Top Down Proteomics

In the Top Down Proteomics platform, a whole cell lysate or nuclear extract go through three different separation steps prior to analysis by mass spectrometry.  The first separation (solution isoelectric focusing) separates the proteins in the sample based on their pI.  Each of the resulting fractions is then separated by GELFrEE (a size based separation).  The next set of fractions are cleaned to remove SDS and separated by reversed phase liquid chromotography prior to MS analysis.  Protein identification and characterization are completed with ProSightPC.

Differences in phosphorylation for proteins identified in asynchronous and M-phase arrested HeLa S3 cells. Top spectra show the identification of proteins in asynchronous HeLa proteins with the phosphorylations designated in red. Lower spectra show the identification of the same proteins in M phase arrested HeLa proteins are identified, again with phosphorylations designated in red. Dynamic changes in phosphorylation between asynchronous and M phase arrest HeLa cells were clearly observed. A fragment map is displayed for the doubly phosphorylated 60S acidic ribosomal protein. Phosphorylated serines are noted in red.

Visualization of 4 mg yeast protein separation by isoelectric focusing (red panel) and subsequent GELFrEE separation of each sIEF fraction (brown panel) by SDS-PAGE. Fractions were eluted off the GELFrEE apparatus and a portion of each fraction was loaded onto gel for visualization.

The worlds first 12 Tesla LTQ-FT Ultra. This 12 Tesla superconducting magnet is coupled to a standard LTQ ion trap and has the capacity for CID, IRMPD and ECD fragmentation. Both 1 mm and nanoLC are options for front end separation. The Kelleher group is also equipped with two 7 Tesla LTQ-FTs.

Natural Products Discovery and BioSynthesis

a) Microbial strains are grown in liquid culture. b) The soluble proteome of the bacterial strain is subjected to shotgun proteomics or 1-dimensional SDS-PAGE to scan for large protein bands for in-gel digestion. c) LC-FTMS analysis is conducted on the resulting peptide mixture, with isolation of expressed T-domain active-site peptides identified by the Ppant ejection assay. d) De novo peptide sequences or database analysis software are used to generate primers to amplify and sequence portions of the expressed biosynthetic gene cluster. e) The gene cluster is identified and sequenced, which informs targeted detection of the natural product produced as depicted in panel f.


The sequence of amino acids added to an NRPS can be determined by analysis of the T domain. Fragmentation of an aminoacylated form of the T domain yields both the pantetheinyl and the phosphopantetheinyl fragment with amino acids attached. (See Dorrestein, P.C. et al, Biochem., 2006, 12756-12766)

Non-ribosomal Peptide Synthetases (NRPS) are comprised of a series of domains with specific functions. The adenylation (A) domain is responsible for the selection/activation of amino acid monomers, the thiolation (T) domain holds the intermediate structure, passing it to the next catalytic site, the condensation (C) domain, which is responsible for the formation of the peptide bond with the upstream thiolation domain. Among other tailoring domains in NRPS synthesis are epimerization (E) domains, responsible for converting the loaded amino acid to its enantiomer and thioesterase (TE) domains, responsible for the release of the NRPS. The Kelleher group analyzes these biosynthetic processes and the kinetics of this biosynthesis, along with discovering new pathways and natural products and their biosynthetic pathways using microbial proteomics and the PrISM workflow described.

Chromatin Oncobiology and DNA-Damage

Credit: Felsenfeld and Groudine, Nature, 2003


Identification of H2A.X in aysynchronous and m-phase arrested HeLa. An increase in phosphorylation can be seen during m-phase arrest (red peak). Confocal microscopy images were also taken at each cellular state, staining with the antibody of phosphorylated H2A.X. An increase in phosphorylation in the m-phase arrested cells is also seen through microscopy.

A) Online separation of mammalian histones.  B) MS profiles of selected histones.

A) Online LC-MS chromatogram of oxidized histone extract.  B) MS data of histones H2B, H4, H2A and H3, obtained using a 7T LTQ FT.  Acetylation forms are labelled.  Histone H3 shows increased acetylation.

Computational Proteomics

To search Top Down data, the intact mass of a protein is queried against our Shotgun Annotated proteome database to select a subset of the total proteins in the database. For each of these candidates, a Poisson-based expectation score is calculated by comparing the predicted protein fragments with the observed fragments.

Capturing Biological Complexity. “Shotgun Annotation” captures the known biological complexity available in the UniProt database. From one base gene, many protein products can be formed. Variation arises from different splice forms, coding polymorphisms and post-translational modifications. All of this complexity is captured in our proteome databases.