SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  • 1
    Potentiation of calcium-activated non-selective cation (CAN) channels was studied in rat hippocampal neurones. CAN channels were activated by IP3-dependent Ca2+ release following metabotropic glutamate receptor (mGluR) stimulation either by Schaffer collateral input to CA1 neurones in brain slices in which ionotropic glutamate and GABAA receptors, K+ channels, and the Na+-Ca2+ exchanger were blocked or by application of the mGluR antagonist ACPD in cultured hippocampal neurones.
  • 2
    The CAN channel-dependent depolarization (ΔVCAN) was potentiated when [Ca2+]i was increased in neurones impaled with Ca2+-containing microelectrodes.
  • 3
    Fura-2 measurements revealed a biphasic increase in [Ca2+]i when 200 μm ACPD was bath applied to cultured hippocampal neurones. This increase was greatly attenuated in the presence of Cd2+.
  • 4
    Thapsigargin (1 μm) caused marked potentiation of ΔVCAN in CA1 neurones in the slices and of the CAN current (ICAN) measured in whole cell-clamped cultured hippocampal neurones.
  • 5
    Ryanodine (20 μm) also led to a potentiation of ΔVCAN while neurones pretreated with 100 μm dantrolene failed to show potentiation of ΔVCAN when impaled with Ca2+-containing microelectrodes.
  • 6
    The mitochondrial oxidative phosphorylation uncoupler carbonyl cyanide m-chlorophenyl hydrazone (2 μm) also caused a potentiation of ΔVCAN.
  • 7
    CAN channels are subject to considerable potentiation following an increase in [Ca2+]i due to Ca2+ release from IP3-sensitive, Ca2+-sensitive, or mitochondrial Ca2+ stores. This ICAN potentiation may play a crucial role in the ‘amplification’ phase of excitotoxicity.

Calcium-activated non-selective (CAN) channels are found in a variety of cells (Partridge & Swandulla, 1988) including many types of neurones (Partridge et al. 1994). The first observation of ICAN in a mammalian cortical neurone was published in 1990 (Hasuo et al. 1990) and CAN channels have since been shown to play important roles in many types of CNS neurones. These channels open in the presence of cytoplasmic Ca2+ and the inward current through them is carried largely by Na+, but can include a component carried by Ca2+ (Poronnik et al. 1991; Partridge et al. 1994). CAN channels are unique in their capacity for positive feedback since they do not inactivate and, in fact, can be further activated either by their own Ca2+ influx or by Ca2+ influx through voltage-dependent calcium channels opened by ICAN-dependent depolarization (Tatsumi & Katayama, 1994). CAN currents underlie such slow depolarizing processes as bursting (Partridge & Swandulla, 1987; Raggenbass et al. 1997), slow after-depolarization (Hasuo et al. 1990) and plateau potentials (Rekling & Feldman, 1997). There is mounting evidence that these channels also play an important role in pathological conditions of cytoplasmic Ca2+ overload (Siesjo & Bengtsson, 1989).

Stimulation of metabotropic glutamate receptors (mGluRs) is an effective means of increasing [Ca2+]i in CA1 neurones (Shirasaki et al. 1994; Jaffe & Brown, 1994) and this increase in [Ca2+]i can activate Ca2+-activated currents including ICAN (Crepel et al. 1994; Congar et al. 1997). Hippocampal CA1 neurones express mainly the group I (mGluR5) type of mGluRs, which are located perisynaptically where they are activated predominantly by high frequency repetitive synaptic inputs (Lujan et al. 1996). Under conditions where most other membrane channels, including ionotropic receptors and calcium-activated potassium channels, are pharmacologically blocked, stimulation of mGluRs activates a slow inward current. The following observations were used to establish this inward current as ICAN. (1) Identical currents are activated by application of the mGluR agonist (±)-1-aminocyclopentane-trans-1,3-dicarboxylic acid (ACPD) or by high frequency stimulation (HFS) of presynaptic Schaffer collateral fibres (Congar et al. 1997). (2) Activation of the current is by means of group I mGluRs, which cause cytoplasmic Ca2+ release through IP3 signalling, and not by group II or group III mGluRs (Congar et al. 1997). (3) A rise in [Ca2+]i is required for activation of the current (Crepel et al. 1994; Congar et al. 1997). (4) The current reverses at the potential expected for non-selective channels and far from the Cl or K+ reversal potentials (Crepel et al. 1994; Congar et al. 1997). Synaptic stimulation of mGluRs is then a convenient means of activating CAN channels in CA1 neurones through the following sequence of events: HFS to Schaffer collaterals [RIGHTWARDS ARROW] presynaptic glutamate release [RIGHTWARDS ARROW] activation of perisynaptic group I mGluRs in CA1 neurones [RIGHTWARDS ARROW] IP3 cascade [RIGHTWARDS ARROW] Ca2+ release from cellular stores [RIGHTWARDS ARROW] Ca2+-dependent ICAN activation [RIGHTWARDS ARROW] CAN channel-dependent depolarization (ΔVCAN).

The preceding description obviously implicates IP3-sensitive stores in the activation of ICAN and such an involvement has been clearly demonstrated in some neurones. For instance, ICAN is activated by IP3 injection (Sawada et al. 1990) or thapsigargin application (Knox et al. 1996) in Aplysia neurones and blocked by internal administration of heparin in neostriatal (Wu & Wang, 1996) or substantia nigra (Wu & Wang, 1995) neurones. On the other hand, in dorsal root ganglion neurones, caffeine activates ICAN (Currie & Scott, 1992) as does intracellular application of βNAD+ (Crawford et al. 1997), both presumably through Ca2+ release from Ca2+-sensitive stores. The interaction of different intracellular sources of Ca2+ in CAN channel activation and modulation has not been investigated.

The results reported here show that HFS of the Schaffer collateral-commissural pathway produces a ΔVCAN in CA1 neurones that can be dramatically potentiated. Ca2+ from IP3-sensitive, Ca2+-sensitive, or mitochondrial Ca2+ stores can significantly enhance ΔVCAN. Possible mechanisms of ΔVCAN potentiation due to an involvement of ryanodine-sensitive stores, of the filling state of Ca2+ stores, of a contribution from Ca2+ influx, and of the effect of Ca2+ on IP3 receptors are considered.

Because CAN channels are activated by [Ca2+]i, cause maintained depolarization, and provide a potential Ca2+ influx pathway, they are potentially important in glutamate-dependent plasticity and toxicity. The observation that Ca2+ from several cellular sources can greatly potentiate CAN channel responses has important consequences for both of these processes. For instance, ICAN potentiation may play a significant role in the ‘amplification’ phase of excitotoxicity (Choi, 1990; Tatsumi & Katayama, 1994).

METHODS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements

Brain slices

Sprague-Dawley rats were anaesthetized with sodium pentobarbital (50 mg kg−1i.p.) and the brain rapidly removed and immediately submerged in ice-cold oxygenated artificial cerebrospinal fluid (aCSF) containing (mm): NaCl 124, KCl 5, NaHCO3 26, NaH2PO4 1.25, MgSO4 1.3, CaCl2 2.5, glucose, 10. Transverse brain slices, 400 μm thick, were cut with a vibratome (Pelco 1000) and slices were incubated for at least 1 h at room temperature in a chamber continuously bubbled with 90 % O2-5 % CO2. Animal protocols were approved by the Institutional Animal Care and Use Committee.

Hippocampal neurone culture

Cultures were prepared from hippocampi isolated from embryonic day 18–19 (E18-19) or postnatal day 3 (P3) Sprague-Dawley rats killed as described above. Hippocampi were dissected in cold sterile aCSF plus 15 mm Hepes, 17 mm dextrose and 17.5 mm sucrose (pH 7.4, with a final osmolarity of 320–335 mosmol l−1). Isolated hippocampi were incubated for 10 min at 37°C in 0.05 % trypsin- EDTA (Gibco-BRL) and gently triturated with a Pasteur pipette. Trituration was repeated with a Pasteur pipette flame-pulled to half its original opening size. Cells were plated at a density of 15 × 103-20 × 103 ml−1 of media in culture dishes containing coverslips coated with polylysine and collagen and maintained at 37°C with 5 % CO2 in a humidified atmosphere. Cells were initially plated in Neurobasal media (Gibco-BRL) containing 10 % fetal bovine serum, 100 U ml−1 penicillin, 0.1 mg ml−1 streptomycin and 25 μm glutamate. After 24 h, fetal bovine serum was substituted with the N2 supplement from Gibco-BRL and after 3 days in culture, glutamate was removed from the culture media. For P3 cultures, B27 supplement (Gibco-BRL) was used instead of N2 supplement. Cells were used for experiments after 7–10 days.

Electrophysiology

For intracellular recording, slices were maintained at 34°C in a submerged brain slice chamber (Scientific Systems Design) with a bath volume of 3 ml and a constant flow of warmed, humidified aCSF, bubbled with 95 % O2-5 % CO2, over the surface at a flow rate of 1 ml min−1. Drugs were either added directly to the bath or applied through a rapid exchange system, which allowed complete exchange of the bath in less than 1 min. Bipolar stimulating electrodes were placed in the Schaffer collateral-commissural pathway and the stimulus intensity was set to obtain the largest subthreshold EPSP. Intracellular recordings from CA1 pyramidal neurones were made using 80–120 MΩ glass microelectrodes filled with 4 M potassium acetate attached to an Axoclamp-2A amplifier (Axon Instruments). Tip potentials were compensated and no noticeable DC drift was observed. Only neurones producing action potentials to depolarizing pulses were used. Input resistance (Ri) was 76.0 ± 7.7 MΩ for a sample of the neurones recorded in the slice. pCLAMP 6 software (Axon Instruments) was used for experimental control and data analysis.

Individual cultured neurones were identified by their pyramidal shape and long processes and were patch clamped using an Axopatch 200A amplifier (Axon Instruments) using standard whole cell techniques. Only neurones with robust, inactivating inward currents following step depolarizations were used. Series resistance was not compensated. Ri was 200.1 ± 41.0 MΩ for a sample of the whole cell patch-clamped cultured neurones. Experimental control and data analysis were by means of pCLAMP 6 software (Axon Instruments). The extracellular recording solution contained (mm): NaCl 140, KCl 5, CaCl2 2, MgCl2 1, Hepes 10, dextrose 25 (pH 7.4, 320 mosmol l−1). The pipette solution contained (mm): KCl 140, MgCl2 4, EGTA 0.1, ATP 2, Hepes 10 (pH 7.4, 290 mosmol l−1). Drugs were applied with a gravity-fed ‘Y-tube’ (Murase et al. 1990), which had a diameter of approximately 100 μm and was placed within 100 μm of the clamped neurone, thereby allowing a complete exchange of the solution bathing a neurone within several hundred milliseconds.

Ca2+ imaging

Cultured hippocampal neurones were loaded with 4 μm fura-2 AM (Sigma) in 0.05 % pluronic acid for 1 h at 37°C in culture medium and rinsed 4 times in recording solution prior to imaging. Images were collected with an inverted microscope (Nikon Diaphot 300) equipped with a cooled CCD video imaging system (SenSys 1400, Photometrics). Metafluor image processing and analysis software (Universal Imaging) was used for data collection at a 0.2 Hz sample rate with an exposure time of 100 ms and 2 × vertical and horizontal binning. Cells were excited at 340 and 380 nm and emitted fluorescence was measured at 510 nm. [Ca2+]i was calculated from background-subtracted ratio images using measured minimum and maximum fluorescence ratios (Rmin and Rmax) values (Grynkiewicz et al. 1985).

Chemicals

Benzamil, dantrolene, carbonyl cyanide m-chlorophenyl hydrazone (CCCP), tetraethylammonium chloride (TEA), and 4-aminopyridine (4-AP) were obtained from Sigma Chemical Co. ACPD, 6-nitro-7-sulphanolylbenzo[f]quinoxaline-2,3-dione (NBQX), D(−)-2-amino-5-phosphonopentanoic acid (AP-5) and (RS)-1-aminoindan-1,5 carboxylic acid (AIDA) were obtained from Tocris Cookson Inc.

