This is a preprint version of our artical published in Journal of Proteomics
Advances in purification and separation of posttranslationally modified proteins
by Martin Černý, Jan Skalák, Hana Černá and Břetislav Brzobohatý
See Journal of Proteomics for the final version:
http://www.sciencedirect.com/science/article/pii/S1874391913003254
10.1016/j.jprot.2013.05.040
A preprint is an author’s own write-up of research results and analysis
that has not been peer-reviewed, nor had any other value added to it by
a publisher (such as formatting, copy editing, technical enhancement
etc...).
Title: Advances in purification and separation of posttranslationally modified proteins
Authors:
Martin Černý, Jan Skalák, Hana Cerna, Břetislav Brzobohatý
Affiliation:
Department of Molecular Biology and Radiobiology, Mendel University in Brno & CEITEC – Central European
Institute of Technology, Mendel University in Brno, Zemědělská 1, CZ-613 00 Brno, Czech Republic
Corresponding author:
Martin Černý
Department of Molecular Biology and Radiobiology
Mendel University in Brno & CEITEC MENDELU
Zemědělská 1
CZ-613 00 Brno
Czech Republic
+420 545 133 374
martincerny83@gmail.com
Posttranslational modifications (PTMs) of proteins represent fascinating extensions of the dynamic complexity
of living cells’ proteomes. The results of enzymatically catalyzed or spontaneous chemical reactions, PTMs form
a fourth tier in the gene – transcript – protein cascade, and contribute not only to proteins’ biological functions,
but also to challenges in their analysis. There have been tremendous advances in proteomics during the last
decade. Identification and mapping of PTMs in proteins have improved dramatically, mainly due to constant
increases in the sensitivity, speed, accuracy and resolution of mass spectrometry (MS). However, it is also
becoming increasingly evident that simple gel-free shotgun MS profiling is unlikely to suffice for comprehensive
detection and characterization of proteins and/or protein modifications present in low amounts. Here, we review
current approaches for enriching and separating posttranslationally modified proteins, and their MS-independent
detection. First, we discuss general approaches for proteome separation, fractionation and enrichment. We then
consider the commonest forms of PTMs (phosphorylation, glycosylation and glycation, lipidation, methylation,
acetylation, deamidation, ubiquitination and various redox modifications), and the best available methods for
detecting and purifying proteins carrying these PTMs.
Key words (6):
posttranslational modification; PTM; protein enrichment; PTM detection; proteome; top-down proteomics.
1. Introduction
1.1. So few, and yet so many
Proteomes are expected to be two to three orders of magnitude more complex than would be predicted from
numbers of protein-encoding genes present in the respective genomes. There are over 300 known naturally
occurring amino acids, but only 64 codons in the genetic code and the actual number of proteinogenic amino
acids is even lower. Altogether, 22 amino acids have been proven to be encoded by DNA. However, after protein
translation these amino acids may undergo further modifications, which considerably increase the diversity of
proteins present in living cells. These modifications can be either transient or permanent, and may result from
either targeted, enzymatically catalyzed reactions or spontaneous chemical reactions in the cell. The Unimod
Database (http://www.unimod.org; 10/2012) lists almost 1000 different protein modifications that have been
detected in mass spectrometric analyses of proteins, but some of them could be artifactual, like methionine and
cysteine oxidations during two-dimensional electrophoresis (2-DE), carbamylation by urea on any free amino
group in the protein sample (at N-termini, or side chains of arginine or lysine), and modifications introduced by
MS analysis (e.g. conversion of phosphoserine to dehydroalanine with neutral loss of phosphoric acid). The
RESID database hosted by the European Bioinformatics Institute, EBI (http://www.ebi.ac.uk/RESID/; 10/2012),
compiles information on protein posttranslational modifications (PTMs) found in nature and its release 70.01
(08/2012) lists 591 entries. Covalent protein modifications can be classified as a covalent addition of some
chemical group, or a covalent cleavage of peptide backbones [1]. Side chain modifications are known for all of
the proteinogenic amino acids except Ala and Pyr, and all can undergo crosslink reactions and/or N- or Cmodification
if they are at the N- or C-terminus, respectively, of their respective proteins (Fig. 1).
<-Figure 1->
2. Methods applied in PTM analyses
2.2. Protein separation or “Thou shalt not use only bottom-up proteomic approaches”
Widely used MS-based approaches in so-called “bottom-up proteomics” (involving digesting samples by
selected proteases, MS analysis of the resulting peptides and finally assembly of the identified sequences) are
excellent for high-throughput protein identification. However, some information is generally lost when proteins
are cleaved into fragments by the chosen proteases prior to MS analysis in typical protocols. Due to the vast
ranges of concentrations of proteins in cells and differences in ionization potentials of peptides, it is always
challenging to obtain full sequence coverage, and even then it is very difficult to distinguish peptides from partly
cleaved protein isoforms. Thus, in bottom-up approaches some of the diversity of isoforms that may result from
a single initially translated protein is inevitably lost (Fig. 2). However, a PTM generally changes a protein’s
chemical and physical characteristics (e.g. molecular weight, shape, charge, pI, hydrophobicity and interactions
with other proteins or ligands). Thus, proteins carrying PTMs can be separated, and the PTMs can be identified,
mapped and characterized by using techniques that exploit these changes.
<-Figure 2->[2]
2.2.1. Size does matter
Protein PTM by a cleavage may result from autocatalytic activity, protease-mediated regulation, or nonenzymatic
protein damage [3]. This is theoretically the easiest kind of PTM to detect, because the average
molecular weight of an amino acid residue is 110 Da [4], hence losses of one or more can be readily detected by
methods separating proteins according to their mass. Apart from mass spectrometry and gradient
ultracentrifugation, there are two suitable techniques for estimating the mass of a protein and separating cleaved
protein isoforms. The first is size-exclusion chromatography (SEC), also called gel filtration. In SEC samples are
introduced in solution to a column containing a solid phase consisting of cross-linked polymer beads with pores
or cavities that large proteins cannot enter. Thus, they pass rapidly through the column around the beads. In
contrast, smaller proteins are retarded to size-related degrees. This technique can fractionate proteins covering a
wide mass range (0.1-100 kDa), depending on the pores’ size [5]. The second option is polyacrylamide gel
electrophoretic separation in the presence of the denaturant sodium dodecyl sulfate (SDS-PAGE).
Approximately one SDS molecule binds per two amino acids, which masks the native charge of a protein and the
resulting negative net charge is proportional to the protein’s mass. During the electrophoresis, the gel acts as a
molecular sieve so the rate of migration of proteins in a sample is inversely related to their mass (high molecular
weight proteins are retarded more than smaller proteins). The original setup presented by Laemmli [6] is not
suitable for separating low molecular weight proteins (<10 kDa), but there are modifications which overcome
this limit [7]. The resolution of both separation techniques allows the estimation of a protein’s molecular mass
with kDa precision, which means that a loss of ten amino acid residues results in a detectable shift in a protein’s
mobility. However, while a protein’s mobility may be mainly dependent on its mass, shifts may be also due to
intramolecular changes. For example, protein phosphorylation can induce calcium ion binding, which reduces
SDS-PAGE mobility [8].
2.2.2. Isoelectric focusing
During isoelectric focusing (IEF), charged proteins migrate through a pH gradient. The net charge of a protein is
the sum of all the negative or positive charges of the amino acid side chains, which depends on the pH
environment and is zero at the protein’s isoelectric point (pI). Thus, electrophoretic migration of a protein stops
at its pI and all proteins are focused around their respective pI. This is only possible when the three-dimensional
protein structure is disrupted by chaotropic agents such as urea, and all charged residues are exposed to the
medium. Basic and acidic amino acid residues that contribute to a protein’s net charge are often sites of PTMs,
and the PTMs often change their charge. For example, acetylation masks positive charge of lysine residues,
while phosphorylation or deamidation introduces an additional negative charge to a protein. Thus, PTMs are
reflected in changes in a protein’s pI and can be detected by IEF. In addition to gel-based IEF, which is suitable
for analytical studies (see 2-DE), IEF can also be applied in preparative mode. Liquid-phase IEF can be used in
preparative analyses, or for separating protein samples into fractions over a specific pI range. This technique has
lower resolution than gel-based IEF and suffers from pH drifts, but it can separate membrane and hydrophobic
proteins, as well as large protein complexes [9]. Off-gel electrophoresis, combining gel and liquid phase modes,
provides better performance than liquid-phase IEF, but transferring samples from the gel to the liquid phase can
be problematic [10]. Another option is “free flow” electrophoresis, which can also be applied in IEF mode. The
method originally developed by Hannig [11] separates protein samples within a thin film of electrolyte that flows
laminarly between two parallel cooling plates. After an electric field is applied (perpendicular to the flow),
proteins are deflected from the direct flow based on their charge and separated into fractions.
2.2.3. Two-dimensional electrophoresis (2-DE)
Two-dimensional gel electrophoresis has been one of the most successful methods for detecting and analyzing
protein PTMs. The method introduced in 1975 [12, 13] combines an IEF separation in the first dimension and an
SDS-PAGE separation in the second dimension. A gel with proteins focused by IEF is treated with SDS to cover
their native charges, then SDS-PAGE is applied to separate them on the basis of their molecular weights in a
plane perpendicular to the plane of the first separation (Fig. 3, A). The reproducibility of 2-DE depends largely
on the IEF step and has been significantly improved by using an immobilizing pH gradient [14]. Theoretically,
2-DE offers unsurpassed resolution and a standard large gel (~ 20 × 20 cm) should be able to resolve over 10,000
protein spots. In practice, a protein present in amounts below 0.5 ng is difficult to detect, even with the most
sensitive detection stains (silver or fluorescent total-protein stains) and the detection limit of the widely used
colloidal Coomassie dye is around 10 ng [7]. Further, PTMs increase the number of spots originating from a
single protein, thus quantitative 2-DE analysis is usually restricted to sets of at most a few thousand proteins.
The greatest potential of 2-DE is in qualitative analysis. The change in the mass of a protein due to a PTM is
often too small to be easily detected by standard SDS-PAGE. However, many PTMs introduce charged groups
into the protein, which leads to a detectable change in the position of the protein on a 2D gel. It has been shown
that IEF in narrow pH ranges can separate proteins that differ in pI values by just 0.001 to 0.01 pH units [15]. In
addition, comparisons of theoretical and observed protein pI and molecular masses can indicate spliced isoforms
as well as proteins with PTMs (Fig. 3, B). Modified proteins can also be visualized in-gel or after Western
blotting using specific stains or antibodies, which are available for all major PTMs. However, in addition to the
scale limitation mentioned above, 2-DE approaches have several drawbacks. One is that they require roughly
100-fold larger samples than routine LC-MS analysis. A second problem lies in the IEF step. Hydrophobic
proteins, large proteins (> 150 kDa), and proteins with extreme pI values cannot easily enter the gel and may be
lost during this step. Furthermore, very small proteins can be washed away during protein equilibration with
SDS. It is also very important to run IEF at an appropriate temperature, generally around 20 °C, otherwise the
presence of urea may lead to artificial carbamylation modification of the proteins in the sample. The presence of
excessive amounts of salts and buffer ions in the sample can also cause sharp temperature rises, which can
promote carbamylation reactions, or even in extreme cases burn the gel strip. Further, IEF is time-consuming and
some proteins may become unstable when focused too long, e.g. cysteine residues may be oxidized [16]. The
problems associated with separating membrane proteins and proteins with extreme pI values can be
circumvented by replacing IEF with blue native (BN) PAGE [17], in which proteins in a sample interact with
Coomassie brilliant blue G-250 dye before their separation. Binding of the dye results in a negative surface
charge, which is essential for electrophoretic separation, but does not denature proteins so their migration (and
separation) is influenced not only by their mass (as in SDS-PAGE), but also by their net surface and molecular
shape.
2.2.4. Liquid chromatography (LC)
Analysis by 2-DE is time consuming. The lack of automation and the necessity to extract proteins prior to MS
analysis led to the development of alternative methods based on liquid chromatography. A two-dimensional
column-based liquid chromatographic technique for resolving complex mixtures was initially introduced in 1978
[18], but its first application in proteomics dates to 1990 [19]. Since then the methodology has been further
improved and accelerated. For example, while a separation by the original set-up took six hours, the method
presented by Stoll et al., (2006) requires only 30 min and provides ca. 100-fold improvement in peak capacity
(the number of peaks/protein species that can be separated from one another)[20]. In standard 2D-LC proteomic
separation (Fig. 3, C), proteins are separated according to their charge in the first dimension, much like IEF,
except that the pI axis is in bands instead of continuous pI increments. This is done by chromatofocusing, which
replaced previously used ion-exchange columns (e.g. strong cation exchange – SCX columns). A pH gradient is
created as the eluting buffer, composed of a large number of buffering substances (polybuffer), titrates a weak
anion exchanger chromatography medium. The protein with the highest pI elutes first and no protein is thus
exposed to a higher pH than its own pI. In contrast to IEF, this method can separate large proteins and
hydrophobic proteins, as well as smaller and hydrophilic proteins [21]. Reversed-phase liquid chromatography is
usually used for the second dimension separation, which separates the proteins by hydrophobicity. There are
numerous possible combinations of LC techniques and theoretically they could all be combined for
multidimensional separation. In practice, mobile phase compatibility restricts the number of on-line modes that
are suitable for proteomic analysis. The compatibility restrictions can be lessen by employment of an off-line
setup in which collected fractions can be either concentrated by evaporating the buffer or diluted prior to
separation in the following dimension [22]. However, only a few successful examples of this approach have been
reported. The combination of chromatofocusing and size-exclusion chromatography is, in principle, similar to 2DE
separation, but it only provides comparable resolution over pH intervals of 3 pH units or less. Like 2-DE,
multidimensional chromatography is capable of separating proteins carrying PTMs (e.g. [23]). However, it has
been more commonly used for peptide fractionation prior to MS analysis, because there is no available technique
for visualizing specific PTMs separated by multidimensional chromatography (unless affinity chromatography
can be used) that provides comparable sensitivity and resolution to gel staining or Western blotting.
2.2.5. Diagonal separation
In the earliest examples of diagonal separation paper electrophoresis was used to characterize disulfide bonds in
peptides[24]. Briefly, mixtures of proteins were separated on a paper strip, treated with performic acid to break
disulfide bonds by oxidizing cysteine residues to cysteic acid, and then subjected to a second dimension
separation. Peptides that had identical electrophoretic mobility lay along a diagonal line following the twodimensional
separation, while peptides whose disulfide bonds were disrupted lay off the diagonal. This method
was later adapted for separating intact proteins in polyacrylamide gels [25], using a similar protocol to 2-DE
expect that the first-dimension gels are also SDS-denaturing (Fig. 3, D). After separation, proteins containing no
interpolypeptide disulfide bond(s) lie in a diagonal line, whereas disulfide-linked protein complexes are
separated into individual components and resolved below the diagonal. The reduction of intrachain disulfide
bonds retards migration in the second dimension and the resulting protein spots appear above the diagonal [26].
Diagonal separation has also been extended to LC applications, which are most suitable for analyzing peptides
with PTMs, but can also be adapted for proteins with PTMs. In a typical experiment, a sample is fractionated by
HPLC, fractions are collected, dried, treated to modify a residue or PTM and again subjected to HPLC analysis.
This technique has been used to study modifications including phosphorylation, nitration and N-glycosylation
[27-30]. The latest development in this field is an automated enzyme-based approach involving capillary
electrophoresis, in which the distal end of the first capillary incorporates an enzyme-based microreactor [31].
Obviously, it is currently designed to detect modified peptides, because the immobilized enzyme requires
unhindered access to the site of the PTM, but it shows promising potential for further development.
<-Figure 3->
2.2.6. Affinity separation and enrichment
The separation of proteins by means of the electrophoresis or chromatographic methods described in the
previous section is based on their mass, charge or hydrophobicity. Thus, protein PTMs can be detected by the
shift in net mass/charge/hydrophobicity they introduce to the modified protein. This approach is excellent if a
PTM is not known, the proteins of interest are abundant, or if there is no available enrichment technique. In other
cases, enrichment is highly advisable, and the main available enrichment techniques are described below.
2.2.6.1. Immunoaffinity
PTM-specific antibodies can be raised against virtually any protein that can be purified. The Human Protein
Atlas Project has provided a massive array of antibodies, covering target proteins encoded by over 14,000 human
genes (http://www.proteinatlas.org/;10/2012). However, it is difficult to obtain specific antibodies targeting only
a PTM group and no other protein epitopes, although this problem can be circumvented (to some degree) by
raising antibodies against synthetic peptides bearing PTMs or artificial substances that mimic the structure of a
targeted PTM [32]. Well-designed anti-peptide antibodies provide much (but not all, as discussed below) of the
specificity required for purifying protein families, e.g. substrates of a specific kinase [33]. Commercially
available antibodies cover several PTMs, including phosphorylation, acetylation, glycosylation, myristoylation,
prenylation, sumoylation and ubiquitination. They are generally either cross-linked to a resin or labeled with a
tag that facilitates their retrieval after a modified protein is captured and they can be used for either
immunoprecipitation or in immunoaffinity columns. For example, proteins with N-trimethylated lysine residues
can be purified by biotinylated antibodies (e.g. ImmuneChem; http://www.immunechem.com).
2.2.6.2. Ligand affinity enrichment
Some PTMs have an affinity for metal or inorganic ions. For example, immobilized Ga ions can separate
sulfated peptides [34] and several combinations of metal ions and metal oxides are widely used for
phosphoproteome analysis (see 3.1.2.). Naturally occurring PTMs often participate in cell signaling and proteinprotein
interactions. Thus, it is also possible to employ innate protein-ligand interaction systems to bind
modified proteins, notably lectin-based affinity columns for glycoproteome enrichment, and media incorporating
a ubiquitin-binding domain for ubiquitinome analyses (see respective parts of this review).
2.2.6.3. Derivatization
In cases where there is no suitable affinity enrichment technique, a PTM can sometimes be labeled with a tag in
vivo, using the native cellular machinery to incorporate modified biomolecules such as tagged carbohydrates or
lipids [35]. In more elaborate approaches for in vitro tagging, purified recombinant enzymes or chemoselective
reactions can be used. Introduced labels can serve for detection (e.g. fluorophores) or isolation, usually via
biotin-avidin affinity enrichment of biotin-tagged proteins/peptides. The purity of enriched PTM proteins can be
further increased by including a cleavable linker in biotin tags, thus allowing bound proteins to be released
without risking contamination by natively biotinylated proteins or a need for harsh denaturing conditions that
may lead to the elution of non-selectively bound proteins [36]. Chemical tagging procedures are available for all
major PTMs, including acetylation, acylation, prenylation, phosphorylation and gylcosylation ([35] and
references therein).
2.3. Deeper mining to detect hidden proteins and PTMs
The past decade has seen huge advances in proteomics. There is no doubt that this is mainly thanks to interest in
the human proteome and its apparent potential for medical applications, but all sectors of proteomics are rapidly
advancing. For example, there were less than 1000 experimentally identified gene products of the model plant
Arabidopsis thaliana in 2004 [37], but the Arabidopsis proteome pep2pro database now includes proteomic data
for over 14 500 proteins (excluding proteins with only single hits) (http://fgcz-pep2pro.uzh.ch; [38]). Ordinary
proteomic experiments based on 2-DE and liquid chromatography do not usually cover more than a fifth of this
number, and it is estimated that alternative splicing and PTMs increase the number of different proteins to at
least 100 000. Thus, the most comprehensive studies, covering several thousands of identified proteins, are only
scratching the surface of the whole proteome. This is partly due to the presence of several highly abundant
proteins, which interfere with the detection of less abundant proteins. Examples include: collagen, actin, myosin
and keratin in many animal tissue samples; albumin, which accounts for up to 60 % of plasma proteomes; and
RuBisCO, which may account for more than 40 % of the protein content in some plant tissues [39]. In marked
contrast, proteins and many PTMs are frequently below the sensitivity threshold of available analytical methods,
and this problem cannot (generally) be solved by crude fractionation, tissue sampling or organelle purification.
2.3.1. Immunoaffinity depletion
Several approaches are used to deplete abundant proteins in samples and increase the proteome coverage. The
first option is to use immunoaffinity separation. Commercially available kits offer immunoaffinity partitioning of
highly-abundant proteins, for example from plasma/serum proteomes, with antibodies for up to 20 proteins.
Further, moderately abundant proteins can be further depleted by antibodies obtained from immunizing chickens
with plasma proteins from which the highly-abundant proteins have been removed. This two-step procedure
theoretically provides a fraction of low-abundance proteins (in the flow-through) and one of moderately
abundant proteins (in the eluted fraction) [40]. Of course, this approach has limits and the price of mixed
antibody columns is quite considerable.
2.3.2. Proteome equalization
The second widely used option is based on peptide affinity subtraction. This technique has been in development
for over two decades, the first reports dating back to the early 1990s [41]. Its commercialized versions employ a
bead-based combinatorial peptide ligand library (CPLL), each unique ligand theoretically binding a unique
protein epitope. In theory, when a complex biological sample is applied to the beads, highly abundant proteins
saturate their high-affinity ligand or ligands, and excess protein is washed away. In contrast, low-abundance
proteins are concentrated on their specific affinity ligand or ligands. The major targets of peptide affinity
subtraction are plasma and serum proteomes, but it can be used to reduce the dynamic range of proteomes from
any complex biological samples [39, 42]. It provides a relatively cheap alternative to immunoaffinity separation
that has several advantages, including cross-sample/species applications. The decomposition of affinity beads
also poses much minimal risks of sample contamination, unlike antibody degradation. Furthermore, samples can
be further divided by sequential elution [43]. On the other hand, results of CPLL extractions depend on proteinligand
affinity interactions and are highly unpredictable. Jiang et al. (2010) have even presented indications that
this method does not improve the quantitative results of proteomic analyses [44], but this may have been due to
their use of a gel-free approach, because we have found that 2-DE benefits from a CPLL application (Fig. 4).
<-Figure 4->[45]
3. Major protein PTMs found in proteomic studies
Many protein modifications have a significant physiological/biological impact. Both enzymatic and nonenzymatic
(chemical) PTMs occur, and with limited knowledge of their origins it is difficult to correctly
distinguish their nature. Historically, the most thoroughly studied type of PTM is phosphorylation, largely due to
the relative ease of detecting protein O-phosphorylation in vivo and in vitro. While the enzymatic origin of Ophosphorylation
was soon accepted, N-phosphorylation was long considered an experimental artifact or a
product of non-enzymatic activity. The existence of N-kinases was recognized much later [46, 47], and the
possibility that this kind of PTM could occur spontaneously has still not been ruled out. A similar controversy
surrounds other PTMs, for example protein S-ubiquitination, which is also considered to be non-enzymatic [48].
We are still far from fully understanding functions of individual gene products and it is likely that future
discoveries will lead to the reevaluation of enzymatic and non-enzymatic PTMs. Some PTMs are clearly results
of protein ageing, especially in proteins that are slowly metabolized, and thus are exposed for longer times to
reactive endo- and exogenous compounds. However, following discovery of the role of oxidative damage in cell
signaling, it is tempting to speculate that at least some of these modifications could be of enzymatic origin, even
if it is through an indirect effect of metabolite production and channeling. In the following text we summarize
techniques for detecting and enriching protein PTMs, focusing on their applicability for large-scale proteomic
analysis. The selected PTMs (Fig. 5) represent not only the most commonly studied protein modifications, with
numbers of entries in the UniProt database [49] reaching up to 10 % of all proteins currently reviewed by its host
consortium, but each presents specific problems, and requires an appropriately adjusted proteomics approach for
detection and enrichment. We do not strictly separate PTMs usually classified as enzymatic from non-enzymatic
PTMs, partly because the methodology for a given type of PTM is easily transferable and partly because a nonenzymatic
modification may result in loss of an enzymatic PTM, as believed e.g. for N-phosphate
dephosphorylation or protein demethylation (see respective parts of the review).
<-Figure 5->[50, 51]
3.1 The phosphoproteome
In 1906 Leven and Alsberg discovered phosphorylation of the egg yolk protein vitellin. It took nearly 30 years to
track down this phosphorylation to serine residues and a further 20 years to discover that it is actively catalyzed
by enzymes [52]. Today, protein phosphorylation is probably the most intensively studied PTM. Numerous
studies have demonstrated that reversible phosphorylation is a key posttranslational modification, and is
involved in nearly all cell processes. It has been shown to participate in signaling and the regulation of diverse
enzyme activities, protein-protein interactions and protein targeting (e.g. [53-58]). Loss of phosphorylation is
also known to render some proteins susceptible to degradation, thus regulating their lifespans [59, 60]. The aim
of so-called phosphoproteomics is to study protein phosphorylation by identifying phosphoproteins, quantifying
phosphorylation, precisely mapping phosphorylation sites, and ultimately revealing their biological functions
[61]. Potential phosphorylation sites are estimated to be approximately three times more numerous than genes
and a third of expressed proteins are assumed to be phosphorylated at any given time. The largest study of the
humans phosphoproteome detected 20 443 phosphosites, and the Arabidopsis phosphoprotein database
PhosPhAt (http://phosphat.mpimp-golm.mpg.de; 10/2012; [62]) lists 12 613 unique phosphosites in 5663
phosphoproteins. Regulation of such extensive and dynamic phosphorylation states requires about 3 % of
human genes and 5 % of the genes in Arabidopsis thaliana, encoding about 1100 protein kinases and 100 to 200
protein phosphatases [63]. Eight amino acids can be phosphorylated (Fig. 5, A), but in eukaryotic cells, protein
kinases and phosphatases are believed to act mainly on Ser, Thr and Tyr residues, typically resulting in
pSer/pThr/pTyr ratios of ca. 1800/200/1. The stability of phosphorylation at the other amino acids is lower.
Phosphoramidates, i.e. phosphohistidine, phosphoarginine and phospholysine, mostly go undetected in
conventional studies of protein phosphorylation because of the instability of the phosphate-nitrogen bond in
acidic solutions [64, 65]. However, reversible His phosphorylation is estimated to occur in a pHis/pTyr ratio of
10/1-100/1 [63]. The stability of acyl-phosphates is even lower, but they play a role for instance in signaling
pathways. While protein phosphorylation is considered to be solely catalyzed by enzymes, protein
dephosphorylation may be either enzymatic or non-enzymatic, and the latter may be simply related to protein
ageing. However, losses of phosphorylations, especially temperature and pH-labile phosphorylations, is also
considered to participate in fast intrinsic responses to stimuli [64].
3.1.1. Visualization of phosphorylated proteins
Radioisotopes have been used since the beginning of phosphorylation research and experiments with
radiolabelled ATP have confirmed that protein phosphorylation is catalyzed by enzymes. This previously widely
used method is based on use of [32
P] labeled J-ATP as a substrate for in vitro kinase reactions followed by
separation of labeled phosphorylated peptides by thin layer chromatography (TLC). Many phosphoproteins can
be visualized by autoradiography, but they are not usually present in sufficient amounts to be identified [52, 66].
In addition, radiolabeling in vivo is problematic and cannot be used to follow natural phosphoproteome
dynamics. Legal and safety restrictions further limit the use of radiolabeled substances and this methodology is
currently in decline, being replaced by immunodetection or fluorescent staining. Increasing numbers of
commercially available antibodies can be used for phosphoprotein detection in Western blot analysis (suppliers
include Qiagen (http://www.qiagen.com), Santa Cruz Biotechnology (http://www.scbt.com/) and Life
Technologies (http://www. lifetechnologies.com). Their use suffers from all disadvantages of antibody-based
methods, i.e. they are rather expensive and have questionable specificity. Further, anti-phospo-PTM antibodies
are only currently available for pTyr, pSer and pThr. The first fluorescent stain that allowed direct, in-gel
detection of phosphate groups attached to tyrosine, serine, or threonine residues (without the need for antibodies
or radioisotopes) was Pro-Q Diamond (Life Technologies). It can be used to label proteins in standard SDSpolyacrylamide
gels or 2-D gels and the staining procedure has been optimized [67]. The second florescent dye
is based on Phos-tag (http://www.phos-tag.com/; [68]), an alkoxide-bridged dinuclear metal (usually Zn2+
and
Mn2+)
complex. These tags preferentially capture phosphomonoester dianions bound to Ser, Thr and Tyr
residues. The detection is based either on a mobility shift of phosphorylated proteins in SDS-PAGE
(polyacrylamide-bound tags) or tag-reporter conjugates [61, 69]. The fluorescent Phos-tag™ Phosphoprotein Gel
Stain produced by Perkin Elmer (www.perkinelmer.com) exploited this technique, but it is no longer available.
The only Phos-tag based detection kit currently on the market is a biotin-tagged version for Western blotting,
Wako Pure Chemical Industries (http://www.wako-chem.co.jp). These phosphospecific stains have certified
ability to detect pThr, pSer and pTyr, but can also detect pAsp and N-phosphates [70]. The detection limit of
florescent stains is reportedly 1-16 ng of phosphoprotein; 5-20× the sensitivity of techniques based on chemical
derivatizations, e.g. by the GelCode Phosphoprotein Staining Kit (Pierce, www.piercenet.com). Phosphospecific
stainings share several of the disadvantages of phospho-specific antibodies: their prices are high and their
specificity must be usually validated by normalization to a negative control. Furthermore, phosphorylation in
proteins detected by phosphostaining is often not detectable by MS analysis, indicating that the stains yield false
positives, although this could be at least partly due to their ability to mark labile N-phosphates, which escape
detection by conventional approaches [65].
3.1.2. Phosphoproteome enrichment by affinity chromatography
The method of choice for identifying phosphorylation site is mass spectrometry, which has replaced Edman
degradation. However, the abundance of many signaling proteins is low in cells (even relative to the attomolar
sensitivity of modern MS systems), and the proportions of these proteins that are phosphorylated at any given
time may also be quite low. Therefore, in a standard LC-MS/MS experiment, the chance of detecting and
sequencing many phosphopeptides is quite low [52]. A number of methods to enrich the phosphorylated fraction
in samples and thus overcome these problems have been developed. Immobilized metal ion affinity
chromatography (IMAC) was originally introduced, in 1975, for purifying native proteins with an intrinsic
affinity for metal ions [71] and was focused on isolating His-tagged proteins. The ability of phosphoserine (in
phosphorylated ovalbumin) to bind selectively to immobilized iron was demonstrated in 1986 [72]. This is based
on the affinity of the negatively charged phosphate group for metal ions (usually Fe3+
or Ga3+
), but they cochelate
proteins with surface histidines, cysteines or tryptophans. Bound proteins are eluted by introducing
excess free ligand (phosphate buffer), or by raising the pH (e.g. by adding NH4OH), which promotes their
dissociation from the IMAC matrix. Non-specific binding of acidic proteins and peptides to the IMAC matrix
can be partly circumvented by chemical derivatization of the sample (esterification of carboxyl groups) or adding
acetate to the sample. Metal oxide affinity chromatography (MOAC) was introduced for analyzing
phosphorylated substances in 1990 [73] and is currently the technique of choice for enriching phosphopeptides
resulting from the tryptic digestion of phosphoproteins prior to MS analysis. Two kinds of beads are commonly
used: ZrO2 and TiO2, which reportedly have specificity (or at least some selectivity) for singly- and multiphosphorylated
peptides, respectively [74]. This methodology has been well optimized (see e.g. [75]), but it is
not frequently used for intact proteins. Aluminum hydroxide (Al(OH)3) matrices have also been used for MOAC
phosphoprotein enrichment. These matrices are relatively inexpensive and provide comparable yields to those
obtained using commercial phosphoprotein purification kits [76]. The numbers of IMAC- and MOAC-based
methods are constantly increasing. For example, Ti4+
-IMAC and Polymer-based metal ion affinity capture
(PolyMAC) techniques have been reported, both of which exploit the titanium atom’s ability to bind
phosphopeptides [77, 78]. Commercial kits with single-use columns and buffers are also now available to
explore the broadening avenues of phosphoproteomics. The manufacturers do not describe the composition of
the affinity matrices in their kits, except those based on affinity to phosphospecific dyes (Phos-tag agarose,
Wako; Pro-Q Diamond Phosphoprotein Enrichment Kit, Life Technologies). However, the similarity of
purification results obtained using these kits to those obtained using IMAC/MOAC matrices (e.g. [76]) suggests
that they are also probably based on IMAC/MOAC technology. Chromatofocusing and ion exchange
chromatography are options that can be used to selectively enrich acidic proteins from a sample. As
phosphoproteins are generally acidic, these methods can be used for crude fractionation of proteomes. Even
separating differentially phosphorylated protein isoforms is possible, provided the samples are not too complex
[21]. The last phosphoprotein-enrichment method that should be mentioned is hydroxyapatite chromatography; a
type of “pseudo-affinity” chromatography or “mixed-mode” ion exchange using a form of calcium phosphate
with the formula Ca10(PO4)6(OH)2. Several types of hydroxyapatite resins have been designed for protein and
DNA separation. Hydroxyapatite has high loading capacity and proteins are eluted by increasing the
concentration of phosphate in the elution buffer. This form of chromatography is also not entirely
phosphoprotein-specific, as the carboxyl moieties of protein interact with the resin in a similar way (albeit more
weakly) to phosphoresidues, but it has been frequently applied for separating proteins that differ only in their
phosphorylation (e.g. [54]).
3.1.3. Uses of antibodies and derivatization in phosphoproteome analyses
Antibodies raised against phosphorylated peptide sequences can be used to immunoprecipitate targeted
phosphoproteins from a cell lysate. Phosphotyrosine antibodies have been available for several years, and
pSer/pThr antibodies are also commercially available, but they reportedly have lower specificity [79]. Producing
anti-phosphohistidine or antiphosphoaspartate antibodies is problematic due to instability of these
phosphorylations. However, many commercially available anti-phosphotyrosine Ig antibodies also recognize
phosphohistidine [64]. Immunoprecipitation is unsuitable for large-scale phosphoproteome studies as several
parallel immunoprecipitation reactions are required for each phosphoamino acid, but it can be beneficial if
specific phospho-site enrichment is desired [80]. Several derivatization-based phosphoprotein enrichment
methods have been developed, but they involve several chemical reactions that are not 100 % efficient. Thus,
yields are limited. Further, their specificity is also questionable. For example, replacement of phosphate groups
with biotinylated tags [81] does not work with pTyr, but can modify Ser/Thr O-glycosylated forms or even nonmodified
residues [82]. Due to the uncertainty of reaction specificity, the wider spread of derivatization-based
techniques for phosphoproteome analyses seems unlikely.
3.1.4. The future of phosphoproteomics
The plethora of methods available for isolating phosphoproteomes does not simplify the task. Many studies
indicate that different approaches provide complementary rather than overlapping coverage of captured
phosphoproteins/peptides. Further, labile phosphorylations cannot be robustly identified, mapped and
characterized using current methodology. Only time will tell which techniques, combinations of approaches or
novel procedures will prevail. A promising approach seems to be an automated diagonal capillary
electrophoresis technique based on use of immobilized phosphatase, which has demonstrated utility for handling
a mixture of phosphopeptides [31]. A similar approach could be theoretically extended to phosphoprotein
samples. Of course, the susceptibility of individual phosphoproteins to the phosphatases applied and increased
sample complexity would be limiting factors. Today, the best enrichment approach for large-scale
phosphoproteome studies seems to be a two-step procedure, starting with phosphoprotein enrichment (using
aluminum hydroxide chromatography, for example) followed by digestion and subsequent phosphopeptide
enrichment using IMAC or MOAC resins [83].
3.2. Redox proteomics
Numerous posttranslational modifications are caused by chemically reactive species such as reactive oxygen and
nitrogen species (ROS/RNS) that are produced via cellular activities (especially, but not solely, under “oxidative
stress conditions”, when the reductive capacity of the cell’s antioxidant systems is exceeded), and are capable of
damaging proteins, lipids, carbohydrates and nucleic acids. The reactions involved may be purely nonenzymatic,
e.g. in uncontrolled responses to excessive energy absorption during photosynthesis, but reactive
species can also be generated to damaging levels in processes linked to physiological cell defense and signaling
processes [84, 85]. Proteins are considered to be major targets of the oxidative species, accounting for 68 % of
oxidized molecules in the cell [86]. Oxidative protein modifications may be either reversible or irreversible, and
may result from either direct oxidation of amino acid residues or the formation of reactive intermediates. Thiol
side chains of cysteine residues are particularly susceptible to redox modifications and cysteine residues are
attracting increasing research attention. However, protein oxidation processes are very complex and redox
proteomics encompassess the full spectrum of PTMs induced by radicals. These can be divided into
modifications that alter side chains of amino acids, and modifications altering protein backbones, causing
fragmentation and cross-linking [87]. Moreover, non-protein components of cells like nucleic acids, lipids or
carbohydrates activated by interaction with reactive species may react with a protein, which further extends the
variability of the redox proteome [88, 89]. Here, we focus on cysteine oxidative PTMs, methionine oxidation,
protein nitration and carbonylation. Protein glycation and protein deamidation are discussed in separate chapters.
3.2.1. Cysteine oxidative modifications
The amino acid that is the most common site of known PTMs is cysteine (Fig. 1). Its oxidative PTMs are
commonly involved in redox signaling and reactive thiol side chains can act as sensors or switches in both signal
transduction and enzyme regulation [90]. Depending on their local concentrations, ROS/RNS can react with
cysteine to form several reversible or irreversible modifications (Fig. 5, C). Cysteine oxidative modifications are
probably the best characterized redox PTMs to date, and a dedicated database of experimentally verified protein
oxidative modifications (RedoxDB) has been established [91].
3.2.1.1. Visualization of modified cysteine residues
The abundance of cysteine in proteins hinders direct detection of modified residues by in vivo [35
S]cysteine
radiolabeling. Several approaches for studying protein glutathionylation involving inhibition of protein
biosynthesis and incorporation of radiolabeled cysteine into glutathione [92] have been developed, but they
severely disrupt cell homeostasis. However, combining protein extraction with thiol (radio)labeling in vitro is
feasible. The easiest method available for detecting disulfide bonds is by electrophoretic diagonal separation
under first nonreducing and then reducing conditions (see Fig. 3, D). In principle, further disulfide labeling is not
essential, but the sensitivity of the procedure is increased. All reversibly oxidized protein thiols can be switched
to labeled thiols [93], as follows. First, the background labeling of unmodified cysteine residues is minimized by
blocking free thiols using thiol-specific reagents such as maleimides, iodoacetic acid or iodoacetamide. Oxidized
thiols are then reduced using modification-appropriate agents (e.g. ascorbate for nitrosylation, arsenite for
sulfenylation, hydroxylamine for acylation, tris(2-carboxyethyl)phosphine for disulfides, and oxidoretuctase
GRX treatment for glutathionylation), and the newly exposed thiols are simultaneously labeled with a thiolreactive
biotin [92, 94]. The biotin switch method can be used for both enrichment and detection. Once the
modified sites have been labeled with the stable biotin group, they can be easily detected by anti-biotin
immunoblotting, or enriched by streptavidin affinity chromatography. Alternatively to biotin replacement,
reduced cysteine residues can be labeled with a fluorescent dye (redox florescence switching), then readily
separated and visualized using 2-DE (particularly the variant DIGE, difference gel electrophoresis, which allows
the resolution of several differentially labeled samples in a single 2D-PAGE experiment). Proteins are labeled by
Cy3 and Cy5 fluorescent dyes, which have a maleimide group that forms a thioether bond with the free thiols of
cysteines in proteins. Comparison of gels stained with these dyes and a total protein stain (e.g. Sypro Ruby)
provides exact quantitative and qualitative information about cysteine PTMs in analyzed sample [95]). Some
cysteine PTMs can be also detected by specific antibodies. Antibodies have been raised against Sglutathionylated
proteins, sulfenic and sulfonic acid PTMs. However, their use has been limited to targeting and
assaying the modification status of individual proteins, and they have not yet been used in any published largescale
proteomic analysis [94, 96, 97].
3.2.1.2. Enrichment of proteins containing modified cysteine residues
The instability of redox cysteine PTMs often poses experimental challenges and thiol-disulfide exchange
between proteins during isolation/enrichment steps may lead to misinterpretation of results. Hence, it is
important to minimize changes in the thioldisulfide status of samples (e.g. alkylation of free cysteine residues)
during enrichment steps. The alkylation followed by a “biotin switch” method described above is currently the
most commonly used technique for purifying S-nitrosylated proteins and can be applied for analyzing any class
of cysteine oxidative PTM for which a suitable reducing agent is available. However, the specificity of reducing
agents is also a limiting factor here. It has been argued that even the selectivity of the ascorbate method for Snitrosylation
might be compromised by ascorbate-mediated reduction of disulfides [98]. On the other hand, this
technique is well established and various useful modifications have been developed. For example, Foresetre et
al. replaced the biotin tagging with fast simultaneous labeling and capture using dipyridyl disulfide coupled to a
resin, which can be followed by direct on-resin digestion and LC-MS/MS analysis [99]. This approach is similar
to activated thiol sepharose chromatography, where the pyridyldithiol group is the moiety that reacts with
reduced thiol groups and binds them. However, the binding efficiency of thiol sepharose for different proteins
may vary, and this method has yet to be further validated [98]. Proteins with reactive disulfides can also be
efficiently trapped by thioredoxin (Trx) affinity chromatography. Natural Trx has two cysteine residues in its
active center, the first catalyzing formation of a mixed disulfide intermediate between Trx and its substrate, and
the second cleavage of this short-lived bond by a nucleophilic attack. If the second cysteine residue is replaced
by serine, the mutated Trx can be used to capture the disulfide proteome, and bound proteins can be
subsequently released by treatment with dithiothreitol [100]. Some proteins can bind to Trx noncovalently in a
protein-protein interaction manner, but this can be circumvented by collecting Trx-protein complexes and
resolving them by nonreducing/reducing diagonal electrophoresis [101]. Trx-based enrichment is specific for Trx
substrates, but not all disulfides are accessible to Trx and affinities may differ even among those that do bind
[98, 100]. Thus, Trx affinity chromatography is not considered suitable for global redox-proteomics experiments.
Glutathionylation results in PTMs that are relatively stable, compared to other oxidative PTMs, and in addition
to the “biotin switch” method they can be targeted by specific antibodies, or by supplying a biotin-labeled
glutathione ester to investigated cells [102, 103]. Proteins that be potentially glutathionylated can also be
captured with agarose-bound glutathione [104], but this method does not help in monitoring the redox status of a
cell. The latest promising strategy is on the border between protein- and peptide-based analyses. It employs two
different alkylation agents, the first to protect free thiols prior to SDS-PAGE protein separation, and the second
after oxidized cysteine in-gel reduction. Proteins are subsequently digested in-gel and peptides are subjected to
16
O/18
O-labeling to differentiate controls and oxidized samples. Peptide-based comparison is then used for
simultaneous analysis of dynamic alterations in the redox state of cysteine sites and protein abundance [105].
However, to address low-abundance proteins, this technique has to be combined with further enrichment steps.
3.2.2. Tyrosine and tryptophan nitration
Reactive nitrogen species are generated in various physiological processes, including mammalian inflammatory
responses. The most common products of protein nitration are nitrotyrosine, and to a much lesser extent,
nitrotryptophan (Fig. 5, B)[106]. Nitrotyrosine can be produced from the reaction between tyrosine and a strong
nitrating agent like peroxynitrite and it is considered a general biomarker of disease progression [107]. In low
complexity samples, nitrotyrosine can be traced by UV-Vis photometry. Like tyrosine, it has an absorption peak
at 280 nm, but nitrotyrosine exhibits additional absorption peaks at ~ 357 and ~ 430 nm in acidic and basic
solutions, respectively. Yang et al. (2010) exploited this technique to follow the kinetics of protein nitration by
coupling HPLC with photo-diode array detection [108]. Similarly, nitrotryptophan formation can be detected by
spectrophotometry. Tryptophan exhibits the strongest fluorescence of any proteinaceous amino acid, with an
excitation maximum at 280 nm and an emission maximum between 305 nm and 355 nm, and its nitration leads
to fluorescence quenching. Both nitrotyrosine and nitrotryptophan can be detected and/or enriched using specific
antibodies [109] or derivatized. The derivatization method involves blocking all primary amines, reducing nitroresidues
to their amino- counterparts and selectively attaching an affinity handle (e.g. biotin) for enrichment or a
reporter group for selective mass spectrometric detection [106, 110].
3.2.3. Methionine oxidation
Like cysteine, the other proteinaceous sulfur-containing amino acid (methionine) is also readily oxidized, usually
yielding methionine sulfoxide and methionine sulfone (Fig. 5, D). Sulfoxide formation can be enzymatically
reversed by the action of specific reductases and it has been suggested that methionine could serve as an
antioxidant that protects other amino acid residues from more deleterious damage [111]. While this possibility
cannot be excluded, it has been shown that methionine oxidation can result in loss of a protein function or
interference with other PTMs close to the oxidized site [112, 113]. Although methionine oxidation is detected in
biological samples relatively often, thorough proteomic methodology for its characterization is still in
development. Anti-methionine sulfoxide antibodies have been raised against oxidized methionine-rich zein
protein [114], but a recent report indicates that their specificity is questionable [115]. Thus, it seems that the only
robust options for analyzing oxidized methionine are peptide-based approaches, which can be based on the 16Da
mass increase upon oxidation [116]. An advanced peptide-based method is diagonal chromatography (see
section 2.2.5.), with enzymatic reduction of methionine sulfoxides by methionine sulfoxide reductase [117]. This
method has been used to identify 35 in vivo methionine oxidations in the mouse serum proteome and could
perhaps also be adapted for protein-level analyses.
3.2.4. Protein carbonylation
Carbonylation is considered to be the third most common oxidative PTM, after cysteine and methionine
oxidation. Direct protein carbonylation predominantly affects side chains of lysine, arginine, proline and
threonine residues [118]. These oxidations are believed to be irreversible and result in glutamic semialdehyde
(arginine and proline), 2-amino-3-ketobutyric acid (threonine) or aminoadipic semialdehyde (lysine) (Fig. 5, G)
[119-122]. Indirect protein carbonylation is more prevalent and involves a reaction with oxidized AGE
(advanced glycation end) products (described in glycoproteomics, section 3.4), or hydroxyl radical-mediated
oxidation of lipids. Lipid-derived aldehydes can form Michael adducts with cysteine, histidine, lysine and
arginine, and in some cases can further undergo Schiff base formation with amines of adjacent lysines,
producing cross-linked amino acids[120, 123, 124]. Analysis of the resulting PTM diversity is hindered by their
instability because carbonyl groups readily undergo Schiff base formation with proximate lysine residues, even
when stored at -80 °C [122].
3.2.4.1. Isolation and detection of carbonylated proteins
To date, 2-DE analysis has been used in most proteomic studies of protein carbonylation [125]. For detection,
reactive carbonyls can be reduced by [3
H]sodium borohydride to corresponding tritium labeled alcohols, but
most methods rely on covalent labeling with nucleophilic hydrazide- or hydrazine- based probes [126].
Originally a qualitative analytical method used in organic chemistry, based on the formation of nitrogen-carbon
bonds (hydrazone) between carbonyls and dinitrophenylhydrazine (DNPH) was adapted to enhance the isolation,
identification and quantification of carbonyl-containing proteins. In this approach, derivatized proteins are
detected/captured by DNPH-targeting antibodies [120, 127]. DNPH can also reportedly react with thioaldehyde
derived from sulfenic acid [128], but no enrichment of corresponding methionine or tryptophan oxidation
products has been observed as yet [125]. Biotin-based hydrazides can be used instead of DNPH and derivatized
proteins are then captured by biotin-avidin affinity chromatography (e.g. [125, 129]). However, the binding
efficiency of biotin-based tags seems to vary considerably, and the most commonly used (biotin hydrazide) gives
the lowest yield [130]. Carbonylated peptides have been successfully enriched by derivatization with Girard’s P
reagent, which forms hydrazones with peptide carbonyl groups and contains a quaternary amine group that is
selectable by SCX chromatography [131]. However, this methodology is unlikely to achieve comparable utility
in analysis of intact proteins mainly because of, in general, modest influence of the newly generated positive
charge on net charge of the respective protein.
3.2.5. Troublesome path of redox proteomics
Unfortunately, current experimental approaches are not able to deal robustly with the diverse spectrum of PTMs
introduced by reactive species. While we are capable of detecting abundant PTMs, using 2-DE and mass
spectrometry, for instance, targeted detection and enrichment techniques are available only for specific types of
redox PTMs. Further, it might be difficult to distinguish between biologically relevant PTMs and artifacts
introduced during an analysis. For example, tryptophan oxidation is considered to be a natural irreversible PTM,
but it has been recently found that kynurenine and N-formyl kynurenine tryptophan oxidation products can be
introduced into a protein during SDS-PAGE separation [132].
3.3. Protein lipidation
Covalent attachment of lipids controls the subcellular localization and activity of diverse proteins. The most
common lipidation processes are prenylation, acylation and glycosyl phosphatidylinositol anchor (GPI)
attachment (Fig. 5, E) [1, 133-135]. Farnesyl and geranylgeranyl-pyrophosphate molecules can be utilized as
electrophilic alkyl donors in posttranslational protein prenylation of conserved cysteine residues at or near the Cterminus
of proteins. These PTM types, which were first reported in 1984 [136], are catalyzed by farnesyl
transferase or geranylgeranyl transferase at the conserved sequence Cys-a-a-X (where a is an alifatic residue and
X any residue), and have been shown to have profound effects on a host of processes involving signal
transduction and intracellular trafficking [137-139]. Protein acylation classes of lipidation PTMs include Nmyristoylation
at N-terminal glycine residues and S-acylation of cysteine residues by addition of palmitic acid or
other long chain fatty acids. The latter is an important mechanism for dynamic targeting of proteins to
membranes, and the labile, reversible thioester linkages are enzymatically formed and cleaved by
acyltransferases and protein thioesterases, respectively [135, 140]. Non-enzymatic S-acylations are also possible,
but apart from modifications reported in mitochondria, these reactions probably occur only in vitro [141]. Nmyristoylation
is a cotranslational modification catalyzed by N-myristoyltransferases at the conserved sequence
Gly-X-X-X-(Ser/Thr/Cys) after cotranslational hydrolysis of the Met1–Gly2 bond by methionine aminopeptidase.
However, it can occur posttranslationally if an internal glycine residue is exposed upon protease-catalyzed
cleavage [1, 135]. Attachment of GPI (which consists of a phosphoethanolamine linker, a glycan core, and
phosphatidyl inositol tail) at the C-terminus of a eukaryotic protein results in its plasma membrane localization
[142]. Other lipid PTMs, including O-acylation and C-cholesterylation, have been reported, but they do not seem
to be widespread [143, 144].
3.3.1. Analysis of protein lipidation
Attachment of a lipid moiety increases a protein’s hydrophobicity, which restricts the methods that can be used
to separate lipidated proteins. Unless lipid PTMs are removed, the best methods for resolving lipidated
proteomes are multidimensional liquid chromatography or BN-PAGE. However, if the aim of an analysis is
solely to identify proteins undergoing lipidation, they can be enriched then released by enzymatic or chemical
cleavage. For example, GPI-modified membrane anchored proteins can be enriched from soluble proteins by
Triton X-114 partitioning [145], as they separate into the micelles fraction with other membrane-associated
proteins. Subsequent cleavage of the GPI anchors by phosphatidylinositol-specific phospholipase C releases
these proteins into the soluble fraction, from which they can be separated by repeated Triton partitioning.
However, this technique does not provide information about GPI composition, or sites of GPI attachment in the
proteins. Moreover, some GPI-anchored proteins are phospholipase-insensitive due to lipid remodeling within
the GPI anchor [146]. However, GPI contains glucosamine, which is generally rare in other glycoproteins. Thus,
instead of enzymatic cleavage, nitrous acid deamination can be used to break the glycosidic bond between
glucosamine and the myo-inositol ring, resulting in cleavage of the lipid portion of the GPI anchor [147].
3.3.2. Detection and enrichment of lipidated proteins
Like the first methods for detecting most of the long-known protein PTMs, the first methods for detecting
protein lipidation were based on radiolabeling, more specifically feeding [3
H] and [125
I]-labeled fatty acids,
followed by protein separation and autoradiography. This is still one of the methods of choice for confirming the
presence of a lipid moiety in a protein since it is not based on indirect detection [135, 138]. In addition, due to
the paucity of lipid antigenicity relatively few antibodies that recognize specific lipids have been described
[148]. Antibodies for specific GPIs are known (e.g. the immune response to malaria involves the production of
anti-GPI antibodies that recognize the unique plasmodium glycan modifications in the Plasmodium falciparum
GPI anchor), but it remains to be shown if any anti-lipid antibodies could be applied for large-scale protein
lipidation analyses. The “Biotin-switch” method described above for cysteine PTMs (see section 3.2.1) has been
successfully applied for detecting S-acylation (with hydroxylamine as a reducing agent), but there are no reports
of a modified version for detecting other types of protein lipidation PTMs. Instead, a new approach has emerged,
based on metabolic labeling with lipid analogs containing chemical groups that are not present in biological
samples. These chemical groups, e.g. alkyne or azide, can be then targeted either in vitro or in vivo by
“bioorthogonal” chemical reactions, such as Staudinger ligation [149] or copper-catalyzed azide-alkyne
cycloaddition (CuAAC) [150, 151]. However, as detection systems these chemical reporters are several orders of
magnitude less sensitive than antibody-based systems, and the high concentrations of reactants that would
perturb cellular homeostasis severely limits application of this strategy in vivo [152]. Nevertheless, azide- and
alkyne-functionalized lipid chemical reporters have already allowed tremendous advances in the detection of
protein S-acylation [153-155], N-myristoylation [156] and prenylation [157]. After protein extraction, the
chemical reporters are reacted with fluorescent or biotinylated tags for in-gel detection or avidin-biotin
enrichment, respectively [135, 158]. The major drawbacks of this technique are the necessity of metabolic
labeling, which means that it cannot be applied to characterize standard samples, and the possibility that the
unnatural substrates used may be processed by cell enzymes in different ways from natural counterparts. Further,
for efficient labeling of prenylated proteins prior to prenylome profiling, the endogenous isoprenoid pool has to
be depleted with statins, which precludes comparative studies of different cellular states and analysis of
regulatory mechanisms without significant metabolic perturbation of cells [159]. An alternative to bioorthogonal
reactions is in vitro labeling, which requires use of engineered enzymes with modified substrate specificity, e.g.
a protein prenyltransferase with biotin-geranylpyrophosphate specificity for prenylome analysis [160]. Such a
modified enzyme can be also introduced into cells, but the technique is better suited for in vitro detection of
potential protein lipidation targets.
3.4. Glycoproteomics
In 1857 the French physiologist Claude Bernard described “glycogenous matter” in liver as a storage form of
glucose. The possible existence of a protein component of glycogen was reported just three decades later, but its
presence was only accepted after nearly a century of further research. Nevertheless, we can consider glycogen,
with its protein core glycogenin, to be the first studied glycoprotein [161, 162]. Today, we know that proteins
can undergo covalent modification by carbohydrates via non-enzymatic glycation or enzymatic N-glycosylation,
O-glycosylation and C-glycosylation (Fig. 5, H). These PTMs play important roles in many key biological
processes, including protein folding, signaling, enzymatic activity, molecular trafficking, cell adhesion and
protein turnover [163, 164]. Glycosylation is believed to be more prevalent than phosphorylation and it is
estimated that at least 50 % of all proteins are modified by carbohydrates [163]. N-linked glycosylation is by far
the most thoroughly studied form of carbohydrate PTM. It predominantly involves cotranslational attachment of
a glycosyl moiety to the carboxamide nitrogen atoms of asparagine, almost always in the consensus sequence
Asn-X-Ser/Thr (where X is any amino acid residue except proline), by N-glycosyltransferases in the lumen of
the endoplasmic reticulum and Golgi apparatus [165]. O-glycosylation has been frequently documented at serine
and threonine residues, less frequently at hydroxyproline and hydroxylysine sites, and very rarely at tyrosine
sites. The glycosyl moieties involved are generally shorter and less complex than N-glycosyl moieties, which
may make it more suitable for reversible regulation by specific transferases and hydrolases. In further contrast to
N- glycosylation, there is no known conserved sequence motif for O-glycosylation [1, 164]. C-glycosylation is
rare, and has only been found at tryptophan residues, the C2 position of the indole ring being a mannosylation
target [163]. Non-enzymatic protein glycation may occur at lysine or arginine amino groups, or the N-terminal
amino acid residues of proteins. In the process, the aldehyde group of the open chain of a reducing sugar
condenses with an amino group to form a Shiff base, and then an Amadori rearrangement of the double bond to
the C2 of the sugar occurs, yielding a stable, covalently linked ketoamine compound (early glycation product
This process may slowly and spontaneously revert, thereby regenerating the free sugar and the amine, or the
glyco-portion may undergo a further series of reactions leading to the formation of various heterogeneous
structures commonly known as advanced glycation end-products (AGE) [166].
The attachment of sugar residues can generate greater structural diversity than any other kind of protein PTM.
While three different amino acids can theoretically generate six different peptide-bond trimers, three different
hexoses could produce anywhere from 1,056 to 27,648 unique trisaccharides; a massive difference [162]. In
nature, the sheer number of combinations is limited by the specificity of glycosyltransferases, but the diversity of
mature glycan chains is still considerable and the remarkable variation in the attached oligosaccharide chains is
the major obstacle in glycoproteomics. For this reason, large-scale analyses have typically focused either on
characterization of the protein parts and mapping the PTM sites, or characterization of the glycan structures. The
latter is beyond the scope of this review and for more information the reader is referred elsewhere, e.g. the
excellent tutorial by An and Lebrilla (2011)[167]. Here, we focus on general approaches to detect and enrich
glycoproteins from a complex sample.
3.4.1. Visualizing a glycoproteome
Detection of carbohydrate protein PTMs can be facilitated by metabolite labeling with either radioactive or
chemically derivatized monosaccharides. In vivo treatment with [3
H]- or [14
C]-labeled glycan precursors
followed by electrophoretic separation allows autoradiographic detection of any glycoprotein into which they are
incorporated [168]. Alternatively, cells can be fed a bioorthogonal chemical tag, usually an azido analog, that
allows chemoselective ligation and hence specific enrichment or detection of glycoconjugates in the same
manner as described above for lipid conjugates [169]. Both of these approaches can be utilized in vitro for
detecting potential glycosylation targets using recombinant glucosyltransferases, or for analyzing spontaneous
non-enzymatic glycation, because the generation of early glycation products with proteins occurs at a rate
proportional to the reducing sugar concentration. Glycoproteins resolved by 2-DE or SDS-PAGE can be
visualized by specific staining, most classically by a periodic acid-Schiff stain (PAS) procedure, in which
glycans’ vicinal diols are oxidized to aldehydes then reacted with Schiff reagent, yielding pink to magenta
adducts [170]. The detection limit is reportedly 25-100 ng for free carbohydrates, but the reported sensitivity for
glycoproteins is in the range 1-10 Pg [171]. Sensitivity has been slightly improved by combining the PAS
reaction with alcian blue or dansyl chloride staining, or by tagging periodate-oxidized aldehydes with biotin
hydrazide and detecting the biotin (by streptavidin-bound peroxidase for instance), but these labeling procedures
are seldom utilized in practice [171-173]. The fluorescent dye Pro-Q Emerald (Life Technologies) also reacts
with periodic acid-oxidized carbohydrate groups, but the protocols are more rapid than previously mentioned
PAS-based labeling procedures and reportedly at least 50-fold more sensitive [174]. Similar alternatives have
been produced, for example by Sigma (Glycoprofile III fluorescent kit; discontinued) and Pierce (Krypton
Glycoprotein staining kit) [172]. Specific protein glycosylation PTMs can be detected after electrophoretic
separation by Western blotting using glycan-specific antibodies, or overlays with radio-labeled or enzymeconjugated
lectins. Lectins are natural proteins characterized by their ability to selectively bind carbohydrate
ligands. Most lectins bind reversibly, do not modify carbohydrate covalent structures, and are non-immune in
origin [175]. Hundreds of lectins have been described and more than 60 are commercially available, with
specificity ranging from very broad (e.g. Concanavalin A binds N-glycoproteins by recognizing a shared
trimannosyl motif of the N-glycan core structure) to very narrow (e.g. Sambucus nigra lectin is specific for sialic
acid attached to galactose/N-acetylgalactosamine by an (D 2-6) linkage [176](See Tab. 1). Lectin-based detection
is also employed in so-called “lectin microarrays”, in which discreet fractions of a proteome are usually spotted
onto nitrocellulose substrates and the resulting array is subsequently incubated with a specific lectin. An
alternative antibody-lectin sandwich version captures specific glycoproteins by antibodies and hybridized arrays
are then incubated with fluorescently or biotin-labeled lectins for detection. This approach is primarily used for
detecting and comparing differences in glycan structures, but MS can be used to identify the glycoproteins
within a sample [177-179]. Glycoproteins may also be detected by mobility shifts, in either the pI or MW
dimension, in diagonal separations after enzymatic or chemical (e.g. trifluoromethanesulphonic acid-induced)
cleavage of glycosylation bonds [29, 180].
<-Table 1->
3.4.2. Enrichment of glycosylated or glycated proteins
Methods described for glycoprotein detection, including metabolite labeling, hydrazide reaction with oxidized
carbohydrates, and especially lectin-based procedures, have also been used to enrich glycoproteins. Lectin
affinity chromatography can be used with single lectin species, or multiple lectins to capture (for example) all Nlinked
glycoproteins [165, 181, 182]. Depending on the lectin(s) used, binding may require the presence of metal
ions or a specific pH. Bound glycoproteins are then desorbed by using a chelating agent or lowering the pH,
respectively, but this can of course be facilitated by simply adding a high concentration of the free carbohydrate
ligand(s). Hydrazide coupling is the major chemical-based enrichment technique for glycoproteomic analysis.
The coupling reagent can be either linked to biotin and captured via biotin-avidin affinity, or immobilized on
beads [183, 184]. Similar chemistry has also been used for more specific enrichment. For example, Zeng et al.
(2009) introduced a method for the selective derivatization of sialic acid by mild periodic oxidation and ligation
with an aminooxy tag in the presence of aniline, and the same group has recently described an approach to
specifically enrich glycoproteins with terminal N-acetylgalactosamine/galactose [185, 186].
Lectin- or hydrazide/aminooxy-bound glycoproteins can be digested on-beads and the remaining glycancontaining
peptide(s) can be released and, if desired, various tags can be incorporated into them for quantitative
purposes [184, 187, 188]. This is suitable for analysis of N-glycans because the known enzyme N-glycosidase F
cleaves the glycosidic bond between the asparagine residue of a glycosylated protein and the core Nacetylgalactosamine,
thereby liberating nearly all N-linked glycans. This is a particularly attractive reaction,
because it results in deamidation of asparagine to aspartate with a mass shift of 1 Da (or 3 Da in [18
O] water)
which is indicative of a PTM site and can be used for comparative quantitation. However, it should be mentioned
that spontaneous deamidation (see section 3.5) has the same effect and may lead to erroneous assignments [189].
A lack of broad-specificity O-glycosidases can be circumvented by reductive E-elimination through the use of a
strong base [190] or hydrazinolysis [191], both of which release O- and N-linked glycopeptides. Another method
for glycoproteome enrichment is based on the reversible reaction of boronic acid with 1,2-cis-diols [192]. These
moieties are present in glycan components like mannose, galactose, glucose, and sialic acid, but are also
preserved in glycated proteins, and boronate affinity chromatography is currently the only antibody-independent
method for enriching glycated proteins [166, 193]. Several derivatives of boronic acid have been introduced for
this purpose, including bisboronic compounds, which can also be conjugated with a fluorescence tag and used
for detecting protein glycosylation/glycation [194], and both sulfonyl- and sulfonamide-phenylboronic acid
derivatives, which can be used at physiological pH. These adjustments have solved the major drawback of the
original setup; the need to use alkaline buffers due to the relatively low ionization constants of the commonly
used phenylboronic acid ligands [195]. Non-specific boronate affinity has also been successfully combined with
lectin affinity in boronic acid-lectin affinity chromatography [196]. In addition, boronate in the form of
methacrylamido phenylboronic acid has been copolymerized in acrylamide gel and applied to detect, analyze and
separate early and AGE glycation products from unaffected proteins [197]. During the electrophoretic
separation, migration of glycated proteins is hindered and gel resolution allows various carbohydrate-protein
adducts to be distinguished. A boronate gel was commercialized by Pierce (Glycogel II boronate affinity gel),
but it is no longer available. Anti-glycosylated/glycated protein antibodies have also been employed in Western
blot analyses, and can be used for specific enrichments. This is especially useful for analyzing protein glycations
that are not recognized by lectins and a number of antibodies have been prepared for analyses of AGE protein
modifications, notably for detecting H-N-carboxymethyllysine and H-N-carboxyethyllysine [198, 199]. These can
be raised, for example, against bovine serum albumin that has been treated with glycolaldehyde, and are
commercially available (e.g. Millipore; http://www.millipore.com).
3.5. Protein deamidation
Asparagine and glutamine residues in both proteins and peptides can be deamidated through spontaneous
chemical reactions that may occur either in vivo or in vitro. The rates of deamidation depend (inter alia) on pH,
temperature and protein structure. Depending on the surrounding amino acids, half-times of deamidations of
specific asparagine and glutamine residues in proteins are within ranges of 1-500 days and 100-1000 days,
respectively, at neutral pH and 37 °C [200]. The rates are significantly higher at alkaline pH, which is typical
(for instance) in trypsin digestion. Thus, it is difficult to distinguish a natural protein deamidation, which could
be a marker of a disease or protein ageing, from artifactual deamidation during proteomic sample preparation
[201, 202]. Deamidation proceeds via formation of a cyclic intermediate by condensation of the side-chain amide
group with the carbonyl group of the following amino acid, and subsequent opening of this ring upon hydrolysis
with water. While this process cannot be completely avoided, even with an improved protocol [202], a potential
way to discriminate in vitro reactions is to prepare samples in [18
O]-water, which increases the molecular weight
of deamidation artifacts [203, 204]. The hydrolysis of a ring intermediate can yield two products; aspartic acid
and isoaspartic acid in the case of asparagine (Fig. 5, I). The second product is more frequent (yield up to 70 %)
and leads to a significant change in the backbone structure of the host protein. Thus it is often accompanied by a
loss of biological activity. Deamidation also introduces an additional negative charge in the protein and can be
detected by charge-sensitive separation methods (IEF, 2-DE or ion-exchange chromatography) [205, 206]).
Isoaspartyl residues can be enzymatically converted to asparagyl residues by protein L-isoaspartyl Omethyltransferase
[207], which can be used to label proteins containing isoaspartyl residues by [3
H]-methylation
in vitro. Proteins can then be separated (e.g. by 2-DE) and converted isoaspartyl residues can be detected by
autoradiography [206, 207]. There is no specific enrichment procedure for isolating a deamidation proteome, and
large-scale analyses thus depend on advances in mass spectrometry [208, 209].
3.6. Protein acetylation
Acetylation of proteins is a specific form of protein acylation (Fig. 5, E) that was discovered in the early 1960s
[210-212]. The most common types are N-acetylation of the amino terminus of a protein and N-acetylation of
lysine residue side chains, but O-acetylations of serine and threonine have also been reported [213]. N-terminal
acetylation rarely occurs in prokaryotes, but it occurs in about 50 % of yeast proteins and more than 80 % of
human proteins [214, 215]. It is believed to be irreversible and Hwang et al. found that it could act as a signal for
ubiquitin-dependent degradation [216]. H-Lysine acetylation is by far the most extensively studied form of
acetylation. It is regulated by N-acetyltransferase and deacetylase enzymes, and plays roles in protein stability,
localization, enzyme activity, and (in histone acetylation) directly regulates gene expression by changing
chromatin structure [217-219]. Analysis of the acetylome is hindered by the high abundance of acetylated
histones in the cell and the lack of suitable physical and biochemical properties of the acetyl group that could be
exploited for its specific enrichment. However, acetylation masks positive charges of lysine residues (or Ntermini)
and can thus be detected by changes in proteins’ pI (IEF, 2-DE) [220]. There has also been some
success in the development of anti-acetyllysine antibodies and immunoaffinity enrichment techniques [221-224].
However, Shaw et al. clearly demonstrated the limitations of this strategy by showing that sets of acetylated
peptides detected by a cocktail of monoclonal antibodies raised against acetylated lysine in different sequence
contexts differed from those detected by the polyclonal antibodies used by other groups [225]. Acetylome
analyses could benefit from diagonal separation, but the only such technique reported to date is a chemical-based
combined fractional diagonal chromatography (COFRADIC) approach to determine N-terminal acetylation
[226], and no method employing enzymatic deacetylation has been presented as yet.
