Tungsten oxide sol, containing highly crystalline nanoparticles of orthorhombic WO3 and having good sedimentation stability, was synthesized using a facile, ultrasonic-assisted technique. An additional steric stabilizer, dextran, was proposed to enhance the stability of WO3 nanoparticles in biological media and to reduce their in vivo toxicity. The cytotoxicity of dextran-stabilized and nonstabilized WO3 sols was studied in vitro using dental pulp stem (DPS) cell lines and breast cancer (MCF-7) cell lines. Both tungsten oxide sols demonstrated low cytotoxicity and low genotoxicity for both stem cells and malignant cells and only slightly reduced their metabolic activity in the concentration range studied (from 0.2 to 200 μg/ml). The data obtained support possible theranostic applications of tungsten oxide colloidal solutions.

1. Introduction

Tungsten trioxide (or tungsten(VI) oxide. WO3 is a semiconductor material with a bandgap width of 2.5–2.8 eV that corresponds to the visible spectrum range. WO3 nanoparticles and nanocrystalline thin films have a wide range of applications in microelectronics and optoelectronics [1], in smart windows [2, 3], in dye-sensitized solar cell engineering [4], in gas-sensing devices [57], in quantum dot-based light-emitting diodes [8], in catalysis [9, 10], in photocatalysis [1012], and in photoelectrocatalysis [13], including water splitting [1416], wastewater purification [17], and disinfection [1822].

Recently, tungsten oxide has attracted a great deal of attention due to its promising biomedical applications [2325]. WO3 nanoparticles strongly enhance the visibility of tissue structures in X-ray-based imaging techniques, namely, computed tomography (CT). The X-ray absorption coefficient of tungsten (4.438 cm2/kg at 100 keV) is much higher than that of convenient CT contrast agent iodine (1.94 cm2/kg at 100 keV) [26]. Tungsten oxide nanoparticles possessing photocatalytic properties [22, 27, 28] have been applied in photothermal [22, 26, 29, 30] and photodynamic [22, 3032] therapies. Tungsten oxide nanoparticles act as a radiation dose-intensifying agent during radiation therapy [32] and can be used as a theranostic agent for simultaneous tumour CT imaging and therapy (trimodal action: photothermal, photodynamic, and radiation [32]). The safety and hazard data on WO3 is available at PubChem [33].

In cancer theranostic applications, photoactive semiconductor nanoparticles should possess two equally important properties: minimum toxicity in the dark (for normal cells) and maximum activity upon irradiation (for tumour cells). Both of these requirements can partially be satisfied by means of varying the particles’ surface state and habitus during synthesis.

The habitus of tungsten oxide nanostructures can easily be adjusted to the application requirements. In this way, 0D (dots), 1D (rods, whiskers, and fibres), 2D (plates, films), or 3D (large particles, blocks) WO3 materials can be synthesized. Various types of nanostructured tungsten oxide have been reported, from simple, spherical nanoparticles [34] to WO3-based aerogel networks [35], quantum dots [3639], nanostructured films [40] (including nanoplate films [41], nanorod films [42], honeycomb-structured films [43], and mesoporous films [44]), nanobelts [45], nanofibres [46], nanowires [30, 46, 47], bundle-like nanowires [30, 48], nanonetworks [49], hollow spheres [50], macroporous spheres [51], wedge-like architectures [52], nanorods [53, 54], nanocuboids [34], square nanoplates [55], nanosheets [56], nanoleaves [57], and urchin-like [30, 58], flower-like [5961], and tree-like nanostructures [62, 63], etc.

Several sophisticated approaches have been developed for the synthesis of WO3 nanostructures using vapour-, liquid-, and solid-phase (both “wet” and “dry”) methods [19]. The vapour-phase route can be realized via laser ablation [64], electron beam irradiation [65], ion bombardment [66], or heat treatment [67] of tungsten-based materials; these techniques are used primarily for the production of nanostructured films and include such processes as sputtering [68] and thermal evaporation (including hot-wire [69, 70] and arc discharge vaporization [71] and spray pyrolysis [7274]). The key liquid-phase approaches for the synthesis of WO3 nanoparticles include precipitation with acids [75], hydro- or solvothermal treatment (using aqueous [7678], nonaqueous [20, 79], or mixed [80] solvents), sol-gel processing (both in aqueous [81, 82] and in nonaqueous [83] systems), reverse microemulsion-mediated routes [8487], and soft [21, 88] and hard [89] templating (including electrodeposition [90]). Solid-phase methods are based mainly on two approaches: tribochemical [91] and thermal [92] decompositions; the latter allows the production of a pure, surfactant-free, well-crystallized material without harmful impurities. For example, ammonium meta- and paratungstate are readily decomposed under heating, to form tungsten trioxide [93102].

It is a well-known fact that the photoactive properties of WO3 nanoparticles depend strongly on the crystallinity of this material: the increase in the crystallite size enhances the photodecomposition rate of organic materials [103] and, similarly, the photocytotoxicity of the nanoparticles [28]. On the other hand, an increase in crystallinity and a decrease in the hydration degree of the tungsten oxide surface (e.g., upon solvent-free synthesis of WO3 nanoparticles) should be accompanied by a decrease in their solubility and toxicity, since the cytotoxicity of WO3 nanoparticles (including genotoxic effects such as DNA damage and micronuclei) is supposedly caused by free tungsten ions [104], which could induce oxidative stress and inflammation. In this paper, we have tried to clarify these controversial issues by analyzing the cytotoxicity of the original surfactant-free aqueous WO3 sol, prepared by solid-state thermal decomposition of ammonium paratungstate, followed by ultrasonic dispersion of the resulting product in water. This sol contains highly crystalline orthorhombic tungsten oxide nanoparticles, which are expected to have low ionic solubility and toxicity and high photoactivity under irradiation. To increase the in vivo stability of the sol, it was additionally modified by the nontoxic stabilizer, dextran. Thus, the paper was aimed at the comparative study of the dark cytotoxicity and metabolic activity of normal stem cells and malignant cells in the presence of stabilized and nonstabilized sols of highly crystalline WO3. For the toxicity studies, we have chosen highly crystalline single-phase WO3 to exclude any other factors which may affect the toxicity of nanoparticles.

2. Materials and Methods

2.1. WO3 Nanoparticles’ Synthesis and Characterization

WO3 aqueous sols were prepared using a recently reported protocol [102]. Briefly, 3 g of ammonium paratungstate (high-purity grade) powder was introduced, for 10 min, into a muffle furnace preheated to 600°С, in air. Thus, the obtained WO3 powder was quenched in air, cooled and mixed with 200 mL of distilled water, and ultrasonicated in an ultrasonic bath for 6 hours to obtain a whitish yellow turbid sol. The sol was allowed to settle for an additional 3 hours and was divided into 2 portions. To one of the portions, dextran powder (high-purity grade, Sigma #31388, 0.162 g) was added and allowed to dissolve under stirring. The concentration of the stock sols as estimated using gravimetric analysis was 2 g/L. Powder X-ray diffraction (XRD) analysis was carried out using a Bruker D8 Advance diffractometer (CuKα-radiation). The scanning electron microscopy (SEM) images were obtained on a Carl Zeiss NVision 40 electron microscope at an accelerating voltage of 1 kV. Before XRD and SEM measurements, the sols were dried in air at 50°C overnight. Time-resolved dynamic light scattering (DLS) measurements of WO3 suspensions’ aggregation behaviour were performed using a Photocor Complex analyzer equipped with a helium-neon laser (). All DLS measurements were conducted at a scattering angle of 90°.

2.2. Cell Culture

Two types of cells were used in the experiments: dental pulp stem (DPS) cells and the breast cancer cell line MCF-7. It should be noted that recent studies have shown a great deal of interest in stem cells for the delivery of antitumour drugs [105108]. DPS cells were isolated from the third molar germ extracted for orthodontic indications from a healthy, 16-year-old patient. The MCF-7 breast cancer cell line was obtained from the cell bank of the Institute of Cell Biophysics of the Russian Academy of Sciences. The cells were extracted with DMEM (PanEko, Russia) containing 200 U/mL penicillin and 200 μg/mL streptomycin (Life Technologies, USA), with a syringe inserted into the dental apex and further treated with 0.25% trypsin and 0.02% EDTA (Life Technologies, USA) for 30 min at 37°C. The isolated cells were centrifuged for 2 min at 1500 rpm and resuspended to a single cell state in a culture medium consisting of DMEM/F-12 (1 : 1; Life Technologies), with the addition of 10% fetal calf serum (FCS). The obtained solution was transferred into 25 mL vials and cultured in a 5% CO2 atmosphere at 37°C with the addition of 10% FCS (HyClone), 100 U/mL penicillin/streptomycin, and 2 mM L-glutamine in DMEM (PanEko, Russia). When the subconfluent cell state was achieved, the cultured cells were treated with 0.25% EDTA-trypsin solution and passaged in 75 cm2 vials in a ratio of 1 : 3. Cells were cultured in DMEM/F-12 (PanEko, Russia), with the addition of 10% FCS, 100 U/mL penicillin/streptomycin, and 2 mM L-glutamine.

