Potential applicability of polyethyleneimine PEI-coated Eu2O3 and Dy2O3 nanoparticles for contrast enhancement in computed tomography

Rare-earth metal oxide nanoparticles considered promising contrast agents for x-ray computed tomography (CT) and magnetic resonance imaging (MRI). The main purpose of this study is to investigate the potential applicability of polyethyleneimine (PEI)-coated Eu2O3 and Dy2O3 nanoparticles (NPs) for CT x-ray attenuation. Morphology and other physicochemical properties of prepared samples were systematically investigated using a range of characterization tools. Preliminary cytotoxicity experiments with L-929 fibroblastic cells suggested that both samples have no significant toxicity at concentrations below 100 μg ml−1. Clinical CT analysis shows that PEI@Eu2O3 NPs exhibit higher x-ray attenuation efficiency (∼8 HU mM−1) as compared to PEI@Dy2O3 NPs (∼5 HU mM−1).


Introduction
Lanthanide elements are widely used for the preparation of phosphorescent powders, glasses, permanent magnets, electronic components, etc Recently, nanostructures based on lanthanide elements attracted considerable interest in biomedical sciences because of the high density of heavy atoms per unit area and their excellent paramagnetic properties [1,2]. For example, small Gd 2 O 3 NPs are often used as potential contrast agents for magnetic resonance imaging (MRI), and computed tomography (CT) [3,4]. Other nanostructures based on lanthanides such as Dy 2 O 3 , Yb 2 O 3 , Eu 2 O 3 , Ho 2 O 3 , and their mixtures were also efficient for CT, MRI, and optical imaging of cells [4][5][6][7][8]. In general, some lanthanide oxide-based NPs show higher water proton relaxivity rates as compared to their molecular counterparts [9]. Typically, each nanoparticle contains a large number of metal ions located on the surface, which in turn explains the contrast enhancement. In addition, lanthanide-based NPs can efficiently attenuate hard x-ray thanks to their high attenuation coefficients [10]. Recent studies suggested that some rare earth oxide-based NPs show better x-ray attenuation as compared to iodine-based complexes making them promising in CT scanning [4,5,10].
To date, nanocrystals of lanthanide oxides were fabricated in organic solvents such as glycols. From one side, the constant usage of toxic organic solvents can cause some occupational health risks. On the other hand, unreacted part of lanthanide salts together with glycols generate chemical wastes that require further utilization. Therefore, a green approach should be introduced for the fabrication of lanthanide-based contrast agents. From Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. this point of view, environment-friendly and low-cost urea homogeneous precipitation is an attractive method for the preparation of lanthanide oxides [8,11]. In this method, urea dissolved in water is decomposed at elevated temperatures, which slowly increases the pH of the solution. As a result, a colloidal solution containing spherical-shaped lanthanide-based NPs can be obtained.
The toxicity of NPs is an important parameter that should be considered for potential applications in biomedicine. Numerous reports suggested that the toxicity of the NPs strongly depends on different parameters such as size, shape, and surface functional groups [12][13][14]. Various surface functional groups such as polyacrylic acid (PAA), polyethylene glycol (PEG), polyvinylpyrrolidone (PVP), and polyethyleneimine (PEI) are commonly used to improve the biocompatibility and colloidal stability of nanostructures [4,8,[15][16][17]. Among them, a cationic polymer such as polyethyleneimine (PEI) can create positive charges on the surface resulting in steric and electrostatic stabilization of nanostructures. To the best of our knowledge, synthesis and PEIfunctionalization of Eu 2 O 3 and Dy 2 O 3 NPs using green approaches were not reported so far. Therefore, the main objectives of this study are a) preparation of PEI-coated Eu 2 O 3 and Dy 2 O 3 NPs using green protocols, b) preliminary cytotoxicity assessment of PEI-coated Eu 2 O 3 and Dy 2 O 3 NPs, and c) testing the PEI-coated Eu 2 O 3 and Dy 2 O 3 NPs as x-ray CT contrast agents.

Synthesis of nanoparticles
High purity materials were purchased from Merck and used as received. For the fabrication of Eu 2 O 3 and Dy 2 O 3 nanoparticles, 0.5 mM of EuCl 3 ×6H 2 O (99.9%) or DyCl 3 ×6H 2 O (99.9%) were dissolved in 50 ml of ultrapure water. Next, 0.5 g of urea (99.0%-100.5%) was added to the prepared solution. The final solutions were heated at 90°C for 2 h in closed Duran glass bottles (100 ml capacity). Formed precipitates were collected by centrifugation process, dried, and calcined in air at 700°C for 1 h. The coating of the NPs with PEI was performed according to a reported protocol [17] with a slight modification. In brief, 60 mg of NPs were ultrasonically dispersed in 5 ml of ultrapure water for several minutes. Next, the solution pH was adjusted to 10 with the help of 0.1 M NaOH. Finally, 0.5 ml of PEI was added and the resulting solution was left for 24 h under constant stirring. PEI-coated NPs were collected, gently rinsed with ultrapure water two times, and naturally dried.

