Abstract
Non-excitable cells (NECs) such as cardiac myofibroblasts that are electrotonically coupled to cardiomyocytes affect conduction velocity (θ) by representing a capacitive load (CL: increased membrane to be charged) and a resistive load (RL: partial depolarization of coupled cardiomyocytes). In this study, we untangled the relative contributions of both loading modalities to NEC-dependent arrhythmogenic conduction slowing. Discrimination between CL and RL was achieved by reversibly removing the RL component by light activation of the halorhodopsin-based hyperpolarizing membrane voltage actuator eNpHR3.0-eYFP (enhanced yellow fluorescent protein) expressed in communication-competent fibroblast-like NIH3T3 cells (3T3 (HR) cells) that served as a model of coupled NECs. Experiments were conducted with strands of neonatal rat ventricular cardiomyocytes coated at increasing densities with 3T3 (HR) cells. Impulse conduction along preparations stimulated at 2.5 Hz was assessed with multielectrode arrays. The relative density of 3T3 (HR) cells was determined by dividing the area showing eYFP fluorescence by the area covered with cardiomyocytes [coverage factor (CF)]. Compared to cardiomyocytes, 3T3 (HR) cells exhibited a depolarized membrane potential (-34 mV) that was shifted to -104 mV during activation of halorhodopsin. Without illumination, 3T3 (HR) cells slowed θ along the preparations from ∼330 mm/s (control cardiomyocyte strands) to ∼100 mm/s (CF = ∼0.6). Illumination of the preparation increased the electrogram amplitudes and induced partial recovery of θ at CF > 0.3. Computer simulations demonstrated that the θ deficit observed during illumination was attributable in full to the CL represented by coupled 3T3 (HR) cells with θ showing a power-law relationship to capacitance with an exponent of -0.78 (simulations) and -0.99 (experiments). The relative contribution of CL and RL to conduction slowing changed as a function of CF with CL dominating at CF ≤ ∼0.3, both mechanisms being equally important at CF = ∼0.5, and RL dominating over CL at CF > 0.5. The finding that RL did not affect θ at CFs ≤ 0.3 is explained by the circumstance that, at the respective moderate levels of cardiomyocyte depolarization, supernormal conduction stabilized propagation. The findings provide experimental estimates for the dependence of θ on membrane capacitance in general and suggest that the myocardium can absorb moderate numbers of electrotonically coupled NECs without showing substantial alterations of θ.