Neural mechanisms of emesis' DAVID0.CAWENTER
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 12/10/14 For personal use only.
Wdswortbe Center for-~ b s r w ~ o r i and e s Research, New York State Department of Health and School of Public Healthp Albany, NY 6223;4 U.I.S.A. Received December 85, 1988 CARPENTER, B. 0,1990. Neural mechanisms of emesis. Can. J. Physiol. Bhamacol. 68: 230-236. Emesis is a reflex, developed to different degrees in different species, that allows an animal to rid itself of ingested toxins or poisons. The reflex can be elicited either by direct n e u r o d connections from visceral afferent fibers, especially those from the gastrointestinal tract, or from humoral factors. Emesis from hmoral factors depends on the integrity of the area postrema; neurons in the area postrem have excitatory receptors for emetic agents. Emesis from gastrointestinal afferents does not depend on the area postrema, but probably the reflex is triggered by projections to some part of the nucleus tractus solitmius. As with a variety of other complex motor fbnctions regulated by the brain stem, it is likely that the sequence of muscle excitation and inhibition is controlled by a centrd pattern generator located in the nucleus tractus sslihriras, and that h f o m t i o n from humoral factors via the area postrem and visceral afferents via the v a p s nerve converge at this point. This centrd pattern generator, like those for motor hnctions such as swallowing, presumably projects to the various motor nuclei, perhaps through intemeurowa%pathways, to elicit the sequential excitatiom and inhibition that controls the reflex. Key words : area postrema, emesis, neurotransmitters, cyclic AMP, neuropeptides.
CARPENTER, D. 0.1990. Neural mechanisms of emesis. Can. J. Physisl. Bhwmaeol. 68 : 230-236. k9Cm&seest un rCflexe, dCveloppC B divers degrCs chez diffkrentes espkces, qui pennet 2 un animal Be se libkrer lui-mi5me de toxines ou poisons absorMs. Le rbflexe p u t Ctre dCc1ench6 soit par des connewions neuronales directes des fibres afferentes viscbrales, spCciaHement ceBles du tube gastrointestinal, ou par des facteurs hurnoraux. L'krnhse provenant des facteurs humorauw dCpend de 19inaCgf.itkde l'arera postrema; les neurones de I'area psstrema ont des rdcepteurs excitatemrs pour les 6mCtiques. L'Cm&seprovenant des fibres affkrentes gastrointestinales ne d$end pas de I'area postrema, mais le rCflexe est prsbablement d6clenchC par des projections dam une partie du woyau du faisceau solitaire. C o m e p u r une variCtC d9autres fowctions rnotrices complexes rkglkes par le tronc ckrkbral, il se peut que la sCquence d9inkibition et d9excitation du muscle soit csntrAlCe par un gkntrateur central dms le noyau du faisceau solitaire et que 19infomtiondes facteurs humraux via 19areapostrema et les fibres affkresates viscCrales via le pneumogastrique convergent B ce point- Ce gCnCrateur central, c o m e ceux pour les fonegions motrices telles que la dCglutition, prsjette vraisemblablement aux divers noyaux moteurs, possiblement par les voies interneuronales, pour produire les skquences d'excitatiosa et d'irhibitisn qui contr6lent le rkflexe. [Traduit par la revue]
Emesis is a complex reflex pathway controlled by the brain stem and involving humoral factors, afferent fibers, and complex excitation and inhibition of both visceral and somatic musculature. The fundamental features of this mechanism were first elucidated in the early 1950s by Wmg, Borison, B r i z m , and their colleagues (see reviews by Wang, '1988 and Barnes, 1984). Forty years later, h e neural mechanisms proposed at h a t time for the most part still stand, and this fact is testimony to the magnitude of the contribution made by these individuals. A particularly important concept, proposed first by Wang and Borison (1952), is that the site at which humord substances act differs from that involved in motor mechanisms, and that these two compnents of the emetic reflex are performed by different neuronal populations. Two of the earliest methods for eliciting emesis were by administration of intravenous apmorphine or intragastric copper sulfate. It was found that while both induced emesis, that to copper sulfate was blocked by cutting the vagus and sympathetic afferents, while that to apomorphine was sensitive to area gostrema ablation (Fig. 1). These results were interpreted to mean that a variety of humsrd agents (including intravenous copper sulfate) induce emesis by actions at the area postrema, one of the circumventricaalar organs which is outside of the blood -brain 'This paper was presented at the symposium Nausea and Vomiting: A Multidisciplinary Perspective, held November 12 and 13, 1988, Ottawa, Ont., Canada, and has undergone the Journal's usual peer review. Printed in Camada / Imprim6 au Canada
barrier (Wisloclai. and h t n a m 1920) and thus has direct contact to circulating dmgs and toxins. In contrast, tntragastrtc copper sulfate apparently activates vagd and sympathetic afferent fibers in the stomach wdl since vagotomy and sympatkctomy are effective in preventing emesis (Wang and Borison 1951). These afferent fibers are able to initiate the emetic reflex but do so by brain-stem pathways that do not involve the area postrema. As illustrated in Fig. 1, it was proposed by Wang and Borison (1952) that these afferent fibers project to a motor emetic center, onto which infomation converges from the neurons of the area postrema. Even in the light of contemporary reexamination of the concept of a defined motor emetic center, there is agreement that these two general patterns of emesis still differ. Emesis is a reflex with survivd vdue for the animal in that it dlows it to rid itself of ingested toxins and poisons. The tight coupling of nausea and emesis makes it likely that these are two components of one reflex and that the unpIeasantness of nausea evolved to reduce the chances that an animal will choose to ingest the same toxin again. Even in rodents, which do not show any motor components of the emetic reflex, taste aversions (bait shyness) remain a major function of the area postrema (&bin et dil.2983; Ossenkopp and Giugno 1985), and it is likely that this reflects the rodent equivdent of nausea. The ultimate experiment has even been done in humans. Lindstrom and Brizzee (1962) treated five humans who suffered from intractable nausea and vomiting with local area postrema ablation, and all patients experienced total relief of these symptoms.
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 12/10/14 For personal use only.
APOMORPHINE MORPHlNE LANABOSlDE C HYDERGINE I.V. COPPER SULFATE
VA AFFE
- SYMPATHETIC AFFERENTS
FIG. I . Schematic representation of the afferent pathways for vomiting by agents acting at central and peripheral sites. TZ, chemoeeptive emetic trigger zone; VC, vomiting center. Agomorphine, morphine, lanatoside C, hydergine, and copper sulfate (iv) all act at the TZ, while intragastric copper sulfate acts through gastric afferents, which project directly to the vomiting center. (From Wang and Tyson (1954).
Merents of the emetic reflex There has been relatively little study of the ability of afferent fiber stimulation to elicit the emetic reflex beyond the demonstration that intragastric copper sulfate causes emesis independent of the area pstrema. Vagal afferents are primarily responsible, but there is some contribution of sympathetic afferents Wang and Borison 1951). There have been several studies investigating the brain-stem distribution of vagal afferent fibers (see Kalia and Mesulam 1980; Chernicky et d o1984). Most vagal afferents project to the nucleus tractus solitarius and the projections from particular organs are to a number of subnuclei. The ody major sites other than nucleus tractus solitarius to which vagal fibers project a p p r to be the area postrema and the dorsal motor vagal nucleus. Anatomic studies demonstrate y connections to and from the area postrema. Morest (1960), in a Golgi study, found many fibers passing both ways between the area postrema and the nucleus tractus solibrius, and later he (Morest 1967) showed ascending pathways in the dorsd and lateral columns of the spinal cord projecting to the area postrema. Other projections have been reported from dorsomedial hypthalmus (Hosoyu and Matsushita 1981) and nucleus armbipus, dorsal motor vagus, nucBeus intracalatus, and XHHth
nucleus (Vigier and Rouviere 1979). There is considerable evidence that vagal afferents project to the area postrema (Kalia and Mesulum 1980; Beckstead and Norgren 1979; Chernicky et al . 1984), including specifically afferents from the stomach wall (Gwyn et al. 1879). The functional significance of these afferents to the area postrema has not been evaluated. While the ablation studies clearly demonstrate that emesis can be elicited without the area postrema, they do not allow one to evaluate the role that afferent fiber innervation of the area postrema might play in the reflex. Certainly the density of vagal terminals in the area postrema is not as dense as in the portions of the nucleus tractus solitarius just deep of the area postrema, but in intact animals vagal activation may promote the emetic reflex through direct synaptic excitation of area postrema neurons as well as through other pathways. One of the major difficulties in the study of the emetic reflex is its great variability among animal species (Borison et al. 1981). As indicated above, rodents do not vomit at all and the sensitivity of other species varies considerably. Man and dog are clearly among the most sensitive species to emesis from humoral agents, while both also show emesis in response to motion, ionizing radiation, and drugs used for cancer chemotherapy. Monkey and cat are considerably less sensitive to most of these stimuli, including apomorphine. Other triggers of emesis include sounds and smells ody associated with injections of chemotherapeutic drugs in what is called anticipatory vomiting, elevated intracranial pressure or brain in~ury,inflammatory bowel disease, especially with distension, and endocrine disturbances such as pregnancy. Neither of the first two examples is thought to utilize the area postrema but rather acts primarily through descending cerebral pathways. Details of their projection pathways are unknown. Bowel disease probably activates vagal and sympathetic afferents, while vomiting of pregnancy is probably hormonal, but there has been little experimental study of either. There remains considerable controversy as to whether or not other forms of emesis (radiation, motion, chemotherapeutic drugs) are mediated ultimately by humoral or nerve afferents. Even here there may be differences between different species, as for example with radiation-induced vomiting. In dog, radiation emesis totally depends on the integrity of the area postrema Wang et al. 1958; Chinn and Wang 1954; Carpenter et d. 1986). In cats, radiation-induced emesis does not depend on the area postrema but is abolished if animals are subjected both to a bilateral dorsal column cordotomy plus a subdiaphragmatic vagotomy @orison et al. 1987). The neuronal pathways for motion sickness are unknown and there is evidence both for W m g and Chinn 1954; Brizzee et A. 1980) and against (Borison 1985) a central role of the area postrema, although these may once again be species differences. In light of the above discussion and the many afferents to the area postrema, it is important to recognize that, while humoral mediators must involve the area postrema, it is very pssible that afferent projections to the area postrema, not just other areas, might also trigger the reflex. However, the principal projection of vagal fibers is to the nucleus tractus solitarius, and since this is dso the site of major projections from the area pstrema, it is a good candidate for a site at which humoral information from the area postrema and visceral afferent information from the v a p s might converge.
