*Article* **The Vitamin A and D Exposure of Cells A**ff**ects the Intracellular Uptake of Aluminum Nanomaterials and Its Agglomeration Behavior: A Chemo-Analytic Investigation**

#### **Fabian L. Kriegel \*, Benjamin-Christoph Krause, Philipp Reichardt, Ajay Vikram Singh, Jutta Tentschert, Peter Laux, Harald Jungnickel and Andreas Luch**

German Federal Institute for Risk Assessment, Department of Chemical & Product Safety, Max-Dohrn-Straße 8-10, 10589 Berlin, Germany; Benjamin-Christoph.Krause@bfr.bund.de (B.-C.K.); Philipp.Reichardt@bfr.bund.de (P.R.); Ajay-Vikram.Singh@bfr.bund.de (A.V.S.); Jutta.Tentschert@bfr.bund.de (J.T.); peter.laux@bfr.bund.de (P.L.); Harald.Jungnickel@bfr.bund.de (H.J.); Andreas.Luch@bfr.bund.de (A.L.) **\*** Correspondence: Fabian.kriegel@bfr.bund.de

Received: 17 December 2019; Accepted: 12 February 2020; Published: 14 February 2020

**Abstract:** Aluminum (Al) is extensively used for the production of different consumer products, agents, as well as pharmaceuticals. Studies that demonstrate neurotoxicity and a possible link to Alzheimer's disease trigger concern about potential health risks due to high Al intake. Al in cosmetic products raises the question whether a possible interaction between Al and retinol (vitamin A) and cholecalciferol (vitamin D3) metabolism might exist. Understanding the uptake mechanisms of ionic or elemental Al and Al nanomaterials (Al NMs) in combination with bioactive substances are important for the assessment of possible health risk associated. Therefore, we studied the uptake and distribution of Al oxide (Al2O3) and metallic Al<sup>0</sup> NMs in the human keratinocyte cell line HaCaT. Possible alterations of the metabolic pattern upon application of the two Al species together with vitamin A or D3 were investigated. Time-of-flight secondary ion mass spectrometry (ToF-SIMS) imaging and inductively coupled plasma mass spectrometry (ICP-MS) were applied to quantify the cellular uptake of Al NMs.

**Keywords:** nanoparticle uptake; ICP-MS; ToF-SIMS; aluminum; vitamin; metabolomics

#### **1. Introduction**

Aluminum is one of the most abundant metals that is used in a wide range of industrial manufacturing processes. It is also present in numerous consumer products such as cosmetics or food contact materials. Studies on Al toxicity revealed a potential risk for neuronal toxicity in humans following chronic Al exposure [1,2] and a possible relation of enhanced Al intake to the development of Alzheimer's disease [3].

The uptake of Al may occur via different routes of exposure. Al NMs (nanomaterials) may cross the barriers of the body because of their small size and thus can significantly increase the overall Al burden [4]. Humans are most likely exposed to Al through cosmetic products due to skin contacts or via food additives [3]. To lower the intake of Al, the first legal actions were taken by the EU regulation No. 380/2012 amending Annex II to Regulation (EC) No. 1333/2008, which became applicable at 1st of August 2014 [5]. Further to this, the use of Al-containing food additives is restricted by the recommendation of the European Food Safety Authority (EFSA) to lower the tolerable weekly intake (TWI) of Al to ≤1 mg/kg body weight. However, it was suggested that this TWI might be significantly exceeded, especially in children [6].

To avoid such a high intake, the conditions simulating use and tolerable quantities for food additives containing Al were adapted in the EFSA regulation. The body's uptake of Al is influenced by several processes. For example, vitamin D3 not only enhances the uptake of essential inorganic elements but also of non-essential and toxic elements such as lead or Al [7]. Vitamin D3 also interferes with retinol metabolism and its uptake in human epidermal keratinocytes [8]. The studies related to the interaction between vitamin D3/A and Al may comprise, among other biological endpoints, Al-mediated toxicity, its distribution pattern at the cellular level, as well as its capability to modulate metabolic patterns of skin cells. The complexation behavior of Al is highly sophisticated and certainly adds to the challenges and elucidation of uptake mechanisms. Furthermore, the ubiquitous occurrence of Al resulting in a high background level hampers the application of sensitive analytical methods. The relatively low density of the light metal, its low mass and the complexity of biological matrices complicate the use of widespread analytical techniques like Raman spectroscopy or transmission electron microscopy (TEM). ToF-SIMS analysis [9] has been previously applied for the detection of cerium dioxide particle clusters in rat lung tissue [10] and for the characterization of Al particles in artificial saliva [11]. In this study, we tried to overcome the mentioned analytical limitations by using ToF-SIMS to explore uptake and distribution patterns of Al and Al2O3 NMs in the human keratinocyte cell line HaCaT. Furthermore, metabolic profile changes of the cell membrane constituents were investigated. Aluminum chloride (AlCl3·6H2O) was used as soluble ionic control. All three Al species were tested with regard to uptake, cellular distribution patterns, and possible cell membrane alterations upon uptake. In addition to single applications, combinations with vitamin A and D3 were tested as well. ICP-MS measurements were utilized to determine the uptake efficacy of each Al species in the different scenarios.

#### **2. Results and Discussion**

#### *2.1. Characterization of Al and Al2O3 NMs*

Both, Al<sup>0</sup> and Al2O3 NMs have been extensivly characterized by our group [12] (see Table 1). The core particle diameter was determined by means of TEM to be between 2–50 nm for the rather spherical Al0 NMs. Single particle (SP) ICP-MS showed a primary particle size range of 50–80 nm for both Al NMs. The results of the small angle x-ray scattering (SAXS) measurments confirmed the findings of TEM and SP-ICP-MS.


**Table 1.** Characterization data for Al0 and Al2O3 nanomaterials (NM). Modified from [12].

The more rod shaped Al2O3 NMs display a width of 10 nm and a length between 20–50 nm as determined by TEM. Dynamic light scattering (DLS) determined the hydrodynamic diameter in dispersion of 250 nm for Al0 NMs, while Al2O3 NMs showed a smaller diameter of 180 nm. Zeta Potential measurements of Al0 and Al2O3 NM showed comparable results with <sup>−</sup>17.2 mV and −17.3 mV in DMEM. X-ray diffraction (XRD) as well as TEM in electron energy loss spectroscopy mode (EELS-TEM) demonstrated the difference of the Al0 and Al2O3 NMs composition and especially their surface. While Al0 NMs had a core-shell structure with Al core and a 2–5 nm oxygen shell (see Figure 1), the Al2O3 NMs were homogeneously oxidized.

**Figure 1.** Transmission electron microscopy (TEM) results: (**A**) TEM pictures of Al<sup>0</sup> NMs; (**B**) oxygen mapping of left TEM picture; (**C**) TEM picture of Al0 NMs; (**D**) aluminum mapping of image in (**C**).

#### *2.2. Cellular Uptake*

To investigate the possible uptake of Al NMs and Al2O3 NMs, we exposed HaCaT cells to 100 μg/mL Al NMs for 24 h in the presence or absence of 1 μmol/L retinol, 5.12 μmol/L of vitamin D3 (high vitamin D3), or a lower vitamin D3 concentration of 80 nmol/L (low vitamin D3). Subsequently we analyzed the exposed cells using ICP-MS to quantify the NM uptake. Quantitative results are shown in Figure 2. Untreated cells had Al levels comparable to ICP-MS blank samples.

**Figure 2.** Inductively coupled plasma mass spectrometry (ICP-MS) measurements of Al content of HaCaT cells exposed to either Al2O3 or Al0 NMs as well as retinol and/vitamin D3. The cellular uptake is normalized to the Al uptake of cells exposed to Al0 NMs or Al2O3 NMs only. \* *p* < 0.05.

For the interpretation of the results, it has to be considered that the data shown are normalized to the Al<sup>0</sup> uptake by HaCaT cells exposed to Al<sup>0</sup> only (without co-exposure to vitamins). The ICP-MS results show that retinol and vitamin D3 appear to have no detectable effect on the uptake of the Al0 NM. In contrast, treatment of HaCaT cells with retinol or vitamin D3 significantly lowered the uptake of the Al2O3 NM (Figure 2).

The different uptake behavior of the two kinds of Al NMs after treatment of the HaCaT cells with vitamins is likely due to differences in their physicochemical properties. It is known that the complexation behavior and thus the agglomeration rate of the used Al NMs differ strongly [12]. Al2O3 is characterized as a rather insoluble and rod shaped oxidized NM, whereas Al<sup>0</sup> NMs are partially soluble and quasi-spherical. Differences in the particle physicochemical properties (e.g., surface area, solubility, etc.) may lead to differences in the uptake mechanism preferred by the cells.

These findings presented in this study are in good accordance with the results of an uptake study of polystyrene NMs on Caco-2 cells [13]. Furthermore, the composition of the NMs might also influence their uptake and distribution [14]. The core-shell structure of the Al0 NM (see Table 1) contains approximately 85% Al. In contrast, Al2O3 is fully oxidized and shows a homogenous distribution of Al as well as oxygen on its surface. Our group was able to demonstrate that the protein corona that is formed during the contact of NMs with cell culture media is less complex for Al<sup>0</sup> NM compared to the protein corona of Al2O3 NM [15]. The surface properties of NMs facilitate interactions with the surrounding environment, which also affects the bioavailability and the interactions of the NM with the cell.

For the assessment of changes of the metabolite patterns of the HaCaT cell membranes and the overall particle distribution, ToF-SIMS analytics was employed. The acquired ToF-SIMS mass spectra for HaCaT cells showed a strong Al peak indicating the presence of Al NMs (Figure 3). Cellular uptake can also be observed in HaCaT cell cultures co-incubated with retinol, vitamin D3, and its combinations. The Al NM uptake could also be observed after treatment with Al2O3 as well as in cultures cultivated with retinol plus vitamin D3 low or high for both NM species.

3D reconstruction of ToF-SIMS images from single HaCaT cells reveals the intracellular presence of Al0 and Al2O3 NMs at all exposure scenarios: NM alone, NM in combination with retinol and/or vitamin D3.

HaCaT cells treated with Al NMs only store the particles in large agglomerates (Figure 4a,e), whereas cells treated with Al NMs in combination with retinol or vitamin D3 show a different uptake and distribution behavior (compare Figure 4b,f vs. Figure 4c,g). Treatment with retinol leads to the accumulation of Al close to the cell membrane with large agglomerates (Figure 4b,f). This process might be due to the increased collagen synthesis and reduced matrix metalloproteinase expression which is known to occur because of vitamin A treatment [16]. The resulting collagen increase facilitates NM collagen interactions [17] which might hinder the NM allocation. Upon co-application of vitamin D3 the ToF-SIMS analyses revealed a different cellular deposition pattern of Al<sup>0</sup> NMs when compared to that after co-exposure to retinol. Co-exposure with vitamin D3 led to an even distribution of particles throughout the entire cell and to the formation of much smaller aggregates (Figure 4c,d,g,h). In addition of being responsible for an enhanced uptake of metals (Figure 2), vitamin D3 also seems to affect particulate distribution patterns within the cell. The ToF-SIMS results are in accordance with the ICP-MS findings, where high vitamin D3 is responsible for an enhanced intracellular uptake (Figure 2).

**Figure 3.** Time-of-flight secondary ion mass spectrometry (ToF-SIMS) mass spectrum (positive mode), showing the Al peak in red color (at *m*/*e* 26.98 u) and a peak in green color (at 27.02 u = C2H3 +), resulting from organic matter in HaCaT cells. The upper line shows the spectra for control HaCaT cells, the lower line for HaCaT cells which were exposed to Al NMs (about 20 nm) for 24 h in addition to high or low vitamin D3, and retinol or their combinations. The x-axis shows the mass to charge ratio (*m*/*z*); y-axis the ion intensities.

The ToF-SIMS results for the exposure of HaCaT cells with Al2O3 NMs show that either treatment with retinol or vitamin D3 leads to a strong decrease in the NM uptake and to smaller agglomerate sizes when compared to cells treated with Al2O3 NMs only (Figure 2). Comparison of Al<sup>0</sup> and Al2O3 NM treatments of HaCaT cells reveals a generally smaller size of NM agglomerates in the latter ones (Figure 5). Based on this it can be assumed that the number of particle agglomerates per cell is much higher for Al2O3 than for Al<sup>0</sup> NMs. Therefore Al2O3 NM aggregation is being largely compromised resulting in a more even intracellular distribution of Al2O3 NMs when compared with Al<sup>0</sup> NM (compare Figure 4b,f vs. Figure 5b,f).

When HaCaT cells were exposed to retinol in combination with either low or high levels of vitamin D3 and Al, the localization of the particles was restricted to the cell membrane region (Figure 6a,c), which corroborates above findings in Figures 4 and 5. ICP-MS results show, however, a similar mass balance for all exposure experiments with Al NMs (Figure 2), indicating again a much larger number of smaller agglomerates in the case of exposure to vitamin D3.

**Figure 4.** Ion reconstruction of a 3D depth profile (depth layer numbers: 50–250) of one single HaCaT cell, which was exposed to Al0 NMs for 24 h. The images show the top-down view of the outline of a cell of a depth profile. The translucent blue outline was reconstructed based on the C3H8N<sup>+</sup> signal that originates from intracellular amino acids. (**a**) Control cells treated with Al0 NM only (red color); and corresponding intracellular localization of Al0 NM agglomerates (**e**). (**b**) Cells treated with retinol and Al<sup>0</sup> NM (red color); and corresponding intracellular localization of Al NM agglomerates (**f**). (**c**) Cells treated with low vitamin D3 and Al0 NM (red color); and corresponding intracellular localization of Al<sup>0</sup> NM agglomerates (**g**). (**d**) Cells treated with high vitamin D3 and Al0 NM (red color); and corresponding intracellular localization of Al0 NM agglomerates (**h**).

Uptake of Al2O3 NMs alone and in combination with retinol, vitamin D3, or both vitamins revealed wide distribution and agglomeration of Al nanomaterials throughout the cytoplasm. In contrast to all co-exposure experiments, formation of larger NM clusters was observed after exposure to Al2O3 NMs alone (Figure 5a,e). ICP-MS data show a significant increase of Al2O3 NMs being present in HaCaT cells following their straight application in contrast to the co-exposure experiments (Figure 2). In addition, ToF-SIMS results show a significant reduction in the sizes of NM agglomerates after the co-exposure experiments (Figure 5 vs. Figure 6).

#### *2.3. Metabolic Changes after Nanomaterial Uptake*

In addition to NM uptake and distribution, we assessed the alterations of the cell membrane constituents of HaCaT cells caused by Al and Al2O3 NM exposures. Treatments with ionic AlCl3\*6H2O and unexposed HaCaT cells were used for comparison (Figure 7).

**Figure 5.** Ion reconstruction of a 3D depth profile (depth layer numbers: 50–250) of one single HaCaT cell exposed to Al2O3 NMs for 24 h. The images show the top-down view of the outline of a cell of a depth profile. The translucent blue outline was reconstructed based on the C3H8N<sup>+</sup> signal that originates from intracellular amino acids. (**a**) Control cells exposed to Al2O3 NMs only (red color); and corresponding intracellular localization of Al2O3 NM agglomerates (**e**). (**b**) Cells treated with retinol and Al2O3 NMs (red); and corresponding intracellular localization of Al2O3 NM agglomerates (**f**). (**c**) Cells treated with low vitamin D3 and Al2O3 NMs (red); and corresponding intracellular localization of Al2O3 NM agglomerates (**g**). (**d**) Cells treated with high vitamin D3 and Al2O3 NMs (red); and corresponding intracellular localization of Al2O3 NM agglomerates (**h**).

The results obtained for cells exposed to retinol, low vitamin D3, and high vitamin D3 reveal significant differences in the composition of the respective cell membranes (Figure 7). Significant differences in the cell membrane composition could also be observed in HaCaT cells co-exposed to retinol and low vitamin D3 or high vitamin D3. In order to determine the differences of lipid membrane constituents we further investigated the significant changed metabolites upon co-epxosure of Al0 or Al2O3 NMs with retinol or high or low vitamin D3 (see Table 2 for Al0 and Table 3 for Al2O3). For further details please see also the Supplementary Material (Figures S1–S6).

The further investigation of changes in the membrane composition was carried out via ToF-SIMS analysis of the major altered cell membrane lipids. Diacylglycerols (DAGs) and phosphatidic acids (PAs) were found to be increased following treatment with either Al0 or Al2O3 NMs in combination with vitamins similarly (see Tables 2 and 3). DAGs were significantly increased after administration of retinol or vitamin D3 together with both Al NMs. DAGs are the product of hydrolysis of the phospholipid phosphatidylinositol 4,5-bisphosphate (PIP2) that serve as activators of the protein kinase C (PKC) pathway [18]. Alterations in the PKC signaling in HaCaT cells lead to strong morphological changes including shape of this cell type [19]. Different PKC isoforms have also different effects on the proliferation and differentiation behavior of HaCaT cells [20]. The PAs and its precursor substances lyso-phosphatidic acid were significantly increased over all treatment regimes. The PAs can be degraded to DAGs and serve as precursors of other membrane lipids [21]. Therefore, we conclude a comparable metabolic alteration of DAGs and PAs in response to Al NM uptake. Phosphatidylethanolamines (PEs) are another important group of membrane constituents. The PEs serve as substrates for phosphatidylcholine (PC) biosynthesis and are crucial for cell surface signaling [22]. Investigations on the ratio of PC to PE in mice showed decreased membrane integrity and the development of a leaky membrane when the ratio decreases [23]. In D3-treatments an increase of the level of PEs was detected because of a shift of the PCs toward PEs thereby causing a leaky membrane of the cells affected. High vitamin D3 addition revealed significantly increased levels of lyso-PC, which is a degradation product of PC and therefore indicates reduced PC levels. This process lowers the PC to PE ratio even more drastically and an enhanced uptake of Al may occur. These findings fit well with the ICP-MS results (see Figure 2).

**Figure 6.** Ion reconstruction of a 3D depth profile (depth layer numbers: 50–250) of one single HaCaT cell exposed to Al NMs (upper panel **a**–**d**) or Al2O3 NMs (lower panel **e**–**h**). The 3D depth profile of the cell is depicted as translucent blue. The images show the top-down view into the outline of a cell of a depth profile. In addition to NM treatment low vitamin D3 (**a**,**b**,**e**,**f**) and high vitamin D3 (**c**,**d**,**g**,**h**) were administered together with retinol. NM agglomerates are shown in detail next to their respective cell.

The significantly increased metabolites after treatment with Al2O3 and retinol or vitamin D3 are more diverse when compared to Al0-treated cells (see Table 3).

In addition to the above mentioned findings, another metabolite, dihydroceramides (DCs), was found to be enhanced after the treatment of HaCaT cells with Al2O3 and vitamin D3 (see Table 3). The increased levels of this compound in HaCaT cells exposed to Al2O3 NM and vitamin D3 provides another explanation for the decreased nanoparticle uptake in cells exposed to Al2O3 NM as DCs enhance the rigidity of the plasma membrane [24]. This could lead to changes in the active transport, vesicle formation, diffusion, and activation of the cell-signaling pathway, all of which are representing processes that depend on plasma membrane dynamics [25].

**Figure 7.** ToF-SIMS analysis of changes in the composition of the cell membranes of HaCaT cells after treatment with Al or Al2O3 NMs and ionic AlCl3·6H2O without or in combination with vitamins ((**a**) retinol; (**b**) high vitamin D3; (**c**) low vitamin D3; (**d**) retinol plus low vitamin D3; (**e**) retinol plus high vitamin D3). The diagram shows the values of the discriminant scores obtained from Fisher's discriminant analysis of 24 single HaCaT cells for each experiment. The performance of the discriminant model was verified by applying the cross-validation procedure based on the "leave-one-out" cross-validation formalism (100%).



**Table 3.** Significantly increased cell membrane constituents upon treatment of HaCaT cells with Al2O3 NMs divided by the additional treatment of either retinol low or high vitamin D3.


The changes in the membrane lipid composition described above serve as an explanation for the decreased uptake of Al2O3 NMs after vitamin A/D3 treatment. The increased levels of DCs after treatment with vitamin D3 could also derive from a protective mechanism of the cell to secure the physiological integrity of mitochondria. It has been previously shown that DCs block the permeabilization of the mitochondrial outer membrane [26].

The comparison of the changes in the constituents of the cell membrane that were introduced upon treatment with either Al0 or Al2O3 NMs clearly shows the necessity to distinguish between the different types of aluminum NMs.

