Research Article

Structural Analysis and Stochastic Modelling Suggest a Mechanism for Calmodulin Trapping by CaMKII

  • Melanie I. Stefan,

    Affiliation: EMBL-European Bioinformatics Institute, Hinxton, UK

    Current address: Biology Division, California Institute of Technology, Pasadena, California, United States of America

  • David P. Marshall,

    Affiliation: EMBL-European Bioinformatics Institute, Hinxton, UK

    Current address: Department of Biochemistry, University of Oxford, UK

  • Nicolas Le Novère mail

    Affiliation: EMBL-European Bioinformatics Institute, Hinxton, UK

  • Published: January 18, 2012
  • DOI: 10.1371/journal.pone.0029406

Reader Comments (1)

Post a new comment on this article

Authors' summary

Posted by mstefan on 07 Mar 2012 at 21:59 GMT

Calcium/calmodulin-dependent kinase II (CaMKII) is a neuronal protein involved in learning and memory. Upon binding to its activator calmodulin, it can undergo phosphorylation, which has two main effects: First, it maintains CaMKII activity, thus rendering it calmodulin-independent. Second, it decreases calmodulin dissociation rates, thereby ``trapping calmodulin in the complex. The exact mechanism by which calmodulin trapping is achieved is so far poorly understood. In this paper, we combine stuctural modelling using the crystal structures of CaMKII and calmodulin with stochastic simulations of interactions between the two molecules. We argue that a single CaMKII unit has not one, but two binding sites for calmodulin, of which oine (the stronger one) becomes more easily accessible if CaMKII is phosphorylated. This hypothesis is consistent with existing biochemical data and with the available protein structures. A computer simulation based on this hypothesis can, indeed, reproduce calmodulin trapping by CaMKII and is the first computational model to do so.

No competing interests declared.