Am J Physiol Regulatory Integrative Comp Physiol
281: R291–R301, 2001.
Factors that increase the contractile tone of the
ductus arteriosus also regulate its anatomic remodeling
HIROKI KAJINO,1 YAO-QI CHEN,1 STEVEN R. SEIDNER,2 NAHID WALEH,1,3
FRANÇOISE MAURAY,1 CHRISTINE ROMAN,1 SYLVAIN CHEMTOB,4
CAMERON J. KOCH,5 AND RONALD I. CLYMAN1,6
1
Cardiovascular Research Institute and 6Department of Pediatrics, University of California,
San Francisco 94143-0544; 3SRI International, Menlo Park, California 94025; 2Department of
Pediatrics, University of Texas Health Science Center, San Antonio, Texas 78284; 4Research
Center, Hôpital Saint-Justine, Montreal, Quebec H3T 1C5, Canada; and 5Department of
Radiation Oncology, University of Pennsylvania, Philadelphia, Pennsylvania 19104
Kajino, Hiroki, Yao-Qi Chen, Steven R. Seidner,
Nahid Waleh, Françoise Mauray, Christine Roman,
Sylvain Chemtob, Cameron J. Koch, and Ronald I.
Clyman. Factors that increase the contractile tone of the
ductus arteriosus also regulate its anatomic remodeling. Am
J Physiol Regulatory Integrative Comp Physiol 281:
R291–R301, 2001.—Permanent closure of the full-term newborn ductus arteriosus (DA) occurs only if profound hypoxia
develops within the vessel wall during luminal obliteration.
We used fetal and newborn baboons and lambs to determine
why the immature DA fails to remodel after birth. When
preterm newborns were kept in a normoxic range (PaO2:
50–90 mmHg), 86% still had a small patent DA on the sixth
day after birth; in addition, the preterm DA wall was only
mildly hypoxic and had only minimal remodeling. The postnatal increase in PaO2 normally induces isometric contractile
responses in rings of DA; however, the excessive inhibitory
effects of endogenous prostaglandins and nitric oxide, coupled with a weaker intrinsic DA tone, make the preterm DA
appear to have a smaller increment in tension in response to
oxygen than the DA near term. We found that oxygen concentrations, beyond the normoxic range, produce an additional increase in tension in the preterm DA that is similar to
the contractile response normally seen at term. We predicted
that preterm newborns, kept at a higher PaO2, would have
increased DA tone and would be more likely to obliterate
their lumen. We found that preterm newborns, maintained
at a PaO2 ⬎200 mmHg, had only a 14% incidence of patent
DA. Even though DA constriction was due to elevated PaO2,
obliteration of the lumen produced profound hypoxia of the
DA wall and the same features of remodeling that were
observed at term. DA wall hypoxia appears to be both necessary and sufficient to produce anatomic remodeling in preterm newborns.
hypoxia; nitric oxide; prostaglandins; oxygen; newborn
IN THE FULL-TERM INFANT, CLOSURE
of the ductus arteriosus (DA) occurs in two phases: 1) initial “functional”
closure of the DA lumen by surrounding smooth muscle
constriction and 2) “anatomic” occlusion of the lumen,
Address for reprint requests and other correspondence: R. I. Clyman, Box 0544, HSE 1492, Univ. of California, San Francisco, CA
94143-0544.
http://www.ajpregu.org
due to extensive neointimal thickening, and loss of
smooth muscle cells (SMC) from the inner muscle media (2). The initial “functional” constriction appears to
be required for the ultimate “anatomic” closure of the
DA. Loss of luminal blood flow produces a zone of
hypoxia in the DA’s muscle media. Hypoxia induces
vascular endothelial growth factor (VEGF) expression,
angiogenesis, neointima formation, and cell death, in
proportion to its severity (2). Failure to develop hypoxia of the muscle media, despite initial constriction,
leads to failure of DA remodeling (2).
The initial “functional” constriction of the DA depends on a balance between dilating and contracting
forces. In the late-gestation fetus, the DA has a high
level of intrinsic tone (24). After delivery, the increase
in arterial oxygen concentration leads to a further
increase in DA tone. The DA also produces several
vasodilator substances that oppose the intrinsic and
oxygen-induced contractile forces. Both vasodilator
prostaglandins (PG), especially PGE2, and nitric oxide
(NO) play a significant role in opposing DA closure and
maintaining its patency during fetal life (10, 13, 19, 24,
32). After delivery, there is an increase in arterial
oxygen concentration, a decrease in circulating PGE2
concentrations, and a decrease in intraluminal blood
pressure (due to the drop in pulmonary vascular resistance); all of these events promote DA constriction in
the full-term neonate (4, 6).
In contrast with the full-term infant, the DA of the
preterm infant frequently remains open for many days
after birth. Even when it does constrict, profound hypoxia in the DA wall and anatomic remodeling often fail
to develop (2). This leads to subsequent DA reopening
(34). Prior studies have attributed the preterm infant’s
weaker “functional” constriction to either 1) its blunted
response to oxygen (1, 30, 36, 37) or 2) its increased
sensitivity to PGE2 and NO (5, 7, 10). The contribution
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
0363-6119/01 $5.00 Copyright © 2001 the American Physiological Society
R291
Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on April 20, 2017
Received 1 September 2000; accepted in final form 28 February 2001
R292
CONTRACTILITY REGULATES DUCTUS REMODELING
of intrinsic tone (24) to DA contractility has not been
examined in the preterm newborn.
In the following study, we examined the relative
importance of the factors that regulate DA tone in the
preterm newborn. We hypothesized that if we could
make the preterm DA constrict more tightly, it might
completely obliterate its luminal blood flow and develop the degree of tissue hypoxia needed to initiate the
remodeling process.
METHODS
Fig. 1. Direct comparison of dot immunoblot assay, relative median
fluorescence intensity [flow cytometric analysis (FACS)] assay, and
radioactive [2-(2-nitro-1H-imidazol-1-yl)-N-(2,2,3,3,3-pentafluoropropyl) acetamide] (EF5) uptake (radiolabel) assay for detection of
EF5 binding in ductus arteriosus (DA) smooth muscle cells (SMC).
SMC were incubated with 10⫺4 M EF5 for 3 h at the indicated oxygen
concentrations. Data from each assay were normalized to the values
obtained at the lowest oxygen concentration, 0.01%.
Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on April 20, 2017
In vitro studies: fetal lambs. Fifty-two late-gestation, nearterm lambs (mixed Western breed: 134 ⫾ 3 days gestation,
term ⫽ 145) and 25 preterm lambs (105 ⫾ 2 days) were
delivered by cesarean section and anesthetized with ketamine HCl (30 mg/kg iv) before rapid exsanguination. These
procedures were approved by the Committee on Animal Research at the University of California, San Francisco.
