Specific and efficient N-propionylation of histones with propionic acid N-hydroxysuccinimide ester for histone marks characterization by LC-MS

Article abstract of DOI:10.1021/ac303171h

Histones participate in epigenetic regulation via a variety of dynamic posttranslational modifications (PTMs) on them. Mass spectrometry (MS) has become a powerful tool to investigate histone PTMs. With the bottom-up mass spectrometry approach, chemical derivatization of histones with propionic anhydride or deuterated acetic anhydride followed by trypsin digestion was widely used to block the hydrophilic lysine residues and generate compatible peptides for LC-MS analysis. However, certain severe side reactions (such as acylation on tyrosine or serine) caused by acid anhydrides will lead to a number of analytical issues such as reducing results accuracy and impairing the reproducibility and sensitivity of MS analysis. As an alternative approach, we report a novel derivatization method that utilizes N-hydroxysuccinimide ester to specifically and efficiently derivatize both free and monomethylated amine groups in histones. A competitive inhibiting strategy was implemented in our method to effectively prevent the side reactions. We demonstrated that our method can achieve excellent specificity and efficiency for histones derivatization in a reproducible manner. Using this derivatization method, we succeeded to quantitatively profile the histone PTMs in KMS11 cell line with selective knock out of translocated NSD2 allele (TKO) and the original parental KMS11 cell lines (PAR) (NSD2, a histone methyltransferase that catalyzes the histone H3 K36 methylation), which revealed a significant crosstalk between H3 protein K27 methylation and adjacent K36 methylation.

Full text of DOI:10.1021/ac303171h

Article
pubs.acs.org/ac
Specific and Efficient N‑Propionylation of Histones with Propionic
Acid N‑Hydroxysuccinimide Ester for Histone Marks Characterization
by LC-MS
†
†
†
†
†
‡
§
Rijing Liao, Haiping Wu, Haibing Deng, Yanyan Yu, Min Hu, Huili Zhai, Pengyuan Yang,
Shaolian Zhou,* and Wei Yi*
,†
,†
†
China Novartis Institutes for BioMedical Research Co. Ltd., Building 8, lane 898 Halei road, Shanghai 201203, China
‡
Novartis Institutes for BioMedical Research, Inc., 250 Massachusetts Avenue, Cambridge, Massachusetts 02139, United States
Department of Chemistry, Fudan University, 220 Handan Road, Shanghai 200433, China
§
*
S Supporting Information
ABSTRACT: Histones participate in epigenetic regulation via
a variety of dynamic posttranslational modifications (PTMs)
on them. Mass spectrometry (MS) has become a powerful tool
to investigate histone PTMs. With the bottom-up mass
spectrometry approach, chemical derivatization of histones
with propionic anhydride or deuterated acetic anhydride
followed by trypsin digestion was widely used to block the
hydrophilic lysine residues and generate compatible peptides
for LC-MS analysis. However, certain severe side reactions
(
such as acylation on tyrosine or serine) caused by acid
anhydrides will lead to a number of analytical issues such as reducing results accuracy and impairing the reproducibility and
sensitivity of MS analysis. As an alternative approach, we report a novel derivatization method that utilizes N-hydroxysuccinimide
ester to specifically and efficiently derivatize both free and monomethylated amine groups in histones. A competitive inhibiting
strategy was implemented in our method to effectively prevent the side reactions. We demonstrated that our method can achieve
excellent specificity and efficiency for histones derivatization in a reproducible manner. Using this derivatization method, we
succeeded to quantitatively profile the histone PTMs in KMS11 cell line with selective knock out of translocated NSD2 allele
(
TKO) and the original parental KMS11 cell lines (PAR) (NSD2, a histone methyltransferase that catalyzes the histone H3 K36
methylation), which revealed a significant crosstalk between H3 protein K27 methylation and adjacent K36 methylation.
ukaryotic DNA is highly compacted by cationic histones to
form chromatin. Besides the function of DNA packaging,
numerous small hydrophilic peptides with poor chromato-
graphic retention. In addition, cleavage at lysine residue is
prevented by their methylation/acetylation, therefore generat-
ing tryptic peptides with different lengths from heterogeneously
modified histones. For example, tryptic digestion on a histone
H3 protein with no modification on K27 would produce a
different tryptic peptide compared to the same protein bearing
K27 trimethylation. The use of Arg-C enzyme that only cleaves
at arginine residues can partially circumvent this problem;
however, the lower efficiency and specificity of Arg-C would
E
histones participate in the regulation of chromatin processes
such as gene transcription, DNA repair, and apoptosis via a
myriad of dynamic post-translational modifications (PTMs) on
1
multiple specific sites of histones. The PTMs are mostly
located on the N-terminal tail of histones, and the prevalent
PTMs include lysine acetylation and methylation, arginine
methylation, and serine and threonine phosphorylation.
Reliable characterization of histone PTMs is the prerequisite
to investigate their biological functions.
