| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Zoological Institute, University of Cologne, 50923 Cologne; and 2Deparment of Biological Cybernetics, University of Bielefeld, 33501 Bielefeld, Germany
Submitted 30 December 2003; accepted in final form 1 March 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Pearson 1993; Prochazka 1996).
It is well known that load or force information is an important sensory parameter used in the generation of a functional walking motor pattern. This information serves in feedback loops in the control of the muscle forces during the stance phase and is used for adapting the motor output to the actual requirements of force production (e.g., cockroach: Pearson 1972; stick insect: Graham 1985; cat: Duysens and Pearson 1980; Orlovsky and Feldman 1972). Previous investigations have analyzed how changes in load during walking can affect the walking motor pattern (review: Duysens et al. 2000). In insects, the main load receptors are the campaniform sensilla (CS), sense organs that are activated by strains in the cuticle. These can result either from load on the leg or from self-generated forces of the muscles acting on the cuticle (Delcomyn 1991; Hofmann and Bässler 1982, 1986; Pringle 1938; Wendler 1964). There is a large amount of information on how CS respond to load or force changes in the leg. The response of CS depends on the direction and magnitude of a given load or force stimulus on the cuticle (Delcomyn 1991; Ridgel et al. 2000, 2001; Zill et al. 1999). Noah et al. (2001) showed for the cockroach that during walking, activity of tibial CS is induced at the onset of the stance and declines while the body weight, carried by the leg, decreases in the course of the stance. In the stick insect walking system, information from CS fields on the trochanterofemur section has been shown to have substantial influence on the motor output of the leg (Bässler 1977; Schmitz 1993). For the stick insect species studied so far, there are four fields of CS, one of which is located on the proximal femur and three are located on the trochanter (Hofmann and Bässler 1986; Schmitz et al. 1991). Akay and coauthors (2001) have shown that the signals from the group of CS located on the proximal femur, the femoral CS (fCS), can reinforce flexor tibiae motoneuronal activity during stance phase. The influence of the fCS signals onto the FTi-joint motoneurons appears to be transmitted both via reflex-pathways modulating the strength of the motor output and by pathways influencing the timing of the pattern generating networks governing the FTi-joint. The three groups of CS located on the trochanter, the trochanteral CS (trCS), do not appear to contribute to this influence. However, there are reports that signals from the trCS influence motoneuronal activity of the thoraco-coxal (TC)-joint without contribution from the fCS (Akay et al. 2001; Schmitz 1993). This leads to the hypothesis that the four groups of CS are functionally segregated in their actions (Akay et al. 2001); this hypothesis significantly differs from the previously implied consideration that trCS and fCS serve as one sensory system for controlling the leg motor output in general (Bässler 1977; Hofmann and Bässler 1982).
In the present investigation, we have set out to test in detail the influence that signals from the trCS have in sculpturing the motor output of the most proximal leg joint, the TC-joint of walking insects. During the stance and swing phase of a leg in freely walking stick insects, back and forth movement of the leg is controlled by the retractor coxae (RetCx) and protractor coxae (ProCx) motoneurons and muscles of the most proximal TC-joint. Previous investigations have shown that the ProCx is innervated at least by eight excitatory motoneurons and the RetCx by seven excitatory motoneurons (e.g., Graham and Wendler 1981). For the generation of useful trajectories, the activities of the antagonistic TC-joint motoneuron groups have to be tightly coupled to the movement of the more distal leg segments, i.e., the femur, the tibia, and the tarsus.
A similar coupling of TC-joint motoneuronal activity was reported in the single-middle-leg preparation of the stick insect, in which one middle leg generates walking-like movements on a treadband; however, with only the CTr- and FTi-joint free to move (Fischer et al. 2001). In this single-middle-leg preparation, in which the TC-joint is fixed, de-efferented, and -afferented the activity of the TC-joint motoneurons is still coupled to walking movements performed by the more distal leg joints. The activity switches from protractor to RetCx motoneuronal activity at the start of the stance phase and switches back from retractor to ProCx motoneurons at the end of the stance and the onset of swing phase. It is yet unknown which influence is underlying this phenomenon.
It is possible to test the effects of selective ablation of CS groups on walking in the single-middle-leg preparation as these animals show movements with clearly distinguishable stance and swing phases. Changes in the activity of TC-joint motoneurons after CS ablation can be determined without having any background effects due to intrajoint influences from the TC-joint sense organs. By sequential ablation of the trCS and fCS in walking preparations and with exclusive stimulation of trCS in reduced preparations, we show that indeed the actions of the four groups of CS are functionally segregated and that sensory signals from the trCS affect the timing of the activity in motoneuron pools supplying the TC-joint. In summary, signals from the trCS control the motor activity in the proximal TC-joint and signals from the fCS influence control of motoneuronal activity of the distal FTi-joint.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Preparation
All legs except for the left middle leg were cut at the level of mid-coxa. The animals were fixed with a dental cement (Protemp II, ESPE) dorsal side up along the edge of a foam platform with only the left middle leg left attached to the body. The TC-joint was fixed with dental cement perpendicular to the body axis so that the leg was extending over the rim of the platform. Care was taken to leave the coxa-trochanter (CTr)-joint free to move and not to cover the cuticular strain measuring sense organs, the CS located on the trochanter (trCS) and on the proximal femur (fCS), close to the trochanter (Delcomyn 1991; Hofmann and Bässler 1986). A window was cut into the tergum at the level of the mesothoracic legs. The gut, connective tissues, and fat were removed to expose the ventral nerve cord. The thoracic cavity was filled with stick insect saline (Weidler and Diecke 1969). The lateral nerves nl2 and nl5, carrying the motor axons innervating the ProCx and the RetCx muscles, which move the TC-joint forward and backward, were crushed distally near their respective muscles. Extracellular recordings were made with hook electrodes (Schmitz et al. 1991) proximal to the crush point of nerves nl2 and nl5 to monitor the activity of the ProCx and RetCx motoneuron pools.
