EBNA1-specific luminescent small molecules for the imaging and inhibition of latent EBV-infected tumor cells

Article abstract of DOI:10.1039/c4cc01589d

An EBNA1-specific small molecule (JLP₂) has been synthesised. As a strong binder and dimerization inhibitor of EBNA1 in vitro, JLP₂ may be used as a selective luminescent agent for the imaging and inhibition of latent EBV-infected ca

Full text of DOI:10.1039/c4cc01589d

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
EBNA1-specific luminescent small molecules for
the imaging and inhibition of latent EBV-infected
tumor cells†
Lijun Jiang,a Yin-Lai Lui,b Hongguang Li,a Chi-Fai Chan,a Rongfeng Lan,c
Wai-Lun Chan,a Terrence Chi-Kong Lau,d George Sai-Wah Tsao,e Nak-Ki Mak*b
and Ka-Leung Wong*a
Cite this: Chem. Commun., 2014,
50, 6517
Received 3rd March 2014,
Accepted 28th March 2014
DOI: 10.1039/c4cc01589d
www.rsc.org/chemcomm
An EBNA1-specific small molecule (JLP2) has been synthesised. As a inhibition of latent EBV-infected cells (Fig. 1a). This study
strong binder and dimerization inhibitor of EBNA1 in vitro, JLP2 may will provide a novel strategy for interfering with the growth of
be used as a selective luminescent agent for the imaging and EBV-associated tumors.1–7 Recently, two groups of novel EBNA1
inhibition of latent EBV-infected cancer cells.
inhibitors, known as Eik1 and peptide inhibitors, have been
introduced. Eik1, developed under a high-throughout screen-
The Epstein–Barr virus (EBV) is etiologically implicated in ing, is a small molecule which can effectively inhibit the
several lymphoid and epithelial malignancies, substantially
contributing to the growth of a diversity of lymphomas and
carcinomas. Identifying the presence of EBV in the cells of
EBV-associated cancers can, therefore, provide an overarching
target for specific therapies.1 Although current treatments for
EBV-associated carcinomas, such as radiotherapy and chemo-
therapy, have long been adopted, the former is inadequate at
killing advanced metastatic tumors or preventing their recur-
rence, while the latter is still under development.2,3 In fact,
a dimeric viral oncoprotein, Epstein–Barr nuclear antigen
(EBNA1), is found to be responsible for the development of
EBV-related malignancies and the maintenance of the EBV
episome. Given that the carcinogenesis of EBV-associated car-
cinomas is symbiotically connected with EBV infection, and
that EBNA1 can function (e.g. replicate, bind DNA and trans-
activate) only upon dimerization (formation of the active form),
we hypothesize that a fluorescent probe consisting of a chro-
mophore and an EBNA1-specific molecule, which hampers the
formation of the dimer, can be used for the imaging and
a Department of Chemistry, Hong Kong Baptist University, Kowloon Tong,
Hong Kong SAR. E-mail: klwong@hkbu.edu.hk; Tel: +852-34112370
b Department of Biology, Hong Kong Baptist University, Kowloon Tong, Hong Kong
SAR. E-mail: nkmak@hkbu.edu.hk; Tel: +852-34117059
c Institute of Research and Continuing Education, Hong Kong Baptist University,
Shenzhen, P. R. China
d Department of Biology and Chemistry, City University of Hong Kong,
Kowloon Tong, Hong Kong SAR
e Department of Anatomy, The University of Hong Kong, Hong Kong SAR
Fig. 1 (a) The schematic model shows the inhibition of the processes of
EBV infection of the host cell and virus reproduction by our compounds;
(b) the chemical structures of EBNA1 specific peptides (P1 and P2) con-
jugated with water soluble luminescent moieties; (c) and (d) the binding
fitting model, obtained via molecular modeling, for the comparison of
interactions between JLP1/JLP2 and EBNA1.
† Electronic supplementary information (ESI) available: Experimental informa-
tion for the synthesis and purification of EBNA1, MTT assays and in vitro imaging;
high resolution mass spectra of P1, P2, JLP1 and JLP2; NMR spectra of inter-
mediates, ligands of JLP1 and JLP2; emission spectra of JLP1 with the addition of
various amounts of EBNA1. See DOI: 10.1039/c4cc01589d
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 6517--6519 | 6517

