Learning Center
Plans & pricing Sign in
Sign Out
Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

Analysis of Nuclear Pore Complex Phosphorylation Sites using a


  • pg 1
									     Analysis of Nuclear Pore Complex Phosphorylation Sites using a MALDI-LTQ
                             Orbitrap Mass Spectrometer

                                    1            1               1                    2
                    Justin Blethrow , Rosa Viner , Vlad Zabrouskov , and Joseph Glavy
                   Thermo Fisher Scientific, 355 River Oaks Parkway, San Jose, CA 95134
              Stevens Institute of Technology, Castle Point on Hudson, Hoboken NJ 07030-5991

Introduction The nuclear pore complex (NPC)             were found to be specific or substantially enriched
contains ~30 proteins and forms the mainl               in mitotic proteins. Additional phosphopeptides
passageway for nucleo-cytoplasmic                       were detected in proteins isolated using mab414,
macromolecular traffic. In higher eukaryotes, the       including several FG-repeat proteins and Nup358.
NPC undergoes mitotic disassembly into                  The ongoing work is focused on expanding our
subcomplexes, thought to be enacted largely via         analysis for fuller coverage of these and other
phosphorylation. We used affinity isolation and         NPC components.
high mass accuracy LC MALDI MS and MS/MS to
identify phosphorylation sites in the NPC and to        Innovative aspects
quantify their relative mitotic and interphase          • Analysis of MALDI-produced ions with an
abundances by stable isotope labeling.                     orbital trap / ion trap hybrid mass
Methods HeLa cells grown in the presence of             • Antibody-based dissection of the Nuclear Pore
labeled arginine or lysine were arrested in G1 and         Complex.
lysed. This was added to unlabeled lysate from
synchronous mitotic cells previously released from      References
G1 arrest. Immunoprecipitations were performed          1: Glavy JS, et al.; Proc. Natl. Acad. Sci. – 104,
using mab414 to isolate approximately ten FG-           3811-3816 (2007); Cell-cycle-dependent
repeat and associated proteins, and an antibody to      phosphorylation of the nuclear pore Nup107–160
Nup107, isolating the Nup107-160 subcomplex.            subcomplex.
Proteins were enzymatically digested, and
peptides were separated by HPLC with automated
fraction spotting. Alternatively, proteins were
separated by PAGE, excised, and processed for
MALDI. Spots were analyzed by MALDI-LTQ
Orbitrap, using survey scans of 60k nominal
resolution. Phosphopeptides were detected by
Neutral Loss ion mapping in the LTQ and selected
for MS3 analysis or HCD fragmentation in the

Results Immunoprecipitation with anti-Nup107
purified the Nup107-160 subcomplex. MS analysis
identified nine proteins, similar to previous
studies(1). Immunoprecipitation with mab414
purified seven NPC components (Nup358,
214,153, 88, 62, 58 and 54) and two karyopherin
proteins (nucleo-cytoplasmic transporters). Full MS
analysis using the Orbitrap consistently provided
high mass accuracies (3 ppm RMS), which
significantly increased the confidence of peptide
identifications. The LC-MALDI analysis format
allowed for fragmentation analysis of every mass
value at every LC time point. This provided
sensitive detection of phosphopeptides by
monitoring for neutral loss of phosphate in MS2
(ion mapping). Subsequent manual MS analysis
to localize the phosphorylation site was performed
if the precursor was not previously selected
automatically in the standard data dependent
experiment. Numerous phosphorylation sites were
observed in the Nup107-160 subcomplex.
Consistent with previous results, several of these
    Analyses of Proteasome Complexes and Proteome Variations in Acute Myeloid
                                Leukemia Cells
               1                            1                   1              2            2              1
 M. Matondo , M.P. Bousquet-Dubouch , S. Uttenweiler-Joseph , B. Payrastre , S. Manenti , B. Monsarrat ,
                                           and O. Burlet-Schiltz
   Institut de Pharmacologie et de Biologie Structurale, CNRS, Université de Toulouse, Toulouse, France
               Centre de Physiopathologie Toulouse-Purpan, INSERM U563, Toulouse, France

Introduction                                               arginine and lysine labelling allowed 90% of
Proteasome is an essential component of the                identified proteins to be quantified. Bioinformatic
ubiquitin-proteasome pathway and plays a critical          data analysis programs were then used to classify
role in protein degradation regulation. Dysfunction        proteins and to visualize pathways in both cell
of this complex machinery can lead to various              lines. Overall, the results obtained by combining
pathologies including cancer (1). Proteasome               different    proteomic     approaches    for   the
inhibitors represent promising new antitumor drugs         characterization of proteasome complexes as well
and one of them is currently used for the treatment        as the proteome content of these cell lines should
of multiple myeloma. Our aim is to compare                 help understanding their differential anticancer
proteasome complexes composition and activity              drug sensitivity.
and to analyze proteome variations in different
leukemic cells to understand better the effect at          Innovative aspects
the molecular level of these drugs.                        • Analysis of proteasome complexes.
                                                           • Quantitative proteome analysis of different
Methods                                                       acute myeloid leukemia cells.
Proteasome complexes were purified by                      • Structure/function relationships of proteasome
immunoaffinity chromatography (2). Their subunits             in acute myeloid leukemia cells.
were separated by gel electrophoresis and
identified by mass spectrometry. Proteasome                References (maximum 3 references)
activity was measured by following the hydrolysis          (1) A. Mani et al., The ubiquitin-proteasome
of fluorescent substrate peptides. Differential            pathway and its role in cancer; J Clin. Oncol. 2005,
quantitative proteome analyses were performed              23, 4776-4789.
using stable isotope labeling with amino acids in          (2) M.P. Bousquet-Dubouch et al, Purification
cell culture (SILAC) with labeled lysine ([ C6]Lys)        and proteomic analysis of 20S proteasomes from
                 13  15
and arginine ([ C6, N4]Arg) amino acids. Mass              human cells; Methods Mol. Biol. 2008 (in press).
spectrometric analyses were performed by                   (3) D. Bouyssié et al, Mascot file parsing and
nanoLC-MS/MS using either a Q-STAR or an LTQ-              quantification (MFPaQ), a new software to parse,
Orbitrap mass spectrometer. Database searches              validate, and quantify proteomics data generated
were performed using the Mascot algorithm and              by ICAT and SILAC mass spectrometric analyses:
quantitative data were analyzed using MFPaQ, a             application to the proteomics study of membrane
in-house developed software (3).                           proteins from primary human endothelial cells;
                                                           Mol. Cell. Proteomics 2007, 6, 1621-1637.
Two human acute myeloid leukemia (AML) cell
lines showing differential sensitivity to proteasome
inhibitors were studied: KG1a cells (AML M0) and
U937 cells (AML M5). First, the proteasome status
in each cell line was compared. We showed that
the proteasome content and its chymotrypsin-like
activity greatly differ in the two cell lines. After 20S
proteasome purification, however, the subunit
pattern observed by 2D gel electrophoresis
revealed only minor differences in the subunit
composition of the complex. Both standard 20S
proteasome and immunoproteasome were present
in both cell lines. Thus, regulatory complexes
associated to the 20S catalytic core were also
studied. Second, the proteomes of the two cell
lines were compared using the SILAC quantitative
approach. NanoLC-MS/MS analyses of nuclear,
cytosolic and membrane protein fractions using an
LTQ-orbitrap mass spectrometer led to the
identification of more than 1000 proteins in each
fraction. Protein quantification based on both
                    1                   1                 1                 1                          1
 Benoit Coulombe , Philippe Cloutier , Annie Bouchard , Célia Jeronimo , Andrée-Anne Lacombe , Mathieu
                                     2                     2                3            1
                     Lavallée-Adam , Mathieu Blanchette , Jack Greenblatt , Diane Forget
 1                                                                                         2
   Institut de recherches cliniques de Montréal, Université de Montréal, Montréal, Canada, McGill Centre for
    Bioinformatics, McGill University, Montréal, Canada, Banting and Best Department of Medical Research,
                                      University of Toronto, Toronto, Canada

Introduction                                                  performance mass spectrometry. Many identified
    Proteins are the central functional components            interaction partners were targeted in reciprocal
of human cells. They are involved in almost all               tagging experiments in order to confirm many
cellular processes, and protein aberrations have              interactions and to enrich the dataset. High-
been shown to have a causative role in many                   confidence      interactions  were      selected
diseases. Most importantly, proteins must interact            computationally using an algorithm that we
with other molecular components of the cell,                  developed and trained using machine learning to
including other proteins, DNA sequences, RNA                  minimize the rate of both false-positives and
molecules and various metabolites, to exert their             false-negatives.
functions. Mapping protein interaction networks in
cells and tissues is expected to produce                      Results
fingerprints of the physiological states of these                The data produced with more than 100 affinity
cells and tissues; similarly, mapping changes in              tagged proteins was used to (i) build a high-
protein interaction networks occurring during                 definition map of interactions that connect
disease progression will generate signatures for              components of the transcription and RNA
specific pathological states. Because protein                 processing machineries in human cells; (ii) show
interaction maps represent multi-variable, complex            that transcription and RNA processing factors
descriptions of physiological/pathological states,            from the soluble cellular fraction are associated
they also provide the descriptions needed to more             with proteins that specifically regulate the
realistically and reliably address issues such as             formation (e.g. assembly, localization and/or
the causes, the diagnoses and, eventually, the                stability) of protein complexes; and, (iii) assign a
cures of human diseases. Mapping protein                      putative function to a number of previously-
interaction networks in health and disease is a               uncharacterized proteins based on ‘‘guilt by
tremendous scientific and technological challenge             association’’.     A    number      of    previously-
that will require huge efforts from many groups of            uncharacterized proteins that we further
scientists world-wide.                                        characterized functionally and biochemically
    Recently, we have developed a technology                  define a novel class of regulatory factors that
pipeline for the systematic characterization of               target RNA polymerase II and other transcription
protein interaction networks from human cells (1).            factors prior and/or after the transcription reaction
Our objective is to leverage this work into a large-          on chromatin DNA.
scale initiative aimed at building a repertoire of
comprehensive maps of protein interaction                     Innovative aspects
networks in health and disease. This repertoire of            • High-precision mapping of human protein
human protein interaction maps, that we propose                  interaction networks
to name the Human Proteotheque, will be built via             • Discovery of a novel class of cellular
a concerted, international initiative that will involve          regulatory factors that target the transcription
a multi-site discovery platform aimed at defining                machinery
protein-protein, protein-DNA, protein-RNA and                 • Creation of the Human Proteotheque, an
protein-metabolite interactions, and integrating the             expanding repertoire of comprehensive maps
data into comprehensive interaction maps through                 of human protein interaction networks in health
bioinformatics. This effort, termed the Human                    and disease
Proteotheque Initiative (HuPI) is at the heart of an
emerging discipline, Integrative Systems Biology              References
(ISB), which is aimed at developing tools and                    (1) C. Jeronimo et al, Systematic analysis of
concepts to generate complex descriptions of                         the protein interaction network for the
biological systems considered globally.                              human transcription machinery reveals the
                                                                     identity of the 7SK capping enzyme; Mol.
Methods                                                              Cell 27, 262-274, 2007
   Here we used our newly developed technology
pipeline to perform a survey of soluble human
protein complexes containing components of the
transcription and RNA processing machineries
using protein affinity purification coupled to high-
    Characterization of a novel subunit of the Drosophila melanogaster chromatin
                             remodeling complex PBAP

 Gillian E. Chalkley 1, Yuri M. Moshkin 1, Karin Langenberg 1, Karel Bezstarosti 2, Andras Blastyak 4, Henrik
                        Gyurkovics 4, Jeroen A. A. Demmers 2, and C. Peter Verrijzer 1

    1. Department of Biochemistry, Center for Biomedical Genetics, and 2. Proteomics Center, Erasmus
 University Medical Center, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands. 3. Hungarian Academy of
            Sciences, Biological Research Center, Institute of Genetics, H-6701 Szeged, Hungary

Introduction                                              consistent with the notion that SAYP is the key
SWI/SNF ATP-dependent chromatin remodeling                architectural subunit required for assembly of the
complexes (remodelers) perform critical functions         tri-partite PBAP module and its linkage to the
in eukaryotic gene expression control. BAP and            SWI/SNF core.
PBAP are the fly representatives of the two               In a genome-wide distribution analysis, SAYP co-
evolutionarily conserved major subclasses of              localized with both PB and BAP170 on larval
SWI/SNF remodelers. Both complexes share 7                salivary gland polytene chromosomes on the
core subunits, including the Brahma (BRM)                 majority of their chromosomal binding sites. These
ATPase, but differ in a few signature subunits:           sites also bind BRM and represent the interband
Polybromo (PB) and BAP170 specify PBAP,                   regions of open, less condensed chromatin. These
whereas OSA defines BAP. Here, we show using              results suggest that PB, SAYP and BAP170
a proteomics approach that the transcriptional co-        appear to form a stably associated PBAP-defining
activator and PHD finger protein SAYP, is a novel         module that is integrally recruited to chromatin.
PBAP subunit. Furthermore, using biochemical              Gene expression profiles from specific RNAi- and
and genetic analyses, we give detailed insight into       mock treated cells were compared, using an
both the architecture and the function of the PBAP        unbiased statistical analysis of the complete data
holoenzyme.                                               set.     Correlation  analysis     of    the   SAYP
                                                          transcriptome and the core, PBAP- and BAP-
Methods                                                   selective subunits showed a very strong
Polyclonal antibodies against BAP and PBAP                relationship between SAYP and BAP170 and PB.
subunits were raised and used for co-                     Principal component analysis revealed that the
immunoprecipitation experiments from Drosophila           SAYP profile is highly correlated with those of PB
embryo nuclear extracts. Eluted proteins were             and BAP170.
resolved by SDS-PAGE and visualised by
coomassie staining. Polypeptides were identified
by mass spectrometry on an LTQ-Orbitrap hybrid
mass spectrometer (Thermo Fischer). Data
analysis was performed using the Mascot search
algorithm searching against FlyBase database.
Further data analysis was performed using in-
house developed software. For RNAi knockdown
and     genome      wide    expression     analysis
experiments, Drosophila S2 cells were cultured in
Schneider’s media and treated with double-
stranded RNA for 4 days. RNA samples from three
completely     independent      biological   SAYP
knockdown experiments were prepared and
hybridized with Affymetrix microarrays.

Mass spectrometric analysis revealed the
presence of SAYP in the BRM- and PB-, but not in
the OSA immunopurification. Stringent co-
immunoprecipitation experiments showed that the           Innovative aspects
core subunits, BAP170 and PB co-purify in an anti-           • Combination of mass spectrometry with
SAYP immunoprecipitation, whereas OSA is                         biochemical and genetic analyses as a
completely absent. Also, the association to the                  novel approach to the investigation of
core by either OSA or the PBAP module was                        transcriptional pathways.
found to be strictly mutually exclusive as no hybrid
complexes were detected. Also, there seems to be          References
a hierarchy in PBAP architecture: SAYP is critical        (1) Chalkley et al, Mol Cell Biol, Feb 25 [Epub
for the integration of assembly. The results are          ahead of print]

               1            2                       2
Mathias Dreger , Tao Deng , and George Brownlee
 Department of Physiology, Anatomy, and Genetics, University of Oxford, Parks Road,
Oxford, OX1 3PT, United Kingdom
  Sir William Dunn School of Pathology, University of Oxford, UK South Parks Road
Oxford, OX1 3RE, United Kingdom

The influenza A virus polymerase complex, a heterotrimeric assembly of the subunits PB1,
PB2, and PA, is critical for virus replication. PB1 requires PA and PB2 subunits for functional
activity. To identify functionally important sites in PB1 sensitive to subunit interactions, we
devised a strategy of SILAC labelling and biotinylation of solvent-accessible lysine residues to
compare the biotinylation pattern of PB1 in its monomeric form or bound by the subunit PA.
We identified a novel site crucial for PB1 function.

TAP-tagged monomeric PB1 or PB1/PA dimers were purified from transiently transfected
                         12               13
HEK293T cells grown in C6-lysine- vs. C6-lysine-containing culture medium. Recovered
proteins were biotinylated in the native state by an amine-specific reagent. The protein
amount was estimated and the samples were combined 1:1, and separated by SDS-PAGE,
followed by trypsin digestion. Biotinylated peptides were enriched by avidin affinity
chromatography, and subjected to nanoflow LC. Fractions were spotted onto a target plate
and analysed by Maldi-MS. SILAC pairs of biotinylated peptides were quantified. Lysine
residues that displayed altered biotinylation between PB1 and PB1/PA were chosen for
mutagenesis to assess their functional relevance for complex formation and polymerase

The levels of PB1 in the PB1 monomer and in the PB1/PA dimer were similar. PA and PB1
were stoichiometrically present in the dimer sample. For mass spectrometric analysis, we first
acquired data in automated nanoflow LC-Maldi MS/MS runs. In this workflow, a portion of the
sample is still accessible on the target plate after the automated run. This allows to assign
further candidate masses and to subject those peptides to a manually operated MS/MS
analysis. The mass spectrometric analysis returned overall 19 biotinylated lysine residues of
PB1, present in more than 30 different SILAC pairs. Six lysine residues displayed altered
accessibility to biotinylation in the PB1 monomer vs. the PB1/PA dimer (changing 2- to 4-fold).
Mutation analysis revealed that one residue that was less accessible in the dimer as
compared to the monomer was found to be a novel site crucial for polymerase activity, but did
not affect dimer formation. Another site displayed increased accessibility in the PB1/PA dimer,
indicating a conformational change in PB1 upon PA binding. Our strategy is generally
applicable to identify protein areas that are sensitive to protein-protein interactions.

Innovative aspects:
   • a generally applicable strategy to semi-quantitatively assess changes in solvent
       exposure of protein surfaces was devised
   • the strategy may yield data on protein-protein interactions as well as on changes of
       protein conformation, and may identify functionally important sites of a protein
   • a novel site critical for PB1 polymerase activity was identified



fig.1 the influenza A polymerase complex

                 x10 4
Intens. [a.u.]







                         1280          1290               1300              1310              1320
fig. 2: SILAC pair of a biotinylated peptide
                                Proteomics of Protein Complexes:
                            An Approach to Analyze Native Proteomes
                            1              1            2,3                    2                 1
              Oliver Drews , Yueju Wang , Emily Chen , John R Yates 3rd , and Peipei Ping
   Department of Physiology, University of California Los Angeles, 675 CE Young Dr S, Los Angeles, CA
  90095, USA; Department of Cell Biology, The Scripps Research Institute, 10550 North Torrey Pines Rd,
 SR11, La Jolla, CA 92037, USA; Department of Pharmacological Sciences, Stony Brook University, BST 8-
                                    125, Stony Brook, NY 11974, USA

Introduction                                                  assemblies and interactions can be retrieved,
Proteins in cells translate signals via interactions,         providing insights into physiology and elevating
convert energy and metabolites in complexes, and              proteomics to deliver a more comprehensive
form structural networks in filaments, pores and              analysis of proteomes.
junctions. Proteins in proteomics though are
predominately analyzed at large scale after                   Innovative aspects
denaturation and hence viewed as singular                     • Native in-solution isoelectric focusing at high
entities. In consideration of the fact that proteins             resolution and the level of proteomes and
interact in biological networks, we developed an                 subproteomes
approach for large scale protein analysis while               • Enables native two-dimensional maps of
preserving proteins in their native state and most               proteomes
importantly maintaining the proteins in complexes.            • Compatible with functional assays and online
                                                                 mass spectrometry
Challenging in analyzing a multitude of native                References
protein complexes is the separation at high                      (1) O. Drews et al., Mammalian proteasome
resolution without perturbation of the fragile                       subpopulations with distinct molecular
interactions between proteins. For our approach,                     compositions and proteolytic activities; Mol
protocols for native isoelectric focusing of protein                 Cell Proteomics. 2007 Nov;6(11):2021-31.
complexes by free flow electrophoresis were
developed to achieve this goal. Furthermore, task
oriented protocols were optimized and tested for
compatibility with prevalent technologies used in
proteomics. Thus, protein complexes were
subsequently subjected to functional assays,
further native or denaturing gel electrophoresis in a
second dimension, and/or identified by online
mass spectrometry.

The approach was applied to native proteomes
containing a widely different number of protein
complexes to demonstrate its effectiveness at
various scales. Separation of crude extracts from
S. cerevisiae consisting of a large variety of
complexes was achieved at a resolution of 0.1 pH
units between pH 3-10. Interestingly, the
complexes focused between pH 4.5-7.5, sharing
only in part the band-with of pIs of denatured yeast
proteins. Subsequent native PAGE provided a 2-D
map of yeast complexes, showing protein bands
from 50-1000 kDa. Identification by MS confirmed
the separation of known and hitherto unknown
homomeric as well as heteromeric complexes. For
example, the homohexamers His1 and Gdh1
separated at 200 and 300 kDa, and focused at pH
5.4 and 5.2 in close proximity to their theoretical
pIs 5.87 and 5.56. At the other end of the dynamic
range, proteasome complexes, purified to
apparent homogeneity, were separated in
subpopulations at a resolution of 0.04 pH units (1).
With the approach, novel information about protein
   Topology and dynamics of protein interaction networks that control cell growth

 Oliver Rinner, Timo Glatter, Alexander Wepf, Katja Köhler, Irena Jevtov, Hugo Stocker, Ernst Hafen, Ruedi
                                       Aebersold and Matthias Gstaiger
                   Institute of Molecular Systems Biology, ETH Zurich, Zurich, Switzerland

Introduction                                                     interaction networks; Nat. Biotechnol.
The control of cellular growth under changing                    2007
environmental conditions (oxygen saturation state,
presence and concentration of growth hormones
and nutrients etc) is an essential physiological
process with important clinical implications. To
date, the response of cells to changes in their
environment has been primarily studied using the
methods of cell biology, biochemistry or genetics in
which complex cellular responses are dissected
into linear sequences of signaling events. Here we
present a systems oriented approach to study a
protein interaction network for the control of cell
growth control.                                                                        QuickTime™ and a

Methods                                                                            TIFF (LZW) decompressor
                                                                                are needed to see this picture.

Building on the results of genetic screens in
Drosophila melanogaster for genes affecting
cellular growth, and by applying advanced
quantitative mass spectrometry techniques we
characterize protein interaction networks in
response to the growth hormone insulin in
drosophila KC167 cells. Protein complexes are
crosslinked with DSP prior to affinity purification
using a single step immuno-affinity purification
approach. Interacting proteins are identified by         Figure 1. Evolutionary conservation of complexes
                                                         containing the target of rapamycin (TOR) kinase
direct LC-MS/MS and relative changes in protein
                                                         between D.melanogaster and humans.
interactions are quantified using the recently
developed “MasterMap” concept (1).
We have generated a collection KC167 cell lines
for the inducible expression of epitope tagged bait
proteins. Following affinity purification and mass
spectrometry (AP-MS) we identified a number of
specific protein interactions that are highly
conserved between humans and D. melanogaster,
including complexes containing the PIK-related
kinase dTOR (Figure1). Relative quantification
from aligned MS1 spectra uncovered a set of
insulin    dependent    interactions     within the
drosophila insulin receptor/target of rapamycin
(InR/TOR) signaling pathway. These hormone
dependent interactions may provide important
cues for understanding the molecular mechanisms
underlying insulin dependent growth control.

Innovative aspects
• Combined genetics and mass spectrometry
• Evolutionary conservation of protein
• Hormone dependent protein-protein

   (1) O.Rinner et al., An integrated mass
       spectrometric and computational
       framework for the analysis of protein
      Pull-downs of endogenously expressed, tagged proteins for interactions and
                 1                        1                1                     1              1               1
Nina C. Hubner , Johannes Graumann , Michiel Vermeulen , Leonie Waanders , Jürgen Cox , Yong Zhang ,
                                    2            2                 2                1
                      Alexander Bird , Ina Poser , Anthony Hyman , Matthias Mann
  Department of Proteomics and Signal Transduction, Max-Planck Institute for Biochemistry, Am Klopferspitz
                                     18, 82152 Martinsried, Germany
 Department for Microtubules and Cell Division, Max-Planck Institute of Molecular Cell Biology and Genetics,
                             Pfotenhauer Str. 108, 01307 Dresden, Germany

Introduction                                              way from a minimal amount of protein. Briefly, we
Multi-protein complexes and modifications are             obtain very pure bait protein by using more
fundamental in most cellular processes. It is of          stringent washing conditions and perform in-depth
great interest to the scientific community to provide     sequencing and assignment of all its peptides.
a network of interactions and modifications. While
there exists already a large-scale interaction            Innovative aspects
dataset for yeast, it has so far not been possible to     • Interaction studies with endogenous
map mammalian interactions and modifications to              expressed proteins in mammalian systems
a similar depth.                                          • High-confidence determination of interaction
Our goal is to establish a standardized pipeline for         partners and modifications using SILAC
mapping specific protein interactions and                 • Unbiased identification of modifications
modifications by SILAC-based quantitative mass
spectrometry (Mann, 2006).                                References
                                                          Mann, M., “Functional and quantitative proteomics
Methods                                                    using SILAC”, Nature reviews 7, 952-958 (2006)
Proteins are tagged by homologous recombination           Muyrers, J.P et al., “Techniques: Recombinogenic
under their endogenous promoter with GFP in                engineering - new options for cloning and
HeLa or embryonic stem cells (Muyrers, 2001). In           manipulating DNA”, Trends in biochemical
contrast to published large-scale interaction              sciences 26, 325-331 (2001).
studies in mammalian systems we do not                    Poser, I. et al., “BAC TransgeneOmics: A high-
overexpress the bait protein and thereby keep the          throughput method for exploration of protein
system in a close to natural state. We combine a           function in mammals”, submitted
highly specific antibody to the GFP-tag with very
small magnetic beads and a novel HPLC
technique, which allows analysis of very low
amounts of material and repeated and targeted
measurement. Using this system with the LTQ-
Orbitrap even low abundance proteins from only
one 15cm cell culture dish can be analyzed. SILAC
is used to distinguish true interaction partners from
background binders.

We demonstrate pulldowns from SILAC-encoded
tagged and control HeLa cell lines. Tagged cell
lines are labeled with heavy arginine and lysine,
An empty control cell line or another tagged cell
line is labeled with light arginine and lysine.
Peptides derived from the two samples can be
distinguished by MS owing to their mass difference
and the signal ratio between SILAC pairs directly
indicates the protein abundance ratio. Unspecific
background binders show a ratio of 1:1 as they are
equally present in both samples. The bait protein         Figure 1. The histogram and boxplot show protein ratios
and specific interactors will have a ratio different of   obtained from a co-immunoprecipitation of GFP-tagged
1 as they are only present in either light or heavy       gamma tubulin (heavy SILAC amino acids) and HeLa wt
from. For tubulin gamma we identified all known           as a control (light SILAC amino acids). Background
                                                          binders are normally distributed (B). Outliers are either
interactors with high ratios and found one protein
                                                          contaminants like trypsin or human keratins introduced
with high confidence that is not yet known to be a        during sample preparation (A) or true tubulin gamma
gamma tubulin binder.                                     interactors like tubulin-gamma complex components 2-6
We also introduce a method to identify even               and T-complex proteins 1-8 (C and encircled) that show
underrepresented modifications in an unbiased             very high heavy/light ratios.
   “Tagless” Strategy of Protein Complex Identification: Towards High Throughput
           1               2             1              2            1          1                1
 Lee Yang , Haichuan Liu , Ming Dong , Simon Allen , Peter Walian , Bing Jap , Terry C. Hazen , Steven C.
          2                 1,2                        1               1                 2         1
      Hall , Susan J. Fisher , John-Marc Chandonia , Mark D. Biggin , H. Ewa Witkowska , Jian Jin
1                                                                    2
 Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA; Department of Cell and Tissue Biology,
                          University of California, San Francisco, CA 94143, USA.

Introduction                                                developed to enable information-based non-
A novel “tagless” strategy for protein complex              redundant MS/MS acquisition. The integration for
identification from native preparations has been            fully automated operation is under way.
proposed (1). Putative protein complexes are
identified based on the MS-monitored co-migration           Innovative aspects
of collections of polypeptides through multiple                • Identification of protein complexes under
orthogonal protein separation steps. As part of the                native     conditions     without    genetic
Genomics: GTL Protein Complex Analysis Project                     manipulation or affinity tags.
we are developing a high throughput pipeline to                • Fully integrated and automated system for
automate the processes of sample preparation,                      instrument control, intelligent MS/MS
MS data acquisition and interpretation to enable                   acquisition,    data      processing    and
analysis of thousands of fractions required for a                  management.
comprehensive       organism-wide    mapping     of
bacterial interactomes.                                     References
                                                               1. Dong et al., “A "Tagless" Strategy for
Methods                                                            Identification of Stable Protein Complexes
Cell lysates undergo a 4-step protein separation                   Genome-wide         by    Multi-dimensional
process (ammonium sulfate precipitation, SCX,                      Orthogonal Chromatographic Separation
HIC and SEC chromatography). Fractions                             and iTRAQ         Reagent Tracking”.     J.
collected at the last step are digested with trypsin,              Proteome Res. 2008 (in press)
labeled with iTRAQ and analyzed by MS/MS using
a LC MALDI workflow. Polypeptides are identified
and their elution profiles are derived on the basis
of iTRAQ-based relative quantitation.        Cluster
analysis of the data groups polypeptides with
similar elution profiles into putative protein
complexes. Methods are being developed for
high-throughput sample processing (digestion and
iTRAQ labeling), automated MS with a feed-back
loop for generating intelligent information-based
iterative MS/MS acquisition routines and
automated submission of polypeptide ID and
quantification data to clustering analysis.

The current scale of biomass production is 400 L
of D. vulgaris (10 g soluble protein).
Comprehensive characterization of the D. vulgaris
interactome necessitates analysis of thousands of
protein fractions. Development of modules and
tools required for automating the sample and data
processing is being pursued. To date, 0.2% of
protein separation space has been analyzed
resulting in detection of at least 6 heteromeric and
45 homomeric complexes. Identified bottle-necks
are being addressed: sample preparation, time
efficiency of MS data acquisition and data
management. Protein digestion and labelling are
performed on PVDF membrane in a 96-well
format. MS data acquisition, database search,
generation of polypeptide elution profiles and
clustering analysis can all be scheduled and
processed in batches. Algorithms for analyzing
and comparing precursors, retention time
predictions and exclusion list creation are being
Characterization of protein-complexes complementing functional-genomic screens in
                                  human cell-lines

  Magno Junqueira; Yusuke Toyoda; Mikolaj Slabicki; Zoltan Maliga; Mirko Theis; Dragomir Krastev; Frank
                          Buchholz; Antony Hyman and Andrej Shevchenko

               Max Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany

Introduction                                             on statistical considerations. In 93% of
Here we report on a rapid and sensitive pipeline         experiments we recovered the bait as one of top
for the systematic characterization of human             hits and in 74% we found potential interaction
protein    complexes,       which    employs    BAC      partners from which 20% have been confirmed by
expression of tagged proteins, digestion of affinity     other techniques.
purified proteins directly in large volumes of eluate
from affinity column, off-line concentration-cleanup     Innovative aspects
of peptides and their accurate identification via        • Identification of 70 new native mammalian
single LC-MS/MS run on a LTQ Orbitrap                       protein complexes
instrument.                                              • Affinity isolation of endogenous mammalian
                                                            complexes is an efficient orthogonal assay that
Methods                                                     complements esiRNA loss-of-function and
BACs with the localisation and affinity purification        localization screens.
(LAP) cassette containing GFP and S-peptide
sequences were inserted as carboxy-terminal
fusion and transfected in HeLa cells to obtain
stable expression. Human protein complexes were
purified by affinity chromatography using an anti-
GFP antibodies and digested with trypsin directly
in the elution buffer containing 100 mM glycine in
100 mM Tris pH 8.0. Tryptic peptides eluted in up
to 0.5 mL volume were cleaned up and
concentrated off-line on UltraMicroSpin C18
cartridge and then sequenced by LC-MS/MS on a
LTQ Orbitrap instrument using 145 min elution
gradient (5 to 60 % of acetonitrile in 0.1% formic
acid). In house developed scriptswere applied to
recognize and validate the interacting partners.

Results                                                  Figure 1- Comparative workflow for direct LC-MSMS
We demonstrate that, because of the high spectra         (shotgun) and for GeLC-MSMS. The new method allows
acquisition rate and dynamic range of LTQ                20 times faster analysis when compared with GeLC-
Orbitrap, no 2D-LC separation was required to            MSMS.
dissect the composition of protein complexes. Our
method provided 20x faster analysis with equal or
better performance compared to Gel-LC-MS/MS
approach – a recognized “golden” standard in the
We applied the method to a large series of baits
selected via genome-wide esiRNA loss-of-function
screens. Knock-down phenotypes usually do not
immediately reveal the unequivocal functional
assignment of targeted genes. However, clues on
protein complex function could be inferred by
following “guilty by association” concept, using
known function or characteristic sequence
domains of identified interaction parteners Within
15 months more than 300 pull-downs were                  Figure 2- APC-complex was isolated by TAP (CDC16
screened for more than 70 individual genes under         “Bait”) and the sample was split into two equal parts.
different experimental conditions. Each experiment       One was analyzed by GeLC-MS/MS, while another one
typically produced over 250 different proteins,          was digested in-solution and directly analyzed by direct
while ~80% were sorted out as background based           LC-MSMS. The chart shows the sequence coverage for
                                                         all subunits achieved by both methods.
Chromatin Proteome Demonstrates System-Wide Response To Replication Inhibition
                                                                          #                      §
                Guennadi A.Khoudoli*, Peter J. Gillespie*, Graeme Stewart , Jens S. Andersen ,
                                     Jason R. Swedlow*and J. Julian Blow*
       * - University of Dundee, College of Life Sciences, Dow Street, Dundee, DD1 5EH, Scotland, UK,

Introduction                                             to progression through replication (Figure 2). The
Stable and accurate propagation of chromosomal           data are consistent with the idea that Mcm2-7
DNA during cell division cycle is a fundamental          licensing complex plays a central role in coordi-
function of all living organisms. The pathways           nating nuclear structure with DNA replication [3].
regulating replication, expression and maintenance
of genetic information are well studied. However,        Innovative aspects
the way that these different processes work              • Functional modules on replicating chromatin
together as a coordinated biological system is              and their temporal order
poorly understood. Here we report a                      • System-wide effect of replication inhibition on
comprehensive analysis of the changing                      the dynamics of the chromatin proteome
association of proteins with chromatin (the              • Central role of Mcm2-7 licensing complex in
chromatin proteome) during progression through              coordinating nuclear structure with DNA
interphase of the cell cycle in the absence or              replication
presence of replication inhibitors.
Methods                                                  1. Futschik, M.E., and Carlisle, B. (2005). J
Nuclei were assembled in vitro in Xenopus cell              Bioinform Comput Biol 3, 965-988.
free extracts and then purified at defined points        2. Dennis, G., at al. Genome Biol 4, P3.
during the progression through interphase during         3. Khoudoli, G.A. at all (Prepared for publication)
replication, as well as in the presence of replication
inhibitors geminin or roscovitine. Associated
proteins were eluted from chromatin, digested with
trypsin and subjected to LC MS/MS analysis. The
proteins abundance was estimated from the
extracted ion chromatograms (XIC) of their
corresponding peptides. Using fuzzy c-mean
clustering (FCM) [1] proteins were grouped into 12
clusters based on the dynamics of their
association with chromatin. The DAVID software
[2] was used to collect ontological terms
associated with all the proteins and to determine if
any of them were enriched in specific FCM
clusters.                                                Figure 1. FCM clustering of temporal profiles of
                                                         polypeptides associated with replicating chromatin
The final set of non-redundant proteins subjected
to the analysis consisted of 746 entries. Of the 606
proteins identified on untreated chromatin, 458
demonstrated more than 30% variation in
abundance during interphase. 12 clusters
generated by FCM algorithm were divided into
three general types that have their peak
abundance on chromatin in early, intermediate or
late interphase (Figure 1).
Annotations analysis demonstrated that FCM
clusters identify a series of modules consisting of
functionally-related protein factors that associate
with interphase chromatin in a defined order.            Figure 2. Hierarchical clustering of the combined data
We examined the effect of blocking DNA                   set.
replication on the chromatin proteome by inhibiting
either replication licensing or S phase CDK              # - . Biomedical Research Centre, University of
activity. This revealed an unexpectedly broad            Dundee, Ninewells Hospital and Medical School,
system-wide effect on the dynamics of the                Dundee. DD1 9SY, UK
chromatin proteome. The behaviour of functional          § - Center for Experimental BioInformatics,
groups reflected a range of different nuclear            University of Southern Denmark, Campusvej 55,
functions that are synchronised with and respond         DK-5230 Odense M, Denmark
         Quantitative Analysis of C/EBPα Transcription Complexes in Leukemia

                   Rositsa I. Koleva, Scott B. Ficarro, Manor Askenazi, Jarrod A. Marto
    Department of Cancer Biology, Dana-Farber Cancer Institute, Department of Biological Chemistry and
        Molecular Pharmacology, Harvard Medical School, 44 Binney St., Boston, MA 02115, USA

Acute myeloid leukemia (AML) remains a highly            Innovative aspects
lethal malignancy with limited therapeutic options.      • Novel multidimensional fractionation with true
Mounting evidence indicates that impaired function          nanoflow chromatography
but not complete absence of the transcription            • Quantitative analysis of protein complexes
factor C/EBPα leads to consecutive accumulation             responding to molecular switch
of myeloid progenitors and blast crisis in AML.          • Target      protein    validation in  myeloid
Recent studies demonstrated that oncogenic Flt3             differentiation model
signaling leads to phosphorylation on serine 21
(pS21) of C/EBPα with concomitant impairment of          References
myeloid differentiation [1, 2]. We hypothesized that     (1) Radomska, H.S., et al., Block of C/EBPα function by
assembly of transcriptionally active protein                  phosphorylation in acute myeloid leukemia with
complex is governed by pS21-regulated protein                 FLT3 activating mutations. J. Exp. Med., 2006.
associations.                                                 203(2): p. 371-381.
                                                         (2) Ross, S.E., et al., Phosphorylation of C/EBPα
                                                              Inhibits Granulopoiesis. Mol. Cell. Biol., 2004. 24(2):
                                                              p. 675-686.
To test this hypothesis we established affinity
tagged C/EBPα expression system under control
of a tet-inducible promoter in myeloid cells with
constitutive Flt3 activity. Treatment of these cells
with Flt3 inhibitors modulated pS21 in a dose-
dependent manner. Next, advanced quantitative
proteomics methodology, including true nanoflow
LC coupled with multidimensional RP/RP
fractionation and on-column iTRAQ stable isotope
labeling, was used to monitor remodeling of
C/EBPα protein complexes as a function of pS21.
siRNA methods were used to abrogate expression
of interactors modulated by pS21 to determine
their functional role in transcription of myeloid-
specific genes.
                                                         Figure 1. C/EBPα putative partners’ network. The edges
Results                                                  reflect HPRD data base annotated protein-protein
Our data represent by far the largest catalog            interactions assessed by various experimental
C/EBPα interactors, including more than 200              approaches. The network is enhanced by addition of
proteins involved in chromatin organization,             new edges (in light green) based on manual curation of
transcriptional modulation, and cell cycle               recent literature. Target proteins selected for further
regulation.    Furthermore,      our     quantitative    studies are highlighted.
proteomics data demonstrate that 1) C/EBPα                      9.0
                                                                                             C/EBPα activation
interacts with proteins genetically linked to                   8.0
leukemia; and 2) many of these interact with
C/EBPα in a phosphorylation-dependent manner.                   5.0
Knock-down       of   newly-identified,    leukemia-            4.0
associated interactors (targets #1-4) reduced the               3.0
ability of C/EBPα to drive expression of
granulocytic target genes. Our data demonstrate                 0.0
that phosphorylation on serine 21 modulates                             Control        A           B
                                                                                  Sub siRNA 1 Sub siRNA (     A
                                                                                                            Meis1#1    B
                                                                      Control                     2)
association of C/EBPα with protein partners that                      siRNA
                                                                                     Target #1                 Target #2
are functionally relevant for myeloid differentiation.                                                          siRNA
Our ability to quantitatively monitor multiple
leukemia-related gene products in the context of         Figure 2. Depletion of target genes #1 and #2,
C/EBPα protein complexes provides valuable               corresponding to novel C/EBPα partners, decreases
insight into the mechanisms by which oncogenic           C/EBPα capacity to induce the transcription of
kinase activity disrupts transcription and leads to      granulocytic genes in myeloid cells.
    Shot-gun membrane proteomics by acid hydrolysis combined with trypsin and
                           1              1            2                 1                       1
             Joseph Kown , Jeehyun Oh , Sunghoon Lee , Seung Il Kim , and Jong-Soon Choi
             Proteome Research Team, Korea Basic Science Institute, Daejeon 305-333, Korea
                      Korean Bioinformatics Center, KRIBB, Daejeon 305-806, Korea

Introduction                                               helix proteins, resulting in a high coverage level of
Proteomics of membrane proteins is essential for           membrane proteins. The predominantly expressed
the understanding of cellular function. So far, mass       proteins by the relative abundance indexing
spectrometric analysis of membrane proteome has            revealed the components of phycobilisome
been done primarily in the identification of soluble       complex (CpcA, CpcB, ApcB, ApcA, ApcE, CpcD),
proteins. In order to attain comprehensive                 PS-II complex (PsbX, PsbY, PsbK, PsbE, PsbD1),
membrane proteome in a cell, the well-developed            PS-I complex (PsaF, PsaM, PsaB) and complex V
sample preparation protocol is required for solving        (AtpG). These are notably involved in the
membrane      hydrophobicity.     Both    gel-based        photosynthesis by capturing light energy.
analysis and shot-gun mass spectrometry method
can be used for membrane proteome analysis.                Innovative aspects
Here we challenged to analyze the whole                    • Acid hydrolysis of multiple trans-membrane
membrane proteome of cyanobacteria by acid                    proteins at aspartic acid for better identification
hydrolysis combined with trypsin and chymotrypsin.         • Acid hydrolysis followed by either trypsin or
Cyanobacteria are a model photosynthetic                      chymotrypsin treatment for membrane proteins
microorganism that is capable of energy                    • Massive identification of membrane proteins
generation and metabolite transport via multiple-             by shot-gun membrane protein profiling
layered membranes and associated proteins (1).                MudPIT
Shot-gun membrane proteomics can give a more
concrete data in mapping the repertoire of                 References
cyanobacterial membrane-linked proteome.                      (1) F. Huang et          al,  Proteomics  of
                                                                  Synechocystis sp. strain PCC 6803; Mol.
Methods                                                           Cellular Proteomics. 2002 1, 956-966.
Using a whole proteome scale prediction method,               (2) C. C. Wu and J. R. Yates III, The
768 proteins were found to be present in the                      application of mass spectrometry to
genome         of     unicellular   cyanobacterium                membrane proteomics; Nature Biotech.
Synechocystis sp. PCC 6803. We employed an                        2003 21, 262-267.
experimental method to verify the membrane
proteome en masse. A shot-gun membrane                     A                          B
proteomic technique was used in combination of
acid hydrolysis at aspartic acid with trypsin or
chymotrypsin       by    multi-dimensional protein
identification technology (MudPIT) (2).

A total of 472 proteins were successfully identified
from 1840 unique peptides and 153 proteins
among them were membrane-linked proteins such
as extracellular [3], outer membrane [11],
periplasmic [7], plasma membrane [98], and
membrane complex [34] were identified by
Synechocystis sp. PCC 6803 proteome database.              Figure 1. Size distribution and predicted region of
In particular, 134 integral membrane proteins              identified peptides resulting from chemical-trypsin
(IMP) with transmembrane helices were detected             method (CT) and chemical-chymotrypsin method (CC)
by Phobius and Soshui prediction methods.                  Peptide masses are plotted at 500 Da unit. The
Among these, the majority of detected IMP was              distribution of peptides produced by the theoretical
done by the chemical digestion with chymtotrypsin          prediction is compared to that produced in experiment
[100/134] rather than trypsin [52/134]. As expected        (B). Each peptide is annotated the portion of signal
in the mass range prediction of peptides cleaved           peptide (pale blue), extracellular region (red),
chemically and enzymatically from whole                    transmembrane region (yellow) and cytoplasmic region
membrane proteome, the high frequency of mass              (blue) as a color.
range 1500~3000 generated by chymotrypsin after
chemical digestion was available for MS/MS
analysis to identify the multi-spanning membrane
                                 Proteomics study on human Ccr4-NOT:
                       a multi-functional complex involved in mRNA metabolism
                   1                       2                 1                 2                         1
    Nga-Chi Lau , Annemieke Kolkman , W.W.M. Pim Pijnappel , Albert J.R. Heck , and H. T. Marc Timmers
 Department of Physiological Chemistry, University Medical Center – Utrecht, Universiteitsweg 100, 3584 CG
Utrecht, The Netherlands; Department of Biomolecular Mass Spectrometry, Utrecht University, Sorbonnelaan
                                 16, 3584 CA Utrecht, The Netherlands

Introduction                                              Results
The Ccr4-NOT complex consists of multiple CNOT            The obtained datasets revealed that not all CNOT
proteins and is evolutionarily conserved from yeast       proteins are stable components of the human
to mammals. Its CNOT subunits have distinct               Ccr4-NOT complex. Also, new subunits of the
functions in synthesis and degradation of mRNA            complex have been identified as well as other
molecules (figure 1). A developmental role is also        interesting novel interactors. In addition, several
indicated for CNOT7 in mouse spermatogenesis              distinct Ccr4-NOT complexes have been identified.
and for CNOT1 in early development of C.elegans.          Future experiments will be focused on the
Here we describe a comprehensive proteomics               functional differences between these Ccr4-NOT
study in human cells to identify the composition          complexes in human cells.
and cellular interactors of the whole Ccr4-NOT
complex in order to understand its functioning in
cells.                                                    Innovative aspect
                                                          • Combining epitope-tagging in human cells,
    protein ubiquitylation      mRNA degradation             double     affinity purification and    mass
     (E3 ubiquitin ligase)       (deadenylation)             spectrometry not only identified novel
                                                             components of macro-molecular complexes,
                       CNOT4                                 but also revealed distinct complexes differing
     CNOT3                                                   in subunit composition
                             CNOT1      CNOT8


        mRNA synthesis

Figure 1. Human Ccr4-NOT complex components
(CNOT proteins) are involved in synthesis (yellow) and
degradation (green) of mRNA molecules. CNOT4 may
play a role in protein ubiquitination, and CNOT3 has an
unknown function.

Clonal stable human cell lines expressing near
endogenous levels of epitope-tagged CNOT
proteins were created. Associated proteins of a
tagged subunit were purified from a cell line via a
two-step affinity purification protocol (flag-HA),
followed by in-solution trypsin digestion and LC-
LTQ-FT       MS/MS      mass      spectrometry    for
identification. Data integration of all CNOT
purifications resulted in a protein network centered
on the Ccr4-NOT complex components.
 Identification of protein phosphorylation sites associated with protein complex
              assembly using an IMAC-based proteomics procedure
                                                    1                  1, 2
                                  Chang-Hung Lee and Yeou-Guang Tsay
             1                                                    2
              Institute of Biochemistry and Molecular Biology and Proteomics Research Center,
                                 National Yang-Ming University, Taipei, Taiwan

It is a common belief that protein modifications,
like phosphorylation, may modulate protein
complex metabolism and thus effectively regulate
protein functions. However, there are only a
limited number of known examples thus far and
most of them are uncovered via the conventional
molecular biology approach.

Here we demonstrate the design and application of
a new proteomics platform that aims to identify
protein phosphorylation sites that are associated
with protein complex assembly. We first undertake
gel filtration liquid chromatography to resolve the
protein complexes from Hela cell at their native
states.      Immobilized ferric ion (Fe ) affinity
chromatography is then employed to enrich the
phosphorylated peptides in each gel filtration
fraction, whose identities and the phosphoamino
acid residues within are documented using liquid
chromatography-tandem mass spectrometry along
with computer programs like TurboSequest and
our in-house software. In order to optimize the
isolation method for phosphopeptides, we have
tested a multitude of the experimental conditions
and systematically examine their impacts on the
specificity and comprehensiveness of the

Among various factors, we found that the pH and
organic solvent in wash and elution solutions
appeared to be particularly important. With this
platform, we have identified a group of
phosphorylation sites from HeLa cells that seem to
be only present in the high-MW protein complexes,
but not in the low-MW ones.

Innovative aspects
• Development of a new proteomics platform
• Optimization of IMAC procedure for
   phosphopeptides purification
• Identification of protein phosphorylation sites
   associated with protein complex assembly
    Molecular recognition in oligomeric enzymes: integration of bioinformatic and
                               biosensoric approaches

                     Yu. Mezentsev, A. Lisitsa, P. Ershov, A. Molnar, A. Ivanov, A. Archakov
                             V.N. Orechovich Institute of Biomedical Chemistry RAMS
                                  Pogodinskaya str. 10, Moscow, 119121, Russia
Introduction                                                  This work was supported in part by Russian
Molecular recognition in protein complexes plays a            Foundation for Basic Research (grant 07-04-
central role in biochemical processes. It                     00575).
understanding is an important task in different               Innovative aspects
fields of biomedical science and drug discovery.              Combination of bioinformatic and biosensoric
The interface areas of protein-protein interactions           methods in protein oligomerization research.
have unique structures and represent prospective              References
targets for a new generation of drugs. The most               (1) Mezentsev Yu.V., Molnar A.A., Gnedenko O.V.,
interesting group of similar targets are oligomeric               Krasotkina Yu.V., Sokolov N.N. and Ivanov
enzymes. This report shows integrative approach                   A.S. Oligomerization of L-Asparaginase from
using bioinformatic and biosensoric methods to                    Erwinia carotovora. Biochemistry (Moscow)
research of molecular recognition in oligomeric                   Supplement Series B: Biomedical Chemistry.
enzymes.                                                          2007, 1(1), 58–67.
Methods                                                       (2) Ivanov A.S., Gnedenko O.V., Molnar A.A.,
We have chosen two oligomeric enzymes as test                     Mezentsev Y.V., Lisitsa A.V., Archakov A.I.
molecular objects most convenient for such                        Protein-protein interactions as new targets for
research — HIV-1 protease (HIVp) (homo-dimer)                     drug design: virtual and experimental
and bacterial L-asparaginases (homo-tetramer) [1].                approaches. J Bioinform Comput Biol. 2007,
We have used some computer methods                                5(2b), 579-592.
(molecular modeling, computational alanine
scanning, molecular dynamics simulation and
energy optimization), as well as direct molecular
interaction measurement by SPR-biosensor. All
calculations were done using Sybyl 6.9.1 (Tripos
Inc.) software running on SGI Origin200 server
and Amber 7. Biosensor measurements were
carried out on Biacore-3000.
Computer analysis of subunits contact areas in
HIVp dimer was done using virtual «alanine
scanning». Several amino acid residues which
bring the significant contribution into the interaction
energy (“hot spots”) have been found. Also such
analysis     of    subunits     contact    areas      in
homotetramers of bacterial L-asparaginases was
done. The basic attention was given the interface
between dimers AC and BD and between
monomers in these dimers. It was shown, that in
each subunit there are 13 residues which play a
key role in interaction between dimers AC and BD.
We modeled chimeric tetramers by substitution of
one subunit on subunit from another L-
asparaginase. The value of calculated interaction
energy decreased.
We tested this results in experiments on optical
biosensor. It was shown, that the interaction
between subunits from different L-asparaginases
is impossible, even between subunits with high
similarity (sequence identity >90%).
Thermodynamics of subunits interaction in both
enzymes, as well as interaction inhibitors of HIVp
dimerization [2] was also studied using optical

Cell surface receptors and their associated signaling pathways on the plasma membrane

are key targets in understanding cellular responses. However, the isolation and

identification of receptor complexes has been elusive. The Fc receptor was captured by

the from the surface of live cells using microbeads coated with the receptor’s cognate

ligand (IgG) and analyzed by LC-MS/MS alongside several controls. Live-cell Affinity

Receptor Chromatography (LARC) resulted in a partially non-redundant list of 288

proteins that were specific to the Fc receptor complex. The proteins identified were in

close agreement with previously determined factors in the Fc receptor complex as

demonstrated by previous genetic and biochemical methods and permitted the discovery

of novel complex members. Confocal microscopy was used to confirm recruitment of

specific members of the Fc, SRC, SYK, PLC, PKC, PI3K, SHIP, TEC, CDC42, RAP,

PAK, GAP, GEF, GRP and CRK to the receptor complex upon activation by the same

ligand microbeads. The expression of mutants and silencing RNA against specific

isoforms were used to demonstrate a functional role for novel members of the Fc receptor

complex including RHOG, P115 RHOGEF, and CRKL. The recruitment of AKT PH

domain GFP was used to quantify the production of phosphorylated inositol at the

activated receptor complex. We conclude that it is feasible to capture an activated

receptor complex from the cell surface of live cells using ligand coated microbeads for

identification of members of a receptor complex or pathway by liquid chromatography

and tandem mass spectrometry.
      Post-Translational Modifications of the Endogenously Expressed Transcription
                                 Factors TFIID and SAGA
                            1                 1              2                  2                 1
             N. Mischerikow , A.F.M. Altelaar , H.T.M. Timmers , W.W.M. Pijnappel & A.J.R. Heck
      Biomolecular Mass Spectrometry and Proteomics Group, Utrecht University, Sorbonnelaan 16, 3584 CA
      Utrecht, The Netherlands; Department of Physiological Chemistry, University Medical Center Utrecht,
                            Universiteitsweg 100, 3584 CG Utrecht, The Netherlands

Introduction                                              References
The general transcription factor TFIID and the            (1) Sanders et al., Proteomics of the Eukaryotic
histone acetyl transferase SAGA are multi-subunit             Transcription Machinery: Identification of
protein complexes in Saccharomyces cerevisiae                 Proteins Associated with Components of
that have a number of subunits in common. Up to               Yeast TFIID by Multidimensional Mass
now, most proteomic studies have led to a static              Spectrometry. Mol Cell Biol, 2002
picture of these complexes (1). However, their            (2) Puig et al., The Tandem Affinity Method: A
subunit composition may vary with time as a                   General Procedure of Protein Complex
mechanism for transcriptional regulation. In this             Purification. Methods, 2001
project we compare the post-translational state of
affinity purified TFIID and SAGA. The aim is to
comprehensively identify post-translational modifi-
cations (PTMs) that might regulate the dynamics of
components shared between the complexes.

TFIID and SAGA were tandem affinity purified from
yeast whole cell extract (2) and precipitated. Both
complex preparations were analyzed by SDS
PAGE, digested in gel with Trypsin and analyzed
by LC-MS. Additionally, both preparations were
also digested in solution with Lys-C/Chymotrypsin,
Lys-C/Glu-C and Lys-C/Trypsin, followed by SCX
chromatography and LC-MS. Peptide analysis was
realized using nano-LC coupled online to a LTQ-
Orbitrap. Peptides and proteins were identified
using the Mascot database search engine.

The generic tandem affinity purification in combi-
nation with tryptic in gel digestion allowed the iden-
tification of all subunits commonly ranked among
TFIID and SAGA with good average sequence co-
verage. However, in some of the 5 TBP-associated
factors (TAFs) present in both TFIID and SAGA,
sequence coverage was relatively low. Moreover,
in these and other subunits segments of sequence
of functional importance were poorly covered. We
therefore followed a multiple protease approach to
enhance sequence coverage as a prerequisite to
comprehensively map PTMs. Using this approach,
we were able to identify a number of different
PTMs on both complexes which to our knowledge
have not been reported.

Innovative aspects
•   comprehensive description of different types
    of PTMs of yeast TFIID and SAGA
•   comparison of PTMs on subunits shared
    between yeast TFIID and SAGA
Spatiotemporal Changes of PKA-AKAP Complexes in Response to Prostaglandin E2.
                               1                2                   1                   2              1
     Nikolaus G. Oberprieler , Thin Thin Aye , Knut Martin Torgersen , Albert J.R. Heck , Kjetil Taskèn
                 1) Biotechnology Centre of Oslo, University of Oslo, N-0317 Oslo, Norway
  2) Department of Biomolecular Mass Spectrometry, Utrecht University, 3584 CA Utrecht, the Netherlands

Introduction                                             PKA-AKAP signalling complexes and provide new
Intracellular signalling events play a crucial role in   insights into the spatiotemporal regulation of PKA-
T-lymphocytes         under    physiological      and    AKAP signalling complexes in T-lymphocytes.
pathological conditions. Prostaglandin E2 (PGE2)
signalling is important in the negative regulation of    Innovative aspects
T-cell receptor signalling and increased PGE2            • Chemical proteomics approaches lead to the
signalling has been implicated in diseases such as          successful     enrichment     of PKA-AKAP
HIV infection and cancer. PGE2 signalling is                complexes.
primarily mediated through cAMP and the                  • A combination of molecular biology and
subsequent activation of PKA. In order to allow             proteomics approaches revealed novel PKA-
targeted action of PKA the kinase is anchored to            AKAP signalosome members.
macromolecular signalling complexes by protein           • Novel aspects of the spatiotemporal regulation
kinase A anchoring proteins (AKAPs).                        of PKA signalling were revealed.

8-AHA-cAMP-agarose beads were used to enrich
cAMP binding proteins and their secondary binding
partners. This resulted in a significant enrichment
of PKA regulatory subunits and their anchoring
proteins (AKAPs). For quantitative proteomic
analysis, peptides from agarose bead fractions
were dimethyl labelled and subjected to ESI-
FTICR MS. Additionally, identified AKAPs were
expressed in SILAC labeled Jurkat T-cells as
FLAG-tagged proteins. Following expression,
AKAP complexes were purified using anti-FLAG
antibody covalently linked to magnetic Dynabeads.
Isolated AKAP complexes were analysed by LC-
orbitrap MS. Furthermore, GFP-tagged AKAPs
were expressed in Jurkat T-cells to determine
changes in AKAP localization following PGE2
treatment using fluorescent microscopy.

The chemical proteomics approach using 8-AHA-
cAMP-agarose beads identified 6 AKAPs which
are expressed in primary human T-lymphocytes.
Furthermore, quantitative analysis of samples
stimulated with PGE2 for 10 and 60min revealed
temporal regulation of PKA-AKAP complexes.
Several proteins were identified which either get
recruited to PKA-AKAP complexes or dissociate            Figure 1: Primary T-lymphocytes were stimulated with
following PGE2 treatment. In order to determine          PGE2 (10μM) for 10 and 60min and compared to
the composition of AKAP-specific signalling              unstimulated controls (CT). Cells were lysed and cAMP-
                                                         binding fractions enriched using 8-AHA-cAMP-agarose
complexes, the identified AKAPs were individually
                                                         beads before subjecting each sample to SDS-PAGE and
expressed as FLAG-tagged proteins in SILAC               Coomassie blue gel staining. Arrows highlight proteins
labelled Jurkat T-cells. This allowed the                which are recruited or dissociate with PGE2 treatment.
identification of several AKAP-specific signalling-
complex members and revealed their temporal
regulation in response to PGE2 treatment. Again, a
number of proteins were specifically recruited or
dissociated from AKAP signalling complexes. The
expression of GFP-tagged AKAPs in Jurkat T-cells
allowed the analysis of spatial regulation of AKAP
complexes in response to PGE2 treatment. Taken
together, the results identify novel members of
                          1                2             3                                3                                                               1
        Nerea Osinalde , Kerman Aloria , Ana Zubiaga , Asier Fullaondo and Jesus M. Arizmendi
    Department of Biochemistry and Molecular Biology, University of the Basque Country, Leioa, Spain.
                  Proteomics Unit SGIker, University of the Basque Country, Leioa, Spain
  Department of Genetics, Physical Anthropology and Animal Physiology, University of the Basque Country,
                                              Leioa, Spain

Introduction                                                 Now these potential E2F2 interacting partners are
E2F2 is member of the E2F transcription factor               being validated. The oligoprecipitates will also be
family, involved in cell cycle progression,                  analyzed by LC-MS/MS to identify new potential
apoptosis, DNA damage repair, differentiation and            E2F2 interacting proteins.
development. Although classically E2F2 has been
classified as a cell cycle activator, previous work in       Innovative aspects
our lab has demonstrated that it can also act as a           • Selective enrichment of E2F2 containing
cell cycle repressor (1). It is known that the                  protein complexes using an specific DNA
transcriptional activity of E2F2 is dependent on its            sequence as a bait.
interacting partners (2), but the ones described so          • Comparison between two different
far are insufficient to understand the way E2F2                 precipitations techniques for the identification
regulates the cell cycle. Here we describe two                  of new E2F2 interacting proteins.
different methods for the identification of new E2F2
interacting proteins.                                        References
                                                                (1) M. Murga et al, Mutation of E2F2 in mice
Methods                                                             causes enhanced T lymphocyte
E2F2 containing protein complexes were                              proliferation, leading to the development of
immunoprecipitated (IP) with a specific antibody                    autoimmunity; Immunity 2001.
against HA and oligoprecipitated (OP) using a                   (2) S. Schlisio et al, Interaction of YY1 with
specific DNA (Chk1 promoter) containing the E2F                     E2Fs, mediated by RYBP, provides a
consensus sequence as a bait, from HA-E2F2 and                      mechanism for specificity of E2F function;
HA-E2F2 / HA-DP1 overexpressing cell lysates,                       EMBO J. 2002.
respectively.       Immunoprecipitations     were

                                                                                                                                             ß- actin
performed 5 times. In both cases the precipitates

were analyzed by western blot to confirm that the                A                                                       B

precipitations     were   performed      correctly.


Components of the immunoprecipitated protein                                                  :IP
complexes were identified by LC-MS/MS after
tryptic digestion.                                                                                  E2F2                                                         DP-1

                                                               IB: antiE2F2                                                               IB: antiHA
We have setup two different and complementary
precipitation methods that allow us to enrich the            Figure 1. (A) E2F2 was immunoprecipiated with anti-HA
cell lysates in E2F2 and its interacting partners.           but not with the control antiT antibody. (B) E2F2 and
Starting from HA-E2F2 overexpressing 293T cell               DP-1 were oligoprecipitated with the DNA containing the
lysates, we were able to immunoprecipitate E2F2              E2F consensus sequence (Chk1) but not with the
using a specific antibody against HA but not using           control DNA (ß- actin).
an unespecific antibody (Fig 1A). Moreover,                  A                                      B                    50
starting     from     HA-E2F2     and      HA-DP1                                                                        45
                                                                                                    Number of proteins

overexpressing 293T cell lysates, we were able to                                antiHA
                                                               antiT                                                     35

oligoprecipitate E2F2 and its dimerization partner                                                                       30

DP1, using a specific DNA (Chk1 promoter)                        17        71         60

containing the E2F consensus sequence as a bait                                                                          15
but not with an unespecific DNA sequence (ß-actin                                                                        5

promoter) (Fig 1B).                                                                                                      0

We identified E2F2 and several potential E2F2                                                                                 1       2        3       4                5
interacting partners by LC-MS/MS from the                                                                                           Number of replicates
immunoprecipitates (Fig 2A). We consider specific
                                                             Figure 2. (A) Number of proteins identified in                                                           5
proteins those that only appeared in the problem
                                                             immunoprecipitations     against   HA      and                                                          5
sample and in none of the negative control                   immunoprecipitations against SV40Ag T. (B) Number                                                       of
replicates. Fig 2B shows the number of specific              specific proteins present in one or more of the                                                         5
proteins present in one or more of the 5 replicates          replicates.
Quantitative phospho-proteomics elucidates phosphorylation-dependent transcriptional regulation of
                                   the yeast Mediator complex

Gideon Oudgenoeg1, Joris Benschop2, Manuel Tzouros1, Tony Miles2, Frank Holstege2, Jeroen Krijgsveld1
Biomolecular Mass Spectrometry and Proteomics group, Utrecht University, Utrecht, the Netherlands
Genomics laboratory, University Medical Center Utrecht, Utrecht, the Netherlands

Introduction                                             application to the Mediator complex has linked
Mediator is a 26-subunit protein complex that is         kinase activity in the Mediator to transcriptional
associated with RNA polymerase II. It is required        regulation by this complex. The developed
for transcription of all protein-coding genes in         methodology will be applicable to other types of
yeast and is functionally conserved from yeast to        modifications, regulatory factors and organisms.
mammals. One of the Mediator subunits is Cdk8, a
protein kinase mainly involved in transcription          Innovative aspects
repression, notably through phosphorylation of the       • Identification of kinase-specific
repeated C-terminal domain of Rpb1 Pol II subunit.          phosphorylation sites
In addition, phosphorylation of other Mediator           • Linking kinase activity to transcriptional
subunits may serve as a mechanism of                        regulation by combination of genomics and
transcription regulation. The aim of this study was         proteomics.
to identify Cdk8-dependent phosphorylation target
sites within the Mediator complex.

Isolation of specific subunits for analysis of               (1) Pinkse W. H. et al., Highly Robust,
phosphorylation was performed by Tandem Affinity                 Automated,      and    Sensitive Online
Purification (TAP), thereby co-purifying associating             TiO2-Based Phosphoproteomics Applied
subunits as well. Phosphorylated peptides in                     To Study Endogenous Phosphorylation in
proteolytic digests of affinity-purified Mediator were           Drosophila melanogaster; J. Proteome
enriched by TiO2 affinity purification followed by               Res., 7 (2), 687–697, 2008.
nanoflow LC-MSMS (1) using a LTQ-FT mass
spectrometer. Kinase- and condition-dependent                (2) van de Peppel, J. et al., Mediator
phosphorylation of Mediator subunits was inferred                expression profiling epistasis reveals a
from quantitative changes in phosphorylation using               signal    transduction   pathway      with
wt and Cdk8 mutant strains, grown in normal and                  antagonistic submodules and highly
15N-labeled      media,      respectively.   Targeted            specific downstream targets. Mol. Cell 19,
quantification of phosphorylated peptides of                     511–522, 2005.
interest was performed by Multiple Reaction
Monitoring (MRM) using nano-RP-LC coupled to a
hybrid triple quad - ion trap instrument.

Multiple phosphorylation sites were identified in
several Mediator subunits. Quantitative analysis
revealed that several of these phosphorylation
events were decreased in the Cdk8 mutant thus
indicating that phosphorylation was Cdk8-
dependent. To test functional relevance and
putative regulatory role of candidate sites, point-
mutants of Cdk8-dependent phospho-serines in
Med15 were made. Expression profiles of these
point-mutants were compared to wt, ∆Cdk8 and
∆Med15 deletion strains. This revealed that mRNA         Figure 1.Modular organization of the Mediator
                                                         Complex (2) and close functional relationships between
profiles of Med15 point mutants were different from
                                                         the Mediator modules and other transcriptional
both wt and ∆Med15 deletion strains, but very            regulators.
similar to the ∆Cdk8 mutant. This indicates that
Cdk8-mediated phosphorylation of Med15 is a
direct modulator regulating gene transcription.
The present work shows that quantitative
phospho-proteomics is a powerful tool to reveal
kinase-specific    phosphorylation    sites.      Its
         Targeting Dynamic Changes of the CFTR Interactome in Cystic Fibrosis

                    Sandra Pankow, Casimir Bamberger, William Balch, John R. Yates IIIrd
                     Department of Chemical Physiology, The Scripps Research Institute,
                          10550 North Torrey Pines Road, La Jolla, CA 92037, USA

Introduction                                             between the different complexes and determine
70% of all cystic fibrosis cases are caused by           the timely manner of the interactions. Eventually,
deletion of phenylalanine 508 (∆F508) of the cystic      close evaluation of these results might elucidate
fibrosis transmembrane conductance regulator             potential therapeutic targets for cystic fibrosis that
(CFTR). In contrast to wild-type (wt) CFTR most of       can be initially tested in the present system.
the ∆F508 CFTR is incorrectly folded and
consequently proteolytically degraded, probably
due to loss of key interaction partners. To
determine and distinguish differences in the             Innovative aspects
composition of protein complexes associated with         • Development of a sensitive IP protocol for
the endogenous wt and mutant CFTR during the                endogenous membrane-embedded proteins
lifecycle    of   the     protein,    we   applied       • Previously non-identified interaction partners
immunoprecipitation and MudPIT (1) in a                     were determined for mutant and wt CFTR
combined approach.                                       • Dynamic changes in protein interactions during
                                                            the protein lifecycle can be monitored

CFTR is a low abundant 12-transmembrane                  References
protein that is hardly accessible by normal                 (1) M.P.Washburn et al., Large scale analysis
experimental procedures. We therefore developed                 of the yeast proteome by multidimensional
an immunoprecipitation (IP) protocol that provided              protein identification technology; Nat.
high sensitivity and allowed us to detect CFTR and              Biotech 2001
associated protein complexes from patient-derived           (2) X.Wang et al., HSP90 Cochaperone Aha 1
cell lines carrying either the ∆F508 CFTR gene or               downregulation rescues misfolding of
the re-inserted wt CFTR gene. Parental cells in                 CFTR in cystic fibrosis; Cell 2006
which the CFTR gene had been silenced were
used as control. For better access to membrane-
embedded proteins, membranes were slightly
disrupted by applying gentle sonication before IP.
Immunoprecipitated proteins were then further
extracted with Methanol/Chloroform to remove
interfering lipids. To possibly most extensively
cover the network of interacting proteins, a
modified MudPIT protocol capable of eluting
hydrophobic peptides was applied.

Using the newly developed strategy we were able
to identify over 350 proteins that constitute directly
or indirectly CFTR containing protein complexes.
Of these, roughly 92% are newly identified as
CFTR interaction partners and about 40% are
specifically associated with the ∆F508 mutation.
Whereas the previously reported network of CFTR
interacting proteins (2) represents a core network
allowing a relatively static view of the complexes
associated with CFTR, the extended network
enables us to follow the dynamic changes in the
CFTR interactome that occur during folding and
trafficking and that determine membrane stability,
e.g. reflect the different cellular processes
comprising the lifecycle of CFTR from nascence to
degradation. Fractionation of the subcellular
compartments will further allow to distinguish
                     Cell cycle regulation of TBP complexes in human cells
                         1                        2                 1                 2
  W.W.M. Pim Pijnappel , Annemieke Kolkman , Marijke P.A. Baltissen , Albert J.R. Heck , and H. Th. Marc
 Department of Physiological Chemistry, University Medical Center – Utrecht, Universiteitsweg 100, 3584 CG
 Utrecht, The Netherlands; Department of Biomolecular Mass Spectrometry, Bijvoet Center for Biomolecular
 Research and Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Sorbonnelaan 16, 3584 CA
                                         Utrecht, The Netherlands

Genome wide studies have revealed waves of
gene expression during the cell cycle. In addition,
mitotic cells exhibit a transcriptional shut down. To
investigate whether these events are regulated at
the level of general transcription factors, we
analyzed the compositions, abundances and post-
translational modifications of TBP complexes
during the human cell cycle using affinity
purification and SILAC.

HeLa cells were transduced with flag-HA tagged
TBP using retroviral gene delivery. Expression of
tagged TBP matched endogenous expression.
Cells were isotopically labelled using heavy
arginine and lysine, and were grown in a 15 liter
bioreactor. Comparisons were made between
cells grown asynchronously and cells blocked in
mitosis (using nocodazole) or in S phase (using
thymidine). TBP complexes were affinity purified,
digested by trypsin and analyzed by nano-LC-
LTQ-FT-MS. Phosphopeptides were enriched
using titanium oxide precolumns.

Our data indicate that the compositions of stable
TBP complexes remained largely intact, whereas
their relative abundances changed during the cell
cycle. In addition, we detected a number of novel
phosphorylations on TBP-associated factors
(TAFs), some of which were enriched in mitosis or
in S phase. The function of one of the mitotic
phosphorylations has been studied in more detail
using a functional replacement with
phosphopreventing or phosphomimetic mutant
proteins and chromatin association and
transcription assays. The results of these analyses
will be presented.

Innovative aspects
• Generic double affinity purification method for
   medium abundant transcription factor
   complexes in human cells
• Adaptation of SILAC methodology and cell
   cycle block and release for large scale
   bioreactor-based human cell culture
• In-depth characterization of cellular changes in
   human protein complexes using state of the art
   mass spectrometry and biological validation.
          Examination of the protein complexes bound on gal operon promoter of
                           Thermoanaerobacter tengcongensis
                                              1,2         1, 2           3         1,2
                                  Zhong Qian , Fan Wei , Li Guo , Siqi Liu
 1                                                   2                                             3
     Beijing Genomics Institute, CAS, Beijing, China; Beijing Proteomics Institute, Beijing, China; Institute of
                                       Microbiology, CAS, Beijing, China

Introduction                                                     (TTE0779) in 2U have not been reported to be
T. tengcongensis is an anaerobic, thermophilic                   related with DNA binding as well as transcriptional
eubacterium. Its genomic and proteomic profiles                  regulation. Further experiments are being carried
were well investigated. In our laboratory, a novel               on to separate these complexes using
gal operon was discovered from this bacterium,                   immunoprecipitation approach for functional
which is composed of five genes, TTE1925,                        analysis.
TTE1926, TTE1927, TTE1928, and TTE1929, and
induced its gene expression by galactose. The T.                 Innovative aspects
tengcongensis gal operon is likely to have two                   • We used the EMSA-LC MS/MS to detect the
promoters, O1 and O2, however, there is lack of                     unknown regulation protein complexes in the
information how the regulatory proteins interact                    regulation mechanism investigation for the
with these DNA regions. Therefore, we initiated the                 thermophilic bacterium.
study to explore the proteins potentially bound to
the promoters.                                                   References
                                                                    (1) D. Stenger, et al, Mass spectrometric
Methods                                                                 identification of RNA binding proteins from
T. tengcongensis cells were lysated in the buffer                       dried EMSA gels; J. Proteome Res. 2004,
containing lysozyme on ice for 4h. Synthesized                          3(3): 662-664.
ssDNA probes were annealed, labeled with [γ- P]
and mixed with the lysates. After incubation for
15m, DNA-protein complexes were separated
through 8% native gel. According to the EMSA
results, the autoradiographic shifted bands were
obviously observed either in galactose or glucose
medium, implying some proteins did interact with
the probes. To detect the components in the
shifted DNA-protein complexes, we excised the
four shifted bands, termed as 1A, 2A, 1U and 2U,
and identified the protein complexes by LC MS/MS
after completely tryptic digestion (1). For data
analysis of LC MS/MS results, a stringent criterion
was employed, an identified protein confirmed
from 3 unique peptides at least.

Results                                                          Figure. EMSA analysis of the DNA-protein complexes.
Totally we have identified more than 30 proteins                 This figure indicates the obvious shifted bands, implying
for each complex. Since the excised bands might                  some proteins have an interaction with the probes
contain some proteins that were non-specifically                 compared to the controls (O1 and O2).
bound to DNA or the background contamination,
the redundant identifications overlapped among
these bands were removed in the list of DNA-
protein complexes. After the treatment, 3 proteins
were identified in complex 1A, 1 in complex 2A, 1
in complex 1U and 4 in complex 2U. In glucose
medium, 2 important proteins were identified, HPr
and PEP-protein kinase. These two proteins were
reported to be involved in a classical carbon
metabolism regulation pathway, carbon catabolite
repression (CCR). However, in galactose medium,
TTE1926 (transcriptional regulator) was identified
as dominant protein bound to 1A and 2A. TTE1926
is a component of T. tengcongensis gal operon
and is expected to possess a DNA binding
structure. ATP:corrinoid adenosyltransferase in
1A, acetate kinase and hypothetical protein
Background proteins in immunoprecipitations: what are they and how reduce them
                Patrice Waridel, Bastienne Jaccard, Alexandra Potts and Manfredo Quadroni
       Protein Analysis Facility, Center for Integrative Genomics, University of Lausanne, Switzerland

Immunoprecipitation (IP) of a protein of interest is
frequently used to identify potential interaction         Innovative aspects
partners in the cell. The ubiquitous presence of
variable levels of unspecific “background” proteins,      •   Analysis of immunoprecipitation background
however, introduces noise in the data and often               proteins on a collection of samples
results in both false positives and false negatives.      •   Identify critical steps and reagents
We have analyzed the composition of this protein          •   Improved protocol with lower background
background through a survey of a number of mock
IPs prepared in different labs. Also, in a parallel
study under controlled conditions we have tried to        References
determine the origins of the background and to
modify protocols to reduce it.                            None

Exact procedures for the IPs of “survey” samples
varied from lab to lab. All IPs were carried out from
human cell extracts, using antibody-based affinity
steps. Proteins bound to beads after IP were
eluted by boiling in SDS and fractionated on mini
SDS-PAGE gels. Molecular weight regions were
cut and in-gel digested. Peptide mixtures were
analysed by LC-MS/MS on a Thermo LTQ-
For the protocol optimization study, cell lysis and
IP were performed with a 1% NP-40, 150 mM
NaCl buffer and with ProteinA-Sepharose beads.
A human IgG fraction and an anti-PCNA
monoclonal antibody were used for the tests.
Several IP parameters were varied for
optimization. Analysis was performed by gel and
fluorescence staining as well as digestion and LC-
MS/MS as described above.

Comparative analysis of 10 Ab-based mock
(negative control) IPs prepared under different
conditions highlighted the recurrent recovery of          Figure 1. Sypro ruby stained gel of IPs performed on
several families of proteins. From these data, a set      HeLa cell extracts with 5 μg human IgG using our
of about 130 proteins was identified, which were          modified protocol vs. a standard one. Bands stronger in
recovered in at least 80% of all experiments and          the standard protocol are indicated by blue circles. The
are thus very likely nonspecific background.              lane on the right contains only IgGs (starting material).
We have then carried out tests under controlled
conditions using a commonly used IP protocol
which employs ProteinA-Sepharose beads, by
subtracting one component at the time. The results
indicate that most background proteins are
recovered because of interaction with the
Sepharose beads. Furthermore, aggregation and
precipitation phenomena occurring in the 1-2 hours
after lysate preparation are responsible for a large
part of the background. We show that treatment
with benzonase coupled with an additional
centrifugation step one hour after lysis and before
addition of the beads are simple and effective
ways to strongly reduce nonspecific background.

Immunoprecipitation and co-immunoprecipitation are classical methods used to isolate specific
proteins or protein complexes from biological samples. Methods use the antibody-antigen
reaction principle to identify a protein that reacts specifically with an antibody in a mixture of
proteins so that its quantity, physical characteristics or interaction partners can be determined.
Traditionally, sepharose and agarose slurries have been used, but more recently magnetic beads
have gained popularity due to shorter and simpler protocols. Magnetic Dynabeads are ideal for
immunoprecipitation. The rapid procedure permits the isolation of labile composites that might
otherwise dissociate during long incubation times (or be damaged by proteases), there is no upper
size limit for the complex to be isolated (ideal for complex pull-down) and the surface properties
give very low non-specific binding. With Dynabeads there is no fear of losing beads (as with
spun down resin) and you can scale down the procedure to reduce the consumption of expensive
antibodies. Here we show data of improved immunoprecipitation and co-immunoprecipitation
protocols, and demonstrate the usefulness of combining GFP-tagged fusion proteins with
Dynabeads to enable the combined visualization and pull-down of proteins.
Your captured primary antibody can be cross-linked to the Dynabeads® or used directly for
affinity purification of pure target protein, peptides, protein complexes or other antigens (fig. 2).
The benefits of magnetic protein separation are reaped in a wide variety of applications, as
documented in numerous publications. For details, protocols and references, please refer to:

Dynabeads® magnetic separation technology is also easy, rapid and efficient for small-scale
magnetic purification of antibodies. Within minutes, you can capture pure and highly
concentrated antibodies directly from samples such as saliva, ascites, serum and tissue culture or
hybridoma supernatants.

 have several limitations compared to magnetic beads which and more .
In this study we show data on improved protocols for immunoprecipitation using Dynabeads and
compare the performance with other technologies.

This can be agarose / sepharose beads (slurries), but more recently Protein A or -G conjugated
magnetic beads have gained popularity[citation needed] due to shorter, simpler protocols and less non-
specific binding
     Proteasome Diversity Analysed by Isotope Labelling and Protein Quantitation
                                       1                 2                        2             2
                     Reinout Raijmakers , Celia R. Berkers , Annemieke de Jong , Huib Ovaa ,
                                                        1                   1
                                   Shabaz Mohammed and Albert J.R. Heck
                   Biomolecular Mass Spectrometry and Proteomics Group, Utrecht University,
                               Sorbonnelaan 16, 3584 CA Utrecht, the Netherlands
        Division of Cellular Biochemistry, Tumor Biology, and Immunology, Netherlands Cancer Institute,
                            Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands

Introduction                                                 direction (2). In addition, we found also the PA28
The proteasome is the protein degradation                    cap proteins to be much more abundant in the
machinery of all eukaryotic organisms and                    spleen proteasome. As this particular cap structure
archaebacteria. It degrades protein to short                 is known to associate mainly with the immuno-
peptides of about seven to eight amino acids,                proteasome, this finding corresponds nicely with
which can be further processed to single amino               the upregulation of the immunoproteasome beta
acids to be reused by the cell in the synthesis of           subunits in that tissue.
new proteins. Inhibitors of the proteasome are
strong inducers of apoptosis and can be used as              Innovative aspects
anti-cancer therapeutics. To enhance therapeutic
applications targeting the proteasome, it is vital to         •    Quantitation of proteasome complexes from
understand in great the detail the diversity of                    cow tissue samples reveals tissue specific
proteasome complexes in various tissues.                           composition of the proteasome

Methods                                                      References
To analyze proteasome diversity, we first purified
20S proteasome core complexes from cow liver                  1.   Hsu, J. L., S. Y. Huang, N. H. Chow, and S. H.
and spleen tissue. Proteasome purification was                     Chen. 2003. Stable-isotope dimethyl labeling for
achieved by sequential ammoniumsulphate                            quantitative proteomics. Anal Chem 75:6843-52.2
precipitation, sucrose gradient centrifugation and            2.   Berkers, C. R., F. W. van Leeuwen, T. A.
                                                                   Groothuis, V. Peperzak, E. W. van Tilburg, J.
weak anion exchange chromatography. Following
                                                                   Borst, J. J. Neefjes, and H. Ovaa. 2007. Profiling
digestion, the two proteasome samples were then                    proteasome activity in tissue with fluorescent
labelled using an improved version of chemical                     probes. Mol Pharm 4:739-48
labelling of protein digests by dimethylation of
peptide N-termini and lysine side chains. This is
achieved with a mixture of formaldehyde and
cyanoborohydride, using either normal or
deuterated formaldehyde (1). Labelled samples
are analyzed by LC-MS/MS and quantitative
differences between the samples are determined
by comparing the intensity of the light and heavy
labelled peptides as determined by the mass

After purification and LC/MS analysis we were able           Figure 1. Quantitation of proteasome subunits in cow
to identify all 14 subunits (7 alpha and 7 beta              spleen and liver tissue. Proteasome subunits are sorted
subunits) of the core of the proteasome complex,             from more abundant in the liver proteasome (left side,
as well as the three alternative beta subunits which         purple area) to more prominent in the spleen
are only present in a specific subclass of the               proteasome (right side, pink area. See the legend for the
                                                             different classes of proteasome subunits included in the
proteasome, called the “immunoproteasome”. In                graph.
addition, we identified two proteins belonging to
the PA28 regulatory cap, which is one of the cap
structures that can regulate proteasome activity.
Differential quantitation of all of these proteins
between the proteasomes purified from cow liver
and spleen revealed distinct differences between
the complex in these tissues (Figure 1). One of the
catalytic beta subunits of the proteasome was
significantly more abundant in the liver, whereas
the three beta subunits of the immunoproteasome
were more present in the spleen sample,
corresponding to previous indications in that
                    Protein complex studies of mouse brain using a new proteomics platform

                                              1                  1                              1, 2
                   Ram al i n gam R aj k um ar , Chen-Chung Liao , a nd Yeou-Guang Tsay
                1                                 2
                Proteomics Research Center and Institute of Biochemistry and Molecular Biology,
                                 National Yang-Ming University, Taipei, Taiwan

                                                          extract. Two major parameters were recorded with
                                                          each identified protein, the molecular masses of its
Protein complex formation is one of the major             primary and quaternary structures. These items of
events that regulate the protein functions (Collins       information were instrumental in establishment of
et al., 2007, Wang and Huang, 2008). Therefore,           the        protein    complex         map.        Functional
comprehensive         mapping    of   protein-protein     categorization of proteins in GFC fractions shows
interactions within the proteome will inevitably          that the proteins for certain functions have a
become a challenging mandate in proteome                  special tendency to become macromolecular
research (Andrew et al., 1999).       In the present      structures in the brain tissue.                Thus, protein
investigation, we report the application of a new         complex mapping analysis may help us to
proteomics platform to the studies of protein             understand the significance of protein complex
complexes in the entire mouse brain proteome.             assembly/disassembly in the overall function of
                                                          cells or tissues.
Seven weeks old C57/B16 mice were maintained
                                                          1. Andrew J.Link et al, Direct analysis of protein
under standard breeding condition.       The brains
                                                                complexes using mass spectrometry; Nature.
were removed after decapitation and immediately
                                                                1999, 17; 676-682.
frozen on dry ice, minced and homogenized in ice-
                                                          2. Collins, S. R., et al., Toward a comprehensive
cold extraction buffer.       The homogenate was
                                                                atlas    of    the   physical          interactome     of
centrifuged and the supernatant was fractionated
                                                                Saccharomyces        cerevisiae;          Mol.       Cell.
with gel filtration chromatography (GFC) and the
                                                                Proteomics. 2007, 6; 439-450.
proteins from each GFC fraction were resolved
                                                          3. Xiaorong Wang and Lan Huang Identifying
with SDS-PAGE. All of the protein bands in the
                                                                dynamic interactors of protein complexes by
gels were digested with trypsin and then analyzed
                                                                quantitative mass spectrometry; Mol. Cell.
with   liquid        chromatography-tandem        mass
                                                                Proteomics. 2008, 7; 46-57.
spectrometry (LC-MS/MS).          Turbo SEQUEST
algorithm and our in-house computer software
were used to interpret the LC-MS/MS data, and to
construct the protein complex map.


With this method, we identified more than 1,500
different proteins from the mouse brain sample.
Based on the protein staining intensities, we
estimated that ~ 60% of proteins were present as
multimeric protein structures in our brain tissue
LC-MS Analysis to Identify Ryanodine Receptor 1 (RyR1) Protein-Protein Interactions
         1                  1                   4                 3                       1,*
Tim Ryan , Parveen Sharma , Alex Ignatchenko , David H. MacLennan , Anthony O. Gramolini and Thomas
   1                           2                     3
    Departments of Physiology, Medical Biophysics, Banting and Best Department of Medical Research,
       University of Toronto, Ontario Cancer Institute, Division of Cancer Genomics and Proteomics
                               Toronto, ON, Canada * Equal senior authors

Introduction                                          immunoprecipitation assays. This protein-protein
Excitation-contraction coupling is defined as the     interaction analysis provides a framework for a
process linking action potential to contraction in    more detailed analysis of RyR1 function.
striated muscle. This process depends on a large
macromolecular protein complex, the central
element of which is the Ca release channel,           Innovative aspects
ryanodine receptor 1 (RyR1) (1). It has been          • To date, there has been no large scale protein-
demonstrated that protein binding partners               protein interaction analysis of RyR1.
regulate RyR1 via its cytosolic domain and            • We have identified novel RyR1 interacting
mutations in RyR1 and its interacting proteins           proteins.
result in disease. Here, we describe the
purification and LC-MS analysis of affinity-tagged
RyR1 protein complexes with the overall aim of        References
understanding calcium ion movement at a               (1)    Du, G.G., et al., Topology of the Ca2+
molecular level.                                             release channel of skeletal muscle
                                                             sarcoplasmic reticulum (RyR1). Proc Natl
                                                             Acad Sci U S A, 2002. 99(26): p. 16725-
Methods                                                      30.
Full length RyR1 rabbit cDNA and a truncated,
cytosolic RyR1 construct were affinity-tagged with
6x-histidine and streptavidin binding peptide
(SBP). These constructs were introduced into
HEK293 cells and C2C12 mouse skeletal
myocytes enabling column purification of RyR1
protein     complexes      under     non-denaturing
conditions. The purified tagged proteins were
subjected to immunoblotting experiments to verify
stable expression and purification of clones. The
purified complexes were analyzed using 1-
dimensional gels and by gel-free LC-MS. A filtering
algorithm was applied to all putative results to
obtain a measure of the statistical reliability
(confidence score) for each candidate identified      Figure 1. Histidine-tagged RyR1 purified from
and only proteins with confidence value ≥ 99% and     transfected C2C12 muscle cells using Ni-NTA
identified with ≥ 2 peptides were accepted.           chromatography. Eluted protein complexes were
                                                      immunoblotted and probed with an anti-RyR1 antibody.

Here, we present data using tandem affinity
purification tagged (streptavidin-binding, and HIS)
proteins transfected into HEK293 and C2C12
muscle cells, followed by gel-free LC-MS to
identify protein-protein interactions in skeletal
muscle. Column purification of tagged proteins
from transfected HEK and C2C12 shows stable
expression and purification of clones by
immunoblotting experiments. Figure 1 shows
ryanodine receptor 1 (RyR1) from C2C12
myocytes, purified on a Ni-NTA column under non-
denaturing conditions. LC-MS analysis of the
purified protein complex showed that we have
identified 5 novel RyR1 interacting proteins and we
are currently validating these interactions using
conventional            co-expression           and
 Quantitative 20S proteasome analysis from 5FU-induced apoptotic Jurkat T cells by
                     SILAC, SDS-PAGE, and 2DE based analysis

Frank Schmidt, Alexander Kloos, Martin von Bergen, Burkhardt Dahlmann, Peter Jungblut, and Bernd Thiede
                             Department of Proteomics, Helmholtz Institute,
                                Permoserstr. 15, 04318 Leipzig, Germany

Introduction                                             Innovative aspects
DNA damage and subsequent induction of                   • Combination of SILAC and 2-DE
apoptosis is possibly the primary cytotoxic
mechanism induced by DNA-binding anticancer
drugs. Apoptosis is a form of programmed cell
death and can be triggered in a cell through either
the extrinsic or the intrinsic pathway. In this study,
quantitative 20S proteasome analysis of 5-fluoro-
uracil-induced apoptosis of Jurkat T cell lysates
was performed in order to identify differentially
regulated proteasomal subunits.

Proteasomal proteins were labelled in cell culture
with stable isotopes of arginines and lysines,
fractionated by SDS-PAGE and 2-DE gels and
further analyzed by MALDI-TOF/TOF-MS and LC-
orbitrap MS. In a first step, cell lysate was
subjected to a gel filtration on a Sepharose 6B
column, in order purify the proteasome.
Afterwards, the proteasome fraction was subjected
to 1D PAGE and bands were digested by trypsin
and identified as well as quantified by LC-orbitrap
MS. In addition, a more detailed analysis was
performed by 2-DE gel analysis where the protein
species were cut out and digested with trypsin.
PMF was achieved by MALDI-TOF/TOF-MS in
order to identify and quantify proteasome subunits.

In total, 26 bands were digested with trypsin and
further analyzed by LC-orbitrap MS. Beside non-
related proteasomal proteins, all α- and β-subunits
were identified within at least one band excluding
the immunosubunits β-1i, β-2i, and β-5i. With
exception of α-7 (also known as C8, α7_sc,
PSA3), all subunits showed an equally ratio of 1:1
using the average ratios belong to all peptides of a
subunit. In case of α-4, α-7, β-4 and β-7 subunits
were detected in two different bands with different
molecular weights. Here, 20S proteasome species
of α-4, β-4, and β-6 showed an equally ratio of 1:1
in contrast to the species of α-7, where the high
molecular one was clearly down regulated during
apoptosis (0.64) compared to the low molecular
one (1.06). In order to explain the different ratios
between the α-7 subunits, a 2-DE gel was
performed. The ratio of low molecular α-7 subunit
showed the same down-regulation as in the 1D
PAGE. The detailed analysis of the MS spectra
yields to a specific methylation only present in the
regulated α-7 species.
                   Proteomic Analysis of Membrane Protein Complexes:
                 Lessons from Affinity-Purified Ion Channels and Receptors
                                    §                    §#              #                    #
                        Uwe Schulte , Catrin S. Müller , Wolfgang Bildl and Bernd Fakler
                      ) Logopharm GmbH, Schlossstr. 14, 79232 March-Buchheim, Germany
     ) Institute of Physiology, University of Freiburg, Hermann-Herder-Straße 7, 79104 Freiburg, Germany

Introduction                                                  examples provide a general basis for studying
Membrane proteins represent a major challenge in              protein networks at the plasma membrane and
proteomics, especially in protein-protein interaction         may serve to establish increased quality standards
studies. Currently, affinity purification coupled to          for affinity-based proteomics.
LC-MS/MS sequencing is regarded as the method
of choice. Although successful in some cases, this            Innovative aspects
approach has multiple pitfalls and lacks general              • Large-scale identification of signaling
quality standards. The most critical factors, namely             supercomplexes at the plasma membrane
protein solubilization, antibody-related effects and          • Native-source based proteomics with highest
unbiased MS data evaluation are addressed based                  sensitivity and reliability
on more than 1000 affinity purifications of ion               • Characterization of previously unapproachable
channels and receptors performed and analyzed                    or functionally unassigned proteins
by our group.
Methods                                                       (1) Berkefeld H, Sailer CA, Bildl W, Rohde V, Thumfart
Membrane       fractions     were    prepared     by              JO, Eble S, Klugbauer N, Reisinger E,
conventional methods from native tissues and                      Bischofberger J, Oliver D, Knaus HG, Schulte U,
subjected to solubilization assays. A selection of                Fakler B: BKCa-Cav channel complexes mediate
                                                                                            2+            +
standardized non-denaturing detergent buffers                     rapid and localized Ca -activated K signaling
(ComplexioLytes) was tested for efficiency and                    (2006). Science 314: 615-20
complex integrity. Solubilized protein complexes              (2) Schulte U: Protein-protein interactions and subunit
                                                                  composition of ion channels (2008). CNS Neurol
were affinity-purified using immobilized antibodies.
                                                                  Disord Drug Targets 7: in press
Pre-immune IgGs or target knockout preparations
served as negative controls. Isolated proteins were
resolved by SDS-PAGE, silver-stained and in-gel
digested with trypsin. Extracted peptide mixtures
were finally resolved on an UltiMate 3000 nano-
HPLC reverse-phase system directly coupled to a
linear ion trap ICR hybrid mass spectrometer
(LTQ-FT). Proteins were identified using Mascot;
LC-MS/MS data was quantitatively evaluated with
proprietary software.

Comprehensive identification of ion channel and
receptor-associated        proteins     requires     a
combination of technical optimization, appropriate
controls and quantitative evaluation. Optimized
solubilization with ComplexioLytes shows more
than 80% efficiency while preserving high
molecular weight target complexes of 0.5-2.5 MDa.
Antibody quality is of critical importance:
unpredictable cross reactivities, biased selection of
target complex populations and competition with
associated proteins are common sources for
artefacts. These problems can by resolved by
using multiple antibodies, different purification
conditions and target knockouts as stringent
controls. Finally, relative quantification by LC-
MS/MS is required for validation of protein
specificity. Application of this technology has led to        Figure 1. Proteomic analysis of BK channel complexes
the identification of novel and functionally relevant         as an example. Optimization of critical steps and careful
interaction partners of a number of ion channels              quality controls resulted in high BKα sequence coverage
(1, 2) and G-protein coupled receptors. These                 and identification of several novel interaction partners.
                            Analysis of subunit composition of the 20S proteasome of
                                            Plasmodium falciparum
                  1,2                             1              2              2                  2                         1           1
      N. Sessler , W. Schütz , R. Fendel , B. Mordmüller , P. G. Kremsner , T. Lamkemeyer , A. Nordheim
             Proteome Center Tübingen, Interfaculty Institute of Cell Biology, University of Tübingen,
                              Auf der Morgenstelle 15, 72076 Tübingen, Germany
         Department of Parasitology, University of Tübingen, Wilhelmstraße 27, 72074 Tübingen, Germany

Introduction                                                                        This analysis revealed at least three variants of the
Malaria due to Plasmodium falciparum causes                                         proteasome complexes at ~700 kDa (a, b, c; see
more than 1 million deaths per year. Symptoms                                       fig. 1B).
occur during the erythrocytic phase of replication,                                 We excised three areas in the gel containing the
characterized by four stages: merozoites, ring-                                     proteasome complexes from schizont stage
stage trophozoites (rings), trophozoites and                                        lysates as determined by Western blotting and
schizonts. The 20S proteasome of P. falciparum is                                   analysed the protein composition by a classical 2D
essential for differentiation and replication, and is a                             PAGE (see fig. 2) All gels of the three excised gel
promising therapeutic target.                                                       slices displayed a comparable spot pattern in the
In order to gain a more detailed knowledge about                                    expected mass range of 20 to 35kDa. We
structure, function and modifications of the                                        observed differences in spot intensity between the
proteasome during the life cycle of this parasite we                                2D gels which correlates with the decreasing
analysed the proteasome by multi-dimensional                                        intensities of the proteasome complexes seen in
electrophoresis and mass spectrometry.                                              BN-PAGE (not shown). We analysed the
                                                                                    corresponding proteins of complex a (see fig. 2) by
Methods                                                                             LC-MS/MS and identified 13 proteasome subunits.
Parasites (P. falciparum strain 3D7A) were
cultured and synchronised using standard                                                    pH 3               2. dim: IEF       pH 10
methods. At a parasitaemia of approx. 5%, cells
of the respective stage were harvested after                                         220
removal of contaminating erythrocytes.
                               [2]                                                   116

                                                                                                                                         3. dim: SDS-PAGE
For three-dimensional analysis two sets of total
protein extracts were separated by blue-native                                         66
                                               [3]                                   55,6
polyacrylamide gel electrophoresis (BN-PAGE) .
After electrophoresis one set was analysed by
Western blot to detect the proteasome complexes                                      36,5                              7
using an antibody against the 20S proteasome. By                                                               4     8
                                                                                      29               1           5
matching the blot to the second set on the non-                                                                       9 12
                                                                                                           2   3 6 10 13
blotted gel the corresponding areas to be excised
were determined. The proteins were eluted and                                         20                               11
subsequently analysed by classical 2-dimensional
(2D) PAGE followed by nano-liquid chromato-
                                                                                    Fig. 2: Classical 2D-gel of the eluted proteins from the
graphy and tandem mass spectrometry (LC-                                            proteasome complex a of schizonts (see fig. 1A, lanes 3 +
MS/MS) analysis.                                                                    1B). The 13 marked spots were analysed by LC-MS/MS.

To date we have analysed the 20S proteasome in                                      Similar analyses of the other erythrocytic phases
three of the four blood stages of P. falciparum                                     of replication are ongoing.
(rings, trophozoites, and schizonts).
                                                                                    Innovative aspects
      A   1   2         3   B                         Fig. 1: Determination         • Separation of total protein extracts of
                                                      of the areas in the gel          P. falciparum     and     detection  of   three
                                                      containing proteasome            proteasome complexes by BN-PAGE.
720                             c                     complexes.                    • Direct identification of the proteasome subunits
                                b                     A: Western blot of BN-
                                a                                                      using three-dimensional electrophoresis.
                                                      PAGE of proteins of
                                                      rings (1), trophozoites
480                                                                                 References
                                                      (2) and schizonts (3).
                                1. dim: BN-PAGE

                                                      B: BN-gel of schizonts        [1] Mordmüller B. et al., Molecular & Biochemical
                                                      proteins, showing                 Parasitology (2006), 148, 79–85.
242                                                   proteasome                    [2] Werhahn W. and Braun H.-P., Electrophoresis
                                                      complexes marked in               (2002), 23, 640-646.
                                                      red (a, b, c).                [3] Schägger H., Cramer W.A., von Jagow G.,
                                                                                        Analytical Biochemistry (1994), 217, 220-230.
     Comparative mass spectrometry of the endogenously expressed nuclear and
                             cytoplasmic exosome

                        Silvia A. Synowsky, Reinout Raijmakers and Albert J. R. Heck
                  Biomolecular Mass Spectrometry and Proteomics group, Utrecht University,
                             Sorbonnelaan 16, 3584 CA Utrecht, The Netherlands

Introduction Proteins are organized in large             RNA binding domain of Csl4, is significantly
protein complexes that form an extensive network         upregulated in the nuclear exosome.
in the cell [1]. Since these protein assemblies are
essential for most cellular functions, it is of high     Innovative aspects
importance to determine their constituents,              • Combination of TAP with macromolecular
topology and subtle differences especially between          mass      spectrometry      and     quantitative
related protein complexes. Here we employ a                 proteomics reveal structural features and
multiplexed      MS     approach     to  investigate        subtle changes in endogenously expressed
endogenously expressed protein complexes by                 heterogeneous protein complexes
native MS and quantitative proteomics. Our               • Macromolecular MS/(MS) identifies subunit
strategy is applied on the yeast nuclear and                composition, the most common stable core of
cytoplasmic exosome to identify subcomplexes                exosomes       and    gas-phase    dissociation
and     to    quantitate    differences   in    their       patterns
phosphorylation status.                                  • Quantitative      proteomics    show     relative
                                                            differences in the phosphorylation status of the
Methods The exosomes were purified using a                  nuclear and cytoplasmic exosome
nuclear and cytoplasmic specific TAP tagged
protein (Rrp6 and Ski7) [1]. The intact exosomes         References
were first analyzed in the MS and MS/MS mode             [1] Gavin AC, et al., Functional organization of
using a modified QToF1 instrument [2]. Typical                the yeast proteome by systematic analysis of
settings were a capillary voltage of 1400 V and a             protein complexes. Nature. 2002; 10(415):
sample cone voltage of 150 V. In the MS/MS mode               p.141-7.
the collision voltage was raised stepwise from 50        [2] van den Heuvel RH, et al., Improving the
to 175 V to observe dissociation patterns of the              performance of a quadrupole time-of-flight
selected precursor. Quantitative information about            instrument    for   macromolecular      mass
the nuclear and cytoplasmic exosome proteins and              spectrometry. Anal Chem. 2006; 78(21):
their phosphorylation status were obtained by                 p.7473-83.
tryptic in-gel digestion followed by dimethyl            [3] Synowsky SA, et al., Probing genuine strong
labeling with heavy and light formaldehyde                    interactions      and       post-translational
respectively. The peptides were mixed and                     modifications in the heterogeneous yeast
analyzed using an LTQ-FT. Prior to MS analysis                exosome protein complex. Mol Cell
the phosphopeptides were enriched using TiO2.                 Proteomics. 2006; 5(9): p. 1581-92

Results The nuclear and cytoplasmic exosome
complexes exhibit the same core complex in native
MS. The mass corresponds to the hexameric ring
plus Dis3, Rrp4 and Rrp40. Therefore the 40+
charge state of the nuclear variant has been
subjected to gas-phase dissociation in MS/MS.
Surprisingly at low collisional voltage Rrp40 is
ejected with few charges from the precursor
indicating a native fold upon dissociation. At further
increasing collision voltage a second dissociation
pathway emerges in which partially unfolded Rrp4
dissociates from the precursor. Quantitative
proteomics identifies all common proteins between
the nuclear and cytoplasmic exosome to be
unchanged. Differences were observed in strictly
nuclear proteins (Rrp6, importin / , Lrp1 and
Ynr024w) and the cytoplasmic protein Ski7. As
shown previously several exosomal proteins
contain at least one phosphorylationsite [3]. Here
we could relatively quantify several of these
phosphorylationsites between the nuclear and
cytoplasmic variant. Most importantly we could
establish that the P-site at serine 94, located in the
Identification of intracellular molecules interacting with the stress-induced, psoriasis
                     susceptibility-related non-coding RNA, PRINS
                                  1            2            3                            1,4
               Krisztina Szegedi , Mária Antal , István Németh , Zsuzsanna Bata-Csörgő ,
                                           1,4               1,4          4
                              Attila Dobozy , Lajos Kemény , Márta Széll
 Department of Dermatology and Allergology, University of Szeged, Korányi fasor 6, 6720 Szeged, Hungary

Introduction                                           when cells are released from cell quiescence,
Previously we have identified a non-coding RNA         while their protein level is highest in differentiating
gene, PRINS (Psoriasis susceptibility-related RNA      cells.
gene Induced by Stress) that is expressed at an        We hypothesize that the non-coding PRINS RNA
elevated level in psoriatic uninvolved epidermis       may be a structural element of a molecular
compared to healthy epidermis and in cells under       complex playing a role in stress-induced cellular
various stress conditions. Those expression data       processes and the abnormal functioning of this
suggests a role for PRINS in psoriasis                 complex may contribute to the pathogenesis of
susceptibility and in cellular stress response (1).    psoriasis.
Our aim was to access the cellular function of
PRINS by identifying intracellular molecules
interacting with this stress-induced non-coding
RNA.                                                   Innovative aspects
                                                       • The non-coding RNA, PRINS physically
                                                          interacts     with       chaperon      molecules
Methods                                                   nucleophosmin and GRP94 in HaCaT
We performed in vitro experiments with a                  keratinocytes. PRINS RNA may be a structural
ribonucleoprotein purification kit. As a target           element of a molecular complex including
template, we used a transcript containing the             nucleophosmin or/and GRP94
19mer sequence that has previously effectively         • The expression of both nucleophosmin and
knocked down the expression of PRINS both in              GRP94 depends on proliferation/differentiation
HeLa and HaCaT cells. We hypothesized that this           state of syncronised HaCaT cells, thus the
region may play an active role in binding other           posttranscriptional regulation might play an
ribonucleic acids and/or proteins. With the help of       important role in their protein expression
Matrix-Assisted Laser-Desorption Ionization Time-      • Here we provide the first data that
of-Flight (MALDI-TOF) method we identified two            nucleophosmin might serve as a marker of
proteins in HaCaT cell lysates that physically            higher proliferation rate in the basal layers of
interact with PRINS RNA. To further characterize          psoriatic involved epidermis
the      identified    proteins,      we      used
immunohistochemistry, RT-PCR and Western blot          References
techniques to detect their mRNA and protein               (1) E.Sonkoly et al, Identification and
expression in HaCaT cell cultures and in healthy              characterization of a novel, psoriasis
and psoriatic skin samples and other various                  susceptibility-related noncoding RNA
tissue samples.                                               gene, PRINS; J. Biol Chem 2005 Jun 24;
                                                          (2) V. Palaniswamy et al, Nucleophosmin is
Results                                                       selectively deposited on mRNA during
We identified two proteins in HaCaT cell lysates              polyadenylation.Nat Struct Mol Biol. 2006
that physically interact with PRINS RNA. One of               May;13(5):429-35
the identified proteins, nucleophosmin is a               (3) P. Fortugno et al, Regulation of survivin
ubiquitously expressed nucleolar phosphoprotein               function by Hsp90; Proc Natl Acad Sci U S
which shuttles continuously between the nucleus               A. 2003 Nov 25;100(24):13791-6
and the cytoplasm (2). The other PRINS RNA
binding protein, GRP94 is a molecular chaperone
heat shock protein (3). Immunohistochemical
experiments revealed that there is no difference in
the expression of GRP94 protein when compared
among healthy, psoriatic uninvolved and involved
skin    samples,     while   the  expression     of
nucleophosmin is significantly elevated in the
dividing cells of the basal layer. In vitro
experiments performed on synchronised HaCaT
cells revealed that the expression of both
nucleophosmin and GRP94 mRNA is highest
                         A Quantitative Proteomics Approach to Reveal
                         Dynamics in Transcription Factor Complexes

     H.Th.M. Timmers*, F. Mousson*, P. de Graaf*, A. Kolkman#, W.W.M. Pijnappel*, B. Gevers^, A.B.
                                      Houtsmuller^ and A.J. Heck#
 *Department of Physiological Chemistry, UMC-U, Universiteitsweg 100, 3584 CG Utrecht; # Department of
 Biomolecular Mass Spectrometry, Utrecht University, Sorbonnelaan 16, 3584 CA Utrecht, The Netherlands
   and ^Department of Pathology, Erasmus MC, Dr. Molewaterplein 50, 3015 GE Rotterdam, Netherlands.

The present methods to study the dynamics of           spectrometric analysis. We used a standard
multi-protein complexes are limited. We have           combination of affinity purification and stable
developed a quantitative proteomics strategy to        isotope labeling (SILAC) to discriminate specific
discriminate stable from dynamic subunits in           from non-specific TBP interactors. By this
protein complexes. This strategy is generally          approach we would classify the known interactor,
applicable and was used to examine the dynamics        BTAF1, as non-specific. Thus, we compared
of TBP transcription complexes.                        different affinity purification protocols and showed
                                                       that a combination of different protocols can reveal
Transcription in eukaryotic cells requires the         information on the specificity and dynamics of
continuous assembly and disassembly of the             protein complexes. We find that the dynamic
transcription machinery on gene promoters. The         behaviour of BTAF1 is inhibited in mitosis and that
association of the TFIID complex to gene               this property is not shared by the TFIID, SL1 and
promoters is the first step in assembly of RNA         TFIIIB complexes.
polymerase II transcription complex. The TATA-
binding protein (TBP) can stably interact with TBP-    Independent support for our quantitative
associated factors (TAFs) to form (at least) four      proteomics strategy comes from in vivo cell
different TBP-TAF complexes (SL1, TFIID, B-            imaging experiments. To measure TBP-complex
TFIID, and TFIIIB) in the eukaryotic nucleus. While    mobility in living cells we performed FRAP
comparison of TBP-binding surfaces for TAF             (fluorescence-recovery-after-photobleaching)
subunits of these complexes explains their             experiments using GFP-TBP expressing human
mutually-exclusive binding to TBP, little is known     cells. In this, the majority of TBP (~70%) displays a
about stability of TBP-complexes and exchange          mobile behavior, which is not an intrinsic
properties of TBP and TAFs. To address this we         biochemical property of TBP. Interestingly, the
applied a SILAC-based proteomics strategy to           mobile fraction of TBP molecules is decreased
analyze subunit exchange in human cell extracts.       after knockdown of BTAF1 expression. In contrast,
We find that the BTAF1/hMot1 subunit of B-TFIID        overexpressed BTAF1 increased the mobility of
exchanges readily, which contrasts with the            TBP molecules in vivo.
behavior of other TAFs.
                                                       Together we provide a generic quantitative
These results were obtained by first generating a      proteomics strategy to investigate regulation of the
human cell line expressing a tagged version of         dynamics in multi-protein complexes. This
TBP. This allows affinity purification of the four     revealed that the BTAF1 protein regulates the
TBP-TAF complexes as shown by mass                     dynamics of the transcription factor TBP.
        Proteomic Analysis of Ubxd1: Identification of Novel Interaction Partners
                                      1                   2             2                   1
                    Julian Vasilescu , Daniel Zweitzig , Dale S. Haines , and Daniel Figeys
                  Ottawa Institute of Systems Biology, University of Ottawa, Ottawa, CANADA
                   Fels Institute for Cancer Research, Temple University, Philadelphia, USA

Introduction                                                  Innovative aspects
Ubxd1 is an adapter protein that was recently                 • First interaction mapping study of Ubxd1
identified as a p53 regulator that plays a role in cell       • Novel Ubxd1 interaction partners identified
cycle arrest and apoptosis (1). For this study, we            • In-house software application created to
performed a series of targeted protein interaction               analyze multiple interaction datasets
mapping experiments (2) in a human cancer cell
line to identify novel interaction partners of Ubxd1.         References
Co-immunoprecipitation and immunofluorescence                    (1) Zweitzig et al, The ubiquitin selective
experiments were then performed to validate a                        chaperone p97 and adapter UBXD1
number of these interactions.                                        promote HAUSP-mediated MDM2
                                                                     deubiquitination and suppression of p53
Methods                                                              function; (submitted) 2008.
Human H1299 cells were transfected with                          (2) Vasilescu et al, Mapping protein-protein
constructs encoding Ubxd1 with a carboxy-                            interactions by mass spectrometry; Curr
terminal Flag- or HA- epitope tag. Cells were                        Opin Biotechnol; 2006.
harvested and lysates were subjected to
immunopurification using anti-Flag or -HA
antibodies. Purified samples were separated by
SDS-PAGE and entire gel lanes, including empty
vector controls, were subjected to an in-gel
digestion step. The resulting peptides were then
analyzed by liquid chromatography tandem mass
spectrometry (LC-MS/MS) using a LTQ linear ion
trap mass spectrometer. After database searching,
an in-house software application was used to
identify proteins that were unique to the Ubxd1-
containing samples. A number of these proteins
were selected for validation experiments using
standard         co-immunoprecipitation      and
immunofluorescence techniques.

Using a targeted interaction mapping approach,
numerous potential interaction partners of Ubxd1
were identified, including components of an
inhibitory complex called the mitotic checkpoint
complex. Although the Flag- and HA- interaction
datasets displayed minimal overlap among the
majority of proteins identified, components of the
mitotic checkpoint complex were consistently
observed. Co-immunoprecipitation experiments
confirmed that these proteins were specifically
purified only in the presence of Ubxd1. Co-
localization experiments also confirmed that these
proteins were present in the same physical
location within the cell as Ubxd1. Finally, results
from several Ubxd1 knock-down experiments
demonstrated an increase in polyploidy, which is a
phenomenon known to be associated with
perturbation of the mitotic checkpoint complex.
         Quantification of Protein Complex Components in Interaction Networks
                                1,2              1,2 ,                1,2,3,4                      1,2
               Alexander Wepf , Timo Glatter , Ruedi Aebersold              and Mat thias Gstaiger
 Institute of Molecular Syste ms Biology, ETH Zurich, Zurich, Switzerland. 2Competence Center for Systems
Physiology and Metabolic Diseases, ETH Zurich, 8093 Zurich, Switzerland.          Faculty of Science, University of
                  Zurich, Zurich, Switzerland, 4Institute for Systems Biology, Seattle, WA, USA

Introducti on
Protein complexes represent major functional units            Innovative aspects
for the control of biological processes.                      •   absolute quantification of protein complex
Understanding on how proteins are organized into                  components in large scale AP -MS studies
protein complexes at a global scale is becoming               •   no need for a large set of heavy peptides
increasingly important for a systems level
conception of cellular processes. Until today,
protein interaction networks are a purely qualitative
representation of com plex interactions, eve n
though affinity -purification coupled to mass
spectrometry (AP -MS), the method of choice for
network analysis, could in principle be applied in a
quantitative fashion.

Here a method is presented w hich combines the
advantages       of      isotope -labelled pept idesfor
absolute quantificationwith lab le-free quantitation.
First, a single heavy peptide of an engineered
amino acid sequence in the affinity tag is used for
absolutequantification of the bait protein . Second,
relative quantification on the intensities of the M S1
signal is performed: Thereby, t he peptides of the
bait protein serve as sta ndard peptides to quantify
the absolute amoun ts of the same protein in the
other affinity -purificat ions.

Particularly highly connected network s like the
human protein phosphatase 2A               network are
suitable for this combined quantification method. In
a large number of interactions the absolute amount
of bait as well as the absolute amount of prey
present after an affinity purification can be
quantified concurrently without neither the need for
knowing or predicting proteotypic peptides nor the
need of an expensive se t of heavy peptides.
Quantitative network information generated in such
a manner allows network model ling and real
complex prediction with much higher accuracy.
Protein associations      can be classified into
abundant and stable associations               versus
interactions present in sub -stoichiometric amounts
or of transient stability. Furthermore complex
composition changes can be monitored and
changes in seco ndary modifications can be
normalized to the pro tein abundance.
       Identification of putative complex I assembly factors by Blue Native Liquid
                                  Chromatography MS/MS

   J.C.T. Wessels, R.O. Vogel, L.P. van den Heuvel, R.J. Rodenburg, L.G. Nijtmans, J.A.M.Smeitink, M.H.
  Nijmegen Center for Mitochondrial Disorders, Laboratory of Pediatrics and Neurology, Radboud University
           Medical Center Nijmegen, Geert Grooteplein 10, 6500HB Nijmegen,The Netherlands

Introduction. Analysis of protein (sub)complexes         proteins identified by this approach is C6ORF66
by Blue Native (BN) electrophoresis produces             which was recently published as an assembly
invaluable information about subunit composition         factor for complex I (1). This protein showed clear
of complexes and assembly intermediates. The             comigration with CI assembly intermediates on
analysis of co-migrating proteins is often               both gradients (figure 1) which was validated by
performed by antibody detection or mass                  westernblot results using NDUFS3 and C6ORF66
spectrometric (MS) identification in a single or two     antibodies.
dimensional separation. Unfortunately, these
approaches are limited by the availability of            Innovative aspects.
antibodies or separation capability for MS analysis.        • First application of LC MS protein
Given these limitations we hypothesized to use                  correlation profiling to blue native
Liquid Chromatography (LC) MS/MS as a second                    electrophoresis.
dimension separation, identification and
quantitation method.                                         •   Most extensive analysis of CI assembly
                                                                 intermediates to date adding new insights
 Methods. To analyze the mitochondrial                           into their exact composition.
complexes we separated proteins on Blue Native
gels using 4-12% and 5-15% acrylamide gradients.             •   Identification of multiple putative CI
Each gel lane was cut into 24 slices and analyzed                assembly factors
in duplo by LC MS/MS. Protein identifications were
performed by database searching. Quantitative            References:
data were extracted as EMPAI, extracted ion                 1. Saada et al, C6ORF66 is an assembly
chromatograms (XIC), and by accurate mass &                     factor of mitochondrial complex I; Am. J.
retention time (AMT) approach to evaluate their                 Hum. Genet. 82, 32-38
applicability to analyze protein migration patterns.
Identification of putative complex I assembly
factors was achieved by searching the dataset for
proteins that comigrate with complex I assembly
intermediates via template matching (Pearson
correlation values). Migration patterns were
validated by 2D BN SDS-PAGE immunodetection.

Results. Evaluation of quantitation methods
showed that all approaches appear applicable to
study migration patterns on BN gels. Based on
close evaluation of the data we decided to
primarily use XIC values and only perform AMT for
a relatively low number of proteins of interest.
Analysis of complexes from the oxidative
phosphorylation system showed focusing of
respective subunits at the expected migration
distances. Close evaluation of complex I (CI)
subunits revealed comigration of NDUFS subunits
2,3,7 and 8 in well separated subcomplexes of
smaller size than holocomplex I. Alignment of the
migration patterns of NDUFS2 and 3 with
westernblot results identified these subcomplexes
as known assembly intermediates of CI. To find
novel CI assembly factors we used the CI
assembly intermediates pattern to screen the
                                                         Figure 1. Normalized migration patterns as determined
datasets for comigrating proteins. Using an              by LC MS of C6ORF66 (yellow) and average template
empirically determined Pearson correlation score         pattern (blue) for 4-12% and 5-15% acrylamide blue
of 0.7 we were able to specifically look for             native gels. Assembly intermediates are respectively in
candidate assembly factors. One of the candidate         slices 8-24 and 6-24 for the 4-12% and 5-15% gradients.
 Contrasting Proteome Biology and Functional Heterogeneity of the 20S Proteasome


                                GW. Young. AV. Gomes, Y. Wang, P. Ping.
                Department of Medicine and Physiology, University of California, Los Angeles.
                             Los Angeles, CA. 90024 United States of America

Introduction: The 20S proteasome complexes are            Results: Our results revealed an organ specific
the major contributors to the intracellular protein       molecular organization of the 20S proteasomes
degradation machinery in mammalian systems.               with distinguished patterns of post-translational
Systematic administration of proteasome inhibitors        modifications as well as unique complex assembly
utilized to combat diseases (e.g. cancer) has             characteristics      and     associating       proteins.
shown targeted actions as well as adversary               Furthermore, the proteome diversities are
effects. The latter was attributed to a lack of           concomitant with a functional heterogeneity of the
understanding pertaining to organ specific                proteolytic patterns exhibited by these two organs.
responses to these inhibitors and the potential           In particular, the liver and heart displayed distinct
diversity of proteome biology of these complexes          activity profiles to two proteasome inhibitors,
in different tissues.                                     epoxomicin and Z-Pro-Nle-Asp-H. Finally, heart
                                                          and liver demonstrated contrasting regulatory
Methods: Accordingly, we conducted a proteomic            mechanisms to the associating partners of these
study to characterize the 20S proteasome                  proteasomes in respect to protein phosphatase 1
complexes and their postulated organ-specific             (PP1) and casein kinase II (CKII). CKII decreased
responses in liver and heart. The cardiac and             all three proteolytic activities of the hepatic 20S
hepatic 20S proteasomes were isolated from the            proteasomes and had no detectable effect in the
same mice with identical genetic information.             heart, while PP1 enhanced just the β5 cardiac 20S
Purified proteasome samples were resolved by              activity, with no detectable effect in the liver. Taken
either denaturing or native electrophoresis. Bands        together, these findings illustrate differential
were tryptically digested and analyzed on an LTQ          proteome biology in an organ specific fashion, i.e.
mass spectrometer. We then examined the                   the unique assembly and functional heterogeneity
molecular composition, complex assembly, post-            of the mammalian 20S proteasome complexes.
translational    modifications    and     associating
partners of these proteasome complexes.
Proteolytic assays of the 20S proteasomes
investigated all three proteolytic activities as they
differed between the heart and liver and in
response to manipulation of the suspect
associating partners.

To top