ionization detector. The gas chromatograph was equipped with a Carbopack B packed column. A five point calibration curve was constructed using t-butanol as internal standard (y = 1.7847x + 0.0410, r2 = 0.9995). An acetone concentration of 29.0 mL/100 mL in the extract was measured. SPE and LC–MS/MS with electro spray source in positive ionization mode was applied for the identification of ricinine in vitreous humour, blood and urine. Multiple reaction monitoring was used for specific detection. Retention times of ricinine and fentanyl-d5 were approximately 2.65 min and 9.18 min, respectively (Fig. 6). The ion suppression experiment noted the presence of matrix effect on the external standard peak in the blood extracts. The calibrators, quality control samples and postmortem tissue samples were analyzed in a single batch. The urine samples (y = 1.1677E 5x + 0.0020, r2 = 0.9964) produced a within-day precision of 0.08% at 0.5 ng/mL and 0.52% at 3.0 ng/mL, and a within-day accuracy of 117.5% at 0.5 ng/mL and 103.3% at 3.0 ng/mL. The extraction recovery (%) was determined by
e18
V. Coopman et al. / Forensic Science International 189 (2009) e13–e20
Fig. 5. ESI+ chromatogram (A) and spectra (B) of the SPE extract of the seeds of Ricinus communis L. at cone voltage 20 V and 45 V.
V. Coopman et al. / Forensic Science International 189 (2009) e13–e20
e19
Fig. 6. MRM chromatograms of a urine sample.
comparing the peak areas of extracted quality controls with the peak areas of unextracted solutions with the same concentration. The calculated values were 13.5% (CV 1.0%) and 8.6% (CV 0.7%) at 0.5 ng/mL and 3.0 ng/mL, respectively. Loss of the analyte occurred during the washing step. 5. Discussion In the presented case, the successive clinical symptoms were consistent with the typical clinical manifestations described in literature in cases of castor bean poisoning by injection [3,14]. On admission, the laboratory tests showed dehydration and metabolic acidosis. The patient deteriorated with haemolysis, liver failure, renal dysfunction and respiratory system failure. No antidote for castor bean poisoning is available. Despite symptomatic and supportive intensive care the patient died due to multiorgan failure. Based on the patient’s statement on admission, it was assumed that the man died approximately 35 h after subcutaneous and intravenous injection of the castor bean extract. Local tissue damage at the subcutaneous injection site was observed during the external body examination. Identification of ricinine in sediment and R. communis seeds was performed by SPE in combination with HPLC/DAD, full-scan GC/MS and full-scan LC/MS, respectively. The novel HPLC/DAD assay has proven to be applicable for the identification of ricinine in castor bean. The sample pre-treatment did not completely remove the castor oil. The triglyceride fatty-acid fraction of castor oil contains between 85.0% and 92.0% ricinoleic acid, an unsaturated omega-9fatty-acid (12-hydroxy-9-cis-octadecenoic acid) [5,6]. In the chromatogram of the full-scan GC/MS analysis, the peak with the highest abundance was identified as ricinoleic acid by NIST computer library search. A ricinine concentration of 21.66 mg/mL was measured in the extract using the LC–MS/MS method. On admission the man stated that he injected approximately 10 mL of the extract. Since the man weighted 83 kg this was consistent with approximately 0.0026 mg ricinine/kg bodyweight. A lethal dose of ricin by means of injection is estimated to be 0.01 mg/kg for an adult [9]. Reported contents in the castor bean vary between 1% and 5% for ricin, and between 0.3% and 0.8% for ricinine [9,19–22]. Considering the lethal dose of ricin
by injection, the ratio (1:5) of ricin to ricinine in castor bean and, given that ricinine contributes to the toxicity of the castor bean, a lethal castor bean poisoning was expected based on the level of ricinine in the extract and the statement of the patient. The laboratory’s standard screening procedure did not reveal the presence of the plant toxin. Hyphenated techniques such as LC– MS/MS are necessary to obtain sufficient sensitivity and specificity to detect low levels of toxins in post-mortem samples. Acetone was detected in blood, urine and vitreous humour. The levels did not indicate an acetone intoxication [23]. Acetone was most likely added to the bean pulp to remove the castor oil while the water soluble toxins remain in the sediment [9]. The homogenization of castor beans with acetone followed by filtration and dissolution in a drink or inhalation of the dry powder can be found on website pages describing manners to commit suicide with plant toxins. In the reported case a considerable amount of acetone remained in the watery castor bean extract. The presented SPE procedure and LC–MS/MS analysis was successfully applied for the identification of ricinine in urine, blood and vitreous humour. To the authors’ knowledge, it is the first time that ricinine is identified in vitreous humour in a case of castor bean poisoning. No ricinine and interfering substances were detected in blank urine and blood samples. The lowest calibration level was 0.2 ng/mL with a signal-to-noise of 8:1. Background levels of ricinine in 30 random