Ciguatera fish poisoning in Hong Kong–A 10-year perspective on the class of ciguatoxins

Toxicon 86 (2014) 96e106

Contents lists available at ScienceDirect

Toxicon journal homepage: www.elsevier.com/locate/toxicon

Ciguatera fish poisoning in Hong KongeA 10-year perspective on the class of ciguatoxins Chun-Kwan Wong, Patricia Hung, Janice Y.C. Lo* Public Health Laboratory Services Branch, Centre for Health Protection, Department of Health, 382 Nam Cheong St., Shek Kip Mei, Kowloon, Hong Kong, China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 17 January 2014 Received in revised form 13 May 2014 Accepted 16 May 2014 Available online 28 May 2014

The present study used liquid chromatography-tandem mass spectrometry (LC-MS/MS) to investigate retrospectively ciguatoxin (CTX)-positive samples as determined by mouse bioassay (MBA) in the past 10 years in Hong Kong. The results showed that Pacific CTXs (PCTX-1, -2 and -3) were the most commonly observed toxins found in the samples, indicating Pacific Ocean areas as the most important origin of ciguatera fish poisoning. Clinical diagnosis from ciguatera patients also revealed the predominance of neurological illnesses in most cases, supporting intoxication of Pacific origin. This study demonstrated the ability of laboratory analysis to identify and quantify Pacific CTXs in suspected fish samples, so as to support the clinical diagnosis of ciguatera. Comparative analysis (Student's t-test and Spearman's rank correlation analysis) on the two CTX detection methods showed approximate linearity for overall P-CTXs (P-CTX-1, -2 and -3)/P-CTX-1 alone as derived by LC-MS/MS and total toxicity levels (P-CTX-1 equivalent) as determined by MBA. The LCMS/MS method coupled with the rapid extraction method could allow the detection of trace amount of CTXs at levels below the clinically relevant limit, 0.1 ppb P-CTX-1 in fish flesh. For practical application, the adoption of a two-tiered approach for testing, chemical analysis by LC-MS/MS for toxic fish screening, coupled with biological assay by MBA for final toxicity confirmation, was proposed for first-line screening of CTX in potentially contaminated fish samples in the market, with an aim to minimizing the use of laboratory mice and at the same time providing reasonably effective means for routine analysis. © 2014 Elsevier Ltd. All rights reserved.

Keywords: Ciguatera fish poisoning Epidemiology LC-MS/MS Pacific CTX Public health

1. Introduction Ciguatera fish poisoning (CFP) is a biotoxin-induced clinical syndrome caused by consumption of tropical and sub-tropical coral reef fish that have been contaminated with natural neurotoxins. The toxins are lipid-soluble and heat-stable polyether sodium channel ichthyosarcotoxins known as ciguatoxins (CTXs). Members of the unicellular marine dinoflagellate genus Gambierdiscus are the main producers of CTXs. These algae-related toxins can be * Corresponding author. Tel.: þ852 2319 8254; fax: þ852 2776 5758. E-mail addresses: [email protected] (C.-K. Wong), [email protected]. hk (P. Hung), [email protected] (J.Y.C. Lo). http://dx.doi.org/10.1016/j.toxicon.2014.05.006 0041-0101/© 2014 Elsevier Ltd. All rights reserved.

bioaccumulated through the food web as a result of grazer herbivorous fish ingesting Gambierdiscus spp. such as Gambierdiscus toxicus. The toxins are subsequently metabolized to more toxic forms at each successive trophic level until reaching humans (Lewis and Holmes, 1993; Lehane and Lewis, 2000). CFP is a debilitating disease characterized by a wide array of symptoms such as gastrointestinal, neurological and cardiovascular disorders (Palafox and Buenconsejo-Lum, 2001; Pottier et al., 2002; Keynan and Pottesman, 2004; Baumann et al., 2010). Symptoms can persist for weeks, months or years and may recur subsequently (Lewis, 2006; Friedman et al., 2008). CFP is an endemic food-borne illness in high-risk areas where reef fish are commonly caught in so-called “ciguatera belt”

C.-K. Wong et al. / Toxicon 86 (2014) 96e106

regions (circumglobal area between the latitudes of 35 N and 35 S) (Lehane and Lewis, 2000; Pearn, 2001), mainly in Pacific and Indian Oceans as well as Caribbean Sea. The CTXs can thus be classified into three main families based on their chemical structures found from the three ciguatera-prone areas, namely Pacific-, Indian- and Caribbean-CTXs (Dickey and Plakas, 2010). To date, more than 400 species of coral reef fish have been identified to be associated with CFP cases in humans (Lehane and Lewis, 2000; Lewis, 2001, 2006). Worldwide, it has been estimated that around 10,000e50,000 individuals suffered from this disease annually (Quod and Turquet, 1996; Daranas et al., 2001; Sumner et al., 2004). Until now, there is no proven effective antidote while symptomatic treatment with mannitol is recommended (Palafox et al., 1988; Friedman et al., 2008; Dickey and Plakas, 2010). With rapid development of tourism and international trade of seafood, the disease has become a potential threat on a global scale. It has been estimated that with misdiagnosis and under-reporting, only 2e10% of cases were reported to heath authorities (Lehane and Lewis, 2000; Lewis, 2001; Wong et al., 2005; Friedman et al., 2008; Caillaud et al., 2010; Dickey and Plakas, 2010). Therefore, the actual public health impact of the disease is likely to be underestimated. In addition, epidemiological perspective of ciguatera and knowledge on the class(es) of CTX are uncertain, limiting further public health research and assessments. Ciguatera has been known for centuries since Alexander the Great (Scheuer, 1994; Pearn, 2001; Stewart et al., 2010b). Halstead (1967) and (1988) also documented the disease in ancient China since at least the T'ang Dynasty (A.D. 618e907). In Hong Kong, cases of CFP have been recorded since the 1980s (Sadovy, 1997; Choi and Wong, 1994; Department of Health (2002)), where two significant outbreaks affecting more than 600 people occurred in 1998 and 2004 which attracted considerable public awareness (Wong et al., 2005). In addition, CFP is also the most common seafood intoxication by marine biotoxins (Choi and Wong, 1994; Wong et al., 2005). Most CFP outbreaks and sporadic poisoning incidents have been associated with the consumption of imported CTX contaminated coral reef fish (Wong et al., 2008). In the absence of reliable, robust and simple assays for rapid screening of potential ciguateric fish, traditional mouse bioassay (MBA) was still the mainstay for toxicity confirmation in fish in the markets and also fish remnants from CFP patients (Lewis, 2000, 2003; Lehane, 2000; Wong et al., 2005, 2008). Although this in vivo assay is superior in analysing total toxicity of the toxin-contaminated fish tissues with reference to the specific symptoms produced from inoculated mice, its routine use has raised concerns on the ethical issue of animal welfare. On the other hand, the bioassay also has problems of non-specificity and insensitivity, resulting in lack of general acceptance by most contemporary testing laboratories. For epidemiological perspectives, in addition to the toxicological data, it is particularly of interest to identify the class(es) of CTX implicated in the cases. In respect of geographical situation and previous clinical information from CFP records (Choi and Wong, 1994; Lewis, 2000; Centre for Health

