INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IN PREVIOUS STUDIES, WE VERIFIED that the neurotransmission of the parasympathetic component (bradycardia) of the baroreflex in the nucleus tractus solitarii (NTS) seems to be mediated by N-methyl-D-aspartic acid (NMDA) receptors (9, 15, 16), although indirect evidence suggested that the sympathoinhibitory component (hypotension) is mediated by non-NMDA receptors (11). To better evaluate the parasympathoexcitatory and sympathoinhibitory components of the baroreflex, we recently developed a new approach involving electrical stimulation (ES) of the aortic depressor nerve (ADN) in awake rats. This approach allows us to evaluate the parasympathoexcitatory and sympathoinhibitory components of the baroreflex through recordings of heart rate (HR) and arterial pressure, respectively (8). It is important to note that the evaluation of the hypotensive response (the sympathoinhibitory component) to baroreflex activation is not feasible when drugs such as phenylephrine are used to produce the activation of the baroreflex, due to their direct vasoconstrictor effect, which masks the vasodilation response (the sympathoinhibitory component) to baroreflex activation.

In the present study, we combined our previous experience in performing experiments with guide cannulas in the direction of the NTS with the placement of bipolar electrodes in the ADN, both in awake rats, to evaluate the role of the different subtypes of excitatory amino acid (EAA) receptors in the processing of the parasympathoexcitatory and sympathoinhibitory components of the baroreflex at the NTS level. To study the role of NMDA and non-NMDA receptors in the processing of the bradycardic (parasympathoexcitation) and hypotensive (sympathoinhibition) responses to ES of the ADN, we used two experimental protocols in distinct groups of rats, in which we performed: 1) bilateral microinjection of DL-2-amino-5-phosphonopentanoic acid (AP-5, a selective NMDA receptor antagonist) into the lateral commissural NTS, followed by microinjection of 6,7-dinitroquinoxaline-2,3 dione (DNQX, a selective non-NMDA receptor antagonist); and 2) bilateral microinjection of DNQX into the lateral commissural NTS, followed by microinjection of AP-5. These experimental protocols were performed in awake rats, and the doses of AP-5 and DNQX used in the present study were in the range of selectivity for NMDA and non-NMDA receptors in the NTS, respectively, in accordance with previous studies from our laboratory (4, 9, 11, 12, 15).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Male Wistar rats weighing 300-320 g were used in the present study. Four days before the experiments, rats under 2.5% (1 ml/100 g ip) tribromoethanol (Aldrich Chemical) anesthesia were placed in a stereotaxic apparatus (David Kopf, Tujunga, CA), and the technique described by Michelini and Bonagamba (17) was used to implant bilateral guide cannulas in the direction of the NTS, in accordance with the coordinates of Paxinos and Watson (19). Additional tribromoethanol was injected when the rats reacted to frequent toe pinching during the stereotaxic surgery. To implant the cannulas, a small window was opened caudal to the lambda, through which a 15-mm-long, 22-gauge stainless steel cannula was introduced perpendicularly 14.5 mm caudal to the bregma, 0.5 mm lateral to the midline, and 7.8 mm below the skull surface of the bregma. The tip of each guide cannula was placed in the cerebellum 1.0 mm above the dorsal surface of the brain stem. The guide cannula was fixed to the skull with methacrylate and watch screws and closed with an occluder until the day of the experiments. The needle (33 gauge) used for microinjection into the NTS was 1.5 mm longer than the guide cannula and was connected by polyethylene-10 (PE-10) tubing to a 1-µl syringe (Hamilton, Reno, NV).

