Toxicity and Pathogenicity Testing of the Insect Pest Control Fungus Metarhizium anisopliae

Renewed interest in the use of Metarhizium anisopliae and its toxins for insect control prompted the following safety assessment. A neutral extract (methylene chloride, pH 7.2), derived from M. anisopliae cultures, was evaluated for toxicity and mutagenicity using aquatic animal bioassays and the Ames test. The average LC50 of the neutral extract obtained in static, acute 96-h tests conducted with </=24-h-old Mysidopsis bahia was 2.41 mg L-1. By partially purifying destruxins from the neutral extract, it was shown that destruxins alone were not responsible for the observed toxicity in mysids. The neutral extract was fetotoxic to developing grass shrimp, Palaemonetes pugio, and frog, Xenopus laevis, embryos; the LC50 values were 52 and 32 mg L-1, respectively. Eye spot abnormalities were observed in shrimp and frog embryos exposed to the neutral extract. In extract-exposed frog embryos, moderate to severe cranial, facial, and gut malformations were also observed. The neutral extract was toxic to juvenile mosquito fish, Gambusia affinis, at an LC50 value of 141 mg L-1. Adult female G. affinis surviving a 24-h exposure to 200 µg ml-1 of the neutral extract produced healthy broods. After 3 months, no mortalities or adverse effects were observed in adult G. affinis fed a diet partially composed of a freeze-dried M. anisopliae culture. The neutral extract did not show mutagenicity in the Ames test using strains TA98 and TA100 with and without metabolic activation by rat liver S9. Significant (p </= 0.05) mortalities were obtained when embryos of grass shrimp and inland silverside fish, Menidia beryllina, were exposed to the same lot of M. anisopliae conidiospores. Exposure of frog embryos to M. anisopliae conidiospores did not cause significant (p > 0.05) mortalities or malformations.