Xuan
Wu‡
ac,
Ming
Liu‡
a,
Jie
Niu‡
b,
Qian
Liu
c,
Xin
Jiang
d,
Yujing
Zheng
a,
Yuna
Qian
ac,
Ying-Ming
Zhang
b,
Jianliang
Shen
*ac and
Yu
Liu
*b
aSchool of Ophthalmology & Optometry, School of Biomedical Engineering, Wenzhou Medical University, Wenzhou, Zhejiang 325035, China. E-mail: shenjl@wiucas.ac.cn
bDepartment of Chemistry, State Key Laboratory of Elemento Organic Chemistry, Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Nankai University, Tianjin 300071, China. E-mail: yuliu@nankai.edu.cn
cUniversity of the Chinese Academy of Sciences Wenzhou Institute, Wenzhou, Zhejiang 325035, China
dDepartment of Urology, Xiangya Hospital, Central South University, Changsha 410008, China
First published on 17th January 2023
An in situ supramolecular self-assembly in the subcellular organelles could provide a new strategy to treat diseases. Herein, we report a protonation-activated in situ supramolecular self-assembly system in the lysosomes, which could destabilize the lysosome membrane, resulting in the selective suppression of cancer cells. In this system, pyridyl-functionalized tetraphenylethylene (TPE-Py) was protonated in the lysosomes of A549 lung cancer cells to form octahedron-like structures with cucurbit[8]uril (CB[8]), which impaired the integrity of the lysosome membrane, resulting in selective suppression of cancer cells. Moreover, its anticancer efficiency was also systematically evaluated in vivo, triggering the apoptosis of tumor tissues with ignorable effects on normal organs. Overall, the protonation-activated self-assembly in the lysosomes based on the host–guest complexation would provide a method for novel anti-cancer systems.
Supramolecular self-assembly,17–24 as a bottom-up method to construct functional materials, has been extensively investigated in stimuli-responsive materials due to its dynamic nature, especially the pH-responsive self-assembly systems, which were widely applied in therapeutic systems.25–28 As is well known, the lysosomes of cancer cells comprise a dynamic system of acidic vesicular compartments (pH 3.8 to 4.7), which is beneficial for pH-mediated therapeutic systems.29 Compared with the traditional pH-responsive therapeutic systems, the pH-mediated self-assembly systems draw our attention. Cucurbit[8]uril (CB[8]),30,31 as a kind of macrocyclic molecule, has attracted tremendous attention due to its ability to encapsulate two guest molecules to form an ordered assembly.32–34 Besides the cationic guest molecules, guest molecules with motifs binding the H+ also could be encapsulated into the cavity of CB[8], such as the amido, which has also been explored to apply in the construction of stimuli-responsive systems.35–37 These phenomena provided a possible way to construct protonation-activated assembly systems in the lysosomes and explore the physiological effect on the lysosomes. To the best of our knowledge, no study has been conducted on the protonation-activated in vivo self-assembly systems based on host–guest complexation, and the exploration of their physiological function in cancer cells.
Herein, pyridyl-functionalized tetraphenylethylene (TPE-Py) was successfully synthesized to form the host–guest complex with CB[8] in the acidic environment, resulting in the formation of octahedron-like structures in the micrometer range. The in vitro experiments confirmed that this protonation-activated self-assembly progress could also be realized in the lysosomes of cancer cells (human lung cell line A549), resulting in the destabilization of the lysosome membrane and suppression of cancer cell proliferation. Moreover, in vivo experiments were also performed, and the results indicated that this self-assembly process could also be realized, resulting in the apoptosis of tumor cells, while having an ignorable effect on the normal tissues (Scheme 1).
Scheme 1 Schematic illustration of the in situ protonation-activated self-assembly between TPE-Py and CB[8]. |
Moreover, from the titration spectra, the obvious decrease in the absorption intensity at 348 nm was detected upon the gradual addition of CB[8] (Fig. S6a, ESI†). Meanwhile, a new absorption peak appeared at 450 nm with an isosbestic point at 424 nm. This phenomenon confirmed the successful construction of the host–guest complex. Then the fluorescence titration was further carried out. As shown in Fig. S8 in the ESI,† the maximum emission peak exhibited a red-shift upon the gradual addition of CB[8], which shifted from 528 nm to 580 nm, at a molar ratio of 1:2 ([TPE-Py·HCl]/[CB[8]]). And at the molar ratio of 1:4, the emission peak could further shift to 605 nm, and this phenomenon was ascribed to the intermolecular charge transfer after the formation of the host–guest complex with CB[8]. An interesting phenomenon could also be observed in this system, the titration equilibrium could not be reached at the molar ratio of binding stoichiometry, which might be ascribed to the relatively weak host–guest interaction between TPE-Py·HCl and CB[8], resulting in a more encapsulation tendency at a molar ratio larger than the binding stoichiometry. Moreover, the fluorescence images also confirmed the above results, due to which the yellow fluorescence could be observed in the TPE-Py·HCl solution. After the addition of CB[8] solution, the fluorescence color turned red at the molar ratio of 1:4 ([TPE-Py·HCl]/[CB[8]]). To further confirm the host–guest interaction between TPE-Py·HCl and CB[8], the UV-vis titration and fluorescence spectra (Fig. S7, ESI†) were also obtained in the neutral solution (pH = 7.4), from which both the decrease in absorption intensity and bathochromic-shift of the fluorescence peak couldn’t be observed. All the above results indicated the successful construction of the host–guest complex between TPE-Py·HCl and CB[8].
Moreover, TEM and SEM were applied to investigate self-assembly morphologies. From the SEM image (Fig. 1C) of TPE-Py·HCl, the square assembly could be observed, and the length of the side was in the range of 2 μm to 6 μm, which was also confirmed by TEM (Fig. 1D). However for the CB[8]⊃TPE-Py·HCl complexation at the molar ratio of 1:2 ([TPE-Py·HCl]/[CB[8]]), well-ordered octahedron-like structures with a diameter of about 2 μm could be observed from both SEM and TEM images (Fig. 1E and F). Even at the molar ratio of 1:4, a similar assembly could be observed (Fig. S9, ESI†). To confirm this observation, the powder XRD profiles were collected, from Fig. S10A in the ESI,† and similar diffraction peaks could be observed at different molar ratios, which could confirm the results in the TEM and SEM images. Moreover, the formed assemblies at different molar ratios were also observed by confocal laser scanning microscopy (Fig. S11, ESI†), from which the red fluorescence could also be observed, indicating that these nano-assemblies were all formed by the host–guest complex. To explore the pattern models in the assemblies, the 2D NOESY spectrum of TPE-Py·HCl was obtained (Fig. S12A, ESI†), from which the correlation peaks (Ha and Hc), (Ha and Hd), and (Hb and Hd) could be observed, which indicated that TPE-Py·HCl exhibited a J-aggregation behaviour to form square nanostructures. This phenomenon was also confirmed by the concentration-dependent 1H NMR spectra (Fig. S12B, ESI†), from which the protons shifted upfield with the concentration increase, indicating the arrangement of pyridyl motifs above the phenyl groups. The above phenomena also accounted for the weak chemical-shift change after the addition of CB[8]. Moreover, from the XRD profile (Fig. S8A, ESI†), the distance between the aromatic rings was determined to be 3.26 Å (2θ = 27.38°), which confirmed this stacking manner.
Due to the intrinsic correlation signals in TPE-Py·HCl, the 2D NOESY spectrum could not provide direct evidence for the binding model between TPE-Py·HCl and CB[8]. But from the pKa shift upon the addition of CB[8] to TPE-Py·HCl solution (Fig. S13, ESI†), we could conclude there might exist strong interaction between CB[8] and the proton on the protonated pyridyl motifs, which would decrease the deprotonation tendency of this proton.37 From the UV-vis spectra (Fig. S6A, ESI†), the addition of CB[8] could result in the hypochromic shift of the absorption peak at 348 nm, and a new charge transfer peak at 450 nm; these results also indicated the destruction of the previous J-aggregation behaviour in TPE-Py·HCl, and the generation of a new stacking manner. Therefore, we could propose a head-to-tail binding model that would be beneficial. And in this way, the remarkable bathochromic shift of the fluorescence peak could be realized.40,41 From Fig. S10A in the ESI,† the diffraction signal assigned to the stacking in TPE-Py·HCl (2θ = 27.38°) could also be observed, which also confirmed our proposal. Moreover, the host–guest complex of TPE-Py·HCl and CB[8] was further aggregated through an interlaced manner, confirmed by the diffraction peaks at 10.35° (d = 8.55 Å) and 12.09° (d = 7.30 Å).42 Therefore, we could conclude that the formed host–guest complex could further stack to form octahedron-like nanostructures. Moreover, the surface potential distribution of TPE-Py·HCl and CB[8]⊃TPE-Py·HCl was measured, which was determined to be +(19.1 ± 3.3) mV and +(16 ± 1.6) mV for [TPE-Py·HCl]/[CB[8]] = 1:2, and +(17.3 ± 3.1) mV for [TPE-Py·HCl]/[CB[8]] = 1:4 (Fig. S13, ESI†), respectively.
Moreover, the pH-mediated fluorescence emission was further investigated (Fig. S14A and B, ESI†). In the neutral solution, the maximum emission peak of CB[8] and TPE-Py was 502 nm. With the acidification of the solution, this emission peak exhibited a bathochromic shift, which moved to about 580 nm in the acidic solution (pH = 4.2). This phenomenon was consistent with the emission behavior of CB[8]⊃TPE-Py·HCl. Therefore, it could be concluded that TPE-Py could form the host–guest complex with CB[8] in the acidic solution. Moreover, this obtained system also exhibited a pH-responsive reactive oxygen species (ROS) generation ability. As shown in Fig. S17 in the ESI,† about 60% of ABDA was decomposed after being irradiated for 10 min in the acidic environment. And nearly no ROS could be generated in a neutral environment. The above results indicated that protonated TPE-Py exhibited higher ROS generation efficiency. To confirm this, the theoretical calculation (Gaussian 16, B3LYP/6-311G (d, p)) was also carried out. After being protonated, the intramolecular charge transfer was beneficial, which would lower the intersystem crossing energy, thus resulting in higher ROS generation efficiency.44 And the theoretical calculation also provided similar results: the energy gap between S1 and T1 decreased from 0.80 eV to 0.53 eV (Fig. S18B and C, ESI†). Moreover, the formation of the host–guest complex with CB[8] could not hinder its ROS generation efficiency, which was also confirmed by our previous research.45 Herein, the ROS generation capability was caused by protonated TPE-Py.
Moreover, the cell viability was improved to 55% after being incubated for 18 h, and negligible cellular toxicity could be detected after being incubated for 12 h (Fig. S20, ESI†). However, the viability of the RS1 cells was not affected even after being incubated for 24 h (Fig. 3B). The ignorable toxicity to the normal cells might be ascribed to the higher pH value of the lysosomes. The pH value was 4.5–6.0 in the normal cells, while the pH value of the lysosomes in cancer cells is 3.8–4.7.46 The confocal laser scanning images (Fig. S25, ESI†) also confirmed that fewer assemblies could be formed in RS1 cells, concluded by the weak red fluorescence that could be observed in the lysosomes. We also investigated the cellular toxicity of pre-prepared CB[8]⊃TPE-Py·HCl on A549 cells, from which negligible toxicity could be detected (Fig. S21, ESI†). The negligible toxicity of prepared nano-assemblies might result from the low uptake efficiency of CB[8]⊃TPE-Py·HCl into A549 cells due to its large size. This was also confirmed using the confocal laser scanning images, as well as flow cytometry results (Fig. S22 and S23, ESI†). From the fluorescence images, weak fluorescence could be observed, indicating that few nano-assemblies were taken in. And the flow cytometry experiment also provided the same result. Therefore, the cellular toxicity of CB[8] and TPE-Py might result from the in situ formed assembly in the cells.
Moreover, the ROS generation properties also confirmed the protonation of TPE-Py in the lysosomes of cancer cells. The cell viability also exhibited a dramatic decrease after being irradiated under white light for 10 min (Fig. S20 and S26A, ESI†). And DCFH-DA was further employed to confirm the generation of ROS in the A549 cancer cells, which presented brighter green fluorescence compared with the control group, indicating the generation of ROS upon white light irradiation (Fig. S26C, ESI†). These results also provided solid evidence for the formation of the host–guest complex between protonated TPE-Py and CB[8] in the lysosomes based on the previous pH-responsive ROS generation experiment. To avoid the influence of concentration, we also investigated the self-assembly behavior of TPE-Py and of the mixture of TPE-Py and CB[8] at a high concentration. From the SEM images, no regular morphologies could be observed (Fig. S27, ESI†), indicating that the formed assemblies should be resulting from protonated TPE-Py and CB[8].
Then the time-dependent confocal laser scanning microscopy was carried out to validate our hypothesis. From the previous fluorescence spectra, it was concluded that the addition of CB[8] could result in the red-shift of the TPE-Py·HCl fluorescence emission peak. As shown in Fig. 4, with the prolongation of incubation time, the red fluorescence intensity (from 551 nm to 650 nm) gradually increased, indicating the tendency of formation of the host–guest complex between CB[8] and TPE-Py. Besides, the cells incubated with TPE-Py were also observed from the red channel (from 551 nm to 650 nm), after being incubated for 24 h, and negligible red fluorescence could be observed under the same conditions (Fig. S28, ESI†); this result also confirmed that the red fluorescence was from the complexation between TPE-Py and CB[8] under acidic conditions, indicating the successful construction of the pH-mediated in vivo self-assembly between TPE-Py and CB[8].
Finally, the protonation-activated self-assembly of CB[8] and TPE-Py in the cells was investigated. From the bright field images of cells incubated with CB[8] and TPE-Py, the octahedron-like structures could be easily observed (Fig. 5A). From the obtained picture, the size of the octahedron-like structures was about 1 to 2 μm, which was consistent with the previous TEM images. Moreover, more and more octahedron-like structures could be observed with the prolongation of the incubation time (Fig. S29, ESI†). And bio-TEM was carried out to investigate the morphologies of the assembly in the A549 cells. As shown in Fig. 5B, a bulky assembly could be observed in the lysosomes, providing solid evidence for the protonation-activated self-assembly of CB[8] and TPE-Py in these cells.
Moreover, the time-dependent fluorescence images (Fig. 4) also confirmed the self-assembly-induced destabilization of the lysosome membrane to some extent. With the prolongation of the incubation time, the fluorescence intensity of the blue channel (from 425 nm to 525 nm) increased, which might result from the escape of TPE-Py from the lysosomes, and due to the relatively high pH value in the cytoplasm, the protonated TPE-Py was neutralized, causing the enhancement in the blue fluorescence. And the time-dependent photodynamic therapy efficiency could also confirm this phenomenon. From Fig. S20 and S26A in the ESI,† it was observed that when the incubation time increased, the photo-induced cell death decreased, indicating the escape of TPE-Py from the lysosomes. All these results provided solid evidence of the damage to the lysosome membranes in the presence of both TPE-Py and CB[8], which impaired the lysosome membranes and triggered cell death.
To further investigate the suppression of the proliferation of tumors caused by TPE-Py and CB[8], we also conducted immunohistochemical staining. First, from the H&E staining of tumor slices (Fig. 7A), an obvious decrease in the number of nuclei could be observed after treatment with the mixture of TPE-Py and CB[8], compared with the compact nuclei in the groups treated with PBS, TPE-Py and CB[8]. This phenomenon indicated that the tumor tissues suffered destruction after the treatment. As is well known, the permeabilization of the lysosomal membrane could release the hydrolytic enzymes to the cytosol, resulting in apoptosis signals in several systems.47,48 Therefore, the Tunel staining was further performed (Fig. 7A). From the obtained pictures, more Tunel+ cells (Fig. 7B) could be found after the treatment of TPE-Py and CB[8], indicating that tumor cells suffered apoptosis. Moreover, the Ki67 staining was also performed to investigate the proliferation ability of tumor cells. From Fig. 7A and C, fewer Ki67+ cells could be observed after the treatment of TPE-Py and CB[8]. Therefore, it could be concluded that TPE-Py and CB[8] could induce the apoptosis of tumor cells, resulting in the suppression of tumor growth.
Fig. 7 (A) H&E, Tunel, and Ki67 staining of tumors after the treatment for 14 days, (B) quantification of Tunel+ cells (n = 3), and (C) quantification of Ki67+ cells (n = 3). |
The safety of TPE-Py and CB[8] was further investigated in Balb/C mice. After injecting the solution containing TPE-Py and CB[8] through the tail vein, aminotransferase (GOT), alanine aminotransferase (GPT), blood urea nitrogen (BUN), and creatinine (CR) were detected using commercially available kits. From Fig. S33A and B in the ESI,† it could be concluded that renal and liver functions were not affected after the injection of TPE-Py and CB[8]. Moreover, the main organs (heart, liver, spleen, liver, and kidneys) were also collected after 14 days, and the H&E staining was performed (Fig. S33C, ESI†). Compared with PBS-treated mice, H&E staining also indicated that TPE-Py and CB[8] possessed ignorable long-term toxicity to the main organs, and these results indicated the excellent biocompatibility of these in situ self-assembly systems.
Footnotes |
† Electronic supplementary information (ESI) available: Details of the characterization of the target compound, UV-vis and fluorescence spectra, zeta potential results, TEM and SEM images, fluorescence images, MTT assays, and in vivo safety evaluation. See DOI: https://doi.org/10.1039/d2sc05652f |
‡ X. Wu, M. Liu, and J. Niu contributed equally to this work. |
This journal is © The Royal Society of Chemistry 2023 |