Baroreflex Modulation by Angiotensins at the Rat Rostral and Caudal Ventrolateral Medulla

Transcription

1 Final Accepted Version Baroreflex Modulation by Angiotensins at the Rat Rostral and Caudal Ventrolateral Medulla Andréia C. Alzamora*, Robson A.S. Santos and Maria J. Campagnole-Santos Departamento de Fisiologia e Biofísica, Instituto de Ciências Biológicas Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil, Running Head: Baroreflex modulation by angiotensins at the VLM *AC Alzamora is presently Associate Professor at the Department of "Ciências Biológicas, Instituto de Ciências Exatas e Biológicas" of the Federal University of Ouro Preto, MG, Brazil Address for correspondence: Maria J. Campagnole-Santos Departamento de Fisiologia e Biofísica Universidade Federal de Minas Gerais Av. Antonio Carlos, 6627 ICB, UFMG , Belo Horizonte, MG, Brasil Phone: (55-31) / FAX: (55-31) mjcs@icb.ufmg.br Copyright 2005 by the American Physiological Society.

2 2 ABSTRACT We determined the effect produced by microinjection of Ang-(1-7) and Ang II into two key regions of the medulla for the control of the circulation, the rostral and caudal ventrolateral medulla (RVLM and CVLM, respectively), on the baroreflex control of heart rate in anesthetized rats. Reflex bradycardia and tachycardia were induced by increases and decreases in mean arterial pressure (MAP) produced by intravenous phenylephrine and sodium nitroprusside, respectively. The pressor effects of Ang-(1-7) or Ang II (25 pmol) after RVLM microinjection (11 ± 0.8 and 10 ± 2 mmhg, respectively) were not accompanied by consistent changes in HR. In addition, RVLM microinjection of these angiotensin peptides did not alter the bradycardic or the tachycardic component of the baroreflex. CVLM microinjections of Ang-(1-7) or Ang II produced hypotension (-11 ± 1.5 and -11 ± 1.9 mmhg, respectively) that was also not accompanied by significant changes in HR. However, CVLM microinjections of angiotensins induced differential changes in the baroreflex control of HR. Ang-(1-7) attenuated the baroreflex bradycardia (0.26 ± 0.06 vs 0.42 ± 0.08 ms/ mmhg, before) and facilitated the baroreflex tachycardia (-0.86 ± 0.19 vs ± 0.10 ms/ mmhg, before), Ang II produced the opposite effect, attenuating the tachycardia (-0.09 ± 0.06 vs ± 0.07 ms/ mmhg, before) and facilitating the baroreflex bradycardia (0.67 ± 0.16 vs 0.41 ± 0.05 ms/ mmhg, before). The modulatory effect of Ang II and Ang-(1-7) on baroreflex sensitivity was completely abolished by peripheral administration of methyl-atropine. These results suggest that Ang II and Ang-(1-7) at the CVLM produce a differential modulation of the baroreflex control of heart rate, probably through distinct effects on the parasympathetic drive to the heart. Key words: Ventrolateral medulla; Baroreflex control of Heart Rate; Renin-Angiotensin System; Arterial Pressure

3 3 INTRODUCTION The baroreceptor afferents terminate primarily in the intermediate portion of the nucleus of the solitary tract (NTS) subjacent to the area postrema in the dorsal medulla (15,32). Although many brainstem and forebrain regions participate in the modulation of the sympathetic and parassympathetic outflows, regions inside ventrolateral medulla are essential for the effectiveness of the baroreceptor reflex (32). The caudal ventrolateral medulla (CVLM) has been functionally defined as a tonicaly active sympathoinhibitory vasodepressor region (15,32,44) that contains a synaptic relay within the baroreflex circuit, connecting the NTS to a site of the sympathoexcitatory reticulospinal neurons, the rostral ventrolateral medulla (RVLM). The RVLM is known to play an essential role in the tonic and reflex control of the sympathetic vasomotor tone (15,32). It has been functionally defined two groups of cardiovascular neurons in the CVLM: neurons that relay baroreflex inputs from the NTS to the RVLM and neurons that are insensitive to the baroreflex input (14,44). In addition, the CVLM depressor response, in part, is associated with vagally mediated decrease in heart rate suggesting a reciprocal connection between the main area containing parasympathetic pre-ganglionic neurons, the nucleus ambiguus, and the CVLM (47). Among the neurotransmitters and neuromodulators that participate in the control of blood pressure at brainstem areas, growing attention has been given to the peptides of the renin-angiotensin system (RAS) (4,21,31,45). High concentrations of angiotensin AT 1 receptor and fibers containing Ang II-immunoreactivity have been described in the dorsomedial and ventrolateral areas of the medulla (2,4,21,22). In addition, important interactions between angiotensinergic peptides and the neuronal elements of the ventrolateral medulla have been shown (47). Although the actions of Ang II are the best characterized, a role for smaller Ang peptides, such as Ang-(1-7), is emerging (4,36,45). We and others have shown that both Ang II and Ang-(1-7) play mainly an excitatory action at RVLM and CVLM, acting through distinct receptors subtypes (3,4,18,19,27,33,38,41). Indeed, we have recently characterized the G protein-coupled receptor MAS as an Ang-(1-7) receptor involved in the biologic actions of this

