Figure 3.

Increase in local cell density leads to decrease in activation, suggesting role for cell-to-cell communication. Cluster stimulation with FluidFM at different cell densities. (A) Concentration gradient during cluster stimulation for C_max = 90 nM (top) and 900 nM (bottom). Dashed line marks where concentration drops below 30 nM (top) or 100 nM (bottom), cutoffs chosen based where activation drops below 50% for that stimulation condition (Supplementary Figure 3A). (B) Cluster stimulation with FluidFM, each plot one representative experiment (Supplementary Figure 4A). Top row C_max = 90 nM, bottom C_max = 900 nM, left column sparse plating, right dense plating, all 5 min. stimulation time. Data binned in 100 μm x 100 μm squares, tile background is agonist concentration, point color is percent activated and point size is the number of cells per bin. Circular dashed line corresponds to dashed line in (A). Negative control (bottom right). (C) Fraction activated in center per independent experiment, cutoff from calibration data (Supplementary Figure 1H). Grouped by stimulation condition (Cmax) and cell density. Significance from permutation ANOVA in R (Mulder et. al, STAR Protocols, in press). (D) Whole culture activation: culture density comparison. Reporter RAW 264.7 cells treated with LPS, LTA, or R848 at the indicated concentrations for 5 minutes, agonist removed by dilution method, incubated for a further 10 minutes, then imaged. NF-κB translocation quantified. Horizontal line indicates activation cutoff from calibration data (Supplementary Figure 1H). Violin plots labelled with fraction activated, significance from Wilcoxon Test in R. Stars: * p ≤ 0.05 *** p ≤ 0.001 **** p ≤ 0.0001 ‘ns’ p > 0.05.