Mass spectrometric methods for the analysis of protein structures are being improved, applied and tested in independent and collaborative research projects in proteomics. Projects in separation methods and data processing are in progress. In the past year, there has been progress on sample treatment (on-bead digestion);isotope labeled fusion proteins as protein standards; understanding a post translational modification in bacterial ribosomes;and multi-institutional collaborative studies on standards in proteomics. After determining the identity of a protein in a mixture by mass spectrometry, quantity is the most important experimental variable required in biological experiments. We previously reported a general approach for the determination of absolute amounts and the relative stoichiometry of proteins in a mixture using fluorescence and mass spectrometry (MS). Toward that objective, we engineered genes to express green fluorescent protein (GFP) with synthetic fusion proteins in Escherichia coli to function as spectroscopic standards for the quantification of analogous stable isotope-labeled, non-fluorescent proteins. However, we found that the limitations imposed by including 27 kDa for GFP severely constrained the size of the resulting expression protein. In order to express ten or more signature peptides (each containing 20-30 amino acids) per fusion protein, we redesigned synthetic genes to include both C-and N-terminal affinity tags, and include several copies of tryptophan to permit quantification by ultraviolet absorption. Proteins thus configured can be synthesized efficiently bearing stable isotope tags using an E coli in vitro translation system. Several labeled proteins have been prepared and are being used in studies to determine the stoichiometry of proteins in immunopurified postsynaptic densities. Combining antibody coated magnetic bead affinity purification as a separation tool with mass spectrometry as a protein identification tool has been a powerful method to determine the composition of interacting proteins and protein complexes. Receptor protein complexes and insoluble complexes offer added challenges, as they may contain hydrophobic membrane proteins. When strong detergents are required to release the complexes from the magnetic beads or prevent their aggregation and precipitation, the subsequent required removal of detergent may cause significant sample losses prior to MS analysis. We are developing a procedure in which protein denaturation and trypsin digestion are performed directly on the affinity bound complex on the magnetic beads, circumventing detergents and reducing sample loss prior to LC/MS/MS analysis. We found that an initial heating, reduction, and alkylation step, followed by wash, disrupts weak antibody complex interactions and allows removal of non-specific binding partners from the bead. Subsequent on-bead tryptic digestion in the presence of non-ionic detergents disrupts stronger antibody complex interactions and leads to detection of antigen and important binding partners, along with some peptide components of the antibody. Finally, an on-bead trypsin digestion in the presence of high urea concentrations disrupts very strong antibody complex interactions, and leads to peptides predominantly from antibody and Protein A and some strongly-bound antigen components. The three step on-bead process ensures thorough solubilization and digestion of protein complexes. This technique is being applied for postsynaptic density analysis as well as ErbB4 complex characterization. The ribosome is the universal macromolecular machine that translates the mRNA transcript into polypeptides. Analytical techniques have facilitated the identification of various ribosomal protein isoforms that result from post-translational modifications (PTMs). Beta-methylthioaspartic acid was previously identified as a novel PTM at position 88 in the Escherichia coli ribosomal protein S12. D88 is universal among all S12 bacterial orthologs and mutations at this position are lethal. Our goal is to elucidate the biological function of this novel PTM by identifying specific binding partners. We have continued to make progress toward characterizing the biological function of this PTM and are in the process of completing data analysis. The beta-methylthioaspartate occurs on a conserved loop region that has been shown through structural studies to be solvent accessible and may be a protein recognition site. Two complementary affinity pull-down strategies were applied to both stationary and mid-log phase samples. The first strategy used recombinant E. coli S12 protein with a C-terminal affinity tag to pulldown proteins that form interactions. This allows S12 to serve as bait for the identification of proteins that form a stable complex and/or form non-stoichiometric transient interactions. We reproducibly identified and tagged 7 candidate S12 binding partners that are specific to log and/or stationary growth phases (stress conditions). Among these is a mid-log phase specific protein (RimO) that was suggested by bioinformatic studies to be at least partially responsible for adding the modification on aspartic acid 88. The second strategy used synthetic biotinylated peptides that represent the conserved loop region (the putative recognition site) on S12. Peptides containing the unmodified and modified Asp 88 serve as baits. The peptide pull-down has the advantage of identifying proteins that directly bind or interact with S12. We reproducibly identified three proteins that bind specifically under mid-log phase conditions and two translational regulatory proteins that bind to S12 peptide under stationary phase conditions. These proteins are among the 7 candidates identified in recombinant S12 pull-downs as potential S12 binders. The importance of transient protein phosphorylation as an regulatory mechanism has been well documented, e.g., cellular signaling pathways. However, much of the evidence for phosphorylation is based on indirect methods, e.g., antibody-based detection and/or phosphate reactive staining of 1D and 2D PAGE gels. In principle, mass spectrometry can be used to detect and determine the sequence location of phosphorylated residues. In practice, the ionization of phosphorylated peptides is suppressed in the presence of non-phosphorylated peptides. Furthermore, the ionization of multiply-phosphorylated is suppressed in the presence of singly-phosphorylated species. In an effort to develop more robust methodologies for the ms-based characterization of phosphorylation, this laboratory is collaborating with the Association of Biomolecular Resource Facilities (ABRF) Proteomics Standards Research Group to develop a modestly complex mixture of singly- and multiply-phosphorylated peptides in a tryptic digest background of their precursor proteins. Multi-institutional studies will be performed in order to optimize qualitative methods this year and will lead to the development of quantitative methods as stable isotope labeled analogs of these reference peptides are prepared for follow-on studies next year. We are also collaborating with the ABRFs Proteome Informatics Research Group to develop improved informatics strategies for recognition and interpretation of tandem mass spectra derived from phosphopeptides contained in a large, previously published dataset. Multi-institutional studies will be performed using a broad array of informatics tools with the goal of defining an optimized workflow for accurate recognition and sequence placement of phosphorylated amino acid residues while minimizing false positive and false negative assignments.
