Vous êtes sur la page 1sur 10

BRAIN

RESEARCH
ELSEVIER Brain Research 647 (1994) 255-264

Research Report

Brain areas essential or non-essential for emesis


Alan D. Miller a,,, Satoshi Nonaka b, Jan Jaku~ c
The Rockefeller Unieersity, 1230 York Ave, New York, NY 10021, USA
h Department of Otolaryngology, Asahikawa Medical School, Asahikawa, Japan
f Department of Physiology, Faculty of Medicine, Comenius University, Martin, Slovak Republic
Accepted 22 February 1994

Abstract

This study was undertaken to better delineate those brain regions that are either essential or non-essential for vomiting.
Fictive vomiting, identified by a characteristic pattern of respiratory nerve discharge, was induced by a combination of emetic
drugs and electrical stimulation of abdominal vagal afferents in decerebrate, paralyzed cats. Regions non-essential for
coordinating vomiting included the entire cerebellum, structures rostral to the medullary retrofacial nucleus, and spinal cord.
Fictive coughing was also elicited following cerebellar removal but was not studied after other procedures. The respiratory-re-
lated components of fictive vomiting were abolished by large lesions or kainic acid injections in the lateral medulla at the level of
the retrofacial nucleus, where respiratory pre-motor and motor neurons are known to exist. Electrical stimulation of this region
of the brainstem failed to evoke vomiting. The results of the present study are consistent with our previous electrical stimulation
[Brain Res., 270 (1983) 154-158] and c-los [J. Neurosci., 14 (1994) 871-888] studies and the hypothesis that emesis is coordinated
not by a unique, well-defined 'vomiting center' but rather by a distributed control system located in the medulla between the
levels of the obex and the retrofacial nucleus.

Key words: Vomiting; Central pattern generator; Retrofacial nucleus; Cerebellum; Brainstem

I. Introduction tributed control systems [10,30,34,36]. It has been pro-


posed, for example, that vomiting may be produced by
Nausea and vomiting (emesis) can be serious condi- the sequential activation of a series of effector nuclei,
tions that occur frequently due to a variety of causes, as opposed to these motor nuclei being activated in
including cancer chemo- and radiation therapy, surgery, parallel by a vomiting center [10]. It has also been
pregnancy, motion and space sickness, and various suggested that a paraventricular system of nuclei, de-
diseases [4,11,27,42]. Although in some cases the pre- fined by their connections with the area postrema and
dominant input pathways by which vomiting is evoked each other, can collectively account for most of the
have been identified, the neuronal substrate underlying p h e n o m e n a associated with nausea and vomiting [30].
the coordination of the vomiting response remains The purpose of the present study was to better
incompletely understood. Previously proposed organi- delineate those regions of the brain that are either
zational schemes have included a 'vomiting center' essential or non-essential for vomiting, using a combi-
located in the medullary dorsolateral reticular forma- nation of different lesioning techniques. Experiments
tion [8,48], a 'pattern generator' in the B6tzinger com- were carried out in cats that were decerebrated, para-
plex in the vicinity of the medullary retrofacial nucleus lyzed, and artificially ventilated. The occurrence of the
[12,13,26], and, in contrast to the above, more dis- act of vomiting in these paralyzed animals (fictive vom-
iting) was identified by a characteristic pattern of a
series of bursts of co-activation of the nerves to the
diaphragm (phrenic nerve) and abdominal muscles that
* Corresponding author. Fax: (1) (212) 327-8530. would produce retching and expulsion in non-para-
E-mail: millera@rockvax.rockefeller.edu or millera(a rockvax.bitnet lyzed animals [35].

0006-8993/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved


SSDI 0 0 0 6 - 8 9 9 3 ( 9 4 ) 0 0 2 4 3 - 6
256 A.D. Miller et al. /Brain Research 647 (1994) 255-264

2. Materials and methods Fictive vomiting was produced using a combination of two proce-
dures. Abdominal vagal afferents were electrically stimulated (con-
2.1. General experimental procedures tinuous trains of 0.8 ms pulses, 25 Hz, 0.5-2 mA) in the lower thorax
just rostral to the diaphragm. In addition, animals were administered
a combination of emetic drugs (Sigma), typically lobeline sulfate (1
Data were obtained from 32 adult cats of either sex. The animals
m g / k g i.v.) and naloxone hydrochloride (3 m g / k g i.v.). Sometimes,
were anesthetized with halothane (Fluothane; Ayerst)/nitrous oxide
apomorphine hydrochloride (1 m g / k g ) and protoveratrine (5-10
and decerebrated at the mid-collicular level after bilateral ligation of
# g / k g ) were added to the injection mixture. Fictive coughing [5,16]
the carotid arteries. All animals except one were paralyzed and
was elicited in a few experiments by mechanical tracheal-bronchial
artificially ventilated. Anesthesia was discontinued following the
stimulation or by electrical stimulation of the superior laryngeal
completion of all surgical procedures and at least 1 h prior to data
nerve (0.2 ms pulses, 2-5 Hz, 80-250 #A).
collection.
The trachea was intubated, and cannulae placed in the femoral
artery to monitor blood pressure and in the femoral veins for drug 2.2. Lesions of the brain
administration. The animals were placed in a stereotaxic frame and
supported using hip pins. The C5 phrenic and L1 abdominal muscle Four types of lesions were carried out. First, the entire cerebel-
nerves were cut and their central ends inserted into bipolar cuff lum was removed by aspiration. Second, complete or partial transec-
recording electrodes. In some experiments, the activity of the whole tions of the brainstem were performed using a brass spatula mounted
recurrent laryngeal nerve was also recorded. Rectal temperature was in a stereotaxic manipulator. Third, electrolytic lesions were made in
kept at 36-37°C using a heating lamp. The animals were paralyzed the region of the retrofacial nucleus by passing current through
with gallamine triethiodide (Sigma; initial injection of 10 m g / k g i.v., glass-insulated, tungsten microelectrodes. Finally, multiple 0.1-0.3
maintained by hourly injections of 5 m g / k g ) and artificially venti- #1 injections of kainic acid (Sigma; 2 m g / m l ) dissolved in phosphate
lated with room air (24 cycles/min). A small expiratory load (l cm buffered saline (pH 7.4) were made in the brainstem using a 1-5 ~1
H 2 0 ) was applied. End-tidal CO2 was kept between 4 - 6 % . Mean Hamilton syringe (26-33 gauge needle) mounted in a stereotaxic
blood pressure was maintained above 90 mm Hg, if necessary using manipulator. The injection solution was saturated with Fast green
i.v. infusion of metaraminol bitartrate (Aramine; Merck, Sharp & dye as a guide for estimating the spread of the solution. Since kainic
Dohme). Nerve activities were amplified, rectified, low- and high-pass acid produces functional inactivation of cell bodies within 30 min
filtered with a 10 ms time constant, and recorded on-line at a while sparing fibers of passage [9], we waited 30 min before testing
sampling rate of 100 Hz using a PDP 11/73 computer (Digital the effects of the injections and then continued testing for another
Equipment Corp.). 1-1.5 h. In addition to using stereotaxic coordinates as a guide for

CONTROL CEREBELLUM REMOVED


A. Respirotion
PHR , , ,

,.,,J', ~-" ,, t./ ~'. ~/' ~..2'.. ,:> ~.~.:


ABD

, r ?li
B. Fictive Vomiting
PHR

ABD

C. D. Fictive Coughing

ABD __ :v. . . . . . . . ",~ ~___

5 sec

Fig. 1. Phrenic (PHR) and abdominal (ABD) nerve discharges during respiration (A), fictive vomiting (B), and fictive coughing (D) after removal
of the entire cerebellum. Data in A, B, and D were obtained from different animals. Each nerve recording is plotted at the same gain pre- vs.
post-lesion in this and following figures. Phrenic discharge in B shows artifact from vagus nerve stimulation used to evoke fictive vomiting. The
end of each fictive vomiting episode is terminated by an expulsion phase, which is characterized by prolonged abdominal discharge [35]. Fictive
coughing was elicited by mechanical tracheal-bronchial stimulation. Histology on the right in C is from the animal with a cerebellar removal in A.
Histology on the left shows location of the infracerebellar nucleus (IFC) in a different (control) animal. P, pyramidal tract; PH, nucleus
prepositus hypoglossi; RB, restiform body; SA, stria acustica; VIN, inferior vestibular nucleus; VLD, lateral vestibular nucleus, dorsal division;
VMN, medial vestibular nucleus; 7, facial nucleus.
A.D. Miller et al. / Brain Research 647 (1994) 255-264 257

placing lesions, some lesions were made in relation to the locations coughing. Data were obtained during respiration and
of spontaneous respiratory-related neurons recorded extracellularly fictive vomiting both before and after cerebellar re-
in the ventrolateral medulla or in relation to antidromic field poten-
moval in five cats, examples from which are shown in
tials recorded in the facial nucleus.
Fig. 1A,B, respectively. Fictive coughing was also
2.3. Electrical stimulation of the brainstem elicited following cerebellar removal in three animals.
Examples of coughs from one animal are shown in Fig.
Electrical stimulation was performed in the region of the retrofa- 1D. Because of recent reports that destruction of the
cial nucleus in an attempt to elicit vomiting (in one non-paralyzed infracerebellar nucleus causes an irreversible loss of
cat) or fictive vomiting (in three paralyzed animals). We usually used spinal expiratory activity [21,22], we were careful to
the stimulus parameters (10 Hz, 0.2 mA, 0.5 ms pulse trains) that
include that nucleus in our cerebellar lesions (Fig. 1C).
were reported by Fukuda and Koga [12] to produce fictive vomiting
in dogs. We also sometimes tried additional stimulus parameters In spite of the removal of the infracerebellar nucleus,
(e.g. 50 Hz, 20-30 p,A). Stimuli were applied using glass-insulated, abdominal discharge continued during expiration, tic-
tungsten microelectrodes at 0.5 m m depth intervals along the elec- tive vomiting, and fictive coughing.
trode track. Small electrolytic lesions were made to mark the loca-
tions of some stimulating sites.
3.2. Removal of the pons and rostral medulla
2.4. Histological reconstructions
The pons and rostral medulla are not essential for
At the end of experiments, animals were euthanized by an producing vomiting, as was determined by making a
overdose of sodium pentobarbital (120 m g / k g i.v.). Neural tissue was series of progressive rostral-to-caudal brainstem tran-
fixed in 10% formalin, sectioned at 100/zm in the frontal plane, and
sections in nine cats. Fictive vomiting, as measured by
stained with thionin. In animals injected with kainic acid, alternate
histological sections were processed with and without counterstain- phrenic and abdominal nerve discharge, was abolished
ing in order to better visualize the extent of spread of green dye. when the transections were near the level of stereo-
Stimulating sites were reconstructed from electrode coordinates in taxic plane P9 and more caudal to it. The level P9
relation to the locations of electrolytic lesions made at some sites. corresponds to about 5-5.5 mm rostral to the obex at
The locations of lesions, kainic acid injections, and stimulating sites
the level of the retrofacial nucleus. An example of
were reconstructed on histological sections with reference to the
atlas of Berman [3]. Stereotaxic planes refer to m m posterior (P) to multiple transections made in one animal is illustrated
the interaural plane. in Fig. 2. Level B in Fig. 2 corresponds to stereotaxic
plane P4.5 (level of trigeminal motor nucleus), level C
to P8 (level of facial and lateral vestibular nuclei), and
3. Results level D to P l l (near caudal end of retrofacial nucleus).
Lesions that abolished fictive vomiting also typically
3.1. Cerebellar removal abolished normal spontaneous respiratory discharge.
However, following lesions in some animals, weak tonic
Removal of the entire cerebellum did not prevent co-activation of the phrenic and abdominal nerves could
the occurrence of phrenic and abdominal nerve dis- sometimes be obtained in response to emetic stimuli
charge during respiration, fictive vomiting, or fictive (e.g. Fig. 2D). Similarly, phasic respiratory discharge

A. CONTROL B.
I
~ ~ ~JI ~i~ ~

ABD
B _

C-

D-

\ /?
t i i i I i
5 Se¢

Fig. 2. Effects of transverse sections made at progressively more caudal levels of the brainstem. Fictive vomiting was not abolished until the lesion
m a d e at level D. Level B corresponds to stereotaxic plane P4.5, level C to P8, and level D to P l l .
258 A.D. Miller et aL ~Brain Research 647 (1994) 255-264

occasionally appeared following the lesion illustrated fects of large knife cuts or kainic acid injections, more
in Fig. 2D. limited damage to the retrofacial nucleus and adjacent
tissue caused by electrolytic lesions in five cats or small
3.3. Lesions at the level of the retrofacial nucleus kainic acid injections in another failed to abolish either
fictive vomiting or respiration. Examples of the regions
Fictive vomiting was abolished by partial transec- effected in four of these animals are illustrated in Fig.
tions encompassing the lateral part of the brainstem 4A, B.
bilaterally at the level of the retrofacial nucleus in four
cats. Data from one animal are shown in Fig. 3A. 3.4. Electrical stimulation of the retrofacial region
Large bilateral kainic acid injections in the same region
in two cats also eliminated fictive vomiting (Fig. 3B), Electrical microstimulation of the region of the
demonstrating that dysfunction of cell bodies (as op- retrofacial nucleus was carried out in four cats (three
posed to axons) in this region is sufficient to abolish paralyzed, one non-paralyzed) in an attempt to elicit
the response. Normal spontaneous respiratory dis- (fictive) vomiting. A total of 488 stimulation sites in 61
charge was also abolished in four of these six cats and vertical electrodes tracks were tested in the region
modified in the other two (post-lesion amplitude de- between 1-7 mm rostral to the obex, 3-5.5 mm lateral
creased by 63% in one, frequency increased by 2.5 to the midline, and 1-6 mm beneath the dorsal
times following kainic acid injections in the other). medullary surface. An example of the stimulating loca-
Gasping [44] was sometimes observed following lesions tions at one brainstem level is shown from one animal
in two animals, including the example shown in Fig. in Fig. 4C. Neither vomiting nor fictive vomiting could
3A. be elicited by stimulation in any animal. However,
Attempts were made in six cats to further delineate several changes in the normal respiratory pattern were
the critical region that when destroyed, resulted in the observed, including changes in respiratory rate and
elimination of fictive vomiting. In contrast to the ef- depth, and the appearance of cough- or sneeze-like
responses. At the end of the electrical stimulation
experiments, administration of emetic drugs produced
vomiting or fictive vomiting in these animals.
CONTROL
A. PHR B. 3.5. Kainic acid injections in the dorsolateral reticular
formation

ABD In an attempt to produce dysfunction of the brain-


stem region that Borison and Wang [8,48] thought
.... t L~
contained a 'vomiting center', kainic acid injections
were made into the dorsolateral reticular formation
POST LESION from about 1 mm caudal to the obex to about 4 mm
PHR rostral to it in two cats. The injections failed to abolish
fictive vomiting, although no clear expulsion phase was
ABD observed after the injections. The expulsion phase at
the end of a vomiting episode can be characterized by
abdominal discharge that is prolonged with respect
both to ongoing phrenic discharge and to abdominal
discharge during the preceding retching phase [35].
Data from both animals are illustrated in Fig. 5.

3. 6. Transection of the caudal medulla

In order to demonstrate that the neuronal circuitry


Fig. 3. Fictive vomiting was abolished by bilateral knife cuts (A) or
that generates vomiting is located within the medulla,
kainic acid injections (B) in the lateral portion of the brainstem at data were obtained from one animal both before and
the level of the retrofacial nucleus (RFN). The extent of the lesion is after a complete transverse section at the level 4 mm
indicated in black in A, and in following figures. The hatched area in caudal to the obex. The lesion abolished phrenic and
B (and in following figures) represents the extent of spread of green abdominal nerve discharge due to transection of bul-
dye that was dissolved in the kainic acid solution. The lesion in A
bospinal pathways; however, the recurrent laryngeal
extended about 0.8 m m in the rostro-caudal direction, while in B the
green dye extended caudally about 2 m m from the illustrated plane nerve still displayed a characteristic pattern of dis-
and rostrally about 4 mm (i.e. from about stereotaxic level P l l to P5. charge [14,25] similar to that observed during fictive
A.D. Miller et aL /'Brain Research 647 (1994) 255-264 259

m. B.

C •

F" ~ I I 1
2ram

Fig. 4. Examples of attempts to delineate the critical region for fictive vomiting at the level of the retrofacial nucleus (RFN). Fictive vomiting was
not abolished by bilateral electrolytic lesions in the 3 animals shown in (A) or by small kainic acid injections in another (B); nor could fictive
vomiting be evoked by electrical stimulation at the sites indicated by horizontal lines on the vertical electrode tracks in C. The effective current
spread would have been in excess of 1 m m from the stimulating sites using a 0.2 m A stimulus [18]. T h e black dot in C dorsal to the R F N
represents the site of a small electrolytic lesion used to mark the location of the stimulating electrode. IO, inferior olive; other abbreviations as in
Fig. 1.

vomiting prior to the transection (Fig. 6). In addition, gions that were shown to be non-essential for coordi-
after the transection, rhythmic fictive swallowing, which nating vomiting.
typically occurs following an episode of fictive vomiting
[14], was still observed immediately after the expulsion 4.1. Central structures non-essential for coordinating
phase. vomiting

4.1.1. Cerebellum
4. Discussion The cerebellum is not essential for vomiting elicited
by emetic drugs or electrical stimulation of abdominal
The present study demonstrated that the brain re- vagal afferents. Similar observations were previously
gion critical for coordinating vomiting lies in the made for vomiting induced by apomorphine, copper
medulla between the levels of the retrofacial nucleus sulfate, or ingestion of spoiled food [1,2,49]. The nodu-
and the obex. The possible organization of the lus and uvula of the cerebellum had at one time been
medullary structures that produce vomiting are dis- thought to be essential for motion-induced vomiting
cussed below after a consideration of other brain re- [1,2,49]. However, more recent studies have shown that
260 A.D. Miller et al. /Brain Research 647 (1994) 255-264

CONTROL
A. PHR t /~ ~ '~ ,I~ a. ,

Aen

KAINIC ACID INJECTION


PHR

ABD
__,., ',, _ ' L J' ~.___/i L._

I I I I I I
5 see

+0.5

+1.5

+3.0

~ig. 5. Fictive vomiting before and after kainic acid injections into the dorsolateral reticular formation. Data from two different animals are
shown in columns A and B. Artifact from electrical stimulation of the vagus nerve appears on the phrenic discharge in the lower panel in B.
Extent of spread of green dye indicated by hatched areas on frontal sections at 0.5, 1.5 and 3.0 m m rostral to the obex. The dye extended from
about 1 m m caudal to the obex to about 4 m m rostral to it in both animals. AMB, nucleus ambiguus; AP, area postrema; CX, external cuneate
nucleus; DMV, dorsal motor nucleus of the vagus; INT, nucleus intercalatus; LR, lateral reticular nucleus; PR, paramedian reticular nucleus; ST,
solitary tract; 5SP, spinal trigeminal nucleus; 12, hypoglossal nucleus; other abbreviations as in previous figures.
A.D. Miller et aL / Brain Research 647 (1994) 255-264 261

A. CONTROL 1, similar patterns of abdominal discharge could be


PHR obtained before and after cerebellar removal.

