Cardiovascular Research 59 (2003) 997–1005
www.elsevier.com / locate / cardiores
Enhanced hindlimb collateralization induced by acidic fibroblast
growth factor is dependent upon femoral artery extraction
James C. Hershey a , *, Halea A. Corcoran a , Elizabeth P. Baskin a , David B. Gilberto b ,
Xianzhi Mao c , Kenneth A. Thomas c , Jacquelynn J. Cook a
a
b
Department of Pharmacology, Merck Research Laboratories, West Point, PA 19454, USA
Laboratory Animal Resources, Merck Research Laboratories, West Point, PA 19454, USA
c
Cancer Research, Merck Research Laboratories, West Point, PA 19454, USA
Received 5 February 2003; received in revised form 11 July 2003; accepted 14 July 2003
Abstract
Recent investigations have established the feasibility of using exogenously delivered angiogenic growth factors to increase collateral
artery development in animal models of myocardial and hindlimb ischemia. Objective: Our aim was to evaluate the ability of a stabilized
form of acidic fibroblast growth factor (aFGF-S 117 ) to stimulate collateralization and arteriogenesis in the rabbit hindlimb following the
surgical induction of ischemia by femoral artery extraction. A secondary objective was to examine angiogenic and arteriogenic effects of
aFGF-S 117 in the absence of a peripheral blood flow deficit. Methods and results: Five days after femoral artery removal, aFGF-S 117 (1,
3, or 30 mg / kg) was intramuscularly delivered into the hindlimb, three times per week for 2 consecutive weeks. End-point measurements
performed on day 20 found that hindlimb reserve blood flow was significantly improved in rabbits that received 3 or 30 mg / kg of
aFGF-S 117 , with no difference in efficacy between these two doses. These hemodynamic results were supported by angiographic evidence
showing enhanced density of collateral vessels in the medial thigh region and histological findings of increased capillary density within
the gastrocnemius muscle from rabbits treated with aFGF-S 117 . When an efficacious dose of 3 mg / kg of aFGF-S 117 was administered to
sham-operated rabbits with intact femoral arteries, there was no change in any of the blood flow, angiographic or histological parameters
measured. Conclusions: These findings demonstrate that a stabilized form of aFGF stimulated the development of functional collateral
arteries in the rabbit hindlimb, an effect which was dependent upon removal of the femoral artery. These results suggest that aFGF-S 117
may have therapeutic potential for the treatment of arterial occlusive disorders.
 2003 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.
Keywords: Angiogenesis; Growth factors; Collateral circulation; Arteries; Blood flow
1. Introduction
Peripheral vascular disease (PVD) is most commonly
caused by arterial obstructions due to advanced atherosclerotic lesions and has been reported to affect |12% of
the elderly population [1]. The most common methods of
treatment for patients with PVD are surgical revascularization techniques such as bypass procedures, endarterectomy
and percutaneous balloon angioplasty. Unfortunately,
many patients are excluded from these procedures due to
*Corresponding author. Tel.: 11-215-652-9943; fax: 11-610-6523811.
E-mail address: james [email protected] (J.C. Hershey).
]
the anatomical extent and distribution of the vascular
obstructions. The ability to augment or accelerate
neovascularization through the delivery of exogenous
growth factors has received much attention since this type
of therapy may be a useful adjunct or alternative to the
conventional procedures used to treat patients suffering
from PVD.
Early pre-clinical and clinical studies that utilized gene
transfer of angiogenic growth factors, such as vascular
endothelial growth factor (VEGF), showed promise as an
innovative approach to treat patients with critical limb
ischemia. However, VEGF gene therapy was also associTime for primary review 21 days.
0008-6363 / 03 / $ – see front matter  2003 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.
doi:10.1016 / S0008-6363(03)00522-4
998
J.C. Hershey et al. / Cardiovascular Research 59 (2003) 997–1005
ated with significant adverse effects such as peripheral
edema (34% of patients in one study), a condition which is
also observed in animal models [2,3]. Other unfavorable
effects associated with VEGF overexpression include
formation of fragile ‘‘angioma-like’’ capillaries and severe
neointimal proliferation [4–6]. Moreover, there have recently been studies using various animal models of hindlimb ischemia and different delivery methods that have
questioned the functional efficacy of VEGF [3,4,7]. These
studies have demonstrated that although VEGF did stimulate angiogenesis, it did not induce arteriogenesis or
improve collateral blood flow. Therefore, due to its uncertain efficacy and mechanism-based adverse effects, the
clinical utility of VEGF for the treatment of arterial
occlusive disorders has been questioned [8,9].
