The PNNL Omics Separations and Mass Spectrometry group develops advanced chromatographic and other separations methods coupled with high resolution mass spectrometry (MS) and applies these in studies of a variety of biological systems (see Bio-Applications pages for examples). These measurement capabilities include global (i.e. untargeted) proteomics (e.g. bottom up, top down, and measurements of various post-translational modifications, such as phosphorylation, glycosylation, etc), metabolomics, lipidomics, and glycomics. We also have interests in targeted measurements, such as targeted proteomics, in applications such as biomarker validation. All of these rely on capillary chromatographic (either liquid or gas) or gas-phase (such as ion mobility spectrometry (IMS)) separations coupled with high resolution mass spectrometry, including time-of-flight (TOF), Orbitrap, and Fourier transform ion cyclotron resonance (FTICR).
Bottom Up Proteomics
Our bottom up quantitative proteomics strategies are diverse and include traditional spectral counting, 16O/18O labeling, isobaric chemical labeling approaches such as iTRAQ and TMT, and the PNNL-developed accurate mass and time (AMT) tag approach (reviewed in Zimmer et al., Mass Spectrom Rev, 2006). These analyses utilize high-efficiency, high-peak capacity capillary liquid chromatography (LC) separations systems coupled with advanced electrospray ionization sources to mass spectrometers to provide quantitative measurements over a dynamic range of at least 103 to 104 without fractionation, and 104 to 106 with fractionation (e.g. high pH reversed-phase chromatography). The flexibility needed for our diverse bottom up proteomics strategies and the need for more sensitive and higher throughput 2D LC separations to "dig deeper" was a primary driver for developing our advanced platforms. Key developments include: 1) an in-house designed capillary LC systems that adopts a flexible plug-and-play-based configuration that allows us to fully automate new configurations in hours rather than days or weeks (Livesay et al., Anal Chem, 2008), 2) in collaboration with the Valco Instrument Company, the ability to make extremely low dead-volume connections with very small-diameter capillary columns, which is vital for achieving high separation power and ultimately, improved proteome coverage, 3) in collaboration with Agilent Technologies, the incorporation of nanoflow LC pumps that can operate at 1200 bar (18,000 psi) for use in our custom LC platforms, 4) an automated, on-line 2D capillary LC system that provides an enormous increase in proteomics data from small samples, and 5) offline chromatographic fractionation (e.g. strong cation exchange or high pH reversed-phase; Yang et al., Expert Rev Proteomics, 2012) of protein digests.
Metabolomics and Lipidomics
Our metabolomics and lipidomics approaches are coupled and begin with an integrated sample extraction protocol that also isolates proteins suitable for high quality proteomics analyses. Based on our previous studies of cells, tissues, and biofluids, we find that the Folch method is amenable for the extraction of the metabolome and lipidome in untargeted studies. Briefly, samples are treated with chloroform/methanol (2:1, v/v) in a 5-fold excess to the sample volume, followed by agitation to thoroughly mix the solution. Samples are then centrifuged to separate precipitated protein from water- and lipid-soluble metabolites. The aqueous and lipid fractions are removed and dried in vacuo.
For polar metabolite analysis, residues from aqueous fractions are chemically derivatized prior to analysis by gas chromatograpy (GC)-MS according to a method developed by the Oliver Fiehn laboratory (Kind et al., Anal Chem, 2009). GC is unsurpassed in terms of separation peak capacity, which is a measure of the number of chromatographic peaks that can be baseline-resolved within the analysis time. Because the stationary phase is coated on the inside of the unpacked GC column, system back pressure is negligible, and separations are typically performed on 30 m capillary columns at carrier gas flow rates of ~1 L/min, leading to highly efficient separations with chromatographic peak widths of a few seconds. In addition, standardization of the electron ionization source to 70 eV by the GC-MS industry, coupled with its overall high reproducibility has enabled development and utilization of commercial and custom mass spectral databases for high-throughput and automated metabolite identifications. Because GC-MS analyses are performed in the gas phase and chemical derivatization is required to increase the volatility of sample constituents, it is not amenable to the analysis of relatively large biomolecules (e.g., >500 Da), as they either exceed the upper mass limit of the mass spectrometer (~1000 m/z for some instruments) or cannot be made volatile within the temperature limits of the instrument. However, GC-MS has proven to be an exceptional tool for the untargeted, quantitative analysis of small, polar metabolites (e.g., amino and small organic acids, sugars, etc.) and fatty acids.
For lipid analysis, dried lipid extracts are reconstituted in isopropanol and analyzed directly using capillary LC coupled with high resolution MS and employing a combined top down/bottom up lipidomics approach (Gao et al., Anal Bioanal Chem, 2012). With this approach, complex lipid molecular species are separated using reversed-phase chromatography and identified based on accurate mass and characteristic fragment ions in sequential full scan and tandem (MS/MS) mass spectra. In addition to adding confidence to the identification of detected lipids, the high mass resolution provided by TOF and Orbitrap MS instruments increases the coverage of the lipidome. We have recently developed a novel software, LIQUID (Lipid Informed Quantitation and Identification) for identification of lipids detected in LC-MS/MS-based lipidomics analyses. For more details on LIQUID, see the Algorithm Development page.
We utilize a variety of state-of-the-art high resolution LC or GC separations combined with a diverse base of mass spectrometry (MS) instrumentation to support the measurement capabilities described above. Over 40 mobile LC systems of various configurations (e.g., specific to lipids, peptides, proteins, glycans, etc.) can readily be coupled with over 30 MS systems that include quadrupole, triple quadrupole, IMS-TOF, Orbitrap, and FTICR. A core team of highly experienced instrument staff assures LC-MS and GC-MS instrumentation is properly maintained and follow quality control (QC) strategies that are specific to the analysis type and project requirements. An in-house developed laboratory information management system (PRISM; Kiebel et al., Proteomics, 2006) is integrated with measurement systems, reducing the risk of errors associated with manually entered sample information and allows for near real-time instrument data file capture, analysis, and archival.
We also have two fully equipped BSL2 lab spaces and a dedicated BSL1 lab space that enable a wide variety of applications in human health and disease, systems biology, and environmental science. Our emphasis on high-throughput and accommodating a diversity of sample types has resulted in significant investment in laboratory automation and building a robust infrastructure for sample management. A modular approach to automation allows for workflow flexibility and utilizes two EpMotion liquid handling systems, King Fisher magnetic bead handler, 2 Gilson automated SPE stations, 6 Agilent 1100 HPLC’s with auto-samplers and fraction collectors, and a wide variety of equipment for biological sample homogenization. These sample preparation labs are staffed by knowledgeable staff that bring decades of experience in preparing biological materials for analysis by mass spectrometry. All metadata associated with sample processing and storage is tracked with electronic note taking and our LIMS.