Neurophysiological investigations of medullary chemosensitive areas of respiration

Regulation of breathing and autonomic outflows by chemoreceptors

Lung ventilation fluctuates widely with behavior but arterial PCO2 remains stable. Under normal conditions, the chemoreflexes contribute to PaCO2 stability by producing small corrective cardiorespiratory adjustments mediated by lower brainstem circuits. Carotid body (CB) information reaches the respiratory pattern generator (RPG) via nucleus solitarius (NTS) glutamatergic neurons which also target rostral ventrolateral medulla (RVLM) presympathetic neurons thereby raising sympathetic nerve activity (SNA). Chemoreceptors also regulate presympathetic neurons and cardiovagal preganglionic neurons indirectly via inputs from the RPG. Secondary effects of chemoreceptors on the autonomic outflows result from changes in lung stretch afferent and baroreceptor activity. Central respiratory chemosensitivity is caused by direct effects of acid on neurons and indirect effects of CO2 via astrocytes. Central respiratory chemoreceptors are not definitively identified but the retrotrapezoid nucleus (RTN) is a particularly strong candidate. The absence of RTN likely causes severe central apneas in congenital central hypoventilation syndrome. Like other stressors, intense chemosensory stimuli produce arousal and activate circuits that are wake- or attention-promoting. Such pathways (e.g., locus coeruleus, raphe, and orexin system) modulate the chemoreflexes in a state-dependent manner and their activation by strong chemosensory stimuli intensifies these reflexes. In essential hypertension, obstructive sleep apnea and congestive heart failure, chronically elevated CB afferent activity contributes to raising SNA but breathing is unchanged or becomes periodic (severe CHF). Extreme CNS hypoxia produces a stereotyped cardiorespiratory response (gasping, increased SNA). The effects of these various pathologies on brainstem cardiorespiratory networks are discussed, special consideration being given to the interactions between central and peripheral chemoreflexes.

Central chemoreception 2005: A brief review

This brief review will place recent findings on specific neurons and receptors identified as putative central chemoreceptors, namely glutamatergic and serotonergic neurons and purogenic receptors, into the context of our working hypothesis that central chemoreception is a distributed property.

Chemosensitivity of serotonergic neurons in the rostral ventral medulla

The medullary raphé contains two subtypes of chemosensitive neuron: one that is stimulated by acidosis and another that is inhibited. Both types of neuron are putative chemoreceptors, proposed to act in opposite ways to modulate respiratory output and other pH sensitive brain functions. In this review, we will discuss the cellular properties of these chemosensitive raphé neurons when studied in vitro using brain slices and primary dissociated cell culture. Quantification of chemosensitivity of raphé neurons indicates that they are highly sensitive to small changes in extracellular pH (pH(o)) between 7.2 and 7.6. Stimulation by acidosis occurs only in the specific phenotypic subset of neurons within the raphé that are serotonergic. These serotonergic neurons also have other properties consistent with a specialized role in chemoreception. Homologous serotonergic neurons are present within the ventrolateral medulla (VLM), and may have contributed to localization of respiratory chemoreception to that region. Chemosensitivity of raphé neurons increases in the postnatal period in rats, in parallel with development of respiratory chemoreception in vivo. An abnormality of serotonergic neurons of the ventral medulla has been identified in victims of sudden infant death syndrome (SIDS). The cellular properties of serotonergic raphé neurons suggest that they play a role in the CNS response to hypercapnia, and that they may contribute to interactions between the sleep/wake cycle and respiratory control.

CO2, brainstem chemoreceptors and breathing

The regulation of breathing relies upon chemical feedback concerning the levels of CO2 and O2. The carotid bodies, which detect O2, provide tonic excitation to brainstem respiratory neurons under normal conditions and dramatic excitation if O2 levels fall. Feedback for CO2 involves the carotid body and receptors in the brainstem, central chemoreceptors. Small increases in CO2 produce large increases in breathing. Decreases in CO2 below normal can, in sleep and anesthesia, decrease breathing, even to apnea. Central chemoreceptors, once thought localized to the surface of the ventral medulla, are likely distributed more widely with sites presently identified in the: (1) ventrolateral medulla; (2) nucleus of the solitary tract; (3) ventral respiratory group; (4) locus ceruleus; (5) caudal medullary raphé; and (6) fastigial nucleus of the cerebellum. Why so many chemoreceptor sites? Hypotheses, some with supporting data, include the following. Geographical specificity; all regions of the brainstem with respiratory neurons contain chemoreceptors. Stimulus intensity; some sites operate in the physiological range of CO2 values, others only with more extreme changes. Stimulus specificity; CO2 or pH may be sensed by multiple mechanisms. Temporal specificity; some sites respond more quickly to changes on blood or brain CO2 or pH. Syncytium; chemosensitive neurons may be connected via low resistance, gap junctions. Arousal state: sites may vary in effectiveness and importance dependent on state of arousal. Overall, as judged by experiments of nature, and in the laboratory, central chemoreceptors are critical for adequate breathing in sleep, but other aspects of the control system can maintain breathing in wakefulness.

