Home

Search

Collections

Journals

About

Contact us

My IOPscience

Synthesis of water-dispersible zinc oxide quantum dots with antibacterial activity and low
cytotoxicity for cell labeling

This content has been downloaded from IOPscience. Please scroll down to see the full text.
2013 Nanotechnology 24 475102
(http://iopscience.iop.org/0957-4484/24/47/475102)
View the table of contents for this issue, or go to the journal homepage for more

Download details:
IP Address: 129.174.21.5
This content was downloaded on 06/06/2014 at 17:43

Please note that terms and conditions apply.

IOP PUBLISHING

NANOTECHNOLOGY

Nanotechnology 24 (2013) 475102 (11pp)

doi:10.1088/0957-4484/24/47/475102

Synthesis of water-dispersible zinc oxide

quantum dots with antibacterial activity
and low cytotoxicity for cell labeling
Shan-hui Hsu1,2 , Ying Yi Lin1 , Sherry Huang1 , Kwok Wai Lem3,4 ,
Dinh Huong Nguyen5 and Dai Soo Lee5
1

Institute of Polymer Science and Engineering, National Taiwan University, Taipei 10617,
Taiwan, Republic of China
2
Research Center for Developmental Biology and Regenerative Medicine, National Taiwan University,
Taipei 10617, Taiwan, Republic of China
3
Department of Physics, MTSE Program, New Jersey Institute of Technology, NJ 07102, USA
4
Super Learning International LLC, NJ 07869, USA
5
School of Chemical Engineering, Chonbuk National University, Jeonju 561-756, Republic of Korea
E-mail: shhsu@ntu.edu.tw and dslee@jbnu.ac.kr

Received 21 May 2013, in final form 9 October 2013

Published 31 October 2013
Online at stacks.iop.org/Nano/24/475102
Abstract
Typical photoluminescent semiconductor nanoparticles, called quantum dots (QDs), have
potential applications in biological labeling. When used to label stem cells, QDs may impair
the differentiation capacity of the stem cells. In this study, we synthesized zinc oxide (ZnO)
QDs in methanol with an average size of 2 nm. We then employed two different types of
polyethylene glycol (PEG) molecules (SH-PEG-NH2 and NH2 -PEG-NH2 ) to conjugate ZnO
QDs and made them water-dispersible. Fourier transform infrared spectroscopy spectra
indicated the attachment of PEG molecules on ZnO QDs. No obvious size alteration was
observed for ZnO QDs after PEG conjugation. The water-dispersible ZnO QDs still retained
the antibacterial activity and fluorescence intensity. The cytotoxicity evaluation revealed that
ZnO QDs at higher concentrations decreased cell viability but were generally safe at 30 ppm
or below. Cell lines of hepatocytes (HepG2), osteoblasts (MC3T3-E1) and mesenchymal stem
cells (MSCs) were successfully labeled by the water-dispersible ZnO QDs at 30 ppm. The
ZnO QD-labeled MSCs maintained their stemness and differentiation capacity. Therefore, we
conclude that the water-dispersible ZnO QDs developed in this study have antibacterial
activity, low cytotoxicity, and proper labeling efficiency, and can be used to label a variety of
cells including stem cells.
S Online supplementary data available from stacks.iop.org/Nano/24/475102/mmedia
(Some figures may appear in colour only in the online journal)

1. Introduction

shells have been developed, these NPs are potentially harmful

to the biological systems because of the leakage of Cd ions
through the shell defect, the decomposition of NPs by oxygen,
and the radicals derived from light irradiation [4, 5].
Mesenchymal stem cells (MSCs) are multipotent cells
that can differentiate into multiple cell lineages and are a
promising cell source for regenerative medicine. Fluorescent
probes including QDs may be used to label and trace the

The use of nanoparticles (NPs) in biological labeling has

received considerable attention. During the past decade, the
typical photoluminescent semiconductor NPs, or quantum
dots (QDs), based on CdSe or CdS species are commonly
applied in biological systems [13]. Although various
protections employing ZnS, polymers, and the other nontoxic
0957-4484/13/475102+11$33.00

c 2013 IOP Publishing Ltd Printed in the UK & the USA

Nanotechnology 24 (2013) 475102

S-H Hsu et al

transplanted MSCs in vivo. However, the uptake of QDs

such as CdSe/ZnS may adversely affect the differentiation
capacity of MSCs, including osteogenic [6] and chondrogenic
differentiation [7]. Moreover, QDs extruded from the labeled
cells may re-enter the adjacent cells either in vitro or in
vivo [8]. Therefore, the development of nontoxic QDs is of
essential importance for stem cell labeling.
Semiconducting ZnO QDs possess unique properties,
such as a high-exciton binding energy (60 meV) and wide
band gap energy (3.37 eV). These properties result in specific
optical properties and high chemical stabilization at room
temperature [911]. In addition, ZnO exhibits not only a
UV excitonic emission peak, but also visible luminescence
(for instance, green luminescence) at different emission
wavelengths due to intrinsic or extrinsic defects. Green
luminescence has been attributed to various types of defects
such as oxygen vacancies [12, 13].
A number of techniques [1418] have been developed to
prepare ZnO QDs. They present new physical properties and
potentials in applications such as blue luminescent devices,
ultraviolet detectors, solar cells [19], and photocatalysts [20].
On the other hand, ZnO NPs are popular active substances
used in ultraviolet protective substances and are generally
regarded to be safe [21, 22]. ZnO NPs possess bacteriostatic
activity against Gram-negative bacteria (Escherichia coli)
[23, 24], Gram-positive bacteria (Staphylococcus aureus) [25], mesophilic, and halophilic bacteria [26]. These
make ZnO QDs attractive nanomaterials in biological
labeling [27, 28].
Although there are a few reports on the cytotoxic effect
of ZnO NPs on mammalian cells [2936], the influence of
ZnO QD labeling on stem cells remains relatively unexplored.
Research has indicated that ZnO in the nano-scale may have
size-dependent cytotoxicity, and the smaller the size, the lower
the cytotoxic effect observed [3638]. In this work, small-size
ZnO QDs were initially synthesized in methanol. These ZnO
QDs were further modified by polyethylene glycol (PEG) to
form water-dispersible ZnO QDs. The antibacterial ability of
the prepared ZnO QDs was evaluated. The water-dispersible
ZnO QDs were used to label cell lines of hepatocytes (HepG2)
and osteoblasts (MC3T3-E1), as well as MSCs isolated from
fat, i.e. the adipose-derived adult stem cells (ADAS). The
cytotoxicity and the generation of reactive oxygen species
(ROS) from cells exposed to water-dispersible ZnO QDs were
analyzed. The differentiation capability of ADAS labeled by
water-dispersible ZnO QDs was examined. We sought to
establish safe and antibacterial QDs for cell labeling.

