- Email: [email protected]
Responses of feline medial medullary reticular formation neurons with projections to the C5–C6 ventral horn to vestibular stimulation
Brain Research 1018 (2004) 247 – 256 www.elsevier.com/locate/brainres
Research report
Responses of feline medial medullary reticular formation neurons with projections to the C5 –C6 ventral horn to vestibular stimulation K.A. Wilkinson, A.P. Maurer, B.F. Sadacca, B.J. Yates* Departments of Otolaryngology and Neuroscience, University of Pittsburgh, Pittsburgh, PA 15213, USA Accepted 1 May 2004 Available online 4 July 2004
Abstract Prior studies have shown that the vestibular system contributes to adjusting respiratory muscle activity during changes in posture, and have suggested that portions of the medial medullary reticular formation (MRF) participate in generating vestibulo-respiratory responses. However, there was previously no direct evidence to demonstrate that cells in the MRF relay vestibular signals monosynaptically to respiratory motoneurons. The present study tested the hypothesis that the firing of MRF neurons whose axons could be antidromically activated from the vicinity of diaphragm motoneurons was modulated by whole-body rotations in vertical planes that stimulated vestibular receptors, as well as by electrical current pulses delivered to the vestibular nerve. In total, 171 MRF neurons that projected to the C5 – C6 ventral horn were studied; they had a conduction velocity of 34 F 15 (standard deviation) m/sec. Most (135/171 or 79%) of these MRF neurons lacked spontaneous firing. Of the subpopulation of units with spontaneous discharges, only 3 of 20 cells responded to vertical rotations up to 10j in amplitude, whereas the activity of 8 of 14 neurons was affected by electrical stimulation of the vestibular nerve. These data support the hypothesis that the MRF participates in generating vestibulo-respiratory responses, but also suggest that some neurons in this region have other functions. D 2004 Elsevier B.V. All rights reserved. Theme: Endocrine and autonomic regulation Topic: Respiratory regulation Keywords: Diaphragm; Phrenic motoneuron; Reticular formation; Vestibular system
1. Introduction Most previous studies considering the regulation of respiratory muscle activity by the brainstem have focused on neurons located in the dorsal and ventral respiratory groups (e.g., Refs. [2,9,12 – 14]). However, recent experiments employing the transneuronal transport of pseudorabies virus have shown that additional neurons located in the medial medullary reticular formation (MRF) also provide direct inputs to motoneurons innervating the diaphragm, abdominal musculature, and the upper airway muscle genioglossus [3 –6,29,38]. The role of MRF neurons in respi-
* Corresponding author. Department of Otolaryngology, University of Pittsburgh, Eye and Ear Institute, Room 106, 203 Lothrop Street, Pittsburgh, PA 15213, USA. Tel.: +1-412-647-9614; fax: +1-412-647-0108. E-mail address: [email protected] (B.J. Yates). URL: http://www.pitt.edu/~byates/yates.html. 0006-8993/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2004.05.080
ratory regulation is unclear, as cells located in this region do not appear to fire in relation to the respiratory cycle [29]. One possible function for neurons in the MRF is to transmit signals from the vestibular system to diaphragm and other respiratory motoneurons. In animal models, electrical or natural stimulation of vestibular afferents [18,22,26, 31,33,35,37], as well as chemical stimulation of vestibular nucleus neurons [34], has been shown to produce alterations in activity of respiratory muscles or nerves innervating these muscles. Electrical stimulation of the vestibular nerve elicits changes in phrenic nerve activity at latencies < 10 ms [35]. Furthermore, bilateral removal of vestibular inputs produces an alteration in the spontaneous activity and posturally related responses of the diaphragm and abdominal musculature [10]. Body rotations that stimulate the vestibular system have also been demonstrated to affect breathing in human subjects [16,21]. However, functional lesions that inactivate the brainstem respiratory group neurons or their axons do not eliminate vestibulo-respiratory responses
248
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
[19,28,37]. Furthermore, vestibular stimulation can produce an increase in activity of respiratory pump muscles without affecting the firing of respiratory group neurons [32,36]. In contrast, many MRF neurons receive vestibular inputs [23,25], and inactivation of this region alters diaphragm and abdominal muscle responses to electrical stimulation of the vestibular nerve [22] or injection of excitatory amino acid analogs into the vestibular nuclei [34]. Nonetheless, there is currently no direct evidence to demonstrate that cells in the MRF relay vestibular signals monosynaptically to respiratory motoneurons. The goal of the present study was to examine the responses of MRF neurons whose axons could be antidromically activated from the vicinity of phrenic motoneurons to whole-body rotations in vertical planes that activated vestibular receptors, as well as to electrical stimulation of the vestibular nerve. For this purpose, a decerebrate cat model was employed, as vestibulo-respiratory responses have been studied extensively using this preparation (e.g., Refs. [22,26,32,33,35,37]). Furthermore, vestibular nucleus and reticular formation neurons are highly active in decerebrate animals [8,11], minimizing the possibility that units in the MRF would fail to respond to vestibular stimulation as a result of low levels of excitability of cells in the brainstem. We tested the hypothesis that the majority of MRF neurons with projections to the ventral horn of the C5 –C6 spinal cord (which contains phrenic motoneurons in the feline [13,14]) respond to vestibular stimulation in a similar fashion as previously documented for the diaphragm [10,26,35], suggesting that these cells contribute to producing vestibular-elicited changes in respiratory activity.
2. Methods and materials All of the procedures used in this study were approved by the University of Pittsburgh’s Institutional Animal Care and Use Committee (IACUC) and conformed to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. 2.1. Surgical procedures Experiments were conducted on 14 adult cats obtained from Liberty Research (North Rose, NY, USA). Animals were initially anesthetized with isoflurane vaporized in oxygen. Catheters were placed in each femoral vein for administration of drugs, and blood pressure was monitored from a femoral artery using a Millar Instruments (Houston, TX, USA) transducer. Rectal temperature was maintained between 36 and 38 jC using an infrared lamp and heating pad. The animal was placed in a modified stereotaxic frame, with the head pitched down 30j to align the horizontal semicircular canals with the earth horizontal plane. As described below, this stereotaxic frame was
mounted on a tilt table capable of simultaneous rotations in the roll and pitch planes. The animal’s body was secured with the use of hip pins and a clamp placed on the dorsal process of the T1 vertebra. A midcollicular decerebration was performed after ligation of the carotid arteries and aspiration of the portion of cerebral cortex overlying the brainstem. Subsequently, a craniotomy was performed to expose the caudal cerebellum, and the caudalmost 5 –6 mm of the cerebellar vermis was aspirated to expose the caudal brainstem. This procedure allowed us to accurately target the MRF for recording. A laminectomy was also performed to expose the C5 – C6 spinal cord. At least 1 h before the beginning of the recording session (and after the decerebration was complete), anesthesia was stopped, and the animal was paralyzed with the use of an intravenous injection of 10 mg/kg gallamine triethiodide (Sigma, St. Louis, MO, USA), which was supplemented by hourly injections of 5 mg/kg. While paralyzed, animals were artificially respired with the use of a positive-pressure ventilator, and end tidal CO2 was maintained near 4%. At the end of the recording session, animals were euthanized by intravenously injecting 120 mg/kg pentobarbital sodium, and the brainstem and C5 – C6 spinal cord were removed and fixed in 10% formalin solution prior to histological analysis (see below). 2.2. Data recording procedures Epoxy-insulated tungsten microelectrodes with an impedance of 5 MV (A&M Systems, Sequim, WA, USA) were used to make recordings from MRF neurons. Systematic tracking was done in 0.5 mm steps in the medial – lateral and anterior – posterior planes from 0 – 2 mm lateral to the midline and from 0– 7 mm rostral to the obex. Typically, multiple electrode tracks were made across one medial – lateral plane, and then the electrode was moved rostral or caudal to that plane, and another series of electrode tracks was performed. While tracking, current pulses were delivered to the C5 – C6 ventral horn, as described below in Section 2.3, for the purpose of stimulating projections of MRF neurons to the vicinity of diaphragm motoneurons. After an antidromically activated unit was isolated, the spontaneous activity of the cell as well as its responses to natural or electrical stimulation of vestibular afferents were determined. Activity recorded from microelectrodes was amplified by a factor of 1000 or 10,000, filtered with a bandpass of 300– 10,000 Hz, and led into a window discriminator for the delineation of spikes from single units. The output of the window discriminator was sampled at 10,000 Hz with the use of a 1401-plus data collection system (Cambridge Electronic Design, Cambridge, UK) and Macintosh G4 computer (Apple Computer, Cupertino, CA, USA). Records of antidromic responses were also obtained by sampling neural activity at 10,000 Hz. Electrolytic lesions were made
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
in the vicinity of two to three recording sites per animal (by passing a 20-AA negative current for 30 s) so that recording locations could be reconstructed. 2.3. Antidromic identification of MRF neurons with projections to the vicinity of diaphragm motoneurons A glass-insulated tungsten microelectrode [30] with an impedance of approximately 0.5 MV was lowered into the C5 – C6 ventral horn using a micromanipulator, for the purpose of stimulating projections of MRF neurons to this region. While lowering the electrode, multiunit activity was recorded and fed to a Grass Instruments (Quincy, MA, USA) AM8 audio monitor; the electrode was moved until discharges at the approximate frequency of respiration (f 20/ min) were detected. The electrode was then connected to a constant-current stimulator for the delivery of square-wave pulses of 0.15 ms duration. When an MRF neuron that responded to stimulation of the C5 – C6 ventral horn was detected, the stimulating electrode was moved dorsally and ventrally to determine the position where the lowest-threshold responses could be elicited. Responses were considered to be antidromic if they exhibited a sharp threshold, fixed latency and collision block with spontaneous spikes; for neurons lacking spontaneous activity, we verified that responses could be consistently elicited when stimuli were delivered at high frequency (50 – 100 Hz). Prior to the end of the experiment, a lesion was produced in the stimulating track by delivering a 20 AA continuous negative current for 60 s. 2.4. Procedures for recording and analyzing responses to rotations in vertical planes In seven animals, we recorded the responses of MRF neurons to stimulation of vertical semicircular canals and otolith organs, which was produced by tilting the entire animal about the pitch (transverse) and roll (longitudinal) axes using a servo-controlled hydraulic tilt table (NeuroKinetics, Pittsburgh, PA, USA). The hydraulics of the tilt table were driven by sinusoidal stimuli delivered by the Cambridge Electronic Design data collection system. We first determined whether a neuron responded to rotations in vertical planes; if a response was present, we ascertained the plane of tilt that produced maximal modulation of the unit’s firing rate (response vector orientation). Response vector orientation was calculated from responses to the ‘‘wobble’’ stimulus, a constant-amplitude tilt whose direction moves around the animal at constant speed [27]. Clockwise wobble stimuli were generated by driving the pitch axis of the tilt table with a sine wave while simultaneously driving the roll axis with a cosine wave. During this stimulus, the animal’s body (viewed from above) appeared to wobble, having in succession nose down, right ear down, nose up, and left ear down. When the signal to the pitch axis of the tilt table was inverted, the stimulus vector rotated in the counterclockwise
249
direction. The direction of the response vector orientation lies midway between the maximal response directions to clockwise and counterclockwise wobble stimulation, because the phase differences between stimulus and response are reversed during the two directions of rotation [27]. Thus, by consideration of both responses, these phase differences can be accounted for. Wobble stimulation was delivered at 0.1, 0.2 and 0.5 Hz, and sometimes at lower frequencies, at amplitudes up to 10j. For neurons that responded to wobble stimulation, the response vector orientation was confirmed by comparing the gain of responses to tilts in a variety of fixed vertical planes, typically delivered at 0.1 Hz and at an amplitude of 7.5– 10j. These tilts always included the roll and pitch planes, as well as planes oriented midway between roll and pitch (i.e., the approximate planes of the vertical semicircular canals). Planar stimuli were generated by applying sine waves to the roll axis, the pitch axis, or simultaneously to both axes of the tilt table, so that during the first half-cycle one side of the body was tilted down, and during the second half-cycle the opposite side was tilted down. Driving both the roll and pitch axes simultaneously produced tilts in a plane oriented between the roll and pitch planes; the orientation was determined by the ratio of the signal sent to the two axes. Planar stimuli were subsequently delivered near the direction of the response vector orientation, at frequencies from 0.05 to 0.5 Hz, to determine the dynamics of responses to vestibular stimulation. Neural activity recorded during rotations in vertical planes was binned (500 bins/cycle) and averaged over the sinusoidal stimulus period. Sine waves were fitted to responses with the use of a least-squares minimization technique [27]. The response sinusoid was characterized by two parameters: phase shift from the stimulus sinusoid (subsequently referred to as phase) and amplitude relative to the stimulus sinusoid (subsequently referred to as gain). Gain and phase measurements were then corrected for the dynamics of the tilt table. Responses were considered significant if the signal-to-noise ratio (see Ref. [27] for method of calculation) was >0.5 and only the first harmonic was prominent (see Fig. 2C for examples of significant responses). 2.5. Procedures for recording and analyzing responses to electrical stimulation of the vestibular nerve The vestibular nerves were prepared for bipolar electrical stimulation in seven animals with the use of a previously described method (e.g., Refs. [17,35,36]). On both sides, the tympanic bulla was exposed using a ventrolateral approach and was opened to expose the promontory. The anterior wall of the promontory was opened to expose the scala vestibuli. One silver – silver chloride ball electrode, insulated except at the tip, was inserted into the scala vestibuli in the direction of the vestibule. The second electrode was placed 1 –2 mm away, in the vicinity of the oval window. The vestibular
250
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
nerve was stimulated using square-wave current pulses that were 0.2 ms in duration. The stimulus intensity required to produce field potentials recordable from the ipsilateral vestibular nuclei was determined; current intensities required to alter the activity of MRF neurons were expressed as a multiple of this threshold. To determine whether an MRF unit received vestibular inputs, a 5-shock train (interpulse interval of 3 ms) at an intensity five times that required to produce vestibular nucleus field potentials was first employed; subsequently, shorter trains were used to determine the minimal number of shocks required to elicit a response. The stimulus repetition rate was one train per 1– 2 s. Response latencies were determined from the first shock of the train as well as from the last shock of the shortest train that evoked a response (this stimulus was deemed the effective shock). In addition, we determined the minimal stimulus intensity required to produce an alteration in neuronal firing when using a 5-shock train. To establish whether the responses to stimulation of the vestibular nerve were due to activation of vestibular afferents (and not to stimulus spread), the thresholds required to produce facial movements were ascertained by delivering a 50-shock train prior to paralysis. The facial nerve, which courses just outside the labyrinth, was the closest non-target nerve to the stimulation site [15]; previous studies have demonstrated that vestibular afferents are selectively stimulated when current intensities are delivered that are subthreshold for activating facial nerve fibers [17,28]. 2.6. Reconstruction of recording and stimulation sites The brainstem and C5 –C6 spinal cord were sectioned at 100-Am thickness in the transverse plane and stained with thionine. Locations of recorded neurons in the medulla were reconstructed on camera lucida drawings of sections with reference to placement of electrolytic lesions, relative positions of electrode tracks, and microelectrode depth. Similarly, the locations where MRF neurons could be antidromically activated from the C5 – C6 gray matter were reconstructed. We assumed that stimulus spread from the lowest-threshold stimulation site in the spinal gray matter was 10 Am per AA of current delivered (e.g., a 20-AA stimulus stimulated a spherical area with a 200-Am radius) [1]. On the basis of this estimate, we classified MRF neurons that could be antidromically activated by current intensities that did not spread beyond the boundaries of the ventral gray matter as projecting to the C5 – C6 ventral horn.
Unless otherwise indicated, pooled data are presented as means F standard deviation.
3. Results Recordings were made from 171 MRF neurons that could be antidromically activated from C5 – C6 using current strengths that did not produce stimulus spread beyond the boundaries of the ventral horn, assuming a current spread of 10 Am/AA delivered [1]. Antidromic thresholds ranged from 8 – 60 AA, and the mean was 33 F 11 AA. Ten neurons were included in the sample even though the lesion marking the stimulation site was not discovered; these cells were antidromically activated using low stimulus intensities (mean of 32 F 10 AA), and recordings conducted while placing the stimulating electrode indicated that it was positioned near the middle of the ventral horn. For all other cells, histological reconstructions confirmed that current spread was confined to the ventral gray matter of the target spinal segments. The large majority of units (135/171 or 79%) were silent, whereas the other 21% (36/171) were spontaneously active. However, even the subpopulation of spontaneously active neurons exhibited low firing rates < 4 spikes/s (average of 0.6 F 0.8 spikes/s). The conduction velocity of MRF neurons with projections to the C5 – C6 ventral horn ranged from 6 – 87 m/s; the mean was 34 F 15 m/s, as indicated in Fig. 1. This figure designates neurons that responded to electrical or natural labyrinthine stimulation, spontaneously active units that failed to respond to stimulation, and other cells (those not tested for vestibular inputs or those that did not respond to stimulation) as separate groups. Neurons lacking spontaneous activity and those untested for vestibular inputs
2.7. Analyses of data Data were tabulated and analyzed with the use of the Prism 4 software package (GraphPad Software, San Diego, CA, USA) running on a Macintosh G5 computer. When two data sets were compared, a Mann – Whitney test was employed, whereas a nonparametric ANOVA (Kruskal – Wallis test) was used to compare three or more data sets.
Fig. 1. Conduction velocity of MRF neurons that were antidromically activated by microstimulation of the C5 – C6 spinal gray matter. Neurons were separated into three groups, which are designated by different shading: cells that responded to vestibular stimulation (black bars), spontaneously active units that failed to respond to vestibular stimulation (white bars), and a combination of silent neurons that failed to respond to stimulation along with neurons that were not tested for the presence of labyrinthine inputs (gray bars).
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
were pooled together, as it would have been impossible to detect inhibition of silent cells, or the units could also have been too hyperpolarized to respond to small excitatory inputs. Thus, we could not be certain whether vestibularelicited alterations in the excitability of silent MRF neurons would have been evident if the cells exhibited baseline spontaneous firing. In seven animals, we considered the effects of wholebody rotations in vertical planes on the activity of 46 MRF neurons that could be antidromically activated from the C5 – C6 ventral horn. Twenty of these cells were spontaneously active, whereas the other 26 were silent. We employed 10j
A. Collision Test, T = 22 µA
251
maximal rotation amplitudes when testing whether the majority of the units (40/46) responded to vertical vestibular stimulation; the other 6 neurons were studied using stimulus amplitudes of 7.5j (5 cells) or 5j (1 cell). The activity of only three of the neurons, all of which were spontaneously active, was modulated by natural vestibular stimulation; an example is shown in Fig. 2. The response vector orientation for this neuron was near contralateral ear down roll, the response gain was near 1 spike/s/j at all stimulus frequencies (0.05 – 0.5 Hz), and the difference in phase between stimulus position and the responses was small (e.g., 16j at 0.1 Hz). The other two neurons whose activity was modu-
C. Responses to Roll Tilt 0.05 Hz, 10º 6 sweeps
1 spike
50 µV 1 ms
0.1 Hz, 10º 10 sweeps
B.
1 spike
Stimulation Site
Ipsi Ear Down
Contra Ear Down
Ipsi Ear Down
1 mm Fig. 2. Responses of an MRF neuron that projected to the C5 ventral horn to natural stimulation of vestibular receptors. (A) Collision test verifying that responses to spinal stimulation were antidromic, and not orthodromic. When the interval between a spontaneously-occurring action potential and the stimulus was reduced from 3 to 1 ms, collision block of the response was noted. Three sweeps were superimposed to produce each trace. The threshold (T) current intensity required to elicit the antidromic response was 22 AA. (B) Area of the spinal gray matter that was stimulated, assuming a current spread of 10 Am/AA in all directions from the electrode tip. (C) Averaged neuronal responses to whole-body tilts in the roll plane at two different frequencies; the number of sweeps averaged for each histogram is indicated. Neuronal activity increased when the animal was tilted in the direction contralateral (Contra) to the side of the brain in which the neuron was located, and decreased when the animal was tilted in the ipsilateral (Ipsi) direction.
252
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
lated by vertical rotations exhibited similar response dynamics. These response characteristics are similar to those previously noted for the diaphragm during rotations in vertical planes [10,26]. Subsequently, in seven other animals we determined whether a more powerful stimulus of vestibular afferents, delivery of current pulses to the vestibular nerve, altered the activity of MRF neurons with projections to the C5 –C6 ventral horn. Responses were recorded from 56 units, 14 of which were spontaneously active and 42 of which were silent, during electrical stimulation of vestibular afferents. Only 3 of the 42 silent neurons responded to electrical vestibular stimulation, although the firing of 8 of the 14 spontaneously active cells was altered by the stimuli. Fig. 3 shows responses to stimulation of the ipsilateral vestibular nerve of an MRF neuron that was activated antidromically from the C6 ventral horn. The effects of current pulses delivered to the ipsilateral and contralateral vestibular nerve on the firing of the 11 MRF neurons that responded to these stimuli are summarized in Table 1. The activity of most units (7/11) was altered by stimulation of the vestibular nerve on one side, but not the other. For the four cells that responded to stimulation of both the ipsilateral and contra-
lateral nerves, the effects from each side were similar. Delivering a single shock altered the firing of five of the neurons, whereas a train of three to four shocks was required to produce a detectable change in activity of the other cells. The minimal response latency from the effective shock noted for any unit was 5.5 ms, and the response latency for most neurons was >15 ms (see Table 1). Whenever electrical stimulation is delivered, there is concern regarding current spread to non-target tissues. We thus evaluated whether MRF neuronal responses could be elicited by current intensities that were subthreshold for activating efferents coursing in the facial nerve, the closest non-target nerve to the electrodes placed in the inner ear [15]. The average threshold for stimulating afferents from the vestibular labyrinth, as indicated by the presence of field potentials recordable from the vestibular nuclei, was 59 F 15 AA. In contrast, the mean threshold for eliciting facial movements using a long train of stimuli was over 10 times higher: 751 F 352 AA. The firing of every MRF unit that was classified as responding to vestibular nerve stimulation could be altered by current intensities that were subthreshold for producing facial contractions. Thus, we are confident that responses to vestibular nerve stimulation
Fig. 3. Responses of an MRF neuron that projected to the C6 ventral horn to electrical stimulation of vestibular afferents. (A) Collision test verifying that responses to spinal stimulation were antidromic; the threshold (T) current intensity required to elicit the responses was 55 AA. Three sweeps were superimposed to produce each trace. (B) Area of the spinal gray matter that was stimulated, assuming a current spread of 10 Am/AA in all directions from the electrode tip. (C) Post-stimulus histograms showing the responses of the neuron to stimulation of the vestibular nerve; approximately 100 sweeps were pooled to generate each trace. The stimulus intensity employed is indicated both in AA and as a multiple of the current strength (T) required to elicit a field potential recordable from the vestibular nuclei. An example of such a field potential, elicited by a 2T stimulus, is provided in the top left panel; this trace represents the average of 60 sweeps. The number of shocks (s) that elicited each response is indicated, and arrows designate the stimulus artifacts resulting from the stimuli.
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256 Table 1 Effects of stimulation of the ipsilateral and contralateral vestibular nerve (VN) on the activity of MRF neurons with projections to the C5 – C6 ventral horn Cell Response to Latency from Response to Latency from number ipsilateral VN effective shock contralateral effective shock stimulation (ms) VN stimulation (ms) 1 2 3 4 5 6 7 8 9 10 11
I E I None E None None I E E None Mean S.D.
7 16 23 – ? – – ? 19 9 – 14.8 6.7
I E I E None E IE None E None E
12 26 ? 35 – ? 12 – 5.5 – 42 22.1 14.5
The onset latency of the response from the last shock of the shortest train that produced an alteration in neuronal activity (effective shock) is also indicated. Abbreviations: E, excited by nerve stimulation; I, inhibited by nerve stimulation; IE, response consisted of inhibition followed by excitation; None, no response to nerve stimulation using a 5 shock train at an intensity five times the threshold for eliciting field potentials recordable from the vestibular nuclei; ?, the response was insufficiently sharp to determine a precise latency.
reflected the activation of labyrinthine afferents, and were not the result of current spread to the facial nerve adjacent to the stimulus site. The locations of all MRF neurons that were antidromically activated from the C5 –C6 ventral horn are indicated in Fig. 4. The neurons were distributed in the medial reticular formation both ipsilateral and contralateral to the side of the spinal cord they projected to, from the level of the obex to approximately 7 mm rostral to the obex. Silent and spontaneously active cells, as well as cells that did and did not respond to natural or electrical stimulation of vestibular afferents, were intermingled in the same regions.
4. Discussion This study provided electrophysiological data that support previous anatomical findings suggesting that MRF neurons make connections with diaphragm motoneurons [4,6,29,38]. The relatively high conduction velocities (mean of 34 m/s) of MRF units that could be antidromically activated from the vicinity of diaphragm motoneurons in the C5 – C6 ventral horn is in accordance with the observation that medial medullary reticular formation neurons that were transneuronally infected by injection of pseudorabies virus into the diaphragm had large cell bodies with diameters of 40– 60 Am [4]. However, the majority of these cells exhibited no spontaneous firing, despite the fact that vestibular nucleus neuronal activity is very high in the decerebrate cat [8,11,27,39]. In addition, the discharges of only 3
253
of 20 spontaneously active MRF units with projections to the C5 –C6 ventral horn were modulated by whole-body tilts in vertical planes at amplitudes up to 10j, despite the fact that such rotations were previously shown to alter the firing of a large number of neurons in the medial medullary reticular formation [7]. Even trains of current pulses delivered to the vestibular nerve (which synchronously activate a large number of vestibular afferents) were ineffective in altering the firing of many MRF units with projections to the vicinity of diaphragm motoneurons, although a majority of MRF neurons respond to such stimuli [23,24]. In combination, these data suggest that a subpopulation of MRF cells participate in generating vestibulo-respiratory responses, but that many MRF neurons that provide inputs to diaphragm motoneurons do not receive labyrinthine inputs. Because these cells also do not have appreciable firing linked to the respiratory cycle [29], they may mediate contractions of respiratory muscles during specialized behaviors unrelated to breathing, such as vomiting [20]. The average onset latency of changes in phrenic nerve activity elicited by electrical stimulation of the vestibular nerve in the cat was previously shown to be 9 ms [35], whereas the responses of most MRF neurons with projections to the C5 –C6 ventral horn occurred at latencies >15 ms. This observation suggests that other neurons in addition to those in the MRF are likely involved in relaying labyrinthine signals to diaphragm motoneurons. This observation is in accordance with the prior finding that inactivation of the feline MRF using muscimol or lidocaine abolished the inhibitory, but not the excitatory, components of vestibulo-respiratory responses [22]. The locations of the additional areas of the nervous system that participate in mediating vestibular influences on diaphragm activity are yet to be established. Transneuronal tracing studies have suggested that in addition to cells in the MRF and the dorsal and ventral respiratory groups, spinal interneurons make direct connections with phrenic motoneurons [4,38]. However, further experiments will be required to establish whether these interneurons play a major role in producing posturally related changes in respiratory muscle activity. Several caveats must be considered when interpreting the present data. First, although we confirmed that the MRF neurons that we studied had projections to the C5 –C6 ventral horn, based on a conservative estimate of stimulus spread [1], there is no direct evidence that these cells made synaptic connections with phrenic motoneurons. However, considering the large number of MRF neurons that were labeled at short latency following injection of the transneuronal tracer pseudorabies virus into the diaphragm [4,6,29,38], it seems probable that many of the antidromically activated cells in this study were premotor respiratory neurons. Furthermore, it is possible that the results from these experiments in felines might not be identical in other species. A previous study in rats showed that changes in diaphragm activity elicited by stimulation of vestibular nucleus neurons were totally abolished following inactivation of the MRF [34], whereas in cats
254
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
Fig. 4. Locations of MRF neurons with projections to the C5 – C6 ventral horn. Neuron locations are plotted on transverse sections of the brainstem. Values to the right of each section indicate the relative distance (in mm) posterior to stereotaxic zero; the level of the obex was approximately P13.5. Neurons on the left side of the diagrams projected to the ipsilateral C5 – C6 ventral horn, whereas cells on the right side projected to the contralateral side of the spinal cord. Different symbols are used to designate neurons tested using natural and electrical vestibular stimulation, and to show silent and spontaneously active neurons that did and did not respond to the stimuli. Units that were not tested for vestibular inputs and silent cells that failed to respond to the stimuli are indicated by the same symbols. Abbreviations: A, nucleus ambiguous; AP, area postrema; C, cuneate nucleus; CN, cochlear nuclei; DMV, dorsal motor nucleus of the vagus; EC, external cuneate nucleus; IO, inferior olivary nucleus; IVN, inferior vestibular nucleus; LRN, lateral reticular nucleus; LVN, lateral vestibular nucleus; MVN, medial vestibular nucleus; P, pyramid; PH, nucleus prepositus hypoglossi; PON, preolivary nucleus; RB, restiform body; RFN, retrofacial nucleus; SA, stria acoustica; SNV, spinal trigeminal nucleus; SON, superior olivary nucleus; ST, solitary tract; STV, spinal trigeminal tract; SVN, superior vestibular nucleus; VII, facial motor nucleus; XII, hypoglossal nucleus.
lesions of this region eliminated only the inhibitory components of vestibulo-respiratory responses [22]. Thus, it is likely that although the MRF participates in control of respiration in a variety of mammals, the precise role played by the area differs between species. For example, the postulated involvement of the MRF in generation of emesis in the cat [20] is not applicable to rodents, which lack the ability to vomit. Further studies will be required to establish the variability in MRF contributions to respiratory regulation across mammalian species. In summary, subpopulations of MRF neurons appear to have different functions in regulating the excitability of diaphragm motoneurons. Some MRF cells receive labyrin-
thine inputs and likely contribute to producing posturally related changes in respiration that are elicited by the vestibular system. However, at least in cats, additional neurons also appear to participate in conveying signals from the vestibular nuclei to phrenic motoneurons. Other MRF neurons with projections to the C5 –C6 ventral horn lack vestibular inputs, as well as appreciable discharges related to the respiratory cycle [29]; such neurons may regulate contraction of the diaphragm during specialized behaviors such as emesis [20]. Prior experiments have shown that MRF neurons with projections to phrenic motoneurons can also be divided into two subpopulations on the basis of neurochemical phenotype: some of the cells are glutamater-
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
gic, whereas others are GABAergic [6]. Considering that MRF lesions abolish the inhibitory, but not the excitatory, components of vestibulo-respiratory responses in felines, it is tempting to hypothesize that the neurons in this region that mediate labyrinthine influences on diaphragm activity are GABAergic, whereas glutamatergic MRF neurons projecting to the C5 – C6 ventral horn have other functions. Nonetheless, this hypothesis is yet to be tested.
