- Email: [email protected]
Suprapontine control of respiration
Respiration Physiology 114 (1998) 201 – 211
Frontiers review
Suprapontine control of respiration Eric M. Horn, Tony G. Waldrop * Department of Molecular and Integrati6e Physiology, 524 Burrill Hall, 407 South Goodwin A6enue, Uni6ersity of Illinois, Urbana, IL 61803, USA Accepted 9 September 1998
Abstract Despite focus on brainstem areas in central respiratory control, regions rostral to the medulla and pons are now recognized as being important in modulating respiratory outflow during various physiological states. The focus of this review is to highlight the role that suprapontine areas of the mammalian brain play in ventilatory control mechanisms. New imaging techniques have become invaluable in confirming and broadening our understanding of the manner in which the cerebral cortex of humans contributes to respiratory control during volitional breathing. In the diencephalon, the integration of respiratory output in relation to changes in homeostasis occurs in the caudal hypothalamic region of mammals. Most importantly, neurons in this region are strongly sensitive to perturbations in oxygen tension which modulates their level of excitation. In addition, the caudal hypothalamus is a major site for ‘central command’, or the parallel activation of locomotion and respiration. Furthermore, midbrain regions such as the periaqueductal gray and mesencephalic locomotor region function in similar fashion as the caudal hypothalamus with regard to locomotion and more especially the defense reaction. Together these suprapontine regions exert a strong modulation upon the basic respiratory drive generated in the brainstem. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Control of breathing; Hypoxia; Brainstem; Suprapontine; Control of respiration; Suprapontine control
1. Introduction Even though the basic neuronal circuitry essential for the generation and maintenance of res* Corresponding author. Tel.: 1-217-2446037; Fax: 1-2172447078; E-mail: [email protected]
piration is located in the medulla and pons, more rostral areas of the brain are known to have a modulatory influence in respiratory control. In contrast to the necessity of the brainstem sites for generating the respiratory rhythm and its basic pattern, suprapontine areas can be lesioned and respiration stays relatively intact (Tenney and Ou,
0034-5687/98/$ - see front matter © 1998 Elsevier Science B.V. All rights reserved. PII: S 0 0 3 4 - 5 6 8 7 ( 9 8 ) 0 0 0 8 7 - 5
202
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
1977). However, suprapontine areas are known to be involved in adjusting respiration in response to alterations in internal or external environmental conditions. Without these sites, the complete respiratory responses elicited by exercise, hypoxia, hypercapnia, and thermal changes are impaired. Early studies employing electrical stimulation techniques demonstrated various suprapontine areas that increase respiratory drive when stimulated. A study by Spencer (1894) described changes in respiration caused by stimulation of forebrain sites, although the exact stereotaxic coordinates of these locations were not determined. One of the earliest studies to stereotaxically identify areas rostral to the pons that when stimulated elicit increases in respiration was performed by Kabat (1936). He identified the periaqueductal gray (PAG), caudal hypothalamus and subfornical regions as the areas that elicit the strongest increases in respiration when stimulated electrically. In contrast, work in the cortex demonstrated that virtually all of the areas stimulated in the telencephalon lead to a decrease in ventilation with the exception of the motor cortex (Kabat, 1936; Kaada, 1951). One major problem with all of these studies, however, is that electrical stimulation cannot definitively identify locations because fibers of passage are also stimulated. Hence, identifying the exact locations and interactions of areas involved in respiratory control had to wait until better methods were developed. Use of these new methods, such as chemical microinjections, single unit recordings, and functional imaging over the past three decades have provided a clearer understanding of the role of suprapontine areas in respiratory control. This review will focus primarily upon those regions in the telencephalon, diencephalon, and mesencephalon that provide a stimulatory influence over respiratory output.
2. Cortex Although respiration is one autonomic function that can be consciously controlled, the specific mechanisms of cortical control over ventilation remain to be elucidated. The cortex has both afferent input and descending projections that
enable a strong influence in respiratory control (see Davenport and Reep (1995) for more detailed review). Retrograde transport of horseradish peroxidase into spinal areas containing respiratory (phrenic and thoracic) motoneurons demonstrated that cortical inputs onto these neurons originate primarily in area 4 of the motor cortex (RikardBell et al., 1985). In addition, cortical inputs impinging upon midbrain PAG neurons arise predominately from the prefrontal cortex (Beitz, 1982). These limbic inputs are likely involved in the respiratory changes observed during different emotional and arousal states, although little evidence exists on the exact function of these connections. The major cortical inputs into the caudal hypothalamus originate from areas 4 and 6 of the motor cortex, suggesting that this area receives cortical respiratory output mainly associated with locomotion (Baev et al., 1985). Furthermore, single shock sensorimotor cortical stimulation has been shown to stimulate respiratory neurons in the medulla through suggested monosynaptic connections (Planche and Bianchi, 1972). Thus, both motor and limbic areas of the cortex have connections with the major regions involved in respiratory outflow. This variable input is to be expected considering that changes in respiratory output need to be matched to all arousal states. Upon stimulation, most areas of the cerebral cortex elicit decreases in respiratory frequency and depth (Spencer, 1894). This is in contrast to the motor controlling areas of the cortex, which are among the only cortical areas that elicit increases in ventilation when stimulated in mammals (Kaada, 1951, 1960). New imaging techniques such as functional magnetic resonance imaging (fMRI) and positron emission tomography (PET) have allowed the analysis of cortical control in conscious humans. Using PET scanning, areas such as the premotor, primary motor and supplemental motor cortices were shown to be active during volitional breathing at rest in human subjects (Colebatch et al., 1991). In addition to control of ventilation during resting conditions, several cortical areas have been identified that display increases in activity during forced or reflexive increases in ventilation. The use of PET
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
203
Fig. 1. Group results of relative regional cerebral blood flow increases associated with all volitional breathing conditions against added inspiratory loads, with unloaded volitional breathing as control. Data were averaged across all relevant runs for six subjects and are shown as projections onto representations of stereotactic space; areas achieving statistical significance (P B0.001) are shown together with areas in which there was a trend toward significance (*0.001B PB 0.01). L, left; R, right; A, anterior; P, posterior; VAC, vertical plane through anterior commissure; VPC, vertical plane through posterior commissure. Areas of activation are shown with increasing significance by an arbitrary gray scale ranging from light gray to black. Anatomic locations of areas of activation are indicated by lettering adjacent to corresponding region of interest: spr, superolateral precentral gyrus; ipr, inferolateral precentral gyrus; ipo, inferolateral postcentral gyrus; ic, intermediate cingulate; mp, medial parietal cortex; s, striatum; m, midbrain; pf, prefrontal cortex; v, vermis (*). For clarity not all areas of activation are labeled bilaterally. (Reprinted from Fink et al., 1996, with permission).
scanning demonstrated that the premotor, primary and supplemental motor areas are activated during periods of increased force of inspiration (Fig. 1) (Fink et al., 1996). Furthermore, a subcortical structure, the basal ganglia, shows increases in activation during increased inspiratory force as imaged by both PET and fMRI (Gozal et al., 1995; Fink et al., 1996). Hence, many structures in the cortex and subcortex traditionally involved with motor function are activated during ventilatory challenges. In addition, some areas more noted for learned motor behaviors were also activated in these studies, including the supplemental and premotor areas. This suggests that learned behavior is involved in
the sensation of changing respiratory loads and the subsequent ventilatory responses to these sensations.
3. Caudal hypothalamus
3.1. Basal respiratory dri6e The caudal hypothalamus has significant connections between midbrain and brainstem nuclei that control respiratory output. Strong descending projections from the caudal hypothalamus to the PAG and ventrolateral medulla, both areas that modulate respiratory outflow, were visualized
204
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
with the anterograde tracer phaseolus vulgaris leucoagglutinin (PHA) (Vertes and Crane, 1996). These descending projections are coupled with reciprocal projections back to the caudal hypothalamus from medullary respiratory areas but not from midbrain PAG sites (Vibert et al., 1979; Cameron et al., 1995a). The role that the caudal hypothalamus plays in basal respiratory drive has been investigated by several studies that blocked or abolished the activity of caudal hypothalamic neurons (see Waldrop and Porter, 1995). Some of these studies demonstrated that inactivation of caudal hypothalamic neurons by injection of barbiturates or placement of electrolytic lesions produces decreases in respiration (Redgate and Gellhorn, 1958; Keller, 1960; Waldrop et al., 1986b). These results indicate that caudal aspects of the hypothalamus provide an excitatory influence over basal ventilatory drive. The hypothalamus appears to be also involved in respiratory drive originating from other brain regions. One early study by Redgate (1963) sought to investigate this interaction by eliciting increases in ventilation through direct electrical stimulation of the medulla. He found that the ventilatory increases could be attenuated by either injecting barbiturates into or coagulating the caudal hypothalamus. Other investigators have further shown that sectioning of the brain rostral to the diencephalon increases ventilation while a midcollicular sectioning has no effect on ventilation, suggesting that areas rostral to the diencephalon exert an inhibitory influence on respiration that is balanced by facilatory drive coming directly from the diencephalon (Fink et al., 1962; Tenney and Ou, 1977).
