I am mainly working on SARM from human and mouse in the past year. I tried to express SARM in E.coli and mammalian cells (HEK293) but both systems produced aggregated proteins. I finally expressed SARM in insect cells through baculovirus expression system and it worked. I designed 6 human constructs and 7 mouse constructs of SARM. All constructs have an N-terminal MBP-His6 tag. The SARM constructs were expressed either in 3L flasks or through a bioreactor. The expression condition is 21C and 72 hours with Multiplicity of infection (MOI) 3. The cell pellets were harvested and sonicated. The proteins were purified through three steps: Talon resin (binding to His6 tag), Amylose resin (binding to MBP tag), and size exclusion column. Four of constructs were purified and confirmed as non-aggregated proteins. Human TIR domain (X06) and mouse TIR domain (X13) were purified as monomeric proteins. The construct X05 from human was a monomer but it has a degradation issue. Full-length human SARM (X01) was also purified but it formed non-aggregated oligomers. The SARM constructs were unstable at room temperature while they were stable at low temperature. I watched heavy aggregation and degradation at 20C in three days for all purified constructs. But I did not find aggregation and degradation at 10C in 7 days and at 4C in 10 days. The N-terminal MBP-His6 tag of SARM constructs could be removed by TEV protease and produced single-band proteins. However, the cleaving efficiency was very low and most of cleaved proteins precipitated during the reaction. Without MBP tag, the solubility of SARM constructs significantly reduced. For example, human TIR domain (X06) could be concentrated 60mg/ml but it only reached 2mg/ml after the MBP tag was removed. In addition, I performed detergent screening to reduce the size of full-length human SARM (X01) oligomer. I found some detergents could reduce molecular weight of X01 from 800kDa to 190kDa and it was probably a trimer. Three SARM constructs X01, X06 and X13 were performed crystallization screening. SARM X06 and X13 crystals could be obtained from multiple conditions and the best crystals were up to 0.1x0.025mm. The crystals were confirmed they were proteins. However, the diffraction pattern was not satisfactory and the low resolution spots were missing. I am continuing to optimize crystallization to get crystals with better diffraction. In addition, I cloned, expressed and purified human TRIF constructs. Full-length TRIF was degraded and only some N-terminal residues were available. The TIR domain from human TRIF was purified as a hexamer and this agrees with previous publication. TRIF should be oligomerized to be functional.
|Botos, Istvan; Segal, David M; Davies, David R (2011) The structural biology of Toll-like receptors. Structure 19:447-59|