Clearly, it would be favorable to deliver a growth factor
that could enhance the growth and development of preexisting vessels (i.e. arteriogenesis), rather than solely
stimulating the sprouting of capillaries (i.e. angiogenesis).
Acidic fibroblast growth factor (aFGF or FGF-1) may have
several therapeutic advantages over VEGF. For example,
unlike VEGF, which is an endothelial cell specific growth
factor, aFGF is a powerful mitogen which stimulates
migration and proliferation of various cell types, such as
smooth muscle cells and fibroblasts which may favor the
development of thicker walled vessels. aFGF also stimulates endothelial cell production of matrix metalloproteinases required for vessel remodeling, and, finally, its high
affinity to heparin sulfates on the surface of endothelial
cells prolongs its half-life and pharmacodynamic action at
the FGF-1 receptor. Indeed, aFGF has been shown to be
active in animal models of dermal wound repair, large
vessel re-endothelialization and myocardial ischemia [10–
13].
The purpose of the present study was to determine if
intramuscular administration of the previously described
stabilized form of aFGF, denoted aFGF-S 117 [12–15],
could stimulate neovascularization and improve collateral
blood flow in ischemic and non-ischemic rabbit hindlimbs.
Compared to other members of the FGF family of proteins,
a cysteine residue at position 117 is unique to aFGF and
replacement of Cys 117 to an isomorphorous serine (S 117 )
does not diminish the intrinsic mitogenic activity of the
protein but substantially increases the half-life of the active
mitogen at physiologic temperature and pH both in the
absence and presence of heparin [14,15].
2. Methods
2.1. Animal care
Fifty-four adult New Zealand White rabbits (male, 3–4
kg) were used for the completion of this study. The rabbits
were individually housed in a temperature controlled
(2061 8C) room with a 12 h / 12 h light / dark cycle with
food and water provided ad libitum. All animal related
procedures were approved by the Institutional Animal Care
and Use Committee at Merck Research Laboratories, West
Point, PA and conform with the Guide for the Care and
Use of Laboratory Animals published by the US National
Institutes of Health (NIH Publication No. 85-23, revised
1996).
2.2. Experimental design
The effect of aFGF-S 117 on collateral-dependent blood
flow was assessed in rabbits after unilateral ligation and
extraction of the femoral artery. We have previously shown
that this method of femoral artery extraction has a transient
(,5 days) effect on resting blood flow, but a marked and
prolonged (,40 days) effect on reserve flow [16]. Thus,
these animals serve as a model of intermittent claudication
and not the more severe critical limb ischemia that is
associated with resting pain.
For the initial surgery, the rabbits were pre-medicated
with xylazine (2.5 mg / kg, i.m.) and anesthetized with
ketamine (70 mg / kg, i.m.). Under sterile surgical conditions, a longitudinal incision was made on the medial
thigh of one ‘‘experimental’’ hindlimb extending inferiorly
from the inguinal ligament to a point just proximal to the
patella. Through this incision, the external iliac artery was
isolated, ligated twice and sectioned between the ligatures.
The distal ligature was used to retract and dissect free the
entire length of the common femoral artery to the point
where it bifurcates into the saphenous and popliteal
arteries. The ligation of the femoral artery results in
retrograde thrombus propagation to the origin of the
external iliac artery. Therefore, blood flow to the hindlimb
after removal of the femoral artery is dependent upon flow
from the internal iliac artery and collateral vessels. All
rabbits were closely monitored by veterinary staff and
received an analgesic (buprenorphine, 0.04 mg / kg s.c.,
b.i.d.) and antibiotic (enrofloxacin, 6.0 mg / kg, s.c.) for 3
days following surgery.
Five days after surgery, the rabbits were randomly
assigned to one of four experimental groups and treated
three times per week for two consecutive weeks with either
the purified recombinant aFGF-S 117 protein at 1 mg / kg
(n510), 3 mg / kg (n512), 30 mg / kg (n511), or the
vehicle (n512), which contained 0.1 mg / ml human serum
albumin and 12 mg / ml heparin in PBS. The dose of
aFGF-S 117 indicates the amount administered during each
of the six treatments. The aFGF-S 117 protein was confirmed to be highly active in a standard Balb / c 3T3 cell
mitogenesis assay (ED 50 595 pg / ml). A separate group of
¨
naıve
rabbits (n59) underwent a sham operation (i.e.
femoral artery remained intact) in order to study the effects
of aFGF-S 117 in the absence of a peripheral blood flow
deficit. For all animals, the aFGF-S 117 was administered as
a purified protein via direct intramuscular injection into the
hindlimb (0.5 ml / site over five muscle sites; adductor
J.C. Hershey et al. / Cardiovascular Research 59 (2003) 997–1005
longus, adductor magnus, vastus medialis, semimembranosus and gastrocnemius). All hemodynamic and
anatomic measurements were performed 20 days after
surgery.
2.3. Measurement of hindlimb blood flow
Twenty days after the initial surgery, resting and reserve
hindlimb blood flow were measured in each rabbit as
previously described [16]. Briefly, anesthetized rabbits
were placed in the supine position, intubated and mechanically ventilated with room air. A midline incision was
made in the lower abdomen and the descending aorta was
isolated at the point of bifurcation into the common iliac
arteries. The common iliac arteries were then isolated and
both epigastric arteries were ligated to prevent blood flow
to the lateral compartments of the abdomen and pelvic
cavity. Perivascular flow probes (1.5 RB, Transonic Systems, Inc.) were placed on the right and left common iliac
arteries for continuous and simultaneous measurement of
blood flow to both limbs. To obtain a functional measurement of collateral blood flow, the hyperemic response to a
brief arterial occlusion (ao) was measured. An arterial clip
was temporarily applied to the common iliac artery and
four separate tests (10, 20, 40 and 60 s in duration) were
performed in duplicate with 10 min recovery time allowed
between tests. The results are presented as the increase in
blood volume (ml), which equals the total area under the
curve during the first 30 s after release of the arterial clip,
minus the baseline blood flow during the time of integration.
999
2.5. Histological analysis
Immediately after the animal was sacrificed (20 days
post-surgery, 15 days after the start of dosing), both
hindlimbs were dissected and transverse sections of the
adductor and gastrocnemius muscles were removed, covered with OCT compound and frozen in liquid nitrogen.
Tissue sections (10 mm thickness) were stained for alkaline
phosphatase using the indoxyl-tetrazolium method to detect capillary endothelial cells and counterstained with
eosin as previously described [16,18,19]. The number of
capillaries was determined using an image analysis system
(Phase 3 Imaging) that was programmed to distinguish
between the dark blue alkaline phosphatase staining of the
capillaries and the pink eosin staining of skeletal muscle
cells. The system automatically identified and counted the
number of capillaries in a chosen field (a total of 60 fields
from six different sections were analyzed per muscle). To
ensure that capillary density was not overestimated due to
tissue edema or underestimated due to muscle atrophy, the
total area of muscle was also determined to produce a
measurement of capillaries / mm 2 of muscle.
2.6. Statistical analysis
All data are presented as mean6S.E.M. Comparisons
within the same animal (e.g. experimental vs. contralateral
limb) were performed using a paired Student’s t-test. For
intergroup comparisons, a one-way ANOVA followed by
Fishers PLSD test for multiple comparisons was used.
Statistical significance was accepted at the P#0.05 level.
2.4. Vascular imaging
Following hindlimb blood flow measurements, angiography was performed as previously described by us and
others to obtain an anatomical index of the growth and
development of the larger collateral vessels in the medial
portion of the upper hindlimb [16,17]. For this, a polyethylene catheter was inserted into the common iliac artery
of the experimental limb, with the tip of the catheter
positioned just proximal to the origin of the internal iliac
artery. The hindlimb was positioned 20 cm below the
output beam of a fluoroscope (General Electric, Stenoscop), and the vascular bed was dilated with an intraarterial
bolus injection of 300 mg sodium nitroprusside. Immediately thereafter, iodinated contrast media (Isovue-370) was
infused intraarterially at a rate of 60 ml / min. Perfusion of
the hindlimb was observed on a monitor in real time and
an angiographic image was taken exactly 4 s after the start
of contrast media infusion. As previously described, a
circular grid overlay was used by one blinded investigator
to count the number of vessels that transect a circle to
obtain a quantitative measurement of the density of
collateral vessels in the medial thigh region [16,17].
3. Results
3.1. Effect of aFGF-S 117 on resting and reserve
hindlimb blood flow
When measured 20 days after femoral artery removal,
there was no significant difference in the resting hindlimb
blood flow between the experimental and contralateral
limbs in any treatment group (Fig. 1). This time course of
reestablished resting flow is consistent with other reports
using similar hindlimb models in rats and rabbits [20,21].
In contrast to the resting flow, reserve blood flow (i.e.
collateral-dependent flow) was significantly increased in
the experimental limbs of rabbits dosed with either 3 or 30
mg / kg of aFGF-S 117 (Fig. 2). There was no difference in
the efficacy observed between the two highest doses of
aFGF-S 117 , and the 1 mg / kg dose was not different from
vehicle-treated rabbits. The reserve blood flow measured in
the contralateral limb, which did not undergo surgery or
drug treatments, was similar among all groups. It is also
noteworthy that the reserve blood flow capacity measured
1000
J.C. Hershey et al. / Cardiovascular Research 59 (2003) 997–1005
3.2. Collateral vessel growth and development
Representative angiographic images of the experimental
limb from vehicle and aFGF-S 117 treated rabbits are shown
in Fig. 3A–D. In response to femoral artery removal, it can
be noticed that the internal iliac artery gives rise to supply
vessels that transverse the medial thigh region of the limb.
