New Technology for the Large-scale Proteomic Comparison of Human Embryonic Stem Cells, Induced Pluripotent Stem Cells, and Somatic Cells

D. Phanstiel¹, J. Brumbaugh¹, G. C. McAlister¹, C. D. Wenger¹, R. Stewart², S. Tian², J. A. Thomson², J. J. Coon¹

¹Department of Chemistry and ²Morgridge Institute for Research, University of Wisconsin, Madison, WI

Induced puripotent stem cells (iPS) represent a breakthrough in stem cell research and pose remarkable therapeutic potential. They circumvent ethical issues surrounding the use of human embryonic stem cells (ES) and could eliminate immune rejection in transplantation therapies. In this work we investigate similarities and differences between human ES and iPS cells that may affect the use of iPS cells for research and therapeutic purposes. To perform these experiments we use novel mass spectrometry-based analyses in combination with isobaric tags for absolute and relative quantitation (iTRAQ). To date, we have identified 77,959 unique peptides and 6,534 unique proteins. The fold-differences observed between pluripotent lines and the somatic line showed a high correlation (R = 0.91) demonstrating remarkable similarity between ES and iPS cells. The set of proteins that was expressed at higher levels in pluripotent cells was mostly nuclear in nature and functionally enriched in the processes of transcriptional regulation and chromatin modification. Among these were numerous transcription factors and other proteins important to the maintenance of pluripotency including SOX2, OCT4, DPPA4, and LIN28. In agreement with current literature, comparison of protein expression changes to changes in mRNA expression revealed only a weak correlation (R = 0.68) highlighting the need for quantitative proteomic analysis.

MS.2

Electron Capture Dissociation in Radio-Frequency-Free Cell

D. F. Barofsky¹, V. G. Voinov^1,3, M. L. Deinzer¹, J. S. Beckman²

Departments of ¹Chemistry and ²Biochemistry and Biophysics, Oregon State University, Corvallis, OR; ³Pacific Institute of Bioorganic Chemistry, Vladivostok, Russia

A radio frequency-free (RFF), analyzer-independent cell has been devised for electron-capture dissociation (ECD) of ions. The device is based on interleaving a series of electrostatic lenses with the periodic structure of magnetostatic lenses commonly found in a traveling wave tube. A five-magnet version of the RFF electromagnetostatic ECD cell was installed in a Finnigan TSQ700 ESI triple quadrupole (QqQ) spectrometer, and its performance was evaluated by recording product-ion spectra of various peptides. These spectra were readily obtained without recourse to a buffering gas or synchronizing electron injection with a specific phase of an RF field. The mass spectra produced with the modified instrument appear in all respects (other than resolution and mass accuracy, which were limited by the mass spectrometer used) to be at least as good for purposes of peptide identification as those recorded with Fourier transform ion cyclotron resonance (FT ICR) instruments; however, the effort and time to produce the mass spectra were much less than required to produce their FT ICR counterparts. A two-magnet version of the electromagnetostatic ECD cell was installed in the same mass spectrometer and used to simultaneously obtain combined ECD/CID product-ion mass spectra that exhibit a-, b-, and c-type ion signals. Details of the cells design, construction, and operation will be presented and discussed.

MS.3

Decoding the Histone Code by Quantitative Proteomics

G. LeRoy¹, M. D. Plazas-Mayorca², N. L. Young¹, and B. A. Garcia¹

Departments of ¹Molecular Biology and ²Chemistry, Princeton University, Princeton, NJ

Epigenetic refers to stable heritable changes in gene expression that are not due to changes in DNA sequence, such as DNA methylation, RNA interference and histone modifications. Histones are small basic proteins that function to package genomic DNA into repeating nucleosomal units (containing ~146 bp of DNA wrapped around two copies of each of histones H3, H4, H2A and H2B) forming the chromatin fiber and hence our chromosomes. In general, the packaging of DNA into chromatin is recognized to be a major mechanism by which the access of genomic DNA is restricted. A wide number of studies show that several covalent histone modifications such as methylation, acetylation, phosphorylation and ubiquitination located in the N-terminal tails correlate with both the regulation of chromatin structure during active gene expression, Oregon, USA heterochromatin formation during gene silencing. Here we are developing novel proteomic strategies to discover differentially expressed histone modifications, identify concurrent combinatorial histone modifications (Histone Codes), and characterize histone codes that are important various processes.

MS.4

Data Processing Algorithms for Analysis of High Resolution MSMS Spectra of Peptides with Complex Post-Translational Modifications

S. Guan, F. Li, S. Eliuk, and A. L. Burlingame

Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA

Data analysis of fragmentation spectra of heavily modified large peptides and small proteins is a challenging task. Efficient fragmentation method, such as electron capture dissociation (ECD) or electron transfer dissociation (ETD), allows for more uniform cleavage along all peptide backbones and for retention of many labile modification groups. However, MSMS spectra from large peptides may contain hundreds perhaps thousands of often overlapping isotopic clusters. We have developed a strategy to directly match the theoretical isotopic distributions to the experimental data. The so called Fragment Assignment by Visual Assistance (FAVA) program detects more fragment ions than by use of deisotopic algorithms.

Heavily modified peptides such as histones, add additional difficulty for the data analysis. Many MSMS spectra come from a mixture of peptides with the same molecular weights but different PTM arrangement. In order to distinguish those PTM "variants" and to estimate their relative stoichiometry, we have extended our FAVA algorithms to examine possible PTM combinations.

We have tested our processing programs with large peptides from Histone samples. Those peptides contain up to 12 lysine methylations. Acetylations and trimethylations can be differentiated by mass accuracy and identification and stoichiometry determination accuracies can be estimated from numbers of unique ions identified and from their intensities.

Financial support was provided by NIH NCRR P41RR001614, NIH NCRR P41RR001614 and NIH NCRR RR019934.

MS.5

Use of Electron Transfer Dissociation to Analyze Combinations of Histone Post-Translational Modifications on an LTQ-Orbitrap

S. Eliuk¹, D. Maltby¹, F. Chu³, B. Panning², and A. L. Burlingame¹

¹Mass Spectrometry Facility, Department of Pharmaceutical Chemistry and ²Department of Biochemistry & Biophysics, University of California, San Francisco, CA; ³College of Life Sciences and Agriculture, University of New Hampshire, Durham, NH

Post-translational modifications of histones are used to regulate DNA-chromatin interactions and ultimately gene expression. In many cases, the various histone PTMs do not regulate function individually. Instead, combinations of modifications are believed to act together to create a ‘histone code’. Elucidating these combinations through analyses of their proteolytic digests since is often not possible as much of this combinatorial information is lost following protein digestion. It is essential to be able to detect and assign the sites of these modifications in combination on particular protein isoforms to assess their significance.

Accordingly, we have developed approaches for both direct infusion and online reverse phase liquid chromatography separation for intact histone and large histone peptides (AspN or GluC digests) on an LTQ Orbitrap with electron transfer dissociation (ETD) fragmentation.

We have applied these methods for the analysis of post-translational modifications of histones in methyltransferase knockout mouse embryonic stem cells compared with wildtype cells. Combined with stable isotope labeling of amino acids in cell culture (SILAC), changes in stoichiometries of modification site occupancies are revealed by quantitative mass spectrometry. Through these techniques, we show that knock out of a specific methyltransferase leads to a variety of histone modification changes.

Support for this research was provided by the Bio-Organic Biomedical Mass Spectrometry Resource at UCSF (A. L. Burlingame, Director) through the Biomedical Research Technology Program of the NIH National Center for Research Resources, NIH NCRR P41RR001614 and S10RR019934.

MS.6

Electron Capture Dissociation for Structural Studies of Integral Membrane Proteins and Their Modifications

J. Whitelegge

The Pasarow Mass Spectrometry Laboratory, The NPI-Semel Institute, David Geffen School of Medicine, University of California, Los Angeles, CA

Full structural analysis of the c-subunit of the Fo domain of ATP synthase demonstrated that the integral membrane proteins that constitute around one third of the proteome are amenable to top-down high-resolution mass spectrometry. Effective electron capture dissociation (ECD) was achieved through thermal excitation of ions using an infra-red laser for activated-ion, or aiECD, implying that it was necessary to break alpha-helix hydrogen bonds prior to electron capture. Collisionally activated dissociation (CAD) was also effective for top-down analysis of the c-subunit but aiECD clearly improved sequence coverage, allowing substantial sequence coverage within the two alpha-helical transmembrane domains. The need to use thermal excitation for effective ECD of this 8 kD protein suggested that integral membrane proteins may have a lower threshold for this requirement than soluble proteins. Subsequent studies of a variety of integral membrane proteins have not been consistent enough to answer this question. The most consistent feature of these experiments has been the frequent observation that while CAD yields useful product ion spectra ECD most often yields familiar charge reduction series even when IR laser is used to activate the ions. The origin of this problem seems to be the fact that integral membrane proteins typically carry fewer charges than soluble proteins after electrospray ionization giving them higher m/z. Strategies to achieve higher charging will be discussed.

MS.7

O-GlcNAcylation: The Post-Translational Modification that Best Highlights the Value of ETD

R. Chalkley

Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California San Francisco, CA

O-GlcNAcylation is a widespread regulatory modification of nuclear and cytoplasmic proteins, analogous to phosphorylation. For a modification that was first discovered 25 years ago, surprisingly little is known about how it regulates protein functions, despite its clear role in problems such as diabetes and Alzheimer disease. Slow progress in characterizing this modification has largely been due to a lack of effective methods for detecting and locating O-GlcNAcylation sites. This is in a large part because the O-glycosidic link that attaches the single N-acetylglucosamine (GlcNAc) moiety to serines and threonines is highly labile under collision induced dissociation in a mass spectrometer (much more so than phosphorylation), so O-GlcNAc site assignment using this mass spectrometric approach has proven largely unsuccessful.

Electron transfer dissociation (ETD) is a recently developed radical-based fragmentation technique that fragments components at locations not defined by bond strength, so is able to maintain labile modifications on fragment ions. Hence, the modification is not cleaved during ETD mass spectrometric analysis, allowing O-GlcNAcylation site assignment.

In this talk I will present how ETD availability, along with strategies for enriching for modified peptides, is starting to transform the characterization of this modification.

This work was supported by NIH NCRR grant RR001614, SIG RR019934 and the Biotechnology and Biological Sciences Research Council of the UK.

MS.8

The Use of ECD for Proteomics-wide Identification and Quantification of iso-Asp Residues

H. Yang, Y. M. E. Fung, A. R. Zubarev, and R. A. Zubarev

Division of Molecular Biometry, Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, Sweden

Protein deamidation is one of the major post-translational modifications (PTM) that lead to protein inactivation in vivo. It has been shown to relate with neurodegenerative diseases (e.g. Alzheimer's disease) and the degradation of commercial protein products. Asparaginyl residue is a major source of deamidation in biological samples, the minor one being the glutaminyl residue. Asparagine deamidation is a non-enzymatic PTM that occurs spontaneously under physiological conditions, which results in a mixture of aspartyl (Asp) and isoaspartyl (isoAsp) residues. Asp isomerization also contributes to the increase in the isoAsp pool, albeit at a lower rate.

Electron capture dissociation (ECD) combined with Fourier transform mass spectrometry (FT MS) are able to distinguish the isoaspartyl peptides by unique ECD fragments of c_n· + 58.0054 (C₂H₂O₂) and z_l-n – 56.9976 (C₂HO₂), where n is the position of the aspartyl residue and l is the peptide length.

In the present study, we tested the specificity of isoAsp detection using the accurate masses of these specific fragments. Totally, 466 isoAsp peptide candidates were identified from 32 whole and partial human proteome samples. Then additional criteria, like adjacent c/z fragments, specific losses from the reduced species, and the shape of the chromatographic peak, were applied to increase the specificity of the method. Upon detailed inspection, 219 isoAsp peptide candidates have been supported by at least one criterion other than the mass of the specific ECD fragments. Most stringent filtering of these candidates yielded several cases where the presence of isoAsp was beyond doubt. Among the identified proteins with isoAsp, actin, heat shock cognate 71 kDa protein and pyruvate kinase have previously been identified as substrates for L-isoaspartyl methyltransferase (PIMT), an important repair enzyme converting in vivo isoaspartyl to aspartyl. Quantification of relative isomerization degree was performed by the label-free approach using in-house developed software. This is the first attempt to analyze the human isoaspartome in a high-throughput manner. The developed workflow allows for further enhancement of the detection rate of isoaspartyl residues in biological samples.

Global Analysis of Small Molecule Interactions with Proteins

X. Li¹ and M. Snyder²

¹Department of Molecular, Cellular and Developmental Biology, Yale University, New Haven, CT; ²Department of Genetics, Stanford University, Stanford, CA

Extensive effort has been made to characterize the interactions of proteins with other proteins, and for the case of transcription factors, their interaction with DNA. Natural small compounds comprise the majority of the cellular molecules, and have been shown to bind proteins as substrates, products, cofactors and regulatory ligands. Although they participate in such diverse processes, a large scale investigation of protein-small molecules has never been performed. We have developed a mass spectrometry assay for the global analysis protein-natural small molecule interactions in yeast, and applied this method to the analysis of molecules that bind lipid pathway biosynthetic proteins and protein kinases. We find that a large number of proteins bind small molecules. Some enzymes bind substrates, others bind their products, and many key regulatory proteins such as protein kinases bind small molecules. One common ligand bound by many proteins is ergosterol suggesting a general role for this compound in regulation. We further explore this by demonstration that the activity of the high conserved AKT protein kinase homolog of yeast depends upon the presence of ergosterol. Overall, our study helps define potential key regulatory steps in biosynthetic pathways and demonstrates that small molecules bind many proteins in a variety of biochemical and regulatory roles and suggest they can serve as regulators of protein activity.

1.2

Membrane-assisted Sample Preparation for Online ESI-MS Analysis of Biomolecules

J. Astorga-Wells^1,2, C. Whitehouse³, and H. Jörnvall¹

¹Karolinska Institutet, Stockholm, Sweden; ²Biomotif AB, Täby, Sweden; ³Analytica of Branford, CT

In addition to further improvements in resolution, mass accuracy, sensitivity and throughput, advances in sample preparation are essential in order to fully unveil the potential of any mass spectrometer. A technology based on ion-selective membranes has been developed in order to perform online sample preparation before electrospray ionization. The technique permits to change the chemical composition of the solution and manipulate analytes a few microliters upstream from the electrospray process. Compounds can be removed or injected from an adjacent channel to the main flow stream through an ion-selective membrane that separates both channels. In this manner, protons can be injected into the flow stream to carry out rapid pH scans; deuterons can be delivered to perform deuterium/hydrogen exchange experiments in order to study protein structural changes; and metal ions can be delivered into the flow stream to study -for example- metal-binding peptides. Furthermore, a device with two membrane sections can be used to introduce an electric filed into the main channel in order to immobilize ions and perform pre-concentration, clean-up, multi-step micro reactions, solvent exchange and separation prior to mass analysis. Examples of each application will be shown as well as preliminary data of other applications.

1.3

Two-Dimensional Liquid Chromatography Coupled with ESI-MS for Protein Identification and Quantification

J. Langridge¹, H. Vissers¹, S. Geromanos³, M. Stapels³, C. Dorschel³, M. V. Gorenstein³, D. Golick³, and H. Aerts²

¹Waters Corporation, MS Technologies Center, Manchester, United Kingdom; ²AMC, University of Amsterdam, Department of Biochemistry, The Netherlands; ³, Waters Corporation, Milford, MA

Mass spectrometry is widely accepted as an essential tool to better understand protein function, facilitating both the identification and quantification of proteins in complex samples. Mass spectrometry based protein identification strategies have previously been described [1–3] that facilitate the simultaneous acquisition of qualitative and quantitative information, Indiana, USA a data independent fashion.

We have extended this approach to generate precise relative quantitation values for proteins contained in biological systems [4–5], and have constructed protein abundance curves for specific tissues, cell lysates and biofluids. This has been shown to be transferrable between different laboratories and independent of instrument type. An important aspect of this quantification approach is that it allows sample loading onto a given analytical column to be determined and optimized, to ensure that ideal chromatographic and mass spectrometric performance is obtained. This results in the maximum number of peptide and proteins being determined from the sample, whilst maintaining maximum accuracy for quantitative measurements. More recently this approach has been extended to cover a wider range of protein abundance by implementing a 2-dimenionsal reverse phase-reverse phase separation strategy, using differential pH. In this manner wide quantitative proteome coverage can be obtained.

Experimental information obtained from such studies will be compared to theoretical models of the given proteome; considering complexity, dynamic range and the inherent physiochemical properties of tryptic peptides in solution and the gas phase.

References

1. Bateman et al. (2002) JASMS 13(7), 792–803.

2. Purvine et al. (2003) Proteomics 3(6), 847–50.

3. Silva et al. (2005) Anal. Chem. 77(7), 2187–200.

4. Silva et al. (2006) Mol. Cell. Proteomics 5(1), 144–56.

5. Hughes et al. (2006) J. Proteome Res. 5(1), 54–63.

1.4

Targeted Proteomic Approaches Provide Insights into Virion Assembly and Chromatin Remodeling during Viral Infection

I. M. Cristea

Department of Molecular Biology, Princeton University, Princeton, NJ

Viruses have evolved finely tuned interactions with their hosts to manipulate and adapt complex cellular processes for their own use. The study of virus-host interactions has therefore emerged as a key driving force in the research of infectious disease during the post-genomic era. Despite these efforts, our understanding of the protein interactome remains, Indiana, USA large part, unknown. The development and incorporation of new approaches that can reveal the dynamics of virus-host protein interactions is a necessity. Modern proteomic techniques have the ability to provide access to such interactions, and the ever increasing sensitivity of mass spectrometry allows the identification and quantification of relatively low levels of proteins. This presentation will describe targeted proteomic approaches for studying virus-host macromolecular assemblies. Highlights will be shown from our studies on infections with human immunodeficiency virus (HIV) and human cytomegalovirus (HCMV).

We employed targeted genetic-proteomic approaches to study the virus-host interface either from the virus or host perspective. Using a library of tagged replication competent HCMV and HIV mutant viruses, we infected primary human fibroblasts (for HCMV) and CEM T cells (for HIV), and employed cryogenic cell lysis and rapid immunoaffinity purifications on magnetic beads to isolate virus-host assemblies. For studies on histone deacetylases (HDAC) during viral infections, we generated cell lines stably expressing green fluorescent protein tagged HDACs and probed their interactions and deacetylation activity. Isolated protein complexes were analyzed using a MALDI LTQ Orbitrap (Thermo Fisher Scientific) and the specificity of observed interactions was confirmed by immunofluorescence, reciprocal immunoprecipitation and metabolic labeling with stable isotopes (I-DIRT).

Two interesting findings will be highlighted: 1) studies on pUL32, pUL99, pUL83 and pTRS1 HCMV proteins demonstrated that parallel processes occur at distinct cellular sites during the assembly of HCMV virions, and 2) chromatin remodeling complexes, including histone deacetylases, are targeted by viruses, possibly in part to gain control over host gene expression and modulate the outcome of an infection.

1.5

Confident Assignment of Post-Translational Modifications Using Top-Down Mass Spectrometry

J. Whitelegge

The Pasarow Mass Spectrometry Laboratory, The NPI-Semel Institute, David Geffen School of Medicine, University of California, Los Angeles, CA

Top-down proteomics uses high-resolution Fourier-transform mass spectrometry (FT-MS) to define proteins by their intact masses in combination with dissociation experiments for unambiguous primary structure determination. FT-MS has been used to characterize the integral and peripheral subunits of the integral membrane Photosystem II complex from the red alga Galdieria sulphuraria. The complex was analyzed by reverse-phase liquid chromatography with online electrospray-ionization mass spectrometry and concomitant fraction collection (LC-MS+). Selected fractions were transferred to static nanospray tips for nano-electrospray ionization on a hybrid linear ion-trap/Fourier-transform ion-cyclotron resonance mass spectrometer (7 Tesla LTQ-FT Ultra; Thermo Scientific) for extended averaging of transients. Collision-activated dissociation (CAD) was achieved in the ion-trap whereas electron-capture dissociation (ECD), IRMPD and activated-ion electron-capture dissociation (aiECD) were performed in the cyclotron cell using an infra-red laser and an electron source. Top-down datasets were deconvoluted using Xtract and matched to a Galdieriaproteome database using Prosight PC (Thermo Scientific). The primary LC-MS+ separation yielded thirty-nine intact mass tags (IMTs) with 100 ppm mass accuracy on a low-resolution instrument. While several subunits could be identified based upon close coincidence of measured and calculated masses, the majority required use of sequence tag functionality of the software to yield candidate identifications for manual assessment. All of the subunits studied carried covalent modifications that required consideration in the analysis, including co-translational N-terminal formylation and post-translational removal of Met1 with or without N-acetylation, removal of signal peptides and attachment of cofactors. Approaches to identify and localize PTMs are considered. The precision afforded through use of FT-MS allows precursor and product ion matching at a 10 ppm tolerance with nearly all assignments achieving less than 5 ppm, such that false positives become highly unlikely. There are however limitations to the accurate assignment of PTMs and these will be illustrated with reference to labile modifications that can be displaced during dissociation. The lessons learned from top-down PTM analyses will be discussed in the context of current strategies for high-throughput bottom-up proteomics.

Protein Quantification through Targeted Mass Spectrometry: The Way Out of Biomarker Purgatory?

S. A. Carr

Broad Institute of MIT and Harvard, Cambridge, MA

The enormous potential of biomarkers to revolutionize clinical practice and improve patient care has been well documented. Given their high potential therapeutic and financial impact it is, on the surface, surprising that so few new protein biomarkers have been introduced into widespread clinical use recently. The reasons for the dearth of new protein biomarkers relate to the high false discovery rate of discovery "omics" methods (regardless of technology used), together with a lack of robust methods for biomarker verification in large clinical sample sets. It is now common for differential analysis of tissue or plasma by multidimensional LC-MS/MS (the workhorse tool for unbiased discovery) to provide confident identification of 1000's of proteins, 100's of which can vary 5-fold or more between case and control samples in discovery studies. Due to the extent of sample fractionation required to access proteins at lower abundance, it is not uncommon for the analysis of a single case/control sample pair to take up to two weeks of on-instrument time. This limits the numbers of samples that can be practically analyzed to typically 10 (or fewer) case vs control comparisons. These numbers are very small relative to the high dimensionality of the proteome (100,000's or more possible components when posttranslational modifications and other variants are taken into account), and the scale of normal variation in the human population. Thus a very large fraction, possibly exceeding 95% of the protein biomarkers "discovered" in these experiments are false positives arising from biological or technical variability. Clearly discovery "omics" experiments do not lead to biomarkers of immediate clinical utility, but rather produce "candidates" that must be "qualified" and "verified" (1,2).

Lack of robust quantitative methods with sufficient sensitivity, reproducibility and throughput has significantly hampered our ability to credential candidates coming from unbiased proteomic discovery efforts since useful Ab reagents for the vast majority do not exist. Our laboratory has focused on addressing this serious barrier by developing robust targeted assay methods employing mass spectrometry to screen and quantify low abundance proteins in plasma. We have recently demonstrated that multiplexed assays for proteins at the low ng/mL level in plasma can be configured using Stable Isotope Dilution (SID) - Multiple Reaction Monitoring (MRM) Mass Spectrometry. Large-scale interlaboratory studies conducted under the auspices of the National Cancer Institute's Clinical Proteomics Assessment for Cancer (CPTAC) program have demonstrated that these assays can be reproducibly configured, deployed and run in multiple laboratories with assay CV's approaching clinical performance. Further improvements will come from the use of peptide immunoaffinity enrichment, referred to as SISCAPA, which holds particular promise for simplifying sample preparation and increasing both throughput and sensitivity of MRM-based assays. Using SISCAPA, assays can be readily configured that enable quantitation of proteins present at low ng/mL levels directly from plasma.

This presentation will focus on the further development and application of MRM-MS and SISCAPA technologies in the context of cancer and cardiovascular disease.

References

1. Rifai, N., Gillette, M. A., and Carr, S. A. (2006) Protein biomarker discovery and validation: the long and uncertain path to clinical utility. Nat. Biotechnol. 24, 971–983.

2. Paulovich, A. G., Whiteaker, J. R., Hoofnagle, A. N., and Wang, P. (2008) The interface between biomarker discovery and clinical validation: The tar pit of the protein biomarker pipeline. Proteom. Clin. Appl. 2, 1386–1402.

2.2

Proteomics Targeted to Sub-Cellular Compartments and Integration with Genomics for Candidate Biomarker Discovery in Colorectal Cancer

C. R. Jimenez

OncoProteomics Laboratory, Department of Medical Oncology, VUmc-Cancer Center Amsterdam, VU University Medical Center, Amsterdam, The Netherlands

Mass spectrometry-based proteomics applied to biomarker-rich sub-cellular compartments in (pre)-clinical samples is emerging as a powerful approach for discovery of tissue-derived biomarkers with close association to the disease. We have applied in-depth proteomics to sub-nuclear compartments, cell surface fractions and tumor secretomes for discovery of colorectal cancer-related proteins with potential use as imaging- and serum-based biomarkers. To this end, we analyzed preclinical (cell lines and genetic mouse model) as well as clinical samples. We have optimized the label-free GeLC-LTQ-FTMS/MS pipeline to allow for quantitation of individual proteins by spectral counting with good reproducibility. Integration of in-depth discovery proteomics with genomics data (microarray and arrayCHG) allows for a powerful approach to prioritize candidates for follow-up and validation by targeted strategies, including immunohistochemistry of tissue microarrays. Promising imageable biomarker candidates have been validated and validation of selected nuclear proteins and secretome differential proteins as serum and stool-based markers is underway.

2.3

Towards the Discovery of Biomarkers in Cerebrospinal Fluid by Combining Peptide Ligand Library Treatment and Label Free Protein Quantification on a LTQ-Orbitrap

F. Roux-Dalvai¹, E. Mouton-Barbosa¹, A. Gonzalez de Peredo¹, D. Bouyssié¹, L. Guerrier², E. Boschetti², F. Berger³, O. Burlet-Schiltz¹, and B. Monsarrat¹

¹CNRS IPBS, Toulouse University, France; ²BioRad, Gif-sur-Yvette, France; ³INSERM, Université Joseph Fourier, Grenoble, France

Cerebrospinal Fluid (CSF) is the biological fluid in closest contact with the brain and thus contains some proteins and other products of neural cell origin, lending itself to proteomic analysis for potential biomarkers of neurological diseases. However, as in the case of other biological fluids, the main analytical challenge in proteomic characterization of the CSF is the very wide concentration range of proteins, largely exceeding the dynamic range of current analytical approaches. The most abundant protein, human serum albumin, constitutes alone around 45% of the total protein content in CSF, which renders extremely difficult the detection of low-abundant species.

Here, we used the peptide ligand library technology Proteominer^TM to reduce the dynamic range of protein concentration in CSF and unmask previously undetected proteins by nanoLC-MS/MS analysis on an LTQ-Orbitrap mass spectrometer. This method was first applied on a large pool of CSF from different sources, with the aim to better characterize the protein content of this fluid, especially for the low abundance components. We were able to identify 1189 proteins in CSF, and among these, 755 were only detected after Proteominer^TM treatment.

One drawback of this method is the large amount of biological fluid that has to be applied on conventional Proteominer^TM columns, precluding its potential use for treatment of patient samples. The method has thus been optimized for clinical studies. First, the Proteominer^TM treatment has been miniaturized to be compatible with the low CSF volume typically obtained after lumbar puncture. We could show that the treatment is still efficient with this miniaturized protocol, and that the dynamic range of protein concentration is actually reduced even with small amounts of beads, leading to an increase of more than 80% of the number of identified proteins in one LC-MS/MS run. Moreover, the reproducibility required for protein quantification has been checked for this new protocol in replicate experiments. Finally, a labelfree quantitative proteomic strategy dedicated to the analysis of large series of samples has been developed. For such studies, fractionation of the samples is generally not possible, but the analysis in one run of a complex sample, even after Proteominer^TM treatment, limits the number of identified and quantified proteins. In order to increase this number, we implemented an approach based on the comparison of MS signals in individual runs with a previously generated MS/MS identification database containing m/z and retention time values associated with peptide sequences. The MFPaQ software was used to assign nonsequenced peptides MS signals to the identification database and to quantify them. The combination of ProteominerTM sample treatment with the bioinformatics workflow developed for data quantification allowed us to increase by a factor 3.4 the number of proteins identified and quantified in the same CSF sample. This work opens the way to future studies aiming at discovering biomarkers in CSF among the low abundance proteins.

Labeling of Peptide Fragmented Mass Spectra in Proteomic Studies

B. Gerrits¹, C. Panse¹, B. Bodenmiller², and R. Schlapbach¹

¹UZH/ETH Functional Genomics Center, Zurich, Switzerland; ²IMSB/ETH, Zurich, Switzerland

Due to the advent of accurate and fast sampling mass spectrometers, proteomic experiments often contain thousands of peptide fragmentation spectra. Although it is commonly accepted that no manual validation of individual spectra in such experiments is feasible, annotated spectra of the peptides assignments with their modifications are required for publication and reviewing purposes. Here we present an algorithm that greatly facilitates the visualisation of peptide fragmentation spectra and aides with the quality assessment of modification sites such as phosphorylation.

The application has two inputs: 1) the Mascot dat-file including the Mascot modifications and 2) the assignment list containing the query number and peptide rank. In the first step using Perl, the application retrieves the peptide assignments and computes the theoretical fragments. These are then mapped to the peak list with the error margin specified during the initial search. During the second stage using R, two different heuristics can be chosen to calculate the appropriate text-labels and print the labelled spectra.

To demonstrate the usefulness of the peakplot application we built a CGI based world wide web accessible userinterface. Also via this interface several data sets are available for testing. As datfile content based filtering aion score cut-off, peptide query hit selection, and a selection by peptide modification based on the mascot servermodification file are provided.

As output on default four different colour schemes areprovided, as well as one multi panel plot, which provides additional graphics and statistics about the assigned peaklist.

A.2

Exact Quantification of Complex Protein Mixtures Using MeCAT—Metal Coded Tagging

R. Ahrends¹, U. Bergmann¹, S. Pieper¹, B. Neumann², C. Scheler², and M. W. Linscheid¹

¹Humboldt-Universitaet zu Berlin, Germany; ²Proteome Factory AG, Berlin, Germany

Quantitative peptide and protein analysis is one of the most promising fields in modern life science. Besides stable isotope coded labeling, metal chelate complexes are an alternative tool for quantification. The development of metal-coded affinity tags (MeCAT) was aimed to provide a robust tool for the quantification of peptides and proteins by utilizing lanthanideharboring metal tags. It was shown that MeCAT is suited for relative quantification of proteins via standard mass spectrometric methods. The approach of tagging biomolecules with MeCAT offers the unique advantage of absolute quantification via inductively coupled plasma mass spectrometry (ICPMS), a well established technique for assessing concentrations down to low attomole ranges. Proteins and peptides are labeled by MeCAT reagents which contain an amino acid residue-reactive labeling group and an element tag loaded with a lanthanide ion. By using different lanthanides such as Lutetium, Holmium, Thulium and Terbium in the MeCAT reagent, multiplex experiments can be performed to analyze several protein samples simultaneously in a proteomic study.

After MeCAT labeling peptides and proteins are separated by common chromatography or electrophoresis techniques and quantified by LC/ESI MS or Inductively Coupled Plasma Mass Spectrometry (ICPMS) detecting the amount of MeCAT metal as a measure for quantity of the protein. If required, proteins of interest are identified by nanoLC/ESI MSn.

In this work we investigated the compatibility of MeCAT labeling to analysis workflows such as nano liquid chromato-graphy/electrospray ionization tandem mass spectrometry (nano-LC/ESI-MSn) and electrophoresis followed by FIA/ICPMS. Focus was given to the separation behavior of labeled peptides and proteins as well as the dynamic range of detection. Furthermore, we demonstrated that MeCAT complexes are stable under a variety of conditions and that by applying LC/ ESI-MS it is possible to cover a dynamic range of 2 orders of magnitude down to the low femtomole range with an average standard deviation below 15%.

Next to the relative quantification pathway applying LC/ ESI-MS we also developed a two dimensional gel based separation system for MeCAT labeled proteins in combination with FIA/ICPMS for absolute quantification of proteinsWith the application of the MeCAT technique to a standard analysis scheme in proteomics, such as the investigation of heat induced expression of recombinant proteins in an Escherichia coli High Cell Density Culture (HCDC), we successfully addressed the suitability to utilize MeCAT on biological samples. Several regulated proteins were identified and quantified including the recombinant Aprotinin::β-galactosidase, heat shock proteins, aconitase, and oligopeptide binding protein.

Besides the obtained relative quantification data, we were able to analyze the recombinant expressed pharmacological active protein Aprotinin (Aprotinin::²-galactosidase) on protein level in an absolute fashion by applying MeCAT-tags in combination with FIA/ICPMS and external calibration. For absolute quantification on the peptide level, metal-coded synthetic peptides which are quantified externally by FIA/ICPMS serve as internal standards in complex peptides mixtures obtained from tryptic proteolysis of biological samples.

A.3

On the Reproducibility of a Fractionation Procedure for Fish Muscle Proteomics

P. Rodrigues¹, T. Silva¹, F. Jessen², and J. Dias¹

¹CCMAR, Universidade do Algarve, FCMA, Centro de Ciências do Mar do Algarve, Campus de Gambelas, Faro, Portugal; ²DTU Aqua, Institut for Akvatiske Ressourcer, Danmarks Tekniske Universitet, Lyngby, Denmark

Sub-cellular fractionation procedures for muscle tissue have been used for some time in proteomics, easing analysis by reducing the number of proteins in a given extract (and therefore improving the dynamic range by allowing larger loads per protein). On the other hand, since it implies a greater number of sample processing steps than a whole extract, it is expected that some noise could be introduced by this fractionation procedure.

The aim of this study was to assess if the amount of noise introduced by a muscle fractionation procedure was significant by comparing it to the baseline technical noise level inherent to 2DE runs, Indiana, USA order to determine if this fractionation method was valid for proteome analysis, using the gilthead seabream (Sparus aurata) as model.

For the experiment, two groups of 5 gilthead seabreams each were subjected to distinct levels of pre-slaughter stress to assess its effects in flesh quality, having obtained three samples of the dorsal muscle from each (post-slaughter, pre-rigor and post-rigor) for a total of 30 samples. These samples were fractionated in five batches, taking care to avoid variable confounding, and separated by 2DE.

The results obtained seem to indicate that the use of this fractionation procedure introduces a low amount of noise, due to the fact that samples fractionated in different batches have the same level of variation as two samples that were fractionated in the same batch. In addition, when attempting to cluster the samples using several different metrics, no grouping made a distinction between fractionation batches.

This shows that this fractionation method can be useful for some proteomic studies involving muscle tissue, especially in cases where the quantity of low abundance proteins is important.

A.4

Optimization of Peak Capacity in One and Two-Dimensional NanoLC

S. Eeltink, B. Dolman, R. Swart, and G. Tremintin

Dionex Corporation, Sunnyvale, CA

To tackle contemporary proteomics samples, different approaches are available for the identification of proteins. In the bottomup approach, proteins are digested and the resulting peptides are separated by high performance liquid chromatography (HPLC). In one-dimensional (1-D) separations, column technology and operating conditions can be optimized to increase peak capacity, but can resolve only relatively simple peptides mixtures. Tryptic digestion of complex proteomics samples containing 1000 proteins can lead up to 50,000 peptides and require powerful separation techniques like multidimensional liquid chromatography (MDLC).

To obtain the best compromise between peak capacity and analysis time in one-dimensional and two-dimensional liquid che4omatography (LC), column technology and operating condiotions were optimized. The effects of gradient time, flow rate, column temperature and column length were investigated in one-dimensional reverse phase (RP) gradient nano-LC, with the aim of maximizing the peak per unit of time for peptides separations.

A.5

Improving the Utility of Electron-Transfer Dissociation

K. F. Medzihradszky, S. P. Salas-Castillo, and A. L. Burlingame

Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA

Captured electron-triggered fragmentation produces almost exclusively peptide backbone cleavages, Indiana, USA form of c and z fragments. Such "limited" fragmentation may be more beneficial for the characterization of longer sequences than CID. In addition, electron-transfer dissociation (ETD) may hold the key for the large scale, reliable site-assignment of such "fragile" post-translational modifications as phosphorylation and O-glycosylation. Evidence has been presented in the recent literature[1] that suggests that charge states equal to 3 or greater provide higher quality ETD spectra of peptides. In addition, it also has been reported that precursor ions below ~ m/z 850 deliver the best results. In the present study we compare the ETD-based information obtained from digests generated with different sequence cleavage specificities. In addition, we report improved ETD results in conjuction with the charge-increasing derivatization of Cys-side-chains as well as carboxyl-groups.

Support for this work was provided by the Bio-Organic Biomedical Mass Spectrometry Resource at UCSF (A. L. Burlingame, Director) through the Biomedical Research Technology Program of the NIH National Center for Research Resources, NIH NCRR P41RR001614.

References

1. Good, D. M., Wirtala, M., McAlister, G. C., and Coon, J. J. (2007) Performance characteristics of electron transfer dissociation mass spectrometry. Mol. Cell. Proteomics 6, 1942–1951.

A.6

Enrichment and Characterization of Secreted Glycopeptides Bearing SA_1-0Galβ1-3GalNAc Structures

Z. Darula¹, and K. F. Medzihradszky^1,2

¹Proteomics Research Group, Biological Research Center of the Hungarian Academy of Sciences, Szeged, Hungary; ²Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA

The lack of consensus sequence, common core structure, and universal endoglycosidase for the release of O-linked oligosaccharides makes O-glycosylation more difficult to tackle than N-glycosylation. Structural elucidation by mass spectrometry is usually inconclusive as the CID spectra of most glycopeptides are dominated by carbohydrate-related fragments. In addition, O-linked structures also undergo a gas-phase rearrangement reaction that eliminates the sugar without leaving a telltale sign at its former attachment site.

In the present study we used electron-transfer dissociation for the characterization of intact glycopeptides affinity-enriched from bovine serum. Some glycopeptide-containing fractions were analyzed also after exoglycosidase treatment. Reducing the size of the carbohydrate chain aided the identification of multiply modified species.

We report the unambiguous identification of 21 novel glycosylation sites. We also detail the limitations of the current methods.

This work was supported by Hungarian Science Foundation grants OTKA T60283 (to KFM) and by NIH grant NCRR P41RR001614 to the UCSF MS Facility (Director, A. L. Burlingame).

A.7

Enrichment of O-GlcNAc Modified Proteins by the Periodate Oxidation – Hydrazide Resin Capture Approach

E. Klement¹, Z. Lipinszky², Z. Kupihar³, A. Udvardy², and K. F. Medzihradszky^1,4

¹Proteomics Research Group, Biological Research Center of the Hungarian Academy of Sciences, Szeged, Hungary; ²Institute of Biochemistry, Biological Research Center of the Hungarian Academy of Sciences, Szeged, Hungary; ³Department of Medical Chemistry, University of Szeged, Szeged, Hungary; ⁴Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA

O-GlcNAc is a post-translational modification found on serine and threonine residues of cytosolic and nuclear proteins. The modification is dynamic and occurs at substoichiometric levels, therefore enrichment is essential. The different strategies so far include chemoenzymatic labeling, beta-elimination of the sugar moiety and lectin weak affinity chromatography. Here, we present a novel enrichment strategy that is based on the periodate oxidation – hydrazide capture approach developed for N-linked glycoproteins. Because of the differences between the two types of carbohydrate modifications the oxidation and elution steps had to be modified. The enrichment protocol was optimized on a mixture of alpha-crystallin and BSA. Then the method was applied to the proteasome complex previously reported as O-GlcNAc modified [1]. Novel modification sites on proteins co-purifying with the proteasome complex will be presented.

This work was supported by Hungarian Science Foundation grant OTKA T60283 (to KFM) and by NIH grant NCRR P41RR001614 to the UCSF MS Facility (Director, A. L. Burlingame).

References

1. Suümegi, M., Hunyadi-Gulyás, E., Medzihradszky, K. F., and Udvardy, A. (2003) 26S proteasome subunits are O-linked N-acetylglucosaminemodified in Drosophila melanogaster. Biochem. Biophys. Res. Commun. 312, 1284–1289.

A.8

Proteome Survey Using Affinity Proteomics and Mass Spectrometry

N. Olsson¹, C. Wingren¹, M. Mattsson², P. James¹, F. Nilsson², and C. A. K. Borrebaeck¹

¹Department of Immunotechnology, Lund University, Lund, Sweden; ²BioInvent International AB, Lund, Sweden

Affinity proteomic methodologies, such as antibody-based microarrays, have shown great promise in several proteome expression profiling applications. The resolution of such proteome analyses is, however, directly related to the number of antibodies included on the array, which currently is a key bottleneck. Here, we present a conceptually new method, denoted Global Proteome Survey (GPS), based on combining the premium features of affinity proteomics and mass spectrometry. The approach will provide novel possibilities for targeting a significant fraction of a proteome in a specie independent manner still using a limited set of antibodies. To this end, we have designed a new class of antibodies, denoted context-independent motif specific (CIMS) antibodies. We have defined sets of short peptide motifs, 4 or 6 amino acids long, where each motif was present in up to a few hundred different proteins (using the human proteome as model system). In this manner, 200 antibodies, binding 50 different motifs commonly distributed among different proteins, would potentially target a protein cluster of 10000 individual molecules, i.e. around 50% of the nonredundant human proteome. To date, we have successfully selected 91 CIMS antibodies against 27 motifs, using our human recombinant scFv antibody library, composed of 2 x 10¹⁰ members and microarray adapted by molecular design, as a renewable probe source. Next, the binders were immobilized in an array format, multiplexed plate format or column format and used to capture and enrich motifcarrying peptides from the digested (trypsinated) proteome(s). The captured peptides were then detected, identified and in some cases even quantified (label-free) using MS and tandem-MS based readout. In this study, we profiled human colon tissue extracts and mouse liver homogenates to demonstrate proof-of-concept of the method. The results showed that the CIMS antibodies were capable of binding and enriching numerous peptides (proteins) harboring the corresponding selection motif, even when targeting crude digests originating from different species. The CIMS antibodies were found to recognize a linear epitope composed of two to four conserved residues, while the identity of the neighboring residues was more flexible. Taken together, the GPS platform has the potential to become a key discovery technology for highthroughput analysis of complex proteomes in health and disease in a specie independent manner.

A.9

Sampling the N-terminal Proteome of Human Serum and Plasma

P. Wildes¹ and J. A. Wells^1,2

Departments of ¹Pharmaceutical Chemistry and ²Cellular and Molecular Pharmacology, University of California, San Francisco, CA

The N-terminal proteomes of serum and plasma are complex and diverse, due to the importance of limited proteolysis in many extracellular signaling pathways. We have developed a method for labeling and enrichment of N-terminal peptides based on specific biotinylation of N-terminal alpha amines using subtiligase, an engineered enzyme. We have employed this method to identify nearly 800 N-terminal peptides in over 200 proteins in human serum and plasma, ranging down to low nM concentrations. While many of these N-termini correspond to known proteolytic processing events (e.g. signal peptide removal, prohormone processing, coagulation or complement activation), nearly 75 percent correspond to exo- or endo-proteolytic cleavages that have not been reported previously. In addition to identifying previously unknown sites of proteolytic processing, N-terminal isolation also allows us to sample one or a few representative peptides from proteins in serum, dramatically reducing the complexity of the sample. The N-terminal peptides can serve as markers of proteolytic events, Oregon, USA surrogates of the intact protein abundance. We are currently developing methods for label-free quantitation of N-terminal peptides to investigate their potential utility as biomarkers.

A.10

Profiling Cell Surface and Secreted Glycoproteins Isolated from Human Thyroid Cancer Cell Lines

T.-Y. Yen, N. Haste, A. Castanieto, A. Arcinas, and B. Macher

San Francisco State University, Department of Chemistry & Biochemistry, San Francisco, CA

We have obtained proteomic profiles from various thyroid cancer cell lines that represent the range of thyroid cancers of follicular cell origin. In this study, we oxidized the carbohydrates of secreted proteins and those on the cell surface with periodate and isolated them via covalent coupling to hydrazide resin. The glycoproteins obtained were identified from tryptic peptides and N-linked glycopeptides released from the hydrazide resin using 2-dimensional liquid chromatography-tandem mass spectrometry in combination with the gas phase fractionation. Thyroid cancer cell lines derived from papillary thyroid cancer (TPC-1), Hürthle cell carcinoma (XTC-1), and metastases of follicular thyroid cancer (FTC-133, FTC-236 and FTC-238) were evaluated. On average more than 100 glycoproteins were identified per cell line, of which about 60 percent are known cell surface or secreted glycoproteins. The usefulness of the approach for identifying thyroid cancer associated biomarkers was validated by the identification of glycoproteins (e.g. CD44 and metalloproteinase inhibitor 1) that have been found to be useful markers for thyroid cancer. In addition to glycoproteins that are commonly expressed by all of the cell lines, we identified others that are only expressed in a specific thyroid cancer cell line. Based on the results obtained by mass spectrometry, a set of glycoprotein biomarker candidates for thyroid cancer is proposed. We are currently quantitatively comparing the relative abundance of a subset of the glycoproteins identified using a label-free quantification method. These results are being compared with quantifications done using antibodies to the glycoproteins.

A.11

Chemical Cross-linking in Complex Mixtures

M. J. Trnka and A. L. Burlingame

Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA

Many fundamental physiological processes are catalyzed by stable protein complexes (e.g. histone remodelling, mRNA splicing by the spliceosome, protein degradation by the proteosome, transit across the nuclear envelope by the nuclear pore complex, initiation of apoptosis). The compositions of these complexes have been intensively studied by immunoprecipitation and affinity purification methodologies combined with mass spectrometry. However, these methods remove all information concerning the spatial arrangement of the constituent proteins within a complex. Furthermore, x-ray structures of larger complexes are difficult to obtain.

Chemical cross-linking reagents are used to preserve topographic information by inserting new covalent bonds between neighboring proteins of a complex. After tryptic digestion, the covalently joined peptides are identified by tandem mass spectrometry, and the sites of modification are used to infer proximity of the original proteins.

Chemical cross-linking methodologies have suffered due to the low yields of true interpeptide cross-links relative to unmodified peptides and so called "dead-end" modified peptides, where only one of the two reactive moieties of the cross-linking reagent reacts with protein. This problem is inherent to all cross-linking reagents that employ activated esters, such as N-hydroxy succinimide esters, as reactive moieties, because succesful cross-linking must compete with hydrolysis.

