- Email: [email protected]
Ciguatera incidence and fish toxicity in Okinawa, Japan
Toxicon 56 (2010) 656–661
Contents lists available at ScienceDirect
Toxicon journal homepage: www.elsevier.com/locate/toxicon
Ciguatera incidence and fish toxicity in Okinawa, Japan Naomasa Oshiro a, b, *, Kentaro Yogi a, c, Shuko Asato a, c, Toshiki Sasaki d,1, Koji Tamanaha a, Masahiro Hirama c, e, Takeshi Yasumoto f, Yasuo Inafuku a a
Okinawa Prefectural Institute of Health and Environment, 2085 Aza-Ozato, Ozato, Nanjo, Okinawa 901-1202, Japan Tokyo University of Marine Science and Technology, 4-5-7 Konan, Minato-ku, Tokyo 108-8477, Japan c SORST, Japan Science and Technology Agency, Japan d Minamata Environmental Research and Develop Center Co., Ltd., 5-98 Hamamatsu, Minamata, Kumamoto 867-0068, Japan e Department of Chemistry, Graduate School of Science, Tohoku University, 6-3 Aramaki, Aoba-ku, Sendai, Miyagi 980-8578, Japan f Okinawa Science and Technology Promotion Center, 12-75 Suzaki, Uruma, Okinawa 904-2234, Japan b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 January 2009 Received in revised form 19 May 2009 Accepted 19 May 2009 Available online 9 June 2009
Okinawa being located in the subtropical region has the highest incidence of ciguatera in Japan. Officially, 33 outbreaks involving 103 patients have been reported between 1997 and 2006. The implicated species were Variola louti, Lutjanus bohar, Lutjanus monostigma, Epinephelus fuscoguttatus, unidentified Lutjanus sp., Plectropomus areolatus, Oplegnathus punctatus, Epinephelus polyphekadion, Caranx ignobilis and moray eel. Toxicities of the leftover meals, as determined by mouse bioassays, ranged from 0.025 to 0.8 MU/g or above (equivalent to 0.175–5.6 ng CTX1B/g). We collected 612 specimens of fish belonging to L. monostigma, L. bohar, Lutjanus argentimaculatus, Lutjanus russellii, V. louti, Variola albimarginata, and E. fuscoguttatus from the coasts around Okinawa and examined the toxicity of the flesh by the mouse bioassay. The rate of toxic fish was as follows: L. monostigma: 32.3%, L. bohar: 11.9%, V. louti: 14.3%, E. fuscoguttatus: 20.8%. Only one out of 36 samples of V. albimarginata and two of 74 samples of L. russellii were found toxic. None of the 35 samples of L. argentimaculatus was toxic. Nor the L. bohar samples weighing less than 4 kg were toxic. In all toxic samples, CTX1B was detected by LC/MS analysis but CTX3C and 51hydroxyCTX3C were not. Ó 2009 Elsevier Ltd. All rights reserved.
Keywords: Ciguatera Incidence Fish Toxicity Ciguatoxin Okinawa Mouse bioassay
1. Introduction Ciguatera fish poisoning (CFP) associated with gastrointestinal, cardiovascular and neurological symptoms and signs is one of the largest scale food poisonings of nonmicrobial origins annually affecting 20,000–60,000 people worldwide (Lehane and Lewis, 2000; Yasumoto, 2005). While ciguatera endemic areas are tropical and subtropical
* Corresponding author at: Okinawa Prefectural Institute of Health and Environment, 2085 Aza-Ozato, Ozato, Nanjo, Okinawa 901-1202, Japan. Tel.: þ81 98 945 0329; fax: þ81 98 945 9366. E-mail address: [email protected] (N. Oshiro). 1 Present address: Minamata Research Center, Chisso Corporation. 1-1 Noguchi-cho, Minamata, Kumamoto 867-8501, Japan. 0041-0101/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.toxicon.2009.05.036
Indo-Pacific Ocean and Caribbean Sea where coral reef developed, increasing world trade of seafood and international tourisms cause outbreaks in other climate (Lehane and Lewis, 2000; Yasumoto, 2005; Wong et al., 2005). Original forms of principal toxins, ciguatoxins (CTXs), are produced by an epiphytic dinoflagellate Gambierdiscus toxicus and transferred to herbivorous and carnivorous fish via food chain (Yasumoto et al., 1977; Yasumoto, 2005). The chemical structures of ciguatoxins in G. toxicus and in fish have been elucidated (Yasumoto, 2001, 2005). Ciguatoxin4A (CTX4A) and ciguatoxin-3C (CTX3C) produced by the dinoflagellate were presumed to undergo structural changes in fish to produce the most representative toxin, ciguatoxin-1B (CTX1B), 51-hydroxyCTX3C and other metabolite toxins (Yasumoto, 2001, 2005). Occurrence of
N. Oshiro et al. / Toxicon 56 (2010) 656–661
CFP is rather rare in Japan, and Okinawa is the only area where sporadic but regular occurrence of CFP is reported. Okinawa is the largest of the Ryukyu archipelago that consists of 48 islands lying in the south-westernmost end of Japan. Its climate is subtropical and CFP has been recognized as endemic in a previous survey (Hashimoto et al., 1969a,b). In this paper, we summarized CFP outbreaks occurred in Okinawa, analyzed fish toxicity, and demonstrated the presence of CTX1B. 2. Materials and methods 2.1. Epidemiological data CFP incidents officially reported by medical doctors to the Department of Health and Welfare, Okinawa Prefectural Government, between 1997 and 2006 are used for epidemiological analysis in this paper. The toxic levels of implicated fish were determined by the official mouse bioassay method (MBA, Satake, 2005), whenever the leftover food and/or remnant fish were available. 2.2. Chemicals Normal saline solution (0.9% NaCl) was purchased from Otsuka Pharmaceuticals Co., Ltd.(Tokyo). The HPLC grade methanol, distilled water, and formic acid obtained from Wako Pure Chemical Industries, Ltd.(Tokyo), were used for the mobile phases in LC/MS analysis. Standard toxins, CTX1B, CTX3C and 51-hydroxyCTX3C for LC/MS were synthesized at Tohoku University (Hirama et al., 2001; Inoue et al., 2006). Other chemicals used were of analytical grade, otherwise stated. The composition of a solvent mixture was expressed in v/v. 2.3. Fish samples Frozen fish specimens of Lutjanus monostigma, Lutjanus bohar, Lutjanus argentimaculatus, Lutjanus russellii, Variola louti, Variola albimarginata, and Epinephelus fuscoguttatus, collected from the Okinawan coasts were purchased from the Fishermen’s Unions and stored at 20 C until use. The fish were thawed, photographed, measured for standard length and weight, dissected into flesh, liver, and other viscera, and frozen again at 20 C until use. Professor Tetsuo Yoshino, University of the Ryukyus, kindly conducted identification of the fish species. 2.4. Extraction Fish samples were extracted following the standard MBA method for ciguatoxin detection. A flesh sample (120 g) was thawed and homogenized with 350 ml of acetone twice. The combined acetone extract was evaporated to produce an aqueous concentrate, which was further extracted with 100 ml of diethyl ether twice. The ether extract was evaporated completely and the residue was partitioned between hexane (50 ml) and methanol– water (9:1, 25 ml). The aqueous methanol layer was completely freed of methanol and the residue was used for
657
the mouse bioassay. In all experiments, evaporation of a solution was carried out under reduced pressure. 2.5. Mouse bioassay (MBA) The standard MBA method was slightly modified to reduce the number of mice for use. The residue from an extract equivalent to 120 g of the flesh was suspended in 3 ml of 1% Tween 60 in normal saline solution. Two 1-ml portions of the suspension were injected respectively into a male mouse of ddY strain weighing 17–20 g by intraperitoneal route (Kyudo co., Ltd., Kumamoto, Japan). The mice were observed for 24 h. When the two mice died or survived, the sample was regarded as toxic or nontoxic, respectively. If one was dead and the other survived, remaining 1 ml of the sample suspension was injected into another mouse. The toxicity of the sample was judged by the survival or death of the third mouse. The flesh samples judged to be toxic by MBA were extracted similarly. Serial dilutions of the extracts were prepared and used for further confirmation of the toxicity levels by MBA. For example, a suspension equivalent to 20 g flesh/ml was prepared and 1 ml and 0.5 ml portions were injected, then the suspension was diluted four times with Tween 60 solution and used as same manner. 2.6. Solid phase extraction (SPE) and LC/MS analysis An extract prepared from 20 g of flesh was dissolved in 20 ml of methanol–water (7:3) containing 1% ammonium hydroxide. The solution was applied to an OASIS HLB cartridge (6 cc, 200 mg, Waters, Massachusetts) preconditioned with methanol followed by distilled water. The toxins retained on the cartridge were eluted first with 3 ml of methanol–water (8:2) containing 1% ammonium hydroxide (Fr. A) and next with 2 ml of acetonitrile–water (7:3) (Fr. B). Fractions A and B were dried and dissolved in 0.5 ml of methanol. A 0.1 ml portion of the each solution was mixed with 0.9 ml of chloroform and passed through a silica gel cartridge (InertSep Si, 100 mg, GL Sciences INC., Tokyo) preconditioned with the same solvent. The eluate was dried, dissolved in methanol (0.1 ml), and the designated test solution was used for LC/MS analysis. The cleanup procedure is presented in Fig. 1. An Agilent 1100 LC/MSD SL system (Agilent Technologies, California) was used for LC/MS analysis. HPLC separations were performed on a Cadenza CD-C18 column (3 mm, 2 150 mm, Imtakt co., Kyoto, Japan) using 0.1% formic acid–methanol (15:85) as the mobile phase. The column temperature and flow rate were kept at 40 C and 0.2 ml/min, respectively. MS parameters were set as follows: Ionization; API-ES, ion mode; positive, fragmenter voltage; 350 V, nebulizer gas; N2 20 psi., capillary voltage; 4000 V, Dry gas; 10 l/min of N2 at 350 C. A 5-ml portion of the test solution was injected into the LC/MS system and positive ions of m/z 1133.5 (0–7.5 min, [CTX1B þ Na]þ), m/z 1061.5 (7.5–15 min., [51-hydroxyCTX3C þ Na]þ), and m/z 1045.5 (15–35 min, [CTX3C þ Na]þ) were monitored to detect respective toxins. The synthetic standards of CTX1B, 51-hydroxyCTX3C and CTX3C used were dissolved in methanol (1 ng/ml). The signal/noise (S/N) values for the
658
N. Oshiro et al. / Toxicon 56 (2010) 656–661 Table 1 CFP incidents officially reported in Okinawa from 1997 to 2006.
flesh (20 g) extract dissolved in 20 ml of methanol/water (7:3) containing 1% ammonium hydroxide OASIS HLB (6cc, 200mg)* 3 ml of methanol/water (8:2) containing 1% ammonium hydroxide ---------------- Fr. A 2 ml of acetonitrile/water (7:3) ---------- Fr. B
Fr. B
Fr. A
Fr. A or B dried dissolved in 0.5 ml of methanol 0.1 ml of the solution 0.9 ml of chloroform InertSep Si (100mg)** 0.5 ml of chloroform/methanol (9:1) non absorbed portion dried dissolved in 0.1 ml of methanol LC/MS analysis Fig. 1. SPE of flesh extract for LC/MS analysis. *Preconditioned with methanol (2 ml) followed by distilled water (2 ml). **Preconditioned with chloroform/methanol (9:1, 5 ml).
toxins in the above order were 3.2, 20.1 and 4.0, respectively. 3. Results and discussion 3.1. Epidemiology Although 33 outbreaks have been reported during the ten years from 1997 to 2006 in Okinawa (Table 1), rumors among fishermen suggest many cases remain unreported. The seasonality of the incidents seems thin, though the highest number (6) was recorded in May (Table 1). As listed in Table 2, the most frequently implicated fish were V. louti and L. bohar (8 incidents for each), followed by L. monostigma (6), E. fuscoguttatus (2), unidentified Lutjanus sp.(2), Plectropomus areolatus (1), Oplegnathus punctatus (1), Epinephelus polyphekadion (1), Caranx ignobilis (1) and moray eel (1). Importantly, leftover foods and/or uncooked fish flesh could be obtained for analysis in 12 incidents. Toxicities of these samples estimated by MBA ranged from 0.025 to above 0.8 MU/g (Table 3). The minimum toxicity to
No.
Date
Number of patients
Number of people consumed
Fish species
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
18/Jan/97 21/Sep/97 17/May/98 17/Jun/98 25/Feb/99 26/Feb/99 08/May/99 24/May/99 03/Aug/99 31/Aug/99 21/Sep/99 18/Oct/99 16/Jan/00 20/May/00 24/Jun/00 19/May/01 31/Aug/01 21/Jun/02 04/Nov/02 30/Nov/02 01/Apr/03 03/Jul/03 22/Jul/03 06/Oct/03 06/Mar/04 18/Nov/04 06/Jul/05 10/Sep/05 24/Jan/06 18/Mar/06 26/May/06 29/Jun/06 31/Dec/06
6 3 3 3 1 1 1 2 1 3 6 1 7 5 4 3 3 2 2 3 2 4 3 4 2 3 7 6 2 2 1 4 3
9 5 7 5 1 2 7 6 4 3 6 11 7 5 26 3 3 5 2 3 5 4 3 4 3 4 9 6 2 2 27 4 8
L. monostigma Lutjanus sp. L. bohar V. louti E. fuscoguttatus E. fuscoguttatus L. bohar L. bohar L. monostigma C. ignobilis Unidentified L. bohar E. polyphekadion L. bohar Snapper Unidentified L. monostigma L. bohar P. areolatus L. monostigma O. punctatus L. monostigma L. monostigma V. louti Moray eel V. louti V. louti V. louti V. louti L. bohar L. bohar V. louti V. louti
induce illness by oral intake was calculated from the data obtained on fish (0.1 MU/g) implicated in the outbreak No. 28 (Oshiro and Tamanaha, 2006). All the six persons who ate the fish were intoxicated. The minimum amount consumed by the patients was estimated to be 140 g. Thus, 14 MU of the toxin was enough to induce illness in an adult. Based on the i.p. lethality of CTX1B in mouse (0.35 mg/kg, Yasumoto, 2001), 1 MU could be regarded as equivalent to 7 ng of CTX1B. Hence, 14 MU is equivalent to 98 ng CTX1B. This value is close to the estimation (10 MU ¼ 70 ng CTX1B) made on fish in French Polynesia (Yasumoto, 2005). On the other hand, the fish No. 31 had the same toxicity level as of Table 2 Frequency of causative species in the outbreaks in Okinawa. Species Variola louti Lutjanus bohar L. monostigma Epinephelus fuscoguttatus Lutjanidae Plectropomus areolatus Oplegnathus punctatus E. polyphekadion Caranx ignobilis Moray eel Unidentified Total
Number of outbreaks 8 8 6 2 2 1 1 1 1 1 2 33
Ratio (%) 24 24 18 6 6 3 3 3 3 3 6
N. Oshiro et al. / Toxicon 56 (2010) 656–661 Table 3 Toxicity (in mice unit per gram of fish flesh, MU/g) of the leftover meals and/or remnant fish from some of CFP incidents reported in Table 1. Fish species
Test sample
Toxicity (MU/g)
2 4 13
Lutjanus sp. V. louti E. polyphekadion
17 20 22
L. monostigma L. monostigma L. monostigma
23 24
L. monostigma V. louti
26
V. louti V. louti V. louti L. bohar V. louti
Cooked flesh Raw flesh Cooked flesh Soupa Cooked flesh Cooked flesh Raw flesh Mixed soupb Mixed soupb Raw flesh Mixed soupb Fleshc Fleshc Raw flesh Cooked flesh Raw flesh
0.29 0.1 0.05 <0.025 >0.2 >0.8 >0.2 0.025 >0.2 0.4 0.1 >0.2 0.1 0.1 0.1 0.05
28 31 32 a b c
500
standard length (mm)
No.
nontoxic toxic
400
300
200
100
0 0.0
Toxica
L. monostigma E. fuscoguttatus V. louti L. bohar V. albimarginata L. russellii L. argentimaculatus
226 24 49 168 36 74 35
73 5 7 20 1 2 0
Total
612
108
a
1.5
2.0
2.5
900
Ratio of toxic specimen (%) 32.3 20.8 14.3 11.9 2.8 2.7 0.0
standard length (mm)
Table 4 Frequency of toxic specimens in representative fish species.
Examined
1.0
Fig. 3. Size dependency of toxic specimens of L. monostigma.
nontoxic toxic
800
Number of specimens
0.5
weight (kg)
Assay was made after removing flesh and bones contained in the soup. Assay was made after removing bones contained in the soup. The flesh had been lightly washed with hot water.
Species
659
700 600 500 400 300 200 100 0
0
5
10
15
20
25
weight (kg) Fig. 4. Size dependency of toxic specimens of E. fuscoguttatus.
Toxicity was evaluated by MBA. Only the flesh was used for assay.
No. 28 (0.1 MU/g) but caused intoxication only in one person. Because the fish was shared by 27 persons, the amount consumed by one person could be smaller than in the incident No. 28. Interestingly, the only person intoxicated had the jaw muscle (Oshiro and Tamanaha, 2007), supporting the reputation among fishermen that head was
more toxic than the body muscle. The lowest mousetoxicity of flesh to cause intoxication was suggested to be 0.05 MU/g (No. 13 and 32). Furthermore, involvement of the fish soup with toxicity as low as 0.025 MU/g or lower (No. 13 and 22) indicates the potential danger due to a meal of such a low toxicity. Thus, the maximum allowance level
600
800
standard length (mm)
700
standard length (mm)
nontoxic toxic
600 500 400 300 200 100 0
0
2
4
6
8
10
weight (kg) Fig. 2. Size dependency of toxic specimens of L. bohar.
12
nontoxic toxic
500 400 300 200 100 0
0
1
2
3
4
weight (kg) Fig. 5. Size dependency of toxic specimens of V. louti.
5
660
N. Oshiro et al. / Toxicon 56 (2010) 656–661
Of 108 fish found to be toxic, 61 specimens belonging to L. bohar, L. monostigma, V. louti, and E. fuscoguttatus were assayed for toxicity by MBA (Table 5). The occurrence of strongly toxic specimens was high in L. monostigma, agreeing with the fishermen’s view that L. monostigma was the most dangerous species (unpublished observation).
Table 5 Toxicity-based profiling of representative fish species: number and percentage (in parentheses) of toxic specimens in each class of toxicity (expressed in mice unit per gram of fish flesh, MU/g) within a toxicity range from 0.025 to 0.4. Species
Number of toxic specimen
Toxicity range 0.025
0.05
0.1
0.2
0.4
L. monostigma L. bohar V. louti E. fuscoguttatus
29 20 7 5
4 9 5 4
7 6 2 1
8 (28) 5 (25)
6 (21)
4 (14)
(14) (45) (71) (80)
(24) (30) (29) (20)
3.3. LC/MS analysis of the flesh extract In our separate work on ELISA analysis using anti-CTX3C and anti-51-hydroxyCTX3C antibodies (Oguri et al., 2003; Tsumuraya et al., 2006), we confirmed that a single treatment on an HLB cartridge column could efficiently remove interfering substances and ensure a good recovery of the two toxins spiked to extracts prepared from nontoxic flesh (unpublished data). In the LC/MS analysis, however, similarly prepared nontoxic flesh-extracts produced many interfering peaks in the chromatograms. Thus, we appended a clean-up step using a silica gel cartridge column, after the treatment on an HLB cartridge (Fig. 1). None of the fish samples exhibited in LC/MS chromatograms peaks attributable to CTX3C or 51-hyrdoxyCTX3C. The absence of the two toxins was also confirmed by ELISA (unpublished observation). On the other hand, CTX1B was detected by LC/MS in all samples judged as toxic by MBA. The intensities of the CTX1B peaks were proportionally correlated with the MBA results, though quantitative determination was unsuccessful (Fig. 6). The discrepancy between the MBA and LC/ MS results could be explained by the two reasons.(1) The technical difficulty of determining the extremely low concentration of CTX1B can be mentioned. From the LC99 of CTX1B (350 ng/kg, i.p., ddY male mouse, 20 g), the MBA toxicities of the fish used in this experiment (0.025– 0.8 MU/g) can be calculated as 0.175–5.6 ng CTX1B/g flesh (Yasumoto, 2005). At such low levels of the toxin, accurate measurements would be difficult. (2) A more likely reason for the discrepancy would be the existence of other CTX congeners that contributed to the MBA toxicity but were undeterminable in LC/MS because of the lack of
as estimated by MBA should be set below 0.025 MU/g (0.175 ng CTX1B equivalence/g).
3.2. Rate for toxic individuals in representative 7 species of fish
amount of CTX1B-equivalent (MU/g)
A total of 612 fish specimens collected in Okinawa were screened for the toxicity of the flesh by MBA. Professor Tetsuo Yoshino (University of the Ryukyus) kindly identified the fish as belonging to L. monostigma, L. bohar, L. russellii, L. argentimaculatus, V. louti, V. albimarginata, and E. fuscoguttatus (Table 4). Of 612 fish, 108 fish (17.6%) were judged to be toxic (Table 4). The rates of toxic individuals were as follows: L. monostigma: 32.3%, E. fuscoguttatus: 20.8%, L. bohar: 11.9%, V. louti: 14.3%. Only one out of 36 samples of V. albimarginata and two of 74 samples of L. russellii were found toxic. None of the 35 samples of L. argentimaculatus was toxic. Among the L. bohar specimens, individuals weighing less than 4 kg were nontoxic (Fig. 2). The toxic ratio rose to 37.7% among specimens above 4 kg and jumped to 61.1% in fish over 7 kg (Fig. 2). Similarly, a parallel increase of toxicity and body weight was observed in L. monostigma, E. fuscoguttatus, and V. louti (Figs. 3–5). Since ciguatoxins accumulate in fish via food chain, it is conceivable to find large individuals more toxic than small ones.
0.30 0.25 y = 0.4386x + 0.0056 R2 = 0.7247
0.20 0.15 0.10 0.05 0.00 0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
total toxicity (MU/g) Fig. 6. Correlation between the LC/MS results expressed in the CTX1B-equivalent and the MBA results.
N. Oshiro et al. / Toxicon 56 (2010) 656–661
standard toxins. As many as 23 congeners have been identified in fish and the causative dinoflagellate, G. toxicus, collected in French Polynesia (Yasumoto, 2001) and most of them are, at present, unavailable for use as calibrants in LC/MS. In all kinds of experiments to optimise and validate toxin analysis, the absence or difficult availability of reference toxins laid obstacles. Nevertheless, the present study is the first to confirm CTX1B in fish from Okinawa and indicates the usefulness of the LC/MS method as a tool to screen individual fish for toxicity. The lowest toxicity in the meal involved in poisoning was found to be 0.025 MU/g (¼0.175 ng CTX1B/g flesh). The data would be useful for setting up a regulation level for CTXs. The high ratio of toxic individuals in L. monostigma, L. bohar, and other species indicated they were inappropriate for food.
Acknowledgments The authors thank to Professor Tetsuo Yoshino (University of the Ryukyus) for identification of fish specimens, Professor Hiroshi Nagai (Tokyo University of Marine Science and Technology) and Mr. Takashi Uehara (Okinawa Prefectural Institute of Health and Environment: OIHE) for comments and discussions, Mr. Tamio Tamaki, Mr. Kaoru Kinjo, Mr. Toshinobu Muroi, Ms. Natsuko Teruya, Ms. Ayuko Koja and Ms. Satsuki Sakugawa (OIHE) for laboratory assistance, and Mr. Atsushi Ohno and Mr. Tsutomu Yamauchi (Pharmaceutical Affairs and Sanitation Division, Department of Health and Welfare, Okinawa Prefectural Government) for providing CFP case reports. This work was partially supported by ‘‘Modeling of Innovative Research Results’’ and ‘‘SORST’’, Japan Science and Technology Agency (JST), and ‘‘City Area Program’’, Ministry of Education, Culture, Sports, Science and Technology, Japan.
661
Conflicts of interest The authors declare that there are no conflicts of interest. References Hashimoto, Y., Konosu, S., Yasumoto, T., Kamiya, H., 1969a. Ciguatera in the Ryukyu and Amami Islands. Bull. Jpn. Soc. Sci. Fish 35, 316–326. Hashimoto, Y., Yasumoto, T., Kamiya, H., Yoshida, T., 1969b. Occurrence of ciguatoxin and ciguaterin in ciguatoxic fishes in the Ryukyu and Amami Island. Bull. Jpn. Soc. Sci. Fish 35, 327–332. Hirama, M., Oishi, T., Uehara, H., Inoue, M., Maruyama, M., Oguri, H., Satake, M., 2001. Total synthesis of ciguatoxin CTX3C. Science 294, 1904–1907. Inoue, M., Miyazaki, K., Ishihara, Y., Tatami, A., Ohnuma, Y., Kawada, Y., Komano, K., Yamashita, S., Lee, N., Hirama, M., 2006. Total synthesis of ciguatoxin and 51-hydroxyCTX3C. J. Am. Chem. Soc. 128, 9352–9354. Lehane, L., Lewis, R.J., 2000. Ciguatera: recent advances and but the risk remains. Int. J. Food Microbiol. 61, 91–125. Oguri, H., Hirama, M., Tsumuraya, T., Fujii, I., Maruyama, M., Uehara, H., Nagumo, Y., 2003. Synthesis-based approach toward direct sandwich immunoassay for ciguatoxin CTX3C. J. Am. Chem. Soc. 125, 7608–7612. Oshiro, N., Tamanaha, K., 2006. Food poisoning and consumer complaint cases caused by chemicals and natural toxins in Okinawa –2005. Annual Report of Okinawa Prefectural Institute of Health and Environment 40, 139–143 (in Japanese). Oshiro, N., Tamanaha, K., 2007. Food poisoning and consumer complaint cases caused by chemicals and natural toxins in Okinawa –2006. Annual Report of Okinawa Prefectural Institute of Health and Environment 41, 167–170 (in Japanese). Satake, M., 2005. Ciguatera. Shokuhin Eisei Kensa Shishin Rikagaku-hen. In: Ministry of Health, Labour and Welfare. Japan Food Hygiene Association, Tokyo, pp. 691–695 (in Japanese). Tsumuraya, T., Fujii, I., Inoue, M., Tatami, A., Miyazaki, K., Hirama, M., 2006. Production of monoclonal antibodies for sandwich immunoassay detection of ciguatoxin 51-hydroxyCTX3C. Toxicon 48, 287–294. Wong, C.K., Hung, P., Lee, K.L.H., Kam, K.M., 2005. Study of an outbreak of ciguatera fish poisoning in Hong Kong. Toxicon 46, 563–571. Yasumoto, T., 2001. The chemistry and biological function of natural marine toxins. Chem. Rec. 1, 228–242. Yasumoto, T., 2005. Chemistry, etiology, and food chain dynamics of marine toxins. Proc. Jpn. Acad. Ser. B Phys. Biol. Sci. 81, 43–51. Yasumoto, T., Nakajima, I., Bagnis, R., Adachi, R., 1977. Finding of dinoflagellate as a likely culprit of ciguatera. Bull. Jpn. Soc. Sci. Fish 43, 1021–1026.
Contents lists available at ScienceDirect
Toxicon journal homepage: www.elsevier.com/locate/toxicon
Ciguatera incidence and fish toxicity in Okinawa, Japan Naomasa Oshiro a, b, *, Kentaro Yogi a, c, Shuko Asato a, c, Toshiki Sasaki d,1, Koji Tamanaha a, Masahiro Hirama c, e, Takeshi Yasumoto f, Yasuo Inafuku a a
Okinawa Prefectural Institute of Health and Environment, 2085 Aza-Ozato, Ozato, Nanjo, Okinawa 901-1202, Japan Tokyo University of Marine Science and Technology, 4-5-7 Konan, Minato-ku, Tokyo 108-8477, Japan c SORST, Japan Science and Technology Agency, Japan d Minamata Environmental Research and Develop Center Co., Ltd., 5-98 Hamamatsu, Minamata, Kumamoto 867-0068, Japan e Department of Chemistry, Graduate School of Science, Tohoku University, 6-3 Aramaki, Aoba-ku, Sendai, Miyagi 980-8578, Japan f Okinawa Science and Technology Promotion Center, 12-75 Suzaki, Uruma, Okinawa 904-2234, Japan b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 January 2009 Received in revised form 19 May 2009 Accepted 19 May 2009 Available online 9 June 2009
Okinawa being located in the subtropical region has the highest incidence of ciguatera in Japan. Officially, 33 outbreaks involving 103 patients have been reported between 1997 and 2006. The implicated species were Variola louti, Lutjanus bohar, Lutjanus monostigma, Epinephelus fuscoguttatus, unidentified Lutjanus sp., Plectropomus areolatus, Oplegnathus punctatus, Epinephelus polyphekadion, Caranx ignobilis and moray eel. Toxicities of the leftover meals, as determined by mouse bioassays, ranged from 0.025 to 0.8 MU/g or above (equivalent to 0.175–5.6 ng CTX1B/g). We collected 612 specimens of fish belonging to L. monostigma, L. bohar, Lutjanus argentimaculatus, Lutjanus russellii, V. louti, Variola albimarginata, and E. fuscoguttatus from the coasts around Okinawa and examined the toxicity of the flesh by the mouse bioassay. The rate of toxic fish was as follows: L. monostigma: 32.3%, L. bohar: 11.9%, V. louti: 14.3%, E. fuscoguttatus: 20.8%. Only one out of 36 samples of V. albimarginata and two of 74 samples of L. russellii were found toxic. None of the 35 samples of L. argentimaculatus was toxic. Nor the L. bohar samples weighing less than 4 kg were toxic. In all toxic samples, CTX1B was detected by LC/MS analysis but CTX3C and 51hydroxyCTX3C were not. Ó 2009 Elsevier Ltd. All rights reserved.
Keywords: Ciguatera Incidence Fish Toxicity Ciguatoxin Okinawa Mouse bioassay
1. Introduction Ciguatera fish poisoning (CFP) associated with gastrointestinal, cardiovascular and neurological symptoms and signs is one of the largest scale food poisonings of nonmicrobial origins annually affecting 20,000–60,000 people worldwide (Lehane and Lewis, 2000; Yasumoto, 2005). While ciguatera endemic areas are tropical and subtropical
* Corresponding author at: Okinawa Prefectural Institute of Health and Environment, 2085 Aza-Ozato, Ozato, Nanjo, Okinawa 901-1202, Japan. Tel.: þ81 98 945 0329; fax: þ81 98 945 9366. E-mail address: [email protected] (N. Oshiro). 1 Present address: Minamata Research Center, Chisso Corporation. 1-1 Noguchi-cho, Minamata, Kumamoto 867-8501, Japan. 0041-0101/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.toxicon.2009.05.036
Indo-Pacific Ocean and Caribbean Sea where coral reef developed, increasing world trade of seafood and international tourisms cause outbreaks in other climate (Lehane and Lewis, 2000; Yasumoto, 2005; Wong et al., 2005). Original forms of principal toxins, ciguatoxins (CTXs), are produced by an epiphytic dinoflagellate Gambierdiscus toxicus and transferred to herbivorous and carnivorous fish via food chain (Yasumoto et al., 1977; Yasumoto, 2005). The chemical structures of ciguatoxins in G. toxicus and in fish have been elucidated (Yasumoto, 2001, 2005). Ciguatoxin4A (CTX4A) and ciguatoxin-3C (CTX3C) produced by the dinoflagellate were presumed to undergo structural changes in fish to produce the most representative toxin, ciguatoxin-1B (CTX1B), 51-hydroxyCTX3C and other metabolite toxins (Yasumoto, 2001, 2005). Occurrence of
N. Oshiro et al. / Toxicon 56 (2010) 656–661
CFP is rather rare in Japan, and Okinawa is the only area where sporadic but regular occurrence of CFP is reported. Okinawa is the largest of the Ryukyu archipelago that consists of 48 islands lying in the south-westernmost end of Japan. Its climate is subtropical and CFP has been recognized as endemic in a previous survey (Hashimoto et al., 1969a,b). In this paper, we summarized CFP outbreaks occurred in Okinawa, analyzed fish toxicity, and demonstrated the presence of CTX1B. 2. Materials and methods 2.1. Epidemiological data CFP incidents officially reported by medical doctors to the Department of Health and Welfare, Okinawa Prefectural Government, between 1997 and 2006 are used for epidemiological analysis in this paper. The toxic levels of implicated fish were determined by the official mouse bioassay method (MBA, Satake, 2005), whenever the leftover food and/or remnant fish were available. 2.2. Chemicals Normal saline solution (0.9% NaCl) was purchased from Otsuka Pharmaceuticals Co., Ltd.(Tokyo). The HPLC grade methanol, distilled water, and formic acid obtained from Wako Pure Chemical Industries, Ltd.(Tokyo), were used for the mobile phases in LC/MS analysis. Standard toxins, CTX1B, CTX3C and 51-hydroxyCTX3C for LC/MS were synthesized at Tohoku University (Hirama et al., 2001; Inoue et al., 2006). Other chemicals used were of analytical grade, otherwise stated. The composition of a solvent mixture was expressed in v/v. 2.3. Fish samples Frozen fish specimens of Lutjanus monostigma, Lutjanus bohar, Lutjanus argentimaculatus, Lutjanus russellii, Variola louti, Variola albimarginata, and Epinephelus fuscoguttatus, collected from the Okinawan coasts were purchased from the Fishermen’s Unions and stored at 20 C until use. The fish were thawed, photographed, measured for standard length and weight, dissected into flesh, liver, and other viscera, and frozen again at 20 C until use. Professor Tetsuo Yoshino, University of the Ryukyus, kindly conducted identification of the fish species. 2.4. Extraction Fish samples were extracted following the standard MBA method for ciguatoxin detection. A flesh sample (120 g) was thawed and homogenized with 350 ml of acetone twice. The combined acetone extract was evaporated to produce an aqueous concentrate, which was further extracted with 100 ml of diethyl ether twice. The ether extract was evaporated completely and the residue was partitioned between hexane (50 ml) and methanol– water (9:1, 25 ml). The aqueous methanol layer was completely freed of methanol and the residue was used for
657
the mouse bioassay. In all experiments, evaporation of a solution was carried out under reduced pressure. 2.5. Mouse bioassay (MBA) The standard MBA method was slightly modified to reduce the number of mice for use. The residue from an extract equivalent to 120 g of the flesh was suspended in 3 ml of 1% Tween 60 in normal saline solution. Two 1-ml portions of the suspension were injected respectively into a male mouse of ddY strain weighing 17–20 g by intraperitoneal route (Kyudo co., Ltd., Kumamoto, Japan). The mice were observed for 24 h. When the two mice died or survived, the sample was regarded as toxic or nontoxic, respectively. If one was dead and the other survived, remaining 1 ml of the sample suspension was injected into another mouse. The toxicity of the sample was judged by the survival or death of the third mouse. The flesh samples judged to be toxic by MBA were extracted similarly. Serial dilutions of the extracts were prepared and used for further confirmation of the toxicity levels by MBA. For example, a suspension equivalent to 20 g flesh/ml was prepared and 1 ml and 0.5 ml portions were injected, then the suspension was diluted four times with Tween 60 solution and used as same manner. 2.6. Solid phase extraction (SPE) and LC/MS analysis An extract prepared from 20 g of flesh was dissolved in 20 ml of methanol–water (7:3) containing 1% ammonium hydroxide. The solution was applied to an OASIS HLB cartridge (6 cc, 200 mg, Waters, Massachusetts) preconditioned with methanol followed by distilled water. The toxins retained on the cartridge were eluted first with 3 ml of methanol–water (8:2) containing 1% ammonium hydroxide (Fr. A) and next with 2 ml of acetonitrile–water (7:3) (Fr. B). Fractions A and B were dried and dissolved in 0.5 ml of methanol. A 0.1 ml portion of the each solution was mixed with 0.9 ml of chloroform and passed through a silica gel cartridge (InertSep Si, 100 mg, GL Sciences INC., Tokyo) preconditioned with the same solvent. The eluate was dried, dissolved in methanol (0.1 ml), and the designated test solution was used for LC/MS analysis. The cleanup procedure is presented in Fig. 1. An Agilent 1100 LC/MSD SL system (Agilent Technologies, California) was used for LC/MS analysis. HPLC separations were performed on a Cadenza CD-C18 column (3 mm, 2 150 mm, Imtakt co., Kyoto, Japan) using 0.1% formic acid–methanol (15:85) as the mobile phase. The column temperature and flow rate were kept at 40 C and 0.2 ml/min, respectively. MS parameters were set as follows: Ionization; API-ES, ion mode; positive, fragmenter voltage; 350 V, nebulizer gas; N2 20 psi., capillary voltage; 4000 V, Dry gas; 10 l/min of N2 at 350 C. A 5-ml portion of the test solution was injected into the LC/MS system and positive ions of m/z 1133.5 (0–7.5 min, [CTX1B þ Na]þ), m/z 1061.5 (7.5–15 min., [51-hydroxyCTX3C þ Na]þ), and m/z 1045.5 (15–35 min, [CTX3C þ Na]þ) were monitored to detect respective toxins. The synthetic standards of CTX1B, 51-hydroxyCTX3C and CTX3C used were dissolved in methanol (1 ng/ml). The signal/noise (S/N) values for the
658
N. Oshiro et al. / Toxicon 56 (2010) 656–661 Table 1 CFP incidents officially reported in Okinawa from 1997 to 2006.
flesh (20 g) extract dissolved in 20 ml of methanol/water (7:3) containing 1% ammonium hydroxide OASIS HLB (6cc, 200mg)* 3 ml of methanol/water (8:2) containing 1% ammonium hydroxide ---------------- Fr. A 2 ml of acetonitrile/water (7:3) ---------- Fr. B
Fr. B
Fr. A
Fr. A or B dried dissolved in 0.5 ml of methanol 0.1 ml of the solution 0.9 ml of chloroform InertSep Si (100mg)** 0.5 ml of chloroform/methanol (9:1) non absorbed portion dried dissolved in 0.1 ml of methanol LC/MS analysis Fig. 1. SPE of flesh extract for LC/MS analysis. *Preconditioned with methanol (2 ml) followed by distilled water (2 ml). **Preconditioned with chloroform/methanol (9:1, 5 ml).
toxins in the above order were 3.2, 20.1 and 4.0, respectively. 3. Results and discussion 3.1. Epidemiology Although 33 outbreaks have been reported during the ten years from 1997 to 2006 in Okinawa (Table 1), rumors among fishermen suggest many cases remain unreported. The seasonality of the incidents seems thin, though the highest number (6) was recorded in May (Table 1). As listed in Table 2, the most frequently implicated fish were V. louti and L. bohar (8 incidents for each), followed by L. monostigma (6), E. fuscoguttatus (2), unidentified Lutjanus sp.(2), Plectropomus areolatus (1), Oplegnathus punctatus (1), Epinephelus polyphekadion (1), Caranx ignobilis (1) and moray eel (1). Importantly, leftover foods and/or uncooked fish flesh could be obtained for analysis in 12 incidents. Toxicities of these samples estimated by MBA ranged from 0.025 to above 0.8 MU/g (Table 3). The minimum toxicity to
No.
Date
Number of patients
Number of people consumed
Fish species
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
18/Jan/97 21/Sep/97 17/May/98 17/Jun/98 25/Feb/99 26/Feb/99 08/May/99 24/May/99 03/Aug/99 31/Aug/99 21/Sep/99 18/Oct/99 16/Jan/00 20/May/00 24/Jun/00 19/May/01 31/Aug/01 21/Jun/02 04/Nov/02 30/Nov/02 01/Apr/03 03/Jul/03 22/Jul/03 06/Oct/03 06/Mar/04 18/Nov/04 06/Jul/05 10/Sep/05 24/Jan/06 18/Mar/06 26/May/06 29/Jun/06 31/Dec/06
6 3 3 3 1 1 1 2 1 3 6 1 7 5 4 3 3 2 2 3 2 4 3 4 2 3 7 6 2 2 1 4 3
9 5 7 5 1 2 7 6 4 3 6 11 7 5 26 3 3 5 2 3 5 4 3 4 3 4 9 6 2 2 27 4 8
L. monostigma Lutjanus sp. L. bohar V. louti E. fuscoguttatus E. fuscoguttatus L. bohar L. bohar L. monostigma C. ignobilis Unidentified L. bohar E. polyphekadion L. bohar Snapper Unidentified L. monostigma L. bohar P. areolatus L. monostigma O. punctatus L. monostigma L. monostigma V. louti Moray eel V. louti V. louti V. louti V. louti L. bohar L. bohar V. louti V. louti
induce illness by oral intake was calculated from the data obtained on fish (0.1 MU/g) implicated in the outbreak No. 28 (Oshiro and Tamanaha, 2006). All the six persons who ate the fish were intoxicated. The minimum amount consumed by the patients was estimated to be 140 g. Thus, 14 MU of the toxin was enough to induce illness in an adult. Based on the i.p. lethality of CTX1B in mouse (0.35 mg/kg, Yasumoto, 2001), 1 MU could be regarded as equivalent to 7 ng of CTX1B. Hence, 14 MU is equivalent to 98 ng CTX1B. This value is close to the estimation (10 MU ¼ 70 ng CTX1B) made on fish in French Polynesia (Yasumoto, 2005). On the other hand, the fish No. 31 had the same toxicity level as of Table 2 Frequency of causative species in the outbreaks in Okinawa. Species Variola louti Lutjanus bohar L. monostigma Epinephelus fuscoguttatus Lutjanidae Plectropomus areolatus Oplegnathus punctatus E. polyphekadion Caranx ignobilis Moray eel Unidentified Total
Number of outbreaks 8 8 6 2 2 1 1 1 1 1 2 33
Ratio (%) 24 24 18 6 6 3 3 3 3 3 6
N. Oshiro et al. / Toxicon 56 (2010) 656–661 Table 3 Toxicity (in mice unit per gram of fish flesh, MU/g) of the leftover meals and/or remnant fish from some of CFP incidents reported in Table 1. Fish species
Test sample
Toxicity (MU/g)
2 4 13
Lutjanus sp. V. louti E. polyphekadion
17 20 22
L. monostigma L. monostigma L. monostigma
23 24
L. monostigma V. louti
26
V. louti V. louti V. louti L. bohar V. louti
Cooked flesh Raw flesh Cooked flesh Soupa Cooked flesh Cooked flesh Raw flesh Mixed soupb Mixed soupb Raw flesh Mixed soupb Fleshc Fleshc Raw flesh Cooked flesh Raw flesh
0.29 0.1 0.05 <0.025 >0.2 >0.8 >0.2 0.025 >0.2 0.4 0.1 >0.2 0.1 0.1 0.1 0.05
28 31 32 a b c
500
standard length (mm)
No.
nontoxic toxic
400
300
200
100
0 0.0
Toxica
L. monostigma E. fuscoguttatus V. louti L. bohar V. albimarginata L. russellii L. argentimaculatus
226 24 49 168 36 74 35
73 5 7 20 1 2 0
Total
612
108
a
1.5
2.0
2.5
900
Ratio of toxic specimen (%) 32.3 20.8 14.3 11.9 2.8 2.7 0.0
standard length (mm)
Table 4 Frequency of toxic specimens in representative fish species.
Examined
1.0
Fig. 3. Size dependency of toxic specimens of L. monostigma.
nontoxic toxic
800
Number of specimens
0.5
weight (kg)
Assay was made after removing flesh and bones contained in the soup. Assay was made after removing bones contained in the soup. The flesh had been lightly washed with hot water.
Species
659
700 600 500 400 300 200 100 0
0
5
10
15
20
25
weight (kg) Fig. 4. Size dependency of toxic specimens of E. fuscoguttatus.
Toxicity was evaluated by MBA. Only the flesh was used for assay.
No. 28 (0.1 MU/g) but caused intoxication only in one person. Because the fish was shared by 27 persons, the amount consumed by one person could be smaller than in the incident No. 28. Interestingly, the only person intoxicated had the jaw muscle (Oshiro and Tamanaha, 2007), supporting the reputation among fishermen that head was
more toxic than the body muscle. The lowest mousetoxicity of flesh to cause intoxication was suggested to be 0.05 MU/g (No. 13 and 32). Furthermore, involvement of the fish soup with toxicity as low as 0.025 MU/g or lower (No. 13 and 22) indicates the potential danger due to a meal of such a low toxicity. Thus, the maximum allowance level
600
800
standard length (mm)
700
standard length (mm)
nontoxic toxic
600 500 400 300 200 100 0
0
2
4
6
8
10
weight (kg) Fig. 2. Size dependency of toxic specimens of L. bohar.
12
nontoxic toxic
500 400 300 200 100 0
0
1
2
3
4
weight (kg) Fig. 5. Size dependency of toxic specimens of V. louti.
5
660
N. Oshiro et al. / Toxicon 56 (2010) 656–661
Of 108 fish found to be toxic, 61 specimens belonging to L. bohar, L. monostigma, V. louti, and E. fuscoguttatus were assayed for toxicity by MBA (Table 5). The occurrence of strongly toxic specimens was high in L. monostigma, agreeing with the fishermen’s view that L. monostigma was the most dangerous species (unpublished observation).
Table 5 Toxicity-based profiling of representative fish species: number and percentage (in parentheses) of toxic specimens in each class of toxicity (expressed in mice unit per gram of fish flesh, MU/g) within a toxicity range from 0.025 to 0.4. Species
Number of toxic specimen
Toxicity range 0.025
0.05
0.1
0.2
0.4
L. monostigma L. bohar V. louti E. fuscoguttatus
29 20 7 5
4 9 5 4
7 6 2 1
8 (28) 5 (25)
6 (21)
4 (14)
(14) (45) (71) (80)
(24) (30) (29) (20)
3.3. LC/MS analysis of the flesh extract In our separate work on ELISA analysis using anti-CTX3C and anti-51-hydroxyCTX3C antibodies (Oguri et al., 2003; Tsumuraya et al., 2006), we confirmed that a single treatment on an HLB cartridge column could efficiently remove interfering substances and ensure a good recovery of the two toxins spiked to extracts prepared from nontoxic flesh (unpublished data). In the LC/MS analysis, however, similarly prepared nontoxic flesh-extracts produced many interfering peaks in the chromatograms. Thus, we appended a clean-up step using a silica gel cartridge column, after the treatment on an HLB cartridge (Fig. 1). None of the fish samples exhibited in LC/MS chromatograms peaks attributable to CTX3C or 51-hyrdoxyCTX3C. The absence of the two toxins was also confirmed by ELISA (unpublished observation). On the other hand, CTX1B was detected by LC/MS in all samples judged as toxic by MBA. The intensities of the CTX1B peaks were proportionally correlated with the MBA results, though quantitative determination was unsuccessful (Fig. 6). The discrepancy between the MBA and LC/ MS results could be explained by the two reasons.(1) The technical difficulty of determining the extremely low concentration of CTX1B can be mentioned. From the LC99 of CTX1B (350 ng/kg, i.p., ddY male mouse, 20 g), the MBA toxicities of the fish used in this experiment (0.025– 0.8 MU/g) can be calculated as 0.175–5.6 ng CTX1B/g flesh (Yasumoto, 2005). At such low levels of the toxin, accurate measurements would be difficult. (2) A more likely reason for the discrepancy would be the existence of other CTX congeners that contributed to the MBA toxicity but were undeterminable in LC/MS because of the lack of
as estimated by MBA should be set below 0.025 MU/g (0.175 ng CTX1B equivalence/g).
3.2. Rate for toxic individuals in representative 7 species of fish
amount of CTX1B-equivalent (MU/g)
A total of 612 fish specimens collected in Okinawa were screened for the toxicity of the flesh by MBA. Professor Tetsuo Yoshino (University of the Ryukyus) kindly identified the fish as belonging to L. monostigma, L. bohar, L. russellii, L. argentimaculatus, V. louti, V. albimarginata, and E. fuscoguttatus (Table 4). Of 612 fish, 108 fish (17.6%) were judged to be toxic (Table 4). The rates of toxic individuals were as follows: L. monostigma: 32.3%, E. fuscoguttatus: 20.8%, L. bohar: 11.9%, V. louti: 14.3%. Only one out of 36 samples of V. albimarginata and two of 74 samples of L. russellii were found toxic. None of the 35 samples of L. argentimaculatus was toxic. Among the L. bohar specimens, individuals weighing less than 4 kg were nontoxic (Fig. 2). The toxic ratio rose to 37.7% among specimens above 4 kg and jumped to 61.1% in fish over 7 kg (Fig. 2). Similarly, a parallel increase of toxicity and body weight was observed in L. monostigma, E. fuscoguttatus, and V. louti (Figs. 3–5). Since ciguatoxins accumulate in fish via food chain, it is conceivable to find large individuals more toxic than small ones.
0.30 0.25 y = 0.4386x + 0.0056 R2 = 0.7247
0.20 0.15 0.10 0.05 0.00 0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
total toxicity (MU/g) Fig. 6. Correlation between the LC/MS results expressed in the CTX1B-equivalent and the MBA results.
N. Oshiro et al. / Toxicon 56 (2010) 656–661
standard toxins. As many as 23 congeners have been identified in fish and the causative dinoflagellate, G. toxicus, collected in French Polynesia (Yasumoto, 2001) and most of them are, at present, unavailable for use as calibrants in LC/MS. In all kinds of experiments to optimise and validate toxin analysis, the absence or difficult availability of reference toxins laid obstacles. Nevertheless, the present study is the first to confirm CTX1B in fish from Okinawa and indicates the usefulness of the LC/MS method as a tool to screen individual fish for toxicity. The lowest toxicity in the meal involved in poisoning was found to be 0.025 MU/g (¼0.175 ng CTX1B/g flesh). The data would be useful for setting up a regulation level for CTXs. The high ratio of toxic individuals in L. monostigma, L. bohar, and other species indicated they were inappropriate for food.
Acknowledgments The authors thank to Professor Tetsuo Yoshino (University of the Ryukyus) for identification of fish specimens, Professor Hiroshi Nagai (Tokyo University of Marine Science and Technology) and Mr. Takashi Uehara (Okinawa Prefectural Institute of Health and Environment: OIHE) for comments and discussions, Mr. Tamio Tamaki, Mr. Kaoru Kinjo, Mr. Toshinobu Muroi, Ms. Natsuko Teruya, Ms. Ayuko Koja and Ms. Satsuki Sakugawa (OIHE) for laboratory assistance, and Mr. Atsushi Ohno and Mr. Tsutomu Yamauchi (Pharmaceutical Affairs and Sanitation Division, Department of Health and Welfare, Okinawa Prefectural Government) for providing CFP case reports. This work was partially supported by ‘‘Modeling of Innovative Research Results’’ and ‘‘SORST’’, Japan Science and Technology Agency (JST), and ‘‘City Area Program’’, Ministry of Education, Culture, Sports, Science and Technology, Japan.
661
Conflicts of interest The authors declare that there are no conflicts of interest. References Hashimoto, Y., Konosu, S., Yasumoto, T., Kamiya, H., 1969a. Ciguatera in the Ryukyu and Amami Islands. Bull. Jpn. Soc. Sci. Fish 35, 316–326. Hashimoto, Y., Yasumoto, T., Kamiya, H., Yoshida, T., 1969b. Occurrence of ciguatoxin and ciguaterin in ciguatoxic fishes in the Ryukyu and Amami Island. Bull. Jpn. Soc. Sci. Fish 35, 327–332. Hirama, M., Oishi, T., Uehara, H., Inoue, M., Maruyama, M., Oguri, H., Satake, M., 2001. Total synthesis of ciguatoxin CTX3C. Science 294, 1904–1907. Inoue, M., Miyazaki, K., Ishihara, Y., Tatami, A., Ohnuma, Y., Kawada, Y., Komano, K., Yamashita, S., Lee, N., Hirama, M., 2006. Total synthesis of ciguatoxin and 51-hydroxyCTX3C. J. Am. Chem. Soc. 128, 9352–9354. Lehane, L., Lewis, R.J., 2000. Ciguatera: recent advances and but the risk remains. Int. J. Food Microbiol. 61, 91–125. Oguri, H., Hirama, M., Tsumuraya, T., Fujii, I., Maruyama, M., Uehara, H., Nagumo, Y., 2003. Synthesis-based approach toward direct sandwich immunoassay for ciguatoxin CTX3C. J. Am. Chem. Soc. 125, 7608–7612. Oshiro, N., Tamanaha, K., 2006. Food poisoning and consumer complaint cases caused by chemicals and natural toxins in Okinawa –2005. Annual Report of Okinawa Prefectural Institute of Health and Environment 40, 139–143 (in Japanese). Oshiro, N., Tamanaha, K., 2007. Food poisoning and consumer complaint cases caused by chemicals and natural toxins in Okinawa –2006. Annual Report of Okinawa Prefectural Institute of Health and Environment 41, 167–170 (in Japanese). Satake, M., 2005. Ciguatera. Shokuhin Eisei Kensa Shishin Rikagaku-hen. In: Ministry of Health, Labour and Welfare. Japan Food Hygiene Association, Tokyo, pp. 691–695 (in Japanese). Tsumuraya, T., Fujii, I., Inoue, M., Tatami, A., Miyazaki, K., Hirama, M., 2006. Production of monoclonal antibodies for sandwich immunoassay detection of ciguatoxin 51-hydroxyCTX3C. Toxicon 48, 287–294. Wong, C.K., Hung, P., Lee, K.L.H., Kam, K.M., 2005. Study of an outbreak of ciguatera fish poisoning in Hong Kong. Toxicon 46, 563–571. Yasumoto, T., 2001. The chemistry and biological function of natural marine toxins. Chem. Rec. 1, 228–242. Yasumoto, T., 2005. Chemistry, etiology, and food chain dynamics of marine toxins. Proc. Jpn. Acad. Ser. B Phys. Biol. Sci. 81, 43–51. Yasumoto, T., Nakajima, I., Bagnis, R., Adachi, R., 1977. Finding of dinoflagellate as a likely culprit of ciguatera. Bull. Jpn. Soc. Sci. Fish 43, 1021–1026.