The Journal of Neurophysiology Vol. 77 No. 1 January 1997, pp. 186-199
Copyright ©1997 by the American Physiological Society

Ca²⁺ Currents in Central Insect Neurons: Electrophysiological and Pharmacological Properties

Dieter Wicher and Heinz Penzlin

Sächsische Akademie der Wissenschaften zu Leipzig, Forschungsgruppe Neurohormonale Wirkungsmechanismen,D-07743 Jena, Germany

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Wicher, Dieter, and Heinz Penzlin. Ca²⁺ currents in central insect neurons: electrophysiological and pharmacological properties. J. Neurophysiol. 77: 186-199, 1997. Ca²⁺ currents in dorsal unpaired median (DUM) neurons isolated from the fifth abdominal ganglion of the cockroach Periplaneta americana were investigated with the whole cell patch-clamp technique. On the basis of kinetic and pharmacological properties, two different Ca²⁺ currents were separated in these cells: mid/low-voltage-activated (M-LVA) currents and high-voltage-activated (HVA) currents. M-LVA currents had an activation threshold of 50 mV and reached maximal peak values at 10 mV. They were sensitive to depolarized holding potentials and decayed very rapidly. The decay was largely Ca²⁺ dependent. M-LVA currents were effectively blocked by Cd²⁺ median inhibiting concentration (IC₅₀ = 9 µM), but they also had a remarkable sensitivity to Ni²⁺ (IC₅₀ = 19 µM). M-LVA currents were insensitive to vertebrate LVA channel blockers like flunarizine and amiloride. The currents were, however, potently blocked by -conotoxin MVIIC (1 µM) and -agatoxin IVA (50 nM). The blocking effects of -toxins developed fast (time constant  = 15 s) and were fully reversible after wash. HVA currents activated positive to 30 mV and showed maximal peak currents at +10 mV. They were resistant to depolarized holding potentials up to 50 mV and decayed in a less pronounced manner than M-LVA currents. HVA currents were potently blocked by Cd²⁺ (IC₅₀ = 5 µM) but less affected by Ni²⁺ (IC₅₀ = 40 µM). These currents were reduced by phenylalkylamines like verapamil (10 µM) and benzothiazepines like diltiazem (10 µM), but they were insensitive to dihydropyridines like nifedipine (10 µM) and BAY K 8644 (10 µM). Furthermore, HVA currents were sensitive to -conotoxin GVIA (1 µM). The toxin-induced reduction of currents appeared slowly ( ~ 120 s) and the recovery after wash was incomplete in most cases. The dihydropyridine insensitivity of the phenylalkylamine-sensitive HVA currents is a property the cockroach DUM cells share with other invertebrate neurons. Compared with Ca²⁺ currents in vertebrates, the DUM neuron currents differ considerably from the presently known types. Although there are some similarities concerning kinetics, the pharmacological profile of the cockroach Ca²⁺ currents especially is very different from profiles already described for vertebrate currents.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

In the vertebrate nervous system, at least six different types of voltage-dependent Ca²⁺ channels have been distinguished on the basis of their biophysical and pharmacological properties. According to the voltage range of activation, channels are divided into low-voltage-activated (LVA) channels (T type) (Nowycky et al. 1985) and mid-voltage-activated (MVA) channels (class E channels expressed in Xenopus oocytes) (Ellinor et al. 1993; Soong et al. 1993), taken together as mid/low-voltage-activated (M-LVA) channels on the one hand (Dunlap et al. 1995) and high-voltage-activated (HVA) channels (L, N, P, and Q type) (Fox et al. 1987; Llinas et al. 1992; Randall et al. 1993) on the other hand (for reviews see Miller and Fox 1990; Olivera et al. 1994; Tsien and Tsien 1990).

Ca²⁺ channels in insect neurons were found to have properties different from T-, L-, and N-type channels in vertebrate cells (Bickmeyer et al. 1994a; Byerly and Leung 1988; Pearson et al. 1993; Pelzer et al. 1989). In particular, long-lasting HVA currents (L-type-like) were blocked by phenylalkylamines (as in vertebrates), but not by dihydropyridines (unlike in vertebrates). -Conotoxin GVIA (-CgTx GVIA), an antagonist of N-type channels (McCleskey et al. 1987; Williams et al. 1992), was ineffective in various invertebrate neurons (Bickmeyer et al. 1994a; Bindokas and Adams 1989; McCleskey et al. 1987; Sun et al. 1987), but was recently shown to block a component of Ca²⁺ currents in dorsal unpaired median (DUM) neurons of the cockroach Periplaneta americana (Wicher and Penzlin 1994).

Both insect neurons (e.g., locust DUM cells) and vertebrate neurons possess Ca²⁺ channels sensitive to -agatoxin I and II (Bindokas and Adams 1989; Bindokas et al. 1991). Furthermore, -agatoxin IVA (-Aga IVA), which is a potent blocker of P-type currents (Mintz et al. 1992), was also found to act on Ca²⁺ currents in neurosecretory locust cells (Bickmeyer et al. 1994b). Although there are some similarities, generally differences between voltage-dependent Ca²⁺ currents in insect (or more generally, invertebrate) and vertebrate neurons are stressed in the literature.

Most somata of insect neurons are not capable of firing action potentials. One exception are DUM neurons in the ventral nerve cord. These neurons, e.g., in the terminal ganglion of the cockroach P. americana, are spontaneously active. Their somata exhibit large fast Na⁺ currents (Lapied et al. 1990). For a maintained repetitive activity, voltage-dependent Ca²⁺ currents were found to be necessary because the activity disappeared in Ca²⁺-free bath solution or in the presence of Ca²⁺ channel blockers like Ni²⁺ (Lapied et al. 1989). Shape of action potential as well as spike frequency can be changed by altering properties of voltage-dependent Ca²⁺ currents that act via Ca²⁺-dependent K⁺ currents (Wicher and Penzlin 1994; Wicher et al. 1994).

First voltage-clamp investigations revealed, in these neurons, the occurrence of at least two types of voltage-dependent Ca²⁺ currents (Wicher and Penzlin 1994). One is a current unaffected by depolarized holding potentials (V_hs) up to 50 mV. This current is sensitive to phenylalkylamines but insensitive to dihydropyridines. The other current is transient and inactivates at depolarized V_h.

In this study we investigated electrophysiological and pharmacological properties of voltage-dependent Ca²⁺ currents in DUM neurons isolated from the fifth abdominal ganglion of the cockroach P. americana. Two different kinds of currents were separated: M-LVA and HVA currents. These currents have different sensitivities to the well-established blockers specific for vertebrate Ca²⁺ channels, -CgTx GVIA (McCleskey et al. 1987), -Aga IVA (Mintz et al. 1992), and -conotoxin MVIIC (-CmTx MVIIC; Hillyard et al. 1992). The electrophysiological and pharmacological properties of the separated currents were compared with those of currents described in invertebrate and vertebrate preparations.

Some of the results have already been presented in abstract form (Wicher and Penzlin 1995).

	METHODS

Abstract Introduction Methods Results Discussion References

Cells

Isolation of cells was performed as described previously (Wicher et al. 1994). Briefly, the fifth and sixth abdominal ganglia of adult cockroaches (P. americana) were excised, desheathed, and incubated for 10 min at room temperature in saline composition, in mM: 190 NaCl, 5 KCl, 5 CaCl₂, 2 MgCl₂, and 10 N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid (HEPES), pH 7.4 containing 1 mg/ml trypsin. After the enzyme was thoroughly washed off and the ganglia were stored in saline for 1 h, the large DUM cells situated in the dorsal midline of the ganglia were separated with the use of thin metal needles. Viability of cells was assessed by microscopic observation: only those having a bright appearance under phase contrast were used.

The investigations started with the use of terminal ganglion DUM cells. In this ganglion 36 large cells containing octopamine are located in the dorsal midline (Dymond and Evans 1979; Eckert et al. 1992). In the same region eight cells of comparable size contain proctolin (Agricola et al. 1985). During the course of this work it became clear that there is a heterogeneity concerning the capability of spontaneous firing and the pattern of ionic currents (e.g., presence or absence of the fast Na⁺ current). For that reason further investigations were restricted to DUM neurons from the fifth abdominal ganglion. Here four octopaminergic DUM neurons form a cluster (Eckert et al. 1992) that can be easily identified and isolated. These neurons send bilateral projections into the segmental nerves (Gundel et al. 1996). Of these four neurons, three had large pear-shaped somata (40-60 µM diam) and were capable of repetitive activity. Under voltage-clamp conditions these three DUM neurons showed large fast Na⁺ currents. Results presented here were obtained from these cells.

Electrophysiology

The ion currents of the isolated neurons were measured at room temperature with the use of the patch-clamp method in the whole cell configuration (Hamill et al. 1981). Current measurements and data acquisition were performed with an EPC9 patch-clamp amplifier (HEKA Elektronic, Lambrecht, Germany) that was controlled by an ATARI computer (MEGA STE). Data were sampled at 10 kHz and filtered at 2.9 kHz. Capacitive and leak currents were compensated by a cancellation routine provided by E9SCREEN. Remaining uncompensated currents were subtracted with an on-line P/4 protocol. For off-line data analysis, M2Lab software (Instrutech, Elmont, NY) was used. Pipettes having resistances of 0.5-0.8 M were pulled from borosilicate capillaries (Hilgenberg, Malsfeld, Germany). The series resistance remaining after compensation did not exceed 1.3 M. Furthermore, to minimize voltage error due to series resistance, only cells with maximum peak Ca²⁺ currents <7.5 nA were used for analysis. V_h was 90 mV. The pipette solution contained (in mM) 50 choline chloride, 30 CsCl, 60 CsOH, 50 tetraethylammonium (TEA)-Br, 2 Mg-ATP, 1 CaCl₂, 10 ethylene glycol-bis(-aminoethyl ether)-N,N,N,N-tetraacetic acid (EGTA), and 10 HEPES. The bath solution for Ca²⁺ current measurements contained (in mM) 160 choline chloride, 30 TEA-Br, 5 CaCl₂, 10 HEPES, and 5*10⁴ tetrodotoxin. This combination of bath and pipette solutions allowed measurements of separated Ca²⁺ currents ~1 min after breaking into the cell. Within this time, contaminating Na⁺ and K⁺ currents disappeared completely.

