A- and C-type rat nodose sensory neurons: Model interpretations of dynamic discharge characteristics

1. Neurons of the nodose ganglia provide the sole connection between many types of visceral sensory inputs and the central nervous system. Electrophysiological studies of isolated nodose neurons provide a practical means of measuring individual cell membrane currents and assessing their putative contributions to the overall response properties of the neuron and its terminations. Here, we present a comprehensive mathematical model of an isolated nodose sensory neuron that is based upon numerical fits to quantitative voltage- and current-clamp data recorded in our laboratory. Model development was accomplished using an iterative process of electrophysiological recordings, nonlinear parameter estimation, and computer simulation. This work is part of an integrative effort aimed at identifying and characterizing the fundamental ionic mechanisms participating in the afferent neuronal limb of the baroreceptor reflex. 2. The neuronal model consists of two parts: a Hodgkin-Huxley-type membrane model coupled to a lumped fluid compartment model that describes Ca²⁺ ion concentration dynamics within the intracellular and external perineuronal media. Calcium buffering via a calmodulin-type buffer is provided within the intracellular compartment. 3. The complete model accurately reproduces whole-cell voltage- clamp recordings of the major ion channel currents observed in enzymatically dispersed nodose sensory neurons. Specifically, two Na⁺ currents exhibiting fast (I(Na(f))) and slow tetrodotoxin (TTX)-insensitive (I(Na(s))) kinetics; low- and high-threshold Ca²⁺ currents exhibiting transient (I(Ca,t)) and long-lasting (I(Ca,n)) dynamics, respectively; and outward K⁺ currents consisting of a delayed-rectifier current (I(K)), a transient outward current (I(t)) and a Ca²⁺-activated K⁺ current (I(K,Ca)). 4. Whole-cell current- clamp recordings of somatic action-potential dynamics were performed on enzymatically dispersed nodose neurons using the perforated patch-clamp technique. Stimulus protocols consisted of both short (≤2.0 ms) and long (≥200 ms) duration current pulses over a wide range of membrane holding potentials. These studies clearly revealed two populations of nodose neurons, often termed A- and C-type cells, which exhibit markedly different action- potential signatures and stimulus response properties. 5. Using a single set of equations, the model accurately reproduces the electrical behavior of both A- and C-type nodose neurons in response to a wide variety of stimulus conditions and membrane holding potentials. The structure of the model, as well as the majority of its parameters are the same for both A- and C-type implementations. Our modeling results suggest that the rather disparate stimulus response properties and action potential signatures exhibited by these two populations of nodose neurons can be attributed to the relative degree of expression of individual ionic membrane conductances. 6. The model predicts a wide range of experimentally derived data from both A- and C-type nodose neurons, including the following: 1) action potentials, which were resistant to the Na⁺ channel blocker TTX, 2) changes in action-potential waveshape with manipulations of external Na⁺ and Ca²⁺ ion concentrations and, 3) transitions in dynamic activity through manipulations in the relative strength of I(t) by the K⁺ channel blocker 4-aminopyridine. 7. Phase-plane analysis applied to both experimental and model-generated action-potential data has been utilized to decompose the total membrane current into the major ionic currents that participate during different phases of A- and C-type somatic action potentials. This technique provides a sensitive measure of the model's ability to mimic the more rapid phases of the recorded action potentials and provides a quantitative basis for explanations regarding the differences in action-potential waveforms. In particular, this analysis reveals a new appreciation for the contributions of I(Ca,n) and I(Na(s)) in the formation of the characteristic 'hump' most often associated with C-type neurons.