The effect of ketanserin on cardiovascular reflexes in conscious normotensive and spontaneously hypertensive rats

European Journal of Pharmacolog),, 186 (1990) 17-28

17

Elsevier EJP 51464

The effect of ketanserin on cardiovascular reflexes in conscious nonnotensive and spontaneously hypertensive rats R o b e r t E. W i d d o p *, A n t h o n y J.M. V e r b e r n e , W i l l i a m J. L o u i s a n d Bevyn J a r r o t t University of Melbourne, Clinical Pharmacolog),and Therapeutics Unit, Department of Medicine, Austin Hospital, Heidelberg~ Victoria 3084~Australia

Received I February 1990, revised MS rec~,~,ed29 May 1990, accepted 12 June 1990

The effect of ketanserin (3 mg/kg i.v.) on the baroreoeptor heart rate reflex and the Bezold-Jari~h reflex was examined in conscious Wistar-Kyoto rats OVKY) and spontaneously hypertensive rats (SHR). In the control situation (before ketanserin treatment), reflex bradycardia in response to phenylephrine (baroreflex) and phenyldiguanide (Bezold-Jarisch reflex)were impaired in SHR as compared with WKY, while reflex tachycardia in response to nitroprusside was similar in the two groups. However, after ketanserin administration in SHR, there was a reversal of the baroreflex-mediated tachycardia in response to nitroprusside into a bradycardic response. The nitroprusside-induced bradycardia was not caused by the release of 5-HT stimulating chemosensitive vagal afferents since the 5-HT3 receptor antagonist MDL 72222 did not block this response. In the same SHR, the Bezold-Jarisch reflex evoked by phenyldiguanide and the phenylephrine-induced bradycardia were potentiated by ketanserin. All the above effects of ketanserin were less evident in the WKY. Ketanserin did not alter vagal efferent function in anaesthetized SHR since it did not affect bradycardia induced by electrical stimulation of the vagus nerve. Therefore, it is suggested that ketanserin has sensitised cardiac vagai afferent mechanisms in SHR, which led to a normalization of reflex bradycardic function to a level normally observed in conscious normotensive WKY (i.e. prior to ketanserin treatment). Baroreceptor heart rate reflex; Bezold-Jarisch reflex; Ketanserin; Vagal afferents; Spontaneously hypertensive rats (SHR); Wistar-Kyoto rats OVKY)

1. Introduction Ketanserin has been widely used as an antihypertensive agent over the past decade although the complex pharmacology of this drug is still not fully understood (see Vanhoutte et al., 1988). The pharmacological classificaton of this drag is as a

A preliminary account of this work was presented to the 22nd Annual Meetingof the Australasian Society of Clinical and Experimental Pharmacologists,Adelaide, Australia, December 5-7, 1988. Correspondence to: R.E. Widdop, University of Melbourne, Clinical Pharmacologyand TherapeuticsUnit, Austin Hospital, Heidelberg, Victoria 3084, Australia

potent 5-HT 2 receptor antagonist which also blocks al-adrenoceptors (Leysen et al., 1981; Fozard, 1982; K.alkman et al., 1982). The contribution of the blockade of both these receptor sites to the antihypertensive effect of ketanserin is still controversial (Vanhoutte et al., 1986; Cohen et al., 1988; Elliot et al., 1988; Hedner and Perssen 1988). In addition, a centrally mediated sympathoinhibitory action of ketanserin, which may contribute to its therapeutic effect, has been described (McCall and Schuette, 1984; Ramage, 1985; McCall and Harris, 1987; Yoshioka et al., 1987; Matsumoto, 1988). Unlike non-selective a-adrenoceptor antagonists such as phentolamine and other peripheral vasodilators such as hydral~7~nc, ~ , n a l reflex

r0014-2999/90/$03.50 © 1990 Elsevier SciencePublishers B.V. (Biomedical Division)

18

tachycardia occurs in response to ketanserin-induced hypotension (see Vanhoutte et al., 1986). This is interpreted as being due to: (i) the lack of presynaptic a2-adrenoceptor blockade, and (ii) the centrally mediated blunting of baroreceptor-mediated reflex tachycardia (McCall and Harris, 1987; Van Zwieten, 1988). However in the few studies which have addressed this latter point, ketanserin was reported to have no effect on baroreflex function (Petersson et al., 1984; Woittiez et al., 1985; Berdeaux et al., 1987; Da~v et al., 1987; Elliot et al., 1988). The aforementioned studies can be criticized in that they usually examined only baroreflex activation, often using a single pressor dose, and this does not directly examine the processes involved in causing reflex tachycardia. The most comprehensive investigation of the effect of ketanserin on baroreflex function was performed by Smits et al. (1987), in that full baroreceptor activation and deactivation was examined in conscious rats using a large range of p~'essor and depressor responses, respectively. This study showed that ketanserin caused a dramatic conversion of nitroprusside (NP)-induced reflex HR respons~ from tachycardic into bradycardic responses. This phenomenon (hereafter called NP reversal) occurred in conscious spontaneously hypertensive rats (SHR), but not Wistar-Kyoto rats (WKY), and was claimed to be unique to ketanserin since the highly selective 5-HT2 receptor antagonist ritanserin did not cause NP reversal (Smits et al., 1988). While the lack of reflex tachycardia in response to NP is consistent with a sympathoinhibitory action of ketanserin (McCall and Schuette, 1984: Ramage, 1985; McCall and Harris, 1987; Yoshioka e~ al., 1987; Matsumoto, 1988), this does not explain the bradycardia which accompanies NP following ketanserin. The fact that the NP-induced bradycardic responses were blocked by atropine (Smits et al., 1987) suggests that ketanserin may have enhanced vagal tone. In the present study, we have confirmed this unusual effect of ketanserin on baroreflex function. Furthermore, we have also examined whether ketanserin can modify other vagal afferent mechanisms. This was achieved by eliciting a vagal cardiopulmonary reflex (Bezold-Jarisch reflex) in conscious SHR and WKY which is caused solely

by the activation of vagal afferent fibres (Thoren. 1979). Thus, phenyldiguanide (PDG) was used to evoke the Bezold-Jarisch reflex via the stimulation of 5-HT3 receptors on vagal afferents (see Widdop et al., 1990).

2. Materials and methods Female WKY and SHR (12-16 weeks; 180-240 g) were anaesthetized with a mixture of methohexitone sodium (32 mg/kg, Eli Lilly) and amylobarbitone sodium (60 mg/kg, Eli Lilly) administered i.p. (1 ml/kg) for the surgical implantation of an arterial and two venous cannulas into the abdominal aorta and right jugular vein, respectively. After surgery, the animals were allowed to recover in individual cages for 2-3 days during which arterial cannulas were flushed twice daily with minimal volumes of heparinised saline (100 U/ml). Baroreceptor heart rate (HR) and BezoldJarisch reflexes were then tested 2-3 days after surgery before and 20 min after the i.v. injection of either 3 mg/kg ketanserin tartrate or saline (1 ml/kg). The dose of ketanserin chosen was based on the study by Stairs et al. (1987). The duration of the experiments was approximately 3.5-4 h.

2.1. Baroreceptor HR reflex (baroreflex) testing Baroreflex activity was assessed in the unrestrained rats in a quiet room after acclimatisation to the laboratory conditions. Mean arterial pressure (MAP) was derived from the blood pressure signal which was measured from the abdominal aorta, using a pressure transducer (Gould P231D), and recorded on a polygraph (Grass, model 7C). H R was derived from the pressure signal using a tachometer (Grass model 7P44B). Pressor responses elicited by i.v. phenylephrine (PE; 1-25 ~ag/kg) and depressor responses elic:'.ed by i.v. NP (1-50/~g/kg) were given alternately by using the two venous cannulas. Doses were chosen randomly by varying the volume of the particular' drug, and this enabled multiple (20-30) injections to be given while avoiding a large cumulative injection volume ( < 0.2-0.4 ml in total). For the analysis of the baroreflex, the peak change~ in HR

19

in response to the bolus doses of either PE or NP were averaged for each experiment per 10 mm Hg range in MAP, as described previously ( S l i t s et al., 1987).

The effect of PDG was tested in WKY and SHR 10-20 min after baroreceptor reflex testing. After this equilibration period, bolus doses of 5, 10, 20 and 40 p g / k g PDG were injected i.v. in ascending order in a volume of 100 /tl/kg and were flushed in with a further 50 pl of normal saline (0.9% w / v NaCI). Thereafter, the reductions in diastolic blood pressure (DBP) and HR were measured. The lower doses of PDG were injected at approximately 4 rain intervals while 4-8 rain was allowed for larger doses to ensure the return of the baseline levels. Following the injection of either saline or ketanserin, the baroreflex and Bezold-Jarisch reflex tests were repeated.

HR and DBP versus log dose-response curves were generally best described by linear partitioning o~ th~ data. All dose-response curves were plotted as changes in H R or DBP against dose of PDG on a log scale. The group standard errors were calculated using the equation ~IEMS/n, where EMS is the error mean square from the ANOVA and n is the number of animals in each group. The significance of differences between pre- and post-ketanserin responses to each dose/frequency (figs. 5 and 6) or MAP grouping (figs. 2, 3 and 8) was determined by the incorporation of the EMS term for the overall ANOVA into the Student's modified t-test formula with the Bonferroni adjustment for multiple comparisons, as follows: t = x, - x21~EMS('In I + l l n 2 ) , where x I and x 2 are mean responses before and after ketanserin treatment at individual doses and nl and n 2 are the number of observations in each group (Wallenstein et al., 1980). The level of significance was taken as 0.05 divided by the number of comparisons.

2.3. Vagal stimulation

2.5. Drugs

Another group of SHR " -e anaesthetized with urethane (1.25 g / k g i.p.), tracheotomized and allowed to breath spontaneously, while the right femoral artery was cannulated to record blood pressure and HR. The right vagus nerve was sectioned distal to the nodose ganglion and the peripheral end was placed over bipolar platinum electrodes and immersed in mineral oil. Changes in HR in response to electrical stimulation of the vagus nerve at 1.5, 2.5, 5, 10, 15, 20 and 25 Hz for 10 s (5-15 V, 1 ms duration) were recorded.

The following drugs were used in this study: atropine sulphate (Astra); ketanserin tartrate (Janssen); MDL 72222 (Merrell Dow); phenylephrine hydrochloride (Koch-Light); phenyldiguanide (Aldrich); sodium nitroprusside (Nipride, Roche). All drugs were dissolved in saline and all doses refer to the salts of the drugs.

2.2. Bezold-Jarisch reflex testing

3. Results

3.1. Resting parameters and effect of ketanserin 2.4. Data analysis The effect of ketanserin on MAP and HR was compared to pre-injection values using Student's paired t-test. All the data from baroreflex, Bezold-Jarisch reflex and vagal stimulation experiments were subjected to an initial analysis of variance (ANOVA) with repeated measures (Snedecor and Cochran, 1980). Subsequent orthogonal partitioning of the sums of squares demonstrated that, for the Bezold-Jarisch reflex, the

The MAP and H R of WKY and SHR, measured during the acclimatization period and immediately before and after ketanserin (3 m g / k g i.v.), are listed in table 1. Initial resting MAP and H R values in SHR were significantly greater than those in WKY. Prior to the injection of either saline or ketanserin, resting MAP and HR had decreased in both WKY and SHR from initial values measured approximately 2 h earlier. This presumably represents the effects of the control

~S00

r

~0

II NP

@ NiP

F'~. L The ~fect of 3 mg/kg keta~sm~ Lv. ( K ] ~

~ KET

@ NP

tt NP

I ndn

on mrte~a] blood pressure (BP) and heart rate (Hit) responses to z~tropmsside

(NP) in a conscious unrestrained SHR.

cardiovascular reflex testing, as well as the m o r e settled phase of the animals to their s u r r o u n d i n g s d u r i n g this period. Indeed, similar changes were

8 120

observed in b o t h the time c o n t r o l (i.e. saline) a n d t r e a t m e n t (i.e. ketanserin) groups. T h e subsequent a d m i n i s t r a t i o n o f saline caused negligible changes

120

b SHR

WKY

• SAUNE

iT mT

.,,. SALINE

80

m

E o

80

Li/

40

0_-10

-20-,t0

-~o_.2o

"g

4o+.so -3o_.4o .so_-so

140 1

WKY

lOO

.. Ks'r

-10_-20

120

SHR

80

~. KET

°1~ 0_10

-20_-30 -40_-50 -10_-20 -30_-40 -50_40

A MAP (mmHg)

-40_-SO .r,o_,.eo

A MAP (mmHg)

L -20

-30_-40

d

I

=

,.~o=..,~o

0..-10

A MAP (mmHg)

C

b

40

0--10

11 II

It it -40_..+0 -20--30 .so_.4o .so .eo -10..-20

A MAP (mmHg)

Fig. 2. Changes in the heart rate (HR) in response to nitroprusside which have been grouped into decades of decreases in mean arterial pressure (MAP). Full and hatched columns refer to nitroprusside responses before and after treatment, respectively, with either saline in (~) WKY and (b) SHR or 3 mg/lcg ketanserin (KET) in (c) WilLY and (d) SHR. * P < 0.05 compared with pretreatment value; n -- 6 WKY and 7 SHR.

21 TABLE 1 Resting MAP and HR of conscious W K Y and SHR and the effect of 3 m g / k g ketanserin i.v. on these values. Values are means + S.E.M. of six W K Y rats and seven SHR. Strain Test period

MAP HR (mmHg) (beats/min)

W K Y Initial 105±3 354±13 Immediately befi~re ketanserin 9 9 ± 3 329±12 20 rain after ketanserm 8 9 ± 6 a 364±10 SHR

3.2. Baroreflex activity

Initial 1 4 3 ± 6 c 419± 9 c Immediately before ketanserin 128 ± 6 364± 10 20 min after ketanserin 104-1-5 b 364±17

a p < 0.05, b p < 0.001 compared with respective control value; c p < 0.001 compared with WKY.

in MAP and HR in both WKY and SHR (data not shown). In the WKY group, ketanserin caused a small but significant decrease in MAP which was accompanied by a variable increase in HR

a

which did not achieve significance. In the SHR group, ketanserin produced a larger hypotensive effect than in the WKY group. This was usually accompanied by an initial, transient bradycardic response, howe~/er when measured at 20 min, there was no change in HR despite the hypotensive effect of ketanserin (table 1).

Under control conditions, the i.v. injection of NP evoked dose-dependent, reflex tachycardic responses, in both WKY and SHR (figs. 1 and 2). For both groups, there were similar changes in reflex tachycardic responses evoked by NP over the experimental period in which saline was given instead of ketanserin (fig. 2a and b). In contrast, ketanserin caused a marked impairmenl of these responses in SHR (fig. 1), in that the reflex tachycardic responses were consistently converted

b

A MAP (mmH9) 0-10

10-20

20.30 30.40 40.50

o " ql/l

50.60

o

.cA -50

.SO

-100

-100

|

n,, -150 =C < -200

10-20

20-30 30-40 40-50 50.60

SHR

,~ -2O0

. SALINE

+ SAUNE

v~ -2,50

C

d

A M A P (mmHg) 0.10

i

10-22 22-a0 a04o

~.s0

o

.50

.50

-100

-100

v

-150

-150


.200

,m

0.10

-150 WKY

-250

--,

A M A P (mmHg)

A M A P (mmHg) 0.10

10-20 20-30 $0.40 40.s0

so-so

E m o .D

nZ

-250

WKY , KET

-2OO

SHR

. t

+ KET -250

Fig. 3. Changes in heart rate (HR) in response to phenylephrine which have been grouped into decades of increases in mean arterial pressure (MAP). Full and hatched c o h i m s refer to phenylephrine responses before and after treatment, respectively, with either saline in (a) W K Y and (b) SHR or 3 m g / k g ketanserin (KET) in (c) WKY and (d) SHR. * P < 0.05 compared with pretreatment value; n -- 6 W K Y and 7 SHR.

TABLE 2 Slope values of P D G dose-response curves for DBP a n d H R before a n d after e~ther saline or 3 m g / k g ketanserin i.v. in conscious W K Y and SHR. Values are means 4- S.E.M. Strain n

Treatment

M A P slope (mmHg/log dose P D G )

H R slope (beats r a i n - i / l o g dose P D G )

WKY

5

Pre-saline Post-saline

- 3 0 4 - 8.2 - 3 3 4 - 6.0

-1444-19,3 -644-13.6 a

6

Pre-ketanserin Post-ketanserin

- 3 9 + 4,5 - 2 5 4 - 4.4

- 9 5 + 14,2 -624-12.3

4

Pre-saline Post-saline

- 74 + 12.4 - 6 8 + 8.2

- 120 + 22.0 -98+26.2

7

Pre-ke~ Post-ketan~in

- 7 6 + 8.0 - 38 + 5.4 a

-124+14.5 - 117 + 13.9

SHR

The control i.v. injection of PE evoked dose-dependent reflex bradycardic responses in both WKY and SHR. Compared with the SHR group, these responses were larger in the WKY and there was also a sul~dl but significant increase in sensitivity with time in this group (fig. 3a and b). Following ketanserin administration, the reflex bradycardic responses in both groups were significantly enhanced, although this was more marked and occurred over a wider range of blood pressure increments in the SHR (fig. 3c and d).

3.3. Bezold-Jarisch reflex

a p < 0,05 compared with pretreatment value.

to bradycardic responses (hereafter called NP reversal). This NP reversal was observed over the complete range of MAP decrements tested (fig. 2d). In WKY, however, there was considerable variation in the NP-induced HR responses after ketanserin. Of the six animals tested, one exhibited complete NP reversal, two exhibited only tachycardic responses, while the remaining three animals showed a mixture of both tachycardic and bradycardic responses to separate NP injections following ketanserin treatment. The averaging cf these data resulted in si?:~ff~cant mean reductions compared with the control reflex tachycardic responses, but without consistent NP reversal (fig. 2c).

The i.v. injections of PDG evoked rapid dosedependent reductions in HR and DBP due to the excitation of cardiopulmonary vagal afferent fibres. The slopes of the mean dose response curves for HR and DBP in response to 5-40/tg/kg PDG in both time control (i.e. saline) and treatment (i.e. ketanserin) groups are listed in table 2. Following saline administration in the WKY group, the slope of the H R dose-response curve was significantly flatter thav control responses as a result of one concentration (20 /~g/kg) of PDG being significantly smaller than control. However, there were similar DBP responses in the WKY group, as well as similar H R and DBP responses and slopes of the dose-response curves in the SHR group following saline administration (table 2). In contrast, ketanserin caused a marked enhancement of the Bezold-Jafisch reflex responses (fig. 4). This was most evident for the PDG-induced HR responses

i= v

E Z

|

v Cb m

loo

® lO PDG

®

T

20 PDG

KET

0 10 PDG

•

1 rain

20 PI)G

Fig. 4. Arterial blood pressure (BP) and heart rate (HR) responses to phenyldiguanide (PDG; 10 and 2 0 / t g / k g i.v.) in a conscious unrestrained S H R before and after 3 m g / k g ketanserin (KET).

23 WKY

SHR

PDG (l~g/kg,i.v.)

PDG (.ug/kg,i.v.)

5

10

20

40

5

10

20

40

I

I

I

I

I

!

!

I

0

0

13PDG R 4.KET

13 l O G

I ÷ KET

A

3: E E

v

Q. Ia <:l

-20

-20

40-

.40.

-60-

-60o

-80

-80.

:)

PDG (~g/kg,l.v.) 5

10

20

40

I

I

I

I

o

PDG (~g/kg,i.v.)

5

10

20

40

I

I

!

I

-so

i

-10

~

-100

~:<1 -150] -200

-150 -200

,

-

,

Fig. 5. Mean dose-response curves of diastofic blood pressure (DBP) and heart rate (HR) responses to phenyldiguanide (PDG) in six WKY (left panels) and seven SHR (fight panels). Open and closed symbols refer to pre- and post- 3 mg/kg ketanserin (KET) values, respectively. * P < 0.05 compared with control value

in the SHR, as seen by the leftward shift in the dose-response curve (fig. 5) without any change in the slopes (table 2). A similar, but less marked effect of ketanserin on HR was seen in the WKY group, while potentiated DBP responses were only observed in the $HR group (fig. 5).

3.4. Vagal stimulation Peripheral vagal stimulation produced frequency-dependent bradycardic responses in anaesthetized SHR. Ketanserin, 3 mg/kg i.v., did not affect these responses when tested 20 min later (fig. 6, table 3).

1.25 I

2.5 I

5 I

10

20

40

I

I

I

c -50 ~mE.lo0 I

.~ -150 rr -r -200 <) -250 -300

-KET

PRE-KET

Fig. 6. The effect of ketanserin i.v. (KET) on heart rate (HR) responses to stimulation of the vagus nerve in four anaesthetized $HR at varying frequencies (1.5-25 Hz).

24

z E

v L m

o

@

NP

~

@

NP

o

20POG T

KET

@

20POG

MDL

,

@

NP ,

~

HP

ATROPINE

I ndn

Fig. 7. The effect of 3 mg/kg i.v. ketanserin (KET), 0.5 m g / k g i.v. MDL 72222 (MDL), and 0.5 m g / k g atropine i.v. on arterial blood pressure (BP) and heart rate (HR) responses to nitroprusside (NP) and phenyldiguanide (PDG) in a conscious unrestrained SHR.

[]

~IP

TABLE 3

[] ÷~'T t ÷l
Changes in H R (beats/min) resulting from electrical stimulation (1 ms, 5-15 V, 10 s) of the peripheral vagus nerve before (control) and after 3 m g / k g i.v. ketanserin (KET) in anaestho. tized SHR. Values are means + S.E.M. of four experiments.

÷WOL

Frequency -v ~

Control

KET

-18+ 8 -38+14 -974-24 - 159+ 33 -194+31 - 223 + 31 - 229 + 25

-13+ 3 -43+ 8 -93+28 - 125 + 14 -!86+28 - 201 :l: 27 - 194 + 24

(Hz)

Q

0

n:z: -40 <]

-80"

~ .

* -10--20

-20 -30

-30_-40

1.5 2.5 5 10 15 20 25

A MAP (mmHg) Fig. 8. Changes in heart rate (HR) in response to nitroprusside (NP), which have been grouped into decades of decreases in mean arterial pressure (MAP), before (full column) and after 3 m g /k g ketanserin (hatched column) alone, or in combination with 0.5 mg/kg MDL 72222 (MDL) in six SHR. * P < 0.05 compared with control NP response.

injections of NP were repeated. In six out of seven animals the NPoinduced bradycardic responses were unaffected by MDL 72222 (fig. 8), although these responses were blocked in the one remaining animal. When atropine (0.5 m g / k g i.v) was given following M D L 72222, this completely blocked the NP reversal (fig. 7).

3.5 Effect of MDL 72222 on NP reversal The ability of the 5-HT3 antagonist MDL 72222 to interfere with the ketanserin-induced NP reversal was examined in a further seven conscious SHR to determine whether NP was releasing ~,-l-ii. In these animals ketanserin (3 mg/kg) converted the NP-induced reflex tachycardia into bradycardic responses, as already described. MDL 72222 was then injected i.v. in a dose (0.5 mg/kg) that was sufficient to completely block the response to 10-20 p g / k g PDG (fig. 7). Ten minutes later, the

4. Discussion

The present study has confirmed the findings ot Smits et ai. (1987) in that the reflex tachycardia in response to NP was converted into bradycardia (NP reversal) in the presence of ketanserin in conscious SHR. Additionally, it was also found in the present study that, in the same animals, there were enhanced reflex bradycardic responses to PE and enhanced Bezold-Jarisch reflex responses to

25 PDG. All of these effects are consistent with ketanserin having a facilitatory action on vagal function. However this does not involve vagal efferent function since ketanserin did not affect the bradycardia evoked by electrical stimulation of the vagus in anaesthetized SHR. The effect of ketanserin to evoke NP reversal was reported to be specific for this drug, since neither prazosin, hydralazine, methysergide nor cyproheptadine reversed the NP-induced tachycardic responses (Smits et al., 1987). In addition, the highly selective 5-HT2 receptor antagonist ritanserin, either alone or in combination with prazosin, failed to evoke NP reversal (Smits et al., 1988). In contrast to the SHR group, there was no consistent NP reversal in the WKY group, as has been described previously (Smits et al., 1987), although occasional NP reversal was observed. Consistent with the NP data, ketanserin itself cause variable, insignificant increases in HR in the WKY group while there was no change in HR in the SHR group, despite ~ greater hypotensive effect in the latter group. Of the few studies which have examined the effect of the ketanserin on reflex tachycardia, no changes in baroreflex deactivation were reported in normotensive subjects (Berdeaux et al., 1987) or normotensive rats (Davy et al., 1987). However, since these previous studies were based on single doses of NP given under normotensive conditions, the previous findings are therefore not surprising in light of the selectivity of the NP reversal phenomenon for the hypertensive state. In the present study, control (i.e. pre-ketanserin) reflex bradycardia in response to PE was impaired in the SHR as compared with the WKY group (fig. 3); a phenomenon which is well documented (Ricksten and Thoren, 1981; Struyker-Boudier et al., 1982; Head and Adams, 1988; Widdop et al., 1990). Following ketanserin treatment, there were enhanced reflex bradycardic responses especially in the SHR group. Indeed, the post-ketanserin responses in the SHR now approximated the control responses in the WKY group (see fig. 3). Several previous clinical and experimental studies reported no change in baroreflex activation following ketanserin (Pettersson et al., 1984; Woittiez et al., 1985; Berdeaux et al., 1987; Davy et al.,

1987; Elliot et al., 1988), however, comparisons with the present study are difficult since the experimental conditions of the fore-mentioned studies differed considerably with respect to dosage, length of ketanserin treatment and the presence or absence of hypertension. In the present study, no attempt was made to explore other doses of ketanserin. Nevertheless, our findings also differed somewhat from those in the analogous study by Smits et al. (1987), in that they observed that ketanserin did not modify the reflex bradycardia using angiotensin II (AII) as the pressor agent. In the present study, ketanserin caused a moderate blockade of PE-induced pressor responses ( -- 3- to 5-fold), although pressor responses before and after ketanserin were always matched. It has been shown that, for similar pressor responses induced by PE and All, AII selectively attenuates the baroreflex control of HR (Guo and Abboud, 1984; Matsumura et al., 1989). Thus, it is possible that smaller reflex bradycardic responses produced by AII per se may have masked any vagal facilitation produced by ketanserin in this previous study (Smits et al., 1987). Cardiopulmonary vagal afferent function was also tested in the same animals since vagal activation by ketanserin in the NP reversal phenomenon was previously implicated as NP reversal was blocked by atropine (Smits et al., 1987). For this purpose, PDG was used to activate vagal afferents resulting in the characteristic Bezold-Jarisch reflex (Thoren, 1979). We have recently reported that Bezold-Jarisch reflex evoked by PDG was impaired in conscious SHR as compared with WKY (Widdop et al., 1990). Similarly, in the present study, PDG responses prior to ketanserin administration were attenuated in the SHR group. However, in the SHR, ketanserin potentiated the Bezold-Jarisch reflex responses such that the HR dose-response curve was shifted to the left, to a similar position as that observed for the control HR dose-response curve in the WKY group (see fig. 5). This enhancement of PDG responses in the SHR produced by ketanserin was similar to that observed with baroreflex activation. Thus, in the same group of SHR, ketanserin evoked NP reversal and normalised reflex bradycardic function (both baroreceptor and cardio-

26 pulmonary receptor reflexes) which otherwise is impaired in this hypertensive strain. All of these effects are consistent with ketanserin causing an increase in vagal tone (Fozard, 1982). This is not due to the modification of vagal efferent function since in the present study peripheral vagal nerve stimulation was unaffected by ketanserin. Rather, it is more likely that the drug has sensitized vagal afferent mechanisms. The cardiopulmonary vagal afferents responsible for activation of cardiopulmonary reflexes are sensitive to changes in cardiac volume (mechanosensitive) or to chemical stimulation (chemosensitive). Thus it is possible that PDG activates a subpopulation of vagal afferents which is distinct from those activated by volume expansion (Baker et al., 1979). In this context, ketanserin-induced NP reversal resembles the paradoxical decrease in both HR and renal sympathetic nerve activity observed during haemorrhage in cats and rats (Oberg and Thoren, 1972; Skoog et al., 1985). While NP reversal and haemorrhage are different haemodynamic events. the haemorrhagic effects were attributed to a preceding increase in the firing rate of mechanosensifive cardiac vagal afferents (Oberg and Thoren, 1972). In the present study, NP reversal was not blocked by the selective 5-HT3 receptor antagonist MDL 72222, which suggests that any release of 5-HT by NP to stimulate chemosensitive vagal afferents was not involved in this phenomenon. Thus, in addition to the sensitization of chemo (i.e. PDG)-sensitive cardiac vagal afferents by ketanserin, the present findings are consistent with the view that PDG-insensitive and/or mechanosensitive cardiac vagal afferents may be similarly affected and thus contribute to NP reversal. While the present findings cannot entirely discount a centrally mediated facilitatory action of ketanserin on vagal tone, this seems unlikely for several reasons. Firstly, both ketanserin and prazosin produce similar central effects, e.g. sympathoinhibition via al-adrenoceptor blockade (McCall and Schuette, 1984; McCall and Harris, 1987; Ramage, 1985), however these drugs exert markedly different effects on baroreflex function since prazosin does not cause NP reversal (Smits et al., 1987; 1988). Secondly, vagal aff~ent activation by ketanserin was implicated in an earlier

study, since vagotomy abolished ketanserin-induced bradycardia in rats while atropine only partially reversed this effect (Fozard, 1982). Thus, the residual bradycardia represents the effect of sympathetic withdrawal in response to vagal afferent activation. Finally, ketanserin occasionally causes a first doze phenomenon in humans involving marked hypotension resulting from the stimulation of vasovagal mechanisms (Waller et al., 1987; Semple et al., 1988). In contrast, Yoshioka et al. (1987) measured aortic depressor nerve activity in anaesthetized rats and found that ketanserin actually decreased aortic baroreceptor afferent nerve activity, when measured 10 min after injection. However, no information concerning the initial effects and time course of this response was reported. In the present study, the vagal facilitatory effect of ketanserin was more evident in SHR compared with WKY. In addition, when compared with WKY rats, we have previously observed greater impairment of Bezold-Jarisch reflex responses in conscious as compared with anaesthetized SHR (Verberne et al., 1988; Widdop et al., 1990). Therefore, these factors, together with a lack of information concerning the influence of ketanserin on cardiac vagal afferent nerve activity, may have contributed to the discrepancies between this and the previous (Yoshioka et al., 1987) study. It is now well recognised that, both in the clinical and experimental setting, vagal afferent modification may contribute to the haemodynamic profile of a range of compounds (Semple et al., 1988). In particular, there is good evidence that both atrial natriuretic peptide (ANP) and arginine vasopressin (AVP) sensitize vagal afferent mechanisms. ANP lowers blood pressure, heart rate and renal sympathetic nerve activity in anaesthetized rats due to vagal afferent activation (Thoren et al., 1986; Imaizumi et al., 1987; Schultz et al., 1988). Furthermore, ANP enhanced PE-induced bradycardia and reduced NP-induced tachycardia in conscious rats (Ferrari et al., 1988). On the other hand, AVP facilitates the inhibition of renal nerve activity evoked by volume expansion in sinoaorticdenervated anaesthetized rabbits (Gupta et ai., 1987), while, in conscious monkeys, AVP potentiates the Bezold-Jarisch reflex evoked by veratri-

27

dine (Barazanji ahd Cornish, 1989). Similarly, captopril has recently been shown to augment the Bezold-Jarisch reflex evoked by veratridine in conscious dogs (Panzenbeck et al., 1988). Thus, the effects of ketanserin in the present study are consistent with the sensitization of cardiovascular reflex mechanisms by other agents known to activate vagal afferent mechanisms, and provide further, indirect evidence for the sensitization of vagal afferents by ketanserin. In conclusion, we have demonstrated that vagal afferent modification is another factor Which should be considered as part of the complex pharmacology of ketanserin (see Introduction). This, together with a central sympathoinhibitory action may explain the lack of baroreflex-mediated tachycardia with hypotensive doses of ketanserin.

Acknowledgements This study was supported by a Programme Grant from the National Health and Medical Research Council of Australia. We would like to thank Janssen-Cilag and Merrell Dow Research Institute for their generous gifts of ketanserin and MDL 72222, respectively. We also wish to thank Ms. L. Crockford for her secretarial assistance in the preparation of this manuscript.

References Baker, D.G., H,M. Coleridge and J.C.G. Coleridge, 1979, Vagal afferent C fibres from the ventricle, in: Cardiac Receptors, eds. R. Hainsworth, C. Kidd and R.J. Linden (Cambridge University Press, London) p. 117. Barazanji, M.W. and K.G. Cornish, 1989, Vasopressin potentiates ventricular and arterial reflexes in the conscious nonhuman primate, Am. J. Physiol. 256, HI546. Berdeaux, A., A. Edouard, K. Samii and J.-F. Giudicelli, 1987, Ketanserin and the arterial baroreceptor reflex in normotensive subjects, European J. Clin. Pharmacol. 32, 27. Cohen, M.L., K.W. Schenck and K.D. Kurz, 1988, 5-HT2-receptor antagonists: at- vs. 5-HT2-receptor blocking properties in blood vessels, J. Cardiovasc. Pharmacol. 11 (Suppl. 1), $25. Davy, M., M. Midol-Monnet, M. Heimburger, F. Beslot and Y. Cohen, 1987, Peripheral and central cardiovascular effects of ketanserin in conscious normotensive rats, Arch. Int. Pharmacodyn. Ther. 290, 193.

Elliot, H.L., R. Donneily and J.L. Reid, 1988, Effects of ketanserin on peripheral vascular pressor mechanisms in essential hypertension, J. Cardiovasc. Pharmacol. 11 (Suppl. 1), $62. Ferrari, A.U., A. Daffonchio0 A. Cavallazzi, S. Gerosa, G. Napoletano and G. Mancia, 1988, Effect of atrial natriuretic factor on arterial baroreceptor control of heart rate and blood pressure in conscious rats, J. Hypert. 6 (Suppl. 4), $284. Foza~'d, J.R., 1982, Mechanism of the hypotensive effect of ketanserin, J. Cardiovasc. Pharmacol. 4, 829. Guo, G.B. and F.M. Abboud, 1984, Angiotensin il attenuates baroreflex control of heart rate and sympathetic activity, Am. J. Physiol. 246, H80. Gupta, B.N., A.L. Abboud, J.S. Floras, P.E. Ayiward and F.M. Abboud, 1987, Vasopressin facilitates inhibition of renal nerve activity mediated through vagal afferents, Am. J. Physiol. 253, HI. Head, G.A. and M.A. Adams, 1988, Time course of changes in baroreceptor reflex control of heart rate in conscious SHR and WKY: contribution of the cardiac vagus and sympathetic nerves, Clin. Exp. Pharmacol. Physiol. !5, 289. Hedner, T. and B. Persson, 1988, Experience with ketanserin and ritanserin in hypertensive patients, J. Cardiovasc. Pharmacol. 11 (Suppl. 1), $44. Imaizumi, T., A. Takeshita, H. Higashi and M. Nakamura, 1987, a-ANP alters reflex control of lumbar and renal sympathetic nerve activity and heart rate, Am. J. Physiol. 253, H1136. Kalkman, H.O., P.B.M.W.M. Timmermans and P.A. Van Zwieten, 1982, Characterization of the antihypertensive properties of ketanserin (R41468) in rats, J. Pharmacol. Exp. Ther. 222, 227. Leysen, J.E., F. Awouters, L. Kenis, P.M. Laduron, J. Vanderberk and P.A.J. Janssen, 1981, Receptor binding pr,~file of R 41 468~ a novel an!agonist at 5-HT2 receptors, Life Sci. 28, 1015. Matsumoto, M., 1988, Effect of ketanserin on adrenal sympatbetic nerve activity in rats, Jap. J. Pharmacol. 48, 57. Matsumura, Y., E.M. Hasser and V.S. Bishop, 1989, Central effect of angiotensin II on baroreflex regulation in conscious rabbits, Am. J. Physiol. 256, R 694. McCall, R.B. and L.T. Harris, 1987, Characterization of the central sympathoinhibitory action of ketanserin, J. Pharmacol. Exp. Ther. 241, 736. McCall, R.B. and M.R. Schuette, 1984, Evidence for an alpha-I receptor-mediated central sympathoinhibitory action of ketanserin, .,I. Pharmacol. Exp. Ther. 228, 704. Oberg, B. and P. Thoren, 1972, Increased activity in left ventricular receptors during haemorrhage or occlusion of caval veins in the cat. - a possible cause of the vaso-vagal reaction, Acta Physiol. Scand. 85, 164. Panzenbeck, M.J., W. Tan, M.A. Hajdu and I.H. Zucker, 1988, Prostaglandins mediate the increased sensitivity of left ventricular reflexes after captopril treatment in consc:,ous dogs, J. Pharmacol. Exp. Ther. 244, 384. Pettersson, A., B. Persson, M. Henning and T. Hedner, 1984,

2$ A~t~ypertensi~e effects of chronic 5-hydroxytryptamine {5-HT2) receptor blockade with ketanserin in the sponta~ y hypert_en~ve rat, Naunyn-Schmiedeb. Arch. PharnmcoL 327, 43. A.G., 1985, The effects of ketanserin, methysergide and LY 53857 on sympathetic nerve activity, European J. ~ . 113, 295. ~ten, S.E. and P. Thoren" 1981, Reflex control of sympathetic nerve activity and heart rate from arterial baroreceptors in conscious spontaneously hypertensive rats, Clin. Sci. 61, 169s. Schultz, H.D., D.G. Gardner, C.F. Deschepper, H.M. Colen ~ and LC.G. Coleridge, 1988, Vag~! C-fibre blockade abolishes sympathetic inhibition by atrial natriuretic factor, Am. j. PhysicS. 255, R6. Semple, P.F., P. Thoren and A.F. Lever, 1988, Vasovagal reactions to ~ , ~ v a s c u l a r dru~- the first dose effect,. J. H~ert. 6, 601. Skoog, P., L Mansson and P. Thoren, 1985, Changes in renal sympathetic outflow during hypotensive haemorrhage in rats, Acta Physiol. Scand. 125, 655. Smi~s, .L, F. Van Dorsten and H. Struyker-Boudier, 1988, Interference of ketanserin with baroreflex control of the circulation in the conscious spontaneously hypertensive rat, Drugs 36 (Supp~ 1), 55. Stairs, £F.M., H. Van Essen, C.M. Tijssen and H.A.J. Struyker-Boudier, 1987, Effects of ketanserin on haemodynamics and baroreflex effects in conscious spontaneously hyperten~ve rats, L Cardiovasc. Pharmacol. 10, 1. Snedecor, G.W. and W.G. Cochran, 1972, Statistical Methods (Iowa State University Press, Ames). Struyker-Boudier, H.A2., R.T. Evenwell, J.F.M. Smits and H. Van Essen" 1982, Baroreflex sensitivity during the developmerit of spontaneous hypertension in rats, Clin. Sci. 62, 589. Thoren, P., 1979, Role of cardiac vagal C-fibres in cardiovascular control, Rev. Physiol. Biochem. Pharmacol. 86, 1. Thore~ P., A.L. Mark, D.A. Mor~an~ T.P. O'Neill, P. Needleman and MJ. Brody, 1986, Activation of vagai depressor

reflexes by atriopeptins inhibits renal sympathetic nerve activity, Am. J. Physiol. 251, H 1252. Vanhoutte, P., A. Amery, W. Birkenhager, A. Breckenridge, F. Buhler, A. Disfler, J. Dormandy, A. Doyle, E. Frolflich, L, Hansson, T. Hedner, N. Hollenberg, H.-E. Jensen, P. Lund-Johansen, P. Meyer, L. Opie, I. Robertson" M. Safar, M. Schalekamp, J. Symoens, J. Trap-Jensen and A. Zanchetti, 1988, Serotoninergic mechanisms in hypertension. Focus on the effects of ketanserin, Hypertension 11, 111. Vanhoutte, P.M., S.G. Ball, A. Berdeaux, M.L. Cohen, T. Hedner, It. McCall, A.G. Ramage, I.H. Reimann" C. Richer, P.R. Saxena, M.A.D.H. Schalekamp, H.AJ. StruykerBoudier, J. Symoens, J.M. Van Nueten and P.A. Van Zwieten" 1986, Mechanism of action of ketansefin in hypertension, Trends Pharmacol. Sci. 7, 58. Van Zwieten" P.A., 1988, Basic pharmacology of alpha-adrenoceptor antagonists and hybrid drugs, J. Hyper~. 6 (Suppl. 2), S3. Verberne, AJ.M., N.A. Young and W.J. Louis, 1938, Impairment of inhibitory cardiopulmonary vagal reflexes in spontaneously hypertensive rats, J. Auton. Nerv. Syst. 23, 63. Wallenstein, S., C.L. Zucker and J.L. Fleiss, 1980, Some statistical methods useful in circulation research, Circ. Res. 47, 1. Waller, P.C., H.A. Cameron and L.E. Ramsay, 1987, Profound hypotension after the first dose of ketanserin, Postgrad. Med. J. 63, 305. Widdop, R.E., A.J.M. Verberne, B. Jarrott and W.J. Louis, 1990, Impaired arterial baroreceptor reflex and cardiopulmonary vagal reflex in conscious spontaneously hypertensive rats, J. Hypert. 8, 269. Woittiez, A.J.J., G.J. Wenting, A.J. Man in "t Veld, F. Boomsma and M.A.D.H. Schalekamp, 1985, Ketanserin: a possible tool for studying the role of serotonin in hypertension, J. Cardiovasc. Pharmacol. 7 (Suppl. 7), S130. Yoshioka, M., M. Matsumoto, H. Togashi, M. Minami and H. Saito, 1987, Central sympathoinhibitory action of ketanserin in rats, J. Pharmacol. Exp. Ther. 243, 1174.