Beta-cell function and mass in type 2 diabetes.

The aim of the work described here was to improve our understanding of beta-cell function (BCF) and beta-cell mass (BCM) and their relationship in vivo using the minipig as a model for some of the aspects of human type 2 diabetes (T2DM). More specifically, the aim was to evaluate the following questions: How is BCF, especially high frequency pulsatile insulin secretion, affected by a primary reduction in BCM or by primary obesity or a combination of the two in the minipig? Can evaluation of BCF in vivo be used as a surrogate measure to predict BCM in minipigs over a range of BCM and body weight? We first developed a minipig model of reduced BCM and mild diabetes using administration of a combination of streptozotocin (STZ) and nicotinamide (NIA) as a tool to study effects of a primary reduction of BCM on BCF. The model was characterized using a mixed-meal oral glucose tolerance test and intravenous stimulation with glucose and arginine as well as by histology of the pancreas after euthanasia. It was shown that stable, moderate diabetes can be induced and that the model is characterized by fasting and postprandial hyperglycemia, reduced insulin secretion and reduced BCM. Several defects in insulin secretion are well documented in human T2DM; however, the role in the pathogenesis and the possible clinical relevance of high frequency (rapid) pulsatile insulin secretion is still debated. We therefore investigated this phenomenon in normal minipigs and found easily detectable pulses in peripheral vein plasma samples that were shown to be correlated with pulses found in portal vein plasma. Furthermore, the rapid kinetics of insulin in the minipig strongly facilitates pulse detection. These characteristics make the minipig particularly suitable for studying the occurrence of disturbed pulsatility in relation to T2DM. Disturbances of rapid pulsatile insulin secretion have been reported to be a very early event in the development of T2DM and include disorderliness of pulses and reduced ability to entrain pulses with glucose. However, the role of reduced BCM and/or obesity in the development of these defects in humans is unknown. Therefore, the investigations were extended to include lean NIA/STZ minipigs where it was shown that a primary reduction of BCM leads to reduced insulin pulse mass but does not change periodicity of the pulses or the ability of glucose to entrain pulses. In contrast, obesity was found to be associated with reduced pulsatile insulin secretion and improved orderliness of glucose entrained pulses in the minipig. Furthermore obesity was associated with pancreatic lipid accumulation and increased beta-cell volume, although BCM relative to body weight was not changed. Finally, a combination of obesity and reduced BCM resulted in severely disturbed insulin secretion and severe morphological changes. Thus, results from NIA/STZ minipigs suggest that not all of the defects of rapid pulsatile insulin secretion seen in human T2DM can be explained by a primary reduction of BCM mass or up to 2 weeks of mild hyperglycemia. Furthermore, based on the results from obese minipigs, obesity in itself induces small defects in rapid pulsatile insulin secretion and the combination of obesity and reduced BCM leads to further deterioration of BCF. Another major characteristic of human diabetes is thought to be reduction of BCM and the ability to follow this parameter over time would greatly improve our understanding of disease progression and allow evaluation of pharmacological methods to increase BCM. BCM cannot, at present, be measured in vivo in humans. We therefore set out to further validate data from smaller studies in lean non-human primates and minipigs showing a correlation between measures of BCF in vivo and BCM. In a large study in lean minipigs with a range of BCM, we found that a strong stimulation of insulin secretion with a combination of glucose and arginine resulted in the best correlation to BCM, as determined using stereology. A similar relationship was also shown in a group of both lean and obese animals, thereby supporting the application of similar methods to estimate BCM in humans over a range of body weights. Since changes in rapid pulsatile insulin secretion are detectable early in the development of diabetes and in obesity, we hypothesized that this parameter could also be highly correlated to BCM as it has been shown in smaller studies in lean minipigs. However, rapid pulsatile insulin secretion did not show a better correlation to BCM than combined stimulation with glucose and arginine, and thus analysis of pulses does not provide a better surrogate marker for BCM in the minipig. To evaluate the weaker correlation of glucose stimulation compared to combined glucose and arginine stimulation in vivo with BCM, we further investigated BCF in lean, beta-cell reduced minipigs by studying BCF in vitro after isolation and perfusion of their pancreases to investigate the ability of the remaining beta-cells to compensate for the loss of BCM by increasing insulin secretion per BCM. The perfused pancreas was chosen in order to allow direct measurement of the insulin secretion without the effects of peripheral tissues. During the perfusion, it was shown that the remaining beta-cells were indeed able to compensate for the loss of BCM to a large extent in response to stimulation with glucose and glucagon-like peptide-1 but not in response to arginine. This shows that the type of stimulus applied is important for the ability to compensate for reduced BCM from the remaining population of beta-cells, and further supports the use of combined stimulation with glucose and arginine for estimation of BCM in vivo. In conclusion, an animal model of reduced BCM and mild diabetes has been developed and characterized. The model has been used to evaluate effects of a primary reduction of BCM, showing a reduced rapid insulin pulse mass but normal periodicity and entrainability of the pulses, whereas obesity was associated with reduced rapid pulsatile insulin secretion. Thus, based on these data, the disturbed rapid pulsatile insulin secretion seen in T2DM humans may not directly be explained by the reduced BCM in diabetes, whereas obesity may be related to the reduced pulsatility. Furthermore, the model has been used to establish a correlation between extensive stimulation of insulin secretion in vivo and BCM obtained by stereology in both lean and obese animals. The ability to estimate BCM based on in vivo experiments in the minipig would allow longitudinal studies on changes in this parameter over time in the intact animal and support application of similar methods in humans. Such methods could be useful for the diagnosis and the measurement of the effectiveness of treatment of diabetes in humans in the future.