Gadolinium oxysulfide-coated gold nanorods with improved stability and dual-modal magnetic resonance/photoacoustic imaging contrast enhancement for cancer theranostics

Abstract

Gold nanorods (GNRs) are emerging as a promising nanoplatform for cancer theranostics because of their unique optical properties. However, they still suffer from many limitations, such as high cytotoxicity, low thermodynamic and biological stability, and a tedious process for integrating other imaging modalities, for further practical biomedical applications. In this work, a strategy by one-step coating of Gd₂O₂S around GNRs is reported to address these limitations of GNRs. After the coating of the Gd₂O₂S shell, the as-fabricated Gd₂O₂S coated GNRs (GNRs@Gd₂O₂S) show enhanced biocompatibility and photostability, and tunable localized surface plasmon resonance. The strong absorption in the near-infrared region renders GNRs@Gd₂O₂S outstanding photoacoustic imaging and photothermal therapy capabilities. Moreover, owing to the T₁ shortening ability of Gd₂O₂S, the GNRs@Gd₂O₂S also show an excellent T₁ MRI contrast performance. The GNRs@Gd₂O₂S thus can serve as a versatile nanoplatform for cancer theranostics combining dual-modal imaging and photothermal therapy.

---

The ability to create agents incorporating multiple functionalities, such as targeting, imaging and therapy, is one of the main advantages of nanomedicine.^1–3 In particular, the combination of diagnostic and therapeutic moieties into a single entity confers the theranostic nanoparticles’ unique capabilities.^4–6 Due to the versatility of these nanoparticles, as well as their potential applications, interest in their synthesis and utility is rapidly increasing in recent years.^7–11 Among the nanostructures being actively explored, gold nanorods (GNRs) are emerging as a promising nanoplatform for cancer theranostics because of their unique optical properties.^12–14 In particular, the high absorption and tunable localized surface plasmon resonance (LSPR) across the near-infrared (NIR) region where tissues are nearly transparent, make GNRs an excellent candidate for photoacoustic imaging (PAI) and as photothermal therapy (PTT) agents.^15–18 Despite these promising prospects of GNRs, many challenges remain to be addressed for further practical bioapplications, especially cancer theranostics. First, the typically synthesized GNRs are usually capped with stabilizing agents, such as cetyltrimethylammonium bromide (CTAB). The presence of these surfactant molecules would bring in toxicities to biological species at different levels.^19,20 The removal of CTAB to reduce cytotoxicity by repeated centrifugation or ligand exchange reactions, however, may cause the aggregation and LSPR blue shift of GNRs. Second, the GNRs often suffer from poor photothermal stability due to the morphology melting and changing under the photothermal heating or femtosecond/nanosecond laser irradiation.^21–23 The laser or high temperature induced melting effects lead to the shift of the longitudinal LSPR peak of GNRs from the desirable near-infrared (NIR) region to the visible region. Besides, when GNRs enter a complex environment, the occurrence of clustering and aggregating could also lead to the blue shift of the LSPR peak of GNRs.^24–26 The loss of NIR deep-tissue penetration capability greatly hinders the application of GNRs for PTT. Third, the integration of other clinically used imaging modalities, such as magnetic resonance imaging (MRI), into the GNR nanoplatform in a facile and efficient way remains challenging. For example, current approaches for the fabrication of GNR nanoplatform with MRI function are typically through covalent chemistry, involving the covalent binding of GNRs and pre-prepared MRI contrast agents (e.g. Fe₃O₄ and Gd-based compounds).^27–30 These covalent attachment strategies are complex and tedious and need a prior surface modification of the nanoparticles. Currently, to improve the biocompatibility and photostability of GNRs, the most common and effective method is coating biocompatible shells, such as silica, polymers, metal oxides and metal chalcogenides, around GNRs.^26,31–38 The addition of protective shells imparts some steric constraint against reshaping and aggregating, thus provides better optical stability of GNRs. However, current coating materials, either silica or polymers, are usually monofunctional that merely serve as protective shells without an intrinsic imaging function. For multimodal cancer theranostics, the exploitation of novel coating materials with imaging abilities is highly desired to combine the LSPR properties of GNRs.

To address these challenges, we herein demonstrate our efforts in fabricating the gadolinium oxysulfide-coated GNRs (GNRs@Gd₂O₂S) as a novel cancer theranostic nanoplatform (Scheme 1). The GNRs@Gd₂O₂S can be synthesized by a one-step coating of the Gd₂O₂S shell without complicated modification. By controlling the thickness of the Gd₂O₂S shell, the LSPR peak of GNRs@Gd₂O₂S can be tuned to the NIR spectral region, which is highly beneficial for PTT and PAI. In addition, the coating of the Gd₂O₂S shell enhances the biocompatibility and photostability during photothermal heating or in complex biological environments of GNRs. More importantly, due to the ability of the Gd₂O₂S shell to shorten T₁ relaxation times, the GNRs@Gd₂O₂S also exhibit excellent T₁ MRI contrast enhancement. Combined with the PAI and PTT capabilities of the GNR core, the as-obtained GNRs@Gd₂O₂S thus can serve as a versatile and effective nanoplatform for in vivo cancer theranostics.

Scheme 1 Schematic illustration of the synthesis of GNRs@Gd₂O₂S by incubating GNRs with Gd³⁺, ascorbic acid (AA), hexamethylenetetramine (HMT) and thioacetamide (TAA) at 80 °C and their potential applications in MRI, PAI and PTT.

Results and discussion

We first synthesized CTAB-capped GNRs according to the previous method with a slight modification.³⁹ The transmission electron microscopy (TEM) image showed that the as-synthesized GNRs were 55.7 ± 7.0 nm in length and 25.4 ± 3.2 nm in width, respectively (Fig. 1a). We then coated the GNRs with a Gd₂O₂S shell via a facile one-step process using Gd(NO₃)₃ as a metal precursor in the presence of ascorbic acid (AA) and thioacetamide (TAA) under mild reaction conditions. Previous studies have demonstrated that AA could induce the formation of AA–metal hydroxide complexes, and the complexes interact with the CTAB capping on the surface of the nanoparticles and subsequently form metal oxide shells via a dehydration process.⁴⁰ TAA serving as a S source could also combine with metal ions and eventually lead to the formation of a metal oxysulfide shell deposited onto GNRs.⁴¹ To obtain better water dispersibility, the as-prepared GNRs@Gd₂O₂S were further modified with polyacrylic acid (PAA) through a simple ultrasonic treatment. The thickness of the Gd₂O₂S shell can be easily tuned by controlling the added concentration of the Gd(NO₃)₃ precursor. For example, we synthesized GNRs@Gd₂O₂S with different Gd₂O₂S shell thicknesses of 4 nm, 8 nm, and 20 nm, respectively (Fig. 1b–d). The high-resolution TEM (HRTEM) image revealed that the lattice spacing in the GNR core was about 0.235 nm, corresponding to the (111) plane of the face-centered cubic (fcc) Au, while the Gd₂O₂S shell was amorphous (Fig. 1e). The X-ray powder diffraction (XRD) pattern of GNRs@Gd₂O₂S showed no diffraction peaks of Gd₂O₂S, confirming that the Gd₂O₂S shell was amorphous (Fig. S1†). The survey scan X-ray photoelectron spectroscopy (XPS) spectrum of GNRs@Gd₂O₂S showed photoelectron lines at a binding energy of about 9 eV, 22 eV, 142 eV, 171 eV, 272 eV, 284 eV, 531 eV, 976 eV, 1187 eV and 1220 eV attributed to Gd 4f, Gd 5p, Gd 4d, S 2p, Gd 4p, C 1s, O 1s, O KLL, Gd 3d_5/2, and Gd 3d_3/2,⁴² respectively (Fig. S2†). All these peaks are very similar to that of previously reported Gd₂O₂S:Eu³⁺ nanostructures.⁴³ TEM-associated energy-dispersive X-ray spectroscopy (EDS) analysis showed typical peaks of Gd, O, and S, and the atomic ratio of Gd : O : S was about 25.92 : 27.54 : 12.03, indicating the successful coating of Gd₂O₂S (Fig. S3†). The energy-dispersive X-ray (EDX) elemental mapping analysis of GNRs@Gd₂O₂S further revealed that the coating of Gd₂O₂S on the nanorods was homogeneous (Fig. 1f). The mass ratio of Au and Gd in GNRs@Gd₂O₂S with a 20 nm shell thickness determined by inductively coupled plasma mass spectrometry (ICP-MS) is about 1.82 ± 0.13.

Fig. 1 TEM images of (a) GNRs, and GNRs@Gd₂O₂S with different shell thicknesses of (b) 4 nm, (c) 8 nm, and (d) 20 nm, respectively. Inset: Corresponding photographs of GNRs and GNRs@Gd₂O₂S with different shell thicknesses. (e) High-resolution TEM image and (f) EDX mapping and HAADF-STEM (inset) images of GNRs@Gd₂O₂S.

In general, a strong resonance peak in the NIR region is desirable for biomedical applications due to the high transparency of soft tissues in this region.^18,44 By controlling the thickness of the Gd₂O₂S shell, we can tune the longitudinal LSPR peak of GNRs@Gd₂O₂S to the NIR region. The as-prepared GNRs had a longitudinal LSPR peak at 700 nm. After coating the Gd₂O₂S shell of 4 nm, 8 nm and 20 nm, the longitudinal LSPR peak was red-shifted to 754 nm, 788 nm and 818 nm, respectively (Fig. 2a). Increasing the Gd₂O₂S shell thickness could increase the effective dielectric constant surrounding the GNR cores, therefore causing an increased red shift of the LSPR modes.¹³ The strong absorbance in the NIR region indicates that GNRs@Gd₂O₂S are highly appropriate for biomedical applications. We therefore only investigated the performance of GNRs@Gd₂O₂S with a 20 nm Gd₂O₂S shell thickness in the subsequent experiments because of their absorbance at a longer wavelength of 818 nm. To evaluate the photothermal conversion capability of GNRs@Gd₂O₂S, the GNR@Gd₂O₂S aqueous solution was exposed to an 808 nm laser and the heating curves and images were collected. Irradiated by the 808 nm laser, the GNR@Gd₂O₂S aqueous solutions showed a rapid increase in temperature within 5 min, and the temperature rising rate and the final temperature were proportional to the concentration and laser power density (Fig. 2b and c). The photothermal conversion efficiency of GNRs@Gd₂O₂S was calculated as about 22.6%, which was similar to that of GNRs (23.2%), indicating that the coating of the Gd₂O₂S shell does not influence the photothermal conversion efficiency of GNRs@Gd₂O₂S (Fig. S4†). However, the GNRs@Gd₂O₂S exhibited enhanced photothermal stability compared to the GNRs. Over four laser irradiation cycles, the photothermal conversion efficiency of GNRs@Gd₂O₂S remained stable without significant changes, while a downtrend could be observed in that of GNRs (Fig. 2d). The laser illumination induced instability of GNRs not only influence the photothermal effect of GNRs, but also may hinder GNRs as PAI agents, since the photon fluencies in PAI always induce a shape deformation of GNRs.⁴⁵ To further confirm the photostability of GNRs@Gd₂O₂S, we investigated the spectrum and morphology changes of GNRs@Gd₂O₂S under the irradiation of high-fluence nanosecond laser pulses (7 ns pulse duration, 10 Hz). We also synthesized GNRs with a longitudinal LSPR peak at about 815 nm (60.9 ± 4.9 nm in length and 13.4 ± 2.5 nm in width) as control samples. Under the nanosecond laser irradiation even up to 50 mJ cm⁻² for 20 min, the longitudinal LSPR absorbance of GNRs@Gd₂O₂S remained highly stable with only a slight shift (Fig. 2e). In contrast, an obvious blue-shift could be observed for GNRs (Fig. S5†). The TEM images further confirmed that the morphology of GNRs@Gd₂O₂S remained the same after the irradiation, while the shape of GNRs changed obviously due to laser melting effects (Fig. S6†). These results clearly demonstrated that GNRs@Gd₂O₂S with the protection of the Gd₂O₂S shell have enhanced photothermal stability. The unpredictable aggregation of GNRs in complicated environments, such as serum, within cells, also limits their further biomedical applications due to the resulted loss of desirable NIR absorbance. To investigate the optical stability of GNRs@Gd₂O₂S in complex biological environments, we measured the absorbance spectrum of GNRs@Gd₂O₂S in different aqueous media including 100 mM NaCl, phosphate buffered saline (PBS, pH 7.4), and cell culture media OptiMEM (containing 10% fetal bovine serum). No matter what solution was dispersed in, the LSPR absorbance at 818 nm of GNRs@Gd₂O₂S remained stable with relatively small changes over 2 days (Fig. 2f). In contrast, after only 1 h incubation, the LSPR peaks of GNRs became broadened, red-shifted or even disappeared, and the absorption intensity greatly decreased (Fig. S7†). Besides, no leakage of Gd ions was detected by ICP-MS in the supernatant 7 days after GNRs@Gd₂O₂S was dispersed in these solutions, indicating that GNRs@Gd₂O₂S is also structurally stable. To better demonstrate the protection of the Gd₂O₂S shell on LSPR of GNRs@Gd₂O₂S, we further performed the finite-difference time-domain (FDTD) simulation. Since the GNRs aggregated mainly in a side-by-side (SS) manner, we simulated the coupling of longitudinal LSPR absorption in the SS dimer.⁴⁶ The simulated longitudinal LSPR peak of the GNR@Gd₂O₂S SS dimer was at 804 nm, exhibiting little blue-shift compared to 815 nm of individual GNRs@Gd₂O₂S, while the longitudinal LSPR peak of GNRs could largely shift from 815 nm to 756 nm upon SS dimer formation (Fig. S8†). Additionally, the electric field distribution of the GNR@Gd₂O₂S SS dimer showed only a slight change (Fig. 2g). In contrast, the electric field intensity of the GNR SS dimer significantly reduced or almost disappeared (Fig. S9†). These results confirmed that the Gd₂O₂S shell can produce steric constraints against plasmon coupling, and therefore provides better optical stability of Gd₂O₂S in complex environments. The stabilizing agent CTAB is essential and widely used for the synthesis of high-quality GNRs. However, the non-covalent adsorption and desorption between the CTAB and GNRs is a dynamic process, and the detached CTAB is highly harmful to cells.^19,20 To evaluate the cytotoxicity of GNRs@Gd₂O₂S, we conducted the tetrazolium-based colorimetric assay (MTT assay). After being incubated with HeLa, Hep G2, and L02 cells for 24 h, the GNRs@Gd₂O₂S had no significant cytotoxicity even at the highest concentration, while CTAB capped GNRs showed obvious cytotoxicity that less than 45% of cells were viable even at the concentration as low as 12.5 μg mL⁻¹ (Fig. 2h). These results indicated that the coating of Gd₂O₂S could reduce the toxicity caused by CTAB, rendering GNRs@Gd₂O₂S good biocompatibility.

Fig. 2 (a) Normalized absorption spectra of GNRs@Gd₂O₂S with different thicknesses of the Gd₂O₂S shell (d_Gd2O2S). (b) Temperature elevation curves of GNR@Gd₂O₂S aqueous solutions of different Au ion concentrations as a function of irradiation time under the 808 nm laser irradiation (1 W cm⁻²). (c) Infrared (IR) thermograph of GNR@Gd₂O₂S aqueous solutions (100 μg mL⁻¹) under the 808 nm laser irradiation with different power densities. (d) Temperature elevation curves over four cycles of GNRs and GNR@Gd₂O₂S aqueous solutions (25 μg mL⁻¹), respectively. (e) Normalized absorption spectra of GNRs@Gd₂O₂S under nanosecond laser (7 ns pulse duration, 10 Hz) irradiation with different laser fluences. (f) Absorption spectra of GNRs@Gd₂O₂S dispersed in water, 100 mM NaCl, PBS, and OptiMEM, respectively. (g) FDTD simulated localized electric field distributions of GNRs@Gd₂O₂S and GNR@Gd₂O₂S SS dimer, respectively. (h) Cell viability of HeLa, Hep G2 and L02 cells after incubated with GNRs and GNRs@Gd₂O₂S with different Au ion concentrations at 37 °C for 24 h, respectively.

