journal of the mechanical behavior of biomedical materials 49 (2015) 105–111

Available online at www.sciencedirect.com

www.elsevier.com/locate/jmbbm

Research Paper

Stiffness of cancer cells measured with an AFM

indentation method

Kozaburo Hayashia,b,c,n,1, Mayumi Iwataa

a
Department of Biomedical Engineering, School of Engineering, Okayama University of Science, 1-1 Ridai-cho, Kita-ku,
Okayama 700-0001, Japan
b
Department of Biomedical Engineering, College of Engineering, National Cheng Kung University, Tainan, Taiwan
c
Division of Bioengineering, Department of Mechanical Science and Bioengineering, Graduate School of Engineering
Science, Osaka University, Toyonaka, Japan

art i cle i nfo ab st rac t

Article history:
Received 29 January 2015
Received in revised form
28 April 2015
Accepted 30 April 2015
Available online 13 May 2015
The stiffness of cancer cells and its changes during metastasis are very important for
understanding the pathophysiology of cancer cells and the mechanisms of metastasis of
cancer. As the first step of the studies on the mechanics of cancer cells during metastasis,
we determined the elasticity and stiffness of cancer cells with an indentation method
using an atomic force microscope (AFM), and compared with those of normal cells. In most
of the past AFM studies, Young's elastic moduli of cells have been calculated from force-

Keywords: indentation data using Hertzian model. As this model is based on several important

Atomic force microscopy (AFM) assumptions including infinitesimal strain and Hooke's linear stress–strain law, in the

Indentation test exact sense it cannot be applied to cells that deform very largely and nonlinearly. To

Cancer cell overcome this problem, we previously proposed an equation F¼a[exp(bδ) 1] to describe

Stiffness relations between force (F) and indentation (δ), where a and b are parameters relating with

Young's elastic modulus
cellular stiffness. In the present study, we applied this method to cancer cells instead of
Young's elastic modulus.
The conclusions obtained are: 1) AFM indentation test data of cancer cells can be very
well described by the above equation, 2) cancer cells are softer than normal cells, and
3) there are no significant locational differences in the stiffness of cancer cells between the
central and the peripheral regions. These methods and results are useful for studying the
mechanics of cancer cells and the mechanisms of metastasis.
& 2015 Elsevier Ltd. All rights reserved.

n
Correspondence to: Suite 602, 2-3-45 Kamokogahara, Higashinada-ku, Kobe 658-0064 Japan.
E-mail addresses: kozaburohayashi@gmail.com (K. Hayashi), pqqdg359@ybb.ne.jp, s-09.06m_t@softbank.ne.jp (M. Iwata).
1
Present Address: Osaka University (Professor Emeritus), Graduate School of Engineering, 1 Machikaneyama-cho, Toyonaka 560-
0043, Japan.

http://dx.doi.org/10.1016/j.jmbbm.2015.04.030
1751-6161/& 2015 Elsevier Ltd. All rights reserved.
106 journal of the mechanical behavior of biomedical materials 49 (2015) 105 –111
1. Introduction
The stiffness of cancer cells and its changes during the
metastatic process are very important for the pathophysiol-
ogy of cancer cells and the mechanisms of the metastasis of
cancer. However, few studies have been done on the mechan-
ical properties of cancer cells during the metastatic process,
although there are many reports on the stiffness of cancer
cells (Cross et al., 2007; Pogoda et al., 2012; Suresh, 2007).
As the first step of the studies on the changes of mechan-
ical properties of cancer cells during metastasis, we deter-
mined the stiffness and elasticity of non-metastatic, primary
cancer cells with an indentation method using atomic force
microscopy (AFM), and compared with those of normal cells.
Several studies have been done on this subject mostly using
AFM indentation tests. For example, Cross et al. (2007)
measured Young's moduli of live metastatic cancer cells
taken from pleural effusions of patients, and found that they
are much larger than those of benign cells obtained from the
same patients. Using the same technique, Li et al. (2008) and
Nikkhah et al. (2011) showed that malignant breast epithelial
cells have significantly lower Young's moduli than their non-
malignant counterparts. Here, Nikkhah et al. (2011) obtained
the results from the indentation depth of less than 200 nm.
On the other hand, Pogoda et al. (2012) found no difference in
Young's moduli between cancerous and benign melanoma
cells at the indentation depth of 200 nm, although the moduli
of the former cells were significantly smaller than those of
the latter ones at the depth of 1400 nm. Like these, the results
are various, and the mechanics of cancer cells has not been
fully understood.
As stated above, in most of the past AFM studies, Young's
elastic moduli have been calculated from force-indentation
data using Hertzian model. This model is based on several
important assumptions including infinitesimal strain and
Hooke's linear stress–strain law. In the exact sense, however,
these assumptions cannot be applied to cells that deform
very largely and nonlinearly. Moreover, many of the data
have been obtained from specific indentation depth in spite
of the inhomogeneous structure of cells. To overcome these
problems, we proposed stiffness parameters obtained from a
formula describing AFM force-indentation relations that
cover a wide range of depth (Miyazaki and Hayashi, 1999).
If we use this method, we do not need to make the assu-
mptions of infinitesimal strain and linear stress–strain
relationship.
Cell structure is heterogeneous, and changes in response
to many factors including mechanical stimulation and mor-
bid state. For example, mechanical stress increases the
density of such microfilaments as actin filaments (stress
fibers) and also rearranges the intracellular microfilament
system (Kataoka et al., 1998; Sumpio et al., 1988). Therefore, it
is highly likely that the transformation into cancer cells and
the advance of carcinogenesis rearrange intracellular struc-
tures and develop non-uniform distributions of microfila-
ments. If so, cellular stiffness may be different depending
upon intracellular locations.
Therefore, the purposes of the present study are: 1) to
represent the stiffness of cancer cells with our stiffness
parameters instead of Young's elastic modulus, 2) to compare
stiffness between cancer and normal cells, 3) to see locational
differences in the stiffness of cancer cells, and 4) to discuss
the difference between our stiffness parameters and Young's
modulus. These method and results would be useful for the
studies of cell mechanics and mechanism during the process
of metastasis.
2. Materials and method
We used HeLa cells (human cervical cancer cell) and End1/
E6E7 cells (squamous epithelial cell from normal human
cervix) as the models of cancerous cells and normal cells,
respectively. We took these cells for the present study,
because these had been widely used as a representative pair
of cancerous and normal cells (for example, Govender et al.,
2014; Jain et al., 2014). These cells were handled with
commonly used methods and procedures. They were seeded
on sterilized glass coverslips placed in polystyrene-made
dishes (FalconTR, Corning Int., Tokyo). Then, HeLa cells were
cultivated in MEM-E solution supplemented with fetal calf
serum, and End1/E6E7 cells were cultured in Keratinocyte
SFM solution, both at 37 1C in a 5% CO2 incubator for 4 days.
The cells were grown up to approximately 90% confluence
after 4 days incubation. The cell-seeded coverslips were
installed into a small container of an atomic force microscope
(AFM, Shimadzu SPM-9600, Japan); the container was filled
with Krebs-Ringer bicarbonate buffer solution kept at room
temperature.
For the AFM indentation tests, we used a cantilever having
a spring constant of 0.03 N/m and being adhered a conical
indenter (half-opening or semi-included angle of 221) to the
tip. Indentation tests were performed at the center of each
cell (designated as “central”) and also at the 1/4 of the major
diameter apart from the cell edge (“peripheral”) (See Fig. 1).
Force (F)–indentation (δ) curves were recorded at 0.2 Hz.
The stiffness of cells was determined from fitting F–δ data
to the following equation (Hayashi, 2006; Miyazaki and
Hayashi, 1999):
F ¼ a½expðbδÞ  1; ð1Þ
where a is an index for the shape of F–δ curves, and b
corresponds to the rate of stiffness change. The differentia-
tion of this equation results in
dF=dδ ¼ bF þ c; ð2Þ
Fig. 1 – Central and peripheral locations for AFM
indentation tests.
 journal of the mechanical behavior of biomedical materials 49 (2015) 105 –111 107

where c ( ¼ ab) represents the initial stiffness (stiffness at zero 0.06 0.06

Slope dF/dδ (nN/nm)

