Expression and Purification: Tips and Tricks

How pure should be the protein for crystallization?

Everyone knows that protein must be "very pure" for crystallization. It is expected that no extra bands will be seen on an overloaded gel. While this stringent requirement may benefit you when growing "x-ray quality crystals" for the purposes of data collection, the protein sample used for first crystallization trials may be quite dirty. In fact, it is good to have some contamination, because it may promote crystallization. This is because the purpose of first crystallization trials is not to obtain perfect crystals (although that would be lovely) but rather find the leading conditions which can be further optimized.

In practice, it is a good idea to take your protein after the first purification step (such as metal affinity column if you have his-tag) and give it a go with crystallization screens. Protein crystals often take two-three weeks to appear and you can use this time to optimize your purification protocol. Of course, it is important to realize that in some situations only final purified product will crystallize, so if preliminary trials with "not-so-pure" protein failed, you should consider running the same screens second time with highly purified batch.