International Journal of Biological Macromolecules 93 (2016) 1465–1478

Contents lists available at ScienceDirect

International Journal of Biological Macromolecules

journal homepage: www.elsevier.com/locate/ijbiomac

Glycol chitosan/nanohydroxyapatite biocomposites for potential bone

tissue engineering and regenerative medicine
Vitor C. Dumont a,b , Herman S. Mansur a,b,∗ , Alexandra A.P. Mansur a,b ,
Sandhra M. Carvalho a,b , Nádia S.V. Capanema a,b , Breno R. Barrioni b
a
Center of Nanoscience, Nanotechnology and Innovation—CeNano2 I, Federal University of Minas Gerais-UFMG, Brazil
b
Department of Metallurgical and Materials Engineering, Federal University of Minas Gerais-UFMG, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: In the last few decades, research on biocomposite nanomaterials has grown exponentially due to the
Received 27 January 2016 global demand for novel solutions in bone tissue engineering and repair. In the present study, it is reported
Received in revised form 7 April 2016 the design and synthesis of biocomposites based on glycol chitosan (GLY-CHI) matrices incorporated with
Accepted 12 April 2016
nano-hydroxyapatite particles (nHA) produced via an eco-friendly chemical colloidal process in water
Available online 13 April 2016
media followed by solvent casting and evaporation methods at room temperature. The structure, mor-
phology, and crystallinity of the components and biocomposites were extensively characterized by light
Keywords:
microscopy (LM), scanning electron microscopy (SEM), transmission electron microscopy (TEM), energy-
Chitosan
Biocomposite
dispersive X-ray spectroscopy (EDX), wavelength dispersive X-ray fluorescence spectroscopy (WD-XRF),
Glycol-chitosan X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), and X-ray micro-computed
Nanoparticle tomography analysis (␮CT). Furthermore, cytotoxicity and cell viability tests were performed on three
Hydroxyapatite nanoparticles cell lines using a 3-(4,5-dimethylthiazol-2yl) 2,5-diphenyl tetrazolium bromide (MTT) assay, an alkaline
phosphatase (ALP) activity test, and LIVE/DEAD® assays. The results demonstrated that the GLY-CHI ligand
played a major role in the nucleation, growth and colloidal stabilization of calcium phosphate particles at
nanoscale dimensions with a narrow distribution and average size of 74 ± 15 nm. The FTIR spectroscopy
associated with the XRD results indicated that nanosized hydroxyapatite (nHA) was the predominant
calcium phosphate phase produced in the colloidal processing route. In addition, the X-ray micro-CT
analysis of the nanocomposite membranes showed that nHA particles were homogenously dispersed in
the glycol-chitosan polymeric matrix. Moreover, according to the in vitro bioassays, the biocomposites
showed an adequate cell viability response and non-cytotoxic behavior toward osteoblastic-like (SAOS)
and embryonic cell lines (HEK293T). Finally, the results of osteogenic differentiation tests demonstrated
that the nHA/GLY-CHI composites are osteoinductive for human bone marrow mesenchymal stem cells
(HBMS), which can be envisioned for prospective use in tissue engineering (e.g., bone, cartilage and
periodontal) applications.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction in the field of biomaterials, the development of innovative bio-

engineered nanomaterials that fulfill all of the requirements for
Millions of patients worldwide are suffering from bone defects bone tissue reconstruction represents an important challenge to
caused by accidents, violence, trauma, cancer, congenital defor- overcome by the regenerative medicine researchers and profes-
mity, surgical reconstruction, and bone-related diseases; however, sionals [1–6]. Essentially, the primary difficulty arises from the
no definitive ideal solution for bone tissue repair and replacement is fact that natural bone possesses a very complex inorganic–organic
available [1–6]. Despite undeniable advances in the recent decades hybrid structure that is well-organized in an intrinsically hierarchi-
cal architecture composed of nanocrystals of hydroxyapatite [HA,
Ca10 (PO4 )6 (OH)2 ] and fibrils of collagen proteins [7,8]. Thus, to
∗ Corresponding author at: Department of Metallurgical and Materials Engi-
mimic the structure and composition of living tissues, the strategy
neering, Federal University of Minas Gerais, Av. Antônio Carlos, 6627—Escola de
of combining different biomaterials, molecules, and cells to form
Engenharia, Bloco 2—Sala 2233, 31.270-901, Belo Horizonte, MG, Brazil. nanostructured biocomposites and hybrids may be considered a
E-mail addresses: [email protected], [email protected] very attractive alternative [1,7,9–11]. In general, hybrid composite
(H.S. Mansur).

http://dx.doi.org/10.1016/j.ijbiomac.2016.04.030
0141-8130/© 2016 Elsevier B.V. All rights reserved.
1466 V.C. Dumont et al. / International Journal of Biological Macromolecules 93 (2016) 1465–1478

materials represent the combination of two or more distinct com- ers [33] and gene delivery [34], no study was found in the consulted
ponents, organic and inorganic, each with important contributions literature using glycol-chitosan combined with HA for producing
to the area of materials science and each with unique characteristics nanocomposite membranes for potential periodontal repair and
that result in improved properties of the whole system [12,13]. regeneration.
More recently, natural polymer-based composites, such as Thus, this study reports for the first time the one-pot synthe-
chitin, chitosan, and collagen, have gradually garnered more sis and characterization of nanocomposites using glycol-chitosan
attention than synthetic polymer-based composites for tissue simultaneously as the capping ligands and the biopolymeric matrix
engineering bio-applications (e.g., bone, cartilage and periodon- for the formation of nano-hydroxyapatite particles via aqueous
tal) primarily due to several aspects, such as abundance in colloidal chemistry. In addition, these biocomposites combin-
nature, worldwide availability, biocompatibility, and environ- ing inorganic nHA with organic glycol-chitosan were extensively
mental compatibility [1,3,14]. Among the many choices of tested for preliminary cytocompatibility using MTT cell prolif-
polysaccharide-based biopolymers for developing composites for eration assay with three human cell cultures (osteoblastic-like,
bio-applications, chitosan has increasingly been used because embryonic cells and human bone marrow mesenchymal stem cells)
of its exceptional amalgamation of properties, such as natural as well as using an alkaline phosphatase (ALP) activity test and a
biocompatibility, biodegradability, and low immunogenicity [15]. LIVE/DEAD® viability-cytotoxicity assay.
Chitosan (CHI) is a linear biopolymer commonly derived from the
deacetylation of chitin formed by ˇ-(1,4)-2-acetamido-2-deoxy-
2. Experimental procedure
d-glucose and ˇ-(1,4)-amino-2-deoxy-d-glucose units, and it has
been widely investigated as the organic component of composites
2.1. Materials
for bone-tissue engineering [1,15]. However, it is only reasonably
soluble in acidic water solutions, mostly due to the protonation
All of the reagents and precursors, phosphoric acid
of amino groups (R-NH3 + ) with poor solubility above a pH of 6.5
(Sigma–Aldrich, USA, 85%, H3 PO4 ), calcium hydroxide
(pKa = 6.5), such as under physiological conditions common to liv-
(Sigma–Aldrich, USA, ≥96%, Ca(OH)2 ), and ammonium hydroxide
ing animals [16,17]. Therefore, to broaden the range of solubility
(Synth, Brazil, 30%, NH4 OH) were used as received. Glycol-chitosan
and to simultaneously add new properties to the polysaccharide
powder (Sigma–Aldrich, St. Louis, MO, USA, PN# G7753; degree
backbone, chitosan derivatives, such as carboxymethyl chitosan
of polymerization = 2000 Mw–410 kDa; degree of deacetylation
(CMC) [18–20], PEGylated-chitosan [21,22] and glycol-chitosan
DD = 76.2%) was used as the polymer matrix. The schematic
(GLY-CHI) [23] have been recently synthesized and investigated
representation of the chemical structure of the glycol-chitosan
for bio-medical and environmental applications. Glycol-based chi-
biomolecule is depicted in Fig. 1. Deionized water (Millipore
tosan derivatives may be suitable candidates as water-soluble
SimplicityTM ) with a resistivity of 18 M cm was used in the
biopolymers for producing composite biomaterials because they
preparation of all of the solutions. All of the syntheses and prepa-
are generally soluble over the entire pH range, i.e., under acidic,
rations were performed at room temperature (25 ± 2 ◦ C) unless
neutral, alkaline and physiological conditions [24,25]. However,
otherwise specified. Potassium bromide (Sigma–Aldrich, USA,
from the bone tissue engineering perspective, polymers (such as
≥99%, KBr) suitable for spectroscopy was used to prepare the FTIR
chitosan and its derivatives) possess low mechanical properties,
pellets.
which are usually not suitable as biomaterials for cortical bone
implants [14]. Hence, ceramic-based biomaterials, such as calcium
phosphate compounds, are of great interest in the field of bone 2.2. Preparation of glycol-chitosan (GLY-CHI) films by solvent
tissue engineering for use as reinforcements of polymer-based cast and evaporation
composites. Hydroxyapatite (HA, Ca10 (PO4 )6 (OH)2 ) is considered
one of the most stable crystalline forms of calcium phosphate, and GLY-CHI solutions (1%, w/v) were prepared by dispersing the
it occurs as a major inorganic component in the bone (in the range polymer powder (0.5 g) in a 50 mL aqueous solution of phospho-
from 60 to 65%) [14]. Consequently, HA has emerged as an impor- ric acid (0.6% v/v). The mixture was placed under constant stirring
tant compound for artificial bone preparation because it stimulates for 24 h until complete solubilization occurred (pH ∼ 2.1). Then,
osteoconduction as it is gradually replaced by the host bone after the solutions were poured into plastic molds (polyethylene round
implantation for orthopedic replacements, especially in treatments plate, diameter = 65 mm) and were allowed to dry for 96 h at room
for bone regeneration and dental implants [14,15,26]. Addition- temperature (Fig. 1S—Supplementary Material).
ally, when HA is combined with biopolymers such as chitosan
to form composites, it might be able to mimic several functions 2.3. Preparation of biocomposite membranes by the colloidal
of natural bone-related tissues [15,26]. The inclusion of inorganic process—solvent cast and evaporation
nanoparticles of calcium phosphates (CaP) into the biopolymer
matrix has several objectives, such as improving the mechanical The synthesis of nHA particles using glycol chitosan as the ligand
properties, biocompatibility, altering the degradation behavior and was adapted based on the procedure developed by our group [31].
incorporating topographic features at the nanoscale that mimic Briefly, GLY-CHI solutions (1%, w/v) were prepared by dispersing
the hybrid nanostructure of natural bone [27]. Current fabrica- the polymer powder (0.5 g) in a 40 mL aqueous solution of phospho-
tion processes can synthesize hydroxyapatite particles within the ric acid (0.75% v/v). These acidic polymer solutions were referred
nanometer range; however, they usually suffer from major aggre- to as “SOL 2”. Approximately 0.55 g of Ca(OH)2 powder was added
gation, agglomeration and heterogeneity, mainly in size, shape, and to 10 mL of deionized water and vigorously stirred for 15 min. This
surface electrostatic charge, which render them inappropriate for calcium suspension was referred to as “SUS 1”. In the sequence,
several biomedical applications [28]. Therefore, by fine-tuning the “SUS 1” was added slowly to “SOL 2”, leading to the immediate
morphological and physicochemical characteristics of hydroxyap- formation of the suspension mixture (“SUS 3”), which was mag-
atite nanoparticles by strict control of the reaction conditions, the netically stirred for 1 h. Next, the pH of “SUS 3” was measured and
overall properties of the produced biocomposites can be signifi- adjusted to 12.0 ± 0.2 with NH4 OH (1.0 mol L1 ), and the suspension
cantly enhanced [29–31]. Unexpectedly, although there are a few was continuously stirred for 24 h (Eq. (1)).
reports published using glycol-chitosan for cancer diagnosis and
treatment [24,32], pharmaceutical applications as drug nanocarri- 6H3 PO4 (aq) − biopolymer + 10Ca(OH)2 (aq)
 V.C. Dumont et al. / International Journal of Biological Macromolecules 93 (2016) 1465–1478 1467

Fig. 1. Representation of the chemical structure of glycol-chitosan macromolecule.

NH4 OH
→ Ca10 (PO4 )6 (OH)2 (s) − biopolymer + 18H2 O(l) (1) analyze the micro-CT datasets in 2D and 3D for morphometry and
densitometry. CTVol software (v.2.3.1.0, Bruker micro-CT) was used
for 3D visualization of the composites.
Then, the polymer/CaP suspensions were poured into plas-
tic molds (polyethylene round plate, diameter = 65 mm) and were
allowed to dry for 96 h at room temperature. The composites pro- 2.4.2. Wavelength dispersive X-ray fluorescence spectroscopy
duced based on nanohydroxyapatite (nHA) were referred to as (WD-XRF)
nHA/GLY-CHI (Fig. 2S—Supplementary Material). The analysis of the inorganic composition of the nHA/GLY-CHI
composite membranes was conducted using wavelength dispersive
X-ray fluorescence spectroscopy (WDXRF SuperMini 200 Rigaku
2.4. Characterization of nHA/GLY-CHI composite membranes
spectrometer) with 50 kV voltage, 4 mA current and 200 W power
at approximately 20 ◦ C and controlled humidity for the elemental
2.4.1. Morphological analysis
analysis (element from F to U).
2.4.1.1. Light microscopy (LM) analysis. Representative samples of
polymer films and composites were examined by light microscopy
(Light microscope Stemi 2000-C) coupled to a Media Cybernetics 2.4.3. X-ray diffraction (XRD)
(PL-A662) camera and using Image-Pro Express imaging software. The crystallinity of the phases present in the biocomposites
was determined by X-ray diffraction (XRD) patterns that were
2.4.1.2. Scanning electron microscopy (SEM) and energy dispersion recorded using a PANalytical X’Pert diffractometer (Cu-K␣ radi-
X-ray spectroscopy (EDX) analysis. The morphology of the films, ation with ␭ = 1.5406 Å). Measurements were obtained in the 2␪
composites, and calcium phosphate particles was evaluated using range of 15–75◦ with steps of 0.06◦ .
scanning electron microscopy (SEM, FEI, model INSPECTTM S50)
coupled with energy dispersion X-ray spectroscopy (EDX, EDAX 2.4.4. Fourier transform infrared spectroscopy (FTIR)
GENESIS). Before examination, the samples were coated with a Calcium phosphate particles were analyzed using the dif-
thin carbon film by sputtering using a low deposition rate. Then, fuse reflectance infrared Fourier transform spectroscopy method
the substrate was cooled, and the maximum distance between tar- (DRIFTS, Nicolet 6700, Thermo-Fischer) over the range of
get and sample was used to avoid sample damage. SEM images 400–4000 cm−1 using 64 scans and a 2 cm−1 resolution with the
were obtained using secondary electrons (SE) at an accelerating subtraction of the KBr background. The samples were mixed in
voltage of 15 kV. The nHA particle size and size distribution data a ratio of 1% (wt.%) to KBr powder dried at 110 ± 5 ◦ C for 2 h.
were obtained based on the SEM images by measuring at least The FTIR spectra of the GLY-CHI films and the nHA/GLY-CHI bio-
200 randomly selected nanoparticles using image-processing pub- composite were obtained using attenuated total reflectance (ATR,
lic domain software (ImageJ, version 1.49, National Institutes of 4000–675 cm−1 using 32 scans and a 4 cm−1 resolution) with back-
Health). ground subtraction.

2.4.1.3. Transmission electron microscopy (TEM) analysis. Nanos- 2.5. Cytotoxicity assays
tructural characterizations of the composites were performed using
a Tecnai G2-20-FEI transmission electron microscope (TEM) based All of the biological tests were conducted according to ISO
on the high-resolution images and also selected area electron 10993-5:2009 (Biological evaluation of medical devices: Tests for
diffraction patterns (SAED) at an accelerating voltage of 200 kV. in vitro cytotoxicity). Cytotoxicity can be assessed by two methods,
The TEM samples were prepared by dropping an aliquot of the referred to as sample extract and direct contact. Both methods are
dispersion of “SUS 3” in ethanol (1.0 mL “SUS 3”: 4 mL ethanol mag- equivalent and globally accepted for the assessment of cytotoxic-
netically stirred for 5 h) onto a holey carbon grid before the analysis. ity of materials for biomedical applications, the choice is essentially
associated with the sample stability at the medium. For instance,
2.4.1.4. X-ray microtomography (-CT) analysis. The three dimen- glycol-chitosan membranes are water-soluble over the entire pH
sional structures of the biocomposite membranes were investi- range, therefore, it is not possible to evaluate cytotoxicity by direct
gated on a 3D microtomographer (SkyScan 1174, Bruker micro-CT) contact only by extract method. Hence, in this study, the samples
at a 12.18 ␮m resolution using 40 kV voltage and 100 ␮A current, were prepared for performing assays using both methods depend-
with a 0.7◦ rotation step and no filter. Images were reconstructed ing on the cell line and/or sample type, which was specifically noted
using NRecon reconstruction software (v.1.6.1.18, Bruker micro- for each bioassay. Ultraviolet irradiation for 30 min on each side of
CT). CTAn software (v.1.15.4.0, Bruker micro-CT) was used to the samples was used as the sterilization method.
1468 V.C. Dumont et al. / International Journal of Biological Macromolecules 93 (2016) 1465–1478

Fig. 2. Morphological analysis of GLY-CHI ((A) digital image and (C) LM image) and nHA/GLY-CHI composite membranes ((B) digital image and (C) LM image and (E) SEM
image). Element chemical analysis: EDX spectra (F) and element (Ca K␣) image mapping (G).

