Harold J. Bell

Comparison of the metabolic and ventilatory response to hypoxia and H2S in unsedated mice and rats.

Hypoxia alters the control of breathing and metabolism by increasing ventilation through the arterial chemoreflex, an effect which, in small-sized animals, is offset by a centrally mediated reduction in metabolism and respiration. We tested the hypothesis that hydrogen sulfide (H(2)S) is involved in transducing these effects in mammals. The rationale for this hypothesis is twofold. Firstly, inhalation of a 20-80 ppm H(2)S reduces metabolism in small mammals and this effect is analogous to that of hypoxia. Secondly, endogenous H(2)S appears to mediate some of the cardio-vascular effects of hypoxia in non-mammalian species. We, therefore, compared the ventilatory and metabolic effects of exposure to 60 ppm H(2)S and to 10% O(2) in small and large rodents (20g mice and 700g rats) wherein the metabolic response to hypoxia has been shown to differ according to body mass. H(2)S and hypoxia produced profound depression in metabolic rate in the mice, but not in the large rats. The depression was much faster with H(2)S than with hypoxia. The relative hyperventilation produced by hypoxia in the mice was replaced by a depression with H(2)S, which paralleled the drop in metabolic rate. In the larger rats, ventilation was stimulated in hypoxia, with no change in metabolism, while H(2)S affected neither breathing nor metabolism. When mice were simultaneously exposed to H(2)S and hypoxia, the stimulatory effects of hypoxia on breathing were abolished, and a much larger respiratory and metabolic depression was observed than with H(2)S alone. H(2)S had, therefore, no stimulatory effect on the arterial chemoreflex. The ventilatory depression during hypoxia and H(2)S in small mammals appears to be dependent upon the ability to decrease metabolism.

Hydrogen sulfide oxidation and the arterial chemoreflex: Effect of methemoglobin

Endogenous H(2)S has been proposed to transduce the effects of hypoxia in the carotid bodies (CB). To test this hypothesis, we created a sink for endogenously produced H(2)S by inducing ∼10% methemoglobinemia via the injection of 250 mg of sodium nitrite in spontaneously breathing anaesthetized sheep. Methemoglobinemia has been shown to catalyze the oxidation of large quantities of sulfide in the blood and tissues. We found that the presence of metHb completely abolished the ventilatory stimulation induced by 10 mg NaHS (i.v.), which in control conditions mimicked the effects of breathing 6-7 tidal volumes of nitrogen, confirming the dramatic increase in the oxidative power of the blood for sulfide. The ventilatory responses to hypoxia (10% O(2)), nitrogen and hyperoxia were in no way depressed by the metHb. Our results demonstrate that the ventilatory chemoreflex is not depressed in the presence of a high oxidative capacity for sulfide and challenge the view that H(2)S transduces the effects of hypoxia in the CB.

Hypocapnia increases the prevalence of hypoxia-induced augmented breaths

Augmented breaths promote respiratory instability and have been implicated in triggering periods of sleep-disordered breathing. Since respiratory instability is well known to be exacerbated by hypocapnia, we asked whether one of the destabilizing effects of hypocapnia might be related to an increased prevalence of augmented breaths. With this question in mind, we first sought to determine whether hypoxia-induced augmented breaths are more prevalent when hypocapnia is also present. To do this, we studied the breath-by-breath ventilatory responses of a group of freely behaving adult rats in a variety of different respiratory background conditions. We found that the prevalence of augmented breaths was dramatically increased during hypocapnic-hypoxia compared with room air conditions. When hypocapnia was prevented during exposure to hypoxia by adding 5% CO2 to the inspired air, the rate of occurrence of augmented breaths was no greater than that observed in room air. The addition of CO2 alone to room air had no effect on the prevalence of augmented breaths. We conclude that in spontaneously breathing rats, hypoxia promotes the generation of augmented breaths, but only in poikilocapnic conditions, where hypocapnia develops. Our results, therefore, reveal a means by which CO2 exerts a stabilizing influence on breathing, which may be of particular relevance during sleep in conditions commonly associated with respiratory instability.

