In addition to its well-known actions on the vasculature,

Transcription

1 NADPH Oxidase Derived Superoxide Anion Mediates Angiotensin II Induced Pressor Effect via Activation of p38 Mitogen Activated Protein Kinase in the Rostral Samuel H.H. Chan, Kuei-Sen Hsu, Chiung-Chun Huang, Ling-Lin Wang, Chen-Chun Ou, Julie Y.H. Chan Abstract The rostral ventrolateral medulla (RVLM), where sympathetic premotor neurons are located, is a central site via which angiotensin II (Ang II) elicits its pressor effect. We tested the hypothesis that NADPH oxidase-derived superoxide anion (O 2 ) in the RVLM mediates Ang II induced pressor response via activation of mitogen-activated protein kinase (MAPK) signaling pathways. Bilateral microinjection of Ang II into the RVLM resulted in an angiotensin subtype 1 (AT 1 ) receptor-dependent phosphorylation of p38 MAPK and extracellular signal-regulated protein kinase (ERK)1/2, but not stress-activated protein kinase/jun N-terminal kinase (SAPK/JNK), in the ventrolateral medulla. The Ang II induced p38 MAPK or ERK1/2 phosphorylation was attenuated by application into the RVLM of a NADPH oxidase inhibitor, diphenyleneiodonium chloride (DPI), an antisense oligonucleotide that targets against p22phox or p47phox subunit of NADPH oxidase mrna, or the superoxide dismutase mimetic tempol. DPI or antisense p22phox or p47phox oligonucleotide treatment also attenuated the AT 1 receptor-dependent increase in O 2 production in the ventrolateral medulla elicited by Ang II at the RVLM. Functionally, Ang II elicited pressor response in the RVLM was attenuated by DPI, tempol, or a p38 MAPK inhibitor, SB The AT 1 receptor-mediated enhancement of the frequency of glutamate-sensitive spontaneous excitatory postsynaptic currents induced by Ang II in RVLM neurons was also abolished by SB These results suggest that NADPH oxidase-derived O 2 underlies the activation of p38 MAPK or ERK1/2 by Ang II in the ventrolateral medulla. Furthermore, the p38 MAPK signaling pathway may mediate Ang II induced pressor response via enhancement of presynaptic release of glutamate to RVLM neurons. (Circ Res. 2005;97: ) Key Words: mitogen-activated protein kinases angiotensin II superoxide anion NADPH oxidase rostral ventrolateral medulla blood pressure In addition to its well-known actions on the vasculature, angiotensin II (Ang II) also plays a critical role in central regulation of circulatory functions. 1,2 One brain target where Ang II exerts its influence on cardiovascular control is rostral ventrolateral medulla (RVLM), 3,4 where premotor neurons that maintain tonic sympathetic vasomotor outflow are located. 5 The RVLM contains a high density of Ang II receptors 6 and is a major site of the sympathoexcitatory and pressor actions of the octapeptide. 7,8 It is generally accepted that activation of Ang II type 1 (AT 1 ) receptors in the RVLM contributes mainly to the cardiovascular effects of Ang II. 9,10 The reactive oxygen species, particularly superoxide anion (O 2 ), is an important intracellular messenger for brain Ang II. The octapeptide increases the activity of NADPH oxidase, the major source of O 2 in the vasculature, 11 and enhances O 2 production in the central nervous system. 12,13 Intracerebroventricular infusion of NADPH oxidase inhibitor antagonizes the increase in renal sympathetic nerve activity or pressor response induced centrally by Ang II. 13,14 In the brain, overexpression of superoxide dismutase (SOD), the enzyme responsible for O 2 breakdown, also abolishes the central pressor effect of the octapeptide. 12,15 A potential intracellular signaling pathway that mediates Ang II elicited responses is for O 2 to activate mitogen-activated protein kinases (MAPKs). 16 At least 6 major mammalian MAPK subfamilies have been identified, of which p38 MAPK, extracellular signal-regulated protein kinase 1/2 (ERK1/2), and stress-activated protein kinase/jun N-terminal Original received April 18, 2005; resubmission received July 26, 2005; revised resubmission received August 28, 2005; accepted August 30, From the Center for Neuroscience, National Sun Yat-sen University (S.H.H.C.), Kaohsiung; Department of Pharmacology, National Cheng Kung University (K.S.H., C.C.H.), Tainan; and Department of Medical Education and Research, Kaohsiung Veterans General Hospital (L.L.W., C.C.O., J.Y.H.C.), Kaohsiung, Taiwan, Republic of China. Correspondence to Julie Y.H. Chan, PhD, Department of Medical Education and Research, Kaohsiung Veterans General Hospital, Taiwan 813, Republic of China. yhwa@isca.vghks.gov.tw 2005 American Heart Association, Inc. Circulation Research is available at DOI: /01.RES C0 772

2 Chan et al O 2 and MAPKs in Angiotensin II Pressor Response 773 kinase (SAPK/JNK) are the best characterized. 17 Activation of MAPKs is involved in Ang II promoted cardiac hypertrophy 18 or contractility of mesenteric resistance arteries 19 in hypertensive rats. In vascular smooth muscle cells and cardiac myocytes, oxidative stress promoted by Ang II leads to activation of MAPKs. 16,18 20 The specific central site at which NADPH oxidase-derived O 2 mediates pressor response to brain Ang II and the O 2 - sensitive intracellular signaling pathway involved remain to be delineated. The present study evaluated the hypothesis that NADPH oxidase-derived O 2 in the RVLM mediates Ang II induced pressor response via activation of MAPK signaling pathways. Our results demonstrate that activation of p38 MAPK by NADPH oxidase-derived O 2 in the RVLM plays a critical role in the short-term AT 1 receptor-mediated pressor response to Ang II, possibly via an enhancement of presynaptic release of glutamate to RVLM neurons. Materials and Methods A brief summary of our experimental strategies is provided. Detailed description of materials and methods can be found in the online data supplement available at Activation of MAPKs in by Ang II Adult male Sprague-Dawley rats (The Experimental Animal Center of the National Applied Research Laboratories, Taiwan) maintained under propofol anesthesia (30 mg kg 1 h 1 ) received bilateral microinjection of Ang II into the RVLM, given alone or together with the AT 1 receptor antagonist losartan or the AT 2 receptor antagonist PD A total volume of 50 nl was delivered to each side of the RVLM over 1 to 2 minutes to allow for complete diffusion of the test agents. The ventrolateral medulla that contained bilateral RVLM was removed at various postinjection time intervals, and the expression of total or phosphorylated p38 MAPK, ERK1/2, or SAPK/JNK was determined by Western blot analysis. Involvement of NADPH Oxidase or Superoxide in Ang II Induced MAPK Phosphorylation in We ascertained a causative involvement of NADPH oxidase or O 2 in Ang II induced MAPK phosphorylation in the ventrolateral medulla by bilateral comicroinjection of Ang II and the NADPH oxidase inhibitor diphenylene iodinium (DPI) or the SOD mimetic tempol into the RVLM. To further investigate the involvement of the p22phox or p47phox subunit of NADPH oxidase in Ang II induced MAPK phosphorylation, antisense (ASON) or sense (SON) p22phox or p47phox oligonucleotide was microinjected bilaterally into the RVLM 24 hours before Ang II administration. The effectiveness of ASON to block p22phox or p47phox mrna expression was confirmed by real-time reverse-transcription polymerase chain reaction analysis. Involvement of NADPH Oxidase in Ang II Induced Superoxide Production in The causative role of NADPH oxidase in Ang II induced O 2 production was examined by pretreatment with losartan or PD123319, NADPH oxidase inhibitors, apocynin (APO) or DPI, or p22phox or p47phox ASON or SON. Losartan, PD123319, APO, or DPI was comicroinjected with, and ASON or SON was microinjected 24 hours before, Ang II application. O 2 production in the ventrolateral medulla was determined by the lucigenin-enhanced chemiluminescence method. Angiotensin Receptor-Dependent Phosphorylation of p47phox Subunit of NADPH Oxidase in Serine phosphorylation of the p47phox subunit of NADPH oxidase induced by Ang II was determined by immunoprecipitation followed by immunoblotting, and the subtype of Ang II receptors involved was studied by bilateral comicroinjection of losartan or PD into the RVLM. Intracellular Signaling for Pressor Response to Ang II Whether NADPH oxidase, O 2, or MAPK is engaged in pressor response to Ang II was studied by bilateral comicroinjection of the octapeptide with DPI, tempol, the p38 MAPK inhibitor SB230850, the ERK1/2 inhibitor U0126, or the JNK inhibitor SP into the RVLM. Cellular Mechanisms Underlying Activation of p38 MAPK in AT 1 Receptor-Dependent Excitation of RVLM Neurons by Ang II To further delineate the cellular mechanisms via which activation of p38 MAPK underlies the AT 1 receptor-dependent pressor action of Ang II in the RVLM, we examined the effects of the octapeptide on spontaneous excitatory postsynaptic currents (sepscs) recorded from RVLM neurons, using the whole-cell patch-clamp technique in brain stem slice preparations. We also evaluated the effects of losartan, PD123319, and SB on the Ang II actions. Results Selective Activation of MAPKs by Ang II in Bilateral microinjection of Ang II (50 pmol) into the RVLM resulted in activation of p38 MAPK or ERK1/2, but not SAPK/JNK, in the ventrolateral medulla (Figure 1A through 1C). Phosphorylation of p38 MAPK or ERK1/2 was detected 5 minutes after Ang II treatment, lasting 10 or 30 minutes, respectively (Figure 1D). The Ang II induced p38 MAPK (Figure 2A) or ERK1/2 (Figure 2B) phosphorylation was significantly antagonized by coadministration of losartan (2 nmol) but not PD (2 nmol). Both Ang II antagonists, on the other hand, did not affect SAPK/JNK phosphorylation in the RVLM (Figure 2C). Furthermore, none of those treatments (Figure 1A through 1C and Figure 2A through C) elicited a discernible effect on total p38 MAPK, ERK1/2, or SAPK/JNK levels in the ventrolateral medulla. Effects of DPI or Tempol on MAPK Activation by Ang II in Bilateral comicroinjection of DPI (1.5 nmol) or tempol (50 nmol) into the RVLM significantly blunted the peak activation of p38 MAPK (Figure 2A) or ERK1/2 (Figure 2B) by Ang II (50 pmol) in the ventrolateral medulla. DPI or tempol, on the other hand, elicited minimal effect on basal expression of MAPKs (Figure 2A through 2C) or SAPK/JNK phosphorylation (Figure 2C).

3 774 Circulation Research October 14, 2005 microinjection of Glu (2 nmol) into the RVLM did not influence p38 MAPK (Figure 3A) or ERK1/2 (Figure 3B) phosphorylation in the ventrolateral medulla. Differential Effects of Losartan, PD123319, or Tempol on Ang II Induced O 2 Production in Figure 4A shows that microinjection of Ang II (50 pmol) bilaterally into the RVLM increased O 2 production significantly at 5 minutes after injection, lasting 20 minutes. The peak O 2 production induced by Ang II was significantly inhibited by coadministration of losartan (2 nmol) or tempol (50 nmol), but not PD (2 nmol). Losartan, PD123319, and tempol alone exerted minimal alteration in basal O 2 level in the RVLM. Effects of p22phox or p47phox ASON or NADPH Oxidase Inhibitors on Ang II Induced O 2 Production in Ang II induced O 2 production in the ventrolateral medulla was also attenuated by pretreatment with p22phox or p47phox ASON (50 pmol; Figure 4B). Control treatment of p22phox Figure 1. A through C, Representative Western blots of phosphorylated (upper left panels) or total (lower left panels) p38 MAPK, ERK1/2, and SAPK/JNK, or densitometric analysis of phosphorylated protein levels (right panels) detected from the ventrolateral medulla in rats 0, 5, 10, or 20 minutes after bilateral microinjection of Ang II (50 pmol) into the RVLM. D, Time course of phosphorylation of p38 MAPK, ERK1/2, and SAPK/JNK in the ventrolateral medulla by Ang II. Values are mean SEM of quadruplicate analyses on samples pooled from 5 to 6 animals in each group. *P 0.05 vs control (C) group in the Scheffé multiple range analysis. Effects of p22phox or p47phox ASON on Ang II Activation of p38 MAPK or ERK1/2 in Whereas application of p22phox or p47phox SON was ineffective, bilateral microinjection of p22phox or p47phox ASON (50 pmol) into the RVLM 24 hours before Ang II administration significantly attenuated p38 MAPK or ERK1/2 phosphorylation induced in the ventrolateral medulla (Figure 3A and 3B) by the octapeptide (50 pmol). Real time reverse transcription polymerase chain reaction analysis (Figure 3C) confirmed that ASON pretreatment selectively suppressed the mrna expression of p22phox or p47phox, but not the p67phox or gp91phox subunit of NADPH oxidase, and that control SON pretreatment was ineffective. Pretreatment with p22phox or p47phox ASON also did not affect total p38 MAPK or ERK1/2 level in the RVLM (data not shown). In a separate series of control experiments, bilateral Figure 2. A through C, Effects of losartan, PD123319, DPI, and tempol on Ang II induced MAPK phosphorylation in ventrolateral medulla after bilateral microinjection of Ang II (50 pmol) into the RVLM. Phosphorylated p38 MAPK, ERK1/2, and SAPK/JNK were measured respectively at 10, 20, and 20 minutes after Ang II. Representative Western blots of phosphorylated (upper left panels) or total (lower left panels) p38 MAPK, ERK1/2, or JNK1/2 are shown on the left panels, and densitometric analysis of phosphorylated protein levels are shown on the right panels. Values are mean SEM of quadruplicate analyses on samples pooled from 5 to 6 animals in each group. *P 0.05 vs corresponding Ang II group in the Scheffé multiple range analysis.

4 Chan et al O 2 and MAPKs in Angiotensin II Pressor Response 775 Effects of Losartan and PD on Ang II Induced p47phox Phosphorylation in Immunoprecipitated p47phox subunit of NADPH oxidase in the ventrolateral medulla examined for serine phosphorylation showed significant elevation in phosphorylated p47phox that was detectable 5 minutes after bilateral microinjection of Ang II into the RVLM, lasting 20 minutes (Figure 4C). Both losartan (2 nmol) and PD (2 nmol) attenuated Ang II induced p47phox phosphorylation, although the AT 1 antagonist was more effective. Bilateral microinjection of Glu Figure 3. A and B, Representative Western blots of phosphorylated p38 MAPK or ERK1/2 (upper panels), or densitometric analysis of phosphorylated protein levels (lower panels) detected from the ventrolateral medulla in rats at 10 minutes after bilateral microinjection of Ang II (50 pmol) or Glu (2 nmol) into the RVLM. C, Real-time polymerase chain reaction results showing fold changes in p22phox, p47phox, p67phox, and gp91phox mrna expression detected from the ventrolateral medulla 10 minutes after microinjection bilaterally of Ang II (50 pmol) into the RVLM of rats, given alone or with additional pretreatment with antisense or sense p22phox or p47phox oligonucleotide, administered into the bilateral RVLM 24 hours before Ang II. Values are mean SEM of quadruplicate analyses on samples pooled from 5 to 6 animals in each group. *P 0.05 vs control (C) group; #P 0.05 vs Ang II group in the Scheffé multiple range analysis. or p47phox SON (50 pmol) was ineffective. Comicroinjection of APO (2 nmol) or DPI (1.5 nmol) also significantly blunted Ang II induced O 2 production in the ventrolateral medulla. APO and DPI, however, exerted no discernible effect on basal O 2 level in the ventrolateral medulla. Figure 4. A, Time course of change in O 2 production in the ventrolateral medulla 0, 5, 10, and 20 minutes after bilateral microinjection into the RVLM of Ang II (50 pmol) alone or in combination with losartan (2 nmol), PD (2 nmol), or tempol (50 nmol). B, O 2 production in the ventrolateral medulla detected at 10 minutes after bilateral microinjection into the RVLM of Ang II (50 pmol), given alone or with additional pretreatment with antisense or sense p22phox or p47phox oligonucleotide (50 pmol), APO (2 nmol), or DPI (1.5 nmol). C, Representative gels (insets) and densitometric analysis from immunoprecipitation (IP) followed by immunoblotting (IB) assays on phosphorylated serine residues and total p47phox NADPH oxidase subunit in the ventrolateral medulla 0, 5, 10, and 20 minutes after bilateral microinjection of Ang II (50 pmol) or Glu (2 nmol, measured at 10 minutes after injection) into the RVLM, as well as effect of losartan (2 nmol) or PD (2 nmol) on serine phosphorylation of p47phox detected at 10 minutes after Ang II. Values are mean SEM of quadruplicate analyses on samples pooled from 5 to 6 animals in each group. *P 0.05 vs control group; #P 0.05 vs Ang II group in the Scheffé multiple range analysis.

5 776 Circulation Research October 14, 2005 Effects of MAPK Inhibitors on Ang II Induced Pressor Response Bilateral coadministration of the p38 MAPK inhibitor SB (500 nmol; Figure 6A), but not the ERK1/2 inhibitor U0126 (500 nmol; Figure 6B) or the JNK inhibitor SP (500 nmol; Figure 6C), into the RVLM significantly attenuated the Ang II promoted pressor effects. None of the MAPK inhibitors, however, exerted discernible effect on basal cardiovascular parameters or Glu (2 nmol)-induced pressor response (Figure 6A through 6C). Effects of Losartan, PD123319, or p38 MAPK Inhibitor on Ang II Induced Cellular Response in RVLM Neurons We measured sepscs in the RVLM neurons that were voltage-clamped at 70 mv and were pharmacologically isolated from spontaneous inhibitory currents by including bicuculline methiodide (10 mol/l) and strychnine hydrochloride (0.5 mol/l) in the perfusate. These sepscs were totally blocked by coapplication of CNQX (20 mol/l) plus D-APV (50 mol/l) in the perfusate (data not shown), Figure 5. Temporal change in MSAP in response to bilateral microinjection of artificial cerebrospinal fluid, Ang II (50 pmol) (A), or Glu (2 nmol) (B) into the RVLM, alone or in combination with DPI (2 nmol), tempol (50 nmol), or 3-CP (50 nmol). Values are mean SEM, n 6 to 7 animals in each group. *P 0.05 vs artificial cerebrospinal fluid group; #P 0.05 vs Ang II group in the Scheffé multiple range test. Horizontal bar denotes time during which microinjection was performed. into the RVLM (2 nmol), on the other hand, did not promote p47phox phosphorylation. Effects of NADPH Oxidase Inhibitor or SOD Mimetic on Ang II Induced Pressor Response Bilateral microinjection of Ang II (50 pmol) into the RVLM evoked a short-term pressor response that lasted for 10 minutes. This pressor response was significantly blunted by coadministration of DPI (1.5 nmol; Figure 5A), APO (2 nmol; data not shown), and tempol (50 nmol; Figure 5B), but not 3-CP (50 nmol; Figure 5C), a compound that is structurally similar to tempol but that exhibits low O 2 scavenging capacity. 21 Treatment with DPI, APO, or tempol, on the other hand, did not alter basal mean systemic arterial pressure (MSAP), heart rate, or the pressor response to bilateral microinjection of Glu into the RVLM (2 nmol; Figure 5A through 5C). In addition, microinjection of Ang II, alone or in combination with various test agents, into areas adjacent to the confines of the RVLM did not result in appreciable alterations in basal MSAP or heart rate (data not shown). Figure 6. Temporal change in MSAP in response to bilateral microinjection of artificial cerebrospinal fluid, Ang II (50 pmol) (A), or Glu (2 nmol) (B) into the RVLM, alone or in combination with SB (500 nmol), U0126 (500 nmol), or SP (500 nmol). Data on artificial cerebrospinal fluid and Ang II are adapted from Figure 6. Values are mean SEM, n 5 to 6 animals in each group. *P 0.05 vs artificial cerebrospinal fluid group; #P 0.05 vs Ang II group in the Scheffé multiple range test. Horizontal bar denotes time during which microinjection was performed.

6 Chan et al O 2 and MAPKs in Angiotensin II Pressor Response 777 Figure 7. Effects of Ang II on glutamatergic sepscs. A, Sample traces (3 traces superimposed) of sepscs before and after application of Ang II (0.5 mol/l). Lower traces are the averaged sepscs of 20 events each before and after Ang II application with increasing time resolution, demonstrating the lack of effect on the amplitude and kinetics of sepscs. B, Amplitude histograms of sepscs before (B1) and during (B2) Ang II application obtained from 255 and 274 sepsc events, respectively. The threshold for peak detection was set at 3 pa, and data were binned in 1 pa intervals. C, Cumulative probability plots of sepscs before (solid line) and during (dash line) application of Ang II. D, Cumulative inter-event interval distribution illustrating a significant decrease in the inter-event interval (ie, increased frequency) during Ang II application. E and F, Summary of the effect of Ang II (0.5 mol/l) on the average amplitude (E) or frequency (F) of sepscs (n 5). Values are mean SEM. *P 0.05 vs baseline in the Student s unpaired t test. K-S denotes Kolmogorov-Smirnov test. The data shown in A, B, C, and D were taken from the same cell. Holding potential, 70 mv. confirming that they are glutamate receptor-mediated synaptic events. In 5 RVLM neurons tested, Ang II (0.5 mol/l) significantly increased the mean frequency of the sepscs from to Hz (Figure 7A and 7F), and significantly shifted the cumulative inter-event interval distribution of sepscs to shorter intervals (Figure 7D). Both amplitude histogram (Figure 7B1 and 7B2) and cumulative probability plots (Figure 7C), however, showed no significant effect of Ang II on the sepsc amplitude. The mean amplitude of sepscs ( pa) recorded in the presence of Ang II (0.5 mol/l) was comparable to that of sepscs ( pa) recorded under baseline condition (Figure 7E). Pretreatment with losartan (5 mol/l) or SB (1 mol/l), but not PD (5 mol/l), significantly blocked the enhancement of Ang II (0.5 mol/l) on sepsc frequency recorded from RVLM neurons (Figure 8B). None of these treatments elicited discernible effects on control Figure 8. Effects of Ang II (0.5 mol/l), given alone or after pretreatment with losartan (5 mol/l), PD (5 mol/l), or SB (1 mol/l), on the amplitude (A) or frequency (B) of sepscs recorded from RVLM neurons. Values are mean SEM. *P 0.05 vs control baseline; #P 0.05 vs Ang II group in the Student s unpaired t test. frequency of sepsc (data not shown), nor did they affect sepsc amplitude (Figure 8A) under baseline conditions or after application of Ang II. Discussion It is well documented that Ang II participates in central cardiovascular control by eliciting sympathoexcitatory and pressor actions 7,8 through activation of AT 1 receptors in the RVLM. 9,10 In the present study, we identified that a novel intracellular signaling mechanism that underlies this action of the octapeptide entails AT 1 receptor-dependent increases in sepsc frequency by activation of p38 MAPK via NADPH oxidase-derived O 2 in the RVLM. Our study provided the first biochemical demonstration that Ang II induces AT 1 receptor-dependent phosphorylation of p38 MAPK or ERK1/2 in the ventrolateral medulla. Activation of p38mapk and ERK1/2 by Ang II has been

7 778 Circulation Research October 14, 2005 reported in mesenteric smooth muscle cells, 19,22,23 aortae, 18,24 or cardiomyocytes. 18,24,25 Similar to previous reports, 19,22 we found that the time course of Ang II induced p38 MAPK phosphorylation was discernibly different from that of ERK1/2. We interpret these temporally different expression profiles to suggest that different upstream signaling pathways exist in linking AT 1 receptor activation to phosphorylation of p38 MAPK or ERK1/2 in the RVLM. In support of this interpretation, inhibition of the Src family of protein tyrosine kinase, which plays a key role in coupling cell surface receptors with the cytoplasmic signaling machinery, 26 abrogates Ang II induced ERK1/2 activation but only partially reverses p38 MAPK activity in vascular smooth muscle cells. 19 In contrast to observations 18,22,24 in aortae or cardiomyocytes, we found that activation of AT 1 receptors by Ang II did not induce SAPK/JNK phosphorylation in the ventrolateral medulla over 60 minutes. Because SAPK/JNK phosphorylation in vascular smooth muscle cells commences at 15 minutes and peaks 30 minutes after application of the octapeptide, 22 the possibility that Ang II induced activation of SAPK/JNK manifests a slower time course in the RVLM than phosphorylation of p38 MAPK or ERK1/2 is deemed unlikely. Another intriguing finding of the present study is the identified role for NADPH oxidase-derived O 2 as a second messenger in the RVLM that mediates Ang II induced activation of p38 MAPK or ERK1/2 via AT 1 receptors. Activation of NADPH oxidase is a multi-step process that is initiated by serine phosphorylation of the cytosolic regulatory p47phox subunit. 27 After translocation to the membrane, the activated p47phox subunit associates with the membranebound gp91phox and p22phox subunits to bring forth enzymatic activity. 28 In this regard, we have detected the presence of membrane-bound p22phox and gp91phox, or cytosolic p47phox and p67phox subunit mrna in the ventrolateral medulla. More importantly, we demonstrated that phosphorylation of serine residues of p47phox induced by Ang II in the RVLM, which was blunted by losartan, manifested a temporal profile that correlated positively with O 2 production induced by the octapeptide. Furthermore, pretreatment with p22phox or p47phox ASON in the bilateral RVLM reversed Ang II induced O 2 production and p38 MAPK or ERK1/2 phosphorylation. It follows that by acting on the AT 1 receptors in the RVLM, Ang II may initiate a signaling cascade that leads to phosphorylation of p38 MAPK or ERK1/2 mediated by NADPH oxidase-derived O 2 after the induction of serine phosphorylation of the p47phox subunit. 29 On the basis of results from vascular smooth muscle cells, the potential upstream cellular signals linking Ang II to p47phox phosphorylation in the RVLM include Rac, 14 phospholipase D, 30 protein kinase C, 30 or Rho-kinase. 31 A MAPK-dependent pathway has been demonstrated to be necessary for Ang II induced pressor response in the RVLM. 32 Functional evaluations in the present study extended this demonstration by showing, for the first time, that O 2 -sensitive activation of p38 MAPK in the RVLM plays an active role specifically in the short-term pressor response to Ang II. In primary cultured brain stem neurons, the octapeptide increases the neuronal firing rate by inhibition of the delayed rectifier potassium current, 33 an action that is associated with an increase in NADPH oxidase activity and O 2 production. NADPH oxidase-dependent O 2 is also involved in Ang II induced increase in intracellular Ca 2 concentration in neurons. 34 We further demonstrated that, at the cellular level, Ang II elicits excitatory actions on RVLM neurons by increasing the frequency, but not the amplitude, of sepscs. A change in the amplitude of sepscs is classically interpreted as a postsynaptic modification, whereas a change in their frequency points toward presynaptic mechanisms that increase the probability of transmitter release. 35,36 Together with the observations that the sepscs were completely antagonized by N-methyl-D-aspartate and non N-methyl-Daspartate antagonist, our results suggest that an enhancement of presynaptic release of glutamate may underlie the excitatory actions of Ang II on RVLM neurons. More importantly, this modulatory action is mediated by AT 1 receptors and requires activation of p38 MAPK. The lack of effect of Ang II on the amplitude of sepscs also implies that synaptic modulation by the octapeptide on RVLM neurons is not mediated by a change in postsynaptic sensitivity to glutamate. Whereas results from our biochemical experiments indicated that Ang II promoted phosphorylation of ERK1/2 in the RVLM via NADPH oxidase-derived O 2, those from our physiological experiments revealed that this MAPK member may not participate in the short-term pressor response to the octapeptide. ERK1/2 is a major growth-signaling kinase that regulates the activity of a number of transcription factors, leading to induction of downstream target genes encoding proteins for smooth muscle cell growth, proliferation, and/or migration. 16,19 In this regard, ERK1/2-dependent, p38 MAPK-independent pathways activate the transcription factor AP-1, leading to upregulation of c-fos mrna in vascular smooth muscle cells. 19 It is therefore possible that Ang II induced ERK1/2 phosphorylation may function in central cardiovascular regulation via cellular events in the RVLM that are involved in transcription regulation. This possibility, however, requires further investigation. Our results showed that peak p38 MAPK phosphorylation (Figure 3) or O 2 generation (Figure 4) was detected at a time point during which Ang II promoted pressor response (Figure 5 and Figure 6) has already reached its peak. This seemingly contradicts our conclusion that activation of p38 MAPK by the NADPH oxidase-derived O 2 in the RVLM mediates the pressor response to Ang II. This discrepancy essentially arises from technical differences between physiological and biochemical experiments. Whereas data collection in the former is in a continuous manner, the earliest feasible time point for collection of samples from the ventrolateral medulla is 5 minutes after application of test agents into the RVLM. Most importantly, substantiation of our conclusion is provided by the demonstration that SB , DPI, and tempol elicited comparable suppression on Ang II induced pressor response in the RVLM (Figure 5 and Figure 6). Our control results with Glu additionally established that our observations are specific to the actions of Ang II in the RVLM. We noted that compared with an almost complete blockade of Ang II promoted pressor response by the O 2 scavenger

8 Chan et al O 2 and MAPKs in Angiotensin II Pressor Response 779 (Figure 5B), NADPH oxidase inhibitor elicited only a partial reversal (Figure 5A). These observations suggest that O 2 generated from other cellular sources, eg, mitochondria, 37 may also participate in Ang II promoted cardiovascular responses in the RVLM. Ang II induced inhibition of the delayed rectifier potassium currents in cultured brain stem neurons is mediated by NADPH oxidase-dependent O 2, but not a hydrogen peroxide mechanism. 33 This dampens the possibility that, as a product of degradation of O 2, hydrogen peroxide in the RVLM may also participate in our observed physiological and biochemical responses to Ang II at the RVLM. Acute microinjection of Ang II into the RVLM in the present study may also account for the lack of upregulation of almost all NADPH oxidase subunits seen after long-term (24 hours to 7 days) treatment with the octapeptide. 30,31 In conclusion, the present study unveiled a novel signaling pathway for the short-term pressor response to Ang II that entails AT 1 receptor-dependent activation of p38 MAPK via NADPH oxidase-derived O 2 in the RVLM. We further showed that the excitatory action of Ang II engages an enhancement of presynaptic glutamate release to RVLM neurons, a synaptic modulation that is AT 1 receptor- and p38 MAPK-dependent. Increased levels of brain Ang II and O 2 in the RVLM are associated with sympathoexcitation during hypertension, 38,39 acute stress, 40 or chronic heart failure. 13 Furthermore, inhibition of MAPKs in the RVLM normalizes arterial pressure in spontaneously hypertensive rats. 32 Antagonism of O 2 -dependent p38 MAPK signaling pathway may therefore represent a novel approach to counteract sympathoexcitation in cardiovascular disorders that are associated with enhanced brain Ang II. Acknowledgments This study was supported by research grants NSC B PAE, NSC B PAE (S.H.H.C.), and NSC B075-B-001-PAE (J.Y.H.C.) from National Science Council, and VGHKS (J.Y.H.C.) from Kaohsiung Veterans General Hospital, Taiwan, Republic of China. References 1. Steckelings U, Lebrun C, Qadri F, Veltmar A, Unger T. Role of brain angiotensin in cardiovascular regulation. J Cardiovasc Pharmacol. 1992; 19:S72 S Phillips MI, Summers C. Angiotensin II in central nervous system physiology. Regul Pept. 1998;78: Fontes MA, Pinge MC, Naves V, Campagnole-Santo MJ, Lopes OU, Khosla MC, Santos RA. Cardiovascular effects produced by microinjection of angiotensins and angiotensin antagonists into the ventrolateral medulla of freely moving rats. Brain Res. 1997;750: Dampney RA, Horiuchi J, Tagawa T, Fontes MA, Potts PD, Polson JW. Medullary and supramedullary mechanisms regulating sympathetic vasomotor tone. Acta Physiol Scand. 2003;177: Ross CA, Ruggiero DA, Park DH, Joh TH, Sved AF, Fernandez-Pardal J, Saavedra JM, Reis DJ. Tonic vasomotor control by the rostral ventrolateral medulla: effect of electrical or chemical stimulation of the area containing C1 adrenaline neurons on arterial pressure, heart rate, and plasma catecholamine and vasopressin. J Neurosci. 1984;4: Allen AM, Moeller I, Jenkins TA, Zhuo J, Aldred GP, Chai SY, Mendelsohn FA. Angiotensin receptors in the nervous system. Brain Res Bull. 1998;47: Hirooka Y, Potts PD, Dampney RAL. Role of angiotensin II receptor subtypes in mediating the sympathoexcitatory effects of exogenous and endogenous angiotensin peptides in the rostral ventrolateral medulla of the rabbit. Brain Res. 1997;772: Dampney RAL, Fontes MAP, Hirooka Y, Horiuchi J, Potts PD, Tagawa T. Role of angiotensin II receptors in the regulation of vasomotor neurons in the ventrolateral medulla. Clin Expt Pharmacol Physiol. 2002;29: 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. 1994;665: Veerasingham SJ, Raizada MK. Brain rennin-angiotensin system dysfunction in hypertension: recent advances and perspectives. Br J Pharmacol. 2003;139: Griendling KK, Minieri CA, Ollerenshaw JD, Alexander RW. Angiotensin II stimulates NADH and NADPH oxidase activity in cultured vascular smooth muscle cells. Circ Res. 1994;74: Zimmermann MC, Lazartigues E, Lang JA, Sinnayah P, Ahmad IM, Spitz DR, Davisson RL. Superoxide mediates the actions of angiotensin II in the central nervous system. Circ Res. 2002;91: Gao L, Wang W, Li YL, Schultz HD, Liu D, Cornish KG, Zucker IH. Superoxide mediates sympathoexcitation in heart failure: roles of angiotensin II and NAD(P)H oxidase. Circ Res. 2004;95: Zimmerman MC, Dunlay RP, Lazartigues E, Zhang Y, Sharama RV, Engelhardt JF, Davisson RL. Requirement for Rac 1-dependent NADPH oxidase in the cardiovascular and dipsogenic actions of angiotensin II in the brain. Circ Res. 2004;95: Zimmerman MC, Lazartigues E, Sharma RV, Davisson RL. Hypertension caused by angiotensin II infusion involves increased superoxide production in the central nervous system. Circ Res. 2004;95: Griendling KK, Sorescu D, Lassegue B, Ushio-Fukai M. Modulation of protein kinase activity and gene expression by reactive oxygen species and their role in vascular physiology and pathophysiology. Arterioscler Thromb Vasc Biol. 2000;20: Pearson G, Robinson F, Beers Gibson T, Xu BE, Karandikar M, Berman K, Cobb MH. Mitogen-activated protein kinase pathways: regulation and physiological functions. Endoc Rev. 2001;22: Izumi Y, Kim S, Zhan Y, Namba M, Yasumoto H, Iwao H. Important role of angiotensin II-mediated c-jun NH 2 -terminal kinase activation in cardiac hypertrophy in hypertensive rats. Hypertension. 2000;36: Touzy RM, He G, Mabrouk ME, Diep Q, Mardigyan V, Schiffrin EL. Differential activation of extracellular signal-regulated protein kinase 1/2 and p38 mitogen activated-protein kinase by AT 1 receptors in vascular smooth muscle cells from Wistar-Kyoto rats and spontaneously hypertensive rats. J Hypertens. 2001;19: Touyz RM, Deschepper C, Park JB, He G, Chen X, Neves MF, Virdis A, Schiffrin EL. Inhibition of mitogen-activated protein/extracellular signalregulated kinase improves endothelial function and attenuates Ang II-induced contractility of mesenteric resistance arteries from spontaneously hypertensive rats. J Hypertens. 2002;20: Krishna MC, Grahame DA, Samuni A, Mitchell JB, Russo A. Oxoammonium cation intermediate in the nitroxide-catalyzed dismutation of superoxide. Proc Natl Acad Sci U S A. 1992;89: Viedt C, Soto U, Krieger-Brauer HI, Fei J, Elsing C, Kubler W, Kreuzer J. Differential activation of mitogen-activated protein kinases in smooth muscle cells by angiotensin II: involvement of p22phox and reactive oxygen species. Arterioscler Thromb Vasc Biol. 2000;20: Touzy RM, Cruzado M, Tabet F, Yo G, Salomon S, Schiffrin EL. Redox-dependent AMP kinase signaling by Ang II in vascular smooth muscle cells: role of receptor tyrosine kinase transactivation. Can J Physiol Pharmacol. 2003;81: Zhang GX, Kimura S, Nishiyama A, Shokoji T, Rahman M, Abe Y. ROS during the acute phase of Ang II hypertension participates in cardiovascular MAPK activation but not vasoconstriction. Hypertension. 2004;43: Wenzel S, Taimor G, Piper HM, Schluter KD. Redox-sensitive intermediates mediate angiotensin II-induced p38 MAP kinase activation, AP-1 binding activity, and TGF- expression in adult ventricular cardiomyocytes. FASEB J. 2001;15: Parsons SJ, Parsons JT. Src family kinases, key regulators of signal transduction. Oncogene. 2004;23: Touyz RM, Chen X, Tabet F, Yao G, He G, Quinn MT, Pagano PJ, Schiffrin EL. Expression of a functionally active gp91phox-containing neutrophil-type NAD(P)H oxidase in smooth muscle cells from human

9 780 Circulation Research October 14, 2005 resistance arteries: regulation by angiotensin II. Circ Res. 2002;90: DeLeo FR, Ulman KV, Dvais AR, Jutila KL, Quinn MT. Assembly of the human neutrophil NADPH oxidase involves binding of p67 phox and flavocytochrome b to a common functional domain in p47 phox. J Biol Chem. 1996;271: Li JM, Wheatcroft S, Fan LM, Kearney MT, Shah AM. Opposing roles of p47 phox in basal versus angiotensin II-stimulated alterations in vascular O 2- production, vascular tone, and mitogen-activated protein kinase activation. Circulation. 2004;109: Touzy RM, Schiffrin EL. Increased generation of superoxide by angiotensin II in smooth muscle cells from resistance arteries of hypertensive patients: role of phospholipase D-dependent, NADPH oxidase-sensitive pathways. J Hypertens. 2001;19: Higashi M, Shimokawa H, Hattori T, Hiroki J, Mukai Y, Morikawa K, Ichiki T, Takahashi S, Takeshita A. Long-term inhibition of Rho-kinase suppresses angiotensin II-induced cardiovascular hypertrophy in rats in vivo: effect on endothelial NAD(P)H oxidase system. Circ Res. 2003;93: Seyedabadi M, Goodchild AK, Pilowsky PM. Differential role of kinases in brain stem of hypertensive and normotensive rats. Hypertension. 2001; 38: Sun C, Sellers KW, Sumners C, Raizada MK. NAD(P)H oxidase inhibition attenuates neuronal chronotropic actions of angiotensin II. Circ Res. 2005;96: Zimmerman MC, Sharma RV, Davisson RL. Superoxide mediates angiotensin II-induced influx of extracellular calcium in neural cells. Hypertension. 2005;45: Katz B. The Release of Neural Transmitter Substances. Liverpool, UK: Liverpool University Press; Huang CC, Chan SHH, Hsu KS. cgmp/protein kinase G-dependent potentiation of glutamatergic transmission induced by nitric oxide in immature rat rostral ventrolateral medulla neurons in vitro. Mol Pharmacol. 2003;64: Chuang YC, Chan JYH, Chang AYW, Sirorska M, Borowy-Borowski H, Liou CW, Chan SHH. Neuroprotective effects of coenzyme Q10 at rostral ventrolateral medulla against fatality during experimental endotoxemia in the rat. Shock. 2003;19: Kishi T, Hirooka Y, Kimura Y, Ito K, Shimokawa H, Takeshita A. Increased reactive oxygen species in rostral ventrolateral medulla contribute to neural mechanisms of hypertension in stroke-prone spontaneously hypertensive rats. Circulation. 2004;109: Tai MH, Wang LL, Wu KLH, Chan JYH. Increased superoxide anion in rostral ventrolateral medulla contributes to hypertension in spontaneously hypertensive rats via interactions with nitric oxide. Free Radic Biol Med. 2005;38: Mayorov DN, Head GA, De Matteo R. Tempol attenuates excitatory action of angiotensin II in the rostral ventrolateral medulla during emotional stress. Hypertension. 2004;44: