Liver regeneration and cellular adaptation to chronic diseases: a systems biology investigation

Date
2016
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
Liver resection is used in the clinic for treatment of hepatocellular carcinoma and for live liver transplant. In otherwise healthy patients following resection, the liver initiates a program of regeneration that involves multiple cell types interacting across multiple length and time scales. As early as 30 seconds after injury, signaling cascades become active within the liver. Dynamic molecular changes continue for approximately 1 week following injury, restoring liver mass to the pre-injury levels. The dynamics of several molecular mediators of this process has been investigated previously to identify transient activation of inflammatory molecules for a few hours and a sustained activation of growth factors over several days. Yet much remains to be understood as to how the regulation of multiple molecular factors is coordinated to control liver repair mechanisms. Additionally, cell-types within the liver can each take on multiple distinguishable phenotypes either contributing to or inhibiting repair. The contributions of these phenotypes to liver repair and disease progression are just beginning to be appreciated. ☐ In contrast to regeneration in otherwise healthy patients, patients requiring resection or transplant likely have multiple comorbidities impairing regeneration. These comorbidities include chronic diseases and common pharmaceutical and recreational drug use (including alcohol abuse). Despite extensive study, the molecular mechanisms governing comorbidity-impaired liver regeneration remain incompletely understood. As a result, there are no robust predictors of liver regenerative capacity in patients undergoing liver resection. ☐ In light of these complexities, we have taken a systems biology approach to understanding liver regeneration in health and disease. We measured the dynamics of genome-wide transcription factor binding of an early, pro-inflammatory responder in liver regeneration (NF-κB) to identify broad features of the pro-inflammatory regeneration response in rat livers. We then investigated a broad range of cytokines, chemokines, and growth factors that respond to pro-inflammatory signals in several cases of successful regeneration to identify possible changes to the liver microenvironment during regeneration. These analyses implicated non-parenchymal cells as important mediators of the successful regeneration response (total mass recovery). ☐ We therefore developed a computational model of liver regeneration that takes into account molecular regulation in hepatocytes contributing to regeneration and course-grained estimations of non-parenchymal cell activation. Using this model, we predicted experimental liver regeneration profiles across multiple species including mice, rats, and humans by tuning a single parameter empirically related to body mass. Additionally, we predicted the molecular mechanisms governing impaired liver regeneration in multiple chronic disease conditions impairing regeneration, including alcoholic steatohepatitis. Our results implicated non-parenchymal cells as important regulators of the dynamics of regeneration in addition to overall mass recovery. ☐ We extended our computational model to synthesize the intrinsically multi- scale nature of liver regeneration by simulating connections between physiological- scale dynamics, transcriptional phenotypes of non-parenchymal cells, and molecular signaling networks. Model analysis showed that shifting balances between populations of non-parenchymal cell activation phenotypes was sufficient to alter regeneration dynamics and overall tissue recovery following partial hepatectomy. As a perturbation to regeneration phenotype, we simulated alcohol-mediated suppression of liver regeneration by fitting our model to experimental data of liver recovery following chronic alcohol consumption and partial hepatectomy. Based on the model simulations, we predict that chronic alcohol consumption acts at a cellular-scale by shifting Kupffer cells from an M1 phenotype to an M2 phenotype following partial hepatectomy and by shifting hepatic stellate cells from a pro-regenerative phenotype to an anti-proliferative phenotype. At a molecular-scale, these changes in cell phenotypes are paralleled by dynamic increases in anti-inflammatory cytokine production and high levels of the anti-regenerative molecules such as fibrous collagens and TGFβ. ☐ We tested these predictions using high-throughput measurements following partial hepatectomy in ethanol-fed rats and controls. We used laser capture microdissection to collect individual hepatic stellate cells from the livers of chronic ethanol-fed animals and controls before and after hepatectomy. We then used a high- throughput gene expression platform to quantify mRNA levels of ~100 genes across these individual cells and used multivariate statistics to identify clusters of hepatic stellate cell transcriptional phenotypes. Our experimental results indicate that multiple transcriptional phenotypes of hepatic stellate cells arise following partial hepatectomy and ethanol exposure, consistent with model predictions. Furthermore, our results suggest that one of the main effects of chronic ethanol consumption is to imbalance hepatic stellate cell populations prior to resection, leading to altered extracellular matrix composition, increased matrix stiffness, and decreased intercalation with pro- regenerative molecules. ☐ We then used similar modeling techniques to simulate homeostatic renewal of hepatocytes in the non-regenerating liver. Our results point to the existence of strong feedbacks within hepatocyte populations governing homeostatic renewal and an external control of hepatocyte renewal by a non-parenchymal cell network that involves multiple cell types and matrix property modulation. Model simulations were able to capture recently observed behaviors of tissue renewal involving stem cells following induced senescence of hepatocytes. ☐ Taken together, these results provide novel insights into how molecular regulation during liver regeneration influences multiple scales to regulate non- parenchymal cell transcriptional phenotype and overall tissue mass recovery.
Description
Keywords
Citation