4 4 heptapeptide (37). Regarding the baroreflex control, the only data available in the literature refer to the effect of Ang II/ or its antagonists on renal sympathetic nerve activity: it has been shown that Ang II induces facilitation of the baroreflex at the RVLM (33-35) and inhibition of the baroreflex at the CVLM (35,40). It is now becoming clear that Ang-(1-7) can act as a contraregulatory peptide of the RAS since Ang-(1-7) presents several effects that are opposite to those of Ang II (16,36). Centrally, lateral ventricle infusion (8,10,29) or microinjection into the NTS (12) of Ang-(1-7) produces facilitation of the baroreflex control of heart rate, in contrast to the well known inhibitory effect induced by Ang II (4,32). In addition, while these peptides may induce similar pressor/ depressor effects at the RVLM/ CVLM, Ang-(1-7), acting probably through its distinct receptor (19,36,37), triggers differential peripheral mechanisms (3,28). In the present study, we attempted to contribute for the understanding of the role of the RAS peptides for the control of blood pressure, by evaluating the modulatory effect on the baroreflex control of heart rate produced by microinjection of Ang-(1-7) or Ang II into the two key areas for the control of the circulation at the ventrolateral medulla, the RVLM and the CVLM. METHODS Surgical Procedures Experiments were performed in male Wistar rats ( g) anesthetized with urethane (1.2 g/ Kg, i.p.). Rats underwent a tracheostomy and a polyethylene catheter was inserted into the abdominal aorta, through the femoral artery for arterial pressure measurement. Another catheter was inserted into the inferior cava vein, through the femoral vein for injection of drugs. Next, the animals were placed in a stereotaxic frame (David Kopf instruments, CA) with the tooth bar 11 mm below the level of the interaural line. The dorsal surface of the brainstem was exposed by a limited occipital craniotomy and an incision of the atlanto-occipital membrane and meninges was performed, as previously described (3,41). The animals were kept on a heating

5 5 pad and the rectal temperature was taken periodically in order to maintain a constant body temperature (aprox. 37 o C). Arterial Pressure Measurements Pulsatile arterial pressure was continuously monitored by a solid state strain-gauge transducer (model TP-200T, Nihon Khoden, Japan) connected to the arterial catheter. Heart rate (HR) was determined with a cardiotachometer (model AT 601G, Nihon Khoden, Japan) triggered by the arterial pressure wave. All variables, pulsatile and mean arterial pressure and HR, were recorded continuously on a direct-writing Nihon Khoden poligraph (Model CP-640G, Nihon Khoden, Japan). Microinjections Procedures Unilateral microinjections of Ang-(1-7), Ang II or sterile saline (vehicle - NaCl 0,9%) in a volume of 100 nl were made over a s period into the RVLM (2.1 mm anterior, 1.8 mm lateral to the obex, and just above pia mater in the ventral surface) or into the CVLM (0.7 mm anterior, 1.8 mm lateral to the obex, and just above pia mater in the ventral surface), as previously described (3,41). Microinjections were made with a triple barreled glass micropipette (outside diameter= Mm), fixed to the stereotaxic manipulator that was inserted in the brain tissue through the dorsal surface. Experiments were made only at sites where the positioning of the micropipette produced a transitory pressor (RVLM) or depressor (CVLM) response (usually mmhg). For all experiments, only one site of the medulla (RVLM or CVLM) was tested per animal. Evaluation of Baroreflex Sensitivity Baroreflex control of HR was determined by recording reflex heart rate changes in response to transient increases or decreases in mean arterial pressure (MAP) produced by repeated bolus injections of graded doses of phenylephrine (0.25 to 5 µg, iv) (baroreflex bradycardia) or sodium nitroprusside (0.5 to 10 µg, iv) (baroreflex tachycardia) (12,13).

6 6 Phenylephrine or sodium nitroprusside doses were injected 1 to 2 minutes apart into a femoral vein in 0.1 ml of isotonic sodium chloride. Blood pressure and heart rate were allowed to return to basal levels before the next dose was given. The baroreflex test was performed in different group of animals before microinjection of the peptides at the RVLM (n=5-7) or CVLM (n=5-8). The dose of phenylephrine or sodium nitroprusside that gave an intermediate MAP change (34-45 mmhg) based on the doseresponse curve was repeated before and at the peak of the response of the peptide. Peak changes in HR occurring during the initial 5-10 seconds of the corresponding maximum change in MAP produced with phenylephrine or sodium nitroprusside were recorded. The HR was converted to Pulse Interval (PI, ms) by the formula: 60,000/ HR. The efficiency of the baroreceptor reflex was estimated by the ratio between changes in HR (as changes in pulse interval) and changes in MAP (PI/ MAP, ms/ mm Hg) in each rat before and after microinjections of peptides at the RVLM or CVLM and it was called baroreflex sensitivity index. In these experiments, only one microinjection site of the medulla (RVLM or CVLM) was tested in each animal. Two or three microinjections (Ang II, Ang-(1-7) and/ or saline in random order) were performed in each animal with an interval of at least 30 min between each other. To test the peripheral mechanisms involved in baroreflex changes, the muscarinic receptor antagonist, methyl-atropine (2.5 mg/ Kg), was initially injected intravenously. After a period of 15 min., the baroreflex control of HR was evaluated before and at the peak of the response induced by unilateral microinjection of Ang-(1-7) (25 pmol, n=5-6) or Ang II (25 pmol, n=6 each) into the CVLM. In these experiments, only one peptide or saline was tested in each animal. The dose of methyl-atropine was chosen in preliminary experiments that showed that 2.5 mg/ kg, i.v., blocked the cardiovascular effects produced by acetylcholine (30 ng) for at least 60 minutes after its administration. In addition, the cardiovascular effects of acetylcholine were tested before and at the end of all experiments in order to verify the effectiveness of the muscarinic blockade.

7 7 Drugs Ang-(1-7) and Ang II were purchased from Bachem (Torance, CA, USA) or Peninsula Laboratories (Belmont, CA, USA). Methyl-atropine nitrate, phenylephrine, sodium nitroprusside were from Sigma Chemical Company (St. Louis, MO, USA). Ang-(1-7) and Ang II were dissolved in sterile isotonic saline (NaCl, 0.9%) at a concentration of 2 mg/ ml, and 10 Ml aliquots were stored at -20 o C. Phenylephrine and sodium nitroprusside were dissolved in sterile saline at 1 mg/ ml concentration and 100 Ml aliquots were stored at -20 o C. At the moment of the experiment, the aliquots were diluted in the desired concentrations and used only once. Methyl-atropine nitrate was dissolved in sterile saline at the moment of the experiment and used only once. Histological Verification of Injection Sites At the end of each experiment, 100 nl of Alcian blue dye (5%) was microinjected into the RVLM or CVLM. The animals were then killed with excess of anesthetic and the brain was carefully removed and fixed in 10% phosphate-buffered formalin. Serial coronal sections (40-50 µm) of the medulla oblongata were made and stained with neutral red for later histological examination. Microinjections sites were identified by the deposition of Alcian blue dye with light microscopy and referred to standard anatomical structures of the brain stem according to the atlas of Paxinos and Watson (30). Statistical Analysis The results are expressed as means ± SEM. Comparisons between before and after injections in the same animal were evaluated by Student s t-test for paired observations. Comparisons among different groups were assessed by one-way ANOVA followed by the Newman-Keuls test. These analyses were performed with the software Graphpad Prism (version 4.00). The criterion for statistical significance was set at p< 0.05.

8 8 RESULTS Effect of RVLM microinjection of Ang-(1-7) and Ang II on baroreflex sensitivity As expected, unilateral microinjection of Ang-(1-7) and Ang II at the RVLM produced increases in blood pressure that were not accompanied by significant changes in HR (Table 1). In addition, microinjection of Ang-(1-7) or Ang II into the RVLM did not alter the bradycardic or tachycardic component of the baroreflex (Table 1). The sensitivity of reflex bradycardia after the microinjection of Ang-(1-7) (0.57 ± 0.10 ms/ mmhg, n=6) or Ang II (0.49 ± 0.09 ms/ mmhg, n=7) was not statistically different from that observed before the microinjections (0.51 ± 0.12 ms/ mmhg, and 0.44 ± 0.11 ms/ mmhg, respectively, Table 1). Similarly, the sensitivity of reflex tachycardia after the microinjection of Ang-(1-7) (-0.63 ± 0.06 ms/ mmhg, n=6) or Ang II (-0.42 ± 0.13 ms/ mmhg, n=6) was similar to that produced before the microinjections (-0.59 ± 0.08 ms/ mmhg and ± 0.05 ms/ mmhg, respectively; Table 1). The microinjection of saline into the RVLM did not significantly alter the bradycardic or tachycardic component of the baroreflex control (Table 1). The lack of a modulatory effect of angiotensins peptides on baroreflex control could be related to fact that the microinjections into the RVLM were unilateral. For these reason, in additional group of animals we tested the effect of bilateral microinjections of Ang-(1-7) (25 pmol, n=9) into the RVLM on the baroreflex control of HR. Bilateral microinjection of Ang-(1-7) into the RVLM produced a similar pressor effect (12 ± 1.7 mmhg; baseline MAP= 86 ± 6 mmhg; n=9, data not shown) to that observed with unilateral microinjection (16 ± 1 mmhg; baseline MAP= 94 ± 5 mmhg n=6; Table 1). No significant effect on HR was also observed (0.3 ± 3.4 beats/ min; baseline HR= 298 ± 15 beats/ min; n=9, data not shown) and as observed with unilateral injection, bilateral microinjection of Ang-(1-7) did not significantly alter the baroreflex bradycardia (0.51 ± 0.09 ms/ mmhg vs 0.53 ± 0.09 ms/ mmhg, before microinjection, n=9, data not shown)

9 Effect of CVLM microinjection of Ang-(1-7) and Ang II on baroreflex sensitivity 9 Unilateral microinjection of Ang-(1-7) into the CVLM produced significant decrease in MAP (Baseline MAP= 110 ± 6 mmhg; RPAM= -11 ± 1.5; n=10), which was similar to that produced by Ang II (Baseline MAP= 99 ± 5 mmhg; RPAM= -11 ± 1.9; n=13). The changes in blood pressure were statistically different from that produced by saline (Baseline MAP= 97 ± 6 mmhg; -2.5 ± 0.6 mmhg, n=9). The hypotensive effect of Ang peptides was not accompanied by significant changes in HR (-5 ± 3 beats/ min and 3 ± 2 beats/ min for Ang-(1-7) and Ang II, respectively). Microinjection of Ang-(1-7) or Ang II into the CVLM induced a differential effect on the bradycardic and tachycardic component of the baroreflex. Figure 1 presents illustrative tracings showing the effect induced by different doses of phenylephrine or sodium nitroprusside before and after CVLM microinjection of Ang II. Microinjection of Ang II into the CVLM (Change in MAP= -11 ± 2.5 mmhg; baseline MAP= 93 ± 5 mmhg and HR= 297 ± 16 beats/ min; n=8) produced a significant increase in the sensitivity of baroreflex bradycardia (0.67 ± 0.16 ms/ mmhg vs 0.41 ± 0.05 ms/ mmhg, before, n=8; Figure 2A). In contrast, after microinjection of Ang-(1-7) into the CVLM (Change in MAP= -12 ± 3 mmhg; baseline MAP= 104 ± 5 mmhg and HR= 286 ± 11 beats/ min; n=5) there was a significant decrease in the baroreflex bradycardia (0.26 ± 0.06 ms/ mmhg vs 0.42 ± 0.08 ms/ mmhg, before, n=5, Figure 2A). On the other hand, the CVLM microinjection of Ang-(1-7) (Change in MAP= -10 ± 1.8 mmhg; baseline MAP= 102 ± 9 mmhg and HR= 335 ± 23 beats/ min; n=5) produced a significant increase in the sensitivity of reflex tachycardia (-0.86 ± 0.19 ms/ mmhg vs ± 0.10 ms/ mmhg, before; n=5; Figure 2B), while microinjection of Ang II (Change in MAP= -11 ± 3 mmhg; baseline MAP= 108 ± 7 mmhg and HR= 304 ± 19 beats/ min; n=5) produced a significant decrease in the sensitivity of reflex tachycardia (-0.09 ± 0.06 ms/ mmhg vs ± 0.07 ms/ mmhg, before; n=5; Figure 2B). Microinjection of saline into the CVLM did not significantly alter the bradycardic (0.53 ± 0.17 ms/ mmhg vs 0.53 ± 0.13 ms/ mmhg, before; n=5) or tachycardic (-0.35 ± 0.09 ms/ mmhg vs ± 0.15 ms/ mmhg, before; n=4) component of the baroreflex (Figure 2).

