PSI Structural Biology Knowledgebase

PSI | Structural Biology Knowledgebase
Header Icons

Related Articles
Automating NMR Structures
April 2015
Protein Folding and Misfolding: A TRiC-ster that Follows the Rules
March 2015
Virology: Making Sensitive Magic
March 2014
Microbiome: Solid-State NMR, Crystallized
September 2013
Membrane Proteome: Making DNA Nanotubes for NMR Structure Determination
August 2013
Protein-Nucleic Acid Interaction: Inhibition Through Allostery
July 2013
Cell-Cell Interaction: Magic Structure from Microcrystals
March 2013
Membrane Proteome: Soft Sampling
December 2012
Membrane Proteome: Specific vs. Non-specific weak interactions
November 2012
Automatic NMR
September 2012
NMR structure test
September 2012
To structure, faster
August 2012
S is for solubility
June 2012
Blind faith
April 2012
Follow the RNA leader
December 2011
Making invisible proteins visible
October 2011
A fragmented approach to membrane protein structures
September 2011
Molecular replacement by magnetic resonance
August 2011
Solutions in the solution
June 2011
No more labeled lipids
May 2011
Capsid assembly in motion
April 2011
NMR challenges current protein hydration dogma
March 2011
Solving homodimeric structures with NMR
November 2010
CASD-NMR: assessing automated structure determination by NMR
June 2010
Peptidoglycan binding: Calcium-free killing
June 2010
Removing the NMR bottleneck
April 2010
NMR has its wiki way
March 2010
Extremely salty
February 2010
The future of NMR
September 2009
Tips for crystallizing membrane proteins
June 2009
Faster solid-state NMR
May 2009
Powerful NMR
April 2009
Activating BAX
December 2008

Tips for crystallizing membrane proteins

PSI-SGKB [doi:10.1038/th_psisgkb.2009.25]
Technical Highlight - June 2009
Short description: A practical guide to cloning, expressing, purifying and crystallizing membrane proteins.Nat. Protoc. 4, 619-637 (2009)

Detailed methods for obtaining pure, soluble membrane protein samples and diffraction-quality crystals are usually published with the solved structure and are often too specific to be generally useful. A general protocol for growth of membrane protein crystals has recently been published in Nature Protocols by Newby et al. 1 in collaboration with the PSI. Here, their method is distilled into ten handy hits for successful membrane protein crystallization.

Tip 1 The best predictors of crystallization success are purity and stability. As a rule of thumb, start with a protein sample that is more than 98% pure, more than 95% homogeneous and more than 95% stable when stored unconcentrated at 4 °C for 1 week.

Tip 2 Add a cleavable amino- or carboxy-terminal tag. N-terminal fusion proteins seem to increase the amount of proteins expressed, and tags in general make purification much simpler. Useful vectors for expression are the pET vectors, which are driven by T7 RNA polymerase and are inducible by isopropyl β-D-1-thiogalactopyranoside (IPTG). An alternative is the pBAD vector system, which uses arabinose for induction. Tags to consider are poly-histidine, maltose-binding protein, glutathione-S-transferase, the PelB leader sequence and the membrane-integrating sequence for translation of integral membrane constructs (Mistic).

Tip 3 Try a variety of temperatures and media for cell growth. A reduction in the growth temperature can make the difference between an improperly folded aggregate in inclusion bodies and one that is correctly folded and inserted into the membrane.

Tip 4 Although not strictly necessary, isolating the membrane faction from Escherichia coli by centrifugation immediately after harvesting and lysing the cells can reduce proteolysis by removing unwanted proteins from the insoluble fraction.

Tip 5 The concentration of detergent used for solubilization will depend in part on its critical micellar concentration (CMC), the concentration at which detergent monomers self-associate into micelles. Concentrations much higher than the CMC are usually used for solubilization to insure that there is sufficient detergent to saturate and disrupt the lipid bilayer and allow the target protein to be extracted from the membrane.

Tip 6 Different detergents can be used for purification and crystallization, and these can be easily exchanged while the sample is immobilized on an ion-exchange column. A longer-chain detergent is often used to extract the protein from the lipid membrane and a shorter-chain one for crystallization.

Tip 7 If the affinity tag proves hard to remove, this might be either because the detergent micelle is preventing the protease from accessing the cleavage site or because the detergent is inhibiting the activity of the protease. Try placing the tag at the other terminus to allow the protease access or try adding five to ten times the normal amount of the recommended protease.

Workflow for generating membrane protein crystals.

Tip 8 Dialysis is not essential for membrane protein crystallization, but it is a useful step to insure that a minimal amount of detergent is used. Detergents dialyze to equilibrium at different rates depending on their CMC, so remember to allow sufficient time for full dialysis.

Tip 9 Reduce phase separation (the tendency for detergent to separate from the aqueous solution) as much as possible, as it can inhibit crystal formation and make results hard to reproduce. One way to do this is to reduce the amount of detergent present by using a concentrator with the largest molecular weight cut-off that will still retain the protein.

Tip 10 Start your crystallization screening with polyethylene glycol (PEG) or commercial screens based on PEG as it is overwhelmingly the most successful precipitating agent for membrane proteins.

Related articles

Predicting crystallization success

A pocket guide to GPCRs

Evolving a better-expressing GPCR

Tuning membrane protein overexpression

Maria Hodges


  1. Z. E. R. Newby et al. A general protocol for the crystallization of membrane proteins for X-ray structural investigation.
    Nat. Protoc. 4, 619-637 (2009). doi:10.1038/nprot.2009.27

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