Hyperinsulinemic Microvascular Blood Flow Response in Skeletal Muscle is Oxygen Mediated

Objective To determine if the blood flow response to insulin is partially mediated by local metabolic demand reflecting increased oxygen consumption required to fix glucose. Methods Two separate protocols were used to address the objective. 8 Sprague‐Dawley (171–204g) rats, 4 per protocol, were anesthetized with sodium pentobarbital via intraperitoneal injection, and mechanically ventilated. Arterial and venous catheters were introduced to maintain and monitor the animals’ cardiovascular state. The extensor digitorum longus (EDL) muscle was isolated and reflected onto the stage of an Olympus IX73 inverted microscope. Intravital video microscopy sequences were recorded during baseline and at hyperinsulinemic euglycemia in both protocols. Euglycemia was achieved through intravenous infusion of 2U insulin/kg/hr and variable rates of 50% D‐glucose were titrated against repeated blood glucose samples until stable normoglycemic levels were reached. In protocol 1, the EDL was reflected over a glass stage insert and isolated from room air with polyvinylidene film and a glass coverslip. In protocol 2, the muscle was instead reflected over a 100μm thick semi‐permeable membrane interfacing the EDL with a gas exchange chamber and similarly isolated from room air. An oxygen (O 2 ) challenge was imposed in protocol 2 where O 2 levels were oscillated at one minute intervals from 7%‐12%‐2%‐7%, with a stable 5% carbon dioxide concentration and the balance of gas composed of nitrogen. For both protocols, offline analysis of microvascular flow was conducted using custom MATLAB software yielding frame‐by‐frame measurements of capillary red blood cell (RBC) velocity, hematocrit, RBC supply rate (SR), and oxygen saturation (SO 2 ). Animal protocols were approved by Memorial University’s Institutional Animal Care Committee. Results No significant changes were found between glucose infusion rates (GIR) at euglycemia between both protocols, as well as blood glucose levels at both baseline and euglycemic conditions between experiments. In protocol 1, SR increased to 11.2 cells/s following euglycemic clamp, compared to 8.0 cells/s at baseline (p = 0.0255) consistent with flow increases reported in the literature. As expected, the imposed O 2 challenges in protocol 2 caused significant changes in SO 2 at 12% and 2% concentrations, which were found to be the same under both baseline and euglycemic conditions. In protocol 2, SR decreased at 12% O 2 and increased at 2% O 2 during the oscillation, compared to 7% O 2 , under both baseline and euglycemic conditions. SR was the same under euglycemia compared to baseline in the last 15 seconds of each step of the O 2 oscillation from 7%‐12%‐2%‐7%, with p values >0.9999, >0.9999, 0.9984, and 0.9963 respectively. Conclusions Gas exchange chamber was effective at manipulating tissue oxygen levels, observed through profound SO 2 changes during chamber oscillations at both baseline and euglycemic conditions. Our results suggest that the increase in muscle blood flow observed in response to insulin, as seen in experiment 1, is eliminated if the tissue oxygen environment is fixed at a given oxygen concentration as demonstrated in experiment 2. Support or Funding Information Project funded through a CIHR project grant awarded to GM Fraser (PJT‐162119)Time series showing the change in capillary red blood cell oxygen saturation over the imposed oxygen oscillations under baseline and hyperinsulinemic euglycemic conditions (N=4 animals, 52 capillaries).Time series of capillary red blood cell supply rate response to imposed oxygen oscillations under baseline and hyperinsulinemic euglycemic conditions (N=4 animals, 135 capillaries).