Monoamine oxidase-A targeting probe for prostate cancer imaging and inhibition of metastasis

Article abstract of DOI:10.1039/c9cc07009e

Mitochondrial enzyme monoamine oxidase (MAO-A) is known to be overexpressed in prostate cancer (PCa) cells. Herein, we have developed a two-photon probe (PCP-1) for selectively targeting and imaging the MAO-A in PCa. Supported by enzymatic docking and in vitro experiments, PCP-1 showed efficiency to visualize MAO-A overexpressing cells and inhibit their growth and metastasis potential.

Full text of DOI:10.1039/c9cc07009e

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Monoamine oxidase-A targeting probe
for prostate cancer imaging and inhibition
of metastasis†
Cite this: Chem. Commun., 2019,
55, 13267
Received 9th September 2019,
Accepted 3rd October 2019
a
Won Young Kim,‡a Miae Won, ‡a Abbas Salimi,‡b Amit Sharma,
Jong Hyeon Lim,c Seung-Hae Kwon,d Joo-Yeong Jeon,d Jin Yong Lee *c and
a
DOI: 10.1039/c9cc07009e
Jong Seung Kim
*
rsc.li/chemcomm
Mitochondrial enzyme monoamine oxidase (MAO-A) is known to
Recent studies have shown that increased levels of mono-
be overexpressed in prostate cancer (PCa) cells. Herein, we have amine oxidase-A (MAO-A) are associated with PCa progression
developed a two-photon probe (PCP-1) for selectively targeting and and inhibition of MAO-A activities prevents the growth of PCa cells
imaging the MAO-A in PCa. Supported by enzymatic docking and in vitro and tumor xenografts in vivo.7 MAO-A is a mitochondria-
in vitro experiments, PCP-1 showed efficiency to visualize MAO-A bound enzyme which catalyzes the oxidative deamination of neuro-
overexpressing cells and inhibit their growth and metastasis potential. transmitters and dietary amines.8 This process produces hydrogen
peroxide (H2O2), a major source of reactive oxygen species (ROS),
Prostate cancer (PCa), is the second leading cause of cancer- which can damage the DNA of cancer cells, the main cause of tumor
related deaths among most male patients in the world.1 The initiation and progression.9 Moreover, previous studies strongly
incidence rate is high with all ages in America followed by support that excessive intracellular levels of H2O2 produced by
Australia/New Zealand and Europe. Cancer metastasis, in MAO-A are responsible for inducing the epithelial to mesenchymal
general, primarily causes human cancer deaths to the extent transition (EMT) in PCa. MAO-A-mediated generation of ROS is
of more than 90%.2 PCa progresses slowly, and as a result many also associated with activation of the transcription factor HIF-1a
patients die because of competing causes. Thus, accurate (a master regulator of hypoxia)10 which further increases the level
detection of PCa and its metastatic potential is considered to of ROS and ultimately drives PCa progression and metastasis.11 In
be needed to decide the exact tumor stage in patients and the this work, MAO-A was identified as an important target for the
correct treatment plan in clinical practices.3,4 Widespread use selective detection of human PCa in vitro.
of prostate specific antigen (PSA) level testing in patients has
Recently, fluorescence imaging has become the most pro-
led to a drastic increase in diagnosis and therapy,6 but some- minent tool in bio-imaging related studies.12 Fluorescence-
times patients don’t receive benefits from intervention due to based diagnosis are fast, noninvasive, sensitive and selective
either the indolent or disseminated state of the disease at the tools for treatment estimation in oncology. Until near-infrared
diagnosis stage. Moreover, this test exhibits low specificity (NIR) fluorophores were used, otherwise, most of the reported
because PSA levels are also elevated in patients with benign fluorophores were associated with common shortcomings like
prostatic hyperplasia.5 This low specificity can lead to excessive poor tissue penetration of the excitation or emission lights
treatments of early-stage and less aggressive cancers6 and it is when applied to mouse or animal models.13 Alternatively, we
both time and money consuming with poor clinical outcomes. chose a reporter with two-photon (TP) excitation capability
Hence, the search for alternate targets and development of because it would be ideal for in vitro bio-imaging as well as
sensitive detection tools with high tumor-targeting specificity in vivo real-time monitoring. Deep tissue penetration of TP
are urgently needed.
fluorescence enhances the microscopic images and its weak
excitation energy reduces undesired tissue damage.14 It may
also facilitate the in vivo tumor imaging and differentiation of
tumors from surrounding normal tissues and organs.
a Department of Chemistry, Korea University, Seoul 02841, Korea.
E-mail: jongskim@korea.ac.kr
b School of Chemical Engineering, Sungkyunkwan University, Suwon 16419, Korea
c Department of Chemistry, Sungkyunkwan University, Suwon 16419, Korea.
E-mail: jinylee@skku.edu
d Seoul Center, Korea Basic Science Institute, Seoul 02841, Korea
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c9cc07009e
In view of the aspects mentioned above, in this investigation,
a novel TP fluorescent probe PCP-1 for targeting MAO-A was
designed. This was achieved by conjugating a MAO-A selective
inhibitor (moclobemide)15 with a TP fluorophore to improve
excitation permeability for imaging. The PCP-1 can selectively
bound to MAO-A, overexpressed in the PCa (Scheme 1). Thus,
‡ These authors contributed equally to this work.
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 13267--13270 | 13267

View Article Online
Communication
ChemComm
Fig. 1 (a) Residual interactions within binding sites of human MAO-A and
probe PCP-1 (pink). (b) Representative structure obtained from the docking
study (PDB Code: 2Z5X).
wavelength of the TP fluorophore for first to third excited states
were 920 nm, 808 nm and 792 nm with a two-photon cross-
sections of 460 GM, 404 GM, and 396 GM (Goeppert–Mayer),
respectively. The calculated data was substantiated by compar-
ison to the reference dye, implying the selected TP fluorophore
can promise ideal functions as a bio-imaging probe for PCa
(Fig. S10, ESI†).
In order to test whether the probe PCP-1 could be used as an
imaging probe for PCa, the optical properties of probe PCP-1 in
aqueous environments were investigated. UV-Vis analysis of
probe PCP-1 (10 mM) in phosphate saline buffer (PBS) solution
(pH 7.4, containing 0.2% DMSO v/v) showed an absorption
maximum at 566 nm, characteristic of the rhodamine moiety
(Fig. S12a, ESI†). Upon excitation at 566 nm, probe PCP-1
exhibited a strong fluorescence emission band centered at
600 nm (Fig. 2a). Further, absorption and emission bands of
probe PCP-1 changed negligibly under both acidic and alkaline
environments (pH 3–10) (Fig. S12 and S2b, ESI†). This result
suggests that probe PCP-1 is unlikely to be affected by the acidic
pH environments of a tumor.
Scheme 1 (a) Design of PCP-1 targeting MAO-A for the detection of
prostate cancer (PCa) and (b) synthetic scheme of PCP-1.
this probe may allow the prostate tumors to be optically dis-
cerned from normal tissues by in vivo TP fluorescence without
surgical methods. In addition, it can prevent cancer cell prolif-
eration and metastasis by inhibition of MAO-A activities.
The probe PCP-1 was successfully synthesized and was
1
analytically characterized by H, 13C NMR and MS (ESI†). The
synthesis of probe PCP-1 is shown in Scheme 1b. Briefly, a TP
fluorophore, i.e., rhodamine derivative with a terminal alkyne
functionality (5), was conjugated with the MAO-A inhibitor
carrying an azide group (3) by using the Cu(I)-mediated azide–
alkyne cycloaddition reaction, generating a triazole moiety which
is easy to synthesize and does not interfere with the biological
environment in vivo.16 To minimize the steric influence of the
fluorophore on the binding interaction between the synthetic
inhibitor (moclobemide) and MAO-A enzyme, a six-carbon alkyl
chain was selected as a linker between the fluorophore and the
inhibitor.17
To initially confirm the interaction of the probe PCP-1 towards
MAO-A enzyme, a docking study and further molecular dynamics
(MD) simulation (Fig. 1) were performed to understand the binding
properties at the atomic level. After MD simulations, a binding
free energy of À118.6 kcal molÀ1 was obtained, and the residual
interactions are shown in Fig. S9 (ESI†). The simulation data
confirmed that the inhibitor was not sterically hindered by the
linker and tagged fluorescent group, indicating that probe PCP-1
was successfully designed for binding with MAO-A.
Fig. 2 (a) Fluorescence spectra of MAO-A-targeted two-photon probe
PCP-1 (10 mM) in 10 mM PBS buffer (pH 7.4, 0.2% DMSO, 37 1C), lex = 566 nm,
lem = 600 nm. (b) Fluorescence intensity change with varying pH range of
the solutions (DMSO–water, 0.2% v/v), lex = 566 nm, lem = 600 nm.
(c) Two-photon action cross section spectra of PCP-1 (10 mM) in PBS buffer
(37 1C). (d) Two-photon fluorescence spectra of PCP-1 in 10 mM PBS buffer
(pH 7.4, 0.2% DMSO, v/v), lex = 830 nm, lem = 600 nm.
In order to verify the photochemical function of a TP fluoro-
phore, density functional theory (DFT) calculations (Fig. S10
and S11, ESI†) were performed. The calculated TP absorption
13268 | Chem. Commun., 2019, 55, 13267--13270
This journal is ©The Royal Society of Chemistry 2019

View Article Online
ChemComm
Next, to explore the TP imaging ability of probe PCP-1 under
Communication
physiological conditions, TP excitation spectra of PCP-1 (10 mM)
was investigated in a PBS buffer. As shown in Fig. 2c, the GM
values (dF) of probe PCP-1 over absorbance labs = 810 nm reached
a maximum, corresponding to results from the calculated TP
absorption data above. Subsequently, PCP-1 was irradiated at
830 nm in PBS solutions (pH 7.4, 0.2% DMSO v/v). As expected,
a fluorescence emission band was observed centered at 600 nm
upon irradiation (Fig. 2d). These experiments successfully demon-
strated the potential of probe PCP-1 for TP imaging of PCa as it
was sufficiently stable in physiological conditions, able to produce
a strong fluorescent emission and has two-photon properties.
These features are inevitably required for small-molecule imaging
in the cells.
Following the encouraging results described above, various
in vitro cell tests were performed. First, in order to examine the
targeting ability of PCP-1 for PCa, TP images were confirmed in
various cell lines with different MAO-A expression levels. Before
that, an immunoblot assay was performed to compare the MAO-A
expression levels in four human cancer cell lines, namely, LNCap
(prostate cancer), MDA-MB-231 (breast cancer), HepG2 (liver
cancer), and A549 (lung cancer) cells, respectively. Among these
four tested cell lines, LNCap cells exhibited the highest levels
of MAO-A enzyme (Fig. S13, ESI†). Subsequently, the probe PCP-1
(10 mM) was treated with each cell line using incubation for 24 h.
As expected, PCP-1 emitted the highest fluorescence intensity with
Fig. 3 (a) TP cell images of human cancer cells; LNCap (prostate adeno-
carcinoma), MDA-MB-231 (breast adenocarcinoma), HepG2 (hepatocellular
carcinoma), and A549 (lung adenocarcinoma) cells treated with PCP-1. The
cells were incubated with HBSS containing the probe (10 mM) and then the
images were obtained. lex = 830 nm; lem = 500–700 nm. (b) Retention time
assay for fluorescence intensity changes of PCP-1 in various tested cancer
cells. (c) Colocalization images of PCP-1 and sub-organelles (Mitochondria,
Lysosome, and ER). (d) Mitochondria membrane potential measurement in
various cancer cells upon treatment with PCP-1.
the LNCap cell line, exhibiting a noticeable difference among due to the fluorophore, but rather to the targeting effect of the
the other tested cell lines, which is in accordance with the MAO-A inhibitor used in PCP-1. Together, these experimental results
MAO-A expression levels in these cells (Fig. 3a). An excellent cell indicated that the probe PCP-1 has excellent MAO-A target ability
permeability of PCP-1 was also confirmed.
and can selectively detect PCa as a promising bio-imaging probe.
