Honokiol, an active compound of Magnolia officinalis, is involved in restoring normal baroreflex sensitivity in hypercholesterolemic rabbits

Abstract This study investigated the effects of methanol extract Magnolia officinalis (MEMO) on baroreceptor reflex sensitivity (BRS) in the hypercholesterolemic rabbits and the involved molecular mechanisms. Male New Zealand white rabbits were randomly divided into Control (normal diet), Cholesterol (0.5% w/w cholesterol diet), and Magnolia groups (0.5% w/w cholesterol diet plus 1% w/w MEMO). The animals were treated with the designated diet for 4 or 8 weeks. BRS in the control of heart rate was assessed by linear regression method. After 8 weeks of treatments, plasma total cholesterol (TC) was significantly elevated in the Cholesterol/Magnolia groups. The arterial blood pressure (aBP) was increased in the Cholesterol and Magnolia groups. The depression of BRS observed in the Cholesterol group was significantly ameliorated in the Magnolia group. After L‐NAME (Nω‐nitro‐Larginine methyl ester, 20 mg/kg, iv), the BRS of the Cholesterol group was significantly improved. Results from our in vitro study further indicated that honokiol, the principle component of MEMO, would protect human umbilical vein endothelial cells (HUVECs) from H2O2‐induced damages and inhibit H2O2‐induced vascular smooth muscles cells (VSMCs) proliferation, which was evident by the decreased expression of pFAK, and p‐Erk1/2. The results of the present study suggested that the improvement of BRS by MEMO in the hypercholesterolemic rabbits might be mediated by the antioxidant property of MEMO as indicated by the results from the L‐NAME and in vitro honokiol studies.

Mechanosensitive nerve endings of the baroreceptors are located in the carotid sinus and aortic arch, which are also the common sites of vascular lesions in atherosclerosis (Chapleau, Hajduczok, & Abboud, 1989). Baroreceptor reflex sensitivity (BRS) was found to be impaired in atherosclerotic subjects (Li, Mao, Abboud, & Chapleau, 1996).
Attenuation of BRS in carotid atherosclerosis has been reported in humans (Nasr, Pavy-Le Traon, & Larrue, 2005). Impairment of baroreflex function in hypercholesterolemia has been attributed to NO, reactive oxygen species (ROS), and other factors. It was found that reduction of vascular wall distensibility in atherosclerosis would attenuate BRS (Mattace-Raso et al., 2007). Depression of BRS in atherosclerosis might be related to the increased vascular stiffness and lesions of baroreceptors (Li et al., 1996). Oxygen free radicals were also shown to play an important role in baroreflex dysfunction in atherosclerosis (Li et al., 1996). Previous study demonstrated that exposure of isolated carotid sinus to superoxide dismutase (SOD) and catalase increased the BRS in hypercholesterolemic rabbits (Li et al., 1996). On the other hand, when carotid sinus treated with exogenous free radicals, the BRS was also found depressed. Therefore, free radicals apparently play a significant role in baroreceptor dysfunction (Hatcher, Gu, & Cheng, 2016).
The Magnolia officinalis L. (M. officinalis) is a known antioxidant and anti-inflammation herb which have been widely prescribed probably on a daily basis in Chinese medicine practice for centuries (Chuang et al., 2013). A 200 mg serving of magnolia bark would contain 95% of honokiol and magnolol. M. officinalis was used clinically for the treatment of stresses and muscle tensions (Weeks, 2009). Extracts of M. officinalis were also used for acute inflammation conditions (Wu et al., 2011). Anti-inflammatory property of M. officinalis was found in Lipopolysaccharide (LPS)-stimulated human gingival fibroblasts (HGF-1) and monocytes (U-937) (Walker et al., 2013). Free radical scavenging property of honokiol and magnolol, the active components of M.
Overproduction of NO and other ROS would account for the endothelial cells apoptosis and VSMCs proliferation in the development of atherosclerosis. Hyperlipidemia and elevated ROS levels might also account for the attenuation of BRS and the development of atherosclerosis. The present study was to investigate the effects of methanol extract of M. officinalis (MEMO) on the lipid profiles, aBP and BRS in high-fat atherogenic diet-induced hypercholesterolemic rabbits and the underlining molecular mechanism. Honokiol, an active compound of MEMO, was used in vitro to evaluate whether the protective effects of M. officinalis would be related to the protection of vascular endothelial cells and inhibition of vascular smooth muscle proliferation. Also, L-NAME was used to determine the role of NO in the depression of BRS in the hypercholesterolemia rabbits.
It was stored under 5°C and dried properly.