RESULTS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements

ΔVCAN

Figure 1A outlines the procedure used to activate the CAN channel-dependent depolarization (ΔVCAN) whose potentiation was studied here. An intracellular recording was made from a CA1 neurone in a hippocampal slice preparation and the presynaptic Schaffer collateral- commissural pathway was stimulated with a bipolar electrode. A single subthreshold presynaptic stimulus under control conditions produced the EPSP which is shown in Fig. 1Aa. This EPSP was reduced when NMDA and GABAA receptors were blocked with, respectively, 50 μm AP-5 and 20 μm bicuculline in Fig. 1Ab, and the EPSP was eliminated altogether in Fig. 1Ac when the AMPA/KA receptors were additionally blocked with 100 μm NBQX. This neurone continued, however, to produce robust, overshooting action potentials to depolarizing current pulses. Next, potassium currents and the Na+-Ca2+ exchanger were also blocked with 10 mm TEA, 5 mm 4-AP and 100 μm benzamil and the neurone now fired very broad action potentials when depolarizing pulses were applied. At this point, HFS (1 s, 100 Hz, 2 ms pulses of the amplitude determined in Fig. 1Aa) produced the slow depolarizing postsynaptic ΔVCAN seen in Fig. 1Ad. These conditions are virtually identical to those used by Congar et al. (1997) to activate ICAN in CA1 neurones. The synaptic nature of ΔVCAN was further demonstrated in two experiments where neurones, which produced a robust ΔVCAN following HFS, showed no measurable depolarization following suprathreshold depolarizations at the same frequency applied directly to the postsynaptic CA1 neurone.

image

Figure 1. PSPs in CA1 neurones

Aa–c, EPSPs recorded in a CA1 neurone in response to a constant-amplitude, single presynaptic stimulus to the Schaffer collaterals; a, control; b, with 50 μm AP-5 and 20 μm bicuculline; c, with AP-5, bicuculline and 100 μm NBQX. Ad, 100 Hz presynaptic stimulus for 1 s in 50 μm AP-5, 20 μm bicuculline, 100 μm NBQX, together with 10 mm TEA, 5 mm 4-AP and 100 μm benzamil. Dashed lines indicateVm=−75 mV. B, normalized time integral of ΔVCAN following repeated HFS. Data represent 105 measurements in 32 CA1 neurones under the same conditions as Ad. Mean initial ΔVCAN= 23 ± 3.4 mV. Error bars are ±s.e.m. Linear regression fit for 0–840 s (large dashed line): slope = 0.0001, y intercept = 1.04. Linear regression fit for 0–1800 s (continuous line): slope = 0.00047, y intercept = 0.985. Small dashed lines indicate the 95 % confidence intervals for 0–1800 s.

Download figure to PowerPoint

Recordings from neurones were generally very stable in the presence of the ionotropic receptor-K+ channel-Na+-Ca2+ exchanger blocking solution and multiple HFSs could be applied to repeatedly elicit ΔVCAN in the same neurone. Evaluation of the time integral is an effective means of including changes in both the amplitude and the duration ΔVCAN (e.g. Thompson, 1997). Figure 1B shows control normalized ΔVCAN time integrals resulting from 105 HFSs in 32 neurones that demonstrate the relative constancy of this response in the absence of any experimental manipulations. Control values of ΔVCAN were especially constant during the first 840 s (representing 93 % of the measurements), but tended to increase somewhat after this time. This increase may have resulted from an increased [Ca2+]i during extended microelectrode impalements.

Cytoplasmic Ca2+

The data shown in Fig. 1B suggest that, under the conditions used here, HFS results in a rather reproducible release of Ca2+ in the postsynaptic CA1 neurone with the consequent activation of a reproducible ΔVCAN. We sought then to determine whether procedures that would alter [Ca2+]i would affect this response. Traces a and b in Fig. 2A show, respectively, the first and seventh ΔVCAN elicited in a neurone that was impaled with a 4 M potassium acetate microelectrode to which was added Ca2+ buffered with EGTA to 500 μm. Between periodic HFSs, constant-current depolarizing pulses were applied through the recording electrode (0.3 nA, 40 ms, 1 Hz). These pulses depolarized the neurone by 10–20 mV and were terminated about 45 s before each HFS application. Not only was there an increase in the amplitude of ΔVCAN, but there was also a dramatic increase in its duration. Figure 2B shows this potentiation of the normalized time integral of ΔVCAN in three neurones tested under these conditions. The peak of ΔVCAN under maximally potentiated conditions was certainly limited by the approach of the membrane potential (Vm) to VCAN, which was found to be approximately −26 mV in these experiments and thus the results reported here may significantly underestimate the overall potentiating effect of a sustained increase in [Ca2+]i on CAN channel activation.

image

Figure 2. Increase in the time integral of ΔVCAN with Ca2+-containing electrode

A, ΔVCAN recorded in a neurone that was periodically stimulated with HFS interposed with continuous trains of depolarizing pulses through the recording electrode. Trace a is the first and trace b is the seventh ΔVCAN. Dashed lines represent Vm=−75 mV B, normalized mean ΔVCAN integral (±s.e.m.) for 21 responses to HFS in 3 neurones as a function of time since the first HFS applied to each neurone. Continuous and dashed lines are, respectively, the linear regression and 95 % confidence intervals for the control data from Fig. 1B.

Download figure to PowerPoint

IP3-sensitive Ca2+ stores

The ΔVCAN response has been shown to be mediated through group I mGluRs which cause Ca2+ release from IP3-sensitive stores (Crepel et al. 1994; Congar et al. 1997). In order to assess the role of IP3-sensitive stores in CAN channel potentiation, we used dissociated hippocampal neurones for measuring [Ca2+]i and ICAN because these neurones were the most conducive to fura-2 imaging, whole cell clamping and drug application. Figure 3 shows representative data in which [Ca2+]i was measured under various conditions with mGluRs stimulated by ACPD. Figure 3A indicates the mean [Ca2+]i response of six neurones during a continued application of 200 μm ACPD followed by application of 1 μm thapsigargin. ACPD consistently caused an immediate rapid increase in [Ca2+]i (a), a more gradual rise in [Ca2+]i, and then a second rather rapid increase in [Ca2+]i (b). There was very little additional change in [Ca2+]i following subsequent addition of thapsigargin. Similar responses were seen in a total of 32 neurones in four different experiments. A very different response was observed when the ACPD was applied in the presence of 200 μm Cd2+. Cd2+ is a broad spectrum blocker of Ca2+ channels, although it is unclear if it is able to block stores depletion-activated (ICRAC) channels. Cd2+ does not directly affect postsynaptic mGluRs in CA1 neurones (Vignes et al. 1996). Figure 3B shows the mean response of 23 neurones in one experiment to 200 μm ACPD applied with 200 μm Cd2+ in which a small initial transient was followed by a gradual, but greatly reduced rise in [Ca2+]i. A similar marked reduction in the ACPD response in the presence of Cd2+ was observed in a total of 50 neurones in three experiments. These results indicate that transmembrane flux participates with the expected IP3-sensitive store-dependent [Ca2+]i response to ACPD application.