View Article Online
Communication
ChemComm
dimerization via association with amino acids 459–607 of the compared the interactions.12 (Fig. 1b and c – JLP1 with EBNA1,
dimerization domain of EBNA1, whilst the peptidic counterpart À7.6 kcal molÀ1 and JLP2 with EBNA1, À9.0 kcal molÀ1.) The
functions similarly in the region of 560–574.8,9 In in vitro result shows that JLP2 does bind EBNA1 more strongly than JLP1
cell line reporter studies, both inhibitors worked well in the and Eik1 (Eik1 = À6.9 kcal molÀ1) (Fig. 1c and d). Based on
EBNA1-dependent, OriP-enhanced transcription of SEAP comprehensive studies (molecular docking, spectroscopy, MBS
(secreted alkaline phosphatase). However, their activities in cross-linked SDS-PAGE assay and in vitro imaging), JLP2 shows
EBV-positive tumor cells have not been reported, and these potential as a specific dual bioprobe that can trace EBNA1 in vitro
inhibitors cannot be visualized by confocal microscopy. Addi- for both the imaging and inhibition of latent EBV-infected tumor
tionally, the poor solubility of the published peptides (e.g. p85) cells. Known EBNA1 specific compounds, namely Eik1 and the
hinders their application.9,10
aforementioned peptides, are not visualized in cells.
There are many validated anticancer drug targets, such as
JLP1 and JLP2 were purified by HPLC and characterized by
cyclins, polo-like kinase (Plk) or even EBV, and recent research NMR and high resolution mass spectroscopy (ESI†). The
has reported that these targets can be inhibited by various absorption and emission bands of JLP1 and JLP2 are 30 nm
tailor-made peptides. However, the major challenges of using red shifted compared with JL (due to the conjugation of
peptides as anti-tumor agents in vitro/in vivo lie in achieving peptides P1/P2 to JL). For JLP1 and JLP2, the absorption and
cellular internalization, their sensitivity towards enzymes excitation bands are very similar at 375 nm (JLP1) and 390 nm
in vivo and the difficulty of monitoring via indirect screening. (JLP2) in aqueous solution (Fig. 2). The emissions of both JLP1
It is widely acknowledged that proteins/peptides presenting and JLP2 are green (lem = B560 nm, lex = 400 nm, Fig. 2) In
cationic surfaces can permeate the cell membranes of eukar- comparison with the EBNA1-specific small molecules/peptides
yotic and endogenous cells. The attachment of luminescent in the literature, our compounds are fluorescent, meaning that
cargos (i.e. organic molecules/nanomaterial/metal complexes) the binding of EBNA1 in situ/in vitro, or their cellular uptake,
can provide real time monitoring of these peptides’ functions can be evaluated directly by fluorescence detection, i.e. flow
and improve their cell permeability and solubility. Herein, cytometry and fluorescence microscopy. The general
we report our newly developed EBNA1-specific dual bioprobe
(JLP2 – Fig. 1b), which conjugates with EBNA1-specific pep-
tides. The design rationale is based on molecular computation
(vide infra) and synthetic simplicity. Experiment-wise, JLP1 and
a well-known mitochondrial marker, chromophore JL,11 served
as the control. As expected, JLP2 illustrated selective and
responsive emission enhancement upon specific binding with
EBNA1, both under aqueous conditions and in vitro, with EBV-
infected tumor cells. The IC50 of JLP2 in latent EBV-infected
tumor cells was much higher than in non-EBV-infected tumor
cells. In addition, we calculated the binding affinities of
Eik1, JLP1 and JLP2 for EBNA1, via molecular docking, and
Fig. 3 The confocal images and cellular uptake (by flow cytometry) of
JLP1 and JLP2 in latent EBV-infected nasopharyngeal carcinoma C666-1
cells (a–c – JLP1; d–f – JLP2) and non-latent EBV-infected human cervical
carcinoma HeLa cells (g–i – JLP1; j–l – JLP2).
Fig. 2 The UV absorption, excitation (JLP1 – lem = 545 nm and lem
=
560 nm) and emission spectra (lex = 405 nm) of JLP1 and JLP2 in aqueous
solution. Inset: the emission enhancement of JLP2 upon the addition of
the 200 nM EBNA1.
6518 | Chem. Commun., 2014, 50, 6517--6519
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
In addition, the cellular uptake of JLP1 differs from that of
JLP2. JLP1 and JLP2 (10 mM) were tested in latent EBV-infected
nasopharyngeal carcinoma C666-1 cells for 12 hours. The
