J Occup Health 2008; 50: 1–6

Received May 7, 2007; Accepted Sep 28, 2007Correspondence to: M.-H. Cho, Laboratory of Toxicology, Collegeof Veterinary Medicine, Seoul National University, Seoul 151-742,Korea (e-mail: mchotox@snu.ac.kr)

Body Distribution of Inhaled Fluorescent Magnetic Nanoparticlesin the Mice

Jung-Taek KWON1, Soon-Kyung HWANG1, Hua JIN1, Dae-Seong KIM3, Arash MINAI-TEHRANI1,Hee-Jeong YOON1, Mansoo CHOI3, Tae-Jong YOON4, Duk-Young HAN5, Young-Woon KANG5,Byung-Il YOON6, Jin-Kyu LEE2,4 and Myung-Haing CHO1,2

1Laboratory of Toxicology, College of Veterinary Medicine, 2Nano Systems Institute-National Core ResearchCenter, 3Nanotechnology & Thermal Processing Laboratory, School of Mechanical and Aerospace Engineering,4Materials Chemistry Laboratory, School of Chemistry, Seoul National University, 5Seoul Center, Korea BasicScience Institute and 6School of Veterinary Medicine, Kangwon National University, Republic of Korea

Abstract: Body Distribution of Inhaled FluorescentMagnetic Nanoparticles in the Mice: Jung-TaekKWON, et al. Laboratory of Toxicology, College ofVeterinary Medicine, Seoul National University,Korea—Reducing the particle size of materials is anefficient and reliable tool for improving the bioavailabilityof a gene or drug del ivery system. In fact,nanotechnology helps in overcoming the limitations ofsize and can change the outlook of the world regardingscience. However, a potential harmful effect ofnanomaterial on workers manufacturing nanoparticlesis expected in the workplace and the lack of informationregarding body distribution of inhaled nanoparticlesmay pose serious problem. In this study, we addressedthis question by studying the body distribution of inhalednanoparticles in mice using approximately 50-nmfluorescent magnetic nanoparticles (FMNPs) as amodel of nanoparticles through nose-only exposurechamber system developed by our group. Scanningmobility particle sizer (SMPS) analysis revealed thatthe mice were exposed to FMNPs with a total particlenumber of 4.89 × 105 ± 2.37 × 104/cm3 ( lowconcentration) and 9.34 × 105 ± 5.11 × 104/cm3 (highconcentration) for 4 wk (4 h/d, 5 d/wk). The bodydistribution of FMNPs was examined by magneticresonance imaging (MRI) and Confocal Laser ScanningMicroscope (CLSM) analysis. FMNPs were distributedin various organs, including the liver, testis, spleen, lungand brain. T2-weighted spin-echo MR images showedthat FMNPs could penetrate the blood-brain-barrier(BBB). Application of nanotechnologies should notproduce adverse effects on human health and the

Rapid Communication

environment. To predict and prevent the potentialtoxicity of nanomaterials, therefore, extensive studiesshould be performed under different routes of exposurewith different sizes and shapes of nanomaterials.(J Occup Health 2008; 50: 1–6)

Key words: Fluorescent Magnetic Nanoparticles,Inhalation, Mice

Rapid development in nanotechnology will result inseveral changes in areas such as nanoscale visualization,insight into living systems, revolutionary biotechnology,synthesis of new drugs for targeted delivery andregenerative medicine, and will offer many otherbenefits1). Reducing the particle size of materials is anefficient and reliable tool for improving the bioavailabilityof a gene or drug delivery system. In fact, nanotechnologyhelps to overcome the limitations of size and can changethe outlook of the world regarding science2). However, apotential harmful effect of nanomaterials on workersmanufacturing nanoparticles is expected in the workplaceand the lack of information regarding body distributionof inhaled nanoparticles may pose a serious problem.Therefore, it is necessary that specialists and researchersin toxicology, chemistry and other fields are aware of theimportance of analyzing the positive aspects ofnanomaterials while avoiding their potential toxiceffects3).

In a previous study, we reported that fluorescentmagnetic nanoparticles (FMNPs) did not cause anysignificant toxicity and penetrated the blood-brain-barrier(BBB) in mice treated by i.p. administration4). However,for the bioapplication of FMNPs, data regarding thetoxicity depending upon exposure routes should beaccumulated. In fact, dominant routes of exposure to

Journal ofOccupational Health

2 J Occup Health, Vol. 50, 2008

workers can be representative organs such as the lung(inhalation) which contains barriers to penetration bysmall particles. Regardless of the presence of such adefense mechanism, nano-sized materials may presentproblems for workers because the defense mechanismmay not respond properly to nano-sized materials.However, studies have not investigated the inhalationtoxicity of nanoparticles yet.

