A model to predict the ATP equivalents of macronutrients absorbed from food

Calculating the physiologically available energy of food at the cellular level (ATP), based on known stoichiometric relationships and predicted nutrient uptake from the human digestive tract may be more accurate than using currently available factorial or empirical models for estimating dietary energy. The objective was to develop a model that can be used for describing the ATP costs/yields associated with the total tract uptake of the energy-yielding nutrients for an adult human in a state of weight loss (sub-maintenance energy intakes). A series of predictive equations for determining ATP yields/costs were developed and applied to the uptake of each energy-yielding nutrient, as predicted separately in the upper-digestive tract and the hindgut using a dual in vivo-in vitro digestibility assay. The costs associated with nutrient ingestion, absorption and transport and with the synthesis and excretion of urea produced from amino acid catabolism were calculated. ATP yields (not including costs associated with digestion, absorption and transport) were predicted as 28.9 mol ATP per mol glucose; 4.7-32.4 mol ATP per mol amino acid and 10.1 mol ATP per mol ethanol, while yields for fatty acids ranged from 70.8 mol ATP per mol lauric acid (C12) to 104 mol ATP per mol linolenic acid (C18 : 3). The energetic contribution of hindgut fermentation was predicted to be 101.7 mmol ATP per g organic matter fermented. The model is not proposed as a new system for describing the energy value of foods in the diet generally, but is a means to give a relative ranking of foods in terms of physiologically available energy (ATP) with particular application in the development of specialised weight-loss foods.