Clinical Science (1994) 87, 427434 (Printed in Great Britain)
421
Role for endothelin in the renal responses to
radiocontrast media in the rat
S. OLDROYD’, S.-J. SLEE’, J. HAYLOR3,S. K. MORCOS‘ and C. WILSON’
Deportment of ’Medicine, ‘Department of Diagnostic Imaging and ‘Shefield Kidney Institute,
Northern General Hospital, Shfleld. U.K.. and =Zeneca Pharmaceuticals, Alderley Park,
Macclesfield, Cheshire, U.K.
(Received 2 November 1993/28 March 1994, accepted 25 May 1994)
1. The involvement of endothelin in the renal res-
ponses to radiocontrast media was examined in the
rat in v i m and in vivo using BQ123, a selective
endothelin (ET,) receptor antagonist, and phosphoramidon, an endothelin-converting enzyme inhibitor.
2. For experiments in vitro, an isolated perfused rat
kidney was employed perfusing in closed circuit with
an albumin-based Krebs-Henseleit solution. The
effects of BQ123 and phosphoramidon on the renal
responses to iotrolan (isossmolar radiocontrast
media) and diatrizoate (high-osmolar radiocontrast
media) were examined. In vivo, renal conductance was
measured using a Doppler flow probe in the anaesthetized rat pretreated with indomethacin, and the
effects of BQ123 and phosphoramidon on the renal
response to intravenous diatrizoate were examined.
3. In vitro, iotrolan and diatrizoate both produced a
biphasic effect on the glomerular filtration rate, characterized by a transient increase followed by a sustained
fall. Pretreatment with BQ123 (10pmol/l), but not
phosphoramidon (1mmol/l), prevented both the
increase and the sustained fall in glomerular filtration
rate induced by radiocontrast media.
4. In v i m , iotrolan and diatrizoate both produced a
sustained fall in renal perfusate flow. An initial
increase in renal perfusate flow was only observed
with diatrizoate. Pretreatment with BQ123 (10pmolll),
but not with phosphoramidon (1 mmol/l), markedly
inhibited the sustained fall in renal perfusate flow
produced by both iotrolan and diatrizoate. BQ123
(10pmol/l), however, markedly potentiated the renal
vasodilatation produced by diatrizoate.
5. In vivo, in the anaesthetized rat treated with
indomethacin, the sustained fall in renal conductance
produced by diatrizoate (2.9 g of iodine/kg, intravenously) was markedly inhibited by pretreatment
with BQ123 (1 mg/kg) but not with phosphoramidon
(10 mg/kg)*
6. A role for endothelin as a mediator of the renal
responses of radiocontrast media is proposed.
INTRODUCTION
Prospective clinical studies indicate an incidence
of 7-10% for the nephrotoxicity which follows the
administration of radiocontrast media (RCM) [ 13.
RCM are well known to produce a biphasic change
in the glomerular filtration rate (GFR) and renal
blood flow, characterized by a transient increase
followed by a more sustained fall [Z]. The possible
involvement of the renin-angiotensin system [3],
adenosine [4] or prostaglandins [S] in this event
has been proposed, but no clear evidence for any
particular pathway has been obtained. RCM have
recently been demonstrated to increase plasma
endothelin levels in the rat [6], in the dog [7] and
in man [8]. The endothelins are a family of isopeptides for which effects on the renal vasculature and
both glomerular and tubular function have been
described (for a review, see [9]). The synthesis of
endothelins involves the formation of inactive precursors known as big-endothelins, which are converted to their mature form by a putative
endothelin-converting enzyme (ECE) [lo]. Mature
endothelins exert their effects via actions on specific
receptors, which to date have been divided into two
sub-types, ETA and ETB [ll]. Since endothelin is a
potent vasoconstrictor peptide it could be involved
in both the acute renal response to RCM and
RCM-induced nephrotoxicity where enhanced
plasma endothelin levels have been demonstrated
PI-
In the present study, we have investigated the
renal effects of RCM, in oitro in the isolated perfused rat kidney (IPRK) and in uiuo in the anaesthetized rat. Evidence obtained using the IPRK shows
that RCM can exert direct effects on kidney function [lZ, 131, including a decrease in renal perfusate
flow (RPF) and GFR, osmotic diuresis, enhanced
protein excretion and cytoplasmic vacuolation of
the proximal tubule [14]. The role of endothelin in
the renal responses to diatrizoate (high-osmolar
Key words contmt media, endothelin. endothelin antagonist, phosphoramidon. renal function.
Abbreviations Cl,, clearance d [“Clinulin; ECE, endothelin-convertingenzyme; ET-I, endothelin-I; FR,, fractional reabsorption of sodium; GFR, glomerular filtration rate;
IPRK. isolated perfured rat kidney; MAP. mean systemic arterial blood pressure; RC, renal conductance; RCM, radiocontrast media; RPF, rend perfurate flow.
Correspondence Dr J Haylor, Floor G. Shdield Kidney Institute. Northern General Hospital. Herrier Road. Shdfield S5 7AU, U.K.
420
S. Oldroyd et al.
RCM) and iotrolan (iso-osmolar RCM) has been
assessed using BQl23, a selective endothelin (ETA)
receptor antagonist [ 151, and phosphoramidon, an
ECE inhibitor [16]. Experiments were also performed to examine the effects of BQ123 and phosphoramidon on the change in renal blood flow
induced by diatrizoate in the anaesthetized rat,
where indomethacin pretreatment was required
before a sustained decrease in renal conductance
(RC) to diatrizoate. RCM are well known to stimulate the synthesis of prostacyclin, a renal vasodilator
[17, 181.
METH0 DS
IPRK
Male Wistar rates (400-450 g), allowed free access
to food and water before experiment, were anaesthetized with thiopentone sodium (125 mg/kg, intraperitoneally). After cannulation of the right ureter and
superior mesenteric artery, the right renal artery was
cannulated by a non-ischaemic technique and the
right kidney was removed to a re-circulating perfusion apparatus, which we have described previously [19]. The right kidney was perfused in
uitro at a constant pressure of l00mmHg (recorded
from the renal artery) with an oxygenated perfusion
medium based on Krebs-Henseleit solution containing (mmol/l): N a + 147, K + 5.2, Ca’+ 2.5, Mg2+
0.96, CI- 128, H C 0 , - 25, SO,’- 0.96, glucose 5.5,
inorganic phosphate, 1.4 together with 6.7% albumin and 14mmol/l mixed amino acids maintained
at 37°C. RPF and renal perfusion pressure were
monitored continuously. Urine and perfusate samples were collected at 5min intervals. The GFR was
assessed as the clearance of [14C]inulin (Cln),measured as the clearance of [‘“C] after the addition of
1 pCi of [‘4C]inulin to the renal perfusate.
Experimental protocol
The first 30min of perfusion was regarded as an
equilibration period and the following 80 rnin as the
observation period. A continuous infusion of angiotensin I1 (5 pmol/min) was sustained throughout the
experiment to increase the vascular tone of the
preparation. Initial experiments were carried out to
establish the effectiveness of the endothelin receptor
antagonist, BQ 123, and the ECE inhibitor, phosphoramidon, under the perfusion conditions
employed. Cumulative concentration-response curves
were performed to the endothelin-1 (ET-1)-induced
decreases in RPF in the presence and absence
of BQ123 (10 pmol/l). Cumulative concentration-response curves were also performed to the
big endothelin-1 (big ET-1)-induced decreases in
RPF in the presence and absence of phosphoramidon ( 1 mmol/l). The cumulative change in RPF
after the addition of each dose of either ET-1 or big
ET-1 was expressed as a percentage of the value
obtained in the 5min period before their addition to
the perfusion system.
Baseline measurements of renal function, including RPF, Cln, urine flow and the fractional reabsorption of sodium (FR,,), were taken in the first
20min of the observation period. Diatrizoate or
iotrolan was added to the perfusion system (n= 6
per group) at 20min (regarded in the Figures as
time 0), giving an initial calculated perfusate concentration of 20 mg of iodine/ml, and measurements
of renal function were taken for a further 60min
period. Experiments with diatrizoate and iotrolan
were repeated after pretreatment with either BQ123
(10 pmol/l) or phosphoramidon (1 mmol/l) 10min
before the addition of RCM ( n = 6 per group). Urine
volume was measured gravimetrically and sodium
content was analysed by flame photometry. In Figs.
1-3, measurements of RPF, CI, and urine flow after
the addition of RCM were expressed as a percentage of the value obtained in the 5min period before
RCM addition.
Experiments in vivo
Male rats (Alderley Park, 300-380 g) were anaesthetized with thiopentone (100 mg/kg, intraperitoneally). The right jugular vein and carotid artery
were isolated, and cannulae, filled with heparinized
saline, were implanted for the administration of
compounds and the measurement of blood pressure,
respectively. The abdomen was opened and a pulsed
Doppler flow probe (model PD-20; Crystal Biotech)
was positioned around the left renal artery. Body
temperature (measured rectally) was maintained at
38°C using a heated pad. Animals were allowed to
stabilize for at least 15rnin before administration of
drugs. Throughout the experiment, continuous
recordings were made of mean systemic arterial
blood pressure (MAP), instantaneous heart rate and
mean Doppler shift signals. Changes in the latter
were taken as indices of flow changes, and RC was
calculated from the MAP and mean Doppler shift
signals.
Rats were divided into four groups and all rats
received an intravenous infusion of diatrizoate (Urogratin 325; 8.9ml/kg) over 2.5min giving a total
dose of 2.9g of iodine/kg. Group 1 (n=4) received
indomethacin vehicle (5% sodium bicarbonate)
5 rnin before diatrizoate. Group 2 (n= 6) received
indomethacin ( 5 mg/kg) 5 min before diatrizoate.
Group 3 ( n= 6) received indomethacin (5 mg/kg)
5 rnin before and BQ123 (1 mg/kg) 3 min before
diatrizoate. Group 4 (n= 6) received indomethacin
(5 mg/kg) followed by phosphoramidon (10 mg/kg)
3 min before diatrizoate.
Materials
Isovist 300 (Schering AG, Berlin, contains iotroIan, 300 mg of iodine/ml, 320 mosmol/kg H,O), Urogratin 325 (Schering AG, Berlin contains diatrizoate,
325 mg of iodine/ml, 1650mosmol/kg H,O), BQ 123
was synthesized by Dr A. Dutta (Zeneca Pharma-
Contrast media and endothelin
429
ceuticals). Phosphoramidon (Cambridge Research
Biochemicals), angiotensin 11, indomethacin (Sigma),
['4C]inulin (Amersham, Amersham, Bucks, U.K.)
and mixed amino acids (Vamin 14, Kabi Pharmaceuticals Sweden) were obtained from the suppliers
as indicated.
Statistics
All results were expressed as means) SEM. For
IPRK data, within-group comparisons were made
using a paired Student's t-test. For comparison
between groups, an analysis of variance was
employed using Minitab software. Data obtained in
oiuo were compared using the paired or unpaired
Student's t-test where appropriate.
I
so I
-20
lotrolan 20 mn/ml
...........................
1
-10
0
10
20
30
40
50
1
60
Time (min)
RESULTS
Studies in the IPRK
In the absence of RCM, all four measured variables of renal function were stable with time when
values taken over 15-20 and 75-80min of the
observation period were compared (n= 6): RPF
(27.2k0.8 to 27.9+0.9mlmin-'g-'), C,, (0.62f
0.06 to 0.63 f0.04 ml min - ' g- '), urine flow (0.11 f
0.01 to 0.12f0.01 mlmin-'g-') and FRN, (90.8f
2.0 to 91.4f 1.8%). Throughout the results section,
the data quoted are the means f S E M of the
values obtained in the 5 min period before the addition of RCM (baseline value) or in the 55-60
period after RCM addition (sustained response).
Where initial transient effects were observed, peak
values are quoted.
Effects of RCM alone
The time course of the effects of RCM on C,, is
show in Fig. 1 . Both RCM exerted a biphasic effect
on C,,. For iotrolan, this consisted of a transient
increase from 0.60 k 0.01 to 1.48 f0.04 ml min - ' g - '
(P< 0.05) followed by a sustained decrease to 0.45 f
0.02 ml min- ' g- ' (P<0.05). Similarly, diatrizoate
produced a transient increase in C,, from 0.69f0.08
to 1.82+0.05mlmin-'g-' (P<0.05) followed by a
sustained fall to 0.44+0.04mlmin-' g-' (P<0.05).
Diatrizoate, however, produced a significantly
greater sustained fall in C,, than iotrolan (P<0.05,
analysis of variance).
The time course of the effects of RCM on RPF is
shown in Fig. 2. Diatrizoate exerted a biphasic effect
on RPF consisting of a transient increase from
29.0f5.8 to 36.8+3.5mlmin-'g-' (P<0.05) followed by a sustained decrease to 19.9f3.4ml
min-' g-' (P<0.05). In contrast iotrolan produced only a sustained decrease in RPF from
27.8f 1.4 to 20.9f 1.2mlmin-'g-' (P<0.05). Diatrizoate produced a significantly greater sustained
Time (min)
Fig. 1. Timecourse of effects of iotrolan (a) and diatrizoate (b) on
4"in IPRK alone (0)
and after treatment with either BQIU ( 0 )
or phosphoramidon (0).RCM were added at time 0. Valuer are
exprerred a a percentage of the baseline obtained in the 5min period
before the addition of RCM to each kidney. Vertical bars indicate SEM
(n=S per group). Where bars are not apparent, the SEM is within the data
point.
fall in RPF than iotrolan (P<0.05, analysis of
variance).
A sustained, monophasic fall in FRN, was
observed after the addition of either iotrolan
(91.0f0.6 to 88.8f0.7%, P<O.O5) or diatrizoate
(87.5f4.4 to 81.3f2.3%, P<0.05) to the IPRK.
Diatrizoate produced a significantly greater fall in
FRN, than iotrolan (P <0.05, analysis of variance).
The time courses of the effect of diatrizoate and
iotrolan on urine flow are shown in Fig. 3. Diatrizoate produced a large initial increase in urine flow
(0.10f 0.02 to 0.39 kO.01 rnl min - ' g - ', P < 0.01)
associated with both the fall in FRN, and the
increase in C,,. Thereafter a modest sustained
increase in urine flow (to 0.15fO.O3mlmin-' g-',
P < 0.05) was observed. A modest sustained increase
in urine flow was also observed with iotrolan
(0.12 f to 0.16 fO.08 ml min- g- ', P < 0.05).
'
S. Oldroyd et al.
_ _ _ _ _lotrolan
_ _ _20mg/ml
____________-------
lotrolan 20mglml
. .. . . . . . . . . . . . . . . . . . . . . . . . .
50
I
I
-10
0
10
20
30
40
M
-10
-20
60
10
20
30
40
M
I
60
Time (min)
Time (min)
k
0
m-
e
e
a
.
B
250-
a4
s 200 P
0)
.-
3
I50
Fig. Z Timecourse of effects of iotrolan (a) and diatrizoate (b) on
RPF in IPRK alone (0)and after treatment with either BQlU
( 0 )or phosporamidon
RCM were added at time 0. Values are
expressed as a percentage of the baseline obtained in the 5min period
before the addition of RCM to each kidney. Vertical bars indicate SEM
( n 4 per group). Where bars are not apparent, the SEM is within the data
point.
(a).
Effexts of RCM after 88123
BQ123 (10pmol/l) itself had no sustained effect
on renal function. The effectiveness of BQ123
(10 pmol/l) as an endothelin receptor antagonist is
shown in Fig. 4, where cumulative concentrationresponse curves are shown to the ET-1-induced
decreases in RPF in the absence (n=4) and presence
(n=4) of BQ123 (lOpmol/l). The cumulative
concentration-response curve to ET-1 was shifted to
the right in the presence of BQ123 (lOpmol/l), in a
parallel fashion without a depression of the maximum, giving a calculated pK, for BQ123 of 6.89.
The time course of the effect of RCM on C,,
after BQ123 (10pmol/l) is shown in Fig. 1 (n=6).
In the presence of BQ123, no initial increase in C,,
and no sustained fall in C,,was observed after the addition of either iotrolan (0.62k0.08 to 0.61
+0.02 ml min- ' g- ', not significant) or diatri-
-
Fig. 3. Timecourse of effects of iotrolan (a) and diatrizoate (b) on
urine flow in IPRK alone ( 0 )and after treatment with either
BQlU ( 0 )or phosphoramidon (0).RCM were added at time 0.
Values are expressed as a percentage of the baseline obtained in the Smin
period before the addition of RCM. Vertical bars indicate SEM ( n 4 per
group). Where bars are not apparent, the SEM is within the data point.
zoate (0.66f0.02 to 0.69+0.06mlmin-'g-1, not
significant) to the IPRK.
The time course of the effect of RCM on RPF
after BQ123 is shown in Fig. 2. In the presence
of BQ123, the effect of iotrolan was markedly
reduced ( P< 0.05, analysis of variance) with only
a small sustained decrease in RPF (25.8+
P<0.05 n=6).
0.7 to 23.8&0.4mlmin-'g-',
For diatrizoate, the presence of BQ123 converted
(P <O.OOl, analysis of variance) the sustained decrease
in RPF into a sustained increase (23.3k1.4 to
26.6& l.Omlrnin-'g-', P(O.05, n=6), whereas the
transient increase in RPF was elevated (23.3 & 1.4 to
37.2 f3.0 ml min- ' g- ') and became markedly
prolonged.
After BQ 123 (10 pmol/l), a sustained monophasic
decrease in FRN, was still observed with diatrizoate
(88.0f2.7 to 84.1 f 1.8%, P<0.05) but not with
iotrolan (91.2f 1.2 to 90.2f lS%, not significant).
Contart media and endothelin
=e8 0 Y
8
6
E60c
..Y
:
g
f
40-
1 2 0 -
0
-13
-12
-11
-10
4
-8
4
4
-5
4
log {[ET-l](mol/l)}
H
.-z
:
also significantly reduced from 63 f5% to 37 f5%
(P<0.05).
As shown in Figs. 1-3, phosphoramidon, in
marked contrast to BQ123, had no effect on any of
the transient or sustained changes in renal function
produced by either iotrolan or diatrizoate (n= 6). In
the presence of phosphoramidon (1 mmol/l), a sustained fall in C,, was observed with either iotrolan
(0.64f0.08 to 0.47+0.05rnlmin-'g-', PcO.05) or
diatrizoate (0.63k0.07 to 0.48f0.08ml min-' g-',
P<O.O5). A sustained fall in RPF with either iotroIan (28.5k2.1 to 22.1 f 1.8mlmin-' g-', PcO.05)
or diatrizoate (26.9f 1.8 to 19.6f2.0mlmin-'g-',
P < 0.05) was also observed. After phosphoramidon,
a sustained increase in urine flow was observed with
both
iotrolan
(0.10+0.09
to
0.16f0.05
ml min-' g-', P<0.05) and diatrizoate (0.11 f
0.07 to 0.16f0.04ml-'min-'g-',
Pe0.05) with
diatrizoate still producing a large initial increase
in urine flow (Fig. 3). A significant reduction
in FRN, after either iotrolan (92.3k2.5 to 91.0k
1.8% not significant) or diatrizoate (89.6f2.0 to
84.9k 1.9%, P < 0.05) was also observed.
Studies in the anaesthetized rat
.-Y
s8
43 I
40-
-E
c
2 20-
-13
-12
-11
-10
-9
-8
-7
4
-5
-4
Fig. 4. Concentration-rrponr curves for (a) endothelin (El-I)
alone ( 0 )and the presence of BQlU ( 0 )and (b) for bigendothe
lin (big El-I) alone ( 0 )and in the presence of phosphoramidon
(0).Vertical bars indicate SEM (n =6 per group).
However, the fall in FRN, produced by both RCM
was significantly reduced (Pe 0.05, analysis of variance) after BQ123. The time course of the effects of
RCM on urine flow after BQ123 is shown in Fig. 3.
BQ123 markedly reduced the transient and the
sustained increase in urine flow observed with both
iotrolan and diatrizoate, however, after BQ123, a
small but sustained increase in urine flow was still
observed with either iotrolan (0.12k0.06 to
0.14+0.02rnlmin-'g~', Pe0.05) or diatrizoate
(0.12 f0.04 to 0.14 f0.02 ml min - ' g- ', P e 0.05).
Effect of RCM after phosphoramidon
Phosphoramidon (1 pmol/l) itself had no effect on
renal function. Its effectiveness as an ECE inhibitor
is shown in Fig. 4. Phosphoramidon produced a
rightward shift in the cumulative concentrationresponse curve to the decrease in RPF induced by
big ET-1. The maximum response to big ET-1 was
The administration of either indomethacin or its
vehicle to the anaesthetized rat had no effect on RC
but produced a transient change (e2min) in MAP
(group 1, 9.6k 1.9%, P<0.05; group 2, 5.0+1.8%,
P < 0.05; group 3, 5.3 f3.0%, not significant, and
group 4, 8.3f2.0%, Pe0.05). Neither BQ123 nor
phosphoramidon had significant effects on cardiovascular parameters. The administration of RCM
induced symptoms of respiratory distress in some
animals and occasionally some animals died of
respiratory failure either during or shortly after the
administration of diatrizoate. The number of animals dying was distributed evenly across the groups
( n = 1,3,2 and 2 in groups 1 to 4, respectively).
Diatrizoate induced initial increases in RC during
the period of administration (Table 1) in all treatment groups. The initial increase in RC was potentiated by pretreatment with BQ123. On ceasing the
administration of diatrizoate RC decreased rapidly
in groups 1 and 2 (Table 1). In the presence of
indomethacin (group 2) this decrease in RC was
maintained throughout the study period. A decrease
in RC to below baseline values after diatrizoate was
not observed in animals receiving both indomethacin and BQ123. The sustained fall in RC induced by
diatrizoate, however, remained unaffected by pretreatment with phosphoramidon (group 4).
Diatrizoate also induced initial decrease in MAP
during the period of administration (Table 2). In
group 1, MAP returned to pretreatment values on
ceasing diatrizoate administration. In groups 2 and
3, MAP remained significantly below pretreatment
values for the duration of the study. In group 4,
MAP was not significantly different from baseline
values on ceasing diatrizoate administration. No
432
S. Oldroyd et al.
Table I. Changes in RC at various time intervals after the start of the administration of diatrizoate. RC is expressed as a percentage of the baseline RC
before the administration of drugs. Statistical significance: *P<O.O5, **P <0.01 versus baseline (paired Student’s t-test); t P <0.05, t t P <0.01 versus group 2 (unpaired
Student’s t-test).
~
Baseline RC
10-3 x
(kH1lmmHg)
Group I (n = 4)
Group 2 (n = 6)
Group 3 (n = 6)
Group 4 (n=6)
13.8 f I.8
22.7 k 3.1
30.3 k 6 . 9
52.9+ 18.5
~~
~
RC (% of baseline)
Time after administration
of diatrizoate (min).
0.5
..
2.5
132.7k15.3 76.1k31.8
115.1 k1.2**121.1 k6.6*
I31.8+5.8*t 140.1 k 11.8
125.4*7.4*
139.3f9.9**
5.0
10
IS
20
2.5
66.7k19.2
84.7*6.3*
101.0+5.6
88.7k8.0
90.2*2.6*t
70.3*5.9**
94.9k4.ltt
78.9k6.F
95.2+4.5t
77.3*5.3**
96.7k4.2t
76.2+6.6*
98.3k3.4t lW.9k6.6
80.3&4.P* 82.9&5.5*
99.0k5.47 98.9k5.47
78.4k8.6
79.8k 11.2
30
101k7.0
85.7k4.6’
97.5k4.9
72.1 k 9 . l
Table 2 Changes in MAP at various time intervals after the start of the administration of diatrizoate. MAP is expressed as a percentage of the baseline MAP before
the administration of drugs. Statistical significance: *P <O.OS. **P <0.01. ***P<O.WI versus baseline (paired Student’s t-test).
MAP (% of baseline)
Baseline
(mmHg)
Group I (n = 4)
Group 2 (n = 6)
Group 3 (n = 6)
Group 4 (n=6)
+
I20 6.0
I29 If:6.0
I23 f 2.0
122k6.0
Time after administration
of diatrizoate (min). . . 0.5
96.7k4.6
101.6&2.5
95.2+ 1.9*
96.1k3.7
2.5
5.0
54.1+9.9* 90.4k4.2
66.6+7.3** 89.0+2.2**
69.7+3.4***83.9*4.6*
63.5*7.1* 95.3k4.4
marked changes in heart rate were seen in any
group.
DlSCUSSlON
We have studied the role of endothelin in the
renal responses to RCM both in uitro in the IPRK
and in uiuo in the anaesthetized rat using an ETA
receptor antagonist, BQ123, and a putative inhibitor of ECE, phosphoramidon. In the IPRK, the
initial transient increase and the sustained decrease
in Cln, an index of GFR, produced by RCM were
entirely inhibited by BQ123, suggesting that both
responses involve endothelin acting via the ETA
receptor. Mechanisms associated with a reduction in
GFR mediated by endothelin include an increase in
afferent arteriolar resistance, reducing glomerular
capillary hydrostatic pressure or an increase in
mesangial cell contractility [ZO], reducing the surface area for filtration. The involvement of endothelin in the initial transient increase in GFR, however,
is less readily explained since in numerous studies
only a decrease in GFR after endothelin administration has been observed (for a review, see [S]). The
possible involvement of secondary mediators,
including nitric oxide [21] and prostaglandins [22],
has been proposed, generating either an increase in
the glomerular ultrafiltration coeflicient or an
increase in glomerular hydrostatic pressure
mediated by a differential increase in efferent rather
than afferent arteriolar resistance. In isolated rat
renal arterioles, endothelin is a selective efferent
vasoconstrictor [23].
In uitro, in the IPRK both iotrolan (an isoosmolar RCM) and diatrizoate (a high-osmolar
RCM) produced a sustained decrease in RPF indicative of sustained vasoconstriction, which was
10
15
99.9k7.1
94.1k2.8
84.5+5.8*
95.6k7.6
102.4f5.3 102.0+5.2
94.3k2.5 95.8+1.7*
82.9+5.3* 82.5k4.5**
89.4k8.7 86.0k10.0
20
25
30
98.6k4.8
94.8+1.9*
86.8f4.0*
82.4kII.I
98.0k4.4
94.0*1.2**
89.3k5.8
79.9k12.4
markedly inhibited by BQ123, suggesting an
ETA-mediated effect independent of contrast osmolality. Endothelin release from endothelial cell
culture stimulated by contrast media is also independent of contrast osmolality [ S , 241. BQ123
(10 pmol/l) produced a parallel rightward shift in the
cumulative concentration-response curve to ET- 1,
suggesting that functional antagonism of the ETA
receptor had been obtained. The potency of BQ123
was similar to that reported for other ETA receptormediated responses [lS]. In contrast to the transient
increase in GFR, the initial transient increase in
RPF observed with diatrizoate was not produced by
iotrolan. Renal vasodilatation induced by diatrizoate is probably an osmotic phenomenon independent of renal endothelin. Its potentiation by BQ123
can be explained by the removal of ETA-mediated
vasoconstriction: physiological antagonism.
In the anaesthetized rat, diatrizoate produced a
transient increase followed by a sustained fall in RC.
RCM have been shown to produce vasodilatation in
other vascular beds, such as the dog hind limb [25].
This may explain the marked decrease in blood
pressure observed during RCM administration,
which may be related to RCM osmolality. As in the
IPRK, BQ123 attenuated the decrease in RC
induced by diatrizoate, suggesting the involvement
of ETA-induced vasoconstriction while the transient
vasodilatation was potentiated. In vivo, BQ123 was
used at a dose which attenuated the pressor
response to ET-1 in the pithed rat [26]. However, it
has previously been reported that exogenous ET-1
administered to anaesthetized rats induced renal
vasoconstriction which was not attenuated by
BQ123, suggesting the involvement of an ET receptor other than ETA [27]. The results obtained in the
current study in uiuo may indicate differences in the
Contrast media and endothelin
mechanism of action of endogenously produced and
exogenously administered endothelin. Our results
also suggest that BQ123 is capable of attenuating
the increase in RPF induced by either endogenously
produced or exogenously applied endothelin when
studied in uitro. The renal response to another highosmolar contrast agent, iothalamate, was recently
reported to be inhibited by an endothelin antagonist
in the anaesthetized rat in uiuo [28].
In the present experiments phosphoramidon, an
ECE inhibitor, was employed to study the involvement of endothelin synthesis in the renal response
to RCM. Pretreatment with phosphoramidon had
no effect on the renal response to RCM in
uiuo at concentrations previously demonstrated to
inhibit the response to big ET-1 in the pithed rat
[24], and in uitro at 1 mmol/l, which reduced the
maximal vasoconstrictor effect of big ET-1 by
approximately 50%. This raises the possibility that
RCM may stimulate the release of preformed endothelin from storage vesicles, as identified in the
pituitary [29], rather than activate the constitutive
pathway for endothelin synthesis. However, the ability of phosphoramidon to inhibit the renal response
to big ET-1, but not to RCM, could also reflect
either differences between ECEs responsible for the
conversion of endogenous and exogenous precursor
peptides or the poor penetration of phosphoramidon to an intracellular target site (for a review, see
~301).
Contrast media have previously been demonstrated to increase endothelin release into the incubating solutions of endothelial cells in culture derived from bovine aorta [5] and human umbilical
vein [24] by a mechanism independent of contrast
osmolality. Renal endothelin release by RCM is
unlikely to be a consequence of generalized osmotic
damage to the vascular endothelium [31, 321, since,
in the IPRK, RCM do not produce ultrastructural
damage to the vascular endothelium or inhibit the
vascular response to acetylcholine, an endotheliumdependent vasodilator (S. Oldroyd, unpublished
work). Stimulation of the constitutive pathway for
endothelin synthesis would also appear unlikely due
to the rapid onset of RCM-induced renal vasoconstriction both in uiuo and in uitro. The ability of
RCM to penetrate capillary endothelial cells, however, has been demonstrated in the rat kidney using
X-ray microanalysis [33]. The hydrophilic nature of
RCM, which may control their cellular penetration,
has recently been suggested as an important determinant of RCM-induced endothelin release [34].
In the present IPRK experiments, both the transient and sustained increases in urine flow induced
by RCM were also inhibited by BQ123. While the
acute increase in urine flow was probably a function
of the GFR, the smaller sustained effect was likely
to be secondary to the fall in FRN,. The fall in FRN,
induced by both RCM was also inhibited by
BQ 123. Tubular sites for endothelin synthesis have
been located histochemically [35] and a direct
433
natriuretic effect of endothelin has been described
[36]. The involvement of tubular events in the renal
vascular properties of RCM mediated through an
effect of endothelin on the tubuloglomerulo feedback system also remains a possibility [37].
Finally, evidence is now accumulating for a role
of endothelin in other forms of renal failure [38],
including the fall in renal function after cyclosporin
[39] and renal ischaemia [MI. Since some of the
renal effects of RCM are mediated by endothelin, a
role for endothelin in the nephrotoxicity induced by
RCM may also be proposed. Such a hypothesis is
supported by the elevated plasma levels of endothelin found in patients suffering from RCM nephrotoxicity compared with control subjects receiving
the same dose but with normal renal function [8].
ACKNOWLEDGMENTS
We thank Trent Regional Health Authority, the
Northern General Hospital NHS Trust, Shefield,
the Kodak Trust of the Royal College of
Radiologists and Schering AG, Berlin, for their
financial support.
REFERENCES
I. Schwab SJ, Hlatky MA, Pieper KS, et al. Contrast nephrotoxicity: a
randomised controlled trial of a nowionic and an ionic radiographic contrast
149-53.
agent. New Engl J Med 1989;
2. Russell SB, S h e r w d T. Monomer/dimer contrast media in the renal
circulation: experimental angiography. Br ] Radiol 1974; 47: 268-71.
3. Lanon TS, Hudson K, Mertz Jl,Romero JC, Knox FG. Renal vasoconstrictive
response t o contrast medium. The role of d i u m balance and the
renin-angiotensin system. J Lab Clin Med 1983; 101: 385-91.
4. Arend L]. Bakris GL, Burnett JC, Megerian C, Spielman WS. Role for
intrarenal adenosine in the renal hernodynamic response t o contrast media. J
Lab Clin Med 1987; 110: 406-11.
5. Porter GA. Experimental contrast-associated nephropathy and its clinical
implications. Am J Cardiol 1990; 6h lS-22F.
6. Heyman SN, Clark BA. Kaiser N, et PI. Radiocontrast agents induce
endothelin release in vivo and in vitro. J Am Soc Nephrol 1992; 3: SMS.
7. Margulies KB. Hildebrand FL, Heublein FM, Burnett JC. Radiocontrast
increases plasma and urinary endothelin. J Am Soc Nephrol 1992; 2: 1041-5.
8. Margulies KB, McKinley LJ, Burnett JC. Endothelin in human and canine
radiocontnst-induced nephropathy. J Vasc Res 1992; 29. 1634.
9. Simonron MS. Endothelins: multifunctional renal peptides. Physiol Rev 1993;
73: 37-11,
10. Yanagisawa M. Kurihara H, Kimura S, et al. A novel vasoconstrictor peptide
produced by endothelial cells. Nature (London) 1388, u1: 41 1-15,
I I. Sakurai T, Yanagisawa M, Takuwa Y, et al. Cloning of a cDNA encoding a
non-isopeptide-selective subtype of the endothelin receptor. Nature (London)
1990;
732-5.
12. Brown P, Haylor J, Morcos SK, Cope GH, El Nahas AM. Effect of diatrizoate
on the function of the isolated perfused rat kidney. Br J Radiol 1992; Is:
1011-17.
13. Brezis M. Greenfeld Z, Herman M, Meyer JJ. Heyman SN, Rosen S.
Experimental nephrotoxicity d the ndiocontnst agents iohexol, ioxaglate and
iothalamate. An in v i m and in vivo study. Invest Radiol 1991: U: 32.5-31.
14. Brown PWG. Haylor ]L, El Nahas AM, Morcos SK. The functional effects of
contrast media on the isolated perfused rat kidney. Contrib Nephrol 1993;
101: 23s-40.
a
434
S. Oldroyd et al.
IS. Ihara M. Noguchi K, Saeki T, et al. Biological profile of highly potent novel
endothelin receptor antagonist selective for the ETA receptor. Life Sci 1992;
W 247.
16. Fukuroda T, Noguchi K. Tsuchida S, et al. Inhibition of biological actions of
big endothelin-l by phorphoramidon. Biochem Biophys Res Commun 1990;
In390-5.
17. Parvez Z, Marsan RE. Moncada R, Patel N. Effect of contrast media on
prortaglandin synthesis in vivo. Invest Radiol 1988; U (Suppl.): 517881.
18. De Maeyer. Deckmyn H, Arnout 1, Baert AL. Vermylen J. Intravenous ionic
contrast media cause local prortacyclin release in man. Invest Radiol 1985; 20:
871-73.
19. El Sayed AA, Haylor J. El Nahas AM. Differential effects of amino acids on the
isolated perfused rat kidney. Clin Sci 1990; Ip: 381-6.
20. Simonsen MS, Dunn MJ. Endothelin-l stimulates contraction of rat glomerular
mesangial cells and potentiates Padrenergicmediated cyclic adenorine
monophosphate accumulation. J Clin Invest 1990; & 790-7.
21. Schramek H, Willinger CC, Gstraunthaler G, Phller W. Endothelin-3
modulates glomerular filtration rate in the isolated perfused rat kidney. Renal
Physiol Biochem 1992; IS:325-33.
22. DOrlansjuste P. Yano M, Telemaque S. ETAdependent pressor effects and
release of prortacyclin induced by endothelin-l in pulmonary and renal
vasculature. 3rd Int Conf Endothelin 1993; 26.
23. Lanese DM, Yuan BH, McMurtry IF, Conger ID. Comparative sensitivities of
isolated rat renal arterioles t o endothelin. Am J Physiol 1992; 26& F89C9.
24. Akagi M. Masaki S, Kituumi K, Mio M, Tasaka K. Comparative study of the
adverse effects of various radiographic contrast media, including ioversol. a
new low-osmolarity medium. (I. The complement system and endothelial cells.
Methods Findings Exp Clin Pharmacol 1991; 13: 449-54.
25. Steiner RM, Grainger RG, Memon N. The effect of contrast media on the
peripheral arterial blood flow of the dog. Clin Radiol 1980; 31: 621-7.
26. Slee S-J, H e y CA, Allen JL, Wilson C. Comparison of pressor responses to
endothelin-l and big endothelin-l in the pithed rat. Br J Pharmacol 1992; 107:
413P.
27. Pollock DM, Opgenorth TJ. Evidence for endothelin induced renal
vasoconstriction independent of ETA receptor activiation. Am J Physiol 1993;
2b& RZZZ-6.
28. Cantley LG. Spokes K. Clark 8. Kulik A, Epstein FH. An endothelin receptor
antagonist decreases the effect of iothalamate on renal blood flow. J Am Soc
Nephrol 1992; 3: 434.
29. Yorhirawa TO, Shinmi A, Giaid M, et al. Endothelin: a novel peptide in the
posterior pituitary system. Science (Washington, DC) UI: 462-4.
30. Opgenorth TJ, Wu-Wong JR. Shioraki K. Endotheliwconverting enzymes.
FASEB J 1992; t 2653-9.
31. Laerum F. Acute damage t o human endothelial cells by brief exposure t o
contrast media in vitro. Radiology 1993: 14: 681-4.
32. Nyman U. Almen T. Effects of contrast media on aortic endothelium. Acta
Radiologica 1980, 362 (Suppl.): 65-71.
33. Nordby A, Tvedt KE, Halgunset J, Haugen OA. Intracellular penetration and
accumulation of radiographic contrast media in the rat kidney. Scanning
Microrc 19% 4: 651-66.
34. Heyman SN. Clark BA. Cantley L. et al. Effects of ioversol versus iothalmate
on endothelin release and radiocontrast nephropathy. Invest Radiol 1993; Z&
313-18.
35. Wilkes BM, Susin M, Mento PF, et d. Localization of endothelin-like
immunoreactivity in rat kidneys. Am J Physiol 1991; Z& F913-20.
36. Ferrario RG, Foulkes R, Salvati P, Patron0 C. Hemadynamic and tubular
effects of endothelin and thromboxane in the isolated perfused rat kidney. Eur
J Pharmacol 1989; 171: 127-34.
37. Takabatake T, lse T. Ohta K, Kobayashi K. Endothelin effects on renal
function and tubuloglomerular feedback. Kidney Int 1991; 39 (Suppl. 32):
s 122-4.
38. Perico N, Remuni G. Role of endothelin in glomerular injury. Kidney Int
1993; 43 (Suppl. 39): 57680.
39. Fog0 A. Hellings SE, lnagami T, Kon V. Endothelin receptor antagonism is
protective in in vivo acute cyclorporine toxicity. Kidney Int 1992; TZ: 770-4.
40. Mino N, Kobayashi M, Nakajima A, et al. Protective effects of a selective
endothelin receptor antagonist, BQIU, in ischaemic acute renal failure in
rats. Eur J Pharmacol 1992; 221: 77-83.