2
result in additional concerns for MS analysis. An alternative
method is the chemical derivatization of lysines on histones
In the past few decades, mass spectrometry (MS) has
8
9−11
2
followed with trypsin digestion. Propionic anhydride
deuterated acetic anhydride
or
emerged as a powerful tool to characterize histone PTMs, and
1
2,13
were commonly used deriva-
discovered many novel PTMs that were previously not detected
3
tization reagents that add acyl groups onto the free or
monomethylated lysines in histones. Acylation of lysines
residues can protect them from tryptic digestion and decrease
the charges of peptides, thereby increasing their retention on
by any other means. Among the three major MS-based
approaches (top-down, middle-down, and bottom-up MS
4
−7
approaches)
for histone PTMs characterization, the
bottom-up MS approach is primarily used for both character-
ization and quantification of histone PTMs. Nevertheless, the
high abundance of lysine and arginine residues on histones is
quite a challenge for the bottom-up MS approach, as
proteolysis with standard proteases such as trypsin will generate
Received: October 31, 2012
Accepted: January 22, 2013
Published: January 22, 2013
©
2013 American Chemical Society
2253
dx.doi.org/10.1021/ac303171h | Anal. Chem. 2013, 85, 2253−2259

Analytical Chemistry
Article
9
Figure 1. MALDI-TOF spectra of the peptide KSAPATGGVKKPHR derivatized at room temperature with (a) propionic anhydride, (b) 50 mM
NHS-propionate in 50 mM TEAB (pH 9.0), and (c) 50 mM NHS-propionate in 50 mM NH HCO (pH 8.0). The methyl-esterified peptides were
4
3
labeled as Me × 1; the quadruply and quintuply propionylated peptides were labeled as pro × 4 and pro × 5, respectively.
reversed-phase columns and enhancing their ionization
efficiency. However, there are still some technical drawbacks
with anhydride derivatization methods. Drogaris et al. found
that double rounds of propionic anhydride derivatization would
significantly decrease the ion abundance of the same peptides
histones which are less-reactive than primary amines in
acylation reaction.
Herein, we report a novel method for the highly specific and
efficient derivatization of the amine groups on histones with
propionic acid NHS ester (NHS-propionate). NHS-propionate
can transfer propionyl groups to both primary and secondary
amines and form corresponding amides (Scheme 1a).
10
compared to the single round of derivatization. This result
reminded us to carefully look into the negative impact of
propionic anhydride derivatization on sample analysis. The
reduction of ion abundance was considered to be likely caused
by the side reactions during the derivatization as observed in
our result (Figure 1a). Since acid anhydrides are very strong
acylation reagents, they can react with hydroxyl groups and
form side products in derivatization. Given the high abundance
of hydroxyl-containing residues (tyrosines, serines, and
threonines) in histones, the nonspecific O-acylation can be
quite significant, eventually leading to inaccurate profiling for
histone PTMs. Some methodologies such as hydroxylamine
Scheme 1. Reactions Involved in Histones Derivatization
a
with Propionate-NHS
1
4
15
treatment and boiling water bath have been employed in
proteomic studies for selective hydrolysis of O-acylation and
restoration of the free hydroxyl groups. Unfortunately, these
methods are not adaptable for histone proteins as the
endogenous phosphorylations of histones would be simulta-
neously removed together with the O-acylation. In this study,
we explored other amine-specific reagents for the chemical
derivatization of the lysine residues on histone with a specific
aim on improving reaction specificity. Several chemicals such as
acid N-hydroxysuccinimide ester (NHS) and acid pentafluor-
ophenyl ester that can rapidly react with amines under mild
condition appeared to be good alternatives for derivatization
a
(a) Propionylation of amines, (b) propionylation of hydroxyl groups,
and (c) ammonia coming from NH
with hydroxyl groups in propionylation.
HCO hydrolysis and competing
4
3
1
6,17
reagents.
Some of them have been utilized in the amino-
Compared to the propionic anhydride, NHS reagents have
been reported as the more amine-specific reagents,
acid derivatization to increase the retention of amino acids in
20,21
which
1
8,19
reversed-phase chromatography.
However, the direct
would contribute to the higher specificity of histone
derivatization. Furthermore, a strategy using ammonia to
competitively inhibit the undesired O-propionylation was
successfully developed in this work. With this strategy, the
nonspecific reactions between hydroxyl groups on peptide/
protein residues and NHS reagents were significantly reduced
to achieve almost complete conversion for histone peptides.
This method was evaluated for reproducibility, specificity, and
efficiency for histones derivatization. The application of this
method was also demonstrated in the quantification of histone
transfer of the derivatization method from small molecules to
intact proteins would be difficult because the specificity and
efficiency of the chemical reactions may be different for the
more complicated and fragile biological substrates. In the case
of cellular sample analysis, histones with heterogeneous
modifications have much more complicated behavior in
chemical reactions than structurally simple small molecules;
furthermore, it would be difficult to complete the derivatization
for secondary amines in the monomethylated lysines of
2
254
dx.doi.org/10.1021/ac303171h | Anal. Chem. 2013, 85, 2253−2259

Analytical Chemistry
Article
PTMs from the KMS11 cells with selective knock out of
translocated NSD2 allele (TKO) and the original parental
KMS11 cells (PAR).
remove any volatile remnants and stored at −80 °C. To
propionylate the free amines newly generated from tryptic
digestion, the second round of derivatization was carried out
with the same protocol used for the first round.
