Activation of nitric oxide synthase by beta 2-adrenoceptors in human umbilical vein endothelium in vitro

Abstract

1. Some animal studies suggest that beta-adrenoceptor-mediated vasorelaxation is in part mediated through nitric oxide (NO) release. Furthermore, in humans, we have recently shown that forearm blood flow is increased by infusion of beta2-adrenergic agonists into the brachial artery, and the nitric oxide synthase (NOS) inhibitor N(G)-monomethyl-L-arginine (L-NMMA) inhibits this response. 2. The purpose of the present study was to determine whether stimulation of human umbilical vein endothelial beta-adrenoceptors causes vasorelaxation and nitric oxide generation, and whether this might be mediated by cyclic adenosine-3',5'-monophosphate (cyclic AMP). 3. Vasorelaxant responses were determined in umbilical vein rings to the nonselective beta-adrenergic agonist isoprenaline and to the cyclic AMP analogue dibutyryl cyclic AMP, following precontraction with prostaglandin F2alpha. 4. NOS activity was measured in cultured human umbilical vein endothelial cells (HUVEC) by the conversion of [3H]-L-arginine to [3H]-L-citrulline, and adenylyl cyclase activity by the conversion of [alpha-32P]-ATP to [32P]-cyclic AMP. 5. Isoprenaline relaxed umbilical vein rings, and this vasorelaxation was abolished by beta2- (but not beta1-) adrenergic blockage, and by endothelium removal or 1 mM L-NMMA. In addition, vasorelaxant responses to dibutyryl cyclic AMP were inhibited by 1 mM L-NMMA, with a reduction in Emax from 90.0+/-9.3% to 50.5+/-9.9% (P<0.05). 6. Isoprenaline 1 microM increased NOS activity in HUVEC (34.0+/-5.9% above basal, P<0.001). Furthermore, isoprenaline increased adenylyl cyclase activity in a concentration-dependent manner; this response was inhibited by beta2 (but not beta1-) adrenergic blockade. Forskolin 1 microM and dibutyryl cyclic AMP 1 mM each increased NOS activity in HUVEC, to a degree similar to isoprenaline 1 microM. The increase in L-arginine to L-citrulline conversion observed with each agent was abolished by coincubation with NOS inhibitors. 7. These results indicate that endothelial beta2-adrenergic stimulation and cyclic AMP elevation activate the L-arginine/NO system, and give rise to vasorelaxation, in human umbilical vein.

Figure 1

Mean±s.e.mean vasorelaxant responses in human umbilical vein rings. Responses are plotted as the percentage of the response to prostaglandin F_2α. (a) Concentration-effect curves in vascular rings with intact endothelium, in response to isoprenaline alone (n=17), in the presence of CGP 20712A 300 nM (n=9), in the presence of ICI 118551 100 nM (n=8), and in the presence of CGP 20712A 300 nM+ICI 118551 100 nM (n=17). (b) Concentration-effect curves to isoprenaline, with intact endothelium (n=17) or following de-endothelialization (n=8), or to isoprenaline with intact endothelium in the presence of L-NMMA 1 mM (n=9). (c) Concentration-effect curves to dibutyryl cyclic AMP in vascular rings with intact endothelium, in the absence or presence of L-NMMA 1 mM (n=7).

Figure 2

[³H]-L-citrulline production from [³H]-L-arginine in HUVEC, expressed as percentage change (mean±s.e.mean) from control values, either in the absence (open bars) or presence (shaded bars) of L-NMMA 0.1 mM. Responses are shown to isoprenaline 1 μM (Isop), forskolin 1 μM (Forsk), dibutyryl cyclic AMP 1 mM (DbcAMP) and histamine 10 μM (Hist). ^**P<0.01 and ^***P<0.001 respectively, vs control. ‡‡‡P<0.001 vs corresponding agonist in the absence of L-NMMA. Numbers adjacent to bars indicate number of umbilical cords tested.

Figure 3

[³H]-L-arginine content in HUVEC, following incubation with [³H]-L-arginine, expressed as percentage change (mean±s.e.mean) from control values, either in the absence (open bars) or presence (shaded bars) of L-NMMA 0.1 mM. Responses are shown to isoprenaline 1 μM (Isop), forskolin 1 μM (Forsk), dibutyryl cyclic AMP 1 mM (DbcAMP) and histamine 10 μM (Hist). ^*P<0.05 and ^***P<0.001 respectively, vs control. ‡‡P<0.01 and ‡‡‡P<0.001 respectively, vs corresponding agonist in the absence of L-NMMA. Numbers adjacent to bars indicate number of umbilical cords tested.

Figure 4

[³H]-L-citrulline production and [³H]-L-arginine content in HUVEC in response to isoprenaline 1 μM, expressed as percentage change (mean±s.e.mean) from control values, either in the absence (open bars) or presence (shaded bars) of L-NAME 0.1 mM. ^**P<0.01 vs control. ‡‡P<0.01 vs isoprenaline in the absence of L-NAME. Results are shown for HUVEC obtained from six different umbilical cords.

Figure 5

Ca²⁺ transients in a population of HUVEC in response to isoprenaline (Isop) 1 μM and to histamine (Hist) 10 μM, as determined by fura-2 fluorescence; the curve shows the ratio of emission at 530 nm following excitation at 340 nm to that following excitation at 380 nm.

Figure 6

Mean±s.e.mean adenylyl cyclase activity in HUVEC homogenates (n=6) as a function of isoprenaline concentration. Responses are shown to isoprenaline alone, isoprenaline in the presence of CGP 20712A 300 nM, and isoprenaline in the presence of ICI 118551 50 nM.