DVMT OF METHODS, INSTR & TECH FOR IMPROVED PROTEOME & PROTEIN COMPLEX ANAL

这个子项目是众多研究子项目之一

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Specific Aim 2 seeks to improve proteome analyses through the development of new methods, instrumentation, and techniques. Specific areas of focus included improved methods for dealing with the phosphoproteome; improved methods for relative and absolute quantitation; advances in MS sensitivity; and elution time prediction for improved protein identification. A very common problem in biomedical proteomics is the desire to track the behavior of informative proteins that are present in low abundance, against a complex background of highly abundant proteins. Strategies for achieving this goal include both sample fractionation approaches to reduce the complexity of the sample analyzed, and instrument development approaches to extend the dynamic range of accurate measurements. These capabilities are the key to any study that involves discovery proteomics for clinical samples. Progress in this area has been driven by collaborative projects that feature non-conventional and/or highly complex samples and the need for dynamic range (D. Smith, Katze, Warren, and Kulkarni). Center advances for analyzing the phosphoproteome  Several technical capabilities have been developed during the last year at the Resource Center to facilitate phosphoproteome analysis, including a special automated LC platform that employs a smaller i.d., (50 ¿m) capillary analytical column with integrated tip and solid phase extraction (SPE) column for sample cleaning. This new LC system provides a high sensitivity and high throughput capability for phosphoproteome analysis. Sample preparation and Immobilized Metal Affinity Chromatography (IMAC) techniques were also further refined. Identification of phosphorylated proteins, changes in the phosphorylation state of a given protein, and identification of specific sites of phosphorylation are all questions of high importance to biologists. In particular, the Klemke collaborative project that focuses on the phosphoproteome of lamellipodia has played a crucial role in driving the development of improved techniques for phosphoproteomics, applying the combined dual-step 18O labeling, IMAC, and LC-MS analysis. Continued improvements in the sensitivity, quantitative accuracy, and completeness of phosphoproteome coverage will continue to be driven by the collaborative projects in the Center (Klemke, Rossie, Kohwi-Shigematsu, Stenoien, and Squier). The Center developed a label-free relative quantitation technology for global targeted phosphoproteome analysis by creating abundance profiles of mass and time features from multiple LC-MS experiments, and performing targeted LC-MS/MS experiments focusing on features of changing abundance profiles. Center researchers applied this technology as a proof-of-concept trial to compare changes in protein phosphorylation in HeLa cells treated with or without calyculin, a potent inhibitor of PP1, PP2A and other members of the phosphoprotein phosphatase (PPP) structural family. This technology was mainly directed toward the identification of phosphopeptides that showed statistically significant changes, and for facilitating the very time-consuming process of phosphopeptide sequencing. Both qualitative and quantitative phosphorylation changes, which are equally important for understanding the dynamic operation of signaling cascades, were revealed from this analysis. This study with collaborator Dr. Sandra Rossie identified more than 1769 unique phosphorylation sites in 1324 unique peptides and 718 unique proteins. More than 300 phosphopeptides corresponding to 277 proteins showed significant changes in response to calyculin treatment, and 83 phosphopeptides of differential abundance were detected in treated samples. These proteins included known targets for PPP enzymes, proteins not previously identified as direct or indirect targets for PPP enzyme regulation, and known and putative proteins not previously shown to be phosphorylated. This study thus reveals a large number of specific phosphorylation sites within new protein targets that are directly or indirectly altered by PPP enzyme activity. The number of changing targets identified in the Centers label-free targeted analysis is comparable to those determined in quantitative studies using differential labeling. The label-free approach has the added advantage of revealing qualitative phosphoproteomic changes in addition to quantitative changes, as differential labeling requires detection of peaks in both samples. This is particularly important when addressing changes in signal transduction-induced phosphorylation events. A similar strategy is being employed to identify PP5 substrates using the automated LC cart for phosphoproteomics combined with the Centers current AMT tag approach. The application of differential labeling for quantitation by Center investigators utilized a modified approach with collaborator Dr. Richard Klemke. This application of an enhanced stable isotope labeling approach for quantitative phosphoproteomics combines trypsin-catalyzed 16O/18O labeling plus 16O/18O-methanol esterification labeling for quantitation, a macro-IMAC trap for phosphopeptide enrichment, and a monolithic capillary column with integrated electrospray emitter. LC separation and MS/MS is followed by neutral loss-dependent MS/MS/MS for phosphopeptide identification using a linear ion trap (LTQ)-FT mass spectrometer and complementary searching algorithms for interpreting MS/MS spectra. To improve the confidence and throughput of phosphopeptide identifications, parameters utilizing accurate mass and reverse database approaches were applied. Advances in MS sensitivity  Two of the major factors that determine MS sensitivity are the efficiency of ion production and the subsequent effectiveness of the transmission of the ions to the detector. To improve sensitivity, Center investigators explored ESI technologies that utilize lower flow rates in the electrospray emitter to increase ion production. Approaches have also been explored that facilitate the use of multiple emitters that operate in parallel and allow the advantages of nL/min flow rate ESI (nano-ESI) to be obtained with traditional higher flow rate liquid separation techniques. In addition, efforts are moving towards reducing ion losses in the ESI interface by incorporating multiple highly efficient heated inlet capillaries. Such inlet modifications take advantage of the large ion capture region of the ion funnel compared to the traditional skimmer interface. The implementation of an electrodynamic ion funnel in Thermo Electron Corporation LTQ and LTQ-FT mass spectrometers improved the sensitivity by ~10 fold. The advantages associated with nano-ESI include: 1) the reduced flow rate decreases initial droplet size, which improves desolvation and increases ion production, 2) there is more excess charge available per analyte, which also increases ion production, and 3) there is less charge competition among different species, which improves quantitation. However, since capillary LC generally employs flow rates greater than ~1 ¿L/min, typical HPLC separations do not achieve the high performance afforded by low flow rate electrospray, which is most pronounced below ~50 nL/min. The approach is to divide the higher flow among multiple ESI emitters, thus enabling the advantages of nano-ESI from elevated flow rate separations. A new ESI emitter emitter fabrication procedure was developed and used to taper the ends of monolithic LC columns, which increased the sensitivity and quantitative ability of proteomic measurements. Silica-based monolithic narrow-bore capillary columns with integrated ESI emitters fabricated directly on the column were developed to provide high quality and robust microSPE-nanoLC-ESI-MS analyses. The integrated ESI emitters added no dead volume to the LC separation, and produced more stable nano-electrosprays at flow rates of ~10 nL/min. The integrated monolithic ESI emitter is more resistant to clogging and also provides good run-to-run reproducibility. Center researchers are using the ESI emitter fabrication method to develop a multiple nano-ESI emitter for improving the sensitivity and quantitative ability of higher flow rate liquid separations. RPLC elution time prediction for improved protein identification  Of benefit to all collaborative projects within the Center are developments that lead to improvements in protein identification. In 2003, the Center introduced an artificial neural network (ANN) method for predicting peptide elution times that was originally based on amino acid composition and later extended to include partial peptide sequence information. Various approaches have been explored for increasing peptide elution time prediction accuracy in reversed phase LC (RPLC). In addition to more complex ANN architectures, several peptide physicochemical (peptide length, hydrophobicity, etc.) and sequence-dependent parameters (peptide sequence, amphipathicity, nearest neighbor, etc.) were examined that have been shown to affect the peptide retention time in LC. To evaluate the sequence-dependent parameters, a much larger training dataset was necessary. As a result, the network was trained using ~345,000 non-redundant peptides identified from a total of 12,059 LC-MS/MS analyses of more than 20 different organisms, and the predictive capability of the model was tested using 1303 confidently identified peptides that were not included in the training set. The present model fully encodes the sequence of peptides up to 50 amino acid residues through an artificial neural network configuration of 1056 input, 24 hidden, and one output node. When compared to previously developed retention time prediction algorithms, the present model provides approximately two-fold better results. Unlike any of the previously developed predictors, this model is now able to accurately predict the retention times of both isobar and isomer peptides. Such capability allows more confident identification of isomer/isobar peptides otherwise indistinguishable by accurate mass measurements. Under current development in the Center are predictive tools for other separation dimensions. A predictive capability developed for strong cation exchange (SCX) liquid chromatographic peptide separations has begun and similar capabilities for IMS and FAIMS separations will follow.