Chapter 35

Neural Control of Respiratory and Cardiovascular Functions

J.L. Feldman, D.R. McCrimmon and S.F. Morrison

Mammals have developed a highly efficient system for tissue delivery of O₂ and removal of CO₂ to support their high metabolic rate. Getting O₂ to the tissues is a two-stage process: a reciprocal pump driven by respiratory muscles exchanges air from outside the body and the lung and a double circulating pump driven by cardiac muscle drives blood through the lungs, where O₂ enters and CO₂ leaves the pulmonary circulation, and then throughout the body, where O₂ leaves and CO₂ enters the systemic circulation. In order to avoid the brain injury that can be triggered by even brief O₂ deprivation, the respiratory and cardiac pumps must work continuously in mammals, from birth until death, without lapses of more than a few minutes.

The role of the brain in assuring O₂ delivery is preeminent. For respiration, the brain generates the rhythmic motor outflows that drive the contraction and relaxation of skeletal muscles that pump the lung and modulate the tone of skeletal and smooth muscles in the upper airways and bronchi to control resistance to air flow. For the circulation, the brain generates the autonomic motor activities that maintain vascular resistance and arterial pressure and that modulate cardiac contractility and heart rate to ensure an appropriate perfusion of all the tissues under a wide range of whole body and local metabolic demands. Through these mechanisms, the brain supports body function over a twenty-fold range in metabolism (in humans from ~0.25 liter × min⁻¹ O₂ consumed at rest to ~3–6 liter × min⁻¹ O₂ consumed during maximal exercise). Although respiratory muscles account for ≤5% of body metabolism at rest, the brain must drive them efficiently and without serious respiratory muscle fatigue. During exercise in humans, for instance, the brain must increase air exchange from a resting value of 5 up to 200 liters × min⁻¹ (maximal performance in world-class athletes) while minimizing the energy costs of breathing to maximize the O₂ available to other working muscles. This chapter summarizes the distinct neural mechanisms underlying the brain’s key roles in controlling respiratory and cardiovascular functions and then describes how these markedly different systems work in tandem.

The drive for continuous ventilation in humans is strong. Typically, breathing is the last consequential movement (the “dying” gasp) to disappear following generalized depression of higher brain function, such as during surgical anesthesia or following insults to the brain—for example, hypoglycemic coma or global brain anoxia following a heart attack. Automatic, homeostatic breathing can remain in people who have lost cortical but maintain brainstem and spinal cord function, resulting in ethical and emotional dilemmas in individuals considered “brain dead.”

Neuroscientists take advantage of this robustness of breathing to study synaptic, neural, and network mechanisms in anesthetized or decerebrate mammals that continue to breathe. These mechanisms are so robust and sufficiently self-contained that a thin in vitro tissue slice (~500 µm thick) at an appropriate medullary level in neonatal or late fetal rodent brainstem continues to generate respiratory-related patterns of motor nerve activity.

The cardiovascular system, in addition to carrying O₂ and CO₂, delivers nutrients and removes waste products from the diverse and widely separated tissues of the body. The circulatory system is a conduit for hormonal and metabolic signals arising in the hypothalamus, adrenal cortex, and gut, and for immune cells from, for example, bone marrow, to sites of infection or injury. The central neural regulation of the cardiovascular system provides the overarching coordination among the various aspects of cardiovascular function required not only to maintain a homeostatic cellular environment for all tissues at rest, but especially to meet the enhanced tissue demands during the challenges of behavior, disease, and environmental adversity. The brain regulates the cardiovascular system primarily through the sympathetic and the parasympathetic divisions of the autonomic nervous system (see Chapter 33) and secondarily via hormones, in particular the sympathetically-driven secretion of epinephrine from the adrenal gland.

Breathing

A Brief Neuroscience-Oriented History

Although the purpose of breathing—to move air into and out of the lungs—was not established until the late eighteenth century, the sites within the body that control breathing have been a constant source of speculation in Western culture (Feldman, 1986). Perhaps the earliest suggestion that the brain was involved came from Galen (ca. A.D. 131–201), who observed that gladiators and animals injured below the neck continued to breathe, but those injured in the neck stopped breathing. Lorry (1760) showed that cerebellectomized rabbits continued to breathe and suggested that the critical circuits lay within the brainstem and upper spinal cord, essentially the contemporary view. Legallois (1813) “extracted” brain tissue to determine what was necessary for breathing; presciently, he localized critical regions to the rostral ventrolateral medulla, close to the preBötzinger Complex (preBötC; Figs. 35.2, 35.7), which we recognize today as a key brain site for generating respiration.

Ramon y Cajal (1909) provided an anatomic rationale for the critical role of the medulla in breathing. Examining the afferent/efferent projections of respiratory-related nerves, he focused on two brainstem nuclei: the nucleus of the solitary tract (NTS), the target of pulmonary afferents, and the nucleus ambiguus, which contains cranial motoneurons that innervate upper airway muscles. His network model for breathing is prescient: afferents containing information about lung volume or blood gases combined with intrinsic properties of brainstem neurons to produce a rhythmic outflow to spinal and cranial respiratory motoneurons (Fig. 35.1). In brainstems isolated from goldfish, Adrian and Buytendijk (1931) recorded slow, rhythmic potentials that were similar in timing to the fish’s gill movements when intact, physiological evidence that the brainstem contained the critical circuits. Gesell and colleagues (1936) subsequently recorded individual medullary neurons discharging bursts of activity in phase with the breathing rhythm. This finding initiated a concerted effort to identify neurons that generate respiratory rhythm (Box 35.1).

Box 35.1 Experimental Analysis Of Ventilation

Neuroscientists are constrained by what they can measure. Noninvasive experiments in humans reveal the phenomenology of breathing—for example, the relationship between ventilation and blood CO₂ levels. Brain imaging may reveal which regions are involved, but at present, spatiotemporal resolution is too coarse to show networks, much less neurons or synapses. Respiration lends itself to study because mammals suitable for experimentation breathe much like humans. Breathing continues following anesthesia or decerebration, and the neural patterns remain following paralysis (and mechanical ventilation), making possible experimental procedures requiring highly invasive techniques, such as single neuron recording. Compared to studies of neurons in culture or in tissue slices, experiments in whole mammals have serious limitations: (1) movements of the heart and lungs make certain types of single neuron recording, as well as modern imaging techniques such as fMRI, difficult in the brainstem and spinal cord; (2) the blood–brain barrier is intact, making precise control of brain extracellular fluid impossible; and (3) acute experiments in large mammals, such as cats, dogs, and goats, are expensive and can last 36 h or more, requiring the dedicated and concurrent efforts of several investigators. Fortunately, the robust features of breathing that allow it to persist when other behaviors are suppressed by anesthesia or decerebration permit further reduction in the experimental preparation that removes these limitations. Moreover, a respiratory-related rhythm persists in the brainstem and spinal cord removed from a neonatal rodent and even in a particular slice of brainstem isolated from this preparation. Using such slices, one can correlate exquisite measures of a neuron’s intrinsic and synaptic properties and projections with measures of endogenous behavior, such as its firing pattern. With many common and otherwise useful in vitro preparations, such as the hippocampal slice, such correlations cannot be made because no endogenous behavior in vitro relates to known behavior in vivo. In general, en bloc in vitro, or slice, preparations play an important role in the study of basic cellular, synaptic, and integrative mechanisms underlying respiratory motor behavior.

Jack L. Feldman and Donald R. McCrimmon

Figure 35.1 Functional organization of the CNS control of breathing. Circuitry centered within the medulla oblongata of the brainstem (blue oval) generates an oscillating inspiratory–expiratory rhythm. Neurons within the oscillator circuit generate rhythmic respiratory motor output without requiring sensory signals related to lung inflation. The rhythm is relayed to networks of premotor and interneurons constituting pattern generators that sculpt the detailed firing patterns relayed to spinal and cranial motoneurons. Spinal respiratory motoneurons innervate skeletal muscles, including the diaphragm and muscles of the rib cage, producing the pumping action of the lungs. Cranial motoneurons, e.g., hypoglossal (cranial nerve XII) motoneurons innervating the tongue and vagal motoneurons (cranial nerve X) innervating the glottis, trachea, and bronchi also receive respiratory input that adjusts airway caliber (and hence the resistance to airflow) over the course of each respiratory cycle. The levels of oxygen (PO₂) and carbon dioxide (PCO₂) are sensed by chemoreceptors located in the brain and major arteries. If a low PO₂ or elevated PCO₂ is detected, there is an increase in afferent signals sent to the oscillator and pattern generating circuits to increase breathing. Also, mechanoreceptors located in the airways sense the magnitude of lung inflation and send signals to the central rhythm and pattern generating circuits to modulate the pattern of breathing. Maintenance of appropriate ventilation also requires that breathing be adjusted during a variety of behaviors such as phonation and sleep as well as in response to exercise or changes in posture. These adjustments are mediated via inputs to brainstem rhythm and pattern forming circuits.

Where are the Neurons Generating the Respiratory Pattern?

