Uptake of magnetic nanoparticles for adipose-derived stem cells with multiple passage numbers

Abstract

With the increasingly promising role of nanomaterials in tissue engineering and regenerative medicine, the interaction between stem cells and nanoparticles has become a critical focus. The entry of nanoparticles into cells has become a primary issue for effectively regulating the subsequent safety and performance of nanomaterials in vivo. Although the influence of nanomaterials on endocytosis has been extensively studied, reports on the influence of stem cells are rare. Moreover, the effect of nanomaterials on stem cells is also dependent upon the action mode. Unfortunately, the interaction between stem cells and assembled nanoparticles is often neglected. In this paper, we explore for the first time the uptake of γ-Fe2O3 nanoparticles by adipose-derived stem cells with different passage numbers. The results demonstrate that cellular viability decreases and cell senescence level increases with the extension of the passage number. We found the surface appearance of cellular membranes to become increasingly rough and uneven with increasing passage numbers. The iron content in the dissociative nanoparticles was also significantly reduced with increases in the passage number. However, we observed multiple-passaged stem cells cultured on assembled nanoparticles to have similarly low iron content levels. The mechanism may lie in the magnetic effect of γ-Fe2O3 nanoparticles resulting from the field-directed assembly. The results of this work will facilitate the understanding and translation of nanomaterials in the clinical application of stem cells.

Full text

mater.scichina.com link.springer.com ...........................Published online 5 September 2017 | doi: 10.1007/s40843-017-9088-9
Sci China Mater 2017, 60(9): 892–902
8SWDNH RI PDJQHWLF QDQRSDUWLFOHV IRU DGLSRVHGHULYHG
VWHP FHOOV ZLWK PXOWLSOH SDVVDJH QXPEHUV
Yan Yang1, Qiwei Wang1, Lina Song1, Xuan Liu2, Peng Zhao1, Feimin Zhang3, Ning Gu1and
Jianfei Sun1
ABSTRACT With the increasingly promising role of nano-
materials in tissue engineering and regenerative medicine, the
interaction between stem cells and nanoparticles has become a
critical focus. The entry of nanoparticles into cells has become
a primary issue for effectively regulating the subsequent safety
and performance of nanomaterials in vivo. Although the in-
fluence of nanomaterials on endocytosis has been extensively
studied, reports on the influence of stem cells are rare.
Moreover, the effect of nanomaterials on stem cells is also
dependent upon the action mode. Unfortunately, the inter-
action between stem cells and assembled nanoparticles is often
neglected. In this paper, we explore for the first time the up-
take of γ-Fe2O3nanoparticles by adipose-derived stem cells
with different passage numbers. The results demonstrate that
cellular viability decreases and cell senescence level increases
with the extension of the passage number. We found the
surface appearance of cellular membranes to become in-
creasingly rough and uneven with increasing passage num-
bers. The iron content in the dissociative nanoparticles was
also significantly reduced with increases in the passage num-
ber. However, we observed multiple-passaged stem cells cul-
tured on assembled nanoparticles to have similarly low iron
content levels. The mechanism may lie in the magnetic effect
of γ-Fe2O3nanoparticles resulting from the field-directed as-
sembly. The results of this work will facilitate the under-
standing and translation of nanomaterials in the clinical
application of stem cells.
Keywords: nanoparticles, assembly, cellular response, cell pas-
sage, uptake
INTRODUCTION
Stem cells are playing an increasingly important role in
tissue engineering and regenerative medicine [1]. Stem
cell-based therapy has shown great potential for the
treatment of many conditions, such as neural lesions [2],
immunological diseases [3] and bone defects [4]. How-
ever, in clinical practice, establishing the source of stem
cells remains a challenge. These sources must be safe,
biocompatible, and free from any ethical dispute. Me-
senchymal stem cells (MSCs) are typically considered to
be capable of differentiation into vascular endothelial cells
[5], osteoblasts [6], chondrocytes [7], and adipocytes [8].
The sources of MSCs are diverse and those derived from
adipose tissue have immune exemption characteristics [9]
and the ability to inhibit activated lymphocyte prolifera-
tion in allografts [10]. Thus, the adipose-derived me-
senchymal stem cells (ADSCs) are especially suitable for
clinical availability and application [11]. Despite this, an
important clinical issue is that the number of freshly
obtained stem cells in vivo is often too small to meet
practical demand. Thus, cellular passaging has become
necessary before cells can be utilized for therapy. Because
the self-renewal and pluripotency of stem cells are closely
dependent upon the passage number, the influence of
passage number on stem cells is worth exploring.
Recently, magnetic nanoparticles have shown an in-
creasingly promising role in the construction of novel
scaffolds that can impose local magnetic, thermal, and
mechanic stimuli [12–14]. As such, tissue growth can be
accelerated and some small molecules or cytokines can be
released in a controlled way [15]. Therefore, the interac-
tion between stem cells and nanoparticles becomes very
critical for the biomedical application of nanomaterials,
and the entry of nanoparticles into cells is a primary issue.
For the cellular uptake of nanoparticles freely suspended
1State Key Laboratory of Bioelectronics, Jiangsu Key Laboratory of Biomaterials and Devices, School of Biological Science and Medical Engineering,
Southeast University, Nanjing 210009, China
2School of Medicine, Southeast University, Nanjing 210009, China
3Stomatological Hospital of Jiangsu Province, Nanjing 210029, China
Corresponding authors (emails: sunzaghi@seu.edu.cn (Sun J); guning@seu.edu.cn (Gu N))
$57,&/(6 SCIENCE CHINA Materials
 September 2017 | Vol. 60 No. 9
© Science China Press and Springer-Verlag Berlin Heidelberg 2017

in solution, researchers have reported transmembrane
pathways to include membrane infiltration [16], phago-
cytosis [17], and pinocytosis [18]. Endocytosis can be
further divided into macropinocytosis, clathrin-mediated
endocytosis, caveolin-mediated endocytosis, and cla-
thrin-/caveolin-independent endocytosis [19]. It is widely
accepted that most nanoscaled objects are transported
across the lipid bilayer through clathrin-mediated en-
docytosis [20–22], which is highly dependent upon the
fluidity of the cell membrane [23]. Although the influence
of nanomaterials has been extensively studied, the influ-
ence of stem cells, especially with respect to passaging, has
rarely been reported. Moreover, cellular uptake should be
explored not only with respect to dissociative nano-
particles but also those that are assembled. The latter are
more important for transplanting stem cells into the body
via a nanoparticle-modified scaffold. Whereas free-float-
ing nanoparticles are often used as drugs and enter the
body via veins, assembled nanoparticles can be integrated
with implantable devices. Their administrative regulation
differs such that assembled nanoparticles are more easily
clinically translated. In addition, assembled nanoparticles
can exhibit some novel collective properties that free-
floating nanoparticles do not. Different aggregation states
of magnetic nanoparticles have shown various effects on
the fates of stem cells [24,25]. Therefore, the endocytosis
of both free-floating and assembled nanoparticles are
worthy of further research.
In this study, we investigated the uptake of γ-Fe2O3
nanoparticles by ADSCs with multiple passage numbers.
In our experiments, we employed both free-floating na-
noparticles in suspension and assembled nanoparticles on
solid substrate. Because ADSCs are readily clinically
available, they are often loaded on nanoparticle-modified
scaffolds to repair bone tissue by implantation. In this
case, the effect of the assembled nanoparticles on cells is
most important. We measured the cell cycle, viability,
proliferation, cellular senescence, intracellular iron con-
tent, and the expression level of several important pro-
teins at various passages. Our data demonstrate that free-
floating and assembled nanoparticles may have different
pathways for entering cells. By increasing the passage
number, we found the phagotrophic capability of the cells
to be significantly reduced. Assembly can effectively in-
hibit the cellular uptake of nanoparticles, which may be
due to the non-covalent magnetic couplings between in-
dividual units. Our results indicate that the entry of dis-
sociative nanoparticles into stem cells is greatly dependent
on the passage number, whereas that of assembled na-
noparticles is independent of passage number. Phagocy-
tosis can be inhibited by the non-covalent interaction
among nanoparticles such that there will be only tiny
amounts of nanoparticles entering the stem cells.
EXPERIMENTAL SECTION
Synthesis, assembly, and characterization of bare γ-Fe2O3
nanoparticles
Synthesis of bare γ-Fe2O3nanoparticles
We slowly added 25% (w/w) N(CH3)4OH to a mixture of
Fe2+ and Fe3+ (molar ratio is 1:2) until reaching a pH value
of 13. This reaction was allowed to continue for 1 h to
obtain black colloidal nanoparticles (Fe3O4). Then, we
pumped air into the reaction system in a 95°C water bath
after adjusting the pH value to 3. Finally, we maintained
this reaction system for 3 h to oxidize the Fe3O4nano-
particles into γ-Fe2O3nanoparticles. Vigorous stirring was
employed throughout the whole reaction.
Assembly of γ-Fe2O3nanoparticles
We purified the synthesized nanoparticles several times
by centrifugal and magnetic separation until reaching a
pH of 7. Then, we placed a drop of 60-μL colloidal sus-
pension onto a rounded glass plate (d= 15 mm), which
had been pre-cleaned using a boiling mixture of H2O2/
H2SO4(volume ratio: 1:3). We generated a magnetic field
using a “C”-shaped solenoid in which the field strength
can be changed by tuning the excitation current. The
cross section was 4 cm×4 cm and the gap was 2 cm. We
placed the glass plate in the middle of the gap. During the
assembly, the field was parallel to the glass plate and re-
mained there until the colloidal suspension was thor-
oughly dried. By iterative experimentation, we found that
a 120-mT field intensity yielded the best nanoparticle
pattern formation, so we fabricated all the assemblies in
our experiments in the presence of a 120-mT magneto-
static field.
We characterized the morphology of the nanoparticles
by transmission electron microscopy (TEM) (832.20B,
Gatan, American), the hydrodynamic diameter and ζ
potential of the bare γ-Fe2O3nanoparticles using a Mal-
vern Zetasizer Nano (Malvern Instruments Ltd., Wor-
cestershire, UK), and the morphology of the nanogranular
assemblies by scanning electron microscopy (SEM) (Ultra
Plus, Zeiss, Germany).
Cell culture of SD-ADSCs cells
We purchased commercial SD-ADSC cells (RASMD-
01001, Cyagen, American) and cultured the cells with an
SCIENCE CHINA Materials................................$57,&/(6
September 2017 | Vol. 60 No. 9 ................................................................................... 
© Science China Press and Springer-Verlag Berlin Heidelberg 2017

SD-ADSC complete medium containing 10% fetal bovine
serum (FBS), 1% penicillin-striptomyxin, and 1% gluta-
mine (RASMD-90011, Cyagen, American) in a humidi-
fied atmosphere with 5% CO2at 37°C. We changed the
culturing medium every three days. When the cells
reached 80% confluence, they required one passaging. We
harvested the cells using a 0.25% (w/v) trypsin-ethylene
diamine tetra-acetic acid solution.
Cell viability
We used flow cytometry to measure the viability of the
cells without nanoparticle treatment. We seeded the
ADSCs at multiple passage numbers in bare 24-well plates
(5 × 104cells per well) for 24 h. Then, we incubated the
cells with RNase A (100 mg mL−1) at 37°C for 30 min,
followed by staining with propidium iodide (PI)
(40 mg mL−1) at 4°C for 30 min in a dark environment.
After staining, we immediately measured the samples
with a FACSCalibur flow cytometer (Becton-Dickinson)
and analyzed the data using FlowJo software (Tree Star,
Ashland, OR, USA).
We measured the viability of cells treated by either
dissociative or assembled nanoparticles using a Cell
Counting Kit-8 (CCK-8). We seeded the cells of multiple
passage numbers in 24-well microtiter plates (0.5 × 104
cells/200 μL culture medium/well) for different experi-
mental groups. For cells cultured on nanoparticle as-
semblies, direct measurement of viability was not
possible. After seeding the cells on the assemblies of na-
noparticles in 24-well plates for 24 h incubation, we wa-
shed the cells twice with phosphate buffer solution (PBS).
Then, we added CCK-8 reagent to each well and co-in-
cubated it with the cells for 1 h at 37°C. Next, we suc-
tioned the reaction liquid onto the bare culture plate to
reduce the impact of the substrate in the detection process
in which bubbles must be avoided. Finally, we determined
the absorbance at two wavelengths (450 nm for the so-
luble dye and 650 nm for the viable cells) using a mi-
croplate reader (SpectraMax M5, Molecular Devices,
Sunnyvale, CA, USA). We then calculated the cellular
viability by the following formula:
Viability(%)=[(ODi−ODblank)/(ODcell−ODblank)]×100%,
where ODiis the OD value of each well, ODblank is the OD
value of the well with the culturing medium but without
the cells, and ODcell is the OD value of the well with cells
cultured in the absence of nanomaterials.
Detection of senescence
We cultured the ADSCs of multiple passage numbers on a
commerical 24-well culturing plate without adding any
nanoparticles, with the addition of dissociative γ-Fe2O3
nanoparticles, and on a glass surface with stripe-like as-
semblies of γ-Fe2O3nanoparticles, respectively, for 24 h.
We fabricated the assemblies on a rounded glass plate that
fit precisely into the well of the commercial 24-well cul-
turing plate. The cell concentration was 2 × 104cells/well.
We then fixed and stained the cells to detect SA β-gal
activity using a Senescence β-Galactosidase Staining Kit
(Beyotime Institute of Biotechnology, China). After 37°C
incubation overnight, we stained a percentage of SA β-gal
positive cells blue.
Morphological characterization of cells
We observed the cellular morphology by fluorescent mi-
croscopy and SEM.
For fluorescent staining and observation, we washed the
ADSCs of multiple passage numbers with PBS and fixed
them in 3.7% formaldehyde for 15 min at room tem-
perature (25°C). We then washed the fixed cells with PBS
and blocked them with a solution of 1% bovine serum
albumin (BSA) and 0.1% Triton X-100 in PBS for 5 min at
room temperature, after which we washed out the re-
dundant reagents with PBS. Then, we incubated the cells
with rhodamine-phalloidin (Keygen) for 20 min. Next, we
washed the cells with PBS and incubated them with
Hoechst 33258 (Keygen) for 5 min to stain the F-actin.
We then observed the stained ADSCs using a confocal
laser scanning microscope system (Olympus, Japan).
For SEM imaging, we cultured the cells at multiple
passage numbers on cell slides at an initial density of 2 ×
104cells/well for 24 h. After washing for three times with
PBS, we fixed the cells with 2.5% glutaraldehyde in PBS
and stored them at 4°C overnight. Prior to SEM imaging,
we dehydrated the ADSCs in a series of ethanol con-
centrations (30%, 50%, 70%, 80%, 90%, 95%, and 100%).
Finally, we maintained the dehydrated ADSCs in a de-
siccator for overnight air drying. We performed SEM
imaging on an SEM system (Ultra Plus, Zeiss, Germany).
Flow cytometry identification of cells
We seeded cells at the 12th passage number in a 6-well
plate (2 × 105cells/well) for 24 h. Then, we digested the
cells and washed them three times with PBS. We then
blocked the washed cells with a solution of 3% BSA for
30 min and mixed the cells (106cells/100 μL) with fluor-
escent-labeled antibodies (0.25 μg L−1) for 1 h at room
temperature. We measured the blank cells (control) and
antibody-labeled cells by flow cytometry (BD FACSCali-
bur, America) and analyzed the data using FlowJo soft-
$57,&/(6 SCIENCE CHINA Materials
 September 2017 | Vol. 60 No. 9
© Science China Press and Springer-Verlag Berlin Heidelberg 2017

ware.
Uptake of γ-Fe2O3nanoparticles (ICP-OES)
We cultured the cells of multiple passage numbers on a
commercial culturing plate without the addition of na-
noparticles (control), with the addition of dissociative γ-
Fe2O3nanoparticles (γ-Fe2O3), and on a glass surface with
stripe-like assemblies of γ-Fe2O3nanoparticles (120 mT),
respectively, with a 5 × 104cells/well concentration. After
24 h of culturing, we digested the cells and washed them
three times with PBS. Finally, we dissolved the cells with
64% nitric acid after counting. We determined the Fe
concentration by inductively coupled plasma optical
emission spectrometry (ICP-OES).
Intracellular localization of γ-Fe2O3nanoparticles (TEM
imaging)
After 24 h incubation, we digested the cells and washed
them three times with PBS. Then, we fixed the cells with
2.5% glutaraldehyde in PBS and stored them at 4°C. We
dehyrated the cell pellets with ethanol in a series of
concentrations (20%, 30%, 40%, 50%, 60%, 70%, and
90%). Then, we treated the cell pellets with 2% uranyl
acetate in 95% ethanol (en bloc stain) for 1 h and further
dehydrated them with 100% ethanol for 1 h. Lastly, we
treated the cell pellets with propylene oxide twice (15 min
each time), followed by treatment with a mixture of
propylene oxide and araldite resin (volume ratio 1:1).
Subsequently, we embedded the cells in araldite resin at
60°C for 48 h to produce ultrathin sections with a mi-
crotome. We then mounted these slices on copper grids
and stained them with 1% aqueous uranyl acetate and
0.2% lead citrate for imaging by TEM (JEM-2000EX,
JEOL) at 80 kV acceleration voltage.
Statistical analysis
To perform our statistical analysis of the experimental
data, we used SPSS software (version 19.0). We obtained
all of the values from more than three independent ex-
periments and express the figure data in mean ± SD
format, where the mean is the averaged value of the data
and SD is the standard deviation. The error bar in the
figures denote the standard deviations. We subjected the
results to one-way analysis of variance using Duncan’s
test to analyze the difference between the experimental
groups. We considered a statistical difference of less than
0.05 to be significant.
RESULTS AND DISCUSSION
The synthesized γ-Fe2O3nanoparticles were bare and
stabilized by surface hydroxylation. These nanoparticles
were about 10 nm (Fig. 1a) in size and their hydro-
dynamic diameters were about 82.65 nm (Supplementary
information, Fig. S1a). The nanoparticles were negatively
charged with a ζ potential of −14.4 mV (Fig. S1b). To
obtain good pattern formation by the magnetic field-di-
rected assembly, we magnetically separated the colloidal
suspension for purification. After magnetic separation,
the nanoparticles formed small clusters (Fig. 1b) and their
hydrodynamic size changed to about 161.5 nm (Fig. S1a).
We used this sample for both the magnetic field-directed
assembly and in the cellular experiments with the dis-
sociative nanoparticles. Fig. 1c shows the stripe-like pat-
tern of the γ-Fe2O3nanoparticles that formed in the
presence of an external magnetostatic field (Fig. 1d shows
the local magnification of Fig. 1c). Here the field intensity
was 120 mT, which we had determined to be the optimal
value after repeated experimentation. In our experiments,
we fabricated all the assemblies of magnetic nanoparticles
at this field intensity.
Then, we cultured the ADSCs of multiple passage
numbers on the commercial culturing plate (control), the
commercial culturing plate with the addition of the dis-
sociative γ-Fe2O3nanoparticles, and the glass surface with
the stripe-like assemblies, respectively, for 24 h (Fig. 1e, f).
Next, we measured the cellular senescence level by SA β-
gal staining. SA β-gal positive cells will be blue in optical
images. Fig. 2a shows optical photos of the staining re-
sults. To quantitatively compare the senescence level with
the passage number, we developed an image processing
method for calculating the percentages of SA β-gal posi-
tive areas (Fig. 2b). A detailed description of this method
can be found in the Supplementary information (Fig. S2).
As we can see in Fig. 2b, the senescence level of cells
increased with additional cell passage numbers in all three
cases. However, the cells cultured on the nanoparticle
assemblies aged faster whereas the dissociative γ-Fe2O3
nanoparticles seemed to slightly inhibit senescence. Here,
we note that the absolute senescence values of the three
samples have little meaning because the substrate for the
assemblies was bare glass rather than a commercial cul-
turing plate. The biocompatibility of the bare glass was
obviously worse than that of the commercial culturing
plate. However, this indicates that the stem cell perfor-
mance may be dependent on the assembly state of the
nanomaterials during application.
We performed flow cytometry in our analysis of the
cellular cycle of different-passaged ADSCs after 24-h
culturing on commercial plates without the addition of
nanoparticles. We used a combination of the S and G2/M
SCIENCE CHINA Materials................................$57,&/(6
September 2017 | Vol. 60 No. 9 ................................................................................... 
© Science China Press and Springer-Verlag Berlin Heidelberg 2017

phases to describe the viability of different-passaged cells.
We can see that the cell viability decreased with additional
passage numbers, especially after the sixth-passaged cells
(Fig. 3a). Fig. 3b–d show morphological images of the
multiple-passaged ADSCs characterized by optical mi-
croscopy, fluorescent microscopy, and SEM, respectively.
The third-passaged cells showed a typical spindle-like
morphology but with an increase in the passage number,
the cells widened and flattened, becoming more like
polygons. The cytoskeleton staining results show that the
extension of the pseudopodium and microfilament be-
came more irregular. From the SEM images, we can also
see that the cellular surface became increasingly rough.
Some hollow microstructures emerged on the cellular
surfaces after the third passage. This indicates that the
permeability and fluidity of the cytomembrane was
greatly altered with the increase of passage numbers.
Here, we note that although there were significant chan-
ges occurring in the cellular morphology, even the 12th-
passaged ADSCs were still MSCs. We employed the po-
sitive marker CD29 and the negative marker CD11b to
identify the 12th-passaged ADSCs [26]. We show the flow
cytometry results in the Supplementary information (Fig.
S3), where the fluorescent peaks of CD29 and CD11b are
in identical locations compared with those of the control.
We performed a CCK8 assay to measure the viability of
different-passaged ADSCs cultured in the presence of
dissociative γ-Fe2O3nanoparticles and on the surface of
the stripe-like assemblies of γ-Fe2O3nanoparticles. The
results show a similar trend with that of the cells in the
absence of nanoparticles (Fig. 3e). We also found that the
dissociative γ-Fe2O3nanoparticles seemed capable of en-
hancing cellular viability. The viability of cells obviously
declined more slowly with the increase of passage num-
bers in the presence of dissociative γ-Fe2O3nanoparticles.
Here, we considered the nanoparticles to have little effect
on the self-renewal and differentiation of ADSCs because
we had conducted all our experiments after just 24 h co-
culturing. However, typically, it takes at least two weeks
for ADSCs to differentiate. So, we consider 24 h to be too
short a time for nanoparticles to renew or differentiate
ADSCs. In addition, stem cells form colonies prior to
20 μm
ab
cd
ef
100 nm
Free nanoparticles
2 μm
100 nm
Figure 1 Characterization of nanoparticles and assemblies. (a) TEM image of γ-Fe2O3nanoparticles before magnetic separation. (b) TEM image of γ-
Fe2O3nanoparticles after magnetic separation. (c) SEM image of the assemblies. (d) Local magnification of (c). (e) Schematic illustration of SD-ADSCs
cultured on the surface of stripe-like assemblies. (f) Schematic illustration of SD-ADSCs treated with dissociative γ-Fe2O3nanoparticles.
$57,&/(6 SCIENCE CHINA Materials
 September 2017 | Vol. 60 No. 9
© Science China Press and Springer-Verlag Berlin Heidelberg 2017

differentiation. However, from the morphological images
of cells during our experiments, we can see that the
ADSCs remained in the discrete state. This also confirms
that the stem cells were prevented from differentiating.
We quantitatively measured the cellular uptake of the
magnetic nanoparticles by ICP-OES after 24 h of cultur-
ing, the results of which are shown in Fig. 4a. For the
dissociative nanoparticles, we found the Fe content inside
the cells to be significantly reduced with an increase in the
passage number. However, multiple-passaged cells cul-
tured on the assemblies of nanoparticles showed little
difference. Moreover, the internal Fe content of different-
passaged cells cultured on the assemblies was much less
than that treated by the dissociative nanoparticles. This
indicates that the field-directed assembly effectively in-
hibited the entry of nanoparticles into cells, which means
better security in vivo [27]. Our microscopic observations
also confirm this point (Fig. 4b–e). We can see that the
nanoparticles located in the endosomes in both the dis-
sociative nanoparticles and the assemblies. However, the
γ-Fe2O3nanoparticles were in isotropic clusters in the
cells treated with dissociative nanoparticles whereas the γ-
Fe2O3nanoparticles formed needle-like aggregates in the
cells cultured on the assemblies. This reveals that the
inhibition of the cellular uptake of nanoparticles possibly
resulted from the field-induced aggregation of the mag-
netic nanoparticles. Due to the weak interaction between
the elemental units, the entry of nanoparticles into the
cells must demand more energy, which makes it more
difficult for the cells to endocytose the nanoparticles [28].
Here, we note that the intracellular γ-Fe2O3nanoparticles
showed a needle-like aggregation in the cells cultured on
magnetic field-directed assemblies. One possibility is that
the assembled nanoparticles entered into the cells in the
shape of small chains because the endocytosis could not
break up the assembled nanoparticles into dissociative
units. The other possibility is that the cellular uptake of
nanoparticles occurred in the form of individual units
100 μm
P3 P6 P9 P12
CON
Ȗ-Fe2O3
120 mT
0.00%
0.20%
0.40%
0.60%
0.80%
1.00%
1.20%
1.40%
1.60%
P3 P6 P9 P12
Senescence level (the area of blue cells)
Control
120 mT
Ȗ-Fe2O3
b
a
Figure 2 Senescence measurements for different-passaged ADSCs using the β-Galactosidase Staining Kit (β-Galactosidase positive cells are blue). (a)
Staining images for cells cultured on commercial culturing plate (CON), treated by dissociative γ-Fe2O3nanoparticles (γ-Fe2O3), and cultured on the
stripe-like assemblies (120 mT). (b) Percentage of β-galactosidase positive area (blue area of (a)) calculated by our image processing method.
SCIENCE CHINA Materials................................$57,&/(6
September 2017 | Vol. 60 No. 9 ................................................................................... 
© Science China Press and Springer-Verlag Berlin Heidelberg 2017

because particles anisotropic in shape could not easily
enter into the cells due to the severe deformation of the
cellular membrane. However, the magnetic nanoparticles
may be assembled into small chains again during in-
tracellular transport due to the magnetic dipolar inter-
action. The formation mechanism of needle-like
aggregates inside the endosome is an interesting issue. As
yet, we cannot explain this phenomenon. However, this
will be our next research goal.
Although the intracellular Fe content was fairly low in
the cells cultured on the assemblies, the transmembrane
pathway could still be activated. As noted above, the main
pathways for the cellular uptake of nanoparticles are
clathrin-mediated endocytosis and phagocytosis. The
clathrin and the scavenger receptor protein (SR-A) are the
typical markers. We detected these two proteins by
fluorescent q-polymerase chain reaction (PCR) assay (Fig.
5a, b). The results reveal that the expression of both cla-
thrin and SR-A decreased with the increase of passage
numbers, irrrespective of whether the cells were treated by
dissociative nanoparticles or cultured on assemblies of
nanoparticles. Moreover, the presence of nanoparticles
greatly up-regulated the expression levels of clathrin and
SR-A compared with the control. This demonstrates that
the signal pathway of cellular uptake was activated by the
nanoparticles although the nanoparticles may have been
prevented from entering into the cells. Also, the influence
on the cells from the dissociative nanoparticles seemed
slightly more significant than that from the assembled
nanoparticles. Based on Fig. 5, we found the alteration of
the clathrin level to be identical in different-passaged cells
treated by the dissociative nanoparticles and cultured on
10 μm 2μm
100 μm
ae
b
cd
P3 P6 P9 P12
P3 P6
P9 P12
P3 P6
P9 P12
0
0.05
0.10
0.15
0.20
0.25
P3 P6 P9 P12
S+G2/M
0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
P3 P6 P9 P12
Cell viability (% vs. control)
120 mT
Ȗ-Fe2O3
Figure 3 Characterization of different-passaged ADSCs. (a) Cell cycle measurement of different-passaged ADSCs without treatment of nanoparticles.
(b) Morphological images of multiple-passaged ADSCs without treatment of nanoparticles characterized by bright field optical microscopy. (c)
Morphological images of multiple-passaged ADSCs without treatment of nanoparticles characterized by fluorescent microscopy. (d) Morphological
images of multiple-passaged ADSCs without treatment of nanoparticles characterized by SEM. (e) Viability measurements using CCK-8 assay for
different-passaged ADSCs cultured on commercial culturing plate (control), treated by dissociative γ-Fe2O3nanoparticles (γ-Fe2O3). and cultured on
the stripe-like assemblies (120 mT).
$57,&/(6 SCIENCE CHINA Materials
 September 2017 | Vol. 60 No. 9
© Science China Press and Springer-Verlag Berlin Heidelberg 2017

the assemblies. With the increase of passage numbers, the
down-regulation of the clathrin level was less significant.
However, the maximal expression of SR-A was at the sixth
passage rather than the third and the expression of SR-A
protein was significantly down-regulated at the ninth and
twelfth passages with respect to the third and sixth pas-
sages. This phenomenon can be interpreted with respect
to the morphological alteration of the cellular membrane
0
2
4
6
8
10
12
14
16
18
20
120 mT assemblies Suspension Ȗ-Fe2O3
Intracellular iron (ng/104 cells)
P3
P6
P9
P12
2m 200 m
bc
de
a
Figure 4 Characterization of cellular uptake for γ-Fe2O3nanoparticles. (a) Iron concentration measurement by ICP-OES for different-passaged
ADSCs treated by the dissociative γ-Fe2O3nanoparticles and cultured on the stripe-like assemblies, respectively. (b–e) Intracellular localization of γ-
Fe2O3nanoparticles characterized by TEM: (b) cells treated by the γ-Fe2O3nanoparticles and (c) local magnification of (b); (d) cells cultured on the
stripe-like assemblies and (e) local magnification of (d).
ab
c
0
0.5
1.0
1.5
2.0
2.5
P3 P6 P9 P12
Clathrin mRNA expression level
Control
120 mT
0
0.5
1.0
1.5
2.0
2.5
3.0
P3 P6 P9 P12
SR-A mRNA
expression
level
Control
120 mT
0
0.5
1.0
1.5
2.0
2.5
3.0
P3 P6 P9 P12
Magnetosensing protein mRNA
expression level
Control
120 mT
Ȗ-Fe2O3
Ȗ-Fe2O3
Ȗ-Fe2O3
Figure 5 mRNA measurements of clathrin, SR-A protein, and magnetosensing protein for different-passaged ADSCs cultured on cellular culturing
plates (control), treated by the dissociative γ-Fe2O3nanoparticles (γ-Fe2O3), and cultured on the stripe-like assemblies (120 mT) with quantitative
PCR. (a) mRNA expression level of clathrin. (b) mRNA expression level of SR-A. (c) mRNA expression level of magnetosensing protein.
SCIENCE CHINA Materials................................$57,&/(6
September 2017 | Vol. 60 No. 9 ................................................................................... 
© Science China Press and Springer-Verlag Berlin Heidelberg 2017

because phagocytosis is highly dependent upon the de-
formation of the cellular membrane. Considering Fig. 4a,
we hypothesize that both the clathrin-mediated en-
docytosis and phagocytosis contributed to the entry of
nanoparticles into the ADSCs. However, from the view-
point of the signal pathway, the dominant means of entry
for the dissociative nanoparticles was clathrin-mediated
endocytosis whereas phagocytosis may play a relatively
major role in assembled nanoparticles. Due to the pre-
sence of magnetic couplings between the elementary
units, cellular phagocytosis was effectively inhibited so
that there were only tiny amounts of nanoparticles
available to enter into the cells. Thus, the difference in the
cellular uptake mechanism between cells treated by dis-
sociative nanoparticles and those cultured on assemblies
of nanoparticles may lie in the pathway of cell entry,
which results from the different interaction states between
individual nanoparticles.
Based on our results, we believe the assembly to have
played an important role. The assembly of γ-Fe2O3na-
noparticles reduced the collective free energy of nano-
particles while augmenting the magnetic energy. This
hypothesis can be partly supported by our conclusions in
previous reports that assemblies of magnetic nano-
particles can mediate the magnetic effect on cells [13,25].
Here, we measured the magnetosensing protein by q-
PCR, which was recently reported to be capable of re-
sponding to magnetic interaction (Fig. 5c) [29]. We found
that with the increase of passage numbers, the expression
of magnetosensing protein was down-regulated and the
expression of magnetosensing protein was maximized
when the cells were cultured on assemblies of nano-
particles. In addition, the difference in the magnetotsen-
sing protein expression between cells treated by
dissociative nanoparticles and those cultured on assem-
blies of nanoparticles was diminished after the third
passage. This result can account for the phenomenon in
which the expression level of the SR-A protein became
approximately identical in both the cells after the sixth
passage. This result also reveals that the magnetic nano-
particles-mediated magnetic effect upon cells grows less
effective in senescent cells. Thus, the magnetism-based
application of iron oxide nanoparticles may be more
suitable for neogenic cells.
CONCLUSIONS
In conclusion, in this study, we explored for the first time
the influence of the number of cellular passages on the
uptake of γ-Fe2O3nanoparticles. We found dissociative
nanoparticles to have increasing difficultly in entering
adipose-derived stem cells with increasing cellular pas-
sages. However, if the nanoparticles are assembled by
magnetic field, their cellular uptake can be effectively
inhibited, with little difference among the multiple-pas-
saged ADSCs. Our fluorescent q-PCR detection results
demonstrated that dissociative and assembled nano-
particles may activate clathrin-mediated endocytosis and
phagocytosis, respectively. Here, we believe the assembly-
mediated magnetic effect plays an important role, which
was partly proved by the measurement of the magneto-
sensing protein. We proposed that the magnetism-based
application of iron oxide nanoparticles may be more
suitable for neogenic cells. Due to the extremely im-
portant role of the cellular uptake of nanoparticles, we
believe our work will facilitate the design and protocol
development of nanoparticles-based biomedical applica-
tions.
Received 21 June 2017; accepted 2 August 2017;
published online 5 September 2017
1 Patel S, Lee KB. Probing stem cell behavior using nanoparticle-
based approaches. WIREs Nanomed Nanobiotechnol, 2015, 7:
759–778
2 Lee IH, Huang SS, Chuang CY, et al. Delayed epidural trans-
plantation of human induced pluripotent stem cell-derived neural
progenitors enhances functional recovery after stroke. Sci Rep,
2017, 7: 1943–1955
3 Laar JMV. Immune ablation and stem-cell therapy in autoimmune
disease: immunological reconstitution after high-dose im-
munosuppression and haematopoietic stem-cell transplantation.
Arthritis Res, 2000, 2: 270–275
4 Wang X, Li G, Guo J, et al. Hybrid composites of mesenchymal
stem cell sheets, hydroxyapatite, and platelet-rich fibrin granules
for bone regeneration in a rabbit calvarial critical-size defect
model. Exp Ther Med, 2017, 13: 1891–1899
5 Tancharoen W, Aungsuchawan S, Pothacharoen P, et al. Differ-
entiation of mesenchymal stem cells from human amniotic fluid to
vascular endothelial cells. Acta Histochemica, 2017, 119: 113–121
6 Tong Y, Niu M, Du Y, et al. Aryl hydrocarbon receptor suppresses
the osteogenesis of mesenchymal stem cells in collagen-induced
arthritic mice through the inhibition of β-catenin. Exp Cell Res,
2017, 350: 349–357
7 Tanthaisong P, Imsoonthornruksa S, Ngernsoungnern A, et al.
Enhanced chondrogenic differentiation of human umbilical cord
Wharton's jelly derived mesenchymal stem cells by GSK-3 in-
hibitors. PLoS ONE, 2017, 12: e0168059
8 Suzuki A, Saeki T, Ikuji H, et al. Brown algae polyphenol, a prolyl
isomerase pin1 inhibitor, prevents obesity by inhibiting the dif-
ferentiation of stem cells into adipocytes. PLoS ONE, 2016, 11:
e0168830
9 Boquest AC, Shahdadfar A, Brinchmann JE. Isolation of stromal
stem cells from human adipose tissue. Methods Mol Biol, 2006,
325: 35–46
10 Lindroos B, Suuronen R, Miettinen S. The potential of adipose
stem cells in regenerative medicine. Stem Cell Rev Rep, 2011, 7:
$57,&/(6 SCIENCE CHINA Materials
 September 2017 | Vol. 60 No. 9
© Science China Press and Springer-Verlag Berlin Heidelberg 2017

269–291
11 Geburek F, Roggel F, van Schie HTM, et al. Effect of single in-
tralesional treatment of surgically induced equine superficial digital
flexor tendon core lesions with adipose-derived mesenchymal
stromal cells: a controlled experimental trial. Stem Cell Res Ther,
2017, 8: 129
12 Hadavi M, Hasannia S, Faghihi S, et al. Novel calcified gum Arabic
porous nano-composite scaffold for bone tissue regeneration.
Biochem Biophys Res Commun, 2017, 488: 671–678
13 Liu X, Zhang J, Tang S, et al. Growth enhancing effect of LBL-
assembled magnetic nanoparticles on primary bone marrow cells.
Sci China Mater, 2016, 59: 901–910
14 Ngadiman NHA, Idris A, Irfan M, et al. γ-Fe2O3nanoparticles
filled polyvinyl alcohol as potential biomaterial for tissue en-
gineering scaffold. J Mech Behav Biomed Mater, 2015, 49: 90–104
15 Lu L, Unsworth LD. pH-Triggered release of hydrophobic mole-
cules from self-assembling hybrid nanoscaffolds. Biomacromole-
cules, 2016, 17: 1425–1436
16 Verma A, Uzun O, Hu Y, et al. Surface-structure-regulated cell-
membrane penetration by monolayer-protected nanoparticles. Nat
Mater, 2008, 7: 588–595
17 Gu JL, Xu HF, Han YH, et al. The internalization pathway, me-
tabolic fate and biological effect of superparamagnetic iron oxide
nanoparticles in the macrophage-like RAW264.7 cell. Sci China
Life Sci, 2011, 54: 793–805
18 Sahay G, Alakhova DY, Kabanov AV. Endocytosis of nanomedi-
cines. J Control Release, 2010, 145: 182–195
19 Conner SD, Schmid SL. Regulated portals of entry into the cell.
Nature, 2003, 422: 37–44
20 Zhao P, Cao M, Song L, et al. Downregulation of MIM protein
inhibits the cellular endocytosis process of magnetic nanoparticles
in macrophages. RSC Adv, 2016, 6: 96635–96643
21 da Luz CM, Boyles MSP, Falagan-Lotsch P, et al. Poly-lactic acid
nanoparticles (PLA-NP) promote physiological modifications in
lung epithelial cells and are internalized by clathrin-coated pits and
lipid rafts. J Nanobiotechnol, 2017, 15: 11–29
22 Banerjee A, Berezhkovskii A, Nossal R. Kinetics of cellular uptake
of viruses and nanoparticles via clathrin-mediated endocytosis.
Phys Biol, 2016, 13: 016005
23 Scheinpflug DK. Measurement of cell membrane fluidity by laur-
dan GP: fluorescence spectroscopy and microscopy. Methods Mol
Biol, 2017, 1520: 159–174
24 Wang Q, Chen B, Cao M, et al. Response of MAPK pathway to
iron oxide nanoparticles in vitro treatment promotes osteogenic
differentiation of hBMSCs. Biomaterials, 2016, 86: 11–20
25 Sun J, Liu X, Huang J, et al. Magnetic assembly-mediated en-
hancement of differentiation of mouse bone marrow cells cultured
on magnetic colloidal assemblies. Sci Rep, 2015, 4: 5125
26 Chamberlain G, Fox J, Ashton B, et al. Concise review: mesench-
ymal stem cells: their phenotype, differentiation capacity, im-
munological features, and potential for homing. Stem Cells, 2007,
25: 2739–2749
27 Sufian MM, Khattak JZK, Yousaf S, et al. Safety issues associated
with the use of nanoparticles in human body. Photodiagnosis
Photodynamic Ther, 2017, 19: 67–72
28 Decuzzi P, Ferrari M. The receptor-mediated endocytosis of
nonspherical particles. BioPhys J, 2008, 94: 3790–3797
29 Qin S, Yin H, Yang C, et al. A magnetic protein biocompass. Nat
Mater, 2015, 15: 217–226
Acknowledgements This work was supported by the National Basic
Research Program of China (2013CB733801) and the National Key
Research and Development Program of China (2017YFA0104301). Sun J
is thankful to the supports from the Fundamental Research Funds for
the Central Universities. All authors are thankful to the supports from
Collaborative Innovation Center of Suzhou Nano Science and Tech-
nology.
Author contributions Sun J designed the experiments, analyzed the
results and modified the manuscript. Yang Y did all the experiments
with the help of Wang Q and Zhao P and wrote the manuscript with the
help from Sun J and Liu X. Song L provided the nanoparticle solution.
Gu N and Zhang F supervised the project. All authors contributed to the
general discussion.
Conflict of interest The authors declare that they have no conflict of
interest.
Supplementary information Supprting data are available in the online
version of the paper.
SCIENCE CHINA Materials................................$57,&/(6
September 2017 | Vol. 60 No. 9 ................................................................................... 
© Science China Press and Springer-Verlag Berlin Heidelberg 2017

Yan Yang received her BSc degree in biotechnology from Anhui University in 2014. Now she is pursuing her master
degree in biomedical engineering at Southeast University. His research interest is the assembly of nanoparticles.
Jianfei Sun received his PhD degree in biomedical engineering from Southeast University in 2008. He is now an associate
professor at the School of Biological Science and Medical Engineering, Southeast University. His research interests include
the fabrication of nanoelectronic devices by self-assembly of nanoparticles and their application in biomedical issues.
Ning Gu received his PhD degree in biomedical engineering from the Department of Biomedical Engineering,Southeast
University, Nanjing, China, in 1996. Currently he is a Cheung Kong Scholar Chair Professor at the School of Biological
Science and Medical Engineering, Southeast University and Director of Jiangsu Key Laboratory of Biomaterials and
Devices. His research interests include the applications of magnetic nanomaterials in biomedicine.
⤜㵍⪌⫛㭞ⰵ䐍ⳟヅ⨅䐫ⶪ㻙⟜㪄㦂⪦㾵㚪㗸㋦㑄⭥䇑㼍䁱㈠
䂏䂁1,㶖儘咠1,㯯㏗㚩1,㒖冐2,䍵㞕1,䍣⳨㘕3,⹬㛟1,㰐ㅄ⳪1
䍋䄋 㰇䓦㚪㗸⤥㑰䊻䔊䐐⹅⧭⼮䊺㪛䄞䁈䐱䊞㎕䊞ⱁ⭥䇇䇤,ⶪ㻙⟜⼮㚪㗸⤥㑰䐏ヅ⭥㼁⿆䔘䇤⧪㸋⹹ポ⿘ㅻ,ⱙ㚪㗸㋦㑄㆙㧌㻙⟜
㬨⹹㻖⭞㚪㗸⤥㑰➓㦌㾵⼮ⶪ㻙⟜㘝䊬⮘㋹⭥㬸䄋㸫㳃.ⶪ㻙⟜⪌⫛㬨㡅䇇䇤䐱⡹⤜㋪㩺⭥⺞⧭,⭌⹹䇻⪌⫛⫛㭞ⰵⶪ㻙⟜㪄㦂㚪㗸㋦
㑄⭥䇑㼍⭥䁱㈠⿚ㅰ㩺.⪬㶃,ⶪ㻙⟜⼮㚪㗸⤥㑰⭥㼁⿆䔘䇤⿚䈌㚪㗸㋦㑄⭥⫇䊻Ⳟ㬞䇱⹹.⡟㸥䊻⤄㑈㠍㩰䔊䓑㑬㳖⫙䓕⭥ư-Fe2O3㚪㗸
㋦㑄䔊䓑ㆂ⹚,⤃䊻ⶤ⢎㘇㩰㞁䂙SD⫔㭔䐍ⳟヅ⨅䐫ⶪ㻙⟜,㦜⽔䁱㈠㑬⤜㵍⫛㭞⭥ⶪ㻙⟜ⰵ䔊䓑⼮䇯㏌⭥㚪㗸㋦㑄⭥㵭㬪㤊㌗.ㆂ⺜ⳃ
㻷,㰇䓦㻙⟜⫛㭞䋗ゴ,㻙⟜』㑇ㅖ⭮,㻙⟜㯆㎰㯏㠞䋗ゴ,⤃㣳㻙⟜㚅⭥⢎㘇⧫㻷⨗⪷⤻⼮⤜㉚䊩⭥㾯㗓.⭒䈌䇯㏌⭥ư-Fe2O3㚪㗸㋦㑄⹓
㞁䂙㬒,㻙⟜㚻㳛⼍㑠㰇䓦⫛㭞⭥䋗ゴⱙブ㩺,⭌䊻䔊䓑㳆㩰㞁䂙⭥⤜㵍⫛㭞⭥㻙⟜㉀䇱㼁㯧⭥㳛⼍㑠,⤃㣳⟜㚻㳛⼍㑠れ㩺.㒎㶃,⪦ⶱ
䇇⭑➸⭥⢎⫐⢎㘘⪦㾵㚪㗸㋦㑄⭥䔊䓑㳆ⰵ㻙⟜䇱⪦㾈䇇.ⶤ䁱㈠⢎㘘,㻙⟜⫛㭞⭥䁂䋒ⰵ䁱㈠㋦㑄㚻⿐㬖䂊㬨䐢⹹䐹䄋⭥,㻙⟜⫛㭞䇇
ⶤ䔘㸋㻙⟜㪄㦂㬖䂊⭥䄜⷗䐹䄋㋝㔨䅓㯹.
$57,&/(6 SCIENCE CHINA Materials
 September 2017 | Vol. 60 No. 9
© Science China Press and Springer-Verlag Berlin Heidelberg 2017