Characterization
Transmission electron microscope TEM JEM 3010 (JEOL Ltd) was used for morphology and size analysis of prepared NPs. The size and zeta potential of prepared NPs were analyzed using Nanotrac Wave II Q (Microtrac MRB). Structural analysis of prepared samples was performed using a Rigaku SmartLab (Rigaku Corporation) x-ray diffraction system. Fourier transform infrared absorption spectrophotometer FTIR Nicolet iS5 (Thermo Fisher Scientific) was used to analyze the surface coating of NPs. Thermal Gravimetric Analysis (TGA) measurements were performed using Simultaneous Thermal analyzer STA-6000 (PerkinElmer Inc.). PEIcoated NPs were heated from ambient temperature up to 600°C at a rate of 10°C per minute. Nitrogen was used as a purge gas (flow rate 20 ml/min). Commercial Phillips Brilliance 64 CT scanner (Phillips) was used for x-ray attenuation experiments. The following conditions were used during measurements: the x-ray source voltage 120 kV, current=482 mA, the field of view FOV=350 mm, and slice thickness=1 mm. All measurements were performed at room temperature.

Cell culture
L-929 cells (a murine fibroblast cell line from subcutaneous connective tissue) were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 1% antibiotic-antimycotic solution (including 10,000 units penicillin, 10 mg streptomycin, and 25 mg amphotericin B per ml) at 37°C in a humidified atmosphere of 5% CO 2 in the air. The cells (1×10 5 cells/well) were seeded into well plates and incubated for growth to confluence overnight.

Cytotoxicity assay
The cell viability was examined by a cell counting kit − 8 (CCK-8, Dojindo Lab.) assay, containing highly watersoluble tetrazolium salt [WST-8, 2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2Htetrazolium, monosodium salt], which is reduced to a yellow color formazan dye by mitochondrial dehydrogenases. After treated with different concentrations of NPs, the cells were incubated with WST-8 for the last 4 h of the culture periods (24 h) at 37°C in the dark. The absorbance was measured at 450 nm using a SpectraMax 340 ELISA reader (Molecular Device Co.). Cell viability was calculated by the percentage ratio of the optical density of the supernatants of each well.

Statistical analysis
All variables were tested in three independent cultures for cytotoxicity assay, which was repeated twice (n=6). Statistical analysis was performed using one-way ANOVA, followed by a Bonferroni test for multiple comparisons. Quantitative data are expressed as the mean±standard deviation (SD). The value of P<0.05 was considered statistically significant.

Results and Discussion
TEM analysis was utilized to evaluate the morphology of the prepared PEI-coated NPs. The inset of figures 1(A) and (B) shows that NPs with nearly spherical morphology were obtained in both cases. Detailed inspection of NPs revealed that both samples have different sizes despite identical synthesis conditions. In particular, the measured size of PEI@Dy 2 O 3 NPs was in the range of 79-102 nm, while the size of PEI@Eu 2 O 3 NPs was in the range of 67-79 nm. It should be noted that no agglomeration of PEI-coated NPs was observed during the TEM analysis. Figures 1(C) and (D) shows the chemical analysis of prepared samples performed by Energy Dispersive x-ray EDX analysis. One can see that Eu, Dy, O, and C elements are easily detectable suggesting the formation of PEI-coated metal oxide NPs.
Zeta potential measurements of PEI@Dy 2 O 3 NPs (+14.3 mV) and PEI@Eu 2 O 3 NPs (+16.9 mV) suggested that positive charges formed on the surface of NPs due to PEI coating. Figure 2 shows digital images of asprepared NPs dispersed in simulated body fluid (SBF) solution. One can see that these NPs can be easily dispersed in SBF, but colloidal dispersions can be stable for a limited time only. In particular, first precipitates can be found after 6 h of storage, while the majority of NPs can be found at the bottom within the first 18 h. This can be explained by the fact that these NPs have moderate zeta potential values. Therefore, additional in-vivo experiments are required to confirm the bio-stability of prepared NPs. Nevertheless, a colloidal solution of NPs can be easily formed again by sonication or even slight shaking. Figure 3(A) shows the XRD patterns of PEI@Dy 2 O 3 and PEI@Eu 2 O 3 NPs. One can easily assign the observed strong peaks to pure cubic phases of corresponding lanthanide oxides (JCPDS # 86-2476 for Eu 2 O 3 and JCPDS# 10-0059 for Dy 2 O 3 ) [18]. The broad peaks at lower angles due to amorphous PEI-coating were not detected. Next, FTIR analysis ( figure 3(B)) was utilized to confirm the presence of the organic PEI coating on the surface of Dy 2 O 3 and Eu 2 O 3 NPs. Absorption peaks observed at 561 cm −1 and 519 cm −1 were assigned to the bending vibrations of Dy-O and Eu-O bands, respectively [19,20]. The peak observed at 850 cm −1 was assigned to the bending vibration of the C-H bond [21]. The split peak around 1512 and 1402 cm −1 was attributed to symmetric and asymmetric stretching vibrations of the carboxyl group [19]. The trapped carbon dioxide displayed absorption around 2350 m −1 [21]. The broad absorption band from 3100 to 3500 cm −1 corresponds to the aliphatic primary amine N-H bond stretching vibrations [21]. The absorption in the region from 3500 and 3700 cm −1 was caused by O-H bond bending vibrations from the absorbed moisture [21]. One can observe that the absorption peaks for Eu 2 O 3 NPs were stronger compared to Dy 2 O 3 NPs. This could be attributed to the size difference of prepared samples. Smaller sized Eu 2 O 3 NPs have a bigger active surface area that can potentially accommodate a larger number of functional groups as compared to Dy 2 O 3 NPs.
Next, TGA analysis was performed to estimate the amount of PEI coating in prepared samples. Figure 4 shows the TGA curves for PEI-coated NPs measured in the range of 20°C-600°C. The PEI-coated NPs exhibited the first weight loss of about 2.8% and 2.9%, for PEI@Dy 2 O 3 and PEI@Eu 2 O 3 respectively, at the range of 20°C-225°C, which could be attributed to the evaporation of physisorbed and chemisorbed water [22]. The second weight loss above 225°C equal to approximately 13.8% and 13.1% for PEI@Eu 2 O 3 and PEI@Dy 2 O 3 NPs, respectively, correspond to the PEI coating amount in the samples. By considering these weight losses, the total amounts of Eu 2 O 3 and Dy 2 O 3 in the samples are 83.3% and 84.1%, respectively.
Preliminary cytotoxicity analysis of PEI-coated NPs has been performed to access the safety of prepared samples. L-929 mouse fibroblast cells were cultured with different NPs concentrations up to 200 μg ml −1 . Figure 5 displays the cytotoxicity profiles of PEI-coated NPs which clearly revealed a dose-dependent trend. In both samples, a significant decrease in cell viability was found after the threshold of 100 μg ml −1 . In literature, 80% of cell viability is usually accepted as a relative indicator of non-toxicity [23]. By considering this assumption, one can assume that these PEI-coated NPs show no significant toxicity at concentrations below 100 μg ml −1 . On the other hand, lanthanide-based NPs can be toxic to other cell lines. Therefore, it is very important to mention that the toxicity of prepared PEI-coated NPs must be studied with other cell lines, by other end-point viability measurements, and in-vivo.
A commercial CT scanner was further utilized to demonstrate the potential applicability of PEI@ Dy 2 O 3 and PEI@Eu 2 O 3 NPs for x-ray attenuation. Figure 6 shows the x-ray phantom images of PEI@Eu 2 O 3 NPs (A) and PEI@Dy 2 O 3 NPs (B) water suspensions at two concentrations. Two different concentrations, i.e. 0.5 mM and 1  mM were tested to observe the contrast enhancement trend with increasing of NPs concentration. Phantom images of pure water (0 mM) were also supplied for comparison. One can observe that the contrast of the phantom images obviously enhanced (become brighter) in suspensions with NPs compared to pure water. In literature, the x-ray attenuation efficiencies (η), i.e. Hounsfield Units HU per 1 mM of metal atomic concentration typically measured for comparison. The HU values per 1 mM of metal concentration for PEI@Eu 2 O 3 and PEI@Dy 2 O 3 were found to be∼8 and∼5 HU mM −1 , respectively. The literature data has shown that the observed x-ray attenuation efficiencies (HU/mM) of the examined NPs are relatively higher than that of commercial iodine contrast agent (Ultravist ® , 4.4 HU mM −1 ) [4]. One can also notice that the x-ray   attenuation efficiency of PEI@Eu 2 O 3 NPs was significantly higher compared to PEI@Dy 2 O 3 NPs. Thus, PEI@Eu 2 O 3 NPs bearing better potential to be applied as a promising x-ray CT contrast agent.

Conclusions
In conclusion, PEI-coated Dy 2 O 3 and Eu 2 O 3 NPs were prepared and investigated their applicability as potential CT contrast agents. The proposed green and low-cost fabrication method allows fast synthesis of the sphericalshaped NPs at large scales. Preliminary cytotoxicity analysis with L-929 cells revealed that prepared samples have no significant toxicity at concentrations below 100 μg ml −1 . X-ray attenuation experiments suggested that PEI@Eu 2 O 3 NPs have better x-ray attenuation efficiency of∼8 HU mM −1 compared to PEI@Dy 2 O 3 NPs with a value of∼5 HU mM −1 . Therefore, PEI@Eu 2 O 3 NPs can be considered as a promising candidate for x-ray attenuation purposes.