News Physiol Sci 17: 144-149, 2002;
doi:10.1152/nips.01385.2002
1548-9213/02 $5.00
News in Physiological Sciences, Vol. 17, No. 4, 144-149,
August 2002
© 2002 Int. Union Physiol. Sci./Am. Physiol. Soc.
Rapidly Exchanging Ca2+ Stores: Ubiquitous Partners of Surface Channels in Neurons
Jacopo Meldolesi
Department of Neuroscience and Excellence Centre in Cell Differentiation, Vita-Salute San Raffaele University, and DIBIT, San Raffaele Institute, 20132 Milan, Italy
 |
Abstract
|
|---|
Neuronal rapidly exchanging Ca2+ stores coincide with the endoplasmic reticulum and possess ample, nonrandom distribution, dual receptor channels, IP3 and ryanodine receptors, and heterogeneous membranes. Because of these properties, the stores are able to reinforce and expand the Ca2+ signals generated at the surface, working as partners of voltage- and receptor-gated channels.
|
Introduction
|
|---|
Ca2+, a fundamental messenger in all cells, is charged with additional, sophisticated roles in neurons. In neurons, in fact, almost no function is carried out without regulation by the cation. For this to occur, the cytosolic Ca2+ concentration ([Ca2+]i) needs to be dynamically controlled, i.e., it does not remain stable (or almost stable, as is the case with the concentrations of other ions) but increases appropriately following stimulation, in such a way that specific intracellular events are switched on, switched off, or, even more importantly, modulated in a positive or negative fashion. In view of the multiplicity and heterogeneity of these regulated events, the [Ca2+]i rises that are necessary must be of variable size. In some cases, in fact, the operative levels remain in the submicromolar range, just a few times higher than the resting level; in others, they need to reach much higher values, in the 10- to 100-µM range, to be effective.
High [Ca2+]i rises are easy to figure out in the proximity of the plasma membrane, where transient equilibria are quickly established between influx (sustained by the opening of surface Ca2+ channels) and efflux (induced by the Ca2+-dependent activation of pumps and transporters). Everyone agrees, therefore, that processes characterized by low Ca2+ affinity occur in the thin (~100 nm) subplasmalemmal cell layer. Deeper in the cytoplasm, however, the [Ca2+]i rises sustained by the opening of surface channels would be inevitably and extensively buffered by cytosolic Ca2+-binding proteins. Therefore, were Ca2+ supply limited to the surface, the [Ca2+]i rises in the bulk of the cell (most of the cytoplasm and the nucleus) would be limited to a narrow range and low-affinity events would be impossible. This is not the case, because Ca2+ supply does take place not from a single but from two types of sources: the extracellular medium and the intracellular, rapidly exchanging Ca2+ stores. The first predominates at the superficial layer, the second in the deep areas of the cell. The two sources do not operate independently but rather coordinately with each other, in such a way that the [Ca2+]i responses they sustain are different in multiple respects, including time and space, with respect to those induced by one source only.
In the present short review, we will focus primarily on neurons and nerve cells, discussing first the major mechanisms operating in the [Ca2+]i control, then a few rapid events in which these mechanisms operate. For more comprehensive presentations of these events, as well as for those of other, slower functions regulated by Ca2+, including mitochondrial metabolism, gene expression, and axonal growth, the reader is referred to other recent reviews by myself (12) and others (2, 19).
|
Cellular and molecular biology of Ca2+ stores and of their environment, the cytosol
|
|---|
Intracellular Ca2+ stores were discovered in the early 1970s; however, understanding of their role has accumulated primarily during the last 10 years. This has been largely due to the development of [Ca2+]i imaging and its progressive integration with patch-clamp electrophysiology, together with increased correlation to classical cell biology, advanced biochemistry, and molecular biology.
The stores consist in the intracellular membrane systems most widely developed in almost all cell types, including neurons: the endoplasmic reticulum (ER), together with a fraction of the Golgi complex. Since, at the moment, no Ca2+ specificities are known in detail, the latter structure will not be discussed further. Traditionally, the ER is considered to be composed by multiple, discrete cisternae. Recent morphological (Ca2+ imaging as well as confocal and electron microscope) and functional evidence in nonneuronal cells has shown, however, the cisternae to be extensively continuous to each other, yielding a vast Ca2+ store network system (Fig. 1
). Such an extensive and stable network might not be present in neurons. In these cells, a higher degree of ER functional and structural complexity is known to exist. Therefore, a corresponding molecular and architectural heterogeneity of the stores does not look unreasonable (see Ref. 3), especially in regions such as the axons, dendrites, and their specialized terminal expansions (12, 18). In this presentation, therefore, membrane continuity of the stores is neither excluded nor taken for granted.
View larger version (142K):
[in this window]
[in a new window]
|
FIGURE 1. Electron microscopy of the neuronal model, PC-12 cell, quick frozen, freeze dried, then exposed to OsO4 vapors and embedded in Epon (details in Ref. 13). The procedure preserved cell ultrastructure much better than conventional fixation. The morphology of the endoplasmic reticulum (ER) appears to be composed not of the usual small cisternae but of long, anastomized elements (arrows) that serial sections show to be largely interconnected (arrows) in an extensive network. Mitochondria (*) appear very dense. DG, dense granules; N, nucleus.
|
|
The ER is distributed throughout the whole neuron. In particular, cisternae are stable components not only of the somata and fiber shafts but also of presynaptic nerve terminals, from which neurotransmitters and neuromodulators are mostly released, and of dendritic spines, specialized for postsynaptic transduction to occur and neuronal signaling to get started. We conclude therefore that the distribution of the stores is fully compatible with their widespread participation in intracellular Ca2+ events.
To understand the functioning of Ca2+ stores, information is needed about the compartment they live in, the cytosol (9). In this compartment, only a small fraction (50–100 nM) of the total calcium content (20–300 µmol/l) remains free and ionizes, while the rest is bound. The low free/total ratio is due to the existence of an efficient buffering system constituted primarily by high-affinity Ca2+-binding proteins, some of which (e.g., calbindin, calretinin, parvalbulin) are without any other known function, whereas others (e.g., calmodulin, calcineurin, etc.) participate in Ca2+ regulation. In addition, due to the presence of these proteins, Ca2+ ions are not free to move throughout the cytosol as in pure water; rather, their diffusion rate is slowed down considerably (100-fold). When surface and/or store channels open, the transmembrane flux of Ca2+ to the cytosol can be such that the local buffering system of proteins is transiently saturated (9). The [Ca2+]i near the mouth of the channels can rise therefore much higher than the average [Ca2+]i of the cytosol. This concept will be fundamental for the following discussion of neuronal Ca2+ homeostasis.
To be operative, rapidly exchanging Ca2+ stores need to possess at least three types of molecular components. These components are appropriately distributed: the first two, pumps and channels, in the membrane and the third, Ca2+-binding proteins, in the luminal space. Until recently, the distribution of these components, whether even throughout the ER system or concentrated at critical areas, was a matter of debate. In the last few years, however, strong functional and immunocytochemical evidence has demonstrated the heterogeneous distribution of channels and probably also of pumps. On the contrary, most (but not all) Ca2+-binding proteins are believed to be widely, if not evenly, distributed.
|
SERCA pumps
|
|---|
Ca2+ pumping from the cytosol to the ER lumen is carried out by SERCAs, the sarcoplasmic-endoplasmic reticulum Ca2+-ATPases. In the brain, as well as in other nonmuscle cells, the most widely expressed SERCA is type 2b. SERCAs are high-affinity pumps, activated any time [Ca2+]i rises higher than resting levels. Results with Ca2+ dyes had initially suggested that empty stores require relatively long periods of time (minutes) to be replenished. Recently, however, replenishment has been shown to be much faster (from seconds down to tenths of milliseconds), especially in restricted areas of the neuron (presynaptic terminals, dendritic spines) (13) (Fig. 2). These findings may be explained, at least in part, by the mechanisms of SERCA regulation (10). The pump, in fact, operates in direct interaction with specific Ca2+-binding proteins. At the cytosolic surface, a Ca2+-induced dephosphorylation mediated by calcineurin causes its very quick detachment from an inhibitor, the ER transmembrane protein calnexin, and thus its activation. At the luminal surface, interaction of SERCA with calreticulin, a ubiquitous Ca2+-binding protein, switches the equilibrium of the store toward partial filling and contributes to the restoration of the resting luminal concentration (~200–400 µM). This, however, can occur only with some delay. Transiently, therefore, the levels of Ca2+ accumulated within the stores can reach values (>10 mmol/l) that exceed the buffering potential of the compartment, with ensuing rise of the luminal [Ca2+]i (10).
View larger version (118K):
[in this window]
[in a new window]
|
FIGURE 2. Ca2+ maps obtained by imaging of electron energy loss spectroscopy analyses of a frog neuromuscular junction. The preparations, incubated in a Ringer solution containing 10 µM Ca2+ and 1 mM 4-aminopyridine, were quick frozen after electrical stimulation. Notice the strong positivity for Ca2+ (red labeling in B) not only within synaptic vesicles (V), which are also labeled at rest, but also in mitochondria (m) and ER, which accumulate the label in the time since stimulation (only 10 ms). Reproduced with permission from Pezzati et al., Biochim Biophys Res Commun 285: 724–727, 2001.
|
|
|
Ca2+ channels: IP3 and ryanodine receptors
|
|---|
The functioning of the pumps is strictly related to that of the ER Ca2+ channels. The multiplicity of the latter and their recognized heterogeneous distribution, with concentration at specific hot spots, appear to play fundamental roles in neuronal physiology. Both inositol 1,4,5-trisphosphate (IP3) and ryanodine (Ry) receptors are composed of specific subunits tetramerized in the ER membrane. Their state, however, is profoundly different. The three distinct IP3 receptor subunits (most abundant in the brain is subunit 2, followed by 3) can oligomerize with each other. This opens the way not only to homologous but also to heterologous tetrameric receptors (15). Ry receptors (also 3 types of subunits) are on the contrary always homologous. Type I is rare in the brain, except in the cerebellum and hippocampus. The most frequent and abundant is type II, which often coexists in the same neuron with the rarer type III.
The two types of receptors appear to be coexpressed within most neurons. Their ratio, however, varies considerably. In terms of expression levels, some differences are impressive. The high density of IP3 receptors (up to 200–500/µm2) in the ER of cerebellar Purkinje neurons is well known. In other neurons, for example cerebellar granules, density is hundreds of times lower (Fig. 3). Also in the latter case, however, the functional role of IP3 receptors remains considerable. With Ry receptors, expression heterogeneities among neurons are less impressive. Also in this case, however, high densities have been described, for example in the pyramidal neurons of the hippocampus. Within neurons, the distribution of the two types coincides only partially, with separate concentrations in distinct areas.
View larger version (114K):
[in this window]
[in a new window]
|
FIGURE 3. Inositol 1,4,5-trisphosphate (IP3) receptor and sarcoplasmic-ER Ca2+-ATPase (SERCA) immunofluorescence in the rat cerebellar cortex. Only Purkinje neurons (left), which are the richest cells, appear positive for the IP3 receptor, whereas all other cerebellar cells (neurons and glia) remain below threshold. In contrast, all cells are positive for SERCA (right), although the highest signals are still shown by Purkinje neurons and also by Golgi cells of the cellular layer. Reproduced with permission from The Journal of Cell Biology, 1991, vol. 113, p. 779–791 by copyright permission of the Rockefeller University Press.
|
|
Regulation of IP3 receptors is very complex and multifarious. In fact, they are phosphorylated by various protein kinases (A, C, G) and dephosphorylated by the protein phosphatase calcineurin. Interaction of the receptor with the latter is not random, because both of them participate in a sovramolecular complex together with the immunophilin-binding protein FKBP12 (3). Proteins of the FKB family are of importance also for the Ry receptor, which in addition appears to be regulated by nitric oxide.
At both receptors, the most important regulatory role is played by the specific activators IP3 and cADP ribose (cADPR) as well as by the direct binding of Ca2+. The previously reported inhibition of type 1 (not type 2 or 3) IP3 receptors by high concentrations of its ligand has been recently questioned (1). Thus the role of IP3 is probably always stimulatory but is dependent, however, on [Ca2+]i. Rises of [Ca2+]i from resting to stimulated values in fact increase the potency of IP3, making the receptor able to open at resting concentrations of the ligand. At the IP3 receptor, therefore, Ca2+ is not an inducer but rather an important modulator of channel opening. In the case of Ry receptors, on the other hand, regulation appears to be opposite. Opening responses are triggered by [Ca2+]i rises [i.e., these are true Ca2+-induced Ca2+ release (CICR) responses]. The role of the second messenger, cADPR, is modulation, dependent on its binding, either directly to the receptor or (more likely) to the associated FKB protein.
|
The ER lumen
|
|---|
At rest, a major fraction (~90%) of the total ER luminal Ca2+ is protein bound, and the proteins that take care of this function are of low affinity (Kd ~1 mM). The latter property is essential to ensure the rapid diffusion of the segregated Ca2+. In fact, at variance with the situation in the cytosol, where Ca2+-binding proteins are of high affinity, diffusion of the cation within the ER is not markedly slowed down by the luminal proteins. Therefore, within the network (but not in isolated cisternae) local depletions induced by receptor activation are quickly compensated by the flow of Ca2+ from adjacent areas of the system (4).
Among the proteins involved, calsequestrin, typically contained within the sarcoplasmic reticulum of striated muscle but expressed also by Purkinje neurons of volatiles (and possibly by other neurons as well), is able to condense at critical sites, where Ca2+ release channels are concentrated. Such condensation (due to the specific interaction of calsequestrin COOH- and NH2-terminal sequences) (6) is in contrast missing in other luminal proteins, which therefore appear widely distributed throughout the ER lumen. Such a localization is not static but results from the continuous recycling of these proteins from the Golgi complex, largely mediated by a receptor specific for their common COOH-terminal tetrapeptide Lys-Asp-Glu-Leu (KDEL). It should be noted, however, that in at least apparent contradiction with the proposed molecular homogeneity of the lumen, studies carried out in quick-frozen neurons have revealed considerable heterogeneities among ER cisternae, strongly suggesting incomplete luminal communication throughout the endomembrane system (18).
At variance with calsequestrin, whose function is apparently dependent on Ca2+ binding only (~50 ions/molecule), other luminal proteins carry out additional functions as well. Calreticulin, the most abundant in neurons and many other cell types, binds 30–50 ions at its COOH-terminal domain. In addition, with the rest of its sequence it works as a chaperone, playing a key role in specific steps of glycoprotein folding, and participates also in the regulation of other protein functions (7). Other chaperones, such as BiP and endoplasmin, are also Ca2+-binding proteins that participate in additional regulatory functions.
|
Physiology of neuronal Ca2+ stores
|
|---|
Many aspects of Ca2+ store functioning in neurons appear similar to those taking place in nonexcitable cells, yet others are peculiar, depending on specific properties such as the surface excitability and complex intracellular geometry. As an example, the stores were shown to play a major role in the dynamic [Ca2+]i processes, invading the neuron from either discrete sites of the plasmalemma (Ca2+ influx from the extracellular space) or intracellular pacemakers. Quite often, these processes were shown to run not at random but along specific intracellular pathways, giving rise to discrete and repetitive [Ca2+]i waves sustained by autoregenerative discharge. The key role of the stores in the latter processes is not difficult to envisage based on the [Ca2+]i dependence of IP3 and, especially, of Ry receptor activation, as previously described. Along the same line, the role of Ca2+ stores extends to events triggered at the surface by action potentials. As already discussed, the influx of Ca2+ triggered at the plasmalemma by depolarization and activation of voltage-sensitive Ca2+ channels would be unable to invade the depth of the cytoplasm due to the extensive Ca2+ buffering of the latter. Work in a variety of neurons, however, has clearly demonstrated that application of short trains or even of single electric stimuli can induce extensive discharge of Ca2+ from the stores, beginning from those located in the surface layers of the cytoplasm, with ensuing invasion of the whole cell (8).
The events described so far can have multiple consequences. First, they can induce local [Ca2+]i rises high enough (>5 µM) to trigger uptake of the cation by mitochondria, followed by release from the latter organelles during the next several seconds. Because of the low Ca2+ affinity of their uptake system, the mitochondria participating in these in-and-out processes can only be those located close to the discharging ER cisternae (17). Overall, the Ca2+ release events sustained by the store appear appropriate to 1) invade the whole cytoplasm; 2) be spatially and temporally heterogeneous, composed also by local high [Ca2+]i spikes; and 3) be more prolonged than those induced by surface channels only. In addition, the prolonged rises of [Ca2+]i can lead to the activation of surface, Ca2+-dependent K+ channels, inducing in neurons afterhyperpolarizations much stronger than those induced by Ca2+ influx only. This mechanism is responsible also for events in neurons, first described in the late 1980s, in which the involvement of Ca2+ stores was envisaged, such as the spontaneous miniature outward potentials. These events are in fact increased in frequency by the administration of caffeine, an activator of Ry receptors. Some of the intracellular events triggered by the coordinate surface + store release, for example nitric oxide generation, are of particular interest because they act not only in the stimulated neurons but also in the adjacent cells. Ca2+-induced signaling, therefore, takes place not only by transmitter release but also via local mechanisms.
|
Synaptic stores
|
|---|
Ca2+ stores are present both pre- and postsynaptically, and their importance at either side is progressively emerging. For quite some time, release of classic neurotransmitters via exocytosis of clear vesicles docked to the presynaptic membrane was believed to depend exclusively on Ca2+ influx. Only the opening of Ca2+ channels strategically positioned in the proximity of the fusion sites can in fact raise the local [Ca2+]i to levels (~100 µM) high enough for exocytosis, a low-affinity process, to take place. Recent experiments, however, have demonstrated a role of intraterminal stores in the course of the response. On a very short time scale (tenths of milliseconds), ER cisternae, which are located at some distance from the presynaptic membrane, accumulate some influxed Ca2+, thus contributing to Ca2+ buffering in the depth of the terminal (Fig. 2
). Shortly thereafter, the accumulated Ca2+ is released from the stores and works therefore to a prolongation and potentiation of the neurotransmitter release response (13). The role of the stores in two other processes occurring presynaptically appears to be even more important: the fusion of dense vesicles, which in the terminals are distributed at some distance from the plasmalemma and which require lower [Ca2+]i rises to be discharged (Fig. 4), and vesicle recycling, no matter whether by the rapid, kiss-and-run, or the slow, coated vesicle mechanism (5).
View larger version (16K):
[in this window]
[in a new window]
|
FIGURE 4. Capacitance assay of exocytotic responses induced in PC-12 cells by the photolysis of a caged Ca2+ compound [maximal cytosolic Ca2+ concentration ([Ca2+]i) rise ~70 µM]. The curve in A illustrates the time course of the response in a cell. Of the two peaks, the first is supported by clear vesicles (CV), the second by dense granules (DG). B: Ca2+ affinity and time dependence of the 2 peaks. CV fusions have a time constant of ~60 m and require [Ca2+]i > 30 µM; DG fusions occur at much slower rates (>10 s) and require much less [Ca2+]i (>3 µM). Reproduced with permission from Kasai et al., Proc Natl Acad Sci USA 96: 945–949, 1999.
|
|
Postsynaptically, the role of the stores also appears to be very important, especially in the numerous spines receiving glutamatergic input and exhibiting three-dimensional scaffolds, the postsynaptic densities (PSDs). The latter structures are no longer considered to be an inert cytoskeletal specialization. Rather, they are recognized to play a major role in the special distribution of the proteins involved in the coordinate signal transduction processes typical of synapses (11). ER cisternae do not appear to participate directly in the very rapid events initiated by glutamate via the activation of 
-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors. However, their IP3 receptors are directly connected to the PSDs via specific proteins named Homer and Shank (20). Because of this interaction, they are located in the ideal position to be activated by the IP3 generated by another class of glutamatergic receptors, the so-called metabotropic group I receptors, which are distributed all around the synaptic contacts. The function of the IP3-mediated Ca2+ release appears to be the establishment of a secondary, slower, and more persistent high [Ca2+]i phase. If alone, this slower phase would remain below threshold. When coordinated with glutamate-induced influx, however, it can induce synergistic responses likely to play fundamental roles in synaptic function.
|
Synaptic plasticity
|
|---|
Various forms of synaptic plasticity, including long-term potentiation and depression (LTP and LTD, respectively), triggered by glutamate stimulation at multiple areas of the brain, depend on the postsynaptic membrane expression of AMPA receptors, working coordinately with N-methyl-D-aspartate (NMDA) receptors. Unexpectedly, however, Ca2+ release from the stores appears to be involved with a modulatory role as well, which can vary substantially depending on the synapse and on the intensity of stimulation. The most impressive examples of this involvement have been shown at hippocampal and cerebellar synapses, where activation of IP3 receptors appears to be necessary for the establishment of LTD (14). A target of Ca2+ store-mediated events is calcium-calmodulin-dependent protein kinase II, an enzyme that, at glutamatergic synapses, is also associated with PSDs in the postsynaptic complex.
Together with postsynaptic responses, plasticity appears to involve changes of neurotransmitter release and possibly also of other presynaptic events. Information in this field is still intensely debated; however, the involvement of Ry receptor, with ensuing increase of the induced [Ca2+]i events, is apparently gaining momentum.
|
Conclusions
|
|---|
Together, the intracellular, rapidly exchanging Ca2+ store system can be envisaged as a comprehensive signaling network that extends nonrandomly, to the depth of the cell, the stimuli received from the surface. In neurons, which differ from most other cells because of their high degree of excitability, the stores operate as strict partners of surface Ca2+ channels, voltage as well as receptor operated. The neuronal surface system is unique for its impressive wealth of specific channels, peculiar in many respects, including excitability, coordination, inhibition, functional up- and downregulation, and heterogeneous distribution in the plasma membrane. Without the stores, however, the impact of surface channels would be limited to a thin, surface layer of the cell.
The stores are inhabitants of the cytosol, perfectly adjusted to its molecular composition and high Ca2+ buffering. This dual, messenger- and Ca2+-induced, excitability of IP3 and Ry receptors enables them to amplify the signals received from the plasmalemma and to master their evolution. In some cases, their heterogeneous distribution at the store surface is important for the generation of waves that can invade the whole cell; in others, it is important to restrict discrete events to small areas, such as single dendritic spines. In addition, high receptor density is necessary for low-affinity Ca2+-regulated events to be possible away from the cell surface. During a [Ca2+]i spike, the rapid uptake-release properties of the stores, often operating in collaboration with mitochondria, can first contribute to the intracellular Ca2+ buffering, then to the prolongation of [Ca2+]i signals.
On the basis of its autoregenerative signaling properties, the store system has been defined as a second excitable membrane system of the cell (2). This definition is largely correct, however, only if unique molecular and cellular properties of the stores are taken into proper account. Neurons, in fact, do not appear to possess two excitable membrane systems working in parallel to each other but a single, coordinate system located in both the plasmalemma and the whole cytoplasm, with profoundly different yet complementary properties in the two areas. It is just because of their complementarity that the two components of the single system are both indispensable and sufficient to warrant the appropriate functioning of cells articulate and bizarre, such as neurons.
|
Acknowledgments
|
|---|
The original work of this paper was supported by grants from Telethon (1118), Italian Ministry of Education and Research (MIUR), European Union (Growbeta), the Italian National Research Council (C0038AA), and the Armenise-Harvard Foundation. This work was carried out in the framework of the Italian MIUR Center of Excellence in Physiopathology of Cell Differentiation.
|
References
|
|---|
- Adkins CE and Taylor CW. Lateral inhibition of inositol 1,4,5-trisphosphate receptors by cytosolic Ca2+. Curr Biol 9: 1115–1118, 1999.[ISI][Medline]
- Berridge MJ. Neuronal calcium signaling. Neuron 21: 13–26, 1998.[ISI][Medline]
- Blaustein MP and Golovina VA. Structural complexity and functional diversity of endoplasmic reticulum Ca2+ stores. Trends Neurosci 24: 602–608, 2001.[ISI][Medline]
- Cameron AM, Steiner JP, Roskams AJ, Ali SM, Ronnett GV, and Snyder SH. Calcineurin associated with the inositol 1,4,5-trisphosphate receptor-FKBP12 complex modulates Ca2+ flux. Cell 83: 463–472, 1995.[ISI][Medline]
- Cancela JM, Gerasimenko OV, Gerasimenko JV, Tepikin AV, and Petersen OH. Two different but converging messenger pathways to intracellular Ca2+ release: the roles of nicotinic acid adenine dinucleotide phosphate, cyclic ADP-ribose and inositol trisphosphate. EMBO J 19: 2549–2557, 2000.[ISI][Medline]
- Coco S, Verderio C, De Camilli P, and Matteoli M. Calcium dependence of synaptic vesicle recycling before and after synaptogenesis. J Neurochem 71: 1987–1992, 1998.[ISI][Medline]
- Gatti G, Trifari S, Mesaeli N, Parker JM, Michalak M, and Meldolesi J. Head-to-tail oligomerization of calsequestrin: a novel mechanism for heterogeneous distribution of endoplasmic reticulum luminal proteins. J Cell Biol 154: 525–534, 2001.[Abstract/Free Full Text]
- Helenius A, Trombetta ES, Hebert DN, and Simons JF. Calnexin, calreticulin and the folding of glycoproteins. Trends Cell Biol 7: 193–200, 1997.[Medline]
- Jacobs JM and Meyer T. Control of action potential-induced Ca2+ signaling in the soma of hippocampal neurons by Ca2+ release from intracellular stores. J Neurosci 17: 4129–4135, 1997.[Abstract/Free Full Text]
- Jafri MS and Keizer J. On the roles of Ca2+ diffusion, Ca2+ buffers, and the endoplasmic reticulum in IP3-induced Ca2+ waves. Biophys J 69: 2139–2153, 1995.[Abstract/Free Full Text]
- John LM, Lechleiter JD, and Camacho P. Differential modulation of SERCA2 isoforms by calreticulin. J Cell Biol 142: 963–973, 1998.[Abstract/Free Full Text]
- Kennedy MB. Signal-processing machines at the postsynaptic density. Science 290: 750–754, 2000.[Abstract/Free Full Text]
- Meldolesi J. Rapidly exchanging Ca2+ stores in neurons: molecular, structural and functional properties. Prog Neurobiol 65: 309–338, 2001.[ISI][Medline]
- Meldolesi J and Grohovaz F. Total calcium ultrastructure: advances in excitable cells. Cell Calcium 30: 1–8, 2001.[Medline]
- Miyata M, Finch EA, Khiroug L, Hashimoto K, Hayasaka S, Oda SI, Inouye M, Takagishi Y, Augustine GJ, and Kano M. Local calcium release in dendritic spines required for long-term synaptic depression. Neuron 28: 233–244, 2000.[ISI][Medline]
- Monkawa T, Miyawaki A, Sugiyama T, Yoneshima H, Yamamoto-Hino M, Furuichi T, Saruta T, Hasegawa M, and Mikoshiba K. Heterotetrameric complex formation of inositol 1,4,5-trisphosphate receptor subunits. J Biol Chem 270: 14700–14704, 1995.[Abstract/Free Full Text]
- Pozzan T and Rizzuto R. High tide of calcium in mitochondria. Nat Cell Biol 2: E25–E27, 2000.[ISI][Medline]
- Pozzo-Miller LD, Pivovarova NB, Leapman RD, Buchanan RA, Reese TS, and Andrews SB. Activity-dependent calcium sequestration in dendrites of hippocampal neurons in brain slices. J Neurosci 17: 8729–8738, 1997.[Abstract/Free Full Text]
- Rizzuto R. Intracellular Ca2+ pools in neuronal signalling. Curr Opin Neurobiol 11: 306–311, 2001.[ISI][Medline]
- Tu JC, Xiao B, Naisbitt S, Yuan JP, Petralia RS, Brakeman P, Doan A, Aakalu VK, Lanahan AA, Sheng M, and Worley PF. Coupling of mGluR/Homer and PSD-95 complexes by the Shank family of postsynaptic density proteins. Neuron 23: 583–592, 1999.[ISI][Medline]
Top
Introduction
Cellular and molecular biology...
