CXMS: An Alternative to X-Ray Crystallography for Proteins

X-ray crystallography is still the most important tool for protein three-dimensional structure elucidation. While it provides high-resolution structural data, it requires a large amount of pure protein and may not allow for native conditions. Nuclear magnetic resonance spectroscopy (NMR) has also been used for high-resolution structural data but has some of the same limitations. So what is the alternative to these very important techniques?

Over the last few years, mass spectrometry has become an important tool in both structure elucidation and protein complexes. While mass spectrometry cannot give the high resolution obtained from X-ray crystallography and NMR studies, it is more tolerant to sample concentration and purity. Chemical cross-linking mass spectrometry (CXMS) has been developed to provide a low-resolution analysis, which is site-specific and managed using distance constraints. CXMS does not require pure samples, nor does it require concentrations as high as those required for X-rays or NMR. It also has a faster turnaround.^1,2 In some cases, CXMS can be performed on in vivo samples, maintaining the native state of the protein in its different forms.²

CXMS works through covalently binding a linker between two amino acids in proximity to each other. These amino acids are constrained by the length of the linker used. Digestion of the protein followed by LC-MS/MS gives information relating to the distal restraints set by the linker. This allows for further insight into protein folding and how proteins interact with each other, which in turn can be used to refine or create in silico models of the protein of interest.^1,2

Warren et al. used intramolecular cross-linking to study human cardiac variant of troponin I (hcTnl).³ Previous work showed potential intramolecular interactions of the N-terminal extension of troponin 1 with the region near the basic inhibitory peptide. Using single cysteine mutations, troponin was labeled using a heterobifunctional cross-linker benzophenone-4-maleimide, which showed cross-linking between the cysteine residue and the methionine residues at 154 and 155. The mixture was simplified using an HPLC with a C4 column to separate out the non-reacted proteins from the intra cross-linked proteins and the inter cross-linked proteins. These were subsequently digested with Arg-C, and the peptides were off-line fractionated by HPLC prior to analysis by direct infusion onto an LTQ-FTICR (Thermo Scientific) for accurate peptide and tandem mass spectrometry data.³

Novak et al. showed that using a crosslinker that reacts to lysine residues does not give as high a crosslinking yield as expected due to reactivity, distribution, and solvent accessibility of the protein residues.⁴ Novak et al. looked at using other crosslinkers that react at other amino acids, such as EDC, which reacts at carboxylic groups and the C-terminus. This type of reaction was labeled DEO (aspartate, glutamate, and C-terminal) and using ubiquitin, they showed that the DEO triad of carboxylic acid residues could also react with a nearby primary amine to form a zero length cross-link, which in turn could form a dihydrazide cross-link. Their results found three new zero cross-links and two new cross-links formed by the DEO dihydrazide in ubiquitin. Their calculations based on these new cross-links determined that the so-called zero length cross-link had a length of 5.8 Angstroms.⁴

Libraries of cross-linkers of varying length and reactive sites are available. These can involve different linker lengths, or the ends can involve homo-linkers, such as Lys/Lys, or hetero linkers, such as Lys/Arg.¹

While mass spectrometry isn’t going to replace X-ray crystallography or NMR as a means to obtain high-resolution structural information on the native protein, it is capable of giving low-resolution data on proximal distances and interactions that are inter- or intra-protein, allowing for both structural information on a protein to be obtained as well as information on protein complexes, which is not something not easily done by X-ray crystallography.

References

1. Sinz, A. (2003) ‘Chemical cross-linking and mass spectrometry for mapping three-dimensional structures of proteins and protein complexes‘, Journal of Mass Spectrometry, 38 (12), (pp. 1225-1237)

2. Singh, P., Panchaud, A., and Goodlett, D.R. (2010) ‘Chemical cross-linking and mass spectrometry as a low-resolution protein structure determination technique‘, Analytical Chemistry, 82 (7), (pp. 2636-2642)

3. Warren, C.M., Kobayashi, T., and Solaro, R.J., (2009) ‘Sites of intra- and intermolecular cross-linking of the N-terminal extension of troponin I in human cardiac whole troponin complex‘, Journal of Biological Chemistry, 284 (21), (pp. 14258-14266)

4. Novak, P. and Kruppa G.H., (2008) ‘Intra-molecular cross-linking of acidic residues for protein structure studies‘, European Journal of Mass Spectrometry, 14 (6), (pp. 355-365)