In some experiments, intracellular recordings were performed of the neuropilar processes of ProCx or RetCx motoneurons. To do so, the ganglion was prepared according to established procedures (Büschges 1989). In brief, the ganglion was fixed on a wax-coated ganglion holder by use of fine cactus needles of Nopalea dejecta. The ganglionic sheath was treated with pronase (Merck KG, Darmstadt) for 60 s to allow electrode penetration. Intracellular recordings were performed with thin-walled sharp microelectrodes in the bridge mode of an intracellular DC-amplifier (SEC-10L, NPI, Tamm, Germany). The electrodes had resistances between 15 and 25 M
when filled with 3 M KAc/0.05 M KCl. The motoneurons were identified by a one-to-one correlation of their intracellular recorded action potentials with spikes in the extracellular recordings from the lateral nerves nl2 and nl5, respectively.
Under the preceding conditions, the middle leg was able to perform stepping-like movements on a treadband with the two free joints, the CTr- and the FTi-joints, (for more details, see Fischer et al. 2001; Gabriel et al. 2003). Sequences of stepping movements were elicited by touching the abdomen of the animal with a paint brush.
During the experiments, the animals were tested under different sensory conditions with the CS being destroyed in the course of the experiment. This was made possible as the whole experimental setup, including the foam platform with the animal, treadband and the hook electrodes, could be rotated without changing their relative positions to each other. To achieve this, the experimental setup was positioned on a turnable metal plate. Thus it was possible to have view and access to all groups of CS on the trochanterofemur. Individual groups of CS were destroyed by pushing a fine insect pin through the pit in the cuticle where the CS are located (Schmitz 1993). Afterward walking sequences were again elicited. In six experiments, all trochantero-femoral CS groups were ablated at once, whereas in other experiments, the CS groups were destroyed sequentially. In four experiments, the three groups of trCS were destroyed first, followed by fCS destruction. In four other experiments, the fCS were destroyed first, followed by trCS destruction.
Experiments on restrained animals
In one set of experiments, the effect of the femoro-trochanteral CS on the TC-joint motoneuronal activity was tested in the partially denervated mesothoracic ganglion, which was denervated except for the innervation of the CS and the anterior and posterior connective nerves. To do so, the lateral nerves were cut except for the nervus cruris (ncr). Nerves originating from the ncr in the coxa, nerves C1 and C2, were also cut. In addition, the ncr was also cut within the femur just distally to the fCS. After this denervation procedure, all sensory input from and motor output to the periphery were excluded except for inputs from the trCS and the fCS. In these preparations, the left middle leg was fixed in a way that it extended over the platform. The femur was attached to a piezoelectric element that was driven by a ramp generator (electronic workshop, University of Cologne). To adequately stimulate the CS, the femur was slightly bent horizontally in caudal and rostral direction with amplitudes of 200-340 µm (Schmitz 1993). The protractor and the retractor motoneuronal activity was recorded extracellularly from the nerve stumps nl2 and nl5 by means of hook electrodes (Schmitz et al. 1988).
Data analysis
Intra- and extracellular recordings and the treadband signal were stored on a computer by using an A/D converter (CED 1401plus interface, Cambridge Electronic) on-line. The recordings were analyzed off-line with the Spike2 software (Version 3.13). Statistical evaluation of data and plotting of graphs was done with Excel 97. 'N' gives the number of experiments and n gives the sample size. The 
2 test of Brandt-Snedecor was used for tests for significance of differences.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Karg et al. 1991). During the stance phase, the animal pulls the treadband toward the body; at the onset of the swing phase, the leg is lifted off the treadband and extended back to a more distal position, where it touches down onto the treadband prior to the next stance (see also: Bässler 1993; Fischer et al. 2001). During these stepping-like movements, the activities of motoneurons innervating the ProCx and RetCx muscles, which in vivo would move the leg in anterior and posterior directions at the TC-joint, are tightly coupled to the stepping movements of the more distal coxa-trochanter (CTr), femur-tibia (FTi), and tibia-tarsus joints (Fig. 1B) (cf. Fischer et al. 2001). In the present paper, we show that this coordination is predominantly due to sensory influences arising from load measuring sense organs of the walking leg. In particular, the CS located on the trochanter—the trCS (Hofmann and Bässler 1982; Schmitz 1993)—contribute to this interjoint coordination.
