The influence of size on domoic acid concentration in king scallop, Pecten maximus (L.)

Concentrations of domoic acid (DA), the biotoxin responsible for amnesic shellfish poisoning (ASP), exceeding the regulatory limit of 20 μg g⁻¹ have caused restricted harvesting and closures of wild king scallop fisheries. Toxin monitoring programmes have reported significant inter-animal variation in DA concentration between scallops from the same area. For the development of reliable sampling and management protocols an understanding of the magnitude and causes of inter-animal variation in toxin concentration are important. Ten samples were collected from an aquaculture site in Clew Bay, Co. Mayo off the west coast of Ireland between February 2003 and February 2004, each sample comprising 12 scallops of each of the following size groups: small (70–85 mm), medium (85–100 mm), large (100–115 mm) and very large (>115 mm). DA concentration in each hepatopancreas and in composite samples of both gonad and adductor muscle from each size group on each sampling occasion were measured. High inter-animal variability in DA concentration in hepatopancreas was recorded; CVs ranging from 12.5% to 82.5%. One negative correlation (R² = 0.7079) between DA concentration in hepatopancreas and scallop shell length, three positive but weak correlations (R² = 0.4536, 0.3459 and 0.4665) and six no correlations were exhibited. Negative correlations were attributed to faster DA uptake by smaller scallops, positive correlations to faster DA depuration by smaller scallops. If only scallops greater than or equal to 100 mm shell length, the minimum commercial size of this species were considered, no correlation occurred on any of the 10 sampling occasions.     

Impact of preparation method on gonad domoic acid levels in the scallop, Pecten maximus (L.)

The king scallop, Pecten maximus (L.), fishery is a valuable economic resource in the UK, and is reliant on supplying premium “roe-on” processed scallops to the continental market. A considerable degree of variability is observed in domoic acid (DA) levels among individual P. maximus and their body components, which complicates the management of the fishery during amnesic shellfish poisoning (ASP) events. This study examined the impact of professional processing and three differing laboratory preparation techniques on final gonadal DA levels. DA analysis was conducted using a LC–MS/MS procedure. The results demonstrate that different methods of preparation can significantly alter gonadal toxicities in scallops from the same site, and the extent to which DA within the digestive loop, which passes through the gonad, contributes to total gonadal DA. Mean gonadal toxicity attributed to the digestive loop contents was estimated at 4.7–24.7 μg DA g⁻¹. Despite large individual variations in toxin levels; in scallops with elevated gonadal toxicities resulting from higher digestive loop content toxicity, the effect of flushing out the contents of the digestive loop significantly reduced the DA content of the tissue and lowered the frequency of individuals harbouring levels above the 20 μg DA g⁻¹ statutory safety limit. Removal of the digestive loop contents can potentially result in an 87% decrease in gonadal DA burden. Furthermore, the method applied by professional processors effectively removed the contents of the digestive loop and reduced gonadal DA to levels comparable with the laboratory techniques. Deliberate contamination with scallop mucus did not increase gonadal DA levels. The extent of toxin variation resulting from differing gonad preparations demonstrates the need to standardize scallop tissue preparation techniques during ASP events. Consequently, detailed protocols aimed at minimizing the contamination of edible components should be developed and adhered to by both processing facilities and monitoring bodies.     

Effect of storage on amnesic shellfish poisoning (ASP) toxins in king scallops (Pecten maximus)

Since 1998, king scallops (Pecten maximus) obtained from Scottish offshore sites have been monitored for domoic acid (DA) and epi-domoic acid (epi-DA), the principal toxic compounds associated with amnesic shellfish poisoning (ASP). The presence of these toxins in king scallops harvested from Scottish waters at concentrations exceeding the current regulatory limit (20 μg g⁻¹ shellfish flesh) is a recurrent event. However, little information was available to determine the effects that different storage conditions experienced during sample transportation to the monitoring laboratory may have on the toxin concentrations, which are subsequently detected. Furthermore, the stability of DA and epi-DA in the solvents (methanol:water (1:1, v/v) and citric acid buffer (0.5 M, pH 3.2)) routinely used for their extraction from shellfish has not previously been assessed. Results from this study demonstrate that when king scallop samples were stored for 2–3 days at 12 °C, a significantly higher toxin concentration was detected in the gonad than when samples were stored at 4 °C and analysed within 48 h. This implies that monitoring programmes must consider transport and storage conditions between harvest and analysis. Stability studies showed rapid decomposition of DA and epi-DA in aqueous methanol extracts while DA and epi-DA seem acceptably stable when stored refrigerated in citrate buffer.     

Detection of the toxin domoic acid from clam extracts using a portable surface plasmon resonance biosensor

The amnesic shellfish poison domoic acid is produced by marine algae of the genus Pseudo-nitzschia. We have developed a portable surface plasmon resonance (SPR) biosensor system for the detection of domoic acid. Because of concerns with domoic acid contamination of shellfish, there is a need for rapid field quantification of toxin levels in both shellfish and seawater. Antibodies were raised against domoic acid and affinity purified. These antibodies were used to develop competition- and displacement-based assays using a portable six-channel SPR system developed in our laboratories. Standard curves for detection of domoic acid in phosphate buffered saline and in diluted clam extracts analyzed by the competition-based SPR assay demonstrated a limit of detection of 3 ppb (10 nM) and a quantifiable range from 4 to 60 ppb (13–200 nM). Comparison of analyses for domoic acid levels in Pacific razor clams, Siliqua patula, containing moderate to high levels of domoic acid by the standard HPLC analysis protocol and the SPR-based assay gave an excellent correlation. This same technology should also function for detection of domoic acid in concentrated algal extracts or high dissolved levels in seawater.     

An assessment of Pseudo-nitzschia population dynamics and domoic acid production in coastal Louisiana

Over 1200 samples were collected from Louisiana estuarine and coastal shelf waters between 1989 and 2002, and analyzed to examine the population dynamics of Pseudo-nitzschia and to assess the potential threat posed by domoic acid (DA), a potent neurotoxin produced by some members within this toxigenic diatom genus. Results demonstrated that three species in this region (Pseudo-nitzschia multiseries, P. pseudodelicatissima complex, P. delicatissima) produce DA, and that particulate toxin levels were highest (up to 3.05 μg L⁻¹) during the spring bloom, while cellular concentrations were highest in the winter/early spring when P. multiseries was most abundant (up to 30 pg cell⁻¹). These particulate toxin levels are comparable to those seen in other regions (e.g., United States west coast) where DA poisoning events have occurred in the past. Pseudo-nitzschia were most abundant under dissolved inorganic nitrogen-replete conditions coupled with lower silicate and/or phosphate concentrations, and in the early spring months when temperatures were cooler. Pseudo-nitzschia were occasionally well-represented in the phytoplankton assemblage (≥10⁶ cells L⁻¹ in 14% of samples, over 50% of total phytoplankton in 5% of samples), indicating that planktivores (e.g., Gulf menhaden, Brevoortia patronus) may have little choice but to consume Pseudo-nitzschia cells, thereby providing potential vectors for DA transfer to higher trophic levels. By comparison, eastern oysters (Crassostrea virginica) present in estuarine waters may be more exposed to this toxin when Pseudo-nitzschia cells are part of a mixed assemblage, reducing selective grazing by these bivalves. C. virginica may thus represent the most effective vector for DA exposure in humans.     

Diatom Nitzschia navis-varingica (Bacillariophyceae) and its domoic acid production from the mangrove environments of Malaysia

The distribution of the toxic pennate diatom Nitzschia was investigated at four mangrove areas along the coastal brackish waters of Peninsular Malaysia. Eighty-two strains of N. navis-varingica were isolated and established, and their identity confirmed morphologically and molecularly. Frustule morphological characteristics of the strains examined are identical to previously identified N. navis-varingica, but with a sightly higher density of the number of areolae per 1 μm (4–7 areolae). Both LSU and ITS rDNAs phylogenetic trees clustered all strains in the N. navis-varingica clade, with high sequence homogeneity in the LSU rDNA (0–0.3%), while the intraspecific divergences in the ITS2 data set reached up to 7.4%. Domoic acid (DA) and its geometrical isomers, isodomoic A (IA) and isodomoic B (IB), were detected in cultures of N. navis-varingica by FMOC-LC-FLD, and subsequently confirmed by LC–MS/MS, with selected ion monitoring (SIM) and multiple reaction monitoring (MRM) runs. DA contents ranged between 0.37 and 11.06 pg cell⁻¹. This study demonstrated that the toxigenic euryhaline diatom N. navis-varingica is widely distributed in Malaysian mangrove swamps, suggesting the risk of amnesic shellfish poisoning and the possibility of DA contamination in the mangrove-related fisheries products.     

Domoic acid production by Pseudo-nitzschia australis and Pseudo-nitzschia calliantha isolated from North Chile

The morphology and toxicity of the four ubiquitous species belonging to the genus Pseudo-nitzschia found in mixed blooms of phytoplankton from northern Chilean waters were studied. The phytoplankton samples and cultures obtained were identified by scanning electron microscopy, revealing the presence of Pseudo-nitzschia australis, P. calliantha, P. pseudodelicatissima and P. subfraudulenta. This is the first report of P. calliantha in northern Chile. Toxin analyses using the LC–MS method confirmed the presence of domoic acid in P. australis and P. calliantha. Domoic acid was not detected in cultures of P. subfraudulenta. This study therefore confirms P. australis and P. calliantha as an unequivocal source of domoic acid in Chilean waters. P. australis is probably the most important producer of amnesic shellfish toxin in view of its domoic acid content. However, more research is needed to evaluate the potential for toxin production in P. pseudodelicatissima.     

Note on the occurrence of Pseudo-nitzschia australis and domoic acid in squid from Monterey Bay, CA (USA)

Over the past decade diatom blooms of domoic acid (DA)-producing Pseudo-nitzschia spp. have been responsible for numerous marine mammal and bird mortalities in Monterey Bay, CA. One possible toxin vector is the market squid, Loligo opalescens, a small pelagic mollusk that plays an important role in the near-shore food web of the California Current ecosystem as a favored vertebrate prey species. This study examined the trophic link between toxic Pseudo-nitzschia and L. opalescens using toxin and stomach content analyses of animals collected from Monterey Bay, CA in 2000. Receptor binding assay data (confirmed by tandem mass spectrometry), demonstrated the presence of DA in squid during a toxic Pseudo-nitzschia event, with P. australis frustules observed in stomach samples. Though DA levels were low (<0.5 μg DA g⁻¹ tissue) in L. opalescens during the study period, it is now clear that this potent neurotoxin can occur in squid and is likely delivered through its krill prey species, which are known to retain DA after feeding on toxic Pseudo-nitzschia. Our findings suggest that further study of the relationship between Pseudo-nitzschia blooms and DA contamination of squid is warranted to better evaluate the potential health risk to humans and wildlife associated with this major commercial seafood species and important prey item.     

Pseudo-nitzschia species population dynamics in the Quoddy Region, Bay of Fundy

The population dynamics of Pseudo-nitzschia species from the Quoddy Region of the Bay of Fundy was examined in a field-based study. Five stations were sampled over a period of 11 weeks during the course of one discrete bloom episode in 2003 with seven species of Pseudo-nitzschia found and enumerated: P. americana, P. delicatissima, P. pseudodelicatissima, P. fraudulenta, P. pungens, P. seriata and P. subpacifica. We related species abundance to physical and chemical properties of seawater (transparency, fluorescence, silicate, phosphate, nitrite + nitrate, ammonia, nitrite, oxygen, sigma-t, tidal level, tidal state and total depth of water column) using canonical correspondence analysis (CCA) to identify factors explaining the greatest amount of variance in their temporal and spatial distribution patterns. Our results indicate that abundance of species and groups of species correlated well with certain specific chemical and physical properties of seawater. P. pseudodelicatissima and P. delicatissima abundance was positively correlated with nitrates and P. americana with depth of the water column. P. pungens was more abundant in samples with higher concentrations of phosphates and lower concentrations of nitrates. P. seriata abundance was negatively associated with total fluorescence. P. subpacifica and P. fraudulenta abundances were not statistically related to any of the variables examined. Our data therefore provides direction for testable, hypothesis driven experiments that could provide predictive insights into the occurrence of certain harmful algal bloom (HAB) species should specific environmental variables be further affected along the gradients extracted in this study.     

Pseudo-nitzschia and domoic acid fluxes in Santa Barbara Basin (CA) from 1993 to 2008

Blooms of domoic acid (DA) producing Pseudo-nitzschia, regularly occur off the coast of California. Although it has been hypothesized that these blooms are increasing in frequency, the lack of historical records limits our understanding of potential causal mechanisms. In this study, an 15-year time-series (1993–2008) of sediment trap samples collected from the Santa Barbara Basin (SBB) at 540 m were analyzed for Pseudo-nitzschia (n = 196, microscopy and SEM) and DA (n = 206, LC–MS/MS) concentrations and fluxes. Results suggest that there was an abrupt shift towards greater frequency and higher magnitude Pseudo-nitzschia blooms and toxic DA flux events in the SBB after the year 2000. SEM analysis of sediment trap material indicates that these events were mainly blooms of P. australis, with cell fluxes increasing by an order of magnitude from a maximum of 4.5 × 10⁶ cells m⁻² d⁻¹ pre-2000, to as high as 3.2 × 10⁸ cells m⁻² d⁻¹ thereafter. Similarly, sediment trap DA fluxes increased by an average of 13.4 μg m⁻² d⁻¹, with only one large event (>5 μg m⁻² d⁻¹) from 1993 to 1999 versus 16 large DA events from 2000 to 2008. While the causes of this abrupt shift remain ambiguous, we suggest that this shift may be related to natural climate variability associated with a change in phase of the North Pacific Gyre Oscillation (NPGO) and its potential influence on the composition and magnitude of waters that are upwelled into the SBB.     

Fatty acid composition in adductor muscle of juvenile scallops (Pecten maximus) from five Norwegian populations reared in the same environment

Adductor muscle fatty acid (FA) composition was compared in Pecten maximus juveniles originating from five locations along the Norwegian coast from 59°N to 65°N to detect possible population differences. Broodstock sized scallops were translocated to sea conditions by a scallop hatchery near Bergen (60°N) before spawning in February 2006. The scallop larvae and juveniles were reared in the same environment for two years and 10 individuals from each population were sampled in May 2007 and in May 2008 for analysis of the FAs in the adductor muscle. The total lipid content determined as total amount of FAs were 5.7 ± 0.3 mg per g tissue, and no significant difference was found among the five populations. The polyunsaturated FAs made up close to 60% of the total, with 20:5n3 and 22:6n3 dominating. The saturated FA content was approximately 30%, while the monounsaturated FA were less abundant (7–10%). The FA composition of the muscles of the five populations was similar within each year, with larger differences between the years. Multivariate, supervised learning method PLS, applied pairwise, showed distinct FA composition, between the scallops from the five locations, indicating population differences. The relatedness between the populations was different in the two years, but the distinct FA profiles of the adductor muscle could be used to distinguish between scallop populations on a local scale. The results indicate habitat-specific lipid metabolism which may have important implications for the scallop aquaculture industry in the context of producing well adapted individuals for the specific locations.     

Purification and partial characterization of chymotrypsin-like proteases from the digestive gland of the scallop Pecten maximus

Three variants of a chymotrypsin-like protease were purified from scallop digestive glands successively by ion-exchange, gel filtration and high-performance liquid chromatographies. Enzyme activity was detected using succinyl-Ala-Ala-Pro-Phe-p-nitroanilide as a specific synthetic substrate for chymotrypsin. This proteinase was inhibited by chymostatin, diisopropylfluorophosphate and phenylmethylsulfonyl fluoride. Estimated molecular mass of the purified enzyme is around 32 kDa. These isoenzymes exhibit very low activities in hydrolyzing small synthetic specific substrates used for trypsic, elastolytic and collagenolytic measurements and referred mainly to a chymotrypsin-like proteinase. Very few differences were measured concerning pH profiles among the three isoenzymes. Stability is higher at low temperature for two variants. An N-terminal analysis was performed on one variant (B) among the three isoenzymes. The alignment of the N-terminal amino acid sequence indicates some homologies with abalone chymotrypsin-like protein and arthropod chymotrypsin proteases as well as with vertebrate serine protease counterparts (trypsin, chymotrypsin and elastase).     

Digestive enzymes in marine species. II. Amylase activities in gut from seabream (Sparus aurata), turbot (Scophthalmus maximus) and redfish (Sebastes mentella)

The amylase activity of the digestive tract of three carnivorous fish species (Sparus aurata, Scophthalmus maximus and Sebastes mentella) has been studied. The activity of seabream and turbot showed its maximum at neutral pH (7.0–7.5); meanwhile, the activity of redfish had an optimum pH at 4.5-5.0. The t° function ranged between 35 and 45°C for the three species. The Arrhenius plots of the intestinal activities of seabream and turbot showed breakpoints at temperatures close to those of their physiological activities. High saline concentrations inhibited the activity of seabream and turbot and activated the activity of redfish. Seabream activity was absolutely dependent on calcium ions. On the contrary, redfish activity was only detected in the absence of this metal. Studies carried out by using several effectors suggested that the activities found in these three species are different. Considering our results from a point of view of the environmental conditions of these species, it might be concluded that enzymatic digestion of dietary carbohydrates proceeds at very low rate. Physiological implications are discussed.     

Alexandrium (Dinophyceae) species in Malaysian waters

A study was carried out to determine the presence of paralytic shellfish poisoning (PSP) toxin-producing dinoflagellates in the coastal waters of Peninsula Malaysia. This followed first ever occurrences of PSP in the Straits of Malacca and the northeast coast of the peninsula. The toxic tropical dinoflagellate Pyrodinium bahamense var. compressum was never encountered in any of the plankton samples. On the other hand, five species of Alexandrium were found. They were Alexandrium affine, Alexandrium leei, Alexandrium minutum, Alexandrium tamarense and Alexandrium tamiyavanichii. Not all species were present at all sites. A. tamiyavanichii was present only in the central to southern parts of the Straits of Malacca. A. tamarense was found in the northern part of the straits, while A. minutum was only found in samples from the northeast coast of the peninsula. A. leei and A. affine were found in both the north and south of the straits. Cultured isolates of A. minutum and A. tamiyavanichii were proven toxic by the receptor binding assay for PSP toxins but A. tamarense clones were not toxic. Mean toxin content for the A. tamiyavanichii and A. minutum clones were 26 and 15 fmol per cell STX equivalent, respectively. This study has provided evidence on the presence of PSP toxin-producing Alexandrium species in Malaysian waters which suggests that PSP could increase in importance in the future.     

DSP toxin production de novo in cultures of Dinophysis acuminata (Dinophyceae) from North America

For decades, many aspects of Dinophysis biology have remained intractable due to our inability to maintain these organisms in laboratory cultures. Recent breakthroughs in culture methods have opened the door for detailed investigations of these important algae. Here, for the first time, we demonstrate toxin production in cultures of North American Dinophysis acuminata, isolated from Woods Hole, MA. These findings show that, despite the rarity of Dinophysis-related DSP events in North America, D. acuminata from this area has the ability to produce DSP toxins just as it does in other parts of the world where this species is a major cause of DSP toxicity. In our cultures, D. acuminata cells were observed feeding on Myrionecta rubra using a peduncle. Culture extracts were analyzed using LC–MS/MS, providing unequivocal evidence for the toxin DTX1 in the Dinophysis cultures. In addition, a significant amount of an okadaic acid diol ester, OA-D8, was detected. These results suggest that this Dinophysis isolate stores much of its OA as a diol ester. Also, toxin PTX-2 and a hydroxylated PTX-2 with identical fragmentation mass spectrum to that of PTX-11, but with a different retention time, were detected in this D. acuminata culture. This demonstration of toxin production in cultured North American Dinophysis sets the stage for more detailed studies investigating the causes of geographic differences in toxicity. It is now clear that North American Dinophysis have the ability to produce DSP toxins even though they only rarely cause toxic DSP events in nature. This may reflect environmental conditions that might induce or repress toxin production, genetic differences that cause modifications in toxin gene expression, or physiological and biochemical differences in prey species.     

Prorocentrum lima (Dinophyceae) in northeastern USA coastal waters: II: Toxin load in the epibiota and in shellfish

The seasonal variation in diarrhetic shellfish poisoning (DSP)-type toxins was followed in the epibiotic community and in shellfish between 41° and 44°N in coastal waters of the northwest Atlantic during a 2-year period. Low levels of okadaic-acid equivalents were detected at all stations in the <90 μm fraction of the collected epibiota as measured by the protein phosphatase inhibition assay, but only 3.5% of the samples had values greater than 100 ng (g dry weight of epibiota)⁻¹. No seasonal pattern could be detected due to differences in intensity, duration and timing of toxin content in the epibiota between the 2 years and between stations. Nevertheless, the concentration of DSP-type toxins in the epibiota correlated weakly but significantly with the abundance of Prorocentrum lima, when data from all stations were considered. A very limited toxin uptake by shellfish was measured at only one station in October and November 2001 and in June and July 2002 at times of maximum cell concentration of P. lima in the epibiota. Toxin levels in shellfish remained well below regulatory limits that would have required quarantine or bans on harvesting. Results from our 2-year survey suggest that, at this time, the threat of DSP events appears minimal. However, the presence of a known toxin producer and its demonstrated ingestion by shellfish would argue for further studies to better understand conditions leading to DSP outbreaks generated by an epiphytic dinoflagellate.     

A study of spatial variability of domoic acid in razor clams: recommendations for resource management on the Washington coast

Domoic acid (DA) levels in Washington State razor clams (Siliqua patula) have been extremely variable and unpredictable, resulting in emergency closures of harvest areas in 1991, 1998, and 1999. Information concerning toxin variability relative to sampling location is important in developing a more reliable plan for managing DA outbreaks. In November 1998, record levels of DA in razor clams (up to 295 ppm) were reported at Kalaloch beach, Washington. A razor clam resource survey conducted there during the summer of 1999, along with the long retention time of DA in the clams, permitted a study of DA levels as a function of tidal elevation and north-south beach location. During the summer low tides (28–31 July 1999) razor clams were collected from six east-west transects, approximately 1.6 km apart. Each clam was individually analyzed for DA to determine the distribution of toxin between (interspecific variability) and within (intraspecific variability) transects. While average DA levels were similar between transects, toxin levels varied substantially among individual clams. The coefficient of variation among all samples (n=445) was 122%, indicating that harvest closures based upon composite analyses of only six clams could be in error. Here we recommend a sampling strategy of razor clams for regulatory purposes that will result in DA measurements more representative of total population toxin values.     

Paralytic shellfish toxin profile in strains of the dinoflagellate Gymnodinium catenatum Graham and the scallop Argopecten ventricosus G.B. Sowerby II from Bahía Concepción, Gulf of California, Mexico

Gymnodinium catenatum Graham is a paralytic shellfish poison (PSP) producer that was described for the first time from the Gulf of California in 1943. During the last decade, its distribution along the Mexican Pacific coastline has increased. In Bahía Concepción, a coastal lagoon on the western side of the Gulf of California, G. catenatum has been linked to significant PSP concentrations found in mollusks. In this study, we describe the saxitoxin profile of 16 strains of G. catenatum, and catarina scallops (Argopecten ventricosus) from Bahía Concepción. Toxins were analyzed by HPLC with post-column oxidation and fluorescence detection. The average toxicity of the G. catenatum strains was 26.0±6.0 pg and 28.0±18.0 pg STX eq/cell after 17 and 22 days of growth, respectively. Ten toxins were recorded, but only dcSTX, dcGTX2, dcGTX3, C1, and C2 were always present in all strains at both growth stages. Since toxin profiles in scallops were similar to the cultures, biotransformations are not significant in catarina scallop. NeoSTX, GTX2, GTX3, and B2 were present in some G. catenatum strains and their presence varied with the age of the culture. In scallop samples, dcSTX, dcGTX2, and dcGTX3 were the most abundant toxins, and from the C-toxin group, only C2 was found. This unique toxin profile can be used as a biomarker for this population, when compared with strains of G. catenatum from other geographic regions.     

Interaction between bacteria and the domoic-acid-producing diatom Pseudo-nitzschia multiseries (Hasle) Hasle; can bacteria produce domoic acid autonomously?

The interaction between bacteria and phytoplankton is increasingly becoming recognised as an important factor in the physiology of toxin production and the dynamics of harmful algal blooms (HABs). Bacteria can play a direct or indirect role in the production of biotoxins once solely attributed to microalgae. Evidence implicating bacteria as an autonomous source paralytic shellfish poisoning biotoxins raises the question of autonomous bacterial toxigenesis of the neurotoxin domoic acid (DA), the cause of amnesic shellfish poisoning. Here, we examine whether the previously observed bacterial enhancement of DA production by Pseudo-nitzschia multiseries (Hasle) Hasle may be attributable to independent biotoxin production by the extra-cellular bacteria associated with this diatom. The growth and toxicity of six cultures of xenic P. multiseries clone CLN-1 were followed for 24 days. Up to day 14 (mid-stationary phase), DA production was not statistically different among culture flasks. On day 14, P. multiseries cells were removed by gentle filtration from a set of triplicate flasks, leaving the bacteria in the filtrate. Following the removal of the algal cells, DA in the filtrate ceased to increase. Instead, DA levels continuously declined. A follow-up experiment determined that this was likely caused by photodegradation rather than by bacterial degradation. We conclude that after removing P. multiseries cells, the extra-cellular bacteria remaining in the filtrate were incapable of autonomous DA toxigenesis, even in the presence of P. multiseries exudates. However, scanning electron microscopy revealed that P. multiseries cells harboured epiphytic bacteria, the importance of which can still not be ruled out in DA production.