as ZnOHMA QDs) were precipitated by adding heptane,

washed, and redispersed in methanol to get 0.5% ZnOHMA
QD dispersion. The dispersion was concentrated to 5% by
evaporating methanol with stirring by a magnetic bar. All the
concentration values of QDs (0, 15, 30, and 60 ppm) in this
study represented those based on the contents of ZnO. The
preparation steps in methanol and the chemical structure of
ZnOHMA QDs are shown in figure 1(A).
To prepare water-dispersible ZnO QDs from ZnOHMA
QDs, 500 l of 1 mg ml1 of ,-bis-amino PEG
(NH2 -PEG-NH2 , Mw 1500 g mol1 , Rapp Polymere,
Germany) in deionized distilled water was mixed with 500 l
0.3 mM ZnOHMA QDs at 25 C. 500 lof 8 mM 1ethyl-3-[3-dimethylaminopropyl] carbodiimide hydrochloride
(EDC) solution was added to form amide bonding between
ZnOHMA and ,-bis-amino PEG. On the other hand,
1 mg ml1 -mercapto--amino PEG (SH-PEG-NH2 , Mw
3000 g mol1 , Rapp Polymere, Germany) was directly mixed
with 500 l 0.3 mM ZnOHMA QDs at 25 C. After reaction
for 2 h, methanol in the reaction mixture was removed by a
rotary evaporator. The water-dispersible PEGylated ZnO QDs
were further purified by dialysis against a dialysis membrane
(3500 Da) for two days. The two types of PEGylated
ZnO QDs were each abbreviated as ZnO-NH-PEG-NH2 and
ZnO-S-PEG-NH2 .
The fluorescence spectra of ZnOHMA QDs and
water-dispersible ZnO QDs were recorded using a quartz
cuvette and a multi-detection microplate reader (Spectramax
M5, Molecular Devices, USA) using an excitation wavelength
of 310 and 280 nm in order to observe the band-edge emission
of ZnO QDs. For transmission electron microscopy (TEM),
the samples were prepared by drop coating on carbon-coated
copper TEM grids and scanned at 200 kV (Philips/FEI Tecnai
20 G2 S-Twin TEM, The Netherlands). The hydrodynamic
diameter and zeta potential were determined by the dynamic
light scattering using a submicron particle size analyzer
(Delsa Nano C, Beckman Coulter, USA). Fourier transform
infrared spectroscopy (FTIR) spectra were obtained on an
FTIR spectrometer (Spectrum 100, Perkin-Elmer, USA)
by the KBr disk method. Thermal gravimetric analysis
was performed by a thermogravimetric analyzer (TGA7,
Perkin-Elmer, USA) from 100 to 750 C at a heating rate of
10 C min1 in nitrogen atmosphere.
2.2. Antibacterial activity test
The antibacterial activities of ZnOHMA QDs and waterdispersible ZnO QDs were evaluated by the number of
bacterial colonies, compared to a negative control, after
incubation with the tested strain Escherichia coli (E. coli) JM
109. The bacteria were grown in nutrient broth inside a shaker
incubator (110 rpm, 35 C) for 24 h, and diluted to 1 107
colony-forming units (CFU) ml1 . 2 ml of 1 107 CFU ml1
E. coli were treated with 2 ml various concentrations (1, 15,
and 30 ppm) of ZnOHMA QDs or water-dispersible ZnO
QDs in 16 ml 2% nutrient broth and placed in the shaker
incubator for 24 h. The grown cultures were diluted to 10, 100,
1000, and 10000-fold in volume and 100 l of each dilution

2. Experimental procedure
2.1. Preparation of water-dispersible ZnO QDs
2 g of Zn(CH3 COO)2 2H2 O, 0.12 g of hydrolyzed
hexahydro-4-methyl phthalic anhydride (HMA) ([HMA]/
[Zn2+ ] = 0.07) and 120 g of methanol were incorporated
into a reactor. 80 g of triethylamine (TEA) was then added
to the mixture under stirring and the reaction was carried
out for four days. After the synthesis, ZnO NPs (denoted
2

Nanotechnology 24 (2013) 475102

S-H Hsu et al

Figure 1. (A) Zn(CH3 COO)2 2H2 O and HMA were used for synthesis of ZnO QDs in methanol. The ZnOHMA QDs were then
precipitated by adding heptane, washed, and redispersed in methanol to get 0.5% ZnOHMA QD dispersions. (B) To prepare
water-dispersible ZnO QDs, either type of the two PEG molecules was added to the ZnOHMA QDs solution. The water-dispersible QDs
generated were each abbreviated as ZnO-S-PEG-NH2 QDs and ZnO-NH-PEG-NH2 .

were re-inoculated onto the plate count agar. The plates were
cultivated at 35 C for 24 h. The numbers of the bacterial
colony were transformed into logarithms of base 10 (log10 ) to
show the antibacterial activity. All experiments were repeated
4 times.

in Hanks buffered salt solution with gentle agitation.

