Abstract
Background. Nanosized titanium dioxide (TiO2) particles are among the most widely used nanoparticles (NPs) worldwide due to their unique properties. The lingual mucosa is still neglected in terms of risk assessment with respect to the NP uptake.
Objectives. The aim of this study was to evaluate the effect of intragastric administration of TiO2 NPs on the mucous membranes of the tongues of albino rats, as well as the potential benefits of a 4-week recovery period.
Material and methods. Twenty-one male albino rats were randomly divided into 3 groups (n = 7 per group). Group I served as a control group and received saline intragastrically daily for 30 days. The experimental groups included group II, which received TiO2 NPs in the amount of 50 mg/kg of body weight (b.w.) intragastrically daily for 30 days, and group III, which also received TiO2 NPs in the amount of 50 mg/kg b.w. intragastrically daily for 30 days, and then was allowed to recover for 4 weeks. Tongue specimens were collected for histological, immunohistochemical and morphometric examinations.
Results. The TiO2 NP group II showed significant atrophic and degenerative changes in the tongue mucosa, which included reduced epithelial thickness in the ventral surface, with disfigurement in the filiform and fungiform papillae. Weak B cell lymphoma-2 (Bcl-2) immunoreactivity was also noted. The 4-week recovery group displayed improvement in the histological picture, with moderate to strong Bcl-2 immunoreactivity in the epithelial cell layers and the underlying connective tissue.
Conclusions. The use of TiO2 NPs caused severe histological and apoptotic changes in the filiform and fungiform papillae, and the ventral surface of the tongue of the rats, while allowing recovery minimized the toxic effect of the NPs.
Keywords: titanium, titanium dioxide, nanoparticle, Bcl-2, rats
Introduction
Continuous improvement in nanotechnology engineering has led to the extensive manufacturing and use of nanomaterials. Usually, nanomaterials are nanosized solid particles with a diameter ranging from 1 nm to 100 nm. Among the popular metals and metal oxides occurring in the nanosized form are gold, silver, iron, and titanium oxide nanoparticles (NPs), which are widely used due to their superior physicochemical properties, such as optical features and magnetic activity, and high thermal and electrical conductivity. Another favorable characteristic is their large surface area to volume ratio.1, 2 However, their escalating consumption raises concerns with regard to their impact on biological systems.3
Titanium dioxide nanoparticles (TiO2 NPs) are widely used, as they exhibit unique properties, including antiviral, antibacterial and antifungal effects. Moreover, TiO2 NPs are highly biocompatible, with great strength, corrosion resistance and low density.4, 5 However, their specific large surface area and quantum effects increase their chemical reactivity in the human body. Strikingly, the incorporation of TiO2 NPs in numerous products, such as food additives, cosmetics and dental care products, raises the exposure risk related to these NPs. Owing to their non-degradability and ability to accumulate in different tissues, elevated exposure to TiO2 NPs may lead to chronic cytotoxicity.6, 7
Previously published research reports that TiO2 NPs exert toxic effects on the human body. These include changes in the cell cycle, damage to the nuclear membrane and apoptosis.8, 9 Additionally, it has been proposed that TiO2 NPs might cause mitochondrial and DNA destruction within cells.10, 11 The mechanism of TiO2 NP toxicity can be summarized in 3 processes: (1) the production of reactive oxygen species (ROS)12; (2) damage to the cell wall and the lipid peroxidation of the cell membrane; and (3) the attachment of TiO2 NPs to intracellular organelles and biological macromolecules, following damage to the cell membrane.13
Exposure to TiO2 NPs can occur by many routes, including inhalation, injection, dermal and mucosal contact, and intragastric absorption. This can lead to the precipitation of TiO2 NPs in different internal organs.14, 15 Weir et al. reported that the highest amount of TiO2 as nanosized particles <100 nm was found in candies and chewing gum.16 Thus, such products expose the oral mucosa to high doses of TiO2.
The aim of the present study was to investigate the histopathological changes in the filiform and fungiform papillae, and the ventral surface of the tongue, following the intragastric administration of TiO2 NPs in albino rats, as well as the possible effect of recovery after the withdrawal of the NPs.
