Rational design of fluorescent light-up probes based on an AIE luminogen for targeted intracellular thiol imaging

Article abstract of DOI:10.1039/c3cc47585a

A water-soluble fluorescent light-up bioprobe based on a luminogen with aggregation-induced emission characteristics was developed for targeted intracellular thiol imaging.

Full text of DOI:10.1039/c3cc47585a

ChemComm
View Article Online
View Journal
COMMUNICATION
Rational design of fluorescent light-up probes
based on an AIE luminogen for targeted
intracellular thiol imaging†
Cite this:DOI: 10.1039/c3cc47585a
Received 3rd October 2013,
Accepted 29th October 2013
Youyong Yuan,‡a Ryan T. K. Kwok,‡b Guangxue Feng,a Jing Liang,a Junlong Geng,a
Ben Zhong Tang*bc and Bin Liu*ad
DOI: 10.1039/c3cc47585a
www.rsc.org/chemcomm
A water-soluble fluorescent light-up bioprobe based on a lumino- reducing energy/charge transfer due to the cleavage of the disulfide
gen with aggregation-induced emission characteristics was devel- bond upon exposure to thiols.6b Some of these probes have been used
oped for targeted intracellular thiol imaging.
for intracellular thiol imaging, but lack cell specificity. In addition, the
traditional fluorophores used for these probe designs often show
Cellular thiols are essential biomolecules that play a critical role in aggregation-caused quenching (ACQ), which reduces their brightness
many biological processes including antioxidant defense, cell signal- and sensitivity in cellular imaging.7 Considering the importance of
ing and cell proliferation.1 Among them, glutathione (GSH) is the thiols in biological functions, it is highly desirable to develop simple,
most abundant (1–10 mM) thiolated tripeptide found within the noninvasive, and specific probes with high signal-to-noise ratios for
human cellular system.2 It is reported that the changes in the cellular targeted imaging of intracellular thiol levels.
GSH level are associated with many health problems such as
Aggregation-induced emission (AIE), a unique phenomenon that
Alzheimer’s disease, leucocyte loss, psoriasis, liver damage and is exactly opposite to the ACQ effect, has emerged as a powerful and
cancer.3 In addition, GSH levels in cancer cells also affect the versatile strategy for the design of novel fluorescent probes.8 Propeller-
effectiveness of chemotherapy.4 Many intelligent drug or gene delivery shaped luminogens, such as tetraphenylethene (TPE) and silole
systems have been developed by taking advantages of the differences derivatives, are non-emissive when molecularly dissolved but are
between intracellular and extracellular thiol concentrations in cancer induced to emit efficiently upon aggregation.8b We have rationalized
cells.5 Detection of intracellular thiol levels is thus of high importance the AIE mechanism as restriction of intramolecular rotations (RIR),
for early diagnosis of diseases, evaluation of disease progression and which prohibits energy dissipation via non-radiative channels, leading
therapy efficiency of new potential drugs.
In recent years, a number of fluorescent sensors have been AIE luminogens particularly attractive for biological applications.8c
developed for the detection of free or intracellular thiols.6 One design
In this contribution, we designed an integrin avb3 targeted
to high quantum yields in aggregated states.8 These advantages make
strategy is to utilize the strong nucleophilicity of the thiols to react light-up probe for cell specific intracellular thiol imaging. The probe is
with unsaturated carbon in electrophiles through thiol-addition composed of a targeted cyclic RGD (cRGD) peptide, a highly water
reactions.6a Many of these probes have shown poor water solubility soluble peptide with five aspartic acids (Asp, D5), a TPE fluorogen and a
or high background signal with limited biological applications. thiol-specific cleavable disulfide linker. cRGD exhibits high binding
Another design is based on dual-labeled fluorescent probes, and the affinity towards avb3 integrin which is a unique molecular biomarker
signal is generated upon separation of dye–quencher pairs or for early detection and treatment of rapidly growing solid tumors. The
probe is highly water soluble and is almost non-fluorescent in aqueous
a Department of Chemical and Biomolecular Engineering, 4 Engineering Drive 4,
media. The cleavage of the disulfide group by thiols leads to enhanced
fluorescence signal output (Scheme 1). This probe has the potential for
real-time monitoring of thiol levels in specific tumor cells.
National University of Singapore, Singapore 117576
b Department of Chemistry, Division of Biomedical Engineering, The Hong Kong
University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China
TPE is an iconic AIE fluorogen.8a It is non-emissive in the
molecularly dissolved state, but shows intense fluorescence in
the aggregates. It is known that water solubility is a prerequisite for
bioprobes, which is typically achieved by attaching hydrophilic
groups to the organic fluorophores. Recently, we have developed
AIE luminogen based light-up probes for monitoring of cellular
proteins.9 The first generation of probe design is limited to peptide
c SCUT-HKUST Joint Research Laboratory, Guangdong Innovative Research Team,
State Key Laboratory of Luminescent Materials and Devices, South China
University of Technology, Guangzhou 510640, China. E-mail: tangbenz@ust.hk
d Institute of Materials Research and Engineering, 3 Research Link, Singapore
117602. E-mail: cheliub@nus.edu.sg
† Electronic supplementary information (ESI) available: Detailed experimental
methods and additional data. See DOI: 10.1039/c3cc47585a
‡ These authors contributed equally.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun.

View Article Online
Communication
ChemComm
Fig. 1 (A) PL spectra of TPE-CH2NH2 and TPE-SS-D5-cRGD in DMSO–PBS
(v/v = 1 : 199). Inset: the corresponding photographs taken under illumina-
tion of a UV lamp. (B) Time-dependent PL spectra of TPE-SS-D5-cRGD
treated with GSH. (C) PL spectra of TPE-SS-D5-cRGD (1.0 mM) in the
presence of different concentrations of GSH. (D) Plot of PL intensity at
470 nm versus concentrations of GSH (mean Æ SD, n = 3).
Scheme 1 (A) General probe design strategy and (B) schematic illustra-
tion of cRGD targeted imaging of intracellular thiols through avb3 integrin
mediated cellular uptake and cleavage of the disulfide bond to induce
fluorescence ‘‘turn on’’. (C) Chemical structure of the probe.
based recognition elements with high water solubility. To extend
the design principle to include a broader range of hydrophobic the concentration range of 0 to 960 mM (Fig. S13, ESI†). Its PL
recognition elements, it is necessary to develop a more general profile also does not change in the presence of the commonly used
strategy. In this work, a thiol specific cleavable disulfide linker cell culture medium, Dulbecco’s Modified Eagle Medium (DMEM).
dithiobis(succinimidyl propionate) (DSP) was utilized to conjugate The probe maintains an ‘‘off’’ state in the complex environment and
aminated TPE and a cRGD using D5 to ensure good water solubility thus has great potential to serve as a specific light-up probe with
of the probe. We found that the incorporation of D5 can efficiently minimum background interference.
transfer TPE into a non-emissive molecular species in aqueous
To study the response of the probe to free thiols, GSH was chosen
media. Using intracellular thiol imaging as an example, we here as the representative thiol due to its high concentration in the human
report a general platform for designing specific light-up probes for cellular system.1 GSH (1 mM) was incubated with 10 mM TPE-SS-D5-
imaging of intracellular molecules or cellular processes.
cRGD in DMSO–PBS mixtures (v/v = 1/199), and the fluorescence
Amine-functionalized TPE (TPE-CH2NH2) was synthesized by redu- spectra were measured at different time points. As shown in Fig. 1B,
cing azide-TPE (TPE-CH2N3) in methanol (Scheme S1 and Fig. S1–S3, the emission intensity of TPE-SS-D5-cRGD increases significantly with
ESI†). Asymmetric functionalization of a DSP linker with TPE-CH2NH2 time, reaching the maximum within 3 h, which is 68-fold higher than
and NH2 terminated D5-cRGD in the presence of N,N-diisopropyl- the intrinsic emission of the probe. The TPE-SS-D5 shows a similar
ethylamine (DIEA) in anhydrous dimethyl sulfoxide (DMSO) afforded time dependent fluorescence increase after incubation with GSH
the probe TPE-SS-D5-cRGD in 45% yield (Scheme S2, ESI†). A control (Fig. S14A, ESI†), while a negligible signal is observed for TPE-CC-D5
probe TPE-SS-D5 with a similar structure but without a cRGD moiety (Fig. S14B, ESI†). TPE-SS-D5-cRGD is further demonstrated to
was also synthesized in 49% yield (Scheme S3, ESI†). In addition, a respond to GSH under acidic conditions (Fig. S15, ESI†).
non-activatable control probe TPE-CC-D5 was prepared in 44% yield by
Next, we investigated the effect of GSH concentration on the
using disuccinimidyl suberate to replace DSP in the coupling reaction emission of the probe. GSH at different concentrations ranging
(Scheme S4, ESI†). NMR and MS characterization confirmed the right from 3.9 mM to 1.0 mM was incubated with TPE-SS-D5-cRGD for 3 h,
structures with high purity of the three probes (Fig. S4–S12, ESI†).