human urine samples were investigated by Johnson et al. Ricinine was not detected in the samples indicating that contamination is not commonly present in general population [9]. In the presented case a ricinine concentration of 5.7 ng/mL in urine was measured. This concentration is in agreement with ricinine levels reported in literature in case of ricin poisoning. Ricinine is expected to be present in human urine at concentrations ranging from 0.08 ng/mL to 10 ng/mL 48 h after lethal exposure [9]. The validation study showed matrix effect in the blood extracts, making the quantification of ricinine inaccurate due to ion suppression on the external standard fentanyl-d5. No ion suppression was observed for ricinine in the blood extract. The use of a internal deuterated ricinine standard is therefore advisable. A semi-quantitative determination of ricinine in the
e20
V. Coopman et al. / Forensic Science International 189 (2009) e13–e20
blood was performed by comparing the peak area of ricinine from the blood samples to that of the spiked calibrators. A ricinine level of 2.3 ng/mL was calculated. 6. Conclusions Identification of ricinine in SPE extracts of castor beans can be performed by full-scan GC/MS, HPLC/DAD and full-scan LC/MS, respectively. The novel HPLC/DAD assay has proven to be applicable for the identification of ricinine in seeds of R. communis L. The exposure to the castor bean extract was confirmed through the detection of the biomarker ricinine in blood, urine and vitreous humour. To the authors’ knowledge, it is the first time that ricinine is reported in vitreous humour in a case of castor bean poisoning. Based on the clinical symptoms and the results of the toxicological analysis we concluded that death was caused by intoxication with plant toxins originated from R. communis L. The manner of death was presumed to be suicide due to intravenous and subcutaneous injection of a castor bean extract. References [1] R. Abrin, Two dangerous poisonous proteins, Jiri Patocka, The ASA Newsletter 85 (4) (2001) 20–25. [2] P.A. Steenkamp, Chemical Analysis of Medicinal and Poisonous Plants of Forensic Importance in South Africa, Ph.D. Thesis, May 2005. [3] J. Audi, M. Belson, M. Patel, J. Schier, J. Osterloh, Ricin poisoning. A comprehensive review, JAMA 294 (18) (2005) 2342–2351. [4] A. Pusztai, Plant Lectins. Chemistry & Pharmacology of Natural Product, Cambridge University Press, 1991. [5] European Pharmacopoeia, 6th edition, Volume 2, 01/2008, Monographs of castor oil virgin, castor oil hydrogenated and castor oil refined. [6] The Merck Index, 14th edition, Merck & Co., USA.
[7] G. Fodor, Colasanti, The pyridine and piperidine alkaloids: Chemistry and pharmacology, in: S. William Pelletier (Ed.), Alkaloı¨ds: Chemical and Biological Perspectives, vol. 3, John Wiley & Sons, 1983. [8] J.B. Harborne Frs, H. Baxter, G.P. Moss (Eds.), Dictionary of Plant Toxins, John Wiley & Sons, 1997. [9] R. Johnson, S. Lemire, A. Woolfitt, M. Ospina, K. Preston, C. Olson, J. Barr, Quantification of ricinine in rat and human urine: a biomarker for ricin exposure, J. Anal. Toxic. 29 (2005) 149–155. [10] P. Mouser, M. Filigenzi, B. Puschner, V. Johnson, M. Miller, S. Hooser, Fatal ricin toxicosis in a puppy confirmed by liquid chromatography/mass spectrometry when using ricinine as marker, J. Vet. Diagn. Invest. 19 (2007) 216–220. [11] T. Robinson, E. Fowell, A chromatographic analysis for ricinine, Nature 183 (1959) 833–834. [12] G.R. Waller, R. Ryhage, S. Meyerson, Mass spectrometry of biosynthetically labelled ricinine, Anal. Biochem. 16 (1966) 277–286. [13] S.M. Darby, M.F. Miller, R.O. Allen, Forensic determination of ricin and the alkaloid marker ricinine from castor bean extracts, J. Forensic Sci. 46 (5) (2001) 1033– 1042. [14] D. Targosz, L. Winnik, B. Szkolnicka, Suicidal poisoning with castor bean (Ricinus communis) extract injected subcutaneously: case report [abstract], J. Toxicol. Clin. Toxicol. 40 (2002) 398. [15] Y. Gaillard, G. Pepin, Poisoning by plant material: review of human cases and analytical determination of main toxins by high-performance liquid chromatography–(tandem) mass spectrometry, J. Chromatogr. B 733 (1999) 181–229. [16] Y. Gaillart, M. Cheze, G. Pepin, Main toxic plants responsible for human deaths: a revue, Ann. Biol. Clin. 59 (6) (2001) 764–765 (electronic version). [17] P.J. Aplin, T. Eliseo, Ingestion of castor oil plant seeds, Med. J. Aust. 167 (5) (1997) 260–261. [18] K.R. Challoner, M.M. McCarron, Castor bean intoxication: review of reported cases, Ann. Emerg. Med. 19 (1990) 1177–1183. [19] S.M. Bradberry, K.J. Dickers, P. Rice, G.D. Griffiths, J.A. Vale, Ricin poisoning, Toxicol. Rev. 22 (2003) 65–70. [20] V.M. Ghandi, K.M. Cherian, M.J. Mulky, Detoxification of castor seed meal by interaction with sal seed meal, JOACS 71 (1994) 827–831. [21] S.S. Kang, G.A. Gordell, D.D. Soejarto, H.D.S. Fong, Alkaloids and flavoids from Ricinus communis, J. Nat. Prod. 48 (1985) 155–156. [22] G.R. Waller, M.S. Tang, M.R. Scott, F.J. Goldberg, J.S. Mayes, H. Auda, Metabolism of ricinine in the castor plant, Plant Physiol. 40 (1965) 803–807. [23] R.C. Baselt, Disposition of Toxic Drugs and Chemicals in Man, Biomedical Publications, California, 1980, pp. 8–10.
