Am J Physiol Heart Circ Physiol 307: H182–H190, 2014. First published May 16, 2014; doi:10.1152/ajpheart.00518.2013.

Overexpression of angiotensin-converting enzyme 2 attenuates tonically active glutamatergic input to the rostral ventrolateral medulla in hypertensive rats Yang-Kai Wang,1* Du Shen,1* Qiang Hao,2* Qiang Yu,3 Zhao-Tang Wu,1 Yu Deng,3 Yan-Fang Chen,4 Wen-Jun Yuan,3 Qi-Kuan Hu,3 Ding-Feng Su,1 and Wei-Zhong Wang1 1

Department of Physiology, Second Military Medical University, Shanghai, China; 2Department of Medical Imaging, Changhai Hospital, Shanghai, China; 3Department of Neurobiology and Physiology, Ningxia Medical University, Yinchuan, China; and 4Department of Pharmacology and Toxicology, Wright State University School of Medicine, Dayton, Ohio Submitted 5 July 2013; accepted in final form 15 April 2014

Wang YK, Shen D, Hao Q, Yu Q, Wu ZT, Deng Y, Chen YF, Yuan WJ, Hu QK, Su DF, Wang WZ. Overexpression of angiotensinconverting enzyme 2 attenuates tonically active glutamatergic input to the rostral ventrolateral medulla in hypertensive rats. Am J Physiol Heart Circ Physiol 307: H182–H190, 2014. First published May 16, 2014; doi:10.1152/ajpheart.00518.2013.—The rostral ventrolateral medulla (RVLM) plays a key role in cardiovascular regulation. It has been reported that tonically active glutamatergic input to the RVLM is increased in hypertensive rats, whereas angiotensin-converting enzyme 2 (ACE2) in the brain has been suggested to be beneficial to hypertension. This study was designed to determine the effect of ACE2 gene transfer into the RVLM on tonically active glutamatergic input in spontaneously hypertensive rats (SHRs). Lentiviral particles containing enhanced green fluorescent protein (lenti-GFP) or ACE2 (lenti-ACE2) were injected bilaterally into the RVLM. Both protein expression and activity of ACE2 in the RVLM were increased in SHRs after overexpression of ACE2. A significant reduction in blood pressure and heart rate in SHRs was observed 6 wk after lenti-ACE2 injected into the RVLM. The concentration of glutamate in microdialysis fluid from the RVLM was significantly reduced by an average of 61% in SHRs with lenti-ACE2 compared with lenti-GFP. ACE2 overexpression significantly attenuated the decrease in blood pressure and renal sympathetic nerve activity evoked by bilateral injection of the glutamate receptor antagonist kynurenic acid (2.7 nmol in 100 nl) into the RVLM in SHRs. Therefore, we suggest that ACE2 overexpression in the RVLM attenuates the enhanced tonically active glutamatergic input in SHRs, which may be an important mechanism underlying the beneficial effect of central ACE2 to hypertension. hypertension; renin-angiotensin system; gene transfer; glutamate; sympathetic activity IT IS WELL KNOWN that the rostral ventrolateral medulla (RVLM) is a key region for the central control of sympathetic outflow and plays a critical role in maintaining resting blood pressure (BP) and sympathetic tone (5, 20). ANG II, a main physiologically active effector peptide of the renin-angiotensin system (RAS), exerts its actions mainly by acting on the ANG II type 1 receptor (AT1R), thus contributing to cardiovascular regulation (1). ANG II signals, including AT1R expression and its sensitivity in the RVLM, are upregulated in hypertensive models, suggesting that increased ANG II signaling in the RVLM is critical in the development and maintenance of hypertension (11, 21, 33). Angiotensin-converting enzyme 2 (ACE2) is a member of the RAS and catalyzes the formation of

* Y.-K. Wang, D. Shen, and Q. Hao contributed equally to this work. Address for reprint requests and other correspondence: W.-Z. Wang, Dept. of Physiology, Second Military Medical Univ., 800 Xiangyin Road, Shanghai 200433, China (e-mail: [email protected]). H182

ANG-(1–7) from ANG II and ANG-(1–9) (45). ANG-(1–7), a major product of ACE2, exerts opposite effects (e.g., vasodilation and antiproliferation) to ANG II, indicating that ACE2 would be a novel therapeutic target for hypertension (8, 26). In the brain, ACE2 is widely expressed in cardiovascular regulatory regions (10, 53). Furthermore, it has been demonstrated that ACE2 expression and activity in the RVLM are decreased in spontaneously hypertensive rats (SHRs), whereas ACE2 gene transfer into the brain, including the RVLM, is beneficial to hypertension (14, 41, 55). These observations suggest that central ACE2 plays an important role in the neural control of BP and sympathetic outflow in hypertension. However, the exact mechanism by which central ACE2 is beneficial to hypertension remains to be determined. Glutamatergic synaptic transmission plays an important role in maintaining the resting activity of RVLM vasomotor neurons, thus acting as a driving element in keeping basal BP and sympathetic tone (5, 6, 20). Kubo et al. (28) reported that the content of glutamate in the RVLM is increased in SHRs compared with normotensive control rats [WistarKyoto (WKY) rats]. Microinjection of the glutamate receptor (GluR) antagonist kynurenic acid (KYN) into the RVLM produces a significant fall in resting BP in hypertensive rats but not in normotensive rats (22, 23). We have also previously reported that enhanced tonic glutamatergic input to the RVLM contributes to hyperactivity of RVLM vasomotor neurons in rats with chronic heart failure, a model of hyperactive brain RAS (49). These observations suggest that the tonically active glutamatergic input to the RVLM is enhanced in hypertension (22, 23, 43). Interestingly, previous evidence has indicated a possible interaction between the RAS and glutamatergic transmission in cardiovascular regions (7, 11); for example, blockade of AT1Rs in the RVLM antagonizes the effects of glutamate (2). Chronic AT1R blockade in the paraventricular nucleus (PVN) leads to a significant reduction in N-methyl-Daspartate (NMDA) receptor expression in rats with chronic heart failure (27). Although it has been demonstrated that ACE2 overexpression in the RVLM leads to a long-term decrease in BP in SHRs (55), there is no direct evidence showing the relationship between ACE2 and tonically active glutamatergic input in the RVLM. Therefore, the major aim of the present study was to determine if the enhanced tonically active glutamatergic input to the RVLM is reduced by ACE2 overexpression in hypertension. METHODS

Animals. Twelve-week-old male SHRs and WKY rats were purchased from Sino-British SIPPR/BK Laboratory Animal (Shanghai,

