Printer Friendly

Use of a liquid chromatographic method for assessment of paralytic shellfish poisoning toxin profiles in mussels and clams from Uruguay.

ABSTRACT For over 30 y, a mouse bioassay has been used to monitor paralytic shellfish poisoning (PSP) toxicity in shellfish harvested along the coast of Uruguay. Although providing a good control mechanism for the safety of shellfish consumers, the assay is known to have several disadvantages. To assess the potential for use of liquid chromatography with fluorescence detection, a study was conducted to analyze shellfish samples harvested over a 20-y period using this alternative method. Mussels and clams were analyzed using the AOAC Official Method 2005.06 with results compared with those obtained from the original mouse bioassay and used to determine the presence of specific toxin congeners. The chromatographic results from nontoxic samples indicated no specificity issues resulting from the potential presence of naturally fluorescent coextractives but with complex and variable toxin distributions in samples designated PSP-positive by the bioassay. Although some species-related differences in toxin profile may be attributed to differences in shellfish metabolism, clear profiles were noted in shellfish harvested during two distinct phytoplankton blooms of Gymnodinium catenatum and Alexandrium tamarense. The work therefore showed evidence for the presence of a wide range of toxin congeners and the need for any replacement quantitative method to be fully capable of analyzing the major toxins of importance. Some differences between the toxicity results determined by the two methods indicated, however, the need for further investigations to assess the potential presence of other toxins, which may remain undetected using AOAC 2005.06.

KEY WORDS: paralytic shellfish poisoning, liquid chromatography, mouse bioassay, toxin profiles, mussels, clams. Uruguay

INTRODUCTION

Filter-feeding bivalve molluscs, such as mussels, scallops, and clams, are known to feed on certain photosynthetic organisms in the marine water column, which naturally contain a range of toxic compounds, known as phycotoxins (Hallegraef 2003). Harmful toxins may accumulate in the shellfish tissue, potentially impacting on the health of the human shellfish consumer, with a number of these naturally occurring toxins resulting in serious illness (Shumway 1990, 1995). Paralytic shellfish poisoning (PSP) is a potentially fatal syndrome resulting from ingestion of shellfish containing toxins known as saxitoxins. These comprise a range of more than 30 different analogues, which are associated with a number of toxic phytoplankton, including the genera of Alexandrium species, together with Gymnodinium and Pyrodinium (Wright 1995, Llewellyn 2006). To ensure consumer protection and comply with European Regulations, monitoring of the phytoplankton in water and PSP toxins (PST) in shellfish is a statutory requirement for European Union (EU) member states and for other non-EU countries wishing to export shellfish products to countries within the EU (Anon 2004). EC Regulation 853/2004 describes the statutory limits of PST in flesh with the maximum permitted limit (MPL) of 800 (ag saxitoxin equivalents (STX eq.) per kg of shellfish flesh (Anon 2005b).

The coast of Uruguay is relatively small (<300 km) in comparison with other South American countries and is influenced greatly by the major water bodies of both the Atlantic Ocean and the La Plata River. The Rio de La Plata is notably the widest river and estuarine system in the world, with studies showing a declining environmental health (Gomez-Erache et al. 2001). Along the coast, shellfish harvesting operations range from small localized fishing communities to larger operations. Bivalve species harvested include natural beds of mussels (Mvtilus edulis), yellow clams (Mesodesma mactroides), and wedge clams (Donax hanky anus), plus offshore harvesting areas of Patagonian scallops (Zigoclamys patagonia) and the venerid clams (Pitar rostrata). At present Z. patagonia and P. rostrata are not subject to commercial harvesting operations.

Since 1980, given the potentially serious impact of toxic phytoplankton bioaccumulation in shellfish within South America (Lagos 2003), the presence of both harmful phytoplankton in the water and toxicity in shellfish has been monitored on a regular basis. Phytoplankton testing revealed the first occurrence of blooms of Alexandrium tamarense during 1991, with repeated blooms occurring through the 1990s (Brazeiro et al. 1997, (Mendez & Ferrari 2003). The presence of this species in the winter is thought to be dependent on the northerly movements of subantarctic waters toward Uruguay in combination with a decrease in the flow of fresh water from the La Plata River, resulting in the movement of oceanic water into the estuarine system (Mendez et al. 1996, Brazeiro et al. 1997, Carreto et al. 1998). Toxic blooms of Gymnodinium catenatum were first reported during the summer of 1992, with repeated occurrence of the species ever because and the highest cell concentrations measured when water temperatures fell between 22-24[degrees]C (Mendez & Medina 2004). These temperatures occur normally during the summer and autumn seasons. A dependency on the presence of cyst seedbeds has also been noted (Mendez 1995). A third species of PSP-producing phytoplankton was reported as blooming during 1992 and originally identified as Alexandrium catenella, although this has because been reidentified as Alexandrium fraterculus (Lagos 2003).

Paralytic shellfish poisoning outbreaks have been observed in shellfish sampled during both summer and winter blooms of toxic phytoplankton. Since the first detection of the algae in 1991, Alexandrium tamarense has resulted in toxicity within shellfish in Uruguayan waters during winter-spring. Maximum PSP toxicity due to this algal species was found to vary considerably from year to year, peaking at 82,850 [micro]g STX eq./kg during 1991 (Medina et al. 1993, Mendez & Medina 2004, Medina, unpublished data). The first outbreak relating to Gymnodinium catenatum was detected during the summer of 1992 with toxicity measured in both wedge clams and mussels. The highest PSP toxicity related to G. catenatum blooms observed to date was in 1992 when 14,780 [micro]g STX eq./kg was measured, associated with a maximum bloom density of 146,000 cells/1 (Mendez & Medina 2004, Medina, unpublished data). When PSP toxicity events are detected during routine monitoring of representative shellfish by the mouse bioassay analysis (MBA), the area in which the toxicity is detected is closed for harvesting and a ban is established until the PSP levels are found to reduce to below the MPL through subsequent retesting. As a result, the proliferation of toxin-producing algal blooms is found to have a significant socioeconomic effect, with reductions in income from tourism and fishing, together with the negative effects relating to the public perception of seafood safety (Mendez et al. 2002).

In the Uruguay, the organization Direccion Nacional de Recursos Acuaticos (DINARA) is the government authority responsible for official control monitoring and testing of fishery products, including bivalve molluscs. Since 1980, the organization has used the reference method for detecting PST; the PSP MBA (McFarren 1959, Anon 1984, 2005a, 2006a). This method has provided a good level of control and protection from PSP intoxication, with the provision of warnings of shellfish toxicity consequently enabling the closure of harvesting areas during period of toxicity. Although the method has not been formally validated, it is simple and relatively cheap to use. The accuracy of the method is, however, known to be affected by matrix coextractives (Turner et al. 2011, 2012, Ben-Gigirey et al. 2012b) and there are ethical issues associated with the use of live animals for food safety control. In addition, the method is insensitive, being less capable of providing any early warnings of increasing toxicity in comparison with alternative and more sensitive nonanimal testing methods (Etheridge 2010). In recent years, a number of countries have moved away from reliance on the MBA with the implementation of an alternative testing method involving precolumn oxidation liquid chromatography with fluorescence detection (LC-FLD). The method has been validated (Lawrence et al. 2005) and adopted by the AOAC as an official, first-action method (AOAC 2005.06; Anon 2005b). Subsequently, the method was approved by the EU as an alternative to the MBA (Regulation EC 2074/2005 as amended) (Anon 2006b). In the United Kingdom, the method has been further validated to incorporate additional toxins and extended to include shellfish species harvested in UK waters (Turner et al. 2009, 2010. 2011, Turner & Hatfield. 2012). This has enabled both the assessment of relative method performance between the LC-FLD and MBA methods, and the generation of quantitative data for specific PST present in contaminated shellfish flesh. Qualitative and quantitative PST profiles' data consequently provide additional useful information on the toxins, which regularly contribute to shellfish toxicity as well as provide an earlier indication of toxin presence during the early growth stages of a phytoplankton bloom.

This manuscript summarizes the work conducted on clam and mussel shellfish samples harvested along the coast of Uruguay between 1991 and 2011. Total PST content was determined by LC-FLD and compared with the MBA, with quantitative data facilitating the assessment of common profiles and any dependency on external factors such as temporal or spatial variability, species or phytoplankton source. The study was aimed at providing potentially useful additional information regarding the occurrence of PSP events, their patterns of dependency, ultimately contributing toward the current state of knowledge with the prevalence and intensity of PSP in Uruguayan shellfish harvesting areas.

MATERIALS AND METHODS

Samples

The study consisted of 15 mussel (Mytilus edulis) and 5 wedge clam (berberecho) (Donax hanleyanus) samples (Table 1). These had been harvested and collected between September 1991 and March 2012 in three different geographical areas along the Uruguayan coast (Fig. 1). Samples were selected to include those harvested during several PSP events as well as two PSP-negative samples. Half of the 20 samples were collected in the winter/spring time during blooms of Alexandrium tamarense between August and October 1991 to 2011 with eight harvested in the summer/early autumn between February and April 1992 to 2012 during blooms of Gymnodinium catenatum. Two further samples were provided, consisting of one clam and one mussel sample found to be free from PSP as determined by the MBA. The PSP-positive samples were chosen to be representative of each of the PSP events recorded in Uruguay between 1991 and 2012.

Mussels and clams taken for official control monitoring were transported to DINARA (where they were homogenized and extracted prior to MBA). Hydrochloric acid extracts not used for the MBA were subsequently placed into long-term frozen storage. During 2012, HCl-extracts were thawed, 10 ml subsamples of a series of randomly selected positive and negative samples were taken, refrozen, and shipped to Cefas for confirmatory analysis. Twenty frozen HCl samples were sent under temperature-controlled conditions and received at the Cefas laboratory in April 2012 after 5 days transportation. On receipt at Cefas, the samples were checked and confirmed as being below +10[degrees]C, before storing at -20[degrees]C until required for analysis.

Analytical Methods

Reagents and Chemicals

Solvents and chemicals used for sample preparation and LC analysis were analytical or LC grade. Certified reference materials (CRM) used were decarbamoyl saxitoxin (dcSTX), neosaxitoxin (NEO), saxitoxin di-hydrochloride (STX), gonyautoxins 1-5 (GTX1-5), A'-sulfocarbamoyl-gonyautoxins 2 and 3 (C1&2), decarbamoylneosaxitoxin (dcNEO), and decarbamoyl gonyautoxins 2 and 3 (dcGTX2&3). These reference standards were purchased from the Institute for Biotoxin Metrology, National Research Council Canada (NRCC, Halifax, Nova Scotia, Canada). In addition, noncertified standards of C3&4 and GTX6 were also obtained from NRCC, both of which were provided with well-characterized concentration declarations. Ampoules of CRM were opened and known volumes diluted in deionized water to produce concentrated stock standard solutions prior to dilution in 0.1 mM acetic acid for preparation of instrument calibration standards.

Shellfish Extraction and MBA

Shellfish homogenates were extracted at DINARA within 1 day of sample collection using the methodology described by AOAC Official Method 959.08 (Anon 2005a). Specifically, 100 g shellfish homogenates were mixed with 100 ml 0.1 M HCl, with the pH adjusted to <4.0. The mixture was boiled gently for 5 min before cooling, readjusting the pH to 2.0-4.0 if required and filtered prior to analysis.

The MBA utilized triplicate mice (CD-I strain, weight range 19-21 g, males, bred in a standard facility laboratory). Testing was conducted following the guidance of McFarren (1959) and Anon (1984, 2005a, 2006a). Sample toxicities were calculated from the median death times of the mice and expressed here in terms of STX eq./kg shellfish flesh.

Sample Clean-Lip

Hydrochloric acid extracts received at Cefas were subjected to semiautomated solid phase extraction (SPE) clean-up using C18-bonded cartridges (C18-T SPE, Phenomenex, Manchester, UK). SPE eluants were pH adjusted to 6.5 [+ or -] 0.5 before diluting to a final volume of 4.0 ml. All samples were further cleaned up using a refined ion-exchange (COOH) SPE clean-up (Turner et al. 2009) resulting in the separation of toxins into three fractions (F1-F3) containing toxins of three different structural classes. Fraction Fl, containing the C toxins (C1&2 and C3&4), F2 (containing the GTX1-6 and dcGTX2&3), and Fraction F3 (containing the carbamates STX, dcSTX, dcNEO, and NEO) were all subjected to further analysis.

Analytical Procedure

C18-cleaned extracts of each sample were derivatized using periodate oxidation (Anon 2005b) prior to qualitative LC-FLD analysis to determine both the presence of PST in the samples and the potential presence of iV-hydroxylated toxins. All samples were subsequently quantified against a minimum of a five level calibration following peroxide oxidation of C18-cleaned extracts. If the screen results showed the potential presence of N-hydroxylated PST, then periodate oxidation of fractions F1-F3 was conducted prior to additional LC analysis. Unoxidized C18-cleaned extracts were also analyzed with the peak area responses of any naturally fluorescent chromatographic peaks with the same retention time as PST subtracted from the toxin peak areas of the oxidized sample. Toxicity equivalence factors (TEF) were taken from those published by EFSA (EFSA 2009). The highest TEF was used for each isomeric pair (GTX1&4, GTX2&3, C1&2, and dcGTX2&3). Concentrations of individual toxins were calculated in units of STX di-HCl eq./kg, to enable the assessment of toxin profiles in terms of toxicity. Paralytic shellfish poisoning toxin concentrations were subsequently used to calculate estimated sample toxicities in terms of [micro]g STX di-HCl eq./kg. Samples were analyzed blind with no prior knowledge of toxicity results obtained using the MBA.

Instrumental Set-Up

Agilent (Stockport, UK) fluorescence detectors (1200 model FLD) were used for the detection of the PST oxidation products. Fluorescence excitation was set to 340 nm and emission to 395 nm. Mobile phase A composed of 0. 1 M ammonium formate, adjusted to pH 6 [+ or -] 0.1 with 0.1 M acetic acid, mobile phase B was prepared from 0.1 M ammonium formate with 5% acetonitrile, also adjusted to pH 6 [+ or -] 0.1 with 0.1 M acetic acid. The mobile phase was delivered by an Agilent 1200 series LC at a flow rate of 2 ml/min. Chromatographic separation was performed using a Gemini C18 reversed-phase column (150 mm x 4.6 mm, 5 [micro]m; Phenomenex, Manchester, UK) with a Gemini C18 guard column (both set at 35[degrees]C). The LC gradient was as follows: 0%-5% mobile phase B in the first 5 min, 5%-70% B for the next 4 min, hold at 70% B for 1 min, and back to 100% A over the next 2 min, 100% A was held for a further 2 min to allow for column equilibration prior to subsequent sample injections.

Data Assessment

Total PSP toxicities were estimated from the LC-FLD results generated following fully quantitative analysis of each of the shellfish sample extracts received. Initially, LC-FLD results were compared against the MBA, to determine any potential issues with the comparison between the two approaches. Individual toxin concentrations were tabulated in each sample and assessed to determine the presence of any visual patterns in toxin profile. In particular, differences in profile were examined in relation to the potential effects of sample species, spatial variability, and both intra- and interannual temporal variability.

RESULTS

Total PSP

Table 1 summarizes the PSP toxicity results as determined by MBA and estimated using the LC-FLD method. Results showed a large variability in levels of PST, with nine samples showing total PST above the MPL (800 [micro]g STX eq./kg) as determined by LC-FLD (Fig. 2). On average, higher PST were measured in the mussel samples as compared with the clams, although it was noted that none of the mussel and clam samples were harvested on exactly the same date other than the two PSP-negative samples (sample number 19 and 20). Total PST levels were found to be significantly higher on average in the shellfish sampled during the blooms of Alexandrium tamarense in comparison with shellfish sampled during blooms of Gymnodinium catenatum, with mean total PST levels of 4,182 and 675 [micro]g STX eq./kg, respectively.

Mean Toxin Profiles

The clam and mussel samples found to be PSP negative by MBA were analyzed to assess the presence of low concentrations of PST or other naturally fluorescent coextractive compounds. Trace levels of C1&2, dcSTX, and STX were observed in both samples (e.g., Fig. 3A), but at concentrations below the limit of quantitation and with total PST levels of ~ 10 [micro]g STX eq./kg well below the designated reporting limit of 160 [micro]g STX eq./kg. Following LC-FLD analysis of both peroxide and periodate oxidized CI8 SPE-cleaned extracts of PSP-positive samples, a wide range of PST was qualitatively identified. Figure 3 illustrates these results, showing chromatograms for sample 1 harvested during the blooms of Gymnodinium catenatum and for sample 15, harvested during an Alexandrium tamarense growth period. Toxins identified included the N-hydroxylated toxins GTX1&4, NEO, and dcNEO, the decarbamoyl toxins dcGTX2&3 and dcSTX together with the remaining carbamates and gonyautoxins STX. GTX2&3, and GTX5. As a result, all the PST currently available as CRM standards from the NRCC were detected in these samples. In addition, both C3&4 and GTX6 toxins were also identified, following periodate oxidation and analysis of fractions F1 and F2, respectively. Figure 3 also shows the chromatograms obtained following the analysis of unoxidized samples 1 and 15. No significant fluorescent peaks were observed, with the exception of one very small peak corresponding in retention time to the primary quantitation peak for dcSTX in sample 15.

Figure 4A illustrates the mean toxin profile quantified in the 18 PSP-positive samples analyzed in this study. Gonyautoxins 1 and 4 was found to be present at the highest concentrations in terms of STX eq., but with significant contributions from GTX2&3, dcSTX, and dcGTX2&3. Smaller relative proportions were also evident for dcNEO, NEO. STX, and GTX5 with only trace quantities of GTX6 and C3&4 quantified. High variabilities of the mean profile were noted.

Toxin Profile Variability

Figure 4B-E illustrate the variability of toxin profiles in relation to a number of different factors, including sample harvesting location, date of harvest, sample species, and source phytoplankton. Figure 4B summarizes the mean toxin profiles determined from shellfish harvested in each of the three harvesting zones (Maldonado, Rocha-La Paloma, and Rocha-Punta del Diablo; Fig. 1). Mean proportions appear similar within each of the zones, with each again associated with large variabilities. Figure 4C shows some differences in the mean profiles quantified for each of the two shellfish species analyzed. On average, the mussel samples were found to contain higher proportions of GTX1&4 and GTX2&3, whereas the clam samples contained higher levels of the decarbamoyl toxins dcSTX and dcGTX2&3. Figure 4D summarizes the mean toxin profiles and associated variabilities in shellfish harvested over three separate time points, 1991 to 1993, 1996 to 1998, and 2003 to 2011. Again, high variabilities were observed in the profiles quantified, with no evidence for any differences associated with the year of harvesting.

Figure 4E graphs the mean profiles associated with samples harvested in the two time frames within each year. One group of samples was harvested during August to October, with the second group harvested between February and April. Although the profile variabilities are still high for some toxins, notably dcGTX2&3, the results show clear and significant differences in the profiles from samples harvested at different time points within the year. The samples harvested between August and October contain very high proportions of GTX1&4, together with significant contributions from GTX2&3 and STX. Lower levels of NEO were also detected in the majority of samples. Only trace levels (<limit of quantitation) of the decarbamoyl toxins dcSTX and dcGTX2&3 were detected, with no dcNEO present in any of the samples. Toxin profile results determined in shellfish harvested between February and April were found to contain predominantly the decarbamoyl toxins dcGTX2&3, dcSTX, and dcNEO, together with lower but significant proportions of C1&2, GTX5, STX, and GTX2&3. Other C toxins were present (C3&4) at trace levels (<0.1% of total STX eq.) in only three of the samples, with GTX6 present in more samples but also at low concentrations (<1% on average).

DISCUSSION

Total PSP

Twenty samples were assessed for PST using the pre-column oxidation AOAC 2005.06 method. Although the total number of samples was small, the number was deemed sufficient to enable the characterization of toxin levels and associated profiles in a range of Uruguayan shellfish harvested with both spatial and temporal variability. Results indicated a wide variability in total toxicity, as demonstrated by both the biological and chemical assays, with notable differences in total PST quantified in the two species examined. On average, higher PST levels were determined in shellfish relating to the source Alexandrium tamarense phytoplankton. This may relate to the relative cell concentrations of the source phytoplankton in the harvesting areas, differences in toxicity of the different algal species, or differences between feeding rates of the algae in both the species of bivalves tested (Ichimi et al. 2001, Asakawa et al. 2005, 2006, Samsur et al. 2007).

The difference between the results obtained by LC-FLD in comparison with MBA is notable for a significant number of samples (Table 1), noting the sample size is relatively small for a full method comparison. The mean LC:MBA ratio was found to be 0.82 but with the individual ratios varying widely. The majority of results were within a factor of 2 from each other, two samples in particular (samples 9 and 17) returned LC-FLD results substantially lower than the original MBA. With no shellfish species or phytoplankton-related effects apparent (Table 1), there were no clear factors explaining these large differences. The MBA conducted in this study was performed to establish guidelines using calibrated systems. Similarly, there were no issues associated with the LC-FLD analysis, with batch controls all acceptable for the analyzed samples. Although comparative method performance was not the goal of this study, the observations are still noteworthy.

The MBA is still the internationally accepted standard for determination of PST in bivalve molluscs. Although the MBA provides a measure of total sample toxicity in STX eq., there are a number of issues associated with the method. These include a high method variability as well as assay interferences resulting from pH and the presence of salts and metals (Stephenson et al. 1955, Shantz, 1960, Wilberg & Stephenson 1960, 1961, Prakash et al. 1971, Park et al. 1986, McCulloch et al. 1989, Nagashima et al. 1991, Aune et al. 1998, Vale & de Sampayo 2001, Turner et al. 2012). The lack of formal validation of the method is also notable (Guy & Griffin 2009). A number of papers have been published describing the comparison between the PSP MBA and LC-FLD methods for the analysis of both mussel and clam species. Previous validation work at Cefas demonstrated a good relationship between the performance of the two methods in both mussels (Turner et al. 2009) and clams (Turner et al. 2010), although the clam species assessed here were different to those formally validated in the United Kingdom. Other authors report variable results with the method. Ben-Gigirey et al. 2012a described a good overall correlation between results obtained by the two methods in a range of shellfish species containing both Alexandrium and Gymnodinium profiles, with toxicity data showing no significant difference between the two methods. Some samples, however, exhibited large differences in toxicity, which could not be fully explained through known factors such as the application of the highest TEF for toxin isomers or the absence of suitable standards for certain toxins.

One factor that may potentially affect such a comparison relates to the elapsed time between the original MBA and the HPLC analysis, which for some samples exceeded 20 y. During the long-term storage of acidic extracts, there is some potential for chemical conversion of toxin analogues to others with different toxic equivalences. The transformation of PST congeners is widely reported in the literature, with previous studies primarily describing toxin conversions either within live shellfish tissues through shellfish metabolism or within in vitro tissue samples due to other chemical, bacterial, or enzymatic action. Such transformations are known to occur in a wide variety of shellfish species and include a variety of mechanisms including epimerization, hydrolytic cleavage of A'-sulfo groups, reductive transformation, and enzymatic decarboxylation (Sakamoto et al. 2000, Sato et al. 2000, Smith et al. 2001, Blanco et al. 2003. Fast et al. 2006, Jaime et al. 2007, Samsur et al. 2007, Cho et al. 2008). The degradation of PST has also been reported through the activity of some bacteria (Donovan et al. 2008). In some instances, these may result in the transformation of analogues with low toxicity such as the 7V-sulfocarbamoyl PST to those with higher toxicity, such as the carbamoyl toxins. Extensive reviews are available on this subject (Cembella et al. 1994. Bricelj & Shumway 1998, Wiesse et al. 2010). Although there is much evidence for such processes within shellfish tissues, there is little published data concerning transformation of PST during long-term storage of acidic extracts of shellfish tissue. Certainly, any potential enzyme-triggered transformation or bacterial degradation should not occur in extracts due to the heating applied during the extraction process (Fast et al. 2006). Some authors have, however, reported stability problems with some PST analogues under storage conditions, notably an instability of STX and degradation of GTX under strongly acidic conditions, although under weak acidic conditions, the stability was deemed reasonable (Shimizu et al. 1975). Other reports have determined toxin stability to be least compromised at higher temperatures when stored at pH 3-4, whereas under frozen conditions no significant toxin degradation or transformation was detectable within 1 y (Indrasena & Gill 2000). Other authors have reported that STX can be stored at temperatures less than +5[degrees]C for up to 2 y in acidic solution without degradation or conversion (Alfonso et al. 1993). In this report, although toxin storage may potentially have been compromised, there were no indications from the data for any consistent decrease or increase in extract toxicity. Table 1 shows the sample extracts from 1991 exhibit both a slight overestimation in toxicity by LC-FLD (samples 3, 4, and 14) and a large underestimation (samples 16 and 17). Furthermore, one other sample showing a large discrepancy in calculated toxicity between the two methods was harvested more recently, during 2010, whereas other samples harvested prior to this date showed a much better agreement between results. Overall, there are no indications that either the storage conditions or pH conditions used during analysis are responsible for any repeatable bias in method performance.

There is the noted potential for other PST to occur, which may contribute significantly to the toxicity of the shellfish but which may not be detected using the AOAC 2005.06 LC-FLD method. The chromatograms obtained from samples 9 and 17 did show one minor additional chromatographic peak, which was unaccounted for during the quantitation, yet may relate to unknown toxins (data not shown). Previous work has also identified the presence of other toxins including a range of hydrophobic PST in Gymnodinium catenatum cultures, including the hydroxybenzoates GC1-GC3 (Negri et al. 2003, 2007), and a number of other related analogues termed GClb-GC6b (Vale 2008a), although these are less prevalent in shellfish (Vale 2008b). If present in these Uruguayan shellfish extracts, the benzoate PST will be retained during the C18-SPE clean-up process, and therefore remain undetected by the current fluorescence method unless changes are made to the clean-up and chromatographic methods. Another suite of analogues, known as the M toxins, are known to show weak fluorescence response, therefore being unsuitable for detection by LC-FLD (Dell'Aversano et al. 2005, 2008). Toxicity equivalence factors have not been formally assessed for these toxins, so it is difficult to assess whether the presence of these toxins would be detected by the MBA. Other work has also demonstrated the presence of the novel analogue 13-nor-decarbanoylsaxitoxin that may have contributed toward the discrepancy between LC-FLD and MBA results for a scallop sample harvested from Argentinean waters (Gibbs et al. 2009). It is noted, however, that low LC: MBA ratios were observed in samples harvested from regions with both Alexandrium tamarense and G. catenatum as the source algae, so a link between poor method agreement and phytoplankton species cannot be inferred from these results alone. Overall, further work would be required to assess the potential presence of these additional congeners using both LC with tandem mass spectrometry and/or LC-FLD with a modified clean-up procedure to assess any other potential causes of the significant differences between the two data sets.

Mean Toxin Profiles

Two samples found to be PSP negative by MBA were also found to contain PST below the LC-FLD reporting limit, trace levels of PST were detected by the chemical method. Other than the early eluting matrix component peaks observed before 2 min in the chromatograms, no other fluorescent peaks were detected. A wide range of PST were identified and quantified in the PSP-positive mussel and clam samples assessed in this study. In addition to the PST commonly associated with shellfish growing in Alexandrium blooms such as STX, NEO. GTX1-5, other toxins associated with Gymnodinium catenatum were quantified including dcNEO. Other toxins that are currently unavailable as certified reference standards were also detected, such as C3&4 and GTX6. Quantitation of these uncertified toxins was undertaken through comparison of toxin oxidation product peak areas against calibration standards prepared from standards received as a gift from the NRCC. Although these additional standards were uncertified, well-characterized concentrations were supplied, thereby enabling direct quantitation of both C3&4 and GTX6. Only trace quantities of these two toxins were determined in the samples analyzed in this study, so without such standards available, the analysis of total sample toxicity would not be significantly compromised; however, if relative concentrations of C3&4 and GTX6 did increase in source phytoplankton, or were found to be present in shellfish harvested from other areas or at other times, then it would be important to have access to certified or well-characterized standards to enable routine quantitation of these toxins. An alternative would be to conduct additional chemical transformation reactions of sample extracts to convert, for example, GTX6 into NEO (Ben-Gigirey et al. 2012b). Subsequent quantitation of the reaction products would then enable the estimation of C3&4 and GTX6 concentrations. It is, however, noted that these steps would add further complexity to the process and introduce additional uncertainty into the overall method. The preferred approach would, therefore, be quantitation against commercially available standards, which is especially important for countries where G. catenatum is prevalent (Ben-Gigirey et al. 2012a).

Analysis of unoxidized sample extracts provided no evidence for the presence of naturally fluorescent matrix coextractives, which might affect the quantitation of toxin oxidation products. Consequently, overestimation of toxin concentrations through the presence of nontoxin components would be unlikely as determined in these samples. If shown to be absent from a larger range of samples from Uruguayan shellfish beds, there could be the potential for eliminating the need for analysis of unoxidized sample extracts. This would result in the reduction in number of analyses required for the method, significantly increasing the throughput of the method.

Gonyautoxins 1 and 4 was present at the highest relative concentration of all the PST, with other major toxins including GTX2&3, dcGTX2&3, and dcSTX. The high variabilities associated with the mean profile shows the profiles to be inconsistent, with other parameters affecting the relative proportions of toxins, potentially including variabilities in toxin content within phytoplankton, and/or differences in shellfish metabolism between species or in different geographical areas. With the presence of other analogues including dcNEO. NEO, STX. and GTX5 together with trace quantities of GTX6 and C3&4, there was evidence for a wide range of toxins in the shellfish analyzed. Consequently, any future attempts to use analytical detection methods for routine testing would need to be capable of detecting and quantifying each of these toxins. Before any such approach could be implemented, the method performance characteristics would need to be verified for all commercial species of importance and for each of the toxins present at significant concentrations within the samples analyzed to date.

Toxin Profile Variability

Toxin concentration results did not indicate any relationship between toxin profiles and either the zone of origin or date of harvesting. Similar mean toxin profiles determined in samples harvested from each of the three harvesting zones, together with a notable high variability in mean profiles, showed little evidence for geographical effects.

Differences in mean toxin profiles were notable between the mussels and clams analyzed. Higher proportions of GTX1&4 and GTX2&3 were present in mussels, whereas decarbamoyl toxins dominated the profiles of the clam samples, including most notably dcGTX2&3 and dcSTX. This could potentially relate to enzymatic decarboxylation of carbamate and jY-sulfocarbamoyl PST as reported in a number of different clam species including Protothaca staminea (Sullivan et al. 1983), Mva arenaria (Buzy et al. 1994. Lin et al. 2004, Fast et al. 2006, Jaime et al. 2007, Samsur et al. 2007, Cho et al. 2008), and Spisula solida (Bricelj & Shumway 1998, Artigas et al. 2007. Turner et al. 2010, 2013a). Usually the A'-sulfocarbamoyl PST that include GTX5 and C1&2 are found to completely transform to decarbamoyl counterparts rapidly once exposed to enzymes in the clam tissues, either in vivo or in vitro (Artigas et al. 2007, Turner et al. 2013a). Here, however, GTX5 is still found in the clam tissue to the same extent on average as the relative proportions in mussels and C1&2 was also detected in two out of the four PSP-positive clam samples analyzed. It was also noted, however, that although the mean profiles appeared notably different between the two species, the variability of the profiles was again high, with Figure 4C showing large overlap between the standard deviations associated with both of the mean profiles for each toxin analyzed. Therefore, although there was some evidence for potential species-related differences between the toxin profiles measured in mussels and clams, it was not possible to confirm that this was the only factor involved in the profile differences from this data alone.

Results determined in samples harvested from either August to October or February to April were found to show the most significant differences in mean toxin profiles. These two periods are associated with the growth of different phytoplankton, with Alexandrium tamarense known to proliferate during August to October, and with the second group harvested between February and April when Gymnodinium catenation was detected (Brazeiro et al. 1997, (Mendez & Ferrari 2003, Mendez & Medina 2004). The samples containing high proportions of carbamate and gonyautoxins, most notably GTX1&4, STX, and GTX2&3 are those associated with A. tamarense. The lower relative concentrations of NEO and trace levels of decarbamoyl toxins resulted in an overall profile similar to those found in a range of shellfish species harvested in Great Britain in areas where Alexandrium species are known to grow (Turner et al. 2013b). Conversely, shellfish associated with blooms of G. catenatum were found to contain predominantly the decarbamoyl toxins dcGTX2&3, dcSTX, and dcNEO as recorded recently by Ben-Gigirey et al. 2012, together with lower but significant proportions of C1&2, GTX5, STX, and GTX2&3. These samples were also notable for the absence of the N-hydroxylated toxins GTX1&4 and NEO. Although present at low concentrations, the determination of C3&4 and GTX6 highlights the importance of the ability to determine these toxins, currently unavailable as certified reference standards.

The profile data, therefore, confirms previous work highlighting the presence and absence of dcGTX2&3 in Gymnodinium catenatum and Alexandrium species, respectively (Oshima et al. 1990, Sekiguchi et al. 2001, Ben-Gigirey et al. 2012a). In addition, these results show an agreement with previous work reported by Mendez et al. 2002, following the analysis of PST distributions by a postcolumn oxidation LC-FLD method (Oshima 1995) in cultures of Alexandrium tamarense and G. catenatum established from germinated sedimentary cysts harvested from Uruguayan waters. The yV-sulfocarbamoyl toxins C1-4, GTX5-6 were found to dominate the G. catenatum cultures, whereas the more potent carbamate toxins GTX1-4 were identified in the more toxic A. tamarense cultures. Consequently, the profile patterns observed in the shellfish analyzed in this study appear to agree well with the profiles described in the source phytoplankton.

Application of Chemical Testing Methods

A number of counties have moved away from use of the MBA in recent years to chromatographic alternatives, with the United Kingdom, Ireland, New Zealand, and Portugal using the pre-column oxidation LC-FLD method (AOAC 2005.06) for routine testing, with other countries and regions of the United States such as Canada, Norway, Alaska, and Maine moving to the use of a postcolumn oxidation LC-FLD method (AOAC 2011.02; van de Riet et al. 2011). Further developments include the interlaboratory validation and acceptance into U.S. legislation of a third alternative method (AOAC 2011.27) based on the use of a receptor binding assay for PST (Anon 2011). As such, there is a clear, albeit slow, global move away from reliance on the MBA for PSP testing, with increasing use and sole reliance on alternative testing methodologies.

There are many advantages associated with the move to nonanimal testing methods, including most notably the ethics associated with the use of animals for food safety testing. In addition, LC-FLD methods are not subjected to the same assay interferences from salts and metals, as noted in the MBA (Turner et al. 2012). The available alternative methods have, without exception, been subjected to full single laboratory and interlaboratory validation programs, and subsequently have been thoroughly assessed prior to becoming accepted as Official Methods of Analysis by the AOAC International. Furthermore, a number of laboratories have been running these methods routinely for many years, and validation has been extended to include both new toxins and additional shellfish matrices (Turner et al. 2009, 2010, 2011, Ben-Gigirey et al. 2012b). Consequently, a great deal of information is already available and published regarding their performance, robustness, and applicability to a large variety of shellfish species in a number of different countries.

There have, however, been various reported disadvantages with implementation of alternative methods to the MBA, including the requirement for CRM, method performance issues and complexity, high-level training requirements, the perception of lower throughput, and turnaround of chemical assays, an inability of the methods to detect emerging toxin congeners, as well as increased delivery costs (Guy & Griffin 2009, Ben-Gigirey et al. 2012b). For these reasons, there are major obstacles for the implementation of replacement PSP testing methods, most notably in countries where the costs of conducting validation, testing, and achieving method accreditation is of huge concern. Results from this study have shown the presence of complex toxin profiles, including toxins not currently available commercially as CRM. With both Alexandrium and Gymnodinium PST-producing phytoplankton affecting the coastal waters of Uruguay, a large amount of further work would be required to facilitate implementation of a LC-FLD method for routine monitoring, including the validation of the method to incorporate additional shellfish species of interest.

Overall, the use of the AOAC 2005.06 LC-FLD method has shown several advantages for the determination of PST in Uruguayan shellfish. Although there was some evidence for species dependency on the toxin profiles quantified, the data indicated very high levels of PST accumulation in both mussels and clams, thus affecting both of the main commercial species in Uruguay. Results have demonstrated a clear dependence on the source phytoplankton for both total sample toxicity and toxin profile type. This work has, therefore, demonstrated the use of an instrumental detection method for the analysis of shellfish from Uruguay for the detection of PST. The performance of any such replacement method would still need to be validated, however, both for the species of interest and to encompass the wide range of toxins identified in this study, and future work would be required to assess the potential presence of other toxins not detectable using precolumn oxidation LC-FLD.

ACKNOWLEDGMENTS

We wish to thank Professor Michael Rychlik from the Technische Universitat Miinchen for funding the time of Sophie Tarnovius during her Masters studies at Cefas. The work described in this article has been conducted in accordance with The Code of Ethics of the World Medical Association(Declaration of Helsinki) for animal experiments.

LITERATURE CITED

Alfonso, A., M. R. Vieytes, A. M. Botana, H. Geonaga & L. M. Botana. 1993. Preparation of mixtures of paralytic shellfish toxin (PSP) standards from mussel hepatopancreas. Fresenius J. Anal. Chem. 345:212-216.

Anon. 1984. AOAC Official Method. Paralytic shellfish poison. Biological method. Final action. Chapter 18: Fish and other marine products. In: L. Gershman, editor. AOAC Official methods for analysis, 14th edition. Arlington, VA: AOAC Inc. pp. 344-345.

Anon. 2004. Regulation (EC) No 854/2004 of the European Parliament and of the Council of 29th April 2004 laying down specific rules for the organisation of official controls on products of animal origin intended for human consumption. Off. J. Eur. Union L 226:83-127.

Anon. 2005a. AOAC Official Method 959.08. Paralytic shellfish poison. Biological method. Final action. Chapter 49: Natural toxins. In: M. W. Truckses, editor. AOAC official methods for analysis. 18th edition. Gaithersburg, MD: AOAC International, pp. 79-80.

Anon. 2005b. AOAC Official Method 2005.06. Quantitative determination of paralytic shellfish poisoning toxins in shellfish using pre-chromatographic oxidation and liquid chromatography with fluorescence detection. Gaithersburg, MD: AOAC International.

Anon. 2006a. Committee on toxicity of chemicals in food, consumer products and the environment. Statement on risk assessment and monitoring of paralytic shellfish poisoning (PSP) toxins in support of human health. COT statement 2006/08. Accessed July 2006. Available at: http://www.food.gov.uk/science/ouradvisors/toxicity/ statements/cotstatements2006/cotstatementpsp200608.

Anon. 2006b. Commission Regulation (EC) No 1664/2006 of 6th Nov. 2006 amending Regulation (EC) No 2074/2005 as regards implementing measures for certain products of animal origin intended for human consumption and repealing certain implementing measures. Off. J. Eur. Union L 320:13-45.

Anon. 2011. AOAC Official Method 2011.27. Paralytic shellfish toxins (PSTs) in shellfish: receptor binding assay. First action 2011. Gaithersburg, MD: AOAC International.

Artigas, M. L., P. J. V. Vale, S. S. Gomes, M. I. Botelho, S. M. Rodrigues & A. Amorim. 2007. Profiles of paralytic shellfish poisoning toxins in shellfish from Portugal explained by carbamoylase activity. J. Chrom. A. 1160:99-105.

Asakawa, M.. R. Beppu, M. Tsubota, K. Ito, H. Takayama & K. Miyazawa. 2005. Paralytic shellfish poison (PSP) profiles and toxification of short-necked clams fed with the toxic dinoflagellate Alexandrium tamarense. Shokuhin Eiseigaku Zasshi 46:251-255.

Asakawa, M., R. Beppu, K. Ito, M. Tsubota, H. Takayama & K. Miyazawa. 2006. Accumulation of paralytic shellfish poison (PSP) and biotransformation of its components in oysters Crassostrea gigas fed with the toxic dinoflagellate Alexandrium tamarense. Shokuhin Eiseigaku Zasshi 47:28-32.

Aune, T., H. Ramstad, B. Heinenreich. T. Landsverk, T. Waaler, E. Eggas & K. Julshamn. 1998. Zinc accumulation in oysters giving mouse deaths in paralytic shellfish poisoning bioassay. J. Shellfish Res. 17:1243-1246.

Ben-Gigirey, B.. M. L. Rodriguez-Velasco, A. Otero, J. M. Vieites & A. G. Cabado. 2012a. A comparative study for PSP toxins quantification by using MBA and HPLC official methods in shellfish. Toxicon 60:864-873.

Ben-Gigirey, B., M. L. Rodriguez-Velasco & A. Gago-Martinez. 2012b. Interlaboratory study for the extension of the validation of 2005.06 AOAC Official Method for dc-GTX-2,3. J. AOAC Int. 95:1-13.

Blanco, J.. M. I. Reyero & J. Franco. 2003. Kinetics of accumulation and transformation of paralytic shellfish toxins in the blue mussel Mytilus galloprovinciatis. Toxicon 42:777-784.

Brazeiro, A., S. M. Mendez & G. Ferrari. 1997. First toxic bloom of Alexandrium tamarense in Uruguay: associated environmental factors. Rev. Atlantica. 19:19-29.

Bricelj, V. M. & S. E. Shumway. 1998. An overview of the occurrence and transfer kinetics of paralytic shellfish toxins in bivalve molluscs. In: B. Reguera, J. Blanco, M. Fernandez & T. Wyatt, editors. Proceedings of the VIII International Conference on Harmful Algae, June 1999. Vigo, Spain: Xunta de Galicia and IOC of UNESCO, pp. 431-436.

Buzy, A., P. Thibalut & M. V. Laycock. 1994. Development of a capillary electrophoresis method for the characterization of enzymatic products arising from the carbamoylase digestion of paralytic shellfish poisoning toxins. J. Chrom. A. 688:301-316.

Carreto, J. I., N. Montoya, A. D. Cucchi Colleoni & R. Akselman. 1998. Alexandrium tamarense blooms and shellfish toxicity in the Argentine Sea: a retrospective view. In: B. Reguera, J. Blanco, M. L. Fernaandez & Y. T. Wyatt, editors. Harmful algae. Xunta de Galicia y IOC-UNESCO, pp. 131-134.

Cembella, A. D., S. E. Sumway & R. Larocque. 1994. Sequestering and putative biotransformation of paralytic shellfish toxins by the sea scallop Placopecten magellanicus: seasonal and spatial scales in natural populations. J. Exp. Biol. Ecol. 180:1-22.

Cho, Y., N. Ogawa. M. Takahashi, H. Lin & Y. Oshima. 2008. Purification and characterization of paralytic shellfish toxin-transforming enzyme, sulfocarbamoylase I, from the Japanese bivalve Peronidia venulosa. Biochim. Biophys. Acta 1784:1277-1285.

Dell'Aversano, C., P. Hess & M. A. Quilliam. 2005. Hydrophilic interaction liquid chromatography-mass spectrometry for the analysis of paralytic shellfish poisoning (PSP) toxins. J. Chrom. A. 1081:190-201.

Dell'Aversano, C., J. A. Walter, I. W. Burton, D. J. Stirling, E. Fattoruso & M. A. Quilliam. 2008. Isolation and structure elucidation of new and unusual saxitoxin analogues from mussels. J. Nat. Prod. 71:1518-1523.

Donovan, C. J., J. C. Ku, M. A. Quilliam & T. A. Gill. 2008. Bacterial degradation of paralytic shellfish toxins. Toxicon 52:91-100.

EFSA. 2009. Scientific opinion. Marine biotoxins in shellfish: saxitoxin group. Scientific opinion of the panel on contaminants in the food chain. The EFSA Journal 1019:1-76.

Etheridge, S. M. 2010. Paralytic shellfish poisoning: seafood safety and human health perspectives. Toxicon 56:108-122.

Fast, M. D., A. D. Cembella & N. W. Ross. 2006. In vitro transformation of paralytic shellfish toxins in the clams Mya areanaria and Protothaca staminea. Harmful Algae 5:79-90.

Gibbs, R. S., K. Thomas, W. A. Rourke, C. J. Murphy, P. McCarron, I. Burton, J. Walter, J. M. van de Riet & M. A. Quilliam. 2009. Detection and identification of a novel saxitoxin analogue in Argentinean scallops (Zygochlamyspatagonica). In: B. W. Halstead (1978), editor. Proceedings of the 7th international conference on Molluscan Shellfish Safety, Nantes, 2009. Poisonous and venomous marine animals of the world. Princeton, NJ: Darwin.

Gomez-Erache. M., D. Vizziano, P. Muniz & G. J. Nagy. 2001. The Health of the Rio De La Plata system. Northern Coast, Uruguay. In: Proceedings of the 4th Bay of Fundy Science Workshop, Saint John. New Brunswick, September 19-21. 2000. Opportunities and challenges for protecting, restoring and enhancing coastal habitats in the Bay of Fundy. Environment Canada: Atlantic Region, Occasional report no. 17, Dartmouth, NS and Sackville, NB.

Guy, A. L. & G. Griffin. 2009. Adopting alternatives for the regulatory monitoring of shellfish for paralytic shellfish poisoning in Canada: interface between federal regulators, science and ethics. Regul. Toxicol. Pharmacol. 54:256-263.

Hallegraeff, G. M. 2003. Harmful algal blooms: a global overview. In: G. M. Hallegraef, D. M. Anderson & A. D Cembella, editors. Manual on harmful marine microalgae. UNESCO, pp. 22-49.

Ichimi, K., T. Suzuki & M. Yamasaki. 2001. Non-selective retention of PSP toxins by the mussel Mytilus galloprovincialis fed with toxic dinoflagellate Alexandrium tamarense. Toxicon 39:1917-1921.

Indrasena, W. M. & T. A. Gill. 2000. Storage stability of paralytic shellfish poisoning toxins. Food Chem. 71:71-77.

Jaime, E., G. Gerdts & B. Luckas. 2007. In vitro transformation of PSP toxins by different shellfish tissues. Harmful Algae 6:308-316.

Lagos. N. 2003. Paralytic shellfish poisoning phycotoxins: occurrence in South America. Comments Toxicol. 9:175-193.

Lawrence, J. F.. B. Niedzwiadek & C. Menard. 2005. Quantitative determination of paralytic shellfish poisoning toxins in shellfish using prechromatographic oxidation and liquid chromatography with fluorescence detection: collaborative study. J. AOAC Int. 88:1714-1732.

Lin, H., Y. Cho, H. Yashiro, T. Yamada & Y. Oshima. 2004. Purification and characterization of paralytic shellfish toxin transforming enzyme from Macta chinensis. Toxicon 44:657-668.

Llewellyn, L. E. 2006. Saxitoxin, a toxic marine natural product that targets a multitude of receptors. Nat. Prod. Rep. 23:200-203.

McCulloch, A. W., R. K. Boyd, A. S. W. de Freitas, R. A. Foxall, W. D. Jamieson, M. V. Laycock, M. A. Quilliam, J. L. C. Wright, V. J. Boyko, J. W. McLaren & M. R. Miedema. 1989. Zinc from oyster tissue as causative factor in mouse deaths in official bioassay for paralytic shellfish poison. J. AOAC Int. 72:384-386.

McFarren, E. F. 1959. Report on collaborative studies of the bioassay for paralytic shellfish poison. J. of AOAC 42:263-271.

Medina, D., G. Inocente & C. Lopez. 1993. PSP in bivalve molluscs along the Uruguayan coast. In: T. J. Smayda & Y. Shimizu, editors. Toxic phytoplankton blooms in the sea. Proceedings of the Fifth International Conference on Toxic Marine Phytoplankton. 1991. Newport, RI: Elsevier, pp. 425-428.

Mendez, S., D. N. Severov, G. Ferrari & C. Mesones. 1996. Early Spring Alexandrium tamarense toxic blooms in Uruguay waters. In: Y. T. Oshima & Y. Fukuyo. editors. Harmful and toxic blooms. Integov.

Oceanog. Commis. Of UNESCO, pp. 113-117. Mendez, S. M. 1995. Atlas concentraciones de quistes del dinoflagelado toxico Gymnodinium catenatum en los sedimentos costeros de Uruguay. CC&I. 1:3.

Mendez, S. & G. Ferrari. 2003. Floraciones toxicas de Gymnodinium catenatum en aguas uruguayas. Publ. Com. Tec. Mix. Frente Maritimo 19:97-102.

Mendez, S. M. & D. Medina. 2004. Twenty-three years of red tide monitoring at fixed stations along the coast of Uruguay. In: K. A. Steidinger, J. H. Landsberg, C. R. Tomas & G. A. Vargo, editors. Harmful algae 2002. Florida Fish and Wildlife Conservation Commission, Florida Institute of Oceanography and Intergovernmental Oceanographic Commission of UNESCO, pp. 341-343.

Mendez, S. M., D. M. Kulis & D. M. Anderson. 2002. PSP Toxin production of Uruguayan isolates of Gymnodinium catenatum and Alexandrium tamarense. In: G. M. Hallegraeff, S. Blackburn, C. J. Bolch & R. Lewis, editors. Harmful algal blooms. Paris, France: UNESCO, pp. 352-355.

Nagashima, Y., T. Noguchi, T. Kawabat & K. Hashimoto. 1991. Dosedeath time curve of paralytic shellfish poisons in ddY strain mice. Nippon Suisan Gakkaishi 57:699-704.

Negri, A., D. Stirling, M. Quilliam. S. Blackburn, C. Bolch, I. Burton, G. Eaglesham, K. Thomas, J. Walter & R. And Willis. 2003. Three novel hydroxybenzoate saxitoxin analogues isolated from the dinoflagellates Gymnodinium catenatum. Chem. Res. Toxicol. 16:1029-1033.

Negri, A. P., C. J. S. Bolch, S. Geier, D. H. Green, T. G. Park & S. I. Blackburn. 2007. Widespread presence of hydrophobic paralytic shellfish toxins in Gymnodinium catenatum. Harmful Algae 6:774-780.

Oshima, Y., K. Sugino, H. Itakura, M. Hirota & T. Yasumoto. 1990. Comparative studies on paralytic shellfish toxin profile of dinoflagellates and bivalves. In: E. Graneli, B. Sundstrom, L. Elder & D. M. Anderson, editors. Toxic marine phytoplankton. New York, NY: Elsevier. 391-396.

Oshima, Y. 1995. Post-column derivatisation liquid chromatography method for paralytic shellfish toxins. J. AOAC Int. 78:528-532.

Park, D. L., W. N. Adams, S. L. Graham & R. C. Jackson. 1986. Variability of the mouse bioassay for determination of paralytic shellfish poisoning toxins. J. AOAC Int. 69:547-550.

Prakash, A., J. C. Medcof & A. D. Tennant. 1971. Paralytic shellfish poisoning in eastern Canada. Information Canada, Ottawa. Bull. Fish. Res. Board Can. 177:1-87.

Sato, S., R. Sakai & M. Kodama. 2000. Identification of thioether intermediates in the reductive transformation of gonyautoxins into saxitoxins by thiols. Bioorg. Med. Chem. Lett. 10:1787-1789.

Sakamoto, S., S. Sato, T. Ogata & M. Kodama. 2000. Formation of intermediate conjugates in the reductive transformation of gonyautoxins to saxitoxins by thiol compounds. Fish. Sci. 66:136-141.

Samsur, M., T. Takatani. Y. Yamagughi, T. Sagara, T. Noguchi & O. And Arakawa. 2007. Accumulation and elimination profiles of paralytic shellfish poison in the short-necked clam Tapes japonica fed with the toxic dinoflagellates Gymnodinium catenatum. Shokuhin Eiseigaku Zasshi 48:13-18.

Sekiguchi, K., S. Sato. S. Kaga, T. Ogata & M. Kodama. 2001. Accumulation of paralytic shellfish poisoning toxins in bivalves and an ascidian fed on Alexandrium tamarense cells. Fish. Sci. 67:301-305.

Shantz. E. J. 1960. Biochemical studies on paralytic shellfish poison. Ann. N. Y. Acad. Sci. 90:843-855.

Shimizu, Y., M. Alam, Y. Oshima & W. E. Fallon. 1975. Presence of four toxins in red tide infested clams and cultured Gonyaulax tamarensis cells. Biochem. Biophys. Res. Commun. 66:731-737.

Shumway, S. E. 1990. A review of the effects of algal blooms on shellfish and aquaculture. J. World Aquacult. Soe. 21:65-104.

Shumway, S. E. 1995. Phycotoxin-related shellfish poisoning: bivalve molluscs are not the only vectors. Rev. Fish. Sci. 3:1-31.

Smith, E. A.. F. Grant, C. M. J. Ferguson & S. Gallacher. 2001. Biotransformation of paralytic shellfish toxins by bacteria isolated from bivalve molluscs. Appl. Environ. Microbiol. 67:2345-2353.

Stephenson, N. R., H. I. Edwards, B. F. MacDonald & L. I. Pugsley. 1955. Biological assay of the toxin from shellfish. Can. J. Biochem. Physiol. 33:849-857.

Sullivan, J.J., W.T. Iwaoka & J. Liston. 1983. Enzymatic transformation of PSP toxins in the littleneck clam. Biochem. Biophys. Res. Comm. 114:465-472.

Turner. A. D.. D. M. Norton. R. G. Hatfield, S. Morris. A. R. Reese, M. Algoet & D. N. Lees. 2009. Single laboratory validation of the AOAC HPLC method (2005.06) for mussels: refinement and extension of the method to additional toxins. J. AOAC Int. 92:190-207.

Turner, A. D., D. M. Norton. R. G. Hatfield, M. Rapkova-Dhanji, M. Algoet & D. N. Lees. 2010. Single laboratory validation of a refined AOAC LC method for oysters, cockles and clams in UK shellfish. J. AOAC Int. 93:1482-1493.

Turner. A. D. & R. G. Hatfield. 2012. Refinement of AOAC Official Method 2005.06 liquid chromatography-fluorescence detection method to improve performance characteristics for the determination of paralytic shellfish toxins in king and queen scallops. J. A OA C International. 95:129-142.

Turner, A. D., R. G. Hatfield, M. Rapkova, W. Higman, M. Aloet, B. A. Suarez-Isla, M. Cordova, C. Caceres, J. van de Riet, R. Gibbs, K. Thomas, M. Quilliam & D. N. Lees. 2011. Comparison of AOAC 2005.06 LC official method with other methodologies for the quantitation of paralytic shellfish poisoning toxins in UK shellfish species. Anal. Bioanal. Chem. 399:1257-1270.

Turner, A. D., M. Dhanji-Rapkova, M. Algoet, B. A. Suarez-Isla. M. Cordova, C. Caceres, J. van de Riet, C. J. Murphy, M. Case & D. N. Lees. 2012. Investigations into matrix components affecting the performance of the official bioassay reference method for quantitation of paralytic shellfish poisoning toxins in oysters. Toxicon 59:215-230.

Turner, A. D.. A. M. Lewis, A. O'Neil & R. G. Hatfield. 2013a. Transformation of paralytic shellfish poisoning toxins in UK surf clams (Spisula solida) for targeted production of reference materials. Toxicon 65:41-58.

Turner, A. D.. B. Stubbs, L. Coates. R. G. Hatfield, M. Dhanji-Rapkova, S. Rowland-Pilgrim, A. O'Neil, P. Stubbs, A. M. Lewis, C. Baker. S. Ross & M. Algoet. 2013b. The assessment of paralytic shellfish poisoning occurrence and toxin profiles in bivalve molluscs from Great Britain during the first five years of official control monitoring as determined by liquid chromatography. Harmful Algae. Submitted for publication.

Vale, P. 2008a. Complex profile of hydrophobic paralytic shellfish poisoning compounds in Gymnodinium catenatum detected by liquid chromatography with fluorescence and mass spectrometry detection. J Chrom. A. 1195:85-93.

Vale, P. 2008b. Fate of benzoate paralytic shellfish poisoning toxins from Gymnodinium catenatum in shellfish and fish detected by pre-column oxidation and liquid chromatography with fluorescence detection. J Chrom. A. 1190:191-197.

Vale, P. & M. A. de Sampayo. 2001. Determination of paralytic shellfish toxins in Portuguese shellfish by automated pre-column oxidation. Toxicon 39:561-571.

Van de Riet, J., R. S. Gibbs. P. M. Muggah. W. A. Rourke. J. D. MacNeil & M. A. Quilliam. 2011. Liquid chromatography post-column oxidation (PCOX) method for the determination of paralytic shellfish toxins in mussels, clams, oysters and scallops: collaborative study. J. AOAC Int. 94:1154-1176.

Wiberg, G. S. & N. R. Stephenson. 1960. Toxicologic studies on paralytic shellfish poison. Toxicol. Appl. Pharmacol. 2:607-615.

Wiberg, G. S. & N. R. Stephenson. 1961. The effect of metal ions on the toxicity of paralytic shellfish poison. Toxicol Appl Pharmacol. 3:707-712.

Wiese, M., P. M. D"Agnostino, T. K. Mihali, M. C. Moffitt & B. A. Neilan. 2010. Neorotoxic alkaloids: saxitoxins and its analogs. Mar. Drugs 8:2185-2211.

Wright, J. L. C. 1995. Dealing with seafood toxins: present approaches and future options. Food Res. Int. 28:347-358.

ANDREW D. TURNER, (1) * SOPHIE TARNOVIUS, (1,2) DINORAH MEDINA (3) AND MARIA SALHI (3)

(1) Centre for Environment, Fisheries and Aquaculture Science (Cefas), Barrack Road, The Nothe, Weymouth, Dorset, DT4 8UB, United Kingdom; (2) Technische Universitat Munchen, Walther-Meissner-Strasse 3, 85748 Garching, German; (3) Direction National de Recursos Acuaticos, P.O. Box 1612, Constituyente 1497, 11200 Montevideo, Uruguay

* Corresponding author. E-mail: andrew.turner@cefas.co.uk

DOI: 10.2983/035.034.0338

TABLE 1.
Summary of samples, sampling areas, harvesting dates, source of
phytoplankton associated (Mendez & Medina 2004, plus unpublished
data) and PSP results by MBA and LC/FLD ([micro]g STX eq./kg).

Sample             Harvesting                  Harvesting
number   Species   area                           date

1        Mussels   Punta del Diablo, Rocha      20/03/98
2        Mussels   Piriapolis, Maldonado        09/10/91
3        Mussels   Punta del Diablo. Rocha      20/09/91
4        Mussels   Jose Ignacio, Maldonado      04/09/91
5        Mussels   Jose Ignacio, Maldonado      29/08/96
6        Mussels   La Paloma, Rocha             04/03/92
7        Mussels   Punta dei Este, Maldonado    17/08/93
8        Mussels   Punta del Diablo. Rocha      09/10/91
9        Clams     La Paloma, Rocha             29/04/10
10       Mussels   Punta dei Este, Maldonado    02/2003
11       Clams     La Paloma, Rocha             29/03/93
12       Mussels   Punta dei Este, Maldonado    24/03/93
13       Clams     La Paloma, Rocha             24/02/92
14       Mussels   Piriapolis, Maldonado        20/09/91
15       Mussels   Punta del Diablo, Rocha      28/08/96
16       Mussels   La Paloma, Rocha             25/09/91
17       Mussels   Punta dei Este. Maldonado    15/10/91
18       Clams     La Paloma. Rocha             26/10/11
19       Clams     La Paloma, Rocha             30/03/12
20       Mussels   Piriapolis, Maldonado        30/03/12

Sample   Source                                    LC:MBA
number   phytoplankton            MBA     LC-FLD   ratio

1        Gymnodinium catenatum    1,670      635    0.38
2        Alexandrium tamarense     <800      371     --
3        A. tamarense             1,160    2,087    1.80
4        A. tamarense               830    1,680    2.02
5        A. tamarense              na      4,439     --
6        G. catenatum              <800      327     --
7        A. tamarense             1,960      454    0.23
8        A. tamarense              na      1,022     --
9        G. catenatum             4,000      193    0.05
10       G. catenatum             1,970      338    0.17
11       G. catenatum              <800      209     --
12       G. catenatum             2,900    1,991    0.69
13       G. catenatum             2,470    1,031    0.42
14       A. tamarense             3,090    7,281    2.36
15       A. tamarense            17,830   25,581    1.43
16       A. tamarense            10,300    2,004    0.19
17       A. tamarense            11,500      560    0.05
18       A. tamarense               630      520    0.83
19       None                      <LOD      10      --
20       None                      <LOD      10      --

Mussels, Mylilus edulis; clams, Donax hanleyanus; nd, not determined;
na, not applicable; LOD. limit of detection.
COPYRIGHT 2015 National Shellfisheries Association, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Turner, Andrew D.; Tarnovius, Sophie; Medina, Dinorah; Salhi, Maria
Publication:Journal of Shellfish Research
Article Type:Report
Geographic Code:3URUG
Date:Dec 1, 2015
Words:9871
Previous Article:Spatial and temporal variability of spawning in the green sea urchin Strongylocentrotus droebachiensis along the coast of Maine.
Next Article:Dinophysis species associated with diarrhetic shellfish poisoning episodes in North Patagonian gulfs (Chubut, Argentina).
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters