Ginkgo biloba extract inhibits oxidized low-density lipoprotein (oxLDL)-induced matrix metalloproteinase activation by the modulation of the lectin-like oxLDL receptor 1-regulated signaling pathway in human umbilical vein endothelial cells

Ginkgo biloba extract inhibits oxidized low-density lipoprotein (oxLDL)-induced matrix metalloproteinase activation by the modulation of the lectin-like oxLDL receptor 1-regulated signaling pathway in human umbilical vein endothelial cells Kun-Ling Tsai, PhD,a,b Yuh-Lih Chang, PhD,c Po-Hsun Huang, PhD,b,d Yung-Hsin Cheng, PhD,c Ding-Hao Liu, MD,e Hsiao-Yun Chen, MS,a and Chung-Lan Kao, MD, PhD,b,e,f,g Tainan and Taipei, Taiwan Background: The overexpression of matrix metalloproteinases (MMPs) induced by oxidized low-density lipoprotein (oxLDL) has been found in atherosclerotic lesions. Previous reports have identified that oxLDL, via the upregulation of lectin-like ox-LDL receptor 1 (LOX-1), modulates the expression of MMPs in endothelial cells. Ginkgo biloba extract (GbE), from Ginkgo biloba leaves, has often been considered as a therapeutic compound for cardiovascular and neurologic diseases. However, further investigation is needed to ascertain the probable molecular mechanisms underlying the antiatherogenic effects of GbE. The aim of this study was to investigate the effects of GbE on oxLDL-activated MMPs of human endothelial cells and to test the involvement of LOX-1 and protein kinase C (PKC)-a, extracellular signalregulated kinase (ERK), and peroxisome proliferator-activated receptor-g (PPAR-g). Methods: Human umbilical vein endothelial cells were stimulated with oxLDL, with or without GbE treatment. LOX-1 signaling and MMPs expression were tested by Western blotting or activity assay. Further, protein expression levels of PKC-a, ERK, nuclear factor-kB, and PPAR-g were investigated by Western blotting. Results: GbE inhibited the oxLDL-caused upregulation of MMP-1, MMP-2, and MMP-3. Pretreating with GbE reduced oxLDL-activated LOX-1 expression. Furthermore, pharmacologic inhibitors of free radicals, CaDD, PKC, and GbE, inhibited the oxLDL-induced ERK and nuclear factor-kB activation. Lastly, GbE ameliorated the oxLDL-inhibited PPAR-g function. Conclusions: Data obtained in this study indicate that GbE actives its protective effects by regulating the LOX-1-mediated PKC-a/ERK/PPAR-g/MMP pathway, resulting in the suppression of reactive oxygen species formation and, ultimately, the reduction of MMPs expression in endothelial cells treated with oxLDL. (J Vasc Surg 2014;-:1-12.) Clinical Relevance: Therapies that inhibit matrix metalloproteinase (MMP) are thought to be new clinical interventions because MMPs have been found to function as participants in atherosclerosis. We confirmed that Ginkgo biloba extract elicits antiatherogenic effects by modulating lectin-like oxidized low-density lipoprotein receptor-1-mediated pathways, resulting in the suppression of reactive oxygen species generation and, ultimately, the inhibition of MMPs in human umbilical vein endothelial cells exposed to oxidized low-density lipoprotein. Our findings suggest that Ginkgo biloba extract is a potential preventive agent against the development of cardiovascular diseases.

Increased levels of oxidized low-density lipoprotein (oxLDL) are thought to be one of the main risk factors for cardiovascular disease and atherosclerosis. OxLDL induces the secretory activities of human endothelial cells. These activities, in turn, cause the endothelial cells to become dysfunctional.1 Previous reports suggested that lectin-like oxLDL receptor 1 (LOX-1), a novel lectin-like receptor of

oxLDL, is expressed primarily in human endothelial cells. LOX-1 also promotes the uptake of oxLDL and regulates some of the biological abilities of oxLDL. The expression of LOX-1 is activated by angiotensin II, oxidative stress, proinflammatory cytokines, and shear stress.2 Matrix metalloproteinases (MMPs) degrade collagen, elastin, and other matrix components of the extracellular

From the Institute and Department of Physical Therapy, National Cheng Kung University, Tainana; the Institute of Clinical Medicine,b Institute of Pharmacology,c School of Medicine,f and Institute of Physical Therapy & Assistive Technology,g National Yang-Ming University, Taipei; and the Division of Cardiology, Department of Medicine,d and Department of Physical Medicine and Rehabilitation,e Taipei Veterans General Hospital, Taipei. This study was supported by the National Science Council Taiwan (NSC-1022314-B-075-004-MY3), Taipei Veterans General Hospital (V102C-021 and DV102-10), Yen-Tjing-Ling Medical Foundation (CI-99/100/101/102), and National Health Research Institutes (NHRI-EX102-10258SI). Author conflict of interest: none.

Additional material for this article may be found online at www.jvascsurg.org. Reprint requests: Chung-Lan Kao, MD, PhD, Department of Physical Medicine and Rehabilitation, Taipei Veterans General Hospital, No. 201, Sec. 2, Shih-Pai Rd, Taipei 11217, Taiwan (e-mail: clkao@ vghtpe.gov.tw). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2014 by the Society for Vascular Surgery. http://dx.doi.org/10.1016/j.jvs.2014.05.098

1

JOURNAL OF VASCULAR SURGERY --- 2014

2 Tsai et al

matrix. Moreover, cardiovascular studies have shown that MMPs activate inflammatory processes, thereby rupturing the atherosclerotic plaque.3 Moreover, MMP-1, MMP-2, and MMP-3 were the MMPs most consistently linked with increased cardiovascular risk in humans.4 LOX-1 has been reported as a principal receptor for oxLDL and has been detected in atherosclerotic plaques.5 Previous reports have demonstrated that oxLDL, LOX-1, and MMPs are colocalized in atherosclerotic plaques.6 The interactions of oxLDL, LOX-1, and MMPs may facilitate matrix breakdown in atherosclerotic plaques and, consequently, predispose these plaques to disruption by oxLDL-mediated LOX-1 upregulation in human endothelial cells. The protein kinase C (PKC)-signaling pathway, linked with extracellular matrix remodeling in atherosclerotic lesions, is an important mediator for intracellular functions. PKC-a has been identified to necessitate nuclear factor (NF)-kB upregulation in the MMPs activation of epithelial cells.7 In addition, PKC promotes MMPs upregulation by oxLDL and is induced by extracellular signal-regulated kinase (ERK), c-Jun NH2-terminal kinases, and P38 mitogen-activated protein kinase (MAPK).8 Peroxisome proliferator-activated receptors (PPARs) are nuclear receptors known to control cellular functions, which include proliferation, differentiation, and lipid as well as glucose metabolism. PPARs modulate gene expression and activation by binding to the retinoid X receptor as a heterodimeric partner for specific DNA sequence elements.9 PPAR-g is expressed in human endothelial cells and is suppressed by the MAPK-derived pathway.10 Martin-Nizard et al11 determined that PPAR-g protects against oxLDL-mediated endothelial-1 secretion, revealing that PPAR-g protects against endothelial cell dysfunction through regulating metabolic disorders and modification of the proinflammatory response. Ginkgo biloba extract (GbE) is obtained from the leaves of the Ginkgo biloba tree and includes 24% of Ginkgo flavone glycoside and 6% of terpene lactones. GbE has been widely regarded to be effective in the treatment of cognitive impairment. The flavonoids extracted from Ginkgo biloba have also been found to possess anticancer, antisenescence, hepatoprotective, and cardioprotective abilities.12 We previously showed that GbE mitigates oxLDLinduced oxidative injury in human endothelial cells by repressing reactive oxygen species (ROS) generation, the earliest signaling event that occurs in oxLDL-caused endothelial dysfunction.13 Ji et al14 reported that GbE reduces glucose-caused accumulation of MMPs in rat mesangial cells. However, the probable mechanism that GbE uses to protect against atherosclerosis development remains to be elucidated. GbE has been shown to be a PPAR-g activator.15 The transcriptional activity of PPAR-g could be inhibited by MAPK/ERK kinase (MEK) activation. Thus, we sought to confirm in this present study whether GbE could inhibit the oxLDL-induced overexpression of MMP-1, MMP-2,

