β-N-Methylamino-l-alanine (BMAA), a neurotoxic nonproteinogenic amino acid implicated in the development of neurodegenerative diseases, has been broadly associated with cyanobacteria and bioaccumulation through the food web. Early high-level detections, from poorly selective high-performance liquid chromatography-fluorescence detection (HPLC-FLD) methods, have never been repeated using modern, selective tandem mass spectrometry methods. We developed a hydrophilic interaction liquid chromatography-tandem mass spectrometry (HILIC-MS/MS) method for total BMAA analysis, using an improved solid-phase extraction sample cleanup to reduce suppression. We identified oxidative degradation products of BMAA, formed during strong acid hydrolysis, due to residual nitrate in culture media and implemented a dewatering step to resolve this problem. The method was validated using Spirulina powder, environmental, and two cultured cyanobacterial samples, achieving a reporting limit of 0.3 mg/kg in dry cyanobacterial samples. The validated method was then applied to an additional 30 cyanobacteria cultures from Aotearoa New Zealand. BMAA was only detected in one strain, Planktothrix sp. CAWBG-35, at 2.3 mg/kg. These results indicate that BMAA is not produced at the high levels originally reported (e.g., g/kg) in freshwater cyanobacteria and suggest that BMAA production is not widespread among freshwater cyanobacteria from Aotearoa New Zealand. These results have broad implications for future risk-based environmental and public health monitoring.

Bioactive venoms and toxins are emerging as a promising source of drug leads. Optimized through evolution, these compounds display remarkable selectivity and ligand affinity toward a range of relevant pharmacological targets. The successful development of new drugs from toxins is hampered in some areas by the chemical complexity of the active compounds, which limits the possibility of using chemical synthesis or recombinant strategies for drug lead generation. Marine paralytic shellfish toxins produced by marine microalgae is one such family of compounds. These compounds are highly potent blockers of voltage-gated ion channels, involved in regulating a range of physiological processes and thus versatile targets for drug development. To overcome the supply issue, the current paper describes the development of a scalable production method to generate gram amounts of gonyautoxin-1,4 by mass cultivation of the dinoflagellate Alexandrium pacificum in artificial seawater. By selecting a high-producing strain and running a series of growth optimization experiments, we have scaled up production from 100 mL to 1150 L, with cellular yields of toxin 30 times higher than in a natural bloom. This allows commercial production of gram amounts of these promising compounds, thereby enabling their use in a range of applications beyond the analytical scale.

Pinnatoxins, a group of marine biotoxins primarily produced by the dinoflagellate Vulcanodinium rugosum, have garnered significant attention due to their potent toxic effects and widespread distribution in marine ecosystems. LC–MS analysis of shellfish and V. rugosum cultures revealed the presence of previously unidentified isomers of pinnatoxins D, E, F, and H, at levels approximately six times lower than those of known isomers. The chemical structures of these isopinnatoxins were determined using a combination of LC–MS/MS and NMR spectroscopy, which demonstrated that the isomerization of each pinnatoxin occurred through the opening and recyclization of the spiro-linked tetrahydropyranyl D-ring to form a smaller tetrahydrofuranyl ring. The acute toxicity of isopinnatoxin E was determined by intraperitoneal injection into mice and was found to be significantly lower than that of pinnatoxin E. Given their low toxicity and low abundance, it is unlikely that isopinnatoxins contribute significantly to the overall toxicity of pinnatoxins.

Indigo naturalis (IN) is a natural product extensively used as a traditional Chinese medicine due to its pharmacological properties. While IN has been proposed to exert its activity, at least partly, via the aryl hydrocarbon receptor (AhR), the pharmacokinetics of its known indole alkaloid AhR ligands and their relative contributions to the net AhR agonistic activity in blood is unknown. We conducted an open-label, placebo-controlled, doseescalation study (0.25-2.0 g) of orally-administered IN in healthy individuals. Blood samples obtained prior and up to 24 h after ingestion were analysed for net AhR activity and pharmacokinetics of circulating IN-derived AhR ligands determined by HPLC-ESI-MS/MS. Oral administration of IN dose-dependently increased net plasma AhR agonistic activity - initially through the early appearance of tryptanthrin, and later sustained by the more potent AhR agonist indirubin. This first pharmacokinetic study of IN-derived indole alkaloids in humans indicates that the increase in systemic AhR agonistic activity is the result of sequential appearance in blood of highly bioavailable / moderately agonistic (tryptanthrin) and moderately bioavailable / highly agonistic (indirubin and indigo) AhR ligands, challenging the commonly held belief that only highly agonistic indole alkaloids contribute to AhR agonism and therapeutic activity of IN.

Increased permeability of the intestinal epithelial layer is linked to the pathogenesis and perpetuation of a wide range of intestinal and extra-intestinal diseases. Infecting humans with controlled doses of helminths, such as human hookworm (termed hookworm therapy), is proposed as a treatment for many of the same diseases. Helminths induce immunoregulatory changes in their host which could decrease epithelial permeability, which is highlighted as a potential mechanism through which helminths treat disease. Despite this, the influence of a chronic helminth infection on epithelial permeability remains unclear. This study uses the chronically infecting intestinal helminth Heligmosomoides polygyrus to reveal alterations in the expression of intestinal tight junction proteins and epithelial permeability during the infection course. In the acute infection phase (1 week postinfection), an increase in intestinal epithelial permeability is observed. Consistent with this finding, jejunal claudin-2 is upregulated and tricellulin is downregulated. By contrast, in the chronic infection phase (6 weeks postinfection), colonic claudin-1 is upregulated and epithelial permeability decreases. Importantly, this study also investigates changes in epithelial permeability in a small human cohort experimentally challenged with the human hookworm, Necator americanus . It demonstrates a trend toward small intestinal permeability increasing in the acute infection phase (8 weeks postinfection), and colonic and whole gut permeability decreasing in the chronic infection phase (24 weeks postinfection), suggesting a conserved epithelial response between humans and mice. In summary, our findings demonstrate dynamic changes in epithelial permeability during a chronic helminth infection and provide another plausible mechanism by which chronic helminth infections could be utilized to treat disease.

Increased permeability of the intestinal epithelial layer is linked to the pathogenesis and perpetuation of a wide range of intestinal and extra-intestinal diseases. Infecting humans with controlled doses of helminths, such as human hookworm (termed hookworm therapy), is proposed as a treatment for many of the same diseases. Helminths induce immunoregulatory changes in their host which could decrease epithelial permeability, which is highlighted as a potential mechanism through which helminths treat disease. Despite this, the influence of a chronic helminth infection on epithelial permeability remains unclear. This study uses the chronically infecting intestinal helminth Heligmosomoides polygyrus to reveal alterations in the expression of intestinal tight junction proteins and epithelial permeability during the infection course. In the acute infection phase (1 week postinfection), an increase in intestinal epithelial permeability is observed. Consistent with this finding, jejunal claudin-2 is upregulated and tricellulin is downregulated. By contrast, in the chronic infection phase (6 weeks postinfection), colonic claudin-1 is upregulated and epithelial permeability decreases. Importantly, this study also investigates changes in epithelial permeability in a small human cohort experimentally challenged with the human hookworm, Necator americanus. It demonstrates a trend toward small intestinal permeability increasing in the acute infection phase (8 weeks postinfection), and colonic and whole gut permeability decreasing in the chronic infection phase (24 weeks postinfection), suggesting a conserved epithelial response between humans and mice. In summary, our findings demonstrate dynamic changes in epithelial permeability during a chronic helminth infection and provide another plausible mechanism by which chronic helminth infections could be utilized to treat disease. The increased permeability of the intestines is implicated in the pathogenesis of many diseases that helminth infection is proposed to protect their host against developing. In this study, we examine changes in intestinal permeability in mice and humans experimentally challenged with chronic helminth infections. We demonstrate dynamic changes in tight junction protein expression and intestinal permeability which could have important health implications for humans infected with helminths. image

