Article review

RESEARCH ARTICLE

Impact of exposure time, particle size and uptake pathway on silver nanoparticle
effects on circulating immune cells in mytilus galloprovincialis

Younes Bouallegui, Ridha Ben Younes, Faten Turki and Ridha Oueslati

Research Unit for Immuno-Microbiology Environmental and Cancerogenesis, Sciences Faculty of Bizerte, University of Carthage, Bizerte, Tunisia

ABSTRACT
Nanomaterials have increasingly emerged as potential pollutants to aquatic organisms. Nanomaterials are
known to be taken up by hemocytes of marine invertebrates including Mytilus galloprovincialis. Indeed,
assessments of hemocyte-related parameters are a valuable tool in the determination of potentials
for nanoparticle (NP) toxicity. The present study assessed the effects from two size types of silver nanopar-
ticles (AgNP: <50 nm and <100nm) on the frequency of hemocytes subpopulations as immunomodula-
tion biomarkers exposed in a mollusk host. Studies were performed using exposures prior to and after
inhibition of potential NP uptake pathways (i.e. clathrin- and caveolae-mediated endocytosis) and over dif-
ferent durations of exposure (3, 6 and 12 h). Differential hemocyte counts (DHC) revealed significant varia-
tions in frequency of different immune cells in mussels exposed for 3 hr to either AgNP size. However, as
exposure duration progressed cell levels were subsequently differentially altered depending on particle
size (i.e. no significant effects after 3 h with larger AgNP). AgNP effects were also delayed/varied after
blockade of either clathrin- or caveolae-mediated endocytosis. The results also noted significant negative
correlations between changes in levels hyalinocytes and acidophils or in levels basophils and acidophils
as a result of AgNP exposure. From these results, we concluded AgNP effects on mussels were size and
duration of exposure dependent. This study highlighted how not only was NP size important, but that dif-
fering internalization mechanisms could be key factors impacting on the potential for NP in the environ-
ment to induce immunomodulation in a model/test sentinel host like M. galloprovincialis.

ARTICLE HISTORY
Received 13 February 2017
Revised 6 May 2017
Accepted 24 May 2017

KEYWORDS
Silver nanoparticles;
endocytosis; hyalinocytes;
granulocytes; Pappenheim
panoptical staining

Introduction

Nanoparticles (NP) are defined as materials with all dimensions
in nanoscale [1–100 nm] (Luoma 2008). Silver nanoparticles
(AgNP) have become the fastest growing product category in
nanotechnology due to their thermoelectrical conductivity, cata-
lytic activity and nonlinear optical behavior and have great value
in the formulation of inks, microelectronic products and biomed-
ical facilities (i.e. imaging devices) (Tiede et al. 2009; Katsumiti
et al. 2015). Their exceptional broad-spectrum bactericidal prop-
erties and biocompatibility (i.e. as drug delivery agent) have also
made AgNP extremely useful in a diverse range of consumer
goods (Luoma 2008; Rainville et al. 2014; Cozzari et al. 2015;
Katsumiti et al. 2015; Marisa et al. 2016).

Worldwide AgNP production is estimated at � 55 tonne/yr
(Piccinno et al. 2012). However, release of AgNP into aquatic
environs can happen through wastewaters generated during
AgNP synthesis and/or incorporation into goods and consumer
products (Canesi et al. 2012; Matranga & Corsi 2012; Katsumiti
et al. 2015; Marisa et al. 2016). As such, AgNP have emerged as
potential stressors that might enter marine environment (Luoma
2008). A lack of appropriate tools to evaluate effective NP
(of AgNP in particular) levels in aquatic environments make
selection of appropriate testing levels a major problem in risk
assessment of engineered NP. As a result, predicted environmen-
tal concentrations for AgNP are often set at a level of
� 0.01 lg/L (Tiede et al. 2009; Katsumiti et al. 2015). Even so,

levels much lower than that have commonly been used in aquatic
species ecotoxicity tests (1–100 lg/L) (Tiede et al. 2009; Canesi &
Corsi 2016), including those with mollusk models.

In the mussel Mytilus galloprovincialis (filter-feeding organ-
ism), hemocytes are hemolymph cells responsible for immune
defence and serve as a first line of defence against foreign substan-
ces (Gosling 2003; Parisi et al. 2008; Giron-Perez 2010; Matozzo &
Bailo 2015). Immune defences carried out by hemocytes constitute
important targets for potential NP toxicity (Canesi et al. 2012;
Canesi & Prochazkova 2013; Katsumiti et al. 2015).

Several studies have shown that different NP types, that is, car-
bon black, C60 fullerenes, TiO2, SiO2, ZnO, CeO2, Cd-based, Au-
based and Ag-based, are rapidly taken up by hemocytes.
Internalization of these NP subsequently impacted on morpho-
logic/functional characteristics including immune responses
(Canesi et al. 2008, 2010a, b, 2012; Katsumiti et al. 2015; Marisa
et al. 2016). Various mussel hemocyte parameters, including total
hemocyte count (THC), differential hemocyte count (DHC),
hemocyte viability, phagocytic activity and lysosomal membrane
stability, have been used as a tool for screening of immunomodu-
latory effects of differing NP (Matozzo et al. 2007; Parisi et al.
2008; Hoher et al. 2013; Matozzo & Bailo 2015; Canesi & Corsi,
2016; Marisa et al. 2016). Specifically, hyalinocytes and granulo-
cytes have been assessed for morphological changes among hemo-
cytes in Mytilus galloprovincialis (Pipe et al. 1997; Chang et al.
2005; Garcia-Garcia et al. 2008).

CONTACT Younes Bouallegui [email protected] Research Unit of Immuno-Microbiology Environmental and Cancerogenesis, Sciences Faculty of
Bizerte, Zarzouna 7021, Bizerte, Tunisia
� 2017 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits
unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

JOURNAL OF IMMUNOTOXICOLOGY, 2017
VOL. 14, NO. 1, 116–124
https://doi.org/10.1080/1547691X.2017.1335810

While granulocytes are large ovoid-shaped cells with a small
eccentric nucleus and granulated cytoplasm (low nucleus/cyto-
plasm [N/C] ratio) that are able to spread out and produce pseudo-
podia), hyalinocytes are small round cells with an agranular (zero-
few granules) small cytoplasm surrounding a large nucleus (high
N/C ratio) (Carballal et al. 1997; Parisi et al. 2008; Cima 2010;
Matozzo & Bailo 2015). Overall, hemocytes can be classified into
two types, granulocytes and hyalinocytes (so-called agranulocytes),
based on morphological characteristics (the presence/absence of
granules in cytoplasm). Staining of the cytoplasm by certain dyes
allows for sub-distinguishing of acidophils from basophils among
the granulocytes. Ultimately, the basophils of M. edulis appear as
granulocytes with small granules, while acidophilic granulocytes
contain large granules. In comparison to the granulocytes, hyalino-
cytes in bivalve have only basophilic properties. Thus, in earlier
studies that described hemocyte subpopulations, the author indi-
cated that basophilic cells (hyalinocytes þ basophils) made up
about 40% of the total hemocyte pool in bivalves/mussels while
eosinophils accounted for the remaining � 60% of all hemocytes
(Chang et al. 2005; Garcia-Garcia et al. 2008).

Cellular uptake by endocytosis (clathrin- or caveolae-mediated
routes) are crucial for a variety of cellular and physiological
activities (i.e. nutrient uptake, immune defence) (Haucke 2006;
Sandvig et al. 2011); each has also been identified as potential
means for NP entry into cells (Moore 2006; dos Santos et al.
2011; Khan et al. 2015). Clathrin-dependent endocytosis involves
formation of a clathrin (protein)-coated pit used in enzymatic
destruction of internalized contents. Caveolae-dependent endo-
cytosis occurs via cell-surface flask-shaped invaginations enriched
with caveolin (cholesterol-binding proteins) (Nichols &
Lippincott-Shwartz 2001; Razani & Lisanti 2002) that permit sub-
cellular movements of ingested materials through a series of
endosomal compartments of increasing acidity allowing for
hydrolytic breakdown (Moore 2006; Puthenveedu & von Zastrow
2006; Doherty & McMahon 2009). Each route can be modified
with inhibitors (Moore 2006; Ivanov 2008; dos Santos et al. 2011;
Khan et al. 2015). Clathrin-mediated endocytosis could be inhib-
ited by the antiviral amantadine through disruption of the cla-
thrin coat, while antibiotic nystatin can impact on cholesterol-
rich microdomains of caveolae-mediated endocytosis (Ivanov
2008; Khan et al. 2015).

In this context, the present study aimed to record the vari-
ation in the percentages of circulating subpopulations of hemo-
cytes, using as method differential hemocytes count [DHC] after
Pappenheim’s panoptical staining [MGG] to: (1) assess effects of
AgNP on circulating hemocyte sub-populations; (2) establish a
relationship linking length of exposure to different size AgNP
and variations in sub-populations [DHC]; and (3) evaluate the
role of uptake pathways (clathrin- and caveolae-dependent endo-
cytosis) – as well as changes in their function – in the effect of
NP on circulating hemocyte subpopulations.

Material and methods

Silver nanoparticles (AgNP) source and characterization

Poly-vinyl-pyrrolidone (PVP)-coated AgNP of <100 nm (99.5%
pure) were purchased from Sigma (Steinheim, Germany). PVP-
coated AgNP <50 nm were produced by a modified process
wherein AgNO3 (Sigma) was dissolved in ethylene glycol (EG)
solvent (ACROS Organics, 98%, Geel, Belgium) in the presence
of PVP (K30, Sigma) as a capping agent (Mezni et al. 2014a,b).

A stock solution of each AgNP size was suspended in artificial
seawater (ASW; 58.5% NaCl; 26.5% MgCl2; 9.8% Na2SO4; 2.8%

CaCl2; 1.65% KCl; 0.5% NaHCO3; 0.24% KBr; 0.07% H3BO3;
0.0095% SrCl2; 0.007% NaF (Pinsino et al. 2015)). Prior to use,
each AgNP stock was mixed several times and an aliquot removed
as a working solution that was sonicated 15 min in alternating
cycles (2 � 30 s) in an ultrasonic bath (VWR, Strasbourg, France).
Primary physicochemical properties of each AgNP was confirmed
by transmission electron microscopy (TEM) coupled with a micro-
analysis characterization (TECNAI G20, Ultra-Twin, FSB, Bizerte,
Tunisia) and ultraviolet-visible (UV-Vis) spectroscopy (T60; PG-
Instruments, Leicestershire, UK). X-ray diffraction (XRD) charac-
terization was performed using a D8 Advance diffracto-meter
(Bruker, Bizerte), with analyses performed in Bragg–Brentano con-
figuration at 40 kV and 40 mA.

Endocytotic internalization blockers

A stock solution of amantadine (3 mg/mL; Sigma, Steinheim,
Germany) was prepared in ultrapure water. Nystatin (Sigma)
stock solution (5 mg/mL; Sigma) was prepared in dimethyl sulf-
oxide (DMSO) vehicle (Sigma); the final concentration of DMSO
in all Nystatin exposures was 0.05% (v/v). Exposures to vehicle
alone or in the presence of AgNP of differing sizes were con-
ducted to assure effects were not caused by any carrier modula-
tion of NP behavior or by the carrier itself. Effective
concentration ranges used were chosen based on previous study
by Khan et al. (2015).

Sampling and experimental design

Mature mussels (M. galloprovincialis) of average shell length 75
[±5] mm were collected from Bizerte lagoon (Tunisia) and main-
tained in oxygenated ASW (35% salinity, pH 8.0; as for local nat-
ural seawater) in static tanks under standard conditions
(aeration, 12/12 h photoperiod, 16 �C). Animals used for exposure
experiments were acclimated for 1–3 days (Canesi et al. 2010b)
and were not fed during either acclimation or exposure.
Exposure in each tank was 1 mussel/0.5 L ASW in all studies. As
only predicted environmental concentrations (PEC) were avail-
able in literature, the chosen dose of 100 lg AgNP/L was selected
as the test concentration; this dose is usually used in ecotoxicity
tests on aquatic species and would be effective in producing
adverse effects that could be correlated with outcomes of previ-
ous in vitro studies (Katsumiti et al. 2015; Canesi & Corsi 2016).

Mussels (n ¼ 10/group) were separately exposed to AgNP
<50 nm (AgNP50) or AgNP <100 nm (AgNP100) for 3, 6 and
12 h with/without initial treatment with the pharmaceutical
inhibitors. For inhibitor-treated groups, mussels were incubated
for 3 h with 100 lM amantadine (AMA), then placed in AgNP
exposure solutions (without AMA) for the required times. For
nystatin (NYS), mussels were exposed with 50 lM NYS for 1 h
before and then continuing over into the AgNP exposure time-
frames (Ivanov 2008; Angel et al. 2013; Khan et al. 2015).
Control groups (n ¼ 10) of mussels were maintained in oxygen-
ated tanks of only ASW and/or ASW with the inhibitors exactly
as above with the AgNP treatments. All exposures were done in
triplicate.

Pappenheim’s panoptical staining (MGG) and differential
hemocyte counts (DHC)

At the completion of the given exposure period, hemolymph
samples were quickly withdrawn (to minimize stress inflicted)

JOURNAL OF IMMUNOTOXICOLOGY 117

from the adductor muscles of each animal, using nn 18-G needle
fitted onto a 3-mL syringe. All samples were collected at 16 �C.
For each sample, hemolymph of all 10 individuals/treatment regi-
men was pooled; the material was then filtered through 1-mm2

mesh sterile gauze into a 5-mL tube at 4 �C to avoid aggregation
(Canesi et al. 2010a). After mixing, 40 lL aliquots were deposited
onto glass slides; after 15 min, the attached cells were fixed with
methanol and then the hemocytes were stained with May-
Gr€unwald solution (Bio-optica, Milan, Italy). Slides were then
counterstained with 5% Giemsa, air-dried and then mounted
using a mounting medium (Entellan Neu, Merck, Darmstadt,
Germany) and cover slipped. Slides were then evaluated using a
GX-10 light microscope (Olympus, Tokyo, Japan); differential
hemocyte counts were made from counts of stained cells in 10
different fields/slide. A minimum of 350 cells/slide was counted.
Ten slides/experimental condition were evaluated.

Statistical analysis

All results are expressed as percentages (±SD) of total hemocytes.
Normal distribution and homogeneity of variance were tested
using Shapiro–Wilk and Bartlett tests prior to statistical analysis.
Statistical analysis of absolute percentages was performed using a
one-way analysis of variance (ANOVA) with a Tukey’s HSD post
hoc test. Modulation in the percentages of hemocyte subpopula-
tions were compared to those of controls (untreated mussels).
Correlation tests were used to determine relationships among
modulated hemocyte subpopulations. Significance overall and
within any correlation (confirmed by linear regression test) was
accepted at p < 0.05.

