Histamine is a potent mediator of local and systemic inflammatory and allergic responses. The complexity of the vascular response to histamine is based on differing actions mediated by histamine-receptor subtypes H1 and H2 in endothelium and smooth muscle. Considerable species differences also have been noted in a variety of responses to histamine, resulting in the release of chemical mediators, such as endothelium-derived relaxing factor (EDRF) and prostaglandins (PG), from endothelial and subendothelial tissues (1-3). There are many reports of histamine-induced smooth-muscle relaxation effected by EDRF released in response to H1-receptor stimulation in rat thoracic aorta (4), human umbilical vein endothelial cells (5,6), guinea-pig aorta (7), monkey and human cerebral artery (8), and bovine corneal endothelial cells (BCECs; 9). On the other hand, Hekimian et al. (10) observed that histamine increases the cyclic adenosine monophosphate (cAMP) content of bovine aortic endothelial cells, a response mediated by H2-receptors. Ottoson et al. (11) showed that the histamine-induced dilatation of small human cerebral arteries was mediated by both H1- and H2-receptors, with the H1 subtype predominating. In these studies, however, NO release was not directly observed because measurement of NO release has proven technically difficult. Our study was undertaken to assess histamine-induced NO release from isolated porcine aorta and NO involvement in the histamine-induced relaxation of porcine coronary arteries, with special reference to the histaminergic receptor subtypes involved.
MATERIALS AND METHODS
Porcine thoracic aorta was cut into ring segments ∼10 mm long after removal of connective tissue. Special care was taken to avoid contact with the luminal surface of the rings to preserve the endothelium. To remove endothelium, the lumens of aortic-ring segments were gently rubbed with cotton thread. These rings were stored in cool normal Tyrode's solution at pH 7.2, which contained (in mM) NaCl, 137.0; KCl, 2.7; (3-[N-morpholino] propane-sulfonic acid; MOPS)-Na, 7.5; CaCl2, 1.2; and glucose, 5.5.
Endothelial cells were cultured by the method of Schmidt et al. (12). The endothelium was scraped from the thoracic aorta of a pig obtained from a local slaughterhouse and suspended in Dulbecco's minimum essential medium (DMEM; Flow Laboratories, Puteaux, France) containing 20% fetal calf serum (Flow Laboratories). They were grown to confluence in a 5% CO2 atmosphere at 37°C and subcultured at a density of ∼20,000 cells/ml in 24-well plates (Greiner Labortechnik, Frickenhausen, Germany).
In this study, we used segments of porcine thoracic aorta to measure NO release by using an NO electrode and electron paramagnetic resonance (EPR) spin trapping. Cultured porcine endothelial cells were used to measure cytosolic Ca2+ with fura-2 acetoxymethylester (fura-2/AM).
Measurement of NO release
An NO electrode (model NO-501; Inter Medical Co., Nagoya, Japan) was used to measure NO release from segments of aorta. The electrode represents a modification by Ichimori et al. (13) of that reported by Shibuki and Okada (14) and measures redox currents with respect to a counterelectrode on the order of picoamperes (pA). The electrode was coated with a three-layered membrane consisting of KCl, NO-selective resin, and silicone. The counterelectrode was made of carbon fiber. The polarographic current was detected with a current-voltage converter circuit. The signal-to-noise ratio (S/N) of electrode current was 1 with 5 μM S-nitroso-N-acetyl-DL-penicillamine (SNAP), which corresponded to 6.5 nM NO. The electrode current increased linearly with NO concentration up to 1 mM and was sufficiently sensitive to detect NO or EDRF produced by endothelial cells of porcine aorta. NO release, expressed as pA of current, was recorded on a chart recorder (T1-106-3; M.T. Giken, Tokyo, Japan). Because of differences in sensitivities between electrodes, we corrected each current by using the current equivalent to the amount of NO released by 5 μM SNAP.
The porcine aortic rings were washed twice before experiments and soaked in 1 ml of Hanks' solution at pH 7.2 (Nissui Chemical, Tokyo, Japan), which contained (in mM) NaCl, 136.9; KCl, 5.4; Na2PO4, 0.34; KH2PO4, 0.45; MgSO4, 0.41; MgCl2, 0.49; CaCl2, 1.3; and glucose, 5.5.
L-Arginine (20 μM) was added to each well, and experiments were carried out at room temperature. Test-solution volumes of 10-20 μl were added to 1 ml of well contents. Repeated applications of drugs were made at intervals >5 min. H1, H2, and cAMP antagonists were added 10 min before the test solution.
Electron paramagnetic resonance
NO release also was measured by EPR spin trapping by using a method developed by Voevodskaya and Vanin (15) and Tominaga et al. (16). Fe-citrate (1 mM) and sodium-N,N-diethyldithiocarbamate (DETC; 1 mM) were dissolved in Tyrode's solution containing aortic ring. Histamine (200 μM), with or without pyrilamine (2 μM), chlorpheniramine (2 μM), or with cimetidine (2 μM) were added. After incubation for 2 min at 34°C, the supernatants of samples were put into 4-mm-diameter EPR tubes (Wilmad Glass, Buena, NJ, U.S.A.), and immediately frozen by immersion in liquid nitrogen at 77°K. EPR spectra were measured by a JEOL FE1X spectrometer (Nikon Denko, Tokyo, Japan). Measurement conditions were as follows: microwave frequency, 9.15 GHz; microwave power, 20 mW; modulation frequency, 100 kHz; modulation amplitude, 0.63 mT; magnetic field, 320 ± 25 mT; response time, 0.1 s; sweep time, 50 mT/4 min; and temperature, 77°K. The concentration of NO trapped was compared semiquantitatively by using the intensity of the first derivative of the EPR signal and its amplitude.
Intracellular Ca2+ measurement
Cytosolic Ca2+ was measured by using fura-2/AM. Endothelial cells were incubated in a culture medium containing 5 μM fura-2/AM for 30 min at 37°C. Cells were then washed with normal Tyrode's solution to remove free fura-2/AM from the culture dishes. Fluorescence was measured by the dual-wave-length excitation method. To measure fluorescence waves, we used a fluorescence microscope (Optiphot; Nikon, Tokyo, Japan) equipped with a photomultiplier tube (R649; Hamamatsu Electronics, Shizuoka, Japan), a photon counter (545A; NF, Hiroshima, Japan), and appropriate combinations of filters (Nihon Shinku-kogaku, Osaka, Japan), so that the cells were excited by alternating wavelengths of 340 and 380 nm. All measurements were carried out at 35-37°C.
Porcine epicardial coronary artery was excised, cleaned of perivascular tissue, and cut into segments 2-3 mm long. Vascular rings in the presence of endothelium were mounted on two stainless steel wires. One wire was connected to a transducer, and the other was anchored to a plastic fiber passed though the vessel lumen within water-jacketed organ baths. The baths contained 5 ml of Krebs bicarbonate solution, which included (in mM) NaCl, 118.4; KCl, 4.7; CaCl2, 2.5; MgSO4, 2.4; NaHCO3, 25.0; KH2PO4, 1.2; and glucose, 11.1; maintained at 37°C while being bubbled with 95% O2/5% CO2. The resting tension was adjusted to 2.0 g and maintained throughout the experiment. Changes in tension were measured by using a force-displacement transducer and recorded on a pen-chart recorder (T1-106-3; M.T. Giken). Vascular rings were allowed to equilibrate for ≥30 min. The response to acetylcholine was assessed in every preparation to evaluate for functioning endothelial cells (17). KCl (30 mM) and H1-and H2-antagonists were added to the bath. After the KCl-induced contraction reached a constant value, histamine or forskolin was added. Contractions or relaxations induced by drugs are presented as a percentage relative to those induced by 30 mM KCl.
Histamine, NG-monomethyl-L-arginine (L-NMMA), chlorpheniramine, and dibutyl cAMP (DBcAMP) were purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.). L-Arginine was obtained from Gibco Laboratories (Grand Island, NY, U.S.A.). Forskolin was obtained from Behring Diagnostics (San Diego, CA, U.S.A.). Cimetidine was a gift from Smith Kline & Beecham (Tokyo, Japan). Rp diastereomer of adenosine cyclic 3′,5′-phosphorothioate (Rp-cAMPS; 18,19) was purchased from Research Biochemicals International (Boston, MA, U.S.A.). DETC was purchased from Aldrich (Milwaukee, WI, U.S.A.). SNAP and Fe-citrate were purchased from Wako Chemical Co. (Tokyo, Japan). Fura-2/AM was purchased from Dojin Kagaku (Kumamoto, Japan).
Data are presented as means ± SEM. The significance of differences between values in two groups was analyzed by the two-tailed Student's t test, and a probability (p) value <0.05 considered significant.
Release of NO in response to histamine stimulation
Figure 1a shows histamine-induced NO release from the porcine aortic segment measured with an NO electrode. Histamine produced NO in a concentration-dependent manner in the range of 10−6−10−3M(Fig. 1e). Application of 200 μM histamine induced NO release with a peak value of 44 ± 6 pA (317 nM NO; n = 6). A second application of histamine produced similar NO release (98 ± 8% of first application; n = 6). Removal of the endothelium completely suppressed histamine-induced NO release from ring segments (n = 4). In the presence of L-NMMA (20 μM), histamine failed to induce detectable NO release (n = 4; data not shown).
To clarify which receptor type (H1 or H2) was responsible for NO production, we tested the effect of pyrilamine and cimetidine, histamine H1- and H2-receptor antagonists, respectively, on NO release. In the presence of 2 μM pyrilamine, histamine released a similar amount of NO (55 ± 12 pA; n = 6, not significantly different from control; Fig. 1b). However, in the presence of 2 μM cimetidine, histamine-induced NO production was reduced to 18 ± 2 pA (n = 4; p < 0.01; Fig. 1c).
Effect of cytosolic adenosine 3′,5′-cyclic adenosine monophosphate (cAMP)
It is reported that H2-receptor-mediated action is linked to the adenylate cyclase system, with cAMP as second messenger (20). Therefore we tested the effect of Rp-cAMPS, a membrane-permeable antagonist of cAMP. Histamine-induced NO release was significantly reduced by the addition of 20 μM Rp-cAMPS (11 ± 7 pA; n = 4; p < 0.01), suggesting that histamine-induced NO release was mediated by cAMP-dependent protein phosphorylation (Fig. 1d).
We also tested the effect of forskolin, an activator of adenylate cyclase, on NO release. Figure 2 shows forskolin-induced NO release from an endothelium-intact porcine aortic ring. Application of 100 μM forskolin induced NO release with a peak value of 335 ± 37 pA (n = 6). Removal of the endothelium completely suppressed forskolin-induced NO release (n = 4); 1 mM DBcAMP also induced NO release with a peak value of 116 ± 28 pA (n = 4). Pretreatment with 20 μM L-NMMA, an inhibitor of NO production, suppressed 0.1-100 μM forskolin-induced NO production (n = 4; data not shown).
NO detection by the EPR spin-trapping method
We further studied the effects of H1 and H2 receptors and cAMP on histamine-induced NO release by using EPR. In this method, DETC and Fe form a complex to trap NO, which is relatively stable as a specific triplet hyperfine structure (hfs). The representative spectra shown in Fig. 3 are attributable to Fe2+(DETC)2NO complex, because the accumulated paramagnetic Fe2+(DETC)2NO complex exhibited a typical EPR signal with g⊥ = 2.035 and g″ = 2.02 with unresolved triplet hfs at g in frozen state (21). The first peak showed the amount of NO release (Fig. 3a). In the presence of pyrilamine, the peak remained unchanged (Fig. 3b), but in the presence of cimetidine (Fig. 3c) or Rp-cAMPS (Fig. 3d), the peak was reduced significantly. We semiquantitatively determined the amount of NO release from a calibration curve. The intensity of the EPR signal increased linearly with the NO concentration, Equation 1
The results were similar to those obtained by the NO electrode. In control samples, histamine produced 431 ± 12 nM NO. Histamine-induced NO release was reduced by 2 μM cimetidine (202 ± 16 nM; n = 5; p = 0.001) and 20 μM Rp-cAMP (206 ± 10 nM; n = 4; p = 0.001), but was not reduced by 2 μM pyrilamine (483 ± 137 nM; n = 4; not significant) or 2 μM chlorpheniramine, another H1-receptor antagonist (420 ± 72 nM to 412 ± 14 nM; n = 4; not significant).
Histamine-induced NO release and cytosolic Ca2+
Production of NO by many substances is considered to be secondary to an increase in cytosolic Ca2+. Therefore we measured changes induced in intracellular Ca2+ by histamine and forskolin by using fura-2/AM. Addition of 100 μM histamine did not produce a detectable increase in intracellular Ca2+(Fig. 4a). Similar results were obtained in experiments with forskolin (n = 8; Fig. 4b). Figure 4c shows the increase in intracellular Ca2+ observed in the same cells with application of other agonists such as adenosine triphosphate (ATP; 200 nM). These results indicated that histamine-induced NO release in cultured porcine aortic endothelium was not associated with cytosolic Ca2+ increase.
Effect of histamine on isometric tension
In porcine coronary artery precontracted with 30 mM KCl, histamine produced further contraction in a concentration-dependent manner (Fig. 5a). Figure 5b shows that pretreatment with cimetidine (2 μM) induced significantly greater contraction by histamine in KCl-precontracted coronary artery (259 ± 37% vs. pretreatment with 30 mM KCl alone; n = 10). On the other hand, pre-treatment with pyrilamine (2 μM) slightly decreased histamine-induced contractile responses (Fig. 5c) without reaching statistical significance (87 ± 6%; n = 11). Forskolin showed vasodilative effects. Figure 5d shows that 10 μM forskolin induced vasodilation of KCl-precontracted coronary artery (22 ± 10% vs. pretreatment of 30 mM KCl; n = 4).
In this study, we investigated the relation of H1 and H2 receptors to histamine-induced NO release from vascular endothelial cells by using an NO electrode and EPR spin trapping. In porcine aortic endothelial cells, histamine-induced NO release was not inhibited significantly by pyrilamine, a histamine H1-receptor antagonist but was inhibited by cimetidine, an H2-receptor antagonist. Forskolin, a stimulator of adenylate cyclase, induced NO production from porcine endothelial cells, and histamine-induced NO release was suppressed by Rp-cAMPS, a membrane-permeable antagonist of cAMP. These results suggest that histamine-induced NO release was mediated mainly by endothelial H2 receptors and not by H1 receptors.
Histamine acting at H1 receptors increases the formation of inositol 1,4,5-triphosphate (IP3), which releases intracellular Ca2+ from organellar Ca2+ stores and also induces influx of extracellular Ca2+(5,22). An increase in cytosolic Ca2+ concentration is needed for hormone-induced signals to stimulate NO synthase, which leads to an increase in NO levels. In our study, however, histamine did not increase cytosolic Ca2+ in cultured porcine aortic endothelial cells as measured by fura-2/AM. In many cell types, two major systems coexist to transduce signals across the cell membrane, the cAMP pathway and the Ca2+/phosphatidylinositol pathway. NO production by histamine was suppressed by H2-receptor antagonist and not by H1-receptor antagonist, suggesting that NO production may relate selectively to the H2 receptor. Moreover, NO production was suppressed by Rp-cAMPS, suggesting that the effect of histamine was mediated by cAMP production.
Although the precise mechanism was not elucidated, a possible mechanism of the cAMP effect in this study could be a stimulation of NO synthase by a new pathway or by increasing sensitivity to Ca2+. The functional role of cAMP in vascular endothelial cells is still controversial. Kuhn et al. (23) showed that cyclic nucleotides including cAMP did not regulate the release of NO from bovine aortic endothelial cells (23). On the other hand, Graier et al. (24) reported that increases in endothelial cAMP amplified bradykinin- and ATP-induced Ca2+ mobilization, in agonist-induced biosynthesis and release of NO, but cAMP by itself did not affect basal NO formation from porcine aortic endothelial cells. In rat hepatocytes, the frequency and the amplitude of IP3-mediated Ca2+ fluxes were enhanced by an increase in cAMP levels (25). Reiser (26), studying a neuronal cell line, stated that a possible mechanism of cAMP effect could involve stimulation of peptide-induced Ca2+ release either by enhancing IP3 formation or by increasing the sensitivity of the IP3 receptor. Gray and Marshall (27) reported that isoprenaline acted through a β-adrenoceptor on the endothelium to increase cAMP and that directly or indirectly, this may release NO to evoke vascular relaxation via an increase in cyclic guanosine monophosphate (cGMP). Mori et al. (28) demonstrated that the increase in cAMP with suppression of phosphodiesterase III (amrinone) or stimulation of adenylate cyclase (forskolin) induced relaxation of rat thoracic aorta via production of NO. In agreement with these studies, our study suggested that increases in cAMP are associated with NO release as measured by both an NO electrode and ESR. Moreover, our results strongly suggest that histamine does not increase cytosolic Ca2+ but increases cAMP, which could regulate NO release from porcine aortic endothelial cells.
It is well documented that V2-type vasopressin receptors are linked to the adenyl cyclase system, with cAMP as second messenger. Aki et al. (29) reported that NO might participate in V2-receptor-mediated renal vasodilation of dogs. This result also indicates the possibility that increases of cAMP may be involved in the release of NO triggered by some vasoactive substances.
The histamine concentration we used in this study is very high. However, the NO electrode shows low amplitude when we use lower concentrations of the agonists (30). Therefore we used high concentrations of histamine to measure NO release. Results were qualitatively similar in low concentrations.
Previously histamine-induced NO release from endothelial cells was believed to involve increased cytosolic Ca2+. Our result is at variance with this conclusion. Differences may involve histaminergic receptor subtypes, species differences, and differences between cultured cells and isolated arteries. However, this study indicates that in some species, NO release from endothelial cells may be induced by increase of cAMP, with or without increased cytosolic Ca2+. Cimetidine is widely used as an H2-receptor antagonist to cure or prevent peptic ulcer. Shimokawa et al. (31) reported that cimetidine had potential vasoconstrictive effects in patients with coronary artery spasm and concluded that cimetidine should be administered with caution in patients with the vasospastic angina pectoris. In agreement with their study, our study suggests that H2-receptor antagonists might evoke an unwanted vasoconstriction by the suppression of NO release from endothelial cells.
Acknowledgment: We thank Dr. Koichiro Tsuchiya for help with EPR measurement.
1. Konishi M, Toda N, Yamamoto M. Different mechanisms of action of histamine in isolated arteries of the dog. Br J Pharmacol
2. Tsuru H, Kohno, S, Iwata M, Shigei T. Characterization of histamine receptors in isolated rabbit veins. J Pharmacol Exp Ther
3. Miller WL, Bove AA. Differential H1
- and H2
-receptor-mediated histamine responses of canine epicardial conductance and distal resistance coronary vessels. Circ Res
4. Van de Voorde J, Leusen I. Role of the endothelium in the vasodilator response of the rat thoracic aorta to histamine. Eur J Pharmacol
5. Lo WWY, Fan T-PD. Histamine stimulates inositol phosphate accumulation via the H1
-receptor in cultured human endothelial cells. Biochem Biophys Res Commun
6. Bull HA, Courtney PF, Rustin MHA, Dowd PM. Characterization of histamine receptor sub-types regulating prostacyclin release from human endothelial cells. Br J Pharmacol
7. Hide M, Fukui H, Watanabe T, Wada H, Yamamoto S. Histamine H1
-receptor in endothelial and smooth muscle cells of guinea-pig aorta. Eur J Pharmacol
8. Toda N. Mechanism underlying responses to histamine of isolated monkey and human cerebral arteries. Am J Physiol
9. Crawford KM, MacCallum DK, Ernst SA. Histamine H1
signaling in cultured bovine corneal endothelial cells. Invest Ophthalmol Vis Sci
10. Hekimian G, Côte S, Van Sande J, Boeynaems JM. H2
receptor-mediated responses of aortic endothelial cells to histamine. Am J Physiol
11. Ottoson A, Jansen I, Edvinsson L. Characterization of histamine receptors in isolated human cerebral arteries. Br J Pharmacol
12. Schmidt K, Mayer B, Kukovetz WR. Effect of calcium on endothelium-derived relaxing factor formation and cGMP levels in endothelial cells. Eur J Pharmacol
13. Ichimori K, Ishida H, Fukahori M, Nakazawa H, Murakami E. Practical nitric oxide measurement employing a nitric oxide-selective electrode. Rev Sci Instrum
14. Shibuki K, Okada D. endogenous nitric oxide release required for long-term synaptic depression in the cerebellum. Nature
15. Voevodskaya NV, Vanin AF. Gamma-irradiation potentiates L-arginine-dependent nitric oxide formation in mice. Biochem Biophys Res Commun
16. Tominaga T, Sato S, Ohnishi T, Ohnishi ST. Potentiation of nitric oxide formation following bilateral carotid occlusion and focal cerebral ischemia in the rat: in vivo detection of the nitric oxide radical by electron paramagnetic resonance spin trapping. Brain Res
17. Nyborg NCB, Mikkelsen EO. 5-Hydroxytryptamine does not release endothelium-derived relaxing factor in rat isolated coronary arteries. Eur J Pharmacol
18. Wang Lu-Y, Salter MW, MacDonald JF. Regulation of kainate receptors by cAMP-dependent protein kinase and phosphatases. Science
19. Yusta B, Ortiz-Caro J, Pascual A, Aranda A. Comparison of the effects of forskolin and dibutyryl cyclic AMP in neuroblastoma cells: evidence that some of the actions of dibutyryl cyclic AMP are mediated by butyrate. J Neurochem
20. Van de Voorde J, Brochez V, Vanheel B. Heterogeneous effects of histamine on isolated rat coronary arteries. Eur J Pharmacol
21. Mülsch A, Vanin A, Mordvintcev P, Hauschildt S, Busse R. NO accounts completely for the oxygenated nitrogen species generated by enzymic L-arginine oxygenation. Biochem J
22. Resink TJ, Grigorian GY, Moldabaeva AK, Danilov SM, Bühler FR. Histamine-induced phosphoinositide metabolism in cultured human umbilical vein endothelial cells: association with thromboxane and prostacyclin release. Biochem Biophys Res Commun
23. Kuhn M, Otten A, Frölich JC, Förstermann U. Endothelial cyclic GMP and cyclic AMP do not regulate the release of endothelium-derived relaxing factor/nitric oxide from bovine aortic endothelial cells. J Pharmacol Exp Ther
24. Graier WF, Groschner K, Schmidt K, Kukovetz WR. Increases in endothelial cyclic AMP levels amplify agonist-induced formation of endothelium-derived relaxing factor (EDRF). Biochem J
25. Schöfl C, Sanchez-Bueno A, Brabant G, Cobbold PH, Cuthbertson KSR. Frequency and amplitude enhancement of calcium transients by cyclic AMP in hepatocytes. Biochem J
26. Reiser G. Nitric oxide formation caused by Ca2+
release from internal stores in neuronal cell line is enhanced by cyclic AMP. Eur J Pharmacol
27. Gray DW, Marshall I. Novel signal transduction pathway mediating endothelium-dependent β-adrenoceptor vasorelaxation in rat thoracic aorta. Br J Pharmacol
28. Mori K, Takeuchi S, Moritoki H, et al. Endothelium-dependent relaxation of rat thoracic aorta by amrinone-induced nitric oxide release. Eur Heart J
29. Aki Y, Tamaki T, Kiyomoto H, et al. Nitric oxide may participate in V2
vasopressin-receptor-mediated renal vasodilation. J Cardiovasc Pharmacol
30. Kishi F, Nakaya Y, Takahashi A, Miyoshi H, Nomura M, Saito K. Intracellular and extracellular Ca2+
regulate histamine-induced release of nitric oxide in vascular endothelial cells as shown with sensitive and selective nitric oxide electrodes. Pharmacol Res
31. Shimokawa H, Okamatsu S, Taira Y, Nakamura M. Cimetidine induces coronary artery spasm in patients with vasospastic angina. Can J Cardiol
Keywords:© Lippincott-Raven Publishers
Nitric oxide; Histamine; Histamine receptors; Calcium; Cyclic AMP; Porcine thoracic aorta