80 74 ± 15 nm
(A) 80
335 ± 70 nm
(B) 80
220 ± 50 nm (C) 70
70 70
60
60 60
Frequency (%)
Frequency (%)

Frequency (%)

50
50 50

40 40
40

30 30 30

20 20 20

10 10 10

0 0 0
0 50 100 150 200 250 300 350 400 450 500 550 600 0 50 100 150 200 250 300 350 400 450 500 550 600 0 50 100 150 200 250 300 350 400 450 500 550 600
Diameter (nm) Diameter (nm) Diameter (nm)

450
(D)
400

350
Nanoparticle Diameter (nm)

300
78%
250

200

150 66%

100

50

0
nHA nHA/CHI nHA/GLY-CHI

Fig. 3. Size distribution obtained using SEM images for CaP without a ligand (A), nHA/CHI (B), and nHA/GLY-CHI (C) composite membranes. (D) Histogram of nHA average
size for the systems under evaluation.
 V.C. Dumont et al. / International Journal of Biological Macromolecules 93 (2016) 1465–1478 1469

Fig. 4. TEM image (a) and SAED pattern (b) of nHA/GLY-CHI composite.

Fig. 5. Typical 3D micro-CT image of nHA/GLY-CHI composite membranes (A). Phase distribution along the thickness of the membrane (B) and in a frontal view (membrane
as produced (a) and membrane without organic-rich phase (b)). Gray and orange colors are associated with organic-rich and inorganic-rich phases, respectively. (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Protocol of Extract method: Extraction procedures were based istry, UFMG. The cells were cultured in DMEM (Dulbecco’s modified
upon the standard ISO 10993-12:1996 (Biological evaluation of eagle medium) with 10% fetal bovine serum (FBS), streptomycin
medical devices, part 7: Preparation of extracts of test materials). sulfate (10 mg mL−1 ), penicillin G sodium (10 units mL−1 ), and
Briefly, the extraction was performed in clean, chemically inert amphotericin-b (0.025 mg mL−1 ), supplied by Gibco BRL (NY, USA),
containers, under static conditions, and preventing contamination. using a humidified atmosphere of 5% CO2 at 37 ◦ C. The cells were
Each sample at the concentration of 3.0 cm2 mL−1 (i.e., surface- used for experiments on passage 23. SAOS cells are used more
area-to-volume ratio) was maintained in contact with DMEM frequently in basic and applied biology research than primary bone-
medium containing 10% FBS in a humidified atmosphere of 5% CO2 derived cells because of their ease of access and repeatability of
at 37 ◦ C for 24 h. After that, the medium extraction was performed results in experiments. In order to have a better prediction on cyto-
and used for the evaluation of the cytotoxicity of the nanomaterials compatibility of tested samples, it is generally preferred to use cell
for each bioassay. lines with similar characteristics and phenotypes to bone tissues.
Additionally, the SAOS cells are broadly accepted as a permanent
line of human osteoblast-like cells because they possess several
2.5.1. Cell cultures osteoblastic features, they are highly proliferative and behave as a
2.5.1.1. Human sarcoma cell line culture (SAOS cells). The immor- source of bone-related molecules. However, it is not always possi-
talized human osteosarcoma-derived (SAOS) cells were provided
by Prof. A. Goes of the Department of Immunology and Biochem-
1470 V.C. Dumont et al. / International Journal of Biological Macromolecules 93 (2016) 1465–1478

ble to directly extrapolate all the effects of osteosarcoma-derived of blue 5,5 -dibromo-4,4 -dichloro indigo (BCI) and the reduced
cell culture with those of bone cell cultures. diformazan (NBT-DF) [35].
ALP using extract method: SAOS cells and HEK293T cells were
2.5.1.2. Kidney cell line of a human embryo culture (HEK293T cells) plated (3 × 105 cells/well) in a 96-well plate. Cell populations were
and human bone marrow mesenchymal stem cells (HBMS). The kid- synchronized in serum-free medium for 24 h, and after this period,
ney cell line of a human embryo (HEK293T) and Human bone the medium was aspirated and replaced with medium containing
marrow mesenchymal stem cells (HBMS) were kindly provided by extracts (prepared according to the standard ISO 10993-12:1996,
Prof. M.F. Leite of the Department of Physiology and Biophysics, as detailed in the protocol) of GLY-CHI and nHA/GLY-CHI samples
UFMG. The cells were cultured in DMEM with 10% FBS, penicillin for 24 h. Controls had been used with cells and DMEM (Dulbecco’s
G sodium (10 units mL−1 ), streptomycin sulfate (10 mg mL−1 ), and modified eagle medium) with 10% FBS, positive control Triton x-
amphotericin-b (0.025 mg mL−1 ), all from Gibco BRL (NY, USA), in 100 (1% v/v in PBS, Sigma–Aldrich, MO, USA) and chips sterile
a humidified atmosphere of 5% CO2 at 37 ◦ C. HEK293T cells were polypropylene Eppendorf as negative control (1.0 mg mL−1 , Eppen-
used for experiments on passage 12 and HBMS cells on passage 4. dorf, Hamburg, Germany). After 24 h, 100 ␮L of BCIP-NBT (Life
Technologies of Brazil Ltd., São Paulo) prepared according to the
manufacturer’s protocol was added to each well and incubated for
2.5.2. MTT assay using 3-(4,5-dimethylthiazol-2yl)2,5-diphenyl
2 h in an oven at 37 ◦ C and 5% CO2 . Images using an inverted optical
tetrazolium bromide
microscope (Leica DMIL LED, Germany) were captured and 40 ␮L
MTT using extract method: SAOS cells and HEK293T cells were
of the SDS solution/4% HCL were added, followed with the incu-
plated (3 × 105 cells/well) in a 96-well plate. Cell populations were
bation for 16 h in an oven at 37 ◦ C and 5% CO2 . Then, 100 ␮L were
synchronized in serum-free medium for 24 h, and after this period,
removed from each well and transferred to a 96-well plate and
the medium was aspirated and replaced with the medium extract
the absorbance was recorded at 595-nm filter (i-Mark, Bio-rad).
(prepared according to the standard ISO 10993-12:1996, as detailed
The values obtained were expressed as percentage ALP activity
in the protocol) of GLY-CHI and nHA/GLY-CHI samples for 24 h.
according to Eq. (3).
Controls were used with the cells and DMEM with 10% FBS,
the positive control Triton x-100 (1% v/v in phosphate buffered Abs of sample and cells
ALP activity (%) = × 100% (3)
saline, PBS, Gibco BRL, NY, USA) and, as a negative control, chips Abs of control
of sterile polypropylene Eppendorf tubes (1 mg mL−1 , Eppendorf,
Hamburg, Germany). After 24 h, the medium was aspirated and 2.5.4. LIVE/DEAD® assay
replaced with 60 ␮L of culture medium with serum in each well LIVE/DEAD using extract method: SAOS cells were plated
and photographed using an inverted optical microscope (Leica (3 × 105 cells/well) in a 96-well plate. Cell populations were syn-
DMIL LED, Germany). Next, 50 ␮L of MTT medium (5 mg mL−1 ) chronized in serum-free medium for 24 h and, after this period,
(Sigma–Aldrich, MO, USA) was added to each well and was incu- the medium was aspirated and replaced with medium containing
bated for 4 h in an oven at 37 ◦ C and 5% CO2 . Subsequently, 40 ␮L extracts (prepared according to the standard ISO 10993-12:1996,
of the SDS solution/4% HCL was placed in each well and incubated as detailed in the protocol) of GLY-CHI and nHA/GLY-CHI sam-
for 16 h in an oven at 37 ◦ C and 5% CO2 . Then, 100 ␮L was removed ples for 24 h. Reference controls were used with cells and DMEM
from each well and transferred to a 96-well plate to quantify the medium with 10% FBS. After 24 h, the medium was aspirated
absorbance (Abs) using a microplate absorbance reader (i-Mark, and washed two times with 100 ␮L of phosphate buffered saline
Bio-Rad) with a 595-nm filter. The values obtained were expressed (PBS, Gibco BRL, NY, USA). The SAOS cells were treated for 30 min
as the percentage of viable cells according to Eq. (2). with a LIVE/DEAD viability/cytotoxicity kit (Life Technologies of
Brazil Ltd., São Paulo) according to the manufacturer’s specifica-
Abs of sample and cells
Cell viability = × 100% (2) tions. Images were obtained with an inverted-optical microscope
Abs of control
(Leica, model DMIL-LED, Germany), and the fluorescence was
MTT using direct contact method: HBMS cells on passage 4 were acquired separately using calcein at ␭ = 530 ± 12.5 nm and EthD-1
synchronized in medium serum free for 24 h. After this period, the at ␭ = 645 ± 20 nm.
cells were trypsinized and seeded (1 × 104 cells/well) on square LIVE/DEAD® using direct contact method: HBMS cells on pas-
samples of nHA/GLY-CHI (4.0 × 4.0 mm2 ) with average thickness sage 4 were synchronized in medium serum free for 24 h. After this
of 213.0 ± 1.0 ␮m in 96-well plates. Controls were prepared as period the cells were trypsinized and seeded (1 × 104 cells/well)
described in the previous section. After 48 h, all medium was aspi- on square samples of nHA/GLY-CHI (4.0 × 4.0 mm2 ) with aver-
rated and replaced with 60 ␮L of culture medium with serum age thickness of 213.0 ± 1.0 ␮m in 96-well plates. Controls were
to each well and photographed using an inverted optical micro- prepared as described in the previous section. After 48 h, the
scope (Leica DMIL LED, Germany). Then, 50 ␮L of MTT (5 mg mL−1 , medium was aspirated and washed two times with 100 ␮L of PBS
Sigma–Aldrich, St. Louis, MO, USA) were added to each well and (Gibco BRL, NY, USA). The HBMS cells were treated for 30 min
incubated for 4 h in an oven at 37 ◦ C and 5% CO2 . Subsequently, with a LIVE/DEAD® viability/cytotoxicity kit (Life Technologies of
40 ␮L of SDS solution/4% HCL was placed in each well and incu- Brazil Ltda., São Paulo) according to the manufacturer’s specifica-
bated for 16 h in an oven at 37 ◦ C and 5% CO2 . Then, 100 ␮L was tions. Images were obtained with an inverted-optical microscope
removed from each well and transferred to a 96-well plate to quan- (Leica, model DMIL-LED, Germany), and the fluorescence was
tify the absorbance using a microplate absorbance reader (Thermo acquired separately using calcein at ␭ = 530 ± 12.5 nm and EthD-1
Plate, TP-reader) with a 595-nm filter. The values obtained were at ␭ = 645 ± 20 nm
expressed as the percentage of viable cells according to Eq. (2).
2.5.5. Evaluation of osteogenic differentiation
2.5.3. Alkaline phosphatase activity (ALP) assay 2.5.5.1. MTT assay. MTT using direct contact method: HBMS
ALP activity was performed biochemically using a combina- cells on passage 4 were synchronized in medium serum free
tion of 5-bromo-4-chloro-3-indolyl phosphate (BCIP) and nitroblue for 24 h. After this period, the cells were trypsinized and
tetrazolium (NBT) referred to as BCIP/NBT. This assay is based on the seeded (1 × 103 cells/well) on square samples of nHA/GLY-CHI
principle that BCIP is hydrolyzed by ALP resulting in the leucoindigo (4.0 × 4.0 mm2 ) with average thickness of 213.0 ± 1.0 ␮m in 96-
that is oxidized by NBT producing an insoluble dark purple mixture well plates. After 1 day, the medium was aspirated and replaced
 V.C. Dumont et al. / International Journal of Biological Macromolecules 93 (2016) 1465–1478 1471

HA (112) 1.2 -PO3-

Absorbance (a.u.)
HA (211) 4
1030
1.0

β-TCP (221)
-C-OH

β-TCP (220)
HA (300)
3-
0.8 -C-O-C- -PO4

β-TCP (214)
965 -C-O-C-
0.6 -C-O-C-

HA (004)
HA (213)
HA (130)
HA (002)

HA (210)
saccharide

HA (222)
Amide Amide saccharide

HA (321)
HA (132)
0.4

HA (402)
1160 880
-C=O -NH
(a) 0.2 1645 1545
(c)
Intensity (a.u.)

0.0

HA (202) 1.2 H3PO4

Absorbance (a.u.)
1.0 -C-OH 971
-C-O-C- -C-O-C-
(b) 0.8
-C-O-C- saccharide
0.6 saccharide 880
+ 1160
0.4 -NH3
(c) 1630 1527
0.2 (b)
0.0
20 30 40 50 60
1.2 3-
2 θ (grades) -PO4

Absorbance (a.u.)
1.0 1035
3-
-PO4
Fig. 6. XRD patterns of uncapped CaP (a), GLY-CHI film (pH ∼ 6.0) (b), and nHA/GLY- 0.8 1098
CHI composite membranes (c).
0.6 3-
2- -PO4
H2O -CO3
0.4 961
1640
0.2 (a)
0.0
2000 1800 1600 1400 1200 1000 800
-1
Wavenumber (cm )

Fig. 8. FTIR spectra of uncapped CaP (a) GLY-CHI ligand as supplied (b), H3 PO4 -
biopolymer (GLY-CHI, pH ∼ 2.1) (c), and nHA/GLY-CHI composite membranes (d).

samples of nHA/GLY-CHI (4.0 × 4.0 mm2 ) with average thickness

of 213.0 ± 1.0 ␮m in 96-well plates. Controls were prepared as
described in the previous section. After 7, 14 and 21days, the
supernatant of each well was removed and treated as previously
described in Section 2.5.3.

2.5.5.3. Cell adhesion and spreading analysis by SEM. The morphol-

ogy of the adhesion and cells proliferation of the nHA/GLY-CHI was
evaluated using scanning electron microscopy (SEM, FEI, model
Fig. 7. FTIR spectrum of GLY-CHI as supplied. Inset: Glycol chitosan chemical struc- INSPECTTM S50) based in a procedure reported by our group [37].
ture with major chemical functions (circles). Before examination, samples were fixed in 2% of glutaraldehyde in
water for 16 h at 37 ◦ C. After that, the samples were washed three
by the osteogenic differentiation medium that was prepared with times with ice-cold PBS (6 ± 2) ◦ C. Then, they were dried by immer-
DMEM with high glucose containing 10% FBS, 50 mg mL−1 of ascor- sion in solutions of crescent concentration of ethanol/water (20%,
bic acid, 10 mM of ß-glycerol-phosphate all from Sigma–Aldrich (St. 50%,90%, 100% v/v) and finally vacuum dried in a desiccator for
Louis, MO, USA) and 0.1 mM of dexamethasone (Aché, SP, Brazil), 24 h. Prior to SEM analysis, the samples were made conductive by
which was changed every 3 days [36]. Controls had been used sputtering with an ultra-thin carbon coating at very low deposition
with cells and osteogenic differentiation medium, positive con- rate.
trol Triton x-100 (1% v/v in PBS, Sigma–Aldrich, St. Louis, MO, USA)
and, as the negative control, chips of sterile polypropylene Eppen-
dorf (1 mg mL−1 , Eppendorf, Hamburg, Germany). After 7, 14 and
21days, the supernatant of each well was removed and treated as 2.5.6. Statistical analysis
previously described in Section 2.5.2. Prism software (GraphPad Software, San Diego, CA, USA)
was used for all data analysis. The statistical significance was
2.5.5.2. ALP assay. ALP using direct contact method: HBMS cells on tested using one-way ANOVA followed by the Bonferroni test. P-
passage 4 were synchronized in serum free medium for 24 h. After values < 0.05 were considered of statistical significance. All of the
this period, the cells were seeded (1 × 104 cells/well) on square experiments were performed with triplicate (n = 3).
1472 V.C. Dumont et al. / International Journal of Biological Macromolecules 93 (2016) 1465–1478

3. Results and discussion which only the inorganic core of nHA can be observed. In addi-
tion, the electron diffraction pattern is presented in Fig. 4B with an
3.1. Characterization of nHA/GLY-CHI composite membranes interplanar distance of approximately 0.27 ± 0.01 nm, which can be
assigned to the (211) plane of the hexagonal unit cell of nanocrys-
3.1.1. Morphological analysis talline HA (JCPDS 86-1203).
The morphological evaluation of GLY-CHI (Fig. 2A and C ) and the X-ray microtomography (micro-CT) is a non-destructive tech-
nHA/GLY-CHI biocomposite (Figs. 2B, 2D and 2E) membranes was nique that resolves the three-dimensional structure of materials
based on observations performed on the set of images obtained revealing its microstructural features (phases, inclusions, cracks,
using a digital camera, light microscopy (LM) and SEM analyses pores). In this sense, micro-CT images allows to obtain the dis-
associated with the EDX spectrum (Fig. 2F) and image mapping tribution of the phases in the biocomposite membrane once the
(Fig. 2G) of the composite. It can be observed that the composite attenuation coefficient of a phase along the X-ray beam is depen-
systems formed homogeneous membranes with no noticeable het- dent on the local composition, mostly on the density and the
erogeneity at the magnifications used. Based on the LM images, atomic number, which are very different for polymers and ceram-
there were distinct morphological aspects of the nHA/GLY-CHI ics [40]. The typical 3D structure of the membrane composed of HA
composite membrane with significant surface roughness (Figs. 2B nanoparticles and GLY-CHI polymer collected by the micro-CT is
and 2D) and the presence of a white-yellowish opaque membrane shown in Fig. 5A indicating a relative uniform texture. Regarding
due to the presence of nHA particles in the composite compared to the micro-CT analysis, it can be considered that the biocompos-
with the smooth surface of the yellowish GLY-CHI transparent poly- ite membrane actually behaved as a 2D system, due to the relative
meric films (Fig. 2A and C). In addition, at a higher magnification thin thickness with minor attenuation of the X-ray beam (i.e., small
using SEM analysis (Fig. 2E), it is possible to observe that the HA par- path-length for the radiation). For evaluating the distribution of
ticles were at the nanoscale dimension and were equally dispersed phases, regions identified as organic-rich (lower attenuation, gray
in the polymer composite matrix without evidence of segrega- color) and inorganic-rich (higher attenuation, orange color) were
tion. A typical EDX spectrum with major signatures from Ca and separated using CTAn software and 3D models were built. Fig. 5B
P elements assigned to nHA nanoparticles immersed in the poly- allows for the visualization of the organic-rich phase at the surfaces
mer matrix is shown in Fig. 2F. The uniformity of the composite of the membrane and inorganic-rich phase in the middle of the bio-
was also confirmed by the homogenous distribution of the sig- composite along the thickness of the material. In addition, Fig. 5C
nal from Ca-K␣ captured by image mapping, as showed in Fig. 2G. shows a frontal view of the original membrane (a) and the same
The quantitative chemical analysis was obtained using WD-XRF. material without the organic-rich phase (b) obtained using CTVol
Ca and P concentrations in nHA were determined from the Ca-K␣ software revealing that the inorganic phase is homogeneously dis-
and P-K␣ line intensities and revealed an average molar ratio [Ca/P] persed in the matrix endorsing the results obtained in the image
of approximately 1.7, which is consistent with the theoretical HA mapping of Ca (Fig. 2G).
stoichiometric ratio.
Aiming at a more in-depth understanding of the effect of the
3.1.2. X-ray diffraction
presence of GLY-CHI ligands during the synthesis of nHA particles
The XRD patterns of the synthesized CaP and biocomposite
compared with chitosan (CHI, as a non-functionalized reference lig-
membranes are shown in Fig. 6. The uncapped calcium phosphate
and) and without polymer in the solution, SEM images were taken
particles (Fig. 6a) produced without the addition of polymer lig-
and analyzed using image processing software. The particle size
and during the synthesis presented the major (2 theta 31.7◦ (211),
distributions of these systems evaluated using over 200 random
32.8◦ (300), 32.2◦ (112), and 25.9◦ (002)) and minor peaks of HA
measurements are presented in Fig. 3. It can be observed that the
(International Centre for Diffraction Data, JCPDS 86-1203), as well
histograms related to the size distribution of nHA produced using
as three peaks associated with the ␤ tri-calcium phosphate (␤-TCP)
no ligand (nHA, Fig. 3A), chitosan (nHA/CHI, Fig. 3B), and GLY-CHI
phase (28.0◦ , 31.2◦ , and 34.5◦ ). The XRD pattern of nHA/GLY-CHI
(nHA/GLY-CHI, Fig. 3C) showed the average sizes of 335 ± 70 nm,
(Fig. 6c) revealed that the composite membrane presented broader
220 ± 50 nm, and 74 ± 15 nm, respectively. The major effect of the
peaks due to the presence of the amorphous polymer (Fig. 6b) and
presence of the glycol-functionalized chitosan is clearly observed
smaller nanocrystals of ceramic material with diffraction halos in
in the nucleation and growth of the HA particles, leading to a very
agreement with the major characteristic peaks of HA at approxi-
relevant decrease in the average size of nHA by approximately 78%
mately 31.7◦ , 32.2◦ , and 25.9◦ . No other CaP phases, such as ␣-TCP
compared to the case without ligands and by 66% compared with
or ␤-TCP, were detected in the biocomposite. Hence, the kinetics
chitosan (Fig. 3D). Here, glycol-modified chitosan ligands are likely
and thermodynamics of the nucleation and growth of nanoparticles
to stabilize the formed nHA inorganic cores by creating an organic
during the synthesis were affected by adding the polymer ligands
shell with the hydrophilic glycol moieties preferentially oriented
(GLY-CHI), altering the composition of calcium phosphates and the
toward water medium and providing steric hindrance that prevents
crystallinity of the biocomposite.
nanoparticle aggregation and agglomeration, which is different
from the electrostatic stabilization by carboxylic groups reported
for carboxymethyl chitosan (CMC) [22,23,31,38,39]. Moreover, the 3.1.3. Fourier transform infrared spectroscopy
formation of the hydrated glycol layer surrounding the nHA parti- The FTIR spectrum of GLY-CHI (as supplied, Fig. 7) revealed the
cles significantly reduces the diffusion of species in solution, thus presence of vibrations at 1660–1640 cm−1 associated with C O
preventing the growth of the nHA core. Consequently, the strategy stretching amide I and over the range of 1590–1545 cm−1 from
developed in this study offers the tools for fine-tuning the final –NH bending of primary amines (1590–1560 cm−1 ) and amide II
dimensions of ultra-small HA particles at nanoscale dimensions (1560–1545 cm−1 ). The bands in the range of 1150–1020 cm−1
embedded in the biopolymer matrices. contain the contributions of the ether (C–O–C, 1150–1050 cm−1 ),
Additionally, the TEM analysis of these systems permitted the primary alcohol (1050–1030 cm−1 ), and secondary alcohol
observation of morphological features of the lowest portion of (∼1100–1070 cm−1 ) from chitosan and ethylene glycol groups
the HA nanoparticle size distribution at a very high resolution, [41–43]. In addition, the bands at 1450–1380 cm−1 are related to
with dimensions typically below 20 nm. A representative image –CH bending, and the vibrations at 1160 cm−1 and 890 cm−1 are
of nHA/GLY-CHI nanoparticle conjugates is shown in Fig. 4A in assigned to the vibrations of the C–O–C saccharide structure [41].
Thus, the FTIR analysis confirmed the glycol chitosan compound
 V.C. Dumont et al. / International Journal of Biological Macromolecules 93 (2016) 1465–1478 1473

Fig. 9. SAOS (A) and HEK293T (B) cells viability response using MTT assay after 24 h of incubation in contact with the extracts of GLY-CHI and nHA/GLY-CHI. SAOS cells
morphology in the control (C) and nHA/GLY-CHI (D) (200×). HBMS cells viability response by MTT assay after 48 h incubation in direct contact with the nHA/GLY-CHI (E).

Fig. 10. ALP activity assay after 24 h incubation with SAOS cell line (A) and HEK293T cell line (B). Presence of the products of metabolism (BCI/NBT-DF) in control (C), GLY-CHI
sample (D), and nHA/GLY-CHI sample (E) after 24 h of contact with HEK293T cells (200×).

supplied with the presence of major chemical functionalities spectrum of HA also revealed the formation of carbonated hydrox-
identified, i.e., amines, amides, hydroxyls and glycols. yapatite due to the presence of CO3 2− bands at 1460–1415 cm−1
The uncapped CaP spectrum (Fig. 8a) shows the typical phos- (␯3 ) and 875 cm−1 (␯2 ), which are characteristic of anionic substi-
phate ␯3 PO4 3− bands at 1098 cm−1 and 1035 cm−1 and the ␯1 tution from phosphate groups by carbonates (B-type HA) [45,46]. A
PO4 3− vibration at 961 cm−1 , which supports the formation of HA band at approximately 1640 cm−1 (␦) was also present, which was
as the main phase in the calcium phosphate ceramic [44]. The FTIR associated with adsorbed water. The presence of these vibrational
1474 V.C. Dumont et al. / International Journal of Biological Macromolecules 93 (2016) 1465–1478

Fig. 11. The LIVE/DEAD® test with SAOS cells after 24 h of contact with extracts. Live cells ((A), green) and Dead cells ((B), red) in control, GLY-CHI and nHA/GLY CHI samples
(200×). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 12. The LIVE/DEAD® test with HBMS cells after 24 h of direct contact. Live cells ((A), green) and Dead cells ((B), red) in control and nHA/GLY-CHI sample (200×). (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

bands related to carbonates and water are typical of hydroxyap- is assigned to the asymmetric stretching (␯3 ) of the molecule. The
atites obtained by wet precipitation routes [47]. energy shift detected in this band is likely related to the formation
In the spectrum of the biopolymer-H3 PO4 film at pH ∼ 2.1 of interactions between the phosphate species and the biopoly-
(Fig. 8b), the antisymmetric and symmetric stretching vibrational mer once the ␯3 vibrations become sensitive to the coordination
modes of the protonated amine were observed at 1630 cm−1 and environment of phosphates [49].
1527 cm−1 , respectively [48]. These bands were expected once In the spectrum of the nanocomposite (Fig. 8c), the presence of
almost all of the amine groups of the polymer were protonated at HA bands at 1030 cm−1 (␯3 PO4 3− ) and 965 cm−1 (␯1 PO4 3− ) was
this pH conditions. A band at 971 cm−1 was also observed that was observed in addition to the characteristic bands of –C–O–C– and
associated with the H3 PO4 species. In an acidic medium at pH ≤ 2 C–OH the biopolymer. The comparison between the spectrum of
(pK1 = 2.20), the phosphoric acid species (H3 PO4 ) are dominant and the nanocomposite (Fig. 8c) and the polymer (Fig. 7) showed the
characterized by the main absorption peak at ∼1006 cm−1 , which disappearance of the band at 1590–1570 cm−1 assigned to the –NH2
 V.C. Dumont et al. / International Journal of Biological Macromolecules 93 (2016) 1465–1478 1475

