] i has not been raised by the first pathway; i.e., as long as glucose has not reached its threshold concentration. The alteration of this hierarchy by long-acting sulfonylureas or genetic inactivation of K ATP channels may lead to inappropriate insulin secretion at low glucose. The amplifying pathway serves to optimize the secretory response not only to glucose but also to nonglucose stimuli. It is impaired in ␤-cells of animal models of type 2 diabetes, and indirect evidence suggests that it is altered in ␤-cells of type 2 diabetic patients. Besides the available drugs that act on K ATP channels and increase the triggering signal, novel drugs that correct a deficient amplifying pathway would be useful to restore adequate insulin secretion in type 2 diabetic patients.

Decreases in both b-cell function and number can contribute to insulin deficiency in type 2 diabetes. Here, we quantified the b-cell mass in pancreas obtained at autopsy of 57 type 2 diabetic (T2D) and 52 non-diabetic subjects of European origin. Sections from the body and tail were immunostained for insulin. The b-cell mass was calculated from the volume density of b-cells (measured by point-counting methods) and the weight of the pancreas. The pancreatic insulin concentration was measured in some of the subjects. b-cell mass increased only slightly with body mass index (BMI). After matching for BMI, the b-cell mass was 41% (BMI < 25) and 38% (BMI 26-40) lower in T2D compared with non-diabetic subjects, and neither gender nor type of treatment influenced these differences. b-cell mass did not correlate with age at diagnosis but decreased with duration of clinical diabetes (24 and 54% lower than controls in subjects with <5 and >15 years of overt diabetes respectively). Pancreatic insulin concentration was 30% lower in patients. In conclusion, the average b-cell mass is about 39% lower in T2D subjects compared with matched controls. Its decrease with duration of the disease could be a consequence of diabetes that, with further impairment of insulin secretion, contributes to the progressive deterioration of glucose homeostasis. We do not believe that the small difference in b-cell mass observed within 5 years of onset could cause diabetes in the absence of b-cell dysfunction.

No abstract

] c oscillations in all beta cells of an individual islet induce pulsatile insulin secretion, but these features of the signal and response are dampened in groups of intrinsically asynchronous islets. Glucose has hardly any influence on the amplitude of [Ca 2+ ] c oscillations and mainly controls the time course of triggering signal. Amplitude modulation of insulin secretion pulses largely depends on the amplifying pathway. There are more similarities than differences between the two phases of glucose-induced insulin secretion. Both are subject to the same dual, hierarchical control over time and amplitude by triggering and amplifying pathways, suggesting that the second phase is a sequence of iterations of the first phase.

Glucose stimulation of insulin release involves closure of ATPsensitive K+ channels, depolarization, and Ca2+ influx in B cells. Mouse islets were used to investigate whether glucose can still regulate insulin release when it cannot control ATP-sensitive K+ channels. Opening ofthese channels by diazoxide (100-250 Mmol/liter) blocked the effects of glucose on B cell membrane potential (intracellular microelectrodes), free cytosolic Ca2+ (fura-2 method), and insulin release, but it did not prevent those of high K (30 mmol/liter). K-induced insulin release in the presence of diazoxide was, however, dose dependently increased by glucose, which was already effective at concentrations (2-6 mmol/liter) that are subthreshold under normal conditions (low K and no diazoxide). This effect was not accompanied by detectable changes in B cell membrane potential.Measurements of 'Ca fluxes and cytosolic Ca2" indicated that glucose slightly increased Ca2+ influx during the first minutes of depolarization by K, but not in the steady state when its effect on insulin release was the largest. In conclusion, there exists a mechanism by which glucose can control insulin release independently from changes in K+-ATP channel activity, in membrane potential, and in cytosolic Ca2+. This mechanism may serve to amplify the secretory response to the triggering signal (closure of K+-ATP channels -depolarization -Ca2" influx) induced by glucose. (J. Clin. Invest. 1992. 89:1288-1295