3.2. Responses to hypoxia Strong evidence exists showing that the caudal hypothalamus modulates the respiratory responses to hypoxia. Tenney and Ou (1977) demonstrated that the respiratory responses to hypoxia are unchanged following midcollicular decerebration in the cat, but are greatly exaggerated following decortication. These results implicated diencephalic structures, which receive tonic
cortical inhibition during hypoxia, as playing a facilitatory role in the ventilatory responses to hypoxia. The role of hypothalamic areas in respiratory responses to hypoxia was further elucidated through the use of microinjection techniques that permit an evaluation of the neurotransmitters involved in this modulation. Horn and Waldrop (1994) demonstrated that the respiratory response to hypoxia is significantly attenuated following microinjection of either a synaptic blocker or an excitatory amino acid antagonist into the caudal hypothalamus (Fig. 2). In contrast, GABAergic activity in the caudal hypothalamus does not appear to be important since microinjection of a GABA synthesis inhibitor does not alter the response to hypoxia (Peano et al., 1992). These results suggest that neurons in the caudal hypothalamus receive excitatory input during hypoxia that facilitates the respiratory response to hypoxia. Several studies have employed electrophysiological techniques to elucidate the properties of caudal hypothalamic neurons involved in the response to hypoxia. Single units were first shown to be stimulated during hypoxia in this region by Cross and Silver (1963), who demonstrated an increase in firing frequency of a large percentage of neurons in the posterior and lateral hypothalamic nuclei. This work was extended by Dillon and Waldrop (1993) who determined that individual neurons in the caudal hypothalamus of anesthetized cats are stimulated by hypoxia and that a large portion of these cells have respiratory related discharge patterns. In addition, these neuronal responses to hypoxia in the cat are maintained following vagotomy and carotid sinus denervation, suggesting that peripheral chemoreceptor and baroreceptor inputs are not necessary for the hypoxic sensitivity of caudal hypothalamic neurons. Subsequent work in the rat provided similar results as in the cat and also demonstrated that many of these respiratory-related neurons that are stimulated by hypoxia in this area send projections to or through the midbrain PAG (Ryan and Waldrop, 1995). Furthermore, more than 75% of neurons in the caudal hypothalamus were shown
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
205
Fig. 2. Typical cardiovascular and respiratory (DEMG: integrated diaphragmatic EMG activity) responses to hypoxia before (A) and after (B) microinjection of kynurenic acid (KYN) into the caudal hypothalamus. Values for PO2 represent the levels delivered to the animal (Reprinted from Horn and Waldrop, 1994, with permission).
206
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
Fig. 3. (A) Response of a hypothalamic neuron to 10% O2 while perfused with control medium. (B) Response is maintained when synaptic transmission is blocked. Note decrease in spontaneous firing rate. (C) Return to control conditions resulted in recovery of spontaneous activity.
to be stimulated by hypoxia in an in vitro brain slice preparation and most of these neurons maintained this response during perfusion with a synaptic blockade medium (Fig. 3); but perfusion with a sodium channel blocker (tetrodotoxin) greatly attenuated the response (Dillon and Waldrop, 1992; Horn and Waldrop, 1997). These findings suggest that caudal hypothalamic neurons not only modulate the ventilatory response to hypoxia but also actively respond to changes in local oxygen concentration and thus may function as central oxygen chemoreceptors.
3.3. Responses to hypercapnia Caudal hypothalamic neuronal modulation of the respiratory responses to hypercapnia has also been investigated. Waldrop (1991) demonstrated that the respiratory response to hypercapnia is accentuated in the cat following microinjection of GABA antagonists into the caudal hypothalamus. A similar effect upon the respiratory response to hypercapnia was observed in the rat following microinjection of either GABA antagonists or synthesis inhibitors (Peano et al., 1992). Thus,
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
neurons in the caudal hypothalamus receive GABAergic input that modulates the ventilatory response to hypercapnic stimulation. Since most of the GABAergic input coming into the caudal hypothalamus is from local interneurons, this modulation is most likely due to a local circuit in this region (Tappaz and Brownstein, 1977). Additional support for caudal hypothalamic involvement in the response to hypercapnia has been provided by electrophysiological studies. An early study reported that single units recorded in this area are stimulated by hypercapnia in the rabbit brain (Cross and Silver, 1963). Recent extracellular studies in the rat and cat have demonstrated that a small subpopulation of caudal hypothalamic neurons are stimulated during hypercapnia both in vivo and in vitro (Cross and Silver, 1963; Dillon and Waldrop, 1992, 1993). These studies suggest that neurons in the caudal hypothalamus are responsive to hypercapnic stimulation, although whether this stimulation arises from CO2 or local pH changes is unknown.
4. Midbrain Electrical stimulation of many areas in the midbrain produces increases in respiration, which are due to increases in both the respiratory rate and depth (Martin and Booker, 1878). Kabat (1936) was the first to use stereotaxic techniques to determine the specific midbrain areas that elicit this stimulation. He demonstrated that various regions, including the PAG and tegmentum, are able to cause increases in respiration when stimulated. It has also been shown that electrical stimulation of the reticular formation in the midbrain alters both the discharge of respiratory-related units in the medulla and phrenic nerve output (Schmid et al., 1988). Furthermore, recording experiments have shown that neurons in midbrain areas have firing patterns temporally related to the respiratory cycle (Orem and Netick, 1982). The periaqueductal gray is an important midbrain area that modulates ventilation when stimulated, receives strong input from caudal hypothalamic neurons and projects to various
207
respiratory-related nuclei in the pons and medulla such as the Ko¨lliker–Fuse nucleus, ventrolateral medulla, and nucleus tractus solitarii (Cameron et al., 1995b). In one study, the effects of stimulating the PAG upon medullary neurons with a respiratory discharge were examined (Duffin and Hockman, 1972). PAG stimulation was found to switch medullary neurons from firing in the expiratory phase to firing in the inspiratory phase, thus demonstrating the strong modulatory control this region has over respiratory output. As mentioned previously, caudal hypothalamic neurons that respond to hypoxia project to the PAG (Ryan and Waldrop, 1995). Thus, it appears that the PAG may be a midbrain relay from the caudal hypothalamus region. In turn, PAG neurons send projections to medullary neurons that are responsible for altering respiratory drive. Hence, this circuitry is an important portion of the neuroaxis involved in respiratory and autonomic control. Another area involved in respiratory outflow is the midbrain tegmentum, which was investigated by Hugelin and colleagues. They recorded the activity of tegmentum neurons in a preparation (paralyzed, ventilated cats) which was designed to prevent the effects of locomotion that accompany stimulation of this area (Hugelin and Cohen, 1963). These recordings demonstrated that tegmental neurons are synchronized to the respiratory cycle via input from medullary neurons since this synchronization was lost following pontomesencephalic sectioning (Vibert et al., 1979). Work of Eldridge and his co-workers has further determined respiratory properties of specific neurons within the midbrain tegmentum. Neurons in this region possess respiratory-related rhythms when ventilatory drive is increased during hypercapnia but this rhythm is missing during basal respiration (Chen et al., 1991). Furthermore, the respiratory-related activity of these neurons is enhanced when the vagus nerve is either cooled or sectioned (Eldridge and Chen, 1992). These results suggest that neurons in the midbrain central tegmental field are not required for the generation or maintenance of respiratory drive, but do play a role in modulating increases in respiratory outflow.
208
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
Fig. 4. Cardiorespiratory responses to microinjection of picrotoxin (78 nl) into the caudal hypothalamus of a vagotomized cat after muscular paralysis. Control conditions are shown in the left panel. Microinjection of picrotoxin produced increases in arterial pressure, heart rate, phrenic nerve activity, and respiratory frequency. Note that these responses were reversed by the GABAA receptor agonist, muscimol (117 nl), injected into the same site (Reprinted from Waldrop et al., 1988, with permission).
In addition to unit recordings, PET scanning and fMRI imaging have shown the midbrain to be activated in humans when breathing hypercapnic mixtures (Gozal et al., 1994; Corfield et al., 1995). Unfortunately, due to the limited resolution of these techniques, specific nuclei in this area could not be analyzed. In addition, activation of the midbrain occurs during increases in inspiratory force caused by resistive loading in human subjects (Gozal et al., 1995; Fink et al., 1996). These studies highlight the fact that midbrain areas have a much higher level of activity during times of increased ventilatory outflow as compared to basal outflow.