Three distinct functional classes of respiratory neurons underlie the generation of the discharge pattern on the motor nerves innervating skeletal muscles of the respiratory pump and airways: respiratory muscle motoneurons, rhythm-generating neurons, and premotoneurons that link and coordinate rhythm-generating neurons to motoneurons and transform that rhythm into a mechanically efficient pattern of respiratory muscle activity. Nonrhythmic neurons that modulate the behavior of these respiratory neurons are also important. These include neurons involved in sensing blood and brain levels of O₂ and CO₂—that is, chemoreceptors—or may reflect sleep-wake state, such as serotonergic or (nor)adrenergic neurons. Brainstem respiratory neurons are concentrated within three regions (Fig. 35.2).

Figure 35.2 Respiratory and cardiovascular cell groups in a sagittal view of the rat brainstem. Functionally distinct brainstem compartments related to breathing or cardiovascular control are indicated by shaded regions of different colors. Red text indicates the three primary respiratory and two primary cardiovascular cell groups. Subregions of two of the primary respiratory groups, the ventral respiratory column (VRC) and pontine respiratory group (PRG), are indicated with black labeling. Nearby brainstem structural landmarks are indicated in gray for prominent neuronal nuclei, or outlined without shading for major fiber bundles. 7, facial nucleus; 7n, facial nerve; AmbC, compact part of nucleus ambiguus; BötC, Bötzinger complex; cVRG, caudal division of ventral respiratory group; DRG, dorsal respiratory group; KF, Kölliker–Fuse nucleus; LRt, lateral reticular nucleus; Mo5, motor nucleus of the trigeminal nerve; NTS, nucleus of the solitary tract; PBr, parabrachial region; Pn, basilar pontine nuclei; preBötC, preBötzinger complex; RTN - pFRG, retrotrapezoid nucleus and parafacial respiratory group; rVRG, rostral division of ventral respiratory group; scp, superior cerebellar peduncle; SO, superior olive; VRG, ventral respiratory group.

Modified from Alheid and McCrimmon (2008).

The dorsal respiratory group (DRG) is located within the caudal half of the NTS, the primary target of afferent input from the glossopharyngeal and vagus nerves. DRG neurons are optimally located to receive respiratory-related sensory input. DRG neurons provide input to other groups of respiratory neurons within the ventral respiratory column (VRC) of the ventral medulla, the pontine respiratory group (PRG), as well as respiratory motoneurons in the spinal cord.

The VRC forms a continuous rostrocaudal field in the ventrolateral medulla with multiple, physiologically distinct subregions, extending from the facial nucleus to the spinomedullary junction. The caudal half of the VRC, termed the ventral respiratory group (VRG), contains respiratory-modulated bulbospinal premotoneurons. Inspiratory neurons in the rostral VRG (rVRG) provide mainly excitatory premotor drive to interneurons projecting to inspiratory muscle motoneurons, including phrenic motoneurons innervating the diaphragm (the principal inspiratory muscle) in the midcervical (C3-C5 in humans) spinal cord and external intercostal muscles of the rib cage in the thoracic spinal cord. Expiratory bulbospinal neurons in the caudal VRG (cVRG) are mainly excitatory and project to interneurons innervating spinal motoneurons that drive internal intercostal and abdominal expiratory pump muscles in the thoracic and lumbar spinal cords, respectively. Inhibitory expiratory premotor neurons in the cVRG and in the Bötzinger complex (BötC) inhibit spinal inspiratory motor pools.

In the rostral VRC, rostral to the rVRG, are the preBötC (an essential structure for normal rhythm generation); the BötC; the retrotrapezoid nucleus (RTN) that contains central chemosensory cells whose activity increases in parallel with increasing local tissue acidity and provides an important source of respiratory drive; and finally, the parafacial respiratory group (pFRG) which plays an important role in the generation of expiratory motor output (Feldman & Del Negro, 2006).

The PRG in the rostral dorsolateral pons encompasses portions of the parabrachial complex and the ventrally adjacent Kölliker-Fuse nucleus. PRG neurons may be necessary for the full expression of the normal breathing pattern, but do not seem essential for rhythm generation per se. Further, they contribute to the reflex changes in breathing elicited by sensory afferents sensitive to lung volume. They also contribute to the reflex changes in breathing induced by input from chemoreceptors sensitive to the levels of arterial blood O₂, CO₂, and pH. PRG neurons may also contribute to the coordination of respiration with the cardiovascular system, and mediate the effects of breathing rhythm on suprapontine structures and vice versa.

Discharge Patterns of Respiratory Neurons

Although inspiration and expiration constitute two obvious phases of the mammalian respiratory cycle, the discharge patterns of inspiratory (diaphragm) and expiratory—for example, internal intercostal—muscle activity suggest two expiratory phases. During the first expiratory phase (E1, termed postinspiration), the diaphragm often exhibits a low level of “postinspiratory” activity that slows diaphragm lengthening at the onset of expiration, thus reducing the initial rate of expiratory airflow (Fig. 35.3). Postinspiratory activity typically declines rapidly, ending by midexpiration. In a healthy individual under resting conditions there is little or no activity of expiratory pump muscles, with expiration occurring via the passive recoil of the lung. At higher levels of ventilation, such as during exercise (note: here exercise means sustained activity of large muscles, such as would be involved in fleeing a predator or chasing prey, or running on a treadmill!), an augmenting pattern of expiratory muscle activity in the second (E2) phase increases the expiratory airflow. These motor patterns are generated by interactions among respiratory neurons primarily localized in the respiratory cell groups illustrated in Figure 35.2.

Figure 35.3 A diverse pattern of respiratory motor activity is exhibited on spinal (phrenic) and cranial nerves innervating the diaphragm and airways in an anesthetized rat. Low-pass filtered traces of nerve activity show the overall pattern of discharge. Red vertical lines indicate phase transitions between inspiration (I) and expiration (E), which itself is divided into two subphases: E1 (postinspiration) and E2. Note that the onset of inspiratory activity on cranial nerves precedes the onset of activity on the phrenic nerve. IX, glossopharyngeal nerve; PhX, pharyngeal branch of vagus nerve; SLN, superior laryngeal nerve XII, hypoglossal nerve. Traces depict activity in a rat, but activity is similar in other mammals although respiratory frequency varies from greater than 100 × min⁻¹ in mice to about 3 × min⁻¹ in giraffes.

Adapted from Hayashi and McCrimmon (1996).

Models for Respiratory Pattern Generation

Oscillations such as the respiratory motor patterns shown in Figure 35.3 are a common form of dynamic activity in physical and biological systems that have attracted modelers since the early seventeenth century when philosopher René Descartes used analytic reasoning to model the mind. Since the 1960s, computational models to explain how rhythm-generating neurons in the preBötC, RTN/pFRG interact with the rest of the network, including premotor and motoneurons, to produce appropriate motor patterns have successfully reproduced human and animal breathing patterns. These models are useful for summarizing interaction mechanisms among components of the respiratory system, but modeling efforts still seek the elusive goal of nontrivial, testable predictions. Although current modeling elucidates plausible mechanisms, further data on network connections, intrinsic membrane properties, and synaptic interactions are needed so models are sufficiently constrained to provide novel insights into the underlying mechanisms.

A Respiratory Rhythm Generator is Discovered in the preBötC

Since the late eighteenth century, the brainstem has been recognized as the source of respiratory rhythm, but only in the past 25 years has the precise locale of the obligate structures been determined. A critical step in this advance was the development of novel in vitro preparations (see Box 35.1). In the 1980s, Suzue and colleagues (Suzue, 1984) noted that the isolated neonatal rat brainstem and spinal cord, termed the “en bloc” preparation, generated a respiratory rhythmic motor output in spinal nerves that innervate muscles of the respiratory pump, as well as in cranial nerves innervating the airways, which control airway resistance. This in vitro preparation permitted a refinement of the “blunt spatula” brainstem transection approach used almost 200 years earlier by Legallois (1813) to identify structures critical for respiratory rhythm generation. In the contemporary case, removing thin (50–75 μm) sections from either rostral or caudal ends of the en bloc preparation, while recording respiratory activity from spinal or cranial nerves resulted in the isolation of a small, ventral medullary region, named the preBötC, as an essential component in the rhythm generating circuitry (Smith, Ellenberger, Ballanyi, Richter, & Feldman, 1991). Several experimental observations indicate that neurons within the preBötC provide a minimum essential circuitry for respiratory rhythm generation (Feldman & Del Negro, 2006). (1) Slices containing the preBötC generate a rhythm indistinguishable from the respiratory-related rhythm en bloc, although this pattern differs from that seen in vivo. (2) In intact adult awake rats, neurotoxic lesions of a subset of preBötC neurons containing the receptor for substance P (i.e., neurokinin-1 receptor), irreversibly induces a pathological, ataxic breathing pattern quite different from normal breathing (Fig. 35.4 and see Box 35.2). (3) Essentially normal inspiratory motor pattern persists after transection of the rat brainstem just rostral to the preBötC, which removes all pontine circuits and the RTN/pFRG although phasic expiratory motor activity is abolished. (4) Rapid transient silencing of a subpopulation of preBötC neurons, those glutamatergic neurons that also produce the peptide somatostatin (~500 neurons/side), leads rapidly to a persistent apnea in awake rats (Fig. 35.5). Taken together, these experimental results underlie the now widely accepted view that the preBötC is an essential component of the circuitry generating breathing rhythm.

