Elucidation of functionally significant structural modifications by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry with post-source decay

Peptide and oligosaccharide ions generated by matrix-assisted laser desorption/ionization (MALDI) undergo in-source collisions as they are accelerated through the concomitant plume of matrix. Post-source decay or decomposition (PSD) of these ions during the relatively long transition to the reflector of a reflecting time-of-flight (TOF) mass spectrometer results in abundant fragmentation at interresidue peptide and oligosaccharide bonds. Although these fragments are not separated at the linear detector, they may be resolved by variation of reflector potential due to differences in their translational energies. Four examples are presented for which PSD has successfully provided structural data on modified or variant peptide sequences, thus demonstrating the utility of this technique. 1. A by-product generated during synthesis of a peptide antigen produced an antigenic response that dominated that of the unmodified form of the peptide sequence, IMIKFNRL, when mice were injected with the unfractionated mixture. As a consequence T cells were cloned that selectively recognized this peptide variant when presented as a complex with major histocompatibility complex molecules. Mass analysis revealed that the dominant peptide was 56 Da greater than the unmodified sequence, which is consistent with presence of a t-butyl substituent. Analysis of sequence-specific PSD fragments enabled localization of this substituent to either Asn6 or Arg7 of the sequence. The presence of an ammonium ion at m/z = 143.2 and automated Edman degradation confirmed the presence of the t-butyl group on Asn6. 2. Abundant PSD occurred with a peptic peptide derived from the attachment protein of human respiratory syncytial virus. This enabled determination that the protein contained a cystine noose. This was possible due to the fact that the peptic peptide produced PSD fragments which included a preserved intermolecular disulfide as well as less abundant disulfide fission fragments. Furthermore, no fragmentation of bonds occurred along the peptide backbone within an intranolecular disulfide loop also present in the peptic peptide. 3. A variant avirulent isolate of Newcastle disease virus was identified by antigenic analysis that revealed an amino acid change in a region of its fusion protein that determines the pathotype potential of the virus. Mass mapping of the subunit of the fusion protein containing the variation identified the presence of a glutamic acid in the C-terminal AspN fragment of the variant subunit compared to glycine in the subunit of another avirulent isolate. Confidence in this conclusion was possible due to the accuracy afforded by use of delayed extraction of ions from the ion source and internal calibration, typically within 20 ppm. This C-terminal AspN peptide underwent PSD to a substantial degree and it was possible to determine that the glycine-to-glutamic acid variation occurred at the C-terminus of the affected subunit. 4. The nucleocapsid protein of Newcastle disease virus is known to be phosphorylated, but the region of the protein and the residue(s) involved have not been defined. Mass analysis of tryptic peptides of the nucleocapsid protein isolated by HPLC revealed a peptide which differed from the mass expected for the C-terminal tryptic peptide of the protein by 80 Da. Compar-isons of PSD spectra of the nonphosphorylated and putatively phosphorylated forms of the peptide revealed the presence of phosphate on a threonine residue in a consensus motif for phosphorylation by a MAP kinase. In summary, our experiences with MALDI-TOF-PSD-MS have been very fruitful in the course of defining the nature and sites of modification of peptides of biological relevance whether synthetic in origin or isolated from viral proteins.