resolutionmini

mise à jour du
1 avril 2004
Physiological Reviews
1995;75(1):1-31
lexique
Central control of breathing in mammals: neuronal circuitry, membrane properties, and neurotransmitters
A Bianchi, M Denavit-saubié, J Champagnat
Laboratoire de Neurobiologie et Neurophysiologie Fonctionnelles, Laboratoire de Biologie Fonctionnelle du Neurone, Institut Alfred Fessard, CNRS, France  
La parakinésie brachiale oscitante Walusinski, Quoirin, Neau  

Chat-logomini

I. INTRODUCTION In mammals, the uptake of oxygen and release of carbon dioxide in the lungs resault from rhytmic contractions of the diaphragm, intercostal, and abdominal muscles. The pressures generated by these "pump" muscles, combined with the flow resistance of the airways determined by the state of contraction of both striated and smooth muscles in them, cause changes in lung volume.
 
Motoneurons driving the respiratory pump muscles are located at various levels of the spinal cord: 1) phrenic motoneurons innervating the diaphragin at C3-C6; 2) motoneurons innervating the intercostal muscles at T1-T12 motoneurons innervating the abdominal muscles at T4-L3.
 
The motoneurons innervating muscles controlling airway dimensions are located in the brain stem. Those for the alae nasi are in the facial nucleus; those for pharyngeal muscles, via the glossopharyngeal and pharyngeal branch of the vagus nerves, are in the nucleus ambiguus; and those for bronchial smooth muscle, via the vagi and the larynx, via the recurrent laryngeal nerves, are also in the nucleus ambiguus. Other cranial motoneurons innervate muscles exhibiting rhythmic respiratory activity. These include 1) hypoglossal motoneurons for genioglossus muscle that governs tongue protrusion in inspiration; 2) naso-labial motoneurons for the muscles surrounding the nasal passages, via the facial nerve and 3) tensor veli palatini motoneurons, via the mandibular branch of the fith nerve, maintain nasopharyngeal isthmus patency in inspiration.
 
These motoneurons are also involved in such nonrespiratory functions as postural control, phonation, protective reflexes of the upper airway (coughing, sneezing, and swallowing), and such expulsive maneuvers as vomiting, defecation, parturition, and micturition.
 
Breathing in mammals relies on a neuronal network located within the bram stem. Central control unplies that the central nervous system (CNS) is intrinsically capable of providing the proper timing of muscle activation, although sensory inputs can modulate respiratory rhythm and pattern and adapt breathing to changes in state. This neuronal network develops a rhythm essentially based on two stable phases, inspiration and expiration. However, a new concept has proposed that the respiratory rhythm generates three phases instead of two, expiration being divided in two phases, ie., stage 1 of expiration (or postinspiration) and stage 2 of expiration.
 
Altough the basic elements making up this network are located within the brain stem, structures outside the brain stem can and do affect respiration. However, these components are not essential to the network because their elimination does not impair rhythmogenesis. These elements external to the network are reminiscent of various structures previously called "centers." The earlier belief that specific respiratory functions resided within circumscribed structures (e.g., the pneumotaxic center) has been modified, by using the term central pattern generator (CPG) or central rhythm generator.
 
As part of the brain stem, neurons of the respiratory CPG share common morphological and electrophysiological characteristics with other brain stem neurons. For example, the axons of respiratory and nonrespiratory cranial motoneurons project dorsomedially then laterally, whereas those of bulbospinal interneurons first project laterally or medially then rostrocaudally. However, the respiratory CPG elaborates a periodic pattern of discharge that remains spontaneously active throughout life. Other rhythmic outputs such as swallowing, coughing, vomiting, or locomotion are also generated by brainstem neuronal networks, the neurons of which exhibit similar augrnenting (ramp) or decrementing discharge patterns and abrupt phase transitions. However, these rhythmic nonrespiratory but vital functions need to be triggered by specific afferent inputs; moreover, they are discontinuous (or irregular), whereas breathing is continuous.
 
A special feature of the respiratory CPG is that it functions automatically but can be controlled voluntarily.
 
In this way, respiration, as a sensory-motor act, can be modulated much like posture and locomotion. The CNS processes afferent inputs to provide an appropriate motor output. In the respiratory system, chemosensitive, pulmonary, and even proprioceptive afférents determine the appropriate respiratory outputs to maintain homeostasis. Voluntary control involves the forebrain, with information being distributed to premotoneurons and motoneurons through pyramidal and extrapyramidal pathways. In contrast, automatic control involves the brain stem, the output of which is distributed to interneurons and motoneurons through proprio-brain stem pathways to cranial motoneurons, and through pathways in the ventrolateral spinal cord to interneurons and motoneurons.
 
In this review, we describe the factors contributing to respiratory rhythmogenesis, ie., the organization of the respiratory CPG in terms of its neuronal elements, synaptic connections, and possible interactions with the different intrinsic membrane properties of different types of neurons, interconnections, and neurotransmitters. It would be an oversimplification to believe that respiratory activity is generated by a single pacemaker or a specific neurotransmitter. Moreover, synaptic interactions alone cannot adequately explain the rhythmic character of respiration. Additional mechanisms, based on intrinsic membrane properties, are likely responsible for the transitions between respiratory phases (also referred to as on- and off-switch functions). Several groups of neurons with différent intrinsic properties, specific interconnections, and neurotransmitters are probably involved.
 
Specific neurotransmitters are now known to be present in fast synaptic transmission and in neuromodulation of the respiratory CPG. Excitatory amino acids play an essential role in the reexcitatory mechanisms, in bulbospinal transmission of inspiratory drive, and in phase termination involving cooperative action of various receptor types. Glycine and gamma-aminobutyric acid (GABA) acting via chloride channels mediate inhibitory postsynaptic effects. Voltage-dependent currents (e.g., low-threshold calcium current) or calcium-dependent potassium current are presumably activated and appear as essential in the determination of respiratory patterns. Unfortunately, most of these ionic conductances have been demonstrated in in vitro preparations, and their existence in vivo remains to be unambiguously demonstrated. Although it is difficult to detennine whether intrinsie membrane properties obtained in vitro by membrane voltage manipulation are compatible with the membrane voltages recorded in vivo, we propose a network model generating oscillations of membrane potentials of brain stem neurons as part of respiratory rhythmogenesis, and based on dynamic cooperation between synaptically induced membrane potential changes and intrinsic membrane properties of the neurons.

V. CONCLUSION

According to our model, respiratory rhythm is not generated by a single conditional pacemaker process. Instead, respiration is a sequential motor behavior, extending over a time scale very long compared with such fast electrophysiological events as action potentials but very slow with respect to long-term neuromodulatory processes. Respiratory neurons included in the model are subjected continuously to excitatory or inhibitory synaptic inputs which modulate their intrinsic membrane properties throughout the respiratory cycle. As already implicit in the three-phase theory, brain stem respiratory activity results from the sequential activation of at least six neuronal populations, leading to two fast phaseswitching transitions and three relatively long respiratory phases. Each process is conditioned by the previous one and initiates the next. A major problem requiring our future attention is the determination of how, in contrast to intermittent motor activity, this sequence of respiratory rhythm-generating processes is maintained throughout life.