The DA was divided into 1-mm-thick rings (3–4 rings/
animal), and isometric tension was measured in a Krebsbicarbonate solution (8, 10, 24). An oxygen electrode (YSI
Model 53 Biological Oxygen Monitor, Yellow Springs, OH),
placed in the 10-ml organ bath, measured the oxygen concentration. The bath solution was equilibrated with gas mixtures containing 5% CO2 and was changed every 30 min.
Rings were stretched to an initial length (preterm: 5.2 ⫾ 0.4
mm; late gestation: 6.6 ⫾ 0.4 mm) that produced a maximal
contractile response to increases in oxygen tension (8) and
were exposed to one of two experimental protocols. In all
experiments, we allowed the tension in the rings to reach a
new steady-state plateau (30 min-1 h) before another experimental change was made.
Protocol 1: effects of oxygen, PG, and NO on near-term and
preterm DA. Rings were equilibrated with a solution containing one of the following oxygen concentrations: 2, 6, 16, 30, or
95%. After the tension reached a plateau, indomethacin
(5.6 ⫻ 6⫺6 M) was added; this was followed by NG-nitro-Larginine methyl ester (L-NAME; 10⫺4 M). Maximal tension
was determined by the response to 100 mM K⫹-Krebs solution (equilibrated with 95% O2-5% CO2). After the maximal
tension had been determined, minimal tension was determined by the response to sodium nitroprusside (SNP; 10⫺4
M).
The difference in tensions between the measured steadystate tension, at a particular oxygen concentration, and the
minimal tension produced by SNP was considered the net
active tension. The difference in tensions between the maximal tension (produced by K⫹-Krebs) and the minimal tension
(with SNP) was treated as the maximal active tension capable of being developed by the rings.
In some experiments, bath solution was collected, and
extracted, to measure PGE2 and 6-keto-PGF1␣ (PGI2 metabolite) by radioimmunoassay (24). Some of the rings from
protocol 1 were used in protocol 2.
Protocol 2: effects of oxygen on near-term and preterm DA
pretreated with indomethacin and L-NAME. Rings of DA that
were used in protocol 2 were first preconditioned (1.5 h) with
solution containing indomethacin and L-NAME and 6% oxygen. Next, the rings were exposed sequentially to 95, 30, 16,
6, and 2% oxygen. The difference in tensions between the
steady-state tension at 2% oxygen and either 6, 16, 30, or
95% oxygen was considered the O2-induced tension due to
that particular oxygen concentration.
Tissues were blotted dry and weighed after the experiments (preterm: 35 ⫾ 8 mg; late gestation: 59 ⫾ 17 mg).
Tensions developed in the rings were expressed as force per
unit cross-sectional area (g/cm2) (8) or as percentage of maximal active tension.
Detection of hypoxia with [2-(2-nitro-1H-imidazol-1-yl)-N(2,2,3,3,3-pentafluoropropyl) acetamide]. To determine the
degree of hypoxia within the rings of DA, we used the [2-(2nitro-1H-imidazol-1-yl)-N-(2,2,3,3,3-pentafluoropropyl) acetamide] (EF5) detection system that we have described previously (24). EF5 binds to cysteine residues of intracellular
proteins after it has been acted on by hypoxia-dependent
nitroreductases (25). When it has been studied in different
cell lines, different tissues, and different species, both in vivo
and in vitro, EF5 has been found to have a similar oxygen
dependency for its rate of binding (25). We have shown
previously that EF5 binding can be assessed quantitatively
using either radioactive EF5-binding, immunofluorescent
flow cytometry or immunoblot techniques (24, 25). We used
vascular SMC, isolated from medial explants of fetal lamb
DA (24), to compare the three different EF5-detection techniques. Approximately 250,000 cells (between passages 4 and
9) were plated onto glass Petri dishes (diameter 50 mm) and
incubated overnight. The next morning, the dishes were
cooled to 4°C and filled with 1 ml fresh medium containing
EF5 (10⫺4 M) in its unlabeled or labeled form (14C-2 position:
43 ␮Ci/mg) (synthesized by Dr. M. Tracy, SRI International,
Palo Alto, CA). Dishes were placed in leak-proof aluminum
chambers that were connected to a manifold allowing the gas
phase of the chambers to be exchanged for the desired oxygen
concentration (24, 25). The chambers were immersed in a
37°C water bath for rapid warming and shaken gently. After
a 3-h incubation, the cells were rinsed three times and
processed for either radioactive counting (25), flow cytometry
(25), or immunoblot assay (24). A mouse monoclonal IgG
primary antibody, ELK 3–51, which is highly specific for EF5
tissue adducts (25), was used to detect bound EF5 in the flow
cytometry and immunoblot assays (24, 25). As can be seen in
Fig. 1, the three methods for detecting EF5 binding were
similar in both their oxygen dependency and their dynamic
range.
Rings of DA were suspended in the organ baths and incubated in Krebs solution containing EF5 (10⫺4 M). The bath
CONTRACTILITY REGULATES DUCTUS REMODELING
Some DA were divided in two before freezing. One-half was
fixed for 5 h in fresh 4% paraformaldehyde at 4°C before
paraffin embedding. Paraffin-embedded sections were rehydrated and incubated with rabbit anti-VEGF antibody (Santa
Cruz Biotechnology, Santa Cruz, CA). Scoring for VEGF was
as follows: 0, no staining; 1⫹, mild staining; and 2⫹, moderate staining. All sections were stained in the same assay.
Assays were reproduced on three separate occasions. Control
sections were treated similarly and had no staining (data not
shown).
Histological measurements were made at the level of minimal luminal area that was determined from serial sections
made through the tissue. Tissue dimensions were determined by averaging measurements made from eight predetermined regions of the section, using a template and National Institutes of Health Image software. The neointimal
zone was defined as the region between the luminal endothelial cells and the internal elastic lamina (identified by phase
contrast microscopy). The depth of vasa vasorum ingrowth,
into the muscle media, was based on the presence of vasa
vasorum within predetermined concentric regions that partitioned the DA wall between the intima and the adventitia
(10).
In some DA, regions of cell loss (absent nuclei) were observed in the intima/inner muscle media. The region of cell
loss was defined as minimal if its area was ⬎3,000 and
⬍15,000 ␮m2 and moderate if it was ⬎20,000 ␮m2.
We have shown previously that EF5 binding in vivo can be
assessed quantitatively using immunofluorescent techniques
(2, 25). Direct digital image acquisition was performed with a
Photometrics “Quantix” CCD camera, and the digitized images were analyzed with National Institutes of Health Image
software and Adobe Photoshop. The microscope was calibrated for constant light output as previously described (2).
For digital analysis, the exposure time was set by the camera’s automatic exposure system. We assigned a “relative
intensity” to a microscopic field by considering the range of
pixel intensities and the total exposure. These procedures
allowed us to adjust the exposures to provide visually acceptable images while preserving the absolute intensity information. The absolute intensity values were corrected for the in
vivo EF5 drug exposure.
Statistics. Statistical analysis was performed by the appropriate Student’s t-test and by analysis of variance. Scheffé’s
test was used for post hoc analysis. Nonparametric data were
compared with a chi square analysis. Results are presented
as means ⫾ SD.
RESULTS
In vitro studies: fetal lamb DA. We used isolated
rings of lamb DA to determine how oxygen concentration affected the contractility of the preterm and lategestation DA. We used five specific bath-solution oxygen concentrations and measured the corresponding
average tissue oxygen concentration by the EF5 immunoblot technique (Table 1). As the bath-solution oxygen
concentration increased from 2 to 95%, the tissue oxygen concentration increased from ⬍0.2 to ⬎10% (Table
1). Preterm and late-gestation DA rings had the same
average tissue oxygen concentration at each of the five
bath-solution oxygen concentrations.
The maximal active tension produced by K⫹ in the
preterm DA (236 ⫾ 36 g/cm2) was similar to the maximal active tension in the late-gestation DA (271 ⫾ 84
g/cm2).
Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on April 20, 2017
solution was preequilibrated with the desired oxygen concentration. After a 3-h incubation, unbound EF5 was rinsed out
of the baths and the rings were frozen with liquid N2 (24).
Immunoblot analyses of EF5 binding in DA rings were compared with measurements made in SMC that were incubated
in known oxygen concentrations (24). Lysates from both the
tissue samples and the SMC standard curve were blotted
onto the same membrane and processed together. Because
the oxygen dependence of EF5 is linearly related to drug
exposure (16, 25), the relative binding of EF5 to a tissue
lysate was corrected for its drug exposure. The in vitro drug
exposure to EF5 was the same in both the tissue samples and
the SMC standard curve (3 ⫻ 10⫺4 M/h). The oxygen concentration in the tissue was determined by comparing the EF5
binding in the tissue with the EF5 binding in the SMC
standard curve lysates. Two types of control tissues were
used in the DA ring experiments: 1) tissue from rings that
were not given EF5, and 2) tissue from EF5-treated rings
that were incubated in 95% oxygen for 3 h. There was no
antibody (ELK 3–51) binding to either of the control tissues
(data not shown).
In vivo studies: preterm baboons. We used preterm baboons
Papio sp. (140 days gestation; full term: 185 days) to examine
the effects of arterial oxygen tension on ductus closure. Animal care, surgery, and necropsy were performed at the
Southwest Foundation for Biomedical Research (San Antonio, TX). Animals were killed at the end of the study with an
overdose of pentobarbital sodium. Preterm fetal baboons
were delivered by cesarean section and euthanized before
breathing. Preterm newborns were delivered by cesarean
section and cared for in the primate intensive care nursery
for the first 6 days after delivery (2). Ventilator management
was designed to maintain the arterial PO2 (PaO2) in either a
normoxic range (50–90 Torr) or a hyperoxic range (⬎200
Torr) [with a fractional inspired oxygen concentration
(FIO2) ⫽ 1.0]. At 6 days after delivery, immediately before
necropsy, the baboons still required mechanical ventilation
with FIO2 ⫽ 0.29 ⫾ 0.14 (normoxia) or 1.0 ⫾ 0.0 (hyperoxia).
There were no differences in PaCO2, pH, or blood pressure
between the groups during the first 6 days (data not shown).
Pulsed Doppler flow studies were performed before necropsy,
using an 8-mHz transducer interfaced with a Biosound
ND256 echocardiographic system to confirm the presence,
direction, and timing of ductus flow (2).
EF5 was given to the baboons (10⫺4 mol/kg iv over 10 min)
36 h before necropsy. Blood samples for EF5 determination
were collected at 1, 6, 12, 24, and 36 h after the dose and
analyzed as previously described (2). The in vivo exposure of
tissues to EF5 was calculated from the area under the curve
of the EF5 serum concentrations and was similar in the two
groups (data not shown). We compared the measurements of
EF5 binding in the two groups of 6-day-old, premature baboons with measurements made in 2-day-old, full-term baboons (n ⫽ 3). The full-term baboons were delivered by the
normal spontaneous vaginal route and were given EF5 36 h
before necropsy.
Immunohistochemistry. Protocols for the immunohistochemistry of endothelial cell nitric oxide synthase (eNOS),
von Willebrand factor (VWF), VEGF, and EF5 were similar
to those reported previously (2, 10). Briefly, DA was frozen in
liquid N2. Frozen sections were incubated with either mouse
monoclonal anti-eNOS (Clone 3, Transduction Lab, Lexington, KY) or rabbit anti-VWF (Dako, Carpenteria, CA) to
detect endothelial cells, or with mouse monoclonal antibody
ELK 3–51 conjugated with the fluorescent dye Cy3 to detect
EF5 binding.
R293
R294
CONTRACTILITY REGULATES DUCTUS REMODELING
Table 1. Comparison of oxygen concentration in the bath solution with mean tissue oxygen concentration in
the near-term and preterm ductus arteriosus rings
Oxygen, %
Bath Solution:
2
6
16
30
95
Tissue oxygen concentration, %
Near term
n
Preterm
n
0.14 ⫾ 0.03
11
0.16 ⫾ 0.02
3
0.26 ⫾ 0.09
8
0.26 ⫾ 0.17
8
0.39 ⫾ 0.12
6
0.52 ⫾ 0.37
9
1.43 ⫾ 2.35
10
1.63 ⫾ 1.74
10
⬎10
5
⬎10
5
Values are means ⫾ SD. n, Number of the ductus rings. Bath-solution oxygen concentration was measured by oxygen electrode. Tissue
oxygen concentration was measured by EF5 immunoblot technique.
Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on April 20, 2017
In protocol 1, as the bath oxygen concentration increased from 6 to 95%, there was a progressive increase
in isometric tension in both preterm and late-gestation
DA rings (Fig. 2A). However, the net active tension of
the preterm DA was significantly (P ⬍ 0.01) less than
that of late-gestation DA (Fig. 2A); in addition, the
preterm DA appeared to be less sensitive to increases
in bath oxygen concentration than the late-gestation
DA [EC50 (bath oxygen concentration that produces
50% of the change in tension observed between 6 and
95% oxygen): preterm, 50% oxygen; late gestation, 28%
oxygen].