Control of breathing and volitional respiratory rhythm in humans

When breathing frequency (f) is imperceptibly increased during a volitionally paced respiratory rhythm imposed by an auditory signal, tidal volume (Vt) decreases such that minute ventilation (Ve) rises according to f-induced dead-space ventilation changes (18). As a result, significant change in alveolar ventilation and Pco(2) are prevented as f varies. The present study was performed to determine what regulatory properties are retained by the respiratory control system, wherein the spontaneous automatic rhythmic activity is replaced by a volitionally paced rhythm. Six volunteers were asked to trigger each breath cycle on hearing a brief auditory signal. The time interval between subsequent auditory signals was imperceptibly changed for 10-15 min, during 1) air breathing (condition 1), 2) the addition of a 300 ml of instrumental dead space (condition 2), 3) an increase in the inspired level of CO(2) (condition 3), and 4) light exercise (condition 4). We found that as f was slowly increased the elaborated Vt decreased in accordance to the background level of CO(2) and metabolic rate. Indeed, for any given breath duration, Vt was shifted upward in condition 2 vs. 1, whereas the slope of Vt changes according to the volitionally rhythm was much steeper in conditions 3 and 4 vs. 1. The resulting changes in Ve offset any f-induced changes in dead-space ventilation in all conditions. We conclude that there is an inherent, fundamental mechanism that elaborates Vt based on f when imposed by the premotor cortex in humans. The chemoreflex and exercise drive to breath interacts with this cortically mediated rhythm maintaining alveolar rather than Ve constant as f changes. The implications of our findings are discussed in the context of our current understanding of the central generation of breathing rhythm.

Hypoxia-induced arterial chemoreceptor stimulation and hydrogen sulfide: too much or too little?

This brief review presents and discusses some of the important issues surrounding the theory which asserts that endogenous hydrogen sulfide (H(2)S) is the mediator of, or at least an important contributor to, hypoxia-induced arterial chemorereceptor stimulation. The view presented here is that before H(2)S can seriously be considered as a candidate for transducing the O(2)-signal in the carotid bodies (CB), fundamental contradictions need to be resolved. One of these major contradictions is certainly the discrepancy between the levels of H(2)S endogenously present in the CB during hypoxia compared to the levels required to stimulate the arterial chemoreceptors in vitro. Very small amounts of H(2)S are thought to be produced endogenously during a given level of hypoxia, yet the partial pressure of tissue H(2)S which is needed to produce an effect commensurate with that of hypoxia is thousands to millions of times higher. This review discusses this and other contradictions in light of what is known about H(2)S concentration and production in various tissues, the lessons we have learnt from the response to exogenous sulfide and the ability of the blood and the mitochondria to oxidize very large amounts of sulfide. These considerations suggest that the increased production of H(2)S in hypoxia and exogenous sulfide cannot produce the same effect on the carotid bodies and breathing. While the effects of the endogenous H(2)S on breathing remains to be established, the effects exogenous sulfide can be accounted for by its long established toxicity on cytochrome C oxidase.

Peripheral oxygen-sensing cells directly modulate the output of an identified respiratory central pattern generating neuron

Breathing is an essential homeostatic behavior regulated by central neuronal networks, often called central pattern generators (CPGs). Despite ongoing advances in our understanding of the neural control of breathing, the basic mechanisms by which peripheral input modulates the activities of the central respiratory CPG remain elusive. This lack of fundamental knowledge vis‐à‐vis the role of peripheral influences in the control of the respiratory CPG is due in large part to the complexity of mammalian respiratory control centres. We have therefore developed a simpler invertebrate model to study the basic cellular and synaptic mechanisms by which a peripheral chemosensory input affects the central respiratory CPG. Here we report on the identification and characterization of peripheral chemoreceptor cells (PCRCs) that relay hypoxia‐sensitive chemosensory information to the known respiratory CPG neuron right pedal dorsal 1 in the mollusk Lymnaea stagnalis. Selective perfusion of these PCRCs with hypoxic saline triggered bursting activity in these neurons and when isolated in cell culture these cells also demonstrated hypoxic sensitivity that resulted in membrane depolarization and spiking activity. When cocultured with right pedal dorsal 1, the PCRCs developed synapses that exhibited a form of short‐term synaptic plasticity in response to hypoxia. Finally, osphradial denervation in intact animals significantly perturbed respiratory activity compared with their sham counterparts. This study provides evidence for direct synaptic connectivity between a peripheral regulatory element and a central respiratory CPG neuron, revealing a potential locus for hypoxia‐induced synaptic plasticity underlying breathing behavior.

The "other" respiratory effect of opioids: suppression of spontaneous augmented ("sigh") breaths.

The purpose of this study was to examine the effects of a clinically relevant opioid on the production of augmented breaths (ABs) in unanesthetized animals breathing normal room air, using a dosage which does not depress breathing. To do this we monitored breathing noninvasively, in unrestrained animals before and after subcutaneous injection of either morphine, or a saline control. The effect of ketamine/xylazine was also studied to determine the potential effect of an alternative sedative agent. Last, the effect of naloxone was studied to determine the potential influence of endogenous opioids in regulating the normal incidence of ABs. Morphine (5 mg/kg) had no depressive effect on breathing, but completely eliminated ABs in all animals in room air (P = 0.027). However, when animals breathed hypoxic air (10% O(2)), animals did express ABs, although their incidence was still reduced by morphine (P < 0.001). This was not a result of sedation per se, as ABs continued at their normal rate in room air during sedation with ketamine. Naloxone had no effect on breathing or AB production, and so endogenous opioids are not likely involved in regulating their rate of production under normal conditions. Our results show that in unanesthetized animals breathing normal room air, a clinically relevant opioid eliminates ABs, even at a dose that does not cause respiratory depression. Despite this, hypoxia-induced stimulation of breathing can facilitate the production of ABs even with the systemic opioid present, indicating that peripheral chemoreceptor stimulation provides a potential means of overcoming the opioid-induced suppression of these respiratory events.

Respiratory effects of changing the volume load imposed on the peripheral venous system

This study was designed to determine if acute distension of the hindlimb venous circulation stimulates breathing, thereby contributing to the respiratory responses to rapid changes in total blood volume. In 10 spontaneously breathing anesthetized sheep, we withdrew 15 ml kg(-1) of blood from a femoral vein over approximately 1-2 min. We then compared the respiratory effects of infusing this venous blood back into the femoral veins across two conditions: the inferior vena cava (IVC) was either unobstructed or obstructed by a balloon-tipped catheter. We found that when blood was withdrawn and blood volume decreased, an absolute increase in breathing often occurred, but more importantly that a relative hyperventilation was always observed. When this blood was re-infused into the animal, effectively increasing blood volume, the respiratory response depended upon whether or not the IVC was open or obstructed. With the IVC unobstructed, a relative hypoventilation occurred, accompanied by an increase in alveolar PCO(2). In contrast, when the venous blood was re-infused and the IVC was obstructed, ventilation increased significantly, and the response was often hypocapnic. These results indicate that increasing the volume load in the venous circulation increases breathing, and that the transduction mechanism is contained within the peripheral venous system. Further, the respiratory drive from this sensory mechanism is subject to modulation via changes in the circulatory status, most likely within the arterial side.

A Peripheral Oxygen Sensor Provides Direct Activation of an Identified Respiratory CPG Neuron in Lymnaea

The mechanisms by which peripheral, hypoxia-sensitive chemosensory cells modulate the output from the respiratory central pattern generator (CPG) remain largely unknown. In order to study this topic at a fundamental level, we have developed a simple invertebrate model system, Lymnaea stagnalis wherein we have identified peripheral chemoreceptor cells (PCRCs) that relay hypoxia-sensitive chemosensory information to a known respiratory CPG neuron, right pedal dorsal 1 (RPeD1). Significance of this chemosensory drive was confirmed via denervation of the peripheral sensory organ containing the PCRCs, and subsequent behavioral observation. This study provides evidence for direct synaptic connectivity between oxygen sensing PCRCs and a CPG neuron, and describes a unique model system appropriate for studying mechanisms of hypoxia-induced, respiratory plasticity from the level of an identified synapse to whole animal behavior.

Control of breathing during acute change in cardiac preload in a patient with partial cardiopulmonary bypass

We recently had the opportunity to investigate the ventilatory effects of changing the rate of venous return to the heart (and thus pulmonary gas exchange) in a patient equipped with a venous-arterial oxygenated shunt (extracorporeal membrane oxygenation (ECMO) support). The presence of the ECMO support provided a condition wherein venous return to the right heart could be increased or decreased while maintaining total aortic blood flow and arterial blood pressure (ABP) constant. The patient, who had received a heart transplant 12 years ago, was admitted for acute cardiac failure related to graft rejection. The clinical symptomatology was that of right heart failure. We studied the patient on the 4th day of ECMO support, while she was breathing spontaneously. The blood flow diverted through the ECMO system represented 2/3 of the total aortic flow (4 l min(-1)). With these ECMO settings, the baseline level of ventilation was low (3.89+/-0.99 l min(-1)), but PET(CO2) was not elevated (37+/-2 mmHg). When Pa(CO2) in the blood coming from the ECMO was increased, no stimulatory effect on ventilation was observed. However, when the diversion of the venous return to the ECMO was stopped then restored, minute ventilation respectively increased then decreased by more than twofold with opposite changes in PET(CO2). These maneuvers were associated with large changes in the size of the right atrium and ventricle and of the left atrium. This observation suggests that the change in venous return affects breathing by encoding some of the consequences of the changes in cardiac preload. The possible sites of mediation are discussed.