Serotonergic mechanisms mediate renal sympathoinhibition during

Serotonergic mechanisms mediate renal sympathoinhibition during severe hemorrhage

in rats

DONALD A. MORGAN, PETER THOREN, ESTHER A. WILCZYNSKI, RONALD G. VICTOR, AND ALLYN L. MARK Veterans Administration Medical Center, Department of Medicine, and The Cardiovascular Center, University of Iowa College of Medicine, Iowa City, Iowa 52242

MORGAN, DONALD A., PETER THOREN, ESTHER A. WILCZYNSKI, RONALD G. VICTOR, AND ALLYN L. MARK. Serotonergic mechanisms mediate renal sympathoinhibition during seuere hemorrhage in rats. Am. J. Physiol. 255 (Heart Circ. Physiol. 24): H496-H502,1988.-Hemorrhage in rats produces reflex decreases in heart rate (HR) and renal sympathetic nerve activity (RSNA). Because serotonergic antagonists attenuate hemorrhage-induced vagal-mediated bradycardia, we determined whether blockade of serotonin synthesis by p-chlorophenylalanine (PCPA) or of serotonin receptors with methysergide would also abolish the renal sympathoinhibition. Mean arterial pressure (MAP), HR, and RSNA were recorded in chloralose-anesthetized rats pretreated with PCPA (300 mg= kg-‘. day-l x 3 days ip, n = 12) or vehicle (0.3 ml saline, n = 9). During hemorrhage, where MAP was maintained at 50 mmHg for 8 min, vehicle-treated rats decreased HR by 27 t 13 beats/min and RSNA by -55 t 7%. In PCPA-treated rats, HR and RSNA did not change. Cervical vagotomy abolished the bradycardia and sympathoinhibition during hemorrhage. After acute administration of methysergide (400 ,ug/kg iv, n = 8) hemorrhage produced increases of RSNA, whereas vehicle (0.5 ml saline, n = 7) preserved the renal sympathoinhibition to hemorrhage in conscious rats. Finally, volume expansion (0.88 ml blood/100 g body wt) produced comparable decreases in RSNA in sinoaortic-denervated rats pretreated with PCPA (n = 11) or vehicle (n = 10) (-58 * 9 vs. 47 -t 7%, respectively). We conclude that serotonergic mechanisms are critically involved in vagal afferent inhibition of RSNA during severe hemorrhage in rats. hypotension; vagal afferent fibers; arterial baroreceptors; chlorophenylalanine; methysergide; volume expansion

p-

IS INCREASING EVIDENCE that hemorrhage can activate bradycardic, sympathoinhibitory, and vasodepressor reflexes in experimental animals (11, 20, 25, 26). In addition, recent observations indicate that hemorrhagic hypotension also is frequently accompanied by bradycardia in patients (17, 18). These responses have been attributed to activation of inhibitory cardiac vagal afferent neurons (fibers) in several species including cats and rats (4, 13-15, 19, 20, 27). For example, Thoren and his colleagues (13, 20) have suggested that during hemorrhagic hypotension, the vigorous contraction of the heart around nearly empty cardiac chambers results in paradoxical activation of inhibitory cardiac vagal afferent fibers. In contrast, vagal afferent fibers do not appear to mediate the sympathoinTHERE

H496

hibitory response to hemorrhage in dogs (11). Serotonergic mechanisms have also been implicated in vasodepressor responses to hemorrhage. Elam et al. (6) reported that serotonergic antagonists facilitated recovery of arterial pressure after hemorrhage in cats. The role of serotonergic mechanisms in the sympathoinhibitory responses to hemorrhage is unknown. Thus the first goal of this study was to determine whether serotonergic antagonists block the renal sympathoinhibitory responses to hemorrhage in rats. We used p-chlorophenylalanine which blocks serotonin biosynthesis (10) and methysergide which blocks serotonergic receptors (1). A second goal was to determine whether serotonergic antagonists also block the sympathetic responses to activation of volume sensitive cardiac vagal afferent neurons (fibers) and arterial baroreceptors. METHODS

