163
Stimulation of Angiogenesis by Adenosine on
the Chick Chorioallantoic Membrane
JERRY W.
DUSSEAU, PHILLIP M. HUTCHINS, AND DANIELLE S. MALBASA
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
The effect of adenosine on the vascular density of the chick chorioallantoic membrane was studied.
Elvax polymer pellets containing 0.2-3.0 mg of adenosine were placed on the chorioallantoic membrane of 10-day embryos. Control pellets containing mannitol were placed at least 1 cm away. After 4
days the membrane was formalin-fixed and removed. A thin plastic coverslip, inscribed with concentric circles (4-8 mm in diameter), was placed over the pellet. A vascular density index was estimated at
20 x by counting the number of intercepts between vessels and the inscribed circles. Adenosine
stimulated a dose-dependent increase (p<0.01) in the vascular density index with the 3-mg pellets
inducing a 15% increase. Inosine, a major metabolite of adenosine, did not cause a change in the
number of intercepts counted. The adenosine-stimulated increase in vascularity was blocked with 110
fig of methyl-isobutyl-xanthine injected daily into the albumin. Partial inhibition was observed with
55 figlday. Methyl-isobutyl-xanthine by itself did not affect the vascular density index. Dipyridamole
enhanced adenosine's stimulation of vascular growth an additional 52%. Given alone, however, it had
no effect on the membrane's vascularity. These data support an angiogenic role for adenosine. The
modest, but consistent, increase in the vascular density index stimulated by adenosine, and the fact
that it may be released during tissue hypoxia, is consistent with an hypothesis that this nucleoside plays
a modulatory role in vessel proliferation accompanying conditions of long-term hypoxia. (Circulation
Research 1986;59:163-170)
N
UMEROUS studies have demonstrated that
the tissue oxygen tension can influence vascular tone and blood flow.1"4 It is generally
agreed that these vascular responses to acute changes
in oxygen tension are autoregulatory in nature, directed toward matching blood flow with local metabolic
requirements. Over extended periods, a proliferation
or involution of blood vessels is also closely related to
changes in local tissue perfusion and metabolic needs.
For example, increased vascularity often follows the
decreased oxygen availability associated with chronic
exercise,5 high altitude adaptation,6 and wound repair.7
Conversely, a hyperoxic environment can result in a
reduced vascular density.8-9 While the role of oxygen is
not understood, its availability to satisfy chronic metabolic demands appears to be an integral component of
the signal initiating a change in vessel number.
During vascular proliferation, locally dilated venules and capillaries characteristically give rise to the
new vascular sprouts. 10 " It is well established that
microvascular dilation occurs in response to a decrease
in oxygen tension.1"3 Whether this change in vessel
caliber (perhaps via changes induced in local blood
flow) provides an adequate stimulus for neovascularization is not known. However, Lombard and Duling12
dissociated the oxygen tension from the time course of
vasodilation during reactive hyperemia, suggesting
that at least one other locally produced substance was
required to maintain the vasodilated state. Locally produced vasodilator metabolites may play such a role.13
Significantly, several vasoactive metabolites produced with tissue hypoxia have been implicated in the
angiogenesis process. Among these are the prostaglandins,14"16 ADP17 and lactic acid.18 In addition, pharmacological studies suggest a role for adenosine. Dipyridamole, a vasodilator that inhibits the activity of
adenosine deaminase and adenosine re-uptake, was
shown to elicit a neovascular response in rat cardiac
and skeletal muscle.1*"21
As various vasoactive metabolites function acutely
to match oxygen supply to oxygen need, they may
also, when produced over extended periods, perform
an analogous role of maintaining metabolic equilibrium via promoting local vascular growth. In this experiment vasoproliferative activity for adenosine, a vasodilator produced in response to tissue hypoxia, is
demonstrated using the chick chorioallantoic membrane. An established model for angiogenesis, the
chorioallantoic membrane exhibits a vasoproliferative
response to selected stimuli. Furthermore, studies employing altered environmental oxygen tensions have
reported changes of this membrane's vascular function,8 as well as various oxygen transport parameters.22-23
From the Department of Physiology and Pharmacology, Bowman Gray School of Medicine, Wake Forest University, WinstonSalem, North Carolina.
Supported by National Institutes of Health Grants HL-13936-12,
HL-O73O4-O8, and HL-07458-05.
Address for reprints: Dr. Jerry W. Dusseau, Department of,
Physiology and Pharmacology, Bowman Gray School of Medicine,
Wake Forest University, Winston-Salem, NC 27103.
