Triggered Photodynamic Therapy and MR/CT ... - Wiley Online Library

4 downloads 96 Views 1MB Size Report
May 26, 2016 - 915 nm Light-Triggered Photodynamic Therapy and. MR/CT Dual-Modal Imaging of Tumor Based on the. Nonstoichiometric Na. 0.52. YbF.
full papers www.MaterialsViews.com

Photodynamic Therapy

915 nm Light-Triggered Photodynamic Therapy and MR/CT Dual-Modal Imaging of Tumor Based on the Nonstoichiometric Na0.52YbF3.52:Er Upconversion Nanoprobes Yanan Huang, Qingbo Xiao,* Huishan Hu, Kunchi Zhang, Yamin Feng, Fujin Li, Jian Wang, Xianguang Ding, Jiang Jiang, Yanfang Li, Liyi Shi, and Hongzhen Lin*

Lanthanide (Ln3+)-doped upconversion nanoparticles (UCNPs) as a new generation of multimodal bioprobes have attracted great interest for theranostic purpose. Herein, red emitting nonstoichiometric Na0.52YbF3.52:Er UCNPs of high luminescence intensity and color purity are synthesized via a facile solvothermal method. The red UC emission from the present nanophosphors is three times more intense than the wellknown green emission from the ≈30 nm sized hexagonal-phase NaYF4:Yb,Er UCNPs. By utilizing Na0.52YbF3.52:Er@SrF2 UCNPs as multifunctional nanoplatforms, highly efficient in vitro and in vivo 915 nm light-triggered photodynamic therapies are realized for the first time, with dramatically diminished overheating yet similar therapeutic effects in comparison to those triggered by 980 nm light. Moreover, by virtue of the high transverse relaxivity (r2) and the strong X-ray attenuation ability of Yb3+ ions, these UCNPs also demonstrate good performances as contrast agents for high contrast magnetic resonance and X-ray computed tomography dual-modal imaging. Our research shows the great potential of the red emitting Na0.52YbF3.52:Er UCNPs for multimodal imaging-guided photodynamic therapy of tumors.

Y. Huang, Dr. Q. Xiao, Y. Feng, F. Li, J. Wang, Dr. X. Ding, Prof. J. Jiang, Dr. Y. Li, Prof. H. Lin i-Lab, Suzhou Institute of Nano-tech and Nano-bionics (SINANO) Chinese Academy of Sciences Suzhou 215123, China E-mail: [email protected]; [email protected] Y. Huang, H. Hu, Prof. L. Shi College of Sciences Shanghai University Shanghai 200444, China H. Hu, Dr. K. Zhang Suzhou Key Laboratory of Nanobiomedicine Division of Nanobiomedicine Suzhou Institute of Nano-tech and Nano-bionics (SINANO) Chinese Academy of Sciences Suzhou 215123, China DOI: 10.1002/smll.201601023

4200 www.small-journal.com

1. Introduction Lanthanide (Ln3+)-doped upconversion (UC) nanoparticles (NPs) have emerged as a new generation of multimodal bioprobes in view of the excellent optical, magnetic, and X-ray attenuation properties of the Ln3+ elements.[1–7] Due to the unique ability of converting near-infrared (NIR) light to short-wavelength visible light, the UCNPs showing superior features including low toxicity, negligible photobleaching, and remarkable light penetration depth in biological specimens.[8–12] More importantly, the long-lived red emissions (sub-ms range) from the UCNPs as energy donors overlap well with the absorptions of many photosensitizers such as zinc phthalocyanine (ZnPc) and Chlorin e6, which enables high efficient energy transfer (ET) for activating the photosensitizers in photodynamic therapy (PDT) of internal tumors upon NIR light irradiation.[11,13,14] These

© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

small 2016, 12, No. 31, 4200–4210

www.MaterialsViews.com

characteristics render the UCNPs excellent versatile nanoplatforms for cancer theranostics, that is, concurrent noninvasive diagnosis and therapy.[3,8,13,15–18] Previously, various UCNPs or UCNP-based nanocomposites have been prepared for cancer theranostics in which the PDTs were mainly triggered by 980 nm laser light because of the high absorption cross section of the sensitization ions Yb3+ at this wavelength.[13,15] Unfortunately, as water and tissues have strong optical absorption in the NIR range from 930 to 1030 nm, 980 nm irradiation causes serious overheating to the biological specimens, especially for PDT that requires long-duration laser exposure (>20 min). To overcome this obstacle, Nd3+sensitized UCNPs those can be excited by the biocompatible 808 nm light have been applied for PDT.[19,20] However, it is rather complicated and tedious to prepare such UCNPs, which have to exploit a carefully designed core–shell–shell nanostructure to avoid the serious luminescence quenching between Nd3+ and the activators. Besides the Nd3+-sensitized UCNPs, the conventional Yb3+-sensitized UCNPs, when being excited at 915 nm instead of 980 nm, also show much less overheating effect and increased penetration depth in biological specimens.[21] However, 915 nm light-triggered PDT based on Yb3+-sensitized UCNPs has not been reported up to now, let alone their application for cancer theranostics. Although some UC materials have demonstrated promising potentials for high contrast in vitro or in vivo bioimaging under 915 nm irradiation,[21,22] their luminescence is still far from being strong enough for effective PDT. One of the main reasons may be that the optical absorption of Yb3+ at around 915 nm is only about half of at 980 nm in UCNPs,[21] which thus results in much weaker UC luminescence when excited by 915 nm than that excited by 980 nm. To achieve present PDT efficiency based on UCNPs, it is urgent to prepare high efficient 915 nm excited UCNPs with their luminescence intensities comparable to or even higher than those of existing 980 nm excited UC materials. Moreover, the UCNPs are strongly desired to possess additional functions such as magnetic resonance (MR) and X-ray computed tomography (CT) imaging properties for the development of multiplex theranostic nanoplatforms.[15,18] In view that the optical transitions of Ln3+ are sensitive to their local coordination, engineering the local structure of Ln3+ in UCNPs has been developed as a facile strategy for improving their UC luminescence recently.[23–25] Especially, selective increment of the red UC luminescence can be achieved through changing the local structure of Er3+ in nonstoichiometric cubic-phase NaxYF3+x or NaxGdF3+x nanocrystals by tuning the composition of the NPs ([Na]/[Ln] ratio).[25] Compared to the NaYF4 and NaGdF4 hosts which boost mainly green emission, the cubic-phase NaYbF4 is a wellknown host for the red emission of Er3+ to be dominant.[26–28] Therefore, it is anticipated that, by tuning the [Na]/[Yb] ratio in NaxYbF3+x:Er3+ UCNPs, the red UC luminescence can be further enhanced to meet the requirement of efficient 915 nm light-triggered PDT. Moreover, the NaxYbF3+x host could also be employed as high contrast agent for MR and CT imaging, in view of the high transverse relaxivity (r2) and strong X-ray attenuation ability of the high content Yb3+.[28–30] The paramagnetic Yb3+ ions have relatively high magnetic moment small 2016, 12, No. 31, 4200–4210

