The relative ability of glucose and ascorbate to glycate and crosslink lens proteins in vitro. off

Nonenzymatic glycation by glucose and/or ascorbate leads to the formation of advanced glycation end products (AGEs), which are thought to be a critical element in lens protein aging and cataract formation. The relative participation of these two glycating agents was evaluated in vitro. The incubation of 100 mM [U-14C]-D-glucose and 10 mM [U-14C]-L-ascorbate with lens proteins resulted in an increasing incorporation over 3 weeks, reaching a maximum of 100 nMol mg-1 protein and 160 nMol mg-1 protein with ascorbate. Glycation was proportional to carbohydrate concentration with both reagents, however ascorbate was 18-fold more reactive with lens proteins than glucose. Protein crosslinking was not obvious with 250 mM glucose as measured by SDS-PAGE, however, ascorbate caused extensive crosslinking even at 3.0 mM. The sugar-dependent incorporation of N alpha-formyl-[U-14C]-L-lysine ([U-14C]Nfl) into proteins, gave values of 1.5 nMol mg-1 protein after 3 weeks with 100 mM glucose compared to 11 nMol mg-1 protein with 10 mM ascorbate. On a molar basis, ascorbate was 70-fold more active than glucose and 100-fold more active than fructose in the crosslinking assay. N alpha-formyl-N epsilon-fructosyllysine (1.0 mM) dissociated to cause the incorporation of 1.2 nMol of [U-14C]NfL, but 1.0 mM 3-deoxyglucosone, the putative active dissociation product of fructosyl-lysine, produced only 1.5 nMol mg-1 protein of crosslinks. The chelator, DTPA, had little or no effect on crosslinking in our assay except at the highest carbohydrate level. These data argue that glucose crosslinking can be shown in vitro with lens proteins, however, it does not proceed significantly via 3-deoxyglucosone, and does not require transition metal ion-mediated oxidation to occur. Quantitatively, however, it is almost two orders of magnitude less than the crosslinking by ascorbate oxidation products in vitro.