Maria Antonieta
Ramirez-Morales‡
*ab,
Elisa
De Luca‡
cd,
Chiara
Coricciati
ce,
Alberto
Rainer
cf,
Giuseppe
Gigli
ce,
Giuseppe
Mele
b,
Pier Paolo
Pompa
g and
Maria Ada
Malvindi
*a
aHiQ-Nano s.r.l., Via Barsanti 1, Arnesano, Lecce, 73010, Italy. E-mail: maryarmzm@gmail.com; mariada.malvindi@hiqnano.com
bDepartment of Engineering of Innovation, Università del Salento, Via Monteroni, Lecce, 73100, Italy
cInstitute of Nanotechnology (NANOTEC)-National Research Council (CNR), Lecce, 73100, Italy
dCenter for Biomolecular Nanotechnology (CBN), Istituto Italiano di Tecnologia, Via Eugenio Barsanti, 1, Arnesano, 73010, Italy
eDepartment of Mathematics and Physics “Ennio De Giorgi”, University of Salento, Lecce, 73100, Italy
fUniversità Campus Bio-Medico di Roma, Via Álvaro del Portillo 21, Roma, 00128, Italy
gNanobiointeractions & Nanodiagnostics, Istituto Italiano di Tecnologia (IIT), Via Morego 30, Genova, 16163, Italy
First published on 25th August 2023
Fluorescent core–shell silica nanoparticles are largely employed in nanomedicine and life science thanks to the many advantages they offer. Among these, the enhancement of the stability of the fluorescent signal upon fluorophore encapsulation into the silica matrix and the possibility to combine in a single vehicle multiple functionalities, physically separated in different compartments. In this work, we present a new approach to the Stöber method as a two-cycle protocol for the tailored synthesis of dual-color fluorescent core–shell silicon dioxide nanoparticles (SiO2 NPs) using two commercial dyes as model. To facilitate the colloidal stability, the nanoparticle surface was functionalized with biotin by two approaches. The biotinylated nanosystems were characterized by several analytical and advanced microscopy techniques including Fourier transform infrared (FT-IR) spectroscopy, dynamic light scattering (DLS), UV-vis, transmission electron microscopy (TEM) and confocal laser scanning microscopy (CLSM). Moreover, advanced super-resolution based on structured illumination was used for the imaging of the double-fluorescent NPs, both on a substrate and in the cellular microenvironment, at nanometric resolution 100 nm, in view of their versatile potential employment in fluorescence optical nanoscopy as nanoscale calibration tools as well as in biomedical applications as biocompatible nanosystems for intracellular biosensing with high flexibility of use, being these nanoplatforms adaptable to the encapsulation of any couple of dyes with the desired function.
Multiple-colored fluorescent NPs can optimize diagnostic imaging and therapeutic techniques by providing accurate control and monitoring of the interactions inside living cells between the target organelles and the used nanomaterial.1,23–25 As reported in the literature, several fluorescent molecules have been attached on NPs surface26,27 and different colours of fluorescent SiO2 NPs have been developed. SiO2 NPs have been functionalized with rhodamine,6 cyanine,28 have been doped with Cd/RhBITC,29 Oregon Green 488 (OG 488),30 fluorescein isothiocyanate (FITC),31 ATTO 647N, STAR 635, Dy-647, Dy-648 and Dy-649.32,33 Multicolor systems, with a size above 100 nm, has been also reported as calibration tools for imaging instruments34,35 for the precise and accurate analysis in life science and biomedicine studies.36
Despite the multiple advantages of fluorescent silica systems, SiO2 NPs tend to agglomerate in solution, especially in cell culture media,34 limiting their studies in biological systems. In order to increase the colloidal stability of NPs, different surfactants/ligands17,18,35–37 can be added to the solution38,39 using two common approaches: surface adsorption and covalent attachment.14 In this work, we adsorbed biotin on NPs surface to improve their dispersion in solution and to increase stability. Moreover, biotin could be used for further studies to improve the nanoparticles uptake in cancer cells,43–45 due to overexpression of biotin receptors in tumoral cells.40–42 Here, we propose a tailored synthesis to obtain well-defined dual-color emitting NPs, with the potential use in the biomedical field as bioimaging/sensing tool.
TEM images of the different core–shell particles (Fig. 2) showed well monodispersed and spherical shaped nanosystems with sizes of 49 ± 3 nm (DC 50), 93 ± 4 nm (DC 90) and 116 ± 3 nm (DC 120). Their size distribution was evaluated also in water solution through DLS measurements. As showed in Fig. 3, at the end of the first cycle NPs had a size respectively of 38 ± 3 nm, 64 ± 5 nm and 92 ± 5 nm, and a final core–shell nanoparticle size of 54 ± 7 nm, 88 ± 9 nm and 118 ± 15 nm.
Fig. 2 TEM images and size distribution of core–shell nanosystems (A) DC 50 (49 ± 3 nm), (B) DC 90 (93 ± 4 nm) and (C) DC 120 (116 ± 3 nm). |
Fig. 3 DLS measurements of core and core–shell silica-based nanosystems in water (A) DC 50, (B) DC 90 and (C) DC 120. |
After the morphological characterization, the fluorescence emission spectra (Fig. 4) were measured for both dyes in a wavelength range from 470 to 750 nm. Although the concentrations of the two dyes in solution were equivalent (confirmed by calibration curves in ESI S1†), comparing the emission spectrum of dual color NPs with that of single color NPs (in ESI S2a†) we observed a decreased emission of OG 488 and an increasing in the emission of ATTO 647N. The lower fluorescence intensity of OG 488 can be attributed to different factors. First, a greater amount of ATTO 647N is encapsulated in the shell, compared to the amount of OG 488 encapsulated in the core, since the shell had a significantly greater volume than the core. Second, it may be due to a partial Förster resonance energy transfer (FRET) phenomenon. According to the literature, for an efficient FRET to occur, the maximum distance between the fluorescent donor and its corresponding acceptor must be less than 10 nm.43–45 In our nanosystems not all fluorophores are within the FRET distance optimal limits, leading to a partial FRET process. In addition, there may be a small contribution of the silica matrix that adsorbs some of the green fluorescence. As confirmed by the spectra in ESI S3,† when the dyes are dispersed in solution, there is not a significant fluorescent emission signal decrease.
Fig. 4 Normalized fluorescence spectra of DC 50, DC 90 and DC 120 in ethanolic solution (λex = 450 nm). |
To improve the dispersion and stability of the colloidal suspensions a biotin surface functionalization step was performed. Biotin not only permits to decrease nanoparticle agglomeration, but also to enhance the cellular uptake,46,47 thanks to the increase of nanoparticle tissue specificity and to the overexpression of biotin receptors on the cell membrane, mainly present in breast cancer and in others types such as colon, leukaemia and lung cancer.48 Both biotin and biotin–NHS were tested for the nanosystem functionalization using three different mass ratio between the biomolecule and the nanosystem DC 50 (1:1, 1:3, 1:5). Zeta potential measurements were carried out (Table 1) to confirm the successful functionalization and evaluate the optimal conditions of dispersion and stability of the NP suspensions. Pristine dual-colored SiO2 NPs used as control showed a negative potential value 39.76 ± 0.47 mV, expected by the negative charges on the surface due to OH groups. The biotin adsorption on the NPs surface led to an important change of the zeta potential to 60.6 ± 2 mV in the ratio 1:1. The increase of biotin amount (ratio 1:3) did not show a significant change of zeta potential (62.7 ± 1 mV), suggesting biotin saturation of the NPs surface. Similar results were obtained using biotin–NHS. The ratio 1:1 (biotin:NPs) confirmed the best results, with a zeta potential of 62.8 ± 1 mV, since by increasing the amount of NPs, the potential change decreases considerably. We considered optimal the zeta potential obtained after the adsorption procedure with the ratio 1:1 (biotin:NPs) and decided to use it in our next experiments, since no additional steps are required for the functionalization, it was a faster and cheaper procedure then the carboxylic acid group activation using biotin–NHS. The presence of biotin on the NPs surface was confirmed using the commercial HABA/Avidin kit (SIGMA H-2153) and through FT-IR measurements. The concentration of biotin measured with HABA/Avidin kit was close to 3 mg mL−1, confirming the amount added for the conjugation (ESI S4†). The presence of biotin on DC 50 surface was also confirmed by the reported signals of both silica and biotin (ESI S5†).
Approach | Mass ratio biotin:NPs | Z-Potential (mV) |
---|---|---|
a Control solution: (−) 39.8 ± 0.5 mV. | ||
Biotin adsorption | 1:1 | 60.6 ± 2 |
1:3 | 62.7 ± 1 | |
1:5 | 14.7 ± 2 | |
Covalent linking | 1:1 | 62.8 ± 1 |
1:3 | 21.8 ± 2 | |
1:5 | 10.3 ± 1 |
After functionalization of DC 50 NPs, their dispersion in cell culture medium was evaluated, comparing their size distribution with pristine NPs without biotin on the surface. As confirmed by DLS measurements (Fig. 5), the NPs without biotin tended to form large aggregates in cell culture medium observed by two groups with average diameter of 95 ± 22 nm and 1066 ± 289 nm having a PDI equals to 0.89, while the biotin on the NP surface leads to the formation of protein corona but significantly reduced the particle aggregation up to 70 ± 16 nm, increasing the stability and dispersion (PDI = 0.05).
Dual-color biotin-functionalized NPs were imaged by CLSM. As shown in Fig. 6, dyes-loaded SiO2 NPs show a good dispersion on the substrate for all the sizes, and a high fluorescence brightness of both fluorescent probes. However, due to the diffraction-limited nature of confocal imaging49,50 (lateral resolution of approximately 200–250 nm) we pushed the imaging of NPs to the nanoscale, adopting structured illumination microscopy (SIM), a super-resolution approach that provides an improvement in spatial resolution of a factor of two when compared to conventional fluorescence microscopy.51 This optical nanoscopy technique has the great advantage to exploit the high frame rate acquisition and the relatively low illumination light power to reconstruct the image of the illuminated object using a conventional wide-field microscope and standard sample preparation protocols. To further improve the quality of the imaging and the achievable resolution for NPs characterization, we employed an evolved version of SIM technology, namely Lattice SIM2, which benefits from the combination of optical 3D lattice illumination patterns and the dual iterative reconstruction algorithm (Lattice SIM2), increasing signal to noise ratio and resolving structures below 100 nm in x and y, achieving an effective lateral resolution down to ∼60 nm.52,53
Fig. 6 Confocal imaging of dual-color SiO2 NPs (a) DC 50, (b) DC 90 and (c) DC 120 immobilized on coverslip. Insets display close-up views (zoom-in) of the regions marked by the dotted white square. |
As shown in Fig. 7 panel (A), with SIM2 processing we achieved a significantly improved image contrast and lateral resolution compared to unprocessed imaging. Indeed, the FWHM values extracted from the Gaussian fit of Oregon Green 488 and ATTO 647N emission intensity profiles, relative to the fluorescent spot in the inset of panel (B), showed a remarkable improvement of the xy resolution down to ca. 90 nm, close to the maximum resolution limit achievable with this technique.
It is worth underlining that the core–shell configuration of the proposed NPs paves the way to the development of theragnostic tools, where stimuli responsive features of the dyes associated to the core and the shell of the nanoparticles could be used to develop nano-sized sensors. As a proof of principle, we leveraged the reported the peculiar feature of Oregon Green 488 dye, which is marketed as insensitive to pH changes in the near-neutral pH range but pH sensitive in moderately acidic solutions, thus displaying pH-emission dependency.54 We then measured the ratio between Green 488 and ATTO 647N emission intensities in DC 50 NP exposed to different pH values in the 3–14 range. We report a response of the nanoparticles to acidic pH levels, associated to a decrease in the ratiometric readout (see ESI S6†).
In view of a future employment of our NPs as theragnostic tools in nanobiomedicine and as probes for high-resolution imaging, we investigated their behaviour within the biological microenvironment. We further confirmed the excellent fluorescent properties of the NPs by visualizing them within cells upon internalization (Fig. 8A) and observed their prominent uptake (thanks to their good dispersion in culture medium) via endocytic pathway and endo-lysosomal accumulation at the proximity of the nucleus, in line with the precedent reported literature on silica NPs.14,55,56
The NP internalization was also followed by Lattice SIM2, to improve the quality of NP detection over confocal microscopy. The imaging of sub-diffraction nanostructures at a resolution close to their effective size is strongly recommended to fully explore their behaviour, particularly with respect to the investigation of nano–bio interactions at the molecular scale. Fig. 8, panel (B) highlights the enhancement of resolution attained with SIM2 processing that enables to distinguish small clusters/single NPs and sub-cellular structures at high contrast with nanoscale spatial resolution, paving the way for advanced multicolour dynamic imaging of nanomaterial–biosystem interactions, essential to improve their rational design for biomedical applications.
For the second cycle, the ATTO 647N doped shell was obtained by stirring 1000 μL of the previous solution with 20 μL of TEOS, 900 μL of ammonium hydroxide (28%), 10 μL of dye (previously activated with APTES) and 5 mL of ethanol for another 24 hours at room temperature.
Lattice SIM2 imaging was performed using Zeiss Elyra 7 with Lattice SIM and SMLM, using the SIM2 reconstruction algorithm. NPs were excited with the laser line at 488 nm (to excite OG 488 fluorophore), at 640 nm (to excite ATTO 647N fluorophore), and the emission signals have been detected with two synchronized cameras (Dual PCO.Edge 4.2 sCMOS camera with Duolink, motorized Dual Camera Adapter for two channel simultaneous acquisitions). DAPI was excited with a 405 nm laser line and phalloidin 568 with a 561 nm laser line and using the appropriate combinations of filters and beam splitters. All the data were acquired using ZEN black software (Zeiss) and analyzed using Origin Pro 8.5 software.
Footnotes |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3na00310h |
‡ These authors have equally contributed to this work. |
This journal is © The Royal Society of Chemistry 2023 |