MATERIALS AND METHODS

Extraction of Anaxagorea luzonensis A. Grey (AL): The dried and chopped heartwood (5 kg) of AL was extracted four times with methanol using soxhlet extractor. After filtration, the methanolic extract was concentrated under reduced pressure and was then partitioned with dichloromethane (CH₂Cl₂), yielding about 10.5 g of CH₂Cl₂-AL after evaporation.

Tissue preparation: In 2009, Experiments were performed using aorta obtained from male Wistar rats (300-350 g) purchased from National Laboratory Animal Center, Mahidol University, Thailand. Rats were housed in standard environmental condition (25°C) under 12 h light/dark cycles and fed with standard laboratory rat chow and tap water ad libitum. All experimental procedures were reviewed and approved by the Animal Research Ethics Committee of the Faculty of Medicine, Srinakharinwirot University.

Male Wistar rats were anaesthetized with Zolitil 50 mg kg^-1 (tiletamine chloridrate and zolazepan chloridrate) into quadriceps muscle and killed by cervical dislocation. Following a thoracotomy, the thoracic aorta was carefully removed, cleaned of fat and connective tissue and cut into 5 mm ring segments. Each ring was transferred to a jacketed organ bath filled with 20 mL of modified Krebs-Henseleit solution, composed of (mM) NaCl 118, KCl 4.7, MgSO₄ 1.2, KH₂PO₄ 1.2, NaHCO₃ 25, CaCl₂ 2, D-glucose 10, that was maintained at 37°C and bubbled continuously with 95 % O₂ and 5% CO₂ mixture. The solution in the organ bath was exchanged every 15 min for 1 h. The rings were mounted between two triangular stainless steel hooks that were passed through the lumen and stretched to an optimal passive tension of about 1 g and maintained at this tension for 1 h.The upper hook was connected to a isometric force transducer (MLT 0210, New South Wales, Australia) and changes in isometric force were recorded on a MacLab recording system (AD instruments, New South Wales, Australia).

Experimental protocol: Following a 1 h equilibration period, methoxamine (10-60 μM) was used to increase tone by approximately 1 g. Once a stable contraction was estabilshed, CH₂Cl₂-AL (1-300 μg mL^-1) was added cumulatively. To explore the mechanisms involved in vasorelaxation induced by CH₂Cl₂-AL, aortic rings were incubated with various inhibitors added to the organ bath before methoxamine was added to increase tone. In vehicle-control experiments, dimethyl sulphoxide (DMSO) alone was added cumulatively in the same volumes as those used in the experiments with CH₂Cl₂-AL.

To examine the contribution of the endothelium in vasorelaxant responses to CH₂Cl₂-AL, the endothelium was mechanically removed by gently rubbing the luminal surface with a cocktail stick. Removal of the endothelium was demonstrated by vasorelaxation to 10 μM carbachol being less than 10% of the induced tone. To investigate the involvement of vasodilator prostanoids via the cyclooxygenase (COX) pathway and Nitric Oxide (NO) in vasorelaxation to CH₂Cl₂-AL, aortic rings were treated with indomethacin (10 μM), a COX inhibitor and N^G-nitro-L-arginine methyl ester (L-NAME, 300 μM), an inhibitor of endothelial nitric oxide synthase, respectively.

To investigate the potential involvement of K⁺ channels in vasorelaxation to CH₂Cl₂-AL, aortic rings were pre-contracted with a high extracellular concentration of KCl (60 mM), which was prepared by replacing an equimolar concentration of NaCl with KCl (Tep-Areenan et al., 2003). These experiments were also performed in endothelium-denuded rings to determine any involvement of Endothelium-Derived Relaxing Factors (EDRFs) in the vasorelaxant effects of CH₂Cl₂-AL on activation of K⁺ channels. To characterized the types of K⁺ channels involved in vasorelaxation to CH₂Cl₂-AL, concentration-response curves to CH₂Cl₂-AL were constructed after incubation with 4-aminopyridine (4-AP, 1 mM), a voltage-gated K⁺ (K_V) channel inhibitor, glibenclamide (10 μM), an ATP-sensitive (K_ATP) inhibitor, or barium chloride (BaCl₂, 30 μM), an inward-rectifier (K_IR) channel inhibitor, clotrimazole (2 μM), an intermediate-conductance calcium-activated potassium (IK_Ca) channel inhibitor and 1-[(2-chlorophenyl)diphenylmethyl]-1H-pyrazole (TRAM-34, 10 μM), a specific inhibitor of IK_Ca channels.