|
Figure 1B shows a stepping sequence together with extracellular recordings of the activities of ProCx and RetCx motoneurons as well as an intracellular recording from a single ProCx motoneuron. At the start of each stance, ongoing ProCx motoneuronal activity is terminated and RetCx motoneuronal activity is initiated. RetCx activity continues throughout the stance phase. At the end of the stance phase, the RetCx motoneuronal activity terminates and activity switches over again to ProCx motoneurons. The intracellular recording of the ProCx motoneuron exemplifies that activation and inactivation of ProCx motoneurons is paralleled by strong depolarizations and hyperpolarizations in membrane potential. Generally, this rigid and fixed coupling of activity in the thoraco-coxal motoneuron pools with the stepping pattern performed by the distal leg joints was observed in all experimental animals in which the CS were kept intact (Fig. 1C, left; N = 11 experiments with extracellular recordings, N = 3 experiments with intra- and extracellular recordings). Whereas 
84% of all steps performed by the animals showed a clear-cut alternating activity pattern (Fig. 4, left), in the remaining 16% of the steps, coxal motoneuronal activity was less clearly modulated. For example, in some stance phases, either no clear burst was detectable in RetCx motoneurons or ProCx and RetCx activity showed multiple transitions throughout one stance phase. The general coupling of coxal motoneuronal activity to the stepping movements of the distal leg segments, observed in the majority of the steps, is also reflected by plotting histograms of the relative thoraco-coxal motoneuronal activity as a function of stance phase for all steps generated (Fig. 1D, left). To do so, the start of the treadband movement was defined as the start of leg stance. For six experiments, all steps (n = 327) were normalized for the duration of the stance phase and each stance was divided into 20 bins. Averaged histograms were generated for ProCx and RetCx motoneuronal activity, each normalized to the respective bin with the maximal counts. An addition 10 bins were evaluated both before the beginning and after the termination of stance. This evaluation (N = 6; n = 327 steps) clearly demonstrates that during stance RetCx motoneuron activity is steadily increasing with ProCx activity being strongly decreased at the onset of stance and maintained only at a low level throughout the whole stance phase. At the end of stance, the activity of coxal motoneurons switches back with the ProCx activity strongly increased and RetCx decreased at the same time.
|
) and the one field of fCS (Tatar 1976), dramatically changed the TC-joint motoneuronal activity pattern during stepping movements of the leg (Fig. 1C). For comparison, sections of two walking sequences from the same animal stepping with all CS intact (left) and with all CS destroyed (right) are presented in Fig. 1C. After the ablation of the CS, activity in thoraco-coxal motoneurons were no longer coupled to the stepping pattern in a phase-locked fashion with RetCx motoneurons being active during stance and the ProCx motoneurons being active during swing. Instead, activity of thoraco-coxal motoneurons was simultaneously increased in both motoneuron pools during the whole execution of each step. This means that the patterning of activity in these pools of motoneurons had ceased after all CS had been ablated (Fig. 1C, right). This is shown in more detail in Fig. 1D by plotting the mean activities of the TC-joint motoneurons during normalized stance-phase before (N = 6; n = 327; left) and after (N = 6; n = 274 steps; right) ablation of the CS. In Fig. 1D, left, the relative activities of ProCx and RetCx motoneurons versus normalized stance duration are shown. This evaluation yielded two findings: first, motoneuronal activity was decreased in both motoneuron pools after ablation of the CS. We observed this in five ProCx and four RetCx nerve recordings in six experiments (not shown). Second, in general, ablation of the CS resulted in the loss of the regular activation/inactivation pattern of coxal motoneurons at the start of leg stance (in 6 of 6 experiments). Furthermore, after ablation of the CS, inactivation of RetCx motoneurons at the end of stance appeared to be far less pronounced. However, some activation of ProCx motoneurons at the end of stance was still present after ablation of the CS (Fig. 1D), presumably due to incomplete removal. From these results, it becomes clear that coupling of the TC-joint motoneuronal activity to the leg stepping pattern is substantially affected by ablation of all CS.
Intracellular recordings from RetCx (N = 4) and ProCx (N = 3) motoneurons revealed that the alteration in coxal motoneuronal activity after ablation of the CS could be attributed to the loss of stance-phase-related synaptic inputs to the motoneurons. Previously we had shown that during stance of the single-middle-leg preparation RetCx motoneurons are excited and ProCx motoneurons are inactivated, whereas the reverse is true during leg swing (Fischer et al. 2001; their Fig. 7). This situation was markedly altered after ablation of the CS (see Fig. 2), and recording these motoneurons after ablation of the CS showed that RetCx motoneurons were no longer depolarized (Fig. 2A) and ProCx motoneurons were no longer hyperpolarized during leg stance (Fig. 2B).
|
From the results presented so far, the question arises as to which of the groups of CSs on the trochanterofemur provide the sensory information underlying the preceding influences. In the stick insect, four groups of CS are located on the trochanterofemur. Of those, the one located on the proximal femur (fCS), specifically influences motoneuronal activity of the FTi-joint (Akay et al. 2001). That is, stimulation of the fCS can reduce and terminate extensor tibiae motoneuronal activity, and it can increase or initiate flexor tibiae motoneuronal activity (Akay et al. 2001). In this study, we provided evidence that the remaining three groups of CS located on the trochanter, the trCS, do not have any influence on the timing of FTi-joint motoneuronal activity, indicating a functional segregation of the CS groups (Akay et al. 2001).
