| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Universités Bordeaux 1 and 2, Centre National de la Recherche Scientifique, Bordeaux, France
Submitted 29 November 2007; accepted in final form 11 March 2008
 |
ABSTRACT |
|---|
|
|
|
INTRODUCTION |
|---|
|
). Although the cellular and synaptic mechanisms engaged in respiratory rhythm generation still remain to be completely elucidated, our understanding of respiratory circuit operation has benefited greatly from the study of various in vitro preparations of the neonatal rat and mouse (for review, see Ballanyi et al. 1999; Feldman and Del Negro 2006; Funk and Feldman 1995; Onimaru et al. 1997). Since the pioneering studies of the respiratory system in isolated brain stem-spinal cord preparations (Smith and Feldman 1987; Suzue 1984), attention over the last two decades has focused mainly on the central neural mechanisms responsible for the inspiratory phase of respiration. In this perspective, accumulated findings have shown that inspiration results from a combination of emergent network and intrinsic membrane properties (reviewed by Ramirez and Viemari 2005) and that medullary neurons likely to be involved in respiratory rhythm generation are located within two distinct but interconnected brain stem populations, the pre-Bötzinger complex (Smith et al. 1991) and the retrotrapezoid nucleus/parafacial respiratory group (RTN/pFRG) (Barnes et al. 2007; Mellen et al. 2003; Onimaru and Homma 2003; Onimaru et al. 2006). However, because expiration has often been considered as a passive component of the respiratory cycle at rest and because it occurs only occasionally in caudal thoracic ventral roots in brain stem-spinal cord preparations in vitro (Smith et al. 1990) or only in response to an acidosis of the extracellular environment (Iizuka 2004), the site(s) of generation, the spatiotemporal organization of spinal motor activity (in particular at the thoraco-lumbar level) and the neural mechanisms underlying the expiratory phase of respiration are less well documented.
Although the central respiratory rhythm-generating networks continue to function alone in isolated brain stem-spinal cord or slice preparations, in intact animals including humans, the respiratory circuitry must interact with other major functions (for example, cardiovascular, thermoregulatory, locomotory) to adapt rhythmic breathing movements to changing environmental and behavioral demands. During vertebrate locomotion, for example, a 1:1 coupling between locomotory and respiratory cycles has been observed in several species of bipeds and quadrupeds when running or galloping speeds have been attained (Boggs 2002; Bramble and Carrier 1983). Furthermore, we have previously reported that in isolated brain stem-spinal cord preparations from neonatal rats, the rhythmic activation of hindlimb sensory pathways by electrical stimulation of lumbar dorsal roots (DR) can reset and entrain respiratory rhythmicity (Morin and Viala 2002). On this basis, we proposed that, in quadrupeds at least, hindlimb somatic sensory inputs could provide timing information to the respiratory rhythm generators to couple the frequency of breathing movements to the locomotor cycle. Moreover, both short-latency excitatory and GABA-mediated inhibitory postsynaptic potentials are produced in inspiratory phrenic motoneurons (PMNs) in response to activation of lumbar afferent pathways (Morin and Viala 2002). Although the exact role of these apparently direct lumbar synaptic inputs to PMNs remains to be established, we proposed that they could contribute to respiratory entrainment by modulating the sensitivity of PMNs to the descending excitatory drive from the medullary respiratory rhythm-generating centers. Whether functioning in synergy or in alternation with the PMNs, other spinal neuronal populations, such as intercostal and abdominal motoneurons, participate actively in respiration to produce inspiratory movements of the rib cage or to facilitate expiration, respectively (reviewed by Iscoe 1998; Monteau and Hilaire 1991). However, to date the actions of limb somatic afferents on intercostal and abdominal respiratory motoneuron activity have not been investigated. Moreover, although stimulation of forelimb somatic afferent pathways can increase the cycle frequency of phrenic nerve discharge (Potts et al. 2000) and entrain the respiratory rhythm (Potts et al. 2005) in heart-brain stem preparations from 6- to 8-wk-old rats, the effects of forelimb sensory inputs on respiration during the perinatal period remain to be determined.
In the present study, in vivo neuroanatomical labeling and in vitro electrophysiological approaches with brain stem-spinal cord preparations were first used to determine the position of intercostal and abdominal motoneurons in the neonatal rat spinal cord. Next we characterized the spatiotemporal organization of all inspiratory (intercostal)- and expiratory (intercostal and abdominal)-related motor patterns along the spinal cord. Finally, the influence of fore- and hindlimb sensory inputs on thoraco-lumbar respiratory motor activity was examined, as was the nature of the intraspinal pathways that mediate such interactions. Part of this work has been presented previously in abstract form (Giraudin et al. 2006).
|
|
METHODS |
|---|
|
Retrograde labeling of intercostal and abdominal motoneurons
Motoneurons innervating intercostal and abdominal muscles were localized in the spinal cord by means of retrograde staining. Newborn animals were anesthetized by hypothermia, and small incisions in the skin were made to expose intercostal or abdominal muscles. Crystals of the cholera toxin subunit B coupled to Alexa Fluor 488 (CTX-AF 488; Molecular Probes; Eugene, OR) were placed on a selected muscle with a pin and then the wound was left to dry for 5 min and disinfected with povidone-iodine (Betadine). Animals were allowed to recover from anesthesia and returned to their nest for 24 h. Spinal cords were then dissected (see following text) and fixed in a 4% paraformaldehyde solution diluted in 0.1M phosphate buffer (PB; pH = 7.4) for 2 h at room temperature or were stored overnight at 4°C. The cords were dehydrated through a graded ethanol series (50, 70, 95, and 100°) and cleared for 2 h in methylsalicylate at room temperature. Spinal cords were then mounted in Fluoromount and observed under a fluorescent microscope (DMRB; Leica) at 488 nm, and images were acquired using a CCD camera (Sony DXC-990P).
In vitro isolated brain stem-spinal cord preparations
Animals were deeply anesthetized and decerebrated from the rostral end of the fifth cranial nerves. The skin and muscles were removed and preparations were then placed in a 100-ml chamber filled with artificial cerebrospinal fluid containing (in mM): 100 NaCl, 4 KCl, 1.2 NaH2PO4, 2 CaCl2, 1.3 MgCl2, 25 NaHCO3, and 30 D-glucose. This standard saline was continuously equilibrated with 95% O2-5% CO2 (pH 7.4) and maintained at 10°C during the dissection. The brain stem and spinal cord with its dorsal and ventral roots still attached were isolated, then the ensemble was placed in a 10 ml recording chamber and fixed on a silicone elastomer (Sylgard) resin block with the ventral surface upwards. The bath temperature was progressively raised and thereafter maintained at 26°C by means of a Peltier system. A postdissection resting period of 30 min was systematically respected before recording procedures began.
Electrophysiology
Respiratory-related activity in both spinal ventral roots and cranial nerves was recorded using glass suction electrodes. Signals were amplified (10,000 times) by differential AC amplifiers (AM System; Phymep, Paris), band-pass-filtered (0.1–3 kHz), rectified, integrated (
= 20 ms; Neurolog System; Digitimer), digitized and stored on a computer hard disk using Spike 2 software (Cambridge Electronic Design, Cambridge, UK) for off-line analysis. Single or train stimulus pulses (0.2–2 V, 0.5 ms at 5–10 Hz) were applied respectively to ventral or dorsal spinal roots via glass suction electrodes using an eight-channel digital stimulator (AMPI; Jerusalem, Israel). The stimulation of dorsal roots followed the procedure used by Morin and Viala (2002), which was based on their finding that the activation of large diameter, and therefore presumed proprioceptive, dorsal root axons in isolated in vitro preparations requires relatively low stimulus intensities with thresholds consistently ranging from 0.6 to 1.1 V. Accordingly in our experiments, the threshold stimulus for so-called "low-threshold" DR afferents was determined by increasing the train shock intensity from a subthreshold value until either a maximum of 1.5 V or an intervening level at which an effect on the timing of the subsequent cycle of ongoing spontaneous respiratory rhythmicity had occurred. Once the latter was observed, and unless otherwise stated in the following, this threshold stimulus intensity was applied to a given root throughout the course of the experiment.
In a series of experiments, a stimulating tungsten electrode (Frederick Haer; Brunswick, ME) was manipulator-positioned in contact with the ventrolateral surface of the brain stem to stimulate (single shocks of 300 µA, 0.1 ms) or to electrolytically destroy (400 µA for 3–5 s) the region containing the pre-Bötzinger complex. The final position of the stimulating electrode was adjusted using the hypoglossal root exits as anatomical landmarks.
Drug application
To reversibly isolate the lumbar spinal cord from the cervical region, axonal conduction in the thoracic spinal segments was prevented by means of sucrose blockade. For this, the recording chamber was partitioned into three compartments with barriers of syringe-ejected petroleum jelly (Vaseline) and the intermediate thoracic cord compartment (containing 
4 contiguous spinal segments) was irrigated with an isotonic sucrose solution (10% in distilled water). The narrow petroleum jelly bridges allowed the cord to remain functionally connected between the different baths, and water tightness was checked at the end of each experiment by adding methylene blue to the bathing medium on either side of a given bridge.
In other experiments, a low calcium concentration (0.1 mM CaCl2, 5 mM MgCl2) solution, which has been found previously to block chemical synaptic transmission in the neonatal rat spinal cord (Cazalets 2005; Morin and Viala 2002; Tresch and Kiehn 2000), was used to reversibly attenuate synaptic transmission in the mid-cord region using the same bath partitioning and verification procedures.
EMG and semi-isolated preparations
In a further set of experiments, a spinal cord-rib cage preparation was used to make electromyographic (EMG) recordings from intact intercostal or abdominal muscles. For this, the spinal cord after brain stem removal was exposed as described in the preceding text but with all ventral and dorsal roots left uncut. The preparation was then pinned down ventral side upwards and bipolar EMG electrodes (made from 50 µm silver wire) were inserted into either external intercostal muscles of the rib cage or medial or lateral abdominal muscles.
Brain stem transections and histological controls
In several isolated brain stem-spinal cord preparations, expiratory activity was recorded from thoracic ventral roots while caudally directed serial transverse sections of the brain stem were performed with a scalpel blade to locate pontine structures involved in the genesis of the expiratory phase of the respiratory cycle. Recordings were performed 
100 min after each section to allow recovery of the preparation. For histological verification of the lesion location, the brain stem region caudal to the section (see Fig. 5C) was fixed for 48H at 4°C in Lillie solution (10% formalin in PB; pH = 7.0). The tissue was then rinsed twice in PB and cryoprotected overnight with 25% sucrose in PB. After embedding in Tissue Tek and freezing by using cooled isopentane until –80°C, frozen 40-µm-thick serial parasagittal sections were cut on a cryostat. Sections were mounted on gelatin-coated slides, and adjacent sections were alternately stained for acetylcholinesterase (AchE) or with cresyl violet to visualize the approximate boundaries of the pontine nuclei. For AchE staining, sections were slide-mounted and left to dry at room temperature (RT). They were then rinsed in 0.2 M acetate buffer (pH = 5.9) for 5 min before being immersed in 0.2 M acetate buffer containing 0.04 M glycine and 0.01 M copper sulfate pentahydrate for 18 h under agitation at RT. The slides were incubated in the same solution containing 1% acetylthiocholine iodide for 2 min under agitation at RT and rinsed three times in acetate buffer. They were then dehydrated in an ascending ethanol series, cleared in two changes of xylene and mounted in Eukitt before observation under microscopy (DMRB; Leica).
|
Inspiratory- and expiratory-related motor activities were characterized in terms of the onset delays and durations of spinal ventral root bursts. For this, 20–30 burst cycles were chosen randomly from sequences of respiratory activity recorded from each preparation. Burst onset delay was taken as the interval (in ms) between the beginning of an inspiratory burst in the C5 ventral root and the onsets of corresponding bursts recorded in more caudal roots. The absolute duration of each burst was also measured. Data analysis was performed with Spike 2 software (Cambridge Electronics Design) and pooled burst onset delay and duration measurements for a given ventral root were expressed as means ± SE. Differences between means were analyzed using a statistical software package (Sigma Stat for Windows; SPSS, Chicago, IL) and assessed by one-way ANOVA with a Student-Newman-Keuls posttest. In a further analysis, values of thoracic burst durations were plotted against the corresponding burst onset delay, and a linear regression line was fitted to the scatter plot. The coefficient of linear regression (r) was calculated and the statistical significance was established using the Pearson test (Prism 4 for Windows, GraphPad software).
