Volume 11 Supplement 10

Highlights from the Sixth International Society for Computational Biology (ISCB) Student Council Symposium

Open Access

Analysis of the functional properties of the creatine kinase system using a multiscale ‘sloppy’ modeling approach

BMC Bioinformatics201011(Suppl 10):O9


Published: 07 December 2010


Distinct functions have been hypothesized for the creatine kinase (CK) enzyme catalyzing the reversible transfer of the high-energy phosphate group of ATP to creatine. In muscle cells, two CK isoforms might mediate (i) temporal energy buffering to maintain ATP homeostasis and (ii) energy transport from mitochondria to myofibrils via the “phosphocreatine shuttle” mechanism. Here we investigate the relative importance of the two roles using a computational model [1]. Model simulations predict a contribution of CK to the net transcytosolic energy transport of less than 1/3.

Materials and methods

The model is validated on experimental data from two scales: kinetic parameter measurements on the enzyme level and response times of oxidative phosphorylation in whole cardiac muscle.

To account for possible inaccuracies in the 22 model parameters, we sample a Bayesian ensemble of parameter sets which allows us to set confidence regions on model predictions. In this ‘sloppy’ modeling approach [2], the likelihood of a parameter combination to be included in the ensemble is proportional to its likelihood to predict the data. We use prior information on single parameter values from enzyme kinetic measurements to constrain them within their measurement errors.


From the ensemble, we predict a contribution of CK to the overall energy transport of 15±8% (mean±SD). The CK system damps the pulses of ATP hydrolysis of 3765 μM*s-1 during cardiac systole about 20 fold. CK inhibition by 98% increases the amplitude of ATP synthesis from 215±23 to 566±31 μM*s-1 (see Figure 1).
Figure 1

Prediction of ATP synthesis rate for full CK activity and CK inhibition by 98%. The plot shows the steady state values of the last second of a simulation over 60 seconds. Note that the plotted regions show the 95% confidence interval between upper and lower bound of the simulation of a parameter ensemble containing ~500 parameter sets. Model input is a forcing function simulating pulsatile ATP hydrolysis in the beating heart, plotted in black.


Our findings clearly support the hypothesis that CK acts as a high capacity temporal energy buffer damping energy peaks rather than being an essential energy transport system.

Authors’ Affiliations

Centre for Integrative Bioinformatics VU (IBIVU), VU University
Department of Clinical Genetics, VU University medical centre


  1. van Beek J: Adenine nucleotide-creatine-phosphate module in myocardial metabolic system explains fast phase of dynamic regulation of oxidative phosphorylation. Am J Physiol Cell Physiol 2007, 293: C815-C829. 10.1152/ajpcell.00355.2006View ArticlePubMedGoogle Scholar
  2. Gutenkunst RN, Waterfall JJ, Casey FP, et al.: Universally sloppy parameter sensitivities in systems biology models. PLoS Comput Biol 2007, 3: 1871–1878. 10.1371/journal.pcbi.0030189PubMedGoogle Scholar


© Hettling et al; licensee BioMed Central Ltd. 2010

This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.