Article

Chronic Treatment With Qiliqiangxin Ameliorates Aortic Endothelial Cell Dysfunction in Diabetic Rats

Journal of Cardiovascular Pharmacology and Therapeutics 1-11 ª The Author(s) 2014 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/1074248414537705 cpt.sagepub.com

Fei Chen, PhD1, Jia-Le Wu, MS2, Guo-Sheng Fu, MD, PhD3, Yun Mou, MD, PhD4, and Shen-Jiang Hu, MD, PhD1

Abstract Qiliqiangxin (QL), a traditional Chinese medicine, has been shown to be beneficial for chronic heart failure. However, whether QL can also improve endothelial cell function in diabetic rats remains unknown. Here, we investigated the effect of QL treatment on endothelial dysfunction by comparing the effect of QL to that of benazepril (Ben) in diabetic Sprague-Dawley rats for 8 weeks. Cardiac function was evaluated by echocardiography and catheterization. Assays for acetylcholine-induced, endotheliumdependent relaxation (EDR), sodium nitroprusside-induced endothelium-independent relaxation, serum nitric oxide (NO), and nitric oxide synthase (NOS) as well as histological analyses were performed to assess endothelial function. Diabetic rats showed significantly inhibited cardiac function and EDR, decreased expression of serum NO and phosphorylation at Ser1177 on endothelial NOS (eNOS), and impaired endothelial integrity after 8 weeks. Chronic treatment for 8 weeks with either QL or Ben prevented the inhibition of cardiac function and EDR and the decrease in serum NO and eNOS phosphorylation caused by diabetes. Moreover, either QL or Ben suppressed inducible NOS (iNOS) protein levels as well as endothelial necrosis compared with the diabetic rats. Additionally, QL prevented the increase in angiotensin-converting enzyme 1 and angiotensin II receptor type 1 in diabetes. Thus, chronic administration of QL improved serum NO production, EDR, and endothelial integrity in diabetic rat aortas, possibly through balancing eNOS and iNOS activity and decreasing renin–angiotensin system expression. Keywords Qiliqiangxin, endothelium, diabetes, NOS, RAS

Introduction Chronic heart failure (CHF) is more common among diabetic patients (12%) compared with the general population (3.9%)1; additionally, heart failure has been shown to predict poorer outcomes for patients with diabetes.2 In some forms of heart failure, increased endothelial dysfunction plays a greater role, and imbalances between endothelial damage and repair or decreased nitric oxide (NO) production are involved in the development of heart failure and other diabetic complications, thus contributing to increased mortality rates.3-6 On the other hand, reduced endothelin and transforming growth factor b1 signaling activity improves survival, ameliorates endothelial dysfunction, and limits cardiac fibrosis in heart failure.7 Additionally, the endothelial dysfunction in large vessels and small resistance vessels or in different circulatory beds may be different in patients with CHF. It is believed that damaged endothelial function appeared early in the progression of CHF and might be associated with impaired perfusion, reduced exercise hyperemia, attenuated exercise capacity, deteriorated prognosis, and increased mortality risk in patients with CHF.8-10 Thus, mending endothelial impairment may be a potential therapeutic strategy for heart failure and other

diabetic complications. Unfortunately, to date, clinical trials of endothelin antagonists in heart failure have failed to produce satisfactory results.11,12 Traditional Chinese herbs have been shown to be safe and efficient modulators of CHF since ancient times. Qiliqiangxin

1

Institution of Cardiology, The First Affiliated Hospital, College of Medicine, Zhejiang University, Hangzhou, People’s Republic of China 2 Department of Cardiology, Xinhua Hospital, Hangzhou, People’s Republic of China 3 Institution of Cardiology, Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University, Hangzhou, People’s Republic of China 4 Department of Ultrasound, The Affiliated Hospital, College of Medicine, Zhejiang University, Hangzhou, People’s Republic of China Manuscript submitted: December 19, 2013; accepted: May 6, 2014. Corresponding Authors: Yun Mou, The First Affiliated Hospital, College of Medicine, Zhejiang University, 79 Qingchun Road, Hangzhou 310003, Zhejiang, People’s Republic of China. Email: [email protected] Shen-Jiang Hu, The First Affiliated Hospital, College of Medicine, Zhejiang University, 79 Qingchun Road, Hangzhou 310003, Zhejiang, People’s Republic of China. Email: [email protected]

Downloaded from cpt.sagepub.com at UNIV OF CO HEALTH SCIENCE CTR on December 15, 2014

2

Journal of Cardiovascular Pharmacology and Therapeutics

(QL) is extracted from a group of Chinese herbs and has been shown to be effective and safe in the treatment of CHF by preventing myocardial remodeling, regulating inflammation, blocking L-type Ca2þ channel activity, reducing Ca2þ overload, downregulating the cardiac chymase signaling pathway and chymase-mediated angitensin II production, and increasing cardiac contractility and kidney blood flow.13-16 However, whether QL can ameliorate endothelial dysfunction in patients with CHF remains unknown. Therefore, we compared the effects of QL on endothelial dysfunction between diabetic and normal rat aortas to explore whether QL confers significant beneficial effects to the endothelium. The effects of QL were compared with benazepril (Ben), a typical angiotensinconverting enzyme inhibitor (ACEI) that has shown protective effects on the endothelium in the treatment of cardiovascular disease.

