Analysis of noncovalent and covalent protein-ligand complexes by electrospray ionisation mass spectrometry

• Main Author: Sundqvist, Gustav
• Format: Doctoral Thesis
• Language: English
• Published: KTH, Glykovetenskap 2008
• Subjects: noncovalent complex; droplet fission; nano-electrospray ionisation; mass spectrometry; Analytical chemistry; Analytisk kemi
• Online Access: http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-4728; http://nbn-resolving.de/urn:isbn:978-91-7178-976-1

In this thesis, the application of electrospray ionisation mass spectrometry (ESI-MS) to the analysis of intact proteins is demonstrated. In papers I and II, the use of ESI-MS for the analysis of noncovalent protein-ligand complexes were discussed. In addition, the interfacing of liquid chromatography (LC) with ESI-MS and the development of an LC-ESI-MS method were demonstrated in paper III for the quality control of recombinant proteins. Furthermore, this method was applied in paper IV for the analysis of covalent glycosyl-enzyme intermediates. The monitoring of noncovalent complexes by ESI-MS is well established. However, the varying characteristic of ESI-MS data, especially in the analysis of noncovalent complexes can make the quantification of such complexes troublesome. In paper I, it was demonstrated how the variation in the position of the ESI-emitter and the initial droplet size of the electrosprayed droplets, together with different partitioning of a protein and its ligand in these droplets, can be the cause of such varying characteristics. Furthermore, it was shown that the partitioning can be of electrostatic and/or hydrophobic/hydrophilic origin. Thus it was demonstrated that if the ligand is more hydrophobic and thereby more surface active relative to the protein, decreasing the droplet size or increasing the distance between the electrospray emitter and the sampling orifice will lead to more efficient sampling of the droplet bulk where the ligand concentration is low. This results in a favoured sampling of free protein relative to the protein ligand complex. The opposite was shown to occur if the ligand is more hydrophilic than the protein. In paper II, Ribonuclease A (RNAse) was used as a model for enzymes acting on polymeric substrates with different chain lengths. Nano-ESI-MS was applied to monitor the noncovalent interactions between RNAse and different target ligands. Among the single building blocks of RNA, including ribose, the bases adenine, guanine, cytosine and uracil, and phosphate, only phosphate was observed to interact at multiple RNAse sites at a higher cone voltage. Furthermore, monobasic singlestranded deoxycytidylic acid oligomers (dCx) of different lengths (X=6, 9 and 12), and RNAse were analysed with nano-ESI-MS. The deoxycytidylic acid with 12 nucleotides was observed with the highest complex to free protein ratio, hence indicating the strongest interaction. Finally, collision induced dissociation of the noncovalent RNAseA-dC6 complex resulted in dissociation of covalently bound cytosine from the nucleotide backbone rather than break up of the noncovalent complex, illustrating the cooperative effect of multiple noncovalent interactions. In paper III an LC-ESI-MS method was presented capable of analysing proteins 10-100 kDa in size, from salt-containing liquid samples. The proteins included human protein fragments for the largescale production of antibodies and human protein targets for structural determination, expressed in E. coli. Also, glycosylated proteins expressed in Pichia pastoris were analysed. The method provides fast chromatography, is robust and makes use of cheap desalting/trap columns. In addition it was used with optimised reduction and alkylation protocols in order to minimize protein aggregation of denatured and incorrectly folded proteins containing cysteins, which otherwise form adducts by disulfide bond formation. Furthermore, the method was used in paper IV for the quantification of covalent proteinligand intermediates formed enzymatically between PttXET16-34, a xyloglucan endo-transglycosylase (XET) from hybrid aspen, and the synthetic substrates GalGXXXGGG and GalXXXGXXXG designed in order to function as donor substrates only. Thus covalent GalG-enzyme and GalGXXXG-enzyme complexes were detected. Moreover, establishing of a pseudo equilibrium for the formation of the covalent GalGXXXG-enzyme complex enabled quantification of the saccharide and enzyme constituents of this equilibrium and determination of the free energy of formation (∆G0). The high mass resolution of the TOF-MS allowed unambiguous assessment of the covalent nature of the glycosyl-enzyme complexes. Morover, the formation of noncovalent complexes between excess substrate and protein, which can deteriorate MS-signal and increase spectrum complexity, was efficiently avoided by the chromatographic step, which separated the saccharide content from the protein content. === QC 20100913