2.3. MTT Assay

The determination of mitochondrial and cytoplasmic dehydrogenases activity in living cells was carried out using a MTT assay based on the reduction of the colourless tetrazolium salt (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT)). After 24 hours of cell incubation with different concentrations of WO3 nanoparticles, 0.5 mg/mL of the MTT reagent was introduced into the wells by replacing the culture media, followed by a standard MTT assay.

2.4. Live/Dead Assay

Assessment of the viability of the cells cultured in the presence of WO3 nanoparticles was performed on a Carl Zeiss Axiovert 200 microscope. An L-7007 LIVE/DEAD BacLight Bacterial Viability Kit (Invitrogen) was used for the assay, which included a SYTO 9 fluorescent dye (absorption 420 nm, emission 580 nm) and a propidium iodide (PI) dye (absorption 488 nm, emission 640 nm). The dyes were added to the medium (1 μg/mL), and the plate was placed in a CO2 incubator for 15 min. Microphotographs were taken after washing the cells with a phosphate-buffered saline.

2.5. Mitochondrial Potential

Mitochondrial membrane potential (MMP) was determined by JC-1 dye, using fluorescence microscopy according to the standard procedure [109]. JC-1 accumulates in the mitochondrial membrane in a potential-dependent manner. The high potential of the inner mitochondrial membrane facilitates the formation of the dye aggregates (J-aggregates) with both excitation and emission shifted towards red light (530 nm/590 nm) when compared with that for JC-1 monomers (485 nm/538 nm) [109]. Cells were seeded into 96-well tissue culture plates (Greiner) at a density of 5·104 cells/well in 100 μL culture medium and cultured in a CO2 incubator at 37°C for 24, 48, and 72 hours. The cells were preincubated with 5 μM JC-1 in the HBSS in a CO2 incubator at 37°C for 30 min. Next, the cells were washed twice using HBSS and analyzed using a 200M Zeiss inverted fluorescence microscope (Zeiss, Germany) at 200x magnification. Results are shown as a ratio of fluorescence, measured at 530 nm/590 nm (aggregates) to that measured at 485 nm/538 nm (monomers).

2.6. Fluorescent Staining of Cell Nuclei with Hoechst 33342 Dye

Cells were cultured in 96-well plates, as described above. After 24, 48, and 72 hours of culturing with WO3 nanoparticles, the cells were washed with HBSS, prior to 20 min staining with Hoechst 33342 (1 mg/mL). Images of stained cells were captured by fluorescence microscopy, and the percentage of apoptosis was calculated by counting (there were >600 cells per group).

2.7. Statistical Analysis

The experiments were carried out in 3-4 replicates, and analytical determinations for each sample were performed in duplicate. The results of the experiments were compared with the control experiment. Methods of variation statistics were applied to estimate the reliability of the results. To assess the statistical significance, the Mann-Whitney test was used (). The obtained data were processed using Microsoft Excel 2007 software.

3. Results and Discussion

WO3 sols, both nonstabilized and stabilized by dextran, demonstrated very good sedimentation stability. After 7 days of storage, the volume fraction of clear liquor did not exceed 7% (Table S1, Supplementary Information). According to X-ray powder diffraction data (Figure 1(a)), the sols obtained were composed of highly crystalline orthorhombic tungsten trioxide (β-WO3) with a particle size of , as calculated using a full-profile analysis of X-ray diffraction patterns. SEM images (Figure 1(b)) were in good agreement with the XRD data and showed that ultrasonication resulted in complete disintegration of WO3 aggregates with the formation of free-standing particles.

The results of the DLS study show that the hydrodynamic diameter of nonstabilized particles was about 54 nm (radius 27 nm); for dextran-stabilized nanoparticles, this value was doubled (Figure 2).

For the cytotoxicity analysis of WO3 nanoparticles, two types of human cell cultures were selected: normal (stem), isolated from the pulp of a healthy donor’s tooth, and transformed, isolated from a breast tumour. This choice was made taking into account their different metabolic activity, morphology, and proliferative activity. From the results of the MTT test, it is clearly seen that MCF-7 cells (Figure 3(a)) proliferated much faster than DPS cells (Figure 3(b)). The analysis of the metabolic activity of MCF-7 and DPS cells using the MTT test after incubation with WO3 nanoparticles, at all the concentrations tested (0.2–200 μg/mL), did not reveal any significant difference with the control group for both types of cell culture. The differential analysis of the ratio of live/dead cells (Figure 4) after incubation with WO3 nanoparticles also did not reveal significant differences with the control group. Morphological features of DPS cells after incubation with WO3 nanoparticles remained unchanged; cells retained the characteristics of fibroblast-like cell cultures, including effective adhesion, spreading, and migration by the leading edge scheme. MCF-7 cells also retained their original morphology and activity after incubation with WO3 nanoparticles, which additionally confirmed the absence of a toxic effect.

Oxidative stress is the major mechanism of the cytotoxic action for the metal-based nanomaterials [110]. The development of oxidative stress can be associated with a disturbance in the mitochondrial metabolism, which leads to an increase of the intracellular ROS level [111]. It is also well known that nanocrystalline tungsten is capable of generating ROS in biological media via the Haber-Weiss reaction [112]. In this regard, we studied the mitochondrial membrane potential (MMP) after the incubation of the cells with WO3 nanoparticles (Figure 5). The quantitative analysis of micrographs revealed a slight decrease of MMP in a dose-dependent manner. This decrease suggests that WO3 nanoparticles still had a certain effect on the cell metabolism, but did not cause cell death, according to the LIVE/DEAD assay.

Next, we performed a morphological analysis of the cell nuclei upon incubation with WO3 nanoparticles, to detect possible signs of genotoxicity (Figure 6). No visible changes in the nuclear apparatus were observed in either normal or malignant cells in the whole WO3 nanoparticle concentration range. Nevertheless, a more detailed study should be performed to confirm the absence of genotoxicity (for example, comet assay).

Thus, it can be concluded that WO3 nanoparticles did not exert short-term (after 24 hours) toxic effects in MCF-7 and DPS cells, although it is worth noting that only the short-term effects of WO3 nanoparticles on human cells were investigated, while the long-term effects of WO3 nanoparticle exposure are still to be studied, including long-term cyto-, geno-, and embryotoxicity studies.

4. Discussion

As a rule, tungsten oxide nanoparticles demonstrate low cytotoxicity and are relatively safe in vitro in the concentration range of up to 1000 μg mL–1. For example, WO3 nanoparticles prepared by the electrical arc discharge method in deionized water [113] and coated by cross-linked chitosan demonstrated no significant cytotoxicity at concentrations up to 5000 μg/mL after 24 hours of incubation. Similarly, PEG-poly-ε-caprolactone encapsulated tungsten oxide nanoparticles (average diameter 108 nm, hydrodynamic diameter 152 nm) were synthesized by thermal decomposition of a tungsten precursor (WCl6) in a polar nonaqueous solvent (diethylene glycol) and modified using a block copolymer [114]. The MTT cytotoxicity assay on HeLa cells demonstrated that obtained nanoparticles exerted a negligible cytotoxic effect within the concentration range of 0.1–5000 μg/mL after 24 h incubation. The same protocol was used for the synthesis of polyacrylic acid-capped tungsten oxide nanoparticles, which also exerted a negligible cytotoxic effect on the A549 human alveolar basal epithelial cell line within the concentration range of 50-1000 μg/mL after 24 hours of incubation [115]. Two-dimensional WO3 nanoplatelets with sizes ranging from 30 to 100 nm and a thickness of around 5–10 nm were synthesized by thermal decomposition of WCl6 in a nonpolar solvent, and coated with a poly-ε-caprolactone layer [116]. The cytotoxicity assays on the HeLa cell line demonstrated a LD50 value of about 0.01 M (about 2300 μg mL–1) and the absence of toxicity for concentrations up to 0.001 M (about 230 μg/mL). Toxicity profiles of uncoated and coated nanoparticles in vitro were very similar, indicating that the polymer did not affect the intrinsic cytotoxicity of WO3. On the contrary, suspensions of uncoated particles were shown to be extremely toxic in vivo, leading to mouse death within a few seconds. Chinde et al. studied the toxicity mechanisms of different concentrations (0–300 μg/mL) of WO3 nano- and microparticles in human lung carcinoma (A549) cells [117]. The microparticles were nontoxic in the entire range of concentrations studied, while the nanoparticles were nontoxic in the concentration range of up to 100 μg/mL only. The WO3 concentration of 200 and 300 μg/mL, after 24 h of exposure, led to a significant increase in the percentage of tail DNA, micronucleus formation, and intrinsic apoptotic cell death. Zhou et al. synthesized tungsten oxide nanorods with a length of and a diameter of by a facile thermal decomposition, then modified them with methoxypoly (ethylene glycol) (PEG) carboxyl acid via ligand exchange [118]. According to the MTT cell viability assay (human epithelial cervical cancer cell line HeLa and normal mouse fibroblast cell line L929), these nanoparticles had no cytotoxicity in the dark (without irradiation) in the concentration range up to 125 μg/mL after 4 h incubation. Here, we demonstrated that our novel, facile, and scalable approach to the synthesis of nanocrystalline WO3 allows for the preparation of stable WO3 sols containing highly crystalline nanoparticles, which ensures their low cytotoxicity and gives them a prospect for biomedical applications.

Actually, the stability of WO3 sols is a key prerequisite for their biomedical application, since, generally, due to the high density of tungsten oxide (7.16 g/cm3), WO3 nanoparticles precipitate readily. The intermolecular forces of attraction (e.g., those of van der Waals) cause nanoparticle agglomeration/aggregation [119] and sol coagulation/sedimentation [120]. The stability of colloidal solutions depends on the repulsion between nanoparticles; there are two main mechanisms of such repulsion: electrostatic and steric [121, 122]. The former is the result of electrical double layer formation around the nanoparticles due to charge separation. When two nanoparticles approach each other, overlapping their double layers, strong repulsion occurs [123]. This mechanism is dominant in weak electrolytes; the thickness of the double layer drastically decreases at higher ionic concentration, nanoparticles approach much closer, and attraction takes place. Since the biological fluids are strong electrolytes, containing many components, this can cause instability of sols and aggregation of particles [124]. This is why the second, steric mechanism of stabilization should preferably be used for biomedical colloidal system engineering [125, 126]. Here, repulsion is attained by adsorbed big organic molecules (uncharged nonionic surfactants or polymers), forming the protective layer on the surface of nanoparticles, preventing their collision due to the energetically unfavourable interaction of hydrated chains (when the steric stabilizer has good hydrophilicity) [127]. A similar approach was used by Zhou et al. [26] to modify WOx nanoparticles with methoxypoly (ethylene glycol) carboxyl acid to provide their good water dispersability and biocompatibility.

Generally, in diluted electrolytes, there are no differences in the properties of sols stabilized via the two mechanisms, but their behaviour drastically changes in biological media [128132]. It is a well-known fact that WO3 nanoparticles having no steric stabilizers (electrostatic stabilization only) are very toxic in vivo, probably due to the particles’ aggregation in blood vessels and capillaries [116]. Thus, despite the fact that the stabilizer (dextran) has little effect on the toxicity of tungsten oxide nanoparticles in vitro, dextran-stabilized sol is more promising for further in vivo applications, for example, in X-ray imaging.

5. Conclusions

Highly crystalline orthorhombic tungsten oxide nanoparticles were synthesized by ammonium paratungstate thermal decomposition, and a very stable aqueous sol was prepared by their ultrasonication. WO3 nanoparticles were shown to be nontoxic for DPS stem cells and MCF-7 breast cancer cells; they did not cause cell death in the concentration range studied (from 0.2 to 200 μg/mL) and only slightly reduced the metabolic activity of the cells. A nontoxic steric stabilizer (dextran) was proposed for further in vivo administration of WO3 nanoparticles. The low toxicity of dextran-stabilized WO3 sol to normal (stem) and malignant cells makes this preparation a possible candidate for biomedical applications including X-ray imaging.

Data Availability

All the data used to support the findings of this study are included within the article and the supplementary information file.

Conflicts of Interest

The authors declare that there is no conflict of interest regarding the publication of this paper.


This research has been supported by the Russian Science Foundation (project 18-73-10150).

Supplementary Materials

Table S1: sedimentation stability of stabilized and nonstabilized WO3 sols after 26 days of storage. (Supplementary Materials)