You are here:
Virtual Tissues in Toxicology
SHAH, I. A. AND J. F. WAMBAUGH. Virtual Tissues in Toxicology. JOURNAL OF TOXICOLOGY AND ENVIRONMENTAL HEALTH - PART B: CRITICAL REVIEWS. Taylor & Francis, Inc., Philadelphia, PA, 13(2-4):314-328, (2010).
Virtual Tissues (VT) are synergistic with ongoing and future toxicology efforts. First, VT directly link with PBPK models to simulate microdosimetry at environmentally relevant exposures. Second, VT also use existing dose-response models that describe molecular alterations and cellular responses. Third, the cell-oriented nature of VT is well suited for using in vitro high-content data to calibrate and evaluate cellular responses. The U.S. EPA ToxCast project is producing such high-throughput data for a large number of environmental chemicals. This will provide mechanistic information concerning pathway perturbations to quantitatively describe the relationship between cellular processes and subsequent cell responses. In the long term, VT offer an in silico approach for extrapolating in vitro endpoints to in vivo human effects across chemicals, environmentally relevant doses, and chronic exposure durations.
New approaches are vital for efficiently evaluating human health risk of thousands of chemicals in commerce. In vitro models offer a high-throughput approach for assaying chemical-induced molecular and cellular changes; however, bridging these perturbations to in vivo effects across chemicals, dose, time, and species remains challenging. Technological advances in multiresolution imaging and multiscale simulation are making it feasible to reconstruct tissues in silico. In toxicology, these “virtual” tissues (VT) aim to predict histopathological outcomes from alterations of cellular phenotypes that are controlled by chemical-induced perturbations in molecular pathways. The behaviors of thousands of heterogeneous cells in tissues are simulated discretely using agent-based modeling (ABM), in which computational “agents” mimic cell interactions and cellular responses to the microenvironment. The behavior of agents is constrained by physical laws and biological rules derived from experimental evidence. VT extend compartmental physiologic models to simulate both acute insults as well as the chronic effects of low-dose exposure. Furthermore, agent behavior can encode the logic of signaling and genetic regulatory networks to evaluate the role of different pathways in chemical-induced injury. To extrapolate toxicity across species, chemicals, and doses, VT require four main components: (a) organization of prior knowledge on physiologic events to define the mechanistic rules for agent behavior, (b) knowledge on key chemical-induced molecular effects, including activation of stress sensors and changes in molecular pathways that alter the cellular phenotype, (c) multiresolution quantitative and qualitative analysis of histologic data to characterize and measure chemical-, dose-, and time-dependent physiologic events, and (d) multiscale, spatiotemporal simulation frameworks to effectively calibrate and evaluate VT using experimental data. This investigation presents the motivation, implementation, and application of VT with examples from hepatotoxicity and carcinogenesis.