Materials and Methods Reagent Streptozocin (STZ), acetylcholine (Ach), sodium nitroprusside (SNP), and phenylephrine (PE) were purchased from Sigma (St Louis, Missouri); a serum NO detection kit was obtained from Jianchen (Nanjing, China), a hydrogen peroxide (H2O2) detection kit was purchased from Beyotime (Haimen, China), the terminal deoxynucleotidyl transferase (TdT)-mediated 20 -deoxyuridine,50 -triphosphate nick end labeling (TUNEL) kit was purchased from Merck (Darmstadt, Germany), the nicotinamide adenine dinucleotide phosphate (NADPH) kit was obtained from AAT (Sunnyvale, California), and Ben was a kind gift from Novartis (Basel, Switzerland). The primary antiphosphoendothelial nitric oxide synthase (antiphospho-eNOS; ser1177), anti-eNOS, anti-inducible nitric oxide synthase (antiiNOS), anti-b-actin antibodies, and secondary antibodies were obtained from Cell Signaling (Denver, Colorado). The primary antiangiotensin-converting enzyme 1 (anti-ACE1) and antiangiotensin II receptor type 1 (anti-AGTR1) antibodies were obtained from Epitomics (Burlingame, California). Qiliqiangxin compounds, consisting of Radix Astragali, Aconite Root, Ginseng, Salvia Miltiorrhiza, Semen Lepidii Apetali, Cortex Periplocae Sepii Radicis, Rhizoma Alismatis, Carthamus Tinctorius, Polygonatum Odorati, Seasoned Orange Peel, and Rumulus Ginnamomi, were kindly provided by Yiling Pharmaceutical Corporation (Shijiazhuang, China). The herbal drugs were authenticated and standardized with marker compounds according to the Chinese Pharmacopoeia 2005. The drug powder was diluted with normal saline.

Animal Model Male Sprague-Dawley rats weighing 180 to 220 g were purchased from the Animal Centre of Zhejiang University. All procedures were approved by the ethics committee for the Use of Experimental Animals at Zhejiang University and conformed with the ‘‘Guidelines for the Care and Use of

Laboratory Animals’’ published by National Academy Press (National Institutes of Health Publication No. 85-23, revised 1996). After starvation for 12 hours, diabetes was induced by intraperitoneal injection of 70 mgkg1 STZ. Approximately 72 hours later, blood glucose levels were measured; rats were considered diabetic when the glucose levels were above 16.4 mmol/L. The same procedures were performed with saline in age-matched rats as the control.

Drug Intervention Rats were randomly divided into 6 groups: (1) the untreated diabetic (DM) rats group; (2) diabetic rats intragastrically administered with QL at 0.8 gkg1d1 (DM-QL); (3) diabetic rats intragastrically administered with Ben at 10 mgkg1d1 (DM-Ben); (4) normal control rats; (5) normal control rats intragastrically administered with QL at 0.8 gkg1d1(CT-QL); and (6) normal control rats intragastrically administered with Ben at 10 mgkg1d1(CT-Ben). Untreated diabetic rats and normal control rats were also intragastrically administered with equal amounts of saline each day. Although diabetes-induced endothelial and cardiac dysfunction could occur much earlier than 8 weeks, our prior experiments have confirmed that STZinduced diabetic rats for 8 weeks were a very mature model for evaluating the endothelial and cardiac dysfunction in our laboratory.17 Therefore, the treatments continued for 8 weeks. The clinical dosage of QL was 0.06 gkg1d1 so the putative dosage of QL was from 0.5 to 3 gkg1d1 in rats.18 The gastric dosage of QL from 0.6 mgkg1d1 to 4 gkg1d1 has been reported15,16,19; however, the most prevalent dosage of QL was from 0.25 to 2 gkg1d1. We first determined the dosage of QL using 3 different concentrations (0.25, 0.8, and 2 gkg1d1). Qiliqiangxin at the dosage of 0.25 gkg1d1did not result in a significant improvement in cardiac function whereas at the dosage of 2 gkg1d1 induced an increasing mortality and only at the dosage of 0.8 gkg1d1 hypothetical results in cardiac function and mortality appeared (data not shown); therefore, we chose 0.8 gkg1d1 as an adequate dosage in our study. Interestingly, another group used 4 gkg1d1 of QL for administration in 8-week-old spontaneously hypertensive rats (SHRs) for 12 weeks and found that QL improves cardiac function.13 The main difference was that the physical qualifications of diabetic rats were much poorer than that of SHRs so the tolerance of QL is much smaller in diabetic rats than in SHRs.

Echocardiography and Hemodynamics After being anesthetized, rats were subjected to transthoracic echocardiography to measure the left ventricular end-systolic internal diameter (LVESd), LV end-diastolic internal diameter (LVEDd), and ejection fraction (EF). The invasive hemodynamics were measured using catheterization. Briefly, a PE50 catheter (1 mm diameter; Jingdong, Inc., Beijing) was inserted into the left ventricle via the right carotid artery to record the heart rate (HR), LV systolic pressure (LVSP), LV enddiastolic pressure (LVEDP), and maximum pressure rising/

Downloaded from cpt.sagepub.com at UNIV OF CO HEALTH SCIENCE CTR on December 15, 2014

Chen et al

3

dropping velocity (+dp/dt) using MEDLAB software (Nanjing, China).

Bioassay of Vasoreactivity Sections of the thoracic aorta were rapidly removed from the rats and immersed into chilled Krebs solution (in mmol/L: NaCl, 118; KCl, 4.7; MgSO47H2O, 1.2; KH2PO4, 1.2; CaCl2, 2.5; NaHCO3, 25; and glucose, 11, bubbled with 95% O2 þ 5% CO2 [pH 7.4]). The perivascular tissue was carefully removed, and the aortic rings were cut into sections 3 to 5 mm in length and resuspended in Krebs solution. After a 60-minute equilibration period at 2.0 g of initial tension, cumulative PE (0.01-10 mmol/L) was generated, and the half-maximal effective concentration (EC50) was calculated to evaluate the vascular contraction capability. After equilibration, other aortic rings were precontracted with PE (1 mmol/L). When the rings reached maximal developed tensions, they were relaxed using cumulative Ach (0.001-10 mmol/L) or SNP (0.001-10 mmol/L), and the EC50s were calculated.20 The EC50 was calculated using GraphPad Prime 5 software (GraphPad Software, Inc., California).

Histological Analysis For the histological analyses, the freshly dissected thoracic aorta was fixed in 10% (v/v) formalin for 24 hours, embedded in paraffin, sliced into 4-mm thick sections, and stained with hematoxylin and eosin. Analyses focused on the presence of dying endothelial cells. The dying endothelium size area–internal elastic lamina area ratio was calculated using Imaging-Pro Plus 5 software (Media Cybernetics, Inc., Maryland).

TUNEL Detection Thoracic aorta sections were deparaffinized and hydrated, and then tissue slices were permeabilized using protease K, allowed to incubate with TdT enzyme for 90 minutes, and then colored using DAB for 30 seconds. The samples were imaged using a microscope.

Measurement of Serum NO Level Serum NO levels were quantified by the nitrite method. Serum was mixed with the reacting substrate, vibrated, allowed to incubate for 10 minutes, and then centrifuged (4500 rpm, 15 minutes). The supernatant was then incubated with the chromogenic reagent. After 15 minutes, the optical density was measured at an absorbance of 540 nm.

Measurement of H2O2 The frozen dissected thoracic aorta were lysed in ice for 30 minutes and centrifuged. Supernatant or H2O2 standard diluents (50 mL) was mixed with the reaction solution (100 mL). After 30 minutes, the optical density was measured at an absorbance of 560 nm.

Measurement of NADPH The frozen dissected thoracic aorta were lysed in ice for 30 minutes and centrifuged. Supernatant or NADPH standard diluents were mixed with the reaction solution. After 1 hour, the optical density was measured at an absorbance of 570 nm.

Western Blot Analysis Proteins extracted from rat aortas were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then transferred to polyvinylidene fluoride membranes (Millipore, Billerica, Massachusetts). The membranes were blocked with 5% nonfat dry milk, incubated with antibodies against eNOS, phosphorylated eNOS, iNOS, and b-actin overnight, and then incubated with horseradish peroxidase-conjugated secondary antibodies. Immunocomplexes were visualized using enhanced chemiluminescence reagent (Bio-Max, Israel).

Statistical Analysis All data are presented as the means + standard error of the means for at least 3 independent experiments. The data for vascular contraction and vasodilatation were analyzed using 2-way analysis of variance (ANOVA). One-way ANOVA was used to compare the significant differences of other data. Values of P < .05 were considered significant. Significant differences were determined using SPSS14.0 software.

Results Qiliqiangxin Affects Biometric and Echocardiographic Parameters in Diabetic Rats Qiliqiangxin was thought to ameliorate CHF in rats16 and reduce serum N-terminal pro-B-type natriuretic peptide and improve LV EF in patients.18 So, we first examined the effect of QL on heart function in rats. In control rats, neither QL nor Ben showed any significant effect on measured physical characteristics, including HR, LVSP, LVEDP, +dp/dt, and EF (Table 1). In contrast, the cardiac contractility was significantly decreased at 8 weeks after STZ injection, including decreased +dp/dt, LVSP, and increased LVEDP. Additionally, heart to body weight (HW/BW) and lung to body weight were similarly elevated in diabetic rats (Table 2). To further confirm the heart dysfunction, we examined the echocardiogram. In parallel, EF and fractional shortening (FS) were significantly decreased in diabetic rats. Also, diabetes resulted in significant increase in LVESd. Interestingly, no change was observed in LVEDd (Table 2 and Figure 1). The main reason for this deviation is that the heart size was smaller, but the HW/BW was higher in diabetic rats. All the measured congestion and systolic responses were significantly attenuated by treatment with QL or Ben (Table 2 and Figure 1). To assess whether QL directly ameliorates the diabetes, we then measured the levels of glucose in diabetic rats. Interestingly, QL or Ben did not change the levels of blood glucose

Downloaded from cpt.sagepub.com at UNIV OF CO HEALTH SCIENCE CTR on December 15, 2014

4

Journal of Cardiovascular Pharmacology and Therapeutics

Table 1. Biometric and Echocardiographic Parameters in Normal Control Rats.a Group

Control 398.73 + 4.25 + 5573.48 + 4025.84 + 147.65 + 85.43 +

HR, bpm LVEDP, mm Hg þdp/dt, mm Hg s1 dp/dt, mm Hg s1 LVSP, mm Hg EF, %

CT-QL 402.64 + 3.19 + 5791.64 + 3957.72 + 133.82 + 87.21 +

23.46 1.88 457.63 586.21 15.64 9.25

CT-Ben 411.54 + 3.51 + 5472.33 + 3987.35 + 139.64 + 83.59 +

16.75 2.47 685.27 682.1 10.38 7.36

11.99 1.86 586.3 622.71 17.28 12.46

Abbreviations; EF, ejection fraction; HR, heart rate; LVSP, left ventricular systolic pressure; +dp/dt, rate of rise/drop of left ventricular pressure; CT-QL, normal control rats intragastrically administered with QL at 0.8 gkg1d1; CT-Ben, normal control rats intragastrically administered with Ben at 10 mgkg1d1. a The effects of QL and benazepril on biometric and echocardiographic parameters in normal control rats. There was no change within different groups.

Table 2. Biometric and Echocardiographic Parameters in Diabetic Rats.a Group HR, bpm LVEDP, mm Hg HW/BW, mg g1 LW/BW, mg g1 þdp/dt, mm Hg s1 dp/dt, mm Hg s1 LVSP, mm Hg LVEDd, cm LVESd, cm FS, % EF, % BW Glucose, mmol/L

Sham 399.2 5.39 2.54 4.45 5012.6 3339 131.8 0.64 0.3 52.49 87.35 457.7 5.23

+ 19.13 + 2.24 + 0.17 + 0.23 + 594.78 + 391.11 + 9.78 + 0.04 + 0.06 + 7.38 + 4.56 + 59.38 + 0.39

DM 413.2 + 25.51 12.62 + 2.5b 4.14 + 0.66b 7.43 + 1.22b 3340.2 + 407.43b 1872.2 + 231.29b 94.6 + 9.1b 0.7 + 0.05 0.5 + 0.05b 28.27 + 6.01b 60.02 + 10.09b 183.83 + 27.64b 25.58 + 4.35b

DM-QL 403.6 + 7.74 + 3.55 + 5.71 + 3995.6 + 2702.8 + 119 + 0.64 + 0.39 + 38.76 + 74.5 + 257.49 + 27.82 +

13.45 1.55c 0.49c 0.89c 145.34c 312.25c 10.46c 0.03 0.05c 5.97c 6.88c 38.65c 9.37

