Although it has been suggested that vasopressin (VP) acts within the central nervous system to modulate autonomic cardiovascular controls, the mechanisms involved are not understood. Using nonpeptide, selective V1a, V1b, and V2 antagonists, in conscious rats, we assessed the roles of central VP receptors, under basal conditions, after the central application of exogenous VP, and after immobilization, on cardiovascular short-term variability. Equidistant sampling of blood pressure (BP) and heart rate (HR) at 20 Hz allowed direct spectral analysis in very-low frequency (VLF-BP), low-frequency (LF-BP), and high-frequency (HF-BP) blood pressure domains. The effect of VP antagonists and of exogenous VP on body temperature (Tb) was also investigated. Under basal conditions, V1a antagonist increased HF-BP and Tb, and this was prevented by metamizol. V1b antagonist enhanced HF-BP without affecting Tb, and V2 antagonist increased VLF-BP variability which could be prevented by quinapril. Immobilization increased BP, LF-BP, HF-BP, and HF-HR variability. V1a antagonist prevented BP and HR variability changes induced by immobilization and potentiated tachycardia. V1b antagonist prevented BP but not HR variability changes, whereas V2 antagonist had no effect. Exogenous VP increased systolic arterial pressure (SAP) and HF-SAP variability, and this was prevented by V1a and V1b but not V2 antagonist pretreatment. Our results suggest that, under basal conditions, VP, by stimulation of V1a, V1b, and cognate V2 receptors, buffers BP variability, mostly due to thermoregulation. Immobilization and exogenous VP, by stimulation of V1a or V1b, but not V2 receptors, increases BP variability, revealing cardiorespiratory adjustment to stress and respiratory stimulation, respectively.
- heart rate
- V2 antagonist
vasopressin (vp) is a neurohypophyseal peptide that acts both as neurohormone and central neurotransmitter/modulator. Peripherally, VP exerts a variety of biological effects in mammals: body fluid balance, blood pressure (BP) control, platelet aggregation, ACTH, aldosterone and clothing factor VIII secretion, liver glycogenolysis, and uterine motility. Centrally, VP affects learning, memory, behavior, and autonomic functions such as BP, body temperature (Tb), and respiration. The role of VP is particularly important in stress (14, 25, 27), and VP dysfunction was found to contribute to the pathogenesis of cardiovascular disease (28, 35) and to underlie central diabetes insipidus.
So far, three types of G protein-linked membrane-bound VP receptor subtypes, V1a, V1b, and V2, have been cloned and characterized by their primary structure, gene localization, mRNA distribution, physiological function, and pharmacology (4, 12, 41, 47). V2 receptors have been found only in the periphery; they are positively coupled to adenylcyclase and play a dominant role in the antidiuretic response to VP. V1a and V1b receptors have been found both peripherally and in the central nervous system (CNS), where they mediate phospholipase C activation and intracellular calcium mobilization (4). The role of VP receptors in the mediation of central effects of VP neurotransmission has not been fully elucidated. It is generally assumed that because of their ubiquitous distribution in the brain, V1a receptors mediate all central effects of VP, including autonomic cardiovascular control (30, 32), while V1b receptors are invoked in stress response and control of ACTH release (1, 11). Although V2 receptors have not been found in CNS, results obtained with both peptide- and nonpeptide-selective V2 antagonists suggest V2-like activity in area postrema (AP), a caudal part of the medulla accessible from both sides of the blood-brain barrier (BBB; 40, 48).
The aim of this work was to investigate the role of putative VP receptors in central autonomic cardiovascular control. In conscious freely moving rats, we studied the central effects of novel nonpeptide and selective V1a (SR49059), V1b (SSR149415), and V2 (SR121463) receptor antagonists (12, 41) under basal and stressed conditions and after the application of exogenous VP. Cardiovascular control was evaluated by a well-established method of spectral analysis of BP and heart rate (HR) oscillations, which provides simultaneous and dynamic insight in autonomic control mechanisms (33, 19). Under basal conditions, the frequencies of the BP spectrum are predominantly distributed around very-low frequencies (VLF) and mostly reflect BP changes associated with thermoregulation (15). Among the vasoactive substances, the renin-angiotensin system (RAS) was found to contribute mostly to the genesis of VLF-BP (38). A minor constituent of the BP spectrum is the high frequency (HF) oscillation. It was found to reflect BP changes induced by respiration (9, 18). In contrast, under basal conditions HR spectrum frequencies are clustered around HF that are associated with vagal modulation of the HR with respiration, a phenomenon well known as the respiratory sine arrhythmia (RSA; 9). Only under stressful conditions does a third component of the spectrum of BP and HR appear at low frequencies (LF), depicting increased sympathetic outflow (19, 22, 33). Recently, VP has been found to be one of the main endogenous antipyretic molecules (37, 43, 46). Because thermoregulation may engender hemodynamic changes and affect cardiovascular short-term variability (15), we also investigated the effect of VP antagonists and of exogenously applied VP on Tb.
The experiments were performed in conscious, unrestrained male Wistar outbred rats weighing 320–350 g, chronically instrumented with a right femoral artery catheter and an intracerebroventricular cannula, housed separately in controlled laboratory conditions (temperature −23°C ± 1.2; relative humidity −60–70%; lighting—12:12-h light-dark cycle) with food and tap water ad libitum. All procedures conformed to EEC Directive 86/609 and the School of Medicine University of Belgrade Guidelines on Animal Experimentation, which are in line with guiding principles for research involving animals recognized by the American Physiological Society (2). The number of animals per group was calculated using software “Power and Sample Size Calculations” for paired designs for a given power 90% and type I error probability of 0.05 (8). At the end of the experiment, rats were killed by an overdose of thiopental sodium.
Rats underwent two surgical interventions under anesthesia 7 days before experimentation: implantation of the intracerebroventricular cannula in the lateral ventricle and 5 days later, insertion of the polyethylene catheter in the right femoral artery. The first intervention was performed while the rat was under ketamine xylazine anesthesia [10% (wt/vol) ketamine and 2% (wt/vol) xylazine] so that the rat could be mounted in a stereotaxic frame. The stereotaxic coordinates derived from Paxinos and Watson (36). The guide cannula (G22) protruded in the lateral ventricle 4 mm beneath the skull and was positioned AP = −1.08 mm (from bregma) and LAT = 1.5 mm (from midline), and fixed with dental cement. The skin above was sutured and sprayed with antibiotics (neomycin and bacytracin), and the guide was plugged with a stainless-steel pin. A second surgical procedure was performed under halothane anesthesia in Univentor 400 Anesthesia Unit, TSE systems, Germany [concentration percentage for induction of anesthesia in the chamber was 4% (vol/vol) and 1.7–2.1% (vol/vol) for maintenance of anesthesia under the mask]. The polyethylene catheter (PE; ED-0.95, ID-0.58) was filled with heparinized saline, inserted into the right femoral artery, and tunneled subcutaneously to exit between the scapulas. Postoperatively, the rats received one injection of penicillin (30,000 IU im) and were allowed 48 h to recover from the second surgery before being subjected to further experimental protocols. At the end of the experiments, rats were anesthetized by thiopental sodium. Then, 5 μl of methylene blue was injected intracerebroventricularly; rats in which the blue coloration of the ventricle could not be observed post mortem were excluded from the study.
All experiments were conducted 48 h after the second surgery in quiet surroundings at ambient temperature (23°C ± 2) with the rats housed separately in Plexiglas cages (25 × 25 × 25 cm). Recordings and drug injections were done using catheter extensions to avoid animal stress due to manipulation. The experiments were started around 10 AM every day 1 h after the exteriorized catheters had been connected to matching 30-cm-long extensions. The arterial line was connected to the pressure transducer (Isotec, Harvard Apparatus) for direct measurement of pulsating BP, while a needle (G26) was protruded 0.5 mm below the tip of the guide in the lateral ventricle and connected to a PE extension (ED = 1.09 mm, ID = 0.38 mm), and a 50-μl Hamilton syringe was prefilled with the drug solution. The rats were then subjected to different protocols. Baseline values of systolic arterial blood pressure (SAP), diastolic arterial blood pressure (DAP), and HR were monitored for at least 60 min before experimentation.
