摘要
Calcium ions (Ca2+) impact nearly every aspect of cellular life. This review examines the principles of Ca2+ signaling, from changes in protein conformations driven by Ca2+ to the mechanisms that control Ca2+ levels in the cytoplasm and organelles. Also discussed is the highly localized nature of Ca2+-mediated signal transduction and its specific roles in excitability, exocytosis, motility, apoptosis, and transcription. Calcium ions (Ca2+) impact nearly every aspect of cellular life. This review examines the principles of Ca2+ signaling, from changes in protein conformations driven by Ca2+ to the mechanisms that control Ca2+ levels in the cytoplasm and organelles. Also discussed is the highly localized nature of Ca2+-mediated signal transduction and its specific roles in excitability, exocytosis, motility, apoptosis, and transcription. In the furnaces of the stars the elements evolved from hydrogen. When oxygen and neon captured successive α particles, the element calcium was born. Roughly 10 billion years later, cell membranes began to parse the world by charge, temporarily and locally defying relentless entropy. To adapt to changing environments, cells must signal, and signaling requires messengers whose concentration varies with time. Filling this role, calcium ions (Ca2+) and phosphate ions have come to rule cell signaling. Here, I describe our current understanding of Ca2+-mediated signaling (complementing several excellent reviews [Berridge, 2005Berridge M.J. Unlocking the secrets of cell signaling.Annu. Rev. Physiol. 2005; 67: 1-21Crossref PubMed Scopus (194) Google Scholar, Burgoyne, 2007Burgoyne R.D. Neuronal calcium sensor proteins: generating diversity in neuronal Ca2+ signalling.Nat. Rev. Neurosci. 2007; 8: 182-193Crossref PubMed Scopus (392) Google Scholar, Carafoli, 2004Carafoli E. Special issue: calcium signaling and disease.Biochem. Biophys. Res. Commun. 2004; 322: 1097Crossref Scopus (15) Google Scholar, Petersen, 2005Petersen O.H. Ca2+ signalling and Ca2+-activated ion channels in exocrine acinar cells.Cell Calcium. 2005; 38: 171-200Crossref PubMed Scopus (84) Google Scholar, Rizzuto and Pozzan, 2006Rizzuto R. Pozzan T. Microdomains of intracellular Ca2+: molecular determinants and functional consequences.Physiol. Rev. 2006; 86: 369-408Crossref PubMed Scopus (872) Google Scholar]) and place particular emphasis on emerging themes related to Ca2+ binding proteins, Ca2+ entry across the plasma membrane, and the localized nature of Ca2+ signals. Protein function is governed by shape and charge. Ca2+ binding triggers changes in protein shape and charge. Similarly, phosphorylation imparts a negative charge, altering protein conformations and their interactions (Westheimer, 1987Westheimer F.H. Why nature chose phosphates.Science. 1987; 235: 1173-1178Crossref PubMed Scopus (1122) Google Scholar). Protein kinases, comprising ∼2% of eukaryotic genomes, remove phosphate from ATP and covalently attach it to the free hydroxyl groups of serine, threonine, or tyrosine residues. The abilities of Ca2+ and phosphate ions to alter local electrostatic fields and protein conformations are the two universal tools of signal transduction. Cells invest much of their energy to effect changes in Ca2+ concentration ([Ca2+]). Underlying the speed and effectiveness of Ca2+ is the 20,000-fold gradient maintained by cells between their intracellular (∼100 nM free) and extracellular (mM) concentrations. In contrast, the concentration of Ca2+'s cousin, Mg2+, barely differs across the plasma membrane. Why is Ca2+ so avidly excluded from the cytosol? One reason is that Ca2+ binds water much less tightly than Mg2+ and precipitates phosphate. Hence, cells have evolved ways to sequester this dangerous