A fluorimetric method based on changes in membrane potential for screening paralytic shellfish toxins in mussels

To prevent the consumption of bivalves contaminated with paralytic shellfish poisoning (PSP), toxin levels in seafood products are estimated by using the official mouse bioassay. Because of the limitations of this bioassay other methods of monitoring toxins are clearly needed. We have developed a test to screen for PSP toxins based on its functional activity; the toxins bind to the voltage-gated Na+ channels and block their activity. The method is a fluorimetric assay that allows quantitation of the toxins by detecting changes in the membrane potential of human excitable cells. This assay gives an estimate of toxicity, since each toxin present in the sample binds to sodium channels with an affinity which is proportional to its intrinsic toxic potency. The detection limits for paralytic shellfish toxins were found to be 1 ng saxitoxin equivalents/ml compared to the regulatory limit threshold of 400 ng/ml (equivalent to 80 microg/100 g) used in most countries. Our results indicate that this fluorescent assay is a specific, very sensitive, rapid, and reliable method of monitoring PSP toxin levels in samples from seafood products and toxic algae.     

Comparison of ELISA and SPR biosensor technology for the detection of paralytic shellfish poisoning toxins

An enzyme labeled immunosorbent assay (ELISA) and surface plasmon resonance (SPR) biosensor assay for the detection of paralytic shellfish poisoning (PSP) toxins were developed and a comparative evaluation was performed. A polyclonal antibody (BC67) used in both assay formats was raised to saxitoxin–jeffamine–BSA in New Zealand white rabbits. Each assay format was designed as an inhibition assay. Shellfish samples (n = 54) were evaluated by each method using two simple rapid extraction procedures and compared to the AOAC high performance liquid chromatography (HPLC) and the mouse bioassay (MBA). The results of each assay format were comparable with the HPLC and MBA methods and demonstrate that an antibody with high sensitivity and broad specificity to PSP toxins can be applied to different immunological techniques. The method of choice will depend on the end-users needs. The reduced manual labor and simplicity of operation of the SPR biosensor compared to ELISA, ease of sample extraction and superior real time semi-quantitative analysis are key features that could make this technology applicable in a high-throughput monitoring unit.     

Determination of paralytic shellfish poisoning (PSP) toxins in UK shellfish

A study was conducted to aid the interpretation of data generated by parallel testing of the qualitative Jellett Rapid Test (JRT) and the mouse bioassay (MBA) for detection of paralytic shellfish poisoning (PSP) toxins within the UK statutory shellfish biotoxin monitoring programme. A selection of stored sample extracts subjected to testing by MBA and/or JRT were further analysed by liquid chromatography with fluorescence detection (LC–FLD) to provide additional information on the concentrations of PSP toxins and toxin profiles.Results, from this study, demonstrate the potential of the JRT to effectively screen out PSP toxin negative shellfish samples and samples containing low concentrations of toxins from UK monitoring programmes. Additionally, data generated using LC–FLD highlights the potential of introducing alternative analytical techniques to completely replace the requirement for the MBA.     

Variation in paralytic shellfish toxin composition within the Protogonyaulax tamaronsis/catenella species complex; red tide dinoflagellates

Unialgal isolates of the Protogonyaulax (—Gonyaulax) tamarensis/catenella species complex, a group of dinoflagellates which causes paralytic shellfish poisoning (PSP), were subjected to toxin analysis by HPLC. Protogonyaulax isolates from widely separated geographical locations were compared, including the northeastern Pacific (British Columbia and Washington State), eastern Canada, Portugal, the United Kingdom and New Zealand. Two distantly related gonyaulacoid species were also analyzed, but the presence of PSP toxins was not detected. Although Protogonyaulax isolates varied markedly in total toxin concentration and toxicity, even through the culture cycle, the toxin ratios of individual isolates were distinctive and relatively constant. No toxins were detected in the Plymouth (U.K.) isolate of P. tamarensis, from the species type locality. Two isolates from Vancouver Island (British Columbia), which were previously considered to be non-toxic according to the mouse bioassay, revealed weak toxin spectra by HPLC. Within populations from English Bay (British Columbia) the toxin profiles of tamarensoid isolates tended to be conservative. However, this was not the case for the catenelloid forms from Washington State, which displayed a greater degree of toxin heterogeneity. Significantly, there was no identifiable relationship between toxicity or toxin profiles and the morphological characteristics conventionally used to separate the two dominant morphotypes into species within this species complex.     

A receptor binding assay for paralytic shellfish poisoning toxins: recent advances and applications

We recently described a high throughput receptor binding assay for paralytic shellfish poisoning (PSP) toxins, the use of the assay for detecting toxic activity in shellfish and algal extracts, and the validation of 11-[3H]-tetrodotoxin as an alternative radioligand to the [3H]-saxitoxin conventionally employed in the assay. Here, we report a dramatic increase in assay efficiency through application of microplate scintillation technology, resulting in an assay turn around time of 4 h. Efforts are now focused on demonstrating the range of applications for which this receptor assay can provide data comparable to the more time consuming, technically demanding HPLC analysis of PSP toxins, currently the method of choice for researchers. To date, we have compared the results of both methods for a variety of sample types, including different genera of PSP toxin producing dinoflagellates (e.g. Alexandrium lusitanicum, r2 = 0.9834, n = 12), size-fractioned field samples of Alexandrium spp. (20-64 microm; r2 = 0.9997, n = 10) as well as its associated zooplankton grazer community (200-500 microm: r2 = 0.6169, n = 10; >500 microm: r2 = 0.5063, n = 10), and contaminated human fluids (r2 = 0.9661, n = 7) from a PSP outbreak. Receptor-based STX equivalent values for all but the zooplankton samples were highly correlated and exhibited close quantitative agreement with those produced by HPLC. While the PSP receptor binding assay does not provide information on toxin composition obtainable by HPLC, it does represent a robust and reliable means of rapidly assessing PSP-like toxicity in laboratory and field samples. Moreover, this assay should be effective as a screening tool for use by public health officials in responding to suspected cases of PSP intoxication.     

Characterization of microalgae and associated bacteria collected from shellfish harvesting areas

Marine algal toxins are an important cause of seafood-associated outbreaks. Some marine bacteria living in association with algae are able to produce channel-blocking substances similar to PSP and TTX toxins and a role of these bacteria in the toxicity of dinoflagellates has been hypothesized. The aim of this study was to monitor, over a period of 2 years, areas used in shellfish production in the northern Adriatic Sea, through the determination of phytoplankton and the characterization of bacteria isolated from algae. Toxicity tests on bacterial extracts were performed using in vivo (mouse) and in vitro (cell culture) tests and by HPLC. The Dinophysis genus was detected throughout the year, while the Alexandrium genus was present in winter and spring. Sixteen bacteria isolated from algae, out of 61 bacterial strains tested by in vitro assay, were found to be producers of toxic substances that could block sodium channels in cells. HPLC analysis for the detection of PSP and TTX toxins always gave negative results, but their presence in concentrations undetectable by HPLC, and/or the production of chemically different substances with similar biological action, could not be excluded.     

Development of a F actin-based live-cell fluorimetric microplate assay for diarrhetic shellfish toxins

A new cytotoxicity assay for detection and quantitation of diarrhetic shellfish toxins (DSP) is presented. This assay is based upon fluorimetric determination of F-actin depolymerization induced by okadaic acid (OA)-class compounds in the BE(2)-M17 neuroblastoma cell line. No interferences were observed with other marine toxins such as saxitoxin, domoic acid, or yessotoxin, thus indicating a good specificity of the assay as expected by the direct relationship between protein phosphatase inhibition and cytoskeletal changes. The proposed method is rapid (<2h) and shows a linear response in the range of 50-300 nM OA. The detection limit of the assay for crude methanolic extracts of bivalves lies between 0.2 and 1.0 microg OA per gram of digestive glands, depending on the type of samples (fresh or canned), thus being similar to that of the mouse bioassay. The performance of this assay has been evaluated by comparative analysis of 32 toxic mussel samples by the F-actin assay, mouse bioassay, HPLC and PP2A inhibition assay. Results obtained by the F-actin method showed no differences with HPLC and significant correlation with PP2A inhibition assay (r(2)=0.71). No false negative results were obtained with this new cell assay, which also showed optimum reproducibility.     

Toxic events in the Northwest Pacific coastline of Mexico during 1992–1995: origin and impact

Previously considered as toxin-free, the Baja California Peninsula has witnessed several toxic algal blooms during the past three years. Apparently these ‘red-tide’ phenomena's outbreaks are not linked to any human related activity. This may just reflect better detection and training. Such events may be periodical and natural rather than induced. The most common types of marine toxins have been detected along the coast of the Peninsula and neighboring waters by mouse bioassay and chromatographic techniques. These are: Tetrodotoxin (TTX), Amnesic Shellfish Poison (ASP), Paralytic Shellfish Poisons (PSP), Diarrhetic Shellfish Poisons (DSP) and even Ciguatera (CFP), which are related to the presence of organisms of Prorocentrum sp. and Alexandrium sp. groups, and the diatom Pseudonitzschia sp. among others. There are also some indications about different kinds of TTX in the puffer fish of the region, and reasons to believe that we are facing a quite different pattern in toxic components, since PSP toxic potency (defined as the number of mouse units per gram(MU/g)of shellfish meat) is very high in spite of low dinoflagellates cell density registered. The ecological and social impact of the above has been considerable, with mass deaths of shellfish, seagulls, dolphins and turtles, and even some human casualties. The locally registered toxicity records: PSP found in one single fanshell reaches to 23 000 MU/100 g of tissue as determined by the mouse bioassay and, on a different event, two persons killed after ingesting puffer fish fillet. The largest reservoir of commercial marine organisms in Mexico is precisely the Northwest coast of the country and important plans for building large harbors and develop aquaculture areas are in progress. Therefore, a monitoring program is essential for an adequate management of such resources. Considering the large extension of the Peninsula (about1600 km)and, at this time, the lack of efficient communication means and scarce population, the implementation of such monitoring programs presents a big challenge. This revised version was published online in August 2006 with corrections to the Cover Date.     

Factors influencing growth and toxin production by cultures of the freshwater cyanobacterium Lyngbya wollei Farlow ex Gomont

Collections of Lyngbya wollei were taken from Guntersville Reservoir, Alabama, over a period of three years. Healthy filaments were isolated and transferred to agar plates of Z-8 and LM6E media. Unialgal isolates were cultured for the study of growth and paralytic shellfish poison (PSP) production. Filaments were extracted and the toxins were detected using high performance liquid chromatography (HPLC) with post column oxidation followed by fluorescence detection. HPLC profiles show that laboratory cultures of L. wollei produced decarbamoyl gonyautoxin 2 and 3, plus several other PSP like toxins whose structures are under investigation. At 26 °C and a light intensity of 11 or 22 μmol m-2 s-1 optimum production of both biomass and toxins occurred. A decrease or increase in temperature or light flux caused a reduction in dry weight or toxicity. Compared to control levels, lower PO4-P and NO3-N and higher calcium levels gave rise to higher biomass and toxicity. Lower calcium, calcium- or PO4-P deficient medium and high NO3-N or PO4-P caused a large decrease in dry weight and toxicity. This revised version was published online in August 2006 with corrections to the Cover Date.     

贝类毒素体外细胞毒性作用研究进展

贝类毒素属于非蛋白类高分子化合物,高温处理不能使之失活,严重影响着贝类食品的安全和我国贝类产品的出口贸易。目前贝类毒素最主要的检测方法是小鼠生物法(MBA),然而随着3R理念的推动,毒性评价的体外细胞毒性替代研究取得很大进展,本文综述了8组贝类毒素体外细胞毒性作用的研究成果。     

Two decades of marine biotoxin monitoring in bivalves from Portugal (1986–2006): A review of exposure assessment

Foodborne outbreaks attributed to marine biotoxins were first reported in Portugal in 1946. A regular monitoring programme was implemented in 1986 for PSP, in 1987 for DSP and in 1996 for ASP. The gradual introduction of HPLC methodologies for DSP and PSP allowed a better understanding of toxin biotransformation by bivalves, supplying more selective and sensitive data than mouse bioassays. A comprehensive exposure assessment from DSP toxins in bivalves from the whole coast was only obtained more recently with the introduction of LC–MS methodology. Data on maximum toxin levels found, geographic distribution, seasonality of toxin families, and frequency of samples above current regulatory limits is presented in order to review the data available on exposure assessment after two decades of monitoring. Contamination with DSP toxins was more severe in estuarine and offshore bivalves from the NW and in offshore Donax spp. from the SW and south coasts. DSP toxins were recurrent every year mainly between late spring/early autumn. PSP toxins appeared intermittently in some years between 1986 and 2006, predominantly in autumn. Bivalves from the entire coast were severely contaminated, although bivalves from the NW coast were affected more often. ASP toxins appeared between spring and autumn around the entire coast, but toxin levels rarely exceeded the regulatory limit. Azaspiracids occurred in trace levels below the regulatory limit. Yessotoxins and pectenotoxins occurred in bivalves but have no known effects on the consumers. Several intoxication outbreaks attributed to PSP and DSP occurred during the two decades of the monitoring programme.     

Erratum to Bacterial diversity in toxic<Emphasis Type="Italic"> Alexandrium tamarense</Emphasis> blooms off the Orkney Isles and the Firth of Forth

The genetic diversity of the bacterial community associated with Alexandrium tamarense blooms was studied in blooms of the toxic dinoflagellates in the waters around the Orkney Isles and the Firth of Forth (Scotland). For toxin and molecular analysis of the bacterial communities associated with the toxic bloom, water samples were taken in 1998 and 1999 from A. tamarense blooms. The bacterial community structure, as determined by DGGE (denaturing gradient gel electrophoresis) showed clear differences between all three investigated size fractions (dinoflagellate-associated bacteria, attached bacteria and free-living bacteria), with high diversity within each sample. DNA sequence analysis of the dominant and most frequent DGGE bands revealed the dominance of α Proteobacteria, mainly of the Roseobacter clade, with similarities of 91–99%. Moreover, DGGE bands occurring at the same position in the gel throughout in most samples corroborate the presence of several specific α Proteobacteria of the Roseobacter clade. Overall, 500 bacteria were isolated from the bloom and partly phylogenetically analysed. They were members of two prokaryotic phyla, the Proteobacteria and the Bacteroidetes, related to Proteobacteria of the α and γ subdivisions (Alteromonas, Pseudoalteromonas and Colwellia). All bacteria were tested for the production of sodium channel blocking (SCB) toxins using mouse neuroblastoma assay. No production of SCB toxins was found and high performance liquid chromatography (HPLC) analysis confirmed these results. The content of total paralytic shellfish poisoning (PSP) toxin in the water samples, as measured within the toxic dinoflagellate blooms using HPLC, ranged from 53 to 2191 ng PSP l^?1 in 1998 and from 0 to 478 ng PSP l^?1 in 1999. Changes in PSP toxin content were not accompanied by changes of DGGE band patterns. We therefore presume that the bacterial groups identified in this study were not exclusively associated with toxic A. tamarense, but were generally associated with the phytoplankton.     

Detection of bacteria originally isolated from Alexandrium spp. in the midgut diverticula of Mytilus edulis after water-borne exposure

Bacteria associated with toxic dinoflagellates have been implicated in the production of paralytic shellfish poisoning (PSP) toxins, but it has not been substantiated that bacteria are truly capable of autonomous PSP toxin synthesis or what role bacteria may play in shellfish toxification. In this study, different putatively PSP toxin producing bacteria originally isolated from toxic Alexandrium spp. were exposed to the blue mussel Mytilus edulis. To document that these bacteria accumulated in the digestive tract of the mussels, hybridization techniques that use rRNA targeted oligonuceotides for in situ identification of these bacteria were applied. The mussel hepatopancreas was dissected and paraffin and frozen sections were made. The dissected glands were hybridized with digoxigenin-labelled 16S rRNA oligonucleotide probes. Results demonstrate that mussels will readily uptake and accumulate these bacteria in the hepatopancreas. However, the mussels were not rendered toxic by the ingestion of the bacteria as determined by HPLC with UV detection for PSP toxins and determination of sodium channel blocking activity using the mouse neuroblastoma assay. Thus, although the role that bacteria play in mussel toxification remains unclear, methods are now available which will aid in further investigation of this relatively unexplored area.     

DETECTION OF TYPE A BOTULINAL TOXIN-PRODUCING ORGANISMS SUBCULTURED FROM CHEESE USING AN AMPLIFIED ELISA SYSTEM

Mascarpone cheese implicated in a botulism outbreak was examined for preformed and cultural botulinal toxins using the mouse bioassay. The cheese was also assayed for cultural toxins and for the most probable number (MPN) of toxin-producing organisms/g using an amplified ELISA. No preformed botulinal toxins were discovered in the cheese samples (pH range 5.84-5.86) using the mouse bioassay. However, after cheese subculture in tryptone-peptone-glucose-yeast extract broth, type A botulinal toxin-producing organisms that formed more than 10,000 MLD (mouse lethal dose)/mL in culture were detected. The ELISA results also revealed that type A toxin was present in the culture with a sensitivity of ∼ 10 MLD/mL. The MPN of type A toxin-producing organisms/g in 12 cheese samples examined ranged from < 0.3-9.33. No ELISA cross-reactivity was noted between the type A toxic cultures and other types (B, E, or F). The ELISA sensitivity was ∼5 MLD/mL casein buffer using purified type A neurotoxin. The advantages of the ELISA test are that the toxin type and approximate lethal dose can be determined within one day compared to the mouse bioassay which takes 3–5 days.     

Toxicity and isolation of the cyanobacterium Nodularia spumigena from the southern Baltic Sea in 1986

Three water bloom samples were collected in August 1986 from the southern Baltic Sea. Acute toxicity of the samples was determined by mouse bioassay and the toxins were further studied by HPLC. The bloom samples contained equal amounts of cyanobacteria Nodularia spumigena and Aphanizomenon flos-aquae and were hepatotoxic. Two hepatotoxic Nodularia spumigena strains were isolated from the samples. The isolates produce a toxic peak indistinguishable from the bloom material in the HPLC analysis. The toxicity of the fractions was verified by mouse bioassay. Thus the toxicity of the bloom samples was in all likelihood caused by Nodularia spumigena.     

Winter accumulation of paralytic shellfish toxins in digestive glands of mussels from Arcachon and Toulon (France) without detectable toxic plankton species revealed by interference in the mouse bioassay for lipophilic toxins

Since January 1993, neurological symptoms and rapid deaths (5 to 10 min) were typically observed in the mouse bioassay of acetone extracts of digestive glands from Arcachon and Toulon (France) during the winter season. It was assumed initially that a new lipophilic toxin was present because tests using the AOAC mouse bioassay for paralytic shellfish toxins on acid extracts of whole shellfish meat were negative, no known lipophilic toxins were detected and no toxic phytoplankton species were observed in the area during the poisoning events. In this study, however, preparative isolation of the toxic factor from toxic mussel digestive glands has revealed the presence of paralytic shellfish toxins, the principal ones being gonyautoxins-2 and -3 at Arcachon and gonyautoxins-1, -4, -2 and -3 at Toulon. The toxin concentrations recorded were below levels harmful to consumers and therefore represent a false positive in the mouse bioassay for lipophilic toxins based upon acetone extraction. The origin of the toxins remains to be determined.     

Paralytic shellfish poisoning (PSP) in Margarita Island, Venezuela

A severe outbreak of Paralytic Shellfish Poisoning (PSP) occurred in Manzanillo and Guayacán, northwestern coast of Margarita Island, Venezuela, between August and October 1991. A bloom of dinoflagellates including Prorocentrum gracile, Gymnodinium catenatum and Alexandrium tamarense seemed to be responsible for this outbreak. Levels of PSP toxins in mussels (Perna perna) exceeded the international safety limit of saxitoxin, 80 microg STX/100 microg meat. PSP toxin values varied between 2548 and 115 microg STX/100 g meat in Manzanillo, and between 1422 and 86 microg STX/100 g meat in Guayacán. At both locations, the highest levels were detected in August, when 24 patients exhibited typical symptoms of PSP toxicity after consuming cooked mussels (16 required hospitalization). A high pressure liquid chromatographic (HPLC) procedure was recently used on the 1991 samples. The major toxin detected in samples of both locations was decarbamoyl saxitoxin (dcSTX), but low concentrations of saxitoxin were also found in Manzanillo samples. Gonyautoxins GTX1, GTX2 and GTX3 were detected only at Guayacán, while in both locations, decarbamoylgonyatouxin (dcGTX2,3) toxins were detected. These findings represent the first time that causative toxins of PSP in Venezuela have been chemically identified, and confirm the presence of dcSTX and dcGTX in mussels from the Caribbean Sea. The presence of dcSTX and dcGTX in shellfish is indicative that Gymnodinium catenatum was a causative organism for outbreak of PSP.     

The toxicity and intraspecific variability of Alexandrium andersonii Balech

The toxicity of Alexandrium andersonii Balech is unclear and its intraspecific variability has yet to be studied. To address these gaps in our knowledge, in the present work five strains of A. andersonii from four different localities were characterized. The results showed that despite genetic homogeneity in the 5.8-ITS (internal transcribed spacer) and large subunit (LSU) regions and similar growth rates, strains originating from different locations varied with respect to cell size, the ratios of certain pigments, and their growth patterns. Cultures of the strains grown at 20 °C were analyzed for toxicity using four different methodologies. The two officially established methods, mouse bioassay and high-performance liquid chromatography with fluorescence detection (HPLC-FLD) and post-column reaction analysis of PSP toxins, failed to show the toxicity of any strain. Strains grown at 14 °C were also negative for PSP toxins by HPLC-FLD. However, strains grown at 20 °C exhibited both a response characteristic of the presence of toxin-inhibiting voltage-gated sodium channels, as demonstrated in a neuroblastoma neuro-2a cell-based assay, as well as hemolytic activity in a sheep red blood cell assay.     

Variability of paralytic shellfish toxin occurrence and profiles in bivalve molluscs from Great Britain from official control monitoring as determined by pre-column oxidation liquid chromatography and implications for applying immunochemical tests

As the official control monitoring laboratory in Great Britain for the analysis of marine biotoxins in shellfish, Cefas have for the past five years conducted routine monitoring for paralytic shellfish poisoning toxins (PST) using a non-animal alternative method to the mouse bioassay reference method; a refined version of the AOAC 2005.06 pre-column oxidation liquid chromatography method. Application of this instrumental methodology has enabled the generation of data not only on the occurrence and magnitude of PST events, but also the quantitation and assessment of different PST profiles. Since implementation of the method in 2008, results have shown huge variabilities in the occurrence of PSTs, with large spatial and temporal variabilities around the coastline. Mean PST profiles were not found to correlate either with total PST content of the shellfish, the year of sampling or with a few notable exceptions, the shellfish species. Toxin profiles were found to fall into one of four distinct profile types, with one relating solely to the exclusive presence of decarbamoyl toxins in surf clams. The other profile types contained variable proportions of gonyautoxins, N-sulfocarbamoyl toxins, neosaxitoxin and saxitoxin. While some indications of geographical repeatability were noted, this was not observed for all profile types. Consequently, the application of rapid immunochemical testing methods to end product testing would need to be considered carefully given the large differences in PST congener cross-reactivities.     

Erratum to: Bacterial diversity in toxic<Emphasis Type="Italic"> Alexandrium tamarense</Emphasis> blooms off the Orkney Isles and the Firth of Forth

The genetic diversity of the bacterial community associated with Alexandrium tamarense blooms was studied in blooms of the toxic dinoflagellates in the waters around the Orkney Isles and the Firth of Forth (Scotland). For toxin and molecular analysis of the bacterial communities associated with the toxic bloom, water samples were taken in 1998 and 1999 from A. tamarense blooms. The bacterial community structure, as determined by DGGE (denaturing gradient gel electrophoresis) showed clear differences between all three investigated size fractions (dinoflagellate-associated bacteria, attached bacteria and free-living bacteria), with high diversity within each sample. DNA sequence analysis of the dominant and most frequent DGGE bands revealed the dominance of Proteobacteria, mainly of the Roseobacter clade, with similarities of 91–99%. Moreover, DGGE bands occurring at the same position in the gel throughout in most samples corroborate the presence of several specific Proteobacteria of the Roseobacter clade. Overall, 500 bacteria were isolated from the bloom and partly phylogenetically analysed. They were members of two prokaryotic phyla, the Proteobacteria and the Bacteroidetes, related to Proteobacteria of the and subdivisions (Alteromonas, Pseudoalteromonas and Colwellia). All bacteria were tested for the production of sodium channel blocking (SCB) toxins using mouse neuroblastoma assay. No production of SCB toxins was found and high performance liquid chromatography (HPLC) analysis confirmed these results. The content of total paralytic shellfish poisoning (PSP) toxin in the water samples, as measured within the toxic dinoflagellate blooms using HPLC, ranged from 53 to 2191 ng PSP l^–1 in 1998 and from 0 to 478 ng PSP l^–1 in 1999. Changes in PSP toxin content were not accompanied by changes of DGGE band patterns. We therefore presume that the bacterial groups identified in this study were not exclusively associated with toxic A. tamarense, but were generally associated with the phytoplankton.An erratum to this article can be found at Communicated by H.-D. Franke