close

Вход

Забыли?

вход по аккаунту

?

Relationships between the morphology and function of gastric and intestinal distention-sensitive neurons in the dorsal motor nucleus of the vagus

код для вставкиСкачать
THE JOURNAL OF COMPARATIVE NEUROLOGY 364~78-91(1996)
Relationships Between the Morphology
and Function of Gastric and Intestinal
Distention-Sensitive Neurons in the
Dorsal Motor Nucleus of the Vagus
RONALD FOGEL, XUEGUO ZHANG, AND WILLIAM E. RENEHAN
Laboratory of Gastrointestinal, Gustatory, and Somatic Sensation, Division of
Gastroenterology, Henry Ford Health Sciences Center, Detroit, Michigan 48202
ABSTRACT
The activity of vagal motor neurons is influenced by sensory information transmitted to
the brainstem. In particular, there is evidence that distention of the stomach increases activity
of motor neurons in the dorsal vagal motor nucleus, whereas distention of the duodenum, small
intestine, and colon reduces neuron firing. In this study, we determined 1)the response of vagal
motor neurons to distention of the stomach and duodenum and 2) whether the response
properties were associated with specific morphological features. Using the single-cell recording
and iontophoretic injection technique, we identified four groups of vagal motor neurons affected
by gastric and/or duodenal distention. Group 1neurons responded to either gastric or duodenal
stimulation. Neurons in groups 2, 3, and 4 were affected by both gastric and duodenal
distention. Group 2 neurons were excited by duodenal distention and were inhibited by gastric
distention. Group 3 neurons were inhibited by duodenal distention and were excited by gastric
distention. Most neurons belonged to group 4. Neurons in this group were inhibited by both
gastric and duodenal distention. Our analyses revealed that the neurons affected by both
stimuli had distinctive structural features. Neurons in group 2 had the largest somata, the most
dendritic branches, and the greatest cell surface area. Neurons in group 3 were the smallest and
had the shortest dendritic length. In addition, we were able to demonstrate that the neurons in
group 4 had a smaller total dendritic length and a smaller cell volume than neurons in group 2
and had more dendritic branch segments than neurons in group 3. These results suggest that
morphological features are associated with specific response properties of vagal motor
neurons. o 1996 Wiley-Liss, Inc.
Indexing terms: gastrointestinal, visceral, mechanoreceptors, intracellular
The vagus nerve, which is the source of parasympathetic
innervation to much of the gastrointestinal tract, plays a
major role in the regulation of gastric acid secretion (Laughton and Powley, 1987; Tache et al., 19911, of endocrine and
exocrine pancreatic secretion (Laughton and Powley, 1987;
Roze, 1989; Li and Owyang, 19931, of gastrointestinal
motility (Bueno and Fioramonti, 19911, and of intestinal
mucosal water and ion absorption (Fogel et al., 1991; Chu
et al., 1993). Recently, there have been advances in our
knowledge of 1)the effects of visceral stimuli on the firing
rate of vagal efferent fibers innervating the abdomen and 2)
the location in the dorsal motor nucleus of the vagus
(DMNV) of the neurons that regulate various aspects of
physiological gastrointestinal function. However, at present, we have only an elementary understanding of the
mechanism(s) by which visceral sensory information influences the vagal control of gastrointestinal physiology.
O 1996 WILEY-LISS, INC.
Electrophysiological studies have shown that the vagal
efferent neurons that project to the viscera are functionally
heterogeneous. This conclusion is based on the results from
studies that described the response to stimulation of gastrointestinal mechanoreceptors. Most vagal efferent neurons
that are responsive to distention of the gastric antrum
appear to exhibit an increased firing rate (Davison and
Grundy, 1978; Grundy et al., 1981; Blackshaw et al., 1987;
Blackshaw and Grundy, 19891, whereas most of the vagal
neurons that are sensitive to distention of the duodenum,
small intestine, or colon exhibit a decrease in activity in the
Accepted July 3, 1995.
Address reprint requests to Ronald Fogel, M.D., Laboratory of Gastrointestinal, Gustatory, and Somatic Sensation, Division of Gastroenterology,
Henry Ford Health Sciences Center, 2799 West Grand Boulevard, Detroit,
MI 48202.
STRUCTURE-FUNCTION RELATIONSHIPS OF DMNV NEURONS
presence of these stimuli (Grundy et al., 1981; Blackshaw
et al., 1987; Zhang et al., 1992; Chu et al., 1993). How does
the site of gastrointestinal tract stimulation influence the
response of the vagal motor neuron? The answer to this
question may be found in the tract-tracing studies that
have demonstrated a visceral topography for the DMNV.
Using retrograde labels such as horseradish peroxidase
(HRP1, investigators have demonstrated that the majority
of motor neurons innervating the stomach (Pagani et al.,
1985; Altschuler et al., 1991) and the pancreas (Rinaman
and Miselis, 1987) are located in the medial aspect of the
DMNV, whereas neurons innervating the cecum are located
in the lateral DMNV (Altschuler et al., 1991).
Powley and collaborators have proposed that a sensorymotor lattice underlies the control of vagally mediated
gastrointestinal reflexes (Powley et al., 1992). According to
this theory, there are geographically separate nodes in the
vagal complex that represent the sites at which sensory
information from a specific location is transmitted to the
vagal efferent neurons that regulate the function of a
particular organ. If we combine this theory with the results
of the electrophysiology studies summarized above, then we
could predict that the gastric nodes are organized in such a
way that the sensory inputs excite DMNV neurons, whereas
the intestinal nodes produce inhibition. A corollary to the
lattice hypothesis would be that the location of the efferent
neuron in the vagal complex determines the nature of the
electrophysiological response.
It is clear, however, that we must reach beyond this level
of understanding if we are to comprehend the more complex
(and potentially more important) features of the vagal
response to gastrointestinal stimulation. Distention of the
gastric antrum identifies two groups of vagal motor neurons. Although most neurons are excited by antral distention, a smaller number respond with a reduction in electrophysiologxal activity (Davison and Grundy, 1978; Grundy
et al., 1981; Blackshaw et al., 1987; Blackshaw and Grundy,
1989). We have recently shown that one neuron in the
lateral DMNV may be excited by distention of the small
intestine, another may be inhibited (the usual response),
and a third may show no response at all (Zhanget al., 1992).
