INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

DURING HEMORRHAGE, a complex array of autonomic and hormonal responses ensues which helps to maintain blood pressure during diminished blood volume. Initial responses are characterized by a sympathoexcitatory phase that effectively shunts blood away from the periphery to preserve perfusion of vital organs. However, if blood loss is severe (>20-30% of blood volume), paradoxical sympathoinhibitory and bradycardic responses result, causing a sudden drop in blood pressure. In cats, it was found that the bradycardia initiated by hypotensive hemorrhage coincided with a burst of activity of vagal afferents arising from the ventricles (18). This finding has led to the hypothesis that vigorous contraction of the empty cardiac chambers during hypotensive hemorrhage stimulates cardiac nonmyelinated vagal mechanoreceptors, leading to the reflex sympathoinhibition and parasympathetic activation characteristic of the decompensatory phase of hemorrhage.

Stimulation of cardiopulmonary chemosensitive receptors produces a pattern of sympathetic responses similar to hemorrhage. In particular, activation of serotonergic 5-HT₃ receptors via intrapericardial administration of the specific 5-HT₃-receptor agonist, phenylbiguanide (PBG), causes bradycardia, hypotension, and renal sympathoinhibition concomitant with adrenal sympathostimulation (13). The cardiodepressor and sympathetic effects of PBG are abolished with bilateral cervical vagotomy but not muscarinic receptor blockade, indicating that the response is mediated by cardiac vagal afferent nerves (22, 24). Though the bradycardic and hypotensive responses to PBG subside quickly, it was recently found that sympathoinhibition, specific to sympathetic nerves innervating the kidney, persists during intravenous infusion of PBG. This led to speculation that hemorrhage might cause the release of endogenous serotonin that could mediate renal sympathoinhibition via cardiopulmonary vagal afferent 5-HT₃-receptor stimulation (22).

Additional serotonergic-receptor subtypes have been implicated in the sympathoinhibitory response to hypotensive hemorrhage. Intravenous administration of the 5-HT₁/5-HT₂-receptor antagonist, methysergide, facilitates restoration of mean arterial pressure (MAP) and heart rate (HR) after hemorrhage in cats (7). Systemic administration of methysergide or inhibitors of serotonin biosynthesis has also been found to abolish the sympathoinhibitory response to hemorrhage in rats (17). Given that such drugs easily cross the blood-brain barrier, it is still not clear whether they mediate their effects on hemorrhage through peripheral or central receptors or through a combination of both. Central administration of serotonergic-receptor antagonists has also been shown to block the reflex hypotension and bradycardia induced by stimulation of cardiopulmonary chemosensitive receptors with PBG (3). Therefore, the following set of experiments were designed to test: 1) whether peripheral 5-HT₃ receptors are involved in the renal sympathoinhibitory response to hemorrhage in rats, and/or 2) whether the same central serotonergic receptors mediate the renal sympathoinhibitory response to hypotensive hemorrhage and to PBG infusion.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Animals

Drugs

From pharmacological studies in the rabbit, it has been speculated that the serotonergic-receptor subtype that mediates the cardiodepressor response to hypotensive hemorrhage is of either the 5-HT₁ or 5-HT₂ type (9). Therefore, methysergide was chosen for use in these studies because of its high potency for both 5-HT₁ and 5-HT₂ serotonergic receptors and relatively low agonist or antagonist activity on dopaminergic, adrenergic, and histamine receptors (2, 12, 14, 19). The high potency of methysergide was desirable particularly in these studies, since only a small quantity of drug could be dissolved in the low injection volume necessary for central drug administration. Although using methysergide precluded determination of the specific serotonergic subtype mediating the response to hemorrhage, methysergide has been shown to be 50- to 100-fold more potent at displacing 5-HT₂-labeled sites than 5-HT₁-labeled sites, making it somewhat selective for 5-HT₂ receptors (12). In contrast, other antagonists, such as ketanserin or metergoline that are more specific for particular serotonergic receptors, show high affinity for other nonserotonergic-receptor subtypes and were therefore not deemed suitable for these studies (2, 5).

