Am J Physiol Renal Physiol 292: F1094 –F1101, 2007.
First published December 5, 2006; doi:10.1152/ajprenal.00351.2005.
Direct vasoconstrictor effect of prostaglandin E2 on renal
interlobular arteries: role of the EP3 receptor
William F. van Rodijnen,1 Iolente J. Korstjens,1
Natalee Legerstee,1 Piet M. ter Wee,2 and Geert-Jan Tangelder1
1
Laboratory for Physiology and 2Department of Nephrology, Institute for Cardiovascular
Research, VU University Medical Center, Amsterdam, The Netherlands
Submitted 26 August 2005; accepted in final form 27 November 2006
van Rodijnen WF, Korstjens IJ, Legerstee N, ter Wee PM,
Tangelder GJ. Direct vasoconstrictor effect of prostaglandin E2 on
renal interlobular arteries: role of the EP3 receptor. Am J Physiol
Renal Physiol 292: F1094 –F1101, 2007. First published December 5,
2006; doi:10.1152/ajprenal.00351.2005.—Evidence indicates that
prostaglandin E2 (PGE2) preferentially affects preglomerular renal
vessels. However, whether this is limited to small-caliber arterioles or
whether larger vessels farther upstream also respond to PGE2 is
currently unclear. In the present study, we first investigated the effects
of PGE2 along the preglomerular vascular tree and subsequently
focused on proximal interlobular arteries (ILAs). Proximal ILAs in
hydronephrotic rat kidneys as well as isolated vessels from normal
kidneys constricted in response to PGE2, both under basal conditions
and after the induction of vascular tone. By contrast, smaller vessels,
i.e., distal ILAs and afferent arterioles, exhibited PGE2-induced va-
sodilation. Endothelium removal and pretreatment of single, isolated
proximal ILAs with an EP1 receptor blocker (SC51322, 1 ␮mol/l) or
a thromboxane A2 receptor blocker (SQ29548, 1 ␮mol/l) did not
prevent vasoconstriction to PGE2. Furthermore, in the presence of
SC51322, responses of these vessels to PGE2 and the EP1/EP3 agonist
sulprostone were superimposable, indicating that PGE2-induced va-
soconstriction is mediated by EP3 receptors on smooth muscle cells.
Immunohistochemical staining of proximal ILAs confirmed the pres-
ence of EP3 receptor protein on these cells and the endothelium.
Adding PGE2 to normal isolated kidneys induced a biphasic flow
response, i.e., an initial flow increase at PGE2 concentrations ⱕ0.1
␮mol/l followed by a flow decrease at 1 ␮mol/l PGE2. Thus our
results demonstrate that PGE2 affects multiple segments of the pre-
glomerular vascular tree in a different way. At the level of the
proximal ILAs, PGE2 had a direct vasoconstrictor action mediated by
EP3 receptors.
interlobular arteries; vasoconstriction
PGE2 IS ONE OF the major cyclooxygenase-derived products
produced by the kidney (17). The actions of PGE2 intrarenally
include both tubular and vascular effects (2). The latter are
important for regulating renal function under various circum-
stances. So far, the effects of PGE2 on smaller pre- and
postglomerular arterioles have mainly received attention. The
postglomerular efferent arterioles were found to be insensitive
to PGE2 (4, 23), while for the preglomerular afferent arterioles
(AAs) and distal interlobular arteries (ILAs) most studies
indicated vasodilation (4, 8, 23), although vasoconstriction in
certain instances has been reported as well (9, 23). Whether
larger vessels further upstream of the glomerulus are also
sensitive to PGE2 is currently unknown. Previous studies
Address for reprint requests and other correspondence: G.-J. Tangelder,
Laboratory for Physiology, Institute for Cardiovascular Research, VU
Univ. Medical Center, van der Boechorststraat 7, 1081 BT Amsterdam, The
Netherlands (e-mail: geertjan.tangelder@vumc.nl).
F1094 0363-6127/07 $8.00 Copyright © 2007 the American Physiological Society
Downloaded from www.physiology.org/journal/ajprenal (027.060.215.086) on January 27, 2020.
have demonstrated that larger preglomerular vessels con-
strict in response to inflammatory mediators, such as sero-
tonin and leukotrienes (5, 19), suggesting a similar mode of
action for PGE2.
In general, the actions of PGE2 are mediated by a family of four
G protein-coupled receptors, designated EP1–EP4 (3, 16). Of
these receptors, EP2 and EP4 have been shown to stimulate
intracellular processes leading to smooth muscle cell relaxation
and, hence, vasodilation. EP1 and EP3 receptors, on the other
hand, stimulate processes leading to vasoconstriction. For exam-
ple, in cell cultures, EP1 receptor activation has been shown to
increase intracellular Ca2⫹ concentration (26). In addition, EP3
receptors have been found to inhibit intracellular cAMP produc-
tion (21, 22) and are thereby able to reduce PKA activity.
The purpose of the present study was to determine the
effects of PGE2 on intermediate and proximal ILAs. Initially,
experiments were performed using the isolated, perfused hy-
dronephrotic rat kidney model. In this preparation, nearly the
entire microvasculature can be visualized, allowing a direct
comparison of responses of different vessel segments. Using
this model, we found that, in contrast to distal ILAs and AAs,
larger preglomerular arterioles manifested a direct vasocon-
strictor response to PGE2. Thereupon, we used normal kidneys
to determine in isolated proximal ILAs the receptor subtype
involved, both pharmacologically and immunohistochemi-
cally. In addition, we assessed the effects of PGE2 on renal
hemodynamics.
MATERIALS AND METHODS
All experiments were performed using 12- to 15-wk-old male
Sprague-Dawley rats (Harlan, Horst, The Netherlands). The animals
were housed in macrolon cages with free access to tap water and a
standard rat chow (Hope Farm, Woerden, The Netherlands). All their
handling was reviewed and approved by the Institutional Animal Care
and Use Committee.
Drugs. PGE2 was obtained from Calbiochem. The EP1 receptor
blocker SC51322, the thromboxane A2 (TXA2) receptor antagonist
SQ29548, and the EP1/EP3 agonist sulprostone were obtained from
Biomol and Cayman Chemical, respectively. ANG II, norepinephrine,
and acetylcholine were purchased from Sigma.
Isolated, perfused hydronephrotic and normal kidneys. To obtain
hydronephrotic kidneys, rats were anesthetized with isoflurane (3% in
O2, 0.7 l/min and N2O, 1.2 l/min, Abbott Laboratory, Queensborough,
Kent, UK) at a younger age, i.e., ⬃8 wk before the in vitro experi-
ments. Subsequently, the left ureter was located and tied off with a
suture. Ureter ligation is known to induce almost complete tubular
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.
http://www.ajprenal.org
 PGE2-INDUCED VASOCONSTRICTION OF PROXIMAL ILAS
atrophy (20), allowing direct microscopic visualization of nearly the
entire renal microvascular bed.
For in vitro perfusion, both hydronephrotic and normal kidneys
were isolated in a similar manner (see Ref. 25 for an extensive
description). Briefly, the rat was anesthetized with a combination of
pentobarbital sodium (50 mg/kg ip, Sanofi Sante, Maassluis, The
Netherlands) and ketamine (25 mg/kg im, Kombivet, Etten-Leur, The
Netherlands). The kidney was exposed by opening the abdominal
cavity, and the renal artery was cannulated via the abdominal aorta.
Perfusion was started in situ with preheated DMEM (Sigma) supple-
mented with (in mmol/l) 23.8 NaHCO3, 5.6 HEPES, 5.5 D-glucose,
and 1 sodium pyruvate (all Sigma). DMEM was equilibrated with
95% air-5% CO2 at 37°C, resulting in a pH of 7.4. Then, the kidney
was excised and moved to a heated chamber on the stage of an
inverted microscope (Axiovert 100, Zeiss) without disruption of renal
flow. For all studies, a single-pass perfusion was employed. The
perfusion apparatus consisted of a small pressurized reservoir (⬃25
ml) connected to the renal arterial cannula, that was filled on demand
from a larger preheated and oxygenated reservoir (⬃600 ml). Agents
were added in a cumulative manner to this larger reservoir. Both
hydronephrotic and normal kidneys were perfused at a constant
pressure of 80 mmHg, monitored at the level of the renal artery.
Pressure changes were eliminated by adjusting the outflow of gas
(95% air-5% CO2) from the small reservoir. In normal kidneys, renal
perfusate flow was measured using an electromagnetic flow probe
(TS410, Transonic Systems) mounted in the perfusion line proximal
to the kidney.
To visualize different vessel segments in hydronephrotic kidneys, a
small hole was made in the cortex through which a fiber optic probe
was inserted, transilluminating a portion of the membranous cortex.
Vessel images were generated using a ⫻40 objective lens (numerical
aperture 0.6, Zeiss) and a CCD camera (7020/20, Philips, Eindhoven,
The Netherlands). The images were recorded on a VCR for offline
analysis using a custom designed vessel wall tracking system (12).
Changes in afferent arteriolar diameters were analyzed just after
branching from ILAs. ILAs were divided into four different groups
based on their location and basal diameters: distal (connected to AAs),
intermediate (30 –50 and 50 –70 ␮m), and proximal (⬎70 ␮m). ILA
diameters were measured near their midpoint.
Isolated cannulated proximal interlobular arteries. Proximal ILAs
were isolated from normal dissected kidneys and kept in ice-cold
dissection DMEM in which NaHCO3 was lowered to 4.2 mmol/l by
replacing NaHCO3 with NaCl. This medium was equilibrated with
atmospheric CO2, and pH was set at 7.4 using NaOH. The isolated
ILAs were mounted with sutures between two glass micropipettes in
a water-jacketed chamber of a pressure myograph. The chamber was
Fig. 1. Effects of PGE2 on successive segments of the preglo-
merular vascular tree in isolated, perfused hydronephrotic rat
kidneys. A: under basal conditions, PGE2 elicited vasoconstric-
tion of interlobular arteries (ILAs) of ⬎30 ␮m. B: in contrast,
during treatment with the vasoconstrictor ANG II (0.1 nmol/l),
PGE2 responses displayed a gradual change from vasodilation
at the small-caliber sites to again vasoconstriction in the largest
ILAs, with a biphasic response in between; n ⫽ 5– 6. *P ⬍ 0.05
vs. basal condition.
AJP-Renal Physiol • VOL
Downloaded from www.physiology.org/journal/ajprenal (027.060.215.086) on January 27, 2020.
F1095
moved to the stage of a microscope, filled with dissection DMEM, and
sealed with a glass cover. One of the micropipettes was connected to
a pressure column, used to gradually pressurize the ILA to 75 mmHg.
The other pipette was clamped off and connected to a micrometer to
stretch the vessel to its in vivo length. The time period from dissection
to mounting the vessel did not exceed 2 h. Superfusion with normal
DMEM was subsequently started, and the temperature was raised to
37°C. Superfusion rate was set at ⬃2 ml/min, with refreshing of the
myograph chamber volume every minute. Agents were added directly
to the superfusion medium in a cumulative manner. Changes in vessel
diameter were measured online using a custom designed measurement
system. Only those vessels exhibiting a clear endothelium-dependent
vasodilator response to 0.3 ␮mol/l acetylcholine (⬎50% reversal of
PGE2-induced vasoconstriction) were used for determining mean
responses. In one series of experiments, the endothelium was removed
deliberately by perfusing the vessels with a bubble of air. When ILAs
were subjected to this treatment, acetylcholine-induced vasodilatation
was completely absent. To exclude the possibility that PGE2-induced
vasoconstriction of ILAs was mediated by TXA2 receptors, isolated
ILAs were treated in another series of experiments with the specific
TXA2 receptor blocker SQ29548 (1 ␮mol/l, 10 min) after which PGE2
concentration-response curves were constructed.
Equilibration and elimination of endogenous prostanoids. All in
vitro preparations described above were allowed to equilibrate for at
least 30 min before experiments began. All were pretreated with 10
␮mol/l ibuprofen (Sigma) to eliminate the influence of endogenous
prostanoids. Sufficient cyclooxygenase (COX)-2 inhibition by this
ibuprofen concentration is indicated by our observation that treatment
with a selective COX-2 inhibitor (NS-398, 10 ␮mol/l, Sigma) in
hydronephrotic kidneys did not change basal diameters of pre- or
postglomerular vessels. Changes in vessel diameter or perfusion flow
caused by a drug were evaluated during the plateau of the response,
usually ⬃10 min after its addition.
Immunohistochemical staining of EP3 and EP1 receptors. Cannu-
lated proximal ILAs were fixated with 2% ice-cold paraformaldehyde
in dissection DMEM, embedded in 15% gelatin, and frozen in liquid
nitrogen. Cut 10-␮m sections were permeabilized with 0.3% Triton
X-100, blocked with 10% goat serum in PBS [1 h, at room temper-
ature (RT)], and incubated with a rabbit polyclonal antibody against
EP3 or EP1 receptors (1:500 in PBS containing 3% goat serum, 48 h,
4°C, both Cayman Chemical). In the negative background control, the
primary antibody was omitted from the incubation medium. The
sections of both types were then washed (PBS, 3 ⫻ 10 min, RT) and
incubated with the secondary antibody consisting of 1:100 diluted
Alexa Fluor 488-conjugated donkey anti-rabbit (1 h, RT, Molecular
Probes). This fluorochrome allows short illumination times, thereby
292 • MARCH 2007 • www.ajprenal.org
F1096 PGE2-INDUCED VASOCONSTRICTION OF PROXIMAL ILAS
precluding autofluorescence. After a final wash (3 ⫻ 30 min, RT),
sections were mounted in medium (Vectashield from Vector) contain-
ing 4⬘,6-diamino-2-phenylindole (DAPI) as a nuclear counterstain. To
distinguish smooth muscle cells from endothelial cells, F-actin was
labeled in additional sections using Cy3-conjungated-rhodamine phal-
loidin (1:60, Molecular Probes), incubated together with the second-
ary antibody. The vascular localization of EP3 and EP1 receptor
protein was studied using an inverted fluorescence microscope (Ax-
iovert 200 Marianas, Zeiss). Images were generated with a ⫻10 air
and ⫻40 oil-immersion objective (numerical aperture 0.50 and 1.30,
respectively; Zeiss) and recorded using a cooled CCD camera
(1,280 ⫻ 1,024 pixels, Cooke Sensicam, Cooke, Tonawanda, NY).
This microscope and camera, as well as the data viewing and pro-
cessing, including deconvolution, were conducted and controlled by
Slidebook software (Slidebook version 4.0, Intelligent Images Inno-
vations, Denver, CO).
Statistical analyses. All values are presented as means ⫾ SE; n
refers to the number of animals studied. Statistical analyses were
performed using Prism 4 (GraphPad Software, San Diego, CA).
Differences within concentration-response curves were assessed using
one-way ANOVA for repeated measurements followed by a New-
man-Keuls post hoc test. P ⬍ 0.05 was considered statistically
significant. Differences between groups were assessed using ANOVA
followed by Student’s t-test. For multiple comparisons, the Bonferroni
correction was applied.
RESULTS
Effects of PGE2 on different segments of the preglomerular
vascular tree in isolated hydronephrotic rat kidneys. As shown
in Fig. 1, marked differences in the response to PGE2 were
seen along the preglomerular vascular tree. Under basal con-
ditions (Fig. 1A), proximal and intermediate ILAs (see Table 1
for basal values) manifested a concentration-dependent con-
striction to PGE2, which appeared to increase in magnitude
with their diameter. Thus, at the highest PGE2 concentra-
tion, i.e., 1 ␮mol/l, mean diameters were reduced by 15.4 ⫾
3.7% in proximal ILAs (⬎70 ␮m, n ⫽ 5) and by 12.0 ⫾ 3.4
and 8.0 ⫾ 3.3% in intermediate ILAs (50 –70 and 30 –50
␮m, respectively; n ⫽ 6 and 5). By contrast, the smallest
caliber ILAs (i.e., distal) and AAs were hardly or not
affected under basal conditions by increasing concentrations
of PGE2 (Fig. 1A).
The vascular bed of isolated hydronephrotic rat kidneys is
almost maximally vasodilated. Thus PGE2 effects were also
AJP-Renal Physiol • VOL
Downloaded from www.physiology.org/journal/ajprenal (027.060.215.086) on January 27, 2020.
Table 1. Diameters of successive segments of the
preglomerular vascular tree in isolated hydronephrotic rat
kidneys under basal conditions and after treatment with
PGE2 (1 ␮mol/l) or ANG II (0.1 nmol/l)
␮m Basal ⫹PGE2 After Wash ⫹ANG II
Proximal ILAs ⬎70 95.5⫾5.4 80.5⫾4.2* 89.0⫾5.1 90.0⫾5.5*
Intermediate ILAs 50–70 59.1⫾2.0 52.0⫾2.3* 59.6⫾2.1 54.5⫾2.8*
30–50 36.4⫾1.7 33.1⫾1.5* 34.8⫾2.2 27.8⫾3.2*
Distal ILAs 24.8⫾0.8 23.9⫾0.9* 24.3⫾1.0 11.6⫾1.5*
AAs 20.3⫾0.9 20.5⫾0.9* 21.0⫾0.6 10.8⫾0.6*
Values are means ⫾ SE; n ⫽ 4 – 6. ILAs, interlobular arteries; AAs, afferent
arterioles. *P ⬍ 0.05 vs. preagonist values.
assessed after the induction of vascular tone with ANG II to
observe possible vasodilator responses. As shown in Fig. 1B,
ANG II-induced vasoconstriction was inversely related to the
basal ILA diameters. Proximal ILAs (n ⫽ 5) failed to respond
to ANG II, while subsequent segments (see Table 1) were
reduced in diameter by 7.8 ⫾ 6.2 (n ⫽ 6), 21.2 ⫾ 6.3 (n ⫽ 6),
and 51.1 ⫾ 7.8% (n ⫽ 6), respectively. AAs exhibited a similar
level of vasoconstriction to ANG II to distal ILAs (48.0 ⫾
3.9%, n ⫽ 4). In all vessels exhibiting tone, PGE2 completely
reversed at lower concentrations (ⱕ10 nmol/l) the ANG II-
induced vasoconstriction. In distal ILAs and AAs, higher PGE2
concentrations (⬎10 nmol/l) had no further effect, with mean
vessel diameters remaining near basal values (24.2 ⫾ 1.2 and
20.7 ⫾ 0.3 ␮m at 1 ␮mol/l PGE2, respectively). In intermedi-
ate ILAs, however, the diameter increase at lower PGE2
concentrations was followed by a subsequent diameter de-
crease. At the level of the proximal ILAs, no vasodilation to
PGE2 was observed. Instead, also in the presence of ANG II,
PGE2 decreased their diameters in a concentration-dependent
manner similar to that observed under basal conditions (to
79.3 ⫾ 4.5 ␮m at 1 ␮mol/l).
Effects of PGE2 on proximal ILAs obtained from normal rat
kidneys. Because in the experiments reported above proximal
ILAs exhibited the largest vasoconstrictor response, proximal
ILAs isolated from normal kidneys were studied as well. As
shown in Fig. 2A, these isolated ILAs also displayed a con-
centration-dependent diameter reduction in response to PGE2.
Significant vasoconstriction occurred from 10 nmol/l PGE2
Fig. 2. PGE2-induced vasoconstriction of proximal ILAs ob-
tained from normal kidneys. A: note enhanced response of
endothelium-denuded vessels (solid symbols). B: furthermore,
also after the induction of vascular tone using norepinephrine
(NE; 30 –70 nmol/l), proximal ILAs only displayed concentration-
dependent vasoconstriction; n ⫽ 4 –7. EC, endothelium. *P ⬍
0.05 vs. basal condition and NE-induced tone, respectively.
#P ⬍ 0.05 vs. control vessels.
292 • MARCH 2007 • www.ajprenal.org
 PGE2-INDUCED VASOCONSTRICTION OF PROXIMAL ILAS F1097

Fig. 3. Failure of the selective EP1 receptor antagonist

SC512322 (1 ␮mol/l) to prevent PGE2-induced vasoconstric-
tion of isolated proximal ILAs from normal kidneys. Shown are
absolute data (A) and %change from pre-PGE2 values (B) for
both the blocked and control condition; n ⫽ 7. *P ⬍ 0.05 vs.
pre-PGE2 value. #P ⬍ 0.05 vs. PGE2 alone.

onward (14.8 ⫾ 3.0%), up to 34.9 ⫾ 2.6% at the highest
concentration of PGE2 (1 ␮mol/l). Removal of the endothelium
did not prevent but enhanced the vasoconstrictor response of
isolated ILAs to PGE2 (Fig. 2A). At 1 ␮mol/l PGE2, diameters
of endothelium-denuded ILAs were decreased by 62.0 ⫾ 6.9%
(n ⫽ 4). Also after induction of a moderate amount of
vascular tone (⬃15%) using norepinephrine, isolated ILAs
only displayed PGE2-induced vasoconstriction (Fig. 2B).
Thus, in the presence of norepinephrine, PGE2 at concen-
trations of 0.01, 0.1, and 1 ␮mol/l decreased mean diameters
of ILAs by 18.4 ⫾ 3.6, 33.3 ⫾ 5.6, and 45.0 ⫾ 5.7% (n ⫽ 6),
respectively.

Fig. 4. Vasoconstrictor effects of the EP1/EP3 receptor agonist

sulprostone on isolated proximal ILAs from normal kidneys.
Shown are absolute data under control conditions (n ⫽ 7, A)
and after pretreatment with the selective EP1 receptor antago-
nist SC512322 (1 ␮mol/l, n ⫽ 6, B), respectively. C: compar-
ison of %change from presulprostone values. D: comparison of
effects of sulprostone and PGE2 during EP1 receptor blockade.
Note that the concentration-response curves for both agonists
are superimposable, indicating that PGE2-induced vasocon-
striction is mediated by EP3 receptor activation. *P ⬍ 0.05 vs.
presulprostone values.

AJP-Renal Physiol • VOL 292 • MARCH 2007 • www.ajprenal.org

Downloaded from www.physiology.org/journal/ajprenal (027.060.215.086) on January 27, 2020.
F1098 PGE2-INDUCED VASOCONSTRICTION OF PROXIMAL ILAS
Fig. 5. Failure of selective thromboxane A2 (TXA2) receptor antagonist
SQ29548 (1 ␮mol/l) to prevent PGE2-induced vasoconstriction of isolated
proximal ILAs from normal kidneys; n ⫽ 5. *P ⬍ 0.05 vs. pre-PGE2
values.
AJP-Renal Physiol • VOL 292 • MARCH 2007 •
Downloaded from www.physiology.org/journal/ajprenal (027.060.215.086) on January 27, 2020.
Characterization of the receptor subtype mediating vaso-
constriction of ILAs to PGE2. Initially, the role of the EP1
receptor in the vasoconstriction of ILAs to PGE2 was assessed.
As shown in Fig. 3A, a selective EP1 blocker had no effect on
basal diameters of isolated proximal ILAs (n ⫽ 7). Further-
more, PGE2-induced vasoconstriction of isolated ILAs was not
inhibited. Instead, the blocker significantly increased their
reactivity to PGE2, as depicted in Fig. 3B. During EP1 receptor
blockade, the PGE2-induced vasoconstriction of ILAs in-
creased from 14.8 ⫾ 3.0 to 23.8 ⫾ 1.9% (P ⫽ 0.03 vs. PGE2
alone), from 26.0 ⫾ 2.4 to 38.3 ⫾ 4.6% (P ⫽ 0.05), and from
34.9 ⫾ 2.6 to 46.1 ⫾ 4.1% (P ⫽ 0.04) at 0.01, 0.1, and 1
␮mol/l PGE2, respectively.
Because a commercially available antagonist for the other
constrictive PGE2 receptor, i.e., EP3, is lacking, its potential
role in ILA vasoconstriction was investigated using the EP1/
EP3 agonist sulprostone. As shown in Fig. 4A, sulprostone
reduced diameters of isolated ILAs in a concentration-depen-
dent manner. The highest concentration of sulprostone, i.e., 1
␮mol/l, reduced diameters by 42.1 ⫾ 5.1% (to 49.0 ⫾ 3.8 ␮m,
n ⫽ 7). Pretreatment with the selective EP1 antagonist did not
affect sulprostone-induced vasoconstriction (Fig. 4, B and C,
n ⫽ 6). Moreover, when the potencies of sulprostone and PGE2
Fig. 6. Presence of EP3 receptor protein in
vascular smooth muscle cells and endothelial
cells of isolated proximal ILAs of normal
kidneys. Shown are EP3 receptor protein
(green), F-actin (red), and nuclei (blue).
A–C: ⫻10 objective. D–F: ⫻40 oil-immer-
sion objective. G: deconvoluted image using
⫻40 objective. White arrows, examples of
dense EP3 receptor labeling near the nucleus
of smooth muscle and endothelial cells.
www.ajprenal.org
 PGE2-INDUCED VASOCONSTRICTION OF PROXIMAL ILAS F1099
in the presence of the EP1 blocker were compared, the con- blockade did not prevent PGE2-induced vasoconstriction, sup-
centration-response curves to both agonists were superimpos- porting our conclusion as mentioned above.
able (Fig. 4D), indicating that the vasoconstrictor actions of Expression of EP3 receptor protein in proximal ILAs. As
PGE2 on ILAs are mediated by the EP3 receptor. shown in Fig. 6, EP3 receptor protein was present in smooth
To exclude a potential involvement of TXA2 receptors in the muscle cells of isolated proximal ILAs. Near the nucleus,
PGE2 response, PGE2 concentration-response curves were also labeling was more dense (Fig. 6G, white arrows), probably
constructed in the presence of a specific TXA2 receptor blocker reflecting receptor protein present in the perinuclear Golgi. In
(SQ29548; 1 ␮mol/l). As shown in Fig. 5, TXA2 receptor addition, EP3 receptor protein was also present on endothelial

Fig. 7. Comparison of EP3 (A–E) and EP1

(F–J) receptor expression in consecutive
proximal ILA slides. Shown are EP3 or EP1
receptor protein (green), F-actin (red), and
nuclei (blue). A–C and F–H: ⫻10 objective.
D and I: ⫻40 oil-immersion objective. E and
J: deconvoluted image using ⫻40 objective.
Note that in contrast to EP3, EP1 receptor
staining is most dense in the endothelium.
Also note that in I, the elastin membrane is
visible as a black line adjacent to the fluo-
rescent endothelium; with the short illumina-
tion times used, the elastin membrane did not
show autofluorescence.

AJP-Renal Physiol • VOL 292 • MARCH 2007 • www.ajprenal.org

Downloaded from www.physiology.org/journal/ajprenal (027.060.215.086) on January 27, 2020.
F1100 PGE2-INDUCED VASOCONSTRICTION OF PROXIMAL ILAS
cells, showing a similar distribution as in smooth muscle cells.
In contrast to EP3, labeling of smooth muscle cells was
relatively weak when an anti-EP1 antibody was used (Fig. 7,
F–J). Immunofluorescence of endothelial cells, on the other
hand, was very high. Dense labeling near the nucleus was not
observed when the anti-EP1 receptor antibody was used.
Effects of PGE2 on renal hemodynamics in isolated normal
kidneys. As shown in Fig. 8, PGE2 predominantly increased
renal perfusion in isolated normal kidneys. Thus, at concen-
trations ⱕ0.1 ␮mol/l, PGE2 increased renal perfusate flow both
under basal conditions (Fig. 8A; from 19.3 ⫾ 2.3 to 20.6 ⫾ 2.5
ml䡠min⫺1 䡠g⫺1; n ⫽ 6) and after pretreatment with ANG II
(Fig. 8B; from 9.1 ⫾ 0.9 to 16.8 ⫾ 1.3 ml䡠min⫺1䡠g⫺1; n ⫽ 6).
In both cases, this initial flow increase was partially reversed at
the highest PGE2 concentration of 1 ␮mol/l (basal: to 18.1 ⫾
2.5 ml䡠min⫺1䡠g⫺1; ANG II to 15.8 ⫾ 1.4 ml䡠min⫺1䡠g⫺1; for
both P ⬍ 0.05 vs. 0.1 ␮mol/l PGE2).
DISCUSSION
We found that PGE2 induced exclusive vasoconstriction of
proximal ILAs both in hydronephrotic rat kidneys and in those
isolated from normal kidneys. Smaller vessels, on the other
hand, i.e., distal ILAs and AAs, were found to dilate to PGE2.
Treatment of isolated proximal ILAs with drugs interacting
with the different PGE2 receptor subtypes indicated a role for
EP3 receptors in PGE2-induced vasoconstriction. The presence
of this receptor subtype on smooth muscle cells and also on
endothelial cells of proximal ILAs was confirmed using im-
munohistochemical staining. Adding PGE2 to normal isolated
kidneys induced a biphasic flow response; i.e., flow increased
at PGE2 concentrations ⱕ0.1 ␮mol/l and subsequently de-
creased at 1 ␮mol/l PGE2.
Our study is the first to demonstrate that the vasoconstrictor
action of PGE2 in the kidney is confined to the larger preglo-
merular vessels. In hydronephrotic kidneys, we mapped the
PGE2 response along the preglomerular tree, finding vasocon-
striction of proximal ILAs, while distal ILAs and AAs only
manifested PGE2-induced vasodilation. This latter observation
is consistent with the response of AAs observed in most other
studies (4, 8, 24), although PGE2-induced vasoconstriction of
these vessels has also occasionally been reported (9, 24). In
AJP-Renal Physiol • VOL
Downloaded from www.physiology.org/journal/ajprenal (027.060.215.086) on January 27, 2020.
addition, we found that intermediate ILAs displayed a biphasic
response to increasing concentrations of PGE2, i.e., vasodila-
tion followed by vasoconstriction. Our data indicate that the
actions of PGE2 along the preglomerular vascular tree gradu-
ally change from exclusively vasoconstriction upstream to
vasodilation downstream. A similar response pattern has been
reported for serotonin (5), suggesting heterogeneity in receptor
expression of the different preglomerular segments.
Renal prostaglandin production has been reported to alter
during hydronephrosis (14, 15), which could have influenced
reactivity of the larger ILAs to PGE2. Therefore, we also
performed single-vessel experiments using tissue from normal
kidneys. We found that PGE2 elicited vasoconstriction of
isolated proximal ILAs, both under basal conditions and after
the induction of vascular tone. Thus responses were similar to
those observed in hydronephrotic rat kidneys, indicating a
general action of PGE2 on proximal ILAs. In addition, we
found in normal isolated kidneys that with increasing con-
centrations of PGE2 only the highest PGE2 concentration
reversed the increase to a decrease in renal perfusate flow,
indicating that, also in this preparation, not all vessels but
probably only the larger ones constrict in response to PGE2.
Of the different PGE2 receptors, EP1 and EP3 have been
shown to activate signal transduction pathways leading to
contraction of smooth muscle cells (3, 16). In proximal ILAs,
our findings indicate that the vasoconstrictor actions of PGE2
involve EP3 receptors. First, we found that PGE2-induced
vasoconstriction of isolated vessels could not be abolished by
selective TXA2 or EP1 receptor blockade. Furthermore, during
EP1 receptor antagonism, concentration-response curves to
PGE2 and to the EP1/EP3 agonist sulprostone were superim-
posable. Removal of the endothelium did not prevent the
vasoconstriction to PGE2, indicating that EP3 receptors are
activated directly on smooth muscle cells. The presence of EP3
receptor protein on smooth muscle cells was confirmed by
immunohistochemical staining. By contrast, EP1 receptor la-
beling was predominantly found at the level of the endothe-
lium. Tang et al. (24), who in contrast to our and various other
studies (4, 8), observed afferent arteriolar constriction to high
PGE2 concentrations, also found that this response was medi-
ated via EP3 receptors.
Fig. 8. Effects of PGE2 on renal perfusate flow in normal
isolated kidneys during control conditions (A) and after pre-
constriction with ANG II (0.1 nmol/l; B); n ⫽ 6. *P ⬍ 0.05 vs.
pre-PGE2 values. †P ⬍ 0.05 vs. 0.1 ␮mol/l PGE2.
292 • MARCH 2007 • www.ajprenal.org
 PGE2-INDUCED VASOCONSTRICTION OF PROXIMAL ILAS
We observed that when isolated proximal ILAs were pre-
treated with a selective EP1 receptor antagonist, PGE2-induced
vasoconstriction was not inhibited but rather potentiated, indi-
cating that EP1 receptor activation stimulates a vasodilator
process. This observation seems consistent with the study of
Audoly et al. (1) showing that PGE2 lowered arterial pressure
to a lesser extent in male EP1 receptor knockout mice than in
wild-type mice. Our immunohistochemical data indicate that
EP1 receptors are predominantly expressed on endothelial
cells. This observation extends the findings of previous studies
demonstrating the presence of EP1 receptors in renal vascular
structures (11, 13, 18). Thus EP1 receptor activation may
stimulate endothelial cells to produce and release vasodilatory
factors modulating EP3-mediated constriction. Indeed, we
found that after deliberate damage of endothelial cells vaso-
constrictor responses to PGE2 were enhanced. Furthermore, in
cultured cells, EP1 receptor activation has been shown to
increase intracellular Ca2⫹ concentration (26), which, in endo-
thelial cells, is known to stimulate nitric oxide production (6).
Our immunohistochemical staining revealed that EP3 recep-
tors were present on both smooth muscle and endothelial cells.
Thus, besides by directly activating smooth muscle cells, PGE2
might also induce vasoconstriction of ILAs by stimulating
endothelial cells to release vasoconstrictor agents. Using en-
dothelium-denuded proximal ILAs, we investigated the possi-
ble contribution of endothelial cells to the vasoconstrictor
actions of PGE2. We found in these vessels that PGE2 was still
able to induce vasoconstriction, suggesting that EP3 as well as
EP1 receptors on endothelial cells fulfill a different role.
A final comment should be made regarding our flow data.
The influence of PGE2 on flow in normal kidneys was biphasic.
At concentrations ⱕ0.1 ␮mol/l, flow increased both under
basal conditions and after preconstriction with ANG II. This
finding is consistent with previous data obtained in this model
(7, 10) and likely reflects the vasodilatory response of the small
preglomerular vessels to PGE2. However, at a concentration of
1 ␮mol/l, flow decreased in both situations. At this concentra-
tion, the arterioles and small arteries, which are the most
important resistance regulators, are completely dilated (Fig.
1B) and, hence, vasoconstriction of the larger arteries becomes
visible in the flow signal.
In summary, the present study demonstrates that PGE2
elicits vasoconstriction of proximal ILAs, mediated via EP3
receptor activation. Toward the glomerulus, the vasoconstrictor
effect of PGE2 gradually changes to PGE2-induced vasodila-
tion.
GRANTS
This study was supported by Grant C 00.1903 from the Dutch Kidney
Foundation.
REFERENCES
1. Audoly LP, Tilley SL, Goulet J, Key M, Nguyen M, Stock JL,
McNeish JD, Koller BH, Coffman TM. Identification of specific EP
receptors responsible for the hemodynamic effects of PGE2. Am J Physiol
Heart Circ Physiol 277: H924 –H930, 1999.
2. Breyer MD, Breyer RM. Prostaglandin E receptors and the kidney. Am J
Physiol Renal Physiol 279: F12–F23, 2000.
AJP-Renal Physiol • VOL 292 • MARCH 2007 •
Downloaded from www.physiology.org/journal/ajprenal (027.060.215.086) on January 27, 2020.
F1101
3. Breyer RM, Bagdassarian CK, Myers SA, Breyer MD. Prostanoid
receptors: subtypes and signaling. Annu Rev Pharmacol Toxicol 41:
661– 690, 2001.
4. Edwards RM. Effects of prostaglandins on vasoconstrictor action in
isolated renal arterioles. Am J Physiol Renal Fluid Electrolyte Physiol 248:
F779 –F784, 1985.
5. Endlich K, Kuhn R, Steinhausen M. Visualization of serotonin effects
on renal vessels of rats. Kidney Int 43: 314 –323, 1993.
6. Fleming I, Busse R. Signal transduction of eNOS activation. Cardiovasc
Res 43: 532–541, 1999.
7. Haylor J, Towers J. Renal vasodilator activity of prostaglandin E2 in the
rat anaesthetized with pentobarbitone. Br J Pharmacol 76: 131–137, 1982.
8. Imig JD, Breyer MD, Breyer RM. Contribution of prostaglandin EP2
receptors to renal microvascular reactivity in mice. Am J Physiol Renal
Physiol 283: F415–F422, 2002.
9. Inscho EW, Carmines PK, Navar LG. Prostaglandin influences on
afferent arteriolar responses to vasoconstrictor agonists. Am J Physiol
Renal Fluid Electrolyte Physiol 259: F157–F163, 1990.
10. Jackson EK, Heidemann HT, Branch RA, Gerkens JF. Low dose
intrarenal infusions of PGE2, PGI2, and 6-keto-PGE1 vasodilate the in vivo
rat kidney. Circ Res 51: 67–72, 1982.
11. Jensen BL, Mann B, Skott O, Kurtz A. Differential regulation of renal
prostaglandin receptor mRNAs by dietary salt intake in the rat. Kidney Int
56: 528 –537, 1999.
12. Loutzenhiser R. In situ studies of renal arteriolar function using the in
vitro-perfused hydronephrotic rat kidney. Int Rev Exp Path 36: 145–160,
1996.
13. Morath R, Klein T, Seyberth HW, Nusing RM. Immunolocalization of
the four prostaglandin E2 receptor proteins EP1, EP2, EP3, and EP4 in
human kidney. J Am Soc Nephrol 10: 1851–1860, 1999.
14. Morrison AR, Nishikawa K, Needleman P. Unmasking of thromboxane
A2 synthesis by ureteral obstruction in the rabbit kidney. Nature 269:
259 –260, 1977.
15. Morrison AR, Nishikawa K, Needleman P. Thromboxane A2 biosyn-
thesis in the ureter obstructed isolated perfused kidney of the rabbit.
J Pharmacol Exp Ther 205: 1– 8, 1978.
16. Narumiya S, Sugimoto Y, Ushikubi F. Prostanoid receptors: structures,
properties, and functions. Physiol Rev 79: 1193–1226, 1999.
17. Navar LG, Inscho EW, Maijd DSA, Imig JD, Harrison-Bernard LM,
Mitchell KD. Paracrine regulation of the renal microcirculation. Physiol
Rev 76: 425–536, 1996.
18. Purdy KE, Arendshorst WJ. EP1 and EP4 receptors mediate prostaglan-
din E2 actions in the microcirculation of rat kidney. Am J Physiol Renal
Physiol 279: F755–F764, 2000.
19. Steinhausen M, Endlich K. Controversies on glomerular filtration from
Ludwig to the present. Eur J Physiol 432: R73–R81, 1996.
20. Steinhausen M, Snoei H, Parekh N, Baker R, Johnson PC. Hydrone-
phrosis: a new method to visualize vas afferens, efferens and glomerular
network. Kidney Int 23: 794 – 806, 1983.
21. Takeuchi K, Takahashi N, Abe T, Abe K. Two isoforms of the rat kidney
EP3 receptor derived by alternative RNA splicing: intrarenal expression
co-localization. Biochem Biophys Res Commun 199: 834 – 840, 1994.
22. Takeuchi K, Takahashi N, Abe T, Ito O, Tsutsumi E, Taniyama Y,
Abe K. Functional difference between two isoforms of rat kidney pros-
taglandin receptor EP3 subtype. Biochem Biophys Res Commun 203:
1897–1903, 1994.
23. Takeuchi K, Takahashi N, Kato T, Abe T, Taniyama Y, Tsutsumi E,
Ito O, Nakagawara K, Abe K. Functional analysis and chromosomal
gene assignment of rat kidney prostaglandin EP3 receptor. Kidney Int
Suppl 55: S183–S186, 1996.
24. Tang L, Loutzenhiser K, Loutzenhiser R. Biphasic actions of prosta-
glandin E2 on the renal afferent arteriole: role of EP3 and EP4 receptors.
Circ Res 86: 663– 670, 2000.
25. van Rodijnen WF, van Lambalgen TA, van Wijhe MH, Tangelder GJ,
ter Wee PM. Renal microvascular actions of angiotensin II fragments.
Am J Physiol Renal Physiol 283: F86 –F92, 2002.
26. Watabe A, Sugimoto Y, Honda A, Irie A, Namba T, Negishi M, Ito S,
Narumiya S, Ichikawa A. Cloning and expression of cDNA for a mouse
EP1 subtype of prostaglandin E receptor. J Biol Chem 268: 20175–20178,
1993.
www.ajprenal.org