| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biological Sciences, University at Albany, State University of New York, Albany, New York 12222
Submitted 24 April 2003; accepted in final form 18 October 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
; Weiss et al. 1986). It contains a uniquely identifiable pair of synaptically connected neurons, C2 and metacerebral cell (MCC), that contribute to arousal, increased excitability in the circuit, and possibly the NO-dependent memory of inedible food (Katzoff et al. 2002). The presynaptic neuron, C2, contains nitric oxide synthase (NOS) and uses the NO produced as an orthograde cotransmitter (Jacklet 1995), with histamine (HA), in excitatory synaptic transmission. We studied the postsynaptic neuron, MCC. It releases serotonin, which modulates the feeding circuit neurons and synapses to feeding muscles (Weiss et al. 1986). MCC also contains guanylyl cyclase (GC), which generates cGMP when stimulated by NO (Koh and Jacklet 1999). NO decreases the membrane potential, increases membrane resistance, enhances excitability, and induces tonic spike activity in MCC (Jacklet 1995; Jacklet and Koh 2001).
The monosynaptic very slow excitatory postsynaptic potential (EPSP) produced in MCC by stimulating C2 is mimicked by bath-applied HA (Weiss et al. 1986), i.e., the membrane depolarizes and the membrane resistance increases. This was done before it was shown that NO was a C2 cotransmitter (Koh and Jacklet 1999). When Weiss et al. measured the ionic currents induced in MCC by HA, they found increased inward current at voltages more positive than –75 mV and believed it was due to the reduction of a potassium current.
To determine the membrane ionic mechanisms that contribute to the depolarization and increased excitability in MCC induced by NO and HA, we recorded membrane currents under voltage clamp in MCC either in situ in the cerebral ganglion or isolated in cell culture. We found that both NO and HA reduce background currents but the currents reduced by each agent are different and are mediated by different second messenger pathways. A preliminary report of this work has appeared in abstract (Jacklet and Tieman 2001).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Aplysia californica (100–150 g) were supplied by the Aplysia Resource Center (University of Miami, FL), kept in an aquarium at 16–18°C, and fed fresh Gracilaria. S-nitroso-N-acetyl-D,L-penicillamine (SNAP) from World Precision Instruments was made up daily as stock (10 mM) in artificial sea water (ASW) and diluted for use. Aliquots of 10 mM HA (Sigma, H-7250) and 5 mM 8-bromoguanosine-3'5'-cyclic monophosphate (8-Br-cGMP; Sigma, B-1381) were dissolved in water and frozen at –20°C for later use. Nifedipine (N7634), cobalt chloride (C3169), and tetraethylammonium (TEA)-Cl (T2265) were obtained from Sigma Chemical.
Neuron culture
We used standard neuron culture techniques for Aplysia neurons (Jacklet and Koh 2001; Kleinfeld et al. 1990; Schacher and Proshansky 1983). Isolated cerebral ganglia were incubated in 1% protease (type IX, Sigma, P-6141) in 2 ml ASW (in mM: 460 NaCl, 10 KCl, 10 CaCl2, 48 MgCl2, and 10 HEPES, pH 7.8) for 1 h at 35°C before desheathing at room temperature. Visually identified single neurons were routinely removed from the ganglion with sharp micropipettes controlled by a micromanipulator and transferred to culture dishes. Culture medium (Kleinfeld et al. 1990) consisted of an isotonic ASW (see following text) and L-15 powder mix without glutamine and organic salts (No. 82–5154EA; Gibco, Grand Island, NY). Neurons were plated on poly- L-lysine- (Sigma, >500,000 MW, p-1524) coated Corning 25000 culture dishes. Plating medium consisted of 50% modified L-15 medium and 50% filtered (0.45 µm) Aplysia hemolymph. Hemolymph in the medium (Schacher and Proshansky 1983) induced prompt attachment to the substrate and neurite outgrowth.
Electrophysiology
Cerebral ganglia were treated with Sigma protease Type IX, as above. The sheath then was removed to expose the neurons. For recording from MCC in the ganglion, a desheathed ganglion was pinned down on a Sylgard-coated dish (2.5 cm in diam) and viewed with an upright dissecting microscope. The MCC axon was cut 300 µm from the cell body to improve membrane space clamping. For cultured neurons, the culture dish containing the isolated MCC was placed on the viewing stage. The culture medium was washed from the chamber using the superfusion system and replaced with standard ASW, with several washes over the course of an hour before recording was started.
Preparations were superfused with ASW containing varying concentrations of SNAP, 8-Br-cGMP, or HA and recordings were made from MCC at room temperature using two electrodes in current clamp or voltage clamp. ASW was superfused with a gravity feed system at a set flow rate (2–3 ml/min) and removed from the dish by aspiration using a suction pump. Solution changes were made using a flow selection manifold. Measurements of NO in a culture dish containing an MCC using a WPI ISO-NO electrode showed that 10 µM SNAP was approximately equivalent to 500 nM NO. Low-resistance electrodes (Re 5–8 MS) pulled from thin-wall glass on a Sutter Instruments Model P-87 Micropipette Puller and filled with 3 M KAc/1 M KCl were used. Recordings were made with an Axon Instruments, GeneClamp 500B amplifier, in either current- or voltage-clamp mode. In voltage-clamp mode the gain was 10 K and the stability was between 800 and 1200 µs. Signals were digitized using Axon Instruments pClamp/Digadata 1200–2 system. Analysis was performed by Axon Instruments software, Clampex and Clampfit. In current-clamp mode the resting membrane potential, action potential characteristics, and membrane resistance were measured. Neurons were then voltage clamped near the resting potential (usually –60 mV) and voltage-clamp protocols were employed to generate current-voltage (I-V) curves. Voltage steps from –100 to –20 mV, 500 to 1000 ms in duration were used. Steady-state currents were measured at 450 or 950 ms. Experiments were performed if a neuron met this criterion: –55 to –65 mV membrane potential, >80 mV action potential, and +2 to –1 nA holding current at –60 mV holding potential. Data from 10 ganglion preparations and 40 isolated neuron preparations were included in our analysis. Differences between experimental and control currents in I-V plots were determined for each voltage step and statistical significance was determined using a two-tailed t-test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
MCC was studied in situ in the isolated cerebral ganglion after the ganglion was desheathed and the axon was cut. The soma was impaled with two microelectrodes for current- and voltage-clamp recordings. When the ganglion was bathed in normal ASW, MCC was quiescent with a resting membrane potential near –60 mV (–61.2 ± 1.09 mV, mean ± SE, n = 10). Action potentials 80–90 mV in amplitude were evoked by 1-nA depolarizing current pulses (Fig. 1A). Moderate ongoing synaptic input was observed in some preparations but usually the synaptic potentials were <1 Hz and 1 mV (see the voltage trace rectangle and inset in Fig. 1A).
|
Steady-state I-V curves obtained in voltage clamp are shown in Fig. 1, D–F. In ASW the curves show steeply increasing outward current at –30 mV, a flattened curve in the –40 to –70 mV range, and steeply increasing inward current at less than –70 mV. HA induced an inward shift in current in the range –30 to –70 mV (Fig. 1D), without current reversal. These curves are similar to those obtained for HA by Weiss et al. (1986). SNAP induced a similar response (Fig. 1E) in the –30 to –70 mV range, but the curves crossed near the expected potassium equilibrium potential of –80 mV. When MCC was treated with both SNAP and HA (Fig. 1F), the HA response added to the SNAP response and the reversal potential was maintained. These differences in responses and the additive nature of the SNAP and HA responses suggest that the SNAP response and the HA response are mediated by separate pathways and ion channels. The addition of HA and SNAP responses was tested further in cultured MCCs (see Fig. 4).
|
MCC responses to SNAP, HA, and 8-Br-cGMP in cell culture
Within 12 h after an MCC was isolated in culture, its axon was securely stuck to the bottom of the culture dish and fine neurite sprouts extended from the cut end of the axon (Fig. 2C, inset). MCCs were examined on day 1 to day 3, with no systematic differences in response. Resting potentials in ASW were about –60 mV (–60.6 ± 0.73 mV, mean ± SE, n = 26) and action potentials of 80–90 mV were evoked by depolarizing current pulses. Hyperpolarizing current pulses evoked potential responses that diminished with hyperpolarization (Fig. 2A), indicative of inward rectification, as expected from MCC recording made in the ganglion (Weiss et al. 1986; Koh and Jacklet 1999). These results show that MCC retains its normal resting membrane and action potential properties after being isolated from the ganglion in cell culture.
|
). These responses are essentially the same as those seen when MCC was in the ganglion (Fig. 1). Threshold concentration for responses to SNAP and HA were 5–10 µM (as in the ganglion) and most experiments were performed with 20 µM. A consistent effect of SNAP, made up within 2 h of use, in all experiments was an inward shift of the holding current at the holding potential of –60 mV. The average holding current was –0.02 ± 0.24 nA, mean ± SE, n = 10 in ASW and –1.32 ± 0.25 nA, n = 10 in 20 µM SNAP. This difference was statistically significant (2-tailed t-test, n = 10, t = 3.6, P < 0.01). The SNAP-induced inward shift in holding current is the result of a decrease in outward current due to a reduction in membrane conductance. Another consistent response to SNAP was the inward shift in steady-state current at potentials between –70 and –30 mV. Data were obtained from 6 MCCs using full I-V curves from –100 to –30 mV (Fig. 3A). Shifts at potentials between –70 and –35 mV for all 6 preparation were statistically significant using a two-tailed t-test (see Fig. 3 legend). Degassed SNAP, tested a day after the SNAP was made up (Fig. 3C), did not alter the response recorded in ASW.
|
Responses to 20 µM SNAP were obtained from seven other MCCs using shorter (500-ms) voltage steps and an abbreviated voltage step protocol (steps to –90, –80, –40, –35, and –30 mV from Vh –60 mV). Inward shifts in steady-state currents at –40, –35, and –30 mV for all seven MCCs were consistent with the shifts obtained from the 6 MCCs obtained using a full protocol (Fig. 3A). The mean shifts were –2.01 ± 0.25 nA at –40 mV, –2.56 ± 0.41 nA at –35 mV, and –3.51 ± 0.7 nA at –30 mV (mean ± SE). The shifts at these potentials were all statistically significant using a two-tailed t-test, t = 7.99, t = 6.22, and t = 4.99, respectively, all P < 0.01. I-V curves from 3 of these showed current reversal consistent with the potassium equilibrium potential similar to those in Fig. 3B, and 4 did not. Thus results from 13 MCCs showed a consistent inward shift in current between –70 and –30 mV, and about half of them (n = 6) showed current reversal near the potassium equilibrium potential. The abbreviated protocol does not allow an exact determination of the reversal voltages, but it appears to be about –80 mV for the average curves of the 3 that showed current reversal using the abbreviated protocol.
HA responses were rapid and very consistent. Average steady-state current responses to 20 µM HA in Fig. 3D show increased inward current at potentials more positive than –65 mV and largest inward shifts at –40 mV. All of the shifts at voltages between –50 and –30 mV were statistically significant (P < 0.05) using a two-tailed t-test (see legend Fig. 3D). Six other MCCs were tested with 20 µM HA using an abbreviated voltage step protocol (steps to –90, –80, –40, –35, and –30 mV from Vh –60 mV). The mean shifts were –1.59 ± 0.40 nA at –40 mV, –2.41 ± 0.44 nA at –35 mV, and –2.60 ± 0.28 nA at –30 mV (mean ± SE). The shifts at these potentials were all statistically significant using a two-tailed t-test, t = 4.40, t = 6.07, and t = 10.10, respectively, all P < 0.01. The average holding current at –60 mV was slightly shifted inward in 20 µM HA. The average holding current was 0.1 ± 0.51 nA, n = 8, in ASW and –0.3 ± 0.41 nA, n = 8, in 20 µM HA (mean ± SE). This difference was not statistically significant (2-tailed t-test, t = 2.0, P < 0.10). There was slight enhancement of transient inward currents at potentials more positive than –40 mV (not shown).
The results of treating MCCs in the ganglion with SNAP and HA together appeared to be additive (Fig. 1F). To test this further MCCs in culture were treated with 20 µM HA alone, 20 µM SNAP alone, and then with 20 µM SNAP and 20 µM HA together. An example is shown in Fig. 4A and the average (n = 4) difference currents for HA alone, SNAP alone, and combined HA and SNAP are shown in Fig. 4B. As expected from previous results (Fig. 3D), HA alone responses were minimal at –60 mV and increased progressively at –50, –40, and –30 mV. SNAP alone responses covered the range from –70 to –30 mV, as expected from the results shown in Fig. 3A. The responses of HA and SNAP were essentially additive (Fig. 4B). In two experiments HA was tested and then HA and SNAP together, in two others the sequence was SNAP and then SNAP plus HA. There was no obvious difference in the response to combined HA and SNAP that depended on the prior exposure to either HA or SNAP. Ratios of the HA and SNAP together currents divided by the SNAP plus the HA currents were calculated for each potential step (see Fig. 4 legend) and found to be very close to 1.0. The ratios were 1.13 at –70 mV, 1.19 at –60 mV, 1.17 at –50 mV, 0.84 at –40 mV, and 0.86 at –30 mV. These results confirm that the HA effects and the SNAP effects are additive and, combined with the results below using cobalt, suggest that there are separate HA and SNAP mechanisms.
The membrane-permeant cGMP analogue 8-Br-cGMP produced shifts in steady-state inward current similar to the SNAP responses that lacked current reversal. 8-br-cGMP, like SNAP, had only a slight effect on early transient current above –40 mV (not shown). Treatment with 40 µM 8-Br-cGMP produced a small response and treatment of the same neuron, after recovery, with 100 µM produced a larger response as shown in Fig. 5A. This neuron's response to 20 µM SNAP (Fig. 5A) also lacked a reversal potential. Full 8-Br-cGMP responses took 10 min or more to develop completely, presumably due to the time required to penetrate the neuron. The average (n = 3) I-V curves of other MCCs in response to 40 µM 8-Br-cGMP did not have a reversal potential (Fig. 5B) and were similar to the SNAP responses (Fig. 3A). The inward shifts in current at –40 and –35 mV were statistically significant (P < 0.05, see Fig. 5 legend). The holding current shifted inward, but the difference (0.56 ± 0.25 nA) was not statistically significant (2-tailed t-test, P < 0.10).
|
Several inorganic calcium and potassium channel blockers (Hille 2001) were tested for their ability to block the SNAP and HA responses. Cobalt was used to block calcium and calcium-dependent currents. It has been used widely to block calcium currents in molluscan neurons (e.g., Walsh and Byrne 1989). Cobalt was preferred because it was effective and it washed out rapidly after a treatment, whereas cadmium did not. In ASW with added 10 or 15 mM cobalt chloride, currents were reduced at several levels (Fig. 6A). The early inward current at –40 mV was reduced and the outward current enhanced. At –80 mV the steady-state current was unchanged, but at –90 mV the inward current was reduced. This calcium-dependent current appears to be a potassium current because the current is unchanged near the potassium equilibrium potential (–80 mV). Inward transient currents were blocked by cobalt at voltages more positive than –40 mV. Difference currents (Fig. 6B) reveal the substantial transient inward currents that were blocked by cobalt at –35 and –30 mV. The changes in steady-state currents produced by cobalt, shown in the I-V curves (Fig. 6C), include reductions of outward currents at –30 and –35 mV and a statistically significant reduction in inward current at –90 mV. There was an inward shift in holding current of 0.9 nA at –60 mV, but it was not statistically significant. These data were obtained with an abbreviated voltage-step protocol, so complete I-V curves are not available. Cobalt did block the HA response (Fig. 6D), suggesting that the HA response is entirely calcium dependent. The HA response appears to be largely a reduction in calcium-activated potassium currents that are normally active at the resting potential and more positive voltages.
|
We tested several well-known potassium channel blockers. Both barium (2–5 mM, n = 3) and cesium (5–10 mM, n = 5) reduced the inward current at –90 mV severely and reduced the outward currents slightly at –30 and –35 mV. Either barium or cesium produced a virtually linear I-V relationship between –90 and –50 mV as shown for cesium (Fig. 6F). They did not block the consistent SNAP effect in the –70 to –30 mV range (Fig. 6F). The consistent reduction of the inward current at –90 mV by barium and cesium indicates that the current is a persistent potassium current, and the blocking effect of cobalt suggests that it is a calcium-dependent potassium current. Since HA appears to block a calcium-activated potassium current we tested the possibility that nifedipine and TEA might block the HA response.
Nifedipine, a Tsien L-type calcium channel blocker (Hille 2001), reduced the early transient inward current at –40 to –30 mV at concentrations of 10–30 µM, similar to the cobalt effect shown in Fig. 6B, and produced a large decrease in steady-state outward current in the voltage range –40 to –30 mV (Fig. 7A). Nifedipine also consistently reduced the inward current at –90 mV and thereby produced a small reversal of current near the potassium equilibrium potential. Both cobalt and nifedipine treatments should block calcium currents (albeit the specific L-type for nifedipine) and indirectly reduce the calcium-activated potassium currents. Although cobalt blocked the HA response, nifedipine did not (Fig. 7B). Average difference current shown in Fig. 7B indicates that the nifedipine alone responses and HA alone responses are additive, and therefore nifedipine does not block the HA response. Ratios of the responses to nifedipine and HA applied together divided by the nifedipine alone plus the HA alone responses were 1.00 at –60 mV, 0.89 at –50 mV, 0.97 at –40 mV, and 1.09 at –30 mV. This result suggests that the calcium-dependent potassium current involved in the HA response is not dependent on a nifedipine-sensitive calcium influx. Also, nifedipine did not block the SNAP response (n = 2, not shown).
|
), was tested at 2–10 mM alone or in combination with other agents. Its effect alone on steady-state currents was similar to, but about one-half, the effect of nifedipine. The difference currents shown in Fig. 7C for 5 mM TEA alone and 20 µM HA alone indicate that the current differences for the combined treatment are essentially the sum of the TEA alone treatment plus the HA alone treatment. Ratios of the currents (responses to combined TEA and HA divided by TEA alone plus HA alone) were 0.90 at –60 mV, 1.08 at –50 mV, 1.10 at –40 mV, and 0.72 for –30 mV. These results suggest that most of the inward current induced by HA near the resting potential is not sensitive to TEA but a small part of the current at –30 mV is. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
). Their effects cannot be separated by blocking calcium influx in C2. The very slow EPSP is completely blocked by low-calcium— high-magnesium solutions (McCamen and Weinreich 1985; Weiss et al. 1986). This is expected because HA release from synaptic vesicles is calcium dependent and NO is produced by the calcium-dependent enzyme NOS.
Both NO and HA depolarize MCC and increase its excitability by decreasing the resting membrane conductance (Koh and Jacklet 1999). The results of the present study show that the decrease in conductance results in an inward shift of background current. NO decreases a different background current than HA does and the second messenger pathways involved are different. NO responses are more complex and variable than HA responses.
HA consistently reduced the background outward currents at and above the resting potential of –60 mV in MCC. A small inward shift in holding current occurred and outward currents were reduced at potentials more positive than the resting potential. The HA response was blocked by cobalt, a calcium channel blocker, and partially blocked by TEA, which blocks some calcium-activated potassium channels (Walsh and Byrne 1989), at –30 mV. Thus the currents affected are calcium and calcium-dependent potassium currents. It appears that HA decreases a persistent calcium-dependent potassium current that is active near the resting potential and at more positive potentials, but not at potentials more negative than the potassium equilibrium potential. The steady-state I-V curve during HA treatment did not reverse at the potassium equilibrium potential, which is reported to be about –80 mV (Weiss et al. 1986). The lack of a reversal potential at the potassium equilibrium potential is the expected result, if the current is not active at potentials more negative than the potassium equilibrium. Weiss et al. (1986) in their studies of MCC in the ganglion also did not find a reversal potential in response to exogenous HA. They found that the HA response became null as the membrane potential approached the equilibrium potential, as we have found. Attempts by them to demonstrate a reversal potential in high-potassium conditions produced weak reversal at best. However, the HA null response shifted in agreement with the calculated shift in the potassium equilibrium potential during high-potassium treatment.
Our experiments with cobalt (Fig. 6, A and B) and nifedipine, the L-type calcium channel blocker, show that there are large transient calcium currents at voltages more positive than –40 mV. There are also persistent calcium-dependent potassium currents below the resting potential, because both cobalt treatment (Fig. 6C) and nifedipine (Fig. 7A) reduced inward currents below the potassium equilibrium potential and cause a reversal of current near the potassium equilibrium potential. There are also persistent calcium-dependent potassium currents above the resting potential, as shown by the nifedipine results (Fig. 7, A and B) and the TEA results (Fig. 7C). A persistent calcium-dependent potassium current at depolarized potentials is not unusual in Aplysia neurons (e.g., Walsh and Byrne 1989), but one active at potentials more negative than the potassium equilibrium potential is unusual. Persistent calcium currents, and possibly calcium release from intracellular stores, play a role in the HA and SNAP responses.
Calcium-activated potassium currents are well known in other Aplysia neurons. Kehoe (1985) found that synaptic input to pleural ganglion neurons blocked calcium-activated potassium currents. Calcium injection activated two potassium currents, one was sensitive to block by TEA and another slower one was not. Walsh and Byrne (1989) found that serotonin (5-HT) and cAMP decreased a TEA-sensitive calcium-activated potassium current in pleural sensory neurons that were tested at select holding potentials between –25 and –38 mV. Injected calcium activated a current with a reversal potential of –65 mV. They concluded that there may be two types of calcium-activated currents, one is a steady-state, TEA-insensitive current that may contribute to the resting membrane potential and another is a depolarization-activated, TEA-sensitive one. Their finding of two distinct currents agrees with Kehoe's report on pleural neurons. Critz et al. (1991) studied a TEA- and cobalt-sensitive current that activated near –10 mV and was reduced by 5-HT and PheMetArgPhe-amide. This appeared to be the fast calcium-dependent potassium current. A slow current, similar to the one identified by Walsh and Byrne (1989), may be the current that is reduced by HA in MCC. It is relatively insensitive to TEA (Fig. 7C) and it is active at –60 mV, which is somewhat lower than the –38 mV activation voltage for the slow current in sensory neurons. However, current activation levels in MCC may be scaled lower than in sensory neurons, since the resting potential of sensory neurons (Walsh and Byrne 1989) is more positive (–43 to –48 mV) than the resting potential (–60 to –65 mV) of MCC.
MCC's response to HA resembles the prolonged depolarization of supraoptic neurons induced by HA blockade of a potassium current, mediated by a G protein (Li and Hatton 1996). Further studies of the HA-induced response in MCC are needed to clarify the type of HA receptor and second messengers that are involved and their similarity to known HA receptor mechanisms.
Each MCC treated with SNAP responded with an inward shift in holding current, associated with a decrease in membrane conductance, at the holding potential of –60 mV, the normal resting potential. Each MCC consistently showed an inward shift in currents between –70 and –30 mV in I-V plots (Fig. 3A). Previously it was shown that the NO donor SNC decreased the membrane conductance and depolarized the membrane potential of MCC (Jacklet 1995; Koh and Jacklet 1999). Taken together these results suggest that a background potassium current that is active in the –70 to –30 mV range is decreased by the NO donors and this decrease causes the depolarization and changes in excitability near the resting potential.
About one-half of the MCCs examined showed a current reversal near the potassium equilibrium potential in I-V plots in response to SNAP (Fig. 3B). A reduction in potassium currents that is active at potentials more negative than the potassium equilibrium potential is expected to lead to rotation of the current curve around the potassium equilibrium potential. The variability of this part of the SNAP response below –80 mV indicates that this component of the response may be subject to state- and time-dependent changes. Our experiments with cobalt suggest that the reversal component may depend on calcium levels in the neuron.
When calcium currents were blocked with cobalt, SNAP induced an inward current that diminished with hyperpolarization but did not reverse (Fig. 6E), suggesting that the current reversal normally involves the reduction of a persistent calcium-dependent potassium current that is blocked by cobalt. The cobalt results also suggest that a persistent inward current with a positive equilibrium potential, likely sodium, is enhanced, because inward currents persist at potentials at and below the potassium equilibrium potential. In cobalt, the SNAP-induced currents at voltages more positive than the potassium equilibrium potential are larger than they are at more negative potentials. This can be explained by a reduction in a persistent background potassium current that is not calcium sensitive. Therefore it appears that the SNAP response is shaped by a combination of changes in currents: a consistent decrease of a background potassium current and increase of a persistent sodium current and a variable decrease in a calcium-dependent potassium current. The latter's contribution seems to vary according to conditions that are not clear. State-dependent changes in the calcium-activated potassium currents may account for most of the variation.
Changes in calcium-activated potassium currents that last for hours are known in Aplysia. A long-lasting refractory period (
18 h) induced by 30 min of spike activity (afterdischarge) occurs in bag cells (Kaczmarek and Kauer 1983). During this time the calcium-dependent potassium current of the BK channel is doubled (Zhang et al. 2002). Although calcium is released from intracellular stores in these neurons, calcium entry through a nonselective cation channel appears to be responsible for initiating the potassium current. Thus it is conceivable that changes in intracellular calcium may contribute to the variation in SNAP responses in MCC.
Calcium currents sensitive to the L-type blocker nifedipine are activated by 5-HT and protein kinase C (PKC) in sensory neurons of Aplysia (Braha et al. 1993) during spike broadening. At a holding potential of –50 mV, 5 µM nifedipine blocks about 0.5 nA of inward current. In our study, the nifedipine-sensitive, steady-state, outward current in MCC was reduced at potentials between –50 and –30 mV (Fig. 7, A and B), suggesting that a persistent, small calcium current is active at those potentials and could stimulate a calcium-dependent potassium current.
Cyclic GMP production is enhanced in MCC by NO treatment and blocked by the GC inhibitor 1H-[1,2,4]oxadiazolo-[4,3a]quinoxaline-1-one (ODQ) (Koh and Jacklet 1999). Also, the membrane resistance increased and depolarization occurred when MCC was treated with the NO donor, SNC, and 20 µM ODQ blocked these NO effects. 8-Br-cGMP also increased the membrane resistance and depolarized MCC. The threshold was 5 µM and the responses leveled off at 40 µM 8-Br-cGMP. Thus the evidence is strong that the NO-induced depolarization and increase in resistance at the resting membrane potential is mediated by NO stimulation of GC and cGMP production. The voltage clamp data from the present study (Fig. 5) support this conclusion. 8-Br-cGMP treatment resembles the SNAP effect, but it is not identical to it. Specifically, the increased inward current near the resting potential (–65 to –30 mV) is present, but neither the current reversal nor the persistent inward current at potentials more negative than the potassium equilibrium potential are present. They may be induced by S-nitrosylation, since NO may act by direct membrane protein S-nitrosylation (Stamler et al. 1992). For example, NO activates calcium-dependent potassium (BK) channels in posterior pituitary nerve terminals, independent of the GC–cGMP pathway (Ahern et al. 1999). S-Nitrosylation frequently is found to enhance persistent sodium currents (Ahern et al. 2002). Its contribution to the NO response in MCC needs to be tested.
The background current that is decreased by the NO–GC–cGMP pathway in MCC is not likely to be the S channel potassium current that is decreased by activation of the cAMP—protein kinase A pathway and augmented by PKC in Aplysia sensory neurons (Sugita et al. 1994; Byrne and Kandel 1996). We have tested cAMP in voltage-clamp experiments on cultured MCCs (Jacklet et al. 2003) and found that it does not mimic the effects of SNAP, but does dramatically increase potassium currents below –75 mV. However, the background current blocked by NO in MCC does have properties similar to the S channel. It is active at the resting potential, lacks inactivation, and is relatively insensitive to most potassium channel blockers. These characteristics also fit the background potassium current carried by two-pore domain KCNK channels (Goldstein et al. 2001).
NO affects a variety of membrane currents in different neuronal and muscular systems (for review see Jacklet 1997). Many effects similar to the ones we have observed in MCC are mediated by activation of the GC–cGMP pathway and result in an increase in excitability. For example, in neuron B7nor of the mollusc Lymnaea (Park et al. 1998), NO reduced a potassium conductance and increased the neurons' excitability. An NO-induced increase in excitability, similar to the MCC increase, was also found in type I hair cells of rat (Chen and Eatock 2000). A voltage-dependent potassium current that was active at the resting potential was blocked by NO. This increased the membrane resistance, depolarized the cell, and thus augmented the receptor potential.
MCC and its presynaptic neuron C2, which uses NO and HA as neurotransmitters, participate in the neural circuits that mediate feeding in Aplysia. The neural circuits are powerfully modulated by 5-HT released from MCC (Morgan et al. 2000; Weiss et al. 1986) resulting in an aroused behavioral feeding state. C2 is a mouth mechanosensory neuron that is activated during feeding, and it excites MCC by its synaptic input. Our results show that the depolarization and increased excitability in MCC caused by the release of NO and HA from C2 are due to the activation of separate pathways, since, the effects of HA and NO are additive, HA responses are calcium dependent, and NO responses are not, and ODQ blocks the NO response but not the HA response (Koh and Jacklet 1999). Activation of these pathways by NO and HA reduces the background currents and persistent calcium-activated potassium currents and increases the excitability of MCC. HA and especially NO are capable of producing state- and time-dependent responses that make important contributions to dynamic changes in the circuit during feeding and changes required for learning and NO-dependent memory of feeding.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
GRANTS
This work was supported by National Institute of Mental Health Grant MH-57746 to J. W. Jacklet.
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: J. W. Jacklet (E-mail: jwj74{at}albany.edu).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Ahern GP, Klyachko VA, and Jackson MB. CGMP and S-nitrosylation: two routes for modulation of neuronal excitability by NO. Trends Neurosci 25: 510–517, 2002.[CrossRef][ISI][Medline]
Braha O, Edmonds B, Sacktor T, Kandel ER, and Klein M. The contribution of protein kinase A and protein kinase C to the actions of 5-HT and the L-type Ca2+ current of the sensory neurons of Aplysia. J Neurosci 13: 1839–1851, 1993.[Abstract]
Byrne JH and Kandel ER. Presynaptic facilitation revisited: state and time dependence. J Neurosci 16: 425–435, 1996.
Chen JW and Eatock RA. Major potassium conductance in type I hair cells from rat semicircular canals: characterization and modulation by nitric oxide. J Neurophysiol 84: 139–151, 2000.
Critz SD, Baxter DA, and Byrne, JH. Modulatory effects of serotonin, FMRFamide and myomodulin on the duration of action potentials, excitability, and membrane currents in tail sensory neurons of Aplysia. J Neurophysiol 66: 1912–1926, 1991.
Goldstein SA, Bockenhauer D, O'Kelly I, and Zilberberg N. Potassium leak channels and the KCNK family of two-P-domain subunits. Nature Rev Neurosci 2: 175–184, 2001.[ISI][Medline]
Hille B. Ion Channels of Excitable Membranes (3rd ed.). Sunderland, MA: Sinauer, 2001, p. 814.
Jacklet JW. Nitric oxide is used as an orthograde cotransmitter at identified histaminergic synapses. J Neurophysiol 74: 891–895, 1995.
Jacklet JW. Nitric oxide signaling in invertebrates. Invertebr Neurosci 3: 1–14, 1997.[CrossRef][Medline]
Jacklet JW, Grizzaffi J, and Tieman DG. Serotonin, nitric oxide and histamine increase excitability in neuron MCC of Aplysia by diverse mechanisms. Soc Neurosci Abstr 29: 44.3, 2003.
Jacklet JW and Koh H-Y. Nitric oxide as an orthograde cotransmitter at central synapses of Aplysia: responses of isolated neurons in culture. Am Zool 41: 82–291, 2001.
Jacklet JW and Tieman DG. Nitric oxide and histamine decrease potassium conductances in Aplysia neurons C4 and MCC under voltage clamp. Soc Neurosci Abstr 27: 31.6, 2001.
Kaczmarek LK and Kauer JA. Calcium entry causes a prolonged refractory period in peptidergic neurons of Aplysia. J Neurosci 3: 2230–2239, 1983.[Abstract]
Katzoff AB, Ben-Gedalya T, and Susswein AJ. Nitric oxide is necessary for multiple memory after learning that a food is inedible in Aplysia. J Neurosci 22: 9581–9594, 2002.
Kehoe J. Synaptic block of a calcium conductance in Aplysia neurones. J Physiol 369: 439–474, 1985.
Kleinfeld D, Raccuia-Behling F, and Chiel HJ. Circuits constructed from identified Aplysia neurons exhibit multiple patterns of persistent activity. Biophys J 57: 697–715, 1990.
Koh H-Y and Jacklet JW. Nitric oxide stimulates cGMP production and mimics synaptic responses in metacerebral neurons of Aplysia. J Neurosci 19: 3818–3826, 1999.
Li Z and Hatton GI. Histamine-induced prolonged depolarization in rat supraoptic neurons: G-protein-mediated, Ca(2+)-independent suppression of K+ leakage conductance. Neuroscience 70: 145–158, 1996.[CrossRef][ISI][Medline]
McCamen RE and Weinreich D. Histaminergic synaptic transmission in the cerebral ganglion of Aplysia. J Neurophysiol 53: 1016–1037, 1985.
Morgan PT, Perrins R, Lloyd PE, and Weiss KR. Intrinsic and extrinsic modulation of a single central pattern generating circuit. J Neurophysiol 84: 1186–1193, 2000.
Park JH, Straub VA, and O'Shea M. Anterograde signaling by nitric oxide: characterization and in vitro reconstitution of an identified nitrergic synapse. J Neurosci 18: 5463–5476, 1998.
Schacher S and Proshansky E. Neurite regeneration by Aplysia neurons in dissociated cell culture: modulation by Aplysia hemolymph and the presence of the initial axonal segment. J Neurosci 3: 2403–2413, 1983.[Abstract]
Stamler JS, Singel DJ, and Loscalzo J. Biochemistry of nitric oxide and its redox-activated forms. Science 258: 1898–1902, 1992.
Sugita S, Baxter DA, and Byrne JH. Activators of protein kinase C mimic serotonin-induced modulation of a voltage-dependent potassium current in pleural sensory neurons of Aplysia. J Neurophysiol 72: 1240–1249, 1994.
Walsh JP and Byrne JH. Modulation of a steady-state Ca2+-activated, K+ current in tail sensory neurons of Aplysia: role of serotonin and cAMP. J Neurophysiol 61: 32–44, 1989.
Weiss KR, Shapiro E, and Kupfermann I. Modulatory synaptic actions of an identified histaminergic neuron on the serotonergic metacerebral cell of Aplysia. J Neurosci 6: 2393–2402, 1986.[Abstract]
Zhang Y, Magoski NS, and Kaczmarek LK. Prolonged activation of Ca2+-activated K+ current contributes to the long-lasting refractory period of Aplysia bag cell neurons. J Neurosci 22: 10134–10141, 2002.
This article has been cited by other articles:
|
|
 |
|
  D. D'Agostino, E. Mazza Jr., and J. A. Neubauer Heme oxygenase is necessary for the excitatory response of cultured neonatal rat rostral ventrolateral medulla neurons to hypoxia Am J Physiol Regulatory Integrative Comp Physiol, January 1, 2009; 296(1): R102 - R118. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Miller, A. Katzoff, and A. J. Susswein Nitric oxide induces aspects of egg-laying behavior in Aplysia J. Exp. Biol., August 1, 2008; 211(15): 2388 - 2396. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. A. Straub, J. Grant, M. O'Shea, and P. R. Benjamin Modulation of Serotonergic Neurotransmission by Nitric Oxide J Neurophysiol, February 1, 2007; 97(2): 1088 - 1099. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Katzoff, T. Ben-Gedalya, I. Hurwitz, N. Miller, Y. Z. Susswein, and A. J. Susswein Nitric Oxide Signals That Aplysia Have Attempted to Eat, a Necessary Component of Memory Formation After Learning That Food Is Inedible J Neurophysiol, September 1, 2006; 96(3): 1247 - 1257. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. G. Hatcher, L. C. Sudlow, L. L. Moroz, and R. Gillette Nitric Oxide Potentiates cAMP-Gated Cation Current in Feeding Neurons of Pleurobranchaea californica Independent of cAMP and cGMP Signaling Pathways J Neurophysiol, May 1, 2006; 95(5): 3219 - 3227. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||