Most experiments were performed with Ca²⁺ as charge carrier. To prevent Ca²⁺-induced rundown of Ca²⁺ currents, Sr²⁺ and Ba²⁺ were occasionally used as charge carriers instead of Ca²⁺. Because Ba²⁺, which is most commonly used for this purpose, had the disadvantage that its currents were larger than Sr²⁺ currents and increased the risk of voltage error due to series resistance, the concentration was reduced to 3 mM. When Ca²⁺ or Sr²⁺/Ba²⁺ were used, solutions had the following compositions. Pipette solution was composed of (in mM) 100 choline methyl sulphate, 60 CsOH, 10 CsCl, 30 TEA-Br, 2 Mg-ATP, 1 CaCl₂, 10 EGTA, and 10 HEPES, pH adjusted to 7.2. Bath solution was composed of (in mM) 190 choline methyl sulphate, 5 CaCl₂ or 5 SrCl₂/3 BaCl₂, 10 HEPES, and 5*10⁴ tetrodotoxin, pH adjusted to 7.4. A comparison of Ca²⁺ currents measured in the two different Ca²⁺ solutions gave no indication of a contribution of Cl currents in the Cl-rich solution. Liquid junction potentials between pipette and bath solutions were corrected. Tetrodotoxin was obtained from Sigma (Deisenhofen, Germany), -CgTx GVIA from RBI (Natick, MA), and -CmTx MVIIC and -Aga IVA from Alomone Labs (Jerusalem, Israel).

Application or washout of blocking agents was performed by transferring the cell (situated on the pipette tip, which had been inserted into a protecting glass tube) into the blocker-containing or control solution, respectively. A total and fast solution change was then achieved by sucking a small amount of solution into the tube.

Tail current measurements were used to assess the amount of current activation. The tail currents were evoked by repolarization to V_h. Tails had a fast activation and deactivation kinetics. Depending on the size of the depolarizing voltage step, the maximum was attained within 100-900 µs. The maximum of tails was used as measure of current activation and is referred to as "tail" in the text.

Results were given as means ± SD; n = number of cells. The evaluation of statistical significance of differences was performed with the use of Student's t-test (error probability P). For data analysis, including nonlinear fitting procedures, the software Prism 2 (Graph Pad Software, San Diego, CA) was used.

	RESULTS

Abstract Introduction Methods Results Discussion References

Kinetic properties

Whole cell recordings of membrane current in DUM neurons from the fifth and the sixth abdominal ganglia of the cockroach P. americana revealed prominent voltage-dependent Ca²⁺ currents. Figure 1A shows a family of Ca²⁺ currents obtained by depolarizing command pulses from a V_h of 90 mV. The threshold for activation was about 50 mV. The currents showed some inactivation, most pronounced on stepping to about 10 mV. Peak currents were maximal at 0 mV, whereas the current-voltage (I-V) curve of currents measured at the end of 50-ms command pulses, i.e., after some decay had been taken place, had the maximum a few millivolts more positive (Fig. 1B). The low threshold for activation indicates the presence of LVA currents, whereas the voltage range in which currents were maximal is rather typical for HVA currents (e.g., Miller and Fox 1990).

View larger version (19K): [in this window] [in a new window]   FIG. 1. A: family of Ca2+ currents from the soma of a dorsal unpaired median (DUM) neuron obtained by voltage jumps from a holding potential (Vh) of 90 mV to potentials ranging from 40 mV to +60 mV in 10-mV steps (terminal ganglion of the cockroach P. americana, membrane capacitance 330 pF). Note the large tail currents. B: comparison of mean current-voltage (I-V) relationships of Ca2+ currents from somata of DUM neurons isolated from the 5th abdominal ganglion. Currents were obtained by depolarizing command pulses 50 ms in duration from Vh = 90 mV. Filled squares: means of peak currents. Open squares: means of currents at the end of the pulse (n = 15 cells). Bars: SD. C: I-V relationships of peak Ca2+/Sr2+ currents. Currents were normalized to the maxima of currents measured with Sr2+ (5 mM). Vh = 90 mV. Data points are means of 5-7 cells from the 5th and 6th abdominal ganglion. Bars: SD. Inset: comparison of a Ca2+ current with a Sr2+ current; currents were obtained by command pulses to 10 mV.

The mean of maximal Ca²⁺ peak currents was 6.7 ± 0.9 (SD) nA (n = 15); the mean cell capacitance was 330 ± 82 pF (n = 26). Assuming a specific membrane capacitance of 1 µF/cm², the maximal current density was 0.2 pA/µm².

Charge carriers

The dependence of current size, voltage dependence, and kinetics on the charge carrier was investigated by substituting 5 mM Sr²⁺ or 3 mM Ba²⁺ for 5 mM Ca²⁺. In the case of Ba²⁺, an equimolar substitution was avoided after some experiments with 5 mM Ba²⁺ because these currents were too large to be adequately clamped. Generally, currents carried by Sr²⁺ or Ba²⁺ were much larger in size than Ca²⁺ currents and showed less decay (Sr²⁺: Fig. 1C, inset; Ba²⁺: compare Fig. 3, D and E). The I-V curve for Sr²⁺ peak currents reaches maximum at 10 mV, compared with 0 mV for Ca²⁺ (Fig. 1C), but there is no shift of the whole I-V curve on the voltage axis. Furthermore, with Sr²⁺ solution, the decay of currents activated by low-voltage commands starts later and is less pronounced than with Ca²⁺ (Fig. 1C, inset). Therefore the difference between the maxima of Ca²⁺ and Sr²⁺ I-V curves seems to reflect the influence of time-dependent inactivation on current size. In contrast, the I-V curve for peak of Ba²⁺ currents is shifted by ~10 mV toward lower voltages (not shown), which might be attributed to the lower concentration of divalent cations and the lower potency of Ba²⁺ in screening negative surface charges (Frankenhäuser and Hodgkin 1957; Hille 1992).

View larger version (22K): [in this window] [in a new window]   FIG. 3. Time-dependent inactivation of Ca2+ currents. A: amplitude of tail Ca2+ currents as a function of pulse duration for 3 potentials. Inactivation is most pronouced at 10 mV. Extracellular Ca2+ concentration: 5 mM; Vh = 90 mV. B: inactivation of Ba2+ tail currents is much smaller than that of Ca2+ currents (same protocol as in A). Extracellular Ba2+ concentration: 3 mM; Vh = 90 mV. C: amplitude of Ca2+ tail currents as a function of the potential of 50-ms command pulses. The tail currents are normalized to those obtained with a command pulse 4 ms in duration. The currents are maximally reduced at 20 to 10 mV (same cell as in A). D and E: superimposed registrations of Ca2+ (D) and Ba2+ (E) currents evoked by jumps 7 and 50 ms in duration to 10 mV. Note the small amount of decay in Ba2+ currents.

The rundown of currents observed in Ca²⁺ solution was substantially decreased in Sr²⁺ and Ba²⁺ solution. This indicates that channel rundown can be partly, although not entirely, accounted for by the influx of Ca²⁺ generated by the repeated activation of Ca²⁺ currents during measurements.

Activation

Ca²⁺ current activation was estimated from the amplitudes of tail currents elicited on repolarization from short command pulses to V_h. To achieve full activation, a pulse 5 ms in duration was chosen, which was close to the time to peak but short enough to avoid substantial inactivation. Activation depended on voltage in sigmoidal fashion (Fig. 2A1). The data could be fitted equally well by assuming a model with m or m² activation kinetics, a situation similar to those previously reported for Helix neurons (Akaike et al. 1978). The curve in Fig. 2A1 is described by a Boltzmann equation where the potential of half-maximal activation V_0.5 = 18 mV and the slope factor S = 8.3 mV (n = 7).

View larger version (22K): [in this window] [in a new window]   FIG. 2. Kinetic properties of Ca2+ current activation (A) and inactivation (B). A1: activation determined from normalized tail currents evoked on repolarization to 90 mV from the various depolarized potentials applied for 5 ms. Data points are means. Bars: SD. n = 7. The curves were fitted to the measured data according to I/Imax = 1/{1 + exp(V  V0.5)/S} with V0.5 = 13 mV and S = 8.6 mV. A2: time to peak for Ca2+ () and Ba2+ () currents. Data points are means. Bars: SD. n = 8. B1 and B2: steady-state inactivation of Ca2+ (B1) and Ba2+ (B2) currents. The peak currents were measured for test pulses to 0 mV after prepulses 5 s in duration to different prepulses as indicated. They were normalized to currents obtained with prepotentials of 120 mV. Data points are means. Bars: SD. n = 8. Currents were fitted by a sum of 2 Boltzmann equations Ii/Ii,max = 1/{1 + exp(Vi,0.5  V)/Si}, i = 1, 2, indicating the existence of 2 inactivation processes with different voltage dependence. The parameters determined for Ca2+ as charge carrier (B1) were I1,max = 0.31*Imax, V1,0.5 = 96 mV, S1 = 14 mV, and I2,max = 0.69*Imax, V2,0.5 = 38 mV, S2 = 10 mV. For Ba2+ (B2) currents the results were I1,max = 0.37*Imax, V1,0.5 = 94 mV, S1 = 15 mV, and I2,max = 0.63*Imax, V2,0.5 = 38 mV, S2 = 10 mV. Note the similarity of the figures for the corresponding parameters. Inset: a fast decaying current component inactivates already at negative potentials. For a Vh of 50 mV the transient component evoked on jumping to 10 mV is reduced by >50%, whereas the slowly decaying current is hardly affected. B3: time of half-maximal decay of Ca2+ and Ba2+ currents. This time is the span between tp and t [I(t) = 0.5*Ip]. Data points are means. Bars: SD. n = 7. For both ions, the fastest time-dependent inactivation occurs in the low-voltage range.

