Article
pubs.acs.org/ac
Separation of Polypeptides by Isoelectric Point Focusing in
Electrospray-Friendly Solution Using a Multiple-Junction Capillary
Fractionator
Konstantin Chingin,† Juan Astorga-Wells,†,‡ Mohammad Pirmoradian Najafabadi,†,‡ Thorleif Lavold,‡
and Roman A. Zubarev*,†,§
†
Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Scheeles väg 2, SE-17177 Stockholm, Sweden
Biomotif AB, Stockholm, Sweden
§
Science for Life Laboratory, Stockholm, Sweden
‡
S Supporting Information
*
ABSTRACT: We introduce an online multiple-junction capillary
isoelectric focusing fractionator (OMJ-CIEF) for separation of
biological molecules in solution by pI. In OMJ-CIEF, the separation
capillary is divided into seven equal sections joined with each other
via tubular Nafion membrane insertions. Each junction is
communicated with its own external electrolytic buffer which is
used both to supply electrical contact and for solvent exchange. The
performance of the fractionator was explored using protein and
peptide samples covering broad pI range. Separation was achieved in ionic and ampholytic buffers, including ammonium formate,
ammonium hydroxide, histidine, and arginine. By maintaining electric potential across upstream segments of the capillary after
the focusing stage, selective release of downstream analyte fractions could be achieved. The selective release mode circumvents
the problem of peak broadening during mobilization and enables convenient comprehensive sampling for orthogonal separation
methods. Using single-component ampholyte buffers with well-defined pI cutoff values, controlled separation of protein mixture
into basic and acidic fractions was demonstrated. The device is cheap and easy to fabricate in-house, simple in operation, and
straightforward in interfacing to hyphened analytical platforms. OMJ-CIEF has a potential of becoming a practical add-on unit in
a wide range of bioanalytical setups, in particular as a first-dimension separation in mass spectrometry based proteomics or as a
preparative tool for analyte purification, fractionation, and preconcentration.
T
separation, excellent concentration capacity, high resolving
power, and sensitivity of CIEF have been found particularly
useful in the analysis of protein mixtures from biological
samples.3−6 In combination with electrospray ionization (ESI)MS detection, CIEF has been applied in a number of protein
studies for identification purposes7−13 as well as to probe
protein refolding,14 phosphorylation,15 glycosylation,16 noncovalent interactions,17 or to determine the isoelectric point of
protein isoforms.18 Because CIEF alone is often insufficient for
comprehensive analysis of real biological samples, the attention
is gradually shifting toward employing this method in
multidimensional platforms as a preparative step for a
second-dimension separation.4 Of particular interest is the
hyphenation between CIEF and RPLC, which offers a
promising alternative to the off-line 2D-PAGE approach19 in
high-throughput proteomics analyses. Unlike 2D LC × LC
protocols, the retention principles of CIEF and RPLC are truly
orthogonal, which greatly benefits the overall peak capacity in
this type of hyphenation.20,21 A number of CIEF−RPLC−MS
remendous developments in mass spectrometry (MS)
over the past 2 decades have brought it to the forefront of
bioanalytical methods. Unequaled specificity, resolving power,
mass accuracy, and high-throughput capabilities offered by
modern MS instrumentation render it essentially indispensable
in many types of biomolecular studies, particularly in the fields
of proteomics and metabolomics.1,2 The practicality of
complementary bioanalytical methods is therefore becoming
increasingly dependent on their online compatibility with MS
analysis rather than their mere stand-alone performance. This
trend is nicely illustrated by the current prevalence of liquid
chromatography (LC), in particular reversed-phase LC
(RPLC), over other separation techniques. While several
other types of separations techniques, such as isoelectric
focusing (IEF), capillary zone electrophoresis (CZE), or
sodium dodecyl sulfate polyacrylamide gel electrophoresis
(SDS−PAGE), are capable of providing superior operation in
a number of biological applications, it is the poor efficiency of
online interfaces with MS detection that limits the wider
application of these methods.
Capillary isoelectric focusing (CIEF) is an electrophoretic
method that separates amphoteric molecules in solution
according to their isoelectric point (pI). Mild conditions of
© 2012 American Chemical Society
Received: May 29, 2012
Accepted: July 9, 2012
Published: July 9, 2012
6856
dx.doi.org/10.1021/ac3013016 | Anal. Chem. 2012, 84, 6856−6862
Analytical Chemistry
Article
Figure 1. Schematic layout of the OMJ-CIEF setup: the six-port injector is shown in the sample-loop loading position. For details on assembling the
fractionator and experimental procedure refer to the Supporting Information.
anticonvective media are either avoided in MS hyphenations or
replaced by less detrimental chemicals, such as glycerol−water
mixtures.13,18,32 Large amounts of carrier ampholytes (CAs) in
CIEF greatly affect the sensitivity of MS detection, suppressing
ionization of analyte molecules and introducing abundant
spectral interferences.29,34,41,42 Although the presence of
ampholytes can be tolerated in some applications, it has a
particularly devastating effect in the analyses of real-life
biological mixtures, hindering observation of trace analytes.
Another problem associated with using high concentrations of
CAs, which is often disregarded in the literature, is their impact
on the performance of a mass spectrometer. Long-term
exposure to the high concentrations of CAs (on the order of
millimolar) and other low-volatility compounds, such as
glycerol, inevitably results in accumulation of these substances
in transfer lines and ion optics of the instrument, gradually
degrading its performance. This affects data reproducibility and
requires frequent workflow interruption for technical maintenance. Carrier ampholytes can be quite efficiently removed
prior to MS detection via online RPLC separation24,28,34,39,43 or
by microdialysis systems in protein analyses.26,35−37,44 In the
absence of purification steps, sensitivity of MS detection can be
improved by employing lower concentrations of CAs compared
to standard CIEF protocols. However, this approach is
commonly associated with decreased buffering capacity and
smoothness of pH gradient, as well as peak broadening and
resolution losses.4,41 Finally, MS-incompatible hydrophilic
polymers, such as polyacrylamide and poly(vinyl alcohol),
which are frequently used as capillary coatings to prevent
electroosmotic flow during separation,14,15,29,34 tend to emanate
into the liquid phase, e.g., upon local heating, and hinder the
analysis.45−49
In this report, we introduce an online multiple-junction
capillary isoelectric focusing (OMJ-CIEF) fractionator for
tandem analyses of biomolecules in solution. In OMJ-CIEF,
the separation capillary is divided into seven equal sections
joined with each other via tubular Nafion membrane insertions
(Figure 1). Each junction is communicated with its own
external electrolytic buffer, which is used both to supply
electrical contact and for solvent exchange. The multiplejunction design adds a number of capabilities to the CIEF
methodology. The possibility of local buffer exchange can be
used to tailor the pH profile along the entire capillary, which
makes it ideologically akin to the compartmental immobilized
pH gradient (IPG) IEF fractionators. Furthermore, by applying
a voltage gradient across certain segments of the capillary after
the focusing stage, analyte fractions can be selectively mobilized
for further interrogation by orthogonal methods. Control over
the mobilization of focused fractions is of particular value for
platforms have been developed and applied in analyses of
proteins and peptides from tumor tissues,22 human saliva,23
yeasts,24,25 and bacteria extracts.26−28
In spite of its unambiguous advantages for protein analysis,
CIEF still remains in the shadow of other capillary separation
methods in terms of routine utilization. A number of issues
associated with online CIEF-MS sampling are still to be
overpassed before this method can rival analytical characteristics and operability of mainstream protocols. The major
constraints are briefly overviewed below. Unlike off-line
analyses, the inherent absence of a terminal vial with electrolyte
buffer in online CIEF-ESI-MS poses considerable technical
challenge to ensure electrical continuity during the experiment.
Most popular approaches to circumvent this problem employ
coaxial sheath-liquid flow, which provides the necessary
electrical contact on the capillary terminus during the focusing
stage and aids subsequent ESI-MS detection.7,9,12,15,29−33
Simply speaking, the ESI source in such schemes is adapted
to replace the outlet vial in classic CIEF experiments. This
allows automated online operation but precludes the possibility
of multidimensional analyses, such as CIEF−RPLC−MS.
Furthermore, the use of sheath liquid causes analyte dilution
at the tip of the capillary, decreasing the sensitivity of detection.
Various microdialysis interfaces have been employed to supply
catholytic and anolytic buffers with electric contacts in
multidimensional CIEF setups,26,27,34−37 at the same time
alleviating online purification of analyte fractions prior to MS
detection.38 The high molecular weight cutoff in microdialysis
membranes (∼3000 Da) makes them particularly suitable for
protein applications but practically unfit for the analyses of
peptides and other small molecules. Comprehensive online
sampling of CIEF with the next-dimension separation in
multidimensional approaches, e.g., with RPLC, represents
another serious challenge. Normally, focused analytes are
mobilized into different storage loops via a splitting interface,
followed by a sequential RPLC−MS analysis of each
loop.22,24,25,27,28,34,39 The efficiency of such sampling, however,
suffers from the decreased resolution of separation during the
mobilization step, especially when hydrodynamic flow is
used.10,40 Besides that, a rapid valving system and the large
number of storage loops are necessary to efficiently fractionate
close CIEF bands. These and other technical requirements, e.g.,
the need for auxiliary pumping, sizably complicate experimental
setup and operation. On the chemistry side, CIEF-MS analyses
suffer from the low tolerance of MS toward classical CIEF
reagents. Thus, inorganic acids and bases traditionally used in
CIEF as catholytes and anolytes are normally replaced by ESIcompatible analogues, such as ammonia and formic acid.32,33
For the same reason, aqueous gels employed in CIEF as
6857
dx.doi.org/10.1021/ac3013016 | Anal. Chem. 2012, 84, 6856−6862
Analytical Chemistry
Article
multidimensional analyses that rely on CIEF separation as a
first dimension. The fractionator is cheap and easy to fabricate
in-house, simple in operation, and straightforward in interfacing
to hyphened analytical platforms, such as ESI-MS or RPLC. We
believe that OMJ-CIEF has the potential to become a practical
add-on unit in a wide range of biomolecular studies, e.g., as a
first-dimension separation in MS proteomics or as a preparative
tool for analyte purification, fractionation, and preconcentration. Here, we report on our proof-of-principle experiments in
which the OMJ-CIEF fractionator was used online with ESI-MS
detection for protein and peptide analyses.
■
METHODS
Details on assembling of the OMJ-CIEF fractionator, CIEF-MS
experiments, protein digestion protocol, and chemicals and
materials used in this study can be found in online Supporting
Information.
Figure 2. Extracted ion chromatogram (EIC) for myoglobin as a
function of focusing time in ammonium formate buffer (5 mM). Five
CIEF-MS runs were conducted successively with different IEF times
(0, 10, 35, 80, 180 min) using the same myoglobin stock solution for
injection (10 μM in 5 mM ammonium formate). All the other
experimental parameters were maintained unchanged in all the five
runs. Separation was carried in the constant current regime (5 μA).
EIC profiles are normalized to the peak intensity of the 80 min
separation run.
■
RESULTS AND DISCUSSION
OMJ-CIEF in Ionic Buffers. The multiple-junction interface
of OMJ-CIEF enables flexible external manipulation of the
solution-phase composition across the separation capillary. In
our first experiment, proteins were loaded in aqueous
ammonium formate buffer, chosen for its good compatibility
with MS.50 The same buffer was used as external electrolytic
solution in all the Eppendorf vials. External solutions were
titrated with ammonium hydroxide and formic acid such as to
produce a linear pH gradient rising from 2.9 in the most
downstream vial (no. 1 in Figure 1, MS end) up to 8.5 in the
most upstream vial (no. 8 in Figure 1, injector end). The pH
difference between each pair of adjacent vials was therefore
equal to 0.8 units. The intention was to enforce a gradient
inside the capillary during separation via local exchange of the
liquid inside the capillary with external buffer solutions. This
approach is conceptually different from the common IEF
methodology of using internally added chemicals, such as
ampholytes. To enhance the exchange rate, molarity of the
outer buffer was 10-fold of the carrier solution (typically, 50 vs
5 mM).
Figure 2 shows how focusing of myoglobin is achieved in
time. It involves two major constituent processes: migration of
the protein plug across the capillary and band narrowing. One
can see that the axial displacement of the maximum of protein
concentration is accomplished relatively fast while band
narrowing takes typically twice as long (Supporting Information
Figure S2). Subject to the external electric force, myoglobin
eventually migrated to the downstream area of the capillary.
From the elution peak time in MS, we estimate the focused
myoglobin band to be located in between the second and the
third junction of the capillary (Figure 1). In the focus point
(zero net force), the pH value of the mobile phase should be
equal to the pI of the protein, which is ca. 7.3 for myoglobin.
Subsequently, pH should be higher than 7.3 upstream from the
focus point and lower downstream in order to supply
myoglobin with overall negative or positive charges, respectively. This conclusion is supported by simple pH-sensing
experiments, in which the OMJ-CIEF was disconnected from
MS after the focusing stage, and the effluent solution was
continuously eluting onto a pH-indicator strip placed next to
the downstream capillary terminal. A sigmoid profile was
observed, started at pH ≈ 3−4 followed by a gradual increase
up to pH ≈ 5−6 within first 3 min, and a subsequent abrupt
jump to pH > 10. Of course, this off-line sensing yields only a
rough estimate, because high-mobility electrolytes rapidly
diffuse inside the capillary as soon as the high voltage is
switched off. Therefore, the actual slope of the sigmoid curve
can be much steeper than measured. The inability to enforce a
linear gradient using external buffers in our experiments can be
explained by the high electrophoretic mobility of ammonium
and formate ions. In the external electrical field abundant
formate anions tend to migrate toward the positive pole,
acidifying the downstream area of the capillary. In the same
way, ammonium cations drift to the negative pole, resulting in
the observed pH splitting. Kinetics of these processes appears
to be faster compared to the solvent exchange rate through the
Nafion membrane. This conclusion is supported by the
reference experiments, in which a similar sigmoid profile was
observed when all the external vials were filled with ammonium
formate at pH 4. Formation of the upstream zone with pH > 10
in such experiments can only be explained by electrolyte
redistribution along the capillary under applied voltage.
A smoother gradient is produced when the two-component
ammonium formate buffer is replaced with single-component
ammonium hydroxide. The ratio between the concentrations of
NH4+ and NH3 in 5 mM ammonium hydroxide solution is
considerably smaller compared to that in neutral ammonium
formate, when [NH4+] ≫ [NH3]. The low concentration of
free NH4+ in the ammonium hydroxide solution prevents
substantial redistribution of electrolytes when the external
electrical field is applied. As a result, a low-sloped pH gradient
from ca. 9 to 11 is established along the entire separation
capillary. Figure 3 shows the separation of protein mixture
containing myoglobin (pI 7.3), cytochrome c (pI 10.3), and
lysozyme (pI 11.3) achieved in ammonium hydroxide. Being
negatively charged in the entire pH range, myoglobin migrates
downstream without constraints until it reaches the end of the
column. It is worth noting that, in ammonium formate,
myoglobin migration had been blocked higher up by the acidic
pH zone rich with formate species, and therefore its elution
took longer time (Figures 2 and 3). In contrast with myoglobin,
lysozyme is positively charged in ammonium buffer and
therefore drifts to the upstream end of the column. Finally,
6858
dx.doi.org/10.1021/ac3013016 | Anal. Chem. 2012, 84, 6856−6862
Analytical Chemistry
Article
the polarity of the IEF voltage was inverted by swapping over
positive and negative electrodes, resulting in a mirror-image
separation. A sharp peak of lysozyme was first detected in the
ion chromatogram, followed by cytochrome c and, ultimately, a
broad band of myoglobin. Besides the body of the fractionator,
other factors contributing to peak distortion are associated with
analyte transition through the T-unit, downstream Nafion
membrane (no. 1, Figure 1), and PEEK tubing and fittings used
to connect the fractionator with the ionization source. As a
result, even the analytes focused at the downstream end of the
OMJ-CIEF fractionator undergo notable broadening before
detection.
Peak broadening during analyte mobilization deteriorates the
achievable separation and represents a general problem in
CIEF. It can be partially alleviated by the use of capillaries with
special coatings, anticonvective solution-phase additives,
electroosmotic flow (EOF), or electrophoretic mobilization,
etc., but these solutions are often associated with various
adverse effects to the overall performance.
Selective Analyte Mobilization in OMJ-CIEF. Complementary to mobilization protocols conventionally employed in
CIEF, the multiple-junction design of OMJ-CIEF offers the
capability to selectively release focused fractions. In this
approach, bands focused downstream are mobilized to the
detector with the carrier flow, while the upstream fractions are
restrained by the electric voltage applied across the respective
junctions. For example, in order to selectively release
myoglobin in the CIEF experiment described above (Figure
3), the downstream electrode was transferred from the first to
the third electrolyte vial and the carrier flow was then started.
This way, myoglobin focused at the downstream end of the
capillary freely eluted to the MS, while the release of
cytochrome c and lysozyme was delayed by the trapping
electric force. This electric force emerges as soon as the
proteins start to migrate downstream and acquire positive
charge in the lower pH region of the capillary. The force is
increasing as the proteins drift further along the capillary and at
some point can become sufficiently high to counterbalance the
hydrodynamic force. The concept of trapping charged
molecular species in membrane capillary interfaces by applied
voltage is known as microfluidic electrocapturing51 and is
employed in various bioanalytical applications for analyte
purification,52,53 concentration,54,55 and separation56,57 purposes. Electrocapturing capacity depends on several experimental
parameters, such as electrophoretic mobility of analyte ions at
given pH, voltage magnitude, and hydrodynamic flow rate. In
our experiments, we added 0.1% aqueous solution of formic
acid in the first two vials right before the release. As a result,
electrophoretic mobility of analyzed proteins sharply increased
as they had reached the acidified downstream region of the
fractionator. Under these conditions, capturing could be
achieved at the carrier flow rate as high as 3 μL min−1, which
enables rapid release of myoglobin (Supporting Information
Figure S3). An alternative approach to enhance the efficiency of
electrocapturing would be to use higher voltage, but it can be
associated with various unwanted effects caused by Joule
heating of the running solution.
The selective analyte mobilization mode has several
important advantages. As long as different bands are allocated
by IEF into different capillary sections, their elution can be
separated in time. This circumvents the problem of peak
broadening and possible band overlapping during simultaneous
mobilization, thereby increasing effective resolution of the
Figure 3. Separation of protein mixture containing myoglobin (10
μM), cytochrome c (65 μM), and lysozyme (56 μM) in ammonium
hydroxide buffer (5 mM). IEF was carried over 80 min in the constant
current regime (5 μA). All the fractions were mobilized for ESI-MS
detection simultaneously via hydrodynamic pumping (3 μL min−1).
EIC profiles of the proteins are normalized to the peak intensity of the
myoglobin chromatogram.
cytochrome c is focused in the middle part of the capillary
where the pH of the solution is equal to its pI value.
To perform CIEF separation in a different pH range,
ammonium hydroxide can be replaced with other ionic buffers
compatible with MS detection. For example, aqueous acetic
acid can be employed to generate a pH gradient in acidic range
(pH ∼ 4−5). The major problem associated with this approach
is that only narrow-range gradients can be produced and their
profiles are highly dependent on electrolyte concentration.
Thus, in the experiments described above the focus point of
cytochrome c across the separation capillary was found to be
particularly sensitive to the concentration of running buffer
solution. Increasing molarity of ammonia would result in both
higher mean value and narrower distribution of pH in the
mobile phase due to the decreased [NH4+]/[NH3] ratio. This
problem is obviated in CIEF when using zwitterionic
ampholyte buffers. In the external electric field applied across
the CIEF capillary, CAs will line up to locally modify the pH of
the solution phase in accordance with their pI, resulting in a
smooth and stable pH gradient. Various CA blends are
nowadays commercially available to cover wide pH ranges for
routine IEF analyses.
Peak Broadening during Analyte Mobilization. It is
important to emphasize that the extracted ion chromatogram
(EIC)-MS profiles of proteins in OMJ-CIEF-MS experiments
only partially reflect their actual in situ distribution in the
fractionator. For example, the observed width of the EIC-MS
profile for unfocused myoglobin (ca. 5 min, Figure 2)
corresponds to ∼15 μL volume of protein band at the
detection point, which is 7−8 times larger than the volume of
the sample injection loop (2 μL). The protein band is
significantly broadened during hydrodynamic mobilization, e.g.,
due to the interaction with capillary walls or sample
diffusion.10,40 As can be seen from Figure 3, EIC peaks of
analyzed proteins are broadening in the following order:
myoglobin < cytochrome c < lysozyme. This trend reflects the
increasing residence time of corresponding proteins in the
separation capillary during the mobilization process. While
myoglobin is focused at the downstream end of the separation
device and is rapidly delivered to the MS, lysozyme has to pass
the entire volume of the device (ca. 15 μL) before ESI-MS
detection. Over this time, the focused protein band is
substantially affected by the interaction with both mobile and
stationary phases of the fractionator. In a reference experiment,
6859
dx.doi.org/10.1021/ac3013016 | Anal. Chem. 2012, 84, 6856−6862
Analytical Chemistry
Article
method. In this respect, selective mobilization can be viewed as
in situ fraction collection, as if protein bands were “cut out”
directly from the focal spot, like in gels or compartmentalized
electrolyzers. In contrast to these techniques, however, OMJCIEF enables online and fully automated transfer of isolated
fractions for hyphenated capillary-based MS or RPLC analyses.
It is also important to note that the analyte transfer can be
flexibly timed, increasing the capacity of OMJ-CIEF to integrate
into multidimensional separation.
pI Fractionation. IEF separation is commonly integrated
into hybrid analyses as a first-dimension technique to reduce
the complexity of biological samples. In proteomics studies,
prefractionation of protein and peptide mixtures by pI greatly
alleviates the identification process, especially with regard to the
discovery of post-translational modifications (PTMs).58−64
PTMs can be difficult to reveal in both MS and LC dimensions
if they have very little or even no effect on the m/z value and
retention time, respectively. In contrast, PTMs can be both
isolated and identified in the IEF domain based on measurable
modification-specific shifts induced in pI values.61 The
employment of IEF as a sample prefractionation technique
has substantially increased over recent years owing to the
progress in commercial IPG-IEF instrumentation, which has
enabled fraction recovery in the liquid phase.65−68 Liquid-phase
analyte recovery nicely fits into LC−MS workflows and
obviates tedious, error-prone analyte extraction and cleanup
procedures that are frequent in gel-based methods.
The multiple-junction interface endues OMJ-CIEF with high
potential for online analyte fractionation. The pH profile across
the separation capillary can be conveniently tailored via
pinpoint solvent exchange to emphasize specific pI regions,
and the focused fractions can then be selectively mobilized. It is
also worth mentioning that the possibility of external electrolyte
introduction via membrane interfaces in OMJ-CIEF increases
practical convenience of its operation. Analyte mixture can be
directly loaded for analysis without chemical pretreatment, and
pure water can be run as carrier solvent, eliminating possible
carryover effects associated with changing the separation buffer.
Figure 4 summarizes a series of experiments in which OMJCIEF was successfully employed to fractionate model protein
mixture according to designated pI cutoff values. The analyte
mixture contained seven proteins dissolved in water covering a
broad pI range: β-lactoglobulin A (pI 5.3), hemoglobin A (pI
6.8), myoglobin (pI 6.9−7.3), α-chymotrypsinogen A (pI 9),
ribonuclease A (pI 9.5), cytochrome c (pI 10−10.5), and
lysozyme (pI 11.3). In the first experiment, 200 mM aqueous
solution of histidine (His) was used as electrolytic buffer in all
external reservoirs. This high buffer concentration was chosen
to allow for the rapid diffusion of His through the Nafion
membranes. Small amphoteric His molecules enter the
separation volume, which is initially filled with pure water,
and modify its pH to their pI value of 7.6. Unlike ionic buffers
such as ammonium formate, His assumes zwitterionic form in
solution, and its density profile is therefore much less sensitive
to the external electrical field. Furthermore, the high buffering
capacity of His around its pI affords the stability of established
pH = 7.6, preventing the formation of gradients across the
capillary. As a result, the entire capillary volume serves as a pIdivider wall with a narrow cutoff value. In the external field,
proteins start to migrate along the fractionator, the direction
being dependent on their net charge at pH 7.6. Proteins with pI
< 7.6 will be negatively charged and will therefore migrate
downstream toward the positive pole (Figure 1), while proteins
Figure 4. OMJ-CIEF separation of a model protein mixture in
ampholytic His (left) and Arg (right) buffers. The analyte mixture
contained seven proteins covering a broad pI range: β-lactoglobulin A
(22 μM, pI 5.3), hemoglobin A (3 μM, pI 6.8), myoglobin (10 μM, pI
6.9−7.3), α-chymotrypsinogen A (31 μM, pI 9), ribonuclease A (58
μM, pI 9.5), cytochrome c (65 μM, pI 10−10.5), and lysozyme (56
μM, pI 11.3) dissolved in water. All the external reservoirs were filled
with the same electrolyte buffer (200 mM aqueous His or Arg), and
pure water was used as a carrier solution. IEF was carried over 1 h in
the constant current regime (5 μA). All the fractions were mobilized
for ESI-MS detection simultaneously via hydrodynamic pumping (3
μL min−1). For better visual perception EIC-MS profiles of the
proteins are plotted separately and normalized to their peak intensity.
with pI > 7.6 will acquire positive charge and move in the
opposite direction. As a result, the protein mixture will separate
into acidic (pI < 7.6) and basic (pI > 7.6) fractions located at
the opposite termini of the fractionator. Note that, due to the
absence of electroosmotic flow in OMJ-CIEF, analytes are
separated merely based on their charge (in other words, pI
value), rather than size-to-charge ratio, as in CZE. Fractionation
of the protein mixture in His buffer following the described
approach is shown in Figure 4, left. The acidic fraction
contained β-lactoglobulin A, hemoglobin A, and myoglobin,
which are negatively charged in His buffer and therefore
migrated downstream to the positive pole of the capillary. The
basic fraction contained α-chymotrypsinogen A, ribonuclease A,
cytochrome c, and lysozyme and was focused at the negative
pole of the fractionator. Despite the notable peak broadening
during analyte mobilization, the fractions arrived to the
detection point without overlapping (Figure 4, left), which
points to a large spatial cushion right after the focusing stage.
Indeed, the acidic (or the basic) fraction could be selectively
released when necessary via electrocapturing of the oppositecharged fraction.
The use of single-component fractionation buffer, such as
His, offers a number of benefits when CIEF is integrated into
multiple-stage analyses. Compared to heterogeneous ampholytic blends, the presence of singular His is easier to
discriminate against in both LC and MS domains, allowing
for more reliable identification of peptides and proteins.
Furthermore, His is a suitable buffer for tryptic digestion,69
and the released fractions can therefore be directly transferred
6860
dx.doi.org/10.1021/ac3013016 | Anal. Chem. 2012, 84, 6856−6862
Analytical Chemistry
Article
agreement with their predicted pI values (Figure 5). The
negatively charged acidic fraction (pI < 5) drifted to the
downstream terminus of the fractionator (RT ∼ 1.5 min), while
the basic fraction (pI > 7.6) drifted upstream (RT ∼ 5 min).
Interestingly, an intermediate fraction was formed (RT ∼ 3
min) containing peptides with pI values ∼6−7. Even though
these peptides are predicted to be negatively charged in His
buffer, their downward migration is probably hindered by more
abundant acidic peptides (pI < 5) occupying the very end of the
separation capillary. Note that both the peptide mixture and
carrier solution already contained His (5 mM) necessary for
IEF, unlike the experiments described in Figure 4, in which
pure water was used. For this reason, it was not necessary to
apply high His concentration in the external vials (5 mM in
Figure 5 vs 200 mM in Figure 4).
online to an immobilized enzyme microreactor (IMER)
without any purification steps. Whenever required, His can be
efficiently removed from analyte solution using online dialysis
interfaces with low molecular weight cutoff or chromatographic
separation.
Figure 4 (right column) illustrates IEF separation of the same
protein mixture performed in arginine (Arg) buffer. Arg is an
ampholytic electrolyte and, analogous to His, acts as a
discriminator wall with a sharp cutoff value equal to its pI
10.8. This higher cutoff results in a different partition, allowing
for the selective isolation of lysozyme, which is the most basic
protein in the mixture. The selective removal of most basic
proteins prior to ESI-MS analysis can sizably advance the
ionization efficiency of other constituents in the analyzed
mixture, which is often hindered in the presence of basic
fraction. On the contrary, preisolation of basic proteins via
removal of acidic fraction is of particular interest in top-down
proteomics approaches relying on electron capture or electron
transfer dissociation (ECD, ETD). Having the largest number
of amino groups, these proteins get most abundantly charged in
positive ion mode ESI-MS, which enables more comprehensive
fragmentation by ECD and ETD compared to the acidic
ones.70,71 For different cutoff values, an appropriate ampholyte
can be chosen from a variety of natural and synthetic products,
the major requirements being the low molecular weight to
allow facile diffusion through Nafion junctions, high buffering
capacity, and tolerance by the MS. For example, glutamic acid
(pI 3.1) could be considered as an option to perform
fractionation in the low-pH range.
Peptide Separation. Unlike microdialysis interfaces with
relatively high molecular weight cutoff values (∼3000 Da),
Nafion membranes have considerably smaller pores, which
allows analyses of smaller molecules, e.g. peptides, without
analyte leakage. Supporting Information Figure S4 and Figure 5
demonstrate the results of OMJ-CIEF separation performed in
5 mM His buffer on a mixture of peptides produced by tryptic
digest of bovine serum albumin (BSA). Altogether, 37 peptides
covering more than 60% of the protein sequence were
identified via searching the peptides’ MS/MS by Mascot. The
observed OMJ-CIEF separation of analyzed peptides is in good
■
CONCLUSION
The growing popularity of recently introduced IPG-IEF
fractionators (e.g., OFFGEL fractionator from Agilent) reflects
the steady demand for carrying separation of biological samples
in solution phase. Direct analyte recovery from solution
precludes the pitfalls associated with gel methods, such as
tedious and error-prone protocols for sample extraction and
purification. Owing to the capillary format of operation, OMJCIEF combines the merits of solution-phase separation with
the possibility of automated online sampling to MS, LC, CE,
and other capillary-based analyses. Advantageous to traditional
CIEF, the segmented interface of OMJ-CIEF enables in situ
manipulation over the fractionation process. The chemical
profile of the mobile phase can be tailored externally via
pinpoint solvent exchange through Nafion junctions, and
analytes focused in different segments of OMJ-CIEF can be
mobilized separately in time. Performed in partitioned volume
with autonomously controlled segments, OMJ-CIEF can be
viewed as a capillary-based analogue of compartmentalized
IPG-IEF electrolyzers. The good online compatibility of OMJCIEF with capillary-based separations renders it particularly
promising in high-throughput proteomics as a first-dimension
prefractionation unit orthogonal to LC and MS domains. Due
to its high loading capacity (>100 pmol), the device can be
useful in semi-online and off-line analyses as a preparative tool
for analyte purification, fractionation, and preconcentration
purposes. The high cost-efficiency and the ease of design and
operation may eventually promote the adoption of OMJ-CIEF
in a wide range of bioanalytical studies.
■
ASSOCIATED CONTENT
S Supporting Information
*
Additional information as noted in text. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Phone: +46 8 524 87594.
Notes
Figure 5. OMJ-CIEF separation of tryptic BSA digest. For the CIEF
analysis peptide mixture was prepared in 5 mM aqueous histidine. The
same buffer system was used as a carrier solution and electrolytic liquid
in external reservoirs. IEF was carried over 1 h in the constant current
regime (5 μA). All the fractions were mobilized for ESI-MS detection
simultaneously via hydrodynamic pumping at 3 μL min−1. Peak elution
time for the analyzed peptides was obtained from their EIC-MS profile
(Supporting Information Figure S4).
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
Financial support from VINNOVA (Grant EuroStar 2010) and
the Swedish Research Council (VR Grant 2011-3699) is
gratefully acknowledged.
6861
dx.doi.org/10.1021/ac3013016 | Anal. Chem. 2012, 84, 6856−6862
Analytical Chemistry
■
Article
(40) Chen, J. Z.; Lee, C. S.; Shen, Y. F.; Smith, R. D.; Baehrecke, E.
H. Electrophoresis 2002, 23, 3143.
(41) Storms, H. F.; van der Heijden, R.; Tjaden, U. R.; van der Greef,
J. Electrophoresis 2004, 25, 3461.
(42) Storms, H. F.; van der Heijden, R.; Tjaden, U. R.; van der Greef,
J. J. Chromatogr., B 2005, 824, 189.
(43) Wang, W. J.; Guo, T.; Rudnick, P. A.; Song, T.; Li, J.; Zhuang, Z.
P.; Zheng, W. X.; Devoe, D. L.; Lee, C. S.; Balgley, B. M. Anal. Chem.
2007, 79, 1002.
(44) Storms, H. F.; van der Heijden, R.; Tjaden, U. R.; van der Greef,
J. Rapid Commun. Mass Spectrom. 2006, 20, 3491.
(45) Li, M. X.; Liu, L.; Wu, J. T.; Lubman, D. M. Anal. Chem. 1997,
69, 2451.
(46) Kelly, J. F.; Ramaley, L.; Thibault, P. Anal. Chem. 1997, 69, 51.
(47) Simo, C.; Elvira, C.; Gonzalez, N.; Roman, J. S.; Barbas, C.;
Cifuentes, A. Electrophoresis 2004, 25, 2056.
(48) Ullsten, S.; Zuberovic, A.; Wetterhall, M.; Hardenborg, E.;
Markides, K. E.; Bergquist, J. Electrophoresis 2004, 25, 2090.
(49) Soga, T.; Ueno, Y.; Naraoka, H.; Ohashi, Y.; Tomita, M.;
Nishioka, T. Anal. Chem. 2002, 74, 2233.
(50) Garcia, M. C. J. Chromatogr., B 2005, 825, 111.
(51) Shariatgorji, M.; Astorga-Wells, J.; Ilag, L. L. Anal. Bioanal.
Chem. 2011, 399, 191.
(52) Astorga-Wells, J.; Jornvall, H.; Bergman, T. Anal. Chem. 2003,
75, 5213.
(53) Vollmer, S.; Astorga-Wells, J.; Bergman, T.; Jornvall, H. Int. J.
Mass Spectrom. 2007, 259, 73.
(54) Astorga-Wells, J.; Swerdlow, H. Anal. Chem. 2003, 75, 5207.
(55) Vollmer, S.; Astorga-Wells, J.; Alvelius, G.; Bergman, T.;
Jornvall, H. Anal. Biochem. 2008, 374, 154.
(56) Astorga-Wells, J.; Tryggvason, S.; Vollmer, S.; Alvelius, G.;
Palmberg, C.; Jornvall, H. Anal. Biochem. 2008, 381, 33.
(57) Astorga-Wells, J.; Vollmer, S.; Tryggvason, S.; Bergman, T.;
Jornvall, H. Anal. Chem. 2005, 77, 7131.
(58) Lengqvist, J.; Uhlen, K.; Lehtio, J. Proteomics 2007, 7, 1746.
(59) Heller, M.; Michel, P. E.; Morier, P.; Crettaz, D.; Wenz, C.;
Tissot, J. D.; Reymond, F.; Rossier, J. S. Electrophoresis 2005, 26, 1174.
(60) Eriksson, H.; Lengqvist, J.; Hedlund, J.; Uhlen, K.; Orre, L. M.;
Bjellqvist, B.; Persson, B.; Lehtio, J.; Jakobsson, P. J. Proteomics 2008,
8, 3008.
(61) Lengqvist, J.; Eriksson, H.; Gry, M.; Uhlen, K.; Bjorklund, C.;
Bjellqvist, B.; Jakobsson, P. J.; Lehtio, J. Amino Acids 2011, 40, 697.
(62) Cargile, B. J.; Bundy, J. L.; Freeman, T. W.; Stephenson, J. L. J.
Proteome Res. 2004, 3, 112.
(63) Tao, D. Y.; Sun, L. L.; Zhu, G. J.; Liang, Y.; Liang, Z.; Zhang, L.
H.; Zhang, Y. K. J. Sep. Sci. 2011, 34, 83.
(64) Heller, M.; Ye, M. L.; Michel, P. E.; Morier, P.; Stalder, D.;
Junger, M. A.; Aebersold, R.; Reymond, F. R.; Rossier, J. S. J. Proteome
Res. 2005, 4, 2273.
(65) Horth, P.; Miller, C. A.; Preckel, T.; Wenz, C. Mol. Cell.
Proteomics 2006, 5, 1968.
(66) Fraterman, S.; Zeiger, U.; Khurana, T. S.; Rubinstein, N. A.;
Wilm, M. Proteomics 2007, 7, 3404.
(67) Cargile, B. J.; Talley, D. L.; Stephenson, J. L. Electrophoresis
2004, 25, 936.
(68) Michel, P. E.; Reymond, F.; Arnaud, I. L.; Josserand, J.; Girault,
H. H.; Rossier, J. S. Electrophoresis 2003, 24, 3.
(69) Cologna, S. M.; Williams, B. J.; Russell, W. K.; Pai, P. J.; Vigh,
G.; Russell, D. H. Anal. Chem. 2011, 83, 8108.
(70) Syka, J. E. P.; Coon, J. J.; Schroeder, M. J.; Shabanowitz, J.;
Hunt, D. F. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 9528.
(71) Zubarev, R. A.; Horn, D. M.; Fridriksson, E. K.; Kelleher, N. L.;
Kruger, N. A.; Lewis, M. A.; Carpenter, B. K.; McLafferty, F. W. Anal.
Chem. 2000, 72, 563.
REFERENCES
(1) Dettmer, K.; Aronov, P. A.; Hammock, B. D. Mass Spectrom. Rev.
2007, 26, 51.
(2) Domon, B.; Aebersold, R. Science 2006, 312, 212.
(3) Kilar, F. Electrophoresis 2003, 24, 3908.
(4) Silvertand, L. H. H.; Torano, J. S.; van Bennekom, W. P.; de Jong,
G. J. J. Chromatogr., A 2008, 1204, 157.
(5) Kasicka, V. Electrophoresis 2010, 31, 122.
(6) El Rassi, Z. Electrophoresis 2010, 31, 174.
(7) Yang, L. Y.; Lee, C. S.; Hofstadler, S. A.; Pasa-Tolic, L.; Smith, R.
D. Anal. Chem. 1998, 70, 3235.
(8) Jensen, P. K.; Pasa-Tolic, L.; Anderson, G. A.; Horner, J. A.;
Lipton, M. S.; Bruce, J. E.; Smith, R. D. Anal. Chem. 1999, 71, 2076.
(9) Jensen, P. K.; Pasa-Tolic, L.; Peden, K. K.; Martinovic, S.; Lipton,
M. S.; Anderson, G. A.; Tolic, N.; Wong, K. K.; Smith, R. D.
Electrophoresis 2000, 21, 1372.
(10) Zhang, C. X.; Xiang, F.; Pasa-Tolic, L.; Anderson, G. A.;
Veenstra, T. D.; Smith, R. D. Anal. Chem. 2000, 72, 1462.
(11) Simpson, D. C.; Smith, R. D. Electrophoresis 2005, 26, 1291.
(12) Tang, Q.; Harrata, A. K.; Lee, C. S. Anal. Chem. 1997, 69, 3177.
(13) Lecoeur, M.; Gareil, P.; Varenne, A. J. Chromatogr., A 2010,
1217, 7293.
(14) Jensen, P. K.; Harrata, A. K.; Lee, C. S. Anal. Chem. 1998, 70,
2044.
(15) Wei, J.; Yang, L. Y.; Harrata, A. K.; Lee, C. S. Electrophoresis
1998, 19, 2356.
(16) Yang, L. Y.; Tang, Q.; Harrata, A. K.; Lee, C. S. Anal. Biochem.
1996, 243, 140.
(17) Martinovic, S.; Berger, S. J.; Pasa-Tolic, L.; Smith, R. D. Anal.
Chem. 2000, 72, 5356.
(18) Lecoeur, M.; Goossens, J. F.; Vaccher, C.; Bonte, J. P.; Foulon,
C. Electrophoresis 2011, 32, 2857.
(19) Gorg, A.; Weiss, W.; Dunn, M. J. Proteomics 2004, 4, 3665.
(20) Giddings, J. C. J. Chromatogr., A 1995, 703, 3.
(21) Francois, I.; Sandra, K.; Sandra, P. Anal. Chim. Acta 2009, 641,
14.
(22) Wang, Y. J.; Balgley, B. M.; Lee, C. S. Expert Rev. Proteomics
2005, 2, 659.
(23) Guo, T.; Rudnick, P. A.; Wang, W. J.; Lee, C. S.; Devoe, D. L.;
Balgley, B. M. J. Proteome Res. 2006, 5, 1469.
(24) Wang, W. J.; Guo, T.; Song, T.; Lee, C. S.; Balgley, B. M.
Proteomics 2007, 7, 1178.
(25) Wang, Y. J.; Balgley, B. M.; Rudnick, P. A.; Evans, E. L.; DeVoe,
D. L.; Lee, C. S. J. Proteome Res. 2005, 4, 36.
(26) Wang, T. T.; Ma, J. F.; Wu, S. B.; Yuan, H. M.; Zhang, L. H.;
Liang, Z.; Zhang, Y. K. Electrophoresis 2011, 32, 2848.
(27) Zhou, F.; Hanson, T. E.; Johnston, M. V. Anal. Chem. 2007, 79,
7145.
(28) Chen, J. Z.; Balgley, B. M.; DeVoe, D. L.; Lee, C. S. Anal. Chem.
2003, 75, 3145.
(29) Tang, Q.; Harrata, A. K.; Lee, C. S. Anal. Chem. 1995, 67, 3515.
(30) Clarke, N. J.; Naylor, S. Biomed. Chromatogr. 2002, 16, 287.
(31) Tang, Q.; Harrata, A. K.; Lee, C. S. Anal. Chem. 1996, 68, 2482.
(32) Mokaddem, M.; Gareil, P.; Varenne, A. Electrophoresis 2009, 30,
4040.
(33) Zhong, X. F.; Maxwell, E. J.; Ratnayake, C.; Mack, S.; Chen, D.
D. Y. Anal. Chem. 2011, 83, 8748.
(34) Zhou, F.; Johnston, M. V. Anal. Chem. 2004, 76, 2734.
(35) Lamoree, M. H.; Tjaden, U. R.; vanderGreef, J. J. Chromatogr., A
1997, 777, 31.
(36) Lamoree, M. H.; van der Hoeven, R. A. M.; Tjaden, U. R.; van
der Greef, J. J. Mass Spectrom. 1998, 33, 453.
(37) Yang, L. Y.; Lee, C. S.; Hofstadler, S. A.; Smith, R. D. Anal.
Chem. 1998, 70, 4945.
(38) Sun, L. L.; Duan, J. C.; Tao, D. Y.; Liang, Z.; Zhang, W. B.;
Zhang, L. H.; Zhang, Y. K. Rapid Commun. Mass Spectrom. 2008, 22,
2391.
(39) Zhou, F.; Johnston, M. V. Electrophoresis 2005, 26, 1383.
6862
dx.doi.org/10.1021/ac3013016 | Anal. Chem. 2012, 84, 6856−6862