Elsevier

Cellular Signalling

Volume 22, Issue 5, May 2010, Pages 759-769
Cellular Signalling

Regulation of CaMKII by phospho-Thr253 or phospho-Thr286 sensitive targeting alters cellular function

https://doi.org/10.1016/j.cellsig.2009.12.011Get rights and content

Abstract

Calcium/calmodulin-stimulated protein kinase II (CaMKII) is an important mediator of synaptic function that is regulated by multi-site phosphorylation and targeting through interactions with proteins. A new phosphorylation site at Thr253 has been identified in vivo, that does not alter CaMKII activity, but does alter CaMKII function through interactions with binding proteins. To identify these proteins, as well as to examine the specific effects following Thr253 or Thr286 phosphorylation on these interactions, we developed an in vitro overlay binding assay. We demonstrated that the interaction between CaMKII and its binding proteins was altered by the phosphorylation state of both the CaMKII and the partner, and identified a CaMKII-specific sequence that was responsible for the interaction between CaMKII and two interacting proteins. By comparing CaMKII binding profiles in tissue and cell extracts, we demonstrated that the CaMKII binding profiles varied with cell type, and also showed that overexpression of a CaMKII Thr253 phospho-mimic mutant in human neuroblastoma and breast cancer cells dramatically altered the morphology and growth rates when compared to overexpression of non-phosphorylated CaMKII. This data highlights the importance of the microenvironment in regulating CaMKII function, and describes a potentially new mechanism by which the functions of CaMKII can be regulated.

Introduction

Calcium/calmodulin-stimulated protein kinase II (CaMKII) is a family of multifunctional serine (Ser)/threonine (Thr) protein kinases encoded by four genes (α, β, γ, and ∂) [1]. One or more members of this family are found in virtually every tissue, and mediate a diverse range of physiological responses. CaMKII is expressed most abundantly in neurons, and is involved in regulating many aspects of neuronal function, including neurotransmitter synthesis and release, cellular morphology and neurite extension, long-term plasticity, and learning [2], [3], [4]. Non-neuronal CaMKII has been implicated in the regulation of other biological processes, such as fertilisation [5], osteogenic differentiation [6], and the maintenance of vascular tone [7].

The biological properties of CaMKII are regulated by multi-site phosphorylation and targeting to specific subcellular locations through interactions with specific target proteins [8]. The roles of two phosphorylation sites, namely Thr286 and Thr305/6, have been well characterised. Phosphorylation at either of these sites has a profound impact on the activity of CaMKII, as well as targeting to various subcellular locations [9], [10]. Recently, the consequences of phosphorylation at Thr253 have begun to be characterised. This phosphorylation site, which is present in every isoform and conserved across species, has been shown to be phosphorylated in vivo [11], [12]. Unlike phosphorylation at Thr286 or Thr305/306, phosphorylation at Thr253 has no direct effect on the activity of CaMKII in vitro, however it has marked effects on CaMKII targeting [11]. In addition, phosphorylation of Thr253 occurs only in a small proportion of cellular CaMKII, much of which is associated with the post-synaptic density (PSD) [11], [12]. This suggests that the functional consequences of Thr253 phosphorylation may be concentrated at this and other specialised cellular locations.

It has been well established that the appropriate targeting of signalling molecules plays an important role in establishing the cellular responses to extracellular stimuli [13], [14]. Changes in the autophosphorylation of CaMKII can regulate translocation and CaMKII binding to specific subcellular locations, such as the PSD. Phosphorylation of Thr305/6 decreases the amount of CaMKII bound to the PSD, stimulating translocation from the PSD to the cytosol [15]. In contrast, phosphorylation of either Thr286 or Thr253 enhances binding to the PSD, with phosphorylation at both Thr286 and Thr253 exhibiting at least an additive increase in binding [11]. Once located in the PSD, CaMKII can phosphorylate many different substrates, including subunits of the NMDA receptor, the AMPA receptor, and ion channels [16], [17], [18].

There is evidence that suggests that the microenvironment in which CaMKII can be found plays a major role in regulating CaMKII function [8]. It is well established that CaMKII located in different subcellular locations behaves differently [19]. In addition, there is indirect evidence that suggests that allosteric changes caused by the microenvironment can bias CaMKII phosphorylation towards different sites. For example, in vitro rates of CaMKII phosphorylation at either Thr286 or Thr253 can be dramatically influenced by different concentrations of ATP, and the rate of Thr286 phosphorylation is more rapid than that observed for Thr253 phosphorylation [11], [20]. In contrast, Thr253 phosphorylation occurs at a relatively rapid rate on CaMKII that is associated with the PSD, even in the absence of added calmodulin [21]. In addition, the reciprocal interaction between CaMKII phosphorylation and CaMKII targeting can be illustrated by studying the interaction of two binding proteins that are present at the PSD (NMDA receptor and CASK). CaMKII that is phosphorylated at Thr286 binds to the NR2B subunit of the NMDA receptor [22], [23]. This binding changes the conformation of CaMKII, keeping it autonomously active. Additionally, this interaction inhibits the phosphorylation of CaMKII at Thr305/306 [23]. In contrast, CaMKII binding to the scaffold protein CASK results in a very different outcome. Whilst CASK binding also changes the conformation of CaMKII, keeping it autonomously active, the conformation is changed in a way that favours autophosphorylation of Thr305/306 [24]. Once the CASK-bound CaMKII becomes phosphorylated on Thr305/306, it dissociates from CASK and releases a pool of CaMKII insensitive to changes in calcium levels. These scenarios highlight the complexity of CaMKII regulation, and demonstrate the importance of molecular environment and the interaction with binding partners in controlling CaMKII function.

Most of the binding sequences on CaMKII responsible for interacting with binding proteins are unknown. Except for the CaMKIIβm splice insert, which contains a putative SH3-domain binding region [25], [26], CaMKII does not contain the well known consensus-binding motifs. Multiple parts of CaMKII can be involved in the binding interactions, with most of the identified sites being located in the C-terminal half of the molecule [8]. For example, BAALC 1-6-8 (Brain and Acute Leukaemia, Cytoplasmic protein) interacts with the C-terminal end of the regulatory domain of α-CaMKII (aa 310–320) [27], and F-actin binds to β-CaMKII in the V1 insert region of the association domain [28].

To identify the proteins responsible for CaMKII targeting, as well as to elucidate the roles of Thr253 and Thr286 phosphorylation in this targeting, we have developed and optimised an in vitro overlay binding assay, based on the method described by McNeill and Colbran [29]. We have examined the interactions between binding proteins and CaMKII, and have investigated the effects of phosphorylation of both the CaMKII and the binding partner on this interaction. In addition, we have identified a major functional consequence following the overexpression of a CaMKII Thr253 phospho-mimic mutant in human neuroblastoma and breast cancer cells, which is potentially mediated by the expression/absence of particular binding proteins.

Section snippets

Primary antibodies

Rabbit antibodies to phospho-Thr253-α-CaMKII were made as described previously [11]. Mouse antibodies to phospho-Thr286-CaMKII, total α-CaMKII, and the FLAG epitope were purchased from Affinity BioReagents (Redfern, NSW, Australia), Chemicon (now Millipore, North Ryde, NSW, Australia), and Sigma (M2, Castle Hill, NSW, Australia), respectively.

Cell lines

COS-7 (ATCC CRL-165; monkey kidney cells), SHSY5Y (ATCC CRL-2266; human neuroblastoma cells), and MDA-MB-231 (ATCC HTB-26; human breast cancer cells) were

Characterisation of the overlay binding assay

We have developed an overlay binding assay in order to examine the interactions of both phosphorylated and non-phosphorylated CaMKII with various proteins. Previous studies that have attempted this have used radio-labelled CaMKII [29], [34], [35], autophosphorylated with 32P-ATP under conditions that result in phosphorylation primarily at Thr286. As we wanted to examine the binding of both phosphorylated and non-phosphorylated CaMKII, we developed a new binding assay that did not involve

Discussion

CaMKII is regulated by multi-site phosphorylation, which can alter enzyme activity, and targeting to cellular microdomains through interactions with binding proteins [8]. These proteins integrate CaMKII into multiple signalling pathways, which may lead to varied functional outcomes following CaMKII phosphorylation, depending on the identity and location of the binding partner. A new phosphorylation site on CaMKII (Thr253) has been identified in vivo [11]. Thr253 phosphorylation does not affect

Acknowledgements

This work was supported by research funds from the National Health and Medical Research Council of Australia, the Hunter Medical Research Institute, and the University of Newcastle. The authors would like to thank Prof. Phil Robinson, Dr. Xiaomei Liao, Dr. Larisa Bobrovskaya, Ms. Lisa Fluechter, Ms. Lucy Murtha, and Ms. Lauren Merritt for their technical assistance on various aspects of this project. The authors declare that they have no conflict of interest regarding any of the work conducted

References (51)

  • S.G. Miller et al.

    J. Biol. Chem.

    (1985)
  • T.R. Soderling et al.

    Trends Neurosci.

    (2000)
  • S. Taha et al.

    Neuron

    (2002)
  • M.K. Shin et al.

    Cell. Signal.

    (2008)
  • J.W. Gurd et al.

    Brain Res.

    (2008)
  • T. Hunter

    Cell

    (2000)
  • Y. Elgersma et al.

    Neuron

    (2002)
  • S. Strack et al.

    J. Biol. Chem.

    (1998)
  • P.I. Hanson et al.

    Neuron

    (1989)
  • A. Dosemeci et al.

    J. Biol. Chem.

    (1994)
  • C.S. Lu et al.

    Neuron

    (2003)
  • V. Urquidi et al.

    FEBS Lett.

    (1995)
  • R.B. McNeill et al.

    J. Biol. Chem.

    (1995)
  • I.T. Lehmann et al.

    J. Biol. Chem.

    (2006)
  • J.P. Liu et al.

    J. Biol. Chem.

    (1994)
  • M. Fährmann et al.

    Biochim. Biophys. Acta

    (2002)
  • T. Tokui et al.

    Biochem. Biophys. Res. Commun.

    (1990)
  • A.B. Jefferson et al.

    J. Biol. Chem.

    (1991)
  • T. Suzuki et al.

    J. Neurochem.

    (2008)
  • T.R. Gaertner et al.

    J. Biol. Chem.

    (2004)
  • J. Li et al.

    Neurosci. Lett.

    (2006)
  • K.P. Giese et al.

    Science

    (1998)
  • K.T. Jones

    Clin. Exp. Pharmacol. Physiol.

    (2007)
  • S. Munevar et al.

    J. Cell. Mol. Med.

    (2008)
  • K.A. Skelding et al.

    Neurochem. Res.

    (2009)
  • Cited by (0)

    View full text