Matrine attenuates high‐fat diet‐induced in vivo and ox‐LDL‐induced in vitro vascular injury by regulating the PKCα/eNOS and PI3K/Akt/eNOS pathways

Abstract Lipid metabolism disorders lead to vascular endothelial injury. Matrine is an alkaloid that has been used to improve obesity and diabetes and for the treatment of hepatitis B. However, its effect on lipid metabolism disorders and vascular injury is unclear. Here, we investigated the effect of matrine on high‐fat diet fed mice and oxidized low‐density lipoprotein (ox‐LDL)‐induced human umbilical vein endothelial cells (HUVECs). Computational virtual docking analyses, phosphoinositide 3‐kinase (PI3K) and protein kinase C‐α (PKCα) inhibitors were used to localize matrine in vascular injuries. The results showed that matrine‐treated mice were more resistant to abnormal lipid metabolism and inflammation than vehicle‐treated mice and exhibited significantly alleviated ox‐LDL‐stimulated dysfunction of HUVECs, restored diminished nitric oxide release, decreased reactive oxygen species generation and increased expression phosphorylation of AKT‐Ser473 and endothelial nitric oxide synthase (eNOS)‐Ser1177. Matrine not only up‐regulates eNOS‐Ser1177 but also down‐regulates eNOS‐Thr495, a PKCα‐controlled negative regulator of eNOS. Using computational virtual docking analyses and biochemical assays, matrine was also shown to influence eNOS/NO via PKCα inhibition. Moreover, the protective effects of matrine were significantly abolished by the simultaneous application of PKCα and the PI3K inhibitor. Matrine may thus be potentially employed as a novel therapeutic strategy against high‐fat diet‐induced vascular injury.


| INTRODUC TI ON
With the rapid development of the global economy and improvements in lifestyle, the overweight and obesity epidemic has become a major health challenge around the world. 1 High-fat diets play a primary role in obesity and can increase the risk for various diseases such as type 2 diabetes mellitus, cardiovascular diseases and other metabolic diseases. [2][3][4] Vascular endothelial dysfunction is closely related to cardiovascular diseases. Vascular endothelial cells play a key role in maintaining vascular permeability, transmitting vascular information and secreting vasoactive substances. Vascular endothelial cell injury plays an important role in the pathological process of atherosclerosis, hypertension, diabetes, cerebrovascular disease and other pathological processes. [5][6][7][8] Obesity can lead to disorders of lipid metabolism, which are characterized by an increase in the rates of triglyceride (TG), total cholesterol (TC) and low-density lipoprotein (LDL) and a reduction in high-density lipoproteins (HDL).
Lipid metabolism disorders can contribute to a reduction in the production and/or bioavailability of nitric oxide (NO), an important endothelium-dependent relaxant factor, thereby leading to vascular dysfunction. Endothelial nitric oxide synthase (eNOS)/NO plays a key role in the pathogenesis of abnormal lipid metabolism-induced cardiovascular complications. 9,10 Therefore, in recent years, the eNOS/NO pathway has been considered as a critical target for mediating endothelial function.
Matrine (Mat), a tetracyclo-quinolizidine alkaloid that is mainly derived from leguminosae such as Sophora flavescens and S subprostrata, has been shown to possess diverse pharmacological activities.
In Asia, S flavescens and S subprostrata are commonly used in meat soups and are thought to improve obesity and diabetes. 11 Mat has been widely used in the clinic for the treatment hepatitis B and also has exhibited a number of therapeutic effects on cardiovascular diseases. 12,13 Mat can protect cardiomyocytes from ischemia/reperfusion injury and also can improve isoproterenol-induced myocardial injury via regulating nitric oxide synthase. 14,15 However, the mechanisms of Mat in endothelial vascular injury due to lipid metabolism disorders have not been studied. Furthermore, details on the molecular mechanism underlying the cardiovascular protective effect of Mat are limited. Thus, the present study explored the possible molecular pathways of Mat in relation to its cardiovascular protective effects.

| Experimental animals
Male C57BL/6 mice (weight range: 16-18 g) were purchased from the animal centre of the Fourth Military University (Xi'an, China) and housed in a controlled environment (22 ± 2°C, 12 hours light/ dark cycle, free access to food and water). The mice were fasted for 12 hours before experimentation. All experiments were conducted between 8:00 am and 13:00 pm in a quiet room with temperature of 22-24°C. All procedures involving animals and their care were conducted in conformity with the NIH guidelines (NIH Pub. No. 85-23, revised 1996) and were approved by the Fourth Military University committee on animal care and use.

| Experimental design
After 2 weeks of adaptive rearing, the mice were randomly divided into five groups: a control group (CON, n = 10), high-fat diet group (HFD, n = 10) and a high-fat diet combined with Mat (0.5, 2.5, 10 mg/ kg) intervention group [HFD+Mat low (L), medium (M) and high (H) dose, respectively, n = 10]. The control group was fed with a normal chow diet and the HFD groups were given the high-fat diet for 12 weeks. Mat was added from 5 to 12 weeks at different concentrations once daily and at the same time. Body weights were monitored every 2 weeks. At the end of the experiment, all mice were fasted for 12 hours, then anaesthetized for blood collection and killed to collect the aorta. Blood samples were centrifuged at 1000 g for 10 minutes at 4°C to isolate the sera.

| Biochemical analyses
Triglyceride, TC, LDL and HDL levels were measured using an automatic biochemical analyzer (200FR; Toshiba, Japan). Pro-inflammatory cytokines (TNF-α, IL-6 and IL-10) and NO levels in the serum were assessed with commercial kits based on the colorimetric method, followed the manufacturer's recommendations and were performed in triplicate.

| Histological examination
Each aorta, which was obtained after decapitation of each mouse, was washed in saline and fixed in 10% formalin for routine haematoxylin and eosin (H&E) staining and histopathological examination.
The fixed tissues were processed routinely, embedded in paraffin wax, sectioned into 5-μm-thick sections in a rotary microtome and then stained with H&E dye. At least three different sections were examined per aorta sample. University. HUVECs were cultured in DMEM/high glucose medium containing 10% foetal bovine serum, 100 U/mL penicillin and 100 U/ mL streptomycin and were incubated at 37°C in 5% CO 2. When the cells had grown to about 80% confluency, the concentration of ox-LDL (100 μg/mL) was added to the medium and the cells were then cultured for another 24 hours. 16,17

| Assessment of cell viability and LDH levels
Human umbilical vein endothelial cells at the logarithmic growth phase were cultured in 96-well plates at a density of 5 × 10 4 cell/ mL and were incubated for 24 hours and then treated as follows:

| Measurement of ROS and NO levels
The cells were treated with 80 μmol/L Mat for 12 hours and then with 100 μg/mL ox-LDL for 24 hours, with or without pretreatment with LY-294002 and L-NAME. Then, intracellular ROS and NO levels were detected as previously described. 19,20 After the indicated treatment, the cells were washed three times with PBS and were then incubated with 10 mmol/L 21,71-dichlorofluorescin diacetate

| Measurement of PKC and eNOS activity
Cells from different groups were washed twice with PBS and scraped, followed by the addition of ATP to initiate a kinase reaction and were then incubated for 90 minutes at 30°C. Subsequently, a phosphor-specific antibody was added to each well, incubated for 60 minutes at room temperature, followed by the addition of an HRP-conjugated secondary for 30 minutes. The optical density of each well was measured at a wavelength of 450 nm and was used in the determination of PKC activity as previously described. 21 The reactions were performed individually in the presence of phospholipids (activated PKC reaction) and in the absence of phospholipids (control reaction). eNOS activity was measured using a NO synthase assay kit (Beyotime Biotech, China) according to the manufacturer's instructions. 22

| Molecular docking studies
The docking studies were conducted using AutoDock Vina software.
The input files were prepared in the graphic interface AutoDock

| Statistical analyses
The statistical analyses of the data to determine significant variations among groups was performed with SPSS 17.0 statistical software.
The data were expressed as the mean ± SD. Multiple comparisons between groups were performed with one-way ANOVA and pairwise comparisons were conducted by independent t test. In all cases, differences with a P < 0.05 were considered statistically significant.

| Mat improves lipid metabolism and inflammation in HFD mice
At baseline, no significant differences in body weight were observed among all groups (data not shown). The weight of the mice that were fed the HFD significantly increased (the weight of mice in the fourth and eighth weeks was measured; the results are presented as supplemental data). After 12 weeks of feeding, the body weight of the HFD mice significantly increased compared to those on a normal diet (P < 0.05). Mat administration reduced the HFD-induced body weight gain in a dose-dependent manner ( Figure 1A). In addition, the mice fed with HFD showed significantly increased serum TC, TG and LDL levels, whereas that of HDL significantly decreased compared with those on the normal diet (P < 0.05). The changes in TG, TC, LDL and HDL levels in the HFD-fed mice were effectively alleviated by Mat treatment (Figure 1B-E). Furthermore, the application of high-dose Mat resulted in a significant reduction in the TG content, reaching a similar level as that in the control mice.
The serum levels of inflammatory factors were also measured.

| Effects of Mat on ox-LDL-induced HUVEC injury
The cell viability of HUVECs that were exposed to 100 μg/mL ox-LDL for 24 hours significantly decreased compared to the control cells. Mat pretreatment effectively alleviated ox-LDL-induced cell injury in a dose-dependent manner, increasing cell viability as determined by MTT ( Figure 3A). On the contrary, when combined with LY-294002, this protective effect significantly decreased compared to the matrine-only group, but could still reflect the protective effect in ox-LDL-exposed HUVECs ( Figure 3B). However, when Mat was combined with the L-NAME (eNOS inhibitor), its protective effects were completely counteracted ( Figure 3C and D). The LDH release levels of HUVECs incubated with ox-LDL significantly increased compared to those of vehicle-treated HUVECs, whereas those subjected to Mat pretreatment showed a significant decrease ( Figure 3E). Interestingly, LY-294002 co-treatment did not completely abolish such protective effects of Mat. In contrast, L-NAME combined with Mat completely diminished the protective effects of Mat ( Figure 3F-H). F I G U R E 3 Effects of matrine on ox-LDL-induced HUVEC injury. Approximately 5 × 104 cells/mL were subjected to MTT and LDH assays for cell viability assessment. A and E, Quantification of the cell viability of HUVECs exposed to ox-LDL with or without matrine pretreatment, n = 3 from three independent experiments. B-D and F-H, Quantification for the cell viability of HUVECs exposed to ox-LDL along with matrine and/or LY-294002/L-NAME. n = 3 from three independent experiments; cell viability was normalized to the control group. Data are shown as the mean ± SD, *P < 0.05, **P < 0.01

| Effects of Mat on NO and ROS in ox-LDLexposed HUVECs
The ROS levels in ox-LDL-exposed HUVECs significantly increased and was counteracted by Mat pretreatment. Figure 5A shows that LY-294002 could not completely block the ROS reduction effect of Mat, but L-NAME co-treatment completely prevented this protective effect, whereas Mat combined with LY-294002 and L-NAME induced the same effect as that of L-NAME-only.
The NO in the ox-LDL-exposed HUVECs significantly dropped relative to the control HUVECs, whereas Mat pretreatment resulted in a significant increase. LY-294002 co-treatment did not completely prevent the protective effect of Mat, but when L-NAME was combined with Mat, the NO content did not increase in ox-LDL-exposed HUVECs and Mat combined with LY-294002 and L-NAME induced the same effect as that using L-NAME alone ( Figure 5B-D).

| Effects of Mat on eNOS in ox-LDLexposed HUVECs
Endothelial nitric oxide synthase activity in HUVECs treated with ox-LDL significantly decreased compared to the control HUVECs (P < 0.05). Mat pretreatment effectively increased eNOS activity in ox-LDL-exposed cells (P < 0.05). LY-294002 co-treatment did not completely abolish this increase in eNOS activity induced by Mat.
F I G U R E 5 Effects of matrine on ROS generation and nitric oxide (NO) production in ox-LDL-exposed HUVECs. A, Intracellular ROS levels were detected using the probe DCFH-DA; fluorescence intensities were measured by fluorescence microscopy; representative fluorescence images are shown (100×). NO levels were measured with the probe DAF-FM DA; fluorescence intensities were measured by fluorescence spectrophotometry and untreated cells were assigned a value of 100%. B, HUVECs exposed to ox-LDL with or without matrine pretreatment and LY-294002. C, HUVECs exposed to ox-LDL with or without matrine pretreatment and L-NAME (D) HUVECs exposed to ox-LDL with matrine, LY-294002 and L-NAME. Three independent samples were used. Data are expressed as the mean ± SD, *P < 0.05, **P < 0.01 However, we also observed that Mat combined with L-NAME did not increase the activity of eNOS, whereas Mat combined with LY-294002 and L-NAME had the same effect as that of L-NAME alone ( Figure 6A).
The PI3K/Akt/eNOS pathway plays an important role in the regulation of NO production and it has been shown that activated Akt phosphorylates eNOS at Ser1177, stimulates eNOS activity and induces NO release. 24

| Mat has significant affinity for PKCα, but not PKCθ
The PKC pathway also participates in the regulation of eNOS. F I G U R E 6 Effects of matrine on eNOS activity and PI3K/Akt/eNOS pathway-related protein expression in ox-LDL exposed HUVECs. A, Changes in eNOS activity in HUVECs exposed to ox-LDL with or without matrine pretreatment along with matrine and/or LY-294002/L-NAME as measured using an eNOS activity assay kit. B-E, Phosphorylation levels of Ser473Akt, Ser1177eNOS, Thr495eNOS, total Akt and the eNOS protein were measured by Western blotting. Representative images of three experiments; densitometric analysis of phosphorylated proteins was normalized to that of total proteins. Data are expressed as the mean ± SD of three independent experiments, *P < 0.05, **P < 0.01

| Effects of Mat on regulating PKC activity, PKCα, eNOS phosphorylation and NO content in ox-LDL-exposed HUVECs
Phosphorylation of Thr495eNOS is PKC-dependent; activation of PKC results in the phosphorylation of eNOS at Thr495. The PKC activity of HUVECs treated with ox-LDL significantly increased compared to the control HUVECs (P < 0.05), whereas Mat pretreatment effectively decreased PKC activity in ox-LDL-exposed HUVECs (P < 0.05) ( Figure 8A). The inhibition effect is similar to that of the PKCα inhibitor Go6976. The PKCα inhibition effect using Mat combined with Go6976 was not significantly different from that using and when combined with Go6976 has no significant effect on this regulatory effect, but was significantly decreased when combined with LY294002 ( Figure 8D and E).

| D ISCUSS I ON
Obesity is a chronic metabolic disorder that is associated with numerous diseases, including hyperlipidemia, diabetes mellitus, hypertension, atherosclerosis, cardiovascular disease and cancer. [26][27][28][29] Abnormal metabolism of blood lipids caused by an HFD is the main cause of cardiovascular disease and increased plasma LDL levels are regarded as a major risk factor. During oxidative stress, LDLs are oxidized at the vessel wall, thereby resulting in ox-LDL, which can cause endothelial dysfunction and is considered as an early and critical step in atherogenesis. 30 phosphorylates Akt at Ser473, resulting in kinase activation and stimulation of Akt enhances eNOS activity by phosphorylating Ser1177, which in turn increases the release of NO. 38,39 We investigated the effects of Mat in this regard. The data show that Mat can increase eNOS F I G U R E 8 Effects of matrine on PKC and eNOS activity, eNOS phosphorylation and nitric oxide (NO) content in ox-LDL exposed HUVECs. A, Changes in PKC activity in HUVECs exposed to ox-LDL with or without matrine pretreatment along with matrine and/or Go6976 as measured using PKC activity assay kits. B and C, Changes in eNOS activity and NO content in HUVECs exposed to ox-LDL with or without matrine pretreatment along with matrine and/or LY-294002/Go6976 measured using eNOS activity assay kits and the (DAF-FM DA) probe for the NO content. D and E, Phosphorylation levels of PKCα, Thr495eNOS, Ser1177eNOS and the total eNOS protein measured by Western blotting. Three independent samples were used. The data are expressed as the mean ± SD, *P < 0.05, **P < 0.01 phosphorylation of Ser1177 and Akt phosphorylation of Ser473 levels. PKCθ and PKCβ are also involved in the regulation of the eNOS pathway and PKCα is mainly involved in the regulation of eNOS495 phosphorylation in ox-LDL-treated endothelial cells. [41][42][43][44][45][46][47] To investigate the affinity of Mat to PKC isoforms, molecular docking studies were performed in this study. 48 Our results showed that Mat could combine with PKCα and its carbonyl group forms H bonds with the Lys368 amino groups, thereby enhancing intermolecular forces. At the same time, it also induces nonpolar interactions with VAL420, ALA480, VAL353, LYS368 and ALA366, but has low affinity for PKCθ compared to PKCα and a PKCθ inhibitor (sotrastaurin). Therefore, to further prove the regulatory effect of Mat on PKCα, the PKCα inhibitor Go6976 was utilized. Our results showed that the PKCα activity in HUVECs treated with ox-LDL significantly increased compared to the control, whereas Mat pretreatment effectively decreased the PKCα activity of ox-LDL-exposed HUVECs (P < 0.05).
This inhibitory effect is similar to that of the PKCα inhibitor Go6976. The application of Mat induced a significant increase in eNOS activity and the NO content in ox-LDL-induced HUVECs and also decreased the expression of phosphorylation of Ser657 PKCα and the phosphorylation of Thr495eNOS. The PKCα inhibitor had the same effect as that of Mat, but the regulatory effect of Go6976 further improved when combined with Mat. It is important to note that when matrine, LY-294002 and Go6976 were given at the same time, the regulatory effect did not significantly differ from that of the Go6976-only treatment group. Furthermore, our results indicated that Mat blocks the phosphorylation of Thr495eNOS, which is similar to the effect of Go6976. Mat could also increase the expression of Ser1177eNOS and its combination with Go6976 did not significant alter its regulatory effect but significantly decreased when combined with LY294002.
In conclusion, Mat effectively improved lipid metabolism, inflammation and the thickness of vascular walls in HFD mice. It also decreased eNOS-Thr497 phosphorylation and increased eNOS-S1177 phosphorylation, thereby increasing NO production, which is mediated by PI3K/Akt and PKCα. Our results shed light on the molecular mechanisms underlying the protective effect of Mat in HFD-induced vascular diseases.

CO N FLI C T S O F I NTE R E S T
The authors declare that they have no conflict of interest.