Abstract Title Hodgkin-Huxley Algorithm to Diagnose Channelopathies from Empirical Action Potentials

Background Human channelopathies are diagnosed by symptom profile or genetic testing which fail to provide a mechanistic basis for patients’ experiences. This project aims to develop an algorithm that decomposes a patient’s electrophysiology data into clinically relevant values that would better focus treatment efforts. Our model system for variation in action potential pathology are garter snakes (Thamnophis spp.) that have evolved resistance to their neurotoxic prey (Taricha spp.). The toxin, tetrodotoxin, occludes the pore of voltage-gated sodium channels that conduct sensory and motor impulses. However, resistance mutations observed in snake skeletal muscle voltage-gated sodium channels significantly reduce channel function by limiting sodium conductance. Our model aims to reproduce that level of mechanistic detail from action potentials recorded in vivo directly in the target tissue rather than by complicated whole-cell electrophysiology techniques.

Methods To observe sodium channel conductivity deficits, we collected action potentials from known healthy and pathological samples and then trained a model based on this dataset. We recorded from muscles carrying tetrodotoxin-sensitive sodium channels (NaV1.4⁺, N=9) from Thamnophis atratus as well as from snakes with mildly resistant channels (A1281P, N=11) suffering mild deficits. We then tailored the Hodgkin-Huxley model of the action potentials to fit skeletal muscle recordings. We deployed the model to decompose action potentials into distinct, molecularly-based mathematical parameters, including sodium channel conductance (GNa), sodium channel activation (m), and inactivation (h). We generated representative action potentials based on averages of each group. Then, R was used to fit a parameter set to each group to assess differences in molecularly-based parameters between groups (e.g. GNa).

Results The model reproduced the expected negative correlation between sodium channel mutations and channel conductance (e.g. reduced GNa). Furthermore, it revealed significant decreases in two kinetic subparameters, m and h, providing an additional explanation of reduced conductance. These data validate our model against peer-reviewed research (del Carlo et al., 2024)

Conclusion This model offers the first estimate of sodium channel function in native snake skeletal muscle. Our model produced results congruent with findings of peer-reviewed reports using gold-standard methods. However, the model currently lacks sufficient specificity to evaluate patient-level data. Ongoing work aims to expand this model as a diagnostic tool by training it on even more extreme snake channelopathies. With further refinement, this model could be a clinically useful tool for the differential diagnosis of human channelopathies, advancing personalized medicine for inherited disorders.