| HPLC analysis of methanol extract of M. officinalis
Our previously described method was slightly modified (Pan et al., 2010). Before HPLC analysis, MEMO was filtered with a 0.2 µm Millipore filter. A total volume of 10 μL was loaded to the column.
External standards of 100 μg/mL were prepared with HPLC grademethanol for calculation of the concentrations of examined samples.
Reverse phase HPLC was performed on a HITACHI HPLC system (Tokyo, Japan) equipped with HITACHI L-7100 pump, HITACHI L-7400 UV detector, and HITACHI L-7200 autosampler. LiChroCART 250-4 C18 HPLC cartridge (5 μm; Merck, Whitehouse Station, NJ, USA) was used for separations. Conditions for separation of examined samples were listed in Table 1. Based on the separation conditions of the standard, two candidate fractions were harvested (collection period: retention time ± 5 min). These fractions were further determined by electrospray ionization-ion trap mass spectrometry (HCT ultra PTM Discovery system; Bruker Daltonics, Billerica, MA, USA) for identification of the target compounds. Capillary voltage was 4,000 V, capillary exit offset 220 V, skimmer potential 60 V, and the trap drive value was 78. Conventional data of electrospray ionization-ion trap mass spectrometry were recorded using a scan range of 150-280 m/z. Nebulizer (nitrogen) pressure was 10 psi with dry gas flow of 5 L/min and dry temperature at 300°C.

| Animals
Male New Zealand White (NZW) rabbits (2.0-3.0 kg) were purchased from the National Laboratory Animal Center (Nankang, Taiwan). The animals were housed in the individual cage with free access to designated food and drinking water under the room temperature of 20 ± 1°C and humidity of 50%-70% for 8 weeks. Light/dark cycle was set to 12/12 hr. Blood sampling was taken prior to treatment (w = 0), at week four (w = 4) and at week eight (w = 8). The study's protocols and procedures were ethically reviewed and approved by a recognized ethical body (Institute Animal Care and Use Committee; IACUC) at China Medical University. After a one week adaptation period, animals were randomly divided into three groups. After 8 weeks of feeding with the designated diet, all rabbits were anesthetized with sodium pentobarbital (30 mg/kg, i.v.) via the marginal ear vein. Blood samples were also collected through the marginal ear vein.

| Treatment protocol
The animals were divided into three groups (n = 8) as described below. In animals with NO synthase blockade, L-NAME was intravenously administered 10-20 min before the phenylephrine injection.
Group 1. Control group was given 100 g/kg BW/day of normal rabbit chow (Fwusow, Taichung, Taiwan) for 4 and 8 weeks, respectively. Group 2. Cholesterol group was given 100 g/kg BW/day added with 10% corn oil and 0.5% cholesterol for 4 and 8 weeks, respectively. Group 3. Magnolia group was administered 100 g/kg BW/day of high-fat atherogenic diet (10% corn oil and 0.5% cholesterol) supplemented with MEMO (1% w/w) for 4 and 8 weeks, respectively.