The needles for microinjection of drugs into the NTS were carefully inserted into the guide cannula, and manual injection was initiated 30 s later. For bilateral microinjection, the microinjection was initially performed on one side, the needle was withdrawn and repositioned on the contralateral side, and then the second injection was made; microinjections were therefore made ~1 min apart. At the end of each experiment, 50 nl of 2% Evans blue dye was microinjected into the same sites for histological analysis. The animals were then submitted to intracardiac perfusion with saline (0.9%) followed by 10% buffered Formalin under ether anesthesia. The brains were removed and stored in buffered Formalin for 2 days, and serial coronal sections (15 µm) were cut and stained using the Nissl method. Only the rats in which the sites of microinjections were located bilaterally in the lateral portion of the commissural NTS (~0.5 mm lateral to the midline) were used for data analysis.

One day before the experiments, while under 40 mg/kg ip pentobarbital sodium anesthesia (Sigma, St. Louis, MO), the rats were submitted to ventral neck surgery with the purpose of isolating the left ADN. After isolation, the ADN was placed on a bipolar stainless steel electrode and the electroneuronographic recording of baroreceptor activity was obtained using a Tektronics oscilloscope after the signal had been properly amplified (21). A loudspeaker allowed us to identify the sound of the action potentials; the combination of the sound of the action potentials with the direct measurement of the pulsatile arterial pressures was used for identification of the ADN. Only rats that presented clear electroneuronographic recordings of the ADN were implanted with chronic electrodes.

After the ADN was identified under the microscope, the bipolar stainless steel electrode supporting a short segment of the ADN was carefully covered with dental covering material (Coltene President/Coltene Whaledent, NJ). It was critical to verify whether the small vessel that irrigates the nerve was patent before covering the electrode and the ADN, because the success of the experiments performed 24 h later depended on the integrity of nerve irrigation. After the electrode and nerve were covered, we allowed at least 30 min to elapse for complete polymerization of the gel, and again the electroneuronographic recording was performed to verify the integrity of the nerve. Once the integrity of the nerve was confirmed, the fine platinum wires of the electrodes were exteriorized through the back of the rat and soldered to a small plug that was later used for connection with the electrical stimulator.

Under the same anesthesia, the femoral artery and vein were catheterized to record the pulsatile arterial pressure and to administer intravenous drugs, respectively. At 24 h after surgery, when the rats had completely recovered from anesthesia, the animals were connected to the recording system (polygraph and pressure transducer, Narco Bio-Systems, Austin, TX) and the electrical stimulator (Department of Physiology, School of Medicine of Ribeirão Preto Bioengineering Facilities), and the experiment was performed in an isolated room. The experiment consisted of an ES period of 5 s (50 Hz, 10 ms, 4 ± 1 V) at intervals of at least 3 min. Pulsatile arterial pressure (PAP), mean arterial pressure (MAP), and HR were recorded.

On the day of the experiments a control recording of the cardiovascular parameters was performed for at least 20 min, and a functional identification of the commissural NTS was performed with microinjections of L-glutamate (1 nmol/50 nl), considering that the microinjection of this excitatory amino acid into the NTS of unanesthetized rats produces pressor and bradycardic responses (11). All microinjections into the comissural NTS were performed in a volume of 50 nl. The solutions were freshly dissolved in 0.9% saline or saline plus 2.5% DMSO (Sigma), and sodium bicarbonate was added to adjust the pH to ~7.0.

Three experimental protocols using distinct groups of rats were used. In protocol 1 (n = 7), the ES of the ADN was performed before (control) and 5 min after bilateral microinjection of AP-5 (10 nmol/50 nl) into the NTS. Immediately after this ES of the ADN, DNQX (0.5 nmol/50 nl) was microinjected into the NTS and 5 min later, i.e., 10 min after the microinjection of AP-5, the ADN was stimulated again; stimulation was then repeated at 20, 30, and 60 min after AP-5 administration. In protocol 2 (n = 6), the ES of the ADN was performed before (control) and 5 min after bilateral microinjection of DNQX (0.5 nmol/50 nl) into the NTS. Immediately after this ES of the ADN, AP-5 (10 nmol/50 nl) was microinjected into the NTS and 5 min later, i.e., 10 min after the microinjection of DNQX, the ADN was stimulated again: stimulation was repeated at 20, 30, and 60 min after DNQX. In protocol 3 (n = 3), the ES of the ADN was performed before (control) and 5 min after bilateral microinjection of the vehicle [saline plus DMSO (2.5%), 50 nl] into the NTS. Immediately after this ES of the ADN, 50 nl of saline was microinjected into the NTS and 5 min later, i.e., 10 min after the microinjection of the vehicle, the ADN was stimulated again, and stimulation was repeated 20 and 30 min after the vehicle.

