Materials Science and Engineering C 58 (2016) 521531

Contents lists available at ScienceDirect

Materials Science and Engineering C

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

Electrospinning of PLGA/gum tragacanth nanobers containing

tetracycline hydrochloride for periodontal regeneration
Marziyeh Ranjbar-Mohammadi a, M. Zamani b,c, M.P. Prabhakaran c,, S. Hajir Bahrami d,, S. Ramakrishna b,c
a

Textile Engineering Group, Department of Engineering, University of Bonab, Bonab, Iran

Mechanical Engineering Department, National University of Singapore, Singapore
Nanoscience and Nanotechnology Initiative, Faculty of Engineering, National University of Singapore, Singapore
d
Textile Engineering Department, Amirkabir University of Technology, Tehran, Iran
b
c

a r t i c l e

i n f o

Article history:
Received 14 May 2015
Received in revised form 23 August 2015
Accepted 27 August 2015
Available online 3 September 2015
Keywords:
Gum tragacanth
PLGA
Periodontal regeneration
Electrospun nanobers
Coaxial electrospinning

a b s t r a c t
Controlled drug release is a process in which a predetermined amount of drug is released for longer period
of time, ranging from days to months, in a controlled manner. In this study, novel drug delivery devices were
fabricated via blend electrospinning and coaxial electrospinning using poly lactic glycolic acid (PLGA), gum
tragacanth (GT) and tetracycline hydrochloride (TCH) as a hydrophilic model drug in different compositions
and their performance as a drug carrier scaffold was evaluated. Scanning electron microscopy (SEM) results
showed that fabricated PLGA, blend PLGA/GT and core shell PLGA/GT nanobers had a smooth and bead-less
morphology with the diameter ranging from 180 to 460 nm. Drug release studies showed that both the fraction
of GT within blend nanobers and the coreshell structure can effectively control TCH release rate from
the nanobrous membranes. By incorporation of TCH into coreshell nanobers, drug release was sustained
for 75 days with only 19% of burst release within the rst 2 h. The prolonged drug release, together with proven
biocompatibility, antibacterial and mechanical properties of drug loaded core shell nanobers make them a
promising candidate to be used as drug delivery system for periodontal diseases.
2015 Elsevier B.V. All rights reserved.

1. Introduction
Drug delivery systems are engineered technologies for the controlled release of therapeutic agents to achieve therapeutic purposes
in humans or animals. In recent years, the market for drug delivery technology (DDT) has increased tremendously [1], and it is forecasted
to reach $136 billion by 2021 [2]. Controlled drug release systems
have shown benets over conventional drugs [3], such as improved
adequacy, reduced side effects, improved patient compliance, and reduced toxicity [4,5]. Electrospinning is one of the developed techniques,
which enables the design and production of nanostructured drug carriers with high loading capacity, encapsulation efciency, multi-drug
delivery with ease of operation, and cost-effectiveness [6,7]. In most
cases, drugs are blended with the polymeric solution to produce drug
incorporated nanobers, which might result in low delivery efciency
and burst release [8], while other electrospinning techniques such as

Correspondence to: M.P. Prabhakaran, Amirkabir University of Technology, 424, Hafez

Ave. Tehran, Iran.
Correspondence to: H. Bahrami, Nanoscience and Nanotechnology Initiative,
E3-05-14, Faculty of Engineering, 2 Engineering Drive 3, National University of Singapore,
Singapore.
E-mail addresses: nnimpp@nus.edu.sg (M.P. Prabhakaran), hajirb@aut.ac.ir
(S.H. Bahrami).

http://dx.doi.org/10.1016/j.msec.2015.08.066
0928-4931/ 2015 Elsevier B.V. All rights reserved.

emulsion or coaxial electrospinning showed capabilities to overcome

some of these problems [9,10]. Use of nanobrous scaffolds loaded
with antibiotic drugs for biomedical applications especially for treatment of infections after tissue damages such as burn, ulcers, surgery
or periodontal disease has evoked considerable interest [11]. Mefoxin incorporated poly(lactide-co-glycolide) membrane displayed a controlled
release of the drug for over 6 days [12]. Kenawy et al. [13], fabricated tetracycline hydrochloride (TCH) loaded poly(ethylene-co-vinyl acetate),
poly(lactic acid) and their blends, then they reported controlled release
of drug over 5 days. However, controlled release of antibiotics is required
for a longer period of time for treatment of some of the chronic infections
such as periodontal diseases. Nevertheless, long term release of hydrophilic drugs (such as TCH) from nanobrous scaffolds is still challenging
due to high solubility of the drug molecule in aqueous mediums. It was
previously demonstrated that the compatibility between hygroscopic
properties of drug and polymer is essential to obtain a sustained drug
release from nanobrous delivery system [14,15].
Poly lactic-co-glycolic acid (PLGA) is a Food and Drug Administration
(FDA) approved synthetic polymer, and is one of the most attractive
polymeric candidates used for the fabrication of drug delivery devices
[16]. Gum tragacanth (GT), on the other hand, is a branched, heterogeneous and anionic polysaccharide with properties such as moisture absorption, hydrocolloid formation, water solubility, drug holding and
releasing abilities and it is categorized as generally recognized as safe

522

M. Ranjbar-Mohammadi et al. / Materials Science and Engineering C 58 (2016) 521531

(GRAS) material at a level of 0.201.30% in food stuffs. This natural biopolymer is a mixture of two soluble and insoluble polysaccharides.
Tragacanthin, a galacturonic acid part of tragacanth which is water
soluble and branched with high molecular weight which gives highly
viscous solutions and bassorin, the other part of tragacanth, is a complex
of methoxylated acids that is insoluble in water and swells to form a gel
or viscous solution [1719]. It is approved as a food additive in European
Union and has the number E 413 in the list of additives conrmed by the
Scientic Committee for Food of the European Community [20]. GT also
exhibited signicant potency for wound healing in the form of mucilage
or blended nanobers with PCL or PVA because of an acceleration in
collagenation and proliferation phases of the wound repair [2123].
Thus, we hypothesized that incorporation of hydrophilic drugs into
composite nanobers of PLGA and GT could provide a more sustained
and prolonged release of the drug, due to better hygroscopic compatibility of the drug and polymeric matrix.
Here, for the rst time, we aim towards the fabrication of composite
scaffolds of PLGA and GT at various ratios via blending and coaxial
electrospinning. Further we investigated the controlled release of
TCH incorporated within these nanobers, along with the physical
characteristics (i.e. wettability, porosity), mechanical properties and
cytocompatibility of the composite nanobers, which are critically important for a nanobrous mat to be employed as scaffolds for periodontal disease treatment.
2. Experimental procedure
2.1. Materials
PLGA with lactic acid:glycolic acid (LA:GA) ratio of 75:25 with an
intrinsic viscosity of 0.72 dl g 1 was purchased from Boehringer
Ingelheim Pharma GmbH & Co. (Ingelheim, Germany). Gum tragacanth
used in this study was a high quality ribbon type, collected from the
stems of Fluccosus species of Astragalus bushes, grown in the central
areas of Iran. TCH (purity N95%) and 1,1,1,3,3,3-hexauoro-2-propanol
(HFP) was purchased from Sigma Aldrich. Human dermal broblasts
(HDFs) were obtained from American Type Culture Collection. Dulbecco
modied Eagle's medium (DMEM), fetal bovine serum (FBS), penicillin
streptomycin solution and trypsin-ethylene diamine tetra acetic acid
were purchased from Gibco, Invitrogen Corp., USA.

were deposited on aluminum wrapped collector at a distance of 15 cm

from the needle tip, dried overnight under vacuum and used for characterization, drug release, and cell proliferation experiments.
2.3. Characterization of nanobers
The morphology of the electrospun nanobers was studied under a
Field Emission Scanning Electron Microscope (FESEM; JEOL JSM-6701F, Japan) after sputter coating with gold (JEOL JFC-1200 ne coater,
Japan) at an accelerating voltage of 15 kV. Diameters of the electrospun
bers were analyzed from the SEM images using image analysis software (Image J, National Institutes of Health, USA). Coreshell structure
of drug loaded PLGA-GT (PG(cs)-TCH) nanobers was examined using
transmission electron microscopy (TEM) (JEOL JFM-3010, Japan). Attenuated total reectance Fourier transform infrared (ATR-FTIR) spectroscopic analysis of the electrospun scaffolds was fullled using a
Nicolet Avatar 380 spectrometer (Thermo Nicolet, Waltham, MA) over
the range of 6003800 cm1 at a resolution of 4 cm.
The pore size of the nanobrous scaffolds was studied using a CFP1200-A capillary ow porometer (PMI, New York, NY). Three samples
of each type of nanobers with the same thickness and a dimension of
2 2 cm2 were used for measuring the pore size. Galwick with a surface
tension of 15.9 dynes/cm (PMI, New York, NY) was used as the wetting
liquid. Wettability of the nanobers was determined via contact angle
measurement, and a sessile drop method-based video contact angle
system (VCA Optima, AST Products, Billerica, MA) was used for this
purpose. The size of the distilled water droplet was set at 1.0 L.
The mechanical properties of the electrospun membranes were determined by a uniaxial tensile machine (Instron5943, Canton, MA)
with a load cell capacity of 10 N and cross head speed of 5 mm min1.
All nanober tape samples were cut in the form of rectangular shape
with dimensions of 30 10 mm2. At rst, a white window paper template was cut and nanobrous tapes were glued onto the top and bottom
areas of the window. It was placed between the grips of the tensile
testing machine and after closing the grips, the other sides of the window papers were cut by a scissor. Tensile test was carried out for the
as obtained dry electrospun mats, and the 48 h phosphate-buffered saline (PBS) hydrated scaffolds. A minimum of six specimens of individual
scaffolds were tested.
2.4. Nanober degradation studies

2.2. Blend and coreshell electrospinning

Electrospinning was performed to prepare blend nanobers of
PLGA-GT (PG) in three different weight ratios, including 100:0, 75:25,
50:50 (wt.%). Polymers were dissolved in HFP for making a total concentration of 16% (w/v). For fabrication of drug containing nanobers,
5% (w/w) TCH (based on the total weight of PLGA and GT) was added
and stirred for 30 min. Prepared solution was loaded into individual
3-mL syringe attached to a 25G blunted stainless steel needle and a
high voltage of 15 kV was applied to the tip of the needle. The ow
rate of the solutions was maintained at 1.0 mL/h using a syringe pump
(KDS 100, KD Scientic, Holliston, MA). For fabrication of core shell
nanobers, PLGA was dissolved in HFP to obtain 16% (w/v) solution
which was used as the shell. GT was dissolved in water to obtain 2%
(w/v) solution and stirred overnight to be used as the core solution.
The polymer solutions were separately fed into 3 mL standard syringes
attached to a coaxial nuzzle. The inner diameter of shell capillary was
0.84 mm, while the smaller capillary had outer and inner diameters of
0.56 mm and 0.30 mm, respectively. A high voltage of 15 kV (Gamma
High Voltage Research, Ormond Beach, FL) was applied, while the ow
rate of the shell and core was maintained at 1.0 mL/h and 0.2 mL/h, respectively, and the polymer solution was drawn into bers. To make
TCH incorporated core shell bers TCH was added to the core solution
(5% w/w) based on the total amount of PLGA and GT, considering the
concentration and aw rate of core and shell solutions. Nanobers

To perform the biodegradability test, the bers on cover slips were

immersed in PBS (pH of 7.4) and incubated for a period of 40 days at
37 C. At each specic time point, the scaffolds were washed and subsequently dried in a vacuum oven for 48 h. The morphology changes were
studied by FESEM.
2.5. TCH release from electrospun nanobers
Release of TCH from electrospun nanobers was measured using a
UV/VIS instrument (Shimadzu 3600, UVVIS-NIR Spectrophotometer).
For this, the drug containing nanobrous mats were accurately weighed,
placed in tightly capped glass bottle, soaked in 1 mL of PBS (pH 7.4) and
kept in shaking incubator at 37 C and 150 rpm. The UV absorbance of
TCH released in buffer solution was determined at max = 362 nm
and converted to the TCH concentration, according to the calibration
curve of TCH in the same media [24]. The cumulative release of TCH
against release time was further plotted. Fig. 1 gives the graphical representation of the preparation of nanobers and studying release properties of them.
2.6. Cell culture and proliferation studies
In vitro biocompatibility of electrospun mats was evaluated
using human dermal broblast (HDF) cells. The proliferation test

M. Ranjbar-Mohammadi et al. / Materials Science and Engineering C 58 (2016) 521531

523

Fig. 1. Graphical representation of the preparation of nanobers and studying release properties.

was done using a colorimetric 3-(4,5-dimethylthiazol-2-yl)-5-(3carboxymethoxyphenyl)-2(4-sulfophenyl)-2H tetrazolium (MTS) assay

(CellTiter 96 AQueous One solution; Promega, USA). In brief, broblasts
were cultured in DMEM augmented with 10% FBS and 1% antibiotic
and antimycotic solutions in a 75 cm2 cell culture ask. Cells were cultured in a humidied incubator at 37 C with 5% CO2, and the media
were replaced every two days. The nanobrous scaffolds were sterilized
under UV radiation for 2 h, washed 3 times with PBS and then sank in
DMEM overnight before cell seeding. For cell seeding, the cells were detached by adding 0.25% of trypsin containing 0.1% EDTA, centrifuged,
counted using a hemocytometer and seeded on the scaffolds placed
in a 24-well plate and tissue culture polystyrene (TCP as control) at a
density of 5000 cells per well. After 1, 3 and 5 days, broblast seeded
scaffolds were rinsed with PBS and 20% of MTS reagent in serum-free
medium was added. After incubation for 3 h, aliquots were pipetted
into a 96-well plate and absorbance of the obtained dye was measured
at 492 nm using a spectrophotometric plate reader (FLUO star Optima,
BMG Lab Technologies, and Germany). The intensity of the obtained
color is directly proportional to the metabolic activity of the cell
population.
2.7. Antibacterial properties
The antimicrobial behavior of PG(cs), PG 75:25-TCH and PG(cs)-TCH
nanobers was studied by agar plate method. Staphylococcus aureus
ATCC 25923 and Pseudo aeruginosa ATCC 27853 were used as Grampositive and Gram-negative bacteria respectively. MuellerHinton
agar media were sterilized in an autoclave at 121 C for 20 min under
15 lbs./in2 pressures. A loop of each bacterium was inoculated on 5 mL
of nutrient broth and incubated at 37 C for 24 h, then cultured in nutrient agar plate. The disk shape samples of nanobrous mat were sterilized by ultraviolet light for 2 h and were placed in a plate. Then the
plates were held in an incubator for 24 h. Images from the samples
were used for assessing the antimicrobial behavior.

2.8. Statistical analysis

All data presented are expressed as mean standard deviation (SD).
Statistical analysis was carried out using one-way analysis of variance
(ANOVA), followed by Tukey post hoc test for multiple comparisons,
and signicance was considered at p 0.05.
3. Results
3.1. Physical and chemical characterization of nanobers
The SEM micrographs of the electrospun nanobers are shown in
Fig. 2. The optimization of the electrospinning conditions with respect
to the concentration of the polymer, applied voltage, ow rate and
distance between the collector and the needle tip was performed, thus
producing a continuous stretch of bers. At the optimized condition,
uniform bead-free bers of PLGA, blend PG and PG(cs) nanobers
were fabricated by blend and coaxial electrospinning. Addition of the
natural polysaccharide (GT) to PLGA produced composite PG nanobers
with diameters much lower than those of PLGA (Table 1). The ber diameter of pure PLGA, PG 75:25 and PG 50:50 without TCH was obtained
as 460 16, 296 25 and 187 26 nm, respectively, which indicates
the reduction in ber diameter of the composite scaffolds using higher
amounts of gum tragacanth in PG nanobers.
At the same time, TCH incorporation into PLGA nanobers decreased the ber diameter, while TCH loaded PG scaffolds did not exhibit signicant changes in ber diameter, compared with drug free
scaffolds. Fig. 2 also exhibited the formation of smooth, continuous
and beadless PG(cs) and PG(cs)-TCH nanobers with a diameter of
399 31 and 180 24 nm, respectively. Similar to pure PLGA nanobers, the diameter of TCH loaded core shell nanobers was signicantly lower than that of drug free PG(cs) nanobers. Fig. 3 shows
the TEM image of drug loaded PG(cs) nanobers prepared by coaxial
electrospinning. The TEM micrograph of the coaxially electrospun bers

524

M. Ranjbar-Mohammadi et al. / Materials Science and Engineering C 58 (2016) 521531

Fig. 2. Morphology and size distribution of electrospun nanobers P as PLGA, P-TCH as PLGA-TCH, PG as PLGA/GT, PG-TCH as PLGA/GT-TCH, PG(cs) as core shell PLGA/GT, and PG(cs)-TCH
as core shell PLGA/GT-TCH).

clearly shows the core compartment embedded within the shell of the
polymer.
The results of pore size measurements are shown in Table 1. The
pore size of the nanobrous mats was signicantly decreased by addition of both GT and drug, compared to pure PLGA and similar PG
drug-free scaffolds, respectively.
Table 1
Diameter and pore size of electrospun nanobers.
Scaffold

Fiber diameter (nm)

Average thickness (m)

Pore size (m)

PLGA
PLGA-TCH
PG 75:25
PG 75:25-TCH
PG 50:50
PG 50:50 -TCH
PG(cs)
PG(cs)-TCH

460 16
288 33
296 25
221 42
187 26
180 24
399 31
197 42

61.20
61.76
61.90
61.34
60.95
60.86
61.56
61.02

2.92 0.02
0.95 0.03
1.80 0.01
0.99 0.04
1.0 0.01
0.63 0.02
1.35 0.02
1.2 0.03

The ATR-FTIR spectra of the electrospun PLGA nanobers (Fig. 4A)

showed peaks at 1757 cm1, 1452 cm1, 1186 cm1 and 1089 cm1
which correspond to the carbonyl group (CO bond), ether group
(COC) and methyl group (CH), respectively. In PG blend nanobers,
the peaks of GT have been partially overlapped by the peaks of PLGA
itself, and hence the OH peaks appeared weaker for these scaffolds.
For drug loaded PLGA nanobers, the carbonyl peak appeared broadened, because of the hydrogen bonding between the hydroxyl groups
of TCH and C = O groups of PLGA. Within the spectra of PLGA/TCH
scaffolds, the stretching frequencies at 1614 and 1579 cm 1 correspond to the carbonyl groups in A and C rings of TCH, respectively
(Fig. 4A,B). These peaks were seen for the drug loaded PG nanobers
prepared by blend electrospinning, but were not obviously visible for
the coreshell PG-TCH nanobers. A mild shift in carbonyl group
wavelength from 1760 to 1754 cm1 for drug loaded PG nanobers
in comparison with PG nanobers might have occurred due to the
weakening of some bonds due to hydrogen bonding between drug
and PLGA or GT.

M. Ranjbar-Mohammadi et al. / Materials Science and Engineering C 58 (2016) 521531

525

properties of TCH. At the same time, the PG(cs) nanobers were more
hydrophilic than the PLGA nanobers. However, the hydrophilicity of
PG(cs) was lesser compared to PG nanobers due to the presence of
the hydrophilic GT within the core of the nanobers.
3.2. Mechanical properties of as-spun and hydrated scaffolds

Fig. 3. TEM micrograph of TCH loaded PG(cs) nanobers.

The surface property (interaction with water) of the electrospun

scaffolds was determined dynamically by water contact angle measurement for a period of 30 s. As shown in Fig. 5, with increasing amounts of
GT within the PG composites, the hydrophilicity of the nanobrous
membrane increased, such that PG 75:25 and 50:50 scaffolds showed
a contact angle value of 85 3 and 76 2, respectively, after 30 s.
The water contact angle of drug loaded PG nanobers was signicantly lower than that of drug free PG scaffolds due to the hydrophilic

The mechanical properties of electrospun nanobrous scaffolds with

or without drug, including their tensile strength, strain at break and
elastic modulus were evaluated and are shown in Fig. 6 and Table 2.
In dry condition, the average tensile strength of PLGA nanobers was
obtained as 4.22 0.02 MPa (Fig. 6A), and it exhibited a strain at
break of 118.74 1.17%. Comparing the stressstrain curves of the different scaffolds, we found that addition of GT into PLGA nanobers signicantly decreased the tensile strength of the nanobers and reduced
the breaking strain (Fig. 6B and C). Moreover, the results of our studies
showed that PG nanobers had a reduced elastic modulus compared
to pure PLGA nanobers (Table 2). The average tensile strength and
elongation at break of PG(cs) were 3.32 0.03 MPa and 102.83
3.01%, respectively, which was signicantly higher than blend nanobers, (Fig. 6D and Table 2). For all formulations, incorporation of TCH
into nanobers reduced the tensile strength and elongation at break,
while the elastic modulus was increased compared to the scaffolds
without drug.
The results of the tensile properties of PBS-hydrated electrospun
scaffolds after 2 days are shown in Fig. 7 and Table 2. All the nanobers
without drug showed reduced tensile strength and strain after 2 days
of hydration compared to the dry state, except PG 50:50 (Fig. 7C) with
no signicant changes. Moreover, none of the TCH loaded nanobers

Fig. 4. (A) ATR-FTIR spectra of nanobers (P as PLGA, P-TCH as PLGA-TCH, PG as PLGA/GT, PG-TCH as PLGA/GT-TCH, PG(cs) as core shell PLGA/GT, PG(cs)-TCH as core shell PLGA/GT-TCH).

526

M. Ranjbar-Mohammadi et al. / Materials Science and Engineering C 58 (2016) 521531

Fig. 5. Water contact angle of nanobers (P as PLGA, P-TCH as PLGA-TCH, PG 75:25 as PLGA/GT 75:25, PG 75:25-TCH as PLGA/GT 75:25-TCH, PG 50:50 as PLGA/GT 50:50, PG 50:50-TCH as
PLGA/GT 50:50-TCH, PG(cs) as core shell PLGA/GT, PG(cs)-TCH as core shell PLGA/GT-TCH).

showed signicant changes in tensile strength and elongation at break

compared to dry state, except PG 50:50 (Fig. 7C) that had a lower
strength and PG(cs) (Fig. 7D) which exhibited a higher strain at hydrated

condition. The average tensile strength and elongation at break of PG(cs)

nanobers were obtained as 1.57 0.03 MPa and 33.41 3.15%, respectively (Fig. 7D). Furthermore, PG(cs)-TCH membrane showed the lowest

Fig. 6. Mechanical properties of the (A) PLGA, PLGA-TCH, (B) PG 75:25, PG 75:25-TCH, (C) PG 50:50, PG 50:50-TCH, (D) PG(cs) and PG(cs)-TCH electrospun nanobers in dry state. (P as
PLGA, P-TCH as PLGA-TCH, PG 75:25 as PLGA/GT 75:25, PG 75:25-TCH as PLGA/GT 75:25-TCH, PG 50:50 as PLGA/GT 50:50, PG 50:50-TCH as PLGA/GT 50:50-TCH, PG(cs) as core shell
PLGA/GT, PG(cs)-TCH as core shell PLGA/GT-TCH).

M. Ranjbar-Mohammadi et al. / Materials Science and Engineering C 58 (2016) 521531

527

Table 2
Tensile properties of the electrospun nanobers under dry and wet conditions.
Samples

PLGA
PLGA-TCH
PG 75:25
PG 75:25-TCH
PG 50:50
PG 50:50-TCH
PG(cs)
PG(cs)-TCH

Dry scaffolds

Wet scaffolds

T (m)

EM (MPa)

UTS (MPa)

SB (%)

T (m)

EM (MPa)

UTS (MPa)

SB (%)

56.80
58.02
54.99
53.07
55.03
54.09
52.04
53.75

76.00 3.60
84.33 7.81
43.34 2.23
64.10 3.34
32.95 1.00
39.62 1.31
40.20 1.71
48.32 4.13

4.22 + 0.02
2.58 0.03
2.30 0.02
1.01 0.05
0.93 0.03
0.72 0.06
3.32 0.03
1.51 0.07

118.74 1.17
15.92 2.12
22.40 2.12
13.50 6.91
13.92 4.06
8.83 7.02
102.83 3.01
14.24 4.51

50.01
52.77
56.86
54.30
53.33
56.02
49.03
48.89

69.00 1.80
62.36 2.83
42.73 1.36
52.11 2.19
32.90 1.10
33.81 1.87
105.32 1.21
22.00 1.67

2.55 0.03
1.71 0.09
1.70 0.08
1.02 0.09
0.89 0.08
0.40 0.10
2.15 0.02
1.57 0.03

64.09 1.12
11.58 1.78
15.17 1.13
11.33 5.01
14.22 1.13
7.75 3.98
36.67 1.13
33.41 3.15

T: average thickness of nanobers, EM: elastic modulus, UTS: ultimate tensile strength; SB: strain at break.

value of elastic modulus (22.00 1.67 MPa) in wet condition, compared

to other formulations (Fig. 7D and Table 2).
3.3. Degradation behavior of nanobers
Drug diffusion can be affected by scaffold's topology and morphology. Thus, the topological and morphological changes during degradation
might be able to control drug release rate from the nanobrous scaffolds. Fig. 8 presents the SEM images of the scaffolds after in vitro degradation for a period of 40 days. The brous structure of PLGA and
PG(cs) scaffolds with and without TCH was partially preserved after
40 days of degradation. However, blend PG scaffolds with and without
drug swelled to a large extent, resulting in loss of nanobrous morphology within similar time scale. These results are in agreement with
higher wettability of PG nanobers compared to PLGA and PG(cs)
nanobers.

3.4. Drug delivery

In this paper, we determined the released amounts of TCH by measuring the absorbance at 362 nm using an ultraviolet visible spectrophotometer. The release of TCH from various electrospun scaffolds is plotted
in Fig. 9A & B. PLGA nanobers exhibited an initial burst release of
23.20% within the rst 2 h (Fig. 9B) and reached a plateau within
7 days, by releasing only 35% of TCH content. The burst release of TCH
from PG 75:25 nanobers was very similar to pure PLGA (24.79%),
while PG 50:50 had a signicantly higher amount of TCH (48.30%) released in the rst 2 h, followed by a very fast release of 90% of drug content within 5 days. However, PG 75:25 composite nanobers showed
a sustained TCH release up to a period of 25 days, the time point that
it reached to the plateau state. For the core shell nanobers, where
TCH was incorporated in the core along with GT, the burst release was
signicantly lower (19%), compared to both pure PLGA and PG blend

Fig. 7. Mechanical properties of PBS hydrated (A) PLGA, PLGA-TCH, (B) PG 75:25, PG 75:25-TCH, (C) PG 50:50, PG 50:50-TCH, (D) PG(cs) and PG(cs)-TCH electrospun membranes after
48 h (P as PLGA, P-TCH as PLGA-TCH, PG 75:25 as PLGA/GT 75:25, PG 75:25-TCH as PLGA/GT 75:25-TCH, PG 50:50 as PLGA/GT 50:50, PG 50:50-TCH as PLGA/GT 50:50-TCH, PG(cs) as core
shell PLGA/GT, PG(cs)-TCH as core shell PLGA/GT-TCH).

528

M. Ranjbar-Mohammadi et al. / Materials Science and Engineering C 58 (2016) 521531

Fig. 8. SEM images of degraded nanobers after 40 days (P as PLGA, P-TCH as PLGA-TCH, PG 75:25 as PLGA/GT 75:25, PG 75:25-TCH as PLGA/GT 75:25-TCH, PG 50:50 as PLGA/GT 50:50, PG
50:50-TCH as PLGA/GT 50:50-TCH, PG(cs) as core shell PLGA/GT, PG(cs)-TCH as core shell PLGA/GT-TCH).

nanobers. Following burst release, PG(cs) scaffolds showed a prolonged

sustained release over the entire study period (75 days), by releasing
68.10% of the total TCH content by this time.
3.5. Proliferation of broblasts on electrospun scaffolds
The cytocompatibility of the electrospun scaffolds was evaluated by
MTS assay after culturing broblasts on the nanobers over a period of
1, 3, and 5 days and the results are shown in Fig. 10. Cell proliferation on

all the scaffolds (with or without drug) was found to increase with culture time, similar to the trend observed on tissue culture plates (TCP). At
days 1 and 3 of cell culture, GT contained nanobers (both blended and
coreshell) without drug exhibited improved cell viability compared
to PLGA membranes. The same trend was observed for TCH incorporated nanobers, though the presence of TCH slightly decreased cell
growth for some of the formulations compared to drug free nanobers.
However, after 5 days, TCH-loaded PG and PG(cs) nanobrous mats did
not exhibit signicant increase in cell viability compared to PLGA-TCH

Fig. 9. Release prole of TCH from PLGA-TCH, PG 75:25-TCH, PG 50:50-TCH, PG(cs)-TCH electrospun scaffolds: A) Entire release time, B) initial burst release.

M. Ranjbar-Mohammadi et al. / Materials Science and Engineering C 58 (2016) 521531

529

Fig. 10. Proliferation of broblasts on electrospun nanobers, measured by MTS assay (P as PLGA, P-TCH as PLGA-TCH, PG 75:25 as PLGA/GT 75:25, PG 75:25-TCH as PLGA/GT 75:25-TCH,
PG 50:50 as PLGA/GT 50:50, PG 50:50-TCH as PLGA/GT 50:50-TCH, PG(cs) as core shell PLGA/GT, PG(cs)-TCH as core shell PLGA/GT-TCH).

nanobers. Moreover, none of the electrospun scaffolds showed significant changes in cell viability after 5 days compared to TCP. This is an indication of cytocompatibility of the scaffolds, essential for applications
such as treatment of periodontal diseases.

may be related to structural differences between two different bacteria.

Gram-negative bacteria are more resistant due to the thick lipopolysaccharide wall structure.
4. Discussion

3.6. Antibacterial properties

In this work, the antibacterial activity of TCH-loaded PG(cs), PG
75:25-TCH and PG(cs)-TCH nanobers was investigated using S. aureus
and Pseudomonas aeruginosa as model bacteria. The drug loaded samples
showed clear bacterial inhibition rings against Gram-positive bacteria
(S. aureus) which is known to be a TCH-sensitive Gram-positive spherical bacterium that causes a wide range of suppurative infections. The
bacterial inhibition ring was smaller on the nanobers containing
Gram-negative bacteria (P. aeruginosa) (Fig. 11). These observations

Periodontitis is a major chronic inammatory disorder that can lead

to the loss of periodontal support for the periodontal ligament which
leads to the formation of an abnormal gap between the tooth and gum
[25]. If the process continues, the tooth can eventually get lost. For
chronic periodontitis, local antimicrobial agents are used as an adjunct
to scaling, root planning and restoring the periodontal health [25]. Multiple investigations have been conducted to incorporate antibiotics into
the polymeric carriers, in order to develop a DDS for treatment of periodontal diseases. Polymeric DDS were designed in different structures

Fig. 11. Antibacterial properties of nanobers; A) PG(cs), B) PG 75:25-TCH and C) PG(cs)-TCH.

530

M. Ranjbar-Mohammadi et al. / Materials Science and Engineering C 58 (2016) 521531

such as lms, microparticles and bers, using both synthetic and natural
polymers [26].
TCH is one of the widely employed antibiotics with proven effectiveness in acceleration of periodontal treatment [27]. TCH with low toxicity
is a broad spectrum antibiotic which can be applied for the treatment of
diseases caused by Gram-negative and Gram-positive microorganisms
by inhibiting protein synthesis in the bacteria [28]. Besides its antibiotic
property, it exhibits anti-inammatory properties and has the ability
to promote the attachment of broblasts. TCH has also been reported
as an inhibitor of the activity of proteinases and hence it can be used
to treat or prevent diseases related to proteinase imbalance, rheumatoid
arthritis, periodontitis and osteomyelitis [29]. Previous studies showed
that the long term routine use of TCH for several months resulted
in clinically favorable effects for periodontal disease [30]. Controlled
release of TCH was attempted by various researchers using various
drug delivery systems such as supramolecular gels based on amphiphilic 3,4,5-trihydroxybenzoic derivatives [31], device based on ethylene
vinyl acetate (EVA) copolymer [32], and porous calcium phosphate/
polyhydroxy butyrate composites [33] for various applications.
Recent advances in the eld of nanotechnology enable the fabrication
of nanobrous constructs containing drugs, such that they have the architectural features and morphological similarities matching the native
extracellular matrix (ECM) [34,35]. Unique properties of nanobers
such as the high surface area, high loading, ease of operation, and cost
effectiveness, make them more suitable as drug delivery vehicles. However, drug release characteristics (e.g. burst release, release rate and
duration) are signicantly inuenced by extent of drug encapsulation
into the nanobrous scaffolds, which is greatly dependent to material selection as well as the drug incorporation method [36,26]. In general,
drugs can be incorporated into nanobers via various methods such as
coatings, blending, co-axial and emulsion electrospinning [26]. Recent
developments in this direction progressed with the application of nanobers as drug delivery systems for periodontal diseases. For example,
Zamani et al. [26] fabricated PCL nanobers containing metronidazole
benzoate, where the drug release was continued for a period of 15 days.
In this study, we explored the incorporation of TCH into a new
bicomponent carrier, PLGA-gum tragacanth (GT), via two different
technics of blend electrospinning and coaxial electrospinning. GT, the
medicinally imported polysaccharide consists of two major fractions: a
water-soluble (tragacanthic acid and small amount of arabinogalactan)
and an insoluble but water-swellable fraction named bassorin. GT exhibited a considerable potency for wound healing in the form of mucilage
[21] or skin regeneration capability in the form of blend nanobers
with PCL or PVA [22,23]. Due to the mentioned structural and compositional advantages, natural availability, antibacterial properties, and low
cost, we believe that TCH loaded PG and PG(cs) bers can be employed
as a proper drug delivery system for multiple applications, including
treatment of periodontal diseases. We aimed to explore the effect of incorporation method (blending vs coaxial electrospinning) on the physical characteristics of the nanobers and TCH release behavior from
PG and PG(cs) nanobers. Moreover, the biocompatibility of the drug
loaded membranes was investigated.
The SEM images (Fig. 2) showed that uniformly distributed nanobers without beads were formed from all formulations. Blending GT
with PLGA decreased the diameter of nanobers. Incorporation of TCH
into PLGA and PG(cs) nanobers also reduced the ber diameter,
while addition of the drug into the blend PG nanobers did not cause
further reduction in diameter of the nanobers (Table 1). The possible
reason for the reduction of the diameter of the nanobers is that both
GT and TCH may improve the polarity of the solution which subsequently increases the electrical conductivity of the solution. Moreover,
PG(cs) nanobers exhibited reduced diameter, which can be attributed
to the use of water as the core solvent. High dielectric constant of water
(80.1 at 25 C) is an indication of the ability of solution to carry more
electrical charges, resulting in higher elongation forces and formation
of thinner nanobers under the electrical eld [37].

The results of mechanical studies showed that PG membranes exhibited less tensile strength compared to PLGA (Figs. 6 and 7). This
can be related to the easier slippage of polymer chains under loading because of less entanglements and weak physical interactions among the
chains of mixed polymers [38]. Another reason for lower strength of
PG bers can be the low mechanical strength of GT itself. Moreover,
TCH could also decrease the tensile strength as well as breaking strain
of all formulations, due to probable plasticizing effect of TCH molecules
for polymer chains. However, the breaking strain of pure PLGA was decreased more than seven times, while blend PG nanobers had less than
50% reduction in strain at the presence of TCH, compared to the similar
membranes without drug. This can be attributed to the highly branched
structure of GT, which intrinsically limited the elongation of its polymer
chains under the loading, resulting in alleviation of the inuence of TCH
on reduction of strain at break. Mechanical behavior of the nanobers
under wet condition is another factor of consideration, since the membranes inserted into the periodontal pocket are exposed to moist condition. Results of mechanical studies under wet conditions showed that
except PG 50:50 nanobers, other TCH loaded membranes preserved
their tensile strength in wet condition. Interestingly, the effect of TCH
on breaking strain of PG(cs) was dependent to dry/wet state of the
membrane. In dry state, addition of TCH caused a drastic decline in
breaking strain similar to pure PLGA, since smaller amount of GT existed
in PG(cs) nanobers compared to blend PG nanobers. However, the reduction in breaking strain of PG(cs) was limited to the extent comparable to blend PG nanobers in wet condition. This can be attributed to the
more pronounced role of GT in hydrated state due to swelling effect of
this natural polymer, resulting in distribution and occupation of a higher
volume of the nanobers by this polymer. For the membranes which are
supposed to be inserted into the periodontal pocket, it is necessary to insure that the membrane possesses enough mechanical strength and rigidity to be inserted into the pocket and retain its integrity during the
release time. On the other hand, the membrane should remain exible
and soft enough in the wet environment to conform to the periodontal
pocket and meet patient compliance [26]. Elastic modulus is a good indicator of the stiffness of a material. Among various TCH incorporated
composite nanobers, PG(cs) exhibited the highest tensile strength
in both dry and wet conditions, while its wet modulus was signicantly lower than blend nanobers. That means, TCH loaded PG(cs)
membranes may provide a proper texture to be easily inserted into
periodontal pocket with a certain amount of back pressure, and remain
comfortably in the pocket with the lowest rigidity/stiffness among the
composite nanobers.
The release kinetics of TCH from electrospun PLGA, PG 75:25,
PG 50:50, and PG(cs) scaffolds was studied for a period of 75 days
(Fig. 9). For all the formulations, initial burst release was attributed to
the release of surface connected drug [39], followed by a controlled release attributed to molecular diffusion through the polymer phase. In
case of blended PG nanobers, the presence of hydrophilic PG segments
which are randomly distributed across the diffusion path can signicantly facilitate water uptake and swelling of the polymeric matrix.
Therefore, faster diffusion of the drug molecules out of the nanobers
occurred, resulting in higher release rate within the rst few days and
reaching a plateau after a certain period of time. As expected, the
explained effect of GT on release rate was enhanced by increasing the
fraction of this polymer in the blend, such that the entire amount of
TCH was released from PG 50:50 in 20 days. These observations are in
agreement with wettability results, which conrmed the promoted
wettability of the scaffolds at higher ratios of GT in the nanobers
(Fig. 5). However, when GT was employed as the core along with TCH,
the likelihood of having drug molecules as well as GT hydrophilic
segments on/near the surface of nanobers is reduced. In this case, the
hydrophobic PLGA shell could control the water uptake and swelling
of the hydrophilic core material, resulting in lowered burst release as
well as prolonged release of TCH in a sustained fashion for 75 days.
The prolonged TCH release from PG(cs) nanobers validates the use of

M. Ranjbar-Mohammadi et al. / Materials Science and Engineering C 58 (2016) 521531

core shell structure for periodontal treatment, since the elimination of

hardened bacteria in periodontal pocket requires sustained exposure
to antibiotics [40].
The cytocompatibility studies showed that GT could successfully
support cell growth on nanobrous membranes in both blending and
coreshell architecture. This might be attributed to the more hydrophilic properties of GT contained membranes, which improves protein
adsorption and subsequent cell attachment and proliferation [41,42].
On the other hand, incorporation of TCH showed to decrease cell
growth in some of the formulations due to the inhibitory effect of this
drug on mitochondrial protein synthesis [43]. However, none of the
GT contained scaffolds showed signicant changes in cell viability compared to the control, which demonstrates good cytocompatibility of
the composite membranes. Antibacterial assessment of drug loaded
PG and PG(cs) nanobers showed that these scaffolds are strong
enough against of S. aureus bacteria.
5. Conclusion
In the present study, TCH-loaded blend and core shell nanobers
with smooth and bead-less morphology were successfully fabricated
from PLGA and GT for application as new and controlled drug delivery
systems. The release rate of TCH in PG blend nanobers increased
with the increase of GT ratio due to enhanced hydrophilicity of the
electrospun nanobers. Compared to PG blend nanobers, PG(cs) membranes showed a more prolonged release of TCH for 75 days with lower
burst release of the drug within the rst 2 h. Among various formulations, PG(cs) nanobers exhibited the highest tensile strength in both
dry and wet conditions, while its wet modulus was signicantly lower
than blend nanobers. Thus, the TCH loaded PG(cs) membranes might
be strong enough to be easily inserted into periodontal pocket and sustainably release the incorporated drug, while affording patient compliance with low rigidity/stiffness of the membrane during the treatment.
These characteristics, along with the proven cytocompatibility and antibacterial properties of TCH loaded scaffolds suggest the particular benets of composite coreshell nanobers to be used as a drug delivery
system for periodontal diseases.
Acknowledgments
This study was supported by the NRF-Technion grant (WBS No:
R-398-001-065-592) and the Nanoscience and Nanotechnology Initiative in the National University of Singapore.

[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]

References
[42]
[1] B. Felice, M.P. Prabhakaran, A.P. Rodrguez, S. Ramakrishna, J. Mater. Sci. Eng. C 4
(2014) 178195.
[2] P. Boisseau, B. Loubaton, C. R. Phys. 12 (2011) 620636.

[43]

531

T. Allen, P. Cullis, Science 303 (2004) 18181822.

Y. Malam, M. Loizidou, A.M. Seifalian, Trends Pharmacol. Sci. 30 (2009) 592599.
J. Zhang, R. Misra, Acta Biomater. 3 (2007) 838850.
B. Wang, Y. Wang, T. Yin, Q. Yu, Chem. Eng. Commun. 197 (2010) 13151338.
Z.X. Meng, X.X. Xu, W. Zheng, H.M. Zhou, L. Li, Y.F. Zheng, X. Lou, J. Colloids Surf. B 84
(2011) 97102.
G. Jin, M.P. Prabhakaran, D. Kai, S. Ramakrishna, Eur. J. Pharm. Biopharm. 85 (2013)
689698.
W. Qian, D.G. Yu, Y. Li, Y.Z. Liao, X. Wang, L. Wang, Int. J. Mol. Sci. 15 (2014) 774786.
H. Qi, P. Hu, J. Xu, A. Wang, Biomacromolecules 7 (2006) 23272330.
T.P. Chaturvedi, R. Srivastava, A.K. Srivastava, V. Gupta, V.P. Kumar, Int. J. Pharm.
Investig. 2 (2012) 213217.
K. Kim, Y.K. Luu, C. Chang, D. Fang, B.S. Hsiao, B. Chu, M. Hadjiargyrou, J. Control.
Release 98 (2004) 4756.
R. Kenawy, G.L. Bowlin, K. Manseld, J. Layman, D.G. Simpson, E.H. Sanders, G.E.
Wnek, J. Control. Release 81 (2002) 5764.
J. Zeng, L. Yang, Q. Liang, X. Zhang, H. Guan, X. Xu, X. Chen, X. Jing, J. Control. Release
105 (2005) 4351.
X. Xu, L. Yang, X. Xu, X. Wang, X. Chen, Q. Liang, J. Zeng, X. Jing, J. Control. Release
108 (2005) 3342.
H.K. Makadia, S. Siegel, Polymers (Basel) 3 (2011) 13771397.
M.U. Adikwu, Bentham Science Publishers, (2009).
A. Rasul, M. Iqbal, G. Murtaza, M.K. Waqas, M. Hanif, Acta Pol. Pharm. 67 (2010)
517522.
B. Singh, V. Sharma, Carbohydr. Polym. 101 (2014) 928940.
D.M.W. Anderson, M.M.E. Bridgeman, Phytochemistry 24 (1985) 23012304.
A. Moghbel, A.A. Hemmati, H. Agheli, I. Rashidi, K. Amraee, Arch. Iran. Med. 8 (2005)
257262.
M. Ranjbar-Mohammadi, S.H. Bahrami, M.T. Joghataei, J. Mater. Sci. Eng. C 33 (2013)
49354943.
M. Ranjbar-Mohammadi, S.H. Bahrami, J. Mater. Sci. Eng. C 48 (2015) 7179.
C.L. He, Z.M. Huang, X.J. Han, J. Biomed. Mater. Res. A 89 (2009) 8095.
S. Pappalardo, O.A. Baglio, C. Cappello, S. Guarrera, M. De Benedittis, M. Petruzzi, R.F.
Grassi, Minerva Stomatol. 55 (2006) 655661.
M. Zamani, M. Morshed, J. Varshosaz, M. Jannesari, Eur. J. Pharm. Biopharm. 75
(2010) 179185.
G. Isik, S. Ince, F. Saglam, U. Onan, J. Clin. Periodontol. 24 (1997) 589594.
A.N. Sapadin, R. Fleischmajer, J. Am. Acad. Dermatol. 54 (2006) 258265.
Z.R. Domingues, M.E. Cortes, T.A. Gomes, H.F. Diniz, C.S. Freitas, J.B. Gomes, A.M.C.
Faria, R.D. Sinisterra, Biomaterials 25 (2004) 327333.
R.A. Seymour, P.A. Heasman, J. Clin. Periodontol. 22 (1995) 2235.
L. Chen, J. Wu, L. Yuwen, T. Shu, M. Xu, M. Zhang, T. Yi, Langmuir 25 (2009)
84348438.
S. Kalachandra, L. Dongming, S. Offenbacher, J. Mater. Sci. Mater. Med. 13 (2002)
5358.
L. Medvecky, R. Stulajterova, J. Briancin, Chem. Pap. 61 (2007) 477484 (composites).
E. Vatankhah, M.P. Prabhakaran, G. Jin, L. GhasemiMobarakeh, S. Ramakrishna,
J. Biomater. Appl. 28 (2013) 909921.
M.C. Bottino, V. Thomas, G. Schmidt, Y.K. Vohra, T.M.G. Chua, M.J. Kowolik, G.M.
Janowski, Dent. Mater. 2 (2012) 703721.
M. Zamani, M.P. Prabhakaran, S. Ramakrishna, Int. J. Nanomedicine 8 (2013)
29973017.
W.K. Son, J.H. Youk, T.S. Lee, W.H. Park, Polymer 45 (2004) 29592966.
Y. Zhang, H. Ouyang, C.T. Lim, S. Ramakrishna, Z.M. Huang, J. Biomed. Mater. Res. B
Appl. Biomater. 72B (2005) 156165.
D.H. Lewis, In: M. Chasin, R. Langer (eds.), New York: Marcel Dekker, Inc. (1990) 143.
W.J. Loesche, N.S. Grossman, Clin. Microbiol. Rev. 14 (2001) 727752.
S. Fleischer, A. Shapira, O. Regev, N. Nseir, E. Zussman, T. Dvir, Biotechnol. Bioeng.
111 (2014) 12461257.
R. Ravichandran, R. Sridhar, J.R. Venugopal, S. Sundarrajan, S. Mukherjee, S.
Ramakrishna, Macromol. Biosci. 14 (2014) 515525.
C. van den Bogert, G. van Kernebeek, L. Leij, A.M. Kroon, Cancer Lett. 32 (1986)
341351.