Oxidative-damage effect of Fe₃O₄ nanoparticles on mouse hepatic and brain cells in vivo

Abstract

To assess the biological safety of Fe₃O₄ nanoparticles (NPs), the oxidative-damage effect of these NPs was studied. Twenty-five Kunming mice were exposed to Fe₃O₄ NPs by intraperitoneal injection daily for 1 week at doses of 0, 10, 20, and 40 mg·kg^-1. Five Kunming mice were also injected with 40 mg·kg^-1 ordinary Fe₃O₄ particles under the same physiological conditions. Biomarkers of reactive oxygen species (ROS), glutathione (GSH), and malondialdehyde (MDA) in the hepatic and brain tissues were detected. Results showed that no significant difference in oxidative damage existed at concentrations lower than 10 mg·kg^-1 for NPs compared with the control group. Fe₃O₄ NP concentration had obvious dose–effect relationships (P<0.05 or P<0.01) with ROS level, GSH content, and MDA content in mouse hepatic and brain tissues at>20 mg·kg^-1 concentrations. To some extent, ordinary Fe₃O₄ particles with 40 mg·kg^-1 concentration also affected hepatic and brain tissues in mice. The biological effect was similar to Fe₃O₄ NPs at 10mg·kg^-1 concentration. Thus, Fe₃O₄ NPs had significant damage effects on the antioxidant defense system in the hepatic and brain tissues of mice, whereas ordinary Fe₃O₄ had less influence than Fe₃O₄ NPs at the same concentration.

Keywords

Fe₃O₄ nanoparticle (NP) / ordinary Fe₃O₄ particle / oxidative damage / reactive oxygen species (ROS) / glutathione (GSH) / malondialdehyde (MDA)

Introduction

Nanoparticles (NPs) are special materials with two or three dimensions ranging from 1 nm to 100 nm. Given the size, chemical composition, and shape of NPs, they have unusual physical and chemical characteristics (Fadeel and Garcia-Bennett, 2010) and are thus used in various fields. NPs can be used in drug delivery, gene delivery, cancer therapy, disease detection, biochemical sensing, bioimaging (Lippacher et al., 2001; Bystrzejewski et al., 2007; Nakamura et al., 2010; Pissuwan et al., 2011), and so on. When NPs are used for imaging or drug delivery, they can be coated with macromolecules such as DNA, proteins, or monoclonal antibodies to attach interested cells. In the progress of the interaction between NPs and cells, ensuring the safety of an organism is very important (Lewinski et al., 2008).

Adverse effects of NPs on organisms have been reported, including damage to aquatic microorganisms caused by boron NPs (Strigul et al., 2009), damage to human lung epithelial cells caused by ZnO and TiO₂ NPs (Hsiao and Huang, 2011), oxidative stress induced by NPs in human lung carcinoma cells (Valodkar et al., 2011), and so on. Four factors affect NP toxicity, i.e., chemical composition, small size, large surface-to-volume ratio, and surface property.

NPs are often classified as organic NPs such as fullerenes and inorganic NPs, which consist of metals (e.g., gold and silver) and metal oxides (e.g., iron oxide, titanium dioxide, silicon dioxide, etc). Given the magnetic, mechanical, and optical properties of iron oxide NPs, they have extensive applications. On account of the magnetic properties of iron oxide NPs, they can be used to conduct tumor hyperthermia treatment (Xie et al., 2009). The magnetic properties of iron oxide NPs are also applied in the super paramagnetic iron oxide particle imaging instrument for magnetic resonance imaging that can be used to study cell tracking (Fleige et al., 2002). Iron oxide NPs can also be applied in drug and gene delivery. A doxorubicin-loaded system based on dextran cross-linked iron oxide NPs has been synthesized, and results show that the use of such NPs in drug delivery is possible (Yu et al., 2008). Fe₃O₄ NPs are a type of iron oxide NPs widely used in many fields, including printing ink manufacture, metal ion removal, magnetite filtration, biomolecule separation, and nuclear magnetic resonance imaging. The unique magnetic properties of Fe₃O₄ NPs have been largely exploited. Given the simplicity of their synthesis and partial solubility in acidic media compared with other nanoparticles like TiO₂ and SiO₂, Fe₃O₄ NPs have become a research hotspot (Ma et al., 2012). However, data on Fe₃O₄ NPs safety are lacking. Inconsistent results have increased confusion regarding this topic. Thus, the safety of Fe₃O₄ NPs must be studied.

Reactive oxygen species (ROS) can be classified into two different types, i.e., radical ROS (superoxide anion and hydroxyl radical) and non-radical ROS (hydrogen peroxide and hypochlorous acid). Considering the existence of active oxygen functional groups, ROS have very high activity in the body. Radicals emerge during the normal metabolic processes of humans. Some mechanisms can clear radicals, such as glutathione (GSH) that can be used to clear ROS. Under normal conditions, a balance exists between the generation and clearance of radicals that does not negatively affect the body (Xia et al., 2008). However, if the amount of ROS produced is too much such that it cannot be handled by normal removal mechanisms such as GSH, ROS accumulate and can damage the structure of cells, thereby affecting their functions. The strong oxidability of ROS can oxidate biomembranes, and the end-product of lipid oxidation is malondialdehyde (MDA). Thus, ROS, GSH, and MDA are usually used as biomarkers of oxidative damage to measure the toxicity of a material.

Materials and methods

Reagents and apparatus

Fe₃O₄ NPs (Sigma, St Louis, MO, USA), ordinary Fe₃O₄ particles, 2’7’-dichlorodihydrofluorescein (DCFH-DA; Calbiochem, La Jolla, CA, USA), 5,5′-dithiobis-(2-nitrobenzoic acid) (DTNB; Sigma), and 2-thiobarbituric acid (TBA; Shanghai Reagent Factory, Shanghai, China) were used in experiments. All other chemicals used were analytical grade.

A low-temperature refrigerated centrifuge (Eppendorf-5417R), a continuous-wave and fluorescent microplate spectrophotometer (Bio-teck FLX800, USA), and a fluorescence spectrophotometer (Hitachi F-4500, Japan) were used in the experiment.

Experimental animals

Thirty SPF Kunming male mice (weight ≈ 24-28 g) were purchased from the Experimental Animal Center of Hubei Province (Wuhan, China). The mice were randomly grouped into 5 treatment groups with 6 mice per group. Three groups were exposed to an Fe₃O₄ nanoparticle environment (exposure doses= 10, 20, and 40 mg·kg^-1). One experiment group was exposed to ordinary Fe₃O₄ particles at a dose of 40 mg·kg^-1. The control group had PBS taking the place of oxide metal material. The experimented mice were housed in different cages at room temperature and fed with normal diet and water. Fe₃O₄ NPs was injected into mice through intraperitoneal injection. The volume of NPs injected to mice was based on their weights at a proportion of 1% mL·g^-1. Mice were given daily intraperitoneal injections of NPs for 7 days continuously. The time of intraperitoneal injections was limited between 12:00 and 13:00βh every day.

Preparation and characterization of Fe₃O₄ NPs

Fe₃O₄ NPs (diameter<50 nm) were suspended in PBS (pH 7.5) to create solutions with concentration of 10, 20, and 40 mg·mL^-1 as required. Normal Fe₃O₄ particles (microscale diameter) were suspended in PBS (pH 7.5) to obtain a solution with 40 mg·mL^-1 concentration. A sonication time of 20 min was used only before each injection to make the solution homogeneous. The diameters of NPs and the crystal appearance were determined by scanning electron microscopy (SEM).

Sample preparation

The mice were sacrificed at noontime on the eighth day. Their brains and livers were quickly separated and rinsed with ice-cold PBS (0.01 mol·L^-1, pH 7.5). All livers and brains were quickly dried, weighed and homogenized with 10 mL·g^-1 ice-cold PBS (pH 7.5) using a homogenizer (about 0.3 g of tissue) on ice. The supernatants were then collected by centrifuging the homogenate at 10⁴β×βg and 4β°C for 10 min, and this supernatant was used to measure ROS, GSH, and MDA.

ROS assay

About 100 μL of the supernatant diluted 40× with PBS (pH 7.5) was transferred to a 96-well microplate. Then, 100 μL of 5 μmol·L^-1 DCFH-DA was immediately added to the microplate, which was slightly shaken and kept from light for 10 min. The fluorescence of dichlorofluoroscein (DCF), the oxidized product of DCFH-DA that can indicate the level of the ROS in a cell, was measured using a microplate spectrofluorometer with an excitation of 485 nm and emission of 525 nm (Elbekai and El-Kadi, 2005).

GSH assay

GSH is a thiol that is the major cleaner of ROS in organisms. GSH contains sulfhydryl that can react with DTNB to produce yellow compounds in the dark (Anderson, 1985). Measurement of the absorbance of yellow compounds (TNB) within the wavelength range of 405 and 414 nm can enable the accurate estimation of GSH in a sample, thereby avoiding disturbance from proteins with thiols. The supernatant was added to the organic solvent (1:3 butyl alcohol: trichloromethane) to precipitate the proteins. After mixing, centrifugation was performed at 10⁴ rpm for 5 min. About 50 μL of the supernatant was transferred to a 96-well microplate, quickly added with 150 μL of 60 μg/mL DTNB solution, and kept in darkness for 5 min. The absorbance value of TNB was analyzed at a wavelength of 412 nm. Based on the standard curve, the GSH concentration was calculated.

MDA assay

At high temperatures (90–100β°C), the reaction of MDA and TBA results in an MDA-TBA adduct and can be measured colorimetrically at 530-540 nm under acidic conditions. Thus, 0.4 mL of tissue supernatant was mixed with 0.6% TBA solution (dissolved in 1βmol/L NaOH and added with 10% TCA) in a centrifugal tube. After a 15 min bath in boiling water, cooling on ice, and centrifuged at 10⁴ × g for 5 min, supernatants were collected for absorbance measurement at 450, 532, and 600 nm. The MDA concentration was calculated according to the formula C (μmol/L) = 6.45 (A₅₃₂ – A₆₀₀) – 0.56A₄₅₀.

Statistical analysis

All data shown in this paper are expressed as the mean±standard error of the mean and analyzed and plotted using Origin 6.0 software. Both one-way ANOVA and Tukey’s test were applied to evaluate significant differences between groups. p<0.05 was considered to be statistically significant, and p<0.01 was considered to be very significant.

Results

After 7 days of continuous intraperitoneal injection of the different concentrations of Fe₃O₄ NPs and normal Fe₃O₄ particles in the mice for 1 week, the characteristics of Fe₃O₄ NPs, ROS levels, GSH content, MDA quantity, and DNA–protein cross-links coefficients of mouse hepatic and brain cells were determined. Results are shown in Figs. 1 to 6.

To characterize the appearance of Fe₃O₄ NPs used in the experiment, the crude particle appearance and dispersive diameter were detected. As shown in the SEM image in Figure 1, the particles were smoothly spherical. After ultrasonication dispersal, the distribution of particle diameters was observed, as shown in Fig. 2. A normal distribution curve with an axis of 35 nm was obtained, and NPs with diameters smaller than 50 nm were>95%.

Characterization of Fe₃O₄ NPs

Generation of ROS

Compared with the blank control, ROS generation both in brains and livers increased with increased Fe₃O₄ NP concentration, and the ROS level significantly increased at 20 mg·kg^-1·d^-1 and 40βmg·kg^-1·d^-1 Fe₃O₄ NPs (p<0.01). Normal Fe₃O₄ particles at 40 mg·kg^-1·d^-1 concentration only led to a significance change in the ROS level of hepatic cells (p<0.05), whereas the ROS level of brain issue did not significantly increase compared with the PBS control group. Fe₃O₄ NPs obviously affected the hepatic cells.

GSH content

The GSH content at different concentrations of Fe₃O₄ NPs and normal Fe₃O₄ particles are shown in Fig. 4. In livers, the decrease in GSH level was significant at 20 mg·kg^-1·d^-1 Fe₃O₄ NPs and 40 mg·kg^-1·d^-1 normal Fe₃O₄ particles (p<0.05). The 40 mg·kg^-1·d^-1 Fe₃O₄ NP group was associated with a very significant decrease in GSH content (p<0.01). The low-concentration Fe₃O₄ NP group did not show significant difference. The trends were similar in the brain, but the 20 mg·kg^-1·d^-1 Fe₃O₄ NPs showed a more significant decrease in GSH content (p<0.01)

MDA concentration

Figure 5 demonstrates that the MDA content increased with increased concentration of Fe₃O₄ NPs in both livers and brain. When the concentration increased to 40 mg·kg^-1·d^-1, a very significant effect (p<0.01) was observed both in livers and brain. Compared with the blank control group, only a slight effect was found upon exposure to normal Fe₃O₄ particles in both livers and brain. At the same concentration, MDA increased more in Fe₃O₄ NPs than in normal Fe₃O₄ particles.

Discussion

The chemical composition and morphology of a material crucially affect its chemical and physical properties. Morphology includes geometric shape, granularity, and particle distribution. In this study, the effects of normal Fe₃O₄ particles and Fe₃O₄ NPs on mouse liver and brain were compared in terms of their composition and particle size.

Particle size is a very important factor affecting NP cytotoxicity. No effect was detected at 100 mg·kg^-1 with≥300 nm nano-SiO₂, whereas liver damage already occurs at 30 mg·kg^-1 with smaller-sized nano-SiO₂ (Nishimori et al., 2009). Single-walled carbon nanotubes are more toxic than multi-walled carbon nanotubes (Jia et al., 2005). A study on mice contaminated with different sizes of Fe₂O₃ particles shows that nano- Fe₂O₃ can markedly decrease pathological response (Zhu et al., 2008). In the present study, ordinary Fe₃O₄ had less influence than Fe₃O₄ NPs at the same concentration. Thus, the size of Fe₃O₄ particles is critical to the toxicity of such particles.

Many reports have shown that NPs can give rise to stress reaction and lead to oxidative damage (Kim et al., 2009; Tedesco et al., 2010; Piao et al., 2011; Novotna et al., 2012). In the current work, we determined the toxicity of a material by detecting oxidative damage in mouse liver and brain after exposure to NPs. Figure 3 shows that in contrast to the PBS control group, increased Fe₃O₄ NP concentration showed obviously increased ROS level. Moreover, the ROS level and NP concentration had an obvious dose–effect relationship and the increasing trend in the liver was more obvious. One explanation for the increase in ROS was that after the Fe₃O₄ NPs entered the body through intraperitoneal injection, the NPs were regarded as exogenous substances and induced some reactions in the body. Consequently, the balance in the body was broken, so the production of ROS exceeded the clearance ability of the body and ROS began to accumulate. NPs can pass organ barriers such as the blood–brain barrier (Borm and Kreyling, 2004), so Fe₃O₄ NPs can lead to more serious oxidative stress than normal Fe₃O₄ particles in the brain. Figure 3 shows that the ROS level at 20 mg·kg^-1·d^-1 Fe₃O₄ NPs obviously overtopped the ROS level at 40 mg·kg^-1·d^-1 normal Fe₃O₄ particles in the brain. The former can induce a very significant change (p<0.01) compared with the PBS control group, but the latter only slightly changed compared with the PBS control group. Thus, Fe₃O₄ NPs can easily accumulate in the brain and induce more substantial damage than normal Fe₃O₄ particles. MDA content is a secondary indicator of oxidative damage; when ROS damages lipids, MDA increases. Similar to ROS, obvious dose–effect relationships (p<0.05 or p<0.01) were observed between Fe₃O₄ NP concentrations and MDA content. Figure 5 shows an obvious increase in MDA content at 40 mg·kg^-1·d^-1 Fe₃O₄ NPs in the brain. By contrast, normal Fe₃O₄ particles showed only slight changes in MDA content in the brain at the same concentration. The results can also lead to the same aforementioned conclusion, i.e., NPs accumulated to a degree that can pass the blood–brain barrier to inflict damage to the brain. The reduction in GSH showed that the proteins fighting oxidative stress decreased because of the lack of proteins that can clear ROS, and protein dysfunction may have occurred. Figure 4 shows that different from the effect on the liver, the particle size had greater effects on brain than liver. In the brain, 20 mg·kg^-1·d^-1 Fe₃O₄ NPs can induce a very significant change (p<0.01), but 40 mg·kg^-1·d^-1 normal Fe₃O₄ particles showed a significant change (p<0.01). These results demonstrated that the Fe₃O₄ NPs definitely induced more damage than normal Fe₃O₄ particles, especially in the brain. The low concentration of Fe₃O₄ NPs (20 mg·kg^-1·d^-1) can induce only limited ROS including superoxide anion, hydrogen peroxide, and hydroxyl radical. Correspondingly, GSH excited in the cytoplasm can clear the excess ROS.

Results showed that at 20 mg·kg^-1·d^-1 NPs, the ROS level in the liver showed a very significant difference from the PBS control group (p<0.01). However, GSH in the liver only showed a significant difference (p<0.05), and MDA in the liver did not show an obvious difference from the PBS control group (p>0.05). When foreign substances invade the body, the first reaction is the accumulation of ROS, so the change in ROS is the most obvious and the significant difference emerges first. Then, the accumulation of ROS can change the GSH and MDA contents, so ROS level is the most sensitive indicator of oxidative damage among the three. With increased concentration, the significance of GSH and MDA can be gradually observed. GSH comparison between liver and brain at 20 mg·kg^-1·d^-1 Fe₃O₄ NPs exceeded our expectation. The reason may be that the brain is more sensitive to changes in the internal environment than the liver. Thus, under the condition of low toxicity, the brain can produce a stronger signal.

Conclusions

The cytotoxicity of Fe₃O₄ on mouse hepatic and brain cells in vivo was studied. The following conclusions were drawn after analyzing the experimental results. First, results of ROS, GSH, and MDA demonstrated obvious dose–effect relationships (p<0.05 or p<0.01) between Fe₃O₄ NP concentrations and ROS level, GSH content, and MDA content in mouse hepatic and brain tissues. When the concentration was 10 mg·kg^-1·d^-1, Fe₃O₄ NPs also exerted some detectable effects. Second, among the three biomarkers, ROS was the most sensitive indicator and oxidative damage was first presented by ROS. Third and last, ordinary Fe₃O₄ had less influence than Fe₃O₄ NPs at the same concentration, so the particle size was very closely related to the toxicity of particles.

References

[1] Anderson M E (1985). Determination of glutathione and glutathione disulfide in biological samples. Methods Enzymol, 113: 548–555

[2] Borm P J, Kreyling W (2004). Toxicological hazards of inhaled nanoparticles—potential implications for drug delivery. J Nanosci Nanotechnol, 4(5): 521–531

[3] Bystrzejewski M, Cudziło S, Huczko A, Lange H, Soucy G, Cota-Sanchez G, Kaszuwara W (2007). Carbon encapsulated magnetic nanoparticles for biomedical applications: thermal stability studies. Biomol Eng, 24(5): 555–558

[4] Elbekai R H, El-Kadi A O (2005). The role of oxidative stress in the modulation of aryl hydrocarbon receptor-regulated genes by As³⁺, Cd²⁺, and Cr⁶⁺. Free Radic Biol Med, 39(11): 1499–1511

[5] Fadeel B, Garcia-Bennett A E (2010). Better safe than sorry: Understanding the toxicological properties of inorganic nanoparticles manufactured for biomedical applications. Adv Drug Deliv Rev, 62(3): 362–374

[6] Fleige G, Seeberger F, Laux D, Kresse M, Taupitz M, Pilgrimm H, Zimmer C (2002). In vitro characterization of two different ultrasmall iron oxide particles for magnetic resonance cell tracking. Invest Radiol, 37(9): 482–488

[7] Hsiao I L, Huang Y J (2011). Effects of various physicochemical characteristics on the toxicities of ZnO and TiO nanoparticles toward human lung epithelial cells. Sci Total Environ, 409(7): 1219–1228

[8] Jia G, Wang H F, Yan L, Wang X, Pei R J, Yan T, Zhao Y L, Guo X B (2005). Cytotoxicity of carbon nanomaterials: single-wall nanotube, multi-wall nanotube, and fullerene. Environ Sci Technol, 39(5): 1378–1383

[9] Kim S, Choi J E, Choi J, Chung K H, Park K, Yi J, Ryu D Y (2009). Oxidative stress-dependent toxicity of silver nanoparticles in human hepatoma cells. Toxicol In Vitro, 23(6): 1076–1084

[10] Lewinski N, Colvin V, Drezek R (2008). Cytotoxicity of nanoparticles. Small, 4(1): 26–49

[11] Lippacher A, Müller R H, Mäder K (2001). Preparation of semisolid drug carriers for topical application based on solid lipid nanoparticles. Int J Pharm, 214(1-2): 9–12

[12] Ma P, Luo Q, Chen J, Gan Y, Du J, Ding S, Xi Z, Yang X (2012). Intraperitoneal injection of magnetic Fe₃O₄-nanoparticle induces hepatic and renal tissue injury via oxidative stress in mice. Int J Nanomedicine, 7: 4809–4818

[13] Nakamura M, Ozaki S, Abe M, Doi H, Matsumoto T, Ishimura K (2010). Size-controlled synthesis, surface functionalization, and biological applications of thiol-organosilica particles. Colloids Surf B Biointerfaces, 79(1): 19–26

[14] Nishimori H, Kondoh M, Isoda K, Tsunoda S, Tsutsumi Y, Yagi K (2009). Silica nanoparticles as hepatotoxicants. Eur J Pharm Biopharm, 72(3): 496–501

[15] Novotna B, Jendelova P, Kapcalova M, Rossner P Jr, Turnovcova K, Bagryantseva Y, Babic M, Horak D, Sykova E (2012). Oxidative damage to biological macromolecules in human bone marrow mesenchymal stromal cells labeled with various types of iron oxide nanoparticles. Toxicol Lett, 210(1): 53–63

[16] Piao M J, Kang K A, Lee I K, Kim H S, Kim S, Choi J Y, Choi J, Hyun J W (2011). Silver nanoparticles induce oxidative cell damage in human liver cells through inhibition of reduced glutathione and induction of mitochondria-involved apoptosis. Toxicol Lett, 201(1): 92–100

[17] Pissuwan D, Niidome T, Cortie M B (2011). The forthcoming applications of gold nanoparticles in drug and gene delivery systems. J Control Release, 149(1): 65–71

[18] Strigul N, Vaccari L, Galdun C, Wazne M, Liu X, Christodoulatos C, Jasinkiewicz K (2009). Acute toxicity of boron, titanium dioxide, and aluminum nanoparticles to Daphnia magna and Vibrio fischeri. Desalination, 248(1-3): 771–782

[19] Tedesco S, Doyle H, Blasco J, Redmond G, Sheehan D (2010). Oxidative stress and toxicity of gold nanoparticles in Mytilus edulis. Aquat Toxicol, 100(2): 178–186

[20] Valodkar M, Jadeja R N, Thounaojam M C, Devkar R V, Thakore S (2011). In vitro toxicity study of plant latex capped silver nanoparticles in human lung carcinoma cells. Mater Sci Eng C, 31(8): 1723–1728

[21] Xia T, Kovochich M, Liong M, Mädler L, Gilbert B, Shi H, Yeh J I, Zink J I, Nel A E (2008). Comparison of the mechanism of toxicity of zinc oxide and cerium oxide nanoparticles based on dissolution and oxidative stress properties. ACS Nano, 2(10): 2121–2134

[22] Xie J, Huang J, Li X, Sun S, Chen X (2009). Iron oxide nanoparticle platform for biomedical applications. Curr Med Chem, 16(10): 1278–1294

[23] Yu M K, Jeong Y Y, Park J, Park S, Kim J W, Min J J, Kim K, Jon S (2008). Drug-loaded superparamagnetic iron oxide nanoparticles for combined cancer imaging and therapy in vivo. Angew Chem Int Ed Engl, 47(29): 5362–5365

[24] Zhu M T, Feng W Y, Wang B, Wang T C, Gu Y Q, Wang M, Wang Y, Ouyang H, Zhao Y L, Chai Z F (2008). Comparative study of pulmonary responses to nano- and submicron-sized ferric oxide in rats. Toxicology, 247(2-3): 102–111

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg