(Principal Investigator's) Lavendamycin and streptonigrin, biosynthetically related compounds, are potent antitumor agents, but their clinical use has been precluded because of high toxicity and poor solubility. We have synthesized 33 variously substituted novel lavendamycins and several of these compounds have shown promising antitumor activity. Some of these analogs appear to be highly ras K oncogene specific (9-130 fold) and had no cytotoxicity against ras K and Lewis Lung oncogene transformed or normal cells. These results suggest that these and other related derivatives may be highly useful drugs for the treatment of tumors whose malignant phenotype is maintained by the ras K oncogene, such as human pancreatic cancer. Treatment with three of the analogs against the K/1-NRK tumor in nude mice resulted in up to 90 percent tumor reduction. In addition, the NCI has screened a number of these compounds against a panel of human tumor lines and eight of these compounds are either in their second (in vitro) or third (in vivo) stages of evaluation. Although our recent work has given us a general understanding of the molecular features which determine the lavendamycins' antitumor activity, it is imperative that we complete the systematic elucidation of the structure-activity relationship (SAR) of this important antibiotic system so that derivatives with better therapeutic indices and greater solubilities can be developed. Approximately 56 new lavendamycins will be prepared by the Picet-Spengler condensation of the desired quinolinedione-2-carbaldehydes with tryptophan derivatives according to our short (5-step) and practical method newly developed for the synthesis of lavendamycin methyl ester. The biological activity of these compounds will be determined on ras K, ras H, ras N, 3LL and parent nontransformed (NRK) cell lines as well as on the NCI's 60 human tumor lines. Furthermore, the inhibitory activity of these analogs toward the cdc25a phosphatase will also be determined. Preliminary studies using this assay have shown several of the analogs to have very promising activity. For more potent derivatives, an assessment of in vivo activity will also be performed. Thus, the main goal of this project is to complete our comprehensive SAR study of the lavendamycin system by synthesizing and screening a series of analogs substituted with various groups ranging in size, shape, electronic effects, oxidation state, polarity and water-solubility. This study will allow us to clearly identify the minimum potent pharmacophore of the lavendamycin system and to determine the role that the parent skeleton and individual substituents play in the activity of the molecule. Based on these findings, we can rationally design analogs which enhance potency, selective cytotoxicity and solubility as potential antitumor drugs.