Focal central chemoreceptor sensitivity in the RTN studied with a CO2 diffusion pipette in vivo

We describe and use a CO2 diffusion pipette to produce a quickly reversible focal acidosis in the retrotrapezoid nucleus region of the rat brain stem. No tissue injection is made. Instead, artificial cerebrospinal fluid (aCSF) equilibrated with CO2 circulates within the micropipette, providing a source for continued CO2 diffusion into the tissue from the pipette tip. Tissue pH electrodes show the acidosis is limited to 500 micron from the tip. In controls (aCSF equilibrated with air), 1-min pipette perfusions increased tissue pH slightly and decreased phrenic nerve amplitude. In moderate- and high-CO2 groups (aCSF equilibrated with 50 or 100% CO2), 1-min perfusions significantly decreased tissue pH and increased phrenic nerve amplitude in a dose-dependent manner. The responses developed and reversed within minutes. Compared with our prior use of medullary acetazolamide injections to produce a focal acidosis, in this approach the acidosis 1) arises and reverses quickly and 2) its intensity can be varied. This allows study of sensitivity and mechanism. We conclude from this initial experiment that retrotrapezoid nucleus region chemoreceptors operate within the normal physiological range of CO2-induced tissue pH changes.

Effect of sinus denervation and vagotomy on c-fos expression in the nucleus tractus solitarius after exposure to CO2

Exposure to hypercapnia and electrical stimulation of the carotid sinus nerve (CSN) has been shown to induce c-fos expression in several brain stem regions including the nucleus tractus solitarius (NTS). To test whether the labeled neurons were activated directly by hypercapnia or secondarily via the carotid bodies (sinus nerve), adult rats were exposed to either air or 14-16% CO2 for 1 h. Experiments were done on eight groups: (1) exposure to air, (2) exposure to CO2, (3) chronic CSN denervation/CO2, (4) chronic unilateral CSN denervation/CO2, (5) chronic sham CSN denervation/CO2, (6) anesthetized/CO2, (7) anesthetized and acute vagotomy/CO2, and (8) premedicated with morphine, 10 mg s.c., 20 min before exposure to CO2. After exposure to CO2 or air the rats were anesthetized, perfused with 4% paraformaldehyde and the brains processed for immunohistochemical staining for c-fos protein using the PAP (i.e. peroxidase anti-peroxidase) technique. Labeled neurons in the area of the NTS in every second 50- "mu"m section were counted and their position plotted using a microscope and camera lucida attachment. Rats exposed to CO2 had a significantly greater number of labeled neurons in the NTS than those exposed to air. Other interventions, such as CSN denervation, surgery, anesthesia, vagotomy or injection of morphine did not significantly affect the level of c-fos expression in rats exposed to hypercapnia, indicative of central stimulation rather than secondary peripheral input. These responsive neurons may be part of a widespread central chemoreceptive complex.

In vivo and in vitro responses of neurons in the ventrolateral medulla to hypoxia

Neurons in the ventrolateral medulla (VLM) are known to be involved in several cardiorespiratory reflexes and to provide tonic drive to sympathetic preganglionic neurons. Recent studies have suggested that VLM neurons modulate the respiratory responses to hypoxia and to hypercapnia. The purpose of the present study was to determine with electrophysiological techniques if the discharge of these neurons is altered by hypoxia and/or by hypercapnia both in vivo and in vitro. Extracellular single-unit activity of VLM neurons (n = 39) was recorded during inhalation of a hypoxic gas (10% O2) and during inhalation of a hypercapnic gas (5% CO2) in anesthetized, spontaneously breathing rats (n = 16). Hypoxia elicited an increase in the discharge frequency in 64% of the VLM neurons studied; hypercapnia stimulated 42% of the neurons. Fifty-two percent of the neurons were stimulated by both hypoxia and hypercapnia. Signal averaging revealed that 76% of the hypoxia-stimulated neurons had a resting discharge related to the cardiac and/or respiratory cycle. Similar percentages of VLM neurons (35/54) were stimulated by hypoxia in a second group of animals (n = 14) that were studied after sinoaortic denervation. A rat brain slice preparation was then used to determine if hypoxia exerts a direct effect upon neurons in the VLM. Perfusing a hypoxic gas over the surface of medullary slices evoked an increase in the discharge frequency in the majority (39/49) of VLM neurons studied; responses were graded in relation to the magnitude of the hypoxic stimulus. Similar responses to hypoxia were observed in VLM neurons studied during perfusion with a synaptic blockade medium. Retrograde labeling of VLM neurons with rhodamine tagged microspheres injected into the thoracic intermediolateral cell column demonstrated that the hypoxia sensitive neurons were located in a region of the VLM that projects to the thoracic spinal cord. These results demonstrate that neurons in the ventrolateral medulla are excited by a direct effect of hypoxia; these neurons may play a critical role in the cardiorespiratory responses to hypoxia.

Responses of respiratory modulated and tonic units in the retrotrapezoid nucleus to CO2

We hypothesized that the retrotrapezoid nucleus (RTN) contains both respiratory modulated (RM) and non-respiratory modulated (NRM) neurons which participate in the ventilatory response to increased CO2. We made extracellular recordings of the activity of 46 single units in the RTN of 9 decerebrate, paralyzed, ventilated cats (5 intact; 4 with carotid body and sinus ablation) under eucapnic (PCO2 = 34.2 +/- 3.5 mmHg; mean +/- SD) and hypercapnic (PCO2 = 47.4 +/- 3.4 conditions. To define a RM unit, we used the eta 2 statistic which is the ratio of the variance of the unit firing rate within respiratory cycles to that across respiratory cycles. We classified the units as RM (N = 17) if the eta 2 values in eucapnia or hypercapnia were > or = 0.25 and as NRM (N = 29) if the values were < 0.25. Overall, 19/46 units (41%) increased their firing rate with increased CO2, 5 decreased their firing rate, and 22 had no significant change in firing rate. Of 17 RM units, 8 (47%) increased their mean firing rate with hypercapnia from 7.6 +/- 3.9 to 23.2 +/- 6.8 spikes/sec. These included 5 inspiratory units, 2 inspiratory units that had an onset of firing in late expiration (Pre-I/I), and 1 expiratory unit. Seven of these also changed their discharge pattern (eucapnic eta 2 = 0.02 to 0.12; hypercapnic eta 2 = 0.34 to 0.79) Of 29 NRM units, 11 (38%) showed a significant increase in mean firing rate with CO2 stimulation from 19.8 +/- 7.2 to 31.3 +/- 8.2 spikes/sec. The RTN has RM units which change their discharge pattern and firing rate in response to increased CO2, as do units within the medulla and pons, and it has NRM units which are also responsive to increased CO2. These data indicate that some neurons of the RTN are involved in the central chemoreceptor response but they provide no direct evidence that chemoreception resides within the RTN.

Hypercapnia and medullary neurons in the isolated brain stem-spinal cord of the rat

We have extracellularly recorded single neuron activity in the ventral medulla of the isolated brain stem-spinal cord preparation of the neonatal rat (37 preparations) in order to test their sensitivity to changes in CO2/H+. Search for neuronal activity was performed when the preparation was superfused with control mock CSF (equilibrated with 2% CO2, 90% O2 in N2; pH = 7.8 at 27 degrees C). Neurons, found down to about 500 microns from the surface, could be classified as R neurons when they showed rhythmic discharge in phase with phrenic activity, recorded from C4 ventral roots; or as Non-R neurons when they did not exhibit such phasic discharge. Among the 89 Non-R neurons, 20 responded to rapidly replacing the control CSF by hypercapnic CSF (8% CO2, 90% O2 in N2; pH = 7.2) with increased, 44 with reduced activity, while 25 did not respond to hypercapnia. Five Non-R neurons became phasic with respiration during hypercapnia. Of the 14 R neurons, 10 fired predominantly in expiration (R-E), 4 in inspiration (R-I). Only one R-E and two R-I neurons were excited by hypercapnia, the remaining were either inhibited or did not respond. Excited Non-R and R neurons were mainly encountered in rostral parts of those areas in the ventral medulla that have been reported as chemosensitive.

A theoretical model of gas transport between arterioles and tissue

A theoretical model of CO2 and O2 diffusion between arterioles and tissue was developed to determine if significant transport could occur in precapillary vessels. There is increasing evidence, both theoretical and experimental, that such exchange does occur. Using a model in which CO2 and O2 were coupled through the Bohr and Haldane effects, we quantified the radial and axial transport. We also examined the roles of axial diffusion in the arteriole wall and tissue and capillary structure on the transport. Capillary arrangements investigated included capillaries independent of the arteriole with the entering capillary PCO2 or PO2 equal to a constant, and capillaries branching off along the length of the arteriole with the entering capillary partial pressure equal to the arteriole partial pressure at the given axial location. We found that for CO2 in arterioles with an inner diameter ranging from 200 to 22 microns, the exiting blood was 6 to 45% of the way to complete equilibrium with the surrounding tissue, respectively. For O2, the range was 8 to 25%, respectively. We also determined that axial diffusion in the arteriole wall and tissue has little effect on the transport and that capillary structure can alter tissue PCO2 by as much as 12 mm Hg in the smallest arteriole, but has little effect on O2 transport.

Acetylcholine and central chemosensitivity: in vitro study in the newborn rat

In vitro experiments were performed in the superfused brainstem-spinal cord preparation of newborn rats in order to analyse the central respiratory effects of acetylcholine. The central motor output was assessed from recording electrical activity in nerves supplying respiratory muscles. Acetylcholine added to the bathing medium induced dose-dependent increases in respiratory frequency which were blocked by muscarinic (but not nicotinic) antagonists and enhanced by physostigmine. These effects originated from the medullary ventral surface where chemosensitive structures have been previously located. The respiratory central chemosensitivity of the isolated brainstem was analysed using a CO2 free, pH 7.9 medium instead of the normal medium (bubbled with 5% CO2, pH 7.3). Decreases at the H+ and CO2 stimuli led to decreased inspiratory activity, resulting mainly from a decrease in the amplitude of the motor output. These responses were enhanced by atropine and diminished by physostigmine. These results obtained in vitro on the newborn rat suggest that cholinergic synapses are not directly involved in the genesis of respiratory rhythmicity but confirm previous results obtained in vivo in adult animal revealing that acetylcholine is implicated in the central respiratory chemosensitivity.

Neurons sensitive to pH in slices of the rat ventral medulla oblongata

The effects of extracellular pH changes on neurons in slices of the rat ventral medulla oblongata were investigated by extracellular recording. Changes in discharge rate were correlated with pH changes in the tissue next to the recorded cell, as measured by H(+)-selective microelectrodes. pH was altered by varying the bicarbonate concentration ([HCO3-]) in the superfusion solution. In 136 out of 316 neurons, the number of spontaneous or electrically evoked discharges per unit time increased with decreasing pH and decreased with increasing pH. Changes of only 0.01-0.04 pH unit were effective in these pH-sensitive neurons. The response was transient; the discharge rate returned to the control value within a few minutes. The pH sensitivity persisted in the presence of 0.5 microM atropine, 20 microM bicuculline and after replacing Ca2+ by Mg2+ in the superfusion solution to reduce synaptic transmission. The response to the same pH decrease was stronger when increasing PCO2 than when reducing [HCO3-]0. The pH-induced response significantly increased during hypoxia. The results show that in the ventral medulla oblongata neurons exist that transiently respond to small decreases and increases of pH. The pH sensitivity is an intrinsic property of these neurons; it is not due to a synaptic mechanism but is modulated by PCO2 and PO2.

Depolarization and stimulation of neurons in nucleus tractus solitarii by carbon dioxide does not require chemical synaptic input

The effects of elevated CO2 (i.e. hypercapnia) on neurons in the nucleus tractus solitarii were studied using extracellular (n = 82) and intracellular (n = 33) recording techniques in transverse brain slices prepared from rat. Synaptic connections from putative chemosensitive neurons in the ventrolateral medulla were removed by bisecting each transverse slice and discarding the ventral half. In addition, the response to hypercapnia in 20 neurons was studied during high magnesium-low calcium synaptic blockade. Sixty-five per cent of the neurons (n = 75) tested were either insensitive or inhibited by hypercapnia. However, 35% (n = 40) were depolarized and/or increased their firing rate during hypercapnia. Nine out of 10 CO2-excited neurons retained their chemosensitivity to CO2 in the presence of high magnesium-low calcium synaptic blockade medium. Our findings demonstrate that many neurons in the nucleus tractus solitarii were depolarized and/or increased their firing rate during hypercapnia. These neurons were not driven synaptically by putative chemosensitive neurons of the ventrolateral medulla since this region was removed from the slice. Furthermore, because chemosensitivity persisted in most neurons tested during synaptic blockade, we conclude that some neurons in the nucleus tractus solitarii are inherently CO2-chemosensitive. Although the function of dorsal medullary chemosensitive neurons cannot be determined in vitro, their location and their inherent chemosensitivity suggest a role in cardiorespiratory central chemoreception.

CO2 decreases membrane conductance and depolarizes neurons in the nucleus tractus solitarii

To identify central sites of potential CO2/H+-chemoreceptive neurons, and the mechanism responsible for neuronal chemosensitivity, intracellular recordings were made in rat tissue slices in two cardiopulmonary-related regions (i.e., nucleus tractus solitarii, NTS; nucleus ambiguus, AMBc) during exposure to high CO2. When the NTS was explored slices were bisected and the ventral half discarded. Utilizing such "dorsal" medullary slices removed any impinging synaptic input from putative chemoreceptors in the ventrolateral medulla. In the NTS, CO2-induced changes in firing rate were associated with membrane depolarizations ranging from 2-25 mV (n = 15). In some cases increased e.p.s.p. activity was observed during CO2 exposure. The CO2-induced depolarization occurred concomitantly with an increased input resistance ranging from 19-23 M omega (n = 5). The lower membrane conductance during hypercapnia suggests that CO2-induced depolarization is due to a decreased outward potassium conductance. Unlike neurons in the NTS, AMBc neurons were not spontaneously active and were rarely depolarized by hypercapnia. Eleven of 12 cells tested were either hyperpolarized by or insensitive to CO2. Only 1 neuron in the AMBc was depolarized and it also showed an increased input resistance during CO2 exposure. Our findings suggest that CO2/H+-related stimuli decrease potassium conductance which depolarizes the cell and increases firing rate. Although our in vitro studies cannot guarantee the specific function of these cells, we believe they may be involved with brain pH homeostasis and cardiopulmonary regulation.

