Letter pubs.acs.org/ac
Intracellular Proteolytic Disassembly of Self-Quenched Near-Infrared Nanoparticles Turning Fluorescence on for Tumor-Targeted Imaging Jinhui Jiang,† Zhibin Zhao,‡ Zijuan Hai,† Hongyong Wang,⊥ and Gaolin Liang*,† †
CAS Key Laboratory of Soft Matter Chemistry, Department of Chemistry, University of Science and Technology of China, 96 Jinzhai Road, Hefei, Anhui 230026, China ‡ Liver Immunology Laboratory, Institute of Immunology and School of Life Sciences, University of Science and Technology of China, Hefei, Anhui 230027, China ⊥ Jiangsu Institute of Nuclear Medicine, Wuxi, Jiangsu 214063, China S Supporting Information *
ABSTRACT: The design of tumor-targeting, intracellular protease-activatable near-infrared fluorescence (NIRF) nanoprobes is broadly interesting but remains challenging. In this work, we report the rational design of a NIR probe Cys(StBu)Lys(Biotin)-Lys-Lys(Cy5.5)-CBT (1) to facilely prepare the self-quenched nanoparticles 1-NPs for tumor-targeted imaging in vitro and in vivo. The biotinylated 1-NPs could be actively uptaken by biotin receptor-overexpressing tumor cells via receptor-mediated endocytosis. Upon intracellular proteolytic cleavage, 1-NPs were disassembled to yield the small molecular probe Lys(Cy5.5)-Luciferin-Lys(Biotin)-Lys-OH (1-D-cleaved), accompanied by fluorescence “Turn-On”. With this NIRF “Turn-On” property, 1-NPs were successfully applied for tumor-targeted imaging. We envision that our nanoparticles could be applied for fluorescence-guided tumor surgery in the near future.
T
tumor imaging.13−15 Recently, aggregation-induced emission (AIE) effect was also utilized to turn “On” the fluorescence for tumor imaging.16 Nanoparticles (NPs) have large surface area to volume ratio and are able to load a large amount of fluorophores, rendering very high sensitivity for tumor imaging.17,18 Moreover, the enhanced permeability and retention (EPR) effect of nanostructures assures that NPs realize passive transport tumor-targeted imaging even without a tumor-targeting warhead.19 Compared with inorganic nanoprobes, organic NIR nanoprobes are emerging as very attractive candidates for tumor imaging due to their better biocompatibility and easier modification.20 Recently, Liang and co-workers developed a self-quencher NIR nanoparticle probe for highly sensitive imaging of furin-overexpressing tumor.12 This NP probe was not functionalized with tumor-targeting warhead and therefore could not distinguish tumor cells from normal cells before entering cells. Its tumor-targeting property was realized by furin-cleaved fluorescence “Turn-On” in furin-overexpressing tumor cells after it was internalized by either normal cells or tumor cells. Inspired by the above poineering studies, as shown in Figure 1, in this work, we rationally designed one type of tumortargeting, self-quenched, organic NIRF nanoparticle probe (i.e.,
umor imaging is important for cancer diagnosis and has progressed fast in recent decades due to the advances in imaging devices and probes.1 Several imaging modalities including magnetic resonance,2 nuclear,3 ultrasonography,4 photoacoustic,5 and Raman spectroscopy6 have been utilized for tumor imaging. Compared with the above imaging approaches, near-infrared fluorescence (NIRF) imaging has the advantages of high sensitivity, low cost, easy operation, and simple instrumentation.7 Moreover, compared with shorter emission wavelength fluorescence probes, NIRF probes have higher tumor penetrating ability due to their reduced tissue absorbance and autofluorescence at NIR wavelengths between 700 and 900 nm.8 Specifically, once the NIR probes are modified with tumor-targeting warheads, they could be promisingly applied for tumor imaging with high sensitivity and selectivety.9 Compared with the probes whose fluorescence “Always On” or “Turn-Off”, fluorescence “Turn-On” probes have lower background signal and thus higher signal-to-noise ratio (or sensitivity) for tumor imaging.10,11 Most of the “Turn-On” probes use the fluorescence resonance energy transfer (FRET) effect to quench their fluorescence before it turns “‘On’” at the tumor site.12 For instance, if the fluorophore of a “Turn-On” probe is linked with the quencher by an enzymatic peptide substrate, its fluorescence remains “Off’’. At the enzymeoverexpressing tumor site, the linker is cleaved by the enzyme and the quencher (acceptor) is separated from the fluorophore (donor); thus, the fluorescence of the probe is turned “‘On’” for © 2017 American Chemical Society
Received: July 27, 2017 Accepted: September 6, 2017 Published: September 6, 2017 9625
DOI: 10.1021/acs.analchem.7b02971 Anal. Chem. 2017, 89, 9625−9628
Letter
Analytical Chemistry
Figure 1. (A) Schematic illustration of reduction-controlled self-assembly of 1-NPs to turn NIRF signal “Off” and subsequent protease-controlled disassembly of 1-NPs to turn NIRF signal “On”. (B) Cartoon illustration of internalization of the biotinylated, self-quenched 1-NPs by biotin receptor-overexpressing tumor cells and intracellular proteolytic disassembly of the nanoparticles to turn the NIR fluorescence “On” for tumortargeted imaging.
1-NP) whose fluorescence was turned “On” upon enzymatic disassembly of the nanoparticles in targeted tumor cells for highly sensitive imaging of a tumor. To facilely prepare the nanoparticles, we rationally designed a precursor Cys(StBu)Lys(Biotin)-Lys-Lys(Cy5.5)-CBT (1), as shown in Figure 1A. In the presence of reducing agent (e.g., tris(2-carboxyethyl)phosphine, TCEP) and at neutral condition, the disulfide bond of 1 was reduced and the N-terminal Cys motif condensed with the cyano group of another monomer 1 to yield cyclized dimer (i.e., 1-Dimer, Figure S1) which self-assembled into the biotinylated self-quenched nanoparticles 1-NPs, as we demonstrated before.21 After being internalized by the biotin receptoroverexpressing tumor cells, as shown in Figure 1B, the amide bond of the Lys-Lys(Cy5.5) motif of 1-D in 1-NPs was specifically cleaved by the intracellular robust proteases (e.g., trypsin),22 and consequently, the 1-NPs were disassembled to yield Lys(Cy5.5)-Luciferin-Lys(Biotin)-Lys-OH (1-D-cleaved, structure shown in Figure 1A); the fluorescence was turned “On”. By this means, the biotin receptor-overexpressing tumors were specifically and highly sensitively imaged by the NIR fluorophore Cy5.5 in 1-NPs. We began the study with the synthesis of compound 1 (Scheme S1). Briefly, the peptide sequence Fmoc-Cys(StBu)Lys(Biotin)-Lys(Dde)-Lys(Boc)-OH (A) with protecting groups was synthesized with solid phase peptide synthesis (SPPS) and then coupling A with 2-cyano-6-aminobenzothiazole (CBT) in the presence of isobutyl chloroformate to yield B. Deprotection of the Boc group of B yielded C after highperformance liquid chromatography (HPLC) purification (Figure S2). Coupling C with Cy5.5-NHS ester yielded D after HPLC purification. Then, the Fmoc and Dde protecting groups on D were sequentially removed to yield compounds E and 1 after HPLC purification, respectively. After characterizing the pure compound 1 (Figures S3 and S4), we first used TCEP to trigger its condensation to assemble 1-NPs and applied asformed 1-NPs for NIR fluorescence detection of protease in vitro. Ten μM 1 in the phosphate-buffered saline (PBS)
containing 40% DMSO (v/v) (pH 7.4) was incubated with 10fold TCEP at 37 °C for 1 h. After that, the reaction mixture was centrifuged, and 1-NPs obtained by centrifugation were redispersed in the same volume of the PBS containing 10% DMSO for fluorescence measurements. Compared with the fluorescence emission at 710 nm of 1, NIR fluorescence emission of 1-NPs dropped to its 1/9.5 (Figure 2A), suggesting efficient self-quenching of the fluorescence after nanoparticle formation. After the 1-NPs dispersion was incubated with trypsin at 37 °C for 6 h, the fluorescence emission at 710 nm of 1-NPs increased 6.2-fold, suggesting disassembly of the nanoparticles by the enzyme and dequenching of the
Figure 2. (A) Fluorescence spectra of 10 μM 1 (blue), 1-NPs dispersion (black), and 1-NPs dispersion incubated with 10 pmol U−1 trypsin at 37 °C for 6 h in trypsin working buffer (red). Excitation: 675 nm. (B) TEM image of 1-NPs dispersion. (C) TEM image of 1-NPs dispersion after being incubated with 10 pmol U−1 trypsin at 37 °C for 6 h. (D) HPLC traces of 1 (black), 1-NPs dispersion (red), and the incubation mixture of 1-NPs dispersion after incubation with 10 pmol U−1 trypsin at 37 °C for 6 h (blue). Absorbance: 675 nm. 9626
DOI: 10.1021/acs.analchem.7b02971 Anal. Chem. 2017, 89, 9625−9628
Letter
Analytical Chemistry fluorescence (Figure 2A). Transmission electron microscopy (TEM) images clearly showed the TCEP-controlled selfassembly and subsequent trypsin-controlled disassembly of 1NPs (Figure 2B,C). Statistic analysis indicated that as-formed nanoparticles (i.e., 1-NPs) have an average diameter of 55.3 ± 14.2 nm (Figures 2B and S5). Since the reduction intermediate of 1 (i.e., 1-Red) is more hydrophilic than 1, we could conclude that there was no nonspecific precipitation of 1-Red in the TEM images of 1-NPs. HPLC traces, together with mass spectra, clearly indicated that 1-NPs were composed of the condensation product of 1 (i.e., 1-Dimer) (red trace in Figures 2D and S6), and the protease trypsin efficiently digested the substrate 1-Dimer to yield 1-D-cleaved (blue trace in Figures 2D and S7). After confirming that 1-NPs could be efficiently digested by protease trypsin to turn the fluorescence “On”, we applied them for tumor cell-targeted imaging. Both robust protease (e.g., cathepsin B) and biotin receptor-overexpressing HeLa cells were selected for this purpose.23 Before that, cytotoxicity of 1NPs on HeLa cells was investigated and the results indicated that, up to 48 h at 2.8 mg/L, 1-NPs did not induce obvious cytotoxicity on the cells (Figure S8), suggesting that 1-NPs at 1.4 mg/L were safe for cell imaging. Time course cell imaging indicated that, while 1.4 mg/L 1-NPs were not fluorescent in the culture media (t = 0 h), fluorescence of the HeLa cells at 37 °C gradually turned on and reached its intensity plateau after 6 h (Figure S9), suggesting that 1-NPs could be efficiently uptaken by the cells and gradually digested by the intracellular proteases. HPLC analysis confirmed that, at this incubation condition, 1-NPs were efficiently hydrolyzed by the proteases in HeLa cell lysate to yield 1-D-cleaved, which was responsible for the “Turn-On” fluorescence (Figure S10). We thus selected the incubation time of 6 h for the following HeLa cell-targeted imaging studies. As shown in the top row of Figure 3, after the
block the biotin receptors on the HeLa cell membranes and prevent (or reduce) the competitive binding of 1-NPs to the receptors. As shown in the bottom row of Figure 3 and in accord with our expectations, the NIR fluorescence inside the cells was significantly reduced. Moreover, we also conducted the NIR fluorescence imaging of HeLa cells incubated with 1NPs at 4 °C. As shown in Figure S11, at the same incubation times, the NIR fluorescence inside the cells at 4 °C was obviously lower than that in cells at 37 °C in Figure S9, respectively. Quantitative analysis indicated that, within the observation time of 1 to 6 h, the NIR fluorescence of the cells at 37 °C increased much quicker than that of cells at 4 °C (Figure S12). These findings are fully consistent with our hypothesis that 1-NPs could actively target the biotin receptorpositive tumor cells via receptor-mediated endocytosis.24 They also suggest that the biotin receptor on the cell membrane is a good biomarker for tumor cell-targeted imaging. We then applied 1-NPs for tumor-targeted imaging using nude mice xenografted with HeLa tumors in the right thighs. Until the tumor sizes were within 5−10 mm in diameter, the nude mice were randomly divided into 2 groups (n = 3 for each group). Twenty-five μL of 1-NPs at 14 mg/L or 25 μL of 1NPs at 14 mg/L with 10 mM biotin at pH 7.4 were, respectively, injected into each of these tumor-bearing nude mice in each group through tumor-direct injection, and the mice were imaged for 72 h in a small animal imaging system. As shown in Figure 4, fluorescence signals from these two groups
Figure 4. Top row, time-course fluorescence imaging of nude mice xenografted with HeLa tumor cells after tumor-direct injection of 25 μL of 1-NPs at 14 mg/L acquired at 2, 6, 12, 24, 48, and 72 h. Bottom row, time-course fluorescence imaging of nude mice xenografted with HeLa tumor cells after tumor-direct injection of 25 μL of 1-NPs at 14 mg/L containing 10 mM biotin acquired at 2, 6, 12, 24, 48, and 72 h.
Figure 3. Top row, fluorescence and overlay images of biotin receptorpositive HeLa cells after incubation with 1.4 mg/L 1-NPs at 37 °C for 6 h. Bottom row, fluorescence and overlay images of the biotin receptor-positive HeLa cells pretreated with 1 mM biotin at 37 °C for 2 h, followed by incubation with 1.4 mg/L 1-NPs for 6 h. Hoechst 33342 (blue) was used for nuclear counterstaining, and the red fluorescence was from the fluorophore Cy5.5 in 1-NPs. All images have the same scale bar: 10 μm.
of tumors all gradually increased from 2 to 72 h with the fluorescence in the 1-NPs group much stronger than that of the control group (i.e., coinjected with biotin and 1-NPs). Quantitative analysis indicated that, from 2 to 72 h after injection, while the fluorescence intensity in HeLa tumors in the experimental group increased 4.4-fold that in the control group only increased 1.8-fold (Figure S13). These results clearly indicated that our 1-NPs could be effectively uptaken by the tumors via the receptor-mediated endocytosis effect, and their NIR fluorescence could be gradually turned “On” by the protease-overexpressing tumors, offering tumor-targeted NIR imaging in vivo. To further validate that the strong NIR fluorescence in 1-NPs-teated mice was actually emitted from HeLa tumors, after imaging of these mice at 72 h postinjection, we sacrificed the mice and took out the tumors and organs
HeLa cells were incubated with 1.4 mg/L 1-NPs for 6 h, bright NIR fluorescence from cytoplasm was observed, suggesting that the fluorescence-quenched 1-NPs were digested by the robust proteases in the lysosomal compartment and disassembled which induced the fluorescence generation in the cells. To validate that above cell uptake of 1-NPs was induced by the active targeting of biotin on the nanoparticle surface, we pretreat the HeLa cells with an excess of biotin for 2 h prior to incubating them with 1-NPs. This treatment was expected to 9627
DOI: 10.1021/acs.analchem.7b02971 Anal. Chem. 2017, 89, 9625−9628
Letter
Analytical Chemistry
(2) Keca, J. M.; Chen, J.; Overchuk, M.; Muhanna, N.; MacLaughlin, C. M.; Jin, C. S.; Foltz, W. D.; Irish, J. C.; Zheng, G. Angew. Chem., Int. Ed. 2016, 55, 6187. (3) Chen, L.; Zhong, X. Y.; Yi, X.; Huang, M.; Ning, P.; Liu, T.; Ge, C. C.; Chai, Z. F.; Liu, Z.; Yang, K. Biomaterials 2015, 66, 21. (4) Weller, G. E. R.; Wong, M. K. K.; Modzelewski, R. A.; Lu, E. X.; Klibanov, A. L.; Wagner, W. R.; Villanueva, F. S. Cancer. Res. 2005, 65, 533. (5) Pu, K. Y.; Shuhendler, A. J.; Jokerst, J. V.; Mei, J. G.; Gambhir, S. S.; Bao, Z. N.; Rao, J. H. Nat. Nanotechnol. 2014, 9, 233. (6) Kang, J. W.; So, P. T. C.; Dasari, R. R.; Lim, D. K. Nano Lett. 2015, 15, 1766. (7) Zhang, X. L.; Tian, Y. L.; Zhang, C.; Tian, X. Y.; Ross, A. W.; Moir, R. D.; Sun, H. B.; Tanzi, R. E.; Moore, A.; Ran, C. Z. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 9734. (8) Choi, H. S.; Gibbs, S. L.; Lee, J. H.; Kim, S. H.; Ashitate, Y.; Liu, F. B.; Hyun, H.; Park, G.; Xie, Y.; Bae, S.; Henary, M.; Frangioni, J. V. Nat. Biotechnol. 2013, 31, 148. (9) Gonçalves, M. S. T. Chem. Rev. 2009, 109, 190. (10) Dong, L.; Miao, Q. Q.; Hai, Z. J.; Yuan, Y.; Liang, G. L. Anal. Chem. 2015, 87, 6475. (11) Liu, Z. Z.; Wang, B. L.; Ma, Z.; Zhou, Y. B.; Du, L. P.; Li, M. Y. Anal. Chem. 2015, 87, 2550. (12) Yuan, Y.; Zhang, J.; Cao, Q. J. W.; An, L. N.; Liang, G. L. Anal. Chem. 2015, 87, 6180. (13) Zhou, J.; Du, X. W.; Li, J.; Yamagata, N.; Xu, B. J. Am. Chem. Soc. 2015, 137, 10040. (14) Ren, C. H.; Wang, H. M.; Mao, D.; Zhang, X. L.; Fengzhao, Q. Q.; Shi, Y.; Ding, D.; Kong, D. L.; Wang, L.; Yang, Z. M. Angew. Chem., Int. Ed. 2015, 54, 4823. (15) Mu, J.; Liu, F.; Rajab, M. S.; Shi, M.; Li, S.; Goh, C.; Lu, L.; Xu, Q. H.; Liu, B.; Ng, L. G.; Xing, B. G. Angew. Chem., Int. Ed. 2014, 53, 14357. (16) Liu, X. M.; Liang, G. L. Chem. Commun. 2017, 53, 1037. (17) Xing, Y.; Zhao, J. H.; Conti, P. S.; Chen, K. Theranostics 2014, 4, 290. (18) Lee, D. E.; Koo, H.; Sun, I. C.; Ryu, J. H.; Kim, K.; Kwon, I. C. Chem. Soc. Rev. 2012, 41, 2656. (19) Allen, T. M.; Cullis, P. R. Science 2004, 303, 1818. (20) Gao, M.; Yu, F. B.; Lv, C. J.; Choo, J.; Chen, L. X. Chem. Soc. Rev. 2017, 46, 2237. (21) Zheng, Z.; Chen, P. Y.; Xie, M. L.; Wu, C. F.; Luo, Y. F.; Wang, W. T.; Jiang, J.; Liang, G. L. J. Am. Chem. Soc. 2016, 138, 11128. (22) Weissleder, R.; Tung, C. H.; Mahmood, U.; Bogdanov, A. Nat. Biotechnol. 1999, 17, 375. (23) Bhuniya, S.; Maiti, S.; Kim, E. J.; Lee, H.; Sessler, J. L.; Hong, K. S.; Kim, J. S. Angew. Chem., Int. Ed. 2014, 53, 4469. (24) Jiang, N.; Tan, N. S.; Ho, B.; Ding, J. L. Nat. Immunol. 2007, 8, 1114. (25) Mahato, R.; Tai, W.; Cheng, K. Adv. Drug Delivery Rev. 2011, 63, 659.
from their bodies and conducted ex vivo imaging. As shown in Figure S14, the HeLa tumors in the experimental group showed stronger NIR fluorescence than those in control group, which was consist with the fluorescence imaging of tumors in Figure 4. Interestingly, fluorescence signals from stomachs and colons of both of these two groups were higher than those of other organs except tumors (Figure S14). This was probably due to the sodium-dependent multivitamin transporter (SMVT) on the intestinal walls that transfers vitamins including biotin to the gastrointestinal organs.25 In conclusion, we rationally designed a NIR probe 1 to facilely prepare the self-quenched 1-NPs for tumor-targeted imaging in vitro and in vivo. Reduction-controlled self-assembly and protease-controlled disassembly of 1-NPs, along with fluorescence “Off” and “On”, were validated by in vitro characterizations. This type of biotinylated nanoparticle could be actively uptaken by biotin receptor-overexpressing tumor cells via receptor-mediated endocytosis. Upon the enzymatic cleavage by intracellular proteases in the lysosomal compartment, 1-NPs were disassembled to yield enzymatic product 1D-cleaved, accompanied by fluorescence “Turn-On”. With the fluorescence “Turn-On” property, 1-NPs were successfully applied for tumor-targeted imaging. We envision that our nanoparticles could be applied for fluorescence-guided tumor surgery in the near future.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.7b02971. General methods; syntheses and characterizations of 1; HPLC conditions; HR-MALDI-TOF/MS and 1H NMR spectra; histogram; cell viability over time; time course fluorescence microscopic images of HeLa cells; HPLC traces; quantification of the mean flux of cells; quantified total photon output; fluorescence imaging of different organs in mice (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Gaolin Liang: 0000-0002-6159-9999 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by Ministry of Science and Technology of China (2016YFA0400904), the National Natural Science Foundation of China (Grants 21675145 and U1532144), the Major program of Development Foundation of Hefei Center for Physical Science and Technology (2016FXZY006), and the Scientific Research Foundation of Jiangsu Provincial Commission of Health and Family Planning of China (H201530).
■
REFERENCES
(1) Kobayashi, H.; Ogawa, M.; Alford, R.; Choyke, P. L.; Urano, Y. Chem. Rev. 2010, 110, 2620. 9628
DOI: 10.1021/acs.analchem.7b02971 Anal. Chem. 2017, 89, 9625−9628