Material and methods
Chemicals
Titanium dioxide NPs were purchased from Nano Gate, Cairo, Egypt. The TiO2 powder consisted of 95–97% anatase phase and 3–5% brookite phase, with NPs of an average size of 15 ±3 nm and a spherical shape (based on transmission electron microscopy (TEM)). The TiO2 anatase particles were prepared by precipitation from a homogeneous solution, using titanium (IV) isopropoxide as a precursor in an aqueous solution acidified with nitric acid to a pH of 2, with a water-to-titanium mole ratio of about 200.17
Animals
Twenty-one male albino rats weighing 150–200 g were used in the study. The rats were 2 months old, which is equivalent to the age of a human child. Children are susceptible to consuming up to 4 times more TiO2/kg of body weight (b.w.) as compared to adults.18
Experimental protocol
The experiment was conducted according to the guidance and approval of the Ethical Committee at the Faculty of Dentistry of Ain Shams University, Cairo, Egypt (approval No.: 615/2017).
The animals were provided by and housed in the Animal Research Center of Ain Shams University. The animals were housed in a sterile, controlled environment, with a temperature of 25 ±2°C and 12-hour dark/light cycles. They were kept in individual cages, with 5 rats per cage. The size of the cage was 20 cm in width and 40 cm in length. The cages were cleaned on a daily basis, and the rats were allowed free access to food and water. Each rat was given a unique number. The rats were randomly and equally allocated into 3 main groups (n = 7):
– group I: the rats received saline intragastrically daily for 30 days to serve as controls;
– group II: the rats received TiO2 NPs in the amount of 50 mg/kg b.w. intragastrically daily for 30 days19; and
– group III: the rats received TiO2 NPs in the amount of 50 mg/kg b.w. daily for 30 days as in group II, and were then allowed to recover for 4 weeks after treatment.
The rats were euthanized with an intracardiac overdose of sodium thiopental (80 mg/kg b.w.), following the different experimental periods.20
Postmortem specimen processing
The rats’ tongues were excised and fixed immediately in 10% phosphate-buffered formaldehyde solution for 48 h. The specimens were washed adequately under running water, dehydrated by transferring through a series of graded alcohol, cleared in xylene, and embedded in paraffin wax according to the standard technique.21
Hematoxylin and eosin staining
Longitudinal sections, 5-micron-thick, were obtained. The right sides of the tongues were stained with hematoxylin and eosin (H&E).21 A light microscope (Leica, Wetzlar, Germany) with ×400 magnification was used to obtain images that were transferred to a computer system for analysis. The epithelial thicknesses of the filiform and fungiform papillae, and the ventral surface of the tongue were measured (in microns) using the ImageJ software, v. 1.41a (National Institutes of Health (NIH), Bethesda, USA; https://imagej.nih.gov/ij/).
Immunohistochemical staining
The left sides of the tongue tissue specimens (formalin-fixed and paraffin-embedded) were used for anti-active B cell lymphoma-2 (Bcl-2) localization.22 The polyclonal primary rabbit anti-active Bcl-2 supplied by Biocare Medical (Pacheco, USA) (1:100 in phosphate-buffered saline (PBS)), code No. CP 229 A, was applied. All sections were counterstained with hematoxylin, incubated for 2–3 min, rinsed in distilled water, dehydrated in graded alcohol, and cleared in xylene. Finally, the specimens were mounted by using an aqueous mounting medium, cover slips were put in place and the samples were examined with a light microscope.
For immunohistochemical evaluation, the Bcl-2-stained cell surface area percentage was measured using the ImageJ software, v. 1.41a.
Statistical analysis
The mean (M) and standard deviation (SD) values for the epithelial thickness (µm) and the Bcl-2 mean area percentage were analyzed using the IBM SPSS Statistics for Windows software, v. 26.0 (IBM Corp., Armonk, USA). The one-way analysis of variance (ANOVA) and Student’s t test were used to compare the data. A p-value equal to or less than 0.05 was considered statistically significant.
Results
Histopathological results (H&E staining)
Filiform papillae
Group I (control)
The control group specimens showed a normal dorsum surface of the tongue with abundant slender filiform papillae that were regularly arranged and covered with keratinized stratified squamous epithelium. These papillae presented a uniform arrangement of the epithelial cell layers, and the keratin layer was evident. There were dispersed clear cells. The underlying lamina propria revealed normal cellular and fibrous elements, and appeared to merge with the subjacent tongue muscles without a clear line of demarcation (Figure 1).
Group II (TiO2 NPs)
As compared to group I, there was noticeable disfigurement in the structure of the filiform papillae, with a marked reduction in their height; they exhibited blunt-ended tops. Also, some areas revealed a complete loss of the papillae with evident hyperkeratosis. Focal separation of the keratin layer from the underlying epithelial cells was also observed. The basal cell layer appeared disorganized, with basilar hyperplasia and some nuclear alterations in the form of pyknosis. Some of the basal and suprabasal epithelial cells had vacuolated cytoplasm. Numerous clear cells were recognized. The lamina propria showed signs of degeneration, with moderate inflammatory cell infiltration. Separation between tongue muscle fibers was noted (Figure 1).
Group III (TiO2 NPs with 4-week recovery)
The tongues of the rats that were allowed a recovery period of 4 weeks after 30 days of TiO2 NP induction were similar to those in the control group in the histological picture of the surface epithelium and the lamina propria. The epithelium appeared to have a considerable thickness and a regular basement membrane. The basal cell layer exhibited mild basilar hyperplasia and large hyperchromatic nuclei. The prickle cell layer displayed a few binucleated cells. The granular cells exhibited distinct keratohyalin granules, but a few cells appeared with vacuolation. The keratin layer was irregular, had nonuniform thickness in certain areas, and had focal areas of separation. A few clear cells were identified throughout the epithelial layers. The underlying lamina propria showed less degenerative areas, deeply stained basophilic fibroblasts and a small number of inflammatory cells (Figure 1).
Fungiform papillae
Group I (control)
The fungiform papillae were of normal mushroom shape and were composed of normal stratified squamous epithelium with a thin uniform layer of keratin and a vascular connective tissue core. A single barrel-shaped taste bud with peripherally arranged cells was observed on the upper surface of the papillae. Lingual muscle fibers ran in different directions (Figure 2).
Group II (TiO2 NPs)
The fungiform papilla was shorter and distorted, and showed signs of epithelial atrophy. Atrophic changes in the shape and orientation of the taste bud cells, which were separated, were also observed. Atrophy in the connective tissue and tongue musculature was noted (Figure 2).
Group III (TiO2 NPs with 4-week recovery)
Similar to the control group, the fungiform papillae assumed a mushroom-like appearance, with almost normal epithelial covering and a barrel-shaped taste bud with normal cells. Reduced signs of degeneration within the lamina propria and the blood vessels engorged with blood cells were observed. Lingual muscle fibers appeared to run in different directions (Figure 2).
Ventral surface of the tongue
Group I (control)
The mucous membrane of the ventral surface of the tongue had a regular, thin keratin layer that lacked the lingual papillae. The lamina propria was formed of the connective tissue. Muscle fibers were composed of the interlacing bundles running in different directions (Figure 3).
Group II (TiO2 NPs)
The ventral surface of the tongue showed the atrophy of the surface epithelium, observed as a reduction in the thickness of the surface epithelium. An apparent loss of the keratin layer was observed. The dissociation of the collagen fibers of the lamina propria and congested blood vessels were noted. Skeletal muscle fibers were apparently atrophied and separated (Figure 3).
Group III (TiO2 NPs with 4-week recovery)
The ventral surface of the tongue regained its normal surface epithelium thickness, with a thin keratin layer and a normal pattern of epithelial ridges. Lingual muscle fibers ran in different directions (Figure 3).
Morphometric analysis
Statistical analysis for the mean epithelial thickness of the filiform and fungiform papillae, and the ventral surface of the tongue in the 3 groups is presented in Table 1 and Table 2.
The ANOVA revealed statistically significant differences between the 3 groups in the mean epithelial thickness of the filiform and fungiform papillae, and the ventral surface of the tongue (p < 0.01). The highest mean epithelial thickness of the filiform and fungiform papillae, and the ventral surface of the tongue was recorded in group I (control), whereas the lowest mean epithelial thickness was recorded in group II (TiO2 NPs).
The pairwise comparison conducted with the use of unpaired Student’s t test revealed a statistically significant decrease in the mean epithelial thickness of the filiform and fungiform papillae, and the ventral surface of the tongue in group II (TiO2 NPs) as compared to both group I (control) and group III (TiO2 NPs with 4-week recovery) (p < 0.01).
Immunohistochemical results
The immunostained sections obtained from the filiform and fungiform papillae, and the ventral surface of the tongue of group I showed numerous cells with positive Bcl-2 immunoreactivity, indicating a strong reaction affecting the epithelial layers and extending to the underlying connective tissue cells in the form of brownish coloration (Figure 4, Figure 5 and Figure 6). The Bcl-2-immunostained sections of group II demonstrated a marked reduction in positively stained cells, indicating a weak reaction (Figure 4, Figure 5 and Figure 6). The Bcl-2-immunostained sections of group III revealed a moderate-to-strong reaction in the epithelial cell layers and the underlying connective tissue (Figure 4, Figure 5 and Figure 6).
Morphometric analysis
Statistical analysis for the mean percentage of the area of the cells positively stained with Bcl-2 immune stain in all groups for the different examined specimens is shown in Table 3 and Table 4.
The ANOVA displayed statistically significant differences between the 3 groups (p < 0.01), with the highest mean Bcl-2-stained cell surface area percentage in the filiform and fungiform papillae, and the ventral surface of the tongue observed in group I (control), and the lowest mean area percentage recorded in group II (TiO2 NPs).
The pairwise comparison conducted with the use of unpaired Student’s t test revealed a statistically significant decrease in the mean Bcl-2-stained cell surface area percentage in the filiform and fungiform papillae, and the ventral surface of the tongue in group II (TiO2 NPs) as compared to both group I (control) and group III (TiO2 NPs with 4-week recovery) (p < 0.01).
Discussion
Titanium dioxide NPs are extensively and frequently utilized in many fields of industry, which increases the risk of exposure and gives rise to serious controversies about their potential effects on the human body. Hence, understanding the effects of TiO2 NPs on biological systems is essential for the development of safe nanotechnology.
Even though TiO2 NPs have been considered inert and nontoxic, and are among the most often used NPs, there is increasing evidence suggesting the contrary. Generally, the cytotoxic effects of TiO2 NPs are accompanied by cell growth inhibition in different cell types.8, 23, 24 With the growing use of TiO2 NPs in the food industry and dentistry, as shown in several studies,25 there is an urgent need to evaluate the histopathological changes in the lingual mucosa following the administration of TiO2 NPs in albino rats, as well as possible recovery after their withdrawal.
The TiO2 administration methods and doses assessed in rats are variable. In this study, the rats received TiO2 NPs at a dose of 50 mg/kg b.w. per day intragastrically for 30 days as subchronic exposure.19 The intake of dietary TiO2 NPs in the United Kingdom has been estimated to be about 5 mg/person/day,26 which is equivalent to approx. 0.1 mg/kg b.w. per day.27 In this study, 500 × the estimated human exposure dose (50 mg/kg b.w.) was used as low-dose TiO2 NP exposure.
Intragastric administration was the route of choice in this study. Previous studies investigated the capacity of TiO2 NPs to cross biological barriers, such as the gastrointestinal tract. Weir et al. provided evidence that TiO2 NPs can pass from the gastrointestinal tract through the circulation to the lungs, spleen, liver, and kidneys up to 2 weeks after exposure.28 Moreover, even when the dose was low, orally administered TiO2 NPs (5, 50 and 500 mg/kg b.w.) accumulated in mice permanently, which caused inflammation, apoptosis and oxidative stress, and consequently gave rise to chronic gastritis.29
The TiO2 powder with an average size of NPs of 15 ±3 nm, consisting of 95–97% anatase phase and 3–5% brookite phase was used in this study. Jin et al. reported that TiO2 NPs with a diameter <25 nm in the anatase crystalline phase caused genotoxicity.30 In comparison with rutile and brookite, anatase has more industrial applications, but it is the most toxic form.30
Usually, the shape of the particle plays a key role in the accumulation of TiO2 NPs in the tissue and cellular response.31, 32 In this study, TiO2 NPs were spherical in shape. An in vitro study demonstrated a correlation between the shape of TiO2 NPs and cell penetration.33 Titanium dioxide NPs penetrated into normal reconstituted human buccal epithelium, with the sphere-shaped NPs found deeper in the epithelium as compared to the spindle-shaped ones.33 This might be explained by the lower cellular uptake being associated with a high aspect ratio, in which it takes longer for the cell membrane to wrap around the elongated particles.34
The light microscopic findings of this study clearly revealed that the administration of TiO2 NPs caused evident structural changes in the tongue mucosa of the rats (Figure 1, Figure 2 and Figure 3). These changes included atrophied filiform papillae, with an alteration in their normal height and number. They exhibited blunt-ended tops, with a total or partial loss of their characteristic conical shape, and evident hyperkeratosis. Moreover, the fungiform papillae were shrunken and displayed malformed architecture with degenerated taste buds. Similarly, the ventral surface of the tongue displayed atrophied epithelium with an apparent loss of the keratin layer.
