Mechanisms of Improved Aortic Stiffness by Arotinolol in Spontaneously Hypertensive Rats

Objectives This study investigates the effects on aortic stiffness and vasodilation by arotinolol and the underlying mechanisms in spontaneously hypertensive rats (SHR). Methods The vasodilations of rat aortas, renal and mesenteric arteries were evaluated by isometric force recording. Nitric oxide (NO) was measured in human aortic endothelial cells (HAECs) by fluorescent probes. Sixteen-week old SHRs were treated with metoprolol (200 mg·kg-1·d-1), arotinolol (30 mg·kg-1·d-1) for 8 weeks. Central arterial pressure (CAP) and pulse wave velocity (PWV) were evaluated via catheter pressure transducers. Collagen was assessed by immunohistochemistry and biochemistry assay, while endothelial nitric oxide synthase (eNOS) and eNOS phosphorylation (p-eNOS) of HAECs or aortas were analyzed by western blotting. Results Arotinolol relaxed vascular rings and the relaxations were attenuated by Nω-nitro-L-arginine methyl ester (L-NAME, NO synthase inhibitor) and the absence of endothelium. Furthermore, arotinolol-induced relaxations were attenuated by 4-aminopyridine (4-AP, Kv channels blocker). Arotinolol produced more nitric oxide compared to metoprolol and increased the expression of p-eNOS in HAECs. These results indicated that arotinolol-induced vasodilation involves endothelium-derived NO and Kv channels. The treatement with arotinolol in 8 weeks, but not metoprolol, markedly decreased CAP and PWV. Biochemistry assay and immunohistochemistry showed that aortic collagen depositions in the arotinolol groups were reduced compared with SHRs with metoprolol. Moreover, eNOS phosphorylation was significantly increased in aortinolol-treated SHR compared with SHRs with metoprolol. Conclusions Arotinolol improves arterial stiffness in SHR, which involved in increasing NO and decreasing collagen contents in large arteries.


Introduction
Beta-blockers (b-blockers) are widely prescribed for the treatment of a variety of cardiovascular pathologies including hypertension, heart failure, primary treatment of myocardial infarction, secondary prevention of ischemic cardiac events as well as other non-cardiovascular diseases. [1] The first generation bblockers, for example propranolol, are nonselective and block both b 1 -and b 2 -adrenoceptors. The second generation b-blockers, including metoprolol, are cardioselective and selectively block b1adrenoreceptors. There are discrepancies concerning the effect of metoprolol, a secondary generation b-blocker, on blood pressure and arterial stiffness. Kosch et al reported that metoprolol decreased central blood pressure and decreased aortic pulse wave velocity in hypertensive patients. [2] In contrast, there are studies showing that metoprolol does not have effect on central systolic and diastolic blood pressures, central pulse pressure, as well as left ventricular wall thickness. [3] The third generation vasodilating bblockers, such as nebivolol, are b 1 -adrenoreceptors specific and have additional ancillary properties including direct vasodilatory effect, and studies show that the vasodilatory effect of nebivolol involves endothelium-derived nitric oxide (NO). [4] Compared to classic b-antagonists, they promote a vasodilation through different mechanisms, which lead to a more favorable hemodynamic profile compared to non-vasodilating b-blockers. Studies show that the third generation of b-blockers have beneficial effect on blood vessels in experimental and human hypertension, [5,6] which are closely related to their vasodilatory effects.
Arotinolol is a nonselective a/b-adrenergic receptor blocker lack of local anesthetic, membrane-stabilizing or intrinsic sympathomimetic properties, which means that arotinolol is a pure a/badrenergic receptor blocker and vasodilating b-blocker as well. [7,8] Arotinolol obtains some advantages from these characteristics since the third-generation b-blocker, such as nebivolol, had no effect on a-adrenergic receptor. Clinically, this drug has been extensively used for the treatment of hypertension. However, it remains unknown whether it has beneficial effect on arterial stiffness and it has few researches on whether vasodilatory effects of arotinolol get involved in endothelial functions.
It has been well documented that endothelial function was impaired and arterial stiffness was significantly higher in spontaneously hypertensive rats (SHR) compared with normotensive Wista Kyoto (WKY) rats. [9] In the present study, we investigated the direct vasodilatory effect of arotinolol and the underlying mechanisms. Moreover, we also comprehensively investigated whether arotinolol had any effect on vascular stiffness and the corresponding changes in endothelial function, arterial stiffness and vascular wall composition in SHR, due to few reports in vessel protection by arotinolol.

Animals and experimental groups
Male SHRs and WKY rats, 16-week-old were purchased from the Shanghai Experimental Animal Center. SHRs were randomly divided into 3 groups, and treated with vehicle (SHR control, n = 12), metoprolol (200 mg?kg 21 ?day 21 , n = 12), or arotinolol (30 mg?kg 21 ?day 21 , n = 12) for eight weeks respectively. Agematched WKY (n = 12) was used as normotensive control. Meanwhile, another 24 16-week-old WKY rats were used in the measurement of isometric force to elucidate the vasodilating effects of arotinolol and underlying mechanisms. All rats were housed in a reversed 12:12-hour light-dark cycle in a temperature-controlled facility (2161uC) with free access to standard laboratory chow and tap water. SHRs in the metoprolol and arotinolol groups were treated with drugs in 0.5 ml water by oral gavage with a feeding needle twice every day for 8 weeks, while WKY group and another control group of SHRs received only 0.5 ml regular water by oral gavage for 8 weeks. Animals were anesthetized via a 60 mg/kg body weight intraperitoneal injection of pentobarbital sodium and then killed by overdose of pentobarbital sodium. The investigation conformed to National Institutes of Health Guidelines for the Care and Use of Laboratory Animals, and was approved by the ethics review board of animal experiments at Ruijin Hospital.
Vessel preparation and measurement of isometric force in rat vascular rings Rat thoracic aortas, renal arteries, and mesenteric arteries were dissected, cleared of fat and connective tissues, cut into 2-3 mm rings, and mounted on isometric force transducers (Danish Myo Technology Model 610 M, Denmark) in a 5-mL organ bath, and aerated with 95% O 2 and 5% CO 2 under an initial resting tension of 19.6 mN, 2.5 mN, and 1 mN, respectively. Force was recorded via a PowerLab/8sp data acquisition system (A.D. Instruments, Castle Hill, Australia). Acetylcholine-induced vasodilations over 80% were regarded as endothelium-intact vessels, whereas lack of relaxation by acetylcholine was taken as endothelium removal.
To evaluate the vasodilatory effect of arotinolol and metoprolol, cumulative concentration-response curves of arotinolol and metoprolol (10 28 -10 25 mol/L) were constructed in three different vascular rings pre-contracted with 1 mmol/L phenylephrine. To evaluate the underlying mechanisms of vasodilations by arotinolol, in some groups, aortic rings were incubated in advanced with N vnitro-L-arginine methyl ester (L-NAME, 10 24 mol/L), tetraethylammonium (TEA, 10 22 mol/L, a potassium channel inhibitor), glibenclamide (10 24 mol/L, an ATP-sensitive potassium channel inhibitor), or 4-aminopyridine (2.5610 23 mol/L, a voltage-gated potassium channel inhibitor) for 30 min before addition of phenylephrine,then cumulative concentration-response curves of arotinolol (10 28 -10 25 mol/L) were constructed. To evaluate the beneficial effect of metoprolol and arotinolol on endothelial function, the vasodilatory response to acetylcholine was evaluated in aortas from WKY, SHR control, and SHR rats treated with metoprolol and arotinolol after 8 weeks' treatment.
Measurement of tail systolic blood pressure, central arterial pressure and pulse wave velocity Tail systolic blood pressures (SBP) were measured at the beginning of experiment and at the end of week 8 in conscious rats by Softron tail-cuff BP-98A non-invasive sphygmomanometer (Softron Incorporated, Tokyo, Japan).
Central arterial pressure (CAP) and pulse wave velocity (PWV) were measured after 8 weeks' treatment as previously described. [10] Briefly, in order to determine the propagation timing, 2-Fr Mikro-tip catheter pressure transducers (Model SPR-407; Millar Instruments, Houston, TX) were implanted into the aortic arch, and the abdominal aortas proximal to the iliac bifurcation. Then two Millar catheters were connected to an amplifier (Powerlab ML118/D; A.D. Instruments) and a PowerLab/8sp data acquisition system at a sampling rate of 1000 Hz to generate the time difference between the diastolic phase centers of the proximal and distal waves. A total of five-nine cycles were averaged as one propagation time. Then, PWV was calculated by dividing the propagation distance between two catheter tips by the propagation time. The central arterial pressure, which represented the systolic pressure of aortic root, was measured by the proximal pressure transducer located in the aortic arch.
Intracellular NO and Ser 1177 phosphorylation of endothelial nitric oxide synthase (p-eNOS) assays in human aortic endothelial cells Human aortic endothelial cells (HAEC) were purchased from BioLeaf (Shanghai BioLeaf Biotech Co., Ltd, Shanghai, China) and cultured in the medium at 37uC under 5% CO 2 . [11] Cells were made quiescent by incubation in serum-free culture medium for 24 h before examination.
To clarify whether arotinolol cause Ser1177 phosphorylation of eNOS, another part of HEACs lysates, which were pretreated with 10 umol/l arotinolol for 2 5 15 30 60 minutes, were subjected to SDS-PAGE gel electrophoresis and electrophoretically transferred to a nitrocellulose membrane as described previously. The membranes were then exposed to primary antibodies of p-eNOS (1:500 dilution; Enzo Life Sciences, Switzerland) or anti-eNOS antibody (1:500 dilution; Enzo Life Sciences, Switzerland) overnight at 4uC. After incubation with the peroxidase-linked secondary antibody for 1 hour at room temperature, immunoreactive proteins were visualized by Western Lightning plus ECL (Perkin Elmer, Boston, MA, USA), and finally exposed to X-ray films for the detection of target proteins.

Western blotting in the aortas of rats
Phosphorylation levels of endothelial nitric oxide synthase (eNOS) protein were evaluated. [12] Briefly, soluble protein extracts (20 mg) from the homogenized aortas were loaded into 8% SDS-polyacrylamide gels and transferred to PVDF membranes. After blocking in non-fat milk, the membranes were exposed to polyclonal rabbit anti-Ser1177 phosphorylation eNOS antibody (1:500 dilution; Abcam Laboratories, UK) or polyclonal rabbit anti-eNOS antibody (1:500 dilution; Enzo Life Sciences, Switzerland) overnight at 4uC. After incubation with HRP-linked secondary antibodies, the membranes were visualized by enhanced chemiluminescence reagent, and finally exposed to x-ray films for the detection of target proteins.

Collagen content and immunohistochemistry
After recording body weights, the thoracic aortas were dissected from rats and subjected to hydrolysis in HCl for 16 h. Next, hydroxyproline content was evaluated by the chloramine T and paradimethylaminobenzaldehyde method. Collagen content was calculated as (hydroxyproline content) 67.46. [13] Immunohistochemistry of the aortic tissues was performed as described previously. [14] Sections were incubated with commercial rabbit polyclonal antibodies against collagen type I (Abcam Laboratories, UK) at 1/1000 dilution overnight at 4uC, and conjugated with horseradish peroxidase (HRP) antibody (1:500 dilution; Santa Cruz Biotechnology, Santa Cruz, CA) at room temperature. Subsequently, all fields were photographed using a Nikon E600 light microscope (Nikon, Tokyo, Japan) at 6200 magnification.
Arotinolol and metoprolol were provided by Sumitomo Pharma Co., Ltd.

Statistical analysis
Vasodilatory effects were expressed as percentage tension of phenylephrine-induced preconstriction. Results are expressed as the mean 6 SEM values. Comparisons were made by using Student's t-test or one way ANOVA analysis with Bonferroni test, if more than two groups were compared. Statistical significance was indicated by P,0.05.

Vasodilatory effects of arotinolol and metoprolol in rat aortas, renal arteries and mesenteric arteries
To determine vasodilatory effects of arotinolol and metoprolol, vasodilations of vascular rings were evaluated by isometric force recording. As shown in Fig. 1, arotinolol dose-dependently relaxed endothelium-intact vascular rings from rat aortas, renal arteries and mesenteric arteries; whereas metoprolol had few vasodilatory effects in the three various vascular rings. Notably compared with the aortas, the vasodilator response curves in the mesenteric arteries dropped more sharply, indicating that the vasodilatory effects of arotinolol were more obvious in mesenteric arteries than in aortas.

Arotinolol-induced vasodilation involves NO and K + channels
The vasodilatory mechanism of arotinolol was then investigated. In endothelium-denuded rings, arotinolol produced less relaxation ( Fig. 2A). Similarly, pretreatment of endothelium-intact rings with eNOS inhibitor L-NAME (100 mmol/L) effectively reduced arotinolol-induced relaxations (Fig. 2B). These results indicate that endothelium-derived NO was involved in arotinolol-induced vasodilation.
The effects of arotinolol on NO production and eNOS phosphorylation were then evaluated since it represented the most important regulatory mechanism in eNOS function. As shown in Figure 2C&D, NO production was significantly increased after HAECs were treated with arotinolol for 30 min. However, this phenomenon did not occur in metoprolol pretreatment. Furthermore, p-eNOS/T-eNOS ratio also went up obviously when HAECs being treated with arotinolol in 2 min. These results indicate that arotinolol-induced vasodilation involves eNOS phosphorylation.
The possible involvement of large conductance Ca 2+ -activated K+ (BK Ca ) channels, voltage-gated K+ (Kv) channels, and ATPsensitive K+ (K ATP ) channels in arotinolol-induced vasodilation was also investigated, since these three channels was closely related to NO-induced vasodilation. As shown in Fig. 3, arotinololinduced relaxations were inhibited by BK Ca channels inhibitor tetraethylammonium ions (TEA+, 1 mmol/L) and Kv channels inhibitor 4-aminopyridine (2.5610 23 mol/L), but K ATP inhibitorglibenclamide did not inhibit the vasodilation by arotinolol. These results indicate that arotinolol relaxed rat aortic rings partially through Kv channels.

Effects of arotinolol on tail SBP, CAP and PWV in SHRs
To evaluate the effects of chronic treatment of arotinolol and metoprolol on blood pressure, SHR were treated with arotinolol and metoprolol for 8 weeks and the changes of tail SBP and CAP were then detected.
As depicted in Fig. 4A, tail SBP of arotinolol and metoprolol groups at week 8 dropped markedly compared with those origins before treatment and those of untreated SHR at week 8, while no differences were observed in tail SBP between arotinolol and metoprolol groups at week 8. Nevertheless, CAP was significantly higher in SHR than that in WKY. Treatment with arotinolol decreased CAP, while metoprolol did not work with the similar drop in tail SBP between arotinolol and metoprolol group at week 8 (Fig. 4B).
Meanwhile, PWV was detected to evaluate the effects of arotinolol and metoprolol on stiffness of large artery after arterial pressures were measured above. The results showed PWV were sharply higher in SHR than that in WKY as well, indicating that stiffness of large arteries was increased in SHR. Moreover, PWV of arotinolol group obviously dropped, compared with those of metoprolol group at week 8 (Fig. 4C).
These results indicate that arotinolol, but not metoprolol, decreased large artery stiffness in SHR.

Arotinolol improved endothelial function and eNOS phsophorylation in thoracic aortas in SHRs
The effects of arotinolol and metoprolol on endothelial function were also investigated since endothelial functions are involved in arterial stiffness. Furthermore, endothelial dysfunction usually is characterized by impaired endothelium-dependent vasodilation to acetylcholine. Fig. 5 showed that the vasodilatory responses to acetylcholine were significantly decreased in SHRs compared with normotensive WKY rats, indicating that endothelial function was impaired in SHR. Arotinolol increased vasodilatory response to acetylcholine, while metoprolol had no effect. These results suggest that arotinolol, but not metoprolol, improved endothelial function in SHR.
Next, we examined whether arotinolol-induced vasodilation improvement involved eNOS phosphorylation in the rat aorta. As shown in Fig. 6, SHR showed a remarkable reduction in p-eNOS compared with the WKY rat as shown in the decreased p-eNOS/ T-eNOS ratio, indicating that endothelial function was impaired in SHR. After treatment with arotinolol, p-eNOS levels were significantly increased in the SHR-rats aortas compared with SHR

Arotinolol decreased collagen content in rat aortas
We then determined whether arotinolol and metoprolol had any effects on collagen I deposition in rat aortas, since it has been well established that collagen content played a critical role in arterial stiffness of large arteries. As shown in Fig. 7, the expression of collagen I was higher in SHR compared with age-matched WKY in immunohistochemistry. Further, biochemical assay showed that collagen content was higher in SHR compared with age-matched WKY. These results indicate the increased collagen deposition in thoracic aorta in SHR, which may be associated with increased arterial stiffness of thoracic aorta. Moreover, as shown in Fig. 7, arotinolol significantly decreased collagen content in both immunohistochemistry and collagen content assay, whereas metoprolol did not happen. These results indicate that arotinolol decreased abnormal collagen deposition in thoracic aortas, which may be related to its beneficial effect on vascular stiffness.

Discussion
The present study demonstrates that arotinolol-induced vasodilation involves eNOS phosphorylation, NO production and Kv channels. Arotinolol effectively decreased arterial stiffness in SHRs as indicated by decreased CAP and PWV, which was accompanied by the increased eNOS phosphorylation/NO production, and decreased collagen deposition in SHRs. Meanwhile, metoprolol takes no effects on vascular protection. Our investigation suggests that the beneficial effect of arotinolol on arterial stiffness may involve the direct influences in endothelial NO in SHRs.   In the present study, arotinolol induced obvious relaxant effects on the three kinds of arteries, while metoprolol did not. Arotinololinduced vasorelaxations were largely inhibited by the eNOS inhibitor L-NAME, and aroninolol significantly increased the eNOS phosphorylation, suggesting that endothelium-derived NO mediates arotinolol-induced relaxations. Furthermore, this study shows that arotinolol increased p-eNOS (Ser1177) level in cultured endothelial cells. These results indicate that arotinolol-induced vasodilation involves enhanced eNOS phosphorlation and increased NO production. This finding is novel since it demonstrates for the first time that arotinolol has direct vasodilatory effect via NO pathway. Moreover, we found that Kv channels were also involved in arotinolol-induced vasodilation, while voltage-gated potassium channels are reportedly involved in the pathogenesis of hypertension, and thus, can be useful as novel therapeutic targets. [15,16] Meanwhile as known to all, NO plays a key role in the aetiology and development of hypertension, and recent researches show that potassium channel dysfunction is one of the determinants to induce arterial hypertension. [17,18,19] This study suggests that arotinolol exactly acts on both targets. It can either enhance Ser1177 phosphorylation of eNOS to release NO in HAECs, or cause vasodilation by affecting potassium channel. Therefore, arotinolol likely has a significant effect on the treatment of hypertension by a novel pathway.
Some studies indicate that arterial stiffness has increasingly become an important prognostic index and potential therapeutic target in hypertension and PWV is a well-recognized classic index for assessing arterial stiffness. [20,21] In the present study, we showed that metoprolol had no effect on arterial stiffness since there was no significant difference in PWV and CAP in metoprolol-treated SHR compared with SHR control, although it caused the similar drop in tail SBP compared with arotinololtreated SHR, which is consistent with previous studies showing that metoprolol had no influence in arterial stiffness in human and animal hypertensive objects. Nevertheless, arotinolol significantly decreased arterial stiffness as shown in the decreased PWV and CAP in arotinolol-treated SHR compared with SHR control. These findings suggest that arotinolol plays a protective role against arterial stiffness.
It has been demonstrated that the composition and properties of vascular walls, such as collagen content, [22]endothelial dysfunction, [9]endothelial nitric oxide synthase (eNOS), [23]and NO levels, [24] all play important roles in the development of vascular stiffness and hypertension. Endothelial dysfunction, usually characterized by impaired endothelium-dependent vasodilation in response to acetylcholine, plays a key role in the development of hypertension and its complications, [9] [25] Furthermore, Wallace SM et al has described that aortic PWV was associated with endothelial dysfunction. [26] Consistent with previous studies, [9] the present study shows that endothelium-dependent vasodilation was significantly lower in SHR compared with age-matched WKY, which might be caused by the lower eNOS phosphorylation level in thoracic aorta in SHR as shown in the present study. Furthermore, the present study shows that arotinolol, but not metoprolol, significantly improved endothelial dysfunction as illustrated in increased vasodilation in response to acetylcholine stimulation. Consistent with the improved vasodilation, the present study also shows that arotinolol, but not metoprolol increased eNOS phosphorylation in SHR. These results indicate that arotinolol had a better effect on improvement of endotheliumdependent vasodilations than metoprolol.
Collagen is the key determinant of arterial stiffness. Barnes MJ et al observed that type I collagen is believed to impart arterial stiffness. [22] It has been shown that the total aortic collagen content or aortic fibrosis of the untreated SHR was considerably higher than that of the normotensive WKY rats. [27,28] Our investigation illustrated that arotinolol decreased aortic collagen contents in the SHR, indicating that arotinolol improved arterial stiffness at least partially by reducing aortic collagen deposition.
In vivo experiments of this study it indicates that arotinolol can reduce PWV and CAP which reflect arterial stiffness by improving endothelium-dependent vasodilations and collagen contents in the SHR aortas. Moreover, some studies present PWV as an independent risk factor of cardiovascular event, closely related to cardiovascular mortality. [29,30] Arterial stiffening per se is the cause of the poor response of blood pressure to chronic drug treatment. [31] In the present study, arotinolol is shown to obviously reduce arterial stiffness in SHRs. Therefore arotinolol would be beneficial to the treatment of hypertension and its complications as a de-stiffening antihypertensive medication. Further studies should be designed to verify vascular function for arotinolol in a clinical trial.
In conclusion, our results suggest that arotinolol has direct vascular protective effect by increasing eNOS phosphorylation and NO production both in vivo and in vitro. Furthermore, arotinolol decreased arterial stiffness and aortic collagen content in SHR. The effect on vessel protection, in combination with the aadrenergic receptor blocking effect, indicates that arotinolol is a superior b-blocker compared with other b-blockers. Few data is available at present with regards to NO production of other a/bblockers, thus additional research is required for further study.