4.1.3. Rostral brain structures


ABD We also demonstrated that structures rostral to the
level of the retrofacial nucleus are not essential for
RLN ~ S S S vomiting. However, these structures undoubtedly serve
as an important input for eliciting emesis under certain
conditions, for example vomiting in anticipation of
cancer chemotherapy [38] and psychogenic vomiting
B. TRANSECTION AT OBEX
[39]. Moreover, vomiting has also been evoked by elec-
PHR
trical stimulation of rostral brain structures, including
the amygdala, ventral anterior thalamic nucleus, hy-
ABD
pothalamus and hippocampus [20,40].
RLN ~' S S S
.\ 4.1.4. Spinal cord
The spinal cord is of course essential for relaying
descending motor commands to the thoracic and ab-
dominal musculature that is critical for producing vom-
Fig. 6. Fictive vomiting before and after transverse section of the iting. However, following transection of the neural axis
brainstem 4 mm caudal to the obex. While the lesion abolished just caudal to the obex, we were still able to elicit a
phrenic (PHR) and abdominal (ABD) nerve discharge, the recurrent
laryngeal nerve (RLN) displayed typical fictive vomiting behavior. vomiting-like response recorded from the recurrent
The expulsion phase is indicated by a downward arrow. Fictive laryngeal nerve. Fictive vomiting was followed by
vomiting is followedby rhythmic fictive swallowing(S). During respi- rhythmic bursts characteristic of the bucco-pharyngeal
ration, the RLN was active during both the inspiratory and early stage of swallowing which typically follows an episode
expiratory(post-inspiratory)phases (not illustrated). of fictive vomiting [14]. The whole recurrent laryngeal
nerve exhibited a complex response during fictive vom-
iting. The activity in phase with co-active phrenic and
abdominal bursts and the large discharge during the
these structures are not essential for vestibular-induced expulsion phase are typical of the expiratory branch of
vomiting elicited by sinusoidal galvanic stimulation of the nerve [14]. The activity between retching bursts is
the vestibular labyrinth using stimulus frequencies that typical of the nerve branch that innervates the esopha-
are most effective for producing motion sickness [37]. gus [25]. The inspiratory branch of the nerve is silent
By analogy, it was argued that the nodulus and uvula during vomiting [14].
are also unlikely to be essential for motion-induced The spinal cord may also serve additional roles
vomiting. during vomiting. Interneurons have recently been
recorded at the level of the phrenic motor nucleus that
exhibit behaviors indicating that they serve more com-
4.1.2. Infracerebellar nucleus plex functions than merely relaying descending com-
It has recently been reported that lesions of the mands to motoneurons [17]. In addition, it has been
infracerebellar nucleus abolish spinal expiratory activ- suggested that spinal afferents may contribute to initi-
ity [21,22]. The effects of the lesions were most striking ating vomiting under certain conditions [7,28,47].
in cats with vagotomies. In the present studies with
animals having intact vagus nerves rostral to the site of 4.2. Brainstem regions essential for vomiting
the vagal stimulating electrode in the lower thorax, we
found that abdominal nerve discharge continued dur- Previous lesion studies in dogs have demonstrated
ing expiration, fictive vomiting, and fctive coughing, that a critical region for producing the respiratory-re-
following removal of the entire cerebellum including lated components of fictive vomiting exists at the level
the infracerebellar nucleus on both sides. Thus, al- of the retrofacial nucleus [12]. We have confirmed
though the infracerebellar nucleus may play some role these findings in cats and extended them to show that
in modulating spinal expiratory activity, its presence is kainic acid injections into this region also abolish f c -
not essential for abdominal nerve discharge. Due to tive vomiting. This latter result is important since it
the variability of the presence and amplitude of ab- demonstrates that dysfunction of cell bodies in this
dominal discharge during respiration in decerebrate region, as opposed to f b e r tracts, is sufficient to abol-
cats we did not attempt to quantify the effects of ish the response. Fukuda and Koga [12,13,26] proposed
infracerebellar removal. However, as illustrated in Fig. that a 'pattern generator' for vomiting is located in the
262 A.D. Miller et al. / Brain Research 647 (1994) 255-264

B6tzinger complex in the region of the retrofacial recorded in decerebrate cats from neurons that were
nucleus. However, since in both their studies and ours not spontaneously active but that we considered to be
the occurrence of fictive vomiting was determined by respiratory-related on the basis of rhythmic variations
monitoring abdominal and phrenic nerve discharge, it in their antidromic activation in phase with the respira-
is still unknown whether lesions in this region abolish tory cycle [32]. Fukuda and Koga [13] did not attempt
all components of vomiting (e.g. including gastrointesti- to identify the axonal projections of their neurons.
nal changes [29]) or only the respiratory-related com- Borison and Wang [8,48] had previously suggested
ponents. Certainly, the effects of the lesions could that a 'vomiting center' existed in the dorsolateral
easily be explained by disruption of the pre-motor and medullary reticular formation. In the present studies,
motor output circuitry for the respiratory components fictive vomiting could still be elicited after kainic acid
of vomiting that are known to exist in the retrofacial injections that included much of the brainstem region
region [13,14,24,31,33]. Our finding that weak, tonic where 13orison and Wang reported that electrical stim-
co-activation of the phrenic and abdominal nerves could ulation could produce vomiting [8] or where large le-
sometimes be obtained in response to emetic stimuli sions induced by chronic radon seed implantation im-
following lesions in this region is consistent with these paired the vomiting response to emetic drugs [48].
lesions having their effect at the level of the pre-motor However, no typical expulsion phase was observed af-
ouput circuitry. ter our injections, suggesting that this region of the
Large lesions that abolished fictive vomiting also brainstem may contain neurons important for produc-
typically abolished, or at least greatly modified, sponta- ing this phase of vomiting. Thus, the structures in-
neous respiratory discharge, consistent with the exis- volved in coordinating vomiting probably include, but
tence at the level of the retrofacial nucleus of neuronal are more extensive than, the region envisioned by
circuitry common to the production of both behaviors. Borison and Wang.
It has previously been shown that lesions at this level
of the brainstem can drastically effect respiration in 4.3. How is uomiting coordinated?
adult cats [43,45]. In the present studies, the failure of
procedures that produced more limited damage to the The results of the present study as well as our
retrofacial nucleus and adjacent tissue to abolish either previous work [34,36] suggest that vomiting is coordi-
fictive vomiting or respiration suggests that the neu- nated by a distributed medullary control system rather
rons critical for the production of these behaviors are than by a unique, well-defined 'vomiting center'. The
not tightly localized within a small area. distribution of c-los protein (Fos), a marker of neu-
Another line of evidence that Fukuda and Koga [12] ronal excitation, induced by vomiting and fictive vomit-
used to suggest that there is a pattern generator for ing in cats suggests that neurons involved in coordinat-
vomiting in the B6tzinger complex is their finding that ing vomiting are embedded in an arc of neurons radiat-
electrical stimulation in this region elicits fictive vomit- ing from the area postrema and nucleus of the solitary
ing in dogs. In contrast, in the present studies with tract (NTS) through the intermediate reticular zone of
cats, we were unable to elicit fictive vomiting in para- the lateral tegmental field to the ventrolateral medulla
lyzed animals or vomiting in a non-paralyzed animal [34]. The area postrema is a chemoreceptive area for
using electrical stimulation of similar areas of the triggering vomiting. Lesions of the area postrema block
brainstem, despite our use of the same stimulus param- vomiting in response to many but not all emetic drugs
eters as Fukuda and Koga. A possible species differ- [6]. The area postrema projects to the NTS [41,46]. The
ence in this regard between dogs and cats has yet to be discharge of NTS neurons is increased by administra-
directly tested. Our present results confirm the find- tion of the emetic drug apomorphine into the fourth
ings of Miller and Wilson [36] that electrical stimula- ventricle in dogs [26]. The NTS also receives input
tion of the brainstem in non-paralyzed cats failed to from the abdominal vagus nerve [19,23,26]. Vomiting
elicit readily reproducible vomiting. induced by activation of abdominal vagal afferents can
Fukuda and Koga [13] also reported that the be blocked by reversible cooling of the NTS [26]. The
B6tzinger complex in decerebrate dogs contains non- recent finding that some NTS neurons receive conver-
respiratory neurons, intermingled with respiratory neu- gent input from the vestibular labyrinth and abdominal
rons, that exhibit firing patterns appropriate for gener- vagus nerve suggests that such neurons may also be
ating the emetic act. They argued that the presence of important for mediating motion sickness [15]. Thus, the
such neurons supports the existence of a vomiting NTS may represent the beginning of a final common
pattern generator in the B6tzinger complex. However, pathway by which different emetic inputs produce vom-
these neurons may well have included latent respira- iting. As discussed above, the region of the retrofacial
tory neurons and upper airway motoneurons that did nucleus contains pre-motor and motor neuronal cir-
not exhibit spontaneous respiratory modulation under cuitry critical for generating the pattern of the respira-
their experimental conditions. We have previously tory-related components of vomiting. The manner in
A.D. Miller et al. / Brain Research 647 (1994) 255-264 263

which various emetic inputs trigger all of the compo- [15] Gr61ot, L., Jakus, J., Miller, A.D. and Yates, B.J., Vestibular
and visceral inputs to nucleus solitarius and adjacent structures
nents of the act of vomiting is still incompletely under-
in cat brainstem. Soc. Neurosei. Abstr. 19 (1993) 1492.
stood. [16] Gr61ot, L. and Milano, S., Diaphragmatic and abdominal muscle
activity during coughing in the decerebrate cat, NeuroReport, 2
(1991) 165-168.
[17] Gr61ot, L., Milano, S., Portillo, F. and Miller, A.D,, Respiratory
Acknowledgments
interneurons of the lower cervical (C4-C5) cord: membrane
potential changes during fictive coughing, vomiting, and swal-
We thank Dr. Laurent Gr61ot for his participation lowing in the decerebrate cat, Pfliigers Arch., 425 (1993) 313-320.
in some experiments and Drs. L. Gr61ot, V.J. Wilson [18] Gustafsson, B. and Jankowska, E., Direct and indirect activation
and B.J. Yates for their comments on the manuscript. of nerve cells by electrical pulses applied extracellularly, J.
Physiol., 258 (1976) 33-61.
This study was supported by a grant from the National
[19] Gwyn, D.G., Leslie, R.A. and Hopkins, D.A., Gastric afferents
Institute of Neurological Disorders and Stroke to the nucleus of the solitary tract in the cat, Neurosci. Lett., 14
(NS20585). (1979) 13-17.
[20] Hess, W.R., (J.R. Hughes (Ed.)), The Functional Organization of
the Diencephalon, Grune & Stratton, New York, 1957.
[21] Huang, Q., Zhou, D. and St. John, W.M., Vestibular and
References cerebellar modulation of expiratory motor activities in the cat, J.
Physiol., 436 (1991) 385-404.
[1] Bard, P., Physiological investigation of causes and nature of [22] Huang, Q., Zhou, D. and St. John, W.M., Cerebellar control of
motion sickness, National Research Council Committee on Avia- expiratory activities of medullary neurons and spinal nerves, J.
tion Medicine, Report No. 485, 1945. Appl. Physiol., 74 (1993) 1934-1940.
[2] Bard, P., Woolsey, C.N., Snider, R.S., Mountcastle, V.B. and [23] Kalia, M. and Mesulam, M.-M., Brain stem projections of sen-
Bromiley, R.B., Delimitation of central nervous mechanisms sory and motor components of the vagus complex in the cat. II.
involved in motion sickness, Fed. Proc., 6 (1947) 72. Laryngeal, tracheobronchial, pulmonary, cardiac and gastroin-
[3] Berman, A.L., The Brainstem of the Cat. A Cytoarchitectonic testinal branches, J. Comp. Neurol., 193 (1980) 467-508.
Atlas with Stereotaxic Coordinates, University of Wisconsin Press, [24] Koga, T. Discharge patterns of bulbar respiratory neurons dur-
Madison, 1968. ing retching and vomiting in decerebrate dogs, Jap. J. Physiol.,
[4] Bianchi, A.L., Gr61ot, L., Miller, A.D. and King, G.L., Mecha- 41 (1991) 233-249.
nisms and Control of Emesis, Colloque INSERM, Vol. 223, John [25] Koga, T. and Fukuda, H., Charactersitic behavior of the respira-
Libbey Eurotext, Montrouge, France, 1992. tory muscles, esophagus, and external anal and urethral sphinc-
[5] Bolser, D.C., Fictive cough in the cat, J. Appl. Physiol., 71 (1991) ters during straining, retching, and vomiting in the decerebrate
2325 -2331. dog, Jap. J. Physiol., 40 (1990) 789-807.
[6] Borison, H.L., Area postrema: chemoreceptor circumventricular [26] Koga, T. and Fukuda, H., Neurons in the nucleus of the solitary
organ of the medulla oblongata, Prog. Neurobiol., 32 (1989) tract mediating inputs from emetic vagal afferents and the area
351-390. postrema to the pattern generator for the emetic act in dogs,
[7] Borison, H.L., McCarthy, L.E. and Johnson, J.R., High dorsal Neurosci. Res., 14 (1992) 166-179.
column cordotomy plus subdiaphragmatic vagotomy prevents [27] Kucharczyk, J., Stewart, D.J. and Miller, A.D., Nausea and
acute ionizing radiation sickness in cats, Exp. Neurol., 98 (1987) Vomiting: Recent Research and Clinical Advances, CRC Press,
645-658. Boca Raton, FL, 1991, 251 pp.
[8] Borison, H.L. and Wang, S.C., Functional localization of central [28] Laffan, R.J. and Borison, H.L., Emetic action of nicotine and
coordinating mechanism for emesis in cat, J. Neurophysiol., 12 lobeline, J. Pharmacol. Exp. Ther., 121 (1957) 468-476.
(1949) 305-313. [29] Lang, I.M., Digestive tract motor correlates of vomiting and
[9] Coyle, J.T. and Schwartz, R., The use of excitatory amino acids nausea, Can. J. Physiol. Pharmacol., 68 (1990) 242-253.
as selective neurotoxins. In A. Bj6rkland and T. H6kfelt (Eds.), [30] Lawes, I.N.C., The central connections of area postrema define
Handbook of Chemical Neuroanatomy, Elsevier, Amsterdam, the paraventricular system involved in antinoxious behaviors. In
1983, pp. 508-527. J. Kucharczyk, D.J. Stewart and A.D. Miller (Eds.), Nausea and
[10] Davis, C.J., Harding, R.K., Leslie, R.A. and Andrews, P.L.R., Vomiting: Recent Research and Clinical Advances, CRC Press,
The organisation of vomiting as a protective reflex. In C.J. Boca Raton, FL, 1991, pp. 77-101.
Davis, G.V. Lake-Bakaar and D.G. Grahame-Smith (Eds.), [31] Miller, A.D. and Ezure, K., Behavior of inhibitory and excita-
Nausea and Vomiting: Mechanisms and Treatment, Springer- tory propriobulbar respiratory neurons during fictive vomiting,
Verlag, Berlin, 1986, pp. 65-75. Brain Res., 578 (1992) 168-176.
[11] Davis, C.J., Lake-Bakaar, G.V. and Grahame-Smith, D.G. (Eds.), [32] Miller, A.D., Ezure, K. and Suzuki, I., Control of abdominal
Nausea and Vomiting: Mechanisms and Treatment, Springer- muscles by brain stem respiratory neurons in the cat, J. Neuro-
Verlag, Berlin, 1986, 184pp. physiol., 54 (1985) 155-167.
[12] Fukuda, H. and Koga, T., The B6tzinger complex as the pattern [33] Miller, A.D. and Nonaka, S., B6tzinger expiratory neurons may
generator for retching and vomiting in the dog, Neurosci. Res., inhibit phrenic motoneurons and medullary inspiratory neurons
12 (1991) 471-485. during vomiting, Brain Res. 521 (1990) 352-354.
[13] Fukuda, H. and Koga, T., Non-respiratory neurons in the [34] Miller, A.D. and Ruggiero, D.A., Emetic reflex arc revealed by
B6tzinger complex exhibiting appropriate firing patterns to gen- expression of the immediate-early gene c-los in the cat, J.
erate the emetic act in dogs, Neurosci. Res., 14 (1992) 180-194. Neurosci., 14 (1994) 871-888.
[14] Gr61ot, L., Barillot, J.C. and Bianchi, A.L., Activity of respira- [35] Miller, A.D., Tan, L.K. and Suzuki, I., Control of abdominal
tory-related orophayrngeal and laryngeal motoneurones during and expiratory intercostal muscle activity during vomiting: role
fictive vomiting in the decerebrate cat, Brain Res., 513 (1990) of ventral respiratory group expiratory neurons, J. Neurophysiol.,
101-105. 57 (1987) 1854-1866.
264 A.D. Miller et al. / Brain Research 647 (1994) 255-264

[36] Miller, A.D. and Wilson, V.J., 'Vomiting center' reanalyzed: an [43] Speck, D.F., B6tzinger complex role in phrenic-to-phrenic in-
electrical stimulation study, Brain Res., 270 (1983) 154-158. hibitory reflex of cat, J. Appl. Physiol., 67 (1989) 1364-1370.
[37] Miller, A.D. and Wilson, V.J., Vestibular-induced vomiting after [44] St. John, W.M., Neurogenesis, control, and functional signifi-
vestibulocerebellar lesions, Brain Behal,. Et:ol., 23 (1983) 26-31. cance of gasping, J. Appl. Physiol., 68 (1990) 1305-1315.
[38] Morrow, G.R. and Dobkin, P.L., Anticipatory nausea and vomit- [45] St. John, W.M., Hwang, Q., Nattie, E.E. and Zhou, D., Func-
ing in cancer patients undergoing chemotherapy treatment: tions of the retrofacial nucleus in chemosensitivity and ventila-
prevalence, etiology, and behavioral interventions, Clin. Psychol. tory neurogenesis, Resp. Physiol., 76 (1989) 159-172.
Ret., 8 (1988) 517-556. [46] Strominger, N.L., Knox, A.P. and Carpenter, D.O., The connec-
[39] Muraoka, M., Mine, K., Matsumoto, K., Nakai, Y. and Naka- tivity of the area postrema in the ferret, Brain Res. Bull., 33
gawa, T., Psychogenic vomiting: the relation between patterns of (1994) 33-47.
vomiting and psychiatric diagnoses, Gut, 31 (1990) 526-528. [47] Sugiyama, H., Hayama, T. and Yagasaki, O., Emetic action of
[40] Robinson, B.W. and Misbkin, M., Alimentary responses to bacterial endotoxin in the cat, Proc. Soc. Exp. Biol. Med., 121
forebrain stimulation in monkeys, Exp. Brain Res., 4 (1968) (1966) 278-281.
330-366. [48] Wang, S.C. and Borison, H.L., The vomiting center: its destruc-
[41] Shapiro, R.E. and Miselis, R.R., The central neural connections tion by radon implantation in dog medulla oblongata, Am. J.
of the area postrema of the rat, J. Comp. Neurol., 234 (1985) Physiol., 166 (1951) 712-717.
344-364. [49] Wang, S.C. and Chinn, H.I., Experimental motion sickness in
[42] Sleisenger, M.H. (Ed.), The Handbook of Nausea and Vomiting, dogs. Importance of labyrinth and vestibular cerebellum, Am. J.
Caduceus Medical Publishers Inc., Pawling NY, 1993, 173 pp. Physiol., 185 (1956) 617 623.

Vous aimerez peut-être aussi