STXs were first detected in the Galician abalone Haliotis tuberculata in 1991. Subsequently, abalone in this region was affected by toxin concentrations sufficiently high enough to enforce indefinite closure of the industry in 1993 [148]. dcSTX was the most abundant toxin reported in abalone, followed by low concentrations of STX [83, 148–149]. The source of these toxins remains unknown. The dinoflagellates Gymnodinium catenatum and Alexandrium minutum are the common STX(s) producers in this region; however, they do not display temporal or geographical distributions corresponding to that of abalone toxicity [83, 151]. Also, the toxin profile of these potential sources differs from that of the abalone; although biotransformation may be responsible for this discrepancy. The authors postulated that cyanobacteria may be the source of the toxin and they report measurable STX(s) concentrations for the cyanobacterium Rivularia sp. It is noteworthy that no other PSP problems were reported for other mollusks or crustaceans in this region [149]. Anatomical distribution showed high toxicity in the epipodial fringe [149, 151] with as much as 2.6 times more toxin in the epithelium compared to the foot. Toxicity generally increased with increasing abalone size. Depuration of toxin in abalone did not occur during three months of monitoring cultured abalone fed a variety of macroalgae [83]. No other abalone PSP reports have been reported for this area in the recent literature. However, there was a report by Huchette and Clavier (2004) [152] that indicated the abalone fishery reopened in Spain in 2002, but was closed again shortly thereafter due to an oil spill.

Abalone harvesting represents an old fishery in South Africa and currently this fishery includes recreational, subsistence and commercial harvesting. In addition to wild harvest, the 1990’s represented a period of movement towards land-based abalone farms. In 1999, STXs were detected in abalone from two farms located along the west coast of South Africa [150]. Subsequent testing also found the presence of toxins in wild abalone. Throughout this evaluation of abalone PSP, toxicity was tested using the AOAC mouse bioassay with levels reported from below the limit of detection to greater than 1600 μg STX eq/100 g. For most of these cases, analysis was conducted on the entire animal; however, some samples were separated into specific body parts to examine anatomical distribution. As with other organisms, there appears to be large variability in toxicity between individuals (5–10 fold variability reported). Observations of detached and paralyzed abalone in the wild were made and analysis confirmed the presence of STXs. Pitcher et al. [150] found a geographical gradient in toxicity with the highest toxicity observed in abalone from the north and a general decrease southwards. The notable distinction in toxin composition for South African abalone compared to those from Spain is that only STX was detected in the former [150]. This profile is different from the known STX(s) source (Alexandrium catenella) and other vectors (e.g. mussels Mytilus galloprovincialis) in the area [150, 153]. Given the toxin profile differences and the feeding behavior of abalone, it is uncertain what the source of STXs is to the abalone. Further investigations by Etheridge et al. [154] indicated the putative source of toxins to be from their natural diet, the macroalgal kelp Eklonia mamixa. Depuration studies suggest that either abalone can retain the toxins for long periods of time or the toxin was still present during the studies. Pitcher et al. (2001) [150] found that abalone retained toxins for at least seven months with no apparent decline in toxicity when kept under controlled laboratory experiments with kelp as the diet. Controlled feeding experiments [154] were conducted using juvenile abalone (2 cm in length, average wet weight 0.6 ± 0.3 g) and demonstrated that depuration did not occur when abalone were either fed kelp or were starved. However, depuration rates of 6.3 μg per 100g per day were observed when abalone was fed artificial feed. Initial toxicity in the abalone was 160 ± 38 μg STX eq per 100g and after being fed artificial feed for two weeks toxin levels decreased to 72.3 ± 12.5 μg STX eq per 100. Thus, it is possible that feeding farmed abalone artificial feed prior to market could reduce the risk of PSP. Toxin distribution among abalone tissues demonstrates differential uptake and compartmentalization. Thus, the contribution of each tissue to total toxin burden is a function of both its absolute toxicity and relative weight contribution. Pitcher et al. [150] found moderate amounts in the foot and viscera and high amounts in the epipodial fringe. Given the high surface area of the epipodial fringe, it contributes significantly to the total toxin burden. Abalone is often marketed with the foot for human consumption; therefore, it has been suggested that scrubbing and/or removing epithelial tissue could decrease toxicity to safe levels for consumption. This could potentially be used as a strategy to reduce toxin levels prior to market.

Periodic PSP events still occur along the west coast of South Africa. In many cases this has resulted in prevention of exporting live abalone. However, shucking and scrubbing (i.e. removing the epithelial layer of the abalone) decreased toxicity to safe levels (aggregate toxicity < 80 μg/100 g whole animal). For example, Pitcher et al. [150] found that toxicity levels in the foot and epipodial fringe (one of the largest reservoirs of STXs containing > 800 μg/100 g in some cases) both decreased significantly (approximately 6 to 9-fold) when scrubbed. Currently, testing for toxins is done regularly under the South African Shellfish Sanitation Program run by Marine and Coastal Management under an MOU. When traces of toxin are detected, sampling frequency increases and farms in the affected area can be prevented from exporting. Again, shucking, scrubbing and cleaning remain processing options (e.g. canning) that can be used to safely market abalone from this region.

During surveillance on the toxicity of invertebrates such as bivalves inhabiting the coasts of Hiroshima Bay in 2001 and 2002, the carnivorous gastropod rapa whelk Rapana venosa, collected in the estuary of Nikoh River, was found to contain toxins which showed paralytic actions in mice; the maximum toxicities (as STXs) were 4.2 MU/g (May 2001) and 11.4 MU/g (April 2002). This equated to total toxicities of 224 and 206 MU/viscera for these specimens (1MU = 0.18 μgSTX). In an attempt to identify the toxic principle(s) in this gastropod, the viscera were extracted with 80% ethanol acidified with acetic acid, followed by defatting with dichloromethane. The aqueous layer obtained was treated with activated charcoal and then applied to a Sep-Pak C18 cartridge. The unbound toxic fraction was analyzed by high-performance liquid chromatography. The gastropod toxin was rather unexpectedly identified as STXs. It was comprised of GTX3, GTX2, and STX as the major components, which accounted for approximately 91 mol% of all components along with STXs Cl and C2, which are N-sulfocarbamoyl derivatives. Judging from their toxin patterns, it was suggested that the STX(s) toxification mechanism of the gastropod was phytoplankton, such as Alexandrium tamarense, transferred to and accumulated in filter-feeders such as the short-necked clam, and then transferred to this carnivorous whelk through predation [137].

Several species of moon snail and whelk are also known to accumulate STXs and such gastropods are often prohibited for harvesting in waters of the states of Maine and Massachusetts, USA. Closures in waters off the coast of Maine are made by the Department of Marine Resources and are posted on their website (http://www.maine.gov/dmr/rm/public_health/closures/-pspclosures.htm [accessed 3 March, 2008]). The moon snail of interest in this area is Lunatia heros, and the whelks impacted by closures are of the family Muricidae and Buccinidae. In Maine state waters, harvesting of moon snails and the whelk Buccinum undatum is closed as a precaution whenever the blue mussel Mytilus edulis exceeds the regulatory limit for STXs, due to the observation that if there are any bivalves carrying STXs then any co-occurring carnivorous gastropods will be toxic as well (D. Couture, pers. comm.). The Division of Marine Fisheries is responsible for the safety of seafood harvested in Massachusetts state waters and their closures can be found on their website (http://www.mass.gov/dfwele/dmf [accessed 3 March, 2008]). Off the coast of Massachusetts, closures are often in effect for the lobed moon snail Polinices duplicatus and the northern moon snail L. heros, as well as the channeled whelk Busycon canaliculatum and the knobbed whelk B. carica (M. Hickey, pers. comm.). Notably, harvesting closures are often extended for moon snails longer than for other species because they accumulate higher levels of toxin by feeding on toxic bivalves. Certain carnivorous mollusks also appear to retain toxins for longer periods of time than the source bivalves. For example, an extensive Alexandrium fundyense bloom occurred off the coast of New England in 2005 resulting in PSP closures of vast regions in state and federal waters [155]. The U.S. Food and Drug Administration is responsible for the safety of seafood harvested in federal waters and they began sampling shellfish in the impacted areas during the 2005 bloom. Sampling continued in 2006 and toxicity levels above the action level were still being detected for moon snails and whelk from federal waters off the coast of Massachusetts (Table 3). In that region, the only other species that remained toxic was the sea scallop (P. magellanicus), in the viscera (Table 3). Sea scallops are known to retain toxins in viscera for long periods of time compared to other co-occurring species [104]. These data demonstrate the need to monitor toxicity for these non-traditional seafood products, even after bloom conditions have dissipated.

In October 2000, an adult male died within hours of ingesting a xanthid crab Zosimus aeneus (Xanthidae) [172]. A second, yet uneaten specimen of Z. aeneus from the same meal contained a total toxicity of 162.8 μg STX eq/100g tissue (comprising GTX2, GTX3, NEO, dcSTX, and STX); these same toxins were identified in the gut contents, blood, liver and urine of the victim. Metabolism of STXs occurred with the ingested crab harboring GTX2, GTX3 and STX, whereas NEO, dcSTX and STX dominated the STXs in the victim's urine. The STX(s) composition in the gut contents, in both their identity and proportion, was intermediate between the eaten crab and the urine suggesting that toxin conversion commenced in the victim's gut. The victim's meal did not consist solely of the toxic crab eaten and the possibility of other food items acting in a synergistic manner with the consumed STXs cannot be discounted. As well as STXs, xanthid crabs are known to harbor tetrodotoxin (TTX) and palytoxin [180–182].

Oikawa et al. [163–164, 183–184] showed that the edible crab Telmessus acutidens both accumulated and retained STXs after consuming contaminated mussels (Mytilus galloprovincialis) in Japan. STXs in two shore crab species, T. acutidens and Charybdis japonica, were compared with the toxin in the prey mussel M. galloprovincialis and causative dinoflagellates Alexandrium tamarense, all having been collected at Onahama, Fukushima Prefecture, in the northern part of Japan. When the toxicities were detected in mussels by mouse bioassays, 73.7% of the sampled T. acutidens were toxic in the hepatopancreas. Charybdis japonica was also expected to be a possible vector species, but only small quantities of STXs were detected in eight specimens of the crab by HPLC analysis. The difference in STX(s) accumulation in both T. acutidens and C. japonica was then investigated at Onahama, Fukushima Prefecture, from 2002 to 2005. The level of toxin accumulation in the hepatopancreas of T. acutidens corresponded to that of mussels when examined on a yearly basis. In 2003, some crabs had a high toxicity of approximately 1000 MU, which compares to one-third of the human minimum lethal dose. Therefore, it was concluded by the authors, that T. acutidens should be monitored as a vector species of PSP toxins. The toxin profile of T. acutidens was also investigated. Because an increase in highly toxic species of STXs with a decrease in low toxic species, such as N-sulfocarbamoyl-11-hydroxysulfate toxins, was not clearly observed between consecutive samples, toxin transformation in T. acutidens was considered to have a minimal impact on toxicity. STXs were also detected in several specimens of C. japonica, but the highest toxicity was only 7.4 MU/g in the hepatopancreas. Lastly, accumulation and depuration rates of STXs in the crab T. acutidens were investigated by feeding toxic and non-toxic mussels under laboratory controlled conditions. The crab accumulated toxins in the hepatopancreas in proportion to the amount of toxic mussels they ingested, and the toxicity in the crab hepatopancreas became 3.2 fold of that in the prey mussels after 20 days of feeding. During depuration, a fast reduction of the total toxicity was observed in the crab, and the retention rate of the toxicity after five days depuration with feeding of non-toxic mussels was 45.8 +/− 18.7%. The reduction of the toxicity was moderated in the later period of depuration, and the retention rates of the total toxicity after 10 and 20 days were 54.1 +/− 29.8% and 14.5 +/− 9.0%, respectively. The toxin profiles in the crab and mussel were investigated by high performance liquid chromatography, and reductive conversions of the toxins were observed when the toxins were transferred from the mussel to the crab. Consequently, high concentrations of GTX2, GTX3, and STX that were not detected in the prey mussels were found in the crab.

Although not thoroughly recorded in the scientific literature, the State of Alaska, Division of Environmental Health, Food Safety and Sanitation Program has been observing elevated levels of STXs in viscera from several species of commercially harvested crabs for years (Figure 1). PSP is endemic to the coastal communities of the State of Alaska [185]. The high frequency of STX producing dinoflagellates coupled with an extensive seafood harvesting industry prompted the state to establish a STX monitoring program. Most commercially harvested crab in Alaska is landed in the open waters of the Bering Sea, but limited harvesting does occur in areas where PSP toxicity is commonly seen in filter-feeding bivalves. In these areas, high regional and species variability in crab STX(s) content exists, with Dungeness crab (Cancer magister) from Kodiak Island appearing to be a consistent food safety concern (Figure 1, 2). To protect public safety, the State of Alaska Food Safety and Sanitation Program, Department of Environmental Conservation, and the Department of Fish and Game perform both pre-season environmental sampling and in season monitoring of both harvesting areas and harvested product. A conservative action level of 70 μg STX eq. /100g viscera (FDA regulatory action level = 80 μg STX eq. /100g tissue) has been established above which product cannot be marketed either live or whole cooked but must be eviscerated at the processing facility where it is landed (http://www.dec.state.ak.us/eh/fss/seafood/PSP/dungeness.htm [accessed 3 March, 2008]). Due to the success of this monitoring program, no reports of PSP due to the consumption of commercially harvested crab have been reported even though visceral concentrations exceeding 500 μg STX eq./100 g have been observed almost yearly in some areas (Figure 2b).

With the puffer fish exception, because STXs do not typically accumulate in fish muscle, humans who consume only the muscle are unlikely to become intoxicated. However, those those who consume whole fish and eat the viscera are likely to become sick. In 1976 in Brunei, 14 nonfatal PSP cases were associated with the consumption of the planktivorous fish Rastrelliger sp. during a bloom of Pyrodinium bahamense var. compressum [48]. One PSP incident in 1983 in Indonesia involved 191 cases and four human fatalities due to the consumption of the planktivorous clupeoid fish Sardinella spp. and Selaroides leptolepis. In a second incident in November 1983, 45 people became ill after consuming fish and suffered numbness, dizziness, and tingling sensations of the lips, tongue, and throat. Although no known toxic dinoflagellate was associated with the event [100], PSP was highly suspected [211]. STXs with toxin profiles similar to Pyrodinium bahamense have been confirmed in gut contents of Sardinella sp. from Brunei [41] and in PSP incidents involving Pyrodinium, toxic shellfish, and fish that were reported from the Philippines [211]. These incidents likely occurred because it is customary in south-east Asia to eat small fish whole, including any potentially toxic viscera [211–212].

Puffer fish became an important source of protein on the east coast of the United States during the Second World War, and supported a commercial fishery in the decades that followed. The primary species landed was the northern puffer fish (Sphoeroides maculatus) but limited numbers of the southern puffer (S. nephelus), primarily from Florida, was also harvested [213]. The industry was centered in the mid-Atlantic states of Virginia, Maryland, New York, and New Jersey with > 6,000 metric tons landed in 1965 (National Marine Fisheries Service Statistics and Economics Division, personal communication). Fish were marketed dressed and skinned under the name “sea squab”. Although the commercial puffer fish industry has steadily declined since the 1970’s, today being only harvested as by-catch, domestic puffer fish can still be found in some U.S. fresh fish markets. In addition, an average of > 500,000 fish was caught annually between 1981 and 2003 by recreational anglers in the U.S. where they are easily obtained by a range of gear including hook and line (National Marine Fisheries Service Statistics and Economics Division, personal communication [213]).

In January 2002, the poison control center in Tampa, Florida, USA received a report of a man hospitalized with symptoms of numbness and tingling of the hands, vomiting, and diarrhea after consuming puffer fish caught during a recreational fishing trip near Titusville, located on the northern Indian River Lagoon (IRL) on Florida’s central east coast [214]. After additional reports of patients with symptoms of neurological illness from Virginia and New Jersey, all associated with what was believed to be S. nephelus originating from the northern IRL, uneaten fish muscle samples from the New Jersey incident sent to the Canadian Institute for Marine Biosciences by the New Jersey Department of Health were, surprisingly, found to contain no detectable TTX but to contain significant amounts of STX, with lesser amounts of the STX congeners B1, and dcSTX [215]. This same combination of toxins was confirmed in meal remnants from two separate poisoning events in 2004 [216]. In total, 28 cases of SPFP were reported from 2002–2004 -- all due to fish originating from the northern IRL [58]. These were the first reports of STXs both in Florida marine waters and in indigenous puffer fish in the U.S. In April 2002, the Florida Fish and Wildlife Conservation Commission (FWC) placed a ban on the commercial and recreational harvesting of all puffer fish species for the entire IRL. At the same time, the FWC initiated intensive sampling for STXs in multiple species of aquatic biota in Florida’s coastal waters with emphasis on the IRL. Partial results of this sampling were reported [58, 217]. Analysis of IRL puffer fish found concentrations of STXs in muscle often well in excess of the 80 μg STX eq./100 g tissue regulatory action limit set for shellfish. After extended monitoring, STX concentrations in puffer muscle in certain regions of the IRL remained well above the action limit. As a result, the puffer fishing ban in the IRL was made indefinite in June 2004. Based on toxin profiles and abundance in the IRL during the first SPFP reports in 2002, Landsberg et al. [58] suggested the dinoflagellate Pyrodinium bahamense, not reported to produce STXs in Florida waters prior to 2002, as the putative toxin source.

Deeds et al. [218] confirmed that S. nephelus from the northern IRL contained elevated concentrations of STX in muscle (1770 ± 159 μg STX/100g tissue) compared to liver (609 ± 432 μg STX/100g tissue), with only low to non-detectable amounts of TTX in all tissues tested. The additional IRL puffer species S. testudineus (checkered puffer) and S. spengleri (bandtail puffer), known to only occur further south in the lagoon system, were found to contain significantly greater concentrations of TTX compared to STX in all tissues (maximum concentration for TTX found in S. testudineus livers 6076 ± 3283 μg TTX/100g tissue – maximum concentration of STX found in S. spengleri livers 74 ± 42 μg STX/100g tissue). This work confirmed S. nephelus, a species not associated with toxicity in the IRL prior to these events, as the likely cause of all 28 cases of SPFP originating from the IRL during 2002–2004. These events on the east coast of the U.S. represented the first confirmed cases of puffer fish poisoning due solely to STX in North America.