Search

Proteomics has enabled the direct investigation of biological material, at first through the analysis of individual proteins, then of lysates from cell cultures, and finally of extracts from tissues and biopsies from entire organisms. Its latest manifestation - quantitative proteomics - allows deeper insight into biological systems. This article reviews the different methods used to extract quantitative information from mass spectra. It follows the technical developments aimed toward global proteomics, the attempt to characterize every expressed protein in a cell by at least one peptide. When applications of the technology are discussed, the focus is placed on yeast biology. In particular, differential quantitative proteomics, the comparison between an experiment and its control, is very discriminating for proteins involved in the process being studied. When trying to understand biological processes on a molecular level, differential quantitative proteomics tends to give a clearer picture than global transcription analyses. As a result, MS has become an even more indispensable tool for biochemically motivated biological research.

Several label-free quantitation strategies have been introduced that obliterate the need for expensive isotopically labeled molecules. However label-free approaches have considerably higher demands in respect of repeatability of sample preparation and fractionation than multiplexing isotope labeling-based strategies. OFFGEL fractionation promises the necessary separation efficiency and repeatability. To test this platform, 12-fraction peptide OFFGEL electrophoresis and online reversed-phase LC connected to a quadrupole TOF mass spectrometer were used to determine differences of the physiological, pathological and biochemical distinct extraocular muscle allotype in comparison to hind-limb muscle. Close to 70% of the peptides separated by OFFGEL electrophoresis were detected only in a single fraction. To determine the separation repeatability of four samples, we compared the ion volumes of multiple peptides deriving from the thick filament-associated protein titin over several fractions and determined a coefficient of variation below 20%. Of the 474 proteins identified, 61 proteins were differently expressed between the two muscle allotypes and were involved in metabolism, muscle contraction, stress response, or gene expression. Several expression differences were validated using immunohistochemistry and Western blot analysis. We therefore consider peptide OFFGEL fractionation an effective and efficient addition to our label-free quantitative proteomics workflow.

Rearrangements in the actin cytoskeleton play a pivotal role for costimulation-induced formation of the immunological synapse and T cell activation. Yet, little is known about the actin-binding proteins that link costimulation to rearrangements in the actin cytoskeleton. Here we demonstrate that phosphorylation of the actin bundling protein L-plastin in response to costimulation through TCR/CD3 plus CD2 or CD28, respectively, is important for the activation of human peripheral blood T lymphocytes (PBT). Mass spectrometry and site-directed mutagenesis revealed that Ser5 represents the only phospho-acceptor site of L-plastin in PBT. Wild-type L-plastin (wt-LPL) and a non-phosphorylatable 5A-L-plastin (5A-LPL) equally relocalized to the immunological synapse between PBT and APC. Yet importantly, cells expressing 5A-LPL showed a significantly lower expression of the T cell activation molecules CD25 and CD69 on the cell surface than cells expressing wt-LPL. This effect is due to a failure in the transport of CD25 and CD69 to the cell surface since the total amount of these proteins within the cells remained unchanged. In conclusion, phosphorylation of the actin bundling protein L-plastin represents a so-far-unknown mechanism by which costimulation controls the transport of activation receptors to the T cell surface.

MS protein identification and quantitation are key proteomic techniques in biological research. Besides identification of proteins, MS is used increasingly to characterize secondary protein modifications. This often requires trimming the analytical strategy to a specific type of modification. Direct analysis of protein modifications in proteomic samples is often hampered by the limited dynamic range of current analytical tools. Here we present a fast, sensitive, multiplexed precursor ion scanning mode--implemented on a quadrupole-TOF instrument--that allows the specific detection of any modified peptide or molecule that reveals itself by a specific fragment ion or pattern of fragment ions within a complex proteomic sample. The high mass accuracy of the TOF mass spectrometer is available for the marker ion specificity and the precursor ion mass determination. The method is compatible with chromatographic separation. Fragment ions and intact molecular ions are acquired quasi-simultaneously by continuously switching the collision energy between elevated and low levels. Using this technique many secondary modifications can be analyzed in parallel; however, the number of peptides carrying a specific modification that can be analyzed successfully is limited by the chromatographic resolution or, more generally, by the depth of the resolved time domain.

In two-dimensional gel electrophoresis-based proteomic experiments matrix-assisted laser desorption/ionization-mass spectrometry (MALDI-MS) peptide mass fingerprinting is often the technique of choice in identifying proteins. Here, we present a novel surface coating technique for MALDI-MS targets that improves manual and automatic sample analysis. A mixture of silicone and graphite is spread in the form of a thin layer over the target. Due to the hydrophobicity of the coating, aqueous solutions can be applied to relatively small spots very precisely using a robotic system. At least four times more liquid can be concentrated on the same area compared to uncoated steel targets. alpha-cyano-4-hydrocinnamic acid crystallizes in form of very small crystals evenly distributed over the surface. The search for "hot spots" during the analysis is not necessary, which supports the automatic acquisition of data. The homogeneous crystal layer can be very effectively washed on-target without encountering major sample losses. This efficient washing and the focused application of aqueous samples replace expensive and time-consuming reversed phase micro column based sample clean-ups. When analyzing peptide mixtures, the signal intensities are up to five times higher than with preparations of the same un-desalted samples on steel targets, since four times more sample can be loaded. The mass resolution remains unaffected by the surface coating. After usage the coating can be removed, followed by a new coating avoiding any carry-over of sample to the next analysis. All these properties make the precoating of MALDI-MS targets with a silicone/graphite layer an ideal technique for routine analysis in large-scale proteomic experiments.

Liquid chromatography tandem mass spectrometry is a major tool for identifying proteins. The fragment spectra of peptides can be interpreted automatically in conjunction with a sequence database search. With the development of powerful automatic search engines, research now focuses on optimizing the result returned from database searches. We present a series of preprocessing steps for fragment spectra to increase the accuracy and specificity of automatic database searches. After processing, the correct amino acid sequences from the database can be related better to the fragment spectra. This increases the sensitivity and reliability of protein identifications, especially with very large genomic databanks, and can be important for the systematic characterization of post-translational modifications.

Despite the progress in genomic DNA sequencing de novo sequencing of peptides is still required in a biological research environment since many experiments are done in organisms whose genomes are not sequenced. A way to unambiguously retrieve a peptide sequence from a tandem mass spectrum is to assign the correct ion type to the fragments. Here we describe a method which improves the specificity in y-ion assignment throughout the spectrum. The differential scanning technique requires that the peptides are partially 18O labelled at their C-terminus and that two fragment spectra are acquired for each peptide, one selecting the 16O/18O isotopic cluster and a second fragmenting only the 18O labelled ions. When the spectra are acquired with a quadrupole time of flight mass spectrometer y-ions can be very specifically filtered from the spectrum using a computer algorithm. Partial or complete peptide sequences can be assigned automatically simply by finding the most abundant series of fragments spaced by amino acid residue masses. This method was used extensively in a project investigating vesicular transport in bovine brain cells. Human or mouse homologues to the bovine proteins were found in EST databases facilitating rapid cloning of the human homologues.