© 2013 Nature America, Inc. All rights reserved.
Large-scale identification of ubiquitination sites by
1950 | VOL.8 NO.10 | 2013 | nature protocols
Protein ubiquitination is an important post-translational modi-
fication (PTM) that is essential for regulating protein turnover
through the ubiquitin-proteasome system. Ubiquitination occurs
most commonly on the ε-amino group of protein lysine residues
through the concerted action of activating (E1), conjugating (E2)
and ligating (E3) enzymes1. As ubiquitin itself has seven lysine
residues, substrates can be monoubiquitinated or polyubiquiti-
nated, resulting in various ubiquitin chain lengths and topologies
that may serve to specifically bind cellular proteins with distinct
Historically, the identification of protein ubiquitination sites
by MS has proven to be challenging because of the low stoi-
chiometry of ubiquitinated proteins, the size of the modifica-
tion itself and the diversity in resulting ubiquitin chain types.
To enhance the identification of low-abundance ubiquitinated
proteins in complex samples, earlier studies have used over-
expressed, affinity-tagged ubiquitin systems to aid in the enrich-
ment and identification of ubiquitinated proteins3,4. These
methods rely on the enrichment of ubiquitinated proteins when
ubiquitin is intact and bound to its substrate. After protein diges-
tion, both formerly ubiquitinated peptides and nonubiquitinated
peptides (from the enriched substrate proteins and the attached
ubiquitin molecules) are present in the sample. The increased
sample complexity resulting from the presence of nonubiquiti-
nated peptides makes detection of specific sites of ubiquitination
challenging. Protein-level enrichment methods have enabled the
detection of several thousand putative, ubiquitinated substrate
proteins, but they lack the necessary enrichment specificity to
enable identification of large numbers of ubiquitination sites.
Robust, large-scale detection of endogenous ubiquitination
sites by MS requires a technique that facilitates the specific enrich-
ment of only the modified lysine-containing peptides of ubiqui-
tinated substrate proteins. To this end, global analysis of protein
ubiquitination has markedly improved with the commercializa-
tion of antibodies specific for the di-glycyl remnant produced on
ubiquitinated lysine residues (K-ε-GG) after trypsin digestion5–8.
Specifically, trypsin digestion of ubiquitinated proteins cleaves off
all but the two C-terminal glycine residues of ubiquitin from the
modified protein. These two C-terminal glycine (GG) residues
remain linked to the ε-amino group of the modified lysine resi-
due in the tryptic peptide derived from digestion of the substrate
protein. The presence of the GG on the side chain of that lysine
prevents cleavage by trypsin at that site, resulting in an internal
modified lysine residue in a formerly ubiquitinated peptide. The
K-ε-GG group is recognized and enriched by using an antibody
specific to K-ε-GG (Fig. 1). It should be noted that modifica-
tion by the ubiquitin-like proteins Nedd8 and ISG15 also results
Namrata D Udeshi, Philipp Mertins, Tanya Svinkina & Steven A Carr
Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA. Correspondence should be addressed to S.A.C. (firstname.lastname@example.org) or
Published online 19 September 2013; doi:10.1038/nprot.2013.120
ubiquitination is essential for the regulation of cellular protein homeostasis. It also has a central role in numerous signaling
events. recent advances in the production and availability of antibodies that recognize the lys--Gly-Gly (K--GG) remnant
produced by trypsin digestion of proteins having ubiquitinated lysine side chains have markedly improved the ability to enrich
and detect endogenous ubiquitination sites by mass spectrometry (Ms). the following protocol describes the steps required to
complete a large-scale ubiquitin experiment for the detection of tens of thousands of distinct ubiquitination sites from cell lines
or tissue samples. specifically, we present detailed, step-by-step instructions for sample preparation, off-line fractionation by
reversed-phase chromatography at pH 10, immobilization of an antibody specific to K--GG to beads by chemical cross-linking,
enrichment of ubiquitinated peptides using these antibodies and proteomic analysis of enriched samples by lc–tandem Ms (Ms/Ms).
relative quantification can be achieved by performing stable isotope labeling by amino acids in cell culture (sIlac) labeling of
cells. after cell or tissue samples have been prepared for lysis, the described protocol can be completed in ~5 d.
Sample enriched with
Reagent as supplied DMP
Figure 1 | Enrichment of K-ε-GG peptides using anti–K-ε-GG antibody.
After the antibody has been chemically cross-linked to a protein A bead
using DMP, peptides are individually enriched for K-ε-GG peptides using the
anti–K-ε-GG antibody, which recognizes the di-glycyl remnant remaining on
modified lysine residues after trypsin digestion.
© 2013 Nature America, Inc. All rights reserved.
nature protocols | VOL.8 NO.10 | 2013 | 1951
in a GG remnant being retained on modified lysine residues,
which makes ubiquitination, Nedd8ylation and ISG15ylation
indistinguishable on the basis of the tryptic remnant. However,
Kim et al.7 have completed experiments in HCT116 cells show-
ing that >94% of K-ε-GG sites are a result of ubiquitination as
opposed to Nedd8ylation or ISG15ylation. The K-ε-GG enrich-
ment method can be performed with SILAC-labeled samples to
enable relative quantification of protein ubiquitination across
differentially perturbed states6,9.
The K-ε-GG–specific antibody has been used in a growing
number of large-scale experiments, including those that studied
the effects of proteasome or deubiquitinase inhibition on the
ubiquitin landscape5–7,10, that globally identified putative cullin-
RING ligase substrates11 and that specifically identified ubiquiti-
nome alterations dependent on the ubiquitin ligase PARKIN12. In
addition, a recent tissue-specific analysis of ubiquitination sites in
mouse tissues revealed both regulation of core signaling pathways
and tissue-specific networks by ubiquitination13. The K-ε-GG–
specific antibody enables the analysis of protein ubiquitination in
a site-specific manner, which has the potential to reveal the degree
of site specificity or site promiscuity of E3 ligases for substrate
Our previous work presented a refined workflow for routine
detection of tens of thousands of distinct ubiquitination sites
from single samples through methodological improvements made
to both the off-line sample fractionation step as well as the K-ε-
GG enrichment step6,14 (Figs. 1 and 2). Specifically, we showed
that fractionation of samples by basic pH reversed-phase (bRP)
chromatography before the enrichment of K-ε-GG peptides
significantly increases the number of identified and quantified
K-ε-GG sites in SILAC-labeled samples6,9,14 (Fig. 2). Our previous
work also demonstrated that chemical cross-linking of the K-ε-
GG–specific antibody to a solid support considerably reduces the
contamination of antibody fragments and non–K-ε-GG peptides
present in the final enriched samples (Figs. 1 and 3)6. The present
protocol provides complete step-by-step instructions for imple-
menting the procedural workflow for large-scale enrichment and
identification of ubiquitinated peptides using MS.
Urea lysis of cells or tissue
LysC and trypsin
digestion of proteins
bRP fractionation of peptides
1 111098765432 12
Fractions are recombined by a
noncontiguous pooling scheme
Figure 2 | Workflow for preparing samples for K-ε-GG enrichment. Samples
are lysed in urea buffer, digested with LysC and trypsin and fractionated
off-line using bRP chromatography. Fractions are pooled to eight final
fractions for enrichment of K-ε-GG peptides using the K-ε-GG–specific
antibody. Fxn, fraction.
Wash 3× after
Figure 3 | Example of an SDS-PAGE gel used to evaluate the efficiency of
antibody cross-linking to protein A beads. TFA eluates from ~30 µg of
pre– and post–cross-linked anti–K-ε-GG antibody were analyzed by
SDS-PAGE. After cross-linking, more than a tenfold decrease in staining
density for heavy and light chains of the antibody is required for optimal
results. Reproduced from Udeshi et al.6.