| Surgical preparation
The rabbits were anesthetized with sodium pentobarbital (30 mg/kg, bolus i.v. plus continuous intravenous drops at about 6 mg kg −1 h −1 ). Absence of pedal reflexes indicated the adequate anesthesia. A cannula was introduced into the trachea through an incision at neck region to prevent respiratory obstruction.
Importantly, rectal temperature of animals were kept at 35-36℃ with a heating plate. A polyethylene catheter (PE-90, Becton Dickinson) filled with heparin solution (500 IU/mL, v/v) was inserted into the left femoral artery for recording of aBP. Through PE-50, polyethylene catheters were also inserted into femoral veins of both limbs for drugs administration. The arterial catheter was connected to a pressure transducer by a three-way stopcock.

| aBP and HR recording
While the experiment, the catheters were maintained patent by flushing with heparinized saline (500 IU/mL) occasionally. The pressure recording system was calibrated before the experiment with a mercury manometer. The aBP was recorded continuously by a pressure transducer (model P23-ID, Gould Instruments, Cleveland, OH, USA), amplified with a pressure processor amplifier (Gould 20-4615-52). Resting aBP was determined after a 20-min stabilization period. Analogue signals including the systolic, diastolic, mean arterial pressures, and HR were digitized and stored on the PowerLab Data Acquisition System (4SP, AD Instruments, Castle Hill, Australia) with a computerized analysis program.

| Measurement of BRS
BRS was measured using a linear regression method. The relationship between the peak increases in systolic blood pressures evoked by phenylephrine (0.25, 0.5, 0.75, and 1 µg/kg, i.v.) and the associated reflex responses in HR was analyzed by linear regression for the individual animal (Hsieh & Hong, 2008), and the slopes of the regression lines were used as the indices of BRS.

| Measurement of plasma lipid level
Blood samples collected in EDTA-coated glass tubes were centrifuged at 1;000 × g for 15 min at 4°C. The plasma was taken for the measurements of total cholesterol (TC) and triglyceride (TG) using the commercial analysis kits from Randox (TR212). All cells were incubated at 37C and humidified 5% CO 2 . Culture media were changed every 2 days.

| MTT cell viability assay
The protective effects of honokiol on the losses of viability induced by H 2 O 2 in HUVECs were determined by MTT assay. In 96-well plates, cells of 4 × 10 4 cells/well were seeded and cultured in medium overnight. Then, the cells were exposed to various concentra-

| Western blot analysis
A density of 5 × 10 6 HUVECs was plated in 10-cm dishes and cul-

| Statistical analyses
The data were presented as mean ± SEM. One-way ANOVA followed by Student-Newman-Keul test was employed for statistical analysis for the differences between groups. A p-value < 0.05 was considered statistically significant.

| HPLC and mass spectrometry of the marked compounds in MEMO
This analytic method was applied to determine honokiol and magnolol in the decoctions of MEMO. A typical chromatogram of MEMO was shown in Figure 1c. The retention times were 7.34 min for honokiol and 10.17 min for magnolol (Table 1). Identification of these compounds was confirmed by comparing the retention time with that of the standards (Figure 1a

| Effects of MEMO on lipid profiles of hypercholesterolemic rabbits
As shown in

| Effect of MEMO on aBP and HR
Rabbits of the Cholesterol group showed significant rise in aBP and HR compared with control rabbits (Table 3). Compared with the Cholesterol group, the rises in aBP, but not the HR, were prevented in the Magnolia group (Table 3).

| Effects of MEMO on the arterial baroreceptor mediated blood pressure regulatory mechanism
BRS was assessed according to the ratios of reflex bradycardic responses to the rises in arterial pressures induced by various F I G U R E 1 HPLC and mass spectrometry of the marked compounds in MEMO. The chromatograph indicated HPLC separation of the standard compounds (100 μg/ml) and MEMO (5 mg/ml), respectively (a, b, and c). Total volume of 10 μl was loaded into HPLC column to measure the relative content of magnolol, honokiol in MEMO according to the concentration of standard compounds. The examination conditions and monitoring wavelength of HPLC analysis were described in Table 1

| NO synthase blockade (L-NAME) restores the BRS in hypercholesterolemic rabbits
To understand the role of NO on the BRS in hypercholesterolemic rabbits, L-NAME was administered 20 min before the phenylephrine injections. BRS was attenuated in Cholesterol group (Figure 2).
However, L-NAME treatment prevented the attenuation of BRS in Cholesterol group (Figure 2). Moreover, L-NAME further restored the BRS in Magnolia group (Figure 2).