Carbon dioxide exchange across the walls of arterioles: implication for the location of the medullary chemoreceptors

The location of the medullary chemoreceptors is not conclusively established. The original experiments, which were believed to suggest a shallow surface location in the ventrolateral medulla, have been questioned because substances, particularly CO2, applied on the surface of the medulla could diffuse into small arterioles. Because the whole tissue blood flow is supplied by surface arterioles, they could transport substances from the surface into the tissue to the respiratory centers. We studied simple transport equations describing movement of CO2 in arterioles bathed by rapidly flowing cerebrospinal fluid (CSF) and arterioles in tissue perfused by capillaries. Substantial exchange of CO2 could occur across the arteriole wall for all expected sizes of vessels when the partial pressure of CO2 at the outside wall was determined by CSF. When an arteriole is surrounded by tissue, only vessels with inside diameters (ID) less than or equal to 50 micron will exchange substantial amounts of CO2 but the smallest arterioles may be nearly in equilibrium with the tissue. The CO2 gradient in tissue around the arteriole will extend approximately 1 mm. Our simple theoretical description of CO2 transport in arterioles predicts substantial exchange in precapillary vessels. CO2 picked up by the smallest surface arterioles when the medulla is perfused at a high rate with CSF will not stay in the blood past the putative depth of the chemoreceptors. In arterioles greater than 30 micron, however, the CO2 could be carried to the respiratory centers.

The roles of medullary extracellular and cerebrospinal fluid pH in control of respiration

To determine the effective stimulus to the central chemoreceptors, we measured CSF and medullary extracellular fluid (ECF) pH and phrenic activity in 11 anesthetized, paralyzed, vagotomized and glomectomized cats. Flat-tipped pH electrodes (2 mm diam.) were used to measure ECF pH on the ventral surface of the medulla and CSF pH 2 mm above the surface. Changes in alveolar/arterial PCO2 were produced by airway occlusions of 10-20 sec durations. Changes in CSF PCO2 and pH were made by infusing 100% CO2 or an acid buffer into the CSF. Airway occlusion caused an increase of alveolar/arterial PCO2. ECF pH began to fall 6-10 sec later, with a maximum decrease of 0.032 pH unit at 21.9 sec. Phrenic activity increased as ECF pH decreased, the greatest activity occurring when ECF pH was most acid. CSF pH decreased after a longer delay. Its maximum decrease at 54.1 sec was smaller (0.026 pH unit) than ECF pH and did not correlate with the increase of phrenic activity. Addition of 100% CO2 or an acid buffer into the CSF produced an acid shift in the CSF pH but no change in ECF pH or phrenic activity. Prolonged (greater than 30 min) increase of acidity of CSF did not alter phrenic activity until ECF pH developed a delayed acid shift. Even then, the change of ECF pH was much smaller than that of CSF. We conclude that medullary chemoreceptors do not respond to changes of CSF pH or PCO2 and that change of pH of CSF minimally affects ECF pH. On the other hand, respiratory responses are closely linked to changes in ECF pH.

Types and locations of respiratory-related neurons in lateral tegmental field of cat medulla oblongata

Extracellular microelectrode recordings were made from a total of 868 neurons in the medullas of cats in regions known to contain high densities of respiratory-related neurons (solitary tract complex, nucleus ambiguus/retroambigualis, lateral tegmental field). Both the discharge patterns and the locations of units were noted and correlated with a recently described substructure of the tegmental field of the cat medulla in which neuronal cell bodies are found associated with sheets of blood vessels supplying the brainstem. The majority of cells were phasically firing (59%) with activity confined to either the inspiratory or the expiratory phase, 21% were tonically firing cells with no discernible respiratory modulation and 20% were silent neurons, responsive to electrical stimulation of the vagus nerves or the dorsolateral pons in the vicinity of nucleus parabrachialis, but not to various respiratory stimuli. Within the solitary tract complex inspiratory discharge patterns were predominant (94%), while in nucleus ambiguus/retroambigualis 26% of the neurons had expiratory patterns with the rest being inspiratory (68%) or tonic (6%). Within the lateral tegmental field, the percentages of inspiratory, expiratory and tonic patterns were 51, 9 and 40%. Thus, inspiratory type patterns were found throughout the medulla, but expiratory patterns were most common in the ambiguus/retroambigualis nuclei. Found within all 3 major regions, but primarily within the lateral tegmental field of the rostral medulla were neurons that discharged with a brief burst at the inspiratory to expiratory phase transition. These cells had properties consistent with the off-switch mechanism: extreme late-inspiratory onset of discharge with the onset time being delayed by lung inflation, peak discharge at or slightly after the peak activity of the diaphragmatic EMG and a discharge rate which was insensitive to lung inflation. Within the lateral tegmental field, where longitudinal sheets of blood vessels running radially with respect to the IVth ventricle have been described, it was found that 85% of the tonically active units and 93% of the respiratory modulated cells were located less than 200 microns from the planes of these sheets. In addition, 87% of the neurons that could be antidromically or synaptically activated from the dorsolateral rostral pons were similarly located.

Does low pH stimulate central chemoreceptors located near the ventral medullary surface?

An in vitro preparation of the medulla oblongata of the rat was used to examine the responses to reducing pH, at constant CO2, of neurons in several identified nuclei. Neurons in an area close to the ventral surface, which is thought to be the location of pH-sensitive central respiratory chemoreceptors, did not respond differently from neurons in other medullary nuclei.

