The in vivo activation of persistent nanophosphors for the
optical imaging of vascularization, tumors and grafted cells
Thomas Maldiney,1 Aurélie Bessière,2 Johanne Seguin,1 Eliott Teston,1 Suchinder K. Sharma,2
Bruno Viana,2 Adrie J.J. Bos,3 Pieter Dorenbos, 3 Michel Bessodes,1 Didier Gourier,2 Daniel
Scherman, 1,* Cyrille Richard.1,*
Supplementary Methods
Chemicals. (3-Aminopropyl)-triethoxysilane (99%) was obtained from Sigma-Aldrich. Zinc nitrate hexahydrate (>99%) was purchased from Fluka. Gallium oxide (99.999%) and chromium (III) nitrate nonahydrate (99.99%) were purchased from Alfa Aesar. Dimethylformamide (>99.9%) was purchased from SDS. Alpha-methoxy-omega-N-hydroxysuccinimide poly(ethylene glycol) PEG MW 5.000 Dalton was bought from Iris Biotech GmbH. Preparation of ZGO. ZnGa1,995Cr0.005O4 nanoparticles (optimized composition, see Fig S1) were synthesized by hydrothermal method and low-temperature sintering in air. First, gallium nitrate was formed by reacting 8.94 mmol of gallium oxide with 10 mL concentrated nitric acid (35 wt%) under hydrothermal condition at 150°C overnight. Then, a mixture of 0.04 mmol of chromium nitrate and 8.97 mmol of zinc nitrate in 10 mL of water was added to the previous solution of gallium nitrate under vigorous stirring. The resulting solution was adjusted to pH 7.5 with an ammonia solution (30 wt%), stirred for 3 hours at room temperature, and transferred into a 25 mL Teflon-lined stainless steel autoclave for 24h heat treatment at 120°C. The resulting compound was washed several times with water and ethanol before drying at 60°C for 2 hours. The dry white powder was finally sintered in air at 750°C for 5 hours. Hydroxylation was performed by basic wet grinding of the powder (500 mg) for 15 minutes, with a mortar and pestle in 50 mL of 5 mM NaOH solution, and overnight vigorous stirring of the resulting suspension at room temperature. Nanoparticles with a diameter of 60 nm were first selected from the whole polydisperse colloidal suspension by centrifugation on a SANYO MSE Mistral 1000 at 4500 rpm for 5 minutes. They were located in the supernatant (assessed by Dynamic Light Scattering). The supernatants were gathered and concentrated to a final 5 mg/mL suspension. Then, nanoparticles with a diameter of 40 nm were selected from the 60 nm concentrated suspension by centrifugation on a SANYO MSE Mistral 1000 at 3500 rpm for 5 minutes. The centrifugation step was repeated 4 times and the resulting suspension concentrated to 5 mg/mL in 5 mM NaOH. Micrometric ZGO powder was prepared by a solid state reaction according to the procedure described earlier.i Nanoparticles functionalization. ZGO-OH nanoparticles were coated according to slightly modified existing protocols.ii Briefly, ZGO-NH2 nanoparticles were obtained by adding 20 µL of 3-
The in vivo activation of persistent nanophosphors for optical imaging of vascularization, tumours and grafted cells
aminopropyl-triethoxysilane (APTES) to a suspension of 5 mg ZGO-OH in 2 mL DMF (Fig. 4.a). The reaction mixture was sonicated for the first 2 minutes using a Branson Ultrasonic Cleaner 1210 and kept under vigorous stirring for 5 hours at room temperature. Particles were washed from the unreacted APTES by three centrifugation and redispersion steps in DMF. ZGO-PEG nanoparticles were obtained by reacting 10 µmol of MeO-PEG5kDa-NHS (50 mg) with 5 mg of ZGO-NH2 nanoparticles in 1 mL DMF (Fig. 4.a). To ensure a maximum PEG density, the last functionalization step was achieved overnight, under vigorous stirring at 90°C. Nanoparticles characterization. X-ray diffraction patterns were obtained on a Panalytical X’Pert Pro diffractometer with an incident-beam Ge monochromator, at U = 45 kV and I = 40 mA. The photoluminescence excitation spectrum was recorded on a Varian Cary Eclipse Fluorescence spectrophotometer at room temperature. Persistent luminescence emission spectrum was measured after 2 minutes excitation under the orange/red LEDs source. The latter is a LEDs array presenting a luminous flux of 5700 lumen (Bridgelux) associated with a 515 nm cut-off filter (Supplementary Fig. S19). Light was collected via an Acton SpectraPro monochromator coupled with a Princeton CCD camera cooled at -70°C. ZGO nanoparticles were characterized using transmission electron microscopy TEM (JEOL JEM-100S) and dynamic light scattering and zeta potential measurements in 20 mM NaCl, performed on a Zetasizer Nano ZS (Malvern Instruments, Southborough, MA) equipped with a 632.8 nm helium neon laser and 5-mW power, with a detection angle at 173° (non-invasive backscattering). From the TEM image (see Fig. 4b), the nanoparticles are spherical with a mean size about 40 nm. TSL excitation spectra. A fibre-coupled grating monochromator in combination with a Xe arc lamp allows the monochromatic excitation of the sample. A Risø Thermoluminescence reader (TL/OSL-DA-15) collects the thermoluminescence glow curves during constant heating at 5°C/s via a cooled photomultiplier tube (Hamamatsu R943-02) and an interference filter (Andover 700FS80). The setup is fully automated and software controlled by a user interface written in LabVIEW. Typically, a measurement time of 10 h was needed for the full TSL excitation spectrum shown in this work. All spectra were corrected for the wavelength-dependent optical output of the excitation source. Note that TSL glow curves and persistent luminescence decay curves from Fig. 1 were recorded after a thermal bleaching at 350°C for 20 minutes. Oral administration. A suspension of 2.4 mg of ZGO in 200 µL of water or small dry grains of 1.4 mg of ZGO were orally given to a mouse. At chosen time, animals were excited for 2 minutes with the LED lamp and placed on their back under the photon-counting device, and the signal acquisitions were recorded for 1 minute. Semi-quantization was achieved through the use of Biospace developed software, PhotoVision+. Experiments were conducted in agreement with a regional ethic committee for animal experimentation.
Figure S1. Influence of chromium concentration on the intensity of luminescence of ZGO. For the hydrothermal synthesis 3 different concentrations of chromium were tested: 0.125, 0.25 and 0.5 %. Among them, we found that the optimal concentration, in term of intensity of the decay curves, is 0.25%.
Characterization of the nanoparticles
Figure S2. X-ray diffractogram of ZGO nanoparticles. The diffraction peaks corresponding to single-crystalline ZnGa2O4 spinel phosphor are marked with an asterisk.
TSL measurements at various temperature and after various excitation wavelengths
The low temperature PL and TSL curves give additional results in particular on the vibronic bands as presented in Fig. S3 and S7.
600 660 720 780
70K
AS-PSBS-PSB
N2
Inte
nsity
(a.u
.)
Wavelength (nm)
318K
R
Figure S3. Cr3+ emission extracted from the TSL spectra at various temperatures
Cr3+ emission extracted from the TSL curves spectra change with temperature. As expected, Anti stokes phonon side bands (AS-PSB) in the above spectrum vary with temperature. In addition as temperature increases, the contribution of N2 becomes more and more important. In the highest temperature (318 K), the contribution of N2 is largely dominant. Indeed the LLP spectrum (corresponding to TSL curve at 318 K) is dominated by N2 line emission and associated phonons side bands. TSL spectra at low temperature, were indeed important in order to better understand the mechanism. TSL spectra under excitation in the conduction band (excitation at 290nm) and under excitation in the trivalent chromium absorption band (at 550 nm) are presented in Fig. S9.
1
10
100
1000
10000
100000
0 50 100 150 200 250 300
coun
ts
Time (s)
UV excitation
H20NaOH 5 mMNaCl 0,9%Glucose 5%
1
10
100
1000
10000
100000
0 50 100 150 200 250 300
coun
ts
Time (s)
LED excitation
H2ONaOH 5 mMNaCl 0.9%Glucose 5%
BA
Figure S4. Persistent luminescence decay after orange/red LED excitation (right) and UV (left) in four different media. ZGO-OH nanoparticles were dispersed in frequently used media (H2O,
NaOH, NaCl, glucose) at a concentration of 1 mg/mL (200 µg), then placed in a 96 wells plate, excited for 2 min either with UV or orange/red LED and the persistent luminescence decay was recorded for 300 s using Optima Photoimager (Biospace). The distance between the lamp and the suspension of nanoparticles is kept at 10 cm and same diaphram is used.
Photostability of n-ZGO
Figure S5. Photostability of n-ZGO. Several 2 minutes LEDs excitation cycles (black arrow), have been performed. Highly stable persistent luminescence signal are obtained. The optical signature and the decay curves remained clearly unaffected after several LEDs excitation cycles
Figure S6. Thermally stimulated luminescence (TSL) glow curves of nanosized-ZGO (n-ZGO) and bulk-ZGO (b-ZGO).
The thermally stimulated luminescence (TSL) glow curves of b-ZGO and n-ZGO after 10 s excitation at 300 nm. Broad glow peaks responsible for the persistent luminescence are observed peaking at 92°C and 108°C for b-ZGO and n-ZGO, respectively. The maximum temperature of TSL peaks being directly related to the trap depth, a broader distribution of traps and deeper traps are
observed in n-ZGO (broad glow peaks extending toward higher temperature) than in b-ZGO. However the overall area intensities of the glow curves are similar showing that both bulk and nano materials store and re-emit a similar amount of light.
100 200 300 400 500
Inte
nsity
(Nor
m)
Temperature (K)
Excitation 290nm Excitation 550nm
100 200 300 400 500 600
104
105
106
107
100 200 300 400 500 600
102
103
104
Log
Inte
nsity
Time (s)
Exc.290nm-150K Exc.290nm - 325K
Log
Inte
nsity
Time (s)
Exc.550nm-150K Exc.550nm-333K
A B
Figure S7. A. Normalized TSL spectra 20K-600K for two different excitations (UV at 290 nm and orange excitation at 550 nm). B. Decay of the persistence luminescence at room temperature and at low temperature for two excitation wavelengths (290 nm 1.2mJ and 550 nm 20mJ)
In the case of excitation in the Cr3+ bands, the TSL intensity is rather weak but still clearly observable. This gives an indication that the mechanism of charge trapping/detrapping is localized in the close vicinity of Cr3+ ions, as 4T2 and 2E are well below the CB edge. With 550 nm excitation (4A2 4T2 excitation), the low temperature TSL peaks (below 250K) have almost disappear and present a weak intensity. Moreover, varying the excitation wavelength induces a shift in the position of the main peak maximum. For all excitations below 400 nm (above 3.1 eV), the peak maximum is observed at around 318 K while excitations above 400 nm (below 3.1 eV) lead to a maximum at around 333 K.
All these results indicate that the LLP mechanism is different according to the excitation energy. In addition to the well known mechanism occurring when high excitation energy is used, a localized mechanism around the Cr3+ ions could be considered.
Stability of Cr3+ state under UV-visible excitation
The LLP/TSL mechanism proceeds first by the excitation of a Cr3+ ion, followed either by a radiative recombination, or by the release of either an electron (giving a Cr4+ ion with S = 1), a hole (giving a Cr2+ ion with S = 2 in weak field or S = 1 in strong field) or an exciton (maintaining the Cr3+ state with S = 3/2). Consequently any change of the oxidation state of chromium modifies its paramagnetism, which can be detected with high sensitivity by Electron Paramagnetic Resonance (EPR) spectroscopy. Figure S2 shows the EPR spectra of Cr3+ in b-ZGO powder recorded at X-band (9.4 GHz), at room temperature in the dark (black spectrum) and under in situ UV excitation in the EPR cavity (red spectrum). The spectra are typical of Cr3+ in a weakly axially distorted octahedral site. Each transition, including the weak ones, belong to the two CrR
3+ and CrN23+ sites. It appeared
clearly that the spectrum recorded under UV excitation was perfectly superimposed to the one
recorded without irradiation. To go further into this point, the magnetic field was set at the maximum amplitude of the strong line at 174 mT, and the EPR intensity was recorded versus time before and during UV excitation. The result is shown in the insert of Fig. S2, and indicates that the EPR intensity decreased only by less than 1% during UV illumination. However this small variation was likely due to a slight heating of the sample during UV excitation, which decreased the magnetic susceptibility according to the Curie’s law (intensity proportional to 1/T). This small decrease can be accounted for by an increase of the sample temperature of ~ 2-3 K. It is concluded that Chromium remained in the 3+ state during UV excitation. We therefore conclude that both electron and hole (exciton) are released from Cr3+ and trapped during excitation.
Figure S8. X-band EPR spectrum of b-ZGO powder recorded at room temperature before (black spectrum) and during (red spectrum) in situ UV-excitation by a lamp from ORIEL Instruments (model -66028; 200 Watts). The incident light was IR-filtered UV. The insert shows the evolution of the EPR amplitude before (black line) and during (red line) UV excitation, recorded at the magnetic field of the maximum of the strong EPR line (arrow). EPR spectra were recorded on a continuous-wave Bruker ELEXSYS E500 spectrometer. Microwave frequency 9.445 GHz; Microwave power: 0.2 mW mW; modulation depth: 1 mT at 100 kHz.
Biodistribution of crude persistent luminescence nanoparticles in vivo and ex vivo
Figure S9. Short-term biodistribution of ZGO-OH nanoparticles. The signal corresponds to near-infrared persistent luminescence from ZGO-OH nanoparticles excited with UV before systemic injection. The persistent luminescence intensity is expressed in false color unit (1 unit = 2800 photons per s.cm2.steradians). The acquisition was performed 10 minutes, 1 hour, and 2 hours after the injection of ZGO-OH nanoparticles in healthy mouse. The acquisition period was respectively 3 minutes, 10 minutes and 20 minutes.
In situ visible excitation of ZGO nanoparticles, not excited ex vivo with UV
A B
Figure S10. Image of mice after ZGO injection, without prior UV excitation. A : Image after the injection 100 µL of non excited ZGO nanoparticles (2 mg/mL). B : Image after 2 min excitation of the mouse with the orange/red LED. The suspension of ZGO was injected into the tail vein 2 hours before the excitation.
Figure S11. Ex vivo biodistribution of ZGO-OH nanoparticles. The acquisition was performed for 3 minutes immediately after a two-minute LEDs illumination of the organs, 15 hours after the injection of ZGO-OH nanoparticles into the tail vein of a healthy mouse. a, Persistent luminescence image after organ collection (1 unit = 2800 photons per s.cm2.steradians). b, Ex vivo optical quantification from persistent luminescence image.
Comparison of background signals with fluorescence acquisition (autofluorescence) and persistent luminescence acquisition
Figure S12. In vivo comparison of background signals prior to quantum dots and persistent luminescence nanoparticles injection in healthy mice. The experiments were conducted by acquiring both fluorescence and persistent luminescence background signals before the injection of
the imaging probes. a, Persistent luminescence image acquired for 4 minutes (53 ms per frame) without any excitation/emission filters. b, Autofluorescence signal acquired before the injection of QDs for 30 seconds (500 ms per frame) with 677 nm-excitation/780 nm-emission filters. The luminescence/fluorescence intensity is expressed in false color count (1 count = 2800 photons per s.cm2.steradians).
Photoluminescence emission and excitation spectra of ZGO-PEG
Figure S13. The PL emission spectrum and excitation spectrum of ZGO-PEG nanoparticles. The emission spectrum (red line) is composed of 2 E 4A2 lines of Cr 3+ ions. The excitation spectrum (black line) evidenced the two 3d-3d bands of Cr3+ at about 425 nm and 560 nm ascribed to 4A2 4T1 and 4A2 4T2 respectively.
Biodistribution of PEGylated persistent luminescence nanoparticles in vivo and ex vivo
Figure S14. Reticulo-endothelial system uptake dynamics of ZGO-PEG nanoparticles, the first six hours after intravenous injection to healthy mice. The persistent luminescence intensity is expressed in false color unit (1 unit = 2800 photons per s.cm2.steradians). A region of interest (ROI) comprising both liver and spleen was drawn on the acquired images to determine the percentage of total luminescence signal retained within the major reticulo-endothelial system (RES) organs. Semi-quantization then allowed relative precise approximation of the half-life of ZGO-PEG nanoparticles, following systemic administration to healthy mice. a, Biodistribution of ZGO-PEG nanoparticles, 2 hours after the intravenous injection (UV excitation of ZGO-PEG nanoparticles before the injection ; 10 minutes acquisition time). b, Biodistribution of ZGO-PEG nanoparticles, 4 hours after the intravenous injection (LED excitation of ZGO-PEG nanoparticles through the animal tissues ; 3 minutes acquisition time). c, Image-based accumulation dynamics of ZGO-PEG nanoparticles within the major RES organs (liver and spleen), after systemic injection.
Figure S15. Biodistribution of ZGO-PEG nanoparticles, 24 hours after intravenous injection. After organ collection, ex vivo quantification was performed from persistent luminescence signal.
Cellular and systemic toxicity of ZGO-PEG nanoparticles
Figure S16. Cellular and systemic toxicity of ZGO-PEG nanoparticles. a, In vitro cellular toxicity of ZGO-PEG nanoparticles towards RAW cells (MTT assay). Nanoparticles at the indicated concentrations were incubated with cells for 48 hours. b, Plasmatic dosage of classical biomarkers for systemic hepatotoxicity. Results are expressed in international units (IU). c, Liver, spleen and kidney histology (n=4). Livers (c, d, e), kidneys (f, g, h) and spleens (i, j, k) were collected from mice before (c, f, i), one day (d, g, j) and four weeks (e, h, k) after intravenous injection of ZGO-PEG nanoparticles (2 mg / mouse). Organs were stained with haematoxylin-eosin-safran. The scale bar is 100 µm for all images.
Results from Fig. S16a indicate no significant cellular toxicity of ZGO-PEG nanoparticles towards murine macrophages after a 48 hours incubation period in vitro. To assess specific hepatotoxicity due to the large accumulation of ZGO nanoparticles in liver, we then dosed the plasmatic level of three major biomarkers commonly referred to for the detection of a loss in liver function (aspartate aminotransferase, alanine aminotransferase and gamma glutamyl-transpeptidase). Results from
Supplementary Fig. S16b compare the plasmatic level of these biomarkers in a control group, and two groups injected with ZGO-PEG nanoparticles (2 mg / mouse). Neither acute toxicity (24 hours after the injection of ZGO-PEG nanoparticles) nor long-term toxicity (1 month after the injection of ZGO-PEG nanoparticles) was detected. Such result is in total agreement with the histology study displayed in Supplementary Fig. S16c to k. Histopathologically, no significant toxicity was observed within liver, spleen, or kidneys. In particular, there is no detectable inflammatory response in the vicinity of Kupffer cells or hepatocytes. In addition, liver, spleen and kidneys showed no significant change in morphology after the injection of ZGO-PEG nanoparticles. Along with the cellular viability assay, these preliminary results regarding systemic as well as hepato-toxicity indicate that ZGO nanoparticles are not responsible for any acute toxicity in healthy mice.
Internalization of ZGO-NH2 in macrophages revealed by TEM
a b
Figure S17. TEM images of macrophages. a, without nanoparticles. b, after incubation with ZGO-NH2 (100 µL, 1 mg/ml) for 6 hours. On the transverse section of the macrophage we can clearly see the nanoparticles into vacuoles.
Optical detection of ZGO after oral administration
Figure S18. In vivo detection of ZGO in mice after oral administration. a-d, Images obtained after the administration of a ZGO nanoparticles dispersed in water (2.4 mg/200 µL). a, Image obtained 2 hours after the administration. b, Image obtained 4 hours after the administration. c, Photograph of an excised mouse 24 hours after the administration of ZGO. d, Luminescence signal of the excised mouse 24 hours after the administration. e-h, Images after the administration of grains of ZGO (1.4 mg). e, Image obtained 2 hours after the administration. f, X-ray image showing the localization of the grains of ZGO 2 hours after the administration (white arrow). g, Image obtained 4 hours after the administration. h, X-ray image showing the localization of the grains of 4 hours after the administration (white arrow). For each luminescence image, the mouse was excited under the LED lamp for 2 minutes, the luminescence signal was then acquired under the Biospace imager for 1 minute.
Figure S19. Emission shape and characteristics of the UV and LED lamps. Two lamps were used for the excitation of the particles (UV 254 Hg Lamp and red LED Philips Lumiled 70W 5700 lm equipped with 515 nm cut‐off filter). The emission of both lamps was collected by an optical fiber connected to a CCD Roper Scientific detector (cooled Pixis 100 camera) coupled with a SpectraPro 2150i monochromator. The monochromator focal lens was 150 mm, grating of 300gr/mm blazed at 300 nm and 500 nm and corrected for the apparatus response. The intensity was measured with a Scientec 372 calorimeter and calibrated lamps.
Supplementary References
i Bessière A., Jacquart, S., Priolkar, K., Lecointre, A., Viana, B. & Gourier, D. ZnGa2O4:Cr3+: a new red long-lasting phosphor with high brightness. Opt. Express 19, 10131-10137 (2011).
ii le Masne de Chermont, Q., Chanéac, C., Seguin, J., Pellé, F., Maîtrejean, S., Jolivet, J.P., Gourier, D., Bessodes, M. & Scherman, D. Nanoprobes with near-infrared persistent luminescence for in vivo imaging. Proc. Natl. Acad. Sci. USA 104, 9266-9271 (2007).
