Rational design of biotin-disulfide-coumarin conjugates: A cancer targeted thiol probe and bioimaging

Article abstract of DOI:10.1039/c3cc49790a

Biotin-disulfide-coumarin conjugates are designed and synthesized as novel fluorescent sensors for cancer targeted intracellular thiol imaging in living organisms. In vitro experiments disclose that probe 6 is preferentially taken up by biotin receptor-positive A549 tumor cells through receptor mediated endocytosis. The Royal Society of Chemistry.

Full text of DOI:10.1039/c3cc49790a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Rational design of biotin–disulfide–coumarin
conjugates: a cancer targeted thiol probe and
bioimaging†
Cite this: Chem. Commun., 2014,
50, 3044
Received 27th December 2013,
Accepted 27th January 2014
Danbi Jung,‡a Sukhendu Maiti,‡a Jae Hong Lee,a Joung Hae Leeb and
Jong Seung Kim*a
DOI: 10.1039/c3cc49790a
www.rsc.org/chemcomm
Biotin–disulfide–coumarin conjugates are designed and synthesized some fluorophores often show aggregation-caused quenching (ACQ)
as novel fluorescent sensors for cancer targeted intracellular thiol properties, which diminishes their brightness and sensitivity in
imaging in living organisms. In vitro experiments disclose that probe cellular imaging.7 Considering the significant role of thiols in cellular
6 is preferentially taken up by biotin receptor-positive A549 tumor functions, it is highly desirable to develop simple, non-invasive, and
cells through receptor mediated endocytosis.
specific probes with high signal-to-noise ratios, and good water
solubility for targeted imaging of intracellular thiols. Therefore, in
Cellular thiols including glutathione (GSH), cysteine (Cys), and homo- view of the demand, we wish to design, synthesise and perform the
cysteine (Hcy) are essential biomolecules which play pivotal roles in spectroscopic characterization of intracellular thiol sensing systems
many biochemical and physiological processes, such as cell prolifera- with precision as well as fluorescence-On, for in vivo and vitro use.
tion, antioxidant defence, and cell signaling.1 Among these metabolites
One of the major challenges in modern research is to develop
GSH (in its free, reduced form) is the most plentiful, which plays an target specific probes which can be used to detect particular cellular
important role in maintaining the oxidation–reduction (redox) state of components that are associated with disease states in vitro. Generally
the cell homeostasis and is the vital source of reducing equivalents for these probes could be used to rapidly test the therapeutic effects. In
enzymes such as superoxide, dismutase, catalase, and glutathione this context, the use of target specific ligands and conjugation
peroxidase.2 In cells, the deviations from optimum ratios of GSH/GSSG strategies that allow for the selective targeting of a probe have
are closely related to various diseases including Alzheimer’s disease, emerged as attractive approaches. Biotin, also known as vitamin
cardiovascular disease, Parkinson’s disease, liver damage and cancer.3 B7 and vitamin H, or coenzyme R is a water soluble B-complex
Moreover, the efficacy of chemotherapy also depends on GSH levels in vitamin that plays an important role in helping the body metabolize
cancer cells.4 Thus, the quantitative detection of these intracellular proteins and process glucose. Biotin is very much well known as a
thiols has drawn a great deal of attention for early identification of cancer targeting unit and also a growth promoter at the cellular
diseases, assessment of disease development and therapeutic effective- level because in the cancer cells biotin receptors such as avidin,
ness of new prospective drugs.
streptavidin are much more than that of normal cells.8 Moreover,
Fluorescent molecules possess excellent light stability, high biotin receptors are overexpressed more than the folate or vitamin
sensitivity, less toxicity and real-time monitoring ability, which have B-12 receptors in many cancer cells, such as leukemia (L1210FR),
been shown to be essential for the quick and precise detection of ovarian (Ov2008, ID8), colon (Colo-26), lung (M109), and breast
intracellular organic biomolecules.5 To date, a small number of (4T1, MMT06056) cancer cell lines respectively.8,14 Therefore the
rationally designed fluorescent probes have been developed for the biotin-appended coumarin unit can go to cancer cell specifically
detection of biologically free or intracellular thiols.6 However, the than that of normal cells through the strong interaction between
maximum number of probes have exhibited poor water solubility, biotin receptor avidin and biotin. The coumarin unit has the ability
high background signal, narrow pH span, and slow response, and of the two-photon photophysical property, which provides the
lack cell specificity with limited biological applications. In addition, advantages of deeper tissue penetration and less photodamage.9
One photon fluorescent spectroscopy is not recommended for use
in living systems because the shorter wavelength light is dispersed
and absorbed by native cellular components. In addition, from a
synthetic point of view, the coumarin moiety is also easily accessible
and has structural tractability that allows for various modifications.
a Department of Chemistry, Korea University, Seoul 136-701, Korea.
E-mail: jongskim@korea.ac.kr; Fax: +82-2-3290-3121
b Korea Research Institute of Standards and Science, Daejon 305-600, Korea
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
Therefore, the biotin-appended coumarin unit (6) is expected to
c3cc49790a
‡ Equally contributed to this research.
monitor intracellular thiols level in cancer cells through fluorescence
3044 | Chem. Commun., 2014, 50, 3044--3047
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
Fig. 1 (a) Absorption spectra of 6 (25.0 mM, 5% DMSO, and 95% PBS buffer)
recorded in the presence and absence of GSH (5.0 mM). (b) Fluorescence
spectra of 6 (5.0 mM) recorded in the presence and absence of GSH (5.0 mM).
(c) Fluorescence changes of 6 (5.0 mM) seen upon treatment with increasing
concentration of GSH (0–60 equiv.). (d) Changes in fluorescence intensity at
450 nm as a function of GSH concentration.
Scheme 1 Synthetic routes to 6.
The absorption and emission nature of the solution 6 after
GSH treatment was indistinguishable with that of reference 7
(Fig. S1, ESI†). We also speculated that on gradual addition of
GSH (0–50 equiv.) to the solution of probe 6 (5.0 mM) in aqueous PBS
buffer, the solution’s fluorescence intensity at lmax = 450 nm reached
saturation (Fig. 1c and d). These results firmly demonstrated that the
disulfide bond of 6 can be immediately cleavable by GSH, the most
essential thiols among other reduced thiols in cancer cells.
Whether the projected GSH-mediated cleavage reaction of probe 6
is affected by the other biologically relevant analytes in cells, the
response of 6 with various thiols, thiol-free amino acids, monovalent,
and bivalent metal ions were investigated under the same reaction
conditions. Similar spectroscopic changes of probe 6 were observed
upon the addition of cysteine (Cys) or homocysteine (Hcy) to those
encountered by GSH (Fig. S22, ESI†). In addition, no substantial
spectroscopic changes were noticed upon disclosure to thiol-free
amino acids such as Asp, Arg, Ala, Val, Tyr, Gly, etc. respectively and
biologically relevant metal cations, including Na+, K+, Ca2+, Mg2+, Zn2+,
Fe2+, Fe3+, Cu2+ respectively (Fig. 2). Only thiol containing amino acids
specifically react with disulfide bonds because the nucleophilicity
character of thiol containing amino acids along with redox potential
imaging, which can be used in practical applications. Here, we report
a cancer targeting fluorescent probe 6 which contains a biotin-
appended coumarin unit that allows for the real-time fluorescence
imaging of reduced thiols such as GSH, Cys, and Hcy respectively
in cells. Conjugates 6 are combined with a cancer targeting unit
as biotin, a disulfide linker cleavable by the intracellular thiols, a
coumarin moiety which gives rise to enhance the fluorescence
intensity after disulfide bond cleavage.
Our aim to develop an intracellular thiol-specific probe relies on
the use of a specific receptor-mediated endocytosis approach. Biotin-
appended coumarin conjugates (6) were synthesized according to
the following synthetic pathway shown in Scheme 1.
Compounds 1 to 3 were synthesized by reported methods.10
Compound 3 was reacted with phosgene and diisopropylethylamine
(DIPEA) followed by reaction with coumarin (7) to provide 4.
The biotin-azide derivative (5) was prepared from the previously
published procedures.11 Finally, our desired compound 6 was
obtained from compounds 4 and 5 by the click reaction in good
yield. The chemical structures of all compounds (1–6) were
1
confirmed by H NMR, 13C NMR, and ESI-MS data.
Before proceeding to cell imaging experiments, we make
clear whether biologically relevant thiols were able to induce
cleavage of the disulfide bond in probe 6. These experiments
were performed by reacting 6 with glutathione (GSH) under
physiological conditions and monitoring the changes in the
optical features using UV-Vis and fluorescence spectroscopy.
