Recently we demonstrated clinically significant levels of a previously unrecognized metabolite of the anticancer drug 5-fluorouracil (FUra) in bile ofcancer patients. In the present study, reanalysis of bile from these patients demonstrated the presence of not one but two previously unrecognized metabolites. The major unrecognized metabolite was purified by reversed-phase HPLC, after which its molecular weight was determined by fast-atom-bombardment mass spectrometry to be 497. The similarity in HPLC retention times and molecular weights of this FUra derivative and the bile acids N-cholylglycine (Mr 465) and N-cholyltaurine (Mr 515), along with the structural similarity of the FUra catabolite 2-fluoro-,B-alanine and the amino acids glycine and taurine, led to the hypothesis that this metabolite could be a conjugate of 2-fluoro-,B-alanine and cholic acid. This hypothesis was tested and confirmed by hydrolyzing the purified metabolite by cholylglycine hydrolase after which: (i) 2-fluoro-,-alanine was demonstrated by using a sensitive HPLC technique capable of resolving all of the known putative FUra metabolites, and (ii) unconjugated cholic acid was identified by both GC and GC-MS. Additionally, chemically synthesized N-cholyl-2-fluoro-8-alanine was shown to cochromatograph on HPLC and TLC with the purified biiary metabolite. In summary, this study demonstrates a unique, so far as we know, pathway of drug metabolism in man in which an amino acid drug metabolite is conjugated with cholic acid and eliminated into the bile.Furthermore, the finding that 2-fluoro-p-alanine is conjugated to bile acids may provide some insight Into the mechanism of cholestasis that is frequently observed after administration of fluoropyrimidine by hepatic arterial infusion.Although first synthesized 30 years ago, fluoropyrimidines are still widely used in the treatment of a variety of cancers.The anabolism of the fluoropyrimidines to cytotoxic nucleotides has been extensively examined (1, 2) and appears to be responsible for the chemotherapeutic action of these agents. On the other hand, relatively few studies have focused on fluoropyrimidine catabolism. This is despite the fact that >80% of the administered dose is metabolized via the catabolic route (3). Additionally, inadequacies in the analytical techniques previously used for the separation of fluoropyrimidine catabolites has until recently prevented a comprehensive examination of the catabolism of these agents. However, these technical problems have been overcome by the development of a specific HPLC method for the separation of all of the putative fluoropyrimidine metabolites (4).Our laboratory has demonstrated the formation of a 5-fluorouracil (FUra) glucuronide conjugate in isolated rat hepatocytes (5) and the presence of an unidentified FUra metabolite in bile of cancer patients (6). In the present study, further analysis of bile from these patients has shown the presence of two previously unrecognized FUra metabolites. These metabolites did not appear to represent FUra...