In experiments conducted to determine the effect of brain stem stimulation on the ongoing respiratory rhythm, the resultant changes in burst timing were expressed in phase-response plots. For this, the reference period (Pref) was taken as the mean cycle period over three spontaneous respiratory cycles prior to stimulation. The ratio of the stimulus latency (time elapsed from the stimulus to the onset of the ensuing evoked inspiratory burst) and Pref then provided the stimulus phase. The shift in phase of the inspiratory burst, expressed as the difference between Pref and the period in which the stimulation occurred and again divided by Pref, was plotted on the ordinate. Differences in mean values for all parameters were taken to be significant at P < 0.05.
|
|
RESULTS |
|---|
|
To determine the location of intercostal motoneurons in the neonatal rat spinal cord, Alexa Fluor 488-conjugated cholera toxin (CTX-AF 488; Fig. 1) was inserted unilaterally into external muscles of the 2nd (n = 4), 5th (n = 4), 8th (n = 4), and 11th (n = 4) intercostal spaces of which there are 12 in the rat rib cage. An example of motoneurons stained after CTX-AF 488 insertion into the fifth intercostal space is shown in Fig. 1A. Retrograde-labeled motoneuron somata were distributed ipsilaterally along the cord in a single column that extended rostrocaudally from the cervical C6 to the thoracic T12 segments. The rib space position of an injected external intercostal muscle within the rib cage was reflected in the cell body location of its innervating motoneurons within the spinal cord. Thus more posterior intercostal muscles had more caudally distributed motoneuron somata (see Fig. 2C, left, filled bars). For a given CTX-AF 488 injected intercostal muscle, the majority of stained somata were located within one to two spinal segments (as in Fig. 1A) although in most preparations (12 of 16), some labeling extended rostrally and/or caudally into one or two immediately adjacent segments.
|
|
Although CTX-AF 488 crystals were applied directly to selected muscles, the possibility of nonspecific staining due to tracer leakage could not be totally excluded. In a series of electrophysiological experiments (n = 4), therefore, spinal cord preparation that included the rib cage and still-attached abdominal muscles (Fig. 2A) was used to identify individual cervico-thoracic ventral roots according to their specific intercostal and abdominal muscle targets. This was achieved by applying single pulse stimuli to successive motor roots while monitoring activation of a specific muscle by EMG recording (Fig. 2B). In agreement with findings from the preceding anatomical experiments, this electrophysiological approach confirmed that motoneurons innervating external intercostal and abdominal muscles project their axons in C7 to T13 (Fig. 2C, left, unfilled bars) and in T9 to L1 (Fig. 2C, right, unfilled bars) ventral roots, respectively. Furthermore, on the basis of EMG signal amplitude, one and sometimes two ventral roots were found to be principally involved in the activation of a given muscle (for example, see the maximal response to T6 stimulation in the seventh intercostal muscle recordings of Fig. 2B). In all preparations studied, however, from three to seven adjacent ventral roots were also found to contribute to the motor command to a given intercostal (Fig. 2C, left) or abdominal (Fig. 2C, right) muscle, as seen in Fig. 2B where attenuated responses in the recorded intercostal muscle occurred with T5, T7, and T8 ventral root stimulation. The close coherence between results obtained with the two anatomical and electrophysiological approaches (compare data in Fig. 2C) therefore confirmed the somatotopic rostrocaudal organization of intercostal and abdominal motoneurons in the spinal cord.
Spatiotemporal organization of spinal inspiratory and expiratory motor activity
Following the pioneering studies of Suzue (1984) and Smith and Feldman (1987) on the respiratory motor system of the neonatal rat brain stem-spinal cord in vitro, a rostrocaudal gradient in spinal respiratory motor outputs was described in which rhythmic bursting in both inspiratory and expiratory phases can occur at various levels along the cord (Iizuka 2004; Smith et al. 1990). Here we wished to complete these earlier descriptions and to determine, when present, the function and the supra-spinal origin of spontaneous respiratory-related output at all cervical to lumbar cord segments.
Respiratory activity in isolated medullary-spinal cord preparations (with the brain stem sectioned at the level of the X cranial nerves; n = 9) consists of cyclic motor bursts occurring conjointly in cranial (hypoglossal) nerves (see Fig. 4A) and in spinal ventral roots throughout the cervico-thoracic cord (Figs. 3A and 4A). As illustrated in Fig. 3A, single shock stimulation (300 µA, 0.1 ms) applied to the surface of the ventrolateral medulla, a region that includes the pre-Bötzinger complex that is thought to be critically involved in the inspiratory phase of respiratory rhythmogenesis (Smith et al. 1991), caused a resetting of ongoing bursts recorded simultaneously from phrenic (C4) and intercostal (T4) ventral motor roots (Fig. 3A, 1 and 2). Consistent with typical features of endogenous biological oscillators (Pinsker 1977), premature (Fig. 3A1) or retarded (Fig. 3A2) spinal root bursts were triggered by the stimulation depending on the phase at which the stimulus occurred in the ongoing rhythm cycle (Fig. 3B). Thus when a stimulus was applied relatively late in the interburst interval (at >50% of the elapsed cycle), the onset of the next burst was advanced causing the ensuing rhythm to be phase-advanced (Fig. 3, A1 and B, right). By contrast, a relatively early stimulus (at <50% of the cycle) retarded the next spontaneous burst and consequently caused a phase-delay in the timing of subsequent cycles (Fig. 3, A2 and B, left). Finally, an electrolytic lesion (400 µA for 3–5 s) performed at the site of stimulation abolished all burst discharge in the C4 and T4 roots (Fig. 3C), thereby further confirming the inspiratory nature of the recorded cervical and thoracic motor activities.
|
|
In ponto-medullary-spinal cord preparations (n = 17), simultaneous recordings from lower thoracic (T12 and T13) ventral roots that carry abdominal motor axons (see Fig. 2C, right) also displayed spontaneous rhythmic bursts that now occurred in alternation with C5 inspiratory discharge (Fig. 5A). Because of their out-of-phase relationship with phrenic discharge and the positions of the spinal ventral roots where they occurred, these thoracic motor bursts were considered to be expiratory in function. Transections of the brain stem at different levels were then conducted to determine the location of the supra-spinal structures involved in the generation of this expiratory-related activity. In a first step, a single transection was performed just rostral to the X cranial nerves to accomplish a global removal of the brain stem pontine structures (n = 10 preparations). In all cases, an increase in respiratory cycle rate that is classically attributed to the removal of the A5 noradrenergic nucleus (Hilaire et al. 1989) was observed. Significantly, moreover, in such reduced medullary preparations, expiratory-related activity was no longer expressed (Fig. 5, B and D, lesion c). Therefore in a second series of experiments on initially intact ponto-medullary-spinal cord preparations, consecutive transverse sections of the brain stem were made in the rostrocaudal direction to more precisely define the location of the pontine structure(s) involved in expiratory-phase rhythm generation. In all seven preparations examined, expiratory-like activity persisted until a section was made just caudally to the VII cranial nerves (Fig. 5, C and D, lesion b). The exact rostrocaudal positioning of this lesion was also verified histologically following physiological experimentation. For this, we made parasagittal sections of the caudal brain stem and stained the tissue for acetylcholinesterase (AchE) or with cresyl violet to visualize the approximate boundaries of remaining pontine motor nuclei (see METHODS) (Fernandes et al. 1998). As seen in Fig. 5C, which shows two AchE-stained parasagittal sections taken at different planes (1 and 2 in right) from the same brain stem, the labeled facial nucleus provided a clear landmark for the slightly more rostral position of the original in vitro brain stem transection. The preceding findings strongly suggest therefore that the brain stem area in the vicinity of the VII motor nucleus in the neonatal rat participates either directly (i.e., in actual rhythmogenesis) or indirectly (i.e., as a necessary relay) in the production of spinal expiratory-phase activity.
To assess the spatiotemporal organization of expiratory activity along the cord, simultaneous ventral root recordings were made from cervical to lumbar cord levels in brain stem-spinal cord preparations that now included the pontine structures (Fig. 6, A and B). Here again, the onsets and durations of individual root discharges were compared with C5 motor bursts as the reference for the inspiratory activity phase of each cycle. Such experiments (n = 10) revealed that expiratory bursting occurred from T8 to L2 ventral roots (Fig. 6, B, traces at left, and C, filled histogram bars in left) in agreement with our previous conclusions (see Fig. 2C, right) that these motor roots innervate the abdominal muscles. Typically the spinal expiratory discharge occurred in a double bursting pattern that consisted of a short preinspiratory burst (mean duration: 0.26 ± 0.02 s) and a prolonged postinspiratory discharge (mean duration: 1.8 ± 0.1 s). However, when a double inspiratory burst in a cycle occasionally occurred spontaneously (Fig. 6B, traces at right), expiratory-related activity was recorded in otherwise solely inspiratory (T5–T7) or previously quiescent (L3) ventral motor roots (Fig. 6C, filled histogram bars in right). Therefore the isolated pons-medulla-spinal cord preparation is capable of generating complex and spatiotemporally variable patterns of fictive respiration in the ensemble of functionally identified (inspiratory, mixed inspiratory-expiratory, and expiratory) motor roots throughout the cervical-to-lumbar cord region.
|
We have previously reported that in the neonatal rat, a phrenic respiratory entrainment-induced polypnea can be achieved by phasic low-threshold lumbar afferent stimulation in the isolated brain stem-spinal cord preparation (Morin and Viala 2002). On this basis, it was proposed that in quadrupeds, hindlimb somatic afferent inputs could serve normally to couple breathing frequency to the locomotor rhythm. Two unaddressed questions that arose from these earlier findings were, first, whether such a respiratory entrainment could also be driven by the cyclic activation of cervical low-threshold sensory afferents originating from forelimb muscles, and second, how intercostal and abdominal respiratory motor outputs might respond to both cervical and lumbar somatic inputs.
To explore these issues, hindlimb and forelimb afferent pathways were rhythmically activated by applying electrical stimuli (0.5 to 1 s pulse trains over a range of 0.2–1.5 V at 10 Hz; see METHODS) to lumbar (L2–L5) or cervical (C7–C8) DRs, respectively, in isolated pontomedullary-spinal cord preparations in which respiratory activity was monitored simultaneously from the C4 (phrenic), T8 (intercostal) and T13 (abdominal) ventral roots (Fig. 7, see schematics). In such preparations (n = 13), the rhythmic activation of either low-threshold lumbar (Fig. 7A) or cervical (Fig. 7B) afferents was able to fully entrain (with 1:1 coupling) ongoing respiratory bursting at all three cord levels. As already described (Morin and Viala 2002), this entrainment derived from a resetting action of the DR stimulation, which was characterized by a near constant interval between each stimulus train and the respiratory burst it evoked (mean C4 burst latency after lumbar DR stimulation: 1.56 ± 0.05 s; Fig. 7A; after cervical DR stimulation: 1.34 ± 0.06 s; Fig. 7B). Importantly, no differences were observed between spontaneous and DR stimulation-induced burst patterns (compare traces in Fig. 7, A and B, with those in Fig. 6B) with both beginning by a brief preinspiratory discharge in thoracic ventral roots. Together these results suggest that entrainment of respiratory activity in cervical (phrenic) and thoraco-lumbar (intercostal and abdominal) motoneurons by lumbar or cervical DR afferent stimulation is mediated by long-loop neuronal pathways via the respiratory rhythm-generating centers in the brain stem (see DISCUSSION).
|
|
|
|
|
DISCUSSION |
|---|
|
) and newborn rats (Ruangkittisakul et al. 2007), thoraco-lumbar expiratory-like activity is abolished by a brain stem transection performed at the rostral boundary of the parafacial nucleus. This implies, therefore, that the rostral pFRG and more anterior pontine structures are necessary for the production of the expiratory phase of respiration. Finally, our study provides new evidence that somatic sensory pathways from both the forelimbs and hindlimbs have direct access to the medullary respiratory centers, thereby providing the substrate for coupling the ensemble of spinal respiratory outputs (including phrenic, intercostal and abdominal motoneurons) to ongoing locomotory movements. Intercostal motoneurons and inspiratory-phase discharge
Intercostal motoneurons are located in discrete rostrocaudally aligned populations in the cervico-thoracic (C6–T13) region of the cord (Fig. 1). Although individual external intercostal muscles appear to receive their motor innervation principally from one or two cord segments, in most cases, a small number of motoneurons located in immediately adjacent rostral and/or caudal segments also contribute to a given intercostal muscle's motor command (Fig. 2). This anatomical organization corresponds closely to that reported in adult cat and rat (reviewed by Monteau and Hilaire 1991), although it differs from the segmental specificity of intercostal motoneuron distributions in the developing chicken embryo (Stirling et al. 1995).
In further agreement with previous studies (Iizuka 2004; Smith et al. 1990), our neonatal rat experiments have shown that all spinal ventral roots containing external intercostal motor axons (i.e., from C6 to T12) continue to express respiratory-related bursting even after the brain stem-spinal cord has been isolated in vitro (Figs. 3 and 4). Multi-site nerve recordings and resetting experiments involving direct medullary stimulation confirmed the inspiratory nature of this rhythmic cervico-thoracic activity. First the intercostal motoneurons located in this cord region fire in phase with C5 ventral roots which normally convey motor axons to the main inspiratory muscle, the diaphragm. Second, as reported in fetal (Di Pasquale et al. 1994) and neonatal rats (Onimaru et al. 1988), electrical stimulation of the rostral ventro-lateral medulla causes a resetting of rhythmicity at both cervical and thoracic levels (Fig. 3), presumably via a direct perturbation of PreI neurons (Onimaru et al. 1988) in the pre-Bötzinger complex, the inspiratory rhythm generator (Smith et al. 1991).
A detailed temporal analysis of these cervico-thoracic inspiratory outputs also showed that the intercostal motoneurons are activated differentially in the respiratory cycle and, on the basis of their burst onsets, endings and durations, can be divided into three populations (Fig. 4). Inspiratory intercostal motoneurons that fire longer bursts are generally recruited earlier in the inspiratory phase of each cycle and are therefore located in more rostral spinal segments, whereas intercostal burst onsets and durations become respectively later and shorter with distance down the cord. The functional significance of the discontinuity in rostrocaudal activation of the intercostal motoneurons remains unclear. Presumably, this reflects regional differences in the organization of descending respiratory pathways in the spinal cord and/or is related to a functional requirement for intercostal muscles to produce discrete regional synergisms in their control of rib cage movements.
Thus the rostrocaudally graded activation of intercostal muscles that was previously described in cat (Greer and Martin 1990), dog (De Troyer and Legrand 1995), and humans (De Troyer et al. 2003) also appears to occur in the neonatal rat. In a functional context, moreover, such a decrementing recruitment pattern would ensure that the inspiratory intercostal muscles with the greater breathing mechanical advantage (i.e., those located in the rostral intercostal spaces of the rib cage) are the most rapidly and extensively activated during each respiratory cycle (for review, see De Troyer et al. 2005).
Although the mechanisms underlying this sequential activation of inspiratory intercostal motoneurons remain unknown, differences in excitability related to cell size or intrinsic membrane properties, and/or a differential modulation of the descending respiratory drive to intercostal motoneurons could be involved (Gandevia et al. 2006). Clearly, however, the possible contribution of peripheral afferent inputs in this recruitment pattern can be eliminated, at least in the case of our isolated medullary-spinal cord preparations. Our experiments have also provided evidence that the descending respiratory drive to intercostal motoneurons in the lower thoracic spinal region is mediated at least in part by indirect segmentally relayed pathways (Fig. 9; see following text) as was recently reported in the upper thoracic segments of the neonatal rat (Juvin and Morin 2005). It is likely therefore that the propriospinal interneurons engaged in the polysynaptic respiratory drive to upper cord segments may also participate in the rostrocaudal activation of the entire intercostal motoneuron population.
Abdominal motoneurons and expiratory-phase discharge
Expiratory-phase activity occurring in alternation with inspiration is also expressed spontaneously in caudal thoracic (T8) to lumbar (L2) segments, with expiratory bursts occurring alone from T13 to L2 (Fig. 6). Combined anatomical and electrophysiological evidence established that these spinal roots contain motor axons that innervate abdominal muscles (Figs. 1 and 2). This conclusion is also supported by previous neonatal rat experiments showing that L1 ventral root activity recorded in vitro corresponded to abdominal EMG activity recorded in vivo (Janczewski et al. 2002). Moreover, with a sudden reduction in inspiratory cycle period during fictive respiration (i.e., when 2 inspiratory bursts per cycle are produced spontaneously), expiratory discharge can appear in more rostral ventral roots (T5–T7) that otherwise appeared to be uniquely inspiratory in function. Presumably this auxiliary recruitment reflects an adaptive ability of the central respiratory system to reinforce expiratory-phase activity in caudal intercostal motoneurons when breathing frequency is suddenly increased. Furthermore, in adult rats, firing patterns in the T6–T8 spinal segments have been shown to be mostly expiratory in function (Tian and Duffin 1996) with the more caudal interspaces of the rib cage having basically an expiratory mechanical advantage (De Troyer et al. 2005). Our present results are therefore consistent with, and add to, the first description of the spatiotemporal patterns of motoneuronal activity during fictive respiration (Smith et al. 1990): respiratory-generating networks in vitro are able to produce a complex pattern of respiration, consisting of alternating inspiratory- and/or expiratory-phase outputs from the cervical to lumbar regions of the spinal cord.
According to these and other authors, expiratory-phase discharge occurs either occasionally in caudal thoracic roots in vitro (Smith et al. 1990) or only in response to central chemoreceptor stimulation (Iizuka 1999, 2004). In contrast to our investigation, however, in these earlier studies the pons was sometimes conserved (Smith et al. 1990) or was consistently removed (Iizuka 1999, 2004). In juvenile vagotomized rats, a transection performed at the caudal end of the facial nucleus that thereby removes the pons also eliminates abdominal EMG expiratory activity while inspiratory bursts are preserved (Janczewski and Feldman 2006) (also see lesion c in Fig. 5D). In agreement with these latter findings and recently reported lesion experiments on the isolated brain stem of newborn rats (Ruangkittisakul et al. 2007), our study has shown that all thoraco-lumbar expiratory-related activity disappears without affecting inspiratory discharge when a brain stem transection is performed near the rostral margin of the parafacial nucleus (Fig. 5), an area that probably includes elements of the RTN/pFRG respiratory networks. Although the structural boundaries of the pFRG have not been clearly established, this brain stem region is known to contain preinspiratory neurons (Onimaru et al. 1987, 1988), which fire in a characteristic pattern consisting of a short preinspiratory burst and longer postinspiratory discharge. This pattern is strikingly similar to the double (preinspiratory and expiratory) bursting activity we observed in T13–L2 motoneurons. Although pFRG preinspiratory neurons do not contact spinal motoneurons directly, they send projections to the caudal ventro-lateral medulla (Janczewski et al. 2002) wherein lies the nucleus retroambiguus, a group of bulbospinal neurons known to monosynaptically excite abdominal expiratory motoneurons (Boers et al. 2006). This therefore accounts for the elimination of spinal expiratory-phase discharge but not inspiratory bursts, that occurred after RTN/pFRG removal (Fig. 5) and is further consistent with current doctrine (Feldman and Del Negro 2006) that mammalian respiratory rhythmogenesis resides with two interconnected oscillatory networks in the RTN/pFRG and the pre-Bötzinger complex, which are responsible respectively for the expiratory and inspiratory phases of respiration.
Respiratory rhythm entrainment by cervical and lumbar somatic afferents
As mentioned in the preceding text, the respiratory system must interact continuously with other CNS regions to produce respiratory patterns adapted to changing behavioral circumstances. During fast locomotion, for example, a variety of mammals coordinate respiratory rhythmicity with ongoing limb movements (Boggs 2002; Bramble and Carrier 1983; Viala 1997). It is now well established that diverse combinations of mechanical (Bramble and Carrier 1983) and neurogenic (Eldridge et al. 1981; Romaniuk et al. 1994; Viala et al. 1987) interactions between the respiratory and locomotor systems are likely to underlie the coupling of these two primary functions. Limb movement-activated sensory inputs have access to the respiratory system (Funk et al. 1992; Palisses et al. 1988) and cyclic somatic afferent stimulation can fully entrain inspiratory phrenic nerve activity in juvenile (Potts et al. 2005) and adult animals (Iscoe and Polosa 1976). Similarly in the isolated brain stem-spinal cord from neonatal rats, periodic activation of low-threshold lumbar afferents in time with bursts of fictive locomotion leads immediately to locomotor-respiratory rhythm coupling (Morin and Viala 2002). However, at birth, freely moving rats use mainly their forelimbs for overground locomotion due to a lack of sufficient hindlimb postural tonus to support their body weight (Altman and Sudarshan 1975; Brocard et al. 1999; Clarac et al. 1998). Our present results now show that the phasic activation of low-threshold sensory inputs to the cervical (C7-C8) spinal cord region, which contains the forelimb locomotor networks (Ballion et al. 2001; Juvin et al. 2005, 2007), is also capable of entraining fictive respiration (Fig. 7). Furthermore we find that periodic electrical stimulation of either cervical or lumbar afferent pathways fully activates the different populations of inspiratory (phrenic and external intercostal) and expiratory (internal intercostal and abdominal) spinal motoneurons in a manner attributable to a resetting action on the brain stem respiratory generators themselves (see also Morin and Viala 2002; Potts et al. 2005). Interestingly, the two-burst patterning seen in abdominal expiratory motoneurons, which is also a particular feature of medullary preinspiratory neuron activity (Onimaru et al. 1987, 1988), is expressed during both spontaneous (Fig. 6B) and DR stimulation-entrained (Fig. 7B) fictive respiration. It is likely, therefore, that the preinspiratory neurons, which are thought to be responsible for the rhythmic activation of abdominal expiratory motoneurons (Janczewski et al. 2002; Onimaru et al. 1987, 1988; Ruangkittisakul et al. 2007), serve as preferential targets for limb afferent circuitry involved in locomotor-respiratory coupling.
Possible sensory-motor pathways that mediate locomotor-respiratory coupling
Figure 10 summarizes our overall findings and proposes neural substrates for locomotor-respiratory coupling in the neonatal rat and possibly quadrupedal mammals in general. During normal resting levels of respiration, a combination of direct-line and segmentally relayed inspiratory and expiratory drives are conveyed down the cord to the various cervical and thoraco-lumbar motoneuron pools responsible for rhythmic inspiratory (diaphragm and external intercostal) and expiratory (internal intercostal and abdominal) muscle contractions (Fig. 10A). During rapid locomotion, the cyclic contractions of fore- and hindlimb muscles activate somatic proprioceptors that in turn, via direct ascending spinal pathways, provide feedback signals that entrain the respiratory rhythm-generating networks through a phasic excitation of the expiratory generator (Fig. 10B). Consequently a coordination of locomotor and respiratory functions occurs in which stride and breathing cycle periods are coupled in a strict 1:1 phase relationship. Earlier evidence from intracellular recordings indicated that the ascending lumbar sensory pathways also provide complex collateral synaptic inputs to phrenic motoneurons en route to the brain stem (Morin and Viala 2002). However, whether the fore- and hindlimb sensory pathways also directly influence other more caudal spinal motor populations and afferent inputs from the four limbs operate independently or in combination and act on the same or different supra-spinal target(s), remain unanswered questions.
|
|
|
GRANTS |
|---|
|
|
|
ACKNOWLEDGMENTS |
|---|
|
|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: D. Morin, Université Victor Segalen Bordeaux 2, UMR CNRS 5227, Laboratoire Mouvement - Adaptation –Cognition, Bâtiment 2A, 146 Rue Léo Saignat, 33076 Bordeaux, France (E-mail: didier.morin{at}u-bordeaux2.fr)
|
|
REFERENCES |
|---|
|
Ballanyi K, Onimaru H, Homma I. Respiratory network function in the isolated brainstem-spinal cord of newborn rats. Prog Neurobiol 59: 583–634, 1999.[CrossRef][Web of Science][Medline]
Ballion B, Morin D, Viala D. Forelimb locomotor generators and quadrupedal locomotion in the neonatal rat. Eur J Neurosci 14: 1727–1738, 2001.[CrossRef][Web of Science][Medline]
Barnes BJ, Tuong C-M, Mellen NM. Functional imaging reveals respiratory network activity during hypoxic and opioid challenge in the neonate rat tilted sagittal slab preparation. J Neurophysiol 97: 2283–2292, 2007.
Boers J, Kirkwood PA, de Weerd H, Holstege G. Ultrastructural evidence for direct excitatory retroambiguus projections to cutaneous trunci and abdominal external oblique muscle motoneurons in the cat. Brain Res Bull 68: 249–256, 2006.[CrossRef][Web of Science][Medline]
Boggs DF. Interactions between locomotion and ventilation in tetrapods. Comp Biochem Physiol 133: 269–288, 2002.[CrossRef][Medline]
Bramble DM, Carrier DR. Running and breathing in mammals. Science 219: 251–256, 1983.
Brocard F, Vinay L, Clarac F. Development of hindlimb postural control during the first postnatal week in the rat. Dev Brain Res 117: 81–89, 1999.[CrossRef][Medline]
Cazalets JR. Metachronal propagation of motoneuron burst activation in isolated spinal cord of newborn rat. J Physiol 568: 583–597, 2005.
Clarac F, Vinay L, Cazalets JR, Fady JC, Jamon M. Role of gravity in the development of posture and locomotion in the neonatal rat. Brain Res Rev 28: 35–43, 1998.[CrossRef][Medline]
De Troyer A, Legrand A. Inhomogeneous activation of the parasternal intercostals during breathing. J Appl Physiol 79: 55–62, 1995.
De Troyer A, Gorman RB, Gandevia SC. Distribution of inspiratory drive to the external intercostal muscles in humans. J Physiol 546: 943–954, 2003.
De Troyer A, Kirkwood PA, Wilson TA. Respiratory action of the intercostal muscles. Physiol Rev 85: 717–756, 2005.
Di Pasquale E, Monteau R, Hilaire G. Involvement of the rostral ventro-lateral medulla in respiratory rhythm genesis during the peri-natal period: an in vitro study in newborn and fetal rats. Dev Brain Res 78: 243–252, 1994.[CrossRef][Medline]
Eldridge FL, Millhorn DE, Waldrop TG. Exercise hyperpnea and locomotion: parallel activation from the hypothalamus. Science 211: 844–846, 1981.
Feldman JL, Del Negro CA. Looking for inspiration: new perspectives on respiratory rhythm. Nat Rev Neurosci 7: 232–242, 2006.[CrossRef][Web of Science][Medline]
Fernandes JL, Kobayashi NR, Jasmin BJ, Tetzlaff W. Acetylcholinesterase gene expression in axotomized rat facial motoneurons is differentially regulated by neurotropins: correlation with trkB and trkC mRNA and isoforms. J Neurosci 18: 9936–9947, 1998.
Funk GD, Feldman JL. Generation of respiratory rhythm and pattern in mammals: insights from developmental studies. Curr Opin Neurobiol 5: 778–785, 1995.[CrossRef][Web of Science][Medline]
Funk GD, Milsom WK, Steeves JD. Coordination of wing beat and respiration in the Canada goose. I. Passive wing flapping. J Appl Physiol 73: 1014–1024, 1992.
Gandevia SC, Hudson AL, Gorman RB, Butler JE, De Troyer A. Spatial distribution of inspiratory drive to the parasternal intercostal muscles in humans. J Physiol 573: 263–275, 2006.
Giraudin A, Cabirol-Pol MJ, Simmers J, Morin D. Respiratory activity of intercostal motoneurons and locomotor-respiratory coupling: neuroanatomical and electrophysiological approaches in neonatal rat. 5th Forum of the Federation of European Neuroscience Societies, Vienna, Austria, 2006.
Greer JJ, Martin TP. Distribution of muscle fibre types and EMG activity in cat intercostal muscles. J Appl Physiol 69: 1208–1211, 1990.
Hilaire G, Monteau R, Errchidi S. Possible modulation of the medullary respiratory rhythm generator by the noradrenergic A5 area: an in vitro study in the newborn rat. Brain Res 485: 325–332, 1989.[CrossRef][Web of Science][Medline]
Iizuka M. Intercostal expiratory activity in an in vitro brain stem-spinal cord-rib preparation from the neonatal rat. J Physiol 520: 293–302, 1999.
Iizuka M. Rostrocaudal distribution of spinal respiratory motor activity in an in vitro neonatal rat preparation. Neurosci Res 50: 263–269, 2004.[CrossRef][Web of Science][Medline]
Iscoe S. Control of abdominal muscles. Prog Neurobiol 56: 433–506, 1998.[CrossRef][Web of Science][Medline]
Iscoe S, Polosa C. Synchronization of respiratory frequency by somatic afferent stimulation. J Appl Physiol 40: 138–148, 1976.
Janczewski WA, Feldman JL. Distinct rhythm generators for inspiration and expiration in the juvenile rat. J Physiol 570: 407–420, 2006.
Janczewski WA, Onimaru H, Homma I, Feldman JL. Opioid-resistant respiratory pathway from the preinspiratory neurons to abdominal muscles: in vivo and in vitro study in the newborn rat. J Physiol 545: 1017–1026, 2002.
Juvin L, Morin D. Descending respiratory polysynaptic inputs to cervical and thoracic motoneurons diminish during early postnatal maturation in rat spinal cord. Eur J Neurosci 21: 808–813, 2005.[CrossRef][Web of Science][Medline]
Juvin L, Simmers J, Morin D. Propriospinal circuitry underlying interlimb coordination in mammalian quadrupedal locomotion. J Neurosci 25: 6025–6035, 2005.
Juvin L, Simmers J, Morin D. Locomotor rhythmogenesis in the isolated rat spinal cord: a phase-coupled set of symmetrical flexion-extension oscillators. J Physiol 583: 115–128, 2007.
Mellen NM, Janczewski WA, Bocchiaro CM, Feldman JL. Opioid-induced quantal slowing reveals dual networks for respiratory rhythm generation. Neuron 37: 821–826, 2003.[CrossRef][Web of Science][Medline]
Monteau R, Hilaire G. Spinal respiratory motoneurons. Prog Neurobiol 37: 83–144, 1991.[CrossRef][Web of Science][Medline]
Morin D, Viala D. Coordinations of locomotor and respiratory rhythms in vitro are critically dependent on hindlimb sensory inputs. J Neurosci 22: 4756–4765, 2002.
Onimaru H, Arata A, Homma I. Localization of respiratory rhythm-generating neurons in the medulla of brain stem-spinal cord preparations from newborn rats. Neurosci Lett 78: 151–155, 1987.[CrossRef][Web of Science][Medline]
Onimaru H, Arata A, Homma I. Primary respiratory rhythm generator in the medulla of brain stem-spinal cord preparation from newborn rat. Brain Res 445: 314–324, 1988.[CrossRef][Web of Science][Medline]
Onimaru H, Arata A, Homma I. Neuronal mechanisms of respiratory rhythm generation: an approach using in vitro preparation. Jpn J Physiol 47: 385–403, 1997.[CrossRef][Web of Science][Medline]
Onimaru H, Homma I. A novel functional neuron group for respiratory rhythm generation in the ventral medulla. J Neurosci 23: 1478–1486, 2003.
Onimaru H, Kumagawa Y, Homma I. Respiration-related rhythmic activity in the rostral medulla of newborn rats. J Neurophysiol 96: 55–61, 2006.
Palisses R, Perségol L, Viala D, Viala G. Reflex modulation of phrenic activity through hindlimb passive motion in decorticate and spinal rabbit preparation. Neuroscience 24: 719–728, 1988.[CrossRef][Web of Science][Medline]
Pinsker HM. Aplysia bursting neurons as endogenous oscillators. I. Phase-response curves for pulsed inhibitory synaptic input. J Neurophysiol 40: 527–543, 1977.
Potts JT, Rybak IA, Paton JF. Respiratory rhythm entrainment by somatic afferent stimulation. J Neurosci 25: 1965–1978, 2005.
Potts JT, Spyer KM, Paton JF. Somatosympathetic reflex in a working heart-brainstem preparation of the rat. Brain Res Bull 53: 59–67, 2000.[CrossRef][Web of Science][Medline]
Ramirez JM, Viemari JC. Determinants of inspiratory activity. Resp Physiol Neurobiol 147: 145–157, 2005.[CrossRef]
Romaniuk JR, Kasicki S, Kazennikov OV, Selionov VA. Respiratory responses to stimulation of spinal or medullary locomotor structures in decerebrate cats. Acta Neurobiol Exp 54: 11–17, 1994.[Medline]
Ruangkittisakul A, Secchia L, Bornes TD, Palathinkal DM, Ballanyi K. Dependence on extracellular Ca2+/K+ antagonism of inspiratory center rhythms in slices and en bloc preparations of newborn rat brain stem. J Physiol 584: 489–508, 2007.
Smith JC, Ellenberger HH, Ballanyi K, Richter DW, Feldman JL. Pre-Bötzinger complex: a brain stem region that may generate respiratory rhythm in mammals. Science 254: 726–729, 1991.
Smith JC, Feldman JL. In vitro brain stem-spinal cord preparations for the study of motor systems for mammalian respiration and locomotion. J Neurosci Methods 21: 321–333, 1987.[CrossRef][Web of Science][Medline]
Smith JC, Greer JJ, Liu G, Feldman JL. Neural mechanisms generating respiratory pattern in mammalian brain stem-spinal cord in vitro. I. Spatiotemporal patterns of motor and medullary neuron activity. J Neurophysiol 64: 1149–1169, 1990.
Stirling RV, Liestol K, Summerbell D, Glover JC. The segmental precision of the motor projection to the intercostal muscles in the developing chicken embryo. A differential labelling study using fluorescent tracers. Anat Embryol 191: 397–406, 1995.[CrossRef][Medline]
Suzue T. Respiratory rhythm generation in the in vitro brain stem-spinal cord preparation of the neonatal rat. J Physiol 354: 173–183, 1984.
Tian GF, Duffin J. Connections from upper cervical inspiratory neurons to phrenic and intercostal motoneurons studied with cross-correlation in the decerebrate rat. Exp Brain Res 110: 196–204, 1996.[Web of Science][Medline]
Tresch MC, Kiehn O. Motor coordination without action potentials in the mammalian spinal cord. Nat Neurosci 3: 593–599, 2000.[CrossRef][Web of Science][Medline]
Viala D. Coordination of locomotion and respiration. In: Neural Control of the Respiratory Muscles, edited by Miller AD, Bianchi AL, Bishop BP. New York: CRC, 1997, p. 285–296.
Viala D, Perségol L, Palisses R. Relationship between phrenic and hindlimb extensor activities during fictive locomotion. Neurosci Lett 74: 49–52, 1987.[CrossRef][Web of Science][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
TOP
ABSTRACT
INTRODUCTION
) applied to the ventral surface of the medulla elicited phase-advanced (A1) or -delayed (A2) inspiratory bursts in the ongoing rhythm cycle. 
, expected time of occurrence of spontaneous C4 and T4 bursts in the absence of resetting. B: phase-response histograms showing phase-advance (>0%) or delayed (<0%) rhythmicity as a function of stimulus timing. Numbers of stimuli in each phase (expressed as a percentage of the cycle) are indicated in parentheses. Data were collected from 9 experiments. *P < 0.05; **P < 0.01. C: suppression of respiratory-related bursting in spinal roots after an electrolytic lesion (400 µA for 5 s) at the site of medullary stimulation in A and B.
) and abdominal expiratory activity (
) in the indicated ventral roots during the production of single (left) or double inspiratory bursts per cycle (right).