This cytotoxic effect of TiO2 NPs on the tongue mucosa might be mediated by oxidative stress, as reported in an in vitro study.10 Human hepatocellular cells were exposed to TiO2 NPs at a concentration of 250 mg/mL, which caused an increase in ROS levels up to two-fold at 5-hour exposure duration, thereby causing DNA damage.10 An in vivo investigation revealed nuclear shrinkage and chromatin condensation in the nuclei of neurons as well as the DNA ladder in the hippocampus of mice as a result of TiO2 NP deposition; these features are considered the classical characteristics of apoptosis.35
Strikingly, there were areas devoid of the lingual papillae on the dorsal surface of the tongue. This might also be due to a direct effect on the genetic constituents of cells, whereby TiO2 NPs were able to induce DNA strand breaks and genetic instability.36 Thus, it could be expected that the widespread use of TiO2 NPs in many products leads to repeated exposure, which might exhaust the proliferative capability of the epithelium, and ultimately lead to epithelial atrophy.33
Occasional inflammatory cell infiltration was observed in the lamina propria of the tongues of the rats exposed to TiO2 NPs, which might designate the activation of the defense mechanism.37, 38 Parallel results were observed in animals exposed to TiO2 NPs in other studies; they exhibited inflammation in the tissues of the kidneys and liver.37, 39 Moreover, in the present study, congested blood vessels were noticed (Figure 3). This coincides with an in vivo study, which determined that the intraperitoneal injection of TiO2 NPs created capillary congestion and hemorrhage in the alveolar wall, which resulted from ROS generation.40
There was remarkable atrophy and separation in the underlying tongue musculature; this is consistent with the findings of Wang et al., who reported damage in cardiac muscles after 30 days of oral exposure to TiO2 NPs (0, 10, 50, 200 mg/kg b.w. perday).41 This muscular damage was further correlated to disrupted levels of proinflammatory cytokines and transcription regulators in the hearts of mice.42
In group III (TiO2 NPs with 4-week recovery), the histological findings (Figure 1, Figure 2 and Figure 3) revealed similar results to those in group I (control) (Figure 1, Figure 2 and Figure 3). The epithelium appeared to have a considerable thickness, with a regular basement membrane, and the underlying lamina propria showed fewer degenerative areas. These results are consistent with those of El-Sheikh et al., who reported moderate improvement in the histopathological changes in rat spleens after the administration of TiO2 NPs was ceased for 8 weeks.43
There was a significant decrease in the mean epithelial thickness in group II (TiO2 NPs) as compared to both group I and group III (p < 0.01 and p < 0.01, respectively). The fact that the epithelial thickness was almost back to normal after the recovery period, despite the marked atrophy, indicates notable basal cell proliferation to compensate for this cell loss. Desquamation is one of the most efficient processes by which the stratified squamous epithelium of the body eliminates potentially harmful agents.44
The immunohistochemical results showed a statistically significant decrease in the mean Bcl-2 area percentage expressed in group II (TiO2 NPs) as compared to group I (control) and group III (TiO2 NPs with 4-week recovery) (p < 0.01, p < 0.01, respectively). Hu et al. clearly demonstrated that hippocampal apoptosis in response to TiO2 NP exposure significantly reduced Bcl-2 expression in the hippocampus of mice.35 Thus, they concluded that TiO2 NP-induced apoptosis in the mouse hippocampus was primarily mediated through the intrinsic apoptosis pathway.35
Conclusions
Our research suggests that a low dose and extended exposure duration, which mimics the potential human exposure to TiO2 NPs, can lead to cytotoxic effects on the lingual mucosa of albino rats. These cytotoxic effects could be moderately improved by discontinuing the administration of TiO2 NPs for a period of time. We recommend increasing the period of discontinuation, as complete improvement may occur. In this sense, it is important that people who consume food containing engineered TiO2 NPs for long periods of time should be particularly concerned. Moreover, our data suggests that the oral cavity should be the subject of TiO2 NP risk assessment studies in the future.
Ethics approval and consent to participate
The experiment was conducted according to the guidance and approval of the Ethical Committee at the Faculty of Dentistry of Ain Shams University, Cairo, Egypt (approval No.: 615/2017).
Data availability
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.
Consent for publication
Not applicable.