(µeff = 4.5 µB) among the Ln3+ ions, which would provide high contrast performance in T2-weighted MR imaging. The X-ray absorption coefficient of Yb3+ (3.88 cm2 g−1 at 100 KeV) is much higher than the clinical iobitridol (1.94 cm2 g−1 at 100 KeV),[30] endowing the Yb3+-based UCNPs a promising contrast medium for CT imaging. All these unique properties make the red emitting NaxYbF3+x:Er UCNPs ideal nanoplatform for 915 nm light-triggered PDT along with MR/CT dual-modal bioimaging in cancer theranostics. Although tuning the Na/Ln ratio in NaxYF3+x and NaxGdF3+x hosts has been successful in generating red UC emissions, thermal decomposition of trifluoroacetate salts is the only way for synthesizing the nonstoichiometric NaxLnF3+x UCNPs currently. Herein, we present a facile solvothermal method to prepare nonstoichiometric Na0.52YbF3.52:Er UCNPs, which show highly efficient red luminescence when irradiated by the 915 nm light. Compared to the thermal decomposition method that may produce toxic fluoride-containing gases, the solvothermal method is more green and economic. More importantly, different from using NaF as the sources of both the Na+ and F− elements in the conventional solvothermal method for preparing NaLnF4 UCNPs, we chose NH4F as the source of F− element and the Na+ was obtained from the “leakage” of the Na-containing reaction solvents. Due to the scarce of Na+ source, our method induces a small [Na]/[Yb] ratio and thus significantly high luminescence intensity and color purity for red UC emission of Er3+. After being simply coated with a thin layer of SrF2 and modified with an amphiphilic polymer 1,2-distearoylsnglycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (ammonium salt) (DSPE-PEG), the UCNPs were successfully demonstrated as multiplex theranostic nanoplatforms for PDT of Hela tumor with greatly diminished overheating effect and MR/CT dual-modal imaging.

2. Results and Discussion 2.1. Sample Characterization The nonstoichiometric NaxYbF3+x:Er UCNPs were synthesized via a similar solvothermal method in preparing NaLnF4 UCNPs, except that NH4F was chosen as the F− source rather than the commonly used NaF. In our strategy, the mixture of NaOH, oleic acid and ethanol not only acts as reaction solvent, but also supplies little amounts of Na+ elements for the NaxYbF3+x:Er UCNPs. As shown in Figure S1a (Supporting Information), the X-ray diffraction (XRD) pattern of the samples could be assigned to the referenced cubic structure (JCPDS 01-077-2043) of NaYbF4 with only slight shift of some diffraction peaks. The energy dispersive X-ray spectroscopy (EDS) reveals the presence of the elements Na+, Yb3+ and Er3+ in the NPs (Figure S1b, Supporting Information). The induction-coupled plasma (ICP) analysis of the as-prepared samples yields a [Na]/[Yb]/ [Er] ratio of 0.522:0.982:0.018, corresponding to a formula of Na0.52YbF3.52:Er of the NPs. In contrast, the [Na]/[Ln] ratio (Ln = Yb+Er) gradually increases from 0.56:1 to 0.59:1 when 2 mmol and 4 mmol NaF were added as F− source,

© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.small-journal.com

4201

full papers www.MaterialsViews.com

respectively, implying the formation of Na0.56YbF3.56:Er and Na0.59YbF3.59:Er UCNPs (see also XRD and transmission electron microscopy (TEM) in Figures S2 and S3, Supporting Information). The addition of NaF also provide Na+ sources for NaxYbF3+x:Er, which thus cause the increase of [Na]/[Ln] ratio. A representative main diffraction peak at around 47° shifted towards a smaller angle as the increasing of [Na]/[Ln] ratio, indicative of a decreased lattice parameter and higher local symmetry for Na0.56YbF3.56:Er and Na0.59YbF3.59:Er (Figure S2b, Supporting Information). The morphology and crystalline structure of the NPs were further investigated by TEM and high-resolution TEM (HRTEM) (Figure S1c,d, Supporting Information). The TEM images show that these NPs were roughly irregular hexagonal plates with a mean diameter of 24 nm. The corresponding HRTEM clearly demonstrates lattice fringes with a d-spacing of 0.27 nm, which agrees well with the lattice spacing for the (200) plane of cubic NaYbF4. The selected area electron diffraction pattern further verified the face centered cubic phase structure of the prepared UCNPs, which is consistent with the XRD and HRTEM analysis (Figure S1e, Supporting Information). To further enhance the UC emission, a thin layer of cubic-phase SrF2 shell was coated on the surface of the Na0.52YbF3.52:Er NPs using a modified thermal decomposition method. The XRD and TEM measurements show that there was no obvious change in the crystal structure and morphology of the Na0.52YbF3.52:Er core upon the introduction of the

SrF2 shell (Figure S1a,f, Supporting Information). The mean diameter of the NPs increases from 24 to 27 nm, indicating a thickness of about 1.5 nm after the SrF2 coating (Figure S1g, Supporting Information). Previously, hybrid nanocomposites combining conventional materials (e.g., organic dyes and quantum dots, etc.) have been prepared for potential multifunctional bioprobing applications, but often involve complicated synthesis processes and cause the enlargement of the original sizes of the NPs.[31] To realize efficient assembly of these different function modalities at the nanoscale, it usually requires the development of novel synthesis strategies, such as gas-phase self-assembly method, as the complementary techniques to the wet chemistry methods.[32–34] As will be discussed later, the Na0.52YbF3.52:Er@SrF2 UCNPs exhibit excellent MR/CT imaging and red UC emission properties simultaneously, which can be applied for cancer theranostics without the assistance of other function modalities. The multifunctional Na0.52YbF3.52:Er@SrF2 UCNPs with relatively small size can be prepared using simple wet chemistry methods, which thus may be of helpful for their further applications as bioprobes. The UC spectra of the uncoated and the SrF2-coated Na0.52YbF3.52:Er UCNPs as well as that of the β-NaYF4:Yb,Er counterparts were recorded under 915 nm excitation at the same condition. As can be clearly seen in Figure 1a,d, the Na0.52YbF3.52:Er NPs exhibit dominant red UC emission centered at 655 nm and weak green UC emissions centered at

Figure 1. Comparison of a,c) the UC emission spectra and (inset) digital luminescence photographs, and d) integrated intensity diagram for the cubic Na0.52YbF3.52:Er core (P2), Na0.52YbF3.52:Er@SrF2 core/shell (P3), and β-NaYF4:Yb,Er UCNPs (P1) with the same concentration of the nanoparticles (1 mg mL−1) upon 915 nm excitation; b) variation of the UC emission spectra and (inset) the ln-ln plots of the UC emission intensity versus the excitation power for the Na0.52YbF3.52:Er UCNPs.