Compared to the vehicle treated rabbits, the angiographic
images from the rabbits that received 3 and 30 mg / kg of
aFGF-S 117 displayed a more dense network of vessels. The
increased vascular density was evident by the increased
angiographic score in rabbits that were treated with 3 and
30 mg / kg of aFGF-S 117 (63.564.7 and 59.766.8, respectively) compared to rabbits that received 1 mg / kg
(44.666.2) or the vehicle (40.562.1) (Fig. 4).
Fig. 1. Resting hindlimb blood flow measured 20 days after femoral
artery extraction was similar between limbs with no differences among
treatment groups. The aFGF-S 117 protein or the vehicle was delivered
intramuscularly into the experimental limb (9), while the contralateral
limb (j) served as a internal control and did not undergo surgery or
receive treatment.
in the experimental limbs from the rabbits treated with
either 3 or 30 mg / kg aFGF-S 117 remained significantly
below the values obtained in the contralateral limbs.
3.3. Skeletal muscle capillary density
Tissue sections from the adductor (composed of slow,
type I fibers located in upper thigh) and gastrocnemius
(composed of fast, type II fibers located in lower limb)
muscles from both the experimental and contralateral limbs
were stained for alkaline phosphatase (AP) to estimate
capillary density. Analysis of the adductor muscle (Fig. 5,
top panel) revealed no statistical differences in the capillary density between the experimental and contralateral
limb muscles among any treatment group, however, as
expected, the capillary density in the white adductor
muscle was lower than that of the red gastrocnemius. As
can be seen, the capillary density in the gastrocnemius
muscles (Fig. 5, bottom panel) of the experimental limb in
all groups was higher compared to the non-operated
contralateral limb. The elevated capillary density in the
gastrocnemius muscle of the experimental limb in the
vehicle group is most likely due to the effect of ischemia
alone since we observe a similar effect in operated, but
untreated rabbits. More importantly, the capillary density
in the gastrocnemius muscles obtained from the experimental limbs of the rabbits treated with 3 or 30 mg / kg
aFGF-S 117 (709669 and 743677 caps / mm 2 , respectively)
was significantly increased compared to the corresponding
experimental limb of vehicle treated rabbits (539651
caps / mm 2 ).
3.4. Effect of aFGF-S 117 on non-ischemic hindlimbs
Fig. 2. Effect of aFGF-S 117 on hindlimb reserve blood flow. The reserve
flow capacities of the experimental (top panel) and contralateral (bottom
panel) limbs were assessed by directly measuring the increase in blood
flow (i.e. hyperemia) in response to a series of temporary arterial
occlusions lasting 10, 20, 40 or 60 s. *P,0.05, compared to the vehicle
group.
A separate group of rabbits (n511) was utilized to
determine the effects of aFGF-S 117 on the existing hindlimb vasculature in the absence of femoral artery extraction. Normal rabbits underwent a sham operation (i.e.
the femoral artery remained intact) and were dosed with
the previously determined efficacious dose of 3 mg / kg
aFGF-S 117 using the identical dosing regimen as described
above for rabbits that underwent femoral artery extraction.
The same hemodynamic and anatomical measurements
J.C. Hershey et al. / Cardiovascular Research 59 (2003) 997–1005
1001
Fig. 3. Representative angiographic images of the experimental limb of rabbits treated with either vehicle (A) or aFGF-S 117 at doses of 1 mg / kg (B), 3
mg / kg (C) or 30 mg / kg (D). Note the increased density of collateral vessels in the medial thigh region of rabbits treated with 3 and 10 mg / kg.
were performed 15 days after the start of dosing. Interestingly, as shown in Table 1, intramuscular administration of
aFGF-S 117 into a normal limb had no effect on any of the
parameters measured (reserve blood flow, angiographic
score or capillary density).
4. Discussion
Fig. 4. Effect of aFGF-S 117 on the growth and development of angiographically visible collateral vessels. The angiographic score represents the
density of collateral vessels in the medial thigh region of the experimental
limb. *P,0.05, compared to the vehicle group.
Following arterial occlusion or major stenosis, blood
flow through collateral arteries is required to maintain
tissue viability. Normally, these collateral vessels are either
not present or do not function as major arterial conduits.
Enhancing the growth and development of these collateral
vessels by the administration of an exogenous growth
factor represents an interesting approach for the treatment
of arterial occlusive diseases. The primary finding of the
present study was that intramuscular administration of a
stabilized form of aFGF enhanced the development of
collateral vessels and improved reserve blood flow in
hindlimbs following femoral artery removal, but had no
1002
J.C. Hershey et al. / Cardiovascular Research 59 (2003) 997–1005
Fig. 5. The effect of aFGF-S 117 on skeletal muscle capillary density was
determined for the adductor (top panel) and gastrocnemius (bottom
panel) muscles from both the experimental (9) contralateral (j) limbs.
Top panel: In the adductor muscle, no differences in capillary density
were observed either between limbs or among treatments groups. Bottom
panel: There was a significant increase in the capillary density of
gastrocnemius muscle from the experimental limb compared to the
contralateral limb. This result was observed in all groups, however
treatment with the two highest doses of aFGF-S 117 resulted in a further
increase in capillary density. *P,0.05 compared to the contralateral limb;
†P,0.05 compared to the experimental limb of the vehicle group.
Table 1
Effect of intramuscular administration of aFGF-S 117 (3 mg / kg / dose) in
sham-operated rabbits (n59) with an intact femoral artery
Parameter
Capillary density (caps /mm 2 )
Gastrocnemius muscle
Adductor muscle
aFGF-S 117
treated limb
Contralateral
non-treated limb
387649
245635
410644
221629
Angiographic score
Resting blood flow (ml / min)
14.662.1
18.660.9
16.562.6
18.161.1
Reserve blood flow (ml)
10 s ao
20 s ao
40 s ao
60 s ao
4.160.4
6.660.5
10.960.8
12.061.1
4.860.6
6.460.4
10.260.9
11.761.2
Values are mean6S.E.M. There are no statistical differences in any of
the values between the aFGF-S 117 treated limb and the contralateral
non-treated limb. ao, arterial occlusion.
angiogenic or arteriogenic effect in normally perfused
limbs.
The results from our study demonstrated that collateraldependent blood flow in the experimental limbs of the
rabbits administered 3 and 30 mg / kg of aFGF-S 117 was
|70% higher compared to vehicle treated rabbits when
measured 20 days after femoral artery removal. The
increases in reserve blood flow and angiographic score
were similar between the rabbits that received 3 and 30
mg / kg, while injections of 1 mg / kg aFGF-S 117 had no
effect and were no different from the vehicle group. Thus,
maximally effective and no-effect doses of aFGF-S 117
were defined. These results are similar to a previous study
by Yang, using a rat model of hindlimb ischemia in which
they showed intraarterial administration of bFGF into the
iliac artery for 2 weeks at 5 or 50 mg / kg / day was equally
effective in stimulating collateral-dependent blood flow,
while 0.5 mg / kg / day was less effective [22]. Tabata et al.
[23] also showed a similar dose–response profile using
plasmid delivery of a secreted form of aFGF in a rabbit
model of hindlimb ischemia. In their study, there was no
improvement in any of the anatomic or functional parameters with low doses of plasmid (100–300 mg), however a
statistically significant increase in these indices was demonstrated at a dose of 500 mg, with no further augmentation at higher doses (700–1000 mg). Furthermore, the
doses found to be efficacious in our study are similar to the
dose used in the phase II study to investigate the efficacy
of recombinant FGF-2 in patients with intermittent claudication (the TRAFFIC study). In that study, a single
intraarterial injection of 30 mg / kg of FGF-2 improved
peak walking time during a graded treadmill-exercise test
performed 90 days after treatment [24]. Interestingly, a
second dose of FGF-2 administered 30 days after the first
injection conferred no additional advantage. Although our
study has shown that aFGF-S 117 stimulated a significant
increase in the reserve flow capacity of the experimental
limb, we recognize that complete restoration did not occur,
as the reserve values of the experimental limb were |30%
of the contralateral limb. This sub-maximal improvement
is similar to the |20% of normal response observed by
others using bFGF [25], and it is possible that the partial
improvements observed in these studies may be due the
time frame of the experiment and / or sub-optimal protocols
for the frequency and duration of growth factor administration.
There are several reports using various animal models of
hindlimb ischemia or peripheral arterial insufficiency that
have addressed the issues of duration and timing of growth
factor treatment in relation to the time of end-point
measurements. The blood flow measurements in our study
were performed 20 days after surgery because there is an
adequate window to observe efficacy beyond the wellknown spontaneous collateralization induced by ischemia
alone. The results obtained with our study design may not
be observed with alternative dosing protocols and different
J.C. Hershey et al. / Cardiovascular Research 59 (2003) 997–1005
days of end-point measurement (e.g. 30 or 40 days after
surgery). However, others have found that bFGF-induced
improvement in hindlimb collateral blood flow in rats was
independent of the duration of bFGF administration [22].
In that report, intraarterial administrations lasting 1, 3, or
14 days yielded similar increases in calf muscle blood flow
when measured 16 days after femoral artery ligation.
Importantly, in those experiments, the bFGF treatment was
initiated on the same day as the surgical ligation of the
femoral artery. Those results suggested that the duration of
treatment was not as important as the timing of the
treatment as long as it coincided with the onset of
ischemia. It has been shown that shortly after the experimental induction of a vascular occlusion, nearby
vessels that likely serve as subsequent collateral conduits
exhibit proliferative activity and upregulation of certain
growth factor receptors that may be necessary for vascular
remodeling [26]. Specifically, using a rabbit model similar
to ours, Deindl and colleagues recently reported increased
expression of FGF receptor-1 (FGFR-1) during the early
phase of arteriogenesis [27]. Their elegant immunohistochemical studies performed 3 days after femoral artery
ligation, showed specific localization of FGFR-1 in smooth
muscle cells of the collateral arteries that crossed the
medial thigh region of the limb. Therefore, administration
of an exogenous growth factor, such as aFGF, may only
need to be present during the ‘‘vascular activation’’ or
early phase of arteriogenesis which temporally coincides
with the development of tissue ischemia or the presence of
a transcollateral pressure gradient. Thus, the results of our
study complement the previous work of others by showing
that responsiveness to aFGF-S 117 was maintained when the
initiation of treatment was temporally dissociated from the
induction of the vascular obstruction by 5 days.
It is well recognized that a flow deficit is a powerful
stimulus for vascular remodeling, even without the intervention of an exogenous growth factor(s) [16,26]. The
remodeling of existing vessels to form functional arterial
conduits has been termed ‘‘arteriogenesis’’, and is clearly
distinct from de novo formation of new capillaries by the
process of angiogenesis. In response to a vascular stenosis
or occlusion, it is arteriogenesis which is believed to be the
more important and compensatory form of vascular growth
required to maintain tissue perfusion [7,16,28]. Although
arteriogenesis was observed via angiography in all animals
in the present study, in contrast to the modestly defined
arterial tree seen in the vehicle group, the aFGF-S 117
treated animals exhibited an expanded vascular tree that
was more dense and extended further distally. These
results are consistent with previous studies using rat and
rabbit models of hindlimb ischemia that showed intraarterial administration of bFGF into the hindlimb led to increased proliferation of endothelial and smooth muscle
cells [29], enlargement of existing conduit vessels and
increased collateral-dependent blood flow [25,29,30]. It is
also interesting to note that aFGF-S 117 had an angiogenic
1003
effect (i.e. increased capillary sprouting) in a skeletal
muscle located in the lower hindlimb. This result agrees
with the work of Ito et al. who, using a similar rabbit
model, showed that in the lower limb where the greatest
perfusion deficits occur, endothelial cell proliferation takes
place in the absence of smooth muscle cell proliferation,
suggesting that angiogenesis predominates over collateral
remodeling in this region [26].
Although the process of arteriogenesis is not completely
understood, it is not likely related to the metabolic
consequences of ischemia, since the tissue surrounding the
collateral conduits has been shown by others to receive
adequate blood flow [25,26,31]. It is more likely that the
hemodynamic changes related to increased flow velocity,
wall stress and shear stress, and / or monocyte involvement
are important in the remodeling process [29]. Thus, it may
be that the FGF receptors in the endothelium are upregulated by these altered flow dynamics leading to an
increased responsiveness to exogenous aFGF. However, it
is unknown whether the beneficial effects of aFGF are a
result of a primary arteriogenic affect or potentiation of
this activity by facilitating or inducing the activity of other
growth factors. We acknowledge that the beneficial effects
of aFGF-S 117 observed in the present study may not be
characteristic of a clinical situation, since the method of
arterial obstruction may have simplified the remodeling
process. Also, prior knowledge of the area of vascular
obstruction allowed site-specific delivery of the growth
factor. Nonetheless, it is possible that the advantageous
clinical situation may exist where following the identification of a peripheral vascular obstruction through imaging,
site-specific delivery of aFGF-S 117 into the tissue that is
most likely to respond could minimize side-effects and
enhance efficacy.
Our results clearly demonstrated that in the absence of
femoral artery extraction, aFGF-S 117 had no effect on any
of the parameters of angiogenesis, arteriogenesis and
hindlimb blood flow. When a previously determined effective dose of aFGF-S 117 was administered to normal rabbit
limbs with an intact femoral artery, we found no changes
in limb blood flow (resting or reserve), angiographic score
or capillary density. These results suggest that adequately
perfused skeletal muscle is unresponsive to exogenous
aFGF and, to our knowledge, this is the first evidence
showing that existing vessels that ultimately serve as
collateral conduits are unresponsive to aFGF, unless a
vascular obstruction exists in the region. This finding
supports the idea that creation or formation of a vascular
obstruction produces site-specific alterations in nearby
vessels which allow them to become responsive to exogenous aFGF.
Although the increased reserve blood flow measured in
response to aFGF-S 117 treatment appeared to be physiologically relevant, we did not evaluate its impact on exercise
tolerance or muscle function. However, using a rat model
of hindlimb ischemia, it has been shown that improve-
1004
J.C. Hershey et al. / Cardiovascular Research 59 (2003) 997–1005
ments in calf muscle function were proportional to the
increases in hindlimb blood flow induced by either bFGF
administration [25], increased physical activity [32], or a
combination of bFGF administration and enhanced physical activity [33]. The goal of future experiments will be to
determine whether the aFGF-S 117 -induced changes observed in this study would translate to enhanced exercise
tolerance or performance. It should also be noted that our
results with the S 117 form of aFGF are consistent with and
complement a previous study showing the efficacy of
WT-aFGF in the same rabbit model [23]. However, it is
difficult to directly compare our results with this study
since the growth factor was delivered in different forms
(plasmid vs. purified protein) and routes of administration
(intra-arterial vs. intramuscular). Although both forms of
aFGF (S 117 and WT) function through the same receptor, it
is unknown whether the prolonged stability and half-life of
aFGF-S 117 would translate into improved efficacy in vivo
and / or increased side-effects compared to the WT form.
In summary, the present study demonstrated that intramuscular administration of a stabilized form of aFGF
stimulated the growth and development of collateral
vessels and produced a marked improvement in hindlimb
reserve blood flow in a rabbit model of peripheral arterial
insufficiency. The surgical removal of the femoral artery
was a prerequisite for aFGF-S 117 activity and suggests that
alterations in blood flow or the induction of ischemia
produces a local environment that is responsive to exogenous aFGF-S 117 . These results suggest aFGF-S 117 may have
therapeutic potential since similar vascular remodeling in
patients with intermittent claudication could stimulate a
meaningful improvement in collateral blood flow and
enhance mobility.
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
Acknowledgements
The authors gratefully acknowledge the surgical and
animal care support from the Department of Laboratory
Animal Resources, Merck Research Laboratories. We also
thank William Huckle and Rosemary McFall for characterization of aFGF-S 117 in cell-based assays.
[18]
[19]
[20]
References
[1] Criqui MH, Fronek A, Barrett-Connor E et al. The prevalence of
peripheral arterial disease in a defined population. Circulation
1985;71:510–515.
[2] Baumgartner I, Rauh G, Pieczek A et al. Lower-extremity edema
associated with gene transfer of naked DNA encoding vascular
endothelial growth factor. Ann Intern Med 2000;6:880–884.
[3] Hershey JC, Baskin EP, Corcoran HA, et al. VEGF stimulates
angiogenesis without improving collateral blood flow following
hindlimb ischemia in rabbits. Heart and Vessels (in press).
[4] Lee RJ, Springer ML, Blanco-Bose WE et al. VEGF gene delivery to
[21]
[22]
[23]
[24]
myocardium: deleterious effects of unregulated expression. Circulation 2000;102:898–901.
Lazarous DF, Shou M, Scheinowitz M et al. Comparative effects of
basic fibroblast growth factor and vascular endothelial growth factor
on coronary collateral development and the arterial response to
injury. Circulation 1996;94:1074–1082.
Masaki I, Yonemitsu Y, Yamashita A 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;90:966–973.
Deindl E, Buschmann I, Hoefer IE et al. Role of ischemia and of
hypoxia-inducible genes in arteriogenesis after femoral artery occlusion in the rabbit. Circ Res 2001;89:779–786.
Helisch A, Schaper W. Angiogenesis and arteriogenesis—not yet for
prescription. Z Kardiol 2000;89:239–244.
Epstein SE, Kornowski R, Fuchs S, Dvork HF. Angiogenesis
therapy: amidst the hype, the neglected potential for serious side
effects. Circulation 2001;104:115–119.
Mellin TN, Mennie RJ, Cashen DE et al. Acidic fibroblast growth
factor accelerates dermal wound healing. Growth Factors 1992;7:1–
14.
Bjornsson TD, Dryjski M, Tluczek J et al. Acidic fibroblast growth
factor promotes vascular repair. Proc Natl Acad Sci USA
1991;88:8651–8655.
Lopez JJ, Edelman ER, Stamler A et al. Angiogenic potential of
perivascularly delivered aFGF in a porcine model of chronic
myocardial ischemia. Am J Physiol 1998;274:H930–H936.
Sellke FW, Li J, Stamler A et al. Angiogenesis induced by acidic
fibroblast growth factor as an alternative method of revascularization
for chronic myocardial ischemia. Surgery 1996;120:182–188.
Linemeyer DL, Menke JG, Kelly LJ et al. Disulfide bonds are
neither required, present, nor compatible with full activity of human
recombinant acidic fibroblast growth factor. Growth Factors
1990;3:287–298.
Ortega S, Schaeffer MT, Soderman D et al. Conversion of cysteine
to serine residues alters the activity, stability, and heparin dependence of acidic fibroblast growth factor. J Biol Chem
1991;266:5842–5846.
Hershey JC, Baskin EP, Glass JD et al. Revascularization in the
rabbit hindlimb: dissociation between capillary sprouting and arteriogenesis. Cardiovasc Res 2001;49:618–625.
Takeshita S, Rossow ST, Kearney M et al. Time course of increased
cellular proliferation in collateral arteries after administration of
vascular endothelial growth factor in the rabbit model of lower limb
insufficiency. Am J Pathol 1995;147:1649–1660.
Zaida AM, Hudlicka O, Tyler KR, Wright AJ. The effect of
long-term vasodilation on capillary growth and performance in
rabbit heart and skeletal muscle. Cardiovasc Res 1984;18:724–732.
Baffour RJ, Berman J, Garb JL 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. J Vasc Surg 1992;16:181–191.
Hendricks DL, Pevec WC, Shestak KC et al. A model of persistent
partial hindlimb ischemia in the rabbit. J Surg Res 1990;49:453–
457.
Mackie BG, Terjung RL. Blood flow to different skeletal muscle
fiber types during contraction. Am J Physiol Heart Circ Physiol
1983;245:H265–H275.
Yang HT, Feng Y, Allen LA, Protter A, Terjung RL. Efficacy and
specificity of bFGF increased collateral flow in experimental
peripheral arterial insufficiency. Am J Physiol Heart Circ Physiol
2000;278:H1966–H1973.
Tabata H, Silver M, Isner JM. Arterial gene transfer of acidic
fibroblast growth factor for therapeutic angiogenesis in vivo: critical
role of secretion signal in use of naked DNA. Cardiovasc Res
1997;35:470–479.
Lederman RJ, Mendelsohn FO, Anderson RD et al. Therapeutic
J.C. Hershey et al. / Cardiovascular Research 59 (2003) 997–1005
[25]
[26]
[27]
[28]
angiogenesis with recombinant fibroblast growth factor-2 for intermittent claudication (the TRAFFIC study): a randomised trial.
Lancet 2002;359:2053–2058.
Yang HT, Deschenes MR, Ogilvie RW, Terjung RL. Basic fibroblast
growth factor increases collateral blood flow in rats with femoral
arterial ligation. Circ Res 1996;76:62–69.
Ito WD, Arras M, Scholz D et al. Angiogenesis but not collateral
growth is associated with ischemia after femoral artery occlusion.
Am J Physiol 1997;42:H1255–H1265.
Deindl E, Hoefer IE, Fernandez B et al. Involvement of the
fibroblast growth factor system in adaptive and chemokine-induced
arteriogenesis. Circ Res 2003;92:561–568.
van Royen N, Piek JJ, Buschmann I et al. Stimulation of arteriogenesis; a new concept for the treatment of arterial occlusive
disease. Cardiovasc Res 2001;49:543–553.
1005
[29] Arras M, Ito WD, Sholz D et al. Monocyte activation in angiogenesis and collateral growth in the rabbit hindlimb. J Clin Invest
1998;101:40–50.
[30] Battler AM, Scheinowitz A, Bor D et al. Intracoronary injection of
basic fibroblast growth factor enhances angiogenesis in infarcted
swine myocardium. J Am Coll Cardiol 1993;22:2001–2006.
[31] Lindner V, Lappi DA, Baird A, Majack RA, Reidy MA. Role of
basic fibroblast growth factor in vascular lesion formation. Circ Res
1991;68:106–113.
[32] Yang HT, Ogilvie RW, Terjung RL. Heparin increases exerciseinduced collateral blood flow in rats with femoral artery ligation.
Circ Res 1995;76:448–456.
[33] Yang HT, Ogilvie RW, Terjung RL. Exercise training enhances basic
fibroblast growth factor-induced collateral blood flow. Am J Physiol
Heart Circ Physiol 1998;274:H2053–H2061.