When the bath oxygen concentration dropped to 2%
(which corresponds to a tissue oxygen concentration
⬍0.2%), the severely hypoxic rings developed a tension
that was greater than that achieved when the oxygen
concentration in the bath solution was 30% (Fig. 2A).
The net active tension achieved by the preterm rings,
which were severely hypoxic, was significantly less
than that achieved by late-gestation rings (P ⬍
0.0001).
We used indomethacin and L-NAME to uncover the
role of endogenous PG and NO in opposing DA contractility. Indomethacin had no contractile effect on DA
rings that were severely hypoxic (bath oxygen concentration ⱕ2%) (Fig. 2B); this is due to the marked
inhibition of endogenous PG production that occurs
when tissue oxygen concentration is ⬍0.2% (Fig. 3)
(24). In contrast, indomethacin caused an increase in
DA tension when the oxygen concentration was ⱖ6%
(Fig. 2B). Although PG production in the preterm DA
was less than or equal to the production in the lategestation DA (Fig. 3), the tension induced by indomethacin, in the preterm DA, represented a significantly
greater percent of the active tension (net ⫹ indomethacin-induced ⫹ L-NAME-induced tension) than it did
near term (P ⬍ 0.01).
L-NAME also had no effect on DA rings that were
severely hypoxic (Fig. 2C); however, when the bath
oxygen concentration was ⱖ6%, L-NAME increased DA
tension. The tension induced by L-NAME represented a
significantly greater percent of the active tension in the
preterm DA than it did near term (P ⬍ 0.01).
In protocol 2, ductus rings were pretreated with
indomethacin and L-NAME before oxygen concentrations in the rings were varied. DA rings had an intrinsic vasoconstrictor tone even at negligible oxygen con-
Fig. 2. Effects of oxygen concentration on DA active tension in
preterm and near-term lambs. A: DA rings were incubated in different bath-solution oxygen concentrations. Height of column represents net active tension (means ⫾ SD) developed by the rings at each
oxygen concentration. Number in parentheses ⫽ number of DA
rings, from separate animals, exposed to each oxygen concentration.
B: height of column represents the steady-state tension after indomethacin was added to the bath (⫾SD of indomethacin-induced
tension; the difference in tensions between the steady-state tension
achieved with indomethacin and the steady-state tension with oxygen alone was considered the indomethacin-induced tension). C:
height of column represents the steady-state tension after indomethacin and NG-nitro-L-arginine methyl ester (L-NAME) were added to
the bath (⫾SD of L-NAME-induced tension; the difference in tensions
between the steady-state tension achieved after L-NAME and the
steady-state tension after indomethacin was considered the
L-NAME-induced tension). *P ⬍ 0.01 compared with net tension,
indomethacin-induced tension, or L-NAME-induced tension at 2%
bath-solution oxygen concentration.
CONTRACTILITY REGULATES DUCTUS REMODELING
R295
centrations (bath concentration 2% oxygen) (Fig. 4).
This oxygen-independent tone was equal to 64 ⫾ 7% of
the maximal active tension in the late-gestation DA,
but only 34 ⫾ 10% in the preterm DA (P ⬍ 0.001). In
the absence of endogenous PG and NO production, the
preterm DA no longer had a blunted response to oxygen, and it was more sensitive to changes in bath
oxygen concentration (EC50: 14 ⫾ 2% oxygen) than was
the near-term DA (EC50: 30 ⫾ 7%) (P ⬍ 0.05) (Fig. 4).
In vivo studies: preterm baboons. Premature newborn baboons do not constrict their DA as tightly and
do not remodel their DA to the same degree as fullterm baboons (2). On the basis of our in vitro experiments, we would predict that increasing the arterial
PaO2 from the normoxic range (50–90 mmHg) to the
hyperoxic range (250–350 mmHg) would more than
double the net active tension produced by the premature DA (Fig. 2A). We hypothesized that if we kept the
arterial PaO2 in the hyperoxic range, we would produce
a greater degree of DA constriction in vivo, which
would increase the likelihood of anatomic remodeling.
Twenty-seven premature newborn baboons were
kept in the normoxic range for the first 6 days after
birth (Fig. 5A). Although 74% had no evidence of a
patent DA by Doppler examination on day 6 (Fig. 5B),
a small patent lumen (⬎0-mm diameter) was still perceptible at the time of necropsy in 86% of the DA (Fig.
5C). In contrast, 96% of the premature newborns that
were kept in the hyperoxic range had a closed DA by
Doppler examination (P ⬍ 0.05) (Fig. 5B), and only
14% had a patent lumen at necropsy (P ⬍ 0.01)
(Fig. 5C).
Fig. 4. Effects of oxygen concentration on DA contractility after PG and nitric oxide (NO) production has
been inhibited. Rings of DA were first equilibrated with
indomethacin and L-NAME; next they were exposed
sequentially to the following bath-solution oxygen concentrations: 95, 30, 16, 6, and 2%. Some of the rings
were not exposed to all of the oxygen concentrations.
Number in parentheses ⫽ number of DA rings, from
separate animals, exposed to each oxygen concentration. Height of column represents active tension
(means ⫾ SD) at a particular oxygen concentration. The
difference in tension between 2% bath oxygen concentration and either 6, 16, 30, or 95% oxygen concentration is considered the oxygen-induced tension. *P ⬍
0.01, †P ⬍ 0.05.
Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on April 20, 2017
Fig. 3. Severe hypoxia inhibits prostaglandin E2
(PGE2) and 6-keto-PGF1␣ released during a 30-min
collection period and expressed per 100 mg tissue wet
wt. Tissue wet weight (in mg): 60 ⫾ 22 near term; 36 ⫾
10 preterm. *P ⬍ 0.01, †P ⬍ 0.05 vs. amount released at
6% bath oxygen concentration. Note that 95% oxygen
concentration also decreased PGE2 release. Number of
DA rings in parentheses.
R296
CONTRACTILITY REGULATES DUCTUS REMODELING
We injected the hypoxia indicator EF5 into full-term
and premature baboons 36 h before necropsy. The DA
of the full-term baboons was closed by Doppler examination at the time of necropsy. We compared photometric analyses of EF5 binding in the DA with measurements made in the ascending aorta. We found
negligible EF5 binding in sections from the ascending
aorta (data not shown) (2). We anticipated this result
because the aorta is a well-oxygenated vessel, with
intramural PO2 values (20–30 mmHg) (2, 23, 35) that
Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on April 20, 2017
Fig. 5. In vivo measurements of arterial oxygen tension (PaO2) and
DA constriction in preterm baboons ventilated with either 100%
oxygen (hyperoxia) or oxygen concentrations needed to keep the
arterial PaO2 in the normoxic range (50–90 mmHg). A: arterial PO2:
means ⫾ SD. Hyperoxia, n ⫽ 27; normoxia, n ⫽ 27. B: luminal
patency of the DA detected by Doppler examination on day 6. Hyperoxia group had a higher rate of closure (P ⬍ 0.05). C: narrowest
measurement of luminal diameter at the time of necropsy on day 6.
