The Whitesides Chamber implemented in the McGrath Lab has now successfully generated a steady and linear gradient

Hi all,

After nearly a month of troubleshooting the fabrication of Whitesides Chamber we finally have the capability to generate a steady and linear gradient of molecules.  In the presented case, we have generated a gradient of rhodamine 6G (R6G), which is a fluorophore that allows us to visualize and confirm the existence of the gradient.  A full-view of the Whitesides Chamber is given in Figure 1 below:

Figure 1. Full-view of the Whitesides Chamber.  The three tubings to the left are connected to three 2 mL eppendorf tubes (each contain 500 uM, 250 uM, and 0 uM of R6G, not shown).  The tubing to the right is connected to a syringe pump (not shown).  As the syringe pump draws fluid through the main channel (the stem part of the “Christmas tree”), R6G from each eppendorf tubes are mixed and split repeatedly to produce a total of 9 different concentrations, which finally merge at the main channel to form a gradient.

Figure 2. (Left) An image of the main channel where the gradient develop, viewed under a 4X objective.  The red, highlighted rectangular region represents a line-scan over which fluorescence is measured. The “halo” to the left is the reflection of the fluorescence light source off the tubing that connects the Whitesides Chamber to a syringe pump.  (Middle) A zoomed-in image of the line-scan.  Technically what we have is a “rectangular-scan”.  The fluorescence along the width is averaged to reduce white noise.  The averaged values along the length of the rectangle is essentially a line-scan.  (Right) The line-scan of the gradient, presented in the form of fluorescence intensities versus pixels.

Note that fluorescence intensities are not very linear near the far left and far right of Figure 2 (Left).  It is most likely caused by the presence of the edges of the main channel and has been seen in the work of others (who also try to generate steady and linear gradient).

We took the middle region of Figure 2 (Right) where the gradient is most linear and further away from the channel walls and replotted it to obtain Figure 3 below.  We also converted the fluorescence intensity and pixel into the respective concentration of R6G and distance in micron with a rough calibration.

Figure 3. The linear gradient produced with the Whitesides Chamber.  The green dots represent measured values obtained from the line-scan.  The blue line represents the best-fit curve obtained from linear regression.

We performed line-scan on the same region for 5 min at a rate of one frame every 15 sec to produce the following concentration profile (Figure 4).  As can be seen the profile does not change very much with time.

Figure 4.  Spatiotemporal profile of R6G concentration.

To further confirm that the gradient produced is steady (does not change significantly over time), we performed linear regression on every line-scan and plotted the obtained slopes (Figure 5).  The difference between the maximum and the minimum of the slopes obtained is only 6.14% that of the maximum, and the coefficient of variation (standard deviation/mean) of the 20 slopes is 1.55%.  The minimum coefficient of determination (R^2) of the 20 fitted curves is 0.9908, further emphasizing that our gradient is linear.

Figure 5. Fluctuations of the gradient over 5 min.