PSI Structural Biology Knowledgebase

PSI | Structural Biology Knowledgebase
Header Icons

Related Theme

Crystals from recalcitrant proteins

PSI-SGKB [doi:10.1038/th_psisgkb.2009.36]
Technical Highlight - August 2009
Short description: Adding small quantities of protease is the most effective rescue strategy for those proteins that just won't form well-diffracting crystals.PLoS ONE 4, e5094 (2009)

You've tried all the usual crystallization tricks—varying the protein concentration, altering salt concentrations and pH, changing the temperature, using additive screens and adding or removing amino- and carboxy-terminal tags—but still no well-diffracting crystals? Don't despair, there are several more options. Straightforward approaches include reductive methylation or the addition of specific ligands or inhibitors, but the most effective is the addition of small quantities of protease directly to the crystallization solution.

Wernimont and Edwards, working with the PSI MCSG and the Structural Genomics Consortium report a nearly 13% success rate with this approach. Their technique is reasonably simple: they purify their target protein with a hexahistidine tag and cleave it with a tobacco etch virus protease. Then they add commercially produced chymotrypsin to the crystallization trials at a starting ratio of 1:100 chymotrypsin to target protein. If chymotrypsin fails, they try trypsin, elastase, papain, subtilisin A and endoprotease Glu-C (V8 protease), at differing ratios.

They took 270 soluble proteins that had previously failed to produce suitable crystals and found that 34 yielded useful crystals after proteolysis. This set included GTPases, chromatin remodeling proteins and tyrosine kinases, demonstrating that this technique works with human proteins as well as those from yeast and prokaryotes.

For example, they had tried for 2–3 years to crystallize the human heme-binding transcription factor Rev-Erbβ and had used dozens of constructs, mutated cysteines and tried various deletions. Almost in desperation, they added a small amount of trypsin to help concentrate the protein and, amazingly, excellent crystals were obtained.

Similarly, the GTPase CGD6_3220 from Cryptosporidium parvum had failed to crystallize despite many attempts. Following treatment with trypsin, it easily crystallized. Close examination of the cleavage site showed that the intact loop would have clashed with its crystal partner and prevented that particular crystal formation.

For years, proteins have been crystallized serendipitously through protease contamination. Now you can make your own luck.

Related articles

Tips for crystallizing membrane proteins

Predicting crystallization success

Methylation boosts protein crystallization

Maria Hodges


  1. A. Wernimont, A. Edwards In situ proteolysis to generate crystals for structure determination: an update.
    PLoS ONE 4, e5094 (2009). doi:10.1371/journal.pone.0005094

Structural Biology Knowledgebase ISSN: 1758-1338
Funded by a grant from the National Institute of General Medical Sciences of the National Institutes of Health