DM-Ben 406.4 + 8.47 + 3.39 + 5.56 + 4112 + 2829.2 + 114.2 + 0.63 + 0.38 + 39.83 + 76.15 + 264.78 + 24.91 +

18.35 2.41c 0.47c 0.8c 390.4c 163.3c 13.14c 0.05 0.04c 3.21c 3.93c 42.2c 5.74

Abbreviations: HR, heart rate; LVEDP, left ventricular end-diastolic pressure; LVEDd, left ventricular end-diastolic dimension; LVESd, left ventricular end-systolic internal diameter; HW/BW, heart weight/body weight; LW/BW, lung weight/body weight; +dp/dt, rate of rise/drop of left ventricular pressure; LVSP, left ventricular systolic pressure; EF, ejection fraction; FS, fractional shortening; DM, diabetic; DM-QL, diabetic rats intragastrically administered with QL at 0.8 gkg1d1; DM-Ben, diabetic rats intragastrically administered with Ben at 10 mgkg1d1. a The effects of QL and benazepril on biometric and echocardiographic parameters in diabetic rats. After 8 weeks of STZ administration, changes in LV function in Sham, DM, DM-QL, and DM-Ben were observed. Qiliqiangxin and benazepril were shown to have a beneficial effect on LV function in diabetic rats. N ¼ 5 rats. b P < .05, compared to sham rats. c P < .05, compared to DM rats.

(Table 2). Furthermore, after 8 weeks of treatment with QL or Ben, the weight loss was significantly attenuated in diabetic rats, which would be related to improved heart function.20

Qiliqiangxin Improves Vasodilatation and Vascular Contraction in the Aortic Rings of Diabetic Rats Ach-induced endothelium-dependent relaxation (EDR) was significantly impaired in the aortic rings of diabetic rats. The maximal diastolic efficiency (Emax) fell to 46.75% + 7.92%, and the EC50 increased to 344.84 + 83.15 nmol/L (Figure 2A). In contrast, SNP-induced endotheliumindependent relaxation (EIR) was similar at every SNP concentration tested (Figure 2B). Compared with the rats in the diabetic group, rats in either the QL treatment or the Ben treatment group exhibited significantly improved endothelial function as shown by increased relaxation at any Ach concentration tested. The EC50 decreased to 125.5 + 11.34 nm or 137.74 + 9.63 nmol/L separately (Figure 2A). Additionally, treatment with QL or Ben could slightly, but not significantly,

improve Ach-induced EDR at certain Ach concentration tested in normal rats, such as at Ach 107 mol/L tested (EDR, P ¼ .08; EC50, P ¼ .13). Furthermore, neither QL nor Ben had any effect on SNP-induced EIR (Figure 2B). To completely evaluate the vascular function, we examined PE-induced vasoconstriction. As shown in Figure 2C, the Emax decreased from 130.34% + 11.58% in normal control rat aortic rings to 85.21% + 14.22% in diabetic rat aortic rings, and the EC50 increased from 68.32 + 2.45 nmol/L in normal rats to 145.24 + 7.85 nmol/L in diabetic rats. Treatment with QL or Ben significantly improved PE-induced vasoconstriction, and the EC50 fell to 93.12 + 3.86 or 82.9 + 4.11 nmol/L, respectively. Additionally, treatment with QL or Ben did not affect the PE-induced vasoconstriction in normal rats.

Qiliqiangxin Could Maintain Endothelial Integrity in Diabetic Rats To further confirm the protective effects of QL in the aortic rings of diabetic rats, we first investigated the effects of QL

Downloaded from cpt.sagepub.com at UNIV OF CO HEALTH SCIENCE CTR on December 15, 2014

Chen et al

5

Figure 1. Representative M-mode tracings of echocardiographic views of the LV from Sham, diabetic (DM) rats, diabetic rats intragastrically administered with QL at 0.8 gkg1d1 (DM-QL), and diabetic rats intragastrically administered with Ben at 10 mgkg1d1 (DM-Ben) groups.

on endothelial apoptosis. We observed a marked increase in TUNEL-positive endothelial cells in diabetic rats. Chronic treatment with QL or Ben resulted in a significant decrease in the number of TUNEL-positive cells (Figure 3A). Sloughing off of the endothelial cells is another important indicator of endothelial integrity. Therefore, we compared the beneficial effects of QL or Ben on the endothelial cells sloughing off. Because the lumen area was discrepant in different aortic ring, the endothelial necrosis size area to internal lamina area ratio (EN–IL ratio) was used to normalize the vessel circumference. A large area of endothelial cellular necrosis size was observed in the aortic endothelium of diabetic rats. However, the increased endothelial necrosis was ameliorated significantly in rats treated with either QL or Ben, the EN–IL ratio decreased from 21.51% + 3.73% to 9.58% + 2.05% and 8.49% + 1.11%, respectively (Figure 3B).

Chronic Treatment With QL Increases the Serum NO Levels in Diabetic Rats Griess detection revealed that serum NO was significantly decreased in diabetic rats after 8 weeks (197.24 + 10.2 vs 32.22 + 13.65 mmol/L). Chronic treatment with either QL or Ben induced a significant increase in serum NO levels compared with the untreated diabetic rats and reached to 79.52 + 8.55 and 92.49 + 16.24 mmol/L, respectively. Neither QL nor Ben significantly increased the serum NO concentrations in normal rats (Figure 4A).

Qiliqiangxin Regulates NOS Protein Expression in Diabetic Rats The expression of phospho-eNOS (ser1177) protein was significantly reduced in the aortic rings of diabetic rats. In contrast, these rats showed significantly upregulated expression of iNOS protein. Qiliqiangxin or Ben treatment stimulated eNOS phosphorylation and suppressed iNOS expression (Figure 5A and B). These data suggest that QL and Ben may provide protection against diabetes through NOS regulation.

Qiliqiangxin Regulates ACE1 and AGTR1 Protein Expression in Diabetic Rats Qiliqiangxin has been proven to alleviate heart failure via angiotensin-dependent manner.16 Therefore, we suspected that QL could also affect angiotensin pathway in diabetic rat aortic rings. As angiotensin II was the major detrimental reagent in renin–angiotensin system (RAS), we measured the protein expression of ACE1, which produces angiotensin II, and AGTR1, which integrates with angiotensin II. As shown in Figure 5A and C, diabetes mellitus caused overexpression of ACE1 and AGTR1, which was on average 1.71-fold or 1.85-fold higher, respectively. Chronic treatment with QL or Ben significantly suppressed the protein expression of ACE1 and AGTR1 in diabetic rat aortic rings.

Downloaded from cpt.sagepub.com at UNIV OF CO HEALTH SCIENCE CTR on December 15, 2014

6

Journal of Cardiovascular Pharmacology and Therapeutics

Figure 2. The effects of QL and benazepril on aortic bioassay response in diabetic rats or normal control rats. A, Acetylcholine-induced endothelium-dependent relaxation (EDR) and Ach-EC50 in rat aortic rings. Qiliqiangxin and benazepril significantly ameliorated diabetesinduced EDR impairment. B, Neither diabetes nor QL had any significant effect on sodium nitroprusside (SNP)-induced endotheliumindependent relation (EIR). C, Phenylephrine-induced vascular contraction and PE-EC50 in rat aortic rings. Qiliqiangxin and benazepril ameliorated PE-induced vascular contraction. *P < .05, compared to sham rats; #P < .05, compared to diabetic (DM) rats. N ¼ 5 rats. Ach indicates acetylcholine; EC50 indicates half-maximal effective concentration; PE, phenylephrine; QL, Qiliqiangxin.

Qiliqiangxin Decreases the Reactive Oxygen Species Activity in Diabetic Rats At present, it is believed that many RAS- or iNOS-initiated effects are currently accepted to be mediated by reactive oxygen species (ROS). Therefore, we first detected the levels of H2O2 in diabetic rat aorta. As shown in Figure 4B, diabetes caused a significant increase in H2O2 levels. Chronic treatment with QL or Ben induced a significant attenuation of H2O2 levels in diabetic rats. Additionally, activated NADPH oxidation increases ROS production in angiotensin-induced endothelial cells.17,21 Therefore, we detected the expression of NADPH in diabetic rats. As shown in Figure 4C, diabetes resulted in a significant decrease in NADPH expression. Chronic treatment with QL or Ben induced a significant increase in NADPH expression in diabetic rats. Additionally, neither QL nor Ben significantly change the expression of H2O2 or NADPH in normal rats.

Discussion Our study suggests that chronic administration of QL and Ben significantly improves serum NO production, EDR, and

endothelial integrity in the aortas of diabetic rats, possibly through a mechanism involving activation of eNOS and inhibition of iNOS. In addition, QL suppresses protein expression of ACE1 and AGTR1. These changes may abrogate the development of heart failure in diabetic rats. In diabetic models, we found that endothelial denudations were more common than in healthy controls. Chronic treatment with QL prevented the disruption of endothelial integrity. These data suggest that QL induces a significant regression of endothelial remodeling in diabetic rats. After 8 weeks of STZ administration, treatment with QL induced a 2-fold increase in serum NO levels and balanced the activity of eNOS and iNOS, resulting in significantly improved Ach-induced EDR. In addition, QL showed beneficial effects on endothelial integrity. Moreover, neither QL nor Ben showed a beneficial effect on SNP-induced EIR. These data suggest that QL effectively abolishes endothelial dysfunction during hyperglycemic overload through an underlying NOS mechanism. Additionally, chronic treatment with QL significantly improved cardiac dysfunction in diabetic rats. Furthermore, QL can not only reduce the expression of ACE1 and AGTR1 but also simulate the protective effects of Ben, an ACE1 inhibitor. Altogether,

Downloaded from cpt.sagepub.com at UNIV OF CO HEALTH SCIENCE CTR on December 15, 2014

Chen et al

7

Figure 3. The effects of QL and benazepril on the endothelial integrity in diabetic rats. A, Diabetic rats induced a significant TUNEL-positive endothelial cell number, and QL or benazepril reduced TUNEL-positive endothelial cell number. B, Representative hematoxylin and eosin (H&E) staining of the endothelium from Sham, DM rats, diabetic rats intragastrically administered with QL at 0.8 gkg1d1 (DM-QL), and diabetic rats intragastrically administered with Ben at 10 mgkg1d1 (DM-Ben) groups. Arrow A, TUNEL-positive cells; B1, B3, and B4, endothelial cellular necrosis area; and B2, endothelial cellular remaining area. The images of 200 are an amplification of the images of 100 within black boxes. *P < .05, compared to sham rats; #P < .05, compared to DM rats. N ¼ 5 rats. DM indicates diabetic; QL, Qiliqiangxin; TUNEL, terminal deoxynucleotidyl transferase-mediated 20 -deoxyuridine,50 -triphosphate nick end labeling.

Downloaded from cpt.sagepub.com at UNIV OF CO HEALTH SCIENCE CTR on December 15, 2014

8

Journal of Cardiovascular Pharmacology and Therapeutics

Figure 4. A, The effects of QL or benazepril on the levels of serum NO in diabetic rats or normal rats. Qiliqiangxin and benazepril significantly increased serum NO levels in diabetic rats. B, The effects of QL or benazepril on the levels of H2O2 in diabetic rats or normal rats. Qiliqiangxin and benazepril significantly induced a decrease in H2O2 level in diabetic rats. C, The effects of QL or benazepril on the expression of NADPH in diabetic rats or normal rats. Qiliqiangxin and benazepril significantly induced an increase in NADPH expression in diabetic rats. *P < .05, compared to sham rats; #P < .05, compared to diabetic (DM) rats. N ¼ 5 rats. H2O2 indicates hydrogen peroxide; NADPH, nicotinamide adenine dinucleotide phosphate; NO, nitric oxide; QL, Qiliqiangxin.

