| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
The Journal of Neurophysiology Vol. 79 No. 1 January 1998, pp. 361-370
Copyright ©1998 by the American Physiological Society
1 Department of Zoology, University of Cambridge, Cambridge CB2 3EJ, United Kingdom; and 2 Institut für Neurobiologie, Freie Universität Berlin, D-14195 Berlin, Germany
 |
ABSTRACT |
|---|
 Abstract
 Introduction
Methods
Results
Discussion
References
|
|---|
Baudoux, Sylvie, Carsten Duch, and Oliver T. Morris. Coupling of efferent neuromodulatory neurons to rhythmical leg motor activity in the locust. J. Neurophysiol. 79: 361-370, 1998. The spike activity of neuromodulatory dorsal unpaired median (DUM) neurons was analyzed during a pilocarpine-induced motor pattern in the locust. Paired intracellular recordings were made from these octopaminergic neurons during rhythmic activity in hindleg motor neurons evoked by applying pilocarpine to an isolated metathoracic ganglion. This motor pattern is characterized by two alternating phases: a levator phase, during which levator, flexor, and common inhibitor motor neurons spike, and a depressor phase, during which depressor and extensor motor neurons spike. Three different subpopulations of efferent DUM neurons could be distinguished during this rhythmical motor pattern according to their characteristic spike output. DUM 1 neurons, which in the intact animal do not innervate muscles involved in leg movements, showed no change apart from a general increase in spike frequency. DUM 3 and DUM 3,4 neurons produced the most variable activity but received frequent and sometimes pronounced hyperpolarizations that were often common to both recorded neurons. DUM 5 and DUM 3,4,5 neurons innervate muscles of the hindleg and showed rhythmical excitation leading to bursts of spikes during rhythmic activity of the motor neurons, which innervate these same muscles. Sometimes the motor output was coordinated across both sides of the ganglion so that there was alternating activity between levators of both sides. In these cases, the spikes of DUM 5 and DUM 3,4,5 neurons and the hyperpolarization of DUM 3 and DUM 3,4 neurons occurred at particular phases in the motor pattern. Our data demonstrate a central coupling of specific types of DUM neurons to a rhythmical motor pattern. Changes in the spike output of these particular efferent DUM neurons parallel changes in the motor output. The spike activity of DUM neurons thus may be controlled by the same circuits that determine the action of the motor neurons. Functional implications for real walking are discussed.
Neuromodulation is essential for the adaptive regulation of an animal's neural and motor machinery because it allows the adjustment of motor patterns to fit different behavioral needs. Neuromodulators can influence profoundly all steps of information processing in neuronal pathways. They alter signaling by sensory neurons (Pasztor and Bush 1989
Preparation
Experiments were carried out using 48 adult desert locusts, Schistocerca gregaria (Forksål), of either sex, taken from our crowded laboratory colony. We used an in vitro preparation as described by Ryckebusch and Laurent (1993) Electrophysiology
The rhythmical motor output evoked by pilocarpine under these conditions has been well described and characterized by Ryckebusch and Laurent (1993 Analysis
Data were stored on a Racal four-channel FM tape recorder and analyzed off-line on a computer running Spike2 software (Cambridge Electronic Design) after digitization through a CED 1401plus interface. Spikes in the DUM neurons were converted into separate spike event channels using a hardware window discriminator, whereas extracellular motor nerve recordings were digitized at 10 kHz. Sorting of the extracellular recordings into different motor units was carried out manually with assistance from a spike-sorting option within Spike2. This was necessary because the precise spike waveform of particular motor units varied from preparation to preparation and even within preparations due to changes in the gain of the recordings over time. In addition, some levator trochanteris motor neuron spikes are so similar in shape and size to SETi spikes that they cannot be distinguished by these criteria alone. However, because SETi is the only unit that spikes in bursts out of phase with the levators in N3B (see Ryckebusch and Laurent 1993
Pilocarpine-induced motor pattern
The pilocarpine-induced motor rhythm described by Ryckebusch and Laurent (1993)
Changes in the synaptic drive and spiking of DUM neurons coincided with generation of the rhythmic
motor pattern
Before pilocarpine application, DUM neurons usually spiked infrequently and never in patterned bursts. All types of DUM neurons received postsynaptic potentials (Fig. 1A) and sometimes spiked at low frequencies (0.02-0.1 Hz). Pilocarpine caused an increase in the frequency of spikes (Fig. 1B). As the motor output became established, the spike activity of DUM neurons settled into patterns that were characteristic for the different types (see next section). Injection of hyperpolarizing and depolarizing current into individual DUM neurons did not alter the frequency of the rhythm nor did it add new motor units to the rhythm. Patterned DUM neuron activity never preceded the occurrence of coordinated motor output. Instead, changes in DUM activity often coincided with the first visible changes in motor output. For example, changes in synaptic inputs to a DUM 3,4 neuron coincided precisely with the first burst of levator and CI1 spikes (Fig. 2). During this first levator burst, the DUM 3,4 neuron produced a pair of spikes that was superimposed on a marked hyperpolarization (Fig. 2,
Spike activity was characteristic for the different types of DUM neuron
The pattern of spikes observed in DUM neurons during established motor patterns were characteristic for different morphological types. A broad distinction could be made between those DUM types that produced rhythmic bursts of spikes and those that did not (Table 2 and Fig. 3). DUM neurons that produced spikes in bursts resulted in autocorrelograms with a peak at lag 0 surrounded by two troughs, and if the period of bursting was regular, there were further positive peaks at intervals corresponding to the cycle period (Fig. 3, 3rd trace). DUM neurons that spiked sporadically did not have a peak at lag 0 in their auto-correlograms (Fig. 3, top 2 traces): such activity was classified as "unpatterned" for the purposes of Table 2.
DUM 1 NEURONS.
None of the DUM 1 neurons recorded showed rhythmical bursting during the pilocarpine-induced motor pattern, but five recordings revealed sporadic spiking. Figure 4 shows an example where DUMDL received many synaptic potentials, but there was no clear patterning in its spike output. Furthermore, DUMDL did not produce spikes at any preferred phase of the motor pattern (Fig. 4B). In the other two recordings made of DUM 1 neurons, they did not spike at all.
DUM 3 AND DUM 3,4 NEURONS.
These neurons had similar spike activity during pilocarpine-evoked motor rhythms. In most preparations (32/42), they produced unpatterned activity of the type shown in Fig. 5. In 10 preparations where 2 of these neurons were recorded simultaneously for periods DUM 3,4,5 AND DUM 5 NEURONS.
Sixteen of 17 DUM 3,4,5 and all 13 DUM 5 neurons recorded during pilocarpine-induced motor patterns produced rhythmical bursts of spikes. An example is shown in Fig. 6, where a DUM 3,4,5 and DUMETi (the DUM 5 neuron that innervates the extensor tibia muscle) showed rhythmical bursts at different phases of the cycle. The frequency of DUM neuron bursting was approximately twice that of the cycle period of one hemiganglion. On other occasions, the two recorded DUM 5 neurons produced bursts during the same phases of the cycle (Fig. 7). The cross-correlogram in Fig. 7B has a similar shape to the autocorrelogram of the upper DUM 5 neuron, indicating that both neurons produced bursts of spikes during the same phases of the cycle.
Activity of individual DUM neurons is coupled to the motor pattern
DUM neurons project bilaterally, so it is possible that they might receive synaptic drive linked to the motor output of both sides of the ganglion. However, in most pilocarpine preparations, the output of the ganglion is not coordinated across the two sides, i.e., levator bursts do not alternate strictly from side to side and the intrinsic cycle frequencies are different for the two sides (Ryckebusch and Laurent 1993
We have shown that the pattern of spike production in specific efferent neuromodulatory neurons is linked to the rhythmical activity in the leg motor neurons. Our study demonstrates that the spike activity of particular efferent DUM neurons is coupled centrally to a rhythmical motor pattern and that these DUM neurons are not activated as a homogeneous group but as individuals. In our locust preparation, a stable and long-lasting rhythmic activity was induced in leg motor neurons of isolated metathoracic ganglia by the application of pilocarpine. Such rhythms were first described in detail by Ryckebusch and Laurent (1993) Different subsets of DUM neurons were linked differently to the pilocarpine-induced motor rhythm
Efferent DUM neurons were activated differently according to their morphological type, and this finding points to close links between the action of a specific locomotory circuit and particular subsets of DUM neurons. Moreover, it suggests that DUM neurons may be allocated more precisely to specific motor acts than previously anticipated by Hoyle and Dagan (1978) What effects might DUM neurons have during real walking?
