Figure 1.
A microfluidic chip for measuring IgE-mediated FcεRI signaling events.
A) General scheme of FcεRI signaling. Crosslinking FcεRI, the high-affinity IgE receptor, through IgE-FcεRI interaction with mulitvalent antigen activates the signal-initiating kinase Lyn, creating receptor binding sites for the signal-propagating kinase Syk. Syk activation leads within minutes to the tyrosine phosphorylation of many proteins, mobilization of cellular Ca2+, the release of inflammatory mediators via degranulation and within hours to cytokine production. B) Micrograph of the monolithic microfluidic chip used in this study. The inlets used for individual cell/reagent loading, buffer washing and intracellular immunostaining are indicated in the figure. C) Selected regions for live-cell imaging. Scale bar: 20 µm. D) Hydrodynamic focusing and detection region for μflow cytometry.
Figure 2.
Analysis of phosphorylation of Syk.
Phosphorylation at Y346 in Syk (rodent numbering) was assessed at the basal condition and at 2, 5 and 30 min after stimulation with 100 ng/ml of DNP-BSA. Phosphospecific immunostaining was followed by μflow or conventional flow cytometry analysis. A) Histograms from μflow cytometry. B) Histograms from conventional flow cytometry. C) Each bar reports the average and standard deviation of median fluorescence determinations made by μflow cytometry in three independent experiments. D) Same as panel C except median fluorescence determinations were made by conventional flow cytometry.
Table 1.
Model parameters.(a)
Figure 3.
Model-based analysis of Syk phosphorylation.
A) Simulated heterogeneity in Syk phosphorylation. The simulated distribution (black line) is comparable to the experimental distribution obtained at either 2 (green line) or 30 min after DNP-BSA stimulation (cf. Figure 2A). The histogram summarizes the predicted steady-state levels of Syk phosphorylation from 103 simulation runs, meaning it represents the distribution of phosphorylated Syk across a population of 103 cells according to the model. In simulations, signaling is induced by a ligand corresponding to DNP-BSA. The copy numbers of the signaling proteins FcεRI, Lyn, and Syk were varied log normally with standard deviation about their nominal means (Table 1). B) The outermost histogram (red line) corresponds to the case where only Lyn copy number is varied, the middle histogram (blue line) corresponds to the case where only receptor copy number is varied, and the innermost histogram (cyan line) corresponds to the case where only Syk copy number is varied. In each case, while the copy number of one protein is varied the copy numbers of the other two signaling proteins are kept constant. C) Predicted effects of Lyn knockdown and Lyn overexpression on the distribution of Syk phosphorylation. The black line represents the wild type case (same as in panel A), the dark red line represents the Lyn knockdown case, and the magenta line represents the Lyn-overexpression case. In the knockdown and overexpression cases, mean copy number of Lyn is taken as 10-fold lower and 10-fold higher than the wild type case. It should be noted that the horizontal axis is logarithmic. Thus, a translation of a given distance from any point along the axis reflects the same fold change but not the same absolute change in copy number.
Figure 4.
Time course of antigen-stimulated Ca2+ mobilization in individual RBL-2H3 cells.
Mobilization was assessed in microfluidic channels using calcium indicator Fluo-3. The arrow indicates the time at which the antigen (DNP-BSA) reached the cells in the region of the microfluidic device under observation.
Table 2.
The variability of time of onset of the calcium response to antigen (DNP-BSA) stimulation.
Figure 5.
Microscopic measurement of released β-hexosaminidase in microfluidic channels.
A) Differential interference contrast (DIC) image of representative live RBL-2H3 cells stimulated with DNP-BSA with no addition of MUG, the fluorogenic substrate of β-hexosaminidase. Scale bar: 10 µm. B) Fluorescence images of live stimulated cells 20 min after addition of both DNP-BSA and MUG. The cells were immersed in supernatant containing released β-hexosaminidase and the fluorescent product methylumbelliferone. The fluorescence signal reveals the spiral geometry of the microfluidic channel. Scale bar: 20 µm. C) A bell-shaped dose-response curve characterizes the degranulation response of live RBL-2H3 cells in microfluidic channels to antigen stimulation. The concentrations of antigen (DNP-BSA) used to stimulate cells ranged from 1 to 1000 ng/ml. D) Results of conventional degranulation assays performed using 24-well tissue culture plates.
Figure 6.
Cytokine production by RBL-2H3 cells was measured by μflow cytometry and ELISA after cells were stimulated with doses of DNP-BSA ranging from 1 to 1000 ng/ml. Cells were challenged in the presence of Brefeldin A to block cytokine transport processes. A) μflow cytometric measurements of TNFα production. B) ELISA measurements of TNFα production.