All values are expressed as means ± SE, and results related to the changes in baseline MAP and HR after bilateral microinjection of AP-5 or DNQX into the NTS were analyzed using the paired Student's t-test. The data related to the changes in the hypotensive and bradycardic responses observed before (control) and after (5, 10, 20, 30, and 60 min) sequential bilateral microinjection of AP-5 and DNQX, or DNQX and AP-5, were analyzed by one-way ANOVA, with the within variable being the changes in MAP and HR in relation to the respective control. The differences were considered significant at the P < 0.05 level.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effect of sequential microinjection of AP-5 and DNQX on bradycardic and hypotensive responses to ES of ADN. Figure 1 shows tracings from one rat that is representative of the group in which the ES of the ADN was performed before (control) and 5 min after administration of AP-5, 5 min after DNQX (10 min after AP-5), and 20, 30, and 60 min after AP-5 was given. The tracings show that the bradycardic response was significantly reduced after AP-5 and almost abolished after DNQX administration, although the hypotensive response was only partially reduced by AP-5 and DNQX. The data for this group are summarized in Fig. 2 and show that AP-5 produced a significant reduction in the bradycardic as well as the hypotensive response to ES of the ADN; the sequential microinjection of DNQX almost abolished the bradycardic response, but produced no significant additional reduction in the hypotensive response. It is important to note that 60 min after microinjection of AP-5 and DNQX, both the bradycardic and hypotensive responses to ES of the ADN were back to control values (Figs. 1 and 2), indicating the reversibility of the blockade of NMDA and non-NMDA receptors in the NTS. Table 1 shows that the bilateral microinjections of AP-5 into the NTS produced no significant changes in baseline MAP or HR 5 min after microinjection.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Changes in heart rate (HR, beats/min), pulsatile arterial pressure (PAP), and mean arterial pressure (MAP) in response to electrical stimulation (ES, solid arrows) of aortic depressor nerve (ADN) before (control) and 5 min (after DL-2-amino-5-phosphonopentanoic acid, AP-5), 10 min (5 min after 6,7-dinitroquinoxaline-2,3 dione, DNQX), 20, 30, and 60 min after sequential microinjection of AP-5 (10 nmol/50 nl) and DNQX (0.5 nmol/50 nl) into nucleus tractus solitarii (NTS, open arrows).

View larger version (45K): [in this window] [in a new window]   Fig. 2.   A: changes in MAP in response to ES of ADN before (control, CON) and 5 min (after AP-5), 10 min (5 min after DNQX), 20, 30, and 60 min after bilateral microinjection of AP-5 (10 nmol/50 nl) and DNQX (0.5 nmol/50 nl) into commissural NTS. B: changes in HR (beats/min) in response to ES of ADN before (CON) and 5 min (after AP-5), 10 min (5 min after DNQX), 20, 30, and 60 min after sequential bilateral microinjection of AP-5 (10 nmol/50 nl) and DNQX (0.5 nmol/50 nl) into commissural NTS (n = 7). * Significantly different compared with control (* P < 0.05).

