Shellfish Toxicity Testing

Non-Animal Test Methods for Marine Biotoxins in Shellfish
Regulatory Considerations
Mouse Bioassay
Why Is the Mouse Bioassay Still Used?

Despite its scientific, technical, and ethical limitations, the mouse bioassay (MBA) or biological method is commonly used to test for marine biotoxins in shellfish. Authorities such as the World Health Organization, European Food Safety Authority (EFSA), and US Food and Drug Administration have noted its lack of sensitivity, specificity, and precision, highlighting the need to make the transition to better test methods.1,2

In the interests of public health and ethics, the MBA should be replaced with the more specific, sensitive, and robust alternative test methods outlined here. Some fisheries have already successfully implemented these methods, and there is a need to define the path for leveraging advances in toxicology in locations where they have not yet been adopted. With proper coordination and planning, industry, regulators, and other stakeholders can identify and analyse impediments to the use of new methodologies and develop strategies to overcome them.

NON-ANIMAL TEST METHODS FOR MARINE BIOTOXINS IN SHELLFISH

Non-animal methods tend to be more specific and sensitive and less time-consuming and expensive per sample analysed. For a list of alternative test methods for diarrhoeal shellfish poisoning (DSP), amnesic shellfish poisoning (ASP), neurotoxic shellfish poisoning (NSP), and paralytic shellfish poisoning (PSP), see Stewart and McLeod and Daneshian et al.3,4 Some of the most commonly used alternatives are described below.

PSP Detection

  • Prechromatographic Oxidation and Liquid Chromatography With Fluorescence Detection (AOAC 2005.06, Lawrence Method, LC-FLD, HPLC-FLD) - Click here for details

    • The method is approved for use in Commission Regulation (EC) No 2074/2005 (as amended by Commission Regulation (EC) No 1664/2006) for mussels, cockles, razor clams, hard clams, minor clam species, scallops, native oysters, and Pacific oysters.
    • Per Commission Regulation (EU) 2017/1980, the method was adopted as the EU’s reference method on 1 January 2019.
    • This method has successfully replaced the MBA in regulatory testing in the UK and other countries since 2006. Because of its higher detection capacity, it has provided an earlier warning of shellfish toxicity.5
    • Comparisons of results from the MBA and HPLC-FLD methods have shown good correlation, except for oysters. It was found that high concentrations of zinc and potentially other metals in oysters were affecting the MBA and not the HPLC-FLD, causing the MBA to underestimate toxicity values significantly.6,7
    • This method provides a full toxin profile and can be conducted in a higher-throughput semi-quantitative screening procedure, followed by a full quantitative analysis as needed.5,8
    • To conduct modern chemical tests such as HPLC, sophisticated analytical equipment and analytical chemistry skills are necessary.9

  • High-Performance Liquid Chromatography With Post-Column Oxidation (AOAC 2011.02, HPLC PCOX) - Click here for details

    • The method is approved for use in the US National Shellfish Sanitation Program (NSSP) “Guide for the Control of Molluscan Shellfish: 2019 Revision” for clams, mussels, oysters, and scallops.
    • HPLC PCOX is more sensitive and reliable than the MBA for detecting PSP.10 Data from a Canadian Food Inspection Agency laboratory show that the switch to the PCOX resulted in an increase in food safety and in confidence in monitoring programmes, thanks to earlier warning of toxic episodes and the potential to improve future PSP monitoring programmes through analysis of chemistry-based data. Thirteen per cent (1,170) of 9,000 molluscan shellfish samples analysed had PSP toxin levels <40 μg saxitoxin (STX) equiv/100 g, which would not have been detected with MBA analysis.11
    • To conduct modern chemical tests such as HPLC, sophisticated analytical equipment and analytical chemistry skills are necessary.9 The PCOX method requires daily attention in order to ensure chromatographic conditions are appropriate as well as routine LC maintenance.11
    • Complicating the use of HPLC PCOX, there is a shortage of certified reference materials and a lack of knowledge of toxin concentration-response relationships and mixture effects of toxins.4,10