5. Suprapontine respiratory control during co-ordinated activity
5.1. Locomotion A crucial role of the respiratory system is its involvement in meeting the increased metabolic demands associated with locomotion in mammals. The increased ventilation is necessary for adjusting oxygen delivery to the increased oxygen demand during locomotion. Exactly how the central nervous system controls the respiratory system in response to increased activity has long been stud-
ied (see Waldrop et al., 1996 for review). Both feed-forward and feedback mechanisms are now known to be important in this regulation (Eldridge and Waldrop, 1991). Orlovskii (1969) was one of the first investigators to demonstrate that stimulation of the caudal hypothalamus elicits locomotion and at the same time increases in respiration. His work suggested that there is a parallel activation of respiratory outflow and locomotion that compensates for the immediate oxygen needs of the animal. In subsequent studies, it was determined that the increases in respiration during stimulated locomotion can be elicited in a paralyzed, ventilated animal, in which feedback from pulmonary and muscle afferents are eliminated (Eldridge et al., 1981, 1985; Waldrop et al., 1988). The results of these studies demonstrated the presence of a ‘central command’ mechanism situated in the caudal hypothalamus that was responsible for the feed-forward activation of respiration during locomotion. In a subsequent study, microinjection of GABA antagonists into the caudal hypothalamus evoked results similar to those produced by electrical stimulation, thus ruling out the effects being due to activation of the electrical axons of passage (Fig. 4) (Waldrop et al., 1988). Recent work has further demonstrated that caudal hypothalamic neurons increase Fos expression following 45 min
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
of treadmill exercise, suggesting that these neurons are activated during exercise in conscious animals (Iwamoto et al., 1996). It is also important to note that sensory input from contracting muscles has been demonstrated to alter the activity of individual neurons in the caudal hypothalamus of anesthetized cats (Waldrop and Stremel, 1989). Thus, although the above results clearly demonstrate that neurons in the caudal hypothalamus can elicit the respiratory changes that are necessary during locomotion, the importance of peripheral input from muscles must also be considered. Another area that is known to elicit a ‘central command’ type response when stimulated is the mesencephalic locomotor region, MLR (Shik and Orlovskii, 1976). The MLR is similar to the caudal hypothalamus in that local stimulation of neurons in this area initiates locomotion with simultaneous increases in respiration and cardiovascular output. The exact influence on respiration of neurons in this area, however, is somewhat obscure. Although stimulation of neurons in this area causes increases in ventilation, a recent study demonstrated that total lung resistance is increased during stimulation of the MLR (Motekaitis and Kaufman, 1996). This is in contrast to the decreases in lung resistance which occur during muscular contraction alone or during spontaneous exercise. Thus, the exact role of the MLR in adjusting ventilation during locomotion remains to be elucidated.
5.2. Defence reaction One aspect of ventilatory control that is partly attributed to hypothalamic and midbrain influence is the so-called ‘defence reaction’, which is a set of well defined physiological and behavioral responses in awake animals occurring in response to a threatening stimulus. The physiological component consists of tachypnea, tachycardia, and a redistribution of blood flow from the viscera to skeletal muscle while the behavioral component consists of hypervigilance, locomotion, jumping and rearing (Hilton, 1982; Waldrop et al., 1986a; Yardley and Hilton, 1986; Duan et al., 1996). Early work looking at the role of the midbrain
209
and hypothalamus in this behavior determined that stimulation of various regions of the midbrain and hypothalamus is sufficient to elicit this response. Some of the strongest responses can be evoked by electrical stimulation of the midbrain PAG, midbrain tegmentum, dorsomedial hypothalamus, and posterior hypothalamic area (Hilton and Redfern, 1986; Yardley and Hilton, 1986). Microinjecting excitatory amino acids or GABA antagonists to stimulate the cell bodies while leaving fibers of passage unaffected provided evidence of the influence of specific hypothalamic neurons in the defence reaction. Both the physiological and behavioral responses to these injections are similar to those produced by electrical stimulation (Shekhar and DiMicco, 1987; Silveira and Graeff, 1992). Since stimulation of many of the regions that evoke locomotion also produces the defence reaction and many of the physiological responses are similar, the question arises of whether or not there is an anatomical difference in the locomotor and defence reaction sites. Existing evidence points to the same neurons controlling parts (i.e. tachypnea) of the response in both paradigms, but that the overall pattern of behavioral and physiological responses is different depending upon whether locomotion or the defence reaction is initiated. How the hypothalamus integrates information leading to the defence reaction was first studied in the early 1960’s by demonstrating that stimulation of the carotid chemoreceptors elicits the defence response in cats that are decorticated just rostral to the optic chiasm (Bizzi et al., 1961). Since input from the carotid sinus nerve is known to influence neurons in the caudal hypothalamus, the activation of the defence response by chemoreceptor stimulation is possibly mediated through the caudal hypothalamus (Thomas and Calaresu, 1972).
6. Conclusions Elimination of suprapontine sites show only minor effects on resting respiration which gives the impression that these areas are not crucial in respiratory control. However, the contribution of these areas is not in basal respiratory generation
210
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
but more in the control of respiration during various conditions such as locomotion, hypoxia, hypercapnia and the defence reaction. Hypothalamic and midbrain areas are richly connected to higher functioning areas, including areas involved in conscious motor control such as the cerebral cortex. These anatomical and function connections suggest that suprapontine centers are involved in the more complicated tasks of receiving sensory information concerning respiratory needs and in turn modulating the outflow of ventilation to match these needs. These tasks are further complicated by the conscious control over ventilation that must be factored into the integration of the ventilatory output during these conditions. Since most of the work in the control of ventilation has focused on medullary and pontine regions, a more thorough examination of suprapontine areas is needed before a complete understanding of ventilatory control can be attained.
References Baev, K.V., Berezovskii, V.K., Kebkalo, T.G., Savos’kina, L.A., 1985. Projection of forebrain structures of the cat to the hypothalamic locomotor area. Neirofiziologiia 17, 255 – 263. Beitz, A.J., 1982. The organization of afferent projections to the midbrain periaqueductal gray of the rat. Neuroscience 7, 133 – 159. Bizzi, E., Libretti, A., Malliani, A., Zanchetti, A., 1961. Reflex chemoreceptive excitation of diencephalic sham rage behaviour. Am. J. Physiol. 200, 923–926. Cameron, A.A., Khan, I.A., Westlund, K.N., Cliffer, K.D., Willis, W.D., 1995. The efferent projections of the periaqueductal gray in the rat: a phaseolus vulgaris-leucoagglutinin study I. Ascending projections. J. Comp. Neurol. 351, 568 – 584. Cameron, A.A., Khan, I.A., Westlund, K.N., Willis, W.D., 1995. The efferent projections of the periaqueductal gray in the rat: a phaseolus vulgaris-leucoagglutinin study II. Descending projections. J. Comp. Neurol. 351, 585–601. Chen, Z., Eldridge, F.L., Wagner, P.G., 1991. Respiratory-associated rhythmic firing of midbrain neurones in cats: relation to level of respiratory drive. J. Physiol. (Lond.) 437, 305 – 325. Colebatch, J.G., Adams, L., Murphy, K., 1991. Regional cerebral blood flow during volitional breathing in man. J. Physiol. (Lond.) 443, 91–103. Corfield, D.R., Fink, G.R., Ramsay, S.C., 1995. Evidence for limbic system activation during CO2-stimulated breathing in man. J. Physiol. (Lond.) 488, 77–84.
Cross, B.A., Silver, I.A., 1963. Unit activity in the hypothalamus and the sympathetic response to hypoxia and hypercapnia. Exp. Neurol. 7, 375 – 393. Davenport, P.W., Reep, R.L., 1995. Cerebral cortex and respiration. In: Dempsey, J.A., Pack, A.I. (Eds.), Regulation of Breathing. Marcel Dekker, New York, pp. 365 – 388. Dillon, G.H., Waldrop, T.G., 1992. In vitro responses of caudal hypothalamic neurons to hypoxia and hypercapnia. Neuroscience 51, 941 – 950. Dillon, G.H., Waldrop, T.G., 1993. Responses of feline caudal hypothalamic cardiorespiratory neurons to hypoxia and hypercapnia. Exp. Brain. Res. 96, 260 – 272. Duan, Y.F., Winters, R., McCabe, P.M., Green, E.J., Huang, Y., Schneiderman, N., 1996. Behavioral characteristics of defence and vigilance reactions elicited by electrical stimulation of the hypothalamus in rabbits. Behav. Brain. Res. 81, 33 – 41. Duffin, J., Hockman, C.H., 1972. Limbic forebrain and midbrain modulation and phase-switching of expiratory neurons. Brain Res. 39, 235 – 239. Eldridge, F.L., Millhorn, D.E., Waldrop, T.G., 1981. Exercise hyperpnea and locomotion: parallel activation from the hypothalamus. Science 211, 844 – 846. Eldridge, F.L., Millhorn, D.E., Kiley, J.P., Waldrop, T.G., 1985. Stimulation by central command of locomotion, respiration, and circulation during exercise. Respir. Physiol. 59, 313 – 337. Eldridge, F.L., Waldrop, T.G., 1991. Neural control of breathing during exercise. In: Whipp, B., Wasserman, K. (Eds.), Exercise: Pulmonary Physiology and Pathophysiology. Marcel Dekker, New York, pp. 309 – 370. Eldridge, F.L., Chen, Z., 1992. Respiratory-associated rhythmic firing of midbrain neurons is modulated by vagal input. Respir. Physiol. 90, 31 – 46. Fink, B.R., Katz, R., Reinhold, H., Schoolman, A., 1962. Suprapontine mechanisms in regulation of respiration, Am. J. Physiol. 202, 217 – 220. Fink, G.R., Corfield, D.R., Murphy, K., Kobayashi, I., Dettmers, C., Adams, L., Frackowiak, R.S., Guz, A., 1996. Human cerebral activity with increasing inspiratory force: a study using positron emission tomography. J. Appl. Physiol. 81, 1295 – 1305. Gozal, D., Hathout, G.M., Kirlew, K.A., 1994. Localization of putative neural respiratory regions in the human by functional magnetic resonance imaging. J. Appl. Physiol. 76, 2076 – 2083. Gozal, D., Omidvar, O., Kirlew, K.A., 1995. Identification of human brain regions underlying responses to resistive inspiratory loading with functional magnetic resonance imaging. Proc. Natl. Acad. Sci. U.S.A. 92, 6607 – 6611. Hilton, S.M., 1982. The defence-arousal system and its relevance for circulatory and respiratory control. J. Exp. Biol. 100, 159 – 174. Hilton, S.M., Redfern, W.S., 1986. A search for brain stem cell groups integrating the defence reaction in the rat. J. Physiol. (Lond.) 378, 213 – 228. Horn, E.M., Waldrop, T.G., 1994. Modulation of the respiratory responses to hypoxia and hypercapnia by synaptic input onto caudal hypothalamic neurons. Brain Res. 664, 25 – 33.
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211 Horn, E.M., Waldrop, T.G., 1997. Oxygen-sensing neurons in the caudal hypothalamus and their role in cardiorespiratory control. Respir. Physiol. 110, 219–228. Hugelin, A., Cohen, M.I., 1963. The reticular activating system and respiratory regulation in the cat. Ann. N.Y. Acad. Sci. 109, 586 – 603. Iwamoto, G.A., Wappel, S.M., Fox, G.M., Buetow, K.A., Waldrop, T.G., 1996. Identification of diencephalic and brainstem cardiorespiratory areas activated during exercise. Brain Res. 726, 109 –122. Kaada, B.R., 1951. Somato-motor, autonomic, and electrocorticographic responses to electrical stimulation of ‘rhinencephalic’ and other structures in primates, cats, and dogs. Acta. Physiol. Scand. 83, 38–89. Kaada, B.R., 1960. Cingulate, posterior orbital, anterior insular, and temporal pole cortex. In: Field, J., Magoun, H.W., Hall, V.E. (Eds.), Handbook of Physiology, Section 1: Neurophysiology, vol II. American Physiological Society, Bethesda, MD, pp. 1345–1372. Kabat, H., 1936. Electrical stimulation of points in the forebrain and midbrain: the resultant alterations in respiration. J. Comp. Neurol. 64, 187–208. Keller, A.D., 1960. Ablation and stimulation of the hypothalamus: circulatory effects. Physiol. Rev. 40, 116–135. Martin, H.N., Booker, W.D., 1878. The influence of stimulation of the midbrain upon the respiratory rhythm of the mammal. J. Physiol. (Lond.) 1, 370–376. Motekaitis, A.M., Kaufman, M.P., 1996. Stimulation of the mesencephalic locomotor region constricts the airways of cats. Respir. Physiol. 106, 263–271. Orem, J., Netick, A., 1982. Characteristics of midbrain respiratory neurons in sleep and wakefulness in the cat. Brain Res. 244, 231 – 241. Orlovskii, G.N., 1969. Spontaneous and induced locomotion of the thalamic cat. Biofizika 14, 1154–1162. Peano, C.A., Shonis, C.A., Dillon, G.H., Waldrop, T.G., 1992. Hypothalamic GABAergic mechanism involved in respiratory response to hypercapnia. Brain Res. Bull. 28, 107–113. Planche, D., Bianchi, A.L., 1972. Modification of bulbar respiratory neuron activity induced by cortical stimulation. J. Physiol. (Paris) 64, 69–76. Redgate, E.S., Gellhorn, E., 1958. Respiratory activity and the hypothalamus. Am. J. Physiol. 193, 189–194. Redgate, E.S., 1963. Hypothalamic influence on respiration. Ann. N.Y. Acad. Sci. 109, 606–618. Rikard-Bell, G.C., Bystrzycka, E.K., Nail, B.S., 1985. Cells of origin of corticospinal projections to phrenic and thoracic respiratory motoneurones in the cat as shown by retrograde transport of HRP. Brain Res. Bull. 14, 39–47. Ryan, J.W., Waldrop, T.G., 1995. Hypoxia sensitive neurons in the caudal hypothalamus project to the periaqueductal gray. Respir. Physiol. 100, 185–194. Schmid, K., Bo¨hmer, G., Fallert, M., 1988. Influence of rubrospinal tract and the adjacent mesencephalic reticular formation on the activity of medullary respiratory neurons and the phrenic nerve discharge in the rabbit. Pflugers Arch. 413, 23 – 31. Shekhar, A., DiMicco, J.A., 1987. Defense reaction elicited by injection of GABA antagonists and synthesis inhibitors into
211
the posterior hypothalamus in rats. Neuropharmacology 26, 407 – 417. Shik, M.L., Orlovskii, G.N., 1976. Neurophysiology of locomotor automatism. Physiol. Rev. 56, 465 – 501. Silveira, M.C., Graeff, F.G., 1992. Defense reaction elicited by microinjection of kainic acid into the medial hypothalamus of the rat: antagonism by a GABAA receptor agonist. Behav. Neural Biol. 57, 226 – 232. Spencer, W.G., 1894. The effect produced upon respiration by faradic excitation of the cerebrum in monkey, cat, dog, and rabbit. Philos. Trans. Royal Soc. Lond. 185, 609 – 660. Tappaz, M.L., Brownstein, M.J., 1977. Origin of glutamate-decarboxylase (GAD)-containing cells in discrete hypothalamic nuclei. Brain Res. 132, 95 – 106. Tenney, S.M., Ou, L.C., 1977. Ventilatory response of decorticate and decerebrate cats to hypoxia and CO2. Respir. Physiol. 29, 81 – 92. Thomas, M.R., Calaresu, F.R., 1972. Responses of single units in the medial hypothalamus to electrical stimulation of the carotid sinus nerve in the cat. Brain Res. 44, 49 – 62. Vertes, R.P., Crane, A.M., 1996. Descending projections of the posterior nucleus of the hypothalamus: phaseolus vulgaris leucoagglutinin analysis in the rat. J. Comp. Neurol. 374, 607 – 631. Vibert, J.F., Caille, D., Bertrand, F., Gromysz, H., Hugelin, A., 1979. Ascending projection from the respiratory centre to mesencephalon and diencephalon. Neurosci. Lett. 11, 29 – 33. Waldrop, T.G., Henderson, M.C., Iwamoto, G.A., Mitchell, J.H., 1986. Regional blood flow responses to stimulation of the subthalamic locomotor region. Respir. Physiol. 64, 93 – 102. Waldrop, T.G., Mullins, D.C., Henderson, M.C., 1986. Effects of hypothalamic lesions on the cardiorespiratory responses to muscular contraction. Respir. Physiol. 66, 215 – 224. Waldrop, T.G., Bauer, R.M., Iwamoto, G.A., 1988. Microinjection of GABA antagonists into the posterior hypothalamus elicits locomotor activity and a cardiorespiratory activation. Brain Res. 444, 84 – 94. Waldrop, T.G., Stremel, R.W., 1989. Muscular contraction stimulates posterior hypothalamic neurons. Am. J. Physiol. 256, 348 – 356. Waldrop, T.G., 1991. Posterior hypothalamic modulation of the respiratory response to CO2 in cats. Pflugers Arch. 418, 7 – 13. Waldrop, T.G., Porter, J.P., 1995. Hypothalamic involvement in respiratory and cardiovascular regulation. In: Dempsey, J.A., Pack, A.I. (Eds.), Regulation of Breathing 2nd ed. Marcel Dekker, New York, pp. 315 – 364. Waldrop, T.G., Eldridge, F.L., Iwamoto, G.A., Mitchell, J.H., 1996. Central neural control of respiration and circulation during exercise. In: Rowell, L.B., Shepherd, J.T. (Eds.), Handbook of Physiology, Section 12: Exercise: Regulation and Integration of Multiple Systems. American Physiological Society, New York, pp. 333 – 380. Yardley, C.P., Hilton, S.M., 1986. The hypothalamic and brainstem areas from which the cardiovascular and behavioural components of the defence reaction are elicited in the rat. J. Auton. Nerv. Syst. 15, 227 – 244.
Frontiers review
Suprapontine control of respiration Eric M. Horn, Tony G. Waldrop * Department of Molecular and Integrati6e Physiology, 524 Burrill Hall, 407 South Goodwin A6enue, Uni6ersity of Illinois, Urbana, IL 61803, USA Accepted 9 September 1998
Abstract Despite focus on brainstem areas in central respiratory control, regions rostral to the medulla and pons are now recognized as being important in modulating respiratory outflow during various physiological states. The focus of this review is to highlight the role that suprapontine areas of the mammalian brain play in ventilatory control mechanisms. New imaging techniques have become invaluable in confirming and broadening our understanding of the manner in which the cerebral cortex of humans contributes to respiratory control during volitional breathing. In the diencephalon, the integration of respiratory output in relation to changes in homeostasis occurs in the caudal hypothalamic region of mammals. Most importantly, neurons in this region are strongly sensitive to perturbations in oxygen tension which modulates their level of excitation. In addition, the caudal hypothalamus is a major site for ‘central command’, or the parallel activation of locomotion and respiration. Furthermore, midbrain regions such as the periaqueductal gray and mesencephalic locomotor region function in similar fashion as the caudal hypothalamus with regard to locomotion and more especially the defense reaction. Together these suprapontine regions exert a strong modulation upon the basic respiratory drive generated in the brainstem. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Control of breathing; Hypoxia; Brainstem; Suprapontine; Control of respiration; Suprapontine control
1. Introduction Even though the basic neuronal circuitry essential for the generation and maintenance of res* Corresponding author. Tel.: 1-217-2446037; Fax: 1-2172447078; E-mail: [email protected]
piration is located in the medulla and pons, more rostral areas of the brain are known to have a modulatory influence in respiratory control. In contrast to the necessity of the brainstem sites for generating the respiratory rhythm and its basic pattern, suprapontine areas can be lesioned and respiration stays relatively intact (Tenney and Ou,
0034-5687/98/$ - see front matter © 1998 Elsevier Science B.V. All rights reserved. PII: S 0 0 3 4 - 5 6 8 7 ( 9 8 ) 0 0 0 8 7 - 5
202
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
1977). However, suprapontine areas are known to be involved in adjusting respiration in response to alterations in internal or external environmental conditions. Without these sites, the complete respiratory responses elicited by exercise, hypoxia, hypercapnia, and thermal changes are impaired. Early studies employing electrical stimulation techniques demonstrated various suprapontine areas that increase respiratory drive when stimulated. A study by Spencer (1894) described changes in respiration caused by stimulation of forebrain sites, although the exact stereotaxic coordinates of these locations were not determined. One of the earliest studies to stereotaxically identify areas rostral to the pons that when stimulated elicit increases in respiration was performed by Kabat (1936). He identified the periaqueductal gray (PAG), caudal hypothalamus and subfornical regions as the areas that elicit the strongest increases in respiration when stimulated electrically. In contrast, work in the cortex demonstrated that virtually all of the areas stimulated in the telencephalon lead to a decrease in ventilation with the exception of the motor cortex (Kabat, 1936; Kaada, 1951). One major problem with all of these studies, however, is that electrical stimulation cannot definitively identify locations because fibers of passage are also stimulated. Hence, identifying the exact locations and interactions of areas involved in respiratory control had to wait until better methods were developed. Use of these new methods, such as chemical microinjections, single unit recordings, and functional imaging over the past three decades have provided a clearer understanding of the role of suprapontine areas in respiratory control. This review will focus primarily upon those regions in the telencephalon, diencephalon, and mesencephalon that provide a stimulatory influence over respiratory output.
2. Cortex Although respiration is one autonomic function that can be consciously controlled, the specific mechanisms of cortical control over ventilation remain to be elucidated. The cortex has both afferent input and descending projections that
enable a strong influence in respiratory control (see Davenport and Reep (1995) for more detailed review). Retrograde transport of horseradish peroxidase into spinal areas containing respiratory (phrenic and thoracic) motoneurons demonstrated that cortical inputs onto these neurons originate primarily in area 4 of the motor cortex (RikardBell et al., 1985). In addition, cortical inputs impinging upon midbrain PAG neurons arise predominately from the prefrontal cortex (Beitz, 1982). These limbic inputs are likely involved in the respiratory changes observed during different emotional and arousal states, although little evidence exists on the exact function of these connections. The major cortical inputs into the caudal hypothalamus originate from areas 4 and 6 of the motor cortex, suggesting that this area receives cortical respiratory output mainly associated with locomotion (Baev et al., 1985). Furthermore, single shock sensorimotor cortical stimulation has been shown to stimulate respiratory neurons in the medulla through suggested monosynaptic connections (Planche and Bianchi, 1972). Thus, both motor and limbic areas of the cortex have connections with the major regions involved in respiratory outflow. This variable input is to be expected considering that changes in respiratory output need to be matched to all arousal states. Upon stimulation, most areas of the cerebral cortex elicit decreases in respiratory frequency and depth (Spencer, 1894). This is in contrast to the motor controlling areas of the cortex, which are among the only cortical areas that elicit increases in ventilation when stimulated in mammals (Kaada, 1951, 1960). New imaging techniques such as functional magnetic resonance imaging (fMRI) and positron emission tomography (PET) have allowed the analysis of cortical control in conscious humans. Using PET scanning, areas such as the premotor, primary motor and supplemental motor cortices were shown to be active during volitional breathing at rest in human subjects (Colebatch et al., 1991). In addition to control of ventilation during resting conditions, several cortical areas have been identified that display increases in activity during forced or reflexive increases in ventilation. The use of PET
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
203
Fig. 1. Group results of relative regional cerebral blood flow increases associated with all volitional breathing conditions against added inspiratory loads, with unloaded volitional breathing as control. Data were averaged across all relevant runs for six subjects and are shown as projections onto representations of stereotactic space; areas achieving statistical significance (P B0.001) are shown together with areas in which there was a trend toward significance (*0.001B PB 0.01). L, left; R, right; A, anterior; P, posterior; VAC, vertical plane through anterior commissure; VPC, vertical plane through posterior commissure. Areas of activation are shown with increasing significance by an arbitrary gray scale ranging from light gray to black. Anatomic locations of areas of activation are indicated by lettering adjacent to corresponding region of interest: spr, superolateral precentral gyrus; ipr, inferolateral precentral gyrus; ipo, inferolateral postcentral gyrus; ic, intermediate cingulate; mp, medial parietal cortex; s, striatum; m, midbrain; pf, prefrontal cortex; v, vermis (*). For clarity not all areas of activation are labeled bilaterally. (Reprinted from Fink et al., 1996, with permission).
scanning demonstrated that the premotor, primary and supplemental motor areas are activated during periods of increased force of inspiration (Fig. 1) (Fink et al., 1996). Furthermore, a subcortical structure, the basal ganglia, shows increases in activation during increased inspiratory force as imaged by both PET and fMRI (Gozal et al., 1995; Fink et al., 1996). Hence, many structures in the cortex and subcortex traditionally involved with motor function are activated during ventilatory challenges. In addition, some areas more noted for learned motor behaviors were also activated in these studies, including the supplemental and premotor areas. This suggests that learned behavior is involved in
the sensation of changing respiratory loads and the subsequent ventilatory responses to these sensations.
3. Caudal hypothalamus
3.1. Basal respiratory dri6e The caudal hypothalamus has significant connections between midbrain and brainstem nuclei that control respiratory output. Strong descending projections from the caudal hypothalamus to the PAG and ventrolateral medulla, both areas that modulate respiratory outflow, were visualized
204
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
with the anterograde tracer phaseolus vulgaris leucoagglutinin (PHA) (Vertes and Crane, 1996). These descending projections are coupled with reciprocal projections back to the caudal hypothalamus from medullary respiratory areas but not from midbrain PAG sites (Vibert et al., 1979; Cameron et al., 1995a). The role that the caudal hypothalamus plays in basal respiratory drive has been investigated by several studies that blocked or abolished the activity of caudal hypothalamic neurons (see Waldrop and Porter, 1995). Some of these studies demonstrated that inactivation of caudal hypothalamic neurons by injection of barbiturates or placement of electrolytic lesions produces decreases in respiration (Redgate and Gellhorn, 1958; Keller, 1960; Waldrop et al., 1986b). These results indicate that caudal aspects of the hypothalamus provide an excitatory influence over basal ventilatory drive. The hypothalamus appears to be also involved in respiratory drive originating from other brain regions. One early study by Redgate (1963) sought to investigate this interaction by eliciting increases in ventilation through direct electrical stimulation of the medulla. He found that the ventilatory increases could be attenuated by either injecting barbiturates into or coagulating the caudal hypothalamus. Other investigators have further shown that sectioning of the brain rostral to the diencephalon increases ventilation while a midcollicular sectioning has no effect on ventilation, suggesting that areas rostral to the diencephalon exert an inhibitory influence on respiration that is balanced by facilatory drive coming directly from the diencephalon (Fink et al., 1962; Tenney and Ou, 1977).
3.2. Responses to hypoxia Strong evidence exists showing that the caudal hypothalamus modulates the respiratory responses to hypoxia. Tenney and Ou (1977) demonstrated that the respiratory responses to hypoxia are unchanged following midcollicular decerebration in the cat, but are greatly exaggerated following decortication. These results implicated diencephalic structures, which receive tonic
cortical inhibition during hypoxia, as playing a facilitatory role in the ventilatory responses to hypoxia. The role of hypothalamic areas in respiratory responses to hypoxia was further elucidated through the use of microinjection techniques that permit an evaluation of the neurotransmitters involved in this modulation. Horn and Waldrop (1994) demonstrated that the respiratory response to hypoxia is significantly attenuated following microinjection of either a synaptic blocker or an excitatory amino acid antagonist into the caudal hypothalamus (Fig. 2). In contrast, GABAergic activity in the caudal hypothalamus does not appear to be important since microinjection of a GABA synthesis inhibitor does not alter the response to hypoxia (Peano et al., 1992). These results suggest that neurons in the caudal hypothalamus receive excitatory input during hypoxia that facilitates the respiratory response to hypoxia. Several studies have employed electrophysiological techniques to elucidate the properties of caudal hypothalamic neurons involved in the response to hypoxia. Single units were first shown to be stimulated during hypoxia in this region by Cross and Silver (1963), who demonstrated an increase in firing frequency of a large percentage of neurons in the posterior and lateral hypothalamic nuclei. This work was extended by Dillon and Waldrop (1993) who determined that individual neurons in the caudal hypothalamus of anesthetized cats are stimulated by hypoxia and that a large portion of these cells have respiratory related discharge patterns. In addition, these neuronal responses to hypoxia in the cat are maintained following vagotomy and carotid sinus denervation, suggesting that peripheral chemoreceptor and baroreceptor inputs are not necessary for the hypoxic sensitivity of caudal hypothalamic neurons. Subsequent work in the rat provided similar results as in the cat and also demonstrated that many of these respiratory-related neurons that are stimulated by hypoxia in this area send projections to or through the midbrain PAG (Ryan and Waldrop, 1995). Furthermore, more than 75% of neurons in the caudal hypothalamus were shown
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
205
Fig. 2. Typical cardiovascular and respiratory (DEMG: integrated diaphragmatic EMG activity) responses to hypoxia before (A) and after (B) microinjection of kynurenic acid (KYN) into the caudal hypothalamus. Values for PO2 represent the levels delivered to the animal (Reprinted from Horn and Waldrop, 1994, with permission).
206
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
Fig. 3. (A) Response of a hypothalamic neuron to 10% O2 while perfused with control medium. (B) Response is maintained when synaptic transmission is blocked. Note decrease in spontaneous firing rate. (C) Return to control conditions resulted in recovery of spontaneous activity.
to be stimulated by hypoxia in an in vitro brain slice preparation and most of these neurons maintained this response during perfusion with a synaptic blockade medium (Fig. 3); but perfusion with a sodium channel blocker (tetrodotoxin) greatly attenuated the response (Dillon and Waldrop, 1992; Horn and Waldrop, 1997). These findings suggest that caudal hypothalamic neurons not only modulate the ventilatory response to hypoxia but also actively respond to changes in local oxygen concentration and thus may function as central oxygen chemoreceptors.
3.3. Responses to hypercapnia Caudal hypothalamic neuronal modulation of the respiratory responses to hypercapnia has also been investigated. Waldrop (1991) demonstrated that the respiratory response to hypercapnia is accentuated in the cat following microinjection of GABA antagonists into the caudal hypothalamus. A similar effect upon the respiratory response to hypercapnia was observed in the rat following microinjection of either GABA antagonists or synthesis inhibitors (Peano et al., 1992). Thus,
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
neurons in the caudal hypothalamus receive GABAergic input that modulates the ventilatory response to hypercapnic stimulation. Since most of the GABAergic input coming into the caudal hypothalamus is from local interneurons, this modulation is most likely due to a local circuit in this region (Tappaz and Brownstein, 1977). Additional support for caudal hypothalamic involvement in the response to hypercapnia has been provided by electrophysiological studies. An early study reported that single units recorded in this area are stimulated by hypercapnia in the rabbit brain (Cross and Silver, 1963). Recent extracellular studies in the rat and cat have demonstrated that a small subpopulation of caudal hypothalamic neurons are stimulated during hypercapnia both in vivo and in vitro (Cross and Silver, 1963; Dillon and Waldrop, 1992, 1993). These studies suggest that neurons in the caudal hypothalamus are responsive to hypercapnic stimulation, although whether this stimulation arises from CO2 or local pH changes is unknown.
4. Midbrain Electrical stimulation of many areas in the midbrain produces increases in respiration, which are due to increases in both the respiratory rate and depth (Martin and Booker, 1878). Kabat (1936) was the first to use stereotaxic techniques to determine the specific midbrain areas that elicit this stimulation. He demonstrated that various regions, including the PAG and tegmentum, are able to cause increases in respiration when stimulated. It has also been shown that electrical stimulation of the reticular formation in the midbrain alters both the discharge of respiratory-related units in the medulla and phrenic nerve output (Schmid et al., 1988). Furthermore, recording experiments have shown that neurons in midbrain areas have firing patterns temporally related to the respiratory cycle (Orem and Netick, 1982). The periaqueductal gray is an important midbrain area that modulates ventilation when stimulated, receives strong input from caudal hypothalamic neurons and projects to various
207
respiratory-related nuclei in the pons and medulla such as the Ko¨lliker–Fuse nucleus, ventrolateral medulla, and nucleus tractus solitarii (Cameron et al., 1995b). In one study, the effects of stimulating the PAG upon medullary neurons with a respiratory discharge were examined (Duffin and Hockman, 1972). PAG stimulation was found to switch medullary neurons from firing in the expiratory phase to firing in the inspiratory phase, thus demonstrating the strong modulatory control this region has over respiratory output. As mentioned previously, caudal hypothalamic neurons that respond to hypoxia project to the PAG (Ryan and Waldrop, 1995). Thus, it appears that the PAG may be a midbrain relay from the caudal hypothalamus region. In turn, PAG neurons send projections to medullary neurons that are responsible for altering respiratory drive. Hence, this circuitry is an important portion of the neuroaxis involved in respiratory and autonomic control. Another area involved in respiratory outflow is the midbrain tegmentum, which was investigated by Hugelin and colleagues. They recorded the activity of tegmentum neurons in a preparation (paralyzed, ventilated cats) which was designed to prevent the effects of locomotion that accompany stimulation of this area (Hugelin and Cohen, 1963). These recordings demonstrated that tegmental neurons are synchronized to the respiratory cycle via input from medullary neurons since this synchronization was lost following pontomesencephalic sectioning (Vibert et al., 1979). Work of Eldridge and his co-workers has further determined respiratory properties of specific neurons within the midbrain tegmentum. Neurons in this region possess respiratory-related rhythms when ventilatory drive is increased during hypercapnia but this rhythm is missing during basal respiration (Chen et al., 1991). Furthermore, the respiratory-related activity of these neurons is enhanced when the vagus nerve is either cooled or sectioned (Eldridge and Chen, 1992). These results suggest that neurons in the midbrain central tegmental field are not required for the generation or maintenance of respiratory drive, but do play a role in modulating increases in respiratory outflow.
208
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
Fig. 4. Cardiorespiratory responses to microinjection of picrotoxin (78 nl) into the caudal hypothalamus of a vagotomized cat after muscular paralysis. Control conditions are shown in the left panel. Microinjection of picrotoxin produced increases in arterial pressure, heart rate, phrenic nerve activity, and respiratory frequency. Note that these responses were reversed by the GABAA receptor agonist, muscimol (117 nl), injected into the same site (Reprinted from Waldrop et al., 1988, with permission).
In addition to unit recordings, PET scanning and fMRI imaging have shown the midbrain to be activated in humans when breathing hypercapnic mixtures (Gozal et al., 1994; Corfield et al., 1995). Unfortunately, due to the limited resolution of these techniques, specific nuclei in this area could not be analyzed. In addition, activation of the midbrain occurs during increases in inspiratory force caused by resistive loading in human subjects (Gozal et al., 1995; Fink et al., 1996). These studies highlight the fact that midbrain areas have a much higher level of activity during times of increased ventilatory outflow as compared to basal outflow.
5. Suprapontine respiratory control during co-ordinated activity
5.1. Locomotion A crucial role of the respiratory system is its involvement in meeting the increased metabolic demands associated with locomotion in mammals. The increased ventilation is necessary for adjusting oxygen delivery to the increased oxygen demand during locomotion. Exactly how the central nervous system controls the respiratory system in response to increased activity has long been stud-
ied (see Waldrop et al., 1996 for review). Both feed-forward and feedback mechanisms are now known to be important in this regulation (Eldridge and Waldrop, 1991). Orlovskii (1969) was one of the first investigators to demonstrate that stimulation of the caudal hypothalamus elicits locomotion and at the same time increases in respiration. His work suggested that there is a parallel activation of respiratory outflow and locomotion that compensates for the immediate oxygen needs of the animal. In subsequent studies, it was determined that the increases in respiration during stimulated locomotion can be elicited in a paralyzed, ventilated animal, in which feedback from pulmonary and muscle afferents are eliminated (Eldridge et al., 1981, 1985; Waldrop et al., 1988). The results of these studies demonstrated the presence of a ‘central command’ mechanism situated in the caudal hypothalamus that was responsible for the feed-forward activation of respiration during locomotion. In a subsequent study, microinjection of GABA antagonists into the caudal hypothalamus evoked results similar to those produced by electrical stimulation, thus ruling out the effects being due to activation of the electrical axons of passage (Fig. 4) (Waldrop et al., 1988). Recent work has further demonstrated that caudal hypothalamic neurons increase Fos expression following 45 min
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
of treadmill exercise, suggesting that these neurons are activated during exercise in conscious animals (Iwamoto et al., 1996). It is also important to note that sensory input from contracting muscles has been demonstrated to alter the activity of individual neurons in the caudal hypothalamus of anesthetized cats (Waldrop and Stremel, 1989). Thus, although the above results clearly demonstrate that neurons in the caudal hypothalamus can elicit the respiratory changes that are necessary during locomotion, the importance of peripheral input from muscles must also be considered. Another area that is known to elicit a ‘central command’ type response when stimulated is the mesencephalic locomotor region, MLR (Shik and Orlovskii, 1976). The MLR is similar to the caudal hypothalamus in that local stimulation of neurons in this area initiates locomotion with simultaneous increases in respiration and cardiovascular output. The exact influence on respiration of neurons in this area, however, is somewhat obscure. Although stimulation of neurons in this area causes increases in ventilation, a recent study demonstrated that total lung resistance is increased during stimulation of the MLR (Motekaitis and Kaufman, 1996). This is in contrast to the decreases in lung resistance which occur during muscular contraction alone or during spontaneous exercise. Thus, the exact role of the MLR in adjusting ventilation during locomotion remains to be elucidated.
5.2. Defence reaction One aspect of ventilatory control that is partly attributed to hypothalamic and midbrain influence is the so-called ‘defence reaction’, which is a set of well defined physiological and behavioral responses in awake animals occurring in response to a threatening stimulus. The physiological component consists of tachypnea, tachycardia, and a redistribution of blood flow from the viscera to skeletal muscle while the behavioral component consists of hypervigilance, locomotion, jumping and rearing (Hilton, 1982; Waldrop et al., 1986a; Yardley and Hilton, 1986; Duan et al., 1996). Early work looking at the role of the midbrain
209
and hypothalamus in this behavior determined that stimulation of various regions of the midbrain and hypothalamus is sufficient to elicit this response. Some of the strongest responses can be evoked by electrical stimulation of the midbrain PAG, midbrain tegmentum, dorsomedial hypothalamus, and posterior hypothalamic area (Hilton and Redfern, 1986; Yardley and Hilton, 1986). Microinjecting excitatory amino acids or GABA antagonists to stimulate the cell bodies while leaving fibers of passage unaffected provided evidence of the influence of specific hypothalamic neurons in the defence reaction. Both the physiological and behavioral responses to these injections are similar to those produced by electrical stimulation (Shekhar and DiMicco, 1987; Silveira and Graeff, 1992). Since stimulation of many of the regions that evoke locomotion also produces the defence reaction and many of the physiological responses are similar, the question arises of whether or not there is an anatomical difference in the locomotor and defence reaction sites. Existing evidence points to the same neurons controlling parts (i.e. tachypnea) of the response in both paradigms, but that the overall pattern of behavioral and physiological responses is different depending upon whether locomotion or the defence reaction is initiated. How the hypothalamus integrates information leading to the defence reaction was first studied in the early 1960’s by demonstrating that stimulation of the carotid chemoreceptors elicits the defence response in cats that are decorticated just rostral to the optic chiasm (Bizzi et al., 1961). Since input from the carotid sinus nerve is known to influence neurons in the caudal hypothalamus, the activation of the defence response by chemoreceptor stimulation is possibly mediated through the caudal hypothalamus (Thomas and Calaresu, 1972).
6. Conclusions Elimination of suprapontine sites show only minor effects on resting respiration which gives the impression that these areas are not crucial in respiratory control. However, the contribution of these areas is not in basal respiratory generation
210
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211
but more in the control of respiration during various conditions such as locomotion, hypoxia, hypercapnia and the defence reaction. Hypothalamic and midbrain areas are richly connected to higher functioning areas, including areas involved in conscious motor control such as the cerebral cortex. These anatomical and function connections suggest that suprapontine centers are involved in the more complicated tasks of receiving sensory information concerning respiratory needs and in turn modulating the outflow of ventilation to match these needs. These tasks are further complicated by the conscious control over ventilation that must be factored into the integration of the ventilatory output during these conditions. Since most of the work in the control of ventilation has focused on medullary and pontine regions, a more thorough examination of suprapontine areas is needed before a complete understanding of ventilatory control can be attained.
References Baev, K.V., Berezovskii, V.K., Kebkalo, T.G., Savos’kina, L.A., 1985. Projection of forebrain structures of the cat to the hypothalamic locomotor area. Neirofiziologiia 17, 255 – 263. Beitz, A.J., 1982. The organization of afferent projections to the midbrain periaqueductal gray of the rat. Neuroscience 7, 133 – 159. Bizzi, E., Libretti, A., Malliani, A., Zanchetti, A., 1961. Reflex chemoreceptive excitation of diencephalic sham rage behaviour. Am. J. Physiol. 200, 923–926. Cameron, A.A., Khan, I.A., Westlund, K.N., Cliffer, K.D., Willis, W.D., 1995. The efferent projections of the periaqueductal gray in the rat: a phaseolus vulgaris-leucoagglutinin study I. Ascending projections. J. Comp. Neurol. 351, 568 – 584. Cameron, A.A., Khan, I.A., Westlund, K.N., Willis, W.D., 1995. The efferent projections of the periaqueductal gray in the rat: a phaseolus vulgaris-leucoagglutinin study II. Descending projections. J. Comp. Neurol. 351, 585–601. Chen, Z., Eldridge, F.L., Wagner, P.G., 1991. Respiratory-associated rhythmic firing of midbrain neurones in cats: relation to level of respiratory drive. J. Physiol. (Lond.) 437, 305 – 325. Colebatch, J.G., Adams, L., Murphy, K., 1991. Regional cerebral blood flow during volitional breathing in man. J. Physiol. (Lond.) 443, 91–103. Corfield, D.R., Fink, G.R., Ramsay, S.C., 1995. Evidence for limbic system activation during CO2-stimulated breathing in man. J. Physiol. (Lond.) 488, 77–84.
Cross, B.A., Silver, I.A., 1963. Unit activity in the hypothalamus and the sympathetic response to hypoxia and hypercapnia. Exp. Neurol. 7, 375 – 393. Davenport, P.W., Reep, R.L., 1995. Cerebral cortex and respiration. In: Dempsey, J.A., Pack, A.I. (Eds.), Regulation of Breathing. Marcel Dekker, New York, pp. 365 – 388. Dillon, G.H., Waldrop, T.G., 1992. In vitro responses of caudal hypothalamic neurons to hypoxia and hypercapnia. Neuroscience 51, 941 – 950. Dillon, G.H., Waldrop, T.G., 1993. Responses of feline caudal hypothalamic cardiorespiratory neurons to hypoxia and hypercapnia. Exp. Brain. Res. 96, 260 – 272. Duan, Y.F., Winters, R., McCabe, P.M., Green, E.J., Huang, Y., Schneiderman, N., 1996. Behavioral characteristics of defence and vigilance reactions elicited by electrical stimulation of the hypothalamus in rabbits. Behav. Brain. Res. 81, 33 – 41. Duffin, J., Hockman, C.H., 1972. Limbic forebrain and midbrain modulation and phase-switching of expiratory neurons. Brain Res. 39, 235 – 239. Eldridge, F.L., Millhorn, D.E., Waldrop, T.G., 1981. Exercise hyperpnea and locomotion: parallel activation from the hypothalamus. Science 211, 844 – 846. Eldridge, F.L., Millhorn, D.E., Kiley, J.P., Waldrop, T.G., 1985. Stimulation by central command of locomotion, respiration, and circulation during exercise. Respir. Physiol. 59, 313 – 337. Eldridge, F.L., Waldrop, T.G., 1991. Neural control of breathing during exercise. In: Whipp, B., Wasserman, K. (Eds.), Exercise: Pulmonary Physiology and Pathophysiology. Marcel Dekker, New York, pp. 309 – 370. Eldridge, F.L., Chen, Z., 1992. Respiratory-associated rhythmic firing of midbrain neurons is modulated by vagal input. Respir. Physiol. 90, 31 – 46. Fink, B.R., Katz, R., Reinhold, H., Schoolman, A., 1962. Suprapontine mechanisms in regulation of respiration, Am. J. Physiol. 202, 217 – 220. Fink, G.R., Corfield, D.R., Murphy, K., Kobayashi, I., Dettmers, C., Adams, L., Frackowiak, R.S., Guz, A., 1996. Human cerebral activity with increasing inspiratory force: a study using positron emission tomography. J. Appl. Physiol. 81, 1295 – 1305. Gozal, D., Hathout, G.M., Kirlew, K.A., 1994. Localization of putative neural respiratory regions in the human by functional magnetic resonance imaging. J. Appl. Physiol. 76, 2076 – 2083. Gozal, D., Omidvar, O., Kirlew, K.A., 1995. Identification of human brain regions underlying responses to resistive inspiratory loading with functional magnetic resonance imaging. Proc. Natl. Acad. Sci. U.S.A. 92, 6607 – 6611. Hilton, S.M., 1982. The defence-arousal system and its relevance for circulatory and respiratory control. J. Exp. Biol. 100, 159 – 174. Hilton, S.M., Redfern, W.S., 1986. A search for brain stem cell groups integrating the defence reaction in the rat. J. Physiol. (Lond.) 378, 213 – 228. Horn, E.M., Waldrop, T.G., 1994. Modulation of the respiratory responses to hypoxia and hypercapnia by synaptic input onto caudal hypothalamic neurons. Brain Res. 664, 25 – 33.
