Afferents integration and neural adaptive control of breathing

Abstract

The respiratory regulatory system is one of the most extensively studied homeostatic systems in the body. Despite its deceptively mundane physiological function, the mechanism underlying the robust control of the motor act of breathing in the face of constantly changing internal and external challenges throughout one's life is still poorly understood. Traditionally, control of breathing has been studied with a highly reductionist approach, with specific stimulus-response relationships being taken to reflect distinct feedback/feedforward control laws. It is assumed that the overall respiratory response could be described as the linear sum of all unitary stimulus-response relationships under a Sherringtonian framework. Such a divide-and-conquer approach has proven useful in predicting the independent effects of specific chemical and mechanical inputs. However, it has limited predictive power for the respiratory response in realistic disease states when multiple factors come into play. Instead, vast amounts of evidence have revealed the existence of complex interactions of various afferent-efferent signals in defining the overall respiratory response. This thesis aims to explore the nonlinear interaction of afferents in respiratory control. In a series of computational simulations, it was shown that the respiratory response in humans during muscular exercise under a variety of pulmonary gas exchange defects is consistent with an optimal interaction of mechanical and chemical afferents. This provides a new understanding on the impacts of pulmonary gas exchange on the adaptive control of the exercise respiratory response. Furthermore, from a series of in-vivo neurophysiology experiments in rats, it was discovered that certain respiratory neurons in the dorsolateral pons in the rat brainstem integrate central and peripheral chemoreceptor afferent signals in a hypoadditive manner. Such nonlinear interaction evidences classical (Pavlovian) conditioning of chemoreceptor inputs that modulate the respiratory rhythm and motor output. These findings demonstrate a powerful gain modulation function for control of breathing by the lower brain. The computational and experimental studies in this thesis reveal a form of associative learning important for adaptive control of respiratory regulation, at both behavioral and neuronal levels. Our results shed new light for future experimental and theoretical elucidation of the mechanism of respiratory control from an integrative modeling perspective.