Abstract

Chemotaxis, the ability of cells to detect and migrate directly towards a source of a chemically active agent, is the result of a sophisticated interplay of proteins within a complex regulatory network. However, partially redundant pathways that simultaneously mediate chemotaxis and dynamic protein distributions complicate the experimental identication of distinct signaling cascades and their inuence on chemotactic migration. Yet, increasingly precise generation and rapid modication of chemotactic stimuliin microuidic devices promise further insight into the basic principles of cellular feedback signaling. I developed a Chemotactic Gradient Generator (CGG) for the exposure of living cells to chemotactic gradient elds with alternating gradient direction based on a double T-junction microuidic chamber. A large extension of the concentration gradients enables the parallel exposure of several dozens of cells to identical chemotactic stimuli, allowing for a reliable quantitative analysis of the chemotactic migration behavior. Two pressure pumps and a syringe pump facilitate accurate control of the inow velocities at the individual ow chamber inlets, pivotal for precise manipulation of the chemotactic stimuli. The CGG combines homogeneous gradients over a width of up to 300 µm and rapid alterations of gradient direction with switching frequencies up to 0.7 Hz. Fast gradient switching in our experimental design facilitates cell stimulation at the intrinsic time scales of their chemotactic response as demonstrated by a gradual increase in the switching frequency of the gradient direction. We eventually observe a "chemotactically trapped" state of Dictyostelium discoideum (D. discoideum) cells at a switching rate of 0.01 Hz. Here, gradient switching proves too fast for the cells to respond to the altered gradient direction by migration. In contrast, we observe oscillatory runs at switching frequencies of less than 0.02 Hz. We distinguish between re-polymerizing cells that exhibit an internal re-organization of the actin cortex in response to chemotactic stimulation and stably polarized cells that gradually adjust their leading edge when the gradient is switched. To experimentally characterize both response types, we record cell shape and the intracellular distribution of actin polymerization activity. Cell shape is readily described by the eccentricity of the cell and to record F-actin polymerization dynamics we introduce a fluorescence distribution moment (FDM). Accurate description of the migratory response behavior facilitates a quantitative analysis of the inuence of both the experimental boundary conditions such as gradient shape, ongoing starvation of the cells, and in particular the inuence of distinct signaling cascades on chemotactic migration. Here, we demonstrate this ability of the GCC by inhibition of PI3-Kinase with LY 294002. PI3-Kinase initiates the formation of fresh pseudopods in the direction of the chemotactic gradient and therefore is one of the key signaling pathways mediating the chemotactic response. In shallow gradients and with ongoing starvation of the cells, we find a decreased ratio of re-polymerizing cells, pointing towards a diminished influence of PI3-Kinase. After inhibition of PI3-Kinase, cell re-polymerization in response to a switch in gradient direction is hindered at 5h of starvation, whereas at 7h of starvation evidence is found that chemotactic migration is more efficient. We observe the astonishing result that in dependency of the boundary conditions of the experiment inhibition of PI3-Kinase promotes an effective chemotactic response. Thus, the CGG for the rst time facilitates a quantitative analysis of the starvation time dependent effect of PI3-Kinase inhibition on D. discoideum chemotaxis.