Recent Reports on the Mechanism of the Action of the Arylsulfonylureas

The current status of this problem was discussed in an editorial in the January-February 1958 issue of DIABETES. Three recently published papers presenting new data warrant further comment. The (3-cytotropic hypothesis of the mechanism of action of these hypoglycemic drugs is gaining increased validity. It postulates that they stimulate the [J cells of the pancreas to secrete an extra amount of insulin into the portal vein which exerts its first action upon the liver. The liver, in consequence, diminishes its output of glucose into the blood by: (1) increasing glycogen storage; (2) decreasing formation of glucose from protein or other precursors; or (3) stimulating lipogenesis from carbohydrate. Hypoglycemia results. Unless the secreted insulin is completely bound by the liver so that none of it is free to reach the systemic circulation, it would be expected that the utilization of carbohydrate in the periphery would be increased. Many authors, however, have been unable to find unequivocal evidence of such increased peripheral action of the extra endogenous insulin presumably secreted in response to the sulfonylureas. This is not in complete accord with the expectations. The P-cytotropic hypothesis, nevertheless, has gained increased support by experiments recently published. Craig, Molzahn, Woodward, and Miller compared the actions of tolbutamide and insulin when injected into the brachial artery on the arteriovenous difference of the forearm. In six nondiabetic subjects they failed to show any effect of tolbutamide on the arteriovenous blood glucose difference. In contrast, insulin in small doses gave highly significant effects. The authors conclude that the data clearly indicate "that under the conditions of these experiments tolbutamide was exerting its blood glucose lowering effect at a nonperipheral site." A second recent paper is that of Madison and Unger. They compared the effects of glucagon-free insulin injected into the portal vein with that following injection into a peripheral vein. They supposed that "insulin secreted into the portal vein, because it passes directly into the liver, may have a metabolic effect different from insulin administered peripherally." In eight dogs under controlled conditions they determined the blood arteriovenous glucose difference as a measure of peripheral glucose utilization. This was significantly diminished when the same amount of insulin was injected endoportally as compared with peripheral injection. In other words, on the basis of these results it would be possible to suppose that, if the endogenous insulin secreted in response to arylsulfonylurea traversed the liver first, its peripheral action might be abolished or reduced to levels which escape quantitation. The conclusions have obvious significance in the consideration of the (3-cytotropic action of these oral hypoglycemic agents. The paper of Jacobs, Reichard, Goodman, Friedman, and Weinhouse adds further supportive evidence to this hypothesis. In forty diabetic and nondiabetic patients in the postabsorptive state they contrasted the effects of intravenous injections of insulin and tolbutamide on the specific radioactivity of blood glucose after a priming dose of this isotopic sugar. From the levels of blood sugar and its specific radioactivity they calculated: (1) the rate of entry of glucose from the liver into the blood, and (2) its peripheral rate of utilization. In the case of insulin they found, concomitant with the onset of hypoglycemia, that the specific activity of the blood glucose remained constant. This plateau of glucose specific activity is the result of the suppression of glucose output by the liver. In addition, the calculated rate of glucose utilization by the periphery was significantly increased. The same type of data was obtained following the intravenous injection of tolbutamide. In this instance the plateau of specific activity of blood glucose associated with the onset of hypoglycemia also was found as in the case of insulin. Again this evidence indicates a suppression of entry of glucose from the liver into the blood. However, the data indicated no increase in peripheral glucose utilization, a result which is in sharp contrast with that following insulin injection. They concluded "that the lack of influence of tolbutamide on peripheral blood utilization . . . coupled with the virtual certainty that this drug requires insulin for its hypoglycemic action suggests to us that the physiological action of the insulin secreted by the pancreas may also be exerted principally or exclusively on the hepatic glucose output." The validity of these conclusions was strengthened by further observations of these