Jean-Pierre Gossard, Réjean Dubuc and Arlette Kolta (Eds.) Progress in Brain Research, Vol. 188 ISSN: 0079-6123 Copyright Ó 2011 Elsevier B.V. All rights reserved.
Looking forward to breathing Jack L. Feldman* Department of Neurobiology, David Geffen School of Medicine, University of California Los Angeles, Los Angeles, California, USA
Abstract: I provide a personal view of the developments since 1986 that underlie the contemporary view(s) about how the rhythm of breathing is generated and how the pattern of breathing is modulated. Two sites in the mammalian brainstem are likely to participate in respiratory rhythm generation: the preBötzinger Complex (preBötC), first described and intensely investigated since 1990, plays a well-documented essential role in normal breathing in mammals of all ages and may be principally involved in controlling inspiratory motor activity, and the retrotrapezoid/parafacial respiratory group (RTN/pFRG) that appears to play at least a modulatory role in neonatal and juvenile rodents and may be a conditional oscillator that controls active expiration. Keywords: preBötzinger Complex; retrotrapezoid nucleus; parafacial respiratory group; respiration; brainstem; medulla; hindbrain.
Breathing is a rhythmic, continuous behavior controlled by the brain and necessary for life in all mammals. The neural circuit underlying respiration has at its core a rhythm generator in the brainstem whose modulation affects the motor output to change pattern and timing of muscle contraction, driving the respiratory pump, for example, diaphragm, abdominals, and modulating airway resistance, for example, tongue and upper airways. Few other vital mammalian functions involve such direct control by a clearly defined relatively compact neural circuit. Breathing as a
centrally driven behavior in mammals is unique in its experimental accessibility from slices to awake/sleeping mammals. The behavior is rhythmic, continuous, and endogenous and the inputs and outputs at every level of analysis are straightforward and measurable. The importance of breathing to life is obvious. We must breathe continuously and reliably from birth, during wakefulness and sleep. Breathing is a behavior that is precisely modulated by metabolic demand that ranges over an order of magnitude, and integrated well with other behaviors such as speech, chewing, swallowing, and locomotion (this book). Disorders of breathing in humans are legion, and their consequences are significant.
*Corresponding author. Tel.: þ1-310-825-0954 DOI: 10.1016/B978-0-444-53825-3.00019-X
Understanding breathing is an essential foundation for efforts to explain more complex behaviors requiring movement of air such as vocalization, speech, and emotion (sighing, laughing, and crying are characterized by rather stereotyped breathing patterns) and provides a basis for development of rational therapeutic approaches to the host of diseases in which breathing is markedly disordered. The historical background for studies of the neural control of breathing can be traced through oldest recorded history and is thoroughly embedded in our culture. Just think of how often the verbs “inspire” and “expire” are used to express important actions. In the West, Aristotle and Galen showed great interest in breathing; in the East, the control of breathing plays a central role in several religions. The early history of these studies is briefly summarized in a previous publication (Feldman, 1986), which also covers the field until 1986. My personal motivation for studying breathing arose from my time as a graduate student in physics. I was trained in a physics culture where the most fundamental problems were solved from first principles. My interests in space and time, that is, relativity, expanded to include its perception, and I became very interested in understanding how the brain did this. My Ph.D. advisor, more aware than I was of the daunting complexity of biological problems, insisted that I start out in my studies of brain function on a much more straightforward and potentially solvable problem. So, I choose what I thought was a simple, even trivial (so goes the physicist's hubris) problem, that is, how does the mammalian nervous system generate respiratory rhythm, with the goal of solving this problem from first principles. When I started working on respiratory rhythmogenesis, the experimental model was the anesthetized cat, and the paradigm for rhythm generation was that it was generated by neurons in the dorsal medulla. I did not readily accept the premises that underlay this view, and my early career as an experimentalist was characterized by
extensive criticism of this idea. Moreover, after several years, I saw the cat being supplanted as a model as more and more interesting work, at first immunohistochemistry and later genetics and molecular biology, was being done in rodents, so we transferred our experimental model to rats. This was also a time of an explosion of in vitro mammalian (rodent) electrophysiology, for example, hippocampal slices, that greatly advanced our understanding of cellular processes, and I dreamed of extending in vitro approaches to studies of behavior, that is, breathing. With this in mind, and needing to find a kindred spirit, I did a sabbatical in Sten Grillner's laboratory (1983–1984) to learn the lamprey preparation as a foundation for developing an in vitro model for mammalian respiration. When Suzue published his important 1984 paper (Suzue, 1984) describing the isolated brainstem spinal cord preparation from neonatal rat that generated breathing rhythm, the experience in Stockholm allowed us to set up the preparation instantly at UCLA, and we started a series of experiments that revolutionized the field (Smith et al., 1991). With Jeff Smith and colleagues, we isolated a small region of the medulla that was essential for breathing rhythm and christened it the “preBötzinger Complex” (preBötC). We then made thin transverse medullary slices of this level and discovered that we could get rhythmic motor nerve output (indicating this slice was generating a bona fide respiratory-related behavior). Since then in a ongoing effort to test our ideas originating from this paper, we performed many experiments using a broad range of techniques that showed that: (i) lesions selectively targeting preBötC neurons that express the neurokinin 1 receptor (NK1R) in intact (unanesthetized and forebrain intact) adult rats induce a disturbed breathing pattern during sleep; with more extensive lesions, ataxic breathing occurs during wakefulness with apneas (complete cessation of breathing) during sleep (Gray et al., 2001); (ii) rapid silencing of preBötC neurons expressing somatostatin (Sst) produces profound apnea in
anesthetized or awake adult rats (Tan et al., 2008) (Fig. 1); (iii) juvenile rats with brainstem transection just rostral to the preBötC continue to generate rhythmic inspiratory-dominated breathing patterns. In quite a turnaround from the initial criticism of the preBötC and the utility of using in vitro preparations for understanding the neural control of breathing, the critical role of the preBötC in breathing is now discussed in major neuroscience and physiology textbooks and routinely taught to graduate and medical students. When doing our initial in vitro studies for breathing, we noticed unusual rhythmic patterns in the spinal cord, and this led to the development of the neonatal rodent spinal cord preparation for locomotion (Smith and Feldman, 1985, 1987; Smith et al., 1988; see pp. 89–90 in Stuart and Hultborn, 2008 for an account of our contribution). For better or worse, we did not have the resources to study locomotion, but it should be noted that our in vitro preparation is being used by > 20 laboratories worldwide to study locomotion (many in this book). At present, we are focusing on two distinct problems: 1. How do 1000 preBötC neurons generate not only a robust but also very labile respiratory rhythm?
What do we understand about rhythmogenesis? Quite frankly, very little that is testable other than issues of regional function (Fig. 1). In spite of a reasonable literature on preBötC neurons, most studies have been rather limited in scope, focusing on a small number of neuron types and never with an extensive dataset that seems truly representative. We cannot even say how many different types of preBötC neurons participate in rhythm generation or modulation, as characterized by their firing patterns under different in vitro conditions, their morphology, the receptor and channel distributions on their somatodendritic membranes, etc. Neither can we describe the relationship of these properties to physiological phenotype, for example, pacemaker versus nonpacemaker. Until recently, most if not all preBötC neurons were presumed to be glutamatergic, whereas it appears that about half are glycinergic (Winter et al., 2010) and that some of these neurons are even pacemakers (MorgadoValle et al., 2010). Perhaps most telling, we know little about the connectivity among preBötC neurons, with but a single study of ours that provides anecdotal information because the dataset was very limited (23 pairs of neurons; Rekling and Feldman, 1997). At two recent international symposia on control of breathing (St. Maximin, France, December
Anesthetized adult rat AL
mechanical ventilation (60 min) apnea recovery 10 s
Awake adult rat AL
apnea 10 s
Fig. 1. 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 recording of breathing movements. Allatostatin administered intracerebrocisternally induces a gradual decline of frequency and tidal volume until apnea develops after several minutes. After 60 min mechanical ventilation, rats resume spontaneous breathing. AAV2: adeno-associated virus 2; EGFP: enhanced green fluorescent protein. From Tan et al. (2008).
2008; Nara, Japan, July 2009), many models for respiratory rhythmogenesis were presented that shared the same problems: they were not testable, their modelers offered no predictions that could falsify them, they did not present any unusual predictions, and they did not provide any insight that could underlie the design of novel experiments. As I mentioned above, my research in breathing started with modeling (Feldman, 1976; Feldman and Cowan, 1975a,b), but I quickly turned to experimentation because any model developed at that time was poorly constrained (and that still seems to be the case). It remains straightforward to construct a mathematical/computational model that can reproduce a (carefully) selected subset of the experimental data. The principal reason is that many key parameters, for example, strength of synaptic connection between two neurons, are unknown, so one could always find a set of parameters, all with reasonable values within the “physiological range,” to fit the data. Unconstrained models that can accommodate new data simply by changing parameters can only establish proof of principle that a particular model can work but offer little insight into the neurobiological system. Yet, that is the “state-of-the-art” for models of respiratory rhythmogenesis. While contemporary models for generation of respiratory rhythm establish proof of principle insofar as they can mimic data, their utility in revealing biological mechanisms of respiratory rhythmogenesis is limited. The models are poorly constrained and represent mostly an exercise in parameterization, typically finding parameters that fit the data, and/or require stipulations that are simply not physiological, sometimes akin to modeling quadrapedal locomotion with wheels instead of limbs. For example, in many models of respiratory rhythmogenesis, neurons are points, ignoring data on the importance of dendrites of preBötC neurons (Morgado-Valle et al., 2008; Pace et al., 2007), and uniformly connected, ignoring data that their connections are most likely sparse (Rekling et al., 2000).
We may need to abandon the notion that this problem can be solved from first principles and accept the fact that we need to obtain tremendously detailed information to produce a dataset sufficient for representative and testable modeling, the neurobiological equivalent to sequencing the genome. The recent finding that the homeobox gene Dbx1 controls the fate of glutamatergic interneurons required for preBötC development (Bouvier et al., 2010; Gray et al., 2010) is a critical step in this direction and opens numerous possibilities for novel experiments to crack this network. 2. Testing the two-oscillator hypothesis (Feldman and Del Negro, 2006; Feldman et al., 2009; Janczewski and Feldman, 2006). When we discovered the preBötc, we hypothesized it was the sole generator of respiratory rhythm, both during development and adulthood. In 2003, we published a paper (Mellen et al., 2003) in which we hypothesized the presence of a second respiratory rhythm generator. Subsequent work in juvenile rats (Janczewski and Feldman, 2006) suggested that this oscillator was located rostral to the preBötc in the region of the retrotrapezoid nucleus/parafacial respiratory group (RTN/ pFRG) and was primarily a conditional oscillator for generating expiratory motor output; this is the core of our “two-oscillator” hypothesis. The existence of a second respiratory oscillator is still controversial (Onimaru et al., 2006), and despite developmental studies that show the critical role of RTN/pFRG in the ontogeny of respiratory rhythms (Jacquin et al., 1996; Thoby-Brisson et al., 2009), whether this second oscillator persists into adulthood is still debated (though with waning intensity). We recently obtained further evidence supportive of our two-oscillator hypothesis in adult mammals and demonstrated that indeed a conditional oscillator in the RTN/pFRG is responsible for active expiratory motor output when properly stimulated (Pagliardini et al., 2011). Furthermore, this expiratory oscillator is tightly coupled with the presumptive inspiratory oscillator in the preBötzinger Complex. We performed experiments
using a variety of methods, including pharmacological disinhibition, optogenetic stimulation, and single unit recordings, to show that activation of the adult RTN/pFRG turns on active expiration associated with silent or tonically firing neurons becoming rhythmic. We conclude that the induced transformation of tonic into rhythmic RTN/pFRG neurons is causal to the onset of active expiration. The most parsimonious explanation of our results is that the RTN/pFRG contains a conditional oscillator responsible for the generation of active expiratory activity in adult mammals. Much work remains to fully test this hypothesis. In summary, our understanding of the basic mechanisms underlying the neural control of breathing has substantially advanced since 1986. Two key sites are widely accepted to underlie the generation of the rhythm, but we are still in the dark about the mechanisms for rhythm generation, as well as how this rhythm gets transformed into a precisely controlled pattern of skeletal muscle activity that ultimately moves the appropriate amount of air to assure that blood gas are regulated and aerobic metabolism is continuously supported. My sense is that finding new “first principles” to understand the neural mechanisms, if they even exist, will be very difficult, and that the next substantial advance, even paradigm shift, will be delayed until we have a database sufficient to build meaningful models.
Acknowledgments This work was only possible through the generous and continuing support from the National Institutes of Health and the extraordinary scientists I have had the privilege to work with in my laboratory. Among them, with respect to the work I discussed here, I owe a special debt to Don McCrimmon, Jeffrey Smith, John Greer, Greg Funk, Jens Rekling, Paul Gray, Nick Mellen, Christopher Del Negro, Victor Janczewski, Consuelo Morgado-Valle, and Wenbin Tan.
Abbreviations AAV2 EGFP preBötC RTN pFRG
adeno-associated virus 2 enhanced green fluorescent protein pre-Bötzinger Complex retrotrapezoid nucleus parafacial respiratory group
References Bouvier, J., Thoby-Brisson, M., Renier, N., Dubreuil, V., Ericson, J., Champagnat, J., et al. (2010). Hindbrain interneurons and axon guidance signaling critical for breathing. Nature Neuroscience, 13, 1066–1074. Feldman, J. L. (1976). A network model for control of inspiratory cutoff by the pneumotaxic center with supportive experimental data in cats. Biological Cybernetics, 21, 131–138. Feldman, J. L. (1986). Neurophysiology of breathing in mammals. In F. E. Bloom (Ed.), Handbook of physiology; section I: The nervous system; volume IV: intrinsic regulatory systems of the brain (pp. 463–524). Bethesda, MD: American Physiological Society. Feldman, J. L., & Cowan, J. D. (1975a). Large-scale activity in neural nets I: Theory with application to motoneuron pool responses. Biological Cybernetics, 17, 29–38. Feldman, J. L., & Cowan, J. D. (1975b). Large-scale activity in neural nets II: A model for the brainstem respiratory oscillator. Biological Cybernetics, 17, 39–51. Feldman, J. L., & Del Negro, C. A. (2006). Looking for inspiration: New perspectives on respiratory rhythm. Nature Reviews. Neuroscience, 7, 232–242. Feldman, J. L., Kam, K., & Janczewski, W. A. (2009). Practice makes perfect, even for breathing. Nature Neuroscience, 12, 961–963. Gray, P. A., Janczewski, W. A., Mellen, N., McCrimmon, D. R., & Feldman, J. L. (2001). Normal breathing requires preBötzinger complex neurokinin-1 receptor-expressing neurons. Nature Neuroscience, 4, 927–930. Gray, P. A., Hayes, J. A., Ling, G. Y., Llona, I., Tupal, S., Picardo, M. C., Ross, S. E., Hirata, T., Corbin, J. G., Eugenin, J., & Del Negro, C. A. (2010). Developmental origin of preBotzinger complex respiratory neurons. Journal of Neuroscience, 30, 14883–14895. Jacquin, T. D., Borday, V., Schneider-Maunoury, S., Topilko, P., Ghilini, G., Kato, F., et al. (1996). Reorganization of pontine rhythmogenic neuronal networks in Krox20 knockout mice. Neuron, 17, 747–758. Janczewski, W. A., & Feldman, J. L. (2006). Distinct rhythm generators for inspiration and expiration in the juvenile rat. Journal de Physiologie, 570, 407–420.
218 Mellen, N. M., Janczewski, W. A., Bocchiaro, C. M., & Feldman, J. L. (2003). Opioid-induced quantal slowing reveals dual networks for respiratory rhythm generation. Neuron, 37, 821–826. Morgado-Valle, C., Beltran-Parrazal, L., Difranco, M., Vergara, J. L., & Feldman, J. L. (2008). Somatic Ca2þ transients do not contribute to inspiratory drive in preBötzinger Complex neurons. Journal de Physiologie, 586, 4531–4540. Morgado-Valle, C., Baca, S. M., & Feldman, J. L. (2010). Glycinergic pacemaker neurons in preBötzinger complex of neonatal mouse. The Journal of Neuroscience, 30, 3634–3639. Onimaru, H., Homma, I., Feldman, J. L., & Janczewski, W. A. (2006). The para-facial respiratory group (pFRG)/ preBötzinger Complex (preBötC) is the primary site of respiratory rhythm generation in the mammal. Journal of Applied Physiology, 100, 2094–2098. Pace, R. W., Mackay, D. D., Feldman, J. L., & Del Negro, C. A. (2007). Inspiratory bursts in the preBotzinger complex depend on a calcium-activated non-specific cation current linked to glutamate receptors in neonatal mice. Journal de Physiologie, 582, 113–125. Pagliardini, S., Janczewski, W. A., Tan, W., Dickson, C. T, Deisseroth, K., & Feldman, J. L. (2011). Active expiration induced by excitation of ventral medulla in adult anesthetized rats. Journal of Neuroscience. (In Press). Rekling, J. C., & Feldman, J. L. (1997). Bidirectional electrical coupling between inspiratory motoneurons in the newborn mouse nucleus ambiguus. Journal of Neurophysiology, 78, 3508–3510. Rekling, J. C., Shao, X. M., & Feldman, J. L. (2000). Electrical coupling and excitatory synaptic transmission between rhythmogenic respiratory neurons in the preBötzinger complex. Journal of Neuroscience, 20, RC113. Smith, J. C., & Feldman, J. L. (1985). Motor patterns for respiration and locomotion generated by an in vitro brainstem-
spinal cord from neonatal rat. Society for Neurocience Abstracts, 11, 24. Smith, J. C., & Feldman, J. L. (1987). In vitro brainstem-spinal cord preparations for study of motor systems for mammalian respiration and locomotion. Journal of Neuroscience Methods, 21, 321–333. Smith, J. C., Feldman, J. L., & Schmidt, B. J. (1988). Neural mechanisms generating locomotion studied in mammalian brain stem-spinal cord in vitro. The FASEB Journal, 2, 2283–2288. Smith, J. C., Ellenberger, H. H., Ballanyi, K., Richter, D. W., & Feldman, J. L. (1991). Pre-Bötzinger complex: A brainstem region that may generate respiratory rhythm in mammals. Science, 254, 726–729. Stuart, D. G., & Hultborn, H. (2008). Thomas Graham Brown (1882–1965), Anders Lundberg (1920-), and the neural control of stepping. Brain Research Reviews, 59, 74–95. Suzue, T. (1984). Respiratory rhythm generation in the in vitro brain stem-spinal cord preparation of the neonatal rat. Journal de Physiologie, 354, 173–183. Tan, W., Janczewski, W. A., Yang, P., Shao, X. M., Callaway, E. M., & Feldman, J. L. (2008). Silencing preBötzinger complex somatostatin-expressing neurons induces persistent apnea in awake rat. Nature Neuroscience, 11, 538–540. Thoby-Brisson, M., Karlen, M., Wu, N., Charnay, P., Champagnat, J., & Fortin, G. (2009). Genetic identification of an embryonic parafacial oscillator coupling to the preBotzinger complex. Nature Neuroscience, 12, 1028–1035. Winter, S. M., Fresemann, J., Schnell, C., Oku, Y., Hirrlinger, J., & Hulsmann, S. (2010). Glycinergic interneurons in the respiratory network of the rhythmic slice preparation. Advances in Experimental Medicine and Biology, 669, 97–100.