Text

theses

Inflammation is a critical component of the physiological response to stress. Ideally, the endpoint of inflammation is restoration of homeostasis. However, when anti-inflammatory processes fail to appropriately balance pro-inflammatory signals, inflammation can have deleterious effects. Clinically, this is problematic in diseases such as sepsis because therapies to control inflammation are limited. Rhythmic biological signals, ranging in time scale from very fast neural oscillations to very slow seasonal patterns, are ubiquitous in physiological systems, including those linked to inflammation. The characteristics of an oscillatory signal reflect the state of underlying rhythm-generating physiological systems. Through studying homeostatic rhythms and their disruption in inflammation, we gained insight into the underlying mechanisms and potential diagnostic utility embedded in physiologic variability via mathematical modeling and systems biology approaches. Based on a mathematical model of human endotoxemia, an experimental model of systemic inflammation, we hypothesized that hormones entrain inflammatory mediators and impose circadian patterns on the endotoxemia response. A model of the interplay between inflammation and circadian rhythms was developed and validated based on the temporal sensitivity of the inflammatory response. In addition to circadian rhythms, ultradian (~1hr) rhythms in cortisol are another prominent hormonal pattern linked to inflammation. By combining models of cortisol production and pharmacodynamics, we evaluated the downstream implications of ultradian rhythms, finding a relationship between the amplitude of homeostatic ultradian rhythms and responsiveness to subsequent stress. Heart rate variability (HRV) has been studied as a potential prognostic marker in inflammation-linked diseases. We modeled the interactions between human endotoxemia and the autonomic nervous system to understand the loss of HRV in response to stress, allowing for the rationalization of experimental observations in the framework of a quantitative model. Finally, we identified a warning signal of transitions between steady states in a mathematical model of chronic endotoxemia, illustrating the potential for our research to be applied in a translational context. These results all point towards the importance of rhythmic patterns in the underlying physiological systems driving the inflammatory response and the potential for useful information about these systems to be derived from the analysis of variability in physiological signals.

ETD_4894

Ph.D.

Includes bibliographical references

by Jeremy D. Scheff

doi:10.7282/T3571922

ETD doctoral

The author owns the copyright to this work.