Fluorescein transport: O-ring position and wafer variability

After diffusing through a polymeric or pnc-Si membrane in the transwell format, diffusing molecules meet an “unstirrred water layer = UWL” on the bottom of the membrane.  The UWL acts as another resistance to transport into the receiver compartment.  According to some PAMPA papers, the UWL can be greater than 1 mm in thickness without stirring.  I hypothesized that the disk of volume between the bottom of the pnc-Si chip and the bottom of the Sepcon housing (due to the O-ring and plastic) might allow the formation of a “void volume” of little mixing with the receiver bulk volume in addition to the UWL.

To test this hypothesis, I assembled Sepcons by adding the O-ring first and then pnc-Si (O-ring bottom) or by adding pnc-Si and then the O-ring (pnc-Si bottom).  The “pnc-Si bottom” set-up eliminates the 1mm spacing due to the O-ring.  I used SC065 chips without pinholes and stirred the samples on the tabletop VWR orbital shaker at 100 rpm.

Experimental details: Vd = 100uL; Vr = 1000uL; Cd = 0.00075%; gain = 0.000125%; 10×10 circle, 1000um border

To get this data, I subtracted the fluorescence from a blocked pnc-Si transwell, i.e., one that I coated with nailpolish to prevent Na-F transport.  This graph shows that there is very little difference between the 2 O-ring configurations, surprisingly.  In fact, the set-up which I though would be less ideal (O-ring bottom) showed slightly more transport.  I believe Christine has some data to back this up…

After I did this experiment and plotted out the data, I realized the data from SC065 didn’t match what I remembered from an earlier experiment with SC048.  Here, I plotted the Na-F transport through SC048 and SC065 – both without pinholes.  This data is also reduced by the “blocked” pnc-Si transwell fluorescence.

Actually, the data doesn’t disagree as much as I expected.  I think it might be useful to do some pore processing of these wafers to see if porosity/pore size might explain this difference.  Over the next couple of weeks, we’ll have a bunch more transport data for pinhole-free membranes (hopefully), so these 2 curves might not be explained by membrane variability.  Rather, this might be due to experimental variability (Sepcon assembly, Sepcon geometry in the 24-well, plastic differences, hands).