Fax:Kluwer Health, 351 West Camden Street, Baltimore, MD 21202-2436. Phone: 410-528-4050. Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, a division of Wolters
http://hyper.ahajournals.org/subscriptions/Subscriptions: Information about subscribing to Hypertension is online at
at Monash University on August 12, 2010 hyper.ahajournals.orgDownloaded from
Role of Angiotensin II Type 1A Receptors in CardiovascularReactivity and Neuronal Activation After Aversive Stress
in MicePamela J. Davern, Daian Chen, Geoffrey A. Head, Carolina A. Chavez,
Thomas Walther, Dmitry N. Mayorov
Abstract—We determined whether genetic deficiency of angiotensin II Type 1A (AT1A) receptors in mice results in alteredneuronal responsiveness and reduced cardiovascular reactivity to stress. Telemetry devices were used to measure meanarterial pressure, heart rate, and activity. Before stress, lower resting mean arterial pressure was recorded in AT1A
�/�
(85�2 mm Hg) than in AT1A�/� (112�2 mm Hg) mice; heart rate was not different between groups. Cage-switch stress
for 90 minutes elevated blood pressure by �24�2 mm Hg in AT1A�/� and �17�2 mm Hg in AT1A
�/� mice (P�0.01),and heart rate increased by �203�9 bpm in AT1A
�/� and �121�9 bpm in AT1A�/� mice (P�0.001). Locomotor
activation was less in AT1A�/� (3.0�0.4 U) than in AT1A
�/� animals (6.0�0.4 U), but differences in blood pressureand heart rate persisted during nonactive periods. In contrast to wild-type mice, spontaneous baroreflex sensitivity wasnot inhibited by stress in AT1A
�/� mice. After cage-switch stress, c-Fos immunoreactivity was less in theparaventricular (P�0.001) and dorsomedial (P�0.001) nuclei of the hypothalamus and rostral ventrolateral medulla(P�0.001) in AT1A
�/� compared with AT1A�/� mice. Conversely, greater c-Fos immunoreactivity was observed in the
medial nucleus of the amygdala, caudal ventrolateral medulla, and nucleus of the solitary tract (P�0.001) of AT1A�/�
compared with AT1A�/� mice. Greater activation of the amygdala suggests that AT1A receptors normally inhibit the
degree of stress-induced anxiety, whereas the lesser activation of the hypothalamus and rostral ventrolateral medullasuggests that AT1A receptors play a key role in autonomic cardiovascular reactions to acute aversive stress, as well asfor stress-induced inhibition of the baroreflex. (Hypertension. 2009;54:1262-1268.)
Cardiovascular reactivity, a rapid sympathetically medi-ated increase in blood pressure in response to aversive
stress, is considered a risk factor for both hypertension andheart disease.1,2 To date, the central mechanisms that controlcardiovascular reactivity are not fully elucidated. However, itis plausible that the regulation of stress reactivity occurs atleast at 3 major central nervous system levels. The first levelincludes the formation of an emotional reaction to a stimulusby cortico-limbic structures. The second consists of the activa-tion of autonomic and endocrine outputs to the periphery byhypothalamic-brain stem circuits. The third involves suppressionof negative feedback signals from baroreceptors, which wouldotherwise effectively counteract cardiovascular activation.
Angiotensin II (Ang II) is increasingly recognized as animportant modulator of cardiovascular reactivity to stress atseveral central nervous system levels. First, Ang II is impli-cated in modulating anxiety3 and may thereby influence theintegration of emotional and behavioral reactions to aversivestimuli at the limbic level. Second, Ang II is criticallyinvolved, at the hypothalamic-brain stem level, in the regu-
lation of autonomic arousal associated with aversive events.4
In particular, we have shown recently that pharmacologicalblockade of Ang II type 1 (AT1) receptors in the rostralventrolateral medulla (RVLM) abolished the pressor responseto emotional stress,5 and inhibition of AT1 receptors in thedorsomedial hypothalamus (DMH) attenuates this response inrabbits.6 Third, Ang II may modulate afferent inputs frombaroreceptors and other peripheral receptors in the brain stemnucleus of the solitary tract (NTS)7 and, thus, influencefeedback autonomic control during stress. Finally, circulatingAng II may affect central pressor responses through receptorslocalized in brain circumventricular organs, which lack theblood-brain barrier.8 It is, thus, conceivable that central AngII is capable of multiple simultaneous actions on cardiovas-cular reactivity. In the rodent, there are 2 forms of the AT1
receptor, and the predominant subtype in the central nervoussystem appears to be the AT1A subtype.9 The development ofa specific AT1A receptor–deficient mouse offered us theopportunity to further explore the role of these receptors inmediating the cardiovascular response to acute stress.
Received July 23, 2009; first decision August 9, 2009; revision accepted October 7, 2009.From the Baker IDI Heart and Diabetes Institute (P.J.D., G.A.H.), Melbourne, Victoria, Australia; Departments of Physiology (D.C.) and Pharmacology
(C.A.C., D.N.M.), University of Melbourne, Victoria, Australia; Centre for Biomedical Sciences (T.W.), Hull York Medical School, University of Hull,Hull, United Kingdom; Excellence Cluster Cardio-Pulmonary System (T.W.), Justus-Liebig-Universität Giessen, Giessen, Germany.
