Capillary Oxygen Triggers Conducted Vasodilation via Connexin 40 Independent of Extracellular K + and Endothelial K IR 2.1

Abstract

Mechanisms that fine-tune red blood cell (RBC) and O2 delivery to capillaries are critical to tissue function. Coupling RBC supply to demand is an intricate process requiring O2 sensing, generation of a capillary stimulus and triggering of a transduction process that alters the diameter of upstream arterioles. The sensor's identity, the stimulus it generates and the mechanisms that enable it are actively debated. Studies have implied the endothelial KIR2.1 channels drive O2 sensing whereas gap junctions, comprised of connexin 40, enable electrical signals to conduct upstream. This idea was tested in KIR2.1-/- and connexin 40-/- mice where the local O2 environment of skeletal muscle was precisely controlled. The extensor digitorum longus muscle was prepared for intravital microscopy; a custom stage insert with a gas chamber allowed for O2 control at the tissue surface. Second-by-second capillary RBC flow responses were recorded as O2 was: 1) reduced from 53 mmHg to 15 mmHg or 0 mmHg for 3 min; or 2) oscillated between 91 mmHg and 15 mmHg (1 cycle/min). Chamber O2 at 15 mmHg induces microvascular responses without diminishing mitochondrial function while 0 mmHg initiates additional metabolic responses. Dropping PO2 on the muscle surface (53 mmHg to 15 mmHg or 0 mmHg) significantly increased RBC supply rate in capillaries of control animals while elevated chamber O2 decreased capillary RBC supply rate in a graded manner. The RBC flow responses in control mice were rapid and tightly coupled to O2 levels as the chamber O2 was oscillated in a sinusoidal fashion. In stark contrast, this blood flow response failed to occur in connexin 40-/- mice. As this response was absent, capillary RBC O2 saturation was lower under resting conditions in the connexin 40-/- animals. Endothelial KIR2.1-/- mice, on the other hand, had normal resting RBC O2 saturation. Furthermore, KIR2.1-/- mice reacted normally to O2 changes, albeit an oscillation or a sustained O2 decrease (to 15 mmHg), with a rise in RBC supply rate. Likewise, the microcirculatory response to 0 mmHg O2 was similar between floxed control and endothelial KIR2.1-/- mice. We show that microvascular O2 responses depend on coordinated electrical signaling via gap junctions comprised of connexin 40 and that endothelial KIR2.1channels do not drive the initiating electrical event. These findings reform our understanding of blood flow regulation and how O2 initiates this process independent of metabolite production.