Low-intensity therapeutic ultrasound effects on intracellular sodium and reactive oxygen species in ex vivo human islets for control of insulin release.

Ultrasound has shown potential to non-invasively stimulate insulin release from pancreatic β-cells, but the underlying mechanisms remain poorly understood. Prior studies have suggested that mechanical effects from ultrasound, such as membrane deformation and cavitation, may play a role in modulating ion channel activity and insulin secretion. However, direct observations of intracellular signaling responses in human islets during ultrasound application are limited. In this work we investigate the effects of ultrasound on human islets in vitro using confocal fluorescence microscopy and enzyme-linked immunosorbent assay (ELISA) to measure intracellular sodium and reactive oxygen species (ROS). Human islets obtained through the NIDDK Integrated Islet Distribution Program (IIDP) were treated with ultrasound, and high glucose. Ultrasound was applied using an 800 kHz unfocused transducer for 3 min at 0.5 W/cm(2) and a 16.6 % duty cycle. Fluorescence imaging showed that high glucose stimulus significantly increased intracellular sodium fluorescence (0.45 ± 0.71 A.U.) compared to ultrasound (0.03 ± 0.10 A.U.) and control (-0.012 ± 0.06 A.U.) (p = 0.001). Ultrasound also increased intracellular sodium fluorescence, though less than glucose. Reactive oxygen fluorescence measured with DCFH-DA indicated increased ROS across all conditions, with ultrasound inducing the highest increase. ELISA measures revealed high glucose caused an initial decrease in extracellular insulin (-14.36 ± 24.69 µU) followed by a moderate increase (2.2 ± 4.76 µU). Our fluorescence microscopy results indicate that high glucose treatment is causing an increase in intracellular sodium levels over the course of the fluorescence recording with that increase being significantly higher (p < 0.05) compared to ultrasound application and control. These results suggest that ultrasound stimulated insulin release does not mimic glucose-driven sodium dynamics but exert bioeffects through shorter time scale mechanical pathways. The trend of higher ROS production in the β-cell with ultrasound application compared to glucose application may suggest that ultrasound induced ROS production occurs at rates higher than physiological glucose induced levels and while ROS generation is metabolic signal in glucose induced insulin release, should be carefully considered when developing ultrasound based therapies for type 2 diabetes.