Received June 17, 1985; accepted May 30, 1986.
Materials and Methods
Fertile eggs were obtained locally and incubated at
38°C and 55% relative humidity for 7 days. The eggs
were rotated twice daily. On Day 7 of incubation a 1 x
1.5 cm window was placed over the chorioallantoic
membrane (CAM). The surface of the egg and all
instruments were disinfected with 70% isopropyl alcohol. The window was closed with clear cellophane tape
164
and the incubation continued without rotation until
Day 10. In placing the window, it was particularly
important to avoid shell dust or fragments falling onto
the CAM, as they can stimulate a local inflammatory
response. The experimental manipulations were started on Day 10 and continued through Day 14. At this
time the experiments were terminated by fixing the
CAM in situ with 10% buffered formalin. A thin piece
of glass was slid under the fixed CAM and the membrane removed and stored in formalin until the vascular
density was estimated, usually within a week.
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Pellet Preparation
Ten percent Elvax-40 polymer pellets (ethylene-vinyl acetate copolymer, 40% vinyl acetate content)
for the sustained release of the experimental materials
were prepared using a simple modification of the method described by Langer.24 The polymer was dissolved
in methylene chloride (1:9; wt: vol). When completely
dissolved, the test material was added to achieve the
desired dose and the mixture thoroughly vortexed.
Ten-microliter aliquots were quickly pipetted onto a
sterile 50 X 75 mm glass slide to form individual
pellets. The pellets were polymerized at — 15°C for
24—48 hours and then placed in a vacuum dessicator at
3°C for an additional 24-36 hours to remove residual
methylene chloride. The pellets were stored in a small
air-tight plastic box at — 15°C until used. All preparative materials and instruments were sterile.
Each substance was applied to the CAM via such
Elvax pellets except the methyl-isobutyl-xanthine
(MIX). MIX, at doses of 55 and 110 fig in 50 and 100
/nl of Normosol-R, respectively, was injected daily
into the albumin during the 4-day experimental period.
Equivalent volumes of Normosol-R were injected into
control eggs. Where studied, individual pellet weights
were obtained using a Cahn electrobalance.
Quantitation of Vascular Density
A vascular density index (VDI) was obtained to
estimate CAM vascularity. An adaptation of a method
described by Harris-Hooker and coworkers25 was used.
A thin plastic coverslip inscribed with 4-, 5-, 6-, and 8mm-diameter concentric circles was centered over the
Elvax pellet on the formalin-fixed CAM. Vascular
density was estimated by counting the number of vessels which intersected the circles. Because this quantative technique employs the continuous lines of concentric circles to sample vessel density, the VDI is not
limited by a predetermined maximum (e.g., the number of intercept points on a grid). This allows a more
thorough sampling of the experimental area, especially
with regard to short newly forming vessels.
If a pellet was not present, the intercepts with all
four circles (72.2 mm total circumference) were counted; with a pellet present, only the 4-, 5-, and 6-mm
circles (47.1 mm total circumference) were counted.
Each response was counted 2-3 times at 20 X magnification under a dissecting microscope, and the average
of the counts recorded. This technique permits counting the intercepts of vessels 10-12 fi in diameter; it
Circulation Research
Vol 59, No 2, August 1986
does not separate arterioles from venules. An experimental and a mannitol-containing control pellet were
placed on each CAM at least 1 cm apart. The responses
on each CAM were evaluated individually as a percent
of control. The site of pellet placement relative to
underlying vessels was not critical. However, pellets
were not placed directly over or immediately next to
the largest vessels in order to avoid a counting bias
resulting from large diameter vessels in the study area.
The positioning of the pellet relative to the large and
small end of the egg was randomized. All quantitation
was done without knowledge of the experimental
group. All data are presented as the mean ± SEM.
Acute in Vivo Responses
The eggs were placed in a heating mantle and the
surface temperature maintained at 38°C. The shell was
then removed down to the level of the CAM. Adenosine (27, 270, or 2700 ng in 10 /xl Normosol-R) or 10
/xl of the vehicle alone was applied locally to the
CAM. After 1 minute, the in vivo VDI was estimated
as described above, except that counting was done at
45 x . The doses of adenosine and the Normosol-R
control were tested in random order and without
knowledge of the concentration. Dipyridamole (1 and
5 pig/10 /xl) was similarly studied.
Pellet Release of Adenosine
Elvax-40 pellets, containing 3 mg of adenosine or 3
mg of adenosine plus 1 mg of dipyridamole, were
prepared as described above, except that tritium-labelled adenosine (2,8-3H adenosine) was also included. Each pellet contained approximately 0.012 /xCi.
The pellets were pre-incubated for 6 hours prior to
study. Individual pellets were incubated in 400 /xl of
Normosol-R at 38°C. Every 24 hours the incubation
medium was replaced and the amount of radioactivity
released determined by liquid scintillation counting.
After the last change, the pellet was dissolved in methylene chloride (200 /xl x 2) and the mixture counted to
determine the total radioactivity added to each pellet.
The release of the tritiated adenosine is presented as
the cumulative percent release.
Materials
The adenosine, inosine, methyl-isobutyl-xanthine,
and dipyridamole were obtained from Sigma. The
methylene chloride (99 mol% pure) was purchased
from Fisher, and the tritiated adenosine from ICN.
Results
During embryonic development of the chick, there
is a progressive increase (p< 0.001) in the vascular
density of the CAM, the site of respiratory exchange
for the embryo. This growth parallels the increase in
embryo weight (Figure 1). The experiments presented
in this study were conducted between Day 10 and 14,
that period when vascularization of the CAM is most
rapid. Despite this rapid growth, the VDI and the
embryo weights exhibited little variation among individuals.
Dusseau et al
Adenosine-Stimulated Anglogenesis
165
25
20
a
z
>
CD
3)
O
15
m
-t
H
Sj
Z
UJ
Z
m
£
2
10
o
| |
u
5
x
H
V)
J 0
8
9
12
15
18
DAYS OF INCUBATION
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
FIGURE I. Age-related changes of the CAM vascular density
index (VDI) and embryo weight during the incubation of fertile
chick eggs. The VDI was determined using the 4-. 5-, 6-, and 8mm concentric circles (72.2 mm total circumference). Data are
mean ± SEM. VDI, n = 14-18; embryo weights, n = 19-30;
p<0.001 for each parameter, 1-WAY ANOVA.
Adenosine released from Elvax polymer pellets
stimulated a local dose-related increase (/?<0.01) in
the vascular density index (VDI) of the CAM (Figure
2). At the 0.2-, 1.0-, and 4.0-mg doses, 7 of 13, 14 of
15, and 14 of 14 CAMs respectively, exhibited increases in vascularity. Inosine, a primary metabolite of
adenosine formed by adenosine deaminase, failed to
elicit a statistically significant change in the VDI at
these same three doses (Figure 2).
The local increase in the VDI stimulated by the 3mg dose of adenosine was a highly consistent response
(Figure 3). Among 44 separate experiments, 43 CAMs
responded with an increase in the VDI. The mean
increase was 15.1 ± 1.5% (p<0.001). These data are
a composite of five separately conducted experiments.
The dose of adenosine incorporated into individual
pellets was confirmed by preparing a sample of pellets
(two preparative batches for each dose), weighing the
individual pellets, and subtracting the mean weight of
similarly prepared pure polymer pellets. The mean
weight of 33 Elvax only pellets was 971.7 ± 85.8 /ig.
The net weight of the 0.2 (n = 25), 1.0 (n = 22), and
3.0 ( n = 1 9 ) mg pellets was 214.6 ± 2 2 . 5 ,
1287.6 ± 45.8, and 3104.6 ± 79.7 /ng, respectively.
Figure 4 illustrates a representative response to a 3mg adenosine pellet and a mannitol control pellet. The
two pellets are 1.2 cm apart. In addition to a greater
VDI, the vessels surrounding the adenosine pellets
generally exhibited a more radial orientation toward
the pellet. This morphological feature was not observed among controls.
Experiments were conducted to determine whether
the increase in the VDI was due to adenosine opening
closed, but preexisting, vessels. The effect of topically
applied adenosine and of dipyridamole on the in vivo
VDI was studied. Adenosine, 27-2700 ng applied in
10 (i\ of Normosol-R, did not significantly affect a
local change of the in vivo VDI among eggs studied at
either 10 or 14 days of incubation (Table 1). Similarly,
dipyridamole (1 and 5 /Ag; n = 12) also failed to alter
the in vivo VDI of 14 Day eggs (p< 0.650) (data not
shown).
MIX, at doses to block the adenosine receptors, was
injected daily into the albumin on Days 10-14. Regression analysis revealed a significant dose-related
inhibition {p < 0.001) by MIX on the adenosine-stimulated VDI (Figure 5). MIX at 110 /Ag/day completely
blocked the angiogenic response to the 3-mg dose of
adenosine. In the absence of MIX, adenosine stimulated a 15.3% increase in the VDI. By itself, MIX (110
/ng/day) did not significantly alter the VDI of unstimulated CAM vessels (Table 2).
118
115
FIGURE 2.
Changes in the VDI of the CAM in
response to various amounts of adenosine or
inosine released from 10% Elvax pellets.
Matching control pellets placed on the same
CAM contained mannitol. The VDI was determined using the 4-, 5-, and 6-mm concentric
circles (47.1 mm total circumference) and presented as percent of control changes. The
mean control VDI for the adenosine dose—response curve was 217.5 (n = 42); for the inosine dose-response curve, 192.7 (n = 33).
Data are mean ± SEM. Number beside each
value is group n. Adenosine dose-response
curve, p<0.0l; inosine. p<0.405 by I-WAY
ANOVA.
109
106
i
CC 103
100
97
0.2
1.0
DOSE (MG)
3.0
Circulation Research
166
250
z
u
200
D
O
(0
*
150
CON
ADN
(3 MO)
FIGURE 3. Increase in the VDI of the CAM in response to the
3 mg dose ofadenosine. The VDI was determined using the 4-,
5-, and 6-mm concentric circles and presented as the actual
number of line-vessel intersections counted. Control pellets
contained I mg of mannitol. Data are mean ± SEM. Number
inside bars is group n. * * * p < 0.001 vs. control by paired t test.
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
The vasodilator, dipyridamole, enhanced adenosine's stimulation of CAM vascularity. One milligram
of dipyridamole added to the 3-mg adenosine pellet
increased the neovascular response an additional 52%
(p< 0.005) (Figure 6). A mannitol control pellet was
also placed on each of the CAMs. Dipyridamole alone,
tested on separate CAMs, had a minimal effect on the
VDI.
The effects of the Elvax polymer itself and of man-
Vol 59, No 2, August 1986
nitol on the CAM vascularure are presented in Table 3.
The presence of the Elvax pellet on the CAM had no
effect on the VDI. Likewise, no change in vascularity
was observed when mannitol was included in the pellet
(Table 3, Figure 4). Neither protocol caused a change
in the orientation of the vessels surrounding the pellet.
The release of tritiated adenosine from Elvax pellets
containing adenosine or a combination of adenosine
and dipyridamole is shown in Figure 7. Over 4 days
approximately 85% of the labelled nucleoside was released from both pellet preparations into the incubation
medium. An initial rapid release occurred during the
first 24 hours, after which the rate of release was more
constant. The rates of labelled adenosine release were
not statistically different (/?<0.382) between the two
curves.
Discussion
This study offers direct evidence that adenosine can
increase the vascularity of the CAM. The quantitative
technique employed in these experiments provides an
estimate of the actual vessel density, the endpoint of
the angiogenic process. Adenosine's augmentation of
vascularization on the CAM appears to be an increase
in the number of new vessels. Acute suffusion of adenosine onto the CAM failed to increase the in vivo VDI
(Table 1) on either 10-day or 14-day CAM's. The
vasodilator, dipyridamole, also failed to affect the in
\
\
FIGURE 4. Formalin-fixed CAM photographed after removal from egg illustrating representative responses to a 3-mg adenosine
pellet (ADN) and a mannitol control pellet (M). The pellets are 1.2 cm apart.
Dusseau et al
Adenosine-Stimulated Angiogenesis
TABLE 1. Acute Effects of Local Adenosine Suffusion on the in
Vivo Vascular Density Index of the CAM at 10 and 14 Days
167
TABLE 2. Effect of MIX on Vascular Density Index of 14-Day
Chick CAM
Adenosine dose (ng)*
0
Group
270
27
2700
10-Day (15) 136 .3±6.6t 136.4±8.1 139.8±6.5 136. 9±8.1
14-Day (15) 177 .0+8.0 175.4±8.4 171.3±8.3 179. 5±8.5
•Adenosine applied in 10 /xl of Normosol-R, 38° C.
tVascular density index = number of intercepts between vessels
and concentric circles over 47.1 mm of circumference.
Data are mean ± SEM; sample size in parentheses.
1-WAY ANOVA (repeated measures): 10-day, p < 0.445; 14day, p < 0.107.
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
vivo VDI. It is likely, therefore, that the observed
increase in the vascularity of the CAM with adenosine
was due to the growth of new vessels and not to a direct
vasodilator effect of this nucleoside opening preexisting vessels.
Among the experiments summarized in Figures 2
and 3, it is clear that the VDI responses to adenosine
are highly consistent, albeit modest. The low variability of this bioassay model (Figure 1) and the absence of
an effect on the VDI by the Elvax pellets (Figure 4,
Table 3) are obvious advantages for discerning these
small changes. Because these experiments were conducted during the time when the normal embryonic
116
13
3 M G ADENOSINE
114
Vascular density index*
Control (6)
29O.0±6.8
-3.4
110 /ig MIX (5)
28O.l±9.6t
•Number of intercepts between vessels and concentric circles
over 72.2 mm of circumference,
tp = not significant (unpaired t test) vs. control.
Data are mean ± SEM; sample size in parentheses.
proliferation of the CAM vessels is most rapid (Figure
1), it is not certain whether the effect of the exogenous
adenosine was only to accelerate an ongoing mechanism or to superimpose an additional effect of its own.
However, the observed VDI increase in the presence of
the 3-mg adenosine pellet was not maximal as dipyridamole enhanced adenosine's effectiveness even further (Figure 6).
Recent experiments directly testing the angiogenic
activity of adenosine have not, however, documented
a clear involvement. Adenosine implanted in the rabbit
cornea stimulated an elongation of the surrounding
limbus vessels toward the adenosine implant. Inosine,
which had no effect in our study (Figure 2), had a more
pronounced effect than adenosine.26
On the CAM, millipore filters saturated with 10
- 4 M adenosine did not elicit an angiogenic response;
nor did a variety of other vasodilators.27 These investigators, using quantitative techniques similar to ours,
only counted vessels that were at least 1 mm in length.
A minimum length was not a criterion in our experiment, although with the 1-mm separation of the lines a
-3 130
112
110
108
-
120
106
10
104
V)
110
102
r :.7247
100
o
55
MIX (UG / DAY)
110
FIGURE 5. Inhibition of the adenosine-stimulated increase in
the VDI by methyl-isobutyl-xanthine (MIX). The MIX (55 or 110
fig/day) or Normosol-R was injected into the albumin. Each
group had an adenosine (3 mg) and a mannitol (I mg) pellet on
each CAM. The VDI was determined using the 4-, 5-. and6-mm
concentric circles. The mean control VDIs for the 0, 55, and
110 fig MIX groups were200.6, 198.7, and 194.1, respectively. Data are mean ± SEM. Number beside each mean value is
group n; p<0.00l by regression analysis.
100
(12)
ADN
(12)
ADN
+
DIP
FIGURE 6. Dipyridamole (DIP) enhancement of adenosine's
(ADN) stimulation of the VDI on the CAM. DIP (1 mg) was
added directly to the Elvax pellet containing 3 mg of adenosine.
DIP alone was tested in a separate group. The VDI was determined using the 4-, 5-, and 6-mm concentric circles. The mean
control VDI was 198.2. Data are mean ± SEM; numbers inside
bars are group n. ***p<0.001 vs. control; **p<0.005 vs
ADN by paired t tests.
Circulation Research
168
TABLE 3. Effect of 10% Elvax Polymer Pellets and of Mannitol
on Vascular Density Index of CAM
Group
No pellet (34)
Elvax only (34)
No pellet (24)
Mannitol - 1 mg (24)
Vascular density
index*
326.2±5.7t
327.3±5.8t
305.0±3.5
%A
0.4±0.54
0.67±0.55
306.8 ± 3.1t
•Number of intercepts between vessels and concentric circles
over 72.2 mm of circumference,
tp = not significant (paired t test).
Data are mean ± SEM; sample size in parentheses.
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
certain percent of the vessels less than 1 mm would not
be counted. Moreover, their results may have been
influenced by the marked reorientation and presence of
vascular loops which formed in response to the millipore filter itself, a response oberved in our own laboratory (unpublished) and by others.28
In other studies, a role for adenosine in angiogenesis
has been implicated using pharmacological interventions. Dipyridamole, which potentiates the biological
activity of adenosine by inhibiting adenosine deaminase and blocking the adenosine re-uptake mechanism,
increased capillary density in the heart20 2I and skeletal
muscle19 of adult rats. The angiogenic response was
evaluated by tritiated thymidine uptake and electron
microscope morphometry of the capillary endothelial
cells.
The absence of vascular proliferation on the CAM
with dipyridamole (Figure 6) as opposed to the response in rats,19"21 may be due to the relatively short
duration of the experiment compared to the 3- to 4week studies with rats; or, to the difference in angiogenic criteria employed. It is also quite possible that
adenosine does not play an active role in regulating
normal embryonic neovascularization, at least on the
CAM. Manipulation of the putative endogenous adenosine response mechanism by either MIX or dipyridamole alone failed to alter the VDI (Table 2, Figure 6).
The effectiveness of these two compounds in affecting
vessel growth was only evident in the presence of
exogenously supplied adenosine. Thus, it appears that
adenosine is not a primary mediator of normal vascular
growth of the CAM.
The data do, however, point to the embryonic presence of functional adenosine receptors. Among the
eggs treated with MIX, the adenosine-induced increase in the VDI was suppressed in a dose-dependent
manner (Figure 5). At low doses, MIX is an effective
adenosine receptor blocker.29 Although this study does
not specifically rule out the well-known1 inhibition of
phosphodiesterase by MIX and other xanthines, the Ki
for this inhibition is in the millimolar range. In contrast, adenosine receptor blockade by MIX is achieved
at significantly lower doses. The K\ for this inhibition
is in the micromolar range, the dose level employed in
this study.
Under other environmental circumstances, such as
long-term hypoxia, adenosine may exert a more active
Vol 59, No 2, August 1986
influence. Periods of reduced oxygen availability are
often accompanied by vascular proliferation,5"7 as well
as the release of adenosine.30-3I Recently, we reported
that eggs incubated in a 15% oxygen environment exhibited a 40% increase in the vascular density of the
CAM.32 That this response was partially inhibited by
MIX at doses comparable to those which blocked
adenosine's stimulation of the VDI in this study (Figure 5), supports a vasoproliferative role for adenosine.
The amounts of adenosine incorporated into the pellets obviously do not approximate physiological levels. However, the incubation of pellets containing tritium-labelled adenosine demonstrated a sustained
release of the adenosine from the polymer. That approximately 85% of the labelled adenosine was released from the Elvax pellets in vitro (Figure 7) should
not be interpreted that 85% of the adenosine incorporated into the experimental pellets was released in vivo.
The release kinetics in the incubation medium and on
the surface of the CAM undoubtedly are very different.
But varying the adenosine content of the experimental
pellets does provide a range of concentration gradients
to be tested. The efficacy of this protocol is demonstrated by the dose-related response of the CAM vascular density index (Figure 2).
Murray and co-workers33 have shown recently that
100
80
60
40
• ADN
(10)
o ADN'DIP
20
24
48
72
(10)
96
HOURS
FIGURE 7. Release of tritium-labelled adenosine from 10%
Elvax pellets incubated in 400 \il ofNormosol-R at 38°C. The
incubation medium was changed every 24 hours. Data are
mean ± SEM. Curves not statistically different (p<0.382; 2WAY ANOVA).
Dusseau et al
169
Adenosine-Stimulated Angiogenesis
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
the rate at which substances are released from Elvax
can depend on the total amount of material incorporated. This suggests that the dipyridamole-augmented response to adenosine (Figure 6) may be due to an increased release of adenosine from the pellets which
contained both compounds. Indeed, the release of tritium-labelled adenosine from these pellets was higher,
but only 7-10% (Figure 7). In view of the slope of the
adenosine VDI dose-response curve in Figure 2, it is
apparent that the magnitude of this increased adenosine release is not sufficient to account for the augmented VDI observed in the presence of dipyridamole.
Qualitatively, a clearly defined response to adenosine is presented in Figure 4. Compared to the control,
the vessels surrounding the adenosine pellet exhibit a
more radial arrangement with a few larger vessels
looping inward toward the adenosine source. This pattern of loops is not as pronounced as that characteristically described for tumor angiogenesis factor,34 a potent angiogenic stimulus on the chick CAM.
A definitive mechanism for adenosine's stimulation
of the increased VDI was not defined in these experiments. During angiogenesis migration and proliferation of endothelial cells are fundamental processes
which may be regulated separately. In the rabbit cornea angiogenesis assay, rabbit wound fluid stimulated
marked neovascularization." When exposed to an in
vitro gradient of this angiogenic factor, capillary endothelial cells exhibited a chemotactic response as well as
a morphological orientation to the gradient. The
wound fluid did not exert a mitogenic effect.353* Separate regulatory mechanisms for endothelial cell proliferation and migration have also been suggested for
inflammation-induced angiogenesis in the rat cornea.37
In this model, endothelial cell mitosis was suppressed
by x-irradiation. However, angiogenic activity, as evidenced by vascular sprouting and cell migration was
not abated until after the fourth day. Analogously,
adenosine may act on a specific regulatory component.
Among cultured aortic endothelial cells, adenosine induced a chemotactic response,38 but this nucleoside
was ineffective in eliciting a mitogenic or proliferative
effect.39 Thus, the elevated VDI with adenosine reported in the present study may be mediated, in part,
through a selective enhancement of endothelial cell
migration.
With our sampling technique, the increased VDI
could be partially explained by the loops and reoriented vessels crossing additional lines of the counting
circles. Such reorientation does not necessarily infer
the concomitant growth of new vesels, but these loops
often form anastomoses between arteries and veins or
other sprouts, subsequently giving rise to additional
sprouts." This loop formation or radial-like arrangement of vessels is an integral component of the CAM
angiogenic mechanism, and therefore its influence on
the counted VDI has been retained in our analysis.
Acknowledgments
The eggs were kindly supplied by Hubbard Farms, Statesville,
North Carolina. The authors also thank Dr. Patricia D'Amore for
generously providing the sterile, washed Elvax-40 polymer.
References
1. Duling BR, Berne RM: Longitudinal gradients in periarteriolar
oxygen tension: A possible mechanism for the participation of
oxygen in local regulation of blood flow. Circ Res 1970;
27:669-678
2. Hutchins PM, Bond RF, Green HD: Participation of oxygen in
the local control of skeletal muscle microvasculature. Circ Res
1974;34:85-93
3. Prewitt RL, Johnson PC: The effect of oxygen on arteriolar red
cell velocity and capillary density in the rat cremaster muscle.
MicrovascRes 1976; 12:59-70
4. Olsson RA: Local factors regulating cardiac and skeletal muscle blood flow. Ann Rev Physiol 1981;43:385-395
5. Hudlicka O: Growth of capillaries in skeletal and cardiac muscle. Circ Res 1982;50:451^61
6. Banchero N: Capillary density of skeletal muscle in dogs exposed to simulated altitude. Proc Soc Exp Biol Med 1975:
148:435-439
7. Knighton DR, Silver IA, Hunt TK: Regulation of woundhealing angiogenesis: Effect of oxygen gradients and inspired
oxygen concentration. Surgery 1981;90:262-269
8. Temple GF, MetcaJf J: The effects of increased incubator oxygen tension on capillary development in the chick chorioallantois. Resp Physiol 1970;9:216-233
9. Roberts RJ, Weesner KM, Bucher JR: Oxygen-induced alterations in lung vascular development in the newborn rat Pedialr Res 1983; 17:368-375
10. Myrhage R, Hudlicka O: Capillary growth in chronically stimulated adult skeletal muscle as studied by intravital microscopy
and histological methods in rabbits and rats. Microvasc Res
1983; 16:73-90
11. Folkman J: Tumor Angiogenesis. Adv Can Res 1985;43:
175-203
12. Lombard JH, Duling BR: Relative importance of tissue oxygenation and vascular smooth muscle hypoxia in determining
arteriolar responses to occlusion in the hamster cheek pouch.
Circ Res 1977;41:546-551
13. HonigCR, FriersonJL: Role of adenosine in exercise vasodilation in dog gracilis muscle. Am J Physiol 1980;238:H703H715
14. BenEzra D: Neovasculogenic ability of prostaglandins, growth
factors, and synthetic chemoattractants. Am J Ophthalmol
1978;86:455-461
15. Ziche M, Jones J, Gullino PM: Role of Prostaglandin El and
copper in angiogenesis. J Natl Cancer Inst 1982;69:475-482
16. Form DM, AuerbachR: PGE2 and angiogenesis. Proc Soc Exp
Biol Med 1983:172:214-218
17. Frazer RA, Ellis EM, Stalker AL: Experimental angiogenesis
in the chorio-allantoic membrane. Biblithca Anal 1979; 18:
25-27
18. Imre G: Studies on the mechanism of retinal neovascularization. BrJ Ophthalmol 1964;48:75-82
19. Tornling G, Adolfsson J, Unge G, Ljungqvist A: Capillary
neoformation in skeletal muscle of dipyridamole-treated rats.
Anneim-Forsch/Drug Res 1980;30:791-792
20. Tomling G, Unge G, Skoog L, Ljungqvist A, Carlsson S,
Adolfsson J: Proliferative activity of myocardial capillary wall
cells in dipyridamole-treated rats. Cardiovasc Res 1978;12:
692-695
21. Mattfeldt T, Mall G: Dipyridamole-induced capillary endothelial cell proliferation in the rat heart — a morphometric investigation. Cardiovasc Res 1983:17:229-237
22. Ackerman NR: The physiological effects of hypoxia on the
erythiocytes of the chick embryo. Dev Biol 1970;23:310-323
23. Baumann R, Padeken S, Haller E-A, Brilmayer T: Effects of
hypoxia on oxygen affinity, hemoglobin pattern, and blood
volume of early chicken embryos. Am J Phvsiol 1983;244:
R733-R741
24. Langer R: Polymers for the sustained release of macromolecules: Their use in a single-step method of immunization, in
Langone JJ, Van Vanakis H (eds): Methods in Enzymology, vol
73. New York, Academic Press, pp 57-75
25. Harris-Hooker SA, Gajdusek CM, Wight TN, Schwartz SM:
170
26.
27.
28.
29.
30.
31.
32.
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
33.
Circulation Research
Neovascular responses induced by cultured aortic endothelial
cells. J Cell Physiol 1983; 114:302-310
Fenselau A: An angiogenic role for adenine nucleotide catabolites (abstract). Fed Proc 1984;43(3):587
Bamhill RL, Ryan TJ: Biochemical modulation of angiogenesis in the chorioallantoic membrane of the chick embryo. J
Invest Dermatol 1983;81:485-488
Jakob W, Jentzsch KD, Mauersberger B, Heder G: The chick
embryo chorioallantoic membrane as a bioassay for angiogenesis factors: Reactions induced by carrier materials. Exp Path
Bd 1978;15:241-249
Wolff J, Londos C, Cooper DMF: Adenosine receptors and the
regulation of adenylate cyclase, in Dumont JE, Greengard P,
Robison GA (eds): Advances in Cyclic Nucleotide Research,
vol 14. New York, Raven Press, pp 199-214
Schrader J, Haddy FJ, Gerlach E: Release of adenosine, inosine, and hypoxanthine from the isolated guinea pig heart during hypoxia, flow-autoregulation and reactive hyperemia.
PflugersArch 1977,369:1-6
Winn HR, Rubio R, Berne RM: Brain adenosine concentration
during hypoxia in rats. Am J Physiol 1981;241:H235-H242
Dusseau JW, Hutchins PM: Low oxygen stimulation of angiogenesis on the chick chorioallantoic membrane (abstract). Fed
Proc 1986;45(3):409
Murray JB, Brown J, Langer R, Klagsburn M: A micro sus-
34.
35.
36.
37.
38.
39.
Vol 59, No 2, August
1986
tained release system for epidermal growth factor. In Vitro
1983;19:743-748
Folkman J, Cotran R: Relation of vascular proliferation to
rumor growth, hit Rev Exp Pathol 1976; 16:207-224
Banda MJ, Knighton DR, Hunt TK, Werb Z: Isolation of a
nonmitogenic angiogenesis factor from wound fluid. Proc Natl
Acad Sci 1982;79:7773-7777
Banda MJ, Dwyer KS, Beckmann A: Wound fluid angiogenesis factor stimulates the directed migration of capillary endothelial cells. J Cell Biochem 1985;29:183-193
Sholley MM, Ferguson GP, Seibel HR, Montour JL, Wilson
JD: Mechanisms of neovascularization: Vascular sprouting can
occur without proliferation of endothelial cells. Lab Invest
1984;51(6):624-634
Teuscher E, Weidlich V: Adenosine nucleotides, adenosine
and adenine as angiogenesis factors. Biomed Biochim Ada
44:1985 ;493-495
Fenselau A, Wallis K, Kaiser D: Studies on a new class of
potential antiangiogenic agents: nucleoside antitnetabolite
drugs. Microvas Res 1981 ;22:58-63
KEY
WORDS • adenosine • angiogenesis
chick chorioallantoic membrane • dipyridamole
vascular density
•
•
Stimulation of angiogenesis by adenosine on the chick chorioallantoic membrane.
J W Dusseau, P M Hutchins and D S Malbasa
Downloaded from http://circres.ahajournals.org/ by guest on June 18, 2017
Circ Res. 1986;59:163-170
doi: 10.1161/01.RES.59.2.163
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1986 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/59/2/163
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/