skip to: page content | site navigation | section menu
Journal of Medical Chemical, Biological and Radiological Defense
Original Received July 11, 2010 J Med CBR Def | Volume 8, 2010
Submitted 11 July 2010 | Accepted 9 October 2010 | Revised 11 October 2010 | Published 27 October 2010

The Crystalline Forms Of Titanium Dioxide Nanoparticles Affect Their Interactions With Individual Cells

Elena I. Ryabchikova1*, Natalia A. Mazurkova2, Nadezdda V. Shikina3, Zinfer R. Ismagilov3

1Institute of Chemical Biology and Fundamental Medicine, Siberian Branch of Russian Academy of Science. Lavrent’eva pr. 8, Novosibirsk. 630090. Russia

2Novosibirsk State University, Pirogova str. 2, Novosibirsk, 630090. Russia

3Boreskov Institute of Catalysis, Siberian Branch of Russian Academy of Science. Lavrent’eva pr.5, Novosibirsk. 630090. Russia

* Corresponding Author:Elena I. Ryabchikova

Tel: +7 (383) 363 51 63 | Email:

Suggested citation: Ryabchikova, E; Mazurkova, N; Shikina, N; Ismagilov, Z; “The Crystalline Forms Of Titanium Dioxide Nanoparticles Affect Their Interactions With Individual Cells”, JMedCBR 8, 27 October 2010,


Nanoparticles were synthesized during the last few decades and have shown great potential to improve the living conditions of mankind in general. Despite nanotechnology´s promise and progress, unconditional acceptance of nanotechnologies is not supported by solid evidence of the safety of nanoparticles for living organisms. Especially in those areas where nanoparticles have direct contact with living organisms, such as in medical treatment, diagnostics, and in agriculture, some safety doubts exist. Titanium dioxide (TiO2) was considered for a long time as biologically inert material in pigment-grade form (spheres of more than 200 nm in diameter) [Ophus et al., 1979; Lindenschmidt et al., 1990]. This compound has been used in the composition of paints, plastics, cosmetics, medical preparations and food for many years, and presently is increasingly applied in nanoforms [Chen, Mao, 2007]. However, the extremely small sizes of the nanoparticles may result in different interactions with cells: nanoparticles have sizes 1—100 nm in two or three dimensions [ASTM E 2456-06]. These interactions could result in toxic properties in materials that were formerly safe when macroforms of the particles were used. The toxic effects of TiO2 nanoparticles have been examined using both routine toxicology and modern cell biology methods. The studies showed that TiO2 nanoparticles could cause various cytotoxic and genotoxic effects, which were though to be attributed to generation of reactive oxygen species [Chan Jin et al., 2010; Cheng-Yu et al., 2008; Kang et al. 2008; Rahman et al., 2002; Vevers, Awadhesh, 2008]. It should be noted, that electron microscopy, which is a powerful tool for nanoworld investigations, has been applied only in few studies of TiO2 nanoparticle interaction with a cell. Damage of mammalian and fish cell cultures after 24—48 h incubation in the presence of nanoparticles and the accumulation of nanoparticles in phagosomes have been reported [Chan Jin et al., 2010; Cheng-Yu, et al., 2008 ; L'Azou, et al., 2008; Vevers, Awadhesh, 2008]. All published electron microscopic studies have been performed on samples incubated with TiO2 nanoparticles for relatively long times (24 and 48 h). No data have been currently reported concerning the early steps of the TiO2 nanoparticle’s interaction with a cell. The early steps of the TiO2 nanoparticle’s interaction with an individual cell are particularly intriguing because these steps could be the main factors determining mechanisms of the cell response to nanoparticles. We pointed our research on examination of early steps of the TiO2 nanoparticles interaction with a cell using electron microscopy.

The TiO2 nanoparticles examined in cell cultures mostly had sizes of more than 10 nm and were the anatase crystalline modification. In a few studies, the rutile modification was used, and some studies did not mention the crystalline modification of the nanoparticles [Cheng-Yu et al, 2008; Kang et al., 2008; L'Azou et al., 2008; Vevers et al., 2008; etc.]. We tried to compare the interaction of different crystalline modifications (amorphous, anatase, rutile, and brookite) of TiO2 nanoparticles having identical sizes (4—5 nm) with the cells of MDCK and Vero cultures. Our data confirm the very close interaction of the nanoparticles with cellular mechanisms and the dependence of the cell’s response on the crystalline form of the compound.



TiO2 nanoparticles having different crystal modifications (amorphous, anatase, rutile, and brookite) were chemically synthesized and used in this study. Briefly, amorphous TiO2 nanoparticles were synthesized by hydrolysis of TiCl4 (99.6 %, Alfa Aesar, Germany) in deionized water at 23—25oC in presence of NaOH (Merck, Germany). Drops of NaOH were added to aqueous solution of TiCl4, with stirring to final pH 6—7 [Ismagilov et al., 2008; Mazurkova et al., 2010]. Anatase TiO2 nanoparticles have been prepared similarly and reported earlier [Zhang, Sun, 2004] by adding to mixture (volume ratio 2:1) of titanium isopropoxide (Ti(i-OC3H7)4, 95%, Alfa Aesar, Germany) and 2-isopropanol (Sigma-Aldrich Chemie GmbH) drops of HNO3 (Merck, Germany) taking [H+]/[Ti] = 0.2 and [H2O]/[Ti] = 100, under intense stirring. The sol then was stirred during 7 h at 70oC, and cooled to room temperature. Brookite TiO2 nanoparticles were prepared by hydrolysis of TiCl4 in water at 3—4oC and pH1, cooled to 4ºC and then TiCl4 was added drop-wise in cold water under intense stirring. The transparent sol was further stirred during 10 min at temperature and pH control. Rutile TiO2 nanoparticles were prepared in two stages. The first stage was identical to brookite preparation. Then the transparent sol was heated to 70oC and stirred at this temperature for 6 h. All prepared products were washed by dialysis against deionized water at 3—4oC to pH=6.7—7.0, using cellulose membrane (ZelluTrans T1, Roth, Carl Roth GmbH+Co.KG). Synthesis of the nanoparticles, their physico-chemical properties and stability were described in details earlier, all nanoparticles had identical sizes 4—5 nm (determined by small-angle X-ray scattering, X-ray diffraction, dynamic laser scattering, atomic force microscopy, transmission and scanning electron microscopy) [Ismagilov et al., 2009; 2010]. Stock suspensions of nanoparticles in deionized water, concentration of 1 mg/mL, were sonicated and diluted before usage by SFM4MegaVir («Hy Clone», USA) cell culture medium with pH 7.2.


MDCK and Vero cells were propagated in SFM4MegaVir and standard DMEM (Invitrogen, Paisley, UK) cell culture mediums correspondingly, supplemented with 2% fetal cow serum «Gibco» (USA), at 37oC in the presence of 5%CO2, 100 U/mL of penicillin and 100 mkg/mL streptomycin. For trypan blue assay the cells were seeded in 12-well plates (Corning, Lowell, USA) at a density of 1.0-1.5×105 cells/mL in 1 mL of culture medium to each well. For electron microscopy the cells were seeded on 6-well plates at a density of 1.0-1.5×105 cells/mL in 3 mL culture medium to each well. The cells were propagated to full monolayer confluence. After washing with fresh cell culture medium the cells were used for experiments with nanoparticles in the same dish. All studies with the nanoparticles have been performed without serum supplement.

Trypan blue assay

For primary evaluation of the toxic effect of different crystalline modifications of TiO2 nanoparticles MDCK cell culture monolayers were treated with a range of TiO2 concentrations (50-1000 µg/mL), incubated during 48 h in the dark as described above, and examined in Axiovert 40 (ZEISS) inverted microscope. Based on visual signs of cell damage non-toxic concentrations were determined: 50-200 µg/mL. Toxic effect of amorphous, anatase, rutile, and brookite TiO2 nanoparticles has been studied by trypan blue staining of MDCK and Vero cells which were incubated at 37oC during 5 and 15 h with 100 µg/mL of the preparations, or 3 and 5 h in presence of 20 and 100 µg/mL of the preparations. Untreated cells served as a control. After the incubation cells were washed with fresh culture medium and removed from the plastic with 1:1 mixture of trypsin (0.25%) and EDTA (0.2%), and stained with 0.025% trypan blue. Number of living and dead cells was counted with a Leica DM 2500 microscope. All experiments have been performed in three independent series, and each figure in the tables represents data from 4—5 independent countings of different samples.

Electron microscopy

Monolayers of MDCK cell culture were treated by amorphous, anatase, rutile and brookite TiO2 nanoparticles in concentrations of 100 and 300 µg/mL during 1, 3 and 5 h at 37oC in the dark. After the incubation, the cells were washed with fresh cell culture medium, gently scrapped, collected in plastic tubes, and centrifuged at 3000 run/min. The pellets were fixed by 4% paraformaldehyde, and routinely processed for electron microscopy, and embedded in epon-araldit resin mixture. Ultrathin sections were routinely stained with uranylacetate and lead citrate and were examined in Jem 1400 (JEOL, Japan) transmission electron microscope. Digital images have been taken by Veleta CCD camera (SIS, Germany).


Trypan Blue Assay Results

Trypan blue assay is a common standard method for determining the number of viable cells in a cell suspension; the trypan blue stains penetrate only the damaged cell plasma membrane, while living cells remain unstained [Altman et al., 1993]. We applied trypan blue assay to examine the effects of TiO2 nanoparticles of different crystalline modifications on two types of cell cultures: Vero and MDCK. The results of the trypan blue assay varied with concentration and incubation period (5 and 15 h) for each crystalline modification of TiO2 nanoparticles are shown in table 1. Table 2 shows the results for MDCK cells with two concentrations of TiO2 and two incubation times. However, in the same experiments Vero cells showed higher damage in comparison with MDCK cells (number of the cells per one well was the same for both cultures). This effect could be related to the different cell structures and surface areas of the two cell cultures. MDCK cells are epithelioid and have tight junctions and columnar shape, so they form solid monolayers and each cell has about 1/3 surface area for contact with culture medium, other contact with neighboring cells and supporting surface. Vero cells are fibroblastoid, composed of flat cells spread on supporting surface, and so each cell has about half of the surface area for contact with culture medium. These different shapes provide different amounts of cell surface areas for contact with the culture medium and with TiO2 nanoparticles.

The ability of TiO2 nanoparticles to damage the cell plasma membrane was predictable; the higher concentration of TiO2 and longer contact time produced more damage to the cells. However differences in alteration of the same cells by the four crystalline modifications of TiO2 were unexpected. Treatment of the cells by identical concentrations of different nanoparticles and incubation times produced different amounts of altered cells (Tables 1 and 2) indicating that the four different crystalline modifications of nanoparticles had different abilities to alter or damage MDCK and Vero cells.

The data in Tables 1 and 2 show slightly different results for exposure of MDCK cells to 100 µL of TiO2 and 5 h contact time. This variation is actually normal because these two experiments were done about five months apart and there were differences in passage levels and other supplemental factors (serum, culture medium). The changes in altered cells are largest for amorphous TiO2, whose characteristics include greater tendencies for aggregation and other structural changes than other crystalline modifications [Ismagilov et al., 2008; 2010]. Electron Microscopy Results: Nanoparticle Structures Electron microscopy has been routinely applied for visualization and measurement of the nanoparticles. Anatase and rutile TiO2 nanoparticles (sizes varied from 10 to 100 nm) appear spherical in published electron microscopic images, which were obtained by scanning or transmission electron microscopy of the samples adsorbed or dispersed on film-covered grids [Chen, Mao, 2007; Cheng-Yu et al., 2008; Hamilton et al., 2009; Kang et al., 2008; L'Azou et al., 2008; Vevers et al., 2008, and others]. We examined the structure of TiO2 nanoparticles on ultrathin sections of MDCK cell culture after 1 h treatment (thickness of a section is 50-60 nm), and found different shapes of the nanoparticles (figure 1).

Amorphous TiO2 nanoparticles were spherical and of middle electron density having size 4-5 nm and rare particles with diameter 10-15 nm. Fine electron dense filamentous material also was recognized between the particles. Particles formed band-like agglomerations looking more or less electron dense, depending on the mass of particles in a section (figure 1A).

Anatase and brookite TiO2 nanoparticles had fine needle-like shapes and formed various figures such as stars and branches, in a whole looking like delicate lace. Anatase nanoparticles appeared more electron-dense and coarse than brookite. The fine needles of both anatase and brookite were individual and did not appear to agglutinate as with the amorphous nanoparticles (figure 1B, C).

Rutile TiO2 nanoparticles formed long needles with 4—5 nm in diameter (dendrites), which composed aggregates resembling palm-leave or fan-like structures (figure 1D). The appearance of the rutile nanoparticles depended on the plane of ultrathin section, and aggregations could look formless. The size of rutile aggregations varied from 30—50 nm to 5—6 µm. The high electron density of rutile nanoparticles altered operation of CCD camera, and clear focus sometimes was possible separately for the rutile particles or for a cell.

The structures of examined TiO2 nanoparticles crystalline modifications on ultrathin sections corresponded to those observed by other methods (see Materials and Methods section). According to our observations, sample processing for ultrathin sectioning does not alter the structure of TiO2 nanoparticles, and that 1 h incubation with living cells also does not influence nanoparticles structure. The ultrathin sections, together with other methods, can be used to examine the structure and measure nanoparticles.

The fine structure of all TiO2 nanoparticles did not change during 3 and 5 h of incubation with cells; however, increases of agglutination and formation of electron dense masses were observed in amorphous, anatase and brookite crystalline modifications.

Electron Microscopy Results: Cellular Responses

Very few publications have examined the details of interactions of TiO2 nanoparticles with a cell at subcellular level using electron microscopy. The main point in these published studies was to show that the nanoparticles entered cells. Thus, presence of 15-nm anatase TiO2 nanoparticles in cytoplasmic vesicles of epithelial proximal tubular LLC-PK1 cells was noted after 24h incubation [L'Azou, et al., 2008]. Phagocytosis of 50-nm diameter TiO2 particles aggregates was found in A549 cell culture (alveolar epithelia type II cells) [Stearns et al., 2001]. Accumulation of anatase and rutile clusters in cytoplasm and cytoplasmic vacuoles was observed in HaCaT cells (immortalized keratinocyte cell line established from adult human skin cells) [Chan Jin et al., 2010]. Studies of L929 mouse fibroblast cells 48 h after treatment with anatase nanoclusters (20—30 nm diameter) revealed increases in the number of lysosomes and disappearances of some cytoplasmic organelles [Cheng-Yu, et al., 2008]. No detailed descriptions and data about mechanisms of TiO2 nanoparticles interaction with a cell have been reported. And, no reports describe any events related to single nanoparticles and cells.

We examined ultrathin sections of MDCK cell culture incubated with different crystalline modifications of TiO2 nanoparticles for 1, 3 and 5 h. Amorphous TiO2 nanoparticles after 1 h incubation covered large area of the cell surface, and the nanoparticles were observed in all folds and cavities of the cell plasma membrane, and thereby deeply penetrated into the cell (figure 2A). It should be noted that clear views of a membrane are possible only when the plane of the section is perpendicular to the cell’s surface, otherwise the membranes in ultrathin sections are invisible. The amorphous nanoparticles seemed to fill all the cavities formed in the cell plasma membrane. Examination of strong perpendicular-plane sections showed two types of direct contact between a single amorphous TiO2 nanoparticle and the cell plasma membrane: contact of single spherical particle (figure 2B) and contact by fine filaments extended between particles and cell plasma membrane (figure 2C). Pictures of “pricking” of the membrane by filamentous material also were observed (figure 2D), however such pictures are ambiguous because the section plane cannot be absolutely perpendicular and the membrane thickness (7—8 nm) is comparable to nanoparticle sizes. Band-like agglomerations contacted with cell plasma membrane by one end or several “feet” (figure 1E). Images of “gluing” of the nanoparticles with the cell plasma membrane were often seen on ultrathin sections that were in “not-perpendicular” plane: one image of this is shown in figure 2F.

How do nanoparticles enter the cells? Electron microscopic examinations showed direct contact of amorphous TiO2 nanoparticles with the cell plasma membrane that are suggestive of possible macromolecule internalization mechanisms (clathrin- and caveolin-dependent endocytosis, macropinocytosis) by the nanoparticles. However, we noted few observations of amorphous TiO2 nanoparticles in “coated” pits (figures 2 G, H). “Coated” pits without nanoparticles were abundant on MDCK cell surface; each section of any cell possessed at least 3—5 pits. These structures represent the initial stages of clathrin-dependent receptor-mediated endocytosis finishing in endosomes [Parkar et al., 2009; Jovic et al., 2010]. Some TiO2 nanoparticles reached the endosomes (figure 2 I,J). The amounts of the “coated” pits and endosomes containing amorphous nanoparticles were very small (12 and 14 findings per sections of 100 cells), indicating that this mechanism is only occasionally used. Receptor-mediated endocytosis does not provide appreciable internalization of amorphous TiO2 nanoparticles. Our electron microscopic examinations of MDCK cells treated by amorphous TiO2 nanoparticles showed that incubation during 1 h resulted in direct contact of the nanoparticles with the cell membrane, penetration into a cell via surface folds and invaginations, and passive internalization of few nanoparticles by receptor-mediated endocytosis.

We examined anatase TiO2 nanoparticles on MDCK cells’ surface and found they penetrated the cell through folds and invaginations (figure 3A) similar to amorphous TiO2 nanoparticles. The mass of anatase directly contacted with cell plasma membrane by single “needles” (figure 3B). Sometimes formation of small electron dense agglomerations, and direct contact of these agglomerations with the cell plasma membrane, was noted (figure 3C). The presence of anatase in “coated” pits and endosomes (figure 3 D — G) was not a rare observation: 148 and 62 findings of anatase per sections of 100 cells were noted in “coated” pits and endosomes, respectively. Additional investigations are needed to understand the exact role of receptor-mediated endocytosis in anatase internalization. However, there were clear differences in the amounts of “coated” pits and endosomes that were observed between the amorphous and anatase modifications. These differences suggest that anatase is capable of a more active interaction with cells using receptor-mediated endocytosis pathway. This suggestion is supported by recent data about the involvement of macrophage receptor with collagenous structure (MARCO) in internalization of anatase TiO2 nanoparticles [Zhou et al., 2008; Thakur et al., 2009; Hussain et al., 2010].

Examination of anatase treated MDCK cells revealed nanoparticles in rare phagosomes (figure 3 H, I). We also observed mitochondria containing electron dense membrane fragments, areas of electron lucent matrix and electron dense spherical granules. Altered mitochondria were subjected to autophagocytosis and autophagosomes with osmiophilic membranes were found inside the cells (figure 3 J, K). Vacuolization of mitochondria and cristae disruption in HaCaT cells by reactive oxygen species were shown directly in studies of anatase nanoparticles (diameter less than 25 nm) [Chan Jin et al., 2010]. The ability of anatase to induce active oxygen radicals also was reported by other authors [Chen, Mao, 2007; Chan Jin et al., 2010; Liu et al., 2010]. The formation of osmiophilic electron dense membranes in our study also was probably related to reactive oxygen radicals produced during anatase-cell interaction. We did not observe damage of mitochondria and autophagosome formation in MDCK cells, either in our control cells or in cells that were treated with other modifications of TiO2 nanoparticles and in untreated control cells in our study.

Brookite TiO2 nanoparticles appeared very similar to anatase, however, after 1 h incubation they remained on the MDCK cell surface, which did not have deep invaginations and folds (figure 4A). MDCK cells treated with brookite appeared rounded and held rare microvilli. Brookite nanoparticles directly contacted with the cell plasma membrane via single “needles” (figure 4 B, C). Thus, in contrast to amorphous and anatase TiO2 nanoparticles, the brookite nanoparticles did not penetrate into MDCK cells and were not found in “coated” pits. No brookite nanoparticles were found in endosomes and phagosomes. Many brookite treated MDCK cells showed a swelling of endoplasmic reticulum cisterns (figure 4C). An appreciable number of the cells appeared to be involved in vacuolization of the endoplasmic reticulum and other pathological changes, which progressed to necrosis (figure 4D). All damaged cells had close contact with brookite nanoparticles, but no nanoparticles were detected in the cytoplasm. There are no publications in relevant literature describing the interaction of brookite TiO2 nanoparticles with cells or animals. Our study found that the brookite modification of the TiO2 nanoparticles possesses properties clearly different from those of anatase and amorphous TiO2, despite identical sizes and similar shapes.

The aggregations of rutile TiO2 nanoparticles were large and physically hard. Consequently when the ultrathin sections were cut, the sections always had holes. The high electron density of these aggregations also affected the operation of CCD camera, and, these rutile aggregations appeared as very dense and formless in the background of “dark” cells or appeared as a distinctive “palm-leave”-like structure in the background of “non-contrast” cell. These are apparent in figure 5. Examination of MDCK cells treated with rutile nanoparticles for 1 h revealed aggregations located on the cell surface and in large and small vacuoles having smooth membranes (figure 5). It is unclear how these vacuoles were formed: possibly by simple buckling under the weight of rutile or perhaps active uptake of rutile particles by a cell. We did not observe the contact of single rutile particles with the cell plasma membrane; the rutile modification shows evidence of an interaction with MDCK cells that is different from those of amorphous, anatase and brookite TiO2 nanoparticles. Treatment of MDCK cells with rutile nanoparticles caused damage of some cells that developed swelling of endoplasmic reticulum cisterns and mitochondria. However, the number of these cells was incomparably less than in brookite treated cells.

Thus, electron microscopic examination of MDCK culture cells treated for 1 h with four crystalline modifications of identical sizes of TiO2 nanoparticles found different patterns in the early steps of cell—TiO2 interactions.

Longer incubation of MDCK cells with TiO2 nanoparticles for 3 and 5 h did not have different patterns in the nanoparticles’ interaction, compared with 1 h incubation; however, we observed an accumulation of amorphous, anatase and rutile nanoparticles inside cells. We also observed an evident increase of the number of phagosomes containing amorphous TiO2 nanoparticles in MDCK cells (figure 6). We noted the same pattern of interaction of anatase TiO2 nanoparticles with MDCK cells for 1, 3, and 5 h incubation, including the formation of “coated” pits and endosomes, and we observed an increase in the total phagosome number. The amorphous and anatase TiO2 nanoparticles appeared to accumulate in MDCK cells during the incubation and increased cell damage.

Treatment of MDCK cells with brookite TiO2 nanoparticles for 3 h resulted in the appearance of some cells with surface folds and invaginations filled with the nanoparticles (figure 7 A, B), but the remaining cells did not show signs of brookite internalization and contacted with brookite “needles” (figure 7 C). Nanoparticles located inside the cell plasma membrane invaginations appeared different from those located outside of the cell: the distance between the “needles” was shorter and mass of the particles appeared denser, however there was no agglutination of “needles” (figure 7 B). This structural difference could reflect the changes in brookite nanoparticles properties, which were responsible for changed pattern in brookite-cell interaction.

The samples of MDCK cells incubated with brookite TiO2 nanoparticles for 5 h showed a further increase in the number of cells penetrated with folds and invaginations containing the nanoparticles, however many cells maintained their structure and did not internalize the nanoparticles. The number of phagosomes increased markedly in all cells of brookite-treated samples. After incubation with MDCK cells, the brookite TiO2 nanoparticles appeared to change their 3D-structure and pattern of interaction with cells.

Examination of MDCK cells treated by all crystalline modifications of TiO2 nanoparticles for 3 and 5 h revealed a noticeable increase in the number of damaged cells that showed signs of altered water-ion balance, i.e., swelling of the cells and decrease of cytoplasm electron density, and vacuolization of endoplasmic reticulum cisterns and mitochondria. Necrotic cells also were observed in all samples. The ratio between unaltered and altered or damaged cells in the samples incubated with different modifications of TiO2 nanoparticles was similar to the of trypan blue viability assay results (Table 2). Many cells were unaffected; their morphology was “normal” and they contained no nanoparticles. In all samples, not each and every cell was able to internalize the nanoparticles during the 5 h incubation, the bulk of all TiO2 nanoparticles, regardless of their crystalline modification, remained outside the cells. Various reports have described the damaging effects of TiO2 nanoparticles on cells after 24—48 h incubation: decrease in cell size, blebbing of cell plasma membrane, chromatin condensation, mitotic disturbances, apoptosis, DNA damage, micronucleus formation, and cell necrosis [Kang et al., 2008a; Liu S et al., 2010; Rahman et al, 2002; Zhao et al., 2009]. Based on our observations, these changes represent late cell reactions to TiO2 nanoparticles, involving complex cellular pathways. Oxygen reactive species generated by TiO2 nanoparticles have been referred to as the main inducers of cell damage [Chan Jin et al., 2010; Cheng-Yu et al., 2008; Fenoglio et al., 2009, and others], but these may not be the only mechanism of the interactions.



Our study has demonstrated that single TiO2 nanoparticles, 4—5 nm, directly contact the cell plasma membrane and that the nanoparticle’s crystalline modification determines the nature of the contact. This interaction suggests that nanoscale, non-organic compounds are able to directly influence cell functions. Our work provides the first direct evidence for earlier suggestions that the very small size of nanoparticles result in extremely close contact with cells and thus generate new types of cell-particle interaction with probable toxic consequences [Nel et al., 2006; Lewinski et al., 2008]. Amorphous, anatase and brookite single nanoparticles directly contact the cell plasma membrane. This contact may alter the function of the cell plasma membrane with a simple mechanical binding to membrane macromolecules. Our observations indicate that contact of brookite nanoparticles with the cell plasma membrane decreases the liquidity of the cell plasma membranes; they appear to “harden” and then lose the surface folds so that the membrane becomes unable to form invaginations. The brookite nanoparticles remain outside the cell.

Amorphous TiO2 nanoparticles probably increase the liquidity of cell plasma membrane, allowing it to readily form deep invaginations and providing penetration of the nanoparticles inside the cell. Formation of “coated” pits by anatase-treated cells suggests anatase particles are able to bind to macromolecules and induce cluster formation, which is necessary for clathrin-mediated endocytosis.

Another mechanism of cell plasma membrane alteration could be related to chemical reactions developing on the surface of TiO2 nanoparticles. Direct contact of single TiO2 nanoparticles with the cell plasma membrane provides sufficient distance for chemical interaction between the TiO2 and plasma membrane molecules. Molecular mechanisms of cell plasma membrane damage can be based on the ability of TiO2 nanoparticles to generate free radicals, including oxygenated free radicals HO2•, O2•, HO•, and carbon-centered radicals, causing cleavage of C-H bonds in organic molecules. These reactions can occur even in the dark and can serve as the first step of oxidative damage of biological molecules [Chen X. Mao, 2007; Fenoglio et al., 2009]. Different crystalline modifications could induce different chemical reactions, and so alter macromolecular structure of cell membrane differently.

We suggest that direct contact of single TiO2 nanoparticles with the cell plasma membrane represents the primary and general step of cell damage, operating for all crystalline modifications of nanoparticles that have sizes allowing such direct interaction. Changes in the water-ion balance in MDCK cells treated by TiO2 nanoparticles represent the first consequence of plasma membrane impairment by TiO2 nanoparticles and the start of alterations in other cellular pathways.

The application of electron microscopy to studies of ultrathin sections of MDCK cells incubated with TiO2 nanoparticles revealed the direct interaction of single amorphous, anatase and brookite nanoparticles with the cell plasma membrane and subsequently different cell responses to the treatment. The crystalline modification determines the pattern of cell plasma membrane reaction to the nanoparticles. Direct contact of the nanoparticles with cell plasma membrane is the primary and critical step of this interaction and defines the subsequent response of the cell.


This work was supported by Russian Ministry of Science and Education, Program “Basic Studies in Natural Sciences” Grant # 2.1.1/5642



Table 1. Cellular Damage as Function of TiO2 Nanoparticle Form . [go to text reference]

Nanoparticle Form
MDCK cell culture Vero cell culture
Incubation times
5 h
15 h
5 h
15 h
Control (untreated cells)



Number of trypan blue stained cells (%) in MDCK and Vero cell cultures treated with different crystalline modifications of titanium dioxide in concentration of 100 µg/mL.


Table 2. Cellular Damage as Function of TiO2 Nanoparticle CONCENTRATION [go to text reference]

Nanoparticle Form
Concentration of TiO2
Concentration of TiO2
Concentration of TiO2
Concentration of TiO2
20 µg/mL
100 µg/mL
Incubation time
3 h
5 h
3 h
5 h
Control (untreated cells)


Number of trypan blue stained cells (%) in MDCK cell culture treated with different concentrations of titanium dioxide crystalline modifications. The longer the incubation time, and the higher the nanoparticle concentration, the greater the amount of cellular damage.


Figure 1: Nanoparticle Forms


Figure 1: Images of TiO2 nanoparticles different crystalline modification in ultrathin sections. A - amorphous, B - anatase, C – brookite, D – rutile. Inserts in figure 1A show large magnification of amorphous nanoparticles. Ultrathin sections. Transmission electron microscopy. [go to text reference for figure 1]

Figure 2: Interaction of amorphous TiO<sub>2</sub> nanoparticles with MDCK culture cells.


Figure 2: Interaction of amorphous TiO2 nanoparticles with MDCK culture cells. Interaction of amorphous TiO2 nanoparticles with MDCK culture cells. I h incubation. A – nanoparticles fill folds and invaginations of a cell; B - E – direct contact of the nanoparticles with cell plasma membrane; F – invagination of plasma membrane containing nanoparticles; G, H – nanoparticles in “coated pits”, clathrin particles are visible on the pit cytoplasmic surface; I, J – nanoparticles in endosomes. Arrows show empty “coated pits”. Ultrathin sections. Transmission electron microscopy. [go to text reference for figure 2]


Figure 3: Disruption of the MAPK signaling cascade by LF leads to diminishing, and eventual absence, of the output signal, which corresponds with cell death.


Figure 3: Interaction of anatase TiO2 nanoparticles with MDCK culture cells. One hour incubation. A – nanoparticles fill folds and invaginations of a cell; B - C – direct contact of the nanoparticles with cell plasma membrane; D, E – nanoparticles in “coated pits” (shown by arrows), clathrin particles are visible on the pit cytoplasmic surface; F, G – nanoparticles in endosomes; H – K – osmiophilic membranes (are shown by arrows) in mitochondria and autophagosomes, arrowheads show the nanoparticles inside phagosome, white arrow shows electron dense spherical granule in mitochondria. Ultrathin sections. Transmission electron microscopy. [go to text reference for figure 3]


Figure 4: Interaction of brookite TiO2 nanoparticles with MDCK culture cells


Figure 4: Interaction of brookite TiO2 nanoparticles with MDCK culture cells. One hour incubation. A – nanoparticles (are shown by arrows) are located on cell surface; B, C – direct contact of the nanoparticles with cell plasma membrane; D – vacuolization of endoplasmic reticulum in a cell, electron lucent vacuoles are scattered in cytoplasm; E –swollen cell with electron lucent cytoplasm. Ultrathin sections. Transmission electron microscopy. [go to text reference for figure 4]


Figure 5: Interaction of brookite TiO2 nanoparticles with MDCK culture cells


Figure 5: Interaction of rutile TiO2 nanoparticles with MDCK culture cells. One hour incubation. A, B – aggregations of electron dense nanoparticles on the surface and in vacuoles; C, D, F – rutile aggregations on cell surface, plasma membrane is shown by arrows; E, G – rutile aggregations in cytoplasmic vacuoles. Ultrathin sections. Transmission electron microscopy. [go to text reference for figure 5]


Figure 6: Interaction of amorphous TiO2 nanoparticles with MDCK culture cells..


Figure 6: Interaction of amorphous TiO2 nanoparticles with MDCK culture cells. Five hour incubation. A – a cell showing swelling of cytoplasm and nucleus (small part of a cell with “normal” electron density is seen in right upper corner of the photo); swollen mitochondria are shown by arrows; phagosomes are shown by white asterisks. Nanoparticles are located between the cells. B – nanoparticles in phagosomes. Ultrathin sections. Transmission electron microscopy. [go to text reference figure 6]


Figure 7: Interaction of brookite TiO2 nanoparticles with MDCK culture cells.



Figure 7: Interaction of brookite TiO2 nanoparticles with MDCK culture cells One hour incubation. A, B – aggregations of electron dense nanoparticles on the surface and in vacuoles; C, D,F – rutile aggregations on cell surface, plasma membrane is shown by arrows; E, G – rutile aggregations in cytoplasmic vacuoles. Ultrathin sections. Transmission electron microscopy.[go to text reference figure 7]




ASTM E 2456-06 “Terminology for Nanotechnology.” ASTM International, 2006

Altman S, Randers L., Rao G. (1993) Comparison of trypan blue dye exclusion and fluorometric assays for mammalian cell viability determinations. Biotechnol Prog. 3, 671–674.

Chan J., Tang Y., Yang G.F., Li L.X., Xu S., Fan Y.X., Huang Y.Y., Yang Y.J. (2010) Cellular toxicity of TiO2 nanoparticles in anatase and rutile crystal phase. Biol Trace Elem Res 2010. DOI 10.1007/s12011-010-8707-0.

Cheng-Yu J., Zhu B.-Sh., Wang X.-F., Lu Q.-H. (2008) Cytotoxicity of titanium dioxide nanoparticles in mouse fibroblast cells. Chem Res Toxicol. 21, 1871–1877.

Chen X., Mao S.S. (2007) Titanium dioxide nanomaterials: synthesis, properties, forms, and applications. Chem Rev. 107, 2891-2959.

Fenoglio I., Greco G., Livraghi S., Fubini B. (2009) Non-UV-induced radical reactions at the surface of TiO2 nanoparticles that may trigger toxic responses. Chem Eur J. 15, 4614–4621.

Hamilton R.F.J., Wu N., Porter D., Buford M., Wolfarth M. Holian A. (2009) Particle length-dependent titanium dioxide nanomaterials toxicity and bioactivity. Particle&Fibre Toxicol. 6:35 DOI:10.1186/1743-8977-6-35.

Hussain S., Thomassen L.C.J., Ferecatu I., Borot M,-C., Andreau K., Martens J.A., Fleury J., Baeza-Squiban A., Marano F., Boland S. (2010) Carbon black and titanium dioxide nanoparticles elicit distinct apoptotic pathways in bronchial epithelial cells. Particle and Fibre Toxicology. 7:10.

Ismagilov Z.R., Tsykoza L.T., Shikina N.V., Zarytova V.F., Zinoviev V.V. Zagrebelnyi S.N. (2009) Synthesis and stabilization of nano-sized titanium dioxide. Russian Chemical Reviews. 78, 873-885.

Ismagilov Z.R., Shikina N.V., Tsykoza L.T., Tuzikov Ph.V., Ushakov V.A., Ischenko A.V., Ryabchikova E.I., Zarytova V.F., Netesova N.A., Zagrebel’nyi S.N. (2010) Synthesis and physicochemical study of nanosized titanium dioxide sols. J Colloid and Interface Science (in press).

Jovic M., Sharma M., Rahajeng J., Caplan S. (2010) The early endosome: a busy sorting station for proteins at the Crossroads. Histol Histopathol. 25, 99–112.

Kang J.L., Moon Ch., Lee H.Su, Lee H.W., Park E-Mi, Kim H.S., Castranova V. (2008a) Comparison of the biological activity between ultrafine and fine titanium dioxide particles in RAW 264.7 cells associated with oxidative stress. J Toxicol Environ Health, Part A. 71, 478-485.

Kang S.J., Kim B.M., Lee Y.J., Chung H.W. (2008b) Titanium dioxide nanoparticles trigger p53-Mediated damage response in peripheral blood lymphocytes. Environ Molec Mutagenesis. 49, 399-405.

L'Azou B., Jorly J., On D., Sellier E., Moisan F., Fleury-Feith J., Cambar J., Brochard P., Ohayon-Courtès C. (2008) In vitro effects of nanoparticles on renal cells. Particle&Fibre Toxicol., 5:22 doi:10.1186/1743-8977-5-22.

Lewinski N., Colvin V., Drezek R. (2008) Cytotoxicity of Nanoparticles. Small 4, 26–49.
Lindenschmidt R.C., Driscoll K.E., Perkins M.A., Higgins J.M., Mourer J.K., Belfiore K.A. (1990) The comparison of a fibrogenic and two non-fibrogenic dusts by bronchoalveolar lavage. Toxicol Appl Pharmacol. 102, 268-281.

Liu S., Xu L., Zhang T., Ren G., Yang Z. (2010) Oxidative stress and apoptosis induced by nanosized titanium dioxide in PC12 cells. Toxicology. 267, 172-177.

Mazurkova N.A., Spitsyna Yu.E., Shikina N.V., Ismagilov Z.R., Zagrebel’nyi S.N., Ryabchikova E.I. (2010) Interaction of titanium dioxide nanoparticles with influenza virus. Nanotechnologies in Russia. 5, 417-420.

Nel A., Xia T., Madler L., Li N. (2006) Toxic Potential of Materials at the Nanolevel. Science. 311, 622-627.

Ophus E.M., Rode L.E., Gylseth B., Nicholson G., Saeed K. (1979). Analysis of titanium pigments in human lung tissue. Scand J Work Environ Health. 5, 290–296.

Parkar N.S., Akpa B.S., Nitsche L.C., Wedgewood L.E., Place A.T., Sverdlov M.S., Chaga O., Minshall R.D. (2009) Vesicle formation and endocytosis: function, machinery, mechanisms, and modeling. Antioxid Redox Signal. 11, 1301–1312.

Rahman Q., Lohani M., Dopp E., Pemsel H., Jonas L., Weiss D.G., Schiffmann D. (2002) Evidence that ultrafine titanium dioxide induces micronuclei and apoptosis in syrian hamster embryo fibroblasts. Environ Health Perspect. 110, 797–800.

Stearns R.C., Paulauskis J.D., Godleski J.J. (2001) Endocytosis of Ultrafine Particles by A549 Cells. Am J Respir Cell Mol Biol. 24, 108–115.

Thakur Sh.A., Hamilton R.Jr., Pikkarainen T., Holian A. (2009) Differential binding of inorganic particles to MARCO. Toxicol Science. 107, 238–246. doi:10.1093/toxsci/kfn210.

Vevers W., Awadhesh N. (2008) Genotoxic and cytotoxic potential of titanium dioxide (TiO2) nanoparticles on fish cells in vitro. Ecotoxicology. 17, 410–420.

Zhao J., Bowman L., Zhang X., Vallyathan V., Young Sh.-H., Castranova V., Ding M. (2009) Titanium Dioxide (TiO2) Nanoparticles induce JB6 cell apoptosis through activation of the caspase-8/Bid and mitochondrial pathways. J Toxicol Environ Health, Part A. 72, 1141–1149.

Zhang A.P., Sun Y.P. (2004) Photocatalytic killing effect of TiO2 nanoparticles on Ls-174-t human colon carcinoma cells. World J Gastroenterol. 10, 3191-3193.

Zhou H., Imrich A., Kobzik L. (2008) Characterization of immortalized MARCO and SR-AI/II-deficient murine alveolar macrophage cell lines. Particle&Fibre Toxicol. 5, doi:10.1186/1743-8977-5-7.