| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Neurobiology and Behavior, Cornell University, Ithaca, New York 14853
Submitted 5 January 2003; accepted in final form 26 March 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
;
Marder and Bucher 2001;
Marder and Calabrese 1996;
Selverston et al. 1998). The
output of this network is very plastic: the network can be reconfigured to
produce multiple variants of a single motor pattern, and its component neurons
can even be recruited into other motor networks
(Dickinson and Moulins 1992;
Nusbaum and Beenhakker 2002).
This plasticity has made the pyloric network a model system in which to
examine the cellular and synaptic mechanisms underlying behavioral flexibility
(Harris-Warrick et al. 1992).
In the spiny lobster, Panulirus interruptus, the pyloric network
contains 14 identified neurons in six major classes: the anterior burster
(AB), two pyloric dilators (PDs), the ventricular dilator (VD), the inferior
cardiac (IC), the lateral pyloric (LP), and eight pyloric constrictors (PYs).
The synaptic connectivity of the network is completely known, including the
identity of the fast synaptic transmitters used by the pyloric neurons
(Marder 1987;
Mulloney 1987), and many of
the intrinsic electrical properties of individual neurons are characterized
(Bal et al. 1988;
Guckenheimer et al. 1997;
Harris-Warrick et al.
1995a,b;
Hartline and Graubard 1992;
Kloppenburg et al. 1999).
Pyloric network plasticity arises in part from the actions of
neuromodulators, delivered via ascending or descending axons or as circulating
hormones, which can transiently change the intrinsic electrical properties of
network neurons and/or the strength of network synapses
(Harris-Warrick and Marder
1991; Marder and Thirumalai
2002; Skiebe
2001). We have been studying the modulatory effects of dopamine
(DA). This monoamine can initiate a unique pyloric motor pattern in quiescent
preparations (Anderson and Barker
1981; Eisen and Marder
1984; Flamm and Harris-Warrick
1986a,b)
and can alter rhythm frequency in actively cycling preparations
(Ayali and Harris-Warrick 1999;
Harris-Warrick et al.
1995b).
DA modulates the pyloric network through its distributed effects on the
ionic currents that shape the neurons' intrinsic electrical properties and by
modulating the strength of chemical and electrical synapses throughout the
network (Ayali et al. 1998;
Harris-Warrick et al.
1995ab,
1998;
Johnson et al. 1995;
Kloppenburg et al. 1999,
2001;
Nusbaum and Beenhakker 2002).
DA evokes rhythmic bursting in the AB neuron, excites the LP, PY, and IC
neurons, and inhibits the VD and PD neurons
(Flamm and Harris-Warrick
1986b; Marder and Eisen
1984; Turrigiano and Marder
1993). Graded transmitter release between the network neurons
organizes the pyloric output (Hartline et
al. 1988). DA enhances the strength of graded synaptic
transmission at most of the glutamatergic synapses by increasing glutamate
release from the AB, LP, and PY neurons
(Johnson and Harris-Warrick
1997). At the same time, it weakens graded synapses from the
cholinergic PD neuron by decreasing presynaptic transmitter release
(Johnson and Harris-Warrick
1990; Johnson et al.
1995). The ionic mechanisms of this synaptic modulation are at
present not known. One reasonable target is the voltage-sensitive calcium
current (ICa), which is a common target by which
modulators can regulate transmitter release
(Dunlap and Ikeda 1998;
Fossier et al. 1999;
Wu and Saggau 1997) and
activate cellular processes that shape the intrinsic firing properties of the
pyloric and other neurons (Hille
2001; Liu et al.
1998).
The purpose of our study was to characterize ICa in the six classes of pyloric neurons in the spiny lobster and to examine whether DA can modify this current in ways that might explain its effects on synaptic transmission and intrinsic firing properties of the neurons. We have found that DA changes ICa in all the pyloric neurons. Most of these effects are consistent with our hypothesis that DA modulates synaptic strength at least in part by targeting presynaptic ICa. In addition, DA modulation of ICa may support its alteration of the intrinsic firing properties in most of the pyloric neurons. Finally, we show that DA acts on different Ca2+ channel subtypes in the different classes of pyloric neurons.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
California spiny lobsters (P. interruptus) were supplied by Don Tomlinson Commercial Fishing (San Diego, CA) and maintained in marine aquaria at 16°C. All chemicals were purchased from Sigma Chemical.
Preparation
Lobsters were anesthetized by cooling in ice. The stomatogastric nervous
system was removed as previously described
(Selverston et al. 1976), and
pinned in a silicone elastomer (Sylgard)-coated petri dish in chilled
Panulirus saline of the following composition (mM): 479 NaCl, 12.8
KCl, 13.7 CaCl2, 3.9 Na2SO4, 10.0
MgSO4, 2 glucose, and 11.1 Tris base, pH 7.4
(Mulloney and Selverston
1974). The STG was desheathed, enclosed in a 1-ml pool walled with
petroleum jelly (Vaseline) and superfused at 5 ml/min with oxygenated
Panulirus saline (17–18°C).
Cell identification and synaptic isolation
Our methods for identifying the six major classes of pyloric neurons and
for isolating them pharmacologically from known synaptic input have been
previously described in detail (Johnson
and Harris-Warrick 1997; Kloppenburg et al.
1999,
2000a;
Peck et al. 2001). We
determined the subclass of most of the PY neurons by injecting hyperpolarizing
current into a PY neuron and noting any change in firing of the LP neuron.
Most PY neurons in this study belonged to a subclass not electrically coupled
to the LP neuron (Hartline et al.
1987; Levini et al.
1994). Because there is relatively strong electrical coupling
between the AB and PD neurons and weaker coupling of these cells with the VD
neuron (Johnson et al. 1993),
we isolated a neuron from its electrically coupled partners by
photoinactivation (Miller and Selverston
1979) in at least one of each series of experiments for each cell
type; this made no obvious differences in our voltage-clamp results (see also
Swensen and Marder 2000).
Voltage clamp, current isolation, and characterization
Two-electrode voltage-clamp was used to examine ICa and
outward currents in pyloric neurons. One electrode, filled with 3 M KCl (7- to
11-M
resistance), was used to inject current, while the second, filled
with 2 M TEA and 2 M CsCl (15- to 25-M resistance), was used to record
voltage and to iontophorese these K+ channel blockers internally.
Neurons were voltage-clamped using an Axoclamp-2B amplifier and pCLAMP8
software (Axon Instruments). Linear leak and capacitative currents were
digitally subtracted using a p/6 protocol
(Armstrong and Bezanilla
1974).
We attempted to isolate ICa from other
voltage-dependent currents by iontophoresing TEA and CsCl into the cell soma
to reduce K+ currents and by using a lobster blocking saline
containing 100 mM tetraethyl-ammonium chloride (TEA) (Na+ reduced
appropriately to maintain osmotic and charge balance), 4 mM 4-aminopyridine
(4-AP), 5 mM CsCl, and 0.1 µM TTX; the pH was adjusted to 7.4 with HCl. We
also added picrotoxin (5 x 10–6 M) to block
all glutamatergic transmission within the STG. Cl– was
substituted for SO42– in all blocking
saline to avoid precipitation when using divalents other than
Ca2+. A depolarizing leak current is initially induced
in PD neurons by 4-AP (Kloppenburg et al.
1999) and other structurally related IA
blockers (Johnson and Kloppenburg, unpublished observations). However, this
disappears within 30 min in the continued presence of 4-AP. In some
experiments, we substituted Ba2+ or
Sr2+ for Ca2+ as the inward charge
carrier. In a few experiments, we further isolated ICa in
the AB neuron from Ca2+-dependent currents by
intracellularly injecting 0.6 M
bis-(2-aminophenoxy)-N,N,N',N'-tetraacetic acid
(BAPTA), in 60 mM KCl, using 500-ms duration, –5-nA pulses at 1 Hz for
10 min. Measurements were begun after 
30 min in blocking saline.
ICa characterization and DA modulation
The I-V relationship for ICa was determined in each pyloric neuron type using 5-mV incremental steps of 150- to 200-ms duration from a holding potential of –60 mV. The effects of DA (100 µM) and the Ca2+ channel blockers CdCl2 (200–600 µM), NiCl2 (100 µM), and nifedipine (10–300 µM) were determined by comparing the mean of five peak ICa measurements from a holding potential of –50 mV to –15 or –10 mV, whichever produced the maximal fast inward current with no late outward current (this was determined by prior I-V characterization), in control (after the peak ICa had attained a steady amplitude) and after 5 (DA) or 20 (Ca2+ channel blockers) min perfusion. DA effects on blocker resistant outward currents were determined in a similar way.
DA was freshly dissolved in blocking solution before each application. A
stock solution of nifedipine in DMSO was prepared daily and was diluted with
blocking saline immediately before use. The final concentration of DMSO
(
0.1%) alone had no effect on ICa. Nifedipine was applied with
room lights off.
Data analysis
We measured the peak current amplitudes with Axograph 4 (Axon Instruments). Results were discarded if the DA effect did not reverse or, in the AB neuron, when DA caused uncontrolled voltage oscillations. An ANOVA, followed by protected t-test, determined the significant differences between individual data groups. Statistical significance was accepted with P < 0.05 for F or t values. Mean current amplitudes and percentages are reported ± SE.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
Under conditions designed to minimize voltage-gated Na+, K+, and H currents, we recorded voltage-dependent inward currents from identified pyloric neurons during depolarizing voltage steps from either –60 or –50 mV. These inward currents had the characteristics of typical Ca2+ currents. First, Cd2+ was an effective blocker of this current. In the PY neuron of Fig. 1A, 200 µM Cd2+ reduced the inward current at –10 mV by 80%. Similar block was seen in PY, PD, and AB neurons. In three ABs, one VD and one IC neuron, increasing the Cd2+ concentration to 600 µM completely blocked the inward current, leaving a slowly rising outward current that was not eliminated by our standard blocking saline (see AB example in Fig. 6C).
|
|
In a few experiments, we compared the inward current in a neuron using Ca2+ or Ba2+ as charge carriers. We did not routinely use Ba2+ saline because the Ba2+ current was often unstable and ran down quickly. Figure 1B shows that in a PY neuron, the Ba2+ current's amplitude was larger and it showed much slower inactivation than ICa, as is characteristic of Ca2+ channels. Similar results were seen in three PY neurons, two ABs, a PD and an IC neuron. These results indicate that under our conditions of pharmacological blockade, the inward current in pyloric neurons is mediated by Ca2+ channels.
We also examined the effects of 100 µM NiCl2 on
ICa in AB, PD, and PY neurons because
Ni2+ blocks the low-threshold ICa
responsible for graded chemical synaptic transmission in the leech
(Angstadt and Calabrese 1991).
The pyloric Ca2+ current was not affected by
Ni2+ (Fig.
1C; n = 3 for all cell types), even after very
long applications (Fig.
1D; this figure also shows the stability of the
ICa in our preparation).
Voltage dependence of ICa in pyloric neurons
Figure 2 shows typical examples of ICa for each pyloric neuron in response to a series of depolarizing steps from –60 to +20 mV. Small voltage steps evoked small, slowly activating currents (see clearest examples in Fig. 2, C and D). Above a critical threshold, most cells showed a sudden jump to a much larger current, but with a variable latency to initiate the current after the voltage step began. With greater depolarizing steps, this large current activated with a shorter latency. In some cells, maximal current amplitude could be reached with the first large current jump (Fig. 2, A, C, and E). In others, the current increased in a more graded manner until a peak was reached (Fig. 2, B, D, and F). With further depolarization, the peak current declined, as expected for a reduction in driving force. Despite the presence of external and internal K+ channel blockers, almost all PD and AB cells, most IC cells and half of the PY cells showed small outward currents at voltage steps above +10 mV (Fig. 2, A, B, E, and F). However, in VD and LP cells, this outward current was not seen (Fig. 2, C and D). In all cell types, a slowly deactivating inward current remained after the end of the voltage step. Repolarization was often accompanied by an active and uncontrolled inward current, especially in AB, LP, and IC cells (for example, Fig. 2C). This uncontrolled inward current was amplified in Ba2+ saline (Fig. 1B), suggesting that it was carried through Ca2+ channels rather than a Ca2+-activated inward current.
|
The characteristics of these inward currents suggest that they did not
arise in the cell body, where our electrodes were placed. Instead, they
probably arose from the neuropil that is electrotonically distant from the
soma and that was under incomplete voltage control
(Graubard and Hartline 1991).
This poor space clamp of a distal current causes the delayed onset of rapidly
activating currents, the sudden jump to near peak current with small step
increases, and the long-lasting inward currents after repolarization.
This imperfect voltage control and the apparent contamination of ICa by residual outward currents prevented a detailed biophysical analysis of the current parameters in each pyloric neuron. Nonetheless, average I-V plots for the inward currents can still give some insight into the approximate voltage dependence of this current and its function in pyloric neurons. Figure 3 shows mean I-V plots for the AB (n = 9), PD (n = 12), LP (n = 9), VD (n = 8), IC (n = 8), and PY (n = 11) neurons. The AB and LP neurons appear to have slightly more hyperpolarized average voltage thresholds (around –45 mV: Fig. 3, A and C) than the other neurons (around –40 mV). The sudden jump in current amplitude (Fig. 3, A, C, and E) reflects the activation of currents in the poorly clamped region of the cells (Fig. 2). The LP neuron appears to have the largest mean peak amplitude of the current, whereas the AB has the smallest. Due to the space-clamp problems described in the preceding text, it is difficult to determine the voltage of half-activation or the voltage for the peak inward current in these neurons. Finally, the extrapolated reversal potential in all cell types was around +40 mV.
|
Poor voltage control of a current arising at a significant electrotonic distance from the recording site will, of course, skew the I-V parameters described in the preceding text. However, in some neurons, much better voltage control was achieved, and in these cases, the I-V characteristics were also within the range of values seen in Fig. 3. For example, the PY neuron in Fig. 4A shows inward Ba2+ currents under much better voltage control as seen by the graded increase in inward current with step depolarizations, the lack of an initial delay in current activation, and good voltage control on repolarization. The I-V plot from this cell (Fig. 4B) shows an activation threshold around –45 to –50 mV, and a peak current of 11 nA at –10 mV. The activation threshold is at the hyperpolarized end of the range for PY cells, perhaps due to better space clamp. However, the values for peak current amplitude and peak voltage are well within the range seen for PY currents recorded under poorer voltage control (Fig. 3). This suggests that the voltage properties in Fig. 3 may roughly approximate the properties of ICa in the different pyloric neurons.
|
DA modulation of pyloric Ca2+ currents
We limited our analysis of the effects of DA to the peak ICa amplitude because it appeared to be less affected by imperfect space clamp than other current parameters. An ANOVA test showed that DA had overall statistically significant effects (P < 0.05) on the peak amplitude of ICa across the different pyloric neuron types.
DA enhanced ICa in most PY and LP neurons and in all IC neurons (Fig. 5A). In seven of eight PY neurons, DA significantly increased the mean peak ICa by 15 ± 4%. In the eighth PY neuron, DA reversibly decreased ICa by 20%. We did not include this cell in our statistical analysis because it was more than 4 SD from the mean DA effect (but see DISCUSSION). Six PY neurons (including the outlier) belonged to the subclass of PY neurons not electrically coupled to the LP; one other PY was electrically coupled to the LP, whereas one PY neuron was not categorized. Thus the effects of DA on ICa in PY neurons did not appear to correlate with PY subclass type. Figure 5B shows an example of the time course of the DA enhancement of ICa in one PY neuron.
|
In seven LP neurons, DA significantly increased the mean peak ICa by 8 ± 4%. DA weakly, but consistently, and thus significantly, increased the peak ICa in IC neurons by 4 ± 1% (n = 5; Fig. 5A). We conclude that the main effect of DA is to enhance ICa in PY, LP, and IC neurons.
In contrast, DA weakly decreased the peak ICa in VD neurons and more strongly decreased ICa in PD and AB neurons (Fig. 5A). Peak ICa in VD neurons was weakly but consistently decreased by an average of 4 ± 1% (n = 5). DA evoked a more robust and statistically significant 23 ± 4% decrease of mean peak ICa in PD neurons (n = 8; example in Fig. 5C). Finally, DA caused a large and significant 50 ± 9% decrease of mean peak ICa in AB neurons (n = 12; example in Fig. 5D).
