- Email: [email protected]
Vestibular influences on hypoglossal nerve activity in the cat
~,~
~,
•
14
,t,, v /
ELSEVIER
Neuroscience Letters 211 (1996) 25-28
HHROIBHa LETTERS
Vestibular influences on hypoglossal nerve activity in the cat C.D. Rossiter, B.J. Y a t e s * Department ~' Otolaryngology and Neuroscience, University of Pittsburgh, Pittsburgh, PA 15213, USA Received 20 March 1996; revised version received 10 May 1996; accepted 10 May 1996
Abstract
The purpose of the present study was to determine if selective activation of vestibular receptors during movement produces changes in hypoglossal nerve activity. Responses were recorded from the hypoglossal nerves during trapezoidal and sinusoidal head rotations in cats with extensive denervations to eliminate non-labyrinthine inputs that could be produced by the movements. Large (50 °) nose-up trapezoidal tilts produced an increase in nerve discharge; ear-down tilt was also effective in one-fourth of the animals• The responses to nose-up tilt were abolished following intracranial transections of the VIIIth cranial nerves. Smaller (20°) sinusoidal head rotations in the roll, pitch and yaw planes were ineffective in producing responses. These data suggest that vestibular inputs elicited by nose-up pitch contribute to tongue protrusion and participate in maintaining airway patency by preventing the tongue from falling to the back of the mouth.
Keywords: Otolith; Labyrinth; Airway; Respiration; Brainstem
Activity in the main tongue protuder, the genioglossal muscle, must be regulated according to body position in order to maintain airway patency [2,9]. In particular, nose-up pitch can result in the tongue falling to the back of the mouth and obstructing the airway; increased activity in the genioglossal muscle must occur to keep the airway open [8,9]. There is evidence to suggest that the vestibular system, which signals body position in space, has influences on the tongue musculature during changes in posture. Activity of the hypoglossal nerve, which innervates both the genioglossal muscle and tongue retractors, is affected by the vestibular system in both rabbits and cats. Electrical stimulation of the vestibular nerve [1,3,6,7] or caloric stimulation of the ear [3,4] produce responses in the hypoglossal nerve or hypoglossal motoneurons. One report also suggested that ear-down tilt in rabbits produced a slight increase in discharges of some hypoglossal motoneurons [5]. Changes in activity produced by electrical stimulation of the vestibular nerve were abolished by transection of the VIIIth nerve [1] or * Corresponding author. University of Pittsburgh, Department of Otolaryngology, Eye and Ear Institute Building, 200 Lotbrop Street, Room 106, Pittsburgh, PA 15213, USA. Tel.: + 1 412 6479614; fax: + 1 412 6472080; e-mail: [email protected]
lesion of the medial and inferior vestibular nuclei [7], demonstrating that these responses were the result of activation of vestibular afferents (and not stimulus spread to non-target tissues). However, appropriate controls were not done to assure that responses to caloric or natural stimulation were due to stimulation of vestibular receptors. The purpose of the present study was to determine if selective activation of vestibular receptors during movement produces changes in hypoglossal nerve activity. Vestibular stimulation was produced by sinusoidal or trapezoidal rotations of the head in animals whose upper cervical dorsal roots were transected to remove neck afferent input, and whose vagus and glossopharyngeal nerves were cut to remove pulmonary, airway and cardiovascular signals that could potentially be elicited by head movement. The possibility that spindle afferents from tongue muscles could contribute to a change in hypoglossal nerve activity during head movement was diminished by cutting the hypoglossal nerves and paralyzing the animals, which assured that the spindles were completely unloaded. Furthermore, the trigeminal nerves were also transected intracranially in some animals to guarantee that cutaneous and proprioceptive inputs from the head did not influence hypoglossal nerve activity during head ro-
0304-3940/96/$12.00 © 1996 Elsevier Science Ireland Ltd. All rights reserved PII S 0 3 0 4 - 3 9 4 0 ( 9 6 ) 1 27 10-3
26
C.D. Rossiter, B.J. Yates / Neuroscience Letters 211 (1996) 2 5 - 2 8
tations. Presumably, the only head-movement related inputs capable of producing hypoglossal nerve responses in this preparation were those from the labyrinth. To support this presumption, we demonstrated that transection of the VllIth cranial nerves abolished hypoglossal nerve responses to head movement. Data were collected from 13 female adult cats. Surgical, stimulation, data recording [12], and data analysis procedures I10] were similar to those in previous studies and will only be described briefly here. Anesthesia was induced and maintained with Fluothane (halothane, Ayerst laboratories) until animals were rendered decerebrate at the midcollicular level. Arterial blood pressure was maintained above 100 mmHg using an intravenous infusion of Aramine (metaraminol bitartrate, Merck, Sharpe & Dohme), if necessary, and rectal temperature was maintained near 38°C using an infrared lamp. Animals were paralyzed using hourly injections qf gallamine triethiodide (5 mg/kg i.v.; initial dose of 10mg/kg) and artificially respired using a positive-pressure ventilator (31-32 cycles/s); the volume of air injected was adjusted to maintain end-tidal CO 2 near 4%. The animal's body was rigidly fixed in place using hip pins and a clamp placed on the TI vertebra. The vagus and glossopharyngeal nerves and the C~-C 3 dorsal roots were transected in all animals, and in four animals the trigeminal nerves were cut just rostral to the decerebration at their intracranial exit from the skull. The cerebellum was aspirated in eight of the animals to provide access to the VIIIth nerve, but in the other five animals the cerebellum remained intact during the recording session. No substantial differences in responses were detected in the two preparations. The animal's head was pitched-down 30 ° and supported by a skull-mounted cylinder that was attached to a head rotator consisting of two servomotors. By altering the attachment between the cylinder and the head rotator, it was possible to produce head rotations in the roll, pitch and yaw planes. The pitch axis passed approximately through the center of the skull-C1 joint, and the axis of rotation for roll and yaw passed approximately through the center of the CI-C2 joint. Both hypoglossal nerves were cut and one or both nerves were inserted into bipolar tunnel electrodes for recording. Hypoglossal nerve activity was amplified by a factor of 10 000, filtered with a bandpass of 10-10 000 Hz, and full-wave rectified (time constant of 0. I ms). The signals were sampled at 500 Hz, averaged, stored and displayed using a Cambridge Electronic Design (CED) 1401-plus data collection system interfaced with a Macintosh Quadra 800 computer. In all 13 animals, increases in hypoglossal nerve activity were produced by 50 ° nose-up trapezoidal movements of the head; examples recorded from two animals are illustrated in Fig. I. The trapezoidal movement (both rising and falling phases) occurred at a velocity of 25°/s; the plateau phase of the stimulus persisted for 11 s. Typically, 25 responses were averaged to determine the effect of
A
2 1
> ::I.
