The Journal of Neurophysiology Vol. 78 No. 6 December 1997, pp. 3359-3370
Copyright ©1997 by the American Physiological Society

Regulation of N- and L-Type Ca²⁺ Channels in Adult Frog Sympathetic Ganglion B Cells by Nerve Growth Factor In Vitro and In Vivo

Saobo Lei, William F. Dryden, and Peter A. Smith

Department of Pharmacology and Division of Neuroscience, University of Alberta, Edmonton, Alberta T6G 2H7, Canada

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Lei, Saobo, William F. Dryden, and Peter A. Smith. Regulation of N- and L-type Ca²⁺ channels in adult frog sympathetic ganglion B cells by nerve growth factor in vitro and in vivo. J. Neurophysiol. 78: 3359-3370, 1997. To examine mechanisms responsible for the long-term regulation of Ca²⁺-channels in an adult neuron, changes in whole cell Ba²⁺ current (I_Ba) were examined in adult bullfrog sympathetic ganglion B cells in vitro. Cells were cultured at low density in defined, serum free medium. After 15 days, total I_Ba was similar to the initial value, whereas I_Ba density was reduced by ~36%, presumably due to an increase in neuronal surface area. By contrast, I_Ba density remained constant after 6-15 days in the presence of murine -NGF (200 ng/ml), and total I_Ba was almost doubled. Inclusion of cytosine arabinoside (Ara-C; 10 µM) to inhibit proliferation of nonneuronal cells, did not affect the survival of neurons in the absence of nerve growth factor (NGF) nor did it attenuate I_Ba. Ara-C did not prevent the effect of NGF on I_Ba. There were three independent components to the action of NGF; during 6-9 days, it increased -conotoxin-GVIA-sensitive N-type I_Ba (I_Ba,N); increased nifedipine-sensitive L-type I_Ba (I_Ba,L) and decreased inactivation of the total Ba²⁺ conductance (g_Ba). The latter effect involved a selective decrease in the amplitude of one of the four kinetic components that describe the inactivation process. Total I_Ba was also 55.8% larger than control in the somata of B cells acutely dissociated from leopard frogs that had received prior subcutaneous injections of NGF. By contrast, injection of NGF antiserum decreased total I_Ba by 29.4%. There was less inactivation of g_Ba in B cells from NGF-injected animals than in cells from animals injected with NGF antiserum (P < 0.001). These data suggest that NGF-like molecule(s) play(s) a role in the maintenance of I_Ba in an adult amphibian sympathetic neuron; the presence of NGF may allow the neuron to maintain a constant relationship between cell size and current density. They also show that I_Ba inactivation in an adult neuron can be modulated in a physiologically relevant way by an extracellular ligand.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Nerve growth factor (NGF) is mandatory for the survival of sympathetic neurons and small nociceptive, sensory neurons at certain critical periods in their development (Davies 1996; Levi-Montalcini and Angelletti 1968; Lewin 1996). These neurons lose their absolute dependence on NGF for survival as they mature. After this, NGF becomes involved in the selection, specification, and maintenance of differentiated neuronal phenotypes (Lewin 1996; Lewin et al. 1992; Lindsay 1996).

Chalazonitis et al. (1987) showed that long-term exposure to NGF decreases action potential (ap) duration of neurons in organotypic cultures of dorsal root ganglia (DRG) from 13 d fetal mice. They proposed that this effect may reflect modulation of Ca²⁺ channel current (I_Ca) in neurons that no longer require NGF for survival. By contrast, Ritter and Mendell (1992) showed that treatment of neonatal rats with NGF in vivo from birth to 5 wk of age produced essentially the reverse effect; ap duration was increased. This effect was confined to high-threshold mechanoreceptor (HTMR) neurons and other types of sensory neuron were unaffected. It therefore remains to be determined whether NGF acts to increase or to decrease I_Ca in mature peripheral neurons whether findings in culture can be extrapolated to the in vivo situation, and whether NGF affects the number, the type, or the properties of Ca²⁺ channels.

Because disconnection of mammalian DRG neurons from their target tissues alters the expression of cell-type-specific proteins (Zhang et al. 1993), cell-surface markers (Persson et al. 1995), and neurotransmitter receptors (Abdulla and Smith 1997), there is no reliable method of identifying one specific subtype of sensory neuron in culture. Dissociated or cultured cells cannot, of course, be classified on the basis of their axonal conduction velocity (Villière and McLachlan 1996). Because as many as 43 different neuronal subtypes are found in mammalian DRG (Lewin 1996), the use of DRG cultures is precluded when exactly the same neuronal subtype is to be studied under different experimental conditions; in the present case, in the presence or absence of NGF. The problem is compounded by progressive changes in ion channel properties as neurons are maintained in culture (Scott and Edwards 1980; Traynor et al. 1992) and by the fact that NGF affects only one subtype of neuron in mammalian DRG in vivo (HTMRs) (Ritter and Mendell 1992).

Unlike DRG neurons, all sympathetic neurons respond to NGF. Murine NGF increases ap duration in B cells of adult bullfrog paravertebral sympathetic ganglia (BFSG) in explant culture (Traynor et al. 1992) in a manner that is reminiscent of its action on mammalian HTMR neurons in vivo (Ritter and Mendell 1992). We have therefore used BFSG B cells to analyze the regulation of Ca²⁺ channels in an adult neuron by NGF both in tissue culture and in vivo.

BFSG neurons are of two types: the smaller C cells and the larger B cells (Horn et al. 1988). When isolated, B cells can be identified on the basis of their size. The remarkable consistency of their electrophysiological characteristics (Adams et al. 1986; Jassar et al. 1993; Smith 1994) allows meaningful comparison of data from a control population of cells with that from a NGF-treated population. Because Ca²⁺ channel kinetics are better characterized in BFSG B cells than in any other vertebrate neuron (Elmslie et al. 1992; Jones and Marks 1989a,b; Jassar et al. 1993; Lipscombe et al. 1988; Sala 1991; Werz et al. 1993), we could investigate actions of NGF on specific, well-characterized kinetic components of channel function and thereby to gain a unique insight into its mechanism of action.

The experiments required the development of a defined-medium, low-density culture system, free of serum or other known growth factors. We found that B cells dissociated and cultured from the sympathetic ganglia of adult bullfrogs could be maintained for 2 wk under such conditions (Lei et al. 1995a,b). The cells were functionally intact at the end of a 15-day experimental period, and there was a significant increase in ap height (P < 0.001) at this time (Lei, unpublished observations). Progressive changes in Ca²⁺ channel properties in the B-cell population therefore could be studied and used as a reliable baseline against which to document and analyze the actions of NGF.

Because NGF decreases ap duration of mammalian DRG cells in culture (Chalazonitis et al. 1987) yet has essentially the reverse effect in vivo, where it selectively increases ap duration of HTMR neurons (Ritter and Mendell 1992), it was necessary to test whether effects of NGF on BFSG Ca²⁺ channels in vivo were the same as those seen in vitro. In vivo effects were examined by injecting NGF or NGF antiserum into the thighs of another species of smaller frog (Rana pipiens). The sympathetic ganglia of these frogs then were removed and the neurons dissociated so that the properties of Ca²⁺ channels in B cells could be examined. It is assumed that R. pipiens sympathetic ganglion (RPSG) B cells, like BFSG B cells, project to mucous glands in the skin of the thighs (Horn et al. 1988), so that subcutaneous injections of NGF or NGF antiserum would have access to B-cell axons and terminals. Some of the results have been published in abstract form (Lei et al. 1995a,b).