Mechanisms sf humor& mediatiom of the emetic reflex Our laboratory has recently attempted to elucidate the
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 12/10/14 For personal use only.
232
CAN. J. PHYSIOL. PHAMACOL. VOL. 68, 1990
mechanisms whereby humord agents activate the emetic reflex through the area postrema. These studies have utilized dogs. We first tested the emetic potentid of common neurotransmitters and peptides to get some indication of the number of possible transmitter receptors involved (Carpenter et dal. 1984) and then made electrophysiologicd recordings from area postrema neurons and applied a wide variety sf agents onto these neurons by ionophoresis (Carpenter et d* 1983, 1988). Table 4 indicates the subsmces found to excite area pstrema neurons. There are several points of interest regarding this infomation. 1. It is remarkable that these very small neurons have receptors for so many substances. The largest neurons of the area postrema are only about 40 pm in diameter and yet they appear to have receptors for almost all of the substances known to function as neuroactive agents within the nervous system. The active agents include conventional neurotrmsmitkrs, peptides, and hormones. In the case of insulin, the area postrema is the ody CNS site where insulin has been shown to be directly excitatory (Carpenter and Briggs 1986). 2. These studies were d l performed using a seven-barrel electrode array from which extracellular recordings were made through the center barrel, while the others were used for ionophoresis or pressure ejection of agonists. The percentage excitation is so high (about 50% on average) as to suggest that every neuron has receptors for d l of these substances. When recording extracellularly, one can only detect suprathreshold excitation, while depolarization which does not reach spike generation would not be detected. With his large array it is unlikely that the transmitter is always applied directly on the neuron. 3. Behavioral pharmacologic studies suggest that each of the receptors is distinct, except in the case of apomorphine md dopamine or acetylcholine and pilocarpine, each pair of which is h o w n to act at a single receptor. For example, enkephdininduced emesis is blocked by ndoxone but that due to apomorphine or mgioknsin is not. In addition, desensitization to emesis secondary to intravenous application of one transmitter did not dter responses to others (Carpenter et al. 1984). These results are consistent with the idea that these neurons have a large number of discrete receptors. 4. The majority of substances for which excitatory responses were found are h o w n to be emetic, as referenced in the right column of Table 1. This is good support for the original hypothesis that humor& agents induce emesis by activating the neurons in the area postrema, which then project to deeper brain-stem structures to stimulate the reflex. The excitatory substances not reported to be emetic have agsgarently not been tested for emetic action. The nature sf the excitatory response obtained from areapostrema neurons is unusual9 as illustrated in Fig. 2. While glutmate caused a brief, high frequency discharge of this neuron, not very different from the excitation by glutmate elsewhere in the nervous system, gastrin caused an excitation of a very long latency (more than 10 s), low frequency (about 0.5 Hz), and long duration (over 30 s). This pattern was typic d for dI of the substances in Table 1 other than glutamate. Figure 3 shows an electronic display, rather than raw data, from mother experiment with glutmate, leucine enkephdin, and apmo~gshine. The long-latency, Bong-duration peptide responses are clear. Frequently multiple applications of several substances would cause the neuron to become spon-
TABLE1. Transmitter actions on area postrem neurons
Substance Glutamate Acetylcholine Nicotine Pilocarpine Serotonin Norepinephrine Histamine Epinephrine Apomorphine Doparnine Insulin Zinc Glucose Angiotensin II Neurotensin TRH VIP Gastrin Substance P Vasopressin Leucine Enkephalin Sesmtostatin CCK LHRH Neuropeptide Y Cdcitonin Prostaglandin A, Prostaglandin A, Prostaglandin B, Prostaglandin B, Brastaglandin D, Prostaglandin D, Prostaglandin El Prostaglandin P,, Prostaglandin F,,
Ns.of % units excitation
Reference for emesis due to substance
Borissn 1959 Douglas 1975 Jenkins and L&y 1971 Bhargava and Dixit 1968 Peng 1963 Hatcher and Weiss 1923 Innes aand Nickerson 1975 Carpenter and Briggs 1986 Carpenter et al. Carpenter et al. Carpenter et al. Carpenter et al. Carpenter et al. Carpenter et al. Carpenter et al.
1984 1984 1984 1984 1984 1984 1984
Carpenter et al. 1984 Carpenter et al. 1984 Harding and McDonald 1989
Maul et al. 1978
NOTE:TRH,thyroid-releasing hormone; CCK, cholecystothiniw; LHRH, luteinking hormone-releasing hormone; VIP, vasoactive intestid polypeptide.
tmeously active at a low frequency; this spontaneous discharge would be maintained for many minutes. We never found spontaneously active units under control conditions. While a prolonged excitatory response after a brief application of a transmitter may very well be the appropriate foundation for a pathway generating a sensation like nausea and a reflex like vomiting, which usually lasts for at least a period of minutes, such a prolonged action from the brief application of a transmitter onto a single neuron is usud. These observations suggest the possibility that responses to d l of these substances have a c o m o n step or are mediated through a common second messenger system. We have tested this hypothesis (Carpenter et al. 1988) and have evidence consistent with the view that cyclic AMP is a c o m o n mediator of the responses to d l but glumate. Cyclic AMP and forskolin, as activator of adenylate cyclase, excite these neurons (Table 2) and show a similar pattern of response (Fig. 4). This conclusion is both interesting and somewhat surprising, especially for substances such as apomovhine, which one expects to act through a D2 receptor (Stefanini md ClementCorrnier 198H), which is not supposed to be coupled to adenylate cyclase (Kebabian md Cdne 1979), and for insulin, which
CARPENTER
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 12/10/14 For personal use only.
B ) Gastrin (588 nC)
FIG.2. Responses of an area postrema neuron to glutamate (Glu) and gastrin. The neuron was at 543 ym below the surface and did not respond to either VIP or angiotensin II. The large deflections are ionophoretic artifacts. Glutarmate was applied as at single pulse of 50 nC. Gastrin was applied +s five pulses, each 180 nC, of which only the Bast two are shown. (From Carpenter et al. (1983) with pespnissisn.)
Leucine Enkephalin [I 40 nC] .----
II
- - --
\Ills lllllljI !I
11
I
Apo morphine (210 nC]
FIG. 3. Raster display of respnses of an area p s t r e m neuron 502 ym &low the surface to ionophoretie application of glutamate (Glu), leucine enkephdie, and apmorphine. The time of ionophoretie application is indicated by the Barge upward deflection, whereas the smdler deflections are electronic pulses triggered by single enkephdin. There was a 7-min interval between the two le~cineewkephalin applications to avoid receptor desensitization. (From Carpenter et d. (1988) with permission.)
at most other sites is associated with an inhibition of adenylate qclase (Uiano and Cuatrecasos 1972) and activation of phosphdiesterase (Fatemi 1985). However, even for these two substances there is additional support for the view that the respnses are indeed mediated through cyclic AMP. Pretreatment of affjlds with a phosphdiesterase inhibitor should retard the breakdown of cyclic AMP and, if involved in mediation of the response, should reduce the threshold dose necessary to induce emesis. We tested theophylline, IBMX,
md RO-17%; thresholds for dl substances tested were reduced. The results with theophylline for insulin, angiotensin, and apmorphine are shown in Table 3. While the evidence for cyclic nucleotide mediation of responses to transmitters, pegtides, and hormones is indirect and derived from only behavioral and extraceildar recording studies, it is not unreasonable that these neurons should use a common mechanism for excitation. Although there is evidence for a role of the area postrem in cardiovascular control (Joy 19%1;
CAN. J. PHYSIOL. PHAMACOL. VOL. 68, 1990
TABLE2. Effects of cyclic nucleotides md related substances -
Substance
-
-
-
-
-
No. of units
-
-
-
-
-
TABLE3. Effects of theophylline on emesis threshold Pretreat with theophylline (25 rnglkg ip)
-
% excitation
ContrsIs
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 12/10/14 For personal use only.