Correspondence to Geoffrey A. Head, Baker IDI Heart and Diabetes Institute, Commercial Rd, Prahran, PO Box 6492, St. Kilda Road Central,Melbourne, Victoria 8008, Australia. E-mail [email protected]
In the present study, we examined the effect of AT1A
receptor deficiency on cardiovascular reactivity and neuronalactivation, as detected by c-Fos immunohistochemistry, inresponse to acute aversive stress in mice. We used cage-switch, which is an olfactory-mediated psychosocial stressmodel that has been shown previously to produce a sustained90-minute–long cardiovascular arousal and to increase inlocomotor activity in mice.10 We hypothesized that alteredcardiovascular reactivity in AT1A
�/� mice may also relate tothe differential responsiveness to stress in one or several ofthe aforementioned brain structures and sought to clarify theirfunctional importance in the actions of Ang II as a stressmediator. We used a telemetry monitoring system to measurecardiovascular changes in conscious freely behaving animalsand also to take into account confounding effects of locomo-tion on cardiovascular function.11
Methods
AnimalsExperiments were conducted in conscious AT1A
�/� mice (n�18)and AT1A
�/� mice (n�18), and the generation of these mice hasbeen described earlier.12 Ten animals were used for the cardiovas-cular responses, and 2 of these were included in the c-Fos experi-ments, with an additional 8 animals to make a total of 10 (n�5 forstress and n�5 no stress). These mice were housed at the Baker IDIHeart and Diabetes Institute and were given ad libitum access to foodand water. Animals were kept on a 12:12 hour light-dark cycle (6 AM
to 6 PM light). The experiments were approved by the Alfred MedicalResearch Education Precinct Animal Ethics Committee and con-ducted in accordance with the Australian Code of Practice forScientific Use of Animals.
implanted with a PA-C10 telemetry device (Data Sciences Interna-tional), with the catheter inserted into the carotid artery and thetransmitter body implanted along the right flank.13 After �1 week ofrecovery, recordings of systolic arterial pressure, diastolic arterialpressure, calculated mean arterial pressure (MAP), calculated heart rate(HR), and locomotor activity were measured continuously at rest andthroughout the stress protocol. Signals were sampled at 1000 Hz usingan analog-to-digital data acquisition card (National Instruments 6024E),as described previously.14 Cage-switch stress was conducted over a90-minute period by removing mice from their original cage and placingthem into a cage occupied previously by a different male mouse. Asecond analysis was performed to calculate MAP and HR during thestress when the activity signal was below the “no-movement” thresholdfor �6 seconds. Some of the AT1A
�/� and AT1A�/� mice subjected to
90-minute cage-switch stress (n�5 per group) or no stress (left in theiroriginal cages for 90 minutes; n�5 per group) were immediatelyperfused and their brains removed for immunohistochemical analysis ofbrain sites recognized as being rich in AT1 receptors.15
Cardiovascular Reactivity to Locomotor ActivityFor each mouse, least-squares regression slopes for the relationshipsbetween MAP and activity and HR and activity were calculatedduring the 90 minutes of stress with 5-minute bins.
Assessment of Baroreflex by Sequence TechniqueSee the online Data Supplement, available at http://hyper.ahajournals.org.
c-Fos Immunohistochemistry and AnalysisSee the online Data Supplement.
Statistical AnalysisCardiovascular data were analyzed by 2-way (repeated-measure)ANOVA to determine the effects of an AT1A receptor deficiency andstress on cardiovascular parameters and locomotion. Values were
Time (min)
Activ
ity (u
nits
)
0
5
10
Time (min)0-20 0-90 21-90
∆ Activity (units)
+0
+5
MA
P (m
mH
g)
100
150
HR
(bea
ts/m
in)
400
600
800
∆HR (beats/min)
+0
+100
+200
+300
∆MAP (mmHg)
+0
+20
+40
**
**
***
***
**
***
***
0 30 60 90
Figure 1. Line graphs represent the MAP,HR, and locomotor activity responsesbetween AT1A
�/� mice (F; n�10) andAT1A
�/� mice (E; n�10) during preoccu-pied cage-switch stress. Each dot repre-sents the mean value averaged across a5-minute period. Bar graphs representaverage changes in MAP, HR, and loco-motor activity in AT1A
�/� (f) and AT1A�/�
(�) mice in response to stress. Theseaverage responses were calculated over90 minutes of stress exposure and a15-minute control period in each animal.Values are mean�SEM. **P�0.01 and***P�0.001 vs AT1A
�/� mice.
Davern et al Effect of Stress in AT1A Receptor–Deficient Mice 1263
at Monash University on August 12, 2010 hyper.ahajournals.orgDownloaded from
expressed as mean�SEM or mean difference�SE of the difference.Statistical evaluation of c-Fos counts was performed by 1-way multi-factor (strain, stress) ANOVA, determining the effect of strains withinfactors of stress and no stress, as well as the overall effect of stressacross groups. Values were considered significant when P�0.05.