3.7. Protein methylation
The first report on the occurrence of a methyl PTM in a protein dates back to 1959, when Ambler and Rees
found H-N-methyl-lysine in a bacterial flagellar protein [227]. Protein methylation predominantly occurs at lysine
and arginine residues, but N- or O-methylation has also been detected at side chains of other amino acids, Nterminal
amino groups and C-terminal carboxyl groups. There have even been reports of cysteine S-methylation
and C-methylation at methylene carbons of arginine and glutamine (Fig. 5, F) [1, 228, 229]. Methylation affects
proteins’ hydrophobicity and can mask a negative charge (by formation of methyl esters). It has demonstrated
roles in various processes, e.g. signal transduction, mRNA splicing, transcriptional control, DNA repair, proteinprotein
interactions, protein stability and protein translocation [229, 230]. Methylation is often catalyzed by
methyltransferase enzymes that utilize S-adenosylmethionine as a substrate, but S-adenosylmethionine can also
readily methylate cysteine and histidine residues non-enzymatically in vitro, and a similar non-enzymatic
mechanism has been proposed for the methylation of some proteins in vivo, e.g. human lens crystallins [231]. It
was long believed that N-methylation is irreversible and that reversibility of O-methylation is due to spontaneous
non-enzymatic decomposition of the ester bond. However, this view was changed by the discovery of lysine
demethylases [232] and the search for arginine demethylases is in progress [233].
3.7.1. Protein methylation detection and enrichment of protein methylomes
The methylation of eukaryotic proteins may be widespread, but the large-scale proteomic techniques that could
rigorously test this hypothesis are just emerging. Protein methylation can be detected by metabolite labeling
using [14
C]- or [3
H]-radiolabeled S-adenosylmethionine, followed by protein extraction, separation and
autoradiography, but to date this procedure has been limited to small-scale analyses and in vitro detection of
possible substrates of methyltransferases [234, 235]. Anti-methyllysine and anti-methylarginine antibodies have
been raised and successfully employed in methylated protein enrichment and Western blot detection [235, 236].
The subtle chemical differences, e.g. between methylated and unmodified arginine, limits the application of
antibodies for intact protein enrichment. However, using antibodies in combination with stable isotope labeling
by amino acids in cell culture (SILAC) by [13
CD3]methionine, Ong et al. detected 59 methylation sites in HeLa
cell extracts [235]. In addition, Uhlmann et al. recently reported a peptide-based method for large-scale
identification of protein arginine methylation that seems to outperform antibody-based enrichment [237]. They
detected 249 methylation sites in peptides from a SILAC-labeled cell lysate and showed that most arginine
methylated peptides were highly basic and hydrophilic, which facilitated their purification. A comparison of the
effectiveness of SCX chromatography, IEF and hydrophilic interaction liquid chromatography (HILIC) clearly
indicated that HILIC [238] is by far the most effective for this purpose. Surprisingly, however, Uhlmann et al.
have not found a similar level of enrichment for peptides containing methylated lysine [237].
3.8. Protein ubiquitination
Ubiquitin is a highly conserved protein of about 8.5 kDa, which can be attached in an ATP-dependent manner to
a protein. The history of protein ubiquitination research dates back to 1977, when Goldknopf and Busch
described an isopeptide bond between a histone 2A lysine residue and the tryptic C-terminal diglycine remnant
of a different protein [239]. PTM by ubiquitin and ubiquitin-like proteins represents a major regulatory system.
The basic enzymatic cascade couples the C-terminus of ubiquitin via an isopeptide bond to the H-amino group of
lysine residues on substrate proteins (Fig. 5, J) or, less frequently, via peptide bonds to the N-terminus, but
evidence has also emerged for ubiquitin ester bonds via cysteine, serine or threonine residues [240]. Proteins can
be monoubiquitinated, or polyubiquitin chains may form via linkages involving any of the molecule’s seven
lysine residues (K6, K11, K27, K29, K33, K48, K63) or its N terminus, which have different structures and
properties, depending on how the chains are assembled [241]. Protein (poly)ubiquitination is known to regulate
protein turnover by influencing proteasome-mediated degradation, gene transcription, subcellular trafficking and
protein localization, enzymatic activity, and to facilitate protein-protein interactions [242-246]. Ubiquitination is
catalyzed through the sequential action of three discrete enzymes: the ubiquitin activating enzyme (E1),
ubiquitin-conjugating enzyme (E2), and ubiquitin-ligating enzyme (E3). Ubiquitin tags can be removed or edited
by deubiquitinating enzymes (DUBs). The number of enzymes involved rivals those involved in
phosphoproteome regulation: the human genome encodes approximately 10 E1, 40 E2, over 600 E3 and 80-90
DUBs. Plant signaling is highly dependent on ubiquitin, which is reflected in over 1,000 Arabidopsis thaliana
genes for E3 [247, 248].
3.8.1. Protein ubiquitination - enrichment
The “ubiquitinome” poses several analytical challenges, including the fact that ubiquitination frequently leads to
protein degradation and loss of modified proteins, unless the proteasome pathway is inhibited [249]. Previously,
proteomic methods for ubiquitin enrichment have predominantly utilized affinity-tagged labeling and antibodybased
capture. Beers and Callis engineered a variant of ubiquitin containing a polyhistidine-tag at its N-terminus
for an in vitro assay, which allowed IMAC enrichment of proteins with incorporated tagged ubiquitin [250]. A
similar approach can also be used for in vivo analyses of transgenic lines expressing tagged ubiquitin [251, 252].
Non-specific binding can be circumvented by employing another specific label in addition to ubiquitin (e.g.
biotin)[253], but this does not eliminate competition between the tagged exogenous ubiquitin and the
endogenous wild-type molecule, which lowers the sensitivity of this method. Further, the possibility that added
tags may influence the physiological activity of the molecule cannot be excluded [254]. Some natural proteins
are known to associate with ubiquitin and their ubiquitin-binding domains have been used as agarose conjugates
for ubiqitinome affinity enrichment [254]. This technique has been further developed by fusing domains into
tandem-repeated ubiquitin-binding entities (TUBEs)[255], which reportedly have higher affinity for
polyubiquitin chains and are capable of protecting ubiquitinated proteins from proteasome cleavage or enzymatic
deubiquitylation, a process difficult to inactivate even in denaturating conditions [256]. Ubiquitinated proteins
are routinely detected by Western blotting with anti-ubiquitin antibodies and antibodies have also been used with
some success for ubiquitinome enrichment [257]. Antibody-based enrichment has also been applied in peptidebased
approaches for ubiquitination detection [258]. Trypsin proteolysis of a ubiquitin-conjugated protein
produces a signature peptide for MS-based detection at the ubiquitination site containing a two-residue remnant
(glycine-glycine) that is derived from the C terminus of ubiquitin and is still covalently attached to the target
lysine residue via an isopeptide bond [251]. Antibodies raised against this sequence are effective tools for
profiling ubiquitinated substrates and have significantly increased the number of verified proteins with
ubiquitination PTMs [259].
4. Summary
Difficulties associated with MS-based analysis have driven the development of alternative approaches to enrich
and detect PTM modified proteins. Advances in MS-based proteomics have greatly facilitated the identification
of thousands of PTM sites in eukaryotic cells and revealed novel modifications, but also shifted the focus of
protein PTM analyses from a proteome to a peptide level. A key element for the success of these studies is the
ability to enrich modified peptides using antibodies or other chemical approaches. This has worked reasonably
well for O-phosphoproteome analysis, but most other PTMs have been more difficult to address and remain
poorly characterized because of the lack of such enrichment methods. However, all known PTMs may be
analyzed using the general workflow illustrated in Fig. 6. Indeed, if it was not for the low abundance of modified
proteins, 2-DE and/or reversed-phase or HILIC chromatography would be sufficient to detect most known
protein PTMs, due to the unsurpassed resolution of the former for pI/MW protein separation and the ability of
the latter to recognize changes in protein/peptide hydrophobicity. As it stands, these methods only provide
satisfactory results for abundant proteins or PTMs tagged with specific (radio)-labels. Sample labeling used to be
an essential step for any PTM study and it still remains advantageous. Labels may be introduced by enzymatic or
chemical reactions in vivo or during/after protein extraction. Labels not only facilitate detection, but can also be
used for specific enrichment. The major drawbacks usually lie in reductions in yields when in vitro derivatization
is applied, and/or the necessity to use enzymes with modified substrate specificity. Analyses of rare protein
PTMs may be improved by proteome fractionation, immunodepletion or proteome equalization. While each
additional protein isolation step provides indisputable benefits, they all prolong the procedure and usually
compromise labile PTMs. Nevertheless, there has been substantial development in PTM enrichment
methodology and a number of modifications may be addressed directly by (immuno)affinity enrichment.
Enriched samples may be further processed (e.g. by 2-DE) or digested. This is the point where bottom-up and
top-down proteomic approaches converge and may proceed through peptide-based enrichment, or directly to MS
analysis. In recent years approaches for analyzing protein PTMs have rapidly developed, but large proportions of
peaks in MS spectra of protein samples are still typically not assigned, and may originate from hidden PTMs.
This hypothesis has been recently corroborated by the detection of a signature fragment of a novel O-
acetylglucosamine-6-phosphate PTM in a re-analysis of a large-scale proteomics dataset [260]. Thus, we are still
clearly far from an optimal procedure for PTM analyses, and further advances are required not only in protein
isolation and MS sensitivity, but also in bioinformatics.
<-Figure 6->
Acknowledgements
This work was supported by the European Social Fund and the state budget of the Czech Republic through the
Education for Competitiveness Operational Programme (grant no. CZ.1.07/2.3.00/30.0017, for Postdocs in
Biological Sciences at MENDELU). The work was done at the Central European Institute of Technology
(CEITEC), with research infrastructure supported by the funds for the project CZ.1.05/1.1.00/02.0068 provided
by the European Regional Development Fund. Additional support for the work came from grants 206/09/2062
and P305/12/2144 (Grant Agency of the Czech Republic), and QI91A229 (Ministry of Agriculture of the Czech
Republic). Access to the MetaCentrum computing facilities provided under the ‘Projects of Large Infrastructure
for Research, Development, and Innovations’ program (ref. LM2010005), funded by the Ministry of Education,
Youth and Sports of the Czech Republic, is highly appreciated.
References
[1] Walsh CT, Garneau-Tsodikova S, Gatto GJJ. Protein posttranslational modifications: the chemistry of proteome diversifications. Angew
Chem Int Ed Engl 2005;44:7342-72.
[2] Seidler J, Zinn N, Boehm ME, Lehmann WD. De novo sequencing of peptides by MS/MS. Proteomics 2010;10:634-49.
[3] Finkel T, Holbrook NJ. Oxidants, oxidative stress and the biology of ageing. Nature 2000;408:239-47.
[4] Lehninger A, Nelson D, Cox M. Lehninger Principles of Biochemistry. New York: W. H. Freeman; 2008.
[5] Irvine G. Size-exclusion high-performance liquid chromatography of peptides: a review. Anal Chim Acta 1997;352:387-397.
[6] Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680-5.
[7] Hafiz A. Principles And Reactions Of Protein Extraction, Purification, And Characterization. United Kingdom: CRC PressINC; 2005.
[8] Alsheikh MK, Svensson JT, Randall SK. Phosphorylation regulated ion-binding is a property shared by the acidic subclass dehydrins.
Plant Cell Environ 2005;28:1114-1122.
[9] Myung J, Lubec G. Use of solution-IEF-fractionation leads to separation of 2673 mouse brain proteins including 255 hydrophobic
structures. J Proteome Res 2006;5:1267-75.
[10] Ly L, Wasinger VC. Protein and peptide fractionation, enrichment and depletion: tools for the complex proteome. Proteomics
2011;11:513-34.
[11] Hannig K. Die trägerfreie kontinuierliche Elektrophorese und ihre Anwendung. 1961;181:244-254.
[12] O'Farrell PH. High resolution two-dimensional electrophoresis of proteins. J Biol Chem 1975;250:4007-21.
[13] Klose J. Protein mapping by combined isoelectric focusing and electrophoresis of mouse tissues. A novel approach to testing for
induced point mutations in mammals. Humangenetik 1975;26:231-43.
[14] Bjellqvist B, Ek K, Righetti PG, Gianazza E, Görg A, Westermeier R, Postel W. Isoelectric focusing in immobilized pH gradients:
principle, methodology and some applications. J Biochem Biophys Methods 1982;6:317-39.
[15] Cossu G, Manca M, Strahler JR, Hanash SM, Righetti PG. Detection of electrophoretically silent mutations by immobilized pH
gradients. J Chromatogr 1986;361:223-9.
[16] Westermeier R, Naven T, Höpker H. Proteomics in practice a guide to successful experimental design. New Jersey: Wiley-Blackwell
2008.
[17] Schägger H, von Jagow G. Blue native electrophoresis for isolation of membrane protein complexes in enzymatically active form. Anal
Biochem 1991;199:223-31.
[18] Erni F, Frei R. Two-dimensional column liquid chromatographic technique for resolution of complex mixtures. J Chromatogr A
1978;149:561-569.
[19] Bushey MM, Jorgenson JW. Automated instrumentation for comprehensive two-dimensional high-performance liquid chromatography
of proteins. Anal Chem 1990;62:161-7.
[20] Stoll DR, Li X, Wang X, Carr PW, Porter SEG, Rutan SC. Fast, comprehensive two-dimensional liquid chromatography. J Chromatogr
A 2007;1168:3-43;discussion 2.
[21] Schmidt SR, Schweikart F, Andersson ME. Current methods for phosphoprotein isolation and enrichment. J Chromatogr B, J
Chromatogr B Analyt Technol Biomed Life Sci 2007;849:154-62.
[22] Murphy, RE, Schure, MR. Instrumentation for Comprehensive Multidimensional Liquid Chromatography. In: Cohen SA (Ed.),
Multidimensional Liquid Chromatography. New Jersey: John Wiley & Sons, Inc.; 2008, 91-125.
[23] Millea KM, Krull IS, Cohen SA, Gebler JC, Berger SJ. Integration of multidimensional chromatographic protein separations with a
combined "top-down" and "bottom-up" proteomic strategy. J Proteome Res 2006;5:135-46.
[24] Brown JR, Hartley BS. Location of disulphide bridges by diagonal paper electrophoresis. The disulphide bridges of bovine
chymotrypsinogen A. Biochem J 1966;101:214-28.
[25] Sommer A, Traut RR. Diagonal polyacrylamide-dodecyl sulfate gel electrophoresis for the identification of ribosomal proteins
crosslinked with methyl-4-mercaptobutyrimidate. Proc Natl Acad Sci USA 1974;71:3946-50.
[26] Samelson LE. Diagonal gel electrophoresis. Curr Protoc Immunol 2001; Chapter 8:Unit 8.6.
[27] Ghesquière B, Colaert N, Helsens K, Dejager L, Vanhaute C, Verleysen K, Kas K, Timmerman E, Goethals M, Libert C,
Vandekerckhove J, Gevaert K. In vitro and in vivo protein-bound tyrosine nitration characterized by diagonal chromatography. Mol Cell
Proteomics 2009;8:2642-52.
[28] Liu P, Feasley CL, Regnier FE. Optimization of diagonal chromatography for recognizing post-translational modifications. J
Chromatogr A 2004;1047:221-7.
[29] Ghesquière B, Van Damme J, Martens L, Vandekerckhove J, Gevaert K. Proteome-wide characterization of N-glycosylation events by
diagonal chromatography. J Proteome Res 2006;5:2438-47.
[30] Gevaert K, Vandekerckhove J. Reverse-phase diagonal chromatography for phosphoproteome research. Methods in Molecular Biology
(Clifton, N.J.) 2009; 527:219-27.
[31] Wojcik R, Vannatta M, Dovichi NJ. Automated enzyme-based diagonal capillary electrophoresis: application to phosphopeptide
characterization. Anal Chem 2010;82:1564-7.
[32] Kee J, Villani B, Carpenter LR, Muir TW. Development of stable phosphohistidine analogues. J Am Chem Soc 2010;132:14327-9.
[33] Döppler H, Storz P, Li J, Comb MJ, Toker A. A phosphorylation state-specific antibody recognizes Hsp27, a novel substrate of protein
kinase D. J Biol Chem 2005;280:15013-9.
[34] Demesa Balderrama G, Meneses EP, Hernández Orihuela L, Villa Hernández O, Castro Franco R, Pando Robles V, Ferreira Batista CV.
Analysis of sulfated peptides from the skin secretion of the Pachymedusa dacnicolor frog using IMAC-Ga enrichment and high-resolution
mass spectrometry. Rapid Commun Mass Spectrom 2011;25:1017-27.
[35] Heal WP, Tate EW. Getting a chemical handle on protein post-translational modification. Org Biomol Chem 2010;8:731-8.
[36] Yang Y, Hahne H, Kuster B, Verhelst SHL. A Simple and Effective Cleavable Linker for Chemical Proteomics Applications. Mol Cell
Proteomics 2012;(ahead of print, doi: 10.1074/mcp.M112.021014).
[37] Giavalisco P, Nordhoff E, Kreitler T, Klöppel K, Lehrach H, Klose J, Gobom J. Proteome analysis of Arabidopsis thaliana by twodimensional
gel electrophoresis and matrix-assisted laser desorption/ionisation-time of flight mass spectrometry. Proteomics 2005;5:1902-
13.
[38] Baerenfaller K, Hirsch-Hoffmann M, Svozil J, Hull R, Russenberger D, Bischof S, Lu Q, Gruissem W, Baginsky S. pep2pro: a new tool
for comprehensive proteome data analysis to reveal information about organ-specific proteomes in Arabidopsis thaliana. Integr Biol (Camb)
2011;3:225-37.
[39] Boschetti E, Bindschedler LV, Tang C, Fasoli E, Righetti PG. Combinatorial peptide ligand libraries and plant proteomics: a winning
strategy at a price. J Chromatogr A 2009;1216:1215-22.
[40] Qian W, Kaleta DT, Petritis BO, Jiang H, Liu T, Zhang X, Mottaz HM, Varnum SM, Camp DG2, Huang L, Fang X, Zhang W, Smith
RD. Enhanced detection of low abundance human plasma proteins using a tandem IgY12-SuperMix immunoaffinity separation strategy. Mol
Cell Proteomics 2008;7:1963-73.
[41] Huse WD, Sastry L, Iverson SA, Kang AS, Alting-Mees M, Burton DR, Benkovic SJ, Lerner RA. Generation of a large combinatorial
library of the immunoglobulin repertoire in phage lambda. Science 1989;246:1275-81.
[42] Fröhlich A, Gaupels F, Sarioglu H, Holzmeister C, Spannagl M, Durner J, Lindermayr C. Looking deep inside: detection of lowabundance
proteins in leaf extracts of Arabidopsis and phloem exudates of pumpkin. Plant Physiol 2012;159:902-14.
[43] Candiano G, Dimuccio V, Bruschi M, Santucci L, Gusmano R, Boschetti E, Righetti PG, Ghiggeri GM. Combinatorial peptide ligand
libraries for urine proteome analysis: investigation of different elution systems. Electrophoresis 2009;30:2405-11.
[44] Jiang J, Opanubi KJ, Coombs KM. Non-biased enrichment does not improve quantitative proteomic delineation of reovirus T3Dinfected
HeLa cell protein alterations. Front Microbiol 2012;3: 1-16.
[45] Černý M, Skalák J, Kurková B, Babuliaková E, Brzobohatý B. Using a commercial method for rubisco immunodepletion in analysis of
plant proteom. Chem listy 2011;105:640-642.
[46] Besant PG, Attwood PV, Piggott MJ. Focus on phosphoarginine and phospholysine. Curr Protein Pept Sci Current 2009;10:536-50.
[47] Attwood PV. PN bond protein phosphatases. Biochim Biophys Acta 2012;DOI:10.1016/j.bbapap.2012.03.001.
[48] Andersen JK. What causes the build-up of ubiquitin-containing inclusions in Parkinson's disease? Mech Ageing Dev2000;118:15-22.
[49] The UniProt Consortium. Reorganizing the protein space at the Universal Protein Resource (UniProt). Nucleic Acids Res 2012;40:D71-
5.
[50] Yamakura F, Ikeda K. Modification of tryptophan and tryptophan residues in proteins by reactive nitrogen species. Nitric Oxide
2006;14:152-61.
[51] Lesgards J, Frayne IR, Comte B, Busseuil D, Rhéaume Ã, Tardif J, Rosiers CD. Differential distribution of 4-hydroxynonenal
adducts to sulfur and nitrogen residues in blood proteins as revealed using Raney nickel and gas chromatography-mass spectrometry. Free
Radic Biol Med 2009;47:1375-1385.
[52] Collins MO, Yu L, Choudhary JS. Analysis of protein phosphorylation on a proteome-scale. Proteomics 2007;7:2751-68.
[53] Černý M, Dyčka F, Bobál'ová J, Brzobohatý B. Early cytokinin response proteins and phosphoproteins of Arabidopsis thaliana
identified by proteome and phosphoproteome profiling. J Exp Bot 2011;62:921-37.
[54] Černý M, Doubnerová V, Müller K, Ryšlavá H. Characterization of phosphoenolpyruvate carboxylase from mature maize seeds:
properties of phosphorylated and dephosphorylated forms. Biochimie 2010;92:1362-70.
[55] Youn JH, Shin J. Nucleocytoplasmic shuttling of HMGB1 is regulated by phosphorylation that redirects it toward secretion. J Immunol
2006;177:7889-97.
[56] Kubicek K, Cerna H, Holub P, Pasulka J, Hrossova D, Loehr F, Hofr C, Vanacova S, Stefl R. Serine phosphorylation and proline
isomerization in RNAP II CTD control recruitment of Nrd1. Genes Dev 2012;26:1891-6.
[57] Shaywitz AJ, Dove SL, Greenberg ME, Hochschild A. Analysis of phosphorylation-dependent protein-protein interactions using a
bacterial two-hybrid system. Sci STKE 2002;2002:PL11.
[58] Calvert MEK, Keck KM, Ptak C, Shabanowitz J, Hunt DF, Pemberton LF. Phosphorylation by casein kinase 2 regulates Nap1
localization and function. Mol Cell Biol 2008;28:1313-25.
[59] Kruse O, Zheleva D, Barber J. Stabilization of photosystem two dimers by phosphorylation: implication for the regulation of the
turnover of D1 protein. FEBS Lett 1997;408:276-80.
[60] Rokka A, Aro EM, Herrmann RG, Andersson B, Vener AV. Dephosphorylation of photosystem II reaction center proteins in plant
photosynthetic membranes as an immediate response to abrupt elevation of temperature. Plant Physiol 2000;123:1525-36.
[61] Kersten B, Agrawal GK, Iwahashi H, Rakwal R. Plant phosphoproteomics: a long road ahead. Proteomics 2006;6:5517-28.
[62] Durek P, Schmidt R, Heazlewood JL, Jones A, MacLean D, Nagel A, Kersten B, Schulze WX. PhosPhAt: the Arabidopsis thaliana
phosphorylation site database. An update. Nucleic Acids Res 2010;38:D828-34.
[63] Laugesen S, Bergoin A, Rossignol M. Deciphering the plant phosphoproteome: tools and strategies for a challenging task. Plant Physiol
Biochem 2004;42:929-36.
[64] Klumpp S, Krieglstein J. Phosphorylation and dephosphorylation of histidine residues in proteins. Eur J Biochem 2002;269:1067-71.
[65] Cieśla J, Frączyk T, Rode W. Phosphorylation of basic amino acid residues in proteins: important but easily missed. Acta Biochim Pol
2011;58:137-48.
[66] Delom F, Chevet E. Phosphoprotein analysis: from proteins to proteomes. Proteome Sci 2006;4:15.
[67] Agrawal GK, Thelen JJ. Development of a simplified, economical polyacrylamide gel staining protocol for phosphoproteins. Proteomics
2005;:4684-8.
[68] Kinoshita E, Takahashi M, Takeda H, Shiro M, Koike T. Recognition of phosphate monoester dianion by an alkoxide-bridged dinuclear
zinc(II) complex. Dalton Trans 2004;1189-93.
[69] Beck M, Komis G, Müller J, Menzel D, Samaj J. Arabidopsis homologs of nucleus- and phragmoplast-localized kinase 2 and 3 and
mitogen-activated protein kinase 4 are essential for microtubule organization. Plant Cell 2010;22:755-71.
[70] Barbieri CM, Stock AM. Universally applicable methods for monitoring response regulator aspartate phosphorylation both in vitro and
in vivo using Phos-tag-based reagents. Anal Biochem 2008;376:73-82.
[71] Porath J, Carlsson J, Olsson I, Belfrage G. Metal chelate affinity chromatography, a new approach to protein fractionation. Nature
1975;258:598-9.
[72] Andersson L, Porath J. Isolation of phosphoproteins by immobilized metal (Fe3+) affinity chromatography. Anal Biochem
1986;154:250-4.
[73] Matsuda H, Nakamura H, Nakajima T. New Ceramic Titania: Selective Adsorbent for Organic Phosphates. Anal Sci 1990;6:911-912.
[74] Kweon HK, Håkansson K. Selective zirconium dioxide-based enrichment of phosphorylated peptides for mass spectrometric analysis.
Anal Chem 2006;78:1743-9.
[75] Kettenbach AN, Gerber SA. Rapid and reproducible single-stage phosphopeptide enrichment of complex peptide mixtures: application
to general and phosphotyrosine-specific phosphoproteomics experiments. Anal Chem 2011;83:7635-44.
[76] Wolschin F, Wienkoop S, Weckwerth W. Enrichment of phosphorylated proteins and peptides from complex mixtures using metal
oxide/hydroxide affinity chromatography (MOAC). Proteomics 2005;5:4389-97.
[77] Zhao L, Qin H, Hu Z, Zhang Y, Wu R, Zou H. A poly(ethylene glycol)-brush decorated magnetic polymer for highly specific
enrichment of phosphopeptides. Chem. Sci. 2012;3:2828-2838.
[78] Iliuk AB, Martin VA, Alicie BM, Geahlen RL, Tao WA. In-depth analyses of kinase-dependent tyrosine phosphoproteomes based on
metal ion-functionalized soluble nanopolymers. Mol Cell Proteomics 2010;9:2162-72.
[79] Grønborg M, Kristiansen TZ, Stensballe A, Andersen JS, Ohara O, Mann M, Jensen ON, 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;1:517-27.
[80] Fíla J, Honys D. Enrichment techniques employed in phosphoproteomics. Amino Acids 2012;43:1025-47.
[81] Oda Y, Nagasu T, Chait BT. Enrichment analysis of phosphorylated proteins as a tool for probing the phosphoproteome. Nat
Biotechnol. 2001;19:379-82.
[82] Rusnak F, Zhou J, Hathaway GM. Reaction of phosphorylated and O-glycosylated peptides by chemically targeted identification at
ambient temperature. J Biomol Tech 2004;15:296-304.
[83] Hoehenwarter W, Thomas M, Nukarinen E, Egelhofer V, Roehrig H, Weckwerth W, Conrath U, Beckers GJM. Tandem metal oxide
affinity chromatography identifies novel in vivo MAP kinase substrates in Arabidopsis thaliana. Mol Cell Proteomics 2012;(ahead of print,
doi: 10.1074/mcp.M112.020560 mcp.M112.020560.).
[84] D'Autréaux B, Toledano MB. ROS as signalling molecules: mechanisms that generate specificity in ROS homeostasis. Nat Rev Mol
Cell Biol 2007;8:813-24.
[85] Spickett CM, Pitt AR. Protein oxidation: role in signalling and detection by mass spectrometry. Amino Acids 2012;42:5-21.
[86] Rinalducci S, Murgiano L, Zolla L. Redox proteomics: basic principles and future perspectives for the detection of protein oxidation in
plants. J Exp Bot 2008;59:3781-801.
[87] Stadtman ER, Levine RL. Protein oxidation. Ann N Y Acad Sci 2000;899:191-208.
[88] Grune T, Shringarpure R, Sitte N, Davies K. Age-related changes in protein oxidation and proteolysis in mammalian cells. J Gerontol A
Biol Sci Med Sci 2001;56:B459-67.
[89] Levine RL. Carbonyl modified proteins in cellular regulation, aging, and disease. Free Radic Biol Med 2002;32:790-6.
[90] König J, Muthuramalingam M, Dietz K. Mechanisms and dynamics in the thiol/disulfide redox regulatory network: transmitters, sensors
and targets. Curr Opin Plant Biol 2012;15:261-8.
[91] Sun M, Wang Y, Cheng H, Zhang Q, Ge W, Guo D. RedoxDB--a curated database for experimentally verified protein oxidative
modification. Bioinformatics 2012;28:2551-2.
[92] Gao X, Bedhomme M, Veyel D, Zaffagnini M, Lemaire SD. Methods for analysis of protein glutathionylation and their application to
photosynthetic organisms. Mol Plant 2009;2:218-35.
[93] Jaffrey SR, Snyder SH. The biotin switch method for the detection of S-nitrosylated proteins. Sci STKE 2001;2001:pl1.
[94] Murray CI, Van Eyk JE. Chasing cysteine oxidative modifications: proteomic tools for characterizing cysteine redox status. Circ
Cardiovasc Genet 2012;5:591.
[95] Izquierdo-Álvarez A, Ramos E, Villanueva J, Hernansanz-Agustín P, Fernández-Rodríguez R, Tello D, Carrascal M, Martínez-Ruiz A.
Differential redox proteomics allows identification of proteins reversibly oxidized at cysteine residues in endothelial cells in response to
acute hypoxia. J Proteomics 2012;75:5449-62.
[96] Seo YH, Carroll KS. Profiling protein thiol oxidation in tumor cells using sulfenic acid-specific antibodies. Proc Natl Acad Sci USA
2009;06:16163-8.
[97] Fujiwara N, Nakano M, Kato S, Yoshihara D, Ookawara T, Eguchi H, Taniguchi N, Suzuki K. Oxidative modification to cysteine
sulfonic acid of Cys111 in human copper-zinc superoxide dismutase. J Biol Chem 2007;282:35933-44.
[98] Lindahl M, Mata-Cabana A, Kieselbach T. The disulfide proteome and other reactive cysteine proteomes: analysis and functional
significance. Antioxid Redox Signal 2011;14:2581-642.
[99] Forrester MT, Thompson JW, Foster MW, Nogueira L, Moseley MA, Stamler JS. Proteomic analysis of S-nitrosylation and
denitrosylation by resin-assisted capture. Nat Biotechnol 2009;27:557-9.
[100] Hisabori T, Hara S, Fujii T, Yamazaki D, Hosoya-Matsuda N, Motohashi K. Thioredoxin affinity chromatography: a useful method for
further understanding the thioredoxin network. J Exp Bot 2005;56:1463-8.
[101] Mata-Cabana A, Florencio FJ, Lindahl M. Membrane proteins from the cyanobacterium Synechocystis sp. PCC 6803 interacting with
thioredoxin. Proteomics 2007;7:3953-63.
[102] Sullivan DM, Levine RL, Finkel T. Detection and affinity purification of oxidant-sensitive proteins using biotinylated glutathione ethyl
ester. Methods Enzymol 2002;353:101-13.
[103] Brennan JP, Miller JIA, Fuller W, Wait R, Begum S, Dunn MJ, Eaton P. The Utility of N,N-Biotinyl Glutathione Disulfide in the
Study of Protein S-Glutathiolation. Mol Cell Proteomics 2006;5:215-225.
[104] Klatt P, Lamas S. Regulation of protein function by S-glutathiolation in response to oxidative and nitrosative stress. Eur J Biochem
2000;67:4928-44.
[105] Martínez-Acedo P, Núñez E, Gómez FJS, Moreno M, Ramos E, Izquierdo-Álvarez A, Miró-Casas E, Mesa R, Rodriguez P, MartínezRuiz
A, Dorado DG, Lamas S, Vázquez J. A novel strategy for global analysis of the dynamic thiol redox proteome. Mol Cell Proteomics
2012;11:800-13.
[106] Nuriel T, Hansler A, Gross SS. Protein nitrotryptophan: formation, significance and identification. J Proteomics 2011;74:2300-12.
[107] Greenacre SA, Ischiropoulos H. Tyrosine nitration: localisation, quantification, consequences for protein function and signal
transduction. Free Radic Res 2001;34:541-81.
[108] Yang H, Zhang Y, Pöschl U. Quantification of nitrotyrosine in nitrated proteins. Anal Bioanal Chem 2010;397:879-86.
[109] Ikeda K, Yukihiro Hiraoka B, Iwai H, Matsumoto T, Mineki R, Taka H, Takamori K, Ogawa H, Yamakura F. Detection of 6nitrotryptophan
in proteins by Western blot analysis and its application for peroxynitrite-treated PC12 cells. Nitric Oxide 2007;16:18-28.
[110] Abello N, Barroso B, Kerstjens HAM, Postma DS, Bischoff R. Chemical labeling and enrichment of nitrotyrosine-containing peptides.
Talanta 2010;80:1503-12.
[111] Levine RL, Berlett BS, Moskovitz J, Mosoni L, Stadtman ER. Methionine residues may protect proteins from critical oxidative
damage. Mech Ageing Dev1999;107:323-32.
[112] Hardin SC, Larue CT, Oh M, Jain V, Huber SC. Coupling oxidative signals to protein phosphorylation via methionine oxidation in
Arabidopsis. Biochem J 2009;422:305-12.
[113] Ciorba MA, Heinemann SH, Weissbach H, Brot N, Hoshi T. Modulation of potassium channel function by methionine oxidation and
reduction. Proc Natl Acad Sci USA 1997;94:9932-7.
[114] Oien DB, Canello T, Gabizon R, Gasset M, Lundquist BL, Burns JM, Moskovitz J. Detection of oxidized methionine in selected
proteins, cellular extracts and blood serums by novel anti-methionine sulfoxide antibodies. Arch Biochem Biophys 2009;485:35-40.
[115] Wehr NB, Levine RL. Wanted and wanting: antibody against methionine sulfoxide. Free Radic Biol Med 2012;53:1222-5.
[116] Rosen H, Klebanoff SJ, Wang Y, Brot N, Heinecke JW, Fu X. Methionine oxidation contributes to bacterial killing by the
myeloperoxidase system of neutrophils. Proc Natl Acad Sci USA 2009;106:18686-91.
[117] Ghesquière B, Jonckheere V, Colaert N, Van Durme J, Timmerman E, Goethals M, Schymkowitz J, Rousseau F, Vandekerckhove J,
Gevaert K. Redox proteomics of protein-bound methionine oxidation. Mol Cell Proteomics 2011;10:M110.006866.
[118] Stadtman ER, Berlett BS. Fenton chemistry. Amino acid oxidation. J Biol Chem 1991;266:17201-11.
[119] Wong CM, Marcocci L, Liu L, Suzuki YJ. Cell signaling by protein carbonylation and decarbonylation. Antioxid Redox Signal
2010;12:393-404.
[120] Grimsrud PA, Xie H, Griffin TJ, Bernlohr DA. Oxidative stress and covalent modification of protein with bioactive aldehydes. J Biol
Chem 2008;283:21837-41.
[121] Requena JR, Chao CC, Levine RL, Stadtman ER. Glutamic and aminoadipic semialdehydes are the main carbonyl products of metalcatalyzed
oxidation of proteins. Proc Natl Acad Sci USA 2001;98:69-74.
[122] Madian AG, Regnier FE. Proteomic identification of carbonylated proteins and their oxidation sites. J Proteome Res 2010;9:3766-80.
