Evolution of air-breathing and central CO(2)/H(+) respiratory chemosensitivity: new insights from an old fish?

The control of breathing in fishes - historical perspectives and the path ahead

The study of breathing in fishes has featured prominently in Journal of Experimental Biology (JEB), particularly during the latter half of the past century. Indeed, many of the seminal discoveries in this important sub-field of comparative respiratory physiology were reported first in JEB. The period spanning 1960-1990 (the 'golden age of comparative respiratory physiology') witnessed intense innovation in the development of methods to study the control of breathing. Many of the guiding principles of piscine ventilatory control originated during this period, including our understanding of the dominance of O2 as the driver of ventilation in fish. However, a critical issue - the identity of the peripheral O2 chemoreceptors - remained unanswered until methods for cell isolation, culture and patch-clamp recording established that gill neuroepithelial cells (NECs) respond to hypoxia in vitro. Yet, the role of the NECs and other putative peripheral or central chemoreceptors in the control of ventilation in vivo remains poorly understood. Further progress will be driven by the implementation of genetic tools, most of which can be used in zebrafish (Danio rerio). These tools include CRISPR/Cas9 for selective gene knockout, and Tol2 systems for transgenesis, the latter of which enables optogenetic stimulation of cellular pathways, cellular ablation and in vivo cell-specific biosensing. Using these methods, the next period of discovery will see the identification of the peripheral sensory pathways that initiate ventilatory responses, and will elucidate the nature of their integration within the central nervous system and their link to the efferent motor neurons that control breathing.

A brainstem preparation allowing simultaneous access to respiratory motor output and cellular properties of motoneurons in American bullfrogs

Breathing is generated by a complex neural circuit, and the ability to monitor the activity of multiple network components simultaneously is required to uncover the cellular basis of breathing. In neonatal rodents, a single brainstem slice can be obtained to record respiratory-related motor nerve discharge along with individual rhythm-generating cells or motoneurons because of the close proximity of these neurons in the brainstem. However, most ex vivo preparations in other vertebrates can only capture respiratory motor outflow or electrophysiological properties of putative respiratory neurons in slices without relevant synaptic inputs. Here, we detail a method to horizontally slice away the dorsal portion of the brainstem to expose fluorescently labeled motoneurons for patch-clamp recordings in American bullfrogs. This 'semi-intact' preparation allows tandem recordings of motor output and single motoneurons during respiratory-related synaptic inputs. The rhythmic motor patterns are comparable to those from intact preparations and operate at physiological temperature and [K+]. Thus, this preparation provides the ability to record network and cellular outputs simultaneously and may lead to new mechanistic insights into breathing control across vertebrates.

Astrocytic contribution to glutamate-related central respiratory chemoreception in vertebrates

Central respiratory chemoreceptors play a key role in the respiratory homeostasis by sensing CO₂ and H⁺ in brain and activating the respiratory neural network. This ability of specific brain regions to respond to acidosis and hypercapnia is based on neuronal and glial mechanisms. Several decades ago, glutamatergic transmission was proposed to be involved as a main mechanism in central chemoreception. However, a complete identification of mechanism has been elusive. At the rostral medulla, chemosensitive neurons of the retrotrapezoid nucleus (RTN) are glutamatergic and they are stimulated by ATP released by RTN astrocytes in response to hypercapnia. In addition, recent findings show that caudal medullary astrocytes in brainstem can also contribute as CO₂ and H⁺ sensors that release D-serine and glutamate, both gliotransmitters able to activate the respiratory neural network. In this review, we describe the mammalian astrocytic glutamatergic contribution to the central respiratory chemoreception trying to trace in vertebrates the emergence of several components involved in this process.

Ventilatory responses of the clown knifefish, Chitala ornata, to arterial hypercapnia remain after gill denervation

The aim of this study was to corroborate the presence of CO₂/H⁺-sensitive arterial chemoreceptors involved in producing air-breathing responses to aquatic hypercarbia in the facultative air-breathing clown knifefish (Chitala ornata) and to explore their possible location. Progressively increasing levels of CO₂ mixed with air were injected into the air-breathing organ (ABO) of one group of intact fish to elevate internal PCO₂ and decrease blood pH. Another group of fish in which the gills were totally denervated was exposed to aquatic hypercarbia (pH ~ 6) or arterial hypercapnia in aquatic normocarbia (by injection of acetazolamide to increase arterial PCO₂ and decrease blood pH). Air-breathing frequency, gill ventilation frequency, heart rate and arterial PCO₂ and pH were recorded during all treatments. The CO₂ injections into the ABO induced progressive increases in air-breathing frequency, but did not alter gill ventilation or heart rate. Exposure to both hypercarbia and acetazolamide post-denervation of the gills also produced significant air-breathing responses, but no changes in gill ventilation. While all treatments produced increases in arterial PCO₂ and decreases in blood pH, the modest changes in arterial PCO₂/pH in the acetazolamide treatment produced the greatest increases in air-breathing frequency. These results strengthen the evidence that internal CO₂/H⁺ sensing is involved in the stimulation of air breathing in clown knifefish and suggest that it involves extra-branchial chemoreceptors possibly situated either centrally or in the air-breathing organ.

Learning to Air-Breathe: The First Steps

Air-breathing in vertebrates has evolved many times among the bony fish while in water. Its appearance has had a fundamental impact on the regulation of ventilation and acid-base status. We review the physico-chemical constraints imposed by water and air, place the extant air-breathing fish into this framework, and show how that the advantages of combining control of ventilation and acid-base status are only available to the most obligate of air-breathing fish, thus highlighting promising avenues for research.

Control of air-breathing in fishes: Central and peripheral receptors

This review considers the environmental and systemic factors that can stimulate air-breathing responses in fishes with bimodal respiration, and how these may be controlled by peripheral and central chemoreceptors. The systemic factors that stimulate air-breathing in fishes are usually related to conditions that increase the O₂ demand of these animals (e.g. physical exercise, digestion and increased temperature), while the environmental factors are usually related to conditions that impair their capacity to meet this demand (e.g. aquatic/aerial hypoxia, aquatic/aerial hypercarbia, reduced aquatic hidrogenionic potential and environmental pollution). It is now well-established that peripheral chemoreceptors, innervated by cranial nerves, drive increased air-breathing in response to environmental hypoxia and/or hypercarbia. These receptors are, in general, sensitive to O₂ and/or CO₂/H⁺ levels in the blood and/or the environment. Increased air-breathing in response to elevated O₂ demand may also be driven by the peripheral chemoreceptors that monitor O₂ levels in the blood. Very little is known about central chemoreception in air-breathing fishes, the data suggest that central chemosensitivity to CO₂/H⁺ is more prominent in sarcopterygians than in actinopterygians. A great deal remains to be understood about control of air-breathing in fishes, in particular to what extent control systems may show commonalities (or not) among species or groups that have evolved air-breathing independently, and how information from the multiple peripheral (and possibly central) chemoreceptors is integrated to control the balance of aerial and aquatic respiration in these animals.

Central control of air breathing in fishes

The diversity of sites and surfaces that are utilized for gas transfer from air to blood in fish is remarkable. While a few species do utilize their gills for gas exchange in air, this is a rare occurrence and most air-breathing fish utilize other surfaces including air-breathing organs and lungs. At present almost nothing is known about the central sites that initiate and regulate air breathing although hypotheses can be put forward based on our rudimentary understanding of the sites involved in water breathing in lampreys and teleost fishes, and those involved in air breathing in pre-metamorphic anuran ampibians. The pumps involved in producing both water and air breathing in fishes are highly conserved, a buccal pump, assisted by pharyngeal and/or parabranchial/opercular pumps, produce both forms of ventilation. What varies between species are the manner in which air breaths are produced (in two versus four phases), and the 'valving' involved in producing water flow over the gills versus air flow in and out of air-breathing organs. The latter suggests that a major step in the evolution of air breathing was the evolution of the mechanisms that control the flow of the respiratory medium. The neural matrix that underlies the co-ordination of the pump and the valving events remains enigmatic and in much need of further research.

How important is the CO2 chemoreflex for the control of breathing? Environmental and evolutionary considerations

Haldane and Priestley (1905) discovered that the ventilatory control system is highly sensitive to CO₂. This "CO₂ chemoreflex" has been interpreted to dominate control of resting arterial PCO₂/pH (P_aCO₂/pH_a) by monitoring P_aCO₂/pH_a and altering ventilation through negative feedback. However, P_aCO₂/pH_a varies little in mammals as ventilation tightly couples to metabolic demands, which may minimize chemoreflex control of P_aCO₂. The purpose of this synthesis is to (1) interpret data from experimental models with meager CO₂ chemoreflexes to infer their role in ventilatory control of steady-state P_aCO₂, and (2) identify physiological causes of respiratory acidosis occurring normally across vertebrate classes. Interestingly, multiple rodent and amphibian models with minimal/absent CO₂ chemoreflexes exhibit normal ventilation, gas exchange, and P_aCO₂/pH_a. The chemoreflex, therefore, plays at most a minor role in ventilatory control at rest; however, the chemoreflex may be critical for recovering P_aCO₂ following acute respiratory acidosis induced by breath-holding and activity in many ectothermic vertebrates. An apparently small role for CO₂ feedback in the genesis of normal breathing contradicts the prevailing view that central CO₂/pH chemoreceptors increased in importance throughout vertebrate evolution. Since the CO₂ chemoreflex contributes minimally to resting ventilation, these CO₂ chemoreceptors may have instead decreased importance throughout tetrapod evolution, particularly with the onset and refinement of neural innovations that improved the matching of ventilation to tissue metabolic demands. This distinct and elusive "metabolic ventilatory drive" likely underlies steady-state P_aCO₂ in air-breathers. Uncovering the mechanisms and evolution of the metabolic ventilatory drive presents a challenge to clinically-oriented and comparative respiratory physiologists alike.

Evolution of lung breathing from a lungless primitive vertebrate

Air breathing was critical to the terrestrial radiation and evolution of tetrapods and arose in fish. The vertebrate lung originated from a progenitor structure present in primitive boney fish. The origin of the neural substrates, which are sensitive to metabolically produced CO2 and which rhythmically activate respiratory muscles to match lung ventilation to metabolic demand, is enigmatic. We have found that a distinct periodic centrally generated rhythm, described as "cough" and occurring in lamprey in vivo and in vitro, is modulated by central sensitivity to CO2. This suggests that elements critical for the evolution of breathing in tetrapods, were present in the most basal vertebrate ancestors prior to the evolution of the lung. We propose that the evolution of breathing in all vertebrates occurred through exaptations derived from these critical basal elements.

Diving into the mammalian swamp of respiratory rhythm generation with the bullfrog

All vertebrates produce some form of respiratory rhythm, whether to pump water over gills or ventilate lungs. Yet despite the critical importance of ventilation for survival, the architecture of the respiratory central pattern generator has not been resolved. In frogs and mammals, there is increasing evidence for multiple burst-generating regions in the ventral respiratory group. These regions work together to produce the respiratory rhythm. However, each region appears to be pivotally important to a different phase of the motor act. Regions also exhibit differing rhythmogenic capabilities when isolated and have different CO2 sensitivity and pharmacological profiles. Interestingly, in both frogs and rats the regions with the most robust rhythmogenic capabilities when isolated are located in rhombomeres 7/8. In addition, rhombomeres 4/5 in both clades are critical for controlling phases of the motor pattern most strongly modulated by CO2 (expiration in mammals, and recruitment of lung bursts in frogs). These key signatures may indicate that these cell clusters arose in a common ancestor at least 400 million years ago.

Does ammonia trigger hyperventilation in the elasmobranch, Squalus acanthias suckleyi?

We examined the ventilatory response of the spiny dogfish, to elevated internal or environmental ammonia. Sharks were injected via arterial catheters with ammonia solutions or their Na salt equivalents sufficient to increase plasma total ammonia concentration [TAmm]a by 3-5 fold from 145±21μM to 447±150μM using NH4HCO3 and a maximum of 766±100μM using (NH4)2SO4. (NH4)2SO4 caused a small increase in ventilation frequency (+14%) and a large increase in amplitude (+69%), while Na2SO4 did not. However, CO2 partial pressure (PaCO2) also increased and arterial pHa and plasma bicarbonate concentration ([HCO3(-)]a) decreased. NH4HCO3 caused a smaller increase in plasma ammonia resulting in a smaller but significant, short lived increases in ventilation frequency (+6%) and amplitude (36%), together with a rise in PaCO2 and [HCO3(-)]a. Injection with NaHCO3 which increased pHa and [HCO3(-)]a did not change ventilation. Plasma ammonia concentration correlated significantly with ventilation amplitude, while ventilation frequency showed a (negative) correlation with pHa. Exposure to high environmental ammonia (1500μM NH4HCO3) did not induce changes in ventilation until plasma [TAmm]a increased and ventilation amplitude (but not frequency) increased in parallel. We conclude that internal ammonia stimulates ventilation in spiny dogfish, especially amplitude or stroke volume, while environmental ammonia only stimulates ventilation after ammonia diffuses into the bloodstream.

Rh protein expression in branchial neuroepithelial cells, and the role of ammonia in ventilatory control in fish

Bill Milsom has made seminal contributions to our understanding of ventilatory control in a wide range of vertebrates. Teleosts are particularly interesting, because they produce a 3rd, potentially toxic respiratory gas (ammonia) in large amounts. Fish are well known to hyperventilate under high environmental ammonia (HEA), but only recently has the potential role of ammonia in normal ventilatory control been investigated. It is now clear that ammonia can act directly as a ventilatory stimulant in trout, independent of its effects on acid-base balance. Even in ureotelic dogfish sharks, acute elevations in ammonia cause increases in ventilation. Peripherally, the detection of elevated ammonia resides in gill arches I and II in trout, and in vitro, neuroepithelial cells (NECs) from these arches are sensitive to ammonia, responding with elevations in intracellular Ca(2+) ([Ca(2+)]i). Centrally, hyperventilatory responses to ammonia correlate more closely with concentrations of ammonia in the brain than in plasma or CSF. After chronic HEA exposure, ventilatory responsiveness to ammonia is lost, associated with both an attenuation of the [Ca(2+)]i response in NECs, and the absence of elevation in brain ammonia concentration. Chronic exposure to HEA also causes increases in the mRNA expression of several Rh proteins (ammonia-conductive channels) in both brain and gills. "Single cell" PCR techniques have been used to isolate the individual responses of NECs versus other gill cell types. We suggest several circumstances (post-feeding, post-exercise) where the role of ammonia as a ventilatory stimulant may have adaptive benefits for O2 uptake in fish.

Effects of maturation and acidosis on the chaos-like complexity of the neural respiratory output in the isolated brainstem of the tadpole, Rana esculenta

Human ventilation at rest exhibits mathematical chaos-like complexity that can be described as long-term unpredictability mediated (in whole or in part) by some low-dimensional nonlinear deterministic process. Although various physiological and pathological situations can affect respiratory complexity, the underlying mechanisms remain incompletely elucidated. If such chaos-like complexity is an intrinsic property of central respiratory generators, it should appear or increase when these structures mature or are stimulated. To test this hypothesis, we employed the isolated tadpole brainstem model [Rana (Pelophylax) esculenta] and recorded the neural respiratory output (buccal and lung rhythms) of pre- (n = 8) and postmetamorphic tadpoles (n = 8), at physiologic (7.8) and acidic pH (7.4). We analyzed the root mean square of the cranial nerve V or VII neurograms. Development and acidosis had no effect on buccal period. Lung frequency increased with development (P < 0.0001). It also increased with acidosis, but in postmetamorphic tadpoles only (P < 0.05). The noise-titration technique evidenced low-dimensional nonlinearities in all the postmetamorphic brainstems, at both pH. Chaos-like complexity, assessed through the noise limit, increased from pH 7.8 to pH 7.4 (P < 0.01). In contrast, linear models best fitted the ventilatory rhythm in all but one of the premetamorphic preparations at pH 7.8 (P < 0.005 vs. postmetamorphic) and in four at pH 7.4 (not significant vs. postmetamorphic). Therefore, in a lower vertebrate model, the brainstem respiratory central rhythm generator accounts for ventilatory chaos-like complexity, especially in the postmetamorphic stage and at low pH. According to the ventilatory generators homology theory, this may also be the case in mammals.

Hypercarbic cardiorespiratory reflexes in the facultative air-breathing fish jeju (Hoplerythrinus unitaeniatus): the role of branchial CO2 chemoreceptors

The aim of the present study was to determine the roles that externally versus internally oriented CO2/H+-sensitive chemoreceptors might play in promoting cardiorespiratory responses to environmental hypercarbia in the air-breathing fish, Hoplerythrinus unitaeniatus (jeju). Fish were exposed to graded hypercarbia (1, 2.5, 5, 10 and 20% CO2) and also to graded levels of environmental acidosis (pH ~7.0, 6.0, 5.8, 5.6, 5.3 and 4.7) equal to the pH levels of the hypercarbic water to distinguish the relative roles of CO2versus H+. We also injected boluses of CO2-equilibrated solutions (5, 10 and 20% CO2) and acid solutions equilibrated to the same pH as the CO2 boluses into the caudal vein (internal) and buccal cavity (external) to distinguish between internal and external stimuli. The putative location of the chemoreceptors was determined by bilateral denervation of branches of cranial nerves IX (glossopharyngeal) and X (vagus) to the gills. The data indicate that the chemoreceptors eliciting bradycardia, hypertension and gill ventilatory responses (increased frequency and amplitude) to hypercarbia are exclusively branchial, externally oriented and respond specifically to changes in CO2 and not H+. Those involved in producing the cardiovascular responses appeared to be distributed across all gill arches while those involved in the gill ventilatory responses were located primarily on the first gill arch. Higher levels of aquatic CO2 depressed gill ventilation and stimulated air breathing. The chemoreceptors involved in producing air breathing in response to hypercarbia also appeared to be branchial, distributed across all gill arches and responded specifically to changes in aquatic CO2. This would suggest that chemoreceptor groups with different orientations (blood versus water) are involved in eliciting air-breathing responses to hypercarbia in jeju.

The phylogeny of central chemoreception

Respiratory chemoreceptors responsive to changes in CO(2)/H(+) appear to be present in all vertebrates from fish to birds and mammals. They appear to have arisen first in the periphery sensitive to the external environment. Thus, in most fish CO(2)/H(+) chemoreceptors reside primarily in the gills and respond to changes in aquatic rather than arterial P(CO)₂ . In the air-breathing tetrapods (amphibians, mammals, reptiles and birds), the branchial arches regress developmentally and the derivatives of the branchial arteries are now exclusively internal. The receptors associated with these arteries now sense only arterial (not environmental) P(CO)₂/pH . Central CO(2)/H(+) chemoreception also appears to have arisen with the advent of air breathing, presumably as a second line of defense. These receptors may have arisen multiple times in association with several (but not all) of the independent origins of air breathing in fishes. There is strong evidence for multiple central sites of CO(2)/H(+) sensing, at least in amphibians and mammals, suggesting that it may not only have originated multiple times in different species but also multiple times within a single species.

Control of respiration in fish, amphibians and reptiles

Fish and amphibians utilise a suction/force pump to ventilate gills or lungs, with the respiratory muscles innervated by cranial nerves, while reptiles have a thoracic, aspiratory pump innervated by spinal nerves. However, fish can recruit a hypobranchial pump for active jaw occlusion during hypoxia, using feeding muscles innervated by anterior spinal nerves. This same pump is used to ventilate the air-breathing organ in air-breathing fishes. Some reptiles retain a buccal force pump for use during hypoxia or exercise. All vertebrates have respiratory rhythm generators (RRG) located in the brainstem. In cyclostomes and possibly jawed fishes, this may comprise elements of the trigeminal nucleus, though in the latter group RRG neurons have been located in the reticular formation. In air-breathing fishes and amphibians, there may be separate RRG for gill and lung ventilation. There is some evidence for multiple RRG in reptiles. Both amphibians and reptiles show episodic breathing patterns that may be centrally generated, though they do respond to changes in oxygen supply. Fish and larval amphibians have chemoreceptors sensitive to oxygen partial pressure located on the gills. Hypoxia induces increased ventilation and a reflex bradycardia and may trigger aquatic surface respiration or air-breathing, though these latter activities also respond to behavioural cues. Adult amphibians and reptiles have peripheral chemoreceptors located on the carotid arteries and central chemoreceptors sensitive to blood carbon dioxide levels. Lung perfusion may be regulated by cardiac shunting and lung ventilation stimulates lung stretch receptors.

Components to the acid–base related ventilatory drives in the South American lungfish Lepidosiren paradoxa

Lungfish are closely related to terrestrial vertebrates (tetrapoda). Like tetrapods, the South American lungfish (Lepidosiren paradoxa) has central chemoreceptors involved in regulation of acid-base status. However, no data were available on peripheral CO(2)/[H(+)] receptors. Therefore, we tested the hypothesis that such receptors exist by measuring the ventilatory responses during a 5h exposure to combined aquatic/gas phase hypercarbia 7% (approximately 49 mmHg). Normocarbic control ventilation was 22 ml BTPS kg(-1)h(-1), and hypercarbia increased ventilation to 175 ml BTPS kg(-1)h(-1) at 5h. This procedure was repeated with the modification that normocarbic mock CSF (pH 7.45; P(CO2) = 20.7 mmHg) was applied to superfuse the cerebral ventricular system during the last 2h of the experiment. This served to eliminate the hypercarbic stimulus to the central chemoreceptors, while possible responses from peripheral chemoreceptors would remain intact. Peripheral receptors were detected, since ventilation became reduced to 62 ml BTPS kg(-1)h(-1) (P<0.05), which exceeds the initial normocarbic control ventilation (P<0.05). Based on this, the peripheral contribution accounted for 20% of the total response to hypercarbia, similar to the contribution of these receptors in man.

