European Journal of Pharrnacology, 213 (1992) 227-233 © 1992 Elsevier Science Publishers B.V. All rights reserved 0014-2999/92/$05.00

227

EJP 52352

Effect of DL-propranolol on nitric oxide production in perfused rat hindquarters H c l d e r M o t a - F i l i p e a, M a t i l d e C a s t r o b a n d Jos~ M. G i f i o - T . - R i c o c c Instituto de Farmacologia, Faculdade de Medicina de Lisboa, b Laboratdrio de Quimica Analitica, Faculdade de Farmdcia de Lisboa and Laborat6rio de Farmacologia, Faculdade de Farmdcia de Lisboa, 1600 Lisboa, Portugal

Received 11 July 1991, revised MS received 16 October 1991, accepted 24 December 1991

The effect of DL-propranolol on NO release in perfused rat hindquarters was studied by using oxyhemoglobin as a capture system to allow the quantitative assay of NO production. In some experiments the stable prostacyclin metabolite 6-keto-PGF~, (6-keto) was simultaneously assayed. We observed that: (1) DL-propranolol induced an increase in NO and 6-keto release. The dextro isomer was inactive; (2) DL-propranolol-induced NO release was only slightly reduced by acetylsalicylic acid in a concentration that inhibits prostacyclin synthesis, and was abolished by the chemical removal of the endothelium with 3-3 cholamidopropyl dimethylammonium 1-propane sulphonate (CHAPS) applied before or during stimulation; (3) NG-nitro-Larginine blocked DL-propranolol-induced NO production, an effect that was antagonized by L-arginine but not by its dextro isomer.

DL-Propranolol; Nitric oxide (NO); EDRF (endothelium-derived relaxing factor); NG-Nitro-L-arginine; CHAPS (3-3 cholamidopropyl dimethylammonium 1-propane sulphonate); Hindquarters (rat)

I. Introduction

The role of the endothelium in vascular dilator responses to many vasoactive compounds such as acetylcholine (Furchgott and Zawadzki, 1980), histamine (Van de Voorde and Leusen, 1983), 5-hydroxytryptamine (Richard et al., 1990), substance P (Rees et al., 1989), bradykinin (Gryglewski et al., 1986), the calcium ionophore A23187 (Rees et al., 1989) etc. has been firmly established. This dependence may be accounted for by the liberation of a very potent vasodilating factor, e n d o t h e l i u m - d e r i v e d relaxing factor (EDRF). This factor has been identified, based on considerable convergent experimental evidence, as nitric oxide (NO), which is derived from the guanidino moiety of L-arginine (Ignarro, 1990; Palmer et al., 1988a). Other non-prostanoid vasodilator compounds may also exist (F6rstermann et al., 1985; Vanhoutte, 1987; Angus and Cocks, 1989). The identification of E D R F with NO is based mainly on the suppression of the in vitro and the in vivo vasodilation induced by L-arginine analoguos that block the synthesis of NO

Correspondence to: H. Mota-Filipe, Instituto de Farmacologia, Faculdade de Medicina de Lisboa, 1600 Lisboa, Portugal. Tel. 351.1.797 9407.

(Palmer et al., 1988a; Schr6der et al., 1990; Moore et al., 1990; Rees et al., 1990; Miilsch and Busse, 1990; Gardiner et al., 1991) and by the scavenging of NO by oxyhemoglobin (Palmer et al., 1988b; Kelm et al., 1988), resulting in the formation of methemoglobin. This reaction is very fast and is used for the spectrophotometric assay of NO (Kelm et al., 1988). Moncada et al. (1988) have also shown the pharmacodynamic and pharmacokinetic parallelism of the effects of E D R F and NO. The aim of the present study was to investigate the effect of an antihypertensive compound, the /3-adrenoceptor blocker DL-propranolol, on NO production in perfused rat hindquarters. In previous experiments Gifio-T.-Rico et al. (1988) have shown that this compound stimulates the production of prostacyclin, as assessed by measuring its stable metabolite 6-ketoP G F ~ (6-keto), in perfused rat hindquarters. As NO shares and potentiates several of the effects of prostacyclin (Shimokawa et al., 1988; Doni et al., 1988; Moncada et al., 1988), namely on peripheral vascular tone and platelet aggregation, it seemed interesting to study the possible effect of DL-propranolol on NO production in a peripheral vascular tissue. Thus in some experiments prostacyclin production was assayed simultaneously in order to correlate both phenomena.

228 2. Materials and methods

2.1. Hindquarters perfusion Male Wistar rats (250-300 g) were used. The animals were stunned and exsanguinated, and the hindquarters were perfused between the aorta and the inferior vena cava, both cannulated below the renal vessels without recycling of the perfusing solution. The perfusion solution was Krebs-Henseleit containing (raM): NaC1 113; KC! 4.7; CaC12 2.55; MgSO 4 1.12; NaHCO 3 24.76; KI--I2PO4 1.19; glucose 11.56. All reagents were analytical grade and were dissolved in borosilicate glass-distilled water. Oxyhemoglobin (4 ~M) was added to the Krebs solution (Krebs-OzHb) and the solution was oxygenated with 95% 0 2 + 5% CO 2. Both the perfusion solution and the preparation were kept at 37°C. The Krebs-O2Hb solution was perfused at 2 ml/min, using a Harvard peristaltic pump. The experiment began after 20 min of perfusion. A basal sample of effluent was then collected before the drugs were added to the Krebs-OzHb reservoir. Twenty-minute samples were collected afterwards for the assay of methemoglobin (metHb) and the stable prostacyclin metabolite 6-keto-PGFla (6-keto). As hemoglobin is known to inhibit the endothelium-dependent vascular responses pressure records were not made.

2.2. Oxyhemoglobin preparation Oxyhemoglobin was prepared using the method Kelm and Schrader (1988). In brief, a molar excess sodium dithionite was added to a solution hemoglobin. Oxyhemoglobin was purified on Sephadex G25 column and diluted in Krebs solution a final concentration of 4/xM.

of of of a to

2.3. Methemoglobin (metHb) assay The amount of metHb present in the Krebs-O2Hb solution and in the effluent was assayed by the spectrophotometric technique described by Harrison and Jollow (1986).

2.4. 6-keto-PGFl, assay

the addition of drugs to the perfusing solution was measured in each period. The amount of metHb is given as a percentage of the total Hb present in the Krebs solution. The results are plotted against time. The overall effect of the drugs was also obtained from the geometrically calculated area under the curve (AUC). Control and experimental conditions were compared with a two-tailed unpaired Student's t-test. The results were considered to be statistically significantly different when P < 0.05. The results are given as means + S.E.M. The number of observations is shown in parentheses.

2.6. Drugs used Hemoglobin bovine Type II, carbachol chloride, atropine sulphate, acetylsalicylic acid, 3-3 cholamidopropyl dimethylammonium-l-propane sulphonate (CHAPS), NC-L-nitroarginine, L-arginine and Darginine were from Sigma. DL-Propranolol and D-propranolol were from ICI. All other reagents were analytical grade. 3. Results

Several experiments were performed under basal conditions in order to: (1) measure the stability .of oxyhemoglobin in oxygenated Krebs solution under our experimental conditions; (2) observe the influence of the erythrocytes eventually present in the vena cava effluent on the metHb assay; (3) determine the basal production of NO up to 120 rain; (4) evaluate the response to cholinergic stimulation, using carbachol as an example of a muscarinic agonist resistant to cholinesterases.

3.1. Stability of oxyhemoglobin in Krebs solution As a routine procedure, the amount of metHb in the Krebs-O2Hb, as a percentage of the total amount of Hb, was measured at the beginning and at the end of each experiment. No difference was observed: metHb 23.13 + 0.84% (70) in the beginning, 23.24 +_ 0.94% (70) P > 0.2, after 80 min, 26.70 _+ 0.20% (6), P > 0.2 after 120 rain.

3.2. Influence of erythrocytes

6-keto-PGFl~ (6-keto) was assayed by RIA after extraction, using the Amersham assay system (% cross reactivity: PGA 1 < 0.014; PGA 2 0.014; PGD 2 < 0.014; PGE 2 5.1; PGF2~ 0.3 TxB 2 < 0.014).

The influence of erythrocytes was studied by assaying the metHb in a sample collected immediately after the beginning of perfusion and comparing the result with the later basal measurements. Erythrocytes had no influence.

2.5. Statist&al analysis

3.3. Basal production of NO

The difference between the amount of metHb present in the effluent under basal conditions and after

The basal production of nitric oxide declined slightly with time, but under our experimental conditions no

229

statistical difference from zero was observed. At the beginning of perfusion, the amount of metHb in the Krebs-O2Hb was 23.13 +0.87% (70) and 25.87+ 1.00% (70) (P > 0.2) in the vena cava effluent. The time course of the basal production of NO, as given by the difference in metHb content of the effluent, with respect to the initial value is shown in fig. 1.

~ A

metHb %

--A

6-keto-PGFllllpha ng/ml

401 8O

6

[ DL-PROPRANOLOLO., ~ / - 1

'°1 10~

/"

S'L /

I

~

4 o___o,o..,u.-, N-l,

*'I*

-4

***

3.4. Effect of carbachol on NO production The validity of the method was also tested using the well-known response to cholinergic stimulus and its blockade by atropine. Carbachol (10/xM) was added to the Krebs-O2Hb solution. The results are shown in fig. 1. When we attempted to block the response with atropine, this drug was present in the Krebs-O2Hb solution (1/zM) from the beginning of the experiment, and carbachol (10 tzM) was added 20 min later. The continuous presence of carbachol steadily increased the amount of metHb in the effluent and this effect was completely blocked by atropine (fig. 1).

3.5. Effect of DL-propranolol Figure 2 shows the response to DL-propranolol (0.33 /zM). The initial increase in metHb concentration in the effluent was faster than with carbachol but the concentration stabilized after about 60 min. Figure 2 also shows the 6-keto response after DLpropranolol. The basal release of prostacyclin, as evaluated with its stable metabolite 6-keto, was 0.70 ± 0.07 (63) ng/ml and tended to decrease during the experiment, which explains the negative value for the area under the curve (-0.71 + 1.40; N = 3). The adrenocep-

~ A

metHb

%

40-

80 -

BACHO

/ ~

N-8

-lO

I

,

,

0

20

40

":"

60

80

-

100

MINLITF8

Fig. 2. Comparison between DL- and D-propranolol on NO and 6-keto-PGFl~ production, with simultaneous assay of both metabolites. Mean + S.E. * * * P < 0.001 relative to the changes in the m e t H b content of the effluent, +++ P < 0.001 relative to the changes in the 6-keto-PGFt, values, in both cases as compared with the initial values. Areas under the curve. NO production: DL-propranolol 1683.24_+ 165.71 (N = 17) P < 0.001; D-propranolol 309.50_+ 106.47 (N = 4), N.S. to the control P < 0.001 relative to DL-propranolol: 6-keto production: control -0.71_+ 1.40 (N = 3); DL-propranolol 31.04_+5.76 ( N = 9 ) P < 0 . 0 0 1 relative to control; D-propranolol -2.63_+ 2.20 (N = 4), N.S. compared to control, P < 0.001 compared to DL-propranolol.

tor fl-blocker significantly increased the production of 6-keto. The effect of DL-propranolol on nitric oxide production was studied in more detail, using several approaches: (1) the comparison of DL- and D-propranolol; (2) the study of the effect of atropine on the response to DL-propranolol; (3) the possible influence of the simultaneous production of prostacyclin, using acetylsalicylic acid (1.6 /zM) to block its synthesis; (4) the importance of the integrity of the endothelium, using 3-3-cholamidopropyldimethylammonium-l-propane sulphonate (CHAPS) (8 mM) to remove the endothelium; (5) the dependence of the response to Larginine by studying the effect of this amino acid alone (300 /zM) and after blockade of its metabolism by NC-nitro-L-arginine (L-NOARG, 10 IxM).

20

3.5.1. Effect of DL- and D-propanolol The effect of DL- and D-propranolol (0.33 tzM) on nitric oxide and 6-keto production is shown in fig. 2. D-Propranolol was significantly less active than DLpropanolol on nitric oxide production and was devoid of activity on 6-keto production.

IOC~,iiBACHOL I O u M • ATROPINE 1 uM

o

~

-10 0

CONTROL , 20

I

I

i F 40 60 MINUTEI3

-...,............t

, 80

.,., ,

100

Fig. 1. Time course of N O production, as evaluated by the change in m e t H b in the effluent (see text). Control experiments and effect of carbachol (10 p~M). Blockade of the response by atropine (1 IzM). M e a n - + S . E . N . S . : non-significant, * * * P < 0.001. Areas u n d e r the curve: control -132.33_+87.63 ( N = 6 ) ; carbachol 1342.33+87.29 (N = 3) P < 0.001; carbachol + atropine 87.00_+ 77.47 (N = 3), N.S.

3.5.2. Effect of atropine on DL-propranolol-induced NO release In order to exclude a non-specific effect of DL-propranolol on cholinergic receptors or of atropine on NO production, atropine (1 /zM) has also used with the /3-blocker. Atropine did not modify the response to

230

DL-propranolol. The area under the curve was 1464.40 + 218.20, N = 5 (P < 0.001 compared to control and non-significant compared to DL-propranolol).

- - / ~ metHb~

- - ~ 6-keto-PGFlalphsng/ml

40

1 ***

.

,).-,.~,,,~.o,.o,. o....j

***

~

I

-0,o

3.5.3. Effect of acetylsalicylic acid The effect of acetylsalicylic acid (1.6 /xM) is shown in fig. 3. Nitric oxide levels were only slightly decreased, while 6-keto production was abolished.

20-

L , ' f ~ ***

10-

~

3.5.4. Effect of endothelium removal with CHAPS

***

*

*

*~

0,4 N.8.

f o ~

.,.s ~ I

, 6 .. ~" ~

I

0,2

~ ~.-~N.8.

o

The effect of removing the endothelium with CHAPS (8 mM), infused during 30 s, was studied in two groups of experiments. In the first CHAPS was applied after 80 rain of perfusion with DL-propranolol (0.33 /zM) and still in presence of this compound (fig. 4). In the second CHAPS was applied at the beginning ot the experiment, before the addition of DL-propranolol. In both cases nitric oxide production in response to the adrenoceptor/3-blocker was abolished. MetHb differed from initial values by 26.49 +_ 2.55% (N = 7) before CHAPS application and by only - 0 . 2 + 1.52% (N = 7) after endothelium removal. This last value was not significantly different from zero (P > 0.2). In the first group of experiments, CHAPS was applied during a high level of stimulation, in the second it prevented any response. The production of 6-keto followed the same pattern.

Arginine induced a small increased in NO production but did not modify the response to the /3-blocker. L-arginine alone did not influence 6-keto production.

3.5.5. Effect of L-arginine

3.5. 6. Effect of N C-nitro-L-arginine

Figure 5 shows the effect of L-arginine alone (300 # M ) and when the amino acid was applied (100 /zM) simultaneously with DL-propranolol (0.33 tzM). L-

~/~

metHb %

--/~

CONTROL8 -10 0

I ' ~

N-6

20

-0,2 40

60 MINUTES

80

100

120

Fig. 4. Effect of CHAPS (8 m M infused for 30 s) on DLpropranolol-induced NO production. C H A P S was applied after 60 min of perfusion with DL-propranolol. Comparison with the controls. Mean_+S.E. * * * P