Mechanical compression elicits NO-dependent increases in coronary flow

Abstract

Our previous studies have demonstrated that a decrease in arteriolar diameter that causes endothelial deformation elicits the release of nitric oxide (NO). Thus we hypothesized that cardiac contraction, via deformation of coronary vessels, elicits the release of NO and increases in coronary flow. Coronary flow was measured at a constant perfusion pressure of 80 mmHg in Langendorff preparations of rat hearts. Hearts were placed in a sealed chamber surrounded with perfusion solution. The chamber pressure could be increased from 0 to 80 mmHg to generate extracardiac compression. To minimize the impact of metabolic vasodilatation and rhythmic changes in shear stress, nonbeating hearts, by perfusing the hearts with a solution containing 20 mM KCl, were used. After extracardiac compression for 10 or 20 s, coronary flow increased significantly, concurrent with an increased release of nitrite into the coronary effluent and increased phosphorylation of endothelial NO synthase in the hearts. Inhibition of NO synthesis eliminated the compression-induced increases in coronary flow. Shear stress-induced dilation could not account for this increased coronary flow. Furthermore, in isolated coronary arterioles, without intraluminal flow, the release of vascular compression elicited a NO-dependent dilation. Thus this study reveals a new mechanism that, via coronary vascular deformation, elicited by cardiac contraction, stimulates the endothelium to release NO, leading to increased coronary perfusion.

Fig. 1

Perfusion system and compression chamber for isolated rat hearts.

Fig. 2

Representative recordings of coronary flow and extracardiac pressure in an isolated and perfused rat heart. A: changes in pulsatile and mean coronary flow (PF and MF, respectively) after the physiological salt solution (PSS) was exchanged with PSS containing a total of 20 mM KCl (PSS-KCl). Arrow indicates the start of perfusion with PSS-KCl. B: changes in mean coronary flow in response to occlusion of coronary perfusion via turning off the inflow stopcock and increasing extracardiac pressure (EP) to 80 mmHg for 10 s in control and after inhibition of nitric oxide (NO) synthesis with N^ω-nitro-l-arginine methyl ester (l-NAME; 2 × 10⁻⁴ M).

Fig. 3

Increase in integrated flow in response to increases in EP and stoppage of coronary perfusion for 10 and 20 s in control and after inhibition of NO synthesis with l-NAME (2 × 10⁻⁴M). Increase in integrated flow, as illustrated by the crosshatched areas in Fig. 2B, was calculated as the difference between the area (flow rate × duration) under a specific response curve and a control area (control flow rate × duration of the response). Data are means ± SE (n = 8–14).

Fig. 4

Increase in integrated flow in response to 10 s of inflow and outflow occlusions and extracardiac compression (0–80 mmHg). A: representative recordings. B: summary data of 5 experiments. Data are means ± SE.

Fig. 5

Release of NO, measured as nitrite in the coronary effluent, in beating hearts and in nonbeating hearts before and after 20 s of EP (80 mmHg). Coronary effluent was collected for 1 min in each experimental condition. Average release of nitrite (in nmol/min) was calculated by the total volume collected during 1 min. Each bar represents the average of nine measurements (means ± SE).

Fig. 6

A: representative Western blots of phospho-endothelial NO synthase (p-eNOS) and eNOS. B: densitometric ratio of p-eNOS and eNOS in beating (I), nonbeating (II), and nonbeating hearts with extracardiac compression (III). Data are means ± SE; n = 3–5. NS, no statistical difference.

Fig. 7

Vasodilatation of isolated rat coronary arterioles induced by release of extravascular pressure (EVP) in control and after administration of l-NAME (2 × 10⁻⁴ M). EVP was increased from 0 to 75 mmHg for 20 and 60 s. Post-EVP dilation is presented as an area that was calculated as the difference between a product of diameter (in μm) and duration (in s) under a specific response curve and a product of the control diameter and the duration of the same response curve. Each bar represents an average of 6 measurements (means ± SE).