Hi Qiangmin, All the comments and references that were already mentioned to you are excellent,
I would stress 3 points. 1- The detergent. A clear distinction should be made between the detergent used for extraction/solubilization and the detergent (or cocktail of detergents/lipids) used for crystallization. These are two very different things. If you are lucky you may not need to change, but once you have extracted your rmembrane protein in one detergent, you should try to characterize its homogeneity by size exclusion chromatography in different detergents ( with shorter or longer chains and/or belonging to a different class (change from a choline or a phospho-glycero-lipid to an alkyloside or from a charged to an uncharged detergent etc). This scouting is tedious but is extremely informative and it can be done on analytical scales (so it does not require too much protein). If you like statistics about detergent use you can look there. *http://www.mpdb.tcd.ie/* depending on the class of membrane protein beta-barrel versus all-alpha helical etc etc you can initially concentrate your efforts on a subset of detergents. 2- The diffraction. As mentioned, starting with very poorly diffracting crystals is not uncommon (as it is for RNA crystals). My own personal experience is that you can get from 40 A resolution to the dreadful 6-4.0 A resolution barrier by tweaking the purification/extraction conditions (1/ changing detergent (shorter chain) and 2/ carefully controlling the amount of detergent present in the sample used for crystallization (to avoid or at least minimize the phase separation problem)). 3- The cryo conditions. Crystallization drops in presence of detergent are actually not as homogenous at it seems. Within the same drop you may have crystals of identical size and morphology and freeze them in the same condition and still get very different diffraction limits. When you freeze your crystals matching the detergent concentration in your cryo-condition with the 'expected' concentration in the drop can be extremely important especially with alkylosides (personal experience). Good luck, -- Pascal F. Egea, PhD Assistant Professor UCLA, David Geffen School of Medicine Department of Biological Chemistry 314 Biomedical Sciences Research Building office (310)-983-3515 lab (310)-983-3516 email pe...@mednet.ucla.edu
