The synthetic approach of the fluorescence “turn-on” nano-aptamer sensor is summarized in Scheme 1. First, AuNPs with a size of 25 nm were synthesized according to the method reported by Wang et al. (2019). The dynamic light scattering (DLS) results showed that AuNPs were well-dispersed with an average hydrodynamic size of 30 nm (Figure 1A). At the same time, the TEM image of AuNPs is exhibited in Figure 1B, where AuNPs are of a round shape and well-dispersed. Subsequently, the ssRNA with the endpoint modified sulfhydryl group was directly bound to the surface of AuNPs through coordination bonds. After the modification of ssRNA, the zeta potential of AuNPs dramatically increased from −58 mV to −36 mV (Figure 1D), indicating the successful modification of ssRNA on AuNPs. In Figure 1D, the UV–Vis spectra of AuNPs and AuNPs-ssRNA showed that the adsorption peak of AuNPs was at 523 nm while that of AuNPs-ssRNA was at 525 nm. The red-shift in the adsorption spectra further demonstrated that ssRNA was successfully conjugated onto the AuNPs’ surface. Since RNA hybridization required annealing, agglomeration would occur in subsequent experiments if the concentration ratio of AuNPs to ssRNA was greater than or less than 1:200 (molar concentration ratio). Thus, the optimum concentration ratio of AuNPs to ssRNA was chosen to be 1:200.

The fluorescence spectra in Figure 2 showed that the emission peak was 525 nm for CdSe/ZnS QDs; there was an overlap between the emission spectrum of QDs and the absorption spectrum of AuNPs around 525 nm, indicating that the QDs and AuNPs could work as a FRET donor–acceptor pair. Then, the CD133-targeted aptamer was conjugated onto CdSe/ZnS QDs by an ammonia carboxylation reaction, and the ratio between aptamer and QDs was optimized. Due to the large surface area of QDs, there were multiple binding sites on the surface of QDs to couple with aptamers. After the overdose aptamer reacted with QDs, the unbound aptamer was removed using an ultrafiltration tube. For proving the aptamer conjugation, QDs and QD-aptamers were monitored by agarose gel electrophoresis (Figure 3), which implied that the conjugation efficiency reached a plateau when the aptamer concentration exceeded 7 times the molar concentration of QDs. Herein, 7:1 was chosen as the optimal concentration ratio of aptamer-bound QDs.

To evaluate whether FRET occurred in our design, these fluorescence spectra were measured, including QD-aptamers, the mixture of QD-aptamer and AuNPs, and the hybridization of QD-aptamers and AuNPs-ssRNA (aptamer sensor) (Figure 4). The fluorescence spectrum for free QDs showed a maximum peak at 525 nm, which was close to the maximum of absorption measured for AuNPs. It suggested that CdSe/ZnS QDs and AuNPs could serve as a suitable donor–acceptor pair. It could also be seen that the fluorescence band (the full width at half-maximum fluorescence intensity, FWHM) was relative narrow and symmetric, which suggested that both QDs and the aptamer sensor were homogenous and monodisperse. According to the past studies, FRET could operate when an energy donor to an acceptor was in close proximity (≤10 nm). The chain length of the CD133-targeted aptamer indicated that the distance between the donor and the acceptor was definitely achieved in the studied system. As could be seen in Figure 4C, when the hybridization between QD-aptamers and AuNPs-ssRNA is driven by the specific pairing of aptamers and partial complementary ssRNA, the fluorescence signal of QDs was quenched dramatically by AuNPs due to the occurrence of FRET. However, when there was no ssRNA conjugated to AuNPs (Figure 4B), the hybridization was not in operation, and the fluorescence intensity of QDs was only partially reduced. The spectra results indicated that the efficiency of FRET was significantly improved by linking QDs and AuNPs through aptamer–ssRNA hybridization. To ensure adequate hybridization between the aptamer and ssRNA, the hybridization time was fixed at 90 min (annealing time). As shown in Figure 5A, the fluorescence signal intensity of QDs decreased with the increase in AuNPs and reached quench equilibrium when the molar concentration ratio of QDs to AuNPs was 10:1 (Figure 5B). Finally, the synthesis conditions of the aptamer sensor were determined where the hybridization reaction time was 90 min and the concentration ratio was 10:1.

To explore the fluorescence “turn-on” efficiency of this FRET aptamer sensor, a series of concentrations of CD133 were tested from 0 to 1.539 μM. Herein, a fixed number of QD-aptamers (5 nM) was used. As a result of competitive replacement of ssRNA by CD133, the distance between QDs and AuNPs increased significantly, and fluorescence recovery was realized. The recovery efficiency was expressed by ΔF = F_C − F₀, where Fc is the fluorescence intensity of QDs after recovery and F₀ is the original fluorescence intensity of the aptamer sensor. As shown in Figure 6A, the recovery efficiency gradually increased with the increase in CD133 and reached a maximum of 68% with 1.539 μM CD133, and then the recovery efficiency was almost unchanged when the concentration of CD133 further increased. Figure 6B indicated the fluorescence signal recovery (F_C − F₀) versus a series of CD133 concentrations. Then the detection limit (LOD) was determined by the standard curve, which was around 6.99 nM.

To evaluate the specificity, the aptamer sensor was treated with multiple common cancer-associated proteins and compared with CD133. All the experiments used the same concentration (2 μM) of proteins for comparison. As shown in Figure 7, these interfering proteins hardly recovered the fluorescence of QDs. For example, the recovery efficiency of CD133 was around 68%, while that of SCC was merely around 5%. The results indicated that this FRET “turn-on” nano-aptamer sensor exhibited high selectivity.

In order to study the biocompatibility of the nano-aptamer sensor, CCK-8 assays were used to evaluate the cytotoxicity of the aptamer sensor. The results indicated that A549 cells still maintained high viability after incubating for 24 h with the aptamer sensor up to 20 nM (Figure 8A). Even if the incubation time was prolonged to 48 h, it would not affect the cell viability (Figure 8B). In other words, if the concentration of the aptamer sensor was not more than 20 nM, then the aptamer sensor had low toxicity to cells and could be used for cellular imaging. Furthermore, to demonstrate the capacity of the fluorescence “turn-on” FRET aptamer sensor in selectively detecting CD133^pos cells, the lung cancer cell A549 (CD133^pos cell) was treated with the aptamer sensor. After incubation with QD-aptamers or the aptamer sensor for 8 h, strong fluorescence signals were detected from A549 cells by confocal microscopy imaging (Figures 9B,C), while the control experiment using QDs without the aptamer at the same concentration showed weak fluorescence (Figure 9A). For verifying that this “turn-on” phenomenon was ascribed to the competitive release of AuNPs-ssRNA, the binding site of A549 cells was saturated with the CD133-targeted aptamer for 24 h before incubation with the aptamer sensor. It turned out that the fluorescence signal from A549 cells was negligible, as shown in Figure 9D. A negative control experiment performed with CD133-negative HuH-7 human hepatoma cells showed that the fluorescence recovery of the aptamer sensor was not observed (Figure 9E), further indicating that the release of AuNPs-ssRNA was from the specific binding between the aptamer and CD133.

Chloroauric acid (HAuCl₄) and sodium citrate were purchased from Shanghai Aladdin Biochemical Technology Co. Ltd. Carboxyl-modified quantum dots (-COOH QDs) were obtained from Suzhou Xingshuo Nanotechnology Co. Ltd. 1-ethyl-(3-dimethylaminopropyl) carboimide (EDC) and N-hydroxysuccinimide (NHS) were purchased from Shanghai Macklin Biochemical Co. Ltd. The amino-modified CD133 aptamer (5′-NH₂-CCCUCCUACAUAGGG-3′) and CD133 aptamer partial complementary pairing part (ssRNA, 5′-SH-CCCUAUG-3′) were synthesized by Shanghai Generay Biotech Co. Ltd. All buffers were prepared with ultrapure water, which was purified to a resistivity of 18.2 MΩ*cm by a milli-Q system (Merck Millipore, United States). A dynamic light scattering particle size analyzer (SZ-100, Horiba, Japan) was used for measuring the zeta potential and particle size. The morphology of nanoparticles was characterized by transmission electron microscopy (JEOL JEM-2100 F, JEOL Ltd., Japan). The UV–Vis absorption spectra were recorded on a microspectrophotometer (NanoDrop ND-1000, Thermo Scientific, United States). Fluorescence spectra were obtained using a fluorescence spectrophotometer (F-7000, Hitachi, Japan). Fluorescence images were recorded using a confocal laser scanning microscope (TCS SP8 Leica, United States).

All reaction instruments needed to be soaked in aqua regia or concentrated alkali and washed with ultrapure water 3 to 4 times. Then 250 μl of 0.4 M chloroauric acid (HAuCl₄) and 50 ml of ultrapure water were added to a round bottom flask, continuously stirred, and heated to boiling. After boiling, 10 ml of 38 mM sodium citrate solution was quickly injected, and then the solution was heated to boiling again. The color of the solution gradually changed from light yellow to wine red; the solution was boiled for another 15 min until the reaction was complete. After cooling, it was filtered by an ultrafiltration membrane and stored at 4°C.

The prepared AuNPs were incubated with a sulfhydryl nucleotide single chain (ssRNA) in a ratio of 1:200 for 24 h, and then 1 M sodium phosphate buffer (PBS, 1 M NaCl, 100 mM Na₂HPO₄ and NaH₂PO₄, pH = 7.4) was added step by step. At least for 30 min between each step, the final concentration of PBS reached 0.1 M. The solution was incubated at room temperature for 40 h. Finally, the excess ssRNA was removed by centrifugation (13,800 rpm, 20 min) and redispersed in 0.1 M PBS.

Since coupling nucleotides with AuNPs would reduce the anions on their surface, the salt solution was added to resist the cations on the surface of AuNPs in order to prevent agglomeration and maintain stability, while uncoupled or incompletely coupled AuNPs would agglomerate and change color. Therefore, salt solution of the same concentration and volume could be added to the reaction mixture of AuNPs and ssRNA to observe whether it changes color and agglomerates, and the reaction degree between AuNPs and ssRNA is judged.

Carboxyl CdSe/ZnS QDs (40 μl, 0.6 nM) were activated with 60 μl (50 mM) EDC and 30 μl (25 mM) NHS under mild stirring for 15 min. The activated QDs were incubated with an amino aptamer at different mass ratios for 24 h, and then the unreacted aptamer was separated using an ultrafiltration centrifuge tube. The obtained QD-aptamers were redispersed in PBS.

The obtained QD-aptamers were divided into two groups. One group was added with the stock solution of AuNPs, and the other was added with AuNPs-ssRNA. The nucleotide chains were complementary paired by annealing (adding the annealing buffer at 95°C for 5 min and gradually reducing to room temperature), and then the fluorescence value of the mixture was measured at the same time to verify that the strong fluorescence quenching effect was due to the complementary pairing of the nucleotide chains, shortening the distance between the energy donor and the acceptor. Based on the fluorescence change in QDs, the reaction time and the molar ratio of AuNPs to QDs were determined. The final fluorescence “turn-on” nano-aptamer sensor was obtained.

After the aptamer sensor reacted with CD133 and other interfering proteins (actin, EGFR, CD44, TGF-β1, PIN 5, SCC, and IL-1α), the fluorescence value of the reaction solution was measured to verify that CD133 could specifically bind to the CD133-targeted aptamer on the surface of QDs by replacing AuNPs-ssRNA. The aptamer did not respond to other interfering proteins, which proved its specificity.

After passage, lung cancer cells A549 were inoculated into confocal dishes and divided into four groups for different experimental operations. The first group was incubated with single QDs for 8 h. The second group was incubated with QD-aptamers for 8 h. The third group was incubated with aptamer sensors for 8 h. The last group was incubated with a single aptamer for 24 h and then with the aptamer sensor for 8 h. The concentration of QDs in the four groups was the same. After incubation, the solution in the dishes was washed with PBS twice, and the fluorescence imaging of the cells was observed under a confocal laser scanning microscope. As control, HuH-7 cells were incubated with an aptamer sensor for 8 h to detect the fluorescence of QDs.