0
v
2
"m
Q_ E
1
c-o¢-3
0
<
|/llHh Nc~raJ~ g'~Hf
cc
2
Time (sec)
Fig. I. Averaged hypoglossal nerve responses to 50 ° nose-up trapezoidal pitch in two animals; the plateau phase of the tilt lasted for 11 s. In both animals, the ClOG 3 dorsal roots and the vagus and glossopharyn-
geal nerves were transected before data collection, in animal B, the trigeminal nerves were also cut intracranially. Responses recorded before and after intraeranial section of the Vlllth cranial nerve are also illustrated for animal B; elimination of vestibular inputs abolished the effecL~of head movementon hypoglossal nerve activity. head movement on hypoglossal nerve activity. Excitability of hypoglossal motoneurons appeared to be elevated as long as the head was held nose-up. In some cases (e.g. the response illustrated in Fig. 1,B1), however, the nerve activity dropped periodically, but remained above baseline during most of the stimulus period. Due to the complexities in the responses, it was difficult to quantify their magnitude. It is noteworthy that pitch-related increases in nerve activity were present in all four animals with trigeminal nerves cut to eliminate proprioceptive and cutaneous inputs from the head, including inputs carried by spindle afferents from tongue muscles (the most likely non-labyrinthine source of the responses). It was impossible to perform nose-down tilts at amplitudes larger than 20 ° , and thus we could not determine if large nose-down head movements altered hypoglossal nerve activity. To demonstrate that the responses to nose-up pitch were the result of activation of labyrinthine receptors, we cut the VIIIth nerve bilaterally in four animals from which responses had been recorded over a period of hours. In all cases, the responses to nose-up tilt disappeared immediately following the nerve transection (see Fig. 1). This finding confirms that the vestibular system was responsible for the increases in hypoglossal nerve activity during nose-up head rotations in these animals. We also tested whether 40 ° trapezoidal ear-down tilt (the largest roll stimulus possible with our stimulation
C.D. Rossiter, B.J. Yates / Neuroscience Letters 211 (1996) 25-28
system) produced changes in hypoglossal nerve activity in 11 of the 13 animals with responses to nose-up pitch. In most animals, only ipsilateral ear-down tilt was employed, although roll in the contralateral direction or in both directions was used in some animals. Roll rotations produced weak effects on nerve activity in 3 of the 11 cats, but no response was elicited in the other eight animals. In one of the three cases where responses were produced, ipsilateral ear-down roll (the only direction tested) elicited an increase in nerve activity; in a second animal, contralateral ear-down roll resulted in a decrease in nerve activity. In the third animal, both ipsilateral and contralateral rotations were used; ipsilateral tilt enhanced hypoglossal nerve discharges and contralateral tilt diminished nerve activity. Sinusoidal rotations (20 ° amplitude) in the roll and pitch planes were also delivered at frequencies between 0.05 and 0.5 Hz (typically 2-3 frequencies including 0.2 Hz) in eight animals. In no case did these stimuli affect hypoglossal nerve activity (the signal-to-noise ratio (see Ref. [10] for method of calculation) for the responses was never systematically > 0.5). We also found that 20 ° sinusoidal yaw rotations at 0.5 Hz in three animals (as well as 1.0 Hz rotations at 20 ° in two of these cats) produced no responses in the hypoglossal nerve. These data show that vestibular receptors activated during large (50 ° ) nose-up rotations of the head in cats produce an increase in hypoglossal nerve activity. Because the increase in nerve activity was maintained as long as the head was in a nose-up position, otolith receptors were presumably mainly responsible for the responses [11]. Furthermore, roll and pitch sinusoidal rotations of smaller amplitude (20 ° ) at frequencies between 0.05 and 0.5 Hz, and yaw rotations of 20 ° amplitude at 0.5 and 1.0 Hz, produced no modulation of hypoglossal nerve activity. These sinusoidal stimuli should have produced considerable angular acceleration and resulted in significant activation of receptors in all 3 semicircular canals [11]; thus, canal afferents do not have strong influences on the firing of hypoglossal motoneurons. Large ear-down tilts had weak influences on hypoglossal nerve activity in a few animals, suggesting that in a small number of cases the best direction of vertical tilt for producing a response was shifted to some degree away from the pitch plane (although even in these animals pitch appeared to be more effective than roll in affecting nerve discharges). These findings are partly at odds with the interpretation of data in previous manuscripts. It was reported earlier that caloric stimulation of the labyrinth, which presumably mainly activates receptors in the horizontal semicircular canal [11], results in an increase in firing in hypoglossal motoneurons [3,4]. In contrast, our 20 ° horizontal rotations did not affect activity in the hypoglossal nerve. The caloric stimuli used in previous studies could have produced retching, a patterned motor response that
27
includes activation of tongue protruder muscles [13]. It is also feasible that increases in hypoglossal nerve discharges elicited by caloric stimulation were due to activation of non-labyrinthine receptors, such as cutaneous receptors in the ear canal. However, another possibility is that our yaw rotations were not large enough in amplitude to generate hypoglossal nerve responses. It was also previously reported that 35 ° static ear-down tilt produced changes in activity of some hypoglossal motoneurons in rabbits [5], whereas our data suggest that roll tilt only elicits hypoglossal nerve responses in a small minority of cats. It is possible, however, that changes in activity of a few motoneurons during roll would not be detected in recordings from whole nerves. The previous investigators did not attempt pitch rotations, so it is not known whether pitch would have produced much stronger responses than roll in their preparation. Nose-up pitch in quadrupeds is analogous to a human assuming a supine position in that both postures result in the tongue falling to the back of the mouth and potentially causing obstructive apnea. Thus, an increase in the activity of tongue protruder muscles during nose-up tilt in cats would assist in maintaining airway patency. It is likely that many of the hypoglossal nerve fibers that were excited during head-up vestibular stimulation innervated fibers in protruder muscles, although effects on tongue retractors (which also receive neural input from the hypoglossal nerve) cannot be ruled out from these data. The current study raises the possibility that damage of the vestibular receptors that are activated by nose-up rotations, or dysfunction in the processing of signals from these receptors, could increase the tendency for obstructive apnea to occur. This possibility remains to be tested. The authors thank Dr. Alan Miller and Dr. Robert Schor for their comments on an earlier version of this manuscript. The technical assistance of Lucy Cotter, Steve Woodring, Gwendolyn Brophy and James Esplen is also appreciated. This work was supported by grant R01 DC02644 from the National Institute on Deafness and Other Communication Disorders of the National Institutes of Health. [1] Elmund, J., Bowman, J.P. and Morgan, R.J., Vestibular influence on tongue activity, Exp. Neurol., 81 (1983) 126-140. [2] Harper,R.M. and Sanerland, E.K., The role of the tongue in sleep apnea. In: C. Guilleminault and W.C. Dement (Eds.), Sleep Apnea Syndromes, Liss, New York, 1978, pp. 219-234. [3] Mameli, O., Melis, F. and Deriu, P.L., Visual and vestibular projections to tongue motoneurons, Brain Res. Bull., 33 (1994) 7-16. [4] Mameli, O. and Tolu, E., Vestibular ampullar modulation of hypoglossal neurons, Physiol. Behav., 37 (1986) 773-775. [5] Mameli, O. and Tolu, E., Hypoglossal responses to macular stimulation in the rabbit, Physiol. Behav., 39 (1987) 273-275. [6] Mameli, O., Tolu, E., Melis, F. and Caria, M.A., Labyrinthine projection to the hypoglossal nucleus, Brain Res. Bull., 20 (1988) 83-88.
28
C.D. Rossiter, B.J. Yates / Neuroscience Letters 211 (1996) 25-28
[7] Miller, A.D. and Siniaia, M.S., Vestibular effects on the upper airway, Soc. Neurosci. Abstr., 21 (1995) 1887. [8] Sauerland, E.K. and Harper, R.M., The human tongue during sleep: electromyographie activity of the genioglossus muscle, Exp. Neurol., 51 (1976) 160-170. [9] Sauerland, E.K. and Mitchell, S.P., Electromyographic activity of the human genioglossus muscle in response to respiration and to positional changes of the head, Bull. Los Angeles Neurol. Soc., 35 (1970) 69-73. [10} Schor, R.H., Miller, A.D. and Tomko, D.L., Responses to head tilt
in cat central vestibular neurons. I. Direction of maximum sensitivity, J. Neurophysiol., 51 (1984) 136-146. [11] Wilson, V.J. and Melvill Jones, G., Mammalian Vestibular Physiology, Plenum Press, New York, 1979. [12] Yates, B.J. and Miller, A.D., Properties of sympathetic reflexes elicited by natural vestibular stimulation: implications for cardiovascular control, J. Neurophysiol., 71 (1994) 2087-2092. [13] Yates, B.J. and Miller, A.D. (Eds.), Vestibular Autonomic Regulation, CRC Press, Boca Raton, FL, 1996.
