ArticlePDF Available

Organization of vestibular inputs to nucleus tractus solitarius and adjacent structures in cat brain stem

Authors:

Abstract and Figures

The vestibular system is involved in maintaining stable blood pressure and respiration during changes in posture and is essential for eliciting motion sickness-related vomiting. Because the nucleus tractus solitarius (NTS) participates in the regulation of sympathetic and inspiratory outflow and the triggering of emesis, we tested the hypothesis that this region receives vestibular inputs in cats. In one set of experiments, microinjections of the tracer Phaseolus vulgaris leucoagglutinin into the medial and inferior vestibular nuclei labeled projections to the middle and lateral regions of the NTS. In electrophysiological experiments, electrical stimulation of the vestibular nerve modified the firing rates of neurons located in the same regions. Some neurons with vestibular inputs received convergent signals from the abdominal vagus nerve and could potentially mediate motion sickness-related vomiting. Others received convergent baroreceptor inputs and could act as a substrate for some components of vestibulosympathetic reflexes. In contrast, inspiratory neurons in the dorsal respiratory group received little vestibular input, suggesting that vestibulorespiratory reflexes are mediated by cells located elsewhere.
Content may be subject to copyright.
Organization of vestibular inputs to nucleus tractus
solitarius and adjacent structures in cat brain stem
B. J. YATES, L. GRELOT, I. A. KERMAN, C. D. BALABAN, J. JAKUS AND A. D. MILLER
Laboratory of Neurophysiology, The Rockefeller University, New York, New York 10021;
and Department of Otolaryngology, The Eye and Ear Institute, University of Pittsburgh,
Pittsburgh, Pennsylvania 15213
Yates, B. J., L. Grklot, I. A. Kerman, C. D. Balaban, J.
Jakug, and A. D. Miller. Organization of vestibular inputs to
nucleus tractus solitarius and adjacent structures in cat brain
stem. Am. J. Physiol. 267 (Regulatory Integrative Comp.
Physiol. 36): R974-R983, 1994.-The vestibular system is
involved in maintaining stable blood pressure and respiration
during changes in posture and is essential for eliciting motion
sickness-related vomiting. Because the nucleus tractus soli-
tarius (NTS) participates in the regulation of sympathetic and
inspiratory outflow and the triggering of emesis, we tested the
hypothesis that this region receives vestibular inputs in cats.
In one set of experiments, microinjections of the tracer
PhaseoZus vulgaris leucoagglutinin into the medial and infe-
rior vestibular nuclei labeled projections to the middle and
lateral regions of the NTS. In electrophysiological experi-
ments, electrical stimulation of the vestibular nerve modified
the firing rates of neurons located in the same regions. Some
neurons with vestibular inputs received convergent signals
from the abdominal vagus nerve and could potentially mediate
motion sickness-related vomiting. Others received convergent
baroreceptor inputs and could act as a substrate for some
components of vestibulosympathetic reflexes. In contrast,
inspiratory neurons in the dorsal respiratory group received
little vestibular input, suggesting that vestibulorespiratory
reflexes are mediated by cells located elsewhere.
motion sickness; cardiovascular; respiration
VESTIBULAR RECEPTORS
located in the inner ear detect
linear and angular acceleration imposed on the head and
thus provide signals to the central nervous system that
indicate head position with respect to gravity and the
direction and velocity of head movements (27). In
addition to stabilizing eye and body position during
movements, vestibular signals appear to have important
effects on autonomic control and play a role in correcting
disturbances in homeostasis that occur after changes in
posture. For example, vestibular influences on sympa-
thetic outflow are presumably involved in compensating
for orthostatic hypotension (28). Vestibular actions on
respiratory motoneurons also occur, but the physiologi-
cal roles of these effects are yet to be determined (30). In
addition, the vestibular system is necessary for eliciting
emesis as an endpoint of motion sickness
(18).
The neural pathways responsible for vestibular influ-
ences on respiration, circulation, and emesis are yet to
be elucidated. One area of the brain stem that may be
involved in all of these interactions is the caudal and
intermediate portions of the nucleus tractus solitarius
(NTS), which has been shown to receive projections
from the medial and inferior vestibular nuclei in rabbits
(2). This region receives primary input from cardiovascu-
lar, gastrointestinal, and pulmonary afferents (1,4, 13),
and relays these signals to neuronal groups that regu-
late autonomic functioning (see Refs. 8 and 15 for
review). In addition, the NTS is interconnected with the
area postrema (10, 22), a circumventricular organ that
detects toxins borne in the circulation (6), and is impor-
tant in activating the brain stem pattern generator for
retching and vomiting. The ventrolateral part of the
NTS also contains a group of inspiratory neurons, most
of which project down the spinal cord and make mono-
or oligosynaptic connections with phrenic and external
intercostal motoneurons (8).
The purpose of this study was to determine whether
neurons in caudal and intermediate aspects of the NTS
and adjacent structures in the brain stem of the cat
receive vestibular input. Two approaches were used. In
one group of animals, injections of the anterograde
neuroanatomic tracer Phaseolus vulgaris leucoaggluti-
nin (PHA-L) were made into the medial and inferior
vestibular nuclei to determine if they have connections
with the NTS. These nuclei were targeted because lesion
studies have demonstrated that they are critical for the
production of vestibulosympathetic and vestibulorespi-
ratory reflexes (30). Furthermore, injection of horserad-
ish peroxidase into the region of the NTS produces
retrograde labeling in the medial and inferior vestibular
nuclei (2,11). In a second group of animals, extracellular
recordings were made from neurons in NTS and adja-
cent areas of the dorsal brain stem to determine if they
responded to electrical stimulation of vestibular affer-
ents. Many of the neurons studied were identified as
belonging to one of three functional groups: cells with
inspiratory-related activity, cells with inputs from the
vagus nerve, or cells with blood pressure-related dis-
charges.
Some of these data have been reported in preliminary
form (9).
METHODS
All of the procedures used in this study were approved by
the Rockefeller University Animal Care and Use Committee.
Neuroanatomic Studies
All surgery was performed under aseptic conditions in
a
sterile operating room. Four adult cats of either sex were
sedated using 0.3 mg/kg im acepromazine and 0.15 mg/kg im
butorphanol, an intravenous catheter was inserted, and the
animals were anesthetized using an intravenous injection of
pentobarbital sodium. Initially, a 25 mg/kg dose was delivered;
supplements were administered as necessary to maintain
areflexia during the surgical procedures. A craniotomy was
performed to expose the vermis and intermediate cerebellum
on the right side, and the cerebellum was gently retracted to
R974 0363-6119/94 $3.00 Copyright o 1994 the American Physiological Society
on November 2, 2008 ajpregu.physiology.orgDownloaded from
VESTIBULAR INPUTS TO NTS
R975
reveal the caudal brain stem to -4 mm rostra1 to the obex;
landmarks on the surface of the fourth ventricle were then
used to estimate the borders of the vestibular nuclei. A
micropipette tip (- 35 km diam), which was filled with a 2.5%
solution of PHA-L (Vector Laboratories) dissolved in sodium
phosphate-buffered saline (PBS) at pH 8.0, was positioned just
beneath the surface of the brain stem, into the medial and/or
adjacent inferior vestibular nucleus. Multiple (3-4) unilateral
injections were made in each animal. A positive current
(maximum of
10
PA) was applied for 20 min to eject the tracer.
After a survival time of
15-21
days, the cats were deeply
anesthetized using pentobarbital sodium
(40
mg/kg ip) and
perfused transcardially with PBS followed by the paraformal-
dehyde-lysine-sodium metaperiodate fixative of McLean and
Nakane (16). The brain stems were extracted, postfixed in 4%
paraformaldehyde with 30% sucrose, and sectioned (40~km
frozen sections, transverse plane), and axonally transported
PHA-L was visualized immunohistochemically as reported
previously (2).
Electrophysiological Studies
Oueruiew. Experiments were conducted on 21 adult cats of
either sex. In all of these animals, we looked for units in the
caudal dorsomedial brain stem that responded to electrical
stimulation of the vestibular nerve. In three initial experi-
ments in decerebrate animals, we analyzed only the pattern of
vestibular inputs and did not attempt to characterize other
responses of the cells. In three subsequent decerebrate ani-
mals we identified cells with inspiratory-related discharges
and tested whether they received vestibular signals. In eight
other cats we determined whether neurons with blood pressure-
related discharges received vestibular signals. It was impos-
sible to perform a decerebration during these experiments
because this procedure requires that the carotid arteries be
ligated (which greatly diminishes baroreceptor inputs to the
brain stem). Instead, the animals were anesthetized using
urethan combined with a-chloralose. The pattern of vestibular
inputs to the caudal dorsomedial brain stem appeared similar
in decerebrate and anesthetized animals, and the results from
the two preparations were pooled. In a final group of seven
animals (also anesthetized using urethan/chloralose), we exam-
ined whether neurons with inputs from the vagus nerve in the
lower thorax (abdominal vagus nerve; n = 3), neck (cervical
vagus nerve, aortic depressor and cervical sympathetic nerves
dissected free; n = 2), or either location (n = 2) responded to
electrical stimulation of the vestibular nerve.
Surgical procedures. In the six animals that were decer-
ebrated, anesthesia was induced with Fluothane (halothane;
Ayerst Laboratories) vaporized in nitrous oxide and oxygen;
the anesthetic dose was adjusted to maintain areflexia to
noxious stimuli. The carotid arteries were ligated, the poste-
rior cerebral cortex was aspirated to expose the rostra1 brain
stem, and an intercollicular decerebration was performed.
Anesthesia was discontinued only after all surgical procedures
were complete, but at least 1 h before the beginning of the
recording session. Paralysis was produced (after the decerebra-
tion) by an initial intravenous injection of
10
mg/kg gallamine
triethiodide (Sigma) and maintained by hourly injections of 5
mg/kg. While paralyzed, animals were artificially respired
using a positive-pressure ventilator (24 cycles/min), and end-
tidal CO2 was maintained at 4-5%. An expiratory load of -
1
cmHzO was also usually applied.
The other
15
animals were anesthetized with an intraperito-
neal injection of 800 mg/kg urethan (Sigma) combined with 40
mg/kg ol-chloralose (Sigma). These animals were not para-
lyzed. We observed the animals carefully during the experi-
ment for two signs that the level of anesthesia was light: the
presence of spontaneous movements or increases in arterial
blood pressure. In the rare instances in which either indication
appeared, an additional intraperitoneal injection of 200 mg/kg
urethan and
10
mg/kg a-chloralose was given.
All animals were placed in a stereotaxic frame and sus-
pended from hip pins. A craniotomy was performed to expose
the caudal cerebellum, and the 2-3 mm of cerebellum that was
most caudal was aspirated to expose the brain stem near the
obex.
Isolation and stimulation of the vestibular nerve. The
vestibular nerve was prepared for stimulation, as in many
previous studies (e.g., Refs. 29-31). Typically, the nerve was
isolated on both sides; however, in the eight experiments in
which we examined units with baroreceptor inputs, only one
side (usually the side contralateral to recording) was im-
planted with stimulating electrodes. This restriction was
necessary because pilot experiments showed that NTS neu-
rons only received weak cardiovascular inputs when both
vestibular nerves were isolated, presumably because the glos-
sopharyngeal nerve (which carries the majority of these sig-
nals and runs adjacent to the tympanic bulla) on both sides
was damaged by the surgery.
To isolate the vestibular nerve, we exposed the ventrolateral
surface of the tympanic bulla, and the bulla was opened to
expose the promontory. The anterior wall of the promontory
was opened to expose the Scala vestibuli and the branches of
the vestibular nerve. A pair of silver-silver chloride ball
electrodes (separation of
l-2
mm) was placed in close proxim-
ity to the nerve, covered with warm semisolid paraffin, and
fixed in place with dental cement.
Square-wave current pulses that were 0.2 ms in duration
were used to test for vestibular inputs to neurons in NTS and
adjacent areas. Both single shocks and trains of two to five
pulses (interpulse interval of 3 ms) were employed. The
stimulus repetition rate was one train per 2-3 s. Thresholds of
vestibular stimulation were expressed as multiples of the
threshold (T) of field potentials recorded from the medial
longitudinal fasciculus (MLF) l-2 mm rostra1 to the obex. The
MLF contains descending vestibulospinal fibers, and thresh-
olds of vestibular-elicited fields recorded there reflect the
threshold for activation of neurons in the vestibular nuclei
(27).
A five-shock t rain at an intensity five times that neces-
sary to elicit MLF field potentials was first used to determine if
a neuron responded to vestibular nerve stimulation. For many
cells, shorter trains at 5T were used to determine the least
number of effective shocks, and five-shock trains at intensities
< 5T were used to determine the minimal effective stimulus
intensity (although all response latencies were determined
using 5T stimuli).
It is unlikely that our stimulus spread beyond the immedi-
ate vicinity of the vestibular nerve, because the stimuli used
were low intensity (typically < 200 PA, and > 300 I-LA for only
1 nerve stimulated in a single animal), and they were delivered
between two closely placed electrodes. Nevertheless, if stimu-
lus spread had occurred, the most likely target would have
been the facial nerve, which courses just outside the labyrinth
(27). In all e xperiments, the minimal stimulus intensity
required with a 50-shock train (pulse width of 0.2 ms, inter-
pulse interval of 3 ms) to produce facial movements was
measured (before paralysis in decerebrate animals) for compari-
son with the stimulus strengths used to elicit changes in unit
activity. We tested both stimulus polarities and used as the
cathode the electrode that produced the lowest threshold eye
movements (these occurred as part of the vestibuloocular
reflex) and the highest threshold facial movements. Supra-
threshold responses could be elicited (with a 5-shock train) in
all but 10% of the units by intensities that, with a 50-shock
on November 2, 2008 ajpregu.physiology.orgDownloaded from
R976
Fig. 1. Charting of’ Pk~s~~us
wdgurh
leucoagglutinin
(PHA-1,) injection sites in cat vestibular nuclei. The
extent of 4 in.jections is charted on a series of transverse
sections through caudal aspect of’vestibuiar nuclei of the
cat; values above each diagram indicate distance rostra1
to obex. Three in.jections kns~s NYO2, NYO3, and
NY041 were confined to vestibular nuclei. The 4th
injection !NYO,5) also involved the reticular formation
subjacent to the inferior vcstibular nucleus. All injec-
tions were made into right side of brain stem. For cIarity,
injection sites for 2 animals are plotted on opposite side
of brain stem. CN, cochlear nucleus; DMV, dorsal motor
nucleus of’the vagus nerve; EC, external cuneate nucleus;
IV, inferior vestibular nucleus; MV, medial vestibular
nucleus; N’I’S, nucleus tractus solitarius; PH, nucleus
prepositus hypoglossi; KB, restiform body; XII, hypoglos-
sal nucleus.
train, did not elicit facial movements. Even if’there was limited
current spread to the facial nerve, it seems unlikely that the
current density would be great enough to activate the higher-
threshold cutaneous and muscle afferents that produce auto-
nomic effects (21) and provide inputs to the NTS (20).
Collectively, this evidence suggests that our vestibuiar stimu-
lation predominantly, if not exclusively, activated vestibular
afferents.
Isolation and stim &&or? of’ the vagus rzervc. The entire
abdominal vagus nerve (both the dorsal and ventral branches)
was transected in the lower thorax, just rostra1 to the dia-
phragm, and placed in a bipolar tunnel electrode for stimula-
tion. The intact cervical vagus nerve was stimulated using
bipolar hook electrodes. Hook electrodes were placed rostra1 to
the stimulation site in all animals to record the evoked afferent
nerve volley.
The vagus nerves were stimulated using 0.5ms square-
wave current pulses at two to five times the intensity neces-
sary to produce a volley recordable proximally on the nerve.
Thresholds ranged from 250 PA to 3 mA. Single shocks as well
as trains of up to five pulses (interpulse interval of 3 ms) were
employed to test for vagal inputs to NTS neurons.
Activity recorded from the vagus nerve was amplified (by a
factor of lOO,OOO), full-wave rectified, and low-pass filtered
with a l-ms time constant. The signals were led into a Digital
Equipment Corporation PDP 1 1 / 73 minicomputer for averag-
ing and storage; the typical bin width was 0.5-l ms.
Recording of’ Aspiratory activity. Neural activation of the
diaphragm was monitoreh by recording from a C5 phrenic
nerve, which was dissected in the neck, cut, and placed in a
bipolar tunnel electrode. Phrenic nerve activity was amplified
by a factor of lO,OOO-100,000, full-wave rectified, low-pass
filtered with a lo-ms time constant, and sampled at 100 Hz as
an analog signal by the PDP 11/73 computer.
NY04 1 mm ‘11111111lIIlII NY0 2
NY05 z NY03
Recording
of’
cardiac-related activitv. Blood pressure was
II
monitored using a pressure-sensitive transducer (Millar Mikro-
Tip) inserted into the femoral artery. A voltage proportional to
the level of blood pressure was generated and fed into the PDP
ll/ 73 minicomputer for storage. The electrocardiogram was
recorded using two low-impedance needle electrodes, one
inserted into the skin overlying the heart and the other
positioned into the skin on the other side of the chest. The
voltage recorded across the two electrodes was amplified by a
factor of 1,000, full-wave rectified, and low-pass filtered with a
l-ms time constant. The R wave of the electrocardiogram was
used to trigger the averaging of blood pressure and unit
activity in the eight experiments in which we looked for
neurons with baroreceptor inputs.
Recording
of’
unit activity. Electrode penetrations were
made O-2 mm lateral to the&midline and within 2 mm of the
anterior-posterior level of the obex using either glass micropi-
pettes containing 2 M NaCl saturated with Fast green dye
(impedance of 1-3 Mi 1) or epoxy-insulated tungsten microelec-
trodes (impedance of lo-12 MI), purchased from Frederick
Haer or A-M Systems). Tracking was done no deeper than 2
mm from the dorsal surf’ace of the brain stem. Action poten-
tials were led into a window discriminator whose output was
led into the 1 l/73 minicomputer for on-line display and
analysis. Poststimulus histograms were generated from unit
responses; the typical bin width was 0.145 ms. Many record-
ing locations were marked by dye injection (when micropipette
electrodes were used) or electrolytic lesions (when tungsten
microelectrodes were employed J.
However, not all neuronal cell types in the caudal dorsome-
dial brain stem were tested for vestibular inputs. We did not
test dorsal vagal motoneurons, which were identified by
having a fixed antidromic latency after vagal nerve stimula-
on November 2, 2008 ajpregu.physiology.orgDownloaded from
VESTII3ULAR INPUTS TO NW
R977
tion, nor cells in area postrema, which could not be held long
enough to study their responses.
Histological procedures. At the end of the experiment,
animals were killed by an overdose of pentobarbital sodium
(120 “g/kg iv) or by injecting 4-5 ml iv of saturated KC1
solution. The brain stem was removed and fixed in Formalin,
and NO-pm transverse sections were stained with thionin.
Locations of recorded neurons were reconstructed on standard
sections with reference to dye marks or electrolytic lesions and
microelectrode depth.
RESULTS
Neuroanatomic Studies
Axonal transport of PHA-L revealed bilateral projec-
tions to the caudal and intermediate aspects of the NTS
from injection sites involving both the inferior vestibu-
lar nucleus and the caudal half of the medial vestibular
nucleus. Axons did not appear to project to the rostra1
NTS. The injection sites from four cats are charted on
drawings of a series of transverse sections in Fig. 1.
Fig. f2. Charting of distribution of vestibular nucleus
projections to caudal aspect of the dorsal medulla in
cat NY02. Injection site is illustrated in Fig. 1.
Locations of PHA-L-positive axons are charted on
camera lucida drawings of a series of transverse
sections showing the ipsilateral (ri&r) and contralat-
era1 (Zef’r ) hypoglossal nucleus (XII 1, DMV. and soli-
tary nucleus. Lateral and medial pathways from
vestibular nuclei to caudal dorsomedial brain stem
are also indicated. AP, area postrema; Int, nucleus
intercalatus; ST, solitary tract; 2, nucleus Z. Subnu-
cIei of the nucleus tractus solitarius (NTS) are as
follows: eommissural (Scorn), intermediate Gint),
lateral (Sk), medial (Sm), parvoceIlular Gpc), ventro-
lateral (Svl J.
Three of these injections included the caudal half of the
vestibular nuclei, without evidence of spread to the
underlying rostra1 NTS, reticular formation, or MLF;
the fourth (NY@% encroached on the reticular forma-
tion lateral and ventral to the rostra1 a spect 0 f the NTS.
Anterograd ely 1 abeled axons could be traced bilaterally
from these injection sites to the caudal dorsomedial
brain stem via the courses identified previously in
rabbits (2 ). The ipsilateral projections either descended
in the m edial vesiibular nucleus and n ucleus Z (lateral
pathway > or proceeded ca udally t h rough nucleus preposi-
tus
hypoglossi and nucleus intercalatus (medi al path-
way
>. The contralateral projections crossed the midline
in the reticular formation and entered the contralateral
vestibular and prepositus nuclei. The axons then fol-
lowed the same courses as the ipsilateral projections to
the solitary nucleus, dorsal motor vagal nucleus, and
nucleus intercalatus.
The regional distribution of vestibular projections to
the caudal dorsomedial brain stem in one animal is
on November 2, 2008 ajpregu.physiology.orgDownloaded from
R978
VESTIBULAR INPUTS TO NTS
charted in a series of camera lucida drawings in Fig. 2.
Rostra1 to the obex (Fig. 2, A and B) there were
projections to nucleus intercalatus, the dorsal motor
nucleus of the vagus, and the medial and lateral regions
of the NTS, except to the ventrolateral region where
labeling was sparse. Caudal to the obex (Fig. 2C) axons
were distributed to the commissural and intermediate
subnuclei. Similar distributions of projections were ob-
served in the three other animals.
Electrophysiological Studies
The neuronal sample in this study falls into two
groups. One group of 44 neurons located in and around
the NTS was determined to have vestibular input, but
other responses of these cells were not identified. The
other group of 87 NTS units was putatively classified by
autonomic responses and was tested for the presence of
vestibular input. Sixteen of the neurons in this group
had abdominal vagal inputs,
15
had cervical vagal
inputs, 28 had blood pressure-related discharges, and 28
had inspiratory-related activity. The proportion of these
neurons receiving vestibular inputs, and the laterality of
the inputs, is indicated in Table
1.
Vestibular inputs to neurons in the caudal dorsome-
dial brain stem. Fifty-nine neurons in NTS and adjacent
structures were identified as having vestibular input.
Stimulation of the vestibular nerve produced a variety of
effects in neurons in the caudal dorsomedial brain stem,
as indicated in Table 2. Only very short trains of one to
three stimuli (median of 2 shocks) were typically neces-
sary to elicit responses in the target neurons. The
maximal stimulus intensities applied to the vestibular
nerve were five times the MLF field potential threshold,
which ranged from
15
to 70 FA (median of 35 J.LA). To
help rule out the possibility that responses were due to
stimulus spread to the facial nerve, we also employed a
five-shock train at lower intensities for many cells. The
minimal stimulus strength tested that elicited responses
was
165
t 67 (SD) PA (median of 150 PA). In contrast,
the minimal intensity required with a 50-shock train to
produce facial movements, reflecting stimulus spread to
the facial nerve, ranged from 90 to 1,000 PA (median of
290 cLA>. All but 6 of the 59 neurons that responded to
vestibular nerve stimulation were shown to be activated
by stimulus strengths that did not result in current
spread to the facial nerve; the thresholds for the six
Table 1. Numbers of cells of different types in caudal
dorsomedial brain stem tested for vestibular inputs
No. of Cells
Vestibular Vestibular
response response Total
present absent
Abdominal vagal inputs 4 12 16
Ipsilateral side 3
Contralateral side 0
Both sides 1
Cervical vagal inputs 4 11 15
Ipsilateral side 3
Contralateral side 1
Both sides 0
Baroreceptor inputs 5 23 28
Ipsilateral side 1
Contralateral side 4
Both sides 0
Inspiratory related 2 26 28
Ipsilateral side
1
Contralateral side
1
Both sides 0
Type unidentified 44 44
Ipsilateral side 26
Contralateral side 11
Both sides 7
Total 59 72 131
Ipsilateral side 34
Contralateral side 17
Both sides 8
Presence of vestibular inputs from ipsilateral, contralateral, or both
sides is indicated.
exceptional cells were no more than 10% higher than
those for the facial nerve.
The responses of two cells in the caudal dorsomedial
brain stem to vestibular nerve stimulation are illus-
trated in Fig. 3. The neuron depicted in Fig. 3A was
excited at short latency (2.5 ms) by single-shock stimula-
tion of the vestibular nerve. The neuron shown in Fig.
3B was also powerfully excited by a single stimulus
applied to the vestibular nerve, but at a much longer
latency ( - 11 ms). Figure 4 indicates the latency at
which neurons were affected by vestibular nerve stimu-
lation, measured from the first of five shocks in Fig. 4A
and from the effective shock in Fig. 4B. Of the 27 cases
in which latency from the effective shock was deter-
mined, five neurons were identified as receiving short-
latency ( < 4 ms) input. Three of these cells were excited,
Table 2. Effects of vestibular nerve stimulation on excitability of neurons in caudal dorsomedial brain stem
No. of Cells
Excited Inhibited Both excited and inhibited
Ipsilateral Contralateral Ipsilteral Contralateral Ipsilateral Contralateral
side side side side side side
Abdominal vagal inputs 1 1 1 2
Cervical vagal inputs 1 2 1
Baroreceptor inputs 3 1 1
Inspiratory related 1 1
Type unidentified 16 8 12 6 5 4
Total 18 12 17 9 7 4
Neurons with both ipsilateral and contralateral vestibular input are represented twice.
on November 2, 2008 ajpregu.physiology.orgDownloaded from
VESTIBULAR INPUTS TO NTS
R979
A 60 pA L
10
spikes l-
2ms
5ms
5ms
Fig. 3. Poststimulus time histograms showing responses of neurons
in NTS and adjacent regions to vestibular nerve stimulation.
A:
short-latency response of a neuron to a single, low-intensity (60 FA)
shock applied to contralateral vestibular nerve. Arrow indicates when
shock was delivered; 132 sweeps were averaged. Threshold for activat-
ing fibers in facial nerve (300 fl) was 5 times higher than intensity
used in this run. B: responses of another neuron to contralateral
vestibular nerve stimulation at 4 different intensities (B1-B4). Stimu-
lus intensity necessary to produce current spread to facial nerve was
220 FA in this animal. No. of sweeps averaged to produce each
histogram:
Bl,
119;
BZ, 78;B3, 48;B4, 75.
one was inhibited, and the fifth was both excited and
inhibited by vestibular nerve stimulation. The inputs to
three of the neurons came from the ipsilateral side,
whereas the other two were driven from the contralat-
eral side. The mean response onset latency (for the cells
for which values could be determined accurately) was
11.5 t 7.4 ms (n = 27) from the effective shock and
22.1 t 13 ms (n = 43) from the first of five shocks.
The locations of the studied neurons with vestibular
input are shown in Fig. 5. Neurons that responded to
vestibular nerve stimulation were scattered throughout
the caudal and intermediate aspects of the NTS, as well
as the nucleus intercalatus; cells with excitatory and
inhibitory inputs were intermingled. Arrows identify
the five neurons with short-latency vestibular inputs,
which were located in the nucleus intercalatus (2 cells),
the commissural subdivision of the NTS (1 cell), and the
ventromedial region of the NTS (2 cells).
Convergence of vagal and vestibular inputs. Sixteen
neurons with abdominal vagal input and 15 cells with
cervical vagal input were tested for responses to stimula-
tion of the vestibular nerve. These units were inter-
mingled throughout the medial and intermediate parts
of the NTS and intercalatus. The vagal responses of
these cells were typically excitatory and were elicited by
stimulus trains of one to five shocks (median of 4 shocks
for the abdominal nerve and 3 shocks for the cervical
nerve). The abdominal vagal effects had a mean latency
of 226 t 55 ms, whereas those elicited from the cervical
vagus had a latency of 75 t 54 ms. Such a long response
latency is not unexpected, because vagal tierents are
predominately unmyelinated or only thinly myelinated.
Four of the neurons (25%) with abdominal vagal inputs
and four cells (27%) with cervical vagal inputs received
convergent vestibular signals; examples are shown in
Fig. 6. Two of the four neurons with abdominal vagal
inputs were driven at short latency ( < 4 ms) by vestibu-
lar nerve stimulation, whereas units with cervical vagal
inputs received only longer-latency vestibular signals
(Fig. 4). The 1 ocations of neurons with convergent
vestibular and vagal inputs are shown in Fig. 5.
Convergence of baroreceptor and vestibular inputs.
We also studied a group of NTS neurons (n = 28, Table
1) that received baroreceptor inputs; these cells were
confined to the dorsal and dorsolateral parts of the
nucleus. We identified these neurons as having firing
patterns that were related to fluctuations in blood
pressure during different phases of the cardiac cycle.
The mean spontaneous blood pressure oscillations dur-
ing these experiments were 47 t 9 mmHg. The majority
B
8
7
ul 6- Aa
00
Convergent Input Unknown
l
E 5- n 00
A A 6aroreceptor
hqnit
q m Vagal Input
2
4-
0.00
O+ Respiratory Related
5
3-
0000
*
2-
0000
l- 000000.00 0
0 ~‘I”I’~I’~,“,‘~,”
0 9 18 27 36 45 54
Response Latency (ms)
Fig. 4. Response latencies to stimulation of ipsilateral (open symbols)
and contralateral (filled symbols) vestibular nerve, measured from the
1st of 5 shocks (i.e., from stimulus onset;
A)
and from the effective
shock
(B).
No. of neurons represented in
A
is less than no. of neurons
in our sample with vestibular input because onset latencies could not
always be determined accurately (despite fact that presence of a
response was clear). This is particularly true for slowly firing neurons
that were inhibited by vestibular nerve stimulation. In addition, 5
neurons that responded to both ipsilateral and contralateral vestibular
nerve stimulation are represented twice in this figure. only 27 neurons are
represented in
B
because it was not possible to hold the other cells long
enough to determine the minimal no. of effective shocks.
on November 2, 2008 ajpregu.physiology.orgDownloaded from
R980
VESTIBULAR INPUTS TO NTS
Fig. 5. Locations of neurons with vestibular input. Posi-
tions are plotted on standardized transverse sections,
1
mm caudal to (- 1 mm), through 2 mm rostra1 to (+2
mm) the obex, as indicated above each panel. Cu, cuneate
nucleus; Gr, gracile nucleus; IVN, inferior vestibular
nucleus; MLF, medial longitudinal fasciculus.
-1 .O
mm
0.0
mm
1.0
mm
-0.5
mm
--s-m--- -1
1.5
mm
2.0 mm
*
B
R
X
of cells (76%) was activated by increases in blood
pressure (example shown in Fig. 7). However, a few
units were inhibited as blood pressure increased, and
activated when it declined.
Five of the 28 barosensitive neurons were identified as
having vestibular input (Tables 1 and 2). An example is
shown in Fig. 7. The latency from the effective shock
was determined for only one of the neurons (the cell
illustrated in Fig. 7, whose response latency was - 10
ms). However, neither this cell nor any of the others
(whose response latencies were at least 25 ms from the
first of 5 shocks) were affected at short latency by
vestibular stimulation.
Vestibular inputs to NTS neurons with inspiratory-
related activity. A third group of 28 neurons located in
the ventrolateral NTS was identified as having activity
in phase with discharges recorded from the phrenic
nerve. Considerable averaging (200-300 sweeps) was
done when testing for vestibular inputs to these cells to
cancel out the spontaneous respiratory-related dis-
charges. During these experiments, stimulation of the
Vestibular Input
1 mm
Convergent Baroreceptor + Vestibular
Inspiratory-Related + Vestibular
Convergent Vagal + Vestibular
Vestibular Latency < 4 ms
vestibular nerve evoked large responses in the phrenic
nerve (for a description of these effects, see Ref. 30);
lmitted to
However
thus vestibular signals were being trans
ratory motoneurons in these animals.
1
.nspi-
only
two of the units showed any responses at
lar nerve stimulation, and the responses all
of to vestibu-
these cells
were extremely weak, indicating that the inspiratory
neurons in the NTS receive little vestibular input.
DISCUSSION
Vestibular inputs to the NTS and adjacent nucleus
intercalatus were demonstrated using both neuroana-
tomic and electrophysiological techniques. Anterograde
transport of PHA-L demonstrated that some of these
inputs came directly from the medial and inferior vestibu-
lar nuclei. Vestibular response latencies of NTS neurons
confirmed the existence of short-latency inputs but also
demonstrated that labyrinthine inputs reach the NTS
through multisynaptic pathways. It was assumed that
NTS and intercalatus units that were driven from the
on November 2, 2008 ajpregu.physiology.orgDownloaded from
VESTIBULAR INPUTS TO NTS
R981
4
Abdominal X
spikes
L
6
Cervical X
spikes
L
Al
50 ms
I 1 A
Bl
50 ms
Al A I I
Vestibular N. (160 pA)
6
spikes
L
Vestibular N. (100 pA)
7
spikes
L
2:Ulll.r.
Vestibular N. (120 pA)
3d I A 1 A
5
spikes
L
50 ms
Y
Fig. 6. Convergence of vagal and vestibular inputs.
Al:
response of a neuron to a 2-mA single shock applied to
abdominal vagus (X) nerves, which were mounted together on stimulating electrode; 38 sweeps were averaged. AZ:
combined excitation and inhibition produced in same cell by 5 stimuli applied to ipsilateral vestibular nerve (N); 59
sweeps were averaged to produce histogram.
Bl:
response of another cell to stimulation of ipsilateral cervical vagus
nerve. Excitation was elicited by single shock of 1-mA intensity; 44 sweeps were averaged.
B2
and
B3:
inhibition
recorded from same unit after ipsilateral vestibular nerve stimulation at 2 different intensities. Large shock artifacts
(which surpassed upper level of window discriminator) obscured counting of action potentials during time in which
stimuli were delivered, explaining apparent silence of unit during this period. However, observations of oscilloscope
tracings during experiment revealed there was no obvious decrease in unit firing during stimulation.
B2
trace was
generated from 64 sweeps,
B3
trace from 52 sweeps.
15
spikes
Bl
4
spikes
25
ms
3
spikes
3
spikes
Fig. 7. Convergence of baroreceptor and vestibular inputs.
Al
: electro-
cardiogram (ECG) used to trigger averaging of blood pressure (BP)
and unit activity; 28 cycles were averaged. AZ: BP fluctuations
entrained by triggering from ECG.
A3:
poststimulus histogram of unit
activity. Firing rate is maximal when BP increases, and minimal when
BP is at its lowest level.
B:
responses of same unit to stimulation of
vestibular nerve on contralateral side using 3 different train lengths.
Stimulus intensity was 225 FA in all 3 runs; minimal stimulus
strength that resulted in current spread to facial nerve was 425 kA.
Traces in
Bl-B3
were generated from 71, 56, and 77 sweeps,
respectively. As in Fig. 6, large shock artifacts (which surpassed upper
level of window discriminator) obscured counting of action potentials
during time in which stimuli were delivered, explaining apparent
silence of unit during: this Period.
labyrinth at latencies < 4 ms received monosynaptic, or
at least relatively direct, inputs from the vestibular
nuclei. This assumption is based on the following calcu-
lation: neurons in the medial vestibular nucleus that
receive monosynaptic vestibular inputs are driven 0.8-
1.5 ms after stimulation of the vestibular nerve (27), an
estimated additional 1.5-2 ms are required for impulse
propagation from cell bodies in the vestibular nuclei to
the caudal dorsomedial brain stem, and a further
0.5-l
ms are needed for synaptic delay time and spike initia-
tion in the second-order cell. The neurons in the sample
that responded to vestibular nerve stimulation at longer
latencies presumably received multisynaptic inputs; the
additional interneurons could have been located within
the vestibular nuclei (so that some of the neurons with
longer-latency vestibular inputs could still have received
direct inputs from the vestibular nuclei) or elsewhere in
the brain stem. Two possible relays include the caudal
medullary raphe nuclei and the reticular formation of
the rostra1 ventrolateral medulla, both of which receive
vestibular inputs (29,
31)
and make connections with
the caudal dorsomedial brain stem (5,25).
PHA-L injections into the medial and inferior vestibu-
lar nuclei produced labeling of axons projecting bilater-
ally to the intermediate and caudal levels of the NTS. A
number of subnuclei received projections, including
those in the lateral and middle parts of the nucleus.
These regions receive inputs from the alimentary tract
(1, 13)
and the area postrema
(10, 22),
suggesting that
they may be involved in the production of emesis. All of
the neurons activated by stimulation of the abdominal
vagus nerve and some of those driven by stimulation of
the cervical vagus nerve (which contains pulmonary and
cardiovascular afferents as well) presumably received
abdominal inputs. One-fourth of the neurons with vagal
inputs responded to vestibular nerve stimulation. Two
of the four cells with abdominal vagal inputs were
on November 2, 2008 ajpregu.physiology.orgDownloaded from
R982
VESTIBULAR INPUTS TO NTS
activated from the labyrinth at latencies ( < 4 ms, suggest-
ing that their vestibular inputs came directly from the
vestibular nuclei. Because stimulation of the abdominal
vagus nerve can serve as a powerful trigger for emesis
(14, 19) that is mediated by the NTS (14), some of the
neurons with convergent vestibular and vagal inputs
might have been involved in producing vomiting. This
observation leaves open the possibility that motion
sickness-related nausea and vomiting are elicited in part
through activation of neurons in the NTS that also
mediate emesis induced by gastrointestinal inputs.
The dorsal and dorsolateral aspects of the NTS receive
cardiovascular afferents in a variety of mammalian
species (4, 13). There was a relatively heavy bilateral
projection from the inferior and medial vestibular nuclei
to the lateral subnucleus of the NTS; however, this
projection was concentrated ventral to the area (the
dorsal margin of NTS) containing the majority of neu-
rons receiving baroreceptor inputs. A population of
barosensitive neurons was identified in the dorsolateral
aspect of the NTS, a fraction (18%) of which was driven
by stimulation of the vestibular nerve. Because we
stimulated the vestibular nerve on only one side during
these studies (so that the glossopharyngeal nerve on the
opposite side would remain intact), we likely underesti-
mated the magnitude of vestibular effects on barosensi-
tive NTS neurons. NTS neurons with convergent baro-
receptor and vestibular inputs had response latencies to
vestibular nerve stimulation > 10 ms, and usually > 25
ms from the first of five shocks. A major synaptic target
of neurons in the NTS with baroreceptor inputs is the
subretrofacial rostra1 ventrolateral medulla (7), which
contains cells with projections to the sympathetic inter-
mediolateral cell column in the thoracic spinal cord (3,
7). The subretrofacial-spinal projection has been postu-
lated as being a major route through which vestibular
signals reach sympathetic preganglionic neurons (3 1).
Subretrofacial cells respond to vestibular nerve stimula-
tion at latencies ranging from
3.4
to 50 ms, the mean
onset latency being 14 t 2 ms (31). Thus neurons in the
NTS with baroreceptor inputs could be responsible for
the longer-latency vestibular effects on subretrofacial
neurons, although other pathways must also be involved
in relaying the shortest-latency vestibular inputs to
subretrofacial-spinal neurons.
The ventrolateral aspect of the NTS, which contains
inspiratory-related neurons (8), received only a modest
vestibular input. Most of these inspiratory neurons,
which form the so-called dorsal respiratory group, project
to the cervical and thoracic spinal cord and convey
central respiratory drive to phrenic and external inter-
costal motoneurons (8). Vestibular inputs to the dorsal
respiratory group were minimal, despite the fact that
strong vestibulorespiratory responses (30) were simulta-
neously recorded from the phrenic nerve. Thus other
pathways must relay vestibular signals to inspiratory
motoneurons. One possibility is the ventral respiratory
group, which is located in the caudal ventrolateral
medulla near the nucleus ambiguus; this is the other
major medullary collection of inspiratory premotor neu-
rons (8). Additional pathways, including the vestibulospi-
nal and/or reticulospinal tracts [which are known to
make synapses on motoneurons in the vicinity of inspira-
tory motoneurons (27)], could also play a role, although
this possibility is yet to be explored experimentally.
Vestibulosolitarius connections are organized differ-
ently in the cat than in the rabbit. In the rabbit, the bulk
of the projection is directed to the ventrolateral part of
the NTS (Z), the region where inspiratory and expira-
tory neurons are located (12), suggesting that the
pathway is likely to influence respiration in this animal.
These differences suggest that vestibular inputs to the
NTS could play disparate roles in different species.
We also demonstrated a projection from the medial
and inferior vestibular nuclei to the nucleus intercala-
tus. Such a connection has previously been indicated
using the retrograde transport of horseradish peroxi-
dase (17). This region receives inputs from neck muscles
(23) and has projections to the superior colliculus (24)
and the fastigial nucleus and other regions of the
cerebellum (26). Thus vestibular inputs to the nucleus
intercalatus are likely to be involved in the control of
head movement.
Perspectives
Anatomic and physiological data from the cat suggest
that through connections with the NTS, the vestibular
system could influence cardiovascular control to main-
tain stable blood pressure during changes in posture and
trigger two of the hallmark manifestations of motion
sickness: nausea and vomiting. However, it seems un-
likely that respiratory control would be prominently
influenced by the vestibulosolitarius projection, because
only a few labeled axons could be traced to the ventrolat-
era1 part of the NTS, and only a small proportion of
neurons with inspiratory-related discharges responded
to vestibular nerve stimulation.
We gratefully acknowledge the assistance of Dr. Kenji Endo in some
of the experiments. We also thank Drs. Victor Wilson, Robert Schor,
David Thomson, and Christian Xerri for helpful comments on the
manuscript.
This work was supported by National Institutes of Health Grants
DC-00693 (to B. J. Yates), DC-00739 (to C. D. Balaban), and NS-20585
(to A. D. Miller), as well as support from the Ministere des Affaires
Etrangeres (France; to L. Grelot) and a career scientist award from the
Irma T. Hirsch1 Trust (to B. J. Yates).
Present address for L. Grelot: Faculte des Sciences Saint Jerome,
Departement de Physiologie et Neurophysiologie, Case 351, URA
CNRS 1832, F-13397 Marseille cedex 20, France.
Present address for I. A. Kerman: School of Medicine, University of
Pittsburgh, Scaife Hall, Pittsburgh, PA 15261.
Present address for J. Jakus: Dept. of Physiology, Comenius Univ.,
Muzealna St. 4,036Ol Martin, Slovakia.
Address for reprint requests: B. J. Yates, Dept. of Otolaryngology,
Univ. of Pittsburgh, Eye and Ear Institute Bldg., 203 Lothrop St.,
Pittsburgh, PA 15213.
Received 27 December 1993; accepted in final form 9 May 1994.
REFERENCES
1. Altschuler, S. M., X. Bao, D. Bieger, D. A. Hopkins, and
R. R. Miselis. Viscerotopic representation of the upper alimen-
tary tract in the rat: sensory ganglia and nuclei of the solitary and
spinal trigeminal tracts. J. Comp. Neural. 283: 248-268, 1989.
2. Balaban, C. D., and G. Beryozkin. Vestibular nucleus projec-
tions to nucleus tractus solitarius and the dorsal motor nucleus of
on November 2, 2008 ajpregu.physiology.orgDownloaded from
VESTIBULAR INPUTS TO NTS
R983
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
the vagus nerve: potential substrates for vestibulo-autonomic
interactions. Exp. Brain Res. 98: 200-212,1994.
Barman, S. M., and G. L. Gebber. Axonal projection patterns
of ventrolateral medullospinal sympathoexcitatory neurons. J.
NeurophysioZ. 53: 1551-1566,1985.
Berger, A. J. Distribution of carotid sinus nerve afferent fibers to
solitary tract nuclei of the cat using transganglionic transport of
horseradish peroxidase. Neurosci. Lett. 14: 153-158, 1979.
Bongianni, F., M. Corda, G. A. Fontana, and T. Pantaleo.
Reciprocal connections between rostra1 ventrolateral medulla and
inspiration-related medullary areas in the cat. Brain Res. 565:
171-174,199l.
Borison, H. L. Area postrema: chemoreceptor circumventricular
organ of the medulla oblongata. Prog. NeurobioZ. 32: 351-390,
1989.
Dampney, R. A. L., J. Czachurski, K. Dembowsky, A. K.
Goodchild, and H. Seller. AflFerent connections and spinal
projections of the pressor region in the rostra1 ventrolateral
medulla of the cat. J. Autonom. Nerv. Syst. 20: 73-86,1987.
Feldman, J. L. Neurophysiology of breathing in mammals. In:
Handbook of Physiology. The Nervous System. Intrinsic Regula-
tory Systems of the Brain. Bethesda, MD: Am. Physiol. Sot., 1986,
sect. 1, vol. IV, chapt. 9, p. 463-524.
Grelot, L., J. Jaku6, A. D. Miller, and B. J. Yates. Vestibular
and visceral inputs to nucleus solitarius and adjacent structures
in the cat brainstem (Abstract). Sot. Neurosci. Abstr. 19: 1492,
1993.
Hay, M., and V. S. Bishop. Interactions of area postrema and
solitary tract in the nucleus tractus solitarius. Am. J. Physiol. 260
(Heart Circ. Physiol. 29): Hl466-H1473, 1991.
Ito, I., and I. Honjo. Central fiber connections of the vestibulo-
autonomic reflex arc in cats. Acta OtoZaryngoZ. 110: 379-385,
1990.
Jiang, C., Z.-H. Wu, and E. Shen. Antidromic mapping of
descending axons of respiratory bulbospinal neurons in the
nucleus tractus solitarius of the rabbit. Brain Res. 413: 189-192,
1987.
Kalia, M., and M.-M. Mesulam. Brain stem projections of
sensory and motor components of the vagus complex in the cat. II.
Laryngeal, tracheobronchial, pulmonary, cardiac and gastrointes-
tinal branches. J. Comp. Neural. 193: 467-508, 1980.
Koga, T., and H. Fukuda. Neurons in the nucleus of the solitary
tract mediating inputs from emetic vagal tierents and the area
postrema to the pattern generator for the emetic act in dogs.
Neurosci. Res. 14: 166-179, 1992.
Loewy, A. D. Central autonomic pathways. In: CentraZ Regula-
tion of Autonomic Functions, edited by A. D. Loewy and K. M.
Spyer. New York: Oxford Univ. Press, 1990, p. 88-103.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
McLean, I. W., and P. K. Nakane. Periodate-lysine-paraformal-
dehyde for immunoelectron microscopy. J. Histochem. Cytochem.
22: 1077-1083,1974.
Mergner, T., 0. Pompeiano, and N. Corvaja. Vestibular
projections to the nucleus intercalatus of Staderini mapped by
retrograde transport of horseradish peroxidase. Neurosci. Lett. 6:
309-313,1977.
Miller, A. D. Motion-induced nausea and vomiting. In: Nausea
and Vomiting: Recent Research and CZinicaZ Advances, edited by
J. Kucharczyk, D. J. Stewart and A. D. Miller. Boca Raton, FL:
CRC, 1991, p. 13-41.
Miller, A. D., L. K. Tan, and I. Suzuki. Control of abdominal
and expiratory intercostal muscle activity during vomiting: role of
ventral respiratory group expiratory neurons. J. Neurophysiol.
57: 1854-1866,1987.
Person, R. J. Somatic and vagal tierent convergence on solitary
tract neurons in cat: electrophysiological characteristics. Neurosci-
ence 30: 283-295,1989.
Sato, A., and R. F. Schmidt. Somatosympathetic reflexes:
tierent fibers, central pathways, discharge characteristics. Physiol.
Rev. 53: 916-947,1973.
Shapiro, R. E., and R. R. Miselis. The central neural connec-
tions of the area postrema of the rat. J. Comp. NeuroZ. 234:
344-364,1985.
Stechison, M. T., and J. A. Saint-Cyr. Organization of spinal
inputs to the perihypoglossal complex in the cat. J. Comp. Neural.
246: 555-567,1986.
Stechison, M. T., J. A. Saint-Cyr, and S. J. Spence. Projec-
tions from the nuclei prepositus hypoglossi and intercalatus to the
superior colliculus in the cat: an anatomical study using WGA-
HRP. Exp. Brain Res. 59: 139-150,1985.
Thor, K. B., and C. J. Helke. Serotonin- and substance
P-containing projections to the nucleus tractus solitarii of the rat.
J. Comp. Neural. 265: 275-293,1987.
Walberg, F. Fastigiofugal fibers to the perihypoglossal nuclei in
the cat. Exp. Neural. 3: 525-541,196l.
Wilson, V. J., and G. Melvill Jones. MammaLian Vestibular
PhysioZogy. New York: Plenum, 1979.
Yates, B. J. Vestibular influences on the sympathetic nervous
system. Brain Res. Rev. 17: 51-59,1992.
Yates, B. J., T. Goto, and P. S. Bolton. Responses of neurons
in the caudal medullary raphe nuclei of the cat to stimulation of
the vestibular nerve. Exp. Brain Res. 89: 323-332,1992.
Yates, B. J., J. Jakug, and A. D. Miller. Vestibular effects on
respiratory outflow in the decerebrate cat. Brain Res. 629:
209-217,1993.
Yates, B. J., Y. Yamagata, and P. S. Bolton. The ventrolateral
medulla of the cat mediates vestibulosympathetic reflexes. Brain
Res. 552: 265-2.72,1991.
on November 2, 2008 ajpregu.physiology.orgDownloaded from
... There is also evidence that both NTS and LTF neurons transmit vestibular signals to RVLM neurons. NTS (18)(19)(20) and LTF (21) receive direct projections from the vestibular nuclei. Projections from the vestibular nuclei to NTS are glutamatergic (20), but the neurotransmitters used by projections from the vestibular nuclei to LTF in felines are unknown. ...
... There was not a significant difference in the magnitude of responses to head-up tilts for NTS, LTF, and RVLM neurons (p = 0.1, one-way ANOVA). Since RVLM neurons receive direct and/or multisynaptic inputs from NTS (1-3) and LTF (14)(15)(16)(17), and the vestibular nuclei project to both regions (18)(19)(20)(21), these data are consistent with the hypothesis that LTF and NTS neurons are components of the vestibulo-sympathetic reflex pathway. ...
... In contrast, neurons in caudal regions of the vestibular nuclei that provide inputs to LTF and NTS responded robustly to small-amplitude body rotations in conscious felines (31,33). Considering that the caudal aspect of the vestibular nuclei has direct projections to NTS and LTF (18)(19)(20)(21), it is unclear how gating of labyrinthine signals occurs in vestibulo-sympathetic responses, such that sympathetic nerve activity only changes during large-amplitude body rotations. It is feasible that NTS and LTF neurons also receive convergent inputs from other regions of the nervous system that modify their responses to vestibular stimuli. ...
Article
Full-text available
Considerable evidence shows that the vestibular system contributes to adjusting sympathetic nervous system activity to maintain adequate blood pressure during movement and changes in posture. However, only a few prior experiments entailed recordings in conscious animals from brainstem neurons presumed to convey baroreceptor and vestibular inputs to neurons in the rostral ventrolateral medulla (RVLM) that provide inputs to sympathetic preganglionic neurons in the spinal cord. In this study, recordings were made in conscious felines from neurons in the medullary lateral tegmental field (LTF) and nucleus tractus solitarius (NTS) identified as regulating sympathetic nervous system activity by exhibiting changes in firing rate related to the cardiac cycle, or cardiac-related activity (CRA). Approximately 38% of LTF and NTS neurons responded to static 40° head up tilts with a change in firing rate (increase for 60% of the neurons, decrease for 40%) of ~50%. However, few of these neurons responded to 10° sinusoidal rotations in the pitch plane, in contrast to prior findings in decerebrate animals that the firing rates of both NTS and LTF neurons are modulated by small-amplitude body rotations. Thus, as previously demonstrated for RVLM neurons, in conscious animals NTS and LTF neurons only respond to large rotations that lead to changes in sympathetic nervous system activity. The similar responses to head-up rotations of LTF and NTS neurons with those documented for RVLM neurons suggest that LTF and NTS neurons are components of the vestibulo-sympathetic reflex pathway. However, a difference between NTS/LTF and RVLM neurons was variability in CRA over time. This variability was significantly greater for RVLM neurons, raising the hypothesis that the responsiveness of these neurons to baroreceptor input is adjusted based on the animal's vigilance and alertness.
... Through mediating the bidirectional communication of the gut-brain axis, it regulates the ingestive behavior and even nausea and vomiting [12]. There are four neural pathways potentially responsible for triggering vomiting by direct projections to the nucleus of the solitary tract (NTS) in the hindbrain: 1) gut vagal afferent fibers innervating the stomach and intestine which are stimulated by paracrine factors (e.g., serotonin) [10,[13][14][15]; 2) motion-related vestibular input from the vestibule in the inner ear [10,16,17]; 3) area postrema (AP) potentially detects circulating toxins [18] and 4) descending pathways from the forebrain [19,20]. Vagotomy and AP ablation in dog and ferret revealed that opioids produce emesis by action on the AP [21] while isoflurane-induced emesis is mediated by an action on the hindbrain rather than the abdominal vagus [22]. ...
Article
Full-text available
Background Postoperative nausea and vomiting (PONV) as a clinically most common postoperative complication requires multimodal antiemetic medications targeting at a wide range of neurotransmitter pathways. Lacking of neurobiological mechanism makes this ‘big little problem’ still unresolved. We aim to investigate whether gut-vagus-brain reflex generally considered as one of four typical emetic neuronal pathways might be the primary mediator of PONV. Methods Three thousand two hundred twenty-three patients who underwent vagus nerve trunk resection (esophagectomy and gastrectomy) and non-vagotomy surgery (hepatectomy, pulmonary lobectomy and colorectomy) from December 2016 to January 2019 were enrolled. Thirty cases of gastrectomy with selective resection on the gastric branch of vagus nerve were also recruited. Nausea and intensity of vomiting was recorded within 24 h after the operation. Results PONV occurred in 11.9% of 1187 patients who underwent vagus nerve trunk resection and 28.7% of 2036 non-vagotomy patients respectively. Propensity score matching showed that vagotomy surgeries accounted for 19.9% of the whole PONV incidence, much less than that observed in the non-PONV group (35.1%, P < 0.01). Multivariate logistic regression result revealed that vagotomy was one of underlying factor that significantly involved in PONV (OR = 0.302, 95% CI, 0.237-0.386). Nausea was reported in 5.9% ~ 8.6% vagotomy and 12 ~ 17% non-vagotomy patients. Most vomiting were mild, being approximately 3% in vagotomy and 8 ~ 13% in non-vagotomy patients, while sever vomiting was much less experienced. Furthermore, lower PONV occurrence (10%) was also observed in gastrectomy undergoing selective vagotomy. Conclusion Patients undergoing surgeries with vagotomy developed less PONV, suggesting that vagus nerve dependent gut-brain signaling might mainly contribute to PONV.
... Dysfunctions of the vestibular system do manifest in various neurological clinical signs, such as head tilt, nystagmus, or ataxia [3,4]. Additionally, signs of kinetosis (motion sickness) can also occur due to the fact that the nucleus of the solitary tract, which is responsible for nausea and emesis, receives vestibular input [5]. Thompson et al. (2009) [6] pointed out that various common central and peripheral vestibular diseases in humans (e.g., peripheral and central vestibular syndrome, Meniere's syndrome, vestibular neuritis, labyrinthitis and vestibular migraine) are associated with symptoms of nausea and vomiting and require intervention with anti-nausea medication. ...
Article
Full-text available
Background Vestibular syndrome is often accompanied by nausea. Drugs currently approved for its treatment have been developed to stop vomiting but not nausea. The efficacy of 5-HT 3 receptor antagonists to reduce nausea has been described for chemotherapy, but not for nausea secondary to vestibular disorders. Methods Sixteen dogs with vestibular syndrome-associated nausea were included in the open-label, multicentre study. The intensity of nausea-like behaviour was analysed before ondansetron administration (0.5 mg/kg i.v.) and 2 h afterwards, using a validated 5-point-scale. The occurrence and frequency of salivation, lip licking, restlessness, vocalisation, lethargy, and vomiting were assessed. Results All dogs initially showed signs of nausea, whereas only 31% showed vomitus. The intensity of nausea was significantly reduced in all dogs ( p ≤ 0.0001) 2 h after ondansetron administration, including the clinical signs of nausea analysed in 11 dogs (salivation [ p = 0.0078], lip licking [ p = 0.0078], restlessness [ p = 0.0039], and lethargy [ p = 0.0078]) except for vocalisation ( p > 0.9999). Conclusions The results provide preliminary evidence of the potential benefit of ondansetron in the treatment of nausea, which was present in all examined dogs. Vomiting was only observed in 5 dogs indicating that nausea can occur separately and should not be perceived only as a preceding stimulation of the vomiting centre.
... There are also afferent (53) and efferent (54) connections to autonomic nuclei in the rabbit. These findings together with the fact that neural activity in the caudal medial, lateral and descending VN (55)(56)(57)(58) are critical for VSR (59,60) may contain a mediator that modulates BP (61)(62)(63) is consistent with the model structure that is being proposed. ...
Article
Full-text available
Vasovagal syncope (VVS) or neurogenically induced fainting has resulted in falls, fractures, and death. Methods to deal with VVS are to use implanted pacemakers or beta blockers. These are often ineffective because the underlying changes in the cardiovascular system that lead to the syncope are incompletely understood and diagnosis of frequent occurrences of VVS is still based on history and a tilt test, in which subjects are passively tilted from a supine position to 20° from the spatial vertical (to a 70° position) on the tilt table and maintained in that orientation for 10–15 min. Recently, is has been shown that vasovagal responses (VVRs), which are characterized by transient drops in blood pressure (BP), heart rate (HR), and increased amplitude of low frequency oscillations in BP can be induced by sinusoidal galvanic vestibular stimulation (sGVS) and were similar to the low frequency oscillations that presaged VVS in humans. This transient drop in BP and HR of 25 mmHg and 25 beats per minute (bpm), respectively, were considered to be a VVR. Similar thresholds have been used to identify VVR's in human studies as well. However, this arbitrary threshold of identifying a VVR does not give a clear understanding of the identifying features of a VVR nor what triggers a VVR. In this study, we utilized our model of VVR generation together with a machine learning approach to learn a separating hyperplane between normal and VVR patterns. This methodology is proposed as a technique for more broadly identifying the features that trigger a VVR. If a similar feature identification could be associated with VVRs in humans, it potentially could be utilized to identify onset of a VVS, i.e, fainting, in real time.
... There are four neural pathways potentially responsible for triggering vomiting by direct projections to the nucleus of the solitary tract (NTS) in the hindbrain: 1) gut vagal afferent bers innervate the stomach and intestine and are stimulated by paracrine factors (e.g., serotonin), [10,[12][13][14] 2) motion-related vestibular input from the vestibule in the inner ear [10,15,16] 3) area postrema (AP) potentially detects circulating toxins [17] and 4) descending pathways from the forebrain. [18,19] Vagotomy and AP ablation in dog and ferret revealed that opioids produce emesis by action on the AP, [20] while iso urane-induced emesis is mediated by an action on the hindbrain rather than the abdominal vagus [21]. ...
Preprint
Full-text available
Background Postoperative nausea and vomiting (PONV) as a clinically most common postoperative complication requires multimodal antiemetic medications targeting at a wide range of neurotransmitter pathways. Lacking of neurobiological mechanism makes this big little problem still unresolved. We hypothesized that gut-vagus-brain reflex generally considered as one of four typical emetic neuronal pathways is the primary mediator of PONV. Methods 3223 patients who underwent vagotomy (esophagectomy and gastrectomy) and non-vagotomy surgery (hepatectomy, pulmonary lobectomy and colorectomy) from December 2016 to January 2019 were enrolled. Nausea and intensity of vomiting (mild, < 3 times;severe, ≥ 3 times) was recorded within 24 h after the operation. Results In the whole PONV incidence, vagotomy surgeries with vagus nerve trunk resection significantly reduced PONV incidence by approximately 4-fold (from 80.6–19.4%). Multivariate logistic regression result revealed that vagotomy was one of underlying factor that significantly involved in PONV (OR = 0.311; 95% CI, 0.246–0.393). Propensity score matching found that 35.3% patients who underwent vagus nerve trunk resection experienced PONV, while PONV was seen in 60.9% patients undergoing surgeries with intact vagus nerve (P < 0.001). Nausea was reported in 5.9%ཞ8.6% vagotomy and 12ཞ17% non-vagotomy patients. Most vomiting were mild, being approximately 3% in vagotomy and 8ཞ13% in non-vagotomy patients. Sever vomiting was much less experienced in patients undergoing vagotomy (ཞ1%) and non-vagotomy (ཞ3%). Conclusion Vagus nerve dependent gut-brain signaling mainly contributes to PONV, highlighting the new approach that modulates vagal activation to alleviate anesthetics and surgical stress-induced nausea and vomiting.
... The copyright holder for this preprint this version posted October 25, 2020. ; https://doi.org/10.1101/2020.10.23.352542 doi: bioRxiv preprint Vestibular Influences on LTF and NTS 19 ...
Preprint
Considerable evidence shows that the vestibular system contributes to adjusting sympathetic nervous system activity to maintain adequate blood pressure during movement and changes in posture. However, only a few prior experiments entailed recordings in conscious animals from brainstem neurons presumed to convey baroreceptor and vestibular inputs to neurons in the rostral ventrolateral medulla (RVLM) that provide inputs to sympathetic preganglionic neurons in the spinal cord. In this study, recordings were made in conscious felines from neurons in the medullary lateral tegmental field (LTF) and nucleus tractus solitarius (NTS) identified as regulating sympathetic nervous system activity by exhibiting changes in firing rate related to the cardiac cycle, or cardiac-related activity (CRA). Approximately 38% of LTF and NTS neurons responded to static 40° head up tilts with a change in firing rate of ~ 50%. However, few of these neurons responded to 10° sinusoidal rotations in the pitch plane, in contrast to prior findings in decerebrate animals that the firing rates of both NTS and LTF neurons are modulated by small-amplitude body rotations. Thus, as previously demonstrated for RVLM neurons, in conscious animals NTS and LTF neurons only respond to large rotations that lead to changes in sympathetic nervous system activity. The similar responses to head-up rotations of LTF and NTS neurons with those documented for RVLM neurons suggest that LTF and NTS neurons are components of the vestibulo-sympathetic reflex pathway. However, a difference between NTS/LTF neurons and RVLM was variability in CRA over time. This variability was significantly greater for RVLM neurons, raising the hypothesis that the responsiveness of these neurons to baroreceptor input is adjusted based on the animal’s vigilance and alertness.
... Activation of NTS inhibits rostral ventrolateral medulla, where sympathetic activity is thought to be mainly controlled. [25,26] Further, it was reported that blood was lowered followed by caloric and rotational vestibular stimulation, and this effect was abolished on the vestibular lesion. [27,28] The relation between mental stress and gastric acid secretion is not clear. ...