How is it that neurons in the same column of the DMNV
(presumably innervating the same region of the gastrointestinal tract and receiving the same sensory input) demonstrate different responses to a visceral stimulus? Is it
possible that differences in the soma-dendritic morphology
and/or connectivity of DMNV neurons are superimposed on
the basic lattice network and may correlate with differences
in the response properties of these cells? Numerous studies
have established the morphological heterogeneity of vagal
motor neurons in the rat (Prechtl and Powley, 1990),
guinea pig (Elfvin and Lindh, 1982; Laiwand et al., 1987;
Engel and Kreutzberg, 1988; Jou et al., 19931, cat (Smolen
and Truex, 1977; McLean and Hopkins, 1981; McLean and
Hopkins, 1982), monkey (McLean and Hopkins, 1985), and
human (Huang et al., 1993), but the functional importance
of these anatomic differences has not been addressed. In a
recent report, Fox and Powley concluded that there was no
meaningful difference in the morphology of the neurons
from different columns in the DMNV (Fox and Powley,
1992). This result would appear to cast doubt on the
existence of any relationships between the structure and
function of DMNV neurons. We suggest, however, that
such a conclusion is premature. To examine the potential
relationship(s) between DMNV morphology and physiology
79
with confidence, we contend that one must utilize an
intracellular recording and labeling technique that permits
the investigator to determine the structure of individual,
physiologically characterized neurons. In fact, we have
recently used such a technique to demonstrate that DMNV
neurons responsive to intestinal distention had at least one
morphological characteristic that correlated with responsivity (Zhang et al., 1992). Specifically, we found that 9 0 9 of
the DMNV neurons that did not project out of the brainstem (i.e., the axon did not join the vagus nerve) were
completely inhibited by intestinal distention, whereas only
5 0 4 of the DMNV neurons that did project out of the
brainstem were completely inhibited by this stimulus.
Although the importance of this observation is not completely clear at this time, this finding does provide strong
evidence for the existence of a relationship between the
structure and function of DMNV neurons. Such a relationship would have broad implications for any theory that
attempted to explain the central nervous system circuitry
that participated in the regulation of gastrointestinal function.
In the present study, we explored the possibility that
DMNV neurons that were responsive to stimulation of
different regions of the gastrointestinal tract (or exhibited
different responses to these stimuli) might exhibit differences in their soma-dendritic morphology. We employed the
single-cell recording and intracellular labeling technique to
obtain morphological and functional information for individual vagal motor neurons affected by gastric and/or
duodenal distention. Our results allowed us to conclude
that certain features of DMNV dendritic architecture are
indeed associated with specific response profiles in the
presence of these two stimuli.
MATERIALS AND METHODS
All studies described in this paper were approved by the
Institutional Animal Care and Use Committee of the Henry
Ford Health Sciences Center. Rats were anesthetized with
intraperitoneal pentobarbital sodium (45 mgikg) before
study. The trachea was cannulated for ventilator-assisted
respiration (Harvard rodent ventilator; South Natick, MA).
A midline abdominal incision exposed the abdominal vagus
nerve, the stomach, and the duodenum. Teflon-insulated
silver wire stimulating electrodes (76 pm diameter) were
placed around the anterior and posterior vagi 2 cm above
the gastroesophageal junction. These electrodes, which
were used to stimulate the vagus nerve, were fixed to the
esophagus and the stomach to prevent displacement.
Thirty-five animals were studied. Gastric and duodenal
distention were performed in the same animal. For gastric
distention, a small incision was made at the most dependent
portion of the greater curve of the stomach. Care was taken
not to damage the extrinsic nerves innervating the stomach
and duodenum. A catheter (I.D. 4.0 mm) for infusion of
fluid was placed into the stomach. The pylorus was transected, and a catheter to permit fluid efflux was sutured into
the gastric outlet. For duodenal distention, the influx
catheter was sutured into the duodenal bulb, and the efflux
catheter was placed 10 cm distal to the ligament of Treitz.
After placement of the gastric and duodenal catheters,
the abdominal incision was closed, and the rats were placed
in a Kopf small animal stereotaxic frame. While rats were in
the stereotaxic frame, their body temperature was maintained by a thermostat-controlled heating pad. The brain-
80
stem was exposed by removing the atlantooccipital membrane and a portion of the occipital bone. Beveled glass
micropipettes (tip diameter 0.2-0.6 pm; resistance, 50-70
Mohm) filled with 2.0% Neurobiotin in 1 M KCl were
lowered into the DMNV between 100 pm rostra1 and 400
pm caudal to the obex (defined as the point at which the
central canal opened into the fourth ventricle). We have
recently shown (Zhang et al., in preparation) that this
region of the DMNV contains the majority of the vagal
motor neurons that project to the duodenum and stomach.
Search stimuli consisted of biphasic pulses (0.5 msec duration, 3 mA, 1 Hz) delivered to the abdominal vagus.
Recording micropipettes were advanced until a driven unit
was encountered. All units driven by the stimulating electrode were tested for a response to duodenal or gastric
distention.
Once a DMNV neuron driven by the vagal stimulating
electrode was identified, the response of the neuron to
perfusion of the stomach and the duodenum with 0.9%
saline was determined. Recordings were made while the
efflux catheter was in the plane of the animal and were
made again when the efflux catheter was elevated 20 cm
above the plane of the animal.
The change in firing rate in response to distention was
tested at least three times. If the response was equivocal,
then the stimulus was presented up to five times. Unit
discharges, which were recorded extracellularly, were amplified by an A-M Systems high-input-impedance microelectrode amplifier (A-M Systems, Everett, WA) and were
displayed and stored on an IBM-compatible 386 MHz
computer with the use of Axotape software (Axon Instruments, Foster City, CA). After response characterization,
the micropipette was advanced until the DMNV neuron was
impaled (this process was facilitated by passing small
positive current pulses from the recording electrode). Penetration of the cell membrane was accompanied by a 20-40
mV drop in resting membrane potential, by an increase in
the amplitude of the action potential, and by a shift from a
bipolar to a monopolar action potential. To confirm that the
cell impaled was the one characterized, receptive fields were
checked prior to Neurobiotin injection. After confirmation
that the response properties were the same, cells were
labeled with Neurobiotin by passing 2-4 nA, 250 msec
positive current pulses at 2 Hz for 2-7 minutes. The
injection was stopped if at any time the resting potential
returned t o prepenetration levels. A maximum of two
injections were attempted on each side. This limitation was
necessary because of the reconstruction protocol for the
injected neurons. To prevent errors in the assignment of
dendritic branches to a specific neuron, it was necessary
that the dendritic arbors of the two neurons did not overlap.
One to six hours after the first injection, rats were
administered a lethal dose of pentobarbital sodium and
perfused through the heart with 500 ml of 0.9% saline
containing 2,000 Uiliter heparin in 0.1 M sodium phosphate buffer, pH 7.3, at room temperature. The rinse
solution was followed by 500 ml fixative containing 1%
paraformaldehyde and 2.5% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4, at 4°C. The brainstem was stored
overnight in 0.1 M phosphate buffer containing 20% sucrose. The tissue was frozen, and 50 pm sections were cut
on a sliding microtome. Tissue was developed with Avidin
D-HRP solution containing cobalt and nickel to enhance
visualization of Neurobiotin-labeled neurons. Sections were
mounted on gelatin-coated slides, and some sections were
R. FOGEL ET AL.
counterstained with neutral red to facilitate identification
of DMNV borders.
Determination of intraluminal pressure
In four animals, we determined the change in intraluminal pressure associated with elevations of the efflux catheter. A PE-50 polyethylene tubing (O.D. 0.97 mm) was
passed through the lumen of the efflux catheter and
positioned approximately half way between the influx and
efflux catheters. The tubing was attached to a calibrated
Statham pressure transducer, and the signal was transcribed by a chart recorder (Gould, Cleveland, OH). The
viscera were returned to the abdomen. The intraluminal
pressure was recorded (during infusion of 0.9% saline)
while the applicable efflux catheter was in the plane of the
rat and was recorded again after the catheter was elevated
by 20 cm.
Statistical analyses
Neurophysiological analysis. Neuronal response properties were examined using the Datapac software system
(Run Technologies, Laguna Niguel, CAI. Peristimulus time
histograms were constructed for the period beginning 30
seconds before and ending 90 seconds after initiation of
distention (distention was maintained for 60 seconds). Each
bin size was 5 seconds. To determine whether a given
response was significantly more or less than the baseline,
the mean activity during the distention period was compared to the mean activity during the pre-and poststimulus
periods using analyses of variance (ANOVA), and the Tukey
B test was employed for posthoc comparisons. Only corrected cx values less than 0.05 were considered statistically
significant.
Morphometric analysis. Three-dimensional reconstructions of individual Neurobiotin-labeled neurons that were
digitized at a magnification of x400 ( ~ 4 dry
0 objective)
were made using the Eutectic Neuron Tracing System
(Eutectic Electronics, Raleigh, NC). This system allows
complete reconstruction of branching structures, even if
the structures are cut during sectioning (Capowski, 1989).
Each reconstruction was verified using the “mathematical
completeness” subroutine of the Neuron Tracing System.
Optical and physical compression (Overdijk et al., 1978;
Uylings et al., 1986) of the tissue did occur in the plane of
section, and the neuronal components in each section were
rescaled to 50 pm (original thickness at the time of
sectioning).
Sixteen features of each neuron were assessed. These
features were grouped into six categories including 1)
indices of overall size [total dendritic length, total surface
area, and total internal volume (including somatic volume)]; 2) measures of somatic shape and size [form factor
FF, a measure of circularity for which a value of 1.0
indicates a perfect circle, and 0 indicates a line; FF =
(47;a)/p2, where a equals soma area, and p equals the
perimeter of the soma in the coronal plane; and somatic
cross-sectional areal; 3) measures related to dendritic
branching [number of primary dendrites, total number of
branch points, total number of branches (a branch segment
is defined as a portion of a dendrite between two consecutive branch points), number of branches per primary
dendrite, and highest branch orderj; 4) measures of dendritic length (average length of each primary dendrite and
average length of individual dendritic branches); and, finally, indices of connectivity such as 5 ) dendritic spines
STRUCTURE-FUNCTION RELATIONSHIPS OF DMNV NEURONS
(total number of spines and spine density) and 6 ) dendritic
swellings (total number of swellings and swelling density).
The neuron was reconstructed with and without the
axon. The reconstruction that did not include the axon was
used to calculate dendritic length and cell areas. Morphological features were analyzed using ANOVA with Tukey B test
correction for multiple comparisons.
Separately, the neurons were reexamined using a x40
objective to determine the destinations of the dendrites.
Areas examined included the ipsilateral DMNV, the contralateral DMNV, the medial subnucleus of the nucleus of the
solitary tract (NST),the subpostremal NST, the gelatinous
subnucleus of the NST, the fourth ventricle, and the central
canal. For each functional group, the number of neurons
with dendrites projecting to each of the above destinations
was tabulated. One-way ANOVA with Tukey B posthoc
tests were used to determine whether there were any
statistically significant differences in dendritic projections
among the functional groups.
RESULTS
Determination of intraluminal pressure
Elevation of the gastric efflux catheter to 20 cm above the
plane of the animal increased intragastric pressure proximal to the efflux catheter by 13.4 ? 0.1 mm Hg but did not
change the pressure in the nondistended duodenum. Similarly, elevation of the duodenal efflux catheter on two
occasions increased intraduodenal pressure by 12 mm Hg
each time. The increase in pressure persisted and remained
constant for the entire time that the efflux catheter was
elevated. Returning the efflux catheter to the plane of the
animal caused a return to baseline pressure. The intragastric pressure was not affected by elevation of the duodenal
efflux catheter.
Neurobiotin-labeled DMNV neurons
Figure 1 shows representative vagal motor neurons that
were characterized and injected with Neurobiotin. All sections are in the coronal plane. The neuron in Figure 1A was
located in the left DMNV in the dorsal aspect of the medial
portion of the nucleus 150 pm caudal to the obex. The
neuron is multipolar and gives rise to a number of ventral
dendrites that remain in the DMNV. A dorsal dendrite (not
seen in this section) reaches the subpostremal subnucleus
of the NST. Figure 1B illustrates a triangular-shaped
neuron in the left DMNV 350 pm caudal to the obex. In this
instance, the axon arises from a primary dendrite (only the
initial portion of the axon can be seen in this section). The
well-labeled axon could be followed in adjacent sections as it
proceeded toward the ventrolateral border of the brainstem
to join the vagus nerve. The neurons in Figure 1C,D were
located in the right DMNV. Both neurons have somata that
are triangular in shape and have axons arising from a
primary dendrite. The neuron in Figure 1C was located 50
Fm caudal to the obex. The large medial dendrite shown in
this section bifurcated, with the larger dorsal branch
reaching the subpostremal NST and the smaller ventral
branch extending to the central canal. The neuron in
Figure 1D was located at the level of the obex. A number of
the dendrites associated with this cell projected to the
gelatinous subnucleus of the NST, whereas the remainder
were restricted to the medial NST.
81
Characterization of DMNV neurons
Fifty-two DMNV neurons were characterized, labeled,
and reconstructed. The locations of the neurons in the
DMNV are shown in Figure 2. Our recordings were restricted to the middle or medial portions of the DMNV.
Forty-nine of the neurons in the present study were located
between the obex and 600 pm caudal to the obex. The other
three neurons were rostral to the obex.
Four groups of DMNV neurons were identified based on
the profile of their response to gastric and duodenal distention: group 1, neurons affected by either duodenal or gastric
distention ( n = 8 ) ; group 2, neurons excited by duodenal
distention and inhibited by gastric distention (n = 7 ) ;group
3, neurons inhibited by duodenal distention and excited by
gastric distention (n = 7); and group 4,neurons inhibited
by both duodenal and gastric distention ( n = 30). The
neurons in the four groups were interspersed in the region
of study (medial and middle DMNV; Fig. 2 ) .
Group 1. The neurons in the first group responded to
either gastric or duodenal distention. Four of the neurons
were inhibited by gastric distention, and four were inhibited by duodenal distention. The mean spontaneous activity
rate for the neurons in this group was 0.83 ? 0.2 Hz. Two of
the neurons affected by duodenal distention and two affected by gastric distention had complete inhibition of firing
for the entire period of distention. The firing rate returned
to prestimulus levels when the distention stimulus was
removed. The other four neurons had inhibition of firing
for only a portion of the period of distention.
A representative group 1 neuron is shown in Figure 3.
The response profile shown in Figure 3A indicates that this
neuron was not affected by duodenal distention. Conversely, gastric distention completely inhibited neuronal
firing for 26 seconds and partially inhibited the neuron for
the remainder of the stimulus period. After the distention
was released, the firing rate increased transiently and then
returned to baseline. The latency of the neuronal response
to electrical stimulation of the vagal nerve (shown in Fig.
3C) was 81 msec.
This neuron was located in the left DMNV. Figure 3D-F
presents computer-generated images of the neuron as it
would be seen in coronal, horizontal, and sagittal sections,
respectively. The neuron was multipolar and had a fusiform
soma. The axon arose from the cell soma adjacent to a
primary dendrite. The horizontal view is very instructive. It
illustrates most clearly the fact that the dendrites tend to
occupy three distinct regions of the nucleus. One set of
dendritic branches is directed toward a region that is rostral
and medial to the soma, another (somewhat smaller) set
proceeds rostrally and laterally, and a third set (arising
from a separate primary dendrite) is directed primarily
toward a region that is caudal and lateral to the soma. The
coronal section shows that one of the branches enters into
the subpostremal region of the NST. A more detailed
discussion of the group morphology with comparison to the
other physiological groups will be presented below.
Group 2. Seven neurons were excited by duodenal distention and were inhibited by gastric distention. These neurons, as a group, had a basal firing rate of 2.0 2 0.3 Hz. This
basal activity was significantly greater than that seen in any
of the other groups. Five of the seven neurons responded to
intestinal distention with a n immediate but transient increase in firing rate, whereas two neurons had a sustained
increase in firine rate for the duration of the intestinal
82
R. FOGEL ET AL.
Fig. 1. Photomicrographs of Neurobiotin-labeled distention sensitive neurons in the dorsal motor nucleus of t h e vagus (DMNV).
Neurons shown in A and B were located in t h e left DMNV. Neurons in
C and D were located in the right DMNV. All four neurons were
multipolar. The arrow indicates t h e neuronal process identified a s t h e
axon. The neuron shown in A belongs to group 2. B-D are examples of
neurons in group 4.
distention. In response to gastric distention, six of these
seven neurons had complete inhibition of firing during the
entire period of gastric distention, whereas the seventh
neuron was completely inhibited for 33 seconds and then
had a gradual increase in firing rate despite persistence of
the gastric distention.
A representative group 2 neuron is shown in Figure 4.
This neuron had a basal activity of 3.6 Hz. Figure 4A
illustrates that distention of the duodenum increased neuronal activity. The neuron responded to distention by an
increase in firing rate to a peak of 7.4 Hz recorded at 18
seconds after the onset of duodenal distention. The effect
persisted for the rest of the distention period. During the
last 30 seconds of distention, the firing rate was 6.6 Hz. At
the end of the distention period, the neuronal firing rate
returned to baseline. Gastric distention (Fig. 4B) caused a
complete inhibition of neuronal activity that started 3
seconds after the onset of gastric distention and that
persisted for 33 seconds. Partial inhibition of neuronal
firing was observed during the remainder of the distention
period. There was an increase in firing rate above baseline
for the 30 seconds after the gastric distention stimulus was
removed. The latency to electrical current stimulation of
the vagus nerve (Fig. 4C) was 85 msec.
The neuron was located in the right medial DMNV 100
km caudal to the obex. The neuron was bipolar (Fig. 4D).
One group of dendrites was oriented in a dorsal and rostra1
direction, whereas the other group was directed caudally,
ventrally, and medially. Dendrites entered the overlying
NST. Note that the scale bar for this neuron is 200 Fm,
indicating that these neurons have a greater dendritic
extension than the neurons in the other groups.
Group 3. The seven neurons in group 3 were inhibited
by duodenal distention and were excited by gastric distention. This group of neurons had a mean spontaneous
activity rate of 0.6 t 0.1 Hz. Five of the seven neurons had
complete inhibition of firing rate with duodenal distention,
whereas two neurons exhibited incomplete inhibition. The
increase in firing rate in response to gastric distention was
present for the entire period of distention in one animal but
took place only during the last 30 seconds of distention in
the other six neurons.
The distention response and morphology of one of these
neurons is shown in Figure 5. Figure 5A illustrates the fact
that this DMNV neuron was completely inhibited by distention of the duodenum. The excitatory effect of gastric
distention was less dramatic (Fig. 5B) but was nonetheless
statistically significant. The increase in activity in response
to gastric distention developed gradually over the 60 second
stimulus period, but the return to baseline following stimulus offset was relatively prompt (approximately 6 seconds).
This particular neuron exhibited a latency of 120 msec to
electrical stimulation of the subdiaphragmatic vagus nerve.
This was one of the longest antidromic stimulation latencies noted in this study.
STRUCTURE-FUNCTION RELATIONSHIPS OF DMNV NEURONS
- v -
83
0 Group 1 neurons
o Group 2 neurons
.
.
.
........:
Fig. 2 . Schematic representation of the DMNV showing the location
of Neurobiotin-labeled neurons. A is 200 pm rostra1 to the obex. B-E
are 0, 200.400, and 600 pm caudal to the obex, respectively. Neurons
are identified by the physiological group to which they were assigned.
.
.
.
-..........
NST, nucleus of the solitary tract; gNST, gelatinous subnucleus of the
NST; mNST, medial subnucleus of the NST; sNST, subpostremal
region of the NST; AP, area postrema; CC, central canal; XII, hypoglossal nucleus.
R. FOGEL ET AL.
84
A
+
Duodenal distention
Release
4
ft
Dorsal
Scale: 12 seconds
Medial
Coronal plane
B
I
1
E
+
+
Release
Gastric distention
Horizontal plane
r,i
Scale: 12 seconds
F
C
Stimulus
artifact
-c
Neuronal
response
4
Dorsal
I, I
Sagittal plane
-1
Scale: 30 mseconds
Fig. 3. Response properties and morphology of a neuron that is
representative of group 1.A illustrates the neuron’s firing rate for a 120
second interval beginning 30 seconds before and ending 30 seconds
after distention of the duodenal portion of the intestine. The onset and
termination of distention are designated by the arrows. B shows that
the neuron is inhibited by gastric distention. C demonstrates the
response to electrical current stimulation of the subdiaphragmatic
vagus nerve. The latency between the stimulus artifact and the
neuronal response is 81 msec. The rate of conduction was 1.2 meters/
second. D-F show the morphology of the neuron as it would be seen in
the coronal (D), horizontal (El, and sagittal (F) planes. The border
between the DMNV and the NST is designated by the dashed line.
85
STRUCTURE-FUNCTION RELATIONSHIPS OF DMNV NEURONS
D
A
Duodenal distention
Release
c
4
\
LLateral
Scale: 12 seconds
L
Coronal plane
E
B
Gastric distention
Release
c
4
/+
r-
i
Lateral
Scale: 12 seconds
Horizontal plane
F
C
Stimulus
artifact
c
Neuronal
response
4
Dorsal
t
Caudal
Sagittal plane
200 um
Scale: 30 mseconds
Fig. 4. Response properties and morpholoa of a group 2 neuron. The conventions arc the same as those used in Figure 2.