~,
•
14
,t,, v /
ELSEVIER
Neuroscience Letters 211 (1996) 25-28
HHROIBHa LETTERS
Vestibular influences on hypoglossal nerve activity in the cat C.D. Rossiter, B.J. Y a t e s * Department ~' Otolaryngology and Neuroscience, University of Pittsburgh, Pittsburgh, PA 15213, USA Received 20 March 1996; revised version received 10 May 1996; accepted 10 May 1996
Abstract
The purpose of the present study was to determine if selective activation of vestibular receptors during movement produces changes in hypoglossal nerve activity. Responses were recorded from the hypoglossal nerves during trapezoidal and sinusoidal head rotations in cats with extensive denervations to eliminate non-labyrinthine inputs that could be produced by the movements. Large (50 °) nose-up trapezoidal tilts produced an increase in nerve discharge; ear-down tilt was also effective in one-fourth of the animals• The responses to nose-up tilt were abolished following intracranial transections of the VIIIth cranial nerves. Smaller (20°) sinusoidal head rotations in the roll, pitch and yaw planes were ineffective in producing responses. These data suggest that vestibular inputs elicited by nose-up pitch contribute to tongue protrusion and participate in maintaining airway patency by preventing the tongue from falling to the back of the mouth.
Keywords: Otolith; Labyrinth; Airway; Respiration; Brainstem
Activity in the main tongue protuder, the genioglossal muscle, must be regulated according to body position in order to maintain airway patency [2,9]. In particular, nose-up pitch can result in the tongue falling to the back of the mouth and obstructing the airway; increased activity in the genioglossal muscle must occur to keep the airway open [8,9]. There is evidence to suggest that the vestibular system, which signals body position in space, has influences on the tongue musculature during changes in posture. Activity of the hypoglossal nerve, which innervates both the genioglossal muscle and tongue retractors, is affected by the vestibular system in both rabbits and cats. Electrical stimulation of the vestibular nerve [1,3,6,7] or caloric stimulation of the ear [3,4] produce responses in the hypoglossal nerve or hypoglossal motoneurons. One report also suggested that ear-down tilt in rabbits produced a slight increase in discharges of some hypoglossal motoneurons [5]. Changes in activity produced by electrical stimulation of the vestibular nerve were abolished by transection of the VIIIth nerve [1] or * Corresponding author. University of Pittsburgh, Department of Otolaryngology, Eye and Ear Institute Building, 200 Lotbrop Street, Room 106, Pittsburgh, PA 15213, USA. Tel.: + 1 412 6479614; fax: + 1 412 6472080; e-mail: [email protected]
lesion of the medial and inferior vestibular nuclei [7], demonstrating that these responses were the result of activation of vestibular afferents (and not stimulus spread to non-target tissues). However, appropriate controls were not done to assure that responses to caloric or natural stimulation were due to stimulation of vestibular receptors. The purpose of the present study was to determine if selective activation of vestibular receptors during movement produces changes in hypoglossal nerve activity. Vestibular stimulation was produced by sinusoidal or trapezoidal rotations of the head in animals whose upper cervical dorsal roots were transected to remove neck afferent input, and whose vagus and glossopharyngeal nerves were cut to remove pulmonary, airway and cardiovascular signals that could potentially be elicited by head movement. The possibility that spindle afferents from tongue muscles could contribute to a change in hypoglossal nerve activity during head movement was diminished by cutting the hypoglossal nerves and paralyzing the animals, which assured that the spindles were completely unloaded. Furthermore, the trigeminal nerves were also transected intracranially in some animals to guarantee that cutaneous and proprioceptive inputs from the head did not influence hypoglossal nerve activity during head ro-
0304-3940/96/$12.00 © 1996 Elsevier Science Ireland Ltd. All rights reserved PII S 0 3 0 4 - 3 9 4 0 ( 9 6 ) 1 27 10-3
26
C.D. Rossiter, B.J. Yates / Neuroscience Letters 211 (1996) 2 5 - 2 8
tations. Presumably, the only head-movement related inputs capable of producing hypoglossal nerve responses in this preparation were those from the labyrinth. To support this presumption, we demonstrated that transection of the VllIth cranial nerves abolished hypoglossal nerve responses to head movement. Data were collected from 13 female adult cats. Surgical, stimulation, data recording [12], and data analysis procedures I10] were similar to those in previous studies and will only be described briefly here. Anesthesia was induced and maintained with Fluothane (halothane, Ayerst laboratories) until animals were rendered decerebrate at the midcollicular level. Arterial blood pressure was maintained above 100 mmHg using an intravenous infusion of Aramine (metaraminol bitartrate, Merck, Sharpe & Dohme), if necessary, and rectal temperature was maintained near 38°C using an infrared lamp. Animals were paralyzed using hourly injections qf gallamine triethiodide (5 mg/kg i.v.; initial dose of 10mg/kg) and artificially respired using a positive-pressure ventilator (31-32 cycles/s); the volume of air injected was adjusted to maintain end-tidal CO 2 near 4%. The animal's body was rigidly fixed in place using hip pins and a clamp placed on the TI vertebra. The vagus and glossopharyngeal nerves and the C~-C 3 dorsal roots were transected in all animals, and in four animals the trigeminal nerves were cut just rostral to the decerebration at their intracranial exit from the skull. The cerebellum was aspirated in eight of the animals to provide access to the VIIIth nerve, but in the other five animals the cerebellum remained intact during the recording session. No substantial differences in responses were detected in the two preparations. The animal's head was pitched-down 30 ° and supported by a skull-mounted cylinder that was attached to a head rotator consisting of two servomotors. By altering the attachment between the cylinder and the head rotator, it was possible to produce head rotations in the roll, pitch and yaw planes. The pitch axis passed approximately through the center of the skull-C1 joint, and the axis of rotation for roll and yaw passed approximately through the center of the CI-C2 joint. Both hypoglossal nerves were cut and one or both nerves were inserted into bipolar tunnel electrodes for recording. Hypoglossal nerve activity was amplified by a factor of 10 000, filtered with a bandpass of 10-10 000 Hz, and full-wave rectified (time constant of 0. I ms). The signals were sampled at 500 Hz, averaged, stored and displayed using a Cambridge Electronic Design (CED) 1401-plus data collection system interfaced with a Macintosh Quadra 800 computer. In all 13 animals, increases in hypoglossal nerve activity were produced by 50 ° nose-up trapezoidal movements of the head; examples recorded from two animals are illustrated in Fig. I. The trapezoidal movement (both rising and falling phases) occurred at a velocity of 25°/s; the plateau phase of the stimulus persisted for 11 s. Typically, 25 responses were averaged to determine the effect of
A
2 1
> ::I.
0
v
2
"m
Q_ E
1
c-o¢-3
0
<
|/llHh Nc~raJ~ g'~Hf
cc
2
Time (sec)
Fig. I. Averaged hypoglossal nerve responses to 50 ° nose-up trapezoidal pitch in two animals; the plateau phase of the tilt lasted for 11 s. In both animals, the ClOG 3 dorsal roots and the vagus and glossopharyn-
geal nerves were transected before data collection, in animal B, the trigeminal nerves were also cut intracranially. Responses recorded before and after intraeranial section of the Vlllth cranial nerve are also illustrated for animal B; elimination of vestibular inputs abolished the effecL~of head movementon hypoglossal nerve activity. head movement on hypoglossal nerve activity. Excitability of hypoglossal motoneurons appeared to be elevated as long as the head was held nose-up. In some cases (e.g. the response illustrated in Fig. 1,B1), however, the nerve activity dropped periodically, but remained above baseline during most of the stimulus period. Due to the complexities in the responses, it was difficult to quantify their magnitude. It is noteworthy that pitch-related increases in nerve activity were present in all four animals with trigeminal nerves cut to eliminate proprioceptive and cutaneous inputs from the head, including inputs carried by spindle afferents from tongue muscles (the most likely non-labyrinthine source of the responses). It was impossible to perform nose-down tilts at amplitudes larger than 20 ° , and thus we could not determine if large nose-down head movements altered hypoglossal nerve activity. To demonstrate that the responses to nose-up pitch were the result of activation of labyrinthine receptors, we cut the VIIIth nerve bilaterally in four animals from which responses had been recorded over a period of hours. In all cases, the responses to nose-up tilt disappeared immediately following the nerve transection (see Fig. 1). This finding confirms that the vestibular system was responsible for the increases in hypoglossal nerve activity during nose-up head rotations in these animals. We also tested whether 40 ° trapezoidal ear-down tilt (the largest roll stimulus possible with our stimulation
C.D. Rossiter, B.J. Yates / Neuroscience Letters 211 (1996) 25-28
system) produced changes in hypoglossal nerve activity in 11 of the 13 animals with responses to nose-up pitch. In most animals, only ipsilateral ear-down tilt was employed, although roll in the contralateral direction or in both directions was used in some animals. Roll rotations produced weak effects on nerve activity in 3 of the 11 cats, but no response was elicited in the other eight animals. In one of the three cases where responses were produced, ipsilateral ear-down roll (the only direction tested) elicited an increase in nerve activity; in a second animal, contralateral ear-down roll resulted in a decrease in nerve activity. In the third animal, both ipsilateral and contralateral rotations were used; ipsilateral tilt enhanced hypoglossal nerve discharges and contralateral tilt diminished nerve activity. Sinusoidal rotations (20 ° amplitude) in the roll and pitch planes were also delivered at frequencies between 0.05 and 0.5 Hz (typically 2-3 frequencies including 0.2 Hz) in eight animals. In no case did these stimuli affect hypoglossal nerve activity (the signal-to-noise ratio (see Ref. [10] for method of calculation) for the responses was never systematically > 0.5). We also found that 20 ° sinusoidal yaw rotations at 0.5 Hz in three animals (as well as 1.0 Hz rotations at 20 ° in two of these cats) produced no responses in the hypoglossal nerve. These data show that vestibular receptors activated during large (50 ° ) nose-up rotations of the head in cats produce an increase in hypoglossal nerve activity. Because the increase in nerve activity was maintained as long as the head was in a nose-up position, otolith receptors were presumably mainly responsible for the responses [11]. Furthermore, roll and pitch sinusoidal rotations of smaller amplitude (20 ° ) at frequencies between 0.05 and 0.5 Hz, and yaw rotations of 20 ° amplitude at 0.5 and 1.0 Hz, produced no modulation of hypoglossal nerve activity. These sinusoidal stimuli should have produced considerable angular acceleration and resulted in significant activation of receptors in all 3 semicircular canals [11]; thus, canal afferents do not have strong influences on the firing of hypoglossal motoneurons. Large ear-down tilts had weak influences on hypoglossal nerve activity in a few animals, suggesting that in a small number of cases the best direction of vertical tilt for producing a response was shifted to some degree away from the pitch plane (although even in these animals pitch appeared to be more effective than roll in affecting nerve discharges). These findings are partly at odds with the interpretation of data in previous manuscripts. It was reported earlier that caloric stimulation of the labyrinth, which presumably mainly activates receptors in the horizontal semicircular canal [11], results in an increase in firing in hypoglossal motoneurons [3,4]. In contrast, our 20 ° horizontal rotations did not affect activity in the hypoglossal nerve. The caloric stimuli used in previous studies could have produced retching, a patterned motor response that
27
includes activation of tongue protruder muscles [13]. It is also feasible that increases in hypoglossal nerve discharges elicited by caloric stimulation were due to activation of non-labyrinthine receptors, such as cutaneous receptors in the ear canal. However, another possibility is that our yaw rotations were not large enough in amplitude to generate hypoglossal nerve responses. It was also previously reported that 35 ° static ear-down tilt produced changes in activity of some hypoglossal motoneurons in rabbits [5], whereas our data suggest that roll tilt only elicits hypoglossal nerve responses in a small minority of cats. It is possible, however, that changes in activity of a few motoneurons during roll would not be detected in recordings from whole nerves. The previous investigators did not attempt pitch rotations, so it is not known whether pitch would have produced much stronger responses than roll in their preparation. Nose-up pitch in quadrupeds is analogous to a human assuming a supine position in that both postures result in the tongue falling to the back of the mouth and potentially causing obstructive apnea. Thus, an increase in the activity of tongue protruder muscles during nose-up tilt in cats would assist in maintaining airway patency. It is likely that many of the hypoglossal nerve fibers that were excited during head-up vestibular stimulation innervated fibers in protruder muscles, although effects on tongue retractors (which also receive neural input from the hypoglossal nerve) cannot be ruled out from these data. The current study raises the possibility that damage of the vestibular receptors that are activated by nose-up rotations, or dysfunction in the processing of signals from these receptors, could increase the tendency for obstructive apnea to occur. This possibility remains to be tested. The authors thank Dr. Alan Miller and Dr. Robert Schor for their comments on an earlier version of this manuscript. The technical assistance of Lucy Cotter, Steve Woodring, Gwendolyn Brophy and James Esplen is also appreciated. This work was supported by grant R01 DC02644 from the National Institute on Deafness and Other Communication Disorders of the National Institutes of Health. [1] Elmund, J., Bowman, J.P. and Morgan, R.J., Vestibular influence on tongue activity, Exp. Neurol., 81 (1983) 126-140. [2] Harper,R.M. and Sanerland, E.K., The role of the tongue in sleep apnea. In: C. Guilleminault and W.C. Dement (Eds.), Sleep Apnea Syndromes, Liss, New York, 1978, pp. 219-234. [3] Mameli, O., Melis, F. and Deriu, P.L., Visual and vestibular projections to tongue motoneurons, Brain Res. Bull., 33 (1994) 7-16. [4] Mameli, O. and Tolu, E., Vestibular ampullar modulation of hypoglossal neurons, Physiol. Behav., 37 (1986) 773-775. [5] Mameli, O. and Tolu, E., Hypoglossal responses to macular stimulation in the rabbit, Physiol. Behav., 39 (1987) 273-275. [6] Mameli, O., Tolu, E., Melis, F. and Caria, M.A., Labyrinthine projection to the hypoglossal nucleus, Brain Res. Bull., 20 (1988) 83-88.
28
C.D. Rossiter, B.J. Yates / Neuroscience Letters 211 (1996) 25-28
[7] Miller, A.D. and Siniaia, M.S., Vestibular effects on the upper airway, Soc. Neurosci. Abstr., 21 (1995) 1887. [8] Sauerland, E.K. and Harper, R.M., The human tongue during sleep: electromyographie activity of the genioglossus muscle, Exp. Neurol., 51 (1976) 160-170. [9] Sauerland, E.K. and Mitchell, S.P., Electromyographic activity of the human genioglossus muscle in response to respiration and to positional changes of the head, Bull. Los Angeles Neurol. Soc., 35 (1970) 69-73. [10} Schor, R.H., Miller, A.D. and Tomko, D.L., Responses to head tilt
in cat central vestibular neurons. I. Direction of maximum sensitivity, J. Neurophysiol., 51 (1984) 136-146. [11] Wilson, V.J. and Melvill Jones, G., Mammalian Vestibular Physiology, Plenum Press, New York, 1979. [12] Yates, B.J. and Miller, A.D., Properties of sympathetic reflexes elicited by natural vestibular stimulation: implications for cardiovascular control, J. Neurophysiol., 71 (1994) 2087-2092. [13] Yates, B.J. and Miller, A.D. (Eds.), Vestibular Autonomic Regulation, CRC Press, Boca Raton, FL, 1996.