Distribution of TTX in animals
Tetrodotoxin (TTX), a pufferfish (“fugu” in Japanese) toxin named after its order name Tetraodontiformes, is the toxic principle of puffer fish poisoning. This toxin (C11H17N3O8; a molecular weight of 319) is one of the most potent nonproteinaceous toxins as well as the best-known marine natural toxins (Figure 1). In Japan, pufferfish have been a traditional food for many years, and since people have become accustomed to eating them, cases of TTX poisoning are frequent. It poses a serious hazard to public health. These cases have occurred on a regular basis not only in Japan but also in Asia for a number of years, sporadically resulting in severe poisoning or even death. On the other hand, the Japanese are aware of the its toxicity and have devised methods to reduce TTX levels especially in the liver. However, TTX poisoning incidents continue to occur in Japan. Since there is no antidote for the toxin, patient mortality is very high. Judging from statistics provided by the Japanese Ministry of Health, Labour and Welfare, the number of deaths due to puffer poisoning has steadily declined, from more than 10 cases every year between 1960 and 1981 to less than 10 cases with low mortality every year since 1982, generally with low mortality. This decline is probably due to not only strict adherence to government regulations but also an increase in cultured puffer rather than a decrease in wild puffer. The toxicosis is characterized by the onset of symptoms in the victim. Treatment of the illness is mainly based on the symptoms of the patient. More fruitful treatment can be provided if the causative toxin is identified. In 1950, TTX was isolated for the first time as a crystalline prism from toxic pufferfish ovaries by Yokoo . Its structure was elucidated by three groups in 1964 [2-4]. TTX is a powerful and specific sodium channel blocker . When ingested by humans, it acts to block the sodium channels in the nerve cells and skeletal muscles, and to thereby block excitatory conduction, resulting in the occurrence of typical symptoms and signs such as respiratory paralysis and even death in severe cases. The lethal potency is 5000 to 6000 MU/mg. One MU (mouse unit) is defined as the amount of toxin required to kill a 20g male mouse within 30 min after intraperitoneal administration, and the minimum lethal dose (MLD) for humans is estimated to be approximately 10,000 MU
equivalent to 2 mg of pure TTX crystals . Many derivatives of TTX have been found, although their toxicities vary widely. As seen in Figures 1, 2 TTX is a heterocyclic guanide compound whose chemical structure has been characterized. Various TTX derivatives from pufferfish and other TTX-bearing organisms have been identified to date as a result of recent progress in the instrumental analysis of TTX (Figure 3). In marine pufferfish species, toxicity is generally highest in the liver and ovary, whereas in brackish and freshwater species, toxicity is higher in the skin [7-13]. TTX was long believed to be present only in the pufferfish. In 1964, Mosher
(A) Hemilactal type, (B) Lacton type, (C) 4,9-Anhydro type
The structure of three types of TTX analogues
2. Isolation of TTX crystals by column chromatography
Of the TTX-bearing animals, our specimens of ribbon worms (“himomushi” in Japanese) adherent to the cultured oyster
3. HPLC – Fluorescence detection
Rapid progress in TTX research, especially in intoxication mechanism of TTX-bearing organisms, is due to recent advancements in instrumental analysis. In particular, postcolumn-HPLC fluorescence detection (HPLC-FLD) methods expected to replace the conventional mouse bioassay, have been explored by many researchers for both qualitative and quantitative analysis of TTX and its analogs. HPLC techniques allow the separation and sensitive detection of individual TTX and its analogs irrespective of their number and group. Therefore, HPLC methods have opened up a new dimension in TTX analysis. However, the results obtained have to be comparable to those of the mouse bioassay. Additionally, accurate HPLC determination of the various TTX components in the samples is a necessity. Using these methods, the toxic principles produced peaks identical to those of authentic TTX and its derivatives. The HPLC- FLD method utilizes a computer controlled by a high pressure pump with a syringe loading sample injector or an autosampler system, a stainless steel column, a reaction pump for delivering reagents, and a fluoromonitor and chromato-recorder for calculation of the peak area. In this method, a strong alkali treatment is applied to TTX which produces a fluorescent compound with excitation and emission wavelengths of 384 and 505 nm, respectively. In this system, first, toxins are separated from the contaminants by a buffer solution on a reversed-phase column packed with C18 resin with an ion-pair reagent (sodium 1-heptanesulfonate; HSA). Then, the isolated toxins are mixed with NaOH, which converts them into fluorescent compounds that are then passed through a stainless steel tube (φ 0.25mm × 100cm) placed in an oven. Eventually, when the fluorescent compounds are passed through a fluoromonitor equipped with a lamp, the retention time of the toxin and fluorescence intensity are recorded. The treated toxins are identified by comparing their retention times with those of authentic TTXs. For the quantitative analysis by HPLC, the detection limit of TTX is approximately 0.03μg, which is satisfactory for practical applications. To date, several continuous improvements have been made to detect TTX and its analogs under different HPLC conditions, and a number of advances in understanding the biochemistry of TTXs are the outcomes of these developments. Briefly, a few promising methodologies are described as follows. In the early 1980’s, a fluorometric continuous TTX analyzer was constructed by combining HPLC and a post-column reaction with NaOH to monitor potentially harmful puffer toxins . In this system, the toxin was first separated from contaminants on a column composed of a weak cation exchange gel with a 0.06 M citrate buffer solution (pH 4.0), and toxin concentrations above 8 MU/g were detected. However, because of the poor performance of the original system in separating and detecting TTX analogs, an improved analyzer was later constructed. Using HPLC-FLD, naturally occurring TTX analogs, 4-
|HPLC control system||JASCO - BORWIN/HSS-2000|
|Column||LiChroCART 250-4 (Merck) (LiChrospher 100 RP-18e, 5μm)||Column size: 4 x 250 mm|
|Column temperature||30oC||CO-2067 plus (JASCO)|
|Mobile phase||60mM ammomium phosphate buffer (pH5.0) containing 10mM HAS and 2% CH3CN||PU-2080 plus (JASCO) 0.5 ml/min|
|Reagent||3 M NaOH||MINICHEMI PUMP SP-D-2502 (Nihonseimitsukagaku) 0.5 ml/min|
|Reaction temperature||110oC||860 CO (JASCO)|
|Detection||Excitation 384nm, emission 505nm||JASCO FP-2025 plus|
4. Mass spectrometry
4.1. Gas-Chromatography-Mass Spectrometry
Gas chromatography (GC) and mass spectrometry (MS) form an effective combination for chemical analysis. GC-MS analysis is an indirect method to detect TTX in a crude extract which is difficult to purify in other advanced analysis methods . In this method, TTX and its derivatives are dissolved in 2 ml of 3 M NaOH and heated in a boiling water bath for 30 min. After cooling to room temperature, the alkaline solution of decomposed compounds is adjusted to pH 4.0 with 1N HCl and the resulting mixture is chromatographed on a Sep-Pak C18 cartridge (Waters). After washing with H2O first and then 10% MeOH, 100% MeOH fraction were collected and evaporated to dryness
4.2. Fast atom bombardment mass spectrometry
Fast atom bombardment mass spectrometry (FAB-MS) is a direct method for the qualitative confirmation of TTX. The analysis was performed on a JEOL JMX DX-300 mass spectrometer . Xenon is used to provide the primary beam of atoms, the acceleration voltage of the primary ion being 3 kV. Scanning is repeated within a mass range of
4.3. Liquid chromatography mass spectrometry
Liquid chromatography-mass spectrometry (LC-MS) is developed to detect TTX with considerable accuracy . The major disadvantage of LC-FLD is the large difference in the structure-dependent fluorescence intensities of the analogs. In particular, the fluorescence intensities of 5-deoxyTTX and 11-deoxyTTX are approximately 1/20 and less than 1/100 of that of TTX, respectively, while those of 6-
4.4. Electrospray ionization – Time of flight – Mass spectrometry
Electrospray ionization time of flight mass spectrometry (ESI-TOF-MS) is applicable to many fields including the analysis of proteins, natural extracts, synthetic mixtures and medical drugs. ESI-TOF-MS is a valuable technique for identification of TTX, although it is not widely used to date in marine toxin determinations. In this analysis, a portion of purified TTX (less than 0.05 mg) is dissolved in a small amount of 1% AcOH, and the resulting solution is added to 50% aqueous MeOH. ESI-TOF-MS is run on a Micromass Q-TOF mass spectrometer. TTX in a tree frog
5. Infrared (IR) spectrometry
IR spectrometry is the analytical technique for the determination of functional groups in TTX. Although the IR spectrum is presumed to be complex, it is a helpful tool to identify TTX. IR-spectra of KBr pellet were acquired using IR spectrophotometer, which was used by Onoue
6. Ultraviolet (UV) spectroscopy
In UV spectroscopy, TTX is generally determined by irradiating a crude toxin with UV light. A small amount of TTX is dissolved in 2 ml of 2 M NaOH and heated in a boiling water bath for 45 min. After cooling to room temperature, the UV spectrum of the solution is examined for characteristic absorptions, associated with C9-base, 2-amino-6-hydroxymethyl-8-hydroxyquinazoline, possibly formed from TTX and/or related substances, if present. In the analysis, the UVspectrum of the alkali decomposed compounds of TTX appears as a shoulder at near 276 nm, indicating the formation of C9-base specific to TTX or related substances (Figure 17). Saito
7. Proton nuclear magnetic resonance (1H-NMR) spectrometry
1H-NMR has played an important role as a complementary method to determine the absolute configuration of TTX. To date, many derivatives of TTX have been isolated, and their 1H-NMR data have been reported by various investigators. In a typical 1H-NMR analysis, 5 mg of TTX crystals have been dissolved in 0.5 ml of 1% CD3COOD in D2O, and placed in a test tube. Figure 18 shows the 1H-NMR spectrum obtained with a 500 MHz JEOL JNM-500 spectrometer, using the methyl group protons of acetone as the internal standard . The 1H-NMR spectrum exhibited a singlet at 2.20 ppm (CH3COCH3), a doublet centered at 2.33 ppm (J =10.0 Hz), a large proton peak at 4.76 ppm (HDO) and a doublet centered at 5.48 ppm (J=10.0 Hz). The pair of doublets around 2.33 and 5.48 ppm, which are the hallmarks of TTX and are assigned to H-4a and H-4, respectively, have been confirmed to be coupled with each other by double irradiation (Figure 19). These results agree well with the corresponding data of TTX. The signals at 4.24, 4.06, 4.28, 3.94, 4.00 and 4.02 ppm are assigned toH-5, H-7, H-8, H-9 and H-11, respectively (Figure 20). A toxin isolated from the horseshoe crab
8. Thin-layer chromatography
TLC is a very commonly used technique in synthetic chemistry for identifying compounds, determining their purity. In TLC analysis, TTX is spotted onto a silica gel 60 precoated plate (Merck). The plate is developed in two different solvent systems of pyridine-ethyl acetate-AcOH -water (15:5:3:4) and 3-BuOH-AcOH-water (2:1:1) in a sealed chamber. The solvent rises by capillary action and an ascending chromatographic separation is obtained. The plate is then sprayed with 10% KOH followed by heating at 100℃ for 10 minutes. The toxin is visualized as a yellow fluorescent spot under UV light (365nm). In TLC analysis, the Rf values of TTX are around 0.71(0.65) and 0.50, respectively [10, 51]. It is also possible to detect TTX on the TLC plate using the Weber reagent that gives pink spot of the toxin. The detection limit is about 2 μg of TTX (10 MU). TLC is a useful technique in those laboratories where HPLC and other costly analytical systems are not available.
Electrophoresis is a relatively simple and rapid method with high resolution detection of polar compounds like TTX [10, 11, 51]. When 1 μl of TTX (10 MU, corresponding to 2 μg) is applied onto a 5 x 18 cm cellulose acetate membrane (Chemetron, Milano), the ion molecules of TTX move toward the cathode with a mobility (Rm) clearly smaller than that of authentic of STX. The analysis is performed for 30 minutes in an electrolytic buffer solution of 0.08 M Tris-HCl (pH8.7), under the influence of an applied electric field with a constant current of 0.8 mA/cm width. The toxin is visualized in the same manner as described for TLC.
10. Capillary isotachophoresis
Capillary isotachophoresis proved to be a very effective technique for the analyses of organic acids, carbohydrates, drugs and amino acids. It is a rapid and accurate detection technique for TTX . It is performed using a cationic system, as TTX exists as cation under acidic and neutral conditions. Conditions for capillary isotachophoresis composed of a leading electrolyte of 5 mM potassium acetate (pH6.0), containing 0.2% Triton X-100 and 0.5 volume of dioxane, and a terminating electrolyte of 10mM β-alanine adjusted to pH 4.5 with acetic acid. When TTX is applied to isotachophoretic analyzer (Shimadzu IR-2A) equipped with a potential gradient 0.32, it is eventually monitored by the detector. The quantitative detection limit by this method is about 0.25μg of TTX. It was possible to quantify TTX content of contaminated puffer extracts without any pretreatment.
In an attempt to protect consumers from TTX-intoxication, the mouse bioassay has historically been the most universally applied tool to determine the toxicity level in monitoring programs. This bioassay, however, shows low precision and requires a continuous supply of mice of a specific size. These potential drawbacks and world-wide pressure to refrain from the unnecessary killing of live animals subsequently led scientists to develop alternative chemical methods to the mouse bioassay for TTX detection and quantification. TTX levels in pufferfish are normally estimated using the mouse bioassay. However, this assay and other techniques such as TLC, electrophoresis, LC, spectrophotometry, and the enzyme-linked immunosorbent assay (ELISA) pose ethical concerns, are not specific and lack sensitivity and precision at low concentrations. HPLC-FLID and LC-MS/GC-MS are sensitive techniques for identification of TTX. However, due to the complexity of sample matrices and insolubility of TTX in organic solvents, HPLC-FLD and LC-MS (or LC-MS/MS) are more preferred methods than GC-MS. MS spectrometry is a powerful technique that also has an important future for the analysis of marine toxins. In addition to high sensitivity and selectivity, MS can provide structural information useful for the confirmation of toxin identity and the identification of new toxins. The drawback of LC-MS and LC-MS/MS analyses is that they involve the use of expensive instruments, which require higher maintenance compared to GC-MS. Nevertheless, for routine analysis of TTXs, HPLC-FID and LC-MS (LC-MS/MS) are expected to replace the conventional mouse bioassay.
This work was supported by a Grant-in-Aid for Scientific Research (C) (2) (No.11660205), (C)(No.18580204), (C)(No.20580220) and (B)(No.20380110) from Japan Society for the Promotion of Science. We are grateful to Dr.Hiroshi Kajihara (Hokkaido University) for his helpful advice on ribbon worms.
Yokoo A. 1950Chemical studies on pufferfish toxin (3)- separation of spheroidine. Nippon KagakuZasshi, 71 590 592.
Tsuda K. Ikuma S. Kawamura M. Tachikawa R. Sakai K. Tamura C. Amakasu D. 1964 Instrumental Analysis of Tetrodotoxin. Chem.Pharm. Bull. 12 1357 1374.
Woodward R. B. 1964 Instrumental Analysis of Tetrodotoxin. Pure Appl. Chem. 9 49 74.
Goto T. Kishi Y. Takahashi S. Hirata Y. 1965Tetrodotoxin. Tetrahedron 21 2059 2088.
Narahashi T. 2001 Instrumental Analysis of Tetrodotoxin. J. Toxicol.-Toxin Rev. 20 67 84
Noguchi T. Ebesu J. S. M. 2001 Instrumental Analysis of Tetrodotoxin. J. Toxicol.-Toxin Rev. 20 1 10.
Hashimoto Y. 1979 Instrumental Analysis of Tetrodotoxin. Japan Scientific Societies Press, Tokyo, 70 91.
Khora S. S. Isa J. Yasumoto T. 1991 Instrumental Analysis of Tetrodotoxin, 57 163 167.
Mahmud Y. Tanu M. B. Takatani T. Asayama E. Arakawa O. Noguchi T. 2001 Instrumental Analysis of TetrodotoxinJ. Nat Toxins, 10 69 74.
Mahmud Y. Yamamori K. Noguchi T. 1999 Instrumental Analysis of TetrodotoxinJ.Food Hyg. Soc., 40 363 367.
Mahmud Y. Yamamori K. Noguchi T. 1999Toxicity and tetrodotoxin as the toxic principle of a brackishwater puffer Tetraodon steindachneri, collected from Thailand, J.Food Hyg. Soc., 40 391 395.
Hwang D. F. Kao C. Y. Yang H. C. Jeng S. S. Noguchi T. Hashimoto K. 1992Toxicity of puffer in Taiwan, Nippon Suisan Gakkaishi, 58 1541 1547.
Mahmud Y. Tanu M. B. Noguchi T. 1999 Instrumental Analysis of Tetrodotoxin. J.Food Hyg. Soc., 40 473 480.
Mosher H. S. Fuhrman F. A. Buchwald H. G. Fischer H. G. 1964Tarichatoxin- tetrodotoxin, a potent neurotoxin. Science 144 1100 1110.
Noguchi T. Kao H. Hashimoto Y. 1971 Instrumental Analysis of Tetrodotoxin. Bull. Jpn. Soc.Sci.Fish., 37 642 647.
Kim Y. H. Brown G. B. Mosher H. S. Furman F. A. 1975Tetrodotoxin : occurrence in Atelopid frogs of Costa Rica. 189 151 152.
Sheumack D. D. Howden M. E. H. Spence I. Quinn R. J. 1978 Instrumental Analysis of Tetrodotoxin. 199 188 189.
Narita H. Noguchi T. Maruyama J. Ueda Y. Hashimoto K. Watanabe Y. Hida K. 1981Occurrence of tetrodotoxin in a trumpet shell ”boshubora”, Charonia sauliae. Bull. Japan. Soc. Sci. Fish., 47 935 941.
Noguchi T. Maruyama J. Ueda Y. Hashimoto K. Harada T. 1981 Instrumental Analysis of TetrodotoxinBull. Jpn. Soc.Sci.Fish., 47 909 913.
Noguchi T. Maruyama J. Narita H. Hashimoto K. 1984 Instrumental Analysis of Tetrodotoxin. 22 219 226.
Noguchi T. Narita H. Maruyama J. Hashimoto K. 1982Tetrodotoxin in the starfish Astropecten polyacanthus, in association with toxification of a trumpet shell, “boshubora”, Charonia sauliae. Bull. Japan. Soc. Sci. Fish. 48 1173 1177.
Kawabata T. 1978In: Ministry of Health and Welfare of Japan, (Ed.), Assay Method for Tetrodotoxin, Food Hygiene Examination Manual, 2Japan Food Hygiene Association, Tokyo, 232 239.
Asakawa M. Toyoshima T. Shida Y. Noguchi T. Miyazawa K. 2000Paralytic toxins in a ribbon worm Cephalothrix species (Nemertean) adherent to cultured oysters in HiroshimaBay, Hiroshima Prefecture, Japan. Toxicon 38 763 773.
Yasumoto T. Nakamura M. Oshima Y. Takahata J. 1982Construction of a continuous tetrodotoxin analyzer. Bull.Jpn.Soc.Sci.Fish., 48 1481 1483.
Nakamura M. Yasumoto T. 1985 Instrumental Analysis of Tetrodotoxin. 23 271 276.
Yasumoto T. Yotsu M. Murata M. Naoki H. 1988New tetrodotoxin analogues from the newt Cynops ensicauda. J.Am.Chem. Soc., 110 2344 2345.
Khora S. S. Yasumoto T. 1989Isolation of 11-oxotetrodotoxin from the puffer Arothron nigropunctatus. Tetrahedron Lett. 30 4393 4394.
Endo A. Khora S. S. Murata M. Naoki H. Yasumoto T. 1988Isolation of 11-nortetrodotoxin-6(R)-ol and other tetrodotoxin derivatives from the puffer Fugu niphobles. Tetrahedron Lett. 29 4127 4128.
Yotsu M. Hayashi Y. Khora S. S. Sato S. Yasumoto T. 1992Isolation and structural assignment of 11-nortetrodotoxin-6(S)-ol from the puffer Arothron nigropunctatus. Biosci. Biotech. Biochem. 56 370 371.
Kotaki Y. Shimizu Y. 19931-hydoxy-5, 11-dideoxytetrodotoxin, the first N-hydroxy and ring-deoxy derivative of tetrodotoxin found in the newt Taricha granulosa. J.Am.Chem.Soc. 115 827 830.
Yotsu-Yamashita M. Yamagishi Y. Yasumoto T. 19955, 6, 11-trideoxytetrodotoxin from the puffer fish Fugu poecilonotus. Tetrahedron Lett. 36 9329 9332.
Yotsu-Yamashita M. Schimmele B. Yasumoto T. 1999 Instrumental Analysis of TetrodotoxinBiosci. Biotech. Biochem. 63 961 963.
Yotsu M. Endo A. Yasumoto T. 1989 Instrumental Analysis of TetrodotoxinAgric. Biol. Chem., 53 893 895.
Yotsu-Yamashita M. Meds D. Kwet A. Schneider M. 2007Tetrodotoxin and its analog 6-epitetrodotoxin in newts (Triturus spp.; Urodela, Salamandridae) from southern Germany. Toxicon 50 306 309.
Nagashima Y. Maruyama J. Noguchi T. Hashimoto K. 1987 Instrumental Analysis of Tetrodotoxin, 53 819 823.
Asakawa M. Delan G. G. Tsuruda S. Shimomura M. Shida Y. Taniyama S. Quilantang M. B. Shindo J. 2010 Instrumental Analysis of Tetrodotoxin. J.Toxicology., Article ID172367.
Noguchi T. Jeon J. K. Arakawa O. Sugita H. Deguchi Y. Shida Y. Hashimoto K. 1986 Instrumental Analysis of Tetrodotoxin.Biochemistry, 99 311 314.
Narita H. Noguchi T. Maruyama J. Nara M. Hashimoto K. 1984 Instrumental Analysis of Tetrodotoxin. Bull.Jpn.Soc.Sci.Fish, 50 85 88.
Asakawa M. Toyoshima T. Ito K. Bessho K. Yamaguchi C. Tsunetsugu S. Shida Y. Kajihara H. Mawatari S. F. Noguchi T. Miyazawa K. 2003 Instrumental Analysis of Tetrodotoxin. , 41 747 753.
Maran B. A. V. Iwamoto E. Okuda J. Matsuda S. Taniyama S. Shida Y. Asakawa M. Ohtsuka S. Nakai T. Boxshall G. A. 2007 Instrumental Analysis of Tetrodotoxin. , 50 779 790.
Man C. N. Noor N. M. Harn G. L. Lajis R. Mohamad S. 2010 Instrumental Analysis of Tetrodotoxin. J.Chrom.A, 1217 7455 7459.
Kurono S. Hattori H. Suzuki O. Yamada T. Seno H. 2001 Instrumental Analysis of TetrodotoxinAnal.Letts., 34 2439 2446.
Noguchi T. Ali A. E. Arakawa O. Miyazawa K. Kanoh S. Shida Y. Nishio S. Hashimoto K. 1991Tetrodonic acid-like substance: a possible precursor of tetrodotoxin. Toxicon, 29 845 855.
Nagashima Y. Nishio S. Noguchi T. Arakawa O. Kanoh S. Hashimoto K. 1988Dtection of tetrodotoxin by thin-layer chromatography/fast atom bombardment mass spectrometry. Anal.Biochem. 175 258 262.
Shida Y. Arakawa O. Onoue Y. Noguchi T. 1998LC/MS of marine toxin-1. In “Proceedings of the 46th Annual Conference on Mass Spectrometry,”
Shoji Y. Yamashita-Yotsu M. Miyazawa T. Yasumoyto T. 2001 Instrumental Analysis of Tetrodotoxin. Analy. Bioch., 290 10 17.
Tsuruda K. Arakawa O. Kawatsu K. Hamano Y. Takatani T. Noguchi T. 2002 Instrumental Analysis of Tetrodotoxin. 40 131 136.
Tanu M. B. Mahmud Y. Tsuruda K. Arakawa O. Noguchi T. 2001 Instrumental Analysis of Tetrodotoxin. , 39 937 941.
Onoue Y. Noguchi T. Hashimoto K. 1984Tetrodotoxin determination method. In “Seafood Toxin” (E.P.Ragelis, ed.) 345 355. American Chemical Society, Washington, DC.
Saito T. Kanoh S. Noguchi T. Shida Y. Hashimoto K. 1987Occurrence of anhydrotetrodotoxin in several nontoxic species of puffer. In “Progress in venom and toxin research” (P.Gopalakrishnakone and C.K.Tan, ed.) 314 324. National University of Singapore, Singapore.
Tanu M. B. Noguchi T. 1999Tetrodotoxin as a toxic principal in the horseshoe crab Carcinoscorpius rotundicauda collected from Bangladesh. J.Food Hyg. Soc., 40 426 430.
Shimada K. Ohtsuru M. Yamaguchi T. Nigota K. 1983 Instrumental Analysis of Tetrodotoxin. J.Food Sci., 48 665 667.