Subscriber access provided by Universiteit Utrecht
Journal of Proteome Research is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Chip-Based Enrichment and NanoLC#MS/MS
Analysis of Phosphopeptides from Whole Lysates
Shabaz Mohammed, Karsten Kraiczek, Martijn W. H. Pinkse,
Simone Lemeer, Joris J. Benschop, and Albert J. R. Heck
J. Proteome Res., 2008, 7 (4), 1565-1571• DOI: 10.1021/pr700635a • Publication Date (Web): 29 February 2008
Downloaded from http://pubs.acs.org on March 13, 2009
More About This Article
Additional resources and features associated with this article are available within the HTML version:
Links to the 2 articles that cite this article, as of the time of this article download
Access to high resolution figures
Links to articles and content related to this article
Copyright permission to reproduce figures and/or text from this article
Chip-Based Enrichment and NanoLC-MS/MS Analysis of
Phosphopeptides from Whole Lysates
Shabaz Mohammed,†Karsten Kraiczek,‡Martijn W. H. Pinkse,†,§Simone Lemeer,†
Joris J. Benschop,†,⊥and Albert J. R. Heck*,†
Biomolecular Mass Spectrometry and Proteomics Group, Bijvoet Center for Biomolecular Research and Utrecht
Institute for Pharmaceutical Sciences, Utrecht University, Sorbonnelaan 16, 3584 CA Utrecht, The Netherlands,
and Agilent Technologies R&D and Marketing GmbH & Company KG, Hewlett-Packard-Strasse 8,
76337 Waldbronn, Germany
Received October 3, 2007
Protein phosphorylation may be the most widespread and possibly most important post-translational
modification (PTM). Considering such a claim, it should be no surprise that huge efforts have been
made to improve methods to allow comprehensive study of cellular phosphorylation events. Neverthe-
less, comprehensive identification of sites of protein phosphorylation is still a challenge, best left to
experienced proteomics experts. Recent advances in HPLC chip manufacturing have created an
environment to allow automation of popular techniques in the bioanalytical world. One such tool that
would benefit from the increased ease and confidence brought by automated ‘nanoflow’ analysis is
phosphopeptide enrichment. To this end, we have developed a reusable HPLC nanoflow rate chip using
TiO2particles for selective phosphopeptide enrichment. Such a design proved robust, easy to use,
and was capable of consistent performance over tens of analyses including minute amounts of complex
Keywords: Automated • Phosphopeptide Enrichment • TiO2• Chip
Post-translational modification (PTM) of proteins is na-
ture’s way to (transiently) modify protein function and/or
activity or to tag proteins for specific subcellular routing.1
Protein phosphorylation may be the most widespread and
thus possibly most important PTM.2–4Therefore, it should
be no surprise that significant efforts have been made to
develop methods to enable the study of cellular phospho-
rylation events. However, progress in attaining site localiza-
tion and expression level in context of cell/protein state has
been slow primarily due to the intrinsic nature of the
phosphorylated residue. For instance, phosphorylated pep-
tides are present at substoichiometric levels, undergo facile
fragmentation inside a mass spectrometer to create poten-
tially difficult to interpret mass spectra, and may provide
poorer signal responses than their unphosphorylated coun-
terparts.5A plethora of techniques have been developed
addressing several of these issues, primarily based on the
specific enrichment of the phosphopeptides of interest.
Phosphopeptide enrichment strategies include the use of
phosphorylated residue specific antibodies,6–8immobilized
metal cation affinity chromatography (IMAC),9–11with its
more recent variants based on metal oxides of zirconium,12
aluminum,13and titanium,14–17strong cation exchange
(SCX)18or strong anion chromatography exchange (SAX).19,20
Comprehensive reviews describing these enrichment tech-
niques are available.21,22Moreover, a number of mass
spectrometric methods have been developed targeted at the
specific analysis of phosphorylated peptides including pre-
cursor ion scanning using diagnostic phosphorylated peptide
fragments,23,24neutral loss scanning exploiting similarly
diagnostic fragmentation pathways,25multiple reaction moni-
toring (MRM),26multiple stages of fragmentation,27,28and,
most recently, electron transfer dissociation.29,30Despite all
these considerable efforts, it remains a challenge to perform
comprehensive identification of protein phosphorylation
Microfluidic HPLC-Chip/MS technology introduced over the
past decade provide novel platforms for peptide separation and
is establishing itself as a robust, reliable alternative to conven-
tional nanocolumn LC-MS systems.33Potential benefits are
provided by the integration of the enrichment column, analyti-
cal columns, connecting capillaries, and nanospray emitter
directly onto the microfluidic HPLC-Chip device, which was
demonstrated to lead to improved sensitivity and enhanced
chromatographic performance leading to superior peptide and
protein identification.34,35Such chips have been shown to be
more than acceptable replacements for traditional nanoLC
setups for a number of applications.36–39Utilizing these
* To whom correspondence should be addressed. E-mail: firstname.lastname@example.org.
‡Agilent Technologies R&D and Marketing GmbH & Company KG.
§Present address: Analytical Biotechnology, Delft University of Technol-
ogy, Julianalaan 67, 2628 BC Delft, The Netherlands.
⊥Present address: Department for Physiological Chemistry, University
Medical Center Utrecht, Universiteitsweg 100, 3584 CG Utrecht, The
10.1021/pr700635a CCC: $40.75
2008 American Chemical Society
Journal of Proteome Research 2008, 7, 1565–1571 1565
Published on Web 02/29/2008
advances in microfluidic HPLC-Chip/MS technologies, we
report here on the optimization of a chip design that combines
the power of TiO2-based phosphopeptide enrichment with the
microfluidic HPLC-chip technology.
Materials and Methods
Sequencing grade trypsin was purchased from Roche Diag-
nostics (Ingelheim, Germany). Bovine serum albumin, alpha
casein, beta casein, and hemoglobin were from Sigma-Aldrich
(Zwijndrecht, The Netherlands). Ammonium bicarbonate, so-
dium phosphate, potassium fluoride, potassium chloride,
sodium orthovanadate, dithithreitol (DTT), iodoacetamide,
acetic acid, and formic acid were from Sigma (Zwijndrecht, The
Netherlands). Titanium oxide was a gift from GL-Sciences (GL-
Sciences, Inc., Japan). Zorbax extend (5 µm) (Agilent, Wald-
bronn, Germany) resin was used for the trap column, and
ReproSil-Pur C18-AQ, 3 µm 120 Å (Dr. Maisch, HPLC GmbH,
Ammerbuch-Entringen, Germany) resin was used for the
analytical column. HPLC grade ACN was purchased from
Biosolve (Valkenswaard, The Netherlands).
Sample Preparations. For the test protein mixtures, serum
albumin, alpha and beta casein, and hemoglobin (100 µM) were
reduced in 1 mM DTT and alkylated in 2 mM iodoacetamide,
following digestion with trypsin overnight at a protein/protease
ratio of 50:1. On the day of analysis, the test mixture was
prepared by dilution of the protein digests (100 µM) to 20 fmol/
µL in 0.5% formic acid.
Human embryonic kidney (HEK) 293T cells were grown to
confluence in Dulbecco’s modified Eagle’s medium containing
10% fetal bovine serum (Invitrogen) and 0.05 mg/mL penicillin/
streptomycin (Invitrogen). Cells were washed with cold phos-
phate-buffered saline (PBS). Cells were consequently centri-
fuged at 1500 rpm to allow removal of the supernatant. Cells
were lysed in 100 µL of 8 M urea and 50 mM ammonium
bicarbonate, containing 5 mM sodium phosphate, 1 mM
potassium fluoride, and 1 mM sodium orthovanadate, pH 8.2.
Cellular debris was pelleted by centrifugation at 14 000 rpm
for 20 min. Approximately 150 µg of protein material was used
for analysis. Proteins were reduced with 1 mM DTT and
alkylated with 2 mM iodoacetamide. The mixture was diluted
4-fold to 2 M urea using 100 µL of 50 mM ammonium
bicarbonate and 50 µL of trypsin solution, 0.1 mg/mL, and
incubated overnight at 37 °C.
Strong Cation Exchange. Strong cation exchange (SCX) was
performed using a Zorbax BioSCX-Series II column (0.8 mm
(i.d.) × 50 mm (l), 3.5 µm), a FAMOS autosampler (LC-packing,
Amsterdam, The Netherlands), a Shimadzu LC-9A binary
pump, and a SPD-6A UV-detector (Shimadzu, Tokyo, Japan).
Prior to SCX chromatography, protein digests were desalted
using a small plug of C18 material (3 M Empore C18 extraction
disk) packed into a GELoader tip (Eppendorf) similar to what
has been previously described,40onto which ∼10 µL of Aqua
C18 (Phenomenex, Torrance, CA) (5 µm, 200 Å) material was
placed. The eluate was dried completely and subsequently
reconstituted in 20% ACN and 0.05% formic acid. After injec-
tion, a linear gradient of 1% min-1solvent B (500 mM KCl in
20% ACN and 0.05% formic acid, pH 3.0) was performed. A
total of 24 SCX fractions (1 min each, i.e., 50 µL elution volume)
were manually collected and dried in a vacuum centrifuge.
TiO2Chips. UV laser ablation in combination with vacuum
lamination of polyimide films was used to create the multilayer
polymer based µ-fluidic devices. The starting material used was
a polyimide film with good chemical resistance to solvents and
biofriendly to proteins and peptides. In a first step, the laser
beam is focused on the polyimide film surface, and by variation
of the energy per pulse, the laser spot size, and the cutting
speed, one can vary width and depth of the channels created.
The laser ablated films are aligned very precisely to each other
and are then vacuum-laminated under temperature and pres-
sure to form the monolithic-like fused chip. UV laser ablation
is also used for 3D trimming of the laminated chip to form an
integral electro-spray tip. Patterned noble metals are applied
by thin film deposition on the polymer film surfaces to create
electrical contacts for electrospray biasing. Finally, sections of
the open microchannels are used for packing, which in our
case are either trapping or enrichment sections or the analytical
sections. For clarity in describing chip component descriptions
and in order to allow a nomenclature similar to that of regular
nanoLC, the terms precolumn and analytical column will be
utilized for filled channels, and in the case of the precolumn,
the term section will be applied to its subdivided compartments.
The MS-mounted chip handler (commonly referred to as the
‘chip-cube’) allows mounting and positioning of the chip tip.
The chip-cube also provides ultra low dead volume valving for
flow path-switching. The microfluidic device is identified by
an integrated RF-Tag, and all hydraulic connections are made
by the system according to chip and application type.
Two precolumn designs were investigated.
1. Design 1: ‘Two Sectioned’ Precolumn. On the basis of
the single precolumn-analytical column design which was first
described in ref 14, a common precolumn bed for reverse phase
and TiO2was created for the chip with a total volume of 320
nL(Figure 1). To achieve equal and accurate packing material
volumes, a second chip that contained a 160 nL section was
filled and used as the material reservoir for packing the ‘two
sectioned’ precolumn. In a first step, RP material (Zorbax
Extend 5µm, 160 nL) was filled into the common channel and
then the rest of the channel was filled with TiO210 µm spheres
(160 nL). The analytical channel (15 cm, 75 µm) was filled with
Reprosil C18 material.
2. Design 2: Three Discrete Sectioned Precolumn. From
our earlier work, we concluded that a three discrete sectioned
precolumn as described31,41could be advantageous to imple-
ment on a microfluidic device as well. To create three discrete
sections for the precolumn, 2 additional layers were required.
The two RP trapping sections of the precolumn were in a
distinct location on one layer, while the TiO2was on a separate
layer with an intermediate layer providing separation. Liquid
transfer was achieved through near-zero dead volume µ-sieve
frit filters formed by UV laser ablation in the intermediate layers
(Figure 2). Approximately 100 single micrometer holes are
forming one thin film sieve frit µ-filter. The two end channels
were individually filled with Zorbax Extend C18 (100 nL), while
the center section was filled with TiO2(45 nL). The analytical
channel (15 cm, 75 µm) was filled with Reprosil C18 material.
Online Enrichment and Analysis. Trapping, enrichment,
and analysis was performed on an Agilent 1100 HPLC system,
consisting of a loading pump operating in the microliter per
minute (µL/min) range and an analytical pump operating in
the nanoliter per minute (nL/min) range.
Two Sectioned Precolumn. Peptides were trapped at 3 µL/
min using 100% solvent A (0.6% acetic acid). (Note: phospho-
rylated peptides will bind to the TiO2section (enrichment),
while all other peptides would be trapped on the subsequent
C18 section.) An initial analysis was performed of what is
trapped on the C18 section by switching the precolumn to be
Mohammed et al.
1566 Journal of Proteome Research • Vol. 7, No. 4, 2008
in-line with the analytical column and nanoflow (analytical)
pump (gradient from 5 to 40% solvent B (80% acetonitrile
(ACN), 0.6%acetic acid) over 40 min at 200 nL/min; total
analysis time 60 min). Elution of phosphorylated peptides from
the TiO2section to the C18 section was achieved by injection
of 30 µL of 250 mM ammonium bicarbonate solution, pH 9.0
(adjusted with ammonia), containing 10 mM sodium phos-
phate, 5 mM sodium orthovanadate, and 1 mM potassium
fluoride. Immediately afterward, an injection of 20 µL of 10%
formic acid was performed to wash and re-equilibrate the
precolumn. Analysis of the eluted (phosphorylated) peptides
was performed by switching the precolumn to be in-line with
the analytical column for a second H2O/ACN gradient.
Three Sectioned Precolumn. Peptides were trapped at 3 µL/
min using 100% solvent A (0.6% acetic acid and 0.5% formic
acid in water) on the first 100 nL C18section. An initial analysis
was performed by switching the precolumn to be in-line with
the analytical column and nanoflow pump, followed by a
gradient from 5 to 40% solvent B (80% ACN, 0.6% acetic acid,
and 0.5% formic acid) over 40 min at 200 nL/min; total analysis
time 60 min. (Note: phosphorylated peptides will bind to the
TiO2section when introduced; the low flow rate allows better
conditions for binding when compared with direct loading of
the sample onto a TiO2section at a flow rate of 3 µL/min.) All
other peptides, with no TiO2affinity were chromatographically
separated at ∼200 nL/min. Elution of phosphorylated peptides
was achieved by injection of 30 µL of 250 mM ammonium
bicarbonate solution, pH 9.0 (adjusted with ammonia), con-
taining 10 mM sodium phosphate, 5 mM sodium orthovana-
date, and 1 mM potassium fluoride. Immediately afterward,
an injection of 20 µL of 10% formic acid was performed to wash
and re-equilibrate the precolumn. Analysis of the eluted
(phosphorylated) peptides was performed by switching the
precolumn to be in-line with the analytical column for a second
Mass Spectrometry. An Agilent MSD XCT+ ion trap mass
spectrometer (Agilent, Walbronn, Germany) was operated in
data-dependent mode, automatically switching between MS
and MS/MS. MS spectra (from m/z 450–1500) were the average
of 2 scans using a target value of 500 000. The three most
intense ions (3 scan averages) were selected for collision-
induced fragmentation using a target value of 500 000 and a
fragmentation amplitude of 1 V (with smart frag. 30%-200%).
Data Analysis. Mascot generic files were generated from raw
data using default values in Distiller (version 126.96.36.199 Matrix
Science, U.K.). These peak lists were searched using an in-
house-licensed Mascot search engine (version 2.1.0, Matrix
Science, U.K.). Files originating from the standard protein
mixture analyses were searched against Swiss-Prot database
(version 53.2), while those related to the HEK293 cells were
against IPI Human (version 3.28). Carbamidomethyl cysteine
was set as a fixed modification, while protein N-acetylation,
oxidized methionines, and phosphorylation of serine and
threonine were set as variable modifications. Trypsin was
specified as the proteolytic enzyme and 1 missed cleavage was
allowed. The mass tolerance of the precursor ion was set to
Figure 1. (a) Schematic of the TiO2-RP HPLC-chip having a common precolumn bed for TiO2and RP. Total volume of the precolumn
section is 320 nL (160 nL + 160 nL). The analytical column is 15 cm in length with an approximately rectangular cross section with a
diagonal of 75 µm. (b) 3D zoom of the actual design which is to scale of the dual phase precolumn. (c) TiO2-RP HPLC-chip analysis of
BSA, alpha casein, and hemoglobin (50 fmol each). The flow-through analysis where the uppermost panel is the BPI chromatogram,
while the lower panels represent extracted ion chromatograms for the main phosphopeptides. (d) Chromatograms of what bound and
eluted from the TiO2enrichment column. All panels are extracted ion chromatograms for the indicated sequences.
Phosphoproteomics on a Chip
Journal of Proteome Research • Vol. 7, No. 4, 2008
400 ppm, and that of fragment ions was set to 0.9 Da. The
threshold for MASCOT identification was set to 41 which
corresponded to p < 0.01. Data is available as a zipped scaffold
file at: https://bioinformatics.chem.uu.nl/supplementary/mo-
Our starting point for chip design was the online TiO2
phosphopeptide enrichment system developed by Pinkse et
al.14The essential part of that design was a two sectioned
precolumn where TiO2preceded a regular C18 trapping section.
During sample loading, phosphorylated peptides would bind
to the TiO2section, while all other peptides would flow through
and be trapped on the C18 section. Figure 1 provides an
overview of the two sectioned TiO2-RP precolumn chip. To
keep the design as close to the robust 1D chip design,37the
channel that would become the precolumn was simply en-
larged to allow 320 nL of packing material. In a first step, a
well-defined volume of RP material (160 nL) was packed and
then the rest of the chip channel was filled with TiO210 µm
Initial testing of the design was performed with 50 fmol of
bovine serum albumin, alpha casein, and hemoglobin digest.
Loading of sample was performed at a ‘typical’ flow rate of 3
µL/min where phosphopeptides selectively bind to the first
section of the precolumn, the TiO2, while the remaining
peptides just flow through and concentrate on the C18 section
of the precolumn. An analysis of what was bound to the C18
was performed, which allowed an evaluation of the enrichment
efficiency of the phosphopeptides to the TiO2. Subsequently,
elution was performed with a pH 9 ammonium bicarbonate
solution containing sodium phosphate, sodium orthovanadate,
and 1 mM potassium fluoride. These Lewis base additives have
been found to improve the efficiency of elution in titanium
enrichment experiments and thus reduce the need to use more
harsh conditions such as higher pH.31,41A final injection of
10% FA was then performed to remove any residual elution
buffer followed by a gradient analysis of the eluted peptides.
Figure 1 represents extracted chromatograms for the main
phosphorylated alpha casein peptides alongside the base peak
chromatogram for both the flow-through and elution analysis.
As can be clearly observed, efficient binding and elution of the
phosphopeptides was achieved. Although, as the BPI chro-
matograms exhibits, a number of nonphosphorylated peptides
are also present in the elution step. Further analysis of these
peptides indicated they were highly acidic in nature. Contami-
nation with acidic peptides represents a common issue in
phosphopeptide enrichment based on metal co-ordination.11
However, in the case of TiO2, it was shown that acidic peptides
show different binding behavior to phosphorylated peptides,
and thus, this difference in conduct can be exploited to improve
selectivity.15A more pressing concern to arise with this two
sectioned precolumn design was the reduction in enrichment
efficiency over time of the TiO2chip. On average, the percent-
Figure 2. (a) Schematic of the RP-TiO2-RP ‘sandwich’ HPLC-chip: The graphic highlights the three discrete sections for precolumn RP
(100 nL)-TiO2(45 nL)-RP (100 nL) and 15 cm, 75 µm analytical column. (b) 3D zoom of the actual design of the precolumn indicating
flow path during loading of sample; note design only allows analysis in forward flush mode. (c) Cross section of precolumn indicating
the use of 6 layers in order to achieve the discrete precolumn sections. Green squares indicate zero dead volume links. (d) RP-TiO2-RP
‘sandwich’ HPLC-chip analysis of BSA, alpha casein, and hemoglobin (50 fmol each). The flow-through analysis where the uppermost
panel is the BPI chromatogram, while the lower panels represent extracted ion chromatograms for the main phosphopeptides. (e)
Chromatograms of what bound and eluted from the TiO2enrichment column. All panels are extracted ion chromatograms for the
Mohammed et al.
1568Journal of Proteome Research • Vol. 7, No. 4, 2008
age of phosphopeptide that did not bind to the TiO2was nearly
50% after only 10 analyses. Possible rationales for such poor
retention are compromise of the TiO2section by components
present in the injected analytes as well as material that may
originate from the autosampler itself such as metal and silica.
Another major issue was that the efficiency of elution became
poorer with each analysis, creating the necessity to perform
multiple elutions. Our main suspicion for this multiple elution
problem was that phase mixing could be occurring in the
precolumn, as there were no discrete independent sections for
the TiO2and RP materials, and furthermore, the TiO2particles
are not uniform in size (Supplemental Figure 1 in Supporting
Information). Although the flow through the precolumn is in
the same direction during both the loading and analysis of
sample, this section will experience quick flow speed and
pressure changes when switching between the two modes. Such
rapid changes can cause agitation and dislodging of the smaller
TiO2particles allowing them to migrate into the RP section.
Such phase mixing can easily explain the precolumn degrada-
tion in a reused chip.
To solve the issues that arose from this two sectioned
precolumn, a new chip was developed, along the lines of the
successful ‘sandwich’ chip design for online TiO2enrichment
we described previously.31,41The ‘sandwich’ refers to an
additional RP section placed in front of the TiO2section which
will behave as a guard and allow binding of phosphopeptides
to the TiO2particles to be more efficient, that is, an RP-TiO2-
RP precolumn. The chip was increased to a 6 layer design to
accommodate discrete regions for each section of the ‘precol-
umn’, thereby eliminating any issues caused by phase mixing
(Figure 2a-c). The size of the TiO2section was also reduced
to improve ability to elute. This is possible because the TiO2
has a far higher capacity than the RP material.15An additional
consequence of this design is that all peptides, including those
that are phosphorylated, will bind initially to the first RP
section. An analysis of what is bound to the first section of the
precolumn will allow the phosphopeptides to migrate and bind
to the TiO2section, while all other peptides will continue onto
the analytical column for separation and MS detection. Figure
2d,e demonstrates typical performance of the new design. What
is noteworthy is the chromatograms displayed in Figure 2
represent analysis number 30 on that particular chip. Thus, the
3-discrete sections of the precolumn eliminate issues relating
to poor binding and elution. The issue relating to selectivity of
phosphopeptides and coenrichment of acidic peptides raised
concerns about efficacy of the online enrichment approach.
However, dramatic improvements in selectivity were achieved
through the use of a modest amount of formic acid (0.5%) in
the LC solvents as exhibited in Figure 2e where the phospho-
peptides have become the base peaks in the elution analysis.31,42
Although the formic acid improves selectivity, one can imagine
the requirement to attain an additional level of improvement
that may only be possible through the use of additives such as
dihydroxybenzoic acid15or lactic acid.43It is also possible to
implement the use of additives into the online chip TiO2system
by performing ‘wash’ steps, while the phosphopeptides are
retained on the TiO2section before the elution step.
Ideal applications for online enrichment analysis are cellular
phosphoproteome studies. A typical method for performing
such an experiment is the use of SCX fractionation at low pH
to perform an initial enrichment of phosphorylated peptides18
followed by a second step of enrichment based on IMAC27or
TiO2.16,44The SCX step will produce a large number of fractions
(usually numbering around 20–50), and so, an automated
online secondary enrichment step would be highly desirable.
To test the applicability of the sandwich TiO2chip, an SCX-
TiO2experiment using 150 µg of human embryonic kidney
(HEK) 293T cell lysate was performed. A new chip was used,
and after an initial analysis of the standard phosphopeptide,
it was ready for use. Figure 3a,b represents the base peak
intensity chromatograms for the flow-through and elution
analysis of the online TiO2enrichment for one of these (HEK)
293T SCX fractions. As the figure highlights, the amount of
material present in the flowthrough is dramatically higher than
the elution indicating a significant level of reduction in sample
complexity. Furthermore, a detailed analysis of this TiO2
analysis shows mostly N-acetylated peptides and protein
C-termini in the flow-through as well as a few phosphorylated
peptides (all of which are present in the elution), while over
80% of peptides present in the elution fraction were phospho-
rylated. Over 100 phosphopeptides were found in this single
60 min gradient (90 min, total) analysis of this fraction. All data
can be viewed through a download of a scaffold file found at
_JPR/. However, the representation of phosphorylated peptides
in the elution can be improved through the use of MS3triggered
Figure 3. An on-chip enrichment/analysis of SCX fraction 7 from a 150 µg HEK293 cell lysate digest. (Left panels) BPI chromatogram
represents the flow-through analysis (top), BPI chromatogram represents the elution analysis (bottom). (Right panels) Mass spectrum
corresponding to the survey scan used to isolate and fragment GNIETTSEDGQVPpSPK (top), tandem mass spectrum of indicated
Phosphoproteomics on a Chip
Journal of Proteome Research • Vol. 7, No. 4, 2008
neutral loss scans,18,27multistage activation,28or an instrument
with better mass accuracy, resolution, and more comprehensive
CAD such as a quadrupole-time-of-flight (Q-ToF) instrument
which will improve sequencing of these awkward species and
create a more realistic picture of enrichment levels. During the
analysis of these SCX fractions, three additional analyses of the
standard proteome mixture were performed (Supplementary
Figure 2 in Supporting Information). Beta casein was added to
the standard mixture to provide the phosphopeptide FQpSE-
EQQQTEDELQDK which is known to have a high affinity (and
therefore strong retention) for TiO2and other metal based
methods due to its highly acidic nature alongside its phosphate
moiety. Such a peptide will highlight issues relating to efficient
elution from TiO2. Supplemental Figure 2 (Supporting Informa-
tion) contains the three additional elution analyses performed
during the use of the new design titanium chip and highlights
a consistent level of performance over 20 analyses. To gauge
reproducibility of the sandwich TiO2chip, the initial analysis
of the standard was used as a reference for the subsequent
robustness checks. The intensities observed in this initial
elution, that is, first analysis, were deemed to represent ‘100%’
efficiency. When the extracted ion chromatograms across these
3 additional ‘standard’ analyses were used, the phosphopeptide
signal average was 150.6% ((62.4%). The standard deviation
across only the 3 additional analyses was 22% which follows
expectation since the chip requires at least one full analysis
cycle to become consistent. Additionally, the retention time of
these peptides differed by less than 10 s on a 60 min analysis.
Furthermore, the only peptide that was observed (once) in the
flow-through analysis was YKVPQLEIVPNpSAEER. This was not
unsurprising since mis-cleavage peptides are known to have
poorer binding efficiencies.45This initial sandwich RP-TiO2-
RP chip proved more than able for an SCX-TiO2experiment
with the chip providing a consistent level of performance across
the analyses with little or no intervention from the operator.
An automated online TiO2chip based liquid chromato-
graphic approach, aimed at enriching phosphorylated peptides
from large complex proteolytic digests, is described. By em-
ploying an RP-TiO2-RP ‘sandwich’ precolumn, as successfully
described for an online capillary approach,31,41a facile and
sensitive device was developed that requires little intervention
from the user to acquire successful operation. This three
sectioned sandwich TiO2precolumn chip allows the same high
selectivity of phosphopeptide-enrichment as the capillary
design but without the need for column preparation and expert
handling. The chip can be simply inserted into the chip cube
and is ready for analysis, overcoming typical experimental
hurdles as dead-volumes and tubing artifacts. In these first
chips, performance could be maintained over tens of analyses
or days of operation with little change in phosphopeptide
enrichment efficiency and analysis. Once the design can survive
close to a hundred analyses, it can be a highly feasible
technology for nonexperts to perform phosphoproteomics
analysis. Improvements can still be envisaged, with maybe even
longer analytical columns and the coupling of the microfluidic
device with a more advanced mass spectrometer.31,41
Technologies foundation. We would also like to thank
Supporting Information Available: Figures of the
schematic representing the cross section TiO2-RP HPLC-chip
precolumn; BPI and extracted chromatograms for three TiO2
elution analyses of a standard protein mixture; table of peak
intensities and peptide retention times for four standard protein
mixture analyses. This material is available free of charge via
the Internet at http://pubs.acs.org.
(1) Jensen, O. N. Interpreting the protein language using proteomics.
Nat. Rev. Mol. Cell Biol. 2006, 7 (6), 391–403.
(2) Manning, G.; Whyte, D. B.; Martinez, R.; Hunter, T.; Sudarsanam,
S. The protein kinase complement of the human genome. Science
2002, 298 (5600), 1912.
(3) Mann, M.; Jensen, O. N. Proteomic analysis of post-translational
modifications. Nat. Biotechnol. 2003, 21 (3), 255–261.
(4) Linding, R.; Jensen, L. J.; Ostheimer, G. J.; van Vugt, M.; Jorgensen,
C.; Miron, I. M.; Diella, F.; Colwill, K.; Taylor, L.; Elder, K.;
Metalnikov, P.; Nguyen, V.; Pasculescu, A.; Jin, J.; Park, J. G.;
Samson, L. D.; Woodgett, J. R.; Russell, R. B.; Bork, P.; Yaffe, M. B.;
Pawson, T. Systematic discovery of in vivo phosphorylation
networks. Cell 2007, 129 (7), 1415–1426.
(5) Steen, H.; Jebanathirajah, J. A.; Rush, J.; Morrice, N.; Kirschner,
M. W. Phosphorylation analysis by mass spectrometry-Myths,
facts, and the consequences for qualitative and quantitative
measurements. Mol. Cell. Proteomics 2006, 5 (1), 172–181.
(6) Pandey, A.; Podtelejnikov, A. V.; Blagoev, B.; Bustelo, X. R.; Mann,
M.; Lodish, H. F. Analysis of receptor signaling pathways by mass
spectrometry: Identification of Vav-2 as a substrate of the epider-
mal and platelet-derived growth factor receptors. Proc. Natl. Acad.
Sc. U.S.A. 2000, 97 (1), 179–184.
(7) Gronborg, M.; Kristiansen, T. Z.; Stensballe, A.; Andersen, J. S.;
Ohara, O.; Mann, M.; Jensen, O. N.; Pandey, A. A mass spectrometry-
based proteomic approach for identification of serine/threonine-
phosphorylated proteins by enrichment with phospho-specific
antibodies - Identification of a novel protein, Frigg, as a protein
kinase A substrate. Mol. Cell. Proteomics 2002, 1 (7), 517–527.
(8) Zheng, H. Y.; Hu, P.; Quinn, D. F.; Wang, Y. K. Phosphotyrosine
proteomic study of interferon alpha signaling pathway using a
combination of immunoprecipitation and immobilized metal
affinity chromatography. Mol. Cell. Proteomics 2005, 4 (6), 721–
(9) Posewitz, M. C.; Tempst, P. Immobilized gallium(III) affinity
chromatography of phoshopetides. Anal. Chem. 1999, 71 (14),
(10) Stensballe, A.; Andersen, S.; Jensen, O. N. Characterization of
phosphoproteinhs from electrophoretic gels by nanoscale Fe(III)
affinity chromatography with off-line mass spectrometry analysis.
Proteomics 2001, 1 (2), 207–222.
(11) Ficarro, S. B.; McCleland, M. L.; Stukenberg, P. T.; Burke, D. J.;
Ross, M. M.; Shabanowitz, J.; Hunt, D. F.; White, F. M. Phosphop-
roteome analysis by mass spectrometry and its application to
Saccharomyces cerevisiae. Nat. Biotechnol. 2002, 20 (3), 301–305.
(12) Kweon, H. K.; Hakansson, K. Selective zirconium dioxide-based
enrichment of phosphorylated peptides for mass spectrometric
analysis. Anal. Chem. 2006, 78 (6), 1743–1749.
(13) Chen, C. T.; Chen, W. Y.; Tsai, P. J.; Chien, K. Y.; Yu, J. S.; Chen,
Y. C. Rapid enrichment of phosphopeptides and phosphoproteins
from complex samples using magnetic particles coated with
alumina as the concentrating probes for MALDI MS analysis. J.
Proteome Res. 2007, 6 (1), 316–325.
(14) Pinkse, M. W. H.; Uitto, P. M.; Hilhorst, M. J.; Ooms, B.; Heck,
A. J. R. Selective isolation at the femtomole level of phosphopep-
tides from proteolytic digests using 2D-nanoLC-ESI-MS/MS and
titanium oxide precolumns. Anal. Chem. 2004, 76 (14), 3935–3943.
(15) Larsen, M. R.; Thingholm, T. E.; Jensen, O. N.; Roepstorff, P.;
Jorgensen, T. J. D. Highly selective enrichment of phosphorylated
peptides from peptide mixtures using titanium dioxide microcol-
umns. Mol. Cell. Proteomics 2005, 4 (7), 873–886.
(16) Benschop, J. J.; Mohammed, S.; O’Flaherty, M. C.; Heck, A. J. R.;
Slijper, M.; Menke, F. L. H. Quantitative phospho-proteomics of
early elicitor signalling in Arabidopsis. Mol. Cell. Proteomics 2007,
6 (7), 1198–1214.
Mohammed et al.
1570Journal of Proteome Research • Vol. 7, No. 4, 2008
(17) Ishihama, Y.; Wei, F. Y.; Aoshima, K.; Sato, T.; Kuromitsu, J.; Oda,
Y. Enhancement of the efficiency of phosphoproteomic identifica-
tion by removing phosphates after phosphopeptide enrichment.
J. Proteome Res. 2007, 6 (3), 1139–1144.
(18) Beausoleil, S. A.; Jedrychowski, M.; Schwartz, D.; Elias, J. E.; Villen,
J.; Li, J. X.; Cohn, M. A.; Cantley, L. C.; Gygi, S. P. Large-scale
characterization of HeLa cell nuclear phosphoproteins. Proc. Natl.
Acad. Sci. U.S.A. 2004, 101 (33), 12130–12135.
(19) Nuhse, T. S.; Stensballe, A.; Jensen, O. N.; Peck, S. C. Large-scale
analysis of in vivo phosphorylated membrane proteins by im-
mobilized metal ion affinity chromatography and mass spectrom-
etry. Mol. Cell. Proteomics 2003, 2 (11), 1234–1243.
(20) Dai, J.; Jin, W. H.; Sheng, Q. H.; Shieh, C. H.; Wu, J. R.; Zeng, R.
Protein phosphorylation and expression profiling by Yin-Yang
multidimensional liquid chromatography (Yin-Yang MDLC) mass
spectrometry. J. Proteome Res. 2007, 6 (1), 250–262.
(21) Pinkse, M. W. H.; Heck, A. J. R. Essential enrichment strategies in
phosphoproteomics. Drug Discovery Today: Technol. 2006, 3, 331–
(22) Zahedi, R. P.; Begonja, A. J.; Gambaryan, S.; Sickmann, A. Phos-
phoproteomics of human platelets: A quest for novel activation
pathways. Biochim. Biophys. Acta 2006, 1764 (12), 1963–1976.
(23) Carr, S. A.; Huddleston, M. J.; Annan, R. S. Selective detection and
sequencing of phosphopeptides at the femtomole level by mass
spectrometry. Anal. Biochem. 1996, 239 (2), 180–192.
(24) Steen, H.; Kuster, B.; Mann, M. Quadrupole time-of-flight versus
triple-quadrupole mass spectrometry for the determination of
phosphopeptides by precursor ion scanning. J. Mass Spectrom.
2001, 36 (7), 782–790.
(25) Covey, T.; Shushan, B.; Bonner, R.; Shroder, W.; Hucho, F.; Jornvall,
H.; Hoog, J.-O.; Gustavsson, A.-M. Methods Protein Sequence Anal.
1991, 24, 9–256.
(26) Unwin, R. D.; Griffiths, J. R.; Leverentz, M. K.; Grallert, A.; Hagan,
I. M.; Whetton, A. D. Multiple reaction monitoring to identify sites
of protein phosphorylation with high sensitivity. Mol. Cell. Pro-
teomics 2005, 4 (8), 1134–1144.
(27) Gruhler, A.; Olsen, J. V.; Mohammed, S.; Mortensen, P.; Faergeman,
N. J.; Mann, M.; Jensen, O. N. Quantitative phosphoproteomics
applied to the yeast pheromone signaling pathway. Mol. Cell.
Proteomics 2005, 4 (3), 310–327.
(28) Schroeder, M. J.; Shabanowitz, J.; Schwartz, J. C.; Hunt, D. F.; Coon,
J. J. A neutral loss activation method for improved phosphopeptide
sequence analysis by quadrupole ion trap mass spectrometry.
Anal. Chem. 2004, 76 (13), 3590–3598.
(29) Chi, A.; Huttenhower, C.; Geer, L. Y.; Coon, J. J.; Syka, J. E. P.; Bai,
D. L.; Shabanowitz, J.; Burke, D. J.; Troyanskaya, O. G.; Hunt, D. F.
Analysis of phosphorylation sites on proteins from Saccharomyces
cerevisiae by electron transfer dissociation (ETD) mass spectrom-
etry. Proc. Natl. Acad. Sci. U.S.A. 2007, 104 (7), 2193–2198.
(30) Molina, H.; Horn, D. M.; Tang, N.; Mathivanan, S.; Pandey, A.
Global proteomic profiling of phosphopeptides using electron
transfer dissociation tandem mass spectrometry. Proc. Natl. Acad.
Sci. U.S.A. 2007, 104 (7), 2199–2204.
(31) Pinkse, M. W.; Mohammed, S.; Gouw, J. W.; van Breukelen, B.;
Vos, H. R.; Heck, A. J. Highly robust, automated, and sensitive
online TiO2-based phosphoproteomics applied to study endog-
enous phosphorylation in Drosophila melanogaster. J. Proteome
Res. 2008, 7 (2), 687–697.
(32) Bodenmiller, B.; Mueller, L. N.; Mueller, M.; Domon, B.; Aebersold,
R. Reproducible isolation of distinct, overlapping segments of the
phosphoproteome. Nat. Methods 2007, 4 (3), 231–237.
(33) Lazar, I. M.; Grym, J.; Foret, F. Microfabricated devices: A new
sample introduction approach to mass spectrometry. Mass Spec-
trom. Rev. 2006, 25 (4), 573–594.
(34) Yin, H.; Killeen, K.; Brennen, R.; Sobek, D.; Werlich, M.; van de
Goor, T. Microfluidic chip for peptide analysis with an integrated
HPLC column, sample enrichment column, and nanoelectrospray
tip. Anal. Chem. 2005, 77 (2), 527–533.
(35) Hardouin, J.; Joubert-Caron, R.; Caron, M. HPLC-chip-mass
spectrometry for protein signature identifications. J. Sep. Sci. 2007,
30 (10), 1482–1487.
(36) Vollmer, M.; Horth, P.; Rozing, G.; Coute, Y.; Grimm, R.; Hoch-
strasser, D.; Sanchez, J. C. Multi-dimensional HPLC/MS of the
nucleolar proteome using HPLC-chip/MS. J. Sep. Sci. 2006, 29 (4),
(37) Fortier, M. H.; Bonneil, E.; Goodley, P.; Thibault, P. Integrated
microfluidic device for mass spectrometry-based proteomics and
its application to biomarker discovery programs. Anal. Chem. 2005,
77 (6), 1631–1640.
(38) Ghitun, M.; Bonneil, E.; Fortier, M. H.; Yin, H.; Killeen, K.; Thibault,
P. Integrated microfluidic devices with enhanced separation
performance: application to phosphoproteome analyses of dif-
ferentiated cell model systems. J. Sep. Sci. 2006, 29 (11), 1539–
(39) Ninonuevo, M.; An, H.; Yin, H.; Killeen, K.; Grimm, R.; Ward, R.;
German, B.; Lebrilla, C. Nanoliquid chromatography-mass spec-
trometry of oligosaccharides employing graphitized carbon chro-
matography on microchip with a high-accuracy mass analyzer.
Electrophoresis 2005, 26 (19), 3641–3649.
(40) Ishihama, Y.; Rappsilber, J.; Mann, M. Modular stop and go
extraction tips with stacked disks for parallel and multidimensional
peptide fractionation in proteomics. J. Proteome Res. 2006, 5 (4),
(41) Lemeer, S.; Pinkse, M. W. H.; Mohammed, S.; van Breukelen, B.;
den Hertog, J.; Slijper, M.; Heck, A. J. R. On-line automated in vivo
zebrafish phosphoproteomics: From large-scale analysis down-to
a single embryo. J. Proteome Res. 2008, 7, 1555–1564.
(42) Schlosser, A.; Vanselow, J. T.; Kramer, A. Mapping of phosphory-
lation sites by a multi-protease approach with specific phospho-
peptide enrichment and nanoLC-MS/MS analysis. Anal. Chem.
2005, 77 (16), 5243–5250.
(43) Sugiyama, N.; Masuda, T.; Shinoda, K.; Nakamura, A.; Tomita, M.;
Ishihama, Y. Phosphopeptide enrichment by aliphatic hydroxy
acid-modified metal oxide chromatography for nano-LC-MS/MS
in proteomics applications. Mol. Cell. Proteomics 2007, 6 (6), 1103–
(44) Olsen, J. V.; Blagoev, B.; Gnad, F.; Macek, B.; Kumar, C.; Mortensen,
P.; Mann, M. Global, in vivo, and site-specific phosphorylation
dynamics in signaling networks. Cell 2006, 127 (3), 635–648.
(45) Klemm, C.; Otto, S.; Wolf, C.; Haseloff, R. F.; Beyermann, M.;
Krause, E. Evaluation of the titanium dioxide approach for MS
analysis of phosphopeptides. J. Mass Spectrom. 2006, 41 (12),
Phosphoproteomics on a Chip
Journal of Proteome Research • Vol. 7, No. 4, 2008