The UNC-Duke Proteomics Center (http://proteomics.unc.edu) provides researchers with the instrumentation, technical expertise and experience to identify and characterize proteins from biological samples from different species, even those species whose genomes have not been completely sequenced. The Center has established routine procedures to identify gel-separated proteins stained with selected visible and fluorescent dyes. In addition, the Center is developing novel techniques to identify a variety of covalent modifications of the proteins, which are often the critical steps in regulation of their activity, localization or turnover. Finally, the Center continues to improve upon techniques used to identify protein-protein interactions.

Central to the Center’s success are the highly-skilled personnel involved in the daily operations of the facility. The Center currently supports one Research Professor, four post-doctoral Research Associates, two Research Technicians, and a Computer Systems Administrator, all of whom are under the administration of a Faculty Director.

The current Faculty Director is Dr. Xian Chen, an Associate Professor in the Biochemistry Department at UNC, who will be responsible for overall design of proteomics experiments at the facility. He is well-experienced in both areas of development of novel proteomics techniques and application of these techniques to solve challenging issues involved in pathogen proteomics and the proteomics associated with innate immune responses. The Facility Director, Dr. Carol Parker, was hired in 2001, and has 36 years experience in mass spectrometry. Dr. Parker is responsible for managing the overall operation of the Proteomics Center and directs the protein characterization component of the Core’s operation. The Assistant Director, Dr. Maria Warren, has had 5 years experience in proteomics and gel-based analyses. Dr. Warren joined the Center in 2004, and comes to UNC from a proteomics core facility at another academic institution. She has extensive expertise in the areas of protein modification and differential expression analysis. Dr. Parker holds a faculty appointment in the Department of Biochemistry & Biophysics in the School of Medicine.

The UNC-Duke Proteomics Center currently supports one Research Professor, five Ph.D. level Research Associates, two Research Technicians, and a Computer Systems Administrator. Bioinformatics support, including support for the Laboratory Information Management System (LIMS), is provided by the Center for Bioinformatics at UNC. Dr. Parker is assisted by Dr. Yan Wu and Dr. Viorel Mocanu in the protein characterization projects. Dr. Viorel Mocanu assists Dr. Parker with protein characterization and performs the epitope mapping experiments as well as a variety of separation and purification techniques, including IMAC and affinity-based separations. Dr. Wu performs some of the LC/MS/MS experiments on the Q-trap and Q-tof. Ms. Mimi Mocanu performs GE Healthcare’s DIGE (differential gel electrophoresis) analyses. Dr. Maria Warren is responsible for protein quantitation on the Q-Trap LC/MS/MS instrument. Dr. David Robinette directs the protein identification from gel studies, which are performed by Ms. Nely Dicheva. Dr. Linhong Jing directs the Core's research projects on the FT-MS , which include "topdown" and LC/LC/MS/MS-based FT-MS projects. Dr. Jun Han is in charge of the new FT-MS-based Metabolomics section using our new 12T FT-MS.

Our services include protein identification, protein characterization, differential protein expression analysis, and FT-MS-based metabolomics. In addition, we are also involved in a number of collaborations and our own research projects, which have resulted in 31 publications and more than 50 papers presented at scientific meetings (as of August 2006). Currently, approximately 150 research groups, at UNC, other universities, and in industry, use the services of the Center. Samples entering the facility are processed following one of several types of proteomic workflows. The progression of protein identification from gels is tracked using the SQL*LIMS. For protein characterization projects, the collaborating individual is responsible for preparing the sample for mass spectrometry, but guidance is available through discussions with the Center staff and reference resources are available on the facility website. Individual investigators are strongly encouraged to discuss their unique applications with Center personnel before submitting their samples.

The services are broadly categorized as:

1) Identification of Gel-Separated Proteins:

The majority of samples we receive are submitted for in-gel digestion and subsequent identification. The MALDI mass fingerprinting and “sequence-tag” methods used at UNC for protein identification are well validated, and tens of thousands of proteins have been identified since the Center opened in January 2002.  The accurate molecular weight and partial sequence information from MALDI TOF/TOF analysis of even a single peptide is often sufficient to identify a protein or a family of homologous proteins containing that peptide. Our work with the MALDI TOF/TOF in the UNC-Duke Proteomics workflow scheme has shown that the success rate for identification of a gel-separated protein can exceed 92% provided that the protein stems from an organism with a completely annotated genome. For proteins from genomes that have not been completely sequenced, protein identification by de novo sequencing using a variety of mass spectrometric approaches (e.g., specific labeling, MS/MS using Q-TOF instruments) can also be arranged.

 

2) The DIGE technique for Relative Expression Levels of Gel-Separated Proteins:

As a result of the expressed need on the part of our users, the Center recruited an Assistant Director (Dr. Maria Warren) who is an expert in the areas of protein quantitation using 2-D differential gel electrophoresis analysis (DIGE), and the Center has greatly expanded its capabilities in this area. DIGE is a technology developed and marketed by GE-Healthcare. It is based on the use of three cyanine dyes or labels that are matched in molecular mass and charge. The labels are fluorescent and have non-overlapping excitation and emission maxima.

In a typical DIGE experiment the samples and control are each labeled with only one of the three dyes. After labeling, the samples and control are pooled, loaded onto a single first dimension strip and separated by isoelectric focusing then SDS-PAGE. A fluorescent imager is then used to image the different fluorophores of the proteins contained in each preparation (i.e., sample or control. Selection of appropriate filters for a particular cyanine label allows each protein preparation to be imaged independently. Software is then used to overlay the images, define spot boundaries and compare quantitation levels. The main benefit of DIGE technology is the ease in comparison of the protein preparations since the samples and control are run on the same gel and are separated under the same electrophoretic conditions.

We currently have IPGPhor and the DALT II systems for processing large format two-dimensional gels. At present samples will be submitted to the facility as protein extracts. The protein quantitation, Cy-dye labeling, two-dimensional separation, imaging and analysis will be completed in the Proteomics Center. The Proteomics Center is equipped with a ProXpress Proteomic Imaging System -- a high-resolution imaging and analysis system that provides 50 micron resolution over the entire gel image, and has a working range for illumination and detection anywhere in the visible or UV spectrum. This imager achieves a linear dynamic range of over three orders of magnitude, and can detect as little as 30 fmol of a SYPRO Ruby-stained protein on a 2D-PAGE gel.  We also have access to a GE Healthcare Typhoon 9400 high performance gel and blot imager with the capability of microarray analysis. This instrument offers three methods of detection: storage phosphor, fluorescence and chemiluminescence. The PhosphorImaging modes of the Typoon 9400 include Blue excited fluorescence (457 nm and 488nm), Green excited fluorescence (532 nm), Red excited fluorescence (633 nm), and Chemiluminescence. 

For spot detection and analysis the Progenesis Workstation Image Analysis Software package (Nonlinear Dynamics) significantly reduces the amount of subjective user intervention required to quantify differential expression. Gel spots containing proteins that have been determined to be up- or down-regulated can then be excised using our BioMachines 2DiD robotic workstation and identified by the in-gel digestion procedure.

3) Full-Length Molecular Weight Analysis and Protein Characterization:

Pure protein samples may be submitted for characterization by accurate molecular weight determination, followed by proteolytic digestion in solution. This is an appropriate scheme if the investigator is attempting to characterize posttranslational modifications of an isolated protein. The sample is typically subjected to HPLC purification prior to submission or purified and desalted according the procedures described on the Center’s webpage. Select fractions are analyzed by MALDI/TOF to determine the full protein molecular weight, which provides information about the physical status and polydispersity of the protein. Samples are then enzymatically digested and subsequent MALDI-MS mass mapping and MALDI-MS/MS sequencing, if necessary, is used to confirm the identity of the protein and to characterize the domains present. An analytical method for identification of phosphorylation sites, based on the combination of immobilized ion affinity chromatography (IMAC) and MALDI (QqTOF)-MS/MS,1 or immobilized antibodies and MALDI/MS/MS,2 was developed by members of the Proteomics Center.  Complementary methods are currently under development for automatic characterization of specific modifications, such as glycosylation and acetylation. These methods are based on capillary liquid chromatography coupled on-line to data-dependent mass spectrometry. The mass spectrometers can be programmed to trigger sequencing automatically in a modification-dependant manner (LC/MS/MS).

The Bruker 12T FTICR(Fourier Transform Ion Cyclotron Resonance)-MS allows high accuracy molecular weight determination of proteins up to ~50 KDa.  It also allows “top-down” sequencing, which we have demonstrated to be a powerful tool for modification site analysis, including phosphorylation site determination, as well as structural studies of enzymes 3, 4.

4) Relative Protein Expression Measurements Without Prior Gel Separations (a developmental project at UNC-CH):

To understand protein behavior in biological systems comprehensively, it is necessary to identify and (often) quantify the expression patterns of low abundance regulatory proteins (at ca. 100 copies per cell) against an intense background of structural and carrier proteins. To perform this feat without selective preconcentration requires extraordinary resolution and a dynamic range of six or more orders of magnitude. Although 2D gels have excellent protein capacity, the limits of resolution and dynamic range mandate subcellular fractionation and (often) ion exchange, size exclusion, or fractionation as prerequisites to sample loading. To overcome the associated losses, most studies start with the preparation of ca. 10 billion cells.

In addition to sample requirements, very acidic, basic, and membrane-associated proteins do not focus well in the first dimension. Also, the 2-D gel method is labor intensive, and the preparation of gels and subsequent image analysis is time consuming and expensive to automate. This is why the UNC-Duke Proteomics Center is developing alternative methods based on LC with online or offline preconcentration and separation, and relative quantification using the MALDI TOF/TOF, or ESI-MS/MS instruments.

One promising method of quantitation is based on differential isotopic labeling.  This method involves quantitation of proteins on the basis of differential acetylation of their peptides with H6 or D6 acetic anhydride.  Differential acetylation of peptides produces pairs of peptides which differ in mw by 3 or 6 Da.  Quantitation is done by digesting two protein samples, differentially acetylating the resulting peptides, recombining the digests, and analyzing them to determine their relative peak heights or peak areas.  This method of isotopic labeling has been developed and successfully applied in our laboratory.5-8 

5) FT-MS-based Metabolomics:

Another project under development at UNC is the relative quantitation of metabolites using FT-MS.  From the ultra-high mass accuracy and precision of the molecular weights, one can determine the elemental composition of these metabolites.  The mass resolution of the FT thus reduces the requirement for separation of the analytes prior to MS analysis, and allows the determination of up- or down-regulated metabolites in direct-infusion experiments.

 

Facilities and Equipment

1)  For Gel-based Samples

  • Two-dimensional Electrophoresis Units: The IPGPhor isoelectric focusing system (GE-Healthcare) is capable of running up to twelve IPG strips (of equal length and pH range) simultaneously. The unit has a Peltier cooler for maintaining constant temperature throughout the first-dimensional separation, which typically takes up to 18 hours. For the 2nd-dimensional separation, Center staff is currently using a DALT II six-gel electrophoresis unit also from GE-Healthcare. The unit processes up to six large format (26 x 20 cm) gels as a single batch, and maintains constant temperature using an auxillary thermostatted circulator.  At present, because of limited capacity, these systems are only used for our DIGE gels  –  all other 1D and 2D electrophoresis is performed by the submitters on their own equipment.

    ProXpress Proteomic Imaging System.     The ProXpress High-resolution Digital Imaging system (PE Biosystems) can generate an image using visible or fluorescently-labeled stains such as the Cy dyes (for 2D-DIGE) and Sypro Ruby. This image can then be marked for cutting by the users, either with or without differential expression analysis, which is performed using the ProGenesis software package. The Progenesis software can then generating a cut-list, and the gel is transferred to the Biomachines 2DiD system for cutting.

    The ProXpress is a high-resolution imaging and analysis system optimized for fluorescent stained large format (23x28 cm) 2D-PAGE gels. The unit provides 50 micron resolution over the entire gel image using >1.5 million 9-micron square pixels on the cooled Kodak KAF-1602E-1 CCD chip by stitching together a series of 5 x 9 cm rectangular frames. The ProXpress can perform this task with a positional accuracy of 50 micron and a repeatability of 25 micron. The camera uses a Xenon arc lamp and six selectable filter wheels to define a working range for illumination and detection anywhere in the visible or UV spectrum. With a user-selectable image illumination time the system can detect as little as 30 fmol of protein stained by SYPRO Ruby and separated on a 2D-PAGE gel. By combining these features, the camera system achieves a linear dynamic range better than three orders of magnitude.

    Progenesis Image Analysis Software. The Progenesis image analysis software platform (Nonlinear Dynamics) significantly reduces the amount of subjective user intervention required to quantify differential expression with state-of-the-art algorithms for robust spot matching and whole image warping. Progenesis hosts a combination of image processing and statistical analysis tools in an open architecture that exports in XML format for seamless integration into the LIMS packages under consideration. The software was purchased with the following NLD-certified hardware for optimal performance:
    • IBM PC workstation with 1.7 GHz Pentium 4 processor
    • 1GB 800MHz 16D RDDRAM RIMM Memory
    • Fire GL4 graphics card
    • 18Gb 10,000 rpm Ultra 160 SCSI Hard drive
  • Gel Cutting: An automated robotic system, the BioMachines 2DiD, is capable of performing both the gel imaging and the gel cutting, and can cut based on “cut-lists” generated by the Progenesis software package or by the user. This robotic system automatically cuts the gels and places the plugs into 96-well plates.

  • Enzymatic digestion of proteins: A ProGest gel processing robot is used to perform automated in-gel tryptic digestion of excised gel spots in a 96-well plate format. The ProGest has a limited capacity of 96 gel slices per day; however, the facility tripled its digestion throughput capacity with the addition of two more ProGest robots in February 2004.

2) Mass Spectrometers and Interfaces

Four instruments are available for nanoESI-MS and nanoESI-MS/MS, and two instruments are available for MALDI-MS and MALDI-MS/MS. This ensures that analyses can always be performed, even during periods of repairs or maintenance.

  • ABI 4700 Proteomics Analyzer (TOF/TOF):  This instrument is a MALDI-MS and MS/MS mass spectrometer with a mass accuracy of 5 to 10 ppm, and a resolution of 25,000. This instrument is the main instrument used for protein identification.  The ABI 4700 TOF/TOF incorporates linear MS, reflector MS and true precursor-selected MS/MS modes. By combining all these capabilities with a 200 Hz laser, high-energy fragmentation cell and superlative ion optics, the ABI 4700 TOF/TOF workstation can deliver a high throughput (100+ samples/hr), flexible, ultra-sensitive approach to protein identification. The ABI 4700 TOF/TOF achieves a resolution for precursor ion selection of +/- 3 Da using a dual-gate Timed Ion Selector (TIS). High-energy fragmentation can produce meaningful sequence from low and high-mass peptides and, since these fragment ion masses are accurate to 0.1-0.2 Da, it is possible to use the ABI 4700 TOF/TOF to identify unknown proteins where the genome is incomplete.
    • Applications Package, Base Layer, Automated Protein ID - This is our Oracle based off-line data station, expressly designed for high throughput automated protein ID. The package includes an off line, Windows 2000-based workstation, and software that allows for batching of the following planned experiment types:
      • MS only for peptide mass fingerprinting
      • MS for peak selection followed by MS/MS for protein ID
      • MS for peptide map fingerprinting (PMF) protein ID followed by MS/MS for confirmation or independent ID (differs from previous option in that this experiment requires data-dependent isolation of precursor ions for MS/MS)
      • MS only for protein quantitation (ICAT)
      • MS for quantitation followed by expression-dependent MS/MS for ID (ICAT)
      • De Novo sequencing package
      • High-throughput posttranslational modification screening tool
      • Protein Librarian

 

Database searching

In database searching, the measured peptide masses are compared with peptide masses and sequences obtained by in silico digestion of each sequence from a protein database (using the same specificity as the enzyme in the experiment) and a score is assigned as a measure of similarity between the protein in the database and the measurement. Proteins in the database are then ranked according to their scores.9, 10 The initial searches are unrestricted, so that all genomes and molecular weights are considered. Subsequent searches can be done on selected species, or against a specific target protein. Members of the UNC-Duke Proteomics Center review the data to confirm protein assignments. Candidate proteins are not considered “confirmed” until quality assurance criteria are met.

The Center has a site license for Mascot9 which allows us to customize searches, databases, and posttranslational modifications. For manual interpretation of MS/MS data, Protein Prospector11 is also used. The Mascot software package has the following features:

    • Automatic analysis of protein mass spectra from any mass spectrometer
    • Can search MS or MS/MS data
    • Batch analysis of multiple spectra without further user intervention
    • Customizable
    • A common set of interpretation tools for clear and efficient visualization of annotated spectra and protein sequences
    • All interpretation tools are hyperlinked to aid intelligent browsing or data mining
    • Automatic identification of protein mixtures
    • Automatic identification of posttranslational modifications.
    • Automated storage of data analysis results in a relational Oracle database
    • All search engines are supported
    • The package can search any database, NCBI GenBank, SwissProt, PIR, dbEST etc. One novelty of this package is that it can be used to search genomic data directly (in all six reading frames) using mass specific information obtained from peptide sequencing experiments from any mass spectrometer. Thus we have the capacity to independently identify proteins by fingerprint and sequence information derived from a single MALDI spot.
    • MASCOT results can be integrated into the ABI LIMS system

Mascot incorporates a probability-based implementation of the Mowse scoring algorithm.10, 12 By casting the Mowse score into a probabilistic framework it is possible to determine the absolute probability that a match is random, and knowing the size of the sequence database being searched, it becomes possible to provide an objective measure of the significance of a result. Mascot reports these scores as -10*LOG 10(P), where P is the absolute probability. Typically Mascot scores higher than 70 are considered significant.

Data processing and database searches are performed using both vendor-specific software platforms (included with the ABI and Micromass instrumentation packages) and the Mascot software from Matrix Science running on a local Dell server (four 1.7 GHz Xenon processors, with 72Gb RAID Hard disk and 4Gb RAM). UNC-CH maintains software licenses and support for Mascot (Matrix Science), Global Server (Micromass), and Progenesis (Nonlinear Dynamics).

The introduction of high-throughput proteomics research requires additional resources for storage, analysis, and retrieval of data. UNC-CH is currently expanding its bioinformatics infrastructure to account for the large numbers of projects based on proteomics technologies. Analysis will be performed on a site-licensed version of Mascot (Matrix Science9), and the ABI high-throughput analysis software, residing on a Dell NT server with four 1 GHz Xeon CPUs. All proteomics data generated in the UNC-Duke Proteomics Center will be stored on an Oracle (version 8i or 9i) database server, a Sun V880 containing 750 MHz processors (4 total), 8 GB of RAM, 400 GB active Raid storage attached to the server and an additional 2.5 TB of Raid storage.

Computational tools will provide data mining from local and public resources such as GenBank, SwissProt, PDB and others. The data management system, a customized ABI LIMS, will track and locate physical resources, but will also manage metadata describing experimental conditions and sample origin. These systems were chosen specifically to allow easy upgrades as required by the capacity of the sample flow and the implementation of new equipment and procedures in the UNC-Duke Proteomics Center. All hardware is upgradeable, and new hardware can seamlessly be integrated into the proteomics infrastructure without the need for replacement of current systems or the permanent migration of data from current storage systems.

Sample Tracking and Data Management with SQL*LIMS:

The SQL*LIMS system for proteomics offer a complete system that integrates sample handling; sample analysis, such as 2D-PAGE and isotope-coded affinity tag (ICAT) experiments; gel image analysis; instruments, such as gel cutting robots and mass spectrometers; and bioinformatics tools for managing and analyzing mass spectroscopy (MS) data, sequence and protein expression data, posttranslational modifications, and other types of data coming from protein identification and characterization experiments. Specifically, this system offers:

    • A complete laboratory information management system (LIMS) for sample, process and results management. Users will have available standard input, retrieval and analysis screens.
    • Complete sample tracking of all parent-child sample relationships allowing the Data Curator to easily drill back and locate original sample information should a sample require reanalysis.
    • Storage of all MS and sequence data in a single database with options that allow data sharing.
    • The ability to accommodate workgroup segmentation for enhanced security
    • The ability to accommodate multiple laboratory workflow schemas if multiple contractors are involved in the same phase of the process.
    • An open architecture that is ISO 9001 compliant. Under the proposed system we will be using SQL*LIMS on a Windows NT application
    • The system architecture will also accommodate most standard data mining, statistical analysis tools and many custom proteomics packages.

The open architecture of SQL*LIMS facilitates interfacing with virtually any piece of laboratory hardware or software package. Working with the manufacturer of the particular package, we obtain file format information for input and output files. SQL*LIMS then simply creates input files that contain the samples and/or tasks to be performed and a SQL*LIMS Task ID. This is used by the target system to guide the desired activity (i.e., performing mass spectrometric analysis and database searching on a series of samples). Once the tasks have been performed and results are generated, a file parsing routine is used to bring the data/results back into the SQL*LIMS system. The unique SQL*LIMS Task ID is the definitive key to associate the results with the appropriate sample(s) being processed.

The SQL*LIMS system is designed to be supported by and provide data to an Oracle database. It creates a standard configuration within the database that is designed to maximize all of the capabilities of the application, and to include the support for data analysis and mining. Within the context of this standard configuration, there is great deal of flexibility for customizing workflows and interfaces at every stage of the process.

 

The Bruker Ultraflex MALDI-TOF/TOF with reflectron and delayed extraction is used for intact protein/large peptide molecular weight determination.  It is also used for high-sensitivity analysis of peptides adsorbed onto affinity beads.  The Ultraflex is capable of mass accuracy of 3.5 ppm or better throughout the peptide mass range when using internal High Precision Calibration (HPC)TM. High mass resolution is achievable over a broad mass range, with typical mass resolution ranging from 10,000 to more than 35,000 for peptides in the m/z range 800-6,000, and up to 30,000 for low molecular weight intact proteins (e.g. cytochrome c). The Ultraflex is capable of detecting as little as 200 attomoles of peptide derived from BSA digest.  Peak detection and sequencing sensitivities of 200 amol peptide applied to the target from a 400 attomole/μL solution (0.5 μL applied) have been achieved in our laboratory using the Ultraflex.

LC/MALDI interface -- LC Packings Probot: The Probot is a robotic device for depositing the effluent from a capillary HPLC, and MALDI matrix, onto a MALDI target. The resulting track is actually a line of MALDI spots, deposited at a rate of up to 1 spot every 5 seconds. Targets prepared on the Probot are analyzed by MALDI MS to create a “chromatogram”, without the need for collecting fractions. This has the result of “uncoupling” the HPLC separation from the MALDI analysis, and allows the HPLC separation to be performed “off-line” with subsequent MALDI analysis of the LC-separated components. The targets can be “archived”, and re-interrogated if, for example, MS/MS is later desired on a particular ion of interest. When used with different column packings, this interface can be used for the separation of a mixture of full-length proteins or a mixture of peptides.   This interface can be used with either the Ultraflex or the ABI 4700 MALDI mass spectrometers.  The Dionex Probot is currently being used for projects involving the detection of phosphopeptides and for the detection of specific peptides that allow the researchers to distinguish between protein isoforms.

ABI Quadrupole/Time of Flight (Q-STAR) capillary LC-ESI/MS and nanoESI-MS and ESI-MS/MS and MALDI-MS and MS/MS with a mass accuracy better than 15 ppm and a resolution in excess of 8,000 for both MS and MS/MS data. Using the nanoelectrospray interface on the Q-Star, Center staff have sequenced peptides at 400 attomole/μL, consuming only 40 attomoles of peptide. The data from this ultratrace amount of sample resulted in a searchable sequence tag and a positive identification. With the nanoESI source, 25 samples can be analyzed per day.   This instrument is used for high-accuracy protein molecular weight determination, and high-sensitivity MS/MS analysis of peptides.

Micromass Q/TOF API-US LC/MS/MS:  This instrument provides capillary LC-MS/MS and nanoESI-MS with a mass accuracy of 10 ppm when calibrating MS/MS data against a known peptide eluting during the LC gradient. The resolution for peptides is in excess of 17,000 (using W mode for ion optics) and the instrument can perform MS/MS experiments on precursor ions up to m/z 3000. The system can isolate and sequence peptides in a data-dependent, “highest-intensity-first” manner or the API-US can be programmed to preferentially isolate lower intensity peptides based upon the results from accurate-mass neutral loss, precursor ion scans or previous MALDI-MS data. Hence, the system can identify modified peptides that might not automatically trigger sequencing in a standard data-dependent sequencing experiment. The API-US was also purchased with a Waters CapLC designed for close coupling of chromatography at capillary (75 μm i.d.) flow rates (~200 nL/min) using a novel Z-spray atmospheric interface.  The Q-tof is a powerful instument for the identification of proteins from complex mixtures, and for modification site determination.  The MS/MS spectra obtained from the Q-tof can be searched with either the Waters ProteinLynx software, or with Mascot9.

ABI Q-Trap mass spectrometer: The ABI 4000 Q-Trap LC/MS/MS System was purchased in late 2005. This instrument is a hybrid triple quadrupole/linear ion trap mass spectrometer, and is especially useful for the identification of phosphorylation sites. The Q-trap LC/MS/MS will also be utilized for detection of post-translational modifications in proteins, and for quantitation projects. The 4000 Q-trap is capable of mass resolution >3,000 at m/z 609, and has a mass range of m/z 5-2800 in Q1/Q3 RF/DC mode, and a mass range of 50-2800 in linear ion trap mode. In addition, it is capable of high-sensitivity full-scan MS, MS/MS, and MS3 with high-selectivity from true triple quadrupole precursor ion (PI) and neutral loss (NL) scans, and can perform multiple reaction monitoring (MRM) for quantitation using the high-sensitivity triple quadrupole.  The Q-Trap is interfaced to an ABI TEMPO capillary LC system for on-line LC/MS/MS.  The MS/MS spectra obtained from the Q-Trap can be searched with either ABI Analyst software, or with Mascot.

FTICR (Fourier Transform Ion Cyclotron Resonance) Mass Spectrometry: A Bruker Daltonics 12 Tesla Apex IV® QqFTICR mass spectrometer equipped with an Apollo® nano-electrospray ionization source was installed in the Proteomics Center in the fall of 2005.   This instrument provides the ultimate in resolution (>100,000), mass accuracy (typically <1 ppm) and sensitivity, and can be used for “top-down” sequencing.13 The accurate molecular weight information provides elemental composition data for small molecules, and has led to the development of a “discovery metabolomics” program. 

The high mass accuracy obtainable from intact proteins allows the determination of the modification status. It will also allow the determination of the exact composition and stoichiometry of high molecular weight (MW) protein complexes. Its CID capabilities will allow the complete elucidation of glycan structures in glycopeptides and glycoproteins, as well as the identification and localization of labile posttranslational modifications. Sequence information can be obtained by Electron Capture Dissociation (ECD), which allows the sequencing of very labile compounds, including γ-carboxyglutamic acid, which cannot be sequenced by conventional ESI CID instruments.14  


References

  1. Raska, C. S., Parker, C. E., Dominski, Z., Marzluff, W. F., Glish, G. L., Pope, R. M., and Borchers, C. H. (2002) Direct MALDI-MS/MS of phosphopeptides affinity-bound to immobilized metal ion affinity chromatography beads. Analytical Chemistry 74, 3429-3433.
  2. Raska, C. S., Parker, C. E., Sunnarborg, S. W., Pope, R. M., Lee, D. C., Glish, G. L., and Borchers, C. H. (2003) Rapid and Sensitive Identification of Epitope-Containing Peptides by Direct Matrix Assisted Laser Desorption Ionization Tandem Mass Spectrometry of Peptides Affinity-bound to Antibody Beads. J. Am. Soc. Mass Spectrom. 14, 1076-1085.
  3. Borchers, C. H., Thapar, R., Petrotchenko, E. V., Torres, M. P., Speir, J. P., Easterling, M., Dominski, Z., and Marzluff, W. F. (2006) Combined top-down and bottom-up proteomics identifies a phosphorylation site in stem-loop-binding proteins that contributes to high-affinity RNA binding. Proceeding of the National Academy of Sciences of the USA 103, 3094-3099.
  4. Borchers, C. H., Marquez, V. E., Schroeder, G. K., Short , S. A., Snider, M. J., Speir, J. P., and Wolfenden, R. (2006) Fourier transform ion cyclotron resonance MS reveals the presence of a water molecule in an enzyme transition-state analogue complex. Proceedings of the National Academy of Sciences of the United States of America 101, 15341-15345.
  5. Glocker, M. O., Borchers, C., Fiedler, W., Suckau, D., and Przybylski, M. (1994) Molecular Characterization of Surface Topology in Protein Tertiary Structures by Amino-Acylation and Mass Spectrometric Peptide Mapping. Bioconjugate Chemistry 5, 583-90.
  6. Borchers, C., Parker, C. E., Deterding, L. J., and Tomer, K. B. (1999) Preliminary comparison of precursor scans and liquid chromatography-tandem mass spectrometry on a hybrid quadrupole time-of-flight mass spectrometer. J Chromatogr A 854, 119-30.
  7. Hochleitner, E. O., Gorny, M. K., Zolla-Pazner, S., and Tomer, K. B. (2000) Mass spectrometric characterization of a discontinuous epitope of the HIV envelope protein HIV-gp120 recognized by the human monoclonal antibody 1331A. Journal of Immunology 164, 4156-4161.
  8. Hochleitner, E. O., Borchers, C., Parker, C., Bienstock, R. J., and Tomer, K. B. (2000) Characterization of a discontinuous epitope of the human immunodeficiency virus (HIV) core protein p24 by epitope excision and differential chemical modification followed by mass spectrometric peptide mapping analysis. Protein Science 9, 487-496.
  9. www.matrixscience.com.
  10. Perkins, D. N., Pappin, D. J., Creasy, D. M., and Cottrell, J. S. (1999) Probability-based protein identification by searching sequence databases using mass spectrometry data. Electrophoresis 20, 3551-3567.
  11. ProteinProspector (http://prospector.ucsf.edu/).
  12. Mascot (www.matrixscience.com).
  13. Kelleher, N. L., Lin, H. Y., Valaskovic, G. A., Aaserud, D. J., Fridriksson, E. K., and McLafferty, F. W. (1999) Top Down versus Bottom Up Protein Characterization by Tandem High-Resolution Mass Spectrometry. Journal of the American Chemical Society 121, 806-812.
  14. Kelleher, N. L., Zubarev, R. A., Bush, K., Furie, B., Furie, B. C., McLafferty, F. W., and Walsh, C. T. (1999) Localization of Labile Posttranslational Modifications by Electron Capture Dissociation: The Case of g-Carboxyglutamic Acid. Analytical Chemistry 71, 4250-4253.

 

Revision: Aug 25, 2006

Copyright © 2003-2006 UNC-CH Proteomics Core Facility, All Rights Reserved.