During longer command pulses Ca²⁺ currents show some decay. Activation curves obtained under these conditions are shifted on the voltage axis toward more positive potentials (V_0.5 = 8 mV, S = 8.5 for 50 ms, not shown). Thus two current components can be distinguished according to their activation: an early, transient component activating in the low- or mid-voltage range (M-LVA) and a late, sustained component activating in the high-voltage range (HVA). Because the voltage difference between the figures for V_0.5 is only 10 mV, the denotation "M-LVA" and "HVA" currents has a more operational meaning.

The existence of different Ca²⁺ currents in DUM neurons can also be discerned in the I-V relationships for the peak and the late current (Fig. 1B) and in the time to peak curve, which shows a hump with a local maximum at 15 mV for Ca²⁺ currents and 25 mV for Ba²⁺ currents (Fig. 2A2). The much longer time to peak of Ba²⁺ currents in the low-voltage range is simply attributed to the late start and the low degree of current decay in Ba²⁺ solution, i.e., with Ca²⁺ there is some overlap between current activation and inactivation.

Deactivation

The deactivation process of currents activated by short pulses (5 ms) occurring on repolarization to the V_h of 90 mV can be described with a single-exponential function. The time constant of the deactivating tail currents is largely independent of the command potential and amounts to 977 ± 38 µs (n = 10). In contrast, with longer depolarizations, the description of tail current deactivation requires two-exponential functions when the preceeding activating pulses were positive to 20 mV. The additional occurring deactivating process has a time constant of 724 ± 31 µs (n = 10). Especially for long pulses (50 ms), the latter process is dominating for activations 0 mV.

These results support the assumption that in DUM cells at least two Ca²⁺ currents exist that differ in their activation (cf. above: M-LVA and HVA currents).

Inactivation

STEADY-STATE INACTIVATION. Depolarizing prepulses led to a reduction of Ca²⁺ currents that depended on the duration of the conditioning pulse. Prepulses of 100 ms reduced the maximal peak currents to ~40%, whereas 5-s prepulses led to complete inactivation. The voltage dependence of steady-state inactivation is shown in Fig. 2B1. Weak depolarizations already reduced considerably a fast inactivating component (decay time constant  ~ 6 ms, Fig. 2B, inset), whereas a slow inactivating component ( ~ 250 ms) was relatively resistant to predepolarizations. The voltage dependence of steady-state inactivation could be fitted by a sum of two Boltzmann equations. Because these processes seem to show some overlap on the voltage axis, a depolarized V_h does not lead to a clear separation of one current component. The remaining as well as the inactivated currents are always a superposition of components. With Ba²⁺ as charge carrier, the voltage dependence was very similar (Fig. 2B2), indicating that these inactivation processes are voltage dependent and not dependent on Ca²⁺ influx.

Time-dependent inactivation

TIME COURSE OF CURRENT DECAY. The time course of the decay of Ca²⁺ currents cannot be described by a single-exponential function. Especially in the voltage range where the decay is most pronounced (about 10 mV), the time course of decay was fitted reasonably well only by a sum of three exponential functions A_k exp(t/_k) with different decay time constants _k. At 10 mV, such a fitting procedure gave the following results: A₁ = 2.7 ± 0.6 nA, ₁ = 6.3 ± 2.5 ms; A₂ = 3.3 ± 0.7 nA, ₂ = 90 ± 21 ms; A₃ = 2.9 ± 0.5 nA, ₃ = 250 ± 45 ms; n = 6. The inactivation of currents carried by Sr²⁺ or Ba²⁺ could be fitted with the use of only two time constants. The fast time constant amounted to 20-40 ms (Sr²⁺) and 20-50 ms (Ba²⁺) and the slow time constant was ~250 ms for both ions. As with Ca²⁺ currents, the faster decaying components of Sr²⁺ or Ba²⁺ currents dominated in the low-voltage range.

Figure 2B3 shows the voltage dependence of Ca²⁺ and Ba²⁺ current decay described by the time to half-maximal decay. As outlined above, the fastest decay occurred in the low-voltage range. Although most pronounced with Ca²⁺, this is also seen with Ba²⁺. Furthermore, in the whole voltage range, Ba²⁺ currents are much more slowly decaying than Ca²⁺ currents.

TIME-DEPENDENT INACTIVATION (DECAY) IS LARGELY Ca²⁺ DEPENDENT. During a command voltage step, Ca²⁺ currents showed some decay. The largest and fastest decay was observed in the voltage range between 20 and 0 mV. The decay was slower at higher voltages and the extent decreased progressively with increasing depolarization. Both effects are illustrated with command potentials to 10, +10, and +50 mV in Fig. 3, A and C. When Ca²⁺ was substituted for by Ba²⁺, the time-dependent decay was strongly reduced (Fig. 3, B and E). This reduction was dramatic in the low-voltage range (compare Fig. 3, D and E). Thus the main part of the observed Ca²⁺ current decay is obviously Ca²⁺ dependent.

On the other hand, there are also differences in the decay between currents carried by Ba²⁺ and Sr²⁺. In the low-voltage range the decay of Sr²⁺ currents is more pronounced than that of Ba²⁺ currents. At 20 mV, Ba²⁺ currents decayed after 50 ms to 79 ± 4% (n = 8) of the peak current; the corresponding Sr²⁺ currents (10 mV) decayed to68 ± 9% (n = 9, for comparison: Ca²⁺: 45 ± 10%,n = 15). At higher voltages the differences disappeared. At 0 mV, Ba²⁺ currents decayed to 82 ± 5% and Sr²⁺ currents (0 mV) to 81 ± 8% (Ca²⁺: 66 ± 10%). The differences in the amount of current inactivation between the charge carriers Ba²⁺ and Sr²⁺ at lower voltages may thus be attributed to differences in the permeability to these ions of channels activated in this potential range.

Pharmacological properties

A previous study (Wicher and Penzlin 1994) indicated the occurrence of at least two pharmacologically separable Ca²⁺ current components. These investigations were extended to further characterize the components and to compare their properties with those of vertebrate calcium currents.

INORGANIC IONS. Ca²⁺ currents in DUM neurons were sensitive to the commonly used inorganic blockers Cd²⁺, Ni²⁺, and Gd³⁺. In vertebrate neurons, LVA, MVA, and HVA currents are found to differ in their sensitivity to Ni²⁺ and Cd²⁺. HVA and MVA currents are potently blocked by Cd²⁺ (Fox et al. 1987; Kasai and Neher 1992; Sather et al. 1993; Soong et al. 1993; Zhang et al. 1993), whereas LVA T-type currents are less sensitive to Cd²⁺ (Fox et al. 1987; Miller and Fox 1990). On the other hand, Ni²⁺ blocks LVA and MVA currents better than HVA currents (Fox et al. 1987; Sather et al. 1993; Soong et al. 1993; Zhang et al. 1993). To test whether these aspects of the vertebrate classification scheme M-LVA/HVA (Dunlap et al. 1995) are also applicable for Ca²⁺ currents in DUM neurons, the effects of Ni²⁺ and Cd²⁺ in these neurons were investigated. At all potentials tested Cd²⁺ blocked more effectively than Ni²⁺. But the sensitivity to Cd²⁺ and Ni²⁺ was voltage dependent. This is shown in Fig. 4 for two representative potentials: 10 mV (M-LVA currents dominating) and +20 mV (HVA currents dominating). Current recordings demonstrating the effects of 1 µM Cd²⁺ and Ni²⁺ are presented in Fig. 4, A2 and B2. The dose-inhibition curves were well fitted by the equation I/I_max = 1/{1+(ion/IC₅₀)S}. In the case of HVA currents, the Hill slope coefficient S was ~1 (Cd²⁺: S = 1.25 ± 0.2; Ni²⁺: S = 1.1 ± 0.1), indicating 1-1 binding. For M-LVA currents, however, S was always <1, indicating multiple binding (Cd²⁺: S = 0.63 ± 0.1; Ni²⁺: S = 0.49 ± 0.05). For HVA currents, half-maximal inhibition was obtained with Cd²⁺ at a half-inactivating concentration IC₅₀ of 5.0 ± 0.8 µM, which is in the range typical for HVA calcium currents in vertebrates (e.g., Fox et al. 1987; Kasai and Neher 1992; Sather et al. 1993). For Ni²⁺, the DUM cell HVA Ca²⁺ channels have a somewhat higher affinity(IC₅₀ = 40 ± 5 µM) than vertebrate HVA channels, e.g., 100 µM Ni²⁺ blocked only 10% of the current in chick sensory neurons (Fox et al. 1987), and in neuroblastoma-glioma cells the IC₅₀ was ~250 µM for Ni²⁺ (Kasai and Neher 1992). In snail neurons, however, the IC₅₀ for Ni²⁺ was 30 µM (Akaike et al. 1978), but in these cells Cd²⁺ was largely ineffective (IC₅₀ = 3 mM!).

View larger version (20K): [in this window] [in a new window]   FIG. 4. Reduction of Ca2+ currents by Cd2+ (A) and Ni2+ (B). A1 and B1: dose-response relationships of Cd2+ block (A1) and Ni2+ block (B1) of peak currents obtained at 2 potentials (10 mV, +20 mV). The curves were fitted according to the equation I/Imax = 1/{1+(ion/IC50)S}, where Ion is Cd2+ or Ni2+. The results for the parameters were as follows. Cd2+: 10 mV, median inhibiting concentration IC50 = 9 µM, S = 0.63; +20 mV, IC50 = 5 µM, S = 0.4. Ni2+: 10 mV, IC50 = 19 nM, S = 0.49; +20 mV, IC50 = 40 nM, S = 1.1. Data points are means. Bars: SD. n = 6. A2 and B2: current registrations in the presence and absence of 1 µM Cd2+ (A2) and 1 µM Ni2+ (B2). Currents were obtained by depolarizations to the potentials indicated.

As is typical for vertebrate LVA and MVA currents, the affinity of DUM M-LVA currents for Ni²⁺ was higher than in HVA currents: IC₅₀ = 19 ± 3 µM. On the other hand, there was also a strong blocking action of Cd²⁺ that compares with that for HVA currents: IC₅₀ = 9 ± 2 µM. Similar dose-response properties were recently found for rbE-II channels expressed in oocytes (Soong et al. 1993). In this system the IC₅₀ for Ni²⁺ amounted to 25 µM, and 10 µM Cd²⁺ blocked >80% of the current. Similar to DUM cell M-LVA currents, rbE-II currents are transient, inactivate at depolarized potentials, and are maximal at depolarizations to 10 mV. Other examples for a Ni²⁺/Cd²⁺ sensitivity similar to DUM cell M-LVA currents are rat cerebellar granule neurons (Ni²⁺: IC₅₀ = 66 µM; Cd²⁺: IC₅₀ = 1 µM) and doe-1 channels expressed in oocytes (Ni²⁺: IC₅₀ = 33 µM; Cd²⁺: IC₅₀ ~ 1 µM) (Ellinor et al. 1993).

The effects of Ni²⁺ and Cd²⁺ on Ca²⁺ currents were completely reversible up to concentrations of 0.1 mM. 85 ± 3% of the currents blocked by 0.1 mM Cd²⁺ and 94 ± 5% of the currents blocked by 0.1 mM Ni²⁺ recovered within 1 min after wash (n = 5).

In some experiments the effect of 1 µM Gd³⁺ on calcium currents in DUM neurons was tested. The inhibiting effect was roughly the same as that of 1 µM Cd²⁺ (not shown). There was not such a strong block as reported for neuroblastoma-glioma cells (Kasai and Neher 1992), but the action was similar insofar as the inhibition took ~1 min to develop.

Organic compounds

HVA L-type currents in vertebrates are sensitive to dihydropyridines, phenylalkylamines, and benzothiazepines (e.g., Catterall and Striessnig 1992). Ca²⁺ currents in invertebrate neurons were shown to be sensitive to phenylalkylamines like verapamil and methoxyverapamil (D600) (Akaike et al. 1978; Pearson et al. 1993; Pelzer et al. 1989; Schäfer et al. 1994; Thomas 1984; Wicher and Penzlin 1994), but, with the exception of bee Kenyon cells (Schäfer et al. 1994), they were never found to be sensitive to dihydropyridines (Bickmeyer et al. 1994a; Byerly and Leung 1988; Pearson et al. 1993; Pelzer et al. 1989; Thomas 1984; Wicher and Penzlin 1994). The benzothiazepine diltiazem in Drosophila neurons had no influence on Ca²⁺ currents that were slightly reduced by phenylalkylamines at high concentrations (Byerly and Leung 1988).

The reduction of peak currents of cockroach DUM cell Ca²⁺ currents after application of 10 µM verapamil or diltiazem is shown in Fig. 5. Both agents blocked an HVA current component activating at potentials more positive than 30 mV and reaching its maximum between 0 and +10 mV. The difference currents (control registration  registration with blocker; not shown) exhibit substantial decay ( ~ 90 ms) in the voltage range between 20 and 0 mV. At more positive potentials the decay becomes slower and less pronounced. The blocker-sensitive currents have large tail currents that deactivate fast (verapamil:  = 470 ± 120 µs, diltiazem:  = 490 ± 130 µs, n = 6). Because voltage dependence and kinetics of the verapamil-sensitive and the diltiazem-sensitive currents are very similar, it seems reasonable to infer that both blockers inhibit the same current. The effects of verapamil and diltiazem were largely reversible. One minute after wash, 75 ± 5% of the verapamil-blocked currents and 80 ± 7% of the diltiazem-blocked currents recovered.

View larger version (25K): [in this window] [in a new window]   FIG. 5. I-V relationships of the reduction of peak Ca2+ currents by 10 µM verapamil (A) and 10 µM diltiazem (B). Filled squares: means of control measurements. Open squares: currents measured 1 min after blocker application. Circles: peak of the blocker-sensitive currents (difference currents); n = 8 (A), n = 5 (B). Both blockers reduced the currents reversibly. The block disappeared within 1-2 min of washing. Insets: registrations of currents evoked by pulses to 0 mV in the presence and absence of 10 µM verapamil (A) and 10 µM diltiazem (B).

The third class of compounds known to block vertebrate L-type currents, the dihydropyridines, had no effect on Ca²⁺ currents in DUM neurons. Neither the antagonist nifedipine (10 µM) nor the agonist BAY K 8644 (10 µM) affected any current, which is in line with the results of previous investigations (Wicher and Penzlin 1994).

Although there were no indications for the occurrance of LVA currents similar to vertebrate T-type currents in DUM neurons, the effect of 10 µM flunarizine was tested. In all experiments (n = 10) flunarizine had no effect. Also, amiloride (10-100 µM) did not reduce any Ca²⁺ current in DUM neurons (H. Achenbach, personal communication). Thus the M-LVA currents found in DUM cells have no similarity with T-type currents.

Peptide toxins

In vertebrate neurons the toxins -CgTx GVIA, -CmTx MVIIC, and -Aga IVA block more or less specificallyN-, Q-, and P-type channels, respectively (Dunlap et al. 1995; Miljanich and Ramachandran 1995; Olivera at al. 1994). In the case of -CgTx GVIA, a blocking effect on Ca²⁺ currents in DUM neurons has already been demonstrated (Wicher and Penzlin 1994). It seemed interesting to find out whether the -toxins are suitable for a separation of Ca²⁺ current components in these cells.

The toxins were applied to the DUM cells in concentrations that lead to 80-90% block of vertebrate N-, P-, and Q-type currents: -CgTx GVIA, 1 µM; -Aga IVA, 50 nM; -CmTx MVIIC, 1 µM. In Fig. 6, A-C, the effects of the three -toxins on peak I-V curves of Ca²⁺ currents are shown. Some characteristic differences in the action on currents between these toxins can be easily discerned. -CgTx GVIA was nearly ineffective in the low-voltage range, but it reduced HVA currents (Fig. 6A). The I-V curve of the -CgTx-GVIA-sensitive currents has a threshold of 30 mV and reaches the maximum at 0 mV. In contrast, -CmTx MVIIC blocked M-LVA currents most potently but only weakly affected HVA currents (Fig. 6C). The -CmTx-MVIIC-sensitive current component activates positive to 50 mV and is maximal at 10 mV. The effect of -Aga IVA was similar to that of -CmTx MVIIC but less pronounced (Fig. 6B).

View larger version (37K): [in this window] [in a new window]   FIG. 6. Effects of -toxins on Ca2+ (A-D) and Ba2+ (E) currents. A-C: I-V relationships of the reduction of peak Ca2+ currents by 1 µM -conotoxin GVIA (-CgTx GVIA, A), 50 nM -agatoxin IVA (-Aga IVA, B), and 1 µM -conotoxin MVIIC (-CmTx MVIIC, C). Data points are means of n = 8 (A), 6 (B), and 9 (C) cells before (filled symbols) and 2 min after application (open symbols) of the peptide toxins. Circles: peak of the toxin-sensitive currents (difference currents). Bars: SD. D and E: current registrations of the effects of -CgTx GVIA and -CmTx MVIIC on Ca2+ (D) and Ba2+ (E) currents in the low-voltage range (D1, D2, E1, and E2) and in the high-voltage range (D3, D4, E3, and E4). The effect of -Aga IVA is not shown because it is similar to -CmTx MVIIC. The calibration for each column is given at the top. Toxin concentrations: 1 µM -CgTx GVIA, 50 nM -Aga IVA, and 1 µM -CmTx MVIIC. The fast decay of low-voltage-activated Ca2+ currents (D1 and D2) was strongly reduced when Sr2+ (not shown) or Ba2+ carried the current (E1 and E2). In this potential range -CmTx MVIIC was the most effective blocker and -CgTx GVIA had weak effects. High-voltage-activated currents, however, were more effectively blocked by -CgTx GVIA than by -CmTx MVIIC.

Furthermore, the toxin-sensitive M-LVA and HVA currents differ in their kinetics. Ca²⁺ currents blocked by -CgTx GVIA were hardly decaying (Fig. 6, D1 and D3) and had pronounced tails with fast deactivation ( = 510 ± 110 µs, n = 7). On the other hand, currents blocked by -CmTx MVIIC (and -Aga IVA) show fast decay (Fig. 6D2). When Ca²⁺ was substituted for by Sr²⁺ or Ba²⁺, the amount of toxin-induced block in the lower-voltage range was increased, whereas the block in the high-voltage range remained of comparable size as with Ca²⁺ (Table 1). Sr²⁺ and Ba²⁺ currents obtained in the lower-voltage range are slower and less decaying than Ca²⁺ currents (Fig. 3). Thus -CmTx MVIIC and -Aga IVA seem to block the predepolarization-sensitive current component that is transient because of Ca²⁺-dependent inactivation (compare Fig. 6, D2 and E2). At higher voltages the decay of -toxin-sensitive currents is less pronounced and largely independent of the charge carrier (Fig. 6, D3-E4).

View this table: [in this window] [in a new window]   TABLE 1. Blocking effect of -toxins on currents carried by Ca2+, Sr2+, and Ba2+

The toxin-induced current block developed with different time course for -CmTx MVIIC and -Aga IVA on the one hand and -CgTx GVIA on the other (Fig. 7A). The blocking effects in the lower-voltage range appeared quickly with a time constant of ~15 s. On wash, the currents recovered quickly with a similar time constant (not shown). Especially in the case of -Aga IVA, no strong depolarizing voltage pulses were necessary to get the relief from block as described for rat Purkinje cells (Mintz et al. 1992). On the other hand, in the presence of -Aga IVA such pulses failed to remove the block, which is also in contrast to the results obtained for the rat neurons. The -CgTx-GVIA-induced current block in the HVA range developed slowly (time constant > 100 s). The recovery of currents after the toxin was washed off appeared slow also and was incomplete after 2 min in most cases.

View larger version (24K): [in this window] [in a new window]   FIG. 7. A: time course of the effect of -CgTx GVIA and -CmTx MVIIC on peak Ba2+ currents. The current reduction on application of 1 µM -CmTx MVIIC in the low-voltage range (10 mV, ) appeared fast. The time course of the toxin-induced block could be fitted well with a single-exponential function (time constant  = 15 s). In contrast, the reducing effect of 1 µM -CgTx GVIA in the high-voltage range (+20 mV, ) developed considerably more slowly. The exponential function fitted on the data points is described by a time constant of 117 s. B and C: specificity of both the toxins -CgTx GVIA and -CmTx MVIIC is limited. Fast measurements of instantaneous Ba2+ I-V relationships were performed with voltage ramps (100 mV, 100 ms). B: 1 µM -CgTx GVIA caused a fast reduction of Ba2+ current in the low-voltage range (10 s) that was followed by a slower decrease of high-voltage currents (45 s, 2 min). C: application of 1 µM -CmTx MVIIC resulted in a fast block of low-voltage-activated currents (10 s) but within 2 min there was an additional slight reduction of the high-voltage-activated component.

The above results, summarized in Table 1, point to a relative specificity of -CmTx MVIIC for the rapidly decaying M-LVA Ca²⁺ current component on the one hand and to a fairly selective block of the slowly decaying HVA component by -CgTx GVIA on the other. But an analysis of the time course and the voltage dependence of peptide toxin effects revealed that both the toxins do not act exclusively selectively. As shown in Fig. 7B for Ba²⁺ currents that were obtained by voltage ramps (allowing fast, time-resolved measurements), 1 µM -CgTx GVIA also affected M-LVA currents slightly within 10 s after application. After some delay a pronounced reduction of HVA currents developed. Furthermore, with -CgTx GVIA at higher concentrations, M-LVA currents were also substantially reduced. The total current at 10 mV was reduced to 64 ± 6% of control when -CgTx GVIA was applied at 5 µM.

On the other hand, 1 µM -CmTx MVIIC blocked M-LVA currents very significantly within 10 s, but in the following 2 min it caused an additional slight reduction of HVA currents (Fig. 7B).

From the above evaluation of the steady-state inactivation of the current components it is already known that a transient M-LVA Ca²⁺ current component is inactivated by depolarized V_hs, whereas the HVA component is not. With the assumption that -CmTx MVIIC and -Aga IVA might act on the depolarization-sensitive component, one would expect that the effects of both blockers depend on V_h, whereas the effect of -CgTx GVIA is largely independent of V_h. Indeed, when V_h was set to 50 mV and the fast decaying component had vanished, -CmTx MVIIC as well as -Aga IVA had no or a very little effect (n = 5 experiments, Fig. 8, A2, A3, B2, and B3). To test the pure -CgTx GVIA effect on HVA currents, a putative action on M-LVA currents was excluded by the presence of 0.5 µM -CmTx MVIIC in the control and test solution (see above). Under these conditions, the blocking effect of -CgTx GVIA was, as expected, not affected by a depolarized V_h of 50 mV (Fig. 8, A1 and B1).

View larger version (18K): [in this window] [in a new window]   FIG. 8. Ca2+ currents affected by the peptide toxins -CgTx GVIA, -Aga IVA, and -CmTx MVIIC differ in sensitivity to a depolarized Vh. A1-A3: effects of the peptide toxins on currents evoked by pulses from 90 to 10 mV. B1-B3: toxin effects on currents evoked by pulses from a depolarized potential (50 mV) to 10 mV. To prevent -CgTx GVIA from acting on the low-voltage-activated component (see Fig. 7), 0.5 µM -CmTx MVIIC was added to the control solution. Note that the action of -CgTx GVIA is unaffected by predepolarization, whereas the block of -Aga IVA was strongly reduced at Vh = 50 mV and that of -CmTx MVIIC was completely eliminated. Toxin concentrations: 1 µM -CgTx GVIA, 50 nM -Aga IVA, and 0.5 µM -CmTx MVIIC.

Although -CmTx MVIIC and -Aga IVA on the one side and -CgTx GVIA on the other seem to act preferentially on different channel types, their specificity was obviously limited. Therefore it was tested whether there is some occlusion between the toxins. When 0.5 µM -CmTx MVIIC was applied, it affected a transient LVA Ca²⁺ current component (20 mV, Fig. 9A1) and was nearly ineffective in the high-voltage range (+20 mV, Fig. 9A2). In the presence of -CmTx MVIIC, -CgTx GVIA reduced very weakly a sustained current at 20 mV, but at +20 mV it blocked 25 ± 6% of the total current (Fig. 9A2, n = 4 experiments), which does not differ significantly from the blocking efficacy without -CmTx MVIIC (30 ± 6%). Thus -CmTx MVIIC, at a concentration of 0.5 µM, does not seem to interfere with -CgTx GVIA in the high-voltage range.

View larger version (28K): [in this window] [in a new window]   FIG. 9. Effects of combinations of -toxins on Ca2+ (A) and Ba2+ (B) currents in the low-voltage range (pulse to 20 mV) and in the high-voltage range (+20 mV). A1: at 20 mV, -CmTx MVIIC (0.5 µM) had a pronounced blocking effect on transient currents, whereas in the presence of -CmTx MVIIC, -CgTx GVIA (1 µM) reduced a sustained current only slightly. A2: at +20 mV, 0.5 µM -CmTx MVIIC was largely ineffective and did not prevent -CgTx GVIA (1 µM) from blocking a part of the high-voltage-activated current. B1: pretreatment with -Aga IVA (50 nM) leads to a weaker response to -CmTx MVIIC (1 µM). The total block was less than expected for agents acting on different targets. Note the similarity of currents sensitive to the toxins. B2: effect of 1 µM -CgTx GVIA on high-voltage-activated currents seems to be independent of the presence of -Aga IVA (50 nM). Note that the -Aga-IVA-sensitive current shows a faster onset of activation than the -CgTx-GVIA-sensitive component.

According to the results given in Figs. 6 and 8, -CmTx MVIIC and -Aga IVA are most likely affecting the same M-LVA current component. When 1 µM -CmTx MVIIC was applied in the presence of 50 nM -Aga IVA, the reduction of Ba²⁺ currents in the low-voltage range was 42 ± 6% (n = 4), which was significantly weaker (P < 0.01) than the 60% block that was observed without -Aga IVA (Fig. 9B1). The additional effect of both toxins was in all cases tested less than the sum of the effect of each toxin. These facts indicate that both toxins might act on the same channel type.

In the above described experiments with depolarized V_h it was observed that -Aga IVA not only blocked a depolarization-sensitive M-LVA component but also a depolarization-insensitive HVA component. Therefore it had to be clarified whether -Aga IVA might reduce the effect of -CgTx GVIA in the high-voltage range. In the experiments addressing this question there was no indication for such an occlusion. Although -Aga IVA blocked a part of HVA currents, the -CgTx GVIA effect was not significantly reduced in the presence of -Aga IVA. These results indicate that -Aga IVA (and possibly -CmTx MVIIC, cf. Fig. 7C) might act on HVA channels that are resistant to -CgTx GVIA.

The I-V relationship of the current blocked by diltiazem and verapamil is very similar to the I-V curve of the -CgTx-GVIA-sensitive current. Because there are no remarkable differences between the kinetic properties like activation and deactivation (large tails) of verapamil-sensitive currents and -CgTx-GVIA-sensitive current, the assumption that these blockers act on the same target seems to be reasonable. It was, however, impossible to prove this assumption with the use of a combination of verapamil and -CgTx GVIA. Each of both blockers reduced the HVA current in the presence of the other by 25-35%. Even at a concentration of 100 µM, verapamil did not weaken the blocking potency of -CgTx GVIA. Therefore the possibility that several ion channels are present in DUM cells that contribute to the total HVA current cannot be ruled out.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Ca²⁺ currents in cockroach DUM cells and invertebrate neurons

In DUM neurons from cockroach abdominal ganglia, large Ca²⁺ currents were obtained by depolarizing pulses. The maximal current density, with Ba²⁺ as charge carrierwhich was approximately twice that of Ca²⁺ currentscompared with that of Ba²⁺ currents in somata of locust neurons (Pearson et al. 1993). The maximum of the I-V curve in DUM cells was within the range reported for the other invertebrate neurons (between 5 and +20 mV). The electrophysiological and pharmacological analysis of the total Ca²⁺ current in cockroach DUM neurons resulted in the separation of M-LVA currents and HVA currents (see below). A summary of the kinetic and pharmacological properties of the M-LVA and HVA Ca²⁺ currents is shown in Table 2.

Many invertebrate Ca²⁺ currents investigated so far have electrophysiological properties similar to those of vertebrate HVA currents (Akaike et al. 1978; Hayashi and Levine 1992; Pearson et al. 1993).

Concerning activation threshold and kinetics, invertebrate Ca²⁺ currents show a heterogeneity comparable with that of vertebrate currents. Activation of Ca²⁺ currents in cockroach DUM neurons started at 50 mV (Fig. 1). Such a low-threshold activation was also found in neurons of snails (Akaike et al. 1978), locusts (Bickmeyer et al. 1994a; Laurent et al. 1993), and bees (Schäfer et al. 1994). In other preparations, including snail neurons (Barnes et al. 1994; Brezina et al. 1987) and various insect neurons (Byerly and Leung 1988; Hayashi and Levine 1992; Pearson et al. 1993), activation started at potentials positive to 40 mV.

Fast decay of Ca²⁺ currents as seen in DUM neurons had also been observed in snail neurons (Akaike et al. 1978), locusts (Laurent et al. 1993), and bees (Schäfer et al. 1994). In nonspiking locust neurons from thoracic ganglia the time course of the decay of Ca²⁺ currents (5 mM Ca²⁺) was fitted by a sum of two exponentials with time constants of 11 ms and 87 ms, which is similar to the situation found in DUM cells (Laurent et al. 1993).

Pharmacological tools successfully used to separate vertebrate Ca²⁺ currents were of limited utility in inverebrates, e.g., currents showing similar voltage dependence, current kinetics, and sensitivity to verapamil to vertebrate L-type currents were insensitive to dihydropyridines (Bickmeyer et al. 1994a; Byerly and Leung 1988; Hayashi and Levine 1992; McClesky et al. 1987; Pearson et al. 1993; Pelzer et al. 1989; Thomas 1984). Only in Kenyon cells of bees did the dihydropyridine derivative nifedipine have a current-reducing effect at a high concentration of 100 µM (Schäfer et al. 1994). In accordance with these results, the Ca²⁺ currents in cockroach DUM neurons are also insensitive to dihydropyridines (Wicher and Penzlin 1994; this paper). On the other hand, in Drosophila brain membranes dihydropyridine-sensitive Ca²⁺ channels were found that were insensitive to phenylalkylamines (Pelzer et al. 1989).

-CgTx GVIA, which blocks vertebrate N-type channels (McCleskey et al. 1987; Williams et al. 1992), failed to affect Ca²⁺ currents in most invertebrate preparations (Bickmeyer et al. 1994a,b; Bindokas and Adams 1989; McCleskey et al. 1987; Sun et al. 1987). The HVA current blocked by -CgTx GVIA in cockroach DUM cells is hardly similar to vertebrate N-type currents (cf. below).

In the venom of the spider Agelenopsis aperta, several types of -toxins were found. In locust DUM neurons, -agatoxin I and -agatoxin II (at nanomolar concentrations) blocked Ca²⁺ channels (Bindokas and Adams 1989; Bindokas et al. 1991). In contrast to the fast development of the -Aga IVA effect on cockroach M-LVA currents, the effects of -agatoxin I and especially of -agatoxin II appeared very slowly in locust neurons.

Another agatoxin, the spider toxin -Aga IVA, was recently shown to block an HVA current in locust neurosecretory brain cells. To achieve a fast block, a concentration as high as 500 nM had to be used. As in vertebrates, the toxin effect could be reversed by a train of strongly depolarizing pulses (Bickmeyer et al. 1994b). This relief was not observed in -Aga-IVA-sensitive DUM cell M-LVA currents, and, additionally, these currents were rapidly blocked by 50 nM -Aga IVA.

Taking the comparison between the electrophysiological properties of Ca²⁺ currents in cockroach DUM cells and other invertebrate neurons together, one can state that, on the basis of the limited availability of data, the cockroach Ca²⁺ currents are more or less similar to other invertebrate Ca²⁺ currents. From the pharmacological point of view, the results obtained for the DUM cell currents concerning the effects of phenylalkylamines and dihydropyridines are in line with previous findings in many other invertebrate preparations. It seems to be typical for neuronal invertebrate Ca²⁺ channels that phenylalkylamine-sensitive channels lack the dihydropyridine sensitivity. For the effects of -toxins on cockroach M-LVA and HVA currents there is no parallel in other invertebrate neurons.

In general, however, there is presently a lack of information about electrophysiological and pharmacological characterization of invertebrate Ca²⁺ currents. The data on invertebrate Ca²⁺ currents available from literature are not comprehensive enough to allow a detailed comparison with the DUM cell currents. One can expect that the application of -toxins or other more specific agents will allow to introduce a typology of Ca²⁺ currents structured in a manner comparable with that in vertebrates.

Ca²⁺ currents in cockroach DUM cells and vertebrate neurons

In vertebrates Ca²⁺ currents are categorized according to the voltage range of their activation into HVA and LVA currents. Currents activating in an intermediate voltage range and showing some properties of LVA currents were previously found when class E channels were expressed in Xenopus oocytes (Soong et al. 1993). Ca²⁺ channels carrying LVA currents and MVA currents are assumed to belong to one family of a1 subunits called M-LVA (Dunlap et al. 1995).

According to this scheme, we referred to the currents separated in our investigations as M-LVA and HVA currents (for kinetic and pharmacological properties see Table 2). This classification does not mean that we consider two channel types as responsible for total Ca²⁺ current in DUM neurons. It might be, for example, that the total HVA current is a superposition of currents flowing through several channel types (see RESULTS).

M-LVA currents in DUM neurons activated at potentials positive to 50 mV and inactivated during depolarized prepotentials. These currents showed a very rapid decay when Ca²⁺ was the charge carrier. Experiments with Ba²⁺ (Fig. 3) revealed that this decay was mainly Ca²⁺ dependent. Such Ca²⁺-dependent decay is a common property of Ca²⁺ channels in vertebrates and invertebrates that is not restricted to LVA currents (Neely et al. 1994). Cd²⁺ blocked M-LVA currents more effectively than Ni²⁺, but the difference in blocking efficacy was less pronounced than for the HVA currents. M-LVA currents were insensitive to the vertebrate T-type blockers flunarizine and amiloride, but they were affected by -Aga IVA and, most potently, by -CmTx MVIIC.

In vertebrates the LVA T-type currents have a very low activation threshold and are sensitive to predepolarizations (Carbone and Lux 1987; Nowycky et al. 1985). T-type currents are, in contrast to M-LVA currents, only mildly blocked by Cd²⁺. Furthermore, the insensitivity of M-LVA currents to amiloride and flunarizine shows that there is hardly a similarity between vertebrate T-type currents and DUM cell M-LVA currents.

A type of Ca²⁺ currents bearing some similarities with M-LVA currents was obtained when class E channels were expressed in Xenopus oocytes. One of these channels comprising the ₁-subunit rbE-II (Soong et al. 1993) carried a current that activated positive to 50 mV (in 4 mM Ba²⁺) and was maximal at 10 mV. Like M-LVA, the rbE-II current was strongly blocked by Cd²⁺ but it was also sensitive to Ni²⁺ (IC₅₀ = 28 µM). Furthermore, -CgTx GVIA did not affect the rbE-II current, but 200 nM -Aga IVA blocked 33% of the current (-CmTx MVIIC was not tested). As observed in DUM cell Ca²⁺ currents, the -Aga-IVA-induced block did not reverse on a train of strong depolarizations. The inactivation rate of rbE-II Ba²⁺ currents ( ~ 340 ms) is, on the other hand, slower than that of M-LVA currents ( = 20-50 ms with 3 mM Ba²⁺), but the decay depends, as demonstrated by Sather et al. (1993), on the coexpressed -channel subunits.

Another vertebrate M-LVA current is the doe-1 current, which is also connected with a class E channel (Ellinor et al. 1993; Zhang et al. 1993). The sensitivity of doe-1 currents to Ni²⁺ and Cd²⁺ compared with that of the cockroach M-LVA current. But doe-1 currents were insensitive to -CmTx MVIIC and -Aga IVA and were only slightly reduced by -CgTx GVIA (5 µM).

Also, vertebrate HVA currents have some pharmacological properties in common with DUM cell M-LVA currents. P- and Q-type currents are sensitive to -CmTx MVIIC and -Aga IVA (Hillyard et al. 1992; Mintz et al. 1992; Sather et al. 1993; for review see Olivera et al. 1994). Class A channels expressed in Xenopus oocytes are sensitive to -CmTx MVIIC (IC₅₀ < 0.15 µM) and -Aga IVA (IC₅₀ = 0.2 µM; Sather et al. 1993) and insensitive to -CgTx GVIA (Mori et al. 1991). In contrast to the fast block of M-LVA currents, it took ~20 min for 1.5 µM -CmTx MVIIC to achieve maximal block of _A1-currents in Xenopus oocytes (Sather et al. 1993). Approximately the same time was necessary for the 1.5 µM -CmTx-MVIIC-induced inhibition of hippocampal synaptic transmission (Wheeler et al. 1994). The time course of toxin action is concentration dependent (Ellinor et al. 1994; Sather et al. 1993; Wheeler et al. 1994) and depends further on the specificity of peptide binding on the channel proteins. Ellinor et al. (1994) demonstrated that the time constant of -CgTx-induced block was raised 20-fold when, instead of N-type channels, mutants with changes in the putative -CgTx-binding site were investigated. The -CmTx MVIIC effect in DUM cells is interesting insofar as the block was fastwhich indicates a specific binding to the channelsbut, on the other hand, there was a quick recovery from block that was in contrast, e.g., to the irreversible action of -CgTx in N-type channels.

Similarly to P-type currents (Mintz et al. 1992), but in contrast to DUM cell M-LVA currents, the block of class A channels by -Aga IVA could be relieved by strong depolarization. The corresponding _1A-currents are referred to as HVA type. In 2 mM Ba²⁺ they activate positive to 30 mV, and the peak I-V curve is maximal at 5 mV (Sather et al. 1993). This voltage dependence is similar to that of M-LVA Ba²⁺ currents (3 mM), but _1A-currents inactivate considerably more slowly. Taking all the differences between P/Q currents and cockroach M-LVA currents together, it is less probable that M-LVA currents are related to vertebrate P- or Q-type currents.

HVA currents in cockroach DUM neurons activated at potentials positive to 30 mV and were resistant to predepolarizations from negative to 50 mV. For potential steps positive to +10 mV, even with Ca²⁺ as charge carrier, current decay was slow and less pronounced. These currents were effectively blocked by Cd²⁺. The IC₅₀ of 5 µM is in the range known from many vertebrate Ca²⁺ channels. Another result that is typical for HVA currents is the much lower current blocking efficacy of Ni²⁺. DUM cell HVA currents were reduced by verapamil and diltiazem, which are known to block vertebrate L-type channels. But, L-type channels are identified by their sensitivity to dihydropyridines, which did not affect the cockroach HVA currents. Another agent that reduced HVA currents is the specific blocker of vertebrate N-type channels, -CgTx GVIA. The effect of this toxin was in most cases, at least partly reversible.

Vertebrate HVA N-type currents are moderately sensitive to depolarized V_hs. They are strongly and irreversibly blocked by -CgTx GVIA (Boland et al. 1994; Fox et al. 1987; McCleskey et al. 1987; Nowycky et al. 1985; Williams et al. 1992). Rather as an exception, in some objects also a reversible block of N-type currents by -CgTx GVIA was observed (Plummer et al. 1989; Sher and Clementi 1991). Furthermore, N-type currents are sensitive to -CmTx MVIIC (Hillyard et al. 1992), but insensitive to -Aga IVA (Olivera et al. 1994). Although the DUM cell HVA currents were affected by -CgTx GVIA, these currents cannot be considered as N like, because this toxin sensitivity is the only one common property with N-type currents, e.g., the pretreatment with -CmTx MVIIC did not occlude the block of -CgTx GVIA. In vertebrates, however, both the toxins competively block N-type channels.

On the other hand, there is no similarity with P- or Q-type currents, which are sensitive to -Aga IVA and -CmTx MVIIC but not to -CgTx GVIA (Mori et al. 1991; Sather et al. 1993).

Vertebrate L-type currents have a high threshold for activation, they are resistant to predepolarization, and they show large tails on repolarization. They are, furthermore, more sensitive to Cd²⁺ than to Ni²⁺. They are blocked by dihydropyridines, phenylalkyamines, and benzothiazepines. A reversible block by -CgTx GVIA as seen for the DUM cell HVA currents was also reported for L-type currents in some vertebrate preparations (Aosaki and Kasai 1989; Wang et al. 1992) and in class D channels expressed in Xenopus oocytes (Williams at al. 1992). Some Ca²⁺ currents described in invertebrates have properties comparable with those of L-type currents (with the exception of missing dihydropyridine sensitivity). The HVA currents found in DUM neurons share this similarity to the L type. The reason why, nevertheless, DUM cell HVA currents should not be considered as L-type-like current is the missing dihydropyridine sensitivity, which is the most important pharmacological criterion in identification of L-type currents.

Summarizing the results of the comparison between the M-LVA and HVA Ca²⁺ currents in cockroach DUM cells and various types of vertebrate Ca²⁺ currents, we conclude that the insect currents, although showing some similarities with vertebrate currents, have no identity or strong similarity with any known vertebrate Ca²⁺ current. Therefore we propose the existence of Ca²⁺ channels in cockroach DUM cells that are different from the channels characterized so far in vertebrates. It remains unclear whether these new channels occur only in invertebrates.

The -toxins tested proved to be useful tools to separate M-LVA and HVA currents. On the other hand, this utility was limited due to the limited target specificity of thetoxins.

Role of Ca²⁺ currents in cockroach DUM neurons

The somata of most cockroach DUM neurons are electrically excitable. The discharge pattern of these cells is a beating, not a bursting. An essential condition for repetitive activity is a Ca²⁺-dependent K⁺ current, I_K,Ca, which is responsible for afterhyperpolarization (Lapied et al. 1989; Wicher et al. 1994). Therefore one important role of voltage-dependent Ca²⁺ currents isvia controlling I_K,Cathe control of afterhyperpolarization.

Furthermore, when the neuronal activity is modulated, Ca²⁺ channels are a possible target for the modulator. Neurohormone D, a peptide belonging to the adipokinetic hormone family, increases the spike frequency of DUM neurons, which is accompanied by changes in the shape of action potentials, e.g., a stronger and shorter afterhyperpolarization (Wicher et al. 1994). The peptide was shown to enhance a transient component of I_K,Ca that was most probably caused by a peptide-induced potentiation of a transient Ca²⁺ current (Wicher and Penzlin 1994).

As in vertebrates (Takahashi and Momiyama 1993; Uchitel et al. 1992; Wheeler et al. 1994), a major role of the non-L-type currents in the release of transmitters and other mediators from nerve endings is conceivable for the DUM cells investigated here, which are thought to secrete octopamine (Dymond and Evans 1979; Eckert et al. 1992).

	ACKNOWLEDGEMENTS

  We thank Dr. C. Walther for critical reading the manuscript and many helpful discussions. We thank H. Achenbach for testing the effect of amiloride on DUM cell calcium current.

  This work was supported by the Deutsche Forschungsgemeinschaft (Pe 564/1-1).

	FOOTNOTES

  Address for reprint requests: D. Wicher, Sächsische Akademie der Wissenschaften zu Leipzig, PF 100322, D-07703 Jena, Germany.

  Received 2 July 1996; accepted in final form 4 September 1996.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• AKAIKE, N., LEE, K. S., BROWN, A. M. The calcium current of Helix neuron. J. Gen. Physiol. 71: 509-531, 1978.[Abstract/Free Full Text]
• AGRICOLA, H., ECKERT, M., UDE, J., BIRKENBEIL, H., PENZLIN, H. The distribution of a proctolin-like immunoreactive material in the terminal ganglion of the cockroach, Periplaneta americana. Cell Tissue Res. 239: 203-209, 1985.
• AOSAKI, T., KASAI, H. Characterization of two kinds of high-voltage-activated Ca-channel currents in chick sensory neurons. Pfluegers Arch. 414: 150-156, 1989.[Medline]
• BARNES, S., SYED, N. I., BULLOCH, A. G. M., LUKOWIAK, K. Modulation of ionic currents by dopamine in an interneurone of the respiratory central pattern generator of Lymnaea stagnalis. J. Exp. Biol. 189: 37-54, 1994.[Abstract/Free Full Text]
• BICKMEYER, U., RÖSSLER, W., WIEGAND, H. Calcium channel currents in cultured pars intercerebralis neurosecretory cells of adult Locusta migratoria. J. Exp. Biol. 197: 393-398, 1994a.[Abstract]
• BICKMEYER, U., RÖSSLER, W., WIEGAND, H. Omega agatoxin IVA blocks high-voltage-activated calcium channel currents in cultured pars intercerebralis neurosecretory cells of adult Locusta migratoria. Neurosci. Lett. 181: 113-116, 1994b.[Medline]
• BINDOKAS, V. P., ADAMS, M. E. -Aga-I: a presynaptic calcium channel antagonist from venom of the funnel web spider, Agenelopsis aperta. J. Neurobiol. 20: 171-188, 1989.[Medline]
• BINDOKAS, V. P., VENEMA, V. J., ADAMS, M. E. Differential antagonism of transmitter release by subtypes of -agatoxins. J. Neurophysiol. 66: 590-601, 1991.[Abstract/Free Full Text]
• BOLAND, L. M., MORRILL, J. A., BEAN, B. P. -Conotoxin block of N-type calcium channels in frog and rat sympathetic neurons. J. Neurosci. 14: 5011-5027, 1994.[Abstract]
• BREZINA, V., ECKERT, R., ERXLEBEN, C. Suppression of calcium current by an endogeneous neuropeptide in neurones of Aplysia californica. J. Physiol. Lond. 388: 565-595, 1987.[Abstract/Free Full Text]
• BYERLY, L., LEUNG, H.-T. Ionic currents of Drosophila neurons in embryonic cultures. J. Neurosci. 8: 4379-4393, 1988.[Abstract]
• CARBONE, E., LUX, H. D. Kinetics and selectivity of a low-voltage-activated calcium current in chick and rat sensory neurons. J. Physiol. Lond. 386: 547-570, 1987.[Abstract/Free Full Text]
• CATTERALL, W. A., STRIESSNIG, J. Receptor sites for Ca²⁺ channel antagonists. Trends Pharmacol. Sci. 13: 256-262, 1992.[Medline]
• DUNLAP, K., LUEBKE, J. I., TURNER, T. Exocytotic Ca²⁺ channels in mammalian neurons. Trends Neurosci. 18: 89-98, 1995.[Medline]
• DYMOND, G. R., EVANS, P. D. Biogenic amines in the nervous system of the cockroach, Periplaneta americana: association of octopamine with mushroom bodies and dorsal unpaired median (DUM) neurones. Insect Biochem. 9: 535-545, 1979.
• ECKERT, M., RAPUS, J., NÜRNBERGER, A., PENZLIN, H. A new specific antibody reveals mine-like immunoreactivity in cockroach ventral nerve cord. J. Comp. Neurol. 322: 1-15, 1992.[Medline]
• ELLINOR, P. T., ZHANG, J.-F., HORNE, W. A., TSIEN, R. W. Structural determinants of the blockade of N-type calcium channels by a peptide neurotoxin. Nature Lond. 372: 272-275, 1994.[Medline]
• ELLINOR, P. T., ZHANG, J.-F., RANDALL, A. D., ZHOU, M., SCHWARZ, T. L., TSIEN, R. W., HORNE, W. A. Functional expression of a rapidly inactivating neuronal calcium channel. Nature Lond. 363: 455-458, 1993.[Medline]
• FOX, A. P., NOWYCKY, M. C., 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]
• FRANKENHÄUSER, B., HODGKIN, A. L. The action of calcium on the electrical properties of squid axons. J. Physiol. Lond. 137: 218-244, 1957.
• GUNDEL, M., MÖBIUS, P., WICHER, D. Different types of dorsal unpaired median (DUM) neurons in a cockroach abdominal ganglion. In: Göttingen Neurobiology Report 1996, edited by N. Elsner, and H.-U. Schnitzler . Stuttgart, Germany: Thieme, 1996, p. 606
• HAMILL, O. P., MARTY, A., NEHER, E., SAKMANN, B., SIGWORTH, F. J. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pfluegers Arch. 391: 85-100, 1981.[Medline]
• HAYASHI, J. H., LEVINE, R. B. Calcium and potassium currents in leg motoneurons during postembryonic development in the hawkmoth Manduca sexta. J. Exp. Biol. 171: 15-42, 1992.[Abstract/Free Full Text]
• HILLE, B. In: Ionic Channels of Excitable Membranes., . Sunderland, MA: Sinauer, 1992
• HILLYARD, D. R., MONJE, V. D., MINTZ, I. M., BEAN, B. P., NADASDI, L., RAMACHANDRAN, J., MILJANICH, G., AZIMI-ZOONOOZ, A., MCINTOSH, J. M., CRUZ, L. J., IMPERIAL, J. S., OLIVERA, B. M. A new conus peptide ligand for mammalian presynaptic Ca²⁺ channels. Neuron 9: 69-77, 1992.[Medline]
• KASAI, H., NEHER, E. Dihydropyridine-sensitive and -conotoxin-sensitive calcium channels in a mammalian neuroblastoma-glioma cell line. J. Physiol. Lond. 448: 161-188, 1992.[Abstract/Free Full Text]
• LAPIED, B., MALECOT, C. O., PELHATE, M. Ionic species involved in the electrical activity of single adult aminergic neurones isolated from the sixth abdominal ganglion of the cockroach Periplaneta americana. J. Exp. Biol. 144: 535-549, 1989.[Abstract/Free Full Text]
• LAPIED, B., MALECOT, C. O., PELHATE, M. Patch-clamp study of the properties of the sodium current in cockroach single isolated adult aminergic neurones. J. Exp. Biol. 151: 387-403, 1990.[Abstract/Free Full Text]
• LAURENT, G., SEYMOUR-LAURENT, K. J., JOHNSON, K. Dendritic excitability and a voltage-gated calcium current in locust nonspiking local interneurons. J. Neurophysiol. 69: 1484-1498, 1993.[Abstract/Free Full Text]
• LLINAS, R., SUGIMORI, M., HILLMAN, D. E., CHERKSEY, B. Distribution and functional significance of the P-type, voltage-dependent Ca²⁺ channels in the mammalian central nervous system. Trends Neurosci. 15: 351-355, 1992.[Medline]
• MCCLESKEY, E. W., FOX, A. P., FELDMAN, D. H., CRUZ, L. J., OLIVERA, B. M., TSIEN, R. W., YOSHIKAMI, D. -Conotoxin: direct and persistent blockade of specific types of calcium channels in neurones but not in muscle. Proc. Natl. Acad. Sci. USA 84: 4327-4331, 1987.[Abstract/Free Full Text]
• MILJANICH, G. P., RAMACHANDRAN, J. Antagonists of neuronal calcium channels: structure, function and therapeutic implications. Annu. Rev. Pharmacol. Toxicol. 35: 707-734, 1995.[Medline]
• MILLER, R. J., FOX, A. P. Voltage-sensitive calcium channels. In: Intracellular Calcium Regulation, edited by and F. Bronner . New York: Liss, 1990, p. 97-138
• MINTZ, I. M., VENEMA, V. J., SWIDEREK, K. M., LEE, T. D., BEAN, B. P., ADAMS, M. E. P-type calcium channels blocked by the spider toxin -Aga-IVA. Neuron 9: 85-95, 1992.[Medline]
• MORI, Y., FRIEDRICH, T., KIM, M.-S., MIKAMI, A., NAKAI, J., RUTH, P., BOSSE, E., HOFMANN, F., FLOCKERZIE, V., FURUICHI, T., MIKOSHIBA, K., IMOTO, K., TANABE, T., NUMA, S. Primary structure and functional expression from complementary DNA of a brain calcium channel. Nature Lond. 350: 398-402, 1991.[Medline]
• NEELY, A., OLCESE, R., WEI, X., BIRNBAUMER, L., STEFANI, E. Ca²⁺-dependent inactivation of a cloned cardiac Ca²⁺ channel ₁ subunit (_1C) expressed in Xenopus oocytes. Biophys. J. 66: 1895-1903, 1994.[Medline]
• NOWYCKY, M. C., FOX, A. P., TSIEN, R. W. Three types of neuronal calcium channel with different calcium agonist sensitivity. Nature Lond. 316: 440-443, 1985.[Medline]
• OLIVERA, B. M., MILJANICH, G. P., RAMACHANDRAN, J., ADAMS, M. E. Calcium channel diversity and neurotransmitter release: the -conotoxins and -agatoxins. Annu. Rev. Biochem. 63: 823-867, 1994.[Medline]
• PEARSON, H. A., LEES, G., WRAY, D. Calcium channel currents in neurones from locust (Schistocerca gregaria) thoracic ganglia. J. Exp. Biol. 177: 201-221, 1993.[Abstract]
• PELZER, S., BERHANIN, J., PAURON, D., TRAUTWEIN, W., LAZDUNSKI, M., PELZER, D. Diversity and novel pharmacological properties of Ca²⁺ channels in Drosophila brain membranes. EMBO J. 8: 2365-2371, 1989.[Medline]
• PLUMMER, M. R., LOGOTHETIS, D. E., HESS, P. Elementary properties and pharmacological sensitivities of calcium channels in mammalian peripheral neurons. Neuron 2: 1453-1463, 1989.[Medline]
• RANDALL, A. D., WENDLAND, B., SCHWEIZER, F., MILJANICH, G., ADAMS, M. E., TSIEN, R. W. Five pharmacologically distinct high voltage-activated Ca²⁺ channels in cerebellar granule cells. Soc. Neurosci. Abstr. 19: 1478, 1993.
• SATHER, W. A., TANABE, T., ZHANG, J.-F., MORI, Y., ADAMS, M. E., TSIEN, R. W. Distinctive biophysical and pharmacological properties of class A (BI) calcium channel ₁ subunits. Neuron 11: 291-303, 1993.[Medline]
• SCHÄFER, S., ROSENBOOM, H., MENZEL, R. Ionic currents of Kenyon cells from the mushroom body of the honeybee. J. Neurosci. 14: 4600-4612, 1994.[Abstract]
• SHER, E., CLEMENTI, F. -Conotoxin-sensitive voltage-operated calcium channels in vertebrate cells. Neuroscience 42: 301-307, 1991.[Medline]
• SOONG, T. W., STEA, A., HODSON, C. D., DUBEL, S. J., VINCENT, S. R., SNUTCH, T. P. Structure and functional expression of a member of the low voltage-activated calcium channel family. Science Wash. DC 260: 1133-1138, 1993.[Abstract/Free Full Text]
• SUN, X. P., TAKEUCHI, H., OKANO, Y., NOZAWA, Y. Effects of synthetic -conotoxin GVIA (-CgTX GVIA) on the membrane calcium current of an identifiable giant neurone, d-RPLN, of an African giant snail (Achatina fulica Ferussac), measured under voltage clamp condition. Comp. Biochem. Physiol. C Comp. Pharmacol. 87: 363-366, 1987.
• TAKAHASHI, T., MOMIYAMA, A. Different types of calcium channels mediate central synaptic transmission. Nature Lond. 366: 156-158, 1993.[Medline]
• THOMAS, M. V. Voltage-clamp analysis of a calcium-mediated potassium conductance in cockroach (Periplaneta americana) central neurons. J. Physiol. Lond. 350: 159-178, 1984.[Abstract/Free Full Text]
• TSIEN, R. W., TSIEN, R. Y. Calcium channels, stores, and oscillations. Annu. Rev. Cell Biol. 6: 715-760, 1990.
• UCHITEL, O. D., PROTTI, D. A., SANCHEZ, V., CHERSKEY, B. D., SUGIMORI, M., LLINAS, R. P-type voltage-dependent calcium channel mediates presynaptic calcium influx and transmitter release in mammalian synapses. Proc. Natl. Acad. Sci. USA 89: 3330-3333, 1992.[Abstract/Free Full Text]
• WANG, X., TREISTMAN, S. N., LEMOS, J. R. Two types of high-threshold calcium currents inhibited by -conotoxin in nerve terminals of rat neurohypophysis. J. Physiol. Lond. 445: 181-199, 1992.[Abstract/Free Full Text]
• WHEELER, D. B., RANDALL, A., TSIEN, R. W. Roles of N-type and Q-type Ca²⁺ channels in supporting hippocampal synaptic transmission. Science Wash. DC 264: 107-111, 1994.[Abstract/Free Full Text]
• WICHER, D., PENZLIN, H. Ca²⁺ currents in cockroach neurones: properties and modulation by neurohormone D. Neuroreport 5: 1023-1026, 1994.[Medline]
• WICHER, D., PENZLIN, H. -Conotoxin GVIA-, -agatoxin IVA-, and -conotoxin MVIIC-sensitive Ca²⁺ currents in central insect neurones. Pfluegers Arch. 429, Suppl.: 149, 1995.
• WICHER, D., WALTHER, C., PENZLIN, H. Neurohormone D induces ionic current changes in cockroach central neurones. J. Comp. Physiol. A Sens. Neural Behav. Physiol. 174: 507-515, 1994.[Medline]
• WILLIAMS, M. E., BRUST, P. F., FELDMAN, D. H., SARASWATHI, P., SIMERSON, S., MAROUFI, A., MCCUE, A. F., ELLIS, S. B., HARPOLD, M. M. Structure and functional expression of an -conotoxin-sensitive human N-type calcium channel. Science Wash. DC 257: 389-395, 1992.[Abstract/Free Full Text]
• ZHANG, J.-F., RANDALL, A. D., ELLINOR, P. T., HORNE, W. A., SATHER, W. A., TANABE, T., SCHWARZ, T. L., TSIEN, R. W. Distinctive pharmacology and kinetics of cloned neuronal Ca²⁺ channels and their possible counterparts in mammalian CNS neurons. Neuropharmacology 32: 1075-1088, 1993.[Medline]

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