We previously showed that serotonin (5-HT2) receptor activation in the nucleus of the tractus solitarius (NTS) produced hypotension, bradycardia, and facilitation of the baroreflex bradycardia. Activation of the preoptic area (POA) of the hypothalamus, which is involved in shock-evoked passive behaviors, induces similar modifications. In addition, previous studies showed that blockade of the infralimbic (IL) part of the medial prefrontal cortex, which sends projections to POA, produced an inhibitory influence on the baroreflex cardiac response. Thus, to assess the possible implication of NTS 5-HT2 receptors in passive cardiovascular responses, we analyzed in anesthetized rats the effects of NTS inhibition and NTS 5-HT2 receptor blockade on the cardiovascular modifications induced by chemical (0.3 M d,l-homocysteic acid) and electrical (50 Hz, 150–200 μA) stimulation of IL or POA. Intra-NTS microinjections of muscimol, a GABAA receptor agonist, prevented the decreases in blood pressure and heart rate normally evoked by IL or POA activation. In addition, we found that intra-NTS microinjection of R(+)-α-(2,3-dimethoxyphenyl)-1-[2-(4-fluorophenylethyl)]-4-piperidine-methanol, a specific 5-HT2A receptor antagonist, did not affect the decreases in cardiovascular baseline parameters induced by IL or POA stimulation but prevented the facilitation of the aortic baroreflex bradycardia normally observed during IL (+65 and +60%) or POA (+70 and +69%) electrical and chemical stimulation, respectively. These results show that NTS 5-HT2A receptors play a key role in the enhancement of the cardiac response of the baroreflex but not in the changes in basal heart rate and blood pressure induced by IL or POA stimulation.
- peripheral nervous system
the arterial baroreceptor reflex is a critical mechanism for maintenance of circulatory homeostasis. However, different responses to stress need this reflex to be controlled. Indeed, the baroreflex cardiac response (bradycardia) is suppressed during “active” coping strategies such as the fight/flight defense reaction, when blood flow is redistributed to vascular beds thereby causing an increase in blood pressure (12, 22). In contrast, hyporeactivity, sympathoinhibition, bradycardia, and sometimes enhanced baroreflex bradycardia occur during shock-like “passive” strategies observed after blood loss, deep pain, or intense exercise (1, 3, 5, 6, 19, 32, 42).
The preoptic area (POA) of the hypothalamus has been claimed to contribute to these passive coping strategies because its activation produces behavioral relaxation (33) and depressive cardiovascular changes through its projection in the ventral part of the periaqueductal gray (vPAG; see Ref. 15), a region responsible for the production of “shock-like” responses to visceral pain (3). In the forebrain, the infralimbic area (IL) of the medial prefrontal cortex, which is known to play a role in “moral emotions” (i.e., guilt and embarrassment; see Ref. 34), could also potentially contribute to passive strategies. Indeed, 1) IL sends massive projections to POA (8, 40), 2) its stimulation in anesthetized rats produces hypotension (38), and 3) IL may exert a tonic facilitatory influence on the baroreflex as shown by the reduction of the maximal cardiac response of this reflex after IL lesion in conscious rats (28, 39).
The nucleus of the tractus solitarius (NTS) is the site of the first synapse of baroreceptor fiber afferent projections (24). Several neurotransmitters in the NTS play a modulatory role in the central control of cardiovascular parameters. In particular, serotonin (5-HT), through the stimulation of NTS 5-HT2 receptors, produces hypotension (18) and a vagus nerve-mediated bradycardia (20, 27, 30). In addition, microinjections of 5-HT2 receptor agonists in the NTS were shown to facilitate the baroreflex bradycardia induced by phenylephrine administration (21).
The aim of our study was to determine whether these NTS 5-HT receptors were involved in the production of all the “passive coping-like” cardiovascular responses to IL or POA stimulation. Thus, as suggested by studies using awake rats, we first characterized, in pentobarbitone sodium-anesthetized rats, the hypotension, bradycardia, and facilitation of the phenylephrine- or aortic depressor nerve-induced cardiac baroreflex bradycardia produced by electrical and chemical [with d,l-homocysteic acid, (DLH) microinjections] stimulation of both IL or POA. Next, we investigated whether these responses were still present after microinjections, in the NTS, of ketanserin, a nonspecific 5-HT2 receptor antagonist, and R(+)-α-(2,3-dimethoxyphenyl)-1-[2-(4-fluorophenylethyl)]-4-piperidine-methanol (MDL-100907) to block specifically local 5-HT2A receptors (16, 37).
MATERIALS AND METHODS
Experiments were performed on 152 male Sprague-Dawley rats, weighing 330–370 g. Animals were kept under controlled environmental conditions (ambient temperature: 21 ± 1°C, 60% relative humidity, food and water ad libitum, alternate 12:12-h light-dark cycles) for at least 1 wk after receipt from the breeding center (CER Janvier, Le Genest-St. Isle, France). Procedures involving animals and their care were all conducted in conformity with the institutional guidelines, which are in compliance with national and international laws and policies (Council directive no. 87–848, 19 October 1987, Ministère de l'Agriculture et de la Forêt, Service Vétérinaire de la Santé et de la Protection Animale, permissions no. 75–116 to M. Hamon, no. 75–117 to R. Laguzzi, and no. 75–855 to C. Sévoz-Couche). The investigations also conformed to the Guide for the Care and Use of Laboratory Animals published by the United States National Institutes of Health (NIH Publication no. 85–23, revised 1996).
Rats were anesthetized with pentobarbitone sodium (60 mg/kg ip), and the depth of anesthesia was regularly assessed by pinching a hindpaw and monitoring the stability of blood pressure and heart rate (HR). In case of withdrawal reflex and/or variations of cardiovascular parameters, a supplementary dose of pentobarbitone sodium was given (5–10 mg/kg iv). A cannula was inserted in the femoral vein for administration of drugs and/or additional doses of pentobarbitone. Systemic blood pressure, mean blood pressure (MBP), and HR were monitored (Pressure Processor and DC Amplifier; Gould, Courtaboeuf, France) through a catheter inserted in the femoral artery. Electrocardiogram (ECG) was recorded using stainless steel pins placed subcutaneously in fore- and hindpaws; signals were amplified and filtered (Universal Amplifier). The R wave of the ECG was discriminated with a window discriminator and used to generate pulses. Arterial blood pressure and ECG pulse signals were relayed to a 1401 interface (1401 Plus; CED, Cambridge, UK) connected to a computer running Spike 2 software (CED). Cardiac intervals and HR were automatically computed from R wave pulses (bin size: 1 s). Rectal temperature was maintained at 37°C with a thermostatically controlled heating blanket. In most rats, the left aortic depressor nerve was dissected out from the vagus nerve by a lateral approach and placed on silver bipolar hook electrodes for electrical stimulation.
Procedures for IL or POA Stimulation
Pentobarbitone-anaesthetized rats were placed in a stereotaxic frame with the head fixed in a horizontal position. A craniotomy was performed, and a bipolar stimulating electrode or a single-barrel glass micropipette (<100 μm external diameter) connected to a Hamilton microsyringe was lowered in the left IL [posterior (P) +2.5 to +3.2; lateral (L) 0.3–0.8; vertical (−4 to −5)] or POA (P: −1.0 to 0.5; L: 1.5–2; V: −6.5 to −7.5, in mm from bregma), using stereotaxic coordinates from Paxinos and Watson's atlas (26). Both regions were identified by monitoring cardiovascular responses: a decrease in both MBP and HR, caused by local electrical stimulation (50 Hz, 1 ms pulse duration, 100 to 200 μA, 10 s) or microinjection of DLH (0.3 M; 100 nl) through the glass micropipette.