10 10 Peripheral Mechanism Involved in the Baroreflex Modulatory effects Induced by CVLM Microinjections of Ang-(1-7) or Ang II We next evaluated the contribution of the parasympathetic tonus for the modulatory effect of the Ang peptides at CVLM on the baroreflex control of HR. Pre-treatment with the muscarinic receptor antagonist, methyl-atropine, produced the expected increase in baseline HR (Average of all animals= 378 ± 10 beats/ min vs 315 ± 11 beats/ min, before treatment; n=19) without significant change in the baseline MAP (90 ± 4 mmhg vs 95 ± 5 mmhg; n=19). As previously described (3), the hypotensive effect of Ang-(1-7) was significantly attenuated after peripheral treatment with methyl-atropine (-3 ± 1 mmhg vs -12 ± 4 mmhg, before treatment; n=10; Table 2). In contrast, the hypotensive effect of Ang II at the CVLM after methyl-atropine was not statistically different from that observed before the antagonist (-9 ± 2 mmhg compared with -13 ± 3 mmhg, before treatment; n=9; Table 2). No significant changes in HR were induced by microinjection of angiotensins into the CVLM, before or after methylatropine treatment (Table 2). In addition, methyl-atropine abolished the hypotensive effect produced by intravenous injection of acetylcholine tested at the end of the experiments (-2 ± 1 mmhg compared with -22 ± 2 mmhg, before treatment; n=19; data not shown), confirming the effectiveness of the muscarinic blockade. Methyl-atropine produced a significant attenuation of the baseline baroreflex sensitivity for both the reflex bradycardia (0.14 ± 0.02 ms/ mmhg vs 0.33 ± 0.10 ms/ mmhg, before, n=10, data not shown) and baroreflex tachycardia (-0.18 ± 0.02 ms/ mmhg vs ± 0.04 ms/ mmhg, before, n=9, data not shown). Further, methyl-atropine prevented the modulatory effect of Ang-(1-7) and Ang II on baroreflex control of HR. Ang-(1-7) microinjection after methylatropine did not change the baroreflex bradycardia (0.12 ± 0.03 ms/ mmhg vs 0.13 ± 0.04 ms/ mmhg, before, n=4; Figure 3A) or the reflex tachycardia (-0.23 ± 0.01 ms/ mmhg vs ± 0.02 ms/ mmhg, before, n=5; Figure 3B). Likewise, Ang II microinjection after muscarinic blockade did not change the baroreflex bradycardia (0.09 ± 0.02 ms/ mmhg vs 0.10 ± 0.02 ms/

11 11 mmhg, before, n=5; Figure 3A) or the reflex tachycardia (-0.18 ± 0.05 ms/ mmhg vs ± 0.06 ms/ mmhg, before, n=4; Figure 3B). Histological examination Figure 4 presents diagrams of frontal sections of the medulla according to the atlas of Paxinos and Watson (30), showing the localization of the microinjectons in the RVLM and CVLM. As it can be seen by the representation of the dye spread (shaded area) in all animals, the microinjections into the RVLM were placed in the ventral portion of the rostroventrolateral reticular and lateral paragigantocellular nuclei (Figure 4A). The micoinjections into the CVLM were located in the ventral portion of the lateral reticular nucleus (Figure 4B). DISCUSSION The major finding of the present study was the observation that while microinjections of Ang II and Ang-(1-7) into the RVLM did not affect the baroreflex control of heart rate, CVLM microinjections of Ang peptides induced differential changes on the bradycardic or tachycardic component of the baroreflex. While Ang-(1-7) attenuated the bradycardia and facilitaded the baroreflex tachycardia, Ang II produced opposite effects, attenuating the tachycardia and facilitating the baroreflex bradycardia. In addition, the modulatory effect of both Ang II and Ang- (1-7) on baroreflex was completely abolished by intravenous methyl-atropine. These results extended previous observations and suggest that Ang II and Ang-(1-7) produce a differential modulation on the baroreflex control of heart rate, probably through a distinct effect on the parasympathetic drive to the heart. It is well accepted that the baroreceptor reflex medullary pathway include GABAergic CVLM neurons that receive excitatory inputs from the NTS and in turn project to presympathetic neurons in the RVLM (15,32,44). Most of the CVLM barosensitive neurons are GABAergic (15,32,44), however some neurons are catecholaminergic or cholinergic (39,43,46). Although these CVLM neurons are often depicted as simple relay to the RVLM, GABAergic cells and

12 12 possible the others, are likely to innervate multiple sites to provide a more widespread baroreceptor-mediated inhibition of other regions of the CNS. Using the microinjection of anterograde tracings into the CVLM, Stocker et al. (42) have shown a dense concentration of labeled axons throughout the lateral medullary reticular formation including the retrofacial nucleus, the nucleus ambiguus (NA), RVLM, hypoglossal nucleus, intercalated nucleus and the facial nucleus. These data suggest that a functional interaction between medullary sites can exist in the control of the parasympathetic and sympathetic nervous system. It is our hypothesis that angiotensin peptides modulate the activity of neurons in the CVLM that may also be involved in controlling baroreceptor modulation of the parassympathetic outflow. Figure 5 presents a simplified schematic model to illustrate our hypothesis. As the baroreflex circuit is activated by the pressor response produced by phenylephrine, an increase in the activity of the parasympathetic pre-ganglionic neurons in the NA is expected based on the well recognized NTS - NA pathway. Ang II acting on CVLM neurons would produce an increase in the activity of cardiac vagal neurons in the NA, through a direct or indirect pathway, yet to be identified. On the other hand, Ang-(1-7) would produce an opposite effect, decreasing the activity of these vagal efferents. One of the possible overall changes in the baroreflex control of HR curve induced by Ang peptides at the CVLM is also illustrated in Figure 5. Several studies have shown depressor responses to microinjection of excitatory amino acids or peptides in the medial and ventral portions of the lateral reticular nucleus (8,26,47). The depressor effect evoked by the microinjections of neuroactive drugs in these sites are consistent but of small magnitude in comparison to those elicited from peri-ambigual area, which are also accompanied by large alterations in HR (26,47). Future studies will be necessary to verify the existence of such pathway, connecting the ventral part of the CVLM to the parasympathetic neurons on the NA or yet other medullary sites. The fact that Ang II and Ang-(1-7) presented distinct modulatory effects on the baroreflex is not surprising. We have shown in previous studies opposite effects for these peptides on baroreflex modulation: ICV infusion (10) or NTS microinjection (12) of Ang-(1-7) produces significant facilitation of the baroreflex bradycardia while Ang II at these same sites

13 13 induces attenuation (9,10,11,25). It is also not unusual that Ang II exerts an excitatory action upon microinjection into a specific area and at the same site attenuation of the baroreflex. For example, at the NTS Ang II induces hypotensive effects, which mimics stimulation of the baroreflex, however the upon its microinjection an attenuation of the baroreflex bradycardia is observed (9,11). At the CVLM the hypotensive effect of Ang peptides is not accompanied by consistent changes in HR or cardiac output (3,38), thus suggesting that both peptides induced stimulatory effects on GABAergic CVLM neurons probably projecting to the RVLM. In fact, the hypotensive effect of Ang II at CVLM is associated with a decrease in renal sympathetic activity (48) and it is blocked by application of muscimol, a GABA agonist, into the RVLM (27). Even though the hypotensive effect of Ang-(1-7) at the CVLM is similar, previous results indicate that it involves a sympatho-nitrergic peripheral mechanism (3). Taken together, these data suggest that differential effects can result from the interaction of angiotensin peptides with barosensitive neurons, which present phasic activity, or with non-barosensitive neurons, which are tonically involved in the control of blood pressure, at least in the CVLM and in the NTS. In keeping with this hypothesis, Kasparov and Paton (24) have shown, using whole-cell patch-clamp technique, that in a subpopulation of NTS neurons that tended to exhibit on going activity, Ang II potentiated TS-evoked excitatory postsynaptic potentials. In a different subpopulation of neurons, characterized as silent cells, Ang II enhances inhibitory postsynaptic potentials. This latter effect could potentially account for the Ang II mediate depression of baroreceptor reflex. In addition, these authors have shown that both effects were blocked by losartan and that the potentiation of excitatory, but not the inhibitory, synaptic transmission involves the release of substance P. Thus, other possible mechanisms, yet to be explored, may involve the release of other neurotrasmitters, the activation of other interneurons or the participation of other receptor subtypes in the effect induced by Ang peptides in the baseline and baroreflex control of blood pressure in different medullary sites such as the CVLM and NTS. Only few studies have evaluated the effect of angiotensins in the RVLM or CVLM on the baroreflex control of arterial pressure. In the rat, Sesoko et al (40) have shown that bilateral microinjection of the Ang II antagonist, sarthran, into the CVLM increased the sensitivity for

14 14 reflex activation of the renal sympathetic nerve activity, whereas the sensitivity for reflex suppression of RSNA was not changed, suggesting that Ang II attenuates baroreflex control of RSNA only for fall in arterial pressure. These data are in agreement with our results showing that Ang II at the CVLM attenuates baroreflex-mediated tachycardia. The studies addressing baroreflex control of sympathetic activity were performed with non-selective angiotensin antagonists, sarthran or saralasin, that may interfere with AT 1, AT 2, and Ang-(1-7) receptor MAS. In these studies we can observe that the effect of Ang II was more profound than that induced by the non-selective angiotensin antagonists. Thus, considering the opposite effects evoked by these peptides, at least at the CVLM, we can conclude that the resulting effect on baroreflex induced by non-specific antagonists will be a balance between the endogenous level of Ang II and Ang-(1-7). Other possibilities to explain our data, although more unlikely, could be related to the spread of the injectate to the NA, or yet, it could be due to the stimulation of the release of AVP, through the stimulation of CVLM neurons that project to the hypothalamus (6). In our study because the site of the microinjection is closer to the ventral surface (Figure 4) it is difficult to believe that the peptide could have direct assessed cholinergic vagal motor neurons in the NA. We cannot completely ruled out the possibility that methyl-atropine have gain access to CVLM neurons upon venous injection. However, this is also a very unlike possibility considering that the quaternary structure of methyl-atropine prevents it from crossing the blood-brain barrier. Previous studies have concluded that Ang II would not present a tonic role in the control of sympathetic tonus at the RVLM because microinjection of Losartan at this site did not induce alteration (4,36), or produced a significant increase in arterial pressure in normotensive rats or rabbits (18). However, subsequent studies in hypertensive animals (1) or in transgenic rats that hold changes in the RAS components (17) have shown that the role of angiotensin peptides at the RVLM may be more complex. Whereas the microinjection of losartan into the RVLM of normotensive rats produced a smaller increase in arterial pressure (18), in animals with over activity of renin-angiotensin system (such as SHR, TRG(mREN2)27, Dahl-salt) this AT 1 antagonist induced a significant reduction in arterial pressure (1,23). Moreover, the

15 15 microinjection of another selective AT 1 receptor antagonist, CV-11974, into the RVLM of transgenic rats with low angiotensin levels in the brain did not produce significant alteration in arterial pressure (5). Taken together, these data suggest that besides its well-known excitatory central action, Ang II may induce an inhibitory effect depending on the endogenous level of the angiotensin peptides, at least at the RVLM. In the present study Ang II and Ang-(1-7) microinjection into the RVLM did not alter the baroreflex control of heart rate for the bradycardic or tachycardic component. Those are not completely unexpected results since this area is primarily involved with the control of peripheral sympathetic activity and the method used in our study to evaluate the baroreflex control is more sensitive for assessing the parasympathetic component of the reflex (13). Interestingly, after RVLM microinjection of Ang-(1-7) there was a tendency for an increase in the tachycardic component of the baroreflex, which may involve changes in the sympathetic outflow (20). Previous studies by Head and colleagues have shown in anesthetized and conscious rabbits that RVLM microinjection of Ang II increases while microinjection of saralasin attenuates the sensitivity of the baroreflex control of renal sympathetic activity. These data suggest an endogenous role for Ang II in the modulation of the sympathetic, but not the parasympathetic, component of baroreceptor reflex at the RVLM (33-35). In conclusion, the data presented in this study show that while microinjections of Ang peptides in the RVLM do not affect the baroreflex control of heart rate, microinjection of Ang II and Ang-(1-7) into the CVLM produce differential effects that lead to changes in the parasympathetic drive to the heart. Further, our data indicate that the non-rvlm CVLM connections, possibly with the nucleus ambiguus, may be an additional pathway for the modulatory influence of angiotensin peptides on the baroreflex control of the heart rate.

