j ourna l homepage: www.e lsev ie r.com/ locate /vph
Vasodilatation produced by Fasudil Mesylate in vivo and in vitro
Qin Li a,b,1, Yuhua Chen c,1, Li Sun c, Gang Fu a, Lianjun Guo a,⁎a Department of Pharmacology, Tongji Medical College of Huazhong University of Science and Technology, Wuhan, Chinab Department of Pharmacology, Shanghai First People's Hospital, School of medicine, Shanghai Jiao Tong University, Shanghai, Chinac Department of neurology, Wuhan Brain Hospital (General Hospital of the Yangtze River shipping), Wuhan, China
To investigate the vasorelaxant effect of fasudil mesylate (FM) in vivo and in vitro. The relaxation effect of FMwas studied using cerebral vasospasm (CVS) model in vivo and isolated aortic rings in vitro. FM (0.35, 1.2,3.5 mg·kg−1) increased cerebrovascular flow (CVF) and femoral blood flow (FBF) dose-dependently in vivo,however, the relaxation effects of FM on cerebral vessels were much stronger than on peripheral vessels; FMshowed dose-dependent relaxation of isolated aortic rings contracted by Methoxamine (Met) or KCl in vitro.The relaxation IC50 of FM and Prasozin (Pra) to the rabbit aortic rings contracted by Met are 27.54 μM and0.01 μM respectively, and the relaxation IC50 of FM to the rabbit and rat aortic rings contracted by KCl are37.15 μM and 0.74 μM respectively. In addition, there is no obvious difference between endothelium-intactand endothelium-removal groups. The Met dose–effect relationship curve was significantly shifted to right byFM (0.3, 3 μM), and Emax was decreased (Pb0.05). The relaxation effect of FM on cerebral vessels was muchstronger than on peripheral vessels in vivo, and the action is in an endothelium-independent manner.
Cerebral vasospasm (CVS) is the most common cause ofdisability andmorbidity in patients with subarachnoid hemorrhage(SAH) that occurs when there is a narrowing of the arteries in thesubarachnoid space (Treggiari-Venzi et al., 2001). Angiographicevidence of vasospasm has been reported in approximately two-thirds of patients with SAH (Riento and Ridley, 2003). Induced SAHin the canine model produces a significant impairment in regionalcerebral blood flow (Bassiouni et al., 2007). Treatment of cerebralvasospasm is considered as a major goal in the management ofpatients surviving SAH. Despite extensive investigations over morethan 40 years, the pathogenesis of CVS and optimal treatment havenot been fully established. Because the pathogenesis of CVS ismultifactorial.
Rho kinase (ROCK), a serine/threonine kinase, is reported to beinvolved in a wide of fundamental cell functions such as smoothmuscle contraction, cell migration, proliferation, and survival(Shimokawa and Takeshita, 2005; Somlyo and Somlyo, 1994).Hydroxy fasudil inhibits ROCK with IC50 value of 0.9–1.8 mM, whileits inhibitory effect is markedly less for myosin light chain kinase(Shimokawa et al., 1999). It was reported that Hydroxy fasudilrelaxed canine basilar or middle cerebral arterial strips precon-tracted by KCl, PGF2α or U-46619(Satoh et al., 2001; Ito et al., 2003),and increased cerebral blood flow (Asano et al., 1987). Fasudil
mesylate (FM) is derived from fasudil hydrochloride (HF) with ahydrochloride group replaced by mesylate, and the effect of FM onvessels is not explained.
In present, blood injection technique is commonly used in the invivo model of CVS through nyxis in the foramen magnum or opticchiasma, followed by injection with nonheparinized autogenousarterial blood to induce delayed cerebral vasospasm (Wang et al.,2011). Then the diameter of basilar artery was measured byangiography, which is used to judge the extent of CVS and the effectof drug. This structure determination method is direct. But in fact, thevessel diameter and blood flow is non-linear relationship. The bloodflow is regulated mainly by blood pressure and vessel resistance. Inshort, if the vessel diameter increases, the blood flow does notnecessarily increase. So there is a certain gap between the structuredetermination method and the therapeutic purpose. In our study,canine CVSwas produced by Cafergot solution administration throughduodenum, followed by cerebral blood flow determination. The bloodflow determination, which belongs to functional determinationmethod, is closer to clinical treatment purposes. In current study,the vasodilatation effects of fasudil mesylate in vivo by blood flowdetermination method and the possible mechanisms in vitro wereinvestigated.
2. Materials and methods
2.1. Drugs and chemicals
Fasudil mesylate (drug grade) (synthesized by Wuhan pharma-ceutical Co Ltd., Wuhan, China) is derived from fasudil hydrochloride
Fig. 1. Comparison of HR and MABP after CVS between Model group and Drugadministration groups. (A) Effect of FM, HF, and Nimo on HR after CVS. There was noobvious change on HR after FM, HF, and Nimo administration. (B) Effect of FM, HF, andNimo on MABP after CVS. There were also no change on MABP after FM (0.35,1.2 mg/kg) administration. (C) MABP decreased from 5th to 60th min after CVS in thepresence of FM 3.5 mg/kg (⁎Pb0.05 vs. Model group).
122 Q. Li et al. / Vascular Pharmacology 55 (2011) 121–126
(HF) with a hydrochloride group replaced by mesylate. It is dissolvedin normal saline and preserved away from light, and its chemicalstructure is as follows (as follow fig). Nimodipine (Nim) was obtainedfrom Bayer healthcare AG, Germany. The following reagents werepurchased: Fasudil hydrochloride (HF), Acetylcholine (Ach), Methox-amine (Met), Prasozin (Pra), all from Sigma (St. Louis, MO, USA) andthe other reagents used were of analytical grade.
2.2. Animals
Adult dogs weighing 11.8±1.4 kg were obtained from Experi-mental Animal Center of Tianjin Institute of Pharmaceutical Research(Tianjin, China), and New Zealand rabbits weighing 2.5–3.5 kg andWistar rats weighing 200–250 g from Experimental Animal Center ofTongji Medical College of Huazhong University of Science andTechnology (Wuhan, China). They were housed in a temperatureand humidity-controlled room (temperature: 22±1 °C, humidity:60%) with free access to food and water. The animals were kept on a12-h light/dark cycle and adapted to these conditions for at least7 days before experiments. All procedures used in this study complywith the Guide for the Care and Use of Laboratory Animals (NationalInstitutes of Health).
2.3. Cerebral vasospasm (CVS)
Dogs were anesthetized using 3% pentobarbital (30 mg/kg i.v.).The right femoral artery and vein were separated. The artery wasused to monitor the systolic blood pressure (SBP), diastolic bloodpressure (DBP), and mean arterial blood pressure (MABP), and thevein was used for the drug injection. At the same time, the leftfemoral artery was separated to monitor femoral blood flow (FBF).The right internal carotid artery and vertebral artery wereseparated respectively to monitor the internal carotid flow (ICF)and vertebral blood flow (VBF). After the blood flow were stable,canine CVS was produced by Cafergot solution ((ergotamine0.5 mg+caffeine 50 mg)/kg) administration through duodenum.The cerebrovascular resistance (CVR) significantly increased45 min after Cafergot injection and sustained stable in thefollowing 120 min. Dogs were randomly divided into six groups,namely, model, FM-treated groups (0.35, 1.2, 3.5 mg/kg), HF-treated group (1.0 mg/kg) and nimodipine-treated group(0.04 mg/kg). The different drugs (i.v. drip) were administrated45 min after Cafergot injection for 60 min, and the dogs were killed120 min after Cafergot injection and the brain was weighted. TheCVR, CVF, FVR were calculated as follows: CBF (mL·min−1·100 gbrain−1)=[(VBF+ICF)×100] / (the weight of brain×0.5); CVR(mm Hg·mL−1·min−1·100 g brain−1)=MABP /CVF; FVR (mmHg·mL−1·min−1)=MABP/FBF.
2.4. Preparation of aortic rings
Rats and rabbits were anesthetized and sacrificed. Fats andconnective tissues were removed from the thoracic aorta, and 3-mm-wide aortic rings were prepared. For an endothelium-free aorta,
the endothelial lining of each ring was removed by pressing the ringand rolling it gently onto a filter paper a few times. The endotheliumwas considered to be intact when relaxation induced by Ach (1 μM)was over 20% of the maximal tension obtained by 60 mMKCl-inducedcontraction, and the removal of the endothelium was confirmed bythe absence of Ach-induced relaxation (Yanaga et al., 2006; Li et al.,2010; Wang et al., 2009).
2.5. Vasodilative effect of fasudil mesylate on isolated aortic rings
Theaortic ringsweremountedon steel hooks ina chamber.Oneendofthe ringwasattached to a force-displacement transducer for recording theisometric contraction. The baths were filled with 10ml of Kreb's solutioncontaining the following (mM): NaCl 120, KCl 4.7, NaHCO3 25.0, KH2PO4
1.2, MgSO4•7H2O 1.2, CaCl2 2.5, and glucose 10.0. The solution wasmaintained at 37 °C and bubbled continuouslywith 5% CO2 and 95% O2 atPH 7.4. The rabbit and rat rings were equilibrated for 45 min at an initial
Fig. 2. Comparison of CBF and CVR after CVS between Model group and Drug administration groups. Compared with model group, the CVR decreased and CBF increased after FMadministration dose-dependently (n=6). (A) Effect of FM, HF, Nimo on CBF after CVS. (B) Effect of FM, HF, Nimo on CVR after CVS. (C) Compared with model group, the CBFincreased from 15th min to 90th min after CVS in FM 0.35 mg/kg group; the CBF increased from 15th min to 120th min in FM 1.2 mg/kg group; and the CBF increased from 15th minto 120thmin in FM 3.5 mg/kg group. (D) Compared withmodel group, the CVR decreased from 15thmin to 30thmin after CVS in FM 0.35 mg/kg group; the CVR decreased from 15thmin to 120th min in FM 1.2 mg/kg group; and the CVR decreased from 5th min to 120th min in FM 3.5 mg/kg group (⁎Pb0.05 vs. Model group).
123Q. Li et al. / Vascular Pharmacology 55 (2011) 121–126
resting tension of 4 g and 1 g respectively. During this period, the Kreb'ssolution in the bath was replaced every 15 min. KCl (60 mM) was addedtocontract theaortic rings.After the contraction reachedaplateau, 1 μMofAch was added. When endothelium was removed, the Ach-inducedrelaxation disappeared.
The rabbit and rat aortic rings with endothelium or withoutendothelium were contracted by Met, KCl. When the contractionreached a plateau, FM or Prazosin (Pra) was added to the bath incumulatively increasing doses. Relaxation was expressed as percent-age of the decrease in maximal tension obtained by Met or KCl-induced contraction.
FM (0.3, 3 μM) were added 10 min before the construction of Metconcentration-response curves. The results were expressed as thepercentage of the maximum contractile tension to Met before andafter pretreatment with FM.
2.6. Statistical analysis
Data are expressed as the means±S.D. and analyzed using generallinear modeling repeated measure analyses and t-test with asignificance level of Pb0.05. IC50 values were calculated by sigmaplot 8.0.
124 Q. Li et al. / Vascular Pharmacology 55 (2011) 121–126
3. Results
3.1. Effect of FM on SBP, DBP, MABP, and HR after CVS in vivo
The cerebrovascular resistance (CVR) significantly increased45 min after Cafergot injection compared with before injection(Pb0.05) and sustained stable in the following 120 min. Kinds ofphysical parameters were monitored and observed for 120 min afterCVS in the presence of FM, HF, nimodipine (Nim) or Saline. There wasno obvious change on HR after FM (0.35, 1.2, 3.5 mg/kg) administra-tion. Therewere also no change on SBP, DBP, andMABP after FM (0.35,1.2 mg/kg) administration, however, SBP, DBP, and MABP decreased
Fig. 3. Comparison of FBF and FVR after CVS betweenModel group and Drug administration gFVR after CVS. (C) Compared with model group, FBF increased at 45th min after CVS in FM 0.FBF increased from 30th min to 120th min in FM 3.5 mg/kg group. (D) Compared with modFVR stayed stable in FM 1.2 mg/kg group; and the FVR decreased from 20th min to 120th m
from 5th to 60th min after CVS after FM (3.5 mg/kg) administration(some data not shown) (Fig. 1).
3.2. Effect of FM on CVR and CBF after CVS in vivo
Compared with model group, the CVR decreased from 15th min to30th min and CBF increased from 15th min to 90th min after CVS inthe presence of FM 0.35 mg/kg (Pb0.05); the CVR decreased from15th min to 120th min and CBF increased from 15th min to 120th minafter CVS in the presence of FM 1.2 mg/kg (Pb0.05); and the CVRdecreased from 5th min to 120th min and CBF increased from 15th
roups. (n=6). (A) Effect of FM, HF, Nimo on FBF after CVS. (B) Effect of FM, HF, Nimo on35 mg/kg group; FBF increased from 45th min to 120th min in FM 1.2 mg/kg group; andel group, there was no significant change on FVR after CVS in FM 0.35 mg/kg group; thein in FM 3.5 mg/kg group (⁎Pb0.05 vs. Model group).
Fig. 4. Dose-dependent vasodilative effect of FM on endothelium-denuded ( ) andendothelium-intact ( ) aortic rings contracted by KCl (n=7–9). A. The vasodilativeeffect of FM on rat aortic ring contracted by KCl. B. The vasodilative effect of FM onrabbit aortic ring contracted by KCl.
125Q. Li et al. / Vascular Pharmacology 55 (2011) 121–126
min to 120th min after CVS in the presence of FM 3.5 mg/kg (Pb0.05)(Fig. 2).
3.3. Effect of FM on FVR and FBF after CVS in vivo
Compared with model group, there was no significant change onFVR, but FBF increased at 45th min in the presence of FM 0.35 mg/kg(Pb0.05); the FVR stayed stable but FBF increased from 45th min to120th min after CVS in the presence of FM 1.2 mg/kg (Pb0.05); andthe FVR decreased from 20th min to 120th min and FBF increasedfrom 30thmin to 120thmin after CVS in the presence of FM 3.5 mg/kg(Pb0.05) (Fig. 3).
3.4. Vasodilative effect of FM on isolated aortic rings contracted by KCl invitro
Fig. 4 shows that FM caused dose-dependent vasorelaxation onKCl-contracted aortic rings with or without endothelium respectively,and there was no significant differences between two groups(PN0.05). Furthermore, the relaxation IC50 of FM to the rabbit andrat aortic rings contracted by KCl are 37.15 μM and 0.74 μMrespectively.
3.5. Vasodilative effect of FM on isolated aortic rings contracted by Metin vitro
Fig. 5 illustrates that FM showed dose-dependent vasorelaxationin Met-induced contraction of rabbit aortic rings with or withoutendothelium respectively. There was no significant difference be-tween two groups (PN0.05). The relaxation IC50 of FM and Prazosin tothe rabbit aorta rings contracted by Met are 27.54 μM and 0.01 μMrespectively.
Fig. 5. Dose-dependent vasodilative effect of FM on endothelium-denuded ( ) andendothelium-intact ( ) aortic rings contracted by Met (n=7–9). A. The vasodilativeeffect of FM on rabbit aortic ring contracted by Met. B. The vasodilative effect of Pra onrabbit aortic ring contracted by Met.
3.6. Effect of FM on Met dose–response curve in vitro
To evaluate the possible α-receptor antagonism of FM, a series ofexperiments were performed, based on contracting the rabbit aorticrings preparations with increasingMet concentrations in the presenceor absence of FM. As shown in Fig. 6, theMet dose–response curvewassignificantly shifted to right by FM (0.3, 3 μM), and the Emax wasdecreased (Pb0.05) (Fig. 6).
4. Discussion
In the present study, CVS was produced by Cafergot solutioninjection through duodenum, and the CVR was decreased and the CVFincreased obviously for two hours by FM (0.35, 1.2, 3.5 mg/kg)administration in a dose-dependent manner (Fig. 2). FVR wasdecreased and FBF increased obviously in FM (3.5 mg/kg) group(Fig. 3). At the same time, there was no obvious change on HR, SBP,DBP, and MABP after FM (0.35, 1.2 mg/kg) administration (Fig. 1).Though MABP dropped from 5th to 60th min after FM (3.5 mg/kg)administration, the HR stand stable (Fig. 1). It may be related to theinhibition of reflex after the anesthesia. These results imply that therelaxation effects of FM on cerebral vessels were much stronger thanon peripheral vessels. FM could ensure the supply of the brain bloodflow and improve the cerebral hypoxia without affecting systemicblood pressure. Although our findings suggest that the ability of FM(1.2 mg/kg) on CBF and CVR was similar to that of HF (1.0 mg/kg), FMmight have better stability and liposolubility than HF. Nimodipine(Nim) is an established vasodilator after cerebral vasospasm (Bilgineret al., 2009) and used in our study as a control to compare thepharmacodynamic action of Nim with FM. Our findings suggest thatthe effect of FM (3.5 mg/kg) on CVR and CBF was stronger than that ofNim (0.04 mg/kg), and the ability of FM (1.2 mg/kg) on CVR and CBFwas weaker than that of Nim (0.04 mg/kg). Although Nim seems to bewell-tolerated in patients, various adverse reactions were observed,such as: decreased blood pressure, abnormal liver function, edema,and so on. Because of the selectivity of FM on blood vessels, FM mightbe associated with fewer side effects.
Methoxamine (Met) is a selective α1-adrenoceptor agonist, andinduce vasoconstriction by intracellular Ca2+ release, and KCl causesvasoconstriction by activating voltage-dependent calcium channel(Putney, 1990; Fasolato et al., 1994; Tunçtan et al., 2000). We foundthat FM was able to inhibit both Met and KCl-induced contractions ina concentration-dependent manner (Figs. 4 and 5), which means thevasodilation effect of FM was independent of intracellular Ca2+
release and voltage-dependent calcium channel activation. Prazosin isa selective α1-receptor blocker, and relaxes the rabbit aorta ringscontracted by Met with IC50 0.01 μM. Compared with Prazosin, therelaxation IC50 of FM is 27.54 μM. Furthermore, the vasorelaxanteffects of FM were confirmed both on the rabbit and rat aortic ringscontracted by KCl. The relaxation IC50 of FM to the rabbit and rat aorticrings contracted by KCl are 37.15 μM and 0.74 μM respectively.
It is known that vasodilation, induced by acetylcholine (Furchgottand Zawadzki, 1980), needs the endothelium layer and endothelium-derived relaxing factor, which has been demonstrated to be nitricoxide (NO). NO acts on the vascular cells, through the stimulation of
Fig. 6. Met dose–response curve of isolated rabbit aorta in the absence ( ) andpresence of FM 0.3 μM ( ), FM 3.0 μM ( ) (⁎Pb0.05 vs. control group).
126 Q. Li et al. / Vascular Pharmacology 55 (2011) 121–126
the soluble enzyme guanylate cyclase and elevation of the cytosoliccGMP (Calderone et al., 1999). Indeed, pretreatment with FM (0.3–100 μM) inhibited Met/KCl-induced contractions in endotheliumdenuded aortic rings, supporting a direct action on the vascularsmooth muscle cells. There were no statistical differences betweenendothelium-intact and endothelium-denuded aorta (Figs. 4 and 5).The relaxation effect of FM was in an endothelium-independentmanner. The Met dose–effect relationship curve was significantlyshifted to right by FM (0.3, 3 μM), and the Emax was decreased. FMinhibitedMet-induced vasoconstriction in a non-competitive manner.
As mentioned above, the relaxation effects of FM on cerebralvessels weremuch stronger than on peripheral vessels in vivo, and theaction is in an endothelium-independent manner. Further studies arerequired to elucidate the underlying mechanistic details.
Acknowledgments
This research was supported by the National Foundation of Natureand Science of China (No. 30772559), and the Hospital Foundation ofFirst People's Hospital Affiliated to Shanghai Jiao Tong University(0913).
References
Asano, T., Ikegaki, I., Satoh, S., Suzuki, Y., Shibuya, M., Takayasu, M., Hidaka, H., 1987.Mechanism of action of a novel antivasospasm drug, HA1077. Pharmacol. Exp. Ther.241, 1033–1040.
Bassiouni, H., Schulz, R., Dörge, H., Stolke, D., Heusch, G., 2007. The impact ofsubarachnoid hemorrhage on regional cerebral blood flow and large-vesseldiameter in the canine model of chronic vasospasm. J. Stroke Cerebrovasc. Dis.16, 45–51.
Bilginer, B., Onal, M.B., Narin, F., Soylemezoglu, F., Ziyal, I.M., Ozgen, T., 2009. The effectsof intravenous cilostazol and nimodipine on cerebral vasospasm after subarachnoidhemorrhage in an experimental rabbit model. Turk. Neurosurg. 19, 374–379.
Calderone, V., Martinotti, E., Baragatti, B., Breschi, M.C., Morelli, I., 1999. Vascular effectsof aqueous crude extract of Atemisia verlotorum Lamotte (Compositae): in vivoand in vitro pharmacological studies in rats. Phytother. Res. 13, 645–648.
Fasolato, C., Innocenti, B., Pozzan, T., 1994. Receptor-activated Ca2+ influx: how manymechanisms for how many channels. Trends Pharmacol. Sci. 15 (3), 77–83.
Furchgott, R.F., Zawadzki, J.V., 1980. The obligatory role of endothelial cells in therelaxation of arterial smooth muscle by acetylcholine. Nature 288, 373–376.
Ito, K., Shimomura, E., Iwanaga, T., Shiraishi, M., Shindo, K., Nakamura, J., Naqumo, H.,Seto, M., Sasaki, Y., Takuwa, Y., 2003. Essential role of rho kinase in the Ca2+
sensitization of prostaglandin F(2alpha)-induced contraction of rabbit aortae.Physiology 546, 823–836.
Li, Y.J., Bao, J.X., Xu, J.W., Murad, F., Bian, K., 2010. Vascular dilation by paeonol-amechanism study. Vascul. Pharmacol. 53, 169–176.
Mol. Cell Biol. 4, 446–456.Satoh, S., Utsunomiya, T., Tsurui, K., Kobayashi, T., Ikegaki, I., Sasaki, Y., Asano, T., 2001.
Pharmacological profile of hydroxy fasudil as a selective rho kinase inhibitor onischemic brain damage. Life Sci. 69, 1441–1453.
Shimokawa, H., Takeshita, A., 2005. Rho-kinase is an important therapeutic target incardiovascular medicine. Arterioscler. Thromb. Vasc. Biol. 25, 1767–1775.
Shimokawa, H., Seto, M., Katsumata, N., Amano, M., Kozai, T., Yamawaki, T., Kuwata, K.,Kandabashi, T., Egashira, K., Ikegaki, I., Asano, T., Kaibuchi, K., Takeshita, A., 1999.Rho-kinase-mediated pathway induces enhanced myosin light chain phosphory-lations in a swine model of coronary artery spasm. Cardiovasc. Res. 43, 1029–1039.
Somlyo, A.P., Somlyo, A.V., 1994. Signal transduction and regulation in smooth muscle.Nature 372, 231–236.
Treggiari-Venzi, M.M., Suter, P.M., Romand, J.A., 2001. Review of medical prevention ofvasospasm after aneurysmal subarachnoid hemorrhage: a problem of neurointen-sive care. Neurosurgery 48, 249–261.
Tunçtan, B., Altuğ, S., Uludağ, O., Abacioğlu, N., 2000. Effects of econazole on receptor-operated and depolarization-induced contractions in rat isolated aorta. Life Sci. 67,2393–2401.
Wang, Y.K., Ren, A.J., Yang, X.Q., Wang, L.G., Rong, W.F., Tang, C.S., 2009. Sulfur dioxiderelaxes rat aorta by endothelium-dependent and -independent mechanisms.Physiol. Res. 58, 521–527.
Wang, Z.P., Chen, H.S., Wang, F.X., 2011. Influence of plasma and cerebrospinal fluidlevels of endothelin-1 and no in reducing cerebral vasospasm after subarachnoidhemorrhage during treatment with mild hypothermia, in a dog model. CellBiochem. Biophys 61, 137–143.
Yanaga, A., Goto, H., Nakagawa, T., Hikiami, H., Shibahara, N., Shimada, Y., 2006.Cinnamaldehyde induces endothelium-dependent and -independent vasorelaxantaction on isolated rat aorta. Biol. Pharm. Bull. 29, 2415–2418.
