I have for several months been trying to develop an assay to measure adhesion between two populations of cells expressing either a ligand or its receptor at the cell surface. The ligand is HIV Env and the receptor is CD4. The setup is something like this:

I seed HeLa cells in a 96-well plate (10⁴ cells per well) and transfect them with HIV provirus to get Env expression. After allowing for 24-36h expression, I add Calcein-AM-labeled Jurkat cells (which natively express CD4). I co-culture the cells for 3 hours at 4ºC (Env is fusogenic so I need low temperature to prevent cell fusion and only measure adhesion). I then gently wash the wells twice, fix, and read Calcein-AM fluorescence on a plate reader. I also simultaneously have wells of the transfected cells alone (I don't add Jurkat cells) which I DAPI-stain and read to normalize for cell survival differences due to different transfections.

This sounds like it should work, right? The more Calcein-AM fluorescence I read out of a well, the more cells adhered. I measure background cell-cell adhesion by having a mock condition (transfected with Env-deficient HIV provirus), so any additional fluorescence above that level should be specific Env/CD4-dependent adhesion.

Well, it turns out I just don't see any appreciable specific adhesion. There is really no difference in the number of cells adhering whether Env is being expressed or not. Here are some things I've tried:

• I have tried different extents of washing (0-4 washes), although the overall number of labeled cells is noticeably decreased with more washing, I still see no specific adhesion.
• I have tried using anti-CD4 antibodies shown to block the Env-CD4 interaction and, although I see a slight decrease in adhesion, this happens whether Env is being expressed or not!
• I have tried adding different numbers of Jurkat cells, ranging from 0.5-2 x 10⁵ per well, still no specificity.
• I co-transfected a small amount of fluorescent protein (DsRed) and verified that my transfections are working with decent efficiency (~60-70%).
• Survival (based on DAPI readings) is very similar between Env-expressing and mock conditions.
• I have tried to visually assess adhesion instead of relying on the plate reader, and my visual estimates generally agree with the plate reader. I can also clearly see the difference between wells that have been washed twice and wells that have been washed 4 times.

I can easily fuse these same cells using the same transfection, that assay just runs at 37ºC (which permits fusion), so I know that the Env-CD4 interaction can take place between these cells. It is also published that Env-CD4-dependent adhesion is observed at 4ºC within 3 hours, but those papers are older and do not give enough detail on how they perform their cell adhesion assay for me to replicate, in principle however I am doing everything they were doing.

It seems to me that this assay should work, but I just can't figure out why I can't see any specific adhesion. Any ideas would be greatly appreciated.