Fifty-seven female Sprague-Dawley rats (BioLab) with body weights from 230 to 290 g were studied. All rats were given regular rat chow and tap water ad libitum. Animal care and experiments complied with the guiding principles of the American Physiology Society for animal experimentation. Animals were anesthetized with methohexital sodium (Brevital, 40 pg/kg ip). Polyethylene catheters were inserted into a femoral artery to measure arterial pressure, into a femoral vein for drug administration and volume expansion, and into the caudal artery for bleeding. Heart rate (HR, beats/min) was measured by a cardiotachometer (Beckman 9857B) triggered from the femoral arterial pressure signal. For experiments in anesthetized rats (n = 42), methohexital was followed by chloralose (50 mg/kg iv in bolus and 25 mg/kg iv per hour). In the studies on chloraloseanesthetized rats, the trachea was cannulated with PE240 and animals breathed room air supplemented with 100% oxygen. In seven rats, arterial blood gases under these conditions were pH 7.38 t 0.01 U, POT 267 +- 21.6 and PCO~ 34.3 t 1.4. For experiments in the conscious state (n = 15), rats were given repeated injections of methohexital (IO-15 mg/kg iv) to maintain a constant level of anesthesia during surgery. The rats were then allowed to awaken and to stabilize for 3 h after the last methohexital injection before beginning the experiments while in the conscious state.

SEROTONERGIC

MECHANISMS

Surgical Procedures Sinoaortic denervation. For studies of cardiopulmonary baroreflex (CPBR) control of renal nerve activity, 21 of 57 rats underwent tracheal intubation and then sinoaortic denervation (SAD). SAD was performed according to the technique of Kreiger (8). Sinoaortic baroreceptors were denervated by removing the aortic depressor nerve along with the superior laryngeal nerve. The cervical sympathetic ganglion and its trunk were also removed. The arterial walls in the carotid sinus region were carefully stripped of neural and connective tissues and painted with 10% phenol in 70% ethanol. Effectiveness of the SAD was confirmed by failure of phenylephrineinduced increases in MAP (20-40 mmHg) to elicit a reflex decrease in HR (>90% inhibition of reflex bradycardia). A polyethylene catheter was inserted into the left ventricle (through the right common carotid artery) to monitor left ventricular end-diastolic pressure (LVEDP). LVEDP was used as a measure of left heart filling pressure and the stimulus to volume sensitive cardiac sensory receptors. Sympathetic nerve recording. The left kidney was exposed with a retroperitoneal dissection. A sympathetic nerve branch to the kidney was dissected free close to the aorta and placed on a thin bipolar platinum electrode. When an optimal nerve recording had been achieved, the nerve and the electrode were covered with silicon rubber (Wacker Sil-Gel 604) as described in detail previously (16). When the recordings were performed in awake animals, the animals were placed in a tubular restraining cage that was large enough to allow movement backwards and forwards. The multiunit nerve signals were detected by a highimpedance probe and amplified 50,000-100,000 by a Grass band pass P511 amplifier with a band width of lOO-1,000 Hz. For monitoring during the experiment, the filtered neurogram was visualized on a Tektronics oscilloscope (502A) and routed through an amplitude discriminator to an audio amplifier and loudspeaker. For permanent recordings and analysis, the filtered neurogram was fed through a nerve traffic analyzer (no. 706C, University of Iowa Bioengineering), which counted nerve spikes exceeding a threshold voltage set just above the noise level. The counter’s time bin was set at 1 s so that the impulse frequency was displayed as the number of spikes collected each second (Hz) on a time-frequency histogram. This display was used to measure changes,in renal sympathetic nerve activity (RSNA) during hemorrhage, volume expansion, and phenylephrine. Measured nerve activity, arterial pressure, HR, and LVEDP were recorded continuously on a Beckman RM Dynograph recorder at a paper speed of 25-50 mm/min (50100 mm/set when LVEDP values were measured). Experimental

Protocols

After surgery, the chloralose-anesthetized animals were allowed to stabilize on a temperature-controlled heating pad maintained at 37.5”C for at least 1 h. The conscious animals were allowed to stabilize for at least 3 h before beginning the experiments.

DURING

HEMORRHAGE

IN

RATS

H497

We performed three studies. Study 1: anesthetized rats with arterial baroreceptors intact were pretreated with either p-chlorophenylalanine (PCPA, 300 mg/kg ip x 3 days, n = 12) or vehicle (n = 9). In these experiments, we obtained responses to hemorrhage and to phenylephrine; study 2: in conscious rats with arterial baroreceptors intact, we obtained responses to hemorrhage and to phenylephrine before and after methysergide maleate (400 pg/kg iv, n = 8) or vehicle (0.5 ml saline iv, n = 7); study 3: in anesthetized SAD rats pretreated with PCPA (300 mg/kg ip x 3 days, n = 11) or vehicle (n = lo), we measured responses to volume expansion. In the first study, 12 rats received PCPA (300 mg/kg ip per day) for 3 consecutive days. Nine rats received intraperitoneal injections of vehicle (0.3-0.5 ml saline, pH 7.3). Daily measurements of body weight and food/ water consumption were recorded. After 3 days of treatment, rats were anesthetized as previously described and femoral arterial pressure, HR, and RSNA were obtained. The rats were intubated and breathed oxygen-enriched air. After 1 h, when rats were stable, phenylephrine (PE, 0.4-4.0 pg= kg-’ . min-‘) was slowly infused by a peristaltic pump for 60-90 s to raise arterial pressure. Arterial baroreflex control of HR and RSNA was evaluated using linear regression analysis to obtain the best-fit reflex slope of HR (change in beats/min) and RSNA (percent change from control) vs. the change of MAP (in mmHg). The slopes for HR and RSNA for each rat were determined and then analyzed as a mean slope for the respective PCPA or vehicle-treatment groups. When arterial pressure, HR, and RSNA had returned to base line, MAP was then decreased in 30-60 s to 50 mmHg by hemorrhage and maintained at this level for 8 min while recording HR and RSNA. This particular maneuver requires hemorrhage of 15-20% of total blood volume in the rat. At the end of 8 min, the blood was reinfused. HR blood pressure, and RSNA returned to base-line values and were allowed to stabilize for l-2 h at which time cervical vagotomy was performed. After another recovery period, responses to phenylephrine and hemorrhage were again obtained. Blood was again infused and chlorisondamine chloride (Ecolid, 5 mg/kg iv) was administered. This abolished all renal nerve activity and was used to establish that the nerve recordings were free of artifact. In the second study, 15 rats were studied in the conscious state. Phenylephrine infusion (0.4-40 pg. kg-‘. min-‘) and hemorrhage were performed before and after methysergide or vehicle. In the third study, 11 rats were pretreated with PCPA and 10 with vehicle. After 3 days of treatment, the rats were anesthetized and sinoaortic baroreceptors were denervated. When experimental variables had stabilized, graded volume expansion was performed by injection of blood into the femoral vein in 0.5-ml increments to a total of 2.5 ml. Each increment in volume was administered over 15-30 s. Data were obtained during the peak response immediately after each increment in volume. After volume expansion, blood was withdrawn to restore preinfusion values of LVEDP. The cervical vagi were then cut and, after 10 min, volume expansion and withdrawal were repeated.

H498

SEROTONERGIC

MECHANISMS

Data Analysis In all studies, on-line acquisition and data analysis were performed with an IBM software routine. Analog values of MAP, HR, and RSNA were digitized on-line during each experimental intervention. In study 3, values of LVEDP and RSNA were measured by inspection of the polygraph record. Statistical comparisons were performed on values in the control state and during the 8th min of hemorrhage and the peak response to phenylephrine and volume expansion. Control values were calculated before each intervention. Comparisons within groups were performed with paired t test; comparisons between groups were performed with unpaired t tests. Values of P c 0.05 were considered significant. Results are expressed as means t SE . RESULTS

DURING

HEMORRHAGE

IN

RATS

t 18 ml in PCPA-treated rats, respectively (P < 0.05). PCPA-treated rats were also more excitable and aggressive when handled compared with the vehicle-treated rats. Responsesto Hemorrhage in PCPA- vs. Vehicle-Treated Rats with Intact Vagi Before hemorrhage, MAP was significantly (P < 0.05) lower in PCPA-treated (102 t 5 mmHg) vs. vehicletreated (122 t 4 mmHg) rats. HR did not differ significantly in the two groups (381 t 10 in PCPA- vs. 394 t 9 beats/min in vehicle-treated rats). Rats were bled to a MAP of -50 mmHg. The amount of blood withdrawn did not differ significantly in the two groups (12.9 t 2 and 11.3 t 1 ml/kg in PCPA- and vehicle-treated rats, respectively). MAP during the 8th min of hemorrhage averaged 53 t 2 mmHg in PCPAtreated rats vs. 47 t 2 mmHg in vehicle-treated rats (P

c 0.05).

Effects of PCPA PCPA decreased body weight and food consumption and increased water consumption. Body weight was comparable in vehicle (279 t 5 g) and PCPA (271 t 7 g) rats before treatment. In contrast, on the 3rd day of treatment, body weight was 286 t 7 in vehicle-treated rats and 254 t 7 in PCPA-treated rats (P < 0.05). Food and water consumption were 51 t 3 g and 132 t 6 ml, respectively, in vehicle-treated rats and 16 t 2 g and 199

MAP (mmHg)

Vehicle

Treated

k

t

Hemorrhage decreased HR and RSNA in vehicletreated rats but did not decrease HR or RSNA in PCPAtreated rats (Figs. 1 and 2, and Table 1). Thus PCPA prevented reflex sympathoinhibition during hemorrhage. Responsesto Hemorrhage in PCPA- vs. Vehicle-Treated Rats with Bilateral Cervical Vagotomy After vagotomy, hemorrhage increased RSNA in both PCPA- and vehicle-treated rats with no significant dif-

PCPA Treated (300 mg/kg i.p x 3 days)

loo 200

r

Renal SNA 100 (Hz) t 560 Heart Rate 4oo @pm)

r

24OL Hemorrhage (2.4 ml)

30 set

1

k Hemorrhage

t

30 set

1

(3 ml)

1. Segments of original records from a vehicle-treated rat (left) and from a p-chlorophenylalanine (PCPA)treated rat (right). MAP, mean arterial pressure; renal SNA, renal sympathetic nerve activity. In vehicle-treated rat, fall in arterial pressure during hemorrhage was followed by a gradual and sustained decrease in renal SNA and heart rate (HR). In PCPA-treated rat, fall in arterial pressure with hemorrhage was not accompanied by a decrease in renal SNA or HR. Absence of renal sympathoinhibition during hemorrhage in PCPA-treated rat could not be explained by a nonspecific depression of cardiovascular reflexes, since an increase in arterial pressure and stimulation of arterial baroreceptors produced by injection of phenylephrine (PE) was accompanied by a prompt and striking decrease in renal SNA (far right). FIG.

k PE

SEROTONERGIC

MECHANISMS

MAP (mmHg)

DURING

Heart

HEMORRHAGE

Rate (bpm)

IN

H499

RATS

Renal

SNA (%)

480 0 Vehicle 0 PCPA

300 -1012345678 t Hemorrhage

-1012345678 t Hemorrhage

-1012345678 t Hemorrhage

Time (min) 2. Hemodynamic and renal SNA fall in arterial pressure with hemorrhage persisted throughout 8 min of hemorrhage. was not accompanied by a decrease in HR FIG.

responses to hemorrhage in vehicle vs. PCPA rats. In vehicle-treated rats, was accompanied by a decrease in heart rate (HR) and renal SNA which In contrast, in PCPA-treated rats, fall in arterial pressure with hemorrhage renal SNA.

1. Responsesto hemorrhage (8th min) in vehicleand p-chlorophenylalanine- treated rats TABLE

AMAP, mmHg Vehicle

Hemorrhage before vagotomy Hemorrhage after vagotomy

AHeart Rate, beats/min

ARSNA, %

PCPA

Vehicle

PCPA

Vehicle

PCPA

-74t4

-5ot4*

-27213

+8t7*

-55t7

+5t8*

-7Ot6

-5525

+13t3

+24t6

+45t9

+37t7

Values are means t SE; n = 9 rats for vehicle group and n = 12 for p-chlorophenylalanine (PCPA) group. AMAP, change in mean arterial pressure; AHR, change in heart rate; ARSNA, change in renal sympathetic nerve activity.

ference in the increases in RSNA in the two groups (Fig. 3 and Table 1). Responsesto Arterial Baroreceptor Stimulation with Phenylephrine in PCPA- vs. Vehicle-Treated Rats Before phenylephrine, base-line MAP and RSNA were 119 2 6 mmHg and 142 t 19 Hz, respectively, in vehicletreated rats and 86 t 4 mmHg (P < 0.05 vs. vehicle) and 127 t 5 Hz in PCPA-treated rats. Stimulation of arterial baroreceptors by increases in arterial pressure during phenylephrine infusion produced reflex decreases in RSNA (Figs. 1 and 4). In animals with intact vagi, the gain of the baroreflex control of RSNA (expressed as ARSNA in %/AMAP in mmHg) was significantly higher in PCPA-treated rats compared with vehicle-treated rats (Fig. 4). After vagotomy, base-line MAP and RSNA were 115 t 7 mmHg and 114 t 10 Hz, respectively, in vehicletreated rats and 101 t 5 mmHg and 127 t 10 Hz,

respectively, in PCPA-treated rats. After vagotomy, the gain of the arterial baroreceptor control of RSNA tended to be greater in PCPA- vs. vehicle-treated rats (Fig. 4). Thus PCPA tended to augment reflex sympathoinhibition during stimulation of arterial baroreceptors. Responsesto Hemorrhage and to Arterial Baroreceptor Stimulation With Phenylephrine Before and After Methysergide in Conscious Rats Intravenous administration of methysergide in the conscious rat did not change base-line MAP or HR but did increase resting RSNA from 123 t 6 to 145 t 7 Hz (P < 0.05). Intravenous injection of saline had no effect on base-line MAP, HR, and RSNA in the vehicle group (Table 2). In the control state, hemorrhage decreased HR and RSNA in the conscious rats. Vehicle did not significantly alter these responses to hemorrhage (Fig. 5A), but methysergide tended to reduce the bradycardic response to hemorrhage and reversed the sympathetic response from a decrease to an increase in RSNA (Fig. 5B). During a recovery period, 1 h after administration of methysergide, hemorrhage produced a decrease in RSNA, which was comparable to that during the control. The amount of blood withdrawn did not differ before vs. after methysergide or vehicle. Methysergide did not alter the decreases in RSNA produced by stimulation of arterial baroreceptors during increases in arterial pressure with phenylephrine. Slopes for arterial baroreflex control of RSNA (expressed as ARSNA in %/AMAP in mmHg) were - 1.81 t 0.37 and -1.54 t 0.28 before and after methysergide, respectively.

H500

SEROTONERGIC

MECHANISMS

MAP (mmHg)

DURING

Heart

Rate

HEMORRHAGE

(bpm)

IN

RATS

Renal

SNA

(%)

480 0 Vehicle 0 PCPA

/ I I I I I I I I -1012345678 t Hemorrhage

-1012345678 t Hemorrhage

-1012345678 t Hemorrhage

Time (min) FIG.

cervical in both

3. Hemodynamic and renal SNA responses to hemorrhage in vehicleand PCPA-treated vagotomy. In these rats with interruption of vagal afferent neurons (fibers), hemorrhage vehicle- and PCPA-treated rats with no significant difference in increase in renal SNA Before

Vagotomy

A Renal AMAP

After

SNA (%) (mmtig)

0

A Renal AMAP 0

Vagotomy SNA (%l (mmHg)

K n=9

n=9

-3

-3 n=12

u

Lp -z0.05J

n=12

rats with bilateral increased renal SNA in the two groups.

2. Base-line values for mean arterial pressure, heart rate, and renal sympathetic nerve activity before and after vehicle or methysergide

TABLE

Vehicle Before

MAP, mmHg HR, beats/min RSNA, Hz

vehicle

Methysergide

-6 Vehicle

PCPA

Vehicle

PCPA

FIG. 4. Sympathetic nerve responses to stimulation of arterial baroreceptors by increases in arterial pressure produced by infusion of phenylephrine in vehicleand PCPA-treated rats. Gain of arterial baroreflex control of renal SNA is plotted as ratio of change in renal SNA (in percent) divided by change in mean arterial pressure (MAP) (in mmHg). Before vagotomy (shown on Left) gain of baroreflex control of renal SNA was significantly higher in PCPAvs. vehicle-treated rats. After vagotomy, gain of arterial baroreflex control of renal SNA tended to be greater in PCPA- vs. vehicle-treated rats. ns, not significant.

Responsesto Volume Expansion in SAD Rats Treated With PCPA or Vehicle In anesthetized rats with SAD, volume expansion increased left ventricular end-diastolic pressure and decreased renal SNA. These responses did not differ significantly in PCPA- and vehicle-treated rats (Table 3). Indeed, if anything, decreases in renal SNA tended to be greater in the PCPA-treated rats (Table 3). Vagotomy greatly attenuated (P < 0.05) the decreases in renal SNA produced by volume expansion in both PCPA- and vehicle-treated rats.

MAP, mmHg HR, beats/min RSNA, Hz

12Ok5 408t12 123k6

(n = 7) After

129t6 46229 122t6

Before methysergide -6

Group

vehicle

1261k7 453tlO 119t7 Group

(n = 8) After methysergide

121t4 430t15 145t7*

Values are means t SE; n = 7 rats for vehicle group and n = 8 rats for methysergide group. See Table 1 for abbreviations. * P < 0.05 before vs. after methysergide.

DISCUSSION

The principal new finding in this study is that serotonergic blockers (PCPA and methysergide) prevented the renal sympathoinhibitory response to activation of vagal afferent neurons (fibers) during hypotensive hemorrhage in rats. In contrast, serotonergic blockers did not attenuate the sympathoinhibitory response to activation of volume-sensitive vagal afferent neurons or arterial baroreceptors. These observations suggest that central serotonergic mechanisms are critically involved in the decreasesin renal sympathetic nerve activity in response to activation of vagal afferent neurons during severe hemorrhage in the rat. We shall consider potential limitations of our experimental methods and preparation. First, could our results be explained by nonspecific depressant effects of PCPA,

SEROTONERGIC

A

A MAP

(mmHg)

A Renal

MECHANISMS

SNA (%)

A Heart

Rate

DURING

(bpm)

100 n = 5-7 mean f SE

0

rp

I”S--J

-100 Control

6

Vehicle

Control

A MAP (mmHg)

0

Vehicle

A Renai

Control

SNA (%)

A Heart

100

-J PC005

-100 Control

Methysergide

(bpm)

-75

L -100

Rate

0

0

-50

Vehicle

n=a mean

f SE

-150 Control

Methysergide

Control

FIG. 5. Hemodynamic and renal SNA responses control state and after vehicle (A) and in control methysergide (B). Entries represent means & SE control at 8th min of hemorrhage. Vehicle did not MAP, heart rate, and renal SNA during hemorrhage methysergide tended to reduce bradycardic response and reversed renal SNA response from a decrease renal SNA with hemorrhage (B).

Methysergide

to hemorrhage in state and after of changes from alter changes in (A). In contrast, to hemorrhage to an increase in

perhaps related to decreased food consumption and body weight? This is unlikely, since the sympathoinhibitory responses to phenylephrine and to volume expansion were both preserved although the sympathoinhibitory response to hemorrhage was abolished by PCPA. In addition, the sympathoinhibitory response to hemorrhage was also blocked by the acute administration of methysergide, which obviously did not decrease body weight. Second, could the effects of PCPA and methysergide be related to actions other than serotonergic blockade, e.g., depletion of catecholamines by PCPA? We did not measure levels of serotonin and catecholamines in the brain in PCPA-treated rats; however, the finding that both methysergide and PCPA blocked the sympathoinhibition during hemorrhage suggests that these effects were related to antiserotonergic activity and

3. Responsesto volume expansion in p-chlorophenylalaninewith sinoaortic denervation

Base line, mmHg Before vagotomy After vagotomy

4.3t0.5 4.1t0.7

+16.3t0.9 +14.6t1.2

and vehicle-treated rats RSNA,

PCPA Group Base line, mmHg 3.2t0.6 4.1t0.8

Hi501

RATS

mmHg

Vehicle Group Response to VE, mmHg

IN

not to catecholamine depletion, since the latter would not be expected with methysergide. Third, could the absence of renal sympathoinhibition during hemorrhage in PCPA-treated rats be related to a lesser stimulus, e.g., a lesser fall in arterial pressure during hemorrhage? This seemsunlikely. The values for MAP during hemorrhage were only slightly higher in the PCPA- than in the vehicle-treated rats. Moreover, there was no difference in the fall in arterial pressure with hemorrhage before and after methysergide. Finally, and perhaps most importantly, the amount of blood withdrawn did not differ significantly in vehicle- vs. PCPA-treated rats or before and after methysergide. Our observations therefore suggest that serotonergic mechanism are importantly involved in the sympathoinhibitory responses to activation of vagal afferent reflexes during hemorrhage but are not involved in responses to activation of arterial baroreceptor reflexes or of mechanosensitive vagal afferent neurons (fibers). We cannot from these experiments precisely define the site and nature of the involvement of serotonergic mechanisms. The finding that phenylephrine and volume expansion decreased renal SNA during PCPA and methysergide excludes the possibility that serotonergic blockade was inhibiting ganglionic transmission. Thus it seems likely that serotonergic mechanisms are involved in the sympathoinhibitory responses to hemorrhage either 1) in the central nervous system or 2) in the activation of vagal sensory receptors and afferent impulses. Central serotonergic neurons appear to be important in the central neural regulation of cardiovascular function (2, 3, 5, 10, 12, 22, 23). Many studies have suggested that the activity of central serotonergic nerves elevates arterial pressure in the rat (3, lo), although others have suggested depressor effects (5,12). In this regard, several investigators have emphasized that serotonergic neurons in different brain regions may subserve opposing functions. We cannot from our observations state whether hemorrhage might increase the activity of inhibitory serotonergic pathways or conversely might decrease the activity of excitatory serotonergic pathways. It might be noted, however, that Sole et al. (22, 23) have demonstrated that serotonergic activity in discrete nuclei in the medulla and posterior hypothalamus is decreased by stimulation of cardiac vagal afferent neurons (fibers) during myocardial infarction in the rat. Electrical stimulation of one of these nuclei (the nucleus centralis

TABLE

LVEDP,

HEMORRHAGE

Response to VE, mmHg +16.lt1.8 +16.4&1.6

Hz

Vehicle Group Base line, Hz 149*7 124t9

Response to VE, % -47t7 -16k6

PCPA Group Base line, Hz 112k7 129t12

Response to VE, % -59t9 -8t4

Values are means t SE; n = 10 for vehicle group and n = 11 for p-chlorophenylalanine (PCPA) group. Volume expansion (VE) was performed with 2.5 ml blood. Renal sympathetic nerve activity (RSNA) responses to VE were reduced (P < 0.05) after vagotomy in both groups. There were no significant differences between vehicle and PCPA groups in responses to VE. LVEDP, left ventricular end-diastolic pressure.

H502

SEROTONERGIC

MECHANISMS

superior) in the rat produces a pressor response that is abolished by PCPA (9, 21). In other words, it appears from previous work that activation of cardiac vagal afferent neurons (fibers) during myocardial infarction decreases the activity of excitatory serotonergic pathways in rats. If stimulation of cardiac vagal afferent neurons (fibers) during hemorrhage resembles that during myocardial infarction, this would suggest that hemorrhage is decreasing the activity of excitatory serotonergic pathways. A second explanation for our findings could be a role of serotonin in ac tivation of vagal sensory receptors and afferent impulses during hemorrhage. It is known that adminis tration of serotonin can activate vagal afferent (7, 24). reflexes that cause bradycardia a nd hypotension Hemorrhage might increase the synthesis and release of serotonin and thereby the activation of vagal afferent reflexes. We cannot from our observations determine whether serotonin is acting at the sensory level or in the central nervous system to produce reflex sympathoinhibition during hemorrhage. In conclusion, the results of this study indicate that serotonergic blockade prevents vagal afferent inhibition of renal sympathetic nerve activity during hemorrhage but does not attenuate renal sympatho inhibi tory responses to stimulation of volume sensitive vagal afferent neurons (fibers) or arterial baroreceptors. The authors thank Sara Jedlicka and Nancy Davin for expert secretarial assistance. This research was supported by Clinical Investigator Award HL01362 to R. G. Victor and Program Project Grant HL-14388 from the National Heart, Lung, and Blood Institute; by research funds from the Veterans Administration; by a travel grant to P. Thoren from the Swedish Medical Research Council; and by a fellowship to E. A. Wilczynski from the Canadian Heart Foundation. Present addresses: P. Thoren, Dept. of Physiology, Box 33031, S40033 Goteborg, Sweden; E. A. Wilczynski, Division of Clinical Pharmacology, Rm. 1418, Toronto Western Hospital, 399 Bathurst St., Toronto, Ontario M5T 258, Canada; R. G. Victor, Cardiology Division, Rm. H8.116, Dept. of Internal Medicine, Univ. of Texas Health Science Center at Dallas, 5323 Harry Hines Blvd., Dallas, TX 75235. Received

21 September

1987; accepted

in final

form

27 April

1988.

REFERENCES 1. BRADLEY, P. B., G. ENGEL, W. FENIUK, J. R. FOZARD, P. P. A. HUMPHREY, D. N. MII~DLEMISS, E. J. MYLECHARANE, B. P. RICHARDSON, AND P. R. SAXENA. Proposals for the classification and nomenclature of functional receptors for 5-hydroxytryptamine. Neuropharmacology 25: 563-576,1986. 2. CHALMERS, J. P. Brain amines and models of experimental hypertension. Circ. Res. 36: 469-480, 1975. 3. CHALMERS, J. P., J. B. MINSON, AND V. CHOY. Bulbospinal serotonin pressor pathways and hypotensive action of methyldopa in the rat. Hypertension Dabs 6: 1116-1121, 1984. 4. CHEN, H. I., H. 0. STINNETT, D. F. PETERSON, AND V. S. BISHOP. Enhancement of vagal restraint on systemic blood pressure during hemorrhage. Am. J. Physiol. 234 (Heart Circ. Physiol. 3): Hl92H198,1978. 5. COOTE, J. H., AND V. H. MACLEOD. The influence of bulbospinal monoaminergic pathways on sympathetic nerve activity. J. Physiol. Lond 241: 453-475,1974.

DURING

HEMORRHAGE

IN

RATS

6. ELAM, R., F. BERGMANN, AND G. FEUERSTEIN. The use of antiserotonergic agents for the treatment of acute hemorrhagic shock of cats. Eur. J. Pharmacol. 107: 275-278, 1985. 7. KALKMAN, H. O., G. ENGEL, AND D. HOYER. Three distinct subtypes of serotonergic receptors mediate the triphasic blood pressure response to serotonin in rats. J. Hypertension 2, Suppl. 3: 143-145,1984. 8. KREIGER, E. M. Neurogenic hypertension in the rat. Circ. Res. 15: 511-521,1964. 9. KUHN, D. M., W. A. WOLF, AND W. LOVENBERG. Pressor effects of electrical stimulation of the dorsal and median raphe nuclei in anesthetized rats. J. Pharmacol. Exp. Ther. 214: 403-409, 1980. 10. KUHN, D. M., W. A. WOLF, AND W. LOVENBERG. Review of the role of the central serotonergic neuronal system in blood pressure regulation. Hypertension Dallas 2: 243-255, 1980. 11. MORITA, H., AND S. VATNER. Effects of hemorrhage on renal nerve activity in conscious dogs. Circ. Res. 57: 788-793, 1985. 12. NEUMAYR, R. J., B. D. HARE, AND D. N. FRANZ. Evidence for bulbospinal control of sympathetic preganglionic neurons by monoaminergic pathways. Life Sci. 14: 793-806, 1974. 13. OBERG, B., AND P. THOREN. Increased activity in left ventricular receptors during hemorrhage or occlusion of caval veins in the cat. A possible cause of the vasovagal reaction. Acta Physiol. Stand. 85: 164-173,1972. 14. OBERG, B., AND S. WHITE. The role of vagal cardiac nerves and arterial baroreceptors in the circulatory adjustment to hemorrhage in the cat. Acta Physiol. Stand. 80: 395-403, 1970. 15. PEARCE, J. W., AND J. P. HENRY. Changes in cardiac afferent nerve-fiber discharges induced by hemorrhage and adrenalin (Abstract). Am. J. Physiol. 183: 650, 1955. 16. RICKSTEN, S. E., E. NORESSEN, AND P. THOREN. Inhibition of renal sympathetic nerve traffic from cardiac receptors in normotensive and spontaneously hypertensive rats. Acta Physiol. Stand. 106: 17-22,1979. R., N. H. SECHER, P. BIE, J. WARBERG, AND T. 17. SANDER-JENSEN, W. SCHWARTZ. Vagal slowing of the heart during haemorrhage: observations from 20 consecutive hypotensive patients. Br. Med. J. 292: 365-366,1986. 18. SECHER, N. H., K. S. JENSEN, C. WERNER, J. WARBERG, AND P. BIE. Bradycardia during severe but reversible hypovolemic shock in man. Circ. Shock 14: 267-274, 1984. 19. SJOSTRAND, T. Circulatory control via vagal afferents, the bleeding bradycardia in the rat, its elicitation and relation to the release of vasopressin. Acta Physiol. Stand. 89: 39-50, 1973. 20. SKOOG, P., J. MANSSON, AND P. THOREN. Changes in renal sympathetic outflow during hypotensive haemorrhage in rats. Acta Physiol. Scund. 125: 655-660, 1985. 21. SMITS, J. F. M., H. VAN ESSEN, AND H. A. J. STRUYKER-BOUDIER. Serotonin-mediated cardiovascular responses to electrical stimulation of the raphe nuclei in the rat. Life Sci. 23: 173-178, 1978. 22. SOLE, M. J., G. R. VAN LOON, A. SHUM, W. LIXFIELD, AND D. C. MACGREGOR. Left ventricular receptors inhibit brain serotonin neurons during coronary artery occlusion. Science Wash. DC 201: 620-622,1978. 23. SOLE, M. J., D. H. G. VERSTEEG, E. R. DE KLOET, N. HUSSAIN, AND W. LIXFELD. The identification of specific serotonergic nuclei inhibited by cardiac vagal afferents during acute myocardial ischemia in the rat. Bruin Res. 265: 55-61, 1983. ’ 24. THAMES, M. D., U. J. JOHANNSEN, AND A. L. MARK. In search of afferent pathways of a cardiogenic hypertensive chemoreflex. Circulation 75: 643-650, 1987. P. Two different inhibitory reflexes elicited from cardiac 25. THOREN, endings of rats. In: Neurul Medianisms and Cardiovascular Disease. Padua, Italy: Liviana, 1986, p. 35-46. 26. UNDESSER, K. P., E. M. HASSER, A. J. TRAPANI, AND V. S. BISHOP. The effect of vasopressin on renal sympathetic nerve activity responses to hemorrhage in conscious rabbits. Circulation 72: 111-17, 1985. 27. VICTOR, R. G., P. THOREN, D. A. MORGAN, AND A. L. MARK. Differential control of adrenal vs. renal sympathetic outflow during hemorrhagic hypotension in rats (Abstract). Clin. Res. 34: 633A, 1986.