Cells or tissue samples (see Reagent Setup for details)
SILAC amino acids (Cambridge Isotope Laboratories)
Tris(hydroxymethyl)aminomethane hydrochloride (Tris-HCl)
Sodium chloride (NaCl)
EDTA (Sigma-Aldrich, cat. no. E7889)
Aprotinin (Sigma-Aldrich, cat. no. A610)•
Leupeptin (Roche, cat. no. 11017101001)
PMSF (Sigma-Aldrich, cat. no. 93482)
2,6 diaminopyradine-3,5-bis(thiocynate) (PR-619; Sigma-Aldrich,
cat. no. SML0430)
Chloroacetamide (CAM; Sigma-Aldrich, cat. no. C0267)
Bicinchoninic acid (BCA) protein assay kit (Pierce, cat. no. 23225)
DTT (Pierce, cat. no. 20291)
Iodoacetamide (IAM; Sigma-Aldrich, cat. no. A3221)
© 2013 Nature America, Inc. All rights reserved.
1952 | VOL.8 NO.10 | 2013 | nature protocols
LysC (Wako Chemicals, cat. no. 129-02541)
Sequencing grade modified trypsin, 1 mg per vial (Promega, cat. no. V511X)
Formic acid (FA; Sigma-Aldrich, 56302)
Trifluoroacetic acid (TFA; Sigma-Aldrich, cat. no. 91707)
Acetonitrile (J.T. Baker, cat. no. 9829-03)
PTMScan ubiquitin remnant motif (K-ε-GG) kit (Cell Signaling
Technology, cat. no. 5562)
Ammonium hydroxide solution, 28% (wt/vol) (NH4OH; Sigma-Aldrich,
cat. no. 338818)
4-Morpholinepropanesulfonic acid (MOPS; Sigma-Aldrich, cat. no. M5162)
Sodium phosphate dibasic (Sigma-Aldrich, cat. no. S9763)
Sodium tetraborate decahydrate (Sigma-Aldrich, cat. no. S9640)
Ethanolamine (Sigma-Aldrich, cat. no. 398136)
Dimethyl pimelimidate dihydrochloride (DMP; Sigma-Aldrich,
cat. no. D8388)
PBS (Invitrogen, cat. no. 10010-023)
Sodium azide, 0.02% (wt/vol)
Lysis buffer (Step 1; see Reagent Setup)
Solid-phase extraction (SPE) desalting solvents (Step 12; see Reagent Setup)
bRP solvents (Step 20; see Reagent Setup)
Antibody cross-linking buffers (Step 24; see Reagent Setup)
Immunoaffinity purification (IAP) solutions (Step 29; see Reagent Setup)
StageTip solvents (Step 40; see Reagent Setup)
UPLC-MS/MS solvents (Step 51; see Reagent Setup)
Solid-phase extraction cartridge for sample desalting, 500 mg sorbent per
cartridge, 37–55 µm particle size (Sep-Pak tC18 6 cc Vac cartridge; Waters,
cat. no. WAT036790)
StageTip adapters for microcentrifuge tubes (Glygen, cat. no. ADP000.24)
Off-line HPLC system for bRP separation: Agilent 1100 series pump
equipped with a degasser, autosampler and fraction collector
Column for bRP separation. We use a Zorbax 300 Extend-C18 column
(9.4 × 250 mm, 300 Å, 5 µm; Agilent) because the silica-based packing
material is stable up to pH 11.5. If an alternative reversed-phase column is
used for the bRP separation, ensure that the silica-based packing material is
stable at high pH. Column should be periodically tested using a simple pep-
tide mixture of the user’s choice to monitor for unwanted peak broadening
or shifts in retention time caused by column deterioration over time.
