Bradykinin Production Associated
with Oxygenation of the Fetal Lamb
By Michael A. Heymann, M.B., B.Ch., Abraham M. Rudolph, M.D.,
Alan S. Nies, M.D., and Kenneth L. Melmon, M.D.
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
ABSTRACT
Previous studies showed that synthetic bradykinin produced in-vitro constriction of the ductus arteriosus of the fetal lamb and guinea pig, constriction
of isolated human and lamb umbilical vessels, and pulmonary vasodilatation in
the fetal lamb.
The present study showed that bradykinin precursor, kininogen, is present in
arterial blood of the fetal lamb by 61 days of gestation and that its concentration
increases toward term. We studied the bradykinin-generating system in 6 exteriorized fetal lambs. Kininogen concentrations in left atrial blood decreased
within 1 to 2 minutes after beginning ventilation with O2, and free bradykinin
was detected in left atrial blood. Pulmonary arterial blood kininogen concentration was not significantly altered. Kininogen concentration in left atrial blood
did not fall in 4 other lambs ventilated with No but did fall after subsequent
ventilation with O2. To study the effects of increased oxygenation in utero
without lung expansion, 6 pregnant ewes were exposed to hyperbaric O2 at 3.63
atmospheres absolute. Associated with a mean increase in brachial arterial blood
Po2 to 44 mm Hg, a value equivalent to that in the ventilated exteriorized fetus,
kininogen concentration fell, and free bradykinin was detected in the brachial
arterial blood of 3 of the 6 fetuses. In 4 other exteriorized lambs we measured
kininogen concentration in brachial arterial and right ventricular blood at frequent intervals up to 30 minutes after ventilation with O2. The concentration
in brachial arterial blood fell before that in right ventricular blood.
The hypothesis is presented that bradykinin is produced from kininogen in
the lungs of the fetal lamb when oxygenated, and that the maximal rate of
production occurs during the first few minutes after expansion of the lungs or
exposure of the ewe and fetus to hyperbaric O2.
ADDITIONAL KEY WORDS
kininogen
hyperbaric oxygenation
pulmonary blood
flow
electromagnetic flowmeter
• The change from the fetal to the neonatal
state is associated with three major readjustments in the circulation. These are pulmonary
From the Cardiovascular Research Institute and the
Departments of Pediatrics, Medicine, and Pharmacology, University of California San Francisco Medical
Center, San Francisco, California 94122.
This investigation was supported in part by U. S.
Public Health Service Crants HE 06285 and HE
09964 from the National Heart Institute and CM
01791 from the National Institute of General Medical
Sciences. Dr. Heymann was supported by a U. S.
Public Health Service Special Research Fellowship
from the National Institutes of Child Health and
Human Development and Dr. Nies by a U. S. Public
Health Service Postdoctoral Fellowship from the
National Heart Institute.
Received August 6, 1968. Accepted for publication
September 2, 1969.
Circulation Research, Vol. XXV, November 1969
vasodilatation, constriction of the ductus
arteriosus and constriction of the umbilical
vessels. Expansion or ventilation of the lungs
with various oxygen-containing gas mixtures
or pure oxygen (1-3) or an increase in
oxygenation of the blood or fluid perfusing the
pulmonary circulation (3-5), ductus arteriosus
(5-8), or umbilical vessels in vitro (9-10),
will produce these changes. However, it has
not yet been determined whether oxygen has
a direct effect on the smooth muscle components of these vessels, or whether its effect is
due to the release of some chemical mediator.
Kovalcik (8) demonstrated that bradykinin,
a potent endogenous vasoactive peptide,
constricted the ductus arteriosus of the fetal
521
522
HEYMANN, RUDOLPH, NIES, MELMON
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
lamb and guinea pig in vitro. Davignon et al.
(11) subsequently showed that bradykinin
constricted the human umbilical artery in
vitro; these studies were confirmed by Eltherington et al. (12). However, Lewis (13) has
recently shown that in one out of four in-vitro
preparations the lamb umbilical artery constricted with bradykinin, and then only with a
large dose (1000 ng). Campbell et al. (14)
have recently reported that small amounts (2
ng) of bradykinin produced pulmonary vasodilatation in the fetal lamb.
The bradykinin-generating system is present
in the human neonate, and free bradykinin
concentrations are significantly higher in
umbilical venous and arterial cord blood
samples than in maternal venous blood (15).
Since bradykinin was present in these samples
in concentrations large enough to produce
physiologic effects, we considered that bradykinin production may play a role in the
circulatory adjustments occurring at birth, and
particularly in pulmonary vasodilatation. Further studies in man on the control mechanism
of bradykinin generation at birth were not
feasible. We therefore decided to see whether
bradykinin could be produced by the intact
fetal lamb, and if so, to determine the
mechanisms of release of bradykinin at the
time of birth. We present data that indicate
that bradykinin generation in the lamb may
be initiated by an increase in arterial blood
Po 2 with or without inflation of the lungs.
Methods
CHEMICAL ASSAYS
Bradykinin.—This was measured on 5 ml of
whole blood, drawn directly into 10 ml of cold
30% perchloric acid, by a modification of the
method of .Webster and Gilmore (16). The
bradykinin was assayed by the four-point analysis, using contraction of the isolated uterus of an
estrus rat. This allows the separation of bradykinin from alkaline amines as previously described
(15). A major limitation of this method is that
bradykinin may be produced or destroyed in
vitro, and care must be taken that these processes
are uniform in all samples. To standardize the
method, 10 to 100 ng of synthetic bradykinin
(Sandoz) was added to a second 5-ml control
blood sample from each fetus. The range of
absolute recovery of this standard from these
control samples varied from 48% to 75% with
different animals; on a single sample, duplicate
recoveries varied from each other by less than
12%. Because a paired determination of the
percent recovery of a known amount of added
bradykinin with each sample would have resulted
in a significant depletion of blood volume, a single
measurement of percent recovery during the
control period was used to standardize all
samples. All concentrations were then corrected
to consider the recovery value as 100% in each
animal. The variability of bradykinin concentrations in duplicate analyses of a single sample or in
two samples obtained in rapid succession was less
than 8% after correction for recoveiy. The lowest
concentration detectable by this method is about
2 ng bradykinin /ml whole blood. Although
bradykinin can cause physiologic vascular
changes in lower concentrations (14), this does
not preclude the useful application of these
methods to detect large changes in bradykinin
production associated with a physiologic event.
CONFIRMATION OF BRADYKININ BIOASSAY
In view of some of the potential difficulties
associated with the variable recovery of bradykinin in the bioassay method as outlined, two
fetal lambs were studied to assess the production
of bradykinin in a different way. The fact that the
fetal lung will respond with pulmonary vasodilatation to extremely small amounts of injected
bradykinin (14) was utilized.
With the ewe under spinal analgesia, a fetal
lamb at term was exteriorized, with placental flow
continuing as previously described (17). Under
local anesthesia, through a left thoracotomy
incision, a cuff type of electromagnetic flow
transducer was placed around the left pulmonary
artery. Flow was then measured with a Statham
M4000-gated sine-wave electromagnetic flowmeter (17). The system was not calibrated
because we were not measuring actual flow rates
but only relative changes in flow. Polyvinyl
catheters were placed in the femoral and main
and left pulmonary arteries. Pressures, heart rate,
and flow were recorded on a Beckman Type R
Dynograph.
A second pregnant sheep was prepared
alongside the above preparation and the fetus
exteriorized; using local anesthesia, the trachea
was cannulated, taking care that no respiratory
effort was made by the lamb and that no air
entered the lungs. The tracheal tube was allowed
to drain freely into a beaker with the catheter tip
under 2 to 6 cm H 2 O. A fetal carotid artery and
maternal femoral artery were cannulated. Blood
samples were then obtained from the carotid
artery of this fetus for Idninogen and bradykinin
estimation by the rat uterus bioassay method
Circulation Research, Vol. XXV, November 1969
523
BRADYKININ PRODUCTION
100
FEMORAL
ARTERY
PRESSURE
mm Hg
5f)
0
100
PULMONARY
ARTERY
PRESSURE
mm Hg
5 0
0
FLOW
LEFT
PULMONARY
ARTERY
—
ZERO FLOW -
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
HEART
RATE
200 i
100 I
2 ml FETAL CAROT/D
ARTERIAL QLOOD
INTO L PA.
2 ml MATERIAL
FEMORAL ARTERIAL
BLOOD INTO L.PA.
4-ng SYNTHETIC
BRAOYKININ
INTO LRA.
FIGURE 1
Fetal femoral and main pulmonary arterial blood pressures, left pulmonary arterial blood flow,
and heart rate. Effects on flow of the injection into the left pulmonary artery of 2 ml carotid
arterial blood from a second fetus 10 minutes after commencing ventilation with 100% 02,
maternal femoral arterial blood, and 4 ng sijnthetic bradykinin are shown.
described. Immediately thereafter, a 2-ml carotid
arterial blood sample was withdrawn and without
delay injected over a 15-second period directly
into the left pulmonary arteiy of the first fetus
while pressures and left pulmonary arterial blood
flow were being continuously measured. These
samples were obtained during a control period,
and then at 1-minute intervals after ventilation
with 100% oxygen.
In both studies, left pulmonary arterial blood
flow was unaffected by the injection of carotid
arterial blood obtained during the control period.
All carotid arterial samples drawn after ventilation produced a transient increase in flow, and
baseline flow remained slightly higher after each
injection. An equivalent volume of well-oxygenated maternal femoral arterial blood injected over
the same time period failed to reproduce this
effect, as did the injection of a blood sample
obtained from the femoral artery of the fetus in
which flow was being measured. No change in
femoral or pulmonary arterial pressure occurred
with any of the injections. Figure 1 shows the
increase in left pulmonary arterial flow following
the injection of 2 ml of carotid arterial blood and
the effect of 4 ng of synthetic bradykinin injected
similarly over 15 seconds for comparison. This
amount of bradykinin would give an approximate
bradykinin concentration of 0.1 to 0.2 ng/ml
Circulation Research, Vol. XXV, November 1969
blood. The lack of effect of the injection of 2 ml
of maternal arterial blood is also demonstrated.
Kininogen.—Blood samples were drawn into
polyethylene syringes containing sufficient heparin to make a final concentration of at least 1
unit/ml. The specimens were centrifuged and the
plasma removed with siliconized or polyethylene
pipettes. Plasma kininogen was measured in
duplicate on 0.4 ml of plasma by the method of
Diniz and Carvalho (18). Kininogen is expressed
as the total amount of bradykinin released from 1
ml of plasma following the addition of trypsin.
The values given are therefore not specific
measurements of the actual amounts of kininogen
present but, rather, indicate the maximal amount
of bradykinin that can be produced from
substrate present in 1 ml of the sample plasma.
Blood kininogen concentrations were measured
more frequently because of the small volume of
blood required. There are also limitations to this
method. Little is known about the kinetics of the
breakdown of kininogen, and nonspecific destruction of kininogen can occur without conversion to
bradykinin. Therefore an in-vivo decrease in
blood kininogen concentration of a specific
amount would not necessarily produce an equivalent amount of circulating bradykinin. Nevertheless, in-vivo kininogen depletion in primates (19)
and in-vitro depletion in human cord blood (15)
524
HEYMANN, RUDOLPH, NIES, MELMON
O
O
2000
KININOGEN
ng/ml PLASMA
I 500
1000
500
60
70
J_
80
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
_L
JL
_L
90
100 110 120 130
GESTATIONAL AGE (DAYS)
14-0
150 MATERNAL
CONTROL
FIGURE 2
Umbilical arterial blood kininogen concentrations in 23 fetal lambs of different gestational ages;
the concentrations in 6 ewes are presented for comparison (open circles).
correlates with bradykinin production, especially
in those physiologic states in which the release of
tryptic enzymes, intravascular thrombosis, and
complement activation are unlikely to occur.
Results
BRADYKININ PRODUCTION IN THE FETAL LAMB
Although the fetal lamb responds to injected synthetic bradykinin (8, 12-14) and
fetal blood on contact with glass is capable of
producing a bradykinin-like substance (14),
there was no knowledge as to the gestational
period at which kininogen first appears or of
changes in kininogen concentration in the
intact fetal lamb during development. To
determine this we placed catheters, by methods previously described (17), in the umbilical vessels of 23 fetal lambs with gestational
ages of 61 to 148 days and obtained blood
samples for kininogen determination from the
umbilical artery without further disturbing the
fetus. Femoral arterial blood samples were
obtained from six ewes for comparison. The
kininogen concentration in umbilical arterial
blood rose gradually from early gestation to
term (Fig. 2). Term fetal and adult kininogen
concentrations were similar (mean maternal
value 1700 ng/ml plasma). We thus demonstrated that kininogen was present in the
blood of the fetal lamb.
EFFECT OF LUNG EXPANSION WITH OXYGEN
ON RELEASE OF BRADYKININ
Studies were designed to determine whether bradykinin was produced and could contribute to the pulmonary vasodilatation occurring in the fetal lungs during the transition
from the fetal to the neonatal circulation.
PULMONARY
ARTERY
3000r
2000 KININOGEN
ng / m l
PLASMA
1000 -
I
CONTROL
EXPANDED
CONTROL
I
EXPANDED
FIGURE 3
Fetal left atrial and pulmonary arterial blood kininogen
concentrations in 6 lambs during a control period and
after expansion of the fetal lungs with O 2 . Left atrial
concentrations fell considerably, whereas three of four
pulmonary arterial concentrations were essentially
unchanged.
Circulation Research, Vol. XXV, November 1969
525
BRADYKININ PRODUCTION
lambs. No bradykinin was detected in any of
these during the control period, but it was
present in all within 2 minutes after ventilation (Fig. 4). Left atrial blood Po2 was
measured in four lambs with a Radiometer
oxygen electrode and gas analyzer; it rose
from an average of 28 (range 23 to 31) mm
Hg to an average of 142 (range 38 to 330)
mm Hg. At the end of the procedure, the
position of the catheters was always confirmed
by direct observation.
These studies showed that following positive-pressure ventilation with oxygen, kininogen concentration fell and bradykinin was
present in left atrial blood.
6 0 r-
50 -
40
-
BRADYKININ
ng/ml
BLOOD
30
20
-
I 0 Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
CONTROL
EXPANDED
FIGURE 4
Fetal left atrial blood bradykinin concentrations in 4
lambs during a control period and after expanding the
fetal lungs with O2. No bradykinin was present during
the control period, but it was present after expansion
in all.
Six fetal lambs at term were exteriorized
with placental circulation continuing, as described before. Through a left thoracotomy,
polyvinyl catheters were inserted within pursestring sutures into the left atrium in all six
lambs and into the main pulmonary artery in
four. The thoracotomy was closed and the
fetus allowed a recovery period of 15 to 20
minutes.
Control blood samples for measurement of
kininogen were obtained from the left atrium
and the pulmonary artery. The fetal lungs
were then expanded and ventilated with 100%
O2. Two minutes later, blood samples were
again obtained from the same sites. Kininogen
concentration in left atrial blood fell from a
mean of 1933 (range 950 to 2940) ng/ml
plasma to a mean of 739 (range 250 to 955)
ng/ml plasma (P<0.01), 1 whereas concentrations in pulmonary arterial blood were
essentially unaltered in three of the four lambs
in which it was measured (Fig. 3).
Bradykinin concentrations were measured
in left atrial blood in only four of the six
•Statistical analysis by paired <-test.
Circulation Research, Vol. XXV, November 1969
ROLE OF OXYGEN VERSUS MECHANICAL EXPANSION
IN THE RELEASE OF BRADYKININ
To determine whether the release of bradykinin was effected simply by mechanical
expansion of the lungs or by oxygenation of
the blood, two series of studies were performed. In the first, four fetal lambs at term
were exteriorized and tracheotomies performed, as in the previous group. We decided
3000 i-
2000
KININOGEN
ng / m l
PLASMA
1000
CONTROL
N2
O2
EXPANDED
FIGURE 5
Fetal left atrial blood kininogen concentrations in 4
lambs during a control period, after expansion of the
lungs with N2 and then with O2. There is no change
after expansion with N2, but after expansion with O2
kininogen concentration fell.
526
HEYMANN, RUDOLPH, NIES, MELMON
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
not to subject the fetus to a thoracotomy and
instead inserted a catheter into the brachial
artery. As the circulation time from left atrium
to brachial artery is only 1 to 2 seconds, a
sample from this site would be representative
of a well-mixed left atrial sample. A control
brachial arterial blood sample was obtained
for measurement of kininogen, and in three
lambs for Po2 as well. The fetal lungs were
then expanded and ventilated with nitrogen.
After 2 minutes of ventilation, blood samples
were obtained and the ventilating gas was
changed to 100% O2. After a further 2-minute
period, brachial arterial blood samples were
again obtained from measurement of kininogen and Po2.
After ventilation with nitrogen, there was
no fall in kininogen concentration in three of
the four lambs and no significant fall in the
group as a whole (P > 0.8) (Fig. 5). After
ventilation with oxygen, there was a significant fall in mean kininogen concentration
from 2108 (range 1620 to 2600) ng/ml plasma
to 1262 (range 950 to 1680) ng/ml plasma
(P<0.05).
The average control brachial arterial blood
Po2 was 20 mm Hg, and this was not
significantly different from the level of 18 mm
Hg after ventilation with nitrogen. After
ventilation with oxygen, the Po2 rose to an
average of 58 mm Hg. These studies thus
demonstrated that mechanical expansion
alone of the fetal lungs did not result in a
decrease in kininogen due to activation of the
bradykinin-generating system.
EFFECTS OF HYPERBARIC OXYGENATION
The second series was concerned with the
effects of increased fetal oxygenation without
exteriorization and without ventilation of the
KININOGEN
ng /ml PLASMA
2000 | -
1500
1000 -
500 -
CONTROL
10
20
RECOVERY
MINUTES
FIGURE 6
Fetal brachial arterial blood kininogen concentrations in 6 lambs when the ewe was breathing
O2 at high pressure. Time 0 represents the time when 3.63 ATA was attained. The shaded
area represents the period of increased pressure. The time of exposure varied in different
animals; the symbol on the shaded area represents the time when decompression began. Control
concentrations were obtained at 1 ATA with the mother breathing room air. Recovery samples
were obtained 30 minutes after the mother returned to 1 ATA and was breathing room air.
Circulation Research, Vol. XXV, November 1969
527
BRADYKININ PRODUCTION
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
fetal lungs. Six pregnant ewes with fetal
gestational ages ranging from 120 days to
term (148 days) were studied in the Duke
University Hyperbaric Unit. Following maternal spinal analgesia, the umbilical vein and
artery were cannulated. Fetal forelimb and
hindlimb veins and arteries were cannulated
after delivery of each limb through a small
hysterotomy. In three instances, an additional
forelimb catheter was advanced into the right
ventricle and then into the pulmonary artery.
When the position of the catheters was
checked after termination of the experiment,
one was found to have passed through the
ductus arteriosus into the aorta; the other two
were in the pulmonary artery. All catheters
were secured in place, and the ewe's abdomen
was closed. More detailed descriptions of
these methods have been published previously
(17). Following a recovery period of 30
minutes, a control sample for measurement of
blood bradykinin and Po-2 and duplicate
samples (three in one fetus) for kininogen
were obtained from the brachial artery.
Kininogen concentration in pulmonary arterial
blood was also measured in two lambs with
pulmonary arterial catheters. The ewe was
then given 100% O2 to breathe, using
an oxygen hood. The ewe and fetus
were exposed to pressure (39 psi) to
reach 3.63 atmospheres absolute (ATA). The
time taken to reach maximal pressure varied
from 4 to 16 minutes, with an average of 7.7
minutes, and the time taken to obtain at least
three complete sets of blood samples while at
3.63 ATA was 15 to 28 minutes. The
individual durations of maintenance at 3.63
ATA for each animal are shown on the shaded
area in Figure 6.
Changes in brachial arterial blood kininogen concentrations are plotted in Figure 6.
There was a decrease in each of the six lambs;
the lowest level occurred in 5 within the first 7
minutes. Comparison by t-test of the mean of
the five paired controls and the mean of the
paired samples obtained between 6 and 15
BRADYKININ
ng/ml BLOOD
80
3.63 ATA
30 20 10 -
CONTROL
10
20
MINUTES
30
RECOVERY
FIGURE 7
Fetal brachial arterial blood bradykinin concentrations in the same 6 lambs as in Figure 5 in
which bradykinin was measured before, during, and after hyperbaric oxygenation.
Circulation Research, Vol. XXV, November 1969
528
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
minutes after reaching 3.63 ATA showed a
significant fall in kininogen concentration
(P < 0.01). After the initial decrease in
brachial arterial blood kininogen, concentrations in two of the lambs showed some
spontaneous recovery toward the control
values, but in the others leveled off with little
further change. In the two lambs in which
pulmonary arterial blood kininogen concentrations were measured, these decreased as well,
paralleling the brachial arterial concentrations
but at a slightly higher level.
Bradykinin was not detected in the control
period in any of the lambs. In three,
bradykinin was present after oxygenation; the
concentrations are shown in Figure 7. In the
other three, none was detected during the
period of hyperbaric oxygenation. Brachial
arterial blood Po2 values rose from an average
of 19 mm Hg during the control period to an
average of 44 mm Hg during hyperbaric
oxygenation.
The ewe was then allowed to breathe room
air and the hyperbaric period terminated for
each animal (Fig. 6). The time required for
decompression varied according to the time
spent at maximal pressure. After again reaching 1 ATA, a 30-minute recovery period was
allowed, and brachial arterial blood samples
were obtained in five lambs for measurement
of kininogen concentrations. These were not
significantly different from the means of the
control values (P>0.8).
Bradykinin was measured in only one lamb
during recovery. In this lamb (Figs. 6 and 7,
solid circles) the kininogen concentration was
lower during the recovery period than at the
termination of the hyperbaric period. Bradykinin concentration was elevated when kininogen was falling and fell as kininogen concentration rose. This indicated active bradykinin
production with kininogen depletion at the
same time.
Brachial arterial blood Poo values during
the recovery phase were not significantly
different from those obtained during the
control period (P > 0.8). Thus an increase in
fetal arterial Po2 alone, without mechanical
expansion and ventilation of the lungs, was
HEYMANN, RUDOLPH, NIES, MELMON
associated with the production of bradykinin
(probably in the lungs) in three of six fetal
lambs, with a resultant depletion of kininogen.
Unlike three of the four animals in the earlier
group of experiments, pulmonary arterial
blood kininogen concentrations did fall, although not to the same degree as in brachial
arterial blood.
PULMONARY BLOOD FLOW DURING HYPERBARIC
OXYGENATION WITH BRADYKININ PRODUCTION
In three of the six fetal lambs studied
during hyperbaric oxygenation, umbilical
blood flow was measured by the antipyrine
method (17), and cardiac output and distribution of flow to organs were measured using
injections of nuclide-labeled microspheres
(20). Flows and distribution were measured
during the control period and then immediately before the end of the period of hyperbaric
oxygenation in each animal. Umbilical blood
flow and cardiac output showed no marked or
consistent changes and pulmonary and systemic arterial blood pressures were unaltered.
Pulmonary blood flow showed a considerable
rise with hyperbaric oxygenation. Expressed
as a percent of cardiac output, pulmonary flow
in the individual lambs increased from 2.3% to
5.4%, 5.6% to 32.5%, and 3.8% to 8.2%. These increases are less marked than at the time of birth
but the presence of an undisturbed, low-resistance placental circulation is possibly responsible for this disparity, since in the fetus changes
in pulmonary blood flow depend on the relative resistances between the pulmonary and
systemic vascular beds. Changes in kininogen
and bradykinin concentrations in these three
lambs are shown in Table 1.
TIME FACTORS INVOLVED IN THE RELEASE
OF BRADYKININ
Pulmonary arterial blood kininogen concentrations fell in the fetuses studied during
hyperbaric oxygenation but not in three of the
four exteriorized ventilated lambs. Since the
former group was followed for a longer time,
we wondered whether this change in pulmonary arterial kininogen concentration was time
related. This led to a group of experiments in
which we examined more closely the time
sequence of kininogen concentration changes
Circulation Research, Vol. XXV, November
1969
529
BRADYKININ PRODUCTION
TABLE 1
Brachial Arterial Blood Kininogen and Bradykinin Concentrations and Pulmonary Arterial Blood
Flow {Percent Cardiac Out-put) in Three Fetal Lambs before and during Hyperbaric Oxygenation
Fetus
Time*
(mln)
Kininogen
(ng/ml plasma)
Control
1200
1460
1350
720
950
1000
3
7
10
23
25
Bradykinin
(ng/ml blood)
Cardiac
output to lungs
2.3
0
0
7.5
5.4
Control
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
3
7
15
17
1475
1500
975
530
500
5.6
0
0
18
32.5
Control
2
6
13
21
24
1070
1000
1170
575
600
675
3.8
0
0
0
8.2
*Time in minutes after reaching 3.63 ATA.
in both the brachial and pulmonary arterial
blood. Four fetal lambs at term were exteriorized and the trachea cannulated as described. In these animals small doses of
succinylcholine (0.25 mg/kg/min iv) were
administered to the fetus to prevent movement and facilitate the procedure. A brachial
artery was cannulated and a catheter passed
from a peripheral forelimb vein into the right
ventricle. Once again, catheter position was
checked carefully at the end of each experiment. Two or three control samples were
obtained from both the brachial artery and
right ventricle to measure kininogen concentrations and from the brachial artery to
measure Po2. The lungs were then ventilated
with 100% Oo, and samples for kininogen from
both sites obtained every minute for 5
minutes, then every 2 minutes for an additional 6 minutes, and at varying times thereafter
for a maximum of 30 minutes. Brachial arterial
Po., was measured as well. In all instances
Circulation Research, Vol. XXV,
November
1969
there was no significant difference between
the right ventricular and brachial arterial
kininogen concentrations during the control
period (P —0.8). After ventilation, the right
ventricular blood concentration remained consistently higher than the brachial arterial
concentration. An example of one of these
studies is shown in Figure 8. Brachial arterial
Po2 rose from an average of 23 mm Hg to an
average of 160 mm Hg.
CONFIRMATION OF BRADYKININ BIOASSAY:
BRADYKININ PRODUCTION IN THE VENTILATED FETUS
Carotid arterial blood kininogen concentration after ventilation in both animals followed
patterns similar to those described earlier.
Bradykinin concentrations in carotid blood
rose immediately in one fetus to peak at a
concentration of 64 ng/ml whole blood 2
minutes after ventilation. In the second fetus,
the appearance of bradykinin in carotid
arterial blood was delayed till 6 minutes after
ventilation, and the peak concentration a
530
HEYMANN, RUDOLPH, NIES, MELMON
KININOGEN
ng /ml PLASMA
3000 i -
2000
RV
1000
BA
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
CONTROL
J_
10
5
MINUTES
-5
FIGURE 8
Fetal right ventricular (RV) and hrachial arterial (BA) blood kininogen concentrations in
relation to time after expansion of the lungs with O2 in a single fetus. Similar patterns were
observed in 3 other animals.
minute later was 3.5 ng/ml whole blood. In
this fetus at the time of the last sample 10
minutes after ventilation, no bradykinin was
detectable by the estrus rat uterus bioassay.
The sample obtained immediately afterwards
and injected into the left pulmonary artery of
the second fetus produced the change shown
in Figure 1.
These studies showed that following ventilation with oxygen, some substance was
released that caused pulmonary vasodilatation. The effect was not due to the injection of
blood with a high POj since maternal arterial
blood failed to produce an effect. Although
this method could not establish the exact
identity of the substance, small amounts were
TABLE 2
Simultaneous Umbilical Arterial and Umbilical Venous Blood Kininogen Concentrations in Four
Lambs before Inmg Expansion and after Expansion with 100% 0s and in Two Animals before and
during Hyperbaric Oxygcnation
Arterial
Venous
Before Expansion
1850
1800
1880
1950
2750
2500
1850
2000
Before Hyperbaric Oxygenalion
985
1075
1000
1070
Arterial
Venous
After Expansion
620
1120
1160
1300
670
1280
1750
1200
During Hyperbaric Oxygenalion
575
820
600
670
Values are ng/ml plasma.
Circulation Research, Vol. XXV, November 1969
BRADYKININ PRODUCTION
detectable, and as the response was similar to
that produced by synthetic bradykinin, possibly it was bradykinin. In addition, increased
bradykinin concentrations were detected in
some samples by the rat uterus bioassay
method; vasodilatation occurred in response to
the injection of samples obtained at the same
time.
Discussion
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
The exact mechanism producing pulmonary
vasodilatation after birth is not yet fully
defined. Colebatch et al. (1) showed that
gaseous expansion of fetal lungs with a
mixture of 3% O2 and 1% CO2 in N2, while not
significantly altering arterial pH and blood
gases, produced pulmonary vasodilatation.
However, further pulmonary vascular dilatation results from a decrease in CO2 content or
increase in O2 content of the ventilating gas
(3). Expansion of the lungs with deoxygenated dextran or saline solution produced little
change in pulmonary vascular resistance, but
when oxygenated dextran solution or arterial
blood was used, pulmonary vascular resistance
fell markedly (2). These responses must be
locally effected, since they are not prevented
by vagotomy and sympathetic blockade.
Dawes (21) has suggested that at least onethird of the total fall in resistance is due to the
gaseous expansion alone, and the remainder is
then produced by the increase in Po2 and fall
in PcOa. However, the exact mechanisms by
which the increase in Po2 actually mediates
these changes in pulmonary vascular resistance are unknown.
We have shown that kininogen is present in
the blood of the fetal lamb as early as 61 days
of gestation and that with maturation of the
fetus the concentration increases. We were
able to demonstrate that the bradykiningenerating system was activated when fetal
arterial Po2 was increased, either by ventilating the fetal lungs with oxygen or by exposing
the ewe and fetus to hyperbaric oxygenation
without inflating the lungs. The evidence we
have obtained suggests that this release of
bradykinin occurred in the lungs. We would
have preferred to obtain pulmonary venous
Circulation Research, Vol. XXV, November 1969
531
blood samples to assess the release more
accurately; however, this was precluded by
the extensive surgical manipulation required.
We fully appreciate that left atrial blood
samples may not represent true pulmonary
venous blood, since during fetal life the left
atrium receives a large volume of blood
shunted across the foramen ovale from the
inferior vena cava, and this in turn comprises
largely umbilical venous blood. Expansion of
the lungs results in an increased left atrial
pressure due to increased pulmonary venous
return with a resultant marked decrease or
even complete cessation of right-to-left shunt
across the foramen ovale. Therefore, after
ventilation, a left atrial blood sample would in
fact represent predominantly pulmonary venous blood. This would also be true in the
animals subjected to hyperbaric oxygenation,
since with an increase in Po,, of blood
perfusing the lungs, pulmonary vascular resistance falls, with expected similar effects on
foramen ovale shunting. To completely exclude the possibility that changes in concentration of kininogen in left atrial blood could
have been produced by changes in that in the
umbilical vein, umbilical venous and arterial
blood samples were obtained before and after
ventilation of the fetal lungs with oxygen in
four lambs, and before and during hyperbaric
oxygenation in two lambs (Table 2). There
were no significant differences between the
umbilical venous and arterial concentrations
in either the control period or following
oxygenation (P>0.08). These studies suggested, therefore, that the decreases in kininogen concentration in left atrial blood were in
fact a true reflection of changes in pulmonary
venous concentrations. However, the remote
possibility that kininogen depletion occurred
in organs of the lower body of the fetus was
not excluded.
Bradykinin could not be detected at all in
left atria! blood in three instances, or was
detected only well after the maximal fall in
kininogen. (Table 1). This does not exclude
the fact that it was released in the lung.
Bradykinin is effective in extremely small
doses, and in some experiments (14) it has
532
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
been shown that injection of 1 to 2 ng of
bradykinin directly into the pulmonary artery
produces pulmonary vasodilatation. Blood
concentrations following the injection of this
amount may not be readily detected by the
assay methods we used. Recent studies by
Ferreira and Vane (22) have shown that in
adult cats, up to 80% of intravenously injected
bradykinin may be taken up by the lungs or
destroyed in a single passage through the
lungs. This, too, and the possibility that the
bradykinin produced was actually utilized in
the lungs to produce pulmonary vasodilatation
could explain the apparent disparity between
the amount of kininogen depletion and the
small and variable amounts of bradykinin
detected in left atrial blood.
The bradykinin concentrations in arterial
blood in several of the fetal lambs were in the
same range as those considered by Nies et al.
(19) to play a role in the cardiovascular
collapse associated with endotoxic shock in
monkeys. There was no effect of these high
bradykinin concentrations on cardiac output,
umbilical blood flow, systemic arterial blood
pressure, and heart rate in the fetal lambs.
Infusion for 15 seconds into the descending
aorta of a fetal lamb of sufficient synthetic
bradykinin (2/u,g) to produce a concentration
approximately equivalent to the highest brachial arterial concentration obtained induced
no change in systemic arterial pressure. Twoand threefold amounts produced a minimal
fall in pressure (2 to 5 mm Hg). The injection
of 100 fjug of synthetic bradykinin directly into
the umbilical artery had no effect on systemic
arterial pressure. Nies et al. (19) have
suggested that other mechanisms such as
endotoxic blockade of sympathetic reflexes,
rather than an increased bradykinin concentration alone, could be responsible for the
sustained peripheral vascular dilatation in
endotoxic shock.
In the present studies, the only physiologic
change required to initiate bradykinin production was an increase in Po,,. Mechanical
expansion of the lungs in the absence of
oxygen could not be demonstrated to play any
role in bradykinin production. Following the
HEYMANN, RUDOLPH, NIES, MELMON
exposure to oxygen, kininogen was rapidly
depleted, with a considerable and rapid fall in
kininogen concentration in left atrial blood.
Our initial studies did not adequately show
the course of these events; however, the latter
two groups of studies demonstrated that the
kininogen concentration of blood perfusing
the lungs remained at control levels for 2 to 3
minutes, and only then started to fall. The fall
in pulmonary arterial kininogen concentration
in the one fetal lamb in Figure 3 could
possibly be explained by the fact that the rate
of fall of pulmonary arterial kininogen may be
variable or that the time taken to oxygenate
the fetus may also be variable.
These observations indicated that sudden
bradykinin production occurred in the fetal
lungs and was dependent only on an increase
in POj regardless of whether this was produced by ventilation with oxygen or by
hyperbaric oxygenation. From these studies,
we were unable to assess the level of Porequired to initiate the process, but in-vitro
studies (15) suggested that a Po- of more
than 35 mm Hg was the critical level. The
duration and amount of total bradykinin
production could not be accurately assessed.
Although the initial rapid decrease in kininogen suggests that bradykinin was rapidly
produced and that the rate of bradykinin
generation then decreased, we do not know
what mechanisms regulate kininogen production at this time of rapid turnover. There is
some evidence in the rhesus monkey (19) that
kininogen may be produced and released
more rapidly than would be expected of most
alpha-2-globulins, and this may also be true in
the lamb. Also, the level of bradykinin in left
atrial blood may not necessarily reflect the
actual amount produced, since it may be
utilized in varying amounts and rates in the
lung.
We have not determined the direct mechanism of bradykinin production in the lamb. In
the human, granulocyte kallikrein or kallikrein
activator has been implicated in bradykinin
production (15). We measured differential
granulocyte counts in the left atrial blood in
some of the lambs in this study but found no
Circulation Research, Vol. XXV,
November
1969
533
BRADYKININ PRODUCTION
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
direct relation between granulocyte count and
bradykinin production. It is also clear that a
temperature fall could not have been a major
stimulus for bradykinin production in these
studies, since in the hyperbaric studies the
fetal lamb remained in utero.
This study has shown that in the lamb
bradykinin is produced in sufficient quantities
to produce pulmonary vasodilatation; bradykinin production is oxygen dependent and not
initiated by mechanical expansion of the lung;
it is produced in the lung but may possibly be
formed in the systemic circulation also, and
the production is associated with pulmonary
vasodilatation occurring with the onset of
respiration in the lamb. However, we have not
as yet been able to determine whether the
effects of oxygen on the pulmonary circulation
are completely mediated through local bradykinin release, or whether oxygen has an
independent direct vasodilator effect as well.
P. V., JR.: Effects of hyperbaric oxygen on
uteroplacental and fetal circulation. Circulation
Res. 22: 573, 1968.
6. BORN, G. V. R., DAWES, G. S., MOTT, J. C , AND
RENNICK, B. R.: Constriction of the ductus
arteriosus caused by oxygen and by asphyxia in
newborn lambs. J. Physiol. (London) 132:
304, 1956.
7. ASSALI, N. S., MORRIS, J. A., SMITH, R. W., AND
MANSON, W. A.: Studies on ductus arteriosus
circulation. Circulation Res. 13: 478, 1963.
8. KOVALCIK, V.: Response of the isolated ductus
arteriosus to oxygen and anoxia. J. Physiol.
(London) 169: 185, 1963.
9. NYBERC,
11.
The authors wish to express their gratitude to Mr.
Sidney Steinberg for his advice on instrumentation,
and to Misses Alice Lytle, Michele Sanda, Dorothy
Reese and Christine Mueller and Mr. Robert Sirbu for
their skillful technical assistance.
The hyperbaric studies were performed at Duke
University Medical Center, Durham, North Carolina.
We are deeply grateful to Drs. Saltzman, Brumley and
Risemberg and the members of the hyperbaric unit for
affording us the opportunity to perform these studies
and for their invaluable assistance during them.
W., AND NADEAU, R. A.: Nervous control of
the circulation in the foetal and newly
expanded lungs of the lamb. J. Physiol.
(London) 178: 544, 1965.
2.
LAUER,
R. M., EVANS, J. A., AOKI,
14.
3.
4.
ASSALI, N. S., KIHSCHBAUM, T. H., AND DILTS,
Circulation Research, Vol. XXV, November 1969
of
DAVIGNON, J., LORENZ, R. R., AND SHEPHERD,
ELTHERINGTON,
L. G., STOFF, J., HUGHES,
T.,
CAMPBELL, A. C. M., DAWES, G. D., FISHMAN,
A. P., HYMAN, A. I., AND PEHKS, A. M.:
MELMON, K. L., CLINE, M.
J., HUGHES, T., AND
NIES, A. S.: Kinins: Possible mediators of
neonatal circulatory changes in man. J. Clin.
Invest. 47: 1295, 1968.
16.
WEBSTER, M. E., AND CILMORE, J. P.: Estimation
of the kallidins in blood and urine. Biochem.
Pharmacol. 14: 1161, 1965.
17.
RUDOLPH,
A.
M.,
AND HEYMANX,
M.
A.:
Validation of the antipyrine method for
measuring fetal umbilical blood flow. Circulation Res. 21: 185, 1967.
18.
DINIZ,
C.
R.,
AND CAKVALHO,
I.
F.:
Micro
method for determination of bradykininogen
under several conditions. Ann. N. Y. Acad. Sci.
104: 77, 1963.
RUDOLPH, A. M., AND YUAN, S.: Response of the
pulmonary vasculature to hypoxia and H+ ion
concentration changes. J. Clin. Invest. 45: 399,
1966.
5.
15.
CASSIN, S., DAWES, G. S., MOTT, J. C , Ross, B.
B., AND STRANC, L. B.: Vascular resistance of
the foetal and newly ventilated lung of the
lamb. J. Physiol. (London) 171: 61, 1964.
Influence
Release of a bradykinin-like pulmonary vasodilator substance in foetal and new-born lambs.
J. Physiol. (London) 195: 83, 1968.
H., AND
KITTLE, C. F.: Factors controlling pulmonary
vascular resistance in fetal lambs. J. Pediat.
67: 568, 1965.
B.:
AND MELMON, K. L.: Constriction of human
umbilical arteries: Interaction between oxygen
and bradykinin. Circulation Res. 22: 747,
1968.
13. LEWIS, B. V.: Response of isolated sheep and
human umbilical arteries to oxygen and drugs.
J. Obstet. Gynaecol. Brit. Commonwealth 75:
87, 1968.
References
1. COLEBATCH, H. J. H., DAWES, C. S., COODWIN, J.
AND WESTIN,
J. T.: Response of human umbilical artery to
changes in transmural pressure. Am. J. Physiol.
209: 51, 1965.
12.
Acknowledgment
G.,
oxygen tension and some drugs on human
placental vessels. Acta Physiol. Scand. 39:
216, 1957.
10. PANIGEL, M.: Placental perfusion experiments.
Am. J. Obstet. Gynecol. 84: 1664, 1962.
19.
NIES, A. S., FORSYTH, R. P., WILLIAMS, H. E.,
AND MELMON, K. L.: Contribution of kinins to
endotoxin shock in unanesthetized rhesus
monkeys. Circulation Res. 22: 155, 1968.
534
20.
RUDOLPH, A. M., AND HEYMANN, M. A.:
Circulation of the fetus in utero: Methods for
studying distribution of blood flow, cardiac
output and organ blood flow. Circulation Res.
21: 163, 1967.
21. DAWES, G. S.: Foetal and Neonatal Physiology.
HEYMANN, RUDOLPH, NIES, MELMON
Chicago, Year Book Medical Publishers, Inc.,
1968, p. 83.
22. FEKREIHA, S. H., ASO VANE, J. R.: Disappearance
of bradykinin and eledoisin in the circulation
and vascular beds of the cat. Brit. J.
Pharmacol. 30: 417, 1967.
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Circulation Research, Vol. XXV, November 1969
Bradykinin Production Associated with Oxygenation of the Fetal Lamb
MICHAEL A. HEYMANN, ABRAHAM M. RUDOLPH, ALAN S. NIES and KENNETH L.
MELMON
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Circ Res. 1969;25:521-534
doi: 10.1161/01.RES.25.5.521
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1969 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/25/5/521
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/