	METHODS

Abstract Introduction Methods Results Discussion References

Tissue culture

Medium-sized adult bullfrogs (10-15 cm; 250-350 g body wt) were maintained in running water at room temperature. Dissection, dissociation, and culture procedures were performed under aseptic conditions. After pithing, the ventral portion of the frog was swabbed with 9% iodine in ethanol. The animal was carefully eviscerated and the VIIth-Xth paravertebral sympathetic ganglia removed from both sides. Neurons were dissociated using a trypsin/collagenase procedure (Selyanko et al. 1990) and suspended in 3 ml of culture medium containing 73% L-15 medium (Gibco) in water supplemented with 10 mM glucose, 1 mM CaCl₂, 100 units/ml penicillin, and 100 µg/ml streptomycin. To minimize interactions between cells, the dissociated cells from one frog were distributed evenly into 3 ml of medium in each of 30 35-mm polystyrene tissue culture dishes (Nunc, Denmark, purchased from Gibco) without additional substrate. This yielded 5-10 neurons per dish. In experiments where neuron-enriched cultures were used, dissociated ganglion cells from one frog were preplated at high-density into two 35-mm tissue culture dishes. After 1-2 h, the nonneuronal cells adhered to the bottom of the dish. The nonadherent cells, which were primarily neurons, were collected, redistributed to 30 tissue culture dishes, and cultured in medium supplemented with 10 µM cytosine arabinoside (Ara-C). This antimetabolite inhibits proliferation of nonneuronal cells in amphibian embryos (Peralta et al. 1995). When used, NGF was added to the medium to a concentration of 200 ng/ml or, in a few experiments, 50 ng/ml. The dishes were placed in a humidified chamber and maintained at room temperature (22°C) for 15 days. Unless otherwise stated, the culture medium was replaced after 7 days.

In vivo experiments

In the interests of economy, leopard frogs (R. pipiens; 60-80 g body wt) were used in preference to bullfrogs. These received subcutaneous injections of 2.5 s NGF (-NGF; 1 µg/g body wt) or rabbit anti-2.5 s NGF fractionated antiserum (33 µg protein/g body wt) into the right thigh. Animals were injected on days 1, 3, 5, and 7 of the experimental period, and the somata of acutely dissociated neurons from the ganglia of the right sympathetic chain were examined electrophysiologically on day 9 or 10. Control data were obtained from B cells from animals that received comparable injections of rabbit serum (33 µg protein/g body wt) or saline.

Electrophysiology

Ca²⁺ channel current was recorded from both acutely dissociated and from cultured cells using the whole cell recording technique (Jassar et al. 1993). Ba²⁺ was the charge carrier. The external solution contained (in mM) 117.5 N-methyl-D-glucamine (NMG) chloride, 2.5 NMG-N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid (HEPES), and 2.0 BaCl₂ (pH 7.2). Internal solution consisted of (in mM): 76.5 NMG-Cl; 2.5, HEPES acid, 10Tris(hydroxymethyl)aminomethane (Tris)-bis-(o-aminophenoxy)N,N,N,N-tetraacetic acid (Tris-BAPTA), 5 Tris-ATP, and 4 MgCl₂, (pH 7.2). For recording, tissue culture medium was replaced with external solution at a flow rate of 2 ml/min. This flow rate allowed exchange of solutions within ~2 min. Medium- to large-sized cells with input capacitance (C_in) > 40 pF from either BFSG (Jassar et al. 1993) or R. pipiens sympathetic ganglion (RPSG) (Selyanko et al. 1990) were selected for recording. Because C_in of BFSG C cells has been found to be <20 pF (Kurenny et al. 1994) and that of RPSG C cells has been found to be <18 pF (Selyanko et al. 1990), we are confident that all recordings were made from B cells. An Axoclamp 2A amplifier (Axon Instruments, Burlingame, CA) in the single-electrode, discontinuous voltage-clamp mode was used to record I_Ba. The use of a "switching amplifier" minimizes series-resistance problems associated with large, rapidly activating currents because the amplifier headstage is clamped to the recorded membrane voltage (Jones 1987). Low-resistance electrodes (DC resistance 1-2 M in the external solution) were used to permit switching frequencies between 40 and 50 kHz and a clamp gain between 8 and 16 nA/mV. During data acquisition, the corner frequency of the filter was set to 10 kHz. Whole cell recordings were first obtained in the current-clamp "bridge" mode, and the amplifier switched to the single-electrode, discontinuous voltage-clamp mode for the study of I_Ba. The holding potential (V_h) was 80 mV unless otherwise stated. C_in was calculated by integrating the capacitive transient, which accompanied a 10-mV depolarizing command from 80 mV. For measurements involving peak currents, data were leak subtracted using a P/4 protocol (Jones and Marks 1989a). Leak subtraction was not done for tail current and kinetic analysis. There were no space-clamp problems with recordings from acutely isolated (spherical) cells but because the cultured cells grew processes, ~5-10% of them were subject to space-clamp error, which was apparent from distortion of the current traces. Recordings from such cells were discarded. Data were digitized with a Labmaster DMA interface and processed on an IBM-compatible computer with the pClamp suite of programs. Experimental records were filtered digitally at 4 kHz before making permanent records with an X-Y plotter. All data are presented as means ± SE. Student's two-tailed t-test or analysis of variance have been used to assess statistical significance. In graphs where no error bars are visible, the error bars are smaller than the symbols used to designate the data points.

NGF (2.5s, -NGF) was either purchased from Alimone Labs., Jerusalem, Israel, or kindly provided by Dr. R. B. Campenot (Department of Cell Biology and Anatomy, University of Alberta). -conotoxin GVIA was dissolved in the external solution to make a final concentration of 300 nM. Nifedipine was dissolved in dimethyl sulfoxide (DMSO) to make a 10-mM stock solution. This solution was diluted with external solution to obtain a final concentration of 10 µM for application from light-proof reservoirs to ganglion cells under subdued lighting conditions. Control experiments with the same concentration of DMSO in the external solution showed no effects on I_Ba. Rabbit anti-2.5sNGF-fractionated antiserum (Sigma lot number 065H8950) was provided as a white powder and reconstituted with sterile deionized water. All other drugs and chemicals were from Sigma.

	RESULTS

Abstract Introduction Methods Results Discussion References

Acutely isolated B cells were usually devoid of neurites but began to generate them after 2-3 days in culture even in the absence NGF (Fig. 1A). The neurites grew longer, thicker, and more branched as the time in culture increased. Cells survived for 15 days in the defined medium, serum-free system. The production of neurites coincided with marked increases in C_in. Figure 1B shows that C_in increased about twofold after 6 days in culture and then remained relatively constant for the duration of the experiment (15 days).

View larger version (62K): [in this window] [in a new window]   FIG. 1. A: phase contrast photomicrographs of an acutely isolated bullfrog paravertebral sympathetic ganglia (BFSG) B cell, a 12-day cultured cell, and a cell cultured for 12 days in the presence of 200 ng/ml nerve growth factor (NGF). Calibration bar is 100 µm and refers to all 3 pictures. B: changes in the input capacitance (Cin) of B neurons as they grow in "normal" (defined medium, serum-free) culture in the presence or absence of NGF C: changes in Cin of B neurons as they grow in neuron-enriched [cytosine arabinoside (Ara-C) treated] culture (n > 20 for all observations).

Inclusion of NGF in the culture did not obviously affect the growth of the somata or the neurites (Fig. 1A). and no significant difference in C_in was measured (Fig. 1B). It therefore would seem that exogenous NGF is neither necessary for neurite extension nor for survival of adult BFSG cells under our culture conditions.

Changes in I_Ba as B cells are maintained in culture

Figure 2 illustrates typical recordings of I_Ba from an acutely dissociated cell, a cell maintained in defined medium for 12 days, and a cell maintained in the presence of NGF for 12 days. Currents were evoked by a series of depolarizing voltage commands (V_h = 80 mV). Tail currents after the depolarizing pulses were recorded at 40 mV to slow their kinetics and facilitate their analysis (see legend to Fig. 4D).

View larger version (20K): [in this window] [in a new window]   FIG. 2. Typical Ca2+ channel currents (IBa) recorded from an acutely isolated BFSG B cell (A) and from cells maintained for 12 days in normal (defined medium, serum-free) culture with (B) or without (C) NGF.

