452313 Cell Population-Based Mathematical Model of Liver Homeostatic Renewal

Thursday, November 17, 2016: 1:42 PM
Continental 8 (Hilton San Francisco Union Square)
Daniel Cook, Department of Chemical and Biomolecular Engineering, University of Delaware, Newark, DE, Babatunde A. Ogunnaike, Department of Chemical Engineering, University of Delaware, Newark, DE and Rajanikanth Vadigepalli, Daniel Baugh Institute for Functional Genomics and Computational Biology, Department of Pathology, Anatomy and Cell Biology, Thomas Jefferson University, Philadelphia, PA

Many chronic diseases impair the self-renewal capability of hepatocytes in the liver. This impaired capacity for renewal may even be one of the major factors contributing to the progression of chronic liver disease through steatohepatitis and fibrosis to cirrhosis. As evidence of this, in the intermediate and late stages of chronic liver disease hepatocyte death leads to scar tissue formation rather than hepatocyte renewal. The harmful effects of chronic diseases on liver regeneration have been widely studied using experimental and computational methods, however, the effects of chronic diseases on homeostatic liver renewal are less well understood.

Even though it is widely recognized that much of liver cellular turnover is accomplished through a continuous process of liver homeostatic renewal, partial hepatectomy followed by liver regeneration is often still the methodology of choice used to investigate dynamic liver function and tissue renewal capability. Regeneration, however, provides a picture of liver function under abnormal operating conditions rather than during normal function; it is therefore of limited utility for investigating everyday tissue function. We showed earlier that an integrated network of multiple cell types is required for liver regeneration following partial hepatectomy and that chronic ethanol use causes an imbalance between populations of pro-regenerative and anti-regenerative hepatic stellate cells prior to hepatectomy. We further showed that this homeostatic imbalance in hepatic stellate cell populations, coupled with the well-documented increase in cytokine production in alcoholic liver disease (ALD), could explain impaired regeneration caused by chronic ethanol abuse. Because diseases such as ALD cause homeostatic imbalances to cell populations, it is possible that such imbalances affect homeostatic renewal in addition to impairing regeneration. Therefore, using a cell population-based mathematical modeling framework, this current study extends that previous work by investigating how homeostatic renewal may be regulated in the healthy liver and how chronic insults and disease adaptation could affect liver homeostatic renewal.

We develop a mathematical model of liver homeostatic renewal that takes into account two recently described populations of hepatocytes: Axin2+ and Axin2-. Hepatocyte renewal occurs predominantly by replication of Wnt-responsive Axin2+ “stem-cell-like” hepatocytes, which have been shown to cluster pericentrally and which respond to factors produced by pericentral endothelial cells (i.e., WNT-2a). As Axin2+ hepatocytes renew, some of the daughter cells migrate from the pericentral region, become Axin2- cells, and populate the remainder of the liver. We investigate the dynamic behavior of this two-state system by simulating tissue renewal, first assuming no feedback between populations, and subsequently investigating the effect of combinations of feedbacks, including product inhibition, cellular competition, and others. We use a systematic design of experiment-based approach to investigate the system behavior under all possible combinations of feedback configurations, specifically characterizing system recovery by calculating total deviation from nominal cell levels and overall recovery time in response to transient perturbations in apoptosis rates and proliferation rates. This exercise allows us to identify which of the investigated models of liver homeostatic renewal responds most robustly to multiple types of disturbances. We assume that nature selects for robust behavior, so that the most robust model indicates the most biologically plausible configuration.

In the presentation, we will discuss our modeling and analysis approach in detail, especially our investigation of the dynamic behavior of our homeostatic renewal model using frequency-based analyses, sensitivity analyses, and stability analyses. We will also discuss our model simulations of chronic insults to healthy and diseased livers. Our simulations and model analyses results lead us to predict that liver homeostatic renewal is governed by a combination of intrinsic (passive) system stability and active control by non-parenchymal cell networks. We further predict that chronic diseases likely impair both the hepatocyte homeostatic renewal system and control by non-parenchymal cell networks.

Research Support: R01 AA018873, R21 AA022417, T32 AA007463, F31 AA023445


Extended Abstract: File Not Uploaded