- Email: [email protected]
Calcium channel subtypes responsible for voltage-gated intracellular calcium elevations in embryonic rat motoneurons
Pergamon
PII:
Neuroscience Vol. 87, No. 3, pp. 719–730, 1998 Copyright 1998 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306–4522/98 $19.00+0.00 S0306-4522(98)00165-1
CALCIUM CHANNEL SUBTYPES RESPONSIBLE FOR VOLTAGE-GATED INTRACELLULAR CALCIUM ELEVATIONS IN EMBRYONIC RAT MOTONEURONS F. SCAMPS,* S. VALENTIN,* G. DAYANITHI† and J. VALMIER*‡ *CNRS UPR 1142, Institut de Biologie, Boulevard Henri IV, 34060 Montpellier, France †CNRS UPR 9055, Biologie des Neurones Endocrines, CCIPE, 141 rue de la Cardonille, 34094 Montpellier Ce´dex 4, France Abstract––The central role of electrical activity and Ca2+ influx in motoneuron development raises important questions about the regulation of Ca2+ signalling induced by voltage-dependent Ca2+ influx. In the purified embryonic rat motoneuron preparation, we recorded barium currents through voltageactivated Ca2+ channels using the whole-cell configuration of the patch-clamp technique. We found that motoneurons express at least four types of high-voltage-activated Ca2+ channels, based on their kinetics, voltage-dependences and pharmacological properties. Of the sustained Ca2+ current activated at 0 mV from a holding potential of 100 mV, approximatively 45% was omega-conotoxin-GVIA (1 µM) sensitive, 25% was omega-agatoxin-IVA (30 nM) sensitive and 20% was nitrendipine (250 nM) sensitive. The residual current, after applying these three antagonists, was an inactivating current that differs from classical T-type Ca2+ currents. Based on this pharmacology, changes in intracellular free Ca2+ concentrations were then monitored by Fura 2 digital imaging microspectrofluorimetry. Upon K+ depolarization, the intracellular Ca2+ transient induced by the activation of each type of Ca2+ channel appeared to be quantitatively proportional to their Ca2+ influx. The existence of a calcium-induced calcium release mechanism through activation of caffeine-, ryanodine-sensitive intracellular stores was then investigated. High doses of caffeine and low doses of ryanodine failed to increase intracellular free calcium concentrations and low concentrations of caffeine and high concentrations of ryanodine did not affect K+-induced intracellular free calcium concentration transients indicating both the absence of Ca2+-gated Ca2+-release channels and of a Ca2+-induced Ca2+ release mechanism. Together, these data provide evidence that embryonic motoneurons express multiple Ca2+ channels that function as important regulators of intracellular Ca2+ signalling and may be involved in their development. 1998 IBRO. Published by Elsevier Science Ltd. Key words: calcium channels, intracellular free calcium, ryanodine, caffeine, spinal motoneuron.
Plasmalemmal voltage-activated Ca2+ channels induce intracellular free Ca2+ concentrations ([Ca2+]i) increase and in this way, control many aspects of neuronal function including generation of Ca2+-dependent action potentials, neurotransmitter release, regulation of neuronal death, synapse formation and elimination, phenotypic differentiation and gene expression.1,10,16,28,31 It has become clear that voltage-dependent Ca2+ transients may reflect not only the entry of Ca2+ through plasmalemmal Ca2+ channels but also Ca2+ release from intracellular stores of the endoplasmic/sarcoplasmic reticulum by a Ca2+-induced Ca2+ release (CICR) mechanism (for review see Ref. 26). First described in striated muscle ‡To whom correspondence should be addressed. Abbreviations: Aga-IVa, omega-agatoxin-IVA; [Ca2+]i, intracellular free Ca2+ concentration; CICR, Ca2+induced Ca2+ release; DHP, dihydropyridine; E, embryonic day; GVIA, omega-conotoxin-GVIA; HEPES, N-2hydroxyethylpiperazine-N -2-ethanesulphonic acid; HP, holding potential; HVA, high-voltage-activated; IBa, barium current; ICa, calcium current; LVA, low-voltageactivated; TEACl, tetraethylammonium chloride.
cells, interest in neuronal CICR has grown with the recognition that it prolongs and amplifies the rise in [Ca2+]i initiated by Ca2+ entry and regulates [Ca2+]i oscillations (for review see Ref. 27). Schematically, Ca2+ stores in neurons responsible for releasing Ca2+ can be separated into at least two types based on their Ca2+ release channel, namely the caffeine- and ryanodine-sensitive Ca2+ channel responsible for the CICR following voltage-dependent Ca2+ entry and the inositol-1,3,4-triphosphate-sensitive Ca2+ channel (for review see Ref. 40). Neuronal voltage-activated Ca2+ channels (ICa) were first classified as low-voltage-activated (LVA) ICa and high-voltage-activated (HVA) ICa: a dihydropyridine (DHP)-sensitive L-type ICa and an omega-conotoxin-GVIA (GVIA)-sensitive N-type ICa (for review see Ref. 42). Then two additional HVA ICa were described which were sensitive to omega-Aga-IVA (Aga-IVa). One, identified in Purkinje neurons30 and referred to as P-type, is blocked with high affinity (IC50 =2–10 nM).34 The other, referred to as Q-type, is blocked with low affinity.21,39 In addition to these different types of
719
720
F. Scamps et al.
currents, a DHP/GVIA/Aga-IVa-resistant HVA ICa, referred to as toxin-resistant ICa, has also been observed.22,44 Recent data suggest that each type of Ca2+ channel is functionally different from one neuron to another and in a single neuron, controls specific pathways for Ca2+ signalling, thus supporting different functions. However, an understanding of the relationship between transmembrane Ca2+ entry, intracellular [Ca2+]i transient and cytoplasmic Ca2+ stores at single cell level is just begining to emerge. In embryonic spinal motoneurons, Ca2+ entry induced by electrical activity has been shown to play a central role in regulating in vivo neuromuscular activity which in turn regulates naturally occurring neuronal death and elimination of supernumerary synapses.17 Although these observations raise many questions about Ca2+ signalling in embryonic motoneurons, to date, no information is available on the regulation of motoneuron intracellular Ca2+ transients by Ca2+ influx. In a previous study, we have shown that it was possible to dissect, using a simple method of purification, the properties of the ICa expressed by mammalian motoneurons and to provide, for the first time, direct measurements of a Aga-IVa-sensitive, probably P-type, Ca2+ current in embryonic rat motoneurons.23 The aim of the present study was to characterize, in this model, the Ca2+ influx through voltage-activated Ca2+ channels, the [Ca2+]i rise induced by membrane depolarization and the possible involvement of a CICR mechanism. EXPERIMENTAL PROCEDURES
All animal experiments were carried out in accordance with the European Communities Council Directive of 24 November 1986 and all efforts were made to minimize animal suffering. Cell cultures Spinal motoneurons from embryonic day 15 (E15) Sprague–Dawley rat embryos (CERJ, Le Genest St Isle, France) were purified by a two-step metrizamide-panning method, as previously described.5 Briefly, ventral spinal cords were dissociated after trypsin digestion and centrifuged over 6.5% metrizamide cushions. The large cells that were not dense enough to pass through the metrizamide were further enriched by immunopanning on Petri dishes coated with antibody to the p75 neurotrophin receptor (low-affinity nerve growth factor receptor), which is specifically expressed by motoneurons at this stage. As described elsewhere,18 all (>95%) of the purified neurons were labelled by antibodies to p75 or to L14, a motoneuron-specific intracellular lectin. A subpopulation of these (50–90%) expressed Islet-1, another motoneuron marker,11,23 and were thus unambiguously identified as motoneurons. To confirm that they were motoneurons, the cells recorded initially were localized on the dish, fixed and stained using an antibody to Islet-1.23 In subsequent experiments, neurons to be patch-clamped were selected, under phasecontrast microscopy, on the basis of their large, multipolar morphology and phase-bright rounded cell body. No differences were seen in comparing the currents from labelled and unlabelled motoneuron-like cells.
Electrophysiological recordings Twenty-four hours after motoneurons seeding, whole-cell recordings were obtained at 20–22C under conditions optimized to ensure isolation of Ca2+ channel currents from other voltage-activated currents. To block Na+ channels, we replaced extracellular Na+ by tetraethylammonium chloride (TEACl) and added tetrodotoxin (1 µM; Sigma, France) to the bath medium. K+ currents were blocked by extracellular TEACl and by internal CsCl. The bathing solution contained: TEACl, 116 mM; BaCl2, 5 mM; HEPES 10 mM; glucose 10 mM; pH 7.35 adjusted with TEA-OH. Recording pipettes were filled with the following solution: CsCl, 134 mM; 1,2-bis(2-aminophenoxy)ethane-N,N,N ,N ,-tetraacetic acid, 20 mM; HEPES, 10 mM; Mg-ATP, 3 mM; Mg-GTP, 1 mM; glucose, 10 mM; at pH 7.35 (adjusted with CsOH). The pipette resistances were c3 MÙ. Whole-cell currents were recorded using a Biologic RK300 patch-clamp amplifier (Claix, France). After seal formation (resistance 5–10 GÙ) and membrane disruption, series resistance was estimated from the capacitative transient evoked by a +10 mV test pulse. Membrane capacity was calculated as Cm=ô/Rs, with ô being the time-constant of capacitative transient and Rs the series resistance. Mean Rs was 16.02.3 MÙ, n=23. Because current amplitudes were always c1 nA, voltage errors resulting from the uncompensated series resistance were negligible (<6 mV). Cells having unstable leak >100 pA were usually discarded, because this phenomenon often induced current run down. Leak was estimated at the end of an experiment by adding 50 µM NiCl2 and CdCl2 to inhibit the remaining inward current. From measurement of the apparent reversal potential of the calcium current when using a ramp protocol (500 ms duration from 100 mV to +50 mV), we obtained a 41 mV negative voltage drift (n=3), when estimated up to 10 min after establishment of the whole-cell configuration. All experimental parameters were controlled with a computer equipped with a Tecmar Labmaster analog interface (Axon Instruments, Chatillon, France). Data acquisition and analyses were performed using the pclamp software (v 6.03, Axon instruments). Current signals were sampled at 5 kHz and filtered at 3 kHz before digitization and storage. Dye loading and measurements of [Ca2+]i Dye loading and measurement of [Ca2+]i was performed as previously described.8 Briefly, the culture dishes were washed, loaded by incubation with 3 µM Fura-2AM and 0.05% w/v Pluronic F-127 (Molecular Probes Inc., U.S.A.) in Locke buffer (NaCl 140 mM; MgSO4 1.2 mM; CaCl2 1.8 mM; glucose 10 mM; KH2PO4 5 mM; HEPES-NaOH 10 mM; pH 7.25) at 34C for 35 min. Subsequently, loaded cells were washed with Locke buffer and fluorescence measurements were performed with buffers maintained at 20–22C. [Ca2+]i levels in single cells were measured using the digital imaging microfluometry system (Axon Instruments, Chatillon, France) based on an inverted microscope (Nikon) equipped with epifluorescence. Interference filters of 340/10 nm and 380/10 nm were alternately mounted on the filter wheel and the excitation light beam was deflected through an oil-immersion objective (40 0.75 NA, Nikon, Champigny-sur-Marne, France). Calibration of the fluorescence ratios was not attempted, because the present study aimed to evaluate relative changes in [Ca2+]i following activation of different types of Ca2+ conductances rather than obtaining absolute values of [Ca2+]i. [Ca2+]i variations were obtained from the ratio between fluorescence emission that resulted from emission at 340 and 380 nm.15 Drugs Nitrendipine (Bayer AG) was dissolved in dimethylsulfoxide to make concentrated stock solutions (10 mM) stored at 20C. Controls showed that the solvent has no effects on Ca2+-channel currents at the final dilutions used here
