Environment-sensitive turn-on fluorescent probes for p53-MDM2 protein-protein interaction

Article abstract of DOI:10.1039/c7md00287d

A series of probes with a turn-on switch for the p53-MDM2 protein-protein interaction were developed. After careful evaluation, these small molecule fluorescent probes exhibited high practical activity and selectivity in vitro and in cellulo. In particular probe 10, which had a K_i value of 0.03 μM, displayed much better binding affinity compared to the positive control Nutlin-3, which had a K_i value of 0.23 μM. These no-wash environment-sensitive turn-on fluorescent probes have been successfully applied to imaging p53-MDM2 interaction in the human lung cancer cell line A549 (wild-type p53) at the micromolar level. Therefore, these fluorescent probes are expected to be used in drug screening and cell staining in p53-MDM2 fields, as well as in pathological and physiological studies of the p53-MDM2 interaction.

Full text of DOI:10.1039/c7md00287d

MedChemComm
View Article Online
View Journal
RESEARCH ARTICLE
Environment-sensitive turn-on fluorescent probes
for p53–MDM2 protein–protein interaction†‡
Cite this: DOI: 10.1039/c7md00287d
Tingting Liu,§a Yan Jiang,§b Zhenzhen Liu,a Jin Li,b Kun Fang,b Chunlin Zhuang,b
a
a
a
Lupei Du, Hao Fang, Chunquan Sheng b and Minyong Li
*
*
A series of probes with a turn-on switch for the p53–MDM2 protein–protein interaction were developed.
After careful evaluation, these small molecule fluorescent probes exhibited high practical activity and selec-
tivity in vitro and in cellulo. In particular probe 10, which had a Ki value of 0.03 μM, displayed much better
binding affinity compared to the positive control Nutlin-3, which had a Ki value of 0.23 μM. These no-wash
environment-sensitive turn-on fluorescent probes have been successfully applied to imaging p53–MDM2
interaction in the human lung cancer cell line A549 (wild-type p53) at the micromolar level. Therefore,
these fluorescent probes are expected to be used in drug screening and cell staining in p53–MDM2 fields,
as well as in pathological and physiological studies of the p53–MDM2 interaction.
Received 2nd June 2017,
Accepted 30th June 2017
DOI: 10.1039/c7md00287d
rsc.li/medchemcomm
type p53 and mdm2 genes have been found to be
overexpressed or amplified, which could down-regulate the
Introduction
As a tumor suppressor, p53 is a mutated gene in human tu-
mor cells that plays a pivotal role in preventing cancer devel-
opment caused by cellular stresses such as oncogenic activa-
tion, DNA damage, hypoxia, and telomere erosion.1,2 It needs
to be emphasized that the p53 protein, which includes several
domains, can regulate the expression of numerous genes with
different biofunctions, such as apoptosis, cell cycle regulation
and differentiation.3 The key role of p53 in the cell demon-
strates that the loss of p53 may lead to dramatic conse-
quences. Normally, the level of p53 is tightly controlled by
two related proteins called MDMX (also known as MDM4)4
and MDM2 (sometimes called HDM2 for its human analog)5–7
through a negative feedback loop.5 It should be noted that
p53 can activate the expression of MDM2 protein so as to in-
crease the MDM2 level, which in turn inhibits p53 through
three mechanisms, including blocking p53 transcription activ-
ity by binding to p53 at the transactivation domain, promot-
ing the nuclear export of p53 by exposing its signal sequence,
and stimulating p53 degradation6,8,9 since MDM2 can serve
as a ubiquitin ligase. In many human malignancies the wild-
p53 protein and impair its tumor suppression activity.10,11
Therefore, stabilization and activation of the p53 pathway by
inhibition of MDM2 has been proposed as a novel therapeutic
approach for cancer therapy.12,13
Historically speaking, it is challenging to develop effective
small-molecule inhibitors for nonenzymatic protein–protein
interactions. Fortunately, the crystal structural basis of the
p53–MDM2 protein–protein interaction was solved by X-ray
crystallography in 1996.14 It was revealed that MDM2 has a
relatively deep hydrophobic cleft which can be filled by three
hydrophobic residues, Phe19, Trp23, and Leu26, from an am-
phipathic α-helix of the peptide in the transactivation do-
main of p53.14 Based on such a well-defined pocket of the
crystal structure, compounds with low molecular weights will
be discovered that could block the interaction of p53–MDM2.
Over the past few years, a number of potent and selective
molecules, such as RO5503781 and RG7112,15 have been
established as p53–MDM2 inhibitors with effective antitumor
activity in vitro and in vivo.
There are many kinds of well-developed techniques, such
as autofluorescent translocation biosensor systems16,17 and
bimolecular fluorescence complementation (BiFC) assays,18
to image and detect the interaction between p53 and MDM2.
Fluorescence techniques have the comparative advantages of
visualizing the protein interaction directly and providing tem-
poral, spatial information in living and intact cells over any
other methods used in perfect picture imaging system (PPIs)
study, such as electrophoresis-based systems, the two-hybrid
system, immune affinity-based methods and mass spectrom-
etry (MS).16,19,20 Many techniques such as autofluorescent
a Department of Medicinal Chemistry, Key Laboratory of Chemical Biology (MOE),
School of Pharmacy, Shandong University, Jinan, Shandong 250012, China.
E-mail: mli@sdu.edu.cn
b Department of Medicinal Chemistry, School of Pharmacy, Second Military
Medical University, 325 Guohe Road, Shanghai 200433, China.
E-mail: shengcq@hotmail.com
† The authors declare no competing interests.
‡ Electronic supplementary information (ESI) available. See DOI: 10.1039/
c7md00287d
§ These authors contributed equally to this work.
This journal is © The Royal Society of Chemistry 2017
Med. Chem. Commun.

View Article Online
Research Article
MedChemComm
translocation biosensors and BiFC techniques often need a
conjugated fluorescent protein as a reporter, which is time-
consuming, complicated and expensive. Therefore, finding
more economical and straightforward methods to image and
detect the interaction between p53 and MDM2 is urgent.
In recent years, small-molecule fluorescent probes have
become very popular in imaging and recognizing proteins of
interest. Compared to the above-mentioned techniques, their
preparations are more convenient and affordable. Further-
more, these small-molecule fluorescent probes with turn-on
mechanisms display more advantages21–24 in which they can
image and detect specific proteins effectively as they allow
for sensitive and specific detection under high signal-to-
background ratios.25 Currently, the development of small-
molecule fluorescent probes for detecting the interactions of
non-enzymatic proteins such as p53 and MDM2 remains a
challenging task.
In our previous studies, we attempted to find small-
molecule fluorescent probes for imaging the p53–MDM2
interaction and successfully discovered an effective fluores-
cent probe.26 However, its fluorescence intensity was not
strong, and the binding affinity to MDM2 was relatively weak.
So we tried to find better probes and fortunately some
protein-specific fluorescent turn-on probes with strong fluo-
rescence and blue shifted emission in hydrophobic surround-
ings were reported by the Tan group.27 Based on the previous
work, compound Z1, which is an effective p53–MDM2 inhibi-
tor developed by the Tan group,28 was chosen as the major
scaffold for MDM2 binding and the fluorophore 4-chloro-7-
nitrobenzoxadiazole (NBD) which has reasonable fluorescent
properties was selected as the recognition moiety in an ali-
phatic spacer. Subsequently, a series of small-molecule fluo-
rescent probes for detecting the p53–MDM2 interaction were
designed and synthesized (Scheme 1).
Scheme 2 The synthetic route of small-molecule fluorescent probes.
(i) 120–130 °C, 6 h. (ii) Triethylamine, r.t., 30 min; dimethylamine hydro-
chloride, THF, 60 min, 0 °C, 43%. (iii) Acetonitrile, 60 °C, 6 h, 43%. (iv)
Acetic acid, microwave, 12%.
amine in THF yielded compound 4, which was alkylated with
different amines to produce compounds 5–7. Finally, com-
pounds 5–7 reacted with the key intermediate 8 to give the
fluorescent compounds 9–11. Further details of the synthetic
procedure can be found in the ESI.‡
Docking study of probe 10
In order to better understand the interactions between the
probes and the MDM2 protein, probe 10 was chosen to be
docked onto MDM2 using the protocol described earlier.28 As
a result, the docking conformation and orientation of com-
pound 10 could be located in the Phe19, Leu26 and Trp23 hy-
drophobic binding site, and the NDB fluorophore was
stretched outside (Fig. 1). These computational results
Results and discussion
Chemistry
A schematic of the synthesis of probes 9–11 is shown in
Scheme 2. In brief, compound 3 was initially obtained by a
sulfonation reaction. Subsequently, compound 3 and di-
methylamine hydrochloride in the presence of trimethyl-
Scheme 1 The design strategy of small-molecule fluorescent probes
for the p53–MDM2 interaction.
Fig. 1 The binding mode of probe 10 (white sticks) with the MDM2
protein.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2017

View Article Online
MedChemComm
Research Article
Table 1 Photophysical properties of the synthesized probes
Compd.
λmax (nm)
λex (nm)
λem (nm)
Φ (%)
9
10
11
375/448
377/449
376/450
370/450
370/440
365/440
575
580
575
1.88
10.09
9.59
apparently proposed the recognition of this probe by MDM2
with multiple hydrophobic interactions. More docking details
are presented in the ESI.‡
Spectroscopic properties of the probes
The spectroscopic properties of probes 9–11 (Fig. S1 and S2‡)
were measured in 5 μM solution in PBS (pH = 7.4) using a
Thermo Scientific Varioskan microplate reader, and the re-
sults were displayed in Table 1. According to the results, all
of the probes possessed excellent fluorescent properties. In
general, their fluorescence quantum yields slightly increased
with the increase in the hydrophobic chain.
Binding affinity of the probes
Subsequently, the ability of probes 9–11 to disrupt the p53–
MDM2 interaction was studied using
a
fluorescence
polarization-based (FP-based) binding assay (details are
showed in the ESI‡). First, we evaluated whether the NBD
fluorophore in our probes affected the FP assay. The result
displayed that the FP value changed very little, with the in-
creased MDM2 protein concentration measured at a wave-
length of 535 nm (excited at 485 nm) (Table S1‡). Moreover,
we evaluated the fluorescence properties of these probes in
the presence of BSA (bovine serum albumin, Energy Chemi-
cal, Shanghai, China), which often forms non-specific bind-
ing with small molecules. The experimental results clearly
demonstrated that there was a slight interaction between the
probes and BSA (Fig. S4‡). For the binding assay, Nutlin-3
with a Ki value of 0.23 μM was selected as a positive control.
It is found that all three probes showed excellent inhibitory
activity. Furthermore, the results revealed that the inhibitory
activities of probes 10–11 towards the p53–MDM2 protein
interaction were much better than that of probe 9. The calcu-
lated Ki values of probes 9–11 were 0.1 μM, 0.03 μM and 0.06
μM (Fig. S3‡), respectively, which were much better than the
positive control Nutlin-3, which had a Ki value of 0.23 μM,
and the previous probe L1, which had a Ki value of 2.29 μM.
Probe 10 in particular had a 10-fold higher potency than the
reference compound (see Table 2).
Fig. 2 Fluorescence microscopic imaging of NCI-H1229 cells (p53
null) and A549 cells (wild-type p53) incubated with 5 μM probe 9 (1,5-
bright field; 2,6-GFP channel). Imaging the inhibition of p53-MDM2
was accomplished by incubating 100 μM Nutlin-3 with 5 μM probe 9
(3-bright field; 4-GFP channel). NCI-H1229 and A549 cells were incu-
bated with the probes at 37 °C for 15 min. The backgrounds of the im-
ages were adjusted by ImageJ software. The imaging was performed
using a Zeiss Axio Observer A1 microscope with a 63× objective lens.
Cytotoxicity assay
Additionally, we assessed the cytotoxicity of these probes and
the positive control Nutlin-3 by an SRB (Sulforhodamine B)
method using a Thermo Scientific Varioskan microplate
reader. Four human cell lines from ATCC (American Type
Culture Collection, Manassas, VA 20110, USA), namely H1299
(p53 null), A549 (wild-type p53), MDA-MB-231 (wild-type p53)
Table 2 The inhibitory activities and in vitro antiproliferative activities of compounds
A549
NCI-H1299
HCT116
MDA-MB-231
Inhibitory activity
Compd.
IC50 (μM)
Ki (μM)
9
10
11
Nutlin-3
>100
>100
>100
19.9 1.3
>100
>100
>100
39.6 1.9
>100
>100
>100
28.0 0.8
>100
>100
>100
31.0 0.2
0.10
0.03
0.06
0.23
This journal is © The Royal Society of Chemistry 2017
Med. Chem. Commun.

View Article Online
Research Article
MedChemComm
Fig. 3 Fluorescence microscopic imaging of NCI-H1229 cells (p53
null) and A549 cells (wild-type p53) incubated with 5 μM probe 10
(1,5-bright field; 2,6-GFP channel). Imaging the inhibition of p53-
MDM2 was accomplished by incubating 100 μM Nutlin-3 with 5 μM
probe 10 (3-bright field; 4-GFP channel).
Fig. 4 Fluorescence microscopic imaging of NCI-H1229 cells (p53
null) and A549 cells (wild-type p53) incubated with 5 μM probe 11 (1,5-
bright field; 2,6-GFP channel). Imaging the inhibition of p53-MDM2
was accomplished by incubating 100 μM Nutlin-3 with 5 μM probe 11
(3-bright field; 4-GFP channel).
and HCT116 (wild-type p53), were selected for the cytotoxicity
assay. All studies were approved by the Ethics Committee
and IACUC of Cheeloo College of Medicine, Shandong Uni-
versity, and were conducted in compliance with European
guidelines. The results revealed that our probes showed ac-
ceptable cell toxicity in all cancer cell lines compared with
Nutlin-3 (Table 2). At the concentration of 100 μM, the inhibi-
tory rate was only 3–55% (Table S2‡). These antiproliferative
results indicated that our probes could be applied to
detecting and imaging the p53–MDM2 interaction in living
cells.
revealed that these probes can exhibit strong fluorescence
and significantly stain the A549 cell line. In contrast, staining
is very low with the NCI-H1299 cell line (Fig. 2–4). In other
words, our probes display particular selectivity for the A549
cell line (p53 wild type). Moreover, the staining occurred pri-
marily in the cytoplasm, which is in accordance with the fact
that MDM2 is mainly located in the cytoplasm. Furthermore,
in order to find whether these probes can selectively detect
the hydrophobic interface of p53–MDM2, another negative
experiment was carried out. The inhibition of the p53–MDM2
interaction was imaged by incubating the wild-type cell line
A549 with 100 μM Nutlin-3 together with each probe. The re-
sult was that the inhibition of the p53–MDM2 interaction by
Nutlin-3 could lead to a decrease in the fluorescence inten-
sity. It confirmed that our probes displayed favorable selectiv-
ity for inhibiting the p53–MDM2 interaction and could be
used in the detection of the p53–MDM2 interaction. Subse-
quently, the binding of probes 9–11 to the living cell A549
was further analyzed by flow cytometry (FCM) (Fig. 5). The to-
tal binding of our probes to the A549 cells was higher than
Fluorescence image assay and flow cytometry assay
Because probes 9–11 exhibited potent inhibitory and accept-
able cell toxicity, we evaluated the fluorescent properties of
these probes for detecting and imaging the p53–MDM2 inter-
action using a Zeiss Axio Observer A1 microscope. NCI-H1299
(p53 null) and A549 (wild-type p53) cells were chosen for the
imaging study in living cells (Fig. 2–4). The imaging results
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2017

View Article Online
MedChemComm
Research Article
Notes and references
1 A. J. Levine, Cell, 1997, 88, 323–331.
2 B. Vogelstein, D. Lane and A. J. Levine, Nature, 2000, 408,
307–310.
3 P. Chène, Mol. Cancer Res., 2004, 2, 20–28.
4 A. Shvarts, W. Steegenga, N. Riteco, T. Van Laar, P. Dekker,
M. Bazuine, R. Van Ham, W. V. D. H. Van Oordt, G.
Hateboer and A. Van der Eb, EMBO J., 1996, 15, 5349–5357.
5 X. Wu, J. H. Bayle, D. Olson and A. J. Levine, Genes Dev.,
1993, 7, 1126–1132.
6 J. Momand, G. P. Zambetti, D. C. Olson, D. George and A. J.
Levine, Cell, 1992, 69, 1237–1245.
7 Y. Barak, T. Juven, R. Haffner and M. Oren, EMBO J.,
1993, 12, 461–468.
8 J. Roth, M. Dobbelstein, D. A. Freedman, T. Shenk and A. J.
Levine, EMBO J., 1998, 17, 554–564.
9 D. Michael and M. Oren, Semin. Cancer Biol., 2003, 13,
49–58.
10 J. Momand, D. Jung, S. Wilczynski and J. Niland, Nucleic
Acids Res., 1998, 26, 3453–3459.
Fig. 5 Flow cytometry analysis of 5 μM probes 9–11 and 100 μM
Nutlin-3 binding to the living cells of A549 (red-blank; blue-Nutlin-3;
yellow-probes; green-probes and Nutlin-3). A- Probe 9, B- Probe 10,
C- Probe 11.
11 W. Wang and Y. Hu, Med. Res. Rev., 2012, 32, 1159–1196.
12 M. Esteller, M. Sanchez-Cespedes, R. Rosell, D. Sidransky,
S. B. Baylin and J. G. Herman, Cancer Res., 1999, 59, 67–70.
13 P. Chène, Nat. Rev. Cancer, 2003, 3, 102–109.
14 P. H. Kussie, S. Gorina, V. Marechal, B. Elenbaas, J. Moreau,
A. J. Levine and N. P. Pavletich, Science, 1996, 274, 948–953.
15 W. Wang and Y. Hu, Med. Res. Rev., 2012, 32, 1159–1196.
16 S. K. Knauer and R. H. Stauber, Anal. Chem., 2005, 77,
4815–4820.
the nonspecific binding to A549 cells treated by probes 9–11
and Nutlin-3. These FCM results are also consistent with the
conclusion above.
17 K. B. Lee, J. M. Hwang, I. S. Choi, J. Rho, J. S. Choi, G. H.
Kim, S. I. Kim, S. Kim and Z. W. Lee, Angew. Chem., Int. Ed.,
2011, 50, 1314–1317.
18 J. D. Amaral, F. Herrera, P. M. Rodrigues, P. A. Dionísio,
T. F. Outeiro and C. M. P. Rodrigues, Biochem. Pharmacol.,
2013, 85, 745–752.
19 N. Wetie, G. Armand, I. Sokolowska, A. G. Woods, U. Roy,
J. A. Loo and C. C. Darie, Proteomics, 2013, 13, 538–557.
20 T. K. Kerppola, Annu. Rev. Biophys., 2008, 37, 465.
21 Z. Ma, Y. Lin, Y. Cheng, W. Wu, R. Cai, S. Chen, B. Shi, B.
Han, X. Shi and Y. Zhou, J. Med. Chem., 2016, 59, 2151–2162.
22 Z. Liu, Z. Miao, J. Li, K. Fang, C. Zhuang, L. Du, C. Sheng
and M. Li, Chem. Biol. Drug Des., 2014, 85, 411–417.
23 Z. Liu, T. Jiang, B. Wang, B. Ke, Y. Zhou, L. Du and M. Li,
Anal. Chem., 2016, 88, 1511–1515.
Conclusions
In conclusion, three small-molecule fluorescent probes 9–11
with excellent fluorescent properties were designed and syn-
thesized for visualizing the p53–MDM2 interaction. Prepara-
tion of these fluorescent probes is convenient and affordable.
After extensive evaluation, these probes have been success-
fully applied to imaging the p53–MDM2 interaction in hu-
man lung cancer cell lines A549 (wild-type p53) at the micro-
molar level. Furthermore, these probes exhibited potent
affinities for the p53–MDM2 interaction and slight cytotoxic-
ity in living cells. As a result, these small-molecule fluores-
cent probes are expected to be used for drug screening and
cell staining in p53–MDM2 fields, as well as in pathological
and physiological studies of the p53–MDM2 interaction.
24 J. Liu, C. Tian, T. Jiang, Y. Gao, Y. Zhou, M. Li and L. Du,
ACS Med. Chem. Lett., 2017, 8, 428–432.
25 L. D. Lavis and R. T. Raines, ACS Chem. Biol., 2008, 3,
142–155.
Acknowledgements
This work was supported by grants from the Taishan Scholar
Program at Shandong Province, the Qilu Scholar Program at
Shandong University, the Program for Changjiang Scholars
and Innovative Research Team in University (No. IRT13028),
the Major Project of Science and Technology of Shandong
Province (No. 2015ZDJS04001) and the Shandong Key Re-
search & Development Project (No. 2015GSF118166).
26 Z. Liu, Z. Miao, J. Li, K. Fang, C. Zhuang, L. Du, C. Sheng
and M. Li, Chem. Biol. Drug Des., 2014, 85, 411–417.
27 Y.-D. Zhuang, P.-Y. Chiang, C.-W. Wang and P. D. K.-T. Tan,
Angew. Chem., 2013, 125, 8124–8128.
28 C. Zhuang, Z. Miao, L. Zhu, G. Dong, Z. Guo, S. Wang, Y.
Zhang, Y. Wu, J. Yao, C. Sheng and W. J. Zhang, Med. Chem.,
2012, 55, 9630–9642.
This journal is © The Royal Society of Chemistry 2017
Med. Chem. Commun.