Glucokinase (ATP:D-glucose 6-phosphotransferase, EC 2.7.1.2) from rat islets of Langerhans was partially purified by chromatography on DEAE-Cibacron blue F3GA agarose. The enzyme eluted in two separate peaks. Sigmoidal rate dependence was found with respect to glucose (Hill coefficient = 1.5) for both enzyme fractions. K. values for glucose were 5.7 mM for the major fraction and 4.5 mM for the minor fraction. Neither fraction phosphorylated GlcNAc. A' GlcNAc kinase (ATP. 2-acetamido-2-deoxy-D-glucose' 6-phosphotransferase; EC 2.7

We evaluated the possible role of islet glucokinase in controlling the rate of islet glucose metabolism, and thereby the rate of glucose-induced insulin release. The activities of glucokinase, hexokinase, P-fructokinase, and glyceraldehyde-P dehydrogenase were quantitated in sonicated or isotonically homogenized islet preparations using pyridine nucleotide-dependent fluorometric assays. In sonicates, about 1/4 of the islet glucose phosphorylating activity was due to an enzyme with kinetic properties similar to glucokinase; 3/4 of the activity was due to hexokinase. The procedure for determining islet glucokinase activity was improved by centrifuging isotonic islet homogenates at 12,000 x g. The supernatant fraction was enriched for glucokinase. About 1/2 of the glucose phosphorylating activity in this fraction was due to glucokinase and 1/2 was due to hexokinase. The glucokinase activity in islet homogenates was !23 of the activity of hexokinase, 1/40 of the activity of P-fructokinase, and 1/400 of the activity of glyceraldehyde-P dehydrogenase. Detailed concentration dependency curves of glucose and mannose utilization were also obtained with intact isolated pancreatic rat islets. Glucose and mannose usage in islets was governed by two superimposed hyperbolic systems differing in Km and Vmax. A high Km system (Km for glucose 11 mM and for mannose 21 mM) predominated. A low Km system (Km for glucose 215 and for mannose 530 microM) contributed about 15% to the total activity. The available data with intact islets could be rationalized by the existence of two distinct hexose phosphorylating enzymes with differing capacities and kinetic properties. These enzymes, tentatively identified as glucokinase and hexokinase, could coexist in the same cell or could be distributed among different cell types. The possible physiologic significance of these results is discussed, emphasizing the idea of dual control of glycolysis and insulin release by glucokinase and hexokinase. An earlier proposal that glucokinase serves as glucoreceptor of beta-cells [J. Biol. Chem. 243:2730 (1968)] is greatly strengthened by the present studies.

Starvation refeeding experiments were conducted in rats to test the hypothesis that adaptation of glucokinase (the high Km component of glucose phosphorylation) could be the major determinant of glucose metabolism of pancreatic islet cells and of glucose-stimulated insulin release. It was found that glucokinase of islet homogenates, glucose use by intact isolated islets, and glucose-induced insulin release as studied in a perifusion system were decreased after 24 h of fasting, whereas P-fructokinase and 3-P-glyceraldehyde DH were unaltered. After extended fasting (e.g., 120 h) all three enzymes were decreased but glucose use did not change any further. Refeeding normalized all parameters. These and previous results support the concept that glucokinase serves as the adaptive beta-cell glucoreceptor relating blood glucose to insulin release.

Alloxan inactivated glucokinase in intact, isolated pancreatic islets incubated in vitro. Inactivation of glucokinase was antagonized by 30 mM glucose present during incubation of islets with alloxan. Glucokinase partially purified from transplantable insulinomas or rat liver was inactivated by alloxan with a half-maximal effect at 2-4 microM alloxan. Inactivation of purified glucokinase was antagonized by glucose, mannose, and 2-deoxyglucose in order of decreasing potency but not by 3-O-methylglucose. Glucose anomers at 6 and 14 mM were discriminated as protecting agents, with the alpha-anomer more effective than the beta-anomer. Glucokinase was not protected from alloxan inactivation by N-acetylglucosamine, indicating that the reactive site for alloxan is not the active site; therefore, glucose may protect glucokinase by inducing a conformational change. Glucokinase is thought to be the glucose sensor of the pancreatic beta-cell. The finding that glucokinase is inactivated by alloxan and protected by glucose with discrimination of its anomers similar to inhibition of glucose-stimulated insulin secretion by alloxan supports this hypothesis and appears to explain the mechanism for inhibition of hexose-stimulated insulin secretion by this agent and the unique role of glucose and mannose as protecting agents.

A previous report indicating reduced glucose metabolism in pancreatic islets isolated from old as compared with young rats was reinvestigated. With a modified islet isolation procedure it was found that islets from 12- to 18-month-old rats had increased glucose use, elevated glucokinase, phosphofructokinase and glucose-6-phosphate dehydrogenase when compared with islets from 2-month-old controls. Glucose-induced insulin release in vitro of islets from the older rats was also improved by the more careful method of islet isolation but did not achieve rates observed with islets from young rats. The data suggest an age-related activation of pancreatic islet cell metabolism, possibly in response to overstimulation by increased peripheral insulin resistance, characteristic of older obese rats.