The liberated cells were obtained from the pellets after
centrifugation and then cultured in Dulbeccos modified Eagle
mediumlow glucose (DMEM-LG)/F12 (1:1) supplemented
by 10% fetal bovine serum (FBS), 100 U ml1 penicillin,
1000 U ml1 streptomycin, 5 mM (4-(2-hydroxyethyl)-1piperazineethanesulfonic acid) (HEPES), and 5% bovine
serum albumin fraction V at 37 C in a 5% CO2 incubator.
The culture medium was refreshed twice a week. To confirm
the genuineness of the MSC phenotype, the cell surface
markers of ADAS were characterized by flow cytometry. The
monoclonal antibodies included CD29, CD31 (BioLegend,
USA), CD34 (Santa Cruz Biotechnology, USA), CD44
(BioLegend), CD45 (Santa Cruz Biotechnology), CD73 (BD

2.3. Isolation and culture of ADAS

All animal procedures followed the ethical guidelines for
experimental animals and were approved by the Institutional
Animal Care and Use Committee. ADAS were isolated
from the subcutaneous adipose tissue of Sprague-Dawley
rats. The adipose tissue was minced into several pieces and
digested for 1 h at 37 C in 900 U ml1 type I collagenase
3

Nanotechnology 24 (2013) 475102

S-H Hsu et al

Biosciences, USA), CD90 (BioLegend), and CD105 (BD

Biosciences) antibodies. A flow cytometer (FACS Caliber, BD
Biosciences) was used for the fluorescence analysis of the
percentage of positive cells. ADAS showed the representative
markers of MSCs including CD29, CD44, CD73, CD90,
and D105 (all >90% positive), but were negative for the
hematopoietic stem cell markers (CD31, CD34, and CD45).
Therefore, ADAS used in this study were considered as
typical MSCs. Cells of the third to the fifth passages were used
in this study.

incubator. After 24 h, the culture medium was replaced by

the medium containing 30 ppm water-dispersible ZnO QDs
and was further incubated for 24 h. After extensive washing
by PBS, cells were observed by a confocal microscope (BD
pathway 435 Bio-imager, USA).
To quantify the amount of ZnO uptake, HepG2,
MC3T3-E1, and ADAS cells were seeded in 24-well culture
plates with a density of 3 104 cells cm2 for 12 h. The
medium was replaced by the medium containing 30 ppm
water-dispersible ZnO QDs for another 24 h. After extensive
washing by PBS to remove the free ZnO QDs, cells were
detached with 0.25% trypsin and collected by centrifugation.
The cell pellets were treated with 6% nitric acid and the
zinc concentration was determined by an inductively coupled
plasma-mass spectrometer (ICP-MS) (Sciex Elan 6100 DRC,
Perkin-Elmer, USA). The cellular zinc ion content was
obtained by deducting the value of the blank group (cells
without the exposure of ZnO QDs) from the measured value.

2.4. Cell culture of HepG2 and MC3T3-E1

The cell lines of hepatocytes (HepG2 human hepatoma
cells) and osteoblasts (MC3T3-E1 mouse pre-osteoblasts)
were purchased from the American Type Culture Collection
(ATCC, USA). The culture medium for HepG2 was Eagles
minimum essential medium (MEM) supplemented by 10%
FBS, 100 U ml1 penicillin, and 1000 U ml1 streptomycin.
The culture medium for MC3T3-E1 was alpha modified
minimum essential medium (-MEM) containing 10% FBS,
100 U ml1 penicillin, and 1000 U ml1 streptomycin. The
culture medium was refreshed twice a week. Cells were
subcultured every three to four days.

2.8. The effect of ZnO QDs on ADAS stemness and

differentiation
ADAS cells (3 104 cells cm2 ) were placed into 24-well
culture plates for 12 h. The expression levels of two stemness
marker genes, octamer-binding transcription factor 4 (Oct4)
and SRY (sex determining region Y)-box containing gene 2
(Sox2), were analyzed by the real-time reverse transcription
polymerase chain reaction (RT-PCR) after cells were labeled
by 30 ppm ZnO QDs for 24 h.
For osteogenic, adipogenic, and chondrogenic differentiation, ADAS cells were cultured in osteogenic,
adipogenic, and chondrogenic induction medium for two
weeks after being labeled by ZnO QDs. The osteogenic
induction medium was -MEM supplemented with 10%
FBS, 10 mM -glycerophosphate (Sigma), 200 mg ml1
ascorbate-2-phosphate (Sigma), and 108 M dexamethasone
(Sigma). The adipogenic induction medium was high
glucose DMEM supplemented with 10% FBS, 0.5 mM
isobutyl-methylxanthine (Sigma), 200 mM indomethacin
(Sigma), 106 M dexamethasone and 10 g ml1 insulin.
The chondrogenic induction medium was DMEM-LG/F12
supplemented with 10% FBS, 10 ng ml1 TGF-3 (CytoLab/PeproTech, Rehovot, Israel), 104 M dexamethasone
e, 50 g ml1 ascorbate-2-phosphate, 40 g ml1 Lproline (Sigma), 1% ITS+premix 100 (Sigma). The gene
expression of osteogenic, adipogenic, and chondrogenic
transcription factors, i.e. runt-related transcription factor
2 (Runx2), peroxisome proliferator-activated receptor 2
(PPAR 2), and SRY-box containing gene 9 (Sox9), was
analyzed by the real-time RT-PCR, respectively. The PCR
reaction and detection were performed by a Thermal Cycler
(Bio-Rad MiniOpticonTM Real-Time PCR Detection System,
USA). Total RNA was isolated from cells by the TRIzol
reagent (Invitrogen, USA). The first strand complementary
DNA was synthesized using RevertAidTM H Minus First
Strand cDNA Synthesis Kit (MBI Fermentas, Germany) from
1 g total RNA. The real-time detection for gene expression
was performed using KAPA SYBR Fast qPCR Kits (Applied
Biosystems, USA). The expression levels were normalized to
those of GAPDH.

2.5. MTT assay for cytotoxicity evaluation of

water-dispersible ZnO QDs
HepG2, MC3T3-E1 osteoblasts, and ADAS cells were seeded
in a 96-well culture plate with a density of 3 104 cells cm2
and incubated for 24 h. The culture medium was replaced
by the medium containing water-dispersible ZnO QDs at
different concentrations and cultured for another 24 h.
The cytotoxicity was evaluated by 3-(4,5-dimethyl-thiazol2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. The
absorbance was measured at 570 nm using a microplate reader
and normalized to the negative control to obtain cell viability.
2.6. Reactive oxygen species (ROS) generation
HepG2, MC3T3-E1, and ADAS cells were seeded into
96-well culture plates with a density of 1.5 105 cells cm2
one day before the experiment [39]. On the day of the
experiment, the culture medium was replaced by the medium
containing water-dispersible ZnO QDs at 30 ppm for 2 h and
then incubated with 20 M 20 ,70 -dichlorodihydrofluorescein
diacetate (DCFH-DA) for 30 min. After DCFH-DA was
removed, cells were washed with phosphate-buffered saline
(PBS) and the fluorescence was detected using an excitation
wavelength of 488 nm and an emission wavelength of 530 nm
with the multi-detection microplate reader and normalized to
the blank [40].
2.7. Cellular uptake of water-dispersible ZnO QDs
HepG2, MC3T3-E1, and ADAS cells (3 104 cells cm2 )
were placed into FalconTM 96-well imaging plates (BD
Bioscience, USA), then incubated in a 37 C/5% CO2
4

Nanotechnology 24 (2013) 475102

S-H Hsu et al

Figure 2. High-resolution TEM images of ZnOHMA QDs, ZnO-S-PEG-NH2 QDs, and ZnO-NH-PEG-NH2 QDs. On the right upper
panel of each image, the enlarged TEM image reveals the crystalline structure of QDs.

(23 nm) water-dispersible green-fluorescence ZnO QDs via

synthesis in methanol and PEGylation in aqueous medium, as
summarized in figure 1(A). The ZnOHMA QDs dispersed
in methanol exhibited green fluorescence when excited with a
UV lamp (figure 1(B)). The two water-dispersible ZnO QDs,
ZnO-S-PEG-NH2 and ZnO-NH-PEG-NH2 , also exhibited
green fluorescence. The photoluminescence spectra (PL) of
ZnOHMA, ZnO-S-PEG-NH2 and ZnO-NH-PEG-NH2 QDs
are presented and discussed in the supplementary information
(available at stacks.iop.org/Nano/24/475102/mmedia).
TEM images are presented in figure 2. All ZnO
NPs were of round shape. The particles of ZnOHMA
had a small size of 2 nm (2.26 0.25 nm), while
ZnO-S-PEG-NH2 and ZnO-NH-PEG-NH2 had an average
diameter of 2.720.36 nm and 3.46 0.48 nm, respectively.
The high-resolution TEM image revealed the small size
(2 nm) and the crystalline structure of the QDs (figure 2).
The results demonstrated that the sizes of the hydrophobic
QDs and water-dispersible QDs were close to each other
and the size distribution was quite uniform. The surface
zeta potential and hydrodynamic diameter are shown in
figures 3(A) and (B). ZnOHMA QDs had a negative surface
zeta potential (16.9 5.5 mV) and a hydrodynamic
diameter of 121.9 2.2 nm. The zeta potentials of waterdispersible ZnO QDs were positive (ZnO-S-PEG-NH2 QDs
19.0 1.0 mV; ZnO-NH-PEG-NH2 QDs 15.7 4.9 mV).
The hydrodynamic diameter of water-dispersible ZnO QDs
(ZnO-S-PEG-NH2 QDs 147.6 16.7 nm; ZnO-NH-PEGNH2 QDs 141.7 2.8 nm) were slightly larger than that of
ZnOHMA QDs.

2.9. Assessment of mitochondrial functionality

HepG2 or ADAS cells were seeded into Seahorse XF-24
plates (Seahorse Biosciences, USA) with a density of 3
104 cells cm2 for 24 h. On the day of the experiment,
the injection ports of the sensors were filled with 100 l
of water-dispersible ZnO QDs in the respective culture
medium (vehicle). The oxygen consumption rate (OCR) and
extracellular acidification rate (ECAR) measurements were
performed using a Seahorse Bioscience XF-24 instrument
(Seahorse Bioscience, USA). Five basal rate measurements
were followed with water-dispersible ZnO QDs or vehicle
addition. All average rates were normalized to the vehicle
control.
2.10. Statistical analysis
Data were expressed as mean standard deviation and
analyzed statistically using one-way ANOVA followed by
Scheffes test using SPSS version 14.0 to establish the
significance of any differences. The level of statistical
significance was set at p < 0.05. Each experiment was
repeated independently 3 times to ensure reproducibility.

3. Results and discussion

3.1. Characterization of ZnO QDs
The hydrophilic properties of QDs are usually required for use
in cell labeling. In this study, we fabricated two small-sized
5

Nanotechnology 24 (2013) 475102

S-H Hsu et al

Figure 3. (A) The zeta potential and (B) hydrodynamic diameter of ZnO QDs. (C) FTIR spectra of ZnOHMA QDs, ZnO-S-PEG-NH2
QDs, and ZnO-NH-PEG-NH2 QDs. (D) Thermal gravimetric analysis of ZnOHMA QDs, ZnO-S-PEG-NH2 QDs, and ZnO-NH-PEG-NH2
QDs.

