Untangling Neuronal Calcium Signaling

From the very moment of conception, calcium plays a pivotal role in fetal development. It rushes in as a wave around the egg to herald the sperm's arrival, binding to proteins that help kick off the whole developmental process. From this first influx, calcium continues to play a critical role in how the body's cells respond to outside signals. Calcium tells muscles to contract and nerves to release neurotransmitters, and is at least part of the signal that helps people form and retain memories. Calcium's role comes full circle with its involvement in cell death. What's so mind-boggling about calcium signaling is its diversity. Not only does calcium play different roles in different cell types, it can play various roles within a single cell depending on both how it enters the cell and how much does so. In neuronal cells, calcium may trigger neurotransmitter release, ion channel gating, kinase activation, nitric oxide (NO) production, and gene transcription. It also helps the cell retain a memory of recent action potentials, which can be important for neuronal plasticity. How calcium activates these diverse responses—and how the responses remain distinct—is the subject of intense study. Scientists have at their disposal a wide range of new and improved tools to monitor and manipulate intracellular calcium levels. Using these tools, they have found that calcium influxes are local events, affecting those channels and receptors that reside nearby. It is the magnitude of the calcium influx, the nature of the channel through which it enters the cell, and the set of proteins associated with that channel, that seem to determine the influx's ultimate downstream response. Regulating Calcium When calcium enters a cell it quickly binds to one of many calcium-binding proteins, the most notable of which is calmodulin (CaM), a 16-kD protein with two pairs of calcium-binding EF hands.2 The calcium/CaM complex activates kinases throughout the cell and can enter the nucleus, where it plays a role in activating gene transcription. Free calcium is quickly pumped out of the cell or into storage in the endoplasmic reticulum (ER) via an ATPase.1,2 The ER plays a critical role in calcium regulation. It is therefore not surprising that it closely follows the cell membrane.1 In some regions the ER consists of multilayered structures called cisternal organelles, located 20 to 80 nm from the cell membrane. In the neuronal synapse the ER is also close to both mitochondria and neurotransmitter-containing vesicles. This positioning allows the ER to act as both a source and a sink for calcium, and to play an active role in regulating its cytoplasmic concentration. Calcium has two major entry routes into the cell.1,2 In the first, an action potential depolarizes the cell membrane, causing voltage-gated calcium channels (VGCC) to open and allowing the ion to enter the cell. The second mode of entry involves ligand-gated channels. When ligands such as glutamate bind receptors on the cell membrane, the receptors change shape. Calcium then enters the cell, inducing a signal transduction cascade that opens calcium channels in the ER. Thus, this second pathway produces a calcium-induced calcium release. The first step in this cascade is the activation of phospholipase C (PLC) by G-proteins. PLC then hydrolyzes phosphatidylinositol 4,5-bisphosphate in the cell membrane into the second messengers inositol-1,4,5-triphosphate (IP3) and diacylglycerol.1,2 IP3 binds to its receptor on the ER (IP3R), which opens to release stored calcium into the cytoplasm. Calcium itself can also induce the IP3R. Low calcium levels make the receptor more sensitive to IP3, and vice versa, enabling the channel to detect the overlap of two weak signals, neither one of which could open the channel on its own.1 One result of this calcium-induced calcium release is that calcium exiting the ER through one channel produces a localized rise in calcium concentration that can trigger a neighboring IP3R to open. This signal can propagate down the ER, carrying its calcium message to other parts of the neuron.1,2 The IP3R is sensitive not only to cytoplasmic calcium concentration and IP3; it also responds to ER calcium levels:1 The channel becomes more sensitive to external calcium or ligands when internal calcium stores are high. This may help cells "remember" when many action potentials have swept through the neuron.1,2 With each action potential that stimulates a calcium influx through the VGCCs, ATPases shuttle calcium into the ER. Several of these action potentials in succession load calcium in the ER