Box 35.2 CNS Lesions Produce Abnormal Breathing Patterns

Many brain injuries and diseases produce abnormal breathing patterns (Plum & Posner, 1980). Because many brain regions provide afferent input to the neurons generating the breathing rhythm, pathology in regions not normally associated with the generation of breathing can produce abnormal breathing patterns. Despite the diffuse nature of many pathologies that give rise to abnormal breathing patterns, Plum and colleagues systematically characterized several breathing disorders (Fig. 35.4) arising from specific CNS pathologies.

Apneustic breathing is marked by prolonged inspiratory periods. In humans, the most frequent pattern is inspirations lasting 2–3 s alternating with prolonged expiratory pauses. In cats, the prolonged inspiratory periods are associated with plateaus in inspiratory drive that can last minutes (leading to death in the absence of mechanical ventilation). Apneusis is observed in people with lesions of the pons, including, or just ventral to, the pontine respiratory group. In experimental animals, apneusis requires not only lesions of the pontine respiratory group, but also interruption of vagal afferent input.

Lesions of the corticobulbar or corticospinal tracts can lead to Cheyne–Stokes respiration, a rhythmic waxing and waning of breathing. Periods of no breathing (apnea) follow each period of waning inspiratory depth. Lesions of the corticobulbar and corticospinal tracts can also result in loss of voluntary control of breathing. In pseudobulbar palsy, for example, voluntary control of breathing and of cranial motoneuron function is lost secondary to a lesion often located dorsomedially in the base of the pons.

Extensive bilateral damage to the medullary respiratory groups can severely disrupt or abolish respiratory rhythm, resulting in death unless artificial ventilation is initiated immediately. Fortunately, unilateral damage does not appear to be sufficient to cause severe disruption of respiratory rhythm- and pattern-generating mechanisms. Because two vertebral arteries supply blood to the medulla, bilateral damage from an infarct or embolism is unlikely.

Less extensive damage to medullary respiratory structures can produce ataxic breathing, an irregular pattern of breathing with randomly occurring large and small breaths, periods of apnea and low breathing frequency.

Spinal injuries that affect the long axons of bulbospinal premotoneurons that transmit respiratory drive to spinal motoneurons can be life-threatening. Damage between C3 and S1 can diminish the ability to generate inspiratory or expiratory movements of abdominal and intercostal muscles, with the effects more pronounced the more rostral the injury. Injuries between the lower brainstem and C3 will also affect the diaphragm. Such injuries require immediate artificial support of ventilation to maintain life. This type of cervical injury is more common in young, active individuals, resulting from dives into shallow water or being thrown from a horse.

Jack L. Feldman and Donald R. McCrimmon

Figure 35.4 Abnormal breathing patterns resulting from CNS disorders. The ordinate is lung volume. Adapted from Plum and Posner (1980).

Figure 35.5 PreBötzinger complex (preBötC) is essential for breathing in adult rats. Rapid silencing of allatostatin receptor (AlstR)-expressing, preBötC somatostatin (Sst) neurons induces persistent (>45 min when mechanically ventilated) apnea in anesthetized or awake adult rats. Traces are plethysmographic recordings of breathing movements. Allatostatin (AL) administered intracerebrocisternally induces a gradual decline in frequency and tidal volume until apnea develops after several minutes. After ~60 minutes of mechanical ventilation, rats resume spontaneous breathing. The ordinate is lung volume.

From Tan et al. (2008).

A Second Respiratory Rhythm Generator is Unveiled

The RTN/pFRG (Fig. 35.2), a region rostral to the preBötC, is now thought to play a critical role in generating respiratory rhythm as well as in its ontogeny and in chemoreception (see Box 35.3). Functional evidence for the presence of this second oscillator initially came from experiments exploiting differences in the opioid sensitivity of preBötC neurons, which are inhibited by opioid agonists, and RTN/pFRG neurons, that appear unaffected. In in vitro slices that contain only the preBötC, opioids slow the inspiratory rhythm in a graded fashion, gradually increasing the interval between inspiratory bursts (Fig. 35.6A, top). In the more intact en bloc preparation that contains both the preBötC and the RTN/pFRG, opioid agonists also cause a slowing of inspiratory rhythm, but instead of progressively lengthening the interval between inspiratory bursts, some inspiratory bursts are skipped. This results in a stepwise lengthening of the interval between inspiratory bursts (termed “quantal slowing” (Mellen, Janczewski, Bocchiaro, & Feldman, 2003)), with the amount of slowing dependent on how many inspiratory bursts were skipped (Fig. 35.6A, bottom). While this is happening in the motor output, RTN/pFRG expiratory neurons continue to burst at the control frequency with no skips and remain phase-locked to motor nerve output when it occurs, whereas preBötC neurons only burst when there is a motor output (Fig. 35.6B). Quantal slowing of the inspiratory rhythm also occurs in vivo, where, the frequency of expiratory motor nerve bursts is unaffected! This is unlikely to be due to a simple block of inspiratory motor output, as the pattern of expiratory motor activity is quite different in cycles with and without inspiration (Fig. 35.6C, right). Excitation or disinhibition of RTN/pFRG neurons can transform a breathing pattern with no active expiratory activity into one with considerable expiratory motor activity (Fig. 35.6D) (Pagliardini, Janczewski, Tan, Dickson, Deisseroth, & Feldman, 2011). Finally, lesions or transections can be used to separate the regions that generate inspiratory and expiratory activity. In vagotomized and anesthetized juvenile rats, complete brainstem transections rostral to the VII nucleus that leave the RTN/pFRG and preBötC intact do not markedly affect inspiratory or expiratory motor bursting activity, whereas transections that disconnect the rostral portion of the RTN/pFRG from the more caudal preBötC abolish expiratory motor activity without affecting inspiratory rhythm. Additionally, continuous lung inflation suppresses inspiratory activity and enhances expiratory drive, whereas lung deflations increase the rate of inspiration, while minimizing rhythmic expiratory efforts.

Box 35.3 Why Two Distinct Oscillators?

Two key events characterize the evolution of breathing from fish to mammals. The first includes transformations that support the transition from aquatic to terrestrial breathing in vertebrates. “Breathing” in fish is driven by cranial nerve innervation of the gills. As primitive fish (such as the ancestors of the lungfish) ventured onto land, rhythmic activity of muscles used for other purposes evolved to drive the emerging lung, and this activity might have had a neural origin that was independent of the rhythm driving the gill muscles. This speculative hypothesis is supported by observations in amphibians. Developing frogs have two distinct breathing patterns driven by different muscles and nerves: a buccal pattern that uses cranial nerve-innervated muscles to essentially swallow air and a lung pattern that uses respiratory muscles to pump the lung. Either the buccal or the lung pattern can be suppressed without abolishing its counterpart, but when both are active they appear to be phase-locked. Notably, the lung rhythm is depressed by opioids (like the mammalian preBötC neurons and inspiration), whereas the buccal rhythm is not (like pre-I neurons and expiration). A prediction from this hypothesis is that the preBötC is absent in fish.

The second key event was the emergence of the diaphragm in mammals. Respiratory physiology in vertebrates has evolved to support higher resting and peak ventilation. For example, lizards have gular pumps that uncouple respiration from locomotor musculature, which enhances O₂ consumption to support high-speed locomotion. Birds utilize sacs to pump O₂-rich air through non-compliant lungs, in support of flight. Uniquely, mammals have muscular diaphragms, a profoundly important ventilatory advance as it allows for continuous high basal rates of ventilation and for the very high levels of ventilation necessary to support substantial increases in metabolism. The improvement in gas exchange associated with the development of the diaphragm was probably crucial for brain evolution, as it was a platform for evolution of a brain that consumes extraordinary amounts of O₂ and is generally intolerant of even transient anoxia. The neural circuits driving the primitive diaphragm could have evolved independently of preexisting circuitry driving the rhythmic pumping machinery of the rib cage, abdominal muscles, or upper airway muscles, or it could have been based on the opioid-sensitive oscillator driving lung breathing in amphibia. Nonetheless, a respiratory movement pattern consisting of alternating inspiratory activity to lower the diaphragm with expiratory motor activity has considerable mechanical advantages for efficient ventilation over a broad range of metabolic demand. Therefore, the neural oscillators for breathing, which may have arisen independently, might be expected to evolve to be coupled.

There may be a special role for two oscillators at birth. In mammals there is a surge of endogenous opioids at birth. This postnatal increase in opioid release would depress the excitability of the preBötC but not the RTN/pFRG, thereby allowing the latter to promote breathing immediately after birth, either by rhythmically driving breathing or by providing a drive necessary to keep the preBötC rhythmic. This view is supported by experiments in Krox-20^−/− mice that have neuroanatomical defects that result in deletion of the brainstem region containing the RTN/pFRG but do not affect the preBötC. These mutant mice die of fatal apneas within 18 hours after birth, but brief postnatal administration of naloxone (an opioid receptor agonist which has no effect in control mice) rescues these pups by eliminating their apneas. Thus, without the RTN/pFRG to provide respiratory drive, the opioid surge at birth could fatally depress preBötC function. We suggest that in Krox-20^−/− mice, in the absence of the RTN/pFRG, the opioid-induced depression of the preBötC causes prolonged and ultimately fatal apneas; naloxone would reverse this depression, which is developmentally relieved within two days, allowing the preBötC to function adequately for survival. This special role for the RTN/pFRG at birth is consistent with the two-oscillator model.

