Toxicon. Vol. 33, No. 5, pp. 699-702. 1995 Copyright 0 1995 Elscvin Bcience Ltd Printed in Great Britain. All rights reserved 0041-0)101/95 S9.50+0.00 SHORT COMMUNICATIONS DETECTION OF A PROTEINACEOUS TOXIN IN THE BRACKISHWATER CLAM (CORBICULA JAPONICA) KAZUO SHIOMI,’ JUNYA ARITA,’ YUJI NAGASHIMA’ and AKIRA SHINAGAWA2 ‘Department of Food Science and Technology, Tokyo University of Fisheries, Koian-4, Minato-ku, Tokyo 108, Japan; and *Gakushuin Women’s Junior College, Toyama, Shinjuku-ku, Tokyo 162, Japan (Received 8 October 1994; accepted 4 Jonuory 1995) K. Shiomi, J. Arita, Y. Nagashima and A. Shinagawa. Detection of a proteinaceous toxin in the brackishwater clam (Corbicula japonica). Toxicon 33, 699-702, l!XV.-Water extracts from the brackishwater clam (Corbicula japonica) were found to be lethal to mice upon i.v. injection. Muscular tissues (foot muscle, adductor muscle, mantle muscle, mantle and siphon) were all toxic while gill and viscera (mid-gut gland, testis and ovary) were nontoxic; the highest toxicity was observed with foot muscle. The toxin was judged to be a thermolabile, basic protein with a mol. wt of about 13,000. The brackishwater clam (Corbicula japonica) is widely consumed in Japan, chiefly as an ingredient for ‘miso’ soup. To our knowledge, there has been no available information on toxicity or toxins of C. japonica. In the course of our recent screening for toxins in aquatic animals, however, water extracts from C. japonica were found to be potently lethal to mice upon i.v. injection. Moreover, the C.japonica toxin was distinguished from the well-known two types of toxins in marine bivalves, paralytic shellfish poisons (PSPs) and diarrhetic shellfish poisons (DSPs). We describe here the toxicity of C. japonica and present evidence that the C. japonica toxin is proteinaceous. Specimens of C. japonica were collected at various lakes and rivers (four brackish lakes and brackish estuaries of two rivers) in Japan from May to September in 1994 (see Table 1 for details). All specimens were transported alive to our laboratory and immediately subjected to experiments’ or kept alive in a cold room (4°C) until used at least within 5 days. The whole soft tissue obtained from each specimen was homogenized in 2 vols of distilled water and centrifuged at 18,800 x g for 5 min. The resultant supernatant was used as crude toxin. In order to clarify the anatomical distribution of toxin, crude toxins were also prepared from various tissues (foot muscle, adductor muscle, mantle muscle, mantle, siphon, gill, mid-gut gland, testis and ovary) pooled from 66 specimens (from Lake Jinzai) comprising 42 male and 24 female specimens. For the determination of lethal activity, groups of two male mice (ddY strain) weighing about 20 g, which were purchased from 699