Hyperoxia group had a narrower lumen (P ⬍ 0.01). In vivo Doppler
findings in necropsied animals: hyperoxia group (closed ⫽ 14); normoxia group (closed ⫽ 11; open ⫽ 3).
lie at the low end of the EF5 dynamic binding curve
(Fig. 1). In contrast, sections of DA, from the full-term
baboons, had intense EF5 binding when compared
with sections from the ascending aorta (median EF5
binding: 25-fold greater than aorta; range: 14- to 50fold greater).
DA from the preterm baboons, whose PaO2 had been
kept in the normoxic range, had significantly less EF5
binding than those at term. Even when the DA was
closed on Doppler examination, the intensity of EF5
binding was only 20% (median; range: 10–37%) of the
EF5-binding intensity at term. In contrast, when preterm baboons were ventilated with 100% oxygen, the
intensity of EF5 binding was similar to the intensity at
term (median: 60%; range: 30–110%) (Fig. 6).
We looked at VEGF expression during DA closure
because VEGF is induced by hypoxia (21) and plays an
important role in endothelial cell proliferation and
migration (39). DA from preterm newborn baboons,
ventilated with 100% oxygen, expressed moderate levels of VEGF in the region of the muscle media that had
intense EF5 binding. The DA from the baboons ventilated with 100% oxygen expressed significantly more
VEGF than the DA from normoxic baboons (P ⬍ 0.05)
(Figs. 7, A and B, and 8). VEGF expression was significantly related to the degree of luminal constriction
(VEGF vs. diameter of DA lumen, P ⬍ 0.01; Fig. 8).
In the full-term DA, proliferation of luminal endothelial cells and migration of medial SMC produce a
neointima that occludes the DA lumen (2, 3, 43). Only
21% of DA from the preterm baboons, with PaO2 in the
normoxic range, had evidence of endothelial cell accumulation in the DA lumen (3 or more layers of luminal
endothelial cells) (Fig. 9B); the rest was lined by a
single layer of luminal endothelial cells (Fig. 7C). In
contrast, all of the DA from the baboons ventilated
with 100% oxygen had neointimal mounds filled with
endothelial cells (Figs. 7D and 9B) (P ⬍ 0.001). The DA
from the baboons ventilated with 100% oxygen also
had a significantly thicker neointima than the DA from
the normoxic baboons (Figs. 7, A, B, C, and D, and 9A)
(P ⬍ 0.01).
In the fetal DA, vasa vasorum are restricted to the
adventitia and rarely penetrate the outer muscle media (Fig. 7G) (3). After birth, vasa vasorum invade the
muscle media, even in the preterm newborn (Fig. 7H)
(3). The baboons ventilated with 100% oxygen had a
much greater ingrowth of vasa vasorum than those
kept in the normoxic range (P ⬍ 0.001; Fig. 9D).
In the full-term DA, cells are lost from the center of
the EF5-stained hypoxic zone due to cell death (2).
Similarly, condensed pyknotic nuclei could be seen
throughout the hypoxic zone of DA obtained from preterm newborns ventilated with 100% oxygen (Fig. 7E).
This was associated with loss of cellular elements from
the intimal-inner media region (Fig. 7, D and F). DA
from premature newborns ventilated with 100% oxygen had more extensive regions of cell loss than those
from normoxic animals (P ⬍ 0.001; Fig. 9C).
CONTRACTILITY REGULATES DUCTUS REMODELING
R297
DISCUSSION
Hypoxia of the DA wall (or events that lead to the
development of hypoxia) seems to be the required stimulus for irreversible closure. Anatomic remodeling occurs only in the presence of moderate/intense hypoxia
(2). Oxygen reaches the muscle media of most blood
vessels through either the vessel’s lumen or its intramural vasa vasorum (derived from small adventitial
arteries) (45). The preterm baboon DA can meet its
oxygen needs from luminal flow alone, because it has
no vasa vasorum in its muscle media (Fig. 7G) (3).
Therefore, when the preterm DA wall fails to become
hypoxic after its initial postnatal constriction, the most
likely cause is a small persistent patent lumen (2, 34).
In the current study, we found that despite the
absence of Doppler-detectable luminal blood flow, a
small residual patent lumen was still present in most
of the preterm baboons, when their arterial PaO2 was
maintained in the normoxic range (Fig. 5, B and C).
Therefore, we studied the four major components that
are known to affect DA contractility (oxygen-induced
tone, intrinsic tone, and PG and NO inhibition of tone)
to see which might explain the poorer contractile response of the preterm newborn.
Although oxygen is an important regulator of DA
tone, the exact mechanisms that mediate the DA response to oxygen are still unknown. Oxygen causes
membrane depolarization in vascular smooth muscle,
which, in turn, is associated with a rise in smooth
muscle intracellular calcium (33). Oxygen inhibits K⫹
channels and releases endothelin-1 in DA smooth muscle (11, 33, 44); however, the importance of K⫹ channels and endothelin-1 in mediating oxygen-induced
contraction is still unclear (R. I. Clyman, unpublished
data) (17).
High oxygen concentrations also decrease PGE2 production by the DA (Fig. 3); their effects on PGI2 (6-ketoPGF1␣) production are less apparent. This suggests an
effect predominantly on PGE synthase rather than
cyclooxygenase. PGE2 synthesis is glutathione dependent (12), and hyperoxia depletes reduced glutathione,
especially in tissues of immature animals (41). These
events may explain the reduction in PGE2 that occurs
in 95% oxygen (Fig. 3).
In the current study, we observed that the oxygeninduced constriction in the preterm DA was less than
that observed near term (Fig. 2A). This is similar to
what has previously been reported (8, 30, 36, 37).
However, we found that in the presence of inhibitors of
PG and NO production, not only was the preterm DA
more sensitive to oxygen, it responded to oxygen with a
greater increment in contractile force than it did near
term (Fig. 4). Therefore, the excessive inhibitory effects
of endogenous PG and NO on DA tone make it appear
as if the preterm DA has a smaller increment in tension, in response to oxygen, than it does near term. The
increased inhibitory effects of PG and NO in the preterm DA are not due to increased production of the
vasodilators (Fig. 3). The increased inhibitory effects of
PG and NO are probably due to the increased sensitivity of the preterm DA to the two vasodilators (5, 7, 10).
