Since current methods of chemotherapy for adenocarcinoma of the colon are essentially ineffective, this study was designed to test for enzymatic differences between tumors and normal colon that might form the basis for more effective treatment. Human colon tumor xenografts also were examined and were found to be very similar to primary tumors when tested for: uridine-cytidine (Urd-Cyd) kinase and orotidine 5'-phosphate (OMP) decarboxylase activity, apparent Michaelis constants of Urd, ATP, and OMP, and temperature and pH optima for Urd-Cyd kinase. However, enzyme activity levels varied from one xenograft line to another, and these differences could not be correlated with growth rate or sensitivity to 5-fluorouracil (5-FU). The xenograft, therefore, may provide a suitable model for the study of human colorectal adenocarcinoma, but care must be taken to screen different lines in order to select ones that are comparable to primary tumors. Primary tumors and xenografts, when compared to normal colon, were found to have significantly higher specific activities of enzymes of both the de novo and salvage pathways of uridine monophosphate (UMP) biosynthesis. The activities of Urd-Cyd kinase and OMP decarboxylase were greater by 132% and 91%, respectively, in primary tumors and 186% and 63%, respectively, in xenografts. Consequently, effective treatment of adenocarcinoma of the colon using inhibitors of pyrimidine nucleotide biosynthesis would probably require the combination of a compound that inhibits the salvage pathway, e.g., inhibitors of Urd-Cyd kinase, with one that inhibits the de novo pathway, e.g., pyrazofurin or N-(phosphonacetyl)-L-aspartate (PALA).