In vitro toxicology of respirable Montserrat volcanic ash.

OBJECTIVES In July 1995 the Soufriere Hills volcano on the island of Montserrat began to erupt. Preliminary reports showed that the ash contained a substantial respirable component and a large percentage of the toxic silica polymorph, cristobalite. In this study the cytotoxicity of three respirable Montserrat volcanic ash (MVA) samples was investigated: M1 from a single explosive event, M2 accumulated ash predominantly derived from pyroclastic flows, and M3 from a single pyroclastic flow. These were compared with the relatively inert dust TiO(2) and the known toxic quartz dust, DQ12. METHODS Surface area of the particles was measured with the Brunauer, Emmet, and Teller (BET) adsorption method and cristobalite content of MVA was determined by x ray diffraction (XRD). After exposure to particles, the metabolic competence of the epithelial cell line A549 was assessed to determine cytotoxic effects. The ability of the particles to induce sheep blood erythrocyte haemolysis was used to assess surface reactivity. RESULTS Treatment with either MVA, quartz, or titanium dioxide decreased A549 epithelial cell metabolic competence as measured by ability to reduce 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT). On addition of mannitol, the cytotoxic effect was significantly less with M1, quartz, and TiO(2). All MVA samples induced a dose dependent increase in haemolysis, which, although less than the haemolysis induced by quartz, was significantly greater than that induced by TiO(2). Addition of mannitol and superoxide dismutase (SOD) significantly reduced the haemolytic activity only of M1, but not M2 or M3, the samples derived from predominantly pyroclastic flow events. CONCLUSIONS Neither the cristobalite content nor the surface area of the MVA samples correlated with observed in vitro reactivity. A role for reactive oxygen species could only be shown in the cytotoxicity of M1, which was the only sample derived from a purely explosive event. These results suggest that in general the bioreactivity of MVA samples in vitro is low compared with pure quartz, but that the bioreactivity and mechanisms of biological interaction may vary according to the ash source.