We investigated which of the groups of CS, i.e., the fCS and/or the trCS, provide the sensory information underlying the coupling of the TC-joint motoneuronal activity to the stepping movements of the distal leg joints. To do so, we selectively destroyed the trCS and the fCS in varying sequence. We compared the activity patterns of the TC-joint motoneurons during walking with the situation of all CS being intact. Figure 3 shows histograms of the relative activity of ProCx (
) and RetCx (
) versus normalized stance phase for four different conditions. In Fig. 3A the TC-joint motoneuronal activity in the control situation (all CS intact, n = 284 steps) is compared with the condition after ablation of the fCS (n = 194 steps). The data evaluation was similar to the one described in Fig. 1D. From both plots, it is clear that the TC-joint motoneuronal activity is not changed by ablation of fCS compared with the control situation. During the stance phase, RetCx activity was increased and ProCx activity reduced, whereas this activation pattern reverses at the end of leg stance and onset of leg swing. This finding suggests that the fCS do not contribute substantially to the coordination of the TC-joint motoneuronal activity. In contrast, specifically ablating the trCS markedly affected the activity pattern of coxal motoneurons during the step cycle (Fig. 3B). As in Fig. 3A, B compares the histogram of the relative activity in coxal motoneurons versus the normalized stance phase for four animals before and after trCS ablation. After ablation of all trCS, the ProCx activity was no longer terminated at the beginning of stance (intact: n = 225 steps, trCS ablated: n = 351 steps). As well, the phasic activation of RetCx motoneurons at the start of the stance was greatly reduced. Still, however, at the end of the stance, ProCx activity resumed. Note the similarity between these data from animals with trCS ablation and the data from all CS ablation shown in Fig. 1D.
|
) and not coordinated (
) TC-joint motoneuronal activity in intact animals (left) and after ablation of the trCS (right). In intact animals, 84.0% of the steps (n = 225, N = 4) showed a coordinated TC-joint motoneuronal activity pattern. After destroying all of the trCS, the number of steps with not coordinated TC-joint motoneuronal activity increased to 68.l% (n = 351, N = 4; Fig. 4A). This difference is highly significant (2 test, P << 0.001%). In experiments in which the fCSs were destroyed first, shown in Fig. 4B, the percentage of observed not coordinated steps changed only slightly from 8.1% in intact (n = 4, n = 284) to 14.9% (n = 4, n = 194) after fCS had been destroyed. This difference is not significant (2 test, P = 9%). Finally, Fig. 4C compares the percentage of coordinated and not coordinated activity of coxal motoneurons in animals before and after ablation of all CS (fCS plus trCS). Ablation of all CS caused a substantial increase of the percentage of not coordinated steps from 11.6% (N = 8, n = 509) to 60.32% (N = 8, n = 373). This difference is highly significant (2 test, P << 0.001%). Ablating all CS led to similar results as with only trCS being removed. A comparison of the distributions of the percentages of coordinated/not coordinated steps in both situations revealed no significant differences (2 test, P = 19.8%). In summary, the preceding results show that destruction of trCS causes a decrease in the probability in the individual step for a coordinated TC-joint motoneuronal activity pattern to be generated. In some steps, however, even after ablation of all CS, the leg muscle control system was capable of generating coordination of TC-joint motoneuronal activity to the stepping pattern of the distal leg joints, however, with a much decreased probability. The data show that the necessary sensory information is provided by the trCS and that the fCS have no or at least no detectable influence.
Stimulation of the CS induces transitions in activity between ProCx and RetCx motoneurons in activated animals
The preceding results indicate that sensory signals from the trCS play a significant role for patterning TC-joint motoneuronal activity during stepping of the distal leg segments. To prove this capability of sensory signals from the CS, we investigated the influence of selective CS stimulation on the activity pattern of TC-joint motoneurons in a reduced preparation. Except for the CS of the middle leg under investigation, the thoracic nervous system of the animal was completely denervated/deafferented. In six experiments, we selectively stimulated the CS by bending of the femur horizontally by means of piezoelectric element, similar to the procedure used by Schmitz (1993). To do so, the animals and the middle leg were restrained to avoid any leg movements and prepared according to the procedure described in METHODS. The locomotor system was activated by touching the abdomen of the animal with a small paint brush (Bässler 1983). This tactile stimulation induced activity in leg motoneurons as well as sequences of transitions in activity between motoneuron pools innervating antagonist muscles of the leg joints (Bässler and Wegner 1983; Cruse 2002). In the present experiments, alternating activity in RetCx and ProCx motoneurons was initiated by tactile stimulation. CS were stimulated by slightly bending the femur in caudal and in rostral direction while the animal was generating alternating bouts of activity in coxal motoneurons on tactile stimulation. Figure 5A shows an example from such an experiment. As becomes clear from this example, ongoing activity of the ProCx motoneurons was terminated, and a burst of activity was initiated in RetCx motoneurons when the femur was bent caudally, which is represented by an upward deflection of stimulus trace. The mean latency between stimulus onset and termination in ProCx activity was 0.05 ± 0.66s (n = 43). The RetCx activity initiated regularly outlasted the ramp-and-hold stimulus and was even maintained throughout and after the reverse ramp-stimulus at the femur with protractor activity resuming some time after rostral bending. Caudal bending of the femur during ongoing RetCx motoneuron bursting did not influence the timing of coxal motoneuronal activity, but could increase the firing frequency of RetCx motoneurons (not shown). In contrast, rostral bending of the femur did not influence ongoing RetCx activity (Fig. 5A).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
