Real-Time Translocation of Fullerene Reveals Cell Contraction

full papers

1986

Fullerenes

DOI: 10.1002/smll.200701279

Real-Time Translocation of Fullerene Reveals CellContractionEmppu Salonen,* Sijie Lin, Michelle L. Reid, Marcus Allegood, Xi Wang,Apparao M. Rao, Ilpo Vattulainen,* and Pu Chun Ke*

Keywords:� aggregation

� fluorescence

� fullerenes

� nanoparticles

Carbon-based nanomaterials possess unique structural, mechanical, and

electronic properties that are exploited in numerous applications. The fate of

nanomaterials in living systems and in the environment is largely unknown,

though there is a reason for concern. Here it is shown how the interaction of

fullerene with natural phenolic acid induces cell contraction. This

phenomenon has a general applicability to carbon-based nanomaterials

interacting with natural amphiphiles. Atomistic simulations reveal that the

self-assembly of C70–gallic acid (GA) favors aggregation. Confocal fluo-

rescence microscopy shows that C70–GA complexes translocate across the

membranes of HT-29 cells and enter nuclear membranes. Confocal imaging

further reveals the real-time uptake of C70–GA and the consequent con-

traction of the cell membranes. This contraction is attributed to the

aggregation of nanoparticles into microsized particles promoted by cell

surfaces, a new physical mechanism for deciphering nanotoxicity.

[�] Dr. E. Salonen, Prof. I. Vattulainen

Laboratory of Physics and Helsinki Institute of Physics

Helsinki University of Technology

P.O. Box 1100, 02015 TKK (Finland)

Fax: (þ358) 9-451-3116

E-mail: [email protected]; [email protected]

Prof. I. Vattulainen

Department of Physics, Tampere University of Technology

P.O. Box 692, 33101 Tampere (Finland)

Fax: (þ358) 3-3115-2600

Prof. I. Vattulainen

MEMPHYS – Centre for Biomembrane Physics

Department of Physics

University of Southern Denmark

Odense (Denmark)

S. Lin, M. L. Reid, M. Allegood, Dr. X. Wang, Prof. A. M. Rao

Prof. P. C. Ke

Department of Physics and Astronomy

Clemson University, Clemson

SC 29634 (USA)

Fax: (þ1)-864-656-0805

E-mail: [email protected]

: Supporting Information is available on the WWW under http://www.small-journal.com or from the author.

� 2008 Wiley-VCH Verlag GmbH & Co

1. Introduction

The progress in nanotechnology in recent years has

revolutionized modern science and technology and, in the

meantime, provoked calls for research into the potential

adverse effects of nanomaterials.[1–7] Of particular concern is

the very real possibility that nanomaterials produced in

research laboratories, and associated with consumer products,

will eventually be discharged into the environment. Yet the

transport and transformation of nanomaterials therein is

largely unknown.[4]

Due to their mutual van der Waals interaction, carbon

nanoparticles readily aggregate and therefore are not usually

considered as potential contaminants in the aqueous phase.

However, carbon nanomaterials in living systems and

discharged in the environment may well be functionalized

or derivatized by biomolecules and natural organic matter[8] to

elicit stable suspensions. To understand these effects, we have

considered fullerene C70 interacting with phenolic compounds

that are ubiquitous in fruits and all plant ecological systems. In

addition to possessing significant therapeutic properties,[9]

phenolic acids represent an ideal class of model molecules for

understanding the bioavailability and the transformations of

nanomaterials in nature. To this end, we have suspended

fullerene C70 in aqueous solution using phenol gallic acid (GA,

. KGaA, Weinheim small 2008, 4, No. 11, 1986–1992

Real-Time Translocation of Fullerene Reveals Cell Contraction

Figure 1. Absorbance, fluorescence, and TEM characterizations of C70–

GA assembly. a) Absorption spectrum of GA (for structure see inset). b)

UV absorbance of C70–GA at a weight ratio of 1:5, recorded at 384 nm.

The concentrations of the C70–GA suspensions were derived from a

calibration using aromatic solvent dichlorobenzene (see Supporting

Information). c) Fluorescence spectra of C70–GA at weight ratios of 1:10

(blue) and 1:20 (orange). The fluorescence of GA of equal amount to that

in C70–GA is negligible (green). Excitation wavelength was 488nm. d)

TEM image of C70–GA filtered and dialyzed consecutively (1:5 weight

ratio).

small 2008, 4, No. 11, 1986–1992 � 2008 Wiley-VCH Verlag Gmb

see Figure 1a inset). Consequently, we have probed the real-

time interaction of C70 and GA using HT-29 cell line and

complemented the experimental studies with detailed atom-

istic simulations. Our combined experimental and in silico

studies have shown the rapid assembly of C70 mediated by GA

and the aggregation of these nanoparticles prompted by cell

membranes. Such physical transformations of nanomaterials

may be connected with nanotoxicity and have important

consequences for biological systems and the environment.

2. Results and Discussion

2.1. Solubilization of C70 by GA

C70 and GA were mixed in Milli-Q water, probe-sonicated

and stored at room temperature for 3 days to allow the C70 to

fully interact with the GA. Within 2 days of incubation the

mixtures of C70–GA turned into stable homogenous suspen-

sions with a green tint. The samples were centrifuged and

filtered to remove large clumps of C70 particles and free GA.

The UV absorbance of C70–GA correlated linearly with the

concentration until a plateau was reached at a value of 0.65 mg

mL�1, indicating the saturation of C70 solubility (Figure 1b).

Distinct fluorescence peaks were observed at 540 and

719 nm when C70–GA suspensions were excited at 488 and

690 nm, respectively (Figure 1c). In contrast, an equal amount

of GA alone yielded minimal fluorescence in the same

wavelength range (green curve in Figure 1c). This confirms

that the fluorescence peaks at 540 and 719 nm were emitted by

C70–GA. The fluorescence intensity at 540 nm is stronger for

the weight ratio of 1:20 than that of 1:10, possibly because of

improved C70 solubility. Centrifugation at 7500 rpm for 3 min

markedly reduced the narrow peak at 719 nm, but did not

significantly alter the broad fluorescence peak at 540 nm,

suggesting that the fluorescence at 719 nm was emitted by less

dissolved C70. This result is reminiscent of that observed for

C60-pyridine suspensions, where fluorescence peaks at 440,

575, and 700 nm were assigned to C60 nanoparticles, C60 lace-

like clusters, and C60 microbulks, respectively.[10] However,

the fluorescence bands of C60 in GA were weaker and less

distinctive than those observed for C70, possibly because of the

greater symmetry and broader fluorescence emission of C60,

which extends into the near-infrared region.[11,12]

Transmission electron microscopy (TEM) imaging showed

filtered C70–GA complexes of an average diameter of �20 nm,

with each complex encasing multiple C70 molecules

(Figure 1d). In addition, prior to the filtrations, ‘‘lace-like’’

clusters were observed. The sizes of these clusters ranged from

tens of nanometers up to micrometers, similar to that observed

for pyridine-solubilized C60.[10] A separate NMR study found

no spectral signatures for C70–GA, indicating their binding to

be of a noncovalent nature.

2.2. Atomistic Molecular Dynamics Simulations ofC70–GA Aggregation

The atomistic aspects of the suspension of fullerene by GA

in aqueous solution and the resulting formation of C70–GA

complexes were elucidated by atomistic molecular dynamics

H & Co. KGaA, Weinheim www.small-journal.com 1987

full papers E. Salonen, P. C. Ke, et al.

Figure 3. a) Normalized PDF versus distance between the CM of the C70 cluster and the CM of GA. The two

peaks corresponding to the GA surface layer (up to about r¼1.84 nm) result from the C70 cluster surface

roughness (cf. Figure 2). Inset: number of GA molecules in the C70 cluster surface layer in time. b) GA order

parameter P2 versus the distance between the CM of a GA molecule and the CM of the nearest C70 molecule.

Statistical margins of error are shown for a few data points to demonstrate typical error bars.

1988

(MD) simulations. The simula-

tions focused on two aspects of

C70–GA interaction: coating of

a C70 cluster by GA and

suspension of single C70 mole-

cules, followed by the cluster-

ing of such C70–GA aggregates.

2.2.1. Icosahedral C70

Cluster

We first considered a system

consisting of an icosahedrally

arranged cluster of 13 neutral

C70 molecules, 64 anionic GA

molecules (pKa¼ 4.4), and 64

Naþ counterions dispersed in

water. The icosahedral arrange-

ment of the fullerenes was

chosen due to its known stabi-

lity for C60 in aqueous solu-

tion.[13] The radius of the C70 cluster was approximately

1.35 nm, as measured from the center of mass (CM) of the

cluster to the outermost carbon atoms. However, the arrange-

ment of carbon atoms in the cluster could not be viewed as

strictly spherical. Instead, the cluster had appreciable surface

roughness, as is also evidenced by the results for GA adsorption

(see below).

The GA molecules rapidly adsorbed onto the surface of

the fullerene cluster (see Figure 2). The aggregation of the GA

was quantified by calculating the C70–GA CM pair distribution

function (PDF), see Figure 3a. The number of GA molecules

in the surface layer was obtained by defining, on the basis of

the PDF, any molecule closer than 1.84 nm from the C70 cluster

CM as adsorbed. The same values could also be obtained by

using a minimum distance criterion of 0.88 nm (again, on

the basis of C70–GA PDF) from any C70 molecule CM in the

cluster. After equilibration of the system over 16.5 ns, the

Figure 2. Icosahedral C70 cluster (light green), caged by GA molecules

(blue) in the atomistic MD simulations. Water, counterions, and some

GA molecules have been omitted from the picture for clarity.

www.small-journal.com � 2008 Wiley-VCH Verlag Gm

number of GA molecules adsorbed on the fullerene cluster

surface reached an average value of 28� 2 (cf. inset of

Figure 3a). A similar value, 22� 3, was also observed in a

simulation with only 36 GA molecules. This suggests that the

number of GA molecules caging the fullerenes was close to the

saturation limit.

The simulations revealed that the formation of the GA

surface layer resulted from two types of interaction: hydrogen

bonding between the GA molecules (on average 0.49� 0.14

hydrogen bonds per adsorbed molecule, see Supporting

Information), as well as C70–GA van der Waals attractions

that were enhanced by a parallelorientation of the GA aromatic

ring with the fullerene surface. The orientational order of GA

was quantified by a second-order Legendre polynomial,

P2 ¼ 1=2ðh3 cos2 ui � 1Þ, where u is the angle between the

GA aromatic ring normal and the vector pointing from the GA

CM to the CM of the nearest C70 in the cluster. The brackets

denote an average over time and all the GA molecules. A value

of P2¼ 1 means that all GA molecules are aligned with their

aromatic ring normal pointing to the nearest fullerene. In turn,

P2¼ 0 means that the GA molecules have completely random

orientations, and P2< 0 reflects a situation where the aromatic

ring plane of GA is perpendicular to the C70 surface.

Figure 3b shows the order parameter P2 as a function of

distance from the CM of the nearest C70 in the cluster. The

ordering of the GA molecules in the surface layer is

profoundly strong. The GA molecules facing C70 lie along

the fullerene cluster surface while, interestingly, right above

the surface layer (at 0.8–1.0 nm) there is a thin region where

the GA molecules assume perpendicular orientations. This is

followed further by another ordered layer of GA molecules at

a separation of about 1.1 nm, stabilized by a stacking of the

aromatic ring structures of the adjacent GA. Finally, at

distances larger than �1.4 nm from the nearest C70, the GA

molecules assume random orientations.

In Figure 4, we provide a complementary view of the

ordering by plotting the same data as a function of the distance

from the total C70 cluster CM. While perhaps corresponding to

bH & Co. KGaA, Weinheim small 2008, 4, No. 11, 1986–1992


Figure 4. GA aromatic ring order parameter P2 as a function of distance

from the C70 cluster CM. The statistical margins of error are shown for

some data points, reflecting the overall statistical uncertainty in the

analysis. An exception is the value of �0.65 at the shortest separation,

resulting from a single data point. Inset: Wireframe image of the C70cluster. The distance from the CM of the cluster to the outermost carbon

atoms is �1.35 nm.

Figure 5. Aggregates of several GA-caged C70 molecules resulting from

the MD simulations.

Figure 6. C70–C70 (CM–CM) PDFs for different time intervals in the

simulation with several solvated C70–GA complexes. The time evolution

of the distribution functions highlights how the aggregate grows in size.

a more straightforward approach in the analysis, it is clear that

the plot in Figure 4 does not appreciably reflect the significant

surface roughness of the C70 cluster. The type of orientational

arrangement observed for the GA surface layer corroborates

the view that p-stacking is important in the formation of

aggregates of aromatic molecules and carbon nanomater-

ials.[14,15]

2.2.2. C70 Suspension and Aggregation by GA

MD simulations were also used to complement the above-

discussed TEM studies, which showed significant clustering of

C70–GA complexes. We followed the self-assembly of such

complexes to elucidate the mechanisms associated with this

nonequilibrium process. We first considered the self-assembly

in a solution containing an initially random distribution of GA

molecules, a single C70 molecule and counterions. As with the

larger aggregate (see above), the C70 was quickly caged by

5.2� 0.7 GA molecules (see Supporting Information), indicat-

ing the stability of the complex. Moreover, ordering of GA on

the fullerene surface was similar to the case of the icosahedral

C70 cluster.

When 13 randomly chosen C70–GA complexes from this

simulation were immersed in the same aqueous environment,

the isolated C70–GA complexes were observed to sponta-

neously form small aggregates, glued together by GA

molecules between the fullerenes, see Figures 5, 6, and 7a.

Figure 6 in particular shows how the fullerene aggregates

increase in size as the system evolves toward the equilibrium

state. This type of binding of GA is similar to that suggested to

take place in the uptake of polycyclic aromatic hydrocarbons

by fullerene aggregates.[16] The role played by GA is

significant. Despite the entropic loss, GA molecules are an

integral part of the aggregates, highlighting the importance of

the hydrogen bond network and van der Waals interactions

driving the self-assembly. Although only about a half of the


GA originally coating the dispersed fullerenes remained on

the aggregated C70, the retained amount of GA is substantial.

GA molecules coat the outer surface of C70 aggregates (cf.

Figures 1d and 5), besides which the internal cavities in the

aggregates also contain significant amounts of GA. Particu-

larly, the GA molecules inside the aggregates are strongly

ordered, with their ring normal pointing to the CM of the

nearest C70 (see Figure 7b). What is more, a comparison of

Figures 3b and 7b demonstrates that the binding of GA is

largely similar on an icosahedral C70 cluster and in an

aggregate comprised of a number of individual C70 caged by

GA molecules. This highlights the prominent role of GA in

stabilizing the clusters over a multitude of sizes.

2.3. Confocal Imaging of C70–GA ComplexesInteracting with HT-29 Cells

To understand the impact of nanomaterials on living

systems, we detected the uptake of C70–GA complexes across

HT-29 cells. Cells seeded in an 8-well sample chamber glass



Figure 7. a) C70–GA PDF and the number of GAmolecules adsorbed on the aggregated C70molecules (inset). b) GA aromatic ring order parameter P2 as a function of distance from

the CM of the nearest C70 molecule. Note the similarity of GA ordering on individual C70molecules here and on the C70 molecules in an icosahedral cluster (Figure 3b).

1990

were excited at 488 nm and their cross sections were imaged by

scanning from the front cell surfaces to the back using confocal

fluorescence microscopy. The presence of C70–GA is ascer-

tained by their green fluorescence, and microsized particles

appear as brighter green spots in Figure 8 (indicated by white

arrows). The uptake of C70–GA complexes was pronounced

and rapid across the cell membranes. Our imaging further

suggests that C70–GA nanoparticles also penetrated into

nuclear membranes (orange arrows in Figure 8c), allowing

them to agglomerate into microsized particles therein.

To obtain information on the real-time translocation of

nanoparticles across cell membranes, HT-29 cells were labeled

with lipophilic DiIC18 (Ex/Em: 644/665 nm, Molecular Probes)

which turned the cell membranes into red fluorescent ‘‘rings’’

surrounding the cell interiors (Figure 9). Imaging was

performed using confocal fluorescence microscopy after

C70–GA of 0.3 mg mL�1 was added to the sample chamber

glass. The sequential images in Figure 9 clearly indicate a

remarkable conformational change of the cells resulting from

the addition of C70–GA. Furthermore, the cells became

increasingly irregular over time because of the clumping of

C70–GA nanoparticles into microparticles on the membrane

surfaces (Figure 9).

The formation of microsized particles was also found inside

cell membranes (Figure 9), although the aggregation therein

presumably proceeded at a slower pace due to the reduced

concentration of nanoparticles in the intracellular space. The

above simulation results propose that the aggregation of

Figure 8. Translocation of C70–GA across HT-29 cell membranes. Confocal sectioning images

of green fluorescent C70–GA (weight ratio 1:5). The dark circles (pointed by orange arrows in [

aggregates are visible in all images (see white arrows).

www.small-journal.com � 2008 Wiley-VCH Verlag GmbH & Co. KGaA, We

nanoparticles is mediated by van der Waals

interactions between the exposed hydrophobic

C70 moieties, and promoted by the higher

affinity of cell membranes for small hydro-

phobic particles. The number of microsized

particles was tabulated for bright-field images

(see Supporting Information) recorded 3 and

48 min after the initial incubation (Figure 9, top

right panel). Due to the diffraction limit, only

particlesover0.4mmindiameterwere counted.

Consequently the cells were found to shrink by

an average of 24.3(�0.6)% in diameter after

48 min of incubation with the nanoparticles

(Figure9 bottom rightpanel).This corresponds

to a change of more than 50% in intracellular

volume, implying substantial water leakage to

avoid lysis. However, such contraction was not

observed for GA alone, or when C70–GA concentration was

reduced from 0.3 to 0.15 or 0.08 mg mL�1. At these lower

concentrations only deposition of particles onto cell mem-

branes was observed (see Supporting Information). This is

likely due to the increased average distance and therefore

reduced interactions between C70–GA nanoparticles.

To further investigate C70–GA induced cytotoxicity on

HT-29 cells, MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxy-

methoxyphenyl)-2-(4-sulfophenyl)-2H-tetazolium) assay for

cellular metabolism was performed on cultures without

C70–GA (control) and with C70–GA of various concentrations

(Figure 10). The pH value of C70–GA solution was first

adjusted to 7.4 and autoclaved to exclude artifacts before

adding to cell cultures. At a concentration of 7 mg L�1 of

C70–GA, the absorbance decreased from 1.847 (control) to

0.713, suggesting significant mortality. At concentrations

above 14 mg L�1, the absorbance dropped even further to

0.23–0.25, indicating severe cytotoxicity. This assay is

consistent with our imaging observations.

3. Conclusions

In summary, we have discovered that cell membranes

promote a remarkable adsorption and aggregation of

suspended fullerene nanoparticles. Such a physical transfor-

mation is nonspecific in nature and is expected to occur

prevalently for carbon-based nanomaterials interacting with

amphiphilic and aromatic amino acids, peptides, proteins and

obtained along the optical axis to confirm the uptake

c]) could be indicative of nuclear membranes. C70–GA

inheim small 2008, 4, No. 11, 1986–1992


Figure 9. Real-time interaction of C70–GA and HT-29 cells. The cell membranes were labeled with lipophilic DiIC18 and their cross sections

appeared as red ‘‘rings.’’ Over time the cells were mechanically contracted due to the mutual interactions between C70–GA nanoparticles (weight

ratio 1:5).

Figure 10. MTS assay on HT-29 cell absorbance versus C70–GA

concentration (056mg L�1). Acute cytotoxicity is shown for C70–GA of

14mg L�1 and above.

nucleic acids that are abundant in biological systems and in the

ecological environment. Our future work will examine these

important aspects in detail. The contraction of cell structure at

high concentrations of C70–GA perturbs membrane elasticity,

affecting ion transport, membrane fluidity and giving rise to

cell stress. At the very least, it impacts upon cell function since

the activation and hence functions of a major fraction of

membrane proteins are sensitive to the elastic properties of

membranes.[17,18] The effects will eventually cause cytotoxicity

and cell death but may also be utilized for drug delivery and

nanomedicine.[19–21]

4. Experimental Section

C70 suspension: C70 (Nano-C, Inc.) and GA (Sigma) of weight

ratios 1:5, 1:10, and 1:20 were mixed in Milli-Q water. The

molecular weights of C70 and GA are 840 and 170 g molS1,


corresponding to molar ratios of approximately 1:25–1:100. The

mixtures of C70–GA were probe-sonicated (VC 130 PB, Sonics &

Materials) at 8W for 30min and then placed at room temperature

for 3 days. C70–GA suspensions were centrifuged at 7500 rpm for

3min to remove large C70 aggregates. The samples were then

filtered out through Microcon (Millipore, MWCO 3000Da) to

remove large clumps of C70 particles and dialyzed for 12 h using

DispoDialyzer filters (Spectrumlabs, MWCO 500Da) to remove free

GA. To determine the concentration of the suspensions, each

sample was dried using SpeedVac. The pellet was then dissolved

in aromatic solvent 1,2-dichlorobenzene. The unsuspended

nanoparticles separated from the suspended ones and entered

the organic phase. The quantity of the unsuspended nanoparticles

was determined using a precalibrated absorbance curve for C70 in

dichlorobenzene (see SI).

Absorbance and fluorescence measurements: The absorption

spectrum of GA was measured at room temperature using a

spectrophotometer (Biomate 3). The UV absorbance of C70–GA

(1:5 weight ratio) versus concentration was read at 384 nm where

GA had a minimal contribution. The absorbance correlated linearly

with concentration until it reached a plateau at �0.65mg mLS1.

The fluorescence of C70–GA was measured for weight ratios of

1:10 and 1:20 using the emission scan function of a spectro-

fluorometer (QM-8/2005, PTI). The green fluorescence at 540 nm

of C70–GA was excited at 488 nm and the fluorescence at 719 nm

was excited at 690 nm. GA dissolved in Milli-Q water showed

minimal fluorescence over 515–800 nm.

Cell culture: HT-29 colon cancer cells were cultured in DMEM

with 1% penicillin streptomycin, 1% sodium pyruvate, and 10%

fetal bovine serum. Approximately 1000 HT-29 cells were seeded

in each (200mL) of an eight-chambered well glass slide and

allowed to attach overnight at 37 -C with 5% CO2.

Confocal fluorescence imaging: For the translocation measure-

ment, a laser of 488 nm was used as the excitation source after

1 h incubation of HT-29 cells with C70–GA. The cell cross sections



1992

were scanned over 20 mm from the front to the back surfaces of

the cells along the optical axis. For the cell contraction

experiment, HT-29 cells were first labeled with lipophilic DiIC18(1,1(-dioctadecyl-3,3,3(,3(-tetramethylindocarbocyanine perchlo-

rate; Ex/Em: 644/665 nm, Molecular Probes). Approximately

100 mL of 5% w/v DiIC18 was added into each well of a chamber

glass slide and incubated with cells for 1 h at 37 -C with 5% CO2.

Imaging was conducted using a confocal fluorescence microscope

(LSM510, Zeiss) with a He–Ne laser of 633 nm as the excitation

source. The initial frame of the image series was recorded �3min

after 150 mL C70–GA suspension (1:5 ratio) was added into the

chamber well. The final concentration of C70–GA in each chamber

glass was estimated to be 0.3mg mLS1 (pH¼7.4). The time lapse

between each consecutive image was 3min.

Atomistic molecular dynamics simulations: The MD simula-

tions were carried out with the GROMACS package, version

3.3.1.[22] We employed the Gromos 53a6 force field,[23] which has

been parameterized with calculations of the free energy of

hydration for various small biomolecules. Hence, the force field

is especially suitable for our simulations of C70 and GA in aqueous

solution. For water we employed the simple point charge (SPC)

model.[24] The standard twin-range scheme for nonbonded

interactions with cutoff values of 0.9 and 1.4 nm was used. All

the simulations were run under constant NpT conditions,

maintained by the weak-coupling thermostat and barostat by

Berendsen et al.[25] Particle Mesh Ewald (PME) method was used

to calculate the electrostatic interactions in the system.[26] All

bond lengths were constrained by the LINCS algorithm.[27] The

time step used in the simulations was 2 fs. The simulation

methodology followed largely our previous work.[28,29] For further

details on the set up of our model systems, see the Supporting

Information.

Acknowledgements

PCK acknowledges the support of NSF grant no. CBET-0736037

and an NSF Career award. We thank Joan Hudson for TEM

imaging. ES and IV acknowledge the support of the Academy of

Finland. Computational resources at the Finnish IT Center for

Science and the Horseshoe cluster at the University of Southern

Denmark are gratefully acknowledged.

[1] V. L. Colvin, Nat. Biotechnol. 2003, 21, 1166.[2] N. Gharbi, M. Pressac, M. Hadchouel, H. Szwarc, S. R. Wilson, F.

Moussa, Nano Lett. 2005, 5, 2578.

www.small-journal.com � 2008 Wiley-VCH Verlag Gm

[3] G. Oberdoerster, E. Oberdoerster, J. Oberdoerster, Envron. Health

Perspect. 2005, 113, 823.[4] A. D. Maynard, R. J. Aitken, T. Butz, V. Colvin, K. Donaldson, G.

Oberdorster, M. A. Philbert, J. Ryan, A. Seaton, V. Stone, S. S.

Tinkle, L. Tran, N. J. Walker, D. B. Warheit, Nature 2006, 444, 267.[5] A. Nel, T. Xia, L. Madler, N. Li, Science 2006, 311, 622.[6] C. M. Sayes, J. D. Fortner, W. Guo, D. Lyon, A. M. Boyd, K. D.

Ausman, Y. J. Tao, B. Sitharaman, L. J. Wilson, J. B. Hughes, J. L.

West, V. L. Colvin, Nano Lett. 2004, 4, 1881.[7] P. C. Ke, R. Qiao, J. Phys: Condens. Matter 2007, 19, 373101.[8] H. Hyung, J. D. Fortner, J. B. Hughes, J. H. Kim, Environ. Sci. Technol.

2007, 41, 179.[9] N. M. Gandhi, C. K. K. Nair, Mol. Cell. Biochem. 2005, 278, 111.[10] F. Zhang, Y. Fang, J. Phys. Chem. B 2006, 110, 9022.[11] B. Ma, Y. P. Sun, J. Chem. Soc, Perkin Trans. 1996, 2, 2157.[12] A. Graja, J. P. Farges, Adv. Mater. Opt. Electron. 1998, 8, 215.[13] G. V. Andrievsky, V. K. Klochkov, A. B. Bordyuh, G. I. Dovbeshko,

Chem. Phys. Lett. 2002, 364, 8.[14] S. Gotovac, H. Honda, Y. Hattori, K. Takahashi, H. Kanoh, K.

Kaneko, Nano Lett. 2007, 7, 583.[15] F. Tournus, S. Latil, M. I. Heggie, J. C. Charlier, Phys. Rev. B 2005,

72, 075431.

[16] K. Yang, B. Xing, Environ. Pollut. 2007, 145, 529.[17] T. J. McIntosh, S. A. Simon, Annu. Rev. Biophys. Biomol. Struct.

2006, 35, 177.[18] J. A. Lundbaeck, J. Phys. : Condens. Matter 2006, 18, S1305.[19] D. Pantarotto, J. Briand,M. Prato, A. Bianco, Chem. Commun. 2004,

16.

[20] N. W. S. Kam, M. O’Connell, J. A. Wisdom, H. J. Dai, Proc. Natl. Acad.

Sci. USA 2005, 102, 11600.[21] Q. Lu, J. M. Moore, G. Huang, A. S. Mount, A. M. Rao, L. L. Larcom, P.

C. Ke, Nano Lett. 2004, 4, 2473.[22] a) E. Lindahl, B. Hess, D. van der Spoel, J. Mol. Mod. 2001, 7, 306;

b) D. van der Spoel, E. Lindahl, B. Hess, G. Groenhof, A. E. Mark, H.

J. C. Berendsen, J. Comp. Chem. 2005, 26, 1701.[23] C. Oostenbrink, A. Villa, A. E. Mark, W. F. van Gunsteren, J. Comp.

Chem. 2004, 25, 1656.[24] H. J. C. Berendsen, J. P. M. Postma, W. F. van Gunsteren, J.

Hermans, Intermolecular Forces (Ed: B. Pullman), Reidel, Dor-

drecht 1981, p. 331.[25] H. J. C. Berendsen, J. P. M. Postma, W. F. van Gunsteren, A. Di Nola,

J. R. Haak, J. Chem. Phys. 1984, 81, 3684.[26] a) T. Darden, D. York, L. Pedersen, J. Chem. Phys. 1993, 98, 10089;

b) U. Essman, L. Perera, M. L. Berkowitz, T. Darden, H. Lee, L. G.

Pedersen, J. Chem. Phys. 1995, 103, 8577.[27] B. Hess, H. Bekker, H. J. C. Berendsen, J. Fraaije, J. Comp. Chem.

1997, 18, 1463.[28] S. Vainio, M. Jansen, M. Koivusalo, T. Rog, M. Karttunen, I.

Vattulainen, E. Ikonen, J. Biol. Chem. 2006, 281, 348.[29] P. Niemela, S. Ollila, M. T. Hyvonen, M. Karttunen, I. Vattulainen,

PLoS Comput. Biol. 2007, 3, 304.

bH & Co. KGaA, Weinheim

Received: December 18, 2007Revised: April 4, 2008Published online: October 23, 2008

small 2008, 4, No. 11, 1986–1992

Real-Time Translocation of Fullerene Reveals Cell Contraction

Documents