Core-shell self-assembly triggered via a thiol-disulfide exchange reaction for reduced glutathione detection and single cells monitoring

A novel core-shell DNA self-assembly catalyzed by thiol-disulfide exchange reactions was proposed, which could realize GSH-initiated hybridization chain reaction (HCR) for signal amplification and molecules gathering. Significantly, these self-assembled products via electrostatic interaction could accumulate into prominent and clustered fluorescence-bright spots in single cancer cells for reduced glutathione monitoring, which will effectively drive cell monitoring into a new era.

Inspired by that mentioned above, we proposed a core-shell DNA self-assembly catalyzed via thiol-disulfide exchange reactions that could realize target-initiated hybridization chain reaction (HCR) for reduced glutathione detection. Importantly, a non-destructive signal amplification tactic for GSH monitoring in living cells was developed. This core-shell self-assembly had a nuclear nanosphere and an outer layer of fluorogen-labeled nucleic acid congeries, attributed to the HCR products triggered via thiol-disulfide exchange reactions. The intracellular core-congeries supplied efficient signal enlargement, fluorescence gathering and enabled super-sensitive monitoring of GSH levels. Moreover, in situ imaging of GSH in living cells could boost the monitoring of the distribution and dynamical expression of GSH, and research on GSH-related cellular processes and diseases.
In Fig. 1, a novel core-shell self-assembly method catalyzed via thiol-disulfide exchange reactions was presented, which was employed to target GSH and trigger the HCR reaction for intracellular GSH monitoring. A key design was the core-shell DNA self-assembly that could realize GSH-initiated core-HCR for signal amplification and molecules gathering. First, three hairpin-structured DNA probes H 1 , H 2 and H 3 were designed, H 1 including a disulfide bond (-S-S-), H 3 labeled with a fluorophore/quencher pair. The H 1 , H 2 and H 3 probes could be flexible enough to synergistically interact with the positively charged surface, enabling H 1 , H 2 and H 3 probes to be steadily assembled on the aminopropyl-modified cores, as core-H 1 H 2 H 3 probes. Second, after the core-H 1 H 2 H 3 probes were incubated into living cells, a target GSH was able to fracture the disulfide bond of H 1 probe via a thiol-disulfide exchange reaction. The fractured H 1 probe initialized the hybridization with H 2 and generated a single-stranded tail in H 2 , which might dissociate or enhance the mobility of H 2 on the surface of cores, benefiting its hybridization with H 3 and renewing a single strand tail in H 3 . Third, by this means, a chain reaction was catalyzed for alternating hybridization between H 2 and H 3 , creating a chain-like assembly of H 2 and H 3 [37][38][39] . These HCR products had a regular conformation, self-assembled on the surface of cores via electrostatic interaction, as core-congeries. It could produce a GSH-activity-related fluorescence signal via the fluorescence resonance energy transfer, due to the fluorophore (FAM) relatively far away from the quencher (BHQ). Because this unique DNA self-assembly resulted in core-shell nanostructure with a positively charged interlayer and nucleic acid congeries, therefore it could offer high efficiency for cellular delivery 35 and fluorescence activation.

Results and Discussion
Characterization of the Materials. The employed materials were displayed by transmission electron microscopy (TEM) as shown in Fig. 2A, the average size of them was about 100 nm. PCG-MSNs (preparation in the ESI) and MSNs exhibited similar type of N 2 adsorption-desorption isotherms. However, it should be pointed out that the amount of the N 2 adsorbed on the PCG-MSNs was much lower than that adsorbed on the MSNs. And correspondingly, PCG-MSNs possessed the Burnauer-Emmett-Teller (BET) surface area of 262.4 m 2 g −1 , which was much smaller than that of the MSNs. From the BJH pore size distributions in Fig. 2, we could find that the pores for the MSNs centered at 4 nm disappeared completely while the pores around 20 nm decreased obviously. These data showed that the massive positively charged aminopropyl groups have been modified into the pores of MSNs.

Preparation and Cytotoxicitiy Tests of of PCG-MSNs Assembled H1, H2 and H3 Probes.
PCGs-MSNs were dispersed in 2 mL hybridization buffer (pH 7.4), and stirred continuously at 37 °C for 90 min to obtain the PCG-MSNs solution (  37 °C for 180 min, the excess reagents were moved away by centrifuging at 10000 rpm for 10 min. The sediment was washed and centrifuged repeatedly for two times to obtain the PCG-MSNs-H 1 H 2 H 3 probes. The cytotoxicity tests of PCG-MSNs assembled H 1 , H 2 and H 3 probes (PCG-MSNs-H 1 H 2 H 3 ) were studied with Hela cells by MTT experiment 40,41 , as shown in Fig. 3. Briefly, the cells were incubated with 200 μ L culture medium containing 15 μ L PCG-MSNs-H 1 H 2 H 3 probes for different times, then were washed with 200 μ L hybridization buffer (pH 7.4) once, MTT (0.5 mg mL −1 , 100 μ L) were seeded in the wells and incubated at 37 °C for 4 h. Afterwards, 150 μ L DMSO was added to each well to dissolve the crystals constituted by the living cells, and the absorbance at 490 nm was tested to get the relative cell viability. The data showed that Hela cells maintained about 90.3% of the cell viability by (A test /A control ) × 100% after incubation with 30 μ L probe for 6 h, revealing that the good biocompatibility.
Analysis of the HCR Reaction, and Fluorescence Responses of Core-Congeries Dispersion and Liquid Supernatants. The HCR reaction activated via GSH was analyzed via polyacrylamide gel   Fig. 5B. The regression equation was depicted as: F − F 0 = 238.6 l gC + 2440.1, the corresponding correlation coefficient (R 2 ) of calibration curve was 0.997, and the detection limit of GSH detection was calculated to be 2.0 × 10 −11 M (3σ). The reproducibility of the core-congeries system was investigated by 11 repetitive measurements of 5.0 × 10 −8 M GSH under the optimal conditions. The relative standard deviation (RSD) was 11.2%, indicating a fine reproducibility of the method.
Cell Imaging for Monitoring GSH. The response and efficient delivery of the core-congeries nanoassembly in vitro offered the possibility for sensitively monitoring of GSH expression in single cells. For the imaging experiments, four types of cells (K562, HepG2, Hela, L-02) were cultured in 6-well slides. In short, the cells were incubated with 700 μ L culture medium including the core-H 1 H 2 H 3 probes (50 mg/ml) for 4 h, then were washed three times with hybridization buffer (pH 7.4). The core-H 1 H 2 H 3 probes could be rapidly adopted by the cells due to the targeted adsorption and endocytosis, which chiefly led to the higher level of nanospheres internalization into cells within 2 h 42,43 . In Fig. 6, upon addition of the core-H 1 H 2 H 3 probes, the cells did not show observable fluorescence within initial 10 min. After 30 min, fluorescence emerged into the cytoplasm and its intensity increasingly heightened ascribed to the GSH-triggered HCR in living cells. The stronger fluorescence-bright spots revealed that the core-H 1 H 2 H 3 probes were successfully transported into the living cells, and the core-HCR nano-assembling was achieved in the single cells. The fluorescence intensity was attained the maximum at 240 min ( Figure S3). The  fluorescence-bright spots in single cells were observed by employing confocal microscopy. The processes of monitoring intracellular GSH were carried out at 37 °C within 4 h, as shown in Fig. 7. The clustered and prominent fluorescence-bright spots could be visualized vividly in living K562, Hela and HepG2 cells, which displayed that low-abundance GSH monitoring in single tumor cells was realizable by this core-HCR method catalyzed via a  thiol-disulfide exchange reaction. Nonetheless, the big fluorescence-bright spots emerged hardly in human normal hepatocytes (L-02). These data implied that the relative expression levels of GSH in living HepG2, K562 and Hela cells were higher than in L-02 cells. So this novel core-HCR method triggered via a thiol-disulfide exchange reaction was practicable to detect changes in GSH expression levels in single tumor cells.

Conclusion
This work reported a novel core-shell DNA self-assembly catalyzed by thiol-disulfide exchange reactions, which could realize target-initiated hybridization chain reaction for signal amplification and molecules gathering. Significantly, the self-assembled core-congeries could accumulate into prominent and clustered fluorescence-bright spots in single cancer cells for GSH monitoring. This finding has exciting potential in the study of pathological variations or biological processes in live cells in virtue of its good selectivity and excellent sensitivity.

Materials and Methods
Chemicals and Reagents. K562 and Hela cells were purchased from KeyGEN biotechnology Company (Nanjing, China). Human hepatocellular liver carcinoma cell line HepG2 was bought from Shanghai Bioleaf Biotechnology Company (Shanghai, China), and human normal hepatocytes L-02 was from Silver Amethyst Biotech. Co. Ltd. (Beijing, China). MirVana Glutathione isolation kit and Fetal bovine serum were purchased from Life Technologies (Carlsbad, California). Tris-HCl, NaCl, MgCl 2 , EDTA, 3-aminopropyltriethoxysilane (APTES), triethanolamine (TEAH 3 ), and tetraethyl-orthosilicate (TEOS) were purchased by Aladdin. All the water used in the work was RNase-free. Hybridization buffer (pH 7.4) contained 10 mM Tris-HCl, 50 mM NaCl, 1 mM EDTA, and 10 mM MgCl 2 . 3-(4,5-dimethyl-thiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and dimethyl sulphoxide (DMSO) were bought from Sigma Chemical Company. Cetyl-trimethylammonium tosylate (CTATos) was purchased from Merck. Unless otherwise mentioned, ultrapure water was used throughout the experiments. All other reagents employed in this work were analytical grade and were used without further purification. All oligonucletides used in the present study were purchased by Sangon Biotech Co., Ltd. Apparatus and Characterization. Fluorescence imaging was performed by a Leica TCS SP8 inverted confocal microscope (Leica, Germany). The cellular images were acquired using a 100× objective. Solid laser (488 nm) was used as excitation source for FAM-labeled probe, and a 495-545 nm bandpass filter was used for fluorescence detection. Transmission electron microscopy (TEM) was measured on a JEOL JEM-2100 instrument. All fluorescence measurements were carried out on a F4600 fluorometer (Hitachi, Japan).
Cell Culture and GSH Preparation. K562, HepG2, Hela and L-02 cells were respectively cultured in RPMI