Lanthanide-doped upconversion nanoparticles: non-invasive trackers for real-time hydrogel degradation monitoring in-vivo

Jalani, Ghulam (Department of Mining and Materials Engineering, McGill University, Montreal, QC, H3A 0C5, Canada)
Naccache, Rafik (Institut National de la Recherche Scientifique-Énergie, Matériaux et Télécommunications, Université )
Rosenzweig, Derek H. (Department of Surgery, McGill University, Montreal, QC, H3G 1A4, Canada)
Makhoul, Georges (Department of Surgery, McGill University, Montreal, QC, H3G 1A4, Canada)
Abddalla, Sherif (Department of Surgery, McGill University, Montreal, QC, H3G 1A4, Canada)
Lerouge, Sophie (Centre de recherche du CHUM, Montreal, QC, H2X 0A9, Canada)
Haglund, Lisbet (Department of Surgery, McGill University, Montreal, QC, H3G 1A4, Canada)
Vetrone, Fiorenzo (Institut National de la Recherche Scientifique-Énergie, Matériaux et Télécommunications, Université )
Cerruti, Marta (Department of Mining and Materials Engineering, McGill University, Montreal, QC, H3A 0C5, Canada)


Biodegradable hydrogels are three dimensional (3D) crosslinked polymeric networks widely used as support matrixes for cell culture in tissue engineering. Hydrogel degradation monitoring is crucial to design a suitable scaffold. It is, however, quite challenging to track the hydrogel degradation in-vivo. The main methods explored so far for real-time in-vivo gel degradation monitoring are magnetic resonance imaging (MRI) and fluorescence spectroscopy. Fluorescence-based assays are the most widely used method for hydrogel tracking in-vivo: they are simple and inexpensive but have some serious drawbacks when used in biological systems. Most of the fluorescent labels consist of organic dyes, which have poor photostability and are prone to photobleaching during long exposures to lasers. Here we use silica coated Lanthanid doped upconversion nanoparticles LiYF4:Yb3+/Tm3+@SiO2 to track the degradation of Chitosan-Hyaluronic acid (CH-HA) gels in living tissues using NIR imaging and photoluminescence (PL) spectroscopy.

Materials and Methods

Yb3+/Tm3+-doped LiYF4 nanoparticles are prepared via thermal decomposition of trifluoroacetate precursors of Y, Yb, Tm in the presence of CF3COOLi, Oleic acid (OA) and 1-Octadecene (OD).(1) The precursor solution prepared in OA and OD is injected into a second flask containing OA and OD at 325 ºC, using a syringe pump at a constant flow rate under argon purging. The decomposition is carried out at this temperature for 60 min. Reverse microemulsion process is used to prepare UCNPs@SiO2. Briefly, OA-capped UCNPs are dispersed in cyclohexane; IGEPAL, TEOS and NH4OH are added and the mixture is stirred for 48 hrs at room temperature.(2) CH-HA-BGP-GN gels are synthesized using our previous method.(3) To prepare gel/NP composites, UCNPs@SiO2 are dispersed in CH-HA-BGP-GN solutions at a concentration of 0.5 mg/ml. 150µl of the gel/NP mix are injected into a live intervertebral disc (IVD) followed culturing at 37 ºC for 3 weeks.


The average length, width and thickness of the NPs are 78±9nm, 43±6nm and 8±1nm respectively (Figure 1A). TEM images UCNPs@SiO2 reveal that an 8±2 nm thick shell of SiO2 is obtained around each NP (Figure 1B). The PL spectra of UCNPs in figure 1C shows bands spanning UV, Vis and NIR regions: spectrum of UCNPs@SiO2 shows similar but less intense bands. X-ray diffraction shows that both UCNPs and UCNPs@SiO2 are crystalline with tetragonal crystal structure (Figure 1D). The experimental setup for NIR imaging is shown in figure 2A. The NIR images collected with laser shining from the top of the IVD (Figure 2B-P1) at different spots on day zero (Figure 2C) show emitted radiation only in the image collected at the center of the IVD (spot O). The image in spot O shows a bright spot at the top (Figure 2C-P1-O), and light emitted throughout the disc. Similar images are obtained in reverse configuration, i.e. laser shining from the bottom. The images collected on different spots on day 3, 5, 8, 15 and 22 are shown in figure 3A, and in figure 3B we plot the intensity of PL spectra at 792nm as a function of time. The images collected on day 3 show emission only at spot O. With the passage of time, the intensity at spot O decreases while it starts increasing gradually at spot A, A′, B and B′ (Figure 3A and 3B).

Discussion and Conclusion

TEM images in figure 1 indicate that both UCNPs and UCNPs@SiO2 are monodispersed and do not aggregate during synthesis. Also, PL spectra show that both types of NPs are highly photoluminescent: the lower intensity of UCNPs@SiO2 is most likely due to quenching due to the silica shell. The NIR images in figure 2C prove that UCNPs are efficiently excited throughout the whole vertical section of the disk (labeled as “t” in Figure 2A) and that the upconverted light travels a distance of ~1.2 cm (labeled as “r” in Figure 2A), since it can be captured by the NIR camera positioned outside the disk. The origin of the brighter spot seen in figure 2C-P1-O is clarified from the images captured in a reversed configuration (Figure 2B-P2) implying that it is not related to a high UCNP concentration in this location, but rather to higher laser intensity. After 3 days, all of the NPs are located at the center of the IVD (spot O), indicating that no degradation is happening (Figure 3A). After 5 days, a small amount of NPs is detected in points A and A′; however, most of the NPs are still visible at the center of the IVD. This phenomenon continues until an equilibrium is reached on day 22. The PL spectra emitted by the UCNPs@SiO2 in figure 3B show the intensity of the band at 792 nm over time at each spot closely match the images shown in figure 3A. The in-vitro NP release is also plotted in figure 3B for comparison (dashed line).

TEM images of (A) UCNPs and (B) UCNPs@SiO2, showing homogenous NPs with narrow size distribution. (C) PL spectra of UCNPs (black line) and UCNPs@SiO2 (red line). (D) X-ray diffraction patterns of UCNPs (black line) and UCNPs@SiO2 (red line) with corresponding (h k l) values.

(A) Schematic showing the IVD and the location of the points irradiated by the laser.(B) Schematic showing the two configurations used to check the effect of laser positioning, either from the top (P1) or the bottom (P2) of the IVD. (C) NIR images of IVD on day zero with two configurations.

(A) NIR images of IVDs while shining with 975 nm light at the locations shown in Figure 2A over a period of 3 weeks. (B) Plots of PL intensity measured at λ=792 nm while shining the laser on positions O, A, A’, B and B’ on the IVD as a function of time. Dashed line shows in-vitro release of NPs.


We Acknowledge Prof. Faleh Tamimi for useful technical discussion and Dr. Marta Quintanilla for optical measurements: NSERC, FRQNT, CSACS, CRC, CFI and MEDA for financial support.


1. Mahalingam V, Vetrone F, Naccache R, Speghini A, Capobianco JA. Colloidal Tm3+/Yb3+-Doped LiYF4 Nanocrystals: Multiple Luminescence Spanning the UV to NIR Regions via Low-Energy Excitation. Adv Mater. 2009;21(40):4025-4028. 2. Li ZQ, Zhang Y, Jiang S. Multicolor Core/Shell-Structured Upconversion Fluorescent Nanoparticles. Adv Mater. 2008;20(24):4765-4769. 3. Jalani G, Rosenzweig DH, Makhoul G, Abdalla S, Cecere R, Vetrone F, et al. Tough, In-Situ Thermogelling, Injectable Hydrogels for Biomedical Applications. Macromolecular Bioscience. 2015: DOI: 10.1002/mabi.201400406

Copyright ©1990 - 2020
Web Development by Inc.

Close Drag