and the corresponding spectra are shown in Fig. 1C. With increasing
The photoluminescence (PL) spectra of TPE-CH2NH2 and GSH concentration, the fluorescence is gradually intensified due to the
TPE-SS-D5-cRGD in DMSO and phosphate buffered saline (PBS, increased amount of TPE aggregates formed in aqueous media. The
pH = 7.4) mixtures (v/v = 1/199) are shown in Fig. 1A. The hydro- molecular dissolution of the probe and the aggregate formation of
phobic TPE-CH2NH2 shows intense fluorescence as nanoaggregates the cleaved products were confirmed by laser light scattering (LLS)
with a quantum yield (F) of 0.23 Æ 0.01 when using quinoline measurements. In the aqueous mixture, no LLS signals could be
sulfate as the standard.10 The TPE-SS-D5-cRGD probe is almost detected from the solution of TPE-SS-D5-cRGD. However, after incuba-
non-fluorescent in the same medium (F = 0.001), due to tion with GSH, the residual hydrophobic AIE luminogen tends to
the easy intramolecular rotations of the TPE phenyl rings in cluster into aggregates (Fig. S16A, ESI†). The formation of aggregates
aqueous media. The significant difference in the PL intensities was further confirmed by AFM (Fig. S16B, ESI†). Under the same
of TPE-CH2NH2 and TPE-SS-D5-cRGD offers opportunity for experimental conditions, a similar increase in fluorescence intensity is
the probe to be used for specific light-up imaging of thiols. observed for TPE-SS-D5 (Fig. S17A, ESI†), but not for TPE-CC-D5
The PL spectrum of TPE-SS-D5-cRGD shows no response to NaCl in (Fig. S17B, ESI†). In addition, plotting the PL intensities at
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
470 nm for TPE-SS-D5-cRGD against the GSH concentration gives a no detectable fluorescence was observed even after incubation for 6 h
perfect linear line (Fig. 1D), suggesting the possibility of using the (Fig. 2C and F). It should be noted that the probe could also be used for
probe for GSH quantification with a detection limit of 1.0 mM.
live cell imaging and the images are shown in Fig. S21 (ESI†).
To monitor the GSH-induced fluorescence activation of TPE-
To provide further evidence for thiol-induced disulfide bond clea-
SS-D5-cRGD, reverse-phase HPLC and MS analyses were used to vage as the trigger of fluorescence turn-on, the U87-MG cells were also
monitor the exposure of the probe to GSH. After incubation of pretreated with buthionine sulfoximine (BSO) before incubation with
TPE-SS-D5-cRGD with GSH for 3 h, the mixture was subjected to TPE-SS-D5-cRGD. BSO is an inhibitor of g-glutamylcysteine synthetase
HPLC analysis. As shown in Fig. S18 (ESI†), in addition to the which can prevent the cells from synthesizing GSH.11 As shown in
TPE-SS-D5-cRGD peak observed at 10.83 min, two new peaks at Fig. S22 (ESI†), the fluorescence of TPE-SS-D5-cRGD treated U87-MG
10.68 min for GSS-TPE and 11.58 min for TPE-SH are observed and cells decreases as the concentration of BSO increases from 25 to
the peaks show mass-to-charge ratios (m/z) of 755.217 and 472.164 100 mM. The significantly reduced fluorescence shown in Fig. S22C
analyzed by IT-TOF, respectively. The fragments of TPE-SH and GSS- (ESI†) as compared to that shown in Fig. 2A reveals that the probe
TPE tend to aggregate in DMSO–PBS (v/v = 1/199), which show blue fluorescence is directly related to GSH concentration in the cells. These
fluorescence with quantum yields of 19 Æ 1% and 12 Æ 1%, results indicate that despite the existence of other free thiols in cells,
respectively, using quinoline sulfate as a reference (Table S1, ESI†). TPE-SS-D5-cRGD could be used as an indicator for intracellular GSH
These results clearly demonstrate that the observed GSH-induced imaging. In vitro cytotoxicity studies also show that the TPE-SS-D5-cRGD
fluorescence intensity change of TPE-SS-D5-cRGD is due to cleavage of probe is biocompatible (Fig. S23, ESI†).
the disulfide bond, which leads to solubility difference between the
In conclusion, we report the synthesis and biological applications of
probe and the fragment. Further titration of TPE-SS-D5-cRGD with a light-up GSH responsive AIE probe. Thanks to the unique nature of
cysteine (Cys), glycine (Gly) and glutamate (Glu), the three amino acids the AIE luminogen, the probe is non-fluorescent in aqueous media but
present in GSH, reveals that the fluorescence turn-on is due to the becomes highly emissive when cleaved by thiols. The probe enables
interaction of free thiol in Cys with the disulfide bond (Fig. S19, ESI†). light-up monitoring of free thiols in solution and in cells with a high
To explore the capability of TPE-SS-D5-cRGD as a specific bioprobe signal-to-noise ratio. The cRGD functionalized peptide allows for
for monitoring intracellular thiol levels in cancer cells, the probe is selective targeting of avb3 integrin of many angiogenic cancers using
incubated with U87-MG human glioblastoma and MCF-7 breast cancer U87-MG as an example, which opens a new opportunity for specific
cell lines. The confocal imaging results are shown in Fig. 2. U87-MG intracellular thiol imaging. Our AIE probe strategy can be generalized to
cells with overexpressed integrin avb3 on the cellular membrane perform various tasks by simply changing the disulfide groups with
were chosen as integrin-positive cancer cells, while breast cancer cells other cleavable linkers in chemical biology. Further development
(MCF-7) with a low level of integrin avb3 expression were used as the of AIE fluorogens with long wavelength emission will facilitate
negative control. After incubation with TPE-SS-D5-cRGD, a strong blue the development of specific bioprobes for in vivo applications.
fluorescence is observed for U87-MG cells (Fig. 2A), whereas for MCF-7 The probe design thus opens new avenues for the construction
cells only a weak fluorescence signal could be found even after of various selective targeting probes for diagnosis, imaging and
incubation for 6 h (Fig. 2D). In contrast, TPE-SS-D5 displays weak drug screening applications.
fluorescence intensity with essentially identical behavior for both cell
We thank the Singapore National Research Foundation (R-279-
lines (Fig. 2B and E). When U87-MG cells were pretreated with free 000-390-281), the SMART (R279-000-378-592), the Research Grants
cRGD prior to TPE-SS-D5-cRGD incubation, weak fluorescence was Council of Hong Kong (HKUST2/CRF/10 and N_HKUST620/11) and
observed (Fig. S20, ESI†). The marked difference reveals that the Guangdong Innovative Research Team Program (201101C0105067115).
selective uptake of TPE-SS-D5-cRGD by U87-MG cells is due to the
Notes and references
integrin receptor-mediated process. For the control probe TPE-CC-D5,
1 C. Hwang, A. J. Sinskey and H. F. Lodish, Science, 1992, 257, 1496.
2 A. Meister, J. Biol. Chem., 1988, 263, 17205.
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 S. Bauhuber, C. Hozsa, M. Breunig and A. Gopferich, Adv. Mater.,
2009, 21, 3286.
6 (a) C. X. Yin, F. J. Huo, J. J. Zhang, M. M. Ramon, Y. T. Yang, H. G. Lv
and S. D. Li, Chem. Soc. Rev., 2013, 42, 6032; (b) H. S. Jung,
X. Q. Chen, J. S. Kim and J. Yoon, Chem. Soc. Rev., 2013, 42, 6019.
7 A. C. Grimsdale, K. L. Chan, R. E. Martin, P. G. Jokisz and
A. B. Holmes, Chem. Rev., 2009, 109, 897.
8 (a) Y. N. Hong, J. W. Y. Lam and B. Z. Tang, Chem. Commun., 2009, 4332;
(b) Y. N. Hong, J. W. Y. Lam and B. Z. Tang, Chem. Soc. Rev., 2011,
40, 5361; (c) D. Ding, K. Li, B. Liu and B. Z. Tang, Acc. Chem. Res., 2013,
DOI: 10.1021/ar3003464; (d) J. Geng, K. Li, D. Ding, X. Zhang, W. Qin,
J. Liu, B. Tang and B. Liu, Small, 2012, 8, 3655; (e) F. Wang, J. Wen,
L. Huang, J. Huang and O. Jin, Chem. Commun., 2012, 48, 7395.
9 H. B. Shi, R. T. K. Kwok, J. Z. Liu, B. G. Xing, B. Z. Tang and B. Liu,
J. Am. Chem. Soc., 2012, 134, 17972.
10 J. N. Demas and G. A. Crosby, J. Phys. Chem., 1971, 75, 991.
Fig. 2 Confocal microscopy images of U87-MG (A–C) and MCF-7 (D–F)
cells after incubation with TPE-SS-D5-cRGD (A, D), TPE-SS-D5 (B, E) and
TPE-CC-D5 (C, F). The nuclei were stained with propidium iodide. All images
share the same scale bar (20 mm).
11 M. Hultberg and B. Hultberg, Chem.-Biol. Interact., 2006, 163, 192.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun.