Increasing the accumulation of aptamer AS1411 and verapamil conjugated silver nanoparticles in tumor cells to enhance the radiosensitivity of glioma

Jing Zhao; Dongdong Li; Jun Ma; Huiquan Yang; Wenbin Chen; Yuyu Cao; Peidang Liu

doi:10.1088/1361-6528/abd20a

1. Introduction

Malignant glioma is a common primary brain tumor with a rapidly fatal clinical course [1, 2]. Radiotherapy, chemotherapy and surgical resection are considered the standard options for the treatment of malignant glioma. Among them, radiotherapy, a common adjuvant treatment approach, performs an important role in the treatment of glioma [3, 4]. However, a dilemma in the application of radiotherapy is the resistance of tumor cells to x-ray irradiation. Radioresistance significantly decreases the efficacy of radiotherapy, which can ultimately lead to tumor recurrence and metastasis [5]. Thus, it is essential to reduce radioresistance and enhance the efficacy of radiotherapy for glioma.

With the rapid development of nanotechnology, nanoscale metal-based materials have been proposed to enhance the radiosensitivity of tumor cells [6, 7]. Among these materials, silver nanoparticles (AgNPs) are particularly attractive owing to their excellent radiosensitizing properties [8]. Xu and colleagues [9] first reported the radiosensitization effect of AgNPs on glioma cells. They synthesized AgNPs with diameters of 20, 50, and 100 nm and compared their radiosensitizing effects. It was found that AgNPs with a smaller particle size had a stronger radiosensitizing effect on U251 glioma cell line. This conclusion was also confirmed in C6 and SHG-44 glioma cell lines. Subsequently, Swanner [10] and Habiba [11] found that AgNPs also exhibited radiosensitizing properties in other tumors, such as breast cancer, colorectal cancer and liver cancer. Moreover, at the same molar and mass concentrations, AgNPs exhibit more profound radiosensitizing ability than gold nanoparticles (AuNPs), which have been widely studied to enhance cancer radiation therapy [12]. However, their ability to efficiently enter and accumulate in tumor cells needs to be improved [13, 14]. Functionally modified AgNPs may be able to meet this demand, thereby enhancing their radiosensitizing effect and effectively reducing their dosage. Bovine serum albumin (BSA) is receiving more and more attention in the field of nanotechnology due to its excellent biocompatibility and easy functionalization [15, 16]. Aptamer AS1411 can specifically bind to nucleolin, which is frequently overexpressed on the plasma membrane of cancer cells, resulting in AS1411-modified nanoparticles being effectively internalized into tumor cells via receptor-mediated endocytosis [17, 18].

P-glycoprotein (P-gp), a glycosylated 170 kDa transmembrane protein encoded by the ABCB1 gene, can transport therapeutic agents out of tumor cells during cancer treatment [19, 20]. Overexpression of P-gp causes tumor cells to become resistant to many promising anti-cancer agents [21, 22]. Verapamil (VRP), the calcium channel blocker, is used in laboratory and in clinical trials that attempt to increase intracellular concentrations of therapeutic substances by inhibiting the activity of P-gp [23]. Since effectively increasing the accumulation of radiosensitizers in tumor cells is essential for successful radiosensitization therapy [24], the P-gp inhibitor VRP may cause an increase in the amount of radiosensitizers in tumor cells, thereby improving the efficacy of radiotherapy.

The aim of the current study was to construct AS1411 and VRP conjugated BSA coated AgNPs (AgNPs@BSA-AS-VRP), and then to evaluate their enrichment in glioma cells and radiosensitization effect (scheme 1). The related radiosensitization mechanisms were also investigated. Our results provide a novel approach for designing highly effective nano-radiosensitizers to enhance the radiosensitivity of tumor cells.

**Scheme 1.** Schematic illustrating the preparation of AgNPs@BSA-AS-VRP and their application in glioma radiotherapy. (a) Schematic illustrating the preparation of AgNPs@BSA, AgNPs@BSA-AS and AgNPs@BSA-AS-VRP. (b) Schematic illustrating the enrichment of AgNPs@BSA-AS-VRP in glioma cells through AS1411-mediated active targeting and inhibition of P-gp activity.
Download figure:
Standard image High-resolution image

2. Materials and methods

2.1. Materials

Polyvinylpyrrolidone (PVP, K30), BSA, VRP, dimethyl sulfoxide (DMSO), 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT), rhodamine 123 (Rho 123), Giemsa staining kit and Thioredoxin Reductase (TrxR) Assay Kit were ordered from Sigma-Aldrich (St. Louis, MO, USA). Aptamer AS1411 (sequence: 5'-GGTGGTGGTGGTTGTGGTGGTGGTGG-3') was synthesized by Sangon Biotech Co., Ltd (Shanghai, China). Fetal bovine serum (FBS) and Dulbecco's modified Eagle's medium (DMEM) were purchased from Gibco (Carlsbad, CA, USA).

2.2. Synthesis of AgNPs

In the present study, PVP-coated AgNPs were prepared using an electrochemical synthesis method [25]. In brief, two silver rods with a diameter of 2 mm were fitted on the electrolytic reactor's cover and used as electrolytic electrodes. Next, 5 mg ml⁻¹ PVP solution was pumped into the reactor at a flow rate of 60 ml h⁻¹ by a peristaltic pump. Meanwhile, a voltage of 10 V was applied to the silver electrodes, and the polarity direction of the electrodes was reversed every minute. At last, the collected reaction mixture was filtered with 0.22 μm filter membrane, and the AgNPs solution was obtained.

2.3. Preparation of AgNPs@BSA-AS-VRP

Firstly, BSA and AgNPs were mixed according to the mass proportion of 10:1 and reacted for 12 h at room temperature. Secondly, the excess BSA was removed by centrifugation (14 000 rpm, 30 min, 4 °C), and the AgNPs@BSA solution was obtained. For the AS1411 conjugation, 100 OD AS1411 was dissolved in 1 ml distilled water and then added to 10 ml of 260 μg ml⁻¹ AgNPs@BSA solution. After stirring overnight, the reaction mixture was dialyzed for 48 h in dialysis bags (molecular weight cut-off 20 000 Da) to remove the non-conjugated AS1411, and the AgNPs@BSA-AS solution was prepared. Finally, verapamil was dissolved in distilled water and reacted with the AgNPs@BSA-AS solution at a molar ratio of 1:19 for 12 h. The resulting AgNPs@BSA-AS-VRP solution was dialyzed thoroughly to remove the unbound reagents and stored at 4 °C.

2.4. Characterization

2.4.1. Transmission electron microscopy (TEM)

The particle size and morphology of the synthesized nanoparticles were observed using TEM (HT7700, Hitachi, Tokyo, Japan) operated at an accelerating voltage of 120 kV. TEM sample was prepared by placing a drop of the sample dispersion on a 200 mesh carbon coated copper grid and drying at room temperature.

2.4.2. Ultraviolet–visible (UV–vis) spectroscopy

The UV–vis absorption spectra of BSA, AS1411, VRP, AgNPs@BSA, AgNPs@BSA-AS and AgNPs@BSA-AS-VRP were recorded at wavelengths ranging from 200 to 700 nm using a UV–vis spectrophotometer (Lamda-950, Perkin Elmer, Germany).

2.4.3. Fourier-transform infrared (FTIR) spectroscopy

The FTIR transmission spectra of BSA, AS1411, VRP, AgNPs@BSA, AgNPs@BSA-AS and AgNPs@BSA-AS-VRP were recorded by an FTIR spectrometer (Nicolet 6700, Thermo Scientific, MA, USA) using KBr pellet technique in the range of 650–4000 cm⁻¹. Spectra were collected from 16 scans at a resolution of 4 cm⁻¹, and the data were analyzed using Omnic software (version 8.2).

2.4.4. Measurement of sizes and zeta potentials of the prepared nanoparticles

The hydrodynamic diameters, polydispersity indexes and zeta potentials of AgNPs@BSA, AgNPs@BSA-AS and AgNPs@BSA-AS-VRP were measured by dynamic light scattering (DLS) using a Zetasizer instrument (Zetasizer Nano ZS90, Malvern Instruments, Malvern, UK). Measurements were performed at a 90° angle after dispersing 20 μl of nanoparticles suspension in 1 ml of distilled water.

2.4.5. Measurement of the densities of AS1411 and VRP conjugated on AgNPs@BSA

The resulting AgNPs@BSA-AS or AgNPs@BSA-AS-VRP reaction mixture was centrifuged for 30 min at 14,000 rpm to separate the non-conjugated AS1411 or VRP. The concentration of non-conjugated AS1411 or non-conjugated VRP was determined by comparing its absorbance at 260 or 227.5 nm with that of the corresponding calibration curve. The calibration curve was established by a UV–vis spectrophotometer using a series of known concentrations of AS1411 or VRP. The density of AS1411 or VRP conjugated on AgNPs@BSA was calculated by the following equation: (amount of AS1411 or VRP added in AgNPs@BSA - amount of non-conjugated AS1411 or VRP) / amount of AgNPs@BSA [26].

2.5. Cell culture and in vitro cytotoxicity assay

The U251 glioma cell line was obtained from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). Cells were cultured in DMEM containing 10% FBS, penicillin (100 U ml⁻¹) and streptomycin (0.1 mg ml⁻¹) in a humidified 5% CO₂ atmosphere at 37 °C.

The cytotoxicity of AgNPs@BSA, AgNPs@BSA-AS and AgNPs@BSA-AS-VRP against U251 cells was evaluated by MTT assay. Briefly, U251 cells (1 × 10⁴ cells ml⁻¹, 100 μl well⁻¹) were seeded in 96-well plates and cultured overnight. Then the cells were incubated with various concentrations (0, 7.5, 15, 30, 60 and 120 μg ml⁻¹) of AgNPs@BSA, AgNPs@BSA-AS or AgNPs@BSA-AS-VRP for 24 h. The corresponding different concentrations of BSA, AS1411 or VRP were also used to incubate the cells under the same conditions. There were three replicates for each tested concentration. After incubation, the cells were washed three times with phosphate buffered saline (PBS) and then cultured with 100 μl fresh medium containing 20 μl MTT solution (5 mg ml⁻¹) for an additional 4 h. Next, the medium was discarded gently, and 150 μl DMSO was added to each well to fully dissolve the formed formazan crystals. Finally, the absorbance of each well was detected using a microplate reader (Thermo Scientific, MA, USA) at a wavelength of 570 nm, and the half maximal inhibition concentration (IC₅₀) values were calculated using SPSS software (version 19.0, SPSS, Chicago, IL, USA).

2.6. Semi-quantitative and quantitative analysis of intracellular nanoparticles

U251 cells were seeded in 24-well plates containing round coverslips at the density of 1 × 10⁴ cells/well. After overnight incubation, U251 cells were treated with AgNPs@BSA or AgNPs@BSA-AS and AgNPs@BSA-AS-VRP mixed in different proportions (1:0, 0:1, 9:1, 19:1 and 29:1) for 12 h. The concentration of AgNPs in all treatment groups was 20 μg ml⁻¹. Subsequently, the cells were washed three times with PBS to remove the non-internalized nanoparticles and then fixed with 4% paraformaldehyde. The internalized nanoparticles were observed under a dark-field microscope (Eclipse E600, Nikon, Tokyo, Japan). Furthermore, inductively coupled plasma mass spectrometry (ICP-MS; 7500a, Agilent, CA, USA) was applied to quantitatively analyze the amount of internalized nanoparticles. U251 cells treated with the above nanoparticles were collected and resuspended in 50 μl PBS at a concentration of 1 × 10⁵ cells ml⁻¹. Then, 6 ml 69.0% nitric acid was used to dissolve the intracellular AgNPs into Ag ions at a temperature of 125 °C. Finally, the resulting Ag ions solution was diluted to 5 ml with distilled water and then analyzed by ICP-MS.

2.7. Rho 123 accumulation assay

Rho 123 accumulation assay was conducted to determine the effects of AgNPs@BSA or AgNPs@BSA-AS and AgNPs@BSA-AS-VRP mixed in different ratios without or with x-ray irradiation on the efflux activity of P-gp in U251 glioma cells. In brief, U251 cells (1 × 10⁴ cells ml⁻¹) were seeded in 6-well plates and incubated at 37 °C. On the second day of seeding, the cells were divided into ten groups: control, AgNPs@BSA, AgNPs@BSA-AS and AgNPs@BSA-AS-VRP mixed in different ratios (1:0, 0:1 and 19:1), and the corresponding irradiation treatment groups. The concentration of AgNPs in all nanomaterial treatment groups was 20 μg ml⁻¹. After 12 h of incubation, the cells in irradiation treatment groups were exposed to 4 Gy of 6 megavolt x-rays. Subsequently, 10 μl Rho 123 (0.5 mg ml⁻¹) was added in each well and incubated for 2 h, and the cells were washed twice with ice-cold PBS before being analyzed by flow cytometry (BD Bioscience, San Jose, CA, USA).

2.8. Colony formation assay

The survival fraction of U251 cells treated with nanoparticles combined with irradiation was analyzed by colony formation assay. U251 cells (1 × 10⁴ cells ml⁻¹) were seeded in 6-well plates and cultured to 80%–90% confluence. Next, the cells were incubated with AgNPs@BSA, AgNPs@BSA-AS or 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP for 12 h. In all nanoparticles treatment groups, the concentration of AgNPs was 20 μg ml⁻¹. Subsequently, the cells were washed three times with PBS and irradiated at a dosage of 0, 2, 4, 6 or 8 Gy. After irradiation, the cells were returned to the incubator to form colonies. Seven days later, U251 cells were fixed with 4% paraformaldehyde and then stained with Giemsa staining kit. After water washing and air-drying the plates, the stained colonies (containing ≥ 50 cells) were counted. At last, the cell survival curves were plotted using the multitarget single-hit model, and the sensitization enhancement ratio (SER) values were calculated. The experiment was repeated three times.

2.9. TrxR activity assay

The TrxR activity was assayed according to the previous literature [27]. Briefly, U251 cells (5 × 10⁵ cells ml⁻¹) were seeded in 6-well plates and incubated overnight. Next, the cells were incubated with AgNPs@BSA, AgNPs@BSA-AS or 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP for 12 h. The concentration of AgNPs in all nanomaterial treatment groups was 20 μg ml⁻¹, and cells without treatment were used as the control. After washing three times with PBS to remove the excess nanoparticles, the cells were detached using 0.25% trypsin and collected by centrifugation (1500 rpm, 5 min, 4 °C). Then the TrxR activity was determined using a commercially available kit. The kit is based on the fact that TrxR catalyzes the reduction of 5,5'-dithiobis (2-nitrobenzoic) acid to 5-thio-2-nitrobenzoic acid, which generates a strong yellow color with maximum absorbance at 412 nm. The linear increase in absorbance at 412 nm over 10 min was monitored using a spectrophotometer (Beckman DU-600, CA, USA), and the TrxR activity rate was determined from the slope of absorbance at 412 nm versus time.

2.10. In vivo anti-tumor study

BALB/c nude mice (female, 16–18 g) were purchased from the Comparative Medicine Center of Yangzhou University (Yangzhou, China) and housed under standard animal care conditions. To investigate the in vivo anti-tumor efficacies of AgNPs@BSA, AgNPs@BSA-AS and 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP plus irradiation, nude mice bearing U251 tumors were used. Tumors were established by subcutaneous injection of 1 × 10⁷ U251 cells into the left posterior leg of nude mice. Nine days after tumor implantation, the mice were randomly divided into eight groups (eight mice per group): saline, AgNPs@BSA, AgNPs@BSA-AS, 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP, and their corresponding irradiation treatment groups. In the present study, saline and nanoparticle solutions were systemically administered via the tail vein. For irradiation, the mice were anesthetized, and then their tumors were locally exposed to a single dose of 6 Gy x-rays. The details of each treatment for mice (composition, concentration, administration volume and dosage) are summarized in table S1. After the above treatments, the tumor volume and mouse body weight were recorded every two days. The tumor volume was calculated with the formula: tumor length × (tumor width)² / 2. Finally, after completing the entire experiment, mice were sacrificed and their tumors and major organs were harvested and stained with hematoxylin and eosin (H&E) for histopathological evaluation. The Ag concentrations in tumors and different organs were measured using ICP-MS. The schedule of the mouse treatment protocols is shown in figure S1 (available online at stacks.iop.org/NANO/32/145102/mmedia).

2.11. Statistical analysis

The obtained data were reported as the mean ± standard deviation (SD). Data were analyzed using Student's t-test when two groups were compared, whereas a one-way analysis of variance was used to analyze differences among three or more groups. P values of less than 0.05 were regarded as statistically significant.

3. Results and discussion

3.1. Characterization results of the prepared nanoparticles

3.1.1. TEM results

TEM was performed to examine the morphology and particle size of AgNPs. As shown in figure 1(a), AgNPs were spherical in shape with good dispersibility. The mean size, determined by counting 200 particles on TEM images, was approximately 18 nm (figure 1(b)). Representative TEM images of AgNPs@BSA, AgNPs@BSA-AS and AgNPs@BSA-AS-VRP showed that the different functionally modified AgNPs were still spherical in shape and well dispersed with the mean size around 18 nm (figures S2(a)–(c)). These results suggested that the modification of BSA, AS1411 and VRP did not significantly change the morphology, dispersibility and particle size of the AgNPs.

3.1.2. UV–vis and FTIR results

UV–vis spectroscopy and FTIR spectroscopy were used to verify the successful construction of AgNPs@BSA-AS-VRP. In the UV–vis spectra, the prepared AgNPs showed a typical absorption peak at 407 nm (figure 1(c)), and then the absorption band of AgNPs was red-shifted to 415 nm after coating with BSA. Although BSA exhibited a characteristic absorption peak at 278 nm, this peak was not clearly shown in the spectrum of AgNPs@BSA. The probable reason was that the prepared AgNPs also had a certain absorption at this wavelength, and the absorption peak of BSA was covered by the strong peak of AgNPs. As illustrated in figure 1(d), aptamer AS1411 exhibited a strong characteristic absorption peak at around 260 nm. The spectrum of AgNPs@BSA-AS contained the characteristic absorption bands of both AS1411 and AgNPs@BSA, suggesting the successful construction of AgNPs@BSA-AS. In addition, the spectrum of AgNPs@BSA-AS-VRP showed a new peak at about 227.5 nm, which was consistent with the characteristic absorption band of VRP, indicating that AgNPs@BSA-AS-VRP were successfully constructed. The results of FTIR spectroscopy showed that the connection of AS1411 to the surface of AgNPs@BSA caused an increase in the intensity of the C=O stretching vibration band at approximately 1680 cm⁻¹ (figure 1(e)). Moreover, a new absorption peak corresponding to the C≡N bonds of VRP at 2235 cm⁻¹ was observed in the FTIR spectrum of AgNPs@BSA-AS-VRP. The FTIR results further confirmed the successful construction of AgNPs@BSA-AS-VRP.

3.1.3. DLS results

The images of particle size distribution, polydispersity index and zeta potential determined through Zetasizer Nano ZS were shown in figures S3(a)–(f), and the values of hydrodynamic diameters, polydispersity indexes and zeta potentials of AgNPs@BSA, AgNPs@BSA-AS and AgNPs@BSA-AS-VRP are summarized in table 1. The polydispersity indexes were all less than 0.3, suggesting that all these prepared nanoparticles were uniform in size [28].

Table 1. Hydrodynamic sizes, polydispersity indexes and zeta potentials of AgNPs@BSA, AgNPs@BSA-AS and AgNPs@BSA-AS-VRP.

Sample	Hydrodynamic size	Polydispersity index	Zeta potential
	(nm)		(mV)
AgNPs@BSA	75.21 ± 2.71	0.24 ± 0.026	−12.28 ± 1.01
AgNPs@BSA-AS	64.98 ± 8.01	0.22 ± 0.025	−13.37 ± 0.78
AgNPs@BSA-AS-VRP	93.13 ± 10.08	0.25 ± 0.014	−12.35 ± 1.12

3.1.4. The conjugation of AS1411 and VRP with the AgNPs@BSA

The calibration curve of AS1411 or VRP was linear with a correlation coefficient of R² = 0.99 (figures S4(a) and (b)), and the densities of AS1411 and VRP conjugated on AgNPs@BSA were 37.3 μg and 9.1 μg per 100 μg AgNPs@BSA, respectively. To elucidate the conjugation mechanisms of AS1411 or VRP with AgNPs@BSA, the AgNPs@BSA-AS, AgNPs@BSA-VRP and AgNPs@BSA-AS-VRP suspensions were mixed with ethylenediaminetetraacetic acid (EDTA), Triton X-100, urea or NaCl solutions, which could destroy the coordination covalent bonds, hydrophobic interactions, hydrogen bonds and electrostatic attraction interactions, respectively [29, 30]. After mixing with EDTA, Triton X-100 or NaCl, the colors of all suspensions mentioned above changed (figure S5), indicating that coordination covalent bonds, hydrophobic interactions and electrostatic attraction interactions played indispensible roles in the conjugation of AS1411 and VRP with AgNPs@BSA. The in vitro release rates of AS1411 and VRP were studied using the standard dialysis method [31]. The results showed that the release rates of AS1411 and VRP from the nanoparticles were relatively slow (figures S6(a) and (b)), suggesting the good stability of AgNPs@BSA-AS and AgNPs@BSA-AS-VRP.

3.2. Cytotoxicity of AgNPs@BSA, AgNPs@BSA-AS and AgNPs@BSA-AS-VRP

When nanomaterials are used in biomedical therapeutic applications, it is of great important to assess their cytotoxic potentials [32]. Particle size, shape, surface charge and modification can all affect the toxicity of nanoparticles [33]. In the current study, the toxicity of BSA, AS1411, VRP and three kinds of AgNPs with different surface modifications was studied by MTT assay. As shown in figure 1(f), AgNPs@BSA, AgNPs@BSA-AS and AgNPs@BSA-AS-VRP decreased viabilities of U251 cells in a concentration-dependent manner, and the corresponding IC₅₀ values were 107, 65 and 16 μg ml⁻¹, respectively. However, at the corresponding concentrations, BSA, AS1411 and VRP had no significant effect on cell viability. The cytotoxicity of AgNPs@BSA-AS was greater than that of AgNPs@BSA, which may be related to the AS1411 modification that can increase the cellular uptake of AgNPs by glioma cells through active targeting [34]. It was also observed that the cytotoxicity of AgNPs@BSA-AS-VRP was much higher than that of AgNPs@BSA-AS. This may be due to efflux pump inhibition by VRP, resulting in more pronounced accumulation of AgNPs in tumor cells. Generally, the best way of using nanomaterials is to strike a well balance between cytotoxicity and therapeutic efficiency [35]. Therefore, in the subsequent experiments, the concentration at which nanoparticles have potential radiosensitizing activity and no obvious toxicity should be considered and used.

3.3. Accumulation of AgNPs@BSA-AS and AgNPs@BSA-AS-VRP mixed in different ratios in glioma cells

To examine the accumulation of AgNPs@BSA-AS and AgNPs@BSA-AS-VRP mixed in different ratios in U251 glioma cells, dark-field imaging and ICP-MS analysis were performed. The data of dark-field imaging showed that only weak cell outlines were observed in the control group, while bright spots corresponding to the nanoparticles were clearly visible in the nanomaterial treatment groups (figure 2(a)). The number of intracellular nanoparticles in the AgNPs@BSA-AS treatment group was much higher than that in the AgNPs@BSA treatment group. This could be due to AS1411 specifically recognizing the nuleolin overexpressed on the plasma membrane of tumor cells, thereby enhancing cellular uptake via receptor-mediated endocytosis [36, 37]. Besides, the amount of intracellular nanoparticles increased with the addition of AgNPs@BSA-AS-VRP in the cell culture medium. When the mixing ratio of AgNPs@BSA-AS and AgNPs@BSA-AS-VRP was 19:1, the number of internalized nanoparticles increased significantly. Moreover, the intracellular concentrations of AgNPs were quantitatively measured by ICP-MS, and the results were consistent with that of dark-field imaging. As clearly revealed in figure 2(b), the modifications of aptamer AS1411 and VRP both helped to enhance the accumulation of AgNPs in tumor cells. Because 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP could already significantly increase the accumulation of nanoparticles in U251 glioma cells, and higher concentrations of AgNPs@BSA-AS-VRP may cause severe toxicity, so 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP were selected for the further experiments.

**Figure 2.** The accumulation of AgNPs@BSA-AS and AgNPs@BSA-AS-VRP mixed in different ratios in U251 glioma cells. (a) Representative dark-field images of U251 cells treated with AgNPs@BSA or AgNPs@BSA-AS and AgNPs@BSA-AS-VRP mixed in different ratios (1:0, 0:1, 9:1, 19:1 and 29:1). (b) Quantification of intracellular Ag concentration by ICP-MS. Data are presented as the mean ± SD (n = 3). ^* P < 0.05, ^** P < 0.01, ^*** P < 0.001 compared with the AgNPs@BSA treated group; ^### P < 0.001 compared with the 1:0 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP treated group.
Download figure:
Standard image High-resolution image

The fluorescent dye Rho 123 has been extensively used as an indicator of P-gp activity [38, 39]. The more Rho 123 accumulates in cells, the more severe the inhibition of P-gp efflux activity. In the current study, the accumulation of Rho 123 in U251 cells was detected by flow cytometry. The results showed that AgNPs@BSA and AgNPs@BSA-AS had no effect on the accumulation of Rho 123, but when AgNPs@BSA-AS-VRP were added, the accumulation of Rho 123 increased significantly (figure 3). The different effects of AgNPs@BSA-AS and AgNPs@BSA-AS-VRP on P-gp activity provided a possible explanation for our findings. That is to say, compared with AgNPs@BSA-AS, AgNPs@BSA-AS-VRP were able to inhibit P-gp efflux activity, thereby causing more nanoparticles being remained in tumor cells. The above results indicated that 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP could effectively accumulate in tumor cells by increasing cellular uptake and inhibiting efflux. It was also found that a single 4 Gy dose of x-ray irradiation caused a remarkable increase in Rho 123 accumulation with respect to the control group, suggesting that 4 Gy irradiation can significantly decrease the P-gp activity. Increased levels of P-gp were observed in fractionally irradiated tumor cells, while a single low-dose of x-ray irradiation often resulted in a decrease in P-gp activity [40]. Our experimental results were consistent with the previous studies. Surprisingly, the effect of 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP plus irradiation on the accumulation of Rho 123 in tumor cells was similar to that of AgNPs@BSA-AS-VRP plus irradiation. One possible reason for this phenomenon is that x-rays have a greater impact on P-gp activity than the above materials, which masks the difference in P-gp efflux activity between AgNPs@BSA-AS-VRP and 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP.

**Figure 3.** Flow cytometry analysis of the effects of AgNPs@BSA or different mixing ratios of AgNPs@BSA-AS and AgNPs@BSA-AS-VRP with or without 4 Gy x-ray irradiation on the accumulation of Rho 123 in U251 cells. U251 cells were incubated with AgNPs@BSA or AgNPs@BSA-AS and AgNPs@BSA-AS-VRP mixed in different ratios (1:0, 0:1 and 19:1). IR in the figure is an abbreviation for irradiation, and MFI in the figure is an abbreviation for mean fluorescence intensity.
Download figure:
Standard image High-resolution image

3.4. Radiosensitizing effect of 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP

In the present study, the radiosensitizing potential of 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP was tested both in vitro and in vivo. In vitro, colony formation assay was used to evaluate the effects of materials on the radiosensitivity of U251 cells, and the dose-survival curves were obtained. As shown in figure 4, the surviving fraction decreased with increasing irradiation dose. Compared with irradiation alone, the surviving fractions of U251 cells treated with AgNPs@BSA, AgNPs@BSA-AS or 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP plus irradiation were significantly decreased, and the SER values were 1.29, 1.42 and 1.55, respectively. These data suggested that 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP had the strongest ability to enhance the radiosensitivity of glioma cells among the above three groups of nanoparticles.

**Figure 4.** Effects of AgNPs@BSA, AgNPs@BSA-AS and 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP plus irradiation on colony formation of U251 cells. Data are presented as the mean ± SD (n = 3).
Download figure:
Standard image High-resolution image

In the in vivo experiment, nude mice bearing U251 tumors were injected with saline, AgNPs@BSA, AgNPs@BSA-AS or 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP via the tail vein, and then their tumors were irradiated 6 h post-injection. The anti-tumor activity of 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP combined with radiotherapy was evaluated by examining changes in tumor volume and tumor weight. As shown in figure 5(a), the tumor volume increased rapidly in the control group and only nanoparticles treatment groups during the experimental period. In marked contrast, the tumor growth was suppressed in the nanoparticles plus irradiation treatment groups. At the endpoint of the experiment (20 d after completing the treatment), the tumor volumes and tumor weights in the nanoparticles plus irradiation treatment groups were significantly smaller than that in only nanoparticles treatment groups (figures 5(a)–(c)). More importantly, in irradiation treatment groups, compared with AgNPs@BSA and AgNPs@BSA-AS, 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP significantly inhibited tumor growth. The therapy efficacy was further confirmed by the pathological results. The representative H&E stained tumor slices from different groups were given in figure 6. The tumor tissues of the AgNPs@BSA, AgNPs@BSA-AS, 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP and irradiation groups were only slightly influenced, and that of the AgNPs@BSA and AgNPs@BSA-AS plus irradiation treatment groups were partially damaged. While the tumor cells of the 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP plus irradiation treatment group were severely destroyed, and the cell number was greatly reduced. Studies have shown that within a certain concentration range, the radiosensitization effect is related to the amount of radiosensitizers in the tumor area [41]. In our study, at the end of the experiment, the amount of Ag in tumor tissues treated with 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP was still greater than that of AgNPs@BSA and AgNPs@BSA-AS (figure 7). Therefore, it was well explained why the radiosensitizing effect of 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP was stronger than that of AgNPs@BSA and AgNPs@BSA-AS. Changes in body weight of nude mice can indirectly reflect the safety of nanoparticles in vivo [42]. The body weights of the mice were monitored and recorded every two days. As shown in figure 5(d), the body weights of mice only treated with nanomaterials increased slowly, while the weights of mice treated with nanomaterials combined with x-rays decreased slightly and then gradually increased to the initial or higher levels, indicating that these nanoparticles have no significant systemic toxicity to tumor-bearing nude mice. Further, the main organs of the mice with different treatments were stained with H&E (figure 6). Compared with the control group, heart, liver, spleen, lung and kidney of the mice treated with AgNPs@BSA, AgNPs@BSA-AS and 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP with or without irradiation had normal cell morphology, no signs of steatosis, inflammation or fibrosis, confirming the good biosafety of the synthesized radiosensitizers.

**Figure 5.** Anti-tumor efficacies of AgNPs@BSA, AgNPs@BSA-AS and 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP plus irradiation in nude mice bearing U251 tumors. (a) Changes in tumor volume with time after different treatments. (b) Representative pictures of tumors collected from tumor-bearing mice with different treatments at the end of the experiments. (c) Tumor weight of each group at the end of the experiments. (d) Changes in body weight of mice bearing glioma with time after different treatments. IR in the figure is an abbreviation for irradiation. All the data are presented as the mean ± SD (n = 8).
Download figure:
Standard image High-resolution image

**Figure 6.** H&E staining of the tumors and major organs of the tumor-bearing nude mice after different treatments. At the end of the experiments, mice treated with AgNPs@BSA, AgNPs@BSA-AS or 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP with or without 6 Gy x-rays irradiation were sacrificed, and their tumors and major organs were stained with H&E for histopathological evaluation. IR in the figure is an abbreviation for irradiation.
Download figure:
Standard image High-resolution image

**Figure 7.** Ag concentrations in tumors and major organs at the end of experiments. Nude mice bearing U251 tumors were injected with saline, AgNPs@BSA, AgNPs@BSA-AS or 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP via the tail vein, and their tumors were irradiated 6 h post-injection. After completing the entire experiment, the concentrations of Ag in tumors and different organs were measured by ICP-MS. IR in the figure is an abbreviation for irradiation. Data are presented as the mean ± SD (n = 8).
Download figure:
Standard image High-resolution image

The ideal radiosensitizers should have at least the following characteristics: (1) significant enhancement of radiation therapy efficacy; (2) excellent tumor targeting ability, so that they can be preferentially accumulated in tumor tissues; (3) good biocompatibility without obvious toxicity [43, 44]. In this study, both in vitro and in vivo experiments distinctly showed that 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP had a significant dose enhancement effect on glioma exposed to x-rays. Additionally, the results of ICP-MS and H&E staining demonstrated that they were highly enriched in tumor tissues, without showing any obvious systemic toxicity to tumor-bearing nude mice. The above results indicated that 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP were promising radiosensitizers.

3.5. 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP decreased the TrxR activity of U251 cells

The mechanism of radiosensitization caused by AgNPs has not been well elucidated. Several mechanisms have been proposed, but the most widely accepted mechanism is that AgNPs enhance the efficacy of radiotherapy by triggering oxidative stress and DNA damage [45, 46]. The thiroredoxin (Trx) system, composed of Trx, TrxR and nicotinamide adenine dinucleotide phosphate, is one of the main detoxification systems in the body and plays an important role in cell growth, defense against oxidative stress and apoptosis [47]. TrxR is an essential enzyme for cellular redox balance and has been suggested as an important therapeutic target for brain cancers [48]. Recent studies have shown that inhibiting TrxR makes cancer cells sensitive to radiotherapy [49–51]. However, it is still unclear whether TrxR is involved in AgNPs-induced radiosensitization. In the present study, the activity of TrxR in U251 cells incubated without or with AgNPs@BSA, AgNPs@BSA-AS or 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP was investigated. As shown in figure 8, the activities of TrxR significantly decreased in all nanoparticles treatment groups compared with the control group. In U251 cells, 20 μg ml⁻¹ AgNPs@BSA, AgNPs@BSA-AS or 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP inhibited TrxR activity by 55%, 64% and 79%, respectively. Researchers have found that metal-based nanomaterials inhibit TrxR activity in a dose-dependent manner [52]. Based on this, we speculated that the significant inhibition of TrxR activity in 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP treatment group may be due to the increased accumulation of AgNPs in tumor cells. Moreover, Penninckx S and colleagues [53] found that AuNPs can inhibit TrxR activity before irradiation exposure, thereby weakening the detoxification systems to exert their radiosensitizing effect. The results of correlation studies confirmed the strong correlation between the AuNPs-induced TrxR activity inhibition and the enhanced radiotherapy [53]. As nano-radiosensitizers, AgNPs and AuNPs have many similarities, especially their radiosensitization mechanisms both involve the induction of oxidative stress. Therefore, it can be speculated that the inhibition of TrxR activity caused by AgNPs may also play an important role in their radiosensitization.

**Figure 8.** Measurement of TrxR activity in U251 cells. (a) Effects of AgNPs@BSA, AgNPs@BSA-AS or 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP exposure on TrxR activity in U251 cells. (b) TrxR activity rate calculated from the slope of TrxR activity curves. Data are presented as the mean ± SD (n = 3).
Download figure:
Standard image High-resolution image

4. Conclusion

In the present study, we synthesized AS1411 and VRP conjugated BSA coated AgNPs (AgNPs@BSA-AS-VRP), and then focused on evaluating the effects of 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP on the P-gp activity and radiosensitivity of glioma cells. It was found that 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP could effectively accumulate in tumor cells by increasing endocytosis and inhibiting efflux. The corresponding SER value was 1.55, suggesting their excellent potential for enhancing the efficacy of radiotherapy. The results of anti-tumor experiment in U251 tumor-bearing mice further confirmed the significant radiosensitizing effect of 19:1 mixed AgNPs@BSA-AS and AgNPs@BSA-AS-VRP. In addition, TrxR might be involved in the AgNPs-induced radiosensitization. These findings are of great significance for the development of high-efficient AgNPs-based radiosensitizers to enhance the radiotherapy of glioma and other cancers.

Acknowledgments

This research was funded by the National Natural Science Foundation of China (81771980, 81571805 and 81703758), and the National Key Basic Research Program of China (973 Program; 2013CB933904).

Conflicts of interest

The authors declare no conflict of interest.

Ethical statement

Animal experiments were conducted in accordance with the principles outlined by the American Physiological Society and the Ethics Committee of Southeast University. All our protocols for animal care and use were reviewed and approved by the Ethics Committee of Southeast University (Approval ID 20180301006).