QFPC Services and Facilities
Quantification of Protein Glycoxidation Products
Oxidative damage of biomolecules is thought to contribute to diabetic vascular disease, but the pathways involved in vivo are poorly understood. One important class of oxidants may be reactive intermediates generated by glycoxidation. The following measurements of markers implicated in oxidative and glycoxidative damage in vivo, including 3-chlorotyrosine, o,o’-dityrosine, ortho-tyrosine, meta-tyrosine, pHA-lysine, carboxymethyl-lysine (CML), and 3-nitrotyrosine, will be offered to DRC affiliate investigators.
Proteins of interest are identified by a variety of factors, including changes in relative abundance or alterations in apparent pI (which suggests a post-translational modification, such as phosphorylation). Proteins are immunostained and isolated, digested enzymatically, and identified by peptide mapping (MALDI-TOF-MS) or tandem mass spectrometric microsequencing (ESI-MS/MS).
The use of shotgun proteomics is available to the DRC affiliate investigator, where enzymatic digests of proteins are separated by liquid chromatography (LC) and subjected to electrospray ionization (ESI) and data-dependent MS/MS analysis. The peptides are identified by searching against a protein database.
Spectral counting and peptide ion intensity are alternative, label-free methods for quantifying relative protein abundance and will also be made available to the DRC affiliate investigator. Spectral counting sums all of the MS/MS spectra observed for peptides derived from a single protein. Because abundant proteins are more likely to be identified during data-dependent MS/MS scanning, spectral counting has the potential to quantify protein levels. Spectral counting is a useful statistic for assessing relative protein abundance in biological samples.
Identifying Differential Protein Expression
Selected reaction monitoring (SRM)
The combination of a single precursor peptide ion and its specific fragment ion is often referred to as the SRM channel. This tandem MS technique greatly reduces chemical noise, markedly improving the signal-to-noise ratio and thereby sensitivity. Furthermore, the instrument’s duty cycle is almost entirely used to monitor the specific SRM channel of interest, further increasing sensitivity. LC-MS/MS using SRM technique is thus capable of extraordinary sensitivity and precision.
Quantification with isotope-labeled peptides
To further increase the power of this approach and to quantify protein abundance, isotope-labeled analogs of the peptides of interest that contain 13C- and/or 15N-labeled amino acids are synthesized. The isotope-labeled peptides are processed in the same manner as the endogenous peptides generated by tryptic digestion so that any losses during sample workup and analysis are identical. The peptides thus serve as internal standards for determining the absolute quantity of a modified peptide. Both the native and isotope-labeled peptides (which are chemically identical but have different masses) are followed concurrently during LC-MS/MS analysis, using SRM.
Mutiplex analysis by SRM
SRM has been used extensively for small molecule drug metabolism and pharmacokinetic studies, which traditionally monitor only a small number of molecular species at one time. In targeted proteomics, however, samples containing hundreds or thousands of components may need to be analyzed. Thus, commercial software tools for analyzing and visualizing raw mass spectrometric data are inadequate for the task.
To overcome these limitations, the Core uses Skyline, a versatile open source software tool developed by the MacCoss laboratory (at the UW), to aid in the development of SRM methods and data analysis (21). Skyline contains an intuitive Windows graphical user interface and a powerful document editor for choosing peptides and transitions. It is fully integrated with peptide MS/MS spectrum libraries from BiblioSpec, the Global Proteomics Machine (GPM), and the National Institute of Standards and Technology (NIST).
Skyline is used to visualize spectra and build scheduled SRM methods directly by predicting the retention time of previously uncharacterized peptides. A state-of-the-art peak-finding algorithm in Skyline then provides rapid and reliable analysis of the acquired SRM data.
Skyline also contains a powerful set of tools that can determine which peak within a complex mixture is formed from a peptide of interest. To do this, Skyline uses the predicted retention time and a new scoring algorithm to compare the rank order of product ion intensities from co-varying transitions with a product ion spectrum stored in a spectrum library. The software provides full support for using standard peptide and stable isotope-labeled peptide internal standards.
Computer Cluster and Protein Identification
The Core has made available to the DRC affiliate investigator use of a 20-multicore-node computer cluster (96 CPU), which is used to run the SEQUEST and Xtandem search engines required for database searching. A user-friendly relational database system, Computational Proteomics Analysis System (CPAS), is used to store and analyze data.
In a typical experiment, MS/MS spectra from a proteomics experiment are searched against the species-specific International Protein Index (IPI) databases, using the SEQUEST and/or Xtandem search engines. Results are validated with PeptideProphet and ProteinProphet, using peptide and protein identification probability criteria and by searching against a randomized protein database to establish the false discovery rate. Results are uploaded into the CPAS system and processed further, using Skyline and quantification software developed in-house. The cluster’s impressive computational power is available for any bioinformatics analysis needed to support DRC work.
Data Storage and Integration
Using the Core’s established computer cluster and database relational system, a centralized database to house and disseminate the raw data and analyzed products generated from this proposal are provided. The existing database system (CPAS) encapsulates user authentication and project management features, and houses raw and analyzed experimental results.
The database will also allow the automated repository of results into PRIDE and file upload to the Tranche network. Tranche is a free and open source file-sharing tool. It is the primary mechanism by which large raw data files generated by mass spectrometry are stored and disseminated. Tranche is set up as a peer-to-server-to-peer distributed network, and will be accessible to all DRC investigators. It also allows outside investigators to access the Core’s proteomic data.
Technology Dissemination and Data Sharing: All data and software generated by the DRC are provided to the research community in a timely manner through a dedicated server, with technical support from the Core’s Bioinformatics Module.
Experimental Design, Power Calculations, and Statistical Analysis of Large Data Sets: The Bioinformatics Module also assists with experimental design, power calculations, and statistical analysis of large data sets. In addition to standard approaches to data analysis, differentially expressed proteins are analyzed for physical interactions and biological functions to identify integrated networks as shown below for the Sterol Responsive Network.
The Core’s newly renovated laboratory is on the first floor of the University’s new SLU campus.
The Core’s Mass Spectrometers
Current mass spectrometers include:
- a Thermo Finnigan LTQ linear ion trap instrument coupled with MichromBioresources Paradigm MS4B multidimensional HPLC and nanospray ionization;
- a Thermo Finnigan LCQ DECA XP Plus electrospray ionization ion trap tandem mass spectrometer interfaced with a 2-D capillary LC system in the ProteomeX configuration and with Agilent 1100 HPLCs;
- a Thermo Finnigan LCQ Classic electrospray ionization ion trap tandem mass spectrometer interfaced with a Waters 2690 microbore HPLC;
- two Hewlett Packard 6890 gas chromatographs equipped with HP5973 mass selective detectors and capable of both electron impact ionization and chemical ionization;
- a Voyager DE STR MALDI TOF mass spectrometer with high resolution mass accuracy for rapid identification of proteins via peptide mass fingerprinting;
- an Applied Biosystems 4700 Proteomics analyzer tandem MALDI-TOF-TOF instrument capable of high-throughput MALDI MS/MS analysis and supported by an LC-MALDI system consisting of Dionex Ultimate two-dimensional capillary HPLC connected to a Shimadzu Accuspot MALDI plate-spotting robot; and
- a Waters QTOF Premier with nanoelectrospray ionization and an Ultra Performance LC.
School of Medicine (SOM) Mass Spectrometry Resource
This facility is located at South Lake Union, which houses the Core laboratory. Because Core personnel are members of the School of Medicine Proteomic Resource, its personnel have full access on a fee-for-service basis to all of the Resource’s instrumentation. Current instrumentation includes a Thermo LTQ-FT Ultra, two Thermo LTQ-Orbitrap XLs, a Thermo LTQ XL with ETD, a Thermo LTQ, a Thermo TSQ Quantum Access, a Thermo TSQ Vantage, and seven Waters nanoACQUITY UPLC systems.
For proteomic data analysis, the Core currently maintains a 22 CPU computer cluster. The CPAS (Computational Proteomics Analysis System) data management and analysis system is used for data storage and high-throughput data analysis.
Robotic Liquid Handling Workstation
A Perkin Elmer Janus integrated liquid handling workstation is used for high-throughput proteolytic digestion, MALDI plate spotting, ELISA assay microtiter plate preparation, and affinity isolation of protein complexes for MS and MS/MS analysis.