Here we present a cross-linking reaction and enrichment strategy that allows discrimination between "dead-end" modified peptides, cross-linked peptides, and unmodified peptides. We have synthesized a trifunctional cross-linking reagent which contains two electrophilic formyl groups and an alkyne moiety. The formyl groups react with lysine residues in the presence of a reducing agent via a reductive amination mechanism. "Dead-end" modified peptides therefore contain an aldehyde moiety that can be used as a chemical handle to deplete these peptides.

Cross-linked peptides can then be enriched by Copper catalyzed Huisgen cylcoaddition of the alkyne handle with a cleavable azido-biotin reagent.

This poster presents optimization of reaction conditions to effect this discrimination in both model proteins and in E. coli cell lysates.

Research support was provided by the Bio-Organic Biomedical Mass Spectrometry Resource (A. L. Burlingame, Director) through the Biomedical Research Technology Program of the NIH National Center for Research Resources, NIH NCRR P41RR001614.

A.12

Rapid MRM Assay Development Strategies — Intelligent Software and Acquisition Strategies for Highest Productivity

S. Mollah, M. M. Champion, and C. L. Hunter

Applied Biosystems, Foster City, CA

Targeted peptide quantitation is a rapidly growing application within proteomics mass spectrometry due to its widespread utility in biomarker verification, protein/peptide confirmation and characterization, as well as pathway mapping. As more extensive protein panels need to be monitored in a targeted way across multiple samples, higher throughput is becoming essential. The need for rapid assay development, higher multiplexing and more robust assays are some of the key challenges. In this work, the combination of unique workflows on the hybrid triple quadrupole linear ion trap mass spectrometer and automated softwares, MRMPilot^TM Software and MultiQuant^TM Software, have been used to automate and simplify creation of highly multiplexed MRM assays. The complete workflow of taking proteomics discovery data, to refinement of MRM transitions, resulting in creation of a final MRM assay of >1000 MRM transitions can now take just a matter of days.

In this study, E. coli samples grown in various growth conditions were used for analysis to illustrate the efficiency of the MRM assay development process. Labeling strategy using mTRAQ reagents 0, 4, and 8 were used with one of the mTRAQ® reagent labeled sample acting as the internal standard (GIS). This provides an added improvement in the robustness of the assay development. This efficient workflow resulted in developing an MRM assay at a rate of ~2 peptide/hr (~48 peptide/day). This has reduced the time required for developing a large MRM assay from weeks to just a matter of days. All assay development/MRM refinement was done from biological matrix, no synthetic peptides are required for this assay development strategy, reducing overall project cost.

A.13

Structural Proteins in the Complex Phage 201phi2-1

S. Weintraub, J. A. Thomas, K. Hakala, P. Serwer, and S. C. Hardies

University of Texas Health Science Center, San Antonio, TX

Tailed phages are bacterial viruses with a DNA genome in a protein head that is connected to a tail used to inject the DNA into a host cell. The wide variety of phage types and infective mechanisms are of interest for use as antibacterial agents. We isolated and sequenced the 316,674-bp genome of the unusual Pseudomonas chlororaphis phage 201phi2-1. Since the function of most of the 460 predicted encoded proteins could not be assigned by comparative methods, proteins identified by MS were used to supplement the informatics. An unprecedented number of virion proteins (88) were identified, with several of them (18) having been cleaved to more than one polypeptide. With the high sequence coverage and large number of semitryptic peptides that were found, we could define many of the polypeptide end points, and hence the cleavage motif of the prohead protease responsible for these cleavages. Most phages cleave the major capsid protein to enable capsid expansion during DNA packaging, accounting for one of the detected 201phi2-1 cleavages. The cleavage patterns were combined with informatic analysis to hypothesize what other maturation processes may be occurring in this phage. One of the cleaved proteins was found by customized profile building methods to be a distant homolog of the beta chain of RNA polymerase. After cleavage, the propeptide as well as the polymerase chain remained in the head. We hypothesize that the propeptide determines capsid localization and that cleavage releases the polymerase chain for injection into the cell. For one of the three RNAP subunits found in the virion, Mississippi, USA showed that there had been a selfsplicing intron. Most phages encode a scaffold protein around which the capsid assembles and is then cleaved during DNA entry and leaves the virion. Scaffold-like sequences were found in the N-terminal propeptides of a family of 6 capsid proteins for which the C-terminal domains were homologous and retained in the mature virion. The stoichiometry of the C-terminal domains within the virion was estimated by spectrum counting, and the mass of propeptides that had been released from the virion was subsequently estimated to be appropriate for the scaffold of a virion of this size. The retained C-terminal domains were presumed to compose the novel inner head body reported for this virus. Finally we sought to resolve a discrepancy by which plots of spectrum counts indicated substantial downgel smears inconsistent with the Coomassie profile. The peptide coverage in the downgel smears was consistent with a degree of nonspecific protein degradation. By comparison of spectrum counts with known stoichiometry of several virion proteins, a saturation effect was found which overestimates the amount of rare degraded species in each gel slice. Efforts are underway to include precursor ion intensities in these assessments as a way to more accurately determine the relative copy number of the 201phi2-1 structural proteins.

A.14

A Quick Method for Differential PTM Analysis of Hypermodified Proteins by FTICR/ECD/MS/MS and Bioinformatics

F. Li¹, S. Guan¹, F. Chu², R. Talroze¹, and A. L. Burlingame¹

¹Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA; ²College of Life Sciences and Agriculture, University of New Hampshire, Durham, NH

Mass Spectrometry has evolved to be the most powerful method for protein identification and characterization of protein covalent modifications. The nature and position of many protein post-translational modifications (PTM) can be identified and even quantified by recently developed mass spectrometry methods. Moreover, different from other methods, mass spectrometry can be employed to discover new proteins and new modifications on proteins.

Despite the impressive progress of mass spectrometry in biological applications in the past decade, determination of post-translational modifications on proteins is far from being routine. Usually labor-intensive and time-consuming sample preparation step is required to enrich or isolate the modified species and success may not be guaranteed. Quick, easy and economical methods for differential PTM analysis without identifying each PTM on multiple samples for comparison are highly desired in many biological research fields. However, there is no such method reported in literature. Here, we present a method to address this unmet need.

Histones are biologically important proteins with many modifications. Characterizing the modifications on histones is even more challenging than a regular PTM determination. Two Histone H4 samples from the wild type and a mutant of mouse embryonic stem cells were used to illustrate this method. Each of the two histone H4 samples was first directly infused to a FTICR mass spectrometer respectively as intact proteins. MS molecular profiles of all modified species can be obtained for each sample. All isomers with the same M/Z were further isolated from other species and subjected to MS/MS fragmentation by ECD. Multiple MS/MS spectra were accumulated to improve the signal to noise ratio of ECD spectra.

A "search all possibility" algorithm was developed to match theoretical MS/MS isotope profiles of all the possible PTM combinations on the sequence to the experimental MS/MS peaks. All the matched experimental peaks are mapped on the sequence indicating numbers of PTMs and the mapping clearly illustrates the "segmental" PTM assignments. In addition, all the modified sequences were ranked in the order of the degree how well the theoretical peaks were matched with the experimental data. Such a list of the best matched modified sequences, combined with the segmental PTM assignments, was used to detect the differences between two samples without having to identify the individual PTM on each sample. The final results show that the nature of PTMs was the same for both samples while the abundances of some certain modified species were different in the detectable dynamic range.

This method requires minimal sample preparation. Neither enzymatic digestion nor analytical separation of PTM variants is required. The difficulty of handling sample complexity is shifted from sample preparation to the MS/MS data analysis stage.

This method can serve as a general screening method for detecting difference among biologically related samples such as (1) wild type and mutant, (2) normal and degraded, (3) healthy and disease samples and it is especially useful for those hypermodified proteins such as histones in a fast and high throughput way.

This work is supported by the Bio-Organic Biomedical Mass Spectrometry Resource at UCSF (A. L. Burlingame, Director) through the Biomedical Research Technology Program of the NIH National Center for Research Resources, NIH Grants NCRR P41RR001614 and NCRR RR019934.

A.15

Improved Data Mining by Using TPP-based Analysis Workflows for Searching MS/MS Data

A. Quandt¹, L. Malstroem¹, H. Lam², D. Shteynberg³, and R. Aebersold¹

¹Institute for Molecular Systems Biology, ETH Zurich, Switzerland; ²Department of Chemical and Biomolcular Engineering, The Hong Kong University of Science and Technology, Kowloon, Hong Kong; ³Institute for Systems Biology, Seattle, WA

The identification and characterisation of peptides from tandem mass spectrometry (MS/MS) data represents a critical aspect of proteomics. In the past, many software packages have been developed to tackle this problem. Beside the development of new analysis tools, recent publications describe also the pipelining of different search programs to increase the identification rate. Unfortunately, it remains still unclear for the practical user when to apply which software or parameter set to retrieve optimal results. Most people rely on the usage of a identification software what consequently often leads to a significant proportion of the experimental spectra which are not going to be identified. Hence, the usage of different tools and parameter combinations is crucial but seldom approached in reality. Among other reasons such as manual result validation and the handling of various data formats, the main problem still remains in the automated combination of different identification tools. Here, we present a workflow approach which is based on the TransProteomicPipeline (Keller et al., 2005; Pedrioli et al., 2004) and combines multiple search tools and strategies with the result to retrieve a more complete list of peptide identifications and a higher protein coverage. For us, a workflow is the combination of various identification tools and search strategies in parallel and/or in sequential order. In the example workflow, we are going to present here, two classical search engines (X!Tandem and OMSSA) have been combined with a spectral library search search (SpectraST). In fact, we use the output of the first two search tools to dynamically create a spectral library which is searched afterwards. To compare the performance of our workflow with the results of the individual search engines, we used the 18-protein-mix dataset, which has been especially created to benchmark different search tools and pipelines. On these initial dataset, we are able to show that the combination of various identification tools in so-called workflows leads to an increased trust into the results by lowering the level of accepted false positives identifications and a significant increased protein coverage leading to a more reliable search results.

A.16

Microwave-assisted Phosphoproteomics

P. Liu and W. Sandoval

Genentech, Inc., S. San Francisco, CA

Reversible protein phosphorylation controls a multitude of important biological functions. Elucidation of the exact site of phosphorylation is often necessary to further understand the intricate mechanisms involved in intracellular sites. We have investigated the use of microwave enhanced tryptic digestions on phosphopeptide recovery. We have incorporated the use of stable isotopically labeled peptides to quantitate differences observed. In addition we have quantitatively evaluated the use of stabilizing chemical modification of phosphopeptides and phosphoproteins for subsequent analysis by mass spectrometry or Edman degradation. Finally we have applied our findings to a global phosphoproteomic analysis of peroxide treated cell lysate.

A.17

Proteome Analysis of Apoplastic Proteins in Rice Shoot Respond to Salt Stress

Y. Song¹, A. L. Burlingame², and Y. Guo^1,2

¹Institute of Molecular and Cell Biology, Hebei Normal University, Shijiazhuang, China; ²Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA

Plants have evolved sophisticated systems to cope with the adverse environmental conditions such as cold, drought and salinity due to their sessile nature. Although a lot of stress response networks have been proposed, the roles of plant apoplast have been obviously ignored in plant stress response. To investigate the role of apoplastic proteins in the salt-stress response, 10-day-old rice plants were treated with 200 mM NaCl for 1, 6 or 12 hours, and the soluble apoplast proteins were extracted and for differential analysis compared to untreated controls by 2-D DIGE saturation labeling techniques. 122 significant changed (1-ANOVA p-value <0.05) spots are identification by LC-MS/MS, and 117 spots representing 69 proteins have been identified. Of these, 37 proteins are apoplastic proteins according to bioinformatic analysis. These proteins are mostly involved in carbohydrate metabolism, oxido-reduction, protein processing and degradation. According to the results of functional categories and cluster analysis, a stress response model of apoplastic proteins has been proposed. These dates indicate that apoplast is an important portion in plant stress signal reception and response.

LC-MS/MS data was provided by the Bio-Organic Biomedical Mass Spectrometry Resource at UCSF (A. L. Burlingame, Director) supported by the Biomedical Research Technology Program of the NIH National Center for Research Resources, NIH NCRR P41RR001614 and NIH NCRR RR019934.

A.18

Investigation into the Use of Lys-N Combined with Electron Transfer Dissociation on a Quadrupole Time-of-Flight Mass Spectrometer for Peptide Sequencing

J. Langridge¹, J. Brown¹, S. Mohammed², N. Taouatas², Iain D G Campuzano¹, and A. J. R. Heck²

¹Waters Corporation, MS Technologies Center, Manchester, United Kingdom; ²Department of Biomolecular Mass Spectrometry, Utrecht University, Utrecht, The Netherlands

Recently, Lys-N was introduced as a near ideal enzyme for proteomics workflows that involve electron transfer dissociation (ETD) or MALDI-CID. Lys-N peptides with a single basic residue, Indiana, USA combination with ETD, have a clear advantage for spectral interpretation due to peptide tandem mass spectra consisting of almost exclusively c-type fragment ions. Furthermore, the combination of Lys-N and SCX enables facile fractionation/enrichment of acetylated, phosphorylated peptides with a single basic residue and non-modified peptides with a single basic residue. Recently, we described the implementation of ETD on a Q-Tof instrument and we describe here the evaluation of the Lys-N strategy on this modified instrument.

BSA was reduced with 45 mM dithiothreitol (50 °C, 15 min), followed by alkylation using 110 mM Iodoacetamide (dark, RT, 15min). Lys-N was added at a ratio of 1:85 (w/w) and incubated over night at 37 °C. Metalloendopeptidase from Grifola Frondosa (Lys-N) was obtained from Seikagaku Corporation (Tokyo, Japan). Doubly or triply charged precursor cations were selected by the quadrupole and allowed to mix and react with either fluoranthene or azobenzene reagent anions previously collected in the travelling wave ion guide (TWAVE) collision cell of the QTOF. Product ions spectra were recorded by the TOF mass analyzer.

We have acquired preliminary data using electron transfer dissociation (ETD) on a modified hybrid quadrupole orthogonal acceleration time-of-flight mass spectrometer (Q-Tof). Several peptides from a Lys-N digest of bovine serum albumin (BSA) have initially been investigated. A variety of the peptides from the digest were selected with the quadrupole, prior to dissociation via electron transfer from the reagent anion. For the doubly charged precursors the MS/MS data acquired show almost exclusively c-ions, these being N-terminal sequence related ions. Cleavage was observed at almost every amide bond in the peptide backbone, yielding easy-to-interpret sequence ladders. This coupled with the inherent mass measurement accuracy and resolution of the oa-TOF mass analyser makes the data easily amenable to de novo sequencing. The signal intensity of the fragment ions seemed to diminish with decreasing m/z, as previously reported. In addition to the ETD experiments the mass spectrometer can acquire alternate scans in CID, and as such data will be compared and contrasted between ETD and CID on a variety of peptides produced from Lys-N digestion and separated by nanoscale LC.

Quantitative Analysis of Proteome Localisation and Dynamics

A. Lamond

Wellcome Trust Centre for Gene Regulation and Expression, MSI/WTB Complex, University of Dundee, Dundee, Scotland, United Kingdom

We are studying the functional organization of mammalian cell nuclei using a dual strategy that combines mass spectrometry (MS) based proteomics with live cell fluorescence imaging (see www.LamondLab.com). This applies two distinct but complementary quantitative techniques to analyse the same biological problem, providing a rigorous approach where potential artifacts or limitations of one method are avoided in the complementary approach and vice versa. The quantitative proteomic methods involve metabolic labeling of cellular proteins in cultured cell lines with the amino acids lysine and arginine containing heavy isotopes such as ¹³C and ¹⁵N. The quantitative imaging experiments, including time-lapse microscopy, FRAP, FLIP, FLIM and FLIM-FRET, are performed on mammalian cell lines stably expressing one or more fluorescent protein-tagged reporters. Both the proteomics and microscopy methods are used to study the same stable cell lines, allowing a direct comparison the resulting data from both techniques.

We have used this dual strategy to characterize in detail the molecular composition of nucleoli under different metabolic and growth conditions and at specific stages of cell cycle progression (see http://lamondlab.com/nopdb/). We have developed a MS-based proteomics strategy to perform quantitative analyses of subcellular protein localization - "spatial proteomics" - including the analysis of protein turnover rates in separate cell compartments. This provides a new approach for annotating the spatial organization of the proteome and for measuring how this changes in response to inhibitors and different cell growth conditions. We have also developed quantitative MS-based approaches for identifying specific protein- protein interactions. These strategies provide a general approach for characterizing the composition, dynamic properties and interactions of either cell organelles or multi-protein complexes.

3.2

Post-Translational Adenosine Monophosphate (AMP) Modification of Proteins

C. A. Worby^1,2,3,9, S. Mattoo^4,9, R. P. Kruger⁵, L. B. Corbeil^6,7, A. Koller⁸, Juan C. Mendez⁶, B. Zekarias⁶, C. Lazar^1,2,3, and Jack E. Dixon^1,2,3,4

Departments of ¹Pharmacology, ²Cellular and Molecular Medicine, ³Chemistry and Biochemistry, and ⁴Howard Hughes Medical Institute, University of California, San Diego, La Jolla, CA; ⁵Department of Biological Chemistry, University of Michigan, Ann Arbor, MI; ⁶Department of Pathology, University of California, San Diego Medical Center, San Diego, CA; ⁷Department of Population Health and Reproduction, School of Veterinary Medicine, University of California, Davis, CA; ⁸Department of Pathology, Stony Brook University, Stony Brook, NY

Eukaryotic cells have devised different strategies to regulate signaling pathways. The best known modification is phosphorylation, which attaches a phosphate group to serine, threonine or tyrosine residues in proteins, thereby regulating their activities. Here, we describe a new modification: the addition of adenosine monophosphate (AMP) on tyrosine residues. AMP addition to Rho GTPases by the Fic domain containing secreted surface antigen IbpA of the respiratory pathogen Histophilus somni leads to cytoskeletal collapse in host cells (1). Specifically, incubation of purified Rho GTPases (RhoA, Rac1 and Cdc42) with GST-tagged and purified Fic domain of IbpA in the presence of ³²P-ATP, but not ³²P-ATP, allows transfer of the ³²P-label to RhoA, Rac1 or Cdc42, thus indicating the addition of AMP versus a phosphorylation event. Unlike VopS, another Fic domain containing protein from V. parahemolyticus, that modifies AMP on threonine residues (2), mass spectrometric analysis of IbpA-treated Rho GTPases show that the IbpA Fic domain adds an AMP to a conserved tyrosine residue in the switch I region of Rho GTPases. In addition, we show that the only human protein containing a Fic domain, HYPE (Huntingtin Yeast-interacting Protein E), also has the ability to add AMP to tyrosine residues in Rho GTPases in vitro. Thus, we identify Fic domain containing proteins as a new class of enzymes that mediate not just bacterial pathogenesis, but also a previously unrecognized eukaryotic post-translational modification that may regulate key signaling events.

Interestingly, threonine and tyrosine AMP modified peptides behave similarly in the mass spectrometer as threonine and tyrosine phosphorylated peptides: whereas threonine modified peptides undergo neutral loss of AMP (plus 18Da) on fragments upon activation of the peptide, peptides fragments modified with AMP on tyrosine mostly stay intact, partially losing adenine as well as adenosine. In addition, AMP-Tyr modified peptides are only identifiable in an ion-trap CID fragmentation experiment, as fragmentation in an HCD cell (or CID in a QSTAR) will lead to a strong signal for adenine and only weak fragmentation peaks of the peptide backbone.

References

1. Worby, C. A., Mattoo, S., Kruger, R. P., Corbeil, L. B., Koller, A., Mendez, J. C., Zekarias, B., Lazar, C., and Dixon, J. E. (2009) The fic domain: regulation of cell signaling by adenylylation. Mol. Cell 34(1), 93–103.

2. Yarbrough, M. L., Li, Y., Kinch, L. N., Grishin, N. V., Ball, H. L., and Orth, K. (2009) AMPylation of Rho GTPases by Vibrio VopS disrupts effector binding and downstream signaling. Science 323(5911), 269–272.

3.3

Dissecting the Structure of the Human Spliceosome by Looking at Its Pieces

P. Coltri¹, J. Ilagan¹, R. J. Chalkley², A. L. Burlingame², and M. S. Jurica¹

¹Department of Molecular, Cell and Developmental Biology and Center for Molecular Biology of RNA, University of California, Santa Cruz, CA; ²Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California San Francisco, San Francisco, CA

Pre-mRNA splicing is the removal of the non-coding introns that interrupt most gene transcripts and serves an essential step in eukaryotic gene expression. The cellular machinery responsible for splicing, termed the spliceosome, is a large protein/RNA macromolecular complex comprised of five structural RNAs and over 100 individual polypeptides. The human complex assembles and functions via a progression of structural intermediates that are not yet fully characterized. The dynamics and complexity of the spliceosome have long posed challenges to detailed biochemical and structural studies that will provide insight into the spliceosome's molecular mechanisms. In particular, isolating distinct conformations of this moving target in the amounts needed for standard biochemical and structural analyses is not simple. We made a key advance in this regard with our development of a substrate-based affinity method to isolate human spliceosomes arrested midway through splicing catalysis (C-complex). Initial mass spectrometry analysis of this complex identified over 200 proteins, ~100 of which were specific to splicing. Using cryo-electron microscopy (cryo-EM) and single particle reconstruction techniques, we solved the structure of C-complex spliceosomes to 30 Å resolution. This model represents an important first step in visualizing the structure of the spliceosome. However, before we can more fully interpret the model in functional terms we must answer questions regarding which components of the spliceosome are visualized/represented in our model and where they are located in the structure. Currently, we are finding ways to take the spliceosome apart and then examining the resultant pieces. Mass spectrometry analysis is critical for defining the protein composition of the pieces, enabling us to define interactions that underpin the spliceosomes architecture. We have examined the contribution of exon sequences in the composition and structure of C-complex and are now looking at the proteins that tightly associate with the intron vs. the region of the upstream exon poised for ligation. In addition to these studies, we have made progress in using chemical modification in conjunction with mass spectrometry to identify regions of proteins that are located at the surface of the spliceosome. This work will allow us to begin localizing these proteins' positions in the complex. By combining the results of these studies with our structural investigation of the spliceosome, we are on the path to assembling a more detailed model of this critical cellular machine.

3.4

Protein Complexes and Functional Pathways in S. cerevisiae and E. coli

M. Babu¹, G. Butland³, J. J. Diaz-Mejia^1,4, P. Hu¹, S. Pu⁵, G. Moreno-Hagelsieb⁴, S. C. Janga¹, S. Wodak^2,5, A. Emili^1,2, and J. Greenblatt^1,2

¹Banting and Best Department of Medical Research, ²Department of Molecular Genetics, University of Toronto, Toronto, ON, Canada, ³Life Science Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA, ⁴Department of Biology, Wilfrid Laurier University, Waterloo, ON, Canada, ⁵Hospital for Sick Children, Toronto, ON, Canada

We have used TAP-tagging and affinity-purification to sort the soluble proteins of S. cerevisiae into complexes. We combined this with systematic synthetic genetic interaction analysis for non-essential gene deletion mutants and essential gene hypomorphs, using the synthetic genetic array (SGA) approach, for genes related to nuclear processes. More recently, we have extended the yeast protein interaction network by focusing on the predicted yeast membrane proteins, purifying each protein three times in the presence of different detergents. We are testing the co-functionality of proteins in various membrane-associated protein complexes by comparing our protein complex data with synthetic genetic interaction data and assessments of the effects of the various proteins in a complex on the morphology of the intracellular compartment in which that complex is located.

We have also used dual affinity tagging followed by affinity purification and mass spectrometry to sort the soluble proteins of E. coli into protein complexes. Although our initial focus was on essential, evolutionarily conserved proteins, we have focused more recently on proteins of unknown function (functional orphans). We integrated our protein-protein interaction network with systematic genome context inferences to derive a probabilistic network of functional inferences encompassing almost all E. coli proteins (98%) and to assign about 57% of the orphans to discrete functional neighborhoods with high confidence. Many of these functional inferences were then confirmed by genome-scale phenotypic assessments. Functional pathways can be derived by systematically identifying genetic interactions, or epistasis, which tends to occur between genes involved in parallel pathways or interlinked biological processes. We have therefore developed a quantitative screening procedure, eSGA (E. coli synthetic genetic arrays), for monitoring bacterial genetic interactions based on conjugation of E. coli deletion or hypomorphic strains to create double mutants on a genome-wide scale. The patterns of synthetic lethality or sickness (aggravating genetic interactions) we observe for certain double mutant combinations provide information about functional relationships and redundancy between pathways and enable us to group E. coli genes into functional modules.

3.5

N-Terminomics: High Confidence, Broad Dynamic Range Coverage Utilizing Novel Polymers for Proteomics Reveals the Functional State of the Proteome

C. M. Overall

UBC Centre for Blood Research, University of British Columbia, Vancouver, British Columbia, Canada

The nature of a protein's N-terminus, its modifications and sequence has profound impacts on the function and localisation of proteins. Moreover, all proteomes are moulded by proteolysis and in all cases this changes the function of a protein, for example in enzyme and protein activation, inactivation, conversion to antagonists, as triggers for secretion, cell surface shedding and finally clearance. Therefore, to functionally annotate proteins, the N and C terminal peptides of proteins must be determined. To focus on these peptides, dedicated techniques are required. Hence, these semitryptic peptides are all too often overlooked in proteomics analyses.

We have developed novel polymers that target primary amine groups on peptides that are invaluable for such proteomics analyses because of their high derivatisation, excellent solubility, no non-specific binding properties, low cost and easy synthesis. Using these polymers in a new procedure termed TAILS (Terminal Amine Isotope Labelling of Substrates) we report a new proteomic and bioinformatics pipeline to rapidly identify natural N-termini and protease cleaved neo-termini of protein substrates after polymer enrichment. MS/MS both identifies the N-terminal peptide and both the substrate and sequence of protease cleavage sites in the same experiment. For most proteins multiple peptides are so identified enabling robust protein identification through multiple peptides. For proteins with single peptides identified, a new statistical analysis enables high confidence protein identification. The key to identifying specific protease substrates is the use of isotopic labelling of all primary amines in order to subtract background proteolysis that is always present. This can be achieved by dimethylation and iTRAQ labelling on primary amines in 8-plex analyses.

We applied TAILS for quantitative N-terminome analysis and for the global analysis of proteolysis in skin inflammation induced by TPA (12-O-tetradecanoyl-phorbol-13-acetate). First, we developed and successfully tested a mass spectrometry-compatible protein isolation and purification method for total skin lysates. Next, we combined this method with TAILS analysis to determine both the skin proteome and skin N-terminome and their perturbations in inflammation. Including wild-type and matrix metalloproteinase (MMP) 2 knockout mice in this multiplex approach allowed us to further identify novel bioactive substrates of this important family of inflammatory matrix metalloproteinase. Thereby, we identified 1,972 proteins with high confidence from murine skin samples with 84 being significantly up-regulated in TPA treated skin including known inflammatory markers such as acute phase proteins and components of the complement system. By TAILS we identified 1,677 N-terminal peptides for 1,032 proteins including 621 that had also been detected prior to N-terminal enrichment. Importantly, among the 411 proteins only identified after enrichment for N-termini were low abundance chemokines like the small inducible cytokines B5 (LIX) and macrophage inflammatory protein 2 (MIP2). As expected, the N-termini of these proteins were also included in a subset of 312 N-terminal peptides assigned to 184 proteins significantly induced by TPA treatment with a statistically significant enrichment of inflammation-related categories by Gene Ontology (GO) analysis. Notably, the analyses were neither skewed by proteins that are highly abundant in skin, such as keratin and filaggrin, nor by serum proteins (only 23 identified).

Hence, N-terminomics analyses using negative peptide selection enables broad proteome coverage with high dynamic range of complex proteomes.

Activity-based Proteomics: Applications for Enzyme and Inhibitor Discovery

B. F. Cravatt III

Department of Chemical Physiology, The Scripps Research Institute, La Jolla, CA

Genome sequencing projects have revealed that eukaryotic and prokaryotic organisms universally possess a huge number of uncharacterized enzymes. The functional annotation of uncharacterized enzymatic pathways, thus, represents a grand challenge for researchers in the post-genomic era. To address this problem, global molecular profiling methods hold great promise, as they provide a relatively unbiased portrait of the biochemical composition of cells and tissues and can reveal unanticipated alterations in their metabolic and signaling networks. Nonetheless, the identification and functional characterization of enzymatic pathways that support human physiology and pathology have, to date, been hindered by a lack of "systems biology" techniques that can evaluate their activity in complex biological samples. To address this problem, we have introduced functional proteomic and metabolomic technologies that record dynamics in enzyme activity in directly in native biological systems. For example, the activity-based protein profiling (ABPP) technology utilizes active site-directed chemical probes to determine the functional state of large numbers of enzymes in proteomes. In this presentation, I will describe the integrated application of ABPP and complementary functional proteomic/metabolomic methods to discover and functionally annotate enzyme activities in mammalian (patho)physiological processes, including cancer and nervous system signaling. The long-term goal of these studies is to map new biochemical pathways that play important roles in human disease and develop selective chemical tools to perturb these pathways in living systems.

4.2

Global Profiling of Proteolytic Cleavage Sites in Apoptosis

S. Mahrus¹, J. Trinidad⁶, D. Barkan⁵, H. Nguyen⁴, A. Sali^1,2, A. L. Burlingame⁶, and J. Wells^1,3

Departments of ¹Pharmaceutical Chemistry, ²Biopharmaceutical Sciences, ³Cellular and Molecular Pharmacology, ⁴Hematology/Oncology, and ⁵Graduate Group in Bioinformatics, University of California, San Francisco, CA; ⁶Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA

The 600 or so proteases encoded in the human genome are involved in a diversity of biological processes. Some function as nonspecific degradative enzymes associated with protein catabolism, exhaustively cleaving many protein substrates at many sites. In contrast, several others function as selective post-translational modifiers, cleaving only a few protein substrates, usually at only one or a few sites. Apoptosis is an important example of a biological process regulated by widespread but specific intracellular proteolysis, predominantly carried out by the caspase family of proteases. This genetically programmed and non-inflammatory form of cell death is a central component of homeostasis, tissue turnover, and development. Since chemotherapeutics typically kill cells by induction of apoptosis, this process is also highly relevant from a therapeutic standpoint. We have developed a novel method for global profiling of proteolytic cleavage sites in complex biochemical mixtures based on use of an engineered peptide ligase, termed subtiligase, for selective biotinylation of free protein N-termini and positive enrichment of corresponding N-terminal peptides. Using this method to study apoptosis, we have sequenced 333 caspase-like cleavage sites distributed among 292 protein substrates in the acute T cell leukemia cell line Jurkat following treatment with the classic chemotherapeutic etoposide. Surprisingly, these sites are generally not predicted by in vitro caspase substrate specificity, but can be used to predict other physiological caspase cleavage sites. Structural bioinformatic studies show that caspase cleavage sites often appear in surface accessible loops and even occasionally in helical regions. We also find that a disproportionate number of caspase substrates physically interact, suggesting that these dimeric proteases target protein complexes and networks to elicit apoptosis, and that targeting of multiple components in each is required for a full commitment to apoptosis. Our current efforts are focused on quantitative analysis of how proteolysis in apoptosis varies as a function of time, target cell type, and apoptotic inducer.

Support for this research was provided by the Bio-Organic Biomedical Mass Spectrometry Resource at UCSF (A. L. Burlingame, Director) through the Biomedical Research Technology Program of the NIH National Center for Research Resources, NIH NCRR P41RR001614.

4.3

N-terminal & ‘Genome free’ Proteomics; de novo Sequence Analysis by a Combination of LysN Protein Digestion and Electron Transfer Dissociation

A. J. R. Heck

Biomolecular Mass Spectrometry and Proteomics Group, Utrecht University, and Netherlands Proteomics Centre, Utrecht, The Netherlands

In this talk targeted novel targeted proteomics technologies will be discussed used to analyze I) protein N-termini and II) proteomes of species of uncharacterized genomes.

Although N-terminal processing of proteins is an essential process, not many large inventories are available, Indiana, USA particular not for human proteins. Using modern day mass spectrometry based proteomics techniques it is now possible to unravel N-terminal processing in a semi-comprehensive way. Strong cation exchange chromatography with improved separation of singly charged peptides was exploited for the targeted analysis of N-acetylated protein termini from human HEK293 cells. Taking advantage of the complementarity between Lys-N, Lys-C, and trypsin for protein digestion, a total of 1391 non-redundant acetylated protein N-termini could be identified in a multi-protease approach, representing the largest dataset of human acetylated protein N-termini to date. Sequence analysis and comparison of the dataset with related datasets from D. melanogaster, S. cerevisiae and H. salinarum provides new insights into N-terminal processing across these species.

For species with un-sequenced or poorly characterized genomes de novo sequencing of MS/MS fragmentation spectra is essential. However, de novo sequencing is challenging due to the complexity of common CID fragmentation spectra. Lys-N enzymatic cleavage in combination with ETD analysis results in fragmentation spectra almost exclusively containing N-terminal fragment ions. These, easy to interpret, ladder sequences open up a completely new window for de novo sequencing. As a proof of concept we analyzed the proteomes of ostrich muscle and hibernating bear heart. We performed a proteomics study of ostrich through Lys-N proteolytic cleavage followed by low-pH SCX fractionation, RP-nanoLC separation and ETD dissociation. The SCX fractionation is used for isolation of the ‘de novo sequence-able’ peptides. These peptides produce fragmentation spectra, after ETD, dominated by c-type ions, which are relatively easy to interpret. De novo analyses of the ETD spectra is performed by an in-house developed algorithm, called LysNDeNovo, which utilizes the presence of a single fragment ion series to assign the peptide sequence. Our de novo sequencing approach results in a significant higher number of peptides identified than searching the ETD Ostrich proteomics dataset using the Mascot search engine. Moreover, the de novo results allow the determination of point mutations as well as conserved regions between proteins of different species.

References

1. Taouatas, N., Altelaar, A. F. M., Drugan, M. M., Helbig, A. O., Mohammed, S., and Heck, A. J. R. (2009) SCX-based fractionation of Lys-N generated peptides facilitates the targeted analysis of post-translational modifications. Mol. Cell. Proteomics 8, 190–200.

2. Gauci, S., Helbig, A. O., Slijper, M., Krijgsveld, J., Heck, A. J. R., and Mohammed, S. (2009) Lys-N and trypsin cover complementary parts of the phosphoproteome in a refined SCX-based approach. Anal, Chem. 81, 4493–4501.

Elucidating Methyltransferase Target Amino Acid Residues and Downstream Modification Events by Quantitative Mass Spectrometry

S. M. Eliuk¹, F. Chu³, B. Panning², and A. L. Burlingame¹

¹Mass Spectrometry Facility, Department of Pharmaceutical Chemistry and ²Department of Biochemistry & Biophysics, University of California, San Francisco, CA; ³College of Life Sciences and Agriculture, University of New Hampshire, Durham, NH

Introduction: Post-translational modifications of histones are used to regulate DNA-chromatin interactions and ultimately gene expression. The various individual histone PTMs do not regulate in isolation. Instead, combinations of modifications are believed to act together to create a ‘histone code’.

Knocking out a specific methyltransferase permits the investigation of its target amino acid residues and how these modifications affect further downstream modification events. Combined with stable isotope labeling of amino acids in cell culture (SILAC), changes in stoichiometries of modification site occupancies are revealed by quantitative mass spectrometry. Through these techniques, we show that knock out of a specific methyltransferase leads to a variety of histone modification changes.

Method: Wild type mouse embryonic stem cells (ESCs) were grown under normal conditions and ESCs mutant for eed, a core component of the histone H3 lysine 27 (H3K27) methyltransferase complex, were grown in SILAC ESC medium. Cells were combined, histones were acid extracted and subsequently fractionated by reverse-phase chromatography. Both pre-fractionated and unfractionated proteins were subjected to digestion with both trypsin and AspN and analyzed by LC ESI MS/MS on a LTQ Orbitrap with ETD fragmentation (fragment ions detected in the Orbitrap). Peptide data were analyzed using two in-house softwares, Protein Prospector and FAVA, allowing assignment of modification sites from fragmentation data and changing stoichiometries between wild-type and mutant cells through quantitative analysis based on the SILAC labeling.

Preliminary Data: As expected, H3K27 methylation was detected in the wild-type sample and absent in the eed mutant sample. In addition, although some modifications were unaffected, we identified other modifications whose abundances were altered in the eed mutant sample. ETD fragmentation with fragment ion detection in the Orbitrap of AspN digested histones was sufficiently sensitive to enable site-specific assignment of modifications on a chromatographic time scale. The online LC separation of the unfractionated histone AspN digests resolved H4 N-terminal tails with varying numbers of acetylations, but not methylations and enabled relative quantitation of differentially modified species. An increase in the level of acetylated species was detected in the methyltransferase mutant cell line. ETD fragment ion spectra were able to confirm the increased level of peptides with K16 acetylation in the mutant cell lines.

Support for this research was provided by the Bio-Organic Biomedical Mass Spectrometry Resource at UCSF (A. L. Burlingame, Director) through the Biomedical Research Technology Program of the NIH National Center for Research Resources, NIH Grants NCRR P41RR001614 and S10 RR019934.

B.2

Ecotoxicoproteomics to Study Microcystin-LR Effects in Medaka Fish

M. Edery, M. Malécot, K. Mezhoud, A. Marie, D. Praseuth, and S. Puiseux-Dao

Muséum national d'Histoire naturelle, Paris, France

Microcystins (MCs) are hepatotoxins with a potent inhibitor activity of protein phosphatases PP1 and PP2A. These nonribosomal peptides are getting more and more attention due to their acute toxicity and potent tumorpromoting activity. These toxins are produced by freshwater cyanobacteria. To date, the detailed mechanisms underlying the toxicity of microcystins are unknown. MC-leucine-arginine (MC-LR) is the most toxic and the most commonly encountered variant of MCs in aquatic environment. It has been used for toxicological investigations on the liver of an aquatic animal model; the medaka fish. We performed differential proteome analyses of MCLR-treated (by 2 exposure routes; water or food contamination) and untreated medaka fish in order to investigate the mechanisms of establishment of early responses to the toxin. Cytosolic, membrane and organelle proteins from livers of exposed or non-exposed medaka were resolved by 2D electrophoresis and detected using stains specific for phosphoproteins and for whole protein content. Overall, more than 100 spots were found to vary significantly on the proteomic 2D maps or on the phosphoproteomic 2D maps. Of these, 32 proteins could be identified by mass spectrometry. Among them, phenylalanine hydroxylase, keratin 18 (type I) and grp78 showed variations in phoshoryl content in agreement with inhibition of PP2A activity after exposure of the fish to MC-LR. The other identified proteins exhibited variations in their expression level. The identified proteins appear to be involved in cytoskeleton assembly, cell signalling, oxidative stress and apoptosis. The methodology described in this report should be widely used to a number of tissues and organisms, thus helping in the search for biomarkers of MC-LR contamination.

B.3

Analysis of the Death Inducing Signaling Complex (DISC) Using Mass Spectrometry

V. Pham, Z. Jin, R. Pitti, Y. Li, D. Lawrence, A. Ashkenazi, and J. Lill

Protein Chemistry Department, Genentech Inc., South San Francisco, CA

Apoptosis is a genetically controlled process in which cells undergo selfdestruction. Under normal conditions apoptosis eliminates damaged cells. However, many cancer cells escape apoptosis, allowing tumors to survive and grow. Apoptosis is triggered by two major pathways: the extrinsic pathway in response to cell stress such as DNA damage, and the intrinsic pathway in response to the activation of death receptors (DR4 & DR5) by death ligands such as Apo2L/TRAIL.

Upon stimulation with Apo2L, DR4 & DR5 trigger apoptosis through the formation of the DISC which recruits the initiator caspase-8 (casp-8). The DISC mediates self processing of casp-8 into its activated form which is then released to the cytoplasm and activates the effector casp-3 & 7, leading to cell destruction. While DISC formation initiates casp8 stimulation, full activation of this protease requires further molecular aggregation events which are not fully understood. Here we describe the utilization of mass spectrometry (MS) in the analysis of components of the DISC leading to a model of casp-8 ubiquitination and activation in Apo2L-induced apoptosis.

Method: For the identification of the DISC components, H460 cells were treated with Apo2L at 4°C for 2 h, the DISC was isolated by size exclusion chromatography (SEC) and was purified with monoclonal antibodies (MAbs) against DR4 & DR5. The integrity of the DISC was confirmed by casp8 Western blot. Proteins were eluted by boiling in 1X SDS sample buffer containing 10 mM DTT at 95°C for 510 min, followed by alkylation with iodoacetamide at RT for 20 min. Proteins were separated on precast 420% Trisglycine gel. Protein bands were visualized by Coomasie Blue R250 and were excised for ingel tryptic digestion at 37°C overnight. For ubiquitination mapping of casp-8, T7 tagged version of casp-8 was expressed in 293S cells cotransfected with Cullin-3 (CUL3) and ubiquitin (Ub). Lysate was immunoprecipitated (IP) with T7 Ab. Protein was eluted using 50:50 of acetonitrile: 0.1% TFA at RT for 5 min followed by insolution tryptic digestion at 37°C overnight. Peptide mixtures were separated on the nanoAquity UPLC and detected on the LTQ-Orbitrap (Fisher Thermo, San Jose) using a top 10 datadependent mode analysis.

Result: Stimulation of H460 cells with Apo2L resulted in the DISC which eluted at around 800 KDa in SEC. Isolation of the complex using SEC prior to IP eliminated the high abundant background proteins which allowed us to better identify the components of the DISC. MS analysis revealed several known DISC components such as Apo2L, DR4, DR5, FADD, casp-8 & 10, FLIP and E3-ubiquitin ligase subunit cullin 3. When 293S cells were transfected with different plasmids express casp-8, CUL3 and Ub, it was noticed that CUL3 enhanced casp-8 ubiquitination. Co-overexpression of CUL3 with wild type Ub further increased casp-8 ubiquitination. However, mutant Ub where all Lys residues were substituted with Arg did not give the same effect, suggesting that CUL3 promoted polyubiquitination of casp-8. Ubiquitination mapping identified K461 of casp-8 was ubiquitinated.

Based on this result it was proposed that upon Apo2L stimulation, the DISC was formed and CUL3 was recruited to the complex. CUL3-mediated ubiquitination of casp-8 might be responsible for the clustering and aggregation of the complex and full activation of casp-8. Activated casp-8 then activated effector casp-3 & 7 leading to apoptosis.

References

1. Jin, Z., Li, Y., Pitti, R., Lawrence, D., Pham, V., Lill, J., and Ashkenazi, A. (2009) Cell 137, P.721735.

B.4

Proteome-scale Identification of Mitochondrial Transit Peptide Cleavage Sites

E. Crawford¹, and J. A. Wells^1,2

Departments of ¹Pharmaceutical Chemistry and ²Cellular and Molecular Pharmacology, University of California, San Francisco, CA

We have used an N-terminus-specific labeling method to identify over 100 sites of mitochondrial transit peptide removal in human proteins. Until recently, proteome-scale study of proteolysis events was confounded by the lack of a robust labeling technology that could enrich for cleavage products from the profusion of proteins in the cell. Our lab developed a technology to overcome this problem. The engineered enzyme subtiligase ligates biotinylated peptide esters onto protein N-termini liberated by proteolysis. Once labeled, the cleavage products are readily extracted and identified by mass spectrometry. Studies of four human cell lines have revealed over 2,000 native N-termini. Over 100 of these (65% of which were previously unidentified) represent cleavages that occur at unique sites between positions 5 and 120 in mitochondrial proteins, making them likely mitochondrial transit peptide cleavage sites. The prediction program TargetP correctly predicts only 30% of these sites. This work represents a significant increase in the number of mitochondrial transit peptide cleavage sites known, and will lead to an improved description of cleavage site recognition and to improved prediction methods.

B.5

Enrichment and Analysis of Phosphopeptides Shaved from Membrane Fraction of Salt-stressed Arabidopsis

I.-F. Chang¹, J.-L. Hsu², L.-Y. Wang³, and S.-Y. Wang¹

¹Institute of Plant Biology, National Taiwan University, Taipei, Taiwan; ²Graduate Institute of Biotechnology, National Pingtung University of Science and Technology, Pingtung, Taiwan; ³Mass Solutions Technology Co. Ltd., Taipei, Taiwan

Under salt stress condition, plants respond by turning on phosphorylating cascade. Membrane is the site where key phosphorylation events occur. Nevertheless, so far limited phosphorylation sites in response to salt stress in plants have been reported. In the present work, the salt stress-stimulated protein phosphorylation of membrane proteins in Arabidopsis was investigated. Membrane fractions from Control, 200 mM or 400 mM salt-treated Arabidopsis suspension plants were isolated, followed by protease shaving with the assistance of 60% methanol and enrichment by zirconium ion-charged magnetic beads. Through the organic solvent promotion, the membrane proteins were prone to be dissolved in this aqueousorganic solvent system and facilitated to the downstream trypsin digestion. The phosphopeptides from different samples were enriched by Zr4+-IMAC (immobilized metal ion affinity chromatography) magnetic beads in parallel within five minutes. Through this simple procedure, we identified 17 phosphopeptides from 15 membrane proteins in Arabidopsis. These proteins include a sugar transporter, AHA1, and aquaporins. A phosphorylation site of AHA1 was identified in 200 mM salt-treated plants. A phosphorylation site of a sugar transporter was identified both in 200 and 400 mM salt-treated plants. In this study, we successfully identify phosphorylation sites from membranes of specific abiotic-stressed plants by organic solvent-assisted trypsin digestion followed by Zr4+-IMAC enrichment. In addition, we also identified distinct phosphorylation sites between different salt stress conditions. Further study such as more comprehensive mapping of phosphorylation sites and quantitative analysis of protein phosphorylation with and without salt stress are in progress.

B.6

Identification of Protein-Protein Interactions Using Chemical Cross-linking and CID and ETD Tandem Mass Spectrometry

X. Du¹, J. Rorie¹, S. Chowdhury², J. Adkins², G. Anderson², and R. Smith²

¹University of North Carolina at Charlotte, NC; ²Pacific Northwest National Laboratory, Richland, WA

Chemical cross-linking combined with mass spectrometry is a powerful method to identify protein-protein interactions and to study structures of protein complexes. A number of strategies have been reported to take advantage of the high sensitivity and high resolution of mass spectrometry. The Aebersold group developed a method that is based on chromatographic enrichment and targeted sequencing of peptides [1]. The cross-linkers are isotopically coded and special software, xQuest, was developed to perform the spectrum matching. The Bruce group used a chemistry-based approach that took advantage of the mass spectrometry cleavable labile bond in the specially designed cross-linker [2].

We report here a software package that is developed for a label-free approach. This approach can use commercially available cross-linkers and takes advantage of the complementary fragmentation capability, Collision-Induced Dissociation (CID) and Electron Transfer Dissociation (ETD), Indiana, USA a Thermo Orbitrap ETD mass spectrometer to increase the confidence of the cross-linked peptide identifications. In addition, the software puts the inter-, intra-, deadend, and unmodified cross-linked peptides in the same scoring framework. This makes it possible to select the best match with an experimental spectrum if different cross-linked peptides with more than one cross-linking types are matched.

The product spectra produced by fragmenting a cross-linked peptide can be very noisy due to the additional mass of the cross-linker. The software package features a denoising algorithm to remove noise peaks in the MS/MS spectra. It takes into account the fact that different segments of the m/z region of a spectrum can have different signal-to-noise ratio. This denoising feature greatly helps the identification of signal peaks from the background noise. In addition, the software has a builtin visualization capability to validate the match between an experimental and a theoretical spectrum.

The software was developed using C++ and the standard template library under Linux with open source tools to provide a high level of cross-platform compatibility. The additional libraries utilized are available under open source licenses. Visualization is accomplished through the generation of W3C compliant HTML documents that are compatible with most web browsers.

The software has been tested on both commercial and in-house synthesized cross-linkers. The latter features an enrichment tag. The identified protein-protein interactions were validated using nuclear magnetic resonance (NMR) and the software shows great promise to identify proteomewide protein-protein interactions using tandem mass spectrometry.

References

1. Rinner, O., Seebacher, J., Walzthoeni, T., Mueller, L. N., Beck, M., Schmidt, A., Mueller, M., and Aebersold, R. (2008) Identification of crosslinked peptides from large sequence databases. Nat. Methods 5(4), 315–318.

2. Zhang, H., Tang, X., Munske, G. R., Tolic, N., Anderson, G. A., and Bruce, J. E. (2009) Identification of protein-protein interactions and topologies in living cells with chemical cross-linking and mass spectrometry. Mol. Cell. Proteomics 8(3), 409–420.

B.7

Quantitative 2D-Gel Analysis of Isotope Encoded Proteins Using SERRS Detection

G. Knudsen¹, B. M. Davis², S. K. Deb², Y. Loethen², R. Gudihal³, P. Perera², D. BenAmotz², and V. J. Davisson²

¹Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California San Francisco, CA; ²Purdue University, West Lafayette, IN; ³Agilent Technologies India Pvt. Ltd., New Delhi, India

A new strategy for quantification of proteins in biological mixtures using comparative 2D gel electrophoresis and in gel detection by surfaceenhanced resonance Raman (SERRS) is presented. Proteomic samples from HCT 116 human cancer cellular lysates were isotope-encoded by treating with Rhodamine-6G (R6G) lysine-labeling reagents. Isoform mixtures of human GMP synthetase were spiked into samples at controlled levels to assess the accuracy of optical imaging methods to quantify small differences in the isoform populations. Post-separation, dye labeled protein concentrations were estimated by water-referenced fluorescence spectral imaging and protein compositions were estimated from averaged comparisons between gels. The gel embedded proteins were subsequently stained by silver nanoparticle deposition and then imaged by Raman spectroscopy. Quantification of the isotope-edited Raman spectra for the R6G labeled proteins in single gels is used to determine the relative concentrations of protein isoforms. The results represent a prototype for future application of surface enhanced isotope-edited optical spectroscopy to comparative proteomics, as an alternative to mass spectrometric methods for quantifying changes in the distribution of low abundance protein isoforms.

This work was supported by NCI F32-CA123662 (GMK), GM067195-04 and GM053155-10 (VJD), NSF CHE 0455968 (DBA, PP), and GAANN Fellowship U.S. Dept. of Education (BMD).

B.8

Global Analysis of Caspase Proteolysis During Inflammation

N. Agard¹ and J. A. Wells^1,2

Departments of ¹Pharmaceutical Chemistry and ²Cellular and Molecular Pharmacology, University of California, San Francisco, CA

Proteolysis is a prevalent post-translational modification, regulating cellsignaling pathways including apoptosis and inflammation. Misregulation of these pathways has been linked to diverse human diseases including cancer, infection, and inflammatory diseases. However, a molecular connection between proteolytic processing and disease progression is limited by our inability to link proteases to their substrates.

The caspases (cysteine aspartyl proteases) are conserved dimeric proteases known to play essential roles in apoptosis and inflammation. In their bestknown functions, initiator and executioner caspases cleave hundreds of identified substrates to induce the complex physiological responses associated with apoptotic cell death. By contrast the inflammatory caspases (Caspases-1, -4, and -5) have only a few known substrates including very important regulatory cytokines (ILb and_IL18). Both biochemical and cellbiological data suggest that many additional substrates should exist. To identify these substrates we have pursued a proteomic strategy based on enzymatic tagging of free N-termini.

Identification of inflammatory caspase targets can be performed in three steps: 1) selective labeling of proteolyzed proteins; 2) isolation of tagged proteins from uncleaved proteins; 3) MS identification. Towards this end, we have developed an engineered enzyme, termed subtiligase, that selectively biotinylates free N-termini of proteins. The biotinylated proteins can then be isolated and identified. Although all endogenous proteins have an N-terminal amino acid, ~80% of eukaryotic proteins are N-terminally acetylated reducing the background of unproteolyzed proteins, and enhancing our ability to detect new proteolytic events.

The application of substiligase to lysates pretreated with caspase-1 has identified 70 caspase substrates more than half of which have not been identified previously. The sites of cleavage are consistent with previously reported primary sequence specificities of caspase-1, and we identify ~40% of previously validated caspase-1 substrates. A small number of these substrates have been identified after cellular activation of caspase-1 mixed with apparent apoptotic caspase substrates, consistent with previous reports of caspase-1 activating caspase-7.

B.9

Two Mass Spectrometry Based Approaches for the Investigation of the Heparin Interactome

A. Ori, M. C. Wilkinson, and D. G. Fernig

University of Liverpool, United Kingdom

Heparan sulfate proteoglycans (HSPGs) are ubiquitous component of cell surface and extracellular space in metazoans. By interacting with a vast number of protein partners, more than two hundred in humans¹, HSPGs participate in molecular networks regulating complex biological phenomena such as development, immune response and disease.

Here, we present two approaches for the identification of new HS/heparinbinding proteins and for the localisation of heparinbinding sites (HBSs) on protein surfaces using MS-based techniques. First, the plasma membrane proteome isolated from rat liver was fractionated by heparinaffinity chromatography. Low-, medium-, and high-affinity sub-proteomes, as defined by the ionic-strength required for their elution, were characterised by filterassisted sample preparation (FASP) and LCQ-TOF mass spectrometry. Second, a strategy for the selective labelling of residues involved in heparin binding was devised. We demonstrated that the presence of heparin can protect against chemical modification amino acids involved in the interaction and that HBS residues can be selectively labelled after dissociation of the protein from the sugar chain. The identification of labelled peptides from well-characterized heparin-binding proteins by tandem mass spectrometry revealed the ability of this new technique to map low- and high-affinity HBSs. These two complementary approaches can shed light on new functional roles of HSPGs and ease the structural investigation of glycosaminoglycans-protein interactions.

References

1. Ori et al. (2008) The heparanome and regulation of cell function: structures, functions and challenges. Frontiers in Biosci. 13, 4309–4338.

B.10

Electron Transfer Dissociation Analysis of the Urinary Peptidome in Pregnancy

S. Hart¹, R. Blankley¹, and S. Guan²

¹University of Manchester, Manchester Interdisciplinary Biocentre, Manchester, United Kingdom; ²Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA

Introduction: The urinary peptidome has strong promise in clinical studies as an abundant, accessible biological sample. Many published studies rely upon MALDI-ToF profiling to compare normal samples with those taken during disease, without characterisation of the sequences of the large, non-specifically-proteolysed peptides. Electron transfer dissociation offers an effective mechanism to determine the sequence of large, highlycharged peptides [1], and where high-resolution, high-efficiency ion/ion reaction is possible, extended sequence tags can be generated for long polypeptides [2].We propose the use of ETD to complement CID in urinary peptidomics. This work forms part of an ongoing study to investigate the causes of preeclampsia and establish a viable panel of biomarkers predicting development of this lifethreatening condition.

Methods: Urine samples were obtained with consent from pregnant women at 15 and 20 weeks gestation. Proteins (>10kDa) were segregated from low molecular weight components, including peptides and bile salts, using a 10kDa nominal molecular weight cut-off centrifugal filter. Peptides were purified from other low molecular weight components of the raw urine filtrate using solid-phase extraction. Briefly, this involved three stages; an initial reversed-phase (RP) step to pre-concentrate urinary peptides from bulk volume, a strong cation exchange step at low pH to segregate peptides and bile salts, and a final RP polishing step [3].Peptides were then subjected to RP HPLC separation, coupled online to a hybrid linear ion trap-orbitrap with electron transfer dissociation capability. Product ion spectra were collected for each selected precursor using both ETD and CAD.

Preliminary data: ETD measurements were found to yield sufficient product ion spectral information to generate confident identifications for highly-charged precursor ions, bearing up to eight protons, on a chromatographic timescale. CID measurements yielded identification information for shorter polypeptides bearing lower charge, Indiana, USA agreement with our previous studies on tryptic and Lys-C peptides [4]. High-resolution ETD measurements, whilst having reduced sensitivity in comparison to similar measurements made using the linear ion trap, provide additional certainty as to both the charge state of products and their odd- vs. even-electron nature, which varies with both fragment length and sequence composition. For larger precursors, high-resolution ETD yields the most confident identification data, and therefore represents the most effective method currently available to characterise such peptidomes without performing proteolytic digestion. Database searching in the absence of predicted cleavage sites is significantly more effective where high mass accuracy data are available.Datasets such as this will also provide valuable information as to the dissociation behaviour of non-tryptic peptides using different fragmentation methods, which can provide insight as to their underlying mechanisms.

Support for this research was provided by the Bio-Organic Biomedical Mass Spectrometry Resource at UCSF (A. L. Burlingame, Director) through the Biomedical Research Technology Program of the NIH National Center for Research Resources, NIH grants NCRR P41RR001614, RR012961 and S10 RR019934 and by a grant from the Engineering and Physical Sciences Research Council of the United Kingdom, EPSRC grant EP/E043143/1.

References

1. Mikesh, L. M., Ueberheide, B., Chi, A., Coon, J. J., Syka, J. E., Shabanowitz, J., and Hunt, D. F. (2006) Biochim. Biophys. Acta 1764(12), 1811–22.

2. McAlister, G. C., Berggren, W. T., GriepRaming, J., Horning, S., Makarov, A., Phanstiel, D., Stafford, G., Swaney, D. L., Syka, J. E., Zabrouskov, V., and Coon, J. J. (2008) J. Proteome Res.7(8), 3127–36.

3. Cutillas, PRC (In Press) Methods in Mol. Biol.

4. Hart, S. R., Lau, K. W., Hao, Z., Broadhead, R., Portman, N., Hühmer, A., Gull, K., McKean, P. G., Hubbard, S. J., and Gaskell, S. J. (2009) J. Amer. Soc. Mass Spectrom. 20(2), 167–75.

B.11

Non-Targeted Characterization of Glycopeptides in Lysosomal Storage Disease Samples

C. Dorschel¹, R. R. Sprenger², J. M. F. G. Aerts³, J. P. C. Vissers¹, and S. J. Geromanos¹

¹Waters Corporation, Milford, MA; ²University of Southern Denmark, Odense, Denmark; ³University of Amsterdam, The Netherlands

Introduction: The post-translational glycosylation of proteins can have a profound effect on their biological function and structure. Hence the ability to identify and characterize glycosylated peptides in systems samples (protein digests) with minimal manipulation of the starting material is advantageous in systems biology. Here we apply a strategy to target glycopeptides in a discovery experiment by observing characteristic fragment ions in the elevated energy channel time-aligned to eluting glycopeptides derived from plasma samples for controls (n = 8) and Gaucher disease patients pre and post-treatment (n = 12).

Methods: Samples were reduced, alkylated and digested with trypsin following standard procedures. Data were acquired on a nanoACQUITY system using gradient elution of peptides on a C18 column coupled to the nano electrospray source of a Synapt HDMS hybrid quadrupole/time of flight mass spectrometer operated in data independent, alternate scanning mode (MSE). Data were processed in ProteinLynx Global Server version 2.4 with glycopeptide detection enabled. Glycopeptide identification is based on the observation of target oxonium ions in the elevated energy channel, the mass difference between the observed precursor and the candidate peptide corresponding to a reasonable combination of carbohydrates, the observation of the glycopeptide Y1 ion (peptide + HexNAc) in elevated energy, and, for N-glycans, the presence of the NxS/T sequence motif.

Results: Prior efforts to identify glycopeptides in complex mixtures have relied on data dependent schemes. The present work uses the same data independent scheme used to identify ordinary tryptic peptides in digest mixtures to identify glycopeptides as well, without any additional sample preparation or data acquisition.

The identified glycopeptides in the samples were tabulated according to chromatographic retention time and intensity, and subjected to hierarchical clustering. The result of the clustering indicated that the identified glycopeptides provided sufficient differentiation to separate the data according to membership in the control, pre-treatment, Oregon, USA, post-treatment groups, but with data from severely affected patients grouping as outliers. This suggests that glycosylation patterns may offer a means to track severity of disease. The data also show the presence of multiple glycoforms for many of the glycopeptides which show that the extent of glycosylation on these peptides is affected by the presence of disease and whether or not the disease is treated.

Further analysis of these results and consideration of clinical implications will be incorporated in the final presentation.

B.12

A Novel Combinatorial Library/Mass Spectrometric Approach for Detection of Autoantibody Signatures to Cancer-Associated O-glycoprotein Epitopes

S. B. Levery, S. K. Kracun, O. Blixt, E. Clo, K. J. Jensen, and H. Clausen

University of Copenhagen/ICMM, Copenhagen, Denmark

Aberrant post-translational glycosylation of the proteome is a hallmark of cancer, although the functional significance of these changes, including their effects in immunity, are poorly understood. Nevertheless, it has been established that aberrant O-glycosylation of the human epithelial cancer mucin MUC1, for example, allows this molecule to break tolerance and induce humoral immune responses. There is thus considerable interest in discovery of Oglycoprotein epitopes (O-PTMs) recognized by natural antibodies in cancer and other patients, since molecules bearing aberrant O-PTM epitopes could be useful as diagnostic and therapeutic targets. A novel high throughput (HT) methodology, that targets circulating autoantibodies rather than the molecules they recognize, has been developed, incorporating a random combinatorial O-glycopeptide library synthesized in the solid phase on polymer beads. The glycopeptides are attached to the beads with a cleavable linker easily released under mild conditions. Thus, beads showing positive reactivity with patient antisera in an immunofluorescence assay can be selected, the linked O-glycopeptides released, and the sequence of the reactive epitopes determined by mass spectrometric methods. In this case, rapid and sensitive sequencing of released glycopeptides incorporating one or more O-GalNAc residues has been achieved with an LTQ-Orbitrap mass spectrometer (ThermoFisher), configured for both higherenergy collision dissociation (HCD) and electron transfer dissociation (ETD), and equipped with a TriVersa NanoMate ESI-Chip interface (Advion BioSciences). In our hands, initial characterization of the reactive epitope by CID-MS2 of the multiply protonated molecular ion is greatly facilitated by employing HCD mode, which removes the GalNAc residue more effectively than conventional ion trap CID, thus generating more abundant peptide fragment ions for sequencing. In a case where only a single glycosylation site is present, this is sufficient; where multiple potential glycosylation sites are possible, ETD is employed to determine the actual location(s) of O-GalNAc residues in the reactive sequence. For example, with a model test sequence having multiple potential O-glycosylation sites, H2N-VTSAPDTRPAPGST(GalNAc)APPAHG-NH2, the single O-GalNAc site was readily identified in ETD mode by an unambiguous set of c and z fragments, despite the presence of a rather high percentage of Pro residues throughout the sequence. This was achieved quickly and sensitively from peptide released from a single bead reactive with an anti-MUC1 monoclonal antibody. The combinatorial synthesis, screening, and MS analysis are all amenable to automated and/or HT protocols, and the library thus has the potential to incorporate a broad range of structures. It is expected that the approach could have wider applicability to other posttranslational modifications, including phosphorylation and sulfation.

B.13

Peak Extraction and Deisotoping of Low-Level MSMS Spectra from Quadrupole/TOF Instruments

J. Chen*, J. Trinidad, S. Guan, and A. L. Burlingame

Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA

Acquisition of LCMSMS spectra from quadrupole/time-of-flight (QTOF) instruments allows for protein identification, quantification, and PTM site assignment in a proteome scale. Although thousands and perhaps tens of thousands of peptides can be identified from a multi-dimension LCMSMS experiment, by use of database search, precise site assignment for many posttranslational modifications (PTM), especially for labile modification such as phosphorylation reminds challenging. Modified peptides in general are of low stoichiometry and their MSMS peaks have low intensity. MSMS spectra of multiply charged peptide precursor ions contain singly and multiply charged product ions. Current and past data processing algorithms of high throughput mass spectrometry data have largely focused on the interpretation of mass spectra with high intensity. The effective analysis of lower signal tandem mass spectrometry data, however, is crucial to accurately determining accurate sites of PTMs such as phosphorylation. In this work, we extract spectral characteristics, by performing statistical analysis of the identified ions in those low-level MSMS spectra. New algorithms are developed to differentiate low-level signal peaks from noise using "rules" learned from the statistical analysis and to determine the charge state, monoisotopic m/z of signal peaks and to remove other isotope peaks from the spectra. The resulting, simplified data is then used to accurately assign sites of PTMs.

Support for this research was provided by the Bio-Organic Biomedical Mass Spectrometry Resource at UCSF (A. L. Burlingame, Director) through the Biomedical Research Technology Program of the NIH National Center for Research Resources, NIH grants NCRR P41RR001614

*UCSF Intern Program from EVHS, San Jose, California, USA 95148

B.14

Network Analysis of Gcn5 HAT Complexes Refines the Organizational Makeup of the SAGA, SliK (SALSA) and ADA Complexes

M. Sardiu, K. Lee, S. Swanson, J. Workman, L. Florens, and M. P. Washburn

Stowers Institute for Medical Research, Kansas City, Missouri, USA

A significant number of proteins perform their functions in conjunction with other proteins by forming distinct complexes which are responsible for specific processes in a cell. Understanding how protein associate into stable protein complexes is a pivotal part of understanding cellular activity. Here, we employed a combination of computational approaches and a systematic collection of quantitative proteomics data from wild-type and deletion strain purifications in order to decipher the relationships between individual proteins and to perform a comparative structure-function analysis of a complex. We applied this approach to a data set generated from components of the Saccharomyces cerevisiae SAGA (Spt-Ada-Gcn5) histone acetyltransferase (HAT) complex, which consists of distinct domains and functional modules. To group the proteins of the complex based on its abundance patterns, we performed hierarchical cluster analysis on the wild-type and deletion strain purifications. Perturbation of the complex by genetic deletion of several subunits resulted in the dissociation of the Gcn5 HAT complexes into five modules: (1) CORE; (2) DeUB; (3) ADA; and two potential novel modules. Furthermore, we could show that Sgf29, a protein of unknown function, formed a subcomplex with ADA. Strikingly, Sgf29 exhibits similar abundance levels and a comparable phenotype after deletion as other members of the ADA subcomplex, indicating that our approach is able to predict the function of novel proteins within a complex. We also explored the functional role of several interactions within the DeUB module.

In summary, we could show by our integrative study that quantitative proteomics analysis of genetic deletion strains provides valuable insights in both structure and function of a multiprotein complex.

B.15

Quantitative Label-free Analysis of Complex Protein Mixtures through the MFPaQ Software

E. Mouton Barbosa, D. Bouyssié, A. Gonzalez de Peredo, F. Roux-Dalvai, O. Burlet-Schiltz, and B. Monsarrat

Institut de Pharmacologie et de Biologie Structurale, Université de Toulouse, CNRS, Toulouse, France

While nanoLC-MS/MS has become state of the art methodology for systematic identification of large numbers of proteins in complex samples, current methods for quantitative analysis using this approach still largely rely on isotopic labeling and have several shortcomings for clinical studies on biological fluids. Label-free approaches, based on direct comparison of the MS signal intensity of tryptic peptides across different runs, appear to be better suited to the analysis of a large set of clinical samples. However, the comparison of multiple MS runs of highly complex biological samples requires highly reproducible and resolutive systems, as well as dedicated bioinformatic tools. Here, we evaluated a quantitative proteomic workflow based on nanoLC-MS/MS analysis of complex mixtures on an Ultimate3000 Dionex system coupled to an LTQ-Orbitrap-XL mass spectrometer, and bioinformatic processing of data with the new labelfree quantitative module of the MFPaQ v4 software. This module uses an identitybased method, Indiana, USA which the precursor ion m/z and retention time (RT) of each identified peptide are extracted from MS/MS database search result files, and used as a starting coordinate to extract the peptide elution peak in the MS survey scans. We present the performances of this quantitative approach in terms of reproducibility, linearity, and number of quantified proteins for complex protein samples.

Conversely to pattern-based strategies in which LC-MS features have to be defined from the analysis of peptide elution and isotopic profiles in LC-MS maps, the approach used in MFPaQ, based on extracted ion chromatograms (XIC) of identified peptides, is driven by experimentally measured RT and by monoisotopic m/z values validated from MS/MS sequencing. Thus, it allows to perform peak detection in a robust an accurate way. A drawback of this method, however, is that only identified peptides can be quantified. For analysis of highly complex peptide mixtures, Mississippi, USA/MS undersampling thus limits the number of identified and quantified proteins. To circumvent this problem, we implemented an approach based on the extraction of MS signals in individual runs using a previously generated MS/MS identification database containing m/z and RT values associated with peptide sequences. In that way, a large number of additional, nonsequenced peptides MS signals can be extracted by the software. Using this strategy, more than 1300 proteins could be efficiently quantified with good accuracy in replicate injections of a highly complex cellular digest.

B.16

Pigment Epithelium-derived Factor: A Biomarker of Preeclampsia with a Role in the Pathogenesis of the Syndrome

K. Williams¹, Y. Zhou¹, and S. Fisher²

Departments of ¹Obstetrics, Gynecology and Reproductive Sciences and ²Cell and Tissue Biology, University of California, San Francisco, CA

During human pregnancy, the placenta, a transient organ that ultimately grows to the size of the liver, is inserted into the maternal circulation. The requisite vascular connections to the uterine vessels are formed by a unique population of placental cells, derived from the embryo, termed cytotrophoblasts. They have tumor-like invasive properties, evade maternal immune surveillance, and line the arterioles that deliver uterine blood to placental chorionic villi. The villi are covered by fused cytotrophoblasts, syncytiotrophoblasts, that line the intervillous space where maternal blood flows and transport substances to and from the embryo/fetus. Preeclampsia (PE), a pregnancy complication that is characterized by the new onset of maternal high blood pressure, proteinuria, and edema, is associated with a failure in the vascular component of cytotrophoblast invasion and placental hypoxia, which is thought to lead to the maternal endothelial dysfunction that is the hallmark of this syndrome. The molecular mechanisms that link underperfusion of the placenta to the maternal signs of preeclampsia include improper differentiation of invasive cytotrophoblasts, which leads to their dysregulated secretion of pro- and anti-angiogenic factors.

We conducted a pilot study to determine if we could use maternal serum and an isobaric mass tagging strategy to identify biomarkers that are beacons of sPE. In this study, we analyzed depleted serum from women with sPE as defined ACOG and women with uneventful pregnancies and normal deliveries at term. LC MS/MS analyses of these samples revealed 10 proteins with different relative abundances in the two sample sets. APOA-4 concentrations dropped along with anti-thrombin-III, which circulates at lower levels as the clinical signs of this pregnancy complication become more severe. Pigment epithelium derived factor (PEDF), a hypoxia-inducible powerful VEGF inhibitor, was found in higher relative abundance in serum in 5/6 PE patients as compared to controls. These data were validated by antibody-based analyses of placental expression of PEDF. ELISA assays showed that smaller differences in protein abundances (cases vs. controls) were evident at 20 weeks of gestation, and significant differences at time of disease were found at early-onset PE vs. control. Immunolocalization showed that PEDF expression was higher in the chorionic villi of placentas obtained from PE patients. In vivo, PEDF inhibited cytotrophoblast differentiation /invasion, phenocopying PE effects on the placenta.

The complexity of the blood proteome in pregnancy, a portion of which is made up of placental proteins, can be exploited to identify circulating biomarkers that are sentinels of relevant cells such as placental trophoblasts and components of the maternal vasculature. We used information obtained from unbiased approaches to discover biomarkers such as PEDF that have a high degree of relevance to the pathophysiology of PE, expanding the current set of placentaderived molecules with vasculogenic/angiogenic functions that are also candidate PE biomarkers.

B.17

Quantitative Comparison of 13-lined Ground Squirrel Kidney Proteins in Six Stages of Mammalian Hibernation

N. M. Heidlebaugh, S. L. Martin, and L. E. Epperson

University of Colorado School of Medicine, Aurora, CO

Kidneys in hibernators demonstrate remarkable functional plasticity in a seasonal fashion that is not found in non-hibernators. During the summer, animals capable of hibernation are homeothermic with body temperature (Tb) of ~37C and feeding cycles like other mammals, but in winter they fast and transition into a heterothermic state in which Tb is maintained at as low as –2.9C for five to 20 days at a time. These bouts of torpor are concomitant with extreme reductions in heart, metabolic and respiratory rates, and are only interrupted by relatively brief (~11 hr) sojourns at euthermic Tb throughout winter. Each of these interbout arousals is characterized by a natural state of hypoxia during torpor when demand is also reduced, and oxidative reperfusion during rewarming, but no measurable tissue damage is accrued as is typically seen in non-hibernators. Such oxidative damage is prevalent in human kidney transplant recipients and results in severely reduced organ function. In addition, ground squirrels like other hibernators establish a closed system in which nitrogen is conserved and the accumulation of urea is greatly reduced or abolished. The biochemical mechanisms for this conservation and reuse of nitrogen remain to be elucidated. Knowledge of these mechanisms that provide protection during hibernation and by which the animals successfully maintain a closed system will inform approaches to organ cold storage and transplant and are likely to improve outcomes.

A quantitative DIGE approach was used to compare kidney proteins taken at several timepoints of the circannual cycle: early summer, later in summer, entrance into torpor, late in torpor, arousing from torpor and interbout aroused, with a sample size of five or six animals for each stage. Protein samples separated by 2D gel were compared and quantified using DeCyder 7.0 and spot data were exported for statistical analysis in R. Significant differences were determined using ANOVA and BH multiple test correction. Spots with q<0.05, Tukey pairwise p < h 0.05, and fold change greater than 1.8 were picked for identification by LC-MS/MS. Functional implications of steady-state hibernator kidney protein level alterations will be discussed.

Characterization of the Velos, an Enhanced LTQ Orbitrap, for Proteomics

J. V. Olsen^1,2, M. L. Nielsen^1,2, N. E. Damoc³, J. Griep-Raming³, T. Moehring³, A. Makarov³, J. Schwartz⁴, S. Horning³, and M. Mann¹

¹Proteomics and Signal Transduction, Max Planck Institute for Biochemistry, Martinsried, Germany; ²NNF Center for Protein Research, Faculty of Health Sciences, University of Copenhagen, Denmark; ³Thermo Fisher Scientific, Bremen, Germany; ⁴Thermo Fisher Scientific, San Jose, CA

Fast and efficient peptide sequencing is a cornerstone of modern proteomics. We have evaluated and characterized a new and improved version of the LTQ Orbitrap for proteomics experiments. The new instrument is different in three key aspects from the current instrument. It's source has a significantly higher ion transmission, it encompasses a significantly faster scanning linear ion trap and it has much more efficient Higher Energy C-trap Dissociation (HCD). These features – in combination with predictive automatic gain control – and parallel detector acquisition enable short cycle times (2.8 seconds) for Top20 MS/MS acquisition in the LTQ in combination with high-resolution full scan MS detection in the orbitrap. The new instrument also present a more than ten-fold improvement in its HCD capabilities in terms of speed and sensitivity, which in combination with the new brighter ion source allow for routine Top10 MS/MS cycles. The HCD spectra provide ppm-mass accuracy on all fragments spanning the full-mass-range (no low mass cut-off). High-resolution HCD in combination with online HPLC separation allow us to identify and quantify more than one thousand proteins in single runs of a complex peptide mixture. As a consequence of the spectacular HCD performance it is now possible to analyze both precursor and fragment ions with high-mass accuracy on a chromatographic scale without sacrificing acquisition speed or sensitivity. We have analyzed a whole yeast proteome lysate by peptide IEF fractionation in combination with HCD sequencing on the Velos and compared the haploid with the diploid state using SILAC. From three replicate analyses of 24 IEF fractions we are able to identify and quantify almost four thousand proteins with very low false discovery rate.

5.2

Analysis of the Yeast Kinase-Substrate Networks by Quantitative Phosphoproteomics

B. Bodenmiller^1,2, S. Wanka^2,3, C. Kraft⁴, J. Urban⁵, D. Campbell⁶, P. Pedrioli⁴, B. Gerrits⁷, P. Picotti¹, H. Lam⁶, O. Vitek⁸, M.-Y. Brusniak⁶, B. Roschitzki⁷, C. Zhang⁹, R. Schlapbach⁷, K. Shokat⁹, A. Colman-Lerner¹⁰, A. Nesvizhskii¹¹, M. Peter⁴, R. Loewith⁵, C. von Mering³, and R. Aebersold^1,6,12

¹Institute of Molecular Systems Biology, ETH Zurich, Switzerland; ²Zurich PhD Program in Molecular Life Sciences, Switzerland; ³Institute of Molecular Biology and Swiss Institute of Bioinformatics, University of Zurich, Switzerland; ⁴Institute of Biochemistry, ETH Zurich, 8093 Zurich, Switzerland; ⁵Department of Molecular Biology, University of Geneva, Switzerland; ⁶Institute for Systems Biology, Seattle, WA; ⁷Functional Genomics Center Zurich, UZH ETH Zurich, Switzerland; ⁸Purdue University, Departments of Statistics and Computer Science, West Lafayette, IN; ⁹Howard Hughes Medical Institute and Department of Cellular and Molecular Pharmacology, University of California, San Francisco, CA; ¹⁰Facultad de Ciencias Exactas y Naturales, University of Buenos Aires, Argentina; ¹¹Department of Pathology, University of Michigan, Ann Arbor, MI; ¹²Faculty of Science, University of Zurich, Switzerland

Reversible phosphorylation of proteins, carried out by kinases and phosphatases, constitutes one of the most important regulatory mechanisms in eukaryotic cells. The various kinases, phosphatases and their substrates form a network that controls and processes the flow of information, from sensors via signaling relays to effector molecules. This network is of fundamental importance for the function and robustness of essentially all biological systems.

Here, we describe a label-free, quantitative phosphoproteomics approach to determine the in vivo relationship between 97 kinases, 28 phosphatases and their cellular substrates in S. cerevisiae. We systematically detected and identified phosphopeptides that showed a significant and reproducible change in their abundance upon removal (or inactivation) of each of these kinases or phosphatases. In total, we identified over 7,000 regulated phosphorylation events on 1,213 substrate proteins, describing the first system-wide in vivo protein phosphorylation network of S. cerevisiae. Our data indicate the extent of biological activity of each kinase and phosphatase under the tested condition, associates many of them with specific cellular functions, and allows correlation of regulated phosphoproteins with yeast growth and morphological phenotypes. Overall, these results expand the systems-level understanding of phosphorylation-modulated signaling in yeast.

5.3

Quantitative Phosphoproteomics to Define Kinase-Substrate Relationships in Cell Division

J. Villén¹, L. J. Holt², B. B. Tuch², A. D. Johnson², D. O. Morgan² and S. P. Gygi¹

¹Harvard Medical School, Boston, MA; ²Departments of Physiology and Biochemistry & Biophysics, University of California, San Francisco, CA

Protein phosphorylation is a main regulatory switch in the cell, controlling processes such as cell division, growth, proliferation, differentiation and survival. This control is performed by intricate signaling networks, which are capable of altering protein activities and rapidly communicating messages from different extracellular or internal cues to ultimately promote adequate cell readjustments. Numerous studies have addressed protein phosphorylation over the past decades, often on a single protein/pathway level. However, the global picture of signaling events can only be accomplished from comprehensive studies, which are becoming attainable by mass spectrometry (MS)-based proteomics.

The main difficulty in MS large-scale phosphorylation studies is the limitation in detecting phosphorylated species within complex sample mixtures due to their low abundance. However, the past five years have seen a steady improvement in phosphopeptide enrichment and MS data acquisition methods along with the development of computational tools for data analysis and validation, allowing us to routinely identify thousands of phosphorylation events from a single experiment. An overview of these efforts and the current status of technologies to profile the phosphoproteome will be given.

Furthermore, these strategies have been combined with stable isotope labeling for quantitative studies where two cell populations in different phosphorylation status are compared. One of the most challenging problems in signal transduction is establishing kinase-substrate relationships. We have combined chemical genetics and large-scale quantitative phosphoproteomics to identify in vivo substrates of the master mitotic kinase Cdk1 and meiotic kinase Ime2 in the budding yeast Saccharomyces cerevisiae. Using this approach, we expanded the number of known bona fide substrates to ~400 for each kinase, and pinpointed the precise sites of phosphorylation. We observed that substrates for Cdk1 and Ime2 vastly overlap; however the sites targeted and the linear motif preferred seem to differ, which provides an example on how different layers of regulation are assembled in complex systems.

5.4

Global Analysis of Cdk1 Substrate Phosphorylation Sites in vivo

L. J. Holt¹, J. Villén², B. B. Tuch³, A. D. Johnson³, S. P. Gygi² and D. O. Morgan¹

¹Departments of Physiology and Biochemistry & Biophysics, and ³Microbiology and Immunology, University of California, San Francisco, CA; ²Department of Cell Biology, Harvard Medical School, Boston, MA

To explore the mechanisms and evolution of cell-cycle control, we analyzed the position and conservation of large numbers of phosphorylation sites for the cyclin-dependent kinase Cdk1 in the budding yeast Saccharomyces cerevisiae. We combined specific chemical inhibition of Cdk1 with quantitative mass spectrometry to identify the positions of 547 phosphorylation sites on 308 Cdk1 substrates in vivo. Comparisons of these substrates with orthologs throughout the ascomycete lineage revealed that the position of most phosphorylation sites is not conserved in evolution; instead, clusters of sites shift position in rapidly evolving disordered regions. We propose that regulation of protein function by phosphorylation often depends on simple nonspecific mechanisms that disrupt or enhance protein-protein interactions. The gain or loss of phosphorylation sites in rapidly evolving regions could facilitate the evolution of kinase signaling circuits.

Characterization and Quantification of Phosphosites in the Proteome of Human Primary T-Lymphocytes

M. Carrascal, V. Casas, D. Ovelleiro, M. Gay, E. Gelpí, and J. Abiaán

Laboratorio de Proteómica-CSIC/UAB, IIBB-CSIC, IDIBAPS, Barcelona, Spain

T lymphocytes mediate cellular and humoral defense against foreign bodies or autoantigens. Protein phosphorylation-dephosphorylation is involved in many aspects of lymphocyte activation. In this context we are interested in the characterization of phosphoproteins and p-sites in human primary T-cells and we are studying the quantitative changes produced by different activators on the phosphoproteome of these cells. For qualitative analysis we used a strategy based on a multidimensional separation, involving preparative SDS-PAGE for prefractionation, Indiana, USA-gel digestion and sequential phosphopeptide enrichment using IMAC and TiO₂ before LC-MSⁿ analysis in an LTQ linear ion trap. Quantitative analyses are being performed by 2D-PAGE as well as by LC MS/MS using ITRAQ labeling.

Using these procedures we have identified more than 400 high confidence p-sites in resting cells, some of which had not been described previously, and quantified near 150 p-sites in resting lymphocytes vs lymphocytes treated for 4-hours with PMA-ionomicine. The complementarily of both enrichment techniques is also shown.¹

The complete set of data obtained so far is stored in our LymPHOS database that is publicly available at http://lymphos.org.² We have implemented an automatic workflow for the annotation of the database that includes tools for MS data filtering and accurate phosphorylation site assignation. The information stored in the database comprises most experimental and spectrometric data and data analysis information. All the spectra supporting each assignation are stored and presented in graphical form. Experimental data for each experiment is provided in a document that follows PSI MIAPE guidelines.

This data constitutes the only phosphorylation map available for human primary T-lymphocytes. Several novel lymphocyte specific p-sites are described and these could be a source of information for future studies on the role of phosphorylation in T-cell functions and the effect of pharmacological and immunological agonists and conditions in T-lymphocyte activation.

References

1. Carrascal et al. (2008) Phosphorylation analysis of primary human tlymphocytes using sequential IMAC and titanium oxide enrichment. J. Proteome Res.

2. Ovelleiro and Carrascal et al. (2009) LymPHOS: Design of a phosphosite database of primary human T cells. Proteomics

6.2

Analysis of Ubiquitin Chain Editing by Quantitative Mass Spectrometry

D. S. Kirkpatrick¹, L. Phu¹, D. J. Bustos¹, J. R. Lill¹, I. Bosanac², S. G. Hymowitz², V. M. Dixit³, and M. H. Glickman⁴

Departments of ¹Protein Chemistry, ²Protein Engineering, and ³Physiological Chemistry, Genentech Inc., S. San Francisco, CA; ⁴Technion Institute, Haifa, Israel

The assembly of a ubiquitin signal on a protein substrate requires the coordinated actions of E1-activating, E2-conjugating and E3-ligase enzymes. Ubiquitin signals are recognized by ubiquitin receptor proteins that recruit modified substrates to the proteasome for degradation or into a multi-protein signaling/trafficking complexes. Both of these processes are opposed by ubiquitin isopeptides which disassemble ubiquitin signals and prevent ubiquitin dependent processes. Recent evidence suggests that fine tuning of cellular processes by the ubiquitin system can occur through the process of ubiquitin editing or chain remodeling- conversion of one ubiquitin signal into another by concerted disassembly and reassembly efforts. To further investigate this process and evaluate its role in cells, we have implemented the Ubiquitin-AQUA method. Ubiquitin-AQUA measures the amount of each polyubiquitin linkage relative to isotopically labeled internal standard peptides designed toward the tryptic branched signature peptides. Isotopically labeled unbranched peptides from ubiquitin are used to quantify the total amount of ubiquitin. Digested peptides and isotopically labeled standards are analyzed either by narrow window extracted ion chromatograms on a high resolution LTQ-Orbitrap or by multiple reaction monitoring on a QTrap mass spectrometer. To improve analysis of the N-terminus of ubiquitin and as well as signature peptides toward K6-linked and linear polyubiquitin chains, we have defined conditions for oxidizing these Met containing peptides to the sulfoxide and sulfone states. Quantification of total ubiquitin based upon multiple loci is performed to minimize possible interference from complex ubiquitin signals. Using these methods we provide evidence that the majority of substrate bound ubiquitin in cells at the steady state is conjugated to substrates as mono- rather than poly-ubiquitin. Furthermore, by combining quantitative mass spectrometry with ubiquitin linkage specific antibodies, we can show that polyubiquitinated substrates purified from cells may be modified by more than one chain linkage. These observations provide a cellular context to our growing understanding of ubiquitin editing.

6.3

Age Determination in the Adult Human Brain and Body Using Bomb-Carbon

K. L. Spalding¹, O. Bergmann¹, S. Bernard², H. Druid³, B. Buchholz⁴, P. A. Arner⁵, and J. Frisen¹

Departments of ¹Cell and Molecular Biology, ³Forensic Medicine, and ⁵Medicine, Karolinska Institute, Stockholm, Sweden; ²Institut Camille Jordan, University of Lyon, France; ⁴Lawrence Livermore National Laboratories, Livermore, CA

Much of the impetus in regenerative medicine is fuelled by the prospect of promoting cell replacement, or blocking unwanted cell production. Without knowing, however, if a specific cell type is renewed in the healthy or pathological situation, it remains uncertain whether it may be realistic and rational to modulate this process. Despite the importance of this information, remarkably little is known about the age of cells in many regions of the adult human brain and body. This is largely due to difficulties in studying this process in humans. Using recently established methodology, which integrates biomedical approaches with recent developments in nuclear physics, it is possible to establish the turnover of cells in human tissues. By measuring ¹⁴C derived from nuclear bomb tests in DNA, it is possible to retrospectively establish the birth date of cells. Findings of neuronal turnover in the adult human brain, fat cell turnover in humans (lean and obese), and age-determination in humans will be discussed.

Epitope Tagging of Endogenous Proteins in Human Cells for Discovery of Novel Protein/Protein Interactions

T. Waldman¹, J. Kim¹, W. Lane², and F. Bunz^3,

¹Georgetown Medical School/Lombardi Cancer Center, Washington, DC; ²Harvard University, Boston, MA; ³Johns Hopkins School of Medicine, Baltimore, MD

Epitope tagging is a powerful and commonly used approach for studying the physical properties of proteins and their functions and localization in eukaryotic cells. In the case of Saccharomyces cerevisiae, it has been possible to exploit the high efficiency of homologous recombination to tag proteins by modifying their endogenous genes, making it possible to tag virtually every endogenous gene and perform genome-wide proteomics experiments. However, due to the relative inefficiency of homologous recombination in cultured human cells, epitope-tagging approaches have been limited to ectopically expressed transgenes, with the attendant limitations of their nonphysiological transcriptional regulation and levels of expression. To overcome this limitation, a modification and extension of adeno-associated virus-mediated human somatic cell gene targeting technology is described that makes it possible to simply and easily create an endogenous epitope tag in the same way that it is possible to knock out a gene. Using this approach, we have created and validated human cell lines with FLAG-tagged alleles of the p53 and PTEN tumor suppressor genes in a variety of untransformed and transformed human cell lines. FLAG immunoprecipitation followed by mass spectrometry enabled the identififcation of both known and novel p53 and PTEN interacting proteins. This straightforward approach makes it possible to study the physical and biological properties of endogenous proteins in human cells without the need for specialized antibodies for individual proteins of interest.

C.2

Quantitative Analysis of the Phosphoproteome of Resting and Activated Human Primary T Cells

P. Ruperez¹, J. A. Oses-Prieto¹, A. Gago², and A. L. Burlingame¹

¹Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA; ²Department of Analytical and Food Chemistry, University of Vigo, Vigo, Spain

Protein phosphorylation-dephosphorylation events play a primary role in regulation of almost all aspects of cell function including signal transduction, cell cycle or apoptosis. The activation of the immune response mediated by T lymphocytes is dependent on changes of the phosphorylated state of different proteins. Consequently, Indiana, USA order to understand the functionality of lymphocytes during the immune response we need to study the changes in phosphorylation profiles that occur in these cells in response to activation.

Here we present a quantitative mass-spectrometry based analysis of the phosphoproteome of primary T cells, comparing resting cells with cells stimulated for 5 min using an anti-CD3 antibody. Relative phosphopeptide levels were quantified by stable isotope labeling methodology with i-TRAQ reagent, using 2 mg of total protein amounts in the combined extract. Titanium dioxide chromatography for phosphopeptide enrichment was performed followed by a fractionation by strong cation exchange chromatography. Fractions were analyzed by nano-LC-ESI-Qq-TOF mass spectrometry on a QStarElite instrument. A total of 139 phosphopeptides were identified from 101 proteins. Changes in relative levels were detected in 59 phosphopeptides; 43 of them increased in the activated T cells and 16 increased in the resting ones. All of the phosphorylation sites identified were on serine and threonine.

Among all the proteins characterized we found several involved in the cytoskeleton reorganization, signal transduction, kinase activity, RNA binding. Some of these proteins are described to be part of the early events of T cell activation.

Research supported by the Bio-Organic Biomedical Mass Spectrometry Resource at UCSF(A. L. Burlingame, Director) through the Biomedical Research Technology Program of the NIH National Center for Research Resources, NIH NCRR P41RR001614 and NIH NCRR RR012961.

C.3

ATAQS Computational Software Tool for High Throughput Transition Optimization and Validation for SRM

Mi-Youn Brusniak¹, D. Campbell¹, J. Chen¹, M. Christiansen¹, E. Deutsch¹, C. Kwok¹, S. Letarte¹, H. Ramos¹, P. Picotti², L. Reiter², J. Watts¹, and R. Aebersold²

¹Institute for Systems Biology, Seattle, WA; ²ETH - Swiss Federal Institute of Technology, Zurich, Switzerland

SRM has recently emerged as a popular approach for targeted proteomics since it provides improved dynamic range, sensitivity and reproducibility vs. shotgun based identification and quantification of peptides. SRM uses a tandem-in-space mass spectrometer, which provides two levels of mass screening (i.e. precursor and fragment ion) yielding better sensitivity and dynamic range. SRM experiments are thus hypothesis-driven, since they require a priori knowledge of the target peptides of interest, the mass of the precursors, and the most intense fragment ions for each target peptide.

There is open source software like TIQAM-ProteinDigestor, for generating in silico digestion and fragmentation transitions and TIQAMViewer, for assisting users with validation of transitions by visualizing MS2-triggered SRM scans. However, these tools lack high throughput validation capability, and do not support heavy and isotopically normal peptide pair measurements for better identification and quantification. To the best of our knowledge, there is no open source software that provides for high throughput methods for validating measured transitions, based on sample type and complexity. SRM validation and quantification using current commercial software obtainable from MS vendors is typically a highly manual process, which is also subjective, and thus inherently variable.

We introduce a new software pipeline: ATAQS (Automated and Targeted Analysis with Quantitative SRM). ATAQS provides an easy to use web browser interface to generate and filter candidate peptides of interest, and to generate, filter and validate peptide transitions. ATAQS also introduces a new proposed file format:TraML (Transition Markup Language) as a common data exchange format for validated transitions by downloading published validated transitions, and publishing your own validated transitions to a public SRM database such as MRMAtlas4.

ATAQS provides algorithms to generate decoy SRM transitions for scoring and validation of transitions in automated manner. Finally, ATAQS provides a flexible pipeline for end users by allowing the workflow to start or end at any point of the pipeline, and for computational biologists by easy extension of java algorithm classes for their own algorithm plug-in classes.

C.4

Mass-Spectrometric Identification and Relative Quantification of N-linked Cell Surface Glycoproteins

B. Wollscheid¹, D. Bausch-Fluck¹, C. Henderson², R. O'Brien², M. Bibel³, R. Schiess¹, R. Aebersold^1,2, and J. D. Watts²

¹ETH - Swiss Federal Institute of Technology, Zurich, Switzerland; ²Institute for Systems Biology, Seattle, Washington, USA; ³Novartis Institutes for Biomedical Research, Basel, Switzerland

Although the classification of cell types often relies on the identification of cell surface proteins as differentiation markers, flow cytometry requires suitable antibodies and currently permits detection of only up to a dozen differentiation markers in a single measurement. We use multiplexed mass-spectrometric identification of several hundred N-linked glycosylation sites specifically from cell surface-exposed glycoproteins to phenotype cells without antibodies in an unbiased fashion and without a priori knowledge. We apply our cell surface-capturing (CSC) technology, which covalently labels extracellular glycan moieties on live cells, to the detection and relative quantitative comparison of the cell surface N-glycoproteomes of T and B cells, as well as to monitor changes in the abundance of cell surface N-glycoprotein markers during T-cell activation and the controlled differentiation of embryonic stem cells into the neural lineage. A snapshot view of the cell surface N-glycoproteins will enable detection of panels of N-glycoproteins as potential differentiation markers that are currently not accessible by other means.

References

1. Wollscheid et al. (2009) Mass-spectrometric identification and relative quantification of N-linked cell surface glycoproteins. Nat. Biotechnol. 27(4), 37886.

C.5

Chronic Ethanol Feeding Affects Proteasome Interacting Proteins

M. P. Bousquet-Dubouch¹, S. Nguen², D. Bouyssié¹, O. Burlet-Schiltz¹, S. W. French², B. Monsarrat¹, and F. Bardag-Gorce²

¹CNRS IPBS, Toulouse University, France; ²Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Torrance, CA

Studies on alcoholic liver injury mechanisms show a significant inhibition of the proteasome activity. To investigate this phenomenon, we isolated proteasome complexes from the liver of rats fed ethanol chronically, and from the liver of their pairfed controls, using a nondenaturing multiple centrifugations procedure to preserve Proteasome Interacting Proteins (PIPs). Isotope-Coded Affinity Tagging (ICAT) and MS/MS spectral counting, further confirmed by Western blot, showed that the levels of several PIPs were significantly decreased in the isolated ethanol proteasome fractions. This was the case of PA28a/b proteasome activator subunits, and of three proteasome-associated deubiquitinases, Rpn11, ubiquitin C terminal hydrolase 14 (Usp14), and ubiquitin carboxyl-terminal hydrolase L5 (UCHL5). Interestingly, Rpn13 C-terminal end was missing in the ethanol proteasome fraction, which probably altered the linking of UCHL5 to the proteasome. 20S proteasome and most 19S subunits were however not changed but Ecm29, a protein known to stabilize the interactions between the 20S and its activators, was decreased in the isolated ethanol proteasome fractions. It is proposed that ethanol metabolism causes proteasome inhibition by several mechanisms, including by altering proteasome interacting proteins and proteasome regulatory complexes binding to the proteasome.

C.6

Phosphorylation Dynamics at Synapses in the Central Nervous System

J. C. Trinidad¹, A. Thalhammer², R. Schoepfer², and A. L. Burlingame¹

¹Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA; ²Department of Pharmacology, University College London, United Kingdom

Glutamate is the main excitatory neurotransmitter in the central nervous system. This neurotransmitter is detected by ligand-gated ion channels in the membrane of postsynaptic neurons. These receptors are associated with a complex protein network known as the postsynaptic density (PSD). The PSD contains proteins of many functional classes including kinases/phosphatases, scaffolding molecules, and structural molecules. In addition to its primary role of mediating fast synaptic transmission, the PSD is a dynamic structure that alters its composition over longer time periods. Regulation of protein levels and post-translational modifications affects the efficacy of future synaptic transmission events and plays a role in such processes as learning and memory. In addition, many nervous system diseases affect synaptic protein levels and phosphorylation states.

Using an established two-step sucrose density fractionation protocol, we have isolated postsynaptic density fractions. In previous work, we demonstrated that tryptic digestion of these PSD preparations, followed by a multidimensional LC-MS/MS approach, can allow for the identification of over 1200 proteins. In addition, we have also previously demonstrated the methodology to isolate and identify over 1400 phosphorylated peptides from these samples, by specifically enriching for such peptides using either immobilized metal affinity chromatography or titanium dioxide beads.

Using stable isotope quantitative mass spectrometry, we profiled molecular changes following synaptic activity in cortical neurons. A mouse model of epilepsy was used, where seizures were induced by intraperitoneal injection of the muscarinic acetylcholine agonist, pilocarpine. Post-synaptic densities were isolated at time-points between zero and sixty minutes post-injection. The molecular composition of synapses was found to change in a complex fashion after seizure. Subsets of proteins displayed distinct patterns of localization at the synapse. Within a given subset of proteins, expression patterns were highly correlated, while different subsets showed distinct patterns. At the level of protein phosphorylation, we observed a greater than 2-fold increase in phosphorylation of serine 845 on the GluR1 receptor, consistent with published data. Overall, less than 5% of all phosphorylation sites had robust changes in response to seizure.

This work has been supported by the Wellcome Trust and BBSRC (to RS) and the Bio-Organic Biomedical Mass Spectrometry Resource at UCSF (A. L. Burlingame, Director) through the Biomedical Research Technology Program of the NIH National Center for Research Resources, NIH NCRR P41RR001614.

C.7

Linaclotide, a Novel Peptide Therapeutic Agent in Clinical Development for the Treatment of IBS-C and Chronic Constipation is Digested in the Mouse and Human Small Intestine to Small Peptides

M. Kessler, R. W. Busby, J. D. Wakefield, W. P. Bartolini, P. Germano, A. P. Bryant, C. B. Kurtz, and M. G. Currie

Ironwood Pharmaceuticals, Inc., Cambridge, MA

Linaclotide (MD-1100) is an investigational peptide therapeutic agent in clinical development for the treatment of irritable bowel syndrome with constipation (IBS-C) and chronic constipation (CC). In Phase 2 clinical trials, linaclotide significantly reduced abdominal pain, abdominal discomfort, and bloating and improved bowel function. The most common adverse event in linaclotide-treated patients was diarrhea. This 14 amino-acid peptide contains three disulfide bonds and is a potent agonist of the receptor guanylate cyclase C (GC-C) located in the intestinal lumen. Linaclotide is locally acting, has low oral bioavailability in animals, and no detectable systemic exposure in humans at therapeutic doses. Linaclotide has an active metabolite, MM-419447, which lacks the C-terminal tyrosine. Both of these peptides are stable in simulated gastric fluid and highly resistant to proteolysis under oxidative conditions. Previous studies have shown that linaclotide incubated in the presence of fluid taken from rat small intestinal ligated loops (RIF) is metabolized to MM-419447 and then the disulfide bonds in both peptides are enzymatically reduced. The reduced peptides are proteolytically degraded to small peptide fragments and ultimately to L-amino acids. The present studies were conducted to determine the metabolic fate of linaclotide in mouse intestinal fluid (MIF), obtained from ligated intestinal loops, and in human intestinal fluid (HIF), obtained from the luminal contents of cadavers. Incubation of linaclotide in MIF resulted in rapid degradation with a t1/2 of 3 min and immediate formation of MM-419447; however, this metabolite was also degraded and was not detectable after 60 min. Linaclotide is degraded in HIF with the immediate formation of MM-419447, followed by degradation to small peptide fragments. These peptide fragments were also observed when linaclotide is metabolized in RIF. HIF contains free thiols in millimolar concentrations and the components of the glutaredoxin/glutathione reductase system, as is the case with RIF. Even with rapid metabolic degradation in mice, orally administered linaclotide significantly increased intestinal fluid secretion and intestinal transit, and it decreased visceral hypersensitivity. This indicates sufficient intestinal residence time for the parent and/or active metabolite to bind to GC-C receptors and mediate the effects described on GI function. In conclusion, these studies support a linaclotide metabolism/digestion pathway in mice and human similar to the pathway observed in rats that involves formation of the active metabolite MM-419447 and reduction of the disulfide bonds in both linaclotide and MM-419447, followed by subsequent proteolytic degradation. Linaclotide is present long enough to have an effect, but its low permeability and rapid degradation result in no detectible exposure as demonstrated in clinical studies.

C.8

Large Scale Multiplex Stable Isotope Dimethyl Labeling Applied to the Quantitative Analysis of Tyrosine Phosphorylation

P. Boersema¹, L. Y. Foong², V. Ding², S. Lemeer¹, S. Mohammed¹, R. Raijmakers¹, B. van Breukelen¹, J. Boekhorst¹, A. B. H. Choo², and Albert J. R. Heck¹

¹Utrecht University, Utrecht, The Netherlands; ²A*STAR (Agency for Science, Technology and Research), Singapore

Introduction: Accurate quantification of protein expression in biological systems is an increasingly part of proteomics research. Incorporation of differential stable isotopes in samples for relative protein quantification has been widely used. Stable isotope incorporation at the peptide level using dimethyl labeling is a reliable, cost-effective and undemanding procedure that can be easily automated and applied in high-throughput proteomics experiments and it uses inexpensive reagents and is applicable to virtually any sample(1). Here, we developed several labeling approaches and apply the method to investigate phospho-tyrosine signaling in human cells.

Methods: Cell lysate digestion derived peptides are dimethyl labeled by reductive amination with isotopomers of formaldehyde (CH2O, light; CD2O, intermediate; 13CD2O, heavy) followed by reduction with cyanoborohydride (light and intermediate) or cyanoborodeuteride (heavy). Small sample amounts (< 1 µg) are labeled while trapped on a nanoLC reversed phase (RP) column that is connected online with a mass spectrometer. Larger protein amounts are labeled in-solution or on-column using RP SPE columns. Peptides derived from digestion of HeLa cells that were mock or pervanadate treated or stimulated with EGF for 0, 10 or 30 min were differentially labeled and combined. Immunoprecipitation of tyrosine phosphorylated peptides was achieved using an immobilized antibody against tyrosine phosphorylation. The eluate was analyzed by LC-MS without further affinity enrichment.

Results: We developed in-solution, online and on-column protocols for stable isotope dimethyl labeling of sample amounts ranging from sub-micrograms to milligrams in a very quick and undemanding manner (1). The online and on-column methods combine the sample clean-up and dimethyl labeling steps thereby minimizing sample loss and speeding up the sample handling process. The online approach is particularly suited for minute amounts of protein since it is performed on a nano-LC system that is online with a mass spectrometer. On-column dimethyl labeling is suited for larger (milligrams) protein amounts and allows consecutive 2D-LC separation. The in-solution approach can be used when several samples are to be labeled in parallel. We applied our methods to the quantification of tyrosine phosphorylation upon pervanadate treatment and EGF stimulation. Immunoprecipitation of differentially labeled tyrosine phosphorylation allowed the quantification of differences in tyrosine phosphorylation of respectively 128 and 73 unique phosphotyrosine peptides. Most of the upregulated phosphotyrosine peptides are involved in the EGF signaling pathway, validating our approach. However, one new site and several others that have not been firmly established earlier to be involved in EGF receptor signaling were also identified. In general, we show that the combination of immuno-affinity purification of tyrosine phosphorylated peptides with large scale stable isotope dimethyl labeling and only a single LC-MS run provides a cost-effective approach to obtain a rather complete qualitative and quantitative picture of a signaling event.

References

1. P. J. Boersema et al. (2009) Multiplex peptide stable isotope dimethyl labeling for quantitative proteomics. Nat. Protoc. 4, 484–494.

C.9

Structural Characterization of Novel Components from the Venom of the Mexican Scorpion Vaejovis mexicanus smithi by Electron Capture Dissociation and Electron Transfer Dissociation

S. P. Salas-Castillo, K. F. Medzihradszky, S. Guan, and A. L. Burlingame

Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California San Francisco, San Francisco, CA

Polypeptides from animal venoms, such as of snakes, spiders, conus snails, scorpions or insects display remarkable biological activities. For example, antimicrobial, enzymatic and ion channels blocking properties. For this reason these polypeptide toxins may be excellent drug candidates and have great potential in agricultural applications. They could have use as tools for biological research as well.

Amino acid sequences and secondary structures (if any) of these toxins have to be deciphered in order to understand the structure-biological activity relationship. In most cases, we have to rely on de novo sequencing for a majority of species because genomic information is not available. Mass spectrometry has been proven a powerful tool for such purposes. However, sequencing of these toxic polypeptides presents a special challenge. They are relatively large peptides (~3–8 kDa), may not contain specific cleavage sites evenly spread, and may feature unexpected and unusual post-translational modifications. At the same time the very same properties may make them excellent candidates for intact peptide de novo sequencing.

This work presents the characterization of some components of the crude venom from the Mexican scorpion Vaejovis mexicanus smithi. ECD and ETD analyses were utilized for sequence determination from the intact polypeptides. We compare the utility of the two dissociation techniques. We illustrate some of the difficulties in peak-picking and fragment assignment. Last but not least, we present that once tentative sequence had been determined, an in-house developed software, FAVA will help to decipher which possibility corresponds to the correct structural assignment.

Research support was provided by the Bio-Organic Biomedical Mass Spectrometry Resource at UCSF (A. L. Burlingame, Director) through the Biomedical Research Technology Program of the NIH National Center for Research Resources, by the grants NIH NCRR RROO1614, NIH NCRR RR019934, NIH NCRR P41RR001614 and NIH NCRR RR015804.

C.10

Proteomics Analysis Reveals a New Player Involved in DNA Repair and/or Replication

J. Gilmore, M. E. Sardiu, S. Venkatesh, B. Stutzman, and M. P. Washburn

Stowers Institute for Medical Research, Kansas City, MO

DNA processes including replication, repair and transcription are tightly regulated by chromatin structure. During DNA replication in eukaryotes, the cellular machinery performing these tasks needs to gain access to the DNA. Likewise, chromatin structure provides an obstacle to the DNA repair machinery attempting to gain access the DNA lesion. Experimental evidence has led to the hypothesis that the histone octomer is removed from the DNA by histone chaperones and chromatin remodeling complexes during theses processes (Chromatin disassembly), and placed back on the DNA after the process is complete (Chromatin assembly). MudPIT analyses of all four purified histones were conducted and many proteins involved in DNA repair/replication were identified. In addition, a protein of unknown function designated YDL156w (Cmr1) was also identified in the analyses. Next, MudPIT analyses were performed on purified Cmr1 and on other proteins that were relatively abundant in the Cmr1 pull down. Hierarchical clustering analysis was performed on the relative protein abundances expressed as dNSAFs. Results indicate that Cmr1 is in close proximity to proteins involved in DNA repair and replication. GOstat analysis was performed on the gene products identified from MudPIT analysis of purified Cmr1. The collection of proteins pulled down by Cmr1 was enriched for several biological processes including DNA damage response and chromatin assembly/disassembly. Cmr1 associated proteins were confirmed using size exclusion chromatography in conjunction with MudPIT analyses, and suggests that Cmr1 does associate with proteins involved in DNA replication and/or repair. Furthermore, proteinprotein interactions with Cmr1 are being validated using Co-IPs.

C.11

Quantitative Comparison of the Shigella dysenteriae Proteome Assessed by Two Label-free Global Profiling Methods, APEX and 2D gels

S. Kuntumalla¹, J. C. Braisted¹, S.-T. Huang¹, P. P. Parmar¹, D. J. Clark¹, H. Alami¹, Q. Zhang², A. Donohue-Rolfe², S. Tzipori², R. D. Fleischmann¹, S. N. Peterson¹, and R. Pieper¹

¹Pathogen Functional Genomics Resource Center, J. Craig Venter Institute, Rockville, MD; ²Cummings School of Veterinary Medicine, Tufts University, North Grafton, MA

The human pathogen Shigella dysenteriae serotype 1 (SD1) is the most virulent of the four Shigella serotypes and is a causative agent of shigellosis. In this study, the stationary phase proteome of SD1 was quantitatively analyzed in Coomassie Brilliant Blue (CBB)-stained 2D gels. More than four hundred and fifty proteins, of which 271 were associated with distinct gel spots, were identified. In parallel, we employed 2D-LC-MS/MS followed by the label-free computationally modified spectral counting method for Absolute Protein Expression Measurements (APEX). Of the 4502 genome-predicted SD1 proteins, 1148 proteins were identified with a false positive discovery rate of 5% and quantitated using 2D-LC-MS/MS and APEX. The dynamic range of the APEX method was approximately one order of magnitude higher than that of CBB-stained spot intensity quantitation. A Spearman rank correlation analysis revealed a reasonably good correlation (Rs = 0.81) for protein quantities surveyed by both methods. The correlation was decreased for protein subsets with specific physicochemical properties, such as low Mr values and high hydropathy scores. To assess APEX abundance measurements in the context of subunit stoichiometries for characterized multi-subunit protein complexes, we compared protein abundance ratios from the SD1 APEX dataset with stoichiometric ratios designated for orthologous E. coli protein complexes. A high correlation was observed for subunits of soluble cellular protein complexes in several cases, including chaperones and proteins involved in energy metabolism. The observed APEX stoichiometry was also indicative of the biological function and dynamics of protein complexes such as peroxidases, where the examined stationary phase of SD1 cells appeared to favor the reduced, active state linked to substrate reduction, thus demonstrating versatile applications of the APEX methodology in quantitative proteomics.

This work was supported by the National Institute of Allergy and Infectious Disease, National Institutes of Health, Department of Health and Human Services (NIAID/NIH/DHHS) under contract number N01-AI-15447 to the Pathogen Functional Genomics Resource Center at JCVI. At Tufts University, this project has been funded in whole or in part with Federal funds from NIAID/NIH/DHHS under contract number N01-AI-30050.

C.12

Characterization of Protein Kinase C-catalyzed CYP3A4 Phosphorylation by LC Tandem Mass Spectrometry

Y. Q. Wang¹, S. Guan², A. L. Burlingame², and M. A. Correia²

¹Department of Cellular and Molecular Pharmacology and ²Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA

CYP3A4, a major human liver P450, is degraded via the ubiquitin-dependent 26S-proteasomal pathway. In the course of this degradation, the protein is phosphorylated and this phosphorylation is enhanced after its structural inactivation by cumene hydroperoxide (CuOOH). We have previously identified three sites: S478 phosphorylated by undefined liver cytosolic kinases, T264 and S420 by protein kinases (PKA and PKC). To determine the physiological relevance of this CYP3A4 phosphorylation we heterologously expressed wild type and CYP3A4 with single, double or triple Ala-mutations of these residues in S. cerevisiae. Our findings indicated that the CYP3A4 S478A mutant was significantly stabilized in yeast strains and this stabilization was further synergized by coupled T264A and S420A mutations. However, these mutations greatly reduced, but did not totally abolish CYP3A4 phosphorylation. To further obtain a more complete characterization of CYP3A4 phosphorylation, we applied mass spectrometric analyses to identify the CuOOH-inactivated CYP3A4 phosphorylation sites after in vitro incubation with several kinases including PKC. Six new sites phosphorylation signals of S116, S119, S131, T92, T284, and S259 together with two known sites of T264 and S420 have been identified thus far. A label free quantification method is currently applied to estimate stoichiometries of phosphorylation at those sites and these findings will aid in further investigation of functional significance of protein phosphorylation for CYP3A4.

Research was supported by NIH grant GM44037, NIH NCRR P41RR001614, NIH NCRR P41RR001614 and NIH NCRR RR019934.

C.13

Analysis of H5N1 Influenza Hemagglutinin Glycosylation by LC/MS/MS Utilizing Hydrazide Capture SPE and HILIC Separation of Intact Glycopeptides

T. A. Blake, T. L. Williams, J. L. Pirkle, and J. R. Barr

Centers for Disease Control and Prevention, Atlanta, GA

Influenza hemagglutinin (HA) is the antigenic glycoprotein that binds and fuses to the cell during infection and is the primary component of seasonal vaccines. Glycosylation of HA is thought to affect the virulence of an influenza strain by interfering with the cell recognition binding site or by masking antigenic regions of the protein. Identification of differences among strains in the number/conservation of glycosylation sites as well as the size/complexity of the glycans themselves is necessary for characterizing HA from emerging strains and for determining the efficacy of alternative virus propagation systems for vaccine production. This work represents a first step towards a platform for investigating differences in HA glycosylation due to virus propagation conditions.

Whole virus reagents from three strains of H5N1 influenza (rgA/Vietnam/1203/2004; A/Indonesia/05/2005; A/Bar-headed goose/Qinghai Lake/1A/05) and a reassortant virus (Ind05/PRB-RG2) grown in embryonated chicken eggs were analyzed. Solid-phase extraction (SPE) via hydrazide capture was utilized to specifically isolate tryptic glycopeptides in order to determine glycosylation site occupation via deglycosylation and subsequent reverse-phase liquid chromatography-tandem mass spectrometry (LCMS/MS). Hydrophilic interaction liquid chromatography (HILIC) was also utilized to improve LC-MS/MS analysis of intact glycopeptides. De novo peptide sequencing was used for peptide confirmation. ExPASy's GlycoMod Tool was utilized to aid in the interpretation of intact glycopeptide MS/MS data.

All six predicted N-linked glycosylation sites within the N-terminal ectodomain of HA were found to be occupied for the reagent strains examined. We identified the presence of glycosylation site 3 for the bar-headed goose strain, even though the exact protein sequence for the selected bar-headed goose strain was not in the database and all other strains within this subset did not predict the presence of this site. This methodology also determined occupied glycosylation sites when the predicted site was ambiguous (i.e. NNST). This approach was then used to propose compositions for multiple glycoforms at the occupied glycosylation sites on HA from the reassortant strain (Ind05/PRB-RG2).

We have applied an approach for determining N-linked glycosylation site occupation and investigating the glycans attached to those sites for tryptic digests of influenza virus samples via selective sample preconcentration and LC/MS/MS analysis. We have used this data on glycosylation site occupation/conservation to assist in the examination of intact glycopeptides generated from a tryptic digest of a reassortant H5N1/H1N1 influenza virus grown in eggs. We have proposed potential sugar compositions for the multiple glycoforms associated with the occupied glycosylation sites on HA. This approach can also be utilized to determine changes in glycosylation incurred by changes in the virus propagation systems used. The information gained from these experiments could then be utilized to examine the relationship between changes in HA glycosylation and changes in virulence/antigenicity.

C.14

Mapping the Chromatin Structure Controlling Life Cycle Progression of the Human Malaria Parasite

A. Saraf¹, S. Cervantes², L. Florens¹, and K. Le Roch²

¹Stowers Institute for Medical Research, Kansas City, MO; ²University of California, Riverside, CA

Post-translation modifications (PTMs) of histones are known to regulate chromatin structure and function in a range of organisms from yeast to human. However, not much is known about how chromatin structure and more specifically nucleosome positioning and occupancy may direct major developmental decisions during the course of the parasite infection cycle in Plasmodium falciparum. While microarray analyses have amassed a great deal of information regarding changes in steady-state mRNA levels, the mechanisms underlying transcriptional regulation are still poorly understood. The paucity of transcription factors identified to date and the absence of apparent tight positive and negative feedbacks in presence of a stress suggest that apicomplexan parasites may use a more global strategy to control their transcriptional levels when compared to their higher eukaryotic counterparts.

Our goal in this study is to identify all potential histone post-translational modifications throughout the parasite erythrocytic cycle. Plasmodium histones were extracted using the standard acetic acid method at the seven main erythrocytic stages. Peptides mixtures generated from digestions with enzymes of diverse specificities were analyzed by Multidimensional Protein Identification Technology (MudPIT) coupled to high resolution tandem mass spectrometry. The dymanics of histone modifications across these seven time points of the erythrocytic cycle were quantitatively analyzed. Many histone posttranslational modifications were identified on all detected parasite histones. In particular, we reproducibly detected the majority of the well-known acetylation events on the lysine residues located in the N-terminal tails of H2B, H3 and H4. We used local spectral counts to estimate how much a particular residue was modified. These acetylations were detected at high levels in agreement with the literature, illustrating how comprehensive and quantitative our approach is.

Our key hypothesis is that histone code as well as nucleosome occupancy play a critical role in controlling the parasite infection throughout the malaria cell cycle. While all eukaryotes undoubtedly employ mechanisms that control chromatin structure and function in their transcriptional mechanisms, the characterization of significant differences between apicomplexa and their human host will open the door to novel targets for chemotherapeutic intervention.

C.15

Developing Optimized Sensitivity for Proteomic Characterization of Minimal Numbers

J. A. Oses-Prieto, D. Maltby, and A. L. Burlingame

Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA

Amounts of material available for proteomic analysis can be very limited. Samples of limited abundance are challenging systems to characterize by mass spectrometry. Internal complexity can still be very high, and absolute sensitivity limits of the instruments for pure components can be greatly altered in the context of a complex mixture. However, a reduction of the inherent complexity of the system may not be a feasible option for small samples. Adsorptive losses during traditional purification/enrichment techniques will represent a significant fraction of the theoretically available material, so final recovery can be below the detection limit of analytical systems. As a consequence, fairly complex but low abundance samples might need to be analyzed without much previous enrichment of particular components. Examples of these types of experimental models are cell types available only in very limited amounts: primary cells present in low number in individuals or cells difficult to culture. The goal of this work is to estimate the analytical limits for qualitative sample characterization imposed by the use of small amounts of sample with rather high internal complexity. We are performing some experiments to determine the number of components that we can identify in total, minimally processed cell lysates, Indiana, USA order to analyze the dependence of the number of components identified on the initial amount of sample.

Our aim is to determine minimal sample size to use without compromising the amount of information we can extract from the sample. As a model system, we use a human embryonic stem cell line H9 (WA09 from Wicell), differentiated as neural precursors. We have analyzed aliquots of tryptic protein digests corresponding to small numbers of these cells (500–5000). We have evaluated the benefits of integration of new chromatographic equipments with more reproducible and precise retention times, and mass spectrometry technologies with higher detection sensitivity and improved mass measurement accuracy. In these aliquots we can still identify more that 1000 peptides from several hundred proteins separated in a 60 min gradient acetonitrile in a reverse phase column. We intend to create an accurate mass-retention time database of peptides and to define a template of the components that we can consistently identify in tryptic digests of this cell type using these small size samples. According to our data, since there is only partial reproducibility of the data between runs (60%), it is possible to perform repetitive analysis using minimal material in a discovery effort to increase the number of components identified on the sample. Characterization of the system will allow then use targeted analysis that would bring down the detection limit for individual components substantially to detect and quantify these components in much smaller samples. This information can be used to compare relative abundances of some of the components during differentiation of these cells into their neural progeny.

Support for this work was provided by the Bio-Organic Biomedical Mass Spectrometry Resource at UCSF(A. L. Burlingame, Director) through the Biomedical Research Technology Program of the NIH National Center for Research Resources, NIH NCRR P41RR001614.

C.16

A Proteomics Approach to Overcoming Bacterial Drug Resistance: The Ribosomal QconCAT

J. Barber¹, Z. Al-majdoub², and S. Gaskell²

¹School of Pharmacy and Pharmaceutical Sciences and ²Manchester Interdisciplinary Biocentre, The University of Manchester, Manchester, United Kingdom

The ribosome is arguably the most important drug target in the bacterial cell, with seven distinct classes of drug inhibiting its action and assembly. These drugs include the macrolides, the safest known anti-bacterials, and the aminoglycosides, the most effective known anti-bacterials. Bacterial disease remains a global health problem not because we lack good drugs but because of the rise of bacterial resistance. The aminoglycosides also suffer from poor therapeutic indices, further limiting their use in the clinic.

We are therefore developing a proteomics approach to enhancing the lifetime and effectiveness of existing anti-bacterial drugs, especially the macrolides and aminoglycosides. We aim to identify targets whose inhibition will lead to synergy with aminoglycosides and macrolides, substantially reducing the likelihood of resistance and enhancing the effectiveness of the drugs. This requires that we can quantify the proteins of the translational machinery and determine the effect of sub-lethal antibiotic doses on the bacterial proteome.

To this end we have designed a flexible QconCAT for the quantification of the E. coli translational machinery by mass spectrometry. A QconCAT is an artificial protein, made up of signature peptides from each of the proteins under study. These peptides are concatenated together and expressed in labelled form using an artificial gene.¹ Our initial QconCAT (the core) contains signature peptides from six central ribosomal proteins L2, L4 and L5 from the 50S subunit and S2, S7 and S8 from the 30S subunit. All these proteins are in place early in the ribosomal assembly process and their presence can be used as markers for ribosomes. The gene encoding the core contains restriction sites into which one or more cassettes encoding other peptides may be inserted. Thus the first cassette encodes signature peptides for the 30S ribosomal proteins.

Most of the ribosomal proteins are small and basic and a digestion protocol involving trypsin alone does not yield sufficient correctly-cleaved peptides for quantification. We therefore employ sequential digestion with endoproteinase Lys-C and trypsin to analyze ribosomal proteins, and the QconCATs are designed for this strategy.

Preliminary results on the effect of the aminoglycoside gentamycin on the bacterial proteome are described.

References

1. Beynon, R. J., Doherty, M. K., Pratt, J. M., and Gaskell, S. J., (2005) Multiplexed absolute quantification in proteomics using artificial QCAT proteins of concatenated signature peptides. Nat. Methods 2, 5879.

C.17

Proteomic Analysis of the Venom from the Mexican Scorpion Centruroides limpidus limpidus

C. Vicente Ferreira Batista¹, O. VillaHernández¹, L. Hernández Orihuela¹, V. Pando², and L. D. Possani¹

¹IBT - Instituto de Biotecnología, UNAM - Universidad Nacional Autónoma de México, Cuernavaca, Morelos, Mexico; ²INSP - National Institute of Public Health, Cuernavaca, Morelos, Mexico

Centruoides limpidus limpidus is an endemic Mexican scorpion distributed in several states of the central-western areas of the country. It is one of the most dangerous arachnids in terms of public health risks in Mexico and it is responsible for deathly accidents, especially among young children. Scorpion venoms contain hundreds of components, among which are enzymes and short chain peptides (toxins) with molecular masses varying from 2.5 to 7 kDa. These toxins are blockers or modifiers of ion-channels present on the membranes of excitable and non-excitable cells (Rodríguez de la Vega, and Possani, Toxicon 43:865875. 2004 and 46: 831844, 2005). From the forty-two amino acid sequences of this scorpion retrieved from NCBI database, only 19 are unique sequences. Differences among these sequences are mainly at the pro-peptide or at the unprocessed C-terminus region of the peptides. Attention was focused to three main questions when studying the soluble part of this venom. (i) Which are the structural differences of the Na+-channel specific toxins from C. limpidus limpidus that makes this scorpion so dangerous to humans, when compared to other species? (ii) Since only 19 distinct peptides are reported thus far, to which family belong the remaining hundred of components known to be present in this venom? Finally, (iii) which is the relative concentration of the Na+-channel specific toxins (NaScTx) expressed in the soluble venom of this scorpion? To answer these questions, the venom was separated by HPLC using an extended gradient system and a complete molecular mass fingerprint was performed on an Orbitrap XL mass spectrometer. More than one hundred HPLC fractions were collected and 219 different molecular masses were determined. The same fractions were digested with trypsin and individually analyzed by CID and HCD simultaneously. Two hundred peptides were unambiguously sequenced and identified as belonging to almost all families of scorpion toxins and enzymes known at present. Some completely unknown components were identified. The relative abundance of the Na-ScTxs in the soluble venom was determined by UV-absorbance. Surprisingly, 70% of the soluble venom is made by the Na-ScTxs, which explains in part its high toxicity. Finally, an interactive structure versus activity diagram was constructed connecting the C. limpidus limpidus components with all other known scorpion components. The diagram shows to which structural and functional families these components belong. It seems that C. limpidus limpidus lethality is probably associated to the high concentration of Na-ScTxs in its venom.

Research supported in part by grants: DGAPA IN227507 to L.D.P and CONACYT 47879Q to C.V.F.B.

Signaling to Transcription Networks in Nerve Injury Response

I. Michaelevski¹, Y. Segal-Ruder¹, O. Shalem², K. F. Medzihradszky³, M. Rozenbaum¹, G. Coppola⁴, D. Geschwind⁴, Y. Pilpel², A. L. Burlingame³, and M. Fainzilber¹

Departments of ¹Biological Chemistry and ²Molecular Genetics, Weizmann Institute of Science, Rehovot, Israel; ³Mass Spectrometry Facility, Department of Pharmaceutical Chemistry, University of California, San Francisco, CA; ⁴Center for Neurobehavioral Genetics, University of California, Los Angeles, CA

Investigations of the molecular mechanisms underlying responses to peripheral nerve injury have highlighted the importance of the retrograde transport system in axons, showing that a protein complex containing importin β1, vimentin and p-ERK 1/2 associates with dynein motors to signal retrogradely from an axonal lesion site to the neuronal cell body. In order to obtain a comprehensive view of the signaling and transcriptional networks involved in retrograde injury signaling, we have now applied proteomics approaches to characterize axonal signaling ensembles, and transcriptomics to characterize the cell body response. These efforts yielded large data-sets, with over 879 proteins and 2465 phosphorylation sites implicated in axonal retrograde injury signaling in rat sciatic nerve, and ~4500 transcripts regulated in the cell body response in dorsal root ganglia. Computational tools were then employed to link these data-sets, thus identifying combined signaling to transcription networks underlying the response to nerve injury. Network redundancies suggest a high level of robustness in the injury response system.

7.2

Regulation of Neuronal Protein Levels at Subcellular Sites Distant from the Cell Body

J. Coleman, S. Alda, D. Vuppalanchi, S. Yoo, D. E. Willis, and J. L. Twiss

Nemours Biomedical Research, A.I. duPont Hospital for Children, Wilmington, DE; Department of Biology, University of Delaware, Newark, DE

Localized synthesis of new proteins in subcellular regions provides a means to rapidly respond to extracellular stimuli and cellular events. In the nervous system, protein synthesis in dendrites plays a role in synaptic plasticity and proteins generated locally in axons are used for growth and injury responses. With geographically separated processes, neurons are a very attractive model system to study protein dynamics in subcellular compartments. We have used several approaches to determine what proteins are generated locally in axonal process and how transport and localized translation of the mRNAs is regulated. mRNAs are transported into axons as ribonucleoprotein complexes. Cues that regulate directionality of axonal growth modify both the populations of mRNAs sent into axons and the localized translation of mRNAs. Response to some stimuli, including axotomy, also requires localized proteolysis. The majority of studies of these mechanisms have been limited to analyses of functional responses or single protein species. Analyses of mRNAs have shown that their transport into axonal processes is regulated with a surprising degree of specificity. It is highly likely that mRNA translation will show similar complexity with specific changes in protein production and protein degradation being linked to extracellular events. Understanding the specificity of these events will require unbiased approaches to quantify proteins dynamics in axons and dendrites. Knowledge of the molecular composition of the transported ribonucleoprotein complexes will also be needed to dissect regulatory mechanisms underlying specificity of mRNA transport into subcellular regions. However, the limited quantities of materials derived from cultured neurons where these cellular processes can be isolated to purity have thus far restricted proteomics approaches.

7.3

Organelle Proteomics: Linking Axonal Transport to Nerve Regeneration

N. Abe¹, A. Almenar-Queralt², C. Lillo², Z. Shen², J. Lozach², S. P. Briggs², D. S. Williams^2,3, L. S. B. Goldstein², and V. Cavalli¹

¹Washington University in St Louis, MO; ²University of California, San Diego, CA; ³University of California, Los Angeles, CA

The extreme polarized morphology of neurons poses a challenging problem for intracellular trafficking pathways. The distant synaptic terminals must communicate via axonal transport with the cell soma for neuronal survival, function and repair. Multiple classes of organelles transported along axons may establish and maintain the polarized morphology of neurons, as well as control signaling and neuronal responses to extracellular cues such as neurotrophic or stress factors. We reported previously that the motorbinding protein Sunday Driver (syd), also known as JIP3 or JSAP1, links vesicular axonal transport to injury signaling. To better understand syd function in axonal transport and in the response of neurons to injury, we developed a purification strategy based on antisyd antibodies conjugated to magnetic beads to identify sydassociated axonal vesicles. Electron microscopy analyses revealed two classes of sydassociated vesicles of distinct morphology. To identify the molecular anatomy of syd vesicles, we determined their protein composition by mass spectrometry. Gene ontology analyses of each vesicle protein content revealed their unique identity and indicated that one class of syd vesicles belong to the endocytic pathway, while another may belong to an anterogradely transported vesicle pool. To validate these findings, we examined the transport and localization of components of syd vesicles within axons of mouse sciatic nerve. Together, our results lead us to propose that endocytic syd vesicles function in part to carry injury signals back to the cell body, while anterograde syd vesicles may play a role in axonal outgrowth and guidance.

7.4

Quantitative Phosphoproteomics Identifies Sites in K-Cl Co-Transporters that Regulate Cell Volume and Neuronal Excitation

J. Rinehart^1,2, Y. D. Maksimova⁴, J. E. Tanis⁵, K. L. Stone^2,3, C. A. Hodson¹, J. Zhang¹, M. Risinger⁶, W. Pan⁷, D. Wu⁷, C. M. Colangelo^2,3, B. Forbush⁵, C. H. Joiner⁶, E. E. Gulcicek^2,3, P. G. Gallagher⁴, and R. P. Lifton^1,2

Departments of ¹Genetics, Howard Hughes Medical Institute, ⁴Pediatrics, ⁵Cellular and Molecular Physiology, and ⁷Pharmacology, Yale University School of Medicine, New Haven, CT; ²Yale/NHLBI Proteomics Center, Yale University, New Haven, CT; ³Keck Biotechnology Resource Laboratory, New Haven, CT; ⁶Cincinnati Comprehensive Sickle Cell Center, Division of Hematology/Oncology, University of Cincinnati College of Medicine & Department of Pediatrics, Cincinnati Children's Hospital Medical Center, Cincinnati, OH

Modulation of intracellular chloride concentration ([Cl^–]_i) plays a fundamental role in cell volume regulation, response to osmotic stress, and neuronal excitation. Cl^– exit via K-Cl co-transporters (KCCs) is a major determinant of [Cl^–]_i, however the mechanisms governing their activities are poorly understood. Using a SILAC approach, we identified two highly conserved phosphorylation sites in human KCC3 that are de-phosphorylated in response to cell swelling. Alanine substitutions at these sites result in constitutively active cotransport. Quantitative studies utilizing multiple reaction monitoring (MRM) confirm that these same sites are modulated during neuronal maturation in vivo. Parallel MRM studies also showed that these sites are highly phosphorylated in plasma membrane KCC3 in isotonic conditions, suggesting that de-phosphorylation increases KCC3's intrinsic transport activity. Studies with phospho antibodies demonstrate how human red blood cells control cell volume via a similar mechanism. Inhibition of the kinase WNK1 via RNAi reduces phosphorylation at these sites. Phosphopeptide mapping via mass spectrometry and antibody based validation studies showed that homologous sites are phosphorylated in all human KCCs. These findings provide new insights into regulation of cell volume and neuronal function and highlight the power of emerging proteomic techniques.

A Proteomics Approach to Overcoming Bacterial Drug Resistance: The Ribosomal QconCAT

J. Barber¹, Z. Al-majdoub², and S. Gaskell²

¹School of Pharmacy and Pharmaceutical Sciences and ²Manchester Interdisciplinary Biocentre, The University of Manchester, Manchester, United Kingdom

The ribosome is arguably the most important drug target in the bacterial cell, with seven distinct classes of drug inhibiting its action and assembly. These drugs include the macrolides, the safest known anti-bacterials, and the aminoglycosides, the most effective known anti-bacterials. Bacterial disease remains a global health problem not because we lack good drugs but because of the rise of bacterial resistance. The aminoglycosides also suffer from poor therapeutic indices, further limiting their use in the clinic.

We are therefore developing a proteomics approach to enhancing the lifetime and effectiveness of existing anti-bacterial drugs, especially the macrolides and aminoglycosides. We aim to identify targets whose inhibition will lead to synergy with aminoglycosides and macrolides, substantially reducing the likelihood of resistance and enhancing the effectiveness of the drugs. This requires that we can quantify the proteins of the translational machinery and determine the effect of sub-lethal antibiotic doses on the bacterial proteome.

To this end we have designed a flexible QconCAT for the quantification of the E. coli translational machinery by mass spectrometry. A QconCAT is an artificial protein, made up of signature peptides from each of the proteins under study. These peptides are concatenated together and expressed in labelled form using an artificial gene.¹ Our initial QconCAT (the core) contains signature peptides from six central ribosomal proteins L2, L4 and L5 from the 50S subunit and S2, S7 and S8 from the 30S subunit. All these proteins are in place early in the ribosomal assembly process and their presence can be used as markers for ribosomes. The gene encoding the core contains restriction sites into which one or more cassettes encoding other peptides may be inserted. Thus the first cassette encodes signature peptides for the 30S ribosomal proteins.

Most of the ribosomal proteins are small and basic and a digestion protocol involving trypsin alone does not yield sufficient correctly-cleaved peptides for quantification. We therefore employ sequential digestion with endoproteinase Lys-C and trypsin to analyze ribosomal proteins, and the QconCATs are designed for this strategy.

Preliminary results on the effect of the aminoglycoside gentamycin on the bacterial proteome are described.

References

1. Beynon, R. J., Doherty, M. K., Pratt, J. M., and Gaskell, S. J. (2005) Multiplexed absolute quantification in proteomics using artificial QCAT proteins of concatenated signature peptides. Nat. Methods. 2, 5879.

8.2

Quantitative Proteomics Analysis of C/EBP Transcription Factor Complexes in Leukemia

J. A. Marto

Department of Biological Chemistry and Molecular Pharmacology, Havard Medical School, Department of Cancer Biology and Blais Proteomics Center, Dana-Farber Cancer Institute, Boston, MA

Acute myeloid leukemia (AML) remains a highly lethal malignancy with limited therapeutic options. Full transformation in AML requires coupling of aberrant molecular events associated with multiple cellular processes. For example, recent work has linked oncogenic FLT-3 kinase, the most common molecular abnormality in acute myeloid leukemia (AML), with functional disruption of otherwise normally expressed C/EBP, a key transcription factor in hematopoiesis. Moreover, there is growing evidence that combined FLT-3 and C/EBP karyotypes provide useful prognostic indicators for AML patients. Despite these strong molecular and clinical links, we know surprisingly little about the mechanisms that underlie normal C/EBP gene activation, and how FLT-3 signaling may interfere with C/EBP function in the context of AML. We are testing the hypothesis that FLT-3 mediated phosphorylation of C/EBP governs the assembly of transcriptionally active or repressed protein complexes. Towards this end we have established inducible expression of dual-affinity tagged C/EBP in myeloid cells that also exhibit constitutive FLT-3 activity. We observe that inhibition of FLT-3 signaling modulates C/EBP phosphorylation in a dose-dependent manner. Next, quantitative proteomics methodology, including iTRAQ labeling and nanoflow LC coupled with online multidimensional RP/RP fractionation was used to monitor remodeling of C/EBP protein complexes as a function of FLT-3 mediated phosphorylation. Our data significantly expand upon the known repertoire of C/EBP interactors, including more than 100 proteins involved in chromatin organization, transcriptional modulation, and cell cycle regulation. Furthermore, our quantitative proteomics data demonstrate that (i.) C/EBP interacts with proteins genetically linked to leukemia; and (ii.) many of these interact with C/EBP in a phosphorylation-dependent manner. Genetic depletion of newly-identified, leukemia-associated protein interactors reduced the ability of C/EBP to drive expression of granulocytic target genes. In addition we have confirmed co-localization of C/EBP and proteins identified in our proteomics analysis at the promoters of early, myeloid-specific genes. Collectively our data demonstrate direct physical and functional links between the tumor suppressor C/EBP and other leukemia-associated, putative oncogenes. Our ability to quantitatively monitor multiple leukemia-related gene products in the context of C/EBP protein complexes provides valuable insight into the mechanisms by which oncogenic kinase activity disrupts transcription and leads to leukemogenesis.

8.3

Rapid, Near Proteome-wide, Quantitative Analysis of Aneuploid Budding Yeast

N. Dephoure¹, E. M. Torres², J. Villén¹, A. Amon², and S. P. Gygi¹

¹Harvard Medical School, Boston, MA; ²Howard Hughes Medical Institute, MIT, Cambridge, MA

Mass spectrometry based proteomics holds the promise to extend the global analysis of gene expression afforded by DNA microarrays to the measurement of cellular proteins. Recent advances in methodology and instrumentation have brought the elusive goal of facile global protein measurement closer. However, the extensive analysis time required to achieve high protein coverage renders routine experimentation impractical for most researchers. Using stable isotope labeling with amino acids in cell culture (SILAC) and strong cation exchange chromatography we are able to routinely and reproducibly quantify ~3,000 budding yeast proteins in less than two days of instrument analysis time. We have used this method to characterize protein level changes in a collection of aneuploid yeast strains, harboring two copies of a single chromosome (disomic). In all strains examined to date, proteins coded on the disomic chromosome were enriched ~2-fold, while levels of other proteins remained constant. Such small changes can be extremely challenging to detect by traditional methods such as immunoblotting, but were easily discernible by our method. We also observed a subset of disomic gene products whose levels were unchanged and are trying to understand the mechanisms that allow these proteins to escape the increase in gene dosage.

8.4

Advancing Epigenetics Research by Proteomics: Technologies, Applications and Perspectives

O. N. Jensen

Centre for Epigenetics and Department of Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark

"An epigenetic trait is a stably heritable phenotype resulting from changes in a chromosome without alterations in the DNA sequence" (Berger et al. 2009).

Histone proteins plays a major role in maintaining chromatin structure, regulating gene activity and DNA integrity. Post-translational modifications of histone proteins that constitute the nucleosome modulate the interactions of these proteins with DNA, transcription factors and chromatin modifying enzymes. There is emerging evidence that combinations of post-translational modifications are key regulators of cellular development and differentiation programs, including epigenetic mechanisms, and that errors in these systems lead to a variety of diseases.

Mass spectrometry plays a prominent role in mapping and quantifying histone modifications, including multi-site, cooperative modifications of histone tails. In this lecture I will describe some of our analytical strategies that are aimed at detailed characterization of histone proteins in the context of pathogenic microorganisms (Salcedo-Amaya et al., 2009; Trelle et al., 2009), drug development (Beck et al., 2006), stem cell research and cancer (Jung et al., submitted). I will also discuss some of the current bottlenecks in functional proteomics and emerging areas of research where proteomics is likely to play a major role.

References

1. Beck et al. (2006) Mol. Cell. Proteomics 5 (7), 1314–25.

2. Berger et al. (2009) Genes Dev. 23, 781–783.

3. Salcedo-Amaya et al. (2009) Proc. Natl. Acad. Sci. U. S. A. 106(24), 9655–60.

4. Trelle et al. (2009) J. Proteome Res. 8(7), 3439–3450.