97

Protection, 2005; Wong et al., 2008), Pacific and/or Indian Oceans CTXs were suggested to be the most likely predominant toxin(s) causing intoxication in Hong Kong. The present study focused on retrospective analysis of available MBA confirmed CTX-contaminated fish samples in the past 10 years by using liquid chromatography tandem mass spectrometry (LC-MS/MS) for toxin identification and quantification. The objectives were (1) to correlate the toxicity values of LC-MS/MS derived as a product of toxin concentration and potency for given toxin with MBA; (2) to investigate adoption of chemical method for supporting the clinical diagnosis of CFP and (3) to identify the possible class(es) of CTX(s) implicated in the past CFP cases/incidents in Hong Kong, so as to provide epidemiological information. 2. Materials and methods 2.1. Sample collection Coral reef fish samples were collected from markets by inspectorate staff of the Food and Environmental Hygiene Department in Hong Kong between year 2004 and 2013. Upon receiving reports of CFP food poisoning, outbreak and complaint cases, implicated fish samples were obtained promptly from victims/complainants for laboratory analysis of CTXs. A piece of giant grouper (Epinephelus lanceolatus) flesh (sample 2-os), provided by an overseas donor exhibiting ciguatera symptoms after eating portions of the flesh, was used as a positive control for clinical and laboratory analysis on CFP cases. All coral reef fish samples collected were immediately kept frozen at 20 or 80  C until testing. Specifically, fish samples collected from years 2004e2011 were stored at 80  C to reduce toxin degradation and/or inter-conversion. CTX-positive samples (27 samples), together with randomly selected negative samples (41 samples covering different types of coral reef fish), as previously determined by MBA, were retrieved for carrying out this retrospective study. 2.2. CTXs standards Pure Pacific-ciguatoxin-1 (P-CTX-1) was procured from Prof. Richard J. Lewis (Institute for Molecular Bioscience, University of Queensland, Australia) for toxin standard calibration and identification in LC-MS/MS. Serial dilutions of P-CTX-1 in 50% aqueous methanol under common coral fish matrix (leopard coral grouper) medium, which was previously tested by LC-MS/MS as P-CTX-negative, were adopted. Eight calibration standards ranging from 0.25 to 50 mg/L P-CTX-1 were constructed for toxin calculation. Correlation coefficient (r2) value of regression analysis greater than 0.99 was applied. The acceptable deviation of each calibration point was less than 20% as compared with its theoretical value. Pure Pacific-ciguatoxin-2 and -3 (PCTX-2 and -3) were subsequently obtained from the same institute for retention times and ion fragments identification in this study. According to our in-house study, no significant difference (Student's t-test, p > 0.05) was found in calibration results between analysis of P-CTX-1 dissolved in solely 50% aqueous methanol and in 50% aqueous

98

C.-K. Wong et al. / Toxicon 86 (2014) 96e106

methanol with common non-oily coral fish matrices (data not shown). Therefore, serial dilutions of P-CTX-2 and -3 in 50% aqueous methanol were then adopted for toxins evaluation. The calibration and acceptance criteria for PCTX-2 and -3 were the same as those of P-CTX-1 calibration.

cleanup. The vial was rinsed with totally 4.5 ml of chloroform and the rinsed solvent was loaded on the cartridge for washing. The final target analyte was then eluted out with 8 ml of chloroform:methanol (9:1, v:v). The eluent obtained was dried under a stream of nitrogen gas and low heat, reconstituted with 200 ml of 50% aqueous methanol prior to injection of sample (20 ml) into LC-MS/MS for chemical analysis.

2.3. Sample extraction All chemicals used for sample extraction were at least HPLC grade reagents or above. Coral reef fish samples were extracted in accordance with the rapid extraction method described by Lewis et al. (2009) and Stewart et al. (2010a) with slight modifications and optimization. Briefly, two grams of sliced fish flesh was weighed in a Falcon tube (50ml) and heated at 70  C water bath for 20 min to denature proteins for enhancing extraction efficiency during homogenization. The sample was then cooled at room temperature, or stored in 20  C freezer until use. For toxin extraction, the cooled fish sample was blended in 8 ml of methanol-hexane (3:1, v:v) for ~30 s using a homogenizer (IKA Ultra Turrax T25, setting 3e4) until homogenous slurry was formed. The homogenate was further sonicated for ~30 s followed by a gentle mixing before centrifugation at 3300 g for 20 min under 4  C/room temperature. The upper hexane layer was aspirated carefully using a pasture pipette and removed, and the lower methanol portion was then transferred into a single use syringe equipped with a 0.45 mm Millipore aqueous membrane filter (Millex®eHA) for filtration. The purified extract was collected in a scintillation vial, adjusted to 50e55% aqueous methanol by addition of Milli-Q water (Ultrapure water, 18.2 MU resistivity at 25  C). The sample extract was then loaded onto a C18 SPE cartridge (Alltech Prevail Maxi-Clean, 900 mg) for solid phase extraction for which the cartridge was preconditioned with approximately 4 ml of Milli-Q water. The cartridge was finally washed with totally 6.5 ml of 65% aqueous methanol before loading 8 ml of 80% aqueous methanol onto the cartridge for eluting the target analyte. The 80% methanol eluent (target analyte) was collected into a capped Falcon tube (50-ml), mixed with 4.2 ml of 1 M sodium chloride solution and 6.7 ml of chloroform followed by well mixing under a vortex for liquideliquid extraction. The mixture was centrifuged at 820 g for 4 min. The lower chloroform phase was transferred carefully to another scintillation vial, dried under filtered air with low heat. The dried residue was then re-dissolved in 4 ml of chloroform and subsequently loaded on a Sep-Pak silica SPE cartridge (Silica Plus, Waters®) which was pre-conditioned with approximately 4 ml of chloroform, for normal phase SPE

2.4. LC-MS/MS analysis A triple-quadrupole-linear ion-trap mass spectrometer, AB/Sciex API4000 QTRAP (AB/MDS Sciex, Concord, Canada) equipped with a turbo ionspray ionization interface was used for the detection of CTXs. An Agilent 1200 series HPLC system (Agilent, Palo Alto, CA, USA) comprising a C18 column (5 mm Phenomenex Luna, 2.1  150 mm) connected with a guard column (5 mm Phenomenex C18, 4  2.1 mm) was used for toxins separation. A flow rate of 0.4 ml/min with a linear gradient setup for two mobile phases (mobile solvent A and B) was employed. The mobile phases consisted of (1) solvent A: 0.1% formic acid in aqueous 2 mM ammonium formate; (2) solvent B: 0.1% formic acid in 95% acetonitrile with 2 mM ammonium formate. The gradient elution program started from 35% of solvent B to 100% over 5 min, held at 100% B for 2 min before returning to 35% B at 7.1 min. The column was finally equilibrated for 5 min with 35% B prior to the next run. The total cycle time was 12 min. The mass spectrometer adopted positive ion mode of detection using multiple reaction monitoring (MRM) with resolution for both Q1 and Q3 set at low condition. The operating parameters for the mass spectrometer are shown in Table 1. Two sets of dominant fragment ions, namely (PCTX-1a [m/z 1075.7], P-CTX-1b [m/z 1057.7] and P-CTX-1c [m/z 1093.7]) and (P-CTX-2a & -3a [m/z 1077.6], P-CTX-2b & -3b [m/z 1059.6] and P-CTX-2c & -3c [m/z 1041.6]), could be generated from the [M þ NH4]þ ions of P-CTX-1 (m/z 1128.7) and P-CTX-2 and -3 (m/z 1112.7), respectively. CTX-positive samples were identified by comparing the retention times and corresponding three ion fragments of P-CTX-1, -2 and -3 with the standard toxins. The acceptable range of retention time allowed for maximum of 3% deviation from the standard (e.g. P-CTX-1 acceptable range for this study: 4.79e5.08 min). Signal-to-noise (S/N) ratio of 3:1 and 10:1 were applied for peak determination at the Lower Limit of Detection (LLOD) level and the Lower Limit of Quantitation (LLOQ), respectively. P-CTX-1, -2 and -3 concentrations (ppb) in samples determined by LC-MS/MS were converted to mouse toxicity (mouse unit per kg of fish