Effects of DA on blocker resistant outward currents in the AB neuron
We were surprised that DA appeared to reduce ICa in AB
neurons because DA strongly enhances release from AB synapses
(Johnson and Harris-Warrick
1997; Johnson et al.
1995). Therefore we examined whether DA enhancement of an outward
current that persists in our blocker saline
(Fig. 2A) might mask a
direct DA effect on ICa. Under control conditions in
blocker saline, this outward current first appeared around –15 mV at the
end of the current trace, and then progressively dominated the current above
+10 mV (Fig. 6A,
control). The current at the higher voltage steps rose relatively slowly (in
part due to the counter inward ICa), and did not
inactivate during the voltage step (Fig.
6A, control). DA increased the peak amplitude of the
outward current at the end of the voltage step while reducing the peak
amplitude of ICa at the beginning of the step
(Fig. 6A, DA). During
DA application, the outward current became apparent at more hyperpolarized
voltage steps (around –25 mV); it dominated the current trace at steps
above –10 mV and did not reach its peak during the 200-ms voltage step
(Fig. 6A, DA). In four
AB experiments where we compared DA's effects on ICa and
the outward current, DA significantly and reversibly reduced the mean peak
ICa by 2.7 ± 1.3 nA and significantly and
reversibly enhanced the peak outward current at +20 mV by 3.5 ± 1.2 nA
(Fig. 6B). In a
separate series of experiments, we tested whether the DA-enhanced outward
current is Ca2+ dependent by blocking
ICa with 600 µM Cd2+. A
significant outward current remained in the presence of
Cd2+, which was activated with little delay at voltages
above –15 mV and did not reach its peak during the voltage step
(Fig. 6C). This
remaining, Cd2+-resistant outward current was not
significantly modified by DA (Fig.
6D; n = 3, P = 0.28). Thus DA
enhancement of a Ca2+-dependent outward current in AB
neurons could possibly mask a direct effect of DA on
ICa.
We then set out to separate ICa from the blocker resistant, Ca2+-dependent outward current. Preliminary experiments using Ba2+ as the inward charge carrier were not successful, because the Ba2+ currents were very unstable and rapidly ran down in the AB neuron. Experiments using Sr2+, reduced external Ca2+ (25–50% of normal) or lower concentrations of Cd2+ (200 µM) failed to separate ICa from the outward current and still allowed enhancement of the outward current by DA (not shown). Intracellular injection of the calcium chelator, BAPTA, however, did separate ICa from the outward current. In three experiments, intracellular BAPTA reduced the outward current at +20 mV to 84 ± 8% of preinjection values. In the experiment shown in Fig. 7, intracellular BAPTA abolished the apparent inactivation of ICa during the voltage step to –20 mV (where the outward current was not yet activated), and enhanced the uncontrolled inward current following the voltage step (Fig. 7A). Under these conditions, DA still reversibly reduced ICa by 47% (Fig. 7B), but this was no longer linked to Iout, which was detectable at higher voltages. Similar results were seen in two other AB neurons. These experiments showed a statistically significant reduction in ICa of 56 ± 11% (measured at –20 mV), and a much more variable and not statistically significant increase in the outward current (57 ± 25%, measured at +20 mV: P = 0.15; Fig. 7C). We conclude that the DA-induced reduction of ICa in the AB neuron is a direct effect and is not caused by a consistently enhanced outward current.
|
Nifedipine blocks the effect of DA on ICa in PD and PY neurons but not in AB neurons
Hurley and Graubard (1998)
demonstrated that the dihydropyridine nifedipine, a selective L-type channel
blocker in vertebrates, blocks ICa in unidentified,
cultured crab STG neurons. We examined the effect of nifedipine on DA-induced
modulation of ICa from Panulirus AB, PD and PY
neurons in situ because DA had the largest effects on ICa
in these cells.
Nifedipine was an effective blocker of ICa in PY neurons (Fig. 8A). Detectable block was obtained with 10 µM nifedipine; 100 µM nifedipine significantly reduced mean peak ICa by 38 ± 8% (n = 5), whereas 200 µM nifedipine had only a slightly greater effect (56 ± 9% reduction, n = 4). Similar effects were seen in PD neurons (Fig. 8B), with detectable reductions at 10 and 50 µM, 55 ± 11% (n = 6) reductions at 100 µM, and only slightly greater block at 200 µM (62 ± 16%, n = 5) with no further increase at 300 µM (n = 2). In one LP neuron, 100 µM nifedipine caused a 42% reduction in peak ICa.
|
AB neurons showed the greatest variability in their response to nifedipine, but in all cases, nifedipine had a much weaker blocking effect than with the other neurons (Fig. 8C). Nifedipine (100 µM) weakly reduced ICa by 20 ± 12% (n = 4, P = 0.18), whereas 200 µM nifedipine caused a nearly significant 28 ± 10% reduction in ICa (n = 5; P = 0.06). These results suggest that nifedipine-sensitive Ca2+ channels carry a major fraction of the total ICa in PY, PD, and possibly LP neurons but a much smaller fraction in AB neurons. In a separate series of experiments, we used a saturating concentration of nifedipine (200 µM) to examine whether the nifedipine-sensitive ICa was modulated by DA. Before nifedipine application, DA evoked a 14 ± 3% increase in the peak ICa (P = 0.06; n = 3) in PY neurons, and a statistically significant decrease of 18 ± 2% (n = 3) in the peak ICa of PD neurons (Fig. 8, D and E). Twenty-minute incubation in nifedipine significantly reduced ICa in PY and PD neurons by 52 ± 11 and 72 ± 15%, respectively (Fig. 8, D and E). In the presence of nifedipine, DA had no detectable effect on the remaining ICa in either cell type (Fig. 8, D and E; examples in Fig. 8, G and H). Nifedipine appeared to occlude the effects of DA on this current in PY and PD neurons.
In contrast, nifedipine did not occlude the DA reduction of ICa in the AB neuron (n = 3). Dopamine significantly reduced the peak ICa by 75 ± 13% before nifedipine treatment; after recovery of ICa, 200 µM nifedipine reduced mean peak ICa by 33 ± 14% (P = 0.16; Fig. 8F). Addition of DA in the presence of nifedipine completely abolished the remaining ICa (Fig. 8F; example in Fig. 8I). Thus DA appears to act on nifedipine-sensitive Ca2+ channels in PY and PD neurons but at least in part on nifedipine-resistant channels in the AB neuron.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
Our results provide an initial characterization of ICa
in pyloric neurons of the spiny lobster. From a holding potential of –60
mV, ICa appeared between –50 and –40 mV in the
different neurons. As discussed in RESULTS, these values are not
very accurate due to poor space clamp of currents arising in the distal
neuropil (Kloppenburg et al.
2000a). It is likely that we could not detect small, distant
Ca2+ currents, and thus the real thresholds for
activation are probably more hyperpolarized than our values (see also
Charlton and Augustine 1990).
In the few cases where we had better voltage control, however, the
I-V parameters of ICa were similar to the
population values, suggesting that our characterization at least approximated
the real values.
In the vertebrates, Ca2+ currents are usually
classified into low-voltage-activated (LVA or T-type, with activation positive
to –70 mV) and high-voltage-activated (HVA, activation positive to
around –30 mV) classes; subtypes of the HVA currents (L-, P-, Q-, N-,
and R-type) are defined by biophysical and pharmacological properties
(Ertel et al. 2000;
Hille 2001;
Triggle 1999). The
ICa in pyloric neurons does exhibit the general
characteristics of some HVA channel types: Ba2+ is a
better charge carrier than Ca2+, inactivation is
Ca2+ dependent and relatively slow compared with LVA
channels, the currents are insensitive to Ni2+, and a
majority of the current in some cells is blocked by the L-type blocker
nifedipine. However, the activation range of our currents is more
hyperpolarized than traditional HVA channels and thus resemble currents with
L-type properties that have more mid-voltage activated ranges. Such currents
in vertebrates are seen in hypothalamic neurons
(Akaike et al. 1989), retinal
bipolar neurons (Heidelberger and Matthews
1992), hippocampal neurons
(Avery and Johnston 1996), hair
cells (Fuchs 1996), and the
L-type subunit CaV1.3
1 expressed in
Xenopus oocytes (Xu and
Lipscombe 2001). The calcium currents in the DG gastric motor
neuron in the STG of the crab Cancer borealis
(Zhang and Harris-Warrick
1995), in unidentified cultured STG neurons from the spiny lobster
(Turrigiano et al. 1995), and
in crayfish motor neurons (Hong and
Lnenicka 1995; Wright et al.
1996) also have mid-voltage activation ranges as do other
invertebrate Ca2+ currents, such as those in squid giant
presynaptic terminals (Charlton and
Augustine 1990), insect neurons
(Wicher and Penzlin 1997), and
snail neurons (Pin et al.
1990). Calcium currents recorded from the LP neuron in the crab,
C. borealis (Golowasch and Marder
1992), and from unidentified cultured STG neurons from the crab
C. productus (Hurley and Graubard
1998) appear to have more depolarized I-V characteristics
and thus, in this way, resemble more classical HVA currents.
Our limited experiments with nifedipine suggest that the total
ICa we recorded in pyloric neurons is composed of multiple
subtypes and that these subtypes can differ between pyloric neurons. Hurley
and Graubard (1998) first
provided evidence that multiple Ca2+ channel types
contribute to the total ICa and the control of synaptic
transmission in crab STG neurons as is the case in many other neurons (for
example, Fossier et al. 1999;
Lnenicka and Hong 1997). Different STG Ca2+ channel
types could be targeted to different cellular sites
(Hartline and Graubard 1992;
see also French et al. 2002).
In our experiments, nifedipine, even at high concentrations, only blocked part
of ICa in PD and PY neurons. The much weaker block by nifedipine of
ICa in the AB neuron suggests that the different pyloric
neuron types may express different ratios of Ca2+
channel types.
In addition, it appears that DA modulates different Ca2+ channel types in different pyloric neurons because nifedipine completely blocked the DA effect on ICa in PY and PD neurons but not in AB neurons. Differing complements of Ca2+ channel types in the different pyloric neurons, especially the AB neuron, should not be surprising, considering the differing roles the neuron types play in pyloric rhythm generation.
Dopamine configuration of the pyloric network
Dopamine reconfigures the Panulirus pyloric network by modifying
the intrinsic properties of the pyloric neurons and altering the strength of
all of the network's synaptic connections
(Harris-Warrick et al. 1998).
DA can initiate a motor pattern from a quiescent preparation, with descending
input blocked, and can alter the pattern in an actively cycling preparation.
It does so by enhancing bursting pacemaker properties in the AB neuron,
exciting the LP, PY, and IC neurons, but inhibiting and reducing activity in
the PD and VD neurons (Eisen and Marder
1984; Flamm and Harris-Warrick
1986a,b;
Makino et al. 2000). In
addition, it strengthens or weakens virtually every synapse in the network
(Johnson et al. 1995).
Dopamine modulation of intrinsic firing properties of pyloric neurons
We are beginning to understand some of DA's actions on the intrinsic
electrical properties of pyloric neurons and their underlying currents, which
contribute to pyloric network configuration. DA inhibits the transient
K+ current (IA) in the AB, LP, PY, and IC
neurons, contributing to their enhanced activity and phase advances in the
pyloric motor pattern (Harris-Warrick et al.
1995a,b;
Peck et al. 2001;
Tierney and Harris-Warrick
1992). At the same time DA enhances IA and a
calcium-dependent outward current (IK(Ca)) in the PD cell,
contributing to its reduced activity
(Kloppenburg et al. 1999). DA
also enhances Ih in the LP, further enhancing its
postinhibitory rebound properties
(Harris-Warrick et al.
1995b).
Dopamine's effects on ICa could contribute to its
modulation of the firing properties of the pyloric neurons. DA increases peak
ICa in the LP, PY, and IC neurons, consistent with its
enhancement of firing in these cells, and reduces peak ICa
in PD and VD neurons, consistent with its inhibition of these cells
(Flamm and Harris-Warrick
1986b; Turrigiano and Marder
1993). These effects could reflect modulation of the direct
depolarizing effects of ICa, which in our measurements had
a rather slow rate of inactivation (see Figs.
1 and
2). Thus in neurons that are
excited, the enhanced ICa, which is activated at
subthreshold voltages (Fig. 3),
could help initiate bursting and bistability and prolong them for a longer
time, as seen, for example, in spinal motor neurons
(Alaburda et al. 2002;
Carlin et al. 2000).
Alternatively, increases in intracellular Ca2+ could
activate calcium-activated nonselective currents (ICAN);
this would boost the effects of calcium entry
(Partridge et al. 1994;
Zhang et al. 1995), and/or
initiate second-messenger cascades that could enhance firing
(Hille 2001). The loss of such
inward current might also contribute to the hyperpolarization and reduction in
firing of the PD and VD neurons.
Dopamine initiates bursting in a quiescent AB neuron
(Flamm and Harris-Warrick
1986b), and enhances burst amplitude and cycle frequency in an
isolated oscillating AB neuron (Ayali and
Harris-Warrick 1999). We show here that DA significantly decreases
peak ICa by 50% in the isolated AB neuron, suggesting that
Ca2+ is not responsible for the main depolarizing drive
underlying DA-evoked bursting in AB neurons as it is for other invertebrate
bursting neurons (Kits and Mansvelder
1996). In a previous study
(Harris-Warrick and Johnson
1987; see also Gola and
Selverston 1981), we showed that reduction of extracellular
Ca2+ and partial blockade of Ca2+
currents can also initiate rhythmic bursting in a quiescent AB neuron. A
sufficient reduction in [Ca2+]out during
DA-induced bursting depolarizes the neuron, and bursting stops at the peak
level of the voltage oscillation
(Harris-Warrick and Flamm
1987; Johnson et al.
1992). These results together suggest that in the AB neuron,
ICa may actually be inhibiting rhythmic oscillatory
capability and a reduction in ICa will paradoxically
enhance bursting. This could occur by a block of IK(Ca),
which might be actively preventing bursting from occurring. Indeed, TEA, which
blocks IK(Ca) as well as the very small
IK(V) in these neurons, also initiates rhythmic bursting
in a quiescent AB neuron (Harris-Warrick
and Johnson 1987). However, our results here suggest that DA
enhances a calcium-dependent outward current, possibly
IK(Ca) in the AB neuron
(Fig. 6). It is possible that
DA has additional or different effects on currents that are not easily
detected from the soma, or that additional, unidentified
Ca2+-dependent outward currents also inhibit and
structure AB bursting activity. Further work will be needed to understand the
dynamics of bursting in this pacemaker cell.
Dopamine modulation of synaptic interactions in the pyloric network
The mechanisms of DA modulation of the pyloric graded chemical synapses are
less clear. All synapses in the pyloric network are targets of DA: the AB, LP,
and PY chemical graded synapses are enhanced by DA, whereas the PD, VD, and
(to a lesser extent) IC chemical synapses are inhibited
(Johnson et al. 1995). We know
little of the synaptic mechanisms or ionic currents targeted by DA to alter
synaptic strength. However, at many of these synapses, DA appears to act at
least in part presynaptically to enhance or decrease release of transmitter
(Johnson and Harris-Warrick
1997). The simplest way for DA to influence transmitter release is
through modulation of ICa, and we have tested this in our
study.
Dopamine modulation of presynaptic ICa
Many previous studies have documented modulation, by DA and other
neuromodulators, of Ca2+ entry into presynaptic
terminals that may alter transmitter or hormone release (for example, see
Dunlap and Ikeda 1998;
Hernandez-Lopez et al. 1997;
Koga and Momiyama 2000;
Nussinovitch and Kleinhaus
1992). We limited our study to the effects of DA on the peak
ICa because this appeared less affected by space clamp
limitations and incomplete K+ channel block.
DA affected the peak ICa in all the pyloric neurons.
The DA enhancement of ICa in LP and PY neurons is
consistent with our hypothesis that DA enhances transmitter release from these
neurons (Johnson and Harris-Warrick
1997). In one outlier experiment, DA reversibly decreased peak
ICa in a PY neuron. Preliminary multiphoton microscopy
Ca2+ imaging results have suggested that in a single PY
neuron, DA can enhance Ca2+ entry into some presynaptic
varicosities, while reducing Ca2+ entry into others
(Kloppenburg et al. 2000b).
Because pyloric neurons have multiple sites of synaptic contact onto their
postsynaptic targets (King
1976), DA-enhancement of ICa and transmitter
release from PY neurons may arise from a net synaptic response dominated by
enhanced Ca2+ entry into the majority of terminals.
Occasionally, the balance may shift, leaving a net reduction in
Ca2+ entry and creating some variability in modulatory
effects on synaptic transmission (Johnson
et al. 1994).
The DA reduction of ICa in PD neurons is also
consistent with DA targeting ICa to reduce, and sometimes
completely abolish, transmitter release from these neurons (Johnson and
Harris-Warrick 1990,
1997). These voltage-clamp
results are consistent with our recent Ca2+-imaging
studies (Kloppenburg et al.
2000a), which showed that DA reduces depolarization-activated
Ca2+ entry into fine varicosities of PD neurons. DA also
markedly weakened VD synapses (Johnson et
al. 1995) while evoking only a very small (though significant)
reduction in ICa. These pronounced inhibitory synaptic
effects are very dramatic, given the rather modest reductions in
ICa; perhaps DA has additional effects at these synapses,
for example to affect the release machinery directly in addition to modulating
ICa in PD and VD terminals (see following text).
DA has only modest effects on the IC neuron, but they appear to be
contradictory: DA weakly enhances ICa while the IC
neuron's single synapse, onto the VD neuron, is slightly weakened. However,
the postsynaptic response of the VD neuron to the IC's inhibitory transmitter,
glutamate, is significantly reduced by DA
(Johnson and Harris-Warrick
1997). Thus at this synapse, the postsynaptic reduction in
glutamate responsiveness appears to outweigh the small expected increase in
transmitter release due to DA's weak enhancement of
ICa.
The major contradiction between DA's effects on synaptic strength and on
ICa is with the AB neuron. DA markedly increases AB
synaptic strength (enough to even activate a silent synapse)
(Johnson et al. 1995), but we
show here that DA reduces ICa by 50% in this cell. We
initially suspected that a Ca2+-dependent outward
current that persisted in our blocker saline and was enhanced by DA might mask
a direct DA enhancement of ICa. However, intracellular
injection of BAPTA allowed us to separate the
Ca2+-dependent outward current from
ICa, and indicated that DA was directly reducing
ICa in the AB neuron. Thus enhancement of AB transmitter
release appears to occur concurrently with a reduction in presynaptic
ICa. Mismatches between DA's effects on
ICa and on transmitter/hormone release have been reported
in other preparations. For example, in the retina, DA suppresses rod inputs
and enhances cone inputs to horizontal cells
(Witkovsky and Dearry 1991),
while enhancing ICa in rods and suppressing it in a large
fraction of cone cells (Stella and
Thoreson 2000). DA can also inhibit Ca2+
currents in melanotropes (Keja et al.
1992), without altering hormone secretion
(Mansvelder et al. 2002). This
mismatch between release and ICa, of course, occurs with
other neuromodulators; for example, serotonin enhances neurotransmitter
release at the crayfish neuromuscular junction by modulating
Ih (Beaumont and
Zucker 2000; see also Saitow
and Konishi 2000) without increasing intracellular
Ca2+ levels (Delaney
et al. 1991). Modulation of transmitter release could occur by a
number of mechanisms, including modulation of presynaptic K+,
Cl–, and H currents, transmitter transporters,
Ca2+-induced Ca2+ release from
internal stores, and direct actions to change the activity of proteins driving
release (Beaumont and Zucker
2000; Bouron 2001;
Congar et al. 2002;
Haydon and Trudeau 1998). For
example, the DA enhancement of presynaptic ICa in rods
decreases transmitter release through activation of a
Ca2+-activated Cl– current
(Thoreson et al. 2002). DA
also reduces IA in the AB
(Peck et al. 2001), which is
present in pyloric neuron synaptic terminals
(Baro et al. 2000) and in
general, increases the input resistance of the AB neuron
(Johnson et al. 1993). Perhaps
these factors contribute to the DA-induced enhancement of AB graded
transmitter release. It is also possible that a subset of
Ca2+ channels with different properties controls release
from AB terminals, and these may not be inhibited by DA.
Functional significance of DA-sensitive Ca2+ currents recorded from pyloric neuron cell somata
Our analysis of DA's effects on synaptic transmission assumed that the
Ca2+ current we recorded is representative of the
presynaptic ICa controlling graded chemical transmitter
release, but it also could help to shape bistability and bursting in pyloric
neurons. As discussed in the preceding text, Ca2+ entry
is probably not the main depolarizing force for AB bursting in the presence of
DA. In other pyloric neurons though, active responses and plateau potentials
can be triggered by small depolarizations or hyperpolarizations that bring the
membrane potential within the activation range of the currents we recorded
(Bal et al. 1988;
Russell and Hartline 1982).
Plateau potentials in the gastric DG neuron of the STG are supported by
Ca2+ currents with a similar activation range as the
ones measured here (Zhang and
Harris-Warrick 1995). The threshold for graded synaptic
transmission between pyloric neurons can vary depending on the synaptic pair
and the modulatory conditions, but it is usually around the resting potential
of the presynaptic neuron (Hartline and
Graubard 1992; Johnson and
Harris-Warrick 1990). When pyloric neurons are held at –50
mV, graded transmitter release begins with depolarizations to –45 mV
(Manor et al. 1997). Thus our
ICa could participate in controlling graded chemical
transmission. The congruence of DA's effects on ICa in the
LP, PY, VD, and PD neurons with its effects on release from those neurons also
supports this ICa as contributing to graded transmitter
release. The Ca2+ currents underlying graded chemical
transmission in invertebrates have been most extensively studied in leech
heart interneurons (Angstadt and Calabrese
1991; Ivanov and Calabrese
2000; Lu et al.
1997). These currents resemble LVA-type currents: they have more
hyperpolarized activation ranges than our ICa and
Ni2+ blocks them. The lack of sensitivity of pyloric
ICa to Ni2+ as well as a report of
enhancement of pyloric graded chemical transmission by
Ni2+ (Zirpel et al.
1993) suggest that the currents underlying leech graded
transmission differ from the ICa controlling pyloric
graded chemical transmission. In its activation range and relatively low
sensitivity to nifedipine, the ICa in pyloric neurons more
closely resembles the L-type current mediating graded transmitter release from
cochlear hair cells (Spassova et al.
2001) and vertebrate retinal neurons
(Berntson et al. 2003;
Taylor and Morgans 1998).
Further work will be needed to clarify the functional role of the
ICa recorded from pyloric neuronal somata.
|
DISCLOSURES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
Present address of P. Kloppenburg: Universität zu Köln, Institut für Zoologie/Physiologie, AG Zelluläre Neurophysiologie, Weyertal 119, D-50923 Köln, Germany.
|
FOOTNOTES |
|---|
Address for reprint requests: B. R. Johnson Dept. of Neurobiology and Behavior, S.G. Mudd Hall, Cornell University, Ithaca, NY 14853 (E-mail: BRJ1{at}Cornell.Edu).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
Alaburda A, Perrier J-F, and Hounsgaard J. Mechanisms causing plateau potentials in spinal motoneurons. Adv Exp Med Biol 508: 219–226, 2002.[ISI][Medline]
Anderson WW and Barker DL. Synaptic mechanisms that generate network oscillations in the absence of discrete post-synaptic potentials. J Exp Zool 216: 187–191, 1981.[ISI][Medline]
Angstadt JD and Calabrese RL. Calcium currents and graded synaptic transmission between heart interneurons of the leech. J Neurosci 11: 746–759, 1991.[Abstract]
Armstrong CM and Bezanilla F. Charge movement associated with the opening and closing
of the activation gates of the Na channels. J Gen
Physiol 63:
533–552, 1974.
Avery RB and
Johnston D. Multiple channels types contribute to the
low-voltage-activated calcium current in hippocampal CA3 pyramidal neurons.
J Neurosci 16:
5567–5582, 1996.
Ayali A and
Harris-Warrick RM. Monoamine control of the pacemaker kernel and cycle
frequency in the lobster pyloric network. J Neurosci
19: 6712–6722,
1999.
Ayali A,
Johnson BR, and Harris-Warrick RM. Dopamine modulates graded and
spike-evoked synaptic inhibition independently at single synapses in pyloric
network of lobster. J Neurophysiol
79: 2063–2069,
1998.
Bal T, Nagy F, and Moulins M. The pyloric central pattern generator in Crustacea: a set of conditional neuronal oscillators. J Comp Physiol [A] 163: 715–727, 1988.
Baro DJ, Ayali
A, French L, Scholz NL, Labenia J, Lanning CC, Graubard K, and
Harris-Warrick RM. Molecular underpinnings of motor pattern generation:
differential targeting of shal and shaker in the pyloric motor system.
J Neurosci 20:
6619–6630, 2000.
Beaumont V and Zucker RS. Enhancement of synaptic transmission by cyclic AMP modulation of presynaptic Ih channels. Nat Neurosci 3: 133–141, 2000.[ISI][Medline]
Berntson A, Taylor WR, and Morgans CW. Molecular identity, synaptic localization, and physiology of calcium channels in retinal bipolar cells. J Neurosci Res 71: 146–151, 2003.[ISI][Medline]
Bouron A. Modulation of spontaneous quantal release of neurotransmitters in the hippocampus. Prog Neurobiol 63: 613–635, 2001.[ISI][Medline]
Carlin KP, Jones KE, Jiang Z, Jordan LM, and Brownstone RM. Dendritic L-type calcium currents in mouse spinal motoneurons: implications for bistability. Eur J Neurosci 12: 1635–1646, 2000.[ISI][Medline]
Charlton MP and Augustine GJ. Classification of presynaptic calcium channels at the squid giant synapse: neither T-, L-nor N type. Brain Res 525: 133–139, 1990.[ISI][Medline]
Congar P, Bergevin A, and Trudeau L-E. D2 receptors inhibit the secretory proce
) and the peak outward current
(at +20 mV, 
) before (control), during (DA), and after (wash) DA
application. *, statistically significant change from the mean control value
(P < 0.05; n = 4). C: typical experiment showing
responses to 5-mV incrementing depolarizing voltage steps from –50 to
+40 mV in control (top) and DA (bottom) conditions after
block of ICa with 600 µM Cd2+.
D: summary of the effects of DA on the outward current remaining
after block of ICa with 600 µM
Cd2+. The mean peak outward current (at +20 mV;
±SE; n = 3) is shown before (control +
Cd2+) and during (DA+ Cd2+) DA
application.
) with DA application (DA, 
) with DA application (Nif + DA,