In this study, we addressed this issue by studying thebody distribution of inhaled nanoparticles in mice usingapproximately 50-nm FMNPs as a model of nanoparticlesthrough a nose-only exposure chamber system developedby our group. Here, we report that inhaled FMNPs weredistributed diversely in organs including the brain andtestis. Our results support the hypothesis that extensivetoxicity evaluation is needed before the practicalapplication of anthropogenic nanomaterials and suggestthat careful regulation of nanoparticles’ application maybe necessary to maintain a high quality of life as well asto facilitate the development of nanotechnology.

Materials and Methods

Fluorescent magnetic nanoparticles (FMNPs)FMNPs were synthesized by the co-precipitation

method in hot basic NaOH solution as described in aprevious study5). Briefly, a ferrite aqueous solution wasadded to polyvinylpyrrolidone (PVP) solution, and themixture was stirred for 1 d at room temperature. RITC-modif ied t r ie thoxysi lane was prepared f romaminopropyltriethoxysilane (APS) and RITC undernitrogen using a standard Schlenk line technique. A mixedsolution of tetraethoxysilane (TEOS) and RITC-modifiedtriethoxysilane (TEOS/RITC-silane molar ratio=0.3/0.04)was injected into the ethanol solution of PVP-stabilizedferrite. The PVP-stabilized FMNPs were separated bythe addit ion of aqueous acetone followed bycentrifugation at 4,000 rpm for 10 min. The supernatantwas removed and the precipitated particles wereredispersed in ethanol. The size and shape of the FMNPswere characterized by a transmission electron microscope(TEM). The prepared core-shell nanoparticles werehomogeneously dispersed in distilled water.

Animals and experiment designSpecific-pathogen-free male and female Slc: ICR mice

(5 wk old) purchased from SLC Inc. (Hamamatusu,Japan) were maintained in our laboratory animal facility(23 ± 2°C, 50 ± 20% relative humidity, 12-h light/darkcycle). The mice were acclimatized for at least 1 wkprior to the beginning of the study. All animalexperiments were performed according to the guidelinefor the care and use of laboratory animal of Seoul NationalUniversity. We randomized mice into 3 groups (10 maleand female mice per group): control, low and high FMNPsexposure groups. In this study, the total particle numbers

in the nose-only exposure chamber system weremaintained at 4.89 × 105 #/cm3 for low concentration and9.34 × 105 #/cm3 for high concentration. The animalswere exposed to FMNPs for 4 wk (4 h/d, 5 d/wk) in thenose-only exposure system. Controls were exposed toair filtered by a high efficiency particulate air (HEPA)filter. The body distribution of FMNPs was evaluatedby magnetic resonance imaging (MRI) and ConfocalLaser Scanning Microscope (CLSM).

FMNPs generation and exposureThe nose-only exposure chamber used in our study

consists of a cylindrical acrylic cylinder and 20 smalltubes (Fig. 1A). A serial aerosol device was constructedon the basis of an 84 × 540 mm acrylic cylinder (mainchamber). Conical acrylic tubes were connected to themain chamber and the mice were placed in the tubes fromthe backside. FMNPs were suspended in distilled water.The atomizer (Model 9302, TSI, MN, USA) generatedthe nanoparticles and these particles were forced to passthrough a heating tube where any remaining distilledwater was allowed to evaporate completely. The heatingtube temperature was maintained at 120–130°C6). Thegenerated aerosols were entered into the dilution chamberwhere the aerosols were mixed with filtered air usingHEPA cartridge-type filter (Model 12144, Pall, NY, USA)before passing into the nose-only exposure chamber.Particle number and size distribution were measured usinga scanning mobility particle sizer (SMPS, Grimm AerosolTechnik, Germany)6, 7). The SMPS, which consists of adifferential mobility analyzer (DMA) and a condensationparticle counter (CPC), is an instrument for measuringthe size distribution of particles in the range from 5 to1,000 nm. The flow rate for the chamber was set at 12 lper minute. When the airflow became stabilized, thenanoparticle number and size in the nose-only exposurechamber were measured at the center of the chamber. Inaddition, to further validate the real size of FMNPs, theFMNPs were sampled by a glass fiber filter (Model66209, Pall) during exposure period and analyzed byscanning electron microscope (SEM).

Fluorescent image analysisFor fluorescent image analysis, brain, liver, nasal

cavity, lungs, kidneys, heart, spleen, testes (male), andovary (female) were removed and the organs wereimmersion-fixed in 10% neutral buffered formalin. Afterroutine tissue processing, the tissues were embedded inlow temperature-melted paraffin. The fixed tissuesamples were sectioned at 4 µm thickness. The slideswere observed under a confocal laser scanningmicroscope (CLSM, Zeiss 510; Atlanta, GA, USA).Under microscopy the photographs were quantified andanalyzed by a computerized system (Image Pro-Plus,Media Cybernetics, MD, USA).

3Jung-Taek KWON, et al.: Body Distribution of Nanoparticles

Magnetic resonance image analysisFor effective anatomical magnetic resonance imaging

of brain, the mice were weighed and anesthetized. MRIexperiments on brain were performed on a 4.7 Tesla MRI/MRS system with a 35-mm vertical bore size (Varian,Unity Inova, Palo Alto , CA, USA) and a 30-mmMillepede quadrature. T2-weighted MR image wasobtained using the following parameters: repetition time

(TR)=3,000 ms, echo time (TE)=80 ms, field of view(FOV)=30 × 30 mm, thickness=1 mm, matrix=512 × 512.

Results

Characteristics and analysis of generated FMNPs innose-only exposure chamber

The TEM study clearly demonstrated that oursynthesized FMNPs were monodisperse and showed

Fig. 1. N a n o p a r t i c l e g e n e r a t i o n s y s t e m a n d c h a r a c t e r i z a t i o n o f F M N P s .(A) Schematic diagram of inhalation system for nanoparticle exposure. The nose-only exposure chamber used in our studyconsists of a cylindrical acrylic cylinder and 20 small tubes. The nanoparticles were generated through an atomizer with theaid of a heating tube to evaporate the remaining water. The generated aerosols were entered into a dilution chamber wherethey were mixed with air filtered by HEPA cartridge-type filter before passing into the nose-only exposure chamber.Particle number and size distribution were measured using the scanning mobility particle sizer (SMPS). The measurementsof particle number and size distribution were performed in total 7 independent analyses with 40 min intervals afterachieving stabilization of the nanoparticle generating system. (B) TEM image of FMNPs. TEM study clearly demonstratesthat our synthesized FMNPs are monodisperse and have unique size. Scale bar represents 100 nm. (C) Size and number ofFMNPs. The size distribution of generated FMNPs was determined using SMPS. G.M.D.(nm) and G.S.D. of the FMNPswas 49 nm and 1.8, and 51 nm and 1.7 for the low and high concentrations, respectively. (D) SEM image of FMNPs.Validation study by SEM of glass fiber captured FMNPs revealed that some FMNPs were aggregated, however, mostFMNPs are distributed evenly and uniformly in NOEC during the study. Scale bar: 100 nm.

4 J Occup Health, Vol. 50, 2008

unique size (Fig. 1B). SMPS study of the particle numberof FMNPs in the nose-only exposure chamberdemonstrated that there was no significant differencebetween the two concentration groups [geometric meandiameter (GMD) and geometric standard deviation (GSD)of the FMNPs were 49 nm and 1.8 for the lowconcentration and 51 nm and 1.7 for the highconcentration]. SPMS analysis also demonstrated that

the particle number distribution of generated FMNPs wasconstantly maintained during the exposure period. SMPSanalysis revealed that the mice were exposed to FMNPswith total particles number of 4.89 × 105 ± 2.37 × 104/cm3 (low concentration) and 9.34 × 105 ± 5.11 × 104/cm3

(high concentration) for 4 wk (4 h/d, 5 d/wk). SPMSanalysis also demonstrated that the particle numberdistribution of generated FMNPs was constantlymaintained during the exposure period (Fig. 1C). Avalidation study by SEM of FMNPs, captured on a glassfiber filter, revealed that some FMNPs were aggregated;however, most FMNPs were evenly and uniformlydistributed in the nose-only exposure chamber during thestudy (Fig. 1D).

Body distribution of Inhaled FMNPsThe body distribution of FMNPs was examined by

CLSM and MRI study. As shown in Fig. 2, FMNPs weredistributed in various organs, including the liver, testis,spleen, lung and brain. In contrast, only a few weredistributed in the nasal cavity, heart, kidney and ovary.In the liver, the fluorescence intensity of FMNPs wasstrongest and distributed throughout the whole organ. Inthe spleen and testis, however, FMNPs were observed inspecific regions. Since we were interested in studyingbrain distribution of FMNPs after inhalation, the braintissue sections were inspected by MRI. In T2-weighted

Fig. 2. Body distribution of FMNPs. (A) CLSM image ofvarious organs of mice treated with FMNPs. FMNPswere frequently detected in the liver, testis, spleen andbrain, but not so much in the nasal cavity and ovary. Inthe testis, FMNPs were mostly found in the interstitialcells. × 400, scale bar: 20 µm. (B) Analysis of imageintensity. Values are means ± S.E. Significantlydifferent from control (* p<0.05, ** p<0.01). N.C.(nasal cavity).

5Jung-Taek KWON, et al.: Body Distribution of Nanoparticles

spin-echo MR images, it could be seen that FMNPs hadpenetrated the BBB (Fig. 3)

Discussion

To understand the potential adverse effects ofnanomaterials on health, it is important to know thegeneral defense mechanism of living organisms and theinteractions between nanoparticles and the immuneresponse. Based on evolutionary history, humans havebeen exposed to small particles, and have developeddefense mechanisms against such particles. Dominantroutes of nanomaterial exposure to workers arerepresented by organs such as the lung (inhalation) andskin (contact) which contain barriers to penetration bysmall particles. Regardless of the presence of suchdefense mechanisms, nano-sized materials may presentseveral problems for workers, because their defensemechanisms may not respond properly to nano-sizedmaterials. Mineral quartz, asbestos and particlesassociated with air pollution are the representativeexamples of such nanomaterials3). There is some limitedresearch related to the impacts of nanomaterials on non-human species and about workers affected bynanomaterials. In our previous study, we showed thati.p. administered 50 nm FMNPs could penetrate theblood-brain-barrier without causing significant toxicity4).The collection of information on risk assessment viadifferent exposure routes of manufactured nanoparticlesis also required. Issues regarding the potential impact ofmanufactured nanomaterials on human health andenvironment have been raised only recently3). In

consideration of these regards, the current inhalation studywas performed to determine the distribution pattern ofnanomaterials using FMNPs. The SEM study showedthat some FMNPs were aggregated (Fig. 1D). A fewaggregations might be caused by Brownian motion andturbulence8, 9). Our results indicate that a nanoparticlegenerating system can be used for an inhalation study ofdiverse kinds of nanomaterials. In fact, extensive size-dependent inhalation toxicities of various nanomaterialsare under investigation. The differences in toxicitydepending upon exposure routes (i.p. vs inhalation) arequite interesting. Most likely, first-pass effects by theportal circulation would account for the differencebecause the majority of i.p. administered FMNPs wouldbe taken up by the liver via the first-pass effect and thenbe redistributed from the liver to the other organs.However, inhalation exposure route would bypass suchliver-related first-pass effect. We studied the organs whichare enriched with the reticuloendothelial system (RES)such as the liver, lung and spleen and non-RES organssuch as the heart and kidney. Among the RES organs,the spleen was affected by inhalation of FMNPs, thereby,suggesting that the body distribution of the FMNPs inthe liver, lung, and spleen might not be associated withthe RES system. Our results strongly suggest that otherfactors may be involved for tissue specific distributionpattern of inhaled FMNPs. The functional mechanismof BBB is thought to be specialized endothelial cells inthe brain microvasculature, which are aided at least inpart, by interactions with glia. Among the uniqueproperties of these endothelial cells is the presence of

Fig. 3. Magnetic resonance image of a mouse brain. Representative cross-sectional MR images of the brain of mouse which had a highconcentration of FMNPs. Note the rectangle represents thecomparison of MR images. The MR image of the brain from a highexposure mouse (right) is darker than that of control mouse (left),indicating that FMNPs were distributed in the brain. T2-weighted MRimage was obtained using the following parameters: repetition time(TR)=3,000 ms, echo time (TE)=80 ms, field of view (FOV)=30 × 30mm, thickness=1 mm, matrix=512 × 512.

6 J Occup Health, Vol. 50, 2008

tight junctions between cells, where the gaps betweenthe junctions are approximately 4 nm10). To gain entry tothe brain, therefore, the FMNPs would probably have topass through the cell membrane of the endothelial cellsof the brain, rather than between the endothelial cells.The penetration of molecules into the brain is largelyrelated to their lipid solubility and to their ability to passthrough the plasma membranes of the cells forming thebarrier11). However, FMNPs do not show this abilitybecause they are highly water soluble. Moreover, thecells of the brain lack of pinocytosis; therefore, FMNPswould have to gain access to the brain by some otherroutes. In the mature central nervous system, the spinaland autonomic ganglia as well as a small number of othersites within the brain, called the circumventricular organs,are not protected by the BBB12). The discontinuity of thebarrier may allow the entry of FMNPs into the brain.Another possibility is that FMNPs may translocate alongthe olfactory nerve into the olfactory bulb13). In fact, arecent report showed that translocation of inhalednanoparticles along the neurons seemed to be moreefficient pathway to the CNS rather than via the BBB14).Such efficient translocation of FMNPs via the olfactoryneuronal pathway in the nasal cavity (Fig. 2) could beresponsible for the observed brain distribution of FMNPs.Moreover, FMNPs also penetrated the testis which isprotected by blood-testis barrier (BTB). The production,differentiation, and presence of male gametes representinimitable challenges to the immune system as theexistence of BTB protects the immune privilege of thetestis15). In this regard, FMNPs may have penetrated BTB,thus, disturbing the maintenance of the immune-privileged status of the testis. Further study is urgentlyneeded to elucidate the precise mechanism by whichFMNPs penetrate BTB and the potential outcome for thedistribution of FMNPs in the testis.

We report here that inhaled FMNPs have penetratedthe BBB and BTB. Applications of nanotechnologiesshould not produce adverse effects on human health andthe environment. Therefore, to predict and prevent thepotential toxicity of nanomaterials, extensive studiesshould be performed under various exposure routes withdifferent sizes and shapes of nanomaterials.

Acknowledgments: Parts of current study weresupported by the research grant of KFDA. J.T.K., S.K.W.,A.M-T., H.J., H.J.Y., and T.J.Y. are grateful for financialsupport from the BK21 program. J.K.L. and M.H.C.thank to the support from the Nano Systems Institute-National Core Research Center (NSI-NCRC) throughKOSEF.

References 1) Roco MC: Nanotechnology: convergence with modern

biology and medicine. Curr Opin Biotechnol 14, 337–346 (2003)

2) Mirkin CA and Taton TA: Semiconductors meetbiology. Nature 405, 626–627 (2000)

3) Oberdorster G, Oberdorster E and Oberdorster J:Nanotoxicology: an emerging discipline evolving fromstudies of ultrafine particles. Environ Health Perspect113, 823–839 (2005)

4) Kim JS, Yoon TJ, Yu KN, Kim BG, Park SJ, Kim HW,Lee KH, Park SB, Lee JK and Cho MH: Toxicity andtissue distribution of magnetic nanoparticles in mice.Toxicol Sci 89, 338–347 (2006)

5) Yoon TJ, Kim JS, Kim BG, Yu KN, Cho MH and LeeJK: Multifunctional nanoparticles possessing a“magnetic motor effect” for drug or gene delivery.Angew Chem Int Ed Engl 44, 1068–1071 (2005)

6) Kim DS, Lee KW and Kim YJ: Characterization of aparticle trap impactor. J Aerosol Sci 37, 1016–1023(2006)

7) Kim DS, Park SH, Song YM, Kim DH and Lee KW:Brownian coagulation of polydisperse aerosols in thetransition regime. J Aerosol Sci 34, 859–868 (2003)

8) Lee KW: Change of particle size distribution duringBrownian coagulation. J Colloid Interface Sci 92, 315–325 (1983)

9) Park SH, Lee KW, Otto E and Fissan H: The log-normalsize distribution theory of brownian aerosol coagulationfor the entire particle size range: part I analyticalsolution using the harmonic mean coagulation kernel.J Aerosol Sci 30, 3–16 (1999)

10) Del Zoppo GJ, Milner R, Mabuchi T, Hung S, Wang Xand Koziol JA: Vascular matrix adhesion and the blood-brain barrier. Biochem Soc Trans 34, 1261–1266 (2006)

11) Stewart PA: Endothelial vesicles in the blood-brainbarrier: are they related to permeability? Cell MolNeurobiol 20, 149–163 (2000)

12) Yonezawa T, Ohtsuka A, Yoshitaka T, Hirano S,Nomoto H, Yamamoto K and Ninomiya Y: Limitrin, anovel immunoglobulin superfamily protein localizedto glia limitans formed by astrocyte endfeet. Glia 44,190–204 (2003)

13) Oberdorster G, Sharp Z, Atudorei V, Elder A, GeleinR, Kreyling W and Cox C: Translocation of inhaledultrafine particles to the brain. Inhal Toxicol 16, 437–445 (2004)

14) Elder A, Gelein R, Silva V, Feikert T, Opanashuk L,Carter J, Potter R, Maynard A, Ito Y, Finkelstein J andOberdorster G: Translocation of inhaled ultrafinemanganese oxide particles to the central nervoussystem. Environ Health Perspect 114, 1172–1178(2006)

15) Fijak M and Meinhardt A: The testis in immuneprivilege. Immunol Rev 213, 66–81 (2006)