Because the motor output of an isolated ganglion treated with pilocarpine reliably produces a well-structured and long-lasting motor output that shows similarities to walking (Ryckebusch and Laurent 1993
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
; Ramirez and Orchard 1990) and the activity of central pattern generating networks (Chiel et al. 1990; Harris-Warrick et al. 1992; Katz and Harris-Warrick 1990; Stevenson and Kutsch 1988), and they have presynaptic effects on the terminals of the motor neurons and postsynaptic effects on muscle fibers (Evans and O'Shea 1977, 1978; O'Shea and Evans 1979). Neuromodulators may be released locally in specific parts of the CNS, onto particular groups of muscle fibers or specific sensory organs, or they may be released more generally into the circulatory system.
) and release octopamine (Morton and Evans 1984; Stevenson and Spörhase-Eichmann 1995). Efferent DUM neurons have their somata at the dorsal midline of each segmental ganglion and project bilaterally to the periphery. The metathoracic ganglion of the locust contains 19 efferent DUM neurons (Campbell et al. 1995) of different types that can be distinguished by their axonal projection patterns through the different lateral nerves (Watson 1984). The peripheral effects of these neurons include the modulation of neurotransmitter release from motor neurons, the reduction of basal muscle tension, the increase of muscular twitch contraction force, the increase of muscle relaxation rate, and the suppression of myogenic rhythms (Evans and O'Shea 1977, 1978; Evans and Siegler 1982; Malamud et al. 1988; Whim and Evans 1988). Neuromodulatory effects of octopamine on wing and leg proprioceptors have been demonstrated in the locust (Eder and Bräunig 1996; Matheson 1997; Ramirez and Orchard 1990). So far there is no direct evidence that efferent DUM neurons also can influence central processing, and although DUM neurons have extensive branching patterns within the CNS, no conventional output synapses have been found (Watson 1984).
). In ballistic kicks of a locust hindleg, however, specific DUM neurons are activated in a distinctive and repeatable pattern that is linked closely to the motor pattern, whereas other DUM neurons are inhibited or unaffected (Burrows and Pflüger 1995). These observations support the hypothesis that these neuromodulatory neurons act individually or in particular combinations in the control of movement. It is, however, not known if DUM neurons are coupled to central-pattern-generating networks or what roles they might play during a rhythmical motor behavior such as walking. If neuromodulators are to modify a specific behavior, then they should be released at the right time and in the appropriate location. This implies that the action of neuromodulatory neurons should constitute part the motor program. Such a link between motor patterns and neuromodulatory action could result from either sensory feedback acting on particular neuromodulatory neurons or from central coupling.
; Delcomyn 1980; Graham 1985; Grillner 1985; Pearson 1993; Zill 1986). Nevertheless, isolated CNSs treated pharmacologically can produce rhythmical motor patterns. In arthropods, isolated parts of the CNS have been induced to produce rhythmic motor activity by application of the muscarinic agonist pilocarpine (e.g., Chrachri and Clarac 1990; Gorczyca et al. 1990; Johnston and Levine 1996). In locusts and stick insects, pilocarpine application to an isolated thoracic ganglion results in coordinated rhythmical activity in the leg motor neuron pools (Büschges et al. 1995; Ryckebusch and Laurent 1993). Ryckebusch and Laurent concluded that pilocarpine activated a central pattern generator, which produced a pattern of motor neuron activity that would be appropriate for the movement of a single leg of a segmental pair of legs during walking. In a second study, using a chain of two or three thoracic ganglia treated with pilocarpine, they showed that the motor output of different segments is coupled centrally in a manner appropriate for the alternating tripod gait of walking (Ryckebusch and Laurent 1994).
View this table:
TABLE 1.
DUM neurons of the metathoracic ganglion
METHODS
Abstract
Introduction
Methods
Results
Discussion
References
. The metathoracic ganglion was dissected out of the animal, leaving the leg nerves long and the longitudinal tracheae intact, and pinned dorsal side uppermost in a silicone elastomer (Sylgard)-lined chamber filled with locust saline [composition (in mM): 140 NaCl, 10 KCl, 4 CaCl2, 4 NaHCO3, and 6 NaHPO4]. The tracheae were opened up at the surface to maintain a supply of air. Protease (SIGMA type XIV) was applied directly to the dorsal surface of the metathoracic ganglion as crystals and left to act for ~40 s before being rinsed off with fresh saline. This facilitated penetration with microelectrodes and enabled pilocarpine to pass through the sheath of the ganglion. A stock solution of 0.8 mM pilocarpine hydrochloride (Sigma) in saline was prepared and added to the bath to give final concentrations in the range 0.2-0.4 mM.
, 1994). We used polyethylene suction electrodes to record the motor output of the ganglion in leg nerves 3B and 5A because these contain the main units used by Ryckebusch and Laurent to characterize the rhythmical motor pattern. Levator trochanteris and slow extensor tibiae (SETi) spikes could be observed in N3B, and the activity of two depressor trochanteris motor neurons (1 fast, Df; 1 slow, Ds) and the common inhibitor motor neuron 1 (CI1) were recorded in N5A.
. DUM neurons were identified by their soma position (dorsal midline), characteristic soma spikes (long-duration, large-amplitude spikes with a prominent afterhyperpolarization) and by their characteristic axonal branch patterns after intracellular staining with cobalt (injected by 5-10 nA depolarizing current pulses of 500-ms duration at 1 Hz for 10-20 min). After cobalt injection, 15 min was allowed for diffusion of ions into the lateral nerves and fine processes. The cobalt then was precipitated with ammonium sulphide and the ganglion washed in saline before fixation in formaldehyde. Silver intensification was carried out following the technique of Bacon and Altman (1977).
), it was possible to separate these units on the basis of burst timing. Differences in spike height of the three motor units in N5A were sufficiently consistent for the computer to use this as a reliable criterion for discrimination. Once units had been separated into different spike event channels, the correspondingtime series were used to construct phase histograms and event auto-/cross-correlograms.
View larger version (50K):
[in a new window]
FIG. 1.
Establishment of the motor pattern after bath application of pilocarpine to an isolated metathoracic ganglion. Motor output was monitored by extracellular recordings of nerves 5A and 3B of the same side and simultaneous intracellular recordings were made from a DUM 5 and a DUM 3,4 neuron. A: before pilocarpine application, the only active motor unit was the slow depressor trochanteris (Ds) in nerve 5A. Both DUM neurons received synaptic inputs but did not spike: the DUM 5 received both excitatory and inhibitory PSPs, whereas the DUM 3,4 received predominantly inhibitory PSPs. B: 5 min after pilocarpine application the motor rhythm consisted of alternate spiking of the levator trochanteris motor neurons (lev) and the common inhibitor 1 (CI1) with the slow depressor trochanteris (Ds). Slow extensor tibiae motor neuron (SETi) was not active. Both DUM neurons spiked but without any clear pattern. C: after 10 min, the motor pattern became fully established, and the lev spiked in phase with CI1 and in antiphase with the slow (Ds) and the fast (Df) depressors trochanteris. DUM 5 neuron now produced rhythmical bursts of spikes, whereas the DUM 3,4 neuron spiked sporadically.
View larger version (27K):
[in a new window]
FIG. 2.
Synaptic inputs to a DUM 3,4 neuron were linked closely to onset of the motor pattern. During the 1st 2 levator bursts the DUM 3,4 neuron produced 2 spikes superimposed on a hyperpolarization of the baseline membrane potential (shown at higher gain in the bottom set of traces). This characteristic waveform ( 
) differed from spikes occurring out of phase with the levator bursts ().
View this table:
TABLE 2.
Activity recorded in 79 different DUM cells from 48 preparations of a metathoracic ganglion treated with pilocarpine
20 to +20 s. The histogram was centered on the first spike event of the reference time series. The occurrence of spikes at different time lags in relation to this reference spike was represented by increasing the count of the relevant bins. The histogram then was centered on the next spike and the relevant bin counts increased. This process was repeated until the histogram had been centered on every spike of the reference time series. The histogram was not normalized but instead presented total event counts. In the autocorrelograms, the correlation of each event with itself at time 0 in the histogram is ignored, and periodic peaks and troughs with a large peak centered at lag = 0 indicated periodic clustering of spikes.
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
was characterized by two phases: a levator phase, during which the anterior coxal rotator, levator trochanteris, common inhibitor and flexor tibiae motor neurons spiked, and a depressor phase, during which the depressors trochanteris, SETi, and depressor tarsus motor neurons spiked.
View larger version (34K):
[in a new window]
FIG. 3.
Spike event autocorrelograms demonstrate the difference between bursting and unpatterned spike activity for 3 different types of DUM neuron. Autocorrelograms for DUMDL (top) and DUM 3 (middle) show no clear structuring. DUM 3,4,5 neuron (bottom) produced clearly separated bursts of spikes resulting in an autocorrelogram that had a prominent peak around lag 0. Period of bursting was fairly regular (~7 s), resulting in periodic peaks and troughs in the autocorrelogram. All DUM types have a narrow trough in their autocorrelograms at lag 0 because of their low spike frequencies (i.e., few spikes occur within 200 ms of each other). Data for DUM 3 and DUM 3,4,5 comes from a 170-s continuous recording from the same preparation, data for DUMDL from a 10-min continuous recording from another preparation. Bin width for autocorrelograms was 200 ms.
View larger version (26K):
[in a new window]
FIG. 4.
A: DUMDL spikes sporadically during the motor rhythm. A single levator burst cycle is marked (lev cycle). B: phase histograms of levator spikes (top) and DUMDL spikes (bottom) plotted in relation to the levator burst cycle. Levator spikes peak in the 1st half of the levator cycle, but DUMDL spikes occur throughout the cycle with no clear preferred phase. Data from 108 cycles over 10-min recording. Same preparation as Fig. 3, top.
View larger version (32K):
[in a new window]
FIG. 5.
Activity of DUM 3 and DUM 3,4 neurons during the motor pattern. Both neurons spike sporadically and share many common inputs, especially hyperpolarizing ones (dotted lines).
).
View larger version (28K):
[in a new window]
FIG. 6.
DUM 3,4,5 and DUMETi (DUM 5 innervating the extensor tibiae muscle) produce rhythmical bursts of spikes during a well-structured motor pattern, monitored by recordings of nerve 5A and nerve 3B of the same side. Each burst of spikes is superimposed on a wave of depolarization.
View larger version (56K):
[in a new window]
FIG. 7.
A: 2 DUM 5 neurons burst rhythmically and in phase with each other during the motor rhythm, monitored by extracellular recordings of nerve 5A (top) and nerve 3B (middle) of the same side. Lower DUM neuron generates more spikes despite being injected with some hyperpolarising current. B: autocorrelogram of the DUM 5* spike train has a pronounced peak at lag 0 bounded by 2 smaller troughs, indicating that this neuron produced bursts of spikes separated by gaps. DUM 5/DUM 5* cross-correlogram has a similar shape around lag 0, indicating that both DUM 5 neurons were bursting in phase with each other. Data from 12-min continuous recording.
). This suggests simultaneous excitation, giving rise to the pair of spikes, and inhibition, producing the hyperpolarization. The spikes do not seem to result from postinhibitory rebound because the neuron spikes during the hyperpolarization not after it. The resulting waveform was very different from that of other spikes immediately preceding and following this first levator burst (Fig. 2, ). Similar inputs were observed during the second levator burst, but by the third levator burst the DUM 3,4 cell showed only a hyperpolarization.
View larger version (44K):
[in a new window]
FIG. 8.
A: Activity of a DUM 3,4,5 and a DUM 3,4 neuron during a pilocarpine rhythm in which only 1 hemiganglion produced a structured output. Within this hemiganglion, the levators spiked synchronously with the common inhibitor 1 and alternately with the slow and fast depressor trochanteris motor neurons, whereas the other hemiganglion produced almost no activity (C). DUM 3,4,5 was depolarized and produced a burst of spikes during the levator phase ( 
), whereas DUM 3,4 stopped spiking during this phase. B: motor spikes and DUM spikes are plotted as phase histograms relative to the levator cycle of the rhythmically active hemiganglion. Slow depressor trochanteris motor neuron (DS) peaks shortly after the levator burst ends. Majority of DUM 3,4,5 spikes occurred during the levator burst, but there was also an increase from phase 0.5 onward, leading up to the next levator burst. DUM 3,4 shows the opposite spike pattern with few spikes during the levator bursts and frequent spiking during the rest of the cycle period. Data from 114 cycles during 15 min of continuous recording. C: recording of N3B on both sides following that in A showing a noncoordinated motor activity in which the levators do not strictly alternate on a cycle to cycle basis.
View larger version (41K):
[in a new window]
FIG. 9.
A: a DUM 3,4,5 neuron generates bursts of spikes around the levator phases ( ) of both sides of the ganglion during a motor pattern that was well coordinated across the ganglion. Extracellular recordings of nerves 3B, on the left and on the right, showed alternating levator and depressor phases within each hemiganglion. Levator bursts (lev) alternated from side to side and coincided with SETi bursts from the opposite side, as would be required for tripod walking. Phase histograms in B confirm that the DUM 3,4,5 neuron is more likely to spike during the levator bursts of either side. Left: occurrence of spikes plotted in relation to the levator cycle of the left side. Right: same data plotted in relation to the right levator cycle. Ipsilateral levators spike during the 1st half of the levator cycle of that side, contralateral levators spike out of phase during the 2nd half of the levator cycle. Because the DUM 3,4,5 neuron produces a burst of spikes around each levator burst from either side, the histogram of DUM spikes has 2 peaks that coincide with the ispilateral and contralateral levator peaks. Data from 74 cycles (left side) and 72 cycles (right side) during 13 min of continuous recording.
30 min, they showed many simultaneous changes in membrane potential that might have resulted from common synaptic inputs. In some cases (10/42), DUM 3 and DUM 3,4 neurons were observed to produce rhythmical bursts of spikes, although hyperpolarizing inputs were still present.
). This would tend to mask any coupling between DUM spiking patterns and motor output of just one side. Coupling between DUM spike activity and the motor pattern was observed in two types of preparation. In the first, and rarer kind, one side of the ganglion produced a well-structured motor pattern, whereas the other side produced a motor pattern with a less stable structure. Although variations did appear over long stretches of time, on a cycle-by-cycle basis there was no strict alternating activity with between the levators of both sides of the ganglion (Fig. 8C) In the second type, the motor output was well coordinated across both sides of the ganglion (Fig. 9) with levator activity in one hemiganglion either preceding or following levator activity of the other side with a short latency of 1-3 s, and never occurring at the same time.
View larger version (15K):
[in a new window]
FIG. 10.
Data from Fig. 7 plotted as phase histograms using a reference cycle that represents the joint output of both hemiganglia. Here a single cycle is defined as from the start of 1 levator burst to the start of the next, irrespective of the side from which the bursts originated. When plotted in this way, the levator spikes of both sides (both lev) fall within the 1st half of the cycle, indicating that there is always a gap between a levator burst from 1 side and the next levator burst from the other side, and that this gap is roughly equal to the levator burst duration. Double peak in DUM 3,4,5 spikes observed in Fig. 7B is converted into a single peak that coincides with the peak in levator spikes, confirming the link between the DUM 3,4,5 and the levator phases of either side. Data from 139 cycles during a 13-min recording.
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
and are characterized by two phases: a levator phase and a depressor phase, which alternate within each hemiganglia. Our results show that these locomotor-like rhythms expressed in leg motor neurons also involve modulation of the activity of efferent DUM neurons and that this rhythmical activation involves only a specific subset of DUM neurons.
). By contrast, in our study, the neuromodulators were efferent components and changes in the spike output of efferent DUM neurons proceeded in parallel with changes in the motor output. In fact, current injection into individual DUM neurons did not have any obvious effects on the centrally generated motor patterns: it did not change the frequency of the rhythms nor did it introduce new motor units into the rhythms.
. This strengthens the hypothesis that different DUM types have specific roles in particular motor behaviors (Burrows and Pflüger 1995). In all DUM neuron types, a general increase in spike activity could be observed. Such an increase could be induced either by muscarinic receptors on the DUM neurons or by the activation of presynaptic sources that provide a nonspecific excitatory drive. However, there were clear differences in the spiking activity of different DUM neuron types during pilocarpine-evoked motor patterns; this suggests that particular DUM neuron types are activated in parallel to the generation of activity in central-rhythm-generating networks. Thus our results provide clear evidence for a specific excitatory drive of efferent DUM neurons.
), but none of the target muscles for DUM 3 or DUM 3,4 neurons via nerve 3 are known (for review, see Stevenson and Spörhase-Eichmann 1995), making interpretation of their activity difficult. However, it is interesting to note that inhibition of DUM 3 and DUM 3,4 neurons also has been observed during the motor program underlying kicking of a hindleg (Burrows and Pflüger 1995).
). Moreover, the period of bursting was irregular and, in some cases, synchronized between the two recorded neurons. These observations are not consistent with the hypothesis that DUM cells become endogenous oscillators under pilocarpine treatment. Instead it seems as if their bursting activity results from the action of specific circuits presynaptic to the DUM neurons because some DUM neurons spiked tonically, whereas others showed rhythmical bursting activity.
).
, 1994), it is worth considering the functional implications of the observed patterns of DUM spike activity.
). Efferent DUM neurons can fulfill both these functions (Evans and O'Shea 1978; O'Shea and Evans 1979). According to this hypothesis, the precise timing of spikes in DUM 3,4,5 and DUM 5 neurons may not have functional significance. Instead, it may simply reflect the available mechanism of central coupling between these efferent neuromodulatory neurons and the rhythmical motor networks that at least ensures increased activity.
). During kicks of the hindleg, however, DUM neurons produced the same pattern of spikes regardless of whether a single leg or both legs kicked (Burrows and Pflüger 1995). Alternatively, we may need to consider central effects when interpreting patterns of DUM activity. Parker (1996) has shown that DUM neurons may have central effects on tibial motor neurons, although these effects appear variable. The final answer to the role of neuromodulation during specific motor patterns can only be addressed by recordings made during real behavior. Nevertheless, our study clearly demonstrates that DUM neurons can be activated individually in the control of movement and that such neuromodulatory processes can be coupled centrally to motor programs.
| |
ACKNOWLEDGEMENTS |
|---|
We thank M. Burrows, P. D. Evans, T. Matheson, P. L. Newland, H.-J. Pflüger, and two anonymous referees for helpful comments on the manuscript.
This work was supported by a studentship from the Wellcome Trust to S. Baudoux, National Institute of Neurological Disorders and Stroke Grant NS-16058 and a Wellcome Trust Grant to M. Burrows and a Deutsche Forschungsgesellschaft Grant (Graduiertenkolleg) to C. Duch.
| |
FOOTNOTES |
|---|
Address for reprint requests: S. Baudoux, Dept. of Zoology, University of Cambridge, Downing St., Cambridge CB2 3EJ, United Kingdom.
Received 3 March 1997; accepted in final form 16 September 1997.
| |
REFERENCES |
|---|
|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Mizutani, H. Ogawa, J. Saito, and K. Oka Fictive locomotion induced by octopamine in the earthworm J. Exp. Biol., January 15, 2002; 205(2): 265 - 271. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Gutovitz, J. T. Birmingham, J. A. Luther, D. J. Simon, and E. Marder GABA Enhances Transmission at an Excitatory Glutamatergic Synapse J. Neurosci., August 15, 2001; 21(16): 5935 - 5943. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Brezina, I. V. Orekhova, and K. R. Weiss Optimization of Rhythmic Behaviors by Modulation of the Neuromuscular Transform J Neurophysiol, January 1, 2000; 83(1): 260 - 279. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Morris, C Duch, and P. Stevenson Differential activation of octopaminergic (DUM) neurones via proprioceptors responding to flight muscle contractions in the locust J. Exp. Biol., January 12, 1999; 202(24): 3555 - 3564. [Abstract] [PDF]
|
|
|
|
|
|
R. Johnston, C Consoulas, H Pfl ger, and R. Levine Patterned activation of unpaired median neurons during fictive crawling in manduca sexta larvae J. Exp. Biol., January 1, 1999; 202(2): 103 - 113. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||
