Minimalist linkers suitable for irreversible inhibitors in simultaneous proteome profiling, live-cell imaging and drug screening

Article abstract of DOI:10.1039/c8cc08685k

Activity-based protein profiling (ABPP) and bioimaging have been powerful approaches for in situ drug screening and target identification. However, these approaches are still hindered by the preparation of high-quality probes. To address this challenge, we developed a series of novel minimalist linkers for irreversible inhibitors by incorporation of various bioorthogonal handles into an α,β-unsaturated amide, a common moiety of many irreversible inhibitors. The linker-containing probes have been demonstrated to be suitable for simultaneous protein labelling, live cell imaging and drug screening.

Full text of DOI:10.1039/c8cc08685k

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Minimalist linkers suitable for irreversible
inhibitors in simultaneous proteome profiling,
live-cell imaging and drug screening†
Cite this: Chem. Commun., 2019,
55, 834
Received 31st October 2018,
Accepted 12th December 2018
Cuiping Guo,‡a Yu Chang,‡a Xin Wang,a Chengqian Zhang,b Piliang Hao,*b
a
Ke Dinga and Zhengqiu Li
*
DOI: 10.1039/c8cc08685k
rsc.li/chemcomm
Activity-based protein profiling (ABPP) and bioimaging have been
powerful approaches for in situ drug screening and target identifi-
cation. However, these approaches are still hindered by the prepara-
tion of high-quality probes. To address this challenge, we developed a
series of novel minimalist linkers for irreversible inhibitors by incor-
poration of various bioorthogonal handles into an a,b-unsaturated
amide, a common moiety of many irreversible inhibitors. The linker-
containing probes have been demonstrated to be suitable for simul-
taneous protein labelling, live cell imaging and drug screening.
Fig. 1 Structures of minimalist linkers for (A) reversible inhibitors and
(B) irreversible inhibitors, respectively.
Drug screening and target identification are critical steps in
drug discovery, but they rely mainly on traditional in vitro
enzymatic assays, which often must first be developed.1 To
solve this problem, activity-based protein profiling (ABPP) and bioimaging studies (Fig. 1A).4 Owing to the improved synthesis
bioimaging have been developed recently as powerful and of photoprobes, which enable simultaneous protein labeling
complementary approaches that can be applied for both target and cellular imaging, they have been broadly applied in various
identification and drug screening in native environments.2
A
small molecules for target identification.5 In this work, we
large number of lead compounds and potential druggable endeavored to create a series of minimalist linkers for irreversible
targets have been identified singly or collectively through these inhibitors (Fig. 1B), with the aim to facilitate the synthesis of
approaches.3 However, both approaches are still hindered by activity-based probes and enable simultaneous imaging of endo-
the lack of high quality probes capable of recapitulating genu- genous kinase activities, studying target engagement in live cells,
ine drug–target engagement. This can be accounted for by and drug screening against druggable targets such as EGFR
different elements such as bioorthogonal handles, fluoro- and BTK.
phores and/or photo-crosslinkers, which must be embedded
With this goal in mind, we examined the structures of
in the parent molecules with minimal perturbation of the various irreversible inhibitors and observed that most of them,
original binding sites. We recently developed several types of especially kinase inhibitors, possess an a,b-unsaturated amide.
photo-crosslinkers with bioorthogonal handles for reversible Importantly, this common moiety is tolerable for modification
inhibitors in affinity-based proteome profiling (AfBP) and based on previously reported structure–activity relationships
(SAR)6 and the co-crystal structures of molecules with proteins
(Fig. S1, ESI†).7 Thus, we postulated that incorporation of bio-
orthogonal handles, especially copper-free ones such as azide,
a School of Pharmacy, Jinan University, Guangzhou City Key Laboratory of Precision
Chemical Drug Development, International Cooperative Laboratory of Traditional
cyclopropene, or trans-cyclooctene (TCO), into an a,b-unsaturated
carboxylic acid could produce minimalist-size linkers (Fig. 1B),
which could confer better bioactivities and dual functions on the
corresponding probes, thus supporting proteome profiling and
live-cell imaging.
Chinese Medicine Modernization and Innovative Drug Development, Ministry of
Education (MOE) of People’s Republic of China, 601 Huangpu Avenue West,
Guangzhou, 510632, China. E-mail: pharmlzq@jnu.edu.cn
b School of Life Science and Technology, ShanghaiTech University, 393 Middle
Huaxia Road, Shanghai 201210, China. E-mail: haopl@shanghaitech.edu.cn
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c8cc08685k
To prove this idea, we incorporated linkers into representa-
tive irreversible inhibitors, afatinib and ibrutinib, to assess
‡ Cuiping Guo and Yu Chang contributed equally to this work.
834 | Chem. Commun., 2019, 55, 834--837
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Communication
ChemComm
Fig. 2 (A) Chemical structures of the activity-based probes (ABPs) and parent inhibitors. (B) IC50 values of the probes against recombinant proteins and
cancer cells with the parent inhibitors as positive controls.
whether they could display desired functions. The introduction of as low as 1 nM, showing that the probes exhibit excellent sensitivity
linkers for each molecule was based on previously reported struc- (Fig. S3, ESI†). Consistent with the inhibition assay results, the
ture–activity relationships.6 Consequently, AF-1–2 and IB-1–3 were IB-1-treated samples failed to show the corresponding labeling band
readily produced by coupling of commercially available inter- (Fig. 3B and C). The major labeling bands were proven to be EGFR
mediates with the linkers (Fig. 2A and Schemes S1–S5, ESI†), and and BTK, respectively, by pull-down/western blotting with the
were fully characterized prior to biological evaluation.
corresponding antibodies (Fig. 3D).
With these probes, we first evaluated their biological activities in
To assess whether the probes could behave as imaging probes to
comparison with those of their parental inhibitors using standard track target location, cellular imaging of live A431 and Toledo cells
in vitro kinase inhibition assays and cell-based, CCK-8 antiprolifera- with AF-1–2 and IB-2–3, respectively, was carried out. The cells were
tion assays. As shown in Fig. 2B and Fig. S2 (ESI†), AF-1–2 and first treated with the probes for 2–4 h. Subsequently, the cells treated
IB-2–3 showed comparable inhibition to the corresponding parent with AF-1 were fixed, permeabilized and reacted with TAMRA-N3
inhibitors under both settings, suggesting that the introduction of under previously optimized click chemistry conditions.4 The cells
these bioorthogonal handles had little effect on protein binding, treated with AF-2 and IB-2–3 were incubated with dibenzocyclooctyne-
under in vitro and cellular conditions. In contrast, the probe TAMRA (DBCO-TAMRA) or tetrazine-Cy5 and then imaged directly.
containing cyclopropene (IB-1) appeared to be significantly less Strong fluorescence signals were observed in the cell membrane or
potent than the parent ibrutinib, implying that cyclopropene is not the cytoplasm (Fig. 3E). Control imaging experiments with DMSO
a suitable substitute for the a,b-unsaturated amide electrophile. under similar conditions showed minimal background fluorescence
This phenomenon could be accounted for by lacking a nitrogen when compared to the labeled cells. Immunofluorescence experi-
atom in the cyclopropene moiety, as the nitrogen atom is essential ments revealed that the probes were largely co-localized with the
for Michael addition by deprotonating the attacking thiol group.
target proteins, EGFR and BTK, respectively (Fig. 3E). These lines of
Next, we determined if the newly developed chemical probes evidence proved that the probes are suitable for simultaneous
could be used in simultaneous protein labeling and bioimaging. proteome profiling and live-cell imaging.
After incubation of AF-1–2 and IB-1–3 with the corresponding kinase
Finally, we proceeded to identify potential cellular targets of
positive cells (A431, Toledo/Raji) for 2–4 h, respectively, the cells afatinib by large-scale chemoproteomics experiments with AF-1–2.
were lysed. The resulting cell lysates were conjugated with the As described above, the probe-labeled proteins were affinity-purified
corresponding reporters (TAMRA-alkyne, TAMRA-N3 or tetrazine- and identified by LC-MS/MS analysis. Control experiments were
Cy5, Table S2, ESI†), and separated by SDS-PAGE, followed by in-gel carried out concurrently with afatinib-treated samples, which were
fluorescence scanning. As shown in Fig. 3A–C and Fig. S3 (ESI†), the used to distinguish real targets from background labeling. A sub-
samples treated with AF-1–2 and IB-2–3 showed a highly selective micromolar probe concentration (0.1 mM) was used to simulate drug
probe labeling profile, with strong fluorescence labeled bands action in situ. The identified protein hits were optimised with label
predominantly at B170 kDa and B70 kDa, respectively, indicating free quantification (LFQ), and the proteins that appeared in dupli-
excellent target selectivity. In the presence of a ten-fold excess of the cate runs and whose LFQ intensity ratios from the probe-treated and
parent molecules as competitors, the major fluorescence bands competitive labeling experiments (AF-1 vs. [AF-1 + afatinib(5Â)], or
disappeared (Fig. 3A–C and Fig. S3, ESI†), indicating that they were AF-2 vs. [AF-2 + afatinib(5Â)]) were greater than 1.5 were considered
probe-targeted proteins rather than the results of non-specific further. Finally, 48 and 32 protein hits were positively identified by
labeling. Concentration-dependent labeling experiments proved AF-1 and AF-2, respectively, and B38% of these proteins were
that the specific labeling band is visible at a probe concentration identical (12 of 32, Fig. 3F, G and Table S3, ESI†). The difference
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 834--837 | 835

View Article Online
ChemComm
Communication
Fig. 3 (A–C) Proteome reactivity profiles of live A431 cells treated with AF-1–2, Toledo cells treated with IB-1–3 and Raji cells treated with IB-1–3, in the
presence or absence of competitors. (D) Pull-down/WB results for target validation of AF-1–2 and IB-2–3 in live cells (in situ). (E) Live-cell imaging of
A431 cells with AF-1–2 (1 mM probe concentration), and Toledo cells with IB-2–3 (1 mM). Immunofluorescence (IF) staining using anti-EGFR and anti-BTK
antibodies. Scale bar = 10 mm. (F) MS-based identification of proteins enriched by probes AF-1–2, ABPP experiments in A431 cells treated with the probes
(AF-1–2; 0.1 mM, 4 h) in the presence or absence of afatinib; data represent the mean LFQ ratio value for the proteins from duplicate runs. The probe-
enriched targets are defined as those with mean LFQ ratio values Z1.5 (probe/(probe + competitor)). (G) Selected high-confidence proteins enriched
with AF-1–2 (0.1 mM) from live A431 cells simultaneously.
could be accounted for by different linkers and the intrinsic Fig. S4 (ESI†), three compounds, 36 (U73122), 42 and 43, were
variability of the instruments. Interestingly, the overlapping proteins identified by screening, which decrease the probe labeling of BTK
include the known target (EGFR) and a series of unknown protein (* marked band), implying that they are potent inhibitors in situ.
targets, such as RAB1B, RAB1CI, TOP3A, PYCR1, AHCY and LY6D, Similar phenomena were also observed in competitive imaging
which could be the potential cellular off-targets of afatinib. These experiments (Fig. 4B). Further validation of the inhibitory proper-
protein hits might be the reason for the anticancer and toxic effects ties revealed IC50 values of 4.14 mM, 1.87 nM, 0.13 mM for the three
of afatinib. These data together proved that the probes AF-1/AF-2 are compounds against BTK, respectively (Fig. 4C). The cell-based
suitable to be applied in target identification of thiol-reactive assay also confirmed the potent inhibition of these screening hits
inhibitors by chemoproteomics studies.