In addition to the oxygen-induced tension, there is
an intrinsic vasoconstrictor tone in the DA that is
independent of ambient oxygen concentration (Figs. 2A
and 4). Whether this sustained intrinsic tone is due to
vasoconstrictors that are released by the endothelial or
SMC of the DA wall or to the increased calcium sensitivity of the DA contractile proteins is currently unknown (14, 20, 26, 27). In the late-gestation DA, the
intrinsic tone is equivalent to 64 ⫾ 7% of the maximal
Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on April 20, 2017
Fig. 6. EF5 binding to 6-day-old premature baboon DA. DA was removed from the newborns 36 h after EF5
injection (A and B; animal in C did not receive EF5). Fourteen-micrometer sections from the middle of the DA were
stained with Cy3-conjugated ELK3–51. Dashed white line, outer edge of DA muscle media (identified by phase
contrast microscopy). Horizontal bar represents 500 ␮m. Each figure is photographed at the same magnification.
There is intense binding of EF5 in the intimal-inner media region of the animal ventilated with 100% oxygen
(hyperoxia) (B); there is only mild EF5 binding in the middle of the muscle media of the normoxemic animal (A);
and there is no antibody, ELK3–51, binding to the control tissue obtained from a newborn ventilated with 100%
oxygen (C).
R298
CONTRACTILITY REGULATES DUCTUS REMODELING
tone that can be developed by the DA. On the other
hand, in the preterm DA, the intrinsic tone is only 34 ⫾
10% of the maximal tone (Fig. 4). The factors that
regulate the intrinsic tone and its increase during late
gestation are currently unknown.
Therefore, despite the increase in arterial PaO2 after
birth, the preterm DA has an ineffective postnatal
constriction due to both its weak intrinsic tone and its
exaggerated response to PG- and NO-mediated vasodilation. These findings offer an explanation for a previously contradictory observation: indomethacin appears
to cause a larger constriction of the fetal DA in vivo in
late-gestation fetuses than in preterm fetuses (31, 40);
this happens despite the fact that endogenous PG have
their greatest inhibitory effects on DA contraction
early in gestation (5, 7, 10). Our findings suggest that
early in gestation, the effects of indomethacin may not
be apparent in vivo, due to the poor intrinsic tone in the
immature DA and its exaggerated response to NO.
Indomethacin may appear to have a greater effect in
vivo, late in gestation, because the increase in intrinsic
tone and decreased responsiveness to NO may make
the absolute change in DA tension more noticeable.
On the basis of our in vitro studies (Fig. 2A), we
predicted (and subsequently found) that preterm newborns kept at a higher arterial PaO2 would have increased DA tone and would be more likely to obliterate
their lumen (Fig. 5C). We found that even though DA
Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on April 20, 2017
Fig. 7. DA from preterm baboons ventilated with 100% oxygen (hyperoxia) increased vascular endothelial growth
factor (VEGF) expression, neointima formation, cell loss, and vasa vasorum ingrowth. VEGF (A, B) and endothelial
NO synthase (eNOS) (C, D, F, G, H) positive immunoreactivity ⫽ brown stain. Cell nuclei counterstained with
hematoxylin (blue). Dashed black line ⫽ internal elastic lamina (identified by phase contrast microscopy); *region
of cell loss in intima/media. Six-day preterm newborn baboon [normoxia group (A, C)]; 6-day preterm newborn
[hyperoxia group (B, D, E, F, H)]; and fetus (140 days gestation) (G). A: no VEGF stain in DA from normoxic
newborn. B: moderate VEGF stain and enlarged neointima in DA from hyperoxic newborn. C: single layer of
endothelial cell line lumen of normoxic newborn. D: widened neointima, filled with luminal endothelial cells and
areas of cell loss in intima/media of hyperoxic newborn. E: pyknotic nuclei (small arrows) in neointima of hyperoxic
newborn. F: cell loss in intima/media of hyperoxic newborn. G: vasa vasorum restricted to adventitia of fetal DA
(arrowheads indicate furthest advance of vasa vasorum toward lumen). H: vasa vasorum invade muscle media of
hyperoxic newborn. Horizontal bar represents micrometers. In other experiments, immunostaining for von
Willebrand factor (VWF) (data not shown) gave the identical pattern as for eNOS.
CONTRACTILITY REGULATES DUCTUS REMODELING
R299
eling process due to its low intrinsic tone (Fig. 2A). In
addition, the preterm infant rarely develops the same
degree of profound DA hypoxia as found at term (Fig. 6A)
(2). At mild-to-moderate degrees of hypoxia, PG and NO
Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on April 20, 2017
Fig. 8. Relationship between VEGF expression and constriction of
the DA lumen in normoxic (n ⫽ 6) or hyperoxic (n ⫽ 6) preterm
newborn baboons. Units of VEGF (0–2) refer to intensity of immunostaining. Diameter of lumen: narrowest measurement of luminal
diameter at the time of necropsy (see Fig. 5C).
constriction was due to elevated PaO2, obliteration of
the lumen produced profound hypoxia in the center of
the constricted vessel (Fig. 6B). In contrast with the
animals whose PaO2 was maintained in the normoxic
range, those ventilated with 100% oxygen developed
both a similar degree of hypoxia and the same features
of anatomic remodeling as observed in the full-term
newborn (Figs. 6, 7, 8, and 9; see Ref. 2). Although
other explanations could account for the individual
appearances of VEGF and cell death in the wall of the
DA (15, 18, 22, 28, 29, 38, 42), their combined appearance and their geographic distribution can be completely accounted for by the intensity and distribution
of hypoxia as measured by EF5.
Perspectives
We used markedly elevated PaO2 to produce permanent DA closure. We are not recommending this as a
clinical approach to be used in preterm infants. Even
short periods of elevated PaO2 have been associated
with retinopathy of prematurity. Rather, these studies
lend support to the concept that functional obliteration
of the DA lumen is both necessary and sufficient to
initiate the remodeling process (even in the preterm
infant). We hypothesize that other more appropriate
treatments that cause sustained closure of the DA (e.g.,
indomethacin or a combination of indomethacin plus
NO inhibition) should also produce DA remodeling (9).
An additional speculation can be made from our findings. The profound hypoxia that occurs during postnatal
contraction in the full-term DA inhibits the production of
PG and NO (Fig. 2, B and C) (24). These two vasodilators
oppose the high intrinsic tone of the full-term DA; their
decreased production more than compensates for the loss
of oxygen-induced tension and enables the DA to remain
constricted as it remodels. On the other hand, even at
similar degrees of profound hypoxia, the preterm DA may
not be able to maintain its constriction during the remod-
Fig. 9. Neointima formation, luminal endothelial accumulation, cell
loss, and vasa vasorum ingrowth are increased in the DA wall of
preterm baboons ventilated with 100% oxygen (hyperoxia). A: neointimal thickness (in ␮m): means ⫾ SD. B: luminal endothelium. The
eNOS- and VWF-positive cells that lined the DA lumen formed either
single or multiple (ⱖ3) layers. C: cell loss was considered minimal if
the area of absent hematoxylin-stained nuclei was ⬎3,000 and
⬍15,000 ␮m2; moderate if ⬎20,000 ␮m2. D: vasa vasorum ingrowth.
Four separate, random sections from the middle of the DA were
stained for VWF and eNOS and photographed. The muscle media
was partitioned into concentric regions based on the percentage of
muscle media thickness between the intima and adventitia. Vessel
ingrowth was categorized based on the innermost region of the wall
that contained at least 1 vasa vasorum.
R300
CONTRACTILITY REGULATES DUCTUS REMODELING
production is no longer inhibited (Fig. 3); by contrast,
mild-to-moderate hypoxia does inhibit oxygen-induced
tone (Fig. 2A). These findings may explain why extremely
immature infants still have a high rate of reopening,
despite the elimination of luminal blood flow during DA
constriction (34).
The authors thank W. Cox for help in performing the Doppler
studies, V. Fontana for editorial assistance, H. Fernandez for technical assistance, and V. Winter and L. Buchanan for organizing and
distributing tissues. We also thank all the technicians in the Neonatal Intensive Care Unit at the Southwest Foundation, San Antonio, Texas, for help in caring for the preterm baboons.
This work was supported in part by National Heart, Lung, and
Blood Institute Grants HL-46691 and HL-56061 and a gift from the
Perinatal Associates Research Foundation.
1. Clyman RI. Developmental responses to oxygen, arachidonic
acid and indomethacin in the fetal lamb ductus arteriosus in
vitro. Prostaglandins Med 1: 167–174, 1978.
2. Clyman RI, Chan CY, Mauray F, Chen YQ, Cox W, Seidner
SR, Lord EM, Weiss H, Wale N, Evan SM, and Koch CJ.
Permanent anatomic closure of the ductus arteriosus in newborn
baboons: the roles of postnatal constriction, hypoxia, and gestation. Pediatr Res 45: 19–29, 1999.
3. Clyman RI, Goetzman BW, Chen YQ, Mauray F, Kramer
RH, Pytela R, and Schnapp LM. Changes in endothelial cell
and smooth muscle cell integrin expression during closure of the
ductus arteriosus: an immunohistochemical comparison of the
fetal, preterm newborn, and full-term newborn rhesus monkey
ductus. Pediatr Res 40: 198–208, 1996.
4. Clyman RI, Mauray F, Heymann MA, and Roman C. Influence of increased pulmonary vascular pressures on the closure of
the ductus arteriosus in newborn lambs. Pediatr Res 25: 136–
142, 1989.
5. Clyman RI, Mauray F, Roman C, Heymann MA, and Payne
B. Effect of gestational age on ductus arteriosus response to
circulating prostaglandin E2. J Pediatr 102: 907–911, 1983.
6. Clyman RI, Mauray F, Roman C, Rudolph AM, and Heymann MA. Circulating prostaglandin E2 concentrations and
patent ductus arteriosus in fetal and neonatal lambs. J Pediatr
97: 455–461, 1980.
7. Clyman RI, Mauray F, Rudolph AM, and Heymann MA.
Age-dependent sensitivity of the lamb ductus arteriosus to indomethacin and prostaglandins. J Pediatr 96: 94–98, 1980.
8. Clyman RI, Mauray F, Wong L, Heymann MA, and Rudolph AM. The developmental response of the ductus arteriosus
to oxygen. Biol Neonate 34: 177–181, 1978.
9. Clyman RI, Seidner S, and Koch C. Combined prostaglandin
and nitric oxide inhibition produces permanent closure of the
PDA in premature baboons (Abstract). Pediatr Res 45: 190A,
1999.
10. Clyman RI, Waleh N, Black SM, Riemer RK, Mauray F,
and Chen YQ. Regulation of ductus arteriosus patency by nitric
oxide in fetal lambs. The role of gestation, oxygen tension and
vasa vasorum. Pediatr Res 43: 633–644, 1998.
11. Coceani F, Armstrong C, and Kelsey L. Endothelin is a
potent constrictor of the lamb ductus arteriosus. Can J Physiol
Pharmacol 67: 902–904, 1989.
12. Coceani F, Huhtanen D, Hamilton NC, Bishai I, and Olley
PM. Involvement of intramural prostaglandin E2 in prenatal
patency of the lamb ductus arteriosus. Can J Physiol Pharmacol
64: 737–744, 1986.
13. Coceani F, Kelsey L, and Seidlitz E. Occurrence of endothelium-derived relaxing factor-nitric oxide in the lamb ductus
arteriosus. Can J Physiol Pharmacol 72: 82–88, 1994.
14. Crichton CA, Smith GC, and Smith GL. ␣-Toxin-permeabilised rabbit fetal ductus arteriosus is more sensitive to Ca2⫹
than aorta or main pulmonary artery. Cardiovasc Res 33: 223–
229, 1997.
Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on April 20, 2017
REFERENCES
15. Cullinan-Bove K and Koos RD. Vascular endothelial growth
factor/vascular permeability factor expression in the rat uterus:
rapid stimulation by estrogen correlates with estrogen-induced
increases in uterine capillary permeability and growth. Endocrinology 133: 829–837, 1993.
16. Evans SM, Joiner B, Jenkins WT, Laughlin KM, Lord EM,
and Koch CJ. Identification of hypoxia in cells and tissues of
epigastric 9L rat glioma using EF5 [2-(2-nitro-1H-imidazol-1-yl)N-(2,2,3,3,3-pentafluoropropyl) acetamide]. Br J Cancer 72: 875–
882, 1995.
17. Fineman JR, Takahashi Y, Roman C, and Clyman RI.
Endothelin-receptor blockade does not alter closure of the ductus
arteriosus. Am J Physiol Heart Circ Physiol 275: H1620–H1626,
1998.
18. Finkenzeller G, Marme D, Weich HA, and Hug H. Plateletderived growth factor-induced transcription of the vascular endothelial growth factor gene is mediated by protein kinase C.
Cancer Res 52: 4821–4823, 1992.
19. Fox JJ, Ziegler JW, Dunbar DI, Halbower AC, Kinsella JP,
and Abman SH. Role of nitric oxide and cGMP system in
regulation of ductus arteriosus tone in ovine fetus. Am J Physiol
Heart Circ Physiol 271: H2638–H2645, 1996.
