Localization of O-Glycosylation Sites in Peptides by Electron Capture

Sep 15, 1999 - Localization of O-Glycosylation Sites in Peptides by Electron Capture Dissociation in a Fourier Transform Mass Spectrometer .... Data-D...
0 downloads 4 Views 100KB Size
Anal. Chem. 1999, 71, 4431-4436

Accelerated Articles

Localization of O-Glycosylation Sites in Peptides by Electron Capture Dissociation in a Fourier Transform Mass Spectrometer E. Mirgorodskaya,† P. Roepstorff,† and R. A. Zubarev*,‡

Department of Molecular Biology and Department of Chemistry, University of Southern Denmark/Odense University, Campusvej 55, Odense M, DK-5230, Denmark

The novel technique electron capture dissociation (ECD) of electrospray generated [M + nH]n+ polypeptide cations produces rapid cleavage of the backbone NH-Ca bond to form c and z‚ ions (in the modified notation of Roepstorff and Fohlman). The potential of the Fourier transform mass spectrometry equipped with ECD in structure analysis of O-glycosylated peptides in the 3 kDa range has been investigated. Totally, 85% of the available interresidue bonds were cleaved in five glycopeptides; more stable c ions accounted for 62% of the observed fragmentation. The c series provided direct evidence on the glycosylation sites in every case studied, with no glycan (GalNAc and dimannose) losses observed from these species. Less stable z‚ ions supported the glycan site assignment, with minor glycan detachments. These losses, as well as the observed formation of even-electron z ions, are attributed to radical-site-initiated reactions. In favorable cases, complete sequence and glycan position information is obtained from a single-scan spectrum. The “mild” character of ECD supports the previously proposed nonergodic (cleavage prior to energy randomization) mechanism, and the low internal energy increment of fragments.

Glycosylation, one of the most common covalent posttranslation modifications, poses a significant challenge to a researcher. Since the development of ESI and MALDI ionization techniques, mass spectrometry proved to be a powerful tool in characterization of protein glycosylation.1 Characterization of Ser- and Thr-linked †

Department of Molecular Biology. Department of Chemistry. (1) (a) Harvey, D. J. J. Chromatogr., A 1996, 720, 429-446. (b) Burlingame A. L. Curr. Opin. Biotechnol. 1996, 7, 4-10. ‡

10.1021/ac990578v CCC: $18.00 Published on Web 09/15/1999

© 1999 American Chemical Society

glycosylation (O-glycosylation) is especially difficult compared to Asn-linked N-glycosylation because of the lack of a reliable amino acid consensus sequence and availability of a general enzyme that will cleave all O-linked structures.2 Addition of N-acetylgalactosamine (GalNAc) to the hydroxyl group of Ser and Thr residues (mucin type) is one of the most common types of mammalian O-glycosylation,2 frequently occurring in sequence regions with a high density of Ser and Thr residues. These regions are often rich in Pro, Gly, and Ala residues, none of which represent cleavage sites for commercially available proteases. Recently, the potential of partial acid hydrolysis in combination with mass spectrometry has been explored for determination of mucin-type glycosylation sites.3 Due to the high density of Ser, Thr, and Gly, the residues that provide efficient acid cleavage sites, the hydrolysis often results in extensive polypeptide bonds cleavages with only partial carbohydrate loss and therefore facilitates identification of the glycosylated sites. The alternative direct mass spectrometric approach to determine O-glycosylation sites by postsource decay (PSD)4a and collisionally activated dissociation (CAD)4b has the advantage of allowing analysis of mixtures without prior separation. The major limitations of PSD and CAD techniques are that glycosidic bonds are more labile, and thus fragment easier than polypeptide bonds, resulting in a low signal intensity or even the absence of fragment ions carrying the glycans. Furthermore, PSD and CAD produce rather (2) Lis, H.; Sharon, N. Eur. J. Biochem. 1993, 218, 1-27. (3) Mirgorodskaya, E.; Hassan, H.; Clausen, H.; Roepstorff, P. Anal. Biochem. 1999, 269, 54-65. (4) (a) Goletz, S.; Thiede, B.; Hanisch, F. G.; Schultz, M.; Peter-Katalinic, J.; Muller, S.; Seitz, O.; Karsten, U. Glycobiology 1997, 7, 881-896. (b) Hanisch, F.-G.; Green, B. N.; Bateman, R.; Peter-Katalinic, J. J. Mass Spectrom. 1998, 33, 358-362. Rademaker, G. J.; Pergantis, S. A.; Bloktip, L.; Langridge, J. I.; Kleen, A.; Thomas-Oates, J. E. Anal. Biochem. 1998, 257, 149-160.

Analytical Chemistry, Vol. 71, No. 20, October 15, 1999 4431

Figure 1. ECD spectrum (50 scans) of the charge states 3+ and 4+ of the peptide TAP25 carrying three GalNAc groups. Numbers of the observed glycan groups are denoted on the sequence for corresponding fragments; these numbers are in agreement with the actual presence except for z7‚ that has lost one glycan group out of the total two.

Figure 2. ECD spectrum (10 scans) of the charge state 4+ of the glycosylated peptide HCGβ carrying three GalNAc groups. Only one glycan loss was observed from a fragment (z20‚).

specific b, y fragmentation5 with preferential cleavage at X-P and D-X (X is any amino acid),6 which is a drawback when extensive sequence information is required. Recently, a new method of ion fragmentation by low-energy electrons has been introduced.7 Electron capture dissociation (ECD) is based on partial recombination of multiply protonated polypeptide molecules with thermal (