RICERCA ITALIANA     

Supramolecular complexes of sorcin in the generation and regulation of Calcium-dependent cellular functions

Università degli Studi di Siena

Abstract

Ca2+ is a second messenger that regulates several cellular functions by activating specific enzymatic activities and gene expression (Berridge, 1993). Ca2+ signalling is based on the ability of cells to keep resting levels of free Ca2+ in the cytosol in the range of 100 nM through the activity of Ca2+ transporters both on the plasma membrane and on the endoplasmic reticulum and to regulate transient increases of Ca2+ in the µM range by allowing Ca2+ to enter from extracellular fluids or to be released from internal stores on the endo/sarcoplasmic reticulum (Pozzan et al., 1994). Key elements in deciphering Ca2+ signalling are many Ca2+-binding proteins that, following Ca2+ binding, become able to interact with their specific targets and thus regulate the function of the interacting proteins. Most Ca2+-binding proteins contain characteristic EF hand motif and CaM is probably the most known of these proteins.

Sorcin is a ubiquitous 22 kDa Ca2+-binding protein of the penta EF-hand family of proteins that following Ca2+ binding undergoes a conformational change that allows reversible interaction with target proteins. In the heart, sorcin is involved in the regulation of the excitation-contraction-relaxation processes, where it plays an important role in the transition between contraction and relaxation cycles.. In particular, sorcin has an important role in the transition between contraction and relaxation. Sorcin has at least three different function: 1) inhibits Ca2+ release from SR by inhibiting RyR, 2) increases Ca2+ entry from cytosol to SR by activating SERCA2a and 3) favours Ca2+ extrusion throughout the sarcolemma increasing NCX activity (Lokuta et al., 1997; Farrell et al., 2003, Seidler et al., 2003, Matsumoto et al., 2005).

In addition to sorcin, the RyR2 activity is known to be regulated by several other proteins and interactors. Besides, results from the last decades have pointed out to phosphorylation as a major post-translational regulatory mechanism of RyRs’ activity and this interaction has been extensively investigated for RyR2 in heart cells. Binding of FKBP12.6 to RyR2 appears to contribute to channel stabilization and this binding appears to be affected by PKA phosphorylation of Ser2808 of RyR2 channels. Accordingly, hyperphosphorylation of RyR2 by PKA induced by chronic beta adrenergic stimulation in human heart failure may result in cardiac arrhytmia (Marx et al., 2000; Ono et al., 2000; Yano et al., 2000; Reiken et al., 2003a; 2003c). Mutations in the RyR2 channels found in patients with genetically inherited exercise-induced arrhythmias also seem to change the channel affinity for FKBP12.6. These mutation would also result in an increased channel activity, suggesting that leaky RyR2 channels may be actually responsible for triggering fatal cardiac arrhythmias in these patients (Weherens et at., 2003; Marks et al 2007). Altogether these data point to an important role of phosphorylative events in RyR2 regulation.

Recently (Arrigoni et al., submitted), the beta subunit of the casein kinase 2, CK2, has been found to interact with sorcin. Interestingly, the amino acid sequence of sorcin contains several consensus sites for CK2 kinase phosphorylation (Pinna e Ruzzene, 1996) making it possible to hypothesize that there could be a regulatory function in the interaction between sorcin and CK2. A further link could be extended between CK2 and RyR2, since also the RyR2 amino acid sequence contains several phosphorylation sites for CK2. Considering the evidence for the important role of phosphorylation in the regulation of RyRs’ function it appears important to verify whether CK2 may either indirectly, through regulation of sorcin, or directly, though the direct phosphorylation of RyR2, have a role in regulating Ca2+ release activity of RyR2.

The present project aims to directly test the role of the molecular interactions of sorcin with RyR2 and NCX and the possible role of CK2 in regulating the interactions between these proteins. Major aims of the projects will be the identification of the sorcin binding site in RyR2 and NCX, investigate the existence of active CK2 phosphorylation sites in RyR2 and whether and how CK2 phosphorylation of sorcin may affect its binding to and/or its regulation of RyR2.

The characterization of these interactions will include structural studies, the determination of the topology of interactions leading to a better understanding of the physiological role of sorcin in the heart. In addition, the project bears important connections of high clinical relevance with major cardiac pathologies and to important processes like apoptosis and Multi Drug Resistance (MDR) for which it may potentially open the way to develop new pharmacological tools. <<<

Principal Investigator

Vincenzo Sorrentino Università degli Studi di SIENA

Research Objectives

The study of sorcin and of its target/interacting proteins represents a common point in the research of the three units participating to this project. Unit Colotti has a long lasting experience in studying both structural and functional properties of sorcin. Conversely, Unit Sorrentino has a research record on genetics and function of ryanodine receptors and Unit Ruzzene can supply relevant support in studying protein phosphorylation patterns. Given this background we expect that the coordinated activity of the three units present in this project can contribute to extend our knowledge on the interaction occurring between sorcin and RyR2 and between sorcin and CK2. The molecular basis of the formation of sorcin NCX will also be studied. Both RyR2 and NCX represent key interactors of sorcin in the context of regulation of cardiac contraction: RyR2 is involved essentially in the excitation-contraction phase, and its opening determines the increase of cytoplasmic calcium concentration, while NCX mostly contributes to the subsequent phase of decontraction. As to the interaction between sorcin and CK2, we expect to know whether this interaction may result in the phosphorylation of sorcin and, should this be the case, we will verify the predicted sites as the actual targets of CK2, and to verify whether phosphylation of sorcin may affect its binding to and/or its regulation of RyR2.

Planned activities are expected also to provide direct information on whether RyR2 may be a substrate for CK2, with possible identification of the phosphorylated sites. The potential involvement of CK2 in the phosphorylation of RyR2 is of particular interest since hyperphosphorylation of RyR2 appears to be a way to regulate channel activity and has been associated to several pathological heart diseases. Infact, PKA phosphorylation of Ser2808 in RyR2 by decreasing the binding of FKBP12.6 appears to contribute to channel destabilization. Accordingly, chronic PKA hyperphosphorylation of RyR2 in heart failure can result in incomplete channel closure and Ca2+ “leak” during diastole, which causes depletion of the SR Ca2+ content and reduced Ca2+ release upon receptor activation in failing human heart (Marx et al., 2000; Marks et al., 2007). Also, some mutations in the RyR2 channels found in patients affected by exercise-induced arrhythmias also seem to reduce the channel affinity for FKBP12.6, and to induce an increase of channel activity that may be responsible for triggering fatal cardiac arrhythmias (Weherens et at., 2003). Although the role of phosphorylation of RyR2 in the development of cardiac pathology still remain to be elucidated and additional elements may contribute the pathogenesis of cardiac dysfunction and heart failure (Benkuski et al., 2007; Ferrero et al., 2007), the role of phosphorylation in the regulation of RyR2 channel activity is certainly a relevant problem.
If CK2 turns out to be involved in sorcin/RyR2 regulation, it is important to consider the possibility of making CK2 a target of pharmacological intervention given the availability of highly specific CK2 cell permeable inhibitors. The relationship between sorcin, CK2 and RyRs in the context of the multidrug resistance (MDR) will also be of high interest considering the poor level of information on mechanisms leading to MDR. Given that MDR is a major reason for the failure of cancer therapy the potential medical relevance of this implications is also very high. <<<

First Results

The study of sorcin and of its target/interacting proteins represents a common point in the research of the three units participating to this project. The dynamic establishment of complexes with ionic channels such as the ryanodine receptor (RyR2), L-type voltage dependent channel, Na+-Ca2+ exchanger (NCX) and the sarcoplasmic reticulum Ca2+-ATPase (SERCA2a) in the heart, allows sorcin to contribute to their regulation. Overall, sorcin participates in lowering cytoplasmic Ca2+ concentration, operating in at least three different modalities: it inhibits calcium release from SR by inhibiting RyR2, increases calcium entry from cytosol to SR and favours its extrusion throughout the sarcolemma, by activating SERCA2a and NCX, respectively. The knowledge of the molecular basis of the interaction of sorcin with the two channels will permit the comprehension of functional alterations in the channels, when they are mutated. Based on the previous experience of the participating units we expect, at the end of the first year, to have defined the regions important for the interaction between RyR2 and sorcin and between the NCX cytoplasmic loop and sorcin.
The characterization of the topology of sorcin-RyR2 and sorcin-NCX interaction will be important to unveil the molecular mechanisms at the basis of the excitation-contraction-decontraction cycle in the myocyte. The knowledge of the molecular details of the supramolecular complexes studied will have an impact for the comprehension of how sensor proteins can regulate complex physiological processes in a modular fashion, such as sorcin (or calmodulin) in the heart. The recent characterization of the natural F112L sorcin mutant confirms the importance of sorcin in the control of heart contraction. The substitution of the F112 residue, located at the end of the D helix next to Asp 113, one of the calcium ligands in the EF3 hand, has been associated with familial hypertrophic cardiomyopathy and hypertension, thus confirming the importance of defining the topology of interaction between sorcin and RyR and NCX to obtain a deeper insight in different cardiac pathologies which may result in the comprehension of the molecular basis of such pathologies.

In addition, we are planning to develop a set of experiments aimed to extend our knowledge on the interaction occurring between sorcin and CK2 and between CK2 and RyR2. About the first point we expect that at the end of the first year the features of the CK2/sorcin interaction will be defined and that we will have results on the phosphorylation of sorcin by CK2. On this basis, the protocol will start for the production of rabbit phosphospecific antibodies. A first outcome of this part of the study will be the determination of a different biochemical and physiological behavior of sorcin upon interaction and/or phosphorylation by CK2. Preliminary results on this aspect are expected, in term of sorcin Ca2+ affinity and interaction with RyR2, by using sorcin in the presence of CK2 subunits (those important for the interaction) or phosphorylated in vitro by CK2. During the second year, we should move the study to cell models, and results are expected on the in vivo association sorcin/CK2 and phosphorylation of sorcin by CK2; we should confirm the effect of CK2 on sorcin function, mostly by the production of phospho-mimetic mutants of sorcin, or by reducing CK2 expression/activity in cells.
As to the interaction between RyR2 and CK2 we expect as a first point to verify CK2-dependent phosporylation of RyR2 by the end of the first year. The biochemical and functional characterization of this interactions are likely to take the most part of the second year, with particular attention in the study of RyR2 phosphorylation in vivo, and to the analysis of its effect on RyR2 function. In particular, evidences are expected on the role of CK2 (alone or in combination with other kinases) in the hyperphosphorylation of the RyR2, which has been frequently reported as related to heart failure. These results could shed some light on the still debated occurrence of RyR2 phosphorylation under pathological conditions. If CK2 turns out to be involved, its consideration as a possible clinical target should be taken into account, given the availability of CK2 inhibitors which, in contrast to what happens for the majority of other kinase inhibitors, are highly specific (besides potent and cell permeable).
The relationship between sorcin, CK2 and RyR in the context of the multidrug resistance (MDR) will also be of high interest considering the poor level of information on mechanisms leading to MDR. If our results will indicate that CK2 is required for the anti-apoptotic role of sorcin (possibly in connection to its association with RyRs, it will be an important finding in the still confused field of MDR, whose medical relevance is obvious (being a major reason for the failure of cancer therapy), but whose molecular mechanisms are only poorly understood. Both sorcin and CK2 are found overexpressed in certain MDR cells, but their exact involvement in this context has not been elucidated yet; the knowledge of their possible connection will disclose new elements for overcoming the MDR phenomenon. Again, in this case, pharmacological inhibition of CK2, already available and well known from a biochemical point of view, should be considered as a potential clinical tool. <<<

Timescale

24 months

National and international background

Ca2+ regulates a number of functions in eukaryotic cells
Ca2+-dependent cellular functions include a wide range of cellular processes, such as secretion, gene expression, muscle contraction, egg fertilization, cell division, apoptosis in virtually all eukaryotic cell types (Berridge, 2006; Rizzuto and Pozzan, 2006). Ca2+ ions have proven to be particularly suitable to act as an intracellular messenger and regulator of cellular function because the very high gradient between the extra-cellular and the cytosolic concentrations. This would allow rapid and reversible changes of its concentration inside the cell, a property that is of key importance for a biological messenger to work properly. In addition, the specific combination of size and charge of Ca2+ allows it to reversibly bind to ligands having sufficient structural complexity to ensure specificity of interaction. The cytosolic concentration of Ca2+ is regulated by transport across plasma and intracellular membranes. Various channels, transport ATPases, uniporters, and antiporters in the plasma membrane, endoplasmic and sarcoplasmic reticulum, and mitochondria are responsible for the transport of Ca2+ (Brini and Carafoli, 2000). Advancements in methods for measuring Ca2+ have revealed that changes in the intracellular Ca2+ concentration occur following specific spatial and temporal patterns. Ca2+ transients may be organized into regular oscillations, whose frequency can be finely modulated to encode specific information. In addition, Ca2+ signalling events can be confined to restricted parts of the cell, giving rise to Ca2+ microdomains controlling different intracellular processes (Berridge, 2006).

Ca2+ signalling requires specific sensor proteins to be decoded

1) the EF-hand protein family
An important aspect of Ca2+ signalling is how complex spatial and temporal signalling arrangements are transduced inside the cells. Ca2+ information is not normally transmitted directly to targets, but it is first processed by sensor proteins. Some of them are committed to the regulation of only one Ca2+-dependent process, such as troponin C in muscle contraction, while others are not target-specific. A number of small cell ligands have been found to bind Ca2+, but the most important class of proteins devoted to decode the information carried by Ca2+ and pass it to targets is represented by a family of proteins known as EF hand proteins, which comprises about 600 members. They may function as a committed separate subunit of a single protein or as a subunit that associates reversibly with different proteins (e.g.,CaM). They may even be an integral portion of the sequence of enzymes (e.g., calpain) (Lewit-Bentley and Rety, 2000). These proteins are based on the EF hand motif, characterized by a helix–loop–helix structure, with a 12 or 14 amino acid long interhelical loop able to bind a calcium ion. The success of the EF-hand domain in nature is related to its structural response to calcium binding. Most of the EF-hand proteins undergo a conformational rearrangement upon passing from the apo to the holo form which then allows the binding of other proteins downstream in the process (Capozzi et al., 2006). The canonical sequence of the EF-hand motif has been detected in small (e.g. calmodulin or S100) proteins and within domains of much larger complex proteins (e.g. myosin or calpain). In most known cases, EF-hand motifs occur in adjacent pairs: parvalbumin and S100 proteins represent the minimal motif. Proteins containing four EF-hand motifs usually have two domains, each formed by a pair of EF-hands, (e.g. calmodulin and troponin C). Calmodulin is the archetype of EF-hand protiens. It is a 17-kDa protein with four calcium-binding sites and serves as a calcium sensor in nearly all eukaryotic cells. In the absence of bound Ca2+, calmodulin consists of two domains, each consisting of a pair of EF-hand motifs joined by a flexible helix. Upon binding of one Ca2+ to each EF hand, these units change conformation, moving hydrophobic residues from the inside to the outside of the domains. The Ca2+-calmodulin complex stimulates a wide array of enzymes, pumps, and other target proteins, inlcuding the Calmodulin-dependent protein kinases (CaM kinases) which, upon calmodulin binding, can phosphorylate many different proteins. In addition, the activated enzyme phosphorylates itself and is thus partly active even after Ca2+concentration falls and calmodulin is released from the kinase (Berg et al., 2002).
2) the penta EF-hand protein family
Penta-EF-hand (PEF) proteins comprise a family of Ca2+-binding proteins that have five EF-hand motifs. In addition to the structural similarities in the EF-hand regions, the PEF protein family members have common features: (i) dimerization through unpaired C-terminal EF5s, (ii) possession of hydrophobic Gly/Pro-rich N-terminal domains, and (iii) Ca2 +-dependent translocation to membranes. Based on comparison of amino acid sequences, mammalian PEF proteins are classified into two groups: Group I PEF proteins (ALG-2 and peflin) and Group II PEF proteins (Ca2+- dependent protease calpain subfamily members, sorcin and grancalcin) (Maki et al., 2002). ALG-2 was found to be involved in the induction of apopotosis in a Ca2+-dependent manner in T cells (Vito et al., 1996). In situ hybridization histochemistry of the rat brain has revealed high levels of ALG-2 mRNAs in the granule and pyramidal cell layers of the hippocampus, choroid plexus, area postrema, and a number of hindbrain nuclei. Moreover, it has been shown that expression levels of ALG-2 in the rat brain increases after temporary focal cerebral ischemia (Li et al., 2000). A novel protein similar to ALG-2 has been cloned and namend peflin (PEF protein with a long N-terminal hydrophobic domain) (Kitaura et al., 1999). Calpains comprise a multigene family of Ca2 +-dependent cysteine proteases, mediating regulatory cleavages of specific substrates involved in a number of processes during differentiation, life and death of the cell (Goll et al., 2003). Sorcin (soluble resistance-related calcium-binding protein), is a 22 kDa protein, first discovered as a gene amplified in multidrug-resistant cancer cells (van der Bliek et al., 1986). Sorcin is expressed in a wide variety of cells and seems to be involved in Ca2 +-homeostasis by modulating Ca2 +-channels (Valdivia, 1998). Finally, grancalcin, a 28 kDa protein, was found to display a Ca2 +-dependent translocation to the granules and plasma membrane of neutrophils, suggesting a role in granule–membrane fusion and degranulation processes (Boyhan et al., 1992).

3) Sorcin, a penta EF-hand protein interacting with different intracellular target proteins
Among PEF proteins, sorcin represents the meeting point of the research interests of the three groups proposing this project. Sorcin has 5 calcium-binding domains (EF1-5) and can bind two Ca2+ ions with high affinity in the ?M range. This binding triggers conformational changes from a "closed" to an "open" configuration leading to the exposure of hydrophobic surfaces. This prompts calcium-mediated translocation of sorcin from soluble (Ca-free state) to membrane sites (Ca-bound state) and target interactions. In addition to Ca2+ sorcin has been proposed to be regulated by phosphorylation. Indeed two potential protein kinase A (PKA) target sites (S149 and S178) are present in the primary sequence of sorcin, although S178 is thought to be the preferential site of sorcin phosphorylation, as in a GST-sorcin fusion protein expressed in bacteria, S178 is the major site phosphorylated by PKA then dephosphorylated by protein phosphatase-1 (Matsumoto et al., 2005). Besides PKA, there are evidences suggesting a possible phosphorylation of sorcin by protein kinase CK2 as well; in fact, sorcin has been recently found as an interacting partner of the ? regulatory subunit of CK2 (Arrigoni et al., submitted); moreover, its primary sequence contains potential targets for CK2, as judged from the well-defined consensus site of this kinase (Pinna and Ruzzene, 1996), which allows a good prediction of phosphorylation occurrence.
Sorcin was originally identified in mammalian cells selected for resistance to a group of natural product anti-cancer drugs, including vincristine and adriamycin (Hamada et al., 1988; Meyers and Biedler, 1981; Wang et al., 1995). Sorcin is overexpressed in many lines of these multidrug-resistance (MDR) cells, in some cases as a result of sorcin gene amplification and co-amplification with the MDR marker, P-glycoprotein (Van der Bliek et al., 1986). Although sorcin's role in eliciting or maintaining drug resistance is not known, increased expression of sorcin has been shown to occur in drug-resistant tumors of cancer patients, suggesting an association between sorcin and MDR (Zhou et al., 2006; Tan et al., 2003; Pareck et al., 2002; Kawakami et al., 2007). In addition, sorcin has been found to participate in different Ca2+–signalling pathways since several different target proteins have been identified. Actually it can interact with the ryanodine receptor (RyR) (Meyers et al., 1995) the ?1 subunit of L-type calcium channels (Meyers et al., 1998), and with SERCA (Matsumoto et al., 2005) in muscle cells, with presenilin 2 in brain (Pack-Chung et al., 2000), with annexin VII (synexin) in adrenal medulla (Brownawell and Creuz, 1997) and in red blood cells (Salzer et al., 2002). In this ensemble, cardiac muscle represents one of the most studied cellular environment for sorcin regulation and activity. Sorcin binds to RyR2 with high affinity and completely inhibits both spontaneous channel and ICa-triggered activity. These effects could be relieved by PKA-dependent phosphorylation of sorcin (Farrell et al., 2003; Lokuta et al., 1997). In addition, sorcin has been found to be also phosphorylated by the ?C isoform of CaMKII. Similarly to what observed for PKA-phosphrylation, CamKII phosphorylation also abolishes the inhibitory effect sorcin on ryanodine receptor (Anthony et al., 2007). Recently, an interaction between sorcin and the ? subunit of caseine kinase II has been also demonstrated (Unit Ruzzene).

However, in vivo overexpression of sorcin has yielded opposite results. Meyers et al. 1995 and Seidler et al., 2003 found depressed Ca2+ transients and contraction (with 20-fold transgenic overexpression), while Suarez et al. 2004 found increased Ca2+ transients, SR Ca2+ load and contractility. Interestingly, a mutation in sorcin (F112L, a mutation within the sorcin D helix) has been proposed to be associated with an inherited form of hypertrophic cardiomyopathy and hypertension (Valdivia et al., 2004). Cardiac myocytes from transgenic mice overexpressing L112-sorcin displayed complex alterations in Ca2+ regulation and contractility, including a slowed inactivation of L-type Ca2+ current, enhanced Ca2+ spark width, duration, and frequency, although no significant alterations were observed in the ventricular function and in blood pressure, suggesting that additional factors may be responsible for the development of cardiac hypertrophy and hypertension in humans (Collis et al., 2007).

references

Anthony D.F et al., J Mol. Cell. Cardiol., in press (2007)

Berg J.M., et al., . In Biochemistry, W. H. Freeman and Company, second edition 2002.

Berridge MJ. Cell Calcium 40: 405-412 (2006).

Boyhan A., et al., J. Biol. Chem. 267:2928–2933 (1992).

Brini M and Carafoli E. . Cell. Mol. Life Sci. 57: 354–370 (2000).

Brownawell A.M., Creuz C.E. .J. Biol. Chem. 272: 22182– 22190 (1997).

Capozzi F., et al., J Biol Inorg Chem 11: 949–962 (2006).

Collis, L. P., et al., FASEB J. 21: 475–487 (2007)

Farrell E.F., et al., J Biol Chem 278:34660–34666 (2003).

Ferrero P., et al., J Mol Cell Cardiol 43:281-291 (2007).

Goll DE, et al., Physiological reviews. 83:731 801 (2003).

Guo T, et al., . Circ Res 99:398–406 (2006).

Hamada H., et al., Cancer Res. 48:3173-3178 (1988).

Kawakami M., et al., Biol Pharm Bull 30:1065-1073 (2007).

Kitaura Y., et al., Biochem. Biophys. Res. Commun. 263: 68–75 (1999).

Lewit-Bentley A. and Rety S. Curr Opin Struct Biol 10: 637-643 (2000).

Li W, Jin ., et al., Neuroscience 96: 161– 168 (2000).

Lokuta A.J et al., J Biol Chem 272: 25333–25338 (1997).

Maki M., et al., Biochimica et Biophysica Acta 1600: 51– 60 (2002).

Matsumoto T., et al., . Basic Res Cardiol. 100:250-262 (2005).

Meyers M.B., Biedler J.L. . Biochem Biophys Res Commun 99:228-235 (1981).

Meyers M.B., et al., J. Biol. Chem. 270: 26411– 26418 (1995).

Meyers M.B et al., J. Biol. Chem. 273: 18930–18935 (1998).

Pack-Chung E., et al., J. Biol. Chem. 275:14440–14445 (2000).

Parekh HK, et al., . Biochem Pharmacol 63:1149-1158 (2002)

Rizzuto R. Pozzan T. Physiol Rev. 86: 369-408 (2006).

Salzer U., et al., Blood. 99:2569-2577 (2002).

Seidler T., et al., . Circ Res 93: 132–139 (2003).

Suarez J., et al., Am J Physiol Heart Circ Physiol 286:H68–H75 (2004).

Valdivia H.H. M Trends Pharmacol. Sci. 19:479–482 (1998).

Valdivia, H. H., et al., J. Muscle Res. Cell. Motil. 25: 605–607 (2004)

Van der Bliek A.M et al. EMBO J. 5:3201–3208( 1986).

Vito P., et al., Science 271: 521–525 (1996).

Wang SL, et al., Biochim Biophys Acta, 1260:285-293 (1995).

Wu Y, et al., Circulation 106:1288–1293 (2002).

Yano M, et al., Circulation 102:2131–2136 (2000).

Zhou Y., et al., . Leukemia Research 30: 469–476 (2006). <<<