Deep-well plates, 96 wells: 2-ml round-bottom wells (Whatman/GE
Healthcare, cat. no. 7701-5200)
Solid-phase extraction disk for StageTips: Empore C18 extraction disk
(3M, cat. no. 98060402181 or cat. no. 98060402173)
Nanospray column for online ultraperformance liquid chromatography
(UPLC)-MS/MS analysis: Self-pack PicoFrit column, 360 µm outer
diameter (o.d.) × 75 µm inner diameter (i.d.), 10 µm i.d. tip, 50 cm length
(New Objective, cat. no. PF360-75-10-N-5); ReproSil-Pur 120 Å, C18-AQ,
1.9 µm (Dr. Maisch)
LC system for online LC-MS analysis: Proxeon Easy-nLC 1000 (Thermo
Fisher Scientific). We use a Proxeon Easy-nLC 1000 and operate under
UPLC conditions; however, any LC instrument that can deliver nanoflow
rates and can operate up to a pressure of 1,000 bar can be used for peptide
separation. Lower pressure, non-UPLC conditions, can also be used for
separations, but we have observed that the number of peptides identified in
a given experiment decreases by at least 1.5-fold
Nanospray column heater, 20 cm (Phoenix S&T, cat. no. PST-CH-20U)
Column heater controller (Phoenix S&T, cat. no. PST-CHC)
Mass spectrometer: Q Exactive or Orbitrap Velos system (Thermo Fisher
Scientific) crItIcal We use a Q Exactive mass spectrometer because we
find it yields higher numbers of identified K-ε-GG peptides relative to the
Orbitrap Velos or Orbitrap Elite systems.
Stock solutions for lysis buffer Stock solutions are: 1 M Tris HCl (pH 8.0),
1 M NaCl, 500 mM EDTA, 1 mg ml − 1 aprotinin, 2 mg ml − 1 Leupeptin,
100 mM PMSF in ethanol, 500 mM chloroacetamide or iodoacetamide,
and 40 mM PR-619 in DMSO
Urea lysis buffer Prepare urea lysis buffer to contain 8 M urea, 50 mM
Tris-HCl (pH 8.0), 150 mM NaCl, 1 mM EDTA, 2 µg ml − 1 aprotinin,
10 µg ml − 1 leupeptin, 50 µM PR-619, 1 mM chloroacetamide or
iodoacetamide and 1 mM PMSF. crItIcal The urea lysis buffer should
always be freshly prepared to prevent carbamylation of proteins. PMSF
should be added to the lysis buffer immediately before use because the
half-life of this inhibitor has been shown to be <35 min in aqueous buffers at
pH 8 (ref. 15).
SPE desalting solvents Wash solvent 1 is 100% acetonitrile, wash solvent 2
is 0.1% (vol/vol) TFA, and wash solvent 3 is 0.1% (vol/vol) FA. Elution
solvent is 50% acetonitrile/0.1% (vol/vol) FA. crItIcal SPE desalting
solvents can be prepared in advance and stored at room temperature
(RT, 20–25 °C); they remain stable for at least 2 weeks.
bRP solvents Stock solution is 200 mM ammonium formate (NH4HCO2);
to prepare 1 liter of NH4HCO2, combine 25 ml of 28% (wt/vol) NH4OH
solution, 45 ml of 10% (vol/vol) FA and 930 ml of H2O. bRP solvent A is
5 mM ammonium formate (pH 10.0)/2% (vol/vol) acetonitrile, and bRP
solvent B is 5 mM ammonium formate (pH 10.0)/90% (vol/vol) acetonitrile.
crItIcal bRP desalting solvents remain stable at RT for several days.
Antibody cross-linking wash buffer Wash buffer is 100 mM sodium borate,
pH 9.0. crItIcal This buffer is stable for at least 2 weeks at RT.
Antibody cross-linking buffer Cross-linking buffer is 20 mM DMP in
100 mM sodium borate (pH 9.0). crItIcal DMP solution should be
freshly prepared immediately before each use.
Antibody blocking buffer Blocking buffer is 200 mM ethanolamine
(pH 8.0; HCl or FA can be used to adjust the pH to 8.0). crItIcal This
buffer is stable for at least 2 weeks at RT.
Stock solutions for IAP Stock solutions for IAP are: 500 mM MOPS that is
brought to pH 7.2 with NaOH, 500 mM sodium phosphate dibasic and 1 M
NaCl. crItIcal IAP solutions can be prepared in advance and stored at
RT, and they remain stable for several weeks.
IAP buffer IAP buffer contains 50 mM MOPS (pH 7.2), 10 mM sodium
phosphate and 50 mM NaCl.
IAP elution solution Elution solution is 0.15 % (vol/vol) TFA.
StageTip solvents Wash solvent 1 is 100% methanol.Wash solvent 2 is
0.1% (vol/vol) FA. Elution solvent is 50% acetonitrile/0.1% (vol/vol) FA.
crItIcal StageTip solvents can be made in advance and stored at RT.
They remain stable for several weeks.
UPLC-MS/MS solvents Solvent A is 0.1% FA/3% acetonitrile (vol/vol).
Solvent B is 0.1% FA/90% (vol/vol) acetonitrile. crItIcal UPLC solvents
remain stable for at least 2 weeks at RT.
Cell or tissue preparation Proteins derived from either cell or tissue samples
can be used in the protocol. For cell culture systems, cells should be cultured
in medium and growth conditions appropriate for the given cell type. Typical
protein starting amounts used in this workflow are ≥5 mg. Cells should be
expanded to an appropriate cell number (typically 50–100 × 106 cells per
sample) to achieve a suitable protein yield. Cells should be washed once or
twice with PBS to remove residual cell culture medium. Proteins are extracted
from the resulting cell pellet.
To increase the stoichiometry of ubiquitinated proteins, cells can be treat-
ed before lysis with a proteasome inhibitor such as MG-132, bortezomib or
epoxomicin5,7,16. The type of proteasome inhibitor used as well as the treat-
ment time and concentration will need to be individually optimized by the
user for each experiment. It should be noted that prolonged treatment (>1 h)
with proteasome inhibitors can make it difficult to differentiate ubiquitina-
tion sites that occur because of stress induced by inhibiting the proteasome
from those sites involved in physiological cellular functions17.
For quantitative analyses, SILAC can be used to differentially label up to
three cellular states in a single experiment following previously described
protocols5,9,18. Note that the use of amine-reactive isobaric tags such as
isobaric tags for relative and absolute quantification (iTRAQ) or tandem
mass tags (TMT) reagents for quantification is not compatible with this
protocol if peptides are labeled before K-ε-GG enrichment. iTRAQ or TMT
reagents will label K-ε-GG sites, and we have found that the K-ε-GG–specific
antibody will not recognize these labeled species. For tissue samples,
we suggest starting with enough tissue to yield at least 5–10 mg of protein.
StageTip setup C18 StageTips are prepared in-house essentially as described
by Rappsilber et al.19, by using Empore C18 extraction disks. We pack two
plugs of C18 material into each StageTip.
Nanoflow C18 column The nanoflow C18 column has dimensions of
360 µm o.d. × 75 µm i.d., a 10-µm-i.d. tip, and it contains 24-cm
ReproSil-Pur 120-Å, C18-AQ 1.9-µm resin (we self-pack our nanoflow
columns in-house using a pressure bomb at 1,000 p.s.i.).
© 2013 Nature America, Inc. All rights reserved.
nature protocols | VOL.8 NO.10 | 2013 | 1953
prepare lysates ● tIMInG 60 min
1| Chill urea lysis buffer to 4 °C and add lysis buffer to the cell pellet (see Reagent Setup) to achieve a protein
concentration of ≤5 mg ml − 1.
2| Clear the lysate by centrifugation at 20,000g for 10 min at 4 °C. Remove the supernatant after centrifugation and
transfer it to a fresh tube.
3| Use a BCA assay to determine the protein concentration. If you are not working with SILAC-labeled samples, continue to
4| (Optional) For SILAC-labeled samples, a single solution should be prepared by mixing an equal amount of protein
(by weight) from each SILAC-encoded sample.
reduction and alkylation ● tIMInG 2 h
5| Reduce disulfide bonds by adding DTT to the sample at a final concentration of 5 mM. Incubate the sample at RT for
6| Carbamidomethylate cysteine residues by adding IAM at a final concentration of 10 mM. Incubate the sample for 30 min
at RT in the dark.
enzymatic digestion ● tIMInG 15–18 h
7| Before digestion, dilute the sample 1:4 with 50 mM Tris HCl (pH 8.0) to reduce the urea concentration to 2 M.