Surgery

Twenty-four hours before the experiment, the rats were reanesthetized (Equithesin, 3 ml/kg ip) for catheter and renal sympathetic nerve recording electrode placement. Animals subjected to the hemorrhage protocol were implanted with a single venous femoral catheter and bilateral femoral arterial polyethylene catheters (PE-50 heat-welded to a length of PE-10) for drug infusion, direct MAP measurement, and arterial blood withdrawal, respectively. The catheters were tunneled subcutaneously to exit at the nape of the neck, and the incisions were sutured closed. All subjects were subsequently instrumented with stainless steel, bipolar recording electrodes.

The electrode was formed from two lengths of single-stranded, Teflon-coated, stainless steel wire (bare diameter = 0.005 in., A-M Systems, Everett, WA) soldered to microconnectors. The leads were isolated within a length of Silastic tubing extending to the microconnector, and the assembly was glued together at the microconnector end with epoxy. Electrodes were implanted through a left flank incision. Microconnector assemblies were externalized along with the vascular catheters at the nape of the neck. Abdominal and back muscles were retracted, and a small (1-2 mm) length of the renal sympathetic nerve extending from the aorticorenal sympathetic ganglion was isolated and placed on hooks formed at the end of the two electrode leads. The nerve and hooks were embedded in a light-weight dental silicon (Bisico, Bielefeld, Germany). The flank incision was sutured closed in two layers with the electrode leads coiled within the subcutaneous space. Rats were allowed to recover in their home cage for 24 h.

In additional groups of animals, the sinoaortic baroreceptors were denervated. Animals were anesthetized with Equithesin (3 ml/kg ip). After anesthesia, a midline incision in the ventral neck region was made, and the muscles were retracted to expose the carotid sinus. The aortic depressor, superior laryngeal, and carotid sinus nerves were bisected bilaterally. The animals were allowed to recover for 3 wk after which they were instrumented as described above. All sinoaortic-denervated (SAD) animals were tested for complete absence of arterial baroreceptor responses by assessing the HR response to an intravenous injection of nitroprusside (30 µg/kg). Only those animals demonstrating a complete lack of tachycardic response to nitroprusside were used in the study.

Animals included in the PBG infusion protocol received intracerebroventricular cannula, bipolar recording electrodes, and unilateral femoral arterial and venous catheters as described above.

Data Acquisition

Analog blood pressure, HR, and RSNA output from the amplified sources were sampled by a computer using an analog-to-digital conversion card (Data Translation, DT2812, Marlboro, MA). The signals were sampled at 100 Hz and the values averaged over 200 ms. The averaged values were collected using LabTech Notebook software (version 7.1.1, Laboratory Technologies, Wilmington, MA).

Experimental Protocols

In initial experiments, studies were performed to determine the dose of methysergide necessary to prevent the bradycardic response to hemorrhage. Doses of 5, 10, 20, and 40 µg were tested in one to two animals. Subsequent tests were performed using the 40-µg dose in larger numbers of animals to provide group data.

In additional experiments, the same protocol was followed except that the methysergide (40 µg/5 µl) or vehicle was injected intravenously rather than intracerebroventricularly to determine whether the observed effects were due to central or peripheral effects of the drug. These rats were not implanted with intracerebroventricular cannulas.

Assessment of arterial baroreceptor involvement in the effect of methysergide on hemorrhage. In this study, SAD animals were prepared as described in the surgical protocol above. They were subjected to the same hemorrhage protocol as described above except they were not instrumented with renal sympathetic nerve electrodes. Because of substantial sympathetic stimulation produced by surgery in baroreceptor-denervated animals, the SAD rats were allowed 3 days to recover from the vascular catheter implantation procedure to minimize the effect of surgery on the response to hemorrhage. This length of recovery precluded the measurement of RSNA due to the short-lived viability of functional renal nerve preparations. In this study, a 20-µg dose of methysergide was used rather than 40 µg to avoid the depressor and bradycardic effects of the higher dose of methysergide. This ensured that group differences in response to hemorrhage were not due to differences in the absolute drop in MAP.

Assessment of the effects of peripheral 5-HT₃-receptor blockade on the cardiovascular and sympathetic responses to hemorrhage. Animals were instrumented with vascular catheters and a nerve recording electrode as described above for the hemorrhage protocol. Twenty-four hours after surgery, the animals were connected with blood pressure and RSNA measurement devices and allowed to habituate for 2 h. The animals were given a 5-µg bolus intravenous injection of PBG, after which they were given 50 µg of the 5-HT₃-receptor antagonist, ICS-205,930 methiodide, dissolved in 200 µl of 20% HBC. Five minutes later, responses to a second 5-µg injection of PBG were measured to ensure full blockade of 5-HT₃ receptors. After an additional 10 min, a 15-min baseline measurement was taken followed by hemorrhage executed in the same manner as described previously. At the conclusion of each experiment, responses to 5 µg iv PBG were again tested to ensure that 5-HT₃-receptor blockade persisted throughout the experiment.

Assessment of the role of central 5-HT₁/5-HT₂ receptors in cardiovascular and sympathetic responses to peripheral 5-HT₃-receptor stimulation. Cardiovascular and sympathetic parameters were recorded as described above. The venous catheter was connected with PE-50 tubing containing PBG. After the animals had 2 h to habituate to the instrumentation, a 15-min baseline recording was taken after which a 40 µg/5 µl intracerebroventricular injection of either methysergide or vehicle was made over 1 min. After an additional 15 min, a 10-min intravenous infusion (25 µg · kg¹ · min¹) of PBG was begun. The 25-µg · kg¹ · min¹ dose of PBG has been shown to produce a partial blockade of RSNA (22). After the experiment, proper intracerebroventricular cannula placement was verified as described above.

Data Analysis

For the hemorrhage studies, MAP, HR, and RSNA were averaged into 1-min intervals. For PBG infusion studies, data were averaged into 1-min intervals with the exception of data collected 1 min before through 1 min after commencement of the PBG infusion. During this period, data were averaged over 10-s intervals to demonstrate the transient cardiovascular and sympathetic responses. The MAP, HR, and percent baseline RSNA responses were compared in a two-way repeated measures analysis of variance (ANOVA) with drug treatment used as a between-groups factor, and time assessed as the within-group repeated measure. Significant interactions were followed by repeated measures analysis over time within each group to determine whether treatments had a significant effect. Significant differences between and within treatment groups were determined with Bonferroni-adjusted t-tests. Results were considered significant when P < 0.05.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Figure 1 demonstrates typical changes in MAP, HR, and RSNA of individual rats given intracerebroventricular injection of either vehicle (Fig. 1A) or methysergide (40 µg/5 µl, Fig. 1B) at time 0 (as indicated by the arrows), followed 15 min later by a blood withdrawal gauged to lower blood pressure to 60 mmHg for 5 min. Figure 2 shows the group cardiovascular and renal sympathetic data from rats given intracerebroventricular injections of either vehicle or methysergide starting at time 0 (1st arrow) followed 15 min later by hemorrhage (2nd arrow). Analyses of the effects of methysergide on baseline parameters (i.e., prehemorrhage data from 5 min before intracerebroventricular injection through 15 min after intracerebroventricular injection) showed that methysergide produced a significant bradycardic response accompanied by a slight, but variable hypotensive response. The methysergide-induced hypotension was significant in rats given subsequent hemorrhage but not in those given the PBG infusion (see Fig. 7). In both studies, methysergide caused variable alterations in RSNA that tended to be sympathoexcitatory but did not differ significantly from control animals. The intracerebroventricular injection of HBC vehicle had no significant effect on baseline levels of the tested parameter.

View larger version (29K): [in this window] [in a new window]   Fig. 1.   Mean arterial pressure (MAP), heart rate (HR), and renal sympathetic nerve activity (RSNA) in individual rats given either intracerebroventricular vehicle (A) or 40 µg/5 µl intracerebroventricular methysergide (B) at time 0 followed 15 min later by hypotensive hemorrhage gauged to decrease blood pressure to 60 mmHg for 5 min. Data were averaged over 200 ms and smoothed over 20 consecutive data points to provide clearer representation of RSNA data.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Responses of MAP, HR, and RSNA to intracerebroventricular injection of 5 µl vehicle (20% 2-hydroxypropyl--cyclodextrin vehicle) or 40 µg/5 µl methysergide in conscious, unrestrained rats starting at time 0 (1st arrow) followed by hemorrhage 15 min later (2nd arrow). Values are group means ± SE. Data represent 1-min averages. Number of rats given in parentheses. Comparisons were performed as follows. 1) Between treatment groups at each interval; Bonferroni-adjusted unpaired t-test, * P < 0.05;  P < 0.01. 2) Between preintracerebroventricular injection baseline and each postintracerebroventricular injection interval up until the start of hemorrhage; Bonferroni-adjusted paired t-test for the control ( P < 0.05) and methysergide-treated group (§ P < 0.05). 3) Between prehemorrhage baseline and each interval during hemorrhage; Bonferroni-adjusted paired t-test for control ( P < 0.05) and methysergide-treated rats (§ P < 0.05).

During hemorrhage, sympathetic activity rose significantly in both groups. Although there was a tendency for a larger rise in sympathetic activity during hemorrhage in the methysergide-treated group, the maximal change between groups was not statistically significant, probably due to the larger variability in the methysergide-treated group. After ~2 min of constant blood withdrawal, the control group showed a rapid decline in MAP that coincided with bradycardia and a fall in sympathetic activity that remained below baseline throughout the hemorrhage. In contrast, methysergide-treated rats showed a delayed hypotensive response that was accompanied by a slight but nonsignificant tachycardia. The onset in the decline of RSNA was also delayed by methysergide. However, RSNA remained above baseline levels throughout the 5-min hemorrhage. It should be noted that withdrawal of blood continued at a constant rate until blood pressure dropped to 60 mmHg in both groups. As shown in Fig. 2, approximately 4 additional minutes of constant blood withdrawal were needed before blood pressure of the methysergide-treated group fell to 60 mmHg, indicating that a larger blood loss was necessary to initiate the cardiodepressor response to hemorrhage compared with controls. Despite the larger blood loss, RSNA in methysergide-treated animals did not fall below baseline levels throughout the 5-min hemorrhage. As shown in Fig. 1B, the level of RSNA remained relatively steady at levels slightly above the original baseline throughout the 5 min of hypotension in the methysergide-treated rats. This was a consistent finding for the methysergide-treated group as a whole. The hemorrhage was not continued beyond 5 min because of the difficulty in maintaining hypotension in the methysergide-treated group. This required additional blood withdrawal, which became increasingly difficult, probably because of the already excessive blood loss and vasoconstriction in this group.

Figure 3 demonstrates the MAP and HR responses of SAD rats subjected to hemorrhage after 20 µg of intracerebroventricular methysergide or vehicle. At this dose, methysergide had no significant effect on either MAP or HR before hemorrhage. Hemorrhage caused an abrupt fall in MAP and HR in SAD animals injected with vehicle. As with baroreceptor-intact animals, methysergide treatment in SAD rats delayed the blood pressure fall during hemorrhage, as indicated by the longer constant-rate blood withdrawal necessary to elicit a significant hypotensive response. The 20-µg dose of methysergide both delayed and attenuated the bradycardic response to hemorrhage as indicated by the longer duration of hemorrhage necessary to significantly lower HR as well as the persistence of a significant between-group difference in HR despite the similar magnitude fall in blood pressure.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   MAP and HR responses to hemorrhage in chronic sinoaortic-denervated rats given prior intracerebroventricular injection with either vehicle or methysergide (n = 6). Hemorrhage commenced at time 0, indicated by arrow. Values are groups means ± SE. Data were averaged over 1-min sampling periods. Number of rats given in parentheses. Comparisons were performed as follows. 1) Between treatment groups at each interval; Bonferroni-adjusted unpaired t-test, * P < 0.05;  P < 0.01. 2) Between prehemorrhage baseline and each interval during hemorrhage; Bonferroni-adjusted paired t-test for control ( P < 0.05) and methysergide-treated rats (§ P < 0.05).