Effect of sequential microinjection of DNQX and AP-5 on bradycardic and hypotensive responses to ES of ADN. Figure 3 shows the tracings of one rat that is representative of the group in which the ES of the ADN was performed before (control) and 5 min after DNQX; 5 min after AP-5 (10 min after DNQX); and 20, 30, and 60 min after DNQX administration. The tracings show that the bradycardic and hypotensive responses were significantly reduced after DNQX was given, and the bradycardic response was abolished after AP-5 was administered, whereas the hypotensive response was additionally reduced by AP-5 but not abolished. The data for this group are summarized in Fig. 4 (n = 6) and show that DNQX produced a significant reduction in the bradycardic and hypotensive response to ES of the ADN; the sequential microinjection of AP-5 almost abolished the bradycardic response but produced no significant additional reduction in the hypotensive response. It is important to note that 60 min after microinjection of DNQX and AP-5, both the bradycardic and hypotensive responses to ES of the ADN were back to control values (Figs. 3 and 4), indicating the reversibility of the non-NMDA and NMDA receptors in the NTS. Table 1 shows that bilateral microinjections of DNQX into the NTS produced a significant increase in baseline MAP and no changes in baseline HR 5 min after their application.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   Changes in HR (beats/min), PAP, and MAP in response to ES (solid arrows) of ADN before (control) and 5 min (after DNQX), 10 min (5 min after AP-5), 20, 30, and 60 min after sequential microinjection of DNQX (0.5 nmol/50 nl) and AP-5 (10 nmol/50 nl) into NTS (open arrows).

View larger version (44K): [in this window] [in a new window]   Fig. 4.   A: changes in MAP in response to ES of ADN before (CON) and 5 min (after DNQX), 10 min (5 min after AP-5), 20, 30, and 60 min after bilateral microinjection of AP-5 (10 nmol/50 nl) and DNQX (0.5 nmol/50 nl) into commissural NTS. B: changes in HR (beats/min) in response to ES of ADN before (CON) and 5 min (after DNQX), 10 min (5 min after AP-5), 20, 30, and 60 min after sequential bilateral microinjection of DNQX (0.5 nmol/50 nl) and AP-5 (10 nmol/50 nl) into commissural NTS (n = 6). * Significantly different compared with control (P < 0.05).

Effect of sequential microinjection of vehicle (saline + DMSO) and saline on bradycardic and hypotensive responses to ES of ADN. The data for this group are summarized in Fig. 5 (n = 3) and show that vehicle (saline + DMSO) and saline produced no significant changes in the bradycardic or hypotensive responses to ES of the ADN. This indicates that bilateral microinjection of vehicle into the NTS produced no effect on the cardiovascular responses to ES of the ADN.

View larger version (42K): [in this window] [in a new window]   Fig. 5.   A: changes in MAP in response to ES of ADN before (CON) and 5 min (after vehicle, DMSO), 10 min (5 min after saline), 20, 30, and 60 min after sequential bilateral microinjection of vehicle and saline into commissural NTS. B: changes in HR (beats/min) in response to ES of ADN before (CON) and 5 min (after vehicle, DMSO), 10 min (5 min after saline), 20, 30, and 60 min after sequential bilateral microinjection of vehicle and saline into commissural NTS (n = 3).

Histology. Figure 6 is a photomicrograph of a 15-µm transverse section of the brain stem of one rat that is representative of the three groups studied, showing the sites of microinjections in the NTS bilaterally.

View larger version (93K): [in this window] [in a new window]   Fig. 6.   Transverse section of brain stem of one rat, representative of groups showing bilateral sites of microinjections in commissural NTS. Arrows indicate sites of microinjections.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

There are several lines of evidence indicating that the EAA L-glutamate is the neurotransmitter of the baroreflex at the NTS level (10, 13, 22), and several studies have evaluated the relative role of the different subtypes of ionotropic receptors in the processing of the autonomic responses in the NTS (1, 2, 18, 23, 24) as well as in the different components of the action potential in the neurons of the NTS (3). In the present study, we evaluated the role of NMDA and non-NMDA receptors in the processing of the parasympathoexcitatory component (bradycardia) and the sympathoinhibitory component (hypotension) at the NTS level in response to ES of the ADN. The major findings indicated that 1) the parasympathetic component seems to be mediated by both NMDA and non-NMDA receptors, considering that sequential microinjection of AP-5 and DNQX produced a complete blockade of the bradycardic response to ES of the ADN; and 2) the sympathoinhibitory component of the responses was only partially blocked by the sequential microinjection of AP-5 and DNQX, suggesting that other receptor classes such as metabotropic ones or a neurotransmitter other than L-glutamate may play a key role in the neurotransmission of this component at the NTS level.

The dose of AP-5 (10 nmol/50 nl) used in the present study was effective in blocking the hypotensive and bradycardic responses to microinjection of NMDA (2 nmol/50 nl) into the lateral commissural NTS of awake rats, although it produced no effect on the hypotensive or bradycardic responses to microinjection of 3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA, 0.05 nmol/50 nl) into the same subregion of the NTS in a distinct group of rats (15). The dose of AP-5 used (10 nmol/50 nl) was effective in blocking the parasympathetic component of the chemoreflex (12) and also in blocking the bradycardic response induced by microinjection of L-glutamate into the most lateral aspect of the commissural NTS (4). On the other hand, the dose of DNQX (0.5 nmol/50 nl) was in the range of selectivity for non-NMDA receptors, because in contrast to the doses of 1.0 and 2.0 nmol/50 nl, it produced no changes in the hypotensive or bradycardic responses to microinjection of NMDA (2 nmol/50 nl) into the lateral commissural NTS (11). In addition, the study by Haibara et al. (11) documented that although microinjection of the 0.5 nmol/50 nl dose produced no effect on the cardiovascular responses to microinjection of NMDA into the NTS, it induced a large increase in baseline MAP, suggesting that the processing of the tonic sympathoinhibitory component of the baroreflex was mediated by non-NMDA receptors. This previous pharmacological evidence obtained under similar experimental conditions indicates that the doses of AP-5 and DNQX used in the present study were in the range of selectivity in blocking their respective receptor subtypes.

In studies performed on anesthetized rats, Leone and Gordon (13) verified that the hypotensive and bradycardic responses to ES of the ADN were abolished by microinjection of kynurenic acid, a nonselective antagonist of EAA receptors, into the NTS. However, in the same study they observed that the hypotensive and bradycardic responses to microinjection of L-glutamate into the NTS were not blocked by antagonism of NMDA and non-NMDA receptors in the NTS. The authors suggested that an EAA or EAA analog other than L-glutamate may be the neurotransmitter of the baroreceptor afferents in the NTS. Although the data of the present study are different from those reported by Leone and Gordon (13), i.e., these investigators blocked the bradycardic and hypotensive responses to ES of the ADN, whereas we blocked the bradycardia and only partially blocked the hypotensive response, we reached a similar conclusion that the neurotransmission of the baroreflex in the NTS may involve an EAA or EAA analog other than L-glutamate, particularly in the neurotransmission of the sympathoinhibitory component.

Studies by Aylwin et al. (3) have documented that both NMDA and non-NMDA receptors coexist in the same second-order NTS neurons and mediate primary visceral afferent transmission in the NTS. In the studies by Aylwin et al. (3), it was not possible to determine whether the second-order neurons were related to the pathways involved in the parasympathoexcitatory or sympathoinhibitory component of the baroreflex, or both. The data of the present study, which included microinjection of a relatively large volume (50 nl) of antagonist into the NTS, suggest that the neurons involved in the parasympathoexcitatory component contain essentially NMDA and non-NMDA receptors, whereas the sympathoinhibitory component of the baroreflex seems to be dependent on other receptor classes or even on a neurotransmitter other than L-glutamate, because the hypotensive response was reduced by only ~50% after NMDA and non-NMDA receptor blockade in the commissural NTS.

Work by Zhang and Mifflin (23) documented that administering kynurenic acid into the NTS inhibited carotid sinus nerve stimulation-evoked discharge in NTS neurons, indicating that EAA receptors are involved in visceral afferent-evoked activation of NTS neurons. Considering that both mono- and polysynaptic inputs were attenuated, these investigators suggested that EAA receptors are used at multiple levels of afferent integration within the NTS. Our data are in agreement with the findings by Zhang and Mifflin (23), particularly in relation to the parasympathoexcitatory component of the baroreflex, which was completely blocked by NMDA and non-NMDA receptors. In addition, these data corroborate those of another recent study by Zhang and Mifflin (24), which shows that the ADN-evoked discharge in some mono- and polysynaptic neurons in the NTS were not attenuated by ionotropic antagonists, suggesting that another receptor or transmitter system may mediate synaptic excitation in these neurons. From our findings, we may suggest that the neurons not affected by ionotropic antagonists in the study by Zhang and Mifflin (24) are part of the sympathoinhibitory pathways of the baroreflex in the NTS.

In a previous study (7), we also verified that the bradycardic responses to microinjection of L-glutamate into the NTS were blocked in a dose-dependent manner by AP-5, indicating that the excitation of NTS neurons projecting to the areas of origin of the preganglionic parasympathetic neurons is mediated by NMDA receptors. We (15) also verified that the bradycardic response to the activation of the baroreflex by phenylephrine infusion was reduced in a dose-dependent manner, indicating that the processing of the parasympathetic component of the baroreflex is mediated mainly by NMDA receptors. In these cases (5, 7, 9, 12, 15, 16), it is important to note that bilateral microinjections of AP-5 into the NTS produced no changes in baseline MAP, suggesting that the tonic sympathoinhibitory component of the baroreflex does not involve NMDA receptors in the neurons projecting from the NTS to the caudal ventrolateral medulla (CVLM).

In studies using vasoactive drugs, we were not able to evaluate the sympathoinhibitory component of the baroreflex due to the direct vasoconstrictor effect of this drug, which masked the vasodilation produced by sympathoinhibition. However, in studies using microinjection of kynurenic acid (a nonselective EAA receptor antagonist) or DNQX (a selective non-NMDA receptor antagonist) into the NTS, we observed a significant increase in baseline MAP, indicating that non-NMDA receptors may play a critical role in the processing of the sympathoinhibitory component of the baroreflex at the NTS level (6, 11). These previous data from our laboratory allowed us to suggest that the parasympathoexcitatory component of the baroreflex at the NTS level may be mediated mainly by NMDA receptors, and the sympathoinhibitory component may be mediated by non-NMDA receptors (11, 16). Recently, we developed a new technical approach to perform ES of the ADN in awake rats (8) to evaluate the parasympathoexcitatory (bradycardia) and sympathoinhibitory (hypotension) components by recording the changes in HR and MAP; in the present study this approach was used to verify whether our hypothesis was correct.

Several important aspects of the present study should be emphasized. In protocol 1, in which we performed sequential microinjection of AP-5 and 5 min later microinjection of DNQX into the NTS, we observed that bilateral microinjection of AP-5 produced no changes in baseline MAP or HR (Fig. 1 and Table 1). However, the bradycardic and hypotensive responses to ES of the ADN were significantly reduced (Fig. 2). In this case, the reduction of the hypotensive response may be secondary to the major reduction in the bradycardic response, which seems to be mainly mediated by NMDA receptors at the NTS level, as we documented in previous studies (9, 15, 16). It is important to note that AP-5 produced no changes in baseline HR, suggesting that tonic activity of the baroreceptor afferents on the parasympathetic component in the NTS is not mediated by NMDA receptors. These receptors seem to play an important role only when an additional activation of the baroreceptors by ES or by an increase in baseline MAP activates neurons in the NTS containing NMDA receptors. In this case, the sequential microinjection of DNQX produced an additional reduction of the bradycardic response, indicating that during the challenge produced by the ES of the ADN both NMDA and non-NMDA receptors participate in the processing of the parasympathoexcitatory component in the neurons of the NTS projecting to the nucleus ambiguus or dorsal motor nucleus of the vagus.