Concentration No. of trials % emesis No, sf trials
5 IU/kg 18 IUIkg 15 %U/kg
12 24 12
Fink et al. 1887; Otsuka et d. 1986) and satiety (vmder Kooy 1984; Edwards and Ritter H986), the major function of the area pstrema appears to be regulation of nausea and vomiting. Thus, it is logical that these neurons respond to a great variety of agents but give the same type of response to all of them.
The motor emetic center The original description of the motor emetic center was based on studies by Borison and Wmg (19491, who performed studies on emesis induced by electrical stimulation of the brain stem and by Borison and Wang (195I), who implanted radon beads to cause emesis. Both studies suggested a discrete brainstem site from which one could evoke vomiting, and the stirnulation experiments of Kum and Sugihara (1955) are d s s consistent with that view. However, there has been no anatomic evidence for such a site, which wodd be expected to receive a major projection from the area postrema. In physiologic studies, Morest (1961) stimulated the area postrem in cats and recorded responses in the dorsolaterd reticular formation near the XIIth nucleus, dong the border and within the substance of the central grey between the posterior c o d s s u r e , more ventrally dong the tectospid tract, and in the inferior border of the superior colliculus. Ikeda et d.(1970) d s o stimulated the area postrema and found in the dog that this would induce emesis. They found that a stimulus would excite about 40% of the neurons in the dorsal portion of the nucleus reticularis parvocellularis and a smaller percentage of neurons in more remote portions of the reticular formation. However, in neither study did the authors claim to have found the motor emetic center. Miller and Wilson (1983) have more recently specifically reexamined this problem with selective brain-stern stimulation and were unable to find a discrete site from which they could consistently elicit emesis. There are other reasons to suspect that a complex reflex such as emesis would not be under the control of a localized motor site. The reflex itself requires the coordinated activity of a great number of motor nuclei (see Barnes 19841, each of which is involved in many different frenctions. These include d l of the abdominal muscles, the diaphragm, the iwtercostds, the muscles of the larynx, pharynx, and tongue, the neurons controlling tone and peristalsis in esophagus, stomach a d smdl intestine, as well as sphincter muscles. The reflex is dso almost always accompmied by changes in autonomic activation, giving gdlor, cold sweats, and hypotension. Most importantly, the motor neurons controlling d l of these muscle groups and autonomic functions must be activated in the appropriate sequence, not just d l activated at once. There are other complex motor acts that are regulated by the brain stem, such as swallowing, breathing, chewing, and even central control of cardiovascular drive. Each of these is regulated by neurons in the nucleus tractus solitarius. These
0.0025 mg/kg 0.0050 mglkg
4
0.0075 mg/kg
4 4
0.0100 mg/kg
4
Insulin (iv) 0 46 47
Apornorphine 0 25 50 100
% emesis
12 12 12
42 66 $3
4
0
4
50
4 4
75 100
NOTE:Effects were studied in 4 days, each before a d after theophylline.
neurons, often restricted in location, are not motoneurons. They are, however, neurons that control the patterns of motor activity and have been described as "central pattern generators." The centrd pattern generators are sites of convergence of afferent information. These neurons provide the trigger stimulus for the generation of the whole of the motor act. However, they operate through a network of difhse interconnections throughout the brain stem to recruit the various motor components in a stereotypic and highly reproducible sequence. The concept of a central pattern generator was derived from studies of movement (walking and swimming) in invertebrates (Hartline m d Gossie 1979; Getting et d. 1980; Weeks 1981). Evidence has been gathered consistent with a similar centrd pattern generator operating in mammdian movement (Grillner and W d e n 1985), breathing (Von Euler 1983), mastication (Dellow and Lund 19T1), and swallowing (Jean and Car 1979; Jean et ale 1975). Similarly, in emesis there is a defined pattern in which muscles must contract or relax and this pattern must be followed over a period of time. Thus, retching is a rhythmic contraction and relaxation, and the sphincter relaxation in the expulsion phase must not occur until the expulsion phase is reached. $0 in emesis, there must be a pattern of excitation and inhibition at the level of brain-stem motor nuclei over a period of tens of seconds and involving many different motor groups. Inherent in the concept of a centrd pattern generator is the notion that the whole reflex is hardwired; once the central pattern generator is activated, the whole sequence of excitation and inhibition follows. By analogy, with these other complex acts, such as swdlowing, it appears very likely that emesis is dso organized with a centrd pattern generator located in the nucleus tractus solitarius. Such a suggestion is not incompatible with the evidence from experiments on electrical stimulation of the brain stem. In d l sf the studies done, including those of Miller and Wilson (19831, emesis was on occasion produced by stimulation; it was not, however, consistent from one site and it could often be elicited from more than one site. This may reflect activation of various cornpwents of the diffkse network.
CARPENTER
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 12/10/14 For personal use only.
i l l .I-PLP..! .-._HI I -__!!-
FIG. 4. Raster display of responses of an area postrem neuron 620 pm below the surface to glutamate (Glu), forsk~lin,and CAMP. The tall and (or) hick black areas indicate time of ionophoresis. (From Carpenter et al. (1988) with permission.)
In summary, emesis is a complex reflex which has afferent, kumord, and motor components, which involves many different trmsmitters, peptides, and hormones, and which is not exactly the same in all species. While we have made some recent progress in understanding this complex behavior, we still have much to understand. In particular, we do not h o w what substances function as physiological h u m r d mediators of emesis of various forms, nor how the nucleus tractus solitarius neurons function to coordinate the motor cornponents of emesis. These questions are important ones that pose challenges for future research. BARNES, J. H. 1984. The physiology and pharmacology of emesis. Mol. Aspects Med. 7: 397 -508. BECKSTEAD, R. M., and N O R G ~ N W., 1979. An autoradiography emmaination of h e central distribution of the trigerninal, facial, glosssphargrngeal and vagd nerves in the monkey. J. Comp. Neurol. l a : 455 -472. BHARGAVA, K. B., and Dnxn~,K. %.1968. Role of the chemoreceptor trigger zone in histamine-induced emesis. Br. J. Phamcol. 34: 508-513. B o m s o ~ H. , L. 1959. Effect of ablation of rn&ullargr emetic chernoreceptor trigger zone on vomiting responses to cerebral intraventricular injection of adrenaline, apomorphine and pilocarpine in the cat. J. PhysioH. (London), 147: 172 - 177. 1985. A misconception of motion sickness leads to false therapeutic expectations. Aviat. Space Environ. M&. 56: 66-68. BONSON,H. L., and WANG,S. C. 1949, Functional localization of central coordinating mechanism for emesis in cat. J. Neurophysisl. 12: 305-313. 1951. Quantitative effects of radon implanted in the medulla oblongata: a technique for producing discrete lesions. J. Comp. Neurol. 94: 33 -56. BOMSON, H. L., BONSON,R., and MCCARTHY, L. E. 1981. Phylogenic and neurologic aspects of the vomiting process. J. Clin. Phamcol. 21: 235 -295. BOIPISON, H. L., MCCARTHY, k. E., and JOHNSON, J. R. 1987. High dorsal column csrdotomy plus subdiaphragmatic vagotomy prevents acute ionizing radiation sickness in cats. Exp. Neurol. 98: 645 -658. BNZZEE,K. R., ORDY,J. M . , and MBHLER,W. R. 1980. Effect of ablation of area postrema on frequency and latency of motion sickness-induced emesis in the squirrel monkey. Physiol. Behav. 24: 849-853.
CARPENTER, D,O., and BRJGGS,D. B. 1986. Insulin excites neurons of the area pstrema and causes emesis. Neurosci. Lett, 68: $5 -89. CAWENTER, B. O., BWIGGS,D. B., and STWOMHNGEW, N. 1983. Responses of neurons of canine area postrema to neurotransmitters and peptides. Cell Msl. Neurobiol. 3: 1 13 - 126. 1984. Peptide-induced emesis in dogs. Behav. Brain Res . 11: 277-281. CARPENTER, D. O., BIP~WS, D. B., KNOX,A. Pe, and STROMINGER, Nok. 1986. Radiation-induced emesis in h e dog: effects of lesions and drugs. Radiat. Res. 1818: 307 -316. 1988. Excitation sf area postrema neurons by transmitters, peptides, and cyclic nucHeotides. J. Neurophysio1. 59: 358-369. CHERNICKY, C. L., BARNES, K.,FERRAWICB, C., and Corsowr~,J. 1984. Afferent projections of the cervical vagus and nodose ganglion in the dog. Brain Res. Bull. 13: 401 -4.1 1. CH~NN H., I., and WANG,S. C. 1954. Locus of emetic action following irradiation. Proc. Soc. Exp. Biol. Med. 89: 472 -474. DELLRIW, P. C., and LUND,J. P. 1971. Evidence for central timing of rhythmical mastication. J. Physiol. (London), 215: H 1 - 13. DCBUGUS, W. W. 1975. Histamines and antihistamines; 5-hydroxytryptarnine and antagonists. In The phmacological basis of &erapeutics. 5th &. Edited by L. S . Cbodmn and A. GiHman. Macmillan. New York. pp. 590 -629. EDWARDS, G. L., and RITTER,R. C. 1986. Area postrema lesions: cause of overingestion is not altered visceral nerve function. Am. J. Physiol. 2511: 8575 -R581. FATEMI, S. H. 1985. Insulin-dependent cyclic AMP turnover in isoHated rat adipscytes. Cell. Mol. Biol. 31: 153 - 161. FINK,G. B., BRUNER, C . A., and MANGIAPANE, M. L. 1987. Area pstrema is critical for angiotensin-induced hypertension in rats. Hypertension (Dallas), 9: 355 - 36 1 . GETTING,P. A., LENNAWD, P. R., and HUME,R. I. 1980. Central pattern generator mediated swimming in Tritonia. 1. Identification and synaptic interactions. J. Neurophysisl. 44: 15 1 -214. GRHLLNEW, S., and WALLEN, P. 1985. Central pattern generators for locomotion, with specid reference to vertebrates. Annu. Rev. Neurosci. 8: 233-261. GWYN,D. G., LESLIE,W. A., and HOPKINS,DmAA.1979. Gastric afferents to the nucleus of the solitary tract in the cat. Neurosci. k#. 84: 13-17. HARDING, R. K., and MCBONALD, T. J. 1989. Identification and characterization sf the emetic effects of peptide - YY. Peptides (Fayetteville, NU), 10: 2 1 -24. HAWLINE, D. K., and GOSSIB,D. V., JR. 1979. Pattern generation in the lobster (PanecUms) stownotogastric ganglion. 1. Pylorie neuron
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 12/10/14 For personal use only.
236
CAN. J. PHYSIOE. PHARMACOL. VOL. 68, 1990
kinetics and synaptic interactions. Biol. Cybern. 33: 209-222. HAXHEW,R. A., and WEIISS,S , 1923. Studies on vomiting. J. Phamacol. 22: 139 - 193. HOSOYA, Pa'. , and MA~USHPTA, M. 1981, A direct projection from the hypothdamus to the area postrema in the rat, as demonstrated by the H W and autoradiographic methods. Brain Res. 214: 144 - 149. IKEDA,M .,Kuso, S., and IWASE,Y. 1970. The response of medullary reticular neurones to electrical stimulation of the chemoreceptor trigger zone in area postrema of the dog. J. Physiol. Soc. Jpn. 32: 600-60.5. ILLHANB, G . , and CUATRECASBS, P. 1972. Modulation of adenylate cyclase activity in liver and fat cell membranes by insulin. Science (Washington, DC), 175: 906 -908. INN=, I. R.,and NICKERSON, M. 1975. Norepinephrine, epinephrine and the sympathomimetic amines. Hta The phamacological basis of therapeutics. 5th ed. Edited by L. S. Goodman and A. Gilmn. MacMiIlan, New York, pp. 4'37-513. JEAN,A., and CAR, A. 1979. Inputs to the swallowing medullary neurons from the peripheral afferent fibers and the swallowing cortical arm. Brain Res. 178: 567 -572. JEAN,A., CAW,A., and ROMAN,C. 1975. Comparison of activity in pontine versus medullary neurones during swaBlowing. Exp. Brain Res. 22: 211 -220. JENKINS,%. C., and LAHAY,D. 19'31. Central mechanisms of vomiting related to catecholamine response: anesthetic implications. Cm. Anesth. Soc. J. 18: 434-441. JOY, M. D. 1971. The intramedullary connections of the area postrema involved in the central cardiovascular response to angiotensin II. Clin. Sci. 41: 89- 100. KALIA,M., and MESULAM, M.-M. 1980. Brain stem projections of sensory and motor components of the vagus complex in the cat. HI. Laryngeal, tracheobronchiall, pulmonary, cardiac, and gastrointestinal branches. J. Comp. Neurol. 193: 467 -508. KAUL,A. E , FEDBWSCHWEIDER, J. M., and STUBBLE~ELD, M. D. 1978. A controlled trial s f antiemetics in abortion of PGF,, and La~~~lr~ards. J . Reprod. Med. 20: 2 13 -2 18. KEBABIAN, J. We, and CALNE,D. B. 1979- Multiple receptors for doparanine. Nature (London), 277: 93 -96. KURU,M., and SWGIHARA, S. 1955. Contributions to the bowlledge of bulbar autonomic centres. %I.Relationship of the vagal nuclei to the gastro-jejunal motility* Jpn. J. Physiol. 5: 21 -36. LEVEY,S%., H A ~ O WJ.NE., , and SMYTH,C. J. 1949. Serum glutamic acid Bevels and the occurrence of nausea and vomiting after the intravenous administration of amins acid mixtures. J. Lab. Clin. Md.34: 1238 - 1248. LHNDSTROM, P. A*, and BMEZEE,K. R. 1962. Relief of intractable vomiting from surgical lesions in the area gostrew. J. Neurosurg. 19: 228-236. MILLER,A. De, and WILSON,V*J. 1983. "Vomiting Center9' reanalyzed: an electrical stimulation study. Brain Res. 270: 9.54 - 158.
MORE ST^ DeK. 1960. A study of the s t m c t ~ ~of r e the area pstrema with Golgi methods. Am. J. Anat. 107: 291 -303. 1961. Ascending pathways from an osmotically sensitive region of the medulla oblongata. Exp. Neurol. 4: 413 -423. 1967. Experimental study of the projections of the nucleus tractus solitarius and the area postrem in the cat. J. C o w . Neurol. 130: 277 -299. OSSENKOP~ K.-P., and GIUGNO,L. 1985. Taste aversions conditioned with multiple exposures to g a m radiation: abolition by area postrema lesions in rats. Brain Res. 3 4 : 1 -7. OTSUKA,A., BARNES,K. %., and F ~ a a ~ w rC. s , M. 1986. Contribution of area postrem to pressor actions of angiotensin HI in dog. Am. J. Physiol. 2511: H538-H546. PEWG,M. 4%. 1963. Locus of emetic action of epinephrine and dopa h dogs. J. Pharmacol. Exp. Ther. 139: 345 -349. RABHN,B. M . , HUNT, W. A., and LEE, J. 1983. Attenuation of radiation- and drug-induced conditioned taste aversions following area pstrema lesions in the rat. h d i a t . Res. 93: 388-394. STEFANHNI, E., and C L E M E N T - C ~ ~ IYo ER 1981. , Detection of dopamine receptors in the area postrema. Eur. J. P b a m c o l . 74: 257 -2m. VAN DER Ko~Y,19. 1984. Area postrema: site where cholecystokinin acts to decrease food intake. Brain Res. 295: 345-347. MGHER, D., and R O U V ~ EA. ~ E 1979. , Afferent and efferent comections of the area postrema demonstrated by the horseradish proxih s e method. Arch. %taleBiol. 117: 325-339. VQNEULER,C. 1983. On the central pattern generator for the basic breathing rhythmicity. J. Appl. Physiol. 55: 1647 - 1659. WANG,S. C. 1980. Physiology and pharmacology of the brain stem. Putura, Mt. Kisco, New York. WANG,S. C., and BONSON,H. L. 1951. Copper sulfate emesis: a study of afferent pathways from the gastrointestinal tract. Am. J. Pbysiol. 164: 520 -526. 1952. A new concept of organization of the central emetic mechanism: recent studies on the sites of action of apomorphine, copper sulfate and cardiac glycosides. Gastroenterology, 22: 1-12. WAWG,S. C., md C a r w ~8%. , 1. 1954. Experimental motion sickness in dogs. Functional importance of chemoreceptive emetic trigger zone. Am. J. Physiol. 178: 11 1 - 116. WANG,S. C., and 'W~SON, R. k. 1954. Central nervous pathways of experimental motion sickness. Int. Rec. Med. 167: 641 -650. WANG,S . C., RENZI,A. A., and CHINN,H. I. 1958. Mechanism of emesis following x-irradiation. Am. J. Physiol. 193: 335-339. WEEKS,9. C. 1981. Neuronal basis of leech swimming: separation of swim initiation pattern generation, and intersegmental coordination by selective lesions. J. Neurophysiol. 45: 698-723. W I S ~ C K 6. I , B., and PUTNAM,T. J. 1920. Note on the anatomy of the areae postremae. Anat. Rec. f 9: 281 -28'3.
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 12/10/14 For personal use only.
Wdswortbe Center for-~ b s r w ~ o r i and e s Research, New York State Department of Health and School of Public Healthp Albany, NY 6223;4 U.I.S.A. Received December 85, 1988 CARPENTER, B. 0,1990. Neural mechanisms of emesis. Can. J. Physiol. Bhamacol. 68: 230-236. Emesis is a reflex, developed to different degrees in different species, that allows an animal to rid itself of ingested toxins or poisons. The reflex can be elicited either by direct n e u r o d connections from visceral afferent fibers, especially those from the gastrointestinal tract, or from humoral factors. Emesis from hmoral factors depends on the integrity of the area postrema; neurons in the area postrem have excitatory receptors for emetic agents. Emesis from gastrointestinal afferents does not depend on the area postrema, but probably the reflex is triggered by projections to some part of the nucleus tractus solitmius. As with a variety of other complex motor fbnctions regulated by the brain stem, it is likely that the sequence of muscle excitation and inhibition is controlled by a centrd pattern generator located in the nucleus tractus sslihriras, and that h f o m t i o n from humoral factors via the area postrem and visceral afferents via the v a p s nerve converge at this point. This centrd pattern generator, like those for motor hnctions such as swallowing, presumably projects to the various motor nuclei, perhaps through intemeurowa%pathways, to elicit the sequential excitatiom and inhibition that controls the reflex. Key words : area postrema, emesis, neurotransmitters, cyclic AMP, neuropeptides.
CARPENTER, D. 0.1990. Neural mechanisms of emesis. Can. J. Physisl. Bhwmaeol. 68 : 230-236. k9Cm&seest un rCflexe, dCveloppC B divers degrCs chez diffkrentes espkces, qui pennet 2 un animal Be se libkrer lui-mi5me de toxines ou poisons absorMs. Le rbflexe p u t Ctre dCc1ench6 soit par des connewions neuronales directes des fibres afferentes viscbrales, spCciaHement ceBles du tube gastrointestinal, ou par des facteurs hurnoraux. L'krnhse provenant des facteurs humorauw dCpend de 19inaCgf.itkde l'arera postrema; les neurones de I'area psstrema ont des rdcepteurs excitatemrs pour les 6mCtiques. L'Cm&seprovenant des fibres affkrentes gastrointestinales ne d$end pas de I'area postrema, mais le rCflexe est prsbablement d6clenchC par des projections dam une partie du woyau du faisceau solitaire. C o m e p u r une variCtC d9autres fowctions rnotrices complexes rkglkes par le tronc ckrkbral, il se peut que la sCquence d9inkibition et d9excitation du muscle soit csntrAlCe par un gkntrateur central dms le noyau du faisceau solitaire et que 19infomtiondes facteurs humraux via 19areapostrema et les fibres affkresates viscCrales via le pneumogastrique convergent B ce point- Ce gCnCrateur central, c o m e ceux pour les fonegions motrices telles que la dCglutition, prsjette vraisemblablement aux divers noyaux moteurs, possiblement par les voies interneuronales, pour produire les skquences d'excitatiosa et d'irhibitisn qui contr6lent le rkflexe. [Traduit par la revue]
Emesis is a complex reflex pathway controlled by the brain stem and involving humoral factors, afferent fibers, and complex excitation and inhibition of both visceral and somatic musculature. The fundamental features of this mechanism were first elucidated in the early 1950s by Wmg, Borison, B r i z m , and their colleagues (see reviews by Wang, '1988 and Barnes, 1984). Forty years later, h e neural mechanisms proposed at h a t time for the most part still stand, and this fact is testimony to the magnitude of the contribution made by these individuals. A particularly important concept, proposed first by Wang and Borison (1952), is that the site at which humord substances act differs from that involved in motor mechanisms, and that these two compnents of the emetic reflex are performed by different neuronal populations. Two of the earliest methods for eliciting emesis were by administration of intravenous apmorphine or intragastric copper sulfate. It was found that while both induced emesis, that to copper sulfate was blocked by cutting the vagus and sympathetic afferents, while that to apomorphine was sensitive to area gostrema ablation (Fig. 1). These results were interpreted to mean that a variety of humsrd agents (including intravenous copper sulfate) induce emesis by actions at the area postrema, one of the circumventricaalar organs which is outside of the blood -brain 'This paper was presented at the symposium Nausea and Vomiting: A Multidisciplinary Perspective, held November 12 and 13, 1988, Ottawa, Ont., Canada, and has undergone the Journal's usual peer review. Printed in Camada / Imprim6 au Canada
barrier (Wisloclai. and h t n a m 1920) and thus has direct contact to circulating dmgs and toxins. In contrast, tntragastrtc copper sulfate apparently activates vagd and sympathetic afferent fibers in the stomach wdl since vagotomy and sympatkctomy are effective in preventing emesis (Wang and Borison 1951). These afferent fibers are able to initiate the emetic reflex but do so by brain-stem pathways that do not involve the area postrema. As illustrated in Fig. 1, it was proposed by Wang and Borison (1952) that these afferent fibers project to a motor emetic center, onto which infomation converges from the neurons of the area postrema. Even in the light of contemporary reexamination of the concept of a defined motor emetic center, there is agreement that these two general patterns of emesis still differ. Emesis is a reflex with survivd vdue for the animal in that it dlows it to rid itself of ingested toxins and poisons. The tight coupling of nausea and emesis makes it likely that these are two components of one reflex and that the unpIeasantness of nausea evolved to reduce the chances that an animal will choose to ingest the same toxin again. Even in rodents, which do not show any motor components of the emetic reflex, taste aversions (bait shyness) remain a major function of the area postrema (&bin et dil.2983; Ossenkopp and Giugno 1985), and it is likely that this reflects the rodent equivdent of nausea. The ultimate experiment has even been done in humans. Lindstrom and Brizzee (1962) treated five humans who suffered from intractable nausea and vomiting with local area postrema ablation, and all patients experienced total relief of these symptoms.
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 12/10/14 For personal use only.
APOMORPHINE MORPHlNE LANABOSlDE C HYDERGINE I.V. COPPER SULFATE
VA AFFE
- SYMPATHETIC AFFERENTS
FIG. I . Schematic representation of the afferent pathways for vomiting by agents acting at central and peripheral sites. TZ, chemoeeptive emetic trigger zone; VC, vomiting center. Agomorphine, morphine, lanatoside C, hydergine, and copper sulfate (iv) all act at the TZ, while intragastric copper sulfate acts through gastric afferents, which project directly to the vomiting center. (From Wang and Tyson (1954).
Merents of the emetic reflex There has been relatively little study of the ability of afferent fiber stimulation to elicit the emetic reflex beyond the demonstration that intragastric copper sulfate causes emesis independent of the area pstrema. Vagal afferents are primarily responsible, but there is some contribution of sympathetic afferents Wang and Borison 1951). There have been several studies investigating the brain-stem distribution of vagal afferent fibers (see Kalia and Mesulam 1980; Chernicky et d o1984). Most vagal afferents project to the nucleus tractus solitarius and the projections from particular organs are to a number of subnuclei. The ody major sites other than nucleus tractus solitarius to which vagal fibers project a p p r to be the area postrema and the dorsal motor vagal nucleus. Anatomic studies demonstrate y connections to and from the area postrema. Morest (1960), in a Golgi study, found many fibers passing both ways between the area postrema and the nucleus tractus solibrius, and later he (Morest 1967) showed ascending pathways in the dorsd and lateral columns of the spinal cord projecting to the area postrema. Other projections have been reported from dorsomedial hypthalmus (Hosoyu and Matsushita 1981) and nucleus armbipus, dorsal motor vagus, nucBeus intracalatus, and XHHth
nucleus (Vigier and Rouviere 1979). There is considerable evidence that vagal afferents project to the area postrema (Kalia and Mesulum 1980; Beckstead and Norgren 1979; Chernicky et al . 1984), including specifically afferents from the stomach wall (Gwyn et al. 1879). The functional significance of these afferents to the area postrema has not been evaluated. While the ablation studies clearly demonstrate that emesis can be elicited without the area postrema, they do not allow one to evaluate the role that afferent fiber innervation of the area postrema might play in the reflex. Certainly the density of vagal terminals in the area postrema is not as dense as in the portions of the nucleus tractus solitarius just deep of the area postrema, but in intact animals vagal activation may promote the emetic reflex through direct synaptic excitation of area postrema neurons as well as through other pathways. One of the major difficulties in the study of the emetic reflex is its great variability among animal species (Borison et al. 1981). As indicated above, rodents do not vomit at all and the sensitivity of other species varies considerably. Man and dog are clearly among the most sensitive species to emesis from humoral agents, while both also show emesis in response to motion, ionizing radiation, and drugs used for cancer chemotherapy. Monkey and cat are considerably less sensitive to most of these stimuli, including apomorphine. Other triggers of emesis include sounds and smells ody associated with injections of chemotherapeutic drugs in what is called anticipatory vomiting, elevated intracranial pressure or brain in~ury,inflammatory bowel disease, especially with distension, and endocrine disturbances such as pregnancy. Neither of the first two examples is thought to utilize the area postrema but rather acts primarily through descending cerebral pathways. Details of their projection pathways are unknown. Bowel disease probably activates vagal and sympathetic afferents, while vomiting of pregnancy is probably hormonal, but there has been little experimental study of either. There remains considerable controversy as to whether or not other forms of emesis (radiation, motion, chemotherapeutic drugs) are mediated ultimately by humoral or nerve afferents. Even here there may be differences between different species, as for example with radiation-induced vomiting. In dog, radiation emesis totally depends on the integrity of the area postrema Wang et al. 1958; Chinn and Wang 1954; Carpenter et d. 1986). In cats, radiation-induced emesis does not depend on the area postrema but is abolished if animals are subjected both to a bilateral dorsal column cordotomy plus a subdiaphragmatic vagotomy @orison et al. 1987). The neuronal pathways for motion sickness are unknown and there is evidence both for W m g and Chinn 1954; Brizzee et A. 1980) and against (Borison 1985) a central role of the area postrema, although these may once again be species differences. In light of the above discussion and the many afferents to the area postrema, it is important to recognize that, while humoral mediators must involve the area postrema, it is very pssible that afferent projections to the area postrema, not just other areas, might also trigger the reflex. However, the principal projection of vagal fibers is to the nucleus tractus solitarius, and since this is dso the site of major projections from the area pstrema, it is a good candidate for a site at which humoral information from the area postrema and visceral afferent information from the v a p s might converge.
Mechanisms sf humor& mediatiom of the emetic reflex Our laboratory has recently attempted to elucidate the
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 12/10/14 For personal use only.
232
CAN. J. PHYSIOL. PHAMACOL. VOL. 68, 1990
mechanisms whereby humord agents activate the emetic reflex through the area postrema. These studies have utilized dogs. We first tested the emetic potentid of common neurotransmitters and peptides to get some indication of the number of possible transmitter receptors involved (Carpenter et dal. 1984) and then made electrophysiologicd recordings from area postrema neurons and applied a wide variety sf agents onto these neurons by ionophoresis (Carpenter et d* 1983, 1988). Table 4 indicates the subsmces found to excite area pstrema neurons. There are several points of interest regarding this infomation. 1. It is remarkable that these very small neurons have receptors for so many substances. The largest neurons of the area postrema are only about 40 pm in diameter and yet they appear to have receptors for almost all of the substances known to function as neuroactive agents within the nervous system. The active agents include conventional neurotrmsmitkrs, peptides, and hormones. In the case of insulin, the area postrema is the ody CNS site where insulin has been shown to be directly excitatory (Carpenter and Briggs 1986). 2. These studies were d l performed using a seven-barrel electrode array from which extracellular recordings were made through the center barrel, while the others were used for ionophoresis or pressure ejection of agonists. The percentage excitation is so high (about 50% on average) as to suggest that every neuron has receptors for d l of these substances. When recording extracellularly, one can only detect suprathreshold excitation, while depolarization which does not reach spike generation would not be detected. With his large array it is unlikely that the transmitter is always applied directly on the neuron. 3. Behavioral pharmacologic studies suggest that each of the receptors is distinct, except in the case of apomorphine md dopamine or acetylcholine and pilocarpine, each pair of which is h o w n to act at a single receptor. For example, enkephdininduced emesis is blocked by ndoxone but that due to apomorphine or mgioknsin is not. In addition, desensitization to emesis secondary to intravenous application of one transmitter did not dter responses to others (Carpenter et al. 1984). These results are consistent with the idea that these neurons have a large number of discrete receptors. 4. The majority of substances for which excitatory responses were found are h o w n to be emetic, as referenced in the right column of Table 1. This is good support for the original hypothesis that humor& agents induce emesis by activating the neurons in the area postrema, which then project to deeper brain-stem structures to stimulate the reflex. The excitatory substances not reported to be emetic have agsgarently not been tested for emetic action. The nature sf the excitatory response obtained from areapostrema neurons is unusual9 as illustrated in Fig. 2. While glutmate caused a brief, high frequency discharge of this neuron, not very different from the excitation by glutmate elsewhere in the nervous system, gastrin caused an excitation of a very long latency (more than 10 s), low frequency (about 0.5 Hz), and long duration (over 30 s). This pattern was typic d for dI of the substances in Table 1 other than glutamate. Figure 3 shows an electronic display, rather than raw data, from mother experiment with glutmate, leucine enkephdin, and apmo~gshine. The long-latency, Bong-duration peptide responses are clear. Frequently multiple applications of several substances would cause the neuron to become spon-
TABLE1. Transmitter actions on area postrem neurons
Substance Glutamate Acetylcholine Nicotine Pilocarpine Serotonin Norepinephrine Histamine Epinephrine Apomorphine Doparnine Insulin Zinc Glucose Angiotensin II Neurotensin TRH VIP Gastrin Substance P Vasopressin Leucine Enkephalin Sesmtostatin CCK LHRH Neuropeptide Y Cdcitonin Prostaglandin A, Prostaglandin A, Prostaglandin B, Prostaglandin B, Brastaglandin D, Prostaglandin D, Prostaglandin El Prostaglandin P,, Prostaglandin F,,
Ns.of % units excitation
Reference for emesis due to substance
Borissn 1959 Douglas 1975 Jenkins and L&y 1971 Bhargava and Dixit 1968 Peng 1963 Hatcher and Weiss 1923 Innes aand Nickerson 1975 Carpenter and Briggs 1986 Carpenter et al. Carpenter et al. Carpenter et al. Carpenter et al. Carpenter et al. Carpenter et al. Carpenter et al.
1984 1984 1984 1984 1984 1984 1984
Carpenter et al. 1984 Carpenter et al. 1984 Harding and McDonald 1989
Maul et al. 1978
NOTE:TRH,thyroid-releasing hormone; CCK, cholecystothiniw; LHRH, luteinking hormone-releasing hormone; VIP, vasoactive intestid polypeptide.
tmeously active at a low frequency; this spontaneous discharge would be maintained for many minutes. We never found spontaneously active units under control conditions. While a prolonged excitatory response after a brief application of a transmitter may very well be the appropriate foundation for a pathway generating a sensation like nausea and a reflex like vomiting, which usually lasts for at least a period of minutes, such a prolonged action from the brief application of a transmitter onto a single neuron is usud. These observations suggest the possibility that responses to d l of these substances have a c o m o n step or are mediated through a common second messenger system. We have tested this hypothesis (Carpenter et al. 1988) and have evidence consistent with the view that cyclic AMP is a c o m o n mediator of the responses to d l but glumate. Cyclic AMP and forskolin, as activator of adenylate cyclase, excite these neurons (Table 2) and show a similar pattern of response (Fig. 4). This conclusion is both interesting and somewhat surprising, especially for substances such as apomovhine, which one expects to act through a D2 receptor (Stefanini md ClementCorrnier 198H), which is not supposed to be coupled to adenylate cyclase (Kebabian md Cdne 1979), and for insulin, which
CARPENTER
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 12/10/14 For personal use only.
B ) Gastrin (588 nC)
FIG.2. Responses of an area postrema neuron to glutamate (Glu) and gastrin. The neuron was at 543 ym below the surface and did not respond to either VIP or angiotensin II. The large deflections are ionophoretic artifacts. Glutarmate was applied as at single pulse of 50 nC. Gastrin was applied +s five pulses, each 180 nC, of which only the Bast two are shown. (From Carpenter et al. (1983) with pespnissisn.)
Leucine Enkephalin [I 40 nC] .----
II
- - --
\Ills lllllljI !I
11
I
Apo morphine (210 nC]
FIG. 3. Raster display of respnses of an area p s t r e m neuron 502 ym &low the surface to ionophoretie application of glutamate (Glu), leucine enkephdie, and apmorphine. The time of ionophoretie application is indicated by the Barge upward deflection, whereas the smdler deflections are electronic pulses triggered by single enkephdin. There was a 7-min interval between the two le~cineewkephalin applications to avoid receptor desensitization. (From Carpenter et d. (1988) with permission.)
at most other sites is associated with an inhibition of adenylate qclase (Uiano and Cuatrecasos 1972) and activation of phosphdiesterase (Fatemi 1985). However, even for these two substances there is additional support for the view that the respnses are indeed mediated through cyclic AMP. Pretreatment of affjlds with a phosphdiesterase inhibitor should retard the breakdown of cyclic AMP and, if involved in mediation of the response, should reduce the threshold dose necessary to induce emesis. We tested theophylline, IBMX,
md RO-17%; thresholds for dl substances tested were reduced. The results with theophylline for insulin, angiotensin, and apmorphine are shown in Table 3. While the evidence for cyclic nucleotide mediation of responses to transmitters, pegtides, and hormones is indirect and derived from only behavioral and extraceildar recording studies, it is not unreasonable that these neurons should use a common mechanism for excitation. Although there is evidence for a role of the area postrem in cardiovascular control (Joy 19%1;
CAN. J. PHYSIOL. PHAMACOL. VOL. 68, 1990
TABLE2. Effects of cyclic nucleotides md related substances -
Substance
-
-
-
-
-
No. of units
-
-
-
-
-
TABLE3. Effects of theophylline on emesis threshold Pretreat with theophylline (25 rnglkg ip)
-
% excitation
ContrsIs
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 12/10/14 For personal use only.