[123] Yuan Q, Zhu X, Sayre LM. Chemical nature of stochastic generation of protein-based carbonyls: metal-catalyzed oxidation versus
modification by products of lipid oxidation. Chem Res Toxicol 2007;20:129-39.
[124] Oe T, Arora JS, Lee SH, Blair IA. A novel lipid hydroperoxide-derived cyclic covalent modification to histone H4. J Biol Chem
2003;78:42098-105.
[125] Moller IM, Rogowska-Wrzesinska A, Rao RSP. Protein carbonylation and metal-catalyzed protein oxidation in a cellular perspective. J
Proteomics 2011;74:2228-42.
[126] Yan L, Forster MJ. Chemical probes for analysis of carbonylated proteins: a review. J Chromatogr B Analyt Technol Biomed Life Sci
2011;79:1308-15.
[127] Sultana R, Butterfield DA. Proteomics identification of carbonylated and HNE-bound brain proteins in Alzheimer's disease. Methods
Mol Biol 2009;566:123-35.
[128] Dalle-Donne I, Carini M, Orioli M, Vistoli G, Regazzoni L, Colombo G, Rossi R, Milzani A, Aldini G. Protein carbonylation: 2,4dinitrophenylhydrazine
reacts with both aldehydes/ketones and sulfenic acids. Free Radic Biol Med 2009;46:1411-9.
[129] Soreghan BA, Yang F, Thomas SN, Hsu J, Yang AJ. High-throughput proteomic-based identification of oxidatively induced protein
carbonylation in mouse brain. Pharm Res 2003;20:1713-20.
[130] Ugur Z, Coffey CM, Gronert S. Comparing the efficiencies of hydrazide labels in the study of protein carbonylation in human serum
albumin. Analytical and Bioanal Chem 2012;404:1399-411.
[131] Mirzaei H, Regnier F. Enrichment of carbonylated peptides using Girard P reagent and strong cation exchange chromatography. Anal
Chem 2006;78:770-8.
[132] Perdivara I, Deterding LJ, Przybylski M, Tomer KB. Mass spectrometric identification of oxidative modifications of tryptophan
residues in proteins: chemical artifact or post-translational modification? J Am Soc Mass Spectrom 2010;21:1114-7.
[133] Nadolski MJ, Linder ME. Protein lipidation. FEBS J 2007;274:5202-10.
[134] Hang HC, Wilson JP, Charron G. Bioorthogonal chemical reporters for analyzing protein lipidation and lipid trafficking. Acc Chem
Res 2011;44:699-708.
[135] Triola G. The Protein Lipidation and Its Analysis. J Glycomics & Lipidomics 2011;S2:1-14.
[136] Schmidt RA, Schneider CJ, Glomset JA. Evidence for post-translational incorporation of a product of mevalonic acid into Swiss 3T3
cell proteins. J Biol Chem 1984;259:10175-80.
[137] Berndt N, Sebti SM. Measurement of protein farnesylation and geranylgeranylation in vitro, in cultured cells and in biopsies, and the
effects of prenyl transferase inhibitors. Nat Protoc 2011;6:1775-91.
[138] Resh MD. Trafficking and signaling by fatty-acylated and prenylated proteins. Nat Chem Biol 2006;2:584-90.
[139] Zhang FL, Casey PJ. Protein prenylation: molecular mechanisms and functional consequences. Annu Rev Biochem 1996;65:241-69.
[140] Fukata Y, Fukata M. Protein palmitoylation in neuronal development and synaptic plasticity. Nat Rev Neurosci 2010;11:161-75.
[141] Bijlmakers M, Marsh M. The on-off story of protein palmitoylation. Trends Cell Biol 2003;13:32-42.
[142] Paulick MG, Bertozzi CR. The glycosylphosphatidylinositol anchor: a complex membrane-anchoring structure for proteins.
Biochemistry 2008;7:6991-7000.
[143] Takada R, Satomi Y, Kurata T, Ueno N, Norioka S, Kondoh H, Takao T, Takada S. Monounsaturated fatty acid modification of Wnt
protein: its role in Wnt secretion. Dev Cell 2006;11:791-801.
[144] Porter JA, Young KE, Beachy PA. Cholesterol modification of hedgehog signaling proteins in animal development. Science
1996;274:255-9.
[145] Vainauskas S, Menon AK. Ethanolamine phosphate linked to the first mannose residue of glycosylphosphatidylinositol (GPI) lipids is
a major feature of the GPI structure that is recognized by human GPI transamidase. J Biol Chem 2006;281:38358-64.
[146] Varma Y, Hendrickson T. Methods to study GPI anchoring of proteins. Chembiochem 2010;11:623-36.
[147] Ferguson MA, Low MG, Cross GA. Glycosyl-sn-1,2-dimyristylphosphatidylinositol is covalently linked to Trypanosoma brucei
variant surface glycoprotein. J Biol Chem 1985;260:14547-55.
[148] Wenk MR. Lipidomics: new tools and applications. Cell 2010;143:888-95.
[149] Saxon E, Bertozzi CR. Cell surface engineering by a modified Staudinger reaction. Science 2000;287:2007-10.
[150] Rostovtsev VV, Green LG, Fokin VV, Sharpless KB. A stepwise huisgen cycloaddition process: copper(I)-catalyzed regioselective
"ligation" of azides and terminal alkynes. Angew Chem Int Ed Engl 2002;41:2596-9.
[151] Tornøe CW, Christensen C, Meldal M. Peptidotriazoles on solid phase: [1,2,3]-triazoles by regiospecific copper(i)-catalyzed 1,3dipolar
cycloadditions of terminal alkynes to azides. J Org Chem 2002;67:3057-64.
[152] Sletten EM, Bertozzi CR. Bioorthogonal chemistry: fishing for selectivity in a sea of functionality. Angew Chem Int Ed Engl
2009;48:6974-98.
[153] Charron G, Wilson J, Hang HC. Chemical tools for understanding protein lipidation in eukaryotes. Curr Opin Chem Biol 2009;13:382-
91.
[154] Martin BR, Cravatt BF. Large-scale profiling of protein palmitoylation in mammalian cells. Nat Methods 2009;6:135-8.
[155] Zhang MM, Tsou LK, Charron G, Raghavan AS, Hang HC. Tandem fluorescence imaging of dynamic S-acylation and protein
turnover. Proc Natl Acad Sci USA 2010;107:8627-32.
[156] Martin DDO, Vilas GL, Prescher JA, Rajaiah G, Falck JR, Bertozzi CR, Berthiaume LG. Rapid detection, discovery, and identification
of post-translationally myristoylated proteins during apoptosis using a bio-orthogonal azidomyristate analog. FASEB J 2008;22:797-806.
[157] Kho Y, Kim SC, Jiang C, Barma D, Kwon SW, Cheng J, Jaunbergs J, Weinbaum C, Tamanoi F, Falck J, Zhao Y. A tagging-viasubstrate
technology for detection and proteomics of farnesylated proteins. Proc Natl Acad Sci USA 2004;101:12479-84.
[158] Yang Y, Hang HC. Chemical approaches for the detection and synthesis of acetylated proteins. Chembiochem 2011;12:314-22.
[159] Hang HC, Linder ME. Exploring protein lipidation with chemical biology. Chem Rev 2011;111:6341-58.
[160] Nguyen UTT, Guo Z, Delon C, Wu Y, Deraeve C, Fränzel B, Bon RS, Blankenfeldt W, Goody RS, Waldmann H, Wolters D,
Alexandrov K. Analysis of the eukaryotic prenylome by isoprenoid affinity tagging. Nat Chem Biol 2009;5:227-35.
[161] Whelan WJ. Pride and prejudice: the discovery of the primer for glycogen synthesis. Protein Sci 1998;7:2038-41.
[162] Kornfeld S. Foreword from First Edition. In: Varki A, Cummings RD, Esko JD (Eds.), Essentials of Glycobiology 2nd edition. New
York: Cold Spring Harbor; 2009.
[163] Rakus JF, Mahal LK. New technologies for glycomic analysis: toward a systematic understanding of the glycome. Annu Rev Anal
Chem (Palo Alto Calif) 2011;4:367-92.
[164] Ohtsubo K, Marth JD. Glycosylation in cellular mechanisms of health and disease. Cell 2006;126:855-67.
[165] Zielinska DF, Gnad F, Wisniewski JR, Mann M. Precision mapping of an in vivo N-glycoproteome reveals rigid topological and
sequence constraints. Cell 2010;141:897-907.
[166] Priego Capote F, Sanchez J. Strategies for proteomic analysis of non-enzymatically glycated proteins. Mass Spectrom Rev
2009;28:135-46.
[167] An HJ, Lebrilla CB. Structure elucidation of native N- and O-linked glycans by tandem mass spectrometry (tutorial). Mass Spectrom
Rev 2011;30:560-78.
[168] Mulloy B, Hart GW, Stanley P. Structural Analysis of Glycans. In: Varki A, Cummings RD, Esko JD (Eds.), Essentials of
Glycobiology 2nd edition. New York: Cold Spring Harbor; 2009.
[169] Laughlin ST, Agard NJ, Baskin JM, Carrico IS, Chang PV, Ganguli AS, Hangauer MJ, Lo A, Prescher JA, Bertozzi CR. Metabolic
labeling of glycans with azido sugars for visualization and glycoproteomics. Methods Enzymol 2006;415:230-50.
[170] Zacharius RM, Zell TE, Morrison JH, Woodlock JJ. Glycoprotein staining following electrophoresis on acrylamide gels. Anal
Biochem 1969;30:148-52.
[171] Miller I, Crawford J, Gianazza E. Protein stains for proteomic applications: which, when, why? Proteomics 2006;6:5385-408.
[172] Gauci VJ, Wright EP, Coorssen JR. Quantitative proteomics: assessing the spectrum of in-gel protein detection methods. J Chem Biol
2011;4:3-29.
[173] Wollscheid B, Bausch-Fluck D, Henderson C, O'Brien R, Bibel M, Schiess R, Aebersold R, Watts JD. Mass-spectrometric
identification and relative quantification of N-linked cell surface glycoproteins. Nat Biotechnol 2009;27:378-86.
[174] Steinberg TH, Pretty On Top K, Berggren KN, Kemper C, Jones L, Diwu Z, Haugland RP, Patton WF. Rapid and simple single
nanogram detection of glycoproteins in polyacrylamide gels and on electroblots. Proteomics 2001;1:841-55.
[175] Kocourek J, Hořejší V. Defining a lectin. Nature 1981;290:188.
[176] Fanayan S, Hincapie M, Hancock WS. Using lectins to harvest the plasma/serum glycoproteome. Electrophoresis 2012;33:1746-54.
[177] Chen S, LaRoche T, Hamelinck D, Bergsma D, Brenner D, Simeone D, Brand RE, Haab BB. Multiplexed analysis of glycan variation
on native proteins captured by antibody microarrays. Nat Methods 2007;4:437-44.
[178] Yue T, Haab BB. Microarrays in glycoproteomics research. Clin Lab Med 2009;29:15-29.
[179] Forrester S, Hung KE, Kuick R, Kucherlapati R, Haab BB. Low-volume, high-throughput sandwich immunoassays for profiling
plasma proteins in mice: Identification of early-stage systemic inflammation in a mouse model of intestinal cancer. Mol Oncol 2007;1:216-
225.
[180] Edge ASB. Deglycosylation of glycoproteins with trifluoromethanesulphonic acid: elucidation of molecular structure and function.
Biochem J 2003;376:339-50.
[181] Yang Z, Harris LE, Palmer-Toy DE, Hancock WS. Multilectin affinity chromatography for characterization of multiple glycoprotein
biomarker candidates in serum from breast cancer patients. Clin Chem 2006;52:1897-905.
[182] Ueda K, Takami S, Saichi N, Daigo Y, Ishikawa N, Kohno N, Katsumata M, Yamane A, Ota M, Sato T, Nakamura Y, Nakagawa H.
Development of serum glycoproteomic profiling technique; simultaneous identification of glycosylation sites and site-specific quantification
of glycan structure changes. Mol Cell Proteomics 2010;9:1819-28.
[183] Medvedev A, Kopylov A, Buneeva O, Zgoda V, Archakov A. Affinity-based proteomic profiling: problems and achievements.
Proteomics 2012;12:621-37.
[184] Lazar IM, Lee W, Lazar AC. Glycoproteomics on the Rise: Established Methods, Advanced Techniques, Sophisticated Biological
Applications. Electrophoresis 2012; (ahead of print; doi: 10.1002/elps.201200445).
[185] Zeng Y, Ramya TNC, Dirksen A, Dawson PE, Paulson JC. High-efficiency labeling of sialylated glycoproteins on living cells. Nat
Methods 2009;6:207-9.
[186] Ramya TNC, Weerapana E, Cravatt BF, Paulson JC. Glycoproteomics enabled by tagging sialic acid or galactose terminated glycans.
Glycobiology 2012; (ahead of print, doi: 10.1093/glycob/cws144).
[187] Sun Z, Qin H, Wang F, Cheng K, Dong M, Ye M, Zou H. Capture and dimethyl labeling of glycopeptides on hydrazide beads for
quantitative glycoproteomics analysis. Anal Chem 2012;84:8452-6.
[188] Zhang Y, Yin H, Lu H. Recent progress in quantitative glycoproteomics. Glycoconj J 2012;29:249-58.
[189] Palmisano G, Melo-Braga MN, Engholm-Keller K, Parker BL, Larsen MR. Chemical deamidation: a common pitfall in large-scale Nlinked
glycoproteomic mass spectrometry-based analyses. J Proteome Res 2012;11:1949-57.
[190] Duk M, Ugorski M, Lisowska E. beta-Elimination of O-glycans from glycoproteins transferred to immobilon P membranes: method
and some applications. Anal Biochem 1997;253:98-102.
[191] Wada Y, Tajiri M, Ohshima S. Quantitation of saccharide compositions of O-glycans by mass spectrometry of glycopeptides and its
application to rheumatoid arthritis. J Proteome Res 2010;9:1367-73.
[192] Liu X, Scouten W. Boronate Affinity Chromatography. Methods Mol Biol 2000;147:119-128.
[193] Zhang Q, Tang N, Schepmoes AA, Phillips LS, Smith RD, Metz TO. Proteomic profiling of nonenzymatically glycated proteins in
human plasma and erythrocyte membranes. J Proteome Res 2008;7:2025-32.
[194] Jin S, Cheng Y, Reid S, Li M, Wang B. Carbohydrate recognition by boronolectins, small molecules, and lectins. Med Res Rev
2010;30:171-257.
[195] Li X, Pennington J, Stobaugh JF, Schöneich C. Synthesis of sulfonamide- and sulfonyl-phenylboronic acid-modified silica phases for
boronate affinity chromatography at physiological pH. Anal Biochem 2008;372:227-36.
[196] Monzo A, Bonn GK, Guttman A. Boronic acid-lectin affinity chromatography. 1. Simultaneous glycoprotein binding with selective or
combined elution. Anal Bioanal Chem 2007;389:2097-2102.
[197] Morais MPP, Mackay JD, Bhamra SK, Buchanan JG, James TD, Fossey JS, van den Elsen JMH. Analysis of protein glycation using
phenylboronate acrylamide gel electrophoresis. Proteomics 2010;10:48-58.
[198] Koito W, Araki T, Horiuchi S, Nagai R. Conventional Antibody against Nepsilon-(carboxymethyl)lysine (CML) shows cross-reaction
to Nepsilon-(carboxyethyl)lysine (CEL): immunochemical quantification of CML with a specific antibody. J Biochem 2004;136:831-837.
[199] Nagai R, Fujiwara Y, Mera K, Motomura K, Iwao Y, Tsurushima K, Nagai M, Takeo K, Yoshitomi M, Otagiri M, Ikeda T. Usefulness
of antibodies for evaluating the biological significance of AGEs. Ann N Y Acad Sci 2008;126:38-41.
[200] Robinson NE, Robinson AB. Deamidation of human proteins. Proc Natl Acad Sci USA 2001;98:12409-13.
[201] Jaisson S, Gillery P. Evaluation of nonenzymatic posttranslational modification-derived products as biomarkers of molecular aging of
proteins. Clin Chem 2010;56:1401-12.
[202] Hao P, Ren Y, Alpert AJ, Sze SK. Detection, evaluation and minimization of nonenzymatic deamidation in proteomic sample
preparation. Mol Cell Proteomics 2011;10:O111.009381-11.
[203] Gaza-Bulseco G, Li B, Bulseco A, Liu HC. Method to differentiate asn deamidation that occurred prior to and during sample
preparation of a monoclonal antibody. Anal Chem 2008;80:9491-8.
[204] Li X, Cournoyer JJ, Lin C, O'Connor PB. Use of 18O labels to monitor deamidation during protein and peptide sample processing. J
Am Soc Mass Spectrom 2008;19:855-64.
[205] Bae N, Yang J, Sitte H, Pollak A, Marquez J, Lubec G. An electrophoretic approach to screen for glutamine deamidation. Anal
Biochem 2012;428:1-3.
[206] Bischoff R, Schlüter H. Amino acids: chemistry, functionality and selected non-enzymatic post-translational modifications. J
Proteomics 2012;5:2275-96.
[207] Reissner KJ, Aswad DW. Deamidation and isoaspartate formation in proteins: unwanted alterations or surreptitious signals? Cell Mol
Life Sci 2003;60:1281-95.
[208] Yang H, Zubarev RA. Mass spectrometric analysis of asparagine deamidation and aspartate isomerization in polypeptides.
Electrophoresis 2010;31:1764-72.
[209] Hao P, Qian J, Dutta B, Cheow ESH, Sim KH, Meng W, Adav SS, Alpert A, Sze SK. Enhanced separation and characterization of
deamidated peptides with RP-ERLIC-based multidimensional chromatography coupled with tandem mass spectrometry. J Proteome Res
2012;11:1804-11.
[210] Narita K. Isolation of acetylpeptide from enzymic digests of TMV-protein. Biochim Biophys Acta 1958;28:184-91.
[211] Phillips DM. The presence of acetyl groups of histones. Biochem J 1963;87:258-63.
[212] Allfrey VG, Faulkner R, Mirsky AE. Acetylation and methylation of histones and their possible role in the regulation of RNA
synthesis. Proc Natl Acad Sci USA 1964;51:786-94.
[213] Mukherjee S, Hao Y, Orth K. A newly discovered post-translational modification--the acetylation of serine and threonine residues.
Trends Biochem Sci 2007;32:210-6.
[214] Arnesen T, Van Damme P, Polevoda B, Helsens K, Evjenth R, Colaert N, Varhaug JE, Vandekerckhove J, Lillehaug JR, Sherman F,
Gevaert K. Proteomics analyses reveal the evolutionary conservation and divergence of N-terminal acetyltransferases from yeast and
humans. Proc Natl Acad Sci USA 2009;106:8157-62.
[215] Polevoda B, Sherman F. N-terminal acetyltransferases and sequence requirements for N-terminal acetylation of eukaryotic proteins. J
Mol Biol 2003;325:595-622.
[216] Hwang C, Shemorry A, Varshavsky A. N-terminal acetylation of cellular proteins creates specific degradation signals. Science
2010;327:973-7.
[217] Kouzarides T. Acetylation: a regulatory modification to rival phosphorylation? EMBO J 2000;19:1176-9.
[218] Spange S, Wagner T, Heinzel T, Krämer OH. Acetylation of non-histone proteins modulates cellular signalling at multiple levels. Int J
Biochem Cell Biol 2009;41:185-98.
[219] Guan K, Xiong Y. Regulation of intermediary metabolism by protein acetylation. Trends Biochem Sci 2011;36:108-16.
[220] Xie H, Bandhakavi S, Roe MR, Griffin TJ. Preparative peptide isoelectric focusing as a tool for improving the identification of lysineacetylated
peptides from complex mixtures. J Proteome Res 2007;6:2019-26.
[221] Kim SC, Sprung R, Chen Y, Xu Y, Ball H, Pei J, Cheng T, Kho Y, Xiao H, Xiao L, Grishin NV, White M, Yang X, Zhao Y. Substrate
and functional diversity of lysine acetylation revealed by a proteomics survey. Mol Cell 2006;23:607-18.
[222] Choudhary C, Kumar C, Gnad F, Nielsen ML, Rehman M, Walther TC, Olsen JV, Mann M. Lysine acetylation targets protein
complexes and co-regulates major cellular functions. Science 2009;325:834-40.
[223] Zhao S, Xu W, Jiang W, Yu W, Lin Y, Zhang T, Yao J, Zhou L, Zeng Y, Li H, Li Y, Shi J, An W, Hancock SM, He F, Qin L, Chin J,
Yang P, Chen X, Lei Q, Xiong Y, Guan K. Regulation of cellular metabolism by protein lysine acetylation. Science 2010;327:1000-4.
[224] Guan K, Yu W, Lin Y, Xiong Y, Zhao S. Generation of acetyllysine antibodies and affinity enrichment of acetylated peptides. Nat
Protoc 2010;5:1583-95.
[225] Shaw PG, Chaerkady R, Zhang Z, Davidson NE, Pandey A. Monoclonal antibody cocktail as an enrichment tool for acetylome
analysis. Anal Chem 2011;83:3623-6.
[226] Van Damme P, Van Damme J, Demol H, Staes A, Vandekerckhove J, Gevaert K. A review of COFRADIC techniques targeting
protein N-terminal acetylation. BMC Proc 2009;3 Suppl 6:S6.
[227] Ambler RP, Rees MW. Epsilon-N-Methyl-lysine in bacterial flagellar protein. Nature 1959;184:56-7.
[228] Grillo MA, Colombatto S. S-adenosylmethionine and protein methylation. Amino Acids 2005;28:357-62.
[229] Paik WK, Paik DC, Kim S. Historical review: the field of protein methylation. Trends Biochem Sci 2007;32:146-52.
[230] Bedford MT, Clarke SG. Protein arginine methylation in mammals: who, what, and why. Mol Cell 2009;33:1-13.
[231] Truscott RJW, Mizdrak J, Friedrich MG, Hooi MY, Lyons B, Jamie JF, Davies MJ, Wilmarth PA, David LL. Is protein methylation in
the human lens a result of non-enzymatic methylation by S-adenosylmethionine? Exp Eye Res 2012;99:48-54.
[232] Shi Y, Lan F, Matson C, Mulligan P, Whetstine JR, Cole PA, Casero RA, Shi Y. Histone demethylation mediated by the nuclear amine
oxidase homolog LSD1. Cell 2004;119:941-53.
[233] Ahmad A, Cao X. Plant PRMTs broaden the scope of arginine methylation. J Genet Genomics 2012;39:195-208.
[234] Mowen KA, David M. Analysis of protein arginine methylation and protein arginine-methyltransferase activity. Sci STKE
2001;2001:pl1.
[235] Ong S, Mittler G, Mann M. Identifying and quantifying in vivo methylation sites by heavy methyl SILAC. Nat Methods 2004;1:119-
26.
[236] Couttas TA, Raftery MJ, Padula MP, Herbert BR, Wilkins MR. Methylation of translation-associated proteins in Saccharomyces
cerevisiae: Identification of methylated lysines and their methyltransferases. Proteomics 2012;12:960-72.
[237] Uhlmann T, Geoghegan VL, Thomas B, Ridlova G, Trudgian DC, Acuto O. A Method for Large-scale Identification of Protein
Arginine Methylation. Mol Cell Proteomics 2012;11:1489-99.
[238] Jian W, Edom RW, Xu Y, Weng N. Recent advances in application of hydrophilic interaction chromatography for quantitative
bioanalysis. J Sep Sci 2010;33:681-697.
[239] Goldknopf IL, Busch H. Isopeptide linkage between nonhistone and histone 2A polypeptides of chromosomal conjugate-protein A24.
Proc Natl Acad Sci USA 1977;74:864-8.
[240] Shimizu Y, Okuda-Shimizu Y, Hendershot LM. Ubiquitylation of an ERAD substrate occurs on multiple types of amino acids. Mol
Cell 2010;40:917-26.
[241] Behrends C, Harper JW. Constructing and decoding unconventional ubiquitin chains. Nat Struct Mol Biol 2011;18:520-8.
[242] Husnjak K, Dikic I. Ubiquitin-binding proteins: decoders of ubiquitin-mediated cellular functions. Annu Rev Biochem 2012;81:291-
322.
[243] Varshavsky A. The ubiquitin system, an immense realm. Annu Rev Biochem 2012;81:167-76.
[244] MacGurn JA, Hsu P, Emr SD. Ubiquitin and membrane protein turnover: from cradle to grave. Annu Rev Biochem 2012;81:231-59.
[245] Reyes-Turcu FE, Ventii KH, Wilkinson KD. Regulation and cellular roles of ubiquitin-specific deubiquitinating enzymes. Annu Rev
Biochem 2009;78:363-97.
[246] Todi SV, Winborn BJ, Scaglione KM, Blount JR, Travis SM, Paulson HL. Ubiquitination directly enhances activity of the
deubiquitinating enzyme ataxin-3. EMBO J 2009;28:372-82.
[247] Cohen P, Tcherpakov M. Will the ubiquitin system furnish as many drug targets as protein kinases? Cell 2010;143:686-93.
[248] Smalle J, Vierstra RD. The ubiquitin 26S proteasome proteolytic pathway. Annu Rev Plant Biol 2004;55:555-90.
[249] Lee DH, Goldberg AL. Proteasome inhibitors cause induction of heat shock proteins and trehalose, which together confer
thermotolerance in Saccharomyces cerevisiae. Mol Cell Biol 1998;18:30-8.
[250] Beers EP, Callis J. Utility of polyhistidine-tagged ubiquitin in the purification of ubiquitin-protein conjugates and as an affinity ligand
for the purification of ubiquitin-specific hydrolases. J Biol Chem 1993;268:21645-9.
[251] Peng J, Schwartz D, Elias JE, Thoreen CC, Cheng D, Marsischky G, Roelofs J, Finley D, Gygi SP. A proteomics approach to
understanding protein ubiquitination. Nat Biotechnol 2003;21:921-6.
[252] Laney JD, Hochstrasser M. Analysis of Protein Ubiquitination. In: Current Protocols in Protein Science. New Jersey: John Wiley &
Sons, Inc.; 2001.
[253] Tagwerker C, Flick K, Cui M, Guerrero C, Dou Y, Auer B, Baldi P, Huang L, Kaiser P. A tandem affinity tag for two-step purification
under fully denaturing conditions: application in ubiquitin profiling and protein complex identification combined with in vivocross-linking.
Mol Cell Proteomics 2006;5:737-48.
[254] Hjerpe R, Rodríguez MS. Efficient approaches for characterizing ubiquitinated proteins. Biochem Soc Trans 2008;36:823-7.
[255] Hjerpe R, Aillet F, Lopitz-Otsoa F, Lang V, England P, Rodriguez MS. Efficient protection and isolation of ubiquitylated proteins
using tandem ubiquitin-binding entities. EMBO Rep 2009;10:1250-8.
[256] Peng J. Evaluation of proteomic strategies for analyzing ubiquitinated proteins. BMB Rep 2008;41:177-83.
[257] Matsumoto M, Hatakeyama S, Oyamada K, Oda Y, Nishimura T, Nakayama KI. Large-scale analysis of the human ubiquitin-related
proteome. Proteomics 2005;5:4145-51.
[258] Xu G, Paige JS, Jaffrey SR. Global analysis of lysine ubiquitination by ubiquitin remnant immunoaffinity profiling. Nat Biotechnol
2010;28:868-73.
[259] Bustos D, Bakalarski CE, Yang Y, Peng J, Kirkpatrick DS. Characterizing ubiquitination sites by peptide based immunoaffinity
enrichment. Mol Cell Proteomics 2012;12:1529-40.
[260] Hahne H, Kuster B. Discovery of O-GlcNAc-6-phosphate modified proteins in large-scale phosphoproteomics data. Mol Cell
Proteomics 2012;11:1063-9.
Primary Sugar Specificity Lectin
Mannose Concanavalin A (ConA), Lens culinaris agglutinin (LCA), Lentil lectin (LCH),
Snowdrop lectin (GNA), Phaseolus vulgaris leucoagglutinin (PHA-L), Pisum
sativum agglutinin (PSA)
Galactose Jacalin (AIL), Coral Tree (ECL), Griffonia simplicifolia lectin I (GSL I), Peanut
Agglutinin (PNA)
N-acetylgalactosamine Dolichos biflorus agglutinin (DBA), Soybean agglutinin (SBA), Sophora
japonica agglutinin (SJA), Vicia villosa lectin (VVL)
N-acetylglucosamine Datura stramonium lectin (DSL), Griffonia simplicifolia lectin II (GSL II),
Lycopersicon esculentum lectin (LEL), Solanum tuberosum lectin (STL), Wheat
germ agglutinin (WGA)
Fucose Aleuria aurantia lectin (AAL), Ulex europaeus agglutinin (UEA),
Sialic acid Elderberry lectin (SNA), Maackia amurensis lectin (MAL)
Tab. 1 - Lectins commonly used in glycoproteomic analyses.
Legends:
Fig. 1 - Frequencies of 22 proteinogenic amino acids and their known PTMs. The percentages indicate
amino acid usage in complete human (gray), model yeast (Saccharomyces cerevisiae, orange) and model plant
(Arabidopsis thaliana, green) proteomes (i.e. the entire sets of proteins expressed by these organisms, according
to information retrieved from http://www.uniprot.org, 10/2012). The bottom part lists numbers of known
naturally occurring PTMs at individual residues (RESID database; release 70.01). This plot illustrates the wide
variability of protein PTMs, but not their amino acid ratios in the proteomes, which also depend on the relative
abundance of the expressed proteins
Fig. 2 - The loss of information in bottom-up proteomic analyses. It is widely accepted that a protein may be
identified based on a single unique (proteotypic) peptide with a well-matched mass spectrum. However, while an
identified fragment may represent a corresponding theoretical protein (A), alternatively it may be part of a
spliced isoform or the result of protein degradation (B), or an indicator of a PTM, especially if the missing
sequence fulfills requirements for MS detection [2] (C).
Fig. 3 - Essential methods for protein multi-dimensional separation.
(A) Schematic workflow of 2-DE analysis. Proteins are separated and focused in the first dimension by IEF in a
strip with a pH gradient according to their pI. The strip with focused proteins is then equilibrated with SDS
buffer to form SDS-protein complexes and usually treated with a reducing agent (e.g. dithiotreitol) and
iodoacetamide to reduce disulfide bridges and alkylate cysteine residues, respectively. The SDS-protein
complexes are then separated in the second dimension by SDS-PAGE. (B) Detection of a protein PTM by 2DE.
A subunit of chloroplastic ATP synthase was identified in three neighboring spots. The apparent molecular
mass of the protein is 55 kDa, in accordance with the theoretical MW, 55.3 kDa. The comparison of apparent pI
values indicates that spot (3) represents unmodified protein, while the proteins in spots (2) and (1) bear
modification(s) that lower their pI to more acidic values. (C) Analysis by 2D-LC. Fractions from a first LC
separation are collected and analyzed by a second LC with contrasting conditions and/or media. The resulting 2D
map is based on UV-VIS profiles and its resolution, especially in the first dimension, is to some extent
proportional to the number of collected fractions. (D) Diagonal separation. Classical separation and detection of
inter- and intra-molecular disulfide bonds, as described in section 2.2.5.
Fig. 4 - Depletion of major abundant proteins can uncover hidden PTMs. Demonstration of immunoaffinity
depletion and proteome equalization in the Arabidopsis thaliana proteome resolved by 2-DE. A sample of 7-dayold
seedlings was divided into three aliquots. One was used for total (acetone/TCA) extraction (A), one was
immuno-depleted of RuBisCO using a Seppro IgY RuBisCO depletion column (B) and one was equalized by
CPLL ProteoMiner (C). Protein spots corresponding to RuBisCO subunits (53 kDa and 20 kDa), show that both
B and C successfully depleted RuBisCO. Demonstration of CPLL effectiveness: Rectangle – no visible effect of
highly abundant protein depletion; Circle – seven protein spots identified as RuBisCO activase; the wide ranges
of pI and MW indicate intensive PTM regulation. Bio-Safe Coomassie stain, 150 Pg of protein, 7 cm IPG strip
(based on [45]; and our unpublished data).
Fig. 5 - Chemical structures of selected PTMs. Protein PTMs described in this review can be separated into 10
distinct but overlapping classes (indicated by colored frames). Individual classes may be further subdivided
(dotted lines). (A) Phosphorylations found in proteins. Ser, Thr and Tyr are amino acids that can be Ophosphorylated;
N-phosphates (amidates) are formed on residues of Lys, His and Arg; acetylphosphorylation is
the least stable protein phosphorylation and has been documented only for Asp (Cys residues are phosphorylated
as intermediates during the enzymatic catalysis, but as this bond is transient they are not usually regarded as
protein PTMs). (B) Protein nitration is documented on Trp and Tyr with 6-nitrotryptophan and 3-nitrotyrosine
being the major nitrated products [50]; Cys residues may be modified by S-nitrosylation. (C) Selected products
of cysteine oxidation. Cysteine is known to undergo a number of oxidative PTMs. The selected PTMs represent
only modifications of the Cys side chain, ignoring protein crosslinking and some less common PTMs that may
occur, e.g. cystenylation (formation of a disulfide bridge between a protein cysteinyl residue and free cysteine)
and thiosulfonate formation (Cys-S-SO2-Cys). (D) Products of methionine oxidation. (E) Protein acetylation
and lipidation. Representatives of S-, O- and N-acylations, an example of a glycosyl phosphatidylinositol
anchor, and a Cys/His/Lys/Arg bound product of lipid oxidation (advanced lipidation end-product), 4-hydroxy-
2-nonenal (which accounts for > 95% of the total unsaturated aldehydes produced during in vitro microsomal
lipid peroxidation [51]); (F) Protein methylation. Representatives of Lys, Arg and His N-methylations; Smethylation
of Cys; C-methylation of Glu; and an example of a methyl ester. (G) Products of protein
carbonylation. The direct oxidation of Arg and Pro (glutamic semialdehyde), Thr (2-amino-3-ketobutyric acid)
and Lys (aminoadipic semialdehyde); examples of indirect protein carbonylation products after protein
interaction with oxidized lipid or carbohydrate. (H) Protein glycosylation. The N-glycan core structure
transferred to the conserved sequence Asn-X-Ser/Thr (where X represents any amino acid except Pro), and
examples of: O-glycosylation, C-mannosylation of Try, an Amadori product of glucose and Lys interaction, and
an AGE-product resulting from further conversion of N6
-fructosamine lysyl and protein crosslinking. (I)
Deamidation and production of an isopeptide bond. (J) Protein ubiquitination. An example of enzymatic
monoubiquitination at Lys, this is essentially a form of protein acylation, but its distinguishing protein-protein
nature demands a separate category.
Fig. 6 - Schematic workflow for protein PTM analysis.