Tailoring Mathematical Models to Stem-Cell Derived Cardiomyocyte Lines Can Improve Predictions of Drug-Induced Changes to Their Electrophysiology

Lei, Chon Lok; Wang, Ken; Clerx, Michael; Johnstone, Ross H.; Hortigón-Vinagre, María P.; Zamora, Víctor; Allan, Andrew; Smith, Godfrey L.; Gavaghan, David J.; Mirams, Gary R.; Polonchuk, Liudmila

doi:10.3389/fphys.2017.00986

Cited by 48 publications

(46 citation statements)

References 78 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, such rapid characterisation using high-throughput systems can benefit precision and personalised medicine. For example, when using human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) in personalised medicine, as described in (26), characterisation of ion current kinetics may need to be taken into account, in order to tailor accurate cell line-specific models.…”

Section: Discussionmentioning

confidence: 99%

Rapid characterisation of hERG channel kinetics I: using an automated high-throughput system

Lei

Clerx

Gavaghan

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

1 Affiliations removed for initial submission as per guidelines. 6 ABSTRACT Predicting how pharmaceuticals may affect heart rhythm is a crucial step in drug-development, and requires a 7 deep understanding of a compound's action on ion channels. In vitro hERG-channel current recordings are an important step in 8 evaluating the pro-arrhythmic potential of small molecules, and are now routinely performed using automated high-throughput 9 patch clamp platforms. These machines can execute traditional voltage clamp protocols aimed at specific gating processes, 10 but the array of protocols needed to fully characterise a current is typically too long to be applied in a single cell. Shorter 11 high-information protocols have recently been introduced which have this capability, but they are not typically compatible with 12 high-throughput platforms. We present a new high-information 15 s protocol to characterise hERG (Kv11.1) kinetics, suitable for 13 both manual and high-throughput systems. We demonstrate its use on the Nanion SyncroPatch 384PE, a 384 well automated 14 patch clamp platform, by applying it to CHO cells stably expressing hERG1a. From these recordings we construct 124 cell-specific 15 variants/parameterisations of a hERG model at 25 • C. A further 8 independent protocols are run in each cell, and are used to 16 validate the model predictions. We then combine the experimental recordings using a hierarchical Bayesian model, which we 17 use to quantify the uncertainty in the model parameters, and their variability from cell to cell, which we use to suggest reasons 18 for the variability. This study demonstrates a robust method to measure and quantify uncertainty, and shows that it is possible 19 and practical to use high-throughput systems to capture full hERG channel kinetics quantitatively and rapidly. 20 21 We present a method for high-throughput characterisation of hERG potassium channel kinetics, via fitting a mathematical 22 model to results of over one hundred single cell patch clamp measurements collected simultaneously on an automated voltage 23 clamp platform. The automated patch clamp data are used to parameterise a mathematical ion channel model fully, opening a 24 new era of automated and rapid development of mathematical models from quick and cheap experiments. The method also 25 allows ample data for independent validation of the models and enables us to study experimental variability and propose its 26 origins. In future the method can be applied to characterise changes to hERG currents in different conditions, for instance 27 at different temperatures (see Part II of the study) or under mutations or the action of pharmaceuticals; and should be easily 28 adapted to study many other currents. Statement of Significance 30The human Ether-à-go-go-Related Gene (hERG) is of great interest in the field of cardiac safety pharmacology. hERG encodes 31 the pore-forming alpha subunit of the ion channel Kv11.1 which conducts the rapid delayed rectifier potassium current, I Kr 32(1)...

show abstract

Section: Discussionmentioning

confidence: 99%

Rapid characterisation of hERG channel kinetics I: using an automated high-throughput system

Lei

Clerx

Gavaghan

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The Paci 2013 model is based on the data collected by Ma et al (20) using hiPSC-CMs. To address the variability between cells that is encountered in experiments, we identified G K1, critical for each I K1 formulation under investigation in a population of 22 cell-specific iPSC-CM models, published earlier by Lei et al (24). In brief, Lei et al (24) adapted the Paci 2013 model on the base of voltage clamp experiments carried out on hiPSC-CMs, by scaling the maximal conductance (S/F) of I Na (Â0.69), I CaL (Â0.80), I Ks , and the maximal activity of I NaCa (pA/pF) (tailored for each cell, the values are reported in Table 2).…”

Section: Comparing the Influence Of Cell-to-cell Variation Between Ipmentioning

confidence: 99%

Required G1 to Suppress Automaticity of iPSC-CMs Depends Strongly on I1 Model Structure

Fabbri

Goversen

Vos

et al. 2019

Biophysical Journal

View full text Add to dashboard Cite

Human-induced pluripotent stem cells derived cardiomyocytes (hiPSC-CMs) are a virtually endless source of human cardiomyocytes that may become a great tool for safety pharmacology; however, their electrical phenotype is immature: they show spontaneous action potentials (APs) and an unstable and depolarized resting membrane potential (RMP) because of lack of I K1 . Such immaturity hampers their application in assessing drug safety. The electronic overexpression of I K1 (e.g., through the dynamic clamp (DC) technique) is an option to overcome this deficit. In this computational study, we aim to estimate how much I K1 is needed to bring hiPSC-CMs to a stable and hyperpolarized RMP and which mathematical description of I K1 is most suitable for DC experiments. We compared five mature I K1 formulations (Bett, Dhamoon, Ishihara, O'Hara-Rudy, and ten Tusscher) with the native one (Paci), evaluating the main properties (outward peak, degree of rectification), and we quantified their effects on AP features (RMP, _ V max , APD 50 , APD 90 (AP duration at 50 and 90% of repolarization), and APD 50 /APD 90 ) after including them in the hiPSC-CM mathematical model by Paci. Then, we automatically identified the critical conductance for I K1 ( G K1, critical ), the minimally required amount of I K1 suppressing spontaneous activity. Preconditioning the cell model with depolarizing/hyperpolarizing prepulses allowed us to highlight time dependency of the I K1 formulations. Simulations showed that inclusion of mature I K1 formulations resulted in hyperpolarized RMP and higher _ V max , and observed G K1, critical and the effect on AP duration strongly depended on I K1 formulation. Finally, the Ishihara I K1 led to shorter (À16.3%) and prolonged (þ6.5%) APD 90 in response to hyperpolarizing and depolarizing prepulses, respectively, whereas other models showed negligible effects. Fine-tuning of G K1 is an important step in DC experiments. Our computational work proposes a procedure to automatically identify how much I K1 current is required to inject to stop the spontaneous activity and suggests the use of the Ishihara I K1 model to perform DC experiments in hiPSC-CMs. SIGNIFICANCE In this work, we aim to contribute a method that will facilitate automated dynamic clamp (DC) experiments in which I K1 is injected in induced pluripotent stem-cell-derived cardiomyocytes (iPSC-CMs). By introducing G K1, critical (minimal I K1 conductance needed to stop automaticity of iPSC-CMs), we are proposing a different approach to setting up DC experiments. These are usually based on the injection of a fixed current density. In contrast, G K1, critical is a parameter that depends on the cell under investigation. Our in silico approach analyzed analogies and differences between I K1 formulations without the confounding factor that can be brought by the variability of iPSC-CMs. It highlighted how much the employed mathematical formulation of I K1 can affect G K1, critical and the action potential waveform in DC experiments.

show abstract

“…However, if the majority of the variability is due to experimental artefacts, then a single set of kinetic parameters could accurately describe the physiology of each current type. If our findings also apply to myocytes, the standard approach to building action potential models with a single set of kinetic parameters for each current is appropriate Ten Tusscher et al Groenendaal et al (2015); Lei et al (2017), and the observed cell-cell variability in kinetics data is not required to be propagated forward in action potential simulations (as some studies have examined Pathmanathan et al (2015)). Nevertheless, we may still need an approach like the one demonstrated here to determine unbiased ion channel kinetic parameters to build the most physiologically-relevant action potential models; as opposed to taking the mean of biased recordings which could accidentally include some experimental artefacts within the kinetics models.…”

Section: Discussionmentioning

confidence: 99%

Accounting for variability in ion current recordings using a mathematical model of artefacts in voltage-clamp experiments

Lei

Clerx

Whittaker

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

Mathematical models of ion channels, which constitute indispensable components of action potential models, are commonly constructed by fitting to whole-cell patch-clamp data. In a previous study we fitted cell-specific models to hERG1a (Kv11.1) recordings simultaneously measured using an automated high-throughput system, and studied cell-cell variability by inspecting the resulting model parameters. However, the origin of the observed variability was not identified. Here we study the source of variability by constructing a model that describes not just ion current dynamics, but the entire voltage-clamp experiment. The experimental artefact components of the model include: series resistance, membrane and pipette capacitance, voltage offsets, imperfect compensations made by the amplifier for these phenomena, and leak current. In this model, variability in the observations can be explained by either cell properties, measurement artefacts, or both. Remarkably, by assuming that variability arises exclusively from measurement artefacts, it is possible to explain a larger amount of the observed variability than when assuming cell-specific ion current kinetics. This assumption also leads to a smaller number of model parameters. This result suggests that most of the observed variability in patch-clamp data measured under the same conditions is caused by experimental artefacts, and hence can be compensated for in post-processing by using our model for the patch-clamp experiment. This study has implications for the question of the extent to which cell-cell variability in ion channel kinetics exists, and opens up routes for better correction of artefacts in patch-clamp data.

show abstract

Tailoring Mathematical Models to Stem-Cell Derived Cardiomyocyte Lines Can Improve Predictions of Drug-Induced Changes to Their Electrophysiology

Cited by 48 publications

References 78 publications

Rapid characterisation of hERG channel kinetics I: using an automated high-throughput system

Rapid characterisation of hERG channel kinetics I: using an automated high-throughput system

Required G1 to Suppress Automaticity of iPSC-CMs Depends Strongly on I1 Model Structure

Accounting for variability in ion current recordings using a mathematical model of artefacts in voltage-clamp experiments

Contact Info

Product

Resources

About