Procedures for Intra-NTS Microinjections
The dorsal surface of the brain stem was exposed through a limited occipital craniotomy. A single-barrel glass micropipette was lowered in the NTS at predetermined coordinates for chemical inhibition and for microinjections of vehicle or 5-HT2 receptor antagonists (0.4 mm rostral and lateral to the calamus scriptorius and 0.4 mm beneath the dorsal surface of the medulla; see Ref. 21). The time interval between two symmetrical microinjections (100 nl) was <1 min. Intra-NTS microinjections of saline produced no changes in basal cardiovascular parameters and baroreflex bradycardia (31). In the present study, we found that intra-NTS microinjections of MDL-100907 vehicle (DMSO) affected neither basal HR (−5 ± 3 beats/min, n = 8) and blood pressure (+4 ± 1 mmHg, n = 8), nor the cardiac response of the baroreflex (−70 ± 4 and −72 ± 2 beats/min before and after DMSO, respectively, n = 8, not significant).
Quantification of the Cardiac Baroreflex Response
The baroreflex response was triggered by electrical stimulation of the aortic nerve (20 Hz, 1 ms, 50–100 μA). By adjusting the stimulation parameters, and following the procedure described in Inui et al. (14, 15), bradycardic responses to stimulation of the aortic nerve were reduced to below the maximum response so that the putative facilitatory effects of IL or POA stimulation on the responses could be clearly demonstrated. In all cases, the cardiac reflex response was defined as the peak decrease in HR (ΔHR) over HR baseline value (aortic-induced cardiac response = ΔHR/HR baseline).
Some rats received phenylephrine (2–5 μg/kg iv) to increase blood pressure for activation of both carotid and aortic baroreceptors. The dose was adjusted to obtain a response below the expected maximum value (see above). Under such conditions, the cardiac reflex response was defined as the ratio of ΔHR, expressed as percent change compared with HR baseline value, over the maximal increase in MBP [phenylephrine-induced cardiac response, in mmHg−1: 100 × (ΔHR/HR baseline)/ΔMBP]. In addition, modifications of HR or cardiac intervals in response to an increase in MBP triggered by phenylephrine administration were analyzed using a four-parameter common sigmoid curve equation (39). Regression analysis of the rectilinear part of both curves allowed calculation of the baroreflex slope for each experimental condition.
Quantification of the Sympathetic Baroreflex Response
Because the hypotension induced by aortic depressor nerve stimulation is the combination of a sympathoinhibition (vasodilation) and a parasympathetic excitation (bradycardia; see Ref. 17), we analyzed the sympathetic reflex response during IL or POA stimulation in a group of rats (n = 10) in which the bradycardic reflex component was suppressed by pretreatment with α-methylatropine (30 μg/kg iv; see Ref. 9). On its own, α-methylatropine produced an increase in HR (+53 ± 5 beats/min from a baseline of 365 ± 5 beats/min) that lasted for >30 min but no significant changes in basal blood pressure (+3 ± 1 mmHg from a baseline of 105 ± 2 mmHg). Under this condition, the aortic-induced sympathetic response was defined as the maximal decrease in blood pressure (ΔMBP) observed during stimulation of the aortic depressor nerve, expressed as a percentage of the MBP baseline value in α-methylatropine-pretreated rats.
IL stimulation induces inhibition of the tonic sympathoexcitation of cardiac β1-adrenergic receptors (29, 38). To analyze the effect of IL (or POA) stimulation on the baroreflex bradycardia in the absence of any sympathetic influence on HR, another group of 10 rats was injected with atenolol (1 mg/kg iv), a specific antagonist at β-adrenoceptors.
Effects of IL or POA Stimulation on the Baroreflex Responses
The aortic depressor nerve was stimulated for 4 s while monitoring cardiovascular changes (“control”). Later (5 min), electrical or chemical stimulation of IL or POA was performed. In case of electrical stimulation of IL or POA, the aortic nerve was stimulated for 4 s precisely at the time when the cardiovascular changes evoked by cortical or hypothalamic stimulation (i.e., bradycardia and hypotension) were maximal (“experimental”). The effects of IL or POA stimulation on the baroreflex responses were calculated as the percent changes in experimental vs. control conditions. In case of chemical stimulation of IL or POA, experimental baroreflex responses were determined 5 min after DLH microinjections, when cardiovascular modifications induced from both regions were maximal.
Phenylephrine (5 μg/kg iv) was first administered (control). Later (10 min), this treatment was repeated concomitantly with IL electrical stimulation (experimental). The effects of IL stimulation on the baroreflex bradycardia were calculated as the percent changes in experimental vs. control conditions.
Effects of IL or POA Stimulation on the Baroreflex Responses after Intra-NTS Microinjections
In experiments aimed at analyzing the effects of intra-NTS microinjections of various receptor ligands on IL or POA stimulation-induced facilitation of the baroreflex bradycardia, three successive determinations of the latter response were performed before (10 min) microinjections and considered stable if they differed from each other by <10%. Next, microinjections were performed, and another determination of the baroreflex bradycardia facilitation was made 15 min later, to be compared with the third determination of the preceding series.
At the end of experiments, electrolytic lesions (50 Hz, 4 mA, 20 s) were made at central stimulation sites, and methylene blue (0.1 μl) was microinjected in the injection sites within the NTS. Rats were then perfused intracardially, and brain coronal sections (60 μm) were cut and stained with neutral red. Methylene blue spread over a maximum of 0.2 mm from the injection site.
Absolute values are expressed as means ± SE of n rats. Statistical comparisons between nonpretreated rats and rats pretreated with either atenolol or α-methylatropine, or subjected to intra-NTS microinjections, were made using a two-way ANOVA followed by post hoc analysis (Bonferroni test). In all other cases, Student's paired t-test was applied.
Atenolol base (Sigma-Aldrich, Saint Quentin-Fallavier, France), α-methylatropine (RBI, Natick, MA), DLH (Sigma Chemicals, St Louis, MO), ketanserin (Janssen, Beerse, Belgium), muscimol (Sigma Chemicals), and phenylephrine hydrochloride (Merck Sharp and Dohme-Chibret, Paris, France) were dissolved in saline.
MDL-100907 was a kind gift of Dr J. G. Wettstein (Hoechst Marion Roussel, Bridgewater, NJ). It was dissolved in DMSO.
The pH of all solutions microinjected in the NTS was adjusted to 7.4.
Baseline values of MBP and HR in pentobarbitone-anaesthetized rats were 101 ± 2 mmHg and 363 ± 6 beats/min, respectively (n = 152).
Effects of IL Stimulation on the Baroreflex Cardiac Response
Microinjections of DLH (0.3 M) in IL (n = 6, Fig. 1A) produced limited but significant bradycardia (−30 ± 1 beats/min, P < 0.05) and a decrease in blood pressure (−14 ± 3 mmHg, P < 0.05) that lasted for ∼10 min. The amplitude of aortic-induced bradycardia was significantly larger (+65%) during IL stimulation than under control, nonstimulated, conditions (−64 ± 8 beats/min from a baseline of 355 ± 10 beats/min and −98 ± 2 beats/min from a baseline of 335 ± 11 beats/min, before and during IL stimulation, respectively, n = 6, P < 0.05, Fig 2). These responses were obtained only when DLH was microinjected at sites located between 2.7 and 3.0 mm rostrally to bregma (Fig. 1A).
Electrical stimulation of IL (n = 35) at the same sites (Fig. 1A) as those found to be effective in chemical stimulation experiments also produced limited but significant decreases in HR (−24 ± 2 beats/min, P < 0.05) and blood pressure (−17 ± 1 mmHg, P < 0.05). Both cardiovascular parameters began to return to normal basal values after stimulation was stopped. The amplitude of aortic-induced cardiac response was significantly enhanced (+60%) during IL stimulation (ΔHR/basal HR: −0.19 ± 0.02 and −0.28 ± 0.01 before and during IL stimulation, respectively, n = 25, P < 0.005).
When the baroreflex was triggered by phenylephrine (5 μg/kg iv), the reflex bradycardia was also increased (+52%) during electrical stimulation of IL (at sites where stimulation facilitated aortic baroreflex bradycardia) compared with that measured under control (i.e., no IL stimulation) conditions [(ΔHR/HR)/ΔMBP: −0.65 ± 0.10 mmHg−1 and −1.05 ± 0.10 mmHg−1 before and during IL stimulation, respectively, n = 10, P < 0.05, Fig. 3]. In addition, IL stimulation (n = 10) shifted the lower HR plateau of the baroreflex sigmoidal curve (see materials and methods) toward a significantly smaller value (Fig. 4 and Table 1). Indeed, regression analysis of the linear part of this curve showed that the baroreflex gain was significantly increased during IL stimulation (P < 0.05, Fig. 4A). Concomitant changes in cardiac interval paralleled those in HR (P < 0.05, Fig. 4B).
Effects of POA Stimulation on the Baroreflex Cardiac Response
POA activation by local DLH microinjections (n = 9, Fig. 1B) induced small decreases in HR (−35 ± 2 beats/min, P < 0.05) and blood pressure (−13 ± 2 mmHg, P < 0.05). In addition, 5 min after DLH administration, the bradycardic response to aortic nerve stimulation was significantly larger (+70%) than that measured before the treatment (−62 ± 7 beats/min from a baseline of 365 ± 10 beats/min and −102 ± 16 beats/min from a baseline of 345 ± 8 beats/min, before and after DLH administration, respectively, P < 0.05). This potentiation lasted ∼10 min. It was observed only when DLH was microinjected at sites located at 0 to −0.5 mm caudally to bregma (Fig. 1B).
Electrical (Fig. 1B) stimulation of POA (n = 27) at the same sites as those where DLH microinjections enhanced the bradycardic baroreflex response produced small decreases in HR (−32 ± 1 beats/min, P < 0.05) and blood pressure (−16 ± 1 mmHg, P < 0.05). Basal levels returned to normal values as soon as stimulation was stopped. The aortic-induced cardiac response was enhanced (+69%) during stimulation of POA (ΔHR/basal HR: −0.18 ± 0.02 vs. −0.30 ± 0.02 before vs. during stimulation, respectively, P < 0.05).
Effects of IL or POA Stimulation on the Baroreflex Cardiac Response in Atenolol-pretreated Rats
On its own, atenolol administration decreased HR (−40 ± 5 beats/min from a baseline of 359 ± 6 beats/min, n = 10, P < 0.05) for approximately 1.5 h and produced a small but not significant reduction of the aortic baroreflex bradycardia (−53 ± 8 vs. −66 ± 7 beats/min, n = 10). In addition, administration of atenolol prevented the bradycardia normally observed during electrical stimulation of IL (−25 ± 1 vs. −3 ± 1 beats/min before vs. after atenolol, respectively, n = 5, P < 0.05) or POA (−35 ± 2 vs. −7.0 ± 3.2 beats/min before vs. after atenolol, respectively, n = 5, P < 0.05). However, the facilitation of the aortic-induced bradycardic response induced by both stimulations was of the same amplitude in atenolol-pretreated (+58%, n = 5, and +68%, n = 5, respectively) as in nonpretreated rats (+60% and +69%, respectively, see above).
Effects of IL or POA Stimulation on the Sympathetic Response in α-Methylatropine-pretreated Rats
Administration of α-methylatropine (30 μg/kg iv, n = 10) totally blocked the aortic baroreflex parasympathetic bradycardia (Fig. 5). In addition, the baroreflex decrease in blood pressure, which is the result of both reflex sympathoinhibition and parasympathoactivation (see Ref. 17), was significantly reduced in α-methylatropine-pretreated rats (−21 ± 1 and −13 ± 1 mmHg before and after α-methylatropine, respectively, P < 0.05, n = 10). In the latter animals, the hypotensive aortic baroreflex response was not significantly changed when baroreflex was triggered during electrical stimulation of IL (−14 ± 1 mmHg, n = 5, Fig. 5) or POA (−13 ± 2 mmHg, n = 5).
Effects of Intra-NTS Microinjections on IL Stimulation-induced Cardiovascular Responses
After microinjections of saline (n = 4) or DMSO (n = 4) in the NTS at levels indicated in Fig. 1C, the effects of IL stimulation on baseline cardiovascular parameters and baroreflex cardiac response (see Table 2) were similar to those observed in naive noninjected rats (see above).
Microinjections of muscimol.
Owens et al. (23) found that the hypotensive response to IL stimulation is abolished during NTS inhibition induced by local microinjection of muscimol, a specific γ-aminobutyric acid (GABAA) receptor agonist. We herein noted that muscimol (50 pmol, n = 4) microinjected at the level of the calamus scriptorius in the NTS (Fig. 1), which produced increases in blood pressure (+15%, P < 0.05) and HR (+10%, P < 0.05, Table 2), was able to prevent not only the hypotension but also the bradycardic response to IL stimulation (Table 2). As expected from NTS inhibition by such local administration of muscimol, reflex bradycardia normally evoked by intravenous phenylephrine administration was completely suppressed in muscimol-pretreated rats [(ΔHR/HR)/ΔMBP: −0.68 ± 0.10 vs. −0.05 ± 0.01, before vs. after intra-NTS muscimol, respectively, n = 4, P < 0.05].
Microinjections of 5-HT2 receptor antagonists.
Bilateral intra-NTS microinjections of ketanserin (10 pmol) at predetermined sites (21) produced a short (1 min) nonsignificant decrease in blood pressure (−8.1 mmHg, n = 17) without affecting HR. Furthermore, ketanserin microinjections modified neither the aortic baroreflex response (−70 ± 5 and −75 ± 4 beats/min before and after ketanserin, respectively, n = 17) nor the hypotensive and bradycardic responses to IL stimulation (Table 2). However, after intra-NTS microinjections of ketanserin, IL stimulation no longer induced a potentiation of the aortic baroreflex bradycardia (n = 13, Table 2). In the same manner, after ketanserin microinjections, IL stimulation was also unable to potentiate the phenylephrine-evoked bradycardic response (+50 and +5% before and after ketanserin pretreatment, respectively, n = 4, P < 0.05).
Microinjections of MDL-100907 (10 pmol, n = 22) in the NTS (Fig 1) produced no significant changes in basal HR (+5 ± 3 beats/min) and blood pressure (−2 ± 3 mmHg) and did not affect the aortic baroreflex response (−71 ± 5 and −69 ± 4 beats/min before and after 10 pmol MDL-100907, respectively). In addition, IL stimulation-induced hypotension and decrease in HR after MDL-100907 (10 pmol) microinjections were not different from those occurring in naive rats (Table 2). However, intra-NTS administration of MDL-100907 (n = 12) dose dependently prevented the expected potentiation of the aortic baroreflex bradycardia by IL stimulation [+50 and −5% before and after 10 pmol MDL-100907, respectively, n = 6, P < 0.05 (Fig. 6); +55 and +14% before and after 5 pmol MDL-100907, respectively, n = 6, P < 0.05].This effect and that of ketaserin (Table 2) lasted at least 30 min.
Effects of intra-NTS Microinjections on POA Stimulation-induced Cardiovascular Responses
Intra-NTS administration of saline (n = 4) or DMSO (n = 4) affected neither the decreases in basal HR (−32 ± 3 and −36 ± 3 beats/min, respectively) and blood pressure (−15 ± 2 and −16 ± 1 mmHg, respectively) nor the facilitation of the baroreflex bradycardia (+69 and +68%, respectively) produced by POA electrical stimulation (see above for the corresponding values in nonmicroinjected rats).
Microinjections of muscimol.
Intra-NTS administration of muscimol (50 pmol, n = 4) was found to prevent the hypotension (−15 ± 2 and −2 ± 1 mmHg before and after muscimol, respectively, P < 0.05) and the bradycardic response (−30 ± 3 and −5 ± 2 beats/min before and after muscimol, respectively, P < 0.05) to POA stimulation.
Microinjections of MDL-100907.
POA stimulation-induced hypotension (−15 ± 4 mmHg, n = 5) and decrease in HR (−37 ± 5 beats/min, n = 5) after intra-NTS microinjection of MDL-100907 (10 pmol) were not different from those observed in naive rats (see above). However, the facilitation of the baroreflex bradycardia normally observed during POA stimulation was prevented by MDL-100907 (n = 10) microinjections [+65 and −8% before and after 10 pmol MDL-100907, respectively, n = 5, P < 0.05 (Fig. 7); +68 and +17% before and after 5 pmol MDL-100907, respectively, n = 5, P < 0.05].
This study demonstrated that, in anesthetized rats, stimulation of both the IL cortex and the POA in the hypothalamus potentiates the cardiac response of the baroreflex and that NTS 5-HT2A receptors are critically involved in this effect.
In conscious animals, chemical lesions of IL caused an inhibition of the baroreflex bradycardia (28, 39), suggesting that IL normally contributes to a tonic facilitatory control of this reflex response. Direct stimulation of IL in awake rats was found to induce cardiovascular changes similar to those associated with active coping strategies, i.e., increases in HR and blood pressure, which is in sharp contrast with that observed in anesthetized rats (38) and conscious rabbits (2), associated with an inhibition of the baroreflex bradycardia (29). Thus comparison between anaesthetized and conscious rats suggests that a group of neurons involved in the defense reaction exists close to IL region and exerts a predominant influence in unanesthetized rats. Because the main goal of our study was to analyze the role of the NTS 5-HT2 receptors in the facilitation of the baroreflex bradycardia induced by IL or POA stimulation, such differences between conscious and anesthetized rats led us to perform relevant experiments in anesthetized rats that exhibit no sign of the defense reaction during these stimulations. Thus we found here that stimulation of both IL or POA was able to enhance the maximal aortic cardiac response of the baroreflex in anesthetized rats. Similar results were obtained after electrical and chemical stimulation, allowing the conclusion that local cells but not fibers “en passage” were involved in the observed effect. In addition, IL stimulation produced an increase of the slope of the baroreflex sensitivity curve (expressed as changes of either HR or cardiac interval) evoked by phenylephrine administration, indicating not only the maximal cardiac reflex response but also the sensitivity of this reflex were enhanced by such stimulation. Interestingly, intravenous administration of atenolol did not affect the potentiation of the baroreflex bradycardia observed during IL or POA stimulation, showing that this effect occurred independent of the cardiac sympathoinhibition triggered by these stimulations.
Concerning the sympathetic component of the baroreflex, previous studies in vagotomized rats (in which the parasympathetic component is inactivated) showed that POA stimulation produced a slight facilitation of the residual reflex hypotension (15). However, in our studies, in α-methyl-atropine-pretreated rats, IL or POA stimulation was unable to produce any facilitatory influence on the residual reflex hypotension.
Previous studies suggested that the NTS could be involved in IL-induced cardiovascular changes. It has notably been shown that IL sent direct projections to the NTS (36) and that the hypotensive responses to IL stimulation are suppressed by chemical inhibition of the NTS (10, 23). In addition, clear-cut data showed that the caudal part of the nucleus raphe magnus, which is known to send serotonergic projections to the NTS (35), is involved in all the cardiovascular changes induced by POA and vPAG stimulation (14, 15). Here we found that both hypotensive and bradycardic responses to IL or POA stimulation were abolished by chemical inhibition of the NTS, confirming the key role of this nucleus in these effects. However, whether NTS cells are activated directly by IL or POA or via the stimulation of the nucleus raphe magnus, to produce these responses, is still an unsolved question.
It has previously been established that microinjection of 5-HT2 receptor agonists in the NTS induces hypotension and bradycardia (20, 27) and that local administration of subthreshold doses only of such ligands can potentiate the cardiac response of the baroreflex (21). Therefore, it could be hypothesized that NTS 5-HT2 receptors are key components in the hypotension, bradycardia, and enhanced cardiac baroreflex response occurring during IL or POA stimulation. Actually, we found that intra-NTS microinjections of ketanserin, a potent 5-HT2 receptor antagonist, were able to prevent the facilitation of the cardiac response of the baroreflex but not the hypotension and bradycardia induced by IL stimulation. Thus NTS 5-HT2 receptors appear to be involved in the facilitation of the baroreflex bradycardia only. Ketanserin has high affinity for 5-HT2A (pKi = 8.9) and 5-HT2C (pKi = 7.0) receptors and lower affinity for 5-HT2B receptors (pKi = 5.4; see Ref. 25), which suggests that 5-HT2A/2C receptors possibly mediated the facilitatory effect of IL stimulation on the baroreflex bradycardia. Like that found with ketanserin, intra-NTS microinjections of MDL-100907, a selective 5-HT2A receptor antagonist (16), also prevented the facilitation of the reflex bradycardia but not the decrease in blood pressure and HR induced by IL stimulation. In addition, MDL-100907, which did not affect POA-induced changes in basal HR and blood pressure, also suppressed the facilitation of the cardiac response of the baroreflex normally evoked by POA activation. Altogether, these results show that the NTS is involved in IL or POA stimulation-induced hypotensive and bradycardic responses and that the facilitation of the baroreflex bradycardia produced by IL or POA stimulation is mediated by NTS 5-HT2A receptor activation. The latter effect might be causally related to 5-HT2A receptor-mediated partial depolarization of the cell membrane, which is known to promote N-methyl-d-aspartate (NMDA) receptor activation, at the origin of baroreflex signaling within the NTS (7). In line with this idea, it is well established that excitatory amino acid synaptic transmission through NMDA receptors requires prior cell membrane depolarization (11) and that 5-HT2A and NMDA receptors are colocalized in some neurons within the NTS (13).
We previously found that the defense reaction involves NTS 5-HT3 receptors to produce inhibition of the baroreflex bradycardia (31). It thus appears that serotonergic fibers, probably originating from different raphe nuclei or different subareas within the raphe magnus, can modulate the baroreflex cardiac response in opposite ways during active and passive coping strategies. On the one hand, some fibers would target NTS cells endowed with 5-HT3 receptors, in which activation produces the inhibition of the baroreflex bradycardia through a local GABAergic system (4, 31). On the other hand, other raphe fibers would target NTS cells endowed with 5-HT2A receptors to promote the cardiac response of the baroreflex (Fig. 8, see details in legend).
In conclusion, the present data provide new insights regarding the neural pathway underlying shock-like passive coping responses. These responses very likely involve activation from the prefrontal cortex at the IL level to the NTS through POA, vPAG, and nucleus raphe magnus, as already suggested in a previous review (22). In addition, our data show that NTS 5-HT2A receptors play a key role in at least one of these responses, i.e., the enhancement of the baroreflex bradycardia.
This research has been supported by grants from Institut National de la Santé et de la Recherche Médicale and Université Pierre et Marie Curie.
We are grateful to Dr. J. G. Wettstein (Hoechst Marion Roussel, Bridgewater, NJ) for generous gifts of MDL-100907.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society