We have identified a variety of antimitochondrial compounds (e.g. gossypol and rhodamine-123) with selective antitumor properties that can be monitored biochemically and noninvasively within several hours of in vitro or in vivo drug administration, using 31P magnetic resonance spectroscopy (MRS). These drugs appear to have minimal effects on normal host tissues (bone marrow, muscle, brain) and their antitumor sensitivity correlates with intracellular content of cathodal LDH isozymes. The overall objectives of this proposal are to extend these preclinical observations and their potential therapeutic applications by comparative in vitro and in vivo studies, giving special emphasis to the newly isolated enantiomers of gossypol (+, -), and a novel synthetic conjugate of platinum-rhodamine (Pt-R). In vitro cytotoxicity, LDH isozymes, and oncogene expression (c-myc, c- Ha-ras) will be measured in two sensitive and one insensitive human tumor cell lines (breast T47-D, pancreas MiaPaCa; colon HCT8), and two cultured rodent tumor lines ( rat pancreatic AR42j and Syrian golden hamster melanoma) both of which can be carried as implanted rodent tumors for concurrent in vivo study. These results will be compared to drug effects on intracellular phosphate metabolism assessed by 31P-MRS. Pt-R will be compared to its individual components, cisplatin (Pt) and rhodamine-123 (R), and the active (-) gossypol will be compared to its inactive (+) enantiomer. Similar studies will include other drugs whose antitumor effects may in part be due to antimitochondrial properties (adriamycin, methylglyoxalbis- guanylhydrazone, difluoromethylornithine, Nile Blue-A, interferon-alpha, and tumor necrosis factor). In vivo studies will also compare early (1-5 hours) and late (7-14 day) antitumor effects of these antimitochondrials with the route of drug injection and choice of drug vehicle. Changes in T1 and T2 relaxation times calculated from selected tumor regions will be determined using proton magnetic resonance imaging (MRI), and these results will be correlated with the observed 31P-MRS changes noted in the same tumors, and histological analyses using a variety of special stains to identify microscopic features associated with the observed 31P-MRS and proton MRI changes.
