Markov models for ion channels: versatility versus identifiability and speed

Markov models (MMs) represent a generalization of Hodgkin–Huxley models. They provide a versatile structure for modelling single channel data, gating currents, state-dependent drug interaction data, exchanger and pump dynamics, etc. This paper uses examples from cardiac electrophysiology to discuss aspects related to parameter estimation. (i) Parameter unidentifiability (found in 9 out of 13 of the considered models) results in an inability to determine the correct layout of a model, contradicting the idea that model structure and parameters provide insights into underlying molecular processes. (ii) The information content of experimental voltage step clamp data is discussed, and a short but sufficient protocol for parameter estimation is presented. (iii) MMs have been associated with high computational cost (owing to their large number of state variables), presenting an obstacle for multicellular whole organ simulations as well as parameter estimation. It is shown that the stiffness of models increases computation time more than the number of states. (iv) Algorithms and software programs are provided for steady-state analysis, analytical solutions for voltage steps and numerical derivation of parameter identifiability. The results provide a new standard for ion channel modelling to further the automation of model development, the validation process and the predictive power of these models.

[1]  Mahmut Ozer,et al.  A comparative analysis of linear, nonlinear and improved nonlinear thermodynamic models of voltage-dependent ion channel kinetics , 2007 .

[2]  F Bezanilla,et al.  Inactivation of the sodium channel. II. Gating current experiments , 1977, The Journal of general physiology.

[3]  F Bezanilla,et al.  A sodium channel gating model based on single channel, macroscopic ionic, and gating currents in the squid giant axon. , 1991, Biophysical journal.

[4]  H. Strauss,et al.  The calcium-independent transient outward potassium current in isolated ferret right ventricular myocytes. II. Closed state reverse use- dependent block by 4-aminopyridine , 1993, The Journal of general physiology.

[5]  Allan R. Willms,et al.  NEUROFIT: software for fitting Hodgkin–Huxley models to voltage-clamp data , 2002, Journal of Neuroscience Methods.

[6]  P. Kienker.  Equivalence of aggregated Markov models of ion-channel gating , 1989, Proceedings of the Royal Society of London. B. Biological Sciences.

[7]  D. Beuckelmann,et al.  Simulation study of cellular electric properties in heart failure. , 1998, Circulation research.

[8]  Y Rudy,et al.  Cellular consequences of HERG mutations in the long QT syndrome: precursors to sudden cardiac death. , 2001, Cardiovascular research.

[9]  H. Strauss,et al.  A quantitative analysis of the activation and inactivation kinetics of HERG expressed in Xenopus oocytes , 1997, The Journal of physiology.

[10]  Yoram Rudy,et al.  Kinetic properties of the cardiac L-type Ca2+ channel and its role in myocyte electrophysiology: a theoretical investigation. , 2007, Biophysical journal.

[11]  Yoram Rudy,et al.  Subunit Interaction Determines IKs Participation in Cardiac Repolarization and Repolarization Reserve , 2005, Circulation.

[12]  Harvey Thomas Banks,et al.  Standard errors and confidence intervals in inverse problems: sensitivity and associated pitfalls , 2007 .

[13]  A. Hodgkin,et al.  A quantitative description of membrane current and its application to conduction and excitation in nerve , 1990 .

[14]  Stefano Severi,et al.  Computer simulation of wild-type and mutant human cardiac Na+ current , 2006, Medical and Biological Engineering and Computing.

[15]  K. Magleby,et al.  Ion channels. Preventing artifacts and reducing errors in single-channel analysis. , 1992, Methods in enzymology.

[16]  R. Rasmusson,et al.  Hodgkin-Huxley and partially coupled inactivation models yield different voltage dependence of block. , 1997, The American journal of physiology.

[17]  Peter J Hunter,et al.  Modeling total heart function. , 2003, Annual review of biomedical engineering.

[18]  R Horn,et al.  Statistical properties of single sodium channels , 1984, The Journal of general physiology.

[19]  David Colquhoun,et al.  The Relation between Classical and Cooperative Models for Drug Action , 1973 .

[20]  K. Summers,et al.  Molecular genetics of long QT syndrome. , 2010, Molecular genetics and metabolism.

[21]  Jonna R. Terkildsen,et al.  The balance between inactivation and activation of the Na+-K+ pump underlies the triphasic accumulation of extracellular K+ during myocardial ischemia. , 2007, American journal of physiology. Heart and circulatory physiology.

[22]  C F Starmer,et al.  Mechanisms of use-dependent block of sodium channels in excitable membranes by local anesthetics. , 1984, Biophysical journal.

[23]  N P Smith,et al.  Development of models of active ion transport for whole-cell modelling: cardiac sodium-potassium pump as a case study. , 2004, Progress in biophysics and molecular biology.

[24]  R. Winslow,et al.  Role of the Calcium-Independent Transient Outward Current Ito1 in Shaping Action Potential Morphology and Duration , 2000, Circulation research.

[25]  R L Winslow,et al.  Molecular Interactions Between Two Long-QT Syndrome Gene Products, HERG and KCNE2, Rationalized by In Vitro and In Silico Analysis , 2001, Circulation research.

[26]  Bruce Smaill,et al.  Hodgkin-Huxley type ion channel characterization: an improved method of voltage clamp experiment parameter estimation. , 2006, Journal of theoretical biology.

[27]  Frederick Sachs,et al.  Maximum likelihood estimation of ion channel kinetics from macroscopic currents. , 2005, Biophysical journal.

[28]  Ronald Wilders,et al.  UvA-DARE ( Digital Academic Repository ) HERG channel ( dys ) function revealed by dynamic action potential clamp technique , 2004 .

[29]  A. Tanskanen,et al.  A simplified local control model of calcium-induced calcium release in cardiac ventricular myocytes. , 2004, Biophysical journal.

[30]  R. Winslow,et al.  Mechanisms of altered excitation-contraction coupling in canine tachycardia-induced heart failure, II: model studies. , 1999, Circulation research.

[31]  T R Chay.  The Hodgkin–Huxley Na+ channel model versus the five‐state Markovian model , 1991, Biopolymers.

[32]  Terrence J. Sejnowski,et al.  Synthesis of models for excitable membranes, synaptic transmission and neuromodulation using a common kinetic formalism , 1994, Journal of Computational Neuroscience.

[33]  H. Strauss,et al.  The calcium-independent transient outward potassium current in isolated ferret right ventricular myocytes. I. Basic characterization and kinetic analysis , 1993, The Journal of general physiology.

[34]  Alain Destexhe,et al.  Nonlinear Thermodynamic Models of Voltage-Dependent Currents , 2000, Journal of Computational Neuroscience.

[35]  D. Noble,et al.  A model for human ventricular tissue. , 2004, American journal of physiology. Heart and circulatory physiology.

[36]  Colleen E. Clancy,et al.  Na+ Channel Mutation That Causes Both Brugada and Long-QT Syndrome Phenotypes: A Simulation Study of Mechanism , 2002, Circulation.

[37]  Martin Fink,et al.  A Respiratory System Model: Parameter Estimation and Sensitivity Analysis , 2008, Cardiovascular engineering.

[38]  Denis Noble,et al.  Contributions of HERG K+ current to repolarization of the human ventricular action potential. , 2008, Progress in biophysics and molecular biology.

[39]  C. Cobelli,et al.  Parameter and structural identifiability concepts and ambiguities: a critical review and analysis. , 1980, The American journal of physiology.

[40]  Claudio Cobelli,et al.  Generalized Sensitivity Functions in Physiological System Identification , 1999, Annals of Biomedical Engineering.

[41]  James Demmel,et al.  The generalized Schur decomposition of an arbitrary pencil A–λB—robust software with error bounds and applications. Part I: theory and algorithms , 1993, TOMS.

[42]  J. Reid,et al.  Structural identifiability in linear time invariant systems , 1976, 1976 IEEE Conference on Decision and Control including the 15th Symposium on Adaptive Processes.

[43]  D. Kass,et al.  Mechanisms of altered excitation-contraction coupling in canine tachycardia-induced heart failure, I: experimental studies. , 1999, Circulation research.

[44]  Francisco Bezanilla,et al.  Inactivation of the Sodium Channel , 2003 .

[45]  R J Bauer,et al.  Theory of the kinetic analysis of patch-clamp data. , 1987, Biophysical journal.

[46]  Yoram Rudy,et al.  Computational biology in the study of cardiac ion channels and cell electrophysiology , 2006, Quarterly Reviews of Biophysics.

[47]  Anthony Varghese,et al.  Computing Transient Gating Charge Movement of Voltage-Dependent Ion Channels , 2002, Journal of Computational Neuroscience.

[48]  D. Doyle,et al.  Structural changes during ion channel gating , 2004, Trends in Neurosciences.

[49]  J. Towbin,et al.  Molecular genetics of long QT syndrome from genes to patients. , 1997, Current opinion in cardiology.

[50]  C. Armstrong.  Interaction of Tetraethylammonium Ion Derivatives with the Potassium Channels of Giant Axons , 1971, The Journal of general physiology.

[51]  R L Winslow,et al.  Cardiac sodium channel Markov model with temperature dependence and recovery from inactivation. , 1999, Biophysical journal.

[52]  Y. Jan,et al.  Molecular basis of K+ channel inactivation gating. , 1993, EXS.