uptake of JLP2 could be observed more apparently: the confocal
images from the C666-1 cells in the presence of JLP1 and JLP2
have been related to the findings from the flow cytometry
uptake studies (Fig. 3). No emission can be obtained from
JLP1 in the C666-1 cells and only JLP2 displays impressive
emission in the latent EBV-infected nasopharyngeal carcinoma
C666-1 cells in the same experiment.
In the cytotoxicity experiment (JLP1 and JLP2 in EBV/
non-EBV infected tumor cells, and Eik1 as the control), JLP2
demonstrated smaller IC50 values in EBV infected cancer cells
(C666-1, BIC50 = 20 mM, similar to the Eik1) than in non-EBV
infected cancer cells (HeLa, BIC50 = 1 mM). On the other hand,
JLP1, P1 and P2 demonstrated similar IC50 values with HeLa and
C666-1 cells under the same experimental conditions. Peptides
P1 and P2 are not cell permeable and alone are not effective in
preventing the growth of carcinoma cells, and so served as the
control experiments in the MTT assays (Fig. 4).
A specific and selective agent for both the imaging and
targeting of this virus-associated tumor is not currently avail-
able in the literature. In this regard, we have synthesised a dual-
function fluorescent probe (JLP2) of verified applicability for the
simultaneous imaging and control of the growth of a tumor
latently infected with EBV (e.g. nasopharyngeal carcinoma), on
account of the probe’s underlying mechanism of specifically
preventing the dimerization of EBNA1.
This work was funded by grants from The Hong Kong
Research Grants Council (HKBU 203013), City University of Hong
Kong, and Hong Kong Baptist University (FRG 2/12-13/069).
Fig. 4 The inhibitory activities, measured by cell viability, of two EBNA1
specific peptides (P1 and P2), two small molecules (JLP1 and JLP2) and Eik1
(control) on (a) non-latent EBV-infected human cervical carcinoma HeLa
cells and (b) latent EBV-infected nasopharyngeal carcinoma C666-1 cells.
Notes and references
1 H. K. Koon, P. S. Chan, R. N. S. Wong, Z. G. Wu, M. L. Lung,
C. K. Chang and N. K. Mak, J. Cell. Biochem., 2009, 108, 1356.
2 L. S. Young and A. B. Rickinson, Nat. Rev. Cancer, 2004, 4, 757;
K. C. Chan, C. M. Ting, P. S. Chan, M. C. Lo, K. W. Lo, J. E Curry,
T. Smyth, A. W. M. Lee, W. T. Ng, G. S. W. Tsao, R. N. S. Wong,
M. L. Lung and N. K. Mak, Mol. Cancer, 2013, 12, 128.
photophysical properties of JLP1 and JLP2 are similar, as only
one amino acid in their structures is different; however, the
in situ and in vitro behaviours of JLP1 and JLP2 are completely
distinct. For instance, after the addition of EBNA1 into the
aqueous solutions of JLP1 and JLP2, emission enhancement
could be detected only for JLP2, more or less 1.5-fold upon the
addition of 200 nM EBNA1 (inset of Fig. 2), whereas, for JLP1, the
emission intensity was slightly diminished with the same dose of
EBNA1 in water (Fig. S14, ESI†). Aside from this, we had also
confirmed the specific binding of JLP2 to EBNA1 by MBS
(3-maleimidobenzoyl N-hydroxysuccinimide) crossed-linked SDS-
PAGE assay. MBS can form covalent cross-linkages with dimerized
3 G. Kennedy and B. Sugden, Mol. Cell. Biol., 2003, 23, 6906.
´
4 A. T. C. Chan, V. Gregoire, J.-L. Lefebvre, L. Licitra and E. Felip, Ann.
Oncol., 2010, 21, 187.
5 L. Frappier, Scientifica, 2012, 2012, 438204.
6 D. W. Smith and B. Sugden, Viruses, 2013, 5, 226.
7 S. J. Duellman, K. L. Thompson, J. J. Coon and R. R. Burgess,
J. Gen. Virol., 2009, 90, 2251.
8 N. Li, S. Thompson, D. C. Schultz, W. Zhu, H. Jiang, C. Luo and
P. M. Lieberman, PLoS One, 2010, 5, e10126.
9 S. Y. Kim, K. A. Song, E. Kieff and M. S. Kang, Biochem. Biophys. Res.
Commun., 2012, 424, 251.
EBNA1 proteins, which can be separated using SDS-PAGE. That is, 10 M. Kenji, Y.-S. Naoko, O. Chikashi, T. Toshiki and M. Hisao,
J. Biol. Chem., 2012, 287(28), 23977.
11 G. R. Rosania, J. W. Lee, L. Ding, H.-S. Yoon and Y.-T. Chang,
EBNA1, disrupted by the compound in the assay, can no longer
exist as a dimer. In Fig. S16 (ESI†) (black box), it can be seen that
J. Am. Chem. Soc., 2003, 125, 1130.
only JLP2 obviously decreased the amount of dimeric EBNA1.
12 O. Trott and A. J. Olson, J. Comput. Chem., 2010, 31, 455.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 6517--6519 | 6519