Solubility of tail tendon collagen from normal, streptozotocin-induced diabetic Lewis rats, and diabetic animals treated with aminoguanidine and two novel advanced glycosylation end products (AGE)-formation inhibitors was investigated by limited pepsin digestion under acidic conditions. Assays were conducted using tail tendon collagen from Lewis rats obtained from two different vendors, Harlan and Charles River Laboratories. Collagen solubility was assessed by following the kinetics of pepsin digestion. The data revealed that the rate of digestion for diabetic animals is markedly slow relative to that of normals. More strikingly, the kinetics of the diabetic animals showed the feature of a lag in digestion regardless of the animal source. Experiments designed to optimize the difference in solubility between normal and diabetic animals demonstrated that Charles River animals exhibit a greater window of solubility than the Harlan animals. More importantly, a pronounced effect of aminoguanidine, an AGE-formation inhibitor, was observed in Charles River animals, but not in the Harlan animals, presumably because of the larger window of solubility between the normal and the diabetic animals in the former. These data indicated that the Charles River Lewis rats are an animal model that demonstrates greater efficacy in this assay. Analysis of in vivo screens designed to test efficacy of aminoguanidine and two novel AGE-formation inhibitors, ALT 462 and ALT 486, demonstrated that monitoring an in vivo dose response is highly dependent on the enzyme concentration as well as the time of digestion, and that 1.5 h of digestion and 10 microg/ml pepsin (5 μg pepsin/mg collagen) appeared optimal. Under these conditions, a 29% normalization of solubility was observed with aminoguanidine at 100 mg/kg body wt, whereas a similar normalization was observed at 10 mg/kg body wt for both ALT 462 and ALT 486. Thus, on a molar basis, ALT 462 and ALT 486 are at least 20 times more potent than aminoguanidine. This is the first demonstration of dose-dependent efficacy for AGE-formation inhibitors in animal models, and as such, this assay provides a method with which to assess the in vivo efficacy of other such inhibitors.