| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institute for Nonlinear Science, University of California, San Diego, La Jolla, California
Submitted 8 June 2006; accepted in final form 25 August 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
S and GTP
S injection into identified neurons. The depolarizations and oscillations were accompanied by an increase of intracellular Ca2+ levels and correlated Ca2+ oscillations. By using cyclopiazonic acid, an endoreticular Ca2+ uptake inhibitor, we show that some internal calcium release may augment the response, but is not crucial for its production. Interestingly, although Ca2+ concentration increase is typically associated with the phosphoinositide pathway, in the lobster, the Ca2+ concentration increase—either voltage dependent or independent—cannot account for the observed depolarization. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
The lobster stomatogastric ganglion (STG) is a well-studied model system for examining the effect of neuromodulation on pattern generation. The main neurotransmitters in this system are acetylcholine (ACh) and glutamate (Glu). ACh acts through muscarinic receptors to produce long-term changes in the STG rhythms (Elson and Selverston 1992
; Katz 1998; Nagy and Dickinson 1983) and it has been suggested that Glu has metabotropic actions (Krenz et al. 2000) through metabotropic Glu receptors (mGluRs).
The STG is composed of about 30 neurons, mostly motoneurons, and controls the movements of the lobster foregut (Selverston and Moulins 1987). The rhythmic output of the system depends on both network and cellular properties (Harris-Warrick et al. 1992). Many of the neurons are capable of oscillating in isolation under certain conditions, such as when neuromodulators (Bal et al. 1988; Selverston 1977; Selverston and Miller 1980) are added to their environment or when neurons are isolated in situ but left connected to higher ganglia, which supply the necessary neuromodulators. Examples of this can be observed in gastric system neurons (Fig. 1; for review see Selverston et al. 1998). They are quiescent if isolated from the rest of the nervous system but typically oscillate at a period of about 0.1 Hz when connected to the commissural ganglion or when the circuit is exposed to specific modulators.
|
). Type I mGluRs in mammals are known to act through phospholipase-C (PLC) that in turn produces inositol 1,4,5-trisphosphate (IP3) (Conn and Pin 1997; Hermans and Challiss 2001) and internal Ca2+ level elevation. In lamprey swimming (El Manira et al. 2002; Kettunen et al. 2002; Krieger et al. 2000) and the STG (Krenz et al. 2000), as in many other systems, type I mGluRs have excitatory effects. Specifically, their activation speeds up ongoing rhythms, increases spike frequency and burst duration, and in the lobster can induce transient oscillations in quiescent gastric systems. In other systems, mGluR I agonists have been shown to increase excitation through a variety of conductance changes (Anwyl 1999), which can be roughly divided into conductance increases such as nonspecific-cation currents (Congar et al. 1997; Guerineau et al. 1995; Tempia et al. 2001) and K+ current conductance decreases (Woodhall et al. 1999; Wu and Barish 1999). We found that Glu has a predominantly excitatory effect on the gastric system when ionotropic receptors are blocked. This excitatory effect can be reproduced with two different mGluR type I agonists with overlapping pathways. We used pharmacological agents to assess the contribution of different steps in the PLC second-messenger pathways to the mGluR I response. Ca2+ imaging combined with voltage clamp, in addition to pharmacological agents, indicated that the internal Ca2+ stores' contribution to the response was minor. By injection of BAPTA into identified neurons, we were able to clamp the Ca2+ concentration and show that, although Ca2+ oscillations are produced by mGluR I agonists, the Ca2+ itself is not the direct cause of the excitatory effect. Rather, excitation and oscillations depend on a second-messenger cascade of PLC, G-proteins, and IP3. As the most downstream in the second-messenger pathway, IP3 is a good candidate for direct action on target channels. However, other mechanisms, or a combination of several, may contribute to the effect of type I mGluRs.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
). The preparation was continuously perfused with saline at room temperature. Individual neurons were identified by correlating intracellular recordings with extracellular recordings from identified nerves. After cell identification the STG was isolated from higher nervous centers (except for experiments with ongoing rhythms) by cutting the stomatogastric nerve. Puffs of drugs were delivered by pressure injection (Picospritzer II, General Valve, Fairfield, NJ) from a glass microelectrode, whose tip was broken to create a fine stream over the ganglion. Electrophysiology
Extracellular stainless steel electrodes were placed on gastric motor nerves and insulated with a petroleum oil/Vaseline mixture. Extracellular signals were recorded using AC amplifiers (A-M Systems, Everett, WA) and intracellular recordings were made with 8- to 10-M
glass electrodes filled with 3 M KCl using intracellular amplifiers (A-M Systems). All data were stored in a PC computer using Axoscope (Molecular Devices, Sunnyvale, CA) for later analysis. Cell photoablation was done by methods described previously (Miller and Selverston 1979). Briefly, the neuron was impaled with electrodes filled with carboxyfluorescein dye. A negative current of about 5 nA was delivered for 10–20 min until the neuron could be observed to be loaded with dye. The preparation was then illuminated with 400- to 500-nm light for 20–30 min until the filled neuron died, as determined electrophysiologically.
Ca2+ imaging
For Ca2+ imaging, identified neurons were impaled with electrodes containing 1 mM Calcium Green (Molecular Probes, Eugene, OR) in water and a negative 5-nA current was delivered for 5–20 min. After filling the neuron with dye, the preparation was moved to a chamber made especially for use on an inverted microscope. The chamber was made of a plastic petri dish coated with Sylgard. In the middle of the dish, beneath the ganglion, the bottom, including the Sylgard, was cut out to the size of a round coverslip (15 mm). A coverslip was glued over the hole and sealed with Sylgard. In this way we could use the usual setup to record from the STG but at the same time image the ganglion through the glass (coverslip). The nerves were pinned to the Sylgard and two rubber strips were placed on top of the two nerves leaving the ganglion, to fix it in place. The chamber and perfusion were placed on an inverted microscope (Diaphot 300, Nikon, Tokyo, Japan). The intracellular recording was done using Axoclamp B2 (Molecular Devices, Sunnyvale, CA). For imaging we used a CCD camera (60 Hz) mounted on the microscope (Ionoptix, Freehold, NJ) with 75-W xenon light source filtered at 480 nm and the emitted light was collected at 530 nm. The systems acquisition board (Ionoptix) integrated the images and intracellular recording (1 kHz). Every three images were averaged and the fluorescence changes in areas of interest were measured as the percentage of 
F/F.
Physiological and drug solutions
The experiments were performed in standard Panulirus saline containing (in mmol/l): 479.1 NaCl, 12.7 KCl, 13.7 CaCl, 10.0 MgSO4, 3.9 NaSO4, 5.0 HEPES, and 5.0 TES (pH 7.4). L-Quisqualic acid (L-quis; Tocris, Bristol, UK) was dissolved in 1 N NaOH at stock concentration of 50 mM. 3,5-Dihydroxyphenylglycine (DHPG, 50 mM; Tocris), picrotoxin (PTX; 10 mM; Sigma), and thimerosal (100 mM; Research Organics, Cleveland, OH) were dissolved in water. Cyclopiazonic acid (CPA; Alexis, San Diego, CA) was dissolved in DMSO at stock concentrations of 100 mM. All drugs and stocks were kept frozen at –20°C and dissolved in Panulirus saline before use.
Caffeine (10 mM; Sigma), concanavalin A (Con A, 100 mM; Worthington Biochemical, Lakewood, NJ), and neomycin (100 mM; Alexis) were dissolved in Panulirus saline before use. Bis-(o-aminophenoxy)-N,N,N',N'-tetraacetic acid (BAPTA, 0.6 N; Sigma) was dissolved in 600 mM KCl and adjusted to 7.4 pH. IP3 (Alexis) was dissolved in water at a stock concentration of 10 mM. Before an experiment it was diluted in 300 mM KCl with HEPES pH 7.4. Guanosine 5'-[-thio]disphosphate (GDPS) and guanosine 5'-[-thio]triphosphate (GTPS) (1 mM, Sigma) were dissolved in 300 mM KCl with HEPES pH 7.4 before the experiment and used to fill the injection electrode.
Analysis and statistics
Spike detection and instantaneous frequency measurements were done with Orbital Spikes. Graphs and time-decay fittings were done with Origin (Microcal Software, Northampton, MA). The data are presented as means ± SE. Student's t-test was used for experiments with n > 30. In all other experiments a Wilcoxon nonparametric test was used. Statistics were calculated with SigmaStat (Jandel Scientific) using a significance level of 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
The existence of the type I mGluR in the lobster STG was first demonstrated (Krenz et al. 2000) using L-quisqualic acid (L-quis) and DHPG as agonists. Applying the drugs on a quiescent gastric system of a different lobster species (Panulirus argus) caused depolarization and oscillation of some neurons. However, it was never clear that this was indeed metabotropic glutamate activation. To verify that glutamate has a similar effect, we applied glutamate puffs on the STG while the ionotropic Glu receptors were blocked with picrotoxin (Bidaut 1980). In five such experiments the predominant response of an ongoing gastric mill rhythm was excitation. However, in some cases neurons were inhibited and the frequency of oscillation decreased. Figure 2 shows an example of a recording from the gastric system. A clear response to a 1-mM puff of Glu can be seen in the LG (lateral gastric) and MG (medial gastric) neurons. The MG depolarized, as can be seen in the intracellular recording, and the rhythm became stronger and regular. Note that the GMs (gastric mills), as recorded from the anterior lateral nerve (ALn), showed only a minor response. The excitatory effect in general was similar to the excitatory mGluR I response observed by Krenz et al. (2000).
|
). Apart from these differences, in our experiments we could see that the application of 100 mM L-quis produced similar depolarizations and transient oscillations in the gastric system as in P. argus (Fig. 3A). The depolarization was specific to certain neuron types (Fig. 3A). The neurons that depolarized were LG, MG, DG, AM, and the GMs. LPG neurons on the other hand were hyperpolarized, stopped spiking, and followed the oscillation in antiphase to LG/MG. In three experiments where we recorded interneuron 1 (Int1) activity it also initially hyperpolarized. Inhibitory mGluRs II and III were also previously described in the lobster (Krenz et al. 2000). In this case Glu might have an inhibitory effect that was being obscured by the excitatory response during the Glu application. To test the possibility, we applied a Glu puff while a quiescent gastric system was excited by L-quis. As can be seen in Fig. 3B both LG and MG were inhibited. Thus when further excitation by mGluR I is impossible, the inhibitory effect of the other mGluR types was revealed. This experiment was repeated three times.
|
Because gastric neurons are extensively interconnected (Fig. 1) it was necessary to remove some neurons from the circuit to see the effect on functionally isolated individual cells. This was done by photoablating specific neurons. For example, ablating all four GMs removed electrical synapses that could provide external excitation to MG and LG (Fig. 4) and depolarization persisted when the neurons were isolated. However, LPG hyperpolarization disappeared when MG and LG were killed, indicating that it was an indirect result of inhibition from the two cells.
|
0.001). The maximal spiking frequency was similar for all cell types: 5–6 Hz (see Table 1).
|
15 min in the presence of L-quis before decaying. Subsequent application of L-quis failed to produce a response. We attributed this result to a desensitization process and indeed the response to bath application of L-quis was indistinguishable from a short puff at a similar concentration. Therefore because of the easier washout process we used 0.5- to 4-s puffs to elicit the response. Does the L-quis act on mGluRs?
The lack of type I efficient mGluR blockers in lobster makes it difficult to prove that the effect is a type I mGluR response. We used the fact that the response strongly desensitizes to examine whether another specific type I mGluR agonist (DHPG, 50-mM puff) would interact with the same receptor pathway. In other words, if the response to L-quis had been occluded shortly after an exposure to DHPG, it was likely that their pathways overlap at either the receptor level or downstream. Such an experiment is shown in Fig. 5. In response to an L-quis puff, MG and LG depolarized and started spiking. The LPG on the other hand was inhibited (as can be seen in the extracellular record of Fig. 5 mixed with GM). The puff of DHPG required a much higher concentration to produce a similar response. In this case the response in GM was not strong enough to produce oscillations; rather the neurons just slowly depolarized. When the L-quis was applied again 5 min after the DHPG, just when the GM and MG resting potentials returned to baseline, the response in MG and LG was much weaker in terms of the number of spikes produced and the spiking frequencies. After a 30-min wash, additional application of L-quis was similar to the first application. The occlusion effect differs between neuron types. In seven experiments, the response to DHPG was somewhat smaller than that to L-quis in LG and MG. However, the same concentration of L-quis immediately after DHPG showed significant reduction in spiking frequency (P < 0.05); in six of the seven experiments the neurons did not fire at all. For GMs, the response to L-quis and DHPG was similar (P > 0.5) and the occlusion was ineffective, with no significant difference between the two L-quis applications. This suggests different mechanisms for LG/MG and the GMs, with the GMs having a nonspecific response to L-quis.
|
-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor blocker] (data not shown). Second-messenger mechanisms
A typical metabotropic response, as opposed to an ionotropic response, would be mediated by G-proteins (Conn and Pin 1997). To explore the role of G-proteins in the L-quis–induced depolarization in the lobster, we ionophoretically injected two forms of nonhydrolyzable guanosine analogs (GDPS and GTPS) to manipulate G-protein activity. The injection of GDPS should lock G-proteins in an inactive position, effectively disabling them. In Fig. 6A we show that GDPS injection into an LG abolished its firing in response to an L-quis puff in the injected neuron and reduced firing in the electrically coupled MG. There was very little reduction in the response of the GMs, which are electrically connected but were not injected with the analog. The LG/MG stopped spiking in three of five experiments and in the two other experiments spiking frequency was reduced by >70% (P < 0.05). In one experiment in which the LG was not spiking initially, the depolarization was reduced by more than half. In the noninjected GMs, on the other hand, the frequency was reduced by 22% (n = 3).
|
S is expected to lock G-proteins in a constantly activated state, thus producing the full extent of its activity. However, as shown in Fig. 6B the effect of GTPS injection was similar to the effect of GDPS injection; that is, the injected LG ceased firing altogether and the spiking in the coupled MG was reduced. To address the inconsistency of the results with GTPS we repeated the injection of the GTPS with a lower concentration in the electrode (300 µM) and a shorter injection duration of 10 min. To follow the changes in membrane potential during and after the injection we impaled the LG/MG neurons with one electrode containing the GTPS solution and a second KCl electrode (Fig. 6C). We did not observe systematic changes in the membrane potential as a result of GTPS injection. In eight experiments we found that before injection the response to an L-quis puff was a depolarization of 8.3 ± 3.8 mV and the neuron started spiking up to an average of 5.5 Hz. At the first application of L-qius after the GTPS injection there was no significant change in either the depolarization (6.0 ± 3.5 mV, P > 0.05) or the spiking frequency (3.6 ± 3.0 Hz, P > 0.05). However, during the second application of GTPS the response to L-quis was substantially diminished. Depolarizations in all experiments were reduced on average by 50.6% to 4.1 ± 2.9 (P < 0.05) and the spiking frequency by 86.8% to 0.7 ± 1.5 (P < 0.05). In half the experiments, the neurons stopped spiking. A possible explanation for the effect of GTPS is that, once activated, the G-proteins are locked in the active state and cannot initiate further responses. We examined this possibility by close examination of the response to the first application of L-quis after the GTPS injection. In three of the experiments the spiking frequency increased by about 300% and in two of the experiments the neurons stopped spiking during the first application of L-quis. These results are consistent with the possibility that the amount of GTPS actually injected into the neuron was variable. In some experiments it was just enough to increase activation but in some cases too much was injected and, with the background activation, the G-proteins became inactivated. This issue can be addressed only with patch recordings and very precise control of GTPS concentration. No clear change in neuron condition after the injection, either in resting potential or unusual spiking, was observed. In most cases LG and MG were as quiescent as before the injection and GMs fired at the same low rate.
In mammals mGluR I activates the PLC pathway (Conn and Pin 1997; Hermans and Challiss 2001), in which increases in IP3 production and phosphokinase-C (PKC) activation have been hypothesized to induce release of Ca2+ from internal stores. To test similar mechanisms in the lobster, we used a broad-spectrum PLC inhibitor, neomycin, to inhibit L-quis–induced depolarization. Figure 7 shows that, after a 30-min incubation in neomycin, the MG spiking in response to the L-quis puff was reduced. The response was partially reversed after a wash. Similar results were seen in other experiments on LG. The neomycin incubation effect on GM was much smaller and statistically insignificant, similar to the results with the G-protein injection. Overall, neomycin reduced LG/MG spiking frequency by 62 ± 20% (P < 0.05 n = 5) and in three experiments after a 30-min wash, spiking frequency recovered to 101 ± 40% of control (P > 0.05).
|
Using Ca2+ imaging, we found that the membrane depolarizations and oscillations in response to L-quis were accompanied by a [Ca2+] increase in specific neurons (Fig. 8A). The internal Ca2+ concentration rose simultaneously with the depolarization and was correlated with fluctuations in membrane potential (n = 3). The temporal resolution did not allow us to discern which occurred first. The global Ca2+ concentration also slowly decayed as the response desensitized and the spike frequency of the bursts decreased. The tight temporal correspondence of the Ca2+ concentration with the voltage changes raises the possibility that all the Ca2+ entry was voltage gated (Haag and Borst 2000; Oertner et al. 2001).
|
The role of internal Ca2+ stores in the response to L-quis
The previous results shows a minor voltage-independent [Ca2+] increase that may indicate release from internal Ca2+ stores, as is the case for other mGluR I systems. To further study the internal Ca2+ stores' contribution, we tried to deplete Ca2+ stores by blocking the activity of the sarcoendoplasmic reticulum Ca2+-ATPase (SERCA) pump. Blocking the pump should stop the refilling of internal stores and eventually deplete them. In preliminary experiments using Ca2+ imaging we saw that this SERCA blocker CPA was more effective in transiently increasing Ca2+ levels in STG neurons than the other commonly used drug, thapsigargin (data not shown). After the transient increase, the Ca2+ levels were reduced to levels below baseline, as expected from SERCA blockage. We therefore used CPA to inhibit SERCA. In Fig. 9A we show that CPA did not reduce the firing frequency in the MG/LG or GM neurons. In fact, the peak-spiking frequency in MG, in this example, increased from 1.1 to 4 Hz and from 1.7 to 5 Hz in LG. In the GM, however, the peak frequency decreased only from 14 to 12 Hz. On average the frequency increased by 84 ± 29.8% in LG/MG (P 0.01, n = 8) and 4 ± 8% in GMs (P > 0.05, n = 4).
|
0.05, n = 4) and recovered to 94 ± 6% of the control after washing. In all the experiments that were performed (n = 4) the MG/LG pair stopped spiking, and the firing recovered after washing out caffeine for 45 min. As with other experiments, the effect on GM spiking frequency was minor.
Because the depletion results were somewhat contradictory, perhaps because of secondary effects of caffeine, we tried a more direct experiment of manipulating internal [Ca2+], by clamping the internal Ca2+ concentration using the Ca2+ buffer BAPTA. BAPTA was ionophoretically injected from an electrode containing 60 mM BAPTA in 300 mM KCl (pH 7) during 15- to 30-min pulses. After BAPTA injection (Fig. 9C) L-quis–induced depolarization did not disappear; rather the LG peak-spiking frequency remained unchanged (8 Hz) but the burst duration increased from 15 to 30 s. On average the spiking frequency in LG/MG increased by 250 ± 106% (n = 7) and the burst duration by 76 ± 37% (P 0.05). In fact, the longer burst duration is an indication that the BAPTA is present inside the neuron. As has been shown by Zhang et al. (1995), BAPTA injection increases burst duration in LG neurons probably by reducing Ca2+-activated K+ currents. Note that the injection protocol was identical to that used with nonhydrolyzable guanosine analogs, but had the opposite effect. This indicates that this is not a nonspecific effect of the current injection, but a real effect of the injected substances.
Can IP3 activation produce the depolarization?
The increase in Ca2+ cannot by itself explain the depolarization because injecting BAPTA into the neurons prolongs, rather than blocks, the response. We therefore sought other mechanisms that could be involved. One possibility was direct activation of channels by IP3 (Kiselyov et al. 1999; Zhainazarov and Ache 1999). We tested this possibility by intracellular injection of IP3 into LG or MG neurons (Fig. 10A), assuming that activating the downstream portions of the pathways would result in decreased responsiveness to subsequent mGluR activation. Because the injection was done ionophoretically, it was impossible to follow its effect on the resting neuron activity during the injection (resulting from the large ionic imbalance produced by the current injection). Nevertheless, we did not see any clear changes in the neuron resting potential or firing rate after injection. After the IP3 injection the robust L-quis–induced spiking in MG completely disappeared, although the spiking in GM was reduced only slightly. On average the spiking rate in response to L-quis decreased by 68 ± 16% (n = 7) in LG/MG and 27 ± 35% (n = 5) in the noninjected GMs. The recovery from the IP3 injection was extremely slow and generally the response to L-quis deteriorated after a 1-h wash. Further support for the involvement of IP3 came from experiments with thimerosal, a drug that inhibits the degradation of IP3 and would therefore be expected to enhance its effect. When the ganglion was incubated in thimerosal (as is seen in Fig. 10B) the response to L-quis was stronger in LG/MG both in duration and spiking frequency, which increased by 223 ± 13% in four experiments, and one MG that was not spiking in response to L-quis before thimerosal began spiking in response to an L-quis puff after bathing in thimerosal (P < 0.05). Together these results suggest the involvement of IP3 in excitation and oscillation production as the result of mGluR type I activation.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
Several lines of evidence suggest that what we report here is indeed a metabotropic effect and not an ionotropic activation: 1) The responses persisted in the presence of known ionotropic blockers such as PTX, a substance that blocks all Glu currents in the STG (Cleland and Selverston 1998). 2) At least in MG and LG the response to L-quis is occluded by another specific type I mGluR agonist, probably through desensitization mechanisms, again demonstrating that these mGluR agonists share pathways. Interestingly, the GM response was not occluded as much, which may indicate that L-quis activated a non-mGluR response in the GMs, perhaps in addition to mGluR responses. 3) Glu and the metabotropic agonist L-quis share the same excitatory pathway. This was shown by the fact that a clear inhibition of Glu was seen when Glu application was followed by L-quis application (desensitization of the excitatory pathway). In addition, the response was unaffected by drugs specific to kainate and AMPA receptors. 4) Guanosine analogs such as GDPS significantly reduced the response. Clearly, if the response is G-protein dependent it must be metabotropic.
A desensitization on the order of tens of seconds was observed in our experiments, as well as in other lobster species (Krenz et al. 2000). Type I mGluR response desensitization is widely seen and has been characterized in mammalian systems. In mammals desensitization occurs through phosphorylation of the receptor by PKC (Alaluf et al. 1995; Francesconi and Duvoisin 2000; Gereau and Heinemann 1998). The time courses described are consistent with the time course of our experiments. A similar mechanism may explain the finding that the two mGluR agonists occluded each other through desensitization of the same receptor pathway.
An alternative possibility that we cannot exclude at this time is that of a more complex interaction between different second-messenger pathways. For instance, it is possible that several G-protein–dependent pathways converge onto a single effector. Activation of any of these pathways would activate the effector, which might result in ineffective activation by another pathway. In this context, it has been shown that multiple neuromodulators in the STG activate the same current (Swensen and Marder 2001), one that has similarities to the L-quis–induced current. If signal transduction pathways overlap, it would not be surprising for them to have, at least in part, similar effects.
G-protein activation
Usually GDPS and GTPS have antagonistic effects, the former locking the G-protein in an inactive state and the latter constantly activating the G-protein. However, in this system with strong desensitization, the effective result is expected to be the same. The first prevents the system from being activated and the second puts the system into a desensitized state, again preventing further activation. It was previously shown that some types of responses to Glu in the lobster are mediated by G-proteins (Miwa et al. 1990). This study used GTPS to demonstrate that after an injection a Glu-induced current in motor neurons was reduced. However, in the case of motor neurons, the evoked current was an outward current accompanied by a conductance increase and inhibition. In our experiments, the metabotropic-evoked response was excitation. In addition to mGluR type I there are also type II and type III mGluRs, which increase inhibition (see Fig. 3B; Krenz et al. 2000), and are more likely to correspond to the above results. Nonetheless, these data demonstrate that such a mechanism exists in the lobster.
In addition, IP3 receptors have been described in other crustaceans (Ukhanov et al. 1998; Zhainazarov and Ache 1999), and in lobsters specifically, although not in the stomatogastric system. This suggests that the described machinery exists in the lobster and we have demonstrated that it is activated by mGluR in the STG. A receptor with a partial sequence close to that of mammalian mGluR I has been cloned from P. argus (Perez-Acevado et al., unpublished results), where it has been shown to be expressed. Preliminary results have shown that it is also expressed in P. interruptus (D. Baro, personal communication).
It is somewhat contradictory that IP3 is involved in the response to L-quis but internal Ca2+ stores are not. In the classical view, IP3 exerts its action by Ca2+ elevation. However, in Purkinje cells IP3 evokes only a small Ca2+ signal (Batchelor et al. 1996) and dopaminergic neurons in substantia nigra show a Ca2+-independent response to mGluR I (Guatteo et al. 1999; Tozzi et al. 2003). One possibility is that the IP3 exerts its effect not through Ca2+ but directly on channels (Kiselyov et al. 1999; Zhainazarov and Ache 1999) or it directly modulates some of the effectors of the channel (Kammermeier et al. 2000). Many kinds of channels are directly modulated by G-proteins (Wickman and Clapham 1995). However, we cannot rule out the possibility that in our manipulation of internal Ca2+ levels, the contribution of internal stores was obscured by the much larger effect of voltage-gated Ca2+ currents, a possibility even more likely if there is close spatial apposition between the processes (Topolnik et al. 2005).
Spatial distribution of caffeine and IP3 stores can explain the differential effects
The involvement of Ca2+ from internal stores in the response to mGluR type I in lobsters appears to be inconclusive on first sight. On the one hand, internal stores depleted by CPA did not eliminate the depolarization, as would have been expected had the internal store been key players. On the other hand, bath application of caffeine does reduce the response. This apparent contradiction can be resolved if there are two functionally separate stores (Berridge 1997; Seymour-Laurent and Barish 1995), but only the caffeine store takes part in the response. This is an unlikely possibility because we have shown that IP3 itself is involved in the response. The more likely possibility is that caffeine directly blocked the IP3 receptors (Ehrlich et al. 1994; Parker and Ivorra 1991). In that case, by applying caffeine, in addition to depletion of caffeine-sensitive stores, we blocked the IP3 receptors, which resulted in a reduced response to the mGluR agonist. In principle this should create the opposite effect to the effect we have seen with thimerosal. Indeed, thimerosal prolonged the IP3-induced response, just the opposite of the caffeine effect.
The mechanism of mGluR type I actions in lobster is similar to some extent to that seen in some vertebrate examples. In summary, the receptors activate the PLC pathway through G-proteins. The G-proteins in addition to activation of PLC also might act directly on the channels, whereas the PLC increases IP3 levels and possibly activates PKC. Although in mammals the main action of IP3 is Ca2+ release from internal stores, in the lobster this does not seem to be the most significant effect in producing a depolarization. Instead the IP3 appears to affect currents more, either directly on the channel or by modulation of other effectors.
The functional role of mGluR I in the STG system is yet to be determined, but application of mGluR I agonists on ongoing activity causes transient excitation and prolonged burst duration in specific gastric neurons. In the intact system the Glu may come from higher, yet undetermined centers, or from within the system (Huang 1998) when excess Glu reaches the mGluRs. When this occurs, it has a combined effect of excitation and inhibition of specific neurons. Because the excitation in our experiments was more dominant, we expect that it will further excite the system in a form of positive feedback and may therefore serve as a gain-control mechanism.
|
GRANT |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: R. Levi, Institute for Nonlinear Science, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0402 (E-mail: rlevi{at}ucsd.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
Anwyl R. Metabotropic glutamate receptors: electrophysiological properties and role in plasticity. Brain Res Brain Res Rev 29: 83–120, 1999.[CrossRef][Medline]
Bal T, Nagy F, and Moulins M. The pyloric central pattern generator in crustacea: a set of conditional neuronal oscillators. J Comp Physiol 163: 715–727, 1988.[CrossRef]
Batchelor AM, Vranesic I, Del Principe F, Garthwaite J, and Knopfel T. The synaptic potential mediated by metabotropic glutamate receptors is not associated with a substantial elevation of cytosolic free calcium concentration in Purkinje cells. Neuroreport 7: 1949–1952, 1996.[Web of Science][Medline]
Berridge MJ. Elementary and global aspects of calcium signalling. J Exp Biol 200: 315–319, 1997.[Abstract]
Bidaut M. Pharmacological dissection of pyloric network of the lobster stomatogastric ganglion using picrotoxin. J Neurophysiol 44: 1089–1101, 1980.
Cleland TA and Selverston AI. Inhibitory glutamate receptor channels in cultured lobster stomatogastric neurons. J Neurophysiol 79: 3189–3196, 1998.
Congar P, Leinekugel X, Ben-Ari Y, and Crepel V. A long-lasting calcium-activated nonselective cationic current is generated by synaptic stimulation or exogenous activation of group I metabotropic glutamate receptors in CA1 pyramidal neurons. J Neurosci 17: 5366–5379, 1997.
Conn PJ and Pin JP. Pharmacology and functions of metabotropic glutamate receptors. Annu Rev Pharmacol Toxicol 37: 205–237, 1997.[CrossRef][Web of Science][Medline]
Ehrlich BE, Kaftan E, Bezprozvannaya S, and Bezprozvanny I. The pharmacology of intracellular Ca(2+)-release channels. Trends Pharmacol Sci 15: 145–149, 1994.[CrossRef][Medline]
El Manira A, Kettunen P, Hess D, and Krieger P. Metabotropic glutamate receptors provide intrinsic modulation of the lamprey locomotor network. Brain Res Brain Res Rev 40: 9–18, 2002.[CrossRef][Medline]
Elson RC and Selverston AI. Mechanisms of gastric rhythm generation in the isolated stomatogastric ganglion of spiny lobsters: bursting pacemaker potentials, synaptic interactions, and muscarinic modulation. J Neurophysiol 68: 890–907, 1992.
Francesconi A and Duvoisin RM. Opposing effects of protein kinase C and protein kinase A on metabotropic glutamate receptor signaling: selective desensitization of the inositol trisphosphate/Ca2+ pathway by phosphorylation of the receptor-G protein-coupling domain. Proc Natl Acad Sci USA 97: 6185–6190, 2000.
Gereau RW and Heinemann SF. Role of protein kinase C phosphorylation in rapid desensitization of metabotropic glutamate receptor 5. Neuron 20: 143–151, 1998.[CrossRef][Web of Science][Medline]
Guatteo E, Mercuri NB, Bernardi G, and Knopfel T. Group I metabotropic glutamate receptors mediate an inward current in rat substantia nigra dopamine neurons that is independent from calcium mobilization. J Neurophysiol 82: 1974–1981, 1999.
Guerineau NC, Bossu JL, Gahwiler BH, and Gerber U. Activation of a nonselective cationic conductance by metabotropic glutamatergic and muscarinic agonists in CA3 pyramidal neurons of the rat hippocampus. J Neurosci 15: 4395–4407, 1995.[Abstract]
Haag J and Borst A. Spatial distribution and characteristics of voltage-gated calcium signals within visual interneurons. J Neurophysiol 83: 1039–1051, 2000.
Harris-Warrick RM, Marder E, Selverston AI, and Moulins M. Dynamic Biological Networks: The Stomatogastric Nervous System. Cambridge, MA: MIT Press, 1992.
Hermans E and Challiss RA. Structural, signalling and regulatory properties of the group I metabotropic glutamate receptors: prototypic family C G-protein-coupled receptors. Biochem J 359: 465–484, 2001.[CrossRef][Web of Science][Medline]
Huang EP. Synaptic transmission: spillover at central synapses. Curr Biol 8: R613–R615, 1998.[CrossRef][Web of Science][Medline]
Kammermeier PJ, Xiao B, Tu JC, Worley PF, and Ikeda SR. Homer proteins regulate coupling of group I metabotropic glutamate receptors to N-type calcium and M-type potassium channels. J Neurosci 20: 7238–7245, 2000.
Katz PS. Neuromodulation intrinsic to the central pattern generator for escape swimming in Tritonia. Ann NY Acad Sci 860: 181–188, 1998.[CrossRef][Web of Science][Medline]
Kettunen P, Krieger P, Hess D, and El Manira A. Signaling mechanisms of metabotropic glutamate receptor 5 subtype and its endogenous role in a locomotor network. J Neurosci 22: 1868–1873, 2002.
Kiselyov KI, Semyonova SB, Mamin AG, and Mozhayeva GN. Miniature Ca2+ channels in excised plasma-membrane patches: activation by IP3. Pfluegers Arch 437: 305–314, 1999.[CrossRef][Web of Science][Medline]
Krenz WD, Nguyen D, Perez-Acevedo NL, and Selverston AI. Group I, II, and III mGluR compounds affect rhythm generation in the gastric circuit of the crustacean stomatogastric ganglion. J Neurophysiol 83: 1188–1201, 2000.
Krieger P, Hellgren-Kotaleski J, Kettunen P, and El Manira AJ. Interaction between metabotropic and ionotropic glutamate receptors regulates neuronal network activity. J Neurosci 20: 5382–5391, 2000.
Miller JP and Selverston AI. Rapid killing of single neurons by irradiation of intracellularly injected dye. Science 206: 702–704, 1979.
Miwa A, Ui M, and Kawai N. G protein is coupled to presynaptic glutamate and GABA receptors in lobster neuromuscular synapse. J Neurophysiol 63: 173–180, 1990.
Nagy F and Dickinson PS. Control of a central pattern generator by an identified modulatory interneurone in crustacea. I. Modulation of the pyloric motor output. J Exp Biol 105: 33–58, 1983.
Oertner TG, Brotz TM, and Borst A. Mechanisms of dendritic calcium signaling in fly neurons. J Neurophysiol 85: 439–447, 2001.
Parker I and Ivorra I. Caffeine inhibits inositol trisphosphate-mediated liberation of intracellular calcium in Xenopus oocytes. J Physiol 433: 229–240, 1991.
Perez-Acevedo NL and Krenz WD. Metabotropic glutamate receptor agonists modify the pyloric output of the crustacean stomatogastric ganglion. Brain Res 1062: 1–8, 2005.[CrossRef][Web of Science][Medline]
Selverston AI. Neural circuitry underlying oscillatory motor output. J Physiol (Paris) 73: 463–470, 1977.
Selverston AI, Elson R, Rabinovich M, Huerta R, and Abarbanel D. Basic principles for generating motor output in the stomatogastric ganglion. Ann NY Acad Sci 860: 35–50, 1998.[CrossRef][Web of Science][Medline]
Selverston AI and Miller JP. Mechanisms underlying pattern generation in the lobster stomatogastric ganglion as determined by selective inactivation of identified neurons. I. Pyloric system. J Neurophysiol 44: 1102–1121, 1980.
Selverston AI and Moulins M. The Crustacean Stomatogastric System. Berlin: Springer-Verlag, 1987.
Selverston AI, Russell DF, Miller JP, and King DG. The stomatogastric nervous system: structure and function of a small neural network. Prog Neurobiol 7: 215–290, 1976.[CrossRef][Medline]
Seymour-Laurent KJ and Barish ME. Inositol 1,4,5-trisphosphate and ryanodine receptor distributions and patterns of acetylcholine- and caffeine-induced calcium release in cultured mouse hippocampal neurons. J Neurosci 15: 2592–2608, 1995.[Abstract]
Swensen AM and Marder E. Modulators with convergent cellular actions elicit distinct circuit outputs. J Neurosci 21: 4050–4058, 2001.
Tempia F, Alojado ME, Strata P, and Knopfel T. Characterization of the mGluR(1)-mediated electrical and calcium signaling in Purkinje cells of mouse cerebellar slices. J Neurophysiol 86: 1389–1397, 2001.
Topolnik L, Congar P, and Lacaille J-C. Differential regulation of metabotropic glutamate receptor- and AMPA receptor-mediated dendritic Ca2+ signals by presynaptic and postsynaptic activity in hippocampal interneurons. J Neurosci 25: 990–1001, 2005.
Tozzi A, Bengtson CP, Longone P, Carignani C, Fusco FR, Bernardi G, and Mercuri NB. Involvement of transient receptor potential-like channels in responses to mGluR-I activation in midbrain dopamine neurons. Eur J Neurosci 18: 2133–2145, 2003.[CrossRef][Web of Science][Medline]
Ukhanov K, Ukhanova M, Taylor CW, and Payne R. Putative inositol 1,4,5-trisphosphate receptor localized to endoplasmic reticulum in Limulus photoreceptors. Neuroscience 86: 23–28, 1998.[CrossRef][Web of Science][Medline]
Wickman KD and Clapham DE. G-protein regulation of ion channels. Curr Opin Neurobiol 5: 278–285, 1995.[CrossRef][Web of Science][Medline]
Woodhall G, Gee CE, Robitaille R, and Lacaille J-C. Membrane potential and intracellular Ca oscillations activated by mGluRs in hippocampal stratum oriens/allveus interneurons. J Neurophysiol 81: 371–382, 1999.
Wu RL and Barish ME. Modulation of a slowly inactivating potassium current, I(D), by metabotropic glutamate receptor activation in cultured hippocampal pyramidal neurons. J Neurosci 19: 6825–6837, 1999.
Zhainazarov AB and Ache BW. Effects of phosphatidylinositol 4,5-bisphosphate and phosphatidylinositol 4-phosphate on a Na+-gated nonselective cation channel. J Neurosci 19: 2929–2937, 1999.
Zhang B, Wootton JF, and Harris-Warrick RM. Calcium-dependent plateau potentials in a crab stomatogastric ganglion motor neuron. II. Calcium-activated slow inward current. J Neurophysiol 74: 1938–1946, 1995.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