Jack L. Feldman and Donald R. McCrimmon

Figure 35.6 Evidence for two respiratory oscillators. (A) Slowing of the inspiratory period in vitro in a slice (top) and en bloc (bottom), before and after DAMGO, an opiate agonist (red bar): top—continuous slowing of rhythm in the slice; bottom—quantal slowing of rhythm en bloc. T(s), time in seconds between successive bursts of inspiratory neuronal activity recorded either from the preBötC (top) or a cervical ventral root (bottom). (B) Simultaneous recordings of inspiratory burst activity in XII nerve and preBötC inspiratory neurons (top two pairs of traces) and pre-I neuron in RTN/pFRG (bottom pair). Arrows in DAMGO traces indicate subthreshold events in preBötC neurons during skipped bursts, which occur at the approximate expected time of inspiratory bursts at control frequency, or unperturbed bursting in pre-I neurons. (C) In vivo recordings (juvenile rat) following fentanyl, an opiate agonist, injection. Left: typical recording with normal cycles (1, 3) interspersed with cycles without inspiratory activity (2, 4). Traces from top: integrated genioglossus muscle EMG; integrated abdominal muscle EMG, tidal volume (V_T). Right: Superposition of ∫EMG_GG and ∫EMG_ABD in a normal cycle (black) compared to an inspiratory skipped cycle (red). Note the marked differences, especially in the ∫EMG_ABD burst. (D) Optogenetic stimulation of RTN neurons turns on abdominal (expiratory) nerve activity, with minimal effect on inspiratory activity.

A, B adapted from Mellen et al. (2003); C adapted from Janczewski and Feldman (2006); D adapted from Pagliardini, et al. (2011).

Summary

The working paradigm for the core engine of breathing is that the preBötC generates inspiratory rhythm and the RTN/pFRG generates expiratory activity, and these two oscillators are coupled (Fig. 35.7).

Figure 35.7 At the core of current models for respiratory rhythmogenesis are two medullary nuclei, the preBötC and the RTN/pFRG. The preBötC is essential for the generation of inspiratory motor activity, sufficient for breathing at rest. Endogenous stimulation of the RTN/pFRG by elevated CO₂, or perhaps exercise, is hypothesized to produce active expiration (absent at rest in adult mammals). The above figure summarizes results discussed in the text.

Mechanoreceptors in the Lungs Adjust Breathing Pattern and Initiate Protective Reflexes

Many different patterns of muscle activity can produce appropriate ventilation; however, the specific pattern should minimize energy expenditure. Mechanical efficiency, in turn, depends on factors such as posture, and lung and chest wall mechanics; for example, lung injury can lead to an increase in fibrous tissue in the lung, making it harder to inflate than a normal lung, and thus for this condition, rapid, shallow breathing is more efficient. Feedback about the mechanical status of the lungs and chest wall is provided by mechanoreceptors. Protective respiratory reflexes, such as sneezing and coughing, are also generated by pulmonary receptors.

There are three distinct groups of lung afferents, and all of them have axons in the vagus nerve that terminate in the NTS (Kubin, Alheid, Zuperku, & McCrimmon, 2006). Two groups have large, myelinated A fibers. These include slowly adapting and rapidly adapting pulmonary stretch receptors. The third group has unmyelinated C fibers and responds to inhaled irritants such as smoke. Most natural stimuli—for example, pulmonary edema or alveolar collapse—probably activate multiple receptor types.

Slowly Adapting Pulmonary Stretch Receptors

Located in airway smooth muscle, slowly adapting pulmonary stretch receptors (SARs) are activated when the airways stretch during lung inflation (Kubin et al., 2006) providing the central pattern generator with information about lung volume. Their activation drives reflexes by which lung inflation shortens inspiration and prolongs expiration (Breuer–Hering reflex), which is an important determinant of inspiratory duration during normal relaxed breathing in most mammalian species, with the notable exception of humans. The activation of these receptors shortens inspiration and increases breathing frequency. In humans, activation of SARs by normal breaths only modestly alters inspiratory timing. Their activation also relaxes airway smooth muscle, dilating the airways to make breathing easier.

Rapidly Adapting Pulmonary Stretch Receptors

Located in airways, rapidly adapting pulmonary stretch receptors (RARs) initiate protective reflexes in response to a variety of stimuli, including large or rapid lung inflation or deflation, inhaled irritants, and, possibly, airway edema (Canning & Spina, 2009). Cigarette smoke activates RARs and elicits a cough to rid the airway of the offending material. RARs elicit a larger than normal inspiration (sigh) in response to alveolar collapse (atelectasis), inflating the lung and popping open the collapsed alveoli. Also contributing to cough is a recently identified group of airway receptors, termed cough receptors, with small myelinated fibers (A-∂ fibers).

Bronchopulmonary C Fibers

C fiber afferent activation elicits apnea followed by rapid shallow breathing. Like RARs, C fibers are polymodal, activated by chemical and mechanical stimuli. Their activation also enhances airway mucus secretion. Inhaled particles are trapped in the mucus and removed from the airways by the action of cilia, which continuously (in the nonsmoker) move the mucus and trapped particles toward the mouth to be swallowed or expectorated.

Summary

Mechanoreceptors, especially those in the lungs, are essential for the precise regulation of the timing and amplitude of breathing. They also participate in reflexes, such as those eliciting coughing and mucous secretion, that protect the airways and lungs from compromises in airflow and inhaled irritants.

Cardiovascular

Organization and Functions of the Innervation of the Cardiovascular System

The heart, unlike skeletal muscle, can beat rhythmically in the absence of its neural input. However, neural inputs are essential for modulating cardiac contractility—that is, force of contraction—and heart rate. Sympathetic and parasympathetic nerves innervate the heart (Fig. 35.8), regulating cardiac function in a push-pull manner. Increasing the sympathetic input to the heart increases heart rate and its contractility, thereby increasing the amount of blood pumped—that is, cardiac output. The sympathetic innervation is via the cardiac nerve that arises from noradrenergic sympathetic ganglion cells (primarily in the stellate ganglion) driven by cholinergic preganglionic neurons in the intermediolateral nucleus of the upper thoracic spinal cord. Increasing the parasympathetic input to the heart decreases the heart rate. The parasympathetic, cholinergic postganglionic neurons are within adipose tissue apposing the heart, driven by cholinergic vagal preganglionic neurons whose somata are in the nucleus ambiguus and the dorsal motor nucleus of the vagus.

Blood vessels, which have only a low contractile tone in the absence of neural input, generally receive only a sympathetic innervation that mediates vasoconstriction (although several notable exceptions are described in the next paragraph). Increased vascular sympathetic activity increases the tone of vascular smooth muscle, resulting in vasoconstriction that increases arterial blood pressure. Decreased vascular sympathetic activity allows the blood vessels to dilate, decreasing blood pressure. Sympathetic input to veins increases venous wall tension, which increases cardiac filling and cardiac output. The sympathetic vascular innervation is from noradrenergic sympathetic ganglion cells in prevertebral (visceral blood vessels) and paravertebral (vessels in the skin and skeletal muscle) ganglia close to their target tissues. Their preganglionic neurons are in the intermediolateral nucleus throughout the thoracolumbar spinal cord segments.

Sympathetic activation can be highly localized or global in effect. Sympathetic preganglionic neurons and ganglion cells are target-specific, providing a substrate for selective CNS regulation of the heart and of the blood vessels in each tissue. Chromaffin cells in the adrenal medulla, driven by adrenal sympathetic preganglionic neurons in the thoracic spinal cord, secrete the catecholamines, epinephrine and norepinephrine, into the circulation as part of the stress response. Although normally at a low level, circulating epinephrine stimulates β1 adrenergic receptors on cardiac tissue to increase cardiac output and β2 adrenergic receptors to dilate skeletal muscle blood vessels, both supporting the “fight-or-flight” response to danger, as well as the muscle hyperemia needed for exercise.

There are several notable exceptions to the concept that blood vessels only receive a sympathetic postganglionic innervation. Innervation of arteries supplying the penis or the clitoris includes a parasympathetic component underlying an active vasodilation that maintains penile erection or clitoral engorgement and determines the success of reproductive behavior. Indeed, the role of nitric oxide as the mediator of this parasympathetic vasodilation is the basis for a variety of successful pharmacological agents to combat erectile dysfunction. Parasympathetic inputs to the eye produce a nitric oxide-mediated vasodilation in the uveal vasculature that contributes both to the fine control of the availability of metabolic fuels to parts of the eye that lack an intrinsic blood supply and to optimizing the retinal image through variations in uveal light absorption. In humans, sympathetic vasodilator nerves to the skin are principally responsible for an acetylcholine-mediated relaxation of vascular smooth muscle that increases cutaneous blood flow for increased heat loss in hyperthermic environments. The parasympathetic innervation of the salivary glands elicits a local vasodilation that supports an increase in salivary secretion. Norepinephrine released from sympathetic nerve terminals innervating brown adipose tissue acts not only on brown adipocytes to induce thermogenesis, but also on local vascular endothelial ß-adrenoceptors to induce nitric oxide synthase and thereby provide the increase in tissue blood flow that distributes the heat throughout the body. Similarly, the increase in adrenal medullary blood flow that accompanies the sympathetic stimulation of adrenal catecholamine release is dependent on a sympathetic preganglionic, cholinergic stimulation of nitric oxide production.

