How does a receptive cell "compute" its response to a ligand? Despite large amounts of accumulated data on signal transduction pathways, our inability to accurately predict the response of cells to ligands or drugs indicates that our answer to this question are still incomplete. Signal transduction networks are large, interconnected and highly dynamic;we still need to understand how cells integrate signals in time and space. Tumor Necrosis Factor (TNF), a regulator of inflammation, is a particularly interesting model system for signal transduction because it is a ligand that induces opposing pro-survival and pro-death signaling pathways. Although TNF receptor 1 (TNFR1) expression is ubiquitous, some cells respond to TNF by differentiating and proliferating, while others commit to cell death. Strikingly, even clonal cancer cells treated with high TNF concentrations show variability: some cells die but others survive. What determines whether a TNF-treated cancer cell survives or dies? To tackle this question, we will: 1. Use same-cell tracking of NF-?B and caspase signaling dynamics with automated image analysis to quantify the respective contributions of NF-?B and caspase signaling dynamics as well as cellular context to TNF- induced cell fate decision. Using these data we will test the hypotheses that: i) NF-?B activation dynamics influence caspase activation dynamics and ii) both signaling dynamics and extracellular context are strong contributors to TNF-induced cell fate. 2. Decode the logic by which cells integrate intracellular signals and external context to commit to a TNF- induced cell fate. Using multivariable regression analysis of our data from Aim 1 we will build models connecting both extracellular cues and intracellular signals to TNF-induced cell fate. By comparing competing models of TNF-induced cell decision processes, we will derive mechanistic insights into the regulatory circuitry driving TNF-induced cell fate. 3. Focusing on the transcriptional arm of the TNF signaling network, we will test whether NF-kB nuclear translocation dynamics can quantitatively predict transcriptional output. Using a novel workflow to perform same-cell imaging of NF-kB dynamics and mRNA counting by single-molecule FISH, we will directly establish the relationship between NF-kB translocation dynamics and its transcriptional activity. Finally, we will use protein-binding microarrays to ask how competition between p65-p50 heterodimers and p50-p50 homodimers contributes to decoding NF-kB activation and to cell line-to-cell line variability in response to TNF. With this work, we take a new approach to studies of signal transduction, harnessing cell-to-cell variability to gain a quantitative understanding of how cells integrate information to determine their behavior in response to a ligand. These approaches will contribute to understanding the multi-factorial control of TNF-induced cell death and will be broadly applicable to the study of other signal transduction networks.
Cancer therapies are generally impeded by a phenomenon called fractional kill, when a proportion of cancer cells evade treatments designed to kill them. Often, genetic variation can explain fractional kill, but this phenomenon is also observed within populations of genetically identical cancer cells. One important consequence of fractional kill is that the surviving cancer cells often develop drug resistance. The goal of our research is to identify non-genetic sources of variability in cancer cells, and ultimately use these insights to predict and improve therapeutic response in cancer patients.
|Xia, X; Owen, M S; Lee, R E C et al. (2014) Cell-to-cell variability in cell death: can systems biology help us make sense of it all? Cell Death Dis 5:e1261|
|Lee, Robin E C; Walker, Sarah R; Savery, Kate et al. (2014) Fold change of nuclear NF-?B determines TNF-induced transcription in single cells. Mol Cell 53:867-79|