Effects of angiotensin II and angiotensin-(1-7) on the release of [3H]norepinephrine from rat
atria.
M M Gironacci, E Adler-Graschinsky, C Peña and M A Enero
Hypertension. 1994;24:457-460
doi: 10.1161/01.HYP.24.4.457
Hypertension is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1994 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/24/4/457
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/
Downloaded from http://hyper.ahajournals.org/ by guest on March 3, 2014
457
Effects of Angiotensin II and Angiotensin-(l-7)
on the Release of [3H]Norepinephrine
From Rat Atria
M.M. Gironacci, E. Adler-Graschinsky, C. Pena, M.A. Enerot
Abstract We examined the effects of angiotensin II (Ang II)
and Ang-(l-7) on the release of [3H]norepinephrine elicited by
nerve stimulation (2 Hz, 0.5 millisecond, for 2 minutes) in rat
atria isolated with their cardioaccelerans nerves. The stimulation-induced release of [3H]norepinephrine was increased 50%
by 3xl0" s mol/L of either peptide. No further increase in
pH] norepinephrine release was observed with peptide concentrations up to 3 x 10"7 mol/L. This effect was completely blocked
by the nonselective angiotensin receptor antagonist saralasin
(1 x 10"7 mol/L). The type 1 angiotensin receptor antagonist DuP
753 ( l x 10"6 mol/L) entirely prevented the increases in [3H]nor-
epinephrine caused by Ang II and Ang-(l-7). On the other hand,
the type 2 angiotensin receptor antagonist PD 123319 (1 x 10"6
mol/L) prevented the increase in pHJnorepinephrine release
elicited by Ang-(l-7) but not by Ang II. These results suggest that
Ang-(l-7), like Ang II, could have a neuromodulatory function in
rat atria via activation of specific angiotensin receptor subtypes,
which could be the subtype 1 angiotensin receptor for Ang II and
subtypes 1 and 2 for Ang-(l-7). (Hypertension. 1994^4:457-460.)
A ngiotensin-(l-7) [Ang-(l-7)] is an active compo/ \
nent of the renin-angiotensin system1 that lacks
JL \~ the phenylalanine present in position 8 of Ang
II. It has been shown that although both peptides are
products of Ang I metabolism, they result from separate
enzymatic pathways. Although angiotensin-converting
enzyme promotes Ang II formation, a prolyl endopeptidase converts either Ang I or II into Ang-(l-7). :w This
heptapeptide was found to be present in central and
peripheral tissues of rats4 and dogs2 and to excite
neurons in the central nervous system.5'6 Although some
of the responses elicited by Ang-(l-7) were similar to
those of Ang II,5-8 different actions for both peptides
were also reported; ie, the heptapeptide is not a dipsogen,9 a constrictor agent of blood vessels,10 or an
aldosterone secretagogue.10 Moreover, different actions
of both peptides in the regulation of cell function were
found in isolated rabbit vasa deferentia, where Ang(1-7) was as potent as Ang II in increasing prostaglandin
synthesis but was devoid of the neuromodulatory activity found for Ang II.11
The aim of the present work was to compare the
effects of the heptapeptide and of Ang II in the release
of [3H]norepinephrine ([3H]NE) in rat atria isolated
with their cardioaccelerans nerves. The effects of both
peptides were assayed in the presence of different
angiotensin receptor antagonists: DuP 753, which is
selective for type 1 angiotensin receptors (AT^; PD
123319, which is selective for type 2 angiotensin receptors (AT 2 ) 1216 ; and saralasin, which blocks both types of
angiotensin receptors.
Received April 1, 1994; accepted in revised form July 6, 1994.
From the Institute) de Qufmica y Fisicoqufmica Biol6gicas
(UBA-CONICET), Facultad de Farmacia y Bioqufmica, Universidad de Buenos Aires (M.M.G., C.P.), and Instituto de Investigaciones Farmacol6gicas (CONICET) (E.A.-G., M.A.E.), Buenos
Aires, Argentina.
tDeceased.
Correspondence to Dr Edda Adler-Graschinsky, Instituto de
Investigaciones Farmacol6gicas, Junfn 956, 5° Piso, Buenos Aires
(1113), Argentina.
© 1994 American Heart Association, Inc.
Keywords • angiotensins • norepinephrine •
angiotensin • angiotensin II
receptors,
Methods
Synthesis of Ang-(l-7)
The Merrifield solid-phase procedure17 was employed, using
Boc-amino acid derivatives. The crude peptide was purified by
high-performance liquid chromatography (HPLC) in a C18
column eluted with a 0% to 40% acetonitrile gradient at a flow
rate of 1.8 mL/min. The purified product was characterized as
a single component by HPLC and thin-layer chromatography.
It showed aspartic acid as the amino-terminal residue as well
as the correct amino acid composition and sequence.
Tissue Preparation
Female Wistar rats (180 to 200 g) were anesthetized with
ether, and the heart was rapidly removed. Both atria were
dissected with their cardioaccelerans nerves in modified
Krebs' solution of the following composition (10~3 mol/L):
NaCl 118.0, KC1 4.7, CaCl2 2.6, MgCl2 1.2, NaHCO3 25.0,
glucose 11.1, EDTA 0.004, and ascorbic acid 0.11. Atropine
(1.4x10"' mol/L) was added to the Krebs' solution to exclude
any influence of muscarinic receptors on the release of norepinephrine. The atria were set up in a 5-mL isolated organ
bath equipped with platinum electrodes for nerve stimulation.
Incubations were carried out in the modified Krebs' solution at
37°C with continuous bubbling of 95% O 2 /5% CO2.
The spontaneous contractions of the preparation were recorded through a Grass FT03C transducer connected to a
Grass polygraph. An equilibration period was allowed to
elapse until the basal resting rate did not differ by more than
10 beats per minute during a 10-minute interval.
[3H]NE Overflow Measurement
Endogenous norepinephrine stores were labeled by incubation of the tissue at 37"C for 30 minutes with 5 ^Ci/mL of
(±)-7-pH]NE (specific activity, 14.3 Ci/mmol; New England
Nuclear Corp) as described by Adler-Graschinsky et al." After
Downloaded from http://hyper.ahajournals.org/ by guest on March 3, 2014
458
Hypertension
Vol 24, No 4
October 1994
2.0r
1.0
1.S
P5"
1.10-
3.10
^•CONTROL
IANG II
3.10"
(1-7)
Fra 1. Bar graph shows effect of angiotensin II (ANQ II) and
ANG-(1-7) (3x10- 8 , 1 x 1 0 - 7 , and 3 x 1 0 " 7 mol/L, n=6) on spontaneous tritium overflow in rat Isolated atria labeled with
pHJnorepinephrine. B2/B1 represents the ratio between basal
tritium overflow obtained before the second and first stimulation
periods. In controls, both B? and B, were obtained after incubation with saline. In the remaining groups, B? was preceded by a
2-minute incubation with the corresponding peptide. Values are
meaniSEM. Drug concentrations are in moles per liter.
incubation, eight consecutive 1-minute washes and then 10
consecutive 5-minute washes with Krebs' solution were performed in every tissue.
Eighty minutes after the end of the incubation of atria with
the 3H transmitter, two consecutive stimulation periods (S, and
S2) were applied 30 minutes apart (square-wave pulses 0.5
millisecond long, at 2 Hz and supramaximal voltage, during 2
minutes).
Antagonists and peptides were added 4 and 2 minutes,
respectively, before S2 and were replaced every 2 minutes
whenever the bath fluid was renewed.
The spontaneous outflow of tritium was measured in 0.5-mL
samples collected every 5 minutes. The tritium release induced
by nerve stimulation was calculated by subtracting the spontaneous outflow assumed to have occurred in each sample
during and after the stimulation period and was expressed as
the fractional release per shock (FR): total evoked overflow
(nCi) per pulse divided by the total nCi remaining in the tissue
at the onset of stimulation. This last value was calculated by
addition of the radioactivity lost during the successive washes
to that measured in the tissue at the end of the experiment.
The spontaneous outflow was the basal resting release obtained in the period immediately before the stimulation.
I
3.10
1.10
3.10
CZ3 ANG (1-7)
I CONTROL
EC3ANG II
FIG 2. Bar graph shows effect of angiotensin II (ANG II) and
ANG-(1-7) (3x10- f l , 1 x i r r 7 , and 3X10" 7 mol/L, n=6) on tritium
overflow elicited by nerve stimulation In rat atria preincubated
with pHJnorepinephrine. Two periods of nerve stimulation were
applied (2 Hz, 2 minutes) 30 minutes apart. S2/S, represents the
ratio between tritium overflow in response to the second and first
stimulation periods. In controls, both Sj and St were obtained
after incubation with saline. In the remaining groups, &? was
preceded by a 2-minute incubation with the corresponding
peptide. Values are mean±SEM. *P<.05 compared with control.
Drug concentrations are expressed in moles per liter.
Ang-(l-7) (Fig 2); ie, the ratio between S2 and Si, which
under control conditions was 0.88 ± 0.02 (n=6), was
increased up to 50% by concentrations as low as 3xl0" 8
mol/L of either peptide. No further increases in [3H]NE
release in response to nerve stimulation were observed
with concentrations of both peptides as high as 3xlO~7
mol/L.
Effects of Angiotensin Receptor Antagonists on
the Increase in [3H]NE Release Elicited by
Ang II and Ang-(l-7)
As shown in Fig 3, the nonselective antagonist of
angiotensin receptors saralasin, at 1 x 10"7 mol/L, abolished the facilitatory effect on [3H]NE release by nerve
stimulation elicited by both Ang II and Ang-(l-7).
Statistical Analysis
All values are mean±SEM. Data were submitted to one-way
ANOVA. Post hoc analysis with the Schefte test was carried
out in every case, and probability values less than .05 were
considered significant.
Drugs
Ang II was purchased from Sigma Chemical Co. DuP 753
was obtained from the Du Pont Co. PD 123319 was a gift from
Dr Jack Hodges at Parke-Davis.
CONTROL
SARALASIN
Results
Effects of Angiotensin Peptides on [ 3 H]N£ Outflow
The spontaneous outflow of [3H]NE 2 minutes before
S! expressed as a fraction of the tissue content was
7.0±1.2xl0~ 3 (n=5) in the control group. This value
was not modified in the presence of either 3xlO~ 8 to
3 x l ( r 7 mol/L Ang-(l-7) or 3 x l f r 8 to 3xl0" 7 mol/L
Ang II (Fig 1). On the contrary, the overflow of tritium
per shock, which in controls was 15.2±4.8xlO~6 during
Si (n=5), was significantly increased by either 3xlO" 8 to
3xlO~ 7 mol/L Ang II or 3xlO" 8 to 3xl0~ 7 mol/L
Fra 3.
Bar graph shows effect of the angiotensin (ANG)
receptor antagonist saralasin on ANG II- and ANG-(1-7)-induced norepinephrine release in rat atria labeled with pH]norepinephrine. Two periods of nerve stimulation were applied (2
Hz, 2 minutes) 30 minutes apart. Tritium overflow was expressed
as the ratio of the second and first stimulation periods (S2/S!). Sj
was preceded by either 4 minutes of incubation with saralasin
(1 x10~ 7 mol/L, n=5) or 2 minutes of incubation with the peptides (1xio~ 7 mol/L) plus saralasin (n=5). The increase in
pHJnorepinephrine release induced by the peptides is shown in
Rg 2. Values are mean±SEM.
Downloaded from http://hyper.ahajournals.org/ by guest on March 3, 2014
Gironacci et al Increase of Norepinephrine Release by Angjotensin Peptides
459
induced release of [3H]NE in the presence of Ang-(l-7)
was prevented even by a 10 times lower concentration of
the AT2 receptor antagonist (lxlO" 7 mol/L), whereas
the effects of Ang II were not prevented by a concentration of this antagonist as high as lxlO~ 5 mol/L (data
not shown).
1.6
Discussion
CONTROL
AN6 II
plni
DUP 753
ANS (1-7)
plai
DUP 763
DOP 763
FIG 4. Bar graph shows effect of DuP 753 on angiotensin II
(ANG IIHnduced or ANG-(1-7)-induced norepinephrine release
in rat atria labeled with pHJnorepinephrine. Two periods of nerve
stimulation were applied (2 Hz, 2 minutes) 30 minutes apart.
Tritium overflow was expressed as the ratio of the second and
first stimulation periods (J^/S,). S2 was preceded by either 4
minutes of incubation with DuP 753 (1 x10~ 6 mol/L, n=5) or 2
minutes of incubation with the peptides (1 x 10" 7 mol/L) plus DuP
753 (n=6). The increase in pHJnorepinephrine release induced
by the peptides is shown in Fig 2. Values are mean±SEM.
When the effects of selective antagonists for AT! and
AT2 receptor subtypes were studied, it was found that
the AT, receptor antagonist DuP 753 ( l x l ( T 6 mol/L)
(Fig 4) reduced the increases in [3H]NE release caused
by Ang II and by Ang-(l-7). The reductions in the
[3H]NE release were 70% in both cases. The antagonist
per se was without effect on [3H]NE release in response
to nerve stimulation.
Unlike what was observed with the AT, receptor
antagonist, the AT2 receptor antagonist PD 123319
(lxlO" 6 mol/L) prevented by 70% the increase in
[3H]NE release elicited by lxlO" 7 mol/L Ang-(l-7) but
did not alter the release of the 3H transmitter in the
presence of 1 x 10"7 mol/L Ang II (Fig 5). The antagonist per se did not modify the [3H]NE release elicited
by nerve stimulation. Note that the nerve stimulated-
CONTROL
ANG II
plus
PD 123319
ANS (1-7)
plai
PD 123319
I
PD 123319
FIG 5. Bar graph shows effect of PD 123319 on angiotensin II
(ANG IIH n duced or ANG-(1-7)-induced norepinephrine release
in rat atria labeled with pH] norepinephrine. Two periods of nerve
stimulation were applied (2 Hz, 2 minutes) 30 minutes apart.
Tritium overflow was expressed as the ratio of the second and
first stimulation periods (J^/St). Sj was preceded by either 4
minutes of incubation with PD 123319 (1x10" 8 mol/L) or 2
minutes of incubation with the peptides (1 X 1 0 " 7 mol/L) plus PD
123319. The increase in pHJnorepinephrine release induced by
the peptides is shown in Fig 2. Values are mean±SEM. *P<.05
compared with control.
The present study shows that in isolated rat atria Ang
II and Ang-(l-7) were equally potent in increasing the
[3H]NE release evoked in response to nerve stimulation.
The observation that neither peptide modified the basal
efflux of [3H]NE probably indicates that they exert their
action at a presynaptic site. Moreover, because the
effects of the peptides were prevented by different
angiotensin receptor antagonists, it appears that this
presynaptic site could be an angiotensin receptor.
Despite the fact that Ang II has a high affinity for AT2
receptors,16 no AT2 receptors are involved for this
agonist but only for Ang-(l-7). The increase in [3H]NE
release by nerve stimulation caused by Ang II was
selectively prevented by the AT! receptor antagonist
DuP 753, suggesting that AT! receptors are involved.
Increases in norepinephrine release by nerve stimulation caused by Ang II through the specific interaction
with the AT, receptors have been reported in both in
vivo and in vitro studies. Among the in vivo effects of
Ang II are the results reported by Wong et al19 and
Suzuki et al,20 who found in anesthetized dogs that the
Ang II-induced enhancement of canine renal sympathetic nerve function results from an increase in norepinephrine release through the activation of AT, receptors. Similar results were also obtained in pithed rats
under spinal cord stimulation21 as well as in the
paraventricular nucleus of conscious rats submitted to
intracerebroventricular injections of pressor doses of
Ang II.22 Among the in vitro assays, a presynaptic
modulation by AT, receptors of norepinephrine release
elicited by field stimulation was found in superfused
slices of rat interscapular fat23 and in atria isolated from
guinea pigs.24
Despite the fact that, as reported for Ang II, Ang(1-7) is devoid of dipsogenic, pressor, or direct myotropic effects in rats and humans,910 the observation that it
is as potent as Ang II in increasing the release of
norepinephrine caused by nerve stimulation of the rat
atria (present results) is coincident with the findings
that the heptapeptide is as potent as Ang II in releasing
vasopressin from the rat hypothalamo-neurohypophysial system,7 producing neuronal excitation in rat
paraventricular neurons,5'6 eliciting cardiovascular effects injected into the dorsal medulla of rats,8 and
stimulating prostaglandin release from rabbit vas deferens11 and Q glioma cells,25 human astrocytes,26 and
porcine aortic endothelial cells.27 Nevertheless, our
results do not agree with those reported by Trachte et
al,11 who found that the noradrenergic neurogenic contractions of the rabbit vas deferens were potentiated by
Ang II but not by the heptapeptide.
The facilitatory action on the sympathetic neurotransmission elicited in the rat atria by both Ang II and
Ang-(l-7) appears to be mediated by stimulation of AT,
receptors. Moreover, the effect of Ang-(l-7) seems to
additionally involve the activation of AT2 receptors.
Identical biological activity caused by both peptides but
Downloaded from http://hyper.ahajournals.org/ by guest on March 3, 2014
460
Hypertension
Vol 24, No 4
October 1994
coupled to the activation of different angiotensin receptor subtypes has also been reported by Jaiswal et al28 for
the increase of prostaglandin synthesis caused by the
peptides in cultured human astrocytes, which was mediated through both ATj and AT2 receptors in the case
of Ang II and solely through AT2 receptors in the case
of Ang-(l-7). Hence, it appears that tissue and species
differences could be critical in determining either similarities or differences between the actions of both
peptides.
Acknowledgments
This work was supported by grant 339700092 from
CONICET, Argentina. This work is dedicated to the memory
of Dr M.A. Enero, to whom belongs the original idea for its
development. The excellent technical assistance of M. Ferrari
is gratefully acknowledged.
References
1. Ferrario CM, Brosnihan KB, Diz DI, JaiswaJ N, Khosla MC,
Milsted A, Tallant E A Angiotensin-(l-7): a new hormone of the
angiotensin system. Hypertension. 1991;18(suppl III):III-126III-133.
2. Ferrario CM, Barnes KL, Block CH, Brosnihan KB, Diz DI,
Khosla MC, Santos RAS. Pathways of angiotensin formation and
function in the brain. Hypertension. 199O,15(suppl I):I-13-I-19.
3. Welches WR, Santos RAS, ChappeU MC, Brosnihan KB, Greene
LJ, Ferrario CM. Prolyl endopeptidase participates in the processing of brain angiotensin. / Hypertens. 1991;9:631-638.
4. Chappell MC, Brosnihan KB, Diz DI, Ferrario CM. Identification
of angiotensin-(l-7) in rat brain. / Biol Chem. 1989;264:1651816523.
5. Felix D, Khosla MC, Barnes KL, Imboden H, Montani B, Ferrario
CM. Neurophysiological responses to angk>tensin-(l-7). Hypertension.
6. Barnes KL, Knowles WD, Ferrario CM. Angiotensin II and angiotensin (1-7) excite neurons in the canine medulla in vitro. Brain Res
Bull. 1990;24:275-280.
7. Schiavone MT, Santos RAS, Brosnihan KB, Khosla MC, Ferrario
CM. Release of vasopressin from the rat hypothalamoneurohypophysial system by angiotensin-(l-7) heptapeptide. Proc
NatlAcad Sci USA. 1988;85:4095-4098.
8. Campagnole-Santos MJC, Diz DI, Santos RAS, Khosla MC,
Brosnihan KB, Ferrario CM. Cardiovascular effects of
angiotensin-{l-7) injected into the dorsal medulla of rats. Am J
Physwl. 1989;257:H324-H329.
9. Fitzsimons JT. The effect on drinking of peptide precursors and of
shorter chain peptide fragments of angiotensin II injected into the
rat's diencephalon. J Physiol (Lond). 1971;214:295-303.
10. Kono T, Taniguchi A, Imura H, Oseko F, Khosla MC. Biological
activities of angiotensin II-(l-6)-hexapeptide and angiotensin
II-(l-7)-heptapeptide in man. Ufe Sci. 1986;38:1515-1519.
11. Trachte GJ, Meixner K, Ferrario CM, Khosla MC. Prostaglandin
production in response to angiotensin-(l-7) in rabbit isolated vasa
deferentia. Pwstaglandins. 1990;39:385-394.
12. Whitebread S, Mele M, Kamber B, de Gasparo M. Preliminary
biochemical characterization of two angiotensin II receptor
subtypes. Biochem Biophys Res Commun. 1989;163:284-291.
13. Chiu AT, Herblin WF, McCall DE, Ardecky RJ, Carini DJ,
Duncia JV, Pease LJ, Wong PC, Wexler RR, Johnson AL, et al.
Identification of angiotensin II receptor subtypes. Biochem Biophys
Res Commun. 1989;165:196-203.
14. Wong PC, Hart SD, Zaspel AM, Chiu AT, Ardecky RJ, Smith RD,
Timmermans PBMWM. Functional studies of nonpeptide angiotensin II receptor subtype-specific ligands: DUP 753 (AII-1) and
PD 123177 (AII-2). / Pharmacol Exp Ther. 1990;255:584-592.
15. Dudley DT, Panek RL, Major TC, Lu GH, Bruns RF, Klinkefus
BA, Hodges JC, Weishaar RE. Subclasses of angiotensin II binding
sites and their functional significance. Mol Pharmacol. 1990;38:
370-377.
16. Timmermans PBMWM, Wong PC, Chiu AT, Herblin WF,
Benfield P, Carini DJ, Lee RJ, Wexler RR, Saye JAM, Smith RD.
Angiotensin II receptors and angiotensin II receptor antagonists.
Pharmacol Rev. 1993;45:2O5-251.
17. Stewart JM, Young JD. Solid Phase Peptide Synthesis. Rockford, 111:
Pierce Chemical Company; 1984:76-80.
18. Adler-Graschinsky E, Langer SZ, Rubio MC. Metabolism of norepinephrine released by phenoxybenzamine in isolated guinea-pig
atria. / Pharmacol Exp Ther. 1972;180:286-301.
19. Wong PC, Hart SD, Timmermans PBMWM. Effect of angiotensin
II antagonism on canine renal sympathetic nerve function.
Hypertension. 1991;17:1127-1134.
20. Suzuki Y, Matsumura Y, Egi Y, Morimoto S. Effects of losartan, a
non-peptide angiotensin II receptor antagonist, on norepinephrine
overflow and antidiuresis induced by stimulation of renal nerves in
anesthetized dogs. / Pharmacol Exp Ther. 1992;263:956-963.
21. Wong PC, Bernard R, Timmermans PBMWM. Effect of blocking
angiotensin II receptor subtype on rat sympathetic nerve function.
Hypertension. 1992;19:663-667.
22. Stadler T, Veltmar A, Qadri F, Unger T. Angiotensin II evokes
noradrenaline release from the paraventricular nucleus in conscious rats. Brain Res. 1992^69:117-122.
23. Cassis LA, Dwoskin LP. Presynaptic modulation of neurotransmitter release by endogenous angiotensin II in brown adipose
tissue. J Neural Transm. 1991;34(suppl):129-137.
24. Brasch H, Sierolawski L, Dominiak P. Angiotensin II increases
norepinephrine release from atria by acting on angiotensin subtype
1 receptors. Hypertension. 1993;22:699-704.
25. Jaiswal N, Diz DI, Tallant EA, Khosla MC, Ferrario MC. Characterization of angiotensin receptors mediating prostaglandin synthesis in Q, glioma cells. Am J Physiol. 1991;260:R1000-R1006.
26. Tallant E A Jaiswal N, Diz DI, Ferrario CM. Human astrocytes
contain two distinct angiotensin receptor subtypes. Hypertension.
1991;18:32-39.
27. Jaiswal N, Diz DI, Chappell MC, Khosla MC, Ferrario CM. Stimulation of endothelial cell prostaglandin production by angiotensin
peptides. Hypertension. 1992;19(suppl II):II-49-II-55.
28. Jaiswal N, Tallant AE, Diz DI, Khosla MC, Ferrario MC. Subtype
2 angiotensin receptors mediate prostaglandin synthesis in human
astrocytes. Hypertension. 1991;17:1115-1120.
Downloaded from http://hyper.ahajournals.org/ by guest on March 3, 2014