  • Receptor Binding Assay (AOAC 2011.27, RBA) - Click here for details

    • The method is approved for use in the NSSP “Guide for the Control of Molluscan Shellfish: 2019 Revision” for mussels. It is also approved for clams and scallops for screening and precautionary closure in the guide.
    • The RBA’s limit of detection is 60 to 100 μg STX equiv/kg, compared to 400 μg STX equiv/kg for the MBA.12
    • When recently assessed for oysters, the RBA was precise and sensitive. The presence of high concentrations of zinc did not affect the RBA as it does the MBA. The authors conclude that the “RBA is safer to use for the determination of PSP”.13
    • While this method does not replace the use of animals, it does not use live animals and does reduce animal use. One rat brain can be used to test 35 to 44 shellfish samples in triplicate at three dilutions,12,14 and the use of porcine membranes from abattoirs is being explored.12

  • Immunological Test Kits - Click here for details

    • The Abraxis shipboard enzyme-linked immunosorbent assay (ELISA) and the lateral flow immunosorbent assay Scotia (formerly Jellett) Rapid Test are approved for limited use in the NSSP “Guide for the Control of Molluscan Shellfish: 2019 Revision”.
    • An ELISA has the ability to detect PSP toxins with greater sensitivity than the MBA at lower levels (50 μg STX equiv/kg shellfish flesh).15 However, the toxin profile of shellfish samples must be taken into account when using an ELISA for screening purposes.15
    • While demonstrating high false positive rates, studies have shown that the Scotia Rapid Test gave no false negatives. This included a study of a wide variety of species that contained a large range of toxicities and toxin profiles.16,17

  • Liquid Chromatography With Tandem Mass Spectrometry (LC-MS/MS) - Click here for details

    • This method has been shown to be more sensitive than the MBA, RBA, and HPLC, with acceptable toxin recovery, repeatability, and within-laboratory reproducibility in a single-laboratory validation of 12 different species of bivalve shellfish from the UK and New Zealand, including a variety of mussel, oyster, clam, and scallop species.18
    • An ultrahigh-performance LC (UHPLC)–tandem MS (MS/MS) method for detecting PSP toxins and tetrodotoxin was validated in a collaborative study with 21 laboratories. The method was demonstrated to be accurate, sensitive, and reproducible using a variety of shellfish, including mussels, oysters, cockles, scallops, and clams.19
    • LC-MS/MS incorporates toxin congeners that other validated methods either cannot detect or have not been validated for.19

ASP Detection

  • High-Performance Liquid Chromatography (AOAC 991.26 for Mussels, HPLC) - Click here for details

    • This method is approved as the reference method for detection of ASP in shellfish in Commission Regulation (EC) No 2074/2005 and for use in the NSSP “Guide for the Control of Molluscan Shellfish: 2019 Revision”. However, it requires expensive equipment, trained staff, and access to toxin standards.15,20
    • The HPLC’s limit of detection for ASP is sufficiently low to detect domoic acid adequately at the concentration of 4.5 mg/kg shellfish flesh.21

  • Biosense ASP ELISA (AOAC 2006.02) - Click here for details

    • This method is approved as a screening method for detection of ASP in shellfish in Commission Regulation (EC) No 2074/2005 (as amended by Commission Regulation (EC) No 1244/2007).
    • Its limit of detection is sufficiently low to detect domoic acid adequately at the concentration of 4.5 mg/kg shellfish meat.21
    • In a study of 58 naturally contaminated shellfish samples, the Biosense ASP ELISA was more sensitive than HPLC-UV, and there were no false negative results.15 In a study of 136 shellfish samples, another ELISA, the Beacon ASP assay, returned no false positive results, and the quantitative data compared well with data from the HPLC-UV method.20
    • This method provides relatively simple, rapid, and quantifiable results. It is also more cost-effective than the current regulatory HPLC method.20

