Iseki and Tsunoda (1952) isolated an organism, identified as Bacillus fulminans, an extract of which was capable of specifically inactivating blood group 0 substance. A recent report by Yosizawa (1957) showed that the enzyme liberated fucose as a result of its action. The present study provides an improved procedure for the isolation and purification of this enzyme, a description of some of its properties, and information relating to its mode of action. MATERIALS AND METHODS B. fulminans, according to the 6th edition of Bergey's Manual of Determinative Bacteriology (Breed et al., 1948) (this organism is not referred to in Vol. I of the 7th edition), is a variety of Bacillus cereus and, using the scheme devised by Smith et al. (1952), the organism which we obtained from Dr. Iseki proved to be a strain of B. cereus; two strains of B. cereus from our stock collection did not, however, produce the blood group 0 inactivating enzyme. Our preparations of blood group 0 substances were isolated from saliva, human ovarian cyst fluid, porcine stomach linings, and from commercial porcine gastric mucin by the phenol extraction method of Annison and Morgan (1952); the isolation procedures and properties of the substances are described in a previous publication by Baer et al. (1954). When saliva was to be tested, it was collected over a period of not more than ½ hr, then, unless used immediately, frozen and maintained at-20 C. For use, the saliva was rapidly thawed and centrifuged at 2 to 6 C to effect clarification. Blood group 0 activity was determined by the test of inhibition of hemagglutination, employing either high titered normal chicken serum or an extract of Ulex europeus seeds as sources of group 1 Supported by research grant 2964 of the

Summary Substances have been isolated from the ovarian cyst of a woman of blood group A who is a nonsecretor of A substance, and from the saliva of a man of group B who is a nonsecretor of B substance. Injection of these materials into chickens resulted in the production of precipitating antibody. The antisera exhibited a specificity for the secretions and derivative substances of A, B and O nonsecretors; there was good correlation between Lea activity of the sample and its ability to precipitate the chicken antisera. It has not, however, proven possible to demonstrate experimentally prepared specific agglutinins for Le(a+) red cells nor can such cells absorb the precipitins from the chicken sera.

No abstract