journal of the mechanical behavior of biomedical materials 14 (2012) 130 –142

Available online at www.sciencedirect.com

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

Research Paper

Investigation of a small-diameter decellularised artery as a

potential scaffold for vascular tissue engineering;
biomechanical evaluation and preliminary cell seeding

E.M. Campbella, P.A. Cahillb, C. Lallya,n

a
School of Mechanical & Manufacturing Engineering, Dublin City University, Glasnevin, Dublin 9, Ireland
b
School of Biotechnology, Dublin City University, Glasnevin, Dublin 9, Ireland

ar t ic l e in f o abs tra ct

Article history: The development of a small-diameter tissue engineered blood vessel (TEBV), with equivalent
Received 4 April 2012 mechanical properties to the vessel being replaced, may provide a potential solution to the
Received in revised form limitations associated with natural and synthetic bypass grafts of small-diameter vessels.
30 May 2012 This study presents the biomechanical properties of small-diameter (o4 mm) porcine
Accepted 5 June 2012 coronary arteries (PCA) and the corresponding natural matrix scaffold of the artery achieved
Available online 15 June 2012 through short-term decellularisation. Tubular segments, up to 50 mm in length, of PCA were

Keywords: perfused with 0.1 M sodium hydroxide (NaOH) for 3 to 12 h to achieve the natural matrix

Tissue engineering scaffold. Uniaxial tensile, inflation and permeability tests were performed on non-decellu-

Vascular scaffold larised and decellularised sections within 24 h of slaughter to determine the alteration in

Decellularised artery mechanical properties as a result of decellularisation. A treatment time of 9 h achieved

Biomechanics decellularisation as all cell nuclei were appropriately disrupted and there was an absence of
smooth muscle in the vascular wall. Uniaxial tensile and inflation tests confirmed the
scaffold maintains its non-linear response, however a less stiff, more distensible low-load
response and stiffer high-load response was found compared to non-decellularised sections.
Vascular smooth muscle cells were successfully seeded to the lumen, abluminal side and
lateral edges of decellularised sections and attachment and infiltration of the xenogeneic
cells after 15 days confirmed the viability of the PCA scaffold as a suitable environment for
cell growth and infiltration. An extended decellularisation treatment time increased the
porosity whilst maintaining the mechanical integrity of the scaffold and this may optimise
the repopulation of the scaffold. This study provides valuable information for the develop-
ment of an optimum TEBV, while also establishing the potential of this natural matrix
scaffold to be used as a graft or vascular tissue engineering scaffold.
& 2012 Elsevier Ltd. All rights reserved.

1. Introduction results in a major economic burden (Roger et al., 2011).

Autologous blood vessel grafts or synthetic grafts can be
Cardiovascular disease remains the principle cause of mor- successfully used to bypass large-diameter diseased vessels,
tality and morbidity worldwide; the treatment of which however, for small-diameter (o6 mm) applications, such as

n
Corresponding author. Tel.:þ353 1 700 7608; fax:þ353 1 700 7148.
E-mail addresses: caitriona.lally@dcu.ie, triona.lally@dcu.ie (C. Lally).