With respect to the bradycardic response to ES of the ADN, we verified in a previous study (8) that it is not dependent on the sympathetic withdrawal because the bradycardia was completely abolished by atropine methyl nitrate. In addition, we verified in a series of current experiments in our laboratory (unpublished data) that the blockade of the sympathetic drive to the heart with atenolol, a ₁-adrenoceptor antagonist, produced no effect on the gain of the baroreflex bradycardic response. Therefore this earlier evidence supports the concept proposed in the present study that the bradycardic component of the baroreceptor afferents is entirely due to the activation of the parasympathoexcitatory component at the NTS level.

It is also important to note that in this protocol the microinjection of DNQX produced a significant increase in baseline MAP, similar to that observed in our previous study (11). In this case we may suggest that the tonic sympathoinhibitory pathways in the NTS are mediated by non-NMDA receptors. However, when we stimulated the ADN, we were not able to block the sympathoinhibitory response (hypotension). Similar to the parasympathoexcitatory component, we may suggest that the sympathoinhibitory component during spontaneous oscillations of arterial pressure is modulated by non-NMDA receptors in neurons of the NTS, but under circumstances of additional increase in baseline MAP, mimicked in the present study by ES of the ADN, receptors other than ionotropic or other neurotransmitters may take part in this neurotransmission. Therefore the present data open two possibilities for the neurotransmission of the baroreflex at the NTS level: 1) neurotransmission of the tonic afferent of the baroreceptors during spontaneous oscillations of arterial pressure variation; and 2) neurotransmission occurring when the baroreceptor afferents are submitted to a challenge such as the increase in baseline MAP and consequent additional activation of the ADN.

In the case of normal MAP levels, we may suggest that the processing of the parasympathoexcitatory component does not involve NMDA receptors, because the microinjection of AP-5 into the NTS produced no changes in baseline HR. However, during the challenge produced by the activation of the baroreceptor afferents, the neurons of the NTS containing these receptors play a critical role, as we demonstrated in recent studies from our laboratory (4, 9). This possibility implies that different subpopulations of neurons in the NTS containing different receptor subtypes participate in the regulation of efferent parasympathetic activity, i.e., one of them in the tonic regulation of baseline HR, and the other during the challenge produced by baroreceptor activation. On the other hand, the data related to microinjection of DNQX into the NTS indicate that the tonic control of the sympathoinhibitory component of baroreceptor afferents in the normal range of arterial pressure variation involves non-NMDA receptors, because microinjection of DNQX into the NTS produced a significant increase in baseline MAP, probably due to blockade of receptors in the neurons projecting from the NTS to the CVLM, a step in the sympathoinhibitory projection to the rostral ventrolateral medulla. However, the hypotensive response to ES of the ADN after DNQX was not blocked. This suggests that during the challenge in the activation of the baroreceptor afferents, neurons containing other subclasses of EAA receptors such as metabotropic ones or even a neurotransmitter other than L-glutamate may play an important role in this neurotransmission, because the hypotensive response to ES after sequential microinjection of AP-5 and DNQX was reduced by ~50%. In addition, it is important to consider that this partial reduction in the hypotensive response could be secondary to the increase in baseline MAP produced by DNQX. If this possibility is true, it may give additional support to the idea that the processing of the sympathoinhibitory component during challenge of the activity of baroreceptor afferents to the NTS does not involve non-NMDA receptors.

In protocol 2, in which DNQX was microinjected first and AP-5 was administered 5 min later, the data also support the aspects discussed above. The microinjection of DNQX into the NTS produced a significant increase in baseline MAP (Figs. 3 and 4, Table 1) and no changes in baseline HR. It is important to note that the sequential microinjection of AP-5 and DNQX or DNQX and AP-5 produced no changes in baseline HR in either protocol. These findings concerning baseline HR suggest that under normal conditions the pulse-to-pulse tonic modulation of the parasympathoexcitatory and sympathoexcitatory components to the heart by baroreceptor afferents does not involve NTS neurons containing ionotropic receptors, and that the EAA receptors located in neurons of both parasympathoexcitatory and sympathoinhibitory projections from the NTS to the CVLM and from the NTS to the nucleus ambiguus are activated only during challenge to the afferents of the baroreceptors to the NTS, such as ES of the ADN.

In protocol 2 we also observed that the hypotensive response to ES of the ADN was reduced after DNQX administration, probably as a consequence of the significant reduction in the bradycardic response. After AP-5 was given, the bradycardic response was almost abolished, but the hypotensive response was maintained at ~50% of the control response, similar to the findings observed in protocol 1. Therefore, the results were similar for the hypotensive and bradycardic responses to ES of the ADN in both protocols using different groups of rats with the opposite sequence of microinjections.

In protocol 3 we performed ES of the ADN before and after the sequential microinjections of the vehicles used for solutions of AP-5 and DNQX into the NTS. The data indicated that the bradycardic and hypotensive responses to ES of the ADN were not affected by volume or mechanical distortion produced by microinjections into the NTS. In addition, it is also important to note that in protocols 1 and 2 the effect of AP-5 and DNQX on the bradycardic (full blockade) and hypotensive (partial blockade) responses to ES of the ADN were reversible, considering that 60 min after bilateral microinjections the responses were back to the control levels.

Although the data of the present study indicate that the hypotensive response was not completely blocked by NMDA and non-NMDA receptor antagonists microinjected into the NTS, we cannot rule out the possibility that part of the population of EAA receptors involved in the sympathoinhibitory component of the baroreflex was not reached by the injection of a total volume of 100 nl into the lateral aspects of the NTS, which may explain the remaining hypotensive response after double blockade of NMDA and non-NMDA receptors. This possibility does not seem to be consistent, because the same microinjections were effective in blocking the bradycardic response of the baroreflex. In addition, it is important to consider that the characteristics of the ES of the ADN used in the present study, such as the frequency and intensity of the stimuli, may not resemble the natural depolarization of the baroreceptor afferent fibers, which is activated by the pulse pressure. Despite the limitations of the methods used in the awake rat model, the present data indicate the following positive aspects: 1) the findings are pharmacologically consistent, considering that in previous studies we verified that the doses of AP-5 and DNQX used in the present study were in the range of selectivity for their respective subtypes of EAA receptors (4, 9, 11, 12, 15); 2) the results are not influenced by the distorting effects of anesthetics on neurotransmission in the NTS (14); and 3) it was possible to evaluate the neurotransmission of the sympathoinhibitory component of the baroreceptor afferents in the NTS of awake rats.

We conclude that the processing of the parasympathetic component of the baroreceptor afferents in the NTS is mediated by both NMDA and non-NMDA receptors, whereas the processing of the sympathoinhibitory component seems to be partially mediated by ionotropic receptors and may involve metabotropic receptors or a neurotransmitter other than L-glutamate, considering that the hypotensive responses to ES of the ADN were only partially blocked.

The perspectives of this study are the following. Although the data of the present study indicate that different subtypes of EAA receptors are involved in the processing of each branch of the autonomic nervous system in the NTS, additional studies combining different experimental approaches are required to confirm this possibility. The use of immunohistochemistry combined with eletrophysiological techniques in tissue slices as well as in whole animals will be useful to evaluate the different subpopulations of neurons in the NTS and their respective subtypes of EAA and other classes of receptors involved in this processing system. The study of the complex network of neurotransmission and neuromodulation of the autonomic processing at the NTS level is very important for the understanding of the physiological mechanisms involved in the central neural control of the circulation, which will make important contributions to the knowledge of the physiopathological mechanisms linked to the neural aspects of different models of experimental hypertension.