Calcium movements in motoneurons (<0.05%). The omega-conotoxin-GVIA (Sigma), omegaAga-IVA (Peptide International, Louisville, KY and Pfizer, U.S.A.) and omega-conotoxin-MVIIC (Latoxan, Bachem Rosans, France), were dissolved in bidistilled water at 0.1 mg/ml, and 1mg/ml respectively to make stock solutions. NiCl2 and CdCl2 were prepared in bidistilled water at 10 mM as stock solution. Test solutions were prepared daily using aliquots from frozen stores to obtain the working concentrations. Nitrendipine and GVIA were applied using a perfusion system (200 µm inner diameter capillary tubing; flow rate of 0.3 ml/min) placed in the vicinity of the cell. A pneumatic pressure pump (Miniframe PPS2, Medical System, NY) was employed for Aga-IVa application (30 to 50 ms ejection time, 200 ms pause). This system allows the use of small volumes of peptide solutions placed in glass microelectrodes (resistance <1 MÙ). Direct applications of Aga-IVa to obtain the same final concentration was also used and no difference was noted. Results are expressed as meanS.D. RESULTS
Characterization of voltage-activated Ca2+ channel currents Distribution and characterization of Ca2+ currents were studied in more than 30 acutely dissociated motoneurons under experimental conditions designed to eliminate other voltage-dependent currents (see Experimental Procedures). Ba2+ was used as a charge carrier on the basis of the larger amplitude of HVA Ca2+ currents and to minimize Ca2+ current rundown. In this condition, Ba2+ currents (IBa) were stable for up to 15 min at a holding potential of 100 mV allowing us to perform electrophysiological and pharmacological experiments. Up to 14 h after seeding, all time- and voltage-dependent currents were blocked by addition of 100 µM NiCl2 and CdCl2 to the external solution and were therefore considered to be Ca2+ channel currents. Immediately after rupture of the membrane, the holding potential (HP) was fixed at 100 mV and depolarizing test pulses to various potentials were delivered every 10 or 15 s. In these conditions, inward currents were recorded in motoneurons showing an amplitude increase (run-up) to reach in 1–2 min a steady-state from which all the investigations were performed. Whereas HVA currents were present in all cells recorded, less than 10% (range 5–30%) of the motoneurons express a LVA current, as checked when applying a ramp protocol from 100 mV to +50 mV. The population expressing only HVA currents represents an excellent model in which to observe HVA currents unimpeded by LVA current. Figure 1A shows the current–voltage (I–V) relationships of the peak and sustained components of IBa obtained from two holding potentials: HP=100 mV where all conductances were fully activated and HP=60 mV which is roughly the values of the membrane resting potential (587 mV; n=10, under 5.4 mM extracellular K+ and 140 mM intracellular K+; fire action potentials were never observed at rest). From HP=100 mV, the peak and
721
sustained currents were larger at all potentials than elicited from HP=60 mV (Fig. 1Aa, b). In the absence of LVA, the threshold for peak and sustained currents activation was similar when activated from 100 mV than from 60 mV (around 40 mV). However, the maximal IBa evoked between 0 to +10 mV from HP=100 mV showed a greater inactivation than it did from HP=60 mV (Fig. 1Ba). So, the current elicited at +10 mV had two components. One component was relatively insensitive to holding potential and was less inactivated during the test pulse (sustained current). The other was sensitive to holding potential and showed greater inactivation (transient current) (Fig. 1Bb). Therefore, the amplitude of IBa evoked at +10 mV from HP=60 mV represented 67%19% and 32% 17% of the sustained and the transient IBa evoked from HP=100 mV (n=9), respectively. To characterize the different Ca2+ current components in embryonic rat motoneurons, we tested the effects of different Ca2+ channel antagonists. Fig. 1C shows that it was possible, on a single motoneuron, to identify four different Ca2+ current subtypes: a (0.3 µM) nitrendipine-, a (1 µM) GVIA-, a (250 nM) Aga-IVa-sensitive and a residual ICa. In this motoneuron, the nitrendipine-sensitive, the GVIAsensitive, the Aga-IVa-sensitive and the residual ICa represented, respectively, 10, 20, 57 and 10% of the total sustained ICa elicited at +10 mV. We next analysed in detail the electrophysiological and pharmacological properties of these ICa components. The GVIA toxin is a selective blocker of the N-type ICa current in different neuronal preparations.42 When applied at 1–5 µM, a saturating concentration for motoneurons (data not shown), GVIA blocked, respectively, 4610% and 4812% of the transient and the sustained IBa elicited by 0 or +10 mV depolarizing pulse in all motoneurons tested (n=10) (Fig. 2Aa). Digital subtraction between control current traces and current traces in the presence of 1 µM GVIA revealed both an inactivating and a sustained component in all neurons tested (Fig. 2Ab). I–V relationships of the peak and sustained components of the GVIA-sensitive current from HP=100 mV is plotted in Fig. 4A. Under control conditions, when HP was varied from 100 mV to 60 mV, a sustained component was partially inactivated as shown in Figs 1B and 2Ba. The sustained component sensitive to HP represented 3319% of global current (n=9). In the presence of 1 µM GVIA, the remaining sustained component was mostly insensitive to HP changes (Fig. 2Bb). Therefore, the difference in the effect induced by depolarizing the HP on the global sustained component may be due to a preferential inactivation of the N-type Ca2+ channel current. The presence of L-type Ca2+ channels has been previously demonstrated in this preparation using the DHP agonist Bay K 8644.23 In order to further analyse this current and to quantify its contribution
722
F. Scamps et al.
Fig. 1. Characterization of ICa in E15 motoneurons. (A) (a) Current–voltage relationships of peak current (Ipeak) measured as the the peak inward current relative to the zero current. Ipeak-voltage relationships were plotted from a 100 mV holding potential (HP) or a 60 mV HP. (b) Current–voltage relationships of the sustained current (Isus) measured at 250 ms relative to the zero current. Isus-voltage relationships were constructed from a 100 mV or 60 mV HP. (Same cell in (a) and (b), Cm=42 pF). (B) (a) Representative current traces elicited from a 100 mV or a 60 mV HP to a +10 mV test pulse. Current traces are taken from A. Global current was composed of a transient component, measured as the difference between peak current and sustained current, and a sustained component, measured relative to zero current. (b) Corresponding difference current obtained by subtracting the current traces measured at 60 mV from the current traces measured at 100 mV. (C) Pharmacological dissection of global ICa. (a) Time-course effects of 0.3 µM nitrendipine, 30 nM Aga-IVa, 1 µM GVIA and 50 µM Cd applications. The same cell was sensitive to all drugs. (b) Corresponding current traces. HP=100 mV; test pulse=+10 mV. (Cm=50 pF).
to the global calcium current, we used nitrendipine, a DHP antagonist of the L-type calcium current. Since nanomolar concentrations of DHP antagonists are
sufficient to block nearly all cardiac and neuronal L-type channels,9 we used 300 nM nitrendipine, a saturating concentration for the DHP high-affinity
Calcium movements in motoneurons
723
Fig. 2. Characterization of the GVIA- and nitrendipine-sensitive currents. (A) (a) Current traces obtained from a 100 mV holding potential to a +10 mV test pulse. Application of 1 µM GVIA induced a decrease in transient and sustained currents. Further addition of 0.3 µM nitrendipine induced a more prominent decrease in sustained current than in transient current. Addition of 50 µM Cd allowed leak correction. (b) Subtraction of the current traces obtained in the presence of GVIA from the control current traces gives the GVIA-sensitive current (from (a)). This current appeared as an inactivating current. (c) Subtraction of the current traces obtained in the presence of nitrendipine from the current remaining after GVIA treatment gives the nitrendipine-sensitive current (from (a)). This current appeared as a slowly activating current. A similar current was obtained when nitrendipine was first applied (see Fig. 1C). (Cm=32 pF). (B) (a) Effects of varying the holding potential from 100 mV (1) to 60 mV (2) on control current measured at +10 mV test potential. (same cell as in A). (b) Effects of holding potential after GVIA application. Note that only the transient component remained sensitive to holding potential. (c) Effects of holding potential after cell treatment with GVIA and nitrendipine.
binding site, even at HP=100 mV, to quantify L-type IBa. At this concentration, nitrendipine inhibited a slowly activating, non inactivating component (Fig. 2Aa,c). The mean inhibition of the sustained
component (measured at 250 ms) was 2817% at +10 mV (n=8). I–V relationships of the ‘‘peak’’ (measured at 10 ms) and sustained components from HP=100 mV is plotted in Fig. 4B. It should be
724
F. Scamps et al.
Fig. 3. Characterization of the Aga IVa-sensitive current and the residual current. (A) (a) Cell was incubated with 1 µM GVIA and 0.3 µM nitrendipine to inhibit N- and L-type currents, respectively. Application of 100 nM Aga-IVa induced a decrease in transient and sustained currents. (b) Subtraction of the current traces obtained in the presence of Aga-IVa from the current remaining after GVIA and nitrendipine treatment gives the Aga-IVa-sensitive current. This current appeared as a slowly inactivating current. (Cm=43 pF). (B) (a) Effects of varying the holding potential from 100 mV (1) to 60 mV (2) on the current remaining after GVIA and nitrendipine treatment. As shown in Fig. 2Bc, only the transient component was sensitive to HP. (b) The residual current (insensitive to GVIA, nitrendipine and Aga-IVa) was composed of a transient component sensitive to HP and a sustained component insensitive to HP.
noted that DHP-sensitive channels have been reported to be rather insensitive to holding potential.20 In this model, the amplitude of the nitrendipinesensitive ICa current elicited from HP=60 mV is about 100% of that from a HP of 100 mV confirming previous study (Fig. 2Bc). Therefore, the decrease of the global sustained component induced by depolarizing the HP of the cell (Figs 1B, 2Ba) was not due to the contribution of the L-type current. We next tested the effect of Aga-IVa, since a functional Aga-IVa-sensitive (probably P-type) component has been previously identified in this model.23 To determine the saturating concentrations of this peptide antagonist, increasing concentrations were applied to motoneurons. A substantial decrease in ICa was found with 30 nM Aga-IVa (249%; n=8) (Fig. 3Aa). The Aga-IVa-sensitive current inactivated very slowly (Fig. 3Ab). This blocking effect had a fast time-course (Fig. 1Ca). Increasing the concentration to 100 nM and 200 nM caused no or little additional decrease in ICa (respectively, 299%; n=3 and 2510%, n=5). I–V relationship of the peak and
sustained components from HP=100 mV is plotted in Fig. 4C. The amplitudes of the Aga-IVa-sensitive current elicited from either HP=100 mV or 60 mV were identical (Fig. 3B). A current remained in the presence of nitrendipine (0.3 µM), GVIA (1 µM) and Aga-IVa (200 nM) (Fig. 3Aa, Bb). This inactivating current activated around 40 mV, reached a maximum amplitude between 0 mV and +10 mV and was always present at +40 mV and thus had the characteristics of a HVA current. I–V relationship of the peak and sustained components from HP=100 mV is plotted in Fig. 4D. Recent reports have shown that µM concentration of omega-conotoxin-MVIIC could inhibit, in some neurons, DHP-, GVIA-, Aga-IVa-insensitive IBa. Further application of 10 µM omega-conotoxinMVIIC to the pre-incubating medium had no effect (n=3). However this toxin-resistant IBa was completely blocked by addition of 50 µM CdCl2 (see Fig. 1C). The transient component represented 757% of the residual ICa (n=5). At the resting potential (60 mV), this transient ICa was greatly
Calcium movements in motoneurons
725
Fig. 4. Current–voltage relationships of the different components of global ICa. (A) Current–voltage relationships of the peak and sustained components of the GVIA-sensitive current (Cm=33 pF). (B) Current–voltage relationships of the peak and sustained components of the nitrendipine-sensitive current (Cm=32 pF). (C) Current–voltage relationships of the peak and sustained components of the Aga-IVa-sensitive current (Cm=51 pF). (D) Current–voltage relationships of the peak and sustained components of the resistant current (Cm=47 pF).
inactivated (Fig. 3Bb). So, it probably represented part of the transient component of the global ICa that inactivated at depolarizing HP of 60 mV. Characterization of [Ca2+]i transients through voltageactivated Ca2+ channels Based on the pharmacology of these ICa, we next studied how Ca2+ entry through voltage-activated Ca2+ channels affected [Ca2+]i and its relation with Ca2+ stores. First, this was done by comparing [Ca2+]i elevations produced by depolarizations before and after applying the three antagonists. Cells were depolarized with 100 mM K+ rather than whole-cell voltage-clamp techniques to minimize the disturbance of Ca2+ homeostasis induced by intracellular perfusion. Here again, neurons to be recorded were selected on the basis of their large, multipolar morphology and phase-bright rounded cell body (see Experimental Procedures and Ref. 23). Under resting conditions, embryonic motoneurons exhibited a
fluorescence ratio within the range 0.5 to 0.7 which was stable throughout the measuring period (up to 30 min). In these neurons, 100 mM K+ evoked a rapid increase in [Ca2+]i in all neurons tested (n=180). The subsequent applications of high K+ did not (or little) alter the peak amplitude of the [Ca2+]i response (as shown in Fig. 6). However, the successive applications of 100–500 nM Aga-IVa, 1 µM GVIA and 250 nM nitrendipine inhibited, respectively, 3011%, 4314% and 1511% of the increase in [Ca2+]i induced by depolarizations (n=58) (Fig. 5). These results confirm the functionality of a P-type Aga-IVa-sensitive, a N-type GVIA-sensitive and a L-type nitrendipine-sensitive voltage-activated ICa, previously characterized using the whole-cell configuration of the patch-clamp technique. When 1 µM GVIA was applied before Aga-IVa and nitrendipine, the percentage of the toxin-sensitive influx was not different (respectively, 4710%, 298%; n=32) (Fig. 5B). We concluded that, here again, there was
726
F. Scamps et al.
Fig. 5. Effects of various Ca2+ channel antagonists on K+-induced [Ca2+]i responses. All traces were monitored in a single neuron. Neurons were first exposed to control K+ (100 mM) depolarizations and [Ca2+]i responses were monitored (left trace). Subsequently, the same neuron was pre-incubated for 2 min with various antagonists as indicated in the trace and then challenged with high K+. After exposure to each antagonist followed by K+ challenge, the dish was washed under fast perifusion. Similar type of results were obtained in 58 and 32 neurons, respectively, for A and B protocols.
no significant cross-reaction between the effects of Aga-IVa and GVIA. We next analysed the presence of caffeine- and ryanodine-sensitive Ca2+ stores and a possible interaction between these stores and voltage-dependent Ca2+ entry. Bath application of caffeine (at 10, 20 and 50 mM), known to activate Ca2+ release from these stores,26 failed to increase [Ca2+]i in E15 rat motoneurons (n=40) (Fig. 6A). To confirm the absence of functional caffeine-sensitive stores in the [Ca2+]i transient induced by depolarization, the effect of caffeine at low concentrations, that potentiated
Ca2+ release through Ca2+-gated release channels, was investigated. Pre-incubating E15 motoneurons with 1 mM caffeine had no effect on [Ca2+]i transients initiated by 50 mM K+ depolarization (n=40) (Fig. 6B). Moreover, the kinetics of [Ca2+]i transients were unchanged before and after caffeine application. We concluded that E15 motoneurons did not express functional caffeine-sensitive intracellular Ca2+ channels and stores. Since some ryanodine receptors are insensitive to caffeine,13 we tested the effects of ryanodine at concentrations known to directly activate (nM
Calcium movements in motoneurons
727
Fig. 6. Effects of caffeine and ryanodine on [Ca2+]i (A) A selected neuron was subjected to high K+ as control [Ca2+]i response and then exposed to caffeine (20 nM) and ryanodine (1 µM) followed by a second K+ challenge (as shown by arrow). Note that the neuron showed no [Ca2+]i response. (B, C) K+-induced [Ca2+]i responses were monitored in the continued presence or absence of 1 mM caffeine (B) or 10 µM ryanodine (C). Similar profiles were obtained in at least 15–20 neurons.
concentrations) or inhibit (µM concentrations) these receptors.19 Direct application of 500 nM ryanodine failed to increase basal [Ca2+]i in this preparation (n=40) (Fig. 6A). We then tested higher concentrations of ryanodine taking into account the fact that ryanodine binding to its receptor has been reported to be use-dependent.40 Ten micromolar ryanodine had no effect on the [Ca2+]i transient induced by 100 mM K+ even when repetitive K+ depolarizations were carried out (Fig. 6C). Here again, no change in the kinetics of the [Ca2+]i transients were observed before and after the drug
application, confirming the absence of functional ryanodine-sensitive intracellular Ca2+ channels and stores. Taken together, these results indicate that embryonic motoneurons are devoid of both caffeineand ryanodine-sensitive Ca2+ channels and stores. DISCUSSION
This study provides the first description of the intracellular Ca2+ transients induced by Ca2+ influx through voltage-activated Ca2+ currents in motoneurons. Based on their pharmacology, kinetic and
728
F. Scamps et al.
voltage-dependence, at least four types of HVA Ca2+ channel currents were characterized and their contribution to the global Ca2+ current quantified: an N-type omega-conotoxin-GVIA-sensitive current, an L-type dihydropyridine-sensitive current, a P-type omega-Aga-IVA-sensitive current and a residual current, resistant to these three antagonists. These currents were present in all embryonic motoneurons recorded here and therefore most likely represent important molecular characteristics of these developing cells. All were available at resting potential and regulated [Ca2+]i when activated by depolarization. The absence of functional intracellular ryanodinesensitive Ca2+ channels and stores reinforces the notion that voltage-activated Ca2+ channel currents may have a major functional importance in regulating [Ca2+]i level during embryonic development. The only previous report on direct measurement of HVA ICa in identified embryonic mammalian motoneurons only described the presence of DHP- and a GVIA-sensitive HVA Ca2+ channels (but Aga-IVa was not tested).36 These data contrast with those obtained in postnatal mammalian motoneurons for which, extracellular and patch-clamp recordings indicate that, at least, motoneurons and motor nerve terminal possess two other ICa insensitive to both DHP and GVIA.38,43 As has been suggested previously,23 we demonstrated unambiguously the existence of four types of HVA Ca2+ currents in embryonic rat motoneurons that differ by their pharmacological properties, their kinetics, their voltagedependences and their different contributions to the total Ca2+ currents and Ca2+ transients. The strength of our demonstration depends essentially on the specificity of actions of the different antagonists. The specificity of Aga-IVa and GVIA has been demonstrated in a large number of clones and neuronal preparations.2 In our hands, even used at concentrations of five- or 10-fold higher than their saturating concentrations, these two antagonists failed to inhibit an additional ICa component. Although non-specific effects have been described for DHP, we and others have shown that at nanomolar concentrations, these compounds affect mainly neuronal L-type Ca2+ channels.9,32 Furthermore, when used together, little cross-reactivity has been seen between these three antagonists in this preparation despite a great variability from one cell to another (see Results). Although Aga-IVa inhibits, in addition to the P-type ICa, another neuronal ICa named Q-type,21,39 and the cloned á1A ICa (presumed to be Q-type) expressed in oocytes,35,41 the component of the non-L, non-N ICa sensitive to Aga-IVa in this preparation appears clearly similar to the P-type Ca2+ channel current first described in cerebellar Purkinje neurons.30 Indeed, the present results show that, despite the use of synthetic Aga-IVa, 30 nM of the toxin was nearly a saturating concentration and its time-course of inhibition was rapid. The current
component inactivated slowly and it is mainly insensitive to the holding potential. Finally, the use of omega-conotoxin-MVIIC did not affect ICa in embryonic motoneurons preincubated with GVIA and Aga-IVa. All these characteristics are in accordance with the notion that the current inhibited by Aga-IVa in motoneurons are mostly a P-like ICa. In the presence of nitrendipine, GVIA and AgaIVa at concentrations known to inhibit L, N and P Ca2+ channels in E15 motoneurons, we have characterized the properties of a non-L, non-N, non-P HVA residual Ca2+ channel current. Although this residual current was insensitive to these four antagonists, a property which is common with LVA T-type Ca2+ channels, these two types of Ca2+ channels appeared clearly different from the T-channel. The residual current was comprised of both fast inactivating and sustained components whereas the T-type current is rapidly inactivating (a feature of all T-type IBa characterized in motoneurons33,36 and in neurons6,25 to date). When comparing the two currents in E15 motoneurons, the residual current has a higher threshold of activation than the T-type current that is completely inactivated at HP=60 mV and the former is unaffected by 500 µM amiloride which completely blocks the T-type current in this preparation (Hivert and Valmier, unpublished observations). Conversely, the functional properties of the residual current in motoneurons seem quite similar to those of both cerebellar granule and dorsal root ganglion neurons9,39 and those of the á1E channel when expressed in oocytes.44 Whole-cell patch-clamp recordings, with Ba2+ as a charge carrier, indicated that all embryonic motoneuron HVA Ca2+ currents were activated at membrane resting potential. In order to analyse the role of these different classes of Ca2+ channels in regulation of cytosolic calcium levels in intact motoneurons, we have measured intracellular calcium transients induced by K+ depolarization, using the Ca2+ indicator Fura 2 and digital imaging microfluometry. This transient represents a physiologically relevant index of Ca2+ channel function. Our results show that all these channels were able to increase intracellular [Ca2+]i upon K+ depolarization. Since depolarizing Ca2+ influx has been shown to induce CICR in neurons by activating caffeine- and/or ryanodinesensitive intracellular stores,27 we next analysed the presence of a caffeine-induced [Ca2+]i increase in this preparation. Even at high concentrations (up to 50 mM), caffeine was unable to reveal a caffeine induced-[Ca2+]i response. Since low concentrations of caffeine can potentiate the response of caffeinesensitive channels to K+-induced Ca2+ influx in sympathetic neurons,12 the effect of K+ in the continued presence of 1 mM caffeine was analysed. Such a strategy failed to potentiate the K+-induced Ca2+ transient. These data demonstrate the absence of functional caffeine-sensitive Ca2+ release channels and caffeine-sensitive stores.
Calcium movements in motoneurons
Some ryanodine-sensitive Ca2+-release channels have been demonstrated to be insensitive to caffeine.13 To test this hypothesis in E15 motoneurons, we have analysed the effect on [Ca2+]i of 500 nM ryanodine, known to activate its receptor and of 10 µM ryanodine, known to inhibit it. Since the effects of ryanodine have been reported to be usedependent, we compared the increases and the kinetics of [Ca2+]i transients after repetitive K+ depolarizations and no significant changes were observed. These data demonstrate unambiguously the absence of functional caffeine-/ryanodine-sensitive Ca2+ release channels and ryanodine-sensitive stores. The absence of effect of Ca2+-gated Ca2+ channels indicates that E15 rat motoneurons are devoid of classical CICR mechanism to regulate [Ca2+]i. This pattern of [Ca2+]i regulation may be specific of this stage of development. Indeed, developmental regulation of other [Ca2+]i-regulating mechanisms have been observed in different systems3,4,7,14,21,29,33,37 and CICR mechanisms have been demonstrated in differentiating spinal neurons in the absence of caffeine-sensitive stores.24 This latter study shows that, although caffeine was unable to increase [Ca2+]i, depletion of internal Ca2+ stores in the endoplasmic reticulum greatly reduced elevation of calcium produced by depolarization. Furthermore, caffeine insensitivity disappeared with the in vitro maturation of amphibian spinal neuron. For these reasons, the existence of a unknown developmental-regulated caffeine-independent CICR mechanism are currently under investigation in E15 rat motoneurons.
729 CONCLUSION
Since Ca2+ influx play a pivotal role in the development of the neuromuscular system, we have characterized voltage-activated Ca2+ channels of embryonic rat motoneurons and explored their contribution in the regulation of the intracellular free Ca2+ transients induced by depolarization. The use of different Ca2+ channel antagonists, i.e. nitrendipine, GVIA, Aga-IVa and CdCl2, demonstrated the presence of at least four different types of HVA calcium channel currents, respectively the L-, N-, P- and R-types. All these calcium channels had the ability to induce intracellular calcium transient upon 100 mM K+ depolarization. Neither caffeine nor ryanodine affected voltage-dependent Ca2+ influx indicating the absence of a Ca2+-induced Ca2+-release mechanism through ryanodine-sensitive Ca2+ channels. These results demonstrate a great diversity of functional pathway for Ca2+ influx and [Ca2+]i signaling through voltage-activated Ca2+ channels in embryonic rat motoneurons which in turn may support the regulation of their activity during development. Acknowledgements—We thank A. Roig and C. Barrere for technical assistance and cell cultures and B. Rouvie`re for typing the manuscript. This work was supported by ‘‘Association Franc¸aise contre les Myopathies’’ (AFM), the Centre National de la Recherche Scientifique (CNRS), the Institut National de la Sante´ et de la Recherche Me´dicale (INSERM). S. Valentin was sponsored by the Ministe`re de la Recherche et de la Technologie (MRT).
REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
Bading H., Ginty D. D. and Greenberg M. E. (1993) Regulation of gene expression in hippocampal neurons by distinct calcium signaling pathways. Science 260, 181–186. Birnbaumer L., Campbell K. P., Catterall W. A., Harpold M. M., Hofmann F., Horne W. A., Mori Y., Schwartz A., Snutch T. P., Tanabe T. and Tsien R. W. (1994) The naming of voltage-gated calcium channels. Neuron 13, 505–506. Brandt C. J., DeLeon S., Martin D. R. and MacLennan D. H. (1987) Adult forms of the Ca2+ ATPase of sarcoplasmic reticulum expression in developing skeletal muscle. J. biol. Chem. 262, 3768–3774. Braun K., Scheich H., Zuschratter W., Heizmann C. W., Matute C. and Streit P. (1988) Postnatal development of parvalbumin-, calbindin-, and adult GABA-immunoreactivity in two visual nuclei of zebra finches. Brain Res. 475, 205–217. Camu W., Bloch-Gallego E. and Henderson E. (1993) Purification of spinal motoneurons from chicken and rat embryos by immunopanning. Neuroprotocols 2, 191–199. Carbone E. and Lux H. D. (1984) A low voltage-activated, fully inactivating Ca channel in vertebrate sensory neurons. Nature 310, 501–502. Cimino M. J. F. C. and Weiss B. (1990) Ontogenetic evelopment of calmodulin mRNA in rat brain using in situ hybridization histochemistry. Devl Brain Res. 54, 43–49. Dayanithi G., Widmer H. and Richard P. (1996) Vasopressin induced intracellular Ca2+ increase in isolated rat supraoptic cells. J. Physiol., Lond. 490, 713–717. Diochot S., Richard S., Baldy-Moulinier M., Nargeot J. and Valmier J. (1995) Dihydropyridines, phenylalkylamines and benzothizepines block N-, P/Q and R-type calcium currents. Pflu¨gers Arch. 431, 10–19. Dunlap K., Luebke J. I. and Turner T. J. (1995) Exocytotic Ca2+ channels in mammalian central neurons. Trends Neurosci. 18, 89–98. Ericson J., Thor S., Edlund T., Jessel T. M. and Yamada T. (1992) Early stages of motor neuron differentiation revealed by expression of homebox gene Islet-1. Science 256, 1555–1560. Friel D. D. and Tsien R. W. (1992) A caffeine- and ryanodine-sensitive Ca2+ store in bullfrog sympathetic neurones modulates effects of Ca2+ entry on [Ca2+]i. J. Physiol., Lond. 450, 217–246. Giannini G., Clementi E., Ceci R., Marziali G. and Sorrentino V. (1992) Expression of a ryanodine receptor-Ca2+ channel that is regulated by TGF-â. Science 257, 91–94. Ghishan F. K. and Arab N. (1988) Active calcium transport by intestinal endoplasmic reticulum during maturation. Am. J. Physiol. 254, G47–80.
730 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44.
F. Scamps et al. Grynkiewicz G., Poenie M. and Tsien R. Y. (1985) A new generation of Ca2+ indicators with greatly improved fluorescence properties. J. biol. Chem. 260, 3440–3450. Gu X. and Spitzer N. C. (1995) Distinct aspects of neuronal differenciation encoded by frequency of spontaneous Ca2+ transients. Nature 375, 784–787. Hamburger V. (1977) The developmental history of the motor neuron. Neurosci. Res. Prog. Bull. 15S, 1–37. Henderson C. E., Camu W., Mettling C., Gouin A., Poulsen K., Karihaloo M., Rullamas J., Evans T., MaMahon S. B., Armanini M. P., Berkmeier L., Phillips H. S. and Rosenthal A. (1993) Neurotrophins promote motor neuron survival and are present in embryonic limb bud. Nature 363, 266–270. Henzi V. and MacDermott A. B. (1992) Characteristics and function of Ca2+- and inositol-1,4,5-triphosphatereleasable stores of Ca2+ in neurons. Neuroscience 46, 251–273. Hess P. (1990) Calcium channels in vertebrate cells. A. Rev. Neurosci. 13, 337–356. Hilaire C., Diochot S., Desmadryl G., Baldy-Moulinier M., Richard S. and Valmier J. (1996) Opposite developmental regulation of P- and Q-type calcium currents during ontogenesis of large diameter mouse sensory neurons. Neuroscience 75, 1219–1229. Hilaire C., Diochot S., Desmadryl G., Richard S. and Valmier J. (1997) Toxin-resistant calcium currents in embryonic mouse sensory neurons. Neuroscience 80, 267–276. Hivert B., Bouhanna S., Diochot S., Camu W., Dayanithi G., Henderson C. E. and Valmier J. (1995) Embryonic rat motoneurons express a functional P-type voltage-dependent calcium channel. Int. J. dev. Neurosci. 13, 429–436. Holliday J., Adams R. J., Sejnowski T. J. and Spitzer N. (1991) Calcium-induced release of calcium regulates differentiation of cultured spinal neurons. Neuron 7, 787–796. Huguenard J. R. (1996) Low-threshold calcium currents in central nervous system neurons. A. Rev. Physiol. 58, 329–348. Kostyuk P. and Verkhratsky A. (1994) Calcium stores in neurons and glia. Neuroscience 63, 381–404. Kuba K. (1994) Ca2+-induced Ca2+ release in neurones. Jap. J. Physiol. 44, 613–650. Larmet Y., Dolphin A. C. and Davies A. M. (1992) Intracellular calcium regulates the survival of early sensory neurons before they become dependent on neurotrophic factors. Neuron 9, 563–574. Leberer E., Seedorf U. and Pette D. (1992) Neural control of gene expression in skeletal muscle. Calcium sequestering proteins in developing and chronically stimulated rabbit skeletal muscle. Biochem. J. 239, 295–300. Llinas R., Sugimori M., Hillman D. E. and Cherksey B. (1992) Distribution and functional significance of the P-type voltage-dependent Ca2+ channels in the mammalian central nervous system. Trends Neurosci. 15, 351–355. Mattson M. P. and Barger S. W. (1993) Roles for calcium signaling in structural plasticity and pathology in the hippocampal system. Hippocampus 3, 73–88. McCarthy R. T. and TanPiengco P. E. (1992) Multiple types of high-treshold calcium channels in rabbit sensory neurons: high-affinity block of neuronal L-type by nimodipine. J. Neurosci. 12, 2225–2234. McCobb D. P., Best P. M. and Beam K. G. (1989) Development alters the expression of calcium currents in chick limb motoneurons. Neuron 2, 1633–1643. Mintz I. M., Venema V. J., Swiderek K. M., Lee T. D., Bean B. P. and Adams M. E. (1992) P-type calcium channels blocked by the spider toxin ù-Aga-IVA. Nature 355, 827–829. Mori Y., Friedrich T., Kim M. S., Mikami A., NakaI} J., Ruth P., Bosse E., Hofmann F., Flockerzi V., Furuichi T., Mikoshiba K., Imoto K., Tanabe T. and Numa S. (1991) Primary structure and functional expression from complementary DNA of a brain calcium channel. Nature 350, 398–402. Mynlieff M. and Beam K. G. (1992) Characterization of voltage-dependent calcium currents in mouse motoneurons. J. Neurophysiol. 68, 85–92. Mynlieff M. and Beam K. G. (1992) Developmental expression of voltage-dependent calcium currents in identified mouse motoneurons. Devl Biol. 152, 407–410. Protti D. A. and Uchitel O. D. (1993) Transmitter release and presynaptic Ca2+ currents blocked by the spider toxin omega-Aga-IVA. NeuroReport 5, 333–336. Randall A. D. and Tsien R. W. (1995) Pharmacological dissection of multiple types of Ca2+ channel currents in rat cerebellar granule neurons. J. Neurosci. 15, 2995–3012. Simpson P. B., Chaliss R. A. J. and Nahorski S. R. (1995) Neuronal Ca2+ stores: activation and function. Trends Neurosci. 18, 299–306. Starr T. V. B., Prystay W. and Snutch T. P. (1991) Primary structure of a calcium channel that is highly expressed in the rat cerebellum. Proc. natn. Acad. Sci. U.S.A. 88, 5621–5625. Tsien R. W., Lipscombe D., Madison D. V., Bley K. R. and Fox A. P. (1988) Multiple types of neuronal calcium channels and their selective modulation. Trends Neurosci. 11, 431–438. Umemiya M. and Berge A. J. (1994) Properties and function of low- and high-voltage-activated Ca2+ channels in hypoglossal motoneurons. J. Neurosci. 14, 5652–5660. Zhang J. F., Randall A. D., Ellinor P. T., Horne W. A., Sather W. A., Tanabe T., Schwartz T. L. and Tsien R. W. (1993) Distinctive pharmacology and kinetics of cloned neuronal Ca2+ channels and their possible counterparts in mammalian CNS neurons. Neuropharmacology 32, 1075–1088. (Accepted 9 March 1998)
PII:
Neuroscience Vol. 87, No. 3, pp. 719–730, 1998 Copyright 1998 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306–4522/98 $19.00+0.00 S0306-4522(98)00165-1
CALCIUM CHANNEL SUBTYPES RESPONSIBLE FOR VOLTAGE-GATED INTRACELLULAR CALCIUM ELEVATIONS IN EMBRYONIC RAT MOTONEURONS F. SCAMPS,* S. VALENTIN,* G. DAYANITHI† and J. VALMIER*‡ *CNRS UPR 1142, Institut de Biologie, Boulevard Henri IV, 34060 Montpellier, France †CNRS UPR 9055, Biologie des Neurones Endocrines, CCIPE, 141 rue de la Cardonille, 34094 Montpellier Ce´dex 4, France Abstract––The central role of electrical activity and Ca2+ influx in motoneuron development raises important questions about the regulation of Ca2+ signalling induced by voltage-dependent Ca2+ influx. In the purified embryonic rat motoneuron preparation, we recorded barium currents through voltageactivated Ca2+ channels using the whole-cell configuration of the patch-clamp technique. We found that motoneurons express at least four types of high-voltage-activated Ca2+ channels, based on their kinetics, voltage-dependences and pharmacological properties. Of the sustained Ca2+ current activated at 0 mV from a holding potential of 100 mV, approximatively 45% was omega-conotoxin-GVIA (1 µM) sensitive, 25% was omega-agatoxin-IVA (30 nM) sensitive and 20% was nitrendipine (250 nM) sensitive. The residual current, after applying these three antagonists, was an inactivating current that differs from classical T-type Ca2+ currents. Based on this pharmacology, changes in intracellular free Ca2+ concentrations were then monitored by Fura 2 digital imaging microspectrofluorimetry. Upon K+ depolarization, the intracellular Ca2+ transient induced by the activation of each type of Ca2+ channel appeared to be quantitatively proportional to their Ca2+ influx. The existence of a calcium-induced calcium release mechanism through activation of caffeine-, ryanodine-sensitive intracellular stores was then investigated. High doses of caffeine and low doses of ryanodine failed to increase intracellular free calcium concentrations and low concentrations of caffeine and high concentrations of ryanodine did not affect K+-induced intracellular free calcium concentration transients indicating both the absence of Ca2+-gated Ca2+-release channels and of a Ca2+-induced Ca2+ release mechanism. Together, these data provide evidence that embryonic motoneurons express multiple Ca2+ channels that function as important regulators of intracellular Ca2+ signalling and may be involved in their development. 1998 IBRO. Published by Elsevier Science Ltd. Key words: calcium channels, intracellular free calcium, ryanodine, caffeine, spinal motoneuron.
Plasmalemmal voltage-activated Ca2+ channels induce intracellular free Ca2+ concentrations ([Ca2+]i) increase and in this way, control many aspects of neuronal function including generation of Ca2+-dependent action potentials, neurotransmitter release, regulation of neuronal death, synapse formation and elimination, phenotypic differentiation and gene expression.1,10,16,28,31 It has become clear that voltage-dependent Ca2+ transients may reflect not only the entry of Ca2+ through plasmalemmal Ca2+ channels but also Ca2+ release from intracellular stores of the endoplasmic/sarcoplasmic reticulum by a Ca2+-induced Ca2+ release (CICR) mechanism (for review see Ref. 26). First described in striated muscle ‡To whom correspondence should be addressed. Abbreviations: Aga-IVa, omega-agatoxin-IVA; [Ca2+]i, intracellular free Ca2+ concentration; CICR, Ca2+induced Ca2+ release; DHP, dihydropyridine; E, embryonic day; GVIA, omega-conotoxin-GVIA; HEPES, N-2hydroxyethylpiperazine-N -2-ethanesulphonic acid; HP, holding potential; HVA, high-voltage-activated; IBa, barium current; ICa, calcium current; LVA, low-voltageactivated; TEACl, tetraethylammonium chloride.
cells, interest in neuronal CICR has grown with the recognition that it prolongs and amplifies the rise in [Ca2+]i initiated by Ca2+ entry and regulates [Ca2+]i oscillations (for review see Ref. 27). Schematically, Ca2+ stores in neurons responsible for releasing Ca2+ can be separated into at least two types based on their Ca2+ release channel, namely the caffeine- and ryanodine-sensitive Ca2+ channel responsible for the CICR following voltage-dependent Ca2+ entry and the inositol-1,3,4-triphosphate-sensitive Ca2+ channel (for review see Ref. 40). Neuronal voltage-activated Ca2+ channels (ICa) were first classified as low-voltage-activated (LVA) ICa and high-voltage-activated (HVA) ICa: a dihydropyridine (DHP)-sensitive L-type ICa and an omega-conotoxin-GVIA (GVIA)-sensitive N-type ICa (for review see Ref. 42). Then two additional HVA ICa were described which were sensitive to omega-Aga-IVA (Aga-IVa). One, identified in Purkinje neurons30 and referred to as P-type, is blocked with high affinity (IC50 =2–10 nM).34 The other, referred to as Q-type, is blocked with low affinity.21,39 In addition to these different types of
719
720
F. Scamps et al.
currents, a DHP/GVIA/Aga-IVa-resistant HVA ICa, referred to as toxin-resistant ICa, has also been observed.22,44 Recent data suggest that each type of Ca2+ channel is functionally different from one neuron to another and in a single neuron, controls specific pathways for Ca2+ signalling, thus supporting different functions. However, an understanding of the relationship between transmembrane Ca2+ entry, intracellular [Ca2+]i transient and cytoplasmic Ca2+ stores at single cell level is just begining to emerge. In embryonic spinal motoneurons, Ca2+ entry induced by electrical activity has been shown to play a central role in regulating in vivo neuromuscular activity which in turn regulates naturally occurring neuronal death and elimination of supernumerary synapses.17 Although these observations raise many questions about Ca2+ signalling in embryonic motoneurons, to date, no information is available on the regulation of motoneuron intracellular Ca2+ transients by Ca2+ influx. In a previous study, we have shown that it was possible to dissect, using a simple method of purification, the properties of the ICa expressed by mammalian motoneurons and to provide, for the first time, direct measurements of a Aga-IVa-sensitive, probably P-type, Ca2+ current in embryonic rat motoneurons.23 The aim of the present study was to characterize, in this model, the Ca2+ influx through voltage-activated Ca2+ channels, the [Ca2+]i rise induced by membrane depolarization and the possible involvement of a CICR mechanism. EXPERIMENTAL PROCEDURES
All animal experiments were carried out in accordance with the European Communities Council Directive of 24 November 1986 and all efforts were made to minimize animal suffering. Cell cultures Spinal motoneurons from embryonic day 15 (E15) Sprague–Dawley rat embryos (CERJ, Le Genest St Isle, France) were purified by a two-step metrizamide-panning method, as previously described.5 Briefly, ventral spinal cords were dissociated after trypsin digestion and centrifuged over 6.5% metrizamide cushions. The large cells that were not dense enough to pass through the metrizamide were further enriched by immunopanning on Petri dishes coated with antibody to the p75 neurotrophin receptor (low-affinity nerve growth factor receptor), which is specifically expressed by motoneurons at this stage. As described elsewhere,18 all (>95%) of the purified neurons were labelled by antibodies to p75 or to L14, a motoneuron-specific intracellular lectin. A subpopulation of these (50–90%) expressed Islet-1, another motoneuron marker,11,23 and were thus unambiguously identified as motoneurons. To confirm that they were motoneurons, the cells recorded initially were localized on the dish, fixed and stained using an antibody to Islet-1.23 In subsequent experiments, neurons to be patch-clamped were selected, under phasecontrast microscopy, on the basis of their large, multipolar morphology and phase-bright rounded cell body. No differences were seen in comparing the currents from labelled and unlabelled motoneuron-like cells.
Electrophysiological recordings Twenty-four hours after motoneurons seeding, whole-cell recordings were obtained at 20–22C under conditions optimized to ensure isolation of Ca2+ channel currents from other voltage-activated currents. To block Na+ channels, we replaced extracellular Na+ by tetraethylammonium chloride (TEACl) and added tetrodotoxin (1 µM; Sigma, France) to the bath medium. K+ currents were blocked by extracellular TEACl and by internal CsCl. The bathing solution contained: TEACl, 116 mM; BaCl2, 5 mM; HEPES 10 mM; glucose 10 mM; pH 7.35 adjusted with TEA-OH. Recording pipettes were filled with the following solution: CsCl, 134 mM; 1,2-bis(2-aminophenoxy)ethane-N,N,N ,N ,-tetraacetic acid, 20 mM; HEPES, 10 mM; Mg-ATP, 3 mM; Mg-GTP, 1 mM; glucose, 10 mM; at pH 7.35 (adjusted with CsOH). The pipette resistances were c3 MÙ. Whole-cell currents were recorded using a Biologic RK300 patch-clamp amplifier (Claix, France). After seal formation (resistance 5–10 GÙ) and membrane disruption, series resistance was estimated from the capacitative transient evoked by a +10 mV test pulse. Membrane capacity was calculated as Cm=ô/Rs, with ô being the time-constant of capacitative transient and Rs the series resistance. Mean Rs was 16.02.3 MÙ, n=23. Because current amplitudes were always c1 nA, voltage errors resulting from the uncompensated series resistance were negligible (<6 mV). Cells having unstable leak >100 pA were usually discarded, because this phenomenon often induced current run down. Leak was estimated at the end of an experiment by adding 50 µM NiCl2 and CdCl2 to inhibit the remaining inward current. From measurement of the apparent reversal potential of the calcium current when using a ramp protocol (500 ms duration from 100 mV to +50 mV), we obtained a 41 mV negative voltage drift (n=3), when estimated up to 10 min after establishment of the whole-cell configuration. All experimental parameters were controlled with a computer equipped with a Tecmar Labmaster analog interface (Axon Instruments, Chatillon, France). Data acquisition and analyses were performed using the pclamp software (v 6.03, Axon instruments). Current signals were sampled at 5 kHz and filtered at 3 kHz before digitization and storage. Dye loading and measurements of [Ca2+]i Dye loading and measurement of [Ca2+]i was performed as previously described.8 Briefly, the culture dishes were washed, loaded by incubation with 3 µM Fura-2AM and 0.05% w/v Pluronic F-127 (Molecular Probes Inc., U.S.A.) in Locke buffer (NaCl 140 mM; MgSO4 1.2 mM; CaCl2 1.8 mM; glucose 10 mM; KH2PO4 5 mM; HEPES-NaOH 10 mM; pH 7.25) at 34C for 35 min. Subsequently, loaded cells were washed with Locke buffer and fluorescence measurements were performed with buffers maintained at 20–22C. [Ca2+]i levels in single cells were measured using the digital imaging microfluometry system (Axon Instruments, Chatillon, France) based on an inverted microscope (Nikon) equipped with epifluorescence. Interference filters of 340/10 nm and 380/10 nm were alternately mounted on the filter wheel and the excitation light beam was deflected through an oil-immersion objective (40 0.75 NA, Nikon, Champigny-sur-Marne, France). Calibration of the fluorescence ratios was not attempted, because the present study aimed to evaluate relative changes in [Ca2+]i following activation of different types of Ca2+ conductances rather than obtaining absolute values of [Ca2+]i. [Ca2+]i variations were obtained from the ratio between fluorescence emission that resulted from emission at 340 and 380 nm.15 Drugs Nitrendipine (Bayer AG) was dissolved in dimethylsulfoxide to make concentrated stock solutions (10 mM) stored at 20C. Controls showed that the solvent has no effects on Ca2+-channel currents at the final dilutions used here
Calcium movements in motoneurons (<0.05%). The omega-conotoxin-GVIA (Sigma), omegaAga-IVA (Peptide International, Louisville, KY and Pfizer, U.S.A.) and omega-conotoxin-MVIIC (Latoxan, Bachem Rosans, France), were dissolved in bidistilled water at 0.1 mg/ml, and 1mg/ml respectively to make stock solutions. NiCl2 and CdCl2 were prepared in bidistilled water at 10 mM as stock solution. Test solutions were prepared daily using aliquots from frozen stores to obtain the working concentrations. Nitrendipine and GVIA were applied using a perfusion system (200 µm inner diameter capillary tubing; flow rate of 0.3 ml/min) placed in the vicinity of the cell. A pneumatic pressure pump (Miniframe PPS2, Medical System, NY) was employed for Aga-IVa application (30 to 50 ms ejection time, 200 ms pause). This system allows the use of small volumes of peptide solutions placed in glass microelectrodes (resistance <1 MÙ). Direct applications of Aga-IVa to obtain the same final concentration was also used and no difference was noted. Results are expressed as meanS.D. RESULTS
Characterization of voltage-activated Ca2+ channel currents Distribution and characterization of Ca2+ currents were studied in more than 30 acutely dissociated motoneurons under experimental conditions designed to eliminate other voltage-dependent currents (see Experimental Procedures). Ba2+ was used as a charge carrier on the basis of the larger amplitude of HVA Ca2+ currents and to minimize Ca2+ current rundown. In this condition, Ba2+ currents (IBa) were stable for up to 15 min at a holding potential of 100 mV allowing us to perform electrophysiological and pharmacological experiments. Up to 14 h after seeding, all time- and voltage-dependent currents were blocked by addition of 100 µM NiCl2 and CdCl2 to the external solution and were therefore considered to be Ca2+ channel currents. Immediately after rupture of the membrane, the holding potential (HP) was fixed at 100 mV and depolarizing test pulses to various potentials were delivered every 10 or 15 s. In these conditions, inward currents were recorded in motoneurons showing an amplitude increase (run-up) to reach in 1–2 min a steady-state from which all the investigations were performed. Whereas HVA currents were present in all cells recorded, less than 10% (range 5–30%) of the motoneurons express a LVA current, as checked when applying a ramp protocol from 100 mV to +50 mV. The population expressing only HVA currents represents an excellent model in which to observe HVA currents unimpeded by LVA current. Figure 1A shows the current–voltage (I–V) relationships of the peak and sustained components of IBa obtained from two holding potentials: HP=100 mV where all conductances were fully activated and HP=60 mV which is roughly the values of the membrane resting potential (587 mV; n=10, under 5.4 mM extracellular K+ and 140 mM intracellular K+; fire action potentials were never observed at rest). From HP=100 mV, the peak and
721
sustained currents were larger at all potentials than elicited from HP=60 mV (Fig. 1Aa, b). In the absence of LVA, the threshold for peak and sustained currents activation was similar when activated from 100 mV than from 60 mV (around 40 mV). However, the maximal IBa evoked between 0 to +10 mV from HP=100 mV showed a greater inactivation than it did from HP=60 mV (Fig. 1Ba). So, the current elicited at +10 mV had two components. One component was relatively insensitive to holding potential and was less inactivated during the test pulse (sustained current). The other was sensitive to holding potential and showed greater inactivation (transient current) (Fig. 1Bb). Therefore, the amplitude of IBa evoked at +10 mV from HP=60 mV represented 67%19% and 32% 17% of the sustained and the transient IBa evoked from HP=100 mV (n=9), respectively. To characterize the different Ca2+ current components in embryonic rat motoneurons, we tested the effects of different Ca2+ channel antagonists. Fig. 1C shows that it was possible, on a single motoneuron, to identify four different Ca2+ current subtypes: a (0.3 µM) nitrendipine-, a (1 µM) GVIA-, a (250 nM) Aga-IVa-sensitive and a residual ICa. In this motoneuron, the nitrendipine-sensitive, the GVIAsensitive, the Aga-IVa-sensitive and the residual ICa represented, respectively, 10, 20, 57 and 10% of the total sustained ICa elicited at +10 mV. We next analysed in detail the electrophysiological and pharmacological properties of these ICa components. The GVIA toxin is a selective blocker of the N-type ICa current in different neuronal preparations.42 When applied at 1–5 µM, a saturating concentration for motoneurons (data not shown), GVIA blocked, respectively, 4610% and 4812% of the transient and the sustained IBa elicited by 0 or +10 mV depolarizing pulse in all motoneurons tested (n=10) (Fig. 2Aa). Digital subtraction between control current traces and current traces in the presence of 1 µM GVIA revealed both an inactivating and a sustained component in all neurons tested (Fig. 2Ab). I–V relationships of the peak and sustained components of the GVIA-sensitive current from HP=100 mV is plotted in Fig. 4A. Under control conditions, when HP was varied from 100 mV to 60 mV, a sustained component was partially inactivated as shown in Figs 1B and 2Ba. The sustained component sensitive to HP represented 3319% of global current (n=9). In the presence of 1 µM GVIA, the remaining sustained component was mostly insensitive to HP changes (Fig. 2Bb). Therefore, the difference in the effect induced by depolarizing the HP on the global sustained component may be due to a preferential inactivation of the N-type Ca2+ channel current. The presence of L-type Ca2+ channels has been previously demonstrated in this preparation using the DHP agonist Bay K 8644.23 In order to further analyse this current and to quantify its contribution
722
F. Scamps et al.
Fig. 1. Characterization of ICa in E15 motoneurons. (A) (a) Current–voltage relationships of peak current (Ipeak) measured as the the peak inward current relative to the zero current. Ipeak-voltage relationships were plotted from a 100 mV holding potential (HP) or a 60 mV HP. (b) Current–voltage relationships of the sustained current (Isus) measured at 250 ms relative to the zero current. Isus-voltage relationships were constructed from a 100 mV or 60 mV HP. (Same cell in (a) and (b), Cm=42 pF). (B) (a) Representative current traces elicited from a 100 mV or a 60 mV HP to a +10 mV test pulse. Current traces are taken from A. Global current was composed of a transient component, measured as the difference between peak current and sustained current, and a sustained component, measured relative to zero current. (b) Corresponding difference current obtained by subtracting the current traces measured at 60 mV from the current traces measured at 100 mV. (C) Pharmacological dissection of global ICa. (a) Time-course effects of 0.3 µM nitrendipine, 30 nM Aga-IVa, 1 µM GVIA and 50 µM Cd applications. The same cell was sensitive to all drugs. (b) Corresponding current traces. HP=100 mV; test pulse=+10 mV. (Cm=50 pF).
to the global calcium current, we used nitrendipine, a DHP antagonist of the L-type calcium current. Since nanomolar concentrations of DHP antagonists are
sufficient to block nearly all cardiac and neuronal L-type channels,9 we used 300 nM nitrendipine, a saturating concentration for the DHP high-affinity
Calcium movements in motoneurons
723
Fig. 2. Characterization of the GVIA- and nitrendipine-sensitive currents. (A) (a) Current traces obtained from a 100 mV holding potential to a +10 mV test pulse. Application of 1 µM GVIA induced a decrease in transient and sustained currents. Further addition of 0.3 µM nitrendipine induced a more prominent decrease in sustained current than in transient current. Addition of 50 µM Cd allowed leak correction. (b) Subtraction of the current traces obtained in the presence of GVIA from the control current traces gives the GVIA-sensitive current (from (a)). This current appeared as an inactivating current. (c) Subtraction of the current traces obtained in the presence of nitrendipine from the current remaining after GVIA treatment gives the nitrendipine-sensitive current (from (a)). This current appeared as a slowly activating current. A similar current was obtained when nitrendipine was first applied (see Fig. 1C). (Cm=32 pF). (B) (a) Effects of varying the holding potential from 100 mV (1) to 60 mV (2) on control current measured at +10 mV test potential. (same cell as in A). (b) Effects of holding potential after GVIA application. Note that only the transient component remained sensitive to holding potential. (c) Effects of holding potential after cell treatment with GVIA and nitrendipine.
binding site, even at HP=100 mV, to quantify L-type IBa. At this concentration, nitrendipine inhibited a slowly activating, non inactivating component (Fig. 2Aa,c). The mean inhibition of the sustained
component (measured at 250 ms) was 2817% at +10 mV (n=8). I–V relationships of the ‘‘peak’’ (measured at 10 ms) and sustained components from HP=100 mV is plotted in Fig. 4B. It should be
724
F. Scamps et al.
Fig. 3. Characterization of the Aga IVa-sensitive current and the residual current. (A) (a) Cell was incubated with 1 µM GVIA and 0.3 µM nitrendipine to inhibit N- and L-type currents, respectively. Application of 100 nM Aga-IVa induced a decrease in transient and sustained currents. (b) Subtraction of the current traces obtained in the presence of Aga-IVa from the current remaining after GVIA and nitrendipine treatment gives the Aga-IVa-sensitive current. This current appeared as a slowly inactivating current. (Cm=43 pF). (B) (a) Effects of varying the holding potential from 100 mV (1) to 60 mV (2) on the current remaining after GVIA and nitrendipine treatment. As shown in Fig. 2Bc, only the transient component was sensitive to HP. (b) The residual current (insensitive to GVIA, nitrendipine and Aga-IVa) was composed of a transient component sensitive to HP and a sustained component insensitive to HP.
noted that DHP-sensitive channels have been reported to be rather insensitive to holding potential.20 In this model, the amplitude of the nitrendipinesensitive ICa current elicited from HP=60 mV is about 100% of that from a HP of 100 mV confirming previous study (Fig. 2Bc). Therefore, the decrease of the global sustained component induced by depolarizing the HP of the cell (Figs 1B, 2Ba) was not due to the contribution of the L-type current. We next tested the effect of Aga-IVa, since a functional Aga-IVa-sensitive (probably P-type) component has been previously identified in this model.23 To determine the saturating concentrations of this peptide antagonist, increasing concentrations were applied to motoneurons. A substantial decrease in ICa was found with 30 nM Aga-IVa (249%; n=8) (Fig. 3Aa). The Aga-IVa-sensitive current inactivated very slowly (Fig. 3Ab). This blocking effect had a fast time-course (Fig. 1Ca). Increasing the concentration to 100 nM and 200 nM caused no or little additional decrease in ICa (respectively, 299%; n=3 and 2510%, n=5). I–V relationship of the peak and
sustained components from HP=100 mV is plotted in Fig. 4C. The amplitudes of the Aga-IVa-sensitive current elicited from either HP=100 mV or 60 mV were identical (Fig. 3B). A current remained in the presence of nitrendipine (0.3 µM), GVIA (1 µM) and Aga-IVa (200 nM) (Fig. 3Aa, Bb). This inactivating current activated around 40 mV, reached a maximum amplitude between 0 mV and +10 mV and was always present at +40 mV and thus had the characteristics of a HVA current. I–V relationship of the peak and sustained components from HP=100 mV is plotted in Fig. 4D. Recent reports have shown that µM concentration of omega-conotoxin-MVIIC could inhibit, in some neurons, DHP-, GVIA-, Aga-IVa-insensitive IBa. Further application of 10 µM omega-conotoxinMVIIC to the pre-incubating medium had no effect (n=3). However this toxin-resistant IBa was completely blocked by addition of 50 µM CdCl2 (see Fig. 1C). The transient component represented 757% of the residual ICa (n=5). At the resting potential (60 mV), this transient ICa was greatly
Calcium movements in motoneurons
725
Fig. 4. Current–voltage relationships of the different components of global ICa. (A) Current–voltage relationships of the peak and sustained components of the GVIA-sensitive current (Cm=33 pF). (B) Current–voltage relationships of the peak and sustained components of the nitrendipine-sensitive current (Cm=32 pF). (C) Current–voltage relationships of the peak and sustained components of the Aga-IVa-sensitive current (Cm=51 pF). (D) Current–voltage relationships of the peak and sustained components of the resistant current (Cm=47 pF).
inactivated (Fig. 3Bb). So, it probably represented part of the transient component of the global ICa that inactivated at depolarizing HP of 60 mV. Characterization of [Ca2+]i transients through voltageactivated Ca2+ channels Based on the pharmacology of these ICa, we next studied how Ca2+ entry through voltage-activated Ca2+ channels affected [Ca2+]i and its relation with Ca2+ stores. First, this was done by comparing [Ca2+]i elevations produced by depolarizations before and after applying the three antagonists. Cells were depolarized with 100 mM K+ rather than whole-cell voltage-clamp techniques to minimize the disturbance of Ca2+ homeostasis induced by intracellular perfusion. Here again, neurons to be recorded were selected on the basis of their large, multipolar morphology and phase-bright rounded cell body (see Experimental Procedures and Ref. 23). Under resting conditions, embryonic motoneurons exhibited a
fluorescence ratio within the range 0.5 to 0.7 which was stable throughout the measuring period (up to 30 min). In these neurons, 100 mM K+ evoked a rapid increase in [Ca2+]i in all neurons tested (n=180). The subsequent applications of high K+ did not (or little) alter the peak amplitude of the [Ca2+]i response (as shown in Fig. 6). However, the successive applications of 100–500 nM Aga-IVa, 1 µM GVIA and 250 nM nitrendipine inhibited, respectively, 3011%, 4314% and 1511% of the increase in [Ca2+]i induced by depolarizations (n=58) (Fig. 5). These results confirm the functionality of a P-type Aga-IVa-sensitive, a N-type GVIA-sensitive and a L-type nitrendipine-sensitive voltage-activated ICa, previously characterized using the whole-cell configuration of the patch-clamp technique. When 1 µM GVIA was applied before Aga-IVa and nitrendipine, the percentage of the toxin-sensitive influx was not different (respectively, 4710%, 298%; n=32) (Fig. 5B). We concluded that, here again, there was
726
F. Scamps et al.
Fig. 5. Effects of various Ca2+ channel antagonists on K+-induced [Ca2+]i responses. All traces were monitored in a single neuron. Neurons were first exposed to control K+ (100 mM) depolarizations and [Ca2+]i responses were monitored (left trace). Subsequently, the same neuron was pre-incubated for 2 min with various antagonists as indicated in the trace and then challenged with high K+. After exposure to each antagonist followed by K+ challenge, the dish was washed under fast perifusion. Similar type of results were obtained in 58 and 32 neurons, respectively, for A and B protocols.
no significant cross-reaction between the effects of Aga-IVa and GVIA. We next analysed the presence of caffeine- and ryanodine-sensitive Ca2+ stores and a possible interaction between these stores and voltage-dependent Ca2+ entry. Bath application of caffeine (at 10, 20 and 50 mM), known to activate Ca2+ release from these stores,26 failed to increase [Ca2+]i in E15 rat motoneurons (n=40) (Fig. 6A). To confirm the absence of functional caffeine-sensitive stores in the [Ca2+]i transient induced by depolarization, the effect of caffeine at low concentrations, that potentiated
Ca2+ release through Ca2+-gated release channels, was investigated. Pre-incubating E15 motoneurons with 1 mM caffeine had no effect on [Ca2+]i transients initiated by 50 mM K+ depolarization (n=40) (Fig. 6B). Moreover, the kinetics of [Ca2+]i transients were unchanged before and after caffeine application. We concluded that E15 motoneurons did not express functional caffeine-sensitive intracellular Ca2+ channels and stores. Since some ryanodine receptors are insensitive to caffeine,13 we tested the effects of ryanodine at concentrations known to directly activate (nM
Calcium movements in motoneurons
727
Fig. 6. Effects of caffeine and ryanodine on [Ca2+]i (A) A selected neuron was subjected to high K+ as control [Ca2+]i response and then exposed to caffeine (20 nM) and ryanodine (1 µM) followed by a second K+ challenge (as shown by arrow). Note that the neuron showed no [Ca2+]i response. (B, C) K+-induced [Ca2+]i responses were monitored in the continued presence or absence of 1 mM caffeine (B) or 10 µM ryanodine (C). Similar profiles were obtained in at least 15–20 neurons.
concentrations) or inhibit (µM concentrations) these receptors.19 Direct application of 500 nM ryanodine failed to increase basal [Ca2+]i in this preparation (n=40) (Fig. 6A). We then tested higher concentrations of ryanodine taking into account the fact that ryanodine binding to its receptor has been reported to be use-dependent.40 Ten micromolar ryanodine had no effect on the [Ca2+]i transient induced by 100 mM K+ even when repetitive K+ depolarizations were carried out (Fig. 6C). Here again, no change in the kinetics of the [Ca2+]i transients were observed before and after the drug
application, confirming the absence of functional ryanodine-sensitive intracellular Ca2+ channels and stores. Taken together, these results indicate that embryonic motoneurons are devoid of both caffeineand ryanodine-sensitive Ca2+ channels and stores. DISCUSSION
This study provides the first description of the intracellular Ca2+ transients induced by Ca2+ influx through voltage-activated Ca2+ currents in motoneurons. Based on their pharmacology, kinetic and
728
F. Scamps et al.
voltage-dependence, at least four types of HVA Ca2+ channel currents were characterized and their contribution to the global Ca2+ current quantified: an N-type omega-conotoxin-GVIA-sensitive current, an L-type dihydropyridine-sensitive current, a P-type omega-Aga-IVA-sensitive current and a residual current, resistant to these three antagonists. These currents were present in all embryonic motoneurons recorded here and therefore most likely represent important molecular characteristics of these developing cells. All were available at resting potential and regulated [Ca2+]i when activated by depolarization. The absence of functional intracellular ryanodinesensitive Ca2+ channels and stores reinforces the notion that voltage-activated Ca2+ channel currents may have a major functional importance in regulating [Ca2+]i level during embryonic development. The only previous report on direct measurement of HVA ICa in identified embryonic mammalian motoneurons only described the presence of DHP- and a GVIA-sensitive HVA Ca2+ channels (but Aga-IVa was not tested).36 These data contrast with those obtained in postnatal mammalian motoneurons for which, extracellular and patch-clamp recordings indicate that, at least, motoneurons and motor nerve terminal possess two other ICa insensitive to both DHP and GVIA.38,43 As has been suggested previously,23 we demonstrated unambiguously the existence of four types of HVA Ca2+ currents in embryonic rat motoneurons that differ by their pharmacological properties, their kinetics, their voltagedependences and their different contributions to the total Ca2+ currents and Ca2+ transients. The strength of our demonstration depends essentially on the specificity of actions of the different antagonists. The specificity of Aga-IVa and GVIA has been demonstrated in a large number of clones and neuronal preparations.2 In our hands, even used at concentrations of five- or 10-fold higher than their saturating concentrations, these two antagonists failed to inhibit an additional ICa component. Although non-specific effects have been described for DHP, we and others have shown that at nanomolar concentrations, these compounds affect mainly neuronal L-type Ca2+ channels.9,32 Furthermore, when used together, little cross-reactivity has been seen between these three antagonists in this preparation despite a great variability from one cell to another (see Results). Although Aga-IVa inhibits, in addition to the P-type ICa, another neuronal ICa named Q-type,21,39 and the cloned á1A ICa (presumed to be Q-type) expressed in oocytes,35,41 the component of the non-L, non-N ICa sensitive to Aga-IVa in this preparation appears clearly similar to the P-type Ca2+ channel current first described in cerebellar Purkinje neurons.30 Indeed, the present results show that, despite the use of synthetic Aga-IVa, 30 nM of the toxin was nearly a saturating concentration and its time-course of inhibition was rapid. The current
component inactivated slowly and it is mainly insensitive to the holding potential. Finally, the use of omega-conotoxin-MVIIC did not affect ICa in embryonic motoneurons preincubated with GVIA and Aga-IVa. All these characteristics are in accordance with the notion that the current inhibited by Aga-IVa in motoneurons are mostly a P-like ICa. In the presence of nitrendipine, GVIA and AgaIVa at concentrations known to inhibit L, N and P Ca2+ channels in E15 motoneurons, we have characterized the properties of a non-L, non-N, non-P HVA residual Ca2+ channel current. Although this residual current was insensitive to these four antagonists, a property which is common with LVA T-type Ca2+ channels, these two types of Ca2+ channels appeared clearly different from the T-channel. The residual current was comprised of both fast inactivating and sustained components whereas the T-type current is rapidly inactivating (a feature of all T-type IBa characterized in motoneurons33,36 and in neurons6,25 to date). When comparing the two currents in E15 motoneurons, the residual current has a higher threshold of activation than the T-type current that is completely inactivated at HP=60 mV and the former is unaffected by 500 µM amiloride which completely blocks the T-type current in this preparation (Hivert and Valmier, unpublished observations). Conversely, the functional properties of the residual current in motoneurons seem quite similar to those of both cerebellar granule and dorsal root ganglion neurons9,39 and those of the á1E channel when expressed in oocytes.44 Whole-cell patch-clamp recordings, with Ba2+ as a charge carrier, indicated that all embryonic motoneuron HVA Ca2+ currents were activated at membrane resting potential. In order to analyse the role of these different classes of Ca2+ channels in regulation of cytosolic calcium levels in intact motoneurons, we have measured intracellular calcium transients induced by K+ depolarization, using the Ca2+ indicator Fura 2 and digital imaging microfluometry. This transient represents a physiologically relevant index of Ca2+ channel function. Our results show that all these channels were able to increase intracellular [Ca2+]i upon K+ depolarization. Since depolarizing Ca2+ influx has been shown to induce CICR in neurons by activating caffeine- and/or ryanodinesensitive intracellular stores,27 we next analysed the presence of a caffeine-induced [Ca2+]i increase in this preparation. Even at high concentrations (up to 50 mM), caffeine was unable to reveal a caffeine induced-[Ca2+]i response. Since low concentrations of caffeine can potentiate the response of caffeinesensitive channels to K+-induced Ca2+ influx in sympathetic neurons,12 the effect of K+ in the continued presence of 1 mM caffeine was analysed. Such a strategy failed to potentiate the K+-induced Ca2+ transient. These data demonstrate the absence of functional caffeine-sensitive Ca2+ release channels and caffeine-sensitive stores.