This distention-sensitive neuron was located in the right
DMNV. The most striking aspect of this neuron's morphology is the orientation of its dendritic arbor. Figure 5E,F,
which presents horizontal and sagittal views of the cell,
respectively, demonstrates that the rostrocaudal extent of
the dendritic field is limited. In fact, the major portion of
the dendritic arbor is restricted to approximately 300 pm in
the rostrocaudal plane.
Group 4. This final group included 30 of the 52 neurons
that were characterized and labeled. All of these neurons
R. FOGEL ET AL.
86
D
A
Duodenal distention
Release
4
4
J
t,
Dorsal
t
Scale: 12 seconds
Lateral
Coronal plane
B
Gastric distention
Release
Scale: 12 seconds
t
Horizontal plane
F
C
Stimulus
artifact
Neuronal
response
4
4
I
Dorsal
L
Caudal
Sagittal plane
100 um
Scale: 30 mseconds
Fig. 5. Response properties and morphology of a group 3 neuron. The conventions are the same as those used in Figure 2
were inhibited by both gastric and duodenal distention.
Fourteen neurons were completely inhibited by both stimuli,
six neurons had a reduction in firing rate to both stimuli
but did not have complete cessation of firing, and ten
neurons were completely inhibited by one stimulus and
were partially inhibited by the other. There were several
STRUCTURE-FUNCTION RELATIONSHIPS OF DMNV NEURONS
patterns of partial inhibition. In some neurons, there was
complete inhibition at the onset of distention, other neurons had complete inhibition only after distention had been
present for 30 seconds, and some neurons had a slowing of
firing rate during the entire period of distention but did not
manifest complete inhibition. These patterns of partial
inhibition were seen with both gastric and duodenal distention.
One neuron that was inhibited by both stimuli is shown
in Figure 6. This DMNV neuron exhibited a spontaneous
activity rate of 1.0-1.2 Hz. Figure 6A,B illustrates that this
neuron’s basal activity was almost completely inhibited by
distention. The duodenal distention effect persisted for 20
seconds after the distention stimulus was discontinued.
Gastric distention caused complete inhibition of neuronal
firing. There was a burst of activity immediately after the
distention stimulus was released that was followed by a 20
second period during which there was no neuronal firing.
The latency for the response to electrical stimulation of the
subdiaphragmatic vagus nerve was 90 msec.
The DMNV neuron shown in Figure 6 was located in the
right DMNV. The neuron was multipolar. The dendrites
had a radial distribution around the soma and extended
into the NST.
Morphologic summary
A summary of the morphological data for each physiologically defined subset is shown in Table 1. Two groups of
neurons had features that distinguished them from the
neurons in each of the other groups. The neurons of group 2
(excited by duodenal distention and inhibited by gastric
distention) were the largest in our sample. These cells had
the largest cell somata, the most dendritic branch segments, and the greatest cell surface areas of any group.
Conversely, the neurons in group 3 (inhibited by duodenal
distention and excited by gastric distention) were the
smallest in our sample. These cells were characterized by
having the shortest dendritic lengths, the smallest surface
areas, and the smallest volumes of the four groups.
In addition to the results presented above, the neurons in
group 4 exhibited a number of features that permitted them
to be distinguished from the neurons in groups 2 and 3. The
neurons of group 4 had a significantly shorter total dendritic length and a smaller cell volume than neurons in
group 2. Group 4 neurons also had significantly more
dendritic branch segments than the neurons in group 3 .
One feature that all of our groups had in common was the
relative average extent in each of the three standard
anatomic axes. We found that the average neuron (regardless of group membership) evidenced a dendritic arbor that
was less extensive in the dorsoventral ( Y )plane than it was
in either the mediolateral (X) or the rostrocaudal (Z) planes.
Thus, the average DMNV neuron in our sample was most
widespread in the X axis and was most restricted in the Y
axis.
Table 2 presents a summary of the destinations of the
dendrites that were associated with the DMNV neurons in
the four physiologically defined groups. All neurons except
for one in group 4 had some dendritic terminations in the
ipsilateral DMNV. Thirteen neurons also had dendrites
that projected to the contralateral DMNV. The frequency of
this finding did not differ among the groups. Most neurons
projected to one or more of the subnuclei of the NST, but,
again, there were no features that distinguished any of the
four groups.
87
DISCUSSION
The present study utilized an intracellular recording and
labeling technique to examine potential morphological correlates for the response of DMNV neurons to distention of
the duodenum and stomach. In contrast to previous studies, in this series of experiments, we used recording electrodes filled with 2% Neurobiotin in 1 M KCl instead of the
6.0%. HRP in 0.05 M Tris buffer and 0.3 M KCl. For the
morphological studies, Neurobiotin is preferred to HRP,
because Neurobiotin produces enhanced filling of injected
neurons and labels neuron collaterals over a greater distance than HRP (Jacquin et al., 1992).
The use of Neurobiotin-filled electrodes did not compromise our electrophysiological recordings. The recorded action potentials had a high signal-to-noise ratio that was
similar to that seen with HRP-filled electrodes (compare to
analogous figures in Zhang et al., 1992). Although we did
not directly compare the characteristics of the two types of
electrodes, there is evidence to suggest that there is no
substantial difference resulting from the use of Neurobiotin. Studying trigeminal primary afferents, Jacquin and
colleagues showed that Neurobiotin-filled electrodes had
recording characteristics that were similar to HRP-filled
electrodes (Jacquin et al., 1992).
Using the single-cell recording and labeling technique, we
have shown that individual DMNV neurons can be placed
into one of four physiologically defined groups based on
their response to distention of the stomach and the duodenum. Furthermore, we have shown that some aspects of the
DMNV neuronal response are related to specific morphologic features. In particular, we demonstrated that neurons
excited by duodenal distention and inhibited by gastric
distention (group 2) were the largest in our sample. Conversely, the neurons that were inhibited by duodenal
distention and excited by gastric distention (group 3) were
the smallest in our sample.
There are several possible explanations for the functional
heterogeneity of vagal motor neurons affected by gastric
and duodenal distention. One possibility is that the neurons
were located in different regions of the DMNV, i.e., in
different “lattices”. We found that the neurons of the four
groups were interspersed in the medial caudal DMNV,
suggesting that a geographic segregation in the DMNV was
not the reason for the functional heterogeneity. Another
interpretation is that the different response patterns were
the result of different sensory inputs. Our protocol does not
allow u s to exclude this possibility, although we were able to
demonstrate that three of the four functional groups
(groups 2, 3, and 4)had similar “receptive fields” (sensitive
to distention of both the duodenum and the stomach).
Nevertheless, it is possible that one subset of DMNV
neurons (e.g., neurons in group 2 ) were innervated by NST
neurons that were excited by duodenal distention and
inhibited by gastric distention, whereas another subset of
DMNV neurons (e.g., group 3 ) received inputs from NST
neurons that had the opposite response to the stimuli. Even
if the four groups of DMNV neurons received different
inputs from the NST (and/or vagal afferents), we must
consider the fact that these functional groups exhibited
distinct morphological features. Could differences in somadendritic morphology be responsible for the different response profiles? There are data that suggest that individual
dendrites may function as distinct channels or integrators
of afferent input (Mifflin, 1993).We did observe statistically
88
R. FOGEL ET AL.
A
Release
Duodenal distention
4
4
Dorsal
t
Lateral
Coronal plane
I
Scale: 12 seconds
E
B
Gastric distention
Release
4
4
-I
I
1
Scale: 12 seconds
F
C
Stimulus
artifact
Neuronal
response
4
4
Lateral
Horizontal plane
Sagittal plane
L
Scale: 30 mseconds
100 urn
Fig. 6 . Response properties and morphology of a group 4 neuron. The conventions are the same as those used in Figure 2
STRUCTURE-FUNCTION RELATIONSHIPS OF DMNV NEURONS
89
TABLE 1. Summary of Morphological Features of DMNV Neurons]
Group I
(n = 8)
Group 2
( n = 71
Group 3
( n = 71
Group 4
( n = 301
026.7 ? 41.6
1.3
20.1
0.57 i 0.05
4,360.4 i 333.7
143.1 I 24.1
35.25 2 4.3
486.4 i 57.82
24 6 i I .4
0.68 i 0.06
5,422.R i 629.5:’
108.5 ? 15.5
53.7 i 6 . 3 2
6.7 i 0 7
19.753.2 ? 1.550.4’
9,163 2 i 1,243 8’
691 6 ? 54.9
409.1 i 60.5
507.0 ? 25.7
271.9 ? 31 4
18.4 i 1.2
0.75 i 0.04
2.7no.x 2 327.8’
109.8 t 19.7
25.4 i 3.3”
5.3 2 0.6
8,291.7 i 1,164.92
2.886.7 i 416.9‘
416.0 i 72.1
2x9 4 i 37.0
931.2 i 36.3
320.0 L 21.9
19.9 t 0.7
Feature
Soma area
Soma diameter
Form lactor
I.ength
Mean len@h
No. segments
Branch order
Surij,, a r r a
Vrrlumt,
Dendritic X plane
Vendritic Y plane
Dendrmc Z plane
6.98
14.091.2
6,403.8
615 2
327.4
499.5
i
0.9
i 1.978.8
i
?
1.450.0
31 7
i 31.8
2 52.8
0.69 i 0.03
X9R7 5 i 182.7’
in9 3 2 5.4
38.5 i 2. I S
6.3 i 0 . 3
13.708.0 i 638.8
5.383.0 2 367.6:’
553.7 I30.0
364 6 i 18.9
425.6 i 23.8
lI,enb<t>15 expressed in k m , arra in k m 2 ,a n d v o l u m ~in p m ” . DMNV, dorsal motor nucleus o f t h e v a ~ w s
‘Significant diflerence from values of all o t h e r groups u s i n g ‘ h k e y B test.
Significant difference in comparisons hetween groups 2 a n d 4
‘Srgn:nlficantd i f k r r n c e hetwren Groups 3 a n d 4
‘TABLE 2. Dendritic Destination of DMNV Neurons’
Neuron
Ipsilateral D M h J
Contralateral DMhT
Medial n . NSI‘
Suhpostrt.mal n . XSI‘
(;elatinou-: n NST
Central canal 1V ventricle
Group 1
Group 2
8
2
8
2
1
4
Group 3
Group 4
7
7
29
1
7
2
8
6
27
1
3
0
9
2
1
2
9
2
‘Numbers tndicatt, t h e n u m b e r of neurons with dendrites projecting to t h a t region. T h e r e
W P ~ Pn o statistically different distrihutians amonlr t h e distributions o f t h e various LmlUDS.
NST. nucleus rifthe solitary tract
significant differences in dendritic length, number of segments, surface area, and volume in neurons with different
response patterns. All of the DMNV neurons that responded to both gastric and duodenal distention (groups 2,
3, and 4)could be distinguished by differences in somadendritic morphology. Although the precise nature of this
relationship is not entirely clear, one potential relationship
may involve cell size and the nature of the response to
duodenal distention. Specifically, we found that the neurons that were inhibited by duodenal distention (groups 3
and 4 ) were also the smallest neurons in the data set.
Given the possible link between the nature of the response to duodenal distention and cell size, it is reasonable
to consider potential morphologic substrates for such a
relationship. One of the features associated with the smaller
cells was shorter dendrites. Although it is difficult to predict
how shorter dendrites might be associated with inhibitory
input from the duodenum, evidence obtained in the gustatory system might provide a clue. Whitehead (1993) has
recently shown that excitatory synapses on taste-sensitive
neurons in the NST tend to be localized on the distal aspect
of the dendrites, whereas the predominant synapse on the
proximal dendrite is inhibitory. If such a differential distribution of excitatory and inhibitory synapses also existed in
the DMNV, then this might influence a given DMNV
neuron’s response to gastrointestinal stimuli. In theory,
neurons with longer dendrites might have a n increased
opportunity to receive excitatory inputs. If this postulate is
valid, however, then it is difficult to explain the fact that the
neurons in group 4,which were inhibited by both stimuli,
were not the smallest neurons in the data set. Instead, this
property was associated with the neurons of group 3, which
were inhibited by one stimulus and excited by another.
Another feature of the dendritic architecture that might
be predicted to influence a given DMNV neuron’s response
properties would be the location of the dendritic terminations. There is substantial evidence that the sensory inputs
to the NST are segregated according to their peripheral
target (Altschuler et al., 1992; Powley et al., 1992). It is also
well established that DMNV neurons extend dendrites into
the overlying NST (Zhang et al., 1992): Therefore, it is
conceivable that vagal motor neurons that are responsive to
a gwen stimulus would extend dendrites into the region of
the NST that received the appropriate sensory input.
Surprisingly, we did not obtain any evidence to support this
postulate. We were able to follow individual dendrites to
their termination in the medial, subpostremal, and gelatinous subnuclei of the NST, but we found no significant
relationship between the response properties of the neurons and the destinations of their dendrites.
We observed that the soma area of group 2 neurons was
significantly greater than that of any of the other groups.
Although soma size has been a variable used to classify
DMNV neurons (McLean and Hopkins, 1981, 1982, 1985;
Laiwand et al., 19871, the relationship between soma size
and neuronal function has not been explored directly. I n
the cat, neurons with larger soma size have more and larger
dendritic processes (McLean and Hopkins, 1981). Different
roles for small and medium-sized neurons have been suggested, because medium-sized neurons projected into the
vagus nerve, whereas small neurons were either interneurons or had ascending projections (McLean and Hopkins,
1982; Laiwand et al., 1987). At present, we do not know the
significance of the variation in soma area with regard to the
response of DMNV neurons to mechanoreceptor excitation.
Although many elements of the relationship(s1 between
structure and function in the DMNV remain to be elucidated, our data does strongly support the contention that
the neurons in this nucleus can be divided into distinct
morphologic groups. Somewhat surprisingly, this is precisely the opposite of the conclusion recently proffered by
Fox and Powley (Fox and Powley, 1992). Fox and Powley
used intracellular injections in a slice preparation to explore
the possibility that the neurons in the different DMNV
“columns” might express unique morphologies, thus providing a structural basis for the different functions that have
been attributed to these columns. These investigators
concluded that “the most parsimonious interpretation of
the data is that DMNV neurons are variants of a single
prototype.” Interestingly, however, they did observe statistically significant differences in a number of the morphological features associated with the neurons in the different
columns. Neurons in the celiac column had larger cell soma
areas than neurons in the other two columns. The celiac
column neurons also had the most extensive dendritic
90
R. FOGEL ET AL.
arbors, the greatest number of medially oriented primary but, rather, we suggest that our data supports the hypothdendrites, and the longest dendritic lengths. Finally, he- esis of heterogeneity of vagal motor neurons.
patic column cells had fewer dorsally directed dendrites
The results of this study support our previous finding
than the neurons of the other two columns. It is difficult to that gastric or intestinal distention inhibits the firing rate
judge how these results might impact on the results of most DMNV neurons (Zhang et al., 1992; Chu et al.,
obtained in our present investigation, because our sample 1993). However, we cannot determine whether the neurons
of labeled cells is restricted to the middle and medial aspect responsive to jejunaliileal distention belong to any of the
of the DMNV (where the majority of the duodenum- and groups recognized in this study. Our results suggest that, in
gastric-sensitive cells reside caudal to the obex). We would order to classify the DMNV neurons affected by mechanoresuggest, however, that Fox and Powley may have been too ceptor stimulation, it is necessary to study the effect
conservative in their final analysis. Their caution is cer- resulting from distention of several different sections of the
tainly understandable, because they were limited by the gastrointestinal tract. We cannot combine the results from
fact that they could only label that portion of any cell that the two papers, because different regions of the gastrointeswas contained in one 100 pm slice. Because our data tinal tract were studied. Moreover, we cannot compare the
indicates that the average DMNV neuron extends approxi- morphology of the neurons because of differences in the
mately 400-500 km in the rostrocaudal plane, it is likely technique of reconstruction. The previous papers used
that Fox and Powley lost a large portion of the dendritic camera lucida reconstruction, whereas our current study
arbor of many neurons. The fact that these investigators employed three-dimensional reconstructions and computerfound morphologic differences despite this limitation sug- derived measurements of dendritic morphology. The morgests that this line of investigation deserves further scru- phology of a neuron in a single plane of section is insufficient for the determination of the quantitative morphological
tiny.
Most previous investigations of the vagal efferent re- measurements that we used to classify motor neurons.
In summary, we conclude that the vagal motor neurons
sponse to gastric distention have shown that most vagal
neurons are excited by this stimulus (Davison and Grundy, innervating the stomachiduodenum are functionally and
1978; Grundy et al., 1981; Blackshaw et al., 1987; Black- morphologically heterogeneous. Our results suggest an
shaw and Grundy, 1989). In the present investigation, association between the response properties and morpholhowever, we recorded from an unexpectedly large number ogy of vagal motor neurons. Two of the most important
of neurons that were inhibited by gastric distention. Al- morphologic features appear to be soma size and dendritic
though the reason for this apparent discrepancy is unclear, architecture. The mechanism(s) by which soma-dendritic
one possible explanation involves potential species differ- architecture influences the electrophysiological response of
ences. With the exception of the recording study performed DMNV neurons remains to be elucidated. We are currently
by Davison and Grundy (19781,virtually all prior investiga- exploring the possibility that these differences in somators used the ferret as their animal model. It is possible that dendritic morphology may influence a neuron’s ability to
the ferret’s response to gastric distention may differ from receive excitatory vs. inhibitory inputs. In particular, we
that of the rat. Another factor that should be considered are interested in possible associations between the neurorelates to the recording technique. We recorded from nal response, morphology, and sensitivity to excitatory and
individual neurons in the brainstem, whereas most previ- inhibitory neurotransmitters. It is hoped that these investious investigations have recorded from teased fiber prepara- gations will not only provide important information regardtions. Depending on the protocol being employed, such ing the organization of the DMNV but that they may also
teased fiber preparations can cause the investigator to contribute to the clinical management of gastrointestinal
mistakenly identify an afferent fiber (which may indeed be diseases.
excited by the stimulus) as efferent. Our data would suggest
that this is particularly problematic if the investigator is
ACKNOWLEDGMENTS
relying on conduction velocities to differentiate afferents
from efferents. The mean latency for the DMNV response
The authors thank D. Randall for technical assistance as
to electrical current stimulation of the vagus nerve in the well as Edward Peterson Ph.D. and J. Massey for help with
present study was 102.7 2.2 msec, yielding a conduction the statistical analyses. This research was supported by
velocity of approximately 1meterisecond. Thus, the major- NIH grant NS30083 (R.F.)
ity of the axons associated with the DMNV neurons we
recorded was probably unmyelinated C fibers.
Although we have demonstrated that DMNV neurons
LITERATURE CITED
innervating the stomachiduodenum can be categorized into Altschuler, S.M., D.A. Ferenci, R.B. Lynn, and R.R. Miselis (1991) Represenfour groups, we do not wish to imply that there are only
tation of the cecum in the lateral dorsal motor nucleus of the vagus nerve
four possible subsets of DMNV neurons. Our sample size
and commissural subnucleus of the nucleus tractus solitarii in rat. J.
Comp. Neurol. 304:261-274.
was not sufficient to permit us to determine whether
neurons in groups 1-3 could be subclassified. Additionally, Altschuler, S.M., L. Rinaman, and R.R. Miselis (1992) Viscerotopic representation of the alimentary tract in the dorsal and ventral vagal complexes
the decision to study DMNV neurons with cell soma in the
in the rat. In S. Ritter, R.C. Ritter, and C.D. Barnes (eds.): Neuromedial DMNV in the region caudal to the obex also has
anatomy and Physiology of Abdominal Vagal Afferents. Boca Raton, FL:
implications for the possible number of subsets. By not
CRC Press, pp. 22-53.
studying the lateral and rostral DMNV, we may miss Blackshaw, L.A., and D. Grundy (1989) Responses of vagal efferent fibers to
stimulation of gastric mechano- and chemoreceptors in the anaesthesubsets of vagal motor neurons with different either morphotized ferret. J. Autonom. New. Syst. 27:39-45.
logical or functional properties. Moreover, more rostral
L.A., D. Grundy, and T. Scratcherd (1987) Involvement of
neurons may be more likely to send dendrites to the Blackshaw,
gastrointestinal mechano- and intestinal chemoreceptors in vagal regelatinous subnucleus of the NST. Therefore, we do not
flexes: An electrophysiological study. J. Autonom. Nerv. Syst. 18.225conclude that there are onlv four sets of DMNV neurons.
284.
*
STRUCTURE-FUNCTION RELATIONSHIPS OF DMNV NEURONS
Bueno, L., and J. Fioramonti 119911 CNS peptidergic regulation of gut
motility. In Y. Tache and D. Wingate ieds.1: Brain-Gut Interactions. Boca
Raton, FL: CRC Press, pp. 231-243.
Capowski. J.J. 119891 Computer Techniques in Neuroanatomy. New York:
Plenum Press.
Chu. X.. X. Zhang, W.E. Renehan, W.H. Beierwaltes, and R. Fogel 119931
Intestinal distention alters vagal efferent activity and small intestinal
transport in vivo. Am. J . Physiol. 264:G341-G350.
Davjson, J.S., and D. Grundy (1978)Modulation of single vagal efferent fiber
discharge by gastrointestinal afferents in the rat. J. Physiol. (London)
284:69-82
Elfvin. L.G., and B. Lindh 11982, A study of the extrinsic innervation o f t h e
guinea-pig pylorus with the horseradish peroxidase tracing technique. J.
Comp. Neurol. 20#:317-324.
Engel. A.K.. and G.W. Kreutzberg 119881 Neuronal surface changes in the
dorsal motor nucleus of the guinea pig in response to axotomy. J. Comp.
Neurol. 275:181-200.
Fogel, R.. X. Zhang, and W. Renehan 119911 Central nervous system
influences on intestinal absorption. In Y. Tache and D. Wingate ieds.):
Brain-Gut Interactions. Boca Raton, FL: CRC Press, pp. 207-216.
Fox. E.A.. and T.L. Powley (1992) Morphology of identified preganglionic
neurons in the dorsal motor nucleus of the vagus. J . Comp. Neurol.
32279-98.
Grundy. U..A.A. Salih, and T . Scratcherd !19811 Modulation of vagal
efferent fiber discharge by mechanoreceptors in the stomach, duodenum
and colon of the ferret. J . Physiol. (London)319:43-52.
Huany, X.-F., I. Tork, and G. Paxinos (1993) Dorsal motor nucleus of the
vakws nerve: A cyto- and chemoarchitectonic study in the human. J.
C h mp . Neu rol . 330: 158- 182.
.Jacquin, M.F.,J.W. Hu, B.J.Sessle, W.E. Renehan,andP.M.E. Waite(1992)
Intra-axonal Neurobiotin'rMinjection rapidly stains the long-range projections of identified trigeminal primary afferents in vivo: Comparisons
with HRP and PHA-L. J . Neurosci. Methods 45:71-86.
J o u . M.-J.. C.-Y. Wen, and J.-Y. Shieh 119931 Localization of the stomachprojecting neurons in the dorsal motor nucleus of the vagus nerve in the
guinea pig. J . Autonom. New. Syst. 43201-208.
Laiwand, R., K.Werman, and Y. Yarom 11987) Time course and distribution
of motoneuronal loss in the dorsal motor vagal nucleus of guinea pig
after cervical vagotomy. J. Comp. Neurol. 256:527-537.
Laughton. W.B., and T.L. Powley (19871 Localization of efferent function in
the dorsal motor nucleus o f t h e vagus. Am. J. Physiol. 252:R13-R25.
Li, Y , and C:. Owyang i 19931 Vagal afferent pathway mediates physiological
action ofcholecystokinin on pancreatic enzyme secretion. J. Clin. Invest.
9 2 4 18-424.
McLean. J.H., and D.A. Hopkins 11981) A light and electron microscopic
study of the dorsal motor nucleus of the vagus in the cat. J . Comp.
Neurol. 1.95157-175.
91
McLean, J.H., and D.A. Hopkins ( 19821 Ultrastructural identification of
labeled neurons in t h e dorsal motor nucleus o f t h e vagus nerve following
in.jections of horseradish peroxidase into the vagus nerve and brainstem.
J . Comp. Neurol. 206:243-252.
McLean, J . H . , and D.A. Hopkins ! 19851 Ultrastructure of the dorsal motor
nucleus of the vagus nerve in monkey with a comparison of synaptolop
in monkey and cat. J . Comp. Neurol. 231:162-174.
Mifflin, S.W. (19931 Laryngeal afferent inputs to the nucleus of the solitarv
tract. Am. J. Physiol. 26.5:K269-R276.
Overdijk, J., H.B.M. Uylings, K. Kuypers, and A.M. KamStrd 11978) An
economical, semi-automatic system for measuring cellular tree structures in three dimensions, with special emphasis on Golpimpregnated
neurons. J. Microsc. 114:271-284.
Pagani, F.D., W.P. Norman, D.K. Kasbekar, and R.A. Gillis (1985)Localization of sites within dorsal motor nucleus of vagus that affect gastric
motility. Am. J. Physiol. 249:G73-G84.
Powley, T.L., H.-R. Berthoud. E.A. Fox, and W. Laughton 11992) The dorsal
vagal complex forms a sensory-motor lattice: The circuitry ofgastrointestinal reflexes. In S. Ritter, R.C. Ritter, and C.D. Barnes !eds.): Neuroanatomy and Physiology of Abdominal Vagal Afferents. Boca Raton, FL:
CKC Press, pp. 57-79.
Prechtl, J.C., and T.1,. Powley (1990)The fiber composition o f t h e abdominal
vagusofthe rat. Anat. Embryol. 1#1:101-115.
Rinaman, I+ and R.R. Miselis (19871 The organization of vagal innervation
of rat pancreas using cholera toxin-horseradish peroxidase conjugate. J .
Autonom. New. Syst. 21r109-125.
Roze, C. (1989) Central regulation of pancreatic secretion. In Y. Tachc and
D.L.Wingate 1eds.l: Brain-Gut Interactions. Boca Raton, FL: CRC Press.
pp. 187-198.
Smolen, A.J., and K.C. 'I'ruex (19771 The dorsal motor nucleus of the vagus
nerve of the cat: Localization of preganglionic neurons by quantitative
histological methods. Anat. Rec. 189555-566.
Tache, Y.. H. Yang, and K. Yanagisawa (19911 Brain regulation of gastric
acid secretion by neuropeptides. In Y. Tache and D. Wingate (eds.):
Brain-Gut Interactions. Boca Raton, FL: CHC Press, pp. 169-186.
Uylings, H.B.M.. A. Kuiz-Marcus, and J . van Pelt 119861The metric analysis
of three-dimensional dendritic tree patterns: A methodological review. J.
Neurosci. Methods 1#:127-151.
Whitehead, M.C. (19931Distribution of synapses on identified cell types in a
gustatory subdivision of the nucleus of the solitary tract. J . Comp.
Neurol. 332:326-340.
Zhang, X., K. Fogel, and W.E. Renehan 119921Physioloby and morphologyof
neurons in t h e dorsal motor nucleus of the vagus and the nucleus o f t h e
solitary tract that are sensitive to distention of the small intestine. J.
Comp. Neurol. 923432-448.
Документ
Категория
Без категории
Просмотров
5
Размер файла
1 372 Кб
Теги
morphology, motor, intestinal, dorsal, distention, neurons, gastric, function, sensitive, vagus, relationships, nucleus
1/--страниц
Пожаловаться на содержимое документа