Selective activation of neuronal functions by Ca2+ is determined by the kinetic profile of the intracellular calcium ([Ca2+]i) signal in addition to its amplitude. Concurrent electrophysiology and ratiometric calcium imaging were used to measure transmembrane Ca2+ current and the resulting rise and decay of [Ca2+]i in differentiated pheochromocytoma (PC12) cells. We show that equal amounts of Ca2+ entering through N-type and L-type voltage-gated Ca2+ channels result in significantly different [Ca2+]i temporal profiles. When the contribution of N-type channels was reduced by ω-conotoxin MVIIA treatment, a faster [Ca2+]i decay was observed. Conversely, when the contribution of L-type channels was reduced by nifedipine treatment, [Ca2+]i decay was slower. Potentiating L-type current with BayK8644, or inactivating N-type channels by shifting the holding potential to −40 mV, both resulted in a more rapid decay of [Ca2+]i. Channel-specific differences in [Ca2+]i decay rates were abolished by depleting intracellular Ca2+ stores with thapsigargin or by blocking ryanodine receptors with ryanodine, suggesting the involvement of Ca2+-induced Ca2+ release (CICR). Further support for involvement of CICR is provided by the demonstration that caffeine slowed [Ca2+]i decay while ryanodine at high concentrations increased the rate of [Ca2+]i decay. We conclude that Ca2+ entering through N-type channels is amplified by ryanodine receptor mediated CICR. Channel-specific activation of CICR provides a mechanism whereby the kinetics of intracellular Ca2+ leaves a fingerprint of the route of entry, potentially encoding the selective activation of a subset of Ca2+-sensitive processes within the neuron.
Depolarizing a neuron opens voltage-gated Ca2+ channels (VGCC), leading to an influx of Ca2+ ions into the cytoplasm, where Ca2+ sensitive signaling cascades are stimulated. Many neuronal functions, including neurotransmitter release, membrane excitability, gene expression, enzyme activity, cell growth, and apoptosis are sensitive to calcium (Berridge 1998). The kinetic profile of intracellular calcium concentration ([Ca2+]i), in conjunction with the colocalization of Ca2+-sensitive signaling proteins with particular ion channels (Marrion and Tavalin 1998; Sheng and Sala 2001) may help to explain how the ubiquitous calcium ion can selectively modulate this large array of neuronal functions (Chawla and Bading 2001; Dolmetsch et al. 1997). The [Ca2+]i profile following a depolarization is the sum of Ca2+ influx, Ca2+-induced Ca2+ release (CICR), buffering, and extrusion from the neuron.
In addition to the absolute levels of [Ca2+]i, the temporal characteristics of [Ca2+]i signals are critical in the integration of coincident signals underlying forms of synaptic plasticity such as long-term potentiation (LTP) and long-term depression (LTD). Mechanisms that render a neuron sensitive to the duration of a [Ca2+]i signal include the activation of CaMKII, which can become Ca2+ independent due to autophosphorylation when neighboring subunits are coincidentally complexed with Ca2+ bound calmodulin (Miller and Kennedy 1986). The temporal regulation of [Ca2+]i signals is also important in the activation of Ca2+-activated potassium channels that regulate the shape and frequency of action potentials and in the activity-dependent changes in gene transcription controlled by CREB phosphorylation. While the relationship between the [Ca2+]i transient and many neuronal functions has been established, it is not well understood how a neuron shapes the [Ca2+]i transient subsequent to depolarization.
This study utilizes concurrent electrophysiology and ratiometric calcium imaging to measure transmembrane Ca2+ current and the resulting rise and decay of [Ca2+]i in differentiated pheochromocytoma (PC12) cells. This combination of techniques allows fine control and monitoring of the amplitude and temporal characteristics of Ca2+ influx. Differentiated PC12 cells are a neuronal cell line previously used to study the specificity of Ca2+ signaling. Examples include the differential induction of gene transcription, regulated by the route of Ca2+ entry into the cell (West et al. 2001), and the presence of fast and slow Ca2+-dependent exocytosis, triggered by synaptotagmins with differing Ca2+ affinities (Sugita et al. 2002). Moreover, recent studies have explored the characteristics and distribution of ryanodine and IP3-mediated calcium stores in this cell line (Johenning et al. 2002). The PC12 cell line contains several VGCCs, the expression of which can be manipulated by nerve growth factor (NGF)–induced differentiation and that can be isolated pharmacologically and with different voltage protocols, allowing experimental manipulation of the route of entry of Ca2+ during a depolarizing pulse. Precedents for channel-specific linkage to CICR in neurons exist (Akita and Kuba 2000; Sandler and Barbara 1999; Usachev and Thayer 1997), although there was a predominance of N-type current over L-type in each case, leaving the possibility that it was the amount of calcium entering each channel type, not the route of entry that was critical. We show that Ca2+ influx through N-type channels is amplified by CICR from intracellular stores while Ca2+ entering through L-type channels does not lead to coupled CICR. Thus equal amounts of Ca2+ entering through these two channel types resulted in significantly different [Ca2+]i temporal profiles.
Patch-clamp studies were performed using the nystatin perforated patch technique with a Dagan 8900 patch clamp amplifier. Current signals were filtered at 3 kHz. Experimental protocols were controlled using Pclamp software (Axon Instruments). Electrodes were coated with Sylgard to reduce pipette capacitance and fire polished just before recording to a resistance of 4–6 MΩ. The patch pipette solution consisted of (in mM) 135 CsCl, 10 CaCl2, 1.2 MgCl2, 25 HEPES, and 10 glucose in the tip. The electrode was backfilled with the same solution, to which 200 μg/ml nystatin was added. After formation of a gigaseal, the series resistance was monitored to evaluate when perforation was complete and stable.
Microscope and perfusion system
An inverted Olympus IX70 microscope equipped with an oil immersion 40× objective lens was used to observe cells loaded on glass coverslips (22 × 22 mm) coated with poly-ornithine and laminin attached to a chamber with a bath volume of ∼50 μl (Warner Instrument, Hamden, CT). Solutions were gravity fed from syringes through an automated snap valve system (Automate Scientific, Oakland, CA) into a micro-manifold with a single output into the chamber. The bath was constantly perfused at a rate of 1 ml/min, providing rapid exchange of the bath solution.
Fluorescence images with excitation at 340 and 380 nm were recorded with an intensified CCD camera. The imaging system (Ionoptix) utilizes a high speed chopper mirror to alternate between wavelengths so that fast calcium events can be measured. Four images were averaged at each wavelength for each time point to improve the signal to noise ratio. The concentration of [Ca2+]i was calculated from the ratio of the fluorescence at two different wavelengths using the equation [Ca2+]i = Kd × (R − Rmin)/(Rmax − R) × β, where Kd is the dissociation constant for fura-2; Rmin and Rmax are the 340/380 nm (background subtracted) ratio for fura-2 free acid in 0 and 1 mM Ca2+, respectively; and β is the ratio (background subtracted) between fura-2 free acid in 0 and 1 mM Ca2+ at 380-nm excitation (Grynkiewicz et al. 1985).
Culturing of PC12 cells
PC12 cells were grown in Dulbecco's modified Eagle's medium (Sigma, St. Louis, MO) supplemented with 5% fetal calf serum (Sigma), 10% horse serum (JHR Biosciences, Lenexa, KS), 50 units/ml penicillin G (Sigma), and 50 mg/ml streptomycin (Sigma). A solution containing (in mM) 130 NaCl, 5 KCl, 2.2 CaCl2, 1 MgCl2, 25 HEPES, and 10 glucose was used for dye loading with 5 μM Fura-2-AM for 30 min at 37°C and perfusion of cells at a rate of 1 ml/min at room temperature (20°C). For recordings, the cells were switched to a perfusion solution containing (in mM) 65 TEA-Cl, 40 NaCl, 5 KCl, 20 CaCl2, 1 MgCl2, 25 HEPES, and 10 glucose.
The decay of [Ca2+]i was fit with linear growth curves on natural log values using an ANOVA for a mixed model using restricted estimation by maximal likelihood. The natural log transformation resulted in approximately linear functions with errors that approximate a normal distribution. A model of the natural log transformation with a single slope was compared with a model with different slopes for each condition. The assumption of normality was evaluated both by investigating the plots of variation and using the Kolmogorov-Smirnoff test. The ratios of [Ca2+]i divided by current (Fig. 5) were compared using a one-way ANOVA with post hoc comparisons using a Sheffe test.
Channel isolation by pharmacological blockade
The reproducibility of the current and [Ca2+]i profile in response to a voltage step (Fig. 1 A) within each cell allowed for the reliable comparison of signals between control and treatment conditions. To achieve reproducible elevations of [Ca2+]i within a cell, it was necessary to wait >180 s between stimulations, presumably reflecting the time for all Ca2+ sequestration processes to return to prestimulus states. Comparisons between [Ca2+]i following influx through different types of VGCC's requires the ability to manipulate and measure both electrical signal and digital imaging signal over a critical range of stimulation. We compared the time-integrated (area under the curve for the 20 s following depolarization) elevation of [Ca2+]i with the time-integrated Ca2+ charge flux. In all cases, the duration of measured [Ca2+]i elevation was at least two orders of magnitude longer than the duration of current flow. Ca2+ influx through voltage-gated channels was modulated in three ways: 1) varying the duration of the depolarizing pulse (Fig. 1B), 2) varying the amplitude of the depolarizing pulse (Fig. 1C), and 3) varying the extracellular Ca2+ concentration while using a constant pulse protocol (Fig. 1D). Each of these protocols produced a level of [Ca2+]i that varied linearly with the level of aggregate Ca2+ current. Work done in other neuronal systems (Hua et al. 1993; Stuenkel 1994) has shown that a nonlinear relationship between Ca2+ entry and [Ca2+]i can occur under heavy stimulation protocols, attributable to activation of a high-capacity, low-affinity buffer (presumably mitochondria) or under light Ca2+ loading, where a low-capacity (quickly saturated), high-affinity buffer (presumably cytoplasmic Ca2+ binding proteins) limits very small Ca2+ loads. The Ca2+ fluxes in experiments reported here, which induced proportional [Ca2+]i elevations (Fig. 1, B–D), did not produce [Ca2+]i levels at either of these extremes. The resting baseline [Ca2+]i was slightly elevated in patched cells compared with those not patched in the dish, likely due to mechanical disturbance by the electrode in 20 mM Ca2+ bath solution.
PC12 Cells treated with NGF (50 ng/ml) for 3–5 days were used since differentiation for this amount of time led to an upregulation of N-type channels, resulting in equivalent Ca2+ flux through N- and L-type Ca2+ channels during a voltage step (Liu et al. 1996; Usowicz et al. 1990). We chose to focus on the 20 s following the voltage step, where [Ca2+]i decay was most pronounced. Over this time period, the decay could be well fit by a first order exponential (Fig. 2, A and B). We choose to present the normalized data to make visual comparison of the kinetics for decaying calcium signals of differing sizes (control vs. pharmacological blockade) easier. Differing amounts of calcium entering each channel type could not have accounted for the differences in decay since the amount of calcium was controlled (see figure legends for average peak [Ca2+]i amplitudes). An advantage of the combined techniques of electrophysiology and Ca2+ imaging is that the amount of calcium entering the cell (Fig. 2, C and D) can be quantitated and compared with the [Ca2+]i profile (Fig. 2, A and B). Our data indicate that the relationship between Ca2+ influx and intracellular Ca2+ dynamics differed, dependent on route of entry. When the contribution of N-type channels is reduced by ω-conotoxin MVIIA treatment, the change in the integrated [Ca2+]i profile is greater than the change in integrated current (−38 ± 2% vs. –26 ± 4%, respectively), while the change of integrated [Ca2+]i profile matches the change in integrated current when the contribution of L-type channels is reduced by nifedipine (−17 ± 1% vs. −18 ± 3%, respectively). This difference in integrated [Ca2+]i signal resulted from a change in the decay rate as opposed to a change in the amplitude of the initial [Ca2+]i rise. The [Ca2+]i decay in the presence of ω-conotoxin MVIIA (−0.052 ± 0.002), described statistically by the slope of the natural log of [Ca2+]i during the 20 s following a depolarizing pulse, was more rapid (P < 0.001) than control (−0.033 ± 0.002; Fig. 2A). The decay following a depolarizing pulse in the presence of nifedipine (−0.053 ± 0.002) was slower (P = 0.036) than control (−0.061 ± 0.002; Fig. 2B). Thus decay is slower when there is a greater contribution of N-type channels versus L-type channels. Although we observed large between cell variability, the change produced by treatment was independent of the starting rate of decay, with 7 of 8 cells increasing their rate of decay following treatment with ω-conotoxin MVIIA and 10 of 10 cells decreasing their rate of decay following nifedipine treatment. The reduction of current and Ca2+ current kinetics are similar for N- and L-type influx (Fig. 2, C and D) and therefore neither account for differences in the [Ca2+]i profile. The use of either ω-conotoxin MVIIA or nifedipine to alter the contribution of N- or L-type channels to the whole cell Ca2+ current results in a decrease in Ca2+ influx compared with control conditions. We also performed experiments in which either the duration or the amplitude of the voltage pulse was increased, such that the influx in the presence of the blocking agents matched that seen in the absence of the agents. The difference in decay kinetics was unaltered (data not shown), indicating that the reduced influx did not play a role in the effects observed. PC12 cells contain VGCCs in addition to N- and L-type. The mRNA encoding three pore-forming α1 subunits (C, B, and A for L-, N-, and P/Q-types, respectively) and three auxiliary β subunits (1, 2, and 3) have been detected in PC12 cells (Liu et al. 1996). There was a minimal contribution of P/Q type voltage-gated Ca2+ channels, with 5% of the current blocked by 300 nM AgaIVA. Combining 300 nM AgaIVA with either ω-conotoxin MVIIA or nifedipine did not alter the results obtained in its absence, indicating that P/Q channels were not playing a role (data not shown). There was also a contribution from R-type channels, with 15% of the total current resistant to block by 10 μM cadmium. This resistant current could be eliminated by the addition of 25 μM nickel.
Channel isolation by pharmacological augmentation
If decay is slower when N-type current makes up more of the total current, we might expect that augmenting the L-type current would lead to faster decay of [Ca2+]i. To test this, Bay K8644 was used to potentiate L-type voltage-gated calcium currents, leading to a change in [Ca2+]i profile (14 ± 2%; Fig. 3 A) that was less than the change in the integrated current (18 ± 2%; Fig. 3C). This indeed resulted from a more rapid decay (P = 0.016) of [Ca2+]i in the presence of 5 μM Bay K8644 (−0.067 ± 0.003; 5-min exposure) than in its absence (−0.055 ± 0.003; Fig. 3A).
Channel isolation by voltage protocol
To confirm the channel specificity of the difference in [Ca2+]i decay independently of pharmacological manipulation, [Ca2+]i decay was measured after voltage steps from holding potentials (Vh) of either −40 or −90 mV (Fig. 3B). Since N-type calcium channels are largely inactivated at −40 mV, we would predict that decay would be faster from a Vh of −40 mV, where the contribution of N-type current is minimal. The difference in the integrated intracellular profile (−32 ± 3%; Fig. 3B) between the holding potentials was greater than the change in integrated current (−24 ± 1%; Fig. 3D). This resulted from a [Ca2+]i decay from a Vh of –40 mV (−0.052 ± 0.001) that was faster (P < 0.001) than observed from a Vh of –90 mV (−0.040 ± 0.001; Fig. 3B), confirming the results obtained with pharmacological current isolation.
CICR is critical for differential effects
Caffeine induces Ca2+ release from ryanodine-gated stores in the endoplasmic reticulum in NGF differentiated PC12 cells (Fasolato et al. 1991; Koizumi et al. 1999; Zacchetti et al. 1991). To explore the contribution of CICR to the slower decay of [Ca2+]i when influx occurs through N-type channels, we examined whether the differences seen in the presence of ω-conotoxin MVIIA and nifedipine were maintained when CICR was blocked. Channel-specific differences in [Ca2+]i decay rates were abolished by depleting intracellular Ca2+ stores by pretreating cells for 60 s with 10 μM thapsigargin (Fig. 4, A and B) or by blocking ryanodine receptors with 100 μM ryanodine (Fig. 4, C and D). When the cells were pretreated with thapsigargin, the [Ca2+]i decay was the same in both the presence (−0.035 ± 0.007) and absence (−0.030 ± 0.003) of ω-conotoxin MVIIA (Fig 4A) and the same in both the presence (−0.028 ± 0.001) and absence of nifedipine (−0.027 ± 0.001; Fig. 4B). When the cells were treated with ryanodine, the [Ca2+]i decay was the same in both the presence (−0.069 ± 0.003) and absence (−0.071 ± 0.002) of ω-conotoxin MVIIA (Fig. 4C) and the same in both the presence (−0.072 ± 0.00) and absence (−0.076 ± 0.002) of nifedipine (Fig. 4D). We conclude that the Ca2+ entering through N-type channels is amplified by ryanodine receptor-mediated CICR. To confirm the role of CICR in the decay of [Ca2+]i following a depolarizing pulse, we used 5 mM caffeine to potentiate CICR (Fig. 4E) and 100 μM ryanodine to inhibit CICR (Fig. 4F). The [Ca2+]i decay rate was slowed (P = 0.032) in the presence of caffeine (−0.052 ± 0.02) compared with control (−0.059 ± 0.01) and increased (P = 0.018) in the presence of ryanodine (−0.094 ± 0.03) compared with control (−0.083 ± 0.02). Neither caffeine nor ryanodine significantly altered the baseline or peak [Ca2+]i values.
Stimulation size needed to induce CICR is altered by biasing the channel type through which Ca2+ enters
Graded amplification of [Ca2+]i by CICR is commonly observed in neurons (Hua et al. 1993; Kostyuk and Verkhratsky 1994), although under some circumstances (Usachev and Thayer 1997), a threshold for inducing CICR can be shown, revealing the potential for regenerative release from intracellular stores. We therefore investigated whether the Ca2+ channel class influenced the relationship between Ca2+ entry and [Ca2+]i across a wide range of stimulation voltages. In control cells, the amplification of [Ca2+]i appears graded, as seen in Fig. 5 C, by a gradual increase in the ratio of the integrated signals for [Ca2+]i for 20 s following stimulus divided by Ca2+ charge (Q). ANOVA indicated that the [Ca2+]i/Q ratio differed significantly (P = 0.008) as a function of voltage step amplitude. Sheffe tests were used for between-group comparisons of the effect of voltage step amplitude, with the [Ca2+]i/Q ratio becoming significantly different from the step to –10 mV at voltage steps to +10 (P = 0.023) and +15 mV (P = 0.017). A comparison of the pattern of [Ca2+]i generated by test pulses of increasing magnitude in the presence of either ω-conotoxin MVIIA (Fig. 5A) or nifedipine (Fig. 5B) reveals amplified [Ca2+]i signals once a certain stimulation size is achieved when the influx is biased through N-type channels. This can be seen in Fig. 5D by a sudden jump in the ratio of [Ca2+]i/Q between stimulations to –5 and 0 mV when influx occurs predominately through N-type channels (nifedipine). ANOVA indicated that the [Ca2+]i/Q ratio differed significantly (P = 0.0006) as a function of voltage step amplitude. Sheffe tests were used for between-group comparisons of the effect of voltage step amplitude, with the [Ca2+]i/Q ratio becoming significantly different from the step to –10 mV at voltage steps to 0 (P = 0.001), +5 (P = 0.002), +10 (P = 0.027), and +15 mV (P = 0.027). This amplification is eliminated when influx occurs predominately through L-type channels (ω-conotoxin MVIIA; Fig. 5E), with the mean [Ca2+]i/Q ratio not different (P = 0.7) across voltages.