E.M. Horn, T.G. Waldrop / Respiration Physiology 114 (1998) 201–211 Horn, E.M., Waldrop, T.G., 1997. Oxygen-sensing neurons in the caudal hypothalamus and their role in cardiorespiratory control. Respir. Physiol. 110, 219–228. Hugelin, A., Cohen, M.I., 1963. The reticular activating system and respiratory regulation in the cat. Ann. N.Y. Acad. Sci. 109, 586 – 603. Iwamoto, G.A., Wappel, S.M., Fox, G.M., Buetow, K.A., Waldrop, T.G., 1996. Identification of diencephalic and brainstem cardiorespiratory areas activated during exercise. Brain Res. 726, 109 –122. Kaada, B.R., 1951. Somato-motor, autonomic, and electrocorticographic responses to electrical stimulation of ‘rhinencephalic’ and other structures in primates, cats, and dogs. Acta. Physiol. Scand. 83, 38–89. Kaada, B.R., 1960. Cingulate, posterior orbital, anterior insular, and temporal pole cortex. In: Field, J., Magoun, H.W., Hall, V.E. (Eds.), Handbook of Physiology, Section 1: Neurophysiology, vol II. American Physiological Society, Bethesda, MD, pp. 1345–1372. Kabat, H., 1936. Electrical stimulation of points in the forebrain and midbrain: the resultant alterations in respiration. J. Comp. Neurol. 64, 187–208. Keller, A.D., 1960. Ablation and stimulation of the hypothalamus: circulatory effects. Physiol. Rev. 40, 116–135. Martin, H.N., Booker, W.D., 1878. The influence of stimulation of the midbrain upon the respiratory rhythm of the mammal. J. Physiol. (Lond.) 1, 370–376. Motekaitis, A.M., Kaufman, M.P., 1996. Stimulation of the mesencephalic locomotor region constricts the airways of cats. Respir. Physiol. 106, 263–271. Orem, J., Netick, A., 1982. Characteristics of midbrain respiratory neurons in sleep and wakefulness in the cat. Brain Res. 244, 231 – 241. Orlovskii, G.N., 1969. Spontaneous and induced locomotion of the thalamic cat. Biofizika 14, 1154–1162. Peano, C.A., Shonis, C.A., Dillon, G.H., Waldrop, T.G., 1992. Hypothalamic GABAergic mechanism involved in respiratory response to hypercapnia. Brain Res. Bull. 28, 107–113. Planche, D., Bianchi, A.L., 1972. Modification of bulbar respiratory neuron activity induced by cortical stimulation. J. Physiol. (Paris) 64, 69–76. Redgate, E.S., Gellhorn, E., 1958. Respiratory activity and the hypothalamus. Am. J. Physiol. 193, 189–194. Redgate, E.S., 1963. Hypothalamic influence on respiration. Ann. N.Y. Acad. Sci. 109, 606–618. Rikard-Bell, G.C., Bystrzycka, E.K., Nail, B.S., 1985. Cells of origin of corticospinal projections to phrenic and thoracic respiratory motoneurones in the cat as shown by retrograde transport of HRP. Brain Res. Bull. 14, 39–47. Ryan, J.W., Waldrop, T.G., 1995. Hypoxia sensitive neurons in the caudal hypothalamus project to the periaqueductal gray. Respir. Physiol. 100, 185–194. Schmid, K., Bo¨hmer, G., Fallert, M., 1988. Influence of rubrospinal tract and the adjacent mesencephalic reticular formation on the activity of medullary respiratory neurons and the phrenic nerve discharge in the rabbit. Pflugers Arch. 413, 23 – 31. Shekhar, A., DiMicco, J.A., 1987. Defense reaction elicited by injection of GABA antagonists and synthesis inhibitors into
211
the posterior hypothalamus in rats. Neuropharmacology 26, 407 – 417. Shik, M.L., Orlovskii, G.N., 1976. Neurophysiology of locomotor automatism. Physiol. Rev. 56, 465 – 501. Silveira, M.C., Graeff, F.G., 1992. Defense reaction elicited by microinjection of kainic acid into the medial hypothalamus of the rat: antagonism by a GABAA receptor agonist. Behav. Neural Biol. 57, 226 – 232. Spencer, W.G., 1894. The effect produced upon respiration by faradic excitation of the cerebrum in monkey, cat, dog, and rabbit. Philos. Trans. Royal Soc. Lond. 185, 609 – 660. Tappaz, M.L., Brownstein, M.J., 1977. Origin of glutamate-decarboxylase (GAD)-containing cells in discrete hypothalamic nuclei. Brain Res. 132, 95 – 106. Tenney, S.M., Ou, L.C., 1977. Ventilatory response of decorticate and decerebrate cats to hypoxia and CO2. Respir. Physiol. 29, 81 – 92. Thomas, M.R., Calaresu, F.R., 1972. Responses of single units in the medial hypothalamus to electrical stimulation of the carotid sinus nerve in the cat. Brain Res. 44, 49 – 62. Vertes, R.P., Crane, A.M., 1996. Descending projections of the posterior nucleus of the hypothalamus: phaseolus vulgaris leucoagglutinin analysis in the rat. J. Comp. Neurol. 374, 607 – 631. Vibert, J.F., Caille, D., Bertrand, F., Gromysz, H., Hugelin, A., 1979. Ascending projection from the respiratory centre to mesencephalon and diencephalon. Neurosci. Lett. 11, 29 – 33. Waldrop, T.G., Henderson, M.C., Iwamoto, G.A., Mitchell, J.H., 1986. Regional blood flow responses to stimulation of the subthalamic locomotor region. Respir. Physiol. 64, 93 – 102. Waldrop, T.G., Mullins, D.C., Henderson, M.C., 1986. Effects of hypothalamic lesions on the cardiorespiratory responses to muscular contraction. Respir. Physiol. 66, 215 – 224. Waldrop, T.G., Bauer, R.M., Iwamoto, G.A., 1988. Microinjection of GABA antagonists into the posterior hypothalamus elicits locomotor activity and a cardiorespiratory activation. Brain Res. 444, 84 – 94. Waldrop, T.G., Stremel, R.W., 1989. Muscular contraction stimulates posterior hypothalamic neurons. Am. J. Physiol. 256, 348 – 356. Waldrop, T.G., 1991. Posterior hypothalamic modulation of the respiratory response to CO2 in cats. Pflugers Arch. 418, 7 – 13. Waldrop, T.G., Porter, J.P., 1995. Hypothalamic involvement in respiratory and cardiovascular regulation. In: Dempsey, J.A., Pack, A.I. (Eds.), Regulation of Breathing 2nd ed. Marcel Dekker, New York, pp. 315 – 364. Waldrop, T.G., Eldridge, F.L., Iwamoto, G.A., Mitchell, J.H., 1996. Central neural control of respiration and circulation during exercise. In: Rowell, L.B., Shepherd, J.T. (Eds.), Handbook of Physiology, Section 12: Exercise: Regulation and Integration of Multiple Systems. American Physiological Society, New York, pp. 333 – 380. Yardley, C.P., Hilton, S.M., 1986. The hypothalamic and brainstem areas from which the cardiovascular and behavioural components of the defence reaction are elicited in the rat. J. Auton. Nerv. Syst. 15, 227 – 244.