  • Lateral Flow Immunosorbent Assays (LFIAs) - Click here for details

    • Reveal® 2.0 for ASP (available from Neogen) is approved as a screening method in the NSSP “Guide for the Control of Molluscan Shellfish: 2019 Revision”.
    • LFIA kits from Neogen and Scotia are simple, low-cost screening assays. They can be used in the field, with results obtained within two hours of sample homogenisation. In a study of 136 shellfish samples, the assays returned no false positives or negatives.20
    • The Scotia assay provides semi-quantitative results of total domoic acid concentrations, and results correlate strongly with HPLC data.20
    • The Neogen kit contains an automated scanner, removing subjectivity from analysis.20

NSP Detection

  • ELISA - Click here for details

    • MARBIONC Brevetoxin ELISA is approved for use under certain parameters in the NSSP “Guide for the Control of Molluscan Shellfish: 2019 Revision” for hard clams, sunray venus clams, and oysters. A negative result in the test can substitute for the MBA for the purpose of controlled relaying or controlled harvest end-product testing or to reopen a previously closed area. A positive result requires additional testing using the MBA or could support the same management actions as samples that fail in the MBA.
    • ELISAs have been developed that are faster than the MBA and are sensitive and accurate at low levels with no complications resulting from matrix interferences.1,22
    • An ELISA was found to be more sensitive than the MBA in detecting brevetoxins (BTXs) in Eastern oysters. In this study, ELISA and LC-MS values were highly correlated, and they correlated well with MBA values (0.74 and 0.73 correlation coefficient, respectively).23

  • LC-MS - Click here for details

    • With appropriate analytical standards, LC-MS provides highly specific identification of individual BTX congeners. It is suggested for confirmation of BTXs in shellfish after screening via ELISA, if needed.1,24
    • While the level of detection of shellfish metabolites of BTX by the NSP MBA is unknown and oyster extracts have been found to cause matrix effects in the MBA for BTXs, a single-laboratory validation of an LC-MS/MS method for six BTXs in mussels, oysters, and clams showed good sensitivity and ruggedness and acceptable recovery and precision.24-26

DSP Detection

  • LC-MS/MS - Click here for details

    • This method is approved as the reference method for detection of DSP in shellfish in Commission Regulation (EC) No 2074/2005 (as amended by Commission Regulation (EU) No 15/2011) and is approved for use in the NSSP “Guide for the Control of Molluscan Shellfish: 2019 Revision” for clams.
    • A study of 196 scallop and mussel samples, including 61 that exceeded the quarantine level by LC-MS but were quantified by MBA as being below the quarantine level, demonstrated that the LC-MS method is applicable to routine monitoring of DSP and other lipophilic toxins in bivalves and is more sensitive and accurate than the MBA in quantifying known lipophilic toxins.27

  • Protein Phosphatase Inhibition Assay (PPIA, OkaTest) - Click here for details

    • This method is specifically listed as an alternative or supplementary method to LC-MS/MS in Commission Regulation (EC) No 2074/2005 and quantitatively measures okadaic acid and other toxins of the same group.28
    • A single laboratory validation as well as a study of eight different test materials from seven different species of molluscs showed that the method is robust and accurate and has good toxin recovery rates.28,29 After sample preparation, it permitted the quantification of up to 43 samples within one hour.29

REGULATORY CONSIDERATIONS

  • As a result of the scientific, technical, and ethical limitations of the MBA, countries such as Australia, Canada, Ireland, New Zealand, and the UK no longer use the test for routine toxicity testing of shellfish.9,11
  • As of 1 January 2019, following implementation of Commission Regulation (EU) 2017/1980, the MBA is no longer the reference method for detecting PSP toxins in the EU. This change should ensure the complete replacement of the MBA in the European Union because EU Directive 2010/63 (on the protection of animals used for scientific purposes) prohibits use of an animal test when an accepted alternative exists.
  • The MBA is not listed in the NSSP “Guide for the Control of Molluscan Shellfish” as a method to test for DSP or ASP.

MOUSE BIOASSAY

The general procedure for the MBA involves injecting homogenised shellfish extracts into the abdomen of mice, observing the mice for toxic effects, and recording the time until they die. For PSP and ASP, up to three mice are injected with the test solution, and the time it takes for them to take their last gasping breath is recorded. If death occurs too quickly (in less than five minutes), the sample is diluted until the time to death is five to seven minutes. If the time to death using an undiluted sample is greater than seven minutes, at least three mice must be used in the test. A table known as Sommer’s Table is then used to convert median death time to mouse units, which is further converted to μg poison/mL using individual laboratories’ conversion factors, which were determined when they standardized the assay. Conversion factors are repeatedly checked using five to ten mice.30 The test procedure is similar for DSP and NSP, except that five mice are used per test for NSP testing and that toxicity for DSP is determined from the smallest dose at which two or three mice in a group of three die within 24 hours.31-33

Numerous conditions can significantly affect the results of the MBA, such as animal strain, sex, and age as well as salt concentration, pH, and treatment of the sample.3,10,31,34 For example, as a result of the “salt effect”, which suppresses toxicity, the MBA may underestimate the toxin content of the sample by as much as 60 per cent.34 Additional limitations of the MBA include the following:

  • While oral toxicity is the relevant human risk factor, toxicities are often based on intraperitoneal injection.15,24
  • Trace metals can result in false positives. It has been reported that zinc accumulation in oysters leads to lethal effects in mice but is not harmful to humans.31,35
  • Test results vary with the choice of solvents used for toxin extraction and injection.4
  • The MBA is unable to discriminate between toxins, including those that are not human health threats, and therefore results in false positives.9
  • The time to death versus toxin level is non-linear.31 Determining the time to death is time-consuming, and a lack of precision in determining death times in the MBA can lead to inaccuracies.12,31
  • The MBA is particularly inhumane, as it involves death as a routine endpoint, it often causes animals to experience severe shock and trauma shortly after dosing, and anaesthesia is not used.32

The MBA has further limitations specific to the type of poisoning it is trying to detect:

DSP
In 2007, EFSA wrote that the MBA has limited capability to detect DSP toxins and is unable to detect toxins below 160 µg okadaic acid equiv/kg shellfish flesh. The MBA  cannot detect dinophysistoxin-3 (DTX3), a DSP-causing DTX analogue that can cause major intoxication, but does detect structures similar to DSP toxins, resulting in false negatives and positives.27,36-38 In addition, elevated levels of free fatty acids in shellfish can lead to false positive results using the DSP MBA procedure.39

NSP
In 2010, EFSA wrote that owing to poor specificity and ethical concerns, the MBA is not considered an appropriate method to detect BTXs.40 Plakas and Dickey also have noted that the “[m]ouse bioassay as performed does not reflect the full complement of bioactive toxins in the brevetoxin-contaminated shellfish”, specifically, polar BTX metabolites are not fully reflected in the MBA but are important for monitoring NSP.1 They added, “Its use is no longer defensible with advancements in instrumental methods of analysis, and in vitro assays.”1

PSP
The receptor binding assay’s (RBA) limit of detection is 60 to 100 μg STX equiv/kg, compared to 400 μg STX equiv/kg for the MBA. Therefore, the RBA allows toxic algal blooms to be detected and responded to earlier.12 While it has been demonstrated that the RBA reliably integrates total toxicity of a sample, the MBA is known to underestimate paralytic shellfish toxins.12 In a study of eight laboratories proficient in the MBA, on average, only 35.1 per cent of toxins were recovered from low-level spiked shellfish samples and 46.6 per cent for the moderate level. The authors note that the variation between laboratories is high for moderate toxin levels because of differing sample dilutions.45 Turner et al reported that high zinc concentrations result in the underestimation of PSP toxicity using the MBA by up to a factor of three.13

