First detection of azaspiracids in mussels in North West Africa.

ABSTRACT Outbreaks of lipophilic toxins have been recorded in the north Atlantic coast of Morocco since 1999, but are rare in the Mediterranean coast. Samples of mussels from the Atlantic coast where toxicity was detected by mouse bioassay were stored for further research. Chemical analysis by LC-MS conducted in mussels harvested from this region showed, in addition to the presence of okadaic acid (OA) and dinophysistoxin-2 (DTX2), the presence of azaspiracid-2 (AZA2) as the dominant form of the azaspiracid's (AZAs) family, followed by AZA1 (13% to 26%). AZA3 was rarely detected, and maximal concentrations found were between 3% and 8% of total AZA1/3. The presence of AZA2 and AZA1 was confirmed by mass spectra. Time series corresponding to the summer of 2004 and 2005 showed maximal concentration of AZAs appeared in July in both years. Correlation with occurrence of OA and DTX2, showed both toxin families could appear simultaneously in Moroccan mussels, but maximal concentrations found were always separated in time. This is the first report of azaspiracids in Morocco (NW Africa) and the first report outside of European coastlines.

KEY WORDS: azaspiracids, AZP, Morocco, LC-MS, DSP, mussels, Mytilus galloprovinciallis

INTRODUCTION

Azaspiracid poisoning (AZP) is a new human gastrointestinal illness discovered in 1995 after consumption of contaminated Irish mussels in the Netherlands (McMahon & Silke 1996). Azaspiracid-1 (AZA1) was the first molecule of the group to be structurally characterized (Satake et al. 1998). Now, at least more than 10 azaspiracid congeners are known (James et al. 2003b). From these, AZA1, AZA2, and AZA3 were the dominant compounds found in shellfish, followed by AZA4 and AZA5. The remaining ones (AZA6-11) were minor components. It is hypothesized so far that these lipophilic toxins accumulate in bivalve molluscs after feeding on the toxic microalgae Protoperidinium crassipes, previously considered to be toxicologically harmless (James et al. 2003a).

The symptoms of acute AZP intoxication include nausea, vomiting, severe diarrhea, and stomach cramps (Satake et al. 1998), which closely resemble the symptoms associated with diarrheic shellfish poisoning (DSP). Toxicological studies showed that repeated administration of AZAs caused prolonged damage in the intestine and induced lung tumors in mice (Ito et al. 2002).

Although shellfish harvested in Ireland provoked the first incidents of human intoxications by azaspiracids, the research in other European countries has led to the discovery of these toxins in shellfish from the United Kingdom, Norway, France, and Spain (James et al. 2002a, Magdalena et al. 2003). Although AZP is a serious concern in Europe, its presence has never been reported outside European waters.

Outbreaks of lipophilic toxins have been recorded in shellfish from the north Atlantic coast of Morocco since 1999, but these toxins are rare in the Mediterranean coast (Taleb 2005). The Moroccan monitoring program to screen for the presence of lipophilic compounds employs the mouse bioassay (MBA), which does not allow distinguishing the contamination from different phycotoxin families, such as DSP or AZP, because these compounds are detected altogether. Although several methodologies exist to study the contamination with DSP, the only methodology available so far to confirm the specific presence of azaspiracids is liquid chromatography coupled to mass spectrometry (LC-MS) (Ofuji et al. 1999). Samples of blue mussels collected in Morocco during the summers of 2004 and 2005 were screened by LC-MS for lipophilic compounds. Besides the presence of DSP toxins (to be reported elsewhere), the presence of azaspiracids was put into evidence and is now reported here for the first time.

MATERIALS AND METHODS

Study Area

Samples of mussels (Mytilus galloprovincialis) were collected regularly from Oulad Ghanem and Dar Hamra at the Atlantic coast of Morocco (south of Casablanca, Fig. 1). Digestive glands (DG) were dissected from whole shellfish, homogenized, and extracted for MBA testing. The samples used for the study were selected from those collected in the period of summer outbreaks of samples positive to lipophilic toxins. Digestive glands of positive samples were stored frozen for ulterior analysis by LC-MS.

[FIGURE 1 OMITTED]

Sample Preparation and LC-MS Analysis

Digestive glands were homogenized, and a 5-g aliquot was taken and extracted by ultraturrax with 20 mL aqueous 90% methanol in a screw-cap plastic centrifuge tube, followed by a 10 min centrifugation at 2500g. A 2-milliliter aliquot of the methanolic supernatant was transferred to 10-mL glass test tubes, washed with hexane (2 x 2 mL), and extracted with dichloromethane (2 x 2 mL). The combined dichloromethane layers were dried with anhydrous sodium sulphate, centrifuged, transferred to small glass test tubes, and dried at 38[degrees]C under reduced pressure on a RapidVap (Labconco, USA). After resuspension with 500-[micro]L aqueous 90% methanol, contents were transferred to a 2 mL autosampler vial. Vortex mixing was used in all liquid-liquid and resuspension steps.

Analyses were performed on a LC-MS system from Hewlett-Packard 1100-Series, consisting of an in-line degasser, a quaternary pump, an autosampler, a column oven, and the 1946A single-quadrupole mass detector. High-purity nitrogen was obtained through an [N.sub.2]-Generator (Dominick-Hunter, Durham, England).

A 1.0-[micro]L aliquot was separated for 12 min on a Merck Lichrospher-100 RP-18 column (5 [micro]m, 125 x 2 mm) protected by a guard colunm packed with the same material (4 x 4 mm) using a mobile phase of acetonitrile-water (86:14, v/v), supplemented with 0.05% trifluoracetic acid, and 50 mM formic acid pumped isocratically at 275 [micro]L/min. The eluent flow was diverted to waste for 2.0 min after sample injection, and MS detection was carried out from 2.0 min forward. Single ion monitoring (SIM) was used to record the signals from the [[M + H].sup.+] ions at: m/z 828.5 (AZA3), 842.5 (AZA1), 844.5 (AZA4, AZA5), and 856.5 (AZA2). Mass spectra were obtained scanning in the range m/z 600-900. Dr. Phillip Hess and Dr. Terry McMahon, Marine Institute, Ireland, supplied a pure standard of AZA1 and mussels contaminated with AZA1-5.

RESULTS

Analysis carded out in SIM mode for azaspiracids in mussel DG collected in the Atlantic coast of Morocco showed the presence in some samples of azaspiracid-2 (AZA2), followed by a minor contribution of AZA1. Only in samples where the concentration of AZA2 was highest, it was possible to detect also AZA3 (Fig. 2a). Analysis in SCAN mode of the most contaminated samples allowed the confirmation of AZA1 and AZA2 by mass spectra (Fig. 2b, c, respectively). AZA3 was not found in concentrations enough to allow confirmation by mass spectra. Although the presence of AZA4/5 was screened, no evidence of their presence was registered.

[FIGURE 2 OMITTED]

The temporal evolution of AZAs was studied in mussels collected in Oulad Ghanem in 2004 and 2005 (Fig. 1). In 2004, concentration of AZA2 and AZA1 was maximal in the beginning of July, then declined, and in autumn increased again (Fig. 3a). In 2005, the major peak of AZAs occurred again in mid July, and throughout the summer it remained fairly constant from 0.20 [micro]g/g down to 0.10 [micro]g/g (Fig. 3b). A similar evolution of toxic profiles was obtained with mussel samples collected in Dar Hamra in 2005 (data not shown). AZA2 was in all samples tested the dominant toxin (Fig. 3c, d). When detected, levels of AZA1 fluctuated between 13% and 26% (Figs. 3c, d). Only in samples from 2005 AZA3 was detected, and maximal concentrations found were between 3% and 8% (Fig. 3d).

[FIGURE 3 OMITTED]

The evolution of the summed concentrations of AZAs in mussels collected in Oulad Ghanem in 2004 and 2005 was compared with the concentration of DSP toxins found in the same samples, corresponding to the sum of hydrolyzed OA and DTX2 (data to be reported elsewhere). From the data presented in Figure 4, one observes that maximal concentrations of AZP toxins appear after the decline of DSP peaks (mid September 2004 and mid July 2005). Because data for the early summer of 2004 is incomplete, it is not possible to conclude if the AZA's peak from mid July follows a declining DSP peak, or simply precedes a new DSP raising peak.

[FIGURE 4 OMITTED]

DISCUSSION

This study reports for the first time the presence of azaspiracids in the NW African coast, and is the first time AZAs are reported in shellfish from such low latitude (32[degrees]50'904N, 8[degrees]53'500W) and outside Europe. The presence of AZA2 in shellfish from the south coast of Portugal was previously suspected, but the low levels found of the suspect compounds did not allow their unambiguous confirmation, mainly because that study used whole flesh (Vale 2004). According to Hess et al. (2005), these toxins are on average circa 5 times more concentrated in DG, compared with the whole mussel.

The simultaneous presence of DSP and AZP toxins may occur in Moroccan mussels, but maximal concentrations of these two distinct toxin families are separated in time. Evidence from Irish mussels points also that contamination with AZP may follow contamination with DSP (Clarke et al. 2006), but appearance of AZP before the rise in DSP toxins may also occur (Hess et al. 2002). Because the species hypothesized to produce AZAs is known to be heterotrophic, its growth might depend on the appearance of conditions that favor the blooming of certain algae species necessary for predation, amongst which also proliferate the DSP producers.

In the Atlantic Moroccan coast, the occurrence of AZP seems to be recurrent, and maximal levels may take place over the summer, although other seasons of the year were not studied yet for the presence of AZP. After the launching of the lipophilic toxin monitoring along the Moroccan coast by mouse bioassay (MBA), the recurrence of lipophilic toxin outbreaks in the Atlantic coast was recorded during the summer time almost every year since 1999, but in the Mediterranean coast occurrence of this type of toxins is rare (Taleb, 2005). For the moment, at least two families of lipophilic compounds are implied in the observed mouse bioassay toxicity. So far, these are the most relevant families of toxins for public health, as other families, such as yessotoxins and pectenotoxins, do not seem to be relevant for human health because of their low oral potency (Anonymous 2004).

The toxin profiles found in European shellfish (hepatopancreas) and plankton samples have been dominated by AZA1 (James et al. 2002a, James et al. 2003a, Magdalena et al. 2003). However, AZA3 has been reported to be the most abundant toxin in meat samples with the hepatopancreas excised (James et al. 2002b). Although variations in toxin profile have been observed, the results reported here of AZA2 dominance might point to a difference in the phytoplankton-producing species. Indeed, the finding of AZAs at such low latitudes may raise the question if more than one species is involved in AZA production and also concerns of a more universal distribution of AZAs than previously believed.

LITERATURE CITED

Anonymous. 2004. Report of the joint FAO/IOC/WHO ad hoc expert consultation on biotoxins in bivalve molluscs. Oslo, Norway, Sept. 26-30, 2004.

Clarke D., L. Devilly, T. McMahon, M. O'Cinneide, J. Silke, S. Burrell, O. Fitzgerald, P. Hess, J. Kilcoyne, M. McElhinney, J. Ronan, R. Gallardo Salas, B. Gibbons, M. Keogh, M. McCarron, S. O'Callaghan & B. Rourke. 2006. A review of shellfish toxicity monitoring in Ireland and review of management cell decisions for 2005. In: Proceedings of the 6th Irish Shellfish Safety Scientific Workshop. Marine Institute, Marine Environment & Health Series, n[degrees]23. pp. 22-34.

Hess, P., T. McMahon, D. Slattery, D. Swords, G. Dowling, M. McCarron & D. Clarke. 2002. Biotoxin chemical monitoring in Ireland-2001. In: Proceedings of the 2nd Irish marine biotoxin science workshop, Galway, 11th October 2001, pp. 8-18.

Hess, P., L. Nguyen, J. Aasen, M. Keogh, J. Kilcoyne, P. McCarron & T. Aune. 2005. Tissue distribution, effects of cooking and parameters affecting the extraction of azaspiracids from mussels, Mytilus edulis, prior to analysis by liquid chromatography coupled to mass spectrometry. Toxicon 46:62-71.

Ito, E., M. Satake, K. Ofuji, M. Higashi, K. Harigaya, T. McMahon & T. Yasumoto. 2002. Chronic effects in mice caused by oral administration of sublethal doses of azaspiracid, a new marine toxin isolated from mussels. Toxicon 40:193-203.

James, K. J., A. Furey, M. Lehane, H. Ramstad, T. Aune, P. Hovgaard, S. Morris, W. Higman, M. Satake & T. Yasumoto. 2002a. First evidence of an extensive northern European distribution of azaspiracid poisoning (AZP) toxins in shellfish. Toxicon 40:909-915.

James, K. J., M. Lehane, C. Moroney, P. Fernandez-Puente, M. Satake, T. Yasumoto & A. Furey. 2002b. Azaspiracid shellfish poisoning: un usual toxin dynamics in shellfish and the increased risk of acute human intoxications. Food Addit. Contain. 19:555-561.

James, K. J., C. Moroney, C. Roden, M. Satake, T. Yasumoto, M. Lehane & A. Furey. 2003a. Ubiquitous "benign" alga emerges as the cause of shellfish contamination responsible for the human toxic syndrome, azaspiracid poisoning. Toxicon 41:145-151.

James, K. J., M. D. Sierra, M. Lehane, A. B. Magdalena & A. Furey. 2003b. Detection of five new hydroxyl analogues of azaspiracids in shellfish using multiple tandem mass spectrometry. Toxicon 41:277-283.

Magdalena, A. B., M. Lehane, S. Krys, M. L. Fernandez, A. Furey & K. J. James. 2003. The first identification of azaspiracids in shellfish from France and Spain. Toxicon 42:105-108.

McMahon, T. & J. Silke. 1996. Winter toxicity of unknown aetiology in mussels. Harmful Algae News 14:2.

Ofuji, K., M. Satake, Y. Oshima, T. McMahon, K. J. James & T. Yasumoto. 1999. A sensitive and specific method for azaspiracids by liquid chromatography mass spectrometry. Nat. Toxins 7:247-250.

Satake, M., K. Ofuji, I. Naoki, K. J. James, A. Furey, T. McMahon, J. Silke & T. Yasumoto. 1998. Azaspiracid, a new marine toxin having unique spirt ring assemblies, isolated from Irish mussels, Mytilus edulis. J. Am. Chem. Soc. 120:9967-9968.

Taleb, H. 2005. Phycotoxines paralysantes (PSP) et diarrhEiques (DSP) le long des cotes marocaines: Evolution spatio-temporelle, profil toxinique et cinetique de decontamination. These de Doctat d'Etat (PhD), Universite Hassan II, Faculte des Sciences Ain Chok, Casablanca, Morocco, pp. 159.

Vale, P. 2004. Is there a risk of human poisoning by azaspiracids from shellfish harvested at the Portuguese coast? Toxicon 44:943-947.

H. TALEB, (1)* P. VALE, (2) R. AMANHIR, (1) A. BENHADOUCH, (1) R. SAGOU (1) AND A. CHAFIK (1)

(1) Institut National de Recherche Halieutique (INRH)--2, Rue de Tiznit, Casablanca, Morocco; (2) Instituto Nacional de Investigacao Agraria e das Pescas-IPIMAR, Av. Brasilia, 1449-006 Lisboa, Portugal

* Corresponding author. E-mail: [e-mail]