Highlights
Modeling the Dynamic Range of a Neuronal Network for Breathing
Submitted by the Mathematical Biosciences Institute (MBI)
For humans and other mammals, breathing is
essential to life. The breathing rhythm relies on
an area of the brain stem known as the preBötzinger
complex, a network of neurons
exhibiting rhythmic bursts of activity that
initiate inspiration. The frequency of the rhythm
varies in response to such challenges as
exercise, sleep, or changes in altitude. The preBötzinger
complex also participates in detecting
reduced concentration of oxygen in the blood,
stimulating a gasping response in order to
restore healthy oxygen levels. This response is critical to the more than 12 million
Americans diagnosed with Obstructive Sleep Apnea, a disorder in which a sleeping
individual's breathing undergoes prolonged pauses broken by gasps or sighs. The failure
to gasp has been implicated in Sudden Infant Death Syndrome. It has become clear that
an understanding of the cellular mechanisms underlying normal and distressed breathing
would be invaluable in treating these and a host of other respiratory ailments.
Experimental and theoretical studies
have led to development of a biophysical
model of a network of preBötzinger
cells that provides a framework for
analyzing the potential roles of cellular
and synaptic processes in the generation
and control of rhythm. The model has
been used to make observations
concerning how a network of cells may
generate different activity patterns than
those of the individual cells that the
network comprises. Of particular
significance were the findings that the model network was both more flexible and more
robust than the individual cells: network activity could exhibit a wider range of
frequencies than could the individual cells, and the network was also able to sustain
rhythmic bursting activity even when properties of cells were varied so that isolated cells
were incapable of such activity.
MBI postdoctoral researchers Janet Best, Alla Borisyuk, and Martin Wechselberger
joined with mathematicians Jonathan Rubin (University of Pittsburgh) and David Terman
(The Ohio State University) to provide a thorough mathematical analysis of the
mechanisms underlying these observations. Using geometric dynamical systems
techniques, predominantly a nonstandard fast/slow decomposition and bifurcation
analysis approach, their work has elucidated the dynamical mechanisms that lead to
various activity patterns seen in the network. Unexpectedly, they have found that the
rhythmic bursting activity of the network qualitatively differs from that of individual,
isolated cells; in particular, cells in the network typically are not firing synchronously,
and this lack of synchronization plays a key role in allowing the network to exhibit
rhythmic bursting even when individual cells would not. Their analysis predicts that these
qualitative differences may be detectable experimentally in the preBötzinger
complex
and provides implications for quantitative aspects of network activity, including
frequency selection. This study also advances the current mathematical understanding of
transitions between activity modes in more general networks of cells.
The MBI brought these researchers together during the thematic year on Mathematical
Modeling of Cell Processes, and most of the research for this project was conducted on
the premises of MBI. As ongoing experiments have revealed further details concerning
intrinsic properties of cells in the preBötzinger
complex, the biophysical model has
continued to be refined, and the analytical approach developed in this project continues to
be applied by these researchers and by others. These cells are now postulated to play an
essential role in generating breathing rhythms in mammals, and delineating their
properties may provide a means to understand and treat lifethreatening
failures of
breathing such as Sudden Infant Death Syndrome and some sleep apneas.
[Note: The figures are legal.] Caption for the burster image: A "top hat" burster can arise
when preBötzinger
cells are coupled in the model.
|