In experiments undertaken to test the effects of 5-HT₃-receptor blockade on hemorrhage, all subjects were tested for efficacy of receptor blockade before hemorrhage. Figure 4 demonstrates a typical response in an individual rat receiving 5 µg iv PBG before (Fig. 4A) and after (Fig. 4C) 5-HT₃-receptor blockade with 50 µg iv ICS-205,930 methiodide (Fig. 4B). After hemorrhage, rats were again tested for responsiveness to PBG to ensure that blockade persisted throughout the hemorrhage. Administration of ICS-205,930 methiodide was effective in blocking PBG responses after hemorrhage in all rats subjected to the protocol (data not shown).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Demonstrates adequacy of 5-HT3-receptor blockade before hemorrhage. Shown are MAP, HR, and RSNA responses of an individual conscious, unrestrained rat during 5 µg iv phenylbiguanide (PBG) (A), 2 min later during intravenous administration of 50 µg of 5-HT3-receptor antagonist, ICS-205,930 methiodide (B), and 5 min later during 5 µg iv PBG (C).

Figure 5 demonstrates the cardiovascular and sympathetic responses to hemorrhage in groups of animals given 40 µg methysergide, 5 µl vehicle, or 50 µg of the 5-HT₃-receptor antagonist, ICS-205,930, directly into the systemic circulation. There were no differences observed among the groups in any of the parameters tested. Therefore, within-group analyses were not performed.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   MAP, HR, and RSNA responses to hemorrhage in conscious, unrestrained rats 15 min after intravenous administration of 5 µl of 20% hydroxypropyl--cyclodextrin vehicle dissolved in 200 µl saline (control), 40 µg methysergide, or 50 µg iv of the 5-HT3-receptor antagonist, ICS-205,930 methiodide. Values are group means ± SE. Data represent 1-min averages. Number of animals given in parenthesis.

Figure 6 shows representative examples of the MAP, HR, and RSNA responses to a 10-min, 25 µl · kg¹ · min¹ iv PBG infusion in individual rats given either 1) 5 µl icv HBC vehicle, or 2) 40 µg/5 µl icv methysergide. As shown in Fig. 6, the initial response to the PPG infusion consisted of a large bradycardic response concurrent with hypotension and sympathostimulation that lasted ~1 min. After the cardiodepressor response subsided, a strong inhibition of renal sympathetic activity occurred. The sympathetic activity typically showed a slow recovery to baseline throughout the continued infusion in control animals. As shown in Fig. 6B, earlier treatment with methysergide prevented the recovery of activity during PBG infusion. However, after the infusion was terminated, nerve activity recovered quickly, demonstrating the continued viability of the nerve in both methysergide- and vehicle-treated rats.

View larger version (22K): [in this window] [in a new window]   Fig. 6.   MAP, HR, and RSNA responses to a 10-min, 25 µg · kg1 · min1 iv infusion of PBG starting at time 0. Individual examples are shown for separate rats treated 15 min before PBG infusion with either 5 µg intracerebroventricular hydroxypropyl--cyclodextrin vehicle (A) or 40 µg/5 µl methysergide (B). Data were averaged into 200-ms blocks and smoothed over 20 consecutive data points to provide clearer demonstration of RSNA responses.

Figure 7 shows MAP, HR, and RSNA responses to the 10-min, 25 µl · kg¹ · min¹ iv PBG infusion in groups of rats given intracerebroventricular methysergide or vehicle 15 min before PBG infusion. There was no difference in MAP, HR, or sympathostimulatory responses to infusion. However, the sympathoinhibitory response took longer to recover in methysergide-treated rats as indicated by the longer duration of the sympathoinhibitory effect of PBG in treated vs. nontreated rats.

View larger version (25K): [in this window] [in a new window]   Fig. 7.   Group data depicting MAP, HR, and RSNA responses to 10-min, 25 µg · kg1 · min1 infusions of PBG in conscious, unrestrained rats treated with intracerebroventricular injections of vehicle or 40 µg methysergide 15 min before infusion. Data are represented as group means ± SE. Number of rats are given in parentheses. Comparisons were performed as follows. 1) Between treatment groups at each interval; Bonferroni-adjusted unpaired t-test, * P < 0.05;  P < 0.01. 2) Between pre-PBG infusion baseline and each interval after initiation of PBG infusion; Bonferroni-adjusted paired t-test for control ( P < 0.05) and methysergide-treated group (§ P < 0.05). For clarity, significance symbols are only demonstrated for maximal responses during transient cardiovascular response to infusion.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The present study demonstrated that in conscious rats, central administration of the 5-HT₁/5-HT₂-receptor antagonist, methysergide, was effective in delaying the onset of the hypotensive response to volume loss and preventing the bradycardia and renal sympathoinhibition that accompanies hypotensive hemorrhage. In addition, it was shown that the ability of central methysergide to delay the hypotensive and bradycardic responses to hemorrhage was not dependent on arterial baroreceptors. Furthermore, it was found that the cardiovascular depression and renal sympathoinhibitory response to hemorrhage are not related to cardiopulmonary chemosensitive receptor stimulation by endogenous serotonin release, since blockade of peripheral 5-HT₃ receptors did not influence the hemorrhage response. In addition, central methysergide administration at a dose that was effective in blocking sympatholytic responses produced by hemorrhage did not affect the HR and blood pressure responses to 5-HT₃-receptor stimulation by PBG infusion but did selectively enhance the renal sympathoinhibitory effect of PBG.

Results of the present study do not support speculation that stimulation of cardiopulmonary chemosensitive receptors contributes to the renal sympathoinhibitory response during hemorrhage in the rat (17). With the use of direct recording of RSNA, this study demonstrated that 5-HT₃-receptor blockade had no effect on renal sympathetic responses to hemorrhage, thereby tending to rule out a role for peripheral 5-HT₃ receptors in the observed effects of hypotensive hemorrhage. Indeed in other studies, systemic 5-HT₃-receptor blockade failed to affect either the hemorrhage-induced decrease in total peripheral resistance in conscious rats or the increase in vascular conductance accompanying caval vein occlusion in conscious rabbits (9, 21).

Methysergide was chosen for these studies because of its highly potent effects on rat serotonergic receptors and low activity on dopaminergic and adrenergic receptors. Although methysergide does have less selectivity for the various serotonergic-receptor subtypes than other serotonergic drugs, it is 50- to 100-fold more potent at displacing 5-HT₂-labeled sites than 5-HT₁-labeled sites, making it relatively selective for 5-HT₂ receptors. Moreover, in preliminary studies, we found that methysergide antagonized the effects of hypotensive hemorrhage at a lower dose than the highly selective 5-HT_2A/5-HT_2C-receptor antagonist, LY-53,857 (unpublished data). This is likely related to the higher potency that methysergide reportedly displays for the 5-HT_2C receptor in comparison to the LY-53,857 compound. In addition, methysergide is relatively more soluble than other more selective serotonergic antagonists. The small volume used for central drug administration in these experiments necessitated the use of a compound with relatively good solubility and therefore made methysergide the drug of choice for these studies.

In the present study, central methysergide did not influence reflex hypotensive and bradycardic responses to a systemically administered 5-HT₃-receptor agonist, PBG. In fact, central methysergide actually augmented rather than inhibited the renal sympathoinhibitory response to PBG. Evidence from an earlier study (3) demonstrated that the transient reflex cardiodepressor responses to intravenous PBG injection were abolished by central administration of the 5-HT-receptor antagonist, spiperone. This drug has high affinity and antagonist activity at 5-HT_1A-, 5-HT_1B-, and 5-HT_2A-receptor subtypes. In contrast, central administration of antagonists selective for ₁-adrenergic, 5-HT_2A/5-HT_2C-, or dopamine D₂ receptors did not affect responses to PBG. This suggests that central 5-HT_1A-receptors influence the cardiodepressor response to systemic 5-HT₃-receptor stimulation (3). Methysergide has been found to have affinity for the various 5-HT₁-receptor subtypes as well as 5-HT_2A- and 5-HT_2C receptors. However, since central methysergide did not attenuate the cardiodepressor response to PBG, it is likely that the cardiodepressor effects of methysergide during hemorrhage are mediated by either 5-HT_2A- or 5-HT_2C receptors.