Table 1 Operating parameters for LCMS/MS analysis of CTXs. Ciguatoxin

Q1

Q3

Compound

CE

DP

TEM

CUR

EP

CXP

P-CTX-1

1128.7 1128.7 1128.7 1112.7 1112.7 1112.7

1075.7 1057.7 1093.7 1077.6 1059.6 1041.6

P-CTX-1a P-CTX-1b P-CTX-1c P-CTX-2&3a P-CTX-2&3b P-CTX-2&3c

32 eV

110 V

300

25

10

15

32 eV

110 V

300

25

10

15

P-CTX-2&-3

Keys: CE: Collision energy; DP: Declustering potential; TEM: Temperature; CUR: Curtain gas; EP: Entrance potential; CXP: Collision cell exit potential. Notes: nitrogen gas was used as collision gas; CEM parameter, the voltage applied to the detector, was set as 2000.

C.-K. Wong et al. / Toxicon 86 (2014) 96e106

flesh, MU/kg, P-CTX-1 equivalent) by multiplying toxicity factors to the concentration values with reference to the published toxicity equivalency factors (TEFs) for CTXs (EFSA Panel on Contaminants in the food chain, 2010). The TEFs (P-CTX-1 equivalent) for P-CTX-1, -2 and -3 are 1, 0.3 and 0.3, respectively. These toxicity factors are based on acute LD50 in mice by intraperitoneal (i.p.) injection of the toxins.

99

2.6. Statistical analysis Student t-test (two-tailed) was adopted to compare data between toxicity values from LC-MS/MS and MBA. Spearman's rank correlation analysis was also performed to assess the relationship between toxicity levels obtained by the two methods. A probability level of 0.05 or smaller was considered as significant for both tests.

2.5. Mouse bioassay (MBA) 3. Results A validated mouse bioassay (MBA) was performed in accordance with the procedures as described in Wong et al. (2005), (2008) and (2009) for sample extraction, mice injection and toxicity calculation. In order to study the toxicity and potential presence of CTXs in the samples, lipid soluble extracts of the coral fish samples (fish ether extracts) were inoculated by i.p. injection into mice. Symptoms or signs of CTXs intoxication and/or survival time(s) of poisoned mice were recorded for CTX identification and calculation. The toxicity was represented by MU/kg of fish flesh (P-CTX-1 equivalent). All experimental animals used were sacrificed in compliance with ethical standards. The mouse toxicity levels produced from MBA were compared with the corresponding toxicity levels derived from LC-MS/MS.

In the past ten years, thirty-three CTX-contaminated (PCTX-1 equivalent) coral reef fish samples were identified by MBA analysis. Among these positive samples (including positive control sample 2-os), 27 frozen fish samples (Table 2) were available for this retrospective study in which two original fish samples (samples 11 and 14) could not be retrieved for sample extraction. As more than one fish samples might have been collected for each case, another fish of the same batch was then taken instead for these two samples. Direct quantitative comparisons of results between MBA and LC-MS/MS analysis for these two samples thus were not feasible. The laboratory results for these two samples were individually determined as “positive” for CTX (P-CTX-1 equivalent) and “not detected” for P-CTX-1, -2, -3,

Table 2 Clinical information of some ciguatera patients after consumption of suspected CTX-contaminated coral reef fish samples between 2004 and 2013.a Year

Sample

Sample status

Coral reef fish involved [species name]

No of person at risk

No. of person ill

No of person hospitalized

Average incubation time (hr)

Symptom(s) reported by patient(s) Bracket ¼ no. of patients, if available

2004 2005 2005 2005 2005 2006

1 2-os 3b 4b 5b 6

SM CL CL CL CL CL

[Cheilinus undulates] [Epinephelus lanceolatus] [Lutjanus argentimaculatus] [Variola albimarginata] [Plectropomus laevis] [Lutjanus argentimaculatus]

e 1 2 4 15 3

e 1 1 4 7 3

e e 1 0 0 1

e e 4.75 10 12 10

2007 2007 2008 2008 2008 2008 2008 2008 2009 2010 2010 2010 2011 2011 2011 2011 2012 2012 2013 2013 2013

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

CL CL CL CL CL CL CL CL CL CL CL CL CL CL CL SM CL CL CL CL CL

Undefined grouper [Lutjanus stellatus] [Lates calcarifer] [Variola albimarginata] [Plectropomus leopardus] [Epinephelus bleekeri] [Lutjanus bohar] [Epinephelus fuscoguttatus] [Epinephelus polyphekadion] Undefined grouper [Plectropomus leopardus] [Lutjanus bohar] [Plectropomus leopardus] [Lutjanus malabaricus] Undefined grouper Undefined grouper [Lutjanus bohar] [Lutjanus bohar] Undefined grouper [Plectropomus leopardus] [Lutjanus bohar]

e 7 4 2 2 e 8 e e 2 e 1 3 e 2 e e 6 3 2 21

e 2 3 2 1 e 8 e e 2 e 1 3 e 2 e e 5 2 2 ~15

e 0 3 1 1 e 8 e e 2 e 1 0 e 1 e e 0 0 2 8

e 9 1.5 0.5e6 24.5 e 3.5 e e 7.5 e 3 8 e 6 e e 5e12.5 9.5 e 7.45

e N, Rhc Di, Fn, Ln, Na Ap(2), Di(2), Dz(2), Ln(4), We(4) Ap(1), Di(3), Ln(7) Ap (1), Di (1), Dz (2), Fr(1), Ln (3), Na (2), Vo(1), We(3) e Ap(1), Di(2), Pn(2) Ap, Di, Dz Ln. Rhc, Vo Ap, Di, Ln, Pn, We, Vo Ln e Ln, Pn e AP, Ln, Pn AP, Di, Dz Fn, Ln, Na, Pn e Di, N (palm, bilateral), toothache Ap, Di, Ln, Na, Rhc, Vo, We e Ap(2), Di(2), Ln(1), Pn(1) e e Ap, Ln, Di, Dz, Na N(2), We(2) N Di, Pn, N, Rhc, Vo

Keys: CL: clinically linked and food remnant from patient(s); SM: sample collected from market; sample 2-os: a raw coral reef fish flesh donated by a patient from overseas who had suffered from CFP after eating portions of the flesh. Ap: abdominal pain; Di: diarrhoea; Dz: dizziness; Fe: fever; Fn: facial numbness; Pn: perioral numbness; Ln: limb numbness; N: numbness; Na: nausea; Rhc: reversal of heat-cold sensation; Vo: vomiting; We: weakness. a For further information on clinical symptoms of ciguatera patients in Hong Kong between 2004 and 2005, please refer to Centre for Health Protection (2005) and Wong et al. (2008). b Clinical information was adopted from Wong et al. (2008).

100

C.-K. Wong et al. / Toxicon 86 (2014) 96e106

respectively, in MBA and LC-MS/MS analysis. A presumably suspected P-CTX peak (SPP) with its retention time at around 6.2 min, however, was identified in one of the samples (sample 11). The presence of other P-CTXs could not be detected by the machine using the pre-determined S/N ratio criteria. For samples 7 and 12, though original samples were used, CTX or its similar congener was found in neither MBA nor LC-MS/MS tests. In addition, clinical information of patients for these four samples was also not available for additional investigations.

calcarifer, Lutjanus malabaricus, Lutjanus stellatus, Plectropomus laevis, two samples each of Lutjanus argentimaculatus, Variola albimarginata, four Plectropomus leopardus, five Lutjanus bohar and five undefined groupers. According to the fish remnants provided by CFP patients for over ten years, P. leopardus (leopard coral grouper) (16.7%) and L. bohar (two-spot red snapper) (20.8%) were the commonest reef fish implicated in the poisonings. 3.3. Toxicity analysis of CTX-positive samples by LC-MS/MS and MBA

3.1. Clinical analysis on CFP patients With reference to clinical picture from CFP patients in the past ten years (Table 2), it showed that neurological and gastrointestinal disorders including numbness, diarrhoea and abdominal pain were the commonest symptoms of CFP in which neurological illnesses (over 50% in most cases) were predominant in each cluster of patients. Overall, the average incubation period ranged between 0.5 and 24.5 h. Among the 88 patients who sought medical attention with suspected CFP, 33% were hospitalized, similar to the previous findings in Hong Kong (Centre for Health Protection, 2005; Wong et al., 2008). 3.2. Coral reef fish species involved in CFP The coral reef fish involved in CFP included one sample each of Cheilinus undulates, Epinephelus bleekeri, Epinephelus fuscoguttatus, Epinephelus polyphekadion, Lates

In twenty-two field samples identified as CTX positive by LC-MS/MS, P-CTX-1, -2 and -3 were the most commonly observed toxins congeners found in the samples. Most samples (73%) were found to have a SPP at retention time around 6.2 min (Fig. 1). Its ion fragments produced by MRM included m/z 1128.7/1075.7, 1128.7/ 1057.7, 1128.7/1093.7; 1112.7/1077.6, 1112.7/1059.6 and 1112.7/1041.6, with 1128.7/1075.7 (P-CTX-1a) being the most abundant when compared with others. In general, the concentrations of P-CTX-1 in the retrospective ciguateric fish samples ranged from 0.02 to 1.43 mg/kg fish flesh, while the range of P-CTX-2 and -3 concentrations were 0.07e0.9 mg/kg and 0.07e0.51 mg/kg, respectively. For positive control sample 2-os, the concentrations of P-CTX1, -2 and -3 were 0.02, 0.25 and 0.21 mg/kg fish flesh, respectively. By comparing toxicities of P-CTXs derived by LC-MS/MS with corresponding total toxicities of CTXs (PCTX-1 equivalent) determined by MBA (Table 3), in

Fig. 1. Total Ion Current (TIC) chromatogram of sample 27 showing the presence of P-CTX-1 (~4.9 min), presumably suspected P-CTX peak (~6.2 min), P-CTX-2 (~6.5 min), P-CTX-3 (~7.1 min) and their corresponding fragment ions chromatograms from Multiple Reaction Monitoring (MRM).

C.-K. Wong et al. / Toxicon 86 (2014) 96e106

101

Table 3 Comparison of toxicities of P-CTXs (MU/kg, P-CTX-1, -2 and -3) derived by LC-MS/MS analysis and total toxicities of CTXs (MU/kg, P-CTX-1 equivalent) determined by mouse bioassay (MBA). Year Sample Sample status

Coral reef fish involved [species name]

P-CTX-1 LC-MS result (MU/kg)

P-CTX-2 LC-MS result (MU/kg)

P-CTX-3 LC-MS result (MU/kg)

Presumably Overall (P-CTX-1, -2 & -3) Total toxicity suspected toxicity (MU/kg) by MBA P-CTX peak derived by LCMS/MS (MU/kg, P-CTX-1 equ.) at ~6.2 min

2004 2005 2005 2005 2005

[Cheilinus undulates] [Epinephelus lanceolatus] [Lutjanus argentimaculatus] [Variola albimarginata] [Plectropomus laevis]

197 3.75 68.0 (95.8)d 10.8 8.05

51.0 13.4 10.2 (13.4)d 10.2 8.06

6.98 11.3 9.67 (18.8)d 4.30 8.06

Yes e e Yes Yes

255 28.5 87.9 (128)d 25.3 24.2

450 47e94 223 160 53e105

[Lutjanus argentimaculatus]

143

58.0

27.4

Yes

228

564

Undefined grouper

ND

ND

ND

e

ND

ND

Lutjanus stellatus] [Lates calcarifer] [Variola albimarginata] [Plectropomus leopardus] [Epinephelus bleekeri] [Lutjanus bohar] [Epinephelus fuscoguttatus] [Epinephelus polyphekadion] Undefined grouper [Plectropomus leopardus] [Lutjanus bohar] [Plectropomus leopardus] [Lutjanus malabaricus] Undefined grouper Undefined grouper [Lutjanus bohar] [Lutjanus bohar] Undefined grouper [Plectropomus leopardus] [Lutjanus bohar]

4.05 101 43.4 ND ND 12.3 ND 4.42 95.5 (73.5)e 16.3 72.5 15.9 9.95 (9.79)f 66.1 24.7 (22.2)f 19.8 23.3 84.4 31.6 255

4.83 15.6 23.6 ND ND 8.59 ND 9.67 25.2 (17.7)e 25.2 22.0 48.9 18.3 (10.7)f 17.7 5.91 (5.37)f 3.76 29.5 16.6 11.1 25.8

4.83 3.76 16.1 ND ND 6.98 ND 7.52 22.0 (4.83)e 12.9 17.2 12.9 15.6 (6.4)f 9.67 3.76 (4.30)f 3.76 14.5 12.9 10.2 13.4

Yes Yes Yes Yes e e e Yes Yes Yes Yes e e Yes Yes e Yes e Yes Yes

13.7 120 83.1 ND ND 27.8 ND 21.6 143 (96.1)e 54.4 112 77.7 43.8(27.0)f 93.4 34.4 (31.9)f 27.3 67.3 114 53.1 294

<59 186 83 <51 ND <57 <57 148 141 <49 153 <35 101 198 <45 167 55.1 110 <69 270

1 2-os 3 4 5

2006 6 2007 7 2007 2008 2008 2008 2008 2008 2008 2009 2010 2010 2010 2011 2011 2011 2011 2012 2012 2013 2013 2013

8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

SM CL CL CLa CL (cooked) CL (cooked) CL (cooked) CL CL CL CLb CL CL CLb CL CL CL CL CL CL CL SMc CL CL CL CL CL

LLOD: 0.025 ppb in fish flesh; LLOQ: 0.1 ppb in fish flesh. Keys: CL: clinically linked and food remnant from patient(s); SM: sample collected from market; sample 2-os: a raw coral reef fish flesh donated by a patient from overseas who had suffered from CFP after eating portion of the flesh; ND: Not detected. a Few minced fish flesh obtained together with fat and fish skin attached. b Original sample was used up; another fish of the same batch (i.e. More than one fish samples were collected from the same source and put into a sample collection bag, they were then given a sample registration number for laboratory analysis) was taken instead for LCMS/MS analysis. c Fish ether extract for mouse bioassay was extracted for LC-MS/MS analysis. d Repeat analysis of sample from original vial after 6 months. e Result from 2nd extraction. f Repeat analysis of sample from original vial.

addition to the highly potent P-CTX-1, both P-CTX-2 and -3 also contributed significantly in the overall LC-MS/MS derived toxicity. In some cases, the concentrations of PCTX-2 and -3 were significantly higher than those of PCTX-1. Their peak height profiles were conspicuously present at around 6.5 and 7.1 min, respectively in the chromatograms. Nevertheless, as the equivalent toxicities (i.p. LD50 in mice) of P-CTX-2 and -3 were about 3-fold less potent than that of P-CTX-1 (EFSA Panel on Contaminants in the food chain, 2010), their toxicity levels were significantly lower as compared with the most toxic P-CTX-1. On the other hand, it should be noted that unambiguous deviations were found between results of individual PCTX (i.e. P-CTX-2 and -3) toxicity levels derived by LC-MS/ MS and the corresponding toxicities determined by MBA (Table 3). Regression analysis (Fig. 2) showed satisfactory linearity when comparing results between toxicities of PCTX-1 alone derived by LC-MS/MS and their corresponding total toxicity levels (P-CTX-1 equivalent) determined

by MBA (r2 ¼ 0.56; Student's t-test p < 0.005). In addition, it was obvious that including the overall LC-MS/MSderived toxicities (P-CTX-1, -2 and -3) for calculation could slightly improve both the slope of the regression line (from 1.49 to 1.32) and the correlation coefficient (from r2 ¼ 0.56 to r2 ¼ 0.61; Student's t-test from p < 0.005 to p < 0.05). Linear regression studies between toxicities of P-CTX-2 and -3 calculated by LC-MS/MS and MBA, however, showed significant differences (Student's t-test p < 0.001 for both P-CTX-2 and -3) between the two methods, where the correlation coefficient values (r2 ¼ 0.36 and 0.14, respectively for P-CTX-2 and -3) between the two methods were unsatisfactory as well. Using Spearman's rank correlation analysis, positive correlation coefficient results (r ¼ 0.669; p < 0.05 for P-CTX-1, and r ¼ 0.593; p < 0.05 for P-CTX-1, -2 and -3) were found between LC-MS/MS and MBA. Similar to the results of previous statistical analysis, no significant relationship was found (r ¼ 0.249; p > 0.05 for P-CTX-2, r ¼ 0.149;

102

C.-K. Wong et al. / Toxicon 86 (2014) 96e106

Fig. 2. The relationship between the total CTX toxicity (MU/kg, P-CTX-1 equivalent) determined by MBA and the overall toxicity of P-CTX-1, -2, -3, and their individual P-CTX toxicities derived by LC-MS/MS (n ¼ 31).

p > 0.05 for P-CTX-3) when comparing toxicities of P-CTX2/P-CTX-3 derived by LC-MS/MS with their corresponding toxicity levels measured by MBA. 3.4. Toxicity analysis of CTX-negative samples by LC-MS/MS and MBA As regards the retrospective analysis of CTX-negative coral reef fish samples determined by MBA, 41 samples including different types of reef fish species (mainly nonoily species) Cheilinus undulates, E. bleekeri, Epinephelus bruneus, Epinephelus coioides, E. fuscoguttatus, E. lanceolatus, E. polyphekadion, L. argentimaculatus, Plectropomus areolatus, P. leopardus, Trachinotus blochii, V. albimarginata and some undefined groupers were tested with LC-MS/MS. Overall, none of them was found to have P-CTX-1, -2, -3 or suspected P-CTX.

4. Discussion CFP is also one of the food-borne diseases in Hong Kong with increasing public health concern in the past years (Choi and Wong, 1994; Wong et al., 2005). The number of CFP outbreaks/incidents showed an increasing trend from the 1980s to the 1990s (Sadovy, 1997). Although no regulatory limit has yet been set in the European Union (EU) for CTXs, measures may be adopted to prevent CTX contaminated fishery products from entering the market (EFSA Panel on Contaminants in the food chain, 2010). It was suggested that consuming fish weighing under 1 kg may reduce the risk of CFP (Lucas et al., 1997). On the other hand, controlling the import of high-risk coral reef fish or fish species from ciguatera endemic areas would also be a possible means (Lewis, 2006). With advanced mass spectrometric techniques and improved toxin extraction

C.-K. Wong et al. / Toxicon 86 (2014) 96e106

methods in the past decades, the detection of CTXs at or even below the clinically relevant level of 0.1 ppb is possible (Lewis et al., 2009; Stewart et al., 2010a), enabling the precise identification and quantification of the toxins in contaminated fish samples. In the present study, we used LC-MS/MS to analyse CTXpositive and -negative samples previously determined by MBA. This is also the first study to investigate the possible classes of CTXs implicated in over the past ten years in Hong Kong. Based on chemical analysis from the available retrospective samples, the results showed that P-CTXs (PCTX-1, -2 and -3) were the most commonly observed ichthyosarcotoxins. However, a major limitation of this study is that only three of many ciguatoxin congeners were tested while only 12 out of 27 samples comprised the majority of the toxins (i.e. the 12 samples contained P-CTX-1, -2, -3 together with presumptive P-CTX substance, and of which the clinical information of ciguatera patients was available). Lewis and Sellin (1992) reported that P-CTX-1, -2 and -3 were common toxins found in coral reef fish collected from the western Pacific Ocean. The toxin profiles obtained in our CTX contaminated samples showed dominance of PCTX-1, with significant amounts of P-CTX-2 and -3 in some samples, indicating diversity of toxin profiles in the field samples. Stewart et al. (2010a, b) and Mak et al. (2013) also documented the prevalence of P-CTX-2 and -3 in ciguatoxic fish of Pacific regions. In the past years, multiple structurally related P-CTX isomers and congeners were determined from CTX contaminated fish and Gambierdiscus spp., where some toxins (e.g. P-CTX-3C, P-CTX-4A and 49-epi P-CTX-3C) isolated from G. toxicus could also be found in ciguateric fish (Lewis and Jones, 1997; Satake et al., 1998; Yasumoto et al., 2000; EFSA Panel on Contaminants in the food chain, 2010). The study from Yogi et al. (2011) showed the presence of other classes of P-CTXs (e.g. P-CTX-3C type toxins) in Pacific fish. In particular, only P-CTX-3C type toxins were identified from fish samples collected from Miyazaki, Japan, while only P-CTX-1B type toxins were found in fish specimens of Okinawa. On the contrary, both types of toxins were determined in a red snapper collected from Minamitorishima Island. This study highlighted distinct regional and species profiles of CTX characteristics in fish from the Pacific region. Previous studies suggested that some precursor toxins were oxidized in fish through the food chain (Satake et al., 1998; Yasumoto et al., 2000). The identification of oxidized toxins (e.g. 51-hydroxy PCTX-3C and 52-epi-54-deoxy P-CTX-1B) in both ciguateric fish and G. toxicus in recent research by Yogi et al. (2011), however, proved that the toxins existed originally in the causative microalgae from particular region. With reference to characterization of CTXs in fish collected in the central Pacific Ocean and Caribbean Sea (Lewis and Holmes, 1993; Lewis et al., 1994; Lewis and Jones, 1997; Vernoux and Lewis, 1997), it has been shown that Pacific- and Caribbean-CTXs are region-specific toxins. It was also suggested that ciguatera causative species G. toxicus in the western and central Pacific Ocean do not produce detectable levels of C-CTX-1, -2 and certain P-CTXs (e.g. P-CTX-2A1, -2B1, -2C1, -3A, -3C) or their precursors (Lewis and Jones, 1997). As regards I-CTXs, they are considered as new toxins different from P-CTXs and C-

103

CTXs, though both I-CTX-1 and -2 have the same molecular mass as C-CTX-1 (Hamilton et al., 2002a, 2002b; FAO, 2004). As a result, even though the standards of C-CTX and I-CTX were not available for investigating the possible presence of other non-Pacific CTXs, the present information still implies that the probable source of ciguateric coral reef fish during the past 10 years in Hong Kong was Pacific Ocean areas. The commonest coral reef fish implicated in CFP were leopard coral groupers and two-spot red snappers, in line with the previous reports (Choi and Wong, 1994; Lehane and Lewis, 2000; Centre for Health Protection, 2005; Wong et al., 2005). Regarding the suspected P-CTX peak (SPP) found at ~6.2 min retention time, it was a rather common peak identified in the chromatograms of some samples, probably owing to the presence of another unknown P-CTX congener/isomer which commonly existed in the field samples. Basically, the occurrence of this peak was normally accompanied by the presence of other common P-CTXs such as P-CTX-1, -2 and -3, except sample 11 from which the suspected P-CTX peak was solely identified. This presumptive P-CTX substance did not appear to be derived from inter-conversion nor differential degradation from the three P-CTX-1, -2 and -3 during or after prolonged storage of samples, since the corresponding peak could also be identified in the chromatograms from some recent samples in 2012 and 2013, where sample extraction and analysis were completed within a week after receipt of samples. Nonetheless, no additional test was available for its confirmation. Further study is necessary to clarify the chemical structure, potency and clinical significance of this suspected P-CTX. In general, CTXs are stable chemicals resistant to normal cooking temperatures, extended periods of storage in freezer and even storage in certain chemical solvents (Bagnis, 1993; Lewis et al., 1998; Hamilton et al., 2002a; FAO, 2004; Friedman et al., 2008; Abraham et al., 2012; Robertson et al., 2014). However, it has been shown that C-CTXs are more susceptible to inter-conversion by high temperature and extended periods of storage (Lewis et al., 1998; Pottier et al., 2002). According to Nukina et al. (1984), the interconversion of CTX might occur in certain species of fish or be induced by alumina chromatography columns. In the present study, there was no evidence of inter-conversion of CTXs in samples stored in 80  C. Regarding the analysis of multiple fish samples from the same batch, the discrepancies in two samples cast doubt on the validity of the “batch” concept for testing. In terms of clinical findings, neurological symptoms were predominant. It has been documented that among the three classes of CTX, neurological symptoms were more pronounced in Pacific- and Indian-CTX poisonings, while gastrointestinal disorders were more prominent for Caribbean-CTX intoxication (Lawrence et al., 1980; Lehane and Lewis, 2000; De Fouw et al., 2001; Baumann et al., 2010), in line with findings from the current study. Between Pacific and Indian CFP, Indian-CTX poisoning is mainly characterized by hallucinogenic symptoms and changes of mental status (Quod and Turquet, 1996; Lewis, 2006; Friedman et al., 2008). The occurrence of IndianCTX poisoning is also less common than Pacific and Caribbean ciguatera poisonings (Quod and Turquet, 1996;

104

C.-K. Wong et al. / Toxicon 86 (2014) 96e106

Pottier et al., 2001; Hamilton et al., 2002a, b). Overall, the clinical presentations of CFP conformed well with the laboratory confirmation of CTXs in food remnant samples, indicating the reliability of clinical diagnosis for CFP. According to this study, LC-MS/MS could allow the detection of trace amount of CTXs at levels below the clinically relevant limit, 0.1 ppb of P-CTX-1 in fish tissues. Moreover, LC-MS/MS could also identify and quantify P-CTXs in field samples or in fish crude extracts prepared for MBA. With a recovery efficiency of 51.6%, based on an in-house study on 16 different types of non-oily coral fish matrices in solvent extraction, together with the favourable LC-MS/MS detection limit (LLOD and LLOQ at 0.25 and 1.0 ppb respectively; i.e. 0.025 and 0.1 ppb respectively in fish flesh, data not shown), LC-MS/MS is expected to be sufficiently sensitive for toxin identification and quantification. The recovery rate obtained here was also comparable to a similar study (51.3%) by Stewart et al. (2010a). Nevertheless, it is important to recognize that the variations of water content in fish flesh might significantly affect the extraction efficiency of P-CTX-1 from samples (Lewis et al., 2009; Stewart et al., 2010a), varying the subsequent final recovery of the toxins. Therefore, further research and improvement on toxin extraction methods are recommended. With reference to the comparison of LC-MS/MS and MBA, toxicities of overall P-CTXs (P-CTX-1, -2 and -3) and PCTX-1 alone derived from LC-MS/MS correlated well with corresponding toxicities determined by MBA, particularly when adopting the overall P-CTX toxicity derived by LCMS/MS for comparison. It is obvious that the calculation of P-CTX-1 alone would under-estimate the overall toxicities of the samples, especially in the presence of high concentrations of P-CTX-2 and -3. Nevertheless, P-CTX-1 is the most potent toxin known when compared with other Pacific-CTXs, Caribbean-CTXs (~10% potency of P-CTX-1) and Indian-CTXs (~20e60% potency of P-CTX-1) (Lewis, 2001; Hamilton et al., 2002b; Lewis, 2006; EFSA Panel on Contaminants in the food chain, 2010). As a result, P-CTX1 is also deemed an appropriate biochemical indicator of the toxicity of samples in the field. Ciguateric fish cannot be distinguished based on taste, odour or appearance. The most potent P-CTX-1 is suggested to induce human illness commonly at levels between 0.1 and 5 ppb (Lewis et al., 1991, 1999; Lehane, 1999; Lehane and Lewis, 2000; Pierce and Kirkpatrick, 2001; Lewis, 2006; EFSA Panel on Contaminants in the food chain, 2010). There have been documentations on fatalities (between <0.1 and 7%) due to consumption of CTX contaminated fish (Centers for Disease Control, 1993; Lewis, 2000; Friedman et al., 2008; Morrison et al., 2008; Caillaud et al., 2010). Development of a validated laboratory analysis protocol for CTXs is vital for public health and clinical management of this foodborne disease. However, one of the major obstacles is lack of available certified reference standards or related toxin surrogates for method development and validation. Moreover, no ongoing proficiency test is in place for CTXs (EFSA Panel on Contaminants in the food chain, 2010). In respect of routine analysis of reef fish samples for CTXs, the adoption of a two-tiered approach, chemical analysis (LC-MS/MS) for first-line screening of toxic fish coupled with biological assay

(MBA) for confirmation, was considered an appropriate strategy, minimizing the use of laboratory mice and at the same time providing a rapid, sensitive and reliable assay for screening potentially toxic fish. In addition, toxin extraction (8e10 samples each time) and LC-MS/MS analysis (12 min turnaround time per sample) can normally be completed within 3 days, comparatively more efficient than extraction and analysis by MBA. In cases where toxin profiles and the chemical identity of the CTXs are not required, a rapid, high-throughput and highly sensitive in vitro cytotoxicity assay (the neuro-2a mouse neuroblastoma cell assay) is considered useful, enabling testing of multiple samples on 96-well plates (Manger et al., 1993, 1995; Van Dolah et al., 1994; Dickey, 2008; Caillaud et al., 2010; Dickey and Plakas, 2010). In addition, the assay can be calibrated using very small amounts of CTX standards and can be more sensitive and less expensive than LC-MS/MS. With the progress in sophisticated mass spectrometric technologies enabling more sensitive and cost effective laboratory testing of CTX, exploration of the diversity of P-CTX profiles and full range of CTXs in varieties of fish samples in different geographic areas would be achievable, contributing to formulating biotransformation mechanisms and understanding the epidemiology of this potent marine biotoxin. Conflicts of interest The authors declare that they have no conflicts of interest. Declarations The authors declare that all commercial products, trade names or materials mentioned in this article do not represent any suggestion, approval or endorsement for use by authors' department, division and organization. All experimental mice used in the bioassay were sacrificed after experiments in compliance with Code of Practice Care and Use of Animals for Experimental Purposes, Agriculture, Fisheries and Conservation Department of Hong Kong, China. Ethical statement All experimental mice used in the bioassay were sacrificed after experiments in compliance with Code of Practice Care and Use of Animals for Experimental Purposes, Agriculture, Fisheries and Conservation Department of Hong Kong, China. Acknowledgements The authors are indebted to all the staff in the Food and Water Laboratory, Public Health Laboratory Services Branch, Centre for Health Protection, Department of Health, for their help and support in this study. The authors are particularly grateful to the anonymous donor for supplying the ciguatoxin-contaminated coral fish samples and the staff of the Food and Environmental Hygiene Department for collecting the coral reef fish samples.

C.-K. Wong et al. / Toxicon 86 (2014) 96e106

Transparency document Transparency document related to this article can be found online at http://dx.doi.org/10.1016/j.toxicon.2014.05. 006. References Abraham, A., Jester, E.L.E., Granade, H.R., Plakas, S.M., Dickey, R.W., 2012. Caribbean ciguatoxin profile in raw and cooked fish implicated in ciguatera. Food Chem. 131, 192e198. Bagnis, R., 1993. In: Falconer, I. (Ed.), Algal Toxins in Seafood and Drinking Water. Academic Press, London, pp. 105e115. Baumann, F., Bourrat, M.B., Pauillac, S., 2010. Prevalence, symptoms and chronicity of ciguatera in New Caledonia: results from an adult population survey conducted in Noumea during 2005. Toxicon 56, 662e667. Caillaud, A., de la Iglesia, P., Darius, H.T., Pauillac, S., Aligizaki, K., Fraga, S., ne, J., 2010. Update on methodologies available for Chinain, M., Dioge ciguatoxin determination: perspectives to confront the onset of ciguatera fish poisoning in Europe. Mar. Drugs 8, 1838e1907. Centers for Disease Control, 1993. Ciguatera fish poisoning: Florida 1991. MMWR Morb. Mortal. Wkly. Rep. 42, 417e418. Centre for Health Protection, 2005. Communicable Diseases Watch Volume 2, Number 13, Weeks 25e26 (June 12e25, 2005). Department of Health, Government of Hong Kong Special Administrative Region. Choi, S.M.Y., Wong, M.M.H., 1994. Epidemiology of ciguatera poisoning in Hong Kong. Public Health Epidemiol. Represent. 3, 12e14. Daranas, A.H., Norte, M., Fern andez, J.J., 2001. Toxic marine microalgae. Toxicon 39, 1101e1132. De Fouw, J.C., Van Egmond, H.P., Speijers, G.J.A., 2001. Ciguatera Fish Poisoning: a Review. National Institute of Public Health and the Environment. RIVM report 388802021. Department of Health, 2002. Topical Health Report no. 2: Statistic on Infectious Diseases in Hong Kong 1946e2001. Government of Hong Kong Special Administrative Region. Dickey, R.W., 2008. Ciguatera toxins: chemistry, toxicology and detection. In: Botana, L.M. (Ed.), Seafood and Freshwater Toxins: Pharmacology, Physiology, and Detection, second ed. CRC Press-Taylor & Francis, Boca Raton, FL, USA, pp. 479e500. Dickey, R.W., Plakas, S.M., 2010. Ciguatera: a public health perspective. Toxicon 56, 123e136. EFSA Panel on Contaminants in the food chain, 2010. Scientific opinion on marine biotoxins in shellfish e emerging toxins: ciguatoxin group. EFSA J. 8, 1627e1665. FAO, 2004. Marine Biotoxins, FAO Food and Nutrition Paper 80. Food and Agriculture Organization of the United Nations, Rome, 278 pp. Friedman, M.A., Fleming, L.E., Fernandez, M., Bienfang, P., Schrank, K., Dickey, R., Bottein, M.-Y., Backer, L., Ayyar, R., Weisman, R., Watkins, S., Granade, R., Reich, A., 2008. Ciguatera fish poisoning: treatment, prevention and management. Mar. Drugs 6, 465e470. Halstead, B.W., 1967. Poisonous and Venomous Marine Animals of the World, vol. 2. U.S. Government Printing Office, Washington, D. C. Halstead, B.W., 1988. Poisonous and Venomous Marine Animals of the World, Second Revised Edition. The Darwin Press, Inc., Princeton, New Jersey. Hamilton, B., Hurbungs, M., Jones, A., Lewis, R.J., 2002a. Multiple ciguatoxins present in Indian Ocean reef fish. Toxicon 40, 1347e1353. Hamilton, B., Hurbungs, M., Vernoux, J.-P., Jones, A., Lewis, R.J., 2002b. Isolation and characterisation of Indian Ocean ciguatoxin. Toxicon 40, 685e693. Keynan, Y., Pottesman, I., 2004. Neurological symptoms in a traveler returning from Central America. J. Intern. Med 256, 174e175. Lawrence, D.N., Enríquez, M.B., Lumish, R.M., Maceo, A., 1980. Ciguatera fish poisoning in Miami. J. Am. Med. Assoc. 244, 154e258. Lehane, L., 1999. Ciguatera Fish Poisoning: a Review in a Risk-Assessment Framework. National Office of Animal and Plant Health, Agriculture, Fisheries and Forestry, Australia. Lehane, L., 2000. Ciguatera update. Med. J. Aust. 172, 176e179. Lehane, L., Lewis, R.J., 2000. Ciguatera: a recent advances but the risk remains. Int. J. Food Microbiol. 61, 91e125. Lewis, R.J., 2000. Ciguatera management. SPC Live Reef Fish Inf. Bull. 7, 11e13. Lewis, R.J., 2001. The changing face of ciguatera. Toxicon 39, 97e106. Lewis, R.J., 2006. Ciguatera: Australian perspectives on a global problem. Toxicon 48, 799e809. Lewis, R.J., Jones, A., 1997. Characterization of ciguatoxins and ciguatoxin congeners present in ciguateric fish by gradient reverse-phase high-

105

performance liquid chromatography/mass spectrometry. Toxicon 35, 159e168. Lewis, R.J., Holmes, M.J., 1993. Origin and transfer of toxins involved in ciguatera. Comp. Biochem. Physiol. 106C, 615e628. Lewis, R.J., Sellin, M., Poli, M.A., Norton, R.S., Macleod, J.K., Sheil, M.M., 1991. Purification and characterization of ciguatoxins from moray eel (Lycodontis javanicus, Muraenidae). Toxicon 29, 1115e1127. Lewis, R.J., Holmes, M.J., Alewood, P.F., Jones, A., 1994. Ionspray mass spectrometry of ciguatoxin-1, maitotoxin-2, and -3, and related marine polyether toxins. Nat. Toxins 2, 56e63. Lewis, R.J., Sellin, M., 1992. Multiple ciguatoxins in the flesh of fish. Toxicon 30, 915e919. Lewis, R.J., Vernoux, J.P., Brereton, I.M., 1998. Structure of Caribbean ciguatoxins isolated from Caranx latus. J. Am. Chem. Soc. 120, 5914e5920. Lewis, R.J., Jones, A., Vernoux, J.-P., 1999. HPLC/tandemelectrospray mass spectrometry for the determination of sub-ppb levels of Pacific and Caribbean ciguatoxins in crude extracts of fish. Anal. Chem. 71, 247e250. Lewis, R.J., 2003. Detection of toxins associated with ciguatera fish poisoning. In: Hallegraeff, G.M., Anderson, D.M., Cembella, A.D. (Eds.), Manual on Harmful Marine Microalgae. IOC Manuals and Guides No. 33. UNESCO, France, pp. 267e277. Lewis, R.J., Yang, A., Jones, A., 2009. Rapid extraction combined with LCtandem mass spectrometry (CREM-LC/MS/MS) for the determination of ciguatoxins in ciguateric fish flesh. Toxicon 54, 62e66. Lucas, R.E., Lewis, R.J., Taylor, J.M., 1997. Pacific ciguatoxin-1 associated with a large common-source outbreak of ciguatera in East Arnhem Land. Nat. Toxins 5, 136e140. Mak, Y.L., Wu, J.J., Chan, W.H., Murphy, M.B., Lam, J.C.W., Chan, L.L., Lam, P.K.S., 2013. Simultaneous quantification of Pacific ciguatoxins in fish blood using liquid chromatography-tandem mass spectrometry. Anal. Bioanal. Chem. 405, 3331e3340. Manger, R.L., Leja, L.S., Lee, S.Y., Hungerford, J.M., Wekell, M.M., 1993. Tetrazolium-based cell bioassay for neurotoxins active on voltagesensitive sodium channels: semiautomated assay for saxitoxin, brevetoxin, and ciguatoxin. Anal. Biochem 214, 190e194. Manger, R.L., Leja, L.S., Lee, S.Y., Hungerford, J.M., Hokama, Y., Dickey, R.W., Granade, H.R., Lewis, R.J., Yasumoto, T., Wekell, M.M., 1995. Detection of sodium channel effectors: directed cytotoxicity assays of purified ciguatoxins, brevetoxins, saxitoxin and seafood extracts. J. AOAC Int. 78, 521e527. Morrison, K., Aguiar Prieto, P., Castro Dominguez, A., Waltner-Toews, D., FitzGibbon, J., 2008. Ciguatera fish poisoning in La Habana, Cuba: a study of local social-ecological resilience. EcoHealth 5, 346e359. Nukina, M., Koyanagi, L.M., Scheuer, P.J., 1984. Two interchangeable forms of ciguatoxin. Toxicon 22, 169e176. Palafox, N.A., Jain, L.G., Pinano, A.Z., Gulick, T.M., Williams, R.K., Schatz, I.J., 1988. Successful treatment of ciguatera fish poisoning with intravenous mannitol. J. Am. Med. Assoc. 259, 2740e2742. Palafox, N.A., Buenconsejo-Lum, L.E., 2001. Ciguatera fish poisoning: review of clinical manifestations. J. Toxicol. Toxin Rev. 20, 141e160. Pearn, J., 2001. Neurology of ciguatera. J. Neurol. Neurosurg. Psychiatry 70, 4e8. Pierce, R.H., Kirkpatrick, G.J., 2001. Innovative techniques for harmful algal toxin analysis. Environ. Toxicol. Chem. 20, 107e114. Pottier, I., Vernoux, J.P., Lewis, R.J., 2001. Ciguatera fish poisoning in the Caribbean islands and Western Atlantic. Rev. Environ. Contam. Toxicol. 168, 99e141. Pottier, I., Vernoux, J.P., Jones, A., Lewis, R.J., 2002. Characterisation of multiple Caribbean ciguatoxins and congeners in individual specimens of horse-eye jack (Caranx latus) by high-performance liquid chromatography/mass spectrometry. Toxicon 40, 929e939. Quod, J.P., Turquet, J., 1996. Ciguatera in Reunion Island (SW Indian Ocean): epidemiology and clinical patterns. Toxicon 34, 779e785. Robertson, A., Garcia, A.C., Flores Quintana, H.A., Smith, T.B., Castillo II, B.F., Reale-Munroe, K., Gulli, J.A., Olsen, D.A., HooeRollman, J.I., Jester, E.L.E., Klimek, B.J., Plakas, S.M., 2014. Invasive lionfish (Pterois volitans): a potential human health threat for ciguatera fish poisoning in tropical waters. Mar. Drugs 12, 88e97. Sadovy, Y., 1997. Ciguatera hits Hong Kong live reef fish trade. SPC Fish. Newslett. 83, 26e28. Satake, M., Fukui, M., Legrand, A.eM., Cruchet, P., Yasumoto, T., 1998. Isolation and structures of new ciguatoxin analogs, 2, 3dihydroxyCTX3C and 51-hydroxyCTX3C, accumulated in tropical reef fish. Tetrahedron Lett. 39, 1197e1198. Scheuer, P.J., 1994. Ciguatera and its off-shoot-chance encounters en route to a molecular structure. Tetrahedron 50, 3e18. Stewart, I., Eaglesham, G.K., Poole, S., Graham, G., Paulo, C., Wickramasinghe, W., Sadler, R., Shaw, G.R., 2010a. Establishing a

106

C.-K. Wong et al. / Toxicon 86 (2014) 96e106

public health analytical service based on chemical methods for detecting and quantifying Pacific ciguatoxin in fish samples. Toxicon 56, 804e812. Stewart, I., Lewis, R.J., Eaglesham, G.K., Graham, G.C., Poole, S., Craig, S.B., 2010b. Emerging tropical diseases in Australia. Part 2. Ciguatera fish poisoning. Ann. Trop. Med. Parasitol. 104, 557e571. Sumner, J., Ross, T., Ababouch, L., 2004. Application of Risk Assessment in the Fish Industry, FAO Fisheries Technical Paper 442. Food and Agriculture Organization of the United Nations, Rome, Italy. Van Dolah, F.M., Finley, E.L., Haynes, B.L., Doucette, G.J., Moeller, P.D., Ramsdell, J.S., 1994. Development of rapid and sensitive high throughput pharmacologic assays for marine phycotoxins. Nat. Toxins 2, 189e196. Vernoux, J.P., Lewis, R.J., 1997. Isolation and characterization of Caribbean ciguatoxins from the horse-eye jack (Caranx latus). Toxicon 35, 889e900.

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, 563e571. Wong, C.K., Hung, P., Lee, K.L.H., Mok, T., Chung, T., Kam, K.M., 2008. Features of ciguatera fish poisoning cases in Hong Kong 2004e2007. Biomed. Environ. Sci. 21, 521e527. Wong, C.K., Hung, P., Lee, K.L.H., Kam, K.M., 2009. Solid-phase extraction clean-up of ciguatoxin-contaminated coral fish extracts for use in the mouse bioassay. Food Addit. Contam. 26, 236e247. Yasumoto, T., Igarashi, T., Legrand, A.eM., Cruchet, P., Chinain, M., Fujita, T., Naoki, H., 2000. Structural elucidation of ciguatoxin congeners by fast-atom bombardment tandem mass spectrometry. J. Am. Chem. Soc. 122, 4988e4989. Yogi, K., Oshiro, N., Inafuku, Y., Hirama, M., Yasumoto, T., 2011. Detailed LC-MS/MS analysis of ciguatoxins revealing regional and species characteristics in fish and causative alga from the Pacific. Anal. Chem. 83, 8886e8891.