View larger version (38K): [in this window] [in a new window]   FIG. 4. A-C: current-density voltage plots for IBa in acutely isolated B cells, cells cultured for 12 days, and cell cultured for 12 days in the presence of NGF. Three different holding potentials (Vh: 90, 60, and 40 mV) were used. n > 20 for data points. D: activation curves for gBa expressed as I/Imax in the 3 experimental situations (n > 20 for all data points). IBa was evoked at a series of different command potentials using 20-ms commands. Tail currents at the end of each pulse were recorded at 40 mV. Tail current amplitude was estimated by fitting the data to a monoexponential function to extrapolate the original amplitude. The solid line represents a plot of a Boltzmann expression of the form It()/It(max) = g/gmax = {1 + expze(VoV)/kT)}1 where It(v) is the amplitude of the IBa tail after a command to a voltage, ; It(max) is the maximum amplitude of IBa tails after commands to positive voltages. Ratio of these currents is equal to g/gmax i.e., the ratio of maximum gBa to that after a command to voltage, , or the fraction of Ca2+ channels open; z is the valency of the gating particle; e is the elementary electronic charge (1.602 × 1019 C), k = Boltzmann's constant(1.381 × 1023 VCK1), T = absolute temperature, and Vo is the potential for half-maximal activation. Deviation of the experimental data from this line probably reflects rapid inactivation (Sala 1991; Jassar et al. 1993). E: normalized inactivation curves for the 3 experimental conditions. IBa was evoked at 0 mV from various holding potentials. F: plot of original (nonnormalized) inactivation data shown in E (n = 22 for acutely isolated cells, n = 16 for cells cultured without NGF, n = 25 for cells cultured with NGF).

Figure 3A shows that there was a transient increase in total I_Ba during the first 3-6 days in culture (P < 0.01; asterisk in Fig. 3A). The current then steadily declined until that present after 15 days in culture was similar to that in acutely dissociated cells (P > 0.05). Inclusion of 200 ng/ml NGF in the culture medium almost doubled the total current. This effect was clearly apparent after 6 days and was maintained throughout the whole experimental period (15 days; Fig. 3A). When a lower concentration of NGF (50 ng/ml) was used, I_Ba seemed larger than that of control neurons yet displayed considerable variability, 200 ng/ml of NGF therefore was used for the remainder of the study.

View larger version (37K): [in this window] [in a new window]   FIG. 3. Time course of the change in IBa recorded at 0 mV from a Vh of 80 mV as B neurons grow in culture with or without NGF. A and B: changes in peak (total) current or current density for cells in normal (defined medium, serum-free) culture. C and D: changes in peak (total) current or current density for cells in neuron-enriched (Ara-C treated) culture (n > 20 for all observations). Definition of symbols in A refers to all 4 graphs.

Changes in current density (I_Ba normalized to cell size; C_in) also were examined (Fig. 3B) to account for differences in cell size among populations of neurons because the largest cells would be expected to have the largest currents. Because C_in increased dramatically during the culture period (Fig. 1B), the density of I_Ba in cells cultured without NGF decreased to ~36% of its initial value within 15 days (Fig. 3B). By contrast, the I_Ba density recorded after 15 days in the presence of NGF was similar to that of acutely isolated cells (P > 0.2; Fig. 3B). Thus the NGF induced-increase in total I_Ba (Fig. 3A) must have been balanced by the increase in cell size (Fig. 1B) so that there was no appreciable increase in current density (Fig. 3B).

Although the above experiments were carried out on isolated B cells in serum-free defined medium, the results pose two additional questions: could the effects of NGF be mediated via the release of other trophic substances from glial cells (Assouline et al. 1987; Lewin and Mendell 1993), macrophages (Guenard et al. 1991), or fibroblasts (Unsicker et al. 1987) and can the survival of neurons in cultures be attributed the release of trophic substances from these nonneuronal cells? To address these questions, the effect of NGF was reexamined in neuron-enriched cultures with media supplemented with Ara-C (10 µM). Figure 3C shows that total I_Ba remains relatively constant throughout the 15-day experimental period and that inclusion of NGF more than doubled the total current as it did in the absence of Ara-C (Fig. 3A). Figure 3D illustrates that I_Ba density in the neuron-enriched cultures was maintained or slightly increased in the presence of NGF yet declined in its absence. Because both total I_Ba and I_Ba density responded similarly in normal (Fig. 3, A and B) and in neuron-enriched cultures (Fig. 3. C and D), the possible presence of nonneuronal cells had little or no effect on the response of cultured neuronsto NGF.

Some trophic support may have been derived from nonneuronal cells because the transient increase in total I_Ba seen after 3-6 days in cultured cells (asterisk in Fig. 3A) was not seen in neuron-enriched cultures (Fig. 3C).¹ The possible presence of nonneuronal cells had little or no effect on neurite production because cell capacitance of cells in neuron-enriched culture increased (Fig. 1C) as did that of neurons in regular culture (Fig. 1B).

Effects of NGF on Ca²⁺-channel kinetics

Figure 4, A-C, illustrates the current-voltage relationships of acute cells and 12-day cultured cells with and without NGF. Experiments were carried out using V_h equaling 90, 60, and 40 mV, and data are expressed as I_Ba density. Several points emerge from the data; peak current density generally occurred at 0 mV for acutely dissociated cells and for cells cultured with or without NGF. This implies that the manipulations have little or no effect on the voltage-dependence of g_Ba activation and hence on its activation kinetics. This observation is confirmed by the activation curves obtained from tail current amplitudes (Fig. 4D, see figure legend for further details). There is no "shoulder" on the I-V relationship for acutely isolated (Fig. 4A), 12-day cultured (Fig. 4B), or NGF-treated (Fig. 4C) cells within the 70 to 20 mV range. This confirms that low-voltage activated T-type channel current (I_Ba,T) (Fox et al. 1987) is not observed in acutely isolated BFSG B cells (see Jassar et al. 1993; Jones and Marks 1989a) and that it does not appear during culture or in response to NGF (see Garber et al. 1989). More significantly, however, shifting V_h altered the I-V relationships in different ways in the acute, cultured, and the NGF groups. This implies that steady state inactivation of I_Ba is changed as cells grow in culture and that inactivation of I_Ba is altered by NGF. In acutely isolated cells, changing V_h from 90 to 40 mV reduces peak I_Ba density by 57% (Fig. 4A). By contrast, shifting V_h of 12-day cultured cells from 90 to 40 mV reduces I_Ba density by 77% (Fig. 4B). The situation in NGF-treated cells is similar to that in acutely isolated cells; shifting V_h to 40 mV reduces I_Ba density by 45%.

These results are reflected in the steady state, normalized inactivation plots shown in Fig. 4E. V_h was set to a range of prepulse potentials from 110 to +10 mV for 20 s before applying a test pulse to 0 mV. The normalized current evoked during the test pulse is plotted against V_h. Compared with acutely dissociated cells, the inactivation curve for 12-day cultured cells was shifted to the left. This means that voltage commands evoked from V_h of 70 to 30 mV would generate less I_Ba as a result of increased steady state inactivation. By contrast, the inactivation curve for NGF-treated cells was shifted to the right so that more current would be available as a result of reduced inactivation.

These changes in inactivation cannot, however, explain all of the changes in I_Ba that occur as the neurons grow in the presence of NGF. Figure 4F shows the inactivation curves plotted as absolute currents. The current attained in NGF-treated cells could never be attained in cultured cells no matter how negative a V_h were to be used because in these cells, the available current reaches a maximum when the holding potential is between 90 and 110 mV. Thus NGF causes an increase in the total amount of I_Ba by decreasing inactivation and by a second process (or processes, see further) that is/are independent of the effect on inactivation.

Analysis of changes in inactivation

Inactivation of g_Ba in BFSG B cells involves four kinetically distinguishable processes: fast, intermediate, slow, and very slow inactivation (Jassar et al. 1993; Jones and Marks 1989b; Werz et al. 1993). We therefore examined which aspects of the inactivation process were affected by growing cells in culture and how these changes were affectedby NGF.

The voltage-dependence of the rate of g_Ba inactivation was determined by using a series of 500-ms conditioning pulses to various potentials before application of a standard 75-ms test pulse to 0 mV. Typical current records from an acutely isolated cell, a 12-day cultured cell, and a cell cultured for 12 days with NGF are illustrated in Fig. 5, A-C, respectively. Inactivation of g_Ba during the conditioning pulse was greater and more rapid in the 12-day cultured cells compared with acutely isolated or NGF-treated cells. Data from all cells studied are summarized in Fig. 6, A-C, which display the peak and end-of-pulse current density for conditioning pulses at various potentials and the peak current density recorded in the subsequent test pulse to 0 mV. In agreement with Jones and Marks (1989b) and Jassar et al. (1993), minimal current flows in the test pulse after conditioning pulses to 0 or 10 mV where I_Ba is maximal. Maximal inactivation (minimal size of test pulse current) for acute, cultured, and NGF cells occurs between 10 and 0 mV. In cultured cells in the absence of NGF, I_Ba (at 10 mV) inactivates by 50.9 ± 1.9% (n = 32) during a 500-ms depolarizing pulse (Fig. 6B). This inactivation is significantly greater than that observed for acutely dissociated cells(29.1 ± 1.6%, n = 24, P < 0.01; Fig. 6A) and for cells cultured for 12 days with NGF (25.9 ± 2.1%, n = 27, P < 0.01; Fig. 6C). The voltage dependence of inactivation of g_Ba in acute, 12-day cultured, and 12-day NGF-treated cells are normalized, replotted, and compared in Fig. 6D (see Jones and Marks 1989b). The difference between the maximal and minimal test current (maximal amount of inactivation) was defined as 1 and the difference between the maximal test current and other test current at different conditioning command potentials was normalized to this value. Comparison of the data shows that g_Ba inactivation in 12-day cultured cells was more pronounced at positive potentials(P < 0.01 from +10 to +70 mV). Inclusion of NGF in the culture medium attenuated inactivation in the cultured cells at both negative potentials (P < 0.01 from 60 to 20 mV) and at positive potentials (P < 0.01 from +10 to +70 mV).

View larger version (29K): [in this window] [in a new window]   FIG. 5. Recordings of Ca2+channel currents (IBa) to show voltage dependence of rate of inactivation in BFSG B cells. Cells were clamped by a series of 500-ms conditioning pulses to various potentials before application of a standard 75-ms test pulse to 0 mV. Typical recordings from an acutely isolated cell (A) and B cells maintained for 12 days in normal (defined medium, serum-free) culture with (B) or without (C) NGF.

View larger version (39K): [in this window] [in a new window]   FIG. 6. Voltage dependence of rate of inactivation of IBa in acutely isolated BFSG B cells (A) and B cells maintained for 12 days in normal (defined medium, serum-free) culture with (B) or without (C) NGF. Data were obtained from experiments such as those illustrated in Fig. 5. Graphs in A-C show changes in peak and end-of-pulse current density recorded during 500-ms conditioning pulses to potentials as indicated on the abscissa as well as peak current densities recorded in subsequent test pulses to 0 mV (n > 20 for all observations). D: normalized summary of data. Difference between maximum currents in test pulses (which followed conditioning pulses to 80 mV) and minimum currents in the test pulses (which followed conditioning pulses to 10 mV) were defined as 1 for each of the plots in A-C. Test pulse currents recorded after conditioning pulses to other potentials then are expressed as fractions of this value.

Figure 7A illustrates typical recordings from acute, 12-day cultured, and NGF treated cells that were used to study the detailed kinetics of inactivation at 0 mV and the rate of recovery at 80 mV. Long-duration (4 s) voltage commands were applied once every 15 s, and increasing delays were introduced between this initial 4-s pulse and subsequent brief pulses that were used to measure the rate of recovery from inactivation. Because there is a very slow component of g_Ba inactivation BFSG B cells, it was necessary to normalize the raw data records in Fig. 7A to the first response before plotting (Jassar et al. 1993; Smith 1994). The onset of inactivation during the pulse could be fit by intermediate (₂) and slow (₃) exponential time constants of amplitudes A₂ and A₃, respectively. A full numerical description of the current required the inclusion of a noninactivating component of amplitude A_o (Werz et al. 1993) (Table 1).²

View larger version (28K): [in this window] [in a new window]   FIG. 7. Kinetics of IBa inactivation. A: currents evoked by 4-s commands to 0 mV from a holding potential of 80 mV recorded from an acutely isolated BFSG B cell (a) and from cells maintained for 12 days in normal (defined medium, serum-free) culture with (b) or without (c) NGF. Note differences in gain for currents in these three traces as indicated by the scale bar. Four-second commands to 0 mV were evoked once every 15 s throughout each experiment. Each long pulse command was followed by a second brief (75 ms) pulse to 0 mV, and the delay between the 1st and 2nd pulses increased for each successive trial. Envelope of the small current amplitudes gave an index of the time course of recovery from inactivation. Because a slow inactivation process produces a progressive decline in successive 4-s pulses, data records in a-c are rescaled digitally to the amplitude of the first record. B: graph showing time course of slow inactivation. Abscissa is pulse number (to 0 mV for 4 s once every 15 s) and ordinate is normalized peak current amplitude. C: time course of recovery from inactivation at 80 mV obtained from envelope of amplitudes of brief (75 ms) pulses to 0 mV.

In 12-day cultured cells, g_Ba exhibited 89 ± 2% (n = 24) inactivation within a 4-s pulse. This is significantly greater than that seen in acutely dissociated cells (77 ± 2%, n = 22, P < 0.01) and in cells cultured with NGF (68 ± 3%, n = 21, P < 0.01). These effects resulted from changes in the amplitude (A₂) of the intermediate component of inactivation (₂). The values of ₂ and ₃ and the amplitude of the slowly inactivating component (A₃) were essentially unchanged (Table 1). The values of total inactivation shown in this table represent the total amount of inactivation that occurred in 4 s.

The rate of onset of very slow inactivation (Jassar et al. 1993) was assessed by measuring the size of peak currents induced by successive 4-s commands to 0 mV that were activated once activated every 15 s. Figure 7B illustrates the change in current for a series of 15 such pulses and demonstrates that there is no significant difference for the onset of slow inactivation of I_Ba (at 0 mV) among acutely dissociated cells and cells cultured for 12 days with or without NGF (P > 0.05). In view of the lack of a significant difference, the kinetics of very slow inactivation were not analyzed further.

Recovery from inactivation at 80 mV is less complete for 12-day cultured cells (without NGF) than for acutely isolated cells. Inclusion of NGF in the culture medium restores the recovery from inactivation to that observed in control cells (Fig. 7C).

NGF-induced changes in L-type Ca²⁺ channels

Studies in pheochromocytoma (PC12) cells (Lewis et al. 1993; Usowicz et al. 1990) and in rat embryonic forebrain neurons (Levine et al. 1995) have shown that L-type current (I_Ca,L), I_Ca,N, and conotoxin/dihydropyridine-resistant current (I_Ca,other) are differentially regulated by NGF. We therefore compared the pharmacology of I_Ba recorded from acutely isolated cells with that from 12-day cultured and 12-day NGF-treated cells. Figure 8A summarizes data from 12 acutely isolated cells that were exposed to -conotoxin GVIA (-CgTx; 300 nM) followed by nifedipine (10 µM). Voltage steps to 0 mV (from V_h = 80 mV) were elicited every 25 s during the perfusion of 300 nM -CgTx and 10 µM nifedipine and once every minute during recovery until the current stabilized. Current density was reduced from 67.5 ± 3.2 to 15.0 ± 5.9 pA/pF by -CgTx, and subsequent addition of nifedipine reduced current by another 3.1 ± 1.4 pA/pF. The average current in acutely isolated neurons was 82.4% -CgTx-sensitive I_Ba,N, 3% nifedipine-sensitive I_Ba,L, and 14.6% I_Ba,other. Similar experiments on cells cultured for 12 days in the absence (n = 12) and presence of NGF (n = 14) are shown in Fig. 8, B and C. Although the pharmacological sensitivity of the current in 12-day cultured cells (Fig. 8B) resembles that seen in acutely isolated cells (Fig. 8A), there is clearly a greater portion of I_Ba,L in NGF-treated cells (20.4%; Fig. 8C). The relative proportions of I_Ba,L, I_Ba,N, and I_Ba,other in cells in the three experimental situations are summarized in Fig. 8D.

View larger version (39K): [in this window] [in a new window]   FIG. 8. Pharmacological properties of IBa. Graphs of IBa density vs, time after exposure to -conotoxin GVIA (-CgTx; 300 nM) followed by nifedipine (10 µM). Data points represent mean current densities from 12 to 14 cells in each series of experiments. IBa was evoked at 0 mV from a Vh of 80 mV once every 20 s before, during, and after exposure to drugs. A-C are data from acutely isolated cells and from cells cultured for 12 days in the presence or absence of NGF. D: summary of data from A-C reexpressed as percentage of -CgTx-sensitive IBa,N, percentage of nifedipine-sensitive IBa,L, and percentage of -CgTx-/nifedipine-resistant IBa,other in the 3 experimental situations.

Thus in addition to decreasing inactivation of total I_Ba and increasing the total current by mechanisms other than decreased inactivation, NGF also appears to alter the proportion of nifedipine-sensitive I_Ba,L in BFSG B neurons.

It could be argued however, that the NGF-invoked increase in I_Ba,L accounts for the observed decrease in the percentage of the total g_Ba that inactivates within a 4-s pulse (see Fig. 7A and Table 1). This is because I_Ba,L may inactivate slowly (Fox et al. 1987) so that the appearance of such a component would attenuate the observed percentage inactivation of the total current in the presence of NGF. To address this, cells were studied in the presence of 10 µM nifedipine. In cells cultured for 12 days in the absence of NGF, the I_Ba recorded in the presence of nifedipine inactivated by 91.9 ± 1.1% (n = 22) after 4 s at 0 mV. By contrast, when NGF had been included in the cultures, the inactivation seen in the presence of nifedipine was reduced to 85.8 ± 1.8% (n = 22; P < 0.01). NGF therefore promoted significant reduction of inactivation under conditions where I_Ba,L was blocked. Thus NGF-induced changes in I_Ba,L do not account for the observed decrease in inactivation.

Time course of the action of NGF

Because there are now several reports of actions of NGF on Ca²⁺ channels that develop within minutes (Shen and Crain 1994; Wildering et al. 1995), we examined the time course of action of NGF on I_Ba in our system. The rate of decline in I_Ba density as cells were maintained in culture was unaffected by a 1-day exposure to NGF (Fig. 9A). With or without 1-day exposure to NGF, current density decreased to 61% of its control level in 5 days. By contrast, exposure to NGF for 3 or 6 days prevented the decline in current density (Fig. 9A). The rate of decline of current density following NGF removal was similar for both 6- and 3-day exposures. After 3-day exposure, current declined to 80% of control within 5 days of the withdrawal of NGF, and after 6-day exposure, current declined to 84% of control in 5 days.

View larger version (23K): [in this window] [in a new window]   FIG. 9. Graphs to show effects of pretreatment of cultures for 1, 3, or 6 days with NGF on IBa density (A) and rate of onset of increased IBa density after application of NGF to cultures that had been maintained for 6 days in its absence (B). Peak IBa recorded at 0 mV from a holding potential of 80 mV.

The rate of onset of the action of NGF was studied by examining its effect on cells that had been maintained in NGF-free medium for 6 days. The maximal effect of NGF took 9 days to develop (Fig. 9B).

In vivo experiments

Frogs (R. pipiens) received four injections of NGF (1 µg/g body wt) during a 7-day period, and I_Ba was examined 2-3 days later. In general, I_Ba in RPSG B cells was less than in BFSG B cells. Total I_Ba (at 0 mV from a V_h of 90 mV) was 55.8% larger in cells from NGF-injected frogs than in cells from control animals; [5.30 ± 0.42 nA (n = 41) compared with 3.40 ± 0.29 nA (n = 34) P < 0.001]. This difference was reflected as a difference in I_Ba density. Figure 10, A and B, shows the averaged I-V relationships. I_Ba density (from all values of V_h) is greater in the B cells from the NGF-treated animals. For example, I_Ba density (at 0 mV; V_h = 90 mV) in cells from NGF-treated animals is ~39% larger than that recorded from B cells of saline-injected animals [64.8 ± 5.2 pA/pF (n = 41) compared with 46.5 ± 1.2 pA/pF (n = 34) P < 0.02].

View larger version (32K): [in this window] [in a new window]   FIG. 10. Current-density-voltage plots for IBa in B cells acutely isolated from Rana pipiens sympathetic ganglia. Three different holding potentials (Vh: 90, 60, and 40 mV) were used. Key to symbols in D applies to all graphs, and unidirectional error bars have been used for clarity. In some places, error bars indicating SE are smaller than the symbols used to define the data points. A: data from cells from control (saline-injected) animals (n = 41). B: data from cells from frogs that had received NGF injections (n = 34). C: data from cells from frogs that had received rabbit serum (n = 32). D: data from cells from frogs that had received NGF antiserum (n = 32).

The situation with acutely dissociated cells, that do not have neurites, differs from the situation in culture, where neurites are produced and C_in increases (Fig. 1B). Although an increase in total I_Ba is seen in neurons that are cultured with NGF (Fig. 3A) and in neurons that are isolated from NGF-treated animals, I_Ba density is only increased in the latter situation. This is because current density is the ratio of total current to C_in and the C_in of B-cell somata from control animals [75.3 ± 4.0 pF (n = 34)] is no different from that of cells from NGF-treated animals [84.6 ± 3.9 pF; (n = 41;P > 0.05)]. By contrast, the increase in C_in seen in culture eventually matches the increase in total I_Ba so that the I_Ba density measured at 15 days is no different from the initial control value (see Figs. 1B and 3B).

Altering the V_h of control cells from 90 to 40 mV decreased peak I_Ba by 43% (n = 34), whereas the same change in V_h for cells from NGF-treated animals changed peak current by ~34% (n = 41; Fig. 10A). This difference and hence the difference in inactivation between the two groups is not significant (0.2 < P < 0.5).

Injection of NGF antiserum did not change the mean C_in of the acutely dissociated cell bodies of RPSG B cells (P > 0.5). The frogs received four injections of 33 µg protein/g body wt during a 7-day period, and I_Ba was examined 3 days later. The control group received four injections of 33 µg serum protein/g body wt during a similar time period. The C_in of B cells from serum-injected (control) animals was 81.8 ± 4.9 pF (n = 32) and that of B cells from NGF antiserum-treated animals was 85.2 ± 4.6 pF (n = 32). Total I_Ba (at 0 mV from a V_h of 90 mV) was smaller in acutely dissociated cells from NGF antiserum-injected animals than in cells from control animals; [2.4 ± 0.21 nA (n = 32) compared with 3.4 ± 0.29 nA (n = 34) P < 0.05; a 29.4% decrease]. The effect of NGF-antiserum on I_Ba is therefore essentially the opposite of the effect of NGF.

Figure 10, C and D, shows the averaged I-V relationships obtained from rabbit serum injected (control) RPSG B cells and from cells from animals treated with NGF antiserum. Peak current density (from all values of V_h) is smaller in the cells from the NGF antiserum-treated animals; I_Ba density (recorded at 0 mV from a V_h of 90 mV) in cells from NGF antiserum-treated animals is 28.9 ± 2.1 pA/pF (n = 32), whereas that from control cells is 45.8 ± 4.8 pA/pF (n = 32; P < 0.005). The antiserum therefore decreases the I_Ba density at this potential by ~37%.

Figure 10D also shows that altering the V_h of cells from NGF-antiserum-treated animals from 90 to 40 mV decreased peak I_Ba density by ~58% (n = 32), whereas the same change in V_h for cells from control (serum-injected) animals changed peak current density by 49% (n = 32). This difference and hence the difference in inactivation between the two groups is not statistically significant (0.2 > P > 0.1).

There is, however, a highly significant difference (P < 0.001) between g_Ba inactivation after NGF antiserum [58% from 28.9 ± 2.1 to 12.3 ± 1.4 pA/pF (n = 32); Fig. 10D] and that seen after NGF [34% from 64.8 ± 5.2 to 44.9 ± 4.2 pA/pF (n = 41); Fig. 10B].

