216
Radical Scavengers of Indapamide in
Prostacyclin Synthesis in Rat
Smooth Muscle Cell
Yoshio Uehara, Hiroaki Shirahase, Taiji Nagata, Toshihiko Ishimitsu, Shigeyoshi Morishita,
Seimei Osumi, Hiroaki Matsuoka, and Tsuneaki Sugimoto
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
Indapamide, a nonthiazide diuretic, exhibits direct vasodilator action as well as natriuretic and
diuretic effects. Although calcium antagonist-like activity has been addressed so far, the
mechanisms for vasodilator effect are still uncertain. To understand the wide range of
indapamide actions, we examined the effects of indapamide on the vascular eicosanoid
generation and investigated its mechanisms by using rat vascular smooth muscle cells in
culture. Indapamide uniquely increased the prostacyclin generation in the vascular smooth
muscle cells in a dose-dependent manner, whereas it did not affect the vasoconstrictor
thromboxane A2. Thiazide diuretics lowered the prostacyclin generation, while nonthiazide
derivatives did not affect the biosynthesis. Enzymatic analysis revealed that indapamide
affected neither [uC]arachidonate liberation nor prostacyclin synthase of the smooth muscle
cells. Indapamide eliminated a stable free radical in a cell-free system, lowered the formation
of malondialdehyde from lipid peroxides in rat brain homogenate, and reduced lipid peroxidation by the free radical generating system of xanthine-xanthine oxidase. Indeed, the
scavenging action of indapamide significantly attenuated the inhibitory activity of 15hydroperoxy-arachidonate to prostacyclin synthase activity. These results indicate that indapamide diuretic increases prostacyclin generation in the vascular smooth muscle cells possibly
through antioxidant effects and that the enhanced prostacyclin generation is partly responsible
for its direct vasodilator action. (Hypertension 1990;15:216-224)
A
n o-chlorobenzenesulfonamide molecule is a
prototype of various diuretics including thiazide and nonthiazide diuretics (Figure 1).
Although natriuretic and diuretic effects are mainly
due to the common structure of o-chlorobenzenesulfonamide, varied side chains give the diuretics characteristic properties. In this context, indapamide
diuretic comprises an indoline molecule and uniquely
exhibits a direct vasodilator action as well as a
natriuretic effect on the cortical collecting duct in the
kidney.1-4 The vasodilator effect is assumed to be
caused by its calcium channel blocker-like activity5 or
endothelium-derived relaxing factor-mediated
From the Second Department of Medicine (Y.U., T.N., T.I.,
H.M., T.S.), University of Tokyo, Tokyo, Japan, and Kyoto Pharmaceutical Industries, Ltd (H.S., S.M., S.O.), Kyoto, Japan.
Supported in part by a Grant-in-Aid from the Ministry of
Education, Science, and Culture of Japan (No. 62570384 and
01570469).
Address for correspondence: Yoshio Uehara, MD, The Second
Department of Medicine, University of Tokyo, 7-3-1 Hongo,
Bunkyo-ku, Tokyo 113, Japan.
Received December 21, 1988; accepted in revised form September 12, 1989.
mechanism.6 However, the exact mechanism is not
fully understood.
On the other hand, the biosynthetic system for
eicosanoid (prostaglandins and thromboxanes) is well
demonstrated in7 8the vascular smooth muscle cells
(VSMC) of rats. - Prostacyclin (PGI2), a major product of arachidonate metabolism in VSMC, has a
powerful vasodilator action and attenuates the9vasoconstrictor response to vasoactive substances. The
eicosanoid metabolism is a major component participating in the regulation of the constriction-relaxation
mechanism of VSMC.
To understand the wide range of indapamide
effects, it seems important to disclose the relevance
of the eicosanoid system in the vascular wall to the
extrarenal indapamide effects. Thus in this study, we
investigated the influences of indapamide on the
eicosanoid generation in VSMC by using cultured
VSMC of rats and attempted to define the mechanisms for its alteration.
Methods
Vascular Smooth Muscle Cell Culture
VSMC were isolated from the thoracic aortas of
7-week-old Wistar rats according to the method of
Uehara et al
Indapamide and Vascular Prostacyclin
217
CHi
o-c W orobenzerwsu Ifo na m td«
/
indapamid*
firosamid*
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
so.
trichJoromethiazid*
FIGURE 1. Schematic drawing showing basal chemical structures of o-chlorobenzenesulfonamidederived diuretics. I, II, III, and TV represent sulfonyl, carbonyl, aminosulfonyl, and aminocarbonyl
group, respectively.
Ross et al.10 Briefly, rats were decapitated without
anesthesia. The descending aortas were immediately
obtained, and the surrounding connective tissue was
carefully removed with forceps and gauze. Then,
explants of ~5 mm2 were placed on a 94x21 mm
polystyrene dish (Corning-Iwaki Glass Co., Tokyo,
Japan) with the intimal side attached to the dish. Ten
milliliters Dulbecco's modified Eagle medium
(DMEM) (GIBCO Labs., Grand Island, New York),
supplemented with 20% (vol/vol) fetal bovine serum
(FBS) (GIBCO Labs.), penicillin (100 units/ml), and
streptomycin (100 /ig/ml), was gently added. The
dishes were kept at 37° C in a humidified atmosphere
of 95% air and 5% CO2. The cells reached confluency
10 days after inoculation. They were harvested by
brief exposure to Hanks' medium, supplemented
with 0.05% (wt/vol) trypsin and 0.02% (wt/vol) ethylenediaminetetraacetic acid disodium, and transferred into a new dish. At this stage, the cells were
designated as the first passage. FBS was reduced to
10% (vol/vol) in the subsequent media. Harvesting
was repeated when the cell growth became confluent.
VSMC from the fourth to the sixth passage were used
in the following studies.
Figure 2A shows the appearance of the cells in
primary culture and in the sixth generation (Figure
2B) on a phase contrast microscope (model IMT Olympus microscope, Olympus Optical Co., Tokyo, Japan).
The cultured VSMC were characterized by the "hills
and valleys" growth pattern on a light microscope,8-11
and the occurrence of cytoplasmic myofilaments was
confirmed by electron microscopy. Actin molecules
were demonstrated by antiactin monoclonal antibody
(Amersham International pic, Buckinghamshire,
England), antimuscle actin immunoglobulin G (IgG)
(Biomed. Technol. Inc., Stoughton, Massachusetts),
and peroxidase-labeled anti-immunoglobulin M (IgM)
or anti-IgG antibody (Kirkegaard & Perry Labs., Inc.,
Gaithersburg, Maryland).8
Effects of Diuretics on Eicosanoid Generation in
Vascular Smooth Muscle Cells
The confluent VSMC in the 35x18 mm six-well
polystyrene dish were washed three times with FBSfree Dulbecco's balanced phosphate-buffered saline
solution (D-PBS) (GIBCO Labs.). Then, the VSMC
were continuously cultured for 1 hour in D-PBS with
10"6 M arachidonic acid and a given concentration of
various diuretics. Eicosanoid released in a D-PBS
medium was measured by direct radioimmunoassay.
The cells were detached with brief exposure to 0.05%
trypsin solution and washed repeatedly with D-PBS.
The cells were homogenized with a Polytron blender
(Kinematica GmbH, Littau, Switzerland), and the
protein content was measured by the method of
Lowry et al.12
Direct radioimmunoassay was performed according to the previous method.1314 Briefly, the assay
mixture consisted of 0.1 ml [3H]eicosanoid (104 disintegrations per minute), 0.1 ml diluted sample or
standard solution, and 0.1 ml diluted antibody solution. The assay mixture was incubated at 4° C for 24
hours. To separate the bound from the free
[3H]eicosanoid, 0.1 ml dextran-coated charcoal solution (2.5% charcoal and 0.25% dextran) was added,
and the mixture was immediately centrifuged at
l,000g at 4° C for 5 minutes. Radioactivity of the
bound [3H]eicosanoid was counted with an automatic
liquid scintillation counter.
218
Hypertension
Vol 15, No 2, February 1990
FIGURE 2.
Photomicrograph of vas-
cular smooth muscle cells (VSMC) on
a phase contrast microscope. Left
graph (Panel a) shows proliferative
VSMC in primary culture and right
graph shows (Panel b) stationary
VSMC at the sixth generation. Arrow in
the right graph indicates typical "hills
and valleys" growth pattern of VSMC.
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
b)
Anti-6-keto-prostaglandin Fi a (6-k-PGFla) serum
cross-reacted 1% with prostaglandin E^ (PGE2), 2%
with prostaglandin F ^ (PGF^), 0.08% with prostaglandin D2 (PGD2), 0.01% with thromboxane B^,
and less than 0.001% with arachidonate. Antithromboxane B2 serum cross-reacted 0.005% with
6-k-PGFla, 0.007% with PGE^ 0.018% with P G F ^
0.06% with PGD2, and less than 0.0001% with arachidonate. Anti-PGEj serum cross-reacted 0.4% with
6-k-PGFla, 0.5% with P G F ^ 0.1% with PGD2,
0.02% with thromboxane B^ and less than 0.001%
with arachidonate.
Effects of Diuretics on Liberation of
[C]Arachidonate From Vascular
Smooth Muscle Cells
Cells (105) were seeded onto a 35x18 mm six-well
polystyrene dish with 2 ml DMEM containing 10%
(vol/vol) FBS. When the VSMC became confluent,
the medium was reduced to 2 ml DMEM with 5%
(vol/vol) FBS and 0.5 /uCi [14C]arachidonate (50-60
mCi/mmol) and maintained for 1 hour to incorporate
[14C]arachidonate into the cellular membrane of the
VSMC. The VSMC labeled with [14C]arachidonate
were repeatedly washed with D-PBS supplemented
with 10~5 M indomethacin. The cells were cultured in
D-PBS with 10"5 M indomethacin and a given concentration of indapamide (10~8 to 10"4 M), bradykinin
(10"' to 10"5 M) (Peptide Institute, Inc., Osaka, Japan),
or calcium ionophore A23187 (10"9 to 10"5 M) (Boehringer Mannheim GmbH, Mannheim, FRG).
[14C]Arachidonate released in the medium for 1 hour
was extracted with Folch's solvent (chloroform:
methanol, 2'\, vol:vol) and purified with silicic acid
thin layer chromatography. The radioactivity in the
zone corresponding to arachidonate was counted by
an automatic scintillation counter. The recovery rate
of this extraction method averaged 96 ±1% (n=6).
Protein content of the cells was measured with the
same procedure as described previously.
Effects of Diuretics on PGI2 Synthase Activity
PGI2 synthase activity was determined with use of
a modified method of Salmon et al.15~17 Briefly, 50 fi\
of the microsomal fraction (105 g pellet) of VSMC,
enriched with PGI2 synthase, was preincubated in 50
fi\ 25 mM Tris (hydroxymethyl)aminomethane HC1
(Tris-Cl) (Sigma Chemical Co., St. Louis, Missouri)
containing 100 mM NaCl and a given concentration
of diuretic at 4° C for 15 minutes (pH 7.5). The
reaction was initiated at 37° C by addition of 10 /xl
prostaglandin H2 (PGH2) (0.5 fig in acetone) (Ono
Pharmaceutical Co., Ltd, Osaka, Japan) and terminated with 50 ul of 50 mM FeCl2 solution after 3
minutes. The generated PGI2 was measured by the
radioimmunoassay as 6-k-PGF1<r To measure the
nonspecific conversion of PGH2 to PGI2, we measured the PGI2 generation by boiled enzymatic fraction. The nonspecific conversion was less than 1%.
Moreover, the amount of PGI2 generated by the
microsomal fraction was negligible when the substrate PGH2 was not added to the assay mixture.
Measurements of Scavenging Effects
The antioxidant activity of indapamide and some
diuretics, or a-tocopherol, was assessed by using a
stable free radical, a,a-diphenyl-/3-picrylhydrazyl
(DPPH) (Wako Pure Chemical Industries, Osaka,
Japan), according to the method of Blois.18 The
reduction of 10~4 M DPPH was initiated by addition
of a given concentration of the test compound. The
conversion of DPPH to its reduced form was
recorded at 25° C for 30 minutes by a Hitachi Model
U-3200 spectrophotometer (Hitachi, Ltd, Tokyo,
Japan) at 517 nm wavelength.
Uchara et al Indapamide and Vascular Prostacyclin
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
The antioxidant activity of indapamide was
assessed by an inhibitory activity to the formation of
malondialdehyde (MDA) in rat brain homogenate,
which according to Kubo et al,19 is an end product of
lipid peroxides. In brief, the brains from 7-week-old
male Wistar rats (Kitayama LABES Co., Kyoto,
Japan) were isolated after the rats were decapitated.
The cortical tissues were homogenized in ice-chilled
50 mM phosphate-buffered saline solution at pH 7.4
using an Ultra-Disperser (Yamato Scientific Co.,
Ltd, Tokyo, Japan). The homogenate was centrifuged at l,300g for 10 minutes at 4° C. The supernatant was diluted with the buffer to 2 mg/ml for the
protein concentration. This prepared supernatant
was incubated at 37° C for 30 minutes with a given
concentration of diuretics or a-tocopherol. The reaction was terminated by addition of 20% trichloroacetic acid (1 ; 1, vol:vol), and the mixture was immediately centrifuged at 2,000g for 15 minutes at 4° C. The
supernatant (2 ml) was heated at 100° C for 15
minutes with 0.4 ml 0.12 M thiobarbituric acid solution at pH 7.4. The optical absorbance of MDA was
measured at 532 nm by the spectrophotometer.
MDA in a medium before the assay was determined
so that the net MDA production during the reaction
could be measured. Protein concentration was measured according to the method of Lowry et al.12
We also explored the antioxidant effects of indapamide using the xanthine-xanthine oxidase free radical generating system.20"22 A given concentration of
the test compound in 15 y\ was mixed with 1.5 ml 50
mM linolenic acid (Sigma Chemical Co.) and 0.15 ml
0.5 mM xanthine (Sigma Chemical Co.) in 10 mM
phosphate-buffered solution including 1% sodium
dodecyl sulfate at pH 7.4. The reaction was initiated
at 37° C by addition of 0.15 ml (5 units/ml) xanthine
oxidase (Sigma Chemical Co.) and terminated with
1.5 ml 20% trichloroacetic acid after 10 minutes.
Peroxidation of linolenic acid was measured as MDA
formation. Briefly, the assay mixture was spun at
2,500g for 10 minutes at 4° C. An aliquot of the
supernatant was mixed with 0.2% butylated hydroxytoluene and 0.02 M thiobarbituric acid solution. The
mixture was treated at 100° C for 10 minutes. Absolute n-butanol was added to the ice-chilled mixture
(1:1, vol:vol), which was spun again at l,200g for 5
minutes at 4° C. The MDA in the organic phase was
measured with a spectrophotometer at 532 nm. The
nonspecific oxidation in this assay system was
+ 54*
P<0.001
2.0
o
+21*
P<aO2
-6*
NS
—*— —$-"
£5
-8*
NS
*• E
•D a
1.0
n-5
n-5
0
NS
T
—
n-5
s
+ 9*
NS
*
y*
^
n-5
n-5
n-5
W* W
W
UP
Concentration of Indapamide (M)
FIGURE 3. Line graph showing dose-response curve of indapamide to prostacyclin generation in vascular smooth muscle
cells (VSMC). Values ( • ) were compared with indapamidefree control value (o). Dose-dependency is analyzed by oneway analysis of variance. F value is 5.71 (p<0.001). NS,
statistically not significant.
checked under the experimental condition without
xanthine-xanthine oxidase. Moreover, we confirmed
that indapamide did not interfere with the xanthinexanthine oxidase reaction by measuring the formation of uric acid in the assay system by a modified
colorimetric method with the uricase-catalase system
of Kageyama23 (Uric Acid C-Test, Wako Pure Chemical Industries, Ltd., Osaka, Japan).
Finally, we attempted to examine the effects of
diuretics on the inhibitory activity of 15-hydroperoxyarachidonate (15-HPETE) to PGI2 synthase.
Enzymatic fraction (50 /A) was preincubated at 4° C
for 15 minutes in a mixture of 50 fil of 25 mM Tris-Cl
buffered solution with 100 mM NaCl and 15-HPETE
(0-2,000 ng/ml) in the absence or presence of 5 x 10"5
M diuretics. PGI2 synthase activity was determined
according to the procedure described in this section.
Reagents
The reagents were all of analytical grade. Authentic
eicosanoids and related compounds were a gift from
Ono Pharmaceutical Co., Ltd (Osaka, Japan). Mefruside was supplied by Yoshitomi Pharmaceutical Co.,
Ltd (Osaka, Japan); furosemide, trichloromethiazide,
and hydrochlorothiazide were obtained from Sigma
Chemical Co. (St. Louis, Missouri), and radioactive
TABLE 1. Effects of Indapamide on Vascular Elcosanold Generation
Concentration (M)
(«=5)
TXA2 (n=5)
219
PGEj (n=5)
PGF^ (n=5)
0.320±0.088
0.331±0.087
0
0.129±0.008
0.173±0.079
io-'
io- 7
0.129±0.010
0.128±0.014
0.446±0.203
0.787±0.341
lO"3
0.128±0.021
0.873±0.298*
0.127±0.035
0.332±0.083
Values are expressed as nanograms per milligrams protein per hour, n, number of rats; TXA2, thromboxane A2;
PGE2, prostaglandin E2; P G F ^ prostaglandtn F ^ .
*p<0.\ versus control (0 M) value.
220
Hypertension
Vol 15, No 2, February 1990
TABLE 2. Effects of Thiazide Diuretics on Vascular Eieosanold Generation
Hydrochlorothiazide
Trichloromethiazide
Concentrations (M)
PGI2 (TI=5)
TXA2 (TI=5)
PGI2 (n=6)
TXA2 (n=6)
0
10"'
lO"7
10"5
p values
1.20±0.04
1.19±0.08
l.ll±0.12
0.14±0.02
0.16±0.04
0.15±0.02
0.98±0.08*
NS
0.15±0.03
NS
1.21±0.08
1.15±0.09
1.15±0.07
0.95±0.06*
<0.01
0.12±0.01
0.14±0.01
0.12±0.01
0.16±0.02
NS
Values are expressed as nanograms per milligram protein per hour. Dose-dependency is assessed by one-way analysis
of variance. F value is 6.66 for trichloromethiazide. NS, statistically not significant; n, number of rats; PGI2,
prostaglandin I2; TXA2, thromboxane A2.
*p<0.05 versus respective control (0 M) values.
materials were purchased from Amersham International pic (Buckinghamshire, England).
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
Statistical Analysis
The values were expressed as mean±SEM. The
difference was assessed by the Student's / test or
one-way analysis of variance.
Results
The direct effect of indapamide, a nonthiazide
diuretic, on the vasodilator PGI2 generation in
VSMC is shown in Figure 3. The PGI2 production
was significantly stimulated with the indapamide
diuretic in a dose-dependent manner, increasing by
54% at a concentration of 10~4 M when compared
with the control value. In contrast, the diuretic did
not influence thromboxane biosynthesis in the VSMC
(Table 1). The VSMC had a small capacity in the
generation of PGEj and P G F ^ Indapamide tended
to increase vasodepressor PGE2 generation in
VSMC, although we could not find a significant
dose-dependency. PGF^ generation was not altered
by the diuretic (Table 1).
The PGI2 stimulatory effect was uniquely observed
in indapamide alone. As shown in Table 2, the
thiazide diuretics, that is, trichloromethiazide and
hydrochlorothiazide, produced a decrease in PGI2
generation, whereas the thromboxane generation was
not altered. The effects on PGI2 generation varied
among the nonthiazide diuretics. Mefruside, a sulfonyl
group diuretic (Figure 1), did not influence the PGI2
biosynthesis in the VSMC (Table 3). Although furosemide is classified as an aminocarbonyl group, which
is structurally related to indapamide (Figure 1), it did
not affect the PGI2 generation (Table 3). Furosemide
exhibited a significant increase in thromboxane A2
production when tested at a higher concentration.
To explore the mechanisms for the unique PGI2
stimulatory effect of indapamide, we analyzed enzymaticalry the eicosanoid biosynthesis. First, we investigated the liberation of [14C]arachidonate from the
cellular membrane of VSMC, into which the radioactive arachidonate had been incorporated. In comparison, bradykinin and the calcium ionophore A23187,
both of which are the stimulators of phospholipase A2,
were also examined. As shown in Figure 4, the control
compounds were found to stimulate, in a dosedependent manner, the [14C]arachidonate liberation
in this assay system. However, the indapamide diuretic
did not produce a significant change in the arachidonate liberation.
Second, we examined the direct effect of indapamide on PGI2 synthase in the microsomal fraction of
the VSMC (Figure 5). The PGI2 synthase activity was
not influenced by indapamide at concentrations ranging from 10~9 to 10~4 M. Furosemide, the control
compound, also did not affect the PGI2 synthase
activity.
Next, we explored the antioxidant effects of indapamide on the eicosanoid generation. Indapamide
diuretic significantly promoted the reduction of a
stable free radical, a,a-diphenyl-/3-picrylhydrazyl, in
a dose-dependent manner (Figure 6). However, neither trichloromethiazide nor furosemide facilitated
the reduction.
TABLE 3. Effects on Nonthiazide Diuretics on Vascular Eicosanoid Generation
Furosemide
Mefruside
Concentration (M)
0
io-'
io- 7
10"5
p values
PGI2 (n=8)
TXA2 (n=8)
PGI2 (n=6)
TXA2 (n=6)
1.20±0.07
1.22±0.06
1.21±0.10
1.33 ±0.07
0.16±0.02
1.08 ±0.05
0.14±0.01
0.15±0.01
0.20±0.03
NS
1.05 ±0.08
1.03 ±0.04
1.09±0.06
NS
0.18±0.01
0.19±0.02
0.18±0.01
0.27±0.01*
NS
NS
Values are expressed as nanograms per milligram protein per hour. Dose-dependency is assessed by one-way analysis
of variance. NS, statistically not significant; n, number of rats; PGI^ prostaglandin I2; TXA2, thromboxane A2.
*p<0.001 versus control value (0 M).
Uehara et al Indapamide and Vascular Prostacyclin
4n
TCMT
^NI
M
s
Jos
|
0
10"*
ft"*
107
10"*
1(T* KT"
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
FIGURE 4. Plot showing effects of indapamide on arachidonate liberation from vascular smooth muscle cells (VSMC). •
represent indapamide (IDP) diuretic; O, D represent bradykinin (BK) and calcium ionophore (A23187), respectively.
Value is an average of six rats. [14C]Arachidonate incorporated into membrane of VSMC for 1 hour averaged
15,449+423 (n=6) dpm/104 cells/hr. Dose-dependency is
analyzed by one-way analysis of variance. F values are 0.887
for indapamide (p>0.1), 4.22 for bradykinin (p<0.02), and
5.29 for calcium ionophore (p<0.005). *p<0.05; *+p<0.025
versus respective control values (0 M).
We examined the indapamide effects on MDA formation in the rat brain homogenate (Figure 7). Indapamide was found to markedly reduce the MDA formation in the rat brain homogenate. Moreover, the dose
for equipotency of the scavenging effect was significantly smaller in indapamide than a conventional scavenger such as a-tocopherol. The thiazide diuretic and
furosemide did not exhibit the antioxidant effects on
the MDA formation in the homogenate.
To ascertain the possible scavenging property of
indapamide, we examined its effects on the peroxidation of linolenic acid by the free radical generating
system of xanthine-xanthine oxidase. We found that
this system caused the peroxidation of linolenic acid,
which was inhibited 37% with superoxide dismutase
coupled with catalase (4.505±0.155 vs. 2.820±0.232
nmol MDA/ml/10 min,p<0.005). By using this assay
system, we assessed the scavenging effects of indapamide. As depicted in Figure 8, indapamide reduced
FRS
\
IDP \
58
Concentration of Compound (M)
221
0
KT* IIT1 10Conc«ntration of Compound (M)
FIGURE 6. Plot showing antioxidant effects of indapamide
on a,a-diphcnyl-fi-picrylhydrazyl. Circles, squares, and triangles represent indapamide (IDP),
trichloromethiazide
(TCMT), and furosemide (FRS), respectively. Value is an
average of four studies. Dose-dependency is analyzed by
one-way analysis of variance. F values are 5,524 for indapamide (p<0.001), 1.00 for trichloromethiazide (p>0.1) and
2.49 for furosemide (p>0.1).
the formation of MDA in a dose-dependent manner,
which became 29% of the control value at 10"4 M. In
contrast, neither trichloromethiazide nor furosemide
affected the formation.
In this context, we examined whether the compound could attenuate the inhibitory activity of free
radicals to PGI2 synthase (Figure 9). As shown in the
open circles, 15-HPETE, a free radical provider and
potent inhibitor to PGI2 synthase, suppressed the
PGI2 synthase activity in a dose-dependent manner.
However, when 5 x 10~5 M indapamide occurred in the
assay mixture, the inhibitory activity of 15-HPETE
was significantly lowered, thereby shifting the doseresponse curve toward the right. Furosemide did not
produce any alteration in the dose-response curve.
0.25(
o
itr*
io-« itr1
Concentration of Compound (M)
°- 05
o IO-* 10-* 10-' itr 1 icr* i r *
Concentration of Compound (M)
FIGURE 5. Plot showing effects of indapamide on PGI2
synthase activity. Circles and triangles represent indapamide
(IDP) and furosemide (FRS), respectively. Value is an average
of six studies. Dose-dependency is analyzed by one-way analysis of variance. F values are 0.46 for indapamide (p>0.1) and
0.35 for furosemide (p>0.1).
FIGURE 7. Plot showing effects of indapamide on malondialdehyde formation in rat brain homogenate. •, Q A, and x
represent indapamide (IDP), trichloromethiazide (TCMT),
furosemide (FRS), and a-tocopherol (at-TOCO), respectively.
Value is an average of four studies. Dose-dependency is
analyzed by one-way analysis of variance. F values are 240 for
indapamide (p<0.001), 4.63 for a-tocopherol (p<0.01), 0.02
for trichloromethiazide (p>0.1), and 0.96 for furosemide
222
Hypertension
Vol 15, No 2, February 1990
o
icr*
i«r*
iir 4
Concentration of Compound (u)
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
FIGURE 8. Plot showing effects of indapamide on linolenic
acid oxidation. •, a, and A represent indapamide (IDP),
trichloromethiazide (TCMT), and furosemide (FRS), respectively. Value is an average of four studies. Dose-dependency is
analyzed by one-way analysis of variance. F values are 52.38
for indapamide (p<0.0001), 2.55 for furosemide (p>0.1),
and 0.16 for trichloromethiazide (p>0.1). *p<0.005 versus
respective control values (0 M).
Discussion
The VSMC were primarily cultured and maintained according to the standard method of Ross et
al.10 The cells were characterized by various properties of growth pattern, which are characteristic of
vascular smooth muscle cells.8-11 Moreover, the origin
of vascular smooth muscle was confirmed as well by
light and electron microscopy and immunohistochemical findings.8
In this study, we demonstrated that indapamide
was a potent stimulator of vasodilator PGI2 generation in VSMC. Indapamide did not affect the vasoconstrictor thromboxane A2, thereby shifting the
PGI2-to-thromboxane A2 ratio toward a vasodilator
state. The enhanced PGE^ generation, although a
small amount, is likely to produce vasodilation as
0
16
80
400 2000
Concentration of 15HPETE
(ng/ml)
FIGURE 9. Plot showing effects of 15-hydroperoxyarachidonate (15-HPETE) on prostaglandin I2 (PGIJ synthase activity, o represent dose-response curve of 15-HPETE. • represent
dose-response curve when 15-HPETE was preincubated with
indapamide. A representpreincubation with furosemide. Dosedependency is assessed by one-way analysis of variance. F
values are 1,514 for 15-HPETE alone (p<0.001), 44 for
furosemide (p<0.001), and 40 for indapamide (p<0.001).
*p<0.02; **p<0.005 versus respective control values.
well. The direct vasodilator action of indapamide is
reportedly observed at the relatively higher concentrations, that is, 10"5 to 10~4 M . w The concentrations that elicit vasodilation are completely correspondent with those that stimulate PGI2 generation
in VSMC. PGI2 and PGE; potentiate adenylate
cyclase activity,24 which finally leads to a decrease in
the cytosolic calcium concentration.25 Conversely,
vasoconstrictor thromboxane A2 enhances the cytosolic calcium through phosphoinositide metabolism
and a receptor-mediated calcium channel mechanism.26 Although the involvement of the eicosanoid
system in the direct vasodilation by indapamide has
not been proven so far, the increase in the PGI2to-thromboxane A2 ratio observed in the VSMC
could be one of the factors that cause vasodilation.
To address these questions more directly, however,
we might be required to test the prostaglandin inhibitors in the direct vasodilator action of the indapamide diuretic.
We showed that indapamide enhanced the elimination of the stable free radical DPPH and reduced
the formation of MDA in the rat brain homogenate.
More directly, we demonstrated that indapamide lowered the oxidation of linolenic acid in the xanthinexanthine oxidase system. It has been reported21 that
xanthine-xanthine oxidase is a free radical-generating
system and that the oxidation of fatty acids is a free
radical-mediated reaction.20-22 We also showed that
0.5 unit superoxide dismutase inhibited the oxidation
37% in this assay system. These data clearly indicate
that indapamide has marked antioxidant effects particularly at the relatively higher concentrations. In
addition, since such antioxidant properties were found
in neither trichloromethiazide nor furosemide, the
scavenging effect is unique to the indapamide diuretic.
Free radicals are more destructive to PGI2 synthase and the cyclooxygenase-hydroperoxidase complex becomes the target as well.2728 Indeed, ONO3144 (Ono Pharmaceutical Co., Ltd, Osaka, Japan), a
potent antioxidant and anti-inflammatory drug, as
well as MK-44727-28 reportedly enhances PGI2 and
PGEj formation while thromboxane A2 generation is
somehow lowered in guinea pig macrophages.29 The
enhanced PGI2 generation could be partly explained
by a diversion of endoperoxide from MDA formation
to PGI2 synthesis. However, it seemed probable that
the antioxidant effect of indapamide protects PGI2
synthase from the attacks of endogenous free radicals, thereby increasing PGI2 formation in VSMC. In
fact, the enzyme analyses revealed that arachidonate
liberation was not responsible for the enhanced PGI2
generation in indapamide-stimulated VSMC. Moreover, we found that indapamide failed to directly
activate PGI2 synthase in a cell-free assay system. In
contrast, this diuretic attenuated PGI2 synthase
inhibitory activity of 15-HPETE, a provider of free
radicals and a powerful inhibitor of PGI2 synthase.
These data are supportive of the involvement of
scavenging effects in the enhanced PGI2 generation
by indapamide diuretic.
Uehara et al
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
In this context, it is reported that PGH synthase
generates superoxide in the presence of reduced
nicotinamide-adenine dinucleotide or nicotinamideadenine dinucleotide B'-phosphate.30 The eumination
of superoxide generated in the formation of PGH2 is
expected to enhance PGH2 production, which would
lead to increased PGE2 generation.29 Hence, the fact
that indapamide tended to increase the PGE2 generation in VSMC could be accounted for by such
scavenging effects on the PGH2 formation.
From a viewpoint of a structure-activity relation,
the common structure of o-chlorobenzenesulfonamide, the basal component for natriuretic effects, is
irrelevant to PGI2 stimulation observed in indapamide (Figure 1). Because PGI2 stimulation and scavenging effects were found in neither mefruside nor
furosemide, the side chains appeared to be largely
responsible for these various properties. Moreover,
there are naturally a few types of free radicals, that is,
superoxide, hydroxy radical, singlet oxygen, or peroxide. It is still uncertain what type of free radicals is
scavenged by the indapamide diuretic. Electrochemical analyses are needed to reveal the mechanisms for
the antioxidant effects of this diuretic.
It has been reported that the oral dosage of
indapamide for obtaining sufficient antihypertensive
effects in rats is more than 1 mg/kg body wt,31-32
which is about 10 times as much as the dosage for
human subjects.33 In rats, a single oral dosage of 1
mg/kg is reported to reach the peak plasma concentration of 1.3 jig/ml, which is equivalent to 3.5 xlO"6
M.34'35 Based on these studies, the indapamide concentrations used in our in vitro study are probably
attainable in rat plasma.
Finally, indapamide exhibited the enhanced PGI2
generation and scavenging effects in vitro. These
properties are assumed to be protective from the
arteriosclerotic changes or tissue damages caused by
blood pressure elevation.36'37 We do not have in vivo
data available for the indapamide effects. In humans,
however, it has been reported that indapamide does
not deteriorate renal function even when given to
patients with renal impairment.38 Thus, our next goal
is to test whether the indapamide diuretic exhibits in
vivo protective effects against organ damage in hypertensive models. Experiments are in progress to
address these questions.
Acknowledgments
We thank Yukari Kawabata for excellent technical
assistance. We acknowledge Ono Pharmaceutical Co.,
Ltd, Osaka, Japan, for the gift of eicosanoids and
related compounds, and Yoshitomi Pharmaceutical
Co., Ltd, Tokyo, Japan, for the supply of mefruside.
References
1. Moore RA, Seki T, Ohsumi S, Oheim K, Kyncl J, Desnoyers P:
Antihypertensive action of indapamide and review of pharmacology and toxicology. Curr Med Res Opin 1977;5(suppl 1):
25-32
Indapamide and Vascular Prostacyclin
223
2. Beregi LG: Antihypertensive and saluretic properties of the
indoline and isoindoline series. Curr Med Res Opin 1977;
5(suppl l):3-8
3. Finch L, Hicks PE: Studies on the marked antihypertensive
properties of indapamide (SE 1520) in rats and cats. Br J
Pharmacol 1976^8:282-283
4. Uhlich E, Troger C, Knoll W: Effects of indapamide in
hypertensive patients and on experimental vascular reactivity.
CurrMcd Res Opin 1977;5(suppl l):71-78
5. Mironneau J: Indapamide-induced inhibition of calcium
movement in smooth muscles. Am J Med 1988;84(suppl
6. Vanhoutte PM, Luscher TF: Role of the endothelium in the
pathophysiology of hypertension. JAMA SEA 1988;
4(suppl):34-42
7. Moncada S, Herman AG, Higgs EA, Vane JR: Differential
formation of prostacyclin (PGX or PGI2) by layers of the
arterial wall. An explanation for the anti-thrombotic properties of vascular endothelium. Thromb Res 1977;ll:323-344
8. Uehara Y, Ishimitsu T, Kimura K, Ishii M, Ikeda T, Sugimoto
T: Regulatory effects of eicosanoids on thymidine uptake by
vascular smooth muscle cells of rats. Prostaglandins 1988;
36:847-857
9. Spokas EC, Quilley J, McGiff JC: Prostaglandins in hypertension, in Genest CJ, Kuchel O, Hamet P, Cantin M (eds):
Hypertension. New York, McGraw-Hill Book Co, 1983, pp
373-393
10. Ross R, Glomset J, Kariya B, Harker L: A platelet-dependent
serum factor that stimulates the proliferation of arterial
smooth muscle cells in vitro. Proc Natl Acad Sci USA 1974;
71:1207-1210
11. Ross R, Kariya B: Morphogenesis of vascular smooth muscle
in atherosclerosis and cell structure, in Bohr DF, Somlyo AP,
Sparks HV (eds): Handbook of Physiology, Section 2: The
Cardiovascular System, Volume II, Vascular Smooth Muscle.
Bethesda, American Physiological Society, 1980, pp 69-91
12. Lowry OH, Rosenbrough NJ, Farr AL, Randall RJ: Protein
measurement with the folin phenol reagent. / Biol Chem
1951;193:265-275
13. Uehara Y, Tobian L, Iwai J, Ishii M, Sugimoto T: Alterations
of vascular prostacyclin and thromboxane A2 in Dahl genetical
strain susceptible to salt-induced hypertension. Prostaglandins
1987;33:727-738
14. Ishimitsu T, Uehara Y, Ishii M, Ikeda T, Matsuoka H,
Sugimoto T: Thromboxane and vascular smooth muscle cells
in genetically hypertensive rats. Hypertension 1988;12:46-51
15. Salmon JA, Flower RJ: Preparation and assay of prostacyclin
synthase. Methods Enzymol 1982;86:91-99
16. Uehara Y, Ishimitsu T, Ishii M, Sugimoto T: Prostacyclin
synthase and phospholipases in the vascular wall of experimental hypertensive rats. Prostaglandins 1987;34:423-432
17. Numabe A, Uehara Y, Hirawa N, Takada S, Yagi S: Effects of
thiazide diuretic on vascular eicosanoid system of spontaneously hypertensive rats. / Hypertens 1989;7:493-499
18. Blois MS: Antioxidant determinations by the use of a stable
free radical. Nature 1958;181:1199-1200
19. Kubo K, Yoshitake I, Kumada Y, Shuto K, Nakamizo N:
Radical scavenging action of flunarizine in rat brain in vitro.
Arch Int Pharmacodyn 1984;272:283-295
20. Fukuzawa K, Takase S, Tsukatani H: The effect of concentration of the antioxidant effectiveness of o-tocopherol in lipid
peroxidation induced by superoxide free radicals. Arch Biochem Biophys 1985;240:117-120
21. Porter N, Lehman LS: Xanthine oxidase initiated oxidation of
model membranes. Effect of position of abstractable hydrogen
atoms in the bilayer on the distribution of products. / Am
Chem Soc 1982;104:4731-4732
22. Sutherland MW, Gebicki JM: A reaction between the superoxide free radical and lipid hydroperoxide in sodium linoleate
micelles. Arch Biochem Biophys 1982;2414:1-11
23. Kageyama N: A direct colorimetric determination of uric acid
in serum and urine with uricaseotalase system. Clin Chim
Ada 1971^31:421-426
224
Hypertension
Vol 15, No 2, February 1990
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
24. Miller OV, Aiken JW, Henker DP, Shebuski RJ, Gorman RR:
Prostacyclin stimulation of dog arterial cyclic AMP levels.
Pmstagtandins 1979;18:915-925
25. Ignarro LJ, Kadowitz PJ: The pharmacological and physiological role of cyclic GMP in vascular smooth muscle relaxation.
Annu Rev Pharmacol Tcoricol 1985;25:171-191
26. Hallam TJ, Sanchez A, Rink TJ: Effects of excitatory and
inhibitory icosanoids on cytoplasmic free calcium directly
measured in human platelets with fluorescent indicator quin-2,
in Branquet P, Garay RP, Frolich JC, Nicosia S (eds):
Prostaglandins and membrane ion transport. New York,
Raven PTCSS, Publishers, 1985, pp 157-163
27. Ham EA, Egan RW, Soderman DD, Gale PH, KuehJ FA:
Peroxidase-dependent deactivation of prostacyclin synthetase.
JBiolChem 1979;254:2191-2194
28. Kuehl FA, Humes TL, Torchiana ML, Ham EA, Egan RW:
Oxygen-centered radicals in inflammatory process. Adv Inflam
Res 1979;l:419-430
29. Crowshaw K: Role of inhibition of prostaglandin biosynthesis
by antiinflammatory drugs in the pathogenesis of renal damage. Adv Inflam Res 1984;6:149-158
30. Kukreja RC, Kontos HA, Hess ML, Ellis EF: PGH synthase
and lipoxygenase generate superoxide in the presence of
NADH or NADPH. Ore Res 1986^9:612-619
31. Kyncl J, Oheim K, Seki T, Solles A: Antihypertensive action of
indapamide. Comparative studies in several experimental
models. Anneimittelforschung 1975;25:1491-1494
32. Delbarre B, Delbarre G: Effect of indapamide on an experimental model of cerebral ischemia in hypertensive rats. Am J
Med 1988;84(suppl lB):20-25
33. Morledge JH: Clinical efficacy and safety of indapamide in
essential hypertension. Am Heart J 1983;106:229-232
34. Campbell DB, Taylor AR, Hopkins YW, Williams JRB:
Pharmacokinetics and metabolism of indapamide: A review.
Curr Med Res Opin 1977;5(suppl l):13-24
35. Coppi G, Mosconi P, Springolo V: Pharmacokinetics of a fixed
combination of atenolol and indapamide in dogs and humans
after oral administration. Curr Ther Res 1987;42:411-418
36. Glavind J, Hartman S, Clemmesen J, Jessen KE, Dan S:
Studies on role of lipoperoxides in human pathology. II. The
presence of peroxidized lipid in the atherosclerotic aorta. Acta
Pathol Microbiol Scand 1952;30:l-6
37. Gryglewski RJ, Dembinska-Kiec A, Zmuda A, Gryglewski T:
Prostacyclin and thromboxane A2 biosynthesis capacities of
heart, arteries and platelets at various stages of experimental
atherosclerosis rabbits. Atherosclerosis 1978;31:385-394
38. Acchiardo SR, Skoutakis VA: Clinical efficacy, safety, and
pharmacokinetics of indapamide in renal impairment. Am
Heart J 1983;106:237-244
KEY WORDS
thromboxane
1
prostacyclin
diuretic therapy
vascular smooth muscle
Radical scavengers of indapamide in prostacyclin synthesis in rat smooth muscle cell.
Y Uehara, H Shirahase, T Nagata, T Ishimitsu, S Morishita, S Osumi, H Matsuoka and T
Sugimoto
Hypertension. 1990;15:216-224
doi: 10.1161/01.HYP.15.2.216
Downloaded from http://hyper.ahajournals.org/ by guest on June 17, 2017
Hypertension is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1990 American Heart Association, Inc. All rights reserved.
Print ISSN: 0194-911X. Online ISSN: 1524-4563
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://hyper.ahajournals.org/content/15/2/216
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Hypertension 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 Hypertension is online at:
http://hyper.ahajournals.org//subscriptions/