0363-6135/14 Copyright © 2014 the American Physiological Society

http://www.ajpheart.org

ACE2 AND GLUTAMATERGIC INPUT TO THE RVLM

China). Rats were housed in a temperature-controlled room and kept on a 12:12-h light-dark cycle with free access to water. All procedures in this study were approved by the Institutional Animal Care and Use Committee of Second Military Medical University. Cloning of human ACE2 in lentiviral vector and production of lenti-ACE2. The construction and production of the lentiviral vector were by Invabio (Shanghai Innovation Biotechnology, Shanghai, China), as previously described (19). Human (h)ACE2 cDNA (Accession No. NM_021804) was used as a template for PCR amplification with the use of the following primer sequences: sense 5=ATGCTGCGCGCCGCACTCAGCAC-3= and antisense 5=-TTACGAGTTCTTCTGTGGCACTT-3=. The hACE2 amplification product was cloned into lentivirus cloning vector pLenO-DCE (Invabio). The pLenO-DCE vector used in these experiments carried a green fluorescent protein (GFP) reporter gene. The presence of an internal ribosome entry site element permits the expression of two gene products (GFP and ACE2) from a single promoter. Note that the control construct (lenti-GFP) contained all sequence elements except for the therapeutic transgene (hACE2). Briefly, lentivirus packaging helper plasmids such as pRSV-Rev, pMDLg/pRRE, and pMD2.G were used to facilitate the transfection of the therapeutic transgene plasmid (pLenO-DCE-ACE2) in 293T cells. Virus-containing media of 293T cells were then harvested at 48h posttransfection, concentrated by ultrafiltration and ultracentrifugation, aliquoted, and stored at ⫺80°C. The number of transducing units, defined as an infectious particle, was determined by estimating the number of GFP-positive cells by FACS analysis in 293T cell suspensions, and the final infectious titer was ⬃5 ⫻ 109 transducing units/ml. Gene delivery into the RVLM. Rats were anesthetized with inhaled isofluorane (3%) and placed in a stereotaxic instrument (Shanghai Alcott Biotech). After the skull was exposed through an incision on the midline of the scalp, a small hole (2.0 mm lateral to the midline and 3.0 mm posterior to the lamda) was drilled bilaterally on the dorsal surface of the occipital according to the standard atlas of rats (35). Using the stereotaxic frame, lentiviral particles (1.25 ⫻ 106 transfection units, 250 nl) were slowly (⬃20 min) injected (10 mm deep to the skull surface) into the RVLM using a 32-gauge Hamilton springe (1 ␮l). The injection speed was controlled by an ultramicropump (WPI). After viral administration, rats were housed individually under specific pathogen-free conditions. Rats were divided into four groups (WKY, SHR, SHR-GFP, and SHR-ACE2) in this study. Measurement of BP by a tail-cuff system in conscious rats. BP and heart rate (HR) in conscious rats were measured noninvasively using a computized noninvasive tail-cuff system (ALC-NIBP, Shanghai Alcott Biotech), as previously described (56). Four parameters were simultaneously monitored: systolic BP, diastolic BP, mean arterial pressure (MAP), and HR. Conscious rats were warmed to an ambient temperature of 30°C to help vasodilate the tail artery by placing them in a holding device mounted on a thermostatically controlled warming plate. Rats were allowed to acclimate to the recording chamber for a period of 10 –15 min before BP was recorded, with rats resting comfortably in the chambers during this process. Using a noninvasive tail-cuff system, pressure was applied to the tail to occlude blood flow and was slowly released until the first pulse of arterial flow could be detected. The cuff was connected to a transducer, and the signal was amplified and recorded in a data-acquisition system. All values were averaged from at least six consecutive cycles. BP and HR were measured before gene transfer to the RVLM and every subsequent week. Measurement of ACE2 activity. Rats were euthanized by an overdose of pentobarbital sodium (200 mg/kg). RVLM tissues were punched on coronal sections of the brain stem according to the rat atlas and stored at ⫺80°C until use. ACE2 activity was measured using a SensoLyte 390 ACE2 Activity Assay Kit (no. 72086, Anaspec). RVLM tissues with lenti-GFP or lenti-ACE2 were rinsed, lysed, sonicated, and centrifuged. Supernatants were used to measure ACE2

H183

activity following the manufacturer’s instructions. The assay of ACE2 activity used Mca/Dnp (7-Mca-YVADAPK) fluorescence resonance energy transfer peptide. In the fluorescence resonance energy transfer peptide, the fluorescence of Mca was quenched by Dnp. Upon cleavage into two separate fragments by the enzyme, the fluorescence of Mca was recovered and monitored at an excitation of 330 nm and emission of 390 nm. The value was monitored using ApectraMax Gemini EM Fluorescence Reader (Synergy 2, Bio-Tek). ACE2 activity was expressed as relative fluorescence units per unit time and normalized for total protein. Western blot analysis. As previously described (37, 56), Western blot analysis was performed for determining the protein levels of ACE2, AT1R, and Mas receptor in the RVLM. In each group, a total of 5 rats were used for RVLM tissue collected for Western blot analysis. After rats had been euthanized by an overdose of pentobarbital sodium (200 mg/kg), the brain was removed, and RVLM tissues were punched on coronal sections of the brain stem according to the standard atlas of rats (35). The analyzed samples from individual rats were not pooled samples. The protein concentration was measured, and protein (30 ␮g) was run on a 7.5% or 12% SDS-PAGE gel and then transferred to a polyvinylidene fluoride membrane (Milipore). After being blocked for 1 h with 5% milk in Tris-buffered salineTween, membranes were subjected to immunoblot analyses with anti-ACE2 (no. 3583-1, Epitomics), anti-AT1R (no. 446p1, Sigma), anti-Mas (no. AAR-013, Alomone Labs), anti-NMDA receptor 1 (no. 2824-1, Epitomics), or anti-GluR5/6/7 (kainate receptor, no. sc28797, Santa Cruz Biotechnology) antibodies. Immunodetection was accomplished with secondary antibodies [ACE2, Mas, NMDA receptor 1, and GluR5/6/7 (peroxidase-conjugated goat anti-rabbit IgG, Jackson Immunoresearch Laboratories); AT1R (peroxidase-conjugated donkey anti-goat IgG, ProteinTech Group)]. After final washes, the membrane was incubated with chemiluminescent agent, and protein bands were visually detected by the Syngene Bio Imaging system (no. 55000, Gene Company) and quantified by GeneTools software (Gene Company). Levels of target proteins were normalized to ␣-tubulin (no. T6074, Sigma), which served as a loading control. Immunohistochemistry. The procedure for immunohistochemistry of GFP expression was carried out as previously described (36). Rats were euthanized with an overdose of pentobarbital sodium (200 mg/kg) and perfused through the aorta with 0.9% NaCl solution and 4% paraformaldehyde in 0.1 M phosphate buffer. Rat brains were removed and refixed in 4% paraformaldehyde in 0.1 M phosphate buffer overnight at 4°C. Brain blocks were transferred to 20% sucrose in 0.1 M phosphate buffer and kept in the solution until they sank to the bottom. After the brain blocks had been rapidly frozen, sections of 20 ␮m thickness were cut in a cryostat and floated in 0.01 M PBS (pH 7.4). The floating sections were washed three times for 5 min in PBS and preincubated in 10% BSA in PBS for 30 min. All incubations and reactions were separated by three 5-min washes in 0.01 M PBS. Finally, sections were then mounted on slides and embedded in antifade mounting medium. The GFP staining (excitation: 488 nm and emission: 530 nm) was detected with a Olympus digital camera DP72 attached to an Olympus microscope (IX71, Olympus). In vivo brain microdialysis. In vivo brain microdialysis was carried out to measure glutamate content in the extracellular fluid in the RVLM, as previously described (37, 56). After rats had been anesthetized with inhaled isofluorane (3%), the RVLM was surgically exposed and chemically identified by a pressor response (at least 20 mmHg) to L-glutamate. A microdialysis probe (MAB6.14.2) was unilaterally inserted into this region. Brain microdialysis was performed by perfusing the probe with artificial cerebrospinal fluid at a rate of 2 ␮l/min with a microdialysis pump (Bioanalytical System). The total volume of each dialysate sample (10 min) was 20 ␮l, and samples were obtained after at least 60 min of animal rest after the surgical operation. Three dialysates were collected for 10-min periods and stored at ⫺80°C until use.

AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00518.2013 • www.ajpheart.org

H184

ACE2 AND GLUTAMATERGIC INPUT TO THE RVLM

HPLC. As previously described (37, 56, 57), HPLC (model 582 pump, ESA) with electrochemical detection (model 5300, ESA) was used to detect the glutamate concentration in dialysate samples from the RVLM and 24-h urinary excretion of norepinephrine (NE). The derivatization of O-phthalaldehyde/␤-mercaptoethanol was used to determine the amino acid glutamate, as described in our previous study (56). Glutamate was separated by HPLC with a reverse-phase C18 column (Shiseido Capcell Pak 75 ⫻ 3 mm, 3 ␮m C18, P/N 88-90816) and further quantified with electrochemical detection. The mobile phase was composed of 100 mM di-sodium hydrogen phosphate anhydrous, 22% methanol, and 3.5% acetonitrile at pH 6.75, and the flow rate was 0.60 ml/min. The detect channel potentials were set at ⫹150 and ⫹550 mV. The HPLC system was controlled, and data were acquired, processed, and analyzed using Coulochem software. The analyte glutamate was identified according to the authentic standard based on its retention time and peak height ratio. The content of glutamate was quantified by a comparison of the area with the standard. Measurement of urinal NE was based on our previous study (56). Briefly, 24-h urinary samples were collected after rats were placed in metabolism cages for 24 h and acidified with glacial acetic acid in 15-ml centrifuge tubes, which were embedded in crushed ice. Dihydroxybenzylamine (Sigma) was used as the internal standard. NE was absorbed onto acid-washed alumina with 3 mol/l Tris(hydroxymethyl) aminomethane buffer at pH 8.6 in 2% EDTA. The alumina was then washed 3 times with 3 ml of distilled water. NE was extracted with 400 ␮l of 0.2 M glacial acetic acid with 5 min of shaking and a final 30-min settlement. The supernatant (40 ␮l) was injected into the HPLC column [reverse phase, ESA, 150 ⫻ 3.2 mm, 3 ␮m C18 (P/N 70 – 0636)], and NE was eluted with the mobile phase (80 mM citric acid monohydrate, 73.4 mM citric acid trisodium salt, 0.12 mM 1-octanesulfonic acid sodium salt, and 0.1 mM EDTA adjust to pH 4.3 with phosphoric acid). The flow rate was set at 0.5 ml/min. Intracerebraventricular infusion. To determine the role of the Mas receptor in mediating the effect of ACE2 overexpression on the release of glutamate in the RVLM, intracerebraventricular (fourth ventricle) infusion of the Mas receptor antagonist A779 was performed on the fifth wk after lenti-ACE2 injected into the RVLM of SHRs. Rats in the fifth week of ACE2 overexpression were anesthetized with inhaled isofluorane (3%) and fixed on the stereotaxic frame. The atlantooccipital membrane was exposed and punctured with a needle. The cannula connect to the osmotic mimipump capsule (0.5 ␮l/h, 1000D, ALZET) was implanted into the fourth ventricle along the pinhole and fixed with biological glue. After 1 wk of continuous infusion of A779 (1 nmol/day) or vehicle (artificial cerebrospinal fluid), rats were subsequently subjected to in vivo brain microdialysis. Responses of BP, HR, and renal sympathetic nerve activity to acute RVLM microinjections. Acute in vivo experiments were performed to detect the cardiovascular sensitivity [BP, HR, and renal sympathetic nerve activity (RSNA)] to injection of GluR antagonist into the RVLM. Surgical procedures, recording of RSNA, and RVLM microinjections were carried out as we have previously described (36, 37, 49, 50, 56). Briefly, rats were anesthetized with urethane (800 mg/kg ip) and ␣-chloralose (40 mg/kg ip), and the trachea was cannulated to facilitate mechanical respiration with a ventilator (SAR-830, CWE). The concentration of end-tidal CO2 was monitored and kept at the level of ⬃4% by a CO2 analyzer (Capstar 100, CWE). The right formal artery was catheterized for BP measurement by PowerLab (8SP, AD Instruments). MAP and HR were derived from the BP pulse. The degree of anesthesia was monitored by the BP response to a nociceptive stimulus paw pinch. An adequate depth of anesthesia was defined by observing a small change (⬍20%) of BP in response to a paw pinch at intervals throughout the experiment, as previously described (13). Additional doses of anesthetic (10 mg/kg iv ␣-chloralose) were administrated as required. The rat was placed in a stereotaxic frame with the head fixed horizontally, and the dorsal surface of the medulla was surgically exposed. The left renal sympa-

thetic nerves were exposed, identified, dissected free of the surrounding connective tissue, and placed on a pair of platinum-iridium recording electrodes. The distal terminal of the renal nerve was cut to avoid afferent activity. RSNA was amplified (band pass: 100 –3,000 Hz) and monitored by the PowerLab system. The preinjection level (control) of RSNA was taken as 100% from the absolute value after the noise level was subtracted. Body temperature was kept at 37°C by a temperature controller. Microinjections in the RVLM were made from a three-barrel micropipette. The coordinates for injections into the RVLM were 2.0 –2.5 mm rostral to the caudal tip of the area postrema, 2.0 –2.2 mm lateral to the medline, and 3.0 –3.2 mm below the dorsal surface of the brain stem. The RVLM injection (100 nl) was made over a period of 5–10 s by a pressure injector (PV820, WPI). A pressor response (at least 20 mmHg) evoked by microinjection of L-glutamate (2 nmol) was used to chemically identify the RVLM. Based on previous studies (22–24, 56), a 2.7 nmol (100 nl) dose of KYN (Sigma) was chosen to block ionotropic GluRs at least 30 min after a functional pressor site in the RVLM was identified. The interval of bilateral injections was within 2 min. Continuous recordings of BP, HR, and RSNA were at least 60 min after bilateral injection of KYN into the RVLM. At the end of the experiment, the rat was euthanized with an overdose of pentobarbital sodium. The maximum of RSNA occurred 1–2 min after the rat was euthanized. As previously described (47), the baseline RSNA value of individual rat was taken as a percentage of the maximum RSNA. Injection sites were also marked by injection of dye blue, and the rat was perfused with 10% formaldehyde solution (100 ml) intracardially. The brain stem was then quickly removed and fixed in 10% buffered formalin. Frozen 50-␮m coronal sections were made using a freezing microtome and mounted on slides. Injection sites were checked according to the standard atlas of Paxinos and Watson (35). Data analysis. Data are presented as means ⫾ SE. Statistical comparisons between experimental groups were made by a Student’s t-test or one-way ANOVA followed by a Newman-Keuls post hoc test. Differences were considered to be significant for P values of ⬍0.05. RESULTS

Efficacy of lenti-ACE2 gene transfer to the RVLM. Figure 1A shows that GFP expression was restricted to the area of the RVLM. We confirmed that the level of ACE2 expression in the RVLM was significantly decreased in untreated SHRs compared with untreated WKY rats. We found that ACE2 was expressed in both neurons and fibers. We further observed an average of an approximately twofold increase (P ⬍ 0.05) in ACE2 expression in SHRs 4 wk after lenti-ACE2 injection into the RVLM compared with lenti-GFP injection (Fig. 1B). Similar to ACE2 protein expression, ACE2 activity in the RVLM was also found to be decreased in SHRs compared with WKY rats. Overexpression of ACE2 produced a significant (P ⬍ 0.05) increase of 69% in ACE2 activity in SHRs (Fig. 1C). Effect of ACE2 overexpression in the RVLM on BP, HR, and 24-h urinary excretion of NE. As shown in Fig. 2, levels of BP and HR began to decrease in SHRs 3 wk after lenti-ACE2 injection compared with lenti-GFP injection. This reduction in BP and HR persisted until the time (6 wk) of termination of the experiment. However, levels of BP and HR in lenti-ACE2transfected SHRs were still higher than those in untreated WKY rats. In addition, lenti-ACE2 injected into the RVLM had no effect on baseline BP and HR in WKY rats (Fig. 2B). Compared with WKY rats, SHRs showed a significant increase in 24-h urinary excretion of NE. However, 24-h urinary excre-

AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00518.2013 • www.ajpheart.org

H185

ACE2 AND GLUTAMATERGIC INPUT TO THE RVLM

A

Bregma -11.96 mm

a

b

c

1 mm WKY

SHR

C

SHR-GFP SHR-ACE2 110 kDa

ACE2 α-Tubulin

50 kDa

2500

Relative levels of ACE2

1.5 #

1.0

* .5

0.0

ACE2 activity (AFU/µg)

B

2000 #

1500

500 0

WKY

SHR

SHR-GFP SHR-ACE2

*

1000

WKY

SHR

SHR-GFP SHR-ACE2

Fig. 1. Overexpression of angiotensin-converting enzyme 2 (ACE2) in the rostral ventrolateral (RVL) medulla (RVLM). A: according to the standard rat atlas (a), green fluorescent protein (GFP) expression was restricted in the RVLM (b, from red square in a) in lenti-ACE2-injected spontaneously hypertensive rats (SHRs). c: higher-magnification image from the white square in b. GFP was expressed on neurons (white arrows) and fibers (red arrows). SP5I, interpolar division of the spinal trigeminal complex; Amb, nucleus ambiguus; LPGi, nucleus paragigantocellularis lateralis; Py, pyramidal tract. Scale bars ⫽ 200 ␮m in b and 50 ␮m in c. B and C: protein level (B) and activity (C) of ACE2 in the RVLM of untreated Wistar-Kyoto (WKY) rats (WKY group), untreated SHRs (SHR group), lenti-GFP-injected SHRs (SHR-GFP group), and lenti-ACE2-injected SHRs (SHR-ACE2 group). AFU, absolute fluorescent units. n ⫽ 5 rats/group. *P ⬍ 0.05 vs. the WKY group; #P ⬍ 0.05 vs. the SHR-GFP group.

tion of NE was significantly reduced in the SHR-ACE2 group compared with the SHR-GFP group (0.36 ⫾ 0.06 vs. 0.59 ⫾ 0.03 ␮g, n ⫽ 5, P ⬍ 0.05; Fig. 2C). Effect of ACE2 overexpression on the release of glutamate in the RVLM. As shown in Fig. 3A, the concentration of glutamate in the microdialysis fluid of the RVLM was significantly higher in untreated SHRs compared with untreated WKY rats (1,476 ⫾ 137 vs. 280 ⫾ 83 ␮g/l, n ⫽ 5, P ⬍ 0.05). The content of glutamate was significantly reduced (1,586 ⫾ 165 vs. 622 ⫾ 70 ␮g/l, P ⬍ 0.05) in SHRs 6 wk after RVLM injection of lenti-ACE2 compared with lenti-GFP, but it was still higher than in WKY rats. Furthermore, the ACE2induced reduction in glutamate release in SHRs was significantly blunted after treatment with intracerebraventricular infusion of the Mas receptor antagonist A779 (1 nmol/day, 1 wk) in the fifth week of lenti-ACE2 injection into the RVLM of SHRs. Moreover, we also observed a relationship between the level of ACE2 protein expression and glutamate release at different time points (baseline, second week, fourth week, and sixth week) after lenti-ACE2 injected into the RVLM of SHRs (Fig. 3C). These results suggest that a gradual decrease in glutamate content was accompanied by a gradual increase in ACE2 expression after lenti-ACE2 injections in SHRs.

Effect of ACE2 overexpression on the decreases in BP, HR, and RSNA evoked by blockade of GluRs in the RVLM. As shown in Table 1, baseline BP, HR, and RSNA in anaesthetized rats were reduced in SHRs 6 wk after ACE2 overexpression in the RVLM. Figure 4 shows original tracings of BP, HR, and RSNA in response to microinjection of the GluR antagonist KYN (2.7 nmol) into the RVLM. Bilateral injection of KYN into the RVLM produced a significant decrease in BP, HR, and RSNA in untreated SHRs but not in untreated WKY rats. However, these decreases in BP (⫺22.7 ⫾ 1.8 vs. ⫺42.4 ⫾ 3.7 mmHg), HR (⫺21.9 ⫾ 4.1 vs. ⫺41.1 ⫾ 3.5 beats/min), and RSNA (⫺11.6 ⫾ 0.9 vs. ⫺20.6 ⫾ 2.6%) induced by KYN in the RVLM were significantly (n ⫽ 5, P ⬍ 0.05) lower in SHRs that received lenti-ACE2 injection compared with lenti-GFP injection (Fig. 5). Effects of ACE2 overexpression on protein levels of AT1R, Mas receptor, and GluR expression in the RVLM of SHRs. Figure 6A shows the effect of ACE2 overexpression on levels of AT1R and Mas protein expression in the RVLM of SHRs. A reduction in AT1R expression was found in the RVLM of lenti-ACE2-injected SHRs compared with lenti-ACE2-injected SHRs, whereas the Mas receptor was upregulated in SHRs with lenti-ACE2 treatment. In addition, ACE2 overexpression produced a reduction of 49% and 50% in protein levels of

AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00518.2013 • www.ajpheart.org

H186

ACE2 AND GLUTAMATERGIC INPUT TO THE RVLM

A

SHR-GFP SHR-ACE2

WKY SHR

500

HR (beats/min)

MAP (mmHg)

180 160 *

#

Injection

140

*#

*#

*#

120 100

400 * Injection

4th

5th

(Week)

6th

B 140

Base 1st 2nd 3rd

C

120

100

Injection

*

4th

5th

6th

2nd 3rd

NMDA receptor 1 and GluR5/6/7 in the RVLM of SHRs, respectively (Fig. 6B). DISCUSSION

The major finding of the present study are as follows: 1) overexpression of ACE2 in the RVLM produced a long-term decrease in BP in SHRs; 2) ACE2 overexpression significantly reduced the release of glutamate in the RVLM of SHRs; and 3) ACE2 overexpression significantly blunted the response of BP,

4th

5th

6th

* .6 #

.4 .2

WKY

(Week)

SHR

SHR-GFP SHR-ACE2

HR, and RSNA to blockade of GluRs in the RVLM of SHRs. On the basis of these results, we conclude that ACE2 overexpression effectively attenuates the enhanced tonically active glutamatergic input to the RVLM, which may be an important contributor to the beneficial effects of central ACE2 on hypertension. It has been previously reported that lentiviral vector-mediated transgene expression begins as early as 3 days postinfection. However, in the present study, levels of BP and HR began

A

B

Base

2nd

4th

6th

α-Tubulin

2000

* *#

*

#

500

WK

Y

*#

.8 .6

#

.4 .2

2 P R 79 SF SH R-GF -ACE aC 2+A7 R 2+ E E H SH S AC R-AC RSH SH

0.0 Base

Content of glutamate (µg/L)

0

*

$

ACE2/Tubulin

*

1000

50 kDa

1.0

#

(week) 110 kDa

ACE2

2500

1500

(Week)

.8

0.0 Base 1st

Content of glutamate (µg/L)

*

1.0

NE in 24-h urine (µg)

WKY-GFP WKY-ACE2

80

Fig. 3. Effect of ACE2 overexpression in the RVLM on glutamate concentration. A: content of glutamate in the dialysis fluid from the RVLM from the WKY, SHR, SHR-ACE2, SHR-GFP groups with or without pretreatment with intracerebraventricular infusion (1 wk) of the Mas receptor antagonist A779. n ⫽ 5 rats/group. *P ⬍ 0.05 vs. the WKY group; #P ⬍ 0.05 vs. the SHR-GFP group; $P ⬍ 0.05 vs. the SHR-ACE2 group ⫹ artificial cerebrospinal fluid (aCSF). B: time course of ACE2 expression levels (top) and glutamate contents (bottom) at baseline and on the second, fourth, and sixth wk after ACE2 gene transfer into the RVLM of SHRs. n ⫽ 4 rats/group. *P ⬍ 0.05 vs. baseline; #P ⬍ 0.05 vs. the value in the fourth wk.

*

300

200 Base 1st 2nd 3rd

MAP (mmHg)

Fig. 2. Effect of ACE2 overexpression in the RVLM on blood pressure (BP), heart rate (HR), and 24-h urinary excretion of norepinephrine (NE). A: time course of mean arterial pressure (MAP; left) and HR (right) measured by tail-cuff system in the WKY, SHR, SHR-GFP, and SHR-ACE2 groups. A timedependent reduction in MAP and HR was induced by lenti-ACE2 injected into the RVLM in SHRs compared with lenti-GFP injection. Base, baseline. n ⫽ 15 rats/group. *P ⬍ 0.05 vs. the SHR-GFP group. B: MAP in WKY rats injected with lenti-GFP (WKYGFP group) and lenti-ACE2 (WKY-ACE2 group). C: effect of ACE2 overexpression on 24-h urinary excretion of NE. n ⫽ 5 rats/ group. *P ⬍ 0.05 vs. the WKY group; #P ⬍ 0.05 vs. the SHR-GFP group.

SHR-GFP SHR-ACE2

WKY SHR

2nd

4th

6th (Week)

2000 1500

* #*

1000 500 0 Base

AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00518.2013 • www.ajpheart.org

2nd

4th

6th

(Week)

ACE2 AND GLUTAMATERGIC INPUT TO THE RVLM

H187

Table 1. Values of baseline MAP, HR, and RSNA in anesthetized rats for acute in vivo experiments Number of rats MAP, mmHg HR, beats/min RSNA, %maximum

WKY Group

SHR Group

SHR-GFP Group

SHR-ACE2 Group

5 120 ⫾ 4 383 ⫾ 8 31.6 ⫾ 3.4

5 172 ⫾ 5* 387 ⫾ 12 46.4 ⫾ 4.4*

5 178 ⫾ 4 395 ⫾ 11 47.3 ⫾ 3.7

5 154 ⫾ 3*† 354 ⫾ 7*† 39.4 ⫾ 3.5*†

Data are means ⫾ SE. Blood pressure was measured by catheterizing the fermoral artery. MAP, mean arterial pressure; HR, heart rate; RSNA, renal sympathetic nerve activity. Rats were divided into the following four groups: untreated Wistar-Kyoto (WKY) rats (WKY group), untreated spontaneously hypertensive rats (SHRs; SHR group), lenti-green fluorescent protein (GFP)-injected SHRs (SHR-GFP group), and lenti-angiotensin-converting enzyme 2 (ACE2)-injected SHRs (SHR-ACE2 group). *P ⬍ 0.05 vs. the WKY group; †P ⬍ 0.05 vs. the SHR-GFP group.

to decrease 3 wk after lenti-ACE2 injected into the RVLM. This observation is agreement with a previous study by Yamazato et al. (55), in which the authors explain the initially compensatory physiological mechanisms responsible to counteract the effect of ACE2 in restoring BP in SHRs. We found that BP and HR were kept at lower levels on the fourth to sixth week in SHRs with lenti-ACE2 treatment. Moreover, ACE2 activity and protein expression in the RVLM were significantly increased 4 wk after ACE2 gene transfer. The presented data show that maximum ACE2 expression is reached at 4 wk after ACE2 gene delivery in SHRs (Fig. 3B). Moreover, the level of ACE2 expression is still higher at 6 wk after gene delivery compared with control. In addition, SHRs 6 wk after lentiACE2 injection into the RVLM showed a significant decrease in 24-h urinary excretion of NE and basal RSNA (percentage of maximum), indicating that ACE2 overexpression in the RVLM reduces sympathetic activity in SHRs. It was notable that ACE2 overexpression in the RVLM of SHRs decreased the BP level, which was still higher than the BP level of WKY rats. However, 24-h urinary excretion of NE in SHRs with ACE2 overexpression was completely restored to its level in WKY rats. In addition to sympathetic activity, alternations in vagal

nerve activity also have an effect on BP level. Lenti-ACE2 injections were located in the sympathetic center of the RVLM but not in the vagal center nucleus ambiguous. Therefore, it is possible that the reduction of sympathetic activity by ACE2 in the RVLM was not sufficient to return the BP level to that of WKY rats. Based on the above evidence, SHRs received lenti-ACE2 or lenti-GFP injections, and results were analyzed after 6 wk to determine the interaction between ACE2 and glutamatergic input at the level of the RVLM. Similar to a previous study (55), we also revealed (Fig. 2B) that ACE2 overexpression in the RVLM did not alter the BP level in WKY rats, suggesting that ACE2 may not contribute to maintaining the resting BP level under normotensive conditions. In this study, untreated WKY rats and SHRs were used to as controls for SHRs with lenti-ACE2 or lenti-GFP gene transfer. In addition, we also note two limitations in our approach of ACE2 gene delivery into the RVLM. The area of the RVLM for ACE2 gene delivery was not functionally identified by a pressor response to L-glutamate. However, according to the distribution of GFP immunostaining, we confirmed that the distribution of gene delivery was accurately located within the RVLMl area. Second, a tail-cuff system was used to

Fig. 4. Representative original recordings of BP, HR, and renal sympathetic nerve activity (RSNA) showing the effects of bilaterally injection of the glutamate receptor (GluR) antagonist kynurenic acid (KYN; 2.7 nmol) into the RVLM of the WKY, SHR, SHR-GFP, and SHR-ACE2 groups. ABP, arterial BP; bpm, beats/min; a.u., arbitrary units; inte-RSNA, integrated RSNA. AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00518.2013 • www.ajpheart.org

H188

ACE2 AND GLUTAMATERGIC INPUT TO THE RVLM MAP

HR

RSNA

0 #

-10 #

-20 -30 -40

*

*

* #

*

*

*

WKY SHR SHR-GFP SHR-ACE2

Fig. 5. Percent changes in MAP, HR, and RSNA induced by microinjection of the GluR antagonist KYN into the RVLM. n ⫽ 5 rats/group. *P ⬍ 0.05 vs. the WKY group; #P ⬍ 0.05 vs. the SHR-ACE2 group.

monitor BP in conscious rats with gene transfer. This tail-cuff method is a form of restraint and, as such, may stress the animal and alter the level of BP. Moreover, the RAS in the RVLM may be involved in the pressor response to stress (3). In this work, however, we found a similar BP level obtained in anesthetized rats by catheterizing the femoral artery compared with in conscious rats by the tail-cuff method. The main goal of the present study was to determine the effect of ACE2 overexpression on the tonically active glutamatergic inputs to the RVLM. Consistent with previous studies (22, 23), we also confirmed that tonically active glutamatergic input to the RVLM is enhanced in SHRs. Furthermore, we found that ACE2 overexpression in the RVLM in SHRs significantly reduced the release of glutamate but also significantly blunted the reduction in BP, HR, and RSNA evoked by microinjection of KYN into the RVLM. Moreover, the gradual decrease in glutamate content was accompanied by a gradual increase in ACE2 expression level after ACE2 gene transfer into the RVLM of SHRs, suggesting that this reduction in glutamate release is probably produced by ACE2 overexpression in SHRs. Consistent with previous studies (10, 55), we confirmed that ACE2 was expressed in both the cell body as

A

SHR-GFP

well as nerve terminal fibers. Therefore, we suggest that ACE2 overexpression is capable of attenuating the enhanced tonically active glutamatergic inputs to the RVLM in SHRs. We also initially observed the changes of glutamate receptors in response to ACE2 overexpression in SHRs. We found that protein levels of NMDA receptor 1 (an ionotropic GluR subtype) and GluR5/6/7 (a non-NMDA receptor) in SHRs were downregulated by ACE2 overexpression. It is likely that ACE2 overexpression reduces the functional state of GluRs in SHRs. The level of NMDA receptor expression is different between WKY rats and SHRs (12), and SHRs are more sensitive to L-glutamate or NMDA in the RVLM (32, 44). However, it has been documented that injection of glutamate into the RVLM affects blood pressure to the same extent in both strains (39). The significance of the functional state of GluRs in the RVLM in mediating the beneficial effect of ACE2 on hypertension needs to be further determined. Clearly, the results of the present study do not adequately address the question of how ACE2 gene transfer into the RVLM reduces tonically active glutamatergic input. ACE2 overexpression leads to an increase in the production of intrinsic ANG-(1–7) (45). Upregulation of the ACE2/ANG-(1–7)/ Mas axis in the brain has been become a potential therapeutic strategy for the treatment of hypertension and chronic heart failure (8, 14, 53, 54). It has been reported central chronic application of ANG-(1–7) is capable of lowering high BP or restoring the blunted baroreflex sensitivity in these cardiovascular diseases (18, 25). In this study, we found that ACE2 overexpression significantly increased protein levels of the ANG-(1–7)-acting receptor Mas; this result is supported by a recent study (41) showing ACE2 overexpression in the PVN upregulated Mas receptor expression in ANG II-induced hypertension. Importantly, it has been suggested that ANG-(1–7) is capable of modulating the release of neurotransmitters (e.g., dopamine, GABA, and glutamate) via Mas receptors in the brain (42, 48). The present data have further demonstrated that the ACE2-mediated reduction in glutamate release in the

38 kDa

AT1R α-Tubulin Mas α-Tubulin

SHR-ACE2 120 kDa

50 kDa

α-Tubulin

50 kDa

KR

50 kDa

α-Tubulin

50 kDa 105 kDa 50 kDa

.6 SHR-GFP SHR-ACE2

.6 .4

SHR-GFP NR1

.8

AT1R and Mas/tubulin

Fig. 6. Representative Western blots (top) and densitometric analyses (bottom; relative levels) of ANG II type 1 receptor (AT1R)/ Mas receptor (A) and N-methyl-D-aspartate receptor 1 (NR1)/kainate receptor (KR; B) in the RVLM of the SHR-GFP and SHR-ACE2 groups. n ⫽ 5 rats/group. *P ⬍ 0.05 vs. the SHR-GFP group.

B

SHR-ACE2

* *

.2 0.0

Glu receptors/tubulin

% changes by KYN

10

SHR-GFP SHR-ACE2

.4

.2

*

*

0.0 AT1R

Mas

AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00518.2013 • www.ajpheart.org

NR1

KR

ACE2 AND GLUTAMATERGIC INPUT TO THE RVLM

RVLM of SHRs was significantly blunted by pretreatment with blockade of the Mas receptor, suggesting that the ANG-(1–7)/ Mas axis contributes, at least partially, to the inhibitory effect of ACE2 on glutamate release in the RVLM. Interestingly, Wang et al. (48) reported that acute injection of exogenous ANG-(1–7) into the RVLM produced an increase in BP accompanied by an increased release of glutamate in normotensive rats. However, the data from the present study and previous study (55) have demonstrated that a long-term increase in endogenous ANG-(1–7) by ACE2 gene transfer into the RVLM significantly reduces BP in SHRs. Differential biological actions of Mas receptor activation by ANG-(1–7) may be dependent on some conditions (e.g., acute vs. chronic, endogenous vs. exogenous, and the BP level of animals). The exact mechanisms underlying the effect of Mas receptor activation on central cardiovascular regulation need to be further defined. On the other hand, recent evidence suggests a possible interaction between ACE2 and the ACE/ANG II/AT1R axis (52, 53). We found that protein expression of AT1Rs in the RVLM is downregulated in SHRs treated with ACE2 overexpression; this finding is agreement with previous studies showing that ACE2 overexpression in the PVN or subfornical organ downregulates AT1Rs in ANG II-induced hypertension (15, 41). It has also been reported that AT1Rs can be expressed in the presynaptic terminal (29). Thus, a downregulated ACE/ ANG II/AT1R axis may be in play to result in the beneficial effects of ACE2 in the RVLM on hypertension. ANG II injected into the RVLM stimulates the release of glutamate (57). In addition to the RVLM, it has also been suggested that ANG II-mediated excitation of magnocellular PVN neurons is dependent on an increase in glutamatergic input (30). Furthermore, it has been suggested that increased ROS, which are mainly derived by ANG II, contribute to enhanced glutamatergic input to the RVLM in SHRs (34). On basis of the above observations, downregulation of the ANG II/AT1R axis may play an important role in mediating the effect of ACE2 overexpression on tonically active glutamatergic input to the RVLM in hypertension. However, the exact mechanism(s) by which alterations of AT1Rs and Mas receptors by ACE2 influence the release of glutamate needs to be further defined. AT1Rs and Mas receptors are reported to be expressed in presynaptic terminals and involved in mediating the release of neurotransmitters (17, 31). Ca2⫹ influx is necessary for the release of neurotransmitters from presynaptic terminals (38). Interestingly, it has been suggested that ANG II increases Ca2⫹ influx (16, 46), whereas ANG-(1–7) may decrease Ca2⫹ influx, perhaps through triggering nitric oxide signaling (9, 40, 51). Therefore, it is likely that the ACE2-induced decrease in glutamate release in the RVLM of SHRs is associated with decreased Ca2⫹ influx, which may be dependent on functional states of the AT1R and Mas receptor. In addition, we also realize that future experiments may be necessary to further differentiate the ACE2 beneficial effect. For example, comparison of ACE2 beneficial effects with AT1R blockade alone or ANG(1–7) infusion alone will reveal more about the involvement of individual components. In summary, the present results have shown that ACE2 gene transfer to the RVLM in SHRs attenuates the increased tonically active glutamatergic inputs to the RVLM in SHRs. Normalization of the enhanced glutamatergic input in RVLM neurons is an important mechanism underlying the beneficial

H189

effects of ACE2 on high BP and sympathetic hyperactivity in hypertension. The present work provides new information to extend our knowledge of the ACE2-induced adjustment of central autonomic networks for cardiovascular activity. ACKNOWLEDGMENTS The authors gratefully acknowledge the statistical assistance of Dr. Jian Lu (Department of Statistical Analysis, Second Military Medical University). GRANTS This work was supported by Shanghai National Natural Science Foundation of China Grants 81170240, 81370363, and 31100831, Second Military Medical University Creativity and Innovation Training Program Grant 2010QN03, and National Heart, Lung, and Blood Institute Grant HL-098637. DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS Author contributions: Y.-K.W., D.-F.S., and W.-Z.W. conception and design of research; Y.-K.W., D.S., Q.H., Q.Y., Z.-T.W., and Y.D. performed experiments; Y.-K.W., D.S., Q.H., Q.Y., Z.-T.W., Y.D., W.-J.Y., and W.-Z.W. analyzed data; Y.-K.W., D.S., and Y.D. prepared figures; Y.-K.W. drafted manuscript; Q.H., Q.Y., Y.D., and Q.-K.H. interpreted results of experiments; Y.-F.C., W.-J.Y., Q.-K.H., D.-F.S., and W.-Z.W. edited and revised manuscript; W.-Z.W. approved final version of manuscript. REFERENCES 1. Allen AM, Zhuo J, Mendelsohn FA. Localization and function of angiotensin AT1 receptors. Am J Hypertens 13: 31S–38S, 2000. 2. Averill DB, Tsuchihashi T, Khosla MC, Ferrario CM. Losartan, nonpeptide angiotensin II-type 1 (AT1) receptor antagonist, attenuates pressor and sympathoexcitatory responses evoked by angiotensin II and L-glutamate in rostral ventrolateral medulla. Brain Res 665: 245–252, 1994. 3. Chen D, Jancovski N, Bassi JK, Nguyen-Huu TP, Choong YT, PalmaRigo K, Davern PJ, Gurley SB, Thomas WG, Head GA, Allen AM. Angiotensin type 1A receptors in C1 neurons of the rostral ventrolateral medulla modulate the pressor response to aversive stress. J Neurosci 32: 2051–2061, 2012. 4. Coleman JE, Huentelman MJ, Kasparov S, Metcalfe BL, Paton JF, Katovich MJ, Semple-Rowland SL, Raizada MK. Efficient large-scale production and concentration of HIV-1-based lentiviral vectors for use in vivo. Physiol Genomics 12: 221–228, 2003. 5. Dampney RA. Functional organization of central pathways regulating the cardiovascular system. Physiol Rev 74: 323–364, 1994. 6. Dampney RA, Tagawa T, Horiuchi J, Potts PD, Fontes M, Polson JW. What drives the tonic activity of presympathetic neurons in the rostral ventrolateral medulla? Clin Exp Pharmacol Physiol 27: 1049 –1053, 2000. 7. Dampney RA, Tan PS, Sheriff MJ, Fontes MA, Horiuchi J. Cardiovascular effects of angiotensin II in the rostral ventrolateral medulla: the push-pull hypothesis. Curr Hypertens Rep 9: 222–227, 2007. 8. Der SS, Huentelman MJ, Stewart J, Katovich MJ, Raizada MK. ACE2: a novel therapeutic target for cardiovascular diseases. Prog Biophys Mol Biol 91: 163–198, 2006. 9. Desole MS, Kim WK, Rabin RA, Laychock SG. Nitric oxide reduces depolarization-induced calcium influx in PC12 cells by a cyclic GMPmediated mechanism. Neuropharmacology 33: 193–198, 1994. 10. Doobay MF, Talman LS, Obr TD, Tian X, Davisson RL, Lazartigues E. Differential expression of neuronal ACE2 in transgenic mice with overexpression of the brain renin-angiotensin system. Am J Physiol Regul Integr Comp Physiol 292: R373–R381, 2007. 11. Dupont AG, Brouwers S. Brain angiotensin peptides regulate sympathetic tone and blood pressure. J Hypertens 28: 1599 –1610, 2010. 12. Edwards MA, Loxley RA, Powers-Martin K, Lipski J, McKitrick DJ, Arnolda LF, Phillips JK. Unique levels of expression of N-methyl-Daspartate receptor subunits and neuronal nitric oxide synthase in the rostral ventrolateral medulla of the spontaneously hypertensive rat. Brain Res Mol Brain Res 129: 33–43, 2004. 13. Farnham MM, Lung MS, Tallapragada VJ, Pilowsky PM. PACAP causes PAC1/VPAC2 receptor mediated hypertension and sympathoexci-

AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00518.2013 • www.ajpheart.org

H190

14.

15.

16.

17.

18.

19.

20. 21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

ACE2 AND GLUTAMATERGIC INPUT TO THE RVLM

tation in normal and hypertensive rats. Am J Physiol Heart Circ Physiol 303: H910 –H917, 2012. Feng Y, Xia H, Cai Y, Halabi CM, Becker LK, Santos RA, Speth RC, Sigmund CD, Lazartigues E. Brain-selective overexpression of human angiotensin-converting enzyme type 2 attenuates neurogenic hypertension. Circ Res 106: 373–382, 2010. Feng Y, Yue X, Xia H, Bindom SM, Hickman PJ, Filipeanu CM, Wu G, Lazartigues E. Angiotensin-converting enzyme 2 overexpression in the subfornical organ prevents the angiotensin II-mediated pressor and drinking responses and is associated with angiotensin II type 1 receptor downregulation. Circ Res 102: 729 –736, 2008. Gebke E, Muller AR, Jurzak M, Gerstberger R. Angiotensin II-induced calcium signalling in neurons and astrocytes of rat circumventricular organs. Neuroscience 85: 509 –520, 1998. Gironacci MM, Valera MS, Yujnovsky I, Pena C. Angiotensin-(1–7) inhibitory mechanism of norepinephrine release in hypertensive rats. Hypertension 44: 783–787, 2004. Guimaraes PS, Santiago NM, Xavier CH, Velloso EP, Fontes MA, Santos RA, Campagnole-Santos MJ. Chronic infusion of angiotensin(1–7) into the lateral ventricle of the brain attenuates hypertension in DOCA-salt rats. Am J Physiol Heart Circ Physiol 303: H393–H400, 2012. Guo JM, Liu AJ, Zang P, Dong WZ, Ying L, Wang W, Xu P, Song XR, Cai J, Zhang SQ, Duan JL, Mehta JL, Su DF. ALDH2 protects against stroke by clearing 4-HNE. Cell Res 23: 915–930, 2013. Guyenet PG. The sympathetic control of blood pressure. Nat Rev Neurosci 7: 335–346, 2006. Ito S, Hiratsuka M, Komatsu K, Tsukamoto K, Kanmatsuse K, Sved AF. Ventrolateral medulla AT1 receptors support arterial pressure in Dahl salt-sensitive rats. Hypertension 41: 744 –750, 2003. Ito S, Komatsu K, Tsukamoto K, Sved AF. Excitatory amino acids in the rostral ventrolateral medulla support blood pressure in spontaneously hypertensive rats. Hypertension 35: 413–417, 2000. Ito S, Komatsu K, Tsukamoto K, Sved AF. Tonic excitatory input to the rostral ventrolateral medulla in Dahl salt-sensitive rats. Hypertension 37: 687–691, 2001. Ito S, Sved AF. Tonic glutamate-mediated control of rostral ventrolateral medulla and sympathetic vasomotor tone. Am J Physiol Regul Integr Comp Physiol 273: R487–R494, 1997. Kar S, Gao L, Belatti DA, Curry PL, Zucker IH. Central angiotensin (1–7) enhances baroreflex gain in conscious rabbits with heart failure. Hypertension 58: 627–634, 2011. Katovich MJ, Grobe JL, Huentelman M, Raizada MK. Angiotensinconverting enzyme 2 as a novel target for gene therapy for hypertension. Exp Physiol 90: 299 –305, 2005. Kleiber AC, Zheng H, Sharma NM, Patel KP. Chronic AT1 receptor blockade normalizes NMDA-mediated changes in renal sympathetic nerve activity and NR1 expression within the PVN in rats with heart failure. Am J Physiol Heart Circ Physiol 298: H1546 –H1555, 2010. Kubo T, Kihara M, Misu Y. Genetically altered brain amino acid metabolism in spontaneously hypertensive rats: a study by using young spontaneously hypertensive rats and renal hypertensive rats. Clin Exp Hypertens A 12: 1191–1201, 1990. Kurihara T, Ozawa Y, Ishida S, Okano H, Tsubota K. Renin-Angiotensin system hyperactivation can induce inflammation and retinal neural dysfunction. Int J Inflam 2012: 581695, 2012. Latchford KJ, Ferguson AV. ANG II-induced excitation of paraventricular nucleus magnocellular neurons: a role for glutamate interneurons. Am J Physiol Regul Integr Comp Physiol 286: R894 –R902, 2004. Li DP, Chen SR, Pan HL. Angiotensin II stimulates spinally projecting paraventricular neurons through presynaptic disinhibition. J Neurosci 23: 5041–5049, 2003. Lin JC, Tsao WL, Wang Y. Cardiovascular effects of NMDA in the RVLM of spontaneously hypertensive rats. Brain Res Bull 37: 289 –294, 1995. Muratani H, Ferrario CM, Averill DB. Ventrolateral medulla in spontaneously hypertensive rats: role of angiotensin II. Am J Physiol Regul Integr Comp Physiol 264: R388 –R395, 1993. Nishihara M, Hirooka Y, Matsukawa R, Kishi T, Sunagawa K. Oxidative stress in the rostral ventrolateral medulla modulates excitatory and inhibitory inputs in spontaneously hypertensive rats. J Hypertens 30: 97–106, 2012. Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates (3rd ed.). New York: Academic, 1998.

36. Peng J, Wang YK, Wang LG, Yuan WJ, Su DF, Ni X, Deng XM, Wang WZ. Sympathoinhibitory mechanism of moxonidine: role of the inducible nitric oxide synthase in the rostral ventrolateral medulla. Cardiovasc Res 84: 283–291, 2009. 37. Peng JF, Wu ZT, Wang YK, Yuan WJ, Sun T, Ni X, Su DF, Wang W, Xu MJ, Wang WZ. GABAergic mechanism in the rostral ventrolateral medulla contributes to the hypotension of moxonidine. Cardiovasc Res 89: 473–481, 2011. 38. Poage RE, Meriney SD. Presynaptic calcium influx, neurotransmitter release, and neuromuscular disease. Physiol Behav 77: 507–512, 2002. 39. Seyedabadi M, Goodchild AK, Pilowsky PM. Differential role of kinases in brain stem of hypertensive and normotensive rats. Hypertension 38: 1087–1092, 2001. 40. Snider BJ, Choi J, Turetsky DM, Canzoniero LM, Sensi SL, Sheline CT, Wang X, Yu SP, Choi DW. Nitric oxide reduces Ca2⫹ and Zn2⫹ influx through voltage-gated Ca2⫹ channels and reduces Zn2⫹ neurotoxicity. Neuroscience 100: 651–661, 2000. 41. Sriramula S, Cardinale JP, Lazartigues E, Francis J. ACE2 overexpression in the paraventricular nucleus attenuates angiotensin II-induced hypertension. Cardiovasc Res 92: 401–408, 2011. 42. Stragier B, Hristova I, Sarre S, Ebinger G, Michotte Y. In vivo characterization of the angiotensin-(1–7)-induced dopamine and gammaaminobutyric acid release in the striatum of the rat. Eur J Neurosci 22: 658 –664, 2005. 43. Sved AF, Ito S, Yajima Y. Role of excitatory amino acid inputs to the rostral ventrolateral medulla in cardiovascular regulation. Clin Exp Pharmacol Physiol 29: 503–506, 2002. 44. Tsuchihashi T, Kagiyama S, Ohya Y, Abe I, Fujishima M. Antihypertensive treatment and the responsiveness to glutamate in ventrolateral medulla. Hypertension 31: 73–76, 1998. 45. Vickers C, Hales P, Kaushik V, Dick L, Gavin J, Tang J, Godbout K, Parsons T, Baronas E, Hsieh F, Acton S, Patane M, Nichols A, Tummino P. Hydrolysis of biological peptides by human angiotensin-converting enzymerelated carboxypeptidase. J Biol Chem 277: 14838–14843, 2002. 46. Wang G, Sarkar P, Peterson JR, Anrather J, Pierce JP, Moore JM, Feng J, Zhou P, Milner TA, Pickel VM, Iadecola C, Davisson RL. COX-1-derived PGE2 and PGE2 type 1 receptors are vital for angiotensin II-induced formation of reactive oxygen species and Ca2⫹ influx in the subfornical organ. Am J Physiol Heart Circ Physiol 305: H1451–H1461, 2013. 47. Wang HJ, Pan YX, Wang WZ, Gao L, Zimmerman MC, Zucker IH, Wang W. Exercise training prevents the exaggerated exercise pressor reflex in rats with chronic heart failure. J Appl Physiol 108: 1365–1375, 2010. 48. Wang J, Shen LL, Cao YX, Sun ZJ, Wang Q, Zhu DN. Role of angiotensin-(1–7) in amino-acid-neurotransmitter-mediated blood pressure regulation in rat rostral ventrolateral medulla. Sheng Li Xue Bao 53: 1–6, 2001. 49. Wang WZ, Gao L, Wang HJ, Zucker IH, Wang W. Tonic glutamatergic input in the rostral ventrolateral medulla is increased in rats with chronic heart failure. Hypertension 53: 370 –374, 2009. 50. Wang WZ, Wang LG, Gao L, Wang W. Contribution of AMPA/kainate receptors in the rostral ventrolateral medulla to the hypotensive and sympathoinhibitory effects of clonidine. Am J Physiol Regul Integr Comp Physiol 293: R1232–R1238, 2007. 51. Whitaker AM, Molina PE. Angiotensin (1–7) contributes to nitric oxide tonic inhibition of vasopressin release during hemorrhagic shock in acute ethanol intoxicated rodents. Life Sci 93: 623–629, 2013. 52. Xia H, Lazartigues E. Angiotensin-converting enzyme 2 in the brain: properties and future directions. J Neurochem 107: 1482–1494, 2008. 53. Xia H, Lazartigues E. Angiotensin-converting enzyme 2: central regulator for cardiovascular function. Curr Hypertens Rep 12: 170 –175, 2010. 54. Xu P, Sriramula S, Lazartigues E. ACE2/Ang-(1–7)/Mas pathway in the brain: the axis of good. Am J Physiol Regul Integr Comp Physiol 300: R804 –R817, 2011. 55. Yamazato M, Yamazato Y, Sun C, ez-Freire C, Raizada MK. Overexpression of angiotensin-converting enzyme 2 in the rostral ventrolateral medulla causes long-term decrease in blood pressure in the spontaneously hypertensive rats. Hypertension 49: 926 –931, 2007. 56. Zha YP, Wang YK, Deng Y, Zhang RW, Tan X, Yuan WJ, Deng XM, Wang WZ. Exercise training lowers the enhanced tonically active glutamatergic input to the rostral ventrolateral medulla in hypertensive rats. CNS Neurosci Ther 19: 244 –251, 2013. 57. Zhu DN, Moriguchi A, Mikami H, Higaki J, Ogihara T. Central amino acids mediate cardiovascular response to angiotensin II in the rat. Brain Res Bull 45: 189 –197, 1998.

AJP-Heart Circ Physiol • doi:10.1152/ajpheart.00518.2013 • www.ajpheart.org