MALDI-TOF. MALDI-TOF mass spectrometric experiments
were performed using a Bruker ultrafleXtreme instrument
MATERIALS AND METHODS
■
Chemicals were purchased from Sigma unless stated elsewhere.
The double deionized water was produced by Milli-Q gradient
A10 system (Millipore, Bedford, MA). The standard peptides
were purchased from GL Biochem Ltd. (Shanghai, China). The
recombinant histone H3 was purchased from Viva Biotech Ltd.
(
Bruker, Bremen, Germany) in the reflectron mode. For regular
MALDI-TOF experiments, all spectra were externally calibrated
using a peptide standard mixture provided by Bruker. A 1.0 μL
aliquot of the peptide solution was mixed with 1.0 μL of α-
cyano-4-hydroxycinnamic acid matrix (5 mg/mL in 1:1
acetonitrile/water and 0.1% formic acid; Sigma). Data from a
set of 100−200 laser shots were accumulated to give an
acceptable spectrum.
(
Shanghai, China).
Synthesis of NHS-Propionate. The scheme for synthesis
of NHS-propionate was shown in Supporting Information
Scheme S1. The procedure was an adaption of a method
22
LC-MS/MS. Peptide mixtures were analyzed by LC-MS/MS
using an LC20A instrument (Shimadzu, Kyoto, Japan)
interfaced to a Q Exactive instrument (ThermoFisher Scientific,
Bremen, Germany). The chromatographic separation was
carried out in a C18 reversed-phase column (150 mm × 4.6
mm, 5 μm, Agilent) at the flow rate of 0.3 mL/min (0−3 min)
and 0.4 mL/min (3−60 min). The chromatographic gradient
was 3% B for 3 min, 3−45% B from 3 to 42 min, 45−95% B
from 42 to 45 min, 95% B from 45 to 48 min, 95−3% B from
described by Kleinmaier et al. Propionyl chloride (3.24 mL,
6.5 mmol) and triethylamine (5.10 mL, 36.5 mmol) were
3
dissolved in dichloromethane (100 mL). Then, N-hydrox-
ysuccinimide (4.0 g, 34.8 mmol) was added under an ice-bath.
The reaction mixture was stirred at room temperature
overnight. Ether was added, and the resultant white precipitate
was removed by filtration. The solvent was concentrated, and
the residue was purified by silica gel chromatography (eluted
with hexane/ethyl acetate (1:1)) to give white solid (3.0 g, yield
1
4
0
8 to 50, and 3% B from 50 to 60 min (solvent A: 99.9% H O,
.1% formic acid; solvent B: 99.9% MeCN, 0.1% formic acid).
5
0%). H NMR (400 MHz, CDCl , δ): 1.27 (t, 3H, J = 7.5 Hz,
2
3
CH ), 2.63 (q, 2H, J = 7.5 Hz, CH ), 2.82 (s, 4H, succinimide
3
2
The electrospray ion source was used with a spray voltage of
CH2).
3
3
.5 kV; sheath, auxiliary gases, and sweep flow rate were set at
0, 20, and 0 arbitrary units, respectively; capillary temperature
Mammalian Cell Culture and Histone Extraction.
KMS11-parental and TKO (translocation knockout) cell lines
were obtained from Horizon Discovery Ltd. (Cambridge, UK)
and were cultured in RPMI1640 (Gibco) supplemented with
was set to 360 °C. The mass spectrometer was operated in a
data-dependent mode to automatically switch between MS and
MS/MS acquisition. In the Q Exactive instrument, full-scan
1
2.5% fetal bovine serum (FBS, Gibco) and 1% penicillin/
(
m/z 200−2000) MS spectra were acquired in the Obitrap
streptomycin (Gibco). Nuclei were isolated, and histone
proteins were fractionated as described by Shechter et al.
23
analyzer with the resolution of 70 000, and then, eight most
intense precursor ions were isolated by the quadrupole for
fragmentation using HCD (higher energy collisional dissocia-
Briefly, core histones were acid extracted from nuclei with 0.4
M H SO4 and then precipitated with trichloroacetic acid
2
2
(
TCA), followed by washing with ice-cold acetone twice. The
tion) with a normalized collision energy of 27%. The MS
resulting pellets were dissolved in deionized water. The core
spectra were acquired in the Orbitrap analyzer with the
resolution of 17 500.
histones were further fractionated on a C column (150 mm ×
8
4
.6 mm, Agilent), using an Agilent series 1200 system
Data Analysis. The raw data from Q exactive were searched
using Proteome Discoverer 1.3 software (ThermoFisher
Scientific) against a custom-made database containing human
histone sequences retrieved from the National Center for
Biotechnology Information (NCBI) database. Enzyme specific-
ity was set to trypsin, allowing for up to 3 missed cleavages.
