The current issue of Nature Reviews Neuroscience came with an incredible article on the control of breathing. What amazed me the most is that currently the mechanism of rhythm generation is not known precisely.

Breathing is the most conspicuous rhythmic behavior of humans, it consists of repeating the cycle of inspiration, postinspiration (a phase in which the muscle contraction achieved by the inspiratory phase is upheld) and expiration. The authors discuss the neural structures generating these processes in depth, and then turn to the models explaining the neural basis of the respiratory cycle.

The first model emerged in the 70’s and proposed that recurrent excitatory activity generates inspiration which, when it reaches a certain limit, activates an inhibitory system. The inhibition is the cause of expiration, as it prevents the activation of the neurons involved in inspiration for the time it takes to exhale. Even in its earliest form the model could explain the activity pattern seen in the nerve controlling the muscles of breathing, but the authors present a convincing analysis of the experimental data and conclude that inhibitory activity is not necessary for the rhythmic activity to emerge. Then in the early 90’s researchers found pacemaker cells in the relevant brain area in brain slices. Pacemaker cells have the ability to generate action potentials without any input (they probably have constantly open ion channels). However, the experimental data is again deemed insufficient to support the model, for example these pacemaker cells have never been seen in live animals.

Another model, the network oscillator, appeared in the 2000s. The idea is that the cells are silent at the beginning of the breathing cycle, but then spontaneous activity emerges in some neurons and they excite their neighbors and then the neighbors excite their neighbors and so on. This collective activity is followed by a refractory period enabling expiration. The main difference to the pacemaker hypothesis is that in this case a network of neurons produces the rhythm and not a few pacemaker cells (even though the starting point of the cycle depends on the spontaneous activity of some cells). According to the authors, this is the only model which currently seems feasible to explain the rhythm generation for breathing.

Other things concerning the models:
– burst firing of neurons may not be needed for rhythm generation (which is considered a basic feature of most models)
– glial cells may have an important role in network dynamics (which are not included in the current models)
– most models are based on rodent physiology, but larger mammals breathe slower and that may necessitate different mechanisms

All this points to the fact that the neural control of breathing is beyond current understanding, even though it is probably one of the simpler functions of the nervous system. Makes one think about the value of BOLD response based computational theories of brain function.

Finally there is a section which discusses the effects of the breathing rhythm on cognitive and emotional functions. The anatomical connections are there, but the means of these effects are not yet clear. There are some references for recent research showing breathing-coupled rhythms in the hippocampus (which is involved in memory):
– Nguyen Chi, V. et al. Hippocampal respiration-driven rhythm distinct from theta oscillations in awake mice. J. Neurosci. 36, 162–177 (2016).
– Zelano, C. et al. Nasal respiration entrains human limbic oscillations and modulates cognitive function. J. Neurosci. 36, 12448–12467 (2016).

That’s all I wanted to share.