A Single Infusion of Polyethylene Glycol-Coated Superparamagnetic Magnetite Nanoparticles Alters Differently the Expressions of Genes Involved in Iron Metabolism in the Liver and Heart of Rats

This study investigated genotype- and tissue-related differences in the biodistribution of superparamagnetic magnetite (Fe3O4) nanoparticles (IONs) into the heart and liver of normotensive Wistar Kyoto (WKY) and spontaneously hypertensive (SHR) rats after a single i.v. infusion of polyethylene glycol-coated IONs (~30 nm, 1mg Fe/kg) 100 min post-infusion. The effects of IONs on the expression of selected genes involved in the regulation of iron metabolism, including Nos, Sod and Gpx4, and their possible regulation by nuclear factor (erythroid-derived 2)-like 2 (NRF2, encoded by Nfe2l2) and iron-regulatory protein (encoded by Irp1) were investigated. In addition, superoxide and nitric oxide (NO) production were determined. Results showed reduced ION incorporations into tissues of SHR compared to WKY and in the hearts compared to the livers. IONs reduced plasma corticosterone levels and NO production in the livers of SHR. Elevated superoxide production was found only in ION-treated WKY. Results also showed differences in the regulation of iron metabolism on the gene level in the heart and liver. In the hearts, gene expressions of Nos2, Nos3, Sod1, Sod2, Fpn, Tf, Dmt1 and Fth1 correlated with Irp1 but not with Nfe2l2, suggesting that their expression is regulated by mainly iron content. In the livers, expressions of Nos2, Nos3, Sod2, Gpx4, and Dmt1 correlated with Nfe2l2 but not with Irp1, suggesting a predominant effect of oxidative stress and/or NO.


Introduction
In organisms, iron absorption, storage and usage are highly regulated. Briefly, the main regulator of systemic iron homeostasis is hepcidin (HEPC, encoded by the Hamp gene)-a hormone primarily secreted by hepatocytes and, to a lesser extent, also in the heart [1] and other tissues, in response to iron loading. Hepcidin binds to ferroportin (FPN, encoded by the Fpn1 gene), the iron exporter on the cell surface. This initiates FPN internalization and degradation resulting in a reduced efflux of iron from the enterocytes, macrophages and other peripheral cells into the circulation [2,3]. In circulation, iron is transported via the circulating transporter transferrin (TF, encoded by the Tf gene). After TF binding to transferrin receptor-1 (TFR1, encoded by the Tfr1 gene) on the peripheral cells, transferrin-bound iron enters the cells in which iron accomplishes its biological functions. Non-transferrin-bound iron enters the cells via the divalent metal transporter (DMT1, In recent studies, we have shown that elevated BP was associated with alterations in the distribution of IONs compared to rats with normal blood pressure [13,32]. However, little is known about differences in iron metabolism and the incorporation of IONs into the heart and liver in hypertensive rats and whether and how IONs affect the expression of genes involved in iron metabolism, antioxidant defence and NO production in these organs after a single i.v. infusion, a condition that is relevant if IONs are used for diagnostic or therapeutic purposes in patients with various BP levels. Thus, this study investigated the influence of a single intravenous infusion of IONs in the liver and the heart in normotensive Wistar Kyoto (WKY) and spontaneously hypertensive rats (SHR). We focused on genotype-and tissue-related differences in the biodistribution of IONs, superoxide and NO productions, gene expressions of the selected genes involved in iron metabolism as well as in genes involved in NO production and maintenance of redox balance in the liver and heart. We hypothesized that chronic elevation of blood pressure (hypertension) significantly alters the biodistribution of IONs and thus alters the expression of the above-mentioned genes, with more pronounced alterations in the liver than in the heart.

Animals and Treatment
In our study, 12-16 weeks old normotensive WKY and SHR male rats were used. They were fed with pelleted diet 1320, variant P (Altromin 1314, variant P; Altromin Spezialfutter, Lage, Germany), with reduced phytoestrogen content and defined iron content (192.51 mg/kg of diet). Rats had free access to food and water. The rats were held in standard conditions (humidity 45-65%; temperature 20-22 • C) with a regular rotation of light and darkness (12/12 h). Before the experiment, all of the rats had implanted catheters in the carotid artery for blood pressure (BP) and heart rate (HR) measurements and a catheter in the jugular vein for infusion of nanoparticles dispersed in physiological saline solution or saline alone (in the control rats). The rats were cannulated under anaesthesia (2.5-3% isoflurane) by the procedure described previously [33] one day before the experiment. The rats were divided into 2 groups in each genotype: control groups to which saline alone was administered (Cont, n = 6-7) and ION treated group (ION, n = 6-7) to which a dispersion of IONs in saline was administered.
Commercially available superparamagnetic PEG-coated IONs were purchased from Merck (previously Sigma-Aldrich) Bratislava, Slovakia (cat. No. 747408, PubChem SID 329765832, accessed on 18 March 2021). Coated IONs were chosen because bare IONs were shown to be toxic due to their ability to induce haemolysis, DNA breakdown and oxidative stress [34,35]. PEG was chosen as a suitable coating as it is a hydrophilic, biocompatible and neutral polymer that enhances the dispersion of IONs in aqueous media. Furthermore, PEGylation of nanoparticles has been demonstrated to reduce immunogenicity, liver uptake, hemotoxic effects and haemolysis, as well as increase the blood circulation time of nanoparticles in vivo [36][37][38]. IONs concentration was 1 mg Fe/mL and they were dispersed in water. The manufacturer declared the size of IONs confirmed by the transmission electron microscope was 28-32 nm, the zeta potential was −12 mV, the polydispersity index was 0.1 and the hydrodynamic size was approximately 45 nm. The exact properties of IONs used in this study were described previously [39]. IONs were autoclaved at 121 • C for 30 min and dispersed in sterile saline to reach a final dose of 1 mg of Fe/kg of body weight and administered in 10 min-lasting infusions into the jugular vein of rats. At the end of the experiment, i.e., 100 min after the IONs (or saline to control rats) administration, a small amount of blood was collected from the carotid artery catheter into ethylenediaminetetraacetic acid-coated test tubes to prepare plasma for determination of plasma corticosterone. Rats were then decapitated unconsciously after short exposure to CO 2 , and trunk blood was collected into heparinized test tubes for determination of plasma iron concentration. Then the hearts and livers were dissected for further analyses.

Blood Pressure and Heart Rate Determination
Before the experiment, the conscious rats were placed into a plastic box 27 × 14 × 9 cm in size, which allowed free movement. BP and HR were recorded with a sampling rate of 1 kHz via a catheter inserted into the carotid artery using the BP-recording device PowerLab system (ADInstruments, Bella Vista, Australia). BP and HR were recorded during the entire experiment, basal mean arterial pressure (MAP) and HR were calculated as the average values of a 200 sec segment between 16 and 20 min of the recording (about 10 min before IONs application) and end MAP and HR were calculated as the average values of a 200 sec segment between 138 and 142 min of the experiment (100 min after ION application) [32].

Determination of Plasma Corticosterone and Iron Concentrations
After collection, blood was centrifuged at 4 • C, 850× g for 10 min and plasma was kept at −80 • C until further analysis. Plasma corticosterone was determined in plasma samples using the Corticosterone Rat/Mouse 125 I radioimmunoassay kit (DRG Instruments GmBH, Germany) according to the manufacturer's instructions. All samples were measured in a single assay with the intra-assay variation coefficient of 4%. The assay sensitivity was 6.3 ng/mL [32].
Iron concentrations were determined in blood plasma from heparinized trunk blood in the accredited laboratory of Laboklin s.r.o. (Bratislava, Slovakia) using the standard laboratory methods [40].

Determinations of NOS Activity and Superoxide Production
Total nitric oxide synthase (NOS) activity was determined in the 20% of tissue homogenates (w:v) of the liver and left heart ventricle (LHV) as described previously [32]. Briefly, tissues were collected into ice-cold buffer (0.05 mol/L Tris-HCl, pH 7.4, containing 1% protease inhibitor cocktail, homogenized 4 • C and centrifuged (4 • C, 10 min, 3000× g). After centrifugation, NO synthase activity was determined in the supernatants by the conversion of [ 3 H]-L-arginine to [ 3 H]-L-citrulline. NO synthase activity was expressed as pkat per gram of protein. The protein concentration was determined using the Lowry method [41].
Superoxide production was determined in approximately 15 mg tissue of the liver and LHV by determination of lucigenin-enhanced chemiluminescence as described previously [42,43]. The results were expressed as cpm/mg of tissue.

Determination of the Relative Content of Biogenic Iron and ION-Originated Iron Content in Plasma, Liver and Heart
To determine the relative content of biogenic iron and iron-containing compounds in the liver and LHV samples, we used a Quantum Design MPMS-XL 7AC (SQUID) magnetometer with a reciprocating sample operation option with a sensitivity of 10 −11 Am 2 . The advantage of this method is its ability to determine all the iron forms in a small sample of biological material with high sensitivity [44][45][46]. It also allows us to distinguish biogenic iron from that originated from IONs based on their different magnetic properties [39].
Samples of the liver and LHV were dissected using ceramic scissors and forceps, frozen immediately in liquid nitrogen and kept at −80 • C until further analysis. Before the analysis, the tissues were defrosted and cut using the cylindrical shaped instrument to obtain repeatedly the same shape of all samples with d =~4.5 mm as described previously [39]. These were vacuum dried and inserted into a plastic measuring straw. For the determination of the relative content of biogenic iron, the magnetic characteristics of the samples were measured in the form of isothermal hysteresis curves at a temperature of 2 K (−271.15 • C) and at a magnetic field up to 7 T once the saturation magnetization (M S ) was reached. M S is the parameter for determining the relative amount of magnetic compounds in the biological samples, of which iron is a dominant component [44,45,47]. Results are expressed as Ms (memu/g of dried weight). For determination of magnetic characteristics of ION-originated iron in tissues, samples were prepared as described above, but the magnetic characteristics were measured at temperature 300 K (26.85 • C) in a magnetic field up to 1 T. ION-originated iron content was calculated as described previously [32,39] and expressed as µg of iron per g of dried weight of the tissue. The magnetic properties of the tissues and iron contents were determined in 5-6 rats per group.
The total RNA of the tissue samples was isolated by using the PureZOL™ RNA Isolation Reagent (Bio-Rad, Hercules, CA, USA) according to the manufacturer's protocols. The amount of total isolated RNA was spectrophotometrically quantified at 260/280 nm and 260/230 nm by using a NanoDrop spectrophotometer (Thermo Scientific, Waltham, MA, USA). In the next step, the isolated RNA was reverse transcribed into cDNA by using Eppendorf Mastercycler (Eppendorf AG, Elbmarsch, Germany) and the reaction mixture of iScript™ cDNA Synthesis Kit (Bio-Rad, Hercules, CA, USA) according to the manufacturer's instructions. Gene amplification by using qPCR was performed on a CFX96 Real-Time PCR detection system (Bio-Rad, Hercules, CA, USA). SsoAdvanced Universal SYBR Green Supermix (Bio-Rad, Hercules, CA, USA) was used for gene amplification.
The PCR mixture contained 1.5 µL of five-fold diluted template cDNA, 10 µL SsoAdvanced mix, 1.5 µL of both forward and reverse primers (Metabion, Germany, 4 µmol/L) and 5.5 µL of nuclease-free water (Sigma-Aldrich, Germany) in a final volume of 20 µL. The thermal cycling conditions were as follows: (1) 50 • C for 1 min, (2) 95 • C for 30 sec, (3) 39 cycles consisting of (a) 95 • C for 10 sec, (b) annealing temperature (depending on the selected primer (see Table 1) for 30 sec and (4) melt curve analysis. Determination of melt curves for amplicon analyses was constructed at 59-95 • C, 5 sec/1 • C. Samples were measured using Bio-Rad CFX Manager software (version 2.0). Expression of studied genes was expressed as the ratio of gene expression with respect to β-actin levels (Actb). Gene-specific primers were designed using the PubMed database (Gene) and program (Primer-BLAST). The following pairs of primer sequences, melting temperatures, accession number and amplicon size of studied genes are described in Table 1. The plates with samples were measured with CFX96 Touch™ Real Time-PCR Detection System (Bio-Rad, Hercules, CA, USA); the program started with heating up the plate to 50 • C (2 min), then increasing the temperature to 95 • C (30 s). The program continued with 39 cycles repeating the steps of denaturation (95 • C, 10 s) and annealing (30 s). The fluorescent dye SYBR Green (Bio-Rad, Hercules, CA, USA) was used for the detection of PCR products. The signal intensity during amplification was detected with Bio-Rad CFX Manager software. The gene expression is presented as the ratio of gene expression to Actb.

MAP, HR and Plasma Corticosterone Levels
The basal MAP of WKY and SHR rats were 130 ± 4 and 178 ± 3 mmHg, respectively (genotype: F (1,21) = 136.2, p < 0.0001). The MAP at the end of the experiment (i.e., 100 min post-infusion) did not change significantly in any group of rats ( Figure 1a).

Nitric Oxide Synthase Activities and Superoxide Productions
Regarding NOS activity, two-way ANOVA revealed significant interaction of genotype and treatment in the liver (F(1,21) = 5.

Nitric Oxide Synthase Activities and Superoxide Productions
Regarding NOS activity, two-way ANOVA revealed significant interaction of genotype and treatment in the liver (F (1,21)

Relative Contents of Biogenic Iron and ION-Originated Iron
Saturation magnetization Ms (relative parameter of biogenic iron content) of the liver of control SHR was significantly higher than in WKY (p < 0.001, Figure 3a). In contrast, the saturation magnetization of the LHV of SHR was reduced compared to WKY (p < 0.001, Figure 3b). ION-originated Fe content was significantly lower in the liver (p < 0.02) and  For superoxide productions, there were significant interactions of genotype and treatment in both the liver (F (1,20) = 4.9, p < 0.04) and the LHV (F (1,20) = 4.7, p < 0.05) with significantly higher levels found in ION-treated WKY compared to control WKY (Figure 2c,d).
Higher superoxide production was found in the liver of control SHR vs. control WKY (Figure 2c). However, no changes in superoxide productions were found in ION-treated SHR vs. control SHR in both the liver and LHV (Figure 2c,d).

Relative Contents of Biogenic Iron and ION-Originated Iron
Saturation magnetization Ms (relative parameter of biogenic iron content) of the liver of control SHR was significantly higher than in WKY (p < 0.001, Figure 3a). In contrast, the saturation magnetization of the LHV of SHR was reduced compared to WKY (p < 0.001, Figure 3b). ION-originated Fe content was significantly lower in the liver (p < 0.02) and the decreasing tendency was also in the LHV (p = 0.06, n = 10) of SHR compared to that in the respective WKY value (Figure 3c,d).

Expression of Genes Involved in Iron Metabolism, Antioxidant Defence and Nitric Oxide Production
Relative mRNA expressions of Nfe2l2, Pparg, Nos2 and Nos3 in the liver and LHV are shown in Figure 4. There were significant main effects of genotype for Nfe2l2 (F(1,20) Figure 5b). In the LHV, significant interactions of genotype and treatment were found only for Sod1 (F (1,20) = 7.20, p < 0.02, Figure 5d); no effects were found for Sod2 (Figure 5e). The mRNA expressions of Gpx4 were unchanged in the liver (Figure 5c), but in the LHV, significant effects of genotype (F (1,20) = 14.71, p < 0.01) and treatment F (1,20) = 5.7, p < 0.03, Figure 5f) were found, with lower levels in SHR vs. WKY and for ION-treated rats vs. controls. In Figure 6, relative mRNA expressions of genes involved in iron metabolism regulation or iron carrying and storage are depicted. A significant effect of ION treatment was found only for Hamp (F(1,20) = 6.74, p < 0.02) in the liver with higher levels in ION-treated rats compared to controls (Figure 6a). Significant interactions of genotype and treatment in the liver were found only for Fth1 (F(1,20) = 9.06, p < 0.01, Figure 6d); there was a significant decrease in Fth1 mRNA in ION-treated SHR vs. control SHR. In the LHV, there was a significant reduction in Tf gene expression in SHR vs. WKY (effect of genotype, F(1,20) = 4.58, p < 0.05, Figure 6g). No differences were found in the gene expression of Irp1 and Tf in the liver (Figure 6b,c) and in the Hamp, Irp1 and Fth1 in the LHV (Figure 6e,f,h). In the liver, Hamp gene expression correlated positively with plasma iron levels (r = 0.58, p < In Figure 6, relative mRNA expressions of genes involved in iron metabolism regulation or iron carrying and storage are depicted. A significant effect of ION treatment was found only for Hamp (F (1,20) = 6.74, p < 0.02) in the liver with higher levels in ION-treated rats compared to controls (Figure 6a). Significant interactions of genotype and treatment in the liver were found only for Fth1 (F (1,20) = 9.06, p < 0.01, Figure 6d); there was a significant decrease in Fth1 mRNA in ION-treated SHR vs. control SHR. In the LHV, there was a significant reduction in Tf gene expression in SHR vs. WKY (effect of genotype, F (1,20) = 4.58, p < 0.05, Figure 6g). No differences were found in the gene expression of Irp1 and Tf in the liver (Figure 6b,c) and in the Hamp, Irp1 and Fth1 in the LHV (Figure 6e,f,h). In the liver, Hamp gene expression correlated positively with plasma iron levels (r = 0.58, p < 0.006, n = 21), which was not found in the LHV. In Figure 7, relative mRNA expressions of genes involved in iron transport are depicted. No genotype and/or treatment-related changes in Tfr1, Dmt1 and Fpn1 gene expressions were found in the liver (Figure 7 a-c). In the LHV, a significant effect of genotype (F(1,20) = 30.02, p < 0.0001), treatment (F(1,20) = 15.27, p < 0.001) and interactions of genotype and treatment were found for Tfr1 (F(1,20) = 757, p < 0.02, Figure 7d). A significant effect of treatment was found in the LHV for Dmt1 (F(1,20) = 5.03, p < 0.04, Figure 7e). No changes were found in the gene expression of Fpn1 in the LHV (Figure 7f). The correlations among In Figure 7, relative mRNA expressions of genes involved in iron transport are depicted. No genotype and/or treatment-related changes in Tfr1, Dmt1 and Fpn1 gene expressions were found in the liver (Figure 7 a-c). In the LHV, a significant effect of genotype (F (1,20) = 30.02, p < 0.0001), treatment (F (1,20) = 15.27, p < 0.001) and interactions of genotype and treatment were found for Tfr1 (F (1,20) = 757, p < 0.02, Figure 7d). A significant effect of treatment was found in the LHV for Dmt1 (F (1,20) = 5.03, p < 0.04, Figure 7e). No changes were found in the gene expression of Fpn1 in the LHV (Figure 7f). The correlations among the genes are in the Supplementary Table S1.

Discussion
Superparamagnetic iron oxide nanoparticles of γ-Fe2O3 (maghemite) and Fe3O4 (magnetite) can be used in various biomedical applications. The advantage of such NPs lies in the possibility to use them for targeted drug delivery in the presence of the magnetic field. However, it is also very important to know the pharmacokinetics of the NPs and their effect on the living organism because iron is an essential biogenic element. In our previous studies, we confirmed the superparamagnetism of IONs used in this study [39] as well as their presence in different tissues depending on the blood pressure [13,32,39,40].
The main findings of this study are that ION-originated iron was found in the livers and hearts of normotensive and hypertensive rats, but considerably less in the tissues of hypertensive SHR than normotensive WKY. ION-originated iron content found in the hearts was roughly around 20% of that determined in the liver of the given genotype. In addition, our study showed considerable tissue differences in the expression of genes involved in iron metabolism, as well as in Sod and Nos expression, in the livers and hearts. Regarding genotype differences, we found that elevated ION content in the tissues of WKY led to elevated superoxide production, which was not seen in SHR. Lower ION con-

Discussion
Superparamagnetic iron oxide nanoparticles of γ-Fe 2 O 3 (maghemite) and Fe 3 O 4 (magnetite) can be used in various biomedical applications. The advantage of such NPs lies in the possibility to use them for targeted drug delivery in the presence of the magnetic field. However, it is also very important to know the pharmacokinetics of the NPs and their effect on the living organism because iron is an essential biogenic element. In our previous studies, we confirmed the superparamagnetism of IONs used in this study [39] as well as their presence in different tissues depending on the blood pressure [13,32,39,40].
The main findings of this study are that ION-originated iron was found in the livers and hearts of normotensive and hypertensive rats, but considerably less in the tissues of hypertensive SHR than normotensive WKY. ION-originated iron content found in the hearts was roughly around 20% of that determined in the liver of the given genotype. In addition, our study showed considerable tissue differences in the expression of genes involved in iron metabolism, as well as in Sod and Nos expression, in the livers and hearts. Regarding genotype differences, we found that elevated ION content in the tissues of WKY led to elevated superoxide production, which was not seen in SHR. Lower ION content in the tissues of SHR was associated with reduced plasma corticosterone, reduced total NOS activity in the liver as well as with reduced liver expressions of Sod1, Sod2 and Fth1. On the other hand, in WKY, Sod2 was elevated in the liver and Sod1 in the heart in ION-treated rats. Despite these genotype-and tissue-related differences, positive correlations were found between plasma iron concentration and hepatic Hamp expression as well as between IONoriginated iron content and Fth1 gene expressions in both tissues investigated, suggesting the considerable influence of ION-derived iron on intracellular iron metabolism in these tissues when all rats were taken into account.
Previously we showed that hypertension (i.e., chronic high BP) was associated with an increased incorporation of IONs into the liver and vascular wall of SHR when determined 24 h after the second ION infusion at the dose of 2 mg Fe/kg/day [13]. On the other hand, an acute stress-induced increase in BP in WKY rats treated with a single infusion of IONs (1 mg Fe/kg) led to a decrease in ION-derived iron in circulation and in the liver when determined 100 min post-ION infusion [32]. Those studies showed that high BP, either chronic or acute, is an important factor affecting ION deposition in the liver. In this study, we focused on the heart and liver after a single ION infusion in normotensive and hypertensive rats, which is relevant to the situation of patients with various levels of BP.
We found that a single ION infusion in rest conditions did not induce significant changes in BP and HR in SHR despite an ION-induced decrease in NO production in the liver of SHR (vs. control SHR) and elevated superoxide production in both tissues investigated in WKY. Unchanged BP after ION administration suggests that mild changes in the above-mentioned parameters as well as the increase in overall hepatic iron content due to the incorporation of ION-derived iron had no effect on BP regulation in any rat strain. This finding corresponds with the results of Seravalle et al., who also found no changes in BP in hypertensive patients with hepatic iron overload compared to hypertensive individuals [9]. Surprisingly, we found a significant decrease in corticosterone levels after ION infusion in SHR, suggesting that IONs can interfere with hypothalamic-pituitary-adrenal axis function, specifically under conditions of hypertension, though the exact mechanisms need to be further studied. In the present study, we did not determine iron content in the adrenal glands, but there are reports in humans showing enhanced acute accumulation of iron nanoparticles within the adrenals that was comparable to that in the tissues of the reticuloendothelial system, such as the liver and spleen [48]. In spite of missing data in hypertensive patients, our study demonstrates that the adrenals of hypertensive individuals may display increased sensitivity to ION administration.
Our study also revealed that plasma iron concentrations were elevated in both rat genotypes after ION infusion. However, plasma iron levels do not reflect ION-originated iron content in the tissues. Using SQUID magnetometry, we found elevated M S , an indirect but very sensitive parameter, reflecting total iron content [39] in the liver of control SHR compared to control WKY. Despite that the liver is considered a rich source of iron in the organism, high levels of iron are also in the heart, in which iron metabolism has to be regulated very precisely to prevent cardiac dysfunction [49]. In addition, SQUID magnetometry allows us to determine the amount of ION-originated iron, which is not possible to distinguish from biogenic iron present in the tissues by biochemical or histological methods [39]. We found that, in conditions of this study, the content of ION-originated iron was lower in the hearts than in the livers and also in tissues of ION-treated SHR compared to the ION-treated WKY. In fact, in three of five hearts of ION-treated SHR we did not find ION-originated iron. However, in both tissues, positive correlations were found between ION-originated iron and Fth1 gene expressions, suggesting the significant influence of ION-originated iron on innate iron metabolism.
This effect was also confirmed by elevated Hamp expression in the liver of ION-treated rats and by a positive correlation between plasma iron and Hamp gene expression in the liver. It is known that the hepatic release of hepcidin is regulated predominantly by intrahepatic iron concentration [50]. However, hepcidin can also be synthesized in smaller amounts by the other tissues, including the heart [51] in which it participates in the regulation of local iron levels [52]. Recent studies show that hepcidin has an irreplaceable role in regulating iron homeostasis in the heart; hepcidin deficiency results in cardiac dysfunction, left ventricular hypertrophy and apoptosis [53]. In our study, no changes in cardiac Hamp expression were found, suggesting that the heart functions should not be considerably altered by ION-originated iron after a single low-dose ION infusion. However, IONs and/or iron released after nanoparticle disintegration can enter redox processes and induce oxidative stress [54,55]. Here we found elevated superoxide production only in the tissues of WKY, which is in agreement with the higher ION-originated iron content found in the tissues of WKY. In addition, Sod2 was elevated in the liver and Sod1 in the heart of WKY, while both Sod1 and Sod2 were significantly reduced in the liver of ION-treated SHR vs. control SHR. Despite our expectation, no corresponding changes (regarding Sod expressions) were found in the expression of the Nfe2l2 gene in both tissues investigated. On the other hand, we found positive correlations between Nfe2l2 and Nos2, Nos3, Dmt1, Gpx4, Sod2 and Pparg in the liver. No such correlations with Nfe2l2 were found in the heart, except for that with Pparg, which correlated positively with Nfe2l2 in both tissues investigated. This study also showed two interesting relations among the gene expressions in the liver and heart when all rats were taken into account. Firstly, Of the 24 correlations found among the genes investigated in this study in the liver, only 1 was negative. In the heart, of the 39 correlations found among the genes investigated, 17 were negative and 22 were positive. Secondly, all correlations with Dmt1 or Fth1 were positive in the liver and only two genes (Fpn1 and Tfr1) correlated positively with Irp1. In contrast, in the heart, all genes which correlated positively with Fth1 or Irp1, correlated negatively with Dmt1. These findings suggest considerable differences in the expression of genes involved in iron metabolism in the liver and heart. Since the optimal iron concentration in the heart is critical for normal cardiac function [49], we assume that the sensitivity of cardiomyocytes to iron is higher compared to that in the liver. Rapid, but not necessarily big, changes in the expression of various genes and corresponding proteins can maintain optimal intracellular iron concentrations in the heart, specifically under oxidative and/or nitrosative stress, as both ROS and NO participate in the regulation of iron metabolism [56,57]. In the heart, we also found a negative correlation between the Fth1 and Dmt1 genes, supposedly under the control of Irp1, which suggests the finely tuned regulation of intracellular iron concentration. In addition, opposite correlations of the Fth1 and Dmt1 genes with other genes mentioned in the supplementary Table S1, especially with Nos and Sod genes, can result in the increase in antioxidant defence and reduction of NO production. NO, in addition to its role in cardiovascular functions [58,59], has an important role in iron metabolism by modulation of cytosolic aconitase (ACO1)/IRP1 function [60]. In conditions of low levels of intracellular iron, the apo form of ACO1 has the function of IRP1, while at high iron levels, its holo form containing the [4Fe-4S] 2+ cluster possesses ACO1 enzymatic activity [61]. NO can promote the release of iron from the Fe-S centres of ACO1 and thus induce the IRP1 function of this bifunctional protein [61]. High levels of IRP1 protein support an increase of Fth1 mRNA translation resulting in elevated iron storage capacity in ferritin cores. In addition, high levels of IRP1 decrease Dmt1 and Tfr1 mRNA stability, which leads to a reduced influx of iron into the cell [62]. Surprisingly, in this study, there were no correlations with the Nfel2l and Pparg genes and other genes investigated in the heart, which is in contrast to the liver. In the liver, the antioxidant genes (Sod2, Gpx4) and Nos2 and Nos3 genes correlated positively with Nfel2l, in agreement with previous findings of elevated NRF2 translocation into the nucleus in the presence of ROS and NO [20,21]. Interestingly, none of the genes investigated in the liver correlated with hepatic Hamp and only two genes (Fpn1, Tfr1) correlated with Irp1 expression. Collectively, the results suggest that while changes observed on the gene levels in the heart seem to be related to changes in iron load, changes in the liver seem to be more related to oxidative stress and/or NO, which may be induced by altered hepatic iron content. Indeed, superoxide production was elevated in the liver of ION-treated WKY as well as in both SHR groups compared to control WKY, which was not found in the heart. Thus, our data showed considerable differences in the expression of genes involved in iron metabolism in the heart and the liver of rats. In agreement, a recent study by Kasztura et al. [63] has shown that the mechanisms of iron uptake, storage and clearance differ between the liver and heart in the porcine model of heart failure progression.

Conclusions
Our results showed that a single administration of IONs intravenously at a relatively low dose was found in the liver and the heart of both rat genotypes 100 min after their infusion. Lower content of ION-derived iron was found in the tissues of SHR compared to WKY as well as in the hearts compared to livers, confirming the hypothesis that high blood pressure significantly alters the biodistribution of IONs with less incorporation into the heart than into the liver. In addition, our results showed differences in the regulation of iron metabolism on gene levels in the heart and liver. In the heart, expressions of various genes involved in iron metabolism, including Sod and Nos, seem to be mediated predominantly by iron content and regulated by IRP1-mediated mechanisms. In the liver, the expression of genes correlated with Nfe2l2, suggesting a predominant effect NRF2-mediated mechanisms in the regulation of iron metabolism due to an altered redox balance and NO bioavailability.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.