Interactions between proteins in vivo are transient. Furthermore, most proteins exist as complex assemblies of different subunits that act as a single unit when interacting with other proteins. Mutations that underly disease can disrupt the formation of essential protein complexes. However, the importance of these mutations remains obscure while the structure of most protein complexes remains unknown.
The search for structures of protein complexes has eluded investigation with traditional methods in structural biology, such as NMR or crystallography. Chemical cross-linking can freeze these transient interactions in place. However, robustly detecting these interactions across the proteome is a daunting challenge. Despite many technological advances, proteome-wide screens typically detect only a fraction of possible cross-link contacts.
The results of these experiments provide researchers with an array of protein side-chains that existed close enough in vivo to be chemically linked. The information provided is equivalent to knowing which pieces of a puzzle touch. With enough binary interactions between protein side-chains, researchers can constrain a computational model of a protein complex to provide a reliable answer. However, detecting a large number of these interactions in a single experiment is difficult.
Researchers led by Alexander Leitner in Ruedi Aebersold’s lab at the Swiss Federal Institute of Technology Zurich have developed and improved method expanding the detection of cross-linked proteins. Following chemical cross-linking, protein isolates were digested using multiple proteases, including Lys-C, Lys-N, Glu-C, and Asp-N, in addition to trypsin, previously used alone.1
The combined digest ensures that cross-linked peptides are uniformly large and unlinked peptides are uniformly small. Digests with a single protease are much more susceptible to variability in the sequences of individual proteins.
Digested peptides were separated using size-exclusion chromatography to enrich cross-linked peptides in high-mass fractions, isolating cross-linked from unlinked peptides further based on size. The purified high-mass peptide fractions were separated and sequenced via liquid chromatography-tandem mass spectrometry (LC-MS/MS) using a LTQ Orbitrap XL Mass Spectrometer (Thermo Scientific). Through LC-MS/MS, researchers were able to identify cross-linked fragments in an automated method.
This combined method significantly increased the number of cross-linked sites that were detectable. The enrichment of cross-linked peptides via size-exclusion chromatography increased the number of detected cross-link sites 3-fold. Digestion using multiple proteases further increased cross-link detection by 70 percent. Results were verified across several model systems including a mixture of model proteins, the rabbit 20S proteasome, and the S. pombe 26S proteasome.
A complement and not a replacement to traditional methods, cross-linked interaction data sets can reveal structure of protein complexes that may not be possible to analyze with other methods, such as X-ray diffraction. The limitation of cross-linking methods is the ability of researchers to separate the mixture of cross-linked and unlinked peptides.
Researchers are combining methods across fields to both improve the isolation of cross-linked peptides and use the data alongside traditional structural methods. For example, the use of size-exclusion chromatography along with chemical cross-linking has previously been used to aid in the determination of a sub-nanometer structure of the S. pombe 26S proteasome.2
The use of combined biochemical, proteomic and structural methods is needed to continue the study of protein complexes. The immediate application is not limited to basic biology in simplified model systems. The ability to perform many of these analyses in automated or semi-automated methods provides the opportunity for a great expansion of the role of protein structure and interaction in normal biology and disease pathology.3
1. Leitner, A., et al. (2012) ‘Expanding the chemical cross-linking toolbox by the use of multiple proteases and enrichment by size exclusion chromatography‘, Molecular and Cellular Proteomics, 11 (3), (p. M111.014126)
2. Bohn, S., et al. (2010) ‘Structure of the 26S Proteasome from Schizosaccharomyces pombe at Subnanometer Resolution‘, PNAS, 107 (49), (p. 20992-20997)
3. Stengel, F., et al. (2012) ‘Joining forces: Integrating proteomics and cross-linking with the mass spectrometry of intact complexes‘, Molecular and Cellular Proteomics, 11 (3), (p. R111.014027)