Imaging pancreatic ß-cell in vivo using manganese-enhanced magnetic resonance imaging

Abstract

Diabetes is characterized by absolute or relative deficiency of insulin secretion by β-cells. Currently, there are no non-invasive diagnostic tools for assessing β-cell mass and function in situ. This thesis aims to develop and implement MRI techniques to image the β-cell in vivo. Calcium ion (Ca2+) entry occurs during insulin secretion and the manganese ion (Mn2+) has been used as a Ca2+ surrogate to study Ca2+ transport in β-cells. Mn2+ is also a positive T1 contrast agent and imaging [Mn2+] changes with manganese-enhanced MRI (MEMRI) may be used to monitor Ca2+ influx during insulin secretion. I hypothesize activated β-cells take up more manganese than resting cells after manganese chloride (MnCl2) administration; therefore, the glucose-stimulated pancreas may show higher signal intensity (SI) than the non-stimulated pancreas by T1-weighted MRI. Being thin and diffuse, the mouse pancreas is difficult to image. It was found to be best delineated by magnetization prepared rapid gradient echo (MP-RAGE) MRI. However, MP-RAGE is not conventionally used for quantitative studies and the relationship between MP-RAGE data and [Mn2+] have to be determined. First, an in vitro study was performed and showed a positive correlation between the effective R1 (R1-effective) values and [Mn2+]. Then, SI profiles and R1-effective values at increasing Mn2+ doses were obtained in the pancreas. Additionally, there was a linear correlation between tissue [Mn2+] by inductively-coupled plasma atomic emission spectrometry and R1-effective by MP-RAGE. The results showed that the MP-RAGE sequence can be used in a semi-quantitative manner. Subsequently, the methodology was applied to image the pancreas in vivo, with and without glucose challenge, in healthy and streptozotocin-induced diabetic mouse models. It revealed a statistically greater signal in the glucose-stimulated pancreas compared to control in healthy mice but not in diabetic mice. Further, Mn2+ infusion appeared to have minimal effects on blood glucose levels and islet morphology. Muscle glucose uptake is also a Ca2+-regulated process and therefore MEMRI was applied in the muscle. Results showed increased manganese uptake in glucose-stimulated muscle, suggesting MEMRI may be used for monitoring muscle glucose uptake. This thesis demonstrates an in vivo methodology to detect enhanced Mn2+ influx in the activated pancreas and skeletal muscle, opening up opportunities for assessing β-cell and skeletal muscle function during normal and abnormal glucose homeostasis.