The use of coupled patch-clamp and Ca2+ imaging techniques allowed the examination of intracellular Ca2+ dynamics after well-controlled and monitored Ca2+ entry through different VGCC types. We show that Ca2+ influx through different classes of VGCC in PC12 cells produces Ca2+ elevations with differing decay kinetics, resulting from the selective activation of CICR by N-type current. Current flux through L-type versus N-type channels was accomplished by both pharmacological (use of channel blockers and facilitators) and voltage protocols. We further demonstrated that thapsigargin and ryanodine eliminate differences in the intracellular profile between Ca2+ channel classes, suggesting CICR mediated by ryanodine receptors as the mechanism. Modulation of the rate of [Ca2+]i decay by caffeine and ryanodine directly demonstrate that CICR shapes the [Ca2+]i signal. A role for Ca2+ released from ryanodine receptors in the ER of neurons has been established for shaping of neuronal [Ca2+]i transients (Friel and Tsien 1992; Garaschuk et al. 1997; Hua et al. 1993; Kano et al. 1995; Lipscombe et al. 1988a; Llano et al. 1994; Shmigol et al. 1995; Solovyova et al. 2002), and we have demonstrated that this shaping of the [Ca2+]i transient can be determined by the class of VGCC utilized. Channel-specific activation of CICR provides a mechanism whereby the kinetics of intracellular Ca2+ leaves a fingerprint of the route of entry, potentially encoding the selective activation of a subset of Ca2+-sensitive targets and processes within the neuron. As an example, activation of D1 dopamine receptors on rat neostriatal neurons decreases N-type and increases L-type Ca2+ currents (Surmeier et al. 1995), which might regulate Ca2+-sensitive processes through an effect on the kinetics of the [Ca2+]i transient.
It is interesting that the memory of route of entry persists for so many seconds beyond the relatively short duration of channel opening, since we might expect that diffusion of Ca2+ would blur the initial segregation of the ion. However, this may be explained by the concentration dependency of the initial CICR activation that then continues to regeneratively release Ca2+ from ryanodine receptors well beyond the depolarization. The initiation of CICR may occur within a microdomain surrounding the pore of an N-type VGCC where the [Ca2+]i would become sufficiently high to activate a co-localized ryanodine receptor. Localized elevations of Ca2+ in the micromolar range are required to activate ryanodine receptors (Fill and Copello 2002), which indeed occurs in the vicinity of VGCCs (Narita et al. 2000). The persistent Ca2+ signal, which is in the nanomolar concentration range, would not impact CICR. Selective activation of CICR is possible since only channels co-localized with ryanodine receptors, and neither Ca2+ entering more distant channels nor residual Ca2+ would create the requisite concentration for ryanodine channel activation. A precedence for functional coupling by co-localization lies in the finding that ryanodine receptors form a functional triad with N-type Ca2+ channels and BK channels in bullfrog sympathetic neurons (Akita and Kuba 2000). An example of the persistence of CICR in neurons over the time course of seconds can be found in the rat visual cortex where a late phase [Ca2+]i increase reflecting CICR lasts many seconds (Kato et al. 1999).