#### **3. Materials and Methods**

#### *3.1. Cell Culture and NM Exposure*

Al NMs (18 nm, 99.9%) and Al2O3 NMs (20 nm, 99+%) were purchased from IoLiTec Ionic Liquids Technologies GmbH, Heilbronn, Germany. The following chemicals were purchased from Sigma-Aldrich (Sigma-Aldrich Corp., St. Louis, MO, USA): AlCl3·6H2O (hexahydrate, ≥97%), retinol (≥97.5%), cholecalciferol (certified reference material), and calcipotriol (European Pharmacopoeia Reference Standard). The chemicals were diluted to the respective concentrations (high vitamin D3 high concentration: 5.12 μmol/L; low vitamin D3 concentration: 80 nmol/L; retinol: 1 μmol/L) in DMSO (≥99.7%), obtained from Sigma-Aldrich (Sigma-Aldrich Corp., St. Louis, MO, USA).

The human immortalized keratinocyte cell line HaCaT was cultured in Dulbecco's modified Eagle's medium (DMEM), 10% fetal bovine serum, and 1% antibiotics (10,000 μg/mL streptomycin and 10,000 units/mL penicillin) at 37 ◦C with 5% CO2. Cells were passaged at 70–80% confluence two times a week.

For ToF-SIMS measurements 0.05 <sup>×</sup> 106 cells were seeded on 1 cm2 silica wafers and left in the incubator for 24 h. Afterwards cells were treated with the respective NM with or without vitamin derivates for 24 h. The wafers were then washed using 150 mM ammonium bicarbonate solution. Samples were fast frozen and lyophilized prior to ToF-SIMS measurements.

NM dispersions were prepared following the NanoGenoTox dispersion protocol: "Final protocol for producing suitable manufactured NMs exposure media" (October, 2011). In brief, a 2.56-mg/mL stock dispersion of each NM was prepared by pre-wetting the powder with 0.5% (vol/vol) ethanol (96%) followed by addition of Millipore water containing 0.05% BSA. Dispersion occurred for 5 min and 9 s at an amplitude of 10% with a probe sonifier, either 200 W Bandelin Sonopuls HD 2200, (BANDELIN Electronic GmbH & Co. KG, Berlin, Germany) (at BfR); or 400 W Branson Sonifier S-450 CE Digital, (Branson Ultrasonics, St. Louis, Missouri, USA) (IMPB). The sample was cooled in an ice-water bath during sonication [27].

#### *3.2. ICP-MS Analysis*

Measurements were performed with a quadrupole ICP mass spectrometer (iCAP Q, Thermo Fisher Scientific GmbH, Dreieich, Germany) equipped with a PrepFast system (ESI Elemental Service & Instruments GmbH, Mainz, Germany), PFA ST Nebulizer, a quartz cyclonic spray chamber, and a 2.5 mm quartz injector (all from Thermo Fisher Scientific, Waltham, MA, USA) using the following isotopes: 27Al and, as an internal standard, 103Rh. Calibrations were performed using ionic standards of Al in a 3.5% HNO3 solution ranging from 2 to 500 μg/L. The internal standard was added using the ICP-MS PrepFast system. The gas flows for the cooling gas and the auxiliary gas were set to 14 L/min and 0.65 L/min, respectively. The sample flow rate was 0.4 mL/min. All isotopes were analyzed using the collision cell technique at 5 mL/min collision gas flow (93% He and 7% H2). Analysis of NM uptake was studied with five replicates per dose. Results were presented as mean values ± standard error of the mean (SEM).

#### *3.3. ToF-SIMS Analysis*

A dedicated cryogenic sample preparation technique with a high cooling rate was used for sample analysis [28,29]. Liquid propane was cooled using liquid nitrogen, thus preventing evaporation of propane at the contact surface of the immersed specimen. Specimens were in contact with liquid propane for 10 s and were afterwards kept in a frozen state using dry ice. The condenser of the Christ Beta 2-8 lyophilizer (Martin Christ GmbH, Osterode am Harz, Germany) was cooled to −80 ◦C. The frozen samples were placed on the frozen heating plate, which was inserted in the freeze-drying chamber and heated to −20 ◦C. Afterwards, vacuum was applied to achieve a pressure of 1.65 mbar. The main drying process started, in which the water is sublimated within 2 h by vacuum and heated to 23 ◦C. The temperature of 23 ◦C is maintained for 30 min. The instrument was ventilated and the freeze-dried samples were stored at −80 ◦C prior to the ToF-SIMS analysis. A ToF-SIMS instrument (ION-TOF V; Ion-TOF GmbH, Münster, Germany) was used for mass spectrometry analyses with a pulsed 30 keV Bi3<sup>+</sup> liquid metal ion gun (LMIG, direct current (dc), 16 nA). Measurement of cell samples was performed at room temperature. Each spectrum was acquired by scanning the ion beam over a sample area of 400 × 400 μm. Positive secondary ions were collected in the mass range up to *m*/*z* 1200 using 106 Bi3<sup>+</sup> pulses. Instrument and analysis conditions were used as described elsewhere for the ToF-SIMS analysis of cell membrane lipids [30].

All depth profiles were performed in dual beam mode on a TOF.SIMS V instrument (ION-TOF GmbH, Münster, Germany) of the reflectron-type, equipped with a 30 keV Bi3 <sup>+</sup> LMIG as primary ion source, a 20 keV argon gas cluster ion source both mounted at 45◦ with respect to the sample surface and an electron flood gun. Bi3<sup>+</sup> was selected as primary ion by appropriate mass filter settings. Primary and sputter ion currents were directly determined at 200 μs cycle time (i.e., a repetition rate of 5.0 kHz) using a Faraday cup located on a grounded sample holder. Scanning area for analysis was <sup>200</sup> <sup>×</sup> <sup>200</sup> <sup>μ</sup>m<sup>2</sup> with 512 <sup>×</sup> 512 pixels. The sputter area for each measurement was 1000 <sup>μ</sup><sup>m</sup> <sup>×</sup> <sup>1000</sup> <sup>μ</sup>m. Surface charging was compensated by flooding with low energy electrons.

ToF-SIMS depth profiles were acquired in positive ion mode. The mass scale was internally calibrated using a number of well-defined and easily assignable secondary ions (C2H5 <sup>+</sup>, C3H7 <sup>+</sup>, and C4H9 <sup>+</sup>) keeping the error of calibration for all spectra below 5 ppm. The data were evaluated using the Surface Lab software (ION-TOF GmbH, Münster, Germany).

Statistical analyses of the ToF-SIMS data were performed as described in detail elsewhere [30–34]. In brief, the acquired data were binned to 1 u. Data processing was carried out with the statistical package SPSS + (version 21) using the mass range between 200 and 1200 mass units to detect significant differences between treated and untreated cells. Ions lower than mass 200 were excluded from the study to avoid contamination of the ions from salts, system contaminants, and other medium components. Each acquired spectrum was then normalized, setting the peak sum to 100%. A principal component analysis (PCA) was performed using all ions. To show that data sets could be separated with a supervised model from each other a Fisher's discriminant analysis was performed. The performance

of the discriminant model was verified by applying the cross-validation procedure based on the "leave-one-out" cross-validation formalism.

#### **4. Conclusions**

The ToF-SIMS measurements have shown that aggregation and incorporation of Al NMs in HaCaT cells are influenced by treatment with vitamins. Experiments with retinol led to the formation of large Al aggregates, which are intercalated in the membrane regions of the cells, while vitamin D3 treatments resulted in the formation of small agglomerates within the entire cell. These findings suggest that, depending on the vitamin treatment, different pathways are used for the uptake of Al NMs.

Furthermore, the results show a decreased uptake rate of Al2O3 NMs after treatment with vitamins (retinol and vitamin D3) in comparison with the control as well as with the exposure to Al0 NMs in combination with both vitamins. A likely explanation for this behavior is the change in the lipid membrane composition, which might lead to an enhanced rigidity of the membrane. Treatment with either retinol or vitamin D3 leads to a drastic decrease in the uptake of Al2O3 NMs.

In contrast to the protective effect observed for Al2O3 NMs, ToF-SIMS measurements revealed a changed lipid metabolite profile for cells in response to co-exposure to vitamin D3 and Al0 NMs. The metabolic changes led to a shift in the PC-to-PE ratio, which contributes to an increased uptake of Al NMs due to a leaky cell membrane.

We conclude a protective effect of the vitamins A and D3 for cells, which are in contact with nanoparticulate oxides such as Al2O3. On the other hand, the presence of these substances may slightly promote the uptake of metallic NMs. Our findings also reflect the high importance of a thorough physicochemical particle characterization, as parameters like agglomeration, solubility, and biokinetics may affect the uptake of NMs with the same elemental constituent.

#### **Supplementary Materials:** The following are available online at http://www.mdpi.com/1422-0067/21/4/1278/s1.

**Author Contributions:** F.L.K. and B.-C.K. conceived and designed the experiments. The project and research was supervised by A.V.S., P.L., and A.L. F.L.K. conducted the cell culture experiments and prepared together with P.R. the ToF-SIMS samples. B.-C.K. carried out the ICP-MS measurements and performed the statistical evaluation of the results. ToF-SIMS measurements and subsequent data analysis was performed by H.J., J.T., and P.R. The manuscript was written by F.L.K. and B.E.K. with input from all authors. All authors have read and approved the final manuscript.

**Funding:** The BfR provided intramural support SFP1322-642 for F.L.K., P.R., H.J. and A.L.

**Acknowledgments:** The authors thank Yves U. Hachenberger for fruitful discussions and his support.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Abbreviations**



#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **Manifestation of Systemic Toxicity in Rats after a Short-Time Inhalation of Lead Oxide Nanoparticles**

**Marina P. Sutunkova 1, Svetlana N. Solovyeva 1, Ivan N. Chernyshov 1, Svetlana V. Klinova 1, Vladimir B. Gurvich 1, Vladimir Ya. Shur 2, Ekaterina V. Shishkina 2, Ilya V. Zubarev 2, Larisa I. Privalova <sup>1</sup> and Boris A. Katsnelson 1,\***


Received: 6 November 2019; Accepted: 19 January 2020; Published: 21 January 2020

**Abstract:** Outbred female rats were exposed to inhalation of lead oxide nanoparticle aerosol produced right then and there at a concentration of 1.30 <sup>±</sup> 0.10 mg/m<sup>3</sup> during 5 days for 4 h a day in a nose-only setup. A control group of rats were sham-exposed in parallel under similar conditions. Even this short-time exposure of a relatively low level was associated with nanoparticles retention demonstrable by transmission electron microscopy in the lungs and the olfactory brain. Some impairments were found in the organism's status in the exposed group, some of which might be considered lead-specific toxicological outcomes (in particular, increase in reticulocytes proportion, in δ-aminolevulinic acid (δ-ALA) urine excretion, and the arterial hypertension's development).

**Keywords:** nanoparticles; lead oxide; inhalation exposure; toxicity

#### **1. Introduction**

Lead oxide nanoparticles (PbO-NP) are engineered for some essential technical applications (e.g., magnetic data storage and magnetic resonance imaging). However, from the standpoint of human health risk assessment, it is much more important that PbO particles pollute workplace and ambient air in aerosol form in such long-established and major industries as copper and lead smelters and refineries. These polluting aerosols, resulting from the evaporation of molten metal in pyrometallurgy and its condensation in the air due to cooling and oxidation, contain a substantial proportion of particles falling within the nano-range [1].

Thus, PbO-NPs pose a real inhalation exposure threat, which makes it essential to assess the latter in toxicological experiments. However, we are aware of only one study of this kind published in 2017 by a group of Czech researchers [2].

The study involved an experiment on ICR white mice continuously exposed in a whole-body inhalation setup for 6 weeks to PbO-NP aerosol generated in situ with a mean particle diameter of 25.9 nm (in the size range of 8–230 nm) with an average concentration of 121.7 μ/m3. The researchers measured the lead contents of organs and tissues with the visualization of retained nanoparticles and described pathological changes in them, concluding that "subchronic exposure to lead oxide nanoparticles has profound negative effects at both cellular and tissue levels". Particularly noteworthy are changes in the hippocampus associated with the penetration of nanoparticles into the brain from the nose as a primary site for their deposition. Such penetration is characteristic of inhaled NPs of any chemical composition, as has been discovered in a number of studies [3–5], including ours [6–8].

However, it is well known that (in contrast to changes in the brain) in a whole-body inhalation chamber, animals' fur is inevitably soiled with particles, which are then licked off by them; thus, not all toxic effects can be attributed precisely and only to inhalation exposure. It is also regrettable that this study did not consider any of the functional or biochemical indicators of organ and systemic toxicity.

There is no doubt that this toxicity is an inherent feature of PbO-NP at all biological levels—from cellular to systemic. It has been shown, in particular, under subchronic intoxication caused in vivo by repeated intraperitoneal injections of PbO nanosuspension to rats [9] and in vitro by its addition to the incubation medium for human fibroblasts [10]. It would be difficult to think of any a priori causes why the inhalation of PbO-NP should not cause lead poisoning. However, this question is so important for occupational medicine and health risk assessment in the above-mentioned industrial conditions that it must be resolved through experimental studies rather than speculatively.

As the first step in this direction, we are presenting the results of an inhalation experiment in which rats were exposed to PbO-NP aerosol at a concentration around 10 times higher than in the above-mentioned experiment [2]. However, in contrast to the latter, it lasted only five days with daily exposures, which were relatively short because of the restrained positioning of the animals in the "nose only" inhalation setup (see Figure 5). We are convinced, though, that this disadvantage is offset by a very important advantage of such setups—they exclude the possibility of nanoparticles penetrating into the organism by routes other than inhalation and the related uncertainty of the causal relationships between just inhalation exposure at a given level and toxic effects.

Overall, the inhalation period in our experiment was 50 times shorter compared with the experiment [2], and the total exposure to nanoparticles may be estimated as about five times lower, even though the concentration of nanoparticles was 10 times higher.

The purpose of this paper is to demonstrate that signs of intoxication were detectable even under an exposure as low and short as in our experiment.

#### **2. Results and Discussion**

Previously, we demonstrated that the deposition of nanoparticles in the lungs as a result of intratracheal administration [11–14] or inhalation exposure [6–8] caused a response in the lower airways that was essentially similar to but a lot more intensive than the one provoked by similar exposures to low-soluble highly cytotoxic micrometer particles, such as, for instance, quartz dust [15,16]. This response manifests itself as an increase in the cellularity of the bronchoalveolar lavage fluid (BALF), mainly at the expense of neutrophil leukocytes. The increase in proteins, in some enzymes (particularly of lysosomal or partly lysosomal origin) and in a number of other biochemical components of the BALF extracellular fraction also testifies for inhaled PbO-NP's pulmonary toxicity.

As follows from the results presented in Tables 1 and 2, we observed the same in our experiment. Although the differences from the corresponding indices of the control group's BALF appear to be rarely statistically significant, the mutual correspondence of the resulting changes allows us to consider them to be an actual effect of inhalation exposure to nanoparticles.


**Table 1.** Cell counts in the bronchoalveolar lavage fluid (BALF) in rats after repeated inhalation of lead oxide nanoparticles (x ± s.e.). PbO-NP: lead oxide nanoparticles.


**Table 2.** Some biochemical indices in bronchoalveolar lavage fluid (BALF) supernatant in rats after repeated inhalation of lead oxide nanoparticles (x ± s.e.).

\*—values statistically significantly different from the corresponding values of the control group (*p* ≤ 0.05 by Student's *t*-test).

As was stated in the Introduction, neither the total exposure to lead nor the period for related adverse changes to develop in the organism might be estimated as considerable, which reduced the probability of development of lead poisoning. Indeed, many of the indices listed in Table 3 indicate that it was of low to moderate intensity.

**Table 3.** Body and inner organs mass and some functional indices in rats after repeated inhalation of lead oxide nanoparticles (x ± s.e.).



**Table 3.** *Cont*.


**Table 3.** *Cont*.

\*—values statistically significantly different from the corresponding values of the control group (*p* ≤ 0.05 by Student's *t*-test).

Note, first of all, that body mass reduction, which was observed in the sham-exposed group as well, was not enhanced by lead exposure. This reduction was likely to be an immediate consequence of the subacute stress caused by immobilization at the beginning of the experiment when the rats were not yet sufficiently well adapted to it. Incidentally, we did not observe this effect during months of exposure in the same setup in three different experiments involving other NP species [6–8].

The masses of the internal organs (both absolute and related to body mass) did not differ statistically significantly from the control values. However, still, these indices were noticeably, even if not significant statistically, increased for lungs, the first target organ for inhaled nanoparticles; for kidneys, the main organ eliminating lead from the blood, which is well-known to suffer damage under lead intoxication (e.g., [17–19]); and for heart, whose electrocardiogram (ECG) signs of impairment will be described below.

The shift in the balance between the processes of excitation and inhibition in the central nervous system (CNS) toward inhibition is evidenced by a statistically significant increase in the index of temporal summation of sub-threshold impulses. Although none of the three behavior indices that we recorded was statistically significantly different from the control values, still, the unidirectional increase in these indices (number of head-dips into holes, number of crossed squares, total number of movements on the "open field") suggests enhanced exploratory activity.

The impairment of the porphyrin metabolism, which is specific to lead toxicodynamics, manifested itself in a more than twofold and statistically significant increase in the concentration of δ-aminolevulinic acid (δ-ALA) in urine, which is one of the early signs of lead poisoning. Typical of it is also an increased concentration of coproporphyrin in urine, which in this study was noticeable, even if statistically not significant. As is known, such impairment is a prerequisite to hem synthesis suppression; however, in this experiment, it did not reach a clearly detectable level, since neither the whole blood nor erythrocyte hemoglobin content was decreased at all. Lead anemia did not develop contrary to expectations judging also by no reduction in the erythrocyte count; however, already there was an appreciable and statistically significant compensatory enhancement of erythropoiesis which manifested itself as a considerable increase in the proportion of reticulocytes, another one of the most sensitive effects of lead intoxication.

It should be noted also that the number of micronuclei in the polychromatophilic erythrocytes of the bone marrow was more than doubled. Although this shift was not statistically significant, it is noteworthy since we practically always discover a systemic genotoxic effect under exposure to various metal oxide nanoparticles judging by increased DNA fragmentation coefficient, which is another informative genotoxicity index. For instance, it was significantly increased under subchronic intoxication caused by repeated intraperitoneal injections of PbO-NPs [9].

Turning back to the hematological features of lead intoxication, only some of the white blood indices revealed a noticeable, although statistically insignificant increase. These shifts are also noteworthy because they are fairly typical of experimental lead intoxications and, in some studies, were even more substantial than in the present one and statistically significant at that (e.g., [20–24]).

Proceeding now to the blood serum, note that the total protein and protein fractions in it were not different from the control value. At the same time, some biochemical indices point to damage to hepatocytes, which, as is well known, may show itself not only as enzyme biosynthesis inhibition (which manifested itself as a statistically significant decrease in the serum activity of γ-glutamyl transpeptidase and an insignificant decrease in the activity of alkaline phosphatase) but also as enhanced enzyme release (which led to an increased serum activity of aminotransferases, lactate dehydrogenase, and amilase). As can be seen in Table 3, some of these shifts are statistically significant.

The small increase in the kidney mass, although statistically insignificant, may assumingly be associated with nephrotoxicity, which is lead's other highly characteristic feature [17–19]. We can interpret in the same way both the increased protein content of the urine and the statistically significant, albeit small decrease in urine's specific density (despite a somewhat lower volume of diuresis) and significantly increased creatinine content (while the creatinine content of the blood serum was not changed). The δ-ALA content of the urine was increased statistically significantly and much higher than in the blood serum. Note that such combination of signs of damage to both tubular epithelium and Malpighian glomeruli (confirmed morphologically) had been observed under subchronic intraperitoneal exposure to PbO-NP [9].

The indices of calcium, myoglobin, troponin, natriuretic peptide, endothelin-1, and vascular endothelial growth factor (VEGF) content of the blood serum were studied as markers of a possible vasocardiotoxic effect of lead intoxication discussed below, but none of them revealed any substantial shift.

The data in Table 4 provide evidence that the lead exposure caused interrelated, although statistically insignificant, hemodynamic changes in the rat tail measured post exposure. All the three blood pressure levels (systolic, diastolic, and mean) were elevated compared with the control values. The natural assumption that this arterial hypertension was caused by increased systemic resistance to blood flow is in agreement with the decelerated blood flow and decreased blood volume. It is important to note that we had also discovered that all these shifts under subchronic intoxication were caused by intraperitoneal injections of lead acetate [24] but, to the best of our knowledge, nobody had previously estimated hemodynamic parameters under PbO-NP exposure of any kind.


**Table 4.** Some hemodynamic indices of rats after repeated inhalation of lead oxide nanoparticles (x ± s.e.).

It should be noted that many of the epidemiological studies have provided evidence of a cause–response relationship between human exposure to lead and the prevalence of arterial hypertension [25–29]. Animal experiments have also been performed, seeking mainly to identify the possible mechanisms of lead-induced hypertension [30–34]. However, the author of a relatively recent overview [35] concluded that "in an occupational setting, the effect of lead exposure on blood pressure remains controversial". In this context, it is to be recalled that the air inhaled by metallurgy workers is contaminated with lead in aerosol form, containing a considerable fraction of nanoparticles (see the Introduction). Therefore, the evidence obtained by us for the first time ever that even a very moderate intoxication caused by exposure to PbO nanoaerosol provokes shifts of a hypertensive type cannot be ignored.

Table 5 presents ECG analysis results obtained in two standard leads, and it is worthwhile to compare them with the data of our ECG study under subchronic lead acetate intoxication [24]. Whereas in the latter we observed elongation of the majority of the interwave intervals pointing to the slowing of the heart rate, in the current experiment, changes in the ECG intervals in the absence of bradycardia were of different types, with the only shift that was statistically significant being of opposite sign (QRS shortening). The statistically significant increase of the P and T amplitudes discovered in both ECG leads in this experiment was not observed in the previous experiment with lead acetate. At the same time, both studies had a common feature, a lowered isoelectric ECG line in the second lead, which can point to some impairment of the myocardium or, at least, metabolic disturbances in it. Such disturbances were indeed revealed in the hearts of rats suffering subchronic lead intoxication [24].



\*—values statistically significantly different from the corresponding values of the control group (*p* ≤ 0.05 by Student's *t*-test).

Transmission electron microscopy of the lung tissue revealed nanoparticles in the cytoplasm of type 1 and type 2 alveolocytes (Figures 1 and 2).

**Figure 1.** Panel (**A**)—nanoparticles (shown by arrows) in alveolocytes of type 1 (TEM, magnification ×51,640) in rat lungs from the exposed group. Panel (**B**)—alveolocytes of type 1 (TEM, magnification ×16,480) in rat lungs from the control (sham-exposed) group.

**Figure 2.** Panel (**A**)—nanoparticles (shown by red arrows) in alveolocytes of type 2 (TEM, magnification ×50,580) in rat lungs from the exposed group. Panel (**B**)—alveolocytes of type 2 (TEM, magnification ×9000) in rat lungs from the control (sham-exposed) group.

Electron microscopy of the olfactory region of the brain showed cytoplasmic vacuolization of neurons, numerous nanoparticles in the neurons' bodies (Figure 3), and pronounced demyelinization of axon membranes (Figure 4).

**Figure 3.** Panel (**A**)—a neuron body with numerous nanoparticles (shown by red arrows) and vacuolized cytoplasm (shown by a red asterisk); TEM, magnification ×41,950; Panel (**B**)—non-damaged neuron bodies in the olfactory region of a rat brain from the control (sham-exposed) group; TEM, magnification ×10,410.

**Figure 4.** Panel (**A**)—an axon with signs of membrane demyelinization (shown by red arrows) in the olfactory region of a rat brain from the exposed group; TEM, magnification ×20,050. Panel (**B**)—non-damaged axon myelin sheath in the olfactory region of a rat brain from the control (sham-exposed) group; TEM, magnification ×24,680.

The mechanism underlying the penetration of inhaled nanoparticles into the brain neurons was mentioned in the Introduction. We should just emphasize that such penetration is also associated with some ultra-structural changes in the brain tissue. It can be assumed that the somewhat controversial shifts in the CNS functional indices that we discussed above are connected not only and even not so much with systemic lead intoxication as with a direct effect of cytotoxic nanoparticles on the brain structures. In this connection, we should recall the key role of the sense of smell in animal behavior control.

#### **3. Materials and Methods**

Airborne Pb-NPs were obtained by sparking from 99.9999% pure lead rods with a diameter of 5.6 mm (supplied by "Giredmet Ltd."—Moscow, Russia) using a Palas DNP-3000 generator set at "Medium energy" regime with the current strength of 5 A and nitrogen flow of 8 L/min. Then, this flow was being mixed with air (6 L/min) for cooling and for oxidizing Pb into PbO-NPs, which were fed into a nose-only exposure tower (CH Technologies, Westwood, NJ, USA) with rats placed into individual restrainers (Figure 5).

**Figure 5.** The nose-only inhalation setup (photographed without the door wings of the draught cupboard) and a similar tower for the sham exposure of the control group.

A setup of the same design obtained from the same supplier was used for a sham exposure of control rats. Particles collected on filters and inspected under a scanning electron microscope (SEM) had a spherical shape and either were singlets or formed small aggregates (Figure 6). The latter, if compact, were measured as one particle. Even so, the particle size distribution (Figure 7) proved fairly clean-cut and restricted to the nanometric range with a mean (± s.d.) diameter of 36 ± 4 nm. The chemical composition of particles sampled on the filters was confirmed by Raman spectroscopy to be PbO (the obtained spectrum had two characteristic peaks at wavelengths 82 cm−<sup>1</sup> and 147 cm−<sup>1</sup> which, according to [36] correspond to PbO).

**Figure 6.** Nanoparticles retained by the Whatman Anodisc membrane filter (mesh diameter 20 nm) of the inhalation setup. SEM, magnification ×100,000.

**Figure 7.** Particle or particle aggregate size normal distribution function (red curve) obtained by statistical processing of 360 measured SEM images of particles accumulated on a polycarbonate filter from the exposed rats' breathing zone.

Our experiment was carried out on outbred white female rats from our own breeding colony with the initial body weight of 252 ± 1.5 g in the NP-exposed 14 rats and 251 ± 2.2 g in the sham-exposed 14 rats. All these rats were housed in conventional conditions, breathed unfiltered air, and were fed standard balanced food. The experiments were planned and implemented in accordance with the "International guiding principles for biomedical research involving animals" developed by the Council for International Organizations of Medical Sciences (1985) and were approved by the Ethics Committee of the Ekaterinburg Medical Research Center for Prophylaxis and Health Protection in Industrial Workers.

After a preliminary training, the rats were exposed or sham-exposed for 4 h a day, 5 times during one working week. Along with each single exposure, a sample of airborne nanoparticles was collected on an acetyl cellulose fine fiber filter attached to the inhalation setup instead of a rat's nose while monitoring the volume velocity of air drawn through the filter. Each daily filter was being sampled during 4 hr in parallel with exposure of rats. The mass of Pb retained on it was determined with an atomic absorption spectrometer, ContrAA 700 (Analytic Jena A, Jena, Germany), and translated into the mass of PbO and then into its air concentration as mg/m3, which proved to be equal to 1.30 <sup>±</sup> 0.10 mg/m3 (The OSHA Permissible Exposure Limit for Pb as well as the respective Russian national standard is 50 μg/m3 averaged over an 8-h period which corresponds to 54 μg PbO/m3. For the 4 h exposure, the equivalent level would be 108 μg/m3. We chose a one order of magnitude higher level to presumably ensure some toxicity outcomes, a short total exposure period notwithstanding).