Mass tolerance for the precursor ions was set to 10 ppm and for
fragment ions was set to 0.05 Da. The propionylation (+56.026
Da) was set to a static modification on the N-terminus of
peptides. The variable modifications included mono- and
dimethylation on lysine and arginine residues (the mono-
methylation of lysine was set as the sum (+70.042 Da) of
propionylation (+56.026 Da) and methylation (+14.016 Da),
propionylation, trimethylation, and acetylation on lysines;
phosphorylation on serine, threonine, and tyrosine. All MS/
MS spectra from histone peptides were also manually inspected
for further fragment assignment. Using the Qual Browser
(ThermoFisher Scientific), the extracted ion chromatograms
(XIC) were constructed for each precursor m/z value with a
mass tolerance of 10 ppm and mass precision up to four
decimal places. For relative quantification of peptides KSAP-
ATGGVKKPHR (K27-R40) containing modifications on K27
or K36, the peak areas of XIC were normalized against an
endogenous standard: a 3-aa histone H3 peptide LVR (L70-
R72) for which no PTMs were observed.
(
Waldbronn, Germany) with a flow rate of 0.8 mL/min and
a gradient of 0% B for 5 min, 0−35% B from 5 to 10 min, 35%
B from 10 to 18 min, 35−65% B from 18 to 70 min, 65−100%
B from 70 to 72, 100% B from 72 to 77 min, 100−0% B from
7
7 to 80 min, and 0% B from 80 to 90 min (A = 5% acetonitrile
in 0.1% TFA and B = 90% acetonitrile in 0.1% TFA). Fractions
were collected with 1 min time intervals, pooled, and dried to
completion in a SpeedVac and stored at −80 °C.
Derivatization of Histones or Peptides and Trypsin
Digestion. The propionic anhydride derivatization wa9s
performed according to the previously published method.
For the derivatization with NHS-propionate, the dried histone
or peptide samples were reconstituted in 50 mM NH HCO3
4
buffer (without any pH adjustment); the final concentration for
histone or peptides was 0.1 μg/μL. Immediately before use,
NHS-propionate was dissolved in acetonitrile at the concen-
tration of 200 mM. Equal volume of NHS-propionate solution
was added to the sample and mixed with the pipet. The
reaction was incubated at 50 °C for 30 min and thoroughly
evaporated by a SpeedVac to remove any volatile remnants.
Propionylated histones were then reconstituted in 50 mM
NH HCO3 buffer and digested with trypsin (Promega,
Madison, WI) at a histones/enzyme ratio of 20: 1 overnight
at 37 °C. The samples were thoroughly dried by a SpeedVac to
4
2
255
dx.doi.org/10.1021/ac303171h | Anal. Chem. 2013, 85, 2253−2259

Analytical Chemistry
Article
Figure 2. MALDI-TOF spectra of the peptide K(Me)SAPATGGVK(Me)KPHR derivatized at (a) room temperature with 50 mM NHS-propionate,
0 mM NH HCO ; (b) 50 °C with 50 mM NHS-propionate, 50 mM NH HCO ; (c) 50 °C with 100 mM NHS-propionate, 50 mM NH HCO ;
5
4
3
4
3
4
3
(
d) 50 °C with 100 mM NHS-propionate, 25 mM NH HCO ; and (e) 50 °C with 100 mM NHS-propionate, 10 mM NH HCO . The doubly,
4
3
4
3
triply, quadruply, and quintuply propionylated peptides were labeled as pro × 2, pro × 3, pro × 4, and pro × 5, respectively.
RESULTS AND DISCUSSION
and form corresponding amides or esters, respectively (Scheme
a,b). However, hydroxyl groups are weaker nucleophiles than
■
1
Reducing Side Products Using Ammonia as the
amines and thereby react with NHS-propionate at a lower rate.
Given the fact that amines and hydroxyl groups have differential
reactivity over NHS-propionate, we took an approach by
adding some ammonium hydroxide to the reaction buffer. The
ammonia can serve as a competition agent that competes with
hydroxyl groups in reacting with NHS-propionate in
derivatization; thus, the O-propionylation can be inhibited.
The results of our experiments confirmed the advantage of
using ammonia to eliminate O-propionylation (Supporting
Information Figure S3). In reality, it is not convenient to use
concentrated ammonium hydroxide in the experimental steps
due to its high volatility. We therefore attempted to replace
ammonium hydroxide with ammonium bicarbonate that can
slightly hydrolyze in water and release ammonia molecules
Competitive Inhibitor. To investigate the specificity of
NHS-propionate toward amines and hydroxyl groups, a
synthetic histone H3 peptide KSAPATGGVKKPHR (K27-
R40) was derivatized with 50 mM NHS-propionate in a 50 mM
triethylammonium bicarbonate (TEAB) buffer (pH 9.0).