Results

Source and characterization of AgNP

Purchased AgNP (<100 nm; AgNP100) were characterized; charac-
terizations met the manufacturer supplied valued (99.5% trace
metal basis). Representative TEM showed homogeneous spherical
characteristics with an approximate primary size of 90 nm (Figure
1(A)); size distribution histograms revealed a median size of 85.0
[±32.6] nm (Figure 1(C)). Representative TEM of synthesized
AgNP (<50 nm; AgNP50) demonstrated homogeneous spherical
characteristics with an approximate size of 50 nm (Figure 1(B));
size distribution histograms revealed a median size of 41.6 [±18.8]
nm (Figure 1(D)). Analyses of each sample indicated that the level
of particles <50 nm within the AgNP100 mixture was � 1.38/each
100 particles from AgNP mixture (i.e. <1.5%).

The XRD pattern recorded from a representative batch of sil-
ver powder is shown in Figure 1(E). The crystalline nature of the
AgNP was demonstrated by diffraction peaks that matched the
face-centered cubic (fcc) phase of silver. The absorption max-
imum of the measured UV-vis spectrum of the colloidal solution
provides information on the average particle size, whereas its full
width at half-maximum (fwhm) can be used to estimate particle
dispersion as demonstrated by Leopold and Lendl (2003).
Agglomeration status analyses performed prior to exposure was
confirmed by absorbance spectra measures at kmax ¼ 400 nm
(Figure 1(F)) that clearly indicated the AgNP had a homogenous
dispersion in aqueous solutions.

Determination of hemocyte subpopulations

Evaluations based on cytoplasmic granules (presence or absence)
and stained granule color (Figure 2) showed that levels of

circulating hemocytes from mussels exposed to AgNP suspen-
sions at the same dose (100 lg/L) varied as a function of differing
particle size. For example, when exposed to AgNP50 for only 3 h,
mussels evinced a significant increase in acidophilic granulocytes
(acidophils) (78.93 [±6.29]%) compared to levels in controls
(60.28 [±8.63]%); however, the AgNP100 at this timepoint
imparted no significant effect. Conversely, exposure to either size
AgNP led to a significant decrease in basophilic granulocyte
(basophils) levels in the same timeframes (i.e. 10.76 [±2.78]% for
AgNP50 and 13.43 [±0.90]% for AgNP100) vs. control (19.77
[±2.89]%).

No significant variations were noted in levels of hyalinocytes
(10.30 [±3.68]% AgNP50, 10.37 [±3.33]% AgNP100, 19.94
[±5.77]% control). Conversely, when exposed to AgNP50 for 6 h,
mussel levels of hyalinocytes displayed a significant increase
(16.21 [±3.69]%) versus control values (7.48 [±3.43]%). No other
significant variations were recorded for basophils (16.24
[±2.49]% AgNP50, 14.27 [± 1.97]% AgNP100, 15.32 [±1.82]% con-
trol) or acidophils (67.54 [±6.07]% AgNP50, 77.49% [±2.69]%
AgNP100, 77.19 [±4.21]% control) in the same timeframe. For the
12-h exposure, no significant variations in hemocyte sub-popula-
tions were noted with either AgNP [hyalinocytes ¼16.63 [±5.37]
% AgNP50, 18.02 [±3.52]% AgNP100, 20.33 [±1.44]% control;
basophils ¼ 24.11 [±7.03]% AgNP50, 19.62 [±2.33]% AgNP100,
17.58 [±0.96]% control; acidophils ¼59.20 [±12.30]% AgNP50,
62.35 [±2.23]% AgNP100, 62.07 [±0.52]% control) (Figure 3(A)).

Effect of uptake pathway on circulating hemocytes

Clathrin-mediated endocytosis inhibition (amantadine [AMA])

Significant increases in basophils were seen [16.02 [±1.62] % vs.
AMA at 12.00 [±0.90] %) in hosts exposed to AgNP100 for 3 h
but not to AgNP50 [15.03 [±1.99] %). No significant variations
were recorded with any 6-h exposures (hyalinocytes: 15.49
[±0.93]% AMA, 13.8 [±2.09]% AMA þ AgNP50, 18.12 [±1.10] %
AMA þ AgNP100; basophils: 15.12 [±0.95]% AMA, 15.62
[±4.09]% AMA þ AgNP50, 14.79 [±2.11]% AMA þ AgNP100;
acidophils: 69.37 [±1.88] % [AMA], 70.57 [±6.15] %
AMA þ AgNP50, 67.07 [±3.21] % AMA þ AgNP100). At 12 h,
acidophil levels were significantly increased in hosts exposed to
either AgNP [74.23 [±2.81] % AgNP50, 73.85 [±0.77] % AgNP100,
68.28 [±0.63] % AMA. Conversely, basophil levels were signifi-
cantly decreased in mussels exposed for 12 h to AgNP100 with
clathrin path blocking (14.51 [±0.15]% vs. AMA at 19.29
[±1.33]%) but not to AgNP50 (19.29 [±1.33]%). Hyalinocyte lev-
els were also significantly reduced in mussels exposed for 12 h to
AgNP50 with clathrin path blocking (8.76 [±0.12] % vs. AMA at
11.79 [±1.03] %); AgNP100 imparted no significant effect (11.63
[±0.76] %) (Figure 3(B)).

Caveolae-mediated endocytosis inhibition

Effect of exposure to AgNP in presence of DMSO (Vehicle)

Percentages of circulating hemocytes in mussels exposed to
DMSO (0.05%) alone for 3, 6 or 12 h were not significantly
changed from levels in untreated mussels (control) (Figure 4(A)).
However, in the presence of AgNP50 or AgNP100, only a signifi-
cant decrease in basophil levels was noted at the 6-h timepoint
(13.46 [±3.78]% and 12.07 [±2.65]%, respectively) as compared
to in hosts exposed only to DMSO (18.25 [±9.06]%). No other
significant changes due to either form of AgNP at all other time-
points was noted (Figure 4(B)).

118 Y. BOUALLEGUI ET AL.

Effect of exposure to AgNP in presence of nystatin
(NYS; caveolae blocker)

No significant changes in circulating hemocytes sub-populations
were evident for either size AgNP with 3 h of exposure in the

presence of NYS (hyalinocytes: 13.91 [±3.64]% AgNP50, 10.39
[±2.31]% AgNP100, 10.13 [±3.37]% NYS). In contrast, exposure
to AgNP50 for 6 h in the presence of NYS caused only a sig-
nificant decrease in acidophils [79.15 [±1.02]% vs. NYS at
84.51 [±2.14]%) and a significant increase in basophils

Figure 1. (A) TEM image of AgNP shows homogenous distribution in size (average size � 50 nm). (B) Histogram of size (diameter) distribution for AgNP <50 nm.
(C) XRD pattern of AgNP powder. (D) Size UV-Vis absorption spectra of PVP-coated AgNP dissolved in MiliQ water. Narrow peak confirms the size of the particles.

JOURNAL OF IMMUNOTOXICOLOGY 119

Figure 2. Representative light micrograph of Mytilus galloprovincialis hemocyte sub-populations. May–Gr€unwald–Giemsa (MGG) staining. AG: acidophilic granulocytes;
Endo: endoplasm (dense stained granules); Ect: ectoplasm (hyaline with thin pseudopodia); Hy: hyalinocytes; BG: basophilic granulocytes. Magnification ¼ 400�.

Figure 3. Variations in circulating hemocyte sub-populations (%) as marker of immunomodulation from AgNP. (A) Exposure to only AgNP. (B) Exposure to AgNP and
Amantadine. Data shown are percentages. Hyalinocytes (dark grey), basophilic granulocytes (light grey), acidophilic granulocytes (medium grey), Cont: untreated,
Aman: amantadine, Ag50: AgNP < 50 nm, Ag100: AgNP< 100nm for 3, 6 or 12 h. N ¼ 10/group. Value significantly different from negative control [�p < 0.05].

120 Y. BOUALLEGUI ET AL.

[11.97 [±3.64] % vs. NYS at 8.71 [±3.37]%). Significant
increases in acidophils were evident only after 12 h of exposure
to AgNP100 in the presence of NYS [73.89 [±0.56] % vs. NYS
at 63.62 [±2.08] %); in contrast, a significant decrease in baso-
phils was noted with exposures to either size AgNP in the
presence of NYS in this same timeframe [20.09 [±0.49]%
AgNP50, 15.74 [±0.89]% AgNP100, 23.23 [±1.08]% NYS]. For
hyalinocytes, a significant increase was only evident with
exposure to AgNP50 in the presence of NYS for 12 h [17.44
[±1.96] % vs. NYS at 13.13 [±1.01]%); no significant effects
were induced with AgNP100 (10.35 [±0.51] %) (Figure 4(C)).

Correlation between variations in hemocyte sub-populations

The variations in hemocyte sub-population levels under the con-
ditions tested here were seen to be intercorrelated. Mussels
exposed under differing conditions for 3 h demonstrated signifi-
cant negative correlations between changes in levels hyalinocytes
and acidophils or in levels basophils and acidophils (r ¼ �0773
and r ¼ �0.900, respectively). No significant correlation was
found between levels of hyalinocytes and basophils (r ¼ 0.466)
(Table 1). With 6-h exposures, a significant [positive] correlation
was seen between changes in levels of hyalinocytes and basophils

Figure 4. (A) Percentages (%) circulating hemocyte sub-populations of mussels exposed to DMSO (0.05%) vehicle, alone for 3, 6 or 12 h compared to untreated mus-
sels (control). Control (black), DMSO (grey). N ¼ 10/group. Data shown are mean percentages± SD. (B) Variations in circulating hemocyte sub-populations (%) due to
AgNP or (C) Nystatin for 3, 6 or 12 h. N ¼ 10/group. Hyalinocytes (dark grey), basophilic granulocytes (light grey), acidophilic granulocytes (medium grey), Cont:
untreated, Nyst: Nystatin, Ag50: AgNP <50 nm, Ag100: AgNP <100nm. Value significantly different from negative control at �p < 0.05, ��p < 0.01.

JOURNAL OF IMMUNOTOXICOLOGY 121

(r ¼ 0.703). In contrast, significant [negative] correlations were
evident for variations in levels of hyalinocytes and acidophils and
basophils and acidophils (r ¼ �0.951 and r ¼ �0.888, respect-
ively) (Table 2). The 12-h exposure gave rise to significant nega-
tive correlations among the variations in levels of hyalinocytes
and acidophils and of basophils and acidophils (r ¼ �0.824 and
r ¼ �0.757, respectively). No significant correlations between
changes in the levels of hyalinocytes and of basophils was noted
(r ¼ 0.255) (Table 3).

Discussion

The present in vivo study aimed to elucidate the ability of AgNP
to enter into Mytilus galloprovincialis marine mussels and modu-
late the percentages of their immune system cell sub-populations.
Previous studies noted the ability of environmental pollutants,
such as mercury and cadmium, to significantly enhanced varia-
tions in hemocyte counts in mussels (Pipe & Coles 1995). In the
same context, changes in immune functions of organisms often
correspond with a presence of environmental stressors (i.e. chem-
icals or toxins) and thus can be used as good indices of local
environmental health status (Parisi et al. 2008; Ottaviani &
Malagoli 2009; Canesi & Corsi 2016; Matozzo 2016; Matozzo &
Gagn�e 2016). The ability of various NP to be taken up by hemo-
cytes and affect immune functions (i.e. lysosomal function,
phagocytic activity, oxyradicals (ROS) production and induce
pro-apoptotic processes) have been investigated in invertebrate
models, as with most invertebrates, mussels possess only innate
immune mechanisms – including phagocytosis, production of
reactive oxygen species (ROS) and nitrogen radicals, etc. – as
means of host protection (Canesi & Prochazova 2013). Canesi
et al. (2008) reported that mussel hemocytes exposed in vitro
from 0.5–4 h to carbon black NP (1–10 lg/mL) displayed
increases in release of lysosomal hydrolytic enzymes, oxidative
burst and NO. In contrast, with C60 fullerene, TiO2 and SiO2,
there were no significant cytotoxic effects in mussel hemocytes
even though each NP-stimulated immune/inflammatory parame-
ters in the exposed hosts (Canesi et al. 2010a,b). Based on all

these studies, Canesi et al. asserted that effects from NP were less
like dependent on the chemical nature of the materials but mor-
eso on associated redox properties that could cause oxidative
stress.

With regard to AgNP, several studies have reported cytotoxic
effects were closely related to increase in production of ROS.
Katsumiti et al. (2015) demonstrated ROS production in mussel
hemocytes reached a peak early (3 h) when exposed to malatose-
stabilized AgNP. Such results could help explain outcomes in the
present study whereby a 3-h exposure to AgNP50 led to signifi-
cant increases in levels of acidophil percentages in mussels, while
no variations were recorded after 6 or 12 h. This short “toxicity
timeframe” may indicate any putative cytotoxic effect caused by
AgNP could potentially be neutralized by the increased presence
of acidophils; this is plausible in that other studies have described
a prominent role for acidophils in host internal defense (Chang
et al. 2005; Garcia-Garcia et al. 2008; Parisi et al. 2008; Matozzo
& Bailo 2015).

Apart from any increased presence of “NP-detoxifying acid-
ophils,” the current results showing that the effect of the AgNP
was duration of exposure–related effect could also be a result of
changes in the bioavailability of these NP over time. As bioavail-
ability of NP is a major factor in ultimate toxicity, surrounding
environment effects on particle size stability, shape, surface
charge, etc. are key variables that will determine effects on
exposed hosts, including mollusks (Levard et al. 2012; Liu et al.
2012; Dobias & Bernier-Latmani 2013; Yu et al. 2014; Katsumiti
et al. 2015; Minetto et al. 2016). Canesi and Corsi (2016)
hypothesize putative trans-formations of NP including how
extracellular proteins could be adsorbed onto a NP surface, form-
ing a protein corona of naturally occurring colloids, particles and
macromolecules in the water column. The protein corona could
then impact how specific cellular receptors, cellular internaliza-
tion pathways, and ultimately in immune responses as well, see
and respond to the now-modified NP.

The results also indicated significant decreases in basophil lev-
els with host exposures for 3 h to either size AgNP (but no sig-
nificant variations with 6- and 12-h exposures) and a significant
increase in hyalinocytes levels only with AgNP50 for 6 h. Here,
the variations showed again that AgNP effects were duration-of-
exposure-dependent. In this same context, the recorded varia-
tions in the different sub-populations could be explained by an
ability of other cell categories, apart from acidophils, to be acti-
vated as part of the immune response. This result was in agree-
ment with outcomes of studies conducted with bacteria in
mussels by Parisi et al. (2008) showed that dramatically varied
proportions of the three cell categories clearly reflected how hya-
linocytes participated in antibacterial responses despite being
reported as “less active” than granulocytes. It was thus concluded
that more than one cell type had been involved in immune
defense. Such activation of different cell types as immune effec-
tors corroborates the hypothesis of Ottaviani et al. (1998) that
suggested that, in bivalve hemolymph (M. galloprovincialis), there
is only one hemocyte type – with two or more different matur-
ation (aging)-related stages, that is, hyalinocytes in a proliferative
stage mature to become granulocytes (Ottaviani et al. 1998).

In the present study, the reasonable choice to have used
AgNP with sizes of <50 and <100 nm was based on the litera-
ture on potential uptake pathways for each size particle. Typical
clathrin-coated pits (vessels for clathrin-mediated endocytosis)
have diameters in the range 120 nm; conversely, internalization
via caveolae-mediated endocytosis is considered the predominant
mechanism of entry for structures of 40–50 nm (and below)

Table 1. Correlations of percentage variations in hemocyte sub-populations
from mussels exposed for 3 h.