Effects of exogenously applied VP on BP, HR, and their short-term variability and Tb.
Vasopressin was applied to conscious rats (n = 6) in increasing doses in a 5-μl volume of pyrogen-free saline in doses of 5, 10, 50, 100, 500, and 1,000 ng icv. Two to three minutes after each intracerebroventricular injection, 20-min-long files were recorded. Between 1 and 1.5 h elapsed between intracerebroventricular injections.
A control group of rats (n = 6) received injections of saline in a volume of 5 μl icv to exclude the nonspecific effects of injected volumes.
The smallest dose of VP that induced changes of cardiovascular short-term variability from a dose-response study was investigated for acute tolerance. The same dose of VP (50 ng icv) was then used to establish functionally selective doses of VP antagonists. For the acute tolerance experiments, four experimental groups were used, six rats in each. The first group of rats received saline followed by two consecutive injections of VP (50 ng icv). The second group of rats was treated with two consecutive saline injections (5 μl icv each time) followed by one injection of VP (50 ng icv). A third group of rats received saline (5 μl icv), VP (50 ng icv), and saline (5 μl icv), while rats in group 4 received three consecutive injections of saline (5 μl icv). The first two intracerebroventricular injections were applied at 1-h intervals, while the third injection was applied at the end of the second recording session. Recording sessions were 20 min long and were started 2 to 3 min after each intracerebroventricular injection.
In a separate group of rats (n = 6), rectal temperature (Harvard Apparatus) was measured 5–10 min after injection of saline (5 μl icv) or VP (50 ng/5 μl icv). On the previous day, rats were trained for rectal temperature measurement to avoid stressful reactions.
Involvement of central vasopressin V1a, V1b, and V2 receptors in the changes of BP, BP short-term variability, and Tb induced by endogenous VP and exogenously applied VP (50 ng icv).
In six experimental groups (6 rats in each) saline (5 μl icv) followed by intracerebroventricular injection of VP-selective nonpeptide V1a or V1b or V2 antagonist in two dose ranges (100 and 500 ng of each antagonist) were followed by 50 ng VP icv injection. The first two intracerebroventricular injections of vehicle and of VP antagonist were applied at 1-h intervals, while the third injection of VP was applied intracerebroventricularly at the end of the recording session with VP antagonist (second recording). Recordings were 20 min long and were started 2 to 3 min after each intracerebroventricular injection.
In a separate group of rats, 50 ng of VP or 100 and 500 ng of V1a or V1b or V2 antagonists were applied. Rectal temperature was measured 10 min after intracerebroventricular injections of 5 μl of vehicle or 50 ng of VP or 100 and 500 ng of V1a or V1b or V2 antagonist.
Association of the changes of BP short-term variability and Tb induced by VP antagonist. Possible contribution of the RAS.
A separate experimental group of rats (n = 6) was designed to investigate whether there is a relationship between the changes in Tb and the cardiovascular changes induced by VP antagonist. Therefore, VP antagonists that concomitantly increased Tb and induced changes in BP variability were applied to rats (n = 6) pretreated with the nonselective cyclooxygenase (COX) inhibitor sodium metamizol (200 mg/kg ip), an antipyretic agent.
To investigate the mechanism by which VP antagonist affects VLF-BP variability, VP antagonists that enhanced VLF-BP oscillation were applied to rats (n = 6) pretreated by quinapril (10 mg/kg ip), an angiotensin-converting enzyme inhibitor (ACEI).
Involvement of central vasopressin V1a, V1b, and V2 receptors in the changes of BP, HR, and their short-term variabilities to acute stress induced by immobilization.
In seven experimental groups (n = 6 rats in each), rats were subjected to the protocol comprising three, 15-min-long recording sessions for baseline, stress, and recovery period. The baseline recording was started while the rat was resting in his home cage (25 × 25 × 25 cm). This was followed by intracerebroventricular injections of either saline (5 μl icv) or V1a, V1b, or V2 antagonists (100 or 500 ng each). Three to five minutes later, rats were covered by an opaque Plexiglas restrainer (6 cm in width by 4 cm in height) and recorded during the stress period. After the restrainer was removed, rats were recorded during recovery.
Nonpeptide and selective V1a (SR49059), V1b (SSR149415), and V2 (SR121463) antagonists were kindly donated by Dr. Claudine Serradeil-Le Gal from Exploratory Research Department of Sanofy-Synthélabo Recherche (Toulouse, France). [Arg8]-vasopressin acetate was purchased from Sigma-Aldrich (UniChem, Belgrade, Serbia), thiopental sodium injections from Rotexmedica (Trittau, Germany), and neomycin bacitracin spray, and penicillin injection from Galenika (Belgrade, Serbia). Halothane was kindly donated by Jugoremedia (Belgrade, Serbia). Ketamine and xylazine were purchased from Richter Pharma (Wels, Austria) and Ceva Santé Animal (Budapest, Hungary), respectively. Metamizol sodium and quinapril chloride were purchased from Hemofarm (Vrsac, Serbia). VP and the V2 antagonist were dissolved in pyrogen-free saline while the V1a and V1b antagonists were dissolved in 10% (vol/vol) and 5% (vol/vol) DMSO, respectively.
Signal processing and spectrum analysis.
Signal processing and spectrum analysis has been described in detail by Japundžić-Žigon (20). Equidistant sampling of SAP, DAP, and HR at 20 Hz allowed direct spectral analysis using fast Fourier transform (FFT) over a 2,048 point time series, or 102.4 s of registration. Thirty overlapping FFTs were then used to compute time spectra over 410 s of the registration period. Prior to FFT calculations, stationarity of the signal was evaluated visually, and data sets were subjected to a 9-point Hanning window and linear trend removal. Spectra were analyzed up to 3 Hz. The sum of the moduli under the volume (modulus vs. time) for 30 FFT segments was calculated for the whole spectrum (TV: 0.00976–3 Hz) and in three frequential domains: VLF (0.00976–0.195 Hz), LF (0.195–0.8 Hz), and HF (0.8–3 Hz). In addition, the position of the peak amplitude of the HF component in HR and in SAP spectra were monitored during experimentation for the estimation of the breathing frequency. Finally, simple statistics (means ± SD) of the distribution of data sets used for time FFT calculations were computed.
Results are expressed as means ± SE. Parametric tests were performed in GraphPad Prism 4 software. One-way ANOVA for repeated measurements followed by a post hoc Bonferroni pairwise comparison was used to assess statistical significance in the same experimental group while statistical significance between experimental groups was assessed by two-way ANOVA for repeated measurements followed by Dunnet’s post hoc test. Statistical significance was considered at P < 0.05.
Effects of exogenously applied vasopressin on BP, HR, and their short-term variability.
Table 1 and Fig. 1, A and B, show that VP injected intracerebroventricularly at 50 ng and at higher doses increases SAP. At 5, 50, and 100 ng, VP increases HF-SAP and HF-DAP variability, whereas only 50 ng of VP increases HR and 1,000 ng of VP increases DAP. The maximal response of SAP to 50 ng of VP occurred 3–5 min after intracerebroventricular injection, and it lasted up to 20 min. Fig. 1, C and D, and Table 2 show that the effect of 50 ng of VP on SAP, HF-SAP, and HF-DAP components is consistent and that no acute tolerance could be observed. However, the tachycardic effect induced by VP in a dose-response study was not consistent after a single injection of 50 ng icv of VP, and bradycardia, or a biphasic response could be observed. Intracerebroventricular injection of 50 ng of VP induced no changes in the HR spectra (not shown).
Involvement of central vasopressin V1a, V1b, and V2 receptors in the BP and BP short-term variability induced by endogenous VP and exogenously applied vasopressin (50 ng icv).
V1a, V1b, and V2 antagonists applied alone did not alter the level of BP and HR (table 3). However, 500 ng of V1a antagonist increased HF-SAP and 100 and 500 ng of the V1b antagonist increased HF-SAP and HF-DAP (Fig. 2, A and B). In addition, 500 ng of the V2 antagonist applied icv increased VLF SAP and VLF-DAP variability (Fig. 2, C and D).
Table 4 shows that 100 and 500 ng of the V1a antagonist and 500 ng of the V1b antagonist prevented the VP-induced SAP increase. In addition, 100 ng of the V1a antagonist and 500 ng of the V1b antagonist prevented VP-induced HF-SAP and HF-DAP increase (Fig. 2, E and F).
Table 5 shows that only the V1a antagonist injected intracerebroventricularly into freely moving rats, at the two different doses, was able to significantly increase Tb.
Association of Tb and HF-SAP increase induced by the V1a antagonist.
The V1a antagonist applied intracerebroventricularly to rats in a dose of 500 ng significantly increased Tb (Table 5) and concomitantly increased the HF-SAP oscillation (Fig. 2A). However, when metamizol (200 mg/kg ip), a nonselective COX inhibitor, was applied to rats before the V1a antagonist (500 ng icv), it prevented the rise in Tb (36.6°C ± 0.2; P > 0.05) and the increase in the HF-SAP oscillation (Figs. 3 and 4). Alone, metamizol neither affected basal cardiovascular parameters (not shown) nor Tb (36.2 ± 0.2°C, P > 0.05).
Involvement of the RAS in the changes of VLF-BP variability induced by the V2 antagonist.
The V2 antagonist applied alone at a dose of 500 ng icv significantly increased the VLF-SAP (Fig. 2C) and the VLF-DAP oscillation (Fig. 2D). To investigate whether this effect involves the RAS, we applied the V2 antagonist to rats pretreated with quinapril, an ACEI. Quinapril (10 mg/kg ip) applied alone to rats half an hour before V2 antagonist, reduced SAP (from 124 mmHg ± 8 to 103 mmHg ± 6, P < 0.01) and increased HR (from 397 bpm ± 13 to 456 bpm ± 14, P < 0.01) and HF-SAP variability (294 mmHg/Hz−1/2 ± 46 to 637 mmHg/Hz−1/2 ± 51, P < 0.01). The V2 antagonist (500 ng icv) administered after quinapril did not modify significantly quinapril’s effects on SAP (114 mmHg ± 7, P < 0.05), HR (446 bpm ± 18, P < 0.05), and HF-SAP (436 mmHg/Hz−1/2 ± 48, P < 0.05), and it failed to increase the VLF-BP oscillation (Figs. 5 and 6).
Neither quinapril alone, nor quinapril plus the V2 antagonist, affected Tb (control Tb − 36.8°C ± 0.2; after quinapril −36.5°C ± 0.4; in rats treated with quinapril plus V2 antagonist, Tb − 36.5°C ± 0.3).
Involvement of central vasopressin V1a, V1b, and V2 receptors in the changes of BP, HR, and their short-term variability in response to acute stress induced by immobilization.
Acute immobilization induced a significant rise in SAP (Fig. 7) and DAP, while the increase of HR did not reach statistical significance (⇓Fig. 9). In BP spectra, a significant increase of both LF-SAP and HF-SAP (Figs. 7 and 8) and LF-DAP and HF-DAP oscillations was noted, whereas in the HR spectrum, an increase in HF-HR was observed (Fig. 9).
In rats pretreated with 100 ng (icv) of the V1a antagonist before immobilization, the increase of LF-SAP (Fig. 7, left) and LF-DAP variability was prevented, while 500 ng of the V1a antagonist prevented the increase of BP, and both LF-SAP, HF-SAP (Fig. 7, right), LF-DAP, and HF-DAP short-term variability. The V1a antagonist-pretreated rats exhibited a significant increase in HR but no change in the HF-HR, following restraint (Fig. 9).
In rats pretreated with 100 and 500 ng icv of the V1b antagonist, immobilization induced no significant changes in BP spectra, while only the higher dose of the V1b antagonist prevented the BP increase induced by restraint (Figs. 7 and 8). The V1b antagonist did not modify the response of the HR to stress (Fig. 9).
The main findings of this study are that in freely moving conscious rats under basal conditions, endogenous VP, by the stimulation of central V1a, V1b, and cognate V2 receptors buffers BP variability in the HF and VLF BP frequency domain, respectively. We also found that V1a and V1b receptors mediate the increase of HF-SAP variability by intracerebroventricularly applied VP and that immobilization stress enhanced BP variability in HF-BP and LF-BP frequency domains involving central V1a and V1b receptors. These data suggest that more than one central structure expressing VP receptors can contribute to BP short-term variability. As VP is a neuropeptide that acts both as a neurotransmitter and a neuromodulator, as well as neurohormone, it can produce patterned rather than simple physiological response (25), a view supported by the morphological finding of widespread distribution of central V1a and V1b receptors in the brain (13, 32).
It is well known that VP plays an important role in the regulation of arterial BP and body fluid balance. Although its contribution to BP maintenance is not evident under basal conditions and it is minor in essential hypertension, it was found to play a key role in the autonomic adjustment of the cardiovascular system to acute hypovolemic stress (27). When “exogenous” VP is injected peripherally it increases BP by the stimulation of V1a receptors located in blood vessels (26). However, in normovolemic animals, the increase of BP is less than expected for a given dose due to a VP-mediated increase in baroreceptor reflex (BRR) sensitivity and the subsequent decrease in HR and cardiac output (3, 6, 16, 26), an effect dependent on the integrity of AP (6, 40, 48). In contrast, centrally administered VP increases BP as a result of sympathoadrenal activation and the stimulation of V1a receptors in the rostral part of the ventrolateral medulla (RVLM), which increases vasomotor drive to the peripheral blood vessels (30). In our experiments, VP antagonists did not affect BP under basal conditions, confirming a minor role of VP in BP maintenance under basal conditions. However, exogenously applied VP and stress significantly increased BP, and this effect could be inhibited by higher doses of the V1a and also V1b, but not by the V2 antagonist pretreatment, in line with a previous report (30). Centrally administered VP does not always increase HR, and even bradycardia was reported (7, 10). In our experiments, the effect of intracerebroventricularly applied VP on HR was inconsistent; all types of HR response occurred, including tachycardia, bradycardia, and a biphasic one. VP injected in the lateral ventricle reaches adjacent areas where it can act differentially to modulate HR. Microinjections of V1 agonist in the AP mimic the effect of peripheral VP and enhance BRR sensitivity and cardiac vagal tone (6), while microinjections of VP in nucleus tractus solitarius (NTS) increases HR and decreases BRR sensitivity (23). In rats submitted to immobilization, there was no change in HR, but the HF, respiratory oscillation of the HR did increase, suggestive of an increased vagal influence on the heart. However, in rats pretreated with the V1a antagonist, HR increased significantly, and the HF-HR increase was diminished. The HF-HR oscillation reflects the respiratory modulation of the cardiac vagal tone or the RSA, and it has been found to be the most sensitive marker of cardiac vagal activity; it can be abolished by atropine or vagotomy, and its diminution in heart disease has been found to be a bad prognostic sign (18, 33). The effect of VP on the HF-HR in our experiments reveals that central V1a mechanisms are implicated in the increase of cardiac vagal tone during stress, and this may be important in protecting the heart in extreme situations.
The role of VP in central autonomic controls under basal conditions is not yet fully understood. Recently, VP has been invoked as one of the main endogenous antipyretic molecules (5, 34, 39, 46), and evidence has been presented that VP plays a tonic role in thermoregulation, although it is uncertain which receptors are involved (46). Our data support a role for VP in thermoregulation under basal conditions, as blockade of V1a receptors significantly increased Tb in freely moving rats. In addition, application of the V2 antagonist intracerebroventricularly to freely moving rats increased the VLF-BP oscillation, which has been shown to reflect changes in cutaneous flow associated with thermoregulation (15, 33). VP acts centrally and peripherally to reduce Tb. The ventral septal area (5) and the preoptic hypothalamus (29) were found to be the central sites of action of VP in thermoregulation, whereas in the AP, baroreflexive suppression of nonshivering thermogenesis (31, 42, 43, 45, 46) and cutaneous (tail) vasodilatation (45) were invoked in VP-induced peripheral thermoregulation. We have previously reported that the V2 antagonist (OPC-31260) applied intravenously increases VLF-BP oscillation under basal conditions in freely moving rats (20, 21). Our present results with SR121463, another nonpeptide V2 antagonist applied intracerebroventricularly in a dose of 500 ng, confirm the previous finding and suggest that V2 antagonist acts most probably at AP, a structure accessible from both sides of the BBB, to modulate VLF-BP oscillation. In addition, the V2 antagonist was found to reduce BRR sensitivity in conscious rats in the AP (40). We, therefore, suggest that the V2 antagonist acts in the AP to blunt BRR sensitivity and debuffer VLF oscillation, which reflects the adjustment of cutaneous blood flow with thermoregulation. Our results also show that the effects of V2 antagonist on VLF-BP oscillation can be prevented by quinapril, an ACEI. It has been well documented that the RAS contributes to VLF-BP oscillation genesis (38). Moreover, ANG II was found to participate in basal thermoregulation both centrally and peripherally (49) and to interact with VP (44). Our results with quinapril confirm the importance of VP and RAS interaction in the modulation of VLF-BP oscillation under basal conditions. However, the V2 antagonist in our experiments did not affect Tb in freely moving rats, suggesting a minor role of V2 receptors in Tb modulation under basal conditions, unless other compensatory mechanisms were engendered to prevent the change in overall Tb.
Another, secondary mechanism that can promote an alteration in Tb is ventilation. Concomitant increase of Tb and HF-SAP by intracerebroventricularly applied V1a antagonist suggests association. There are several brain structures expressing VP receptors involved in the modulation of respiration. However, convincing evidence about the contribution of VP in the modulation of the respiration under basal conditions has not been provided. The HF-BP oscillation reflects respiratory perturbations of the hemodynamics of the great thoracic vessels and the heart (9, 18). It is therefore strongly influenced by the respiratory pattern (depth and frequency of breathing) on one side, and central volemia and hemodynamics (venous return, stroke volume, and cardiac output) on the other. In our experiments, the HF increases evoked by V1a and V1b antagonists suggest an increase in the depth of breathing (tidal volume increase), since the breathing frequency, as judged by the position of the HF peak in BP and HR spectra, was in eupnoic range throughout the experiments (1.2–1.4 Hz), and animals were well hydrated during experimentation. Our results suggest that the HF-SAP increase by V1a antagonist could occur as a compensatory response to the increase of Tb. This view is substantiated by the finding that metamizol, an antipyretic agent, prevented the increase of both the Tb and the HF-SAP oscillation induced by the V1a antagonist. However, the differential effect of V1a and V1b antagonist on Tb implies that other central mechanisms unrelated to thermoregulation are also involved in HF-BP increase.
An interesting finding of the present study was that intracerebroventricularly applied VP increased the HF-SAP oscillation, although it did not alter Tb. Moreover, this effect was prevented by V1a or V1b antagonists. In our experiments, the effects of VP antagonists on BP variability reveal the effect of very small, basal, concentrations of endogenous VP present in neuronal synapses subserving classical, synaptic or “wire” neurotransmission. However, challenged by stress, VP is also being released from neuronal somata and dendrites of the magnocellular neurons in the hypothalamic paraventricular nucleus (25), wherefrom it can reach remote receptor sites in the brain, subserving nonsynaptic or “volume” transmission. Such variable and multiple modes of neurotransmission by VP provide the basis for involvement of VP in the coordination of functions in a patterned response to stress, rather than the modulation of only one physiological function. Therefore, it is not surprising that basal concentrations of VP in the brain and VP challenged by stress or injected intracerebroventricularly exert opposite effects on one physiological function, such as breathing, which underlies HF-BP variability. Present results imply that exogenously applied VP, as well as endogenously released VP by acute stress, exerts a stimulant effect on respiration involving both V1b and V1a receptors. This is supported by morphological identification of V1a receptors in the pre-Bötzinger complex and the phrenic nuclei, as well as in the circum ventricular organs, where they were found to stimulate, and not to inhibit respiration (24). For instance, microinjections of VP in the pre-Bötzinger complex have been reported to stimulate respiration and to involve the V1a receptor (24). Similar effects on respiration were obtained with microinjections of l-glutamate into the paraventricular nucleus (PVN; 50) that increases the release of endogenous VP and accentuates its effects on respiration. Although there is no morphological evidence regarding the localization of V1b receptors in the brain stem areas associated with cardiorespiratory regulation, immunohistochemistry studies indicate that V1b receptors are located in the lateral septum, where infusion of a V1b antagonist was found to reduce the emotional reaction to stress, fear, and anxiety (11). Therefore, it is logical to assume that the V1b antagonist in our experiments, by alleviating the emotional reaction to stress, pacified the inspiratory activity and prevented the increases of the HF-BP oscillation. Functionally, the HF-BP increase in stress is important because it depicts increased blood oxygenation and better aspiration of blood by the lungs toward the thorax, thus improving heart filling and circulation.
Immobilization also increased the LF component of BP variability, and this effect could be reduced by pretreatment of rats by V1a or V1b antagonists. The LF oscillation or the Mayer wave of BP has been shown to reflect the sympathetic drive to blood vessels (17, 18, 22). This can be attenuated by peripheral blockade of α-adrenergic receptors, as well as sinoaortic desafferentation, and has been shown to grow in amplitude during stress and in early stage of stress-related disorders (17, 18, 22, 33). Hence LF-BP oscillation was proposed to be a very sensitive marker of the sympathetic drive to blood vessels (19, 33). It is well known that VP is released from the hypothalamic PVN during stress to act peripherally in the stimulation of ACTH release and centrally to modulate the emotional reaction to the stress (11, 25, 41). Our results now provide pharmacological evidence that central V1a and V1b receptors also contribute to the changes in BP and LF-BP induced by acute immobilization stress. This is supported by the finding of cellular localization of V1a receptor mRNA in the RVLM of rats (32), a well-known integrative site that receives inputs both from the periphery and the higher brain structures and sets the level of efferent vasomotor drive. Nevertheless, we cannot exclude the possibility that V1a receptors, and particularly V1b receptors located in higher brain structures and affecting emotions and behavior, contribute to LF-BP increase by immobilization.
In conclusion, our results emphasize the role of central V1a, V1b, and cognate V2 receptors in buffering HF-BP and VLF-BP variability under basal conditions and suggest that they are, at least in part, due to thermoregulation. Moreover, the results point out the role of central V1a and V1b receptors in the genesis of LF-BP and HF-BP variability during cardiorespiratory adjustment to stress, as well as in the mediation of HF-SAP variability increase by intracerebroventricularly applied VP.
We greatly value the generous support of the Royal Society (UK) and the Serbian Ministry of Science and Technology.
The authors thank Dr. Claudine Serradeil Le Gal from Sanofy Synthélabo research labs (Toulouse, France) for providing us with the selective nonpeptide vasopressin antagonists. We also thank Dr. Vesna Starčević, from the Department of Physiology School of Medicine, University of Belgrade, for the initial help in intracerebroventricular cannulation techniques in rats and the skillful technical assistance of Verica Savić.
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