Abstract
The active fraction obtained from B. coli filtrate by the method described in the preceding paper was dissolved in water and evaporated to dryness in vacuo. A white residue was obtained, largely crystalline in nature. Strong heating produced some charring, but much of the crystalline residue was unaffected indicating that a considerable portion was inorganic in nature.
On repeated treatment of the active fraction with methyl alcohol, most of it (82%) went into solution. The methyl alcohol solution was evaporated to dryness and taken up in water, but most of it now failed to dissolve. However, addition of sodium hydroxide effected solution. Acidification with hydrochloric acid produced a heavy, flocculent precipitate which dissolved again on further addition of alkali. This solution was neutralized and tested on tumor-bearing mice; it was completely inactive as far as production of hemorrhage was concerned.
The methyl alcohol-insoluble portion (18%) dissolved readily in a small volume of water, with formation of a permanent foam. The presence of surface tension reducing material had been noted early in the purification, and its occurrence paralleled the presence of the active agent in the course of the fractionation. A stock solution was prepared containing 1 mg. per cc. of this fraction. This solution produced hemorrhage in mouse tumors in doses of 0.2 cc. of a 1 :500 dilution, i. e., the minimum hemorrhage-producing dose contained 0.0004 mg. of active material.
Upon evaporation to dryness in vacuo the active solution gave a residue which was largely non-crystalline. A comparatively small amount of crystalline material was present in which 3 crystal types were noted; the bulk of the residue consisted, however, of a film of non-crystalline glassy material.
The above-described stock solution of the active fraction gave a negative biuret test and a strongly positive Molisch test.
Get full access to this article
View all access options for this article.