●

Slope dF/dδ (nN/nm)

Hela ● End1/E6E7
force or at zero indentation). Therefore, the stiffness of each ● ●
●
cell can be expressed by the parameters b and c. The validity 0.04 0.04 ●
●● ●
●
of Eq. (1) is evaluated from the linearity of the relation ●● ●
0.02 0.02 ●
●●
dF/dδ=0.0044F+0.0004 dF/dδ=0.0052F+0.0009
between dF/dδ and F. ●
●● ●
Moreover, we calculated Young's elastic modulus (E) of
0 0.5 1.0 1.5 0 0.5 1.0 1.5
each cell from the following equation derived on the basis of
Force F (nN) Force F (nN)
Hertzian model (Costa, 2003/2004; Touhami et al., 2003)
 Fig. 3 – Examples of the plotting of dF/dδ against F, and the
F ¼ ð2Eδ2 tan αÞ= πð1−v2 Þ ; ð3Þ
straight regression lines for the data obtained form
where α and ν are the half-opening or semi-included angle of individual cells.
the cantilever tip and Poisson's ratio, respectively. As cells are
assumed to be incompressible, 0.5 is commonly used for Central (n=6, Mean+SD)
Poisson's ratio. We also used 0.5 for ν. In the present study, x10
-3
x10
-3

the half-opening angle α was equal to 221 (0.38 rad). As 8

Parameter c (nN/nm)
2.0

Parameter b (/nm)
Hertzian model is based on a number of simplifying assump- p < 0.05
6
tions, including homogeneous structures, isotropic and linear 1.5
elastic material properties and infinitesimal deformation, 4 1.0
caution must be paid when Eq. (3) is applied to more complex
AFM indentation data (Costa, 2003/2004). The differentiation 2 0.5
of the above equation gives:
0 0
dF=dδ ¼ 0:686Eδ ð4Þ HeLa End1/ HeLa End1/
E6E7 E6E7
If the relation between dF/dδ and δ is linear, then we can
obtain Young's elastic modulus E peculiar to each cell. Fig. 4 – Values of parameters b and c obtained from the
AFM studies were performed on six coverslips (6 batches; central areas of cancerous (HeLa) and normal (End1/
n¼ 6) for each of HeLa and End1/E6E7 cells. From each coverslip, E6E7) cells.
we selected 1 to 3 cells and carried out the indentation tests, and
used the average of their results as the data for the coverslip.
HeLa (n=6, Mean+SD)
Results are expressed as means7SDs (standard deviations).
x10-3 x10-3
Student's t-test was used to analyze data between cells at each 8 2.0

Parameter c (nN/nm)
Parameter b (/nm)

location and between locations in each cell. Significance was set

at the probability (p) level of 0.05 (5%), that is p¼0.05 (5%). 6 1.5

4 1.0
3. Results
2 0.5

Force (F)–indentation (δ) curves are demonstrated in Fig. 2, 0 0

where each curve is a representative one obtained from each Central Peripheral Central Peripheral

of 6 coverslips for each type of cell. All cells showed nonlinear
F–δ curves and large deformation. Although there are fairly
large differences in the curves among cells, cancer cells
(HeLa) seem to have smaller slopes than normal cells (End1/
E6E7). The relations between dF/dδ and F were linear and very
well fitted to Eq. (2) (Fig. 3). The regression coefficients of the
data below F¼ 1 nN (equivalent to approximately δ¼ 300 nm)
were 0.937 in average for all cells, which implies that Eq. (1)
Fig. 2 – Examples of AFM force–indentation curves obtained
from the central areas of cancerous (HeLa) and normal (End1/
E6E7) cells.
Fig. 5 – Values of parameters b and c obtained from the
central and peripheral areas of cancerous (HeLa) cells.
can be applied to the analysis of cell stiffness. The slopes of
the relations shown in Fig. 3 were 4.4  10  3 and 5.2  10  3
nm  1 for a HeLa cell and an End1/E6E7 cell, respectively,
The values of parameters b and c in central locations were
smaller in cancer cells (b ¼3.7370.75  10  3 nm  1; c¼ 0.437
0.13  10  3 nN/nm) than in normal cells (b ¼5.3872.67  10  3
nm  1; c ¼0.9970.36  10  3 nN/nm) (Fig. 4). There was a sig-
nificant difference in parameter c between cancer and normal
cells, although this was not the case for parameter b. In both
cells, no significant differences were observed in the stiffness
(c) and the rate of stiffness change (b) between central and
peripheral locations (Figs. 5 and 6).
The relations between dF/dδ and δ were nonlinear in the
range below 300 nm indentation depth, and did not fit to Eq. (4)
as a whole (Fig. 7). However, the data between 0 and 150 nm
depths seem to fit to straight lines, indicating that the relations
are expressed by the equation. The slopes of the relations in this
range were 1.92 and 4.09 nN/nm2 in a HeLa cell and an End1/
108 journal of the mechanical behavior of biomedical materials 49 (2015) 105 –111

End1/E6E7 (n=6, Mean+SD) strained solid material (Hertz, 1881). Those assumptions are
x10-3 x10-3 not completely fulfilled in cells. As Fig. 7 shows and also as
8 2.0 Dokukin et al. (2013) noted, Young's moduli obtained from

Parameter c (nN/nm)
Parameter b (/nm)

Hertzian model-based equations strongly depend on the

6 1.5
4 1.0
2 0.5
0 0
Central Peripheral
Fig. 6 – Values of parameters b and c obtained from the
central and peripheral areas of normal (End1/E6E7) cells.
Fig. 7 – Examples of the relation between dF/dδ and δ, and
the straight regression line for each data below 150 nm.
p < 0.05
7
Elastic modulus E (kPa)
indentation depth due to the non-linearity of stress–strain
relations of cells that is against the assumption for the
model. Therefore, rigorously speaking, this model should
not be applied to cells. However, thus calculated elastic
modulus has been rather widely used as a parameter of cell
Central Peripheral stiffness for the reason why it may be a good approximation
of the true elastic modulus (Li et al., 2008).
In the case that the tip shape of an AFM cantilever is, for
example, conical like the present study, Hertzian model gives
Eq. (3), and the relation between force (F) and indentation (δ)
forms the parabola. Eq. (4) is derived from Eq. (3), which
indicates that the relation between dF/dδ and δ is linear and
its slope gives Young's elastic modulus. To obtain the elastic
modulus of a cell, we should confirm that F–δ data and/or dF/
dδ–δ data fit well with Eqs. (3) and (4), respectively. In reality,
however, the relation between dF/dδ and δ is not always
linear, for example, as seen from Fig. 7; in these results, linear
relations were observed only in the range of small indenta-
tion, say less than 150 nm, but this was not the case above
this range. These results imply that the elastic moduli of cells
obtained from AFM indentation tests generally change
depending on indentation depth primarily because of their
(n=6, Mean+SD) nonlinear stress–strain relations and inhomogeneous struc-

tures. If the tip is of spheroidal shape, the relation between

6 force and indentation becomes more complex; the square of
5 force is proportional to the cube of indentation (Costa, 2003/
4 2004; Li et al., 2008). As stated above and as estimated from
Fig. 7, the elastic moduli of cells obtained from AFM methods
3
are different at different indentation depths, and therefore
2
we should demonstrate their values together with each
1 corresponding indentation depths.
0 To overcome the drawbacks coming from Hertzian model,
Central Peripheral Central Peripheral
we proposed an empirical equation (Eq. (1)) for describing the
HeLa End1/E6E7
elastic behavior of cells (Miyazaki and Hayashi, 1999), and
Fig. 8 – Values of Young's elastic modulus E in the central have been using parameters b and c included in Eqs. (3) and
and peripheral areas of cancerous (HeLa) and normal (End1/ (4) for quantitatively representing cellular stiffness (Hayashi,
E6E7) cells. 2006). The most important advantage of the method is that
we do not need to think of the assumptions used for Hertzian
model. We can validate Eq. (1) from the linearity of Eq. (2). As
E6E7 cell, respectively; these values are equivalent to Young's seen from Fig. 3 as an example, dF/dδ–δ data very well fitted to
elastic moduli of 2.80 and 5.96 kPa, respectively. Eq. (2) with very high regression coefficients, which supports
At central location, the elastic modulus of cancer cells the validity of Eq. (1). There have been no studies which
(2.4870.50 kPa) was significantly smaller than that of nor- compared stiffness between cancer cells and normal cells
mal cells (5.5070.54 kPa) (Fig. 8); however, there was using Eqs. (1) and (2). As a similar maneuver, Kim et al. (2012)
no statistically significant difference at peripheral location used the slope of force–distance (force–indentation) curve
(2.4470.70 kPa vs. 4.1171.09 kPa). Locational differences were after the AFM tip approached closely or contacted cell sur-
not significant in both cells. face. This is essentially equivalent to the parameter c.
Both of parameters b and c included in Eq. (2) were smaller
in HeLa cells than in End1/E6E7 cells (Fig. 4), indicating that
4. Discussion cancer cells are softer than normal cells. Most of the past
studies on the stiffness of various types of cancer cells, which
In many of the past AFM studies on the mechanical stiffness were determined from not only AFM indentation tests
of cells, Young's elastic moduli were calculated from Hertzian (Table 1) but also the other techniques like microfluidic
model (for example, Table 1). As is well known, this model optical stretchers (Guck et al., 2005), revealed their higher
was made on the basis of several important assumptions like deformability compared to normal or benign cells (Suresh,
homogeneous, isotropic, linear elastic, and infinitesimally 2007).
 journal of the mechanical behavior of biomedical materials 49 (2015) 105 –111 109

Table 1 – Examples of Young's elastic modulus values E obtained from past studies and the present study using AFM
indentation tests, where the values of the present study were obtained from Fig. 8 (central area). RT means room
temperature. Fitting# indicates that the modulus values were determined from fitting studies. See the text for the details.

Source & Method Cell Elastic modulus (kPa) Depth or force

Cross et al. (2007) Cells in pleural effusions Tumor 0.53 o400 nm

Conical, 37 1C Benign 1.97
Li et al. (2008) Breast epithelial cells Cancerous (MCF-7) 0.39 o200 pn
Spheroidal, 37 1C, Fitting# Benign (MCF-10A) 0.54 (o400 nm?)
Nikkhah et al. (2011) Breast epithelial cells Malignant (MDA-MB-23) 0.50 o196 nm
Spheroidal, RT Non-Malignant (MCF-10A) 1.11
Pogoda et al. (2012) Melanoma cells Metastatic (A375) 5.10 (0.76*) at 200 nm
Conical, RT, Fitting# Primary (WM35) 5.25 (3.07*) (*at 1400 nm)
Lekka et al. (2012) Prostate cells Cancerous (PC-3) 1.95 at 400 nm
Conical, RT Normal (PZHPV-7) 3.09
Ramos et al. (2014) Bladder cells Cancerous (HTB-9) 3.0 o300 nm
Conical, RT, Fitting# Non-Malignant (HCV29) 16.0
Present study Cervical cells Cancerous (Hela) 2.48 o 150 nm
Conical, RT, Fitting# Normal (End1/E6E7) 5.50

Lower stiffness of cancer cells is attributable to the
disorganization of cytoskeletal structures and also to the
reduction of actin filaments or their bundles. In fact, Li
et al. (2008) stated from their observations using AFM and
confocal fluorescence images that cancer cells have less
organized stress fibers and weaker actin filamentous network
than benign cells. Cancer cells grow and divide into undiffer-
entiated cells at unregulated, quickened pace, the cytoplasm
also undergoes changes, and new structures appear or
normal structures disappear. Due to these structural changes,
cancerous cells become more deformable than normal cells,
and therefore may have a high potency to more easily
migrate and invade through surrounding tissue matrix, and
induce metastasis (Cai et al., 2010; Li et al., 2008).
There were almost no locational differences in stiffness
between the peripheral region and the central region,
although normal cells (End1/E6E7) had a tendency of having
lower stiffness in the former than in the latter region (Figs. 5
and 6). As stated above, carcinogenesis may rearrange intra-
cellular structures and develop non-uniform distributions of
microfilaments. If so, cellular stiffness may change depend-
ing upon intracellular locations. However, this was not the
case in our study. These results agree with the past studies
showing that there were no locational differences in filamen-
tous structures in cancer cells (Li et al., 2008) and cultured
normal cells (Kataoka et al., 1998).
Pogoda et al. (2012) carried out depth-sensing studies of
the stiffness (Young's modulus) of metastatic and primary
melanoma cells by means of AFM indentation tests. They
observed in both cells that stiffness decreased greatly with
increase in the indentation depth within the range of
approximately 500 nm, and that the rate of the reduction
decreased over 500 nm and stiffness approached constant
level over 1000 nm. They obtained essentially similar results
from cancerous and normal breast cells (Lekka et al., 2012),
and cancerous and non-malignant bladder cells (Ramos et al.,
2014). And they ascribed the depth dependent changes of cell
stiffness to the non-homogeneity of cytoskeleton, although
they did not study the microstructures of those cells in detail.
As long as we know, such depth-sensing studies have
been done only by the above-cited single research group
(Lekka et al., 2012; Pogoda et al., 2012; Ramos et al., 2014).
They used Eq. (3) to calculate Young's elastic modulus E. If so,
the elastic modulus E is equivalent to the slope of the curve
(or line) for the relation between dF/dδ and δ. If the relation is
linear, its slope gives a single value of E. In reality, however,
this relation is not linear in general (Fig. 7), and the slope
progressively increases with the increase of indentation δ,
because most cells and soft biological tissues have J-shaped
stress–strain curves, and their slopes progressively increase
with increase in strain. Therefore, Young's modulus should
be smaller at smaller indentation depths than at larger
indentation depths in AFM indentation tests. In fact, Chiou
et al. (2013) reported the results showing that Young's moduli
of NIH3T3 and 7-4 cells were significantly larger at the
indentation force of 1 nN than at that of 0.2 nN; the former
force corresponds to larger indentation depth than the latter.
Moreover, they recommended that indentation force should
be limited to 1 nN to prevent cell damages and misreading of
cellular stiffness. The results obtained by Chiou et al. (2013)
and by the present study are contrary to the above results
obtained by Lekka and their group. We do not know the
reason for the contradictory results, because they did not
explain how to determine Young's modulus at large depths.
In our present study, we applied Eq. (3) to the data at the
depths smaller than 150 nm (equivalent to approximately
0.0035 nN) where the relations between dF/dδ and δ were
approximately linear (Fig. 7).
To evaluate the elasticity or the stiffness of cells, we
applied Eq. (1) to AFM force–indentation data, and used
parameters included in the equation. In many studies, how-
ever, Young's elastic moduli were calculated from, for exam-
ple, Eq. (3) for the conical shape of cantilever tip. Such
equations as used for Young's modulus are obtained from
Hertzian model that is based on several important assump-
tions as stated above. However, this model has been widely
used for analyzing the data from AFM indentation tests on
cells, because it is believed to give a good approximation of
elastic (Young's) modulus. To compare our results with the
others, we also calculated this modulus applying Eq. (4) to the
data in the range of small indentation depth, say less than
150 nm, where the relations between dF/dδ and δ were
110 journal of the mechanical behavior of biomedical materials 49 (2015) 105 –111
approximately linear (Fig. 7). Thus obtained results demon-
strated that the elastic moduli of cancer cells were smaller
than those of normal cells (Fig. 8), similarly to the results
obtained by most of the past AFM studies (Table 1). The
elastic modulus values shown in this table are very much
scattered probably because of limited accuracies in the
measurements of force and indentation (Costa et al., 2006);
many of the results seem to have been calculated from only a
couple of data of force and indentation using, for example,
Eq. (3). Therefore, it would be recommended to analyze force–
indentation data by fitting them to Eq. (3) and (4).
We carried out the AFM indentation tests keeping cells at
room temperature. Sunyer et al. (2009) reported that the
complex shear moduli of human alveolar epithelial cells
determined with AFM were significantly larger at the physio-
logical temperature than at 131 C. However, they observed no
statistical difference in the elasticity between the physiolo-
gical temperature and room temperature (211 C). Recently,
Chiou et al. (2013) reported that there were no significant
differences in Young's modulus of NIH3T3 cells between body
temperature and room temperature (311 C). Many of past AFM
studies on the elasticity of cancer cells have been done at
room temperature as seen from Table 1. Considering these
results, we did the tests on cells kept at room temperature to
compare stiffness between cancerous and normal cells. As
the change of temperature should change the contractile
activities of molecular motors and therefore may affect the
stiffness of cells, further studies should be done.
It is well known that the structure and the stiffness of cells
are changed by stress (Kataoka et al., 1998; Sumpio et al.,
1988). Cancer cells move from primary locations to metastatic
lesions. Lower stiffness of cancer cells is advantageous for
their motility and metastasis. During these processes, they
are exposed to normal and shear stresses, which would affect
their morphology, structure, and mechanical properties.
Therefore, it is prerequisite to study the effects of stresses
on cellular stiffness, the remodeling of cells, and the effects of
stiffness on the motility and metastasis of cancer cells. As the
first step, the present study and results would be very useful.
5. Conclusions
The conclusions obtained from the present study are:
1) We can apply the equation F¼ a[exp(bδ)  1] to AFM inden-
tation test data of cancer cells, where F and δ are force and
indentation, respectively, and a and b are parameters
relating cellular stiffness,
2) cancer cells are softer than normal cells, and
3) there is no locational difference in the stiffness of cancer
cells between the central and the peripheral regions.
Compliance with ethical standards
This work was performed at the Department of Biomedical
Engineering, School of Engineering, Okayama University of
Science, where both of the authors were belonging to, using a
part of the regular budget from the University.
Acknowledgment
We appreciate the technical assistance of Dr. Takeru Naiki,
Department of Biomedical Engineering, School of Engineer-
ing, Okayama University of Science, Okayama, Japan.
r e f e r e n c e s
Cai, X., Xing, X., Cai, J., Chen, Q., Wu, S., Huang, F., 2010.
Connection between biomechanics and cytoskeleton
structure of lymphocyte and Jurkat cells: an AFM study.
Micron 41, 257–262.
Chiou, Y.-W., Lin, H.-K., Tang, M.-J., Lin, H.-H., Yeh, M.-L., 2013.
The influence of physical and physiological cues on atomic
force microscopy-based cell stiffness assessment. PLoS One 8
(10), e77384.
Costa, K.D., 2003/2004. Single-cell elastography: probing for
disease with the atomic force microscope. Dis. Mark. 19,
139–154.
Costa, K.D., Sim, A.J., Yin, F.C-P., 2006. Non-Hertzian approach to
analyzing mechanical properties of endothelial cells probed
by atomic force microscopy. Trans. J. Biomech. Eng. 128,
176–184.
Cross, S.E., Jin, Y.-S., Rao, J., Gmzewski, J.K., 2007.
Nanomechanical analysis of cells from cancer patients. Nat.
Nanotech. 2, 780–783.
Dokukin, M.E., Guz, N.V., Sokolov, I., 2013. Quantitative study of
the elastic modulus of loosely attached cells in AFM
indentation experiments. Biophys. J. 104, 2123–2131.
Govender, Y., Avenant, C., Verhoog, N.J.D., et al., 2014. Rhe
injectable-only contraceptive medroxyprogesterone acetate,
unlike norethisterone acetate and progesterone, regulates
inflammatory genes in endocervical cells via the
glucocorticoid receptor. PLoS One 9 (5), e96497.
Guck, J., Schinkinger, S., Lincoln, B., et al., 2005. Optical
deformability as an inherent cell marker for testing malignant
transformation and metastatic competence. Biophys. J. 88,
3689–3698.
Hayashi, K., 2006. Tensile properties and local stiffness of cells.
In: Holzapfel, G.A., Ogden, R.W. (Eds.), Mechanics of Biological
Tissue. Springer, Berlin/Heidelberg, pp. 137–152.
Hertz, H., 1881. On the elastic contact of elastic solids. J. Reine
Angew. Math. 92, 156–171.
Jain, A., Kunmar, L., Kushwaha, B., Sharma, M., et al., 2014.
Combining a synthetic spermicide with a natural
trichomonacide for safe, prophylactic contraception. Hum.
Reprod. 9, 242–252.
Kataoka, N., Ujita, S., Sato, M., 1998. Effect of flow direction on the
morphological responses of cultured bovine aortic endothelial
cells. Med. Biol. Eng. Comput. 36, 122–128.
Kim, K.S., Cho, C.H., Park, E.K., et al., 2012. AFM-detected
apoptotic changes in morphology and biophysical property
caused by paclitaxel in Ishikawa and HeLa cells. PLoS One 7
(1), e30066.
Lekka, M., Gil, D., Pogoda, K., et al., 2012. Cancer cell detection in
tissue sections using AFM. Arch. Biochem. Biophys. 518,
151–156.
Li, Q.S., Lee, G.Y.H., Ong, C.N., et al., 2008. AFM indentation study
of breast cancer cells. Biochem. Biophys. Res. Commun. 374,
609–613.
Miyazaki, H., Hayashi, K., 1999. Atomic force microscopic
measurement of the mechanical properties of intact
 journal of the mechanical behavior of biomedical materials 49 (2015) 105 –111
endothelial cells in fresh arteries. Med. Biol. Eng. Comput. 37,
530–536.
Nikkhah, M., Strobl, J., Schmelz, E.M., et al., 2011. Evaluation of
the influence of growth medium composition on cell
elasticity. J. Biomech. 44, 762–766.
Pogoda, K., Jaczewska, J., Wiltowska-Zuber, J., et al., 2012. Depth-
sensing analysis of cytoskeleton organization based on AFM
data. Eur. Biophys. J. 41, 79–87.
Ramos, J.R., Pabijan, J., Garcia, R., et al., 2014. The softening of
human bladder cancer cells happens at an early stage of the
malignancy process. Beilstein J. Nanotechnol. 5, 447–457.
Sunyer, R., Trepat, X., Fredberg, J.J., et al., 2009. The temperature
dependence of cell mechanics measured by atomic force
microscopy. Phys. Biol. 6, 025009.
111
Suresh, S., 2007. Biomechanics and biophysics of cancer cells.
Acta Biomater. 3, 413–438.
Sumpio, B.E., Barnes, A.J., Buckley, M., et al., 1988. Alterations in
aortic endothelial cell morphology and cytoskeletal protein
synthesis during cyclic tensional deformation. J. Vasc. Surg. 7,
130–138.
Touhami, A., Nysten, B., Dufrene, Y.F., 2003. Nanoscale mapping
of the elasticity of microbial cells by atomic force microscopy.
Langmuir 19, 4539–4543.