Download Swallowing reflex and brain stem neurons activated by superior

Am J Physiol Gastrointest Liver Physiol
280: G191–G200, 2001.
Swallowing reflex and brain stem neurons activated by
superior laryngeal nerve stimulation in the mouse
Q. SANG AND RAJ K. GOYAL
Center for Swallowing and Motility Disorders, West Roxbury Division of Veterans Affairs
Boston Healthcare System, West Roxbury 02132; and Harvard Medical School,
Boston, Massachusetts 02215
Received 28 February 2000; accepted in final form 24 August 2000
peristalsis is a centrally programmed sequence of oropharyngeal and esophageal
contractions that is orchestrated by a so-called swallowing program generator (SPG) in the brain stem (10,
13, 14, 20). The SPG neurons receive input from peripheral oropharyngeal afferents and central cortical
projections, which are responsible for activating reflex
swallowing (10, 13). The premotor neurons send commands to motoneurons that execute the motor peristaltic sequence.
It has been shown that distinct components of the
SPG are responsible for oropharyngeal and esophageal
phases of swallowing (10, 13, 14). The oropharyngeal
swallowing is evoked by oropharyngeal afferents carried primarily in the superior laryngeal nerves (SLN)
that project onto the second-order interneurons in interstitial solitary subnuclei (SolI) and intermediate
solitary subnuclei (SolIM) (16, 24). Studies (3) using
pseudorabies virus as a transneuronal tracer have
shown that these subnuclei also contain premotor neurons for motoneurons in semicompact formation of the
nucleus ambiguus (NAsc), loose formation of the nucleus ambiguus (NAl), and other cranial nerve nuclei
that innervate the oropharyngeal striated muscles. Interneurons in SolI and SolIM also project onto central
solitary subnuclei (SolCe) interneurons (8). SolCe neurons are premotor neurons for motoneurons in the
compact formation of the nucleus ambiguus (NAc) that
innervate the esophageal striated muscle (4, 15). During swallow-evoked primary peristalsis, the oropharyngeal and esophageal components of SPG act in
concert. However, the oropharyngeal and esophageal
phases may be dissociated (10). Oropharyngeal activity
without esophageal peristalsis may occur when SolI
and SolIM interneurons fail to activate SolCe interneurons. On the other hand, secondary peristalsis (esophageal peristalsis alone) is elicited when SolCe interneurons are activated by afferents arising from the
esophagus (17).
The above conclusions apply primarily to the activity
in the striated muscle of the oropharynx and esophagus (6, 8, 13, 20). The preganglionic motoneurons for
the smooth muscle portion of the esophagus and the
lower esophageal sphincter (LES) are located in the
dorsal motor nucleus of vagus (DMV) (1, 9, 28, 29, 30);
however, location of their premotor neurons is unclear
(2, 27).
We investigated the swallowing responses elicited by
SLN stimulation at various frequencies in mice. Electrical stimulation of the SLN at 10 Hz produced a
primary peristalsis, 5 Hz produced oropharyngeal contractions and LES relaxation without esophageal peristalsis, and 1 Hz produced no response. We also examined the location of putative swallow-related neurons
in the brain stem that may be involved in these reflex
Address for reprint requests and other correspondence: R. K.
Goyal, Research and Development (151), VA Boston Health Care
System, 1400 VFW Parkway, West Roxbury, MA 02132 (E-mail:
[email protected]).
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
c-fos expression; solitary subnuclei; nucleus ambiguus; dorsal
motor nucleus of vagus; swallowing program generator
SWALLOW-INDUCED PRIMARY
http://www.ajpgi.org
G191
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.247 on May 12, 2017
Sang, Q., and Raj K. Goyal. Swallowing reflex and brain
stem neurons activated by superior laryngeal nerve stimulation in the mouse. Am J Physiol Gastrointest Liver Physiol
280: G191–G200, 2001.— The purpose of the present study
was to identify vagal subnuclei that participate in reflex
swallowing in response to electrical stimulation of the left
superior laryngeal nerve (SLN). SLN stimulation at 10 Hz
evoked primary peristalsis, including oropharyngeal and
esophageal peristalsis, and LES relaxation. It also induced
c-fos expression in interneurons in the interstitial (SolI),
intermediate (SolIM), central (SolCe), dorsomedial (SolDM)
and commissural (SolC) solitary subnuclei. Neurons in parvicellular reticular nucleus (PCRt) and area postrema (AP)
and motoneurons in the semicompact (NAsc), loose (NAl),
and compact (NAc) formations of the nucleus ambiguus and
both rostral (DMVr) and caudal (DMVc) parts of the dorsal
motor nucleus of vagus were also activated. The activated
neurons represent all neurons concerned with afferent SLNmediated reflexes, including the swallowing-related neurons.
SLN stimulation at 5 Hz elicited oropharyngeal and LES but
not esophageal responses and evoked c-fos expression in
neurons in SolI, SolIM, SolDM, PCRt, AP, NAsc, NAl, and
DMVc but not in SolCe, NAc, or DMVr. These data are
consistent with the role of SolI, SolIM, SolDM, NAsc, NAl,
and DMVc circuit in oropharyngeal peristalsis and LES relaxation and SolCe, NAc, DMVc, and DMVr in esophageal
peristalsis and LES responses.
G192
BRAIN STEM NEURONS IN SWALLOWING
activities by identifying the c-fos-activated neurons at
different SLN stimulus frequencies in separate animals. c-fos is an early response gene that is expressed
in transsynaptically activated neurons. Identification
of c-fos expression may serve as an important technique to identify neural circuits in reflex activities (11,
22, 26, 34).
MATERIAL AND METHODS
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.247 on May 12, 2017
Twenty-eight adult male Swiss Webster mice (Taconic,
Germantown, NY) weighing 20–40 g were used. The procedures were approved by the Animal Committee of the Brockton/West Roxbury Veterans Affairs Medical Center.
Esophageal manometry. Esophageal manometry was performed using a custom-designed catheter assembly (Dentsleeve,
Parkside, SA, Australia). The assembly consisted of three
silicon catheters (ID, 0.3 mm; OD, 0.6 mm) that were glued
together. The catheter was 6.5 cm long and had an outside
diameter of 1.2 mm. Each catheter had a 0.3-mm side opening; the side openings were located 2 mm apart. The catheters were filled and continuously perfused with bubble-free
preboiled water from a reservoir at a rate of 7 ␮l/min using a
pressurized nitrogen chamber. The catheters provided a lowcompliance system and were connected to flow through pressure transducers that were connected to ETH 400 Bridge
amplifiers (CB Science, Dover, NH). The output of pressure
signals was displayed and recorded on MacLab software (AD
Instruments, Castle Hills, NSW, Australia).
The mice were anesthetized with a subcutaneous injection
of pentobarbital sodium (60 mg/kg) and placed on a heating
pad. The catheter assembly was introduced into the esophagus through a gastrostomy. The catheters were then withdrawn gradually in 0.5- to 1-mm increments to scan for the
high-pressure zone of the LES. The three catheter openings
were positioned in the LES and at 2 and 4 mm above the
LES. Esophageal and LES pressure changes in response to a
spontaneous swallow and electrical stimulation of the SLN
were then recorded. The occurrence of artifactual LES relaxation-associated axial movement of the LES was excluded by
careful visual monitoring. Pharyngeal swallowing activity
was identified by visually observing the motion of the exposed laryngopharyngeal area in the neck. The pharyngeal
pressures were not recorded for technical reasons.
A midline neck incision was then made, and the left SLN
was exposed. In the mouse, the SLN arises from the vagus
below the nodose ganglion and lies dorsal to the carotid
artery. The SLN was isolated and severed at the point where
it enters the laryngeal musculature. The central end of the
SLN was placed on a bipolar (silver-platinum) hook electrode
and covered with warm mineral oil and stimulated using a
Grass stimulator (model S11; Quincy, MA). Current spread
to adjoining tissues caused local muscle contractions and was
avoided. The stimulus consisted of square wave pulses of 0.5
ms at 8 V, and stimulation was performed for a period of 9 s
at frequencies of 1, 5, or 10 Hz.
c-fos expression. The electrical stimuli used to induce c-fos
expression were similar to those used for the manometric
studies, except that the 9-s stimuli were repeated with a
stimulus free interval of 10 s for a period for 30 min. Repeated stimuli with 10-s stimulus-free intervals were applied
to avoid decay of the swallowing responses with continuous
stimuli (33). At the end of the experiment, the mice were
killed with a subcutaneous injection of pentobarbital sodium
(100 mg/kg) and perfused transcardially with 60 ml PBS
(0.9% NaCl in 0.01 M sodium phosphate buffer, pH 7.0)
followed by 60 ml of 4% formaldehyde (4% formaldehyde in
0.1 M sodium phosphate buffer, pH 7.0). The brain stem with
attached upper cervical spinal cord was removed and postfixed in the same fixative overnight at 4°C. The fixative was
removed with three washes in PBS, and the tissue was stored
in PBS containing 1% sodium azide plus 30% sucrose for 24 h
at 4°C before being sectioned. Frozen coronal serial sections
of 25-␮m thickness were cut and collected directly on silanecoated slides (Sigma Chemical, St. Louis, MO) so that the
orientation of the sections was always the same. The sections
were then processed for histochemical and immunohistochemical staining.
The first of every three sections was stained with cresyl
violet for anatomic landmark studies. The section was
washed in water and stained with 1% cresyl violet (Fisher
Scientific, Pittsburgh, PA) containing 0.25% acetic acid for 5
min. After being rinsed in water, the section was differentiated in 50% ethanol for 5 min and rinsed again in water. The
section was then mounted with the mounting medium (Sigma Chemical).
The second section was immunostained for choline acetyltransferase (ChAT) and c-fos. The section was incubated in a
mixture of primary antisera that contained a rabbit anti-c-fos
(1:3,000, Oncogene Research, Cambridge, MA) and a goat
anti-ChAT (1:100, Chemicon International, Temecula, CA)
for 16–24 h. The unbound antibodies were removed by three
washes in PBS. The section was then incubated in a biotinylated donkey anti-rabbit IgG (1:200, Jackson ImmunoResearch, West Grove, PA) and a biotinylated donkey antisheep IgG (1:500, Jackson ImmunoResearch) for 2 h and
Vectastain avidin-biotin complex kit (Vector Laboratory,
Burlingame, CA) for another 2 h. Immunoreactivity was
revealed using a peroxidase substrate kit (diaminobenzidine,
DAB; Vector Laboratory). We distinguished c-fos and ChAT
immunoreactivities by the location of the immunoreactive
products. c-fos stained the cell nucleus, whereas ChAT
stained the cell cytoplasm. ChAT immunoreactivity can be
revealed by a donkey anti-sheep IgG even though the primary antibody is a goat anti-ChAT (30).
The third section was immunostained for c-fos and neuronal nitric oxide synthase (nNOS). The section was incubated
in a mixture of a rabbit anti-c-fos and a rabbit anti-nNOS
(1:400, Santa Cruz Biotechnology, Santa Cruz, CA), and the
immunoreactivity was revealed by DAB reaction as described
above. Again, c-fos and nNOS immunoreactivities were also
distinguished by the location of immunoreactive products.
c-fos stained the cell nucleus, whereas nNOS stained the cell
cytoplasm.
All the preparations were examined with a Zeiss Axioplan
microscope. Images were captured on a Hamamatsu camera
with Openlab software.
c-fos-positive neurons per coronal sections that showed the
maximum number of activated neurons were quantitated
semiquantitatively and scored as follows: ⫺, for none; ⫹, for
less than 5; ⫹⫹, for between 5 and 10; ⫹⫹⫹, for between 10
and 50; and ⫹⫹⫹⫹, for greater than 50 c-fos-positive neurons.
In the present study, we used posterior limits of area
postrema (pAP) as the reference point for describing brain
stem levels in craniocaudal orientation. AP neurons are a
group of small cresyl violet-stained neurons with a subpopulation of ChAT-positive neurons. The craniocaudal extent of
the AP in the mouse is ⬃500 ␮m. Obex was not used as a
reference point, because the level of obex is defined differently by different investigators.
The solitary subnuclei (Sol) were identified by their relative location to anatomic landmarks, such as the fourth
ventricle (IV), AP, solitary tract (SolT), and DMV, and corre-
BRAIN STEM NEURONS IN SWALLOWING
RESULTS
Motor response of esophagus and LES to SLN stimulation. As shown in Fig. 1, during a spontaneous swallow manometric recording showed a propagating pressure wave along the esophageal body and relaxation
followed by an aftercontraction of the LES. Pharyngeal
swallowing activity that was associated with esophageal
manometric response was observed by visual inspection
of the exposed pharynx. Electrical stimulation of the
central end of the SLN induced esophageal pressure
changes in the esophagus and LES with frequency-dependent responses. Electrical stimulation of the SLN at 1
Hz did not induce any obvious pressure changes in either
the esophageal body or the LES. Electrical stimulation at
5 Hz induced pharyngeal response and a slight decrease
in pressures in the LES, but no esophageal response.
SLN stimulation at 10 Hz evoked a pharyngeal as well as
an esophageal response, including relaxation followed by
an aftercontraction of the LES. Each 9-s stimuli typically
evoked two swallowing responses.
c-fos expression in vagal subnuclei with SLN stimulation. Successful studies were obtained in 4 or 5 animals in
each group receiving SLN stimulation at 1, 5, and 10 Hz.
Electrical stimulation of the left SLN induced c-fos expression in neurons that were located over a wide area
along the entire rostrocaudal extent of the vagal nuclei.
c-fos was expressed bilaterally with activated neurons
being more numerous on the ipsilateral than the contralateral side.
In the solitary nucleus, c-fos expression was found in
several different solitary subnuclei (see Figs. 2, 3, and
Fig. 1. Examples of esophageal and lower esophageal sphincter (LES) pressure responses to spontaneous swallowing and superior laryngeal nerve (SLN) stimulation at 1, 5, and 10 Hz. A spontaneous swallow is characterized
by an esophageal peristaltic contraction associated with LES relaxation and aftercontraction. SLN stimulation at
5 Hz induced an LES relaxation without esophageal contraction. SLN stimulation at 10 Hz for 9 s elicited 2
episodes of propulsive contractions of the esophageal body and relaxation of the LES followed by an aftercontraction at 10-Hz stimulation. Stimuli were square wave pulses of 0.5 ms.
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.247 on May 12, 2017
lated with the cytoarchitectural characteristics as defined by
cresyl violet staining and nNOS staining.
From a coronal perspective, Sol is broadly divided into a
smaller lateral and a larger medial subdivision based on
their position in relation to the SolT. The lateral subdivision
is further subdivided into four subnuclei: dorsolateral
(SolDL), ventrolateral (SolVL), ventral (SolV), and SolI. The
medial subdivision is further subdivided into seven subnuclei: SolIM, SolCe, medial (SolM), commissural (SolC), dorsomedial (SolDM), gelatinosus (SolG), and rostrocentral
(SolRCe). SolCe can be easily distinguished from the other
Sol subnuclei as it stands out as a distinct group of compactly
packed nNOS-immunoreactive neurons. SolG is recognized
by its acellular gelatinous appearance due to the preponderance of the unmyelinated fibers. In the mouse, parvicellular
reticular nucleus (PCRt) is a group of scattered cells that lies
just ventral to the Sol and extends further ventrally to NA. In
the craniocaudal orientation, PCRt extends ⬃600 ␮m rostrally from the anterior AP (aAP).
In the mouse, DMV extended from 1,250 ␮m rostral to and
500 ␮m caudal to the pAP. It was arbitrarily divided into
rostral DMV (DMVr) and caudal DMV (DMVc). Immunohistochemical staining with ChAT or nNOS revealed that a
subpopulation of DMV neurons were ChAT- or nNOS-immunoreactive neurons.
The NA is located in ventral regions of the medulla and in
the mouse extends throughout its craniocaudal extent of the
medulla. It is easily recognized by the cholinergic nature of
its neurons. Similar to that described in the rat (7), in the
mouse the dorsal division of NA is composed of three rostrocaudally aligned subdivisions, including NAc, NAsc, and NAl
formations, respectively. The ventral division of the NA corresponds to the external formation that extends along the
entire length of the medulla.
G193
G194
BRAIN STEM NEURONS IN SWALLOWING
7 and Table 1). In SolI, c-fos immunoreactivity was
seen with all three stimulus frequencies examined. The
number of c-fos-positive neurons increased as the stimulation frequency increased (Fig. 2, A–C, and Table 1).
In SolIM, c-fos-positive neurons were observed
throughout its rostrocaudal extent, but they were most
numerous at a level just rostral to the aAP (500–600
␮m rostral to pAP). The number of c-fos-positive neurons also increased as the stimulus frequency increased (Fig. 2, D–F, and Table 1). Sparse c-fos-reac-
tive neurons were also seen in SolDM and SolCe. In
SolCe, c-fos-positive neurons were observed with 10-Hz
SLN stimulation but not with 1- or 5-Hz SLN stimulation (Fig. 3). In SolM, few c-fos-positive cells were seen
at 1 Hz, and their number did not change with higher
stimulus frequencies. We have interpreted these observations as indicating a nonspecific response.
In SolC, c-fos-positive neurons were seen with all the
stimulus frequencies throughout the rostocaudal extent of SolC. The number progressively increased in
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.247 on May 12, 2017
Fig. 2. c-fos expression in the solitary subnuclei (Sol) with SLN stimulation at different frequencies. Coronal
sections are shown from different animals processed for c-fos and neuronal nitric oxide synthase (nNOS)immunohistochemical staining. c-fos expression is identified as nuclear staining and nNOS as cytoplasmic staining
in neurons. A–C: sections from rostral part of the Sol showing that c-fos-reactive neurons are present in the
interstitial Sol (SolI; I) [present lateral to solitary tract (SolT; T)] with SLN stimulation at 1 Hz (A). The
c-fos-reactive neurons increased with stimulus frequency of 5 Hz (B) and increased further with 10-Hz stimulation
(C). D–F: sections at a level just rostral to the area postrema (AP). In these sections, the central Sol (SolCe; Ce) is
easily identifiable as a group of densely packed nNOS-positive neurons. The area between the SolCe and SolT
represents the region of the intermediate Sol (SolIM). Note an increase in c-fos-reactive neurons in SolIM with
increasing stimulus frequencies. DMV, dorsal motor nucleus of vagus. Some c-fos-positive neurons in the dorsal
part of the parvicellular reticular nucleus (PCRt) are shown clearly in E. Note also some c-fos-positive neurons in
the dorsomedial Sol (SolDM). Only a rare c-fos-reactive neuron was seen in the medial Sol (SolM), and the number
of neurons did not increase with increasing stimulus. IV, fourth ventricle. G–I: sections through caudal part of Sol
showing c-fos-positive neurons in commissural subnucleus (SolC; C). There was no obvious increase in c-fospositive neurons with increasing SLN stimulus frequencies in the SolC. cc, Central canal. J–L: sections through
mid-AP showing that the number of c-fos-positive neurons increased as the stimulation frequency increased. Scale
bars, 100 ␮m.
BRAIN STEM NEURONS IN SWALLOWING
Table 1. c-fos-Positive neurons in vagal subnuclei
with SLN stimulation
SolI
SolIM
SolM
SolCe
SolDM
SolC
AP
PCRt
DMVr
DMVc
NAsc
NAc
1 Hz
5 Hz
10 Hz
⫹ to ⫹⫹
⫹
⫺ to ⫹
⫺
⫺
⫹ to ⫹⫹
⫹⫹
⫹
⫺
⫹
⫹
⫺
⫹⫹ to ⫹⫹⫹
⫹⫹⫹
⫺ to ⫹
⫺
⫺ to ⫹
⫹⫹
⫹⫹⫹
⫹⫹ to ⫹⫹⫹
⫺
⫹ to ⫹⫹
⫹ to ⫹⫹
⫺
⫹⫹⫹
⫹⫹⫹
⫺ to ⫹
⫹
⫹
⫹⫹
⫹⫹⫹⫹
⫹⫹ to ⫹⫹⫹
⫹
⫹ to ⫹⫹
⫹ to ⫹⫹
⫹
the caudal part of the subnucleus (at levels of 700–850
␮m). However, the number of active neurons did not
change with increasing stimulus frequencies, suggesting that these neurons may not be related to swallowing function.
In addition to SolI, SolIM, SolDM, SolCe, and SolC,
SLN stimulation also evoked c-fos in the ventrolateral
solitary subnucleus (SolVL), which is concerned with
respiratory functions. Other solitary subnuclei including SolDL, SolV, SolRCe, SolM, and SolG did not show
any c-fos expression with SLN stimulation.
SLN stimulation also evoked c-fos expression in the
PCRt ,which lies just ventral to the Sol (Fig. 2E). The
c-fos was expressed bilaterally with ipsilateral predominance. SLN stimulation also induced c-fos in neurons
in the AP (Fig. 2, J–L). The active neurons were
distributed throughout the core of the AP and not
restricted to the rim of the AP as seen with the stimulation of subdiaphragmatic vagal afferents (29).
As in the premotor neurons in solitary subnuclei,
c-fos expression also showed a frequency-dependent
pattern in the vagal motor subnuclei. SLN stimulation
evoked c-fos in NAc and rostral part of NAsc. In the
NAsc, c-fos-positive neurons were seen with all the
frequencies examined (Fig. 4, A–C). However, in the
NAc no c-fos-positive neurons were found with 1- or
5-Hz stimulation but some c-fos-positive neurons were
found with 10-Hz stimulation (Fig. 4, D and E).
In the DMV, the pattern of c-fos-positive neurons in
response to SLN stimulation was different in the rostral and caudal parts. In the DMVr, c-fos-positive neurons were seen with 10-Hz stimulation but not with 1or 5-Hz stimulation (Fig. 5). Even with 10-Hz stimulation, only a small number of DMVr neurons showed
c-fos immunoreactivity. In contrast, in the DMVc, c-fospositive neurons were found with all three frequencies
of SLN stimulation (Fig. 6).
The number of neurons expressing c-fos with SLN
stimulation is summarized in Table 1. The c-fos expres-
Fig. 3. c-fos expression in the SolCe
with 5- and 10-Hz SLN stimulation.
Coronal sections were processed for cfos (nuclear) and nNOS (cytoplasmic)
double immunostaining. In A (low
power) and A⬘ (high power), lack of c-fosimmunoreactive neurons is shown in the
SolCe with 5-Hz stimulation. In B (low
power) and B⬘ (high power) 1 c-fos-positive neuron (arrow) is shown in the
SolCe with 10-Hz stimulation. A and B:
scale bars, 100 ␮m. A⬘ and B⬘: scale bars,
50 ␮m.
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.247 on May 12, 2017
⫺, Not detected; ⫹, the highest number of positive cells on a single
section was ⬍5; ⫹⫹, the highest number of positive cells on a single
section was between 5 and 10; ⫹⫹⫹, the highest number of positive
cells on a single section was between 10 and 50; ⫹⫹⫹⫹, the highest
number of positive cells on a single section was ⬎50. SLN, superior
laryngeal nerve; Sol, solitary subnucleus; SolI, interstitial Sol;
SolIM, intermediate Sol; SolM, medial Sol; SolCe, central Sol;
SolDM, dorsomedial Sol; SolC, commissural Sol; AP, area postrema;
PCRt, parvicellular reticular nucleus; DMV, dorsal motor nucleus of
vagus. DMVr, rostral DMV; DMVc, caudal DMV; NA, nucleus ambiguus; NAsc, semicompact formation of NA; NAc, compact formation of NA.
G195
G196
BRAIN STEM NEURONS IN SWALLOWING
sion of vagal subnuclei in their craniocaudal extent
with 10-Hz SLN stimulation is summarized in Fig. 7.
DISCUSSION
The brain stem circuit involved in swallowing has
been investigated using a variety of techniques (2, 5, 6,
14). These include studies (24, 31) of swallowing responses after microlesions or microstimulation of welldefined medullary regions, electrical activity in the
identified neurons in response to afferent stimuli that
elicit reflex swallowing, and morphological tracer studies (2–4, 9, 18, 19) that show connectivity of neurons in
the circuit. Together, they have produced important
information on the neural circuit concerned with reflex
swallowing. The above studies have been carried out in
a number of different animal species. Although no
studies have been reported in the mouse, available
studies show that despite some obvious species differences, the general organization of the swallowing circuit across animal species is fairly similar.
The present studies using esophageal manometry
and c-fos expression in brain stem nuclei provide information on functional anatomy of the medullary neurons that may constitute the swallowing-related neurons in the mouse. Moreover, by using the different
stimulus parameter of the SLN to elicit different
esophageal and LES responses, these studies also help
identify the medullary neurons that may be involved in
different components of swallowing responses. Although identification of the brain stem swallowing
circuit using c-fos expression as a marker has not been
described before, this technique has been used widely
to identify neurons involved in a variety of reflexes (11,
22, 26, 34). The results of c-fos expression studies
require careful interpretation and correlation with results of studies using different techniques. This is
because c-fos expression may not directly correlate
with a reflex response. For example, c-fos is not expressed in neurons with inhibitory links to SLN input.
Even in neurons with an excitatory link to SLN input,
c-fos expression is most marked in the first-order neurons, and it decreases progressively in second- and
third-order neurons. Stimulation of the SLN afferents
elicits a variety of swallow-related and other reflexes,
including gagging, retching, coughing, laryngeal
spasm, bronchoconstriction, and apnea. These reflexes
use overlapping neural circuits. However, the type of
reflex evoked by the SLN stimulation depends on the
stimulus parameters used (5, 10, 25). In the mouse,
SLN stimulation at 10 Hz elicited a full swallowing
response, consisting of oropharyngeal and esophageal
components including LES relaxation. Stimulation at 5
Hz produced only oropharyngeal and LES relaxation
without evoking an esophageal response. Much stronger SLN stimulation (⬎10 Hz) produced apnea. Therefore we did not use stimuli with frequencies ⬎10 Hz to
avoid activation of other SLN-mediated reflexes. Even
so, it is possible that some of the c-fos-expressing neurons observed in this study may be concerned with
functions other than reflex swallowing.
Although fiber tracing or electrophysiological studies
dealing with reflex swallowing are not available in the
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.247 on May 12, 2017
Fig. 4. c-fos expression in nucleus ambiguus (NA) with SLN stimulation.
Coronal sections processed for c-fos
(nuclear) and choline acetyltransferase
(ChAT) (cytoplasmic) double immunostaining. The NA can be recognized by
the presence of a group of ChAT-immunoreactive neurons. A–C: c-fos-positive
neurons (arrows) in the semicompact
formation of the NA (NAsc) with stimuli at 1, 5, and 10 Hz in 3 different
mice. D and E: c-fos reactivity in compact formation of NA (NAc) with stimuli at 5 and 10 Hz in 2 different mice.
Note the absence of nuclear c-fos expression with 5 Hz (D) and the presence of c-fos-positive neurons (arrow)
with a 10-Hz stimulus (E). Cytoplasmic ChAT staining in neurons is evident in both D and E. Scale bars, 50
␮m.
BRAIN STEM NEURONS IN SWALLOWING
G197
mouse, such data in other animal species provide good
correlation with c-fos data in the mouse. Studies (5, 12,
18, 19, 24) have shown that SLN afferents project onto
a large number of solitary subnuclei, such as SolVL,
SolI, and SolIM, that are concerned with respiratory
protective reflexes, including swallowing. In the
mouse, neurons in all these areas expressed c-fos in
response to SLN stimulation. In the rat, studies (2, 3,
8) using pseudorabies virus as a neurotracer have
shown that SolI and SolIM contain premotor neurons
for the NAsc and NAl that house motoneurons for
pharyngeal muscles. In the mouse, we found that electrical stimulation of the SLN evoked c-fos expression in
interneurons in SolI and SolIM and in motoneurons in
NAsc and NAl. SolI and SolIM neurons have also been
shown to project onto SolCe (8) that contain premotor
neurons for NAc motoneurons innervating the esophageal striated muscle (4). SLN stimulation at 5 Hz
produced only the oropharyngeal phase of swallowing,
and stimulation at 10 Hz produced a full swallowing
response. Therefore, in the mouse, interneurons in SolI
and SolIM and motoneurons in NAsc and NAl may
constitute the brain stem circuit for oropharyngeal
swallowing. Interneurons in SolI, Sol IM, and SolCe
and motoneurons in NAsc, NAl, and NAc may be responsible for the full swallowing reflex. These results
are consistent with the results of microstimulation and
electrophysiological studies (10, 16) showing that the
dorsal part of the SPG that includes the region of the
SolT nucleus is responsible for the coordination of the
oropharyngeal phase as well as the complete swallowing sequence.
Studies (4, 6, 17) using extracellular unit recordings,
microstimulation, and morphological circuit tracing
have also shown that the SolCe is the main site of
termination of esophageal afferents. In our studies,
SLN stimulation at 10 Hz but not at 5 Hz showed
c-fos-expressing neurons in SolCe, NAc, and esophageal body contractions. Moreover, the number of activated neurons even at 10 Hz was small. The reason for
this low level of activated neurons is not clear; however, it may be related at least in part to the fact that
SolCe interneurons represent second- or third-order
neurons in the swallowing circuit. These observations
suggest that the neural circuit responsible for the
esophageal phase of the peristalsis requires higherintensity SLN simulation than that for the pharyngeal
phase of the peristalsis and are consistent with the
view that SolCe-NAc may form the brain stem circuit
for esophageal peristalsis (4, 8, 17).
The SPG was originally mapped into two regions (13,
14). The dorsal region is constituted by Sol and adjoining PCRt and a ventral region encompassing NA and
the adjoining PCRt. These studies (13, 14) and others
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.247 on May 12, 2017
Fig. 5. c-fos expression in rostral DMV
(DMVr) with 5- and 10-Hz SLN stimulation. Coronal sections processed for
c-fos (nuclear staining) and ChAT (cytoplasm staining) double immunostaining. The location of DMVr can be
recognized as a group of ChAT-positive
neurons. In A (low power) and A⬘ (high
power), a lack of c-fos-immunoreactive
neurons is shown in DMVr with 5-Hz
stimulation. In B (low power) and B⬘
(high power), a c-fos-positive neuron is
shown (arrow) in the DMVr with 10-Hz
stimulation. A and B: scale bars, 100
␮m. A⬘ and B⬘: scale bars, 50 ␮m.
G198
BRAIN STEM NEURONS IN SWALLOWING
Fig. 7. Diagrammatic representation of subnuclei of vagal complex showing c-fos-active neurons (shaded area)
with 10-Hz SLN stimulation. The level of 0 represents the posterior border of the AP. The darkness of the shade
represents the density of c-fos-positive neurons in vagal subnuclei, with the darkest being the densest. Note that
SLN stimulation at 10 Hz that evoked a full swallowing motor response also induced c-fos in neurons in SolI,
SolIM, SolCe, NAc, NAsc, DMVc, and DMVr. SolDL, dorsolateral Sol; SolVL, ventrolateral Sol; SolV, ventral Sol;
SolRCe, rostrocentral Sol; SolG, gelatinous Sol; aAP, anterior AP; pAP, posterior AP.
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.247 on May 12, 2017
Fig. 6. c-fos expression in caudal DMV (DMVc) with different frequency SLN stimulation. Coronal sections were
processed for c-fos (nuclear staining) and ChAT (cytoplasm staining) double immunostaining. The location of DMVc
can be recognized as a group of ChAT-positive neurons. Neurons expressing c-fos (arrowheads) were found with
stimulation at all 3 frequencies shown. A–C: scale bars, 100 ␮m. A⬘–C⬘: scale bars, 50 ␮m.
BRAIN STEM NEURONS IN SWALLOWING
these neurons without producing a motor response. It
has been shown that SLN stimulation that does not
evoke swallowing activity can evoke electrical activity
in the swallowing neurons (16). The frequency-dependent motor responses and c-fos expression in response
to SLN stimulation further support the validity of the
results with SLN stimulation.
In summary, these studies show that SLN stimulation evokes swallowing and peristalsis and c-fos expression in interneurons constituting the putative SPG
and motoneurons in the brain stem. These findings are
consistent with the view that SolI, SolIM, NAsc, and
NAl neurons are responsible for oropharyngeal swallowing and that SolCe and NAc neurons form the
circuit for esophageal striated muscle peristalsis. Both
these circuits are involved in primary peristalsis. Further studies are needed to more precisely determine
the roles of PCRt in reflex swallowing and define links
that are responsible for bilateral integration of the two
hemi SPGs. Studies are also needed to test whether
SolDM, rather than SolCe or SolM, house premotor
neurons for the DMVc neurons that mediate LES relaxation during reflex swallowing.
We thank Drs. Priyattam Shiromani and D. V. Sivarao for technical advice and Donna Kantarges for editorial assistance.
This work was supported in part by National Institute of Diabetes
and Digestive and Kidney Diseases Research Grant DK-31092 and a
Merit Review Award from the Office of Research and Development,
Medical Research Service, Department of Veterans Affairs.
REFERENCES
1. Abrahams TP and Hornby PJ. Lower esophageal sphincter
relaxation evoked by stimulation of the dorsal motor nucleus of
the vagus in ferrets (Abstract). Gastroenterology 116: A589,
1999.
2. Altschuler SM, Bao X, Bieger D, Hopkins DA, and Miselis
RR. Viscerotopic representation of the upper alimentary tracts
in the rat: sensory ganglia and nuclei of the solitary and spinal
trigeminal tract. J Comp Neurol 283: 248–268, 1989.
3. Bao X, Wiedner EB, and Altschuler SM. Transsynaptic localization of pharyngeal premotor neurons in rat. Brain Res 696:
246–249, 1995.
4. Barrett RT, Bao X, Miselis RR, and Altschuler SM. Brain
stem localization of rodent esophageal premotor neurons revealed by transneuronal passage of pseudorabies virus. Gastroenterology 107: 728–737, 1994.
5. Bellingham MC and Lipski J. Morphology and electrophysiology of superior laryngeal nerve afferents and postsynaptic
neurons in the medulla oblongata of the cat. Neuroscience 48:
205–216, 1992.
6. Bieger D. Central nervous system control mechanisms of swallowing: a neuropharmacological perspective. Dysphagia 8: 308–
310, 1993.
7. Bieger D and Hopkins DA. Viscerotopic representation of the
upper alimentary tract in the medulla oblongata in the rat: the
nucleus ambiguus. J Comp Neurol 262: 546–562, 1987.
8. Broussard DL, Lynn RB, Wiedner EB, and Altschuler SM.
Solitarial premotor neuron projections to the rat esophagus and
pharynx: implications for control of swallowing. Gastroenterology
114: 1268–1275, 1998.
9. Collman PI, Tremblay L, and Diamant NE. The central
vagal efferent supply to the esophagus and lower esophageal
sphincter of the cat. Gastroenterology 104: 1430–1438, 1993.
10. Doty RW, Richmond WH, and Storey AT. Effect of medullary
lesions on coordination of deglutition. Exp Neurol 17: 91–106,
1967.
11. Fields RD, Eshete F, Stevens B, and Itoh K. Action potentialdependent regulation of gene expression: temporal specificity in
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.247 on May 12, 2017
using microelectrode recordings and chemostimulation
techniques have suggested that PCRt may play an
important role in reflex swallowing. In the present
study, we found that SLN stimulation evoked c-fos
expression in PCRt neurons, which is consistent with
the view of a possible role for PCRt neurons in reflex
swallowing (13, 14, 31, 32).
The LES in all animal species, including the mouse,
and the distal esophagus in many species are composed
of smooth muscles that receive preganglionic motor
innervation from DMV motoneurons (1, 9, 27–29). Several studies (1, 28) have suggested that DMVr and
DMVc neurons provide excitatory and inhibitory inputs, respectively, to the smooth muscle of the distal
esophagus and LES. In the present study, 10-Hz SLN
stimulation elicited a full swallowing response, including LES relaxation and aftercontraction, and induced
c-fos expression in both the DMVr and DMVc. On the
other hand, 5-Hz SLN stimulation elicited isolated
LES relaxation only and induced c-fos expression solely
in DMVc neurons. These observations are consistent
with those of previous studies (1, 21, 25, 27, 28) in
other species finding that inhibitory responses of the
LES are more sensitive to afferent stimulation at intensities lower than the excitatory responses.
Dendrites of DMV motoneurons have been shown to
penetrate and make contacts with neurons in SolM and
SolDM but not in SolCe (2, 23). In the present study,
SLN stimulation at 5 Hz activated neurons in SolDM
and DMVc but not in SolM or DMVr. Therefore it is
possible that SolDM may contain premotor neurons for
DMVc motoneurons that are involved in swallowevoked LES relaxation. Recent studies (27) have shown
that SolCe neurons project onto DMVc as well as DMVr
motoneurons, suggesting that the SolCe-DMVc/DMVr
neural circuit may also regulate LES function in primary peristalsis. Further studies are needed to resolve
this important issue.
Studies in several animal species have shown that
SLN afferents project onto SolI, SolIM, and SolCe neurons only ipsilaterally. However, ipsilateral SLN stimulation activates these nuclei bilaterally. The neural
connections responsible for bilateral integration of
swallowing with ipsilateral SLN stimulation are not
known. Studies (5, 18, 19, 31) in the cat have shown
that afferent fibers of ipsilateral SLN project bilaterally onto SolC over a 2.5-mm area. However, functional
studies (33) do not reveal a role for SolC in swallowing.
Therefore, c-fos activity in SolC may not be related to
swallowing. Recent studies (31) in the cat have suggested that the bilateral integration of swallowing activity may occur at the level of PCRt that receives
afferent projections from SLN bilaterally and make
contralateral connections between the half SPGs on
either side of the midline.
Low-frequency (1 Hz) SLN stimulation evoked no
oropharyngeal or esophageal motor activity. However,
it induced c-fos in a few neurons in some of the subnuclei. c-fos expression has been shown to correlate with
electrical spike potentials in neurons (26). Therefore, it
is possible that SLN stimulation at 1 Hz activates
G199
G200
12.
13.
14.
15.
16.
17.
19.
20.
21.
22.
23.
Ca2⫹, cAMP-responsive element binding proteins, and mitogenactivated protein kinase signalling. J Neurosci 17: 7252–7266,
1997.
Hashim MA and Bieger D. Excitatory amino acid receptormediated activation of solitarial deglutitive loci. Neuropharmacology 28: 913–921, 1989.
Jean A. Brainstem organization of the swallowing network.
Brain Behav Evol 25: 109–116, 1984.
Kessler JP and Jean A. Identification of the medullary swallowing regions in the rat. Exp Brain Res 57: 256–263, 1985.
Kobler JB, Datta S, Goyal. RK, and Benecchi EJ. Innervation of the larynx, pharynx, and upper esophageal sphincter of
the rat. J Comp Neurol 349: 129–147, 1994.
Lang IM, Medda BK, Ren J, and Shaker R. Characterization
and mechanisms of the pharyngoesophageal inhibitory reflex.
Am J Physiol Gastrointest Liver Physiol 275: G1127–G1136,
1998.
Lu WY and Bieger D. Vagal afferent transmission in the NTS
mediating reflex responses of the rat esophagus. Am J Physiol
Regulatory Integrative Comp Physiol 274: R1436–R1445, 1998.
Lucier GE, Egizii R, and Dostrovsky JO. Projections of the
internal branch of the superior laryngeal nerve of the cat. Brain
Res Bull 16: 713–721, 1986.
Mifflin SW. Laryngeal afferent inputs to the nucleus of the
solitary tract. Am J Physiol Regulatory Integrative Comp Physiol
265: R269–R276, 1993.
Miller AJ. The search for the central swallowing pathway: the
quest for clarity. Dysphagia 8: 185–194, 1993.
Mittal RK, Holloway RH, Penagini R, Blackshaw LA, and
Dent J. Transient lower esophageal sphincter relaxation. Gastroenterology 109: 601–610, 1995.
Morgan JI and Curran T. Stimulus-transcription coupling in
the nervous system: involvement of the inducible proto-oncogenes fos and jun. Annu Rev Neurosci 14: 421–451, 1991.
Norgren R and Smith GP. Central distribution of subdiaphragmatic vagal branches in the rat. J Comp Neurol. 207–223,
1988.
24. Ootani S, Umezaki T, Shin T, and Murata Y. Convergence of
afferents from the SLN and GPN in cat medullary swallowing
neurons. Brain Res Bull 37: 397–404, 1995.
25. Paterson WG, Rattan S, and Goyal RK. Experimental induction of isolated lower esophageal sphincter relaxation in anesthetized opossums. J Clin Invest 77: 1187–1193, 1986.
26. Rinaman L, Verbalis JG, Stricker EM, and Hoffman GE.
Distribution and neurochemical phenotypes of caudal medullary
neurons activated to express c-fos following peripheral administration of cholecystokinin. J Comp Neurol 338: 475–490, 1993.
27. Rogers RC, Hermann GE, and Travagli RA. Brainstem pathways responsible for esophageal control of gastric motility and
tone in the rat. J Physiol (Lond) 514: 369–383, 1999.
28. Rossiter CD, Norman WP, Jain M, Hornby PJ, Benjamin S,
and Gillis RA. Control of lower esophageal sphincter pressure
by two sites in dorsal motor nucleus of the vagus. Am J Physiol
Gastrointest Liver Physiol 259: G899–G906, 1990.
29. Sang Q and Goyal RK. Lower esophageal sphincter relaxation
and activation of medullary neurons by subdiaphragmatic vagal
stimulation in the mouse. Gastroenterology. In press.
30. Sang Q and Young HM. The origin and development of the
vagal and spinal innervation of the external muscle of the mouse
esophagus. Brain Res 809: 253–268, 1998.
31. Sugimoto T, Umezaki T, Narikawa K, and Shin T. Crossing
inputs of the superior laryngeal nerve afferents to medullary
swallowing-related neurons in the cat. Neurosci Res 30: 235–
245, 1998.
32. Umezaki T, Matsuse T, and Shin T. Medullary swallowingrelated neurons in the anesthetized cat. Neuroreport 9: 1793–
1798, 1998.
33. Weerasuriya A, Bieger D, Hockman CH. Interaction between
primary afferent nerves in the elicitation of reflex swallowing.
Am J Physiol Regulatory Integrative Comp Physiol 239: R407–
R414, 1980.
34. Willing AE, and Berthoud H-R. Gastric distension-induced
c-fos expression in catecholaminergic neurons of rat dorsal vagal
complex. Am J Physiol Regulatory Integrative Comp Physiol 272:
R59–R67, 1997.
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.247 on May 12, 2017
18.
BRAIN STEM NEURONS IN SWALLOWING