SH-PEG-NH2 , which was double that of NH2 -PEG-NH2 .

Even when fewer molecules of SH-PEG-NH2 were present
on the ZnO QDs, they may account for a higher weight.

The FTIR spectra of ZnOHMA QDs and waterdispersible ZnO QDs are displayed in figure 3(C). The
characteristic absorption band at 1415 cm1 (CO stretching)
was observed for ZnOHMA. After PEG conjugation,
the absorption band at 1415 cm1 disappeared. Instead,
absorption bands ranging from 1000 to 1250 cm1 were
observed, which was attributed to the amine group (CN) in
the structure of PEG. A new absorption peak at 1657 cm1
showed up in the spectrum of ZnO-NH-PEG-NH2 , which
was assigned to the amide I band. The appearance of this
new peak represented a conjugation between the carboxylic
group derived from ZnOHMA and the amine group on
NH2 -PEG-NH2 via EDC [41, 42]. Comparing the spectra
of ZnOHMA and ZnO-S-PEG-NH2 , the specific adsorption
bands at 14001600 cm1 for ZnOHMA disappeared in
ZnO-S-PEG-NH2 . This suggested that SH-PEG-NH2 may
directly bind with ZnO QDs by competing with HMA for
binding sites.
The corresponding TGA curves are shown in figure 3(D).
A weight loss of 27.4% at 500 C for ZnOHMA QDs
was ascribed to the HMA on the surface of the QDs.
The corresponding weight loss of ZnO-S-PEG-NH2 and
ZnO-NH-PEG-NH2 was 64.6% and 51.8% respectively. The
per cent weight loss at high temperature represented the
organic fraction (HMA or PEG) of the ZnO QDs. The
greater weight loss (or organic fraction) of ZnO-S-PEG-NH2
may be attributed to the molecular weight of the stabilizer

3.2. Antibacterial efficacy of ZnO QDs

Concentration is one of the main factors that affect the
antibacterial properties of ZnO NPs. In this study, the
antibacterial efficacy was examined in three concentrations
of QDs (1, 15, and 30 ppm). The activity of various ZnO
QDs against bacterial growth is shown in figure 4(A). It
was clear that ZnOHMA QDs, ZnO-S-PEG-NH2 QDs,
and ZnO-NH-PEG-NH2 QDs at 15 ppm and 30 ppm
significantly inhibited the growth of E. coli. However, none
of them significantly reduced bacterial growth at 1 ppm. In
comparison to the antibacterial properties of silver NPs at
low concentrations [40], ZnO QDs were not as effective. It
was worth mentioning that at the concentration of 15 ppm the
antibacterial efficacy of ZnOHMA QDs was better than the
PEGylated ZnO QDs. ZnO QDs were reported to distort and
damage the bacterial cell membrane via generation of ROS
and altering the surface electrostatic interactions, resulting in
a leakage of cytoplasmic content [24, 43]. We hypothesize that
conjugation of the PEG molecules onto ZnOHMA QDs may
introduce a barrier (e.g. steric hindrance) between the bacteria
cell wall and QDs [22], causing a reduction of the antibacterial
efficiency.
6

Nanotechnology 24 (2013) 475102

S-H Hsu et al

Figure 4. (A) The antibacterial efficacy of ZnOHMA QDs, ZnO-S-PEG-NH2 QDs, and ZnO-NH-PEG-NH2 QDs. (B)(D) The viability
of HepG2, MC3T3-E1, and ADAS cells exposed to ZnO-S-PEG-NH2 QDs and ZnO-NH-PEG-NH2 QDs of different concentrations for
24 h.

ZnO-NH-PEG-NH2 QDs is demonstrated in figure 5. The

level of ROS was increased in ADAS (figure 5(C)) upon
exposure to ZnO QDs but not in HepG2 (figure 5(A)) or
MC3T3-E1 (figure 5(B)). The above results were consistent
with the general findings in the literature that stem cells are
more susceptible to QD toxicity [24].

Many investigations about the cytotoxicity and antibacterial properties of silver NPs have been published. In spite of
their remarkable antibacterial properties, silver NPs demonstrate mitochondrial dysfunction and induction of ROS which
in turn set off DNA damage and chromosomal aberrations.
Future applications of silver NPs as an antiproliferative agent
could be limited by the fact that they are equally toxic to
normal cells [44]. In comparison, the major advantage of ZnO
QDs is the biocompatibility and low cytotoxicity, especially
for human stem cells.

3.4. Cellular uptake of water-dispersible ZnO QDs

The fluorescent images for cells labeled with waterdispersible ZnO QDs are shown in figure 6(A). In comparison
to the negative control (cells without labeling, as shown in the
supplementary information available at stacks.iop.org/Nano/
24/475102/mmedia), HepG2 and ADAS labeled with either
ZnO QDs exhibited bright green fluorescence at 30 ppm,
suggesting that both QDs were successfully taken up by the
cells under these circumstances. However, the fluorescence
signal of labeled MC3T3-E1 was substantially weaker. The
fluorescent images of labeled HepG2 were not as clear
because HepG2 tended to form clusters. The amount of
cellular uptake of ZnO QDs, expressed as the mass of zinc,
is shown in figure 6(B). The amount of cellular uptake was
the greatest for ADAS (ZnO-S-PEG-NH2 QDs 0.99 pg/cell
on average and ZnO-NH-PEG-NH2 QDs 1.7 pg per cell on
average), followed by HepG2 (0.33 pg and 0.73 pg/cell
for each type of QD), and finally MC3T3-E1 (0.13 pg
and 0.29 pg/cell for each type of QD). Based on these
results, all cells were successfully labeled at 30 ppm of
ZnO QDs. Literature has reported that the concentration

