Cardiovascular Research 61 (2004) 132 – 142
www.elsevier.com/locate/cardiores
Conduction performance of collateral vessels induced by vascular
endothelial growth factor or basic fibroblast growth factor
Keisuke Kondoh a,b,c, Hiroyuki Koyama a,b,c,*, Tetsuro Miyata c, Tsuyoshi Takato b,
Hirohumi Hamada d, Hiroshi Shigematsu c
a
Department of Vascular Regeneration, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan
b
Division of Tissue Engineering, The University of Tokyo Hospital, Tokyo, Japan
c
Division of Vascular Surgery, Department of Surgery, Graduate School of Medicine, The University of Tokyo, 7-3-1, Hongo, Bunkyo, Tokyo 113-8655, Japan
d
Department of Molecular Medicine, Sapporo Medical University, Sapporo, Japan
Received 7 July 2003; received in revised form 30 September 2003; accepted 6 October 2003
Time for primary review 24 days
Abstract
Objective: In the present study, we delivered vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF)
gene to a rabbit model of hind limb ischemia utilizing an ex vivo method of gene transfer, and evaluated the functional performance of the
developed collateral vessels. Method: The left femoral artery of a male Japanese White rabbit was excised to induce limb ischemia, and a
section of skin was resected for culture of auto-fibroblasts. Twenty days later, the VEGF gene, bFGF gene or h-galactosidase gene (LacZ)
was adenovirally transferred to the cultured auto-fibroblasts (5 106 cells), and the next day, a pair of specifically infected fibroblasts
(total 1 107 cells) was injected via the left internal iliac artery of the same rabbit. Pairs of transferred genes into the fibroblasts were as
follows: LacZ/LacZ (control group), VEGF/LacZ (VEGF group), bFGF/LacZ (FGF group) and VEGF/bFGF (combination group).
Twenty-eight days after cell administration, collateral development and its function were evaluated. Results: Calf blood pressure ratio,
resting blood flow of the left iliac artery and capillary density of ischemic muscle showed similar degrees of angiogenic effects in the
VEGF and FGF groups, which were significantly greater than those in the control group. On the contrary, angiographic score, collateral
conductance and smooth muscle cell (SMC)-positive vessel density in the FGF group were significantly greater than those in the VEGF
group. In the combination group, collateral conductance showed synergistic effects, and in vivo blood flow and smooth muscle cellpositive vessel density revealed additive effects of VEGF and bFGF. Conclusion: These findings suggested that bFGF-induced collateral
development exceeded VEGF-induced collateral development in the induction of arteriogenesis, and that combined gene delivery of VEGF
and bFGF produced additive or synergistic effects of collateral development as compared with the effects induced by transfer of each gene
alone.
D 2003 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.
Keywords: Vascular endothelial growth factor; Basic fibroblast growth factor; Therapeutic angiogenesis; Arteriogenesis; Ex vivo gene transfer; Collateral
vessels
1. Introduction
Induction of angiogenic reactions to enhance collateral
vessel development is a promising approach for the treatment of vascular occlusive disease, and this concept has
* Corresponding author. Division of Vascular Surgery, Department of
Surgery, Graduate School of Medicine, The University of Tokyo, 7-3-1,
Hongo, Bunkyo, Tokyo 113-8655, Japan. Tel.: +81-3-5800-8653; fax: +813-3811-6822.
E-mail address: [email protected] (H. Koyama).
been termed ‘‘therapeutic angiogenesis’’ [1]. A variety of
procedures have been presented for the effective development of collateral vessels, and the most widely studied
strategy is local delivery of angiogenic growth factors, such
as vascular endothelial growth factor (VEGF) [1], basic
fibroblast growth factor (bFGF) [2] and others [3]. Direct
administration of recombinant protein was initially investigated to deliver these growth factors [1,2], and subsequently, several methods using gene transfer techniques were
established to achieve more efficient and sophisticated
delivery [3 –5]. In previous studies, we presented a new
0008-6363/$ - see front matter D 2003 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.cardiores.2003.10.003
K. Kondoh et al. / Cardiovascular Research 61 (2004) 132–142
delivery strategy, in which auto-fibroblasts, adenovirally
transduced with angiogenic growth factor gene, were selectively injected into the donor artery (ex vivo method). The
donor artery indicates a potential origin of collateral development to ischemic lesion with sufficient blood in-flow, and
we consider that the peripheral site of the donor artery is an
ideal delivery target in angiogenic therapy. The reason for
utilizing fibroblasts as delivery carrier was that the fibroblast
is a non-blood cell which is readily feasible for primary
culture. Thus, the most of gene-transduced fibroblasts were
mechanically trapped in peripheral small vessels of the
donor artery, achieving specific gene delivery. Indeed, our
previous studies showed excellent development of collateral
vessels in animal models of limb ischemia [6] and myocardial ischemia [7].