QL appears to regulate the balance between proinflammatory and anti-inflammatory cytokines as well as improve the regression of cardiac remodeling in cardiomyocytes13-16 and endothelial dysfunction during hyperglycemic overload, RAS seems to participate in this progression. Accumulating evidence indicates that the endothelium plays an essential role in the underlying processes of heart failure, indicating that modulation of endothelial function may be a potential target for CHF treatment.5,6 The healthy endothelium is a monolayer of cells, covering the inner walls of blood vessels, and regulates vascular function through the balanced production of vasodilators and vasoconstrictors, the balanced induction of antiproliferative and anti-inflammatory responses, and modulation of the fibrinolysis pathway in response to stimuli.22,23 Nitric oxide is considered to be the predominant mediator in these processes.24 In states of diabetes or CHF, the abnormal release of or response to NO has been observed in the vascular endothelium.20,25 Usually, NO is considered as the initial trigger of vascular dysfunction, and decreased serum NO level is a strong predictor of negative cardiovascular outcomes.26 Although eNOS and iNOS are all precursors of NO, there is a paradox between serum NO level and activities of eNOS and iNOS in diabetes. Nitric oxide is mainly derived from the eNOS isoform in the healthy endothelium, but in diabetes or CHF, excessive NO release triggers a shift to the iNOS isoform. Current evidence supports the facts that the NO produced by iNOS appears to mediate cytotoxic effects and inhibit myocardial contractility. In contrast, the NO produced by eNOS shows antioxidant capability and vascular relaxing effects.20,24,25,27,28 The expression of eNOS or iNOS is different in the various stages of STZ-induced diabetic rats or diabetic patients. Indeed, eNOS expression decreases in STZinduced diabetic rats at 4 weeks in parallel with an increase in the expression of iNOS.29 Our study20 and other previous studies24,25,27,28 reveal that the expression of eNOS has dual characteristics: increases at 2 weeks or decreases at 4 weeks

in STZ-induced diabetic rats. In contrast, the expression of iNOS is traditionally been thought to be boosting; therefore, it is a candidate for being aging, insulin resistance, mesangial fibrogenesis, and vascular inflammatory.30-33 Moreover, the levels of serum NO begin to decrease at 4 weeks.33 This mainly explains that iNOS accumulation is typically interpreted to form peroxynitrite (ONOO), a powerful oxidant that can easily induce substrate nitration. Protein S-nitrosylation inactivates eNOS and decreases NO release.33 Knockdown of iNOS is believed to inhibit the S-nitrosation of protein, alleviate vascular inflammatory, attenuate insulin resistance, and delay the aging and mesangial fibrogenesis. In addition, hyperglycemia can directly inhibit the synthesis of endogenous NO.25 Decreased production of endothelial NO is associated with increasing proinflammation, prothrombosis, and vascular contraction through enhanced activity of vascular constrictor factors such as angiotensin II and endothelin 1.34 In the present study, not only was the expression of iNOS sharply upregulated but also the activity of eNOS (phosphorylation of eNOS [ser1177]) was markedly downregulated in STZ-induced diabetic rats. Together, these changes may contribute to decreased serum levels of NO, leading to the impairment of endothelial function. Periodic acid-Schiff is a pluripotent toxic factor and an important contributor to cardiovascular remodeling and cardiovascular dysfunction. Renin–angiotensin system pathway initiates from the biogenesis of angiotensin I converted to angiotensin II under ACE1, and then angiotensin II couples with AGTR1 to generate adverse effects. Sulfhydryl ACEI promotes endothelial survival, whereas overexpression of Gaq, a subunit of AGTR1, recruits promyocardial hypertrophic effects.35 At present, it is believed that many RAS- or iNOSinitiated effects are currently accepted to be mediated by ROS. Superoxide dismutase has been shown to delay endothelial senescence; furthermore, overexpression of catalase in mice seems to reverse angiotensin II-induced cardiac hypertrophy

Downloaded from cpt.sagepub.com at UNIV OF CO HEALTH SCIENCE CTR on December 15, 2014

Chen et al

9

Figure 5. A. The effects of QL and benazepril on the expression of NOS isoforms, ACE1, and AGTR1 in diabetic rats as detected by Western blotting. b-Actin expression served as an internal control. Qiliqiangxin and benazepril effectively increased the phosphorylation of endothelial nitric oxide synthase (eNOS) and decreased the expression of inducible nitric oxide synthase (iNOS), ACE1, and AGTR1 in diabetic rats. B, Quantitative analysis of the NOS isoforms expression. C, Quantitative analysis of the ACE1 and AGTR1 expression. *P < .05, compared to sham rats; #P < .05, compared to diabetic (DM) rats. N ¼ 3 rats. ACE1 indicates angiotensin converting enzyme 1; AGTR1, angiotensin II receptor type 1; NOS, nitric oxide synthase; QL, Qiliqiangxin.

and heart failure; moreover, activated NADPH oxidation increases ROS production in angiotensin II-treated endothelial cells.36 Reactive oxygen species induces S-nitrosocysteine activation and iNOS expression and then leads to vascular inflammation and serum NO decrease. Scavenging of ROS or treatment with ACEI increases NO expression and reactivates eNOS.37 In the present study, increased expression of ACE1 and AGTR1 proteins was observed in DM rats, whereas inhibition of RAS using Ben increased serum NO level, rebalanced eNOS/iNOS expression, and improved vasodilatation in DM rats that indicated RAS was a key regulator of vascular dysfunction. Additionally, a decrease in the expression of NADPH was

observed in diabetic rats. The use of QL or Ben could ameliorate diabetes-mediated NADPH decrease. Taken together, these data reveal that the RAS–ROS–NOS play key roles in DM rats. Recently, the disruption of endothelial integrity was shown to be positively correlated with abnormal vasoconstriction in CHF.38 Lip et al39 revealed a 3-fold increase in circulating endothelial cells in the blood and a 6-fold decrease in flowmediated dilation in patients with CHF. The increased number of circulating endothelial cells suggests general endothelial denudation in CHF. A positive correlation between circulating endothelial cells and von Willebrand factor has also been observed in patients with CHF. During endothelial denudation,

Downloaded from cpt.sagepub.com at UNIV OF CO HEALTH SCIENCE CTR on December 15, 2014

10

Journal of Cardiovascular Pharmacology and Therapeutics

collagen is exposed, increasing the risk of thromboembolism in patients with CHF. In our study, a large area of endothelial denudation and more apoptotic endothelial cells were observed in the aortic rings of diabetic rats. These data revealed that disruption of endothelial integrity is also involved in the development of CHF in diabetic rats. Unlike Ach-induced EDR or SNP-induced EIR, the PEinduced contraction was controversial in diabetic aorta: decreases40 in vascular contraction or increases41 in vascular have been reported. Additionally, Xavier et al found a PE-induced vascular contraction change in a time-dependent manner in diabetic rats.42 Interestingly, in our previous20 and present studies, diabetes mellitus induced a decrease in vascular contraction to PE. In our previous study, we demonstrated that the progressive decrease in a-adrenoceptor number and the desensitization of a-adrenergic receptor-mediated vascular contraction in DM aortas may explain this change. Furthermore, antioxidant seems to improve the responses of PE.20 In the present study, QL or Ben could ameliorate PE-induced vascular contraction in DM rats; its properties as both an AGTR2 and an antioxidant may be useful. But whether there are desensitization and decrease in the number of AGTR2 is still unclear in diabetes mellitus. It needs further study. Qiliqiangxin capsules have already been reported to be safe and effective for the treatment of CHF. The main active constituents of QL are radix astragal and aconite root, which have been shown to have positive inotropic, chronotropic, and diuretic effects and positive effects on vasodilation, the antiinflammatory response, blockage of the Ca2þ channel, and downregulation of chymase-mediated angiotensin II production in the management of congestive heart failure.13-16 Although Ma et al43 reported that QL balances the production of NO and endothelin in patients with CHF, there is no evidence that QL produces beneficial effects on the endothelium in CHF. In the present study, we observed that QL and Ben significantly suppress the development of endothelial damage while regulating the activity of NOS isoforms; moreover, QL inhibited the expression of ACE1 and AGTR1 proteins. Perhaps, RAS and NOS were the target receptors of QL, which may be involved in the effects of QL on the vascular dysfunction. In conclusion, QL balances the NO pathway, inhibits the expression of ACE1 and AGTR1, increases serum NO levels, affects endothelial function, and maintains endothelial integrity in diabetic rats. Although further studies should focus on the exact molecular mechanisms underlying the effects of QL, our study suggests an application for QL in the prevention of endothelial damage in diabetic patients. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This work

was supported by the China National Natural Science Foundation, Nos: 81170242 and 81200191.

References 1. Nichols GA, Hillier TA, Erbey JR, Brown JB. Congestive heart failure in type 2 diabetes: prevalence, incidence, and risk factors. Diabetes Care. 2001;24(9):1614-1619. 2. Wells BJ, Roth R, Nowacki AS, et al. Prediction of morbidity and mortality in patients with type 2 diabetes. Peer J. 2013;11:e87. 3. Davignon J, Ganz P. Role of endothelial dysfunction in atherosclerosis. Circulation. 2004;109(23 suppl 1):III27-III32. 4. Marti CN, Gheorghiade M, Kalogeropoulos AP, Georgiopoulou VV, Quyyumi AA, Butler J. Endothelial dysfunction, arterial stiffness, and heart failure. J Am Coll Cardiol. 2012;60(16): 1455-1469. 5. Natali A, Ferrannini E. Endothelial dysfunction in type 2 diabetes. Diabetologia. 2012;55(6):1559-1563. 6. Van Craenenbroeck EM, Conraads VM. Mending injured endothelium in chronic heart failure: a new target for exercise training. Int J Cardiol. 2013;166(2):310-314. 7. Kapur NK, Wilson S, Yunis AA, et al. Reduced endoglin activity limits cardiac fibrosis and improves survival in heart failure. Circulation. 2012;125(22):2728-2738. 8. Tousoulis D, Charakida M, Stefanadis C. Inflammation and endothelial dysfunction as therapeutic targets in patients with heart failure. Int J Cardiol. 2005;100(3):347-353. 9. Katz SD, Hryniewicz K, Hriljac I, et al. Vascular endothelial dysfunction and mortality risk in patients with chronic heart failure. Circulation. 2005;111(3):310-314. 10. Chong AY, Blann AD, Patel J, Freestone B, Hughes E, Lip GY. Endothelial dysfunction and damage in congestive heart failure: relation of flow-mediated dilation to circulating endothelial cells, plasma indexes of endothelial damage, and brain natriuretic peptide. Circulation. 2004;110(13):1794-1798. 11. Kelland NF, Webb DJ. Clinical trials of endothelin antagonists in heart failure: a question of dose? Exp Biol Med (Maywood). 2006; 231(6):696-699. 12. Kirkby NS, Hadoke PW, Bagnall AJ, Webb DJ. The endothelin system as a therapeutic target in cardiovascular disease: great expectations or bleak house? Br J Pharmac. 2008; 153(6):1105-1119. 13. Liu W, Chen J, Xu T, et al. Qiliqiangxin improves cardiac function in spontaneously hypertensive rats through the inhibition of cardiac chymase. Am J Hypertens. 2012;25(2):250-260. 14. Wei Y, Liu X, Hou L, Che W, The E, Jhummon MV. Qiliqiangxin affects L type Ca(2þ) current in the normal and hypertrophied rat heart. Evid Based Complement Alternat Med. 2012;2012:131830. 15. Xiao H, Song Y, Li Y, Liao YH, Chen J. Qiliqiangxin regulates the balance between tumor necrosis factor-alpha and interleukin-10 and improves cardiac function in rats with myocardial infarction. Cell Immunol. 2009;260(1):51-55. 16. Zou Y, Lin L, Ye Y, et al. Qiliqiangxin inhibits the development of cardiac hypertrophy, remodeling, and dysfunction during 4 weeks of pressure overload in mice. J Cardiovasc Pharmacol. 2012;59(3):268-280.

Downloaded from cpt.sagepub.com at UNIV OF CO HEALTH SCIENCE CTR on December 15, 2014

Chen et al

11

17. Chen F, Chen B, Xiao FQ, et al. Autophagy protects against senescence and apoptosis via the RAS-mitochondria in highglucose-induced endothelial cells. Cell Physiol Biochem. 2014; 33(4):1058-1074. 18. Li X, Zhang J, Huang J, et al. Efficacy and Safety of Qili Qiangxin Capsules for Chronic Heart Failure Study Group. A multicenter, randomized, double-blind, parallel-group, placebo-controlled study of the effects of qili qiangxin capsules in patients with chronic heart failure. J Am Coll Cardiol. 2013;62(12):1065-1072. 19. Zhang J, Wei C, Wang H, et al. Protective effect of qiliqiangxin capsule on energy metabolism and myocardial mitochondria in pressure overload heart failure rats. Evid Based Complement Alternat Med. 2013;2013:378298. 20. Fu GS, Huang H, Chen F, et al. Carvedilol ameliorates endothelial dysfunction in streptozotocin-induced diabetic rats. Eur J Pharmacol. 2007;567(3):223-230. 21. Imanishi T, Tsujioka H, Akasaka T. Endothelial progenitor cells dysfunction and senescence: contribution to oxidative stress. Curr Cardiol Rev. 2008;4(4):275-286. 22. Furchgott RF, Zawadzki JV. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature. 1980;288(5789):373-376. 23. Libby P, Aikawa M, Jain MK. Vascular endothelium and atherosclerosis. Handb Exp Pharmacol. 2006;(176 pt 2):285-306. 24. Rubanyi GM, Romero JC, Vanhoutte PM. Flow-induced release of endothelium-derived relaxing factor. Am J Physiol. 1986; 250(6 pt 2):H1145-H1149. 25. Li H, Forstermann U. Nitric oxide in the pathogenesis of vascular disease. J Pathol. 2000;190(3):244-254. 26. Paneni F, Beckman JA, Creager MA, Cosentino F. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: part I. Eur Heart J. 2013;34(31):2436-2443. 27. Agnoletti L, Curello S, Bachetti T, et al. Serum from patients with severe heart failure downregulates eNOS and is proapoptotic: role of tumor necrosis factor-alpha. Circulation. 1999; 100(19):1983-1991. 28. Ikeda U, Shimada K. Nitric oxide and cardiac failure. Clin Cardiol. 1997;20(10):837-841. 29. Oyadomari S, Gotoh T, Aoyagi K, Araki E, Shichiri M, Mori M. Coinduction of endothelial nitric oxide synthase and arginine recycling enzymes in aorta of diabetic rats. Nitric Oxide. 2001; 5(3):252-260. 30. Gao L, Huang W, Li J. NOX1 abet mesangial fibrogenesis via iNOS induction in diabetes. Mol Cell Biochem. 2013;382(1-2): 185-191.

31. Jia J, Liu Y, Zhang X, Liu X, Qi J. Regulation of iNOS expression by NF-kB in human lens epithelial cells treated with high levels of glucose. Invest Ophthalmol Vis Sci. 2013;54(7):5070-5077. 32. Kim DH, Chung JH, Yoon JS, et al. Ginsenoside Rd inhibits the expressions of iNOS and COX-2 by suppressing NF-kB in LPSstimulated RAW264.7 cells and mouse liver. J Ginseng Res. 2013;37(1):54-63. 33. Ropelle ER, Pauli JR, Cintra DE, et al. Targeted disruption of inducible nitric oxide synthase protects against aging, S-nitrosation, and insulin resistance in muscle of male mice. Diabetes. 2013;62(2):466-470. 34. Hare JM, Nguyen GC, Massaro AF, et al. Exhaled nitric oxide: a marker of pulmonary hemodynamics in heart failure. J Am Coll Cardiol. 2002;40(6):1114-1119. 35. Porrello ER, Amore AD, Curl CL, et al. Angiotensin II type 2 receptor antagonizes angiotensin II type 1 receptor-mediated cardiomyocyte autophagy. Hypertension. 2009;53(6):1032-1040. 36. Enzo RP, Lea MDD. Cardiomyocyte autophagy is regulated by angiotensin II type 1 and type 2 receptors. Autophagy. 2009; 5(8):1215-1216. 37. Sandra D, Erika T, Marina Z, Lucia M. Sulfhydryl angiotensinconverting enzyme inhibitor promotes endothelial cell survival through nitric-oxide synthase, fibroblast growth factor-2, and telomerase cross-talk. J Pharmacol Exp Ther. 2010;332(3): 776-784. 38. Blum A. Heart failure—new insights. Isr Med Assoc J. 2009; 11(2):105-111. 39. Lip GY, Gibbs CR. Does heart failure confer a hypercoagulable state? Virchow’s triad revisited. J Am Coll Cardiol. 1999;33(5): 1424-1426. 40. Paskaloglu K, Sener G, Ayangolu-Dulger G. Melatonin treatment protects against diabetes-induced functional and biochemical changes in rat aorta and corpus cavernosum. Eur J Pharmacol. 2004;499(3):345-354. 41. Weber LP, MacLeod KM. Contractile responses of caudal arteries from diabetic rats to adrenergic nerve stimulation. J Vasc Res. 1994;31(1):25-32. 42. Xavier FE, Davel AP, Rossoni LV, et al. Time dependent hyperreactivity to phenylephrine in aorta from untreated diabetic rats: role of prostanoids and calcium mobilization. Vascul Pharmacol. 2003;40(1):67-76. 43. Ma FF, Lu FY, Feng SW, Li YX, Xiao WG. Influence of qiliqiangxin capsule on vascular endothelial function and heart function in patients with chronic heart failure. Mod Jour Integ Trad Chin West Med. 2008;17(17):260-2603.

Downloaded from cpt.sagepub.com at UNIV OF CO HEALTH SCIENCE CTR on December 15, 2014