In Vivo Mouse Abdominal Oxygen Imaging And Assessment of Subcutaneously Implanted Beta Cell Replacement Devices

Type 1 diabetes (T1D) is an autoimmune disease that leads to the loss of insulin-producing pancreatic beta cells. Beta cell replacement devices or bioartificial pancreas (BAP) have shown promise in curing T1D and providing long-term insulin independence without the need for immunosuppressants. Hypoxia in BAP devices damages cells and imposes limitations on device dimensions. Noninvasive in vivo oxygen imaging assessment of implanted BAP devices will provide the necessary feedback and improve the chances of success. Pulse-mode electron paramagnetic resonance (EPR) oxygen imaging (EPROI) using injectable trityl OX071 as the oxygen-sensitive agent is an excellent technique for obtaining partial oxygen pressure (pO ) maps in vitro and in vivo. In this study, our goal was to optimize in vivo mouse abdominal EPROI and demonstrate proof-of-concept pO imaging of subcutaneously implanted BAP devices. All EPROI experiments were performed using a 25 mT EPROI instrument, JIVA-25®. For in vivo EPROI experiments, trityl OX071, a whole-body mouse resonator (∅32 mm × 35 mm), C57BL6 mice, and the inversion recovery electron spin echo (IRESE) pulse sequence were utilized. We investigated the signal amplitude and pO in mouse abdomen region for intravenous (i.v.) and intraperitoneal (i.p.) injection methods with either only a single bolus (B) or bolus plus infusion (BI) for 72.2 mM OX071 and the effect of OX071 concentrations from 18 to 72.2 mM for the i.p.-B injection method. We also investigated the impact of animal respiratory rate on mouse abdominal pO . Finally, we performed proof-of-concept pO imaging of two subcutaneously implanted BAP devices, OxySite and TheraCyte. At the end of the four-week study, the TheraCyte devices were extracted and analyzed for fibrosis, vascular differentiation, and immune cell infiltration. We established that mouse abdominal pO remains stable irrespective of trityl injection methods, concentrations, imaging time, or animal breathing rate. We demonstrate that the i.p.-B and i.p.-BI methods are suitable for EPROI, and i.p.-B method provides higher signal amplitude compared to i.v.-BI and up to 75 min of imaging. An injection with a reduced trityl concentration and higher volume provides higher signal amplitude for i.p.-B method at the beginning. We also highlight the advantage of milder anesthesia for consistent, reliable mouse pO imaging. Finally, we demonstrate that EPROI could provide longitudinal noninvasive oxygen assessment of subcutan