All the IP protocols I've seen start with making a lysate using a buffer with mild non-ionic detergents, etc. which makes sense if you're trying to get cytoplasmic proteins into solution, and then have them interact with a specific antibody to facilitate their subsequent recovery.
I'm interested in ECM proteins, and their specific interactions with each other and membrane proteins on the cells. We've got an epitope tagged membrane protein that we can express in zebrafish embryos, and I'd really like to try co-immunoprecipitation (coIP) to see if we can pull down ECM molecules it is interacting with.
However, matrix proteins are notoriously insoluble, and the vast majority of our putative target molecules will not be solubilized by something like conventional RIPA buffer. But if we use harsher extraction conditions (chaotropes, reducing agents, etc), we will denature the target molecules and disrupt any interactions. Furthermore, we would obviously have to remove these denaturants before adding our antibody, or the antibody would be denatured as well.
Has anyone had any success or even heard of a good immunoprecipitation protocol for extracellular matrix proteins?
I don't think co-IP is an appropriate assay to address this question for the reasons stated in your post. ECM molecules have evolved to function in an adsorbed/insoluble physical state and they are intrinsically sticky/aggregation-prone making it exceedingly difficult to interpret specificity in protein-protein interactions such as co-IP. Moreover, if you are looking for interactions with a membrane-bound receptor, then addition of detergents to your assay could alter the activity of your membrane-bound protein in undesirable ways, adding another layer of complexity.
Classically, the way this problem has been solved has been to map bioactive regions of the ECM protein (i.e. small proteolytic fragments that mediate or antagonize cell-matrix adhesion), and then affix those fragments as baits in an affinity chromatography assay to pull out proteins from fractionated membranes.
A more modern approach would be to use cross-linking mass-spectrometry to try and map interacting peptides between your receptor of interest and potential ECM ligands. To do this, you would need to be able to solubilize the epitope tagged extracellular region of your receptor of interest using limited proteolysis.
Thanks for your response Daniel. We're considering various cross linking approaches and epitope-tagged expression approaches. But part of the problem here is that we're not using a tissue culture system in which expressing variously engineered constructs and/or obtaining loads of material suitable for affinity chromatography is trivial.
If the interactions of interest to us are occurring between our molecules of interest and molecules that are effectively insoluble and/or tightly bound to insoluble ECM that is not represented in our clarified homogenates, they're going to be very difficult to detect.
I suppose I'm probably asking the impossible, but I was hoping someone might know of a good approach for solubilizing ECM proteins in such a way that they could be subsequently re-natured enough to interact meaningfully.
Cheers
That is a tall order. You might get some solubilization with limited proteolysis, especially using a mix of collagenases and MMPs, although I don't know how well those work across species (you would probably have difficulty getting zebrafish-sourced enzymes). The only efficient ECM solubilization protocol I've seen involves harsh denaturation (5M GuHCl plus 10mM DTT), and so wouldn't be amenable for your assay.
Another strategy that could be worth exploring if you have candidate ECM-receptor pairs and suitable antibodies is proximity ligation assays to look at co-localization in situ in your embryos. That wouldn't prove direct binding and would involve advance knowledge of molecules of interest, but positive results might help suggest co-assembling proteins and therefore increase your chances of finding biochemical interactions.
Ultimately, though, I think you'd need to be able to recombinantly express a soluble bioactive fragment of your ECM protein of interest to have a decent shot at characterizing receptor binding and specificity.
Thanks Daniel; we're already working on the PLAs, so I completely agree with you're thinking... it's just a tough slog and I was hoping there might be a short cut :)
Cheers
If you do try the denaturation/renaturation route, one thing going for you is that many of the binding interactions are with linear peptides, so it is more about surface accessibility of the ligand than proper protein folding (at least extrapolating on the integrin-RGD paradigm).
In theory, you could also consider the protein binding microarray route. Genentech has applied it to receptor-ECM questions (Article A secreted protein microarray platform for extracellular pro...
), but to the best of my knowledge the approach hasn't gained a major foothold in the cell adhesion field.- Badges
- Science topic
- Similar topics
- Biological Science
- Immunology
- Antibodies