New approaches are nescessary for improved chemotherapy of solid tumors. To this purpose, we have taken advantage of the cancer-specific metabolic defect of methionine dependence, which causes many types of cancer cells to reversibly arrest in lateS/G2 of the cell cycle in medium where methionine (Met) has been replaced by homocysteine (Hcy) (Met- Hcy+ medium). In co- cultures of cell lines derived from four types of solid human tumors and equal numbers of normal human fibroblasts, the use of an S-phase drug (e.g., adriamycin) while the cultures were in Met- Hcy+ medium and subsequent use of an Mphase drug (e.g., vincristine) after the cultures were shifted to Med+ Hcy- medium completely eliminated the cancer cells while allowing flourishing growth of the normal cells. We have termed this procedure """"""""Methioninedependent chemotherapy"""""""" (P.H. Stern and R.M. Hoffman. J. Natl. Cancer Inst. 76: 629-639, 1986). To bring methioninedependent chemotherapy closer to clinical trials we propose here to isolate a methioninase that upon administration to the animal or patient will allow the modulation of circulating levels of methionine. We have thus far identified the bacteria Alcaligenes fascalis and Pseudomonas pseudoalcaligenes which contain a methioninase that apparently does not cleave homocysteine. We propose to isolate with stateoftheart genetic selection and protein purification procedures a methioninase from these organisms which has a low Km, a long halflife in the circulation, and a low toxicity for the experimental animal or human patient. The methionine cleaving properties, halflife and toxicity of the enzyme will be tested in rats. The purification and characterization of a desirable enzyme justifies cloning the methioninase gene to be carried out in Phase II methioninase, cellcycle specific chemotherapy, solid tumors, methionine dependence, enzymedrug.