J Mater Sci: Mater Med (2017) 28:32
DOI 10.1007/s10856-017-5849-z
TISSUE REGENERATION OR REGENERATION OF ENGINEERED TISSUE?
Dual-acting biofunctionalised scaffolds for applications in
regenerative medicine
Camilo Chaves1,2,3 Chuanyu Gao1 Jerome Hunckler1 Moustafa Elsawy1,4
Josette Legagneux3 Gilles Renault5 Alain Charles Masquelet6 Achala de Mel1
Received: 22 November 2016 / Accepted: 6 January 2017 / Published online: 20 January 2017
Springer Science+Business Media New York 2017
Abstract Off the shelf scaffolds for replacing ultra-small
diameter vascular grafts are valuable for reconstruction of
diseased or damaged vessels. The limitations for such grafts
include optimal handling with ready availability of varied
lengths of grafts, graft patency with the ability to replace the
function of active cellular mechanisms and adequate
mechanical properties to maintain physicochemical func-
tion. We used a well-established, solvent casting method for
potential tissue replacement scaffold fabrication with
incorporated bioactive molecules, which we have pre-
viously explored to confer haemocompatibility. These
grafts were tested in-vivo within the abdominal aorta of 10
Wistar rats and the patency was clinically and echo-
graphically evaluated. Haemocompatibility and
Electronic supplementary material The online version of this article
(doi:10.1007/s10856-017-5849-z) contains supplementary material,
which is available to authorized users.
Camilo Chaves and Chuanyu Gao contributed equally to the study.
* Achala de Mel
demelach@gmail.com
1
Division of Surgery and Interventional Science, University
College London, London, UK
2
Universit Pierre et Marie Curie, 4 Place Jussieu, Paris 75005,
France
3
Ecole de Chirurgie, AGEPS, AP-HP, 7 Rue du Fer Moulin,
Paris 75005, France
4
University College London Hospitals, London, UK
5 material (ECM) and promote cell-material interactions. Sig-
Institut CochinINSERM U1016, 27 rue du fbg Saint Jacques,
Paris 75014, France
6 host tissue with an adequate blood supply. Thus angiogenesis
Hpital St Antoine, Service de Chirurgie Orthopdique &
Traumatologique, Unit Chirurgie Rparatrice & Chirurgie de la
Main, Paris F-75571, France
Original Research
endothelialisation were assessed on explants. Biofunctio-
nalised scaffolds were also grafted subcutaneously and
intraperitoneally to evaluate integration, inammation and
angiogenesis reactions. The potential wider applications of
this dual acting scaffold were evaluated for its interactions
with human dermal broblasts as well as bronchial epithe-
lial cells. Physicochemical property evaluation of the
functionalised grafts has claried the mechanical strength
and permeability. This study conrmed the microsurgical
suturability of tubular grafts and graft patency of functio-
nalized scaffolds. The study demonstrated the potential of a
dual acting biofunctionalised scaffolds use for a wide range
of tissue engineering applications where micro-porous, yet
impermeable scaffolds are needed.
1 Introduction
There is a need for biomimetic materials that can be used as
tissue substitutes or medical devices that can address damaged
and diseased tissue [13]. Biostable materials are ideal as
permanent tissue replacements. In particular, when such
replacements are required for emergency surgery, it is
important to have access to off the shelf scaffolds, which
through interactions with biological systems can induce bio-
logical functionality in situ. Elastomers have versatile prop-
erties, which are highly relevant for surgical implants,
especially their exible mechanical properties as well as the
ability to incorporate functional moieties or induce surface
topographical features that could biomimic the extra cellular
nicantly, tissue replacement scaffolds must integrate with the
is an essential feature for a tissue engineering scaffold [46].
Among many factors that can promote angiogenesis,
32 Page 2 of 12
bioactive molecules that encourage vessel formation and a
degree of scaffold porosity are signicant for efcacious graft
function. Whilst porosity is a signicant feature in tissue
engineering [79], the process of tubular structure construc-
tion, requires the graft exterior to be impermeable so that the
luminal content is contained and provides a channel of
delivery for the relevant content. Therefore, direct fabrication
of a porous yet impermeable scaffold is useful. This study
demonstrates a single-step fabrication method of obtaining
porous, non-permeable scaffold through elution of bioactive,
yet biologically favourable molecules.
A previous study has considered the above properties
in vitro where a poly(carbonate-urea) urethane (PCU), was
functionalised with L-Arginine Methyl Ester (L-AME) [10].
L-AME is an analogue of L-Arginine (L-Arg), which is an
endogenous substrate of nitric oxide synthase and thus plays a
signicant role in the cardiovascular [11], central nervous
system [12], respiratory system [13], as well as in wound
healing. This study aimed to clarify the in vivo biocompat-
ibility of the preliminary in vitro results and the angiogenesis
potential.
Considering the 3Rs (replacement, reduction, renement)
linked to in vivo studies, we have opted for a rat model as a
small animal model to test material properties [14]. Despite
their suitability for the assessment of micro-vascular con-
duits, rat models are generally recognised to have a robust
vascular system that can resist many features associated with
test-graft failure. Therefore, in this study as a compromise of
observing 3Rs and obtaining meaningful results, we intro-
duced a graft mismatch at the site of anastomosis in order to
induce local turbulence in blood ow, [15] with the aim of
comparing and evaluating the better performing scaffold in
the presence of adverse vascular events. Therefore, the aim
of the aortic vessel anastomosis was to evaluate the scaffold
that can better resist thrombosis inducing factors, indepen-
dent of the scaffold mechanical properties in vivo.
This scaffold in previous studies has shown positive
interactions with human dermal broblasts, (HDFs) and
endothelial progenitor cells (EPC) in vitro [10] and here we
demonstrate its interactions with epithelial cells, with the
intention to evaluate the practical application and suitability
to apply this functionalised material in medical devices and
tissue replacements. Figure 1 presents a summary of the
experimental plan aimed to address the main aims of this
article.
2 Materials and methods
2.1 PCU and L-AME-PCU polymer
Scaffolds were generated from 18% PCU synthesised as
below. Polycarbonate polyol, 2000 mwt was placed in a
J Mater Sci: Mater Med (2017) 28:32
250 ml reaction ask, equipped with mechanical stirrer and
nitrogen inlet, and heated to 90 C.
4,4-Methylenebis(phenyl isocyanate), (MDI), was added
to the polyol and then reacted, under nitrogen, at 7585 C
for 90 min to form a pre-polymer. Dimethylacetamide was
added slowly to the pre-polymer to form a solution; the
solution was cooled to 40 C. Chain extension of the pre-
polymer was carried out by the drop wise addition of
ethylenediamine and isobutylamine in dimethylacetamide to
form a solution of polycarbonate urea-urethane in dime-
thylacetamide. After completion of the chain extension 1-
butanol in dimethylacetamide was added to the polymer
solution.
90, 180 and 270 mg of L-arginine methyl ester dihy-
drochloride (L-AME) (Sigma-Aldrich) were dissolved in
500 l of deionised water and vortexed to produce a clear
solution. The solution was added drop by drop to 5 g of
18% PCU with 2 ml of dimethylacetamide (DMAC), fol-
lowed by vortexing and gentle heating in a water bath to
produce a clear solution.
2.2 Tubular scaffold fabrication
Five millimeter diameter stainless steel mandrels (Metal
Mania online retailer) were attached to a exible lament,
(to adopt a vertical balance during solvent evaporation)
when suspended to hang freely in an oven (Fig. 1). The
mandrels were dipped into 50 ml centrifuge tube containing
PCU with 180 mg of L-AME (L-AME-PCU) and a control
with PCU only. The coated mandrels were placed in an
oven set at 65 C for 3 h. They were then inverted and re-
dipped into the centrifuge tube to ensure that the polymer
was evenly distributed. This process was repeated to pro-
duce scaffolds with 6, 8 and 10 coats. Tubes were then
immersed overnight in propanol-lled test tubes before
separating from the mandrels.
2.3 Scaffold fabrication for in vitro tests
PCU and L-AME-PCU polymer solution were poured on
10 cm-diameter circular glass plates and left in an air-
circulating oven at 65 C for at least 12 h to produce a cast
polymer. Thirteen millimeter diameter disks were cut from
the casted polymer and sterilised in 70% ethanol for 1 h, and
left under UV light for 1 h, followed by three rounds of
washing in phosphate-buffered saline (PBS) of 5 min each.
Sterilised disks were incubated in respective cell culture
medium for at least 3 h prior to cell seeding.
2.4 Scaffold sterilisation
Scaffolds were immersed in 70% ethanol solution overnight
(Sigma) and ethanol was evaporated in a biosafety cabinet
J Mater Sci: Mater Med (2017) 28:32 Page 3 of 12 32

Fig. 1 A summary of the

in vitro, and in vivo
experimental plan. In vitro and
in vivo characterisation of L-
AME functionalised PCU
scaffolds. a and b L-AME and
PCU in DMAC solvent were
mixed in centrifuge. c Stainless
steel mandrel as suspended by a
exible tubing and was dip
coated with PCU or L-AME-
PCU. d DMAC (solvent)
evaporates in the oven (e) the
mandrels with polymer coatings
were inverted and dipped in the
polymer solution to achieve its
desired number of even coatings
(f). Tubular scaffolds (g) were
then anastomosed to Wister rat
aorta (h). Mixture was poured
onto a glass plate (i). Plate with
the mixture is placed into the
oven (d). DMAC evaporates in
the oven (j) to produce casted
polymer (k) which was then cut
into 13 mm diameter disks (l).
Human bronchial epithelial cells
(HBEC) and human dermal
brobalsts (m) were grown in
cell culture asks (n). Cells were
seeded onto scaffolds placed in
48-well plates (o)

under ultra violet light to ensure sterility of the scaffolds. 2% PFA and 1.5% GTA) and critical point dried before
Scaffolds that have been removed from the mandrel were being processed for analysis.
immersed in deionized water-lled 50 ml centrifuge tubes
for 48 h. The tubes were placed over a roller and was
changed with fresh water every 24 h. 2.6 Permeability testing

2.5 Scanning electron microscope (SEM)
Samples were sputter-coated with 5 nm of gold-
nanoparticles using a plasma sputter-coater (Quorum
Q15ORS) and samples were imaged using JEOL JSM-
7401F eld emission scanning electron microscope or Zeiss
EVO HD15 microscope. Test samples for SEM with cells
were xed (in 300 L 0.1 M sodium cacodylate buffer with
Test scaffolds completely covered the 1 cm2 cut-out in the
acrylic holding sheet as part of the main permeability testing
device. (Supplementary Fig. 1AF) The holding sheets
were tightened to create a water-tight chamber. Water
pressure in the device was increased and measured by
C9557 Pressure Meter (Comark). The volume of water
that escaped from the main device through the scaffold was
collected by the beaker.
32 Page 4 of 12
2.7 Burst pressure testing
A dynamic model was assembled to test PCU and L-AME-
PCU tubes using a water bath, pressure monitor, connection
tubes. The test grafts were introduced with a solution with a
blue coloured dye (Supplementary Fig. 1GI)
The 6, 8, and 10 coated tubes were subjected to burst
pressure measurements. Each tube was connected to the
infusion pump on one side and to the pressure monitor on
the other. Blue died water was infused at a rate of 0.2 ml/
min and the rate was subsequently increased up to 50 ml/
min. The pressure inside the tube was recorded simulta-
neously using the pressure monitor.
2.8 Tensile strength
Test polymers were cut into dumbbell shape samples
(Fig. 2b and c, Supplementary Fig 2) according to ISO 37
Type 3. Instron 5565 gripped two ends of the dumbbell at
20 mm, and scaffolds were pulled to their breaking point.
Tensile strengths were subsequently measured by Bluehill
software.
2.9 In vivo study
Tubular scaffolds were excised into 10 mm. 10 Wister rats
were included in this study as was considered optimal to
obtain the experimental clarication, whilst observing
replacement, reduction, renement (3Rs) within the allo-
cated research funds for this study. Five male rats were
implanted with PCU scaffolds (control), and 4 female and 1
male rats with bio-functionalised L-AME-PCU scaffolds.
(Fig. 3, Supplementary Fig 3) This study was approved by
the Regional Committee of Ethics on Animal Experiments
(Paris Descartes University) (registration number
CEEA34.15-021).
The rats were anesthetized with intramuscular injection
of 10 ml of ketamine (50 mg/ml) and 1.5 ml of chlorpro-
mazine (5 mg/ml). Then, they were placed in supine posi-
tion with their heads at the left of the surgeon. Their four
limbs were attached to the table with adhesive elastic bands
(Elastoplast).
A 7 cm median xipho-pubic skin incision was made.
Followed by dissection of muscular and peritoneal layers to
expose the aorta and the vena cava, and the intestines were
mobilised to the left. A further dissection was performed
under a microscope (Carl Zeiss) to free the subrenal
branches of the abdominal aorta. The collaterals between
the renal arteries and the iliac bifurcation were cauterised
(Supplementary Fig 3).
The aorta was clamped with Gilbert double clamp 5 mm
proximal to the iliac bifurcation and distal to the renal
vessels, and 5 mm of the aorta was excised. The lumen of
J Mater Sci: Mater Med (2017) 28:32
the proximal and distal aortic segments was rinsed to pre-
vent thrombi formation (Fig. 3a and b). Rounded 10/0
Ethilon was used to perform end-to-end anastomosis
between the vascular scaffolds and ends of the aorta. The
lumen was rinsed again to ush the air bubbles and the
anastomosis was checked under saline. The clamp was
released distally, and then proximally. Anastomosis was
checked 3 min after unclamping to check for leakage. The
peritoneum and muscular layer were closed with 4/0 Mer-
suture. No anticoagulants were introduced in order to
avoid a bias and reduce confounding. The rats were
immediately warmed and placed in cages. Their weights
were monitored post-surgery.
2.10 Ultrasound evaluation
Prosthetic patency was evaluated by evaluating the ow
proximal to the implanted scaffold. The rats that survived
more than a day were brought for Doppler ultrasound
evaluation. Rats were anesthetised with 4% isourane in air
for induction and maintained using 2.5% isourane. They
were positioned in supine position and their vitals mon-
itored. The abdomen was shaved and then depilated using
depilatory cream. We used VEVO2100 ultrasound ima-
ging system (Visualsonics, Toronto, CA), with a MS250
probe at 18 MHz, to visual blood ow proximal and distal
to the anastomotic sites at 21, and 120 days after the sur-
gery. Pulsed Doppler measurements were performed
upstream and downstream to the implanted scaffold when
possible.
2.11 Histology
Amongst the ve control rats: four died within 1 day fol-
lowing the procedure. one control rat died on day 4 after
grafting (this rat was sacriced due to display of pain and
hind limb immobility). Two test rats were sacriced on day
120 as one of test rats died during surgical procedure, and
one during transport between labs for analysis (but the
explanted graft demonstrated a patent graft). All samples of
of PCU and L-AME-PCU explant scaffold and rat tissues
were xed on with 4% paraformaldehyde then embedded in
parafn and stained with hematoxylin-eosin. PCU and L-
AME-PCU scaffolds were also implanted intraperitoneal
and subcutaneously , which were xed with 4% paraf-
ormaldehyde and embedded in parafn and stained with
hematoxylin-eosin to evaluate the inammatory response.
The histology sample sectioning was met with difcul-
ties in obtaining full and complete sections of the scaffold
attached to live tissues. Due to this damage to the samples
during processing, no quantication analysis could be per-
formed. The interface between the scaffold and tissues were
not embedded, thereby making comparison between the
J Mater Sci: Mater Med (2017) 28:32
Fig. 2 a Mandrels of 1.5 mm (second from left), scaffold on mandrel
(third from left), 6 coats (fourth from right), 8 coats (third from right),
10 coats (second from right) Mandrels with varying diameters can be
used to fabricate grafts with required diameters. d 6 coats of L-AME-
PCU under uorescence microscopy. e 8 coats of L-AME-PCU under
uorescence microscopy. f 10 coats of L-AME-PCU under
samples difcult. A modied protocol, which was used to
study the interface between bones and implants could be
implemented here to strengthen the samples analysis [16].
2.12 Cell cultures
Human bronchial epithelial cells (HBEC) (Caltag Medsys-
tems) were cultured in human bronchial epithelial cell
medium with 1% antibiotic (50 l/ml streptomycin, 50 l/ml
penicillin) and supplements (Caltag Medsystems). HBEC
were seeded onto sterilised 13 mm disks of scaffold at a
density of 5*105 cells per well on 48-well plates. The cell
Page 5 of 12 32
uorescence microscopy. Scaffold in dumbbell shape before elution
(b) after elution demonstrating greater compliance with expanded and
elongated scaffold (c). SEM images of g luminal surface of L-AME-
PCU graft immersed in propanol. h External surface of L-AME-PCU
graft immersed in propanol. i Luminal surface of L-AME-PCU graft
immersed in water
medium in each well was made up to 1 ml and changed
every two days.
2.13 Fluorescence microscopy
The disk was removed on day and xed in 500 l of 4%
formaldehyde in PBS (Gibco) for 1 h at room temperature.
Formaldehyde was aspirated and disks were then washed
with PBS twice. Disks were protein-blocked in 500 l 1%
bovine serum albumin (BSA) with 0.25% triton in PBS for
45 min at room temperature and washed in PBS twice.
Disks were then submerged in 200 l of 0.5% phalloidin
(Sigma) in PBS for 1 h, followed by 100 l of DAPI
32 Page 6 of 12
Fig. 3 Grafts prepared for
anastomosis and placed next to
exposed aorta; Control PCU (a,
c, e) and L-AME-PCU (b, d, f).
Microsurgical end-to-end
anastomosis of the PCU
(control) and functionalised
grafts before (c, e) and after (d,
f) unclamping
(Sigma) for 15 min, at room temperature and in the dark.
Stained cells were visualised using EVOS FL Cell Ima-
ging System (Life Technologies).
2.14 Statistical analysis
DAgostino & Pearson omnibus normality tests were con-
ducted. P < 0.05 was considered statically signicant, and
the null hypothesis that the data is normally distributed is
rejected. Ordinary ANOVA test was conducted for nor-
mally distributed data with Bonferroni post hoc test.
Kruskal-Walli test was conducted for data that is not nor-
mally distributed with Dunns post hoc test.
3 Results
3.1 Macro and micro structure of tubular scaffolds
Mandrel with diameters of 1.5 mm (second from left
Fig. 2a) were used, for in vivo studies. The cross-sections
of 6, 8 and 10 coated tubular scaffold under uorescence
J Mater Sci: Mater Med (2017) 28:32
microscopy (Fig. 2df) were used to measure the average
scaffold wall thickness as 330.25 20.57 m for six coats,
348.70 14.82 m for eight coats, 497.28 49.23 m for
10 coats (Table 1). The scaffold wall thickness increased
with increasing number of coats. Mandrels with a range of
diameters can be used to manufacture these grafts with
ease.
SEM images of luminal and external surface of L-AME-
PCU graft immersed in propanol (Fig. 2g and h) demon-
strated that L-AME did not elute from the scaffold in the
presence of propanol while it was eluted when immersed in
water (Fig. 2i). This proved to be a signicant nding as
immersion of the polymer coated mandrels in water is
required to allow the grafts formed to be detached away
from the mandrels, but substitution in propanol facilitated
this whilst retaining L-AME within the graft. Luminal
surface of L-AME-PCU scaffolds had regular, honeycomb
structure. The fabrication resulted in a transparent structure
for the control and an opaque surface for the functionalised
scaffold.
L-AME concentrations ranging up to (3.75, 7.5, 15, 30
and 60 mg) were tested and found to generate smaller pores.
(results not included) However, there was no signicant
J Mater Sci: Mater Med (2017) 28:32 Page 7 of 12 32

0.191 0.0183
Table 1 The average tube thickness measured using ImageJ for tubes

27.198 3.563
with 6, 8 and 10 coats (n = 9)

374.3 6.372
Number of coats 6 8 10

10 coats
Average tube 330.25 20.57 348.70 14.82 497.28 49.23
thickness (m)

0.167 0.0163
25.177 3.619
314.3 6.197
difference between biomechanical properties measured by

L-AME-P (Porous/after elution)

8 coats
tensile strength, contact angle and cell-material interactions
(both HDFs and HBEC) when visualised with uorescence

Table 2 Tensile strength of control scaffold, L-AME-PCU non-porous scaffold, and L-AME-PCU porous scaffold (where L-AME was eluted in water)
microscopy (results not included).

0.1548 0.00121
22.193 3.219
3.2 Permeability and burst pressure

298.2 5.785
The scaffold was impermeable to de-ionised water at pres-

6 coats
sures of 200 mmHg when it was tested by the permeability
testing device as a at scaffold (Supplementary Fig. 1AF).
The 6, 8 and 10 coated L-AME-PCU tubes resisted burst

0.229 0.0286
47.062 25.662
pressures of 450 mmHg, 700 mmHg and 1028 mmHg

520.67 152.7
respectively. Distension of the tubes was noted, and there
was an increase in their diameters. 2 0.25 cm for the six
coats tube, 1.5 0.15 cm for the eight coats tube and 1 10 coats
0.1 cm for the ten coats tube (Supplementary Fig. 1, Sup-
plementary video).
0.193 0.0211
39.536 19.261
509.51 150.6
L-AME-P (non-Porous/before elution)

3.3 Tensile strength

8 coats

There is a decrease in maximum stress from 72.02 4.138

MPa in the control to 39.53 19.26 MPa in L-AME-PCU
non-porous scaffold and to 25.17 3.619 MPa in L-AME-
0.173 0.0151
34.825 12.152

PCU porous scaffold. Table 2 illustrates a trend on tubes

399.62 134.6

with increasing layers of L-AME incorporated polymer

coatings.
6 coats

3.4 Doppler study

0.057 0.008

Sagittal view of the scaffold showed the inner and outer

72.02 4.138
Control samples

679.3 59.400

border of the tubing, where they are clearly delineated

(Fig. 4 c2). However, the blood ow within the scaffold in
pulsed wave (PW) or colour Doppler was unable to be
assessed due to strong attenuation. However, colour and
PW Doppler assessment of blood ows proximal and distal
to the scaffolds were obtained (Fig. 4 c35). PW Doppler
blood ow velocity measurement distal to bionfunctiona-
Sample thickness (mm)
Maximum stress (MPa)

lised L-AME-PCU scaffold in abdominal aorta showed

Strain at break(%)

adequate blood ow (Fig. 4 c3). While the control PCU

scaffold was non-functional with slowed velocity proles,
distal Doppler showed a suboptimal scaffold with slowed
velocities proles thus suggesting there are collateral
arteries ensuring vascularisation of hind limbs (Fig. 4 c5).
32 Page 8 of 12
Fig. 4 Photographs of representative images of explants of the test
scaffolds, which were placed in the rat model in situ a1 and b1 and a
longitudinal section of the scaffold a2 and b2; Histological sections
illustrating the interface between the scaffold (indicated in dotted line),
with b3, b4, b5, b6, b7 and b8 L-AME, and the aorta, stained with
hematoxylin-eosin, showing, at different magnication (4; 10;
20; 40), the implant (IM), the endothelium of the tunica intima
(Endo), the tunica intima (TI), the tunica media (TM), the tunica
3.5 Histology
All rats in the control group and indicated thrombus for-
mation, and L-AME effect within PCU ensured graft
patency and anti-thrombogenic properties in vivo. The
interface between the implant and the aorta was evaluated,
as well as the degree of endothelialisation of luminal surface
of the scaffolds. An early aorta structure from within the
implant with some endothelial cells in the tunica intima in
sagittal and transversal views were evident with L-AME-
PCU samples (Fig. 4b38).
J Mater Sci: Mater Med (2017) 28:32
adventitia (TA), the blood vessels (BV) and some red blood cells
(RBC). Ultrasound of the test rat 21 days after the implantation in
transversal (c1) and sagittal (c2) view; (c3) Pulsed Wave Doppler
blood ow velocity measurement distal to bionfunctionalised L-AME-
PCU scaffold in abdominal aorta. (c4) Pulsed Wave Doppler blood
ow velocity measurement proximal to biofunctionalised L-AME-
PCU scaffold in abdominal aorta. (c5) Pulsed Wave Doppler blood
ow velocity measurement in distal abdominal aorta, t
The histology analysis of PCU and L-AME-PCU scaffolds
implanted intraperitoneal and subcutaneously, but did not
show any sign of pronounced inammation (Fig. 5ad).
Intraperitoneal PCU and L-AME-PCU scaffolds did not
exhibit any difference. (Fig. 5a and b) The subcutaneous
implant showed a more vascularized interface in sagittal sec-
tion (Fig. 5c and d) suggesting an effect of L-AME on blood
vessel formation. Overall, the results of histology have brought
relevant information about the inammatory response, the
enhanced tissue formation on the implant surface, and the
induced vascularization of L-AME. Graft mismatch is
J Mater Sci: Mater Med (2017) 28:32
Fig. 5 Photographs of the inammatory studies of the PCU-scaffold
implanted intraperitoneally a1 and b1 and subcutaneously c1 and d1 in
the same rat; Histological sections showing the interface between the
scaffold (indicated in dotted line), without b2,b3,d2, d3, and d4 and
demonstrated in Supplementary Fig. 4 showing the difference
in diameter between rat aorta and PCU implant.
3.6 Human bronchial epithelial cells (HBEC)
interactions with functionalised scaffold
Human bronchial epithelial cells were seeded on PCU and
L-AME-PCU scaffolds, where the latter had L-AME eluted.
On day 7, on the control scaffolds, epithelial cells appear to
be round with a small nucleus and cytoplasm. They dis-
played abnormal bronchial epithelial cell morphology and
Page 9 of 12 32
with a2,a3,a4,a5,a6,a7,c2,c3 L-AME, and the aorta, stained with
hematoxylin-eosin, showing, at different magnication (4; 10;
20; 40), the implant (IM), the blood vessels (BV) and some red
blood cells (RBC)
poor cell adhesion (Supplementary Fig 5A). The test sam-
ples with L-AME functionalised scaffolds, had bronchial
epithelial cells with healthy cobblestone morphology and
sheet-like appearance, which demonstrated healthy adhe-
sion and proliferation. (Supplementary Fig. 5BD) SEM of
bronchial epithelial cells, cultured on L-AME-PCU at
scaffolds, demonstrated microivili-like protrusions on cell
membranes. There were healthy HBEC cell morphology
and cell-cell interaction, as well as complete covering of the
scaffold surface indicating good cell adhesion and pro-
liferation after seeding (Supplementary Fig. 5E, F).
32 Page 10 of 12
4 Discussion
This study has demonstrated the applicability of L-AME-
PCU in vivo for the rst time. L-AME has been recognised
for its dual function as a bioactive molecule and a porogen
when incorporated within PCU-based polymer, in a pre-
vious study [10]. This critical rst step of in vivo evaluation
supports its potential wide applications in regenerative
medicine and demonstrated the survival of a cell-free
scaffold. The in vivo model with the choice of end-to-end
anastomosis of a mismatched graft presents a novel view of
conrming material properties, and haemocompatibility
independent of biomechanics of the material or the geo-
metrical inuence of the graft in dynamic, physiological
conditions (Fig. 3).
The relevance of the current model with regards to
endothelialisation as applied to humans is limited as rat
models are recognised for spontaneous endothelialisation.
Multicomponent grafts, which comprise of central graft
segments that avoid tissue in-growth to separate transmural
and haematogenous endothelialisation can be evaluated in
addition to relatively shorter implantation time [17], and
thus we have opted for a maximum 120 days. A rabbit
model could be more appropriate as a small animal model to
evaluate endothelialisation as well as to evaluate a relatively
longer length of vascular graft, if not limited by cost and
strict 3Rs, where the primary question of haemocompat-
ibility can be answered by a murine model as applied here.
Non-human primates, are indeed the closest to human
endothelialisation, but are extremely difcult to obtain with
regards to ethical considerations and associated cost [17
19].
Future studies should encompass a large animal model
with end-to-side anastomosis [20] or with a perfect match of
a diameter with the native vessel for an end-to-end ana-
stomosis and thus to avoid turbulence at the site of ana-
stomosis, which would form a potential pre-clinical study of
this promising graft as a vascular substitute.
The functionalised graft following L-AME elution
adopts a porous structure and therefore despite using sol-
vent evaporation, method for graft fabrication offers sui-
table compliance to the graft, closely matching the foamy
structure resulting from solvent exchange method of fabri-
cation for elastomeric polymers [21]. The functionalised
grafts were optimal for suturability and gains more soft and
elastic material properties following L-AME elution
(Fig. 2b and c) yet remaining uid impermeable. Non-
functionalised grafts, however retain a relatively rigid
structure with solvent cast fabrication and is not complaint
with native vessels (Fig. 4 c5). Scaffolds for tissue regen-
eration or replacement should ideally have sufcient
mechanical strength that match native tissue and withstand
physiological stress [22]. Supporting our in vitro studies of
J Mater Sci: Mater Med (2017) 28:32
permeability (Supplementary Fig. 1, Supplementary Video)
of the functionalised, porous ultrasmall diameter graft,
which indicated to resist pressures up to 1028 mmHg,
remained patent within a major vessel such as the aorta in
the rat model. L-AME-PCU tubular scaffolds have shown a
decrease in maximum stress from 72.02 4.138 MPa in the
control to 39.53 19.26 MPa in L-AME-PCU non-porous
scaffold and to 25.17 3.619 MPa in L-AME-PCU porous
scaffold. This trend was observed in our previous study,
which was performed on at sheets [10]. However, the
tensile strength of human aorta, which is 131 56 kPa [23],
can be comfortably managed by PCU and L-AME-PCU
with much higher tensile strength enabling it to withstand
physiological forces.
In a previous study of using elastomers with solvent
casting fabrication, the grafts (non-porous) resisted high
pressures up to 1413.66 134.33 mmHg, while in its coa-
gulated (porous) form, they resisted 215.67 19.81 mmHg
[24]. Therefore, this study presents a more rapid, ultra-
simple fabrication method of inducing pores with a single-
step where the porogen itself imparts favourable biological
properties, compared to sodium bicarbonate. Sodium
chloride which can be used as porogen, in elastomer fab-
rication, but need to be completely removed prior to
implantation [24]. The link between porosity with regards to
protein and cell interactions are complex and it has been
shown that high porosity materials (>45 m) can facilitate
brovacular inltration and lead to reduction in graft
compliance, leading to failure [17]. The method of material
functionalisation, which is introduced here can be applied to
pre-polymer systems, where it can be applied as a surface
coating for metallic stents, or even 3D printed scaffolds thus
imparting biofunctionality and surface topography [24, 25].
The graft fabrication, which result in non-permeable
scaffold has potentially wider applications in repairing or
replacing tubular structures that include oesophagus, sto-
mach and urethral [2629] tissue, where impermeable
scaffolds are paramount to prevent luminal content leakage,
and a surface porous topography to support cell-materials
interactions to ensure patency [9]. A previous study showed
improved endothelial progenitor cell (EPC) viability with
increased concentration of L-AME incorporated PCU
scaffold due to porous, honeycomb scaffold for both
broblasts and EPCs [10]. The present study has shown that
there is a good interaction with human bronchial epithelial
cells on functionalised scaffolds (Supplementary Fig. 5).
Histology analysis showed early aorta formation in
luminal surface of the L-AME-PCU scaffolds (Fig. 4). This
early aorta consists of tunica intima, media and adventitia as
well as an endothelial layer, where L-AME increased the
viability of endothelial progenitor cells in vivo. Angiogen-
esis is crucial to providing nutrients to and removing waste
products from the cells on the scaffold to ensure survival of
J Mater Sci: Mater Med (2017) 28:32
the implanted scaffolds and regeneration of the surrounding
tissue [4, 5]. The histology study provides good evidence of
an enhanced angiogenic response with increased vascular
formation towards the scaffold externally and internally in
contact with the aorta and in subcutaneous environment. In
both cases the blood vessel formation was higher in contact
with L-AME-PCU than with control PCU, suggesting an
increased angiogenic response due to the bioactive porogen,
L-AME. Inammation modulated by VEGF and nitric oxide
plays an important role in extravascularisation of inam-
matory mediators that can induce angiognesis [6, 30].
Therefore, conrming a role for L-AME functionalised
scaffold in angiogenesis through possible NO production.
The histology of the samples from those sutured to the aorta
and subcutaneously, demonstrated no visibly pronounced
inammatory response (Figs. 4 and 5). However PCR/
ELISA could be potentially used to clarify the presence of
inammatory response [6]. The scaffolds placed in the
peritoneum have exhibited no chronic inammation or
brous capsule formation, suggesting a good biocompat-
ibility of the scaffold within the body. However, it is sug-
gested that a weak inammatory response that subsides may
be benecial for better angiogenesis response since some
mediators of inammation are able to induce angiogenesis
[6].
Biodegradable scaffolds are widely used in the eld of
tissue engineering that include PGA [31], PCL [8, 27].
However, with rapid degradation rates such with PGA
scaffolds their stability is dependent on functional tissue
formation. Furthermore, the long culture periods between
two and six months in biodegradable scaffolds and the poor
regenerative capacity of cells in old patients make the
applications of biodegradable scaffolds less feasible in
emergencies and geriatrics [32]. Hence, the biostability and
impermeable nature of PCU-based functionalised scaffolds
as studied here can be desirable for surface coatings, and
potential off-the-shelf applications in regenerative medicine.
Acknowledgements Authors would like to thank Arnold Darbyshire
at UCL for synthesising and providing with polyurethane for this study
and Nathalie Poitevin and Jean Luc Vignes at Ecole de Chirurgie, Paris
for their support in the microsurgical laboratory. ME is sponsored by
University of Alexandria, Egypt and JH is sponsored by Globe
Microsystems Ltd UK.
Compliance with ethical standards
Conict of interest The authors declare that they have no com-
peting interests.
Page 11 of 12 32
References
1. Jen MC, Serrano MC, van Lith R, Ameer GA. Polymer-based
nitric oxide therapies: recent insights for biomedical applications
Michele. Adv Funct Mater. 2012;22:23960.
2. Kim TG, Shin H, Lim DW. Biomimetic scaffolds for tissue
engineering. Adv Funct Mater [Internet]. 2012;22:244668.
doi:10.1002/adfm.201103083.
3. Zorlutuna P, Vrana NE, Khademhosseini A. The expanding world
of tissue engineering: the building blocks and new applications of
tissue engineered constructs. IEEE Rev Biomed Eng.
2013;6:4762.
4. Briquez PS, Clegg LE, Martino MM, Gabhann FM, Hubbell JA.
Design principles for therapeutic angiogenic materials. Nat Rev
Mater [Internet]. 2016;1:15006 Available from http://www.nature.
com/articles/natrevmats20156.
5. Novosel EC, Kleinhans C, Kluger PJ. Vascularization is the key
challenge in tissue engineering. Adv Drug Deliv Rev.
2011;63:30011. Available from http://www.ncbi.nlm.nih.gov/
pubmed/21396416.
6. Oates M, Chen R, Duncan M, Hunt JA. The angiogenic potential
of three-dimensional open porous synthetic matrix materials.
Biomaterials. 2007;28:367986.
7. Geesala R, Bar N, Dhoke NR, Basak P, Das A. Porous polymer
scaffold for on-site delivery of stem cells - Protects from oxidative
stress and potentiates wound tissue repair. Biomaterials [Internet].
2016;77:113. doi:10.1016/j.biomaterials.2015.11.003.
8. Guarino V, Causa F, Ambrosio L. Porosity and mechanical
properties relationship in PCL porous scaffolds. J Appl Biomater
Biomech. 2007;5:14957.
9. Loh QL, Choong C. Three-dimensional scaffolds for tissue
engineering applications: role of porosity and pore size. Tissue
Eng Part B Rev [Internet]. 2013;19:485502. Available from
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=
3826579&tool=pmcentrez&rendertype=abstract.
10. Everett W, Scurr DJ, Rammou A, Darbyshire A, Hamilton G, de
Mel A. A material conferring hemocompatibility. Sci Rep
[Internet]. 2016;6:26848. Available from http://www.ncbi.nlm.
nih.gov/pubmed/27264087%0Ahttp://www.pubmedcentral.nih.
gov/articlerender.fcgi?artid=PMC4893622.
11. de Mel A, Murad F, Seifalian AM. Nitric oxide: a guardian for
vascular grafts? Chem Rev [Internet]. 2011;111:574267. doi:10.
1021/cr200008n.
12. Ockelford F, Saada L, Mel EG and de A Is nitric oxide assuming a
janus-face in the central nervous system? Curr Med Chem. 2016
May 27;23(16):162537 Available from: http://www.eurekaselect.
com/node/140453/article.
13. Akter F, Coghlan G, de Mel A. Nitric oxide in paediatric
respiratory disorders: novel interventions to address associated
vascular phenomena? Ther Adv Cardiovasc Dis [Internet].
2016;10:25670. Available from http://tak.sagepub.com/content/
10/4/256.abstract
14. Iannaccone PM, Jacob HJ. Rats! Dis. Model. &amp;amp; Mech.
[Internet]. 2009;2:206 LP-210. Available from: http://dmm.
biologists.org/content/2/5-6/206.abstract.
15. Haruguchi H, Teraoka S. Intimal hyperplasia and hemodynamic
factors in arterial bypass and arteriovenous grafts: a review. J Artif
Organs [Internet]. 2003;6:22735. Available from doi:10.1007/
s10047-003-0232-x.
16. Guillot R, Pignot-Paintrand I, Lavaud J, Decambron A, Bourgeois
E, Josserand V, et al. Assessment of a polyelectrolyte multilayer
lm coating loaded with BMP-2 on titanium and PEEK implants
in the rabbit femoral condyle. Acta Biomater [Internet] Acta
Materialia Inc. 2016;36:31022. doi:10.1016/j.actbio.2016.03.
010.
32 Page 12 of 12
17. Byrom M, Bannon P, White G, Ng MKC. Animal models for the
assessment of novel vascular conduits. J Vasc Surg.
2010;52:17695.
18. Lepidi S, Grego F, Vindigni V, Zavan B, Tonello C, Deriu GP,
et al. Hyaluronan biodegradable scaffold for small-caliber artery
grafting: preliminary results in an animal model. Eur J Vasc
Endovasc Surg. 2006;32:4117.
19. Swartz DD, Andreadis ST. Animal models for vascular tissue-
engineering. Curr Opin Biotechnol. 2013;24:91625.
20. Koobatian MT, Row S, Smith RJ Jr, Koenigsknecht C, Andreadis
ST, Swartz DD. Successful endothelialization and remodeling of a
cell-free small-diameter arterial graft in a large animal model.
Biomaterials. 2016;76:34458.
21. Ghanbari H, de Mel A, Seifalian AM. Cardiovascular
application of polyhedral oligomeric silsesquioxane nanomater-
ials: a glimpse into prospective horizons. Int J Nanomedicine.
2011;6:77586.
22. Leong KF, Chua CK, Sudarmadji N, Yeong WY. Engineering
functionally graded tissue engineering scaffolds. J Mech Behav
Biomed Mater. 2008;1:14052.
23. Sommer G, Sherifova S, Oberwalder PJ, Dapunt OE, Ursomanno
PA, DeAnda A, et al. Mechanical strength of aneurysmatic and
dissected human thoracic aortas at different shear loading modes. J
Biomech. 2016;49:237482.
24. de Mel A, Yap T, Cittadella G, Hale LR, Maghsoudlou P, de
Coppi P, et al. A potential platform for developing 3D tubular
J Mater Sci: Mater Med (2017) 28:32
scaffolds for paediatric organ development. J Mater Sci Mater
Med. 2015;26:18.
25. de Mel A, Punshon G, Ramesh B, Sarkar S, Darbyshire A,
Hamilton G, et al. In situ endothelialisation potential of a bio-
functionalised nanocomposite biomaterial-based small diameter
bypass graft. Biomed Mater Eng. 2009;19:31731.
26. Fu W, Wang Z, Li G, Zhang B, Zhang L, Hu K. A surface-
modied biodegradable urethral scaffold seeded with urethral
epithelial cells. Chin Med J. 2011;124:308792.
27. Kuppan P, Sethuraman S, Krishnan U. PCL and PCL-gelatin
nanobers as esophageal tissue scaffolds: optimization, char-
acterization and cell-matrix interactions. J Biomed Nanotechnol.
2013;9:154055.
28. Lv J, Chen L, Zhu Y, Hou L, Liu Y. Promoting epithelium regen-
eration for esophageal tissue engineering through basement mem-
brane reconstitution. ACS Appl Mater Interfaces. 2014;6:495464.
29. Maemura T, Kinoshita M, Shin M, Miyazaki H, Tsujimoto H,
Ono S, et al. Assessment of a tissue-engineered gastric wall patch
in a rat model. Artif Organs. 2012;36:40917.
30. Arroyo AG, Iruela-Arispe ML. Extracellular matrix, inammation,
and the angiogenic response. Cardiovasc Res. 2010;86:22635.
31. Atala A, Bauer SB, Soker S, Yoo JJ, Retik AB. Tissue-engineered
autologous bladders for patients needing cystoplasty. Lancet.
2006;367:12416.
32. Ravi S, Chaikof E. Biomaterials for vascular tissue engineering.
Regen Med. 2010;5:121.