Acknowledgements The authors thank Lucy Cotter and Heather Arendt for valuable technical assistance in all phases of the experiment, as well as Christopher Olsheski, Michael Knepp, Adam Anker, and Michael Devinney for their support. Funding was provided by Grant R01 DC03732 from the National Institutes of Health. Electronics support was provided through core grants EY08098 and DC05205 from the National Institutes of Health.
References [1] C. Abzug, M. Maeda, B.W. Peterson, V.J. Wilson, Cervical branching of lumbar vestibulospinal axons. With an appendix by C.P. Bean, J. Physiol. 243 (1974) 499 – 522. [2] A.L. Bianchi, M. Denavitsaubie, J. Champagnat, Central control of breathing in mammals: neuronal circuitry, membrane properties, and neurotransmitters, Physiol. Rev. 75 (1995) 1 – 45. [3] I. Billig, J.M. Foris, J.P. Card, B.J. Yates, Transneuronal tracing of neural pathways controlling an abdominal muscle, rectus abdominis, in the ferret, Brain Res. 820 (1999) 31 – 44. [4] I. Billig, J.M. Foris, L.W. Enquist, J.P. Card, B.J. Yates, Definition of neuronal circuitry controlling the activity of phrenic and abdominal motoneurons in the ferret using recombinant strains of pseudorabies virus, J. Neurosci. 20 (2000) 7446 – 7454. [5] I. Billig, K. Hartge, J.P. Card, B.J. Yates, Transneuronal tracing of neural pathways controlling abdominal musculature in the ferret, Brain Res. 912 (2001) 24 – 32. [6] I. Billig, J.P. Card, B.J. Yates, Neurochemical phenotypes of MRF neurons influencing diaphragm and rectus abdominis activity, J. Appl. Physiol. 94 (2003) 391 – 398. [7] P.S. Bolton, T. Goto, R.H. Schor, V.J. Wilson, Y. Yamagata, B.J. Yates, Response of pontomedullary reticulospinal neurons to vestibular stimuli in vertical planes. Role in vertical vestibulospinal reflexes of the decerebrate cat, J. Neurophysiol. 67 (1992) 639 – 647. [8] J.H. Carey, E.C. Crosby, H.N. Schnitzlein, Decorticate versus decerebrate rigidity in subhuman primates and man, Neurology 21 (1971) 738 – 744. [9] M.I. Cohen, Central determinants of respiratory rhythm, Annu. Rev. Physiol. 43 (1981) 91 – 104. [10] L.A. Cotter, H.E. Arendt, J.G. Jasko, C. Sprando, S.P. Cass, B.J. Yates, Effects of postural changes and vestibular lesions on diaphragm and rectus abdominis activity in awake cats, J. Appl. Physiol. 91 (2001) 137 – 144. [11] R.A. Davis, L. Davis, Decerebrate rigidity in animals, Neurosurgery 9 (1981) 79 – 89. [12] J. Duffin, K. Ezure, J. Lipski, Breathing rhythm generation: focus on the rostral ventrolateral medulla, News Physiol. Sci. 10 (1995) 133 – 140. [13] K. Ezure, Synaptic connections between medullary respiratory neu-
[14]
[15]
[16]
[17]
[18] [19]
[20] [21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31] [32]
[33]
[34]
255
rons and considerations on the genesis of respiratory rhythm, Prog. Neurobiol. 35 (1990) 429 – 450. J.L. Feldman, Neurophysiology of breathing in mammals, in: F.E. Bloom (Ed.), Handbook of Physiology. The Nervous System, IV. Intrinsic Regulatory Systems of the Brain, American Physiological Society, Bethesda, MD, 1986, pp. 463 – 524. R.R. Gacek, The course and central termination of first order neurons supplying vestibular end organs in the cat, Acta Oto-Laryngol., Suppl. 254 (1969) 1 – 66. H. Kaufmann, I. Biaggioni, A. Voustianiouk, A. Diedrich, F. Costa, R. Clarke, M. Gizzi, T. Raphan, B. Cohen, Vestibular control of sympathetic activity. An otolith-sympathetic reflex in humans, Exp. Brain Res. 143 (2002) 463 – 469. I.A. Kerman, B.J. Yates, Regional and functional differences in the distribution of vestibulosympathetic reflexes, Am. J. Physiol., Regul. Integr. Comp. Physiol. 275 (1998) R824 – R835. D. Megirian, Vestibular control of laryngeal and phrenic motoneurons of cat, Arch. Ital. Biol. 106 (1968) 333 – 342. A.D. Miller, T. Yamaguchi, M.S. Siniaia, B.J. Yates, Ventral respiratory group bulbospinal inspiratory neurons participate in vestibular-respiratory reflexes, J. Neurophysiol. 73 (1995) 1303 – 1307. A.D. Miller, S. Nonaka, J. Jakus, B.J. Yates, Modulation of vomiting by the medullary midline, Brain Res. 737 (1996) 51 – 58. K.D. Monahan, M.K. Sharpe, D. Drury, A.C. Ertl, C.A. Ray, Influence of vestibular activation on respiration in humans, Am. J. Physiol., Regul. Integr. Comp. Physiol. 282 (2002) R689 – R694. R.L. Mori, A.E. Bergsman, M.J. Holmes, B.J. Yates, Role of the medial medullary reticular formation in relaying vestibular signals to the diaphragm and abdominal muscles, Brain Res. 902 (2001) 82 – 91. B.W. Peterson, M. Filion, L.P. Felpel, C. Abzug, Responses of medial reticular neurons to stimulation of the vestibular nerve, Exp. Brain Res. 22 (1975) 335 – 350. B.W. Peterson, J.I. Franck, N.G. Pitts, N.G. Daunton, Changes in responses of medial pontomedullary reticular neurons during repetitive cutaneous, vestibular, cortical, and tectal stimulation, J. Neurophysiol. 39 (1976) 564 – 581. O. Pompeiano, D. Manzoni, U.C. Srivastava, G. Stampacchia, Convergence and interaction of neck and macular vestibular inputs on reticulospinal neurons, Neuroscience 12 (1984) 111 – 128. C.D. Rossiter, N.L. Hayden, S.D. Stocker, B.J. Yates, Changes in outflow to respiratory pump muscles produced by natural vestibular stimulation, J. Neurophysiol. 76 (1996) 3274 – 3284. R.H. Schor, A.D. Miller, D.L. Tomko, Responses to head tilt in cat central vestibular neurons: I. Direction of maximum sensitivity, J. Neurophysiol. 51 (1984) 136 – 146. K. Shiba, M.S. Siniaia, A.D. Miller, Role of ventral respiratory group bulbospinal expiratory neurons in vestibular-respiratory reflexes, J. Neurophysiol. 76 (1996) 2271 – 2279. T. Shintani, A.R. Anker, I. Billig, J.P. Card, B.J. Yates, Transneuronal tracing of neural pathways influencing both diaphragm and genioglossal muscle activity in the ferret, J. Appl. Physiol. 95 (2003) 1453 – 1459. S.D. Stoney, W.D. Thompson, H. Asanuma, Excitation of pyramidal tract cells by intracortical microstimulation: effective extent of stimulating current, J. Neurophysiol. 31 (1968) 659 – 669. P.C. Tang, B.E. Gernandt, Autonomic responses to vestibular stimulation, Exp. Neurol. 24 (1969) 558 – 578. S.F. Woodring, B.J. Yates, Responses of ventral respiratory group neurons of the cat to natural vestibular stimulation, Am. J. Physiol., Regul. Integr. Comp. Physiol. 273 (1997) R1946 – R1956. F.D. Xu, Z. Zhang, D.T. Frazier, Transient respiratory augmentation elicited by acute head-down tilt in the anesthetized cat, J. Appl. Physiol. 85 (1998) 490 – 496. F. Xu, J. Zhuang, T.-R. Zhou, T. Gibson, D.T. Frazier, Activation of different vestibular subnuclei evokes differential respiratory and pressor responses in the rat, J. Physiol. 544 (2002) 211 – 223.
256
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
[35] B.J. Yates, J. Jakus, A.D. Miller, Vestibular effects on respiratory outflow in the decerebrate cat, Brain Res. 629 (1993) 209 – 217. [36] B.J. Yates, L. Gre´lot, I.A. Kerman, C.D. Balaban, J. Jakus, A.D. Miller, Organization of vestibular inputs to nucleus tractus solitarius and adjacent structures in cat brain stem, Am. J. Physiol., Regul. Integr. Comp. Physiol. 267 (1994) R974 – R983. [37] B.J. Yates, M.S. Siniaia, A.D. Miller, Descending pathways necessary for vestibular influences on sympathetic and inspiratory out-
flow, Am. J. Physiol., Regul. Integr. Comp. Physiol. 268 (1995) R1381 – R1385. [38] B.J. Yates, J.A. Smail, S.D. Stocker, J.P. Card, Transneuronal tracing of neural pathways controlling activity of diaphragm motoneurons in the ferret, Neuroscience 90 (1999) 1501 – 1513. [39] B.J. Yates, B.J. Jian, L.A. Cotter, S.P. Cass, Responses of vestibular nucleus neurons to tilt following chronic bilateral removal of vestibular inputs, Exp. Brain Res. 130 (2000) 151 – 158.
Research report
Responses of feline medial medullary reticular formation neurons with projections to the C5 –C6 ventral horn to vestibular stimulation K.A. Wilkinson, A.P. Maurer, B.F. Sadacca, B.J. Yates* Departments of Otolaryngology and Neuroscience, University of Pittsburgh, Pittsburgh, PA 15213, USA Accepted 1 May 2004 Available online 4 July 2004
Abstract Prior studies have shown that the vestibular system contributes to adjusting respiratory muscle activity during changes in posture, and have suggested that portions of the medial medullary reticular formation (MRF) participate in generating vestibulo-respiratory responses. However, there was previously no direct evidence to demonstrate that cells in the MRF relay vestibular signals monosynaptically to respiratory motoneurons. The present study tested the hypothesis that the firing of MRF neurons whose axons could be antidromically activated from the vicinity of diaphragm motoneurons was modulated by whole-body rotations in vertical planes that stimulated vestibular receptors, as well as by electrical current pulses delivered to the vestibular nerve. In total, 171 MRF neurons that projected to the C5 – C6 ventral horn were studied; they had a conduction velocity of 34 F 15 (standard deviation) m/sec. Most (135/171 or 79%) of these MRF neurons lacked spontaneous firing. Of the subpopulation of units with spontaneous discharges, only 3 of 20 cells responded to vertical rotations up to 10j in amplitude, whereas the activity of 8 of 14 neurons was affected by electrical stimulation of the vestibular nerve. These data support the hypothesis that the MRF participates in generating vestibulo-respiratory responses, but also suggest that some neurons in this region have other functions. D 2004 Elsevier B.V. All rights reserved. Theme: Endocrine and autonomic regulation Topic: Respiratory regulation Keywords: Diaphragm; Phrenic motoneuron; Reticular formation; Vestibular system
1. Introduction Most previous studies considering the regulation of respiratory muscle activity by the brainstem have focused on neurons located in the dorsal and ventral respiratory groups (e.g., Refs. [2,9,12 – 14]). However, recent experiments employing the transneuronal transport of pseudorabies virus have shown that additional neurons located in the medial medullary reticular formation (MRF) also provide direct inputs to motoneurons innervating the diaphragm, abdominal musculature, and the upper airway muscle genioglossus [3 –6,29,38]. The role of MRF neurons in respi-
* Corresponding author. Department of Otolaryngology, University of Pittsburgh, Eye and Ear Institute, Room 106, 203 Lothrop Street, Pittsburgh, PA 15213, USA. Tel.: +1-412-647-9614; fax: +1-412-647-0108. E-mail address: [email protected] (B.J. Yates). URL: http://www.pitt.edu/~byates/yates.html. 0006-8993/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2004.05.080
ratory regulation is unclear, as cells located in this region do not appear to fire in relation to the respiratory cycle [29]. One possible function for neurons in the MRF is to transmit signals from the vestibular system to diaphragm and other respiratory motoneurons. In animal models, electrical or natural stimulation of vestibular afferents [18,22,26, 31,33,35,37], as well as chemical stimulation of vestibular nucleus neurons [34], has been shown to produce alterations in activity of respiratory muscles or nerves innervating these muscles. Electrical stimulation of the vestibular nerve elicits changes in phrenic nerve activity at latencies < 10 ms [35]. Furthermore, bilateral removal of vestibular inputs produces an alteration in the spontaneous activity and posturally related responses of the diaphragm and abdominal musculature [10]. Body rotations that stimulate the vestibular system have also been demonstrated to affect breathing in human subjects [16,21]. However, functional lesions that inactivate the brainstem respiratory group neurons or their axons do not eliminate vestibulo-respiratory responses
248
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
[19,28,37]. Furthermore, vestibular stimulation can produce an increase in activity of respiratory pump muscles without affecting the firing of respiratory group neurons [32,36]. In contrast, many MRF neurons receive vestibular inputs [23,25], and inactivation of this region alters diaphragm and abdominal muscle responses to electrical stimulation of the vestibular nerve [22] or injection of excitatory amino acid analogs into the vestibular nuclei [34]. Nonetheless, there is currently no direct evidence to demonstrate that cells in the MRF relay vestibular signals monosynaptically to respiratory motoneurons. The goal of the present study was to examine the responses of MRF neurons whose axons could be antidromically activated from the vicinity of phrenic motoneurons to whole-body rotations in vertical planes that activated vestibular receptors, as well as to electrical stimulation of the vestibular nerve. For this purpose, a decerebrate cat model was employed, as vestibulo-respiratory responses have been studied extensively using this preparation (e.g., Refs. [22,26,32,33,35,37]). Furthermore, vestibular nucleus and reticular formation neurons are highly active in decerebrate animals [8,11], minimizing the possibility that units in the MRF would fail to respond to vestibular stimulation as a result of low levels of excitability of cells in the brainstem. We tested the hypothesis that the majority of MRF neurons with projections to the ventral horn of the C5 –C6 spinal cord (which contains phrenic motoneurons in the feline [13,14]) respond to vestibular stimulation in a similar fashion as previously documented for the diaphragm [10,26,35], suggesting that these cells contribute to producing vestibular-elicited changes in respiratory activity.
2. Methods and materials All of the procedures used in this study were approved by the University of Pittsburgh’s Institutional Animal Care and Use Committee (IACUC) and conformed to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. 2.1. Surgical procedures Experiments were conducted on 14 adult cats obtained from Liberty Research (North Rose, NY, USA). Animals were initially anesthetized with isoflurane vaporized in oxygen. Catheters were placed in each femoral vein for administration of drugs, and blood pressure was monitored from a femoral artery using a Millar Instruments (Houston, TX, USA) transducer. Rectal temperature was maintained between 36 and 38 jC using an infrared lamp and heating pad. The animal was placed in a modified stereotaxic frame, with the head pitched down 30j to align the horizontal semicircular canals with the earth horizontal plane. As described below, this stereotaxic frame was
mounted on a tilt table capable of simultaneous rotations in the roll and pitch planes. The animal’s body was secured with the use of hip pins and a clamp placed on the dorsal process of the T1 vertebra. A midcollicular decerebration was performed after ligation of the carotid arteries and aspiration of the portion of cerebral cortex overlying the brainstem. Subsequently, a craniotomy was performed to expose the caudal cerebellum, and the caudalmost 5 –6 mm of the cerebellar vermis was aspirated to expose the caudal brainstem. This procedure allowed us to accurately target the MRF for recording. A laminectomy was also performed to expose the C5 – C6 spinal cord. At least 1 h before the beginning of the recording session (and after the decerebration was complete), anesthesia was stopped, and the animal was paralyzed with the use of an intravenous injection of 10 mg/kg gallamine triethiodide (Sigma, St. Louis, MO, USA), which was supplemented by hourly injections of 5 mg/kg. While paralyzed, animals were artificially respired with the use of a positive-pressure ventilator, and end tidal CO2 was maintained near 4%. At the end of the recording session, animals were euthanized by intravenously injecting 120 mg/kg pentobarbital sodium, and the brainstem and C5 – C6 spinal cord were removed and fixed in 10% formalin solution prior to histological analysis (see below). 2.2. Data recording procedures Epoxy-insulated tungsten microelectrodes with an impedance of 5 MV (A&M Systems, Sequim, WA, USA) were used to make recordings from MRF neurons. Systematic tracking was done in 0.5 mm steps in the medial – lateral and anterior – posterior planes from 0 – 2 mm lateral to the midline and from 0– 7 mm rostral to the obex. Typically, multiple electrode tracks were made across one medial – lateral plane, and then the electrode was moved rostral or caudal to that plane, and another series of electrode tracks was performed. While tracking, current pulses were delivered to the C5 – C6 ventral horn, as described below in Section 2.3, for the purpose of stimulating projections of MRF neurons to the vicinity of diaphragm motoneurons. After an antidromically activated unit was isolated, the spontaneous activity of the cell as well as its responses to natural or electrical stimulation of vestibular afferents were determined. Activity recorded from microelectrodes was amplified by a factor of 1000 or 10,000, filtered with a bandpass of 300– 10,000 Hz, and led into a window discriminator for the delineation of spikes from single units. The output of the window discriminator was sampled at 10,000 Hz with the use of a 1401-plus data collection system (Cambridge Electronic Design, Cambridge, UK) and Macintosh G4 computer (Apple Computer, Cupertino, CA, USA). Records of antidromic responses were also obtained by sampling neural activity at 10,000 Hz. Electrolytic lesions were made
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
in the vicinity of two to three recording sites per animal (by passing a 20-AA negative current for 30 s) so that recording locations could be reconstructed. 2.3. Antidromic identification of MRF neurons with projections to the vicinity of diaphragm motoneurons A glass-insulated tungsten microelectrode [30] with an impedance of approximately 0.5 MV was lowered into the C5 – C6 ventral horn using a micromanipulator, for the purpose of stimulating projections of MRF neurons to this region. While lowering the electrode, multiunit activity was recorded and fed to a Grass Instruments (Quincy, MA, USA) AM8 audio monitor; the electrode was moved until discharges at the approximate frequency of respiration (f 20/ min) were detected. The electrode was then connected to a constant-current stimulator for the delivery of square-wave pulses of 0.15 ms duration. When an MRF neuron that responded to stimulation of the C5 – C6 ventral horn was detected, the stimulating electrode was moved dorsally and ventrally to determine the position where the lowest-threshold responses could be elicited. Responses were considered to be antidromic if they exhibited a sharp threshold, fixed latency and collision block with spontaneous spikes; for neurons lacking spontaneous activity, we verified that responses could be consistently elicited when stimuli were delivered at high frequency (50 – 100 Hz). Prior to the end of the experiment, a lesion was produced in the stimulating track by delivering a 20 AA continuous negative current for 60 s. 2.4. Procedures for recording and analyzing responses to rotations in vertical planes In seven animals, we recorded the responses of MRF neurons to stimulation of vertical semicircular canals and otolith organs, which was produced by tilting the entire animal about the pitch (transverse) and roll (longitudinal) axes using a servo-controlled hydraulic tilt table (NeuroKinetics, Pittsburgh, PA, USA). The hydraulics of the tilt table were driven by sinusoidal stimuli delivered by the Cambridge Electronic Design data collection system. We first determined whether a neuron responded to rotations in vertical planes; if a response was present, we ascertained the plane of tilt that produced maximal modulation of the unit’s firing rate (response vector orientation). Response vector orientation was calculated from responses to the ‘‘wobble’’ stimulus, a constant-amplitude tilt whose direction moves around the animal at constant speed [27]. Clockwise wobble stimuli were generated by driving the pitch axis of the tilt table with a sine wave while simultaneously driving the roll axis with a cosine wave. During this stimulus, the animal’s body (viewed from above) appeared to wobble, having in succession nose down, right ear down, nose up, and left ear down. When the signal to the pitch axis of the tilt table was inverted, the stimulus vector rotated in the counterclockwise