Central chemosensitivity and the reaction theory

Images

Morphological observations on superficial medullary CO2--chemosensitive areas

Physiological investigations have indicated that the ventrolateral surface of the medulla oblongata is involved in the chemical drive to respiration. In this investigation, light and electron microscopic investigations of the 3 chemosensitive regions reveal the following. (1) Evaginations of the ventral surface abut the overlying pia mater thereby delimiting discrete compartments; invaginations of the surface delimit wide cisternae lined with basement membrane. Neuronal elements with numerous synapses, were found scattered among astrocytic processes of the marginal glia in intermediate and caudal chemosensitive areas Microvasculature are conspicuously absent from the marginal glia. Intramedullary vessels are surrounded by perivascular spaces and the endothelium shows zonulae occludentes at cell junctions. (2) Horseradish peroxidase (HRP) applied to the ventral surface diffused throughout the interstitial and perivascular compartments, into synaptic clefts and neuronal soma. Diffusion of HRP into blood vessels was blocked at zonulae occludentes. Following intravenous injection of HRP, no reaction product was found outside cerebral vasculature in chemosensitive areas. (3) In spontaneously breathing cats, 2% procaine applied to the caudal chemosensitive area resulted in respiratory depression which began with the second breath. It is proposed, that substances which stimulate or depress respiration, when applied to the ventral medullary surface, produce their effects on superficial neurons located in the intermediate and caudal chemosensitive areas after diffusion through interstitial spaces.

Neurophysiological studies on superficial medullary chemosensitive area for respiration

In cats anesthetized with chloralose-urethane (40 mg/kg chloralose; 200 mg/kg urethane) pH sensitivity of neurons in the caudal chemosensitive area on the ventrolateral surface of the medulla oblongata was examined while monitoring phrenic nerve activity simultaneously. pH was varied by superfusion of the ventral medullary surface with mock cerebrospinal fluid (CSF) of different pH (pH 7.4 control, 7.0 and 7.8). A total of 130 units from 21 cats changed their firing rate in response to CSF-pH changes. These were subdivided into 3 groups. In Group I, 31 respiratory pH sensitive units increased their firing rate in response to decreased mock CSF-pH, noxious pinch, joint movement in contralateral forelimb, and increased inspired CO2. These responses may have originated from respiratory center neurons. Group II consisted of 59 non-respiratory pH sensitive units whose firing rate changed in an inverse manner with CSF-pH changes. Of these, 30 responded to contralateral distal forelimb movement, 15 to hair manipulation, 9 to heavy pressure and 5 to noxious pinch. Increased inspired CO2 (rebreathing) did not modify activity. The response to pH is believed to be from non-specific neurons. Group III consisted of 40 non-respiratory pH sensitive units responding to CSF-pH changes and to increased inspired CO2. The firing rate was irregular, the interval distribution approaching an exponential function. It may be postulated that the impulse frequency of chemosensitive impulses may be irregular at the site of impulse generation, the irregularity decreasing by convergence during transmission to the respiratory centers. The time course of Group III chemosensitive units was similar to phrenic nerve responses.

The respiratory role of the ventral surface of the medulla studied in the anaesthetized rat

1. The respiratory role of the ventral surface of the medulla was studied in rats anaesthetized with a urethane-chloralose mixture. 2. In fifty-eight studies on twelve animals, direct superfusion of the medullary surface with artificial c.s.f. made acid by the reduction of bicarbonate content or by the increase of PCO2 produced no significant stimulation of respiration provided that the temperature of the brain surface was unaltered. 3. Superperfusion of the medullary surface with c.s.f. of low bicarbonate content produced an inhibition of respiration in fourteen of thirty-eight experiments. 4. Electrical stimulation on the surface revealed a localized area lateral to the pyramids and rostral to the XIIth nerve where stimulation at low intensity produced an increase in the frequency and depth of respiration. 5. The application of carbachol to a similar region increased both the frequency and amplitude of ventilation at lower concentrations than were required to obtain effects from surrounding areas. 6. Sudden switching between perfusates at different temperatures produced changes of ventilation within 1-2 sec of a change of surface temperature. The Q10 for the ventilation/temperature relationship was approximately 6. 7. The experiments confirm that the ventral surface of the medulla contains neural elements which, at least during urethane-chloralose anaesthesia, have a significant effect on respiration. The stimulus for these effects in the rat does not appear to be a change in H+ concentration. It appears more probable that the primary role of the area lies in the link between thermal and respiratory regulation.

Elimination of central chemosensitivity by coagulation of a bilateral area on the ventral medullary surface in awake cats

Breathing and respiratory response to CO2 were observed in 6 awake cats and 1 control before and after bilateral coagulation of the formerly described area S (Schläfke and Loeschcke, 1967) on the ventral medullary surface under hyperoxic conditions. Ventilation decreased, PCO2 rose and CO2 response was almost or completely abolished in 4 cats, and moderately reduced in 2 cats. Inhalation of CO2 had an inhibitory effect on ventilation in two cases. In some instances the respiratory frequency was increased by CO2. Periodic breathing as well as spontaneous hyperventilation elicited by 'arousal' indicate parallels to the Pickwickian or Ondine's curse syndrome. No respiratory changes were produced by a lesion on the pyramidal tract medial to the area S. It is concluded that central chemosensitivity can be eliminated within the superficial layer of the area S. The loss of CO2 response seems to be correlated with complete destruction of the superficial nerve cells located within the area S (Petrovický, 1968) and degeneration within the ventral part of the nucleus paragigantocellularis.

Influence of the CSF bicarbonate concentration on the ventilatory response to CO2 in relation to the location of the central chemoreceptors

In anaesthetized cats, in which the cerebrospinal fluid bicarbonate concentration was varied by a ventriculocisternal perfusion technique, the ventilatory response to CO2 during hyperoxia could be satisfactorily described by VE = S(PCSFCO2 -B). Both the slope S and the intercept B were positively and linearly related to the CSF bicarbonate concentration. Assuming that the PCSFCO2 is equal to the PCO2 in extracellular fluid, it can be shown that VE is a linear, but not a unique function of the [H+] at the site of the chemoreceptors; the slope of this relation varies with the bicarbonate concentration at that site, possibly due to chemical complex formation between HCO-3 and Ca2+ or Mg2+. Changes in the B-value were related to the location of the central chemoreceptors with the models of Pappenheimer and Berndt aand their coworkers. It was found that changes in the CSF bicarbonate concentration are reflected for 60 per cent at the site of the central chemoreceptors, and that this was independent of the cerebral perfusion. Using Berndt's model a distance between CSF and central chemoreceptors of approximately 100 micron was found; this calculated distance is relatively insensitive to relationship (logarithmic or not) between ventilation and H+ concentration and to changes in cerebral perfusion, owing to the approximate nature of the diffusion model.

On the transmission of the stimulating effects of carbon dioxide to the muscles of respiration

1. Electromyography was used to measure the response of the diaphragm and intercostal muscles to CO2 in artificially ventilated decerebrate cats. 2. Hypocapnia produced tonic activity in either inspiratory or expiratory muscles or both, according to the preparation. 3. A graded effect of CO2 on both rhythmic and tonic activity was observed and for the latter this could be seen at as low as 10 torr PA,CO2. 4. In one human subject tonic firing of expiratory motoneurones was also induced by hypocapnia and this activity showed a graded increase with increasing (CO2. 5. A saggital incision of the medulla aimed at interrupting inspiratory bulbospinal axons abolished activity in inspiratory muscles and at eupnoeic levels of CO2 converted the activity of expiratory muscles from a periodic to a topic firing pattern. 6. Following such lesions the threshold for rhythmic excitation of expiratory muscles was elevated and this revealed that the graded effect of CO2 on tonic expiratory activity extends to as high as 60 torr. 7. The tonic activation of respiratory muscles in response to CO2 ceased after cervical cord transection or when the saggital incision in the medulla was extended caudally to the first cervical segment. 8. It is concluded that the CO2 dependent activation of spinal respiratory motoneurones is conveyed by bulbospinal axons which decussate in the vicinity of the obex and that this activation can be rhythmic or tonic. 9. It is suggested that the rhythmic excitation of expiratory muscles derives from a periodic inhibition of expiratory bulbospinal neurones which are subjected to a tonic CO2 dependent excitation which is continuously variable over the physiological range.

Effect of H+ on spontaneous neuronal activity in the surface layer of the rat medulla oblongata in vitro

The effect of changing extracellular pH (pHe) on the spontaneous activity of neurons in brain slices taken from the ventral layer of the rat medulla oblongata was compared to the response of neurons in dorsal slices. In the ventral medulla, more than 50% of the neurons were excited by H+. These neurons were found just lateral to the pyramidal tract between the root of the hypoglossal nerve and the trapezoid body. In the dorsal medulla, low pHe caused an inhibition of activity in most neurons, although a few were excited. The fact that H+ elicted excitation predominantly in the ventral medullary substrate to respond to pHe changes. Depression of synaptic transmission within the neuronal network in the slice by reducing the [Ca2+]e and increasing the [Mg2+]e altered the nature of responses of neurons to H+: In the ventral medulla, the majority of neurons were inhibited by H+, whereas in the dorsal medulla more than 50% of neurons were excited. Therefore, "specificity" of the ventral medullary neurons seemed to be dependent upon intact synaptic connections. A possible role of acetylcholine-acetylcholinesterase system in the response of ventral medullary neurons to H+ is discussed.

Carbon dioxide sensitivity of the central chemosensitive mechanisms. An exploration by direct stimulation in rats

In urethane-anaesthetised adult albino rats ventral surface of the brainstem was stimulated chemically by increasing the local CO2 concentration and electrically. Two areas were demarcated on the ventral surface of the brainstem, one which showed an increase in pulmonary ventilation on chemical and electrical stimulation, and another which showed a decrease in pulmonary ventilation and sometimes even respiratory arrest. EEG activity recorded from the area from where increased pulmonary ventilation was obtained showed a synchronous slow wave activity during chemical stimulation and inhalation of a CO2-air mixture. This area is situated 0.5--1 mm lateral to the mid-line extending up to the rootlets of the VIIth to IXth cranial nerves. The response increased proportionately on increasing the strength of the chemical stimulus, till it reached a plateau. In carotid body denervated and chronic hypoxic animals, the magnitude of the responses was shown to be increased, probably due to increased sensitivity of the central chemosensitive mechanisms.

'Exponential peeling' of ventilatory transients following inhalation of 5, 6 and 7% CO2

The 'exponential peeling' technique has been applied to minute ventilation and tidal volume transients occurring after the abrupt removal of 7, 6 and 5% CO2 in inspired air. These transients, in many cases, were found to be composed of three exponential components, each contributing to the total ventilatory response and each having individual time responses. Gelfand and Lambertsen (1973) have attributed these components to the peripheral chemoreceptors as a group and to two central chemoreceptors. Statistical analysis to determine the constancy of the contribution of the three components over the range of CO2 values studied showed that, although the values for each at the different stimulus levels were not significantly different, the great subject-to-subject variation in the data precluded a firm conclusion about the constancy of the components. Because of a number of considerations it was concluded that exponential peeling of respiratory transients following abrupt removal of CO2 inhalation is not a satisfactory way to approach the problem of the numbers, relative contributions and time responses of the various receptor groups comprising the respiratory controller.

Metabolic control of respiratory neuronal activity and the accompanying changes in breathing movements of the rabbit. 1. Mainpulation of inspiratory and expiratory-inspiratory neurons

The property of the neuronal membrane to be permeable to metabolic modifiers of two regulatory enzymes has been utilized to manipulate the spike activity of inspiratory (I) and expiratory-inspiratory (EI) neurons of the bulbar respiratory centre. The neurons have been classified according to their response to lung distention or collapse (alpha- or beta-type) and to hyperventilation (tonic firing denoted by "+", cessation of activity by "-"). Using extracellular microelectrodes for single unit recording, the medulla oblongata was superfused with a metabolite-containing CSF. The various neuronal sub-types exhibited a differential activating or inhibitory response to one or several metabolic effectors. For example Ialpha+ units were activated by 5 mM glucose-6-phosphatase (G-6-P) and 3.5 mM 3-phosphoglycerate (3-PGA), which both inhibited Ibeta+ neurons, while 5 mM AMP inhibited Ialpha+ much more strongly than Ibeta+ cells. The spike density of Ialpha- and Ibeta- neurons was increased in the presence of 2.5 mM fructose-6-phosphate and 3.5--5 mM AMP, but became reduced by G-6-P. In contrast, 3 mM fructose-1,6-diphosphate and 5 mM 3-PGA activated the Ialpha- but inhibited the Ibeta- neurons. The EIbeta units were characteristically activated by 10 mM citrate, which inhibited all I-type neurons. Activations of the Ialpha and Ibeta neurons led to an accelerated respiratory rate and a higher tidal volume, while the opposite was true for EIbeta neurons. Intravenous injection of metabolites could not duplicate the striking effects under local applications.

pH sensitivity of cells located at the ventrolateral surface of the cat medulla oblongata in vitro

pH sensitivity of cells located at the ventral surface of the medulla oblongata was examined in a thin brain slice of the cat in vitro and the following results were obtained: (1) Transmembrane potential of the surface cells located in the area medial to the XIIth cranial nerve was reduced slightly by application of low pH solution; (2) In the rostral part of the area medial to the XIIth cranial nerve regular neuronal discharges could be observed extracellularly. The rate of firing of these cells was increased by lowering the external pH. These results were considered to support the idea the H+ receptor cells may exist in the surface layer of the ventral medulla.

The effects of changes in pH and PCO2 in blood and water on breathing in rainbow trout, Salmo gairdneri

The effect of sustained hypercapnia on the acid-base balance and gill ventilation in rainbow trout, Salmo gairdneri, was studied. The response to an increase in PICO2 from 0.3 to 5.2 mm Hg was a five-fold increase in gill ventilation volume and a slight increase in breathing frequency. There was a concomitant rise in PACO2 and an immediate fall in pHa. If PICO2 was maintained at 5.2 mm Hg for several days, ventilation volume gradually returned to the initial, prehypercapnic level within three days. Arterial pH also returned to the initial level within 2-3 days. These results are consistent with the hypothesis that under these conditions fish regulate pH via HCO3/C1 exchange across the gills rather than by changes in ventilation and subsequent adjustment of PACO2. A reduction in environmental pH causes a reduction in pHa but only a slow gradual increase in VG. Injections of HC1 or NaHCO3 into the blood have opposite effects on pHa but both cause a marked increase in VG. It is concluded that a rise in PACO2 results in a rise in VG and that changes in pH in blood or water have little direct effect on VG in rainbow trout. Possible location for receptors involved in this reflex response are discussed.

The control system regulating breathing in man

This review takes the regulation of breathing in man as an example of a highly developed and much studied control system in higher mammals.