divalent, perhaps at first to simply reduce its cytosolic levels but later to use its binding energy for signal transduction. Unlike complex molecules, Ca2+ cannot be chemically altered. Thus, to exert control over Ca2+, cells must chelate, compartmentalize, or extrude it. Hundreds of cellular proteins have been adapted to bind Ca2+ over a million-fold range of affinities (nM to mM), in some cases simply to buffer or lower Ca2+ levels, and in others to trigger cellular processes. The local nature of Ca2+ signaling is intimately tied to this large range of affinities. The calcium ion can accommodate 4–12 oxygen atoms in its primary coordination sphere, with 6–8 being most common. The chelating compounds EDTA and EGTA cage Ca2+ via a doublet of two amine and four carboxylate groups. In comparison, nature's specialized Ca2+ binding proteins, the oxygen atoms of carboxyl and carbonyl groups (and sometimes water) coordinate binding to Ca2+. Typically, six to seven oxygen atoms surround Ca2+ at ∼2.5 Å in a pentagonal bipyramid (Figure 1A) (Strynadka and James, 1989Strynadka N.C. James M.N. Crystal structures of the helix-loop-helix calcium-binding proteins.Annu. Rev. Biochem. 1989; 58: 951-998Crossref PubMed Google Scholar). The professional protein chelator of Ca2+ is the EF hand domain (named after the E and F regions of parvalbumin) (Nakayama and Kretsinger, 1994Nakayama S. Kretsinger R.H. Evolution of the EF-hand family of proteins.Annu. Rev. Biophys. Biomol. Struct. 1994; 23: 473-507Crossref PubMed Scopus (179) Google Scholar), which is present in hundreds of proteins. Helix-turn-helix motifs are common in proteins ranging from channel voltage sensor “paddles” to DNA-binding proteins. In EF hand helix-turn-helix motifs, negatively charged oxygen atoms cradle Ca2+ within a ∼12 amino acid loop between two orthogonal α helices (Figure 1B). The affinities of EF hand domains for Ca2+ vary ∼100,000-fold depending on a variety of factors ranging from critical amino acids in the Ca2+ binding loop to side-chain packing in the protein core. Calmodulin (CaM1-4) is a small, ubiquitous adaptor protein that amplifies Ca2+'s diminutive size to the scale of proteins. No other molecule more dramatically emphasizes the evolutionary importance of Ca2+ signaling. Having changed only slightly over 1.5 billion years of evolution and being transcribed from three separate chromosomes in humans, expression levels of this protein shaped the beaks of Darwin's finches (Abzhanov et al., 2006Abzhanov A. Kuo W.P. Hartmann C. Grant B.R. Grant P.R. Tabin C.J. The calmodulin pathway and evolution of elongated beak morphology in Darwin's finches.Nature. 2006; 442: 563-567Crossref PubMed Scopus (377) Google Scholar). When Ca2+ binds, the shape of the calmodulin domains change, triggering their ability to relieve protein autoinhibition, remodel active sites, and dimerize proteins (Hoeflich and Ikura, 2002Hoeflich K.P. Ikura M. Calmodulin in action: diversity in target recognition and activation mechanisms.Cell. 2002; 108: 739-742Abstract Full Text Full Text PDF PubMed Scopus (548) Google Scholar). Hundreds of proteins contain calmodulin recruitment sites characterized by interspersed basic and bulky hydrophobic amino acids bracketed by aromatic residues. Calmodulin is shaped like a dumbbell, but with a flexible joint in its middle (Meador et al., 1992Meador W.E. Means A.R. Quiocho F.A. Target enzyme recognition by calmodulin: 2.4 A structure of a calmodulin-peptide complex.Science. 1992; 257: 1251-1255Crossref PubMed Scopus (921) Google Scholar, Meador et al., 1993Meador W.E. Means A.R. Quiocho F.A. Modulation of calmodulin plasticity in molecular recognition on the basis of x-ray structures.Science. 1993; 262: 1718-1721Crossref PubMed Scopus (602) Google Scholar). The EF hands of calmodulin have distinct affinities for Ca2+, and their binding affinities are often increased by interaction with target proteins. Binding of Ca2+ is associated with a large change in conformation and exposure of hydrophobic surfaces within each domain, which triggers calmodulin's Ca2+ sensor activity (binding to its targets). Hydrophobic residues, usually containing methionine, wrap around amphipathic regions of target proteins, such as the α helices in myosin light chain kinase (MLCK; Figures 1B and 1C) and calmodulin dependent kinase II (CaMKII). In many cases, both domains wrap around the target, compacting the structure into a globular shape (for movies, see http://www.molmovdb.org/cgi-bin/morph.cgi?ID=180968-23252). This Ca2+ switch has also been cleverly adapted to a fluorescence resonance energy transfer-based Ca2+ sensor (Palmer and Tsien, 2006Palmer A.E. Tsien R.Y. Measuring calcium signaling using genetically targetable fluorescent indicators.Nat. Protocols. 2006; 1: 1057-1065Crossref PubMed Scopus (343) Google Scholar) and, mimicking nature, will likely be engineered to do much more. Calmodulin also extends the reach of Ca2+ by activating phosphorylation pathways. Ca2+/calmodulin binding relieves autoinhibition of the catalytic domain of calmodulin kinase (CaMK) family enzymes. CaMKIIs multimerize, leading to auto- and interphosphorylations that prolong kinase activity. S100 Ca2+-sensing proteins are the largest family of EF hand proteins (>25 human genes), putatively targeting more than 90 proteins. Like calmodulin, Ca2+ binding in S100 proteins triggers exposure of hydrophobic surfaces to target proteins. Some S100 proteins, as homo- and heterodimers of two 2-EF-hand subunits, assemble/disassemble protein complexes (such as those containing tubulin or p53), although much remains to be discovered (Santamaria-Kisiel et al., 2006Santamaria-Kisiel L. Rintala-Dempsey A.C. Shaw G.S. Calcium-dependent and -independent interactions of the S100 protein family.Biochem. J. 2006; 396: 201-214Crossref PubMed Scopus (458) Google Scholar). Neuronal Ca2+ sensor (NCS) proteins have four EF hands, binding three (frequenin, neurocalcin-δ, GCAP) or two (KChIP) Ca2+ ions (Burgoyne, 2007Burgoyne R.D. Neuronal calcium sensor proteins: generating diversity in neuronal Ca2+ signalling.Nat. Rev. Neurosci. 2007; 8: 182-193Crossref PubMed Scopus (392) Google Scholar). Their structures are also compact and globular, relying on surface charge and membrane association for specificity. Several NCS proteins (recoverin, hippocalcin, VILIP1-3) are Ca2+-triggered switchblades in which a myristoyl group is unsheathed, enabling attachment to membranes. Lipid bilayers are like a workbench, holding proteins on a surface to organize their function and increase speed by reducing diffusion from three dimensions to two. Not surprisingly, many proteins have evolved domains that place, or remove, them from the lipid bilayer. Ca2+, with its positive charge, is often used to change a protein's location in the cell from the cytoplasm to a membrane surface (translocation). A C2 domain is an ∼120 amino acid segment with a common fold, an 8-stranded antiparallel β sandwich connected by variable loops (Cho and Stahelin, 2005Cho W. Stahelin R.V. Membrane-protein interactions in cell signaling and membrane trafficking.Annu. Rev. Biophys. Biomol. Struct. 2005; 34: 119-151Crossref PubMed Scopus (449) Google Scholar). In many C2 domains, binding of two or three Ca2+ ions in the three variable loops creates a substantial electrostatic potential that accelerates a protein's association with anionic membrane leaflets, such as the cytoplasmic surface of eukaryotic plasma membranes and the outer leaflet of outer mitochondrial membranes (Figure 1D). Neutralization of charge by Ca2+ binding in the variable loops of some proteins may allow penetration of hydrophobic and aromatic amino acids into the bilayer. The C2 domain is common in signal-transduction proteins; there are approximately 650 human proteins listed in the protein family (Pfam) database with C2 architectures. These include well-known signaling proteins such as phospholipases, protein-kinase C (PKC), phosphoinositide 3-kinase (PI3K), synaptotagmins, rabphilin, and Munc. Additional specificity is conferred upon this group of proteins by a second domain. For example, C1 domains bind diacylglycerol, while pleckstrin homology (PH) and PX domains bind phosphatidylinositol lipids with specificity determined by phosphate positions in the inositol ring. Thus, increases in [Ca2+] initiate translocation of proteins with C2 and other domains (e.g., protein kinase C family proteins) to specific regions of membranes containing their substrate. Another Ca2+-dependent membrane targeting scheme is employed by annexins, where phosphoryl moieties of the membrane replace charge from carbonyl oxygens and water in a unique Ca2+-binding fold (Gerke et al., 2005Gerke V. Creutz C.E. Moss S.E. Annexins: linking Ca2+ signalling to membrane dynamics.Nat. Rev. Mol. Cell Biol. 2005; 6: 449-461Crossref PubMed Scopus (1056) Google Scholar). Both phosphatidylinositol 4, 5 bisphosphate (PIP2) and calmodulin (either Ca2+ bound or free) are highly negatively charged, but PIP2 is bound to inner leaflets of plasma membranes by its acyl chains, whereas calmodulin is soluble and cytosolic. Both are ubiquitous and abundant. Clusters of positively charged residues on many peripheral (e.g., K-Ras, MARCKS) and integral (e.g., the juxtamembrane regions of ion channels and the EGFR) proteins produce a local positive potential that acts as a basin of attraction for PIP2, both enhancing the local concentration of PIP2 and pulling the cluster close to the membrane. Negatively charged calmodulin competes with PIP2 and may pull the positively charged protein segment off the membrane (Figure 2) (McLaughlin and Murray, 2005McLaughlin S. Murray D. Plasma membrane phosphoinositide organization by protein electrostatics.Nature. 2005; 438: 605-611Crossref PubMed Scopus (675) Google Scholar). Further elaboration of this mechanism of competitive switching is enabled by PIP2 generation/hydrolysis, PKC phosphorylation of the basic cluster, and the degree of Ca2+ binding to calmodulin. Membrane-bound phosphatidylinositol 4, 5 bisphospate (PIP2) and cytosolic calmodulin are both highly negatively charged (red field lines). PIP2 accumulates near a positively charged (blue field lines), amphipathic region of a protein, or corresponding peptide, and pins it to the inner leaflet. An increase in local [Ca2+]i activates calmodulin, which “pulls” the basic region off from the membrane. Activation of phospholipase C (PLC) hydrolyzes PIP2, which also can reduce the basic cluster's interaction with the membrane. The best evidence for this mechanism comes from peripheral proteins, such as K-Ras, and MARCKS (McLaughlin and Murray, 2005McLaughlin S. Murray D. Plasma membrane phosphoinositide organization by protein electrostatics.Nature. 2005; 438: 605-611Crossref PubMed Scopus (675) Google Scholar), but similar mechanisms may also affect the Ca2+-permeant, PIP2- and Ca/CaM-gated TRP channels. This is modified from original figures provided by Murray and McLaughlin (McLaughlin et al., 2005McLaughlin S. Smith S.O. Hayman M.J. Murray D. An electrostatic engine model for autoinhibition and activation of the epidermal growth factor receptor (EGFR/ErbB) family.J. Gen. Physiol. 2005; 126: 41-53Crossref PubMed Scopus (99) Google Scholar, McLaughlin et al., 2002McLaughlin S. Wang J. Gambhir A. Murray D. PIP2 and proteins: interactions, organization, and information flow.Annu. Rev. Biophys. Biomol. Struct. 2002; 31: 151-175Crossref PubMed Scopus (662) Google Scholar). Ca/CaM in the cytoplasm is shown in its open conformation. PIP2-charged head groups are yellow; the hydrophobic peptide with interspersed basic amino acids is green. Blue lines, +25 mV; Red lines, −25 mV electrostatic potentials. Calmodulin image reproduced from The Journal of General Physiology (2005) 126, 41–53. Copyright 2005 The Rockefeller University Press. Like Sisyphus, ATPase pumps are condemned to push Ca2+ uphill for eternity into the endoplasmic reticulum (ER) (via sarcoendoplasmic reticular Ca2+ ATPases; SERCA pumps) or out of the cell (via plasma membrane Ca2+ ATPases; PMCA pumps) (Figure 3A). To maintain low cytoplasmic [Ca2+], ATPases exchange protons for two (SERCA) or one (PMCA) Ca2+ per ATP hydrolyzed. SERCA (ATP2A1-3) and PMCA (ATP2B1-4) Ca2+ pumps are P type ATPases, defined by an obligatory aspartyl phosphate intermediate in the pump cycle (Strehler and Treiman, 2004Strehler E.E. Treiman M. Calcium pumps of plasma membrane and cell interior.Curr. Mol. Med. 2004; 4: 323-335Crossref PubMed Scopus (132) Google Scholar). A second mechanism, the Na+/Ca2+ exchangers (NCX, or SLC8A1-3), and the Na+/Ca2+-K+ exchangers (NCKX; SLC24A1-5) exchange one Ca2+ ion for three Na+ ions (NCX) or cotransport one K+ ion with one Ca2+ ion in exchange for four Na+ ions (NCKX). Running in their “forward” modes, inward (depolarizing) Na+ current drives Ca2+ extrusion. The high-affinity, low-capacity PMCAs and the low-affinity, high-capacity Na+/Ca2+ (−K+) exchangers complement each other. The PMCAs are effective at maintaining low internal [Ca2+] over long durations, whereas NCX and NCKX can make the rapid adjustments needed during generation of action potentials in neurons (Hilgemann et al., 2006Hilgemann D.W. Yaradanakul A. Wang Y. Fuster D. Molecular control of cardiac sodium homeostasis in health and disease.J. Cardiovasc. Electrophysiol. 2006; 17: S47-S56Crossref PubMed Scopus (40) Google Scholar). Calmodulin can substantially increase both PMCA Ca2+ affinity and ATPase pump rate. Voltage-gated Ca2+-selective channels (CaVs) are the fastest Ca2+ signaling proteins and initiate dramatic changes within a single cell (Figure 3B). Each channel conducts roughly a million Ca2+ ions per second down the 20,000-fold gradient; a few thousand channels/cell can effect >10-fold changes in intracellular levels within milliseconds. Like transistors, the triggering messengers are the photons of the electromagnetic field. The channel's antenna is a paddle-shaped helix-turn-helix loop containing positively charged residues (usually arginines) (Long et al., 2005Long S.B. Campbell E.B. MacKinnon R. Voltage sensor of Kv1.2: structural basis of electromechanical coupling.Science. 2005; 309: 903-908Crossref PubMed Scopus (768) Google Scholar). A change in voltage moves the paddle that in turn pulls the channel “gate” open. Hodgkin and Huxley first documented that, unlike transistors, conductances in biological membranes can be highly selective. Calcium selectivity is a consequence of moderately high-affinity Ca2+ binding in the pore of the channel. Based on mutagenesis of Ca2+ channels and inferences from other Ca2+-binding proteins such as Ca2+-ATPases, seven oxygens contributed by aspartate and glutamate side chains are likely to form the Ca2+ cage (Gouaux and Mackinnon, 2005Gouaux E. Mackinnon R. Principles of selective ion transport in channels and pumps.Science. 2005; 310: 1461-1465Crossref PubMed Scopus (592) Google Scholar). In the presence of normal extracellular [Ca2+], Ca2+ block is essential for selectivity as only Ca2+ can bind and then enter the pore. When external Ca2+ is removed, Ca2+ channels become nonselective and allow Na+ and K+ to transit the membrane. The simplest type of Ca2+ compartmentalization is established by the cell's plasma membrane. Ca2+ signaling occurs between cells in two ways. First, cellular Ca2+ autonomy can be circumvented by gap junction (connexin) channels, as often occurs in epithelia and always in cardiomyocytes. More commonly, cell-to-cell signaling is effected by transmitter-gated, usually Ca2+ permeant, ion channels (e.g., NMDA, nicotinic, purinergic ionotropic). Voltage-gated Ca2+ channels (CaV) rapidly increase periplasmic [Ca2+] that in turn trigger protein-fusion machines (e.g., synaptotagmins and SNARE complexes), enabling vesicles containing transmitter molecules to fuse to the plasma membrane. Small molecules (ATP, acetylcholine) and single amino acids (e.g., glutamate) released outside the cell gate Ca2+-permeant channels (P2X, nicotinic receptors, NMDA receptors) on adjacent cell membranes. These Ca2+-mediated events dominate much of neuroscience, and corollaries are now appreciated in extracellular communication between almost all cell types. Ca2+ control of synaptic release and Ca2+ entry via neurotransmitter channels is reviewed elsewhere (Jahn and Scheller, 2006Jahn R. Scheller R.H. SNAREs–engines for membrane fusion.Nat. Rev. Mol. Cell Biol. 2006; 7: 631-643Crossref PubMed Scopus (1734) Google Scholar, Sudhof, 2004Sudhof T.C. The synaptic vesicle cycle.Annu. Rev. Neurosci. 2004; 27: 509-547Crossref PubMed Scopus (1772) Google Scholar). As these were recently reviewed (Ramsey et al., 2006Ramsey I.S. Delling M. Clapham D.E. An introduction to TRP channels.Annu. Rev. Physiol. 2006; 68: 619-647Crossref PubMed Scopus (1120) Google Scholar), only a few points about transient receptor potential (TRP) ion channels are summarized here. TRP ion channels are formed by tetrameric assembly around a pore. Most are weakly voltage-sensitive, nonselective ion channels. The sole yeast, and some mammalian, TRP channels span only intracellular membranes. The majority of the 28 mammalian TRPs comprise plasma membrane channels that depolarize cells and increase intracellular Na+ and Ca2+. Many TRP channels are greatly potentiated by phospholipase C (PLC) activation by G protein-coupled or tyrosine-kinase receptors (TKR). Intracellular Ca2+ gates some mammalian TRP channels but modulates practically all TRP channels. Their physiological roles are most clearly established in sensory systems, and indeed many are activated by environmental signals such as temperature change, pH, volatile chemicals, and plant compounds, but their functions are probably much broader. The central unanswered question in this field is how TRP channels are normally activated in vivo. TRP channels are frequently, but incorrectly, called store-operated channels. In a strict definition of a store-operated channel, the channel is activated by store depletion even when cytoplasmic [Ca2+] levels are buffered to low levels (otherwise, all Ca2+-activated, and modulated, Ca2+-permeant channels would be store operated channels). Thus far, TRP channels by themselves do not satisfy these conditions. Ca2+ is constantly seeping out of the ER into the cytoplasm. SERCAs tirelessly pump it back into the ER. If these pumps are blocked, ER [Ca2+] runs down. Similarly, prolonged incubation of many cells in low Ca2+ media allows the PMCAs to extrude the leaked Ca2+, depleting ER Ca2+. In many nonexcitable cells, Ca2+ entry across the plasma membrane is infrequent (blood cells for instance are exceptionally “tight”). IP3 receptor (IP3R)-mediated release of Ca2+ from the ER in response receptor activation empties the ER as PMCAs pump Ca2+ out of the cell faster than it can be repleted. Slowly over seconds after such store depletion, a Ca2+ entry mechanism is activated. This mechanism is called store-operated Ca2+ entry (Putney, 2005Putney Jr., J.W. Capacitative calcium entry: sensing the calcium stores.J. Cell Biol. 2005; 169: 381-382Crossref PubMed Scopus (139) Google Scholar). Researchers have paid special attention to a slow, tiny, but highly selective Ca2+ conductance that is activated when ER [Ca2+] drops (CRAC or Ca2+-release activated current; ICRAC) (Parekh and Penner, 1997Parekh A.B. Penner R. Store depletion and calcium influx.Physiol. Rev. 1997; 77: 901-930Crossref PubMed Scopus (1265) Google Scholar). Most importantly, CRAC is activated by a decline in ER [Ca2+] but not by a rise in cytoplasmic [Ca2+]. Although TRPs and other channels were widely proposed for this role, only CRAC, so far, demonstrates activation that is independent of cytoplasmic Ca2+, consistent with it being critical for store-operated Ca2+ entry. The single-channel conductance of ICRAC is ∼15 femtosiemens (fS), approximately 1000-fold lower than that of other ion channels (Prakriya and Lewis, 2006Prakriya M. Lewis R.S. Regulation of CRAC Channel Activity by Recruitment of Silent Channels to a High Open-probability Gating Mode.J. Gen. Physiol. 2006; 128: 373-386Crossref PubMed Scopus (113) Google Scholar) (total CRAC current from an entire cell is only ∼5–10 pA). T cells from two brothers suffering from immunodeficiency had long ago been shown to lack ICRAC (Partiseti et al., 1994Partiseti M. Le Deist F. Hivroz C. Fischer A. Korn H. Choquet D. The calcium current activated by T cell receptor and store depletion in human lymphocytes is absent in a primary immunodeficiency.J. Biol. Chem. 1994; 269: 32327-32335Abstract Full Text PDF PubMed Google Scholar). In T cells, the primary Ca2+ entry pathway is CRAC; Ca2+-dependent dephosphorylation of the nuclear factor of activated T cells (NFAT) by calcineurin initiates its translocation to the nucleus for regulation of chemokine genes (Feske, 2007Feske S. Calcium signalling in lymphocyte activation and disease.Nat. Rev. Immunol. 2007; 7: 690-702Crossref PubMed Scopus (723) Google Scholar). The first clue as to the molecular component of store-operated Ca2+ entry came when STIM1, a single transmembrane-spanning domain protein primarily residing in the ER, was found to be essential for ICRAC activation (Roos et al., 2005Roos J. Digregorio P.J. Yeromin A.V. Ohlsen K. Lioudyno M. Zhang S. Safrina O. Kozak J.A. Wagner S.L. Cahalan M.D. et al.STIM1, an essential and conserved component of store-operated Ca2+ channel function.J. Cell Biol. 2005; 169: 435-445Crossref PubMed Scopus (1431) Google Scholar). STIM1's N terminus sterile α motif may seed multimerization, whereas its Ca2+-binding EF hand is presumably the sensor of ER [Ca2+] (Lewis, 2007Lewis R.S. The molecular choreography of a store-operated calcium channel.Nature. 2007; 446: 284-287Crossref PubMed Scopus (410) Google Scholar). Subsequently, a four-transmembrane domain plasma-membrane protein, Orai1 (Feske et al., 2006Feske S. Gwack Y. Prakriya M. Srikanth S. Puppel S.H. Tanasa B. Hogan P.G. Lewis R.S. Daly M. Rao A. A mutation in Orai1 causes immune deficiency by abrogating CRAC channel function.Nature. 2006; 441: 179-185Crossref PubMed Scopus (1708) Google Scholar, Zhang et al., 2006Zhang S.L. Yeromin A.V. Zhang X.H. Yu Y. Safrina O. Penna A. Roos J. Stauderman K.A. Cahalan M.D. Genome-wide RNAi screen of Ca2+ influx identifies genes that regulate Ca2+ release-activated Ca2+ channel activity.Proc. Natl. Acad. Sci. USA. 2006; 103: 9357-9362Crossref PubMed Scopus (700) Google Scholar), was shown to be required for CRAC activity, and mutagenesis of its pore proved that it was the channel-forming subunit (Prakriya et al., 2006Prakriya M. Feske S. Gwack Y. Srikanth S. Rao A. Hogan P.G. Orai1 is an essential pore subunit of the CRAC channel.Nature. 2006; 443: 230-233Crossref PubMed Scopus (1027) Google Scholar, Yeromin et al., 2006Yeromin A.V. Zhang S.L. Jiang W. Yu Y. Safrina O. Cahalan M.D. Molecular identification of the CRAC channel by altered ion selectivity in a mutant of Orai.Nature. 2006; 443: 226-229Crossref PubMed Scopus (666) Google Scholar). Upon ER Ca2+ depletion, STIM1 aggregates in the ER just below Orai1 in the plasma membrane (Figure 4A). Although apposition of the ERM domain-containing C terminus of STIM1 is within 25 nm of Orai1, a direct link has not been established (Wu et al., 2006Wu M.M. Buchanan J. Luik R.M. Lewis R.S. Ca2+ store depletion causes STIM1 to accumulate in ER regions closely associated with the plasma membrane.J. Cell Biol. 2006; 174: 803-813Crossref PubMed Scopus (633) Google Scholar). What causes STIM1 to cluster at plasma-membrane regions adjacent to ER membranes, and what gates Orai? Related to these details is a more important question. Like the interaction between the CaV-Ryanodine receptor Ca2+ calcium channels in muscle (see section below), the Stim/Orai channel complex brings Ca2+ into a narrow region between the plasma membrane and the ER. What are the targets of this localized release? Interestingly, submembranous mitochondria receive (and buffer) much of this localized Ca2+ increase. Surely there is much cell biology remaining to be uncovered, and it may have little to do with ER Ca2+ repletion. Like an aged homeowner, evolution has crowded every nook and cranny of the cell with its handiwork of lipids and proteins. Thus, nonuniformity, cooperativity, and compartmentalization are essential features of cell biology, even on the nanomolar scale. Hydrated Ca2+ can diffuse 40 μm in 1 s (40 nm/ms) in simple saline solution (Einstein, 1905Einstein A. Über die von der molekularkinetischen Theorie der Wärme geforderte Bewegung von in ruhenden Flüssigkeiten suspendierten Teilchen.Annalen der Physik. 1905; 17: 549-560Crossref Scopus (3935) Google Scholar). This mobility is never realized in the crowded, charged cytosolic world where the freedom of Ca2+ is measured in nanometers and microseconds. Ca2+ exits single ion-channel pores at rates >1 per μs, but fixed and mobile endogenous buffers limit CaV-mediated changes in [Ca2+] to 10 μM levels within 20 nm (Naraghi and Neher, 1997Naraghi M. Neher E. Linearized buffered Ca2+ diffusion in microdomains and its implications for calculation of [Ca2+] at the mouth of a calcium channel.J. Neurosci. 1997; 17: 6961-6973Crossref PubMed Google Scholar), a volume that can accommodate several 40 Å calmodulin molecules (Figure 5). The steep Ca2+ gradient around entry sites can give rise to nonhomogeneous activation of Ca2+ binding proteins with similar Ca2+ affinities. Countering these steep gradients are mobile buffers and mobile Ca2+-trigger proteins, which prolong the Ca2+ signal and increase its effective length constant. The fine reticular ER and mitochondria spread like a vast three-dimensional spider web within cells, actively sequestering Ca2+. The distributed nature of these Ca2+ compartments insures that Ca2+ is only briefly free before encountering an extrusion (PMCA, NCX) or uptake (SERCA/MiCa) mechanism. Intracellular Ca2+ compartments are also Ca2+ distribution systems. The pinnacle of evolutionary success in this regard is skeletal muscle, where the sarcoplasmic reticulum