The mass deposition of inhaled particles in the lower airways can be estimated only very roughly because this estimation is based on physiological parameters that are rather uncertain:

(1) The rat's minute respiratory ventilation as assessed experimentally by different authors varies between 78 mL [37] and 210 mL [38].

The so-called multi-path particle dosimetry (MPPD) model for rats used in [39] assumes the breathing frequency to be equal to 102 min−<sup>1</sup> and tidal volume to be 2.1 mL. It gives minute ventilation, 214 mL, which almost exactly corresponds to the above given highest experimental value and thus seems to be an extreme estimate. In our previous inhalation studies with iron oxide [6], silica [40], and nickel oxide [41] nanoparticles, we calculated particle deposition based on the minute ventilation value of 100 mL, which is close to the median of the above experimental range but of course somewhat arbitrary.

(2) The pulmonary deposition fraction of inhaled NPs may be estimated to be equal to 0.52, which is close to the lowest of the differently substantiated assessments given by [42] but much higher as compared with 0.124, which value is assumed to be the alveolar region deposition fraction for 20 nm particles according to the same MPPD model.

Thus, the pulmonary PbO-NP deposition per each single exposure may be tentatively estimated as mean NP concentration in the inhaled air, mg/m3 <sup>×</sup> minute respiratory ventilation, mL <sup>×</sup> exposure time, min <sup>×</sup> mL to m<sup>3</sup> ratio <sup>×</sup> mcg to mg ratio <sup>×</sup> deposited fraction = 1.30 <sup>×</sup> 100 <sup>×</sup> 240 <sup>×</sup> 10−<sup>6</sup> <sup>×</sup> 10<sup>3</sup> <sup>×</sup> 0.52 = 16.2 mcg.

After the end of the exposure week, the following procedures were performed for all rats:


Bronchoalveolar lavage was carried out 24 h after the last inhalation exposure. A cannula connected to a Luer's syringe containing 10 mL of normal saline was inserted into the surgically prepared trachea of a rat under hexenal anaesthesia. The fluid entered the lungs slowly under the gravity of the piston, with the animal and syringe positioned vertically. Then, the rat and the syringe were turned 180◦, and the fluid flowed back into the syringe. The extracted BALF was poured into siliconized refrigerated tubes. An aliquot sample of the BALF was drawn into a white blood cells (WBC) count pipette together with 3% acetic acid and methylene blue. Cell count was performed in a standard hemocytometer (the so-called Goryayev's Chamber). For cytological examination, the BALF was centrifuged for 4 min at 1000 rpm; then, the fluid was decanted, and the sediment was used for preparing smears on two microscope slides. After air drying, the smears were fixed with methyl alcohol and stained with azure eosin. The smears were microscoped with immersion at a magnification of

×1000. The differential count for determining the percentage of alveolar macrophages (AM), neutrophil leucocytes (NL), and other cells counted together was conducted up to a total number of 100 cells. Allowing for the number of cells in the BALF, these percentages were recalculated in terms of absolute AM and NL counts.

Seven out of 14 not rats not subjected to that lavage in both exposed and sham-exposed groups were killed by cervical dislocation under ether anesthesia. The liver, spleen, kidneys, heart, lungs, and brain were weighed. Blood was collected from the tail vein under ether anesthesia at the end of the experiment. The biochemical indices determined in the blood included total serum protein, albumin, globulin, alanine and aspartate transaminases (ALT, AST), bilirubin, catalase, glutathione, ceruloplasmin, malondialdehyde (MDA), lactate dehydrogenase (LDH), alkaline phosphatase, amilase, glucose, thiol groups (SH-groups), gamma-glutamyl transpeptidase (GGTP), urea, acid, calcium, myoglobin, troponin, natriuretic peptide, endothelin-1, and vascular endothelial growth factor (VEGF).

For determining the hemoglobin content, hematocrit, thrombocrit, mean erythrocyte volume, and for counting a red blood cells (RBC), WBC, and thrombocytes, we used a MYTHIC-18 auto-hematology analyzer (C2 Diagnostic, Montperllier, France). Reticulocyte percentage was counted on smears under optical microscopy after supravital staining with brilliant cresyl blue. The number of micronucleus in the polychromatophilic erythrocytes of the bone marrow was counted in smears under optical microscopy after fixing in methanol and staining by Pappenheim's stain.

All the clinical laboratory blood and urine tests with the exception of the specially considered ones were performed using the well-known techniques described in many manuals (e.g., [48]). Before performing euthanasia, we measured the heart rate, arterial pressure, blood flow rate, and blood volume in the rat tails using the noninvasive blood pressure system CODA-HT8 (Kent Scientific, Torrington, CT, USA) and recorded electrocardiograms (ECG) in the first and second leads with programmed analysis using the ecgTUNNEL system (emka TECHNOLOGIES, Paris, France).

In addition, the pulmonary and brain accumulation of NPs and the ultrastructure of respective tissues were visualized by means of transmission electron microscopy (TEM). To this end, pieces of an organ were fixed in 2% paraformaldehyde and 2.5% glutaraldehyde in a cacodylate buffer with 5% sucrose at pH 7.3, post-fixed in 1% osmium tetroxide, contrasted with uranyl acetate en bloc, and embedded in epoxy resin (Spurr). This sample preparation procedure was carried out in a microwave tissue processor, HISTOS REM (Milestone, Milan, Italy). Semi-thin (900 nm thick) sections of epoxy blocks were stained in toluidine blue with the addition of 1% borax and examined under the optical microscope for choosing a site for TEM. The 60 nm ultrathin sections of this site obtained with the help an ultramicrotome (Power Tome, RMC, Tucson, AZ, USA) were contrasted with uranyl acetate and lead citrate. Grid-mounted sections were investigated in an electron microscope, AURIGA (Carl Zeiss; MT, Oberkochen, Germany) in the STEM mode in the range of magnifications 1200–200,006.

#### **4. Conclusions**

The experimental results and their discussion with reference to the literature data enable us to argue that inhalation exposure to lead oxide nanoparticles associated with their retention in the organism (olfactory brain included) demonstrated with TEM, led (even though the exposure level was relatively low and was of short duration) to disturbances in the organism, some of which are specific to lead intoxication (in particular, increase in reticulocytes proportion, in δ-ALA urine excretion, and signs of arterial hypertension).

**Author Contributions:** Formal analysis, I.N.C.; Investigation, S.N.S., S.V.K., V.Y.S. and E.V.S.; Project administration, M.P.S.; Resources, V.B.G.; Visualization, I.V.Z.; Writing – original draft, B.A.K.; Writing – review & editing, L.I.P. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was funded by the budget of the Ekaterinburg Medical Research Center for Prophylaxis and Health Protection in Industrial Workers.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