249
direction. The direction of the response vector orientation lies midway between the maximal response directions to clockwise and counterclockwise wobble stimulation, because the phase differences between stimulus and response are reversed during the two directions of rotation [27]. Thus, by consideration of both responses, these phase differences can be accounted for. Wobble stimulation was delivered at 0.1, 0.2 and 0.5 Hz, and sometimes at lower frequencies, at amplitudes up to 10j. For neurons that responded to wobble stimulation, the response vector orientation was confirmed by comparing the gain of responses to tilts in a variety of fixed vertical planes, typically delivered at 0.1 Hz and at an amplitude of 7.5– 10j. These tilts always included the roll and pitch planes, as well as planes oriented midway between roll and pitch (i.e., the approximate planes of the vertical semicircular canals). Planar stimuli were generated by applying sine waves to the roll axis, the pitch axis, or simultaneously to both axes of the tilt table, so that during the first half-cycle one side of the body was tilted down, and during the second half-cycle the opposite side was tilted down. Driving both the roll and pitch axes simultaneously produced tilts in a plane oriented between the roll and pitch planes; the orientation was determined by the ratio of the signal sent to the two axes. Planar stimuli were subsequently delivered near the direction of the response vector orientation, at frequencies from 0.05 to 0.5 Hz, to determine the dynamics of responses to vestibular stimulation. Neural activity recorded during rotations in vertical planes was binned (500 bins/cycle) and averaged over the sinusoidal stimulus period. Sine waves were fitted to responses with the use of a least-squares minimization technique [27]. The response sinusoid was characterized by two parameters: phase shift from the stimulus sinusoid (subsequently referred to as phase) and amplitude relative to the stimulus sinusoid (subsequently referred to as gain). Gain and phase measurements were then corrected for the dynamics of the tilt table. Responses were considered significant if the signal-to-noise ratio (see Ref. [27] for method of calculation) was >0.5 and only the first harmonic was prominent (see Fig. 2C for examples of significant responses). 2.5. Procedures for recording and analyzing responses to electrical stimulation of the vestibular nerve The vestibular nerves were prepared for bipolar electrical stimulation in seven animals with the use of a previously described method (e.g., Refs. [17,35,36]). On both sides, the tympanic bulla was exposed using a ventrolateral approach and was opened to expose the promontory. The anterior wall of the promontory was opened to expose the scala vestibuli. One silver – silver chloride ball electrode, insulated except at the tip, was inserted into the scala vestibuli in the direction of the vestibule. The second electrode was placed 1 –2 mm away, in the vicinity of the oval window. The vestibular
250
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
nerve was stimulated using square-wave current pulses that were 0.2 ms in duration. The stimulus intensity required to produce field potentials recordable from the ipsilateral vestibular nuclei was determined; current intensities required to alter the activity of MRF neurons were expressed as a multiple of this threshold. To determine whether an MRF unit received vestibular inputs, a 5-shock train (interpulse interval of 3 ms) at an intensity five times that required to produce vestibular nucleus field potentials was first employed; subsequently, shorter trains were used to determine the minimal number of shocks required to elicit a response. The stimulus repetition rate was one train per 1– 2 s. Response latencies were determined from the first shock of the train as well as from the last shock of the shortest train that evoked a response (this stimulus was deemed the effective shock). In addition, we determined the minimal stimulus intensity required to produce an alteration in neuronal firing when using a 5-shock train. To establish whether the responses to stimulation of the vestibular nerve were due to activation of vestibular afferents (and not to stimulus spread), the thresholds required to produce facial movements were ascertained by delivering a 50-shock train prior to paralysis. The facial nerve, which courses just outside the labyrinth, was the closest non-target nerve to the stimulation site [15]; previous studies have demonstrated that vestibular afferents are selectively stimulated when current intensities are delivered that are subthreshold for activating facial nerve fibers [17,28]. 2.6. Reconstruction of recording and stimulation sites The brainstem and C5 –C6 spinal cord were sectioned at 100-Am thickness in the transverse plane and stained with thionine. Locations of recorded neurons in the medulla were reconstructed on camera lucida drawings of sections with reference to placement of electrolytic lesions, relative positions of electrode tracks, and microelectrode depth. Similarly, the locations where MRF neurons could be antidromically activated from the C5 – C6 gray matter were reconstructed. We assumed that stimulus spread from the lowest-threshold stimulation site in the spinal gray matter was 10 Am per AA of current delivered (e.g., a 20-AA stimulus stimulated a spherical area with a 200-Am radius) [1]. On the basis of this estimate, we classified MRF neurons that could be antidromically activated by current intensities that did not spread beyond the boundaries of the ventral gray matter as projecting to the C5 – C6 ventral horn.
Unless otherwise indicated, pooled data are presented as means F standard deviation.
3. Results Recordings were made from 171 MRF neurons that could be antidromically activated from C5 – C6 using current strengths that did not produce stimulus spread beyond the boundaries of the ventral horn, assuming a current spread of 10 Am/AA delivered [1]. Antidromic thresholds ranged from 8 – 60 AA, and the mean was 33 F 11 AA. Ten neurons were included in the sample even though the lesion marking the stimulation site was not discovered; these cells were antidromically activated using low stimulus intensities (mean of 32 F 10 AA), and recordings conducted while placing the stimulating electrode indicated that it was positioned near the middle of the ventral horn. For all other cells, histological reconstructions confirmed that current spread was confined to the ventral gray matter of the target spinal segments. The large majority of units (135/171 or 79%) were silent, whereas the other 21% (36/171) were spontaneously active. However, even the subpopulation of spontaneously active neurons exhibited low firing rates < 4 spikes/s (average of 0.6 F 0.8 spikes/s). The conduction velocity of MRF neurons with projections to the C5 – C6 ventral horn ranged from 6 – 87 m/s; the mean was 34 F 15 m/s, as indicated in Fig. 1. This figure designates neurons that responded to electrical or natural labyrinthine stimulation, spontaneously active units that failed to respond to stimulation, and other cells (those not tested for vestibular inputs or those that did not respond to stimulation) as separate groups. Neurons lacking spontaneous activity and those untested for vestibular inputs
2.7. Analyses of data Data were tabulated and analyzed with the use of the Prism 4 software package (GraphPad Software, San Diego, CA, USA) running on a Macintosh G5 computer. When two data sets were compared, a Mann – Whitney test was employed, whereas a nonparametric ANOVA (Kruskal – Wallis test) was used to compare three or more data sets.
Fig. 1. Conduction velocity of MRF neurons that were antidromically activated by microstimulation of the C5 – C6 spinal gray matter. Neurons were separated into three groups, which are designated by different shading: cells that responded to vestibular stimulation (black bars), spontaneously active units that failed to respond to vestibular stimulation (white bars), and a combination of silent neurons that failed to respond to stimulation along with neurons that were not tested for the presence of labyrinthine inputs (gray bars).
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
were pooled together, as it would have been impossible to detect inhibition of silent cells, or the units could also have been too hyperpolarized to respond to small excitatory inputs. Thus, we could not be certain whether vestibularelicited alterations in the excitability of silent MRF neurons would have been evident if the cells exhibited baseline spontaneous firing. In seven animals, we considered the effects of wholebody rotations in vertical planes on the activity of 46 MRF neurons that could be antidromically activated from the C5 – C6 ventral horn. Twenty of these cells were spontaneously active, whereas the other 26 were silent. We employed 10j
A. Collision Test, T = 22 µA
251
maximal rotation amplitudes when testing whether the majority of the units (40/46) responded to vertical vestibular stimulation; the other 6 neurons were studied using stimulus amplitudes of 7.5j (5 cells) or 5j (1 cell). The activity of only three of the neurons, all of which were spontaneously active, was modulated by natural vestibular stimulation; an example is shown in Fig. 2. The response vector orientation for this neuron was near contralateral ear down roll, the response gain was near 1 spike/s/j at all stimulus frequencies (0.05 – 0.5 Hz), and the difference in phase between stimulus position and the responses was small (e.g., 16j at 0.1 Hz). The other two neurons whose activity was modu-
C. Responses to Roll Tilt 0.05 Hz, 10º 6 sweeps
1 spike
50 µV 1 ms
0.1 Hz, 10º 10 sweeps
B.
1 spike
Stimulation Site
Ipsi Ear Down
Contra Ear Down
Ipsi Ear Down
1 mm Fig. 2. Responses of an MRF neuron that projected to the C5 ventral horn to natural stimulation of vestibular receptors. (A) Collision test verifying that responses to spinal stimulation were antidromic, and not orthodromic. When the interval between a spontaneously-occurring action potential and the stimulus was reduced from 3 to 1 ms, collision block of the response was noted. Three sweeps were superimposed to produce each trace. The threshold (T) current intensity required to elicit the antidromic response was 22 AA. (B) Area of the spinal gray matter that was stimulated, assuming a current spread of 10 Am/AA in all directions from the electrode tip. (C) Averaged neuronal responses to whole-body tilts in the roll plane at two different frequencies; the number of sweeps averaged for each histogram is indicated. Neuronal activity increased when the animal was tilted in the direction contralateral (Contra) to the side of the brain in which the neuron was located, and decreased when the animal was tilted in the ipsilateral (Ipsi) direction.
252
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
lated by vertical rotations exhibited similar response dynamics. These response characteristics are similar to those previously noted for the diaphragm during rotations in vertical planes [10,26]. Subsequently, in seven other animals we determined whether a more powerful stimulus of vestibular afferents, delivery of current pulses to the vestibular nerve, altered the activity of MRF neurons with projections to the C5 –C6 ventral horn. Responses were recorded from 56 units, 14 of which were spontaneously active and 42 of which were silent, during electrical stimulation of vestibular afferents. Only 3 of the 42 silent neurons responded to electrical vestibular stimulation, although the firing of 8 of the 14 spontaneously active cells was altered by the stimuli. Fig. 3 shows responses to stimulation of the ipsilateral vestibular nerve of an MRF neuron that was activated antidromically from the C6 ventral horn. The effects of current pulses delivered to the ipsilateral and contralateral vestibular nerve on the firing of the 11 MRF neurons that responded to these stimuli are summarized in Table 1. The activity of most units (7/11) was altered by stimulation of the vestibular nerve on one side, but not the other. For the four cells that responded to stimulation of both the ipsilateral and contra-
lateral nerves, the effects from each side were similar. Delivering a single shock altered the firing of five of the neurons, whereas a train of three to four shocks was required to produce a detectable change in activity of the other cells. The minimal response latency from the effective shock noted for any unit was 5.5 ms, and the response latency for most neurons was >15 ms (see Table 1). Whenever electrical stimulation is delivered, there is concern regarding current spread to non-target tissues. We thus evaluated whether MRF neuronal responses could be elicited by current intensities that were subthreshold for activating efferents coursing in the facial nerve, the closest non-target nerve to the electrodes placed in the inner ear [15]. The average threshold for stimulating afferents from the vestibular labyrinth, as indicated by the presence of field potentials recordable from the vestibular nuclei, was 59 F 15 AA. In contrast, the mean threshold for eliciting facial movements using a long train of stimuli was over 10 times higher: 751 F 352 AA. The firing of every MRF unit that was classified as responding to vestibular nerve stimulation could be altered by current intensities that were subthreshold for producing facial contractions. Thus, we are confident that responses to vestibular nerve stimulation
Fig. 3. Responses of an MRF neuron that projected to the C6 ventral horn to electrical stimulation of vestibular afferents. (A) Collision test verifying that responses to spinal stimulation were antidromic; the threshold (T) current intensity required to elicit the responses was 55 AA. Three sweeps were superimposed to produce each trace. (B) Area of the spinal gray matter that was stimulated, assuming a current spread of 10 Am/AA in all directions from the electrode tip. (C) Post-stimulus histograms showing the responses of the neuron to stimulation of the vestibular nerve; approximately 100 sweeps were pooled to generate each trace. The stimulus intensity employed is indicated both in AA and as a multiple of the current strength (T) required to elicit a field potential recordable from the vestibular nuclei. An example of such a field potential, elicited by a 2T stimulus, is provided in the top left panel; this trace represents the average of 60 sweeps. The number of shocks (s) that elicited each response is indicated, and arrows designate the stimulus artifacts resulting from the stimuli.
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256 Table 1 Effects of stimulation of the ipsilateral and contralateral vestibular nerve (VN) on the activity of MRF neurons with projections to the C5 – C6 ventral horn Cell Response to Latency from Response to Latency from number ipsilateral VN effective shock contralateral effective shock stimulation (ms) VN stimulation (ms) 1 2 3 4 5 6 7 8 9 10 11
I E I None E None None I E E None Mean S.D.
7 16 23 – ? – – ? 19 9 – 14.8 6.7
I E I E None E IE None E None E
12 26 ? 35 – ? 12 – 5.5 – 42 22.1 14.5
The onset latency of the response from the last shock of the shortest train that produced an alteration in neuronal activity (effective shock) is also indicated. Abbreviations: E, excited by nerve stimulation; I, inhibited by nerve stimulation; IE, response consisted of inhibition followed by excitation; None, no response to nerve stimulation using a 5 shock train at an intensity five times the threshold for eliciting field potentials recordable from the vestibular nuclei; ?, the response was insufficiently sharp to determine a precise latency.
reflected the activation of labyrinthine afferents, and were not the result of current spread to the facial nerve adjacent to the stimulus site. The locations of all MRF neurons that were antidromically activated from the C5 –C6 ventral horn are indicated in Fig. 4. The neurons were distributed in the medial reticular formation both ipsilateral and contralateral to the side of the spinal cord they projected to, from the level of the obex to approximately 7 mm rostral to the obex. Silent and spontaneously active cells, as well as cells that did and did not respond to natural or electrical stimulation of vestibular afferents, were intermingled in the same regions.
4. Discussion This study provided electrophysiological data that support previous anatomical findings suggesting that MRF neurons make connections with diaphragm motoneurons [4,6,29,38]. The relatively high conduction velocities (mean of 34 m/s) of MRF units that could be antidromically activated from the vicinity of diaphragm motoneurons in the C5 – C6 ventral horn is in accordance with the observation that medial medullary reticular formation neurons that were transneuronally infected by injection of pseudorabies virus into the diaphragm had large cell bodies with diameters of 40– 60 Am [4]. However, the majority of these cells exhibited no spontaneous firing, despite the fact that vestibular nucleus neuronal activity is very high in the decerebrate cat [8,11,27,39]. In addition, the discharges of only 3
253
of 20 spontaneously active MRF units with projections to the C5 –C6 ventral horn were modulated by whole-body tilts in vertical planes at amplitudes up to 10j, despite the fact that such rotations were previously shown to alter the firing of a large number of neurons in the medial medullary reticular formation [7]. Even trains of current pulses delivered to the vestibular nerve (which synchronously activate a large number of vestibular afferents) were ineffective in altering the firing of many MRF units with projections to the vicinity of diaphragm motoneurons, although a majority of MRF neurons respond to such stimuli [23,24]. In combination, these data suggest that a subpopulation of MRF cells participate in generating vestibulo-respiratory responses, but that many MRF neurons that provide inputs to diaphragm motoneurons do not receive labyrinthine inputs. Because these cells also do not have appreciable firing linked to the respiratory cycle [29], they may mediate contractions of respiratory muscles during specialized behaviors unrelated to breathing, such as vomiting [20]. The average onset latency of changes in phrenic nerve activity elicited by electrical stimulation of the vestibular nerve in the cat was previously shown to be 9 ms [35], whereas the responses of most MRF neurons with projections to the C5 –C6 ventral horn occurred at latencies >15 ms. This observation suggests that other neurons in addition to those in the MRF are likely involved in relaying labyrinthine signals to diaphragm motoneurons. This observation is in accordance with the prior finding that inactivation of the feline MRF using muscimol or lidocaine abolished the inhibitory, but not the excitatory, components of vestibulo-respiratory responses [22]. The locations of the additional areas of the nervous system that participate in mediating vestibular influences on diaphragm activity are yet to be established. Transneuronal tracing studies have suggested that in addition to cells in the MRF and the dorsal and ventral respiratory groups, spinal interneurons make direct connections with phrenic motoneurons [4,38]. However, further experiments will be required to establish whether these interneurons play a major role in producing posturally related changes in respiratory muscle activity. Several caveats must be considered when interpreting the present data. First, although we confirmed that the MRF neurons that we studied had projections to the C5 –C6 ventral horn, based on a conservative estimate of stimulus spread [1], there is no direct evidence that these cells made synaptic connections with phrenic motoneurons. However, considering the large number of MRF neurons that were labeled at short latency following injection of the transneuronal tracer pseudorabies virus into the diaphragm [4,6,29,38], it seems probable that many of the antidromically activated cells in this study were premotor respiratory neurons. Furthermore, it is possible that the results from these experiments in felines might not be identical in other species. A previous study in rats showed that changes in diaphragm activity elicited by stimulation of vestibular nucleus neurons were totally abolished following inactivation of the MRF [34], whereas in cats
254
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
Fig. 4. Locations of MRF neurons with projections to the C5 – C6 ventral horn. Neuron locations are plotted on transverse sections of the brainstem. Values to the right of each section indicate the relative distance (in mm) posterior to stereotaxic zero; the level of the obex was approximately P13.5. Neurons on the left side of the diagrams projected to the ipsilateral C5 – C6 ventral horn, whereas cells on the right side projected to the contralateral side of the spinal cord. Different symbols are used to designate neurons tested using natural and electrical vestibular stimulation, and to show silent and spontaneously active neurons that did and did not respond to the stimuli. Units that were not tested for vestibular inputs and silent cells that failed to respond to the stimuli are indicated by the same symbols. Abbreviations: A, nucleus ambiguous; AP, area postrema; C, cuneate nucleus; CN, cochlear nuclei; DMV, dorsal motor nucleus of the vagus; EC, external cuneate nucleus; IO, inferior olivary nucleus; IVN, inferior vestibular nucleus; LRN, lateral reticular nucleus; LVN, lateral vestibular nucleus; MVN, medial vestibular nucleus; P, pyramid; PH, nucleus prepositus hypoglossi; PON, preolivary nucleus; RB, restiform body; RFN, retrofacial nucleus; SA, stria acoustica; SNV, spinal trigeminal nucleus; SON, superior olivary nucleus; ST, solitary tract; STV, spinal trigeminal tract; SVN, superior vestibular nucleus; VII, facial motor nucleus; XII, hypoglossal nucleus.
lesions of this region eliminated only the inhibitory components of vestibulo-respiratory responses [22]. Thus, it is likely that although the MRF participates in control of respiration in a variety of mammals, the precise role played by the area differs between species. For example, the postulated involvement of the MRF in generation of emesis in the cat [20] is not applicable to rodents, which lack the ability to vomit. Further studies will be required to establish the variability in MRF contributions to respiratory regulation across mammalian species. In summary, subpopulations of MRF neurons appear to have different functions in regulating the excitability of diaphragm motoneurons. Some MRF cells receive labyrin-
thine inputs and likely contribute to producing posturally related changes in respiration that are elicited by the vestibular system. However, at least in cats, additional neurons also appear to participate in conveying signals from the vestibular nuclei to phrenic motoneurons. Other MRF neurons with projections to the C5 –C6 ventral horn lack vestibular inputs, as well as appreciable discharges related to the respiratory cycle [29]; such neurons may regulate contraction of the diaphragm during specialized behaviors such as emesis [20]. Prior experiments have shown that MRF neurons with projections to phrenic motoneurons can also be divided into two subpopulations on the basis of neurochemical phenotype: some of the cells are glutamater-
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
gic, whereas others are GABAergic [6]. Considering that MRF lesions abolish the inhibitory, but not the excitatory, components of vestibulo-respiratory responses in felines, it is tempting to hypothesize that the neurons in this region that mediate labyrinthine influences on diaphragm activity are GABAergic, whereas glutamatergic MRF neurons projecting to the C5 – C6 ventral horn have other functions. Nonetheless, this hypothesis is yet to be tested.
