| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Physiology, Spinal Cord Research Centre, University of Manitoba Winnipeg, Manitoba, Canada
Submitted 7 February 2005; accepted in final form 3 May 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
1, 2) receptors are not critical for PPR-evoked locomotor-like activity. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Grillner et al. 1997; Jordan 1991, 1998; Rossignol 1996). In recent years, considerable progress on the neural mechanisms for the control of locomotion has been made using isolated spinal cord preparations from neonatal rats and mice (Bonnot et al. 2002; Clarac et al. 2004; Jiang et al. 1999; Kiehn and Butt 2003; Kiehn and Kjaerulff 1998; Kudo and Yamada 1987; Nishimaru et al. 2000; Schmidt and Jordan 2000; Smith and Feldman 1987; Whelan et al. 2000). For this purpose, most studies have employed applications of neurotransmitters or other drugs into the bath to elicit locomotor-like activity. These studies have shown that a variety of neurochemicals, including N-methyl-D-aspartate (NMDA), noradrenaline, 5-hydroxytryptamine (5-HT), dopamine, and cholinergic agonists can elicit rhythmic activity with a locomotor-like pattern (Alford et al. 2003; Rossignol et al. 2002). One of the most reliable means for inducing locomotion, at least in the neonatal rat isolated spinal cord, is with bath-applied 5-HT (Cazalets et al. 1990, 1992; Cowley and Schmidt 1994; Schmidt and Jordan 2000). Furthermore, the most effective site for 5-HT to induce locomotion has been localized to the supralumbar segments of the spinal cord (Cowley and Schmidt 1997), and transplantation of 5-HT neurons into the thoracic cord activates locomotion in adult chronic spinal rats (Gimenez y Ribotta et al. 2000).
Several 5-HT receptor types have been identified in the spinal cord (Hochman et al. 2001; Schmidt and Jordan 2000), and blockage of certain of these receptors types can disrupt 5-HT–induced locomotor activity. In both the neonatal rat and mouse spinal cord, selective antagonists of 5-HT2A and 5-HT7 receptors have been shown to block 5-HT–induced rhythmicity (Beato and Nistri 1998; Bracci et al. 1998; Cazalets et al. 1992, 1995; Hochman et al. 2001; Jordan and Schmidt 2002; Madriaga et al. 2004; Schmidt and Jordan 2000). Nevertheless, the cells on which these antagonists act to block locomotion are not known. It would be possible to block the locomotor output recorded in these experiments if the affected receptors were on interneurons of the CPG or on motoneurons. It has been shown that 5-HT–induced locomotion requires 5-HT be applied at the lumbar level as well as in the supralumbar region, although 5-HT applied to the lumbo-sacral region alone induced only tonic activity (Cowley and Schmidt 1997). There is ample evidence that 5-HT exerts an excitatory effect directly on motoneurons (reviewed in Schmidt and Jordan 2000). Both 5-HT2A and 5-HT7 receptors have been shown to be involved in mediating 5-HT effects on motoneurons (Gilmore and Fedirchuk 2004; Inoue et al. 2002; Jackson and White 1990; Takahashi and Berger 1990; Wang and Dun 1990).
Although virtually all of the 5-HT terminals in the rat spinal cord originate in brain stem nuclei (reviewed in Lakke 1997), the descending 5-HT cells responsible for inducing locomotion have not been identified. In fact, it has never been shown that stimulation of brain stem neurons containing 5-HT is effective for the induction of locomotion. Brain stem–evoked locomotion has been examined in isolated rodent brain stem–spinal cord preparations using chemical stimulation of the entire brain stem (Atsuta et al. 1991; Smith et al. 1988) or with electrical stimulation of sites within the brain stem, although without detailed anatomical identification of the effective sites (Atsuta et al. 1988, 1990, 1991; Gilmore and Fedirchuk 2004; Zaporozhets et al. 2004).
Here, we attempt to identify a population of brain stem 5-HT–containing neurons that can be stimulated to evoke locomotor-like activity in the isolated neonatal rat brain stem–spinal cord preparation. Furthermore, we performed experiments designed to determine the receptors at the spinal level that are responsible for the brain stem–evoked locomotor-like activity and to identify and localize the spinal cord cells that are influenced by the effective antagonists. Preliminary experiments that preceded this study have been published (Fyda and Jordan 1999; Jordan and Schmidt 2002), but none of the data in this study were included in those reports. A preliminary report of the data presented here was published in abstract form (Liu and Jordan 2004).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
All experiments were performed in accordance with Canadian Council on Animal Care guidelines and were approved by the University of Manitoba Animal Protocol Committee. The experiments were performed on brain stem–spinal cord preparations isolated from neonatal Sprague-Dawley rats (0–3 days old). After anesthesia with halothane, the animal was immediately decapitated just posterior to bregma, eviscerated, and removed to a sylgard-coated recording chamber. A laminectomy was performed, first removing the dorsal portions of the vertebrae and then the ventral side of the vertebrae. The dorsal and the ventral roots were cut from C1 to S3. The brain stem was transected at a site just rostral to the exit of the trigeminal nerve. The isolated brain stem–spinal cord was superfused with artificial cerebrospinal fluid (ACSF, concentration in mM: 128 NaCl, 3.0 KCl, 0.5 NaH2PO4, 1.5 CaCl2, 1.0 MgSO4, 21 NaHCO3, and 30 glucose) and oxygenated with 95% H2O-5% CO2 at room temperature. The preparation was pinned ventral side up to the sylgard surface, and petroleum jelly barriers were placed at the C5/6, T8/9, and L2/3 levels (unless otherwise specified in the text) to form separate pools. This was done because preliminary experiments showed that specific 5-HT antagonists had differential effects when applied to the lower thoracic/rostral lumbar or the lower lumbar regions of the spinal cord (Jordan and Schmidt 2002; Schmidt and Jordan 2000). The barriers were tested for leakage before and after the experiments by their ability to maintain different depths of superfusate. Thus the spinal cord could be separated into rostral (C9–L2) and caudal (below L3) compartments.