Concentration No. of trials % emesis No, sf trials
5 IU/kg 18 IUIkg 15 %U/kg
12 24 12
Fink et al. 1887; Otsuka et d. 1986) and satiety (vmder Kooy 1984; Edwards and Ritter H986), the major function of the area pstrema appears to be regulation of nausea and vomiting. Thus, it is logical that these neurons respond to a great variety of agents but give the same type of response to all of them.
The motor emetic center The original description of the motor emetic center was based on studies by Borison and Wmg (19491, who performed studies on emesis induced by electrical stimulation of the brain stem and by Borison and Wang (195I), who implanted radon beads to cause emesis. Both studies suggested a discrete brainstem site from which one could evoke vomiting, and the stirnulation experiments of Kum and Sugihara (1955) are d s s consistent with that view. However, there has been no anatomic evidence for such a site, which wodd be expected to receive a major projection from the area postrema. In physiologic studies, Morest (1961) stimulated the area postrem in cats and recorded responses in the dorsolaterd reticular formation near the XIIth nucleus, dong the border and within the substance of the central grey between the posterior c o d s s u r e , more ventrally dong the tectospid tract, and in the inferior border of the superior colliculus. Ikeda et d.(1970) d s o stimulated the area postrema and found in the dog that this would induce emesis. They found that a stimulus would excite about 40% of the neurons in the dorsal portion of the nucleus reticularis parvocellularis and a smaller percentage of neurons in more remote portions of the reticular formation. However, in neither study did the authors claim to have found the motor emetic center. Miller and Wilson (1983) have more recently specifically reexamined this problem with selective brain-stern stimulation and were unable to find a discrete site from which they could consistently elicit emesis. There are other reasons to suspect that a complex reflex such as emesis would not be under the control of a localized motor site. The reflex itself requires the coordinated activity of a great number of motor nuclei (see Barnes 19841, each of which is involved in many different frenctions. These include d l of the abdominal muscles, the diaphragm, the iwtercostds, the muscles of the larynx, pharynx, and tongue, the neurons controlling tone and peristalsis in esophagus, stomach a d smdl intestine, as well as sphincter muscles. The reflex is dso almost always accompmied by changes in autonomic activation, giving gdlor, cold sweats, and hypotension. Most importantly, the motor neurons controlling d l of these muscle groups and autonomic functions must be activated in the appropriate sequence, not just d l activated at once. There are other complex motor acts that are regulated by the brain stem, such as swallowing, breathing, chewing, and even central control of cardiovascular drive. Each of these is regulated by neurons in the nucleus tractus solitarius. These
0.0025 mg/kg 0.0050 mglkg
4
0.0075 mg/kg
4 4
0.0100 mg/kg
4
Insulin (iv) 0 46 47
Apornorphine 0 25 50 100
% emesis
12 12 12
42 66 $3
4
0
4
50
4 4
75 100
NOTE:Effects were studied in 4 days, each before a d after theophylline.
neurons, often restricted in location, are not motoneurons. They are, however, neurons that control the patterns of motor activity and have been described as "central pattern generators." The centrd pattern generators are sites of convergence of afferent information. These neurons provide the trigger stimulus for the generation of the whole of the motor act. However, they operate through a network of difhse interconnections throughout the brain stem to recruit the various motor components in a stereotypic and highly reproducible sequence. The concept of a central pattern generator was derived from studies of movement (walking and swimming) in invertebrates (Hartline m d Gossie 1979; Getting et d. 1980; Weeks 1981). Evidence has been gathered consistent with a similar centrd pattern generator operating in mammdian movement (Grillner and W d e n 1985), breathing (Von Euler 1983), mastication (Dellow and Lund 19T1), and swallowing (Jean and Car 1979; Jean et ale 1975). Similarly, in emesis there is a defined pattern in which muscles must contract or relax and this pattern must be followed over a period of time. Thus, retching is a rhythmic contraction and relaxation, and the sphincter relaxation in the expulsion phase must not occur until the expulsion phase is reached. $0 in emesis, there must be a pattern of excitation and inhibition at the level of brain-stem motor nuclei over a period of tens of seconds and involving many different motor groups. Inherent in the concept of a centrd pattern generator is the notion that the whole reflex is hardwired; once the central pattern generator is activated, the whole sequence of excitation and inhibition follows. By analogy, with these other complex acts, such as swdlowing, it appears very likely that emesis is dso organized with a centrd pattern generator located in the nucleus tractus solitarius. Such a suggestion is not incompatible with the evidence from experiments on electrical stimulation of the brain stem. In d l sf the studies done, including those of Miller and Wilson (19831, emesis was on occasion produced by stimulation; it was not, however, consistent from one site and it could often be elicited from more than one site. This may reflect activation of various cornpwents of the diffkse network.
CARPENTER
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 12/10/14 For personal use only.
i l l .I-PLP..! .-._HI I -__!!-
FIG. 4. Raster display of responses of an area postrem neuron 620 pm below the surface to glutamate (Glu), forsk~lin,and CAMP. The tall and (or) hick black areas indicate time of ionophoresis. (From Carpenter et al. (1988) with permission.)
In summary, emesis is a complex reflex which has afferent, kumord, and motor components, which involves many different trmsmitters, peptides, and hormones, and which is not exactly the same in all species. While we have made some recent progress in understanding this complex behavior, we still have much to understand. In particular, we do not h o w what substances function as physiological h u m r d mediators of emesis of various forms, nor how the nucleus tractus solitarius neurons function to coordinate the motor cornponents of emesis. These questions are important ones that pose challenges for future research. BARNES, J. H. 1984. The physiology and pharmacology of emesis. Mol. Aspects Med. 7: 397 -508. BECKSTEAD, R. M., and N O R G ~ N W., 1979. An autoradiography emmaination of h e central distribution of the trigerninal, facial, glosssphargrngeal and vagd nerves in the monkey. J. Comp. Neurol. l a : 455 -472. BHARGAVA, K. B., and Dnxn~,K. %.1968. Role of the chemoreceptor trigger zone in histamine-induced emesis. Br. J. Phamcol. 34: 508-513. B o m s o ~ H. , L. 1959. Effect of ablation of rn&ullargr emetic chernoreceptor trigger zone on vomiting responses to cerebral intraventricular injection of adrenaline, apomorphine and pilocarpine in the cat. J. PhysioH. (London), 147: 172 - 177. 1985. A misconception of motion sickness leads to false therapeutic expectations. Aviat. Space Environ. M&. 56: 66-68. BONSON,H. L., and WANG,S. C. 1949, Functional localization of central coordinating mechanism for emesis in cat. J. Neurophysisl. 12: 305-313. 1951. Quantitative effects of radon implanted in the medulla oblongata: a technique for producing discrete lesions. J. Comp. Neurol. 94: 33 -56. BOMSON, H. L., BONSON,R., and MCCARTHY, L. E. 1981. Phylogenic and neurologic aspects of the vomiting process. J. Clin. Phamcol. 21: 235 -295. BOIPISON, H. L., MCCARTHY, k. E., and JOHNSON, J. R. 1987. High dorsal column csrdotomy plus subdiaphragmatic vagotomy prevents acute ionizing radiation sickness in cats. Exp. Neurol. 98: 645 -658. BNZZEE,K. R., ORDY,J. M . , and MBHLER,W. R. 1980. Effect of ablation of area postrema on frequency and latency of motion sickness-induced emesis in the squirrel monkey. Physiol. Behav. 24: 849-853.
CARPENTER, D,O., and BRJGGS,D. B. 1986. Insulin excites neurons of the area pstrema and causes emesis. Neurosci. Lett, 68: $5 -89. CAWENTER, B. O., BWIGGS,D. B., and STWOMHNGEW, N. 1983. Responses of neurons of canine area postrema to neurotransmitters and peptides. Cell Msl. Neurobiol. 3: 1 13 - 126. 1984. Peptide-induced emesis in dogs. Behav. Brain Res . 11: 277-281. CARPENTER, D. O., BIP~WS, D. B., KNOX,A. Pe, and STROMINGER, Nok. 1986. Radiation-induced emesis in h e dog: effects of lesions and drugs. Radiat. Res. 1818: 307 -316. 1988. Excitation sf area postrema neurons by transmitters, peptides, and cyclic nucHeotides. J. Neurophysio1. 59: 358-369. CHERNICKY, C. L., BARNES, K.,FERRAWICB, C., and Corsowr~,J. 1984. Afferent projections of the cervical vagus and nodose ganglion in the dog. Brain Res. Bull. 13: 401 -4.1 1. CH~NN H., I., and WANG,S. C. 1954. Locus of emetic action following irradiation. Proc. Soc. Exp. Biol. Med. 89: 472 -474. DELLRIW, P. C., and LUND,J. P. 1971. Evidence for central timing of rhythmical mastication. J. Physiol. (London), 215: H 1 - 13. DCBUGUS, W. W. 1975. Histamines and antihistamines; 5-hydroxytryptarnine and antagonists. In The phmacological basis of &erapeutics. 5th &. Edited by L. S . Cbodmn and A. GiHman. Macmillan. New York. pp. 590 -629. EDWARDS, G. L., and RITTER,R. C. 1986. Area postrema lesions: cause of overingestion is not altered visceral nerve function. Am. J. Physiol. 2511: 8575 -R581. FATEMI, S. H. 1985. Insulin-dependent cyclic AMP turnover in isoHated rat adipscytes. Cell. Mol. Biol. 31: 153 - 161. FINK,G. B., BRUNER, C . A., and MANGIAPANE, M. L. 1987. Area pstrema is critical for angiotensin-induced hypertension in rats. Hypertension (Dallas), 9: 355 - 36 1 . GETTING,P. A., LENNAWD, P. R., and HUME,R. I. 1980. Central pattern generator mediated swimming in Tritonia. 1. Identification and synaptic interactions. J. Neurophysisl. 44: 15 1 -214. GRHLLNEW, S., and WALLEN, P. 1985. Central pattern generators for locomotion, with specid reference to vertebrates. Annu. Rev. Neurosci. 8: 233-261. GWYN,D. G., LESLIE,W. A., and HOPKINS,DmAA.1979. Gastric afferents to the nucleus of the solitary tract in the cat. Neurosci. k#. 84: 13-17. HARDING, R. K., and MCBONALD, T. J. 1989. Identification and characterization sf the emetic effects of peptide - YY. Peptides (Fayetteville, NU), 10: 2 1 -24. HAWLINE, D. K., and GOSSIB,D. V., JR. 1979. Pattern generation in the lobster (PanecUms) stownotogastric ganglion. 1. Pylorie neuron
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by CONCORDIA UNIV on 12/10/14 For personal use only.
236
CAN. J. PHYSIOE. PHARMACOL. VOL. 68, 1990
kinetics and synaptic interactions. Biol. Cybern. 33: 209-222. HAXHEW,R. A., and WEIISS,S , 1923. Studies on vomiting. J. Phamacol. 22: 139 - 193. HOSOYA, Pa'. , and MA~USHPTA, M. 1981, A direct projection from the hypothdamus to the area postrema in the rat, as demonstrated by the H W and autoradiographic methods. Brain Res. 214: 144 - 149. IKEDA,M .,Kuso, S., and IWASE,Y. 1970. The response of medullary reticular neurones to electrical stimulation of the chemoreceptor trigger zone in area postrema of the dog. J. Physiol. Soc. Jpn. 32: 600-60.5. ILLHANB, G . , and CUATRECASBS, P. 1972. Modulation of adenylate cyclase activity in liver and fat cell membranes by insulin. Science (Washington, DC), 175: 906 -908. INN=, I. R.,and NICKERSON, M. 1975. Norepinephrine, epinephrine and the sympathomimetic amines. Hta The phamacological basis of therapeutics. 5th ed. Edited by L. S. Goodman and A. Gilmn. MacMiIlan, New York, pp. 4'37-513. JEAN,A., and CAR, A. 1979. Inputs to the swallowing medullary neurons from the peripheral afferent fibers and the swallowing cortical arm. Brain Res. 178: 567 -572. JEAN,A., CAW,A., and ROMAN,C. 1975. Comparison of activity in pontine versus medullary neurones during swaBlowing. Exp. Brain Res. 22: 211 -220. JENKINS,%. C., and LAHAY,D. 19'31. Central mechanisms of vomiting related to catecholamine response: anesthetic implications. Cm. Anesth. Soc. J. 18: 434-441. JOY, M. D. 1971. The intramedullary connections of the area postrema involved in the central cardiovascular response to angiotensin II. Clin. Sci. 41: 89- 100. KALIA,M., and MESULAM, M.-M. 1980. Brain stem projections of sensory and motor components of the vagus complex in the cat. HI. Laryngeal, tracheobronchiall, pulmonary, cardiac, and gastrointestinal branches. J. Comp. Neurol. 193: 467 -508. KAUL,A. E , FEDBWSCHWEIDER, J. M., and STUBBLE~ELD, M. D. 1978. A controlled trial s f antiemetics in abortion of PGF,, and La~~~lr~ards. J . Reprod. Med. 20: 2 13 -2 18. KEBABIAN, J. We, and CALNE,D. B. 1979- Multiple receptors for doparanine. Nature (London), 277: 93 -96. KURU,M., and SWGIHARA, S. 1955. Contributions to the bowlledge of bulbar autonomic centres. %I.Relationship of the vagal nuclei to the gastro-jejunal motility* Jpn. J. Physiol. 5: 21 -36. LEVEY,S%., H A ~ O WJ.NE., , and SMYTH,C. J. 1949. Serum glutamic acid Bevels and the occurrence of nausea and vomiting after the intravenous administration of amins acid mixtures. J. Lab. Clin. Md.34: 1238 - 1248. LHNDSTROM, P. A*, and BMEZEE,K. R. 1962. Relief of intractable vomiting from surgical lesions in the area gostrew. J. Neurosurg. 19: 228-236. MILLER,A. De, and WILSON,V*J. 1983. "Vomiting Center9' reanalyzed: an electrical stimulation study. Brain Res. 270: 9.54 - 158.
MORE ST^ DeK. 1960. A study of the s t m c t ~ ~of r e the area pstrema with Golgi methods. Am. J. Anat. 107: 291 -303. 1961. Ascending pathways from an osmotically sensitive region of the medulla oblongata. Exp. Neurol. 4: 413 -423. 1967. Experimental study of the projections of the nucleus tractus solitarius and the area postrem in the cat. J. C o w . Neurol. 130: 277 -299. OSSENKOP~ K.-P., and GIUGNO,L. 1985. Taste aversions conditioned with multiple exposures to g a m radiation: abolition by area postrema lesions in rats. Brain Res. 3 4 : 1 -7. OTSUKA,A., BARNES,K. %., and F ~ a a ~ w rC. s , M. 1986. Contribution of area postrem to pressor actions of angiotensin HI in dog. Am. J. Physiol. 2511: H538-H546. PEWG,M. 4%. 1963. Locus of emetic action of epinephrine and dopa h dogs. J. Pharmacol. Exp. Ther. 139: 345 -349. RABHN,B. M . , HUNT, W. A., and LEE, J. 1983. Attenuation of radiation- and drug-induced conditioned taste aversions following area pstrema lesions in the rat. h d i a t . Res. 93: 388-394. STEFANHNI, E., and C L E M E N T - C ~ ~ IYo ER 1981. , Detection of dopamine receptors in the area postrema. Eur. J. P b a m c o l . 74: 257 -2m. VAN DER Ko~Y,19. 1984. Area postrema: site where cholecystokinin acts to decrease food intake. Brain Res. 295: 345-347. MGHER, D., and R O U V ~ EA. ~ E 1979. , Afferent and efferent comections of the area postrema demonstrated by the horseradish proxih s e method. Arch. %taleBiol. 117: 325-339. VQNEULER,C. 1983. On the central pattern generator for the basic breathing rhythmicity. J. Appl. Physiol. 55: 1647 - 1659. WANG,S. C. 1980. Physiology and pharmacology of the brain stem. Putura, Mt. Kisco, New York. WANG,S. C., and BONSON,H. L. 1951. Copper sulfate emesis: a study of afferent pathways from the gastrointestinal tract. Am. J. Pbysiol. 164: 520 -526. 1952. A new concept of organization of the central emetic mechanism: recent studies on the sites of action of apomorphine, copper sulfate and cardiac glycosides. Gastroenterology, 22: 1-12. WAWG,S. C., md C a r w ~8%. , 1. 1954. Experimental motion sickness in dogs. Functional importance of chemoreceptive emetic trigger zone. Am. J. Physiol. 178: 11 1 - 116. WANG,S. C., and 'W~SON, R. k. 1954. Central nervous pathways of experimental motion sickness. Int. Rec. Med. 167: 641 -650. WANG,S . C., RENZI,A. A., and CHINN,H. I. 1958. Mechanism of emesis following x-irradiation. Am. J. Physiol. 193: 335-339. WEEKS,9. C. 1981. Neuronal basis of leech swimming: separation of swim initiation pattern generation, and intersegmental coordination by selective lesions. J. Neurophysiol. 45: 698-723. W I S ~ C K 6. I , B., and PUTNAM,T. J. 1920. Note on the anatomy of the areae postremae. Anat. Rec. f 9: 281 -28'3.