PAI has emerged as a versatile imaging modality for various kinds of biomedical applications.⁴⁷ Gold nanomaterials have been widely used as contrast agents for PAI because of their tunable LSPR peak positions and large absorption cross sections.¹² We then measured the photoacoustic signals from GNR@Gd₂O₂S aqueous solutions with different concentrations. The photoacoustic signal increased with the increase of the concentration, demonstrating that GNRs@Gd₂O₂S could effectively convert the energy harvested from light into heat and thus serve as a novel contrast agent for PAI (Fig. 3a and b). Gadolinium-based complexes or nanostructures have been widely exploited as T₁ MRI contrast agents owing to the high magnetic moment and long electron spin relaxation time of Gd³⁺ ions.^48,49 To investigate the MRI performance of the GNRs@Gd₂O₂S, we conducted phantom studies and relaxivity measurements on a 0.5 T MRI system. T₁-Weighted phantom images showed that GNRs@Gd₂O₂S exhibit a stronger positive contrast effect (brighten signal) than Gd-DTPA for a given Gd concentration, suggesting the excellent T₁ contrast ability of GNRs@Gd₂O₂S (Fig. 3c). We then measured the longitudinal relaxivity (r₁) value according to the linear relationship of longitudinal relaxation rate (1/T₁) versus Gd ion concentrations (Fig. 3d). The r₁ value of GNRs@Gd₂O₂S was determined as 9.13 ± 0.3 mM⁻¹ s⁻¹, which was more than twice as high as that of commercial T₁ contrast agents Gd-DTPA (4.44 ± 0.5 mM⁻¹ s⁻¹), further confirming that GNRs@Gd₂O₂S could afford more effective chemical exchange relaxations to the water protons around the nanoparticles and therefore resulted in the greater T₁ relaxivity. Inspired by the excellent PAI and MRI contrast performance of GNRs@Gd₂O₂S, we further investigated the potential of GNRs@Gd₂O₂S for in vivo PAI/MRI dual-modal tumor imaging using Hep G2 tumor bearing BALB/c nude mice as models. We used intratumoral injection for animal studies because it enables dispensing extremely high doses of drug throughout the tumor with minimal systemic toxicity. After injection of GNRs@Gd₂O₂S (100 μL, 50 μg mL⁻¹), PAI images of the tumor site were collected. The tumor region presented a significant increase in the photoacoustic signal after the injection, confirming the feasibility of GNRs@Gd₂O₂S for PAI contrast enhancement (Fig. 3e and f). We then examined the in vivo MRI contrast effects of GNRs@Gd₂O₂S on a 7 T MRI scanner. After the injection, a brighter signal in the tumor region could be observed on T₁ weighted MRI images, suggesting that GNRs@Gd₂O₂S can produce strong T₁ contrast enhancements in the tumor (Fig. 3g and h). The presented results therefore demonstrated that GNRs@Gd₂O₂S hold great promise in serving as a dual-modal contrast agent combining PAI and MRI for cancer diagnosis.

Fig. 3 (a) PAI images and (b) PAI intensity of GNR@Gd₂O₂S aqueous solutions with different Au concentrations. (c) T₁-Weighted phantom images of GNR@Gd₂O₂S and Gd-DTPA aqueous solutions with different Gd concentrations at 0.5 T, respectively. (d) The linear fitting of longitudinal relaxation rates at 0.5 T versus Gd concentrations for GNRs@Gd₂O₂S and Gd-DTPA, respectively. The r₁ values were determined by three independent experiments. (e, f) In vivo PAI and corresponding quantification of intensity, and (g, h) T₁-weighted MRI images and corresponding quantification of the signal-to-noise ratio (SNR) of mice tumor sites before and after the injection of GNRs@Gd₂O₂S (100 μL, 50 μg mL⁻¹).

PTT is emerging as one of the most powerful techniques for cancer therapy, because of its high selectivity and minimal invasiveness.^50–53 Nanomaterials with NIR absorption have been widely developed as PTT agents for cancer treatment.¹⁸ We have showed that GNRs@Gd₂O₂S exhibit strong NIR absorption and high photothermal conversion efficiency under 808 nm laser irradiation, indicating that GNRs@Gd₂O₂S are highly appropriate for use as PTT agents. We first investigated the in vitro PTT performance of GNRs@Gd₂O₂S. Hep G2 cells were incubated with GNRs@Gd₂O₂S of different concentrations for 12 h, and then treated with 808 nm laser irradiation (1 W cm⁻²) for 5 min. The MTT assay showed that the cell survival rate decreases sharply with the increase of GNR@Gd₂O₂S concentration. When the concentration of GNRs@Gd₂O₂S reached 50 μg mL⁻¹, the cell survival rate was only less than 20% (Fig. S10†). The merged fluorescence images also confirmed that almost all the cells were dead after they were incubated with GNRs@Gd₂O₂S of 50 μg mL⁻¹ for 4 h and treated with 1 W cm⁻² laser irradiation for 5 min (Fig. S11†). Motivated by the in vitro results that GNRs@Gd₂O₂S can efficiently kill the cancer cells under NIR laser irradiation, we further investigated the in vivo PTT performance of GNRs@Gd₂O₂S. We first examined the in vivo photothermal conversion effect of GNRs@Gd₂O₂S by using an infrared thermograph to record the temperature changes in the tumor site of the mice. The IR thermographic images showed that the temperature at the tumor area could rapidly reach ∼52.1 °C in the presence of GNRs@Gd₂O₂S (100 μL of 50 μg mL⁻¹) under 808 nm laser irradiation (1 W cm⁻²) within 8 min (Fig. 4a). In comparison with the injection of PBS, the maximum temperature at the tumor area reached only about 34.9 °C under the same irradiation conditions, confirming that GNRs@Gd₂O₂S have an excellent photothermal conversion capability even in solid tumors. The tumor bearing mice were then randomly divided into four groups, mice were injected with GNRs@Gd₂O₂S (100 μL, 50 μg mL⁻¹) and then irradiated with a 808 nm laser (1 W cm⁻²) for 8 min (GNRs@Gd₂O₂S + laser group), mice without any treatment (blank group), mice were only injected with GNRs@Gd₂O₂S (GNRs@Gd₂O₂S only group), and mice injected with PBS and then irradiated with a laser (PBS + laser group). The co-treatment with both GNRs@Gd₂O₂S and 808 nm laser irradiation could induce serious empyrosis at the tumor sites of mice, leading to the complete elimination of tumor in several days. In marked contrast, the mice in the other control groups exhibited rapid tumor growth within 16 days (Fig. 4b and c). The body weights of the mice for all groups were measured during the treatments, and no significant weight drop was observed, indicating the low toxicity of all treatments (Fig. S12†). The hematoxylin and eosin (H&E) staining of tumor sections revealed the pycnosis and karyolysis of cancer cells in the GNRs@Gd₂O₂S + laser group, confirming that the tumor was severely burned after PTT treatment (Fig. 4d). Meanwhile, no noticeable organ damage or inflammatory lesion was observed in major organs, such as heart, liver, spleen, lung, kidney, of mice 16 days after the therapy, demonstrating the safety of PTT treatments (Fig. S13†).

Fig. 4 (a) IR thermograph of mice under laser irradiation (808 nm, 1 W cm⁻²) at different time points with (upper row) and without (lower row) the injection of GNRs@Gd₂O₂S. (b) Representative photographs of Hep G2 tumor-bearing BALB/c mice with (upper row) and without (lower row) PTT treatments. Photos were collected at day 0 and day 16, respectively. (c) Tumor growth curves of the Hep G2 tumor-bearing mice after different treatments. (d) Representative H&E stained histological images of tumor sections after different treatments. All images share the same scale bar.

Conclusions

In summary, we have successfully synthesized GNRs@Gd₂O₂S and investigated their potential to serve as a multifunctional cancer theranostic nanoplatform. Owing to the protection of the Gd₂O₂S shell, the GNRs@Gd₂O₂S exhibit high stability either under high-fluence nanosecond laser irradiation or in complex biological environments. Additionally, the coating of Gd₂O₂S also reduces the toxicity caused by CTAB, rendering GNRs@Gd₂O₂S good biocompatibility. Both in vitro and in vivo studies demonstrate that the GNRs@Gd₂O₂S have excellent PAI and MRI contrast enhancement, showing great capability for the use of dual-modal contrast agents. Furthermore, the GNRs@Gd₂O₂S with strong NIR absorption and high photothermal conversion efficiency can effectively inhibit tumor growth through the photothermal effect. We thus believe that the stable and powerful GNR@Gd₂O₂S nanoplatform holds a great promise for biomedical applications, particularly cancer imaging and therapy.

Acknowledgements

This research was supported by the National Natural Science Foundation of China (no. 21475026, U1505221, 21505021, 21622502, 21635002, 81501461 and 11547172), Natural Science Foundation of Fujian Province of China (no. 2015H6011, 2016J05035), the Program for Changjiang Scholars and Innovative Research Team in the University (no. IRT15R11), and the Independent Research Project of State Key Laboratory of Photocatalysis on Energy and Environment (no. 2014B02).

References

1. K. Riehemann, S. W. Schneider, T. A. Luger, B. Godin, M. Ferrari and H. Fuchs, Angew. Chem., Int. Ed., 2009, 48, 872–897.
2. T. Lammers, S. Aime, W. E. Hennink, G. Storm and F. Kiessling, Acc. Chem. Res., 2011, 44, 1029–1038.
3. Y. Gao, J. Lim, S.-H. Teoh and C. Xu, Chem. Soc. Rev., 2015, 44, 6306–6329.
4. J. Xie, S. Lee and X. Chen, Adv. Drug Delivery Rev., 2010, 62, 1064–1079.
5. E. K. Lim, T. Kim, S. Paik, S. Haam, Y. M. Huh and K. Lee, Chem. Rev., 2015, 115, 327–394.
6. R. Wang, H. Hu, Q. Cai, N. Zhao, Y. Zhu and F. Xu, Sci. China: Chem., 2015, 58, 1461–1470.
7. L. S. Lin, Z. X. Cong, J. B. Cao, K. M. Ke, Q. L. Peng, J. Gao, H. H. Yang, G. Liu and X. Chen, ACS Nano, 2014, 8, 3876–3883.
8. L. S. Lin, Z. X. Cong, J. Li, K. M. Ke, S. S. Guo, H. H. Yang and G. N. Chen, J. Mater. Chem. B, 2014, 2, 1031–1037.
9. X. R. Song, X. Wang, S. X. Yu, J. Cao, S. H. Li, J. Li, G. Liu, H. H. Yang and X. Chen, Adv. Mater., 2015, 27, 3285–3291.
10. G. Huang, X. Zhu, H. Li, L. Wang, X. Chi, J. Chen, X. Wang, Z. Chen and J. Gao, Nanoscale, 2015, 7, 2667–2675.
11. L. Cheng, A. Kamkaew, H. Sun, D. Jiang, H. F. Valdovinos, H. Gong, C. G. England, S. Goel, T. E. Barnhart and W. Cai, ACS Nano, 2016, 10, 7721–7730.
12. X. Yang, M. Yang, B. Pang, M. Vara and Y. Xia, Chem. Rev., 2015, 115, 10410–10488.
13. H. Chen, L. Shao, Q. Li and J. Wang, Chem. Soc. Rev., 2013, 42, 2679–2724.
14. X. Huang, S. Neretina and M. A. El-Sayed, Adv. Mater., 2009, 21, 4880–4910.
15. X. Huang, I. H. El-Sayed, W. Qian and M. A. El-Sayed, J. Am. Chem. Soc., 2006, 128, 2115–2120.
16. S. Manohar, C. Ungureanu and T. G. Van Leeuwen, Contrast Media Mol. Imaging, 2011, 6, 389–400.
17. J. V. Jokerst, A. J. Cole, D. Van de Sompel and S. S. Gambhir, ACS Nano, 2012, 6, 10366–10377.
18. L. Cheng, C. Wang, L. Feng, K. Yang and Z. Liu, Chem. Rev., 2014, 114, 10869–10939.
19. A. M. Alkilany, P. K. Nagaria, C. R. Hexel, T. J. Shaw, C. J. Murphy and M. D. Wyatt, Small, 2009, 5, 701–708.
20. N. Khlebtsov and L. Dykman, Chem. Soc. Rev., 2011, 40, 1647–1671.
21. S. Link, C. Burda, M. B. Mohamed, B. Nikoobakht and M. A. El-Sayed, J. Phys. Chem. A, 1999, 103, 1165–1170.
22. S. Link, C. Burda, B. Nikoobakht and M. A. El-Sayed, J. Phys. Chem. B, 2000, 104, 6152–6163.
23. M. B. Mohamed, K. Z. Ismail, S. Link and M. A. El-Sayed, J. Phys. Chem. B, 1998, 102, 9370–9374.
24. H. C. Huang, S. Barua, D. B. Kay and K. Rege, ACS Nano, 2009, 3, 2941–2952.
25. X. Hu and X. Gao, Phys. Chem. Chem. Phys., 2011, 13, 10028–10035.
26. Z. Zhang, L. Wang, J. Wang, X. Jiang, X. Li, Z. Hu, Y. Ji, X. Wu and C. Chen, Adv. Mater., 2012, 24, 1418–1423.
27. C. Wang, J. Chen, T. Talavage and J. Irudayaraj, Angew. Chem., Int. Ed., 2009, 48, 2759–2763.
28. H. W. Yang, H. L. Liu, M. L. Li, I. W. Hsi, C. T. Fan, C. Y. Huang, Y. J. Lu, M. Y. Hua, H. Y. Chou, J. W. Liaw, C. C. Ma and K. C. Wei, Biomaterials, 2013, 34, 5651–5660.
29. H. Sun, Q. Yuan, B. Zhang, K. Ai, P. Zhang and L. Lu, Nanoscale, 2011, 3, 1990–1996.
30. J. Wang, J. Liu, Y. Liu, L. Wang, M. Cao, Y. Ji, X. Wu, Y. Xu, B. Bai, Q. Miao, C. Chen and Y. Zhao, Adv. Mater., 2016, 28, 8950–8958.
31. C.-S. Chang and L. J. Rothberg, Chem. Mater., 2015, 27, 3211–3215.
32. Z. Zhang, J. Wang, X. Nie, T. Wen, Y. Ji, X. Wu, Y. Zhao and C. Chen, J. Am. Chem. Soc., 2014, 136, 7317–7326.
33. Y. Zhang, J. Qian, D. Wang, Y. Wang and S. He, Angew. Chem., Int. Ed., 2013, 52, 1148–1151.
34. H. Hou, L. Chen, H. He, L. Chen, Z. Zhao and Y. Jin, J. Mater. Chem. B, 2015, 3, 5189–5196.
35. M. Ji, M. Xu, W. Zhang, Z. Yang, L. Huang, J. Liu, Y. Zhang, L. Gu, Y. Yu, W. Hao, P. An, L. Zheng, H. Zhu and J. Zhang, Adv. Mater., 2016, 28, 3094–3101.
36. M. A. Muhammed, M. Doblinger and J. Rodriguez-Fernandez, J. Am. Chem. Soc., 2015, 137, 11666–11677.
37. H. Wang, Z. Sun, Q. Lu, F. Zeng and D. Su, Small, 2012, 8, 1167–1172.
38. J. Zhao, P. Xu, Y. Li, J. Wu, J. Xue, Q. Zhu, X. Lu and W. Ni, Nanoscale, 2016, 8, 5417–5421.
39. B. Nikoobakht and M. A. El-Sayed, Chem. Mater., 2003, 15, 1957–1962.
40. Y. Yang, S. Han, G. Zhou, L. Zhang, X. Li, C. Zou and S. Huang, Nanoscale, 2013, 5, 11808–11819.
41. L. Ma, S. Liang, X.-L. Liu, D.-J. Yang, L. Zhou and Q.-Q. Wang, Adv. Funct. Mater., 2015, 25, 898–904.
42. J. F. Moulder, W. F. Stickle, P. E. Sobol and K. D. Bomben, Handbook of X-ray photoelectron spectroscopy, Perkin-Elmer Corporation, Eden Prairie, 1992.
43. J. Thirumalai, R. Chandramohan, S. Valanarasu, T. A. Vijayan, R. M. Somasundaram, T. Mahalingam and S. R. Srikumar, J. Mater. Sci., 2009, 44, 3889–3899.
44. E. Hemmer, A. Benayas, F. Légaré and F. Vetrone, Nanoscale Horiz., 2016, 1, 168–184.
45. L. C. Chen, C. W. Wei, J. S. Souris, S. H. Cheng, C. T. Chen, C. S. Yang, P. C. Li and L. W. Lo, J. Biomed. Opt., 2010, 15, 016010.
46. P. K. Jain and M. A. El-Sayed, Chem. Phys. Lett., 2010, 487, 153–164.
47. M. Xu and L. V. Wang, Rev. Sci. Instrum., 2006, 77, 041101.
48. E. J. Werner, A. Datta, C. J. Jocher and K. N. Raymond, Angew. Chem., Int. Ed., 2008, 47, 8568–8580.
49. H. Chen, G. D. Wang, W. Tang, T. Todd, Z. Zhen, C. Tsang, K. Hekmatyar, T. Cowger, R. B. Hubbard, W. Zhang, J. Stickney, B. Shen and J. Xie, Adv. Mater., 2014, 26, 6761–6766.
50. K. Dong, Z. Liu, Z. Li, J. Ren and X. Qu, Adv. Mater., 2013, 25, 4452–4458.
51. Z. Zha, X. Yue, Q. Ren and Z. Dai, Adv. Mater., 2013, 25, 777–782.
52. X. R. Song, S. X. Yu, G. X. Jin, X. Wang, J. Chen, J. Li, G. Liu and H. H. Yang, Small, 2016, 12, 1506–1513.
53. L. S. Lin, X. Yang, G. Niu, J. Song, H. H. Yang and X. Chen, Nanoscale, 2016, 8, 2116–2122.

---

Footnote

† Electronic supplementary information (ESI) available: Details on experimental information and supplementary figures. See DOI: 10.1039/c6nr08281e

---