Indirect Quantitation of Saxitoxin by HPLC with Post-Column Oxidation and Fluorometric Detection

ANALYTICAL SCIENCES SEPTEMBER 2014, VOL. 30 931 2014 © The Japan Society for Analytical Chemistry

Indirect Quantitation of Saxitoxin by HPLC with Post-column Oxidation and Fluorometric Detection

Ryuichi WATANABE,* Tomoko HARADA,* Ryoji MATSUSHIMA,* Hiroshi OIKAWA,** Yasukatsu OSHIMA,*** Masaki KANENIWA,* and Toshiyuki SUZUKI*†

*National Research Institute of Fisheries Science, 2-12-4 Fukuura, Kanazawa, Yokohama 236–8648, Japan **National Research Institute of Fisheries and Environment of Inland Sea, 2-12-4 Maruishi, Hatsukaichi, Hiroshima 739–0452, Japan ***School of Marine Bioscience, Kitasato University, 1-15-1 Kitasato, Minami, Sagamihara 252–0373, Japan

The indirect identification and quantification of saxitoxin (STX) using other STX analogues by high-performance liquid chromatography with post-column oxidation and fluorescent detection (HPLC-FD) was investigated. Decarbamoylsaxitoxin (dcSTX) among the many STX analogues was selected as an external standard to identify and quantify STX. The retention time of STX in shellfish extracts by HPLC-FD was reproducibly estimated by using the retention time of dcSTX and the separation factor (α) between STX and dcSTX. Almost all of the columns tested to setup the method were useful to identify STX. Because a molar fluorescent coefficient of dcSTX was slightly different from that of STX, a factor used to correct the fluorescent coefficient in STX/dcSTX was determined to be 1.30. The indirect quantification of STX in scallop extracts by using the correction factor agreed to 80 – 100% precision with direct quantification using STX as an external standard.

Keywords Saxitoxin, post-column HPLC, paralytic shellfish toxins, decarbamoylsaxitoxin

(Received May 15, 2014; Accepted July 17, 2014; Published September 10, 2014)

high-performance liquid chromatography coupled with Introduction fluorescent detection (HPLC-FD),12,13 has been used in the research field, and the applicability of this method as the official Saxitoxin (STX, Fig. 1) is a natural toxin with potent testing method for STX and other STX analogues in shellfish neurotoxicity to humans by acting on voltage-gated sodium has been demonstrated. Although HPLC-FD methods can channels on nerve cells,1 and can cause death in severe cases. identify and quantify individual toxins by chromatographic STX and other paralytic shellfish toxins (PSTs) are produced by separation on a column, a series of toxin standards are essentially toxic algae, and are accumulated in shellfish. Contaminations required in these methods. Almost all of the PST standards are of shellfish with PSTs have become a worldwide public health available at least in Japan in our governmental research program; problem. To prevent human poisoning from PSTs, shellfish however, STX cannot be produced or provided in Japan due to a have been monitored by the mouse bioassay (MBA) as the Japanese domestic law based on the international treaty in official testing method in many countries. CWC. Various instrumental detection methods for STX and PSTs In the HPLC-FD method reported by Oshima,12 STX can be have been developed as an alternative method to MBA.2–6 A detected together with neosaxitoxin (neoSTX) and drawback in the instrumental methods is insufficient availability decarbamoylsaxitoxin (dcSTX) on the same chromatographic of STX and PSTs standards for the identification and run;12 therefore, either neoSTX or dcSTX can be a candidate as quantification of toxins. Especially STX is listed as chemical an external standard for the analysis of STX on the basis of the warfare agents in schedule 1 in the chemical weapons convention Oshima method. In our present study, we found that the (CWC),7 and it is prohibited to produce, stockpile and utilize retention time of neoSTX was relatively unstable compared to STX for any purpose. Liquid chromatography–tandem mass dcSTX due to the pH dependence in the dissociation of the spectrometry (LC/MS/MS) is one of the most potentially useful N1-hydroxyl group. DcSTX can be simply prepared by methods for the identification and quantification of STX and chemical conversion from C1/C2 from toxic algae,14–17 and it is PSTs.2,6, 8–11 In terms of the analysis of STX, it is noteworthy useful as a calibrant standard owing to its chemical stability.18 that some of the LC/MS/MS methods have quantified STX in For these reasons, we chose dcSTX as an external standard to shellfish extract and human urine samples by using gonyautoxin identify and quantify STX in shellfish samples. In our present 10 15 11 1 and N7- isotopically labeled STX, respectively, as an study, a detailed investigation of the identification and internal standard. Another instrumental method, quantification of STX on HPLC-FD by using a dcSTX external standard is described. † To whom correspondence should be addressed. E-mail: [email protected] 932 ANALYTICAL SCIENCES SEPTEMBER 2014, VOL. 30

