Protein Crystallography Course

• Is it pure?
• Purity is the most important prerequisite for crystallizability
• Run an overloaded gel to see contaminants present at 1% of the total protein
• Purity also means that there is no heterogeneity in posttranslational modifications
• If you are crystallising a protein-protein complex that will stay together at concentrations of 1mg/ml, purify the complex from uncomplexed protein before setting up trials
• Is it folded?
• Check the activity of your protein if you have an assay
• Check the CD spectrum of your protein to see if it has secondary structure
• Is it fresh?
• Proteins break down with time and the mixture becomes heterogeneous
• Always try to set up your trials as soon as possible, preferably the same day as the last step of the purification
• Is it monodispersed?
• Monodispersity means that the protein exists in solution as a single oligomeric species eg. monomer or dimer
• It means the protein is free of non-specific oligomers and aggregates
• Use a size-exclusion column as the last step in the purification
• Use the dynamic light scatterer to see if your protein is monodispersed
• (Set up trials anyway)

If you are getting nowhere with your protein, try crystallizing something else.

• Ligand-protein complex. Does your protein bind a ligand? If it does, it can help because the binding of a ligand
• is likely to order the region of the protein that binds ligand
• may bring two subdomains together and reduce flexibility>
• will change the surface properties of the protein
• may cause a conformational change in your protein
• Different constructs. Heterogeneous tertiary/quaternary structure can hinder crystallization
• Does your protein break down to a stable proteolytic fragment (either "spontaneously" or with the help of an added protease)?
• Does the homology of your protein to others in the same family drop off at the N- and C-termini?
• Does your protein have a domain structure?
• Check the Pfam database
• Check the ProDom database
• Do a PsiBlast. search
• Does your protein have low complexity regions?
• Different species. Sometimes, a point mutation is all that is required to stop/start a protein crystallizing. Working on a different species is the easiest way to get a collection of point mutations that do not affect function.
• Deglycosylation. If your protein is glycosylated the floppy and heterogeneous carbohydrates may be interfering with the crystallisation. Try enzymatic deglycoylation.

© 1999-2005 Airlie J McCoy, University of Cambridge. All rights reserved.

Last updated: 7 June, 2005