Biomaterials 32 (2011) 1185e1192

Contents lists available at ScienceDirect

Biomaterials
journal homepage: www.elsevier.com/locate/biomaterials

Fluorescence-enhanced gadolinium-doped zinc oxide quantum dots for magnetic resonance and uorescence imaging
Yanlan Liu a, b, Kelong Ai a, Qinghai Yuan c, Lehui Lu a, *
a

State Key Laboratory of Electroanalytical Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun, Jilin 130022, PR China Chinese Academy of Sciences, Beijing 100039, PR China c Department of Radiology, The Second Hospital of Jilin University Norman Bethune, Changchun 130022, PR China
b

a r t i c l e i n f o
Article history: Received 2 August 2010 Accepted 12 October 2010 Available online 4 November 2010 Keywords: Nanocomposite Enhanced uorescence Cytotoxicity In vitro test MRI

a b s t r a c t
We report here the development of Gd-doped ZnO quantum dots (QDs) as dual modal uorescence and magnetic resonance imaging nanoprobes. They are fabricated in a simple, versatile and environmentally friendly method, not only decreasing the difculty and complexity, but also avoiding the increase of particles size brought about by silica coating procedure in the synthesis of nanoprobes reported previously. These nanoprobes, with exceptionally small size and enhanced uorescence resulting from the Gd doping, can label successfully the HeLa cells in short time and present no evidence of toxicity or adverse affect on cell growth even at the concentration up to 1 mM. These results show that such nanoprobes have low toxicity, especially in comparison with the traditional PEGylated CdSe/ZnS or CdSe/ CdS QDs. In MRI studies, they exert strong positive contrast effect with a large longitudinal relaxivity (r1) of water proton of 16 mM1 s1. Their capability of imaging HeLa cells with MRI implies that they have great potential as MRI contrast agents. Combining the high sensitivity of uorescence imaging with high spatial resolution of MRI, We expect that the as-prepared Gd-doped ZnO QDs can provide a better reliability of the collected data and nd promising applications in biological, medical and other elds. 2010 Elsevier Ltd. All rights reserved.

1. Introduction Magnetic resonance imaging (MRI) is regarded as one of most powerful techniques in modern diagnostic medicine, because it can penetrate deep into tissue, providing anatomical details and highquality three-dimensional images of soft tissue in a non-invasive monitoring manner [1,2]. Unfortunately, this technique has just been able to resolve objects larger than a few micrometers in size, thus exhibiting much lower sensitivity than radioactive or optical methods [3e5]. Fluorescence imaging (FI), in contrast, has much higher sensitivity and the potential for real-time imaging, but with limited depth perception, which restricts their application just to surface or near-surface phenomena [6,7]. These high complementary characteristics between them allow their respective limitations to be effective overcome by integrating magnetic resonance and optical imaging functionalities into a single nanostructure [8]. As a consequence, intensive efforts have been drawn into the development of MRI-FI nanoprobes due to their prominent advantages in terms of medical diagnose and molecular imaging [9e18].

* Corresponding author. Tel.: 86 431 85262418; fax: 86 431 85262406. E-mail address: lehuilu@ciac.jl.cn (L. Lu). 0142-9612/$ e see front matter 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2010.10.022

Up to now, developing MRI-FI nanoprobes usually adopted two different strategies. One logical strategy was based on incorporating paramagnetic ions into quantum dots (QDs) such as CdSe [9e11]. These nanoprobes showed good photo-stability, high relaxivity and quantum yield (QY), but their high toxicity and potential pollution hazard as the released Cd and Se from the surface QDs might limit their biological and medical applications. In addition, selenides or suldes were intrinsically less-resistant to oxidation by the O2 in air [19,20]. The other was based on coating a magnetic core and uorophores with shells such as silica or modifying silica-capped uorescent cores with paramagnetic ions [12e18]. Generally, these nanoprobes had lower toxicity compared with those obtained by the above strategy thanks to the encapsulation by silica shell. Meanwhile, they allied high sensitivity of FI to the high spatial resolution of MRI, thus ensuring a better reliability of the collected data. Silica encapsulating involved in the fabrication of these nanoprobes, however, not only made the synthesis more complex and difcult, but also increased the particles size [21]. From the viewpoint of practical applications, larger particles were not very suitable for biological and medical elds, especially for labeling functional subcellular or proteins, as larger particles often affect their biological function and are more likely to be recognized and cleared by the phagocyte [22,23]. For these reasons,

1186

Y. Liu et al. / Biomaterials 32 (2011) 1185e1192

Fig. 1. aec) TEM images of Gd-doped ZnO QDs with x 0, 0.08 and 0.3, the insets show the corresponding electron diffraction patterns. def) High-resolution TEM images of Gd-doped ZnO QDs with x 0, 0.08 and 0.3.

it is still a big challenge to develop reasonable strategies for preparing excellent MRI-FI nanoprobes which should be guided by following criteria: 1) high relaxivity and quantum yield, 2) small size, 3) simple and cost-effective synthesis method, 4) low toxicity, and 5) chemical stability in air. Moreover, the nanoprobes will provide better reliability for clinical diagnosis if they are T1-positive agents, since they can enhance the signal intensity instead of reducing the signal intensity in T1-weighted MRI, avoiding the confusion with the reduced signals from bleeding, calcication or

metal deposits [24,25]. Also, we expect that these nanoprobes can give the emission which is more sensitive for human eye instead of blue emission since most cells and tissues also appear blue under UV light [26,27]. In this study, we presented a simple and versatile method to develop dual modality nanoprobes that satised all of these criteria by doping Gd3 ions in low toxic ZnO QDs. Detail on their synthesis, optical properties, cytotoxicity and applications for in vitro MRI-FI imaging were investigated. Experimental results indicated that

Y. Liu et al. / Biomaterials 32 (2011) 1185e1192 Table 1 Comparison between Gd-doped ZnO QDs with different x values. ZnO QDs with different x values (ICP data) Sizea (nm) 6.0 5.18 5.02 4.39 4.14 3.95 3.76 3.55 Sizeb (nm) 5.7 5.11 4.96 4.36 3.9 3.8 3.56 3.27

1187

Band gapc (eV) 3.42 3.435 3.44 3.47 3.49 3.50 3.52 3.55

Intensity (a.u.)

0 0.02 0.04 0.05 0.08 0.3

0 0.02 0.04 0.05 0.08 0.10 0.12 0.30

20

30

40

50 2 /deg.

60

70

80

a Measured by XRD. The average particle size was evaluated based on DebyeeScherer formula d 0.89l/(bcosq), where d is the average diameter of the particles, l is the X-ray wavelength (CuKa, 0.154 nm). q and b are the Bragg diffraction angle and the peak width in radians at half-height, respectively. b Determined using Meulenkamps empircal formula. c measured by UVeVis spectra

Fig. 2. XRD patterns of Gd-doped ZnO QDs with different x values.