1751-6161/$ - see front matter & 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jmbbm.2012.06.001
 journal of the mechanical behavior of biomedical materials 14 (2012) 130 –142
coronary artery bypass grafting (CABG), a mismatch between
the biomechanical properties of the graft and the native
vessel can compromise the long-term patency of synthetic
vascular bypass grafts making them less than ideal substi-
tutes (Conklin et al., 2002; Salacinski et al., 2001). The
development of a small-diameter bypass graft with corre-
sponding biomechanical properties may overcome this lim-
itation and improve the long-term patency rates for small-
diameter applications.
Tissue engineered blood vessel (TEBV) scaffolds prepared
from the natural arterial scaffold may provide a solution,
provided that the native biomechanical properties are retained.
In an effort to produce TEBVs that match the structural,
chemical, and mechanical properties of the native vessel,
allogeneic and xenogeneic blood vessels have been decellu-
larised using a range of different techniques creating scaffolds
for the migration and growth of native endothelial cells and
smooth muscle cells (Allaire et al., 1997; Conklin et al., 2002;
Dahan et al., 2012; Sheridan et al., 2012; Teebken et al., 2000;
Wilshaw et al., 2011). For complete decellularisation, the ECM
must be appropriately disrupted to allow for access to all cells
and to provide a path for their subsequent removal while
minimising the disruption in order to retain the native ECM
mechanical and biological properties (Gilbert et al., 2006).
Many different animals (Allaire et al., 1997; Chow and
Zhang, 2011; Conklin et al., 2002; Dahan et al., 2012; Goissis
et al., 2000; Liu et al., 2004; Schaner et al., 2004; Sheridan
et al., 2012; Williams et al., 2009; Zhao et al., 2010) have been
used to determine the viability of decellularisation methods
on a variety of blood vessels including the aorta (Goissis et al.,
2000; Liu et al., 2004), carotid artery (Conklin et al., 2002;
Dahan et al., 2012; Sheridan et al., 2012; Williams et al., 2009;
Zhao et al., 2010), iliac artery (Goissis et al., 2000), saphenous
vein (Schaner et al., 2004) and iliac vein (Goissis et al., 2000)
for the development of a TEBV. Combinations of chemical,
enzymatic and physical treatments have been used (Gilbert
et al., 2006) and the minimum decellularisation time reported
has been 4 h for rat aorta where agitation in NaOH for 1 h
followed by 3 h of vigorous agitation in PBS achieved rapid
decellularisation (Liu et al., 2004). The majority of protocols,
however, take from several days (Goissis et al., 2000; Sheridan
et al., 2012; Zhao et al., 2010) to weeks (Dahan et al., 2012;
Williams et al., 2009). The most commonly used methods
involve an enzymatic process, such as Trypsin digestion,
followed by immersion in chemical detergents, such as
TritonX-100 and ammonium hydroxide (Amiel et al., 2006;
Williams et al., 2009; Sheridan et al., 2012; Dahan et al., 2012).
Other studies have shown that the mechanical properties
of biological materials alter dramatically with time once
harvested from the body (Chow and Zhang, 2011; Stemper
et al., 2007). Therefore the longer the decellularisation time
the greater the natural degradation of the tissue which will
have an effect on the mechanical properties of the resulting
scaffold. Consequently, it is critical to establish a short-term
decellularisation method to retain the native mechanical
properties and architecture aiding the development of a
potentially viable TEBV. Lengthy decellularisation times may
also reduce the clinical and commercial viability of such
TEBVs given the high costs associated with prolonged culture
periods.
131
CABG is the most common bypass surgery performed,
however test data on the mechanical properties of coronary
vessels is limited (van Andel et al., 2003). Clearly, the devel-
opment of bypass grafts from coronary arteries would be
highly advantageous because the vessel size and compliance
would match that of the host vessel. Due to the small size of
coronary vessels, their numerous branches and their limited
accessibility no study has been carried out to-date to exam-
ine decellularisation of human or porcine coronary arteries
and the resulting effect on their mechanical properties. A
comprehensive understanding of the biomechanical proper-
ties of non-decellularised coronary arteries and the corre-
sponding natural matrix scaffold is necessary therefore since
it provides vital insights into the ideal biomechanical proper-
ties for TEBVs and TEBV scaffolds. The objective of this study
is to obtain the natural matrix scaffold of coronary arteries
using a short-term decellularisation protocol to minimise
alteration in mechanical properties. A thorough examination
of the mechanical properties and permeability of non-decel-
lularised and decellularised porcine coronary arteries (PCA) is
conducted and the cell attachment capabilities of the corre-
sponding natural matrix scaffold ascertained to determine its
potential as a TEBV scaffold.
2. Materials and methods
2.1. Tissue preparation
Hearts from pigs aged 6 months and weighing approximately
90 kg were obtained within 1 h of slaughter and transported
in 0.9% saline at room temperature. The small-diameter
(internal diameter of o4 mm) PCA were excised immediately.
Segments, up to 50 mm in length, of the left anterior
descending (LAD), left circumflex (LCX) and right (RCA)
coronary arteries were excised using standard dissection
tools, with the branches ligated where required. The coronary
arteries excised had similar thicknesses and internal dia-
meters and sections of PCA were only used where the internal
diameter was greater than 1.5 mm (see Table 3). Segments
were stored in 0.9% saline at room temperature until further
use. Decellularisation and mechanical tests were performed
within 24 h of slaughter to minimise the effect of degradation
as a result of harvesting.
2.2. Decellularisation
A chemical decellularisation protocol by Liu et al. (2004) was
modified for tubular segments of PCA, where 0.1 M sodium
hydroxide (NaOH) was used to disrupt the cells for time
periods of 3, 6, 9 and 12 h followed by 0.9% saline for 3 h to
remove chemical residue and cellular debris. Preliminary
tests of flat and short cylindrical sections of PCA were carried
out by immersion under constant agitation using a magnetic
stirrer. As segments up to 40 mm in length were required for
biomechanical tests, a perfusion system was developed
similar to that of Montoya and McFetridge (2009), to improve
the infiltration of 0.1 M NaOH along the length of the artery.
Natural polypropylene barbs were inserted into the arterial
lumen at either end and mounted horizontally in a flow
132 journal of the mechanical behavior of biomedical materials 14 (2012) 130 –142
chamber. A fluid reservoir at a height of 100 mmHg contain-
ing either 0.1 M NaOH or 0.9% saline was frequently replen-
ished. A peristaltic pump circulated the fluid at a frequency of
2 Hz and a pressure transducer inserted into the lumen of the
vessel monitored the pressure.
2.3. Histological and immunohistochemical analysis
Sections of PCA were processed for staining using standard
methodologies. Hematoxylin and eosin (H&E) stained for
nuclear material, elastin-van Gieson stained for elastin, and
Masson’s Trichrome stained for collagen. Briefly, ring seg-
ments, 2 mm in length, were fixed in 10% formalin overnight,
dehydrated in increasing concentrations of ethanol, cleared
in xylene solution, embedded in paraffin, sectioned into 5 mm
thick sections, and stained. Sections were also stained with a
1/1000 dilution of 40 ,6-diamidino-2-phenylindole (DAPI) for
1 min to image nuclear material. Images were obtained using
a fluorescent microscope (Olympus BX-51) with integrated
Olympus DP30BW and Olympus DP50 cameras.
2.4. Mechanical analysis
Mechanical tests were performed on non-decellularised PCA,
decellularised PCA perfused for 6 h with 0.1 M NaOH and 0.9%
saline for 3 h, and non-decellularised PCA stored in 0.9%
saline for 9 h to correspond with the time required for
decellularisation. Unsuccessful tests were a result of unsui-
table diameters or lengths of PCA and branches located in the
Fig. 1 – (A) Inflation test rig where artery is mounted in the flow chamber and a precise pre-stretch applied by the linear
micrometer. (B) Permeability rig.
test region. All tests were performed within 24 h of slaughter
at room temperature.
2.4.1. Uniaxial tensile tests
Uniaxial tensile tests in the longitudinal direction were per-
formed to determine the stress–strain response of non-decel-
lularised PCA and the corresponding change in the response as
a result of decellularisation. 26 RCAs were divided into seg-
ments 20 mm in length, opened in the longitudinal direction
and cut into dog-bone shaped sections, 3.7 mm in width and
with a gauge length of 12 mm, using a custom-made die. Non-
decellularised (n¼ 25), 9 h non-decellularised (n¼ 18), 9 h decel-
lularised (n¼19) and 15 h decellularised (n¼ 6) sections were
tested using a Zwick Z005 displacement controlled tensile
testing machine with a 20 N load cell. Pre-conditioning was
performed to a maximum force of 1 N over 5 cycles and the
sections were then tested to failure at a strain rate of 60%/min
of the initial gauge length at a pre-load of 0.001 N.
2.4.2. Inflation tests
Inflation tests were performed to determine the response of
intact PCA under physiological pressures and the correspond-
ing change as a result of decellularisation. Seven segments of
LAD, approximately 40 mm in length and with all branches
ligated, were cannulated at both ends and mounted horizon-
tally into a custom-made flow chamber (Fig. 1(A)). A pre-stretch
of 30% was applied using a linear micrometer attached to the
flow chamber to imitate in vivo conditions for PCA (van Andel
et al., 2003). The flow chamber was connected to a pressure
 journal of the mechanical behavior of biomedical materials 14 (2012) 130 –142
head of 0.9% saline, the height of which was altered using a
pulley system. The luminal pressure was measured using a
paediatric pressure catheter (Abbott, IL, USA) connected to a
pressure transducer (DTXPlus, BD, NJ, USA), amplifier and
monitoring device (TA-100, CWE, PA, USA). After five pre-
conditioning cycles of 10–200 mmHg the LAD was inflated from
10–200 mmHg in increments of 10 mmHg. At each step a video
extensometer (MESSPHYSIK Material Testing, Austria) imme-
diately recorded an image. The LAD was decellularised over 9 h
and the test was repeated. A vessel was discarded if it leaked
during either test. Image analysis at each time step resulted in
multiple measurements of the external diameter along the
length of the LAD.
2.4.3. Permeability tests
Permeability tests were performed to determine the change as a
result of decellularisation and also to determine whether decel-
lularised PCA can maintain the ability to withstand high pres-
sures. Twenty-one LCXs were divided into segments, each 15 mm
in length, and were tested non-decellularised (n¼17), 9 h non-
decellularised (n¼ 10) or 9 h decellularised (n¼ 11). The segments
were opened in the longitudinal direction and 8 mm circular
sections were obtained using a custom-made die. The section
was placed in a well in the bottom plate of a custom-built
permeability rig obstructing a 4 mm central hole (Fig. 1(B)). The
top plate and a spacer were secured to the bottom plate by four
screws. A water-filled inflation device (Basix 25, Merit Medical
Systems, USA) connected to the top plate increased the pressure
to 1500 mmHg for 60 min. A petri dish under the bottom plate
Fig. 2 – Non-decellularised tissue (0- and 9-h) and decellularised tissue (6-, 9-, 12-, and 15-h) stained with H&E, van Gieson,
Masson’s Trichrome, and DAPI. Scale bar: 200 lm.
133
collected fluid perfused through the section throughout the test.
Permeability was calculated using Darcy’s Law:
QLm
k¼ ð1Þ
ADP
where, k is the permeability coefficient (m2), Q is flow rate (m3/s),
L is tissue thickness (m), m is viscosity of water (Pa s), A is surface
area (m2) and DP is pressure drop across the material.
2.4.4. Dimensional analysis
The removal of cells from the vascular wall and the potential
damage to the ECM may alter the dimensions of the vessel
when compared to fresh material. Fourteen non-decellularised
sections of PCA and 14 corresponding 9 h decellularised sec-
tions of PCA were obtained, each 2 mm in length. Images of the
cross-section were taken under a microscope with an inte-
grated camera. The sections were cut in the longitudinal
direction and a second image was immediately taken. The
images were analysed for measurements of thickness, internal
and external circumference, and opening angle using ImageJ
1.42 software (National Institutes of Health, Bethesda, MD).
2.5. Cell seeding
To evaluate the suitability of decellularised PCA for cell
attachment bovine aortic smooth muscle cells (BASMC,
Coriell AG08504, NJ, USA) were seeded either to the lumen
and lateral edges or to the abluminal side and lateral edges
of 5  5 mm flat segments of 9 h decellularised and 15 h
134 journal of the mechanical behavior of biomedical materials 14 (2012) 130 –142

decellularised PCA at a concentration of 1  105 cells/cm2
for 72 h and at a concentration of 4  104 cells/cm2 for 15
day in a 24-well plate. Cells were routinely cultured
using RPMI 1640 media (Sigma R8758) supplemented with
10% fetal bovine serum (Sigma F9665) and 1% Penicillin–
Streptomycin (Sigma P4333) without the addition of any
adhesion molecules. The 24-well plate was placed in an
incubator in a humidified atmosphere of 95% air and 5% CO2
at a temperature of 37 1C. The samples were repeatedly washed
in PBS to ensure that any non-adhering cells were removed
from the surface prior to fixing in 10% formalin for H&E and
DAPI analysis.

Fig. 3 – (A) Representative stress–strain response for non-decellularised, 9 h non-decellularised and 9 h decellularised sections
of RCA. (B) Determination of initial slope, mi, pre-conditioned slope, mp, final slope, mf, and x-intercepts for mp and mf.
(C) Tensile data for non-decellularised, 9 h non-decellularised, 9 h decellularised and 15 h decellularised sections of RCA
Significance, po0.05.
 journal of the mechanical behavior of biomedical materials 14 (2012) 130 –142 135

Fig. 4 – Non-decellularised (~) and 9 h decellularised (m) pressure-external diameter response for inflation tests of LAD.
Determination of initial slope, mi, physiological slope, mphy, and final slope, mf.
136 journal of the mechanical behavior of biomedical materials 14 (2012) 130 –142

Bovine aortic endothelial cells (BAEC) were seeded to the
lumen of 5  5 mm flat segments of 15 h decellularised PCA at
a concentration of 2  104 cells/cm2 for 10 day in a 96-well
plate. The segments were maintained in an incubator and
processed for analysis under the same conditions as BASMC-
seeded sections.
unequal variation, and t-tests to ascertain the p-value were
performed on data sets. A p value o0.05 was considered
significant. In some instances the removal of outliers was
required to achieve normality. Outliers were defined as data
points outside of the mean72SD.

3. Results
2.6. Statistics
3.1. Decellularisation
All data is presented as mean7standard deviation (SD). Error
bars indicate one standard deviation. D’Agostino and Pearson A combination of stains were used to determine the presence
omnibus K2 normality tests, F-tests to determine equal or or absence of cells from the vascular wall (Fig. 2(A)–(X)). The

Table 1 – Non-decellularised and 9 h decellularised pressure-external diameter data analysis from inflation tests of LAD.

Initial slope, Physiological slope, Final slope, Initial external diameter (mm)
mi (mm/mmHg) mphy (mm/mmHg) mf (mm/mmHg)

Non-decellularised, n ¼ 7 0.024470.0099 0.003270.0006 0.001770.0008 3.8070.47

9 h decellularised, n ¼ 7 0.042370.0083 0.001370.0007 0.000770.0003 4.3370.57

n
Significance, po0.05.

Fig. 5 – The (A) non-decellularised and (B) corresponding 9 h decellularised pressure-diametrical stretch response for
inflation tests of LAD. (C) Non-decellularised (~) and 9 h decellularised (m) response for specimen # 4 indicating analysis of
the initial (mi) and final slopes (mf).
 journal of the mechanical behavior of biomedical materials 14 (2012) 130 –142 137

compact medial layer and the looser adventitial layer of the
vascular wall can be clearly distinguished in the images. Nuclei
are clearly visible throughout the vascular wall of non-decel-
lularised and 9 h non-decellularised PCA in both the H&E and
DAPI stains (Fig. 2(A), (B) and (S), (T)). While there is a degree of
cell disruption, intact cells were still visible after 6 h decellu-
larisation (Fig. 2(C) and (U)). A decellularisation time of 9 h or
more was found to disrupt all cell nuclei (Fig. 2(D)–(F) and
(V)–(X)). As a result of 0.1 M NaOH treatment there is a loosen-
ing of both medial and adventitial layers with increasing effect
corresponding to increased treatment times. After 15 h decel-
lularisation large pores are visible in both the media and
adventitia; in some instances the two layers separate from
each other. Van Gieson stain and Masson’s Trichrome stain
highlight the presence of elastin and collagen, respectively,
throughout the vascular wall which are visible at all stages of
treatment (Fig. 2(G)–(R)). Smooth muscle and cytoplasm stain
red with Masson’s Trichrome staining (Fig. 2(M)–(R)). While
smooth muscle and cytoplasm are visible in the media of non-
decellularised sections, there is partial removal after 6 h treat-
ment and total removal after 9 h of treatment onwards.
3.2. Mechanical analysis
diameter at all recorded pressures compared to non-decel-
lularised LAD. The external diameter-pressure curves were
analysed for initial, physiological and final slopes (Fig. 4 and
Table 1). Statistical significance was found in the slopes for
9 h decellularised LAD in comparison to non-decellularised
LAD at all stages. The pressure-diametrical stretch curves
show that 9 h decellularised LAD is initially less stiff when
compared to non-decellularised LAD (Fig. 5 and Table 2).
However in all but one instance, as the pressure increased
the 9 h decellularised LAD exhibited stiffer behaviour than
non-decellularised LAD.
3.2.3. Permeability tests
A statistically significant increase in the permeability of 9 h
decellularised sections was exhibited in comparison to both
non-decellularised sections and 9 h non-decellularised sec-
tions (Fig. 6). There was no statistically significant difference
in the permeability of non-decellularised LCX and 9 h non-
decellularised LCX. There was large variability in the perme-
ability of the 9 h decellularised segments which was not
present in the non-decellularised segments. All test groups
withstood pressures up to 1500 mmHg.

3.2.1. Uniaxial tensile tests 3.2.4. Dimensional analysis

All sections exhibited the typical non-linear J-shaped curve Results for thickness, internal circumference, external cir-
associated with soft tissues for representative stress-strain cumference and opening angle are presented in Table 3.
responses (Fig. 3(A)). The variability in the elastic behaviour There is a statistically significant decrease in the thickness of
exhibited between hearts was evident for all groupings. The
slope of the curve was analysed in three regions; the initial
Table 2 – Non-decellularised and 9 h decellularised
slope, the pre-conditioned slope, and the final slope; by a linear
pressure-diametrical stretch data analysis from inflation
fit of the data in that region as well as the x-intercepts for the tests of LAD.
pre-conditioned and final slopes (Fig. 3(B)). The 9 h decellu-
larised sections exhibited less stiff behaviour in the initial Initial slope, Final slope,
mi (mmHg) mf (mmHg)
linear region but statistically significant stiffer behaviour in
both the pre-conditioned and final regions of the curve when Non-decellularised, n ¼7 167.49745.66 2087.997584.37
compared to non-decellularised and 9 h non-decellularised 9 h decellularised, n¼ 7 100.77722.82 5769.3871187.68
sections (Fig. 3(C)). The x-intercepts for the pre-conditioned
n
and final slopes occur at a statistically significant higher strain Significance, po0.05.

for 9 h decellularised sections than for non-decellularised or

9 h non-decellularised sections. Furthermore, after pre-condi-
tioning the gauge length of the decellularised sections does not
extend to the same degree as the non-decellularised and 9 h
non-decellularised sections. The non-decellularised and 9 h
non-decellularised sections exhibit no significant difference
at any stage of the stress-strain response. After 15 h decellu-
larisation the same trends were exhibited in comparison to
non-decellularised sections however there was no significant
increase in the change in mechanical properties found as a
result of a prolonged treatment time when compared to the 9 h
decellularised sections (Fig. 3(C)).