image

Figure 3. [Ca2+]i response of dissociated neurones to ACPD

A, response of 6 neurones in one experiment to 200 μm ACPD followed by 1 μm thapsigargin (TGN). B, response of 23 neurones in one experiment to simultaneous application of 200 μm ACPD and 200 μm Cd2+ followed by 1 μm thapsigargin. Data represent means ±s.e.m.

Download figure to PowerPoint

The role of IP3-sensitive stores in the potentiation of ΔVCAN by HFS was assessed in experiments such as that shown in Fig. 4A and B. A typical ΔVCAN was elicited in a neurone under the conditions used in Fig. 1Ad; then 1 μm thapsigargin was added to the bath and ΔVCAN was activated by periodically applied HFS. As can be seen in the mean results in Fig. 4B, there was a maintained increase in time integral of ΔVCAN in the presence of thapsigargin.

image

Figure 4. Effect of thapsigargin on ΔVCAN and ICAN

A, representative ΔVCAN elicited by HFS in blocking conditions (trace a) or after 600 s in 1 μm thapsigargin (trace b). Dashed lines represent Vm=−75 mV. B, normalized mean ΔVCAN integral (±s.e.m.) for 23 responses to HFS in 4 neurones as a function of time since adding 1 μm thapsigargin at t= 0. Continuous and dashed lines are, respectively, the linear regression and 95 % confidence intervals for the control data from Fig. 1B. C, ICAN in whole cell patch-clamped cultured neurones held at −75 mV. Mean ±s.e.m. for 6 neurones of peak ICAN activated by 200 μm ACPD in the presence of 10 mm TEA and 5 mm 4-AP either alone or following a minimum 3 min bath application of 1 μm thapsigargin. Potentiation is significant at P < 0.05.

Download figure to PowerPoint

Whole cell patch-clamp recordings of cultured neurones were used to assess the potentiation of ICAN activated by bath application of ACPD. In the neurones shown in Fig. 4C, 200 μm ACPD was applied in the presence of 10 mm TEA and 5 mm 4-AP to block Ca2+-activated potassium currents. Under these conditions, ICAN was activated with similar characteristics to those reported by Congar et al. (1997). Thapsigargin (1 μm) was then added to the bath and, after a minimum of 3 min, ACPD was again applied and a significant potentiation of ICAN was observed. A total of six neurones were studied under these conditions and the mean ICANs.e.m.) before and after thapsigargin is shown in Fig. 4C.

Ca2+-sensitive Ca2+ stores

The potentiating effect of increasing [Ca2+]i on ΔVCAN (Fig. 2) and the biphasic [Ca2+]i response to ACPD (Fig. 3A) suggested that Ca2+ released from IP3-sensitive stores might lead to a secondary increase in [Ca2+]i perhaps through activation of Ca2+-sensitive stores. We used ryanodine, which causes Ca2+ release from Ca2+-sensitive stores (Rousseau et al. 1987), to demonstrate that this source of Ca2+ could be effective in the potentiation of ΔVCAN. Figure 5A is an example of a neurone to which 20 μm ryanodine was bath applied after an initial control period of HFS activation of ΔVCAN under the conditions used in Fig. 1Ad. Similar to the effect of thapsigargin, there was a dramatic increase in both the amplitude and duration of ΔVCAN. Figure 5B shows the mean time integral of ΔVCAN for four neurones under these conditions. Some of these neurones exhibited small spontaneous depolarizations that were increased following HFS and were frequently large enough to produce small bursts of action potentials (see Fig. 5Ab).

image

Figure 5. Effect of ryanodine on ΔVCAN

Aa, control ΔVCAN response; Ab, eighth ΔVCAN after 1000 s in 20 μm ryanodine. Dashed lines indicate Vm=−75 mV. B, normalize mean ΔVCAN integral (±s.e.m.) for 29 responses to HFS in 4 neurones as a function of time since adding 20 μm ryanodine at t= 0. Continuous and dashed lines are, respectively, the linear regression and 95 % confidence intervals for the control data from Fig. 1B.

Download figure to PowerPoint

The combined contribution of IP3- and Ca2+-sensitive stores to ΔVCAN potentiation was tested by applying 20 μm ryanodine with 1 μm thapsigargin (data not shown). Both of the neurones subjected to this combined treatment showed potentiation similar to that produced by thapsigargin or ryanodine alone, with a mean 7.8 ± 1.8-fold increase in the time integral of ΔVCAN. Furthermore, the increase was maintained for 20 min, similar to the response to ryanodine or thapsigargin alone.

To investigate further the involvement of Ca2+-sensitive stores in ΔVCAN potentiation, we repeated the experiments shown in Fig. 2 on neurones pretreated with dantrolene. Dantrolene has been shown to prevent release of Ca2+ from Ca2+-sensitive stores (Tekkok & Krnjevic, 1996). Every neurone tested without dantrolene pretreatment exhibited a marked increase in amplitude and time integral of ΔVCAN when Ca2+ was introduced into the cytoplasm from the recording electrode (e.g. Fig. 2). On the other hand, when pretreated with 100 μm dantrolene, the neurone in Fig. 6A and four additional neurones shown in Fig. 6B exhibited no increase in amplitude or in time integral under the same experimental conditions.

image

Figure 6. Effect of dantrolene on ΔVCAN amplification with Ca2+-containing electrode

These neurones were pretreated in 100 μm dantrolene and then periodically stimulated with HFS interposed with continuous trains of depolarizing pulses through the recording electrode. Aa is the first and Ab is the ninth ΔVCAN in one neurone. Dashed lines represent Vm=−75 mV B, normalized mean ΔVCAN integral (±s.e.m.) for 22 responses to HFS in 5 neurones as a function of time since the first HFS applied to each neurone. Continuous and dashed lines are, respectively, the linear regression and 95 % confidence intervals for the control data from Fig. 1B.

Download figure to PowerPoint

Mitochondrial Ca2+ stores

The involvement of mitochondrial Ca2+ stores in ΔVCAN potentiation was tested using CCCP. CCCP is an uncoupler of oxidative phosphorylation that dissipates the mitochondrial membrane potential, which drives mitochondrial uniporter Ca2+ uptake (Tang & Zucker, 1997). Figure 7A is an example of a neurone to which 2 μm CCCP was bath applied after an initial control period of ΔVCAN activation by HFS under the conditions used in Fig. 1Ad. Similar to the effect of thapsigargin or ryanodine, there was a dramatic increase in both the amplitude and duration of ΔVCAN in the presence of CCCP. Figure 7B shows the mean time integral of ΔVCAN for four neurones under these conditions.

image

Figure 7. Effect of carbonyl cyanide m-chlorophenyl hydrazine (CCCP) on ΔVCAN

Aa, control ΔVCAN response; Ab, the fifth ΔVCAN after 1200 s in 2 μm CCCP. Dashed lines indicate Vm=−70 mV. B, normalized mean integral of ΔVCANs.e.m.) for 18 responses to HFS in 4 neurones as a function of time since adding 2 μm CCCP at t= 0. Continuous and dashed lines are, respectively, the linear regression and 95 % confidence intervals for the control data from Fig. 1B.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements

Our intent in these experiments was to analyse the potentiation of CAN channels whose activation through mGluRs had been previously characterized (Crepel et al. 1994; Congar et al. 1997). Although our experiments were completed under very similar conditions to those of the previous authors, we undertook the following control experiments to demonstrate further that the potentiation we were studying was an effect on CAN channels. (1) All of these experiments were carried out in the presence of 100 μm benzamil to rule out a contribution from the Na+-Ca2+ exchanger (Agrawal, 1996). (2) A reversal potential of approximately −26 mV was measured for ΔVCAN under current-clamp conditions. This is close to the reversal potential for non-selective channels with expected intracellular cation concentrations, but considerably removed from the chloride reversal potential (ECl) of −69 mV determined from the reversal of GABAA IPSPs under our experimental conditions. (3) ΔVCAN was blocked by bath application of 1 mm AIDA, but not by 100 μm AIDA. This strongly suggests that ΔVCAN is activated by HFS through mGluR5 s - one of the IP3-linked group I mGluRs (Moroni, 1997).

In these experiments, mGluRs were stimulated over three different time courses. Presynaptic HFS was applied for 1 s with consequent activation of ΔVCAN which could be potentiated to last for up to about 30 s. ACPD was applied to whole cell-clamped neurones for 10–20 s activating a moderately constant-amplitude ICAN for this same length of time. Finally, ACPD was bath applied for several minutes resulting in an increasing [Ca2+]i if Cd2+ was not present. The processes responsible for ΔVCAN potentiation could be initiated on each of these time scales.

Potentiation is a postsynaptic effect

We have shown that CAN channels, activated as a result of mGluR stimulation, exhibit the potential for considerable potentiation. Three lines of evidence argue that this potentiation is a postsynaptic and not a presynaptic effect. (1) Potentiation occurs when Ca2+ is injected into the postsynaptic neurone (Fig. 2). (2) Thapsigargin-induced potentiation is observed in cultured hippocampal neurones in which ICAN is activated by ACPD applied directly to the recorded neurone (Fig. 4C). (3) Congar et al. (1997) studied ICAN activated both by bath application of ACPD and by presynaptic stimulation under the conditions used in this study. They found that the amplitude of ICAN increased with stimulation frequency and was maximal at 100 Hz (the frequency used in this study). The amplitude of the maximal ICAN following HFS was close to that following bath application of 200 μm ACPD. Thus the baseline condition for ΔVCAN activation in our studies probably represented a saturated presynaptic response and the observed potentiation would then have to be postsynaptic.

The observed potentiation follows an increase of [Ca2+]i by any of several means including: (1) direct injection of Ca2+ into the cytoplasm, (2) release of Ca2+ from IP3-sensitive stores, (3) release of Ca2+ from Ca2+-sensitive stores, and (4) block of Ca2+ uptake into mitochondria. In each instance, the resulting increase in [Ca2+]i potentiated the ΔVCAN activated by subsequent mGluR stimulation. Possible mechanisms for this [Ca2+]i-dependent potentiation will be considered below.

Potentiation of ΔVCAN

The observed potentiation ΔVCAN could occur directly as a result of modulation of CAN channels or indirectly as a result of an increase in the postsynaptic Ca2+ signal. The latter is more likely, given that ACPD causes a multiphase continued increase in [Ca2+] (Fig. 3) and that modulation of CAN channels by phosphorylation has previously been shown to depress their activity (Partridge et al. 1990; Razani-Boroujerdi & Partridge, 1993). There are at least four possibilities for a [Ca2+]i-dependent potentiation of ΔVCAN. (1) Ca2+-sensitive Ca2+ stores could be activated in addition to the mGluR-stimulated Ca2+ release from IP3-sensitive stores. (2) Cytoplasmic Ca2+ loads could increase the filling state of Ca2+ stores and hence the amount of Ca2+ available to be released by mGluR stimulation. (3) Depletion of intracellular Ca2+ stores could result in a transmembrane ICRAC that would potentiate refilling of mGluR-releasable stores. (4) Increased cytosolic [Ca2+] following mGluR-dependent Ca2+ release could cause an increase in the sensitivity of IP3 receptors to IP3. Each of these possibilities will be considered further below.

Ca2+-sensitive stores release

The release and sequestration of Ca2+ by ryanodine-sensitive stores has been thoroughly documented in CA1 neurones (Garaschuk et al. 1997) and these stores are a potential candidate for the ΔVCAN potentiation described here. In dorsal root ganglion neurones, ICAN can be activated by Ca2+ released from Ca2+-sensitive stores by caffeine (Currie & Scott, 1992). Depletion of Ca2+-sensitive stores by pretreatment with caffeine and ryanodine blocks the ability of mGluRs to activate the Ca2+-dependent K+ current (IK-Ca) in CA1 neurones (Shirasaki et al. 1994) and to activate an inward current in dorsal root ganglion neurones (Crawford et al. 1997). Thus Ca2+-sensitive stores are present in these neurones and Ca2+ released from these stores can activate Ca2+-activated ion channels. Furthermore, an interaction between IP3-sensitive stores and Ca2+-sensitive stores has been demonstrated such that depletion of one store diminishes the ability of the other store to activate Ca2+-dependent channels (Imanishi et al. 1996). Finally, the Ca2+ influx during an action potential is sufficient to trigger Ca2+ release from Ca2+-sensitive stores (Usachev & Thayer, 1997).

A consistent explanation for the biphasic increase in [Ca2+]i following ACPD application (Fig. 3A) is that Ca2+ released from IP3-sensitive stores causes a subsequent additional release from Ca2+-sensitive stores. The ability of dantrolene to prevent the Ca2+-induced potentiation of ΔVCAN (Fig. 6) argues strongly for an important role for Ca2+-sensitive stores in this process. The potentiation of ΔVCAN by ryanodine (Fig. 5) would then reflect the ability of Ca2+ released from Ca2+-sensitive stores to combine with that released by IP3-sensitive stores in the activation of CAN channels much as caffeine potentiates depolarizing after-potentials in supraoptic nucleus neurones (Li & Hatton, 1997).

Filling state of stores

The filling state of intracellular stores is a crucial factor in determining their ability to cause a significant change in [Ca2+]i. In PC12 cells, depletion of IP3- or Ca2+-sensitive stores activates stores refilling with a half-time of about 1 min (Bennett et al. 1998). In some instances, stores may need to be primed before they can generate a large regenerative release (Berridge, 1998); for instance, in CA3 pyramidal neurones, HFS causes significant increases in [Ca2+]i only after intense loading of Ca2+ stores (Pozzo Miller et al. 1996). Furthermore, as little as 1 min of KCl-dependent stores filling increases the amplitude and frequency of both IP3- and Ca2+-dependent elementary Ca2+ release events (Koizumi et al. 1999).

In some of the experiments reported here, oscillations of ΔVCAN were observed in the presence of ryanodine (e.g. Fig. 5Ab). One model for such oscillations in the presence of Ca2+ release agonists is based on feedback control of stores filling state (Henzi & MacDermott, 1992). The amplification of ΔVCAN described in these experiments may indicate an increase in the filling state of the stores following cytoplasmic Ca2+ loads from any of a number of sources. This would underlie a subsequently larger Ca2+ release with a consequent potentiation of ΔVCAN.

Transmembrane Ca2+ flux

Intracellular Ca2+ stores are the basis for a second messenger signalling pathway that is not initially dependent on extracellular Ca2+. However, depletion of these stores signals transmembrane Ca2+ influx through ICRAC channels by means of a diffusable messenger (Randriamampita & Tsien, 1993). A recent study reported the presence in CA1 neurones of ICRAC channels that are structurally related to the trp channel of Drosophila (Philipp et al. 1998). This pathway is activated following depletion of IP3-sensitive stores by thapsigargin (Takemura et al. 1989) or of Ca2+-sensitive stores by caffeine (Garaschuk et al. 1997). In both hippocampal neurones (Jaffe & Brown, 1994) and dorsolateral septum neurones (Zheng et al. 1996), transmembrane Ca2+ influx affects the amplitude of the [Ca2+]i response following mGluR stimulation. While voltage-dependent Ca2+ channels are not directly responsible for the activation of ICAN by ACPD (Crepel et al. 1994), extracellular Cd2+ does reduce the [Ca2+]i response to ACPD (Fig. 3B) and the amplitude of ICAN (Congar et al. 1997). The potentiation of ΔVCAN reported here may reflect a dependence of the Ca2+ available to activate CAN channels upon transmembrane Ca2+ flux and hence the filling state of Ca2+ stores.

Sensitivity of IP3 receptors to IP3

Both IP3 and ryanodine receptors are sensitive to the resting level of cytoplasmic Ca2+ (Berridge, 1998). Type I IP3 receptors, which are the most common family of these receptors in the brain, have a large cytoplasmic domain where Ca2+ can act as modulator of the transduction of ligand binding to channel opening (Joseph, 1996). There is a bell-shaped relationship between cytoplasmic [Ca2+]i and cerebellar IP3 receptor channel opening with a maximum at 0.2 μm (Bezprozvanny et al. 1991). Thus an increase in [Ca2+]i, from any of the several sources shown in Figs 2, 4, 5 or 7, could lead to the potentiation of ΔVCAN because of an increased sensitivity of IP3 receptors to the IP3 generated by mGluR stimulation.

Implications for excitotoxicity

Ischaemic neuronal death in the CA1 region of the hippocampus follows a large increase in [Ca2+]i in CA1 pyramidal neurones (Mitani et al. 1993). Glutamate- (or NMDA-) induced excitotoxicity is dependent on the amplitude (Milani et al. 1991) and on the duration of the rise in [Ca2+]i (Limbrick et al. 1995). In addition to the well-established role of ionotropic glutamate receptors, mGluRs have been shown to be involved in glutamate excitotoxicity. For instance, group I mGluRs have been linked to the pathology of ischaemic brain damage (Nicoletti et al. 1996). An involvement of ICAN in excitotoxicity is less well established. One proposal suggested that when [Ca2+]i reaches a threshold level, ICAN is activated, allowing rapid dissipative ion fluxes (Siesjo & Bengtsson, 1989). The results presented here indicate that ICAN can be greatly potentiated by several processes that increase [Ca2+]i including Ca2+ released from mitochondria (Fig. 7). This is an especially significant observation because of the involvement of mitochondrial Ca2+ in excitotoxic cell death (Khodorov et al. 1996) and the potential of ICAN to depolarize cells and provide an additional Ca2+ influx pathway (Poronnik et al. 1991; Partridge et al. 1994). The implication is that a potentiated ICAN could play a crucial role in the ‘amplification’ phase of excitotoxicity (Choi, 1990).

  • Agrawal, S. K. & Fehlings, M. G. (1996). Mechanisms of secondary injury to spinal cord axons in vitro: a role of Na, Na-K-ATPase, Na/H exchanger and the Na/Ca exchanger. Journal of Neuroscience 16, 545552.
  • Bennett, D. L., Bootman, M. D., Berridge, M. J. & Cheek, T. R. (1998). Ca2+ entry into PC12 cells initiated by ryanodine receptors or inositol 1,4,5-trisphosphate receptors. Biochemical Journal 15, 349357.
  • Berridge, M. J. (1998). Neuronal calcium signaling. Neuron 21, 1326.
  • Bezprozvanny, I., Watras, J. & Ehrlich, B. F. (1991). Bell-shaped calcium-response curves of Ins(1,4,5)P3 and calcium-gated channels from endoplasmic reticulum of cerebellum. Nature 351, 751754.
  • Choi, D. W. (1990). Cerebral hypoxia: some new approaches and unanswered questions. Journal of Neuroscience 10, 24932501.
  • Congar, P., Leinekugel, X., Ben-Ari, Y. & Crepel, V. (1997). IS generated by synaptic stimulation or exogenous activation of group I metabotropic glutamate receptors in CA1 pyramidal neurons. Journal of Neuroscience 17, 53665379.
  • Crawford, J. H., Wootton, J. F., Seabrook, G. R. & Scott, R. H. (1997). Activation of Ca2+-dependent currents in dorsal root ganglion neurons by metabotropic glutamate receptors and cyclic ADP-ribose precursors. Journal of Neurophysiology 77, 25732584.
  • Crepel, V., Aniksztejn, L., Ben-Ari, Y. & Hammond, C. (1994). Glutamate metabotropic receptors increase a Ca2+ activated nonspecific cationic current in CA1 hippocampal neurons. Journal of Neurophysiology 72, 15611569.
  • Currie, K. P. M. & Scott, R. H. (1992). Calcium-activated currents in cultured neurones from rat dorsal root ganglia. British Journal of Pharmacology 106, 593602.
  • Garaschuk, O., Yaari, Y. & Konnerth, A. (1997). Release and sequestration of calcium by ryanodine-sensitive stores in rat hippocampal neurons. The Journal of Physiology 502, 1330.
  • Grynkiewicz, G., Poenie, M. & Tsien, R. Y. (1985). A new generation of Ca2+ indicators with greatly improved fluorescence properties. Journal of Biological Chemistry 260, 34403450.
  • Hasuo, H., Phelan, K. D., Twery, M. J. & Gallagher, J. P. (1990). A calcium-dependent slow afterdepolarization of rat dorsolateral septal nucleus neurons in vitro. Journal of Neurophysiology 64, 18381846.
  • Henzi, V. & MacDermott, A. B. (1992). Characteristics and function of Ca2+ and inositol 1,4,5-trisphosphate-releasable stores of Ca2+ in neurons. Neuroscience 46, 251273.
  • Imanishi, T., Yamanaka, H., Rhee, J. S. & Akaike, N. (1996). Interaction between the intracellular Ca2+ stores in rat dissociated hippocampal neurones. NeuroReport 7, 14211426.
  • Jaffe, D. B. & Brown, T. H. (1994). Metabotropic glutamate receptor activation induces calcium waves within hippocampal dendrites. Journal of Neurophysiology 72, 471474.
  • Joseph, S. K. (1996). The inositol triphosphate receptor family. Cellular Signalling 8, 17.
  • Khodorov, B., Pilelis, V., Vergun, O., Storozhevykh, T. & Vinskaya, N. (1996). Mitochondrial deenergization underlies neuronal calcium overload following a prolonged glutamate challenge. FEBS Letters 397, 230234.
  • Knox, R. J., Jonas, E. A., Kao, L. S., Smith, P. J., Connor, J. A. & Kaczmarek, L. K. (1996). Ca2+ influx and activation of cation current are coupled to intracellular Ca2+ release in peptidergic neurons of Aplysia californica. The Journal of Physiology 494, 627639.
  • Koizumi, S., Bootman, M. D., Bobanović, L. A., Schell, M. J., Berridge, M. J. & Lipp, P. (1999). characterization of elementary Ca2+ release signals in NGF-differentiated PC12 cells and hippocampal neurons. Neuron 12, 125137.
  • Li, Z. & Hatton, G. I. (1997). Ca2+ release from internal stores: role in generating depolarizing after-potentials in rat supraoptic neurones. The Journal of Physiology 498, 339350.
  • Limbrick, D. D., Churn, S. B., Sombati, S. & DeLorenzo, R. J. (1995). Inability to restore resting intracellular calcium levels as an early indicator of delayed neuronal cell death. Brain Research 690, 145156.
  • Lujan, R., Nusser, Z., Roberts, J. D. B., Shigemoto, R. & Somogy, P. (1996). Perisynaptic location of metabotropic glutamate receptors mGluR1 and mGluR5 on dendrites and dendritic spines in the rat hippocampus. European Journal of Neuroscience 8, 14881500.
  • Milani, D., Guidolin, D., Facci, L., Pozzan, T., Buso, M., Leon, A. & Skaper, S. D. (1991). Excitatory amino acid induced alterations of cytoplasmic free Ca2+ in individual cerebellar granule neurons: role in neurotoxicity. Journal of Neuroscience Research 28, 434441.
  • Mitani, A., Yanase, H., Sakai, K., Yake, Y. & Kataoka, K. (1993). Origin of intracellular Ca2+ elevation induced by in vitro ischemia like condition in hippocampal slices. Brain Research 601, 103110.
  • Moroni, F., Lombardi, G., Thomsen, C., Leonardi, P., Attucci, S., Peruginelli, F., Torregrossa, S. A., Pellegrini-Giampietro, D. E., Luneia, R. & Pellicciari, R. (1997). Pharmacological characterization of 1-aminoindan-1,5-dicarboxylic acid, a potent mGluR1 antagonist. Journal of Pharmacology and Experimental Therapeutics 281, 721729.
  • Murase, Y., Ishibashi, H., Koyama, S. & Akaike, N. (1997). Ca2+ activated K currents in rat locus coeruleus neurons induced by experimental ischemia, anoxia, and hypoglycemia. Journal of Neurophysiology 78, 26742681.
  • Nicoletti, F., Bruno, V., Copani, A., Casabona, G. & Knopfel, T. (1996). Metabotropic glutamate receptors: a new target for the therapy of neurodegenerative disorders. Trends in Neurosciences 19, 267271.
  • Partridge, L. D., Mòller, T. H. & Swandulla, D. (1994). Calcium activated non-selective channels in the nervous system Brain Research Review 19, 319325.
  • Partridge, L. D. & Swandulla, D. (1987). Single Ca2+-activated cation channels in bursting neurons of Helix. Pflügers Archiv 410, 627631.
  • Partridge, L. D. & Swandulla, D. (1988). Calcium-activated non-specific cation channels. Trends in Neurosciences 11, 6972.
  • Partridge, L. D., Swandulla, D. & Müller, T. H. (1990). Modulation of calcium-activated non-specific cation currents by cyclic AMP-dependent phosphorylation of neurons of Helix. The Journal of Physiology 429, 131145.
  • Philipp, S., Hambrecht, J., Braslavski, L., Schroth, G., Freichel, M., Murakami, M., Cavalie, A. & Flockerzi, V. (1998). A novel capacitative calcium entry channel expressed in excitable cells. EMBO Journal 17, 42744282.
  • Poronnik, P., Cook, D. I., Allen, D. G. & Young, J. A. (1991). Diphenylamine-2-carboxylate (DPC) reduces calcium influx in a mouse mandibular cell line (ST885). Cell Calcium 12, 441447.
  • Pozzo Miller, L. D., Petrozzino, J. J., Golarai, G. & Connor, J. A. (1996). Ca2+ release from intracellular stores induced by afferent stimulation of CA3 pyramidal neurons in hippocampal slices. Journal of Neurophysiology 76, 554562.
  • Raggenbass, M., Pierson, P., Metzger, D. & Alberti, S. (1997). Action of a metabotropic glutamate-receptor agonist in rat lateral septum: Induction of a sodium-dependent inward after current. Brain Research 776, 7587.
  • Randriamampita, C. & Tsien, R. Y. (1993). Emptying of intracellular Ca2+ stores releases a novel small messenger that stimulates Ca2+ influx. Nature 364, 809814.
  • Razani-Boroujerdi, S. & Partridge, L. D. (1993). Activation and modulation of calcium-activated non-selective cation channels from embryonic chick sensory neurons. Brain Research 623, 195200.
  • Rekling, J. C. & Feldman, J. L. (1997). Calcium-dependent plateau potentials in rostral ambiguous neurons in the newborn mouse brainstem in-vitro. Journal of Neurophysiology 78, 24832492.
  • Rousseau, E., Smith, J. S. & Meissner, G. (1987). Ryanodine modifies conductance and gating behavior of single Ca2+ release channels. American Journal of Physiology 253, C364368.
  • Sawada, M., Ichinose, M. & Maeno, T. (1990). Activation of a non-specific cation conductance by intracellular injection of inositol 1,3,4,5-tetrakisphosphate into identified neurons of Aplysia. Brain Research 512, 333338.
  • Shirasaki, T., Harata, N. & Akaike, N. (1994). Metabotropic glutamate response in acutely dissociated hippocampal CA1 pyramidal neurons of the rat. The Journal of Physiology 475, 439453.
  • Siesjo, B. K. & Bengtsson, F. (1989). Calcium fluxes, calcium antagonists and calcium-related pathology in brain ischemia hypoglycemia and spreading depression: a unifying hypothesis. Journal of Cerebral Blood Flow and Metabolism 9, 127140.
  • Takemura, H., Hughes, R. A., Thastrup, O. & Putney, J. (1989). Activation of calcium entry by the tumor promoter thapsigargin in parotid acinar cells. Journal of Biological Chemistry 264, 1226912271.
  • Tang, Y.-G. & Zucker, R. S. (1997). Mitochondrial involvement in post-tetanic potentiation of synaptic transmission. Neuron 18, 483491.
  • Tatsumi, H. & Katayama, Y. (1994). Brief increases in intracellular Ca2+ activate K current and non-selective cation current in rat nucleus basalis neurons. Neuroscience 58, 553561.
  • Tekkok, S. & Krnjevic, K. (1996). Calcium dependence of LTP induced by 2-deoxyglucose in CA1 neurons. Journal of Neurophysiology 76, 23432352.
  • Thompson, A. M. (1997). Activity-dependent properties of synaptic transmission at two classes of connections made by rat neocortical pyramidal axons in vitro. The Journal of Physiology 502, 131147.
  • Usachev, Y. M. & Thayer, S. A. (1997). All-or-none Ca2+ release from intracellular stores triggered by Ca2+ influx through voltage-gated Ca2+ channels in rat sensory neurons. Journal of Neuroscience 17, 74047414.
  • Vignes, M., Blanc, E., Davos, F., Guiramand, J. & Recasens, M. (1996). Cadmium rapidly and irreversibly blocks presynaptic phospholipase C-linked metabotropic glutamate receptors. Neurochemistry International 29, 371381.
  • Wu, T. & Wang, H. L. (1995). Neurotensin increases the cationic conductance of rat substantia nigra dopaminergic neurons through the inositol 1,4,5-trisphosphate-calcium pathway Brain Research 683, 242250.
  • Wu, T. & Wang, H. L. (1996). The excitatory effect of cholecystokinin on rat neostriatal neurons: ionic and molecular mechanisms. European Journal of Pharmacology 307, 125132.
  • Zheng, F., Gallagher, J. P. & Connor, J. A. (1996). Activation of a metabotropic excitatory amino acid receptor potentiates spike-driven calcium increases in neurons of the dorsolateral septum Journal of Neuroscience 16, 60796088.

Acknowledgements

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements

The authors would like to thank Drs Benjamin Walker and Thomas Resta for use of the Ca2+ imaging apparatus, Dr Nancy Kanagy for providing animals, and Drs Bill Shuttleworth and Bridget Wilson for helpful comments on the manuscript. This work was supported in part by NIH grant AA00227 to C.F.V.