Calcium movements in motoneurons
Some ryanodine-sensitive Ca2+-release channels have been demonstrated to be insensitive to caffeine.13 To test this hypothesis in E15 motoneurons, we have analysed the effect on [Ca2+]i of 500 nM ryanodine, known to activate its receptor and of 10 µM ryanodine, known to inhibit it. Since the effects of ryanodine have been reported to be usedependent, we compared the increases and the kinetics of [Ca2+]i transients after repetitive K+ depolarizations and no significant changes were observed. These data demonstrate unambiguously the absence of functional caffeine-/ryanodine-sensitive Ca2+ release channels and ryanodine-sensitive stores. The absence of effect of Ca2+-gated Ca2+ channels indicates that E15 rat motoneurons are devoid of classical CICR mechanism to regulate [Ca2+]i. This pattern of [Ca2+]i regulation may be specific of this stage of development. Indeed, developmental regulation of other [Ca2+]i-regulating mechanisms have been observed in different systems3,4,7,14,21,29,33,37 and CICR mechanisms have been demonstrated in differentiating spinal neurons in the absence of caffeine-sensitive stores.24 This latter study shows that, although caffeine was unable to increase [Ca2+]i, depletion of internal Ca2+ stores in the endoplasmic reticulum greatly reduced elevation of calcium produced by depolarization. Furthermore, caffeine insensitivity disappeared with the in vitro maturation of amphibian spinal neuron. For these reasons, the existence of a unknown developmental-regulated caffeine-independent CICR mechanism are currently under investigation in E15 rat motoneurons.
729 CONCLUSION
Since Ca2+ influx play a pivotal role in the development of the neuromuscular system, we have characterized voltage-activated Ca2+ channels of embryonic rat motoneurons and explored their contribution in the regulation of the intracellular free Ca2+ transients induced by depolarization. The use of different Ca2+ channel antagonists, i.e. nitrendipine, GVIA, Aga-IVa and CdCl2, demonstrated the presence of at least four different types of HVA calcium channel currents, respectively the L-, N-, P- and R-types. All these calcium channels had the ability to induce intracellular calcium transient upon 100 mM K+ depolarization. Neither caffeine nor ryanodine affected voltage-dependent Ca2+ influx indicating the absence of a Ca2+-induced Ca2+-release mechanism through ryanodine-sensitive Ca2+ channels. These results demonstrate a great diversity of functional pathway for Ca2+ influx and [Ca2+]i signaling through voltage-activated Ca2+ channels in embryonic rat motoneurons which in turn may support the regulation of their activity during development. Acknowledgements—We thank A. Roig and C. Barrere for technical assistance and cell cultures and B. Rouvie`re for typing the manuscript. This work was supported by ‘‘Association Franc¸aise contre les Myopathies’’ (AFM), the Centre National de la Recherche Scientifique (CNRS), the Institut National de la Sante´ et de la Recherche Me´dicale (INSERM). S. Valentin was sponsored by the Ministe`re de la Recherche et de la Technologie (MRT).
REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
Bading H., Ginty D. D. and Greenberg M. E. (1993) Regulation of gene expression in hippocampal neurons by distinct calcium signaling pathways. Science 260, 181–186. Birnbaumer L., Campbell K. P., Catterall W. A., Harpold M. M., Hofmann F., Horne W. A., Mori Y., Schwartz A., Snutch T. P., Tanabe T. and Tsien R. W. (1994) The naming of voltage-gated calcium channels. Neuron 13, 505–506. Brandt C. J., DeLeon S., Martin D. R. and MacLennan D. H. (1987) Adult forms of the Ca2+ ATPase of sarcoplasmic reticulum expression in developing skeletal muscle. J. biol. Chem. 262, 3768–3774. Braun K., Scheich H., Zuschratter W., Heizmann C. W., Matute C. and Streit P. (1988) Postnatal development of parvalbumin-, calbindin-, and adult GABA-immunoreactivity in two visual nuclei of zebra finches. Brain Res. 475, 205–217. Camu W., Bloch-Gallego E. and Henderson E. (1993) Purification of spinal motoneurons from chicken and rat embryos by immunopanning. Neuroprotocols 2, 191–199. Carbone E. and Lux H. D. (1984) A low voltage-activated, fully inactivating Ca channel in vertebrate sensory neurons. Nature 310, 501–502. Cimino M. J. F. C. and Weiss B. (1990) Ontogenetic evelopment of calmodulin mRNA in rat brain using in situ hybridization histochemistry. Devl Brain Res. 54, 43–49. Dayanithi G., Widmer H. and Richard P. (1996) Vasopressin induced intracellular Ca2+ increase in isolated rat supraoptic cells. J. Physiol., Lond. 490, 713–717. Diochot S., Richard S., Baldy-Moulinier M., Nargeot J. and Valmier J. (1995) Dihydropyridines, phenylalkylamines and benzothizepines block N-, P/Q and R-type calcium currents. Pflu¨gers Arch. 431, 10–19. Dunlap K., Luebke J. I. and Turner T. J. (1995) Exocytotic Ca2+ channels in mammalian central neurons. Trends Neurosci. 18, 89–98. Ericson J., Thor S., Edlund T., Jessel T. M. and Yamada T. (1992) Early stages of motor neuron differentiation revealed by expression of homebox gene Islet-1. Science 256, 1555–1560. Friel D. D. and Tsien R. W. (1992) A caffeine- and ryanodine-sensitive Ca2+ store in bullfrog sympathetic neurones modulates effects of Ca2+ entry on [Ca2+]i. J. Physiol., Lond. 450, 217–246. Giannini G., Clementi E., Ceci R., Marziali G. and Sorrentino V. (1992) Expression of a ryanodine receptor-Ca2+ channel that is regulated by TGF-â. Science 257, 91–94. Ghishan F. K. and Arab N. (1988) Active calcium transport by intestinal endoplasmic reticulum during maturation. Am. J. Physiol. 254, G47–80.
730 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44.
F. Scamps et al. Grynkiewicz G., Poenie M. and Tsien R. Y. (1985) A new generation of Ca2+ indicators with greatly improved fluorescence properties. J. biol. Chem. 260, 3440–3450. Gu X. and Spitzer N. C. (1995) Distinct aspects of neuronal differenciation encoded by frequency of spontaneous Ca2+ transients. Nature 375, 784–787. Hamburger V. (1977) The developmental history of the motor neuron. Neurosci. Res. Prog. Bull. 15S, 1–37. Henderson C. E., Camu W., Mettling C., Gouin A., Poulsen K., Karihaloo M., Rullamas J., Evans T., MaMahon S. B., Armanini M. P., Berkmeier L., Phillips H. S. and Rosenthal A. (1993) Neurotrophins promote motor neuron survival and are present in embryonic limb bud. Nature 363, 266–270. Henzi V. and MacDermott A. B. (1992) Characteristics and function of Ca2+- and inositol-1,4,5-triphosphatereleasable stores of Ca2+ in neurons. Neuroscience 46, 251–273. Hess P. (1990) Calcium channels in vertebrate cells. A. Rev. Neurosci. 13, 337–356. Hilaire C., Diochot S., Desmadryl G., Baldy-Moulinier M., Richard S. and Valmier J. (1996) Opposite developmental regulation of P- and Q-type calcium currents during ontogenesis of large diameter mouse sensory neurons. Neuroscience 75, 1219–1229. Hilaire C., Diochot S., Desmadryl G., Richard S. and Valmier J. (1997) Toxin-resistant calcium currents in embryonic mouse sensory neurons. Neuroscience 80, 267–276. Hivert B., Bouhanna S., Diochot S., Camu W., Dayanithi G., Henderson C. E. and Valmier J. (1995) Embryonic rat motoneurons express a functional P-type voltage-dependent calcium channel. Int. J. dev. Neurosci. 13, 429–436. Holliday J., Adams R. J., Sejnowski T. J. and Spitzer N. (1991) Calcium-induced release of calcium regulates differentiation of cultured spinal neurons. Neuron 7, 787–796. Huguenard J. R. (1996) Low-threshold calcium currents in central nervous system neurons. A. Rev. Physiol. 58, 329–348. Kostyuk P. and Verkhratsky A. (1994) Calcium stores in neurons and glia. Neuroscience 63, 381–404. Kuba K. (1994) Ca2+-induced Ca2+ release in neurones. Jap. J. Physiol. 44, 613–650. Larmet Y., Dolphin A. C. and Davies A. M. (1992) Intracellular calcium regulates the survival of early sensory neurons before they become dependent on neurotrophic factors. Neuron 9, 563–574. Leberer E., Seedorf U. and Pette D. (1992) Neural control of gene expression in skeletal muscle. Calcium sequestering proteins in developing and chronically stimulated rabbit skeletal muscle. Biochem. J. 239, 295–300. Llinas R., Sugimori M., Hillman D. E. and Cherksey B. (1992) Distribution and functional significance of the P-type voltage-dependent Ca2+ channels in the mammalian central nervous system. Trends Neurosci. 15, 351–355. Mattson M. P. and Barger S. W. (1993) Roles for calcium signaling in structural plasticity and pathology in the hippocampal system. Hippocampus 3, 73–88. McCarthy R. T. and TanPiengco P. E. (1992) Multiple types of high-treshold calcium channels in rabbit sensory neurons: high-affinity block of neuronal L-type by nimodipine. J. Neurosci. 12, 2225–2234. McCobb D. P., Best P. M. and Beam K. G. (1989) Development alters the expression of calcium currents in chick limb motoneurons. Neuron 2, 1633–1643. Mintz I. M., Venema V. J., Swiderek K. M., Lee T. D., Bean B. P. and Adams M. E. (1992) P-type calcium channels blocked by the spider toxin ù-Aga-IVA. Nature 355, 827–829. Mori Y., Friedrich T., Kim M. S., Mikami A., NakaI} J., Ruth P., Bosse E., Hofmann F., Flockerzi V., Furuichi T., Mikoshiba K., Imoto K., Tanabe T. and Numa S. (1991) Primary structure and functional expression from complementary DNA of a brain calcium channel. Nature 350, 398–402. Mynlieff M. and Beam K. G. (1992) Characterization of voltage-dependent calcium currents in mouse motoneurons. J. Neurophysiol. 68, 85–92. Mynlieff M. and Beam K. G. (1992) Developmental expression of voltage-dependent calcium currents in identified mouse motoneurons. Devl Biol. 152, 407–410. Protti D. A. and Uchitel O. D. (1993) Transmitter release and presynaptic Ca2+ currents blocked by the spider toxin omega-Aga-IVA. NeuroReport 5, 333–336. Randall A. D. and Tsien R. W. (1995) Pharmacological dissection of multiple types of Ca2+ channel currents in rat cerebellar granule neurons. J. Neurosci. 15, 2995–3012. Simpson P. B., Chaliss R. A. J. and Nahorski S. R. (1995) Neuronal Ca2+ stores: activation and function. Trends Neurosci. 18, 299–306. Starr T. V. B., Prystay W. and Snutch T. P. (1991) Primary structure of a calcium channel that is highly expressed in the rat cerebellum. Proc. natn. Acad. Sci. U.S.A. 88, 5621–5625. Tsien R. W., Lipscombe D., Madison D. V., Bley K. R. and Fox A. P. (1988) Multiple types of neuronal calcium channels and their selective modulation. Trends Neurosci. 11, 431–438. Umemiya M. and Berge A. J. (1994) Properties and function of low- and high-voltage-activated Ca2+ channels in hypoglossal motoneurons. J. Neurosci. 14, 5652–5660. Zhang J. F., Randall A. D., Ellinor P. T., Horne W. A., Sather W. A., Tanabe T., Schwartz T. L. and Tsien R. W. (1993) Distinctive pharmacology and kinetics of cloned neuronal Ca2+ channels and their possible counterparts in mammalian CNS neurons. Neuropharmacology 32, 1075–1088. (Accepted 9 March 1998)