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Our principal findings are that NGF treatment potentiates I_Ba in a sympathetic neuron by three independent mechanisms, one of these actions is novel in that it demonstrates that inactivation of I_Ba in an adult neuron can be modulated by an extracellular ligand, NGF selectively alters one particular kinetic component of the inactivation process, and effects of NGF seen in culture resemble those seen in vivo. These findings, together with the in vivo effects of NGF-antiserum are consistent with the hypothesis that NGF or similar molecules are involved in the maintenance of Ca²⁺ channel function in adult sympathetic neurons. We suggest that the presence of appropriate neurotrophic support allows the expression and activity of Ca²⁺ channels to keep pace with the growth or size of the neuron. Further analysis of the data provides insights into how these effects may be achieved.

Potentiation of I_Ba by NGF

The three mechanisms that contribute to NGF-induced potentiation of I_Ba in BFSG are increases in I_Ba,L, increases in I_Ba,N, and decreases in inactivation. These three effects reflect independent phenomena rather than different experimental manifestations of a single effect. The inactivation curves shown in Fig. 4F show that the NGF-induced increase in total I_Ba is not solely a consequence of decreased steady-state inactivation; if this were so, removal of inactivation by holding cells at 110 mV would restore current seen in 12-day cultured cells to the same value seen in NGF-treated cells. It is also unlikely that this NGF-induced increase in total current simply reflects the increase in I_Ba,L. NGF more than doubles the I_Ba evoked from 110 mV (approximately a 120% increase; see Fig. 4F), and if this were solely due to an increase in I_Ba,L, 55% of the current seen in NGF-treated cells would be L type. The amount of I_Ba,L in these cells is however much lower (20.4%; Fig. 8D). The NGF-induced increase in I_Ba,L cannot account for the overall decrease in I_Ba inactivation because NGF produced a decrease in inactivation of total current under conditions where L channels were blocked with a supramaximal concentration of nifedipine.

NGF-induced decrease in I_Ba inactivationa novel mechanism of modulation?

Although an increase in I_Ca inactivation has been seen during NGF-induced differentiation of PC12 cells (Streit and Lux 1987, 1990), to the best of our knowledge, regulation of g_Ca inactivation by an extracellular ligand has not previously been described in an adult neuron.

Treatment of BFSG B neurons with phosphatase inhibitors increases the amplitude (A₂) of the intermediate component of I_Ba,N inactivation (Werz et al. 1993) without major alteration of the other kinetic parameters: ₂, ₃, and A₃. This suggests that Ca²⁺ channels are subject to tonic phosphorylation in vivo and that preservation of phosphorylation of a site on I_Ba,N channels leads to increased inactivation. Because NGF attenuates inactivation by selectively reducing this same kinetic parameter (A₂), it may produce its effect by reducing phosphorylation of a site on the N-type Ca²⁺ channel. Indeed, Aoki et al. (1996) recently reported that NGF can increase the mRNA for a novel tyrosine phosphatase in PC12 cells.

Alteration in expression of -subunits of Ca²⁺-channel proteins (Varadi et al. 1991) would not adequately explain the present results because -subunits regulate both the inactivation and the activation kinetics of Ca²⁺ channels, whereas NGF affected only the inactivation process. It is possible, however, that NGF favors the appearance of slowly inactivating I_Ca,N channels that are characteristic of nerve terminals. This subtype of N-type Ca²⁺ channel derives from a splice variants of the _1B Ca²⁺-channel subunit gene (Fisher and Bourque 1996).

Those actions of NGF that are not a consequence of decreased inactivation, i.e., the increase in I_Ba,N seen from a V_h of 110 mV and its effects on I_Ba,L may reflect alterations in channel expression at the translational or transcriptional level. Perhaps mechanisms similar to those activated during differentiation of PC12 cells (Cavalié et al. 1994; Lewis et al. 1993) can be provoked in adult BFSG B cells by NGF. The NGF-induced increase in I_Ca may be relevant to the proposed involvement of increased Ca²⁺ influx in the transduction process for some growth factors (Levine et al. 1995) by activation of mitogen-activated protein kinase (Finkbeiner and Greenberg 1996; Rosen et al. 1994).

Maintenance of Ca²⁺ channels by NGF

Cultured BFSG B neurons extend neurites either in the absence or in the presence of NGF. In the absence of NGF, the total I_Ba remains roughly constant, but in its presence, total I_Ba increases such that current density³ keeps up with the growth and increased cell size. NGF does not therefore appear to initiate growth per se, but once neurite production occurs, it seems to regulate the properties of the neuron so that the demands of a greater neuronal size can be met. NGF injection in vivo may promote sprouting of the terminal fields of BFSG neurons as it does in mammalian DRG (Diamond et al. 1987); the increased I_Ba seen in cell bodies under these conditions may reflect increased channel synthesis to meet the demands of an enlarged neuron.

Although murine NGF promotes a pronounced enhancement of neurite outgrowth in serum supplemented explant cultures of whole BFSG (Kelly et al. 1989), no such response was seen in the present experiments (Fig. 1). This may be a consequence of the lack of nutritional support and/or the absence of other types of trophic influences in the low-density, defined medium culture system that we have used. The present data show, however, that the NGF-induced increase in g_Ca in BFSG is not secondary to NGF-induced neurite extension.

Physiological significance of NGF's action

The main advantage of tissue culture experiments on BFSG B cells is that one specific cell type can be identified so that the direct effects of NGF on a variety of aspects of Ca²⁺-channel function can be analyzed. Despite this, the use of an amphibian system to study the actions of a mammalian growth factor requires justification. Therefore the following should be noted. 1) A bivalent antibody that activates mammalian TrkA, TrkB, and TrkC receptors increases I_Ba in BFSG B cells in a similar fashion to NGF. Frog neurons therefore express neurotrophin receptors that are immunologically-similar to mammalian Trk receptors (Lei et al. 1997). 2) The gene coding for amphibian NGF has been cloned from Xenopus laevis and the predicted aminoacid sequence of the mature peptide bears significant similarity to mammalian NGF (Carriero et al. 1991). 3) Antibodies raised against mouse NGF can inhibit a neurotrophic action of denervated frog sciatic nerve on frog DRG neurons (Kuffler and Megwinoff 1994). The use of mammalian antibodies to block actions of amphibian NGF is justified by the observation that they bind to mature X. laevis NGF (Carriero et al. 1991).

Because adult neurons tend to dedifferentiate in culture, their response to NGF in vitro may no longer be relevant to their behavior in vivo. Our data show however, that NGF increases total I_Ba both in vitro and in vivo. Because of this similarity, the various indirect mechanisms whereby NGF might influence BFSG B cells (Assouline et al. 1987; Guenard et al. 1991; Korsching and Thoenen 1983; Lewin and Mendell 1993; Unsicker et al. 1987), do not seem to be invoked in vivo. These data, and the result that NGF antiserum decreases total I_Ba amplitude in vivo are consistent with the hypothesis that Ca²⁺ channels in BFSG are regulated in vivo by (amphibian) NGF.

Injections of NGF in vivo do not seem to decrease g_Ba inactivation as NGF does in vitro. A possible reason for this is that amphibian sympathetic neurons have access to a certain amount of target-derived NGF in vivo (Carriero et al. 1991) in much the same way as mammalian neurons (Nja and Purves 1978). It therefore may be more appropriate to compare cells from animals treated with anti-NGF to cells from animals treated with NGF. When this is done, there is a highly significant difference in inactivation.

The increased inactivation of g_Ba that is seen as the neurons are maintained in culture resembles the effect of axotomy on intact BFSG B neurons (Jassar et al. 1993). Moreover, the amplitude of the `intermediate' component (A₂) of inactivation is increased by both manipulations and ₂, ₃, and A₃ are little changed. It will be recalled from Table 1 that A₂ is the component that is decreased selectively by NGF. One hypothesis suggests that the changes induced in BFSG B neurons after disconnection with their target reflects loss of retrogradely supplied NGF (see Verge et al. 1996). Indeed, the electrophysiological properties of BFSG neurons are reestablished once target contact is restored (Kelly et al. 1988; Petrov et al. 1996). A selective effect of NGF on A₂ therefore might be predicted from the hypothesis that loss of target-derived NGF is involved in the effects of axotomy on N-type Ca²⁺ channels. This hypothesis cannot, however, explain all effects of axotomy on B cells because Na⁺ currents are increased both by axotomy (Jassar et al. 1993) and by treatment of cultures with NGF (Lei, unpublished observations). The data, however, do argue against the contrary hypothesis that axotomy-induced increases in g_Ba inactivation result from increased release of NGF as part of the inflammatory response that accompanies nerve injury (see Lewin and Mendell 1993). If this were so, NGF would increase rather than decrease g_Ba inactivation.

	ACKNOWLEDGEMENTS

  This work was supported by the Medical Research Council of Canada. S. Lei was supported by a Rick Hansen Man-in-Motion/Alberta Paraplegic Foundation studentship award.

	FOOTNOTES

¹   Because Ara-C failed to influence the current between 4 and 15 days, Ara-C has no direct toxic effects on cultured neurons. For example, for cells cultured for 12 days, I_Ba density in the presence of both Ara-C and NGF is not significantly different from that of the cells cultured with NGF alone (P > 0.05) ²   The terminology of Werz et al. (1993) has been used to describe the kinetics of I_Ca inactivation. The fastest component of inactivation (₁) of amplitude A₁ can only clearly be distinguished in cells that have been treated with phosphatase inhibitors. ³   Because current density is expressed as the ratio of I_Ba to C_in, the present experiments provide no information as to where new channels are expressed. Ca²⁺ channels normally are thought to be expressed in cell bodies and growth cones, and Lipscombe et al. (1988) have shown that they appear on neurites of BFSG in culture. It is not known to what extent channels in neurites and/or growth cones of cultured cells contribute to the total current measured at the cell body.

  Address reprint requests to P. A. Smith.

  Received 2 May 1997; accepted in final form 21 August 1997.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• ABDULLA, F. A., SMITH, P. A. Ectopic ₂-adrenoceptors couple to N-type Ca²⁺ channels in axotomized rat sensory neurons. J. Neurosci. 17: 1633-1641, 1997.[Abstract/Free Full Text]
• ADAMS, P. R., JONES, S. W., PENNEFATHER, P. S., BROWN, D. A., KOCH, C., LANCASTER, B. Slow synaptic transmission in frog sympathetic ganglia. J. Exp. Biol. 124: 259-285, 1986.[Abstract/Free Full Text]
• AOKI, N., YAMAGUCHI-AOKI, Y., ULLRICH, A. The novel protein-tyrosine kinase phosphatase PTP20 is a positive regulator of PC12 cell neuronal differentiation. J. Biol. Chem. 271: 29422-29426, 1996.[Abstract/Free Full Text]
• ASSOULINE, J. G., BOSCH, P., LIM, R., KIM, I. S., JENSEN, R., PANTAZIZ, N. J. Rat astrocytes and schwann cells in culture synthesize nerve growth factor-like neurite promoting factors. Dev. Brain Res. 31: 103-118, 1987.
• CARRIERO, F., CAMPIONI, N., CARDINALI, B., PIERANDREI-AMALDI, P. Structure and expression of nerve growth factor gene in Xenopus oocytes and embryos. Mol. Reprod. Dev. 15: 313-322, 1991.
• CAVALIÉ, A., BERNINGER, B., HAAS, C. A., GARCIA, D. E., LINDHOLM, D.., LUX, H. D. Constitutive upregulation of calcium channel currents in rat phaeochromacytoma cells: role of c-fos and c-jun. J. Physiol. (Lond.) 479: 11-27, 1994.[Abstract/Free Full Text]
• CHALAZONITIS, A., PETERSON, E. R., CRAIN, S. M. Nerve growth factor regulates the action potential duration of mature sensory neurons. Proc. Natl. Acad. Sci. USA 84: 289-293, 1987.[Abstract/Free Full Text]
• DAVIES, A. M. The neurotrophic hypothesis: where does it stand? Philos. Trans. R. Soc. Lond. B Biol. Sci. 351: 389-394, 1996.[Medline]
• DIAMOND, J., COUGHLIN, M., MACINTYRE, L., HOLMES, M., VISHEAU, B. Evidence that endogenous -nerve growth factor is responsible for the collateral sprouting, but not the regeneration, of nociceptive axons in adult rats. Proc. Natl. Acad. Sci. USA 84: 6596-6600, 1987.[Abstract/Free Full Text]
• ELMSLIE, K. S., KAMMERMEIER, P. J., JONES, S. W. Calcium current modulation in frog sympathetic neurones: L-current is relatively insensitive to neurotransmitters. J. Physiol. (Lond.) 456: 107-123, 1992.[Abstract/Free Full Text]
• FINKBEINER, S., GREENBERG, M. E. Ca²⁺-depenedent routes to Ras: mechanisms for neuronal suvival, differentiation and plasticity. Neuron 16: 233-236, 1996.[Medline]
• FISHER, T. E., BOURQUE, C. W. Calcium channel subtypes in the somata and axon terminals of magnocellular neurosecretory cells. Trends Neurosci. 19: 440-444, 1996.[Medline]
• FOX, A. P., NOWYCKY, M. S., TSIEN, R. W. Kinetic and pharmacological properties distinguishing three types of calcium currents in chick sensory neurones. J. Physiol. (Lond.) 394: 149-172, 1987.[Abstract/Free Full Text]
• GARBER, S. S., HOSHI, T., ALDRICH, R. W. Regulation of ionic currents in pheochromocytoma cells by nerve growth factor and dexamethasone. J. Neurosci. 9: 3976-3987, 1989.[Abstract]
• GUENARD, V., DINARELLO, C. A., WESTON, P. J., AEBISCHER, P. Peripheral nerve regeneration is impeded by interleukin-I receptor antagonist released from polymeric guidance channel. J. Neurosci. Res. 29: 396-400, 1991.[Medline]
• HORN, J. P., FATHERAZI, S., STOFER, W. D. Differential projections of B- and C-sympathetic axons in peripheral nerves of the bullfrog. J. Comp. Neurol. 278: 570-580, 1988.[Medline]
• JASSAR, B. S., PENNEFATHER, P. S., SMITH, P. A. Changes in sodium and calcium channel activity following axotomy of B-cells in bull-frog sympathetic ganglion. J. Physiol. (Lond.) 472: 203-231, 1993.[Abstract/Free Full Text]
• JONES, S. W. Sodium currents in dissociated bull-frog sympathetic neurones. J. Physiol. (Lond.) 389: 605-627, 1987.[Abstract/Free Full Text]
• JONES, S. W., MARKS, T. N. Calcium currents in bullfrog sympathetic neurons. I. Activation, kinetics and pharmacology. J. Gen. Physiol. 94: 151-167, 1989a.[Abstract/Free Full Text]
• JONES, S. W., MARKS, T. N. Calcium currents in bullfrog sympathetic neurons. II. Inactivation. J. Gen. Physiol. 94: 169-182, 1989b.[Abstract/Free Full Text]
• KELLY, M.E.M, BISBY, M. A., LUKOWIAK, K. Regeneration restores some of the altered electrical properties of axotomized bullfrog B-cells. J. Neurobiol. 19: 357-372, 1988.[Medline]
• KELLY, M.E.M., TRAYNOR, P., SMITH, P. A. Amphibian sympathetic ganglia as a model system for investigating regeneration in the vertebrate peripheral nervous system. Comp. Biochem. Physiol. A Physiol. 93: 131-140, 1989.
• KORSHING, S., THOENEN, H. Nerve growth factor in sympathetic ganglia and corresponding target organs of the rat: correlation with density of sympathetic innervation. Proc. Natl. Acad. Sci. USA 80: 3513-3516, 1983.[Abstract/Free Full Text]
• KUFFLER, D. P., MEGWINOFF, O. Neurotrophic influence of degenerated sciatic nerve on adult dorsal root ganglion neurons. J. Neurobiol. 25: 1267-1282, 1994.[Medline]
• KURENNY, D. E., CHEN, H., SMITH, P. A. Effects of muscarine on K⁺ channel currents in the C-cells of bullfrog sympathetic ganglion. Brain Res. 658: 239-251, 1994.[Medline]
• LEI, S., DRYDEN, W. F., SMITH, P. A. Nerve growth factor-induced reduction in Ba²⁺ current inactivation in adult sympathetic neurons is independent of neurite elongation (Abstract). Biophys. J. 68: A209, 1995a.