against cancer cells (Fig. S2, ESI†). The effects of the positive
Observing that AF-1/AF-2 and IB-2/IB-3 specifically label and screening hits on the BTK signaling pathway also confirmed that
image the target proteins, EGFR and BTK, in situ, we envisioned these compounds can dose-dependently suppress the phosphory-
that competitive labeling and competitive imaging could be lation of BTK-downstream proteins (Fig. 4D). It is noteworthy
developed as general methods for in situ drug screening against that U73122 is a potent phospholipase C (PLC) inhibitor,8 which
these druggable targets. To test this idea, IB-3 was used as a tool reduces agonist-induced Ca2+ increase in platelets and poly-
probe to test the inhibition of members of a B210-membered morphonuclear neutrophils (PMNs), and little information is
compound library against BTK with Toledo cells. The compound available to date concerning its inhibitory action against BTK.
library consists of commercially available natural products and Compounds 42 and 43 are specific inhibitors against EGFRT790M
,
synthetic small molecules. Upon incubation of IB-3 with or with- which were previously developed by our group.9 These results
out the screening inhibitors for 4 h, the cells were lysed and show that BTK is an off-target of these inhibitors. Taken together,
conjugated with tetrazine-Cy5, and the labeled proteomes were these data show that the minimalist-linker containing probes
separated by SDS-PAGE, and the fluorescence intensity of the BTK can be used in simultaneous protein labeling, live cell imaging
band exhibited by IB-3 was measured. As shown in Fig. 4A and and in situ drug screening.
836 | Chem. Commun., 2019, 55, 834--837
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Communication
ChemComm
Fig. 4 (A) In situ drug screening by competitive labeling with IB-3 (1 mM), lane 1 is a DMSO control, lane 2 is a positive control, and lanes 3–6 are treated
with screening compounds (10 mM final). (B) Competitive imaging of IB-3 (1 mM) in the presence or absence of screening compounds (10 mM final).
(C) IC50 values of the positive screening hits against BTK. (D) The effects of the positive screening hits on the phosphorylation of BTK-downstream
proteins. (E) Chemical structures of the positive screening hits.
656–666; (d) H. Shi, C. Zhang, G. Y. J. Chen and S. Q. Yao, J. Am. Chem.
Soc., 2012, 134, 3001–3014.
We have developed a series of minimalist linkers for irreversible
inhibitors. Upon introduction of the linkers to afatinib and ibrutinib,
the resulting probes exhibited similar bioactivities to those of the
parent compounds, and were shown to be suitable for simultaneous
proteome profiling and live-cell imaging. Moreover, competitive
labeling and competitive imaging can be general methods for
in situ drug screening. We expect that these linkers, especially
the azide- and TCO-containing linkers (L8/10), could be widely
used in various irreversible inhibitors for drug screening and
target identification.
We thank the National Natural Science Foundation of China
(21602079, 21877050), the Science and Technology Program of
Guangdong Province (2017A050506028), the Science and Technology
Program of Guangzhou (201704030060, 201805010007) and
the China Postdoctoral Science Foundation (2016M602611)
for their financial support. We also thank Prof. Shao Q. Yao
(NUS, Singapore) for the invaluable suggestions on this work.
3 (a) D. Medina-Cleghorn and D. K. Nomura, Chem. Biol., 2014, 21,
1171–1184; (b) C. G. Parker, A. Galmozzi, Y. Wang, B. E. Correia,
K. Sasaki, C. M. Joslyn, A. S. Kim, C. L. Cavallaro, R. M. Lawrence,
S. R. Johnson, I. Narvaiza, E. Saez and B. F. Cravatt, Cell, 2017, 168,
527–541; (c) K. Lee, H. S. Ban, R. Naik and Y. S. Hong, Angew. Chem.,
Int. Ed., 2013, 52, 10286–10289.
4 (a) H. Guo and Z. Li, MedChemComm, 2017, 8, 1585–1591;
(b) K. Cheng, J. S. Lee, P. Hao, S. Q. Yao, K. Ding and Z. Li, Angew.
Chem., Int. Ed., 2017, 56, 15044–15048; (c) Z. Li, P. Hao, L. Li,
C. Y. J. Tan, X. Cheng, G. Y. J. Chen, S. K. Sze, H.-M. Shen and
S. Q. Yao, Angew. Chem., Int. Ed., 2013, 52, 8551–8556; (d) Z. Li,
D. Wang, L. Li, S. Pan, Z. Na, C. Y. J. Tan and S. Q. Yao, J. Am. Chem.
Soc., 2014, 136, 9990–9998; (e) Z. Li, L. Qian, L. Li, J. C. Bernhammer,
H. V. Huynh, J. S. Lee and S. Q. Yao, Angew. Chem., Int. Ed., 2016, 55,
2002–2006.
5 (a) M. Soethoudt, S. C. Stolze, M. V. Westphal, L. van Stralen,
A. Martella, E. J. van Rooden and W. Guba, J. Am. Chem. Soc., 2018,
140, 6067–6075; (b) H. Guo, J. Xu, P. Hao, K. Ding and Z. Li, Chem.
Commun., 2017, 53, 9620–9623; (c) D. Zhu, H. Guo, Y. Chang, Y. Ni,
L. Li, Z.-M. Zhang, P. Hao, Y. Xu, K. Ding and Z. Li, Angew. Chem.,
Int. Ed., 2018, 57, 9284–9289; (d) B. Zheng, H. Guo, J. Xu, N. Ma, Y. Ni,
L. Li, P. Hao, K. Ding and Z. Li, Chem. – Asian. J, 2018, 13, 2601–2605.
6 (a) L. A. Honigberg, A. M. Smith, M. Sirisawad, E. Verner, D. Loury,
B. Chang, S. Li, Z. Pan, D. H. Thamm, R. A. Miller and J. J. Buggy,
Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 13075–13080; (b) J. B. Smaill,
G. W. Rewcastle, J. A. Loo, K. D. Greis, O. H. Chan, E. L. Reyner,
H. D. Showalter, P. W. Vincent, W. L. Elliott and W. A. Denny, J. Med.
Chem., 2000, 43, 1380–1397.
Conflicts of interest
There are no conflicts to declare.
7 (a) A. T. Bender, A. S. Gardberg, A. Pereira, T. Johnson, Y. Wu,
R. Grenningloh, J. Head, F. Morandi, P. Haselmayer and L. Liu-Bujalski,
Mol. Pharmacol., 2017, 91, 208–219; (b) F. Solca, G. Dahl, A. Zoephel,
G. Bader, M. Klein, C. O. Kraemer, F. Himmelsbach and E. G. R. Adolf,
J. Pharmacol. Exp. Ther., 2012, 343, 342–350.
References
1 (a) J. G. Moffat, J. Rudolph and D. Bailey, Nat. Rev. Drug Discovery,
2014, 13, 588–602; (b) M. Schenone, V. Dancık, B. K. Wagner and P. A.
Clemons, Nat. Chem. Biol., 2013, 9, 232–240.
ˇ´
˜
2 (a) D. S. Tyler, J. Vappiani, T. Caneque, E. Y. N. Lam, A. Ward, 8 Y. Pang, K. Wang, Y. Wang, Z. Chenlin, W. Lei and Y. Zhang,
O. Gilan, Y. C. Chan, A. Hienzsch, A. Rutkowska and T. Werner, et al., Chem.-Biol. Interact., 2018, 292, 110–120.
Science, 2017, 356, 1397–1401; (b) M. A. Miller and R. Weissleder, Nat. 9 S. Chang, L. Zhang, S. Xu, J. Luo, X. Lu, Z. Zhang, T. Xu, Y. Liu, Z. Tu,
Rev. Cancer, 2017, 17, 399–414; (c) P.-Y. Yang, K. Liu, M. H. Ngai,
M. J. Lear, M. Wenk and S. Q. Yao, J. Am. Chem. Soc., 2010, 132,
Y. Xu, X. Ren, M. Geng, J. Ding, D. Pei and K. Ding, J. Med. Chem.,
2012, 55, 2711–2723.
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 834--837 | 837