INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

BESIDES THE WELL-KNOWN EFFECTS of oxytocin (OT) on uterine contraction and milk ejection in females, this peptide is important in other complex physiological functions, behavioral, gastrointestinal, renal, and cardiovascular (29). OT is present in males and females in similar levels and is secreted in response to stress, volume, osmotic and satiety stimuli (12, 15, 21, 22). The role of OT in cardiovascular regulation has been the subject of limited investigation. OT is a weaker vasoconstrictor than vasopressin (VP), but it has significant effects on blood pressure, vascular tone, and renal function (25, 26, 39). In terms of central actions, OT is present in brain stem integrative cardiovascular centers, the result of projections from parvicellular neurons of the paraventricular nucleus (PVN) of the hypothalamus (3, 23, 34, 35). Manipulation of PVN OT systems with antisense oligonucleotides, antagonists, or lesions results in prominent alterations in cardiovascular responses to stress and peptidergic stimulation and alterations in salt consumption (4, 16, 22). Brain stem administration of OT or OT-receptor antagonists changed local neuronal activity (6, 8) and autonomic control of the heart (32).

Reflex control of the circulation involves a continuous interaction of primary integrating brain stem nuclei [nucleus of the solitary tract (NTS), dorsal motor nucleus of the vagus (DMV), nucleus ambiguous (NA), caudal and rostral ventrolateral medulla] with suprabulbar-modulating areas, such as the PVN (7). Brain stem regions are innervated by multiple peptidergic systems (OT, VP, enkephalins, substance P, and others), all of which have been implicated in cardiovascular control (37). An interesting possibility is that OT (as well as VP) projections from the PVN to the NTS, the first synaptic relay of peripheral afferents to the brain, act as important links in modulating autonomic control of the circulation (17, 20). Previous data demonstrated that oxytocinergic projections to the solitary vagal complex regulate exercise-induced tachycardia. OT is released into the NTS/DMV when trained rats exercise, but exogenous OT administered into this area caused a marked blunting of tachycardic response in both sedentary and trained rats (2). From these data, we hypothesized that OT projections to the NTS/DMV would modulate baroreflex control of the heart. Actually, new studies, using a genetically modified mouse strain lacking the OT gene, showed changes in baroreflex function (19) that are consistent with the blunted tachycardic response observed in the rat (2). Nevertheless, there are some controversies concerning the nature of OT regulation of cardiac function. For example, OT has been shown to have no effect or to decrease basal heart rate (HR) (10, 25), although there are reports of both a reduction and an enhancement in baroreflex gain (27, 33).

The objective of the present study was to investigate the role of OT in baroreceptor reflex control of HR using a model in which drugs can be injected into specific brain stem regions of the conscious rat. We studied the effects of OT peptide administration and OT-receptor blockade in the solitary vagal complex on HR changes produced by pressor increases and decreases. We further investigated the relative participation of vagal and sympathetic outflow in baroreflex modulation by OT. Because the solitary vagal complex in the rat is an elongated structure that receives various afferent projections, we first determined the pattern of dorsal brain stem oxytocinergic innervation before proceeding to the functional studies.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals. Male Wistar-Kyoto rats, 3-4 mo old, weighing 250-330 g were used. They were kept on a 12:12-h light-dark cycle with standard laboratory chow and water provided ad libitum. All surgical procedures and protocols used are in accordance with "Ethical Principles in Animal Research," adopted by the Brazilian College of Animal Experimentation, and were approved by the University of Sao Paulo Ethical Committee for Animal Research.

Immunohistochemistry and radioimmunoassay. Rats were killed by decapitation, and the brains were rapidly removed. Brains were fixed by immersion in Bouins solution (20% sucrose) for 4 days. Series of 20-µm cryostat sections (1 in 5) were collected in 0.01 M PBS. Sections were processed for Nissl (general histology) and OT immunohistochemistry. The immunostaining protocol used 48-h antisera incubation followed by the peroxidase/3-3'diamino benzidine enhanced with nickel (Vectastain Elite, Vector Labs, Burlingame, CA). The rabbit OT antisera were produced in our laboratory and used at a dilution of 1/50,000.

Animal surgery. Groups of rats were submitted to chronic cannulation of the dorsal brain stem, according to a technique developed in our laboratory (18). Briefly, the rats were anesthetized with Nembutal (40 mg/kg iv) and placed in a stereotaxic apparatus (Kopf, Tujunga, CA) with their head in a horizontal position. The skull was widely exposed, and one screw was fixed in the parietal bone. A small window was opened caudally to the lambda to allow the introduction of a stainless steel guide cannula (17-mm length, 0.6-mm outside diameter) at an angle of 24°. The stereotaxic coordinates for the solitary vagal complex (comprising the NTS and DMV or NTS/DMV) placement were as follows: 1 mm caudal to interaural line, 0.4 mm lateral (right or left) to the midline, and 8.92 mm below the skull surface. The guide cannula did not reach the NTS/DMV area, but it lay in the ventral cerebellum, avoiding tissue damage of the area to be studied. For the fourth ventricle cannulation, the rostrocaudal and lateral coordinates were the same, but the guide cannula was introduced 8.8 mm from the skull surface. The cannula was fixed in the skull with fast polymerizing methacrylate cement and was closed by an occluder. The rats were given penicillin (Pentabiotico Veterinario, Fontoura Wyeth, São Paulo, Brazil) and kept in individual Plexiglas cages. After a recovery period of 3-4 days, catheters (PVC tubing, Critchley, Silverwater, New South Wales, Australia) were inserted into a femoral artery and vein for cardiovascular monitoring and drug administration. The catheters were tunneled subcutaneously to the back of the neck where they were fixed with sutures. The rats were tested 1 day after the vascular surgery.

Experimental protocols. Arterial pressure (AP; systolic, diastolic, and mean) and HR (Biotach, triggered by AP pulses) were recorded continuously (P23XL transducer + preamplifier, connected to RS3400 recorder, Gould, Cleveland, OH) in freely moving rats. At the beginning of the experimental session, the occluder was removed from the guide cannula, and 20-30 min were allowed for stabilization of cardiovascular parameters before the measurement of basal values of AP and HR. The first set of experiments consisted of the determination of control values and responses to loading/unloading of baroreceptors in two specific situations: 1) after administration of OT antagonist [Deamino-Cys¹, D-Tyr (Et)², Thr⁴, Orn⁸-Oxytocin (OT_ant); Bachem] at a dose of 20 pmol/200 nl into the NTS/DMV or fourth ventricle and 2) after administration of OT, at a dose of 20 pmol/200 nl into the same area. In both protocols, vehicle (Veh; 0.9% NaCl, 200 nl) was administered before the control measurements. The OT dose used was selected in preliminary experiments (n = 4) as the highest dose that did not change baseline levels of mean AP (MAP) and HR when administered into the NTS/DMV of conscious freely moving rats. The efficacy and dose of the OT_ant used to block OT effects were also tested in the preliminary experiments; in two tests made 1 and 6 h after the NTS/DMV administration, the effect of the antagonist was shown to be present. For microinjection of Veh or peptides, a 33-gauge needle (18-mm length, connected by PE-10 tubing to a microliter syringe 701-N, Hamilton, Reno, NV) was introduced into the guide cannula. Microinjections lasted 15-20 s, and the needle was removed 5-10 s later. Control values of MAP and HR were obtained 20-30 min later. For baroreceptor stimulation, 100-µl bolus injections of graded doses of phenylephrine (0.1 up to 6.4 µg/kg) and sodium nitroprusside (0.2 up to 12.8 µg/kg) were given into the femoral vein. Phenylephrine and nitroprusside injections were made in a random order, and subsequent injections were not made until the recorded parameters had returned to preinjection levels. Therefore, for each animal, the protocol was repeated twice: Veh and OT or Veh and OT_ant treatments. Because of the long-lasting effects of centrally administered OT (28), different groups of rats were used for OT_ant and OT administration into the NTS/DMV.

Data and statistical analysis. Results are presented as means ± SE. Functional data are expressed as absolute values and responses. Baroreceptor reflex control of HR, determined for each rat, was estimated by the sigmoidal logistic equation (11, 13) fitted to data points. It correlates absolute HR and MAP values during transient pressure changes induced by phenylephrine and sodium nitroprusside injections
where P1 = lower HR plateau, P2 = HR range, P3 = a curvature coefficient that is independent of range, and P4 = MAP₅₀, i.e., the MAP at half the HR range. The average gain (G) or slope of the curve between the two inflection points is given by G = P2 × P3/4 and the upper plateau = P1 + HR range (P2). To compare our data to previously published studies (27, 33), we also adjust linear regression equations to data points. This analysis correlates HR changes with changes in MAP (HR = aMAP + b) during both loading and unloading of baroreceptors. Parameters for both sigmoidal and linear fitting were used to compare the effects of OT and OT_ant on baroreceptor function.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Oxytocinergic projections/OT content. Figure 1 shows that OT-like immunoreactivity is present in fibers in the NTS/DMV, mainly from the calamus scriptorium up to 500-550 µm rostral. OT-like-projecting fibers are very dense in the DMV and permeated a large extension of the medial NTS. Surrounding regions such as the area postrema, hypoglossal nucleus, gracilis and cuneate nuclei are largely devoid of OT-like immunoreactivity. Measurement of peptide content in the dorsal brain stem area (including the NTS/DMV) showed much higher levels of OT than VP. OT content was 55 ± 6 pg/area (n = 6), 3.7-fold higher than VP content (15 ± 5 pg/area, n = 6).

View larger version (151K): [in this window] [in a new window]   Fig. 1.   Immunoreactivity for oxytocin (OT; golden fibers and dots) in dorsal brain stem areas showing dense projections in the medial part of the nucleus of the solitary tract (mNTS) and dorsal motor nucleus of the vagus (DMN), extending from the calamus scriptorium up to 500-550 µm rostral. Neurons in the area postrema (AP), hypoglossal (XII), and gracilis nuclei were devoid of reactivity. Left insets: magnifications of the white box shown at right (675 × 520 µm). Anterior-posterior coordinates are +500 µm (A) and +150 µm rostral to the calamus scriptorium (B).

Basal values and pressor/depressor sensitivity. Table 1 summarizes the basal values of MAP and HR in the four groups. MAP and HR values after Veh administration did not differ among groups. Veh, OT, or OT_ant administration into the NTS/DMV or in the fourth cerebral ventricle caused only transient changes, but it did not alter baseline values of MAP and HR. Intravenous bolus injections of graded doses of phenylephrine and sodium nitroprusside caused dose-related changes in MAP, with parameters depicted in Table 2. Maximal pressor/depressor responses and respective effective dose (ED₅₀) did not differ between groups. OT or OT_ant administration into the NTS/DMV or fourth ventricle did not change MAP responses to phenylephrine and sodium nitroprusside.

Baroreceptor reflex control of HR. Reflex control of the heart was analyzed by fitting sigmoidal curves to data points for each condition in each rat. The average sigmoidal fittings for the groups administered Veh and OT or Veh and OT_ant into the NTS/DMV are presented in Fig. 2. OT and its receptor antagonist changed baroreceptor reflex control of HR in opposite directions. OT displaced the inferior plateau toward a lower HR value, increasing significantly the HR range (149 ± 11 to 182 ± 11 beats/min; Table 3). On the other hand, OT_ant treatment displaced the lower HR plateau toward a higher HR value, decreasing significantly both the operating range (166 ± 12 to 131 ± 11 beats/min) and gain of the reflex (5.68 ± 1.62 to 2.83 ± 1.05 beats · min¹ · mmHg¹; Table 3). Therefore, maximal bradycardic response was increased by OT administration and reduced by OT-receptor blockade (75 ± 11 vs. 56 ± 9 and 35 ± 2 vs. 61 ± 13 beats/min, treatment vs. control, OT and OT_ant, respectively; Fig. 2). Accordingly, OT increased significantly the sensitivity of reflex bradycardia, whereas OT_ant decreased it (parameters of linear regression equations in Table 3). There were no alterations in the tachycardic responses during unloading of baroreceptors (Fig. 2 and Table 3). During baroreflex testing, OT-receptor blockade of the NTS/DMV also caused a small increase in control MAP, whereas OT administration produced a small decrease in control HR.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   A: average logistic function curve showing the relationship between mean arterial pressure (MAP) and heart rate (HR) during baroreflex testing in conscious rats treated with vehicle (Veh) and OT or OT antagonist (OTant) into the solitary vagal complex (NTS/DMV). Points on curves represent average control values of MAP and HR. B: comparison of maximal tachycardic and bradycardic responses observed following treatments. Significance at *P < 0.05 vs. respective Veh treatment.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   A: average logistic function curve showing the relationship between MAP and HR during baroreflex testing in conscious rats treated with Veh and OT or OTant into the fourth cerebral ventricle. Points on curves represent average control values of MAP and HR. B: comparison of maximal tachycardic and bradycardic responses observed following treatments.

Sympathetic and parasympathetic blockade. Selective ₁- or cholinergic antagonist treatment did not affect baseline MAP, but it caused a significant decrease (13-16%) or significant increase (12-17%) in baseline HR, respectively (values on Table 4). As expected, sympathetic and vagal blockade determined marked blunting of both bradycardic and tachycardic responses to loading/unloading of baroreceptors. The sensitivity of reflex bradycardia was reduced by 53 and 73% after Atn and Atr treatments, whereas the gain of reflex tachycardia was 70 and 79% smaller following Atn and Atr, respectively. However, in the presence of intact vagal outflow (Atn treatment), OT administration restricted to the NTS/DMV still potentiated the bradycardic response to pressor challenges. The gain of reflex bradycardia after OT was significantly higher than Veh treatment into this area (1.03 ± 0.24 vs. 0.57 ± 0.13 beats · min¹ · mmHg¹) and not different from that observed in the control situation (1.22 ± 0.06 beats · min¹ · mmHg¹; Table 4). On the other hand, OT was without effect in Atr-treated rats. After both treatments, OT was unable to change reflex tachycardia.

Injection sites. Microscopic examination of the brain stem revealed the microinjections were directed to the medial part of the NTS and adjacent DMV. The injected dye followed the elongated structure of both the NTS and DMV. The dye-stained areas were mostly confined to the NTS/DMV, with the injections centered near the border between these nuclei; a larger spread out, with an average extension of 475 ± 47 µm, was observed in the rostrocaudal direction. For all brain stems analyzed, the mean point of injection in the anteroposterior plane was located between 150 and 350 µm rostral to the calamus scriptorium, mainly at the level illustrated in Fig. 1B, which we showed to present a very high OT immunoreactivity. In the rats in which injections targeted the fourth ventricle, macroscopic examination immediately after removal of the brain revealed the presence of the dye in the cerebrospinal fluid. Histological examination showed no injection sites in the brain stem of these rats.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present data show that OT plays an important role in modulating baroreceptor reflex control of HR. It appears that brain stem oxytocinergic projections function to maintain a lower HR during pressor challenges. Specifically, our results showed that 1) the NTS/DMV is an important site at which OT modulates reflex control of HR; 2) OT released by oxytocinergic terminals, but not OT present in the cerebrospinal fluid, is the functional modulator of HR; 3) OT facilitates the slowing of the heart by augmenting vagal outflow, thus increasing both the bradycardic response and the gain of the reflex during baroreceptor loading; and 4) OT does not change sympathetic outflow to the heart and is not involved in the modulation of the tachycardic responses to transient pressure decreases. In addition, our data confirmed previous observations on both the high OT content in the dorsal brain stem (14, 23) and the presence of OT terminals specifically in the NTS/DMV area (3, 30, 35).

The involvement of OT in the autonomic control of the circulation has been previously demonstrated (6, 8, 16, 22, 29, 32). There were specific effects of changes in OT turnover in the PVN (16, 22) and of central OT injections on HR control (6, 8, 32). In our study, blockade of OT receptors in a specific area of the solitary vagal complex (the site of dense oxytocinergic projections) attenuated pressor-induced bradycardia, thus showing for the first time that oxytocinergic projections to this area facilitate bradycardic responses. Further evidence for a role of OT was the opposite effect achieved after OT administration into the NTS/DMV. Exogenous OT, through a vagal effect, facilitated the slowing of the heart and increased the magnitude of bradycardia (downward shift of the lower plateau), without changing the reflex gain. Thus endogenous OT provides for the maintenance of both a normal HR range and reflex sensitivity during baroreceptor activation. Our data are supported by the results of Russ and Walker (33), who showed that intravenous OT augmented, whereas a selective OT antagonist completely blocked, the augmentation of reflex bradycardia. In addition, the present data suggest that the NTS/DMV region is a specific site for modulation of baroreceptor reflex control of HR by OT. Previous observations showed that OT administration into the DMV caused a concentration-dependent increase in firing of local neurons (8) as well as a reduction of baseline HR (32). It is important to recognize that OT-containing pathways from the PVN also project to the NA in the ventral brain stem (3, 34, 35). The NA contains many parasympathetic preganglionic cell bodies that project to the heart (7). The functional effects of OT in the NA on HR control, as well as the relative participation of NA and DMV neurons in the modulation of HR function by OT, remain to be determined.

OT is also present in the cerebrospinal fluid, where it exhibits a circadian rhythm, with high concentrations seen during the light period (24). Although it was hypothesized that ventricular OT could modulate cardiac function, in our experiments administration of OT or OT_ant into the cerebrospinal fluid of the fourth ventricle had no effect on baroreflex function. It is true that we used similar doses of OT and OT_ant for both tissue and intraventricular administrations in such a way that an actual administered dose was smaller in the cerebrospinal fluid. However, one should consider that the doses used (in the pmol or ng range) were much higher than brain stem levels of the peptide (in the fmol range) (14) and that blunting of baroreflex control of the heart was described following intracisternal administration of an OT dose similar to ours (27). The reasons for this discrepancy between the findings are not apparent, because both experiments were performed with conscious chronically instrumented rats and used similar peptide doses. However, the observation that changes after intracisternal OT administration were only evident 15 to 30 min later (27) associated with the observation that OT and its 7-9 amino-acid fragment were more effective in changing cardiovascular parameters when injected into the fourth rather than into the lateral ventricle (40) strongly suggests that the cardiovascular site of action is in the brain stem.

The different results (potentiation vs. no change) achieved with the two routes of OT administration (NTS/DMV vs. fourth ventricle) suggest a specific role for the brain stem-released peptide. This is confirmed by our observation that, when acting at the brain stem level, OT increased vagal tone to the heart. The present set of data also fits with previous results from our laboratory, which showed that OT projections to the solitary vagal complex are activated when trained (but not sedentary) rats exercise and that OT released into this area acts to restrain the exercise-induced tachycardia (2). It is well known that trained individuals present a smaller tachycardic response than sedentary ones at similar submaximal exercise intensities (2, 17). In trained rats, the release of OT, while not directly affecting tachycardia, could oppose the tachycardic response by reducing vagal withdrawal during dynamic exercise. Similar to the present study, we also observed that OT injected into the fourth ventricle had no effect on HR control during exercise (2). Taken together, the results using two different paradigms (exercise and baroreflex control) indicate that central oxytocinergic neurons projecting to the NTS/DMV area constitute an important mechanism to modulate HR control, favoring a beneficial slowdown of the heart under situations in which the organism is stressed.

Receptors for OT as well as V₁ are widely distributed in brain stem areas (36, 38). Several studies using receptor autoradiography or in situ hybridization for mRNA established a nonoverlapping nature of brain OT and V₁ receptors in the dorsal brain stem: OT receptors are very dense in the DMV, whereas V₁ receptors are abundant in the NTS (1, 36, 38). Our results are consistent with these findings, because a slowdown of the heart was observed following OT administration into the NTS/DMV (present experiments, 2, 17), whereas facilitation of exercise-induced tachycardia and a smaller bradycardic response were observed with administrations of VP restricted to the NTS (9, 18, 20). Therefore, it seems that the PVN oxytocinergic projections facilitate bradycardic responses by acting on OT receptors mainly located in the NTS/DMV border, whereas vasopressinergic projections act on the NTS V₁ receptors to facilitate tachycardic responses. These opposite findings, showing a balance between excitatory and inhibitory PVN inputs on HR control at the dorsal brain stem, indicate an efficient controlling mechanism. A similar "arrangement" in which these closely related peptides have opposite effects has already been shown (2, 17). The present findings do reinforce the importance of this hypothalamic peptidergic controlling mechanism. The relevance of oxytocinergic modulation in cardiovascular control was also confirmed by two other observations: the high OT content (vs. VP) measured in the dorsal brain stem areas and our recent observation in knockout mice (19) that deletion of the OT gene caused a marked reduction of the bradycardic response during the loading of baroreceptors. The similar effects on baroreflex control with the knockout mice (19) and rats with a blockade of OT receptors restricted to the NTS/DMV (present data) also highlight the important role of OT in the solitary vagal complex to modulate bradycardic responses.

It is important to stress that we gave only unilateral injections. The different stimulation produced by administration of OT or OT_ant on one side is sufficient to change vagal control of the heart during baroreceptor reflex testing, thus providing evidence for the strong modulatory effect of this peptide at the NTS/DMV level. The OT_ant used in the present study is very selective for OT receptors, with a high in vivo anti-OT/anti-V₁ selectivity (5). OT_ant data, depicting the true effect of OT-receptor activation, are very consistent with the results observed following OT administration into the NTS/DMV. It confirmed that the bradycardic effect observed after OT administration into the NTS/DMV was due exclusively to OT-receptor activation and not to both OT and V₁ receptors, as proposed previously (33).

In summary, our data demonstrate that oxytocinergic projections to the solitary vagal complex are functional modulators of cardiovascular control. They act to facilitate the vagal outflow to the heart during pressure challenges, thus maintaining the magnitude and the gain of the baroreceptor-mediated bradycardia. The improvement of the bradycardic response is a specific effect of the neurotransmitter OT, and it is mediated by OT receptors in the NTS/DMV area.

Perspectives