Cajaninstilbene Acid Relaxes Rat Renal Arteries: Roles of Ca2+ Antagonism and Protein Kinase C-Dependent Mechanism

Cajaninstilbene acid (CSA) is a major active component present in the leaves of Cajanus cajan (L.) Millsp. The present study explores the underlying cellular mechanisms for CSA-induced relaxation in rat renal arteries. Vascular reactivity was examined in arterial rings that were suspended in a Multi Myograph System and the expression of signaling proteins was assessed by Western blotting method. CSA (0.1–10 µM) produced relaxations in rings pre-contracted by phenylephrine, serotonin, 9, 11-dideoxy-9α, 11α-epoxymethanoprostaglandin F2α (U46619), and 60 mM KCl. CSA-induced relaxations did not show difference between genders and were unaffected by endothelium denudation, nor by treatment with NG-nitro-L-arginine methyl ester, indomethacin, ICI-182780, tetraethylammonium ion, BaCl2, glibenclamide, 4-aminopyridine or propranolol. CSA reduced contraction induced by CaCl2 (0.01–5 mM) in Ca2+-free 60 mM KCl solution and by 30 nM (−)-Bay K8644 in 15 mM KCl solution. CSA inhibited 60 mM KCl-induced Ca2+ influx in smooth muscle of renal arteries. In addition, CSA inhibited contraction evoked by phorbol 12-myristate 13-acetate (PMA, protein kinase C agonist) in Ca2+-free Krebs solution. Moreover, CSA reduced the U46619- and PMA-induced phosphorylation of myosin light chain (MLC) at Ser19 and myosin phosphatase target subunit 1 (MYPT1) at Thr853 which was associated with vasoconstriction. CSA also lowered the phosphorylation of protein kinase C (PKCδ) at Thr505. In summary, the present results suggest that CSA relaxes renal arteries in vitro via multiple cellular mechanisms involving partial inhibition of calcium entry via nifedipine-sensitive calcium channels, protein kinase C and Rho kinase.


Introduction
Cajaninstilbene acid (CSA, Figure 1), one of the main effective ingredients, is present in the leaves of Cajanus cajan (L.) Millsp (pigeon pea) [1] which is commonly used to treat ischemic necrosis of femoral head in traditional Chinese medicine. Recent studies show that the extracts or CSA, possess anti-microbial [2,3], antitumor [4], hepatoprotective [5,6] and anti-hyperglycemic [7] properties. CSA-containing extracts also protect against amyloidb 25-35-induced cognitive deficits in mice through increasing the activity of choline acetyl transferase and anti-oxidation [8]. CSA reduces radical and peroxide generation, inhibits xanthine oxidase activity, and protects from DNA damage in vitro [9,10]. The stilbene extracts containing CSA reverse the elevation of the concentration of follicle stimulating hormone and luteinizing hormone and improve femoral morphological structure similar to the effect produced by 17b-estradiol supplementation without affecting the serum 17b-estradiol level and uterine weight in ovariectomized rats, suggesting that CSA may exert a phytoestrogenic activity [11]. In addition, the extract containing 76% CSA markedly lowers levels of serum and hepatic total cholesterol, triglyceride and LDL cholesterol in diet-induced hypercholesterolemic mice, indicating that CSA could be potentially useful for the attenuation of atherosclerosis [12,13].
The pharmacological activity of CSA in the vascular system is unknown. It is possible that CSA also benefits vascular function. Therefore, the present study was designed to examine the cellular mechanisms for CSA-induced relaxation and roles of signaling molecules involved in the regulation of contractility in rat arteries.

Results
The Effect of Cajaninstilbene Acid (CSA) on Agonistsinduced Contraction CSA produced concentration-dependent relaxations to similar degrees in isolated renal arteries pre-contracted with 60 mM KCl, phenylephrine, serotonin and U46619 ( Figure 2, Table 1), while the vehicle (DMSO) showed no effect on contraction.

Roles of the Endothelium and Estrogen Receptor
CSA-induced relaxations showed no gender difference since they were comparable in both male and female rat arteries ( Figure 3A). Treatment with ICI-182780 (estrogen receptor antagonist at 10 mM, Figure 3B), L-NAME (nitric oxide synthase inhibitor at 100 mM) and indomethacin (non-selective cyclooxygenase inhibitor at 3 mM) ( Figure 3D), or mechanical removal of endothelium ( Figure 3C) did not modulate the relaxant effect of CSA.

Roles of Calcium Channel Inhibition in CSA-induced Relaxations
To test the possible role of inhibiting Ca 2+ influx in CSAinduced relaxation, the arteries were incubated in a Ca 2+ -free depolarizing solution containing 60 mM KCl. The representative trace in Figure 4A shows that the addition of CaCl 2 into this bathing solution caused concentration-dependent contractions. Thirty-minute treatment with CSA (0.1-10 mM) inhibited CaCl 2evoked contraction while nifedipine (L-type calcium channel blocker at 100 nM) was used as positive control ( Figure 4B). In addition, CSA inhibited the contraction induced by (2)-Bay

Roles of Inhibition of Rho Kinase and Protein Kinase Cdependent Mechanisms
CSA partly suppressed the sustained contraction evoked by phorbol 12-myristate 13-acetate (PMA, protein kinase C activator at 10 mM) in a Ca 2+ -free Krebs solution ( Figure 6A and B).
Treatment of arteries with 30 nM U46619 for 30 min increased the phosphorylation of MLC at Ser19 ( Figure 7A) and MYPT1 at Thr853 ( Figure 7B), which were reversed by co-treatment of 10 mM CSA. Furthermore, after treating the arteries with 10 mM PMA in a Ca 2+ -free solution for 60 min, the levels of phosphorylation of MLC at Ser19 ( Figure 7C), MYPT1 at Thr853 ( Figure 7D) and PKCd at Thr505 ( Figure 7E) were all elevated as compared to the control. Co-treatment with 10 mM CSA reversed the phosphorylation of these signaling molecules ( Figure 7).

Discussion
The present study examined the vascular reactivity of cajaninstilbene acid in rat renal arteries and provided novel findings regarding its pharmacological properties. CSA produced endothelium-independent relaxation (1) partly through antagonism of Ca 2+ influx via nifedipine-sensitive Ca 2+ channel and (2) partly through inhibition of Rho kinase and PKC-dependent contractile mechanisms in vascular smooth muscle cells.
Previous studies showed that CSA possesses an estrogen-like activity on osteoblast and osteoclast [14] or ovariectomy-induced bone loss in rats [11]. The present results, however, do not show a gender difference as CSA-induced relaxations were similar in arteries from both male and female rats. In addition, treatment with ICI-182780, a classic estrogen receptor antagonist, did not affect the relaxant effect of CSA.
The endothelium regulates vascular tone, while hypertension and atherosclerosis are associated with the impaired endothelial function; the latter is usually caused by disturbed balance in endothelium-derived relaxing and contracting factors [15]. CSAinduced relaxations were independent of the presence of the intact endothelium nor affected by L-NAME, indomethacin or in arteries contracted by 60 mM KCl, suggesting negligible roles for endothelium-derived relaxing factors. CSA relaxed renal arteries constricted by U46619, serotonin and phenylephrine with similar potency and propranolol was without effect on the CSA relaxation, thus ruling out the interaction of CSA with receptors. It is probable that CSA directly acts on vascular smooth muscle to cause relaxation. CSA-induced relaxations were not affected by elevated KCl nor by individual blockers for various types of K + channels, thus discounting the involvement of K + channel. Elevated KCl is known to activate voltage-gated Ca 2+ channels via membrane depolarization in VSMCs. Therefore, high KCl-induced contraction was reduced by CSA, suggesting that CSA is likely to interfere with Ca 2+ influx via Ca 2+ channels. This notion was further supported by the following three observations. First, CSA progressively inhibited Ca 2+ -triggered contraction in a Ca 2+ -free, 60 mM KCl-containing Krebs solution. Second, CSA also concentration-dependently reduced contractions evoked by the Ca 2+ channel activator (2) The present results also suggest that in addition to calcium antagonism other cellular mechanisms may contribute to CSAinduced relaxations as CSA is more effective in relaxing arteries contracted by receptor agonists than by elevated KCl. Constrictive agonists used in this study can activate MLCK, PKC and Rho kinase; the latter two are independent of intracellular Ca 2+ rise [16,17,18]. Both PKC and Rho kinase are involved in the development of hypertension, cerebral and coronary vasospasm, ischemia/reperfusion injury and athero-  sclerosis [19,20,21]. Hence, they can become promising therapeutic targets for the treatment of cardiovascular events. The present results show that CSA partly decreased contraction triggered by the exogenous PKC activator PMA in a Ca 2+ -free Krebs solution, suggesting that PKC inhibition may be involved. The downstream targets of both calcium-dependent andindependent mechanisms mediating vascular smooth muscle contraction are phosphorylation of myosin light chain (MLC) and myosin phosphatase target subunit 1 (MYPT1) [18]. Phosphorylation of MYPT1 decreases its activity to dephosphorylate MLC; and subsequently leading to sustained contraction induced by phosphorylation of MLC [22]. PMA was shown to stimulate phosphorylation of MYPT1 at Thr853 [23,24] and MCL at Ser19 [25]. U46619 is known to trigger Rho kinase and then phosphorylate MLC and MYPT1 while PMA is the PKC activator. The present study shows that CSA reduced U46619-and PMA-induced phosphorylation of MYPT1 and MLC and phosphorylation of PKCd. Taken together with the functional results, inhibition of PKC and Rho kinase signaling pathways with reduced phosphorylation of MYPT1 and MLC is likely to account for the part of CSA-induced vasorelaxation.
In summary, the present study provides novel evidence showing that CSA relaxes renal arteries in vitro likely through both antagonism of calcium entry via nifedipine-sensitive Ca 2+ channel and inhibition of cellular pathways in association with PKC and Rho/Rho kinase. The present results indicate that CSA and CSA-containing herbs can be of potential benefits in reducing the elevated VSMC tension which is implicated in cardiovascular pathogenesis although such benefit needs in-depth investigation in animal models of hypertension and other vascular pathologies.

Artery Rings Preparation
The investigation conforms to the Guidelines for the Care and Use of laboratory animals published by the National Institutes of Health. Both male and female Sprague-Dawley rats (250,300 g) were killed by cervical dislocation. The kidneys were removed and dissected in ice-cold oxygenated Krebs solution. Krebs solution contained the following composition (in mM): 119 NaCl, 4.7 KCl, 1 MgCl 2 , 2.5 CaCl 2 , 25 NaHCO 3 , 1.2 KH 2 PO 4 and 11 Dglucose.

Isometric Force Measurement
Rings (,2 mm in length) of renal arteries were isolated from adhering connective tissues. Each segment was mounted in a Multi Myograph System (Danish Myo Technology A/S, Denmark), bathed in Krebs solution bubbled with 95% O 2 -5% CO 2 and maintained at 37uC at pH ,7.4 as described previously [26]. Renal arterial rings were set to an optimal tension of 2 mN and stabilized for 90 min. The rings were then contracted by 0.5 mM phenylephrine and challenged with 3 mM acetylcholine to confirm the integrity of the endothelium. In some protocols, the endothelium was mechanically removed via rubbing the internal surface of arteries with a stainless steel wire and verified by the lack of relaxation in response to 3 mM acetylcholine.
Finally, the calcium-independent mechanisms were studied in endothelium-denuded rings. Each ring was contracted by phorbol 12-myristate 13-acetate (PMA at 10 mM) in a Ca 2+ -free Krebs solution containing 100 mM EGTA and then relaxed by CSA.

Calcium Fluorescent Imaging in Renal Artery
Renal arteries were dissected free from connective tissue and cut open longitudinally. The vascular strips were incubated with 5 mM calcium indicator fluo-4 AM (Molecular Probes) in Krebs solution for 1 h at 37uC as described before [30]. Some strips were cotreated with 10 mM CSA or 100 nM nifedipine for 30 min. The strips were then bathed in organ chambers filled with Krebs solution bubbled by 95% O 2 and 5% CO 2 and 10 mM BDM (2,3butanedione monoxime, myosin inhibitor, Sigma) at 37uC. Using an Olympus Fluoview FV1000 laser scanning confocal system (Olympus), fluorescence was measured continuously every 20 s (excitation: 495 nm and emission: 505-525 nm). The fluorescence intensity at a certain time point (F 1 ) was compared to that at the starting point of image recording (F 0 ) to show the changes of intracellular Ca 2+ concentration [Ca 2+ ] i upon the addition of 60 mM KCl.

Statistical Analysis
Data are means6SEM of n rats. pD 2 refers to the negative logarithm of the dilator concentration that caused half of the maximal relaxation and E max % refers to the maximum relaxation. Concentration-response curves were analyzed via GraphPad software (Version 4.03) and statistical significance was compared by two-tailed Student's t-test or one-way analysis of variance followed by Newman-Keuls test. Values of p,0.05 was considered statistically significant.