Content uploaded by Hamid akbari javar
Author content
All content in this area was uploaded by Hamid akbari javar on May 30, 2020
Content may be subject to copyright.
Acta Biomaterialia 108 (2020) 326–336
Contents lists available at ScienceDirect
Acta Biomaterialia
journal homepage: www.elsevier.com/locate/actbio
Full length article
In situ bone tissue engineering using gene delivery nanocomplexes
Atefeh Malek-Khatabi
a
, Hamid Akbari Javar
a
, Erfan Dashtimoghadam
b
, Sahar Ansari
c
,
Mohammad Mahdi Hasani-Sadrabadi
c , d , e
, Alireza Moshaverinia
c , d , e , ∗
a
Department of Pharmaceutical Biomaterials, Faculty of Pharmacy, Tehran University of Medical Sciences, Tehran , 1417614411, Iran
b
Department of Chemistry, University of North Carolina, Chapel Hill, North Carolina, 27599-3220, United States
c
Weintraub Center for Reconstructive Biotechnology, Division of Advanced Prosthodontics, School of Dentistry, University of California, Los Angeles, USA
d
California NanoSystems Institute, University of California, Los Angeles, USA
e
Department of Bioengineering, University of California, Los Angeles, USA
a r t i c l e i n f o
Article history:
Received 31 October 2019
Revised 29 February 2020
Accepted 4 March 2020
Available online 8 March 2020
Keywo rds:
Nanocomplexes
Gene delivery
Microfluidics micromixing
Plasmid DNA
Nanofiber scaffolds
Bone tissue regeneration
a b s t r a c t
Gene delivery offers promising outcomes for functional recovery or regeneration of lost tissues at cel-
lular and tissue levels. However, more efficient carriers are needed to safely and locally delivery of ge-
netic materials. Herein, we demonstrate microfluidic-assisted synthesis of plasmid DNA (pDNA)-based
nanocomplexe (NC) platforms for bone tissue regeneration. pDNA encoding human bone morphogenesis
protein-2 (BMP-2) was used as a gene of interest. Formation and fine-tuning of nanocomplexes (NCs)
between pDNA and chitosan (CS) as carriers were performed using a micromixer platform. Flow char-
acteristics were adjusted to tune mixing time and consequently size, zeta potential, and compactness
of assembled NCs. Subsequently, NCs were immobilized on a nanofibrous Poly( ε-caprolactone) (PCL) scaf-
fold functionalized with metalloprotease-sensitive peptide (MMP-sensitive). This construct can provide an
environmental-sensitive and localized gene delivery platform. Osteogenic differentiation of bone marrow-
derived mesenchymal stem cells (MSCs) was studied using chemical and biological assays. The presented
results converge to indicate a great potential of the developed methodology for in situ bone tissue en-
gineering using immobilized microfluidic-synthesized gene delivery nanocomplexes, which is readily ex-
pandable in the field of regenerative nanomedicine.
Statement of significance
In this study, we demonstrate microfluidic-assisted synthesis of plasmid DNA (pDNA)-based nanocom-
plexes (NCs) platforms for bone tissue regeneration. We used pDNA encoding human bone morphogene-
sis protein-2 (BMP-2) as the gene of interest. Using micromixer platform nanocomplexes (NCs) between
pDNA and chitosan (CS) were fabricated and optimized. NCs were immobilized on a nanofibrous poly-
caprolactone scaffold functionalized with metalloprotease-sensitive peptide. In vitro and in vivo assays
confirmed the osteogenic differentiation of mesenchymal stem cells (MSCs). The obtained data indicated
great potential of the developed methodology for in situ bone tissue engineering using immobilized
microfluidic-synthesized gene delivery nanocomplexes, which is readily expandable in the field of re-
generative nanomedicine.
© 2020 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
1. Introduction
Bone tissue possesses an inherent regenerative ability to make
up for unwitting deformations, but some defects such as severe
trauma, pathological conditions, cancer, osteoporosis, and surgical
∗Corresponding author: Diplomate, American Board of Prosthodontics, Division
of Advanced Prosthodontics, UCLA School of Dentistry, 108 33 LeConte Ave, B3-023
CHS, Los Angeles, California 90095-1668.
E-mail address: amoshaverinia@ucla.edu (A. Moshaverinia).
revisions can disrupt this ability [1–4] . Over 2.2 million bone re-
generation procedures, including autologous, banked, and allograft
bone, are implemented annually. The bone graft substitute market
stood at US$2.35 bn in 2014 and is foreseen to exceed US$3.48
bn by the end of 2023 [ 1 , 5 , 6 ]. Nevertheless, these procedures
face some drawbacks such as donor site morbidity, donor tissue
limitation, high cost, and postoperative complications [7] . Bone
tissue engineering as a decent alternative to bone regeneration
not only has attracted significant interest thanks to its potential
https://doi.org/10.1016/j.actbio.2020.03.008
1742-7061/© 2020 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
A. Malek-Khatabi, H.A. Javar and E. Dashtimoghadam et al. / Acta Biomaterialia 108 (2020) 326–336 327
in reducing costs and surgical trauma but also has brought about
an efficient treatment of bone defects [8-10] . Scaffolds are known
as a critical part of tissue engineering. Polymeric nanofibers are
a unique candidate as scaffolds as they are capable of mimicking
the three-dimensional (3D) microenvironment of extracellular
matrix (ECM) and support growth and infiltration of the cells in
vitro and in vivo [ 11 , 12 ]. Besides, their fibrous structure provides
a large surface area for functionalization and/or encapsulation of
bioactive agents [13-17] . Here we select Polycaprolactone (PCL) as
a base material for nanofiber preparation. PCL is a biodegradable,
biocompatible, semi-crystalline, and aliphatic polyester with a tun-
able degradation and low melting point ranging. It is nontoxic in
nature and possesses a rubbery structure under the physiological
condition, which provides high toughness and superior mechanical
properties such as high strength and proper Young’s modulus and
elasticity to support osteogenic regeneration of stem cells [18] . On
the other side, the enormous potential of mesenchymal stem cells
(MSCs) toward multi-lineage differentiation makes them the most
reliable candidate among cell sources for regenerative medicine
[19] . Moreover, MSCs are capable of mediating cellular activities
through secretion of a wide range of bioactive molecules such as
growth factors (GFs) and cytokines [20] . Despite having all these
benefits, MSCs can also be genetically engineered to differentiate
toward osteoblastic phenotype and morphology. Presentation of
regulatory factors, including GFs, small molecules, and plasmid
DNAs (pDNAs), have been reported to manipulate the fate of MSCs
toward the target lineage [21–26] . Recombinant human bone mor-
phogenetic protein-2 (BMP-2) is considered as widely used GFs for
osteogenesis [27–29] . Considering the toxicity and cost of rhBMP-2
[ 30 , 31 ], pDNA encoding BMP-2 can be used as an interesting
low-cost alternative that can provide stable and long-term gene
expression in MSCs [32–34] . Various viral and non-viral carriers
have been used for gene delivery. Despite their lower efficiency,
non-viral systems can offer immuno-saf ety, cost-effectiveness,
and having an efficient transgenic capacity [ 35 , 36 ]. Non-viral
gene delivery platforms are mostly based on cationic polymers
as a vehicle to interact with negatively charged circular pDNA
electrostatically [37–40] . Chitosan biopolymer, for example, can
be used as a vehicle with enhanced cell transfection, endosome
escape, limiting the accessibility of DNase. Physical properties (like
size, surface charge, compactness) of the formed nanocomplexes
(NCs) between the carrier and pDNA can drastically affect the gene
transfection efficacy of the platform. Microfluidic systems can be
used to precisely control the formation of NCs. The application of
different types of microfluidic devices to synthesize nanocarriers
has offered a prominent advancement in various fields of drug
delivery. Nanoparticle synthesis based on microfluidic techniques
has brought about many benefits, including more homogeneity,
uniformity, and compactness, narrower size distribution, and
lower cytotoxicity to make tunable and reproducible nanoparticles
while providing operator-independent process [ 11 , 41–46 ]. The
flow rate, solute concentrations, and diameter of the designed
channels tune the size, shape, and compactness of particles in
comparison to the uncontrollable particles produced through
the conventional “bulk” method. Here, a microfluidic micromixer
was designed to help the formation of chitosan (CS)/pDNA NCs.
Moreover, the immobilization of NCs on the scaffold surface may
reduce their exposure to immune cells, NCs aggregation, and
increase cellular uptake with long-term gene expression. There-
fore, it offers well-controlled, sustained spatiotemporal delivery,
and localized gene expression. A combination of a gene delivery
approach with an electrospun nanofiber scaffold can present an
efficient, affordable, and safe process for gene therapy-based tissue
engineering.
2. Experimental section
2.1. Chemicals and biologicals
Unless noted otherwise, all chemicals were bought from Sigma-
Aldrich, Inc. (St. Louis, MO). Medium molecular weight chitosan
with molecular weight of 280,0 0 0 g/mol and degree of deacety-
lation of 83% (Fluka) was used in this study. Glassware was acid
cleaned overnight and then thoroughly rinsed with Milli-Q wa-
ter. Cell culture reagents, solutions, and dishes were obtained from
Thermo Fisher Scientific (Waltham, MA), except as indicated oth-
erwise. Mesenchymal stem cells (MSCs) were isolated from the fe-
murs and tibias of 30-day-old rats. In brief, density gradient cen-
trifugation was performed to isolate the cells from bone marrow.
Cells were cultured in DMEM-low glucose supplemented with 10%
fetal bovine serum at 37 °C in humid air containing 5% CO
2
. The
culture medium was replaced to remove non-adherent cells. Sub-
confluent cells in the seed cultures were removed from the cul-
ture flasks by treating with 0.25% trypsin/0.01% EDTA. MSCs with
passage 3–4 were used in all the experiments. Flow cytometric
study was used to evaluate the stemness of the isolated cells. Spe-
cific MSC markers including STRO-1, CD166, and CD146 were as-
sessed. The cytometric analysis demonstrated that the rat BMMSCs
expressed specific MSC markers, STRO-1, CD146, and CD166.
2.2. Microfluidic chips fabrication
Silicon wafers spin-coated with SU-8 photocurable epoxy to
a thickness of 60 μm on the normal 4-inch silicon wafer and
then baked. Conventional photolithography is used to fabricate the
master mold that contains patterned structures on the surface in
the cleanroom facility. The surface of molds was modified us-
ing trimethylethoxy silane as a self-assembled monolayer to pre-
vent PDMS (Sylgard 184, Dow Corning, Switzerland) sticking. PDMS
resin and its curing agent (10:1), poured over the patterned master
mold, degassed in desiccators, and baked in the oven. After curing,
PDMS was removed from the wafer, and the holes were punched.
The PDMS was then bonded to a glass slide after oxygen plasma
activation (100 mW, 1 min). Each of microfluidic devices consists
of three inlets and one outlet. Fluorescent flows were used to de-
termine the range of stable flow rates.
2.3. Formation of NCs
The PDMS microfluidic micromixer device has two lateral inlets
for the aqueous solution of chitosan (1 mg.mL
−1 in DNAse/RNAse-
free water) with the pH of 6 as sheath flows, the central inlet for
pDNA (pcDNA3.1/His/hBMP2 from BD Biosciences) flow, and one
outlet. The chitosan solution stream was split into two, in order to
achieve two lateral chitosan streams at the same flow rate. Follow-
ing the preparation, nanocomplexes isolated from unbound poly-
mers and pDNAs by centrifugation at 20,0 0 0 ×g for 40 min. The
supernatant also removed and lyophilized to be used to quantify
unbound polymer.
For the bulk synthesis of NCs, polymeric solutions were pre-
pared by dissolving chitosan (1 mg.mL
−1
) in acetic acid/ (20 mM)
HEPES solution under constant stirring. The nanoprecipitation of
chitosan chains performed by the dropwise addition of pDNA con-
taining solution (in nuclease-free water; pH 7. 4 ) followed vortex
mixing by 1 min.
Fluorescein-labeled chitosan (FITC
–CS) was synthesized and
used to characterize the amount of unreacted polymers, as re-
ported before [11] . Briefly, fluorescein isothiocyanate (FITC) dis-
solved in methanol (2 mg.mL
–1
) and was added gradually to the
328 A. Malek-Khatabi, H.A. Javar and E. Dashtimoghadam et al. / Acta Biomaterialia 108 (2020) 326–336
CS solution (1% w/v) in acetic acid and dialyzed (10 kDa MWCO)
after 5 h of reaction in the dark.
2.4. Electrospinning
Ester-terminated Medical grade (poly( ε-caprolactone) (PCL; M
w
:
80 kg.mol
−1
) were used for this study. The 500 μL PCL solution
(10% w/w in hexafluoroisopropanol) per scaffold was electrospun
using a laboratory-made electrospinning device at 15–20 kV at the
constant infusion rate of 2.5 mL.h
−1
.
2.5. Nanofiber surface activation
To activate the nanofiber surface, PCL mats were treated with
oxygen plasma for 5 min and then immersed in 1 M Sodium Hy-
droxide (NaOH) and placed on gentle shaking for 4 h at room tem-
perature. The treated PCL nanofibrous were washed three times in
deionized water and dried using nitrogen gas flow.
2.6. Peptide and nanocomplex conjugation
Following nanofiber surface activation, the peptide sequences
were conjugated on the surface of the scaffold. 8 mm diame-
ter electrospun mat was immersed in 500 uL of crosslinking so-
lution containing 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide
(EDC; 200 mg.mL
−1
) / N-hydroxysuccinimide (NHS; 50 mg.mL
−1
)
in 2-(N-morpholino)ethanesulfonic acid (MES, pH 5.9) for 10 min
and then washed with PBS. After the addition of selected peptides,
the scaffolds left under gentle shaking overnight at 4 °C. Then the
membranes were washed with NaCl (0.15 M; pH 8) for 30 min. The
following peptide sequences were selected based on the published
reports [47] and were synthesized on solid resin using an auto-
mated peptide synthesizer based on standard F-moc chemistry.
MMP-sensitive sequence: GCRDGPQG ↓ IAGQDRCGC
MMP-insensitive sequence: GCRDGDQGIAGFDRCGC
integrin-binding peptide cyclic RGD (cRGD): RGD(d)FK
sequence;
Fluorescent sequence: GCRD-FITC
The surface of CS-based NCs is also activated by sulfos-
uccinimidyl 6-(3
-[2- pyridyldithio ]-propionamido)hexanoate (Sulfo-
LC-SPDP) crosslinker to create sensitive disulfide bonds to react
with the thiol groups of cysteine residues on the MMP peptide
(conjugated on the surface of the scaffold. Immobilization of NCs
was performed by wetting the peptide-modified scaffolds with NCs
containing solution. The reaction will be conducted under gentle
shaking overnight at 4 °C.
2.7. Flow simulation
A 2D modeling of 10 distinctive mixer units was implemented,
and the finite element approach for various geometries was per-
formed to solve the continuity and constitutive Navier-Stocks equa-
tions in the company with the convection-diffusion equation. The
Coanda effect is to account for bending fluid streamlines and in-
creasing the mixing effect. It leads different numbers of entrance
flows to divide into two narrow straight and curved directions,
which merge together in the next junction and experience another
Coanda effect and split again in the same manner. This loop, flow
mixing, would be repeated until the complete mixing comes out.
2.8. Characterization
The size, size distribution, and Zeta potential of the prepared
NCs at pH 7.4 were investigated using dynamic light scattering (Ze-
tasizer 30 0 0HS, Malvern Instruments Ltd., Worcestershire, UK).
The morphology of nanofibers was characterized using scanning
electron microscopy (SEM; Zeiss Supra 40VP) after sputter coat-
ing of disks with an iridium layer using South Bay Technology Ion
Beam Sputtering (San Clemente, CA).
Number of carboxyl groups on the surface of PCL nanofibers be-
fore and after treatment were quantified by a toluidine blue O as-
say as reported before [48] .
Nanocomplex stability was analyzed by the protocol reported
before [49] . Briefly, Quant-iT PicoGreen reagent was added to solu-
tions containing NCs and incubated for 30 min. Then fluorescence
signal was measured, and background signal subtracted and nor-
malized to the signal in free pDNA controls (no chitosan).
To quantify the amount of unreacted polymer remaining in so-
lution following NCs formation, NCs will be removed by centrifu-
gation at 20,0 0 0 ×g for 20 min at 4 °C. The fluorescence signal
of the supernatant was then quantified to identify the unbonded
(unreacted) CS amount and corrected for the background signal.
The amount of encapsulated pDNA in the NCs was estimated
by measuring the difference between the initial amount of pDNA
added and the amount of non-encapsulated plasmid remaining in
the aqueous suspension after the formation of NCs process. The
amount of free pDNA was determined using PicoGreen reagent
as mentioned earlier. Encapsulation efficiency of NCs were deter-
mined according to previously published reports [50] .
Kinetic of MMP-2 dependence of NCs release from MMP-
sensitive sequences was measured using a modified fluorescent
quantification assay as previously reported [ 47 , 51 ]. Briefly, Nanofi-
brous scaffolds were equilibrated at 37 °C in PBS overnight be-
fore adding 5 nM recombinant human MMP-2 (rhMMP-2) (R&D
Systems). The supernatant was collected, and media replaced with
PBS containing the same concentration of fresh MMP-2 every day
to preserve enzymatic activity. The collected supernatants were an-
alyzed by quantification of fluorescein signal (FITC
–CS NCs). For
flow cytometry analysis, cells were analyzed on FACSVerse using
FlowJo software (Treestar).
The cytotoxicity (cell viability) tests were assessed using an
MTT colorimetric assay. Cytotoxicity of the NCs was determined af-
ter 72 h incubation with MSCs cultured in DMEM medium at 37 °C
in a 5% CO
2
incubator. To determine cytotoxicity, cells were plated
at a density of 10, 0 0 0 per well in 96-well plates and then incu-
bated overnight. For cell viability studies, GFP encoding plasmids
(pEGFP-N1 from BD Biosciences) was used. Commercially available
lipofectamine 20 0 0 was used to form control NCs according to pre-
viously published protocols [52] .
For cell transfection and differentiation experiments, 50,0 0 0
cells were cultured in 24-well plates for 24 h before experiments
at 37 °C and 5% CO
2
in the stem cell growth media [11] . GFP or
BMP-2 encoding NCs were added to the cells using the serum-free
medium. Transfection efficiency analyzed 24 h post-transfection
using fluorescence microscopy imaging (Leica TCS SP5 confocal mi-
croscope) and flow cytometry (FACSVerse using FlowJo software).
The amount of conjugated NCs were adjusted to present about
5 μg of plasmid DNA per square centimeter of the nanofiber mem-
brane. The control samples with soluble BMP-2 protein also tested
as positive control. The other tested group was unconjugated BMP-
2 plasmid DNA encapsulated NCs which added to the MSCs while
being cultured on RGD-modified nanofibers.
Quantitative real-time PCR assays were used to analyze gene
expression. Cultured cells were recovered from membranes after
the course of treatment, and RNA was isolated via TRIzol reagent.
The RNA levels of alkaline phosphatase (ALP), core-binding factor
alpha l (cbfa1), osteocalcin (OCN), and the alpha l chain of colla-
gen type I (COL1A1) were reverse-transcribed and single-stranded
cDNA was made using Superscript III cDNA synthesis kits. The
relative gene expression was determined using the 2
- Ct tech-
nique, normalizing to the Ct of the reference gene, glyceraldehyde-
A. Malek-Khatabi, H.A. Javar and E. Dashtimoghadam et al. / Acta Biomaterialia 108 (2020) 326–336 329
3- phosphate dehydrogenase (GAPDH): 50-TGT TCC TAC CCC CAA
TGT ATC CG-30; and 50-TGC TTC ACC ACC TTC TTG ATG TCA T-30 .
In order to quantify the ALP activity, cell culture supernatants
were collected on three introduced time points (1st, 2nd, and 4th
weeks), centrifuged at 20 0 0 ×g for 10 min at 4 °C to remove de-
bris and analyzed for ALP using an ALP Kit according to the manu-
facturer’s instructions. The absorbance was read according to man-
ufacturer protocol. The ELISA assay (Novus Biologicals, Centennial,
CO, #NBP2-68,153) was used based on two goats OCN antibodies,
which bring both the C-terminal and N-terminal proteins to deter-
mine rat OCN concentration [53] . Supernatants were collected at
days 7, 14, and 28 of the culture periods and stored at −20 °C for
later measurements. Samples were run according to the manufac-
turer’s instructions and compared to rat OCN standards. ALP activ-
ity normalized by the protein content in terms of unit/mg and OCN
value normalized by DNA content of MSCs in terms of ng/μg over
four weeks, and results are reported in first, second, and fourth. Af-
ter 4 weeks of culture, the amount of secreted calcium was quan-
tified using the O
–Cresolphthalein Complex one method, as men-
tioned before [11] . The amount of calcium present was normalized
to the total protein content of the cells. Xylenol orange (XO) was
used to stain deposited calcium after four weeks of culture. Briefly,
samples were imaged with identical conditions including same ex-
posure time and magnification, and without performing any au-
tomatic or manual signal amplification. Then the color threshold
was adjusted for red fluorescence channel. The mineralized area
quantified with ImageJ software according to the protocol reported
before [54] . All animal experiments were performed in accordance
with the Provisions and General Recommendation of Experimen-
tal Animals Administration and approved by the Institutional Ani-
mal Care and Use Committee of Tehran University of Medical Sci-
ences. Sprague-Dawley rats (male, 4-month old) were used for in
vivo functional assays as reported before [55] . Calvaria was drilled
to make 5 mm full-thickness defects without touching the underly-
ing dura mater. Each defect was washed with saline to remove de-
bris before the placement of the scaffolds. The implanted scaffolds
were harvested four 4- and 10-weeks post-surgery and micro-CT
analysis was performed to analyze the amount of bone regenera-
tion.
2.9. Statistical analysis
The Kruskal-Wallis rank sum test, one-way ANOVA and two-
tailed Student’s t -test were utilized as appropriate to analyze the
data at a significance of αor p < 0.05. Quantitative data were ex-
pressed as mean ±standard deviation (SD). For all the tests, the
threshold was set to p < 0.05 for “statistically significant,” p < 0.01
for “statistically very significant” and p < 0.001 for “statistically ex-
tremely significant.”
3. Results and discussion
Previously we used a T-shape microfluidic platform for pro-
ducing reproducible, monodisperse, well-controlled, and finely de-
signed nanostructured carriers for small molecule delivery [56–59] .
However, for the current application ( i.e., synthesis of CS/pDNA
nanocomplexes), the T-shape microfluidic platform would not be
efficient due to the existence of laminar flow as well as slow diffu-
sion kinetic of pDNA related to its high average molecular weight.
Hence, an improved mixing procedure is required. Vortex mixing
has been used to synthesize NCs via electrostatic self-assembly. But
this approach leads to heterogeneous particles which suffer high
variability in properties and efficiency originated from metastable
preparation and subsequent aggregation. In order to tackle this
crucial point, a tesla micromixer (TMM) was selected as a mi-
croreactor in this work to improve mixing efficiency to attain uni-
form, compactness, and monodisperse CS/pDNA NCs. TMM have
been utilized to provide an efficient mixing in a variety of chemical
and biological applications [ 60 , 61 ]; but, to the best of our knowl-
edge TMM has not been used to synthesize NCs except our recent
study that this platform was used to coat nanoparticles in a con-
trollable manner [62] . Both NCs properties and reactor condition
hinge on the micromixer geometrical characteristics and flow pa-
rameters such as flow ratios, flow rate, channel diameter, and so
forth [ 34 , 35 ]. Flow analysis (using COMSOL Multiphysics) was used
to study the optimum flow/concentration profiles to determine the
final design of the micromixer, as reported before [62] . In this re-
search, the TMM has three inlets, chitosan (CS) is injected from
two lateral inlets as sheath flow and pDNA (p-hBMP-2) is injected
via central channels. This micromixer will provide favorable condi-
tions for polycations (here chitosan) and polyanions (here plasmid
DNA) to face each other and form compact nanoscale aggregates.
Previously we have shown that nanoparticles those forming inside
microfluidic platforms are significantly smaller than those formed
with the bulk mixing process [56] . This method can also help for-
mation of more compact particles as these particles will form un-
der the shear [63] . The concentration distribution which provides
a proper mixing after four loops ( Fig. 1 A). The effects of process-
ing parameters including flow ratio (Polymers flow rates/total flow
rates) from 0.03 to 0.3 (μF-1: 0.07, μF-2: 0.1, μF-3: 0.2) and rela-
tive concentration of p-hBMP2 (pDNA:CS = 1:1, 1:3, and 1:5) were
studied to optimize the formulation ( Fig. 1 B–E). As rapid mixing
decreased the population of residual unreacted polymer, increas-
ing the flow ratio leads to an increase in size and reduce surface
charge (zeta potential) of resulting NCs. On the other side, by in-
crease in the abundant of chitosan the tendency of formation of
more compact NCs will increase which will be resulted in forma-
tion of smaller NCs. the presence of more NCs formed at the flow
ratio of 0.1 (μF-2), and the relative concentration of 1:3 were se-
lected for further characteristic properties due to its optimal size
( ≈130 nm), low polydispersity index (PDI < 0.2), and high sta-
bility ( Fig. 1 E). Polydispersity indexes of other NCs, μF-1 and μF-
3 are 0.18 and 0.36, respectively. Formation of NCs in fast mixing
regimes (μF-1 and μF-2) will also increase the yield of intermolecu-
lar electrostatic bonds between chitosan polycations and negatively
charged plasmid DNA chains as tested by measuring the amount of
unbonded polymers ( Fig. 1 F). Use of microfluidic mixing approach
can also increase the encapsulation efficiency of pDNA from 85%
for bulk mixing to around 98% while using microfluidic platform
( Fig. 1 G). Picogreen competition assay [ 51 , 64 ] was used to deter-
mine the compactness and binding stability of NCs ( Fig. 1 H). Ob-
taining NCs with high compactness avoids nonspecific interactions.
Cell viability of resultant μF-2 NCs was tested 72 h after incu-
bation of MSCs with NCs cultured in regular media and compared
to bulk synthesized and lipofectamine NCs as control ( Fig. 2 A). In-
significant change in the cell viability at high NC concentrations
(20 μg/ml) confirm the nontoxic nature of chitosan carrier com-
pared to commercially available lipofectamine.
Transfection efficiency of synthesized NCs in three different
flow ratios (μF-1, μF-2, and μF-3) was compared to the correspond-
ing bulk synthesized NCs, lipofectamine, and naked p-BMP2 as ad-
ditional control groups ( Fig. 2 B). It should be noted that for cell
transfection studies, GFP encoding plasmids were used instead of
pBMP-2. GFP expressing cells were visualized using fluorescence
microscopy ( Fig. 2 B) and also quantified using flow cytometry
( Fig. 2 C). NCs formed at a rapid mixing regime (μF-1 and μF-2)
show about 80 percent of transfection, which is among the high-
est reported values for MSCs to transfection [65–68] . For example,
traditional mixing, which was used to synthesize NCs of plasmid
pAcGFP1-C1 (pDNA) and spermine modified cationic starch (CS)
provided transfection efficacy of about 30 percent in HepG2 cells
[69] . In another study, Ho et al. introduced a reliable system to
330 A. Malek-Khatabi, H.A. Javar and E. Dashtimoghadam et al. / Acta Biomaterialia 108 (2020) 326–336
Fig. 1. (A) A schematic representation of a Tesla micromixer to make chitosan-based gene delivery nanocomplex. COMSOL simulation results showing concentration distri-
butions. Hydrodynamic diameter (based on DLS) and Zeta potential of the resulted particles as different chitosan (CS) to BMP-2 plasmid DNA (pDNA); CS:pDNA 1:1 (B), 5:1
(C), 3:1 (D). (E) release of pDNA from nanocomplexes formed at different conditions as a function of time. (F) The amount of unreacted FITC-labeled chitosan remaining in
the solution after the formation of nanocomplexes with different methods and constant CS:pDNA of 3:1 as evaluated after removing the formed nanocomplexes by centrifu-
gation. (G)
Encapsulation efficiency of prepared NCs formed at different flow ratios. (H) Binding stabilities of pDNA to the nanocomplexes formed at different flow ratios as
quantified based on competition assay. Fluorescence signals were corrected for background signal and normalized such that polymer-free controls = 1.0 arbitrary units (AU).
The data are expressed as mean ±SD. number of independent experiments (n): 5. The results were statistically analyzed using one-way ANOVA with post-hoc analysis. For
all the tests, the threshold was set to p < 0.05 for “statistically significant”, p < 0.01 for “statistically very significant” and p < 0.001 for “statistically extremely significant”.
Statistical significance is indicated by
∗(significant),
∗∗ (very significant), and
∗∗∗ (extremely significant) for differences between samples with different formulations.
A. Malek-Khatabi, H.A. Javar and E. Dashtimoghadam et al. / Acta Biomaterialia 108 (2020) 326–336 331
Fig. 2. (A) The effect of nanocomplex formulation and concentration on the viability of hBMMSC (based on MTT assay). (B) Fluorescence images showing transfected (GFP + )
cells with microfluidic and bulk prepared nanocomplexes loaded with 2.5 μg of GFP encoding plasmid. Scale bar: 250 μm. (C) Transfection efficiency of nanocomplexes as
quantified with flow cytometry. The data are expressed as mean ±SD ( n = 5). The results were statistically analyzed using one-way ANOVA with post-hoc analysis. For all
the tests, the threshold was set to p < 0.05 for “statistically significant”, p < 0.01 for “statistically very significant” and p < 0.001 for “statistically extremely significant”.
Statistical significance is indicated by
∗(significant),
∗∗ (very significant), and
∗∗∗ (extremely significant) for differences between samples with different formulations.
Fig. 3. (A) Schematic of the electrospinning and conjugation processes used to make poly( ε-caprolactone) (PCL) nanofibers and to immobilized CS/pDNA nanocomplexes to
PCL nanofibers via MMP-responsive linkage. (B) Scanning electron microscopy (SEM) image of plain nanofibers. (C) Fluorescent micrograph showing successful conjugation
of FITC-labeled MMP on nanofibers’ surfaces. (D) SEM image of CS/pDNA NC immobilized on the surface of PCL nanofibers. (E) Quantification of remaining pDNA in the
construct before and after conjugation and washing steps. The data are expressed as mean ±SD ( n = 5). The results were statistically analyzed using one-way ANOVA with
post-hoc analysis. The threshold was set to p < 0.0 01 for “statistically extremely significant”. Statistical significance is indicated by
∗∗∗ (extremely significant) for differences
between samples with different formulations.
332 A. Malek-Khatabi, H.A. Javar and E. Dashtimoghadam et al. / Acta Biomaterialia 108 (2020) 326–336
Fig. 4. (A) Cumulative in vitro release of pDNA from Chitosan/pDNA NCs conjugated via MMP sensitive and MMS non-sensitive linkages. Release profiles were evaluated
at 37 °C and pH of 7.4 in the presence or absence of MMP-2 enzyme. (B) Flow cytometry analysis of internalization of surface-immobilized nanocomplex in the
presence
or absence of MMP-2. The data are expressed as mean ±SD ( n = 5). The results were statistically analyzed using one-way ANOVA with post-hoc analysis. The threshold
was set to p < 0.001 for “statistically extremely significant”. Statistical significance is indicated by
∗∗∗ (extremely significant) for differences between samples with different
formulations.
improve the delivery efficacy and provide a thoroughly adjustable,
controllable microfluidic process of NCs formation with the accept-
able (50–70%) transfection efficacy [42] .
Although the formation of gene delivery carriers is very criti-
cal but localized delivery of these therapeutic agents can provide
additional benefits, especially in the context of tissue engineering.
Here we utilized nanofibrous electrospun mats as support scaffold
to immobilize and deliver engineered NCs. PCL was used as a syn-
thetic biocompatible and bioresorbable materials with adjustable
mechanical and biological properties [70–72] . We have optimized
properties of PCL based nanofiber membranes for periodontal tis-
sue regeneration [ 73 , 74 ]. The formation of 2 μm thick fibers was
demonstrated using scanning electron microscopy (SEM) ( Fig. 3 A).
Mechanical (stress-strain curve) and degradation properties of de-
veloped PCL nanofibrous scaffold summarized in Supporting In-
formation section (Supplementary Figure 2). The designed nanofi-
brous scaffold will provide Young’s modulus of about 1.5 MPa,
which can support osteogenic differentiation of stem cells due to
its stiffness [73] . The surface of the PCL nanofibrous scaffolds was
further functionalized to promote the localized delivery of trans-
fection agents, reduce cytotoxicity, and maximize the cellular at-
tachment. Metalloprotease (MMP)-sensitive (or MMP-insensitive)
and cyclic RGD (cRGD) peptide sequences were immobilized on
the surface of nanofiber scaffold after oxygen plasma following
by controlled hydrolysis. Oxygen plasma and hydrolysis can in-
crease the number of carboxylic groups at the surface of nanofibers
from about 7 ±5 to 121 ±17 nmol of -COOH per mg of PCL. Pro-
vided oxygen-enriched reactive groups on the surface of the PCL
scaffold will be functionalized with target peptides via EDC/NHS
chemistry. Chitosan-based NCs will be activated using sulfosuccin-
imidyl 6-(3
-[2- pyridyldithio ]-propionamido) hexanoate (Sulfo-LC-
SPDP) crosslinker to provide reactive site toward thiol groups (cys-
teines amino acid) at the end of MMP peptide sequences ( Fig. 3 B).
The bioconjugation mechanism was described in detail in supple-
mentary Figure 1. This immobilization strategy can provide con-
trolled (MMP responsive) cleavage of peptide sequence and sub-
sequent release of conjugated NCs. Also, the interaction between
MSCs integrin ( αv
β3
) receptors and cRGD peptide sequences fa-
cilitates the process of interaction and adherence of stem cells to
the substrate. Conjugation of fluorescently (FITC) labeled peptide
was used to verify homogenous peptide coating ( Fig. 3 C). SEM mi-
crograph of immobilized NCs on the surface of the functionalized
scaffold was shown in Fig. 3 D.
Due to the applied reaction conditions and several washing
steps, NCs may lose their cargo during the immobilization process.
Hereby comparing the pDNA contents before and after conjuga-
tion, the stability of NCs were tested for both microfluidic-assisted
and bulk synthesized NCs ( Fig. 3 E). microfluidic-assisted synthe-
sized NCs (μF-2) show higher stability (around 70%) due to its
more compact structures compare to bulk synthesized NCs (around
30%). This stability is highly desirable to protect the cargo (here p-
BMP-2) and may reduce the required therapeutic dose. [ 11 , 24 ]
After the functionalization and immobilization of NCs on the
fibrous scaffold surface, the release profile was examined. Both
release profiles of immobilized NCs through sensitive and non-
sensitive MMP crosslinkers are indicated at the presence and ab-
A. Malek-Khatabi, H.A. Javar and E. Dashtimoghadam et al. / Acta Biomaterialia 108 (2020) 326–336 333
Fig. 5. (A) Schematic representation of a cell-construct interaction. RGD containing peptide will provide the cell-interaction sites. The presence of a digestive enzyme (like
MMP-2) accelerated the release of plasmid DNA nanocomplexes, followed by the uptake of these particles by neighbor stem cells. Analysis of gene expression profiles of
osteogenic markers. Real-time polymerase chain re action (qPCR) analysis of the osteogenic differentiation of MSCs cultured on PCL-based membranes after 4 weeks in regular
culture. Expression levels of five osteogenic genes, BMP-2 (B), ALP, Col1, RUNX2, and OCN (C), we re evaluated relative to GAPDH, which was used as a housekeeping gene.
(D) ALP
activity by protein content and (E) Osteocalcin per DNA content of MSCs after culturing in regular media containing various formulation of either bonded or free
plasmid-loaded nanocomplexes for 4 weeks. (F) Calcium deposition on extracellular matrix and quantification of mineralized area based on Xyle no l orange calcium detection
assay. (G) Xyle nol orange staining indicating the mineralized area for cultured MSCs on PCL-based nanofibrous membranes after 4 weeks. The data are expressed as mean
±SD ( n = 5). The results were statistically analyzed using one-way ANOVA with post-hoc analysis. For all the tests, the threshold was set to p < 0.05 for “statistically
significant”, p < 0.01 for “statistically very significant” and p < 0.001 for “statistically extremely significant”. Statistical significance is indicated by
∗(significant),
∗∗ (very
significant), and
∗∗∗ (extremely significant) for differences between samples with different formulations.
334 A. Malek-Khatabi, H.A. Javar and E. Dashtimoghadam et al. / Acta Biomaterialia 108 (2020) 326–336
Fig. 6. In vivo evaluation of osteogenesis in rat calvarial model after 4 and 10 weeks. Scale bar: 5 mm. The data are expressed as mean ±SD ( n = 5). The results were
statistically analyzed using one-way ANOVA with post-hoc analysis. For all the tests, the threshold was set
to p < 0.05 for “statistically significant”, p < 0.01 for “statistically
very significant” and p < 0.001 for “statistically extremely significant”. Statistical significance is indicated by
∗(significant),
∗∗ (very significant), and
∗∗∗ (extremely significant)
for differences between samples with different formulations.
sence of MMP-2 as a function of time ( Fig. 4 ). Immobilized NCs via
insensitive linkers do not show a significant release whether MMP-
2 is presented or not; however, immobilized NCs by sensitive link-
ers (PCL-MMP-NCs) show a remarkable NCs release (around 60%
after 24 h) at the presence of MMP-2. MMP-2 inhibitor control was
used to show that the process can be modulated by the environ-
mental factor to accelerate or inhibit the release of immobilized
plasmids.
In the next step, MSCs were cultured on the functionalized scaf-
fold surfaces, MMP-2 secretion leads to sustain cleavage of immo-
bilized NCs by sensitive MMP linkers from the scaffold surface that
release NCs and make them accessible to MSCs ( Fig. 4 A). Stem cells
can also internalize released NCs, which can direct the cells to-
ward osteogenesis. Here, we analyzed internalization of nanocom-
plexes in the presence and absence of soluble MMP-2. As shown
in Fig. 4 B, incorporation of MMP-sensitive peptide can significantly
increase the released of NCs. These NCs can easily internalized
by cultured MSCs and affect their fates. Quantitative polymerase
chain reaction (qPCR) was used to investigate the expression level
of BMP2 as well as various osteoblastic markers including alkaline
phosphatase (ALP), runt-related transcription factor2 (RUNX2), os-
teocalcin (OCN), and alpha-1 type I collagen (col1A1) after one and
four weeks of culturing cells in regular media in presence of free or
immobilized plasmid delivery particles (NCs) ( Fig. 5 B, C). The solu-
ble addition of recombinant hBMP-2 was used as a positive control.
A significant increase ( > 70 folds) was observed in BMP2 ex-
pression level while having NCs conjugated via MMP-2 sensitive
peptide ( Fig. 5 B). In addition to BMP-2, higher expression levels of
other osteogenic markers also demonstrated ( Fig. 5 C). For exam-
ple, a higher expression level of ALP after four weeks and com-
pare to the other control groups was observed. ALP has a criti-
cal role in the initiation of mineralization but is not required for
the continuation of bone nodules mineralization [75] . RUNX2 is
known as a central gene that contributes to the osteoblast pheno-
type induction. qPCR data show significantly higher RUNX2 expres-
sion levels after four weeks of culture compared to other control
groups ( Fig. 5 C). Besides, an increase in col1A1 mRNA, as another
marker of bone tissue regeneration, we also detected the expres-
sion of late-stage osteogenic and bone-specific markers like OCN.
OCN is produced right before and along with matrix mineraliza-
tion. Relative mRNA expression of OCN shows a significant increase
compared to the samples prepared without MMP-2 sensitive pep-
tides. It should also be noted that although, the presence of MMP-
insensitive peptides limits the release and potential internalization
of the plasmid loaded-NCs, but as the stem cells are cultured on
the nanofibers which are conjugated with NCs, they may engulf
the NCs and also the release of plasmid DNA over time may trig-
ger the osteogenesis of cultured cells.
Chemical (p-nitrophenol) assays also performed to test the ac-
tivity of expressed ALP. High ALP activity indicates a high differen-
tiation rate of MSCs and, subsequently, a higher bone regeneration
rate. The amount of secreted OCN into the culture medium also
quantified and compared between different groups after 1, 2, and
4 weeks of the culture ( Fig. 5 E). Results demonstrate OCN protein
expression follows the same trend as quantified osteogenic genes,
especially in the later time points. Calcium phosphate (CaPs) is a
major component of bone tissue. Calcium deposition can be de-
tected during the in vitro osteogenesis. Here, calcium deposition
content on ECM ( Fig. 5 D) and also the percentage of the miner-
alized area ( Fig. 5 F) were assessed using calcium colorimetric as-
say and Xylenol Orange (XO) staining, respectively. Similar to ge-
netic and protein assays, calcium deposition measurements proved
the statistically significant improvement ( p < 0.01) in osteogenesis
while using engineered construct developed in this work.
We also performed in vivo experiments to test the function-
ality of the proposed strategy in the animal model. Calvarial de-
fect model was created in rats followed by insertion of MSCs-
loaded (2 ×1010
5 cell/cm
2
) onto PCL membranes with different
formulations. Our preliminary results demonstrated a significant
increase in the regenerated bone volume and also forming more
dense bone-like structures ( Fig. 6 ). Our results and previously re-
ported studies confirm the importance of proteolytic sensitivity of
the scaffold for remodeling and new bone formation [47] . Thus, the
healing response in vivo depended on the enzymatic sensitivity of
the immobilized gene delivery carriers.
In the current study, the design and characterization of a na-
noengineered platform to repair and regenerate a segment of bone
tissue was presented. Controlled formation of plasmid DNA-based
nanocomplexes (NCs) were described. Immobilization of these NCs
through a digestive enzyme (here MMP-2) sensitive peptide was
used to localize the effect as well as offering the environmen-
tally sensitive presence of plasmids. Structure, stability, and func-
tionality of these constructs were tested to engineer a platform to
promote osteogenic differentiation of stem cells. Although, further
characterization is needed to highlight the mechanism of action
and to optimize the in vivo functionality, however, we believe such
a smart and environmentally sensitive platform may offer fine-
tuning of the tissue regeneration and recovery.
A. Malek-Khatabi, H.A. Javar and E. Dashtimoghadam et al. / Acta Biomaterialia 108 (2020) 326–336 335
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared to
influence the work reported in this paper.
Supplementary material
Supplementary material associated with this article can be
found, in the online version, at doi: 10.1016/j.actbio.2020.03.008 .
References
[1] J.R. Lieberman , G.E. Friedlaender , Bone Regeneration and repair: Biology and
Clinical Applications, Springer, 2005 .
[2] R. Dimitriou , E. Tsiridis , P.V . Giannoudis , Current concepts of molecular aspects
of bone healing, Injury 36 (12) (20 05) 1392–1404 .
[3] R. Dimitriou , E. Jones , D. McGonagle , P. V . Giannoudis , Bone regeneration: cur-
rent concepts and future directions, BMC Med 9 (1) (2011) 66 .
[4] A .S. Mistry , A .G. Mikos , in: Tissue Engineering Strategies For Bone regenera-
tion, Regenerative medicine II, Springer, 2005, pp. 1–22 .
[5] B. Trajkovski , M. Jaunich
, W.- D. Müller , F. Beuer , G.-G. Zafiropoulos , A. Housh-
mand , Hydrophilicity, viscoelastic, and physicochemical properties varia tions
in dental bone grafting substitutes, Materials (Basel) 11 (2) (2018) 215 .
[6] M. Bohner , Resorbable biomaterials as bone graft substitutes, Mater. To da y 13
(1–2) (2010) 24–30 .
[7] C.J. Damien , J.R. Parsons , Bone graft and bone graft substitutes: a review of
current technology and applications, J. Appl. Biomater. 2 (3) (1991) 187–208 .
[8] K.J. Burg , S. Porter , J.F. Kellam , Biomaterial developments for bone tissue engi-
neering, Biomaterials 21 (23) (20 0 0) 2347–2359 .
[9] S. Bose , M. Roy , A. Bandyopadhyay , Recent advances in bone tissue engineering
scaffolds, Trends Biotechnol. 30 (10) (2012) 546–554 .
[10] M.M. Stevens , Biomaterials for bone tissue engineering, Mater. Tod ay 11 (5)
(2008) 18–25 .
[11] M.M. Hasani-Sadrabadi , S.P. Hajrezaei , S.H. Emami , G. Bahlakeh , L. Danesh-
mandi , E. Dashtimoghadam , E. Seyedjafari , K.I. Jacob , L. Tayebi , Enhanced os-
teogenic differentiation of stem cells via microfluidics synthesized nanoparti-
cles, nanomedicine, Nanotechnol., Biol., and Med. 11 (7) (2015) 1809–1819 .
[12] D.W. Hutmacher , J.T. Schantz , C.X. Lam
, K.C. Tan , T.C. Lim , State of the art and
future directions of scaffold-based bone engineering from a biomaterials per-
spective, J. Tissue Eng. Regen. Med. 1 (4) (2007) 245–260 .
[13] S. Agarwal , J.H. We ndo rff, A. Greiner , Progress in the field of electrospinning
for tissue engineering applications, Adv. Mater. 21 (32–33) (20 09) 3343–3351 .
[14] A . Martins , A .R. Duarte , S. Faria , A .P. Marques , R.L. Re is , N.M. Neves , Osteogenic
induction of hBMSCs by electrospun scaffolds with dexamethasone release
functionality, Biomaterials 31 (22) (2010) 5875–5885 .
[15] P.O. Rujitanaroj
, Y.C. Wang , J. Wang , S.Y. Chew , Nanofiber-mediated controlled
release of siRNA complexes for long term gene-silencing applications, Bioma-
terials 32 (25) (2011) 5915–5923 .
[16] T.J. Sill , H.A. von Recum , Electrospinning: applications in drug delivery and tis-
sue engineering, Biomaterials 29 (13) (2008) 1989–2006 .
[17] T. Uyar , R. Havelund , J. Hacaloglu , F. Besenbacher , P. Kingshott , Functional
electrospun polystyrene nanofibers incorporating
α-, β-, and γ-Cyclodextrins:
comparison of molecular filter performance, ACS Nano 4 (9) (2010) 5121–5130 .
[18] R. Dwivedi , S. Kumar , R. Pandey , A. Mahajan , D. Nandana , D.S. Katti , D. Mehro-
tra , Polycaprolactone as biomaterial for bone scaffolds: review of literature, J.
Oral Biol. Craniofac. Res. 10 (1) (2020) 381–388 .
[19] J. Maia , T. Santos , S. Aday , F. Agasse , L. Cortes , J.O. Malva , L. Bernardino , L. Fer-
reira , Controlling the neuronal differentiation of stem cells by the intracellular
delivery of retinoic acid-loaded nanoparticles, AC S Nano 5 (1) (2011) 97–106 .
[20] A.I. Caplan , Adult mesenchymal stem cells for tissue engineering versu s regen-
erative medicine, J. Cell. Physiol. 213 (2) (2007) 341–347 .
[21] M. Bouyer , R. Guillot , J. Lavaud , C. Plettinx , C. Olivier , V. Curry , J. Boutonnat ,
J.L. Coll , F. Peyrin , V. Josserand , G. Bettega , C. Picart , Surface delivery of tunable
doses of BMP-2 from an adaptable polymeric scaffold induces volumetric bone
regeneration, Biomaterials 10 4 (2016) 168–181 .
[22] T.H. Kim , R.K. Singh , M.S. Kang , J.H. Kim , H.W. Kim , Gene delivery nanocarriers
of bioactive glass with unique potential to load BMP2 plasmid DNA and to in-
ternalize into mesenchymal stem cells for osteogenesis and bone regene ration,
Nanoscale 8 (15) (2016) 8300–8311 .
[23] C.T. Laurencin , K.M. Ashe , N. Henry , H.M. Kan , K.W. Lo , Delivery of small
molecules for bone regenerative engineering: preclinical studies and potential
clinical applications, Drug Discov. To da y 19 (6) (2014) 794–800 .
[24] N. Monteiro , D. Ribeiro , A. Martins , S. Faria , N.A. Fonseca , J.N. Moreira , R.L. Reis ,
N.M. Neves , Instructive nanofibrous scaffold comprising runt-related transcrip-
tion
factor 2 gene delivery for bone tissue engineering, ACS Nano 8 (8) (2014)
8082–8094 .
[25] H. Nie , M.L. Ho , C.K. Wang , C.H. Wang , Y.C. Fu , BMP-2 plasmid loaded
PLGA/HAp composite scaffolds for treatment of bone defects in nude mice,
Biomaterials 30 (5) (2009) 892–901 .
[26] C.W. Pouton , K.M. Wagstaff, D.M. Roth , G.W. Moseley , D.A. Jans , Targe ted de-
livery to the nucleus, Adv. Drug Deliv. Rev. 59 (8) (2007) 698–717 .
[27] A. Valentin-Opran , J. Wozney , C. Csimma , L. Lilly , G.E. Riedel , Clinical evalua-
tion of recombinant human bone morphogenetic protein-2, Clin. Orthop. Relat.
Res. 395 (2002) 110– 12 0 .
[28] B. Chen , H. Lin , J. Wa ng , Y. Zhao , B. Wa ng , W. Zhao , W. Sun , J. Dai , Homoge-
neous osteogenesis and bone regeneration by demineralized bone matrix load-
ing with collagen-targeting bone morphogenetic protein-2, Biomaterials 28 (6)
(2007) 1027–1035 .
[29] H. Yo ne zawa , K. Harada , T. Ikebe , M. Shinohara , S. Enomoto , Effect of recombi-
nant human
bone morphogenetic protein-2 (rhBMP-2) on bone consolidation
on distraction osteogenesis: a preliminary study in rabbit mandibles, J. Cran-
io-Maxillofacial Surg. 34 (5) (2006) 270–276 .
[30] F. Wegman , R.E. Geuze , Y. J. van der Helm , F. Cumhur Oner , W. J. Dhert , J. Alblas ,
Gene delivery of bone morphogenetic protein-2 plasmid DNA promotes bone
formation in a large animal model, J. Tissue Eng. Regen. Med. 8 (10) (2014)
763–770 .
[31] Y. Zhang , W. Li , T. Laurent , S. Ding , Small molecules, big roles– the chemical
manipulation of stem cell fate and somatic cell reprogramming, J.
Cell. Sci. 125
(Pt 23) (2012) 5609–5620 .
[32] S.-Y. Park , K.-H. Kim , S. Kim , Y.- M. Lee , Y.-J . Seol , BMP-2 gene delivery-based
bone regeneration in dentistry, Pharmaceutics 11 (8) (2019) 393 .
[33] A. Kolk , M. Boskov , S. Haidari , T. Tischer , M. van Griensven , O. Bissinger ,
C. Plank , Comparative analysis of bone regeneration behavior using recombi-
nant human BMP-2 versus plasmid DNA of BMP2, J. Biomed. Mater. Res. Part A
107 (1) (2019) 163–173 .
[34] B. Khorsand , N. Nicholson , A.-V. Do , J.E. Femino , J.A. Martin , E. Petersen ,
B. Guetschow , D.C. Frederick s , A.K. Salem , Regeneration of bone using nanoplex
delivery of FGF-2 and BMP-2 genes in diaphyseal long bone radial defects in a
diabetic rabbit model, J. Controlled Release 24 8 (2017) 53–59 .
[35] R. Gardlik , R. Palffy , J. Hodosy , J. Lukacs , J. Tur na , P. Celec , Vectors and delivery
systems in gene therapy, Med. Sci. Monitor 11 (4) (2005) Ra110–Ra121 .
[36] N. Nayerossadat , T. Maedeh , P.A . Ali , Viral and nonviral delivery systems for
gene delivery, Adv. Biomed. Res. 1 (2012) 27
-27 .
[37] A.B. Hill , M. Chen , C.-K. Chen , B.A. Pfeifer , C.H. Jones , Overcoming gene-del iv-
ery hurdles: physiological considerations for nonviral vectors, Trends Biotech-
nol. 34 (2) (2016) 91–105 .
[38] B.R. Olden , Y. Cheng , L.Y. Jonathan , S.H. Pun , Cationic polymers for non-viral
gene delivery to human T cells, J. Controlled Release 282 (2018) 140 –147 .
[39] Y. Rui , D.R. Wilson , J.J. Green , Non-Viral delivery to enable genome editing,
Trends Biotechnol. (2018) .
[40] Y. Liu , C.-F. Xu , S. Iqbal , X.-Z. Yan g , J. Wang , Responsive nanocarriers as an
emerging platform for cascaded delivery of nucleic acids to cancer, Adv. Drug
Deliv. Rev. 115 (2017) 98–114 .
[41] M.M. Hasani-Sadrabadi , V. Karimkhani , F.S. Majedi , J.J. Van Dersarl , E. Dash-
timoghadam , F. Afshar-Taromi , H. Mirzadeh , A. Bertsch , K.I. Jacob , P. Renaud ,
F.J. Stadler , I. Kim , Microfluidic-assisted self-assembly of complex dendritic
polyethylene drug delivery nanocapsules, Advanced Materials 26 (19) (2014)
3118–3123 .
[42] Y.- P. Ho , C.L. Grigsby , F. Zhao , K.W. Leong , Tuning physical properties of
nanocomplexes through microfluidics-assisted confinement, Nano Lett. 11
(5)
(2011) 2178–2182 .
[43] A.T. Hsieh , N. Hori , R. Massoudi , P. J . Pan , H. Sasaki , Y.A. Lin , A.P. Lee , Nonvi-
ral gene vector formation in monodispersed picolitre incubator for consistent
gene delivery, Lab Chip 9 (18) (2009) 2638–2643 .
[44] R. Karnik , F. Gu , P. Basto , C. Cannizzaro , L. Dean , W. Kyei-Manu , R. Langer ,
O.C. Farokhzad , Microfluidic platform for controlled synthesis of polymeric
nanoparticles, Na no Lett. 8 (9) (20 08) 2906–2912 .
[45] F.S. Majedi , M.M. Hasani-Sadrabadi , J.J. VanDersarl , N. Mokarram , S. Hojjati-E-
mami
, E. Dashtimoghadam , S. Bonakdar , M.A. Shokrgozar , A. Bertsch , P. Re-
naud , On-Chip fabrication of paclitaxel-loaded chitosan nanoparticles for can-
cer therapeutics, Adv. Funct. Mater. 24 (4) (2014) 432–441 .
[46] S. Soleimani , M.M. Hasani-Sadrabadi , F. S. Majedi , E. Dashtimoghadam , M. Ton -
dar , K.I. Jacob , Understanding biophysical behaviours of microfluidic-synthe-
sized nanoparticles at nano-biointerface, Colloids and Surf. B, Biointerfaces 14 5
(2016) 802–811 .
[47] M. Lutolf , J. Lauer-Fields , H. Schmoekel , A.T. Metters , F. Web er , G. Fields ,
J.A. Hubbell , Synthetic matrix metalloproteinase-sensitive hydrogel s for the
conduction of tissue regeneration: engineering cell-invasion characteristics,
Proc. Nat . Acad. Sci. 100 (9) (2003) 5413–5418 .
[48] X. Li , B. Cho , R. Martin , M. Seu , C. Zhang , Z. Zhou , J.S. Choi , X. Jiang , L. Chen ,
G. Walia , Nanofiber-hydrogel composite–mediated angiogenesis for soft tissue
reconstruction, Sci. Transl. Med. 11 (490) (2019) eaau6210 .
[49] C.L. Grigsby , Y. -P. Ho , C. Lin , J.F. Engbersen , K.W. Leong , Microfluidic prepara-
tion of polymer-nucleic acid nanocomplexes improves nonviral gene transfer,
Sci. Rep. 3 (2013) 3155 .
[50] A. Bozkir , O.M. Saka
, Chitosan nanoparticles for plasmid DNA delivery: effect of
chitosan molecular structure on formulation and release characteristics, Drug
Deliv. 11 (2) (2004) 107–112 .
[51] R.J. Wad e , E.J. Bassin , C.B. Rodell , J.A. Burdick , Protease-degradable electrospun
fibrous hy droge ls, Nat. Commun. 6 (2015) 6639 .
[52] B. Dalby , S. Cates , A. Harris , E.C. Ohki , M.L. Tilkins , P. J . Price , V.C. Ciccarone ,Ad-
vanced transfection with lipofectamine 20 0 0 reagent: primary neurons, siRNA,
and high-throughput applications, Methods 33 (2) (2004) 95–103 .
[53] J.L. Santos , E. Oramas , A.P. Pêgo , P.L . Granja
, H. Tomás , Osteogenic differenti-
ation of mesenchymal stem cells using PAMAM dendrimers as gene delivery
vectors, J. Controlled Release 134 (2) (20 09) 141–148 .
336 A. Malek-Khatabi, H.A. Javar and E. Dashtimoghadam et al. / Acta Biomaterialia 108 (2020) 326–336
[54] J. Cardeira , P.J . Gavaia , I. Fernández , I.F. Cengiz , J. Moreira-Silva , J.M. Oliveira ,
R.L. Reis , M.L. Cancela , V. Laizé, Quantitative assessment of the regenerative
and mineralogenic performances of the zebrafish caudal fin, Sci. Rep. 6 (2016)
39191 .
[55] T. Aghaloo , C.M. Cowan , Y.- F. Chou , X. Zhang , H. Lee , S. Miao , N. Hong ,
S.I. Kuroda , B. Wu , K. Ting , Nell-1-induced bone regeneration in calvarial de-
fects, Am. J. Pathol. 169 (3) (2006) 903–915 .
[56] F.S. Majedi , M.M. Hasani-Sadrabadi , J.J. VanDersarl , N. Mokarram , S. Hojjati-E-
mami , E. Dashtimoghadam , S. Bonakdar , M.A. Shokrgozar , A. Bertsch , P. Re-
naud , On-chip fabrication of paclitaxel-loaded chitosan nanoparticles for can-
cer therapeutics, Adv. Funct. Mater. 24 (4) (2014) 432–441 .
[57] F.S. Majedi , M.M. Hasani-Sadrabadi , S.H. Emami , M. Taghipoor , E. Dashti-
moghadam , A. Bertsch , H. Moaddel , P. Renaud , Microfluidic synthesis of chi-
tosan-based nanoparticles for fuel cell applications, Chem. Commun. 48 (62)
(2012) 7744–7746 .
[58] M.M. Hasani-Sadrabadi , F.S. Majedi , J.J. VanDersarl , E. Dashtimoghadam ,
S.R. Ghaffarian , A. Bertsch , H. Moaddel , P. Renau d , Morphological tuning of
polymeric nanoparticles via microfluidic platform for fuel cell applications, J.
Am. Chem. Soc. 13 4 (46) (2012) 18904–18907 .
[59] M.M. Hasani-Sadrabadi , V. Karimkhani , F.S. Majedi , J.J. Van Dersarl , E. Dash-
timoghadam , F. Afshar-Taromi , H. Mirzadeh , A. Bertsch , K.I. Jacob , P. Renaud ,
Microfluidic-Assisted self-assembly of complex dendritic polyethylene drug de-
livery nanocapsules, Adv. Mater. 26 (19) (2014) 3118–3123 .
[60] A.-S. Yan g , F.-C. Chuang , C.-K. Chen , M.-H. Lee , S.-W. Chen , T.-L. Su , Y.- C. Yang
,
A high-performance micromixer using three-dimensional Tesla structures for
bio-applications, Chem. Eng. J. 263 (2015) 4 4 4–451 .
[61] C.-Y. Lee , W.-T. Wang , C.-C. Liu , L.-M. Fu , Passive mixers in microfluidic sys-
tems: a review, Chem. Eng. J. 288 (2016) 146–160 .
[62] M.M. Hasani-Sadrabadi , S. Tarane jo o , E. Dashtimoghadam , G. Bahlakeh , F. S. Ma-
jedi , J.J. VanDersarl , M. Janmaleki , F. Sharifi, A. Bertsch , K. Hourigan , L. Ta yeb i ,
P. Renaud , K.I. Jacob , Microfluidic manipulation of core/shell nanoparticles for
oral delivery of chemotherapeutics: a new treatment approach
for colorectal
cancer, Adv. Mater. 28 (21) (2016) 4134–4141 .
[63] S. Soleimani , M.M. Hasani-Sadrabadi , F. S. Majedi , E. Dashtimoghadam , M. Ton -
dar , K.I. Jacob , Understanding biophysical behaviours of microfluidic-synthe-
sized nanoparticles at nano-biointerface, Colloids and Surf. B: Biointerfaces 145
(2016) 802–811 .
[64] P.L . Ma , M. Lavertu , F. M. Winnik , M.D. Buschmann , Stability and binding affin-
ity of DNA/chitosan complexes by polyanion competition, Carbohydr. Polym.
176 (2017) 167–176 .
[65] T. Yoshikawa , E. Uchimura , M. Kishi , D.P. Funeri u , M. Miyake , J. Miyake , Trans-
fection microarray of human mesenchymal stem cells and on-chip siRNA gene
knockdown, J. Controlled Release 96 (2) (2004) 227–232 .
[66] H.-Q. Mao , K. Roy , V.L. Troung-Le , K.A. Janes , K.Y. Lin , Y. Wa ng , J.T. August ,
K.W. Leong , Chitosan-DNA nanoparticles as gene carriers: synthesis, character-
ization and transfection efficiency, J. Controlled Release 70 (3) (2001) 399–421 .
[67] Y. Wu , A.E. Smith , T.M . Reineke , Lipophilic polycation vehicles display high
plasmid DNA
delivery to multiple cell types, Bioconjug. Chem. 28 (8) (2017)
2035–2040 .
[68] T.- H. Kim , R.K. Singh , M.S. Kang , J.-H. Kim , H.-W. Kim , Gene delivery nanocar-
riers of bioactive glass with unique potential to load BMP2 plasmid DNA and
to internalize into mesenchymal stem cells for osteogenesis and bone regener-
ation, Nanoscale 8 (15) (2016) 8300–8311 .
[69] H. Wang , X. Li , L. Chen , X. Huang , L. Li , Cationic starch/pDNA nanocomplexes
assembly and their nanostructure changes on gene transfection efficiency, Sci.
Rep. 7 (1) (2017) 14 84 4 .
[70] M.A. Woodruff, D.W.
Hutmacher , The return of a forgotten polymer—Polycapro-
lactone in the 21s t century, Prog. Polym. Sci. 35 (10) (2010) 1217–1256 .
[71] J. Venugopal , S. Ramakrishna , Applications of polymer nanofibers in
biomedicine and biotechnology, Appl. Biochem. Biotechnol. 125 (3) (20 05)
147–157 .
[72] E. Malikmammadov , T.E. Tanir , A. Kiziltay , V. Hasirci , N. Hasirci , PCL and
PCL-based materials in biomedical applications, J. Biomater. Sci., Polym. Ed. 29
(7–9) (2018) 863–893 .
[73] M.M. Hasani-Sadrabadi , P. Sarrion , N. Nakats uka , T.D . Yo un g , N. Taghdiri ,
S. Ansari , T. Aghaloo , S. Li
, A. Khademhosseini , P. S. Weiss , Hierarchically pat-
terned polydopamine-containing membranes for periodontal tissue engineer-
ing, AC S Nano 13 (4) (2019) 3830–3838 .
[74] A. Nasajpour , S. Ansari , C. Rinoldi , A.S. Rad , T. Aghaloo , S.R. Shin , Y. K. Mishra ,
R. Adelung , W. Swieszkowski , N. Annabi , A. Khademhosseini , A. Moshaverinia ,
A. Tama yol , A multifunctional polymeric periodontal membrane with os-
teogenic and antibacterial characteristics, Adv. Funct. Mater. 28 (3) (2018)
170 3 43 7 .
[75] H.C. Blair , Q.C. Larrouture , Y. Li , H. Lin , D. Beer-Stoltz , L. Liu
, R.S. Tuan ,
L.J. Robins on , P.H. Schlesinger , D.J. Nelson , Osteoblast differentiation and bone
matrix formation in vivo and in vitro , Tissue Eng. Part B: Rev. 23 (3) (2017)
268–280 .