Editors' ChoicePathway Crosstalk

Double the Inputs Does Not Mean Double the Outputs

See allHide authors and affiliations

Science's STKE  13 Jun 2006:
Vol. 2006, Issue 339, pp. tw199
DOI: 10.1126/stke.3392006tw199

Although it is convenient to think of cell signaling cascades as linear responses to extracellular inputs, the reality is that cells receive multiple stimuli and the signaling network is highly interconnected with many points of crosstalk and feedback. Crosstalk among signaling pathways results in cellular responses that may not simply be the additive effects of the ligands present, but rather a context-dependent response. Natarajan et al. prove this by analyzing signaling responses to 22 single ligands and 231 pairwise combinations in RAW 264.7 macrophages (data from the Alliance for Cellular Signaling). The responses were changes in the concentrations of intracellular calcium and cAMP, phosphorylation status of 21 signaling proteins, and secretion of 18 cytokines. For single ligand experiments, the raw data were transformed into a mathematical value, the Z-score, to allow uniform statistical analysis. For the double ligand experiments, a new score representing the nonadditive effects of the two ligands on the cellular responses was calculated (ΔΔZ). In many cases, crosstalk between the two pathways was evident, with examples including synergy (response greater than the predicted additive response) and saturation of the response (the response was the same as that for either ligand alone). These nonadditive responses were consistent with previously reported crosstalk. In some cases, the authors chose to further investigate the mechanism of crosstalk experimentally. Each of the 22 ligands showed a nonadditive interaction with at least one other ligand for regulation of cytokine secretion. The authors suggest that the grouping of ligands based on their effects on cytokine secretion suggest a limited set of "interaction agents," which are defined as signaling circuits that couple distinct signaling pathways. In some aspects, this analysis confirms existing data. Also the analysis points to new points of crosstalk and suggests that the complexity of the cellular response to any ligand is highly context-dependent, with the potential to vary with the cell's signaling history and current signaling environment (presence of one or more ligands).

M. Natarajan, K.-M. Lin, R. C. Hsueh, P. C. Sternweis, R. Ranganathan, A global analysis of cross-talk in a mammalian cellular signalling network. Nat. Cell Biol. 8, 571-580 (2006). [PubMed]

Stay Connected to Science Signaling