The entry of the calcium ion into the cytoplasm of cells from the extracellular medium or from intracellular stores plays an important signaling role. Cytoplasmic calcium acts as a signal by binding to high-affinity calcium-binding proteins. These proteins in turn act as transducers of the signal by activating other proteins or may be activated directly to carry out enzymatic or structural changes. In this way, many extracellular signals are converted into intracellular activities.
An important subset of these intracellular calcium-binding proteins are proteins that also interact with membranes in a calcium-regulated fashion. In the resting cell, many of these proteins are freely soluble in the cytoplasm (or nucleoplasm). However, when calcium enters the cell, these proteins move onto membrane surfaces. In this way, they make fundamental changes in the character of the membrane surface. Some, such as the annexins (Gerke et al. 2005), are so abundant that up to one half of the intracellular surface of all membrane systems may be covered by annexins in stimulated cells. This influences the organization of the lipids in the membrane, may promote membrane–membrane contacts, and may protect or repair the membrane lipid bilayer structure from chemical or mechanical disruption (Creutz et al. 2012).
One of the most interesting members of this class of calcium-dependent, membrane-binding proteins was not characterized until the membrane-binding proteins of the important model secretory cell Paramecium tetraurelia were studied (Creutz et al. 1998; Tomsig and Creutz 2002). In contrast to either mammals or green plants, in which the majority of membrane-binding proteins are annexins, Paramecium extracts contain only a single major membrane-binding protein. Sequencing of this protein revealed that it was not an annexin, but a protein that bound to membranes through a pair of C2 domains, homologous to the domain of protein kinase C that binds calcium and phospholipids. Further sequence analyses revealed that EST and genomic sequencing databases included many copies of homologous sequences from plants, animals, and slime molds, although the encoded proteins had not been previously studied or described, possibly because they are of much lower abundance than annexins in mammalian or plant cells. This new class of proteins was named “copines,” derived from the French word copine (pronounced ko-peen’), which means a friend, because of the association (poetically, the friendship) of these proteins with membranes. Interestingly, not only are copines almost ubiquitously present in eukaryotes (although absent from yeast and Drosophila), they are generally present in a given organism as a family of multiple homologs – for example, humans have copines 9, nematodes 6, Dictyostelium 6, and plants at least 3. This high degree of conservation and multiplicity of this protein family strongly suggests that the copines play a fundamentally important role in cell biology.
Nomenclature of Copine Family Members in Diverse Organisms
In Humans: copine I, copine II, copine III, copine IV, copine V, copine VI, copine VII, copine VIII, copine IX; alternatively, copines 1 through 9
In Paramecium: CPN1, CPN2
In Arabidopsis: copine 1, copine 2, copine 3; bon1, bon2, bon3
In Dictyostelium: copine A, copine B, copine C, copine D, copine E, copine F; cpnA, cpnB, cpnC, cpnD, cpnE, cpnF
In Caenorhabditis: gem-4 copine, NRA-1 copine
Domain Structure of Copines
Functions of Copines
A key to understanding the biological functions of the copines may be to understand the role of their unique A domain. The similarity to the integrin A domain suggests two rather distinct possibilities. First, the integrin A domain has the classic form, called a Rossman fold, that is present in many nucleotide-binding proteins (Lee et al. 1995). This suggests the copines might require a nucleotide cofactor in order to function. There is one report in the literature that copine III has an associated protein kinase activity (Caudell et al. 2000), although the specific enzymatic activity was extraordinarily low and has not been reported for other members of the protein family.
The other hypothesis suggested by the similarity of the copine A domain to the integrin A domain is that the copine A domain is a site for protein–protein interactions. This hypothesis was tested by screening for proteins that interact with the copine A domain. Using a yeast two-hybrid system with the A domains of three human copines as baits, a number of proteins were identified that interact with the copine A domain (Tomsig et al. 2003). The ability of these proteins to interact directly with copine was tested in in vitro assays using recombinant proteins in which the interactions were found to be of very high affinity, stable to extensive washing. Furthermore, the copines were found to be able to recruit the target proteins to phospholipid surfaces in vitro. In contrast to the integrin A domain, the proteins that were found to interact with the copine A domain are almost exclusively intracellular proteins (Tomsig et al. 2003).
Interaction of Copines with Signaling Proteins
Regulators of phosphorylation: MEK1, protein phosphatase 5, CDC42 binding kinase (homolog of the myotonic dystrophy kinase), TAB2 (TAK1 kinase-binding protein)
Regulators of transcription: Generally these potential targets are not DNA-binding transcription factors but act as regulators of known transcription factors – Myc-binding protein (PAM), Sno proto-oncogene product, Wilm’s tumor associated protein, BcoR (BCL-6 corepressor)
Calcium-binding proteins: Copine I (target of copine IV), ALG2
Regulators of ubiquitination/NEDDylation: UBC12, E2-230
Cytoskeletal regulation: Radixin, BICD2 (dynamitin-binding protein)
Although there is obviously considerable variety in the structures and functions of these targets, it was found that in the majority of cases the binding site for the copine could be mapped to a coiled-coil region of the target protein. Significant specificity for the particular coiled-coil domains in these targets was evident since the copines did not bind coiled-coils of abundant structural or fibrous proteins such as keratins or myosins. A consensus sequence for the coiled-coil copine target motif was determined and found to have predictive value for identifying new copine targets, in particular, the kinase MEK1 (Tomsig et al. 2003).
A Dual-Function Hypothesis
Additional Diverse Studies of the Biology of Copines
Recent studies in animal and plant systems have implicated copines in a variety of functional contexts. Although the mechanistic roles of the copines in these systems are not well defined in most cases, they presumably rely on the ability of calcium to regulate the C2 domains and on the functions of target proteins interacting with the copines.
In the brain, copine VI (or N-copine for neural copine) is one of a number of proteins that is upregulated in the hippocampus after nerve stimulation with kainic acid (Nakayama et al. 1998). It has therefore been speculated that copine may be involved in synaptic potentiation. In addition, copine VI has been observed to migrate into hippocampal cell dendritic spines in response to NMDA stimulation and calcium entry (Reinhard et al. 2016). In a copine VI knockout mouse, long-term potentiation and learning abilities in behavioral tests are reduced (Reinhard et al. 2016). This correlates with the failure of dendritic spines to increase in size in response to hippocampal cell stimulation in the knockout. Evidence suggests that copine VI normally recruits the small GTPase Rac1 to the dendritic spines, leading to stabilization of the actin cytoskeleton and the morphological changes in the spines. In the absence of copine VI, this stabilization of the spines is reduced, and this is possibly responsible for the learning deficits in the knockout mouse (Reinhard et al. 2016).
A copine has been identified in C. elegans that interacts genetically with a plasma membrane cation channel (Church and Lambie 2003). The copine may be involved in the trafficking or insertion of the ion channel into the cell membrane since mutation of the copine gene suppresses certain phenotypes of mutations in the ion channel gene.
A C. elegans copine is also associated in a multiprotein complex with the nicotinic acetylcholine receptor (Gottschalk et al. 2005). Furthermore, mutation of the copine, or downregulation of its expression by siRNA, reduces the number of receptors in the cell membrane therefore implicating the copine in the process of receptor trafficking or membrane insertion.
Genetic experiments in the green plant Arabidopsis have demonstrated that copine plays important roles in controlling cell growth and the hypersensitivity response. In a random screen for mutations causing dwarfism, it was found that plants with a copine mutation grow to a small size (hence the proposed name BON1 (Bonsai) for the copine gene) (Hua et al. 2001). In independent studies, a mutant Arabidopsis plant was isolated that was hypersensitive to low humidity and had enhanced apoptotic responses to stress (Jambunathan et al. 2001). The mutation was mapped to the same copine gene responsible for dwarfism. Green plants have multiple copine genes, similar to the situation in mammals. Therefore, the observation that modification of a single plant copine gene has significant phenotypic effects suggests that copines are not redundant in function.
Dictyostelium discoideum has six copine genes that are differentially regulated during development, indicating the corresponding gene products may each have unique functions (Damer et al. 2005). A knockout of the gene for copine A results in defects in cytokinesis and contractile vacuole function, suggesting roles in membrane biogenesis or function (Damer et al. 2007). In addition, the knockout model has a partial arrest in a late developmental stage.
Copine I appears to regulate signaling from the TNF-α receptor (Tomsig et al. 2004). TNF-α is a potent cytokine that elicits critical biological responses such as inflammation and apoptosis. One specific role of copine in this context is apparently to confer calcium sensitivity on the TNF-α//NF- κB signaling pathway. This leads to cross talk with signaling pathways that introduce calcium into the cytoplasm of cells such as activation of muscarinic receptors coupled to phospholipase C.
Copine III has recently been found in association with ErB2, the human epidermal growth factor receptor 2 in breast cancer cells, where it may play a role in regulation of cancer cell motility (Heinrich et al. 2010).
The copines are a family of calcium-dependent, membrane-binding proteins that consist of nine different gene products in humans. Homologous proteins are expressed in most plants, animals, and protists. The copines are soluble proteins that associate with negatively charged phospholipids in a calcium-dependent fashion through the action of two C2 domains in the N-terminal half of the molecule. The C-terminal half of the copine molecule has an “A domain,” or Von Willebrand domain, that may bind to and recruit target proteins to membrane surfaces. A hypothetical function for the copines is to provide a mechanism for the regulation of various signaling pathways by recruiting components of these pathways to membranes in a calcium-dependent fashion. Additional studies are needed to define the full set of copine-interacting proteins and to determine the effects on these proteins of relocalization or modification by copines.