Electrically evoked locomotion
Locomotor-like activity was evoked through electrical stimulation of the parapyramidal region (PPR) of the mid-medulla. A monopolar tungsten electrode (10–50 K
, tip diameter 1–2 µm, 25-µm exposed tip; MicroProbe) was lowered into the PPR just lateral to the pyramidal tract (PT), 0.5–1.2 mm caudal to the ponto-medullary junction. Small adjustments were made in the position of the electrode until locomotion could be reliably evoked. Square pulses of constant current, 5-ms duration, 2.5–3 Hz, were given at variable intensities (80–200 µA). Other nearby sites were also stimulated to determine the specificity of the PPR for evoking locomotion (see Fig. 1B ). Glass or plastic suction electrodes were applied to record the L2 and L5 ventral discharges of both sides to monitor locomotor-like activity (Cowley and Schmidt 1997; Kjaerulff and Kiehn 1996). Ventral root recordings were amplified, filtered (30 Hz to 2 KHz), digitized, stored, and analyzed using software developed at the Winnipeg Spinal Cord Research Centre (http://www.scrc.umanitoba.ca/doc/). Locomotor-like activity consisted of alternating activity recorded between both the left and right L2 and L5 ventral root pairs and alternating activity between ipsilateral flexor (L2) and extensor (L5) ventral root pairs (Fig. 1).
|
Chemically evoked locomotion
To elicit locomotor-like activity by activation of neurons in the vicinity of the stimulating probe, while excluding activation of fibers of passage, chemical stimulation was employed. Glutamate (10–50 mM/1 µl) or the GABA receptor antagonist bicuculline (10 mM/1 µl) was injected through a glass micropipette (4–6 µm tip diam) connected to a pneumatic PicoPump (World Precision Instruments). Control saline injections (1 µl, n = 4) did not produce locomotor-like activity. In each case, the injection site was the same as previously identified as effective for electrically evoked locomotor-like activity in the same preparation.
Drugs
Various antagonists were tested to block the following receptors during electrically and chemically evoked locomotor-like activity: 5-HT2A (ketanserin 5–20 µM, spiperone 5–20 µM), 5-HT7 (clozapine 0.5–1.0 µM, SB269970 5–15 µM), noradrenergic 1 (prazosin 1–10 µM), 2 (yohimbine 5–25 µM, RX 821002 10–25 µM), dopaminergic D1 (SCH 23390 10–20 µM), and D2 (sulpiride, 10–20 µM). All antagonists were dissolved in distilled water. Drugs prepared for injection into the brain stem (glutamate or bicuculline) were dissolved in saline. Drug doses were adjusted to correspond approximately to the doses of 5-HT required to elicit locomotor-like activity in the neonatal rat spinal cord (Cazalets et al. 1990, 1992; Cowley and Schmidt 1994, 1997). For example, the EC50 for 5-HT at the 5-HT7 receptor is 1.8 nM (Hochman et al. 2001), whereas the doses required for 5-HT–induced locomotor-like activity in the neonatal rat spinal cord were in the range of 10–125 µM (Cowley and Schmidt 1997). The IC50 of the selective 5-HT7 antagonist SB269970 is 
1.3 nM (Bengtson et al. 2004), so that the doses of SB269970 used here (5–15 µM) are well within the range of effective doses for this preparation. All drugs were obtained from Sigma.
Immunohistochemistry
The brain stems were placed in 4% paraformaldehyde for 24 h at 4°C and postfixed in 15% sucrose solution for 
72 h. Thirty- to 50-µm-thick sections were cut on a cryostat. The sections were presoaked in 0.1 M PBS for 1 h to remove the holding medium and pretreated with 5% donkey serum in PBS-T (0.1%) for 30 min; after washing in PBS, sections were incubated with rabbit anti 5-HT antibody (1:200) in 0.1% PBS-T and 1% donkey serum for 48 h at 4°C and incubated with donkey anti-rabbit Cy3 antibody (1:250) in 0.1% PBS-T and 1% donkey serum for 2 h at room temperature. Sections were washed in PBS and Tris-HCI (50 mM) and coverslipped with vectashield. Section were examined and photographed using a Nikon fluorescence microscope.