3.3. Cytotoxicity and quantification of ROS generation in

cells exposed to water-dispersible ZnO QDs
Cytotoxicity tests of water-dispersible ZnO QDs were
conducted in various concentrations from 0 to 60 ppm and
in three different types of cells, i.e. HepG2, MC3T3-E1,
and ADAS. Results are shown in figures 4(B)(D). When
the concentration of water-dispersible ZnO QDs was below
or equal to 45 ppm, nearly all (>95%) HepG2 or
MC3T3-E1 survived. At 60 ppm, ZnO-S-PEG-NH2 was
not significantly cytotoxic to HepG2 and MC3T3-E1, while
ZnO-NH-PEG-NH2 caused some cytotoxicity, in particular
to MC3T3-E1. In contrast to HepG2 and MC3T3-E1, ADAS
were more susceptible to the cytotoxic effect of ZnO QDs.
As shown in figure 4(D), both water-dispersible ZnO QDs
at the concentration of 45 and 60 ppm had caused apparent
cytotoxicity to ADAS. The ROS generation for different
cells exposed to 30 ppm of ZnO-S-PEG-NH2 QDs or
7

Nanotechnology 24 (2013) 475102

S-H Hsu et al

Figure 5. ROS levels for HepG2, MC3T3-E1, and ADAS cells exposed to ZnO-S-PEG-NH2 QDs and ZnO-NH-PEG-NH2 QDs at the
concentration of 30 ppm for 2 h.

a supplementary report (figure S3 available at stacks.iop.

org/Nano/24/475102/mmedia). Although these QDs may
aggregate slightly during storage, they could be again
dispersed upon gentle shaking. Because the prepared ZnO
QDs were well dispersed in an aqueous environment, they
could enter cells efficiently and remained fluorescent inside
the cells. Therefore, the aggregation did not affect the
application in cell labeling.
A comparison of cytotoxicity and cellular uptake data
revealed that the ZnO QD-induced cytotoxicity as well as
ROS generation was positively correlated with the amount of
cellular uptake of ZnO QDs. The successful uptake of ZnO
QDs at the safe concentration of 30 ppm ensured a good
labeling efficiency for the cells. The cytotoxicity of ZnO QDs
may be attributed to the dissolved zinc ions (Zn2+ ) in the
medium, which raise the intracellular Zn2+ and ROS to induce
cell apoptosis [29, 47, 48].

of magnetic Fe3 O4 NPs intended for ADAS labeling was

25 ppm [45]. In addition, most commercial QDs require
surface coating by transfection agents in order to be taken
up by cells effectively. Therefore, the water-dispersible
ZnO QDs developed in this study were endocytosed quite
efficiently.
In the synthetic process of ZnO QDs, hexahydro-4methyl phthalic anhydride (HMA) and PEG were used
to modify the surface, and improved the stability of
QD in aqueous environment so they could be dispersed
homogeneously. The prepared ZnO QDs remained stable in
dispersion for at least six months. The literature indicates
that the value of zeta potential is highly associated with
the size of QDs, which was about 10 mV as the QDs
size was about 5 nm. The lower potential means weak
resistance to flocculation [46]. In our study the zeta
potential of ZnOPEG was about 15 mV, larger than
that in the literature, indicating that our synthesized ZnO
QDs had good resistance to aggregation. In addition, the
experimental results of cell labeling (figure 6) in this
study showed that all cells (including HepG2, MC3T3-E1,
and ADAS) were successfully labeled at 30 ppm of ZnO
QDs. In comparison, the fluorescence images of ADAS
cells without being labeled by ZnO QDs is shown in

3.5. The effect of water-dispersible ZnO QDs on the stemness

and differentiation of stem cells
The expression of stemness marker genes Sox2 and Oct4
for ADAS labeled with 30 ppm of ZnO-S-PEG-NH2
QDs is demonstrated in figure 7(A). The expressions of
8

Nanotechnology 24 (2013) 475102

S-H Hsu et al

function of the cellular energy metabolism. The mitochondrial

activity of HepG2, as shown in the electronic supplementary
data (available at stacks.iop.org/Nano/24/475102/mmedia),
revealed a high tolerance for both ZnO QDs. This may
account for the high cell viability observed for HepG2 upon
exposure to ZnO QDs. The change of OCR for ADAS exposed
to ZnO-S-PEG-NH2 QDs is shown in figure 8(A). It is
evident that the exposure to ZnO QDs decreased the OCR of
ADAS. On the other hand, the exposure did not affect ECAR
(figure 8(B)) and at high concentration the exposure caused
an active burst of glycolysis in ADAS during 6090 min. This
burst represented that cells strived to compensate for the lost
oxygen consumption. This maintenance and compensation
effect of ECAR also occurred in HepG2. We hypothesize that
the reduced oxygen and increased glycolysis consumption of
ADAS may be related to the enhanced differentiation capacity
of ADAS after ZnO QD exposure.
The observation that cells strived to compensate for
the insufficient adenosine triphosphate (ATP) by glycolysis
is particularly interesting. Based on the literature, a small
amount of cerium oxide NPs may elicit antioxidant protection
in exposed cells [49]. We hypothesize that ZnO QDs in
safe concentrations may provoke a protective mechanism by
reducing oxygen consumption. This may present a condition
similar to hypoxia, which was recently found to increase the
stemness and multipotency of ADAS [50]. Besides, stem cells
heavily rely on glycolysis, or sugar fermentation, to drive
their metabolic activities. The literature has demonstrated
a significant correlation between the metabolic phenotype
and pluripotent developmental stage of the stem cells and
their functions [51]. The maintenance of stemness and the
increased differentiation capacity of ADAS in our study
may be associated with the metabolic shift from oxidative
phosphorylation to glycolysis in the presence of QDs. In
summary, the ZnO-S-PEG-NH2 QDs developed in this study
possess good antibacterial activity and can be taken up by cells
efficiently. At 30 ppm, they are not cytotoxic and do not impair
the functions of stem cells. They can be used in cell labeling
and tracking.

