JOURNAL OF TISSUE ENGINEERING AND REGENERATIVE MEDICINE RESEARCH ARTICLE

J Tissue Eng Regen Med (2011).

Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/term.413

Development of a total atherosclerotic occlusion

with cell-mediated calcium deposits in a rabbit
femoral artery using tissue-engineering scaffolds
Beili Zhu1,2∗ , Steven R. Bailey1,2 , James Elliott3 , Xiankai Li1,4 , G. Patricia Escobar1 ,
Eva M. Rodriguez1 and C. Mauli Agrawal2
1 Janey Briscoe Center for Cardiovascular Research, Janey and Dolph Briscoe Division of Cardiology, Department of Medicine, University

of Texas Health Science Center at San Antonio, TX, USA

2 Department of Biomedical Engineering, College of Engineering, University of Texas at San Antonio, TX, USA
3 Department of Laboratory Animal Resources, University of Texas Health Science Center at San Antonio, TX, USA
4 Department of Cardiology, Shanghai Tenth People’s Hospital, Tongji University, Shanghai, People’s Republic of China

Abstract
This study sought to establish a chronic total occlusion (CTO) model with cell-mediated calcium
deposits in rabbit femoral arteries. CTO is the most severe case in atherosclerosis and contains
calcium deposits. Previous animal models of CTO do not mimic the gradual occlusion of arteries
or have calcium in physiological form. In the present study we tested the strategy of placing
tissue-engineering scaffolds preloaded with cells in arteries to develop a novel CTO model. Primary
human osteoblasts (HOBs) were first cultured in vitro on polycaprolactone (PCL) scaffolds with 5 ng
TGFβ1 loading for 28 days for precalcification. The HOB–PCL construct was then implanted into a
rabbit femoral artery for an additional 3, 10 or 28 days. At the time of sacrifice, angiograms and
gross histology of arteries were captured to examine the occlusion of arteries. Fluorescent staining
of calcium and EDS detection of calcium were used to evaluate the presence and distribution of
calcium inside arteries. Rabbit femoral arteries were totally occluded over 28 days. Calcium was
presented at CTO sites at 3, 10 and 28 days, with the day 10 specimens showing the maximum
calcium. Chronic inflammatory response and recanalization were observed in day 28 CTO sites. A
novel CTO model with cell-mediated calcium has been successfully established in a rabbit femoral
artery. This model can be used to develop new devices and therapies to treat severe atherosclerotic
occlusion. Copyright  2011 John Wiley & Sons, Ltd.

Received 25 May 2010; Accepted 30 November 2010

Keywords atherosclerosis; chronic total occlusion; calcium; tissue engineering scaffolds;

polycaprolactone; primary human osteoblast; TGFβ1

1. Introduction et al., 2007). The chronic total occlusion is a ≥ 3

Atherosclerosis is the leading cause of mortality and
morbidity in North America and chronic total occlusion
(CTO) is the progression of this disease, featuring cal-
cium deposits in completely obstructed arteries (Soon
*Correspondence to: Beili Zhu, Janey and Dolph Briscoe
Division of Cardiology, Department of Medicine, University of
Texas Health Science Center at San Antonio, 7703 Floyd Curl
Drive, TX 78229-3900, USA. E-mail: beilizhu@yahoo.com
Copyright  2011 John Wiley & Sons, Ltd.
month-old obstruction of an artery consisting of vari-
ous degrees of fibro-atheromatous plaque and thrombus
(Fiorilli et al., 2006; Puma et al., 1995). Calcium depo-
sition in plaques is a surrogate marker for CTO (Hsu
et al., 2008; Li et al., 2007). Its development is an active,
cell-mediated and highly regulated process similar to
osteogenesis (Giachelli, 2004; Montecucco et al., 2007).
Although CTO occurs in only 10% of all percutaneous
coronary interventions (PCIs), it reflects the most severe
case in atherosclerosis and remains a big challenge in
cardiology (Bhambhani et al., 2007). Despite remarkable

Figure 1. H&E stain of an atherosclerotic human right iliac artery. Arrow indicates bone; ∗ haematopoietic elements, which are
bone marrow-like tissue. Courtesy of Felipe J. Solano, South Texas Veterans Health Care System, San Antonio
advances in treatment strategies, expertise and equip-
ment, there is a strong need to create animal models for
CTO for the development of novel devices and techniques
to treat risky lesions (Bhambhani et al., 2007).
Many approaches have been explored to create ani-
mal models of CTO. Balloon-initiated injury and surgical
partial ligation can produce occluded arteries by intima
hyperplasia (Ishii et al., 2006; Raval et al., 2006; Sampath
et al., 2007). High-cholesterol diets can induce arteries to
accumulate collagen and calcify vesicles, which are steps
in vascular calcification that occur prior to calcium depo-
sition (Hsu et al., 2004). The combination of mechanical
damage in arteries and a high-cholesterol/low-calcium
diet in animals has been seen to cause partial occlusion
in vessels, with calcification in the artery walls (Hong
et al., 1995; Ishii et al., 2006; White et al., 1988). This
is in contrast to CTO in humans, where calcium is often
present in the lumen area. Besides, in these models, in
addition to target arteries, calcification was also seen in
non-specific organs, such as bile ducts and renal cortex
(Hong et al., 1995). Figure 1 shows a longitudinal section
of an atherosclerotic human right iliac artery, where the
tissue has been stained by haematoxylin and eosin (H&E).
It shows that in advanced atherosclerosis, dense bone-like
structure and haematopoietic elements (bone marrow-like
tissue) form in arteries. Placement of ameroid constrictors
on the external surface of an artery and injection of throm-
bin in one segment of an artery are some other methods
for inducing CTO (Operschall et al., 2000; Radke et al.,
2006; Segev et al., 2005; Tomaru et al., 1996; Tuzun
et al., 2009; Tyagi et al., 1996; Zhu et al., 2008). How-
ever, the down-side is that calcium is often missing in
these models and they require major surgery.
Some new CTO models have been using minimal-
invasive interventional techniques to place polymer
sponges or copper stents at the exact target location in an
artery (Prosser et al., 2006; Reffelmann et al., 2004; Song
et al., 2005; Thind et al., 2007). Reproducible animal
models of totally occluded vessels were achieved in this
fashion. However, calcium was often not present in the
occlusions (Kipshidze et al., 2005; Prosser et al., 2006).
More recent studies report the implantation of gelatin
sponges with bone powder mixture or Ca2+ , PO4 4− dip-
coated poly(L-lactide) scaffolds in arteries (Suzuki et al.,
2008, 2009). Calcium was detected at occlusion sites in
these studies. However, this calcium was not deposited
Copyright  2011 John Wiley & Sons, Ltd.
B. Zhu et al.
by cells and there is still a concern regarding whether it
is similar to the pathological form found in human CTO.
The primary goal of our study was to create a repro-
ducible CTO model in a rabbit femoral artery that exhibits
total occlusion and includes calcium deposits which
mimic pathological calcium deposits in the atheroscle-
rotic lumen. The iliac and femoral arteries of New Zealand
white rabbits have been used extensively in vascular stud-
ies. These include creating models of CTO by injection
of thrombin solution in iliac-femoral arteries, restenosis
studies of balloon angioplasty injury and the follow-up
stent implantation at the sites to reduce neointima hyper-
plasia (Buerke et al., 2007; Busk et al., 2008; Farb et al.,
2000; Hehrlein et al., 1994; Lysitsas et al., 2007; Segev
et al., 2005; Strauss et al., 2003; Tomaru et al., 1996;
Yang et al., 2006; Zhang et al., 2008; Zhang and Yin,
2008). Thus, rabbit femoral arteries are a well-established
model in vascular investigation.
Several studies have reported the application of human
cells in rabbit in vivo studies. Human corneal endothelial
cells have been transplanted in rabbit corneas, and they
retained their function of corneal dehydration (Hitani
et al., 2008; Mimura et al., 2004). Human bone marrow-
derived mesenchymal stem cells have been placed in a
damaged heart or dorsal skin in rabbits (Stoff et al., 2009;
Wang et al., 2005). The heart function was improved by
increased cardiac muscle contraction and the mechanical
strength of dorsal wounds was enhanced, respectively.
Also, human umbilical cord blood stem cells have been
transplanted into rabbit cartilage defects with encourag-
ing results (Yan and Yu, 2007). The outcomes of these
studies have shown that human-origin cells implanted in
rabbits can maintain themselves in vivo and do not induce
a large immunological response and phagocytosis.
One mechanism for the formation of calcified CTO
is that vascular cells such as smooth muscle cells and
pericytes undergo a phenotypic transition to express
osteoblastic lineage markers (Giachelli, 2005). Since it is
this osteoblastic modulation that generates the calcium-
rich extracellular matrix in CTO, here we used primary
human osteoblasts (HOBs) to develop this model of cal-
cified CTO (Giachelli, 2004). Polycaprolactone (PCL) is a
biodegradable polymer whose total degradation can reach
2–4 years, depending on the molecular weight (Woodruff
J Tissue Eng Regen Med (2011).
DOI: 10.1002/term
Total atherosclerotic occlusion with cell-mediated calcium deposits in rabbit femoral artery

Figure 2. (A) Polycaprolactone (PCL) scaffolds (diameter, 3 mm; length, 5 mm). (B) Cross-sections of polycaprolactone scaffolds
under light microscopy

and Hutmacher, 2010). PCL has long been applied in med-

ical devices and bone tissue engineering (Woodruff and
Hutmacher, 2010).
In our previous in vitro study, we used primary human
osteoblasts (HOBs) cultured on polycaprolactone (PCL)
scaffolds loaded with the growth factor TGFβ1 to achieve
a construct with calcium deposits (Zhu et al., 2010).
The hypothesis of the present investigation was that by
implantation of this HOB/PCL construct in rabbit femoral
arteries using interventional techniques, a CTO model can
be generated with cell-mediated calcium in the occlusion
sites.

2. Materials and methods

Figure 3. (A) Schematic description of scaffold delivery to create
chronic total occlusion (CTO) in a rabbit femoral artery.
2.1. Fabrication of PCL scaffolds (B) Location of CTO sites in targeted femoral arteries

PCL scaffolds were prepared by vibration particle/salt

leaching techniques, as previously reported (Zhu et al.,
2010). Poly-ε-caprolactone was obtained from Birming-
ham Polymers (Birmingham, AL, USA). The inherent
viscosity of the material as received was 1.21 dl/g and the
average molecular weight was 150 kDa. The solvent used
in the scaffold fabrication process was dichloromethane
(Sigma, St Louis, MO, USA). The final PCL scaffolds for
animal studies were cylindrical (diameter 3 mm × length
5 mm; Figure 2). Before cell culture, the scaffolds were
soaked in 75% ethanol for 15 min and then rinsed with
phosphate-buffered saline (PBS) three times for steriliza-
tion purposes.
2.2. TGFβ1 coating on PCL scaffolds
After sterilization, PCL scaffolds were first soaked in
20 µg/ml fibronectin solution (Sigma-Aldrich) overnight.
TGFβ1 powder was purchased from R&D Systems (Min-
neapolis, MN, USA). It was reconstituted in 4 mM
HCl to prepare a stock solution at a concentration of
40 ng/µl. The stock solution of TGFβ1 was then diluted to
5 ng/20 µl volume using 1% BSA. Each 20 µl aliquot was
then drop-loaded onto fibronectin-coated PCL scaffolds,
using a pipette.
Copyright  2011 John Wiley & Sons, Ltd.
2.3. Precalcification of primary human
osteoblast cultures on TGFβ1-coated scaffolds
Primary human osteoblasts (HOBs; catalogue no. CC-
2538) were purchased from Lonza Walkersville (Walk-
ersville, MD, USA). HOBs from the third passage were
seeded on PCL scaffolds with 5 ng TGFβ1 loading
at 2 × 105 cells. The medium used for precalcifcation
treatment was α-minimal essential medium (α-MEM;
Invitrogen, Carlsbad, CA, USA) supplemented with 7%
fetal bovine serum (FBS), 1% antibiotic/antimyotic stabi-
lized (10 000 U/ml penicillin, 10 mg/ml streptomycin),
100 µg/ml L-ascorbic acid, 5 mM β-glycrolphosphate
and 10−10 M dexamethasone. After 28 days of culture,
HOB–PCL constructs were harvested for implantation in
rabbit femoral arteries. Our previous studies have shown
that such constructs had calcium deposits in PCL scaffolds
(Zhu et al., 2010).
2.4. Animal experiments
New Zealand white rabbits weighing 4.5–5 kg (Myrtle’s
Rabbitry, TN, USA) were used in this study. The experi-
mental protocol was approved by the Institutional Animal
J Tissue Eng Regen Med (2011).
DOI: 10.1002/term
 B. Zhu et al.

Figure 4. Angiogram of both left and right femoral arteries before implantation: baseline (A) and after implantation of scaffolds
(B) from the same animal. Typical images of chronic total occlusion (CTO) sites of both left and right femoral arteries at 3 (C), 10
(D) and 28 days (E) showed total occlusions in vessels. Arrows indicate CTO occlusion sites in arteries

Figure 5. (A) Gross histology of a fresh 28-day chronic total occlusion (CTO) artery with a control scaffold inside. (B) A 28-day
CTO artery with a treated scaffold implanted inside: p, proximal end; d, distal end; s, scaffold. Arrows indicate scaffold constructs
inside the artery

Care and Use Committee at the University of Texas Health
Science Center at San Antonio. The care and use of
animals was in accordance with the National Research
Council (NRC) publications and other federal regulations.
The animals were pre-anaesthetized using a mixture
of ketamine (30 mg/kg) and midazolam (0.2 mg/kg),
administered intramuscularly (i.m.). The animals were
intubated and anaesthesia was maintained using isoflu-
rane inhalant (1–2%) administered with 100% O2 .
Three time points (3, 10 and 28 days) were used for the
study, with six animals at each time point. Each animal
received a control PCL scaffold and a test scaffold with
osteoblasts and TGFβ1, which had undergone an in vitro
precalcification culture for 28 days beforehand.
A right common carotid artery cut-down was per-
formed, and a 5 French introducer sheath (Boston Sci-
entific, Natick, MA, USA) was placed into the artery.
Copyright  2011 John Wiley & Sons, Ltd.
A 0.014 inch guide wire was first advanced across the
aortic arch. A 5 Fr angled Glidecath (Terumo Medical,
Somerset, NJ, USA) was followed over the wire to pass
the aortic arch and advanced to a point proximal to
the iliac bifurcation. A baseline angiogram of both of
the femoral arteries was performed at this time. The
guide wire was advanced down one femoral artery to
just proximal to the stifle. The Glidecath was retracted,
leaving the 0.014 wire in place. A treated PCL scaf-
fold was threaded onto the end of the wire and was
pushed down to the mid-thigh area of the femoral
artery (left or right), using a 5 Fr straight XP Glide-
cath (Terumo) (Figure 3A). A control PCL scaffold was
then placed in a similar position of the contralateral
femoral artery, using the same technique (Figure 3B).
A post-scaffold implantation angiogram was performed
immediately after placement to document the deployment
J Tissue Eng Regen Med (2011).
DOI: 10.1002/term
Total atherosclerotic occlusion with cell-mediated calcium deposits in rabbit femoral artery

Figure 6. Stains for calcium (green fluorescence) at chronic total occlusion (CTO) sites of rabbit femoral arteries with scaffold
implantation at 3, 10 and 28 days. Nuclei were counterstained with DAPI (blue fluorescence) from the same location. Artery sections
that had control scaffolds implanted inside were stained with calcium (A–C), and DAPI (D–F). CTO sites with treated scaffolds were
stained for calcium (G–I) and DAPI (J–L). Arrows identify the elastic lamina of arteries, and point toward the lumen. Scale bar
= 250 µm. (M) Quantitative area of calcium deposits in treated and control CTO sites at 3, 10 and 28 days; ∗ p < 0.05; ∗∗ p < 0.01

of the constructs and to verify the percentage occlusion
of the vessels. The catheter and wire were removed
and the carotid artery was ligated with 4-0 silk suture.
The incisions were closed in two layers, using 3-0
absorbable suture. The rabbits were allowed to recover
and monitored until they were sacrificed at days 3, 10
and 28.
Prior to euthanasia, the animal was sedated and intu-
bated as stated above. An abdominal incision was made
and the abdominal aorta isolated. A 5 Fr introducer sheath
(Boston Scientific Corp.) was placed inside the vessel and
an angiography was performed to document the per-
centage occlusion of the femoral arteries (both left and
right sides) at the end of each time point. The rabbits were
then euthanized while still under anaesthesia; the femoral
arteries were flushed with saline and then excised.
Copyright  2011 John Wiley & Sons, Ltd.
2.5. Tissue preservation
Femoral arteries were fixed with 10% neutral buffered
formalin for 24 h and then embedded in Tissue-Tek OCT
compound (Electron Microscopy Sciences, Hatfield, PA,
USA). Sections were cut in 60 µm thickness using a Leica
cryostat, and mounted onto Superfrost Plus microscope
slides (Erie Scientific, Portsmouth, NH, USA).
2.6. Detection of calcium element
The tissue sections were first washed in Ca2+ -free
PBS (Invitrogen) three times at 5 min intervals. They
were then air-dried overnight. The dry specimens
were examined using a scanning electron microscope
(JEOL JSM-6610LV, Peabody, MA, USA) equipped
J Tissue Eng Regen Med (2011).
DOI: 10.1002/term

Figure 7. EDS profile of chronic total occlusion (CTO) arteries
at day 28 with control scaffolds (A) and treated scaffolds (B)
with an Energy-dispersive X-ray Spectroscopy (EDS)
detector (EDAX Apollox system) to detect calcium
elements.
2.7. Fluorescent imaging of calcium and cell
nuclei
Tissue sections were first washed in PBS three times
at 5 min intervals. Staining was then performed as pre-
viously described (Zhu et al., 2010). Calcification was
stained using an OsteoImage mineralization assay kit
(Lonza). The slides were mounted in mounting medium
with DAPI (Vector Laboratories, Burlingame, CA, USA).
They were later observed under an Olympus FV-1000
laser scanning microscopy system with an Olympus IX-81
microscope at ×10 magnification. Calcium deposits were
stained green, while cell nuclei were stained blue.
2.8. Quantification of calcium area
Areas of calcium deposits were measured based on fluo-
rescent calcium images, using Image J software (version
1.4, National Institutes of Health). Three samples were
taken from each study group at each time point, and
three images were used from one specimen to measure
the amount of calcium. The field of view of each image
was 0.41 mm2 .
Copyright  2011 John Wiley & Sons, Ltd.
B. Zhu et al.
2.9. Haemotoxylin and eosin staining
Tissue sections were routinely stained with H&E, using
Gill’s III haematoxylin, based on standard protocols.
After the staining procedure, the slides were air-dried
and mounted in mounting medium (Vector Laboratories)
and covered with coverslips. Artery sections were later
observed under light microscopy.
3. Results
3.1. Angiographic outcomes
A total of 18 rabbits were used in this study; in 17
(94%) of them successful CTO was established in the
femoral arteries. PCL scaffolds in one rabbit for the day
28 time point were missing at the occlusion site after
artery dissection. Figure 4 illustrates angiograms of arter-
ies at baseline before implantation (A) and immediately
after implantation (B) from the same animal. Figure 4C–E
shows typical images of CTO at days 3 (C), 10 (D) and 28
(E). The femoral arteries on both sides were patent before
implantation and then were obstructed immediately after
the placement of the scaffolds. At the same time, the
collateral circulation was able to feed the vessels of lower
limbs. CTO sites remained 100% occluded at days 3, 10
and 28. After 28 days, a rich collateral circulation had
been developed in the lower limbs (Figure 4E).
3.2. Gross histology
Figure 5 shows the gross histology of 28 day CTO arteries
with control scaffolds (Figure 5A) and treated scaffolds
(Figure 5B) inside. In both groups, the scaffolds were
observed in the middle of arteries.
3.3. Fluorescent stain of calcium and cell nuclei
Figure 6A–L shows the calcium (green fluorescence) and
nuclei of cells (blue fluorescence) present in occluded
arteries. At day 3, CTO with treated scaffolds showed
calcium deposits all over the artery, while the control
CTO also showed the presence of calcification. At day
10, CTO with treated scaffolds showed an extensive and
higher degree of calcium than day 3 specimens, while
the control CTO still had very little calcification. At
day 28, treated CTO did not have as much calcium as
day 3 and 10 specimens. At this time point, the control
CTO did not show any calcification. Thus, calcium was
detected in arteries with treated scaffolds at days 3,
10 and 28, with the day 10 arteries showing the most
extensive calcification and day 28 specimens with few
calcium deposits. The control CTOs at days 3 and 10 had
minor degrees of calcification compared to their treated
counterparts at the same time point.
J Tissue Eng Regen Med (2011).
DOI: 10.1002/term
Total atherosclerotic occlusion with cell-mediated calcium deposits in rabbit femoral artery
Figure 8. H&E staining of chronic total occlusion (CTO) arteries with control scaffolds or treated scaffolds implanted at 3 and
10 days, respectively. Arrows identify the elastic lamina of arteries; ∗ scaffold structure inside CTO sites. Scale bars labelled in each
panel
3.4. Quantification of calcium area
Figure 6M shows the area of calcium deposits in CTO
arteries. At days 3, 10 and 28, treated CTO had more
calcium than control CTO (p < 0.05). Most calcium was
observed on day 10. An average of 1.2 × 104 µm2 area
was covered by calcium in 41 × 104 µm2 of captured
imaged area. At days 3 and 10, control CTO sites also had
little calcium deposits.
3.5. Detection of calcium element
Figure 7 demonstrates the EDS profiles of day 28 CTOs,
showing a typical control scaffold (Figure 7A) and a
treated scaffold (Figure 7B). EDS is an analytical method
that can examine the presence of chemical elements in
specimens, including calcium. At day 3, treated CTO had
calcium signal which the control did not show. At day
10, both treated and control CTO had signal peaks for
calcium. At day 28, treated CTO had calcification in the
arteries, while control CTO did not have calcification. In
comparison with a negative control (a normal artery),
treated CTOs at 3, 10 and 28 days all had calcification in
occlusion sites. The EDS profile indicated that calcification
was found in treated CTO at 3, 10 and 28 days.
Some control CTOs also had a trace of calcium in
arteries.
Copyright  2011 John Wiley & Sons, Ltd.
3.6. Histological staining with haematoxylin
and eosin
Figures 8–10 illustrate the H&E staining of occluded CTOs
in arteries at different magnifications. Figure 8 shows that
both control and treated CTO sites were 100% occluded at
3 and 10 days. Figure 10A indicates the total occlusion of
control and treated CTO sites at day 28. In both Figure 8
and Figure 10A, the lumina of the arteries were filled
with cells and the structure of the scaffolds. PCL material
was seen through the occlusion sites. Fewer struts of PCL
scaffolds remained inside the occlusion sites at day 28
compared to those at days 3 and 10.
Figure 9 shows the inflammatory response in CTOs at
3, 10 and 28 days. At day 3, lymphocytes and leukocytes
were seen near the PCL struts in treated CTO. Leuko-
cytes are the cell type present during acute inflammation
(Hersh and Bodey, 1970). At day 10, leukocytes resided in
the adventitia of both control and treated CTOs. In addi-
tion, fibroblasts, an indicator of the chronic inflammation
(Buckley et al., 2001), were shown inside the occlusion
sites in treated CTOs. At day 28, in treated CTOs, leuko-
cytes resided close to the scaffold struts. They were mixed
with fibroblasts and infiltrated into occlusion sites (images
not shown).
Figure 10A shows that at day 28, new blood vessels
were formed inside both control and treated CTOs
(recanalization). Vessels from the adventitia migrated
J Tissue Eng Regen Med (2011).
DOI: 10.1002/term

Figure 9. Acute and chronic inflammatory response in control and treated CTO sites at 3, 10 and 28 days using H&E staining.
Arrows indicate leukocytes and arrowheads lymphocytes. ∗ Blue spiny cells, which are fibroblasts. Scale bars labelled in
each panel
inside the lumen and spilled over in the treated CTOs.
Figure 10B identifies an endothelial cell nucleus residing
in the internal elastic lamina of the treated CTO.
Intact endothelial cells indicate that the vessel wall
was not damaged during the interventional procedure
of polymer implantation. In consistency with Figures 6
and 7, Figure 10C shows the presence of calcium deposits
in the day 28 treated CTO.
4. Discussion
The goal of this study was to develop a CTO model with
cell-mediated calcium in animal arteries using tissue-
engineering scaffolds. Primary human osteoblasts (HOBs)
were first cultured on PCL scaffolds and under calcification
Copyright  2011 John Wiley & Sons, Ltd.
B. Zhu et al.
treatment in vitro for 28 days and then implanted in vivo
for an additional 3, 10 and 28 days.
Angiograms showed that the targeted femoral arteries
were 100% occluded after placement of scaffolds and the
vessels remained totally occluded over 28 days. Similar
angiographic results were found in several catheter-based
CTO models (Kipshidze et al., 2005; Prosser et al., 2006;
Suzuki et al., 2008, 2009) over similar time periods. Our
gross histology confirmed the total occlusion of arteries,
with the presence of implanted scaffolds in both control
and treated groups.
Fluorescent staining of calcium showed that calcifica-
tion was successfully established in CTO sites at all time
points, with the day 10 specimen having the maximal
calcium. Figure 11 shows fluorescent staining of cal-
cium in cellular scaffolds under the same precalcification
J Tissue Eng Regen Med (2011).
DOI: 10.1002/term
Total atherosclerotic occlusion with cell-mediated calcium deposits in rabbit femoral artery

Figure 10. (A) Recanalization, the new vessel formation in control and treated CTO sites at 28 days using H&E staining; arrowheads
indicate small new blood vessels; arrows identify the elastic lamina of arteries; ∗ location of scaffolds. (B) Intact endothelial cell
nucleus resides on the lumen of the blood vessel in a day 28 CTO; arrow indicates endothelial cell. (C) Calcium deposits in a day
28 CTO; arrow indicates calcium. Scale bars labelled in each panel

Figure 11. Fluorescent calcium staining (green) of scaffolds pretreated with HOBs and TGFβ1 loading for calcification in cell
culture at 28 days before implantation (A). Calcium staining of scaffolds with same precalcification treatment implanted inside
rabbit femoral arteries for an additional 10 days (B). Quantitative area of calcium deposits in treated CTO sites at day 0 before
implantation and day 10 after implantation (C). Arrows indicate calcium deposits. Scale bar = 250 µm (A, B)

Copyright  2011 John Wiley & Sons, Ltd. J Tissue Eng Regen Med (2011).
DOI: 10.1002/term

Figure 12. H&E staining of HOBs on PCL scaffolds in in vitro cell cultures (A) and inside animal arteries at day 10 (B). H&E staining
of control scaffolds with no HOBs seeded in arteries for 10 days (C). Arrows indicate blue elongated cells bridging the struts of the
scaffolds. Scale bar = 200 µm (A,B,C)
treatment in vitro before implantation (Figure 11A) and
after implantation in rabbits for an additional 10 days
(Figure 11B). As seen, calcium was extensively developed
in rabbit femoral arteries, and more calcium was observed
in the scaffolds after placement than before implantation.
Our previous in vitro study demonstrated cellular calcium
deposits in HOBs using the same staining technique (Zhu
et al., 2010). Besides, local delivery of TGFβ1 has been
reported to improve wound healing of bone fractures in
animals (Tattini et al., 2008; Wildemann et al., 2005).
These findings suggest that, in this present study, the
TGFβ1 coating on the scaffolds, together with the HOBs,
played a role in inducing more calcium deposits in arteries.
Figure 11 clearly shows that, after scaffold implanta-
tion, the calcification in our CTO model progressed inside
the arteries. Using osteoblast growth factor strategy, we
were able to achieve increased calcium contents over time
in our animal model. Our unique cell-mediated calcium
deposits mimicked the progression of human CTO in clin-
ical cases. As in humans, calcification and occlusion of
arteries is a cell-mediated process and is highly similar to
osteogenesis (Giachelli, 2004; Montecucco et al., 2007).
To our knowledge, other published models did not have
calcium that would grow over time, as in a clinical dis-
ease process. Also, their calcium was in the form of either
Ca2+ ions or bone powders that are not biologically active
(Suzuki et al., 2008, 2009).
EDS profiles again demonstrated the calcium element
in treated CTO arteries. These data indicated that calcium
deposits were present in occlusion sites and matched the
findings of the calcium staining. When HOBs were cul-
tured on bioactive glass scaffolds in vitro, similar EDS
profiles of calcium were detected after 12 and 21 days
(Jones et al., 2007). Similar peaks for calcium were also
Copyright  2011 John Wiley & Sons, Ltd.
B. Zhu et al.
seen in poly(L-lactide) scaffolds dip-coated with Ca2+ and
PO4 4 – ions (Suzuki et al., 2009).
Figure 12 shows H&E staining of HOBs on scaffolds.
Figure 12A shows that when HOBs were cultured on PCL
in vitro, the cells stained as blue needle-like shapes on the
scaffolds. Cells with the same shape were seen on scaffolds
10 days after the constructs were implanted into the artery
(Figure 12B). This indicated that HOBs were still present
on the scaffold and resided at the same location. In
contrast, the control CTO at day 10 (Figure 12C) did not
have cells with this shape, because only PCL itself was
implanted into these rabbit arteries, with no HOBs.
When human osteoblasts were implanted in the dorsi
muscles of mice, needle-shaped crystals were seen in the
tissue at 8 weeks, using transmission electron microscopy
(Yamanouchi et al., 2001). When our human osteoblasts
are left long enough in arteries, it is very likely that bony
spicules will be found at the treated CTO sites. Human
CTOs often contain tissue that is pathologically similar
to bone (Doherty et al., 2003; Sage et al., 2010). Leaving
HOBs long enough in arteries will develop bone archi-
tecture in lumen that greatly mimics the human disease
of CTO.
H&E staining in our study illustrated that the acute
and chronic inflammatory response occurred in sequence
at treated CTO sites. In addition, recanalization was
observed at day 28 at the occlusion sites. Intact endothelial
cells demonstrated that our mature interventional tech-
nique did not damage the blood vessels when the scaffolds
were inserted. Here, calcium deposits were emphasized
again by H&E staining, besides the proof of fluorescent
staining and EDS detection to show its presence.
Total atherosclerotic occlusion in humans is a chronic
vascular inflammation (Stoll and Bendszus, 2006).
Microvascular invasion occurs in occlusion (Sage et al.,
J Tissue Eng Regen Med (2011).
DOI: 10.1002/term
Total atherosclerotic occlusion with cell-mediated calcium deposits in rabbit femoral artery

2010) and bone tissue is present in the arteries as well
(Doherty et al., 2003). Based on these facts, our CTO
model bears great similarities to this human disease,
and contains the most important characteristics of a
human CTO.
response and recanalization were observed at day 28
CTO sites. This novel CTO model with cell-deposited
calcium can be used to support the future development
of new techniques and devices to treat more risky CTOs.

5. Conclusions
A CTO model was developed, using osteoblast-seeded PCL
scaffolds treated with TGFβ1, which were cultured in vitro
for 28 days to initiate calcification and then inserted into
the femoral arteries of a rabbit model. The results indi-
cate that the femoral arteries stayed totally occluded over
28 days. Cell-deposited calcium was observed at CTO
sites at 3, 10 and 28 days, with the day 10 specimens
showing the maximum calcium. Chronic inflammatory
Acknowledgements
The authors are grateful to Robert L. Reddick MD, Professor
and Chair of the Department of Pathology at UT Health Science
Center at San Antonio, for his help in reading the H&E staining
and its interpretation. Fluorescent images were generated in the
Core Optical Imaging Facility at the University of Texas Health
Science Center at San Antonio, which is supported by NIH-NCI
P30 CA54174 (CTRC at UTHSCSA), NIH-NIA P30 AG013319
(Nathan Shock Center) and NIH-NIA P01AG19316. This work
was funded by the Janey Briscoe Center for Cardiovascular
Research at The University of Texas Health Science Center at
San Antonio.

References
Bhambhani A, Gadkar N, Cook S, 2007;
Intravascular ultrasound guided percuta-
neous coronary intervention for chronic
toal occlusion of left anterior descending
artery. Kardiovask Med 10: 78–80.
Buckley CD, Pilling D, Lord JM, et al. 2001;
Fibroblasts regulate the switch from acute
resolving to chronic persistent inflamma-
tion. Trends Immunol 22(4): 199–204.
Buerke M, Guckenbiehl M, Schwertz H,
et al. 2007; Intramural delivery of
Sirolimus prevents vascular remodeling
following balloon injury. Biochim Biophys
Acta 1774(1): 5–15.
Busk M, Mertz H, Espersen GT, et al. 2008;
Effects of pentoxifylline on the vascular
response to injury after angioplasty in rab-
bit iliac arteries. Basic Res Cardiol 103(3):
257–264.
Doherty TM, Asotra K, Fitzpatrick LA, et al.
2003; Calcification in atherosclerosis:
bone biology and chronic inflammation
at the arterial crossroads. Proc Natl Acad
Sci USA 100(20): 11201–11206.
Farb A, Tang AL, Shroff S, et al. 2000;
Neointimal responses 3 months after 32 P
β-emitting stent placement. Int J Radiat
Oncol Biol Phys 48(3): 889–898.
Fiorilli R, De Felice F, Violini R. 2006;
[Chronic total coronary occlusions: an
overview]. G Ital Cardiol (Rome) 7(12):
780–797.
Giachelli CM. 2004; Vascular calcification
mechanisms. J Am Soc Nephrol 15(12):
2959–2964.
Giachelli CM. 2005; Inducers and inhibitors
of biomineralization: lessons from patho-
logical calcification. Orthod Craniofac Res
8(4): 229–231.
Hehrlein C, Zimmermann M, Pill J, et al.
1994; The role of elastic recoil after bal-
loon angioplasty of rabbit arteries and
its prevention by stent implantation. Eur
Heart J 15(2): 277–280.
Hersh EM, Bodey GP. 1970; Leukocytic
mechanisms in inflammation. Annu Rev
Med 21: 105–132.
Hitani K, Yokoo S, Honda N, et al. 2008;
Transplantation of a sheet of human
corneal endothelial cell in a rabbit model.
Mol Vis 14: 1–9.
Hong MK, Vossoughi J, Haudenschild CC,
et al. 1995; Vascular effects of diet-
induced hypercalcemia after balloon
artery injury in giant Flemish rabbits. Am
Heart J 130(4): 758–764.
Hsu HH, Tawfik O, Sun F. 2004; Mechanism
of dystrophic calcification in rabbit aor-
tas: temporal and spatial distributions
of calcifying vesicles and calcification-
related structural proteins. Cardiovasc
Pathol 13(1): 3–10.
Hsu JJ, Tintut Y, Demer LL. 2008; Murine
models of atherosclerotic calcification.
Curr Drug Targets 9(3): 224–228.
Ishii A, Vinuela F, Murayama Y, et al. 2006;
Swine model of carotid artery atheroscle-
rosis: experimental induction by surgical
partial ligation and dietary hypercholes-
terolemia. Am J Neuroradiol 27(9):
1893–1899.
Jones JR, Tsigkou O, Coates EE, et al. 2007;
Extracellular matrix formation and min-
eralization on a phosphate-free porous
bioactive glass scaffold using primary
human osteoblast (HOB) cells. Biomate-
rials 28(9): 1653–1663.
Kipshidze N, Sadzaglishvili K, Panarella M,
et al. 2005; Evaluation of a novel endo-
luminal vascular occlusion device in a
porcine model: early and late follow-up.
J Endovasc Ther 12(4): 486–494.
Li JJ, Zhu CG, Yu B, et al. 2007; The role of
inflammation in coronary artery calcifica-
tion. Ageing Res Rev 6(4): 263–270.
Lysitsas DN, Katsouras CS, Papakostas JC,
et al. 2007; Antirestenotic effects of a
novel polymer-coated d-24851 eluting
stent. Experimental data in a rabbit iliac
artery model. Cardiovasc Intervent Radiol
30(6): 1192–1200.
Mimura T, Yamagami S, Yokoo S, et al.
2004; Cultured human corneal endothe-
lial cell transplantation with a collagen
sheet in a rabbit model. Invest Ophthalmol
Vis Sci 45(9): 2992–2997.
Montecucco F, Steffens S, Mach F. 2007;
The immune response is involved in
atherosclerotic plaque calcification: could
the RANKL/RANK/OPG system be a
marker of plaque instability? Clin Dev
Immunol. [Epub ahead of print]. PMID:
18320012.
Operschall C, Falivene L, Clozel JP, et al.
2000; A new model of chronic cardiac
ischemia in rabbits. J Appl Physiol 88(4):
1438–1445.
Prosser L, Agrawal CM, Polan J, et al. 2006;
Implantation of oxygen enhanced, three-
dimensional microporous L-PLA polymers:
a reproducible porcine model of chronic
total coronary occlusion. Catheter Cardio-
vasc Interv 67(3): 412–416.
Puma JA, Sketch MH Jr, Tcheng JE, et al.
1995; Percutaneous revascularization of
chronic coronary occlusions: an overview.
J Am Coll Cardiol 26(1): 1–11.
Radke PW, Heinl-Green A, Frass OM, et al.
2006; Evaluation of the porcine ameroid
constrictor model of myocardial ischemia
for therapeutic angiogenesis studies.
Endothelium 13(1): 25–33.
Raval AN, Karmarkar PV, Guttman MA,
et al. 2006; Real-time magnetic resonance
imaging-guided endovascular recanaliza-
tion of chronic total arterial occlusion
in a swine model. Circulation 113(8):
1101–1107.
Reffelmann T, Sensebat O, Birnbaum Y,
et al. 2004; A novel minimal-invasive
model of chronic myocardial infarction in
swine. Coron Artery Dis 15(1): 7–12.

Copyright  2011 John Wiley & Sons, Ltd. J Tissue Eng Regen Med (2011).
DOI: 10.1002/term

Sage AP, Tintut Y, Demer LL. 2010; Regula-
tory mechanisms in vascular calcification.
Nat Rev Cardiol 7(9): 528–536.
Sampath S, Raval AN, Lederman RJ, et al.
2007; High-resolution 3D arteriography of
chronic total peripheral occlusions using
a T1-W turbo spin-echo sequence with
inner-volume imaging. Magn Reson Med
57(1): 40–49.
Segev A, Nili N, Qiang B, et al. 2005;
Human-grade purified collagenase for the
treatment of experimental arterial chronic
total occlusion. Cardiovasc Revasc Med
6(2): 65–69.
Song W, Lee J, Kim H, et al. 2005; A new
percutaneous porcine coronary model of
chronic total occlusion. J Invasive Cardiol
17(9): 452–454.
Soon KH, Selvanayagam JB, Cox N, et al.
2007; Percutaneous revascularization of
chronic total occlusions: review of the
role of invasive and non-invasive imaging
modalities. Int J Cardiol 116(1): 1–6.
Stoff A, Rivera AA, Sanjib Banerjee N, et al.
2009; Promotion of incisional wound
repair by human mesenchymal stem
cell transplantation. Exp Dermatol 18(4):
362–369.
Stoll G, Bendszus M. 2006; Inflammation
and atherosclerosis: novel insights into
plaque formation and destabilization.
Stroke 37(7): 1923–1932.
Strauss BH, Goldman L, Qiang B, et al.
2003; Collagenase plaque digestion for
facilitating guide wire crossing in chronic
total occlusions. Circulation 108(10):
1259–1262.
Suzuki K, Saito N, Zhang G, et al. 2008;
Development of a novel calcified total
occlusion model in porcine coronary arter-
ies. J Invasive Cardiol 20(6): 296–301.
Suzuki Y, Oyane A, Ikeno F, et al. 2009;
Development of animal model for calcified
Copyright  2011 John Wiley & Sons, Ltd.
chronic total occlusion. Catheter Cardio-
vasc Interv 74(3): 468–475.
Tattini C, Manchio J, Zaporojan V, et al.
2008; Role of TGFβ and FGF in the treat-
ment of radiation-impaired wounds using
a novel drug delivery system. Plast Recon-
str Surg 122(4): 1036–1045.
Thind AS, Leung G, Munce NR, et al. 2007;
Investigation of micro-ultrasound for
microvessel imaging in a model of chronic
total occlusion. Ultrason Imaging 29(3):
167–181.
Tomaru T, Fujimori Y, Morita T, et al. 1996;
Local delivery of antithrombotic drug pre-
vents restenosis after balloon angioplasty
in atherosclerotic rabbit artery. Jpn Circ J
60(12): 981–992.
Tuzun E, Oliveira E, Narin C, et al. 2009;
Correlation of ischemic area and coronary
flow with ameroid size in a porcine model.
J Surg Res 164(1): 38–42.
Tyagi SC, Kumar S, Cassatt S, et al. 1996;
Temporal expression of extracellular
matrix metalloproteinases and tissue plas-
minogen activator in the development of
collateral vessels in the canine model of
coronary occlusion. Can J Physiol Pharma-
col 74(8): 983–995.
Wang JA, Fan YQ, Li CL, et al. 2005; Human
bone marrow-derived mesenchymal stem
cells transplanted into damaged rabbit
heart to improve heart function. J Zhejiang
Univ Sci B 6(4): 242–248.
White CJ, Ramee SR, Card HG, et al. 1988;
Laser angioplasty: an atherosclerotic
swine model. Lasers Surg Med 8(3):
318–321.
Wildemann B, Sander A, Schwabe P, et al.
2005; Short-term in vivo biocompati-
bility testing of biodegradable poly
(D,L-lactide)–growth factor coating
for orthopaedic implants. Biomaterials
26(18): 4035–4040.
B. Zhu et al.
Woodruff MA, Hutmacher DW. 2010; The
return of a forgotten polymer – polycapro-
lactone in the 21st century. Progr Polym
Sci 35: 1217–1256.
Yamanouchi K, Satomura K, Gotoh Y, et al.
2001; Bone formation by transplanted
human osteoblasts cultured within colla-
gen sponge with dexamethasone in vitro.
J Bone Miner Res 16(5): 857–867.
Yan H, Yu C. 2007; Repair of full-thickness
cartilage defects with cells of different ori-
gin in a rabbit model. Arthroscopy 23(2):
178–187.
Yang W, Ge J, Liu H, et al. 2006; Arsenic
trioxide eluting stent reduces neointima
formation in a rabbit iliac artery injury
model. Cardiovasc Res 72(3): 483–493.
Zhang MZ, Li CL, Jiang YT, et al. 2008; Por-
phyromonas gingivalis infection acceler-
ates intimal thickening in iliac arteries in
a balloon-injured rabbit model. J Periodon-
tol 79(7): 1192–1199.
Zhang ZY, Yin XH. 2008; Impact of
adenovirus-mediated local expression of
human tissue factor pathway inhibitor on
vascular smooth muscular cell prolifera-
tion and apoptosis in the stent-implanted
femoral artery of the rabbit. J Int Med Res
36(3): 567–571.
Zhu B, Bailey SR, Agrawal CM. 2010;
Engineering calcium deposits on poly-
caprolactone scaffolds for intravascu-
lar applications using primary human
osteoblasts. J Tissue Eng Regen Med. [Epub
ahead of print]. PMID: 20827712.
Zhu CC, Chen SL, Lin XF, et al. 2008; [Cre-
ate a standard mini-swine model of
chronic ischemic myocardium by thora-
coscopy]. Zhonghua Wai Ke Za Zhi 46(15):
1163–1165.
J Tissue Eng Regen Med (2011).
DOI: 10.1002/term