Tài liệu Drugs and Poisons in Humans - A Handbook of Practical Analysis (Part 53) pptx

6.3 6.3 © Springer-Verlag Berlin Heidelberg 2005 II.6.3 Tetrodotoxin by Sunao Fukushima and Yukio Ohtsuka Introduction Japanese people are very fond of eating fugu (pu er)  shes (especially Takifugu rubripes) as a feast especially in the winter season. However, the  sh contains highly poisonous toxin tetro- dotoxin (TTX) especially in its liver and ovary.  erefore, the number of fugu (pu er)  sh poisoning incidents in Japan were 554, in which 912 people were poisoned during 20 years in 1980–1999; among the 912 people, 106 subjects were fatal (mortality rate, 11.6 %) ( > Table 3.1) a . However, in recent 10 years, its incidence has been reduced to about half due to the improvement of early life-saving systems in emergency medicine.  e origins of tetrodotoxin, the presence of its analogs ( > Figure 3.1) and its mechanisms of action have been being clari ed [1,2]. However, for the tetrodotoxin analogs ( > Figure 3.1), their presence, toxicokinetics and toxicities especially in human pu er  sh poisoning cases have not been studied.  erefore, in this chapter, the target is limited to tetrodotoxin only. As a test for tetrodotoxin, a biological method using mice is long being used for testing  sh tissues; but it is not suitable for blood and urine specimens of poisoned patients. In this chapter, the methods of analysis of tetrodotoxin in human specimens by GC/MS and HPLC are described [3]. ⊡ Table 3.1 Incidence of fugu (puffer) fish poisoning cases in Japan Year The number of incidents The number of patients The number of fatality Mortality ratio (%) 1980–1984 143 242 47 19.4 1985–1989 147 256 29 11.3 1990–1994 138 221 13 5.9 1995–1999 126 193 17 8.8 Total 554 912 106 11.6 Structures of tetrodotoxin (TTX) and its analogs. ⊡ Figure 3.1 482 Tetrodotoxin GC/MS analysis TTX is soluble only in acidic alcohols; it cannot be analyzed it by GC/MS in its unchanged form. TTX should be treated with alkali to form 2-amino-6-hydroxymethyl-8-hydroxyquinazoline (so-called “ C 9 base”, M.W. 191), which can be trimethylsilylated [C 9 base-(TMS) 3 , M.W. 407] for GC/MS analysis ( > Figure 3.2). Reagents and their preparation i. Reagents TTX b , diethylamine, tri uoroacetic acid (TFA), β-estradiol (BD) and N,O-bis(trimethylsilyl)tri-  uoroacetamide (BSTFA) can be purchased from Sigma (St. Louis, MO, USA); SylonBFT (BSTFA/TMCS, 99:1) from Supelco (Bellefonte, PA, USA); trimethylchlorosilane (TMCS) from Pierce (Rockford, IL, USA). Other common chemicals used were of the highest purity commercially available. ii. TTX standard solution A 1-mg aliquot of TTX is dissolved in 10 µL acetic acid and diluted with distilled water to prepare the standard 10 mL solution (100 µg/mL). iii. C 9 base standard solution A 1-mg aliquot of TTX is dissolved in 0.5 mL of 5 % KOH solution and heated at 100 °C for 30 min. A er cooling it to room temperature, the solution is neutralized with concentrated HCl solution, saturated with KCl (by addition of KCl if necessary) and extracted with 5 mL of n-butanol three times.  e combined n-butanol extracts are evaporated to dryness under re- Pretreatment and derivatization procedures for TTX for its GC/MS analysis. ⊡ Figure 3.2 483 duced pressure; the resulting yellow residue (C 9 base) is dissolved in 10 mL of ethanol contain- ing 2 % acetic acid (C 9 base standard solution, 100 µg/mL). GC/MS conditions GC column c : a DB-5 fused silica capillary column (30 m × 0.25 mm i.d.,  lm thickness 5 µm, J&W Scienti c, Folsom, CA, USA). Conditions; instrument: a quadrupole GC/MS instrument d ; column (oven) temperature: 220 °C → 5 °C/min → 250 °C; injection temperature: 250 °C; detector temperature: 280 °C; carrier gas: He; its  ow rate: 1.0 mL/min; injection mode: splitless. Procedures i. Procedure 1 (body fluid and tissue specimens) [4] i. A 2-mL volume of a body  uid specimen (or well-homogenized suspension of a 2-g tissue specimen) is extracted with 30 mL methanol containing 2 % acetic acid 2–3 times with re uxing for 10 min each. ii.  e extracts are combined, passed through a  lter paper and evaporated to dryness under reduced pressure. iii. If necessary, the residue is degreased with diethyl ether e , dissolved in 5 mL of 0.5 % acetic acid solution and neutralized with 5 % KOH solution. iv.  e above solution is passed through the  rst Sep-Pak C 18 cartridge f (Waters, Milford, MA, USA), which had been pretreated by passing 5 mL methanol and 10 mL distilled water. A 3-mL volume of distilled water is also passed through it; both  ltrates are combined. v.  e  ltrate solution is mixed with 20 % KOH solution to adjust its pH to 9–10, and heat- ed at 100 °C for 20 min g . vi. A er cooling to room temperature, the above solution is neutralized with concentrated HCl solution, and poured into the 2nd Sep-Pak C 18 cartridge h , followed by washing with 2 mL water and 2 mL of methanol/distilled water (2:8); the target compound is eluted with 5 mL methanol. vii.  e eluate is evaporated to dryness under reduced pressure, and the residue is dissolved in a small amount of methanol containing 0.5 % acetic acid and transferred to a small glass vial with a Te on cap, and again evaporated to dryness under reduced pressure i . viii. A 10-µL aliquot of dimethylformamide and 30 µL of SylonBFT are placed in the above vial, capped airtightly and heated at 100 °C for 10 min. A er cooling to room tempera- ture, 10 µL diethylamine is added to the mixture to neutralize it. A 1-µL aliquot of the  nal solution is injected into GC/MS j,k . ii. Procedure 2 l (blood plasma) [5] i. A 2-mL volume of a plasma specimen is mixed with 30 mL of methanol containing 2 % acetic acid and extracted with re uxing in a water bath with heating. ii.  e mixture is centrifuged at 3,000 rpm for 5 min, and the resulting supernatant solution is evaporated to dryness under reduced pressure. iii.  e residue is shaken with 10 mL of 0.1 % acetic acid aqueous solution/chloroform (1:1). GC/MS analysis 484 Tetrodotoxin iv. It is centrifuged at 3,000 rpm for 5 min; the aqueous phase is passed through the  rst Sep-Pak PS-2 cartridge m (Waters). v.  e  ltrate is mixed with a half volume of 3 M KOH aqueous solution and heated for 15 min in a boiling water bath; a er cooling to room temperature, the solution is neutral- ized with 2 M HCl solution. vi.  e above solution is mixed with 17 µL TFA and 10 mL of 0.1 M phosphate bu er (pH 7.0) and poured into the 2nd Sep-Pak PS-2 cartridge m . vii. Just before the completion of the  ow through the cartridge, the  ow is stopped and le for 60 min n ; then the cartridge is washed with 10 mL of puri ed water and dried by pass- ing air through it. viii.  e C 9 base is eluted from the cartridge with 5 mL of acetonitrile containing 0.5 % acetic acid and evaporated to dryness under a stream of nitrogen. ix.  e residue is dissolved in a small amount of acetonitrile containing 0.5 % acetic acid and transferred to a small glass vial with a Te on cap, and again evaporated to dryness in it. x.  e residue is dissolved in 19 µL dimethylformamide containing 0.2 % BD, mixed with 22 µL BSTFA and 1 µL TMCS, capped airtightly and heated at 95 °C for 10 min for derivatization. x. A er cooling to room temperature, the above solution is neutralized with diethylamine; a 2-µL aliquot of the  nal solution is injected into GC/MS. Assessment of the methods i. Procedure 1 By GC/MS analysis of C 9 base-(TMS) 3 , a mass spectrum is obtained as shown in > Figure 3.3. Peaks appear at m/z 407, 392 and 376 [6].  e mass spectral pro le can be used for identi ca- tion.  e base peak at m/z 392 is used for quantitation with its peak areas by the external cali- bration method.  e detection limit by this method is about 1 ng/mL. ii. Procedure 2  e three ions are used for qualitative analysis.  e quantitation is performed using the peak area ratios of the ion at m/z 392 to that at m/z 285 [the base peak of BD-(TMS) 2 ].  e detection limit by this method is 0.5 ng/mL l in blood plasma. EI mass spectrum of the C 9 base-(TMS) 3 . ⊡ Figure 3.3 485 HPLC analysis Yasumoto et al. [7, 8] developed a TTX analyzer by combining an HPLC instrument with a  uorophotometer; it enables separation of TTX from crude biological matrices and its quanti- tation. Fuchi et al. [9] determined TTX in sea foods by a similar method.  e authors [10] also tried to measure the compound in urine of poisoned patients; the scheme of the system is shown in > Figure 3.4. A specimen mixed with a mobile phase enters a separation column by the action of Pump A. A er separation, TTX is decomposed to a  uorescent compound (a stable intermediate in the middle of reaction into the C 9 base) in a reaction box by alkaline solution, which is supplied by Pump B; the  uorescent compound is measured with the  uorescence detector a er cooling the reaction solution to room temperature. To obtain the best conditions of this system, the concentration of the alkaline solution, re- action temperature, and excitation and emission wavelengths should be optimized. Reagents and their preparation  ey are almost the same as described in the GC/MS section. Sodium dodecyl sulfate can be obtained from Sigma. HPLC conditions HPLC column: Inertsil ODS-2 (250 × 4.6 mm i. d., GL Sciences, Tokyo, Japan) or other columns of similar quality o . Conditions; pumps: LC-10AD for both;  uorescence detector: RF-550; reaction box: CRB-6A; reaction coil in the box: stainless steel tube to be used for HPLC (15 m × 0.25 mm i.d.) (all obtained from Shimadzu Corp., Kyoto, Japan); mobile phase for Pump A ( ow rate): 50 mM sodium potassium phosphate bu er containing 2 mM sodium dodecyl sulfate (pH 6.8) Schema of an HPLC system for analysis of TTX. ⊡ Figure 3.4 HPLC analysis 486 Tetrodotoxin (0.4 mL/min); mobile phase for Pump B ( ow rate): 4 M NaOH (0.4 mL/min) p ; reaction box temperature: 120 °C; exitation wavelength: 400 nm; emission wavelength: 495 nm q . Procedure i. A 2-mL volume of a urine specimen is mixed with 0.5 mL of 0.5 % acetic acid solution, and poured into a Bond Elut SCX (Varian, Harbor City, CA, USA) cartridge, which had been equilibrated with 5 mL puri ed water, 5 mL methanol and 5 mL of 0.1 % acetic acid solu- tion. ii.  e cartridge is washed with 2 mL of 0.1 % acetic acid solution, 2 mL methanol and 4 mL puri ed water. iii. TTX is eluted from the cartridge with 4 mL of 0.1 % sodium potassium phosphate bu er (pH 7.0); 20-µL of the eluate is injected into HPLC. iv.  is method is applicable only to urine specimens at the present time. Assessment of the method A er addition of a  xed amount of TTX to urine, its concentration is measured by HPLC and conventional GC/MS (Procedure 1); the values obtained by the methods were almost the same.  erefore, it was concluded that the quantitation of TTX can be made even by HPLC [10]. For urine specimens obtained from TTX poisoning cases, the quantitation was made by both methods as shown in > Table 3.2; each case showed very similar values obtained by both methods [11]. Except for Case No. 1, the three patients listed in > Table 3.2 showed poison- ing symptoms of intermediate severity, but recovered a er treatments. ⊡ Table 3.2 Comparison of measurements of tetrodotoxin (TTX) in urine by HPLC with those by GC/MS in actual poisoning cases Victim (age/sex) TTX concentration (ng/mL) HPLC GC/MS No. 1 (22/M) 2,550 2,480 No. 2 (51/F) 146 132 No. 3 (52/M) 152 135 No. 4 (39/M) 301 288 Poisoning cases and toxic concentrations Blood TTX concentrations and fatal levels  e authors also experienced the analysis of TTX in blood and/or urine of 13 subjects in TTX poisoning. As shown in > Table 3.3, the number of blood specimens was 12 in 13 cases. Among the poisoned subjects, 3 subjects were fatal and 10 recovered. In Case No. 1 of the table, 487 the victim cooked Fugu niphobles in a large amount, was poisoned by eating its liver and died just a er arrival at a hospital. In Case No. 3, the victim was found dead in a ship on ocean navigation. In Case No. 4, the victim was found dead in an automobile in a public park.  e cases Nos. 3 and 4 were treated as unnatural death. In other cases, all patients recovered a er being admitted in hospitals. In Cases Nos 2, 5, 9 and 11, the subjects landed fugu  shes by themselves and ate them a er cooking to be poisoned; they were treated as self-negligence cases. In Case No. 6, the victim bought the skin of a fugu  sh at a store and was poisoned; in Cases Nos. 7, 8, 10 and 13, the victims ate fugu dishes containing the organs (the liver and/or skin) of the  sh at restaurants. All of these cases were treated as negligent homicide incidents; the store and restaurants were ordered to suspend their business.  e accidents occurred due to erroneous knowledges on cooking methods of fugu  shes. > Figure 3.5 shows the plots of blood TTX concentrations on a calibration curve; its con- centrations in the fatal cases are higher than those in the survived cases. It can be estimated that borderline blood concentrations between the fatal and survived cases seem to be about 100 ng/mL, which is in accordance with that reported by Suenaga et al. [6]. At 25–100 ng/mL of TTX in blood, paresthesia, verbal paralysis, disappearance of various re exes and  nally respiratory paralysis appear as poisoning symptoms; however, in these cases, by early emer- gency treatments, they could survive. ⊡ Table 3.3 TTX concentrations in human specimens in its poisoning Victim (age/sex) Specimen Concentration (ng/mL) Outcome No. 1 (93/F) blood urine 93.0 650 dead No. 2 (62/M) blood 36.3 recovered No. 3 (34/M) blood 175 dead No. 4 (65/M) blood 320 dead No. 5 (39/M) blood urine 12.0 295 recovered No. 6 (57/M) blood urine 35.5 78.5 recovered No. 7 (50/M) blood urine 6.5 443 recovered No. 8 (?/M) blood urine 2.5 105 recovered No. 9 (35/M) blood urine 63.9 27.2 recovered No. 10 (39/M) urine 68.5 recovered No. 11 (51/M) blood 37.9 recovered No. 12 (?/M) blood 16.1 recovered No. 13 (49/M) blood urine 15.3 245 recovered Poisoning cases and toxic concentrations 488 Tetrodotoxin Periods for TTX excretion into urine Urine is a very advantageous specimen, because it can be obtained noninvasively and contains relatively high concentrations of poisons and relatively low contents of impurities; these advan- tages make its analysis simple and rapid. As shown in > Table 3.3, among 13 victims, urine specimens were available for 8 victims; only one was fatal among the 8 victims with her urinary concentration being 650 ng/mL.  e TTX excretion into urine was monitored as a function of time a er ingestion. > Figure 3.6 shows the time course of TTX excretion into urine obtained by sampling stockpiled urine every 12 h up to 3 days for Cases 6 and 9 listed in > Table 3.3. For both subjects, the excretion was highest in the period from 12 to 24 h; TTX was detectable from urine even on the 3rd day. Unfortunately, urine specimens could not be obtained a er 3 days; therefore, it is not clear how long the urinary TTX is detectable a er ingestion. Calibration curve for blood TTX. ⊡ Figure 3.5 Urinary excretion of TTX as a function of time after ingestion. ⊡ Figure 3.6 489 Notes a)  ese  gures were obtained from data collected by Bureau of Medical Drugs, Ministry of Health, Labour and Welfare of Japan. b) TTX is not commercially available as pure crystals; TTX powder for a biochemical use (about 99 %) is available. c) Packed columns (5 % SE-52, 1 m × 3 mm i.d.) or wide-bore capillary columns ( DB-17, 15 m × 0.53 mm i.d.) can be also used. However, in view of separation ability and con- tamination, medium-bore capillary columns are preferable. d) For example, Shimadzu QP5050A, Shimadzu 1100EX or HP5971A can be used. e)  e extent and times of washings of the residue with diethyl ether are di erent in di erent specimens (impurities or lipid contents); some skillfulness based on experience is required for the technique. f)  e  rst Sep-Pak cartridge is not used for extraction of TTX, but used only for removal of hydrophobic impurities being contained in TTX specimens. g) As shown in > Figure 3.7, the conversion of TTX into C 9 base is completed in about 10 min; a er 30 min of heating, the recovery becomes much lower. h)  e 2nd Sep-Pak C 18 cartridge is used for extraction of the C 9 base produced by the alkali treatment. i) It is essential to dry it up completely for silylation; it is sometimes dried up under reduced pressure in the presence of phosphorus pentaoxide. j)  e column should be  lled with the silylating reagent gas. When the silylating reagent only is injected into GC/MS between the injections of sample extracts, reproducibility of the assay may be enhanced. k) When a packed column is used for GC/MS, the residue is mixed with 80 µL DMF, 200 µL BSTFA and 10 µL TMCS and heated at 100 °C for 10 min; a er cooling to room tempera- ture, the mixture is neutralized with 10 µL diethylamine [C 9 base-(TMS) 3 derivative mix- ture, total volume 300 µL] and 3-µL of the solution is injected into GC/MS. Formation rates of the C 9 base as a function of time of heating TTX at 100 °C. ⊡ Figure 3.7 Poisoning cases and toxic concentrations 490 Tetrodotoxin l)  e procedure 2 gives higher sensitivity than the procedure 1. However, for actual mea- surements of poisoning specimens ( > Table 3.3), the procedure 1 had been used. m) For activation of Sep-Pak PS-2 cartridges, methanol, puri ed water, acetonitrile containing 0.5 % acetic acid, puri ed water, 1.68 % (50 mM) EDTA aqueous solution and puri ed water, 3 mL each, were passed though them. For the 2nd cartridge, 3 mL of 0.1 M sodium potassium phosphate bu er (pH 7.0) was added at the end for activation. Like the above Sep-Pak C 18 cartridges, the  rst Sep-Pak PS-2 cartridge is used only for removal of hydro- phobic impurities and the 2nd one for extraction of the C 9 base produced by alkali treat- ment from TTX. n) To secure the complete adsorption of the neutral C 9 base to the cartridge, the 60 min inter- val is necessary. o) For the TTX analyzer, Hitachi gel 3011C (400 × 5 mm i. d.) (GL Sciences) is being used. p) Since the concentration of NaOH is very high, the whole line system should be washed completely a er use. q)  e ranges of various conditions used for preliminary optimization experiments were: NaOH concentration: 1–5 M; reaction box temperature: 100–140 °C; excitation wave- length: 390–410 nm; emission wavelength: 485–505 nm. References 1) Hashimoto C (ed) (1988) Recent Advance of Fugu Toxin Studies. Koseisha-koseikaku, Tokyo (in Japanese) 2) Noguchi T, Arakawa O, Hashimoto C (1989) Fugu poison: its origins and the mechanisms of toxigenicity. Jpn J Food Hyg Assoc 30:281–288 (in Japanese with an English abstract) 3) Matsui T, Ohtsuka Y, Sakai J (2000) Recent advance of studies on fugu toxin. Yakugaku Zasshi 10:825–837 (in Japanese with an English abstract) 4) Fukushima S (1992) Analysis of tetrodotoxin in body fluids and tissues. Reports of Studies by the 9th Trainees of Forensic Science Training Institute, Forensic Science Training Institute, Tokyo, pp 285–295 (in Japanese) 5) Ohtsuka Y, Tokunaga H, Fukushima S et al. (1997) Sensitive determination of tetrodotoxin in serum by GC/MS. Proceedings of TIAFT XXXV Annual Meeting, Padova, Italy, pp 614–618 6) Suenaga K, Kotoku S (1980) Detection of tetrodotoxin in autopsy material by gas chromatography. Arch Toxicol 44:291–297 7) Yasumoto K (1989) Analysis and applications of marine toxins. Chemistry and Biology 27:401–406 (in Japanese) 8) Yasumoto T, Mitishita T (1985) Fluorometric determination of tetrodotoxin by high performance liquid chroma- tography. Agr Biol Chem 49:3077–3080 9) Fuchi Y, Morisaki S, Nagata T et al. (1988) Determination of tetrodotoxin in sea foods by high-performance liquid chromatography. Jpn J Food Hyg Assoc 5:306–312 (in Japanese with an English abstract) 10) Fukushima S (1996) Trace analysis of tetrodotoxin in human specimens. Simple determination by HPLC with fluorescence detection. Abstracts of the 116th Annual Meeting of the Pharmaceutical Society of Japan, p 188 (in Japanese) 11) Fukushima S (1996) Analysis of tetrodotoxin in specimens of poisoned victims. Jpn J Toxicol 9:473–474 (in Japanese) . dissolved in a small amount of acetonitrile containing 0.5 % acetic acid and transferred to a small glass vial with a Te on cap, and again evaporated to. A 2-mL volume of a plasma specimen is mixed with 30 mL of methanol containing 2 % acetic acid and extracted with re uxing in a water bath with heating. ii.