Longitudinal noninvasive PET-based beta cell mass estimates in a spontaneous diabetes rat model

Abstract

Diabetes results from an absolute or relative reduction in pancreatic beta cell mass (BCM) leading to insufficient insulin secretion and hyperglycemia. Measurement of insulin secretory capacity is currently used as a surrogate measure of BCM. However, serum insulin concentrations provide an imprecise index of BCM, and no reliable noninvasive measure of BCM is currently available. Type 2 vesicular monoamine transporters (VMAT2) are expressed in human islet beta cells, as well as in tissues of the CNS. [11C]Dihydrotetrabenazine ([11C]DTBZ) binds specifically to VMAT2 and is a radioligand currently used in clinical imaging of the brain. Here we report the use of [11C]DTBZ to estimate BCM in a rodent model of spontaneous type 1 diabetes (the BB-DP rat). In longitudinal PET studies of the BB-DP rat, we found a significant decline in pancreatic uptake of [11C]DTBZ that anticipated the loss of glycemic control. Based on comparison of standardized uptake values (SUVs) of [11C]DTBZ and blood glucose concentrations, loss of more than 65% of the original SUV correlated significantly with the development of persistent hyperglycemia. These studies suggest that PET-based quantitation of VMAT2 receptors provides a noninvasive measurement of BCM that could be used to study the pathogenesis of diabetes and to monitor therapeutic interventions.

Figures

Figure 1. Representative reconstructed serial PET images of coronal abdominal planes of BB-DP rats during progression of autoimmune diabetes.

The bodies of the pancreata of animals A and G are identified by large arrows; the livers are identified by the letter L. Approximately 300 μCi of (+)-α-[11C]DTBZ was used for imaging. Euglycemic male rats were imaged to establish a baseline at approximately 7 weeks of age; these animals were reimaged 2 or 3 times during the following 5 weeks. Whole-venous-blood glucose concentrations of 7-week-old animals fasted for 4 hours ranged from 90 to 135 mg/dl. Reconstructed images represent the summed data of the entire 1-hour-long scanning period. Display ranges of images are not equivalent. Quantitation of activity within pancreas and liver regions of interest during the scan period is shown in the time-activity curves in the panels at the right. The age of the animal in weeks appears in the lower right corner. The average activity of [11C]DTBZ in nCi/cc of tissue within the liver or pancreatic regions of interest versus the duration of the PET scan in seconds is plotted on the ordinates and abscissae, respectively. Solid lines represent activity in the liver, and open circles represent activity in pancreas regions of interest.

Figure 2. Relationship between the max SUVs of [11 C]DTBZ and AUC IPGTT during progression of diabetes in the BB-DP rat.

The max SUVs obtained from the time-activity curves for each animal imaged during the study are plotted against the corresponding glucose tolerance test index (AUC IPGTT) (diamonds). The age of the animal (in weeks) appears next to the AUC IPGTT and max SUV values. The animal ID (A–G) appears in the upper-right corner of each graph.

Figure 3. Transaxial reconstruction of a male-baboon abdominal dynamic PET scan with [11 C]DTBZ.

Kidney (K), pancreas (P), and stomach (S) are indicated in these summed early frames.