Fig. 13. HBMS cell viability response by MTT assay after 7, 14 and 21 days of incubation in direct contact with the nHA/GLY-CHI samples (A). HBMS cells morphology in
control (B) and nHA/GLY-CHI (C) samples after 21 days of contact (100×).

bending, which can be associated with the formation of coordina- of the biocomposites are considered very promising for potential
tion complexes between these primary amino groups of GLY-CHI bone-related biomedical applications.
and the Ca2+ from HA possibly through dative bonds [31,50–54].
3.2.2. ALP activity assay
Alkaline phosphatase is an important enzyme involved in bone
3.2. Cell toxicity assay formation and mineralization and is a well-known and widely used
a marker of osteoblast differentiation for in vitro tests [52,53]. In
3.2.1. MTT assay this study, the ALP activity of SAOS and HEK293T cells using the
In the current study, the viability responses of SAOS and extract method with GLY-CHI and nHA/GLY-CHI samples were ana-
HEK293T cells with the extracts of GLY-CHI and nHA/GLY-CHI sam- lyzed and the results are presented in Fig. 10A (SAOS cells) and
ples were analyzed using the MTT assay, which is specifically used Fig. 10B (HEK 293). As it can be observed, there was no signifi-
to assess mitochondrial function and cell viability. When analyzing cant difference in ALP activity of the biohybrids when compared
the response of the SAOS cells tested, no significant difference in to the control group. The images obtained from the products NBT-
viability was detected compared with the control group (Fig. 9A). DF and BCI visualized by light microscopy are shown in Fig. 10C
Fig. 9C shows the image of the SAOS cells of the controls and (HEK293T—control), Fig. 10D (HEK293T—GLY-CHI), and Fig. 10E
nHA/GLY-CHI (Fig. 9D) samples after 24 h. In addition, regarding to (HEK293T—nHA/GLY-CHI) endorsing the similar behavior mea-
the confluence of these cells, no relevant difference was observed sured for all the samples. These findings suggest that extracts of
in all of the samples as compared with the control group. Simi- GLY-CHI and nHA/GLY-CHI samples did not interfere in the pro-
larly, no significant difference in the HEK293T cell viability results duction of ALP in either cell types after 24 h of incubation.
was observed compared with the control group (Fig. 9B). Therefore,
these results demonstrated that the novel biocomposites nHA/GLY- 3.2.3. LIVE/DEAD® assay
CHI were not cytotoxic toward SAOS and HEK293T cell types, which The LIVE/DEAD® viability/cytotoxicity assay provides a two-
are model cell-lines (permanent) for the evaluation of potential color fluorescence cell viability test, which is essentially based
cytotoxicity of materials for biomedical applications. To have more on the simultaneous determination of live (stained in green, ␭ex
specific evidence of the cytocompatibility of the biocomposites, 495 nm, ␭em 515 nm) and dead cells (stained in red, ␭ex 495 nm,
further MTT assay was performed using HBMS cell line in direct ␭em 635 nm) with two probes, Calcein AM (acetoxymethyl ester of
contact with the samples, with the results presented in Fig. 9E. It calcein) and EthD-1 (ethidium homodimer), that measure known
can be clearly observed an equivalent response of the HBMS cell parameters of cell viability, the intracellular esterase activity and
viability with no detected cytotoxicity toward the biocomposites. plasma membrane integrity [54,55]. Therefore, LIVE/DEAD® cyto-
Moreover, as mesenchymal stem cells are a critical component of toxicity assay was performed to validate the results of the MTT
the bone-healing process as they represent the precursors for both and ALP tests using SAOS cells (human osteoblast-like cells) of the
osteoblasts and chondrocytes, these preliminary in vitro results extracts of GLY-CHI and nHA/GLY-CHI. In Fig. 11 it is noted that the
1476 V.C. Dumont et al. / International Journal of Biological Macromolecules 93 (2016) 1465–1478

Fig. 14. Measurements of ALP activity of HBMS cells after 7, 14, and 21 days of incubation in direct contact with the nHA/GLY-CHI (A). HBMS cells morphology in control (B)
and nHA/GLY-CHI (C) samples after 21 days of contact (100×).

Fig. 15. Typical representative SEM image of the biocomposite-HBMS cell biointerfaces (Secondary Electrons—SE at 5,000×) indicating cell adhesion, spreading, and
proliferation (inset drawing). White arrow: cytoplasmic extensions of HBMS cell anchored to the biocomposite.

cells tested with the extracts of samples GLY- CHI and nHA/GLY- marized in Fig. 12. It can be observed that the HBMS cells in contact
CHI had a fluorescence pattern similar to the control group, i.e., with the nHA/GLY-CHI sample had a fluorescence pattern similar
high green fluorescence (viable cells) and little or no red fluores- to the control group, i.e., high green fluorescence (viable cells) and
cence (dead cells). Thus, as expected, the LIVE/DEAD test confirmed little or no red fluorescence (dead cells). Moreover, it is essential
the previous results showed in the cell viability assays by MTT and to highlight that no dissimilarity was observed on the cell via-
ALP. bility responses when comparing the results using extracts from
Additionally, the non-toxic behavior of the new biohybrids samples and direct contact methods. Thus, based on the three
developed in this study was performed the LIVE/DEAD® cytotox- assays (MTT, ALP and LIVE-DEAD® ), using extracts and direct con-
icity assay using direct contact of the samples with human bone tact methods, and three cell cultures, the overall results confirmed
marrow mesenchymal stem cells (HBMS), and the results are sum- qualitatively and quantitatively the cytocompatibility of the nHA-
 V.C. Dumont et al. / International Journal of Biological Macromolecules 93 (2016) 1465–1478 1477

GLY-CHI biohybrids produced. In that sense, it may be suggested orthopedic surgery, particularly where mechanical strength is not
that the GLY-CHI and nHA/GLY-CHI systems have shown promising crucial for the performance [3,15]. Nonetheless, it should be men-
potential to find applications as biomaterials for tissue engineering. tioned that further in vitro and in vivo studies are needed before
recommending these novel biohybrids for clinical applications.
3.3. Evaluation of osteogenic differentiation
4. Conclusion
Stem cells have three stages to differentiate into osteoblasts;
i.e., cell proliferation, development of mineralized matrix, and This study demonstrates for the first time that the GLY-CHI
mineral deposition. The cell proliferation and differentiation are ligands used during the synthesis of HA nanoparticles markedly
interconnected characteristics; therefore, the proliferation poten- affected the final average size by modulating the nucleation and
tial decreases slowly as the cells begin the process of differentiation growth kinetics of the process compared to chitosan. The GLY-
[36]. In the current study, to investigate the osteogenic differenti- CHI-based biocomposites presented the smallest nHA particle
ation in HBMS cells in direct contact with nHA/GLY-CHI samples, sizes (74 ± 15 nm) and the system without ligand in the synthe-
we evaluate cell proliferation through the mitochondrial viability sis showed the largest average size (335 ± 70 nm). As expected,
by MTT and ALP tests. The MTT tests of the HBMS cells (after 48 h) the system with chitosan ligands presented particles with inter-
showed no significant difference in the viability results after 7, 14, mediate dimensions (220 ± 50 nm). Moreover, the results showed
and 21 days of incubation with the samples compared to the control that these aminopolysaccharide ligands have driven to the for-
group (Fig. 13A). Fig. 13 (bottom) shows the images of HBMS cells mation of uniform biocomposites composed of CHI or GLY-CHI
in the control condition (B) and in contact with nHA/GLY-CHI (C) polymer with nHA homogeneously embedded and distributed in
samples, where the morphological aspects of the cells appear not the matrices. Furthermore, regarding the in vitro assays, no cyto-
to have changed. Assessment of cell proliferation and viability is a toxicity was observed for these nanobiocomposites evaluated using
valuable tool for analyzing cell cultures in contact with biomateri- SAOS, HEK293T and HBMS cells as preliminary models to mimic
als, however, alone does not allow comprehensive characterization the bone tissue response. Moreover, the results demonstrated that
of the cell-response. The osteoblastic differentiation step can be the nHA/GLY-CHI composites are osteoinductive for HBMS cells. In
determined by the activity of alkaline phosphatase enzyme, which summary, novel nHA-based biocomposites were effectively pro-
is required to generate inorganic phosphate from hydrolysis of duced at room temperature using an aqueous co-precipitation
extracellular inorganic pyrophosphate associated with the crystal- route via an entirely eco-friendly chemical process, which may be a
lization of hydroxyapatite. Moreover, it is an important marker of promising biomaterial suitable for cell ingrowth and osteoconduc-
differentiation of osteoblast cells [36,56]. Thus, ALP activity was tion in bone tissue repair and restauration.
measured during the seventh, fourteenth and twenty-first day of
direct contact of the nHA/GLY-CHI biocomposites with HBMS cells.
In Fig. 14A, it can be observed the increase of the ALP activity of Authors’ contributions
HBMS cells of approximately 35%, 45% and 55% after 7, 14, and 21
days, respectively, compared to the control group. This crescent H.S. Mansur designed the experimental procedure of the study,
gradual increase of alkaline phosphatase activity of HBMS cells with performed the analysis and drafted the manuscript. V. Dumont
the time of incubation with the biocomposites is very interesting as and A.A.P. Mansur performed the synthesis, characterization and
it demonstrates that the nHA/GLY-CHI samples are biocompatible analysis of the biocomposites and drafted the manuscript. S.M. Car-
and favored cell proliferation. The images of HBMS cells in the con- valho performed and analyzed the cytotoxicity assays and drafted
trol condition (Fig. 14B) and in contact with nHA/GLY-CHI (Fig. 14C) the manuscript. Nádia S.V. Capanema and B.R. Barrioni performed
indicated that the morphological aspects of the cells appear not to the micro-CT characterization and analysis of the composites. All
have changed. Moreover, the MTT assays after 7, 14, and 21 days, authors have read and approved the final version manuscript.
did not show increased mitochondrial activity of HBMS cells.
Therefore, combining the ALP and MTT results, it is suggested Conflicts of interest
that, from the seventh day, the HBMS cells may have reached a
critical cell density to initiate the cell differentiation stage and the The authors declare that they have no conflict of interest.
proliferation remained relatively stable until the twenty-first day
of analysis. This assumption is supported by the literature [56],
Acknowledgments
which suggested that stem cells need a critical number of cells to
begin differentiating and, after that, they stop dividing and the min-
The authors acknowledge financial support from the fol-
eralization step is initiated. In addition, some cells may undergo
lowing Brazilian research agencies: CAPES—Coordenação
apoptosis and, therefore, can maintain the appropriate number of
de Aperfeiçoamento de Pessoal de Nível Superior (PROEX-
cells during the stage of differentiation.
433/2010;PNPD); FAPEMIG—Fundação de Amparo à Pesquisa do
A typical SEM image of the interaction of the nHA/GLY-CHI bio-
Estado de Minas Gerais (PPM-00202-13; BCN-TEC 30030/12);
composites with HBMS cells is shown in Fig. 15. It can be observed
CNPq—Conselho Nacional de Pesquisa (PQ1B–306306/2014-0;
that the nanocomposite allowed the embryonic stem cells (HBMS)
UNIVERSAL-457537/2014-0); FINEP—Financiadora de Estudos
to settle on the surfaces, through cell adhesion and spreading. The
e Projetos (CTINFRA-PROINFRA 2008/2010). The authors also
morphology of the cells showed predominantly elongated sprawl-
express their gratitude to the staff of the Microscopy Centre/UFMG
ing on membranes emitting cytoplasmic processes that facilitate
for the TEM analysis, Prof D.B. Santos and P. Trigueiro for the SEM
adhesion and cell communication [37]. Hence, this result also
images.
demonstrates the biocompatibility of the HBMS cells toward the
biocomposites.
Therefore, it can be suggested that these biocomposites made Appendix A. Supplementary data
of nano-hydroxyapatite and glycol-chitosan offer promising per-
spectives to be used in a wide range of bioapplications, such as Supplementary data associated with this article can be found,
for periodontal treatment, as barrier membranes in guided tis- in the online version, at http://dx.doi.org/10.1016/j.ijbiomac.2016.
sue regeneration (GTR) and moldable biomaterials for dental and 04.030.
1478 V.C. Dumont et al. / International Journal of Biological Macromolecules 93 (2016) 1465–1478

References [28] T.N. Do, W.H. Lee, C.Y. Loo, A.V. Zavgorodniy, R. Rohanizadeh, Hydroxyapatite
nanoparticles as vectors for gene delivery, Ther. Delivery 3 (2012) 623–632.
[1] L. Pighinelli, M. Kucharska, Chitosan–hydroxyapatite composites, Carbohydr. [29] L. Kong, G. Yuan, L. Guangyuan, G. Yandao, Z. Nanming, Z. Xiufang, A study on
Polym. 93 (2013) 256–262. the bioactivity of chitosan/nano-hydroxyapatite composite scaffolds for bone
[2] S. Pina, J.M. Oliveira, R.L. Reis, Natural-based nanocomposites for bone tissue tissue engineering, Eur. Polym. J. 42 (2006) 3171–3179.
engineering and regenerative medicine: a review, Adv. Mater. 27 (2015) [30] S.V. Dorozhkin, Nanosized and nanocrystalline calcium orthophosphates, Acta
1143–1169. Biomater. 6 (2010) 715–734.
[3] A. Di Martino, M. Sittinger, M.V. Risbud, Chitosan: a versatile biopolymer for [31] V.C. Dumont, A.A.P. Mansur, S.M. de Carvalho, F.G.L.M. Borsagli, M.M. Pereira,
orthopaedic tissue-engineering, Biomaterials 26 (2005) 5983–5990. H.S. Mansur, Chitosan and carboxymethyl-chitosan capping ligands: effects
[4] K. Jahana, M. Tabrizian, Composite biopolymers for bone regeneration on the nucleation and growth of hydroxyapatite nanoparticles for producing
enhancement in bony defects, Biomater. Sci. 4 (2016) 25–39. biocomposite membranes, Mater. Sci. Eng. C 59 (2015) 265–277.
[5] A.R. Short, D. Koralla, A. Deshmukh, B. Wissel, B. Stocker, M. Calhoun, D. [32] S.J. Lee, H.S. Min, S.H. Ku, S. Son, I.C. Kwon, S.H. Kim, K. Kim, Tumor-targeting
Deand, J.O. Winter, Hydrogels that allow and facilitate bone repair glycol chitosan nanoparticles as a platform delivery carrier in cancer
remodeling, and regeneration, J. Mater. Chem. B 3 (2015) 7818–7830. diagnosis and therapy, Nanomedicine 9 (2014) 1697–1713.
[6] P. Wang, L. Zhao, J. Liu, M.D. Weir, X. Zhou, H.H.K. Xu, Bone tissue engineering [33] H.Y. Yoon, S. Son, S.J. Lee, D.G. You, J.Y. Yhee, J.H. Park, M. Swierczewska, S.
via nanostructured calcium phosphate biomaterials and stem cells, Bone Res. Lee, I.C. Kwon, S.H. Kim, K. Kim, M.G. Pomper, Glycol chitosan nanoparticles as
2 (2014) 14017. specialized cancer therapeutic vehicles: sequential delivery of doxorubicin
[7] C. Sanchez, H. Arribart, M.M.G. Guille, Biomimetism and bioinspiration as and Bcl-2 siRNA, Sci. Rep. 4 (2014) 6878.
tools for the design of innovative materials and systems, Nat. Mater. 4 (2005) [34] C.A. Holmes, M. Tabrizian, Substrate-mediated gene delivery from
277–288. glycol-chitosan/hyaluronic acid polyelectrolyte multilayer films, ACS Appl.
[8] M. Vallet-Regí, D.A. Navarrete, Biological apatites in bone and teeth, in: M. Mater. Interfaces 5 (2013) 524–531.
Vallet-Regí, D.A. Navarrete (Eds.), Clinical Use: From Materials to [35] L.A. Trinh, M.D. McCutchen, M. Bonner-Fraser, S.E. Fraser, L.A. Bumm, D.W.
Applications, The Royal Society of Chemistry, Cambridge, 2016, pp. 1–29. McCauley, Fluorescent in situ hybridization employing the conventional
[9] J. Lewandowska-Łańcucka, S. Fiejdasz, L. Rodzik, A. Łatkiewicz, M. NBT/BCIP chromogenic stain, Biotechniques 42 (2007) 756–759.
Nowakowska, Novel hybrid materials for preparation of bone tissue [36] W.A. Ribeiro Neto, A.C.C. de Paula, T.M.M. Martins, A.M. Goes, L. Averous, G.
engineering scaffolds, J. Mater. Sci. Mater. Med. 26 (2015) 231. Schlatter, R.E.S. Bretas, poly(Butylene
[10] X.N. Qi, Z.L. Mou, J. Zhang, Z.Q. Zhang, Preparation of chitosan/silk adipate-co-terephthalate)/hydroxyapatite composite structures for bone
fibroin/hydroxyapatite porous scaffold and its characteristics in comparison tissue recovery, Polym. Degrad. Stab. 120 (2015) 61–69.
to bi-component scaffolds, J. Biomed. Mater. Res. Part A 102 (2014) 366–372. [37] H.S. Costa, A.A.P. Mansur, E.F. Barbosa-Stancioli, M.M. Pereira, H.S. Mansur,
[11] F. Wang, Y.C. Zhang, H. Zhou, Y.C. Guo, X.X. Su, Evaluation of in vitro and Morphological mechanical, and biocompatibility characterization of
in vivo osteogenic differentiation of macroporous alumina scaffolds coated with calcium phosphate/PVA, J. Mater.
nano-hydroxyapatite/chitosan/poly(lactide-co-glycolide) scaffolds with Sci. 43 (2008) 510–524.
human umbilical cord mesenchymal stem cells, J. Biomed. Mater. Res. Part A [38] C. Fang, N. Bhattarai, C. Sun, M. Zhang, Functionalized nanoparticles with
102 (2014) 760–768. long-term stability in biological media, Small 5 (2009) 1637–1641.
[12] P. Gomez-Romero, Hybrid organic–inorganic materials-in search of synergic [39] R.A. Sperling, W.J. Parak, Surface modification, functionalization and
activity, Adv. Mater. 13 (2001) 163–174. bioconjugation of colloidal inorganic nanoparticles, Philos. Trans. R. Soc. A
[13] R.-M. Wang, S.-R. Zheng, Y.-P. Zheng, Introduction to polymer matrix 368 (2010) 1333–1383.
composites, in: R.-M. Wang, S.-R. Zheng, Y.-P. Zheng (Eds.), Polymer Matrix [40] E. Maire, J.Y. Buffière, L. Salvo, J.J. Blandin, W. Ludwig, J. Létang, On the
Composites and Technology, Woodhead Publishing Limited, Cambridge, 2011, application of X-ray microtomography in the field of materials science, Adv.
pp. 1–25. Eng. Mater. 3 (2001) 539–546.
[14] M. Swetha, K. Sahithi, A. Moorthi, N. Srinivasan, K. Ramasamy, N. [41] F.G.L.M. Borsagli, A.A.P. Mansur, P. Chagas, L.C.A. Oliveira, H.S. Mansur,
Selvamurugan, Biocomposites containing natural polymers and O-Carboxymethyl functionalization of chitosan: complexation and adsorption
hydroxyapatite for bone tissue engineering, Int. J. Biol. Macromol. 47 (2010) of Cd(II) and Cr(VI) as heavy metal pollutant ions, React. Funct. Polym. 97
1–4. (2015) 37–47.
[15] J. Venkatesan, S.-K. Kim, Chitosan composites for bone tissue engineering—an [42] Y. Arai, D.L. Sparks, ATR–FTIR spectroscopic investigation on phosphate
overview, Mar. Drugs 8 (2010) 2252–2266. adsorption mechanisms at the ferrihydrite–water interface, J. Colloid
[16] P. Solier, A. Denuzière, C. Viton, A. Domard, Relation between the degree of Interface Sci. 241 (2001) 317–326.
acetylation and the electrostatic properties of chitin and chitosan, [43] Z. Li, Y. Du, Z. Zhang, D. Pang, Preparation and characterization of CdS
Biomacromolecules 2 (2002) 765–772. quantum dots chitosan biocomposite, React. Funct. Polym. 55 (2003) 35–43.
[17] F.P. Ramanery, A.A.P. Mansur, F.G.L.M. Borsagli, H.S. Mansur, Green and facile [44] M.H. Santos, L.G.D. Heneine, H.S. Mansur, Synthesis and characterization of
synthesis of water-soluble ZnS quantum dots nanohybrids using chitosan calcium phosphate/collagen biocomposites doped with Zn2+ , Mater. Sci. Eng.
derivative ligands, J. Nanopart. Res. 16 (2014) 2504–2519. C 28 (2008) 563–571.
[18] H.S. Mansur, A.A.P. Mansur, E. Curti, M.V. de Almeida, [45] J.C. Elliot, Structure and Chemistry of the Apatites and Other Calcium
Functionalized-chitosan/quantum dot nano-hybrids for nanomedicine Phosphates, 1st ed., Elsevier, Amsterdam, 1994.
applications: towards biolabeling and biosorbing phosphate metabolites, J. [46] M.E. Fleet, X. Liu, Coupled substitution of type A and B carbonate in
Mater. Chem. B 1 (2013) 1696–1711. sodium-bearing apatite, Biomaterials 28 (2007) 916–926.
[19] H.S. Mansur, A.A.P. Mansur, E. Curti, M.V. de Almeida, Bioconjugation of [47] C.J. Liao, F.H. Lin, K.S. Chen, J.S. Sun, Thermal decomposition and reconstitution
quantum-dots with chitosan and N,N,N-trimethyl chitosan, Carbohydr. Polym. of hydroxyapatite in air atmosphere, Biomaterials 20 (1999) 1807–1813.
90 (2012) 189–196. [48] R. Rangel-Mendez, R. Monroy-Zepeda, E. Leyva-Ramos, P.E. Diaz-Flores, K.
[20] J.C.C. Santos, P.M.D. Moreno, A.A.P. Mansur, V. Leiro, H.S. Mansur, A.P. Pêgo, Shirai, Chitosan selectivity for removing cadmium(II), copper(II), and lead(II)
Functionalized chitosan derivatives as nonviral vectors: physicochemical from aqueous phase: pH and organic matter effect, J. Hazard. Mater. 162
properties of acylated N,N,N-trimethyl chitosan/oligonucleotide (2009) 503–511.
nanopolyplexes, Soft Matter 11 (2015) 8113–8125. [49] J.T. Bamgbose, S. Adewuyi, O. Bamgbose, A.A. Adetoye, Adsorption kinetics of
[21] M. Malhotra, C. Lane, C. Tomaro-Duchesneau, S. Saha, S. Prakash, A novel cadmium and lead by chitosan, Afr. J. Biotechnol. 9 (2010) 2560–2565.
method for synthesizing PEGylated chitosan nanoparticles: strategy, [50] M. Benavente, Adsorption of metallic ions onto chitosan: equilibrium and
preparation, and in vitro analysis, Int. J. Nanomed. 6 (2011) 485–494. kinetic studies, in: Licentiate Thesis, Royal Institute of Technology, Stockholm,
[22] A.A.P. Mansur, C.G. de Almeida, S.M. de Carvalho, L.V. de Faria, M.V. de 2008.
Almeida, H.S. Mansur, Cytocompatible fluorescent quantum dot/PEG-chitosan [51] A.J. Varma, S.V. Deshpande, J.F. Kennedy, Metal complexation by chitosan and
bioconjugates for nanomedicine applications, Eur. J. Inorg. Chem. 27 (2015) its derivatives: a review, Carbohydr. Polym. 55 (2004) 77–93.
4555–4564. [52] G.C. Reilly, S. Radin, A.T. Chen, P. Ducheyne, Differential alkaline phosphatase
[23] A.A.P. Mansur, H.S. Mansur, Quantum dot/glycol chitosan fluorescent responses of rat and human bone marrow derived mesenchymal stem cells to
nanoconjugates, Nanoscale Res. Lett. 10 (2015) 172. 45S5 bioactive glass, Biomaterials 28 (2007) 4091–4097.
[24] I.C. Sun, J.H. Na, S.Y. Jeong, D.E. Kim, I.C. Kwon, K. Choi, C.H. Ahn, K. Kim, [53] E. Verné, S. Ferraris, C. Vitale-Brovarone, S. Spriano, C.L. Bianchi, A. Naldoni, M.
Biocompatible glycol chitosan-coated gold nanoparticles for tumor-targeting Morra, C. Cassinelli, Alkaline phosphatase grafting on bioactive glasses and
CT Imaging, Pharm. Res. 31 (2014) 1418–1425. glass ceramics, Acta Biomater. 6 (2010) 229–240.
[25] M.M. Chen, Y.Q. Huang, H. Cao, Y. Liu, H. Guo, L.S. Chen, J.H. Wang, Q.Q. Zhang, [54] C.D.F. Moreira, S.M. Carvalho, H.S. Mansur, M.M. Pereira, Thermogelling
Collagen/chitosan film containing biotinylated glycol chitosan nanoparticles chitosan–collagen–bioactive glass nanoparticle hybrids as potential injectable
for localized drug delivery, Colloids Surf. B 128 (2015) 339–346. systems for tissue engineering, Mater. Sci. Eng. C 58 (2016) 1207–1216.
[26] S. Nath, A. Dey, A.K. Mukhopadhyay, B. Basu, Nanoindentation response of [55] J. Walker, S. Shadanbaz, T.B.F. Woodfield, M.P. Staiger, G.J. Dias, The in vitro
novel hydroxyapatite-mullite composites, Mater. Sci. Eng. A 513–514 (2009) and in vivo evaluation of the biocompatibility of Mg alloys, Biomed. Mater. 9
197–201. (2014) 015006.
[27] X. Cai, H. Tong, X. Shen, W. Chen, J. Yan, J. Hu, Preparation and [56] S. Kale, S. Biermann, C. Edwards, C. Tarnowski, M. Morris, M.W. Long,
characterization of homogeneous chitosan–polylactic acid/hydroxyapatite Three-dimensional cellular development is essential for ex vivo formation of
nanocomposite for bone tissue engineering and evaluation of its mechanical human bone, Nat. Biotechnol. 18 (2000) 954–958.
properties, Acta Biomater. 5 (2009) 2693–2703.