To examine the vascular effect of CH₂Cl₂-AL on extracellular Ca²⁺ influx, concentration-response curve to CaCl₂ (10 μM-30 mM) were constructed in the presence and absence of CH₂Cl₂-AL (1, 10 and 100 μg mL^-1) for 30 min. Aortic rings were first allowed to equilibrate at 1 g tension in a Ca²⁺-free Krebs solution and then the rings were bathed with Ca²⁺-free, high KCl (100 mM) Krebs solution. In vehicle-control experiments, DMSO was added in the same volume as that used in the experiments with CH₂Cl₂-AL.

Data and statistical analysis: The concentration of vasorelaxant giving half-maximal relaxation (EC₅₀) and maximal responses (R_max) were obtained from the concentration-response curve fitted to a sigmoidal logistic equation using the GraphPad Prism package described by Tep-Areenan et al. (2003). R_max and pEC₅₀ values (negative logarithm of the EC₅₀) were compared by analysis of variance (ANOVA) with statistically significant differences between groups being determined by Bonferroni’s post-hoc test. These were expressed as Mean±SEM. The results were considered statistically significant when p value was less than 0.05. The number of animals in each group is represented by n.

Chemicals: All drugs and chemicals were purchased from Sigma Chemical Company (St. Louis, Missouri, USA), but zoletil was purchased from Virbac (Carros Cedex, France). Indomethacin was dissolved in ethanol. CH₂Cl₂-AL and glibenclamide were dissolved in DMSO. 4-AP, BaCl₂ and TRAM-34 were dissolved in distilled water. The remaining drugs were dissolved in the Krebs solution. All drugs were made up on the day of the experiment.

RESULTS

The effects of endothelial denudation, indomethacin and L-NAME on vasorelaxation to CH₂Cl₂-AL in rat aortic rings: CH₂Cl₂-AL (1-300 μg mL^-1) caused vasorelaxation in a concentration-dependent manner (pEC₅₀ = 4.87±0.08 with R_max = 118±4%, n = 6, Fig. 1). Removal of the endothelium significantly (p<0.05) reduced the effects of CH₂Cl₂-AL (R_max: control = 118±4%, n = 6; denuded = 96.1±7.5%, n = 6, Fig. 1). Similarly, vasorelaxant responses to CH₂Cl₂-AL were significantly (p<0.05) reduced after pretreatment with a combination of indomethacin plus L-NAME, which were not different from L-NAME alone (R_max: L-NAME alone = 99.3±5.5%, n = 6; indomethacin plus L-NAME = 95.3±8.6%, n = 6, Fig. 1).

Fig. 1:	Effects of removal of the endothelium (denuded), indomethacin (10 μM), NG-nitro-L-arginine methyl ester (L-NAME, 300 μM) and a combination of indomethabin and L-NAME on vasorelaxation to CH2Cl2-AL in aortic rings precontracted with methoxamine. Data were shown as Mean±SEM

Fig. 2:	Effects of a high concentration of extracellular K+ (60 mM KCl) on vasorelaxation to CH2Cl2-AL in endothelium-intact and -denuded aortic rings precontracted with methoxamine. Data were shown as Mean±SEM

Fig. 3:	Effects of 4-aminopyridine (4-AP, 1 mM), barium chloride (BaCl2, 10 μM) and clotrimazole (2 μM) on vasorelaxation to CH2Cl2-AL in aortic rings precontracted with methoxamine. Data were shown as Mean±SEM

Fig. 4:	Effects of glibenclamide (10 μM) and 1-[(2-chlorophenyl)diphenylmethyl]-1H-pyrazole (TRAM-34, 10 μM) on vasorelaxation to CH2Cl2-AL in aortic rings precontracted with methoxamine. Data were shown as Mean±SEM

The effects of high extracellular potassium and potassium channel inhibitors on vasorelaxation to CH₂Cl₂-AL: Raising extracellular K⁺ concentration (60 mM KCl) significantly (p<0.001) inhibited vasorelaxation induced by CH₂Cl₂-AL in endothelium-intact and-denuded rings (R_max: 60 mM KCl = 63.1±9.1%, n = 6; 60 mM KCl with denuded = 44.4±6.1%, n = 6, Fig. 2). However, there was no significant difference between two groups.

Fig. 5:	Effects of CH2Cl2-AL on CaCl2-induced contraction in aortic rings depolarized by 100 mM KCl. Data were shown as Mean±SEM

In intact endothelium, vasorelaxant effects of CH₂Cl₂-AL were significantly (p<0.05) inhibited by either BaCl₂, or clotrimazole, but not 4-AP (R_max: BaCl₂ = 101±5%, n = 6; clotrimazole = 91.7±10.8%, n = 6, Fig. 3). Moreover, pretreatment with glibenclamide or TRAM-34 significantly (p<0.001) inhibited vasorelaxation induced by CH₂Cl₂-AL (R_max: glibenclamide = 82.5±4.35%, n = 6; TRAM-34 = 91.0±4.7%, n = 6, Fig. 4).

The effects of CH₂Cl₂-AL on CaCl₂-induced contraction in rat aortic rings: CaCl₂ (10 μM-30 mM) elicited concentration-dependent contraction of KCl (100 mM) depolarized rings in Ca²⁺-free medium. CH₂Cl₂-AL concentration-dependently reduced contractions induced by CaCl₂ (R_max: control = 1.09±0.04 g, n = 6; 1 μg mL^-1 CH₂Cl₂-AL = 0.92±0.04 g, n = 6; 10 μg mL^-1 CH₂Cl₂-AL = 0.54±0.01 g, n = 6; 100 μg mL^-1 CH₂Cl₂-AL = 0.17±0.02 g, n = 6, Fig. 5).

DISCUSSION

The present study has demonstrated, for the first time, about the vasorelaxant effects of CH₂Cl₂-AL and mechanisms involved in its action. CH₂Cl₂-AL causes vasorelaxation in the isolated rat aorta, which is partly due to endothelium-derived No. Activation of K⁺ channels and inhibition of Ca²⁺ influx are largely involved in the vasorelaxant effects of CH₂Cl₂-AL.

Previous studies have reported that the extract of Anaxagorea luzonensis A.Gray contains several flavones and flavonoids (Kitaoka et al., 1998; Gonda et al., 2000). Therefore, vasorelaxant responses of rat aortic rings to CH₂Cl₂-AL may be caused by the effects of these compounds. The present findings are in agreement with several earlier investigations with plant extract showing that flavonoids exert a vasorelaxant property (Herrera et al., 1996; Ajay et al., 2003; Gilani et al., 2006; Morello et al., 2006). Recently, studies by our group have shown that the ethanolic extract of Kaempferia parviflora and its methoxyflavones cause vasorelaxation in the rat isolated aorta (Tep-Areenan et al., 2010; Tep-Areenan and Sawasdee, 2010).

The vascular endothelium plays an essential role in regulating vascular tone via synthesis and release Endothelium-Derived Relaxing Factors (EDRFs), such as NO and prostacyclin (Lokhande et al., 2006; Vanhoutte, 2009; Vanhoutte et al., 2009). The present experiments show that relaxations induced by CH₂Cl₂-AL were inhibited by removal of the endothelium. Moreover, the effects of CH₂Cl₂-AL were inhibited by L-NAME, a NOS inhibitor, but not inhibition of a COX pathway by indomethacin. These results indicate the participation of endothelium-derived NO in the effects of CH₂Cl₂-AL. The present study also found that pretreatment with L-NAME plus indomethacin showed no further inhibition of responses to CH₂Cl₂-AL than those observed with L-NAME alone. Therefore, these results indicate that the relaxant responses to CH₂Cl₂-AL involve NO, but not vasodilator prostanoids via the COX pathway. Although, previous studies have demonstrated that vasorelaxant effects of some flavones and flavonoids involve the COX pathway (Ajay et al., 2003; Uydes-Dogan et al., 2005), our CH₂Cl₂-AL constituents may be different. Further studies would be needed to identify the active ingredients in our extract.

Several types of K⁺ channels are located in vascular smooth muscle cells, including K_ATP, K_Ca, K_V, and K_IR channels. Opening of K⁺ channels in the cell membrane of vascular smooth muscle cells increases K⁺ efflux, causing hyperpolarization, which closure of voltage-gated Ca²⁺ channels and subsequently vasorelaxation (Jackson, 2000; Sobey, 2001). To investigate the involvement of K⁺ channels in vasorelaxant effects of CH₂Cl₂-AL, a high concentration of KCl was used to increase vascular tone. It was found that 60 mM KCl inhibited the effects of CH₂Cl₂-AL in both endothelium-intact and denuded aortic rings. However, there was no further inhibition of responses to CH₂Cl₂-AL in denuded rings, compared to intact rings. Taken together, it is suggested that opening of K⁺ channels on smooth muscle cells by CH₂Cl₂-AL is unlikely due to EDRFs.

In order to investigate the contribution of exact types of K⁺ channels involved in vasorelaxant effects of CH₂Cl₂-AL, we used different types of K⁺ channel inhibitors. It was found that vasorelaxation induced by CH₂Cl₂-AL were inhibited by barium chloride, a K_IR channel inhibitor, glibenclamide, a K_ATP channel inhibitor, clotrimazole, an intermediate-conductance K_Ca channels, TRAM-34, a specific inhibitor of intermediate-conductance K_Ca (IK_Ca) channels. Conversely, 4-aminopyridine, a K_V channel inhibitor, did not affect vascular responses to CH₂Cl₂-AL. These results suggest that vasorelaxant responses to CH₂Cl₂-AL are mediated by increasing K⁺ efflux, at least in part, through K_IR, K_ATP, IK_Ca channels.

Another aspect examined in the present study was whether vasorelaxant responses to inhibition of extracellular Ca²⁺ influx involved. We found that contractile responses of rat aortic rings to CaCl₂ in Ca²⁺ -free medium containing KCl were inhibited by CH₂Cl₂-AL. These findings support the notion that CH₂Cl₂-AL can block Ca²⁺ influx from the extracellular space. In addition, in the presence of CH₂Cl₂-AL, there were rightward shifts in concentration-response curves for CaCl₂ with reduction in the R_max values, thus suggesting that CH₂Cl₂-AL acts as a non-competitive calcium antagonist. Clinically, Ca²⁺ antagonist are potentially used to treat hypertension (Ishimitsu et al., 2009; Stokes, 2009; Adatia and Shekerdemian, 2010). Concerning the vasorelaxant effects of CH₂Cl₂-AL, it is likely to suggest that the extract of Anaxagorea luzonensis A. Grey may acts as antihypertensive agent.

CONCLUSION

From our data, it is possible to conclude that CH₂Cl₂-AL exerts its vasorelaxant effects by acting on multiple sites of actions. Vascular responses to CH₂Cl₂-AL are partly mediated by endothelium-dependent No. Additionally, inhibition of Ca²⁺ entry and activation of K⁺ channels are required for the vasorelaxant effects of CH₂Cl₂-AL. Further chemical and pharmacological experiments are required to isolate and identify the active constituents responsible for this vascular effect and to investigate the in vivo effect of GL extract.

ACKNOWLEDGMENTS

This study was funded by Srinakharinwirot University (Grant No. 375/2552). Financial support for P. Sawasdee through the National Research University Project of CHE and Ratchadaphiseksomphot Endowment Fund (FW001A) is very gratefully acknowledged. We would like to express our deepest gratitude to Dr. Alfredo Villarroel for improving the English. We also thank Mr. Phongphat Wetchasit for his technical support. The authors have no conflict of interest to report.