). Here we have shown that the tight coupling of the RetCx and ProCx motoneuronal activities to the stepping movements substantially gets worse after ablation of the cuticular strain sensitive CS. Intracellular recordings from the RetCx and ProCx motoneurons revealed that the massive deterioration of the coordinated activity is due to the lack of depolarizing input to the RetCx motoneurons and lack of hyperpolarizing input to the ProCx motoneurons during stance phases. Results from experiments where the destruction of the three trochanteral trCS and the femoral fCS were performed sequentially indicate that mainly signals from the trCS underlie the coupling of the TC-joint motoneuronal activity. Ablation of fCS alone has no detectable effect, suggesting that only signals from the trCS have access to the motor pattern generating system of the TC-joint. In reduced preparations, in which the middle leg was completely immobilized, stimulation of the CS by bending the trochanterofemur in the posterior direction switches the motoneuronal activity from ProCx to RetCx. In contrast, bending the trochanterofemur in the anterior direction does not change the ongoing activity pattern of TC-joint motoneurons. Influence of cuticular strain information on patterning TC-joint motoneuronal activity in the single middle leg preparation
Intracellular recordings of the RetCx and ProCx motoneuronal activity show that at the start of the stance phase the ProCx hyperpolarizes and stops firing action potentials and this hyperpolarized situation remains during the stance. At the same time, RetCx depolarizes and starts to fire action potentials, and the depolarized situation remains until the end of the stance. At the end of the stance, the ProCx depolarizes and the RetCx hyperpolarizes again. In experiments with intracellular recordings from ProCx or RetCx motoneurons and destroyed trCS, we showed that the substantially degraded coordination after ablation of the trCS is due to the lack of this de- and hyperpolarizing pattern of RetCx and ProCx motoneurons during the stance phase. In the totally restrained preparation, increased load on the leg applied during ProCx burst activity immediately stopped the protractor burst and elicited RetCx burst activity. This strongly indicates that signals from the trCS underlie the activation of the RetCx and inactivation of the ProCx at the start of and throughout the stance. In freely forward walking insects, it is these muscles that contribute most to the stance and swing movements of the leg in the step cycle. RetCx motoneurons, exclusively active during stance, mainly determine the forward propulsion and speed of the body during walking while ProCx motoneurons, active during swing, determine trajectory, velocity and duration of the swing phase (Dean 1984; Epstein and Graham 1983; Graham and Wendler 1981; Schmitz and Hassfeld 1989). Our data suggest that the patterning of the TC-joint motoneuronal activity during a step cycle depends on the load of the leg. At the end of the swing phase, when the tarsus regains ground contact, the suddenly increasing cuticular stress might be a well-defined trigger signal for the motor pattern generator to switch from ProCx to RetCx activity and hereby changing from swing to stance mode. Several behavioral studies have demonstrated that ground contact is one of the key signals for the transition from swing to stance (e.g., Cruse et al. 1995), and this mechanism was successfully tested with algorithms that simulate stick insect walking (e.g., Cruse et al. 1995; Schmitz et al. 2001). Although a contribution of tarsal signals cannot be excluded, it seems at least plausible that the trCS are better suited to detect ground contact. The mechanical conduction time from the tarsus to the proximally located CS groups is by far shorter than the nervous conduction time from tarsal sensors (e.g., Höltje and Hustert 2003). The animals by exploiting this physical advantage and due to the advantageous geometrical arrangement of the CS groups near the thoracic ganglion could gain some temporal advantage. During the stance phase, the leg is under load as it has to support and to propel the body. Thus the trCS information also in this phase is well suited to keep the stance muscle active and prevent the swing muscle from being activated. For cockroaches, Pearson (1972) reported even a reinforcing action of load monitoring sense organs during stance, a principle also found in vertebrate locomotion (e.g., cat: Prochazka et al. 1997; for a review, see Duysens et al. 2000).
One issue needs further attention, however: are the trCS the only sense organs patterning TC-joint motorneuronal activity in the walking cycle? Our data presented show that after ablation of trCS still 30% of the steps show properly coordinated TC-joint motoneuronal activity. This could either indicate that removal was incomplete, or, more likely, that another sensory or central signal is involved. The latter possibility is favored by the finding that after trCS removal at the end of stance ProCx motoneuronal activity was still somewhat increased while in the restrained preparation release of the load increasing stimulus fails to elicit ProCx activity. Tarsal receptors are a potential source of this influence.
Differential role of trCS and fCS in patterning leg motoneuronal activity during walking
Previously it has been demonstrated that the loading status of the leg also influences the patterning of the FTi-joint motoneuronal activity (Akay et al. 2001). This study showed that the influence on the FTi-joint is carried out solely by signals coming from the fCS and that the trCS have no influence on the FTi-joint motoneuronal activity. The fCS were shown to inhibit the activity of the extensor tibiae motoneurons and to increase EMG activity of the flexor tibiae muscle. Destruction of the fCS caused a decrease of the flexor tibiae activity during stance phase but did not change extensor tibiae motoneuronal activity, in neither stance nor swing phase.
Interestingly, in the present study we find that solely signals from the trCS affected the TC-joint motorneuronal activity patterns while the fCS did not influence the TC-joint. In our experiments, only the destruction of the trCS caused a substantial worsening of the coordination of the TC-joint motoneuronal activity with the walking pattern performed by the distal CTr- and FTi-joints. Animals with only the fCS destroyed still had normal coordinated TC-joint motoneuronal activity which was comparable to the pattern in intact situations.
Our data suggest that the four groups of CS, located on the trochanterofemur (trCS and fCS), previously treated as "one" sensory system (Hofman and Bässler 1982; Schmitz 1993; Schmitz and Stein 2000), can be considered as two functional subgroups. The fCS is the one group that influences the motoneuronal activity of the more distal FTi-joint, and the trCS are the three groups influencing the motoneuronal activity of the more proximal TC-joint. Further detailed analysis is necessary to see whether the three groups of trCS can be divided into three functional subgroups. One indication for further subgrouping can be obtained from data from Schmitz (1993), who showed that ablation of the anterior group of the trCS led to a reduction of ProCx reflex responses while ablation of the posterior group of trCS affected mainly the RetCx reflex response in the resting animal.
Sensory information can be used to pattern motoneuronal activity by modifying its amplitude within a step (magnitude control) as well as by affecting the timing of the activity of the central-pattern-generating networks (timing control) (Pearson 1993, 1995). These two different control mechanisms by which sensory signals can act can be used for intrajoint as well as for interjoint coordination. In stick insects, detailed knowledge is available about magnitude (Bässler 1993; Bässler and Büschges 1998; Büschges et al. 2000) and timing effects (Bässler and Büschges 1998) in FTi-joint control. Regarding interjoint influences of the sensory signals, much less is known. Previously, it has been shown that the movement and position of the FTi-joint can influence the motoneuronal activity of the more proximal CTr-joint via an interjoint reflex pathway (magnitude control) as well as influencing the timing of the activity of the central-rhythm-generating network (CRG) of the CTr-joint (timing control) (Bucher et al. 2003; Hess and Büschges 1997). Akay et al. (2001) showed that this mechanism is not a general rule. They showed that, in the distal direction, stimulating the movement and position sensitive sense organs of the CTr-joint did not influence the FTi-joint motoneuronal activity. However, stimulation of the load-sensitive sense organs, the CS, terminates or decreases the extensor tibiae motoneuronal activity and activates or increases the flexor tibiae activity. In completely isolated mesothoracic ganglia (only the innervation of the fCS were left intact), they showed that the fCS signals can reset the rhythmic activity of the FTi-joint motoneurons, induced by pilocarpine application, suggesting that the fCS signals have access to the CRG of the FTi-joint. It would be interesting to know whether the signals from the trCS can also influence the activity of the pilocarpine-activated CRG of the TC-joint in the otherwise totally denervated animal. The more detailed study of this open question is the focus of current investigations (unpublished results), in which we show that the trCS indeed influence the timing cues of the TC-joint CRG.
Walking pattern generation in stick insect middle leg
Coordination of all leg joints is necessary in walking. However, from previous findings it is known that in stick insects there is no central coupling between the activities of motoneurons controlling different leg joints (Büschges et al. 1995). Several recent papers have reported putative sensory mechanisms that could underlie the coupling of the three centrally uncoupled CRGs for the three different leg joints (Akay et al. 2001; Bässler 1986, 1988; Bucher et al. 2003; Cruse 1985a; Hess and Büschges 1997, 1999). The results of the mentioned papers, completed by the new results of the present paper are summarized in Fig. 6.
|
). There exists no cycle-to-cycle coupling in their activities when they are activated in isolated thoracic ganglia by the presence of the muscarinic acetylcholine receptor agonist pilocarpine (Büschges et al. 1995). This situation is depicted as three networks, each consisting of two antagonistic neuron pools, represented by circles and named according to the set of motoneurons/muscles they drive. This drive is indicated by arrows that connect the CRGs to their respective motoneurons and muscles (depicted by squares). The diagram (Fig. 6) shows the four states of the CRGs (1-4) within one complete stepping cycle, starting at the second half of the swing phase, throughout the stance phase, and until the initiation of the next swing phase. Active neuronal components are represented by filled symbols.
We would like to point out that the mechanisms described in the following merely depict the timing of activity transitions during a step cycle and are not meant as mechanisms which also determine the actual shaping of the motor output. During the second half of the swing phase (state 1 in Fig. 6), the ProCx (Pro) part of the TC-CRG is active together with the levator trochanteris (Lev) part of the CTr-CRG and the extensor tibiae (Ext) part of the FTi-CRG. Activity of the extensor tibiae motoneurons extends the FTi-joint, and this movement is sensed by the femoral chordotonal organ (fCO). Extension signals from the fCO now excite the depressor trochanteris (Dep) part of the CRG governing the CTr-joint and inhibit the levator trochanteris (Lev) part (Bucher et al. 2003; Hess and Büschges 1999). The activation of the depressor motoneurons and inactivation of the levator motoneurons cause a depression of the leg, and this leads the tarsus of the leg to be set on the ground. The tarsus touching the ground causes an increase of the load on the leg, which is sensed by the trochanteral and femoral CS (trCS and fCS) (Höltje and Hustert 2003; Noah et al. 2001). The trCSs excite the RetCx (Ret) part of the CRG governing the TC-joint and inhibit the ProCx (Pro) part of the CRG of the TC-joint, which leads to a transition in motoneuronal activity from Pro to Ret. This would start the stance movement if the leg were free to move Ret activity would cause the leg to move backward around the TC-joint and propel the body forward (present paper). The fCS-signaling load from the leg excites the flexor tibiae (Flx) part of the CRG governing the FTi-joint and inactivates the extensor tibiae (Ext) part of this CRG. Resulting Flx activity causes the leg to pull the treadband (Akay et al. 2001). During the stance, the flexion movement of the FTi-joint, sensed by the fCO, reinforces activity of the Flx part of the CRG governing the FTi-joint and thereby its own flexion movement (positive feedback, first part of the active reaction) (Bässler 1986, 1988). At a particular position, signals from the fCO terminate activity of the Flx part of the CRG governing the FTi-joint. Flx activity is terminated and Ext starts to fire again (2nd part of the active reaction) (Bässler 1986, 1988). At the same time, flexion signals from the FTi-joint activate the Lev part of the CRG governing the CTr-joint. Lev motoneurons are activated (Bucher et al. 2003; Hess and Büschges 1999). At this point, the load on the leg decreases, which is sensed by the trCS and fCS. Signals from the fCS would assist Ext activity (Akay et al. 2001; Bässler 1986, 1988), and signals from the trCS would activate the Pro part of the CRG governing the TC-joint and inactivate the Ret part of this CRG. This results in an inactivation of Ret motoneurons and an activation of Pro motoneurons (Cruse 1985a,b; present paper), and a new swing is initiated.
As a conclusion, from our present data combined with those from previous investigations, it is now possible to generate a sequence of sensory-central interactions that can account for the timing of the motor output of a complete stepping cycle as it occurs in the single-middle-leg preparation of the stick insect. Interestingly, the coupling of the motoneuronal activity in the TC-joint found in the present study represents a forward walking motor pattern. Subsequent studies have to examine whether these mechanisms are modified for generating curve walking or backward walking.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Present address of T. Akay: University of Pennsylvania, School of Medicine, 121 Johnson Pavilion, Philadelphia, PA 19104-6060.
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: A. Büschges, Dept. Animal Physiol./Zool. Institute, University of Cologne, Weyertal 119, 50923 Köln, Germany (E-mail: ansgar.bueschges{at}uni-koeln.de).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Bässler U. Sense organs in the femur of the stick insect and their relevance to the control of position of the femur-tibia-joint. J Comp Physiol [A] 121: 99–113, 1977.[CrossRef]
Bässler U. Neural Basis of Elementary Behavior in Stick Insects. Berlin: Springer-Verlag, 1983.
Bässler U. Afferent control of walking movements in the stick insect Cuniculina impigra. II. Reflex reversal and the release of swing phase in the restrained foreleg. J Comp Physiol [A] 158: 351–362, 1986.[CrossRef]
Bässler U. Functional principles of pattern generation for walking movements of stick insect forelegs: the role of the femoral chordotonal organ afferences. J Exp Biol 136: 125–147, 1988.
Bässler U. The walking- (and searching-) pattern generator of stick insects, a modular system composed of reflex chains and endogenous oscillators. Biol Cybern 69: 305–317, 1993.[CrossRef]
Bässler U and Büschges A. Pattern generation for stick insect walking movements-multisensory control of a locomotor program. Brain Res Rev 27: 65–88, 1998.[CrossRef][Medline]
Bässler U and Wegner U. Motor output of the denervated thoracic ventral nerve cord in the stick insect Carausius morosus. J Exp Biol 105: 127–145, 1983.
Bucher D, Akay T, DiCaprio RA, and Büschges A. Interjoint coordination in the stick insect leg-control system: the role of positional signaling. J Neurophysiol 89: 1245–1255, 2003.
Büschges A. Processing of sensory input from the femoral chordotonal organ by spiking interneurones of stick insects. J Exp Biol 144: 81–111, 1989.
Büschges A, Schmitz J, and Bässler U. Rhythmic patterns in the thoracic nerve cord of the stick insect induced by pilocarpine. J Exp Biol 198: 435–456, 1995.[Abstract]
Büschges A, Sauer AE, and Bässler U. Flexibility of a proprioceptive feedback system results from ist "parliamentary" (distributed) organisation. In: Prerational Intelligence: Adaptive Behavior and Intelligent Systems Without Symbols and Logic, edited by Cruse H, Dean J, and Ritter H. Dordrecht, The Netherlands: Kluwer Publ., 2000, pp. 267–286, 2000.
Chrachri A and Clarac F. Fictive locomotion in the fourth thoracic ganglion of the crayfish Procambarus clarkii. J Neurosci 10: 707–719, 1990.[Abstract]
Cruse H. Which parameters control the leg movement of a walking insect? I. Velocity control during the stance phase. J Exp Biol 116: 343–355, 1985a.[ISI]
Cruse H. Which parameters control the leg movement of a walking insect? II. The start of the swing phase. J Exp Biol 116: 357–362, 1985b.[ISI]
Cruse H. The functional sense of "central oscillations" in walking. Biol Cybern 86: 271–280, 2002.[CrossRef][ISI][Medline]
Cruse H, Brunn D, Bartling Ch, Dean J, Dreifert M, Kindermann Th, and Schmitz, J. Walking—a complex behavior controlled by simple networks. Adapt Behav 3: 385–418, 1995.
Dean J. Control of leg protraction in the stick insect: a targeted movement showing compensation for externally applied forces. J Comp Physiol [A] 155: 771–781, 1984.[CrossRef]
Delcomyn F. Activity and directional sensitivity of leg campaniform sensillae in a stick insect. J Comp Physiol [A] 168: 113–119, 1991.[CrossRef][Medline]
Duysens J, Clarac F, and Cruse H. Load-regulating mechanisms in gait and posture: comparative aspects. Phys Rev 80: 83–133, 2000.
Duysens J and Pearson KG. Inhibition of flexor burst generation by loading ankle extensor muscles in walking cats. Brain Res 187: 321–332, 1980.[CrossRef][ISI][Medline]
Epstein S and Graham D. Behaviour and motor output of stick insects walking on a slippery surface. J Exp Biol 105: 215–229, 1983.
Fischer H, Schmidt J, Haas R, and Büschges A. Pattern generation for walking and searching movements of a stick insect leg. I. Coordination of motor activity. J Neurophysiol 85: 341–353, 2001.
Gabriel J, Scharstein H, Schmidt J, and Büschges A. Control of flexor motoneuron activity during single leg walking of the stick insect on an electronically controlled treadwheel. J Neurobiol 56: 237–251, 2003.[CrossRef][ISI][Medline]
Graham D. Pattern and control of walking in insects. Adv Insect Physiol 18: 31–140, 1985.
Graham D and Wendler G. The reflex behaviour and innervation of the tergo-coxal retractor muscles of the stick insect Carausius morosus. J Comp Physiol 143: 81–91, 1981.[CrossRef]
Hess D and Büschges A. Sensory pathways involved in interjoint reflex action in insect leg. J Neurobiol 33: 891–913, 1997.[CrossRef][ISI][Medline]
Hess D and Büschges A. Role of proprioceptive signals from an insect femur-tibia joint in patterning motoneuronal activity of an adjacent leg joint. J Neurophysiol 81: 1856–1865, 1999.
Hofmann T and Bässler U. Anatomy and physiology of throchanteral campaniform sensilla in the stick insect, Cuniculina impigra. Physiol Entomol 7: 413–426, 1982.
Hofmann T and Bässler U. Response characteristics of single trochanteral campaniform sensillae in the stick insect, Cuniculina impigra. Physiol Entomol 11: 17–21, 1986.
Höltje M and Hustert R. Rapid mechanosensory pathways code leg impact and elicit very rapid reflexes in insects. J Exp Biol 206: 2715–2724, 2003.
Karg G, Breutel G, and Bässler U. Sensory influences on the coordination of two leg joints during searching movements of stick insects. Biol Cybern 64: 329–355, 1991.[CrossRef]
Knop G, Denzer L, and Büschges A. The role of central bistable networks in generating reflex reversal in the insect leg muscle control system. J Neurophysiol 86: 3065–3068, 2001.
Noah JA, Quimby L, Frazier SF, and Zill SN. Force detection in cockroach walking reconsidered: discharges of proximal tibial campaniform sensilla when body load is altered. J Comp Physiol [A] 187: 769–84, 2001.[CrossRef][ISI][Medline]
Orlovsky GN, Deliagina TG, and Grillner S. Neuronal Control of Locomotion. Oxford, UK: Oxford Univ. Press, 1999.
Orlovsky GN and Feldmann AG. Role of afferent activity in the generation of stepping movements. J Neurophysiol 4: 304–310, 1972.
Pearson KG. Central programming and reflex control of walking in the cockroach. J Exp Biol 56: 173–193, 1972.
Pearson KG. Common principles of motor control in vertebrates and invertebrates. Annu Rev Neurosci 16: 265–297, 1993.[ISI][Medline]
Pearson KG. Proprioceptive regulation of locomotion. Curr Opin Neurobiol 5: 786–791, 1995.[CrossRef][ISI][Medline]
Pearson KG. Neural adaptation in the generation of rhythmic behavior. Annu Rev Physiol 62: 723–753, 2000.[CrossRef][ISI][Medline]
Pringle JWS. Proprioception in insects. II. The action of the campaniform sensilla on the legs. J Exp Biol 15: 114–131, 1938.[Abstract]
Prochazka A, Gillard D, and Bennett DJ. Positive force feedback control of muscles. J Neurophysiol 77: 3226–3236, 1997.
8 excitatory motoneurons the axons of which travel in the nerve nl2, while the RetCx is innervated by 