Measurements and statistical analysis
The cycle duration was defined as the time interval between the onsets of successive bursts of ventral root activity and was measured for 10 consecutive cycles in each episode of locomotor-like activity. Amplitude was measured from these same consecutive cycles as the average peak-to-peak amplitude of the rectified and filtered waveforms. The effects of drugs on amplitude and duration were expressed quantitatively as a fraction of respective control values in the same experiment. The statistical values are indicated as the mean ± SD; n refers to the number of experiments. One-way ANOVA or Student's t-test was used for statistical analysis, and the level of significance was set as P < 0.05. Data were pooled as required if the one-way ANOVA test did not reveal significant differences between individual samples. The coupling strength between left/right or flexor/extensor discharges of the ventral roots (see Fig. 1) was assessed using circular statistics (Kjaerulff and Kiehn 1996; Kriellaars et al. 1994). The phase value (
) indicated by the direction of the vector was defined by dividing the latency between the onsets of paired root cycles by the step cycle period, so that a value of 0.5 represents out-of-phase activity in the two roots, and values of 0 or 1 would occur when the roots are completely in phase. The mean value (r), indicating the concentration of phase values, was expressed by the length of the vector, which ranged from 0 to 1. The Rayleigh test (Zar 1974) was used to determine the coupling strength.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
The effective sites for evoking locomotor-like activity were located in a very restricted area of the mid-medulla just lateral to the PT that others (Helke et al. 1989; Sasek and Helke 1989) have termed the PPR. The effective area was 0.5–1.2 mm caudal to the junction between the pons and medulla (just rostral to the inferior olivary nucleus at the level of the nucleus of the VIIth cranial nerve), 0.5–1.0 mm lateral to the midline, and 0.2–0.5 mm beneath the ventral surface. Stimulation 0.5 mm or more lateral or medial to an effective site did not produce locomotor-like activity. It is well known that 5-HT neurons in the raphe nuclei (i.e., raphe magnus, raphe pallidus, and raphe obscurus) project to the spinal cord, and stimulation of these midline regions was not effective for producing locomotor-like activity. Figure 1A shows a typical lesion at the effective stimulus site, which is surrounded by many 5-HT immunoreactive neurons. The sites of stimulation from 12 representative experiments are shown in Fig. 1B. The sites that were effective for evoking locomotor-like activity are just lateral to the PT, in an area rich in 5-HT neurons (Fig. 1A). In the adult rat, a high proportion of the 5-HT–containing cells in this area project to the spinal cord (Jones and Light 1992). Our effective stimulus sites may overlap with the most effective sites for brain stem–evoked locomotion reported by Atsuta et al. (1988, 1990). They did not specifically seek to stimulate within areas containing 5-HT neurons, however, and the relationship of their stimulus sites to 5-HT neurons is unknown.
Properties of locomotor-like activity evoked by electrical stimulation of the PPR
Locomotor-like activity with right/left and flexor/extensor alternation is shown in Fig. 1. The raw recordings from the right (R) and left (L) L2 (flexor) and L5 (extensor) ventral roots over a 60-s period are shown in Fig. 1C, top four traces, and rectified and filtered versions of these same recordings are repeated in the bottom four traces. Alternation between the right (RL2 and RL5) and left (LL2 and LL5) flexor and extensor ventral roots is shown in Fig. 1D by the rasterized overlay of 15 successive step cycles in these same rectified and filtered traces triggered at the onset of activity in RL2. In Fig. 1E, the alternating activity is analyzed using polar plots derived from 11 preparations. The Rayleigh test reveals that mean values () were 0.44 ± 0.03 (RL2/RL5), 0.43 ± 0.05 (LL2/LL5), 0.51 ± 0.04 (RL2/LL2), and 0.51 ± 0.05 (RL5/LL5). r values reflecting the coupling strength for each of these root pairs were significant (P < 0.001): 0.93 ± 0.07 (RL2/RL5), 0.93 ± 0.04 (LL2/LL5), 0.93 ± 0.04 (RL2/LL2), and 0.91 ± 0.05 (RL5/LL5). Only preparations in which well-coordinated locomotor-like activity occurred, such as shown in Fig. 1, are reported here.
The effective stimuli in these experiments were 80- to 200-µA square pulses of 5-ms duration at 2.5–3.0 Hz. The cycle duration of the PPR-evoked locomotion ranged from 2.75 to 5.20 s (mean, 3.88 ± 0.76 s; n = 86). During the locomotor-like activity, both the cycle duration and amplitude were remarkably stable. In most cases, the onset of the stimulus was followed immediately by locomotor-like activity, which ceased almost coincidentally with termination of the stimulus (Fig. 1C). Rarely, one to three cycles persisted after switching off the stimulus. Stimuli were typically applied for periods of 60–90 s, separated by 5-min intervals during the control period before drug application. On application of an antagonist, stimuli were applied after 1, 3, 5, 10, 15, and 20 min (see Fig. 3).
|
To verify that electrically evoked locomotor-like activity was mainly due to the activation of cells in this area instead of descending fibers, the excitatory amino acid glutamate (1 µl/10 mM, n = 4) and the GABAA receptor antagonist bicuculline (1 µl/10 mM, n = 4) were microinjected into the sites that were effective for electrically evoked locomotor-like activity. As would be predicted if the stimulus were effective due to activation of neuronal somas in or around the stimulus site, locomotor-like activity was induced after 3–5 min of drug injection. The locomotion evoked in this manner was characterized by 5–10 episodes of alternating rhythmic activity lasting 30–40 min (Fig. 1F). In some cases, episodes of well-coordinated locomotor-like activity were separated by one or two bursts of synchronous activity (see Fig. 5D) in all four ventral roots.
|
5-HT2A RECEPTOR ANTAGONISTS ARE ONLY EFFECTIVE WHEN APPLIED TO THE CAUDAL COMPARTMENT. Although several authors have previously shown that the 5-HT2A receptor antagonist ketanserin blocks locomotion induced by 5-HT (see Introduction), a striking feature of the effects of this drug in our experiments, shown in Fig. 2 A, was that it has no effect on PPR-evoked locomotion when applied to the rostral compartment (20 µM, n = 5), but it consistently blocks locomotion within 3 min when applied below L2 (20 µM, n = 6, Fig. 2B). A second antagonist with high affinity for the 5-HT2A receptor, spiperone, also blocked PPR-evoked locomotion (Fig. 2C) when applied to the caudal compartment (15 µM, n = 6) but was without effect (data not shown) when the same concentration of the drug was applied to the rostral compartment (n = 5). The output of L2 motoneurons was preserved when low doses of the 5-HT2A antagonists were applied in the caudal compartment (Fig. 2D). L2 output was blocked only with high doses of these antagonists applied in the caudal compartment.
|
The effects of ketanserin on the amplitude of the L5 ventral root discharges and the duration of the step cycle for all experiments are summarized in Fig. 3A. A dose of 20 µM rapidly reduced the amplitude (by 42.5%) and increased the duration of the step cycle (by 31%) at 1 min after addition of ketanserin, whereas lower doses (5 and 10 µM) reduced the amplitude over a longer period, and in the case of the 5-µM dose, recovery was observed after 10 min. The lower doses had very little effect on the step cycle duration, with the exception of a slight increase (by 25%) after 20 min with the 20-µM dose. These results suggest that blocking the 5-HT2A receptor may serve to reduce motoneuron output, rather than to disturb the central pattern generating mechanisms controlling cycle duration. The fact that L2 activity is eventually blocked, even though ketanserin is applied only below L3, suggests CPG elements and/or interneurons related to motoneuron output in L2 may be affected by the drug (see DISCUSSION).
5-HT7 RECEPTOR ANTAGONISTS ARE EFFECTIVE ONLY WHEN APPLIED TO THE ROSTRAL COMPARTMENT.
Clozapine (1 µM, n = 5) a nonspecific antagonist with a high affinity for the 5-HT7 receptor (Hochman et al. 2001) was without effect when applied to the caudal compartment (Fig. 4 A). At the same dose (1 µM, n = 7), clozapine completely blocked PPR-evoked locomotor-like activity when applied to the rostral compartment (Fig. 4B). The blockage achieved with this dose of clozapine was complete within 3 min of application. When clozapine was applied at a lower concentration (0.5–0.75 µM, n = 6) to the rostral compartment, it caused a substantial prolongation of cycle duration (Fig. 4C). As shown in the histograms in Fig. 3C, the cycle duration was approximately doubled at 10 min after drug application (to 2.02 ± 0.19, P < 0.001). A decrease in amplitude also occurred after 10 min (to 0.70 ± 0.07, P < 0.05). The 1-µM dose of clozapine rapidly increased cycle duration before complete block of locomotion (Fig. 3C). These results indicate that clozapine, rather than blocking locomotion through an action on motoneurons, does so by reducing an excitatory 5-HT–mediated input to elements of the CPG for locomotion, thereby altering cycle duration. The amplitude of the ventral root discharges was also reduced, suggesting that the suppression of activity in interneurons of the CPG results in a reduced excitatory drive to the motoneurons. Clozapine also has a high affinity to 5-HT2A receptors, but it did not mimic the action of ketanserin in the caudal compartment, suggesting that its action was primarily on 5-HT7 receptors.
|
SB269970 also blocked locomotor-like activity induced by microinjection of bicuculline or glutamate into the PPR (Fig. 5D), first increasing the cycle duration (at 5 min), and then later blocking locomotor-like activity (10 min). This shows that locomotor-like activity evoked by activation of only the neurons in the PPR and not the passing fibers can also be blocked by the 5-HT7 receptor antagonist. Episodes of synchronous activity persisted after SB269970 in cases of locomotor-like activity produced by injections of bicuculline into the PPR, however (see Fig. 5D, bottom four traces). Locomotor-like activity produced by injections of glutamate into the PPR was not accompanied by synchronous activity.
Effects of dopaminergic and noradrenergic antagonists
DOPAMINERGIC (D1 AND D2) ANTAGONISTS DO NOT ALTER PPR-EVOKED LOCOMOTOR-LIKE ACTIVITY.
To examine if dopamine is involved in the regulation of the locomotor-like activity evoked by electrical stimulation of the PPR, dopaminergic antagonists (the D1 antagonist SCH23390and the D2 antagonist sulpiride) were applied to the whole spinal cord below the C5 segment. Sulpiride (20 µM, n = 3) altered neither the amplitude of the ventral root recordings (0.94 ± 0.18, P > 0.05) nor step cycle duration (0.98 ± 0.13, P > 0.05). Similarly, SCH23390(20 µM, n = 3) had no significant effect on ventral root discharge amplitude (0.89 ± 0.26) or cycle duration (1.05 ± 0.21). Although clozapine has moderate to high affinity for dopamine receptors (van Tol et al. 1991), the above results suggest that the action of clozapine is not due to blocking the D1 or D2 dopamine receptor. It also indicates that, although 5-HT–evoked locomotion in the neonatal mouse can be disrupted by either D1 or D2 antagonists (Madriaga et al. 2004), these receptors are not involved in the induction of locomotion from the PPR. According to Madriaga et al. (2004), the 5-HT–evoked locomotion could be most effectively disrupted by a combination of the D1 and D2 antagonists. We therefore attempted to disrupt PPR-evoked locomotor-like activity with SCH23390and sulpiride in combination (20 µM each, n = 3), but there was no significant effect (data not shown).
NORADRENERGIC ANTAGONISTS.
The 2 antagonist yohimbine was applied to the whole bath. Low concentrations of the drug (1–10 µM, n = 8) produced either no effect (amplitude: 0.94 ± 0.17, P > 0.05; duration: 1.06 ± 0.21, P > 0.05, n = 4) or an obvious decrease in cycle duration (amplitude: 0.88 ± 0.19, P > 0.05; duration 0.78 ± 0.19, P < 0.05, n = 4) after 5–10 min of drug application, as shown in Fig. 6 A. A high concentration of yohimbine (20 or 25 µM, n = 6) always blocked locomotor-like activity after 5–10 min of drug application (Fig. 6B). Addition of the specific noradrenergic 2 antagonist RX 821002 (10–25 µM, n = 6) caused no change either in amplitude (1.02 ± 0.23, P > 0.05) or duration (1.05 ± 0.20, P > 0.05). In addition, no changes in amplitude (0.92 ± 0.14, P > 0.05) or duration (1.10 ± 0.19, P > 0.05) were obtained when the 1 antagonist prazosin (10 µM, n = 4) was applied to the bath. In a recent study (Gabbay and Lev-Tov 2004) on the locomotor-like activity induced by noradrenaline in the isolated neonatal rat spinal cord, the 1 antagonist prazosin antagonized the drug-induced rhythmic activity, whereas high doses (20 µM) of the 2 antagonist yohimbine were required for blockage. Although our finding that similar doses of yohimbine block PPR-evoked locomotor-like activity is consistent with these results, the absence of any effect of the more specific 2 antagonist RX 821002 or prazosin in our experiments suggest that neither 1 nor 2 noradrenergic receptors play a significant role in locomotor-like activity produced by PPR stimulation.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
We have shown that electrical and chemical stimulation of a specific group of 5-HT neurons in the brain stem of the isolated neonatal rat brain stem and spinal cord preparation is sufficient for evoking locomotor-like activity. These cells are localized in the PPR of the mid-medulla. 5-HT neurons that project to the ventral horn of the lumbar spinal cord via the ventral and ventrolateral funiculi have been localized in the raphé pallidus, raphé obscurus, and the ventral medulla (para-raphé area), including an area close to the ventral surface and lateral to the pyramidal tract, the PPR (Basbaum et al. 1978; Bowker et al. 1981; Dahlstrom and Fuxe 1965; Helke et al. 1989; Hokfelt et al. 2000; Holstege 1987; Holstege and Kuypers 1987a,Holstege and Kuypers 1987b; Jones and Light 1992; Martin et al. 1978; Schmidt and Jordan 2000; Skagerberg et al. 1985; Steinbusch 1981; Westlund et al. 1983). These and other studies show that fibers from the raphé and para-raphé nuclei are known to terminate at multiple levels of the spinal cord and on a variety of neuron types, including autonomic preganglionic neurons, neurons in laminae VII, VIII, and X, and -motoneurons. It is noteworthy that a high proportion of the 5-HT neurons in the PPR (so-called para-raphé neurons) project to the lumbar spinal cord (Jones and Light 1992) in adult rats. Cells in the PPR region have been labeled trans-synaptically with pseudorabies virus injected into the gastrocnemius muscle of adult rats (Kerman et al. 2003). Cells in this region were shown to be activated during treadmill exercise, because they were positive for the activity-dependent marker c-fos (Iwamoto et al. 1996). We have evidence that PPR stimulation gives rise to release of 5-HT, dopamine, and noradrenaline in the thoraco-lumbar spinal cord (Fyda et al. 1997).
Chemical activation with glutamate or bicuculline shows that excitation of neurons in the PPR, and not fibers of passage, accounts for the evoked locomotion. In adult rat experiments, microinjections of acetylcholine, substance P, GABA antagonists, or NMDA into the medioventral medulla induced locomotion (Kinjo et al. 1990) The effective site was similar to the PPR location we report here for the neonatal rat preparation. Neurons in the raphé pallidus and raphé obscurus did not give rise to locomotion when stimulated electrically, and are therefore not likely to be the source of the 5-HT terminals on neurons that induce locomotor activity. Although nonserotonergic neurons in the PPR are simultaneously activated during stimulation, they may not play a central role in the production of locomotor-like activity because 5HT7 and 5-HT2A receptor antagonists block the PPR-evoked activity. This supports the notion that PPR stimulation functions primarily through release of 5-HT in the spinal cord. The possibility that non–5-HT–containing cells in the PPR are involved in the modulation of locomotor-like activity cannot be excluded in our experiments, but if non–5-HT neurons are involved, they are not sufficient for this purpose in the absence of concurrent actions of simultaneously activated 5-HT neurons.
Our novel finding that 5-HT neurons in the PPR are sufficient for the production of locomotion has implications for future research on the function of specific 5-HT neurons in the control of movement and for studies on the use of transplanted 5-HT neurons for the restoration of locomotion after spinal cord injury. Previous studies have shown that 5-HT neurons of the raphé pallidus and raphé obscurus are active during locomotion (Fornal et al. 1985; Heym et al. 1982; Veasey et al. 1995), but there are no electrophysiological studies on the activity of PPR neurons during locomotion. Efforts to restore locomotion have involved the transplantation of 5-HT neurons into the spinal cord (Gimenez y Ribotta et al. 2000; Orsal et al. 2002), and serotonergic reinnervation of the L1–L2 level of the spinal cord from these implants is correlated with locomotor recovery. It is possible that PPR serotonergic neurons were included in the transplants (Gimenez y Ribotta et al. 2000). The effectiveness of reinnervating the upper lumbar segments is consistent with our finding that the rhythm generating portions of the CPG activated by PPR stimulation is in the region above L3. Future attempts at reinnervation for restoring locomotion should focus on the use 5-HT neurons of the PPR.
5-HT7 receptor activation is required for PPR-evoked locomotor-like activity
In a preliminary report of some of these results (Jordan and Schmidt 2002), we found that clozapine, a nonspecific antagonist with a high affinity for the 5-HT7 receptor, blocked brain stem–evoked locomotor-like activity. We also provided preliminary evidence that clozapine applied intrathecally could block brain stem–evoked locomotion in the decerebrate cat preparation (Schmidt and Jordan 2000). Here we show that both clozapine and the more specific 5-HT7 antagonist SB-269970 block PPR-evoked locomotor-like activity and that their action is limited to regions of the spinal cord rostral to the L3 segment. Application of either antagonist into the compartment caudal to the L2 segment was ineffective. The restriction of the actions of the 5-HT7 antagonists to the rostral compartment is consistent with the suggestion that 5-HT–medicated excitation of the locomotor CPG requires the supralumbar regions of the spinal cord (Bertrand and Cazalets 2002; Cowley and Schmidt 1997; Jordan and Schmidt 2002; Schmidt and Jordan 2000). It is also in accordance with the notion that the CPG for locomotion in the neonatal rat, although distributed throughout the thoraco-lumbar region, is most readily activated from the thoracic and rostral regions of the lumbar segments (Bertrand and Cazalets 2002; Cazalets et al. 1995; Kiehn and Kjaerulff 1998; Kjaerulff and Kiehn 1996). The role of 5-HT7 receptors in the initiation of locomotion by direct application of 5-HT has been pointed out by other groups (Hochman et al. 2001; Madriaga et al. 2004). Various pharmacological and lesioning studies have localized the dominant components of the locomotor CPG to the lower thoracic-upper lumbar region (Bertrand and Cazalets 2002; Cazalets et al. 1995; Cowley and Schmidt 1997; Kjaerulff and Kiehn 1996; Kremer and Lev-Tov 1997). Hochman et al. (2001) have shown a similar distribution for cells stained with a 5-HT7 receptor antibody. These same authors also showed that a substantial proportion of cells in this region of the cord labeled with the activity-dependent label sulforhodamine after 5-HT–induced locomotor-like activity were positive for the 5-HT7 receptor. We also have reported preliminary data (Jordan and Schmidt 2002) showing that many cells in the thoraco-lumbar area that are active during locomotion in the adult rat (as shown with c-fos immunostaining after a treadmill locomotion task) are positive for the 5-HT7 receptor. Thus it is plausible that this distribution of 5-HT7 receptors underlies our finding that 5-HT7 receptor antagonists block PPR-evoked locomotor-like activity only if applied rostral to the L3 segment.
Neurons with 5-HT7 receptors are involved in rhythmogenesis
The finding that 5-HT7 receptor blockade results in a progressive reduction in step cycle duration before finally blocking PPR-evoked locomotor-like activity altogether (Figs. 3–5) strongly suggests that the affected 5-HT7 receptors are located on cells that play a role in the rhythm-generating capacity of the locomotor network. These cells may either be involved in providing an excitatory drive to the CPG neurons producing the rhythm or they may form part of the CPG. This is consistent with the apparent gradient of 5-HT–responsive CPG neurons from thoracic to lumbar cord (see Jordan and Schmidt 2002; Schmidt and Jordan 2000 for review). Concomitant with blockage of this rhythm-generating function by the 5-HT7 antagonists, the amplitude of the locomotor output recorded from motoneurons in the ventral roots is also reduced (Figs. 3–5). This does not seem to be due to a direct suppression of motoneuron excitability caused by blockage of 5-HT7 receptors on motoneurons, because the antagonists did not alter the motoneuron activity recorded from the L5 ventral root when applied into the caudal compartment, which included the motor nuclei of the L5 segment. Rather, it is more likely due to a progressive reduction in the output of the more rostrally located elements of the rhythm generating network, resulting in a reduced excitatory drive to the motoneurons, either directly or through more caudally located elements of the CPG.
5-HT2A receptor activation is also required for PPR-evoked locomotor-like activity
Consistent with numerous prior studies showing that antagonists with high affinity for the 5-HT2A receptor block locomotor-like activity (Bracci et al. 1998; Cazalets et al. 1992; MacLean et al. 1998; Madriaga et al. 2004), we found that PPR-evoked locomotor-like activity was blocked when the higher doses of ketanserin or spiperone were applied to the caudal compartment. The results with these two antagonists, taken together, strongly suggest that their effects are exerted on 5-HT2A receptors, because both have high affinity for this receptor, with nonoverlapping affinities for other 5-HT receptors (Glennon et al. 2002; Hochman et al. 2001). No effects were observed when they were applied above L3. It is important to note that the 5-HT2A antagonists blocked rhythmic activity in the L2 ventral root when applied at high doses below the L2 segment and also changed the frequency of the rhythm (Fig. 3). This may be taken as evidence that CPG elements in the lower compartment are affected by the 5-HT2A antagonists. These neurons in the lower cord may strongly interact with the locomotor networks in the upper lumbar cord (Gabbay and Lev-Tov 2004). This is consistent with the well-established concept that the CPG for locomotion is distributed along the length of the spinal cord (see Kiehn and Kjaerulff 1998 for review).
It is noteworthy that with the higher doses of the 5-HT2A antagonists, the activity in L5 was always reduced before any effect on L2. This sequence of effects was also observed by Madriaga et al. (2004) for ketanserin antagonism of rhythmic activity in the neonatal mouse spinal cord induced by 5-HT. Thus even with antagonist application over both motor nuclei simultaneously, the effect on L2 lags behind that on L5. One possible explanation for these observations is that motoneurons located in L2 receive an excitatory drive from more caudally situated interneurons that possess 5-HT2A receptors, whereas the extensor motoneurons recorded in the L5 root are more directly affected by the antagonist due to their expression of 5-HT2A receptors. There is evidence that interneurons in the ventral horn are immunoreactive for the 5-HT2A receptor (Cornea-Hébert et al. 1999; Doly et al. 2004). In the neonatal mouse spinal cord, -methyl-5-HT, a 5-HT2 agonist, was shown to elicit locomotor activity (Madriaga et al. 2004), suggesting that cells with 5-HT2A receptors can contribute to the initiation of locomotion. It is possible that blockage of L2 rhythmic output by applying the 5-HT2A antagonists in the caudal compartment in our experiments is due to suppression of PPR-evoked activity in such interneurons.
L2 rhythmic activity persisted, whereas L5 activity was reduced or blocked when lower doses of the 5-HT2A antagonists were applied to the caudal compartment. The 5-HT2A antagonists reduced the amplitude of the L5 ventral root discharges when applied below L2, suggesting that many of the cells possessing 5-HT2A receptors involved in locomotor-like activity are part of the output rather than the rhythm-generating component of the locomotor system. These output cells may be the motoneurons themselves or premotor interneurons driven by of forming part of the CPG. A substantial portion of the effects of 5-HT2A antagonists are likely to be directly on motoneurons, because immunolabeling for these receptors is particularly dense in lamina IX around motoneurons (Cornea-Hébert et al. 1999; Doly et al. 2004). Moreover, the excitatory actions of 5-HT on motoneurons can be blocked by 5-HT2A antagonists (Gilmore and Fedirchuk 2004; Jackson and White 1990; Takahashi and Berger 1990; Wang and Dun 1990).
There is evidence that maturation of 5-HT effects is responsible for the development of repetitive firing properties in extensor but not flexor motoneurons (Pflieger et al. 2002). In the chick spinal cord, 5-HT fibers were differentially distributed among motor nuclei, and dense clusters of 5-HT terminals were found preferentially on extensor motoneurons (Okado et al. 1988). Hounsgaard et al. (1988) found that bistable properties were induced in extensor but rarely in flexor motoneurons of spinal cats by intravenous 5-hydrpxytryptophan. In our study, 5-HT2A receptor antagonists, when applied above L3, have no effect on locomotor-like activity, even though the antagonists were applied in the compartment containing the flexor motoneurons recorded in the L2 ventral root. This could be explained if flexor motoneurons do not possess 5-HT2A receptors. Alternatively, 5-HT2A receptors on flexor motoneurons may be functionally immature in the neonatal rat. There is currently no information on the distribution of 5-HT2A receptors on motoneurons innervating specific muscles, although it is clear that not all motoneurons possess 5-HT2A receptors (Doly et al. 2004).
Dopaminergic (D1, D2) and noradrenergic (1, 2) receptors are not necessary for PPR induced locomotor-like activity
Dopaminergic and noradrenergic agonists can induce rhythmic activity in rodent in vitro preparations (Gabbay and Lev-Tov 2004; Jiang et al. 1999; Kiehn and Kjaerulff 1996; Kiehn et al. 1999; Whelan et al. 2000). In the functionally mature neonatal mouse preparation, dopamine was required for chemically evoked locomotion (Jiang et al. 1999). Madriaga et al. (2004) showed that dopaminergic D1 and D2 receptors potentiate the ability of 5-HT to evoke locomotor activity in the neonatal mouse spinal cord. In the case of PPR-evoked locomotor-like activity, however, neither D1 nor D2 antagonists had an effect, indicating that these dopaminergic receptors do not contribute to PPR-evoked locomotor-like activity.
The 1 noradrenergic antagonist prazosin blocked noradrenaline-evoked rhythmic activity in the isolated neonatal rat spinal cord (Gabbay and Lev-Tov 2004) but was without effect in the case of PPR-evoked locomotor-like activity in our experiments. High doses of the 2 antagonist yohimbine also blocked noradrenaline-induced locomotor-like activity (Gabbay and Lev-Tov 2004), and our results with yohimbine were similar for PPR-evoked locomotor-like activity. However, the absence in our experiments of any blocking effect of the more specific 2 antagonist RX 82002 (Clarke and Harris 2002) does not confirm a role for 2 receptors in PPR-evoked locomotor-like activity. Yohimbine is known to have a marked affinity for the 5-HT1A receptor (Newman-Tancredi et al. 1998), where it acts as an agonist. There is evidence that 5-HT, acting through 5-HT1 receptors, can inhibit the locomotor rhythm in the isolated rat spinal cord (Beato and Nistri 1998). Perhaps this explains the locomotor blocking action of yohimbine in our experiments. We argue, therefore, that neither 1 nor 2 noradrenergic receptors are required for PPR-evoked locomotion. The ability of yohimbine at low concentrations to shorten the locomotor cycle is consistent with the finding that 2 noradrenergic agonists decrease the frequency of NMDA-induced locomotor-like activity (Sqalli-Houssaini and Cazalets 2000) and suggests that these receptors can also modulate locomotion induced by PPR stimulation.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: L. M. Jordan, Dept. of Physiology, Spinal Cord Research Ctr., Univ. of Manitoba, 730 William Ave., BMSB 425, Winnipeg MB R3E 3J7, Canada (E-mail: larry{at}scrc.umanitoba.ca)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Atsuta Y, Abraham P, Iwahara T, Garcia-Rill E, and Skinner RD. Control of locomotion in vitro: II. chemical stimulation. Somatosens Mot Res 8: 55–63, 1991.[ISI][Medline]
Atsuta Y, Garcia-Rill E, and Skinner RD. Electrically induced locomotion in the in vitro brainstem-spinal cord preparation. Dev Brain Res 42: 309–312, 1988.[CrossRef]
Atsuta Y, Garcia-Rill E, and Skinner RD. Characteristics of electrically induced locomotion in rat in vitro brain stem-spinal cord preparation. J Neurophysiol 64: 727–735, 1990.
Atsuta Y, Garcia-Rill E, and Skinner RD. Control of locomotion: I. deafferentation. Somatosens Mot Res 8: 45–53, 1991.[ISI][Medline]
Basbaum AI, Clanton CH, and Fields HL. Three bulbospinal pathways from the rostral medulla of the cat: an autoradiographic study of pain modulating systems. J Comp Neurol 178: 209–224, 1978.[CrossRef][ISI][Medline]
Beato M and Nistri A. Serotonin-induced inhibition of locomotor rhythm of the rat isolated spinal cord is mediated by the 5-HT1 receptor class. Proc R Soc Lond B Biol Sci 265: 2073–2080, 1998.[Medline]
Bengtson CP, Lee DJ, and Osborne PB. Opposing electrophysiological actions of 5-HT on noncholinergic and cholinergic neurons in the rat ventral pallidum in vitro. J Neurophysiol 92: 433–443, 2004.
Bertrand S and Cazalets JR. The respective contribution of lumbar segments to the generation of locomotion in the isolated spinal cord of newborn rat. Eur J Neurosci 16: 1741–1750, 2002.[CrossRef][ISI][Medline]
Bonnot A, Whelan PJ, Mentis GZ, and O'Donovan MJ. Locomotor-like activity generated by the neonatal mouse spinal cord. Brain Res Brain Res Rev 40: 141–151, 2002.[CrossRef][Medline]
Bowker RM, Westlund KN, and Coulter JD. Origins of serotonergic projections to the spinal cord in rat: an immunocytochemical-retrograde transport study. Brain Res 226: 187–199, 1981.[CrossRef][ISI][Medline]
Bracci E, Beato M, and Nistri A. Extracellular K+ induces locomotor-like patterns in the rat spinal cord in vitro: comparison with NMDA or 5-HT induced activity. J Neurophysiol 79: 2643–2652, 1998.
Cazalets JR, Borde M, and Clarac F. Localization and organization of the central pattern generator for hindlimb locomotion in newborn rat. J Neurosci 15: 4943–4951, 1995.[Abstract]
Cazalets JR, Borde M, and Clarac F. The synaptic drive from the spinal locomotor network to motoneurons in the newborn rat. J Neurosci 16: 298–306, 1996.
Cazalets JR, Grillner P, Menard I, Cremieux J, and Clarac F. Two types of motor rhythm induced by NMDA and amines in an in vitro spinal cord preparation of neonatal rat. Neurosci Lett 111: 116–
) are concentrated in the para-pyramidal region, among the para-raphe 5-HT neurons. Other effects observed were alternating discharge of 1 or 2 ventral roots of 3- to 5-s duration (
), increased tonic activity in 1 or more ventral roots (
), or no response (
). Sites of the pyramidal tract (PT), the nucleus of the VIIth cranial nerve (VII), the gigantocellular reticular nucleus, pars alpha (GiA), the nucleus raphé obscurus (ROb), and the nucleus raphé pallidus (RPa) are indicated (modified from Paxinos et al. 1991