Triethylammonium is a tertiary amine that will not react with
NHS ester, and its volatility facilitates the following MS
analysis. The derivatization products were analyzed with
MALDI-TOF (Figure 1b). The major product quadruply
propionylated (pro × 4) peptide (m/z 1657.93) was confirmed
by MS/MS as the desired propionylated peptide (Supporting
Information Figure S1a), showing the better amine-specificity
of NHS reagents. However, a quintuply propionylated (pro ×
5) product (m/z 1713.97) was also detected. This over-
(
Scheme 1c). From the subsequent results (Figure 1c), a full
propionylated product was confirmed by MS/MS as an
overpropionylated product with four N-propionylations on
amines and one O-propionylation on the serine residue
conversion of the peptide into the desired quadruply
propionylated product with negligible side products was
achieved, showing the great improvement in overreaction
reduction by using ammonium bicarbonate as reaction buffer.
Improving the N-Propionylation Efficiency for Secon-
dary Amines. Other than primary amines, histones from cells
contain secondary amines originating from the endogenously
monomethylated lysine residues. The secondary amines are less
reactive in reacting with acylation reagents to form correspond-
ing amides. To investigate the efficiency of the derivatization on
the secondary amines by NHS-propionate, a standard histone
H3 K27-R40 peptide K(Me )SAPATGGVK(Me )KPHR was
(Supporting Information Figure S1b). Previous studies have
shown that NHS esters could also append acyl-groups on
hydroxyl-containing amino acids such as tyrosine, serine, and
7
,9
threonine, which is evidently reflected from our data as well
Figure 1b). We first attempted to reduce the over-reaction by
(
adjusting the pH of TEAB buffer. However, the pH level had
little impact on the reaction specificity within a range from 8.0
to 10.0 (Supporting Information S2).
From the chemistry perspective, the NH groups or OH
2
1
1
groups can both serve as nucleophiles to react with NHS
reagents with the same nucleophilic substitution mechanism
synthesized and used as a testing substrate. In this peptide, both
K27 and K36 are monomethylated as secondary amines. Figure
2
256
dx.doi.org/10.1021/ac303171h | Anal. Chem. 2013, 85, 2253−2259

Analytical Chemistry
a showed the products of the peptide after derivatization
Article
2
increased propionylation on hydroxyl groups (Figure 2e). The
data clearly suggested a critical concentration needed for
efficient elimination of side reactions when keeping the
substrate concentration constant. On the basis of these above
results, the optimal reaction condition for both free and
monomethylated lysines was determined as the condition we
under the condition as described in the previous section; the
data from MALDI-TOF MS analysis showed that there were
doubly (pro × 2, m/z 1573.95), triply (pro × 3, m/z 1629.98),
and quadruply (pro × 4, m/z 1686.00) propionylated peptides
in the derivatization products. Further confirmed by MS/MS,
the quadruply propionylated peptides were the completely
propionylated products while the doubly and triply propiony-
lated peptides were the products of incomplete propionylation
in which none or only one of the secondary amines was
propionylated (Supporting Information Figure S4a,b,c). This
result demonstrated the relatively weaker reactivity of
secondary amines than primary amines on peptide substrates
as only 40% secondary amines were propionylated. To address
this issue, we attempted to optimize the reaction conditions
including the temperature of reaction and concentration of
NHS reagent. As shown in Figure 2a,b, the reaction at an
increased temperature (50 °C) converted more doubly and
triply propionylated peptides into quadruply products. (The
conversion of secondary amines was enhanced from 40% to
Table 1. m/z Value, Retention Time (RT), and
Reproducibility of Derivatization Combined with Trypsin
Digestion (RSD of Peak Area) for Histone H3 Peptides
a
peptide
m/z
RT (min) RSD (%, n = 5)
T
K
K
K
KQTAR
408.7323
535.3037
577.8482
553.3179
544.8142
473.2718
681.8850
450.7687
443.2977
724.3750
748.9289
417.2092
15.78
19.36
27.52
22.18
23.61
17.44
31.85
29.85
22.67
34.24
34.77
8.12
4.4
4.1
5.3
4.1
4.7
5.1
5.0
6.4
3.3
2.0
2.6
4.5
3
8
STGGKAPR17
9
18QLATKAAR26
27SAPATGGVKKPHR40
Y
E
Y
41RPGTVALR49
50IR52
54QKSTELLIR63
K
64LPFQR69
6
9%.) We further carried out the derivatization by escalating
L70VR72
the concentration of NHS-propionate reagents from 50 to 100
E
73IAQDFKTDLR83
mM. As expected, the efficiency was further enhanced from
V117TIMPKDIQLAR128
6
(
9% to 87% (Figure 2c). Instead of the double round approach
i.e., repeating the derivatization procedure under the same
condition) where nearly complete propionylation was achieved
data not shown), we decided to further optimize the reaction
G132ER134
a
Peptides were those with complete N-specific propionylation.
(
system by changing the concentration of the ammonium
bicarbonate. As discussed above, the ammonia produced from
ammonium bicarbonate hydrolysis plays a key role in the
competitive reaction that prevents hydroxyl groups of peptide
residues from reacting with NHS-propionate to generate
undesired O-propionylation. However, the ammonia could
also compete with the histone amines (especially the secondary
amines) in reaction with NHS-propionate, therefore reducing
the derivatization efficiency. Thus, the concentration of
ammonia or ammonium bicarbonate in reaction buffer is
considered to be a key factor affecting both side reaction
elimination and efficiency of derivatization. Since NH HCO at
used in Figure 2d. In particular, we also confirmed that this
condition would not affect the integrity of phosphorylations on
peptides (data not shown).
