Signal transduction through dedicated molecular pathways and networks in live cells is subject to substantial variability. Thus the precision of detection of the signaling inputs is not absolute, limiting the amount of transmitted information and thus the spectrum of signal-conditioned phenotypic outcomes. In the previous funding periods, we developed a new framework for analysis of information transfer in noisy signaling networks and applied it to various networks, most notably the tumor necrosis factor (TNF) stimulated activation of diverse targets, including, most prominently the NF-kappaB transcription factor. We have also built a progressively experimentally validated deterministic model of this signaling network, which has been extensively used both in the analysis of the innate immune response and in development and validation of diverse systems biology tools. To enable high-throughput analysis of signaling networks, we have also developed a number of novel micro-fabricated tools and streamlined high throughput single cell analysis. Although this work and the work of many other labs have done much to uncover multiple insights into the workings of the TNF-activated signaling networks and many other eukaryotic signaling pathways and networks, much remains to be explored. First, we still lack complete understanding of how the variability (noise) is distributed across complex multi-node signaling networks. Knowledge of such noise distributions can both help us train more realistic stochastic mathematical models of the signaling processes and better understand the mechanisms of information transfer through complex biochemical `channels'. Second, in many cases, an in TNF induced signaling processes in particular, analysis is frequently performed in model cell lines, often without a clear connection to phenotypic outputs (the cell decision making). Thus it is not clear how much of the information transferred through signaling networks is used to inform cell decisions. Third, the analysis of signaling processes is mostly performed by imposing simplistic inputs, rather than reconstituting the signaling inputs that can be generated by live cells communicating TNF and other ligand-based signals. Thus, in experiments in cell culture, we might not study the response to the most relevant or informative inputs. In this application, we propose a set of modeling and experimental studies to address these challenges, by focusing on more relevant cell types: macrophages and endothelial cells, and the relevant phenotypic outcome of TNF signaling process: an increase in the endothelial layer permeability. We propose to build detailed stochastic analysis of TNF signaling in these cell types and use it to describe heterotypic cell-cell communication. A result of this analysis will be a set of tools to analyze the effect of lipopolysaccharide (LSP) and TNF signaling on an important aspect of the innate immune response, the endothelial layer integrity control leading to monocyte recruitment, which will also be universally applicable to other signaling and cell-cell communication systems.
Information transfer in living cells through dedicated biochemical pathways is subject to variability, or noise, thus limiting the extent to which cell functions can be informed by external cues. In this proposal, we explore how the behavior of endothelial cells is informed by the signals representing bacterial infection, presented either directly or through the intermediate communication with macrophage cells. As a result of this analysis, we will develop a set of tools informing us on the outcome of noisy inflammatory signaling resulting in different blood vessel permeability, within a widely applicable framework describing more generally information transfer in biochemical networks and pathways.
Showing the most recent 10 out of 46 publications
