Minimalist cyclopropene-containing photo-cross-linkers suitable for live-cell imaging and affinity-based protein labeling

Article abstract of DOI:10.1021/ja502780z

Target identification of bioactive compounds within the native cellular environment is important in biomedical research and drug discovery, but it has traditionally been carried out in vitro. Information about how such molecules interact with their endogenous targets (on and off) is currently highly limited. An ideal strategy would be one that recapitulates protein-small molecule interactions in situ (e.g., in living cells) and at the same time enables enrichment of these complexes for subsequent proteome-wide target identification. Similarly, small molecule-based imaging approaches are becoming increasingly available for in situ monitoring of a variety of proteins including enzymes. Chemical proteomic strategies for simultaneous bioimaging and target identification of noncovalent bioactive compounds in live mammalian cells, however, are currently not available. This is due to a lack of photoaffinity labels that are minimally modified from their parental compounds, yet chemically tractable using copper-free bioorthogonal chemistry. We have herein developed novel minimalist linkers containing both an alkyl diazirine and a cyclopropene. We have shown chemical probes (e.g., BD-2) made from such linkers could be used for simultaneous in situ imaging and covalent labeling of endogenous BRD-4 (an important epigenetic protein) via a rapid, copper-free, tetrazine-cyclopropene ligation reaction (k₂ > 5 M^-1 s^-1). The key features of our cyclopropenes, with their unique C-1 linkage to BRD-4-targeting moiety, are their tunable reactivity and solubility, relative stability, and synthetic accessibility. BD-2, which is a linker-modified analogue of (+)-JQ1 (a recently discovered nanomolar protein-protein-interaction inhibitor of BRD-4), was subsequently used in a cell-based proteome profiling experiment for large-scale identification of potential off-targets of (+)-JQ1. Several newly identified targets were subsequently confirmed by preliminary validation experiments.

Full text of DOI:10.1021/ja502780z

Subscriber access provided by George Mason University Libraries & VIVA (Virtual Library of Virginia)
Article
“Minimalist” Cyclopropene-Containing Photo-Cross-Linkers
Suitable for Live-Cell Imaging and Affinity-Based Protein Labeling
Zhengqiu Li, Danyang Wang, Lin Li, Sijun Pan, Zhenkun Na, Chelsea Y J Tan, and Shao Q. Yao
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/ja502780z • Publication Date (Web): 27 Jun 2014
Downloaded from http://pubs.acs.org on July 1, 2014
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
Zhengqiu Li, Danyang Wang, Lin Li, Sijun Pan, Zhenkun Na, Chelsea Y. J. Tan, and Shao Q. Yao*
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543
ABSTRACT: Target identification of bioactive compounds
within the native cellular environment is important in biomedi-
cal research and drug discovery, but has traditionally been car-
ried out in vitro. Information about how such molecules interact
with their endogenous targets (on and off) is currently highly
limited. An ideal strategy would be one that recapitulates pro-
tein-small molecule interactions in situ (e.g. in living cells), and
at the same time enables enrichment of these complexes for
subsequent proteome-wide target identification. Similarly,
small molecule-based imaging approaches are becoming increasingly available for in situ monitoring of a variety of proteins includ-
ing enzymes. Chemical proteomic strategies for simultaneous bioimaging and target identification of non-covalent bioactive
compounds in live mammalian cells, however, are currently not available. This is due to a lack of photo-affinity labels that are
minimally modified from their parental compounds, yet chemically tractable using copper-free bioorthogonal chemistry. We have
herein developed novel minimalist linkers containing both an alkyl diazirine and a cyclopropene. We have shown chemical probes
(e.g. BD-2) made from such linkers could be used for simultaneous in situ imaging and covalent labeling of endogenous BRD-4 (an
important epigenetic protein) via a rapid, copper-free, tetrazine-cyclopropene ligation reaction (k2 > 5 M-1 s-1). The key features of
our cyclopropenes, with their unique C-1 linkage to BRD-4-targeting moiety, are their tunable reactivity and solubility, relative
stability and synthetic accessibility. BD-2, which is a linker-modified analog of (+)-JQ1 (a recently discovered nanomolar protein-
protein-interaction inhibitor of BRD-4), was subsequently used in a cell-based proteome profiling experiment for large-scale
identification of potential off-targets of (+)-JQ1. Several newly identified targets were subsequently confirmed by preliminary
validation experiments.
has been a terminal alkyne because it is small, chemically inert
and can be further modified for downstream proteomic
applications with azide-containing reporters via copper-
Target identification of small molecule-based bioactive
compounds within the native cellular environment is important
catalyzed alkyne-azide cycloaddition (CuAAC).10 In the case
of non-covalent bioactive compounds, which account for > 95%
in biomedical research and drug discovery, but has
of all FDA-approved drugs and candidates, another key
traditionally been carried out in vitro by using recombinant
proteins or crude cellular lysates.1 Information about how such
consideration is the introduction of an additional photo-
reactive moiety within the probe for on-demand (e.g. by UV
molecules interact with their endogenous targets (on and off)
is therefore highly limited.2 An ideal strategy would be one
irradiation), in-cell conversion of transient protein-ligand
complexes into stable, covalent ones (Figure 1A).8,9 Such
that recapitulates protein-small molecule interactions in situ
(e.g. in living cells), and at the same time enables enrichment
photo-affinity labeling (PAL) approaches had previously been
used to study different types of non-covalent interactions.11 In
of these complexes for subsequent proteome-wide target
identification.3 In the last several years, by taking cue from
the context of “drug profiling”, however, the key requirement
concepts developed for activity-based protein profiling,4 cell-
is the chemically tractable photo-cross-linker must be made as
small as possible.3 Our recently developed 1st-generation
based proteome profiling methods have emerged; drug-like
chemical probes minimally modified from their parental
“minimalist” linkers, which contain an alkyl diazirine and a
terminal alkyne connected by short aliphatic chains, fulfill
compounds have been used for large-scale interrogation of
such a requirement, and have been used to tag many kinase
protein-small molecule interactions and rapid identification of
potential cellular targets in live cells.3,5,6 Such an “in situ drug-
profiling” approach is applicable to compounds that form
inhibitors with minimal loss in their native biological activities
(Figure 1A).9 But they fall short of being ideal due to the need
of Cu(I) catalysts in CuAAC, thus are ineffective in events
either irreversible or reversible complexes with their intended
targets.7-9 The probe design involves introduction of a
where both photo-cross-linking and reporter-tagging (e.g. in
bioimaging applications12) are conducted simultaneously in
chemically “tractable” handle which should not disrupt
situ. Weissleder et al recently reported trans-cyclooctene
protein-ligand interaction in situ. So far, the handle-of-choice
(TCO)-modified kinase inhibitors with tetrazine-containing
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. (A) Two drug-profiling PAL approaches: (left) previously developed approach in which bioactive compounds were “tagged”
with 1st-generation “minimalist” linkers;9 (right) 2nd-generation approach reported in the current work, with cyclopropenes as chemically
tractable tags suitable for copper-free bioorthogonal chemistry. (B) Structures of previously reported cyclopropenes (1-4),18-20 and newly
reported ones (L4 & L7; boxed). (C) Known BRD-4-targeting compounds. (D) the x-ray complex (PDB code: 3MXF) of BRD-4/(+)-
JQ1 showing the active-site binding (right panel).24a
fluorescent reporters for live-cell imaging of kinase
activities.12a,b Other small molecule-based imaging approaches
have also become increasingly available for in situ monitoring
of a variety of proteins including enzymes.12c-h While elegantly
designed and effective, Weissleder’s kinase-imaging probes
have some room for further improvement. For example, the
replacement of the relatively bulky TCO moiety with smaller
“clickable“ tags and the introduction of a photo-cross-linker
would render this strategy suitable for above-mentioned in situ
imaging/profiling applications.13 We previously showed that
chemical modifications of bioactive compounds with “tags“ of
different sizes had a significant effect on target recognition.8
Furthermore, introduction of a covalent linkage between a
potentially diffusible small molecule imaging probe and its
intended targets has been shown to further improve imaging
resolution in situ by effectively “fixing“ the probe near the
reaction site.14 Herein, we report the development of L4 & L7,
our 2nd-generation minimalist linkers (Figure 1B; boxed),
which contain a novel cyclopropene handle and a diazirine
within the same molecule, and their successful incorporation
into protein-protein-interaction (PPI) inhibitors of BET
bromodomains (e.g., BRD-4; Figure 1C & 1D). We show, for
the first time, chemical probes made from such linkers could
be used for simultaneous imaging and protein labeling in live
cells via copper-free tetrazine-cyclopropene ligation reaction.
