Background: High-protein diets have been found to protect mice from the lethal effects of cytotoxic pyrimidine analogues and to reduce the toxicity of the antipyrimidine fluorouracil (5-FU), but the biochemical explanation for these effects is not known. PALA potentiates the chemotherapeutic efficacy of 5-FU, and each of the two agents can produce dose-limiting intestinal toxic effects. We have shown that intraperitoneal infusion of ammonium chloride stimulates intestinal de novo pyrimidine synthesis. This stimulation with excess ammonia, which can also result from high-protein intake, is dependent on the presence of carbamoyl phosphate synthetase I, an enzyme in the liver and intestine but not in most tumors. These findings suggest that a high-protein diet can stimulate pyrimidine synthesis in the liver and intestine but leave it unchanged in tumor tissue.
Purpose: The purpose of this study was to determine whether varying dietary protein causes pharmacologically relevant and preferential changes in de novo pyrimidine synthesis.
Methods: Mice were fed diets containing 18%, 35%, or 50% casein. Dietary effects on de novo pyrimidine synthesis were measured in the intestine, liver, and B16 mouse melanoma in mice treated with PALA and in untreated mice. De novo synthesis was measured by infusion of [15N]alanine into intact animals, determination of 15N incorporation into uracil by use of gas chromatography-mass spectrometry, and calculation of the fraction of the uracil nucleotide pool formed by de novo synthesis.
Results: In mice on a 50% casein diet (high protein), de novo pyrimidine synthesis increased substantially in the liver and intestine, compared with synthesis in mice receiving 18% casein. Increase in pyrimidine synthesis in B16 tumor tissue was negligible. The high-protein diet protected the intestine and liver from depletion of uracil nucleotide pools by PALA, and toxicity in tumor-free animals was reduced, as determined by mortality after PALA treatment. Sensitivity of the B16 tumor to the biochemical and cytotoxic effects of PALA was not diminished.
Conclusions: We propose that the basis for these effects of a high-protein diet is the generation of excess carbamoyl phosphate in tissues containing carbamoyl phosphate synthetase I. This carbamoyl phosphate can stimulate de novo pyrimidine synthesis and compete with drugs that interact with enzymes of the de novo pathway, thereby selectively protecting the liver and intestine.
Implications: These data provide a biochemical explanation for reported effects of high-protein diet on toxicity of antipyrimidines like 5-FU. Studies are underway to determine if stimulation of pyrimidine synthesis by excess ammonia improves therapy with 5-FU alone or combined with PALA.