Cellular Decision Making by Non-Integrative Processing of TLR Inputs

• Published in: Cell reports (Cambridge) Vol. 19; no. 1; pp. 125 - 135
• Main Authors: Kellogg, Ryan A., Tian, Chengzhe, Etzrodt, Martin, Tay, Savaş
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 04.04.2017; Elsevier
• Subjects: 3T3 Cells; Animals; Bacterial Infections - metabolism; Cell Communication; digital; dynamics; Humans; immunity; information processing; Lipopeptides - pharmacology; Lipopolysaccharides - pharmacology; live-cell imaging; Mice; Microfluidic Analytical Techniques; microfluidics; MicroRNAs - genetics; MicroRNAs - metabolism; NF-kappa B - metabolism; Signal Transduction; signaling; single cell; synergy; TNF Receptor-Associated Factor 6 - genetics; TNF Receptor-Associated Factor 6 - metabolism; Toll-Like Receptor 2 - antagonists & inhibitors; Toll-Like Receptor 2 - metabolism; Toll-Like Receptor 4 - antagonists & inhibitors; Toll-Like Receptor 4 - metabolism; toll-like receptors; digital; information processing; single cell; dynamics; signaling; synergy; toll-like receptors; live-cell imaging; immunity; microfluidics
• ISSN: 2211-1247; 2211-1247
• DOI: 10.1016/j.celrep.2017.03.027

Cells receive a multitude of signals from the environment, but how they process simultaneous signaling inputs is not well understood. Response to infection, for example, involves parallel activation of multiple Toll-like receptors (TLRs) that converge on the nuclear factor κB (NF-κB) pathway. Although we increasingly understand inflammatory responses for isolated signals, it is not clear how cells process multiple signals that co-occur in physiological settings. We therefore examined a bacterial infection scenario involving co-stimulation of TLR4 and TLR2. Independent stimulation of these receptors induced distinct NF-κB dynamic profiles, although surprisingly, under co-stimulation, single cells continued to show ligand-specific dynamic responses characteristic of TLR2 or TLR4 signaling rather than a mixed response, comprising a cellular decision that we term “non-integrative” processing. Iterating modeling and microfluidic experiments revealed that non-integrative processing occurred through interaction of switch-like NF-κB activation, receptor-specific processing timescales, cell-to-cell variability, and TLR cross-tolerance mediated by multilayer negative feedback. [Display omitted] •Bacterial infection involves co-stimulation of TLR4 and TLR2 receptors•TLR4 and TLR2 induced distinct dynamic NF-κB profiles when stimulated separately•Under co-stimulation, single cells responded to either one ligand or the other•Switch-like NF-κB response and TLR cross-tolerance mediate non-integrative processing Kellogg et al. study NF-κB responses under TLR co-stimulation and observe that single cells respond with a dynamic NF-κB profile characteristic of either TLR2 or TLR4 activation, rather than a mixed response. Modeling and microfluidic experiments revealed that switch-like NF-κB activation and TLR cross-tolerance mediate this phenomenon, termed non-integrative processing.