Selection of growth factors for delivery is also critical for
promoting appropriate development of collateral vessels to
ischemic tissues. Several growth factors are known to
induce angiogenic processes in vivo [1 – 3], and the
expected effects might differ according to the growth factor,
since each growth factor has a unique mechanism for
expressing its activity. Indeed, there are three processes in
forming new vessels—vasculogenesis, angiogenesis and
arteriogenesis [8], and the functional performance of developed collateral vessels possibly depends upon which process is predominantly induced [9]. Although previous
studies reported favorable results of collateral augmentation
by using a variety of growth factors [1– 3,5], few studies
have characterized the acquired angiogenic effects and
compared them between different growth factors [10].
In the present study, we delivered VEGF or bFGF by
using the ex vivo gene delivery system described above, and
compared the characteristics of collateral development promoted by these factors in a rabbit model of chronic hind
limb ischemia. Further, combined delivery of VEGF and
bFGF was carried out in the same manner, and the possibility of synergistic effects was evaluated.
133
control vector for the animal experiment was AxCALacZ,
which varies from the Escherichia coli LacZ cDNA.
2.2. In vitro study
Rabbit fibroblasts (passage 3), cultured in 60-mm
dishes, were infected with AxCAhVEGF at 20 plaqueforming units (p.f.u)/cell in 1 ml Dulbecco’s modified
Eagle’s minimum essential medium (DMEM, Gibco
BRL, NY, USA) with 2% FBS. One hour later, the cells
were washed twice with PBS and cultured in 3 ml of the
same medium. The medium was changed daily, and stored
at 1, 4, 7, 14, 21, and 28 days after infection. Another set
of rabbit fibroblasts not infected was cultured in the same
culture medium, and the medium was changed daily and
stored as the control. The concentration of VEGF in each
medium was measured by ELISA using a VEGF ELISA kit
(R&D Systems, MN, USA) according to the supplier’s
instructions.
2.3. Animal model of chronic hind limb ischemia
We used a rabbit model of hind limb ischemia to
evaluate in vivo angiogenic responses. Male Japanese
White rabbits weighing 3.0 – 3.5 kg (Saitama Rabbitry,
Saitama, Japan) were anesthetized by intramuscular injection of a mixture of ketamine (50 mg/kg) and xylazine (2.5
mg/kg), and the left femoral artery was completely removed from its proximal origin to the bifurcation formed
by the saphenous and popliteal arteries as described
previously [1,5,6]. Before closure of the wound, 10 10
mm skin section was obtained for primary culture of
fibroblasts. All protocols were approved by the Guide for
Care and Use of Laboratory Animals by the US National
Institute of Health (NIH publication No. 85-23, revised
1996).
2.4. Ex vivo gene transfer
2. Materials and methods
2.1. Adenovirus vector
Replication-deficient recombinant adenovirus vector
containing a modified human bFGF gene with the secretory
signal sequence of interleukin-2 (AxCAMAssbFGF) was
used for gene transfer of bFGF [6]. Synthesized bFGF is
secreted from infected cells because of the signal sequence,
and our previous study showed that cultured rabbit fibroblasts significantly secreted bFGF into the medium from 1
to 28 days after gene transfer with AxCAMAssbFGF [6]. To
transfer the VEGF gene, we constructed an adenovirus
vector containing the human VEGF165 gene (AxCAhVEGF)
in the same expression unit. The cDNA of the human
VEGF165 was obtained using RT-PCR, which matched with
the sequence presented in GenBank (AF 022375). The
Fibroblasts were cultured from the resected skin tissue,
and grown to confluence in 100-mm dishes as described
previously [6,7]. At 20 days after femoral artery excision,
5 106 fibroblasts (three passages) were infected in vitro
with AxCALacZ, AxCAMAssbFGF or AxCAhVEGF at
20 p.f.u/cell and incubated for 24 h. Subsequently at day
21, a 3-French end-hole infusion catheter (Rapid Transitk, Cordis, FL, USA) was introduced into the left
internal iliac artery via the carotid artery under fluoroscopic guidance. After double washing with PBS, the
infected fibroblasts in DMEM with 2% FBS were injected
as a bolus through the infusion catheter. The injected cells
were composed of two sets of 5 106 fibroblasts infected
with different adenoviruses as follows; AxCALacZ/
AxCALacZ as control (control group: n = 7), AxCALacZ/AxCAMAssbFGF (FGF group: n = 7), AxCALacZ/
AxCAhVEGF (VEGF group: n = 7), and AxCA-
134
K. Kondoh et al. / Cardiovascular Research 61 (2004) 132–142
Fig. 1. Experimental protocol.
