Automated High-Throughput Patch Clamp and Modelling to Capture hERG Kinetics and Temperature Dependence using Optimised Voltage Protocols

Abstract

Automated platforms for assessing ion channel functionality are now widely used to allow high-throughput electrophysiology measurements. Mathematical models of ion channels, which constitute the core of whole-cell electrophysiology models, are typically constructed through fitting to patch-clamp data. The patch-clamp data is usually generated using long, dedicated protocols for studying particular gating processes, which are usually unable to characterise all the kinetics of interest in a single cell. We present a new 15-second protocol to characterise hERG1a (Kv11.1) kinetics, suitable for both manual and high-throughput systems. We demonstrate its use on the Nanion SyncroPatch 384PE, a 384-well automated patch-clamp platform. An open data interface provided by Nanion allows for export of trace data and online analysis data as .json files, which we used for further processing. We are able to construct up to 124 cell-specific variants/parameterisations of a hERG model using data from CHO cells stably transfected with hERG1a at five distinct temperatures between 25 and 37°C. We have validated all our cell-specific models against 8 other voltage-clamp protocols run in the same cells (Lei et al. 2019, doi:10.1016/j.bpj.2019.07.029). The experimental recordings are combined using a hierarchical Bayesian model to quantify the uncertainty in the model parameters and their variability from well-to-well; we use this hierarchical model to suggest reasons for the variability. Our models also reveal that activation is far more temperature sensitive than inactivation, and we observe that the temperature dependency of the kinetic parameters is not represented well by Q10 coefficients; it broadly follows a generalised, but not the standardly-used, Eyring relationship (Lei et al. 2019, doi:10.1016/j.bpj.2019.07.030).