Enriched Au nanoclusters with mesoporous silica nanoparticles for improved fluorescence/computed tomography dual‐modal imaging

Abstract Objectives Au nanoclusters (AuNCs) have been used widely in fluorescence bio‐imaging because of their good fluorescence, small particle size and non‐cytotoxicity. AuNCs are also efficient in computed tomography (CT) imaging. Hence, a dual‐modal imaging probe can be constructed without any complicated modification processes by exploiting the excellent performance of AuNCs. In the present study, AuNCs were enriched with mesoporous silica nanoparticles (MSNs) to obtain enhanced fluorescence/CT dual‐modal imaging, which was capable of acquiring more imaging information for diseases compared with single‐mode imaging. Materials and methods Biocompatible bovine serum albumin (BSA)‐capped AuNCs were prepared and loaded into amine‐functionalized MSNs to form MSN@AuNCs. BSA‐AuNCs, MSNs, and MSN@AuNCs were characterized by ultraviolet‐visible (UV‐vis) spectra, transmission electron microscopy (TEM), fluorescence spectra, and zeta potential. CT imaging was recorded using micro‐CT scanning. Fluorescence imaging was measured using confocal laser scanning microscopy and flow cytometry. Results The prepared AuNCs and MSNs possessed good properties as previously reported. The fluorescence intensity and CT value of the AuNCs were enhanced after being enriched with MSNs. The nanoparticles were both non‐cytotoxic. Confocal laser scanning microscopy and flow cytometry indicated that MSN@AuNCs in CAL‐27 cells showed improved fluorescence imaging compared with simple AuNCs at the same concentration. Conclusions The results revealed that the strategy of enriching AuNCs with MSNs can obtain highly sensitive fluorescence/CT dual‐modal imaging, which indicated the potential of this nanoparticle in the diagnosis and treatment of disease.


| INTRODUC TI ON
Near-infrared (NIR) fluorescence imaging has shown great potential in the field of biological imaging because of its advantages, such as high imaging sensitivity and strong operability. [1][2][3] However, the further application of NIR fluorescence imaging probes to obtain high brightness is limited by concentration requirements and low luminous efficiency. Moreover, fluorescence imaging lacks high spatial resolution, which adversely affects the identification of anatomical information. Computed tomography (CT) is an approved diagnostic tool that is able to obtain abundant anatomical information; however, it has relatively low sensitivity. [4][5][6] Dual-modal imaging, integrating two imaging methods, can provide complementary information and thus might satisfy the requirements for accurate diagnosis of diseases compared with single-modal imaging. [7][8][9] To construct fluorescent/CT dual-modal imaging probes, two imaging techniques are required to realize in one imaging reagent. Currently, dual-modal imaging probes based on NIR fluorescence usually use nanoparticles as carriers, and different agents are capped with the nanoparticles through physical or chemical approaches. [10][11][12] Generally, the nanoparticle carriers include liposomes, dendrimers, polymer micelles, carbon nanotubes and SiO 2 . 13,14 With the fluorescence and CT imaging reagents loaded on, the modified nanoparticles are then used for dual-modal imaging. Nevertheless, the conjugation of different media requires complex processes such as synthesis, modification, assembly and purification, which will inevitably cause various deficiencies, such as an increase in the particle size, poor dispersion and a decrease in the respective imaging effect. 15 Hence, it is important to develop a simple and high brightness dual-modal imaging probe for imaging analysis.
Among numerous fluorescence agents, gold nanoclusters (AuNCs) are acknowledged for their excellent NIR fluorescence performance and low cytotoxicity. 16,17 Furthermore, AuNCs possess CT imaging efficiency, which makes them suitable as CT contrast reagents. [18][19][20] Therefore, AuNCs can be used directly as fluorescence/CT dual-modal imaging probes. For instance, Mao's group reported the application of AuNCs in fluorescence/CT dual-modal imaging for tumour localization. 21 However, imaging probes based on simple AuNCs exhibited low luminous efficiency and poor imaging sensitivity in vitro and in vivo. 22 Furthermore, the quantum yield of AuNCs was only 6.85% according to our previous study. As an exogenous nanomaterial, the high concentration required to improve imaging efficiency might lead aggregation-induced quenching and bring a certain level of toxicity. Hence, it remains challenging to develop an enhanced fluorescence and CT dual-modal imaging probe. Specially, it is feasible to improve the imaging sensitivity of fluorescence by increasing the concentration of the AuNCs without aggregation-caused quenching because of the large Stokes shift of the red-emitting AuNCs. 23,24 Recently, the specific structure and uniform pores of biocompatible mesoporous silica nanoparticles (MSNs) have led to their approval as an enriching contrast medium to obtain increased concentrations of nanoparticles. 25 Bovine serum albumin (BSA), a type of qualified protein modifier, can be capped with AuNCs to supply extra properties such as reducing and stabilizing the nanoparticles. 28,29 Besides, BSA is widely used as a coating material in drug delivery because of its high biocompatibility and simple capping procedure. 30,31 In the present study, BSA-functionalized AuNCs were synthesized in one step and then enriched into the pores of biocompatible MSNs through electrostatic interactions and carboxy-amine bonds (Scheme 1). The local concentration of AuNCs was increased apparently, and both fluorescence and CT imaging were improved simultaneously. To compare the fluorescence and CT imaging sensitivity of enriched nanoparticles with that of simple AuNCs, the nanoparticles were characterized using physical and chemical measurements, and in vitro experiments.

