Targeted Measurement and Manipulation of Hydrogen Peroxide in the Cardiovascular System

Abstract

Hydrogen peroxide (H2O2) is a reactive oxygen species (ROS) with a diverse range of reported roles in both normal physiology and disease. While H2O2 and other ROS have long been reported to be drivers of acute and chronic disease, the past several decades of research have seen the emergence of a crucial physiologic role for H2O2 in signal transduction. Thus, the field is faced with the paradox that the same molecule necessary for normal physiology can also induce pathologic effects in the very same cell type or organ. The transition between the salutary and pathologic roles of H2O2 is still not well understood and remains an active area of investigation. The development of spatially targeted sensors for ROS has led to the notion that H2O2's many different possible effects in a given cell type may depend on both the spatial origin and enzymatic source of ROS production, but the tools to rigorously test this hypothesis both in vitro and in vivo are lacking. This work identifies distinct roles for H2O2 simultaneously produced by two different NADPH oxidase (NOX) isoforms in the heart in response to insulin stimulation. Fluorescent H2O2 biosensors reveal that both NOX2 and NOX4 generate H2O2 and lead to an increase in cytosolic H2O2. However, H2O2 derived from NOX2 appears to augment the canonical anabolic pathways activated by insulin, while H2O2 from NOX4 appears to attenuate the inotropic effects of beta-adrenergic stimulation, effectively acting as an endogenous ``beta blocker." A new hypothesis is proposed that suggests that the enigmatic cardiac dysfunction associated with diabetes (diabetic cardiomyopathy) results in part from reduced NOX4 activity due to cardiac insulin resistance. Next, a flexible system for generating H2O2 in experimental biologic systems is refined. In order to improve the toolkit of investigators examining subcellular roles of H2O2, a yeast D-amino acid oxidase (DAAO), which generates H2O2 during the degradation of D-amino acids, is fused to a fluorescent H2O2 biosensor and expressed in endothelial cells and cardiac myocytes. Upon addition of millimolar concentrations of D-alanine in vitro, the enzyme is observed to generate H2O2 in both cell types. Cardiac myocytes are found to have several distal insulin signaling proteins as well as members of the mitogen activated protein kinase (MAPK) and glycogen synthase kinase (GSK) families activated upon production of H2O2 in the cytosol. Finally this DAAO technology is applied to endothelial cells to examine the role of local redox dynamics in shear stress responses. Using subcellularly targeted forms of the H2O2-sensitive fluorescent biosensor HyPer, endothelial cells were observed to show greater increases in nuclear H2O2 with laminar shear stress than with oscillatory shear. The opposite was true for the cytosolic form of HyPer, where oscillatory shear stress elicited greater increases in HyPer fluorescence than laminar shear stress. These observations were correlated with previous reports showing that oscillatory shear induces cytokine transcription via the transcriptions factor NF-kappa B while laminar shear increases activity of the atheroprotective transcription factor Nrf2. To test whether the subcellular differences in H2O2 production were causative of the differential activation of these redox-sensitive transcription factors, H2O2 was generated in the nucleus and cytosol of endothelial cells via expression of differentially-targeted DAAO constructs. Nuclear H2O2 production primarily induced the Nrf2 transcript heme-oxygenase 1 while oscillatory flow mainly activated NF-kappa B as measured by TNF-alpha induction. Finally, further applications of the DAAO technology are proposed, including its use in probing other subcellular redox signaling pathways and future applications in vivo.