| Honokiol inhibits H 2 O 2 -induced proliferation in VSMCs
To determine the inhibitory effect of honokiol on VSMCs, the cells were incubated with various concentrations (0, 1.25, 2.5, 5, and 10 μM) of honokiol plus 200 μM H 2 O 2 for 24 hr. Cell viability was measured by MTT assay. Our results showed that honokiol at concentrations of 5 and 10 μM significantly inhibited the VSMCs proliferation ( Figure 4).

| Expression of pFAK and pErk1/2 effected by honokiol treatment in VSMCs
To investigate how honokiol affected VSMCs proliferation, pFAK and pErk1/2 were measured by Western blotting assay. Our results showed that treatment with honokiol (1.25, 2.5, 5, and 10 μM) dosedependently reduced the phosphorylation of pFAK and pErk1/2 ( Figure 5). Our data suggested that honokiol might play a significant role in regulation of VSMCs proliferation.

| D ISCUSS I ON
In the present study, we demonstrated that MEMO improved BRS (Table 3) and ameliorated the elevation of TC (Table 2)  At week 4, it was shown that aBP and HR did not increase in cholesterol diet-feeding rabbits, which was different from previous experiments (Z. Li et al., 1996;Wilfert et al., 2000). The possible reasons might be related to the use of oil, the proportion of different cholesterol feeding time. At week 8, Cholesterol group showed a higher TC, increased aBP, and depressed BRS. When diet was supplemented with MEMO, the Magnolia group had a lower TC compared with Cholesterol group (Table 2), but the aBP was still higher than that in the Control group (Table 3). It was consistent with previous study that administration of magnolol in rabbits did not improve the hypertension status (Huang, Hong, Tsai, & Lai, 2000). However, compared with the Cholesterol group, the BRS of the Magnolia group was improved (Table 3). As previously mentioned, the protective effects of MEMO might be related to its anti-inflammation and antioxidant properties and the possible involvement of ROS, particularly, NO.

