We conduct research in areas of chemistry, biochemistry, and medicine in which mass spectrometry is the primary analytic tool. Our current research focuses on two areas: the reaction energetics of gas phase ions and protein characterization.
Reaction energetics of gas phase ions
Yergey; in collaboration with Blank, Campbell, Stein, Vestal
We work on the relationship between the energy applied to the formation of a gas phase peptide ion and the nature of its fragmentation. The type and extent of fragmentation are the determining factors of tandem mass spectrometry (MS-MS) spectra, which provide the foundation on which mass-spectrometric characterization of proteins is based. We are using matrix-assisted laser desorption ionization (MALDI) of peptides as a model system to study peptide ion fragmentation. The study of ion energetics relationships between laser fluence and peptide ion fragmentation is fundamental to optimizing MALDI TOF/TOF (time-of-flight) experiments for the purpose of peptide sequencing. In our studies, we obtain peptide fragmentation spectra, obtained typically with 5,000 laser shots, in both the unimolecular decomposition and collision-induced dissociation (CID) modes. We have the ability to follow two time points easily for each peptide decomposition, i.e., the in-source fragmentation consisting of ions formed within 1 microsecond after the laser firing and the slower, mass-dependent fragmentation occurring within the instrument’s collision cell.
We used the fragmentation of YGGFL, the model peptide leucine enkephalin (LeuEnk), over the full range of laser fluence as the basis for our initial studies. While not a peptide of the type normally encountered in protein characterizations, LeuEnk is an excellent model for studies of short-lived processes in the laser plume. We acquired LeuEnk fragmentation spectra in both MS and MS-MS by using three common matrices and a laser pulse length of 600 psec rather than the 5 nsec previously used. We acquired spectra as a function of laser fluence beginning at the onset of ionization and extending to the maximum fluence available in the instrument. The spectra obtained with a-cyano-hydroxycinnamic acid (ACHA) revealed several distinct processes in LeuEnk fragmentation. First, the MS mode spectra showed a region of extensive fragmentation occurring in what must be a very short time frame following the onset of ionization. The rapid fragmentations, leading only to detectable immonium ions, were associated with the laser pulse–induced direct vaporization of molecules from the sample surface. A second set of process took place within the first several hundred nanoseconds following the laser pulse. These processes, also manifest in MS mode, were most likely associated with desorption of LeuEnk ions from particles consisting principally of matrix ablated from the sample surface. The desorbed ions underwent a large number of collisions with the high-temperature gases present in the laser plume and began to fragment; the fragmentations proceeded in a series of consecutive reactions in which the amide backbone bonds ruptured. Our spectra show that the initial direct desorption processes reached a maximum at about 50 percent of the total laser fluence and then increased no further; at that point, the consecutive fragmentation reactions supplanted the direct desorption processes in intensity. Finally, the MS-MS mode spectra exhibited little fragmentation, most likely because the high-energy portions of the energy distributions associated with the second stage, which is the particle desorption process as described above, were depleted. Use of either 2,5-dihydroxy benzoic acid (DHB) or di-methoxy-hydroxy cinnamic acid (sinapinic acid, SA) led to spectra with much less fragmentation than observed with ACHA and a much higher fluence for the onset of the protonated LeuEnk molecule itself.
In contrast to the extensive production of immonium ions observed with ACHA, we observed none at all with the use of SA; with DHB, the level of ions was about 30 percent of those seen with ACHA. The lower levels of immonium ions in SA and DHB versus ACHA were also associated with much lower levels of backbone fragmentation. We also compared ACHA fragmentation with a laser pulse length of 600 psec, whereas the earlier studies employed 5 nsec pulse lengths. The 600 psec pulse length gave rise to the onset of ionization at much lower levels of fluence and was associated with much higher levels of immonium ion formation than was the 5 nsec pulse length. Our observations are almost certainly the result of the dominance of the first mechanism described above, the direct desorption from the sample surface.
Protein characterization
Antal, Backlund, Olson, Vieira, Yergey; in collaboration with Blank, Coorssen, Crouch, DePamphilis, Epstein, Garland , Harrington, Howard, Humphrey, Leppla, Robbins, Rouault, Sackett, Schneerson, Schuck, Sheeley, Spellman, Spiegel, Zimmerberg
We conduct research on the mass-spectrometric characterization of proteins, both collaboratively with groups in NICHD and independently. To identify unknown proteins, we query genomic databases by using the MS data to determine whether any of the protein sequences present in the databases have expected proteolytic cleavage products with theoretical masses that match the empirically determined masses of the peptides generated from the unknown. We take advantage of three mass-spectrometric approaches:: Matrix Assisted Laser Desorption Ionization (MALDI) with Time-of-Flight (TOF) mass analysis; liquid chromatography (LC) followed by electrospray ionization with mass analysis in an instrument capable of using fragmentation reactions to generate peptide sequences, i.e., LC-MS/MS; and MALDI followed by tandem TOF analysis to determine peptide sequences from fragment ion spectra. With this combination of approaches, we are confident that, given enough material in a gel band (100 fmole), we can identify a protein that is described in a database.
We have also developed a novel approach to providing sequence information for proteins that are not described in databases as a consequence of database error, incomplete splice variants, or SNPs; such incompleteness is most frequently associated with organisms whose genomes are unknown or partially characterized, e.g., Xenopus laevis. We have termed the approach De Novo Peptide Sequencing through Exhaustive Enumeration of Peptide Compositions (EEPC); the so-called sequence tag for a peptide is found within peptide fragmentation spectra, typically by employing mass accuracies of 0.5-1Da. We measure fragment ion masses to 0.05 Da or better and use ions arising from the decomposition of energetic ions alone, without the collision-induced dissociation upon which all other current methods rely. We then extend a base sequence or match potential amino acid compositions with those in a novel database generated specifically for this work and consisting of an exhaustive list of all amino acid compositions up to a maximum of 2 kDa. The database, a length-indexed peptide composition lookup table (LIPCUT), is indexed by both peptide length and mass to facilitate access during execution of the extension or the matching of de novo algorithms; to date, we have devised five algorithms of which two, a simple extension approach and a bit-mapped matching approach, are slated for further development. Using spectra from the MALDI tandem TOF instrument, we were able to obtain correct sequences from several protein sources for 21 peptides ranging in length from seven to 15 residues. Our approach gives better results than several commercially available de novo algorithms. A sample-spotting robot enables spatial separations of LC eluents onto a MALDI target plate, thus reducing the number of near-isobaric interferences and permitting more extensive coverage of the protein under investigation.
As regards protein identification and sequencing, we have made progress in characterizing the C-terminal post-translational modifications of tubulins, specifically implementing improvements in the cyanogen bromide digestion protocol and in software for assigning glycation, glutamylation, and de-tyrosination mass-spectral peaks within the families of both alpha- and beta-tubulins in samples containing several isotypes. We were thus able to assign more than 60 peaks in spectra of rat brain tubulins, which are second only to those of the testis in complexity. In addition, we improved the detection of phosphorylation sites by using our differential MALDI spectra approach that compares positive and negative ion spectra. The method employs the esterification of carboxylic acid sites by using methanolic HCl; the esterified acidic residues do not ionize efficiently in negative ion MALDI, but the phosphorylated peptides are unaffected. Preliminary studies suggest that the method may allow us to quantify endogenous levels of phosphorylation.
Characterizing the protein mass fingerprints of amniotic fluid from patients who have undergone premature labor has progressed to the stage of identifying patients with evidence of a bacterial infection along with premature labor. We are attempting to differentiate premature labor leading to pre-term delivery from labor that does not lead to pre-term delivery. We compare MALDI mass spectra in the range of 2 to 10 kDa obtained from amniotic fluid samples that have been desalted and then applied directly to the MALDI sample stage. We have developed a mathematical/statistical approach in MatLab to automate both ANOVA and Principal Component Analysis and to differentiate reliably between classes of samples. We are using these mass-spectrometric and mathematical approaches to characterize cells isolated by an improved form of laser capture microdissection.
Harrington PdeB, Vieira NE, Espinoza J, Nien JK, Romero R, Yergey AL. Analysis of variance/principal component analysis: a soft tool for proteomic discovery. Anal Chem Acta 2005;544:118-127.
Sergeev YV, Soustov LV, Chelnokov EV, Bityurin NM, Backlund PS, Wingfield PT, Ostrovsky MA, Hejtmancik JF. Increased sensitivity of amino-arm truncated betaA3-crystallin to UV-light induced photo-aggregation. Invest Ophthalmol Vis Sci 2005;46:3263-3273.
Vance BA, Harley PH, Backlund PS, Ward Y, Phelps TL, Gress RE. Human CD69 associates with an N-terminal fragment of calreticulin at the cell surface. Arch Biochem Biophys 2005;438:11-20.
Wassif CA, Krakowiak PA, Wright BS, Gewandter JS, Sterner AL, Javitt N, Yergey AL, Porter FD. Residual cholesterol synthesis and simvastatin induction of cholesterol synthesis in Smith-Lemli-Opitz syndrome fibroblasts. Mol Gen Metabol 2005;85:96-107.
Yocum AK, Oe T, Yergey AL, Blair IA. Novel lipid hydroperoxide-derived hemoglobin histidine adducts as biomarkers of oxidative stress. J Mass Spectrom 2005;40:754-764.
Collaborators
Paul Blank, PhD, Laboratory of Cellular and Molecular Biophysics, NICHD, Bethesda, MD
Jennifer Campbell, PhD, Applied Biosystems, Framingham, MA
Jens Coorssen, PhD, University of Calgary, Calgary, Canada
Robert Crouch, PhD, Laboratory of Molecular Genetics, NICHD, Bethesda, MD
Melvin DePamphilis, PhD, Laboratory of Molecular Growth Regulation, NICHD, Bethesda, MD
Jonathan Epstein, MS, Unit on Biologic Computation, NICHD, Bethesda, MD
Donita Garland, PhD, Laboratory of Mechanisms of Ocular Disease, NEI, Bethesda, MD
Peter de B. Harrington, PhD, Ohio University, Athens, OH
Bruce Howard, MD, Laboratory of Molecular Growth Regulation, NICHD, Bethesda, MD
Glen Humphrey, PhD, Laboratory of Cellular and Molecular Biophysics, NICHD, Bethesda, MD
David Klein, PhD, Section on Neuroendocrinology, NICHD, Bethesda, MD
Steven Leppla, PhD, Bacterial Toxins and Therapeutics Section, NIAID, Bethesda, MD
Thomas Neubert, PhD, Skirball Institue of Biomolecular Medicine, New York University School of Medicine, New York, NY
Juan Rivera, PhD, Molecular Immunology and Inflammation Branch, NIAMS, Bethesda, MD
John Robbins, MD, Laboratory of Developmental and Molecular Immunity, NICHD, Bethesda, MD
Tracey Rouault, MD, Cell Biology and Metabolism Branch, NICHD, Bethesda, MD
James Russell, DVM, Section on Cell Biology and Signal Transduction, NICHD, Bethesda, MD
Dan Sackett, PhD, Laboratory of Integrative and Medical Biophysics, NICHD, Bethesda, MD
Rachel Schneerson, MD, Laboratory of Developmental and Molecular Immunity, NICHD, Bethesda, MD
Peter Schuck, PhD, Division of Bioengineering and Physical Science, ORS, NIH, Bethesda, MD
Douglas Sheeley, ScD, Division of Biomedical Technology, NCRR, Bethesda, MD
Dan Spellman, BS, Skirball Institute of Biomolecular Medicine, New York University School of Medicine, NY
Alan Spiegel, MD, Director, NIDDK, Bethesda, MD
Stephen Stein, PhD, MS Data Center, NIST, Gaithersburg, MD
Gisela Storz, PhD, Cell Biology and Metabolism Branch, NICHD, Bethesda, MD
Akos Vertes, PhD, George Washington University, Washington, DC
Marvin Vestal, PhD, Virgin Instruments, Framingham, MA
Joshua Zimmerberg, PhD, MD, Laboratory of Cellular and Molecular Biophysics, NICHD, Bethesda, MD
For further information, contact aly@helix.nih.gov.