16 16 REFERENCES 1. Allen AM. Blockade of Angiotensin AT 1 receptors in the rostral ventro-lateral medulla of spontaneously hypertensive rats reduces blood pressure and sympathetic nerve discharge. J Renin Angiotens-Aldost Syst 2(suppl 1): S120-S124, Allen AM, Moeller I, JenkinsTA, Zhuo J, Aldred GP, Chai SY and Mendelsohn FAO. Angiotensin receptors in the nervous system. Brain Res Bull 47: 17-28, Alzamora AC, Santos RAS and Campagnole-Santos MJ. Hypotensive effect of Ang II and Ang-(1-7) at the caudal ventrolateral medulla involves different mechanisms. Am J Physiol (Regul Integr Comp Physiol) 283: R1187-R1195, Averill DB and Diz D. Angiotensin peptides and baroreflex control of sympathetic outflow: Pathways and mechanisms of the medulla oblongata. Brain Res Bull 51(2): , Baltatu O, Fontes MAP, Campagnole-Santos MJ, Caligiorne SM, Ganten D, Santos RAS and Bader M. Alterations of the renin-angiotensin system at the RVLM of transgenic rats with low brain angiotensinogen. Am J Physiol (Regul Integr Comp Physiol) 280(2): R , Blessing WW and Willoughby. Inhibiting the rabbit caudal ventrolateral medulla prevents baroreceptor-initiated secretion of vasopressin. J Physiol 367: , Bonham AC, Jeske I. Cardiorespiratory effects of DL-homocysteic acid in caudal ventrolateral medulla. Am J Physiol. 1989, 256(3 Pt 2): H688-H Britto RR, Santos RAS, Fagundes-Moura CR, Khosla MC and Campagnole-Santos MJ. Role of angiotensin-(1-7) in the modulation of the baroreflex in renovascular hypertensive rats. Hypertension 30:, , Campagnole-Santos MJ, Diz DI and Ferrario CM. Baroreflex modulation by angiotensin II at the nucleus tractus solitarii. Hypertension 11(Suppl I): I-167-I-171, Campagnole-Santos MJ, Heringer SB, Batista EN, Khosla MC and Santos RAS. Differential baroreceptor reflex modulation by centrally infused angiotensin peptides. Am J Physiol (Regul Integr Comp Physiol) 267: R89-R94, Casto R and Phillips MJ. Angiotensin II attenuates baroreflexes at nucleus tractus solitarius of rats. Am. J. Physiol. (Regul Integr Comp Physiol) 250: R193-R198, Chaves GZ, Caligiorne SM, Santos RAS, Khosla MC and Campagnole-Santos MJ. Modulation of the baroreflex control of heart rate by angiotensin-(1-7) at the nucleus tractus solitarii of normotensive and spontaneously hypertensive rats. J Hypertension 18: , Coleman TG. Arterial baroreflex control of heart rate in the conscious rat. Am J Physiol (Heart and Circ Physiol) 238: H515-H520, Cravo SL, Morrison SF and Reis D. Differentiation of two cardiovascular regions with caudal ventrolateral medulla. Am J Physiol (Regul Integr Comp Physiol) 261: R985-R994, Dampney RAL. Functional organization of central pathways regulating the cardiovascular system. Physiol Rev 74(2): , 1994.

17 Ferrario CM, Chappel MC, Tallant EA, Brosnihan KB and Diz DI. Counterregulatory actions of angiotensin-(1-7). Hypertension 30(2): , Fontes MAP, Baltatu O, Caligiorne SM, Campagnole-Santos MJ, Ganten D, Bader M, and Santos RAS. Angiotensin peptides acting at rostral ventrolateral medulla contribute to hypertension of TGR (mren2)27 rats. Physiol. Genomics 2: , Fontes MAP, Martins Pinge MC, Naves V, Campagnole-Santos MJ, Lopes OU, Khosla MC and Santos RAS. Cardiovascular effects produced by microinjection of angiotensins and angiotensin antagonists into the ventrolateral medulla of freely moving rats. Brain Res 750: , Fontes MAP, Silva LCS, Campagnole-Santos MJ, Khosla MC, Guertzenstein PG and Santos RAS. Evidence that angiotensin-(1-7) plays a role in the central control of blood pressure at the ventrolateral medulla acting through specific receptors. Brain Res 665: , Guo GB and Abboud FM. Angiotensin II attenuates baroreflex control of heart rate and sympathetic activity. Am J Physiol (Heart and Circ Physiol) 246: H80-H89, Head GA, Saigusa T and Mayorov DN. Angiotensin and baroreflex control of the circulation. Braz J Med Biol Res 35: , Hu L, Zhu D, Yu Z and Wang JQ. Expression of angiotensin II type 1 (AT1) receptor in the rostral ventrolateral medulla in rats. J Appl Physiol 92: , Ito S, Hiratsuka M, Komatsu K, Tsukamoto K, Kanmatsuse K and Sved A. Ventrolateral medulla AT 1 receptors support arterial pressure in Dahl-salt sensitive rats. Hypertension 41(part 2): , Kasparov S, Paton JF. Differential effects of angiotensin II in the nucleus tractus solitarii of the rat--plausible neuronal mechanism. J Physiol 521(Pt 1): , Matsumura Y, Hasser EM and Bishop VS. Central effect of angiotensin II on baroreflex regulation in conscious rabbits. Am J Physiol (Regul Integr Comp Physiol) 256: R694- R700, Marchenko V, Sapru HN. Cardiovascular responses to chemical stimulation of the lateral tegmental field and adjacent medullary reticular formation in the rat. Brain Res 977(2): , Muratani H, Ferrario CM and Averill DB. Ventrolateral medulla of spontaneously hypertensive rats: role of angiotensin II. Am J Physiol (Regul Integr Comp Physiol) 264: R388-R395, Oliveira DR, Santos RAS, Santos GPF, Khosla MC and Campagnole-Santos MJ. Changes in baroreflex control rate produced by central infusion of selective angiotensin antagonists in hypertensive rats. Hypertension 27(6): , Oliveira RC, Campagnole-Santos MJ and Santos RAS. The pressor effect of angiotensin-(1-7) at the rostral ventrolateral medulla is multimediated. J Hypertens 16(suppl 2): S129 (P16.28-abstract), Paxinos G and Watson C. The rat brain in stereotaxic coordinates. New York: Academic Press, San Diego, 1986.

18 Phillips MI and Sumners C. Angiotensin II in central nervous system physiology. Regul Peptides 78: 1-11, Pilowsky PM and Goodchild AK. Baroreceptor reflex pathways and neurotransmitters: 10 years on. J Hypertens 20: , Saigusa T and Head GA. Renal sympathetic baroreflex effects of angiotensin II infusions into the rostral ventrolateral medulla of the rabbit. Clin Exp Pharmacol Physiol 20(5): 351-4, Saigusa T, Granger NS, Godwin SJ and Head GA. The rostral ventrolateral medulla mediates sympathetic baroreflex responses to intraventricular angiotensin II in rabbits. Auton Neurosc: Basic and Clinical 107: 20-31, Saigusa T, Iriki M and Arita J. Brain angiotensin II tonically modulates sympathetic baroreflex in rabbit ventrolateral medulla. Am J Physiol (Heart and Circ Physiol) 271: H1015-H1021, Santos RAS, Campagnole-Santos MJ and Andrade SP. Angiotensin-(1-7): an update. Regul Peptides 91: 45-62, Santos RAS, Simoes e Silva AC, Maric C, Silva DMR, Machado RP, Buhr I, Heringer- Walther SB, Pinheiro SVB, Lopes MT, Bader M, Mendes EP, Lemos VS, Campagnole- Santos MJ, Schultheiss HP, Speth R and Walther T. Angiotensin-(1-7) is an endogenous ligand for the G protein-coupled receptor Mas. Proc Nat Acad Sci 100(14): , Sasaki S and Dampney RAL. Tonic cardiovascular effects of angiotensin II in the ventrolateral medulla. Hypertension 15: , Schreihofer AM and Guyenet PG. Baro-activated neurons with pulse-modulated activity in the rat caudal ventrolateral medulla express GAD67 mrna. J Neurophysiol 89(3): , Sesoko S, Muratami H, Takeshita S, Teruya H, Kawazoe N and Fukuyama K. Modulation of baroreflex function by angiotensin II endogenous to the caudal ventrolateral medulla. Brain Res 671: 38-44, Silva LC, Fontes MAP, Campagnole-Santos MJ, Khosla MC, Campos Jr RR, Guertzenstein PG and Santos RAS. Cardiovascular effects produced by microinjection of angiotensin-(1-7) on vasopressor and vasodepressor sites of the ventrolateral medulla. Brain Res 613: , Stocker SD, Steinbacher BC, Balaban CD and Yates BJ. Connections of the caudal medullary reticular formation in the cat brainstem. Exp Brain Res 116(2): , Stornetta RL and Guyenet PG. Distribution of glutamic acid decarboxylase mrnacontaining neurons in rat medulla projecting to thoracic spinal cord in relation to monoaminergic brainstem neurons. J Comp Neurol 407(3): , Sved AF, Ito S and Madden CJ. Baroreflex dependent and independet roles of the caudal ventrolateral medulla in cardiovascular regulation. Brain Res Bul 51(2): , Veerasingham SJ and Raizada MK. Brain renin-angiotensin system dysfunction in hypertension: recent advances and perspectives. Brit J Pharmacol 139: , 2003.

19 Verbene AJ, Stornetta RL and Guyenet PG. Properties of C1 and other ventrolateral medullary neurons with hypothalamic projections in the rat. J Physiol 517(2): , Willete RN, Barcas PP, Krieger AJ and Sapru HN. Vasopressor and depressor areas in the rat medulla. Neuropharmacology 22: , Willette RN, Punnen-Grand S, Krieger AJ and Sapru HN. Differential regulation of regional vascular resistance by the rostral and caudal ventrolateral medulla in the rat. J Auton Nerv Syst 18: , 1987.

20 20 ACKNOWLEDGEMENTS This study was supported by grants from the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, Programa de Grupos de Excelência PRONEX and Edital Universal to M.J.C-S.), Coordenadoria de Aperfeiçoamento e Capacitação de Pessoal Técnico Administrativo (CAPES). Andréia Carvalho Alzamora was a recipient of a CAPES-PICDT doctoral fellowship and is presently Associate Professor at the Department of Biological Sciences, at the Exact and Biological Sciences Institute of the Federal University of Ouro Preto, MG, Brazil. The authors wish to thank the skillful technical assistance of José Roberto da Silva.

21 Table 1: Baseline values of mean arterial pressure (MAP, mmhg) and heart rate (HR, beats/ min), MAP and HR changes and the sensitivity index of the baroreflex bradycardia (RPI/ RMAP, ms/ mmhg) and baroreflex tachycardia (RPI/ RMAP, ms/ mmhg) before and after RVLM microinjection of Ang-(1-7) (25 pmol), Ang II (25 pmol) or saline (100 nl). n Baseline Values MAP (mmhg) HR (beats/ min) Cardiovascular Changes RMAP (mmhg) RHR (beats/ min) Baroreflex Bradycardia Index (RIP/ RMAP, ms/ mmhg) Before After n Baseline Values MAP (mmhg) HR (beats/ min) Cardiovascular Changes RMAP (mmhg) RHR (beats/ min) Baroreflex Tachycardia Index (RIP/ RMAP, ms/ mmhg) Saline 5 92 ± ± 12 3 ± ± ± ± ± ± 18 3 ± ± ± ± 0.05 Ang-(1-7) 6 94 ± ± ± 1* 6 ± ± ± ± ± ± 1.8* 1 ± ± ± 0.06 Ang II ± ± ± 3* 0.4 ± ± ± ± ± ± 3.0* -4 ± ± ± 0.13 Before After Values are mean ± SE. *p< 0.05 compared with saline group (ANOVA followed by Newman-Keuls).

22 Table 2. Baseline values of mean arterial pressure (MAP, mmhg) and heart rate (HR, beats/ min) and MAP and HR changes induced by microinjection of Ang-(1-7) (25 pmol), Ang II (25 pmol) or saline (100 nl) into the CVLM before and after methyl-atropine (2.5 mg/ kg, i.v.). Baseline Values Angiotensins at CVLM MAP HR RMAP RHR (mmhg) (beats/ min) (mmhg) (beats/ min) Ang-(1-7) Before 94 ± ± ± 4-4 ± 2 (n=10) After 89 ± ± 11* -3 ± 1* -3 ± 1 Ang II Before 100 ± ± ± 3-6 ± 5 (n=9) After 91 ± ± 16* -9 ± ± 1 Values are mean ± SE. *p< 0.05 compared with before (Student t test for paired observations).

23 23 FIGURE LEGENDS Figure 1. Blood pressure (PAP= pulsatile arterial pressure; MAP=mean arterial pressure, mmhg) and heart rate (HR, beats/ min) tracings obtained in a Nihon-Khoden polygraph illustrating the reflex changes in HR induced by MAP changes produced by bolus intravenous injections of different doses of phenylephrine (Phe, 1-5 Mg; A) or sodium nitroprusside (NP, Mg; B) before and after the microinjection of Ang II (25 pmol) into the CVLM. Arrows indicate the moment of injection and dotted line indicates the microinjection into the CVLM. Figure 2A- Baroreflex bradycardia sensitivity index (Changes in PI/ Changes in MAP, ms/ mmhg) and 2B- Baroreflex tachycardia sensitivity index (Changes in PI/ Changes in MAP, ms/ mmhg) before and after CVLM microinjections of saline (100 nl; n=4,5), Ang-(1-7) (25 pmol; n=5 each) or Ang II (25 pmol; n=5,8). *p< 0.05 in comparison to before (Student t test for paired observations). Figure 3A- Baroreflex bradycardia sensitivity index (Changes in PI/ Changes in MAP, ms/ mmhg) and 3B- Baroreflex tachycardia sensitivity index (Changes in PI/ Changes in MAP, ms/ mmhg) before and after CVLM microinjections of Ang-(1-7) (25 pmol; n=5 each) or Ang II (25 pmol; n=4-5) in animals treated with methyl-atropine (2.5 mg/ Kg, i.v.). None of the indexes of baroreflex sensitivity after CVLM microinjections were statistically different from before. Figure 4. Drawings of frontal sections of the medulla oblongata from the atlas of Paxinos and Watson (1986) showing the location of the microinjections in the RVLM (A) or in the CVLM (B) obtained from the histological examination of the deposition of the Alcian blue dye in frontal sections of the medulla of the rats used in the present study. Numbers on the bottom of each drawing refer to the distance from the bregma. AP= area postrema; Amb= nucleus ambiguus; IO= inferior olive; LR= lateral reticular nucleus; LPGi= lateral paragigantocellular nucleus; py= pyramidal tract; RVL= rostroventrolateral reticular nucleus; Sol= nucleus of solitary tract; XII= hypoglossal nucleus.

24 24 Figure 5. Simplified schematic representation of the medullary pathway of the baroreflex control of heart rate (A). Baroreceptor afferents make their first synapse at the nucleus of the solitary tract (NTS). Neurons at the NTS drive the baroreceptor information mainly to two groups of areas/ nuclei: 1) the nucleus ambiguus (NA) and the dorsal motor nucleus of the vagus (not represented), which contain the pre-ganglionic cells of the parasympathetic system and 2) the caudal ventrolateral medulla (CVLM), which in turn, modulates the pre-motor neurons of the rostral ventrolateral medulla that control the activity of the pre-ganglionic cells of the sympathetic system in the intermediolateral column of the spinal cord. Lines on the graph (B) represent the correlation between reflex changes in HR that are expected for MAP changes induced by vasoactive drugs, in the control and after Ang II or Ang-(1-7) microinjection at the CVLM. The data of the present study showed that Ang II and Ang-(1-7) acting at the CVLM can differentially modulate the parasympathetic output of the baroreflex, whether this effect involves a direct or indirect pathway is yet to be explored. Positive sign (+) indicate excitatory effect and negative sign (-) indicate inhibitory effect.

25 200 A 200 B PAP mmhg PAP mmhg MAP mmhg 100 MAP mmhg 75 HR beats/ min HR beats/ min Phe 1 µg Phe 2.5 µg Phe 5.0 µg Phe 2.5 µg NP 2.5 µg NP 7.5 µg NP 7.5 µg Ang II CVLM Ang II CVLM

26 Figure 2 A Changes in PI/ Changes in MAP (ms/ mmhg) CVLM Microinjections Baroreflex Bradycardia Before After * * Saline Ang-(1-7) Ang II B CVLM Microinjections Baroreflex Tachycardia Changes in PI/ Changes in MAP (ms/ mmhg) Saline Ang-(1-7) Ang II Before After * *

27 27 Figure 3 A CVLM Microinjections Baroreflex Bradycardia after Methyl-Atropine Changes in PI/ Changes in MAP (ms/ mmhg) 0.4 Before After Ang-(1-7) Ang II B CVLM Microinjections Baroreflex Tachycardia after Methyl-Atropine Changes in PI/ Changes in MAP (ms/ mmhg) Ang-(1-7) Before After Ang II

28 28 Figure 4 A mm Sol mm Amb RVL LPGi mm IO py mm B mm Sol AP XII mm Amb LR mm IO py mm

29 29 Figure 5 A Baroreceptors afferents Ang II NTS CVLM? NA Ang-(1-7) + - Heart Rate, beats/ min Parasympathetic drive to the Heart B Ang-(1-7) 200 Control Ang II Mean Arterial Pressure, mmhg