Dear Wenhe, this sounds a bit like our struggles to solubly express and purify the dsRNA-binding avian reovirus sigmaA protein, please find the two papers here: http://www.ncbi.nlm.nih.gov/pubmed/?term=van+Raaij+MJ%2C+sigmaA The key to solving our problem was to realise the protein aggregated especially at 4 degrees, but was more stable at room temp., i.e. it appeared to be cold-sensitive. So we did all functional assays, including gel-shift assays, with protein always kept at room temp. However, it was finally crystallised at 5 degrees. MBP is "notorious" for pulling partially-folded proteins into solution, which then aggregate or precipitate upon cleavage. Although your protein is active in gel-shift assay, it may still be partially folded I guess. Perhaps you can isolate a well-folded population by using a nucleic-acid column? Perhaps some mild unfolding-refolding?
Greetings, Mark Mark J van Raaij Lab 20B Dpto de Estructura de Macromoleculas Centro Nacional de Biotecnologia - CSIC c/Darwin 3 E-28049 Madrid, Spain tel. (+34) 91 585 4616 http://www.cnb.csic.es/~mjvanraaij On 6 Mar 2014, at 08:12, WENHE ZHONG wrote: > Dear CCP4 friends, > > I have been searching around the CCP4 mails for the discussion of soluble > protein aggregations, as I have the same problem recently. Many efforts I > have tried but without success. Here, I would like to summarise what I have > done and maybe any of you come across some ideas. > > 1. My protein (not membrane protein) is around 35kDa with a pI of 8.6. It was > expressed in E.coli as a fusion protein where its N-ter was fused by MBP-tag > with a flexible linker (Factor Xa cleavage site available). The whole fusion > protein is ~77kDa (pI 6.1). > 2. After MBPTrap affinity purification, the protein was run through > gel-filtration column it came out at void volume. The aggregation was also > confirmed by DLS. The buffer condition is: 20 mM HEPES, pH7.4, 125 mM KCl, > 20% glycerol, 40 mM maltose, 5 mM DTT. > 3. I made several truncations and only one gave me a monomer (The other > truncations were all soluble aggregates). However, this monomeric truncation > is small (~5 kDa) and not important for the function (not interesting > fragment). At least, this monomeric truncation suggests that the aggregation > is not due to the aggregation of MBP. The truncations were designed in both > two versions: long linker (including TEV cleavage site) and short linker > (-Ser-Ser-Ser-). > 4. Hampton detergent screens and high salt (1-2 M KCl or NaCl) were tried on > both truncations and full-length protein, but none of them work (DLS > confirm). Divalents and other ions should not bind to this protein. > 5. The full-length aggregated protein is still functional in Gel-shift assay, > which indicates the protein is corrected folding. > > I guess "soluble aggregation" is a common problem for tough proteins. If > anyone has experience on it and can share with us, please drop an email. > > Kind regards, > Wenhe