Acknowledgements The authors thank Lucy Cotter and Heather Arendt for valuable technical assistance in all phases of the experiment, as well as Christopher Olsheski, Michael Knepp, Adam Anker, and Michael Devinney for their support. Funding was provided by Grant R01 DC03732 from the National Institutes of Health. Electronics support was provided through core grants EY08098 and DC05205 from the National Institutes of Health.
References [1] C. Abzug, M. Maeda, B.W. Peterson, V.J. Wilson, Cervical branching of lumbar vestibulospinal axons. With an appendix by C.P. Bean, J. Physiol. 243 (1974) 499 – 522. [2] A.L. Bianchi, M. Denavitsaubie, J. Champagnat, Central control of breathing in mammals: neuronal circuitry, membrane properties, and neurotransmitters, Physiol. Rev. 75 (1995) 1 – 45. [3] I. Billig, J.M. Foris, J.P. Card, B.J. Yates, Transneuronal tracing of neural pathways controlling an abdominal muscle, rectus abdominis, in the ferret, Brain Res. 820 (1999) 31 – 44. [4] I. Billig, J.M. Foris, L.W. Enquist, J.P. Card, B.J. Yates, Definition of neuronal circuitry controlling the activity of phrenic and abdominal motoneurons in the ferret using recombinant strains of pseudorabies virus, J. Neurosci. 20 (2000) 7446 – 7454. [5] I. Billig, K. Hartge, J.P. Card, B.J. Yates, Transneuronal tracing of neural pathways controlling abdominal musculature in the ferret, Brain Res. 912 (2001) 24 – 32. [6] I. Billig, J.P. Card, B.J. Yates, Neurochemical phenotypes of MRF neurons influencing diaphragm and rectus abdominis activity, J. Appl. Physiol. 94 (2003) 391 – 398. [7] P.S. Bolton, T. Goto, R.H. Schor, V.J. Wilson, Y. Yamagata, B.J. Yates, Response of pontomedullary reticulospinal neurons to vestibular stimuli in vertical planes. Role in vertical vestibulospinal reflexes of the decerebrate cat, J. Neurophysiol. 67 (1992) 639 – 647. [8] J.H. Carey, E.C. Crosby, H.N. Schnitzlein, Decorticate versus decerebrate rigidity in subhuman primates and man, Neurology 21 (1971) 738 – 744. [9] M.I. Cohen, Central determinants of respiratory rhythm, Annu. Rev. Physiol. 43 (1981) 91 – 104. [10] L.A. Cotter, H.E. Arendt, J.G. Jasko, C. Sprando, S.P. Cass, B.J. Yates, Effects of postural changes and vestibular lesions on diaphragm and rectus abdominis activity in awake cats, J. Appl. Physiol. 91 (2001) 137 – 144. [11] R.A. Davis, L. Davis, Decerebrate rigidity in animals, Neurosurgery 9 (1981) 79 – 89. [12] J. Duffin, K. Ezure, J. Lipski, Breathing rhythm generation: focus on the rostral ventrolateral medulla, News Physiol. Sci. 10 (1995) 133 – 140. [13] K. Ezure, Synaptic connections between medullary respiratory neu-
[14]
[15]
[16]
[17]
[18] [19]
[20] [21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31] [32]
[33]
[34]
255
rons and considerations on the genesis of respiratory rhythm, Prog. Neurobiol. 35 (1990) 429 – 450. J.L. Feldman, Neurophysiology of breathing in mammals, in: F.E. Bloom (Ed.), Handbook of Physiology. The Nervous System, IV. Intrinsic Regulatory Systems of the Brain, American Physiological Society, Bethesda, MD, 1986, pp. 463 – 524. R.R. Gacek, The course and central termination of first order neurons supplying vestibular end organs in the cat, Acta Oto-Laryngol., Suppl. 254 (1969) 1 – 66. H. Kaufmann, I. Biaggioni, A. Voustianiouk, A. Diedrich, F. Costa, R. Clarke, M. Gizzi, T. Raphan, B. Cohen, Vestibular control of sympathetic activity. An otolith-sympathetic reflex in humans, Exp. Brain Res. 143 (2002) 463 – 469. I.A. Kerman, B.J. Yates, Regional and functional differences in the distribution of vestibulosympathetic reflexes, Am. J. Physiol., Regul. Integr. Comp. Physiol. 275 (1998) R824 – R835. D. Megirian, Vestibular control of laryngeal and phrenic motoneurons of cat, Arch. Ital. Biol. 106 (1968) 333 – 342. A.D. Miller, T. Yamaguchi, M.S. Siniaia, B.J. Yates, Ventral respiratory group bulbospinal inspiratory neurons participate in vestibular-respiratory reflexes, J. Neurophysiol. 73 (1995) 1303 – 1307. A.D. Miller, S. Nonaka, J. Jakus, B.J. Yates, Modulation of vomiting by the medullary midline, Brain Res. 737 (1996) 51 – 58. K.D. Monahan, M.K. Sharpe, D. Drury, A.C. Ertl, C.A. Ray, Influence of vestibular activation on respiration in humans, Am. J. Physiol., Regul. Integr. Comp. Physiol. 282 (2002) R689 – R694. R.L. Mori, A.E. Bergsman, M.J. Holmes, B.J. Yates, Role of the medial medullary reticular formation in relaying vestibular signals to the diaphragm and abdominal muscles, Brain Res. 902 (2001) 82 – 91. B.W. Peterson, M. Filion, L.P. Felpel, C. Abzug, Responses of medial reticular neurons to stimulation of the vestibular nerve, Exp. Brain Res. 22 (1975) 335 – 350. B.W. Peterson, J.I. Franck, N.G. Pitts, N.G. Daunton, Changes in responses of medial pontomedullary reticular neurons during repetitive cutaneous, vestibular, cortical, and tectal stimulation, J. Neurophysiol. 39 (1976) 564 – 581. O. Pompeiano, D. Manzoni, U.C. Srivastava, G. Stampacchia, Convergence and interaction of neck and macular vestibular inputs on reticulospinal neurons, Neuroscience 12 (1984) 111 – 128. C.D. Rossiter, N.L. Hayden, S.D. Stocker, B.J. Yates, Changes in outflow to respiratory pump muscles produced by natural vestibular stimulation, J. Neurophysiol. 76 (1996) 3274 – 3284. R.H. Schor, A.D. Miller, D.L. Tomko, Responses to head tilt in cat central vestibular neurons: I. Direction of maximum sensitivity, J. Neurophysiol. 51 (1984) 136 – 146. K. Shiba, M.S. Siniaia, A.D. Miller, Role of ventral respiratory group bulbospinal expiratory neurons in vestibular-respiratory reflexes, J. Neurophysiol. 76 (1996) 2271 – 2279. T. Shintani, A.R. Anker, I. Billig, J.P. Card, B.J. Yates, Transneuronal tracing of neural pathways influencing both diaphragm and genioglossal muscle activity in the ferret, J. Appl. Physiol. 95 (2003) 1453 – 1459. S.D. Stoney, W.D. Thompson, H. Asanuma, Excitation of pyramidal tract cells by intracortical microstimulation: effective extent of stimulating current, J. Neurophysiol. 31 (1968) 659 – 669. P.C. Tang, B.E. Gernandt, Autonomic responses to vestibular stimulation, Exp. Neurol. 24 (1969) 558 – 578. S.F. Woodring, B.J. Yates, Responses of ventral respiratory group neurons of the cat to natural vestibular stimulation, Am. J. Physiol., Regul. Integr. Comp. Physiol. 273 (1997) R1946 – R1956. F.D. Xu, Z. Zhang, D.T. Frazier, Transient respiratory augmentation elicited by acute head-down tilt in the anesthetized cat, J. Appl. Physiol. 85 (1998) 490 – 496. F. Xu, J. Zhuang, T.-R. Zhou, T. Gibson, D.T. Frazier, Activation of different vestibular subnuclei evokes differential respiratory and pressor responses in the rat, J. Physiol. 544 (2002) 211 – 223.
256
K.A. Wilkinson et al. / Brain Research 1018 (2004) 247–256
[35] B.J. Yates, J. Jakus, A.D. Miller, Vestibular effects on respiratory outflow in the decerebrate cat, Brain Res. 629 (1993) 209 – 217. [36] B.J. Yates, L. Gre´lot, I.A. Kerman, C.D. Balaban, J. Jakus, A.D. Miller, Organization of vestibular inputs to nucleus tractus solitarius and adjacent structures in cat brain stem, Am. J. Physiol., Regul. Integr. Comp. Physiol. 267 (1994) R974 – R983. [37] B.J. Yates, M.S. Siniaia, A.D. Miller, Descending pathways necessary for vestibular influences on sympathetic and inspiratory out-
flow, Am. J. Physiol., Regul. Integr. Comp. Physiol. 268 (1995) R1381 – R1385. [38] B.J. Yates, J.A. Smail, S.D. Stocker, J.P. Card, Transneuronal tracing of neural pathways controlling activity of diaphragm motoneurons in the ferret, Neuroscience 90 (1999) 1501 – 1513. [39] B.J. Yates, B.J. Jian, L.A. Cotter, S.P. Cass, Responses of vestibular nucleus neurons to tilt following chronic bilateral removal of vestibular inputs, Exp. Brain Res. 130 (2000) 151 – 158.