Paracellular intestinal absorption of glucose, creatinine, and mannitol in normal animals: relation to body size

Mice, rats, or rabbits were provided with a liquid diet of 10-12% glucose (Glc), 0.5-1% creatinine, 1-2% mannitol, and mannitol labeled with 3H on a terminal carbon. Average rates of ingestion of Glc were greater than maximum rates of active, carrier-mediated Glc transport reported for the intestines of these species. The discrepancy was small in mice but increased exponentially with body weight (BW). Ingestion-absorption of Glc increased exponentially with the 0.73 power of BW as expected from metabolic rate, whereas active transport, estimated from the literature, varied exponentially with the 0.50 power of BW. It is estimated that in humans the ingestion-absorption rate of Glc may be 10-20 times greater than active transport. In the presence of Glc, 50-65% of the ingested creatinine was recovered in urine compared with 75-85% recovered after intraperitoneal or subcutaneous injections. The amount of creatinine recovered in urine depended on the amount of ingested Glc, as predicted from the effects of Glc on width and permeability of absorptive cell junctions. Eighty percent or more of the 3H label on mannitol was recovered in urine or other body fluids, although most of the [3H]mannitol was oxidized to [3H]water after being absorbed intact from the intestine. It is concluded that in the presence of Glc, creatinine and mannitol (together with Glc, amino acids, and other small nutrients) are absorbed passively by solvent drag between absorptive cells, as found previously in anesthetized rats (J. R. Pappenheimer and K. Z. Reiss. J. Membr. Biol. 100: 123-136, 1987). The ratio of solvent drag to carrier-mediated transport increases exponentially with BW and may account for the capacity of human intestines to absorb large amounts of Glc during prolonged exercise, Glc tolerance tests, or oral Glc-saline therapy for dehydration.