Basal Neural Tone to the Cardiovascular System

The sympathetic and parasympathetic nerves to the heart and the sympathetic nerves to the vasculature all have a basal level of discharge ultimately controlled by central neural circuits innervating the various populations of preganglionic neurons (Fig. 35.8). Due to the intrinsic rhythmicity of cardiac pacemaker cells, the heart beats without extrinsic innervation (as is the case for human heart transplants). Heart rate is modulated by the combined sympathetic, parasympathetic, and adrenal epinephrine effects on cardiac pacemakers. For instance, resting heart rate is slower in trained athletes due to an increased vagal (parasympathetic) tone—a reflex that maintains a constant resting cardiac output in the face of the increased stroke volume resulting from the stronger cardiac muscle produced by training. In contrast, although there are a variety of chemical signals that affect vascular smooth muscle contraction, under normal conditions, the level of sympathetic nerve activity to blood vessels is the primary factor determining vascular diameter and, in turn, the resistance to blood flow throughout the vascular network. Vascular smooth muscle, in the absence of sympathetic input, has insufficient tone to provide enough resistance to keep arterial blood pressure at a level sufficient to provide adequate blood flow to meet the metabolic needs of the brain and other tissues in the body. Thus, a basal level of sympathetic tone is essential.

Figure 35.8 Functional organization of the CNS regulation of cardiovascular function. The medulla contains sympathetic neurons, including sympathetic premotor neurons that are responsible for the significant level of basal sympathetic nerve discharge to the heart and vasculature. Although the medullary mechanisms underlying the generation of basal sympathetic tone remain to be identified, they result in the activity of medullary sympathetic premotor neurons that provides the essential excitation to spinal sympathetic preganglionic neurons that maintains the basal sympathetic tone to the heart and vasculature. The basal activity of sympathetic nerves maintains a release of norepinephrine in the heart and vascular beds that supports, respectively, the rate and strength of cardiac pumping and the constriction of blood vessels that produces the resistance to blood flow required to maintain blood pressure at a level sufficient to ensure adequate blood flow to the brain, even against the pull of gravity. Selective activation or inhibition of the sympathetic outflow to the vascular beds in different tissues allows relative reductions or increases in blood flow to those tissues as needed to support a wide range of behaviors and of homeostatic adjustments to environmental challenges. Many of the stimuli for changes in sympathetic and vagal outflow to the cardiovascular system, from homeostatic reflexes, to diseases, dehydration, drugs, and metabolism, elicit their responses through inputs, including those from the hypothalamus, to the medullary networks, including the premotor neurons, that generate sympathetic tone.

Sympathetic Premotor Neurons and Generation of Basal Sympathetic Tone

The cardiovascular sympathetic preganglionic neurons receive their primary excitatory drive from premotor neurons in the rostral ventrolateral medulla (RVLM). These bulbospinal neurons (Fig. 35.9) are glutamatergic, but many also contain the enzymes for catecholamine synthesis (C1 cell group). Silencing these neurons with local injections of muscimol (a GABA_A receptor agonist), or transection of their bulbospinal axons as in spinal cord injury, results in a precipitous fall in arterial pressure due to marked reductions in cardiac output and in vascular resistance. RVLM neurons have a significant basal discharge rate that supports basal cardiovascular sympathetic nerve activity. Modulation of this activity underlies a variety of reflex- and centrally-evoked changes in sympathetic nerve activity, including the baroreceptor and chemoreceptor reflexes and the response to physical exercise, which are described below. Other brainstem and hypothalamic neurons also project to the intermediolateral nucleus in the thoracic spinal cord and thus also act as sympathetic premotor neurons. However, interruption of their activity does not eliminate basal sympathetic tone.

Figure 35.9 Anatomical schematic of the organization of the arterial baroreceptor reflex pathways providing short-term stabilization of arterial pressure through feedback control of cardiac and vascular performance (see text for description). Amb, nucleus ambiguus, containing cardiac vagal preganglionic neurons; Cb, cerebellum; CVLM, caudal ventrolateral medulla, containing GABAergic, inhibitory interneurons; IML, intermediolateral nucleus of the thoracic spinal cord, containing cardiac and vasoconstrictor sympathetic preganglionic neurons; NTS, nucleus of the tractus solitarius, containing second-order, barosensory neurons; RVLM, rostral ventrolateral medulla, containing sympathetic premotor neurons projecting to the spinal cord to control the heart and blood vessels.

How and where is basal sympathetic tone generated? Understanding the generation and regulation of the basal sympathetic tone to the cardiovascular system is highly relevant, since exaggerated sympathetic outflow to the vasculature contributes to the elevated blood pressure in several forms of hypertension and an enhanced sympathetic outflow to the heart is a precipitating factor in cardiac arrhythmias leading to sudden cardiac death (Box 35.4). Blockade of various RVLM inputs has not been fruitful in identifying the primary sources controlling their activity, a key mystery in explaining the generation of the tonic cardiovascular sympathetic nerve activity. Although the RVLM neuronal pacemaker-like discharge seen under certain in vitro conditions was once proposed to provide the tonic drive, this does not appear to be the case because, under normal conditions, their rhythmic discharge is not endogenous, but rather is due to synaptic drive. Thus, although the RVLM sympathetic premotor neurons are essential elements in the circuit generating the basal cardiovascular sympathetic outflow (Fig. 35.8), the network or cellular mechanisms responsible for their basal activity remain unknown.

Box 35.4 Cardiovascular Disease

Alterations in the autonomic neural regulation of the cardiovascular system characterize the major cardiovascular diseases—contributing significantly to the etiology of the disease and to exacerbation of the symptoms and morbidity. Elevated sympathetic vasoconstrictor outflow plays a role in the increased vascular resistance seen in essential (primary) hypertension, the most prevalent form of high blood pressure and a major risk factor for stroke and congestive heart failure. Although the cause or trigger for this augmented stimulation of the CNS circuits generating vasoconstrictor sympathetic nerve activity is unknown, elevated dietary salt or increased central production of the hormone angiotensin II, both of which increase the discharge of RVLM sympathetic premotor neurons, are thought to contribute. Following myocardial infarction (heart attack), the cardiac sympathetic outflow is increased, at least partially the result of reflex drive to compensate for the reduced cardiac output, and this sympathetic drive to increase heart rate and cardiac contractility increases the stress on the weakened cardiac muscle, contributing to congestive heart failure and predisposing to arrhythmias. An imbalance in the neural regulation of the heart beat, with exaggerated cardiac sympathetic nerve activity and reduced cardiac vagal outflow, is a main cause of lethal arrhythmias and sudden cardiac death. Attempts to reduce mortality in such patients include surgical removal of the right stellate ganglion which contains the sympathetic ganglion neurons controlling heart rate. Increased understanding of the neural circuits and neurotransmitters controlling and generating sympathetic nerve activity to the cardiovascular system holds the promise of improved therapeutic approaches to a variety of cardiovascular diseases.

Shaun F. Morrison

The parasympathetic (vagal) preganglionic neurons influencing cardiac function are strongly excited by inputs from the NTS which has neurons receiving input from arterial baroreceptors that sense pressure in the major arteries, thereby providing the afferent drive for vagal baroreceptor reflex bradycardia (Fig. 35.9). Although the central circuits underlying the parasympathetic vasodilatory control of certain vascular beds (see above) remain relatively unknown, this parasympathetically mediated redistribution of blood flow is critical for the successful performance of certain physiological functions, such as reproduction.

Summary

Neural circuits, primarily within the brainstem, are responsible for generating the modest level of vasoconstrictor sympathetic nerve activity to blood vessels that must be sustained to provide the vascular tone necessary to maintain a sufficient level of arterial pressure to perfuse all the capillary beds in the body, but especially those in the brain and heart, with the O₂- and nutrient-rich blood necessary for survival. Although the cellular and neural network mechanisms through which this basal sympathetic nerve activity is generated remain unknown, the vasoconstrictor sympathetic premotor neurons in the RVLM are the key medullary output neurons of this generating network, providing the glutamatergic excitation necessary to drive the spinal sympathetic circuits whose activity maintains vascular tone. Similarly organized, but distinct brainstem and spinal circuits generate and control the cardiac and other sympathetic outflows, contributing to the optimized tissue function necessary for homeostasis, the support of behavior and organismal survival.

Sensory Regulation of Respiratory and Cardiovascular Systems

The principal homeostatic goal in the control of breathing is the regulation of the pressures of O₂ and CO₂ and pH in arterial blood—that is, arterial PO₂ = 80–100 mmHg, PCO₂ = 35–45 mm Hg and pH = 7.35–7.45. Such regulation requires their constant measurement by sensors connected in a feedback network to control breathing. Decreases in either PO₂ or pH or an increase in PCO₂ stimulate breathing, whereas increases in PO₂ or pH or a decrease in PCO₂ depress breathing. Similarly, a principal homeostatic goal in the regulation of the cardiovascular system is the maintenance of adequate blood flow to the brain. To meet this requirement, the CNS receives baroreceptor sensory input related to the stretch in the walls of the major arteries and modifies cardiovascular performance to sustain an arterial pressure sufficient to perfuse the brain with oxygenated blood.

Controlling Breathing and Cardiovascular Function; O₂ and CO₂ Are Measured By Chemoreceptors

The consequences of anoxia are catastrophic (particularly for the brain), but the effects of modestly depressed or elevated PO₂ are benign (due in part to the unique way blood hemoglobin reversibly binds O₂). However, small changes in PCO₂ (with consequent changes in pH) can profoundly affect cellular metabolism. Ventilation is very sensitive to small changes in PCO₂: a 1 mmHg increase in PCO₂ leads to a 2 liter × min⁻¹ increase in ventilation at rest (~5 liter × min⁻¹ in an adult human). In other words, a 2.5% increase in PCO₂ at rest leads to a 40% increase in ventilation! Thus, at rest and typically in exercise, ventilation is controlled mostly to regulate CO₂; however, in severe hypoxia, everything else is ignored. For example, compared to the normal ambient PO₂ at sea level (~160 mmHg), the exceptionally low PO₂ (~40 mm Hg) at the top of Mount Everest results in such a strong ventilatory drive that PCO₂ in climbers without supplemental O₂ is exceptionally low (~10 mm Hg compared with normal ~40 mm Hg). The sites and mechanisms of CO₂ chemoreception are the subject of intense scrutiny (Feldman, Mitchell, & Nattie, 2003). While the carotid bodies are sensitive to changes in PCO₂, a robust CO₂ response remains in peripherally chemo-denervated, vagotomized mammals, indicating the existence of intracranial sensors for CO₂ or related variables (pH, HCO₃⁻). A number of structures are implicated in intracranial chemotransduction (Feldman et al., 2003; Huckstepp & Dale, 2011). Among these, neurons within the RTN (Guyenet et al., 2010) and in the raphe (Richerson, 2004; Ray et al., 2011) respond robustly to changes in CO₂/pH within the physiological range and project to respiratory nuclei.

The mammalian brain is extremely sensitive to O₂ deprivation: several minutes of anoxia can initiate a cascade leading to loss of consciousness and neuronal death. Accordingly, the arterial chemoreceptors, the principal O₂ sensors for the entire body, are located in the carotid bodies in the bifurcation of the common carotid artery, the main portal through which O₂-enriched blood is transported to the brain. Arterial chemoreceptor afferents enter the brain via the glossopharyngeal nerve (cranial nerve IX) and terminate on second-order sensory neurons located primarily in the commissural region of the NTS.

Arterial chemoreceptor signals account for only a small part of the chemical drive to breathe during wakefulness, as breathing 100% O₂ reduces peripheral chemoreceptor discharge to near zero but only decreases ventilation by ~15% in awake mammals. Decreasing PO₂ has relatively little influence on chemoreceptor activity and ventilation until it falls to ≤60 mm Hg. With further reductions in PO₂, chemoreceptor discharge and ventilation increase exponentially.

In contrast to the global effects of hypoxia on breathing, the cardiovascular response is compartmentalized, directed toward conserving the O₂ availability for the brain by increasing brain blood flow while simultaneously reducing blood flow to other tissues. This is accomplished primarily by a widespread sympathetically induced vasoconstriction, which is driven by the arterial chemoreceptors and mediated by a strong activation of RVLM sympathetic premotor neurons. Although the brainstem circuitry underlying arterial chemoreceptor-stimulated excitation of RVLM sympathetic premotor neurons remains to be identified, a role for the RTN has been strongly implicated (Moreira, Takakura, Colombari, & Guyenet, 2006). In extreme cases, vasoconstriction evoked by severe hypoxia can reduce blood flow sufficiently to cause organ damage—for example, renal failure. Interestingly, although diving mammals such as seals also conserve brain O₂ availability by slowing their heart rate (by vagal activation) and maximally constricting their muscle blood vessels during their dives (hypoxic breath-holding), they avoid hypoxic tissue damage and maintain use of skeletal muscle through adaptations that include a prominent increase in myoglobin concentration.

Baroreceptor Reflex Regulation of Cardiovascular Performance

With the primary goal of maintaining adequate perfusion of the brain with sufficient O₂-rich arterial blood to support its continuous and high level metabolism, two regulatory circuits are critical for the maintenance of blood pressure (via the baroreceptor reflex) and for the distribution of blood flow (via the arterial chemoreceptor reflex). The baroreceptor reflex can be modeled as a simple negative feedback circuit in which changes in arterial pressure evoke compensatory changes in cardiovascular sympathetic nerve activity and cardiac parasympathetic vagal nerve activity to affect cardiac output and vascular resistance in an effort to return arterial pressure toward its baseline level. Figure 35.9 illustrates the pathways mediating the baroreceptor reflex-mediated changes in autonomic outflows. The sensors in the baroreceptor reflex are stretch receptors (baroreceptors), sensitive to changes in the pressure across the walls of the major arteries, and with axons in the vagus and glossopharyngeal nerves that project to second-order NTS neurons (Fig. 35.9). Thus, the activity of NTS barosensory neurons tracks increases and decreases in arterial pressure, including the rapid rise and fall in arterial pressure that occurs during the systolic phase of each cardiac cycle. Second-order glutamatergic barosensory neurons in the NTS project to GABAergic neurons in the caudal ventrolateral medulla (CVLM) that, in turn, project to the RVLM to inhibit cardiovascular sympathetic premotor neurons (Fig. 35.9). Thus, for example, a fall in arterial pressure will induce the following sequence: decreased activity of barosensory NTS neurons leading to disfacilitation of CVLM inhibitory interneurons and subsequent disinhibition of RVLM sympathetic premotor neurons causing increased preganglionic activity resulting in cardiac stimulation and vasoconstriction, acting together to counter the fall in arterial pressure. Concurrently, the reduced firing of NTS barosensory neurons decreases the excitation of cardiac vagal preganglionic neurons in the nucleus ambiguus (Fig. 35.9), thereby diminishing their inhibitory influence on the heart, effectively amplifying the effects of increases in cardiac sympathetic nerve activity. Each of the synaptic integration sites in the baroreceptor reflex pathway present the potential for other inputs to modulate the effectiveness of the reflex. For instance, both arterial pressure and heart rate must rise to provide adequate perfusion of skeletal muscle during escape behavior (or exercise, see below) and this is accomplished, in part, by presynaptic inhibition of the baroreceptor afferent inputs to second-order barosensory NTS neurons from hypothalamic neurons activated during these behaviors.

Summary

Chemoreceptors in the carotid bodies and in the brain provide sensory input to the central circuits controlling breathing and cardiovascular function. PCO₂ is normally regulated within narrow limits over a wide range of metabolic demands and this usually assures an adequate O₂ supply to the brain. However, during significant hypoxia, arterial chemoreceptors sensitive to O₂ can provide a powerful stimulus to increase breathing. Central chemoreceptors are exquisitely sensitive to changes in PCO₂ and in pH and can evoke large and rapid increases in respiration. Baroreceptors transduce the pressure in the major arteries to neural signals that modulate the autonomic outflows to the cardiovascular system, thereby stabilizing arterial pressure and maintaining adequate blood flow to the brain.

Modulation and Plasticity of Respiratory Motor Output

In meeting the homeostatic goal of maintaining normal arterial PO₂ and PCO₂, breathing must continually be adapted to meet changes in the state of the body—for example, posture—and in the environment (e.g., sojourning to or living at high altitude), and, in particular, to compensate for a variety of chronic cardiopulmonary diseases that result in low arterial PO₂.

Adaptations in the control of breathing in response to hypoxemia are examples of neural plasticity (Feldman et al., 2003). Inhaling an hypoxic gas mixture for 10–20 min elicits a progressive increase in ventilation. This slow increase in breathing at the onset of hypoxia is matched with a slow posthypoxic decline back to normal ventilation that is termed short-term potentiation (arrowheads in Fig. 35.10). Thus, even though the afferent input to the respiratory system from arterial and central chemoreceptors can change abruptly, the respiratory response consists of slower, smoother changes. This is important since abrupt changes can result in an unstable breathing pattern such as Cheyne-Stokes respiration (Fig. 35.4 and see Box 35.2), often seen in patients with severe congestive heart failure.

Figure 35.10 Respiration-related neural activity in an anesthetized and artificially ventilated rat exhibits respiratory long-term facilitation. Peak electrical activity in the phrenic nerve corresponds to the magnitude of inspiration. When arterial O₂ and CO₂ are normal (prehypoxia), the phrenic bursts are rhythmic and consistent (shown with expanded time scale). Upon exposure to intermittent (upper trace) or continuous (lower trace) hypoxia, phrenic bursts increase in amplitude, reflecting the drive to take deeper breaths when hypoxia-sensitive chemoreceptors are activated. In both patterns of hypoxia, phrenic nerve activity slowly returns nearly to baseline over a period of several minutes in a process termed short-term potentiation (STP; open arrowheads). Following intermittent, but not continuous hypoxia, phrenic burst amplitude progressively increases over an hour. This slow increase reflects respiratory long-term facilitation (LTF).