Recent reports suggested that overexpression of MAO-A in
Next, to examine the retention ability of PCP-1, time-dependent
fluorescence imaging was performed in the cell lines. As shown in human PCa cells enhances cancer cell growth and induces
Fig. 3b, the probe PCP-1 showed significant fluorescence emission EMT.10 Therefore, we examined cell viability and migration
in the LNCap cell line with sufficient time (48 h), which is the assays to test the ability of PCP-1 to inhibit MAO-A activity in
required characteristic for successful in vivo imaging. In addition, to PCa cells. LNCap cells were treated with rhodamine fluoro-
validate whether the probe PCP-1 binds with MAO-A in the cancer phore, moclobemide and clorgyline as control groups and
cells, inhibition studies were conducted. Clorgyline, a selective PCP-1, respectively. As shown in Fig. 4a and Fig. S16a (ESI†),
inhibitor of MAO-A enzyme, was pretreated (100 mM) in LNCap the group treated with PCP-1 exhibited a remarkable reduction
cells prior to incubation with PCP-1. The results showed a notice- in cell viability in a dose-dependent manner, while cells treated
able reduction in fluorescence intensity in the LNCap cell (Fig. S14 the fluorophore (rhodamine) showed only a little effect even at
and S15, ESI†). These results support that a strong fluorescence high doses. This result showed that PCa cell growth was sup-
and increased retention in the PCa cell are caused by the MAO-A pressed by the MAO-A inhibitor of PCP-1 impeding the MAO-A
targeting unit of PCP-1 which selectively binds with MAO-A.
activity. Compared with PCP-1, the groups treated with clorgyline
Moreover, the colocalization of probe PCP-1 was also examined and moclobemide, respectively also showed a decrease in cell
by confocal microscopy upon co-incubation with the commercial viability but less than PCP-1 (Fig. 4b and Fig. S16b, ESI†). Thus,
Mito (mitochondrial), Lyso (lysosomal) and ER (endoplasmic reti- PCP-1 has a better targeting and inhibition effect on MAO-A’s
culum) trackers. As shown in confocal merged images (Fig. 3c), the activities, which may have resulted from the preferential mito-
fluorescence of PCP-1 was completely localized with the Mito chondria targeting effect of PCP-1. The effective cell death
tracker. Since the MAO-A enzyme is bound to the outer membrane mechanism of PCP-1 was also investigated by Western blot
of mitochondria in a cell, PCP-1’s localization in the mitochondria and Annexin V/PI (Fig. 4c and d). The reduced expression of
is the inevitable result. Subsequently, changes in the membrane the MAO-A protein by PCP-1 in LNCap cells induced cell death,
potentials of mitochondria in each of the cell lines were determined which was demonstrated by increased BAX (pro-apoptotic
by a membrane potential (MMP) assay. Clearly, the LNCap cell protein) and decreased BCL-2 (anti-apoptotic protein) expres-
exhibited the lowest mitochondria membrane potential (Fig. 3d) sions. Also, the level of cleaved caspase 3 was enhanced by
upon treatment with PCP-1. These results demonstrated that the PCP-1 treatment (Fig. 4c). Further, the Annexin-V/PI staining
reason for PCP-1 accumulation in cancer cell mitochondria is not assay confirmed that PCP-1 treatment increased the Annexin
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 13267--13270 | 13269

View Article Online
Communication
ChemComm
results were attributed to PCP-1’s ability to inhibit MAO-A’s
activity, thereby suppressing prostate cancer cell growth, pro-
liferation and preventing their metastatic potential.
In conclusion, a new TP imaging probe (PCP-1) was devel-
oped. PCP-1 was preferentially taken up and retained in the
MAO-A overexpressing PCa cells (LNCap) over other tested
human cancer cell lines. Furthermore, the remarkable selectivity
towards MAO-A was also supported by enzymatic docking
studies. In addition, PCP-1 showed a significant inhibition
effect on PCa cell growth, proliferation, and cancer metastasis
by inhibiting MAO-A’s activities. These results, therefore,
suggest that the probe PCP-1 will be highly applicable for
further in vivo imaging and the design of targeted therapeutics
for PCa in the future.
This work was supported by the National Research Foundation
of Korea (NRF) funded by the Ministry of Science and ICT (CRI
project no. 2018R1A3B1052702 and NRF-2019M3E5D1A01068998,
J. S. K.), and the Basic Science Research Program (2017R1D1A1-
B03030062, M. W.) funded by the Ministry of Education. This work
was also supported in part by Korea Basic Science Institute Grant
(T38662).
Fig. 4 PCP-1 induced cell death. Cell viability of PCP-1, and moclobemide.
Cells were incubated with 0, 1, 5, 10, 30 and 50 mM of PCP-1 (a), and
moclobemide (b) for 24 h in LNCap cells. Values represent mean Æ SE of
three independent experiments performed in triplicate; *p o 0.05. (c) Cell
death-related protein expression levels by PCP-1 in LNCap cells are shown
by Western blot assay. (d) AnnexinV/PI staining assay in LNCap cells
incubated with the PCP-1 (10 mM) and then the values were obtained for
24 h (control group treated with 1% DMSO).
Conflicts of interest
There are no conflicts to declare.
Notes and references
1 R. L. Siegel, K. D. Miller and A. Jemal, Ca-Cancer J. Clin., 2016, 66,
7–30.
2 C. L. Chaffer and R. A. Weinberg, Science, 2011, 331, 1559–1564.
3 T. Aikou, Y. Kitagawa, M. Kitajima, Y. Uenosono, A. J. Bilchik,
S. R. Martinez and S. Saha, Cancer Metastasis Rev., 2006, 25, 269–277.
4 R. Siegel, E. Ward, O. Brawley and A. Jemal, Ca-Cancer J. Clin., 2011,
61, 212–236.
5 L. D. Papsidero, M. C. Wang, L. A. Valenzuela, G. P. Murphy and
T. M. Chu, Cancer Res., 1980, 40, 2428–2432.
6 D. Bates, S. Abraham, M. Campbell, I. Zehbe and L. Curiel, PLoS
One, 2014, 9, e97220.
7 V. Flamand, H. J. Zhao and D. M. Peehl, J. Cancer Res. Clin. Oncol.,
2010, 136, 1761–1771.
8 M. Bortolato, K. Chen and J. C. Shih, Adv. Drug Delivery Rev., 2008,
60, 1527–1533.
9 S. Josson, T. Nomura, J. T. Lin, W. C. Huang, D. Q. Wu, H. E. Zhau,
M. Zayzafoon, M. N. Weizmann, M. Gururajan and L. W. K. Chung,
Cancer Res., 2011, 71, 2600–2610.
10 A. Sharma, J. F. Arambula, S. Koo, R. Kumar, H. Singh, J. L. Sessler
and J. S. Kim, Chem. Soc. Rev., 2019, 48, 771–813.
11 J. B. Wu, C. Shao, X. Li, Q. Li, P. Hu, C. Shi, Y. Li, Y. T. Chen, F. Yin,
C. P. Liao, B. L. Stiles, H. E. Zhau, J. C. Shih and L. W. K. Chung,
J. Clin. Invest., 2014, 124, 2891–2908.
Fig. 5 PCP-1 inhibits EMT related to metastasis in prostate cancer cells.
(a) Cell migration assay in LNCap cells incubated with the PCP-1 (10 mM)
and then the images were obtained for 24 h (control group treated with
1% DMSO). (b) The expression of EMT-related markers, E-cadherin,
N-cadherin, and Twist-1 was observed after PCP-1 (10 mM) treatment
using Western blot assay.
V-positive apoptotic cells population (by 40%) compared to
the control (Fig. 4d).
12 X. Q. Chen, T. Pradhan, F. Wang, J. S. Kim and J. Yoon, Chem. Rev.,
2012, 112, 1910–1956.
To investigate the metastatic potential inhibition of LNCap
cells by PCP-1, we used a cell migration assay. After treatment 13 H. W. Liu, K. Li, X. X. Hu, L. M. Zhu, Q. M. Rong, Y. C. Liu,
X. B. Zhang, J. Hasserodt, F. L. Qu and W. H. Tan, Angew. Chem., Int.
Ed., 2017, 56, 11788–11792.
14 X. L. Xie, J. L. Fan, M. W. Liang, Y. Li, X. Y. Jiao, X. Wang and
with PCP-1 (10 mM, 48 h) in LNCap cell lines, a significant
reduction in migration of PCP-1 treated cells were observed as
compared to the control group (Fig. 5a). Further, E-cadherin
was upregulated whereas N-cadherin and Twist-1 were signifi-
cantly downregulated in LNCap cells treated with PCP-1, sug-
gesting that PCP-1 could inhibit EMT in PCa (Fig. 5b). These
B. Tang, Chem. Commun., 2017, 53, 11941–11944.
15 B. Fulton and P. Benfield, Drugs, 1996, 52, 450–474.
16 M. Meldal and C. W. Tornoe, Chem. Rev., 2008, 108, 2952–3015.
17 S. A. Kularatne, Z. G. Zhou, J. Yang, C. B. Post and P. S. Low, Mol.
Pharmacol., 2009, 6, 790–800.
13270 | Chem. Commun., 2019, 55, 13267--13270
This journal is ©The Royal Society of Chemistry 2019