3.2.2. Inflation tests

The external diameter-pressure curves for each LAD demon-
strate that non-decellularised and decellularised arteries
exhibit initial and final linear regions linked by a non-linear
region (Fig. 4). The external diameter at the initial reading of
10 mmHg increased from 3.8070.47 to 4.3370.57 mm after Fig. 6 – Permeability results of non-decellularised, 9 h non-
9 h decellularisation however this was not statistically sig- decellularised and 9 h decellularised sections of LCX.
nificant. 9 h decellularised LAD experience a larger external Significance, npo0.05.
138 journal of the mechanical behavior of biomedical materials 14 (2012) 130 –142

9 h decellularised PCA in comparison to non-decellularised
sections. The internal diameter of 9 h decellularised PCA
increased in comparison to non-decellularised PCA however
there is no trend found for the external circumference. There
was a statistically significant decrease in the opening angle for
9 h decellularised PCA compared to non-decellularised PCA.
3.3. Cell seeding
Both H&E and DAPI staining confirmed a confluent layer of
cells attached to the luminal and abluminal sides of 9 and 15 h
decellularised PCA after 72 h when BASMCs were seeded at an
initial cell concentration of 1  105 cells/cm2 (Fig. 7(A) and (D)).
A degree of infiltration can be seen on the abluminal side of the
more porous 15 h decellularised scaffold (Fig. 7(D)) whilst no
infiltration was visible from the luminal surface for either
decellularisation times. After 15 day of static culture, cell
infiltration from the abluminal side and lateral edges of the
scaffolds is apparent for both the 9 and 15 h decellularised
sections (Fig. 7(B), (C) and (E), (F)). In parallel sections, H&E and
DAPI staining confirmed that BAEC seeded on the luminal side
formed a confluent layer in the 15 h decellularised PCA (Fig. 8).
The images chosen are representative of all sections viewed.
Non cell-seeded decellularised tissue samples were used as
controls for all experiments and there was no evidence of cells
present in those samples by H&E and DAPI (data not shown).
4. Discussion
Due to the wide range of species, blood vessels, timescales and
decellularisation protocols used it is difficult to determine the

Table 3 – Dimensional analysis of porcine coronary arteries.

Thickness (mm) Internal circumference (mm) External circumference (mm) Opening angle (1)

Non-decellularised, n ¼ 14 0.6370.18 6.2471.48 10.1972.06 72.22726.42

9 h decellularised, n¼ 14 0.4670.11 7.2071.44 9.9371.72 18.59727.52

n
Significance, po0.05, from corresponding non-decellularised.

Fig. 7 – H&E of BASMC-seeded decellularised PCA (9- and 15-h) at concentrations of (A), (D) 1  105 cells/cm2 and (B), (C), (E), (F)
5  104 cells/cm2 to the (A) lumen and lateral edges and (B)–(F) ablumenal side and lateral edges after 72 h and 15 day static
culture. Scale bar: 200 lm.
 journal of the mechanical behavior of biomedical materials 14 (2012) 130 –142
Fig. 8 – H&E of the cross-section and DAPI staining of the surface of 15 h decellularised PCA (A) and (C) luminally seeded with
BAEC at a concentration of 2  104 cells/cm2 for 10 days and (B) and (D) abluminally seeded with BASMC at a concentration of
1  105 cells/cm2 7 for 72 h. Scale bar: 200 lm.
optimum approach to achieve a decellularised scaffold suitable
for a TEBV. While the removal of cellular components is vital to
suppress the immune response, the biomechanical properties
of the scaffold are equally important (Gilbert et al., 2006). For
TEBVs in particular, the scaffold should withstand physiologi-
cal pressures and match the compliance of the native vessel.
Many decellularisation studies determine the effectiveness of
the protocol but without any biomechanical analysis (Kaushal
et al., 2001; Liu et al., 2004; Teebken et al., 2000). In addition,
studies which have included biomechanical evaluation have
generally tested small test specimens as opposed to full tubes
suitable for use as replacement arteries (Amiel et al., 2006;
Sheridan et al., 2012; Williams et al., 2009). We have established
that treatment by perfusion is necessary for decellularisation
of full tubular lengths of arteries, with all branches ligated,
similar to Dahan et al. (2012) and Montoya and McFetridge
(2009). Following 9 h decellularisation no intact cell nuclei were
visible throughout the vascular wall demonstrating that all
smooth muscle content was removed (Fig. 2(D), (P) and (V)).
Further treatment resulted in scaffolds with increased porosity
which could aid in the repopulation of the scaffold.
The alteration in biomechanical properties as a result of
decellularisation is due to the removal of cellular components
and also potential damage to ECM components. Natural degra-
dation of the tissue also occurs once tissue is removed from the
body. For preservation, arteries are generally stored in physiolo-
gical solutions for short periods of time (Carew et al., 2004; van
Andel et al., 2003) or frozen for longer periods (Conklin et al.,
2002; Sheridan et al., 2012; Venkatasubramanian et al., 2006;
Wilhjelm et al., 1997). Significant changes in the mechanical
properties of arteries have been reported for tissues stored in
physiological solution for more than 48 h (Chow and Zhang,
139
2011; Stemper et al., 2007) and also following storage at 20 and
80 1C (Chow and Zhang, 2011; Venkatasubramanian et al.,
2006). This study presents a protocol which results in the
successful decellularisation of fresh tubular sections of PCA
within 9 h of excision thereby removing the variable of natural
degradation of the arterial tissue. This is verified by the compar-
ison of the non-decellularised and 9 h non-decellularised uni-
axial tensile test data and the permeability data where there is
no statistically significant difference (Fig. 3(C) and Fig. 6). The
alteration in biomechanical properties in this study is therefore
due to the decellularisation technique alone. Staining of decel-
lularised sections does not indicate a degradation of the collagen
and elastin matrix so the change in mechanical properties can
be linked to the removal of vascular cells alone (Fig. 2(G)–(R)).
A small-diameter TEBV with appropriate biomechanical
and structural properties is required for procedures such as
CABG. There is very limited data on the biomechanical
properties of human and porcine coronary arteries in terms
of uniaxial tensile, biaxial tensile and inflation tests
(Holzapfel et al., 2005; Lally et al., 2004; van Andel et al.,
2003). The most common blood vessel used in decellularisa-
tion and repopulation studies to produce a TEBV is the elastic
carotid artery (Conklin et al., 2002; Dahan et al., 2012;
Sheridan et al., 2012; Williams et al., 2009; Zhao et al., 2010)
which has a considerably larger internal diameter than the
coronary arteries and may prove to be an unsuitable replace-
ment for the coronary arteries. This study is the first to
provide an insight into the mechanical properties of the
small-diameter muscular coronary artery and its correspond-
ing natural matrix scaffold. Vessel lengths of LAD, RCA, and
LCX can be excised with all branches ligated and decellu-
larised to act as vascular tissue engineering scaffolds
140 journal of the mechanical behavior of biomedical materials 14 (2012) 130 –142
however the LCX tapers more rapidly than the LAD and RCX
and is unsuitable for lengths greater than 25 mm.
A key factor for vascular substitutes is burst pressure or
ultimate strength since an inadequate burst pressure could
lead to acute rupture and anastomotic degeneration (Schaner
et al., 2004). In this study burst pressure tests have not been
explicitly performed however the inflation tests confirm that
9 h decellularised tubular sections can withstand luminal
pressures up to 200 mmHg. The permeability tests indicate
that flat sections of 9 h decellularised matrix can withstand
pressures up to 1500 mmHg, well above the physiological
range of an artery. Since this pressure only indicates the
normal radial pressure that the tissue can withstand, without
explicit consideration of circumferential stress, it can there-
fore not be used as a true measure of burst pressure but it
does show that the tissue retains considerable integrity and
strength following decellularisation. The burst pressure of
coronary arteries has not been previously reported, however
the burst pressures of human saphenous veins and internal
mammary arteries, which are favourable substitutes for CABG
procedures, are reported to be 15997877 and 31967164 mmHg,
respectively (Konig et al., 2009).
In addition, the permeability test results, showing an
increase in permeability of the tissue following 9 h decellu-
larisation (Fig. 6), provide key information on the permeabil-
ity of the natural scaffold and therefore the optimum
environment for vascular SMCs seeded in tissue engineered
scaffolds.
Although changes in the mechanical properties occur as a
result of decellularisation, the scaffold maintains the highly
non-linear response present in fresh arteries. Similar to other
studies, the tensile results (Fig. 3(C)) show the decellularised
arteries to be initially less stiff and more distensible in the
low load range due to the removal of vascular cells (Dahan
et al., 2012; Fitzpatrick et al., 2010). Subsequently as collagen
is recruited the decellularised arteries become stiffer than
non-decellularised arteries (Amiel et al., 2006; Dahan et al.,
2012; Sheridan, et al. 2012). The absence of smooth muscle
cells results in a more pronounced response from the col-
lagen and elastin content; an increased elasticity in the low-
load range due to the elastin and an increase in stiffness in
the latter stages from the contribution of collagen fibres. The
study of the opening angle also supports this finding (Table 3).
The non-decellularised samples spring open indicating the
presence of residual stresses within the wall; however the
decellularised samples failed to do so. This decrease in
opening angle further confirms full decellularisation since
the opening angle has been reported to significantly decrease
as a result of the removal of smooth muscle cells from the
arterial wall (Roy et al., 2005; Williams et al., 2009). The
inflation tests show similar trends where the initially less
stiff decellularised PCA becomes increasingly stiffer in com-
parison to the non-decellularised segments (Fig. 5) similar to
the findings of Conklin et al. (2002).
A limitation of this work includes the use of a low resolu-
tion imaging technique which cannot provide conclusive data
on disruption of collagen or elastin fibres following decellu-
larisation with NaOH. It should be noted however that
degradation of collagen crosslinks is highly unlikely given
that the final slope of the stress–strain curve, where collagen
predominantly bears the load, is stiffer for the decellularised
arteries when compared to fresh tissue.
A short-term decellularisation method is clearly more
commercially viable than longer term decellularisation pro-
tocols. Another limitation for the commercial viability of a
TEBV scaffold production is the time it takes to repopulate
the scaffold. BASMC attachment after 72 h of static culture
indicates the viability of the PCA scaffold as an environment
for xenogeneic cells; however, the internal basal lamina
proves to be a barrier to cell infiltration from the luminal
surface irrespective of treatment time. This is a positive
outcome since it indicates that the basal lamina of the vessel
is intact and could therefore support endothelial cell (EC)
seeding without risk of SMC neointimal formation. The more
loosely packed adventitial layer appears the most favourable
option for pre-seeding the scaffold and cell infiltration since
it is more porous. The 15 h decellularised scaffold provides an
improved opportunity for cell infiltration due to the adventi-
tial layer being more loosely packed as a result of longer 0.1 M
NaOH treatment. Uniaxial tensile tests performed after 15 h
treatment demonstrate that there is no further significant
alteration in the mechanical properties of the scaffold and
verify its ability to withstand loads. This also supports the
finding that the vascular cells are compromised after 9 h
0.1 M NaOH treatment and no longer play a role in the
mechanical response of the vessel. To overcome issues of
lack of cell infiltration in decellularised arteries, novel tech-
niques other than the standard method of seeding cells from
the luminal and abluminal sides, have been proposed by
Soletti et al. (2006) and more recently by Sheridan et al. (2012).
Soletti et al. (2006) accomplished more uniform scaffold cell
seeding using combined vacuum pressure, centrifugal force
and flow whilst Sheridan et al. (2012) embedded micro-
needles in the artery during the decellularisation procedure
and subsequently used the needles to inject cells and re-seed
the scaffold. The aim of this work is to obtain a natural
matrix scaffold with minimum disruption to the underlying
architecture thereby facilitating cell infiltration from cells of the
host artery as well as those pre-seeded on the lumen and/or
abluminal side. BASMC were seeded on the scaffold to verify the
suitability of the PCA scaffold for xenogeneic cell attachment
and growth. Initial static cell tests of 15 day in duration on 9-
and 15-h decellularised sections show significant potential for
smooth muscle cells to populate the artery from an anasto-
mosed host artery and from the abluminal side (Figs. 7(B), (C)
and 7(E), (F)). In addition, luminal seeding of 15 h decellularised
sections show potential to form an endothelial layer after 10 day
static culture (Fig. 8). Future cell culture tests will further assess
the viability of full scaffold repopulation by SMCs from the host
artery by seeding the lateral edges of the tissue with high cell
densities. SMCs will also be pre-seeded on the abluminal side
and ECs on the lumen to provide for optimum scaffold re-
population. It has recently been shown that the repopulation of
a scaffold can reverse the changes in mechanical properties that
are due to removal of vascular cells (Dahan et al., 2012). A
similar trend for the decellularised PCA scaffolds could show
recovery of the native mechanical properties. In this context,
future tests will investigate further the migration and infiltration
of cells into the natural matrix scaffold over longer time periods
in both static culture conditions and in a vascular bioreactor
 journal of the mechanical behavior of biomedical materials 14 (2012) 130 –142
which simulates physiological conditions (Colombo et al., 2010)
to determine the changes in mechanical properties that occur
as a result. This will provide further insights into the potential of
the porcine natural matrix scaffold as a viable TEBV scaffold.
5. Conclusions
The removal of the vascular cells from porcine coronary
arteries has been achieved through a short-term decellular-
isation method. The scaffold has been verified to be a suitable
environment for xenogeneic cell growth. It also has potential
for use simply as a vascular graft given that its acellular
nature does not initiate an immune response, it retains its
structural strength and it has mechanical properties very
close to that of the fresh artery whilst also maintaining
sufficient burst strength. In addition, the biomechanical
properties of PCA and the corresponding natural matrix
scaffold of the artery presented provide vital information
for the development of an optimum TEBV for small diameter
vessels from other synthetic materials, whilst also establish-
ing the potential of this natural matrix scaffold to be used as
a vascular tissue engineering scaffold.
Acknowledgements
This project is funded by Science Foundation Ireland under
the Research Frontiers Programme (08/RFP/ENM1378). In
addition, the authors thank Joseph Mackle from the School
of Mechanical and Manufacturing Engineering in Dublin City
University for his assistance with cell seeding.
r e f e r e n c e s
Allaire, E., Bruneval, P., Mandet, C., Becquemin, J.P., Michel, J.B.,
1997. The immunogenicity of the extracellular matrix in
arterial xenografts. Surgery 122, 73–81.
Amiel, G.E., Komura, M., Shapira, O., Yoo, J.J., Yazdani, S., Berry, J.,
Kaushal, S., Bischoff, J., Atala, A., Soker, S., 2006. Engineering
of blood vessels from acellular collagen matrices coated with
human endothelial cells. Tissue Engineering 12, 2355–2365.
Carew, E.O., Garg, A., Barber, J.E., Vesely, I., 2004. Stress relaxation
preconditioning of porcine aortic valves. Annals of Biomedical
Engineering 32, 563–572.
Chow, M.J., Zhang, Y., 2011. Changes in the mechanical and
biochemical properties of aortic tissue due to cold storage.
Journal of Surgical Research 171, 434–442.
Colombo, A., Zahedmanesh, H., Toner, D.M., Cahill, P.A., Lally, C.,
2010. A method to develop mock arteries suitable for cell
seeding and in-vitro cell culture experiments. Journal of the
Mechanical Behaviour of Biomedical Materials 6, 470–477.
Conklin, B.S., Richter, E.R., Kreutziger, K.L., Zhong, D.-S., Chen, C.,
2002. Development and evaluation of a novel decellularized
vascular xenograft. Medical Engineering & Physics 24,
173–183.
Dahan, N., Zarbiv, G., Sarig, U., Karram, T., Hoffman, A., Machluf,
M., 2012. Porcine small diameter arterial extracellular matrix
supports endothelium formation and media remodeling form-
ing a promising vascular engineered biograft. Tissue Engineer-
ing: Part A 18, 411–422.
Gilbert, T.W., Sellaro, T.L., Badylak, S.F., 2006. Decellularization of
tissues and organs. Biomaterials 27, 3675–3683.
141
Goissis, G., Suzigan, S., Parreira, D.R., Maniglia, J.V., Braile, D.M.,
Raymundo, S., 2000. Preparation and characterization of
collagen-elastin matrices from blood vessels intended as
small diameter vascular grafts. Artificial Organs 24, 217–223.
Fitzpatrick, J.C., Clark, P.M., Capaldi, F.M., 2010. Effect of decel-
lularization protocol on the mechanical behavior of porcine
descending aorta. International Journal of Biomaterials 2010,
1–11.
Holzapfel, G.A., Sommer, G., Gasser, C.T., Regitnig, P., 2005. Deter-
mination of layer-specific mechanical properties of human
coronary arteries with nonatherosclerotic intimal thickening
and related constitutive modelling. American Journal of
Physiology—Heart and Circulatory Physiology 289, H2048–H2058.
Kaushal, S., Amiel, G.E., Guleserian, K.J., Shapira, O.M., Perry, T.,
Sutherland, F.W., Rabkin, E., Moran, A.M., Schoen, F.J., Atala,
A., Soker, S., Bischoff, J., Mayer Jr., J.E., 2001. Functional small-
diameter neovessels created using endothelial progenitor
cells epanded ex vivo. Nature Medicine 7, 1035–1040.
Konig, G., McAllister, T.N., Dusserre, N., Garrido, S.A., Iyican, C.,
Marini, A., Fiorillo, A., Avila, H., Wystrychowski, W., Zagalski,
K., Maruszewski, M., Jones, A.L., Cierpka, L., de la Fuente, L.M.,
L’Heureux, N., 2009. Mechanical properties of completely
autologous human tissue engineered blood vessels compared
to human saphenous vein and mammary artery. Biomaterials
30, 1542–1550.
Lally, C., Reid, A.J., Prendergast, P.J., 2004. Elastic behavior of
porcine coronary artery tissue under uniaxial and equibiaxial
tension. Annals of Biomedical Engineering 32, 1355–1364.
Liu, S.Q., Tieche, C., Alkema, P.K., 2004. Neointima formation
on vascular elastic laminae and collagen matrices scaffolds
implanted in the rat aortae. Biomaterials 25, 1869–1882.
Montoya, C.V., McFetridge, P.S., 2009. Preparation of ex vivo-based
biomaterials using convective flow decellularization. Tissue
Engnieering: Part C 15, 191–200.
Roger, V.L., Go, A.S., Lloyd-Jones, D.M., 2011. Heart disease and
stroke statistics—2011 update: a report from the American
Heart Association. Circulation 123, e18–e209.
Roy, S., Silacci, P., Stergiopulos, N., 2005. Biomechanical properties
of decellularized porcine common carotid arteries. American
Journal of Physiology—Heart and Circulatory Physiology 289,
H1567–H1576.
Salacinski, H.J., Goldner, S., Giudiceandrea, A., Hamilton, G.,
Seifalian, A.M., Edwards, A., Carson, R.J., 2001. The mechanical
behaviour of vascular grafts: a review. Journal of Biomaterials
Applications 15, 241–278.
Schaner, P.J., Martin, N.D., Tulenko, T.N., Shapiro, I.M., Tarola,
N.A., Leichter, R.F., Carabasi, R.A., Dimuzio, P.J., 2004. Decel-
lularized vein as a potential scaffold for vascular tissue
engineering. Journal of Vascular Surgery 40, 146–153.
Sheridan, W.S., Duffy, G.P., Murphy, B.P., 2012. Mechanical char-
acterization of a customized decellularized scaffold for vas-
cular tissue engineering. Journal of the Mechanical Behavior
of Biomedical Materials 8, 58–70.
Soletti, L., Nieponice, A., Guan, J., Stankus, J.J., Wagner, W.R., Vorp,
D.A., 2006. A seeding device for tissue engineered tubular
structures. Biomaterials 27, 4863–4870.
Stemper, B.D., Yoganandan, N., Stineman, M.R., Gennarelli, T.A.,
Baisden, J.L., Pintar, F.A., 2007. Mechanics of fresh, refriger-
ated, and frozen arterial tissue. Journal of Surgical Research
139, 236–242.
Teebken, O.E., Bader, A., Steinhoff, G., Haverich, A., 2000.
Tissue engineering of vascular grafts: human cell seeding of
decellularised porcine matrix. European Journal of Vascular
and Endovascular Surgery 19, 381–386.
van Andel, C.J., Pistecky, P.V., Borst, C., 2003. Mechanical proper-
ties of porcine and human arteries: implications for coronary
anastomotic connectors. The Annals of Thoracic Surgery 76,
58–64.
142 journal of the mechanical behavior of biomedical materials 14 (2012) 130 –142
Venkatasubramanian, R.T., Grassl, E.D., Barocas, V.H., Lafontaine,
D., Bischof, J.C., 2006. Effects of freezing and cryopreservation
on the mechanical properties of arteries. Annals of Biomedi-
cal Engineering 34, 823–832.
Wilhjelm, J.E., Vogt, K., Jespersen, S.K., Sillesen, H., 1997. Influence
of tissue preservation methods on arterial geometry and echo-
genicity. Ultrasound in Medicine and Biology 23, 1071–1082.
Williams, C., Liao, J., Joyce, E.M., Wang, B., Leach, J.B., Sacks, M.S.,
Wong, J.Y., 2009. Altered structural and mechanical properties
in decellularized rabbit carotid arteries. Acta Biomaterialia 5,
993–1005.
Wilshaw, S.P., Rooney, P., Berry, H., Kearney, J.N., Homer-Vannia-
sinkam, S., Fisher, J., Ingham, E., 2011. Development and
characterization of acellular allogeneic arterial matrices. Tis-
sue Engineering: Part A.
Zhao, Y., Zhang, S., Zhou, J., Wang, J., Zhen, M., Liu, Y., Chen, J., Qi,
Z., 2010. The development of a tissue-engineered artery using
decellularized scaffold and autologous ovine mesenchymal
stem cells. Biomaterials 31, 296–307.