Mapping of Cysteine Knot and Disulfide Bonds in Recombinantly Expressed Proteins by Mass Spectrometry

A cysteine knot is a triplet of disulfide bonds (six cysteine residues total) where one of the disulfide bonds passes through the ring formed by the other pair of disulfides. This is a structural motif found in several proteins in a wide range of cellular functions and is known to provide stability to the protein’s overall structure.^1,2 Understanding these structures and how they affect protein structure and function is important. Several diseases are associated with the malfunction of cysteine knot proteins, including metachromatic leukodystrophy (MLD). MLD is caused by a mutation in the gene encoding arylsulfatase A (rhASA), resulting in a cysteine-to-arginine amino acid substitution and blocking cysteine knot formation.² Replacement of the enzyme is a possible therapeutic for patients suffering from MLD, but in order for any molecular mimic to be effective, an accurate mapping and understanding of the structure of the cysteine knot is required.

There are several challenges associated with the study of this motif. First, the assignment of the disulfide pairing has to be done correctly to understand the knot. In the simplest case of six cysteine residues, there is a total of 15 different sets of disulfide bonds that can result in a cysteine knot, but only one is likely physiologically relevant. Second, the proteins that have this motif share almost no sequence homology, so the initial identification of proteins likely to contain this motif is difficult.^2,3 Third, disulfide bonds are sensitive to a number of chemical insults that may occur in the processing of samples for different analyses, including changes in pH and changes in reduction/oxidation conditions of the sample. Disulfide “scrambling” can occur where random disulfide bonds that do not reflect the actual physiological state of the protein, can form in a protein in solution.^3,4 Current experimental approaches to mapping and understanding the underlying structure of cysteine knots has relied on X-ray crystallography and nuclear magnetic resonance spectroscopy, both of which can be time consuming and don’t address the aforementioned problems of scrambling and sample handling.³ Recently, progress has been made using mass spectrometry technologies in the mapping and identification of disulfide bonds and cysteine knot motifs.^3,4

In Ni et al.,³ a variety of digestions and fragmentation techniques were used on recombinant human arylsulfatase A to determine if the disulfide bond pattern resulted in the physiologically determined knot structure required for function when expressed and purified recombinantly. Understanding the disulfide and knot formation in rhASA is required for the manufacture of an effective biopharmaceutical in order for the drug to maintain proper function. Arylsulfatase provides interesting challenges since in addition to the knot motif, there are nested sulfide linkages and cysteine residues not found in disulfide bonds. Conditions were optimized for pH for each of the cysteine residues in rhASA, and the status of each of the disulfide bonds and cysteine residues was successfully determined. Using an LTQ-Orbitrap ETD XL mass spectrometer (Thermo Scientific), the researchers were able to map the disulfide linkages in the rhASA. The LTQ provided the researchers with multiple MS/MS fragmentations for assignment between CID and ETD. Coupled with the use of a variety of digestion and sample treatments (PNGase to remove glycosylations and pepsin/Lys-C/trypsin/Asp-N for peptide generation), researchers were able to differentiate between free cysteines, modified cysteines and peptides with one, two, or three disulfide linkages.³ The use of pepsin (digestion at pH 2 where disulfide scrambling is unlikely) allowed for the identification of the unpaired cysteine residues in the protein and was able to show that cysteine 51 was modified to formylglycine, a known modification for some cysteine residues.

The mapping of the cysteine knot in the rhASA was possible through the use of ETD on the pepsin digestion of the protein. The knot is made up of cysteine 470/482, cysteine 471/484, and cysteine 475/481. The ETD fragmentation from both the 4+ and 5+ charge state showed that the cysteine linkage forming the knot was correctly conserved in the rhASA protein. In fact, the use of the high-resolution mass spectrometer (LTQ-Orbitrap) provided even more convincing evidence of the knot motif and assignment than previously shown.³

Cysteine knot-containing proteins are involved in a variety of cellular processes and are associated with various disease states in humans.² Previously, the study and correct mapping of this knot motif has been laborious and difficult to confirm experimentally. The use of high-resolution mass spectrometry, the thoughtful handling of samples, and the use of a variety of digestion techniques allows for the relatively quick and accurate identification and characterization of disulfide bonds in proteins and greater structural motifs, such as knots. This advance will allow for the generation of more effective pharmaceuticals and aid in the identification of additional cysteine knot proteins.

References

1. Alvarez, E., Cahoreau, C., and Combarnous, Y. (2009) ‘Comparative structure analyses of cystine knot-containing molecules with eight aminoacyl ring including glycoprotein hormones (GPH) alpha and beta subunits and GPH-related A2 (GPA2) and B5 (GPB5) molecules‘, Reproductive Biology and Endocrinology, 7, (p. 90)

2. Kolmar, H. (2009) ‘Biological diversity and therapeutic potential of natural and engineered cystine knot miniproteins‘, Current Opinions in Pharmacology, 9 (5), (pp. 608-614)

3. Ni, W., et al. (2013) ‘Complete mapping of a cystine knot and nested disulfides of recombinant human arylsulfatase A by multi-enzyme digestion and LC-MS analysis using CID and ETD‘, Journal of the American Society for Mass Spectrometry, 24 (1), (pp. 125-133)

4. Wang, Y., et al. (2011) ‘Characterization and comparison of disulfide linkages and scrambling patterns in therapeutic monoclonal antibodies: using LC-MS with electron transfer dissociation‘, Analytical Chemistry, 83 (8), (pp. 3133-3140)