Hyalinocytes Basophils Acidophils

Hyalinocytes 1.0000 – –
Basophils 0.4661 1.0000 –
Acidophils �0.7738�� �0.9008�� 1.0000
��Value significantly correlated at p < 0.01.

Table 2. Correlations of percentage variations in hemocyte sub-populations
from mussels exposed for 6 h.

Hyalinocytes Basophils Acidophils

Hyalinocytes 1.0000 – –
Basophils 0.7034� 1.0000 –
Acidophils �0.9511�� �0.8886�� 1.0000
Value significantly correlated at �p < 0.05 or ��p < 0.01.

Table 3. Correlations of percentage variations in hemocyte sub-populations
from mussels exposed for 12 h.

Hyalinocytes Basophils Acidophils

Hyalinocytes 1.0000 – –
Basophils 0.2550 1.0000 –
Acidophils �0.8243�� �0.7577�� 1.0000
��Value significantly correlated at p < 0.01.

122 Y. BOUALLEGUI ET AL.

in diameter. Thus, while effects on clathrin-mediated endocytosis
would reflect how the cells interacted with both size AgNP here,
any impact of exposure on caveolae-mediated endocytosis would
then be more directly impactful upon the AgNP <50 nm only
(Moore 2006; Doherty & McMahon 2009; Khan et al. 2015). This
is an important distinction in that these studies did not segregate
out the relatively few particles <50 nm from the AgNP100 parent
sample so as to provide hypothetical data for AgNP50 versus
AgNP51–100. While such analyses would be interesting and
informative, the reality is that there is no way in the real world
to face such segregated selections from a parent mixture of par-
ticles (any type) even if the original cutoff value was set at
100 nm. Further, as the AgNP100 samples only contained �1.4%
particles <50 nm, their relative contribution to the observed out-
comes for the AgNP100 would be expected to be nominal.

Apparently in keeping with this assumption, an AgNP size-
dependent effect variation in the percentages of cell categories
was in fact observed here. Other studies also reported size-
dependent toxicity of AgNP, that is, with maltose-stabilized
AgNP (Katsumiti et al. 2015). In that study, small NP (Ag20-Mal)
were significantly more toxic than larger NP (Ag40-Mal and
Ag100-Mal). Such outcomes were expected based on a concept
proposed by Hine (1999) that posited differences in phagocytosis
between granulocytes and hyalinocytes were related to character-
istics of the involved particles (i.e. differences in size properties
here) rather than differences in immune cell ability to phagocyt-
ize/process the particles.

The present study also sought to clarify the role of varying
uptake mechanisms for NP (here AgNP) in influencing effects on
the frequency of immune cell types. The variations in the percen-
tages of different sub-populations seen here showed that when
clathrin- or caveolae-mediated endocytosis was inhibited, effects
caused by either size AgNP were delayed. Such results might be
due to a potential ability of either uptake route to initially
“mitigate” toxic effects of AgNP as each pathway enables any
early-internalized particles to be broken-down/digested. While
this might reduce initial levels of intracellular AgNP, it con-
versely increases the presence of the AgNP externally (such as in
an actual water environment) to putatively serve as continuous
source of Ag ions due to particle oxidation (involving dissolved
O2 and protons in aqueous system) (Dobias & Bernier-Latmani
2013; Gliga et al. 2014; Yu et al. 2014). Over time, the now
increasingly present Agþ ions could then impart their own forms
of cytotoxicity as was demonstrated in studies by Park et al.
(2013) and Katsumiti et al. (2015).

Conclusions

Overall, the results here showed how silver nanoparticles (AgNP)
may influence the frequency of different hemocyte sub-popula-
tions as biomarker of the immunomodulation of mussel hemo-
cytes by NP. It was clearly noted that nanotoxicity of AgNP was
size and indirectly duration of exposure dependent. The internal-
ization mechanism of NP most likely considered as major factor
underlying NP effects in hemocytes of M. galloprovincialis.
Lastly, it is highly recommended further research be undertaken
to clarify how specific uptake routes could be involved in deter-
mining NP toxicity.

Acknowledgements

This study is funded by the immunomicrobiology, environmental
and cancerogesis IMEC Research Unit, Sciences Faculty of Bizerte,

University of Carthage, Tunisia. The authors acknowledge Prof.
David Sheehan at the Proteomic Research Group in the School of
Biochemistry and Cell Biology at University College Cork (Ireland),
for reviewing this paper.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This study is funded by the immunomicrobiology, environmental
and cancerogesis IMEC Research Unit, Sciences Faculty of Bizerte,
University of Carthage, Tunisia.

References

Angel B, Batley G, Jarolimek C, Rogers N. 2013. The impact of size on the
fate and toxicity of nanoparticulate silver in aquatic systems.
Chemosphere. 93:359–365.

Canesi L, Corsi I. 2016. Effects of nanomaterials on marine invertebrates. Sci
Total Environ. 565:933–940.

Canesi L, Proch�azov�a P. 2013. Invertebrate immune system as a model for
investigating the environmental impact of nanoparticles. In: Boraschi D,
Duschl A, editors. Nanoparticles and the Immune System. Oxford:
Academic Press; p. 91–112.

Canesi L, Ciacci C, Betti M, Fabbri R, Canonico B, Fantinati A, Marcomini
A, Pojana G. 2008. Immunotoxicity of carbon black nanoparticles to blue
mussel hemocytes. Environ Int. 34:1114–1119.

Canesi L, Ciacci C, Betti M, Vallotto D, Gallo G, Marcomini A, Pojana G.
2010a. In vitro effects of suspensions of selected nanoparticles (C60 fuller-
ene, TiO2, SiO2) on Mytilus hemocytes. Aquat. Toxicol. 96:151–158.

Canesi L, Ciacci C, Fabbri R, Marcomini A, Pojana G, Gallo G. 2012. Bivalve
mollusks as a unique target group for nanoparticle toxicity. Marine
Environ Res. 76:16–21.

Canesi L, Fabbri R, Gallo G, Vallotto D, Marcomini A, Pojana G. 2010b.
Biomarkers in Mytilus galloprovincialis exposed to suspensions of selected
nanoparticles (Nano-carbon black, C60 fullerene, Nano-TiO2, Nano-SiO2).
Aquat Toxicol. 100:168–177.

Carballal M, Lopez M, Azevedo C, Villalba A. 1997. Hemolymph cell types of
the mussel Mytilus galloprovincialis. Dis Aquat Org. 29:127–135.

Chang S, Tseng S, Chou H. 2005. Morphological characterization via light
and electron microscopy of the hemocytes of two cultured bivalves: A
comparison study between hard clam (Meretrix lusoria) and Pacific oyster
(Crassostrea gigas). Zool. Studies. 44:144–153.

Cima F. 2010. Microscopy methods for morpho-functional characterization of
marine invertebrate hemocytes. In: M�endez-Vilas A, & D�ıaz J, editors.
Microscopy: Science, technology, applications and education. Mexico City:
Formatex; p. 1100–1107.

Cozzari M, Elia A, Pacini N, Smith B, Boyle D, Rainbow P, Khan F. 2015.
Bioaccumulation and oxidative stress responses measured in the estuarine
ragworm (Nereis diversicolor) exposed to dissolved, nano- and bulk-sized
silver. Environ Pollut. 198:32–40.

Dobias J, Bernier-Latmani R. 2013. Silver release from silver nanoparticles in
natural waters. Environ Sci Technol. 47:4140–4146.

Doherty G, McMahon H. 2009. Mechanisms of endocytosis. Annu Rev
Biochem. 78:857–902.

dos Santos T, Varela J, Lynch I, Salvati A, Dawson K. 2011. Effects of trans-
port inhibitors on the cellular uptake of carboxylated polystyrene nanopar-
ticles in different cell lines. PLoS One. 6:e24438.

Garcia-Garcia E, Prado-Alvarez M, Novoa B, Figueras A, Rosales C. 2008.
Immune responses of mussel hemocyte sub-populations are differentially
regulated by enzymes of the PI3-K, PKC, and ERK kinase families.
Develop. Comp. Immunol 32:637–653.

Giron-Perez M. 2010. Relationships between innate immunity in bivalve mol-
lusks and environmental pollution. Invert Survival J. 7:149–156.

Gliga A, Skoglund S, Wallinder I, Fadeel B, Karlsson H. 2014. Size-dependent
cytotoxicity of silver nanoparticles in human lung cells: The role of cellular
uptake, agglomeration and Ag release. Part Fibre Toxicol 11:27.

Gosling E, editor. 2003. Bivalves mollusks: Biology, ecology and culture.
Oxford: Blackwell Publishing.

Haucke V. 2006. Cargo takes control of endocytosis. Cell. 127:35–37.

JOURNAL OF IMMUNOTOXICOLOGY 123

Hine P. 1999. The inter-relationships of bivalve haemocytes. Fish Shellfish
Immunol. 9:367–385.

H€oher N, Regoli F, Dissanayake A, Nagel M, Kriews M, K€ohler A, Broeg K.
2013. Immunomodulating effects of environmentally-realistic copper con-
centrations in Mytilus edulis adapted to naturally low salinities. Aquat
Toxicol. 140:185–195.

Ivanov A, editor. 2008. Pharmacological inhibition of endocytic pathways: is
it specific enough to be useful? In: Exocytosis and endocytosis. New York:
Humana Press; p. 15–33.

Katsumiti A, Gilliland D, Arostegui I, Cajaraville M. 2015. Mechanisms of
toxicity of Ag nanoparticles in comparison to bulk and ionic Ag on mussel
hemocytes and gill cells. PLoS One. 10:e129039.

Khan F, Mirsa S, Bury N, Smith Brian D, Rainbow P, Luoma S, Valsami-
Jones E. 2015. Inhibition of potential uptake pathways for silver nanopar-
ticles in the estuarine snail Peringia ulvae. Nanotoxicology. 9:493–501.

Leopold N, Lendl B. 2003. A new method for fast preparation of highly sur-
face-enhanced Raman scattering (SERS) active silver colloids at room tem-
perature by reduction of silver nitrate with hydroxylamine hydrochloride.
J Phys Chem. 107:5723–5727.

Levard C, Hotze E, Lowry G, Brown G. 2012. Environmental transformations
of silver nanoparticles: Impact on stability and toxicity. Environ Sci
Technol. 46:6900–6914.

Liu J, Wang Z, Liu F, Kane A, Hurt R. 2012. Chemical transformations of
nanosilver in biological environments. ACS Nano. 6:9887–9899.

Luoma S. 2008. Silver nanotechnologies and the environment. Washington
(DC): Woodrow Wilson International Center for Scholars and The Pew
Charitable Trusts.

Marisa I, Matozzo V, Munari M, Binelli A, Parolini M, Martucci A,
Franceschinis E, Brianese N, Marin M. 2016. In vivo exposure of the mar-
ine clam Ruditapes philippinarum to zinc oxide nanoparticles: Responses
in gills, digestive gland and haemolymph. Environ Sci Pollut Res Int.
23:15275–15293.

Matranga V, Corsi I. 2012. Toxic effects of engineered nanoparticles in the
marine environment: Model organisms and molecular approaches. Mar
Environ Res. 76:32–40.

Matozzo V, Bailo L. 2015. A first insight into haemocytes of the smooth
venus clam Callista chione. Fish Shellfish Immunol. 42:494–502.

Matozzo V, Gagn�e F. 2016. Immunotoxicology approaches in ecotoxicology:
lessons from mollusks. In: Ballarin l, Cammarata M, editors. Lessons in
immunity: From single-cell organisms to mammals. London: Academic
Press; p. 29–51.

Matozzo V, Rova G, Marin M. 2007. Hamocytes of the cockle Cerastoderma
glaucum: Morphological characterization and involvement in immune
responses. Fish Shellfish Immunol. 23:732–746.

Matozzo V. 2016. Aspects of eco-immunology in mollusks. Invert Survival J.
13:116–121.

Mezni A, Mlayah A, Serin V, Smiri L. 2014b. Synthesis of hybrid Au-ZnO
nanoparticles using a one pot polyol. Mater Chem Phys. 147:496–503.

Mezni A, Dammak T, Fkiri T, Mlayah A, Abid A, Smiri L. 2014a.
Photochemistry at surface of gold nanoprisms from SERS blinking. J Phys
Chem C. 118:17956–17967.

Minetto D, Ghirardini V, Libralato G. 2016. Saltwater ecotoxicology of Ag,
Au, CuO, TiO2, ZnO, and C60 engineered nanoparticles: An overview.
Environ. Intl. 92:189–201.

Moore M. 2006. Do nanoparticles present ecotoxicologic risks for the health
of the aquatic environment? Environ Intl. 32:967–976.

Nichols B, Lippincott-Shwartz J. 2001. Endocytosis without clathrin coats.
Trends Cell Biol. 11:406–412.

Ottaviani E, Malagoli D. 2009. Around the word stress: Its biological and
evolutive implications. Invert. Survival J. 6:1–6.

Ottaviani E, Franchini A, Barbieri D, Kletsas D. 1998. Comparative and mor-
pho-functional studies on Mytilus galloprovincialis hemocytes: Presence of
two aging-related hemocyte stages. Ital J Zool. 65:349–354.

Parisi M, Li H, Jouvet Lionel B, Dyrynda E, Parrinello N, Cammarata M,
Roch P. 2008. Differential involvement of mussel hemocyte sub-popula-
tions in the clearance of bacteria. Fish Shellfish Immunol. 25:834–840.

Park K, Tuttle G, Sinche F, Harper S. 2013. Stability of citrate-capped silver
nanoparticles in exposure media and their effects on the development of
embryonic zebrafish (Danio rerio). Arch Pharm Res. 36:125–133.

Piccinno F, Gottschalk F, Seeger S, Nowack B. 2012. Industrial production
quantities and uses of 10 engineered nanomaterials for Europe and world.
J. Nanopart. Res. 14:1109–1120.

Pinsino A, Russo R, Bonaventura R, Brunelli A, Marcomini A, Matranga V.
2015. Titanium dioxide nanoparticles stimulate sea urchin immune cell
phagocytic activity involving TLR/p38 MAPK-mediated signaling pathway.
Sci Rep. 5:14492.

Pipe R, Coles J. 1995. Environmental contaminants influencing immune func-
tion in marine bivalve mollusks. Fish Shellfish Immunol. 5:581–595.

Pipe R, Farley S, Coles J. 1997. The separation and characterisation of haemo-
cytes from the mussel Mytilus edulis. Cell Tissue Res. 289:537–545.

Puthenveedu M, von Zastrow M. 2006. Cargo regulates clathrin-coated pit
dynamics. Cell. 127:113–124.

Rainville L, Carolan D, Varela A, Doyle H, Sheehan D. 2014. Proteomic
evaluation of citrate-coated silver nanoparticles toxicity in Daphnia magna.
Analyst. 139:1678.

Razani B, Lisanti M. 2002. Role of caveolae and the caveolins in mammalian
physiology. Rev. Undergrad Res. 1:44–50.

Sandvig K, Pust S, Skotland T, van Deurs B. 2011. Clathrin-independent
endocytosis: Mechanisms and function. Curr Opin Cell Biol. 23:413–420.

Tiede K, Hassell€ov M, Breitbarth E, Chaudhry Q, Boxall A. 2009.
Considerations for environmental fate and ecotoxicity testing to support
environmental risk assessments for engineered nanoparticles.
J Chromatogr A. 1216:503–509.

Yu S, Yin Y, Chao J, Shen M, Liu J. 2014. Highly dynamic PVP-coated silver
nanoparticles in aquatic environments: chemical and morphology change
induced by oxidation of Ag0 and reduction of Agþ. Environ Sci Technol.
48:403–411.

124 Y. BOUALLEGUI ET AL.

Copyright of Journal of Immunotoxicology is the property of Taylor & Francis Ltd and its
content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder’s express written permission. However, users may print, download, or email
articles for individual use.

Research Article
Evaluation of the Response of HOS and Saos-2 Osteosarcoma Cell
Lines When Exposed to Different Sizes and Concentrations of
Silver Nanoparticles

Konstantinos Michalakis ,1,2,3 Athina Bakopoulou ,1 Eleni Papachristou,1

Dimitra Vasilaki ,1 Alexandros Tsouknidas ,4 Nikolaos Michailidis ,5

and Elaine Johnstone6

1School of Health Sciences, Aristotle University of Thessaloniki, Thessaloniki, Greece
2Tufts University, Boston, MA, USA
3University of Oxford, Oxford, UK
4Laboratory for Biomaterials and Computational Mechanics, Department of Mechanical Engineering, University of
Western Macedonia, Kozani, Greece
5Department of Mechanical Engineering, School of Engineering, Aristotle University of Thessaloniki, Thessaloniki, Greece
6Department of Oncology, University of Oxford, Oxford, UK

Correspondence should be addressed to Konstantinos Michalakis; [email protected]

Received 12 September 2021; Revised 20 November 2021; Accepted 22 November 2021; Published 13 December 2021

Academic Editor: Aziz ur Rehman Aziz

Copyright © 2021 Konstantinos Michalakis et al. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work
is properly cited.

Osteosarcoma is considered to be a highly malignant tumor affecting primarily long bones. It metastasizes widely, primarily to the
lungs, resulting in poor survival rates of between 19 and 30%. Standard treatment consists of surgical removal of the affected site,
with neoadjuvant and adjuvant chemotherapy commonly used, with the usual side effects and complications. There is a need for
new treatments in this area, and silver nanoparticles (AgNPs) are one potential avenue for exploration. AgNPs have been found to
possess antitumor and cytotoxic activity in vitro, by demonstrating decreased viability of cancer cells through cell cycle arrest and
subsequent apoptosis. Integral to these pathways is tumor protein p53, a tumor suppressor which plays a critical role in
maintaining genome stability by regulating cell division, after DNA damage. The purpose of this study was to determine if p53
mediates any difference in the response of the osteosarcoma cells in vitro when different sizes and concentrations of AgNPs are
administered. Two cell lines were studied: p53-expressing HOS cells and p53-deficient Saos-2 cells. The results of this study
suggest that the presence of protein p53 significantly affects the efficacy of AgNPs on osteosarcoma cells.

1. Introduction

Osteosarcoma is considered a relatively uncommon malig-
nant disease. Nevertheless, it is the most common cancer
arising from bone [1]. It usually affects adolescents and
young adults. In recent years, much advancement has been
made in treating osteosarcoma, which combines surgery,
chemotherapy, and sometimes radiotherapy. Currently, the
5-year survival rate for patients diagnosed with osteosar-

coma is 60-70% [2–5]. The chemotherapy agents employed
include cisplatin, doxorubicin, ifosfamide, and methotrexate.
Other cytotoxic agents such as etoposide and different com-
binations have also been suggested in the literature [6]. Nev-
ertheless, the use of these drugs has several side effects and
complications including neutropenia, mouth ulcers, fatigue,
severe diarrhea, nausea, and vomiting. The side effects can
be very serious and commonly require hospitalization. Car-
diomyopathies and irreversible lung fibrosis have also been

Hindawi
BioMed Research International
Volume 2021, Article ID 5013065, 16 pages
https://doi.org/10.1155/2021/5013065

described, illustrating that severe side effects present a major
drawback for the use of chemotherapeutic agents [7]. This
along with therapeutic limitations, due to the systemic cyto-
toxic effects, has motivated scientists to start exploring dif-
ferent directions in an attempt to find innovative therapies
for several types of cancer, including osteosarcoma [8–15].
Some novel therapeutic agents have been tested for that pur-
pose, including tumor microenvironment inhibitors, which
target signal-transduction pathways and immunomodula-
tory agents. Methods for overcoming resistance mechanisms
as well as new delivery mechanisms have also been tested
[16]. One of these avenues of interest is silver nanoparticles.
Although the exact action by which AgNPs act on cells is not
fully understood, it is speculated that a Trojan horse mecha-
nism is involved [17]. Upon entering the cell, the AgNPs
release silver ions in the cytoplasm which then induce the
formation of ROS, thus causing an imbalance of the cell’s
redox homeostasis [18, 19]. It is not known yet whether
the observed oxidative damage is due to the action of AgNPs
per se, accumulation of silver ions in the cytoplasm, or a
combination of both [20, 21] (Figure 1). A recent in vitro
study testing the antibacterial effect of AgNPs with different
sizes has shown that smallest-sized AgNPs are more effica-
cious on two different types of Gram-negative bacteria
[22]. According to Gliga et al., smaller AgNPs are more
active due to the increased Ag ion release from the increased
total surface area [23] (Figure 2).

Tumor protein p53, whose gene TP53 is located on the
short arm of chromosome 17, plays a critical role in regulating
cell division, after DNA damage occurs. It is crucial in deter-
mining if the DNA damage can be repaired or if the cell will
undergo apoptosis [24, 25]. When DNA damage in the form
of a double-strand break occurs, there is recruitment of
ATM serine protein kinases and/or ATR kinases, which are
then activated. These kinases phosphorylate p53, leading the
protein to evade degradation by ubiquitin. As a result, the
levels of p53 increase markedly; the protein is stabilized and
activates transcription of p21(Cip1/Waf1) [26]. The latter acts
by binding and inhibiting the activity of several complexes,
including cyclin E-CDK2, cyclin E-CDK1, and cyclin E-
CDK4/6, and prevents cell cycle progression at phase G1 [27,
28]. This arrest gives time to the cell to repair the damage of
the DNA. Furthermore, p53 is responsible for the production
of DNA repair enzymes and proapoptotic proteins [29].

In this way, p53 acts as a tumor suppressor, and its inac-
tivation seems to play a key role in the development of
human cancer. For the pivotal role in maintaining genome
integrity, p53 has been named “guardian of the genome”
[30]. If DNA is damaged and p53 is present and functional,
the cell cycle arrests in phase G1. On the contrary, in the
absence of functional p53, cells continue to grow and divide.
The p53 protein is unique in the sense that it exists in very
small quantities in normal cells, due to its instability and
rapid degradation. Mouse models have shown that the
absence of p53 is associated with the development of several
types of tumors [31]. Furthermore, p53 is mutated in more
than half of all human cancers, and in more than 80% of
tumors, there is a p53 signaling pathway disruption of some
kind [32–34].

Several human osteosarcoma cell lines have been isolated
so far, including the HOS, U-2OS, MG-63, G-292, and Saos-
2. An analysis of these cell lines with p53 genomic probes
has revealed some key differences. p53 was found to be pres-
ent in G-292, MG-63, HOS, and U-2OS cell lines, with a
rearrangement in the first intron of the gene described in
G-292 and MG-63. A point mutation within the p53 coding
sequence has been described in HOS cells which results in
overproduction of mutant p53 [35, 36].

There has been speculation that the AgNP-induced
mechanism of cytotoxicity may be affected by the presence
of functional p53 [37], although the evidence is rather lim-
ited. Therefore, the possible differences in the effect that
AgNPs have on the viability of different human osteosar-
coma cell lines, in which p53 is expressed or not, should
be further investigated. The aim of this study was to deter-
mine if there are any differences in the response of two
cell lines: p53-expressing HOS cells and p53-deficient
Saos-2 cells, after different sizes and concentrations of
AgNPs are administered. The null hypotheses were the
following:

(a) The size of AgNPs would not affect the response of
p53-expressing HOS cells and p53-deficient Saos-2
osteosarcoma cell lines

(b) The AgNP content of the colloid would not affect the
response of p53-expressing HOS cells and p53-
deficient Saos-2 osteosarcoma cell lines

(c) The presence or absence of p53 would not affect the
response of osteosarcoma cells to AgNP treatment

2. Methods and Materials

2.1. Osteosarcoma Cells. The HOS (p53-expressing)
(Figure 3) and Saos-2 (p53-deficient) (Figure 4) cells needed
for this research project were obtained from the American
Type Culture Collection (ATCC No. HTB 85).

2.2. Silver Nanoparticle Preparation. Two commercially
available colloidal suspensions (PLiN Nanotechnology)
with monodispersed populations of spherical AgNPs, i.e.,
7 nm and 60nm in size, respectively, were synthesized as
summarized below. Silver nitrate (99.9% AgNO3, Mr =
169:873 g/mol) was used as a silver precursor (Duchefa
Biochemie) for the reduction into AgNPs, with compo-
nents conventionally found in literature, while a protein
with a molecular mass of 20-25 kg/mol (Sigma-Aldrich)
was employed as the stabilizer. AgNPs were produced via
liquid chemistry, by adding the reduction agent to the pre-
heated aqueous solution of the silver nitrate, stirred along
with the stabilizer, to ensure complete dissolution. The
characteristics of the AgNPs are presented in Table 1
and Figure 5.

p53-expressing HOS and p53-deficient Saos-2 osteosar-
coma cells were treated with colloid silver (PLiN Nanotech-
nology, Thessaloniki, Greece) of 7 nm and 60 nm positively
charged AgNPs, as determined by dynamic light scattering.
In addition to control (c), which contained no AgNPs, six

2 BioMed Research International

Inhibition of e-
transport chain

e-
Peptidoglycan damage Cell membrane breakage

Silver nanoparticles

e-

DNA
damage

Oxidative stress
Mitochondrial dysfunction

Enzyme inactivation

Protein
denaturation

Ribosome
disassembly

ROS

Figure 1: Mechanisms of action of AgNPs on cells.

15 nm
50 nm

100 nm

Si
lv

er
n

an
op

ar
tic

le
s

Lysosome

Lysosome

Lysosome

Mitochondrion Mitochondrion Mitochondrion

Nucleus

Nucleus

Nucleus

Figure 2: Efficacy of AgNPs according to their size.

HOS

Figure 3: HOS (p53-expressing) osteosarcoma cells used for the
purposes of this study.

SAOS-2

Figure 4: Saos-2 (p53-deficient) osteosarcoma cells used for the
purposes of this study.

3BioMed Research International

different concentrations (c1-c6) were tested for three time
periods, i.e., 24, 48, and 72 hours. These concentrations were
as follows: c1 = 10ppm, c2 = 5 ppm, c3 = 2:5ppm, c4 = 1:25
ppm, c5 = 0:625 ppm, and c6 = 0:3125 ppm [37].

2.3. HOS and Saos-2 Cell Culture. HOS and Saos-2 cells were
expanded in cell culture media (CCM) in 75cm2 flasks. Cell
cultures were maintained in an incubator at 37°C, in 5%
CO2 and 95% humidity until reaching 80-90% confluency.
Cell harvesting from the flask surface was performed using
0.25% Trypsin/1mM EDTA solution (Invitrogen). For cell

counting and determination of cell density and percentage of
dead cells before each experimental assay, an improved Neu-
bauer hemocytometer (Laboroptik, Lancing, UK) and Trypan
blue exclusion tests were used [38].

2.4. Evaluation of Cell Viability with the MTT Assay. The
viability of HOS and Saos-2 cells was investigated by the
MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide) assay. Cells were cultured in direct contact with
the specimens in 96-well plates (104 cells/well) for 24, 48,
and 72h, at 37°C and 5% CO2. After these three time points,

Table 1: Material characterization for the 7 nm and 60 nm colloid silver suspension.

(a) 7 nm

Average diameter (nm) 6.93 Solvent Deionized water

Standard deviation (%) 19.01 Viscosity (cP) 0.888

Concentration (ppm) 1500 Capping agent type Organic

Zeta-potential (mV) — pH 4.15

(b) 60 nm

Average diameter (nm) 59.97 Solvent Deionized water

Standard deviation (%) 13.44 Viscosity (cP) 0.888

Concentration (ppm) 1710 Capping agent type Organic

Zeta-potential (mV) — pH 4.30

0.16

0.14

0.12

0.06

0.04
0.02

0.10

0.08

1 10 100 1 10 100
Diameter (nm)

Avg. diameter: 6.93 nm±19.01 Avg. diameter: 59.97 nm±13.40

N
um

be
r (

a.
u.

)

10 nm 60 nm

(a)

3.0
2.5

2.0
1.5
1.0
0.5

0

A
bs

or
ba

nc
e

(a
.u

)

300 400 500 600 700 800
Wavelength (nm)

max = 420 nm, Abs = 1.719 max = 430 nm, Abs = 0.896

300 400 500 600 700 800

(b)

Figure 5: Characteristics of the colloidal silver suspensions, namely, (a) size distribution and indicative TEM image (for the 7 nm colloid)
and (b) UV-Vis spectra.

4 BioMed Research International

MTT (5mg/ml in CCM) was added to each well containing
the specimens, and the plates were incubated for 4h at 37°C
and 5% CO2. During this period, the NAD(P)H-dependent
cellular oxidoreductase enzymes of mitochondria reduce the
tetrazolium dye MTT to its insoluble formazan, which has a
purple color. After this period, the medium containing the
MTT solution was discarded and 500μl of DMSO (dimethyl
sulfoxide) was added to each well and incubated for 1h at
37°C to dissolve the insoluble purple formazan product into
a colored solution. Then, the optical density (OD) was mea-
sured against blank (DMSO), at a wavelength of 545nm and
a reference filter of 630nm by a microplate reader (Epoch,
Biotek, Biotek Instruments, Inc., Vermont, USA). The experi-
ments were repeated three times, with 6-8 replicates for each
repetition. All results were expressed as an average percentage
of the control value [38].

2.5. Evaluation of Cell Proliferation with the BrdU Assay. The
proliferation rates of HOS and Saos-2 cells seeded of each
group were investigated by the BrdU (5-bromo-2′-deoxy-
uridine) assay (Sigma-Aldrich, Roche Diagnostics,
Manheim, Germany).

Cells were cultured in 96-well plates (104 cells/well) for 24,
48, and 72h, at 37°C and 5% CO2, as described earlier. After-
wards, BrdU was added at a concentration of 10μΜ, and the
plates were incubated for 6h at 37°C and 5% CO2. Then,
treated cells were fixed with FixDenat® solution (at 15-25°C,
for 30min), according to the manufacturer’s recommenda-
tions, and exposed to a peroxidase-conjugated BrdU antibody
(anti-BrdU-POD) at a concentration of 10μΜ for 90min.
Afterwards, 200μl of 3-3′-5-5′-tetra-methyl-benzidine sub-
strate (TMB) was added to each well. The blue color
peroxidase-substrate reaction ended after 5min, by an H2SO4

solution (stop solution, 50μl/well). The incorporated BrdU
were quantified by measuring the OD in a microplate reader
(Epoch, Biotek, Biotek Instruments, Inc., Vermont, U.S.A.), at
a wavelength of 450nm and a reference filter of 690nm. Cell-
free and BrdU-free wells served as internal controls for this
assay. The resulting OD values of those wells were used as
blank (negative control) and background control (positive con-
trol), respectively. The experiments were repeated three times,
with 6-8 replicates for each repetition. All results were
expressed as an average percentage of the control value [38].

Live/dead double staining was utilized to detect viable and
dead Saos-2 and HOS cells when exposed to AgNPs. Calcein-
AM, which is a highly lipophilic and cell membrane-
permeable dye, and the nuclei-staining dye Propidium Iodine,
which cannot pass through a viable cell membrane, were uti-
lized for that purpose. A 490nm light was used for simulta-
neous monitoring of viable and dead cells with a single-
excitation fluorescence microscope (Figures 6 and 7).

2.6. Statistical Analysis. For statistical analysis, Prism 6
(GraphPad, CA, U.S.A.) software was utilized. A two-way
Analysis of Variance (ANOVA) was performed for the via-
bility assays, while for follow-up comparisons between
groups and time points, Tukey’s post hoc test was employed.
Normal distribution was confirmed by Kolmogorov-
Smirnov normality tests. The level of statistical significance
was set to 0.05 (α = 0:05).

3. Results

3.1. Evaluation of Cell Viability by the MTT Assay. HOS
(p53-expressing) and Saos-2 (p53-deficient) osteosarcoma
cell viability was assessed by the MTT assay, for two different

(a) (b)

(c)

Figure 6: Live/dead double staining detecting viable and dead HOS cells when exposed to 2.5 ppm concentration AgNPs, after 48 hours: (a)
control, (b) 7 nm, and (c) 60 nm (magnification ×100).

5BioMed Research International

AgNP sizes (7 and 60 nm) and six different concentrations
(c1 = 10 ppm, c2 = 5 ppm, c3 = 2:5ppm, c4 = 1:25 ppm, c5
= 0:625 ppm, and c6 = 0:3125 ppm), at three time points
(24 hours, 48 hours, and 72 hours).

3.2. Evaluation of HOS Cell Viability for 7 nm and
60 nm AgNPs

3.2.1. 7nm. The 10 ppm and the 5 ppm concentrations of the
7 nm AgNPs demonstrated a remarkably decreased meta-
bolic activity for all three time points which was statistically
significant (P < 0:0001).

At lower concentrations, a small increase in viability was
observed. The biggest increase in cell viability (110% (±
10.29%)) was noticed in 24 hours at the 1.25ppm concentra-
tion. The smallest increase (65.58% (±1.86%)) was also
noticed at the 1.25 ppm concentration, in 72 hours
(Table 2 and Figure 8).

3.2.2. 60 nm. Unlike the 7 nm AgNPs, in the 60 nm AgNPs,
only the 10 ppm concentration demonstrated a remarkably
decreased cell viability, for all three examined time
periods. Specifically, these values were 5% (±1.20%),
4.84% (±3.05%), and 12.88% (±9.85%) at the 24-hour,

(a) (b)

(c)

Figure 7: Live/dead double staining detecting viable and dead Saos-2 cells when exposed to 2.5 ppm concentration AgNPs, after (a) control,
(b) 7 nm, and (c) 60 nm (magnification ×100).

Table 2: Average percentage values and standard deviations for HOS cell viability, when subjected to exposure of different concentrations of
7 nm and 60 nm AgNPs.

(a) MTT

7 nm Control 10 ppm 5 ppm 2.5 ppm 1.25 ppm 0.625 ppm 0.3125 ppm

24 h 100.00 (±11.73) 4.50 (±0.50) 6.67 (±2.25) 94.39 (±12.36) 110.00 (±10.29) 88.33 (±13.45) 87.00 (±14.77)
48 h 100.00 (±5.42) 4.68 (±0.90) 8.25 (±7.03) 97.02 (±9.70) 95.04 (±23.89) 86.23 (±20.14) 80.20 (±2.10)
72 h 100.00 (±5.22) 10.11 (±2.69) 9.66 (±5.98) 72.93 (±4.93) 65.58 (±1.86) 74.29 (±7.92) 87.70 (±4.00)

(b) Average

60 nm Control 10 ppm 5 ppm 2.5 ppm 1.25 ppm 0.625 ppm 0.3125 ppm

24 h 100.00 (±13.76) 5.00 (±1.20) 85.67 (±5.57) 92.33 (±4.65) 85.11 (±5.11) 80.56 (±1.73) 82.06 (±7.22)
48 h 100.00 (±2.89) 4.84 (±3.05) 77.54 (±6.17) 74.37 (±6.99) 81.39 (±14.76) 77.06 (±18.64) 80.16 (±6.01)
72 h 100.00 (±6.35) 12.88 (±9.85) 57.90 (±2.82) 56.13 (±3.23) 57.28 (±7.64) 71.77 (±11.89) 87.25 (5.65)

6 BioMed Research International

48-hour, and 72-hour time periods, respectively. A big
increase in cell viability was noticed for all other concen-
trations, ranging between 56.13% (±3.23%) and 92.33%
(±4.65%). The smallest value was noticed at 72 hours,
while the biggest one at 24 hours, both in the 2.5 ppm
concentration (Table 2 and Figure 8).

The 2-way ANOVA (α = 0:05) for the MTT assay of HOS
cells subjected to 7nm AgNPs revealed a statistically signifi-
cant effect of the concentration factor (F = 5:618, P = 0:0069
), the time factor (F = 144:3, P < 0:0001), and their interaction
(F = 2:869, P = 0:0057), while for the 60nm AgNPs, the
ANOVA revealed a statistically significant effect of the
concentration factor (F = 12:30, P < 0:0001), the time factor
(F = 112:5, P < 0:0001), and their interaction
(F = 3:976, P = 0:0004).

3.3. Evaluation of Saos-2 Cell Viability for 7nm and 60 nm
AgNPs. An entirely different behavior of the Saos-2 osteosar-
coma cells is observed when compared to the HOS cells.

3.3.1. 7 nm. The smallest percentage in cell viability was
90.03% (±2.50%), noticed at 72 hours in the 10 ppm concen-
tration, while the biggest one was 121.35% (±7.42%). The
latter was observed at 24 hours in the 2.5 ppm concentration.
In general, very little cytotoxicity was observed at all concen-
trations and time points (Table 3 and Figure 9).

3.3.2. 60 nm. Similarly, with the 7 nm AgNPs, the 60 nm
AgNPs at 10 ppm did not produce a decreased Saos-2 cell
viability for any concentration, when compared to the con-
trol. In general, it can be observed that in the 24-hour

140

120

100

80

60

40

20

0

C
el

l v
ia

bi
lit

y
(%

o
f c

on
tr

ol
)

C
el

l v
ia

bi
lit

y
(%

o
f c

on
tr

ol
)

Control 10 ppm 5 ppm 2.5 ppm 1.25ppm 0.625 ppm 0.3125 ppm

MTT HOS_7 nm

MTT HOS_60 nm

120

100

80

60

40

20

0
Control 10 ppm 5 ppm 2.5 ppm 1.25 ppm 0.625 ppm 0.3125 ppm

⁎⁎⁎

⁎
⁎
⁎

⁎⁎⁎
⁎
⁎

⁎⁎
⁎

⁎

⁎

⁎

⁎ ⁎⁎

⁎

⁎ ⁎

⁎ ⁎

⁎

⁎

⁎

⁎

⁎

⁎
⁎

⁎

⁎⁎
⁎

⁎ ⁎⁎
⁎

⁎

⁎

⁎
⁎

⁎

⁎

⁎⁎
⁎

⁎ ⁎
⁎

⁎

⁎ ⁎

⁎ ⁎

⁎

⁎⁎
⁎

⁎

⁎
⁎

24 h

48 h

72 h

Figure 8: HOS cell viability (percentage values), when subjected to exposure of different concentrations of 7 nm and 60 nm AgNPs (same
color asterisks indicate statistically significant differences between the control and suspensions of different AgNP concentrations. Absence of
asterisks or absence of same color asterisks indicates no statistically significant differences, according to Tukey’s HSD test for α = 0:05).

7BioMed Research International

period, all concentrations did not have a negative effect on
Saos-2 cells. On the opposite, it seems that it promoted cell
viability, as it reached 112.28% (±5.57%), for the
0.625 ppm concentration (Table 3 and Figure 9).

The 2-way ANOVA (α = 0:05) for the MTT assay of Saos-
2 cells subjected to 7nm AgNPs revealed a statistically signif-
icant effect of the concentration factor (F = 18:10, P < 0:0001),
the time factor (F = 3:005, P = 0:0156), and their interaction
(F = 2:690, P = 0:0088), while for the Saos-2 cells subjected
to 60nm AgNPs, the ANOVA revealed a statistically signifi-
cant effect of the concentration factor (F = 21:17, P < 0:0001
). However, a statistically significant effect was not demon-
strated for the time factor (F = 1:899, P = 0:1036) and the
interaction of concentration and time (F = 1:476, P = 0:1720).

3.4. Evaluation of Cell Proliferation by the BrdU Assay. HOS
(p53-expressing) and Saos-2 (p53-deficient) osteosarcoma
cell proliferation was assessed by the BrdU test, for the two
different AgNP sizes (7 and 60 nm), six different concentra-
tions (c1 = 10 ppm, c2 = 5 ppm, c3 = 2:5ppm, c4 = 1:25 ppm,
c5 = 0:625 ppm, and c6 = 0:3125 ppm), and three time
periods (24 h, 48 h, and 72h).

3.5. Evaluation of HOS Cell Proliferation for 7nm and
60 nm AgNPs

3.5.1. 7nm. Like in the MTT cell viability assay, the 10 ppm
and the 5 ppm concentrations of the 7 nm AgNPs demon-
strated a remarkably decreased cell proliferation activity for
all three time periods, when the BrdU assay was performed
(Table 4 and Figure 10).

3.5.2. 60 nm. Unlike the 7 nm AgNPs, in the 60 nm AgNPs,
only the 10 ppm concentration demonstrated a remarkably
decreased cell proliferation, for all three examined time
periods. Specifically, these values were 5.04% (±5.51%),
1.65% (±2.45%), and 4.29% (±4.81%) at the 24-hour, 48-
hour, and 72-hour time periods, respectively. A big increase
in cell viability was noticed for all other concentrations
(Table 4 and Figure 10).

The 2-way ANOVA (α = 0:05) for the BrdU assay of HOS
cells subjected to 7nm AgNPs revealed a statistically significant
effect of the concentration factor (F = 6:534, P = 0:0034), the
time factor (F = 56:95, P < 0:0001), and their interaction
(F = 2:428, P = 0:0169), while for the BrdU assay of HOS cells
subjected to 60nm AgNPs, the ANOVA revealed a statistically
significant effect of the concentration factor
(F = 13:14, P < 0:0001), the time factor (F = 72:19, P < 0:0001
), and their interaction (F = 4:573, P = 0:0001).

3.6. Evaluation of Saos-2 Cell Proliferation for 7nm AgNPs.
An entirely different behavior of the Saos-2 osteosarcoma
cells is observed when compared to the HOS cells. While cell
proliferation values for the HOS cell ranged between -0.69%
and 9.20%, for the examined time periods of the 10ppm and
5 ppm concentrations, the corresponding values for the
Saos-2 cells were 83.93% (±9.43%) and 97.73% (±12.41%)
(Table 5 and Figure 11).

The 2-way ANOVA (α = 0:05) for the BrdU assay of
Saos-2 cells subjected to 7 nm AgNPs did not reveal a sta-
tistically significant effect of the concentration factor
(F = 2:762, P = 0:0746). However, a statistically significant
effect was demonstrated for the time factor
(F = 3:936, P = 0:0033) and the interaction of concentra-
tion and time (F = 3:453, P = 0:0014). For the BrdU assay
of Saos-2 cells subjected to 60 nm AgNPs, the 2-way
ANOVA (α = 0:05) revealed a statistically significant effect
of the concentration factor (F = 10:23, P = 0:0002). How-
ever, a statistically significant effect was not demonstrated
neither for the time factor (F = 1:481, P = 0:2081) nor for
the interaction of concentration and time
(F = 1:219, P = 0:3025).

4. Discussion

The objective of this study was to investigate if AgNPs of dif-
ferent sizes and concentrations could have a potential appli-
cation in osteosarcoma treatment and if cytotoxic efficacy
was affected by the presence or absence of p53. Although
there is published evidence that AgNPs can be used

Table 3: Average percentage values and standard deviations for Saos-2 cell viability, when subjected to exposure of different concentrations
of 7 nm and 60 nm AgNPs.

(a) MTT

7 nm Control 10 ppm 5 ppm 2.5 ppm 1.25 ppm 0.625 ppm 0.3125 ppm

24 h 100.00 (±10.20) 107.88 (±13.32) 119.61 (±6.57) 121.35 (±7.42) 115.80 (±4.85) 114.31 (±10.05) 106.09 (±9.01)
48 h 100.00 (±12.30) 114.44 (±3.21) 104.02 (±8.26) 105.68 (±7.54) 110.19 (±10.49) 100.61 (±1.07) 91.39 (±4.77)
72 h 100.00 (±3.45) 90.03 (±2.50) 95.71 (±1.36) 101.4 (±3.32) 102.43 (±2.05) 104.02 (±1.34) 101.03 (±1.94)

(b) Average

60 nm Control 10 ppm 5 ppm 2.5 ppm 1.25 ppm 0.625 ppm 0.3125 ppm

24 h 100.00 (±12.56) 102.38 (±7.59) 109.51 (±8.74) 110.45 (±12.50) 109.96 (±11.33) 112.28 (±5.57) 106.19 (±15.02)
48 h 100.00 (±2.46) 91.83 (±16.49) 88.27 (±5.02) 84.27 (±10.55) 100.43 (±5.15) 91.86 (±4.47) 87.29 (±7.33)
72 h 100.00 (±1.23) 81.44 (±3.07) 87.02 (±1.13) 95.80 (±3.93) 98.24 (±4.84) 98.89 (±3.18) 97.57 (±3.62)

8 BioMed Research International

effectively against certain types of osteosarcoma cells [37],
there is no study to the authors’ knowledge, determining
the efficacy of AgNPs against HOS osteosarcoma cells, which
express p53 protein.

Two methods, targeting different biological endpoints,
were selected to evaluate the impact of AgNPs on the p53-
expressing HOS and p53-deficient Saos cell lines. The
MTT assay is a typical method to assess cell viability through
the evaluation of the active metabolic activity of living cells,
whereas the BrdU assay is used for evaluating cell proliferation
through DNA intercalation. The combination of the two
methods can answer the question of whether a reduction in
the metabolic activity observed through the MTT assay is pri-
marily caused by cell death or by cell cycle delays leading to
reduced cell proliferation. The latter is a common mechanism

of action of several antineoplastic drugs that primarily act by
causing cell cycle arrest in different phases of the cell cycle
(G1, S, or G2). Based on the above, our main goal was to eval-
uate such a potential mechanism and balance between cell
death and cell cycle delays—which shows an effort of the cell
to repair the damage—while morphological observations were
performed through phase-contrast microscopy, showing a
typical rounding and detachment of the cells at the higher
NP concentrations. A range of concentrations between
0.3125ppm and 10ppm of both 7nm and 60nm AgNPs were
tested for effects on cell viability and proliferation in p53-
expressing HOS and p53-deficient Saos-2 osteosarcoma cell
lines. The concentrations used in this study were selected to
align with those used in the main comparatory study [32].
The sizes of AgNPs used in the present study, 7nm and
60nm, were selected to be in accordance with other studies
reporting on that subject [37, 39, 40].

The results of the present study indicate that all three
null hypotheses have to be rejected, as it was demonstrated
that the size of AgNPs affected the response of the tested
human osteosarcoma cell lines, the concentration of AgNPs
affected the response of the tested human osteosarcoma
cells, and finally the presence of protein p53 affected the
response of osteosarcoma cells to AgNP treatment. How-
ever, the results indicate that the size of the AgNPs and the
presence of p53 seem to have a stronger impact on the fate
of osteosarcoma cells, as a dose response with the concentra-
tions used in this study was not defined.

4.1. Effects of Size and Concentration on AgNPs on Cell
Viability. The first two hypotheses of this study examined
the effect of size and concentration of AgNPs on osteosar-
coma cells.

The findings of this study, regarding cell viability, are in
partial agreement with those of Kovacs et al. [37]. Our
results have clearly demonstrated that the 7 nm AgNPs are
very effective in significantly lowering the cell viability of
the p53-expressing HOS osteosarcoma cells, at both the
10 ppm and 5 ppm concentrations, at all time points. The
same efficacy was not observed at lower concentrations.
Kovacs and coworkers found that smaller AgNPs (5nm)
had a stronger cytotoxic effect on wild-type p53-containing
U2Os and p53-deficient Saos-2 osteosarcoma cells than
larger AgNPs (35 nm) [37]. According to additional pub-
lished data, there is a faster cellular uptake of smaller rather
than larger nanoparticles [41]. Moreover, another study has
shown that small AgNPs present a large total surface area
and demonstrate greater cytotoxicity, due to the increased
release of silver ions [23]. According to that study, the higher
silver release is associated with higher cytotoxicity in eukary-
otic cells, a finding which was verified by the results of the
present study, as well.

In the present study, the 60nm AgNPs were effective at
the 10 ppm concentration, at all time points, while these
AgNPs at lower concentrations did not display marked cyto-
toxicity against the HOS osteosarcoma cells. Unlike the p53-
expressing HOS cells, the cell viability of the p53-deficient
Saos-2 cells was not markedly affected by the AgNPs used
in this study, at any time point, a finding which is not in

140

120

100

80

60

40

20

0

C
el

l v
ia

bi
lit

y
(%

o
f c

on
tr

ol
)

C
on

tr
ol

10
p

pm

5
pp

m

2.
5

pp
m

1.
25

p
pm

0.
62

5
pp

m

0.
31

25
p

pm
140

120

100

80

60

40

20

0

C
el

l v
ia

bi
lit

y
(%

o
f c

on
tr

ol
)

MTT SAOS-2_7 nm

MTT SAOS-2_60 nm

⁎

⁎ ⁎
⁎

⁎

⁎

⁎
⁎

⁎

⁎

⁎

⁎

⁎

⁎

C
on

tr
ol

10
p

pm

5
pp

m

2.
5

pp
m

1.
25

p
pm

0.
62

5
pp

m

0.
31

25
p

pm

24 h

48 h

72 h

Figure 9: Saos-2 cell viability (percentage values), when subjected
to exposure of different concentrations of 7 nm and 60 nm AgNPs
(same color asterisks indicate statistically significant differences
between the control and suspensions of different AgNP
concentrations. Absence of asterisks or absence of same color
asterisks indicates no statistically significant differences, according
to Tukey’s HSD test for α = 0:05).

9BioMed Research International

agreement with the results of Kovacs and coworkers [37].
The results of the present study indicate that only the 7 nm
AgNPs at 10 ppm and 5 ppm and the 60 nm AgNPs at
10 ppm reach the threshold toxicity, which significantly
lowers the metabolic activity of the HOS osteosarcoma cells
[23, 42]. This effect of AgNPs may be explained by differen-
tial cellular uptake. It has been previously documented that
AgNPs can be incorporated into eukaryotic cells via endocy-
tosis mediated by caveolae and clathrin [43–45]. Further-
more, scanning electron microscopy has verified the
presence of AgNPs on cell membranes [46]. However,
AgNPs have not been detected either in the nucleus or in
the mitochondria [37]. Nevertheless, the endocytosed
AgNPs, according to many authors, act as “Trojan horses”
carrying and delivering silver ions into the cells [17, 41,
47]. It has been hypothesised that these ions are responsible
for all the biological phenomena observed.

All p53-expressing HOS cell groups treated with 7 nm
AgNPs at concentrations ranging between 2.5 ppm and
0.3125 ppm or 60nm AgNPs at concentrations ranging
between 5 ppm and 0.3125 ppm revealed an enhanced meta-
bolic activity. The same finding was observed for p53-
deficient Saos-2 cell groups treated with 7 nm and 60 nm
AgNPs at all tested concentrations. This could be explained
by elevated mitochondrial biogenesis, perhaps induced by
the oxidative stress that the endocytosed Ag ions caused
[48]. Cells suffer from oxidative stress when the cell cannot
detoxify and inactivate the reactive oxygen species (ROS)
that are produced [49]. It has been well documented in the
past that the production of ROS in the mitochondria is a
physiological process with ROS being a natural byproduct
of oxidative phosphorylation. The electron transport chain
on the inner mitochondrial membrane involves ATP syn-
thase and complexes I-IV. Eighty percent of the superoxide,
which is produced by complexes I and III, is released into
the intermembranous space, while the remaining 20% is
released at the mitochondrial matrix [50]. From the interior
of the mitochondria, superoxide leaks to the cytoplasm due
to the mitochondrial permeability transition pore (mPTP),
which is a protein existing in the mitochondrial outer mem-

brane [51, 52]. Subsequently, superoxide dismutase (SOD)
catalyzes the partitioning of superoxide to O2 and H2O2
(hydrogen peroxide). This process can take place either in
the mitochondrial matrix, where it is catalyzed by MnSOD,
or in the cytosol, where it is catalyzed by Cu/ZnSOD.
Hydrogen peroxide is considered a highly diffusible second
messenger. Crucially, the behavior of tumor cells is influ-
enced by the signaling events which are related to oxidation
stress [53–55]. Several events in cancer cell biology are asso-
ciated with ROS, including adhesion, angiogenesis, survival
and apoptosis, metabolism, progression, proliferation, motil-
ity, and tumor stemness [56].

4.2. Presence of Protein p53 and Cell Viability. The third
hypothesis of this study was to determine if the presence of
protein p53 affects the response of osteosarcoma cells to
treatment with AgNPs.

Many published papers have demonstrated that p53 is a
multitasking protein [57], shielding the cells against cancer
on many levels, including nucleotide excision repair
[58–61], base excision repair [62, 63], mismatch repair,
DNA double-strand break repair and recombination [64,
65], nonhomologous end joining [66, 67], homologous
recombination [68, 69], and interactions with REcQ heli-
cases [70, 71]. The results of the present study align with
these studies, i.e., a differential response in cell viability is
apparent after treatment with AgNPs that is dependent on
the p53 status of the cell lines. It should be mentioned how-
ever that the results of the present study are opposite from
those reported by Kovacs et al. [37], who have found that
p53-expressing U2Os and p53-deficient Saos-2 cells were
killed at approximately the same degree, when exposed to
AgNPs. The present study demonstrated that the p53-
expressing HOS osteosarcoma cells presented a significantly
diminished viability when subjected to high concentrations
of 7 nm (10 ppm and 5 ppm) and 60nM (10 ppm) AgNPs,
thus suggesting that p53 does play a role in AgNP-
mediated cytotoxicity. The putative mechanism is via high
oxidative stress leading to the production of increased ROS
levels [72]. Excessively high levels of ROS cause damage to

Table 4: Average percentage values and standard deviations for HOS cell proliferation, when subjected to exposure of different
concentrations of 7 and 60 nm AgNPs.

(a) BRDU

7 nm Control 10 ppm 5 ppm 2.5 ppm 1.25 ppm 0.625 ppm 0.3125 ppm

24 h 100.00 (±15.70) 1.17 (±0.86) 9.20 (±2.88) 112.47 (±25.79) 96.11 (±18.95) 82.67 (±10.46) 61.35 (±2.94)
48 h 100.00 (±6.68) 0 (±0.06) 2.44 (±1.23) 72.62 (±12.73) 75.68 (±18.66) 81.46 (±29.14) 80.94 (±31.82)
72 h 100.00 (±8.54) -0.69 (±0.27) 4.06 (±1.66) 65.82 (±28.57) 39.49 (±11.78) 61.99 (±5.45) 71.47 (±14.23)

(b) Average

60 nm Control 10 ppm 5 ppm 2.5 ppm 1.25 ppm 0.625 ppm 0.3125 ppm

24 h 100.00 (±15.70) 5.04 (±5.51) 107.67 (±13.20) 112.89 (±5.62) 91.03 (±4.28) 75.78 (±2.56) 89.00 (±8.63)
48 h 100.00 (±6.68) 1.65 (±2.45) 107.91 (±14.10) 97.75 (±16.07) 68.58 (±13.54) 86.65 (±6.10) 115.76 (±23.11)
72 h 100.00 (±8.54) 4.29 (±4.81) 71.43 (±4.17) 52.65 (±16.85) 59.21 (±9.91) 85.10 (±7.20) 92.33 (±24.92)

10 BioMed Research International

essential cell ingredients and structures, such as nucleic
acids, proteins, lipids, membranes, and organelles. This is
followed by activation of certain processes, leading eventu-
ally to apoptosis [73]. The main difference between the
Saos-2 and the HOS cells is that the latter expresses the
p53 protein. Therefore, it is logical to assume that p53 is
the factor that is responsible for the diminished viability of
the HOS osteosarcoma cells when exposed to AgNPs at high
concentrations. As the levels of ROS in the HOS cells
increase due to the encapsulation of AgNPs, there is an
interaction between the ROS and p53. A previous study
which conducted a microarray examination of cells treated
with hydrogen peroxide has found 16 genes, highly respon-

sive to H2O2, which were targeted by p53 [74]. Increased
levels of ROS in HOS cells stimulate certain pathways com-
bining p53 and redox signaling. The levels of ROS play a
very important role in the signals that will be initiated, in
order for p53 to target certain genes which will determine
the fate of the osteosarcoma cell. Two studies have found
that p53 suppresses antioxidant genes, and as a result, cellu-
lar ROS levels increase, leading to oxidative stress. Specifi-
cally, the manganese superoxide dismutase (MnSOD) gene,
which encodes an antioxidant enzyme (SOD2) that protects
cells from oxidative damage, is suppressed at the promoter
level by either p53 activation or by p53 overexpression [75,
76]. However, other antioxidant genes, such as ALDH4

24 h

48 h

72 h

160

140

120

100

80

60

40

20

0

C
el

l v
pr

ol
ife

ra
tin

(%
o

f c
on

tr
ol

)

BrdU HOS_7 nm

BrdU HOS_60 nm

160

−20

140

120

100

80

60

40

20

0

−20

C
el

l v
pr

ol
ife

ra
tin

(%
o

f c
on

tr
ol

)

0.
31

25
p

pm

C
on

tr
ol

10
p

pm

5
pp

m

2.
5

pp
m

1.
25

p
pm

0.
62

5
pp

m

⁎
⁎ ⁎

⁎

⁎
⁎

⁎
⁎
⁎ ⁎

⁎
⁎

⁎
⁎
⁎
⁎ ⁎
⁎

⁎
⁎
⁎
⁎

⁎
⁎

⁎
⁎

⁎
⁎

⁎
⁎

⁎
⁎

⁎
⁎ ⁎

⁎
⁎

⁎
⁎

⁎
⁎

⁎

⁎ ⁎

⁎
⁎ ⁎
⁎

⁎
⁎

⁎ ⁎
⁎

⁎

⁎

⁎

⁎

⁎
⁎

⁎

⁎
⁎
⁎
⁎
⁎
⁎
⁎ ⁎

⁎ ⁎
⁎

⁎
⁎

⁎

⁎
⁎

⁎
⁎

−20

0.
31

25
p

pm

C
on

tr
ol

10
p

pm

5
pp

m

2.
5

pp
m

1.
25

p
pm

0.
62

5
pp

m

Figure 10: HOS cell proliferation (percentage values), when subjected to exposure of different concentrations of 7 nm and 60 nm AgNPs
(same color asterisks indicate statistically significant differences between the control and suspensions of different AgNP concentrations.
Absence of asterisks or absence of same color asterisks indicates no statistically significant differences, according to Tukey’s HSD test for
α = 0:05).

11BioMed Research International

(aldehyde dehydrogenase 4), and PIG12, which is a novel
member of the microsomal glutathione S-transferase gene
family, have been shown to be concurrently upregulated
with p53 overexpression and seem to be like an adaptive
response to oxidative stress induced by p53 [77–79].
Increased ROS levels causing oxidative stress lead to mito-
chondrial lipid degradation, as well as morphological
changes, i.e., chromatin condensation and fragmentation,
and biochemical alterations, i.e., poly ADP-ribose polymer-
ase (PARP) caspase-mediated degradation, which are defi-
nite signs of cellular apoptosis [77].

4.3. Effects of Treatment Time on Cell Viability. Time-depen-
dent viability was also observed in the present study. A
slight increase in cell viability was noted from 24 to 72
hours for the 7 nm at 10ppm and 5 ppm and for the
60 nm for the 10 ppm. However, for the big majority of
the remaining concentrations (2.5 ppm, 1.25 ppm,
0.625 ppm, and 0.3125 ppm), this trend was not observed.
On the contrary, a marked decrease in HOS cell viability
was observed from 24 to 72 hours, reaching values to
about 60-70% of the control. This finding is in alignment
with the results of Kovacs and coworkers, who checked
cell viability at 24 and 48 hours [37]. Decreased cell viabil-
ity may indicate that the release of silver ions from the
AgNPs in the lowest concentrations takes more time, but
after 24 hours, the silver ion concentrations reach a level
which is capable of contributing to the generation of
ROS, which finally reach toxic levels and trigger apoptosis.
Although Saos-2 cell viability at 72 hours was decreased
compared to that of 24 hours, the cell viability values were
very close to those of the control.

4.4. Effect of AgNPs’ Size and Concentration on Cell
Proliferation. In addition to cell viability, which was tested
with the MTT assay, the BrdU test was used to target the cell
proliferation rates of p53-deficient Saos-2 and p53-
expressing HOS osteosarcoma cells. As BrdU is incorporated
into newly synthesized DNA, it can detect which cells are in
the S-phase of the cell cycle [80]. The same trend that was
noticed for metabolic activity, measured by the MTT assay,

was also noticed when the BrdU assay was employed. A
marked difference was noticed between the p53-deficient
Saos-2 cells and the p53-expressing HOS cells. In general,
cell proliferation of the HOS cells was decreased from the
24-hour to 72-hour time interval. Once again, the 7 nm
AgNPs had a marked effect on reduced HOS cell viability
at both the 10ppm and 5 ppm concentrations, while the
60 nm AgNPs presented a noticeable effect at only the high-
est concentration. A clear effect on reduced Saos-2 cell via-
bility was not noted at either the 7 nm or the 60nm
AgNPs, at any concentration.

Regarding cell proliferation, the results of the present
study are not in accordance with those reported by Kovacs
et al., who have found that p53-expressing U2Os and p53-
deficient Saos-2 cells presented similar proliferation rates
(in relation to the control), when exposed to AgNPs
[37]. The present study demonstrated that the p53-
expressing HOS osteosarcoma cells presented significantly
less proliferation compared to p53-deficient Saos-2 cells
when subjected to 7 nm AgNPs at 10 ppm and 5 ppm con-
centrations, and 60 nm at 10 ppm concentration. The
diminished cell proliferation noted for all time periods
when HOS cells were treated with 7 nm AgNPs at
10 ppm and 5 ppm, and with 60 nm AgNPs at 10ppm is
probably due to DNA damage through the elevation of
ROS at toxic levels. Several studies have demonstrated that
p53 apoptotic signals lead to activation of caspase [81, 82].
However, the exact mechanism by which this activation
occurs is still not well defined. It has also been speculated
that mitochondrial cytochrome c (mtCyt c), which is
needed for ATP production, is also involved in the apo-
ptotic procedures. Gao and colleagues reported that the
cytosolic release of cytochrome c, which activates caspases,
and membrane translocation of Bax are both mediated by
protein p53 [83]. Caspase-3 in particular is responsible for
morphological changes in the nucleus, through cleavage of
a variety of substrates, as well as for disintegration of the
DNA [82].

In addition to these apoptotic procedures, p53 is
responsible for cell cycle arrest, through interaction with
protein p21 (WAF1/C1P1), which acts as a signal to halt

Table 5: Average percentage values and standard deviations for Saos-2 cell proliferation, when subjected to exposure of different
concentrations of 7 nm and 60 nm AgNPs.

(a) BRDU

7 nm Control 10 ppm 5 ppm 2.5 ppm 1.25 ppm 0.625 ppm 0.3125 ppm

24 h 100.00 (±6.45) 108.73 (±3.20) 97.73 (±12.41) 94.23 (±5.28) 89.12 (±0.37) 80.88 (±3.92) 101.99 (±4.68)
48 h 100.00 (±6.62) 59.76 (±16.90) 93.42 (±14.61) 95.31 (±8.78) 91.08 (±0.60) 87.58 (±13.32) 99.91 (±21.24)
72 h 100.00 (±7.07) 83.93 (±9.43) 89.84 (±2.81) 89.75 (±5.69) 90.60 (±1.38) 90.89 (±3.83) 97.02 (±6.91)

(b) Average

60 nm Control 10 ppm 5 ppm 2.5 ppm 1.25 ppm 0.625 ppm 0.3125 ppm

24 h 100.00 (18.91) 102.21 (4.83) 81.25 (21.52) 90.01 (3.52) 83.03 (16.40) 86.82 (21.28) 95.95 (10.43)

48 h 100.00 (6.62) 100.23 (2.60) 99.28 (14.65) 118.66 (2.05) 110.94 (8.61) 108.84 (4.82) 104.26 (6.75)

72 h 100.00 (7.07) 93.12 (11.91) 85.25 (14.54) 91.60 (13.22) 84.21 (0.60) 93.57 (5.98) 104.29 (7.52)

12 BioMed Research International

cell division. p21 binds to cyclin-CDK complexes, which
are responsible for promoting the cell cycle. This binding
has as a result the inhibition of kinase activity and there-
fore arrest of the cell cycle. It has been demonstrated that
gene p21 has numerous elements mediating p53 binding,
activating in this way the gene which is responsible for
p21 protein encoding [84, 85].

In general, the results of the present study confirm that
lower size AgNPs are more effective at decreasing cell viabil-
ity than the larger ones. However, in conflict with previously
published work [37], we found that AgNP treatment mark-
edly decreased cell viability of p53-expressing HOS cells,
but not of p53-deficient Saos-2 cells, suggesting a role for
p53 in AgNP-mediated cytotoxicity. Therefore, this study
supports the notion that the treatment with AgNPs is more
effective if the osteosarcoma cells express protein p53.

5. Conclusions

The results of the present study suggest that AgNPs of a
smaller diameter (7 nm) are more effective on osteosarcoma
cell viability and cell proliferation than those of a bigger
diameter (60 nm). Furthermore, a higher concentration of
AgNPs is more effective than that of a smaller concentration.
The 5 ppm concentration is effective only for the 7 nm
AgNPs. Within the 72-hour period, treatment with AgNPs
of 7 nm at 5 or 10 ppm is highly effective against p53-
expressing osteosarcoma cells, but it is not effective against
p53-deficient osteosarcoma cells. Concentrations of less than
5 ppm for the 7 nm silver nanoparticles and less than 10 ppm
for the 60 nm silver nanoparticles are not effective. The pres-
ence of protein p53 affects significantly the efficacy of AgNPs
on osteosarcoma cells.

140

120

100

80

60

40

20

0

C
el

l p
ro

lif
er

at
io

n
(%

o
f c

on
tr

ol
)

C
on

tr
ol

10
p

pm

5
pp

m

2.
5

pp
m

1.
25

p
pm

0.
62

5
pp

m

0.
31

25
p

pm

140

120

100

80

60

40

20

0

C
el

l p
ro

lif
er

at
io

n
(%

o
f c

on
tr

ol
)

C
on

tr
ol

10
p

pm

5
pp

m

2.
5

pp
m

1.
25

p
pm

0.
62

5
pp

m

0.
31

25
p

pm

BrdU SAOS-2_7 nm

BrdU SAOS-2_60 nm

⁎

⁎

⁎

⁎

⁎ ⁎
⁎

⁎

⁎

⁎

⁎
⁎

⁎
⁎

⁎
⁎

⁎

24 h

48 h

72 h

Figure 11: Saos-2 cell proliferation (percentage values), when subjected to exposure of different concentrations of 7 nm and 60 nm AgNPs
(same color asterisks indicate statistically significant differences between the control and suspensions of different AgNP concentrations.
Absence of asterisks or absence of same color asterisks indicates no statistically significant differences, according to Tukey’s HSD test for
α = 0:05).

13BioMed Research International

The results of the present study suggest that the use of
AgNPs against certain types of osteosarcoma, which involve
the presence of protein p53, seems to be effective. However,
preclinical testing is needed to further establish the efficacy
and the safety of AgNP use. Other parameters, including the
best route of administration, the therapeutic window, pharma-
codynamics, pharmacokinetics, and any potential side effects,
all need to be established. This in vitro study contributes to the
growing body of evidence that AgNPs might be a useful addi-
tion to the armamentarium of osteosarcoma treatment, but
there is much to be done before AgNPs can be shown to be
effective in the clinical arena. There is however a glimmer of
hope that AgNPs may be translated into a useful treatment
in the fight against osteosarcoma.

Data Availability

All data are available upon request.

Ethical Approval

The HOS and Saos-2 cells needed for this research project
were obtained from the American Type Culture Collection
(ATCC No. HTB 85). All procedures were performed in
accordance with the Declaration of Helsinki.

Conflicts of Interest

Drs. Nikolaos Michailidis and Alexander Tsouknidas are
shareholders of the PLiN Nanotechnology S.A., Thessalo-
niki, Greece. The principal investigator and the rest of the
authors have no conflict of interest.

Authors’ Contributions

KM was the principal investigator of this study, AB and EP
were tissue engineering and biology experts and assisted in
the experiments and explanation of the observed phenomena,
AT and NM were involved in the synthesis of the AgNPs and
assisted in the explanation of the biological action of silver
nanoparticles, and EJ was the clinical expert who assisted in
organising the project and helped in manuscript revision.

Acknowledgments

The authors would like to thank PLiN Nanotechnology S.A.,
Thessaloniki, Greece, for providing the silver nanoparticles.

References

[1] M. L. Broadhead, J. C. M. Clark, D. E. Myers, C. R. Dass, and
P. F. M. Choong, “The molecular pathogenesis of osteosarcoma:
a review,” Sarcoma, vol. 2011, Article ID 959248, 12 pages, 2011.

[2] M. P. Link and E. Eilber, “Osteosarcoma,” in Principles and
Practice of Pediatric Oncology, P. A. Pizzo and D. G. Poplack,
Eds., pp. 889–920, Lippincott-Raven, Philadelphia, 3rd ed. edi-
tion, 1997.

[3] P. A. Meyers and R. Gorlick, “Osteosarcoma,” Pediatric Clinics
of North America, vol. 44, no. 4, pp. 973–989, 1997.

[4] G. Rosen, C. A. Forscher, and H. J. Mankin, Cancer Medicine,
J. F. Holland and E. Frei, Eds., Hamilton, BC Decker, 5th ed.
edition, 2000.

[5] S. S. Bielack, B. Kempf-Bielack, G. Delling et al., “Prognostic
factors in high-grade osteosarcoma of the extremities or trunk:
an analysis of 1, 702 patients treated on neoadjuvant coopera-
tive osteosarcoma study group protocols,” Journal of Clinical
Oncology, vol. 20, no. 3, pp. 776–790, 2002.

[6] M. W. Bishop, K. A. Janeway, and R. Gorlick, “Future direc-
tions in the treatment of osteosarcoma,” Current Opinion in
Pediatrics, vol. 28, no. 1, pp. 26–33, 2016.

[7] C. Carr, J. Ng, and T. Wigmore, “The side effects of chemo-
therapeutic agents,” Current Anaesthesia and Critical Care,
vol. 19, no. 2, pp. 70–79, 2008.

[8] D. Kerr, L. Rowett, and A. Young, “Cancer biotherapy – an
introduction,” in Cancer Biotherapy: An Introductory Guide,
Y. A. Rowett and D. L. Kerr, Eds., Oxford University Press,
Oxford, 2006.

[9] R. A. Ismail, G. M. Sulaiman, M. H. Mohsin, and A. H. Saa-
doon, “Preparation of silver iodide nanoparticles using laser
ablation in liquid for antibacterial applications,” IET Nanobio-
technology, vol. 12, no. 6, pp. 781–786, 2018.

[10] H. Barabadi, H. Vahidi, K. Damavandi Kamali et al., “Emerg-
ing theranostic gold nanomaterials to combat lung cancer: a
systematic review,” Journal of Cluster Science, vol. 31, no. 2,
pp. 323–330, 2020.

[11] H. Barabadi, H. Vahidi, M. A. Mahjoub et al., “Emerging anti-
neoplastic gold nanomaterials for cervical cancer therapeutics:
a systematic review,” Journal of Cluster Science, vol. 31, no. 6,
pp. 1173–1184, 2020.

[12] A. Khatua, A. Prasad, E. Priyadarshini et al., “Emerging anti-
neoplastic plant-based gold nanoparticle synthesis: a mecha-
nistic exploration of their anticancer activity toward cervical
cancer cells,” Journal of Cluster Science, vol. 31, no. 6,
pp. 1329–1340, 2020.

[13] H. Barabadi, H. Vahidi, K. Damavandi Kamali et al., “Emerg-
ing theranostic silver nanomaterials to combat colorectal can-
cer: a systematic review,” Journal of Cluster Science, vol. 31,
no. 2, pp. 311–321, 2020.

[14] M. S. Jabir, Y. M. Saleh, G. M. Sulaiman et al., “Green synthesis
of silver nanoparticles using Annona muricata extract as an
inducer of apoptosis in cancer cells and inhibitor for NLRP3
inflammasome via enhanced autophagy,” Nanomaterials,
vol. 11, no. 2, p. 384, 2021.

[15] M. S. Jabir, A. A. Hussien, G. M. Sulaiman et al., “Green syn-
thesis of silver nanoparticles fromEriobotrya japonicaextract:
a promising approach against cancer cells proliferation,
inflammation, allergic disorders and phagocytosis induction,”
Artificial Cells, Nanomedicine, and Biotechnology, vol. 49,
no. 1, pp. 48–60, 2021.

[16] K. O’Day and R. Gorlick, “Novel therapeutic agents for osteo-
sarcoma,” Expert Review of Anticancer Therapy, vol. 9, no. 4,
pp. 511–523, 2009.

[17] E. Park, J. Yi, Y. Kim, K. Choi, and K. Park, “Silver nanoparti-
cles induce cytotoxicity by a Trojan-horse type mechanism,”
Toxicology In Vitro, vol. 24, no. 3, pp. 872–878, 2010.

[18] S. Kim, J. E. Choi, J. Choi et al., “Oxidative stress-dependent
toxicity of silver nanoparticles in human hepatoma cells,” Tox-
icology In Vitro, vol. 23, no. 6, pp. 1076–1084, 2009.

[19] A. Avalos, A. I. Haza, D. Mateo, and P. Morales, “Cytotoxicity
and ROS production of manufactured silver nanoparticles of

14 BioMed Research International

different sizes in hepatoma and leukemia cells,” Journal of
Applied Toxicology, vol. 34, no. 4, pp. 413–423, 2014.

[20] J. Liu, D. A. Sonshine, S. Shervani, and R. H. Hurt, “Controlled
release of biologically active silver from nanosilver surfaces,”
ACS Nano, vol. 4, no. 11, pp. 6903–6913, 2010.

[21] T. Verano-Braga, R. Miethling-Graff, K. Wojdyla et al.,
“Insights into the cellular response triggered by silver nano-
particles using quantitative proteomics,” ACS Nano, vol. 8,
no. 3, pp. 2161–2175, 2014.

[22] M. A. Raza, Z. Kanwal, A. Rauf, A. N. Sabri, S. Riaz, and
S. Naseem, “Size- and shape-dependent antibacterial studies
of silver nanoparticles synthesized by wet chemical routes,”
Nanomaterials, vol. 6, no. 4, p. 74, 2016.

[23] A. R. Gliga, S. Skoglund, I. Odnevall Wallinder, B. Fadeel, and
H. Karlsson, “Size-dependent cytotoxicity of silver nanoparti-
cles in human lung cells: the role of cellular uptake, agglomer-
ation and Ag release,” Particle and Fibre Toxicology, vol. 11,
no. 1, p. 11, 2014.

[24] C. A. Finlay, P. W. Hinds, and A. J. Levine, “The p53 proto-
oncogene can act as a suppressor of transformation,” Cell,
vol. 57, no. 7, pp. 1083–1093, 1989.

[25] A. J. Levine, C. A. Finlay, and P. W. Hinds, “P53 is a tumor
suppressor gene,” Cell, vol. 116, pp. S67–S70, 2004.

[26] N. Fenouille, G. Robert, M. Tichet et al., “The p53/p21Cip1/
Waf 1 pathway mediates the effects of SPARC on melanoma
cell cycle progression,” Pigment Cell & Melanoma Research,
vol. 24, no. 1, pp. 219–232, 2011.

[27] C. Deng, P. Zhang, J. Wade Harper, S. J. Elledge, and P. Leder,
“Mice Lacking p21 CIP1/WAF1_ undergo normal development,
but are defective in G1 checkpoint control,” Cell, vol. 82,
no. 4, pp. 675–684, 1995.

[28] A. L. Gartel and S. K. Radhakrishnan, “Lost in transcription:
p21 repression, mechanisms, and Consequences,” Cancer
Research, vol. 65, no. 10, pp. 3980–3985, 2005.

[29] H. Lodish, A. Berk, C. A. Kaiser et al., Molecular Cell Biology,
WH Freeman and Co, New York, 7th ed. edition, 2013.

[30] D. P. Lane, “p53, guardian of the genome,” Nature, vol. 358,
no. 6381, pp. 15-16, 1992.

[31] S. Venkatachalam, S. Tyner, and L. A. Donehower, “The p53-
deficient mouse as a cancer model,” in Tumor Models in Can-
cer Research. Cancer Drug Discovery and Development, B. A.
Teicher, Ed., pp. 247–261, Humana Press, Totowa, NJ, 2002.

[32] M. Levine, “P53, the cellular gatekeeper for growth and divi-
sion,” Cell, vol. 88, no. 3, pp. 323–331, 1997.

[33] G. Lozano and S. Elledge, “p53 sends nucleotides to repair
DNA,” Nature, vol. 404, no. 6773, pp. 24-25, 2000.

[34] B. Vogelstein, D. Lane, and A. J. Levine, “Surfing the p53 net-
work,” Nature, vol. 408, no. 6810, pp. 307–310, 2000.

[35] J. W. Romano, J. C. Ehrhart, A. Duthu, C. M. Kim, E. Apella,
and P. May, “Identification and characterization of a p53 gene
mutation in human osteosarcoma cell line,” Oncogene, vol. 4,
p. 1483, 1989.

[36] N. Chandar, B. Billig, J. McMaster, and J. Novak, “Inactivation
of p53 gene in human and murine osteosarcoma cells,” British
Journal of Cancer, vol. 65, no. 2, pp. 208–214, 1992.

[37] D. Kovacs, N. Igaz, C. Keskeny et al., “Silver nanoparticles
defeat p53-positive and p53-negative osteosarcoma cells by
triggering mitochondrial stress and apoptosis,” Scientific
Reports, vol. 6, no. 1, p. 27902, 2016.

[38] Y. Kirmanidou, M. Sidira, A. Bakopoulou et al., “Assessment
of cytotoxicity and antibacterial effects of silver nanoparticle-

doped titanium alloy surfaces,” Dental Materials, vol. 35,
no. 9, pp. e220–e233, 2019.

[39] J. Xu, J. Ma, X. Sun et al., “Ag nanoparticles sensitize IR-
induced killing of cancer cells,” Cell Research, vol. 19, no. 8,
pp. 1031–1034, 2009.

[40] M. Jeyaraj, G. Sathishkumar, G. Sivanandhan et al., “Biogenic
silver nanoparticles for cancer treatment: an experimental
report,” Colloids and Surfaces, B: Biointerfaces, vol. 106,
pp. 86–92, 2013.

[41] I. L. Hsiao, Y. K. Hsieh, C. F. Wang, I. C. Chen, and Y. J.
Huang, “Trojan-horse mechanism in the cellular uptake of sil-
ver nanoparticles verified by direct Intra- and extracellular sil-
ver speciation analysis,” Environmental Science & Technology,
vol. 49, no. 6, pp. 3813–3821, 2015.

[42] S. Kaba and E. M. Egorova, “In vitro studies of the toxic effects
of silver nanoparticles on Hela and U937 cells,” Nanotechnol-
ogy, Science and Applications, vol. 8, pp. 19–29, 2015.

[43] R. R. Miranda, A. G. Bezerra Jr, C. A. Oliveira Ribeiro et al.,
“Toxicological interactions of silver nanoparticles and non-
essential metals in human hepatocarcinoma cell line,” Toxicol-
ogy In Vitro, vol. 40, pp. 134–143, 2017.

[44] M. Milić, G. Leitinger, I. Pavičić et al., “Cellular uptake and
toxicity effects of silver nanoparticles in mammalian kidney
cells,” Journal of Applied Toxicology, vol. 35, no. 6, pp. 581–
592, 2015.

[45] I. Hsiao, Y. Hsieh, C. Chuang, C. Wang, and Y. Huang, “Effects
of silver nanoparticles on the interactions of neuron- and glia-
like cells: toxicity, uptake mechanisms, and lysosomal track-
ing,” Environmental Toxicology, vol. 32, no. 6, pp. 1742–
1753, 2017.

[46] X. Sun, J. Shi, X. Zou, C. Wang, Y. Yang, and H. Zhang, “Silver
nanoparticles interact with the cell membrane and increase endo-
thelial permeability by promoting VE-cadherin internalization,”
Journal of Hazardous Materials, vol. 317, pp. 570–578, 2016.

[47] V. De Matteis, M. A. Malvindi, A. Galeone et al., “Negligible
particle-specific toxicity mechanism of silver nanoparticles:
the role of Ag+ ion release in the cytosol,” Nanomedicine,
vol. 11, no. 3, pp. 731–739, 2015.

[48] C. F. Lee, C. Y. Liu, R. H. Hsieh, and Y. H. Wei, “Oxidative
stress-induced depolymerization of microtubules and alter-
ation of mitochondrial mass in human cells,” Annals of the
New York Academy of Sciences, vol. 1042, no. 1, pp. 246–254,
2005.

[49] P. Ray, B. Huang, and Y. Tsuji, “Reactive oxygen species (ROS)
homeostasis and redox regulation in cellular signaling,” Cellu-
lar Signalling, vol. 24, no. 5, pp. 981–990, 2012.

[50] D. Han, E. Williams, and E. Cadenas, “Mitochondrial respira-
tory chain-dependent generation of superoxide anion and its
release into the intermembrane space,” The Biochemical Jour-
nal, vol. 353, no. 2, pp. 411–416, 2001.

[51] M. Crompto, “The mitochondrial permeability transition pore
and its role in cell death,” The Biochemical Journal, vol. 341,
no. 2, pp. 233–249, 1999.

[52] P. Stor, “Reactive oxygen species-mediated mitochondria-to-
nucleus signaling: a key to aging and radical-caused diseases,”
Science’s STKE, vol. 2006, no. 332, p. re 3, 2006.

[53] T. P. Szatrowski and C. F. Nathan, “Production of large
amounts of hydrogen peroxide by human tumor cells,” Cancer
Research, vol. 51, pp. 794–798, 1991.

[54] A. Gupta, S. F. Rosenberger, and G. T. Bowden, “Increased
ROS levels contribute to elevated transcription factor and

15BioMed Research International

MAP kinase activities in malignantly progressed mouse kerati-
nocyte cell lines,” Carcinogenesis, vol. 20, no. 11, pp. 2063–
2073, 1999.

[55] P. Storz, “Reactive oxygen species in tumor progression,”
Frontiers in Bioscience, vol. 10, no. 1-3, pp. 1881–1896, 2005.

[56] G. Y. Liou and P. Storz, “Reactive oxygen species in cancer,”
Free Radical Research, vol. 44, no. 5, pp. 479–496, 2010.

[57] A. B. Williams and B. Schumacher, “p53 in the DNA-damage-
repair process,” Cold Spring Harbor Perspectives in Medicine,
vol. 6, no. 5, article a026070, 2016.

[58] S. Lee, B. Elenbaas, A. Levine, and J. Griffith, “p53 and its 14
kDa C-terminal domain recognize primary DNA damage in
the form of insertion/deletion mismatches,” Cell, vol. 81,
no. 7, pp. 1013–1020, 1995.

[59] X. W. Wang, H. Yeh, L. Schaeffer et al., “p53 modulation of
TFIIH-associated nucleotide excision repair activity,” Nature
Genetics, vol. 10, no. 2, pp. 188–195, 1995.

[60] R. Mirzayans, L. Enns, K. Dietrich, R. D. Barley, and M. C. Pat-
erson, “Faculty DNA polymerase δ/ɛ-mediated excision repair
in response to γ radiation or ultraviolet light in p53-deficient
fibroblast strains from affected members of a cancer-prone
family with Li-Fraumeni syndrome,” Carcinogenesis, vol. 17,
no. 4, pp. 691–698, 1996.

[61] Y. Liu and M. Kulesz-Martin, “p53 protein at the hub of cellu-
lar DNA damage response pathways through sequence-
specific and non-sequence-specific DNA binding,” Carcino-
genesis, vol. 22, no. 6, pp. 851–860, 2001.

[62] L. Jayaraman, N. C. Moorthy, K. G. Murthy, J. L. Manley,
M. Bustin, and C. Prives, “High mobility group protein-1
(HMG-1) is a unique activator of p53,” Genes & Development,
vol. 12, no. 4, pp. 462–472, 1998.

[63] C. Gaiddon, N. C. Moorthy, and C. Prives, “Ref-1 regulates the
transactivation and pro-apoptotic functions of p53 in vivo,”
The EMBO Journal, vol. 18, no. 20, pp. 5609–5621, 1999.

[64] L. Deriano and D. B. Roth, “Modernizing the nonhomologous
end-joining repertoire: alternative and classical NHEJ share
the stage,” Annual Review of Genetics, vol. 47, no. 1, pp. 433–
455, 2013.

[65] M. Jasin and R. Rothstein, “Repair of strand breaks by homol-
ogous recombination,” Cold Spring Harbor Perspectives in
Biology, vol. 5, no. 11, article a012740, 2013.

[66] Y. Gao, D. O. Ferguson, W. Xie et al., “Interplay of p53 and
DNA-repair protein XRCC4 in tumorigenesis, genomic stability
and development,” Nature, vol. 404, no. 6780, pp. 897–900, 2000.

[67] M. J. Difilippantonio, S. Petersen, H. T. Chen et al., “Evidence
for replicative repair of DNA double-strand breaks leading to
oncogenic translocation and gene amplification,” The Journal
of Experimental Medicine, vol. 196, no. 4, pp. 469–480, 2002.

[68] H. Willers, E. E. McCarthy, B. Wu et al., “Dissociation of p53-
mediated suppression of homologous recombination from G1/
S cell cycle checkpoint control,” Oncogene, vol. 19, no. 5,
pp. 632–639, 2000.

[69] C. Arias-Lopez, I. Lazaro-Trueba, P. Kerr et al., “p53 modu-
lates homologous recombination by transcriptional regulation
of the RAD51 gene,” EMBO Reports, vol. 7, no. 2, pp. 219–224,
2006.

[70] R. J. Monnat, “Human RECQ helicases: roles in DNA metab-
olism, mutagenesis and cancer biology,” Seminars in Cancer
Biology, vol. 20, no. 5, pp. 329–339, 2010.

[71] D. L. Croteau, V. Popuri, P. L. Opresko, and V. A. Bohr,
“Human RecQ helicases in DNA repair, recombination, and

replication,” Annual Review of Biochemistry, vol. 83, no. 1,
pp. 519–552, 2014.

[72] D. Guo, L. Zhu, Z. Huang et al., “Anti-leukemia activity of
PVP-coated silver nanoparticles via generation of reactive oxy-
gen species and release of silver ions,” Biomaterials, vol. 34,
no. 32, pp. 7884–7894, 2013.

[73] M. Redza-Dutordoir and D. A. Averill-Bates, “Activation of
apoptosis signalling pathways by reactive oxygen species,” Bio-
chimica et Biophysica Acta, vol. 1863, no. 12, pp. 2977–2992,
2016.

[74] S. Desaint, S. Luriau, J. C. Aude, G. Rousselet, and M. B. Tole-
dano, “Mammalian Antioxidant Defenses Are Not Inducible
by H2O2,” The Journal of Biological Chemistry, vol. 279,
no. 30, pp. 31157–31163, 2004.

[75] P. Drane, A. Bravard, V. Bouvard, and E. May, “Reciprocal
down-regulation of p53 and SOD2 gene expression -implica-
tion in p53 mediated apoptosis,” Oncogene, vol. 20, no. 4,
pp. 430–439, 2001.

[76] S. K. Dhar, Y. Xu, Y. Chen, and D. K. St Clair, “Specificity pro-
tein 1-dependent p53-mediated suppression of human manga-
nese superoxide dismutase gene expression,” The Journal of
Biological Chemistry, vol. 281, pp. 21698–21709, 2006.

[77] K. Polyak, Y. Xia, J. L. Zweier, K. W. Kinzler, and
B. Vogelstein, “A model for p53-induced apoptosis,” Nature,
vol. 389, no. 6648, pp. 300–305, 1997.

[78] K.-A. Yoon, Y. Nakamura, and H. Arakawa, “Identification of
_ALDH4_ as a p53-inducible gene and its protective role in
cellular stresses,” Journal of Human Genetics, vol. 49, no. 3,
pp. 134–140, 2004.

[79] B. Liu, Y. Chen, and D. K. St Clair, “ROS and p53: a versatile
partnership,” Free Radical Biology & Medicine, vol. 44, no. 8,
pp. 1529–1535, 2008.

[80] R. Leif, J. H. Stein, and R. M. Zucker, “A short history of the
initial application of Anti-5-BrDU to the detection and mea-
surement of S phase,” Cytometry. Part A, vol. 58A, no. 1,
pp. 45–52, 2004.

[81] M. Schuler, E. Bossy-Wetzel, J. C. Goldstein, P. Fitzgerald, and
D. R. Green, “p53 Induces Apoptosis by Caspase Activation
through Mitochondrial Cytochrome _c_ Release,” The Journal
of Biological Chemistry, vol. 275, no. 10, pp. 7337–7342, 2000.

[82] N. Kanipandian, D. Li, and S. Kannan, “Induction of intrinsic
apoptotic signaling pathway in A549 lung cancer cells using
silver nanoparticles from _Gossypium hirsutum_ and evalua-
tion of _in vivo_ toxicity,” Biotechnology Reports, vol. 23, arti-
cle e00339, 2019.

[83] C. Feng Gao, S. Ren, L. Zhang et al., “Caspase-Dependent
Cytosolic Release of Cytochrome _c_ and Membrane Translo-
cation of Bax in p53-Induced Apoptosis,” Experimental Cell
Research, vol. 265, no. 1, pp. 145–151, 2001.

[84] A. A. Elbendary, F. D. Cirisano, A. C. Evans Jr. et al., “Rela-
tionship between p21 expression and mutation of the p53
tumor suppressor gene in normal and malignant ovarian
epithelial cells,” Clinical Cancer Research, vol. 2, pp. 1571–
1575, 1996.

[85] I. K. Bukholm, J. M. Nesland, R. Kåresen, U. Jacobsen, and
A. L. Børresen, “Relationship between abnormal p53 protein
and failure to express p21 protein in human breast carcino-
mas,” The Journal of Pathology, vol. 181, no. 2, pp. 140–145,
1997.

16 BioMed Research International

Copyright of BioMed Research International is the property of Hindawi Limited and its
content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder’s express written permission. However, users may print, download, or email
articles for individual use.