20. Gaine SP, Hales MA, and Flavahan NA. Hypoxic pulmonary
endothelial cells release a diffusible contractile factor distinct
from endothelin. Am J Physiol Lung Cell Mol Physiol 274:
L657–L664, 1998.
21. Goldberg MA and Schneider TJ. Similarities between the
oxygen-sensing mechanisms regulating the expression of vascular endothelial growth factor and erythropoietin. J Biol Chem
269: 4355–4359, 1994.
22. Harada S, Nagy JA, Sullivan KA, Thomas KA, Endo N,
Rodan GA, and Rodan SB. Induction of vascular endothelial
growth factor expression by prostaglandin E2 and E1 in osteoblasts. J Clin Invest 93: 2490–2496, 1994.
23. Heistad DD and Marcus ML. Role of vasa vasorum in nourishment of the aorta. Blood Vessels 16: 225–238, 1979.
24. Kajino H, Chen YQ, Chemtob S, Waleh N, Koch CJ, and
Clyman RI. Tissue hypoxia inhibits prostaglandin and nitric
oxide production and prevents ductus arteriosus reopening.
Am J Physiol Regulatory Integrative Comp Physiol 279: R278–
R286, 2000.
25. Koch CJ, Evans SM, and Lord EM. Oxygen dependence of
cellular uptake of EF5 [2-(2-nitro-1H-imidazol-1-yl)-N-(2,2,3,3,3pentafluoropropyl)a cet amide]: analysis of drug adducts by fluorescent antibodies vs bound radioactivity. Br J Cancer 72:
869–874, 1995.
26. Kovitz KL, Aleskowitch TD, Sylvester JT, and Flavahan
NA. Endothelium-derived contracting and relaxing factors contribute to hypoxic responses of pulmonary arteries. Am J Physiol
Heart Circ Physiol 265: H1139–H1148, 1993.
27. Leach RM, Robertson TP, Twort CH, and Ward JP. Hypoxic
vasoconstriction in rat pulmonary and mesenteric arteries. Am J
Physiol Lung Cell Mol Physiol 266: L223–L231, 1994.
28. Li J, Perrella MA, Tsai JC, Yet SF, Hsieh CM, Yoshizumi
M, Patterson C, Endege WO, Zhou F, and Lee ME. Induction
of vascular endothelial growth factor gene expression by interleukin-1 beta in rat aortic smooth muscle cells. J Biol Chem 270:
308–312, 1995.
29. Liu Y, Christou H, Morita T, Laughner E, Semenza GL,
and Kourembanas S. Carbon monoxide and nitric oxide suppress the hypoxic induction of vascular endothelial growth factor
gene via the 5⬘ enhancer. J Biol Chem 273: 15257–15262, 1998.
30. McMurphy DM, Heymann MA, Rudolph AM, and Melmon
KL. Developmental changes in constriction of the ductus arteriosus: responses to oxygen and vasoactive agents in the isolated
ductus arteriosus of the fetal lamb. Pediatr Res 6: 231–238,
1972.
31. Moise KJ Jr, Huhta JC, Sharif DS, Ou CN, Kirshon B,
Wasserstrum N, and Cano L. Indomethacin in the treatment
of preterm labor: effects on the fetal ductus. N Engl J Med 319:
327–331, 1988.
32. Momma K and Toyono M. The role of nitric oxide in dilating
the fetal ductus arteriosus in rats. Pediatr Res 46: 311–315,
1999.
CONTRACTILITY REGULATES DUCTUS REMODELING
40. Sharpe G, Larsson K, and Thalme B. Studies on closure of
the ductus arteriosus. XII. In utero effect of indomethacin and
sodium salicylate in rats and rabbits. Prostaglandins 9: 585–
596, 1975.
41. Shattuck KE, Grinnell CD, Keeney SE, Noworyta K, and
Rassin DK. Hyperoxia and glutathione depletion in the isolated
perfused rat liver. J Investig Med 45: 576–583, 1997.
42. Shweiki D, Neeman M, Itin A, and Keshet E. Induction of
vascular endothelial growth factor expression by hypoxia and by
glucose deficiency in multicell spheroids: implications for tumor
angiogenesis. Proc Natl Acad Sci USA 92: 768–772, 1995.
43. Tannenbaum JE, Waleh NS, Mauray F, Gold L, Perkett
EA, and Clyman RI. Transforming growth factor-␤ protein and
messenger RNA expression is increased in the closing ductus
arteriosus. Pediatr Res 39: 427–434, 1996.
44. Tristani-Firouzi M, Reeve HL, Tolarova S, Weir EK, and
Archer SL. Oxygen-induced constriction of rabbit ductus arteriosus occurs via inhibition of a 4-aminopyridine-, voltage-sensitive potassium channel. J Clin Invest 98: 1959–1965, 1996.
45. Wolinsky H and Glagov S. Nature of species differences in the
medial distribution of aortic vasa vasorum in mammals. Circ Res
20: 409–421, 1967.
Downloaded from http://ajpregu.physiology.org/ by 10.220.33.2 on April 20, 2017
33. Nakanishi T, Gu H, Hagiwara N, and Momma K. Mechanisms of oxygen-induced contraction of ductus arteriosus isolated from the fetal rabbit. Circ Res 72: 1218–1228, 1993.
34. Narayanan M, Cooper B, Weiss H, and Clyman RI. Prophylactic indomethacin: factors determining permanent ductus arteriosus closure. J Pediatr 136: 330–337, 2000.
35. Niinikoski J, Heughan C, and Hunt TK. Oxygen tensions in
the aortic wall of normal rabbits. Atherosclerosis 17: 353–359,
1973.
36. Noel S and Cassin S. Maturation of contractile response of
ductus arteriosus to oxygen and drugs. Am J Physiol 231: 240–
243, 1976.
37. Oberhansli-Weiss I, Heymann MA, Rudolph AM, and
Melmon KL. The pattern and mechanisms of response to oxygen by the ductus arteriosus and umbilical artery. Pediatr Res 6:
693–700, 1972.
38. Pertovaara L, Kaipainen A, Mustonen T, Orpana A, Ferrara
N, Saksela O, and Alitalo K. Vascular endothelial growth factor
is induced in response to transforming growth factor-beta in fibroblastic and epithelial cells. J Biol Chem 269: 6271–6274, 1994.
39. Plate KH, Breier G, Millauer B, Ullrich A, and Risau W.
Up-regulation of vascular endothelial growth factor and its cognate receptors in a rat glioma model of tumor angiogenesis.
Cancer Res 53: 5822–5827, 1993.
R301