In another study in which nine laboratories conducted the MBA for PSP toxins, results were so variable (from 1 to 383 mg/100 g expressed as total PSP toxins on a fresh-weight basis) that a statistical evaluation was not conducted.31

ASP
The MBA can provide only a detection limit of 40 mg/kg, while the regulatory limit in the EU is 20 mg/kg.20,41 Thus, the limit of detection for ASP in the MBA is not low enough for this method to be used for regulatory purposes.31

WHY IS THE MOUSE BIOASSAY STILL USED?

The MBA continues to be used because it has been used for decades; therefore, scientists and regulators are comfortable with it. In addition, high levels of technical expertise are not needed, and it is seen as less expensive than alternative tests in many countries. However, Steward and McLeod have argued that if the MBA is conducted following rigorous standards of hygiene and husbandry with appropriately trained and supervised workers, the costs are substantial – exceeding those of non-animal test methods.9 They note that standards of mouse colony breeding and maintenance, husbandry, disease prevention, and ethical oversight must be inadequate in countries in which the MBA is less expensive than alternatives, or in their words, “countries that conduct the MBA on the cheap, so to speak, are simultaneously cutting corners on animal welfare”.9

Additionally, there is a misconception that the MBA is able to identify emerging toxins; however, there is no scientific evidence of this, and in fact, there are many known toxins that the MBA is not well suited to detect.10,42 On the other hand, in the case of HPLC methods, emerging toxins that are similar in structure to those in the suite are likely to appear in the chromatogram. Analysts familiar with evaluating chromatograms and the toxin profile of the sampled geographic area should be able to identify unusual results that could indicate an emerging toxin.10

  • WORKS CITED

    1Plakas SM, Dickey RW. Advances in monitoring and toxicity assessment of brevetoxins in molluscan shellfish. Toxicon. 2010;56(2):137-149.

    2Food and Agriculture Organization of the United Nations, World Health Organization, Intergovernmental Oceanographic Commission of UNESCO. Report of the Joint FAO/IOC/WHO ad hoc Expert Consultation on Biotoxins in Bivalve Molluscs. Oslo, 2004.

    3Stewart I, McLeod C. The laboratory mouse in routine food safety testing for marine algal biotoxins and harmful algal bloom toxin research: Past, present and future. J AOAC Int. 2014;97(2):356-372.

    4Daneshian M, Botana LM, Dechraoui Bottein MY, et al. A roadmap for hazard monitoring and risk assessment of marine biotoxins on the basis of chemical and biological test systems. ALTEX. 2013;30(4):487-545.

    5Turner AD, Hatfield RG, Maskrey BH, et al. Evaluation of the new European Union reference method for paralytic shellfish toxins in shellfish: A review of twelve years regulatory monitoring using pre-column oxidation LC-FLD. Trends Anal. Chem.. 2019;113:124-139.

    6Turner AD, Dhanji-Rapkova M, Algoet M, et al. Investigations into matrix components affecting the performance of the official bioassay reference method for quantitation of paralytic shellfish poisoning toxins in oysters. Toxicon. 2012;59(2):215-230.

    7Turner AD, Hatfield RG, Rapkova M, et al. Comparison of AOAC 2005.06 LC official method with other methodologies for the quantitation of paralytic shellfish poisoning toxins in UK shellfish species. Anal Bioanal Chem. 2011;399(3):1257-1270.

    8Hatfield RG, Punn R, Algoet M, Turner AD. A rapid method for the analysis of paralytic shellfish toxins utilizing standard pressure HPLC: Refinement of AOAC 2005.06. J AOAC Int. 2016;99(2):475-480.

    9Stewart I, McLeod C. A three Rs perspective on the mouse bioassay in routine seafood safety testing for algal biotoxins – 1: replacement. Altern Lab Anim. 2014;42(5):P53-P56.

    10Guy AL, Griffin G. Adopting alternatives for the regulatory monitoring of shellfish for paralytic shellfish poisoning in Canada: Interface between federal regulators, science and ethics. Regul Toxicol Pharmacol. 2009;54(3):256-263.

    11Rourke WA, Murphy CJ. Animal-free paralytic shellfish toxin testing – the Canadian perspective to improved health protection. J AOAC Int. 2014;97(2):334-338.

    12Ruberu SR, Langlois GW, Masuda M, Kittredge C, Perera SK, Kudela RM. Receptor binding assay for the detection of paralytic shellfish poisoning toxins: Comparison to the mouse bioassay and applicability under regulatory use. Food Addit Contam Part A Chem Anal Control Expo Risk Assess. 2018;35(1):144-158.

    13Turner AD, Broadwater M, Van Dolah F. Use of the receptor binding assay for determination of paralytic shellfish poisoning toxins in bivalve molluscs from Great Britain and the assessment of method performance in oysters. Toxicon. 2018;148:155-164.

    14Van Dolah FM, Fire SE, Leighfield TA, Mikulski CM, Doucette GJ. Determination of paralytic shellfish toxins in shellfish by receptor binding assay: Collaborative study. J AOAC Int. 2012;95(3):795-812.

    15Garet E, González-Fernández A, Lago J, Vieites JM, Cabado AG. Comparative evaluation of enzyme-linked immunoassay and reference methods for the detection of shellfish hydrophilic toxins in several presentations of seafood. J Agric Food Chem. 2010;58(3):1410-1415.

    16Dorantes-Aranda JJ, Campbell K, Bradbury A, et al. Comparative performance of four immunological test kits for the detection of paralytic shellfish toxins in Tasmanian shellfish. Toxicon. 2017;125:110-119.

    17Turner AD, Tarnovius S, Johnson S, Higman WA, Algoet M. Testing and application of a refined rapid detection method for paralytic shellfish poisoning toxins in UK shellfish. Toxicon. 2015;100:32-41.

    18Turner AD, McNabb PS, Harwood DT, Selwood AI, Boundy MJ. Single-laboratory validation of a multitoxin ultra-performance LC-hydrophilic interaction LC-MS/MS method for quantitation of paralytic shellfish toxins in bivalve shellfish. J AOAC Int. 2015;98(3):609-621.

    19Turner AD, Dhanji-Rapkova M, Fong SYT, et al. Ultrahigh-performance hydrophilic interaction liquid chromatography with tandem mass spectrometry method for the determination of paralytic shellfish toxins and tetrodotoxin in mussels, oysters, clams, cockles, and scallops: Collaborative study. J AOAC Int. 2020;103. doi: 10.5740/jaoacint.19-0240.

    20Johnson S, Harrison K, Turner AD. Application of rapid test kits for the determination of amnesic shellfish poisoning in bivalve molluscs from Great Britain. Toxicon. 2016;117:76-83.

    21European Food Safety Authority Panel on Contaminants in the Food Chain. Marine biotoxins in shellfish – domoic acid. EFSA Journal. 2009;1181:1-61.

    22Naar J, Bourdelais A, Tomas C, et al. A competitive ELISA to detect brevetoxins from Karenia brevis (formerly Gymnodinium breve) in seawater, shellfish, and mammalian body fluid. Environ Health Perspect. 2002;110(2):179-185.

    23Plakas SM, Jester EL, El Said KR, et al. Monitoring of brevetoxins in the Karenia brevis bloom-exposed Eastern oyster (Crassostrea virginica). Toxicon. 2008;52(1):32-38.

    24Turner AD, Higgins C, Davidson K, et al. Potential threats posed by new or emerging marine biotoxins in UK waters and examination of detection methodology used in their control: Brevetoxins. Mar Drugs. 2015;13(3):1224-1254.

    25McNabb PS, Selwood AI, Van Ginkel R, Boundy M, Holland PT. Determination of brevetoxins in shellfish by LC/MS/MS: Single-laboratory validation. J AOAC Int. 2012;95(4):1097-1105.

    26Wong CK, Hung P, Kam KM. Development of an ICR mouse bioassay for toxicity evaluation in neurotoxic poisoning toxins-contaminated shellfish. Biomed Environ Sci. 2013;26(5):346-364.

    27Suzuki T, Quilliam MA. LC-MS/MS analysis of diarrhetic shellfish poisoning (DSP) toxins, okadaic acid and dinophysistoxin analogues, and other lipophilic toxins. Anal Sci. 2011;27(6):571-584.

    28Smienk H, Domínguez E, Rodríguez-Velasco ML, et al. Quantitative determination of the okadaic acid toxins group by a colorimetric phosphatase inhibition assay: Interlaboratory study. J AOAC Int. 2013;96(1):77-85.

    29Smienk HG, Calvo D, Razquin P, Domínguez E, Mata L. Single laboratory validation of a ready-to-use phosphatase inhibition assay for detection of okadaic acid toxins. Toxins (Basel). 2012;4(5):339-352.

    30AOAC International. AOAC official method 959.08 paralytic shellfish poison biological method.

    31Food and Agriculture Organization of the United Nations. Marine biotoxins. Rome, 2004.

    32Combes RD. The mouse bioassay for diarrhetic shellfish poisoning: A gross misuse of laboratory animals and of scientific methodology. Altern Lab Anim. 2003;31(6):595-610.

    33National Shellfish Sanitation Program. Laboratory evaluation checklist – analysis for NSP (mouse bioassay). Guide for the control of molluscan shellfish: 2017 revision. 2017.

    34LeDoux M, Hall S. Proficiency testing of eight French laboratories in using the AOAC mouse bioassay for paralytic shellfish poisoning: Interlaboratory collaborative study. J AOAC Int. 2000;83(2):305-310.

    35Aune T, Ramstad H, Heidenreich B, et al. Zinc accumulation in oysters giving mouse deaths in paralytic shellfish poisoning bioassay. J Shellfish Res. 1998;17(4):1243-1246.

    36European Food Safety Authority Panel on Contaminants in the Food Chain. Marine biotoxins in shellfish – okadaic acid and analogues. EFSA Journal. 2008;589:1-62.

    37Lee TC, Fong FL, Ho KC, Lee FW. The mechanism of diarrhetic shellfish poisoning toxin production in prorocentrum spp.: Physiological and molecular perspectives. Toxins (Basel). 2016;8(10).

    38Gerssen A, Mulder PP, McElhinney MA, de Boer J. Liquid chromatography-tandem mass spectrometry method for the detection of marine lipophilic toxins under alkaline conditions. J Chromatogr A. 2009;1216(9):1421-1430.

    39Lawrence JF, Chadha RK, Ratnayake WM, Truelove JF. An incident of elevated levels of unsaturated free fatty acids in mussels from Nova Scotia and their toxic effect in mice after intraperitoneal injection. Nat Toxins. 1994;2(5):318-321.

    40European Food Safety Authority Panel on Contaminants in the Food Chain. Marine biotoxins in shellfish – emerging toxins: Brevetoxin group. EFSA Journal. 2010;8(7):1677-1706.

    41Regulation (EC) No 853/2004 of the European Parliament and of the Council of 29 April 2004 laying down specific hygiene rules for food of animal origin.

    42James KJ, Fidalgo Sáez MJ, Furey A, Lehane M. Azaspiracid poisoning, the food-borne illness associated with shellfish consumption. Food Addit Contam. 2004;21(9):879-892.