carbohydrates and lipids.15 We were however worried about its
susceptibility to be reduced by endogenous thiols. Strained
alkenes/alkynes (i.e. TCO, norborenes, cyclooctynes) as well
as tetrazines are also amenable to metabolic incorporation, but
most of them are too bulky for our applications.16,17 We chose
substituted cyclopropenes, which are similar in size to a
terminal alkyne, metabolically stable, and undergo rapid
bioorthogonal reaction with a suitable tetrazine (e.g. a second-
order rate constant k2 of up to 13 M-1 s-1 for 3).18 We were also
intrigued by the tunability of both the stability and reactivity in
cyclopropenes, which are affected sterically and electronically
by their C-1/C-2/C-3 substituents (Figure 1B, 1-4).18,19 These
features might be further explored in multiplex experiments by
using mutually exclusive cyclopropene pairs.20 The relative
chemical instability of both alkyl diazirines and cyclopropenes
was another concern in our linker design. As unhindered
cyclopropenes are prone to self-polymerization and attack by
biological nucleophiles,18,19 and both cyclopropenes and
diazirines are sensitive to harsh reaction conditions, having
them present within the same linker poses significant synthetic
challenges. Finally, we settled on the two diazirine-containing
cyclopropene linkers, L4/L7. Unlike other reported
cyclopropenes, where compounds were linked at C-3
position,18-21 we chose to attach compounds via C-1 linkage
(Figure 1B). This offers several advantages: (1) C-3 is now
available for further tuning of the cyclopropene’s reactivity
and solubility; (2) C-1 attachment offers the needed steric
hindrance in the resulting cyclopropenes for improved stability
without introducing an additional methyl at either C-1 or C-2
position (e.g. 1-4 in Figure 1B); (3) L4/L7 are now
synthetically accessible from their terminal-alkyne precursors
(e.g. our 1st-generation linkers9) by using the well-established
rhodium-catalyzed reaction of -diazo esters with alkynes.22
Design and Synthesis. In order to render our 1st-generation
minimalist linkers copper-free, we surveyed numerous
“handles” used in bioorthogonal chemistry as possible
replacement of terminal alkynes.15-17 An –N3 handle was
initially considered as it is small and has been widely used for
metabolic incorporation of unnatural amino acids,
ACS Paragon Plus Environment

Page 3 of 10
Journal of the American Chemical Society
BD-1 and BD-2 should maintain similar protein-binding
Scheme 1. Synthesis of Diazirine Containing Cyclopropene
Linkers (L4/L7) and BRD-4 Targeting Probes (BD-1/3).
1
2
property as their parental compound GW841819X (a close
analog of (+)-JQ1).26 To synthesize BD-1/2, L4 was first
converted to L5 with NaN3 (86% yield), then reduced to L6
with PPh3/H2O/THF (76%). Subsequent coupling between L6
and S1 afforded BD-1 (65%), which was further reduced to
give BD-2 (43%). BD-3 (a reference probe tagged with our 1st-
generation minimalist linker) and NP-1/NP-2 (negative
control probes for BD-3/BD-2, respectively) were similarly
synthesized. With an electron-donating hydroxymethyl group
at C-3 position in the cyclopropene of BD-2, we expected this
probe, compared to BD-1, to possess better water solubility
and faster ligation toward tetrazine-containing reporters.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ligation Studies with Different Tetrazines. With these
probes in hand, we first examined the tetrazine-ligation
reactivity and biocompatibility of L4/L7 (Figure 2).
Cyclopropene 5 was tested concurrently as a reference
compound.18 Different tetrazines (6-8; Figure 2C), including
Biotin-TZ2 (8), were used, and the ligation reactions were
monitored by both LCMS (Figures S1-S7), and spectrometric
measurement of disappearance in the tetrazine absorption band
at 520 nm to obtain the second-order rate constant (Figure 2B-
D; Figures S10 & S11). We obtained a rate constant of 5.03 ±
0.34 M-1 s-1 for L7 with 8, which is similar to one of the fastest
tetrazine-cyclopropene ligations reported (e.g. k2 up to 13 M-1
s-1 with cyclopropene 318). As expected, L4, with an electron-
withdrawing C-3 ester, reacted sluggishly (k2 = 0.31 x 10-2 M-1
s-1 with 8), comparable to that of 5 under identical assay
conditions (k2 = 0.56 x 10-2 M-1 s-1). Similar reactivity trends
between L4 and L7 with other tetrazines were observed
(Figure 2C); in all cases, L7 consistently reacted between 10-
to 2000-fold faster than L4, with k2 ranging from 0.1 to 5.03
M-1 s-1. While these reactions are slower than some other
tetrazine-based ligations,16,17 they are still much faster than the
Staudinger ligation (k2 < 0.01 M-1 s-1) – a bioorthogonal
Synthesis of L4 was accomplished from the previously
reported L3 by treatment with commercially available ethyl
diazoacetate and a catalytic amount of Rh2(OAc)4 (Scheme
1).9 The single-step synthesis of L4 from L3 clearly outweighs
its relatively low yield (21%). Further DIBAL reduction of the
C-3 ester in L4 afforded L7 in excellent yield (85%). It should
be noted that, unlike reported cyclopropenes (1-4),18-20 which
needed obligatory TMS protection at either C-1 or C-2
position during synthesis, both L4 and L7 were stable under
our reaction conditions without any special handling. With
these 2nd-generation minimalist linkers, we next used them to
make BRD-4 targeting chemical probes (BD-1 and BD-2).
BRD-4 is a BET bromodomain that recognizes acetylated
lysine (Kac) residues such as those located on histones. It is an
important epigenetic “reader” involved in chromatin
remodeling, DNA damage, cell-cycle control and other critical
cellular processes.23 Recent findings that benzodiazepine-
containing compounds such as (+)-JQ1, I-BET and
GW841819X (Figure 1C) are nanomolar PPI inhibitors of
BRD-4 have spurred enormous interests to further develop
such compounds into potential drugs against diseases
including cancer, HIV infection and heart disease.24,25 Previous
proteomic study was carried out, in an in vitro pull-down
experiment by using an immobilized I-BET analog together
with mammalian cell lysates, to unequivocally identify BRD-4
as the endogenous target of these compounds.26 Their potential
off-targets, however, have not been investigated at the
proteome level in situ (e.g. in live cells, not lysates), due to a
lack of suitable chemical probes. Based on the published x-
ray structure of (+)-JQ1/BRD-4 complex (Figure 1D),24a
substituents at C6 and on both the triazole and thiophene rings
in (+)-JQ1 were critical in binding to the Kac-binding pocket
of BRD-4, while the t-butylacetyl group at C4 was
nonessential. A recent study showed surface-immobilized (+)-
JQ1 via C4 linkage retained its full BRD-4 binding property.27
We thus expected that, with an C4 appendage to L4/L7, both
reaction widely used in live cells and animals.15
A
representative ligation reaction is shown in Figure 2A, with
the structure of the expected ligation products confirmed by
both LCMS and NMR (Figure S1). Although asymmetric
cyclopropene/tetrazine ligation could produce up to 4 different
diastereomers,18-20 we were unable to chromatographically
separate them. Ligation products from cyclopropenes that
contain a suitable C-3 nucleophile were previously shown to
undergo further intramolecular cyclization (Figure S2);19 no
such evidence was found in our system.
Biocompatibility and Turn-ON Properties of L7. The
cyclopropene moiety in L7 was found to be highly stable in
aqueous buffers and in the presence of biological nucleophiles
(L-cysteine) for extended periods of time (Figures S8 & S9),
thus making them suitable for in situ applications. Weissleder
et al recently showed tetrazines could effectively quench the
fluorescence of several organic fluorophores, and the authors
further exploited this unique feature in live-cell bioimaging to
minimize nonspecific background fluorescence labeling, by
making use of tetrazine-containing fluorogenic reporters
whose fluorescence was “Turned-ON” only upon ligation with
a strained dienophile.12a,b We wondered whether such a system
might work with our newly developed diazirine-containing
cyclopropenes. As shown in Figure 2D & 2E, with a cell-
permeable, tetrazine-containing fluorogenic reporter FL-TZ1,
the intrinsic fluorescence of its deacetylated form, S10, went
from the “OFF” to the fully “ON” state within 5 min of
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 10
1
(D)
(A)
2
3
4
5
6
7
8
9
8 + L4
8 + 5
8 + L7
(B)
k2 = 0.31 ± 0.07
(10-2 M-1s-1
k2 = 0.56 ± 0.07
(10-2 M-1s-1)
k2 = 502.5 ± 33.9
(10-2 M-1S-1
)
)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.75
0.5
0.25
(E)
65000
FL-TZ1
FL-TZ1/L7
50 min
100 min
50 min
100 min
Time
50 s
100 s
(C)
cyclopropene
Tetrazine
k2 (10-2M-1 s-1
0.45 ± 0.07
20.0 ± 1.98
2.6 ± 0.95
)
L4
L7
L4
L7
L4
L7
5
25.7 ± 1.7
O
0.31 ± 0.07
502.5 ± 33.9
0.56 ± 0.07
HN
NH
H
H
H
H
N
N
N
N
N
0
N
N
Biotin-TZ2
(8)
O
S
2
550
600
Wavelength (nm)
650
N
O
O
Figure 2. (A) Scheme of tetrazine 8 reacting with cyclopropene L4, L7 or 5. Different diazanorcaradiene regioisomeric products could
form, but for simplicity, only one regioisomer is depicted. (B) Plots of tetrazine absorbance (520 nm) versus time between ligation of 8
and L7, L4 or 5. Data was fitted to a first-order exponential decay to obtain the corresponding second-order rate constant (k2). (C) k2 of
different tetrazines 6, 7 or 8 reacting with L7, L4 or 5. (D) Schematic showing of ligation reaction between FL-TZ1 and L7. (E) Emission
spectra of deacetylated FL-TZ1 alone, and 5 min after ligation with L7. 10 L of FL-TZ1 (10 mM in DMSO) was first treated with 20
L K2CO3 solution (100 mM) for 2 h at room temperature. Next, 40 L L7 (15 mM in 10% DMSO/H2O) was added and the reaction mix-
ture was incubated for 5 min. Upon dilution (500 x with 10% DMSO/H2O), the emission fluorescence was measured (ex = 488 nm). Con-
trol sample was processed under the same conditions without addition of L7.
ligation with L7. The same effect was observed with a
tetraethylrhodamine-containing reporter (Figure S10).
3/protein labeling was detected only after 5-60 min of click
reaction under CuAAC conditions. This highlights the
extremely fast ligation reaction between the cyclopropene in
BD-2 and the tetrazine reporter. In comparison, fluorescent
bands of BD-1/protein labeling reactions started to appear
after 1 min of ligation (with TER-TZ1), but only reached a
comparable level after 60 min. This agrees well with the
ligation rate difference of L4 and L7 as earlier observed
(Figure 2B & 2C). As BD-2 appeared to be a more superior
probe than BD-1 in both its labeling efficiency and ligation
speed, it was chosen for all subsequent studies. We next
confirmed BD-2 could be used to selectively label BRD-4 in a
more complex environments (Figure 3C); a fluorescent band
corresponding to the molecular weight of (His)6-tagged BRD-
4 was visible only in bacterial lysates over-expressing this
protein, and the fluorescence intensity was abolished in the
presence of excessive (+)-JQ1. Similar results were obtained
with BD-3, the terminal alkyne-containing reference probe.
Subsequently, in order to evaluate the live-cell imaging
capability of BD-2 towards BRD-4 endogenously expressed in
mammalian cells, we directly added the probe to the medium
of growing HepG2 cells, with or without 10 times of (+)-JQ1
as a competitor. Control experiments were carried out with
NP-2. Upon UV irradiation to cross-link endogenous
probe/protein complexes, the cells were directly treated with
the cell-permeable FL-TZ1, then imaged (Figure 3D); strong
green fluorescence signals were detected throughout the
Bioimaging and Affinity-Based Protein Labeling. We
next assessed whether the two newly developed,
cyclopropene-containing chemical probes could be used for
simultaneous imaging and covalent labeling of BRD-4 protein.
We first confirmed, in an isothermal titration calorimetry (ITC)
experiment, that the probes could still bind to recombinant
(His)6-tagged BRD-4 protein with similar affinity as (+)-JQ1
(Figure 3A); the Kd values of BD-2 (860 nM) and BD-3 (810
nM) were both within ~ two-fold range of that of (+)-JQ1 (390
nM) under identical assay conditions, thus confirming the
minimal effect caused by C4 modifications in (+)-JQ1 as
previously reported.24-27 We next showed both probes were
able to efficiently label recombinant BRD-4 protein under
UV-irradiation conditions (Figure 3B & 3C). First we carried
out labeling with recombinantly purified BRD-4. Briefly, upon
UV irradiation, the cross-linked protein/BD-2 and protein/BD-
1 complexes were treated with TER-TZ1 for different lengths
of ligation time (1 to 60 min). For comparison, protein/BD-3
complexes were “clicked” with TER-N3 under standard
CuAAC conditions.9 Labeled samples were separated and
visualized by in-gel fluorescence scanning (Figure 3B); we
observed strong fluorescent bands of BD-2/protein labeling in
as little as 1 min of tetrazine-cyclopropene ligation reaction
(right), with 1-20 M of the probe (left). On the contrary, BD-
ACS Paragon Plus Environment

Page 5 of 10
Journal of the American Chemical Society
nucleus of labeled cells, which colocalized well with signals indicates the possible presence of both common and unique
1
2
3
4
5
6
7
8
obtained from the immunofluorescence (IF) results with anti-
BRD-4 antibody (see panel iv). No fluorescence was detected
in cells treated with NP-2 (panel vii), or with 10 x competitive
(+)-JQ1 (panel iii).
cellular targets of each probe (vide infra). We nevertheless
could conclude that the newly developed cyclopropene-
containing probe performed similarly as our 1st-generation,
terminal alkyne-containing minimalist probe in cell-based
proteome profiling experiments.
In Situ Target Identification and Validation. Next, the
proteome reactivity profiles of the same in situ-labeled HepG2
cells were further analyzed (Figure 3E); upon cell lysis, the
labeled samples were ligated with TER-TZ1 (TER-N3 for
BD-3-labeled samples), separated by SDS-PAGE, then
analyzed by in-gel fluorescence scanning (top gel).
Concurrently, a portion of the same sample was ligated with
Biotin-TZ2, affinity-enriched on avidin-agarose beads, then
separated, followed by Western blotting (WB) (bottom gel). A
73-kDa band in the pull-down (PD) samples treated with BD-2
(1 and 5 M) was successfully detected by anti-BRD-4
antibody, but not in samples labeled with NP-2 or in the
presence of 10 x competitive (+)-JQ1, clearly showing the
successful labeling of endogenous BRD-4 by BD-2. Similar in
situ proteome labeling results were obtained with the reference
probe BD-3. Interestingly, the proteome reactivity profiles of
HepG2 cells labeled by BD-2 and BD-3, though largely
similar as one might expect, did show some minor but distinct
differences (compare lanes 2 & 4 in Figure 3E top gel). This
Finally, large-scale PD/LC-MS/MS analysis was carried out
to delineate other unknown cellular targets (on and off) of (+)-
JQ1, as well as to compare the difference in the performance
between BD-2 and BD-3 (Figure 3F & 3G, Table 1 &
Supporting Information). In order to minimize “false
hits“ caused by non-specific protein binding and intrinsic non-
specific labeling of PAL, proteins that appeared in LC-MS/MS
results of experiments performed with negative probes (NP-1
& NP-2) and with 10 x (+)-JQ1 were removed. In addition,
only proteins that appeared in duplicated/triplicated samples
were considered further. At the end, we identified 420 and 326
candidates for BD-2 and BD-3, respectively, in which ~41%
of the proteins (132; relative to 326) were pulled down by both
probes (Figure 3F). Further analysis of these 132 proteins
indicated a large number of them (49) were nuclear proteins
(Table S3). By analysis of putative (+)-JQ1 targets under more
stringent criteria, we obtained 161 and 117 higher-confidence
candidate proteins for BD-2 and BD-3, respectively, with 48
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(B)
[Probe]/μM
BD-2
BD-1
5
BD-1
15 30 60
BD-2
15 30 60
BD-3
BD-3
5
(A)
1
20
1
5
20
1
20
1
5
1
5
t/min
1
5
15 30 60
FL
CBB
(C)
[JQ1]/μM
BD-2
BD-3
80 40 20 10
5
2.5
0
UI
80 40
20 10
5
2.5
0
UI
kDa
70
34
23
17
FL
CBB
FL
Merged
CBB
DIC
BD-2/FL-TZ1
BD-2/(+)-JQ1
IF (anti-BRD-4)
Nu
NP-2/FL-TZ1
(D)
I
ii
iii
iv
v
vi
vii
(E)
(F)
BD-2
BD-2
BD-3
5
-
BD-3
(G)
[Probe]/μM
1
5
1
1
5
+
1
5
5
5
BD-2
BD-3
-
-
-
+
+
+
-
-
10 x JQ1
kDa
70
55
127 kDa
51 kDa
anti-DDB1
anti-RAD23B
FL
288
132
194
25
PD/WB
anti-BRD-4
PD/WB
Figure 3. (A) ITC results of (+)-JQ1/BD-2/BD-3 binding to (His)6-tagged BRD-4. See Figure S12 for details. (B) Concentration-
dependent labeling of BD-1/2/3 with recombinant BRD-4 (2 g; left gels), and time-dependent ligation (1 to 60 min; 5 M probe) of
probe-labeled BRD-4 with TER-TZ1 (for BD-1/2) and TER-N3 (for BD-3). FL = in-gel fluorescence scanning. CBB = Coomassie gel.
(C) Labeling of (His)6-tagged BRD-4 over-expressing bacterial lysates by 1 M of BD-3 (left gels) or BD-2 (right gels), in the presence
of different amounts of (+)-JQ1 (0-80 M), with TER-TZ1 (10 M; 2 h ligation). (D) Live-cell imaging of BD-2/NP-2 (5 μM) followed
by ligation with FL-TZ1. IF = immunofluorescence. Merged: panels i + ii. (Insets in panels iv & v): merged panels ii/iv (with a Pearson’s
coefficient R = 0.76) & ii/v, respectively. Scale bar = 10 m. (E) In-gel fluorescence scanning showing the proteome reactivity profiles of
live HepG2 cells labeled by BD-2/BD-3, with or without 10 x (+)-JQ1. The corresponding pull-down (PD)/Western blotting (WB) results
are shown (bottom gel). For BD-2-labeled cells, ligation was done with TER-TZ1/Biotin-TZ2. NP-1/NP-2: negative control PD. (F)
Venn diagram showing the number of proteins labeled/enriched by BD-2/3 (1 M) upon in situ labeling/PD/LC-MS/MS with live HepG2
cells. Non-specific protein labeling was minimized by control labeling experiments with NP probes, and with 10 x (+)-JQ1. (G) Prelimi-
nary validation of the two newly identified off-targets of (+)-JQ1 by PD/WB from BD-2 (1 M)-labeled HepG2 cells. BD-3 labeling was
carried out concurrently as a reference. The corresponding control experiments were done with negative probes (NP-1 & NP-2) and with
10 x (+)-JQ1.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 10
Table 1. Selected High-Confidence Nuclear Proteins Enriched
by Both BD-2 and BD-3.
1
In conclusion, the copper-free, minimalist photo-cross-
2
3
4
5
6
7
8
9
mass
(in Da)
protein
protein scorea
emPAIa
linkers disclosed in this work have led to successful
development of the first affinity-based probes capable of both
imaging and covalent labeling of endogenous BRD-4 in live
mammalian cells. Key features of cyclopropenes used in these
novel linkers, with unique C-1 linkage to protein-targeting
moiety, are their tunable reactivity and solubility, relative
stability and synthetic accessibility. With an increasing
number of terminal alkyne-containing chemical probes
available in the literature,3 some may be converted to their
cyclopropene-containing counterparts using the chemistry
developed herein for future copper-free, chemical profiling
applications. From our study, we have successfully carried out
in situ proteome profiling in live HepG2 cells. Subsequent
large-scale pull-down/LC-MS/MS experiments have resulted
in the identification of several hundred candidate proteins
targeted by (+)-JQ1. With the dual capability of our 2nd-
generation minimalist probes in simultaneous live-cell
imaging and in situ proteome profiling, and the availability of
different reporter tags, these unique features were further
exploited to narrow down a list of high-confidence off-targets
of (+)-JQ1, two of which were subsequently validated by
preliminary experiments.
MYOF
DDB1
USP7
TXNRD1
HNRNPK
NAP1L1
RAD23B
CAPG
236100
128142
117951
60725
51230
44972
43202
38760
36142
2195
295
396
1993
1168
706
224
282
460
0.75
0.19
0.24
3.15
0.98
1.03
0.25
0.28
0.55
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
PP2CA
aProtein scores and emPAI values from BD-2 samples are shown.
For values from BD-3 sample, see Table S3. Shaded proteins
were further validated (Figure 3G).
overlapping proteins in which 14 were nuclear localized
(Table S2). Selected proteins are summarized in Table 1. It is
somewhat surprising that, given the high degree of structural
homology in BD-2 and BD-3, only 41% of high-confidence
proteins pulled-down by both probes were identical. This
could be due to a number of factors: (1) the intrinsic
variability of the instrument used in LC-MS/MS experiments.
In fact, even with identical PD samples, duplicated/triplicated
injections could only achieve, on average, 75% coverage of
the same proteins; (2) the linker difference in BD-2/BD-3.
Despite careful choices of linker appendage point and
minimized linker size, the two probes afforded insignificant
but noticeable differences in their proteome reactivity profiles
as shown in Figure 3E. Future studies will need to be carried
out to better understand the exact effect of linker on the
outcome of off-target identification. Notwithstanding, with the
dual capability of our new minimalist probes for both live-cell
imaging and in situ proteome profiling, and the availability of
different reporter tags (i.e. terminal alkyne and cyclopropene),
these features were used to further improve the confidence
level of candidate hits identified form our PD/LC-MS/MS
experiments. Table 1 represents a list of selected proteins
which are likely genuine off-targets of (+)-JQ-1, as they were
both nuclear-localized (as supported by our imaging results)
and successfully enriched by both BD-2 and BD-3. We rea-
soned that a true cellular target of (+)-JQ1 would bind to both
probes with similar affinity. In future proteomic experiments,
such a combination approach by using both alkyne- and cy-
clopropene-tagged bioactive compounds might prove to be
highly effective to help in the elimination of many potential
false positives. To further confirm some of these proteins were
indeed true cellular targets of (+)-JQ1, we carried out
preliminary target validation experiments on two candidate
proteins, DDB1 and RAD23B (Figure 3G). These two proteins
were chosen because they are key proteins involved in DNA
damage/repair pathways, and BRD-4 was recently found to be
activated in response to DNA damage.25b By performing the
same in situ proteome labeling/PD with live HepG2 cells and
WB analysis of the enriched lysates with anti-DDB1 and anti-
RAD23B antibodies, both proteins were successfully
identified from both BD-2 and BD-3 labeled cells, but not
with 10 x (+)-JQ1, indicating they are likely true cellular
targets of (+)-JQ1. We caution, however, more extensive
biological experiments will be needed in future to further
substantiate our findings.
General Information. All chemicals were purchased from
commercial vendors and used without further purification,
unless indicated otherwise. All reactions requiring anhydrous
conditions were carried out under argon or nitrogen
atmosphere using oven-dried glassware. HPLC-grade solvents
were used for all reactions. Reaction progress was monitored
by TCL on pre-coated silica plates (Merck 60 F254 nm, 0.25 µm)
and spots were visualized by UV, iodine or other suitable
stains. Flash column chromatography was carried out using
silica gel (Merck 60 F254 nm). All NMR spectra (1H-NMR, 13C-
NMR) were recorded on Bruker 300/500 MHz NMR
spectrometers. Chemical shifts are reported in parts per
million referenced to appropriate internal standards or residual
solvent peaks (CDCl3 = 7.26 ppm, DMSO-d6 = 2.50 ppm).
The following abbreviations were used in reporting spectra, br
s (broad singlet), s (singlet), d (doublet), t (triplet), q (quartet),
m (multiplet), dd (doublet of doublets). Mass spectra were
obtained on Shimadzu IT-TOF-MS or Shimadzu ESI-MS
system. All analytical HPLC were carried out on Shimadzu
LCMS (IT-TOF) system or Shimadzu LCMS-2010EV system
equipped with an auto-sampler using reverse-phase
Phenomenex Luna 5 μm C18(2) 100 Å 50 × 3.0 mm columns.
Water with 0.1% TFA and acetonitrile with 0.1% TFA were
used as eluents and the flow rate was 0.6 mL/min.
HepG2 cells were cultured in Dulbeccoʼs modified Eagle
medium (DMEM; Invitrogen) containing 10% heat-inactivated
fetal bovine serum (FBS; Invitrogen), 100 units/mL penicillin,
and 100 μg/mL streptomycin (Thermo Scientific) and
maintained in a humidified 37 °C incubator with 5% CO2. To
generate protein lysates, cells were washed twice with cold
phosphate-buffered saline (PBS), harvested with 1 × trypsin or
by use of a cell scraper, and collected by centrifugation. Cell
pellets were then washed with PBS and lysed with 25 mM N-
(2-hydroxyethyl)piperazine-N-2-ethanesulfonic acid (HEPES)
buffer (with 150 mM NaCl, and 2 mM MgCl2, pH 7.5)
ACS Paragon Plus Environment

Page 7 of 10
Journal of the American Chemical Society
containing 0.1% NP-40. Protein concentration was determined Ethyl
2-(2-(3-(2-aminoethyl)-3H-diazirin-3-
1
2
3
4
5
6
7
8
by Bradford protein assay. For Western blotting (WB)
experiments, samples from HepG2 cells were resolved by
SDS−PAGE and transferred to poly(vinylidene difluoride)
membranes. Membranes were then blocked with 3% BSA in
TBST (0.1% Tween in Tris-buffered saline) for 1 h at room
temperature. After blocking, membranes were incubated with
the corresponding primary antibody for another hour. After
incubation, membranes were washed with TBST (4 × 10 min)
and then incubated with an appropriate secondary antibody.
Finally, blots were washed again with TBST before being
developed with SuperSignal West Dura Kit (Thermo
Scientific). BRD-4 recombinant protein was expressed and
purified as described previously.24a,27 Antibodies against BRD-
4 (ab75898), DDB1 (EPR6089) & RAD23B (ab86781) were
purchased from Abcam.
yl)ethyl)cycloprop-2-enecarboxylate (L6). To a solution of L5
(348 mg, 1.4 mmol) dissolved in THF/water (10:1, 5 mL) was
added triphenylphosphine (0.41 g, 1.6 mmol) at room tem-
perature. The reaction mixture was stirred for 10 h before ad-
dition of 1 N HCl (3 mL). Upon extraction with diethyl ether,
the aqueous layer was neutralized with 1 N NaOH, and the
resulting mixture was further extracted with diethyl ether,
before being concentrated in vacuo to yield the desired prod-
uct L6 (223 mg, 76%). 1H NMR (300 MHz, CDCl3) δ 6.46 (s,
1H), 4.14 (m, 2H), 2.51 (t, J = 7.5, 2H), 2.36 (m, 2H), 1.74 (t,
J = 7.2, 2H), 1.62 (t, J = 7.2, 2H), 1.27 (t, J = 7.5, 2H); 13C
NMR (75 MHz, CDCl3) δ 176.19, 114.11, 95.53, 60.34, 36.42,
35.45, 32.84, 30.84, 27.27, 19.92, 14.27.
BD-1. Compound S1 was synthesized by following previ-
ously reported procedures.27 To a stirred solution of S1 (66.4
mg, 0.2 mmol) in DMF (2 mL) was added HOBt (32 mg, 0.24
mmol), EDCI (46 mg, 0.24 mmol), DIEA (52 mg, 0.4 mmol)
and L6 (49 mg, 0.22 mmol). The mixture was stirred for 12 h
and diluted with water. Subsequently, the mixture was extract-
ed with 2 × 10 mL EtOAc. The organic phase was washed
with brine. Upon solvent evaporation, the residue was purified
by flash column (100:1 to 20:1 DCM:MeOH) to afford BD-1
as a white solid (69 mg, 65%). 1H NMR (500 MHz, CDCl3) δ
7.70 (m, 1H), 7.51-7.30 (m, 8H), 6.84 (br, 1H), 6.49 (s, 1H),
4.58 (t, J = 7.5 Hz, 1H), 4.13 (m, 2H), 3.54 (m, 1H), 3.36 (m,
1H), 3.13 (m, 2H), 2.66 (s, 3H), 2.30 (m, 2H), 2.13 (s, 1H),
1.69 (m, 2H), 1.64 (m, 2H), 1.28 (t, J = 7.5 Hz, 3H); 13C NMR
(125 MHz, CDCl3) δ 176.27, 170.62, 167.74, 138.73, 131.97,
131.54, 130.65, 129.41, 129.30, 128.19, 127.14, 123.12,
111.78, 95.90, 60.34, 53.71, 39.31, 34.81, 32.31, 30.10, 26.65,
19.94, 19.86, 14.26, 12.15. HR-MS (ESI) calcd for [M+Na]+
560.2386; Found 560.2392.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Kinetic studies of tetrazine-cyclopropene ligation were
performed mostly based on references 18-20, with details
provided in the Supporting Information.
Ethyl
2-(2-(3-(2-iodoethyl)-3H-diazirin-3-
yl)ethyl)cycloprop-2-enecarboxylate (L4).9 A 50-mL two-neck
round bottom flask was charged with L3 (1.48 g, 6.0 mmol,
3.0 equiv.), Rh2(OAc)4 (44 mg, 0.01 mmol, 0.05 equiv.) and
dichloromethane (DCM; 20 mL).
A solution of ethyl
diazoacetate (0.23 g, 2.0 mmol, 1 equiv.) in DCM (5 mL) was
added via a syringe pump in 10 h. The solvent was removed in
vacuo and the residue was purified by flash column
chromatography (10:1 hexane:EtOAc) to afford the desired
1
product L4 as a colorless oil (0.14 g, 21%). H NMR (500
MHz, CDCl3) δ 6.46 (s, 1H), 4.13 (m, 2H), 2.82 (t, J = 7.5,
2H), 2.34 (m, 2H), 2.16 (s, 1H), 2.09 (t, J = 7.5, 2H), 1.77 (m,
2H), 1.25 (t, J = 7.2, 3H); 13C NMR (125 MHz, CDCl3) δ
175.94, 113.74, 96.05, 60.38, 37.20, 29.88, 28.35, 19.96,
19.86, 14.38, -4.16. HR-MS (ESI) calcd for [M+Na]+
357.0076; Found 357.0063.
BD-2. To a stirred solution of BD-1 (10 mg, 0.019 mmol) in
2 mL toluene was added 0.1 mL of 1.0 M DIBAL (in Hexane,
0.1 mmol, 5 equiv.) at -78 oC. The mixture was stirred for 2 h
and subsequently quenched by addition of 1 N HCl (2 mL).
Upon extraction with EtOAc (2 × 10 mL), the combined or-
ganic layers were dried over Na2SO4 and concentrated in vac-
uo, and the resulting residue was purified by flash column
(50:1 to 20:1 DCM:MeOH), affording the desired product BD-
(2-(2-(3-(2-Iodoethyl)-3H-diazirin-3-
yl)ethyl)cycloallyl)methanol (L7). To a stirred solution of L4
(33.4 mg. 0.1 mmol) in 3 mL toluene was added 0.5 mL of 1.0
o
M DIBAL (0.5 mmol, 5 equiv.) at -78 C. The mixture was
stirred for 1 h and quenched by addition of 1 N HCl (3 mL).
Upon extraction by 2 × 10 mL of EtOAc , the organic phase
was concentrated in vacuo and purified by flash column
chromatography (1:3 EtOAc:hexane) to afford L7 as colorless
1
2 (4 mg, 43%). H NMR (500 MHz, CDCl3) δ 7.73 (m, 2H),
7.54 (m, 5H), 7.39 (m, 2H), 7.16 (br, 1H), 6,72 (m, 1H), 4.63
(m, 1H), 3.75 (m, 1H), 3.68 (m, 1H), 3.48 (m, 1H), 3.35 (m,
1H), 3.28 (m, 1H), 3.10 (m, 1H), (d, J = 7.0 Hz, 1H), 2.67 (s,
3H), 2.37 (m, 3H), 1.71 (m, 2H), 1.67 (m, 2H); 13C NMR (125
MHz, CDCl3) δ 173.23, 139.08, 137.54, 133.45, 131.90,
130.24, 128.74, 127.56, 126.18, 124.14, 120.36, 117.80,
112.90, 103.68, 71.71, 67.98, 54.16, 53.40, 39.27, 38.51,
33.91, 32.09, 27.47, 22.86, 21.10, 14.12, 12.34. HR-MS (ESI)
calcd for [M+Na]+ 518.2280; Found 518.2271.
BD-3. Linker L2 was obtained by following previously re-
ported procedures.9 To a stirred solution of compound S1 (70
mg, 0.22 mmol) in DMF (2 mL) was added HOBt (32 mg,
0.24 mmol), EDCI (46 mg, 0.24 mmol), triethylamine (52 mg,
0.4 mmol) and L2 (30 mg, 0.22 mmol). The mixture was
stirred for 12 h and diluted with water. Subsequently, the mix-
ture was extracted with EtOAc (2 x 10 mL), and the organic
phase was washed with brine (2 x 10 mL). Upon solvent evap-
oration, the resulting residue was purified by flash column
(100:1 to 20:1 DCM:MeOH), affording the desired product
1
oil (24 mg, 85%). H NMR (300 MHz, CDCl3) δ 6.75 (s, 1H),
3.54 (t, J = 7.5, 2H), 2.80 (t, J = 7.5, 2H), 2.31 (t, J = 7.5, 2H),
2.08 (m, 3H), 1.79 (m, 3H); 13C NMR (75 MHz, CDCl3) δ
123.90, 103.66, 68.09, 37.38, 30.02, 28.46, 20.97, 20.69, -
4.07. ESI-MS: m/z calcd for 292.0; Found 292.2.
Ethyl
2-(2-(3-(2-azidoethyl)-3H-diazirin-3-
yl)ethyl)cycloprop-2-enecarboxylate (L5). To a solution of L4
(334 mg, 1.0 mmol) in 10 mL of DMF was added sodium az-
ide (78 mg, 1.2 mmol), and the resulting mixture was stirred at
room temperature for 12 h, then quenched with 10 ml water.
Upon extraction with EtOAc (2 × 10 mL), the organic layers
were washed with saturated aqueous NaCl (10 mL) and dried
with anhydrous Na2SO4. The solvents were removed in vacuo
and the crude product was purified by flash column chroma-
tography (10:1 hexane:EtOAc) to yield L5 (214 mg, 86%) as a
1
colorless oil. H NMR (300 MHz, CDCl3) δ 6.44 (s, 1H), 4.13
(m, 2H), 3.14 (t, J = 7.5, 2H), 2.34 (t, J = 7.5, 2H), 2.16 (s,
1H), 1.74-1.69 (m, 4H), 1.27 (t, J = 7.5, 3H); 13C NMR (75
MHz, CDCl3) δ 176.41, 113.77, 96.81, 60.17, 46.06, 32.13,
30.22, 26.21, 19.85, 19.78, 14.26.
1
BD-3 as a white solid (60 mg, 67%). H NMR (500 MHz,
CDCl3) δ 7.70 (m, 1H), 7.55-7.29 (m, 8H), 6.84 (br, 1H), 4.64
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 10
(t, J = 7.5 Hz, 1H), 3.65 (m, 1H), 3.41 (m, 1H), 3.21 (m, 2H), remove excessive probe, followed by UV irradiation for 20
2.68 (s, 3H), 2.02 (m, 2H), 1.65 (m, 4H); 13C NMR (125 MHz,
CDCl3) δ 170.58, 167.78, 138.72, 133.25, 131.98, 131.55,
130.67, 129.41, 129.29, 128.20, 127.36, 123.34, 82.85, 69.30,
53.71, 39.34, 34.34, 32.60, 31.94, 29.62, 26.67, 13.15, 12.16.
HR-MS (ESI) calcd for [M+Na]+ 474.2018; Found 474.2018.
Isothermal Titration Calorimetry. Experiments were per-
formed on an ITC200 calorimeter (GE Amersham) interfaced
with a computer for data acquisition and analysis using Origin
7 software (Microcal Inc., Northampton, MA). The binding
isotherms were best-fit to a one-set binding-site model by
Marquardt non-linear least-square analysis to obtain binding
stoichiometry (N), association constant (Ka), change in enthal-
py (ΔH), and change in entropy (ΔS). The free energy (ΔG)
and ΔS were calculated by the following equations, respective-
ly: ΔG = -RT lnKa; ΔS = (ΔH-ΔG)/T. Recombinantly purified
(His)6-tagged BRD-4 protein27 was extensively dialysed
against 20 mM HEPES buffer (with 150 mM NaCl, pH 7.4)
and the same buffer was used for reconstituting small mole-
cule ligands. The experiments were carried out by titration of
20 µM of the protein in the sample cell with 400 µM of a lig-
and in the titration syringe. The cell was thermostated at 25 oC
and uniform mixing was ensured by continuous stirring of
titration syringe at 400 rpm. All titration experiments were
performed by addition of 2 µL titrant per injection, with 20
injections spaced at 200-second intervals.
In Vitro and In Situ Protein Labeling. For gel-based pro-
tein/cell lysate labeling experiments, procedures were based
mostly on previously published protocols with some modifica-
tions.9 For labeling of recombinant BRD-4, to 2 µg of the pro-
tein in 50 μL PBS buffer was added different concentrations of
the probe, and the reaction was incubated for 30 min at room
temperature with gentle shaking. Subsequently, the reactions
were UV-irradiated (350 nm) for 20 min followed by conju-
gating with a suitable “click” reporter. For probes BD-3 and
NP-1, rhodamine azide TER-N3 (in 50 μM final concentration
from 1 mM DMSO stock9) under catalysis of CuSO4 (1 mM
final concentration from 100 mM freshly prepared stock solu-
tion in deionized water), TBTA (100 μM final concentration
from 10 mM freshly prepared stock solution in DMSO) and
TCEP (1 mM final concentration from 100 mM freshly pre-
pared stock solution in deionized water) were used. For BD-1,
-2 and NP-2, TER-TZ1 (50 μM final concentration from 1
mM freshly prepared DMSO stock) was used with ligation
time of 2 h (for concentration-dependent labeling experi-
ments), or 1-60 min (for time-dependent labeling). For bacte-
rial lysate labeling, 1 M of the probe and 2 g of the lysate
were used, together with 10 μM of TER-TZ1 (with 2 h te-
trazine-cyclopropene ligation time). Subsequently, 10 μL of 6
× SDS loading dye was added and the mixture was heated at
95 ºC for 10 min. The resulting proteins were separated by
SDS-PAGE. In-gel fluorescence scanning (FL) was used to
visualize the labeled protein bands. Coomassie staining (CBB)
was used to detect the total amount of loaded proteins.