Fig. 1 Chemical structure of saxitoxins.

tSTX = t0 + α(tDC – t0). (1) Experimental

Here tSTX is the retention time (min) of STX, t0 the void Reagents and chemicals volume (2.75 min) in the HPLC-FD system, α the separation Ion pair chromatograph-grade ion pair reagents, sodium factor and tDC the retention time (min) of dcSTX. These 1-hepatanesulfonate and tetrabutyl ammonium phosphate, were parameters were determined as follows. A mixture standard purchased from Tokyo Chemical Industry (Tokyo, Japan) and solution containing neoSTX (2.07 μmol/L), dcSTX Sigma-Aldrich Japan (Tokyo, Japan), respectively. Ammonia (0.57 μmol/L) and STX (0.92 μmol/L) was analyzed in triplicate solution, HPLC-grade acetonitrile (MeCN) and methanol under the chromatographic conditions reported by Oshima12 to

(MeOH) were purchased from Kanto Chemical (Tokyo, Japan). determine the retention time of each toxin. The void volume (t0) Other reagents (ortho-periodic acid, dipotassium hydrogen of the HPLC-FD system was determined by using another PST phosphate and potassium hydroxide) were purchased from standard, C1/C2, which is not retained on the column under this Wako Pure Chemical (Osaka, Japan). C1/C2 and neoSTX used chromatographic condition. were prepared under the shellfish monitoring standard program The separation factor (α) between STX and dcSTX was in Japan;19 dcSTX and STX were in-house standards prepared at calculated using Tohoku University by the former Prof. Oshima’s laboratory.

Separation factor (αSTX/DC) = [(tSTX – t0)/t0]/[(tDC – t0)/t0], (2) HPLC-FD 12 HPLC-FD followed by the method reported by Oshima was where DC is dcSTX, αSTX/DC the separation factor between STX performed by using Acquity UPLC H-class (Waters, Milford, and dcSTX. MA) equipped with a quaternary solvent manager pump, an autosampler, a column heater (30°C) and a fluorometric detector. Indirect quantitation of STX using dcSTX as the STX alternatives A post-column reaction system was comprised of a double The molar fluorescence coefficient of STX in HPLC analysis plunger pump (SPD-2502 U, Nihon Seimitsu Kagaku, Tokyo, slightly differs from that of dcSTX. STX can be indirectly Japan) for the oxidizing reagent and acidifier, and of a dry quantitated by using the dcSTX calibrant if the ratio of the reaction bath (DB-5, Shimamura Tech, Tokyo, Japan) equipped molar fluorescence coefficient between STX and dcSTX is with a 10-m teflon tube (0.25 mm inner diameter and 1/16 inch determined. A mixture of the STX and dcSTX standard solution outer diameter). Separation was achieved on a Mightysil RP-8 was analyzed in triplicate under a previously reported condition12 GP (150 × 4.6 mm i.d., 5 μm, Kanto Chemical) connected with to investigate the relative molar coefficient ratio between STX a guard column (5 × 4.6 mm i.d., 5 μm, Kanto Chemical). The and dcSTX. From the data obtained in our present study, a mobile phase consisted of a 30 mM ammonium phosphate correction factor, 1.30, was determined by the ratio between buffer containing 2 mM sodium 1-heptanesulfonate at pH STX/dcSTX. Therefore, the concentration of STX in a sample 7.1/MeCN (96/4, v/v). Five microliters of each analyte and can be quantified by using the dcSTX standard solution by the reference material were injected into HPLC-FD. Other following equation: analytical conditions were in accordance with those reported by 12 Oshima. Prior to sample injections, the injection needle was CSTX = (PASTX × CDC × VSTD)/(PADC × m × VSAM), (3) washed using MeOH/distilled water (10/90, v/v). Data was TM acquired with Empower2 . where CSTX is the concentration of STX in a sample, PASTX the peak area of STX in the sample, DC is dcSTX, CDC the Estimation of the retention time of STX calculated from that of concentration of dcSTX used as a calibrant standard, VSTD the dcSTX injection volume (μL) of a standard solution, PADC the peak area The retention time of STX in a chromatogram can be estimated of dcSTX as a calibrant standard, m a correction factor (1.30) by the following equation using the retention time of dcSTX and VSAM the injection volume (μL) of a sample. and the separation factor, which is also defined as ratio of capacity factor (k′) between two analytes: ANALYTICAL SCIENCES SEPTEMBER 2014, VOL. 30 933

Fig. 2 Chromatograms of STX standards (A) and scallop extract (B) under the analytical condition of STX.

Table 1 Capacity ratio (k′) of STX, neoSTX, and dcSTX under Extraction of toxins form scallops various chromatographic conditions Toxins were extracted from scallops Patinopecten yessoensis Mobile phase and column temperaturea fed on toxic dinoflagellate Alexandrium tamarense according to the AOAC method 959.08.20 The extracts were passed through Parameter neoSTX dcSTX STX a Sep-pak C18 plus cartridge (Waters) and ultracentrifuged as 12 Phosphate buffer in the mobile phase reported in a previous paper. 10 mM 4.97 10.35 11.44 20 mM 2.79 5.06 5.71 30 mMb 1.85 3.15 3.60 Results and Discussion pH of the mobile phase 6.5 4.15 5.32 5.96 Peak identification of STX by the STX alternatives 7.1b 2.03 3.43 3.93 Figure 2A shows a typical chromatogram of a mixture of a 7.5 1.25 2.71 3.15 STX, dcSTX, and neoSTX standard solution obtained on a Acetonitrile in the mobile phase 3% v/v 2.78 4.80 5.53 Mightysil RP-8 column with LC conditions reported by 4% v/vb 1.99 3.39 3.88 12 Oshima. STX and its two analogues were eluted from the 5% v/v 1.48 2.50 2.85 column in order of neoSTX, dcSTX, and STX, as reported in a Column temperature previous study.12 In our present study, a mixture of STX, 20°C 2.92 4.40 5.32 dcSTX, and neoSTX standard solution was analyzed under 30°Cb 2.31 3.95 4.54 various chromatographic conditions differing mobile-phase 40°C 1.55 2.92 3.22 compositions, pH and column temperatures to determine the a. Average values in triplicate. ratio of the capacity factor (k′) between STX and other STX b. Parameters those reported for analyses of STX, neoSTX, and dcSTX analogues. Table 1 gives the k′ values of STX analogues by Oshima.12 obtained under different chromatographic conditions. Table 2 lists the separation factor (α) between STX and other STX analogues obtained on simultaneous chromatographic runs under different conditions. The α was calculated from the ratio 1.29 ± 0.26 (12.9% RSD). The α between STX and neoSTX of k′ listed in Table 1. The α between STX and dcSTX obtained greatly increased with an increase of the pH. Because the under different chromatographic conditions was 1.14 ± 0.03 dissociation of the N1-hydroxyl group on the neoSTX at higher (2.4% RSD), whereas the α between STX and neoSTX was pH conditions resulted in a weaker retention of the neoSTX 934 ANALYTICAL SCIENCES SEPTEMBER 2014, VOL. 30 paired ion complex on the reversed phase column, in contrast conditions compared with those obtained between STX and the pH did not relatively affect the retention of STX, the α neoSTX. This indicates that dcSTX is a more suitable compound between neoSTX and STX increased at high pH ranges. than neoSTX to identify STX from their α values. Therefore, Although the column temperatures significantly affected the α dcSTX was chosen as an external standard to calculate the between STX and dcSTX among the four parameters, the α retention time by Eq. (1). values were relatively constant under different chromatographic Although Eq. (1) is useful to calculate the retention time of STX from that of dcSTX, the tolerance of the α values is of practical importance for applying our present research to analyses of STX in other laboratories. The minimum α value Table 2 Separation factors (α) between STX and other two STX analogues obtained under various chromatographic obtained in our present study was 1.10, as shown in Table 2. conditions This value was also the minimum required value in terms of chromatographic separation between dcSTX and STX. Mobile phase and column temperaturea Therefore, 1.10 was set as the minimum α value for calculating the retention time of STX. Although the maximum α value was Parameter neoSTX dcSTX 1.21, the maximum tolerance α value was set at 1.35 by Phosphate buffer in the mobile phase assuming the application of various HPLC columns. Therefore, 10 mM 2.30 1.11 it was concluded that the α values with tolerance of between 20 mM 2.05 1.13 1.10 and 1.35 with our fundamental chromatographic conditions 30 mMb 1.94 1.14 (30 mM phosphate buffer containing 2 mM sodium pH of the mobile phase 1-heptanesulfonate and 4% MeCN, pH 7.1, column temperature 6.5 1.43 1.12 7.1b 1.94 1.15 30°C) ensure a robust identification of STX. It is also noteworthy 7.5 2.51 1.16 that the α values with tolerance values of between 1.10 and 1.35 Acetonitrile in the mobile phase allow us to use the various chromatographic conditions listed in 3% v/v 1.99 1.15 Table 2 for identification of STX. 4% v/vb 1.95 1.14 To confirm the applicability of our method, the α values 5% v/v 1.92 1.14 between STX and dcSTX were obtained on four other different Column temperature columns (Mightysil RP-18 GP II, Xbridge BEH C8, Unison 20 C 1.83 1.21 ° UK-C8, Zorbax Bonus-RP) under chromatographic conditions. 30°Cb 1.97 1.15 40°C 2.08 1.10 The conditions with 30 mM phosphate buffer containing 2 mM sodium 1-heptanesulfonate and 2 – 4% MeCN, pH 7.1, column Average 1.29 1.14 temperature 30°C were used to obtain α. The α values obtained S.D. 0.26 0.03 on a Mightysil RP-18 GP II, an Xbridge BEH C8, a Unison R.S.D., % 12.9 2.4 UK-C8 and a Zorbax Bonus-RP with the chromatographic Maximum 2.51 1.21 conditions were 1.17, 1.15, 1.22 and 1.32, respectively. These Minimum 1.43 1.10 values were within the tolerance values in the identification of a. Average values in triplicate. STX on a Mightysil RP-8 GP. These results suggest that our b. Parameters those reported for analyses of STX, neoSTX, and dcSTX method using α between STX and dcSTX with tolerance values by Oshima.12

Table 3 Molar fluorescent coefficient ratio between STX and other two STX analogues under various chromatographic conditions

Mobile phase and column temperaturea Post-column reactiona

Parameter neoSTX dcSTX Parameter neoSTX dcSTX

Phosphate buffer in the mobile phase Reaction temperature 10 mM 5.87 1.72 80°C 3.24 1.27 20 mM 3.71 1.49 85°Cb 2.93 1.27 30 mMb 2.89 1.34 90°C 2.62 1.26 pH of the mobile phase pH of the oxidizing reagent 6.5 1.83 1.46 8.5 2.04 1.24 7.1b 2.90 1.27 9.0b 2.93 1.27 7.5 3.38 1.34 9.5 2.29 1.22 Acetonitrile in the mobile phase Phosphate buffer in the oxidizing reagent 3% v/v 3.00 1.32 45 mM 2.04 1.24 4% v/vb 3.07 1.30 50 mMb 2.93 1.27 5% v/v 3.01 1.30 55 mM 2.29 1.22 Column temperature Ortho-periodic acid in the oxidizing reagent 20°C 2.95 1.39 6.3 mM 1.97 1.21 30°Cb 2.88 1.41 7.0 mMb 2.93 1.27 40°C 3.01 1.37 7.7 mM 2.39 1.26

Average 3.21 1.39 Average 2.55 1.25 S.D. 0.94 0.12 S.D. 0.43 0.02 R.S.D., % 29.4 8.8 R.S.D., % 17.0 1.8

a. Average values in triplicate. b. Parameters in a normal analytical condition. ANALYTICAL SCIENCES SEPTEMBER 2014, VOL. 30 935

Fig. 3 Linearity of saxitoxins.

Table 4 Information on scallop extract used in the study of between 1.10 and 1.35 could be widely applicable on several Concentration Mouse toxicity reversed-phase HPLC columns. Toxin μmol/L ng/mL MU/mLa μg STX equiv./mLb, c Indirect quantitation of STX using dcSTX of the STX alternatives Because the ratio of the molar fluorescence response is neoSTX 1.59 501 3.65 0.635 dcSTX 0.05 79 0.06 0.011 applicable as a correction factor for the indirect quantification of STX 1.49 446 3.70 0.643 STX by using other STX analogues as a calibrant, a mixture of STX, neoSTX and dcSTX standard solutions was analyzed a. Specific toxicity12: neoSTX; 2295 MU/μmol, dcSTX; under modified analytical conditions of the mobile-phase 1247 MU/μmol, STX; 2483 MU/μmol. composition and post-column reaction system. The results are b. Toxicity equivalent factors (TEF)9: neoSTX; 0.924, dcSTX; 0.513, summarized in Table 3. The correction factors of STX/dcSTX STX; 1.000. 12 obtained under different post-column reaction and c. Mouse toxicity (μg STX equiv./mL) : Conc. (μM) × 0.3722 × TEF × 1.16. chromatographic conditions were 1.25 ± 0.02 (1.8% RSD) and 1.39 ± 0.12 (8.8% RSD), respectively. These values were clearly lower that those obtained with neoSTX. Because the large differences in the correction factors obtained between various reaction and chromatographic conditions were not external standard was 11.81 under this HPLC condition. observed, these values were averaged. The averaged correction Therefore, the calculated retention time of STX by the retention factor under our fundamental chromatographic conditions was time of dcSTX and α with the minimum and maximum tolerance 1.32 ± 0.05 (4.0% RSD), and the factor was set to 1.30. values on the Eq. (1) was 12.72 and 14.98 min, respectively. The linearity of each toxin standard solution differing by the The retention time of STX (13.14 min) observed in the scallops concentrations is shown in Fig. 3. Each toxin showed a extract was within the range of the calculated retention time concentration-dependent linearity. Since the slopes in dcSTX (12.72 – 14.98 min). On the other hand, the STX contents in and STX were 54 and 69, respectively, the ratio of the slope the scallop extracts obtained by indirect quantification with (STX/dcSTX) was 1.27, which was close to the above correction dcSTX in this HPLC condition was 1.43, which was 96.0% of factor of 1.30, indicating that 1.30 was a proper value as the that obtained by the direct quantification of STX in the sample. correction factor. A good correlation on the retention time of STX and quantification results between direct and our indirect method Indirect quantitation and peak identification of STX in scallop was obtained on four other columns (Mightysil RP-18 GPII, extract Xbridge BEH C8, Unison UK-C8, Zorbax Bonus-RP), as shown The contents of STX analogues and the mouse toxicities in Table 5. These results suggest that our indirect identification determined by previously reported HPLC-FD12 are given in and quantification method for STX with an external dcSTX Table 4. The STX contents in the scallops were 3.7 MU/mL standard is widely applicable to bivalve samples contaminated shellfish extract and 0.64 μg STX/mL shellfish extract, with STX on various HPLC columns. respectively. These values were about 2-fold those of the regulatory levels of 4 MU/g (2 MU/mL shellfish extract) in Japan and the internationally accepted regulatory level Conclusions (0.8 mg/kg), respectively. Table 5 gives chromatographic parameters on the identification and quantification of STX in We developed an indirect analytical method to identify and the scallops. Figure 2 shows the HPLC-FD chromatogram quantify STX by using an external dcSTX standard on the obtained from a mixture of standard toxins (A) and from the HPLC-FD analysis. Chromatographic parameters including the scallop extract (B) on a Mightysil RP-8 GP. The retention time retention time of dcSTX and separation factor (α) between STX of STX in the scallop extract identified by an external standard and dcSTX with tolerance between 1.10 and 1.35 obtained on a of STX on the Mightysil RP-8 GP was 13.14 min (Table 5). On Mightysil RP-8 GP column ensured a robust identification of the other hand, the retention time of dcSTX obtained by an STX in bivalves under various mobile-phase conditions. The 936 ANALYTICAL SCIENCES SEPTEMBER 2014, VOL. 30

Table 5 Separation factors of STX to dcSTX and indirect quantitation of STX from scallop extract in different analytical columns Expected time width/min a Raw data/min Concentration (μmol/L) Analytical column b tSTX of STX in sample A tdcSTX t1 (α: 1.10) t2 (α: 1.35)

Mightysil RP-8 GP 13.14 11.81 12.72 14.98 1.43 (96.0) Mightysil RP-18 GP II 13.94 12.27 13.22 15.60 1.46 (98.0) Xbridge BEH C8 9.15 8.30 8.86 10.25 1.20 (80.5) Zorbax Bonus RPc 10.08 8.29 8.84 10.23 1.31 (87.9) Unison UK-C8d 8.02 7.03 7.46 8.53 1.22 (81.8)

a. Information on analytical columns tested was as follows: Mightysil RP-8 GP (150 × 4.6 mm i.d., 5 μm, Kanto Chemical), Mightysil RP-18 GP II (150 × 4.6 mm i.d., 5 μm, Kanto Chemical), Xbridge BEH C8 (150 × 4.6 mm i.d., 5 μm, Waters), Zorbax Bonus RP (150 × 4.6 mm i.d., 3.5 μm, Agilent Technologies, Santa Clara, CA), and Unison UK-C8 (150 × 4.6 mm i.d., 3 μm, Imtakt, Kyoto, Japan). b. Scallop extract by AOAC 959.08., values in parenthesis shows percentile (%) to STX by direct quantitation. c. STXs were analyzed using the mobile phase containing 2% (v/v) MeCN. d. STXs were analyzed using the mobile phase containing 3% (v/v) MeCN.

applicability of our method was demonstrated on four other Chromatography Post-Column Oxidation (PCOX) Method, reversed phase HPLC columns (Mightysil RP-18 GP II, Xbridge First action 2011”, Official method of Analysis, 19th ed., BEH C8, Unison UK-C8, Zorbax Bonus-RP) with our 2012, Chap. 49.10.05, Arlington, Virginia, 105. fundamental chromatographic conditions (30 mM phosphate 5. AOAC Official method 2011.27, “Paralytic shellfish toxins buffer containing 2 mM sodium 1-heptanesulfonate and 2 – 4% (PSTs) in Shellfish, Receptor Binding Assay First action MeCN, pH 7.1, column temperature 30°C). Therefore, robust 2011”, Official method of Analysis, 19th ed., 2012, Chap. identification of STX without the STX standard could be 49.10.06, Arlington, Virginia, 116. possible when more than two different HPLC columns, 6. M. Diener, K. Erler, B. Christian, and B. Luckas, J. Sep. described in the present study, are used. Due to the applicability Sci., 2007, 30, 1821. of our method under several different chromatographic 7. http://www.opcw.org/chemical-weapons-convention/annexes/ conditions, it was supposed that our method is useful on wide annex-on-chemicals/schedule-1/. varieties of reversed-phase HPLC columns. However, we 8. C. Dell’Aversano, G. Eaglesham, and M. A. Quilliam, J. strongly recommend that the applicability of our method for the Chromatogr. A, 2004, 1028, 155. identification of STX on other HPLC columns not described in 9. R. Watanabe, R. Matsushima, T. Harada, H. Oikawa, M. our present study should be demonstrated by confirming the Murata, and T. Suzuki, Food Addit. Contam., Part A, 2013, retention time of STX obtained by the STX standard. We plan 30, 1351. to investigate the applicability of our method on several other 10. M. Halme, M.-L. Rapinoja, M. Karjalainen, and P. reversed-phase HPLC columns. These results will be reported Vannienen, J. Chromatogr. B, 2012, 880, 50. elsewhere. 11. R. C. Johnson, Y. Zhou, K. Statler, J. Thomas, F. Cox, S. Hall, and J. R. Barr, J. Anal. Toxicol., 2009, 33, 8. 12. Y. Oshima, J. AOAC Int., 1995, 78, 528. Acknowledgements 13. J. M. van de Riet, R. S. Gibbs, F. W. Chou, P. M. Muggah, W. A. Rouke, G. Burns, K. Thomas, and M. A. Quilliam, J. This study was financially supported by Ministry of Economy, AOAC Int., 2009, 6, 1690. Trade and Industry. We would like to gratefully acknowledge 14. R. Watanabe, T. Suzuki, and Y. Oshima, Mar. Drugs, 2011, Dr. J. Hungerford, Food and Drug Administration, USA, for the 9, 466. assistance with English corrections on this draft paper. 15. Y. Shimizu and M. Yoshioka, Science, 1981, 212, 547. 16. S. Sato, R. Sakai, and M. Kodama, Bioorg. Med. Chem., 2000, 10, 1787. References 17. M. V. Laycock, P. Thibault, S. W. Ayer, and J. A. Walter, Nat. Toxins, 1994, 2, 175. 1. W. A. Catterall, S. Cestèle, V. Yarov-Yarovoy, F. H. Yu, K. 18. H. P. van Egmond, A. Mouriño, P. A. Burdaspal, and A. Konoki, and T. Scheuer, Toxicon, 2007, 49, 124. Boenke, J. AOAC Int., 2001, 84, 1668. 2. C. Dell’Aversano, P. Hess, and M. A. Quilliam, J. 19. H. Goto, T. Igarashi, R, Sekiguchi, K. Tanno, M, Satake, Y. Chromatogr. A, 2005, 1081, 190. Oshima, and T. Yasumoto, in “Harmful algae. Xunta de 3. AOAC Official method 2005.06, “Paralytic shellfish Galicia and Intergovernmental Oceanographic Commission Poisoning toxins in Shellfish, prechromatographic oxidation of UNESCO”, eds. B. Reguera, J. Blanco, M. L. Fernández, and liquid chromatography with fluorescence detection, and T. Wyatt, 1998, Spain, 216. First action 2005”, Official method of Analysis, 19th ed., 20. AOAC Official method 959.08, “Paralytic shellfish Poison, 2012, Chap. 49.10.03, Arlington, Virginia, 89. Biological Method, First Action 1959 Final Action”, 4. AOAC Official method 2011.02, “Paralytic shellfish toxins Official method of Analysis, 19th ed., 2012, Chap. 49.10.01, in Mussels, Clams, Oysters and Scallops, Liquid Arlington, Virginia, 86.