Courtesy of T. L. Baker and G. S. Mitchell.

A longer-lasting plasticity is revealed following a repetitive series of brief episodes of hypoxia, as may occur in a patient with sleep apnea (see Box 35.5). As illustrated in Figure 35.10, following the final hypoxic episode, ventilation transiently returns toward control levels, but is followed by a progressive augmentation that lasts for several hours termed respiratory long-term facilitation. Like other forms of neural plasticity in the brain, respiratory long-term facilitation is dependent on activation of serotonergic and noradrenergic inputs to respiratory motoneurons. This activates protein kinases, such as protein kinase C, thereby initiating an intracellular signaling cascade that amplifies the synaptic currents arising from glutamatergic respiratory premotoneurons. In response to a single prolonged hypoxic exposure, protein phosphatases constrain activity in this pathway and prevent long-term facilitation. However, repetitive bouts of hypoxia decrease the activity of phosphatases such as the serine/threonine phosphatases, thereby removing this constraint and permitting the development of long-term facilitation (Wilkerson, Macfarlane, Hoffman, & Mitchell, 2007).

Box 35.5 Disorders of Breathing are Widespread and Serious

Breathing at rest is normally controlled to maintain arterial PCO₂ at 40±2 mm Hg. However, cardiopulmonary pathology can increase PCO₂ markedly. Narcotics, barbiturates, and most general anesthetics depress breathing and at high doses, as can happen when self-administered, these drugs can cause death by respiratory failure. For opioids, respiratory depression appears to be due to depressant actions directly on preBötzinger complex neurons.

Sleep apnea. Sleep is associated with a modest decrease in ventilation and a reduced responsiveness to deviations in PCO₂ and PO₂. With the onset of sleep, the breathing pattern can become unstable, and apnea (defined as at least 10 s without breathing) can occur. In the most common form, obstructive sleep apnea, there is a loss of tone in airway muscles and tongue, so that the reduction in airway pressure during inspiration can pull in on the walls of the upper airway sufficiently to cause vibration—that is, snoring, even airway obstruction. Obstruction reduces or abolishes ventilation, raising CO₂ and lowering O₂. This in turn can cause arousal, and with it restoration of airway muscle tone and airway patency. In serious cases, the cycle of sleep, airway obstruction, and hypoxia-induced waking repeats hundreds of times every night. The marked disruption of sleep can cause debilitating hypersomnolesence, and severe obstructive sleep apnea can also have other sequelae, such as hypertension. In central sleep apnea, pauses in breathing result from the failure of respiratory drive. Mechanisms underlying central apneas are not understood, but the problem is prevalent during sleep at the end stages of a variety of neurodegenerative diseases, including amyotrophic lateral sclerosis, Parkinson’s disease, and multiple systems atrophy (where there is evidence of loss of preBötC neurons).

Sudden infant death syndrome (SIDS). SIDS is defined as the unexpected sudden death of an infant or young child that cannot be explained by a postmortem examination. SIDS is the leading cause of death of infants between 1 month and 1 year of age in the United States. Although a number of etiologies are likely to contribute to SIDS, current hypotheses tend to focus on abnormalities of cardiorespiratory control. The apnea hypothesis of SIDS attributes apnea to disorders of both the chemical control of breathing and the arousal response to insufficient ventilation. Infants who later succumb to SIDS have been described as exhibiting irregular breathing patterns, including periods of apnea, and depressed arousal responses to hypoxia or hypercapnia. Epidemiologically, SIDS has been associated with infants sleeping face down, where pillows, blankets, and mattress can limit the diffusion of expired air, resulting in elevated CO₂ and lower O₂ in inspired air and, consequently, in arterial blood. This in turn could depress breathing sufficiently to cause respiratory arrest and death.

Genetic Diseases Affecting Breathing

Congenital Central Hypoventilation Syndrome (CCHS; Weese-Mayer et al., 2009). CCHS is a potentially life-threatening genetic disorder typically presenting in newborns and characterized by respiratory insufficiency especially during sleep. This breathing phenotype is the consequence of an impaired ventilatory response to hypercapnia (increased blood PCO₂) and hypoxemia (decreased blood PO₂). Additionally, the disease is often associated with functional impairments of the autonomic nervous system and tumors of neural crest derivatives. The defining mutation in CCHS is a polyalanine expansion in Phox2b. The extent of the polyalanine repeat mutation correlates with the severity of the disease, with the most severe cases requiring continuous artificial ventilation. The offspring of mice genetically modified to contain the polyalanine expansion have irregular breathing at birth, do not increase their breathing in response to CO₂ stimulation of chemoreceptors, and die shortly after birth. Postmortem examination of these mice provided a likely explanation for the depression of breathing. Chemoreceptor neurons in the RTN (Fig. 35.2) normally express Phox2b but in mouse pups with the Phox2b polyalanine expansion these neurons were absent. Thus, the Phox2b mutation appears to result in a loss of RTN neurons that reduces the normal respiratory drive arising from chemoreceptors.

Rett Syndrome (RTT; Lioy, Wu, & Bissonnette, 2011). RTT is a developmental disorder of the central nervous system resulting from mutations in the X-linked gene methyl-CpG-binding protein 2 (Mecp2) and is usually seen in girls. It results in severe neurological symptoms that include speech and movement disorders as well as irregular breathing (e.g., Cheyne-Stokes breathing; Fig 35.4 and Box 35.2) that can include life-threatening periods of apnea. The onset of symptoms typically begins at about 6–18 months of age at which time there is a slowing down of motor skills and then a regression in communication skills.

Jack L. Feldman and Donald R. McCrimmon

Exercise: An Example of Cardiopulmonary Integration

Physical activity involving repetitive movement of large muscles significantly increases the metabolic consumption of O₂ and production of CO₂. Even during mild activity, such as walking, the metabolic changes require substantial ventilatory and cardiac output increases to maintain arterial and brain levels of CO₂ and O₂ within normal limits. Impressively, the cardiorespiratory changes are proportional to the changes in metabolism and there is almost no change in arterial PCO₂ or PO₂ with mild or moderate intensity exercise. During high intensity exercise, the increase in breathing can result in a decrease (!) in arterial PCO₂. The primary cardiovascular adjustments accompanying exercise include not only the sympathetically mediated increases in heart rate and cardiac contractility necessary to elevate cardiac output required to support increased muscle metabolism, but also a widespread sympathetic vasoconstriction. This increased sympathetic input to the vasculature plays a significant role in optimizing exercise performance. Blood volume “stored” in the veins at rest is brought into the arterial circulation to increase the capacity for O₂ transport to exercising muscle. Arterial blood pressure is maintained in the face of a marked reduction in total peripheral vascular resistance due to the substantial vasodilation occurring in exercising muscle (driven mostly by local metabolic signals—for example, adenosine—rather than by a selective inhibition of sympathetic tone to exercising muscle). Metabolism and O₂ consumption in nonexercising tissues is significantly reduced—for example, the reduced blood flow to the GI tract accounts for the cramping that occurs with intense exercise immediately after a large meal. Sympathetically driven adrenal epinephrine secretion further supports both the increased cardiac output and the skeletal muscle hyperemia. Because the large amount of heat produced by exercising muscles (and the respiratory muscles engaged to support exercise) must be eliminated to prevent a significant rise in body temperature (potentially lethal hyperthermia), secondary, but critically important cardiovascular changes are engaged as exercise progresses: a reduction in sympathetic vasoconstrictor tone to the skin increases skin blood flow and heat loss, and sympathetic vasoconstriction to the kidney is increased to retain water and electrolytes that are required to maintain blood volume during the increased sweat production (as much as 1.5 l/h) required to enhance heat loss from the skin.

Despite its utility as a model of neural integration and widespread scientific fascination with the topic, the physiological mechanisms underlying the matching of changes in cardiorespiratory function to exercise remain mysterious. Since there is normally little change in CO₂ or O₂, classic negative feedback from arterial or brain chemoreceptors cannot adequately account for these cardiorespiratory adjustments. Thus, feedforward, descending command from higher brain systems and sensory input from joint and muscle receptors appears to be important.

The volitional initiation of exercise, signaled by suprapontine structures, likely involves activation of spinal pathways for locomotion and concurrent feedforward activation of brainstem mechanisms controlling breathing and cardiovascular function. This allows pulmonary ventilation and gas transport to change appropriately and quickly before there are any significant changes in blood gases (as would occur if the mechanism was strictly based on feedback). Commands for the activation of pulmonary and locomotor muscles presumably arise in similar ways in the motor cortex, thalamus, and basal ganglia. Once the motor command has been initiated, respiratory and locomotor activities are likely coordinated at subcortical levels, including the hypothalamus and periaqueductal gray. Electrical or chemical activation of the hypothalamus in the anesthetized, paralyzed cat elicits fictive locomotion paired with proportional increases in respiratory motor output (suggestive of exercise-related increases in ventilation, termed exercise hyperpnea) and redistribution of blood flow consistent with the induction of locomotion (Waldrop, Eldridge, Iwamoto, & Mitchell, 1996). Because the animals were paralyzed, muscle activation did not occur, and thus the changes in breathing were the result of feedforward control rather than sensory feedback from joint or muscle receptors.

At the level of the brainstem, coordination of the respiratory and cardiovascular responses likely involves circuitry in the dorsolateral pons, in the vicinity of the PRG. Descending inputs from the hypothalamus, periaqueductal gray, and dorsolateral pons activate medullary respiratory circuits to increase both the depth of each breath—that is, tidal volume—and respiratory frequency, which almost certainly involves modulation of the activity of the rhythm generating circuitry in the preBötC. The need for an increase in the size of each breath means there is also a parallel increase in the activity of pattern forming circuits in, for example, the rVRG. There is also an increase in the contribution of expiratory muscles. Expiratory rhythm generating neurons in the region of the pFRG appear to be synaptically inhibited at rest and this inhibition may be removed by signals related to exercise, possibly through the influence of descending inputs from the higher brain structures. Through reciprocal connections between the pFRG and preBötC, activation of the pFRG may contribute to the increase in breathing frequency. Activation of the pFRG will also increase tidal volume through activation of expiratory pattern-forming circuits in the caudal VRG.

The increases in cardiac, vasoconstrictor, and adrenal sympathetic outflows necessary for exercise are mediated by enhanced discharge of the relevant sympathetic premotor neurons in the RVLM. Although the exercise-related stimulus and the source of the increased excitatory drive to these premotor neurons are unknown, their excitatory inputs from central respiratory neurons are likely to contribute significantly. Also contributing to the sympathetic activation is a centrally-mediated reduction in the effectiveness of the inhibitory baroreceptor reflex, due to modulation of the synaptic efficacy in the NTS, which would otherwise inhibit sympathetic outflow as arterial blood pressure increases during exercise. The increased central respiratory drive may also contribute to the activation of adrenal sympathetic premotor neurons in the RVLM during exercise, which leads to secretion of adrenal epinephrine and a potent stimulation of cardiac output and skeletal muscle vasodilation. As the body heat load rises during exercise, warm-sensitive, thermoregulatory neurons in the preoptic hypothalamus increase their inhibitory influence on the cutaneous vasoconstrictor premotor neurons in the raphe nuclei of the ventromedial medulla to dilate skin blood vessels. Simultaneously, these preoptic neurons excite the sympathetic outflow to sweat glands to augment heat loss. As the osmolarity of the blood increases due to sweating, a specialized, osmosensitive neural network in the anterior hypothalamus increases the activity of renal sympathetic premotor neurons in the RVLM to augment the sympathetic outflow to the kidney to reduce renal filtration and help sustain blood volume.

Summary

Breathing and cardiovascular function are fundamental and intertwined physiologic functions that must be controlled vigilantly by the brain. This chapter has explored the basic mechanisms underlying rhythm generation, sensory processing, and motor output for these regulatory systems.

References

1. Adrian ED, Buytendijk FJJ. Potential changes in the isolated brainstem of goldfish. Journal of Physiology (London). 1931;71:121–135.

2. Alheid GF, McCrimmon DR. The chemical neuroanatomy of breathing. Respiratory Physiology & Neurobiology. 2008;164:3–11.

3. Canning BJ, Spina D. Sensory nerves and airway irritability. Handbook of Experimental Pharmacology. 2009;194:139–183.

4. Feldman JL. Neurophysiology of breathing in mammals. In: Bloom FE, ed. Handbook of physiology; section I: the nervous system; volume IV: intrinsic regulatory systems of the brain. Bethesda, MD: American Physiological Society; 1986:463–524.

5. Feldman JL, Del Negro CA. Looking for inspiration: new perspectives on the generation of respiratory rhythm. Nature Reviews Neuroscience. 2006;7:232–242.

6. Feldman JL, Mitchell GS, Nattie EE. Breathing: Rhythmicity, plasticity, chemosensitivity. Annual Review of Neuroscience. 2003;26:239–266.

7. Gesell R, Bricker J, Magee C. Structural and functional organization of the central mechanism controlling breathing. The American Journal of Physiology. 1936;117:423–452.

8. Guyenet PG, Stornetta RL, Abbott SB, Depuy SD, Fortuna MG, Kanbar R. Central CO2 chemoreception and integrated neural mechanisms of cardiovascular and respiratory control. Journal of Applied Physiology. 2010;108(4):995–1002.

9. Hayashi F, McCrimmon DR. Respiratory motor responses to cranial nerve afferent stimulation in rats. The American Journal of Physiology. 1996;271(4 Pt 2):R1054–1062.

10. Huckstepp RT, Dale N. Redefining the components of central CO2 chemosensitivity - towards a better understanding of mechanism. Journal of Physiology. 2011;589(Pt 23):5561–5579.

11. Janczewski WA, Feldman JL. Distinct rhythm generators for inspiration and expiration in the juvenile rat. Journal of Physiology. 2006;570(Pt 2):407–420.

12. Kubin L, Alheid GF, Zuperku EJ, McCrimmon DR. Central pathways of pulmonary and lower airway vagal afferents. Journal of Applied Physiology. 2006;101(2):618–627.

13. Legallois M. Experiments on the principle of life Philadelphia, PA: Thomas; 1813.

14. Lioy DT, Wu WW, Bissonnette JM. Autonomic dysfunction with mutations in the gene that encodes methyl-CpG-binding protein 2: insights into Rett syndrome. Autonomic Neuroscience: Basic & Clinical. 2011;161(1–2):55–62.

15. Lorry M. Les mouvements du cerveau. Mem Math Phys Pres Acad Roy Sci Div Sav Paris. 1760;3:344–377.

16. Mellen NM, Janczewski WA, Bocchiaro CM, Feldman JL. Opioid-induced quantal slowing reveals dual networks for respiratory rhythm generation. Neuron. 2003;37(5):821–826.

17. Moreira TS, Takakura AC, Colombari E, Guyenet PG. Central chemoreceptors and sympathetic vasomotor outflow. The Journal of Physiology. 2006;577(Pt 1):369–386.

18. Pagliardini S, Janczewski WA, Tan W, Dickson CT, Deisseroth K, Feldman JL. Active expiration induced by excitation of ventral medulla in adult anesthetized rats. Journal of Neuroscience. 2011;31:2895–2905.

19. Plum F, Posner JB. The diagnosis of stupor and coma Contemporary Neurology Series. Vol. 19 3rd ed. Philadelphia, PA: Davis; 1980.

20. Ramon y Cajal S. Histologie du Systeme Nerveux de l’homme et des Vertebres Paris: Maloine; 1909.

21. Ray RS, Corcoran AE, Brust RD, et al. Impaired respiratory and body temperature control upon acute serotonergic neuron inhibition. Science. 2011;333(6042):637–642.

22. Rekling JC, Feldman JL. Calcium-dependent plateau potentials in rostral ambiguus neurons in the newborn mouse brain stem in vitro. Journal of Neurophysiology. 1997;78(5):2483–2492.

23. Richerson GB. Serotonergic neurons as carbon dioxide sensors that maintain pH homeostasis. Nature Reviews Neuroscience. 2004;5(6):449–461.

24. Smith JC, Ellenberger HH, Ballanyi K, Richter DW, Feldman JL. Pre-Bötzinger complex: a brainstem region that may generate respiratory rhythm in mammals. Science. 1991;254(5032):726–729.

25. Suzue T. Respiratory rhythm generation in the in vitro brain stem-spinal cord preparation of the neonatal rat. The Journal of Physiology. 1984;354:173–183.

26. Tan W, Janczewski WA, Yang P, Shao XM, Callaway EM, Feldman JL. Silencing preBötzinger complex somatostatin-expressing neurons induces persistent apnea in awake rat. Nature Neuroscience. 2008;11(5):538–540.

27. Weese-Mayer DE, Rand CM, Berry-Kravis EM, et al. Congenital central hypoventilation syndrome from past to future: model for translational and transitional autonomic medicine. Pediatric Pulmonology. 2009;44(6):521–535.

28. Wilkerson JE, Macfarlane PM, Hoffman MS, Mitchell GS. Respiratory plasticity following intermittent hypoxia: roles of protein phosphatases and reactive oxygen species. Biochemical Society Transactions. 2007;35(Pt 5):1269–1272.

Suggested Readings

1. Guyenet PG. The sympathetic control of blood pressure. Nature Reviews Neuroscience. 2006;7:335–346.

2. Malpas SC. Sympathetic nervous system overactivity and its role in the development of cardiovascular disease. Physiological Reviews. 2010;90:513–557.

3. Waldrop TG, Eldridge FL, Iwamoto GA, Mitchell JH. Central neural control of respiration and circulation during exercise. In: Rowell LB, Shepherd JT, eds. Handbook of physiology sec 12 exercise: regulation and integration of multiple systems. New York: Oxford University Press; 1996:333–380.