Histone Derivatization. To expand the method for the
intact histone proteins, the recombinant histone H3 was
derivatized using the same reaction condition we previously
optimized on the basis of the synthetic peptide substrates. Five
aliquots of histone samples were first derivatized under the
optimized conditions and then digested with trypsin. The
digested samples were further derivatized to propionylate the
newly generated N-terminal amines. The samples were
subjected to LC-MS/MS for analysis. (For more details, please
refer to “Materials and Methods”.) Table 1 summarizes the
RSD data, demonstrating the great reproducibility of the
method for the recombinant histone H3 samples (with RSD
less than 5% in general).
4
3
a concentration of 50 mM used in the previous experiments
was adequate to prevent side products, we started out using
NH HCO3 at decreased concentrations. As expected, the
4
derivatization efficiency was apparently enhanced when the
NH HCO concentration decreased from 50 to 25 mM (Figure
4
3
2
d) and the conversion rate was increased to 95%. Further
In order to assess the efficiency and specificity of the method,
we calculated the conversion rates for each of the peptides
using the following equation:
reduction of the NH HCO concentration to 10 mM led to the
complete propionylation of all the amines groups but with
4
3
XIC for desired products
XIC for desired products + XIC for incomplete products + XIC for overreacted products
conversion =
× 100%
As shown in Table 2, the general conversion rates of
derivatization were greater than 95% except for Y RPG-
Profiling of Histone PTMs in KMS11 Cells. The nuclear
receptor SET domain-containing protein 2 (NSD2, also named
MMSET and WHSC1) is a histone lysine methyltransferase
that mediates the mono- and dimethylation on histone H3 Lys
41
TVALR and E IAQDFKTDLR (conversion rates were still
49
73
83
above 90%). MS/MS results showed the N-terminal amines
were not propionylated in the incomplete propionylated
E IAQDFKTDLR peptides (Supporting Information Figure
24
3
6 (H3K36). The NSD2 protein plays an important role in
chromatin regulation and mammalian development, and the
aberrant expression of NSD2 is associated with a number of
human diseases including multiple myeloma. 15−20% of
multiple myeloma patients carry a gene translocation between
chromosomes 4 and 14 [t(4;14)(p16.3;q32)], which leads to
73
83
S5). This is probably due to the intramolecular salt bridge
between the α-NH and the γ-COOH on the Glu residue,
2
which may significantly reduce the nucleophilicity of this α-
25
NH and therefore lower the conversion.
abnormal up-regulation of NSD2 gene expression. KMS11 is
2
2
257
dx.doi.org/10.1021/ac303171h | Anal. Chem. 2013, 85, 2253−2259

Analytical Chemistry
Article
Table 2. Comparison of the MS Abundances of Histone H3 Peptides with Correct Propionylation (Desired Products),
a
Incomplete Propionylation (Incomplete Products), or Nonspecific Propionylation (Overreacted Products)
peptide
desired products
incomplete products
overreacted products
conversion, %
T KQTAR8
(3.35 ± 0.15) × 10 7
(4.69 ± 0.19) × 10 7
(8.58 ± 0.45) × 10 7
(2.60 ± 0.11) × 10 7
(3.43 ± 0.16) × 10 7
(1.40 ± 0.07) × 10 7
(1.25 ± 0.21) × 10 5
(7.63 ± 0.76) × 10 4
(1.89 ± 1.26) × 10 4
(5.59 ± 1.84) × 10 4
(1.23 ± 0.13) × 10 6
(1.21 ± 0.22) × 10 5
(2.56 ± 0.57) × 10 5
(2.12 ± 0.17) × 10 6
nd
99.5
97.3
99.8
98.9
94.2
100.0
96.7
100.0
100.0
90.5
99.0
99.9
3
K STGGKAPR
9
17
K QLATKAAR
26
18
4
K SAPATGGVKKPHR
40
(3.03 ± 1.25) × 10
27
b
Y RPGTVALR
49
nd
41
E IR
52
nd
50
7
4
5
Y QKSTELLIR
63
(1.62 ± 0.08) × 10
(2.15 ± 0.14) × 10 7
(8.81 ± 1.30) × 10
(1.20 ± 0.00) × 10 4
nd
(2.00 ± 0.19) × 10 6
(1.13 ± 0.15) × 10
(9.06 ± 1.50) × 10 3
(4.69 ± 0.32) × 10
54
K LPFQR
69
nd
nd
nd
nd
nd
64
7
L VR
72
(3.16 ± 0.11) × 10
70
E IAQDFKTDLR
83
(1.91 ± 0.04) × 10 7
73
7
5
V117TIMPKDIQLAR128
G132ER134
(1.09 ± 0.03) × 10
(1.06 ± 0.05) × 10 7
a
b
Standard deviations were obtained from 5 replicates. Not detected.
Figure 3. The global changes on histone PTMs from TKO and PAR cells. The error bars were established on the basis of triply technical replicates.
a model t(4;14)+ multiple myeloma cell line with the
overexpression of NSD2 gene. In order to investigate the
biological consequence of NSD2 translocation in cancer cell
lines, we further applied our method and compared the histone
modification levels between the KMS11 cells with selective
knock out of translocated NSD2 allele (TKO) and the original
Moreover, to our current knowledge, no modifications have
been reported on this peptide in the literature.
The changes of methylation on K27 and K36 were shown in
Figure 3. As expected, the methylation (including mono-, di-,
and trimethylation) on K36 dramatically decreased in TKO
cells while the methylation of K27 was strikingly increased,
which was consistent with the previous reports about the
26
parental KMS11 cells (PAR).
Histone proteins were first isolated by acid extraction from
25,27
crosstalk between K36 methylation and K27 methylation.
23
PAR or TKO Cells as described by Shechter et al., and
histone H3 were then fractionated by HPLC. Due to the high
efficiency and specificity of derivatization, only a little amount
of histone H3 (∼0.5 μg) was required to generate sufficient MS
signal intensity even with normal LC-MS for PTMs
quantification. After the first round of NHS-propionate
derivatization, the histone was digested by trypsin. After trypsin
digestion, the second round of derivatization was performed to
propionylate the newly generated N-terminal amines on tryptic
peptides. The samples were then analyzed by LC-MS/MS. (See
Because the trypsin cleavage after lysines was blocked by our
derivatization, some key lysine residues involved in epigenetic
regulation such as K9 and K14 (aa K9-R17) and K27 and K36
coexist on the same tryptic peptides (aa K27-R40), and
therefore, the crosstalks of combinatorial PTMs on these sites
can be investigated. Our results indicated that in either PAR or
TKO cells there was no certain peptide carrying both K27me3
and K36me3 marks, and very few peptides contained both
K27me3 and K36me2. This observation suggested an
antagonizing effect between the trimethylation of K36 and
the trimethylation of K27.
“Materials and Methods” for details.)
For the relative quantification of PTMs-containing peptides
In addition to the PTMs on peptide K27-R40, we also
profiled the other PTMs on histone H3 (see Figure 3). Our
results demonstrated that the coexisting K9 methylation and
the K14 acetylation were up-regulated to a certain degree in the
TKO cells, which was not reported previously. However, the
increase of coexisting K9 methylation and K14 acetylation was
not clear in terms of physiological association with loss of
translocated NSD2 gene expression. In addition, there were no
from both TKO and PAR cells, XIC peak areas corresponding
to each modified peptide were measured. The XIC peak areas
were further normalized against an internal “house-keeping”
peptide from histone H3 trypsin digestion to correct the
variations in histone amounts used in analysis. Peptide L VR
70
72
was selected as the internal reference peptide as it is not
endogenously modified in both TKO and PAR cell lines.
2
258
dx.doi.org/10.1021/ac303171h | Anal. Chem. 2013, 85, 2253−2259

Analytical Chemistry
Article
other significant PTMs changes on histone H3 between these
two cell lines.
(11) Plazas-Mayorca, M. D.; Zee, B. M.; Young, N. L.; Fingerman, I.
M.; LeRoy, G.; Briggs, S. D.; Garcia, B. A. J. Proteome Res. 2009, 8,
5
(
367−5374.
12) Zhang, Q.; Zhang, K.; Zou, Y.; Perna, A.; Wang, Y. J. Am. Soc.
CONCLUSIONS
■
Mass Spectrom. 2007, 18, 1569−1578.
In this paper, we have reported a novel derivatization method
for the specific and efficient derivatization of histones with
NHS-propionate. Our method can be conveniently executed
and meanwhile achieve excellent derivatization specificity and
efficiency in a reproducible manner. A competitive inhibiting
strategy was designed and then successfully used in our method
to achieve the excellent specificity for derivatization. This
strategy could also be applied in other derivatization methods
for the side reactions prevention. The effects of temperature,
pH value, and reagents concentration on the efficiency and
specificity of derivatization were examined, which would be
informative for other researchers to optimize their acylating
derivatization system. Lastly, since our method can achieve
above 90% conversion rate of histone peptides, it should be of
great value in the absolute quantification of histone modified
peptides.
(13) Mackeen, M. M.; Kramer, H. B.; Chang, K. H.; Coleman, M. L.;
Hopkinson, R. J.; Schofield, C. J.; Kessler, B. M. J. Proteome Res. 2010,
9, 4082−4092.
(
14) Miller, B. T.; Rogers, M. E.; Smith, J. S.; Kurosky, A. Anal.
Biochem. 1994, 219, 240−248.
15) Abello, N.; Kerstjens, H. A.; Postma, D. S.; Bischoff, R. J.
Proteome Res. 2007, 6, 4770−4776.
16) Anderson, G. W.; Zimmerman, J. E.; Callahan, F. M. J. Am.
Chem. Soc. 1963, 85, 3039.
(
(
(
17) Cline, G. W.; Hanna, S. B. J. Org. Chem. 1988, 53, 3583−3586.
(18) Yang, W. C.; Mirzaei, H.; Liu, X.; Regnier, F. E. Anal. Chem.
2006, 78, 4702−4708.
(19) Abello, N.; Geurink, P. P.; van der Toorn, M.; van Oosterhout,
A. J.; Lugtenburg, J.; van der Marel, G. A.; Kerstjens, H. A.; Postma, D.
S.; Overkleeft, H. S.; Bischoff, R. Anal. Chem. 2008, 80, 9171−9180.
(20) Kvaratskhelia, M.; Miller, J. T.; Budihas, S. R.; Pannell, L. K.; Le
Grice, S. F. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 15988−15993.
(21) Siezen, R. J.; Coppin, C. M.; Benedek, G. B. Biochem. Biophys.
Res. Commun. 1985, 133, 239−247.
ASSOCIATED CONTENT
■
(22) Kleinmaier, R.; Gschwind, R. M. J. Labelled Compd. Radiopharm.
2009, 52, 29−32.
*
S
Supporting Information
Additional information as noted in text. This material is
available free of charge via the Internet at http://pubs.acs.org.
(23) Shechter, D.; Dormann, H. L.; Allis, C. D.; Hake, S. B. Nat.
Protoc. 2007, 2, 1445−1457.
24) Li, Y.; Trojer, P.; Xu, C. F.; Cheung, P.; Kuo, A.; Drury, W. J.,
rd; Qiao, Q.; Neubert, T. A.; Xu, R. M.; Gozani, O.; Reinberg, D. J.
(
3
AUTHOR INFORMATION
■
*
Biol. Chem. 2009, 284, 34283−34295.
Corresponding Author
Phone: +86-216-1606328 (S.Z.); +86-216-1606085 (W.Y.).
Fax: +86-216-1606155 (S.Z.); +86-216-1606626 (W.Y.). E-
mail: shaolian.zhou@novartis.com (S.Z.); wei.yi@novartis.com
W.Y.).
(25) Martinez-Garcia, E.; Popovic, R.; Min, D. J.; Sweet, S. M.;
Thomas, P. M.; Zamdborg, L.; Heffner, A.; Will, C.; Lamy, L.; Staudt,
L. M.; Levens, D. L.; Kelleher, N. L.; Licht, J. D. Blood 2011, 117,
2
(
11−220.
26) Lauring, J.; Abukhdeir, A. M.; Konishi, H.; Garay, J. P.; Gustin, J.
(
P.; Wang, Q.; Arceci, R. J.; Matsui, W.; Park, B. H. Blood 2008, 111,
Notes
856−864.
The authors declare no competing financial interest.
(27) Yuan, W.; Xu, M.; Huang, C.; Liu, N.; Chen, S.; Zhu, B. J. Biol.
Chem. 2011, 286, 7983−7989.
ACKNOWLEDGMENTS
■
R.L. is supported by novartis presidential postdoctoral
fellowship in CNIBR. This work is also partially supported by
MOST 973 project (2012CB910504).
REFERENCES
■
(
(
1) Strahl, B. D.; Allis, C. D. Nature 2000, 403, 41−45.
2) Garcia, B. A.; Shabanowitz, J.; Hunt, D. F. Curr. Opin. Chem. Biol.
2
(
007, 11, 66−73.
3) Peng, C.; Lu, Z.; Xie, Z.; Cheng, Z.; Chen, Y.; Tan, M.; Luo, H.;
Zhang, Y.; He, W.; Yang, K.; Zwaans, B. M. M.; Tishkoff, D.; Ho, L.;
Lombard, D.; He, T.-C.; Dai, J.; Verdin, E.; Ye, Y.; Zhao, Y. Mol. Cell.
Proteomics 2011, 10, M111.012658.
(
4) Han, X.; Jin, M.; Breuker, K.; McLafferty, F. W. Science 2006, 314,
1
(
(
09−112.
5) Siuti, N.; Kelleher, N. L. Nat. Methods 2007, 4, 817−821.
6) Young, N. L.; DiMaggio, P. A.; Plazas-Mayorca, M. D.; Baliban,
R. C.; Floudas, C. A.; Garcia, B. A. Mol. Cell. Proteomics 2009, 8,
266−2284.
7) Bogdanov, B.; Smith, R. D. Mass. Spectrom. Rev. 2005, 24, 168−
00.
8) Bonaldi, T.; Imhof, A.; Regula, J. T. Proteomics 2004, 4, 1382−
396.
9) Garcia, B. A.; Mollah, S.; Ueberheide, B. M.; Busby, S. A.;
Muratore, T. L.; Shabanowitz, J.; Hunt, D. F. Nat. Protoc. 2007, 2,
33−938.
10) Drogaris, P.; Wurtele, H.; Masumoto, H.; Verreault, A.;
Thibault, P. Anal. Chem. 2008, 80, 6698−6707.
2
(
2
(
1
(
9
(
2
259
dx.doi.org/10.1021/ac303171h | Anal. Chem. 2013, 85, 2253−2259