The lack of effect of methysergide on the transient cardiodepressor responses was not due to the use of an excessive dose of PBG. Indeed, a prior study demonstrated that this dose caused only an intermediate renal sympathoinhibitory response in anesthetized rats (22). Preliminary studies in our lab demonstrated that a 25-µg · kg¹ · min¹ infusion of PBG produced submaximal depressor and bradycardic responses in the conscious rat as well.

The transient cardiodepressor response induced by PBG is a well-characterized reflex mediated by activation of chemosensitive receptors of vagal afferent neurons (6). It is possible that the responses to PBG observed in this study were also mediated by other noncardiopulmonary 5-HT₃ receptor populations. Peripheral 5-HT₃ receptors are found on vagal afferent terminals of the gut and the cardiopulmonary organs and vasculature (1, 6, 16). In addition, peripherally administered PBG has access to 5-HT₃ receptors of the area postrema, a circumventricular organ, that lacks the blood-brain barrier that bars access of PBG to other regions of the brain (15). Prior studies have demonstrated that the cardiovascular and renal sympathoinhibitory responses to intravenous PBG infusion are completely abolished by cervical but not by subdiaphragmatic vagotomy (22, 23). In turn, it has been shown that the response is mediated by 5-HT₃-receptor activation, since a highly specific 5-HT₃ receptor blocker completely abolishes the response (22). Therefore, it is most likely that stimulation of the transient cardiodepressor response observed in this study was due to activation of cardiopulmonary vagal afferent chemosensitive receptors.

Methysergide administered directly into the cerebral ventricles caused a delay in the onset of hypotension, bradycardic, and sympathoinhibitory responses to hemorrhage. At the 40-µg dose, intracerebroventricular methysergide was effective in completely abolishing the bradycardic and sympathoinhibitory responses, but the same dose given intravenously had no effect on either baseline or hemorrhage measurements, indicating that methysergide mediated its effects through the central nervous system. These results confirm other reports that have implicated central serotonin in the sympatholytic response to hemorrhage. The hypotension and bradycardia that accompanied a fixed volume of blood withdrawal in anesthetized cats were attenuated by prior serotonin depletion via blockade of endogenous serotonin biosynthesis with p-chlorphenylalanine (7). In addition, it was shown that recovery of blood pressure and HR occurred more rapidly, and morbidity was significantly reduced when methysergide was administered intravenously immediately after hemorrhage (7). Morgan et al. (17) demonstrated that either peripherally administered methysergide or p-chlorphenylalanine (given at doses previously shown to reduce central serotonin levels) prevented the renal sympathoinhibitory response to hemorrhage in chloralose-anesthetized rats. Because methysergide and p-chlorphenylalanine cross the blood-brain barrier, it was suggested that endogenous central serotonin mediated the response to hemorrhage. However, the present study is the first to indicate a specific central serotonergic-receptor-mediated component to the hemorrhagic response in conscious rats.

Additional evidence that central serotonergic receptors are involved in the sympatholytic response to hemorrhage comes from studies in rabbits. Evans et al. (10) initially reported that both methysergide and an alternative 5-HT₂-receptor antagonist, LY-53,857, administered in the fourth cerebral ventricle of conscious rabbits were efficacious in preventing the increased vascular conductance that occurs following simulated hemorrhage with caval vein occlusion. The same authors showed a significant negative correlation between the binding affinity of various serotonergic and ₂-adrenergic ligands for 5-HT_1A receptors isolated from rabbit brainstem homogenate and the dose of the ligands necessary to prevent the increased conductance during caval vein occlusion in conscious rabbits. The authors speculated that activation of 5-HT_1A receptors was responsible for prevention of the conductance response to simulated hemorrhage (9). An alternative explanation is that activation of 5-HT_1A receptors located on serotonergic cell bodies caused an autoinhibition of endogenous serotonin release thus preventing postsynaptic serotonergic-receptor activation. The most likely postsynaptic serotonergic receptor involved in the sympatholytic response to hemorrhage is the 5-HT_2C receptor, since both methysergide and LY-53,857 have high affinity for both 5-HT_2A and 5-HT_2C receptors (5, 12). Ketanserin, a more specific 5-HT_2A-receptor antagonist (11), has no effect on the response to simulated hemorrhage in rabbits (9). Furthermore, the notion that the effects of methysergide are mediated by 5-HT_2C receptors is consistent with the observed lack of effect of methysergide on PBG-induced cardiodepressor responses, which can be blocked by spiperone, an antagonist with high affinity for 5-HT_1A and 5-HT_2A receptors (3, 14).

The mechanism by which methysergide influenced the renal sympathoinhibitory response to hemorrhage in rats was not clearly defined in this study. One possibility is that methysergide augmented reflex arterial baroreceptor responses to low pressure, thus counteracting the sympathoinhibitory effect of the cardiopulmonary afferent stimulation. Studies in rabbits indicate that baroreflex control of RSNA is severely impaired following hypotensive hemorrhage, whereas HR baroreflex responses are only mildly affected. The loss of RSNA baroreflex sensitivity can be reversed by blockade of cardiac vagal afferent activity with procaine, suggesting that vagal afferent activation during hemorrhage suppresses arterial baroreceptor reflex control of RSNA, but not HR, a notion consistent with the fact that rabbits do not demonstrate bradycardia during hypotensive hemorrhage (4). It is possible that central endogenous serotonin plays a role in the suppression of arterial baroreflex control of both HR and RSNA during hypotensive hemorrhage in rats. Although studies indicate that peripheral serotonergic-receptor blockade reduces baroreflex sensitivity in normovolemic animals (8, 17), there is still a lack of convincing evidence that central serotonin mediates baroreflex pathways. In the present study, intracerebroventricular methysergide was still effective in delaying the onset of hemorrhage-induced bradycardia in SAD rats, indicating that the effect of methysergide on hemorrhage-induced HR responses is, at least in part, independent of the arterial baroreceptors. However, it is possible that the effect of methysergide on RSNA during hemorrhage is mediated through a different pathway and could be dependent on arterial baroreflex sensitization.

An alternative possibility is that methysergide reduced the sensitivity of the cardiopulmonary mechanoreceptor reflex, which could attenuate the afferent signal for sympathoinhibition. Reports using volume loading as a stimulus for cardiopulmonary mechanoreceptor activation have failed to demonstrate an effect of methysergide on the cardiopulmonary reflex (17). However, it is possible that a central serotonin-mediated sensitization of cardiopulmonary mechanoreceptors might only be observable under conditions of hypovolemia. As yet, this possibility has not been tested.

In summary, the current study demonstrated that intracerebroventricular administration of methysergide prevents the bradycardia and renal sympathoinhibition that accompanies hypotensive hemorrhage. The effect of methysergide is not due to arterial baroreceptor sensitization, since SAD did not prevent the delay in hypotension or the delay and attenuation of bradycardia. Furthermore, the bradycardic and sympathoinhibitory responses to hemorrhage were not associated with activation of cardiac chemosensitive receptors responsive to serotonin, since systemic 5-HT₃-receptor blockade did not alter the hemorrhage response. Further support for the notion that cardiopulmonary chemoreceptors are not involved in the response to hemorrhage is provided by the opposing effects of methysergide on the renal sympathoinhibitory responses to hemorrhage and 5-HT₃-receptor stimulation. The mechanism by which methysergide attenuated responses to hemorrhage and augmented the sympathoinhibitory response to 5-HT₃-receptor stimulation remains to be determined, but it is likely to involve the blockade of serotonergic 5-HT_2C receptors within the central nervous system.