• LEI, S., DRYDEN, W. F., SMITH, P. A. Effects of nerve growth factor on N- and L-type Ca²⁺current in bullfrog sympathetic neurones (Abstract). Proc. Can. Fed. Soc. Exp. Biol. 38: 78, 1995b.
• LEVI-MONTALCINI, R., ANGELETTI, P. U. Nerve growth factor. Physiol. Rev. 48: 534-569, 1968.[Free Full Text]
• LEVINE, E. S., DREYFUS, C. F., BLACK, I. B., PLUMMER, M. R. Differential effects of NGF and BDNF on voltage-gated calcium currents in embryonic basal forebrain neurons. J. Neurosci. 15: 3084-3091, 1995.[Abstract]
• LEWIN, G. R. Neurotrophins and the specification of neuronal phenotype. Philos. Trans. R. Soc. Lond. B Biol. Sci. 351: 405-411, 1996.[Medline]
• LEWIN, G. R., MENDELL, L. M. Nerve growth factor and nociception. Trends Neurosci. 16: 353-359, 1993.[Medline]
• LEWIN, G. R., RITTER, A. M., MENDELL, L. M. On the role of nerve growth factor in the development of myelinated nociceptors. J. Neurosci. 12: 1896-1905, 1992.[Abstract]
• LEWIS, D. L., AIZPURUA, H.J.D., RAUSCH, D. M. Enhanced expression of Ca²⁺ channels by nerve growth factor and the v-src oncogene in rat phaeochromocytoma cells. J. Physiol. (Lond.) 465: 325-342, 1993.[Abstract/Free Full Text]
• LINDSAY, R. M. Role of neurotrophins and Trk receptors in the development and maintenance of sensory neurons: an overview. Philos. Trans. R. Soc. Lond. B Biol. Sci. 351: 365-373, 1996.[Medline]
• LIPSCOMBE, D., MADISON, D. V., POENIE, M., REUTER, H., TSIEN, R. Y., TSIEN, R. W. Spatial distribution of calcium channels and cytosolic calcium transients in growth cones and cell bodies of sympathetic neurones. Proc. Natl. Acad. Sci. USA 85: 2398-2402, 1988.[Abstract/Free Full Text]
• NJA, A., PURVES, D. The effects of nerve growth factor and its antiserum on the synapses in the superior cervical ganglion of the guinea pig. J. Physiol. (Lond.) 277: 55-75, 1978.[Abstract/Free Full Text]
• PERALTA, M. A., AYBAR, M. J., SANCHEZ, S. S. Alteration in morphogenic cell interactions by 1-beta-D-arabinofuranosylcytosine in Bufo arenarum embryo. Biocell. 19: 159-173, 1995.[Medline]
• PERSSON, J. K. E., LINDH, B., ELDE, R., ROBERTSON, B., RIVEROMELIAN, C., ERIKSSON, N. P., HOKFELT, T., ALDSKOGIUS, H. The expression of different cytochemical markers in normal and axotomized dorsal root ganglion cells projecting to the nucleus gracilis of the rat. Exp. Brain Res. 105: 331-344, 1995.[Medline]
• PETROV, T., SHAPIRO, J., DUFF, J. P., PENNEFATHER, P. S., GORDON, T., SMITH, P. A. Cell body calcium current depends on contact of axon with target. Can. J. Physiol. Pharmacol. 74: A27-A28, 1996.
• RITTER, A. M., MENDELL, L. M. Somal membrane properties of physiologically-identified sensory neuron in the rateffects of nerve growth factor. J. Neurophysiol. 68: 2033-2041, 1992.[Abstract/Free Full Text]
• ROGERS, M., HENDRY, I. Involvement of dihydropyridine-sensitive calcium channels in nerve growth factor dependent outgrowth by sympathetic neurons. J. Neurosci. Res. 26: 447-454, 1990.[Medline]
• ROSEN, L. B., GINTY, D. D., WEBER, M. J., GREENBERG, M. E. Membrane depolarization and calcium influx stimulate MEK and MAP kinase via activation of Ras. Neuron 12: 1207-1221, 1994.[Medline]
• SALA, F. Activation kinetics of calcium currents in bull-frog sympathetic neurones. J. Physiol. (Lond.) 437: 221-238, 1991.[Abstract/Free Full Text]
• SCOTT, B. S., EDWARDS, A. V. Electric membrane properties of adult mouse DRG neurons and the effect of culture duration. J. Neurobiol. 11: 291-301, 1980.[Medline]
• SELYANKO, A. A., SMITH, P. A., ZIDICHOUSKI, J. A. Effects of muscarine and adrenaline on neurones from Rana pipiens sympathetic ganglia. J. Physiol. (Lond.) 425: 471-500, 1990.[Abstract/Free Full Text]
• SHEN, K. F., CRAIN, S. M. Nerve growth factor rapidly prolongs the action potential of mature sensory neurons in culture and this effect requires activation of G_s-coupled kappa opioid receptors on these cells. J. Neurosci. 14: 5570-5579, 1994.[Abstract]
• SMITH, P. A. Amphibian sympathetic ganglia: an owner's and operator's manual. Prog. Neurobiol. 43: 439-510, 1994.[Medline]
• STREIT, J., LUX, H. D. Voltage dependent calcium currents in PC12 growth cones and cells during NGF-induced cell growth. Pflügers Arch. 408: 634-641, 1987.[Medline]
• STREIT, J., LUX, H. D. Calcium current inactivation during nerve-growth-factor-induced differentiation of PC12 cells. Pflügers Arch. 416: 368-374, 1990.[Medline]
• TRAYNOR, P., DRYDEN, W. F., SMITH, P. A. Trophic regulation of action potential in bullfrog sympathetic neurones. Can. J. Physiol. Pharmacol. 70: 826-834, 1992.[Medline]
• UNSICKER, K., REICHERT PREIBSH, H., SCHMIDT, R., PETTMANN, B., LABOURDETTE, G., SENSENBRENNER, M. Astroglial and fibroblast growth factors have neurotrophic functions for cultured peripheral and central nervous system neurons. Proc. Natl. Acad. Sci. USA 84: 5459-5463, 1987.[Abstract/Free Full Text]
• USOWICZ, M. M., PORZIG, H., BECKER, C., REUTER, H. Differential expression by nerve growth factor of two types of Ca²⁺ channels in rat phaeochromocytoma cell lines. J. Physiol. (Lond.) 426: 95-116, 1990.[Abstract/Free Full Text]
• VARADI, G., LORY, P., SCHULTZ, D., VARADI, M., SCHWARTZ, A. Acceleration of activation and inactivation by -subunit of skeletal muscle calcium channel. Nature 352: 159-162, 1991.[Medline]
• VERGE, V.M.K., GRATTO, K. A., KARCHEWSKI, L. A., RICHARDSON, P. M. Neurotrophins and nerve injury in the adult. Philos. Trans. R. Soc. Lond. B Biol. Sci. 351: 423-430, 1996.[Medline]
• VILLIÈRE, V., MCLACHLAN, E. M. Electrophysiological properties of neurons in intact rat dorsal root ganglion classified by conduction velocity and action potential duration. J. Neurophysiol. 76: 1924-1941, 1996.[Abstract/Free Full Text]
• WERZ, M. A., ELMSLIE, K. S., JONES, S. W. Phosphorylation enhances inactivation of N-type calcium channel current in bullfrog sympathetic neurons. Pflügers Arch. 424: 538-545, 1993.[Medline]
• WILDERING, W. C., LODDER, J. C., KITS, K. S., BULLOCH, A.G.M. Nerve growth factor (NGF) acutely enhances high-voltage activated calcium currents in molluscan neurons. J. Neurophysiol. 74: 2778-2781, 1995.[Abstract/Free Full Text]
• ZHANG, X., MEISTER, B., ELDE, R., VERGE, V.M.K., TERENIUS, L., HOKFELT, T. Large caliber primary afferent neurones projecting to the gracile nucleus express neuropeptide Y after sciatic nerve lesions. Eur. J. Neurosci. 5: 1510-1519, 1993.[Medline]

0022-3077/97 $5.00 Copyright ©1997 The American Physiological Society