these nanoprobes had great potential applications in biological and medical elds.
2. Experimental section 2.1. Materials Zinc acetate dihydrate was purchased from Beijing Chemical Reagent factory (Beijing, China). Tetramethylammonium hydroxide (TMAH) and Gadolinium acetate hydrate were procured from Alfa Aesar. Oleic acid was obtained from Sigmae Aldrich. Other chemicals were of analytical grade and used without any further purication. Water used throughout all experiments was puried with the Millipore system. 2.2. Apparatus The nanoparticle size was examined using JEOL 2000-FX transmission electron microscopy (TEM). The molar ratios of Gd/Zn in the products were determined by an ELAN 9000/DRC ICP-MS system. UV/Vis spectra were recorded on a VARIAN CARY 50 UV/Vis spectrophotometer. The uorescence spectra were obtained using a PerkineElmer LS 55 luminescence spectrometer. XPS was performed using a VG ESCALAB MKII spectrometer, utilizing the XPSPEAK software (Version 4.1) to deconvolute the narrow-scan XPS spectra of Gd 3d of Gd-doped ZnO QDs and Gd2O3, using adventitious carbon to calibrate the binding energy of C1s (284.5 eV). 2.3. Synthesis of pure and Gd-doped ZnO QDs ZnO QDs were synthesized according to a method described in a previous report [28]. Briey, Zn(OAc)2$2H2O (1.2 mM) was rst dissolved in 20 mL of ethanol under vigorous stirring. 70 mL of Oleic acid was then added and the mixture was heated to reux. TMAH (0.526 mL) in 5 mL of reuxing ethanol was rapidly mixed with this mixture under vigorous stirring. Three minutes later, the reaction was stopped with 50 mL of ethanol and cooled down immediately in an ice bath. After centrifugation, the resulting oleate-stabilized ZnO QDs were washed with ethanol several times, and nally redispersed in 10 mL of toluene. Gd-doped ZnO QDs with different x values were synthesized by adding stoichiometric amount of gadolinium acetate to the zinc acetate precursor solution, keeping the total initial concentration of metal ions constant. 2.4. Surface modication of the ZnO QDs 1 mL of AEAPS in toluene (0.1 M) was added to the solution of ZnO QDs in10 mL of toluene under stirring at room temperature. After 5 min, 1 mL of TMAH in ethanol (0.2 M) was injected. The mixture was reuxed at 85  C for 45 min and then cooled down to room temperature. The functional ZnO QDs with amido on their surface were obtained by centrifugation and washed several times with toluene, acetone and water. Finally, QDs were redispersed in water and stored at 4  C in the dark for further experiments. 2.5. Cell cytotoxicity HeLa cells were incubated in the culture medium at 37  C in an atmosphere of 5% CO2 and 95% air. After 24 h, the culture medium was removed. The cells were

incubated in culture medium containing Gd-doped ZnO QDs (x 0.08) with different concentrations for another 24 h and washed with medium twice. 100 mL of the new culture medium containing 10% MTT reagent was introduced followed by further incubating for 4 h to allow formation of formazan dye. After removing the medium, the purple formazan product was dissolved with DMSO for 15 min. Finally, the optical absorption of formazan at 570 nm was measured by an enzyme-linked immunosorbent assay reader.

2.6. Confocal imaging of labeled cell HeLa cells were incubated with Gd-doped ZnO QDs (x 0.08) at 37  C in an atmosphere of 5% CO2 and 95% air for different time. Then, the cells were isolated by

12000 11000

Intensity (a.u.)

10000 9000 8000 7000 1180 1190 1200 1210 1220 1230 1240 Binding Energy (eV)

12000 11000

Intensity (a.u.)

10000 9000 8000 7000 6000 1180 1190 1200 1210 1220 Binding Energy (eV) 1230 1240

Fig. 3. XPS spectra of Gd 3d core-level for a) the Gd-doped ZnO QDs (x 0.08). b) Gd2O3.

1188

Y. Liu et al. / Biomaterials 32 (2011) 1185e1192

800 700 600 0.00 0.02 0.04 0.05 0.08 0.10 0.12 0.30

750 700 650 Intensity (a.u.) 600 550 500 450 400 350 300

Intensity (a.u.)

500 400 300 200 100 0 400

0.00

0.04

0.08 x

0.12

0.30

450

500

550 600 650 Wavelength (nm)

700

750

800

Fig. 4. a) Fluorescence emission spectra of Gd-doped ZnO QDs with different x values at an excitation wavelength of 340 nm. b) Fluorescence of Gd-doped ZnO QDs with different x values under UV light at 365 nm.

centrifugation, washed several times, and subsequently redispersed in PBS buffer. Images were collected using confocal microscopy excitated at 405 nm. 2.7. T1-weighted imaging For in vitro MRI, Gd-doped ZnO QDs (x 0.08) were dispersed in water with Gd3 concentrations in the range from 0 to 0.05 mM. HeLa cells were incubated with Gd-doped ZnO QDs (x 0.08) for 2 h after centrifugation, the cell pellet were covered with agarose. T1-weighted images of Gd-doped ZnO QDs (x 0.08) with varied Gd3 concentrations and cells treated and untreated with Gd-doped ZnO QDs were collected using 1.5 T human clinical scanner.

3. Results and discussion 3.1. Synthesis and characterization ZnO QDs were produced through hydrolyzing zinc acetate by TMAH in ethanol [28]. Gadolinium doping of ZnO QDs were achieved by replacing partially zinc acetate with gadolinium acetate, and the actual molar ratios of Gd/Zn (designed as x) in the nal

products was ascertained by inductively coupled plasma mass spectrometry (ICP-MS). Typical transmission electron microscopy (TEM) images in Fig. 1a,b illustrated the diameter of ZnO QDs and Gd-doped ZnO QDs (x 0.08) was approximately 6 nm and 4 nm, respectively. Nevertheless, for the sample with a higher Gd concentration (x 0.3), the particles were so small that they could not be seen clearly (Fig. 1c). Meanwhile, selected area electron diffraction analysis presented similar patterns for ZnO QDs and Gddoped ZnO QDs (x 0.08 and 0.3), and all the diffraction rings could be indexed to wurtzite ZnO. High-resolution TEM images shown in Fig. 1d,e revealed lattice fringes with a spacing of about 0.26 nm for both ZnO QDs and Gd-doped ZnO QDs (x 0.08), which were well consistent with the lattice spacing in the (002) planes of ZnO wurtzite phase, suggesting that Gd doping did not induce any signicant lattice distortions of ZnO QDs. However, when x increased to 0.3, the crystallinity of the ZnO QDs became so weak that no clear lattice fringe could be observed (Fig. 1f). X-ray diffraction (XRD) was next carried out to determine the crystal structure of the nal products. As seen in Fig. 2, the pattern of

Y. Liu et al. / Biomaterials 32 (2011) 1185e1192

1189

140 120 100

80 60 40 20 0 -20 0.005mM 0.05mM 0.5mM 1mM 5mM

Whereas, remarkable Gd element peaks could be seen in the EDAX patterns of Gd-doped ZnO QDs, providing an important conrmation for the Gd doping (Supplementary Fig. 1). XPS spectra of Gd-doped ZnO QDs (x 0.08) showed two detectable peaks centered at 1192 eV and 1225 eV, corresponding to Gd 3d5/2 and Gd 3d3/2, respectively, which were in good agreement with those of Gd3 in Gd2O3 (Fig. 3). 3.2. Optical properties Optical properties of Gd-doped ZnO QDs with x values varied from 0 to 0.3 were further explored by measuring their uorescent emission spectra (Fig. 4a). ZnO QDs exhibited a strong yellow emission centered at about 565 nm excited at a wavelength of 340 nm. Interestingly, with increasing concentration of Gd3, there was a signicant enhancement in uorescence intensity, coupled with a blue shift of emission band. When x increased to 0.08, the uorescence intensity approximately achieved the maximum. The corresponding QY value (Rhodamine 6G as reference uorescent dye) enhanced from 13.5% (x 0) to 34% (x 0.08), and the maximum emission blue shifted from 565 nm to 550 nm. Further increasing the concentration of Gd3 led to a notable decrease in uorescence intensity and a further blue shift. Up to now, the visible emission of ZnO QDs was most widely considered to be derived from oxygen vacancies [29e31]. The more the oxygen vacancies were, the stronger the emission intensity of ZnO QDs was. According to previous report, the reduction of particles diameter generally induced the increase in the content of oxygen vacancies [32]. In our present experimental system, Gd doping decreased the size of ZnO QDs, which might increase the content of oxygen vacancies and thus result in an improvement in QY of ZnO QDs. However, excessive Gd3 ions consumed the ZnO nanoparticles, decreasing the uorescence intensity, reversely. Generally, the blue-shifts of uorescence spectra of ZnO QDs was closely related to their band-gap enlargement owing to the quantum size effect [33,34]. To evaluate whether the band gap enlargement of ZnO QDs was present in the case of Gd doping, typical UV/Vis absorption spectra of the Gd-doped ZnO QDs with

Fig. 5. Cell viability of HeLa cells incubated with various concentrations of Gd-doped ZnO QDs (x 0.08) for 24 h.

pure ZnO QDs illustrated that all the reection peaks can be indexed to ZnO structure (JCPDS No. 65-3411 and JCPPDS No. 21-1486). In the case of the Gd-doped ZnO QDs, the patterns revealed the presence of a single hexagonal phase of ZnO (JCPDS No. 65-3411), and the crystallinity becomes weaker as x value increased. Surprisingly, no reection peak of Gd2O3 could be observed in these XRD patterns, which may result from the formation of amorphous species. In addition, the average particles diameter calculated on the basis of DebyeeScherer formula decreased as the increase of x value (Table 1). These results were in accordance with those obtained from TEM images. In order to demonstrate the Gd doping and its oxidation state, energy-dispersive X-ray analysis (EDAX) and X-ray photoelectron spectroscopy (XPS) were next employed. For pure ZnO QDs, the EDAX pattern manifested a strong correlation of these elements with the ZnO QDs, no Gd or other impurity elements were tested.

Fig. 6. a) Confocal laser scanning microscopic images of HeLa cells incubated without Gd-doped ZnO QDs (x 0.08). b, c) HeLa cells incubated with Gd-doped ZnO QDs (x 0.08) for 30 min and 2 h at the same concentration (0.625 mM).

1190

Y. Liu et al. / Biomaterials 32 (2011) 1185e1192

Fig. 7. a) T1-weighted magnetic resonance image for various Gd3 concentrations of Gd-doped ZnO QDs (x 0.08) in water from a 1.5 T clinical MRI system. b) The linear relationship between T1 relaxation rates (1/T1) and Gd3 ion concentrations for Gd-doped ZnO QDs (x 0.08). c) T1-weighted image of blank HeLa cells pellet (left) and HeLa cells incubated with Gd-doped ZnO QDs at 0.01 M Gd3 ions for 2 h.

different x values were given (Supplementary Fig. 2). As ZnO is a direct band gap semiconductor, its band gap originated from the equation [35]:

a A,

hn E1=2 hn

(1)

where a, E and A are absorption coefcient, band gap and constant, respectively. According to equation (1), the optical band gaps (Eg) of ZnO QDs with different x values can be estimated by extrapolating the linear part near the onset in a plot of (ahv)2 versus hv. A clear blue shift in band gap was discernible as x value increased (Supplementary Fig. 2). For instance, the band gaps of ZnO QDs with x 0.08 and 0.3 increased from 3.42 eV (x 0) to 3.49 eV and

Y. Liu et al. / Biomaterials 32 (2011) 1185e1192

1191

3.55 eV, respectively. Consequently, it was reasonable to conclude that the blue shift in emission of ZnO QDs was attributed to the band-gap widening effect of Gd doping. Based on these band gap values, the particles size could be also determined as 5.7 nm (x 0) and 3.9 nm (x 0.08), which were well consistent with the results from TEM images and XRD (See supplementary for the calculation method). 3.3. Cytotoxicity The signicantly enhanced emission of Gd-doped ZnO QDs (x 0.08), observed above, incited us to evaluate their potential application as imaging probes in vivo. The cytotoxicity of Gd-doped ZnO QDs, which impeded their use for in vivo applications, was rst investigated by MTT cell proliferation assay (where MTT is 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, a tetrazole). Modication of Gd-doped ZnO QDs with N-(2-aminoethyl) aminopropyltrimethoxysilane (AEAPS) prior to their use was performed because the presence of amine groups at the surface of QDs favored their dispersion in aqueous solutions and provided a functionalized surface, simultaneously. The cell viability results were depicted in Fig. 5. As expected, the cell viability and proliferation were not hindered by the presence of the Gd-doped ZnO QDs at the concentration up to 1 mM. Further, the cell viability was still about 50% at the highest dose of 5 mM. In comparison, some traditional Cd-based QDs showed toxicity even at a very low concentration [36e38]. For example, a cell viability loss of 50% was observed for procine renal proximal cells after exposure to PEGylated CdSe/ZnS QDs [39]. These results illuminated a remarkably low toxicity of the as-synthesized Gd-doped ZnO QDs. Another concern derived from the possible release of Gd3 from Gd-doped QDs due to high toxicity of free Gd3 ions. The leaching experiment was conducted by ICP analysis after incubating the Gd-doped ZnO QDs in PBS buffer for one week. The result displayed no Gd3 ion was detected in the supernatant after separation, revealing that our nanoprobes were stable in physiological environment. 3.4. Applications for in vitro uorescence imaging The favorable data obtained above make Gd-doped ZnO QDs interesting candidates for cellular imaging. Confocal laser scanning microscopy images of HeLa cells incubated with a solution of Gddoped ZnO QDs (x 0.08) for different time were next collected under the same condition and presented in Fig. 6b,c, respectively. For comparison, the confocal microscopy image of HeLa cells incubated with media only was also shown (Fig. 6a). As regards the cells incubated with QDs, the yellow emission from cells became stronger as the incubation time increase, but no emission can be observed from the cells incubated without QDs. Moreover, Gddoped ZnO QDs could attach the surface of cells and exhibit strong yellow emission in relatively shorter time (30 min) owing to their positive surface charges, indicating Gd-doped ZnO QDs could be used as effective labeling probes. Additionally, their exceptionally small size could impede cytophagy of phagocyte, providing a sufciently long time for labeling imaging. 3.5. Applications for in vitro MR imaging The Gd3 ions in the Gd-doped ZnO QDs can accelerate longitudinal (T1) relaxation of water protons and exert bright contrast in regions where the nanoprobes localize, and then we proceeded to investigate the ability of Gd-doped ZnO QDs as MRI contrast agents. T1-weighted MR image was evaluated at a 1.5 T human clinical scanner. Our result showed an enhancement of MR signal as the increasing of Gd3 ion concentrations in the range

from 0 to 0.05 mM (Fig. 7a). The specic relaxivity values (r1) evaluated from a plot of the relaxation rate (T1) versus the Gd3 1 ion concentration was determined to be 16 mM1 S1, which was much larger than that of Magnevist. Given their high relaxivity and stability, T1-weighted MR imaging was also performed on the cell pellets under the same magnetic strength. As seen in Fig. 7c, the T1-weighted MR image of HeLa cells treated with Gd-doped ZnO QDs were evidently brighter than that of untreated HeLa cells, demonstrating the Gd-doped ZnO QDs could serve as effective T1-MRI contrast agents. 4. Conclusions In summary, we have presented a facile strategy for the fabrication of novel MRI-FI nanoprobes named as Gd-doped ZnO QDs with reduced size. Such nanoprobes exhibited signicantly enhanced yellow emission due to the Gd doping. The utility of Gd-doped ZnO QDs as optical imaging agents has been clearly demonstrated. HeLa cells could be successfully imaged in short time and remained high viability even at high dose of QDs, revealing the weak toxicity of Gd-doped ZnO QDs. Furthermore, Gd-doped ZnO QDs showed contrast enhancement in MRI due to their affect on the T1 relaxation times. These properties allowed Gd-doped ZnO QDs to function as effective dual modal imaging nanoprobes. Finally, we envisaged that functionalizing these nanoprobes with target ligands could be achieved due to the amine groups on their surface, and thus these nanoprobes would nd a broad range of applications in biomedical eld. Acknowledgements Financial support by the Hundred Talents Project of Chinese Academy of Sciences, National Basic Research Program of China (973 Program; No.2010CB933600), NSFC (No. 21075117; 20873138), and Alexander von Humboldt Foundation is gratefully acknowledged. Appendix. Supplementary information Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.biomaterials.2010.10.022. Appendix Figures with essential color discrimination. Certain gures in this article, Fig. 4 is difcult to interpret in black and white. The full color images can be found in the online version, at doi:10.1016/j. biomaterials.2010.10.022. References
[1] Mulder WJM, Strijkers GJ, vanTilborg GAF, Grifoen AW, Nicolay K. Lipidbased nanoparticles for contrast-enhanced MRI and molecular imaging. NMR Biomed 2006;19:142e64. [2] Lauffer RB. Paramagnetic metal complexes as water proton relaxation agents for NMR imaging: theory and design. Chem Rev 1987;87:901e27. [3] Reid T. Magnetic resonance imaging: forcing the nanoscale. Nat Nanotechnol; 2009; doi:10.1038/nnano.2009.14. [4] Caravan P, Ellison JJ, McMurry TJ, Lauffer RB. Gadolinium (III) chelates as MRI contrast agents: structure, dynamics, and applications. Chem Rev 1999;99:2293e352. [5] Anderson EA, Isaacman S, Peabody DS, Wang EY, Canary JW, Kirshenbaum K. Viral nanoparticles donning a paramagnetic coat: conjugation of MRI contrast agents to the MS2 capsid. Nano Lett 2006;6:1160e4. [6] Tsai CP, Hung Y, Chou YH, Huang DM, Hsiao JK, Chang C, et al. High-contrast paramagnetic uorescent mesoporous silica nanorods as a multifunctional cell-imaging probe. Small 2008;4:186e91. [7] Bridot JL, Faure AC, Laurent S, Rivie`re C, Billotey A, Hiba BJ, et al. Hybrid gadolinium oxide nanoparticles: multimodal contrast agents for in vivo imaging. J Am Chem Soc 2007;129:5076e84.

1192

Y. Liu et al. / Biomaterials 32 (2011) 1185e1192 [24] Werner EJ, Datta A, Jocher CJ, Raymond KN. High-relaxivity MRI contrast agents: where coordination chemistry meets medical imaging. Angew Chem Int Ed 2008;47:8568e80. [25] Bulte JWM, Kraitch-man DL. Iron oxide MR contrast agents for molecular and cellular imaging. NMR Biomed 2004;17:484e99. [26] Xiong HM, Xu Y, Ren QG, Xia YY. Stable aqueous ZnO@polymer core-shell nanoparticles with tunable photoluminescence and their application in cell imaging. J Am Chem Soc 2008;130:7522e3. [27] McLellan JS, Marcos S, Prieto PM, Burns SA. Imperfect optics may be the eyes defence against chromatic blur. Nature 2002;417:174e6. [28] Jana NR, Yu HH, Ali EM, Zheng YG, Ying JY. Controlled photostability of luminescent nanocrystalline ZnO solution for selective detection of aldehydes. Chem Commun; 2007:1406e8. [29] Vanheusden K, Warren WL, Seager CH, Tallant DR, Voigt JA, Gnade BE. Mechanisms behind green photoluminescence in ZnO phosphor powders. J Appl Phys 1996;79:7983e90. [30] Dijken AV, Meulenkamp EA, Vanmaekelbergh D, Meijerink A. Identication of the transition responsible for the visible emission in ZnO using quantum size effects. J Lumin 2000;90:123e8. [31] Dijken AV, Meulenkamp EA, Vanmaekelbergh D, Meijerink A. The kinetics of the radiative and nonradiative processes in nanocrystalline ZnO particles upon photoexcitation. J Phys Chem B 2000;104:1715e23. [32] Xiong HM, Shchukin DG, Mohwald H, Xu Y, Xia YY. Sonochemical synthesis of highly luminescent zinc oxide nanoparticles doped with magnesium(II). Angew Chem Int Ed 2009;48:2727e31. [33] Spanhel L, Anderson MA. Semiconductor clusters in the sol-gel process: quantized aggregation, gelation, and crystal growth in concentrated zinc oxide colloids. J Am Chem Soc 1991;113:2826e33. [34] Zhang L, Yin L, Wang C, Lun N, Qi Y, Xiang D. Origin of visible photoluminescence of ZnO quantum dots: defect-dependent and size-dependent. J Phys Chem C 2010;114:9651e8. [35] Monticone S, Tufeu R, Kanaev AV. Complex nature of the UV and visible uorescence of colloidal ZnO nanoparticles. J Phys Chem B 1998;102:2854e62. [36] Zhang LW, Yu WW, Colvin VL, Monteiro-Riviere NA. Biological interactions of quantum dot nanoparticles in skin and in human epidermal keratinocytes. Toxicol Appl Pharmacol 2008;228:200e11. [37] Bhang SH, Won N, Lee TJ, Jin H, Nam J, Park J, et al. Hyaluronic acidquantum dot conjugates for in vivo lymphatic vessel imaging. ACS Nano 2009;3:1389e98. [38] Wu C, Shi L, Li Q, Jiang H, Selke M, Ba L, et al. Probing the dynamic effect of Cys-CdTe quantum dots toward cancer cells in vitro. Chem Res Toxicol 2010;23:82e8. [39] Stern ST, Zolnik BS, McLeland CB, Clogston J, Zheng J, McNeil SE. Induction of autophagy in porcine kidney cells by quantum dots: a common cellular response to nanomaterials? Toxicol Sci 2008;106:140e52.

[8] Lewin M, Carlesso N, Tung CH, Tang XW, Cory D, Scadden DT, et al. Tat peptide-derivatized magnetic nanoparticles allow in vivo tracking and recovery of progenitor cells. Nat Biotechnol 2000;18:410e4. [9] Santra S, Yang H, Holloway PH, Stanley JT, Mericle RA. Synthesis of waterdispersible uorescent, radio-opaque, and paramagnetic CdS: Mn/ZnS quantum dots: a multifunctional probe for bioimaging. J Am Chem Soc 2005;127:1656e7. [10] Wang S, Jarrett BR, Kauzlarich SM, Louie AY. Core/shell quantum dots with high relaxivity and photoluminescence for multimodality imaging. J Am Chem Soc 2007;129:3848e56. [11] Li IF, Yeh CS. Synthesis of Gd doped CdSe nanoparticles for potential optical and MR imaging applications. J Mater Chem 2010;20:2079e81. [12] Yi DK, Selvan ST, Lee SS, Papaefthymiou GC, Kundaliya D, Ying JY. Silica-coated nanocomposites of magnetic nanoparticles and quantum dots. J Am Chem Soc 2005;127:4990e1. [13] van Schooneveld MM, Vucic E, Koole R, Zhou Y, Stocks J, Cormode DP, et al. Improved biocompatibility and pharmacokinetics of silica nanoparticles by means of a lipid coating: a multimodality investigation. Nano Lett 2008;8:2517e25. [14] Selvan ST, Patra PK, Ang CY, Ying JY. Synthesis of silica-coated semiconductor and magnetic quantum dots and their use in the imaging of live cells. Angew Chem Int Ed 2007;46:2448e52. [15] Kim J, Kim HS, Lee N, Kim T, Kim H, Yu T, et al. Multifunctional uniform nanoparticles composed of a magnetite nanocrystal core and a mesoporous silica shell for magnetic resonance and uorescence imaging and for drug delivery. Angew Chem Int Ed 2008;47:8438e41. [16] Lee JE, Lee N, Kim H, Kim J, Choi SH, Kim JH, et al. Uniform mesoporous dyedoped silica nanoparticles decorated with multiple magnetite nanocrystals for simultaneous enhanced magnetic resonance imaging, uorescence imaging, and drug delivery. J Am Chem Soc 2010;132:552e7. [17] Santra S, Bagwe RP, Dutta D, Stanley JT, Walter GA, Tan W, et al. Synthesis and characterization of uorescent, radio-opaque, and paramagnetic silica nanoparticles for multimodal bioimaging applications. Adv Mater 2005;17:2165e9. [18] Rieter WJ, Kim JS, Taylor KML, An H, Tarrant T, Lin W. Hybrid silica nanoparticles for multimodal imaging. Angew Chem Int Ed 2007;46:3680e2. [19] Kirchner C, Liedl T, Kudera S, Pellegrino T, Munoz JA, Gaub HE. Cytotoxicity of colloidal CdSe and CdSe/ZnS nanoparticles. Nano Lett 2005;5:331e8. [20] Hardman R. Toxicologic review of quantum dots: toxicity depends on physicochemical and environmental factors. Environ Health Perspect 2006;114:165e72. [21] Ai KL, Zhang BH, Lu LH. Europium-based uorescence nanoparticle sensor for rapid and ultrasensitive detection of an anthrax biomarker. Angew Chem Int Ed 2009;48:304e8. [22] Medintz IL, Uyeda HT, Goldman ER, Mattouss H. Quantum dot bioconjugates for imaging, labelling and sensing. Nat Mater 2005;4:435e46. [23] Drummod DC, Meyer O, Hong K, Kirpotin DB, Papahadjopoulos D. Optimizing liposomes for delivery of chemotherapeutic agents to solid tumors. Pharmacol Rev 1999;51:691e744.