Box-modeling of 15N/14N in mammals.

The 15N/14N signature of animal proteins is now commonly used to understand their physiology and quantify the flows of nutrient in trophic webs. These studies assume that animals are predictably 15N-enriched relative to their food, but the isotopic mechanism which accounts for this enrichment remains unknown. We developed a box model of the nitrogen isotope cycle in mammals in order to predict the 15N/14N ratios of body reservoirs as a function of time, N intake and body mass. Results of modeling show that a combination of kinetic isotope fractionation during the N transfer between amines and equilibrium fractionation related to the reversible conversion of N-amine into ammonia is required to account for the well-established approximately 4 per thousand 15N-enrichment of body proteins relative to the diet. This isotopic enrichment observed in proteins is due to the partial recycling of 15N-enriched urea and the urinary excretion of a fraction of the strongly 15N-depleted ammonia reservoir. For a given body mass and diet delta15N, the isotopic compositions are mainly controlled by the N intake. Increase of the urea turnover combined with a decrease of the N intake lead to calculate a delta15N increase of the proteins, in agreement with the observed increase of collagen delta15N of herbivorous animals with aridity. We further show that the low delta15N collagen values of cave bears cannot be attributed to the dormancy periods as it is commonly thought, but inversely to the hyperphagia behavior. This model highlights the need for experimental investigations performed with large mammals in order to improve our understanding of natural variations of delta15N collagen.