TA B L E 2
Effects of on the total cholesterol of NZW rabbits fed with a high-fat diet for 8 weeks
Values are presented as mean ± SE. The Cholesterol group was fed with normal rabbit chow plus 10% (w/w) corn oil and 0.5% (w/w) cholesterol. The Magnolia group was fed with the same diet as the Cholesterol group, but plus 1% w/w MEMO.  Note: All values are expressed as means ± SE. The Control group was fed with normal rabbit chow. The Cholesterol group was fed with normal rabbit chow plus 10% (w/w) corn oil and 0.5% (w/w) cholesterol. The Magnolia group was fed with the same diet as the Cholesterol group, but plus l% (w/w) M. officinalis methanol extract (MEMO, see text for detail). * p < .05, compared with the Control group of the same week. # p < .05, compared with the Cholesterol group of the same week.
Accumulation of oxygen-derived free radicals would inhibit BRS during the generation of atherosclerotic arteriosclerosis (Li et al., 1996). In addition, studies have shown that chronic inflammation, the excessive accumulation of foam cells, and proliferation of vascular smooth muscle would increase vascular stiffness (Mozos et al., 2017) which would in turn change the sensitivity of baroreceptor reflex (Lee et al., 2011). NO plays an important role in the relaxation of vascular smooth muscle. Inhibition of NO would result in vasoconstriction and elevated aBP. It has been suggested that long-term treatment with NOS inhibitor, L-NAME, would induce hypertension, sympathetic hyperactivity, and BRS attenuation (Souza, Ballejo, Salgado, Da Silva, & Salgado, 2001). However, our data showed that acute administration with L-NAME to hypercholesterolemic rabbits significantly restored their BRS (Figure 2). It was consistent with the findings that in the sober rabbit, acute intravenous administration of NO inhibitors also increased BRS (Liu, Murakami, & Zucker, 1996). In the present study, changes in BRS in hypercholesterolemia rabbits might be associated with NO. It was demonstrated that the antiarrhythmic effects of magnolol (an active compound in MEMO) could be inhibited by the L-NAME. It was also suggested that the action of MEMO on BRS could be related to the NO signaling pathways (Hong, Huang, & Tsai, 1996).
Oxidative stress was thought to play an important role in the damages of endothelial cells and to be one of the etiology in atherosclerosis. ROS could induce inflammation, cell death, platelet aggregation, and VSMCs proliferation (Lee et al., 2011). These events would lead to cardiovascular lesions. It was shown that ROS-induced injuries of endothelial cells involved in the early development of atherosclerosis (Y. Li, Liu, & Zhan, 2000). Excess ROS could cause oxidative stress and damages to plasma membranes, important cellular proteins and nucleic acid (Young & Woodside, 2001 (1-10 mM) increased intracellular Ca 2+ and stimulated the synthesis of L-citrulline from L-arginine in bovine aortic endothelial cells (Shimizu, Ishii, Yamamoto, & Momose, 1997). Therefore, we also used H 2 O 2 to investigate the effects of honokiol on the oxidative injuries in endothelial cells (HUVECs). Several studies have evaluated the antioxidant activities of honokiol. It was shown to prevent the cerebral ischemia/reperfusion injuries by modulating the enzyme systems related to ROS production and metabolism in neutrophils (Liou, Shen, Chen, Tsao, & Tsai, 2003). It also ameliorated the expression of eNOS which was depressed by oxLDL and reduced the oxLDL-induced adhesion molecules and the adherence of THP-1 cells to HUVECs (Ou, Chou, Lin, Yang, & Sheu, 2006  HUVECs injury would provoke migration of VSMCs. Abnormal proliferation of VSMCs played a critical role in the development of hypertension and atherosclerosis (Ross, 1999). In the present study, we determined the effects of honokiol on the oxidative injury of vascular smooth muscle cells induced by H 2 O 2 and examined the expression of pFAK and pERK1/2 proteins in injured cells (Newby & George, 1993), which were thought to be related to the cell proliferation (Guyton, Gorospe, Kensler, & Holbrook, 1996;Zhao & Guan, 2011). Our results demonstrated that honokiol could inhibit the H 2 O 2 -induced VSMCs proliferation.

| CON CLUS ION
We have successfully induced hypercholesterolemia and atherosclerosis in NZW rabbits [40]. Rabbits treated with high cholesterol diet showed significant increases in aBP and HR and significant decreases in BRS. Treatment with MEMO in addition to the cholesterol feeding prevented the rise in aBP, and HR and significant restored of BRS.
After NO inhibition with L-NAME, BRS was increased in Cholesterol and Magnolia groups. These effects might be attributed to NO.
Effects of honokiol were found mediated by inhibiting ERK and FAK signaling pathways, which would subsequently prevent endothelial cells from ROS and/or NO oxidative injuries and suppressed vascular smooth muscle cells proliferation. Honokiol could be a useful compound, especially in the management of hypercholesterolemia and atherosclerosis. Hospital, Changhua, Taiwan (RD107042). We also appreciate the English-editing assistance from Stephen Hsu. We appreciate Wen-Hsin Lin for her technical assistance.

CO N FLI C T O F I NTE R E S T
The authors declare that they do not have any conflict of interest.

E TH I C A L S TATEM ENT
This study does not involve any human testing. The animal study's protocols and procedures were ethically reviewed and approved by a