MAssbFGF/AxCAhVEGF (combination group: n = 7).
Heparin and other anticoagulants were not used during
any of the procedures (Fig. 1).
Pharmaceutical, Osaka, Japan) through the same catheter. In
vivo blood flow was calculated as previously described [6].
2.7. Collateral conductance
2.5. Calf blood pressure ratio
Calf blood pressure was measured in both hind limbs,
and calf blood pressure ratio, defined as the ratio of left
systolic pressure to right systolic pressure, was calculated as
previously described [6]. The measurement was carried out
at six time-points as follows: immediately after femoral
artery removal, and immediately before, immediately after,
and 7, 14 and 28 days after administration of gene-transduced fibroblasts.
2.6. Angiographic score and in vivo blood flow
Selective angiography of the left internal iliac artery was
performed at 28 days after cell injection. A 3-French endhole catheter was introduced into the left iliac artery
through the carotid artery, and the catheter tip was positioned at the level of the middle of the first sacral vertebra.
Angiography was performed as described previously, and
the angiographic score was determined [6]. Additionally,
the diameter of the proximal part of the left caudal gluteal
artery was measured.
Before selective internal iliac arteriography, a 0.014-inch
Doppler guide wire (Endo Sonics, CA, USA) was introduced
via the 3-Fench catheter to the proximal part of the left
internal iliac artery. Average peak velocity (APV) was measured at rest, and subsequently maximum APV was determined after bolus injection of 2 mg papaverine (Dainippon
At 28 days after injection of gene-transduced fibroblasts,
we measured collateral conductance according to the method of Schaper et al. [11– 15] with modification. Animals
were intubated via tracheotomy and ventilated with room
air. The jugular vein was cannulated for infusion of lactated
Ringer’s solution and occasional injections of pentobarbital
to maintain an appropriate level of anesthesia. After bolus
injection of 2000 Units heparin, the right femoral artery was
cannulated with a 24-gauge polyethylene infusion catheter
(inner diameter: 0.47 mm, Terumo, Tokyo, Japan) for
sampling of a microsphere reference, and the left saphenous
artery was also cannulated with a 24-gauge infusion catheter
just above the ankle. Then, the distal aorta was exposed and
cannulated in proximal and distal directions with 14-gauge
polyethylene infusion catheters (inner diameter: 1.73 mm,
Terumo). To perfuse both hind limbs at a stable flow rate, a
pump-driven shunt was inserted between the proximal and
distal catheters; the proximal-directed catheter was used for
blood withdrawal, and the distal-directed catheter for infusion of blood. The catheter of the saphenous artery and the
distal-directed catheter in the aorta were connected to a
pressure transducer system to record proximal pressure (PP)
and distal pressure (DP) of collateral vessels. Subsequently,
adenosine (Kowa Shinyaku, Tokyo, Japan) was continuously infused (1 mg/kg/min) into the shunt circuit to achieve
maximum dilation of the hind limb vessels, and the shunt
system was driven to perfuse blood flow at two different
K. Kondoh et al. / Cardiovascular Research 61 (2004) 132–142
135
2.9. Serum level of bFGF and VEGF
To assess the systemic level of bFGF and VEGF after
injection of the gene-transduced fibroblasts, serum sampling
was carried out immediately before and 7, 14, and 28 days
after cell administration. The serum samples were also
obtained from normal rabbits (n = 4) as normal control.
Serum concentrations of bFGF and VEGF were measured
by ELISA using a bFGF ELISA kit and a VEGF ELISA kit
(R&D Systems).
2.10. In vivo expression of VEGF
Fig. 2. ELISA showing in vitro secretion of VEGF from AxCAhVEGFtreated rabbit fibroblasts. There was significant secretion of VEGF in the
culture medium from 1 to 10 days after gene transfer. C, control sample; d,
days after infection. *P < 0.01 (versus values at 14, 21 and 28 days).
proximal pressure level between 45 and 70 mm Hg. At each
pressure level, colored microspheres (diameter: 15 Am, DyeTrak, Triton Technology, CA, USA) with different color
(eosin and yellow) were injected into the shunt circuit, and a
reference blood sample was withdrawn for 3 min at rate of
0.6 ml/min. After the above procedures, the distal half of the
whole thigh muscles and whole crural muscles were dissected. The number of microspheres accumulated in the
dissected muscles was counted according to the supplier’s
manual, and muscle blood flow (BF) was calculated as
described previously [11]. Collateral conductance was defined as 100 BF/(PP DP). The mean of collateral conductance analyzed under different proximal pressure levels
was used for statistical analysis.
Rabbits of the ischemic model were killed immediately
before and 4, 7 and 14 days after the injection of 5 106
AxCAhVEGF-treated fibroblasts (n z 4 at each time point),
and the left adductor muscles were taken. One group of
normal rabbits (n = 4) was killed as normal control. The
muscle samples ( c 1 g) were lysed in 2 ml of lysis buffer,
and the concentrations of VEGF were measured by using a
VEGF ELISA kit (R&D Systems). The lysis buffer, used in
this analysis, was cell culture lysis reagent (Promega, WI,
USA) diluted with PBS (1:5) to minimize the influence of
detergent.
2.11. Statistical analysis
Results are expressed as mean F S.D. To adjust for
multiple comparisons, Tukey – Kramer method was used
for analyzing all angiogenic effects. ANOVA was carried
out to examine the interaction between VEGF and bFGF
(synergistic effect) in collateral conductance, in vivo blood
flow, capillary density and SMC-positive vessel density.
2.8. Histological study
Tissue specimens were obtained as transverse sections
from the left semimembranous muscle of each animal at 28
days after cell administration, and embedded in OCT
compound (Miles, NJ, USA). Frozen sections (5 Am thick)
were cut, and capillary endothelial cells were stained by the
indoxyl-tetrazolium method. A total of 20 different fields
from three sections per animal were randomly selected, and
capillary/muscle fiber ratio were determined as described
previously (capillary density) [6]. To evaluate the maturation of arteries, three sections per animal were immunostained for smooth muscle actin. The monoclonal antibody
against a-smooth muscle actin (1A4, DAKO, Copenhagen,
Denmark) was applied at a 1:500 dilution after blocking
with 1% normal horse serum. Subsequent incubation with
biotinylated horse anti-mouse IgG (1.2 Ag/ml, Vector Laboratories, CA, USA) and an ABC Elite kit (Vector Laboratories) was performed. The number of vessels surrounded by
smooth muscle cells (SMC) was counted in each section, the
area of the whole section was measured as described
previously [16], and density of SMC-positive arteries was
analyzed (SMC-positive vessel density).
Fig. 3. Time course of calf blood pressure ratio (ratio of left calf systolic
pressure to right calf systolic pressure). d, days after infection. *P < 0.01
(versus control group and VEGF group at 14 days); yP < 0.01 (versus other
groups at 28 days), zP < 0.01 (versus control group at 28 days); NS, no
significant difference.
136
K. Kondoh et al. / Cardiovascular Research 61 (2004) 132–142
Serum levels of VEGF and bFGF were also evaluated by
repeated measures ANOVA. All data were considered
significant at P < 0.05
10 days after gene transfer, and maximum secretion was
observed at day 4 (Fig. 2).
3.2. Calf blood pressure ratio
3. Results
3.1. In vitro study
The secretion of VEGF in the culture medium from
infected rabbit fibroblasts increased significantly from 1 to
The ratio of left calf systolic pressure to right calf systolic
pressure (calf blood pressure ratio) showed no significant
difference in all the groups before and at 7 days. At 14 days,
the calf blood pressure ratio in the combination group was
significantly higher than that in the control group and VEGF
group. At 28 days, the calf blood pressure ratio in the VEGF
Fig. 4. Selective left internal iliac angiograms of control group (a), VEGF group (b), FGF group (c) and combination group (d) at 28 days after cell
administration. Arrow indicates left internal iliac artery. Development of collateral vessels was quantified by the angiographic score at day 28 (e). The diameter
of left caudal gluteal artery was also measured (f). *P < 0.01 (versus other groups), yP < 0.05, zP < 0.01(versus FGF and combination groups), §P < 0.05 (versus
FGF and VEGF groups).
K. Kondoh et al. / Cardiovascular Research 61 (2004) 132–142
137
and FGF groups was significantly higher than that in the
control group, and the ratio in the combination group was
even greater than that in the VEGF and FGF groups. No
difference was detected between the VEGF and FGF groups
(Fig 3). We also measured the calf blood pressure ratio
immediately after cell administration to assess the influence
of intra-arterial injection of 1 107 fibroblasts, and no
decrease was observed as compared with the ratio immediately before injection (data not shown). Further, no significant difference was detected between the right calf systolic
pressure immediately before and after cell injection (data
not shown).
3.3. Angiographic score
At 28 days after injection of gene-transduced cells,
selective internal iliac arteriographs showed few collateral
vessels in the control group (Fig 4a), whereas marked
development of collateral vessels was observed in the VEGF,
FGF and combination groups (Fig 4b,c and d). To quantify
the development of collateral vessels, angiographic score
was calculated. The score was highest in the combination
Fig. 5. In vivo blood flow of left internal iliac artery, measured with
Doppler guide wire. APV was recorded at rest (a), maximum APV (b) was
measured after papaverine injection, and then blood flow was calculated.
*P < 0.01(versus other groups); yP < 0.01, zP < 0.05, §P < 0.05 (versus other
groups); NS, no significant difference.
Fig. 6. Collateral conductance, the reciprocal of blood flow resistance in
collateral vessels, representing the net function of developed collateral
vessels. *P < 0.01 (versus other groups), yP < 0.05.
Fig. 7. (a) Capillary density of left semimembranous muscle measured in
fresh frozen cross-section stained by indoxyl-tetrazolium method. (b) SMCpositive vessel density of left semimembranous muscle evaluated in crosssection immunostained with anti-smooth muscle actin antibody. *P < 0.05
(versus other groups), yP < 0.05 (versus control and VEGF groups),
z
P < 0.01 (versus control and VEGF groups), §P < 0.05.
138
K. Kondoh et al. / Cardiovascular Research 61 (2004) 132–142
group, followed by the FGF group, and the VEGF group (Fig
4e). Significant differences were detected between every pair
of groups; the FGF and VEGF groups, and the VEGF and
control groups. The diameter of the left caudal gluteal artery
was significantly larger in the FGF and combination groups
than in the VEGF and control groups (Fig 4f).
significant increase in the VEGF, FGF, and combination
groups as compared with the control group. There was no
significant difference between the VEGF and FGF groups
(Fig 5a). Maximum blood flow was highest in the combination group, and there was significant difference between
every pair of groups (Fig 5b).
3.4. In vivo blood flow
3.5. Collateral conductance
In vivo blood flow of the left internal iliac artery at rest,
measured with a Doppler guide wire at day 28, showed a
At 28 days after cell injection, the collateral conductance
in the combination group was highest in all groups, which
Fig. 8. Additive results of VEGF gene-promoted effect and bFGF gene-promoted effect (left bar), and the result after combined gene delivery of VEGF and
bFGF (right bar). Synergistic effects was detected in collateral conductance (a), and additive effects were observed in blood flow (resting and maximum) (b),
capillary density (c), and SMC-positive score (d). Add, additive effects of VEGF group and FGF group. Combi, combination group. Error bars are not shown.
*P < 0.05 (positive synergistic effects produced by VEGF and bFGF).
K. Kondoh et al. / Cardiovascular Research 61 (2004) 132–142
139
was approximately four times that in the control group and
close to the level of normal animals (Fig 6). The next high
collateral conductance was observed in the FGF group,
followed by the VEGF group. There were significant differences between every pair of groups.
VEGF groups. SMC-positive vessel density in the combination group was even higher than that in the FGF group
(Fig 7b).
3.6. Histological analysis
A significant synergistic effect of VEGF and bFGF was
detected only in collateral conductance (Fig 8a). In vivo
blood flow of the left internal iliac artery (resting and
maximum) and SMC-positive score showed no synergism
with combined use, while the data of the combination group
were significantly higher than those of the VEGF and FGF
groups, which represented the additive effects by the combination (Fig 8b,c). We did not carry out ANOVA analysis of
Capillary density in the VEGF, FGF and combination
groups was significantly higher than that in the control
group, while no difference was detected between the three
groups (Fig 7a). SMC-positive vessel density was significantly higher in the FGF group than in the control group,
though no difference was detected between the control and
3.7. Synergistic effect
Fig. 9. Time course of serum VEGF (a) and bFGF (b) levels measured by ELISA. d, days after infection.
140
K. Kondoh et al. / Cardiovascular Research 61 (2004) 132–142
Fig. 10. Time course of VEGF expression in left adductor muscle. The data
were presented as a ratio of VEGF content to total protein content. d, days
after infection. *P < 0.01 (versus values of before cell administration, 14
days and normal groups).
synergism for calf blood pressure ratio and angiographic
score, because both evaluations were not expressed as
continuous variable.
3.8. Time course of serum level of bFGF and VEGF
ANOVA detected no significant change in serum VEGF
(Fig 9a) or bFGF (Fig 9b) through the all time-points.
3.9. In vivo expression of VEGF
The expression of VEGF in the left adductor muscle was
significantly increased at 4 and 7 days after cell administration, though the expression level at day 14 was similar to
that before cell injection (Fig 10).
4. Discussion
The present study tried to compare the characteristics of
VEGF-promoted collateral development with that of bFGFpromoted collateral development in a rabbit model of hind
limb ischemia, in which an ex vivo gene transfer method
was utilized to achieve specific delivery to peripheral
vessels of the donor artery [6,7]. The in vivo expressions
of VEGF and bFGF after the delivery were confirmed in the
present and previous studies. In order to fairly compare the
effects promoted by such different growth factors, certain
references indicating the conditions of delivery must be
standardized, since each growth factor possibly possesses an
independent profile as bioactive substance. The present
study showed no significant difference in the calf blood
pressure ratio, resting blood flow of the left internal iliac
artery and capillary density in ischemic muscle tissue
between the VEGF and FGF groups, while these VEGF
and FGF group data were significantly higher than those in
the control group. Calf blood pressure ratio and in vivo
blood flow are conventional parameters representing general
degrees of collateral vessel development [1]. Capillary
density of the ischemic muscle is a parameter of angiogenesis and vasculogenesis, since angiogenesis and vasculogenesis refer to the process of capillary formation from
endothelial cells or endothelial progenitor cells. These three
evaluation methods have been frequently utilized and established in several previous studies investigating angiogenic
reactions in vivo [1,3,5,6]. Consequently, it was possible to
consider that the gross angiogenic effects observed in the
VEGF group were of a similar level to those in the FGF
group, which might enable comparison of the angiogenic
character between VEGF and bFGF. Indeed, the in vitro
data of the present study showed VEGF secretion of 1.8 ng/
1000 cells/day at 4 days after AxCAhVEGF infection, and
our previous study showed bFGF secretion of 25 pg/1000
cells/day at 4 days after AxCAMAssbFGF infection [6];
thus, the protein secretion ratio of VEGF to bFGF was
approximately 72 under these conditions. Asahara et al. [17]
reported that a 50-fold greater dose of VEGF was required
for the same degree of bFGF-induced angiogenic effects,
supporting the above results.
In such a situation, angiographic score, collateral conductance and SMC-positive vessel density in the FGF group
were significantly greater than those in the VEGF group.
Angiographic score reflects the number of vessels with
relatively large diameter, since only visible vessels on
angiograms are counted for calculation of this score. Collateral conductance represents the function of collateral
vessels to conduct blood flow to ischemic sites [11– 15],
and functional vessels require the maturation of wall structure and a certain luminal diameter. SMC-positive vessel
density also indicates the development of mature vessels,
because mature vessels possess SMC around the endothelial
layer [8]. Thus, angiographic score, collateral conductance
and SMC-positive vessel density are all regarded as valuable parameters of arteriogenesis. The findings of the
present study, therefore, showed that bFGF-induced collateral development exceeded VEGF-induced collateral development in the induction of arteriogenesis, when the same
level of angiogenic reaction was promoted in the both
groups.
Arteriogenesis refers to the maturation processes of
preexisting small vessels, and these processes comprise an
enlargement of vessel size through vascular remodeling
[8,9]. The diameter of vessels defines the resistance of
blood flow, and it is assured that a larger vessel can conduct
a greater volume of blood. Meanwhile, vasculogenesis and
angiogenesis are other forms of vessel growth. Vasculogenesis in adults is the process by which endothelial
progenitor cells differentiate and replicate to form a primitive network of blood vessels, and angiogenesis represents
the process of endothelial replication and migration resulting in the formation of capillary networks [18]. Since newly
formed vessels in vasculogenesis and angiogenesis are
constructed mainly from endothelial cells, the diameter of
K. Kondoh et al. / Cardiovascular Research 61 (2004) 132–142
these vessels would be too small to conduct adequate blood
flow to ischemic tissues. Even if vasculogenesis and angiogenesis generate new vessel networks, subsequent arteriogenesis is necessary to develop functional collateral vessels
[8,9]. Thus, arteriogenesis might be the most important
process in angiogenic therapy. Indeed, recent studies demonstrated that the delivery of some factors potentially
inducing arteriogenesis, such as MCP-1 [12], LPS [13],
GM-CSF [14], TGF-h1 [15], augmented favorable development of functional collateral vessels in animal models of
ischemia, suggesting the importance of arteriogenesis in
angiogenic therapy. In these studies, measurement of collateral conductance was carried out to evaluate the net
function of the developed collateral vessels. We also measured collateral conductance in the present study, and
showed a clear difference in the quality of developed
collaterals after the delivery of VEGF or bFGF.
It might be reasonable to conclude that bFGF-induced
collateral development was advantageous in the induction of
arteriogenesis, as compared with VEGF-induced collateral
development. Previous studies showed that an increase in
sheer stress on the vessel wall induced infiltration of macrophages around vessels [19], which released several bioactive
substances and then promoted arteriogenesis [13,20].
Among angiogenic growth factors, bFGF is known to be a
potent stimulator of arteriogenesis [9,21]. The processes of
arteriogenesis involve a variety of vessel wall cells including endothelial cell, SMC and fibroblasts, and bFGF has
broad potential to act as a growth factor for these cells [22].
In addition to arteriogenesis, bFGF also promotes angiogenesis by stimulating endothelial cells [9,21]. Contrarily,
VEGF, which is another well-known angiogenic growth
factor, is a specific promoter of angiogenesis, since VEGF
is a growth factor for endothelial cells. Further, a recent
study showed that VEGF possibly mobilizes endothelial
progenitor cells from the bone marrow [23]. Since the target
cell type of bFGF differs from that of VEGF, some difference in the form and function of the developed collateral
vessels could be expected. Masaki et al. [10] also delivered
the VEGF gene or bFGF gene to the mouse ischemic limb
by their Sendai virus-based gene delivery system, and the
developed vessels after gene transfer of bFGF were more
mature than those after VEGF gene transfer, supporting
above results of this study.
Another important finding of the present study was that
combined gene delivery of VEGF and bFGF promoted
significantly greater angiogenic effects in some evaluations,
as compared with sole gene delivery of VEGF or bFGF.
Especially significant synergistic effects on collateral conductance were produced by combined gene delivery. Similar
evidence has been reported by other groups. Asahara et al.
[9] administrated a single intra-arterial bolus of VEGF and
bFGF protein to the rabbit ischemic hind limb, and showed
synergistic or additive effects in some evaluations of angiogenic responses. Further, Chae et al. [24] injected naked
plasmid DNA of VEGF and angiopoietin-1, and showed
141
additive effects on angiographic score and capillary density.
One possible explanation for the synergistic and additive
effects is that each growth factor described above has an
independent receptor system and specific character for
inducing angiogenic effects. Namely, although VEGF is a
strong angiogenic factor specific to endothelial cells, it is
also known that VEGF increases the permeability of capillaries [25]. Previous studies indeed described the occurrence
of local edema after VEGF delivery [10,26]. Since leakage
of the fluid component and worsening of edema might
decrease the efficiency of tissue perfusion, there is a
possibility that the net blood flow conduction in capillaries
formed by VEGF would be considerably less than the
expected blood flow matching to the gross volume of the
capillary bed. In contrast, bFGF does not only promote the
proliferation of endothelial cells, but also induces development of the medial layer and adventitia [22], which could
sustain and support the endothelial layer from the outside.
Therefore, bFGF possibly increases the function of VEGFinduced capillaries by preventing leakage of fluid, resulting
in synergistic angiogenic effects after combined gene delivery of VEGF and bFGF.
In summary, we delivered the VEGF or bFGF gene to the
rabbit ischemic hind limb by utilizing an ex vivo method of
gene transfer, and demonstrated in vivo that induction of
arteriogenesis after bFGF delivery was greater than that after
VEGF delivery in certain condition. Since arteriogenesis,
which represents vessel maturation and enlargement, is
considered to be the most fundamental process in angiogenic therapy for ischemic diseases, bFGF might be suitable
as an angiogenic growth factor to induce the development of
functional collateral vessels. Furthermore, the present study
showed that combined gene delivery of VEGF and bFGF
produced additive or synergistic effects on collateral development, as compared with the effects induced by gene
transfer of VEGF or bFGF alone.
Acknowledgement
We thank Dr. Takuhiro Yamaguchi (Department of
Biostatistics, School of Health Science and Nursing, The
University of Tokyo) for statistical pieces of advice.
References
[1] Takeshita S, Zheng LP, Brogi E. Therapeutic angiogenesis. A single
intraarterial bolus of vascular endotherial growth factor augments
revascularization in a rabbit ischemic hind limb model. J Clin Invest
1994;93:662 – 70.
[2] Unger EF, Banai S, Epstein SE, et al. Basic fibroblast growth factor
enhances myocardial collateral flow in a canine model. Am J Physiol
1994;266(4 pt 2):H1588 – 95.
[3] Taniyama Y, Morishita R, Ogihara T, et al. Therapeuic angiogenesis
induced by human hepatocyte growth factor gene in rat and rabbit
hindlimb ischemia models: preclinical study for treatment of peripheral arterial disease. Gene Ther 2001;8(3):181 – 9.
142
K. Kondoh et al. / Cardiovascular Research 61 (2004) 132–142
[4] Feldman LJ, Steg PG, Isner JM, et al. Low-efficacy of percutaneous
adenovirus-mediated arterial gene transfer in the atherosclerotic
rabbit. J Clin Invest 1995;95:2662 – 71.
[5] Tsurumi Y, Takeshita S, Isner JM, et al. Direct intramuscular gene
transfer of naked DNA encoding vascular endothelial growth factor
augments collateral development and tissue perfusion. Circulation
1996;94(12):3281 – 90.
[6] Ohara N, Koyama H, Miyata T, et al. Adenovirus-mediated ex vivo
gene transfer of basic fibroblast growth factor promotes collateral
development in a rabbit model of hind limb ischemia. Gene Ther
2001;8(11):837 – 45.
[7] Ninomiya M, Koyama H, Miyata T, et al. Ex vivo gene transfer of
basic fibroblast growth factor improves cardiac function and blood
flow in a swine chronic myocardial ischemia model. Gene Ther
2003;10(14):1152 – 60.
[8] Carmeliet P. Mechanisms of angiogenesis and arteriogenesis. Nat
Med 2000;6:389 – 95.
[9] van Royen N, Piek JJ, Schaper W, et al. Stimulation of arteriogenesis;
a new concept for the treatment of arterial occlusive disease. Cardiovasc Res 2001;49:543 – 53.
[10] Masaki M, Yonemitsu Y, Sueishi K, et al. Angiogenic gene therapy
for experimental critical limb ischemia: acceleration of limb loss by
overexpression of vascular endothelial growth factor 165 but not of
fibroblast growth factor-2. Circ Res 2002;90:966 – 73.
[11] Hoefer IE, van Royen N, Buschmann IR, Piek JJ, Schaper W. Time
course of arteriogenesis following femoral artery occlusion in the
rabbit. Cardiovasc Res 2001;49:609 – 17.
[12] Ito WD, Arras M, Winkler B, Schaper W. Monocyte chemotactic
protein-1 increases collateral and peripheral conductance after femoral
artery occlusion. Circ Res 1997;80(6):829 – 37.
[13] Arras M, Ito WD, Schaper W, et al. Monocyte activation in angiogenesis and collateral growth in the rabbit hindlimb. J Clin Invest
1998;101:40 – 50.
[14] Buschmann IR, Hoefer IE, Schaper W, et al. GM-CSF: a strong
arteriogenic factor acting by amplification of monocyte function.
Atherosclerosis 2001;159:343 – 56.
[15] van Royen N, Hoefer I, Schaper W, et al. Exogenous application of
transforming growth factor beta 1 stimulates arteriogenesis in the
peripheral circulation. FASEB J 2002;16:432 – 4.
[16] Koyama H, Reidy MA. Reinjury of arterial lesions induces intimal
smooth muscle cell replication that is not controlled by fibroblast
growth factor 2. Circ Res 1997;80:408 – 17.
[17] Asahara T, Bauters C, Zheng LP. Synergistic effect of vascular endothelial growth factor and basic fibroblast growth factor on angiogenesis in vivo. Circulation 1995;92:365 – 71.
[18] Asahara T, Murohara T, Isner JM, et al. Isolation of putative progenitor endothelial cells for angiogenesis. Science 1997;275:964 – 7.
[19] Shyy YJ, Hsieh HJ, Usami S, Chien S. Fluid shear stress induces a
biphasic response of human monocyte chemotactic protein 1 gene
expression in vascular endothelium. Proc Natl Acad Sci U S A 1994;
91:4678 – 82.
[20] Arras M, Strasser R, Schaper J, et al. Tumor necrosis factor-alpha is
expressed by monocytes/macrophages following cardiac microembolization and is antagonized by cyclosporine. Basic Res Cardiol 1998;
93:97 – 107.
[21] Baffour R, Berman J, Friedmann P, et al. Enhanced angiogenesis and
growth of collaterals by in vivo administration of recombinant basic
fibroblast growth factor in a rabbit model of acute lower limb ischemia: dose – response effect of basic fibroblast growth factor. J Vasc
Surg 1992;16:181 – 91.
[22] Klagsburn M. The fibroblast growth factor family: structural and biological properties. Prog Growth Factor Res 1989;1:207 – 35.
[23] Asahara T, Takahashi T, Isner JM, et al. VEGF contributes to postnatal neovascularization by mobilizing bone marrow-derived endothelial progenitor cells. EMBO J 1999;18(14):3964 – 72.
[24] Chae JK, Kim I, Koh GY, et al. Coadministration of angiopoietin-1
and vascular endothelial growth factor enhances collateral vascularization. Arterioscler Thromb Vasc Biol 2000;20(12):2573 – 8.
[25] Connolly DT, Heuvelman DM, Feder J, et al. Tumor vascular permeability factor stimulates endothelial cell growth and angiogenesis.
J Clin Invest 1989;84:1470 – 8.
[26] Vajanto I, Rissanen TT, Yla-Herttuala S, et al. Evaluation of angiogenesis and side effects in ischemic rabbit hindlimbs after intramuscular injection of adenoviral vectors encoding VEGF and LacZ.
J Gene Med 2002;4(4):371 – 80.