We are making a GST-fusion protein for the first time. Typically these are eluted from an affinity column with 10 mM glutathione: From an article on the subject, "10 mM glutathione buffer: 50 mM Tris, 10 mM reduced glutathione, pH 8.0 (make fresh daily)"

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3584333/

For some applications the next step is removal of GST tag by cleavage with a protease (thrombin in our case). Thrombin contains several disulfide bonds. Some years ago, I looked into the issue of reducing agents and how they might inactivate thrombin. My recollection is that low concentrations of certain biochemical reducing agents could be tolerated, but high concentrations could not be. What I am having a hard time learning is whether or not 10 mM glutathione will reduce and inactivate thrombin. On the one hand, 10 mM is a decent concentration; on the other hand, glutathione might be too bulky to react with thrombin quickly. Ordinarlly I would just do a buffer exchange, but we are trying to minimize the number of steps of things like buffer exchanges, because our protein can drop out of solution under conditions that we have not been able to predict or control yet.

Does anyone know whether thrombin that is inactivated by PMSF would bind to a benzamidine affinity column? I can argue it either way.