The genus Gambierdiscus produces an array of bioactive hydrophilic and lipophilic secondary metabolites that range in mode of action and toxicity. In this study, the metabolite fingerprint was mapped for thirteen Gambierdiscus, five Coolia and two Fukuyoa species (34 isolates) by assessing the production of 56 characterised secondary metabolites. Gambierdiscus polynesiensis was the only species to produce Pacific-ciguatoxin-3B (P-CTX3B), P-CTX3C, iso-P-CTX3B/C, P-CTX4A, P-CTX4B and iso-P-CTX4A/B. G. australes produced maitotoxin-1 (MTX-1) and MTX-5, G. cheloniae produced MTX-6 and G. honu produced MTX-7. Ubiquitous production of 44-methylgambierone was observed amongst all the Gambierdiscus isolates, with nine species also producing gambierone. Additional gambierone analogues, including anhydrogambierone (tentatively described herein), were also detected in all Gambierdiscus species, two Coolia and two Fukuyoa species. Gambieroxide was detected in G. lewisii and G. pacificus and gambieric acid A was detected in ten Gambierdiscus species, with G. australes (CAWD381) being the only isolate to produce gambieric acids A–D. This study has demonstrated that the isolates tested to date produce the known CTXs or MTXs, but not both, and highlighted several species that produced ‘unknown’ compounds displaying characteristics of cyclic polyethers, which will be the focus of future compound discovery efforts.

Identifying compounds responsible for the observed toxicity of the Gambierdiscus species is a critical step to ascertaining whether they contribute to ciguatera poisoning. Macroalgae samples were collected during research expeditions to Rarotonga (Cook Islands) and North Meyer Island (Kermadec Islands), from which two new Gambierdiscus species were characterized, G. cheloniae CAWD232 and G. honu CAWD242. Previous chemical and toxicological investigations of these species demonstrated that they did not produce the routinely monitored Pacific ciguatoxins nor maitotoxin-1 (MTX-1), yet were highly toxic to mice via intraperitoneal (i.p.) injection. Bioassay-guided fractionation of methanolic extracts, incorporating wet chemistry and chromatographic techniques, was used to isolate two new MTX analogs; MTX-6 from G. cheloniae CAWD232 and MTX-7 from G. honu CAWD242. Structural characterization of the new MTX analogs used a combination of analytical chemistry techniques, including LC–MS, LC–MS/MS, HR–MS, oxidative cleavage and reduction, and NMR spectroscopy. A substantial portion of the MTX-7 structure was elucidated, and (to a lesser extent) that of MTX-6. Key differences from MTX-1 included monosulfation, additional hydroxyl groups, an extra double bond, and in the case of MTX-7, an additional methyl group. To date, this is the most extensive structural characterization performed on an MTX analog since the complete structure of MTX-1 was published in 1993. MTX-7 was extremely toxic to mice via i.p. injection (LD50 of 0.235 µg/kg), although no toxicity was observed at the highest dose rate via oral administration (155.8 µg/kg). Future research is required to investigate the bioaccumulation and likely biotransformation of the MTX analogs in the marine food web.

Most marine biotoxins are produced by microalgae. The neurotoxin tetrodotoxin (TTX) has been reported in many seafood species worldwide but its source is unknown, making accumulation and depuration studies in shellfish difficult. Tetrodotoxin is a water-soluble toxin and cannot be directly ingested by shellfish. In the present study, a method was developed which involved binding TTX to solid particles of humic acid and encapsulating them in agar-gelatin capsules. A controlled quantity of TTX-containing microcapsules (size range 20–280 μm) was fed to Paphies australis, a bivalve known to accumulate TTX in the wild. The TTX-containing microcapsules were fed to P. australis every second day for 13 days. Ten P. australis (including five controls fed non-toxic microalgae) were harvested after 7 days and ten after 13 days. Paphies australis accumulated TTX, reaching concentrations of up to 103 µg kg−1 by day 13, exceeding the European Food Safety Authority recommended concentration of 44 μg kg−1 in shellfish. This novel method will allow future studies to explore the effects, accumulation and depuration rates of TTX in different animals and document how it is transferred through food webs.

Saxitoxin and its derivatives, the paralytic shellfish toxins (PSTs), are well known to be toxic to humans, and maximum permitted levels in seafood have been established by regulatory authorities in many countries. Monitoring of PSTs is typically performed using chemical methods which quantify the concentration of the individual PST analogues, of which there are many. However, since the toxicities of analogues are different, they do not equally contribute to the overall toxicity of the sample. To account for these differences, toxicity equivalency factors (TEFs) need to be determined for each analogue and applied. Currently there are no established TEFs for decarbamoyl gonyautoxin 1&4 (dcGTX1&4), which occurs in some clam species such as Mactra chinensis contaminated with PSTs due to metabolism within the shellfish. In this study the median lethal dose of purified, equilibrated epimeric mixture of dcGTX1&4 has been determined by intraperitoneal injection (i. p.) (4.75 mu mol/kg) and by feeding (34.9 mu mol/kg). The most relevant route of exposure is orally with feeding being more representative of human consumption and more reliable than gavage. Based on the median lethal dose by feeding, a TEF of 0.1 is recommended for dcGTX1&4. Receptor binding activity and i.p. toxicity results showed dcGTX1&4 to be much less toxic than STX (140-170-fold). However, by feeding a much smaller difference in toxicity was observed with dcGTX1&4 being only 11-fold less toxic than STX. Analysis of the gut contents of mice dosed with dcGTX1&4 showed the presence of decarbamoyl gonyautoxin 2&3, decarbamoyl saxitoxin and decarbamoyl neosaxitoxin, all of which are of greater toxicity. This conversion of dcGTX1&4 within the digestive track to more toxic congeners may explain the high relative toxicity of dcGTX1&4 by feeding compared to that determined by i.p. and by sodium channel activity.

Understanding the toxicity and production rates of the various secondary metabolites produced by Gambierdiscus and cohabitating benthic dinoflagellates is essential to unravelling the complexities associated with ciguatera poisoning. In the present study, a sulphated cyclic polyether, gambierone, was purified from Gambierdiscus cheloniae CAWD232 and its acute toxicity was determined using intraperitoneal injection into mice. It was shown to be of low toxicity with an LD50 of 2.4 mg/kg, 9600 times less toxic than the commonly implicated Pacific ciguatoxin-1B, indicating it is unlikely to play a role in ciguatera poisoning. In addition, the production of gambierone and 44-methylgambierone was assessed from 20 isolates of ten Gambierdiscus, two Coolia and two Fukuyoa species using quantitative liquid chromatography–tandem mass spectrometry. Gambierone was produced by seven Gambierdiscus species, ranging from 1 to 87 pg/cell, and one species from each of the genera Coolia and Fukuyoa, ranging from 2 to 17 pg/cell. The production of 44-methylgambierone ranged from 5 to 270 pg/cell and was ubiquitous to all Gambierdiscus species tested, as well as both species of Coolia and Fukuyoa. The relative production ratio of these two secondary metabolites revealed that only two species produced more gambierone, G. carpenteri CAWD237 and G. cheloniae CAWD232. This represents the first report of gambierone acute toxicity and production by these cohabitating benthic dinoflagellate species. While these results demonstrate that gambierones are unlikely to pose a risk to human health, further research is required to understand if they bioaccumulate in the marine food web.

The cyanobacterium Microcoleus autumnalis grows as thick benthic mats in rivers and is becoming increasingly prevalent around the world. M. autumnalis can produce high concentrations of anatoxins and ingestion of benthic mats has led to multiple dog deaths over the past two decades. M. autumnalis produces a suite of different anatoxin congeners including anatoxin-a (ATX), dihydroanatoxin-a, (dhATX), homoanatoxin-a and dihydrohomoanatoxin-a. Benthic mat samples often contain high levels of dhATX, but there is little toxicology information on this congener. In the present study, natural versions of dhATX and ATX were purified from cyanobacteria to determine the acute toxicity by different routes of administration using mice. Nuclear magnetic resonance spectroscopy was used to confirm the putative structure of dhATX. By intraperitoneal (ip) injection, the median lethal dose (LD50) for dhATX was 0.73 mg/kg, indicating a reduced toxicity compared to ATX (LD50 of 0.23 mg/kg). However, by oral administration (both gavage and feeding), dhATX was more toxic than ATX (gavage LD50 of 2.5 mg/kg for dhATX and 10.6 mg/kg for ATX; feeding LD50 of 8 mg/kg for dhATX and 25 mg/kg for ATX). The relative nicotinic acetylcholine receptor-binding affinities of ATX and dhATX were determined using the Torpedo electroplaque assay which showed consistency with the relative toxicity determined by ip injection. This work highlights that toxicity studies based solely on ip injection may not yield LD50 values that are relevant to those derived via oral administration, and hence, do not provide a good estimate of the risk posed to human and animal health in situations where oral ingestion is the likely route of exposure. The high acute oral toxicity of dhATX, and its abundance in M. autumnalis proliferations, demonstrates that it is an important environmental contaminant that warrants further investigation.

A new analogue of gambierone, 44-methylgambierone, was isolated from the benthic dinoflagellate Gambierdiscus australes collected from Raoul Island (Rangitahua/Kermadec Islands). This molecule has been previously reported as maitotoxin-3. The structure of 44-methylgambierone was elucidated using 1D- and 2D-nuclear magnetic resonance spectroscopy and mass spectrometry techniques. The nine-ring polyether backbone (A-I) and functional groups (carbonyl, terminal diol, 1,3-diene and monosulphate) are the same for both compounds with the addition of an olefinic methyl group being the only modification in 44-methylgambierone. (C) 2019 The Authors. Published by Elsevier Ltd.

Fluorescent molecules are regularly utilised to study ligand receptor interactions. Many ligands for nicotinic receptors have been conjugated with fluorophores to study receptor kinetics, recycling and ligand binding characteristics. These include small agonist molecules, as well as large peptidic antagonists. However, no small molecule antagonists have been investigated using this method. Pinnatoxin F is a newly discovered non-peptidic muscle type nicotinic receptor antagonist produced by the marine dinoflagellate species Vulcanodinium rugosum. This molecule has the potential for conjugation to a fluorophore, allowing subsequent visualisation of interactions with nicotinic receptors. Pinnatoxin F was modified by addition of diaminopolyether spacers, to which a fluorophore (VivoTag (R) 645) was conjugated. The fluorescent pinnatoxin was then applied to muscle sections from thy1-YFP-H transgenic mice, which express YFP in motor nerves, to allow direct visualization of fluorescent binding at the neuromuscular junction. The addition of both the diaminopolyether spacer and the VivoTag (R) 645 reduced the potency of pinnatoxin F, as evidenced by a reduction in in vitro neuromuscular blocking activity and in vivo toxicity. Despite this reduced potency, the fluorescent molecule selectively labelled endplate regions in thy1-YFP mouse muscle sections and this labelling was inhibited by pre-exposure of muscle sections to native pinnatoxin F or the nicotinic antagonist alpha-bungarotoxin. This study proves nicotinic receptor binding activity of pinnatoxin F and is the first example of a fluorophore-conjugated small-molecule antagonist for nicotinic receptors. These results indicate the potential for other small-molecule nicotinic receptor antagonists to be fluorescently labelled and used as probes for specific nicotinic receptor subtypes. (C) 2014 Elsevier Ltd. All rights reserved.

For the first time wild-caught Tasmanian abalone, Haliotis rubra, have been reported to contain paralytic shellfish toxins (PSTs). This observation followed blooms of the toxic dinoflagellate Gymnodinium catenatum. No illnesses were reported, but harvesting restrictions were enforced in commercial areas. Abalone were assayed using HPLC-FLD methodology based on AOAC official method 2005.06. An uncommon congener, deoxydecarbamoyl-STX (doSTX), was observed in addition to regulated PSTs as unassigned chromatographic peaks. A quantitative reference material was prepared from contaminated Tasmanian abalone viscera and ampouled at 54.2 μmol/L. The LD50 of doSTX via intraperitoneal injection was 1069 nmol/kg (95% confidence limits 983-1100 nmol/kg), indicating it is nearly 40 times less toxic than STX. A toxicity equivalence factor of 0.042 was generated using the mouse bioassay. Levels of PSTs varied among individuals from the same site, although the toxin profile remained relatively consistent. In the foot tissue, STX, decarbamoyl-STX and doSTX were identified. On a molar basis doSTX was the dominant congener in both foot and viscera samples. The viscera toxin profile was more complex, with other less toxic PST congeners observed and was similar to mussels from the same site. This finding implicates localised dinoflagellate blooms as the PST source in Tasmanian abalone.

In 2010 the Cawthron Institute adopted AOAC official method 2005.06 (Lawrence method) for regulatory testing of paralytic shellfish toxins. This included adapting the method to a UPLC format and developing a rapid periodate screen to eliminate the vast majority of samples with no PSTs present. The method gained New Zealand regulatory approval and has since been used to test >2000 samples. Soon after implementation a major HAB of the toxic dinoflagellate Alexandrium catenella occurred in a prime shellfish growing area of New Zealand. This event was the most serious to date in this country with extremely high cell concentrations observed in some locations (>4 x 10(6) cells L-1). Toxin levels observed in Greenshell (TM) mussels (Perna canaliculus) and Flat oysters (Ostrea chilensis) exceeded the regulatory level of 0.8 mg/kg shellfish meat as saxitoxin equivalents. Closures of commercial shellfish harvesting areas were enforced for a period of up to three months as toxin levels remained above the regulatory level for an extended period, even after the bloom had crashed. Analysis of several hundred positive shellfish samples during this event allowed us to better understand the technical performance of the method during a bloom event. The periodate screen substantially overestimated the true PST level in the samples because several PSTs gave co-eluting oxidation products, and it was assumed that the entire peak was due to the presence of the more toxic congener. The ratio between the screen and confirmation test results remained relatively constant throughout the bloom events. This information supports an amendment to the overly conservative regulatory control scheme employed in New Zealand for PST testing. Despite overestimation, the periodate screen has proved highly useful as it allows a quick determination of PST-free samples and provides a high level of security against harvesting contaminated products. (C) 2013 Elsevier B.V. All rights reserved.

A single-laboratory validation is reported for an LC/MS/MS quantification of six brevetoxins in four matrixes (Greenshell mussel, eastern oyster, hard clam, and Pacific oyster). Recovery and precision data were collected from seven analytical batches using shellfish flesh at 0.05 mg/kg. Method recoveries and within-laboratory reproducibility ranged from 73 to 112%, with an RSD between 14 and 18% for brevetoxin-3, brevetoxin B5, brevetoxin B2, and S-desoxy brevetoxin B2. The recovery and within-laboratory reproducibility for brevetoxin-2 was 61%, with an RSD of 27%. Brevetoxin B1 gave an RSD of 12%, but no reference material was available and this toxin was only recorded in a hard clam sample naturally contaminated with brevetoxins. One naturally contaminated sample of each shellfish matrix, with brevetoxin levels ranging from 0.012 to 9.9 mg/kg, was tested in multiple batches, and the RSDs were similar to those for fortified samples at 0.05 mg/kg. Comparisons with limited data for the neurotoxic shellfish poisoning mouse bioassay for four naturally contaminated shellfish samples showed that the regulatory action limit of 0.8 mg/kg is conservative with respect to the bioassay regulatory limit of 20 mouse units/100 g.

Palytoxin is a highly toxic non-proteinaceous marine natural product that can pass through the food chain and result in human illnesses. A recent review by the European Food Safety Authority concluded that palytoxin requires regulation in seafood and a limit of 30 μg kg⁻¹ for shellfish flesh was suggested. Current methods based on LC-MS detection of intact palytoxins do not have sufficient sensitivity to enforce this limit for palytoxin. To improve sensitivity for trace analysis, a novel screen approach has been developed that uses LC-MS/MS analysis of substructures generated by oxidative cleavage of vicinal diol groups present in the intact toxin. Oxidation of palytoxins, ovatoxins or ostreocins using periodic acid generates two nitrogen-containing aldehyde fragments; an amino aldehyde common to these toxins, and an amide aldehyde that may vary depending on toxin type. Conditions for micro-scale oxidation of palytoxin were optimised, which include a novel SPE cleanup and on-column oxidation step. Rapid analysis of cleavage fragments was established using LC-MS/MS. Linear calibrations were established for the amino aldehyde from a palytoxin reference standard, which is suitable for all known palytoxin-like compounds, and for the confirmatory amide aldehydes of palytoxin and ostreocin-D. Palytoxin recoveries (at 10 μg kg⁻¹) from shellfish and fish tissues were 114-119% (as amine aldehyde) and 90-115% (as amide aldehyde) with RSDs for both of ≤ 18% (all tissues, n = 12). The method LOD was determined to be approximately 1 ng mL⁻¹ and the LOQ 4 ng mL⁻¹, which corresponds to 10 μg kg⁻¹ in tissue (flesh of shellfish or fish). The method has potential for use in research and is sufficiently sensitive for regulatory testing, should it be required.

Farmed greenlip abalone Haliotis laevigata were fed commercial seaweed-based food pellets or feed pellets supplemented with 8 × 10⁵ Alexandrium minutum dinoflagellate cells g⁻¹ (containing 12 ± 3.0 μg STX-equivalent 100 g⁻¹, which was mainly GTX-1,4) every second day for 50 days. Exposure of abalone to PST supplemented feed for 50 days did not affect behaviour or survival but saw accumulation of up to 1.6 μg STX-equivalent 100 g⁻¹ in the abalone foot tissue (muscle, mouth without oesophagus and epipodial fringe), which is ∼50 times lower than the maximum permissible limit (80 μg 100 g⁻¹ tissue) for PSTs in molluscan shellfish. The PST levels in the foot were reduced to 0.48 μg STX-equivalent 100 g⁻¹ after scrubbing and removal of the pigment surrounding the epithelium of the epipodial fringe (confirmed by both HPLC and LC-MS/MS). Thus, scrubbing the epipodial fringe, a common procedure during commercial abalone canning, reduced PST levels by ∼70%. Only trace levels of PSTs were detected in the viscera (stomach, gut, heart, gonad, gills and mantle) of the abalone. A toxin reduction of approximately 73% was observed in STX-contaminated abalone held in clean water and fed uncontaminated food over 50 days. The low level of PST uptake when abalone were exposed to high numbers of A. minutum cells over a prolonged period may indicate a low risk of PSP poisoning to humans from the consumption of H. laevigata that has been exposed to a bloom of potentially toxic A. minutum in Australia. Further research is required to establish if non-dietary accumulation can result in significant levels of PSTs in abalone.