[CANCER RESEARCH 38. 210-214. January 1978] Hydroxylated Metabolites of R,S-1-(Tetrahydro-2-furanyl)-5- fluorouracil (Ftorafur) in Rats and Rabbits1 A. T. Wu, J. L. Au, and W. Sadee School of Pharmacy, University of California, San Francisco, California 94143 ABSTRACT Two hydroxylated metabolites (M and M ) have been isolated from rabbit urine after administration of Ftorafur (FT). The structures of 3'-OH-FT and 4'-OH-FT were as signed to M, and M , respectively. A reverse-phase high performance liquid chromatography assay was developed for measuring FT, M,, M . and 5-fluorouracil (FU) plasma levels. M,, M , and FU were present in rabbit and rat plasma in greatly varying concentrations after FT admin istration. Pharmacokinetic studies suggest that FU for mation proceeds via metabolic intermediate(s) and that the extent of FT activation is variable. A horse liver thymidine phosphorylase preparation capable of catalyz ing the conversion of /¿-ribo-2-deoxy-5-fluorouracN to FU was inactive against FT and M,. However, 20% of M was converted to FU by this enzyme, which suggests that the urinary metabolite M consisted of a mixture of enan- tiomers with 20% present in the natural /3-D configuration. The stereochemistry of M, remains unknown. Hydroxyla- tion of FT to /i-D-4 -OH-FT and subsequent cleavage to FU by thymidine phosphorylase represents one possible activation mechanism of FT to FU. However, lack of cor relation between plasma levels of M and FU indicates that this mode of metabolic activation may account for only part of the overall activation of FT in vivo. INTRODUCTION The pyrimidine antimetabolite FT3 has shown significant activity in several adenocarcinomas with a spectrum of activity similar to but less toxicity than FU (3, 12, 17). FT is considered to be a chemical depot form of FU (4, 6, 7, 16) that is generated in vivo, presumably by hepatic metabolism involving cytochrome P-450 (8, 13). This concept is in agreement with the rather low in vitro cytotoxic activity of FT in the absence of significant amounts of cytochrome P-450 (10). The mechanism of FT conversion to FU remains unknown. Several reports support the hypothesis that FU formation 'Supported by USPHS Grant GM-16496 from the National Institutes of General Medical Sciences and CA-05186 from the National Cancer Institute. Further support was obtained from Training Grant GM 00728-15; from the Earl C. Anthony Fund. University of California. San Francisco; and from Grant RR 00892-01A1 from the Division of Research Resources. NIH, to the Magnetic Resonance Laboratory, of the University of California, San Fran cisco. 3To whom requests for reprints should be addressed. 3The abbreviations used are: FT, Ftorafur [1-(tetrahydro-2-furanyl)-5-fluo- rouracil); FU. 5-fluorouracil; FUdR, /3-ribo-2'-deoxy-5-fluorouracil; GC-MS, gas chromatography-electron impact-mass spectrometry; HPLC. high-per formance liquid chromatography; NMR, nuclear magnetic resonance; TLC. thin-layer chromatography; GC. gas chromatography. Received June 20. 1977; accepted October 14. 1977. may not be the only mechanism of FT activation. Smolyan- skaya and Tugarinov (16) have demonstrated the in vivo presence of a microbiologically active metabolite fraction of FT in addition to FU. Differences in FT and FU toxicities to mice also support an activation mechanism of FT other than, or in addition to, formation of FU (11), while therapeu tic cross-resistance between FT and FU suggests that the major part of the therapeutic mechanism of action is com mon to FT and FU (9). Pharmacokinetic analysis of FT and FU plasma levels following administration of FT to rabbits and rats (18) further indicates the presence of metabolites other than FU that might be independently active or serve as intermediate(s) in the formation of FU. In this study we have used FT specifically labeled either in the tetrahydrofuranyl ring or in the FU moiety in order to investigate metabolic pathways of FT other than formation and subsequent metabolism of FU. With this technique, 2 hydroxylated metabolites of FT were isolated and character ized, and their presence in circulating peripheral blood of rats and rabbits was established. MATERIALS AND METHODS Chemicals and Reagents. All chemicals and reagents were of analytical reagent grade. FT was supplied by the Chemical and Drug Procurement Section, Division of Can cer Treatment, National Cancer Institute, Washington, D. C. [2-14C]FT [1-(tetrahydro-2 -furanyl)-5-fluoro-2-14C-uracil; specific activity, 46 ¿iCi/rng]and [2',5'-14C]FT [1-(tetrahy- dro-2-furanyl-2',5'-14C)-5-fluorouracil, specific activity, 33 ¿iCi/mg]were obtained through the National Cancer Insti tute from Isotope Synthesis Laboratory, Stanford Research Institute, Menlo Park, Calif. Bis-15A/-5-fluorouracil (15N,,-FU) was synthesized from bis-'5/V-thiourea (99% 15N enrich ment; Koch Isotope, Cambridge, Mass.) (15). /3-Ribo-5-fluo- rouracil and FUdR were obtained from Nutritional Biochem- icals Corp., Cleveland, Ohio. Apparatus. GC was performed on a Varian Aerograph 2700 equipped with a glass column (6 ft x 0.25 inch outside diameter; 2 mm inside diameter) packed with 3% OV-1 on Gas-Chrom Q (100 to 120 mesh; Applied Science Laborato ries, State College, Pa.). The injector, column, and detector temperatures were 245, 180, and 275°,respectively. Flow rates for helium, oxygen, and hydrogen were 40, 300, and 30 ml/min, respectively. Mass spectra were obtained by direct insertion on an AEI MS902 mass spectrometer with isobutane chemical ionization and by use of a gas Chro matographie inlet system on a Varian MAT CH-7 electron impact mass spectrometer connected with a Varian Aero graph 2700 gas Chromatograph and controlled by a Nova 2-10 computer (GC-MS). 14C activity was measured in a Searle Analytic Mark III liquid scintillation counter. 210 CANCER RESEARCH VOL. 38 Research. on August 19, 2017. © 1978 American Association for Cancer cancerres.aacrjournals.org Downloaded from