min on ice. The cells were trypsinized and pelleted by centrif-
ugation. Eventually, the cell pellets were resuspended in
HEPES buffer (50 μL), homogenized by sonication, and dilut-
ed to 1 mg/mL with PBS. To 50 μL of the resulting proteome
solution was added TER-TZ1 (50 μM final concentration
from 1 mM freshly prepared DMSO stock). The reaction was
further incubated for 2 h with gentle mixing, before being
terminated by addition of pre-chilled acetone (0.5 mL; 30 min
incubation at −20 °C). Precipitated proteins were subsequently
collected by centrifugation (13000 rpm × 10 min at 4 °C). The
supernatant was discarded and the pellet was washed with 200
μL of prechilled methanol, 2 × loading buffer was added and
heated for 10 min at 95 °C. Around 20 μg (per gel lane) of
proteins were separated by SDS−PAGE and visualized by in-
gel fluorescence scanning. Subsequently, the remaining por-
tion of the probe-labeled proteome (prior to tetrazine ligation
reaction) was added Biotin-TZ2 (50 μM final concentration
from 1 mM freshly prepared DMSO stock). The mixture was
incubated at room temperature for 2 h with gentle mixing,
acetone precipitated, and resolubilized in 1% SDS (in PBS)
with brief sonication. This resuspended sample was then incu-
bated with avidin-agarose beads (100 µL/mg protein) over-
night at 4 ºC. Upon centrifugation, the supernatant was re-
moved and the beads were washed with 0.1% SDS once and
PBS (4 x), then boiled in 1 × SDS loading buffer for 15 min.
Control PD using the negative probe (NP-2) and the reference
probe (BD-3) were carried out concurrently.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Live-Cell Imaging. The experiments were mostly based on
previously published protocols.9,12a,12c HepG2 cells seeded in
glass bottom dishes (Mattek) and grown until 70−80% conflu-
ency were treated with 0.3 mL of DMEM with a probe or NP-
2 at different indicated concentrations, with or without 10 x
(+)-JQ1. After incubation for 2 h, the medium was removed
and cells were gently washed twice with PBS, followed by UV
irradiation (350 nm) for 20 min on ice. Cells were then incu-
bated in DMEM containing FL-TZ1 (50 μM final concentra-
o
tion) for 1-60 min at 37 C, and washed with fresh DMEM
medium (10 min to 2 h) before being imaged. We also tried
the no-wash, direct imaging protocol,12a but it resulted in high-
er fluorescence background. Images shown in Figure 3D were
results obtained from 2 h DMEM wash. In the last 20 min of
incubation, Hochest nuclear stain (1:30000 dilution) was add-
ed to the incubation medium. For immunofluorescence (IF)
experiments, after live-cell imaging, the cells were fixed for 1
h at room temperature with 3.7% formaldehyde in PBS,
washed twice with cold PBS again, and permeabilized with
0.1% Triton X-100 in PBS for 10 min. Cells were then
blocked with 2% BSA in PBS for 30 min, washed twice with
PBS, and further incubated with anti-BRD-4 antibody (1:100
dilution) for 1 h at room temperature, washed twice with PBS,
and then incubated with Cy3-labeled goat anti-rabbit (Invitro-
gen 81-6115, 1:100 dilution) for 1 h, following by washing
again with PBS before imaging. All imaging data were col-
lected on a Leica TCS SP5X confocal microscope system and
images were processed as previously described.12c
For in situ proteome labeling and PD/WB experiments (e.g.
Figure 3E & 3G), either 1 or 5 μM of each probe, with or
without 10 x (+)-JQ1 as the competitor, was used. Briefly,
after cells were grown, the medium was removed, and cells
were washed twice with cold PBS and then treated with 0.5
mL of the DMEM-containing probe (diluted from DMSO
stocks whereby DMSO never exceeded 1% in the final solu-
tion). After 2-4 h of incubation at 37 °C/5% CO2, the medium
was aspirated, and cells were washed gently with 2 x PBS to
Large-Scale Protein Identification by LC-MS/MS. Pull-
down experiments were carried out as above described, sepa-
rated on 10% SDS-PAGE gels, followed by Coomassie stain-
ing and tryptic digestion as previously described.9 Digested
peptides were extracted from the gel, separated on a Shimadzu
UFLC system (Shimadzu, Japan) coupled to an Orbitrap Elite
(Thermo Electron, Germany). The Orbitrap Elite was set to
ACS Paragon Plus Environment

Page 9 of 10
Journal of the American Chemical Society
(6) Park, J.; Koh, M.; Park, S. B. Mol. BioSyst. 2013, 9, 544-550.
perform data acquisition in the positive-ion mode, except that
(7) Yang, P. -Y, Liu, K.; Ngai, M. H.; Lear, M. J.; Wenk, M. R.;
Yao, S. Q. J. Am. Chem. Soc. 2010, 132, 656-666.
(8) Shi, H.; Zhang, C. -J.; Chen, G. Y. J.; Yao, S. Q. J. Am. Chem.
Soc. 2012, 134, 3001-3014.
(9) Li, Z.; Hao, P.; Li, L.; Tan, C.Y. J.; Cheng, X.; Chen, G. Y. J.;
Sze, S. K.; Shen, H. -M.; Yao, S. Q. Angew. Chem. Int. Ed.
2013, 52, 8551-8556.
1
2
3
4
5
6
7
8
the m/z range of 350−1600 was used in the full MS scan. The
raw data were converted to mgf format. The database search
was performed with an in-house Mascot server (version
2.2.07, Matrix Science) with MS tolerance of 10 ppm and
MS/MS tolerance of 0.8 Da. Two missed cleavage sites of
trypsin were allowed. Carbamidome-thylation (C) was set as a
fixed modification, and oxidation (M) and phosphorylation (S,
T, and Y) were set as variable modifications. All LC-MS/MS
samples were injected at least twice, and proteins that ap-
peared in duplicated/triplicated runs (~75% on average) were
further processed. Results obtained from the above experi-
ments (in situ, and negative probe) were processed as shown
below, and complete MS results are summarized in the Sup-
porting Information (SI_2). Finally, we ranked the protein hits
by their protein scores and emPAI values. The Exponentially
Modified Protein Abundance Index (emPAI) offers approxi-
mate, label-free, relative quantitation of proteins in a mixture
based on protein coverage by the peptide matches in a data-
base search result.28 Selected MS results are summarized in
Tables 1, S2 & S3. “False” hits that appeared in negative con-
trol PD/LC-MS/MS experiments were eliminated. For runs
obtained with 10 x (+)-JQ1, proteins were deemed potential
hits if their protein scores were significantly lowered (or did
not appear) in the presence of the competitor. Subcellular lo-
calizations of potential hits were determined by checking indi-
vidual proteins against www.genecards.org.
(10) Kolb, H. C.; Sharpless, K. B. Drug Discovery Today 2003, 8,
1128-1137.
(11) For a recent review, see Dubinsky, L.; Krom, B. P.; Meijler, M.
M. Bioorg. Med. Chem. 2012, 20, 554-570, and references cited
therein.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(12) (a) Yang, K. S.; Budin, G.; Reiner, T.; Vinegoni, C.; Weissle-
der, R. Angew. Chem. Int. Ed. 2012, 51, 6598-6603. (b) Budin,
G.; Yang, K. S.; Reiner, T.; Weissleder, R. Angew. Chem. Int.
Ed. 2011, 50, 9378-9381. (c) Li, L.; Shen, X.; Xu, Q. -H.; Yao,
S. Q. Angew. Chem. Int. Ed. 2013, 52, 424-428. (d) Li, L.; Ge,
J.; Wu, H.; Xu, Q.- H.; Yao, S. Q. J. Am. Chem. Soc. 2012, 134,
12157-12167. (e) Hu, M.; Li, L.; Wu, H.; Su, Y.; Yang, P. -Y.;
Uttamchandani, M.; Xu, Q. -H.; Yao, S. Q. J. Am. Chem. Soc.
2011, 133, 12009-12020. (f) Edgington, L. E.; Berger, A. B.;
Blum, G.; Albrow, V. E.; Paulick, M. G.; Lineberry, N.; Bogyo,
M. Nat. Med. 2009, 15, 967-973. (g) Pratt, M. R.; Sekedat, M.
D.; Chiang, K. P.; Muir, T. W. Chem. Biol. 2009, 16, 1001-
1012. (h) Watzke, A.; Kosec, G.; Kindermann, M.; Jeske, V.;
Nestler, H. P.; Turk, V.; Turk, B.; Wendt, K. U. Angew. Chem.
Int. Ed. 2008, 47, 406-409.
(13) Yang, K. S.; Budin, G.; Tassa, C.; Kister, O.; Weissleder, R.
Angew. Chem. Int. Ed. 2013, 52, 10593-10597.
Preliminary Target Validation. The pull-down samples
were prepared as above described, separated by 10% SDS gel
and transferred to poly(vinylidene difluoride) membranes. The
WB procedures were carried out as above described, with the
respective antibodies (1:5000 and 1:2000 dilutions for DDB1
and RAD23B, respectively).
(14) (a) Kwan, D. H.; Chen, H. -M.; Ratananikom, K.; Hancock, S.
M.; Watanabe, Y.; Kongsaeree, P. T.; Samuels, A. L.; Withers,
S. G. Angew. Chem., Int. Ed. 2011, 50, 300–303. (b) Ge, J.; Li,
L.; Yao, S. Q. Chem. Commun. 2011, 47, 10939-10941.
(15) Sletten, E. M.; Bertozzi, C. R. Angew. Chem. Int. Ed. 2009, 48,
6974-6998.
(16) Devaraj, N. K.; Weissleder, R. Acc. Chem. Res. 2011, 44, 816-
827.
(17) Selvaraj, R.; Fox, J. M. Curr. Opin. Chem. Biol. 2013, 17, 753-
760.
(18) Yang, J.; Šečkutė, J.; Cole, C. M.; Devaraj, N. K. Angew. Chem.
Int. Ed. 2012, 51, 7476-7479.
Supporting Information. Other relevant experimental sections,
characterization of new compounds, supplementary chemical,
biological and LC-MS/MS results. This material is available free
of charge via the Internet at http://pubs.acs.org.
(19) Patterson, D. M.; Nazarova, L. A.; Xie, B.; Kamber, D. N.; Pre-
sche, J. A. J. Am. Chem. Soc. 2012, 134, 18638-18643.
(20) Kamber, D. N.; Nazarova, L. A.; Liang, Y.; Lopez, S. A.; Pat-
terson, D. M.; Shih, H. -W.; Houk, K. N.; Prescher, J. A. J. Am.
Chem. Soc., 2013, 37, 13680-13683.
(21) Yu, Z.; Pan, Y.; Wang, Z.; Wang, J.; Lin, Q. Angew. Chem. Int.
Ed. 2012, 51, 10600-10604.
(22) Yan, N.; Liu, X.; Pallerla, M.; Fox, J. M. J. Org. Chem. 2008,
73, 4283-4286.
chmyaosq@nus.edu.sg
(23) Berger, S. L. Nature 2007, 447, 407-412.
(24) (a) Filippakopoulos, P.; Qi, J.; Picaud, S.; Shen, Y.; Smith, W.
B.; Fedorov, O.; Morse, E. M.; Keates, T.; Kickman, T. T.;
Felletar, I.; Philpott, M.; Munro, S.; McKeown, M. R.; Wang,
Y.; Christie, A. L.; West, N.; Cameron, M. J.; Schwartz, B.;
Heightman, T. D.; Thangue, N. L.; French, C. A.; Wiest, O.;
Kung, A. L.; Knapp, S.; Brandner, J. E. Nature 2010, 468,
1067-1073. (b) Nicodeme, E.; Jeffrey, K. L.; Schaefer, U.;
Beinke, S.; Dewell, S.; Chung, C. –W.; Chandwani, R.; Maraz-
zi, I.; Wilson, P.; Coste, H.; White, J.; Kirilovsky, J.; Rice, C.
M.; Lora, J. M.; Prinjha, R. K.; Lee, K.; Tarakhovsky, A. Na-
ture 2010, 468, 1119-1123.
(25) (a) Delmore, J. E.; Issa, G. C.; Lemieux, M. E.; Peter B. Rahl, P.
B.; Shi, J.; Jacobs, H. M.; Kastritis, E.; Gilpatrick, T.; Paranal,
R. M.; Qi, J.; Chesi, M.; Schinzel, A. C.; McKeown, M. R.;
Heffernan, T. P.; Vakoc, C. R.; Bergsagel, P. L.; Ghobrial, I.
M.; Richardson, P. G.; Young, R. A.; Hahn, W. C.; Anderson,
K. C.; Kung, A. L.; Bradner, J. E.; Mitsiades, C. S. Cell 2011,
146, 904-917. (b) Floyd, S. R.; Pacold, M. E.; Huang, Q.;
Funding was provided by the National Medical Research Council
(CBRG/0038/2013) and Ministry of Education (MOE2012-T2-1-
116/MOE2012-T2-2-051). We acknowledged contributions from
Dr. Chongjing Zhang for his initial work, and Ramya Chandra-
sekaran for expression of recombinant BRD-4 protein.
(1) Ziegler, S.; Pries, V.; Hedberg, C.; Waldmann, H. Angew.
Chem. Int. Ed. 2013, 52, 2744-2792.
(2) Simon, G. M.; Niphakis, M. J.; Cravatt, B. F. Nat. Chem. Biol.
2013, 9, 200-205.
(3) Su, Y.; Ge, J.; Zhu, B.; Zheng, Y. -G.; Zhu, Q.; Yao, S. Q. Curr.
Opin. Chem. Biol. 2013, 17, 768-775.
(4) Cravatt, B. F.; Wright, A. T.; Kozarich, J. W. Annu. Rev. Bio-
chem. 2008, 77, 383-414.
(5) Lee, J.; Bogyo, M. Curr. Opin. Chem. Biol. 2013, 17, 118-126.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 10
Clarke, S. M.; Lam, F. C.; Cannell, I. G.; Bryson, B. D.; Ram-
Clement, C. A.; Boullay, A.-B.; Grimley, R. L.; Blandel, F. M.;
eseder, J. R.; Lee, M. J.; Blake, E. J.; Fydrych, A.; Ho, R.;
Greenberger, B. A.; Chen, G. C.; Maffa, A.; Del Rosario, A. M.;
Root, D. E.; Carpenter, A. E.; Hahn, W. C.; Sabatini, D. M.;
Chen, C. C.; White, F. M.; Bradner, J. E.; Yaffe, M. B. Nature
2013, 498, 246-250.
Prinjha, R. K.; Lee, K.; Kirilovsky, J.; Nicodeme, E. J. Med.
Chem. 2011, 54, 3827-3838.
(27) Zhang, C.; Tan, C. Y. J.; Na, Z.; Ge, J.; Chen, G. Y. J.; Uttam-
chandani, M.; Sun, H.; Yao, S. Q. Angew. Chem. Int. Ed., 2013,
52, 14060-14064.
(28) (a) Ishihama, Y.; Oda, Y.; Tabata, T.; Sato, T.; Nagasu, T.;
Rappsilber, J.; Mann, M. Mol. Cell. Proteomics 2005, 4, 1265-
1272. (b) Hao, P.; Guo, T.; Li, X.; Adav, S. S.; Yang, J.; Wei,
M.; Sze, S. K. J. Proteome Res. 2010, 9, 3520−352.
1
2
3
4
5
6
7
8
(26) Chung, C. -W.; Coste, H.; White, J. H.; Mirguet, O.; Wilde, J.;
Gosmini, R. L.; Delves, C.; Magny, S. M.; Woodward, R.;
Hughes, S. A.; Boursier, E. V.; Flynn, H.; Bouillot, A. M.;
Bamborough, P.; Brusq, J. -M. G.; Gellibert, F. J.; Jones, E. J.;
Riou, A. M.; Homes, P.; Martin, S. L.; Uings, I. J.; Toum, J.;
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment