| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biology, Georgia State University, Atlanta, Georgia 30303-3088
Submitted 4 September 2003; accepted in final form 26 April 2004
 |
ABSTRACT |
|---|
|
|
|
INTRODUCTION |
|---|
|
), thalamo-cortical function (Steriade 1999), and motor control (Calabrese 1995). A considerable amount is known about the mechanisms underlying cellular membrane potential oscillators (see Benson and Adams 1987; Jacklet 1989), but much less is understood about the mechanisms that underlie the generation of rhythmic activity by so-called network oscillators, which produce rhythmic patterns of activity as an emergent property of the interactions of neurons that lack intrinsic oscillatory properties (for examples, see Jacklet 1989). In some network oscillators, such as half-center oscillators, it has been shown that the spiking properties and the membrane potentials of component neurons help determine the periodicity of the oscillations (Calabrese 1995; Marder and Calabrese 1996). We tested whether such factors affect the periodicity of a slightly more complex network oscillator, the central pattern generator (CPG) underlying the escape swim CPG of the mollusc, Tritonia diomedea.
The Tritonia escape swim is a stereotyped fixed action pattern (Getting 1989a; Willows and Hoyle 1969). A single episode of escape swimming lasts up to a minute and consists of a series of alternating dorsal/ventral whole body flexions. It has been proposed that the Tritonia swim CPG is a pure network oscillator in which none of the component neurons exhibits intrinsic oscillatory activity (Getting 1989a). Rhythmic activity has been proposed to be produced through a combination of reciprocal inhibition combined with delayed excitation among the 3 CPG cell types (Getting 1989a).
Computer simulations of the CPG suggest that the cycle period is strongly influenced by the time course of synaptic potentials and spiking properties such as spike frequency adaptation (Getting 1983a, 1989b). However, previous results indicated that the periodicity of the motor pattern is not determined by the level of spiking activity in some CPG neurons (Fickbohm and Katz 2000). Here we more thoroughly investigate whether the membrane potential or activity levels of the CPG neurons play a role in determining the cycle frequency of the swim motor pattern.
We found that the cycle period is insensitive to the level of activity exhibited by individual neurons in the CPG, but that the Tritonia escape swim motor pattern can exhibit a range of cycle periods when the nervous system is exposed to different temperatures. These results have important implications for the mechanisms governing motor pattern generation in this model network oscillator.
Portions of this work were presented previously in abstract form (Katz et al. 2002).
|
|
METHODS |
|---|
|
Tritonia diomedea were obtained from Living Elements Ltd. (Delta, B.C., Canada) and maintained in recirculating artificial seawater (Instant Ocean; Aquarium Systems, Mentor, OH) at 11°C.
Electrophysiology
The Tritonia brain, composed of the fused cerebropleural and pedal ganglia, was dissected from the animal as previously described (Fickbohm and Katz 2000), dipped in 0.5% glutaraldehyde for 20–30 s to immobilize contractile elements in the ganglionic sheath, and pinned dorsal side up in a 35-mm petri dish lined with Sylgard (Dow Corning, Midland, MI). The ganglionic sheath was then mechanically removed with forceps and scissors. Normal saline, consisting of (in mM) 420 NaCl, 10 KCl, 10 CaCl2, 50 MgCl2, 10 D-glucose, and 10 HEPES, pH. 7.6, was continuously superfused over the ganglia using a gravity flow system at a rate of about 1 ml/min. The saline was chilled before entry into the petri dish by running the polyethylene inflow tube through a Plexiglas chamber surrounding the petri dish atop a Peltier cold plate. Temperature was monitored with a flexible probe (Physitemp, Clifton, NJ) positioned close to the ganglion in the dish and adjusted by changing the voltage to the Peltier device.
Intracellular recordings were obtained with glass microelectrodes (1–15 M
) filled with 3 M KCl. Signals were recorded using Axoclamp 2B amplifiers (Axon Instruments, Union City, CA) and digitized with a 1401plus A/D system and analyzed with Spike2 software (both from Cambridge Electronic Design, Cambridge, UK). To monitor and trigger the swim motor program, suction electrodes were placed on pedal nerve 3 (PDN3), PDN5, and PDN6 (Willows et al. 1973). Fictive swim motor programs were elicited by unilaterally stimulating (2 ms, 10 to 20 V pulses at 10 to 20 Hz for 1 s) PDN3, which contains the axons of sensory neurons.
The pathway for eliciting a swim motor pattern is illustrated in Fig. 1A. Sensory neurons converge on a trigger interneuron (Tr1) and a commandlike neuron, the dorsal ramp interneuron (DRI) (Frost and Katz 1996; Frost et al. 2001). DRI synapses on the dorsal swim interneurons (DSIs). The DSIs, together with the ventral swim interneurons (VSI-A, VSI-B) and interneuron C2 (cerebral neuron 2) form the central pattern generator (CPG) that produces the swim motor pattern (Getting 1989a). All neurons are bilaterally represented. The CPG neurons synapse on efferent flexion neurons: the ventral flexion neurons (VFN) and the dorsal flexion neurons (DFN-A, DFN-B).
|
, 1989a). Therefore VSI-A was not examined here except as a monitor of the swim motor pattern.
There are 3 DSIs in each hemisphere of the cerebral ganglion: DSI-A, DSI-B, and DSI-C (Getting et al. 1980). DSI-B and DSI-C are completely equivalent neurons and differ from DSI-A only slightly in their connectivity (Getting 1981). It is difficult to classify the DSIs without recording from all 3, so when we recorded from multiple DSIs, they were assigned an arbitrary number.
Current was injected into the somata of individual neurons through a balanced bridge circuit. Different amounts of positive and negative current were injected in different experiments in the range of 0.5 to 16 nA. The amount of current used in each experiment was dependent on the properties of the electrode and the cell. The maximum amount of current that could be injected without injuring the neuron was used. In all cases, the current had a substantial effect on the activity of the injected neuron; depolarizing current always evoked strong spiking responses. The precise extent of hyperpolarization produced in each case is not known because a bridge circuit was used. However, based on the input resistances of the neurons [C2: 13 M; DSI: 31 M; VSI: 15 M (Getting 1983a, 1989b)], it is expected that even with the minimum negative current used (–1 nA), the resulting hyperpolarization of the membrane potential in the soma should be around –13 mV. In most cases, the hyperpolarization was substantially greater.
To elicit tonic spiking, repeated 20 ms, 7 to 12 nA current pulses were injected at a set periodicity (5–20 Hz). Each current pulse elicited a single action potential. The success of stimulation was monitored either by observing the peak waveform of the action potentials, by observing the spike afterhyperpolarizations, or by recording one-for-one excitatory postsynaptic potentials (EPSPs) in follower neurons.
Swim cycle analysis
The cycle period progressively lengthens during the course of a single swim episode, starting around 7–8 s and ending around 10–12 s (Lennard et al. 1980). Therefore to compare motor patterns, which had varying numbers of cycles, we measured the initial cycle period (T). This was calculated as the time interval between the mean spike times of the first 2 complete bursts after the onset of the motor pattern (see Fig. 1D). The mean spike time was chosen because it is a more reliable marker than either the first or last spike. C2 was often used as the monitor of bursting because its bursts are the most regular, although equivalent results were obtained from any of the CPG neurons or the extracellular nerve recording.
All quantitative results are expressed as means ± SE. Paired t-tests were used to compare the effects of depolarization and hyperpolarization on the swim period.
Behavioral experiments
Behavioral data were obtained from 3 animals treated in parallel over a 3-week period. To evoke the escape swim behavior, animals were stimulated with 5 ml of a 5 M NaCl salt solution applied by use of a syringe over the posterior region of the body surface. The resulting swimming response was recorded by video camera and analyzed for cycle period. Animals were videotaped and the time was recorded on the tape with a date/time generator (WJ 810, Panasonic). A single cycle was defined as the time from one maximum ventral flexion until the next. Animals were stimulated 4 times at each test temperature at 10-min intervals. After each animal was tested at a given temperature, the tank temperature was adjusted in increments of 1°C. The animals were allowed 2.5 h to equilibrate to the new temperature.
|
|
RESULTS |
|---|
|
Cycle period was not affected by manipulating VSI-B membrane potential
There is a single VSI-B on each side of the brain. We tested the effect of changing VSI-B firing on the periodicity of the fictive swim motor pattern. A typical pattern of bursting activity was recorded intracellularly in CPG neurons, efferent flexion neurons, and extracellularly on pedal nerves in response to electrical stimulation of a body wall nerve (at arrow, Fig. 1B). VSI-B inhibits the other CPG members (see Fig. 1A) and strong tonic firing of a single VSI-B is sufficient to inhibit rhythmic motor pattern production (A. Sakurai, unpublished observation). A lower level of depolarizing current injected into VSI-B (here: 2 nA) caused the normally quiescent VSI-B to fire tonically, but did not suppress the motor pattern (Fig. 1C). The initial cycle period (T), as indicated by the DSI spike times, was not affected by the constant VSI-B firing (Fig. 1D).
The participation of the VSI-Bs is necessary for motor pattern generation: hyperpolarization of both VSI-Bs generally prevents the production of the swim motor pattern (Getting 1983b). Suppressing the firing of a single VSI-B with hyperpolarizing current injection did not halt motor pattern production, allowing us to study the effect of VSI-B activity on the rhythmicity of the motor pattern (Fig. 2). Beginning before the nerve stimulus, constant current (–7 nA) was injected into VSI-B, causing the membrane potential to hyperpolarize more than 50 mV (Fig. 2B, open arrow; Fig. 2C, red trace). Nerve stimulation caused normal bursting in the other CPG neurons; however, spiking in VSI-B was greatly attenuated and full spikes were completely absent from most of the cycles. Despite the reduction in VSI-B spiking, the initial cycle period was unaffected, as indicated by the C2 spike times (Fig. 2D).
|
Cycle period was not affected by manipulating DSI membrane potential
There are 6 DSIs, 3 on each side of the brain. We were able to inject current (1 to 5 nA) simultaneously into 2 or 3 of the 6 neurons to bias their membrane potentials either in the depolarizing or hyperpolarizing direction while simultaneously recording from CPG neuron C2 (Fig. 3). Tonic depolarization of the DSIs caused them to increase their rate of spiking before initiation of the swim motor program (Fig. 3B, open arrows). However, the initial cycle period was unaffected, as indicated by the mean spike times of the C2 bursts (Fig. 3D).
|
). However, in preparations in which hyperpolarization of 2 or 3 DSIs was not sufficient to prevent rhythmic motor pattern production, it had no effect on the initial cycle period, although it strongly reduced the DSI firing during the swim motor program and reduced the number of cycles (Fig. 3, C and D).
On average, the initial cycle period was not significantly affected by either depolarization or hyperpolarization of 2 or 3 DSIs (Fig. 3E). The average initial period under control conditions was 6.7 ± 0.6 s (n = 6). With the DSIs depolarized, the initial period was 6.9 ± 0.1 s, which was not a significant change from control values in those preparations (P = 0.5, n = 3, paired t-test). When the DSIs were hyperpolarized, the initial period was 6.7 ± 0.7 s, which also was not a significant change from control in those preparations (P = 0.8, n = 5, paired t-test). Thus biasing the membrane potentials of 2 or 3 DSIs did not alter the periodicity of the motor pattern, until it exceeded a certain level, at which point the motor pattern was completely suppressed (Getting and Dekin 1985).
Tonic DSI firing elicited bursting of the same periodicity as the nerve-evoked swim motor pattern
The DSIs are the only known targets of the commandlike neuron DRI (see Fig. 1A) (Frost and Katz 1996) and tonic activation of the DSIs has been shown to elicit a swim motor pattern in some preparations (Fickbohm and Katz 2000; Frost et al. 2001). In the set of experiments reported here, tonic DSI stimulation elicited rhythmic bursting activity in other CPG neurons in 20 out of 27 preparations tested. The spike frequency required to elicit bursting varied between preparations. Increasing the spike frequency or the number of DSIs firing sometimes allowed the bursting to occur (Fig. 4A). In the example shown in Fig. 4A, when one DSI was stimulated at 5 Hz, it caused a slight depolarization of the membrane potential of C2 and a slight inhibition of the firing in the other DSIs. Adding a second DSI at 5 Hz further excited C2, causing it to spike. The addition of a third DSI at 5 Hz caused C2 and other neurons recorded on PDN6 to fire action potentials in a rhythmic bursting pattern.
|
For all of the preparations examined, there was no significant difference in the initial period of bursting, regardless of the number of DSIs stimulated or their spike frequency (Fig. 4F). The average initial cycle period in response to nerve stimulation in these preparations was 6.9 ± 0.4 s (n = 9). When a single DSI was stimulated to fire at 10 or 20 Hz, the average initial period was 7.7 ± 0.5 and 6.8 ± 0.6 s, respectively, neither of which was significantly different from control (n = 4, P = 0.6 and 0.5, respectively, paired t-test). Stimulating 2 DSIs at 10 or 20 Hz did not significantly affect the initial cycle period (n = 4, mean values of 7.2 ± 0.5 and 6.3 ± 0.4 s, P values of 0.8 and 0.7, respectively). Thus although spiking in the DSIs could elicit bursting activity, the level of spiking activity in the DSIs did not determine the periodicity.
Cycle period was not affected by manipulating C2 membrane potential
Biasing the membrane potential of a single C2 in either depolarizing or hyperpolarizing direction altered the firing in that cell but did not affect the periodicity of the nerve-evoked motor pattern (Fig. 5). Injection of a maintained constant depolarizing current (1 to 5 nA) into one C2 caused it to begin firing tonically before nerve stimulation (Fig. 5B, open arrow) but did not affect the periodicity of bursting in the other CPG neurons during a nerve-evoked swim motor pattern (beginning at the solid arrow). In the same preparation, injection of hyperpolarizing current (–1 to –5 nA) into one C2 also did not significantly affect the burst period (Fig. 5C). Although C2 started from a more hyperpolarized potential, it rapidly depolarized to spike and remained depolarized throughout the motor pattern (Fig. 5C), resulting in a larger voltage excursion (Fig. 5D). The initial cycle period measured from the spike times of the contralateral C2 was unaffected by any level of current injection (Fig. 5E).
|
As with VSI-B, there is only a single C2 on each side of the brain. C2 is a necessary component of the CPG; hyperpolarization of both the left and right C2s with injection of as little as –0.5 nA prevented production of the swim motor pattern as previously reported (Getting and Dekin 1985). In the example shown in Fig. 6, A–C, –1 nA injected into both C2s prevented the swim motor pattern (Fig. 6B). However, injection of –1 nA into just a single C2 did not alter the periodicity of the motor pattern, although it decreased the burst duration of the injected cell and increased the amplitude of its membrane potential oscillations (Fig. 6C).
|
C2-triggered motor patterns had the same periodicity as nerve-evoked motor patterns
It was previously reported that brief depolarization of C2 is sufficient to trigger a swim motor pattern (Taghert and Willows 1978). We observed this only in a minority of preparations (7 out of 28 in this study). However, when it occurred, the periodicity of the motor pattern evoked in this fashion was very similar to that caused by nerve stimulation (Fig. 7). Although self-sustained motor patterns caused by brief C2 stimulation were rarely observed, it was previously reported that maintained current injection into C2 also can elicit a swim motor pattern for as long as C2 is depolarized (Getting 1977). We obtained this response in 2 preparations and observed that the periodicity of the swim motor pattern produced in this manner was also similar to that caused by nerve stimulation (data not shown). Thus the periodicity of the motor pattern was the same when triggered by C2 depolarization or by stimulation of the afferent axons in the body wall nerve.
|
In some preparations, we were able to elicit rhythmic bursting in CPG neurons when C2 was stimulated to fire tonically, using repetitive 20-ms current pulses (5 to 12 nA at either 10 or 20 Hz). Tonic firing in C2 caused rhythmic activity in 6 of 11 preparations where it was tested (Fig. 8). In each case, the periodicity of the motor pattern elicited by tonic C2 firing was similar to that caused by nerve stimulation regardless of the frequency of C2 spiking (Fig. 8, C and D). Thus the periodicity of the motor pattern was not dependent on the level of spiking activity exhibited by a single C2.
|
|
None of the manipulations of CPG neuron activity in this study caused any change in the periodicity of the swim motor pattern. It is possible that, unlike other known CPGs, this oscillator might be capable of operating at only one cycle period. In an effort to see whether the CPG was capable of producing different cycle periods, we tested its performance at varying temperatures.
We found that the ability to evoke a swim motor pattern was temperature sensitive. The percentage of nerve stimuli that failed to evoke a swim motor pattern increased at the extremes of the temperature range tested (6–20°C). However, if a swim motor pattern was evoked, then the cycle period of the swim motor pattern was strongly temperature dependent (Fig. 10). At 5°C, the initial cycle period in the example shown in Fig. 10A was 11.9 s (see also Fig. 10D). It decreased to 7.2 s at 10°C (Fig. 10, B and D), and decreased further to 4.2 s at 15°C (Fig. 10, C and D). The average initial cycle period over the whole temperature range for all preparations tested is shown in Fig. 10E. There was a progressive decrease in cycle period with increasing temperature. Thus unlike changes to the membrane potentials or spiking levels of the CPG neurons, varying the temperature profoundly affected the periodicity of the motor pattern.
|
Although the cycle period of the motor pattern was greatly affected by temperature, the percentage of each period that the neurons were active remained constant. The burst duration of the swim interneurons increased as the period increased (Fig. 10, A–C, Fig. 11). In addition, the phasing between dorsal and ventral phase interneurons was unaffected by temperature (Fig. 11). Thus the shapes of the bursts and the phasing of the components did not differ despite the large difference in cycle period.
|
|
|
DISCUSSION |
|---|
|
). In an effort to better understand motor pattern generation in this model system, we manipulated the membrane potentials and spiking activity of component neurons in the CPG. To our surprise, we found that manipulating the membrane potentials or levels of spiking activity of CPG interneurons over a wide range of values did not affect the periodicity of the swim motor pattern. Our expectation was that the period of oscillation would be influenced by the level of activity of the CPG neurons. For example, one simple scenario is that if the membrane potentials of the CPG neurons were closer to threshold or, indeed, above threshold, then the period ought to decrease because it would take less time to begin the next cycle, whereas if the membrane potentials were farther from threshold, then it should take longer to depolarize and thus increase the period. Or alternatively, depolarizing a neuron might have increased the period by increasing the inhibition on other neurons. However, when any of the CPG neurons, even the normally silent C2 and VSI-B, was depolarized to the point of spiking, the cycle period was unaffected. Conversely, when any of these neurons was hyperpolarized, it did not slow down the motor pattern, but merely caused the hyperpolarized neuron to exhibit a greater voltage excursion during the bursts (Fig. 2C, Fig. 5D). These results suggest either that the circuit can compensate for differences in activity of component neurons or that activity levels of these neurons do not play important roles in determining the periodicity of the motor pattern.
It is possible that current injection into the cell body does not affect the membrane potential at important integrative areas that are electrically distant from the soma. Previous work suggested that the length constant for the DSIs was about 2 mm (Getting 1983a), which is approximately the distance from the soma to the farthest point in the pedal ganglion. Therefore based on this measurement, even if the synapses were located as distant as physically possible, one would expect a maximum voltage attenuation of 36% of the value in the soma. Thus a –50 mV hyperpolarization in the soma would still produce a substantial –18 mV change in these distant sites. The ability to affect distant sites in the neuron using current injection into the soma is further suggested by the fact that hyperpolarization of the neurons decreased or eliminated spiking in all cases. Furthermore, hyperpolarization of both C2s with as little as –0.5 nA of current was able to suppress the motor pattern, whereas higher levels of current injected into a single C2 had no effect on periodicity. Thus the lack of an effect of current injection on cycle period cannot be attributed to an inability to alter the membrane potential at distant synapses.
The inability to prevent motor pattern generation by strong hyperpolarization of a single C2 or VSI-B indicates that the contralateral counterpart is capable of maintaining the rhythm without its partner. Thus although activity in at least one cell is necessary for motor pattern generation, altering the membrane potential of either cell did not affect the periodicity, whereas hyperpolarizing both cells often caused the pattern to cease.
The periodicity was not affected by the spike frequency in DSI or C2 when these neurons were driven to fire at defined frequencies. With the DSIs, we were able to stimulate over a wide range of frequencies (5–20 Hz) and include one to 3 neurons, effectively producing a 12-fold change in the total number of spikes per second. Regardless of the DSI tonic spike frequency, the cycle period of the resulting swim motor pattern was consistent. Thus the cycle period was independent of the level of spiking activity in DSI or C2. This suggests further that the periodicity is not controlled by the sensory or command inputs to the CPG since stimulating the DSIs and C2 bypasses the input pathway (Fig. 1A).
A permissive role for DSI and C2 in rhythmogenesis
DSI and C2 are necessary components of the CPG. Strong hyperpolarization of both C2s or of 3 DSIs blocks motor pattern generation (Getting and Dekin 1985). Furthermore, although the swim motor pattern could be produced with one C2 hyperpolarized, interfering with G-protein signaling in a single C2 is sufficient to halt the motor pattern (Clemens and Katz 2003). However, we found that changing the spiking activity of these neurons within the range tested did not affect the periodicity of the motor pattern. These results suggest that DSI and C2 each have a permissive, gating effect on the oscillator. That is, their spiking activity is necessary to allow the motor pattern to proceed, but their individual membrane potentials and temporal spiking patterns are not directly involved in generating the oscillations.
A permissive rather than rhythmic role is supported by the observation that the motor pattern can be elicited by tonic firing in either DSI or C2. When a single DSI is made to fire at a constant frequency, the other DSIs are often silent (see Fig. 4C), presumably because of recruitment of I-cell inhibition (Getting and Dekin 1985). Thus it is unlikely that rhythmic activity arises as a consequence of the bursting in other DSIs. It is possible that the tonic firing in C2 and DSI is sculpted by rhythmic presynaptic inhibition. That is, although the neurons fire tonically, they could be releasing neurotransmitter rhythmically onto other CPG neurons. An underlying membrane potential oscillation, which could serve to gate synaptic output, was often observed during periods of tonic spiking (see Fig. 8B). Such a translation of tonic activity into phasic bursting has been reported in the lobster stomatogastric system (Coleman and Nusbaum 1994).
A permissive role for DSI activity in rhythmic motor pattern production is further supported by the behavior of the DSIs after treatment with the serotonin precursor, 5-hydroxytryptophan (5-HTP). 5-HTP treatment increases the serotonin content of the DSIs and potentiates the actions of these serotonergic neurons (Fickbohm and Katz 2000). However, after 5-HTP treatment, the DSIs drop out of the nerve-evoked motor pattern after initially bursting, perhaps because of enhanced recruitment of inhibition, whereas other neurons continue to burst. This suggests that serotonin (5-HT) released from the DSIs is sufficient to elicit the motor pattern and that bursting may be merely a mechanism for delivering enough 5-HT to the oscillator; bath application of serotonin produces a swim motor program with a periodicity similar to that of the swim evoked by stimulation of afferent nerves (McClellan et al. 1994).
Temperature sensitivity
After determining that the activity levels of CPG neurons did not affect the periodicity of the motor pattern, we looked for other factors that might play a role. No neuromodulatory substances have been found to affect the period of the motor pattern (McClellan et al. 1994). We observed that the period of the fictive motor pattern was strongly dependent on temperature; there was a 3-fold decrease in period with a 10°C increase in temperature. The cycle period of the motor pattern produced by the isolated nervous system exhibited a temperature sensitivity similar to that of the animal's behavior, suggesting that the sensitivity of the behavior to changes in temperature are attributed to mechanisms in the CNS rather than neuromuscular or sensory responsiveness. Furthermore, the temperature sensitivity was apparent even when the motor pattern was elicited from within the circuit (by stimulating C2), suggesting a temperature-sensitive mechanism intrinsic to the CPG, rather than a temperature-sensitive afferent input.
Although temperature altered the cycle period, the phase relationships of the CPG neurons did not change with temperature. This suggests that the mechanisms underlying the production of the rhythmic activity are likely to be the same at the different temperatures and that the cause of the change in period is not attributed to a shift to a different mode of oscillation. Thus the CPG is capable of producing a wide range of periods, although this periodicity is not determined by the electrical activity of the CPG neurons.
The periodicities of many rhythmic behaviors, particularly those produced by poikilotherms, show a strong dependency on temperature. Examples are widespread and diverse and include Aplysia brasiliana swimming (von der Porten et al. 1982), locust flight (Xu and Robertson 1994), electric fish communication signals (Dunlap et al. 2000), and cricket and cicada calling songs (Fonseca and Revez 2002; Pires and Hoy 1992). The oscillation frequency of individual bursting pacemaker neurons also shows temperature sensitivity resulting from the temperature sensitivity of ion channels (Johnson et al. 1992; Thompson et al. 1986; Treistman and Bablanian 1985). Many neuronal processes, such as synaptic transmission, action potential duration, and enzyme activity, are temperature dependent (Hakozaki et al. 1989; Janssen 1992; Johnson et al. 1991; Joyner 1981; Montgomery and Macdonald 1990; Thompson et al. 1986). Thus the fact that the cycle frequency of the Tritonia swim motor pattern is temperature sensitive does not help elucidate the mechanisms underlying frequency regulation in this system; it merely establishes that the system is capable of oscillating at different periodicities.
Where is the oscillator?
Our results appear enigmatic because the CPG neurons are all necessary for the motor pattern, yet none is capable of altering the periodicity of the pattern by itself. It is possible that there are important neurons missing from the known CPG circuit. For example, the unidentified I-cells whose inhibitory PSPs are recruited by the DSIs (Getting and Dekin 1985) could be a part of the network oscillator.
An alternate hypothesis is that the oscillations are based on the dynamics of second-messenger signaling. Getting's simulation of the circuit suggested an important role for the time courses of slow synaptic potentials in setting the periodicity of the motor pattern (Getting 1989a,b). Recent work demonstrates that changes in the time course of synaptic potentials in a CPG can have a greater effect on cycle period than changes in their amplitude (Prinz et al. 2003).
Slow synaptic potentials in the Tritonia CPG are likely to be mediated by second-messenger signaling (Clemens and Katz 2001) and G-protein signaling is necessary for the production of the motor pattern (Clemens and Katz 2003). Thus the reason that the oscillator is insensitive to the resting potential level could be that it is a biochemical oscillator (Goldbeter 1996) rather than a voltage-controlled oscillator (Benson and Adams 1987). Calcium oscillations can arise through an interaction of intracellular calcium stores with membrane Ca2+ channels (Friel 1995). Similarly, it has been proposed that cAMP levels can oscillate with a periodicity of about 10 s (Cooper et al. 1995), which is close to the period of the Tritonia swim motor pattern. Such biochemically based oscillators would be temperature sensitive and might require spiking activity in CPG neurons to allow a Ca2+ influx, but would not be easily perturbed by voltage. Further experiments are needed to test this intriguing hypothesis.
Other rhythmic behaviors that have been hypothesized to be produced by network oscillators include leech swimming (Brodfuehrer and Thorogood 2001), Clione swimming (Arshavsky et al. 1998), Xenopus embryo swimming (Dale and Kuenzi 1997), lamprey swimming (Arshavsky et al. 1993), gastropod feeding (Elliott and Susswein 2002), the crustacean gastric mill movements (Manor et al. 1999), and mammalian respiration (Del Negro et al. 2002). These networks differ in the number of neurons involved and in their complexity. Some are too complex to test the roles of spiking activity in individual CPG neurons. The Tritonia swim CPG is of intermediate complexity; it is more complex than a half-center oscillator, but still has a manageable number of neurons. The insights into the control of rhythmic pattern generation gained by the study of this model system are likely to be applicable to other network oscillators.
|
|
GRANTS |
|---|
|
|
|
ACKNOWLEDGMENTS |
|---|
|
Present address of S. Clemens: Emory University School of Medicine, Department of Physiology, 615 Michael St., Atlanta, GA 30322 (E-mail: scleme2@emory.edu).
|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: P. S. Katz, Department of Biology, Georgia State University, MSC 8L0389, 33 Gilmer St. SE, Unit 8, Atlanta, GA 30303-3088 (E-mail: pkatz{at}gsu.edu).
|
|
REFERENCES |
|---|
|
Arshavsky YI, Orlovsky GN, Panchin YV, Roberts A, and Soffe SR. Neuronal control of swimming locomotion: analysis of the pteropod mollusc Clione and embryos of the amphibian Xenopus. Trends Neurosci 16: 227–233, 1993.[CrossRef][Web of Science][Medline]
Benson JA and Adams WB. The control of rhythmic neuronal firing. In: Neuromodulation: The Biochemical Control of Neuronal Excitability, edited by LK Kaczmarek and IB Levitan. New York: Oxford Univ. Press, 1987, p. 100–118.
Brodfuehrer PD and Thorogood MSE. Identified neurons and leech swimming behavior. Prog Neurobiol 63: 371–381, 2001.[CrossRef][Web of Science][Medline]
Calabrese RL. Oscillation in motor pattern-generating networks. Curr Opin Neurobiol 5: 816–823, 1995.[CrossRef][Web of Science][Medline]
Clemens S and Katz PS. Identified serotonergic neurons in the Tritonia swim CPG activate both ionotropic and metabotropic receptors. J Neurophysiol 85: 476–479, 2001.
Clemens S and Katz PS. G-protein signaling in a neuronal network is necessary for rhythmic motor pattern production. J Neurophysiol 89: 762–772, 2003.
Coleman MJ and Nusbaum MP. Functional consequences of compartmentalization of synaptic input. J Neurosci 14: 6544–6552, 1994.[Abstract]
Cooper DMF, Mons N, and Karpen JW. Adenylyl cyclases and the interaction between calcium and cAMP signalling. Nature 374: 421–424, 1995.[CrossRef][Medline]
Dale N and Kuenzi FM. Ion channels and the control of swimming in the Xenopus embryo. Prog Neurobiol 53: 729–756, 1997.[CrossRef][Web of Science][Medline]
Del Negro CA, Morgado-Valle C, and Feldman JL. Respiratory rhythm: an emergent network property? Neuron 34: 821–830, 2002.[CrossRef][Web of Science][Medline]
Dunlap KD, Smith GT, and Yekta A. Temperature dependence of electrocommunication signals and their underlying neural rhythms in the weakly electric fish, Apteronotus leptorhynchus. Brain Behav Evol 55: 152–162, 2000.[CrossRef][Web of Science][Medline]
Elliott CJH and Susswein AJ. Comparative neuroethology of feeding control in molluscs. J Exp Biol 205: 877–896, 2002.
Fickbohm DJ and Katz PS. Paradoxical actions of the serotonin precursor 5-hydroxytryptophan on the activity of identified serotonergic neurons in a simple motor circuit. J Neurosci 20: 1622–1634, 2000.
Fonseca PJ and Revez MA. Temperature dependence of cicada songs (Homoptera, Cicadoidea). J Comp Physiol A Neuroethol Sens Neural Behav Physiol 187: 971–976, 2002.[CrossRef][Web of Science][Medline]
Friel DD. Calcium oscillations in neurons. Ciba Found Symp 188: 210–223, 1995.[Medline]
Frost WN, Hoppe TA, Wang J, and Tian LM. Swim initiation neurons in Tritonia diomedea. Am Zool 41: 952–961, 2001.[CrossRef]
Frost WN and Katz PS. Single neuron control over a complex motor program. Proc Natl Acad Sci USA 93: 422–426, 1996.
Getting PA. Neuronal organization of escape swimming in Tritonia. J Comp Physiol A Sens Neural Behav Phsyiol 121: 325–342, 1977.[CrossRef]
Getting PA. Mechanisms of pattern generation underlying swimming in Tritonia. I. Neuronal network formed by monosynaptic connections. J Neurophysiol 46: 65–79, 1981.
Getting PA. Mechanisms of pattern generation underlying swimming in Tritonia. II. Network reconstruction. J Neurophysiol 49: 1017–1035, 1983a.
Getting PA. Mechanisms of pattern generation underlying swimming in Tritonia. III. Intrinsic and synaptic mechanisms for delayed excitation. J Neurophysiol 49: 1036–1050, 1983b.
Getting PA. A network oscillator underlying swimming in Tritonia. In: Neuronal and Cellular Oscillators, edited by JW Jacklet. New York: Marcel Dekker, 1989a, p. 215–236.
Getting PA. Reconstruction of small neural networks. In: Methods in Neuronal Modeling: From Synapses to Networks, edited by C. Koch and I. Segev. Cambridge, MA: MIT Press, 1989b, p. 171–194.
Getting PA and Dekin MS. Mechanisms of pattern generation underlying swimming in Tritonia. IV. Gating of central pattern generator. J Neurophysiol 53: 466–480, 1985.
Getting PA, Lennard PR, and Hume RI. Central pattern generator mediating swimming in Tritonia. I. Identification and synaptic interactions. J Neurophysiol 44: 151–164, 1980.
Goldbeter A. Biochemical Oscillations and Cellular Rhythms. Cambridge, UK: Cambridge Univ. Press, 1996.
Hakozaki S, Matsumoto M, and Sasaki K. Temperature-sensitive activation of G-protein regulating the resting membrane conductance of Aplysia neurons. Jpn J Physiol 39: 115–130, 1989.
Jacklet JW. Neuronal and Cellular Oscillators. New York: Marcel Dekker, 1989.
Janssen R. Thermal influences on nervous system function. Neurosci Biobehav Rev 16: 399–413, 1992.[CrossRef][Web of Science][Medline]
Johnson BR, Peck JH, and Harris-Warrick RM. Temperature sensitivity of graded synaptic transmission in the lobster stomatogastric ganglion. J Exp Biol 156: 267–285, 1991.
Johnson BR, Peck JH, and Harris-Warrick RM. Elevated termperature alters the ionic dependence of amine-induced pacemaker activity in a conditional burster neuron. J Comp Physiol A Sens Neural Behav Physiol 170: 201–209, 1992.[Medline]
Joyner RW. Temperature effects on neuronal elements. Fed Proc 40: 2814–2818, 1981.[Web of Science][Medline]
Katz PS, Sakurai A, Clemens S, and Davis D. Cycle period regulation in the Tritonia escape swim. Soc Neurosci Abstr 67.5: 2002.
Lennard PR, Getting PA, and Hume RI. Central pattern generator mediating swimming in Tritonia. II. Initiation, maintenance, and termination. J Neurophysiol 44: 165–173, 1980.
Manor Y, Nadim F, Epstein S, Ritt J, Marder E, and Kopell N. Network oscillations generated by balancing graded asymmetric reciprocal inhibition in passive neurons. J Neurosci 19: 2765–2779, 1999.
Marder E and Calabrese RL. Principles of rhythmic motor pattern generation. Physiol Rev 76: 687–717, 1996.
McClellan AD, Brown GD, and Getting PA. Modulation of swimming in Tritonia: excitatory and inhibitory effects of serotonin. J Comp Physiol A Sens Neural Behav Physiol 174: 257–266, 1994.[Medline]
Montgomery JC and Macdonald JA. Effects of temperature on nervous system: implications for behavioral performance. Am J Physiol Regul Integr Comp Physiol 259: R191–R196, 1990.
Perez-Orive J, Mazor O, Turner GC, Cassenaer S, Wilson RI, and Laurent G. Oscillations and sparsening of odor representations in the mushroom body. Science 297: 359–365, 2002.
Pires A and Hoy RR. Temperature coupling in cricket acoustic communication. II. Localization of temperature effects on song production and recognition networks in Gryllus firmus. J Comp Physiol A Sens Neural Behav Physiol 171: 79–92, 1992.[Medline]
Prinz AA, Thirumalai V, and Marder E. The functional consequences of changes in the strength and duration of synaptic inputs to oscillatory neurons. J Neurosci 23: 943–954, 2003.
Steriade M. Coherent oscillations and short-term plasticity in corticothalamic networks. Trends Neurosci 22: 337–345, 1999.[CrossRef][Web of Science][Medline]
Taghert PH and Willows AOD. Control of a fixed action pattern by single, central neurons in the marine mollusk, Tritonia diomedea. J Comp Physiol 123: 253–259, 1978.[CrossRef]
Thompson S, Smith SJ, and Johnson JW. Slow outward tail currents in molluscan bursting pacemaker neurons: two components differing in temperature sensitivity. J Neurosci 6: 3169–3176, 1986.[Abstract]
Treistman SN and Bablanian GM. Effects of sea water temperature on bursting pacemaker activity in cell R15 in the intact Aplysia. Brain Res 346: 155–159, 1985.[CrossRef][Web of Science][Medline]
von der Porten K, Parsons DW, Rothman BS, and Pinsker H. Swimming in Aplysia brasiliana: analysis of behavior and neuronal pathways. Behav Neural Biol 36: 1–23, 1982.[CrossRef][Web of Science][Medline]
Willows AOD, Dorsett DA, and Hoyle G. The neuronal basis of behavior in Tritonia. I. Functional organization of the central nervous system. J Neurobiol 4: 207–237, 1973.[CrossRef][Web of Science][Medline]
Willows AOD and Hoyle G. Neuronal network triggering of fixed action pattern. Science 166: 1549–1551, 1969.
Xu H and Robertson RM. Effects of temperature on properties of flight neurons in the locust. J Comp Physiol A Sens Neural Behav Physiol 175: 193–202, 1994.[CrossRef]
This article has been cited by other articles:
|
|
 |
|
  J. M Newcomb and P. S Katz Different functions for homologous serotonergic interneurons and serotonin in species-specific rhythmic behaviours Proc R Soc B, January 7, 2009; 276(1654): 99 - 108. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Yamaguchi, D. Gooler, A. Herrold, S. Patel, and W. W. Pong Temperature-Dependent Regulation of Vocal Pattern Generator J Neurophysiol, December 1, 2008; 100(6): 3134 - 3143. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Calin-Jageman, M. J. Tunstall, B. D. Mensh, P. S. Katz, and W. N. Frost Parameter Space Analysis Suggests Multi-Site Plasticity Contributes to Motor Pattern Initiation in Tritonia J Neurophysiol, October 1, 2007; 98(4): 2382 - 2398. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. J. Norris, A. L. Weaver, L. G. Morris, A. Wenning, P. A. Garcia, and R. L. Calabrese A Central Pattern Generator Producing Alternative Outputs: Temporal Pattern of Premotor Activity J Neurophysiol, July 1, 2006; 96(1): 309 - 326. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
TOP
ABSTRACT
INTRODUCTION