and increase the IP3R's sensitivity until eventually, even a small calcium influx from the VGCC causes the IP3R to open, producing a greater increase in calcium concentration than the VGCC alone could. Using the ER as a memory device might regulate neurotransmitter release. Local calcium increases due to VGCCs in the synapse cause exocytosis of small synaptic vesicles located near the plasma membrane. These vesicles contain amino acids such as GABA or glutamate. However, neuropeptide-containing vesicles, located farther from the cell membrane, are not as sensitive to a localized calcium influx through the VGCC. Instead, evidence suggests that these vesicles are released only after several action potentials in succession load the ER to such a point that it releases its calcium stores.1-3 In addition to the IP3R, the ER supports three other calcium channels.1 Like the IP3R, the ryanodine receptor (RyR) is calcium-inducible and can, with the IP3R, establish calcium waves or release ER calcium stores in response to a VGCC-induced calcium influx. And like the IP3R, the RyR is sensitive to the ER calcium concentration. The RyR is named after the reagent ryanodine, which blocks the channel open; the receptor's normal, biological ligand is the second messenger, cyclic ADP ribose. Neither of the two other ER-localized calcium channels, the nicotinic acid dinucleotide phosphate (NADP) receptor and the sphingolipid calcium release-mediating protein on the ER (SCaMPER), is well characterized. Nuclear Signaling Each class of membrane-bound calcium channels plays a unique role in calcium signaling. L-type VGCCs are primarily located around the cell soma and dendrites and are involved in nuclear signaling. P/Q-type channels at the synapse are involved in neurotransmitter release, while the N-methyl-D-aspartate (NMDA)-type glutamate receptor plays a critical role in long-term potentiation, which is associated with learning and memory. Other channels, in conjunction with ER-resident channels, regulate gene expression, NO production, and neurotransmitter release. Courtesy of Upstate Biotechnology Inc.Got the Message? Immunohistochemical analysis of paraffin-embedded rat brain stained with Upstate Biotechnology's anti-mGluR1 (06-310). The question is how a cell keeps these signaling pathways distinct. The answer, according to Richard Tsien, a professor at Stanford University's Beckman Institute, is that the individual channels don't produce a global rise in calcium but rather a small, localized calcium bubble. "Calcium in the cell is not like water in a bathtub," Tsien says. Calcium coming in through an L-type VGCC near the soma, for example, won't activate a response from a P/Q-type channel at the synapse. Instead, he says that localized changes in calcium concentration near the channel itself are necessary for most calcium signaling to occur. Tsien's lab demonstrated that the membrane-proximal calcium concentration regulates nuclear signaling by showing that L-type channel-induced nuclear signaling was affected by a fast buffer that began binding calcium 0.1 mm from the membrane, but not by the use of a slow buffer that diffuses 1 mm into the cell before mopping up calcium. "That implied that the signal had to be really, really close to the channel," Tsien concludes. A previous discovery from Tsien's lab helps explain one aspect of what occurs at the channel entrance.4 His group found that the L-type calcium channel *1C subunit is bound to CaM in the subunit's C-terminal region. When calcium enters the channel in response to an action potential, it binds to CaM, which then undergoes a conformational change and interacts with the channel's IQ domain. This interaction inactivates the channel so that it no longer allows calcium into the cell. "L-type channels, through this calmodulin IQ signaling, are making a temporarily local signal," Tsien says. Michael Greenberg's lab at Harvard Medical School tied this channel inactivation to nuclear signaling and changes in gene expression. "The real issue in [calcium- induced] gene expression is how you achieve specificity," Greenberg says. He has long speculated that proteins bound to the cytosolic region of calcium channels give each channel its specificity, a theory supported by the lab's recent work. In a recent paper, Greenberg's lab tied calcium passing through the L-type channel to MAPK pathway activation and cAMP response element-binding protein (CREB) phosphorylation.5 CREB is a transcription factor that drives transcription of several genes involved in neuronal development and plasticity, including c-Fos, brain-derived neurotrophic factor (BDNF), and Bcl-2. In this study, Greenberg and colleagues used dihydropyridine (DHP), which specifically keeps L-type channels closed. In doing so, they also blocked CREB's long-term phosphorylation. However, DHP did not significantly decrease the bulk rise in calcium within the cell, indicating that it is the calcium passing through the L-type channel rather than the net rise in calcium levels that activates CREB phosphorylation. Greenberg and his lab then inserted a DHP-insensitive L-type channel into neurons. As expected, this new channel prevented DHP from blocking long-term CREB phosphorylation in response to membrane depolarization. But when the researchers added a second mutation to the DHP-insensitive channel, which prevents CaM binding, they found that calcium passage through this channel could no longer trigger long-term CREB phosphorylation, indicating that CREB phosphorylation depends upon CaM binding to the L-type channel. The Missing Link? The missing link in this pathway is how the calcium/CaM complex triggers the pathway leading to CREB phosphorylation. Both Greenberg and Tsien speculate that further proteins must be located near the channel entrance. "We need to figure out what it is and how it is induced when calcium binds the calmodulin," Greenberg says. Paul Mermelstein, a postdoctoral fellow from Tsien's lab and now an assistant professor at the University of Minnesota, may have already identified a link between CaM and the MAPK pathway. He and Cornelia Kurschner, an assistant member of the developmental neurobiology faculty at St. Jude Children's Research Hospital in Memphis, Tenn., found that the L-type channels apparently bind the PDZ proteins CIPP and NIL-16.6 NIL-16 has several PDZ domains, which anchor channels in place on the cell membrane. One of the most well-known PDZ proteins, PSD-95, aggregates the NMDA-type calcium channels near protein kinases.7 Mermelstein says that like PSD-95, "NIL-16 is not only interacting with the L-type channel but is bringing it in proximity to other signaling proteins." He doesn't know what these proteins are, but he suspects they are CaM and the kinases that are immediately downstream of the CaM/calcium complex. Although each of these studies has focused on CREB phosphorylation, this event is just one effect of calcium entry through one particular type of channel. There are other pathways as well, and they sometimes converge. For example, Greenberg is finding that some gene activation requires the integration of several different signals.8 His students have found that the BDNF promoter contains three distinct binding sites, one of which binds CREB. Greenberg says the other two also bind calcium-regulated transcription factors, but it remains to be seen which pathways activate them. "One of these is a totally new transcription factor—all motifs are novel," he says. "That's surprising in this day and age." Greenberg speculates that the pathways emanating from other channels will follow a model much like the one he identified, with a unique subset of CaM and kinases aggregated around that channel. This could explain how each channel controls a different set of responses. "There is a lot of tantalizing data that these complexes will be the mediators to the different responses to calcium," Greenberg says. Working with Calcium Studying calcium's movements throughout the cell depends on the ability to manhandle the individual receptors. Researchers can use ionophores, such as ionomycin and A23187, to shuttle calcium and other ions through the cell membrane. Reagents that block individual channels open or closed also exist. These products can selectively prevent channels from opening, to investigate the effects of other channels, or from closing, to flood the cell with calcium. Courtesy of V. Lev-Ram, R. Hilal-Dandan, A. Miyawaki, L. Brunton, and R.Y. TsienCardiac Cameleon: A spontaneously beating single heart muscle cell from a transgenic mouse expressing a "cameleon." The upper panel shows the cell at the most relaxed phase of its beat (diastole), the lower panel shows the same cell at peak contraction (systole). The green vs. red pseudocolors indicate low vs. high intracellular Ca2+. Researchers can examine calcium's movement throughout the cell using one of many fluorescent calcium indicators, which can be visualized using fluorescence microscopy, flow cytometry, or fluorescence spectroscopy. These indicators are primarily derivatives of calcium chelators like EGTA and APTRA, whose fluorescent properties change when bound to calcium. The original indicators were excited by UV light, but UV produces extensive background fluorescence and can generate free radicals that damage the cell, says Jianwei Ho, president of Hayward, Calif.-based Biotium. "Indicators with a longer wavelength means excitation is less harmful," says Ho. Newer indicators such as fluro-3 and rhod-2 respond to a 488-nm argon laser, which induces very little cellular damage. Indicators now exist with a range of affinities for calcium, and which have excitation and emission spectra ranging from the original blue to green to orange and red. Biotium, Molecular Probes of Eugene, Ore., and many other companies sell fluorescent indicators as both salts and esters. The cell-permeable esters can be passively loaded and are useful for experiments involving large numbers of cells, but whichever fluorescent indicator a researcher uses, it must be introduced into the cell. This can be achieved using microinjection, electroporation, patch-pipette perfusion, or with Molecular Probes' Influx™ pinocytic cell-loading reagent. All of these techniques cause some disruption to the cell. However, says Iain Johnson, Molecular Probes' product manager for calcium indicators, "They have been in routine use now for 15 years so the nature of problem is well understood. With an understanding you can avoid those issues." Some new indicators developed by the University of San Diego's Roger Tsien can help researchers overcome the problems inherent in introducing indicators into the cell.9 His lab has developed indicators based on green fluorescent protein (GFP) variants and CaM that can be used to transform an interesting cell population. He has tethered two different GFP variants to either side of CaM and a CaM-binding peptide to create these indicators, which he calls "cameleons." When calcium binds to CaM, the latter binds the adjoining peptide. The resulting conformational change brings the two fluorescent proteins into closer proximity so that the emission from one protein can excite the second protein through fluorescence resonance energy transfer (FRET). Researchers detect calcium's presence by exciting the cyan variant of GFP and then looking for increased yellow emission. Researchers can express the cameleon transgene in many cell types and can include tags to localize the indicator to the ER, cytosol, or nucleus. Other tools for studying calcium include "caging" reagents. For instance, researchers who require a sudden, controlled increase in calcium concentration can use caged calcium, which contains chelators that bind calcium until stimulated by UV light. The reverse product is also available: Diazo-2 quickly depletes a cell of calcium in response to 360-nm light. Finally, Molecular Probes offers caged ionophores, which release the ionophore in response to UV light. From conception to death, calcium has a hand in many biological processes. In neurons alone, the ion mediates a dizzying array of roles. Aided by new and improved tools, researchers continue to untangle calcium's signaling circuitry, but there is no shortage of work. As Harvard's Greenberg points out, these studies must all be linked to actual biological processes: "What's important is how this plays out in real neural circuits and in mature animals." Amy Adams (amya@nasw.org) is a freelance writer in Mountain View, Calif. References 1. M.J. Berridge, " Neuronal calcium signaling," Neuron, 21:13-26, 1998. 2. M.D. Bootman et al., "The organisation and functions of local Ca2+ signals," Journal of Cell Science, 114:2213-22, June 2001. 3. Y. Peng, "Ryanodine-sensitive component of calcium transients evoked by nerve firing at presynaptic nerve terminals," Journal of Neuroscience, 16:6703-12, 1996. 4. R.D. Zuhlke et al., "Calmodulin supports both inactivation and facilitation of L-type calcium channels," Nature, 399:105, 107-8, 1999. 5. I.R. Dolmetsch et al., "Signaling to the nucleus by an L-type calcium channel-calmodulin complex through the MAP kinase pathway," Science, 294:333-9, Oct.12, 2001. 6. C. Kurschner, M. Yuzaki, " Neuronal interleukin-16 (NIL-16): A dual function PDZ domain protein," Journal of Neuroscience, 19:7770-80, 1999. 7. T. Tezuka et al., "PSD-95 promotes Fyn-mediated tyrosine phosphorylation of the N-methyl-D-aspartate receptor subunit NR2A," Proceedings of the National Academy of Sciences (PNAS), 96:435-40, 1999. 8. A.E. West et al., "Calcium regulation of neuronal gene expression," PNAS, 98:11024-31, Sept. 25, 2001. 9. A. Miyawaki et al., "Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin," Nature, 388:882-7, 1997. Supplemental Materials Calcium Research Tools

Untangling Neuronal Calcium Signaling

Interested in reading more?

Become a Member of