Acid-base balance and CO2 excretion in fish: unanswered questions and emerging models

Carbon dioxide (CO(2)) excretion and acid-base regulation in fish are linked, as in other animals, though the reversible reactions of CO(2) and the acid-base equivalents H(+) and HCO(3)(-): CO(2)+H(2)O<-->H(+)+HCO(3)(-). These relationships offer two potential routes through which acid-base disturbances may be regulated. Respiratory compensation involves manipulation of ventilation so as to retain CO(2) or enhance CO(2) loss, with the concomitant readjustment of the CO(2) reaction equilibrium and the resultant changes in H(+) levels. In metabolic compensation, rates of direct H(+) and HCO(3)(-) exchange with the environment are manipulated to achieve the required regulation of pH; in this case, hydration of CO(2) yields the necessary H(+) and HCO(3)(-) for exchange. Because ventilation in fish is keyed primarily to the demands of extracting O(2) from a medium of low O(2) content, the capacity to utilize respiratory compensation of acid-base disturbances is limited and metabolic compensation across the gill is the primary mechanism for re-establishing pH balance. The contribution of branchial acid-base exchanges to pH compensation is widely recognized, but the molecular mechanisms underlying these exchanges remain unclear. The relatively recent application of molecular approaches to this question is generating data, sometimes conflicting, from which models of branchial acid-base exchange are gradually emerging. The critical importance of the gill in acid-base compensation in fish, however, has made it easy to overlook other potential contributors. Recently, attention has been focused on the role of the kidney and particularly the molecular mechanisms responsible for HCO(3)(-) reabsorption. It is becoming apparent that, at least in freshwater fish, the responses of the kidney are both flexible and essential to complement the role of the gill in metabolic compensation. Finally, while respiratory compensation in fish is usually discounted, the few studies that have thoroughly characterized ventilatory responses during acid-base disturbances in fish suggest that breathing may, in fact, be adjusted in response to pH imbalances. How this is accomplished and the role it plays in re-establishing acid-base balance are questions that remain to be answered.

Phylogeny of vertebrate respiratory rhythm generators: the Oscillator Homology Hypothesis

A revolution is underway in our understanding of respiratory rhythm generation in mammals. Until recently, a major focus of research within the field has centered around the question of locating and elucidating the mechanism of rhythmogenesis of a single respiratory neuronal oscillator which is reiterated bilaterally within the brainstem. Now it appears that each hemisection may contain at least two oscillators that interact to generate the respiratory rhythm in mammals. Comparative studies have hinted at the existence of multiple respiratory oscillators in non-mammalian vertebrates for some time, raising the possibility of homologous oscillators. Here, we consider available tools to identify neuronal oscillators and critically review the evidence for the importance and existence of multiple respiratory oscillators in vertebrates. First focusing on a comparison between frogs and mammals, we then evaluate the hypothesis that ventilatory oscillators in extant tetrapods evolved from ancestral oscillators present in fish (the Oscillator Homology Hypothesis). While supporting data are incomplete, the Oscillator Homology Hypothesis will likely serve as a useful framework to motivate further studies of respiratory rhythm generation in lower vertebrates.

Reflex Cardioventilatory Responses to Hypoxia in the Flathead Gray Mullet (Mugil cephalus) and Their Behavioral Modulation by Perceived Threat of Predation and Water Turbidity

In hypoxia, gray mullet surface to ventilate well-oxygenated water in contact with air, an adaptive response known as aquatic surface respiration (ASR). Reflex control of ASR and its behavioral modulation by perceived threat of aerial predation and turbid water were studied on mullet in a partly sheltered aquarium with free surface access. Injections of sodium cyanide (NaCN) into either the bloodstream (internal) or ventilatory water stream (external) revealed that ASR, hypoxic bradycardia, and branchial hyperventilation were stimulated by chemoreceptors sensitive to both systemic and water O2 levels. Sight of a model avian predator elicited bradycardia and hypoventilation, a fear response that inhibited reflex hyperventilation following external NaCN. The time lag to initiation of ASR following NaCN increased, but response intensity (number of events, time at the surface) was unchanged. Mullet, however, modified their behavior to surface under shelter or near the aquarium edges. Turbid water abolished the fear response and effects of the predator on gill ventilation and timing of ASR following external NaCN, presumably because of reduced visibility. However, in turbidity, mullet consistently performed ASR under shelter or near the aquarium edges. These adaptive modulations of ASR behavior would allow mullet to retain advantages of the chemoreflex when threatened by avian predators or when unable to perceive potential threats in turbidity.

Ancient gill and lung oscillators may generate the respiratory rhythm of frogs and rats

Though the mechanics of breathing differ fundamentally between amniotes and "lower" vertebrates, homologous rhythm generators may drive air breathing in all lunged vertebrates. In both frogs and rats, two coupled oscillators, one active during the inspiratory (I) phase and the other active during the preinspiratory (PreI) phase, have been hypothesized to generate the respiratory rhythm. We used opioids to uncouple these oscillators. In the intact rat, complete arrest of the external rhythm by opioid-induced suppression of the putative I oscillator, that is, pre-Bötzinger complex (PBC) oscillator, did not arrest the putative PreI oscillator. In the unanesthetized frog, the comparable PreI oscillator, that is, the putative buccal/gill oscillator, was refractory to opioids even though the comparable I oscillator, the putative lung oscillator, was arrested. Studies in en bloc brainstem preparations derived from both juvenile frogs and metamorphic tadpoles confirmed these results and suggested that opioids may play a role in the clustering of lung bursts into episodes. As the frog and rat respiratory circuitry produce functionally equivalent motor outputs during lung inflation, these data argue for a close homology between the frog and rat oscillators. We suggest that the respiratory rhythm of all lunged vertebrates is generated by paired coupled oscillators. These may have originated from the gill and lung oscillators of the earliest air breathers.

The Multifunctional Fish Gill: Dominant Site of Gas Exchange, Osmoregulation, Acid-Base Regulation, and Excretion of Nitrogenous Waste

The fish gill is a multipurpose organ that, in addition to providing for aquatic gas exchange, plays dominant roles in osmotic and ionic regulation, acid-base regulation, and excretion of nitrogenous wastes. Thus, despite the fact that all fish groups have functional kidneys, the gill epithelium is the site of many processes that are mediated by renal epithelia in terrestrial vertebrates. Indeed, many of the pathways that mediate these processes in mammalian renal epithelial are expressed in the gill, and many of the extrinsic and intrinsic modulators of these processes are also found in fish endocrine tissues and the gill itself. The basic patterns of gill physiology were outlined over a half century ago, but modern immunological and molecular techniques are bringing new insights into this complicated system. Nevertheless, substantial questions about the evolution of these mechanisms and control remain.

Ontogeny of central CO2 chemoreception: chemosensitivity in the ventral medulla of developing bullfrogs

Sites of central CO2 chemosensitivity were investigated in isolated brain stems from Rana catesbeiana tadpoles and frogs. Respiratory neurograms were made from cranial nerve (CN) 7 and spinal nerve 2. Superfusion of the brain stem with hypercapnic artificial cerebrospinal fluid elicited increased fictive lung ventilation. The effect of focal perfusion of hypercapnic artificial cerebrospinal fluid on discrete areas of the ventral medulla was assessed. Sites of chemosensitivity, which are active continuously throughout development, were identified adjacent to CN 5 and CN 10 on the ventral surface of the medulla. In early- and middle-stage tadpoles and frogs, unilateral stimulation within either site was sufficient to elicit the hypercapnic response, but simultaneous stimulation within both sites was required in late-stage tadpoles. The chemosensitive sites were individually disrupted by unilateral application of 1 mg/ml protease, and the sensitivity to bath application or focal perfusion of hypercapnia was reassessed. Protease lesions at CN 10 abolished the entire hypercapnic response, but lesions at CN 5 affected only the hypercapnic response originating from the CN 5 site. Neurons within the chemosensitive sites were also destroyed by unilateral application of 1 mM kainic acid, and the sensitivity to bath or focal application of hypercapnia was reassessed. Kainic acid lesions within either site abolished the hypercapnic response. Using a vital dye, we determined that kainic acid destroyed neurons by only within 100 microm of the ventral medullary surface. Thus, regardless of developmental stage, neurons necessary for CO2 sensitivity are located in the ventral medulla adjacent to CN 5 and 10.

Effects of afferent input on the breathing pattern continuum in the tambaqui (Colossoma macropomum)

This study used a decerebrate and artificially-ventilated preparation to examine the roles of various afferent inputs in breathing pattern formation in the tambaqui (Colossoma macropomum). Three general breathing patterns were observed: (1) regular breathing; (2) frequency cycling and (3) episodic breathing. Under normoxic, normocapnic conditions, 50% of control fish exhibited regular continuous breathing and 50% exhibited frequency cycling. Denervation of the gills and oro-branchial cavity promoted frequency cycling. Central denervation of the glossopharyngeal and vagus nerves produced episodic breathing. Regardless of the denervation state, hyperoxia produced either frequency cycling or episodic breathing while hypoxia and hypercarbia shifted the pattern to frequency cycling and continuous breathing. We suggest that these breathing patterns represent a continuum from continuous to episodic breathing with waxing and waning occupying an intermediate stage. The data further suggest that breathing pattern is influenced by both specific afferent input from chemoreceptors and generalised afferent input while chemoreceptors specific for producing changes in breathing pattern may exist in fish.

Central CO2 chemoreception in developing bullfrogs: anomalous response to acetazolamide

Central CO(2) chemoreception and the role of carbonic anhydrase were assessed in brain stems from Rana catesbeiana tadpoles and frogs. Buccal and lung rhythms were recorded from cranial nerve VII and spinal nerve II during normocapnia and hypercapnia before and after treatment with 25 microM acetazolamide. The lung response to acetazolamide mimicked the hypercapnic response in early-stage and midstage metamorphic tadpoles and frogs. In late-stage tadpoles, acetazolamide actually inhibited hypercapnic responses. Acetazolamide and hypercapnia decreased the buccal frequency but had no effect on the buccal duty cycle. Carbonic anhydrase activity was present in the brain stem in every developmental stage. Thus more frequent lung ventilation and concomitantly less frequent buccal ventilation comprised the hypercapnic response, but the response to acetazolamide was not consistent during metamorphosis. Therefore, acetazolamide is not a useful tool for central CO(2) chemoreceptor studies in this species. The reversal of the effect of acetazolamide in late-stage metamorphosis may reflect reorganization of central chemosensory processes during the final transition from aquatic to aerial respiration.

Central respiratory activity of the tadpole in vitro brain stem is modulated diversely by nitric oxide

Nitric oxide (NO) is a potent central neuromodulator of respiration, yet its scope and site of action are unclear. We used 7-nitroindazole (7-NI), a selective inhibitor of endogenous neuronal NO synthesis, to investigate the neurogenesis of respiration in larval bullfrog (Rana catesbeiana) isolated brain stems. 7-NI treatment (0.0625-0.75 mM) increased the specific frequency of buccal ventilation (BV) events, indicating influence on BV central rhythm generators (CRGs). The drug reduced occurrence, altered burst shape, and disrupted clustering of lung ventilation (LV) events, without altering their specific frequency. LV burst occurrence and clustering also differed between pH conditions. We conclude that NO has diverse effects on respiratory rhythmogenesis, being necessary for the expression of respiratory rhythms, inhibiting the frequency of BV CRG, and affecting both shape and clustering of LV bursts through conditional modulation of LV CRG. We confirm central chemosensitivity in these preparations and demonstrate chemomodulation of LV burst clustering and occurrence but not specific frequency. Results support distinct oscillators underlying LV and BV CRGs.

Branchial innervation

Inspection of the dorsal end of fish gills reveals an impressive set of nerve trunks, connecting the gills to the brain. These trunks are branches of cranial nerves VII (the facial) and especially IX (the glossopharyngeal) and X (the vagus). The nerve trunks carry a variety of nervous pathways to and from the gills. A substantial fraction of the nerves running in the branchial trunks carry afferent (sensory) information from receptors within the gills. There are also efferent (motor) pathways, which control muscles within the gills, blood flow patterns and possibly secretory functions. Undertaking a more careful survey of the gills, it becomes evident that the arrangement of the microanatomy (particularly the blood vessels) and its innervation are strikingly complex. The complexity not only reflects the many functions of the gills but also illustrates that the control of blood flow patterns in the gills is of crucial importance in modifying the efficiency of its chief functions: gas transfer and salt balance. The "respiratory-osmoregulatory compromise" is maintained by minimizing the blood/water exchange (functional surface area of the gills) to a level where excessive water loss (marine teleosts) or gain (freshwater teleosts) is kept low while ensuring sufficient gas exchange. This review describes the arrangement and mechanisms of known nervous pathways, both afferent and efferent, of fish (notably teleosts) gills. Emphasis is placed primarily on the autonomic nervous system and mechanisms of blood flow control, together with an outline of the afferent (sensory) pathways of the gill arches.

Phylogeny of CO2/H+ chemoreception in vertebrates

The traditional view has been that respiratory chemoreceptors responsive to changes in P(CO(2))/pH first evolved in air breathing vertebrates at both peripheral and central sites. Recent evidence, however, suggests that fish also possess chemoreceptors responsive to changes in P(CO(2)) per se. In many species these receptors reside in the gills and respond primarily to changes in aquatic rather than arterial P(CO(2)). There is also scattered evidence to suggest that central CO(2)/H(+)-sensitive chemoreceptors may be present in representatives of all fish groups but only the data for air breathing fish are strong and convincing. The phylogenetic trends that are emerging indicate that the use of CO(2) chemoreception for cardiorespiratory processes arose much earlier than previously believed, (arguably) that CO(2) chemoreception may first have arisen in the periphery sensitive to the external environment and that central CO(2)/H(+) chemoreception subsequently arose multiple times in association with several of the independent origins of air breathing, and that the mechanisms of CO(2)/H(+) chemotransduction may be as varied as the different receptor groups involved.

Evolution of central respiratory chemoreception: a new twist on an old story

Evolution of central respiratory chemosensitivity has been linked traditionally to the need for carbon dioxide regulation that accompanied the evolution of air breathing in terresterial animals. We examined the validity of this linkage by investigating the possibility of central chemoreception in air breathing fish that diverged from the amphibian lineage long before the appearance of terrestriality. We showed that the isolated brainstem preparation of the long nose gar (Lepisosteus osseus) produces a putative motor pattern for lung ventilation, which is responsive to CO(2). These findings, together with more inferential evidence, suggest an association between air breathing and central chemosensitivity in aquatic animals that spans the major branches in vertebrate phylogeny. Furthermore, developmental observations in tadpoles suggest that the neural substrates for central chemoreception exist in proximity to that for rhythm generation. We postulate that a primitive ancestral CPG, sensitive to CO(2) is conserved and is evidenced in the intrinsic coupling of respiratory CPG and central chemoreception in modern tetrapods.

The relative roles of external and internal CO2versusH+ in eliciting the cardiorespiratory responses ofSalmo salarandSqualus acanthiasto hypercarbia

Fish breathing hypercarbic water encounter externally elevated PCO2 and proton levels ([H+]) and experience an associated internal respiratory acidosis, an elevation of blood PCO2 and [H+]. The objective of the present study was to assess the potential relative contributions of CO2versus H+ in promoting the cardiorespiratory responses of dogfish (Squalus acanthias) and Atlantic salmon (Salmo salar) to hypercarbia and to evaluate the relative contributions of externally versus internally oriented receptors in dogfish.

In dogfish, the preferential stimulation of externally oriented branchial chemoreceptors using bolus injections (50 ml kg–1) of CO2-enriched (4 % CO2) sea water into the buccal cavity caused marked cardiorespiratory responses including bradycardia (–4.1±0.9 min–1), a reduction in cardiac output (–3.2±0.6 ml min–1 kg–1), an increase in systemic vascular resistance (+0.3±0.2 mmHg ml min–1 kg–1), arterial hypotension (–1.6±0.2 mmHg) and an increase in breathing amplitude (+0.3±0.09 mmHg) (means ± s.e.m., N=9–11). Similar injections of CO2-free sea water acidified to the corresponding pH of the hypercarbic water (pH 6.3) did not significantly affect any of the measured cardiorespiratory variables (when compared with control injections). To preferentially stimulate putative internal CO2/H+ chemoreceptors, hypercarbic saline (4 % CO2) was injected (2 ml kg–1) into the caudal vein. Apart from an increase in arterial blood pressure caused by volume loading, internally injected CO2 was without effect on any measured variable.

In salmon, injection of hypercarbic water into the buccal cavity caused a bradycardia (–13.9±3.8 min–1), a decrease in cardiac output (–5.3±1.2 ml min–1 kg–1), an increase in systemic resistance (0.33±0.08 mmHg ml min–1 kg–1) and increases in breathing frequency (9.7±2.2 min–1) and amplitude (1.2±0.2 mmHg) (means ± s.e.m., N=8–12). Apart from a small increase in breathing amplitude (0.4±0.1 mmHg), these cardiorespiratory responses were not observed after injection of acidified water.

These results demonstrate that, in dogfish and salmon, the external chemoreceptors linked to the initiation of cardiorespiratory responses during hypercarbia are predominantly stimulated by the increase in water PCO2 rather than by the accompanying decrease in water pH. Furthermore, in dogfish, the cardiorespiratory responses to hypercarbia are probably exclusively derived from the stimulation of external CO2 chemoreceptors, with no apparent contribution from internally oriented receptors.

Relationship between cerebro-spinal fluid pH and pulmonary ventilation of the South American lungfish, Lepidosiren paradoxa (Fitz.)

The respiratory control in land vertebrates (Tetrapoda) is mainly linked to regulation of acid-base status, which involves peripheral and central chemoreceptors. The lungfish (Dipnoi) might constitute the sister group of all land vertebrates (Tetrapoda) and possess a combination of real lungs and reduced gills. In this context, we evaluated the possible presence of central respiratory chemoreceptors in the South American Lungfish, Lepidosiren paradoxa. Pulmonary ventilation and respiratory frequency increased significantly with reductions of CSF pH by means of mock CSF solutions. This suggests that Lepidosiren possess central acid-base receptors.

The CO2/pH ventilatory drive in fish

That ventilation in fish is driven by O2 has long been accepted. The O2 ventilatory drive reflects the much lower capacitance of water for O2 than for CO2, and is mediated by O2 receptors that are distributed throughout the gill arches and that monitor both internal and external O2 levels. In recent years, however, evidence has amassed in support of the existence of a ventilatory drive in fish that is keyed to CO2 and/or pH. While ventilatory responses to CO2/pH may be mediated in part by the O2 drive through CO2/pH-induced changes in blood O2 status, CO2/pH also appear to stimulate ventilation directly. The receptors involved in this pathway are as yet unknown, but the experimental evidence available to date supports the involvement of branchial CO2-sensitive chemoreceptors with an external orientation. Internally-oriented CO2-sensitive chemoreceptors may also be involved, although evidence on this point remains equivocal. In the present paper, the evidence for a CO2/pH-keyed ventilatory drive in fish will be reviewed.

Which came first, the lung or the breath?

Lungs are the characteristic air-filled organs (AO) of the Polypteriformes, lungfish and tetrapods, whereas the swimbladder is ancestral in all other bony fish. Lungs are paired ventral derivatives of the pharynx posterior to the gills. Their respiratory blood supply is the sixth branchial artery and the venous outflow enters the heart separately from systemic and portal blood at the sinus venosus (Polypteriformes) or the atrium (lungfish), or is delivered to a separate left atrium (tetrapods). The swimbladder, on the other hand, is unpaired, and arises dorsally from the posterior pharynx. It is employed in breathing in Ginglymodi (gars), Halecomorphi (bowfin) and in basal teleosts. In most cases, its respiratory blood supply is homologous to that of the lung, but the vein drains to the cardinal veins. Separate intercardiac channels for oxygenated and deoxygenated blood are lacking. The question of the homology of lungs and swimbladders and of breathing mechanisms remains open. On the whole, air ventilatory mechanisms in the actinopterygian lineage are similar among different groups, including Polypteriformes, but are distinct from those of lungfish and tetrapods. However, there is extreme variation within this apparent dichotomy. Furthermore, the possible separate origin of air breathing in actinopterygian and 'sarcopterygian' lines is in conflict with the postulated much more ancient origin of vertebrate air-breathing organs. New studies on the isolated brainstem preparation of the gar (Lepisosteus osseus) show a pattern of efferent activity associated with a glottal opening that is remarkably similar to that seen in the in-vitro brainstem preparation of frogs and tadpoles. Given the complete lack of evidence for AO in chondrichthyans, and the isolated position of placoderms for which buoyancy organs of uncertain homology have been demonstrated, it is likely that homologous pharyngeal AO arose in the ancestors of early bony fish, and was pre-dated by behavioral mechanisms for surface (water) breathing. The primitive AO may have been the posterior gill pouches or even the modified gills themselves, served by the sixth branchial artery. Further development of the dorsal part may have led to the respiratory swimbladder, whereas the paired ventral parts evolved into lungs.