Figure 6. (A) The fluorescence images of HepG2, MC3T3-E1, and

ADAS cells each labeled with ZnO-S-PEG-NH2 QDs and
ZnO-NH-PEG-NH2 QDs respectively (24 h treatment) were
acquired by confocal microscopy. Cells were incubated for 24 h.
The culture medium was then replaced by the medium containing
30 ppm water-dispersible ZnO QDs. After 24 h, cells were washed
extensively by PBS and observed by the confocal microscope.
(B) The amount of ZnO QD uptake by HepG2, MC3T3-E1, and
ADAS cells, was quantified by ICP-MS. To quantify the amount of
ZnO QDs uptake, labeled cells were digested and the zinc
concentration was determined by the ICP-MS.

4. Conclusion

Sox2 and Oct4 were not significantly alerted for labeled

cells. This indicated that ADAS cells maintained their
stemness after taking up the water-dispersible ZnO QDs.
The differentiation ability of labeled ADAS is shown in
figure 7(B). The gene expression of transcription factors
associated with the osteogenic, adipogenic, and chondrogenic
differentiations (Runx2, PPAR 2, and Sox9, respectively)
significantly increased after induction. This suggested that
water-dispersible ZnO QDs may not adversely affect the
differentiation capacity of ADAS, and instead may improve
it.

Water-dispersible ZnO QDs were prepared from chemical

synthesis and PEG conjugation in this study. They
demonstrated small size (23 nm) and green fluorescence.
The prepared ZnO QDs had antibacterial activity and good
biocompatibility to HepG2 and osteoblasts. Although stem
cells such as ADAS were more susceptible to ZnO QD
cytotoxicity, a safe and efficient labeling was observed at
30 ppm of ZnO QDs. The ROS analysis indicated that ADAS
were easily affected by ZnO QDs because of the greater
ROS generation. However, the reduced oxygen consumption
of ADAS exposed to a low amount of ZnO QDs may facilitate
their differentiation by metabolic compensation through
glycolysis. Therefore, the water-dispersible ZnO-S-PEG-NH2
QDs at 30 ppm can be developed as potential fluorescent cell
trackers.

3.6. The effect of water-dispersible ZnO QDs on energy

metabolism
The in situ metabolic OCR and ECAR ratios each represent
the oxidative phosphorylation function and glycolysis
9

Nanotechnology 24 (2013) 475102

S-H Hsu et al

Figure 7. (A) The expression of stemness marker genes Sox2 and Oct4 for ADAS cultured for 24 h and then challenged with
water-dispersible ZnO QDs (30 ppm) for 36 h, analyzed by the real-time RT-PCR. p < 0.05 among the indicated groups. (B)(D) The gene
expression of Sox9, Runx2, and PPAR 2 for ADAS labeled with water-dispersible ZnO QDs each after chondrogenic, osteogenic, and
adipogenic induction for two weeks. p < 0.05 among the indicated groups. These data suggest that the stemness and multipotency of
ADAS may not be adversely affected by QD labeling.

Figure 8. (A) The oxygen consumption rate (OCR) and (B) extracellular acidification rate (ECAR) of ADAS upon exposure to
ZnO-S-PEG-NH2 QDs at 30 and 60 ppm. In the plots, the challenge was given at 35 min (shown as the dotted line) and last for a period of
85 min (end point: 120 min).

Acknowledgments

References
[1] Iyer G, Michalet X, Chang Y-P, Pinaud F F, Matyas S E,
Payne G and Weiss S 2008 Nano Lett. 8 4618
[2] Jayagopal A, Su Y R, Blakemore J L, Linton M F, Fazio S and
Haselton F R 2009 Nanotechnology 20 165102
[3] Marks K M and Nolan G P 2006 Nature Methods 3 591
[4] Kirchner C, Liedl T, Kudera S, Pellegrino T, Munoz Javier A,
Gaub H E, Stolzle S, Fertig N and Parak W J 2004 Nano
Lett. 5 331

This research was supported by the National Research

Program on Nanoscience and Technology sponsored by
the National Science Council (NSC 101-2120-M-002-002
and NSC 101-2321-B-002-039). Technical support for the
extracellular flux analyzer was kindly provided by Cell-Bio
Biotechnology Co., Ltd.
10

Nanotechnology 24 (2013) 475102

S-H Hsu et al

[30] Heng B C, Zhao X, Xiong S, Kee W N, Boey F Y-C and

Loo J S-C 2010 Food Chem. Toxicol. 48 1762
[31] Heng B C, Zhao X, Xiong S, Ng K W, Boey F Y-C and
Loo J S-C 2011 Arch. Toxicol. 85 695
[32] Nair S, Sasidharan A, Divya Rani V V, Menon D, Nair S,
Manzoor K and Raina S 2009 J. Mater. Sci.-Mater. Med.
20 235
[33] Premanathan M, Karthikeyan K, Jeyasubramanian K and
Manivannan G 2011 Nanomed. Nanotechnol. Biol. Med.
7 184
[34] Wahab R, Kaushik N, Verma A, Mishra A, Hwang I H,
Yang Y-B, Shin H-S and Kim Y-S 2011 J. Biol. Inorg.
Chem. 16 431
[35] Wang Y, Aker W G, Hwang H-M, Yedjou C G, Yu H and
Tchounwou P B 2011 Sci. Tot. Environ. 409 4753
[36] Zhang L W, Yu W W, Colvin V L and Monteiro-Riviere N A
2008 Toxicol. Appl. Pharmacol. 228 200
[37] Jeng H A and Swanson J 2006 J. Environ. Sci. Health A
41 2699
[38] Reddy K M, Feris K, Bell J, Wingett D G, Hanley C and
Punnoose A 2007 Appl. Phys. Lett. 90 213902
[39] Heng B C, Zhao X, Tan E C, Khamis N, Assodani A, Xiong S,
Ruedl C, Kee W N and Loo J S-C 2011 Arch. Toxicol.
85 1517
[40] Chi T-Y, Yeh H-Y, Lin J-J, Jeng U S and Hsu S-H 2013
J. Mater. Chem. B 1 2178
[41] Araki J, Wada M and Kuga S 2000 Langmuir 17 21
[42] Gooding J J and Hibbert D B 1999 TrAC-Trends Anal. Chem.
18 525
[43] Liu Y, He L, Mustapha A, Li H, Hu Z Q and Lin M 2009
J. Appl. Microbiol. 107 1193
[44] AshaRani P V, Low Kah Mun G, Hande M P and
Valiyaveettil S 2008 ACS Nano 3 279
[45] Hsu S-H, Ho T-T and Tseng T-C 2012 Biomaterials 33 3639
[46] Wood A, Giersig M, Hilgendorff M, Vilas-Campos A,
Liz-Marz L M and Mulvaney P 2003 Aust. J. Chem.
56 1051
[47] Deng X, Luan Q, Chen W, Wang Y, Wu M, Zhang H and
Jiao Z 2009 Nanotechnology 20 115101
[48] Simon H-U, Haj-Yehia A and Levi-Schaffer F 2000 Apoptosis
5 415
[49] Xia T, Kovochich M, Liong M, Madler L, Gilbert B, Shi H,
Yeh J I, Zink J I and Nel A E 2008 ACS Nano 2 2121
[50] Zhou W et al 2012 EMBO J. 31 2103
[51] Kondoh H, Lleonart M E, Nakashima Y, Yokode M,
Tanaka M, Bernard D, Gil J and Beach D 2007 Antioxid.
Redox Signaling 9 293

[5] Lin C-A J, Liedl T, Sperling R A, Fernandez-Arguelles M T,

Costa-Fernandez J M, Pereiro R, Sanz-Medel A,
Chang W H and Parak W J 2007 J. Mater. Chem. 17 1343
[6] Hsieh S-C, Wang F-F, Lin C-S, Chen Y-J, Hung S-C and
Wang Y-J 2006 Biomaterials 27 1656
[7] Hsieh S-C, Wang F-F, Hung S-C, Chen Y-J and Wang Y-J
2006 J. Biomed. Mater. Res. B 79B 95
[8] Ranjbarvaziri S, Kiani S, Akhlaghi A, Vosough A,
Baharvand H and Aghdami N 2011 Biomaterials 32 5195
[9] Bagnall D M, Chen Y F, Zhu Z, Yao T, Shen M Y and Goto T
1998 Appl. Phys. Lett. 73 1038
[10] Kong X Y, Ding Y, Yang R and Wang Z L 2004 Science
303 1348
[11] Wong E M and Searson P C 1999 Appl. Phys. Lett. 74 2939
[12] Djurisic A B, Leung Y H, Tam K H, Ding L, Ge W K,
Chen H Y and Gwo S 2006 Appl. Phys. Lett. 88 103107
[13] Gong Y, Andelman T, Neumark G, OBrien S and
Kuskovsky I 2007 Nanoscale Res. Lett. 2 297
[14] Chu S-Y, Yan T-M and Chen S-L 2000 J. Mater. Sci. Lett.
19 349
[15] Damonte L C, Mendoza Zelis L A, Mar Soucase B and
Hernandez Fenollosa M A 2004 Powder Technol. 148 15
[16] He C, Sasaki T, Usui H, Shimizu Y and Koshizaki N 2007
J. Photochem. Photobiol. A 191 66
[17] Li M, Hari B, Lv X, Ma X, Sun F, Tang L and Wang Z 2007
Mater. Lett. 61 690
[18] Starowicz M and Stypula B 2008 Eur. J. Inorg. Chem.
2008 869
[19] Yoshida T, Terada K, Schlettwein D, Oekermann T, Sugiura T
and Minoura H 2000 Adv. Mater. 12 1214
[20] Jang E S, Won J-H, Hwang S-J and Choy J-H 2006 Adv.
Mater. 18 3309
[21] Burnett M E and Wang S Q 2011 Photodermatol.
Photoimmunol. Photomed. 27 58
[22] Diffey B L and Grice J 1997 Br. J. Dermatol. 137 103
[23] Perelshtein I et al 2013 J. Mater. Chem. B 1 1968
[24] Zhang L, Jiang Y, Ding Y, Povey M and York D 2007
J. Nanopart. Res. 9 479
[25] Jones N, Ray B, Ranjit K T and Manna A C 2008 FEMS
Microbiol. Lett. 279 71
[26] Sinha R, Karan R, Sinha A and Khare S K 2011 Bioresour.
Technol. 102 1516
[27] Tang X, Choo E S G, Li L, Ding J and Xue J 2009 Langmuir
25 5271
[28] Xiong H-M, Xu Y, Ren Q-G and Xia Y-Y 2008 J. Am. Chem.
Soc. 130 7522
[29] Brunner T J, Wick P, Manser P, Spohn P, Grass R N,
Limbach L K, Bruinink A and Stark W J 2006 Environ. Sci.
Technol. 40 4374

11