Results
Cardiovascular and Locomotor ResponsesFigure 1 illustrates the values for MAP, HR, and locomotoractivity during the basal and experimental periods. Beforestress exposure, lower resting MAP was recorded in AT1A
�/�
mice (85�2 mm Hg) than in AT1A�/� mice (112�2 mm Hg),
whereas HR levels and locomotor activity were not differentbetween groups. Cage-switch stress elicited prompt andsustained pressor and tachycardia responses, as well aslocomotor activation, in AT1A
�/� mice (�24�2 mm Hg,�203�9 bpm, and �5.5�0.6 U, respectively, averaged over90 minutes; n�10). The MAP, HR, and locomotor activityincreased by �17�2 mm Hg, �121�9 bpm, and 2.5�0.4 U,respectively, averaged over the 90-minute period in AT1A
�/�
mice (n�10), which was less than that observed in AT1A�/�
mice (P�0.01). However, when expressed as a percentage ofbasal levels, the pressor responses (�90 minutes) wereidentical in both strains (21.6�1.4 versus 21.1�2 mm Hg inAT1A
�/� and AT1A�/� mice, respectively). Comparing only
the initial 20-minute period, the changes in blood pressure andactivity were similar in both strains, but the tachycardia responseto cage-switch stress was attenuated in AT1A
�/� mice by 25.3%(P�0.001).
A further analysis of selected periods of little activity wasdetermined by a minimum activity threshold (Figure 2). Averagevalues for activity during stress were reduced to 0.07�0.04 and0.05�0.01 U over the 90-minute period in AT1A
�/� andAT1A
�/� mice, respectively, and represented 1.2% and 2.0% ofthe normal activity level. During these still periods, the pressorand tachycardia responses to cage-switch stress in AT1A
�/�
mice were 57% and 66% of those observed in AT1A�/� mice,
respectively (P�0.01; Figure 2; n�8 per group).
Relationship Between Locomotor Activityand MAPThere was a positive relationship between activity and MAPduring cage-switch stress, which was similar between AT1A
�/�
(slope: 4.2�1.0 mm Hg per unit of activity; r�0.52; n�10) andAT1A
�/� mice (slope: 5.7�0.6 mm Hg per unit of activity;r�0.63; n�10; P�0.25 for difference between strains). Simi-larly, the HR relationship with activity during cage-switch stresswas also similar in both strains (AT1A
�/� slope:24.8�5.7 mm Hg per unit of activity, r�0.62, n�10; AT1A
�/�
mice slope: 32.1�4.2 mm Hg per unit of activity, r�0.79,n�10).
Spontaneous Baroreceptor Reflex SensitivityBaroreflex sensitivity as assessed as the slope of spontaneousup and down sequences was similar in AT1A
�/� (n�9)compared with AT1A
�/� mice (n�10). The number ofbaroreflex-dependent sequences and the baroreflex activationindex (percentage of baroreflex to total number of sequences)was also similar (Figure 3). Cage-switch stress was accom-
Time (min)
0 30 60 90
Activ
ity (u
nits
)
0
1
2
Time (min)
0-20 0-90 20-90
∆ ACT (units)
0
1
2
MAP
(mm
Hg)
100
150
HR
(bea
ts/m
in)
400
600
800
∆HR (beats/min)
+0
+100
+200
+300
∆MAP (mmHg)
+0
+20
+40
*
**
***
**
******
Figure 2. Line graphs represent the MAPand HR responses when locomotor activ-ity is at or below a nominal thresholdbetween AT1A
�/� mice (F; n�8) andAT1A
�/� mice (E; n�8) during preoccu-pied cage-switch stress. Each dot repre-sents the mean value averaged across a5-minute period. Bar graphs representaverage changes in MAP, HR, and loco-motor activity in AT1A
�/� (f) and AT1A�/�
(�) mice in response to stress. Theseaverage responses were calculated over90 minutes of stress exposure and a15-minute control period in each animal.Values are mean�SEM. **P�0.01,***P�0.001 vs AT1A
�/� mice.
1264 Hypertension December 2009
at Monash University on August 12, 2010 hyper.ahajournals.orgDownloaded from
panied by a marked decrease in baroreflex sensitivity ofAT1A
�/� mice, as well as a marked increase in the number ofbaroreflex sequences (P�0.05). Conversely, AT1A
�/� miceexhibited no baroreceptor reflex gain inhibition during stress,nor did they increase the number of sequences per minuteduring stress (Figure 3). Thus, the level of baroreflex sensi-tivity during stress was �69% greater in AT1A
�/� comparedwith AT1A
�/� mice (average of up and down sequences,P�0.0004). Stress did not alter the baroreflex activationindex in either group (Figure 3). A control experiment wasalso performed with no cage-switch or disturbance thatshowed no changes to baroreflex gain, numbers of sequences,or baroactivation index (Figure 4; n�10 per group).
c-Fos ImmunohistochemistryCage-switch stress was accompanied, in AT1A
�/� mice, byan increase in c-Fos expression in all of the brain regions
studied, except the subfornical organ (Table). This stress-induced increase in c-Fos levels was reduced by 73% in thecentral nucleus of the amygdala (CeAm; Figure 5A and 5B),65% in the raphe pallidus, 47% in the paraventricular nucleus(PVN; Figure 5C and 5D), 44% in the RVLM (P�0.001), andby 69% in the bed nucleus of the stria terminalis, and 47% inthe DMH (Figure 5E and 5F; P�0.001) in AT1A
�/� mice(n�5) compared with AT1A
�/� mice (n�5; Table). Con-versely, c-Fos expression after cage-switch stress was greaterby 126% in the NTS, 99% in the medial nucleus of theamygdala (MeAm; Figure 5G and 5H), 53% in the caudalventrolateral medulla (P�0.001), and by 85% in the organumvasculosum of the lamina terminalis (P�0.02) in AT1A
�/�
mice compared with AT1A�/� mice (Table). c-Fos expression
was also observed in other brain regions recognized as beingrich in AT1A receptors,15 including the median preopticnucleus in the forebrain and area postrema in the hindbrain;but cell counts were quite low, and we did not observe a
Dow
n sl
ope
b/m
in/m
mH
g-15
-10
-5
0
Up
slop
eb/
min
/mm
Hg
-15
-10
-5
0
Num
ber /
min
Dow
n se
quen
ces
0
1
2
Num
ber /
min
up s
eque
nces
0
1
2
**
rest stress rest stress
Bar
o Ac
tivat
ion
Inde
x
0.00
0.25
0.50
0.75
***
*
AT1A +/+ AT1A -/-
*
*
**
Figure 3. Average baroreflex gain (bpm per millimeter of mer-cury) is determined from the slope from down and upsequences (top). The total number of sequences (middle) andbaroreflex activation index (number of baroreflex-relatedsequences as the percentage of total sequences) in AT1A
�/�
mice (left; n�10) and AT1A�/� mice (right; n�9) before and dur-
ing cage-switch stress. *P�0.05 for effect of stress.
Dow
n sl
ope
b/m
in/m
mH
g
-15
-10
-5
0
Up
slop
eb/
min
/mm
Hg
-15
-10
-5
0
Num
ber /
min
Dow
n se
quen
ces
0.0
0.5
1.0
1.5
Num
ber /
min
Up
sequ
ence
s
0.0
0.5
1.0
1.5
rest No Stress rest No Stress
Bar
o A
ctiv
atio
nIn
dex
0.00
0.25
0.50
0.75
AT1A -/-AT1A +/+
Figure 4. Average baroreflex gain (bpm per millimeter of mer-cury) is determined from the slope from down and upsequences (top). The total number of sequences (middle) andbaroreflex activation index (number of baroreflex relatedsequences as the percentage of total sequences) in AT1A
�/�
mice (left; n�10) and AT1A�/� mice (right; n�10) before and
during a period of no stress.
Davern et al Effect of Stress in AT1A Receptor–Deficient Mice 1265
at Monash University on August 12, 2010 hyper.ahajournals.orgDownloaded from
significant difference between groups for these regions (Ta-ble). There was little c-Fos expression in most brain regionsof control unstressed animals in either group (n�5 pergroup). Activation was observed in the DMH, but this wasone tenth of the effect of cage-switch stress.
DiscussionThe major findings from the present study were that thesustained pressor, tachycardic, and locomotor activity re-sponses to a 90-minute cage-switch stress were attenuated inAT1A
�/� compared with normal AT1A�/� mice together with
a reduced number of c-Fos–stained cells in specific brainregions, including the DMH, RVLM, PVN, and raphe palli-dus. These regions are known to be important for theexpression of the autonomic responses to stress. Together,these findings suggest that activation of AT1A receptors,presumably by Ang II, normally facilitate the autonomicmanifestations through the hypothalamic-brain stem path-ways. This suggests that the lesser pressor response to stressin AT1A
�/� is not attributed to a “nonspecific” effect of lowerblood pressure or to possible differences in spinal16 organglionic17 transmission or peripheral vascular reactivity.
We observed a 2-fold greater level of c-Fos immunoreactivityin AT1A
�/� mice in the MeAm, which is a critical region knownto be activated by an aversive olfactory stimulus in the rodent18
and also for mediating the pressor responses to stress.19 Thus, thestimulus may have even been greater in the AT1A
�/� mice.Despite the higher c-Fos–positive cell counts in the MeAm, thepressor, tachycardic, and also locomotor responses were lessin the AT1A
�/� mice. Therefore, an even greater attenuationmight be expected with an equal level of emotional response.
Our study provides evidence for a link between the physio-logical response to stress and the activation of the MeAm andDMH involving the AT1A receptor. However, we cannot deter-mine from our study whether this is a direct effect of the absenceof receptors in the region or an indirect one from an altered input
to these regions. Furthermore, the c-Fos technique is limited inthat it only detects neurons that are activated and not those thatare inhibited. There is electrophysiological evidence for anangiotensinergic projection from the MeAm to the anteriorhypothalamic area that involves AT1 receptors20; however, weare not aware of whether a projection to the DMH has beenstudied. Contrasting the findings in the MeAm, we observedlesser c-Fos expression in the CeAm and bed nucleus of the striaterminalis, which are the only regions in the basal forebrain thathave been shown to respond to an increase in blood pressure.21
Thus, the lower c-Fos counts in these regions might reflect thelesser overall increase in blood pressure and HR in the AT1A
receptor–deficient mice. Although the c-Fos counts are likely toreflect a somewhat greater contribution from the first 20-minuteperiod when there was little difference in the rise in bloodpressure between strains, there is still a significant contributionfrom the later times when a difference was observed.
Previous studies have shown an increase in blood pressureduring experimentally induced acute aversive stress in severalspecies, including mice,14 rats,22,23 and rabbits.5,6,24 The DMHhas been described as the core of the hypothalamic defense areabecause of its essential involvement in regulating the cardiovas-cular responses to stress.6,24 In addition, it is the common outputcenter both for the vasoconstrictor (via the RVLM) and for theexpression of the tachycardia via spinally projecting raphepallidus neurons receiving inputs from the DMH.25 Further-more, the DMH has high levels of AT1 receptor binding andAT1A receptor mRNA,26 which makes it a most likely site forthe effect of AT1A receptor deletion on the cardiovascularresponsivity to aversive stress. Our recent findings suggest animportant role for AT1A receptors in the DMH, becausemicroinjections of candesartan into the DMH dose-dependently attenuated the hypertensive effect of aversivestress in rabbits.24 Like the DMH, the RVLM is rich in AT1
receptors,27 and functional studies have identified that sym-pathetic premotor neurons in the RVLM are heavily inner-
Table. No. of c-Fos–Labeled Neurons per Section in Nonstressed and Stressed Mice
Mean No. of activated neurons, as detected by c-Fos-immunoreactivity, in nonstressed mice (left) and mice exposed to 90-mins of cage-switch stress (right). CVLMindicates caudal ventrolateral medulla; OVLT, vascular organ of the lamina terminalis; RPa, raphe pallidus; BST, bed nucleus of the stria terminalis; AP, area postrema;MnPO, median preoptic nucleus; SFO, subfornical organ; NS, not significant.
1266 Hypertension December 2009
at Monash University on August 12, 2010 hyper.ahajournals.orgDownloaded from
vated by excitatory inputs arising from the DMH.28 Earlier,we showed that blockade of AT1 receptors in the RVLM ofrabbits attenuates pressor responses induced by aversive (air-jet)stress. This is further supported by other studies that showed thatICV administration of the angiotensin peptide antagonist sarala-sin inhibited the pressor response to restraint stress in rats.29 Inthe present study, reduced cardiovascular reactivity (pressorresponse and tachycardia) and attenuated neuronal activation inthe DMH and RVLM were observed in AT1A
�/� mice com-pared with AT1A
�/� mice. Although we cannot be certain wherethe lack of AT1 receptors is affecting the cardiovascular re-sponses, our studies are consistent with an important role forcentral Ang II and AT1A receptors in modulating the response toacute aversive stress probably at the level of the hypothalamusand the brain stem.
Cage-switch responses involve not only pressor responsesand tachycardia but also a marked increase in locomotoractivity, which may complicate the interpretation of compar-isons between strains. We have performed a further analysisthat selected periods of no appreciable movement and showedattenuated pressor and tachycardic responses in AT1A
�/�
mice over the entire 90-minute period. These findings, inconjunction with other studies where we have shown that
restraint stress produced a similar degree of reduced pressorresponses in AT1A
�/� mice with little change in activity (themice were restrained), suggest that locomotor activity is notconfounding the current findings.30
Our results are also relevant to the autonomic regulationby the baroreflex, because greater c-Fos expression wasobserved in the caudal ventrolateral medulla and NTSaccompanied by an absence of stress-induced baroreflexinhibition in AT1A
�/� mice compared with AT1A�/� mice.
These brain regions are important for baroreflex regula-tion, and the greater c-Fos expression quantitatively re-flects a difference in baroreflex responsivity betweenstrains during stress. Previous studies have demonstratedthat endogenously formed Ang II acts in the NTS to releaseNO, which leads to reduced neuronal activity and inhibi-tion of the baroreflex pathways.7,31 These data suggest thatactivation of AT1A receptors, possibly in the NTS, mayplay a key role in the stress-induced inhibition of the HRbaroreflex in mice. Greater neuronal activation in thecaudal ventrolateral medulla of AT1A
�/� mice, a majorrelay site in baroreflex transmission, could further contrib-ute to the lack of baroreflex function during stress in theseanimals. No differences in baroreflex gain were observed
BA C
D FE
HG
Figure 5. Photomicrographs of coronal sections through the CeAm (A and B), PVN (C and D), DMH (E and F), and MeAm (G and H) ofAT1A
�/� mice (n�5; A, C, E, and G) and AT1A�/� mice (n�5; B, D, F, and H) after 90 minutes of cage-switch stress. c-Fos immunore-
activity can be seen throughout many nuclei in the CeAm, PVN, and DMH in AT1A�/� mice (A, C, and E) with less activation in AT1A
�/�
mice in each of these brain regions (B, D, and F). By contrast, greater c-Fos expression can be seen in the MeAm of AT1A�/� mice (H)
compared with AT1A�/� mice (G). Scale bar: 100 �m.
Davern et al Effect of Stress in AT1A Receptor–Deficient Mice 1267
at Monash University on August 12, 2010 hyper.ahajournals.orgDownloaded from
in AT1A�/� mice at rest suggesting that, under normal
conditions, there is little activation of AT1A receptors.
PerspectivesOur study using genetic deletion of AT1A receptors in micesuggests that brain Ang II is an important modulator ofpressor, cardiac, and locomotor responses to acute aversivestress. Although our study does not specify the precise role ofthese receptors at the level of individual nuclei, the differen-tial pattern of c-Fos between the strains suggests that AT1A
receptors may play several discrete roles. We propose that,under conditions of acute stress, AT1A receptors in the limbicsystem or cortex may play an important role in suppressingthe fear/anxiety response but may act within limbic andhypothalamic nuclei to facilitate both cardiovascular andlocomotor activation. Because both HR and pressor responseswere similarly affected by the absence of AT1A receptors, theregions likely responsible are the DMH and RVLM, both ofwhich are known to be important for the regulation of theseresponses. We speculate that AT1 receptor activation may,therefore, reinforce the cardiovascular adaptive reactions ofincreasing cardiac output and directing blood to muscles forflight or fight. In addition, Ang II can also limit the negativefeedback of the pressor response by the baroreflex that would tryto counteract the rise in pressure, possibly by modulating thepathway from the hypothalamus to the NTS. Conversely, AT1A
receptors in circumventricular organs appear to play a limitedrole in the acute stress-induced cardiovascular activation.
AcknowledgmentsWe thank Pascal Carrive for his sage advice on the article and LuisaLa Greca and Thu-Phuc Nguyen-Huu for their very abletechnical support.
Sources of FundingThis work was funded by project grants from the National Health andMedical Research Council, Australia (472680 and 472652).
DisclosuresNone.
References1. Light KC, Girdler SS, Sherwood A, Bragdon EE, Brownley KA, West
SG, Hinderliter AL. High stress responsivity predicts later blood pressureonly in combination with positive family history and high life stress.Hypertension. 1999;33:1458–1464.
2. Rozanski A, Blumenthal JA, Kaplan J. Impact of psychological factors onthe pathogenesis of cardiovascular disease and implications for therapy.Circulation. 1999;99:2192–2217.
3. Saavedra JM, Ando H, Armando I, Baiardi G, Bregonzio C, Jezova M,Zhou J. Brain angiotensin II, an important stress hormone: regulatorysites and therapeutic opportunities. Ann N Y Acad Sci. 2004;1018:76–84.
4. Watanabe T, Fujioka T, Hashimoto M, Nakamura S. Stress and brainangiotensin II receptors. Crit Rev Neurobiol. 1998;12:305–317.
5. Mayorov DN, Head GA. AT1-receptors in the RVLM mediate pressorresponses to emotional stress in rabbits. Hypertension. 2003;41:1168–1173.
6. De Matteo R, Head GA, Mayorov DN. Tempol in the dorsomedialhypothalamus attenuates the hypertensive response to stress in rabbits.Am J Hypertens. 2006;19:396–402.
7. Wong LF, Polson JW, Murphy D, Paton JF, Kasparov S. Genetic andpharmacological dissection of pathways involved in the angiotensin II-mediateddepression of baroreflex function. FASEB J. 2002;16:1595–1601.
8. McKinley MJ, Allen AM, Burns P, Colvill LM, Oldfield BJ. Interactionof circulating hormones with the brain: the roles of the subfornical organ
and the organum vasculosum of the lamina terminalis. Clin ExpPharmacol Physiol. 1998;25(suppl):S61–S67.
9. Oldfield BJ, Allen AM, Hards DK, Mckinley MJ, Schlawe I, MendelsohnF. Distribution of angiotensin II receptor binding in the spinal cord of thesheep. Brain Res. 1994;650:40–48.
10. Lee DL, Webb RC, Brands MW. Sympathetic and angiotensin-dependenthypertension during cage-switch stress in mice. Am J Physiol RegulIntegr Comp Physiol. 2004;287:R1394–R1398.
11. Leary AC, Donnan PT, MacDonald TM, Murphy MB. The influence ofphysical activity on the variability of ambulatory blood pressure. Am JHypertens. 2000;13:1067–1073.
12. Ito M, Oliverio MI, Mannon PJ, Best CF, Maeda N, Smithies O, CoffmanTM. Regulation of blood pressure by the type 1a angiotensin II receptorgene. Proc Natl Acad Sci U S A. 1995;92:3521–3525.
13. Butz GM, Davisson RL. Long-term telemetric measurement of cardio-vascular parameters in awake mice: a physiological genomics tool.Physiol Genomics. 2001;5:89–97.
14. Jackson K, Head GA, Morris BJ, Chin-Dusting J, Jones E, La Greca L,Mayorov DN. Reduced cardiovascular reactivity to stress but not feedingin renin enhancer knockout mice. Am J Hypertens. 2007;20:893–899.
15. Phillips MI, Shen L, Richards EM, Raizada MK. Immunohistochemicalmapping of angiotensin AT1 receptors in the brain. Regul Pept. 1993;44:95–107.
16. Suter C, Coote JH. Intrathecally administered angiotensin II increasessympathetic activity in the rat. J Auton Nerv Syst. 1987;19:31–37.
17. Ma XY, Sigmund CD, Hingtgen SD, Tian X, Davisson RL, Abboud FM,Chapleau MW. Ganglionic action of angiotensin contributes to sympa-thetic activity in renin-angiotensinogen transgenic mice. Hypertension.2004;43:312–316.
18. Dielenberg RA, Hunt GE, McGregor IS. “When a rat smells a cat”: thedistribution of Fos immunoreactivity in rat brain following exposure to apredatory odor. Neuroscience. 2001;104:1085–1097.
19. Kubo T, Okatani H, Nishigori Y, Hagiwara Y, Fukumori R, Goshima Y.Involvement of the medial amygdaloid nucleus in restraint stress-inducedpressor responses in rats. Neurosci Lett. 2004;354:84–86.
20. Hagiwara Y, Nishigori Y, Fukumori R, Kubo T. The medial amygdaloidarea is involved in activation of angiotensin II-sensitive neurons in theanterior hypothalamic area. Brain Res. 2005;1033:128–134.
21. Li Y-W, Dampney RAL. Expression of fos-like protein in brain followingsustained hypertension and hypotension in conscious rabbits. Neuroscience.1994;61:613–634.
22. McDougall SJ, Lawrence AJ, Widdop RE. Differential cardiovascularresponses to stressors in hypertensive and normotensive rats. Exp Physiol.2005;90:141–150.
23. Kanbar R, Orea V, Barres C, Julien C. Baroreflex control of renalsympathetic nerve activity during air-jet stress in rats. Am J Physiol RegulIntegr Comp Physiol. 2007;292:R362–R367.
24. De Matteo R, Head GA, Mayorov DN. Angiotensin II in dorsomedialhypothalamus modulates cardiovascular arousal caused by stress but notfeeding in rabbits. Am J Physiol Regul Integr Comp Physiol. 2006;290:R257–R264.
25. Samuels BC, Zaretsky DV, DiMicco JA. Tachycardia evoked by disin-hibition of the dorsomedial hypothalamus in rats is mediated throughmedullary raphe. J Physiol. 2002;538:941–946.
26. Allen A, Oldfield B, Giles M, Paxinos G, McKinley M, Mendelsohn F.Location of angiotensin receptors in the nervous system. In: Quirion A,Bjorklund A, Hokfelt T, eds. Handbook of Chemical Neuroanatomy.Amsterdam, Netherlands: Elsevier; 2000:79–124.
27. Mendelsohn FAO, Allen AM, Clevers J, Denton DA, Tarjan E, McKinleyMJ. Localization of angiotensin II receptor binding in rabbit brain in vitroautoradiography. J Comp Neurol. 1988;270:372–384.
28. Horiuchi J, McAllen RM, Allen AM, Killinger S, Fontes MA, DampneyRA. Descending vasomotor pathways from the dorsomedial hypothalamicnucleus: role of medullary raphe and RVLM. Am J Physiol Regul IntegrComp Physiol. 2004;287:R824–R832.
29. Saiki Y, Watanabe T, Tan N, Matsuzaki M, Nakamura S. Role of centralANG II receptors in stress-induced cardiovascular and hyperthermicresponses in rats. Am J Physiol. 1997;272:R26–R33.
30. Chen D, La Greca L, Head GA, Walther T, Mayorov D. Blood pressurereactivity to emotional stress is reduced in AT(1A)-receptor knockout miceon normal, but not high salt intake. Hypertension Res. 2009;32:559–564.
31. Matsumura K, Averill DB, Ferrario CM. Angiotensin II acts at AT1receptors in the nucleus of the solitary tract to attenuate the baroreceptorreflex. Am J Physiol. 1998;275:R1611–R1619.
1268 Hypertension December 2009
at Monash University on August 12, 2010 hyper.ahajournals.orgDownloaded from
Online Supplemental Short title: The effect of stress in AT1A receptor deficient mice
THE ROLE OF AT1A RECEPTORS IN
CARDIOVASCULAR REACTIVITY AND NEURONAL ACTIVATION FOLLOWING AVERSIVE STRESS IN MICE
Pamela J. Davern1, Daian Chen2, Geoffrey A. Head1*,
Carolina A. Chavez3, Thomas Walther4 and Dmitry N. Mayorov3
1Baker IDI Heart & Diabetes Institute, Melbourne, Victoria, Australia 2Dept of Physiology, University of Melbourne, Victoria, Australia 3Dept of Pharmacology, University of Melbourne, Victoria, Australia 4Centre for Biomedical Sciences, Hull York Medical School, University of Hull, UK
7 October, 2009
*Corresponding author: Geoffrey A. Head Baker IDI Heart & Diabetes Institute Commercial Road, Prahran, P.O. Box 6492, St Kilda Road Central, Melbourne, Victoria, 8008, Australia Phone 61 3 8532 1332 Fax 61 3 8532 1100 Email: [email protected]
at Monash University on August 12, 2010 hyper.ahajournals.orgDownloaded from
Materials and Methods Assessment of Baroreflex by Sequence Technique
The spontaneous baroreceptor reflex slope was calculated as the slope of the regression lines between MAP and HR, with a delay of 0-6 heartbeats, using the sequence technique.1, 2 The number of baroreflex sequences compared with non-baroreflex sequences was highest with a four-beat delay in both groups and therefore used to calculate baroreflex gain. Sequences with at least three intervals (>0.3 mmHg MAP changes, >1 beat/min HR changes) were analysed only if the correlation coefficients were >0.85. Spontaneous baroreflex slope was calculated as the mean value of the significant slopes obtained. A third order polynomial smoothing function was used to reduce the contribution from respiration.3 Baroreflex activation index was calculated as the ratio of baroreflex type sequences (+/- and -/+) divided by the total number of sequences.4 c-Fos Immunohistochemistry
Mice were deeply anaesthetised with an intraperitoneal injection of sodium pentobarbitone (100 mg/kg). The animals were perfused transcardially with 20 ml of 0.9 % saline and 60 ml of 4 % paraformaldehyde dissolved in 0.1 M phosphate buffer, pH 7.2 (PB). Subsequently, the brain was removed and postfixed for 1 hour in 20 % sucrose in paraformaldehyde, and placed in 20 % sucrose in PB and refrigerated overnight at approximately 4 °C. Coronal sections (40 μm) were cut on a cryostat and placed in PB. Free-floating sections were incubated in 10 % normal horse serum at room temperature for 1 hour. Sections were then incubated in primary antibody, sheep anti-c-Fos (Chemicon) diluted 1:2000 in a solution of 2 % normal horse serum and 0.3 % Triton X-100 (Sigma) in PB at room temperature overnight. Sections were washed in PB prior to incubation in biotinylated donkey anti-sheep immunoglobins (1:200, Jackson) in PB containing 2 % normal horse serum for 1 hour. Thereafter, the sections were washed and incubated in avidin-biotin peroxidase complex (1:100, Vector) in PB for 1 hour. Following washes in 0.05 M Tris buffer (pH 7.6), sections were incubated in a solution of 40 mg nickel ammonium sulphate and 50 mg 3-3′ diaminobenzidine hydrochloride per 100 ml Tris buffer for 10 min, 15 μl of 30 % hydrogen peroxide was added for a further 6 min. Following final washes sections were mounted on gelatine coated microscope slides.
c-Fos Immunohistochemical Analysis Bright-field illumination using a Motic BA400 microscope and Motic images plus 2.0 were used to assess sections that exhibited c-Fos-immunoreactivity as detected by black stained nuclei. Brain sites examined include the vascular organ of the lamina terminalis (OVLT), median preoptic nucleus (MnPO), bed nucleus of the stria terminalis (BST), subfornical organ (SFO), paraventricular nucleus of the hypothalamus (PVN), central nucleus of the amygdala (CeAm), medial nucleus of the amygdala (MeAm), DMH, raphe pallidus (RPa), RVLM, caudal ventrolateral medulla (CVLM), NTS and area postrema. The mouse brain atlas5 was utilized in a blind analysis of up to four sections per animal for each brain site and counts of total black stained nuclei within the known boundaries of the nucleus were recorded.
at Monash University on August 12, 2010 hyper.ahajournals.orgDownloaded from
Comparing spectral and invasive estimates of baroreflex gain. IEEE Eng Med Biol Mag. 2001;20:43-52.
2. Adams DJ, Head GA, Markus MA, Lovicu FJ, van der Weyden L, Köntgen F,
Arends MJ, Thiru S, Mayorov DN, Morris BJ. Renin enhancer is critical for regulation of renin gene expression and control of cardiovascular function. J Biol Chem. 2006;281:31753-31761.
3. Oosting J, Struijker-Boudier HA, Janssen BJ. Validation of a continuous
baroreceptor reflex sensitivity index calculated from spontaneous fluctuations of blood pressure and pulse interval in rats. J Hypertens. 1997;15:391-399.
4. Parati G, DiRienzo M, Bonsignore MR, Insalaco G, Marrone O, Castiglioni P,
Bonsignore G, Mancia G. Autonomic cardiac regulation in obstructive sleep apnea syndrome: evidence from spontaneous baroreflex analysis during sleep. J Hypertens. 1997;15:1621-1626.
5. Paxinos G, Franklin KBJ: The mouse brain in stereotaxic coordinates. San
Diego, California, Academic Press, 2001.
at Monash University on August 12, 2010 hyper.ahajournals.orgDownloaded from
