Modification of DiFrancesco-Noble equations to simulate the effects of vagal stimulation on in vivo mammalian sinoatrial node electrical activity

We present a new mathematical model for vagal control of rabbit sinoatrial (SA) node electrical activity based on the DiFrancesco-Noble equations. The original equations were found to be unstable, resulting in progressive cycle by cycle depletion or accumulation of ions in intra- and extracellular compartments. This problem was overcome by modifying the maximum Na-K pump current and the time constant for uptake of intracellular calcium. We also included a formulation for the acetylcholine (ACh)-activated potassium current which was consistent with experimental data. This formulation was based on kinetics first proposed by Osterrieder and later modified by Yanagihara. The resulting model exhibits cycle-cycle ionic stability, and includes an ACh-activated potassium current which accurately reproduces experimentally observed effects of vagal stimulation on both the membrane potential and its time-derivative. Simulations were performed for both brief-burst and prolonged vagal stimulation using simplified square wave profiles for the concentration of ACh in the synaptic cleft space. This protocol permits the isolation of cardiac period dynamics caused by changes in membrane potential and intra- and extracellular ionic concentrations from those caused by other mechanisms including the dynamics of ACh release, diffusion, hydrolysis and washout. Simulation results for the effects of brief-burst single cycle stimulation on the cardiac period agree closely with experimental data reported in the literature, accurately reproducing changes in membrane potential and the phasic dependency of the response to the position of vagal stimulus bursts within the cycle. Simulation of the effects of prolonged vagal stimulation accurately reproduced the steady-state characteristics of heart period response, but did not yield the complex multimodal dynamics of the recovery phase, or the pronounced post vagal tachycardia observed experimentally at the termination of the stimulus. Our results show that the major chronotropic effects of vagal stimulation on the SA cell membrane can be explained in terms of the ACh-activated potassium current. The effects of this membrane current however are generally fast acting and cannot contribute to any long lasting dynamics of the cardiac period response. The modified DiFrancesco-Noble model presented in this article provides a valuable theoretical tool for further analysis of the dynamics of vagal control of the cardiac pacemaker.

Dynamics of cardiac period responses after prolonged vagal stimulation in the dog

In 12 dogs anesthetised with alpha-chloralose and urethane, and beta-adrenergic blocked with propranolol, the cervical vagi were stimulated for 60 seconds with supramaximal constant current pulses at frequencies between 2.0 and 10 Hz. The time course of the recovery of the cardiac period response, after cessation of vagal stimulation, was analyzed using nonlinear curve fitting techniques. It was found that the recovery phase could be reliably fitted with a function consisting of the sum of up to three exponential terms. The first term has a moderate rate constant of 0.2260 +/- 0.0112 S-1 (SE) and is independent of vagal stimulus frequency and the preceding bradycardia. We propose that this term is associated with a recovery from the bradycardia with a rate constant indicative of the reduction of acetylcholine at the pacemaker caused by hydrolysis and washout. The second term has a small rate constant of 0.0499 +/- 0.0014 s-1 (SE) and a negative gain. This term describes the time course of the post vagal tachycardia observed at cessation of stimulation. It is also independent of stimulus frequency and the preceding bradycardia. The third term has a large and variable rate constant (range: 0.247-8.01 s-1) and becomes increasingly dominant when the preceding bradycardia is large. We propose that this third component arises from a rapid return of the pacemaker focus to the dominant location prior to vagal stimulation. The mathematical characterisation of this component is important to permit the accurate derivation of the time courses of the remaining two components. Curvilinear relationships exist between the gain parameters for each of the exponential terms and the preceding bradycardia.