As shown in Fig. 1a, probe 6 revealed a strong absorption band
at lmax = 325 nm. This absorption peak intensity increases
by B1.4-fold with some red shift upon the addition of (5.0 mM)
GSH. The solution of 6 displayed very weak fluorescence
centred at 450 nm on excitation at 325 nm. The fluorescence
emission signal of 6 (5.0 mM) at lmax = 450 nm in aqueous PBS
buffer was enhanced by B5-fold following exposure to GSH
(5.0 mM) as shown in Fig. 1b.
Fig. 2 Fluorescence spectra of 6 (5.0 mM) recorded in the absence and
presence of GSH, Cys, Hcy (5.0 mM, respectively) and various metal
cations (1.0 mM: monovalent metal ions; 0.1 mM: divalent metal ions,
respectively) with excitation effected at 325 nm. All spectra were recorded
in 5% DMSO & 95% PBS buffer (pH 7.4) at 37 1C.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 3044--3047 | 3045

View Article Online
ChemComm
Communication
play a vital role in the physiological environment.6 In the cancer cells
For the promising bio-application toward thiol-mediated cleavage
generally reactive oxygen species (ROS) such as hydrogen peroxide of the vulnerable S–S bond and concomitant fluorescence-On in vitro
(H2O2), super oxide radicals (O2À), and nitric oxide (NO) are present as test, the conjugate 6 was applied to living cells. Initially, we used one
interfering particles. But this ROS species does not break the disulfide photon fluorescence spectroscopy to take images in living cells using
linkage. These outcomes of the above experiment indicate that the probe 6. Due to shorter wavelength light scattered and autocaptivated
disulfide bond in 6 will undergo specific thiol-mediated cleavage by native cellular components the cellular imaging was not good.
without substantial interference from other chemical analytes which Thus, we implemented two photon fluorescence spectroscopy, which
might be ubiquitous in a biological system.
has the following advantages such as minimized light scattering,
Furthermore, we have also inspected the pH-dependent changes reduced photodamage to living biological samples and fluoro-
in fluorescence of 6 (5.0 mM), with and without addition of 5.0 mM phores, and improved spatial resolution, and sensitivity as well as
GSH. As shown in Fig. S2 (ESI†), in the absence of GSH, 6 is stable the ability to image thicker specimens.9
over a pH range from 4 to 8. However, in the presence of GSH, a
In order to apply probe 6 to the living systems, firstly we investigated
substantial enhancement of fluorescence intensity was observed whether the biotin moiety can lead the probe especially to biotin
over a wide pH range from 5 to 8. Therefore, these results lead us receptor-positive tumor cells. Probe 6 was treated with biotin receptor-
to suggest that the biotin-coumarin conjugates 6 can be used to positive A549 and biotin receptor-negative WI-38 cells. The A549 cells
sense the existence of cellular thiols without interference from pH with high levels of the biotin receptor, exhibit strong fluorescence
effects in biological environment.
intensity within 20 min of incubation with 6 (5.0 mM), whereas very little
To rationalize the fluorescence enhancement of 6 after the cleavage or no fluorescence signal was observed in the biotin receptor-negative
of vulnerable S–S bond, we performed time dependent fluorescence WI-38 cancer cells, under similar experimental conditions, as shown in
spectra with a different GSH concentration (Fig. S3, ESI†). In the Fig. 3. We performed another cell experiments where 6 and compound
presence of 5.0 mmol GSH, the enhancement of fluorescence intensity 4 without biotin were treated in the two cell lines (A549 and WI38),
of 6 reached a saturation point within 15 min. It was observed that the Fig. S4 (ESI†). In these two experiments we found that noticeable
rate of reaction is faster when we used 5.0 mmol GSH rather than that difference in the fluorescence intensity inside the cells only in the
of 3.0 mmol, 1.0 mmol GSH. This observation suggested that free presence of the biotin group. Therefore, we concluded that the selective
coumarin units could be generated within a short span of time once it uptake of 6 by A549 cells can be accredited by the interaction of its
entered into the cells. In contrast, the fluorescence intensity of 6 was biotin moiety with a biotin receptor on the tumor cells, possibly
not enhanced in the absence of GSH.
through receptor-mediated endocytosis.
We performed another set of investigations, to justify whether the
To provide further confirmation for thiol-induced disulfide bond
fluorescence enhancement of 6 as shown in Fig. 1, is an indication of cleavage as well as concomitant fluorescence-On, probe 6 was treated
free coumarin unit formation. Probe 6 was treated with 5.0 mmol with NEM (N-ethylmaleimide). NEM is well known to react with thiol
GSH at 37 1C for 2 h, then aliquot was subjected to mass analysis. groups.12 Therefore, in the presence of this thiol inhibitor, the
From the mass spectra, the main peak of the biotin moiety ([M + Na] = fluorescence intensity of cells treated with 6 was expected to be
539.10) (Fig. S21, ESI†) was observed. Together, these results lead us to relatively decreased. The results are shown in Fig. 4. It is obvious
suggest that the disulfide bond present in 6 was cleaved upon the that the fluorescence intensity of 6 in A549 cells decreases as the
treatment of GSH and thiol related enzymes. This cleavage occurred concentration of NEM is increased (0–0.5 mM). This phenomenon
through an intramolecular nucleophilic substitution at the neigh- provided another confirmation that the fluorescence emission inten-
bouring carbonate bond to form a 5-membered thio lactone ring. sity of 6 depends on the intracellular thiol levels.
This, in turn, will lead to the release of free coumarin units as well as
To prove our design expectation as well as receptor mediated
fluorescent intensity increases which are described in Scheme 2. This endocytosis process, we carried out the cell viability experi-
type of self-emulative disulfide linker can be readily used in a variety ments of compound 6 with biotin positive A549 cells and biotin
of cancer-specific drug conjugates suitable for further applications in negative WI 38 cells (Fig. S5, ESI†). We observed that at low to
biological systems.
high concentration 6 showed less cell viability in the A549 cells
Fig. 3 The left side panels show the confocal microscopy images of A549
and WI-38 cells treated with 5.0 mM of 6 in PBS buffer, total incubation
period of 20 min. The right side panels show contrast images. Cell images
were obtained using two photon excitation wavelengths at 740 nm and
emission wavelengths of 400–500 nm.
Scheme 2 Reaction mechanism of 6 with thiols under physiological
conditions.
3046 | Chem. Commun., 2014, 50, 3044--3047
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
Fig. 5 Confocal laser fluorescence microscopic images of A549 cells incu-
bated with 5.0 mM of 6 containing (b) Mito tracker deep red (0.5 mM) excited at
633 nm, collected at 650–750 nm; (e) ER-tracker red (0.5 mM) excited at 514 nm,
collected at 570–600 nm; (h) Lyso tracker red DND-99 (1.0 mM) excited at
514 nm, collected at 570–650 nm. (a, d, g) are the fluorescence images of 6
(5 mM) for 20 min at 37 1C, excited at 740 nm, collected at 400–500 nm. (c, f, i)
overlay of the merged images, respectively.
Fig. 4 Confocal laser fluorescence microscopic images of A549 cells
treated with 5.0 mM of 6 in PBS buffer. The cells were pre-incubated with
media containing N-ethylmaleimide (NEM) of various concentrations (0,
0.2, and 0.5 mM) for 30 min at 37 1C. Cell images were obtained using two
photon excitation wavelengths of 740 nm, and emission wavelengths of
400–550 nm, green signal, respectively.
rather than WI38 cells. This is because 6 can easily enter into
the high levels of the biotin receptor positive A549 cells through
the strong interaction between avidin and biotin. Together
these results provide support that cellular uptake of probe 6
into A549 cells takes place through receptor-mediated endo-
cytosis and gives rise to fluorescence enhancement as the result
of thiol-induced disulfide bond cleavage. Therefore, it is clearly
confirmed again that the biotin moiety in 6 obviously plays as a
targeting unit to tumor cell in this fluorescence probe.
powerful new approach for the specific tumor targeting thiol
sensor and bio imaging. In addition this system may have a role
in opening new possibilities for the screening of new potential
drug agents, diagnosis and bioimaging applications in vitro
and in vivo.
J. S. K acknowledges the CRI project (2009-0081566) from the
National Research Foundation of Korea.
Notes and references
To recognize the intracellular location of coumarin moiety released
from 6, after get into the cells, colocalization experiments were
performed using selective fluorescent trackers like as mitochondria
(Mito-), lysosome (Lyso-) and endoplasmic reticulum (ER). The results
from these studies are shown in Fig. 5. The fluorescence image of
probe 6 was mainly overlapped with Mito-, ER markers. On the other
hand, the fluorescence describable to 6 does not exactly co-localize with
Lyso Tracker. As the ER membrane is closely associated with the inner
nuclear membrane (INM), it may be assumed that molecules intended
for the INM diffuse through the ER membrane.13 We, therefore,
conclude that thiol mediated disulfide cleavage of 6 occurs in the
mitochondria as well as ER; this scission process assists in the release
of the free coumarin moiety and concomitant fluorescence-On.
In conclusion, we have designed and synthesized novel coumarin–
disulfide–biotin conjugates (6) for the precise detection of intercellular
thiols including GSH, Cys and Hcy and its cell viability as well as
biological applications in vivo. Upon addition of thiols, the fluores-
cence intensity of 6 increases by about 5-fold. Moreover, probe 6
shows good water solubility, is inactive towards other biologically
relevant analytes such as various thiols, thiol-free amino acids,
monovalent, and bivalent metal ions, is easy-to-visualize, and is
reactive over a wide pH range. From confocal microscopic experi-
mental studies, we revealed that our system goes to precisely biotin
1 (a) X. Chen, Y. Zhou, X. Peng and J. Yoon, Chem. Soc. Rev., 2010,
39, 2120; (b) C. Hwang, A. J. Sinskey and H. F. Lodish, Science, 1992,
257, 1496.
2 S. C. Lu, Curr. Top. Cell. Regul., 2000, 36, 95.
3 D. M. Townsend, K. D. Tew and H. Tapiero, Biomed. Pharmacother.,
2003, 57, 145.
4 K. A. Conklin, Integr. Cancer Ther., 2004, 3, 294.
5 (a) J. S. Kim and D. T. Quang, Chem. Rev., 2007, 107, 3780;
(b) S. H. Lee, J. Y. Kim, J. Ko, J. Y. Lee and J. S. Kim, J. Org. Chem.,
2004, 69, 2902; (c) Y. H. Lee, M. H. Lee, J. F. Zhang and J. S. Kim,
J. Org. Chem., 2010, 75, 7159; (d) H. J. Kim and J. S. Kim, Tetrahedron
Lett., 2006, 47, 7051; (e) J. S. Kim, H. J. Kim, H. M. Kim, S. H. Kim,
J. W. Lee, S. K. Kim and B. R. Cho, J. Org. Chem., 2006, 71, 8016;
( f ) K. C. Ko, J.-S. Wu, H. J. Kim, P. S. Kwon, J. W. Kim, R. A. Bartsch,
J. Y. Lee and J. S. Kim, Chem. Commun., 2011, 47, 3165.
6 (a) H. S. Jung, X. Q. Chen, J. S. Kim and J. Yoon, Chem. Soc. Rev.,
2013, 42, 6019; (b) J. H. Lee, C. S. Lim, Y. S. Tian, J. H. Han and
B. R. Cho, J. Am. Chem. Soc., 2010, 132, 1216.
7 A. C. Grimsdale, K. L. Chan, R. E. Martin, P. G. Jokisz and
A. B. Holmes, Chem. Rev., 2009, 109, 897.
8 S. Maiti, N. Park, J. H. Han, H. M. Jeon, J. H. Lee, S. Bhuniya, C. Kang
and J. S. Kim, J. Am. Chem. Soc., 2013, 135, 4567.
9 H. C. Kim, S. Kreiling, A. Greiner and N. Hampp, Chem. Phys. Lett.,
2003, 372, 899.
10 F. Friscourt, C. J. Fahrni and G. J. Boons, J. Am. Chem. Soc., 2012,
134, 18809.
11 C. Wang, Z. Li, D. Cao, Y. L. Zhao, J. W. Gaines, A. Bozdemir,
M. W. Ambrogio, M. Frasconi, Y. Y. Botros, J. I. Zink and
J. F. Stoddart, Angew. Chem., Int. Ed., 2012, 51, 5460.
receptor-positive A549 cells rather than WI 38 cells (biotin negative) 12 J. D. Gregory, J. Am. Chem. Soc., 1955, 77, 3922.
13 S. Saksena, Y. Shao, S. C. Braunagel, M. D. Summers and
through receptor mediated endocytosis. Furthermore, the fluores-
cence image of probe 6 was mainly overlapped with Mito-, ER
markers. Therefore, our probe, described above, could afford a
A. E. Johnson, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 12537.
14 G. Russell-Jones, K. McTavish, J. McEwan, J. Rice and D. Nowotnik,
J. Inorg. Biochem., 2004, 98, 1625.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 3044--3047 | 3047