Elsevier

Toxicon

Volume 45, Issue 7, 1 June 2005, Pages 891-900
Toxicon

Cytotoxic and cytoskeletal effects of azaspiracid-1 on mammalian cell lines

https://doi.org/10.1016/j.toxicon.2005.02.015Get rights and content

Abstract

Azaspiracid-1 (AZA-1) is a newly identified phycotoxin reported to accumulate in molluscs from several northern European countries and documented to have caused severe human intoxications. The mechanism of action of AZA-1 is unknown. Our initial investigations have shown that AZA-1 is cytotoxic to a range of cell types. Cytotoxicity was evident in all seven cell types tested, suggesting a broad-spectrum mode of action, and was both time- and concentration-dependent. However, AZA-1 took an unusually long time (>24 h) to cause complete cytotoxicity in most cell types, with the exception of the rat pituitary GH4C1. Extended exposure times did not always lower the EC50 value for a given cell line, but always resulted in more complete cytotoxicity over a very narrow concentration range. The Jurkat cell line (human lymphocyte T) appeared to be very sensitive to AZA-1, although the EC50 values (24–72 h) for all the cell types were in the low nanomolar range (0.9–16.8 nM). The effect of AZA-1 on membrane integrity was tested on Jurkat cells and these data confirm our visual observations of cytotoxicity and necrotic cell lysis following exposure of Jurkat cells to AZA-1 and suggest that AZA-1 has some properties unique among marine algal toxins. Additionally, there were dramatic effects of AZA-1 on the arrangement of F-actin with the concurrent loss of pseudopodia, cytoplasmic extensions that function in mobility and chemotaxis. Although these phycotoxin-specific effects of AZA-1 suggest a possible site of action, further work using cell-based approaches is needed to determine the precise mode of action of AZA-1.

Introduction

Azaspiracids (AZA) are nitrogen-containing polyether toxins with a unique spiral ring assembly, a cyclic amine, and a carboxylic acid, and were first detected in mussels (Mytilus edulis) in Ireland in 1995. This toxin has subsequently been detected in other bivalve species, including oysters (Crassostrea gigas, Ostrea edulis), scallops (Pecten maximus), clams (Tapes phillipinarium), cockles (Cardium edule), and razor fish (Ensis siliqua) (Hess et al., 2001, Furey et al., 2003, Hess et al., 2003). Cases of AZA intoxications and/or contaminated shellfish in several countries, including Ireland, UK, Norway, Netherlands, France, Spain, and Italy have been documented (Satake et al., 1998a, Ito et al., 2002, James et al., 2002a, Magdalena et al., 2003a). Although the human symptoms resembled those of diarrhetic shellfish poisoning (DSP), the illness was subsequently named azaspiracid poisoning (AZP) once azaspiracid was identified as a novel toxin (Ito et al., 1998, Ofuji et al., 1999). The regulatory limit for AZA in shellfish in the European Union has been set at 160 μg azaspiracid equivalents per kg (European Commission, 2002).

To date, 11 different congeners have been identified and named AZA-1 through AZA-11 (Ito et al., 1998, Satake et al., 1998b, James et al., 2003b). Consistent with their highly oxygenated polyether structure and the seasonal occurrence of shellfish toxicity, AZAs are thought to originate from a dinoflagellate source (Ofuji et al., 1999). Initially, AZA was identified in association with a heterophic dinoflagellate, Protoperidinium sp (Yasumoto, 2001) later reported as P. Crassipes (James et al., 2003a); however, definitive confirmation that this Protoperidinium species produces AZAs de novo will require demonstration of AZA synthesis in laboratory cultures grown on defined prey species.

Current studies are aimed at assessing the distribution and long-term dynamics of AZA accumulation in shellfish. There is some evidence that AZAs, unlike the DSP toxins, may accumulate not only in the hepatopancreas (digestive glands), but may also occur in the gonads (roe) (Magdalena et al., 2003b) and the adductor muscle (meat) (James et al., 2000, James et al., 2002b). Yet, given the extensive variation between individual animals and among shellfish species, no general trend has been established. The fact that AZAs can persist in M. edulis for up to 8 months following the initial exposure, possibly reflecting tissue migration and a slow depuration rate of the toxin (James et al., 2000), does indicate that protection of shellfish consumers will require vigilant monitoring.

Following human consumption of AZA-contaminated shellfish, there is generally a rapid onset of symptoms very similar to those of DSP, including nausea, vomiting, severe diarrhea, and stomach cramps (McMahon and Silke, 1996, McMahon and Silke, 1998). In mice and rats, intraperitoneal (i.p.) injections induce neurological, ‘PSP-like’ symptoms with progressive paralysis, fatigue, difficulty breathing, and subsequent death as soon as 35 min following injection (Ito et al., 1998, McMahon and Silke, 1998, Satake et al., 1998a). Pathological effects include histological alterations in the liver, pancreas, spleen, and necrotic lymphocytes in the thymus. More detailed in vivo studies with mice have since demonstrated that oral AZA-1 exposure induces a variety of changes in the intestinal tract, including accumulation of fluid, necrosis and edema in the lamina propria of the mid-intestinal tract, and fused, shortened villi. Necrosis of T and B lymphocytes were also documented in the spleen and thymus, as well as fatty changes in the liver, hyperplasia of the epithelial lining in the stomach, and tumors in the lungs (Ito et al., 1998, Ito et al., 2000, Ito et al., 2002).

Recent efforts to determine the mode of action of AZA have been conducted by several investigators using in vitro techniques. Due to the similar nature of the symptoms of AZP with DSP, a mode of action similar to the DSP toxins on protein phosphatase (PP) activity was proposed by Flanagan et al. (2001). However, unlike DSP toxins, bioactive extracts containing AZAs did not have any inhibitory effect on PP1 activity, yet were clearly cytotoxic to both human hepatoblastoma and bladder carcinoma cell lines (Flanagan et al., 2001). This suggests that the AZA-1 mode of action differs from that of the DSP toxins, although human symptomatology of these two syndromes is similar. Using a combination of neuroblastoma cells and human lymphocytes, it was shown that AZA-1 reduces cellular F-actin content in a non-apoptotic manner following the elevation of cytosolic calcium and cAMP levels (Roman et al., 2002). The authors suggested that AZA-1 might be targeting the cytoskeleton. In contrast to AZA-1, the congeners AZA-2 and AZA-3 have different effects on the levels of cytosolic calcium, possibly suggesting a unique mode of action for each toxin analogue (Botana, 2002).

Herein we report on our investigation of the AZA-1 mode of action using cytotoxicity assays and morphological analyses. Following initial experiments examining the effects of AZA-1 on protein phosphatase 2A, we used a combination of in vitro cell-based approaches to screen a panel of seven mammalian cell lines for cytotoxicity to assess cell-specific sensitivities to AZA-1. Detailed experiments were then performed on an AZA-1-sensitive, Jurkat lymphocyte T cell line in order to characterize AZA-induced effects on cellular morphology and cytoskeletal arrangements. We anticipate that our findings will lead to the development of new methods for AZA detection and aid in defining the AZA-1 mode of action, which will, in turn, facilitate monitoring of the ecological and human health effects associated with the AZAs. Ultimately, this information may also be of use in the development of treatments for humans exposed to AZAs.

Section snippets

Toxins and chemicals

Azaspiracid (AZA-1) was extracted from 2 kg of mussels (M. edulis) that were collected in 1996 from Killary Harbour, on the west coast of Ireland and in 1999 from Bantry Bay, on the southwest coast of Ireland. Toxins were extracted in 2001, as described by Satake et al., 1998b, Ofuji et al., 1999 with slight modifications. Stock AZA-1 (2.4 mg) was determined to be >93% pure by NMR and showed <1% impurity of other AZA subtypes/congeners by liquid chromatography–mass spectrometry (LC–MS).

Effect of AZA-1 on protein phosphatase 2A activity

OA inhibited PP2A activity in a concentration-dependent manner (EC50=1.5 nM), whereas it is clear that AZA-1 did not have any effect on PP2A activity at all concentrations tested (Fig. 1).

Effect of AZA-1 on cell line viability

Screening experiments were conducted on seven mammalian cell lines (Table 1) to determine the cell type-specificity and sensitivity of AZA-1-induced cytotoxicity. After 24 h of exposure, AZA-1 concentrations of ≤1 nM did not affect viability for any of the seven mammalian cell lines tested. However, all cell

Discussion

Since AZA was first detected during a 1995 outbreak in Ireland, a considerable amount of work has been carried out on the structural characterization of AZA-1 and its analogs, the development of LC–MS based analytical methods for AZA detection and quantification, as well as ecological studies related to its spatial and temporal distribution. However, no studies have investigated the toxicological effects of AZA on aquatic organisms, no AZA specific bioassays have been developed, and only a few

Acknowledgements

This work was funded by the National Oceanic and Atmospheric Administration/National Ocean Service (NOAA/NOS) and in part by work supported by the National Science Foundation under REU Grant No. DBI-0244007 to L.E. and K.G. Burnett. Funding for the Marine Institute, Ireland, and collaborative efforts was also obtained from the Irish National Development Plan (NDP) under the Marine Research Strategic Project ST-02-02, Azaspiracids Standards and Toxicology (ASTOX). MJT was supported by a National

References (32)

  • A. Tubaro et al.

    Suitability of the MTT-based cytotoxicity assay to detect okadaic acid contamination of mussels

    Toxicon

    (1996)
  • M.R. Vieytes et al.

    A fluorescent microplate assay for diarrheic shellfish toxins

    Analytical Biochemistry

    (1997)
  • B. Wehrle-Haller et al.

    Actin, microtubules and focal adhesion dynamics during cell migration

    The International Journal of Biochemistry and Cell Biology

    (2003)
  • Botana, L.M., 2002. Receptor candidates to the new toxins. 10th International Conference on Harmful Algae, pp....
  • Colman, J.R., Twiner, M.J., Hess, P., McMahon, T., Yasumoto, T., Satake, M., Doucette, G.J., Ramsdell, J.S., in press....
  • European Commission, 2002. Commission Decision 2002/225/EC laying down detailed rules for the implementation of Council...
  • Cited by (106)

    View all citing articles on Scopus
    View full text