The coupling of a specific class of VGCC with ryanodine receptors could occur either by nonhomogeneous distribution of each in specific regions of a cell or by colocalization within microdomains throughout a cell. A slower decay in the neurites would be predicted if both N-type channels and ryanodine receptors were more highly expressed in this region. In neurites of differentiated PC12 cells, both a predominance of N-type current (Reber and Reuter 1991) and preferential occurrence of elementary Ca2+ release from ryanodine receptors in response to caffeine have been shown (Koizumi et al. 1999). We favor colocalization within microdomains throughout a cell, because we were unable to detect a slower decay of [Ca2+]i in the neurites than in the cell bodies (data not shown), consistent with the finding of ryanodine receptors types 2 and 3 distributed throughout the cytoplasm of differentiated PC12 cells (Johenning et al. 2002). L-type and N-type Ca2+ channels appear concentrated in local hot spots in frog sympathetic neurons, sometimes dominated by one channel type (Lipscombe et al. 1988b). Subsurface cisterns, extensions of the endoplasmic reticulum containing ryanodine receptors, exist in close apposition to the cell membrane (Berridge 1998), allowing the colocalization necessary for functional coupling.
Our ω-conotoxin MVIIA and nifedipine treatments did not fully isolate each channel type, but rather shifted the contribution of N- versus L-type channels during influx. In all cases the influx occurs through a mixed population of VGCC's, but with a predominance of current flowing through L-type channels in the presence of ω-conotoxin MVIIA and through N-type channels in the presence of nifedipine. A small contribution from R-type or P/Q type current cannot be ruled out. Our interpretations are strengthened by the correlative data from BayK and voltage protocols, which do not depend on channel blockade. Although a functional coupling of N-type channels to CICR can explain the differential shaping of the [Ca2+]i profile, the possibility of coupling to other processes, such as extrusion by Ca2+-ATPase or store-operated capacitative Ca2+ entry, also exists.
Although an all or none release of calcium can be demonstrated in some neurons when ryanodine receptors are sensitized by caffeine (Usachev and Thayer 1997), it is more common for CICR to be graded with increasing stimulus strength (Hua et al. 1993; Kostyuk and Verkhratsky 1994). Our data indicate that, under normal conditions, the amplification of Ca2+ influx by CICR varies in a graded fashion with stimulation size. The presence of an apparent threshold when calcium enters through a channel type that is privileged in its ability to trigger CICR (Fig. 5D) indicates that CICR has regenerative capacity in neurons. Additionally, smaller depolarizations may more successfully activate Ca2+ sensitive cascades depending on the route of entry due to the channel specific amplification by CICR.
CICR is directly involved in many neuronal functions, such as modulating firing patterns by altering the afterhyperpolarization (Akita and Kuba 2000), promoting synaptic plasticity by integrating coincident inputs with residual Ca2+ following stimulation (Svoboda and Mainen 1999), by mediating neurotransmitter release (Emptage et al. 2001; Narita et al. 2000; Smith and Cunnane 1996), and by altering gene expression through mechanisms such as the induction of specific forms of phospho-CREB (Deisseroth and Tsien 2002). Genes whose transcription is mediated by CREB phosphorylation can show expression patterns that reflect the temporal features of Ca2+ transients (Bito et al. 1996; Curtis and Finkbeiner 1999). The same amount of Ca2+ entering through different VGCC's have been reported to selectively modulate release of vasopressin and oxytocin in preparations from the rat neurohypophysis (von Spreckelsen et al. 1990; Wang et al. 1997, 1999) and acetylcholine release in rat superior cervical ganglion (Gonzalez Burgos et al. 1995). CICR can contribute to Ca2+ signals triggered by a single action potential in some neurons (Sandler and Barbara 1999). The regenerative release of Ca2+ from ryanodine receptors can stimulate processes locally in the vicinity of ryanodine channels and also alter the duration of the global cytosolic Ca2+ rise. The linkage between CICR and a specific class of VGCC within a neuron couples discrete Ca2+ activated processes with the route of Ca2+ entry.
This work was supported by National Institute on Alcohol Abuse and Alcoholism Grant AA-08003 to S. N. Treistman and predoctoral fellowship AA-05552 to K. Tully.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 by the American Physiological Society