4202 www.small-journal.com

© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

small 2016, 12, No. 31, 4200–4210

www.MaterialsViews.com

522/543 nm, which can be ascribed to the intra-4f transitions of 4F9/2→4I15/2 and 2H11/2/4S3/2→4I15/2 of Er3+, respectively. The intensity of the red emission for the Na0.52YbF3.52:Er UCNPs is about 2.1 times higher than the green emission of the 30 nm sized β-NaYF4:Yb,Er (structure and size determined from Figure S4, Supporting Information), reflecting the superiority of the nonstoichiometric systems. Moreover, the intensity ratio between the red and green emissions (R/G ratio) of the Na0.52YbF3.52:Er UCNPs (≈20.5) is much higher than that of the stoichiometric NaYbF4:Er ones (≈3.8),[26] demonstrating the success of the local structure engineering in increasing the color purity of the red emission. The R/G ratio of the Na0.52YbF3.52:Er UCNPs remains nearly the same as varying the pump power from 0 to 3 W cm−2, which excludes the possible influence of the pump power on the color purity (Figure 1b). The pump-power dependence of the UC emission intensity suggests an excitation photon number of 2.1 and 1.7 for the red and green emissions, respectively, indicative of twophoton pumping nature of the UC processes. The UC emission of Na0.52YbF3.52:Er UCNPs was also compared with Na0.56YbF3.56:Er and Na0.59YbF3.59:Er with the same concentration (1 mg mL−1) upon 915 nm excitation (Figure S5, Supporting Information). Due to the lower local symmetry and more Na+/F− vacancies, the Na0.52YbF3.52:Er UCNPs show superior red UC emission property in aspect of both luminescence intensity and color purity, demonstrating the unique role of NH4F as F− source in reducing [Na]/[Ln] ratio and thus improving the red UC emission of NaxLnF3+x:Er UCNPs when being prepared via the solvothermal method. After epitaxial growth of the SrF2 shell, the red emission intensity of Na0.52YbF3.52:Er can be further enhanced by 17 times, which can be attributed to the sufficiently hindered ET from the excited Yb3+ to the surface quenchers (Figure 1c,d). The R/G ratio of the Na0.52YbF3.52:Er@SrF2 core/shell UCNPs decreased to 5.8, which is still much higher than that of the stoichiometric NaYbF4:Er UCNPs. The intensive red emission with high color purity upon excitation at 915 nm makes the Na0.52YbF3.52:Er@SrF2 UCNPs a favorable choice for deep-tissue PDT by avoiding the interference and damages to tissues that often occur at other excitation and emission wavelengths.

2.2. Biocompatible Modification and Drug Loading The red emission of Er3+ in UCNPs overlaps well with the absorption of the photosensitizer ZnPc (650–680 nm), and therefore could be utilized for triggering the ZnPc to generate singlet oxygen (1O2) and kill cancer cells.[35] To facilitate their application for PDT in biological environments, the Na0.52YbF3.52:Er@SrF2 UCNPs were rendered waterdispersible and biocompatible via modification with a commonly used amphiphilic polymer DSPE-PEG.[21,30] The resultant DSPE-PEG modified UCNPs (Lipo-UCNPs) could be well dispersed in aqueous solutions without any apparent morphological change (Figure 2a). Dynamic light scattering (DLS) analysis shows that the Lipo-UCNPs have a narrow size distribution with an average hydrodynamic size of 80.7 nm (inset in Figure 2a). In correspondence to the large small 2016, 12, No. 31, 4200–4210

number of CH2 and C O groups brought by the DSPEPEG molecules, significantly increased IR absorption peaks at 2925, 2855 cm−1 and 1566, 1463, 1382 cm−1 can be clearly seen in the Fourier transform infrared spectroscopy (FTIR) spectra of the Lipo-UCNPs, which demonstrates the success coating of PEG on the NP surfaces (Figure 2b). The LipoUCNPs exhibit low toxicity in HeLa cells and strong red UC luminescence upon excitation at 915 nm (Figure 2c,d). Furthermore, by injecting 50 µL of 2.0 mg mL−1 LipoUCNPs solution into nude mice with an injection depth of about 5 mm, high-contrast red luminescence images could be obtained upon irradiation at 915 nm (inset of Figure 2c). These results clearly reflect the excellent optical and biological properties of the UCNPs after PEG modification. The Lipo-UCNPs are efficient drug carriers in view that the lipophilic molecules such as ZnPc could be tightly adsorbed onto the hydrophobic oleic acid layer near the surface of the UCNPs.[36] The Lipo-UCNPs were loaded with ZnPc (Lipo-UCNPs-ZnPc) via soaking in concentrated solution of ZnPc for 24 h. The saturated loading capacity of ZnPc in the nanocomposites was determined to be 9.6% according to the standard calibration curve of ZnPc (Figure 3a), which is similar to those of in other UCNP-based carriers.[35,37] ZnPc loading has no obvious influence on the both the particle size and stability of the nanocomposites in aqueous solution. The Lipo-UCNPs-ZnPc nanocomposites could be well dispersed with no obvious precipitation observed in aqueous solution in at least one week. The average hydrodynamic size of the Lipo-UCNPs-ZnPc was determined to be 86 nm from DLS analysis, which is similar to that of 80.7 nm for Lipo-UCNPs before ZnPc loading (Figure S6, Supporting Information). To assess the stability of ZnPc in Lipo-UCNPs-ZnPc, the amount of ZnPc leaking from the nanocomposites was measured in Dulbecco’s phosphate-buffered saline (D-PBS) (pH = 7.4) similar to that reported before.[38,39] As can be seen in Figure S7 (Supporting Information), above 95% of ZnPc still remains in Lipo-UCNPs-ZnPc after 16 h of incubation, which indicates that ZnPc could be firmly attached on Lipo-UCNPs through the intermolecular hydrophobic force in our experimental conditions. Our result is consistent with that of ZnPc in amphiphilic chitosan modified UCNPs, which shows only slight leakage at pH = 7.4.[39] Due to the occurrence of efficient ET from the UCNPs to ZnPc, the intensity of the red UC emission was dramatically reduced as increasing the loading amount of ZnPc (Figure 3b). The ET efficiency, defined as (I0 −I1)/I0, where I0 and I1 are the integrated intensities of red UC emission of the NPs before and after ZnPc loading, reaches 92.5% for Lipo-UCNPs-ZnPc with saturated loading capacity. The capability of 1O2 generation by Lipo-UCNPs-ZnPc upon irradiation at 915 nm was evaluated by the bleaching of 1,3-diphenylisobenzofuran (DPBF) in D-PBS. As shown in Figure 3c and Figure S8 (Supporting Information), the absorption of DPBF in solution of Lipo-UCNPs-ZnPc decreased by 76% as the irradiation time increased from 0 to 55 min, indicating the success of ZnPc activation and singlet oxygen generation under the driving of the 915 nm light. The absorption of DPBF showed a quick bleaching within 20 min, which is comparable to the bleaching rates triggered by the conventional 980 nm light

© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.small-journal.com

4203

full papers www.MaterialsViews.com

Figure 2. Characterization of the Lipo-UCNPs: a) TEM image and (inset) corresponding size distributions; b) FTIR spectrum; c) UC emission spectrum (1 mg mL−1) and (inset) in vivo UC emission imaging (50 µL, 2 mg mL−1) of a nude mouse with pump power of 0.5 W cm−2; d) cell viability of Hela cells treated by different concentrations of Lipo-UCNPs. For comparison, FTIR and UC emission spectra of the oleic acid capped UCNPs (OA-UCNPs) were also presented in panels (b) and (c), respectively.

(typically within 15–20 min for complete bleaching).[15,40–42] Notably, we chose a relatively low power density of 0.5 W cm−2 for the 915 nm light throughout the experiment, identical to that commonly adopted for the 980 nm light in PDT. Although the absorption cross-section of the sensitizer ion Yb3+ at 915 nm is much lower than that at 980 nm, the dramatic efficiency enhancement of the red UC luminescence thanks to the nonstoichiometric composition could compensate the absorption loss and provide sufficient energy power for activating the ZnPc molecules without increasing the intensity of the irradiation source. As a result, high efficiency of 1O2 generation was obtained in the LipoUCNPs-ZnPc system under mild 915 nm light. To further assess the efficacy of the PDT triggered by 915 nm light, in vitro therapy was performed by incubating the HeLa cells with Lipo-UCNPs-ZnPc (0.3 mg mL−1) for 24 h followed by irradiation with 915 nm light at 0.5 W cm−2. As can be seen in Figure 3d, the cell viability dramatically decreased after 20 min of irradiation, and ≈51% of the cancer cells were killed by the singlet oxygen generated from Lipo-UCNPsZnPc. The cell viability was reduced by 65% upon further extension of the irradiation time to 30 min. The above results indicate the promising potential of the present UCNP com-

4204 www.small-journal.com

posites in carrying photosensitizer drugs for killing cancer cells in the aid of 915 nm laser light with a rather low power density.

2.3. In Vivo PDT Triggered by 915 nm To evaluate the in vivo therapeutic efficacy of the 915 nmlight-triggered PDT systems, 50 µL of Lipo-UCNPs-ZnPc (10 mg mL−1) were intratumorally injected into HeLa tumorbearing mice. Mice that received intratumoral injection of pure PBS solution were applied as growth control. After 12 h of the postinjection, the mice were irradiated with the 915 nm light laser for 20 min at the tumor site. For a better comparison, the PDT treatment was performed with a 5 min interval for every 2 min of 915 nm irradiation, similar to those usually done for PDTs using 980 nm light.[37,42,43] Then the tumor volumes and the body masses of these mice were measured over 10 d. The tumor volumes in the control group increased remarkably from ≈26.4 to ≈428 mm3 after the 10 d treatment due to the lack of inhibition of the cancer cell growth. In sharp contrast, the tumor volumes in the experiment group increased much more slowly during the same treatment

© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

small 2016, 12, No. 31, 4200–4210

www.MaterialsViews.com

Figure 3. Drug loading and photorelease of Lipo-UCNPs-ZnPc: a) quantification of ZnPc loading into Lipo-UCNPs at different ZnPc concentrations; b) UC emission spectra of the Lipo-UCNPs with different concentrations of ZnPc loading; c) photorelease of singlet oxygen from Lipo-UCNPs-ZnPc (200 µg mL−1) upon excitation by 915 nm light (0.5 W cm−2); d) HeLa cell viability exposed to Lipo-UCNPs-ZnPc (300 µg mL−1) and irradiated with the 915 nm light (0.5 W cm−2).

period, changing from ≈25.9 to ≈137.5 mm3 only, which indicates the successful reduction of the tumor growth rate by the Lipo-UCNPs-ZnPc based PDT driven with 915 nm light (Figure 4a,c). Moreover, obviously increased body weight of the mice in the experiment group was observed as compared to that in the control group (Figure 4b), implying the improved health condition of the mice in the former case. The in vivo therapeutic effects of our 915 nm light-triggered PDT systems are comparable to those conventional ones triggered by 980 nm.[37,42,43] To confirm the validity of our systems in avoiding overheating of tissues in contrary to that often occurs under 980 nm irradiation, it is necessary to evaluate the temperature increase induced by the 915 nm light in PDT applications. The 915 nm laser induced heating effect was characterized using an anesthetized nude mouse with its back exposed to a circular laser beam at a certain incident angle. The spatiotemporal temperature of the mouse surface was recorded and analyzed using an infrared thermal imaging camera that put in front of the mouse. As displayed in Figure 5, upon the mouse being irradiated for 0–5 min by a 915 nm laser, the temperature in the irradiated area experienced very little change when the pump power was kept below 0.58 W cm−2. Further increment of the pump power small 2016, 12, No. 31, 4200–4210

up to 1.94 W cm−2 induced the temperature to increase from about 27 to 38.5 °C after 5 min of irradiation. In contrast, the temperature in the irradiated area of a mouse can increase by about 12 °C after only 1 min of exposure to a 980 nm laser with pump power of 0.5 W cm−2.[21] These results clearly confirm that the 915 nm light-triggered PDT based on the red emitting nonstoichiometric Na0.52YbF3.52:Er@SrF2 UCNPs could effectively overcome the overheating problem while maintain the same high therapeutic efficacy in comparison to the conventional 980 nm light-triggered systems.

2.4. MR and CT Imaging By utilizing the high transverse relaxivity and the high X-ray absorption capability of Yb3+, the Na0.52YbF3.52:Er@SrF2 UCNPs were also applied for MR and CT imaging of tumors. The transverse relaxation time (T2) of the Lipo-UCNPs was measured in aqueous solutions with Yb3+ concentrations of 0–3.0 × 10−3 m (Figure 6a). From the slope of the plot of 1/T2 versus the Yb3+ concentration, the ionic transverse relaxivity (r2) for the Lipo-UCNPs was determined to be 9.77 s−1 mM−1, which are much higher than that of clinical Gd-DTPA

© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.small-journal.com

4205

full papers www.MaterialsViews.com

Figure 4. Variation of a) tumor volumes and b) body weights of mice, and c) representative photos of a nude mouse and the corresponding tumor at 10 d after treatment for the experiment and control groups, respectively.

(r2 value of 5.8 s−1 mM−1).[30] The high magnetization of the Na0.52YbF3.52:Er@SrF2 UCNPs is due to the large effective magnetic moment of Yb3+ (µeff = 4.5 µB). The MR images of the Lipo-UCNPs solutions exhibit significantly enhanced effect

on the T2-weighted sequences with the increment of Yb3+ concentration, which reflects the effectiveness of the UCNPs as T2-weighted MR imaging agents (inset of Figure 6a). Meanwhile, the phantom CT images and Hounsfield unit (HU)

Figure 5. Temperature distributions on the mouse skin that irradiated by 915 nm laser of different power densities for various irradiation durations.

4206 www.small-journal.com

© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

small 2016, 12, No. 31, 4200–4210

www.MaterialsViews.com

Figure 6. a) Relaxation rate (1/T2, r2) and (inset) in vitro T2-weighted MR images, and c) CT values (HU) and (inset) in vitro CT images of Lipo-UCNPs as functions of the Yb3+ concentration; in vitro b) T2-weighted MR image and d) CT image of rats after intratumorally injection of Lipo-UCNPs-ZnPc with Yb concentrations of 2.0 and 5.0 mg mL−1, respectively.

values of the Lipo-UCNPs in aqueous solutions showed remarkable signal enhancements at increased concentrations of Yb3+ (Figure 6c). The line slope of the HU value versus ion concentration of Yb3+ is about 5.38 HU mM−1, which is comparable to the commercial iodine-based X-ray contrast agents (Iopromide, 5.19 HU mM−1).[29] Benefiting from the excellent MR and CT performance of Yb3+, in vivo T2-weighted MR and CT imaging of the HeLa tumor were carried out by intratumorally administration of Lipo-UCNPs-ZnPc. As illustrated in Figure 6b,d, high contrast imaging of the tumors were observed in the T2-weighted MR and the CT investigations after 24 h postinjection. The combination of MR and CT imaging with the 915 nm light-triggered PDT using the same UCNP system could provide multiple complementary imaging data and meet the various requirements in cancer theranostics.

3. Conclusion We have synthesized the intensively red emitting Na0.52YbF3.52:Er UCNPs via a facile solvothermal method for simultaneous 915 nm light-triggered PDT and MR/CT dual-modal imaging. Due to the successful local structure engineering, the luminescence intensity and color purity of the red emission from the Na0.52YbF3.52:Er UCNPs were significantly enhanced upon excitation at 915 nm. Moreover, the intensity of the red emission was further increased by ≈17 times by coating the as-prepared UCNPs with a thin layer of SrF2 while a relatively high R/G ratio of 5.8 was still retained. Upon irradiation by the 915 nm light, the ZnPcloaded Na0.52YbF3.52:Er@SrF2 UCNPs not only showed high small 2016, 12, No. 31, 4200–4210

efficacy in generating singlet oxygen species and kill cancer cells, but also avoided the overheating effect which is often associated with the 980 nm laser irradiation. Meanwhile, the Na0.52YbF3.52:Er@SrF2 UCNPs exhibit high r2 and HU values, which were also employed as MR and CT dual-modal bioimaging agents of tumors. The high 915 nm-light-triggered PDT efficacy against cancer cells and tumors, in combination with the good performance in MR and CT imaging, demonstrates that the red emitting Na0.52YbF3.52:Er@SrF2 UCNPs are highly promising for the development of multiplex theranostic nanoplatforms.

4. Experimental Section Materials: Yb(NO3)3·5H2O (99.99%), Er(NO3)3·5H2O (99.99%) were purchased from Jinan Henghua Sci.& Tec. Co., Ltd. Oleic acid (technical grade, 90%), 1-octadecene (ODE, 90%), ZnPc, DPBF and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) were purchased from Sigma-Aldrich. Ethanol, cyclohexane, and chloroform were purchased from Sinopharm Chemical Reagent Co., Ltd (China). Ammonium fluoride (NH4F) and dimethyl sulfoxide (DMSO) were purchased from Aladin. DSPE-PEG was purchased from Avanti Polar Lipids. All reagents were used as received without further purification. Synthesis of Na0.52YbF3.52:Er UCNPs: The cubic-phase Na0.52YbF3.52:Er UCNPs were synthesized by a modified solvothermal method. Briefly, 0.98 mmol of Yb(NO3)3, 0.02 mmol of Er(NO3)3, and 3 mL of deionized water were added to a mixture of NaOH (600 mg), deionized water (2 mL), oleic acid (20 mL), and ethanol (10 mL) under thorough stirring. Then, 4 mL of deionized

© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.small-journal.com

4207

full papers www.MaterialsViews.com

water containing 4 mmol of NH4F was drop-wisely added to the mixture. After vigorous stirring, the colloidal solution was transferred into a 50 mL Teflon-lined autoclave, sealed and heated at 200 °C for 8 h and cooled naturally to room temperature (RT). The final products were collected by means of centrifugation, washed with ethanol and cyclohexane for three times, and finally dispersed in 10 mL of cyclohexane for further use. For comparison, Na0.56YbF3.56:Er and Na0.59YbF3.59:Er UCNPs were synthesized using the similar procedure as Na0.52YbF3.52:Er, except that the 4 mmol of NH4F was replaced as 2 mmol of NaF + 2 mmol of NH4F, and 4 mmol of NaF, respectively. The hexagonalphase NaYF4:Yb,Er NPs were synthesized via a modified coprecipitation route. Generally, 0.80 mmol of YCl3·6H2O, 0.18 mmol of YbCl3·6H2O, and 0.02 mmol of ErCl3·6H2O were mixed with 6 mL of OA and 15 mL of ODE in a 100 mL three-neck round-bottom flask. The resulting mixture was heated to 150 °C under N2 flow with constant stirring for 60 min to form a homogeneous and transparent solution. After the mixture being cooled down to RT, 10 mL of methanol containing 4 mmol NH4F and 2.5 mmol NaOH was added drop-wisely. The resulting solution was stirred for 30 min and then heated to 70 °C to evaporate the methanol. The mixture was then kept at 120 °C for 20 min and heated to 310 °C and held at this temperature for 60 min. After cooled to RT, the NPs were precipitated, centrifuged and washed with ethanol, and finally dispersed in cyclohexane. Synthesis of Na0.52YbF3.52:Er@SrF2 Core/Shell UCNPs: The coating of SrF2 was carried out by a facile thermal decomposition method. Generally, 0.5 mmol of Sr(CF3COO)2 and 0.5 mmol of Na0.52YbF3.52:Er NPs were added into the solutions of 10 mL OA and 10 mL ODE in a three-necked flask at RT. The resulting solution was stirred at RT, heated to 70 °C, and then to 120 °C to entirely evaporate the methanol and the water, respectively. Thereafter, the mixture was heated at 280 °C for 1 h and then cooled to RT naturally. The resulting Na0.52YbF3.52:Er@SrF2 NPs were collected by means of centrifugation, washed with ethanol, and cyclohexane for three times, and finally dispersed in cyclohexane. Modifying the UCNPs with DSPE-PEG (Lipo-UCNPs): In a typical experiment, a dispersion containing 25 mg of the core/shell UCNPs and 5 mL of chloroform was added into a 50 mL round-bottom flask equipped with a magnetic stir bar. Another 5 mL of chloroform solution containing 50 mg of DSPE-PEG was then added. The mixture was gently stirred for 30 min. A Labconco rotary evaporator with a water bath of 30 °C was used to evaporate the solvent, and the residual was readily dispersed in water under ultrasonication for 5 min. The resulting dispersion was filtered through a 0.22 µm membrane filter to remove large aggregates and then kept at 4 °C for further use. Characterization: XRD patterns of the UCNPs were carried out on a BRUKER D8 Discover power diffractometer with Cu-Ka1 radiation (λ = 0.154 nm). Both TEM and HRTEM measurements were performed using a Tecnai G2 F20 S-Twin field-emission TEM equipped with the EDS. The precise concentrations of Yb3+ were determined by an Inductively Coupled Plasma Optical Emission Spectrometer (PerkinElmer ICP-OES 2100DV). UC luminescence spectra were recorded on a Princeton Instruments (SP-2500i) spectrofluorimeter equipped with a 3W semiconductor laser (915 nm). FTIR spectra were measured on a Perkin–Elmer IR spectrometer using the KBr pellet technique. Absorption spectra were recorded using Perkin–Elmer’s Lamda 25 UV/vis spectrophotom-

4208 www.small-journal.com

eter in the range of 190–1100 nm. DLS experiments were carried out on an ALV-5000 spectrometer-goniometer equipped with an ALV/LSE-5004 light scattering electronic and multiple tau digital correlator and a JDS Uniphase He–Ne laser (632.8 nm) with an output power of 22 mW. In Vivo Animal Imaging: One nude mouse was chosen and placed under anesthesia using chloral hydrate for in vivo studies. For deep tissue imaging, the anesthetic mouse was injected subcutaneously at the thorax with 50 µL of PBS solution (pH 7.4) containing 2 mg mL−1 of Lipo-UCNPs. The depth of injection was estimated from needle penetration and was controlled about 5 mm. UC luminescence was observed in a darkened room by excitation with a 915 nm laser and recorded using a CCD-based digital camera with an 800 nm filter to eliminate NIR scatter. At the end of the experiments, the animals were disposed according to standard protocol approved by Suzhou Institute of Nano-Tech and NanoBionics (SINANO), Chinese Academy of Sciences. Loading of Zinc Phthalocyanine on Lipo-UCNPs (Lipo-UCNPsZnPc): ZnPc was loaded by soaking the Lipo-UCNPs (2 mL) in a 5 mL solution of ZnPc in DMSO for 24 h at RT in the dark. For the saturated ZnPc loading experiment, Lipo-UCNPs (2 mL) was soaked in 5 mL solutions containing different concentrations (50–500 × 10−6 M) of ZnPc. Free ZnPc was removed by centrifugation at 15 000 rpm for 10 min. The Lipo-UCNPs-ZnPc was washed three times with deionized water and resuspended in phosphate buffer solution (PBS, pH = 7.4) by ultrasonication. To evaluate the drug loading capacity, ZnPc was extracted from Lipo-UCNPs by ethanol, and the residual ZnPc content was determined using the calibration curve of ZnPc standard solutions by the UV-vis measurement at 650–680 nm. The measurement was carried out in triplicate for each formulation (n = 3). Drug loading capacity of ZnPc (%) was calculated as following: loading capacity = [amount of ZnPc in the Lipo-UCNPs (g)]/[amount of Lipo-UCNPs-ZnPc (g)] × 100. ZnPc release efficiency was performed as follows: 1 mL of Lipo-UCNPs-ZnPc (5 mg mL−1) was dispersed in 50 mL of PBS containing 2% sodium dodecylsulphate (SDS) at 37 °C with continuous stirring. A 2 mL amount of solution was withdrawn and centrifuged at 13 000 rpm for 10 min at different time points. ZnPc in the supernatant was determined using the standard curve of ZnPc by the UV–vis measurement at 650–680 nm that calibrated in PBS (pH = 7.4) containing 2% SDS. Cytotoxicity Assessment: The cytotoxicity of the Lipo-UCNPs was assessed by cell viability based on the MTT assay. The assay was performed in triplicate in the same manner. Briefly, HeLa cells were seeded into 96-well plates at a density of 1 × 104 cells per wells in 150 µL of media. After overnight growth, the cells were incubated with various concentrations of Lipo-UCNPs (0, 20, 50, 100, 200, 300, 400, 500, 600 µg mL−1) for 24 h (37 °C, 5% CO2). Then 100 µL of MTT solution (1 mg mL−1) was added to each well and the cells were further incubated for 3 h at 37 °C. After the MTT solution was removed, 150 µL of DMSO was added to each well and the plate was gently shaken for 10 min to dissolve the precipitated violet crystals. The optical density (OD) was measured at 490 nm using a microplate reader (Perkin Elmer, Victor X4). Cell viability was evaluated as a percentage compared to control cells. The measured OD values of the blank, control, and experimental groups were coded as ODbla, ODcon, and ODexp, respectively. Cellular survival rates were calculated by using Equation (1)

© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

small 2016, 12, No. 31, 4200–4210

www.MaterialsViews.com ⎛ (ODexp − ODbla )⎞ survival rate (%) = ⎜ × 100% ⎝ (OD con − ODbla )⎟⎠

(1)

Generation of Singlet Oxygen upon 915 nm Irradiation: Singlet oxygen generation from ZnPc loaded Lipo-UCNPs was determined by the DPBF bleaching method. An ethanol solution of DPBF (200 µL, 1 × 10−3 M) was added to ZnPc (500 × 10−6 M) loaded Lipo-UCNP (2 mL, 200 µg mL−1) aqueous solution. The samples were continuously irradiated by the 915 nm light (0.5 W cm−2) for 60 min under proper magnetic stirring, and the absorbance of DPBF around 350–500 nm was collected. For comparison, the absorbance of DPBF in solution of PEG-UCNPs with no ZnPc loaded was measured upon 915 nm excitation. In Vitro PDT: The cellar photodynamic effects were carried out using the standard MTT assay. Based on the cell viability evaluation, the Lipo-UCNPs-ZnPc with the concentration of 300 µg mL−1 was selected to assess the photodynamic effects of cancer cells. Briefly, the seeded-cell culture medium was replaced with the serum-free medium containing Lipo-UCNPs-ZnPc. After incubation for 24 h, the cells were carefully rinsed twice with PBS in the dark. The cells were irradiated with a 915 nm laser with an output power density of 0.5 W cm−2 for different durations. After irradiation, the cells were allowed to incubate for an additional 24 h. The in vitro photodynamic effects were assessed by cell viability according to the standard MTT assay method as mentioned above. In Vivo PDT: For in vivo PDT measurements, the Hela cells (2 × 106) subcutaneously injected into the ≈4 week old nude mice (n = 8). As the tumors grew up to a volume of about 26 mm3, the nude mice were randomly divided into groups A and B with four mice per group. For Group A, the tumor-bearing nude mice were treated with 50 µL of Lipo-UCNPs-ZnPc (10 mg mL−1 in PBS) via intratumoral injection. For comparison, the Group B was treated with 50 µL of PBS via intratumoral injection. After 12 h of postinjection, the mice were irradiated by the 915 nm laser at the tumor site for 20 min (power density of 0.5 W cm−2). For better comparison with that of the 980 nm laser, the laser treatment was performed with 5 min interval for every 2 min of light exposure. The therapeutic efficacy of the Lipo-UCNPs-ZnPc on the mice was assessed for every two days by measuring the tumor volume that is calculated by length × width × width/2. Laser Heating Effect: Infrared thermal imaging was performed with an IR thermal camera (FLIR, USA). Generally, a nude mouse was chosen and anesthetized by injecting chloral hydrate, then put on the table with its back exposed to a circular laser beam (915 nm) at a certain incident angle. The laser beams were expanded to a circular beam with a diameter of about 10 mm and the output power density was turned to work at various levels (0.35, 0.58, 0.82, 1.04, 1.45, 1.72, and 1.94 W cm−2) under a stable RT of 20 °C. The infrared thermal imaging camera was put in front of the mouse to record and analyze the spatiotemporal temperature of the mouse surface at 0, 1, 3, and 5 min. In Vitro and In Vivo T2-Weighted MR Imaging: The T2-weighted MR images were measured using a Bruker 11.7 T micro-MRI system. The measurement was performed under the following conditions: repetition time (TR) = 3000 ms, echo time (TE) = 24 ms, and the number of excitations (NEX) = 4. The acquired images had a matrix size of 156 × 156 cm, a field of view of 2.5 × 2.7 cm, and a slice thickness of 0.5 mm. The r2 relaxivity values were determined through the curve fitting of 1/T2 relaxation time (s−1) versus the small 2016, 12, No. 31, 4200–4210

Yb concentration (mM). For in vivo T2-weighted MR imaging, LipoUCNPs-ZnPc (50 µL, Yb concentration of 2.0 mg mL−1) were intratumorally injected into a tumor-bearing mouse in situ, and the same dose physiological saline was intratumorally injected into the tumor of another mouse as control. The two mice were scanned after post-injection of 24 h. In Vitro and In Vivo X-Ray CT Imaging: The CT imaging experiments were performed on a Philips 256-slice CT scanner (Philips Medical System) at 120 kVp voltages. Imaging parameters were given as follows: thickness, 0.9 mm; pitch, 0.99; 120 kVp, 300 mA; field of view, 350 mm; gantry rotation time, 0.5 s; table speed, 158.9 mm s−1. For in vitro CT imaging, the Lipo-UCNPs-ZnPc were dispersed in water with various concentrations (10 × 10−3, 25 × 10−3, 50 × 10−3, and 100 × 10−3 mM) and then placed in a series of 1.5 mL tubes for CT imaging. For in vivo CT imaging, LipoUCNPs-ZnPc (50 µL, Yb concentration of 5.0 mg mL−1) were intratumorally injected into the tumor-bearing mouse in situ, and the same dose physiological saline was also intratumorally injected into the tumor of another mouse as control. The two mice were scanned after post-injection of 24 h.

Supporting Information Supporting Information is available from the Wiley Online Library or from the author.

Acknowledgements This work is supported by the National Natural Science Foundation of China (Nos. 21403288, 21173253, 21273271, and 21403289) and the National Natural Science Foundation of Jiangsu Province (No. BK20140382).

[1] G. Y. Chen, J. Damasco, H. L. Qiu, W. Shao, T. Y. Ohulchanskyy, R. R. Valiev, X. Wu, G. Han, Y. Wang, C. H. Yang, H. Agren, P. N. Prasad, Nano Lett. 2015, 15, 7400. [2] B. Zhou, B. Y. Shi, D. Y. Jin, X. G. Liu, Nat. Nanotechnol. 2015, 10, 924. [3] D. M. Yang, C. X. Li, J. Lin, Nanomedicine 2015, 10, 2573. [4] M. Bettinelli, L. Carlos, X. Liu, Phys. Today 2015, 68, 38. [5] S. Watanabe, T. Asanuma, T. Sasahara, H. Hyodo, M. Matsumoto, K. Soga, Adv. Funct. Mater. 2015, 25, 4390. [6] M. Bettinelli, Nat. Nanotechnol. 2015, 10, 203. [7] M. K. Gnanasammandhan, N. M. Idris, A. Bansal, K. Huang, Y. Zhang, Nat. Protoc. 2016, 11, 688. [8] Z. Chen, W. Zheng, P. Huang, D. T. Tu, S. Y. Zhou, M. D. Huang, X. Y. Chen, Nanoscale 2015, 7, 4274. [9] R. Arppe, I. Hyppanen, N. Perala, R. Peltomaa, M. Kaiser, C. Wurth, S. Christ, U. Resch-Genger, M. Schaferling, T. Soukka, Nanoscale 2015, 7, 11746. [10] P. Alonso-Cristobal, P. Vilea, A. El-Sagheer, E. Lopez-Cabarcos, T. Brown, O. L. Muskens, J. Rubio-Retama, A. G. Kanaras, ACS Appl. Mater. Interfaces 2015, 7, 12422. [11] N. Mauser, D. Piatkowski, T. Mancabelli, M. Nyk, S. Mackowski, A. Hartschuh, ACS Nano 2015, 9, 3617. [12] K. T. Lee, J. H. Park, S. J. Kwon, H. K. Kwon, J. Kyhm, K. W. Kwak, H. S. Jang, S. Y. Kim, J. S. Han, S. H. Lee, D. H. Shin, H. Ko,

© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.small-journal.com

4209

full papers www.MaterialsViews.com

[13] [14] [15] [16] [17]

[18] [19] [20] [21] [22] [23] [24] [25] [26]

[27]

I. K. Han, B. K. Ju, S. H. Kwon, D. H. Ko, Nano Lett. 2015, 15, 2491. G. Tian, X. Zhang, Z. J. Gu, Y. L. Zhao, Adv. Mater. 2015, 27,7692. N. M. Idris, M. K. G. Jayakumar, A. Bansal, Y. Zhang, Chem. Soc. Rev. 2015, 44, 1449. S. Lu, D. T. Tu, P. Hu, J. Xu, R. F. Li, M. Wang, Z. Chen, M. D. Huang, X. Y. Chen, Angew. Chem., Int. Ed. 2015, 54, 7915. F. He, G. X. Yang, P. P. Yang, Y. X. Yu, R. C. Lv, C. X. Li, Y. L. Dai, S. L. Gai, J. Lin, Adv. Funct. Mater. 2015, 25, 3966. X. Z. Ai, C. J. H. Ho, J. Aw, A. B. E. Attia, J. Mu, Y. Wang, X. Y. Wang, Y. Wang, X. G. Liu, H. B. Chen, M. Y. Gao, X. Y. Chen, E. K. L. Yeow, G. Liu, M. Olivo, B. G. Xing, Nat. Commun. 2016, 7, 10432. W. Zheng, P. Huang, D. T. Tu, E. Ma, H. M. Zhu, X. Y. Chen, Chem. Soc. Rev. 2015, 44, 1379. F. J. Ai, Q. Ju, X. M. Zhang, X. Chen, F. Wang, G. Y. Zhu, Sci. Rep. 2015, 5, 10785. D. Wang, B. Xue, X. G. Kong, L. P. Tu, X. M. Liu, Y. L. Zhang, Y. L. Chang, Y. S. Luo, H. Y. Zhao, H. Zhang, Nanoscale 2015, 7, 190. Q. Q. Zhan, J. Qian, H. J. Liang, G. Somesfalean, D. Wang, S. L. He, Z. G. Zhang, S. Andersson-Engels, ACS Nano 2011, 5, 3744. A. Xia, Y. Y. Deng, H. Shi, J. Hu, J. Zhang, S. S. Wu, Q. Chen, X. H. Huang, J. Shen, ACS Appl. Mater. Interfaces 2014, 6, 18329. C. Z. Zhao, X. G. Kong, X. M. Liu, L. P. Tu, F. Wu, Y. L. Zhang, K. Liu, Q. H. Zeng, H. Zhang, Nanoscale 2013, 5, 8084. A. Xia, X. F. Zhang, J. Zhang, Y. Y. Deng, Q. Chen, S. S. Wu, X. H. Huang, J. Shen, Biomaterials 2014, 35, 9167. H. Dong, L. D. Sun, Y. F. Wang, J. Ke, R. Si, J. W. Xiao, G. M. Lyu, S. Shi, C. H. Yan, J. Am. Chem. Soc. 2015, 137, 6569. A. Punjabi, X. Wu, A. Tokatli-Apollon, M. El-Rifai, H. Lee, Y. W. Zhang, C. Wang, Z. Liu, E. M. Chan, C. Y. Duan, G. Han, ACS Nano 2014, 8, 10621. G. Tian, X. P. Zheng, X. Zhang, W. Y. Yin, J. Yu, D. L. Wang, Z. P. Zhang, X. L. Yang, Z. J. Gu, Y. L. Zhao, Biomaterials 2015, 40, 107.

4210 www.small-journal.com

[28] Z. G. Yi, S. J. Zeng, W. Lu, H. B. Wang, L. Rao, H. R. Liu, J. H. Hao, ACS Appl. Mater. Interfaces 2014, 6, 3839. [29] L. J. Zhou, X. P. Zheng, Z. J. Gu, W. Y. Yin, X. Zhang, L. F. Ruan, Y. B. Yang, Z. B. Hu, Y. L. Zhao, Biomaterials 2014, 35, 7666. [30] D. L. Ni, W. B. Bu, S. J. Zhang, X. P. Zheng, M. Li, H. Y. Xing, Q. F. Xiao, Y. Y. Liu, Y. Q. Hua, L. P. Zhou, W. J. Peng, K. L. Zhao, J. L. Shi, Adv. Funct. Mater. 2014, 24, 6613. [31] Q. Ju, D. T. Tu, Y. S. Liu, R. F. Li, H. M. Zhu, J. C. Chen, Z. Chen, M. D. Huang, X. Y. Chen, J. Am. Chem. Soc. 2012, 134, 1323. [32] J. H. Byeon, Y. W. Kim, J. Mater. Chem. B 2014, 2, 3185. [33] J. H. Byeon, Y. W. Kim, Small 2014, 10, 2331. [34] J. H. Byeon, Y. W. Kim, ACS Appl. Mater. Interfaces 2014, 6, 763. [35] Q. B. Xiao, Y. T. Ji, Z. H. Xiao, Y. Zhang, H. Z. Lin, Q. B. Wang, Chem. Commun. 2013, 49, 1527. [36] G. Tian, Z. J. Gu, L. J. Zhou, W. Y. Yin, X. X. Liu, L. Yan, S. Jin, W. L. Ren, G. M. Xing, S. J. Li, Y. L. Zhao, Adv. Mater. 2012, 24, 1226. [37] S. S. Cui, H. Y. Chen, H. Y. Zhu, J. M. Tian, X. M. Chi, Z. Y. Qian, S. Achilefu, Y. Q. Gu, J. Mater. Chem. 2012, 22, 4861. [38] E. Ricci-Junior, J. M. Marchetti, Int. J. Pharm. 2006, 310, 187. [39] S. S. Cui, D. Y. Yin, Y. Q. Chen, Y. F. Di, H. Y. Chen, Y. X. Ma, S. Achilefu, Y. Q. Gu, ACS Nano 2013, 7, 676. [40] X. L. Chen, Z. X. Zhao, M. Y. Jiang, D. P. Que, S. G. Shi, N. F. Zheng, New J. Chem. 2013, 37, 1782. [41] H. J. Wang, R. Shrestha, Y. Zhang, Part. Part. Syst. Charact. 2014, 31, 228. [42] M. Wang, Z. Chen, W. Zheng, H. M. Zhu, S. Lu, E. Ma, D. T. Tu, S. Y. Zhou, M. D. Huang, X. Y. Chen, Nanoscale 2014, 6, 8274. [43] C. Wang, H. Q. Tao, L. Cheng, Z. Liu, Biomaterials 2011, 32, 6145.

© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Received: March 25, 2016 Revised: May 26, 2016 Published online: June 23, 2016

small 2016, 12, No. 31, 4200–4210