8| To increase the number of fully cleaved peptides generated after enzymatic digestion, treat samples with endoproteinase
LysC before trypsin digestion20. If a more-rapid digestion procedure is desired, move to Step 9 and complete only trypsin
digestion. For LysC digestion, use an enzyme-to-substrate ratio of 1:50 (wt/wt) and digest for 2 h at RT while shaking.
crItIcal step Check the pH of the reaction using a pH indicator strip at the beginning and end of digestion to ensure
that the pH is in the optimal range for both LysC and trypsin (pH 8–8.5).
9| For trypsin digestion, use an enzyme-to-substrate ratio of 1:50 (wt/wt). Complete the trypsin digestion overnight at RT
10| Quench the digestion reaction by adding TFA to the sample to a final amount of 0.1% (vol/vol) and vortex. Make sure
that the pH of the solution is ≤3 using a pH indicator strip before desalting.
11| Centrifuge the peptide solution for 5–10 min at 3,000g to remove precipitate (contains mainly urea). Transfer the
supernatant into a new tube.
peptide desalting by spe ● tIMInG ≥5 h
12| Use a 500-mg tC18 SepPak cartridge with a vacuum manifold to desalt peptide samples.
crItIcal step As has been previously described in the protocol by Villén and Gygi21, the choice of cartridge size should
be based on sample input amount. Capacities for SepPak cartridges are 3–5% (wt/wt) of the sorbent weight21. We have
observed that letting the cartridges run dry during washing or loading can increase sample losses during desalting from 5%
13| Condition the cartridge with 5 ml of wash solvent 1 (see Reagent Setup).
14| Condition the cartridge with 5 ml of elution solvent.
15| Equilibrate the cartridge with four washes of 5 ml of wash solvent 2.
© 2013 Nature America, Inc. All rights reserved.
1954 | VOL.8 NO.10 | 2013 | nature protocols
16| Load the sample onto the cartridge in 5-ml aliquots.
17| Wash the cartridge four times with 5 ml of wash solvent 2.
18| Elute peptides into 15-ml conical tubes with two washes of 3 ml of elution solvent.
19| Dry the eluted peptides completely by lyophilization (preferred) or by vacuum centrifugation. We find that lyophilization
tends to be more consistent at generating a dry, fluffy peptide sample that is easily reconstituted. However, we find that
lyophilizing peptide samples can be two or three times slower than vacuum centrifugation.
pause poInt Dried samples are stable and can be stored at −20 or −80 °C for several weeks.
brp fractionation ● tIMInG ≥ 3 h
crItIcal See Figure 2 for an overview of this section.
20| Reconstitute peptide samples in bRP solvent A (see Reagents) to achieve a concentration <10 mg ml−1, and then
centrifuge the samples at 20,000g to remove any material that did not go into solution.
21| For separation of peptides, use the gradient and flow rate settings outlined in the table below. Across the entirety of the
bRP separation, a total of 96 2-ml fractions are collected at a flow rate of 3 ml min–1. Before separating the peptide sample,
a blank gradient should be run to wash and equilibrate the RP column. During the blank gradient wash, fractions do not need
to be collected.
time interval (min)lc gradient (% B) lc flow rate (ml per minute)
crItIcal step We use a Zorbax 300 Extend-C18 column (9.4 × 250 mm, 300 Å, 5 µm; Agilent) for separating 6–15 mg
of total peptide. However, you must adjust the scale of the RP C18 column if you are using peptide amounts outside of this
range. Do not store Zorbax 300 Extend columns at high pH for extended periods of time (multiple days) when the column is
not in use, as this can permanently decrease column performance.
22| Inject the samples onto the bRP column at a flow rate of 3 ml min–1 and separate them as described above. Collect
fractions every 0.66 min in a 2-ml 96-well plate (Fig. 2). The UV absorbance measurement can be monitored at 214 nm.
23| After separation, pool the bRP fractions in a serpentine, noncontiguous manner (as described below and shown in
Fig. 2) to generate eight final fractions6,14,22,23. Specifically, to generate eight total fractions, every eighth fraction is
combined (final fraction 1 = 1,9,17,25,33,41,49,57,65; final fraction 2 = 2,10,18,26,34,42,50,58,66; …). We find that