| Preparation of BSA-capped AuNCs
Following a previously method reported by Ding's group, the BSAcapped AuNCs were prepared by a simple process. 32 Briefly, HAuCl 4 (5 mL, 10 mM) and BSA (5 mL, 50 mg/mL) were mixed and stirred quickly at 37 ℃ for 2 minutes. NaOH solution (0.5 mL, 1 M) was subsequently added to adjust the pH value. The reaction mixture was kept under vigorous stirring at 37 ℃ for 12 hours. Finally, a dark brown solution was obtained. The BSA-functionalized AuNCs solution was maintained at 4℃ for further experimentation.

| Preparation of NH2-MSNs
According to a previously reported approach, amine-functionalized MSNs were prepared based on using cationic surfactant CTAC and TEA as the template. 24 The CTAC solution (48 mL, 25 wt%) and TEA (0.36 g) were added simultaneously into 72 mL of distilled water and rigorously stirred at 60 ℃ for 1 hour. Then, TEOS in cyclohexane (40 mL, 5 v/v%) was slowly dropped into the reaction system and incubated under gentle stirring for another 12 hours. The obtained white jelly was allowed to cool to room temperature naturally before being washed with ethanol (10 864 × g, 10 minutes) for several times. The resultant MSNs were dispersed in NH 4 NO 3 solution and then refluxed at 60°C for 6 hours twice to remove the template.
To gain amine-functionalization MSNs, 0.1 mL APTS was dropped into an MSN ethanol solution and refluxed at 78°C for 12 hours under gentle stirring. After that, to purify the products, they were centrifuged three times at 10 864 × g for 10 minutes and finally dispersed in distilled water. The NH 2 -MSNs were stored in 4°C for further use.

| Preparation of MSN@AuNCs
Typically, 1 mL BSA-AuNCs solution was first reacted with EDC (2 mg) and NHS (1 mg). After a few minutes, 1 mL of NH 2 − MSNs (10 mg/mL) was added into the reaction system, which was further stirred at room temperature for 4 hours. The obtained MSN@ AuNCs were washed three times by centrifugation (10 864 × g, 10 minutes) using distilled water to remove the supernatant and then re-dispersed with distilled water. The resultant products were store at 4°C.

| Characterizations of AuNCs, MSNs and MSN@AuNCs
The particle sizes and zeta potential were measured using a nanoparticle analyser SZ-100 (Horiba Scientific, Kyoto, Japan). Inductively

| CT imaging in vitro
AuNCs and MSN@AuNCs at a concentration range of 0-5 mM were  supplemented with 10% FBS and 1% penicillin-streptomycin. The medium was renewed every two days.

| Cell cytotoxicity of AuNCs and MSN@AuNCs
When the cells had proliferated to 80%, they were harvested for experimentation. Cell cytotoxicity was measured using the CCK-8 assay. Briefly, L929 cells, CAL-27 cells, ACC-2 cells and SCC-25 cells were counted and seeded in 96-well plates at a density of 10 4 cells per well, which contained 100 μL of complete medium. Then, the cells were cultured in a 37°C incubator with 5% CO 2 overnight to adhere. Afterwards, the cells were washed with PBS, and new complete medium with various concentrations of AuNCs and MSN@AuNCs (calculated by the concentration of AuNCs) from 100 to 800 nM was added into wells. Five wells for each condition were set up as positive controls. Then, the cells were incubated for another 48 hours. After exposure, the medium was removed, and cells were washed with PBS. Subsequently, 100 μL FBS-free medium with 10% CCK-8 was added to wells. After approximately 2 hours, the medium turned to yellow or brown. The cell cytotoxicity of AuNCs and MSN@AuNCs was analysed by detecting the absorbance at 450 nm with a microplate reader (Thermo Scientific, Waltham, MA, USA).

| Statistical analysis
The experiments in this study were all repeated at least three times under the same conditions. All data were analysed using an independent t test or one-way ANOVA using SPSS 26.0, and a P value < 0.05 was considered statistically significant.

| Preparation and characterization of AuNCs, MSNs and MSN@AuNCs
Based on AuNCs and MSNs, a highly sensitive fluorescence/CT dual-modal imaging probe was constructed through enrichment, without complicated modification and assembly, aiming to realize The TEM images indicated that the AuNCs were loaded into the MSNs. As shown in Figure 2B  Furthermore, the fluorescence lifetime was determined by the luminescence decay curves in Figure 3D. There was no significant difference between the AuNCs and the MSN@AuNCs, which indicated that the fluorescence of the AuNCs was stable, without enrichment caused self-absorption after assembly.

| CT imaging compared between AuNCs and MSN@AuNCs in vitro
AuNCs enriched into MSN were expected to enhance the CT intensity compared with that of simple AuNCs. The feasibility of MSN@AuNCs as a sensitive CT reagent was tested by CT scanning in vitro as shown in Figure 4. The CT intensity of both AuNCs and MSN@AuNCs continuously increased as the concentration of AuNCs increased from 1 to 5 mM. In addition, the intensity of MSN@AuNCs was obviously higher than that of the simple AuNCs at the same concentrations ( Figure 4A). Moreover, several highdensity shadows, representing MSN@AuNCs aggregation and sedimentation, could be observed, which reflected the enhanced CT imaging capabilities of the MSN@AuNCs. Figure 4B showed that there was a good linear relationship between the MSN@ AuNCs and the HU values, and between the AuNCs and the HU values. The slope of the MSN@AuNCs curve was 2-fold greater than that of the AuNCs curve.

| Cell cytotoxicity of AuNCs and MSN@AuNCs
The cytotoxicity of the AuNCs and MSN@AuNCs towards L929 cells, CAL-27 cells, ACC-2 cells, and SCC-25 cells was measured using the CCK-8 assay. After cultured with AuNCs and MSN@AuNCs in a diverse concentration range of 100 to 800 nM for 48 hours, the L929 cells were not affected by any concentration, and the cell viability in 600 nM MSN@AuNCs was statistically higher than that in the control ( Figure 5A). For CAL-27 cells, there was no increase in cytotoxicity by AuNCs and MSN@AuNCs incubation after 48 hours, even at 800 nM ( Figure 5B). Therefore, no significant decline in the cell viability at concentrations from 100 to 800 nM was detected after 48 hours of incubation. Furthermore, the cytotoxicity of AuNCs and MSN@AuNCs was verified in ACC-2 cells and SCC-25 cells, and no significant cytotoxicity was detected ( Figure 5C and 5D). These

| Fluorescence imaging and intensity of cells in vitro
The fluorescence BSA-AuNCs and MSN@AuNCs were prepared with the aim of comparing the feasibility of enhancing the fluorescence of CAL-27 cells. Confocal microscopy analysis was used to observe the fluorescence intensity of the AuNCs and MSN@AuNCs. As exhibited in Figure 6A, after incubation for 6 hours, red-emitted fluorescence was observed distinctly in the cytoplasm and membrane of CAL-27 cells. Furthermore, the cells incubated with MSN@AuNCs emitted brighter fluorescence compared with those incubated in simple AuNCs, according to the confocal laser microscopy images and interactive 3D surface plots, which indicated that AuNCs enrichment could enhance the fluorescence imaging effect. In addition, L929 cells were measured under the same conditions. However, the fluorescence in the L929 cells was weak using both nanoparticles. A previous study showed that BSA is a functional protein that could stabilize nanomaterials and selectively recognize tumour cells. 32 The capacity of BSA and the ability to absorb external substances of tumour cells led to the higher fluorescence intensity in CAL-29 cells than in L929 cells. Therefore, enriching AuNCs into MSNs was feasible to enhance the fluorescence imaging effect in CAL-27 cells.
To obtain the quantitative analysis results of the fluorescence in-  High brightness NIR fluorescence/CT dual-mode imaging probes can be used widely for tumour imaging analysis. Moreover, as a superior drug delivery medium, the imaging probes could be loaded with anti-tumour drugs for tumour-targeted treatment. The dualmodal imaging probes are of great scientific significance to achieve F I G U R E 6 (A) Confocal laser scanning microscopy images and interactive 3D surface plots of CAL-27 cells and L929 cells incubated with AuNCs and MSN@AuNCs for 6 hours; (B) average fluorescence intensity of CAL-27 cells and L929 cells incubated with AuNCs and MSN@ AuNCs for 12 and 16 hours by flow cytometry. Data are presented as mean ± SD (n = 3). *P < .05, **P < .005, ***P < .002 high-sensitivity live tumour multimode imaging, to construct novel drug delivery systems, and could promote the development of clinical cancer precision diagnosis and treatment.

| CON CLUS IONS
In summary, multimodal imaging technology, achieved by integrating two or more imaging functions, can complement single-mode imaging technology and obtain more information on lesions through the detection of multiple types of images, which is significant to promote the development of accurate clinical disease diagnosis. [33][34][35] In our study, AuNCs with both fluorescence imaging and CT imaging functions were enriched into the non-toxic biocompatible material MSNs using simple methods, which made up for the lack of spatial resolution in NIR fluorescence imaging and enhanced the imaging sensitivity of each mode, thereby obtaining high brightness fluorescence and CT imaging effects. We believe our dual-modal imaging probes have a powerful potential for application in the diagnosis and treatment of disease.

ACK N OWLED G EM ENTS
We gratefully acknowledge the financial support from the Sichuan

CO N FLI C T O F I NTE R E S T
No conflict of interest was declared of this article.

AUTH O R CO NTR I B UTI O N S
Yifang Yuan performed the experiments, analysed the data, con-

DATA AVA I L A B I L I T Y S TAT E M E N T
The data, supporting the findings of this work, are available from the corresponding author upon reasonable request.