I’m not sure how you could implement this idea with a cell-free medium. I haven’t been able to think of anything (that could work) so far, but here’s an idea using a medium containing bacterial cells. What if we used the Beta-galactosidase assay (wiki has a pretty good summary of what this is, for those that are unfamiliar with it) such that we used ONPG as the substrate? I think ONPG can’t on its own get inside bacterial cells to be in contact with the b-gal enzyme. So, if we could somehow make the allergen cause lysis of the bacteria, b-gal would be exposed and could act on ONPG to cause yellowing of the solution. In the absence of allergen, the solution will be colourless. From reading, it seems to be that the ideal temperature for the enzymatic reaction is 28°C, which is a fair approximation of room temperature, but I’m not sure how accurately this needs to be maintained.
In terms of reaction kinetics, it seems like we could get a strong enough yellow colour within a few minutes if we have a high concentration of cells. We should perhaps use IPTG (at 5 uL it acts as a strong inducer of lacZ gene expression) or insert genes to up-regulate the transcription of beta-gal to ensure a fast reaction.
One of the things that worries me a little is: what happens when the bacterial cells die naturally? Hopefully, b-gal will disintegrate too so that we don’t get false positives. If not, I suppose we should genetically engineer this connection.(?) In any case, we should probably have a negative control too.
So the BIG question is how do we get an allergen to lyse bacteria? I was doing some research and found that there exists a peptidoglycan hydrolase (so this would lyse the cell wall) and a lipase that could destroy the cell membrane and together they can actually be expressed as a functioning fusion protein. I’m not sure about the details of how we could make the allergen control the activity of this fusion protein (especially since we cannot rely on gene expression after the addition of the allergen since transcription and translation would take too long). Protein-protein interactions will be our best bet. Any ideas?