and MMP-3 in human endothelial cells by LOX-1 and whether the PKC/ERK/PPAR-g/MMPs signaling pathway mediates the course. METHODS Reagents. GbE extracts were gained from Dr Willmar Schwabe (Karlsruhe, Germany). The vehicle control was 30% polyethylene glycol. Fetal bovine serum, M199, and trypsin-ethylenediaminetetraacetic acid were obtained from Gibco BRL Life Technologies (Grand Island, NY). Low serum growth supplement was obtained from Gibco. BAPTA-AM (1,2-Bis(2-amino-5-methylphenoxy)ethaneN,N,N0 ,N0 -tetraacetic acid tetrakis(acetoxymethyl) ester), diphenyleneiodonium chloride (DPI), Gö6976, PD98059, and U0126 were obtained from Sigma-Aldrich (St. Louis, Mo); anti-LOX-1, - 2, and -3 and MMPs enzyme-linked immunosorbent assay (ELISA) kits were from R&D Systems (Minneapolis, Minn), and MMPs activity kits were from AnaSpec (Fremont, Calif). Anti-ERK, antiphosphoERK, anti-PKC-a/b, anti-proliferating cell nuclear antigen, anti-NF-kB/p65, anti-IkB, and anti-PPAR-g were bought from Cell Signaling Technology (Danvers, Mass). Short interfering (si)RNAs for nontargeting si-control and si-PPAR-g (NM_005037) were obtained from Dharmacon Research (Lafayette, Colo). Cell cultures. Human umbilical vein endothelial cells (HUVECs), obtained from American Type Culture Collection (Manassas, Va), were cultured in 199 Medium with a low serum growth supplement. Culture dishes were coated with gelatin for 2 hours. Penicillin and streptomycin were used as an antibiotic. Trypsinethylenediaminetetraacetic acid was used for cell passage. Immunoblotting. After treatment with oxLDL, the cytosolic/nuclear protein of cells was isolated by a Cytoplasmic Extraction kit (Pierce, Rockford, Ill), according to the manufacturer’s instructions. Total protein was determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and an immunoblot assay. The blots were incubated with 5% milk for 1 hour and then incubated with primary antibodies 1 hour at room temperature, respectively, followed by incubation with horseradish peroxidaseconjugated secondary antibody for 1 hour. An enhanced chemiluminescent assay (Amersham, Berkshire, United Kingdom) was used to detect the bound immunoproteins. The intensities of protein expressions were quantified by densitometric analysis (Digital Protein DNA Imagineware, Huntington Station, NY). Protein expressions were normalized by internal control genes and indicated by a bar chart. Isolation of messenger RNA and quantitative realtime polymerase chain reaction. The RNeasy kit (Qiagen, Valencia, Calif) was used for total RNA isolation from endothelial cells. Oligonucleotide specificity was confirmed by a homology search dependent on the human genome (Basic Local Alignment Search Tool, National Center for Biotechnology Information, Bethesda, Md) and validated using a dissociation curve analysis. The oligonucleotide primer sequences are reported in the Table.

JOURNAL OF VASCULAR SURGERY Volume -, Number -

Tsai et al 3

Table. The oligonucleotide sequences Gene LOX-1 MMP-1 MMP-2 MMP-3 TIMP-1 b-Actin

Sense

Antisense

50 -GATGCCCCACTTGTTCAGAT-30 50 -TGGGATATTGGAGCAGCAAGAGGCT-30 50 -CGCTCAGATCCGTG GTGA-30 50 -ACCTGACTCGGTTCCGCCTGT-30 50 -TATCCGGTACGCCTACACCC-30 50 -CGGGAAATCGTGCGTGAC-30

50 -CAGAGTTCGCACCTACGTCA-30 50 -GCAGCAGCAGCAGTGGAGGA-30 50 -CGCCAAATAAACCGGTCCTT-30 50 - CAGTTGGCTGGCGTCCCAGG-30 50 -TGGGCATATCCACAGAGGCT-30 50 -TGCCCAGGAAGGAAGGCT-30

LOX-1, Lectin-like oxidized low-density lipoprotein receptor 1; MMP, matrix metalloproteinase; TIMP-1, tissue inhibitor of MMP 1.

Fig 1. Effect of Gingko biloba extract (GbE) on oxidized low-density lipoprotein (oxLDL)-induced matrix metalloproteinase (MMP)-1, MMP-2, and MMP-3 activation and tissue inhibitor of MMP-1 (TIMP-1) expression. A, MMP-1, MMP-2, MMP-3 and (B) TIMP-1 messenger RNA (mRNA) or (C-E) protein levels in human umbilical vein endothelial cells (HUVECs) that were pretreated with GbE (12.5-100 mM) for 2 hours, followed by exposure to oxLDL (130 mg/mL) for an additional 24 hours. F, Enzyme linked immunosorbent assay (ELISA) assay and (G) MMP-1, MMP-2, and MMP-3 activity assays were used to test MMP-1, MMP-2, and MMP-3 release and activity under oxLDL stimulation. Cells were lysed at the end of the incubation period, and lectin-like oxLDL receptor 1 (LOX-1) mRNA and protein were analyzed by real-time polymerase chain reaction and Western blotting, respectively. The mRNA and protein levels of MMP-1, MMP-2, and MMP-3, and TIMP-1 were normalized to the level of b-actin. The data illustrated in the graphs represent the mean 6 standard error of the mean of three different experiments. #P < .05 vs untreated control; *P < .05 vs oxLDL treatment.

Measurement of ROS production. The ability of GbE to inhibit ROS formation in endothelial cells was investigated by the superoxide indicator dihydroethidium (DHE). Confluent endothelial cells were pretreated with

GbE for 2 hours, and oxLDL was incubated to the medium with or without GbE. A fluorescence microplate reader was used to analyze fluorescence intensity by an excitation at 540 nm and an emission at 590 nm.

4 Tsai et al

JOURNAL OF VASCULAR SURGERY --- 2014

Fig 2. Effect of Gingko biloba extract (GbE) on oxidized low-density lipoprotein (oxLDL)-induced lectin-like oxLDL receptor 1 (LOX-1) activation. Human umbilical vein endothelial cells (HUVECs) were pretreated with GbE (12.5100 mM) or diphenyleneiodonium chloride (DPI; 5 mM) for 2 hours, followed by exposure to oxLDL (130 mg/mL) for 24 hours. Cells were lysed at the end of the incubation period, and (A) LOX-1 messenger RNA (mRNA) and (B and C) protein were analyzed by real-time polymerase chain reaction and Western blotting, respectively. GAPDH, Glyceraldehyde-3-phosphate dehydrogenase. The mRNA and protein levels of LOX-1 were normalized to the level of b-actin. The data illustrated in the graph represent the mean 6 standard error of the mean of three different experiments. #P < .05 vs untreated control; *P < .05 vs oxLDL treatment.

The formula [(Ft2  Ft0)/Ft0]  100, was used to calculate the percentage increase of fluorescence, with Ft2 indicating the fluorescence at 2 hours of oxLDL exposure and Ft0 indicating the fluorescence at 0 hours of oxLDL exposure. Details of the Methods can be found in the Supplementary Methods (online only). Statistical analyses. The data are shown as the mean 6 standard deviation. Differences between groups were analyzed by a one-way analysis of variance, followed by the Bonferroni post hoc test. A statistical significance was defined as a P value of <.05. RESULTS GbE suppressed the oxLDL-facilitated MMPs overexpression. We previously documented that GbE attenuates oxLDL-mediated oxidative injury in HUVECs at concentrations from 12.5 to 100 mg/mL.13 In this study, endothelial cells were pretreated with GbE (12.5100 mg/mL) for 2 hours and were stimulated with oxLDL (130 mg/mL) for an additional 24 hours. The RNA and protein levels of MMP-1, MMP-2, and MMP-3 and tissue inhibitor of MMP-1 (TIMP-1) were investigated by realtime polymerase chain reaction and Western blotting (Fig 1). The messenger RNA (mRNA) levels (Fig 1, A) and protein levels of MMP-1, MMP-2, and MMP-3 (Fig 1, C and D) were active by oxLDL; however, the pretreatment of cells with GbE attenuated the expression

of MMPs. TIMP-1 is the inhibitor of MMP.16 We confirmed pretreatment of GbE reverses oxLDL-inhibited TIMP-1 expression in messenger RNA (Fig 1, B) and protein (Fig 1, C and E). We also used ELISA and MMPs activity assay to test the MMP-1, MMP-2, and MMP-3 concentration and activity in medium (Fig 1, F and G). Pretreatment with GbE significantly protected endothelial cells against oxLDL-promoted MMP-1, MMP-2, and MMP-3 release and activity. GbE mitigated oxLDL-induced LOX-1 activation in HUVECs. LOX-1 is considered a novel target for atherosclerosis. Li et al17 demonstrated that LOX-1 mediated the oxLDL-caused upregulation of MMPs in endothelial cells. Pretreatment of endothelial cells with GbE for 2 hours, followed by exposure to oxLDL for 24 hours, caused the inhibition of LOX-1 mRNA and protein expression (Fig 2). In addition, pretreatment with the ROS inhibitor DPI inhibited oxLDL-derived LOX-1 activation, indicating that ROS plays a key role in the increased protein expression of LOX-1. Our findings suggested that GbE might also act as a LOX-1 repressor when HUVECs are exposed to oxidative stress. GbE inhibited oxLDL-facilitated ROS production by the regulation of LOX-1. To evaluate whether LOX-1 was involved in the protective effects of GbE on oxLDLmediated ROS formation, we confirmed that GbE inhibited oxLDL-facilitated ROS formation (Fig 3, A and B).

JOURNAL OF VASCULAR SURGERY Volume -, Number -

Tsai et al 5

Fig 3. Inhibitory effects of Gingko biloba extract (GbE) on oxidized low-density lipoprotein (oxLDL)-induced reactive oxygen species (ROS) generation in HUVECs. After preincubation for 2 hours with the lectin-like oxLDL receptor 1 (LOX-1) antibody (ab), GbE (12.5-100 mM), and ROS inhibitor (diphenyleneiodonium chloride [DPI]), cells were treated with 130 mg/mL oxLDL for 2 hours, followed by a 1-hour incubation with the superoxide-sensitive fluorescent probe dihydroethidium (DHE; 10 mM). A, Fluorescence images exhibited the ROS level in control cells and human umbilical vein endothelial cells (HUVECs) stimulated with oxLDL in the presence GbE or LOX-1 antibody. (B) The fluorescence intensity of cells was measured with a fluorescence microplate reader. The fluorescence distribution of DHE oxidation is expressed as a percentage of increased intensity. mAb, monoclonal antibody. The activity of (C) superoxide dismutase (SOD) and (D) catalase in HUVECs stimulated with oxLDL in the absence or presence of indicated concentrations of GbE was determined. The data illustrated in the graph represent the mean 6 standard error of the mean of three different experiments. #P < .05 vs untreated control; *P < .05 vs oxLDL treatment.

Pretreatment with an anti-LOX-1 monoclonal antibody and DPI destroyed oxLDL-derived ROS, suggesting that ROS (a key regulator after oxLDL stimulation) was significantly based on the interaction of oxLDL to LOX-1 and the subsequent activation of reduced nicotinamide adenine dinucleotide phosphate (NADPH) oxidase. This result corresponds with Fig 2, in which GbE was able to modulate LOX-1 expression during HUVEC exposure to oxLDL, thereby mediating a negative feedback for NADPH oxidase upregulation and ROS formation. To validate the mechanisms of GbE’s antioxidant function in endothelial cells exposed to oxLDL, we tested antioxidant enzyme activities in endothelial cells exposed with oxLDL stimulation. Antioxidant enzyme activities, such as superoxide dismutase and catalase, were reduced in endothelial cells treated with oxLDL (Fig 3, C and D). Yet in HUVECs with GbE pretreatment, superoxide dismutase and catalase activities were significantly potentiated. Consistent with our previous findings, pretreatment with an anti-LOX-1 monoclonal antibody reversed oxLDLrepressed antioxidant enzymes function, indicating that inhibition of antioxidant enzymes functions were mainly based on the interaction of oxLDL and LOX-1 and the subsequent production of ROS.

GbE reduced oxLDL-induced PKC and ERK activation. A previous report revealed that ROSs are capable of oxidizing targets that participate in the regulation of MMP, such as PKC.18 Thus, we wanted to know whether GbE regulates PKC phosphorylation in endothelial cells by oxLDL stimulation. As shown in Fig 4 (A and B), oxLDL remarkably up-regulated the phosphorylation of PKC-a/b in the non-GbE-pretreated samples. OxLDL-induced PKCa/b activation was reduced by treatment with the ROS inhibitor (DPI) and LOX-1 antibody, suggesting that GbE inhibits the oxLDL-caused phosphorylation of PKC-a/b by mitigating LOX-1-mediated ROS generation. We also found that oxLDL increases PKC-a activity relative to the control. No upregulation in PKC-a activity, however, was seen in endothelial cells that had been intervened on with GbE, LOX-1 antibody, or DPI (Fig 4, C). ERK is known to regulate MMP function in a variety of cells.19 PKC activation has also been identified as an upstream promoter for ERK phosphorylation.20 A previous study by our group showed that oxLDL-induced MMP overexpression occurred by the MEK/ERK mechanism and not through the p38 or phosphoinositide-3 kinase (PI-3K) molecules.21 We tested the probable participation of ERK in oxLDL-induced MMPs overexpression by

6 Tsai et al

JOURNAL OF VASCULAR SURGERY --- 2014

Fig 4. Inhibitory effect of Gingko biloba extract (GbE) on oxidized low-density lipoprotein (oxLDL)-induced protein kinase C (PKC) activation. Human umbilical vein endothelial cells (HUVECs) were preincubated for 2 hours with GbE (12.5-100 mM) and the reactive oxygen species (ROS) inhibitor (diphenyleneiodonium chloride [DPI]), followed by exposure to oxLDL (130 mg/mL) for 2 hours. A and B, At the end of the incubation period, the levels of phosphorylated (p)-PKC were determined by immunoblotting. The protein levels of p-PKC-a and PKC-b were normalized to the level of total PKC. ab-LOX-1, antibody to lectin-like oxLDL receptor 1. C, PKC-a activity in whole-cell lysates was measured by a fluorescein green assay kit. The data illustrated in the graph represent the mean 6 standard error of the mean of three different experiments. #P < .05 vs untreated control; *P < .05 vs oxLDL treatment.

Fig 5. Protective effect of Gingko biloba extract (GbE) on oxidized low-density lipoprotein (oxLDL)-activated expression of extracellular signal-regulated kinase (ERK) and phosphorylated (p)-ERK. Human umbilical vein endothelial cells (HUVECs) were preincubated for 2 hours with GbE (12.5-100 mM) and the reactive oxygen species (ROS) inhibitor (diphenyleneiodonium chloride [DPI]), followed by exposure to oxLDL (130 mg/mL) for 1 hour. A and B, At the end of the incubation period, the levels of p-ERK were determined by immunoblotting. The protein level of p-ERK was normalized to the level of total ERK. The data illustrated in the graph represent the mean 6 standard error of the mean of three different experiments. #P < .05 vs untreated control; *P < .05 vs oxLDL treatment.

HUVECs. Our results revealed that levels of oxLDLactivated ERK phosphorylation were reversed after treatment with GbE. DPI, a ROS inhibitor, also protected endothelial cells against oxLDL-mediated ERK activation, indicating that the protective effects provided by of GbE are mainly embedded in its antioxidant capabilities (Fig 5, A and B). GbE repressed oxLDL-increased ERK phosphorylation and NF-kB activation by modulating the ROS/ CaDD/PKC axis. The oxLDL-facilitated ROSs are able to modulate several signaling transduction pathways, including MAPKp38 and PI-3K, thereby inducing endothelial cell dysfunction, p38 and PI-3K can both result in the upregulation of NF-kB, which subsequently promotes proinflammatory events. A previous study determined that an increased level of ROS could induce the upregulation of MMPs by MAPK/NF-kB signaling.22 To investigate whether NF-kB upstream modulators participate in the inhibitory functions of GbE-mitigated oxLDL-caused activation of MMPs, endothelial cells were pretreated with pharmacologic inhibitors of ROS (DPI), Caþþ, (1,2-bis(o-aminophenoxy) ethane-N,N,N0 ,N0 -tetraacetic acid [BAPTA]), PKC-a/b (Gö6976), and ERK (PD98059). As demonstrated in Fig 6, the expression of oxLDL-promoted ERK phosphorylation and NF-kB activation, which is downstream of ERK phosphorylation, was conspicuously restrained in cells pretreated with GbE, DPI, BAPTA, Gö6976, and PD98059. Taken together, our findings conclude that ROS formation and Caþþ-derived PKC-mediated ERK signaling may be participating in the inhibition of oxLDL-caused NF-kB upregulation by GbE.

JOURNAL OF VASCULAR SURGERY Volume -, Number -

Tsai et al 7

Fig 6. Oxidized low-density lipoprotein (oxLDL)-induced activation of extracellular signal-regulated kinase (ERK) and nuclear factor (NF)-kB in the presence of pharmacologic inhibitors or Gingko biloba extract (GbE). The pretreatment of human umbilical vein endothelial cells (HUVECs) with the indicated concentrations of diphenyleneiodonium chloride (DPI), 1,2-bis(o-aminophenoxy)ethane-N,N,N’,N’-tetraacetic acid (BAPTA), Gö6976, PD98059, or GbE attenuated ERK activation, NF-kB inhibitor-a (I-kBa) degradation, and NF-kB activation induced by oxLDL. HUVECs were pretreated with each inhibitor or GbE for 2 hours, followed by incubation with oxLDL (130 mg/ml) for 1 hour. A-D, The cells were lysed at the end of the incubation period, and the proteins were analyzed by Western blotting. The protein levels of phosphorylated (p)-ERK, NF-kB, and IkB-a were normalized to the levels of total ERK, proliferating cell nuclear antigen (PCNA), and b-actin, respectively. The data illustrated in the graph represent the mean 6 standard error of the mean of three separate experiments. #P < .05 vs untreated control; *P < .05 vs oxLDL treatment.

Knockdown of PPAR-g impaired the inhibitory effects of oxLDL-activated MMP expressions by GbE. The ERK pathway is fundamental in the signal transduction of free radicals. PPAR-g functions as a transcription factor that controls proliferation and differentiation and is known to have key roles in obesity, diabetes, and inflammation.23 Previous studies demonstrated that upregulated MEK can inhibit the transcriptional abilities of PPAR-g10 and that PPAR-g is able to downregulate the expression of MMPs in human myeloid leukemia cells.24 We hypothesized that oxLDL-activated LOX-1/PKC signaling was the mechanism by which PPAR-g expression was mitigated. Stimulation with oxLDL causes inhibition in the protein level of PPAR-g (Fig 7, A and B). However, intervention with GbE reversed the level of PPAR-g protein. We also confirmed the ROS inhibitor (DPI), antibody to LOX-1, Gö6976, and PD98059 noticeably reduced oxLDL-repressed PPAR-g expression and NF-kB inhibitor (pyrrolidine dithiocarbamate), and partially reduced oxLDLrepressed PPAR-g expression (Fig 7, C and D). This finding suggested that the LOX-1/PKC/ERK/NF-kB axis is

important for the regulation of PPAR-g function in oxLDLtreated endothelial cells. This novel finding indicated that GbE may also play the role of a PPAR-g activator, because PPAR-g activators are recognized as being protective in the clinical intervention of human cardiovascular diseases. We silenced PPAR-g gene expression with siRNA to validate the protective role of GbE in oxLDL-induced oxidative injury. As demonstrated, a PPAR-g siRNA transfection for 72 hours effectively mitigated PPAR-g protein expression (Fig 8, A). The protective effects of GbE in oxLDLinduced MMP-1, MMP-2, and MMP-3 overexpression were ablated in HUVECs transfected with PPAR-g siRNA (Fig 8, B and C). The phenomenon was also confirmed by ELISA assay. Fig. 8, D shows the protective effects of GbE on oxLDL-promoted MMP-1, MMP-2, and MMP-3 release were absent when PPAR-g was silenced. Rosiglitazone, a PPAR-g agonist, was used as positive control, proving that oxLDL mediated MMP-1, MMP-2, and MMP-3 activation by modulating the function of PPAR-g.

8 Tsai et al

JOURNAL OF VASCULAR SURGERY --- 2014

Fig 7. Gingko biloba extract (GbE) reversed oxidized low-density lipoprotein (oxLDL)-diminished peroxisome proliferator-activated receptor-g (PPAR-g) expression. A and B, Human umbilical vein endothelial cells (HUVECs) were preincubated for 2 hours with GbE (12.5-100 mM) or the reactive oxygen species (ROS) inhibitor (diphenyleneiodonium chloride [DPI]), antibody to lectin-like oxLDL receptor 1 (ab-LOX-1), protein kinase C (PKC) inhibitor (Gö6976), extracellular signal-regulated kinase (ERK) inhibitor (PD98069), and nuclear factor (NF)-kB inhibitor (pyrrolidine dithiocarbamate [PDTC]), (C and D) followed by exposure to oxLDL (130 mg/mL) for 1 hour. At the end of the incubation period, the levels of PPAR-g for both were determined by immunoblotting. The protein levels of PPAR-g were normalized to the level of b-actin. The data illustrated in the graph represent the mean 6 standard error of the mean of three different experiments. #P < .05 vs untreated control; *P < .05 vs oxLDL treatment.

GbE consists of 24% Ginkgo flavone glycoside; thus, the therapeutic abilities of GbE may be due to its different components. The antioxidant ability of quercetin, one of the major components of GbE, has been reported.13 We compared the efficacy of GbE and quercetin in inhibiting MMP secretion. Quercetin, at the concentration of 12.4 mg/mL (equivalent to 100 mg/mL of GbE), was not as effective as GbE in oxLDL-facilitated MMP-1, MMP-2, and MMP-3 release (Fig 8, E). Therefore, we assume that quercetin may not be the main component in the mitigation of oxLDL-regulated MMPs activation by GbE. DISCUSSION MMP-inhibition therapies are considered to be novel clinical interventions. The MMPs are expressed in human vascular muscle cells and in endothelial cells, both of which are involved in the progression of atherosclerosis.25 LOX-1, a key receptor for oxLDL, is able to mediate severe atherosclerosis-related molecules and signaling transduction pathways, including the MAPK, PI-3K, and NF-kB pathways.26 Our group previously found that GbE exerts its protective effects against oxLDL-triggered endothelial oxidative injury and apoptosis.13 In the current study, we well demonstrated that GbE attenuated oxLDL-induced MMPs expression through the regulation of the LOX-1mediated ROS/PKC/ERK/NF-kB signaling. Our results also revealed that the inhibitory effects of GbE on oxLDL-

induced MMP-1, MMP-2, and MMP-3 activation were ablated by PPAR-g siRNA, suggesting that PPAR-g activation participates in the GbE-mediated mechanism. Furthermore, DPI suppressed the oxLDL-activated LOX-1/PKC/ ERK/NF-kB pathway. This finding indicates that the repression of oxLDL-facilitated ROS formation acts an important role in the cytoprotective effect of GbE (Fig 9). Being a major receptor for oxLDL in human endothelial cells, LOX-1 was originally known to be a molecule that caused endothelial oxidative damage under the exposure of oxLDL. Once oxLDL interacts to LOX-1, it induces rapid activation of NADPH oxidase, resulting in ROS generation.27 Because LOX-1 upregulation is considered an upstream inducer for endothelial cell and cardiomyocyte death, the suppression of LOX-1 upregulation and the repression of LOX-1-derived oxidative pathways may potentially be useful to treat cardiovascular diseases. One report showed that the mitigation of LOX-1 with a specific antibody reduced cardiomyocyte apoptosis and infarct size in rats with ischemic injury and reduced hypertension-induced endothelial cell injury,28 reiterating that LOX-1 repression may act against cardiovascular diseases. In addition, MMP activation has been demonstrated to be essential in the development of atherosclerosis and heart failure.29 ROS formation facilitated by oxidative stress has been reported to occur by recurring LOX-1 and MMP. Sugimoto et al30 demonstrated that the inhibition of LOX-1

JOURNAL OF VASCULAR SURGERY Volume -, Number -

Tsai et al 9

Fig 8. Inhibition of peroxisome proliferator-activated receptor-g (PPAR-g) with short interfering (si)RNA antagonized the effects of Gingko biloba extract (GbE) on oxidized low-density lipoprotein (oxLDL)-induced matrix metalloproteinase (MMP)-1, MMP-2, and MMP-3 activation. Human umbilical vein endothelial cells (HUVECs) were transfected with PPAR-g siRNA for 72 hours and were treated with 100 mg/mL GbE for 1 hour, followed by exposure to 130 mg/mL oxLDL for 24 hours. A, The Western blot analysis of the si-5’ adenosine monophosphate-activated protein kinase-a knockdown efficiency. B and C, The cell lysates were analyzed by Western blot using anti-MMP-1, anti-MMP-2, and anti-MMP-3, or anti-b-actin antibodies. mRNA, Messenger RNA. D, Enzyme-linked immunosorbent assay (ELISA) was used to examine the protective effects of GbE on oxLDL-promoted MMP-1, MMP-2, and MMP-3 release. E, Inhibitory efficiency of GbE and quercetin were compared in the same concentration. The values represent means 6 standard error of the mean from four separate experiments. #P < .05 vs untreated control; *P < .05 vs oxLDL treatment; &P < .05 vs GbE þ oxLDL treatment.

and MMP expression suppressed oxidative stress-mediated NADPH oxidase activation. Most importantly, LOX-1, oxLDL, and MMPs have been shown to be involved in atherosclerotic plaques.31 We therefore hypothesized that GbE protects endothelial cells against oxLDL-activated MMPs overexpression by modulating LOX-1-derived signaling. As indicated in Fig 2, GbE is able to mitigate LOX-1 expression in oxLDL-treated HUVECs. We then proved that GbE inhibited ROS production by inhibiting the LOX-1-mediated signaling. In addition, oxLDLfacilitated ROS production was eliminated by pretreatment with GbE and a monoclonal antibody to LOX-1, revealing that GbE has a major role in oxLDL-facilitated ROS generation and LOX-1 expression.

Increasing evidence has identified that PKC, an important upregulator of NADPH oxidase, can be negatively controlled by 5’-adenosine monophosphate-activated protein kinase. With the downregulation of PKC, NADPH oxidaseinduced ROS production can be reduced.32 PKC contains multiple cysteine remains, and those could be oxidized and activated by free radicals.33 Li et al17 reported that PKC plays an important role in oxLDL-activated MMP overexpression by modulating LOX-1 function. In our investigation, we hypothesized that GbE exerts its antiatherogenic function by obstructing ROS-mediated PKC activation. In line with previous reports that PKC signaling is important in the modulation of MMP expression, we confirmed that GbE reduced the oxLDL-caused overexpression of MMP-1, MMP-2, and

JOURNAL OF VASCULAR SURGERY --- 2014

10 Tsai et al

Fig 9. Schematic diagram shows the signaling cascades involved in the attenuation of matrix metalloproteinase (MMP)-1, MMP-2, and MMP-3 expression in cells exposed to oxidized low-density lipoprotein (oxLDL) that were treated with Gingko biloba extract (GbE). As depicted, GbE inhibited oxLDL-induced lectin-like oxLDL receptor 1 (LOX-1) activation, thereby inhibiting oxLDL-facilitated reactive oxygen species (ROS) formation and protein kinase C (PKC) activation. We also confirmed that Caþþ, PKC, mitogen-activated protein kinase (MAPK), and ROS were involved in the protective actions of GbE. Moreover, we revealed that GbE pretreatment reversed oxLDL-repressed peroxisome proliferator-activated receptor-g (PPAR-g) expression. Finally, we found that GbE mitigates oxLDLpromoted MMP-1, MMP-2, and MMP-3 activation by modulation of PPAR-g. The / indicates activation or induction, and indicates inhibition or blockade. w

MMP-3 by mitigating the formation of ROS and the phosphorylation of PKC. This finding was further validated by the observation that the intervention of endothelial cells with DPI repressed the oxLDL-derived phosphorylation of PKC. NADPH oxidase-generated ROSs are able to activate MAPK and NF-kB activation. Previous studies have demonstrated that oxLDL-induced ERK activation occurs through LOX-1 and that this effect further produces the activation of MMP-1, MMP-2, and MMP-3 in THP-1 macrophages.34 Because our evidence showed that GbE reduced oxLDL-caused phosphorylation of ERK and the activation of NF-kB, we therefore further investigated whether the use of pharmacologic inhibitors of ROSs (DPI), Caþþ (BAPTA), and PKC-a/b (Gö6976) could attenuate the activation of NF-kB. Our results indicated that MMP-1, MMP-2, and MMP-3 downregulation by GbE may cross-talk with the mitigation of ROS-derived Caþþ-dependent PKC activation and the subsequent activation of ERK and NF-kB. Our data were comparable with previous findings that quercetin inhibits MMP-1 activation by repressing ERK phosphorylation and that ascochlorin mitigates oxLDL-activated MMP-9 upregulation by mitigating the MEK/ERK pathway.34

PPAR-g suppresses oxidative injury generation and progression by modulating proinflammatory reactions in human endothelial cells. PPAR-g also inhibits adhesion molecular expression by a process mediated by tumor necrosis factor-a,35 suggesting that PPAR-g acts to protect against oxidative damage. Our data in the current study suggest that pretreatment of GbE inhibits oxLDLrepressed PPAR-g expression (Fig 6, A and B). This result is consistent with a previous report by Zhou et al15 that indicated that GbE could functionally upregulate PPARg expression. The protective effects of GbE on oxLDL-activated expression of MMPs were ablated in cells transfected with PPAR-g siRNA. Therefore, our findings suggest that GbE suppresses the oxLDL-induced expression of MMPs through the reduction of ROS formation, and which in turn, upregulates the Caþþ-dependent PKC-a and ERK/PPAR-g/NF-kB signaling pathways. The limitations of this study are, first, that we used in vitro investigations to test the protective effects of GbE. Animal studies will be the major part in our further study. Second, the protective effects of GbE should be confirmed in other cells, for example, smooth or skeletal muscle cells.

JOURNAL OF VASCULAR SURGERY Volume -, Number -

CONCLUSIONS Our study demonstrates that GbE can inhibit oxLDLinduced endothelial damage. The protective effects of GbE may arise from enhanced nitric oxide bioavailability, maintenance of intracellular Caþþ, mitochondrial stabilization, and apoptosis prevention.13 We also found GbE mitigated the activation of NADPH oxidase through the oxLDLpromoted 5’ adenosine monophosphate-activated protein kinase/PKC axis.35 Inhibition of those events is thought to be a potential approach for treatment of cardiovascular diseases in clinical practice. Our findings concluded that GbE elicits its antiatherogenic abilities by LOX-1-mediated PKC-a/ERK/PPARg/NF-kB pathways, resulting in the inhibition of ROS formation and, ultimately, the repression of MMP-1, MMP-2, and MMP-3 in endothelial cells exposed to oxLDL. We propose that GbE may potentially become a preventive compound against cardiovascular diseases. AUTHOR CONTRIBUTIONS Conception and design: KT, YLC, PH, DL, HC, CK Analysis and interpretation: KT, YLC, YC, DL, HC, CK Data collection: KT, PH, YC, DL, CK Writing the article: KT, YLC, PH, YHC, CK Critical revision of the article: KT, YLC, PH, CK, Final approval of the article: KT, DL, CK Statistical analysis: KT, YHC, CK Obtained funding: KT, YLC, PH, YC, YHC, CK Overall responsibility: CK REFERENCES 1. Galle J, Lehmann-Bodem C, Hubner U, Heinloth A, Wanner C. CyA and OxLDL cause endothelial dysfunction in isolated arteries through endothelin-mediated stimulation of O(2)(-) formation. Nephrol Dial Transplant 2000;15:339-46. 2. Murase T, Kume N, Korenaga R, Ando J, Sawamura T, Masaki T, et al. Fluid shear stress transcriptionally induces lectin-like oxidized LDL receptor-1 in vascular endothelial cells. Circ Res 1998;83:328-33. 3. Galis ZS, Sukhova GK, Libby P. Microscopic localization of active proteases by in situ zymography: detection of matrix metalloproteinase activity in vascular tissue. FASEB J 1995;9:974-80. 4. Vuilleumier N, Bas S, Pagano S, Montecucco F, Guerne PA, Finckh A, et al. Anti-apolipoprotein A-1 IgG predicts major cardiovascular events in patients with rheumatoid arthritis. Arthritis Rheum 2010;62: 2640-50. 5. Mehta JL, Chen J, Hermonat PL, Romeo F, Novelli G. Lectin-like, oxidized low-density lipoprotein receptor-1 (LOX-1): a critical player in the development of atherosclerosis and related disorders. Cardiovasc Res 2006;69:36-45. 6. Ehara S, Ueda M, Naruko T, Haze K, Itoh A, Otsuka M, et al. Elevated levels of oxidized low density lipoprotein show a positive relationship with the severity of acute coronary syndromes. Circulation 2001;103:1955-60. 7. Shin Y, Yoon SH, Choe EY, Cho SH, Woo CH, Rho JY, et al. PMAinduced up-regulation of MMP-9 is regulated by a PKCalpha-NFkappaB cascade in human lung epithelial cells. Exp Mol Med 2007;39:97-105. 8. Wang HH, Hsieh HL, Wu CY, Yang CM. Oxidized low-density lipoprotein-induced matrix metalloproteinase-9 expression via PKCdelta/p42/p44 MAPK/Elk-1 cascade in brain astrocytes. Neurotox Res 2010;17:50-65.

Tsai et al 11

9. Kliewer SA, Umesono K, Noonan DJ, Heyman RA, Evans RM. Convergence of 9-cis retinoic acid and peroxisome proliferator signalling pathways through heterodimer formation of their receptors. Nature 1992;358:771-4. 10. Hu E, Kim JB, Sarraf P, Spiegelman BM. Inhibition of adipogenesis through MAP kinase-mediated phosphorylation of PPARgamma. Science 1996;274:2100-3. 11. Martin-Nizard F, Furman C, Delerive P, Kandoussi A, Fruchart JC, Staels B, et al. Peroxisome proliferator-activated receptor activators inhibit oxidized low-density lipoprotein-induced endothelin-1 secretion in endothelial cells. J Cardiovasc Pharmacol 2002;40:822-31. 12. Tunali-Akbay T, Sener G, Salvarli H, Sehirli O, Yarat A. Protective effects of Ginkgo biloba extract against mercury(II)-induced cardiovascular oxidative damage in rats. Phytother Res 2007;21:26-31. 13. Ou HC, Lee WJ, Lee IT, Chiu TH, Tsai KL, Lin CY, et al. Ginkgo biloba extract attenuates oxLDL-induced oxidative functional damages in endothelial cells. J Appl Physiol 2009;106:1674-85. 14. Ji L, Yin XX, Wu ZM, Wang JY, Lu Q, Gao YY. Ginkgo biloba extract prevents glucose-induced accumulation of ECM in rat mesangial cells. Phytother Res 2009;23:477-85. 15. Zhou L, Meng Q, Qian T, Yang Z. Ginkgo biloba extract enhances glucose tolerance in hyperinsulinism-induced hepatic cells. J Nat Med 2011;65:50-6. 16. Shiryaev SA, Remacle AG, Golubkov VS, Ingvarsen S, Porse A, Behrendt N, et al. A monoclonal antibody interferes with TIMP-2 binding and incapacitates the MMP-2-activating function of multifunctional, pro-tumorigenic MMP-14/MT1-MMP. Oncogenesis 2013;2:e80. 17. Li D, Liu L, Chen H, Sawamura T, Ranganathan S, Mehta JL. LOX-1 mediates oxidized low-density lipoprotein-induced expression of matrix metalloproteinases in human coronary artery endothelial cells. Circulation 2003;107:612-7. 18. Wu WS. The signaling mechanism of ROS in tumor progression. Cancer Metastasis Rev 2006;25:695-705. 19. Borden P, Heller RA. Transcriptional control of matrix metalloproteinases and the tissue inhibitors of matrix metalloproteinases. Crit Rev Eukaryot Gene Expr 1997;7:159-78. 20. Wiejak J, Dunlop J, Stoyle C, Lappin G, McIlroy A, Pediani JD, et al. The protein kinase C inhibitor, Ro-31-7459, is a potent activator of ERK and JNK MAP kinases in HUVECs and yet inhibits cyclic AMPstimulated SOCS-3 gene induction through inactivation of the transcription factor c-Jun. Cell Signal 2012;24:1690-9. 21. Kuo MY, Ou HC, Lee WJ, Kuo WW, Hwang LL, Song TY, et al. Ellagic acid inhibits oxidized low-density lipoprotein (OxLDL)induced metalloproteinase (MMP) expression by modulating the protein kinase C-alpha/extracellular signal-regulated kinase/peroxisome proliferator-activated receptor gamma/nuclear factor-kappaB (PKCalpha/ERK/PPAR-gamma/NF-kappaB) signaling pathway in endothelial cells. J Agric Food Chem 2011;59:5100-8. 22. Tung WH, Tsai HW, Lee IT, Hsieh HL, Chen WJ, Chen YL, et al. Japanese encephalitis virus induces matrix metalloproteinase-9 in rat brain astrocytes via NF-kappaB signalling dependent on MAPKs and reactive oxygen species. Br J Pharmacol 2010;161:1566-83. 23. Chinetti G, Fruchart JC, Staels B. Peroxisome proliferator-activated receptors (PPARs): nuclear receptors at the crossroads between lipid metabolism and inflammation. Inflamm Res 2000;49:497-505. 24. Liu J, Lu H, Huang R, Lin D, Wu X, Lin Q, et al. Peroxisome proliferator activated receptor-gamma ligands induced cell growth inhibition and its influence on matrix metalloproteinase activity in human myeloid leukemia cells. Cancer Chemother Pharmacol 2005;56:400-8. 25. Newby AC. Matrix metalloproteinase inhibition therapy for vascular diseases. Vascul Pharmacol 2012;56:232-44. 26. Tsai KL, Chen LH, Chiou SH, Chiou GY, Chen YC, Chou HY, et al. Coenzyme Q10 suppresses oxLDL-induced endothelial oxidative injuries by the modulation of LOX-1-mediated ROS generation via the AMPK/PKC/NADPH oxidase signaling pathway. Mol Nutr Food Res 2011;55(Suppl 2):S227-40. 27. Sawamura T, Kume N, Aoyama T, Moriwaki H, Hoshikawa H, Aiba Y, et al. An endothelial receptor for oxidized low-density lipoprotein. Nature 1997;386:73-7.

12 Tsai et al

28. Li D, Williams V, Liu L, Chen H, Sawamura T, Romeo F, et al. Expression of lectin-like oxidized low-density lipoprotein receptors during ischemia-reperfusion and its role in determination of apoptosis and left ventricular dysfunction. J Am Coll Cardiol 2003;41:1048-55. 29. Pasterkamp G, Schoneveld AH, Hijnen DJ, de Kleijn DP, Teepen H, van der Wal AC, et al. Atherosclerotic arterial remodeling and the localization of macrophages and matrix metalloproteases 1, 2 and 9 in the human coronary artery. Atherosclerosis 2000;150:245-53. 30. Sugimoto K, Ishibashi T, Sawamura T, Inoue N, Kamioka M, Uekita H, et al. LOX-1-MT1-MMP axis is crucial for RhoA and Rac1 activation induced by oxidized low-density lipoprotein in endothelial cells. Cardiovasc Res 2009;84:127-36. 31. Chen H, Li D, Sawamura T, Inoue K, Mehta JL. Upregulation of LOX-1 expression in aorta of hypercholesterolemic rabbits: modulation by losartan. Biochem Biophys Res Commun 2000;276:1100-4. 32. Ceolotto G, Gallo A, Papparella I, Franco L, Murphy E, Iori E, et al. Rosiglitazone reduces glucose-induced oxidative stress mediated by NAD(P)H oxidase via AMPK-dependent mechanism. Arterioscler Thromb Vasc Biol 2007;27:2627-33.

JOURNAL OF VASCULAR SURGERY --- 2014

33. Shih RH, Cheng SE, Hsiao LD, Kou YR, Yang CM. Cigarette smoke extract upregulates heme oxygenase-1 via PKC/NADPH oxidase/ ROS/PDGFR/PI3K/Akt pathway in mouse brain endothelial cells. J Neuroinflammation 2011;8:104. 34. Kang JH, Kim JK, Park WH, Park KK, Lee TS, Magae J, et al. Ascochlorin suppresses oxLDL-induced MMP-9 expression by inhibiting the MEK/ERK signaling pathway in human THP-1 macrophages. J Cell Biochem 2007;102:506-14. 35. Ou HC, Hsieh YL, Yang NC, Tsai KL, Chen KL, Tsai CS, et al. Ginkgo biloba extract attenuates oxLDL-induced endothelial dysfunction via an AMPK-dependent mechanism. J Appl Physiol (1985) 2013;114:274-85.

Submitted Feb 14, 2014; accepted May 26, 2014.

Additional material for this article may be found online at www.jvascsurg.org.

JOURNAL OF VASCULAR SURGERY Volume -, Number -

SUPPLEMENTARY METHODS (online only) Reagents. Chemicals were obtained from the following companies: Ginkgo biloba extract (GbE), a defined complex mixture containing 24% Ginkgo flavone glycoside (primarily composed of quercetin, kaempferol, and isorhamnetin) and 6% terpene lactones (ginkgolides A, B, C, and bilobalide) extracted from Ginkgo biloba leaves, was obtained from Dr Willmar Schwabe (Karlsruhe, Germany). The vehicle control was 30% polyethylene glycol. Fetal bovine serum, M199, and trypsin-ethylenediaminetetraacetic acid were obtained from Gibco (Grand Island, NY); low serum growth supplement (Cascade Biologics Inc, Portland, Ore), 2-Bis(2-aminophenoxy)ethane-N,N,N0 ,N0 -tetraacetic acid tetrakis(acetoxymethyl) ester (BAPTA-AM), diphenyleneiodonium chloride (DPI), Gö6976, pyrrolidine dithiocarbamate, PD98059, U0126, penicillin, and streptomycin were obtained from Sigma-Aldrich (St. Louis, Mo); antilectin-like oxidized low-density lipoprotein (oxLDL) receptor 1 (LOX-1), anti-matrix metalloproteinase (MMP)-1, anti-MMP-2, anti-MMP-3, MMP-1, MMP-2, and MMP-3 enzyme-linked immunosorbent assay kits were obtained from R&D Systems (Minneapolis, Minn); MMP-1, MMP2, and MMP-3 activity kits were obtained from AnaSpec (Freemont, Calif). Antiprotein kinase C (PKC) a/b, antinuclear factor-kB/p65, and anti-nuclear factor-kB inhibitor were obtained from Cell Signaling (Danvers, Mass); anti-peroxisome proliferator-activated receptor-g was obtained from Santa Cruz Biotechnology (Santa Cruz, Calif); antiproliferating cell nuclear antigen was obtained from Transduction Laboratories (San Diego, Calif); and antiERK and anti-phospho-ERK were obtained from BD Biosciences (Franklin Lakes, NJ). Isolation of messenger RNA and quantitative realtime polymerase chain reaction. Total RNA was isolated from human umbilical vein endothelial cells using the RNeasy kit (Qiagen, Valencia, Calif). Oligonucleotides for LOX-1 and b-actin were designed using the computer software package Primer Express 2.0 (Applied Biosystems, Foster City, Calif). All of the oligonucleotides were synthesized by Invitrogen (Breda, The Netherlands). Oligonucleotide specificity was determined by a homology search within the human genome (Basic Local Alignment Search Tool, National Center for Biotechnology Information, Bethesda, Md) and confirmed by a dissociation curve analysis. The oligonucleotide sequences were as follows: LOX-1: sense primer 50 -GATGCCCCACTTGTTCA GAT-30 ; antisense primer 50 -CAGAGTTCGCACCTACG TCA-30 ; MMP-1: sense primer 50 -TGGGATATTGGAGCAG CAAGAGGCT-30 , antisense primer 50 - GCAGCAGCAG CAGTGGAGGA-30 ; MMP-2: sense primer 50 -CGCTCAGATCCGTG GT GA-30 , antisense primer 50 -CGCCAAATAAACCGGTCC TT-30 ;

Tsai et al 12.e1

MMP-3: sense primer 50 - ACCTGACTCGGTTCCG CCTGT-30 , antisense primer: 50 - CAGTTGGCTGGCGT CCCAGG-30 ; TIMP-1: sense primer 50 -TATCCGGTACGCCTAC ACCC-30 , antisense primer 50 - TGGGCATATCCACAGA GGCT-30 ; b-actin: sense primer 50 -AGGTCATCACTATTGGC AACGA-30 ; antisense primer 50 -CACTTCATGATGGA ATTGAATGTAGTT-30 . Polymerase chain reaction was performed with SYBR Green in an ABI 7000 sequence detection system (Applied Biosystems), according to the manufacturer’s guidelines. Lipoprotein oxidation. Human plasma LDL was obtained from Sigma-Aldrich. Copper-modified LDL (1 mg protein/mL) was prepared by exposing LDL to 10 mM CuSO4 for 16 hours at 37 C.16 After oxidation, the amount of thiobarbituric acid reactive substances in LDL ranged from 15 to 20 nmol/mg LDL. Transfection with small interfering RNA. The Ontarget Plus SMART pool small interfering (si)RNAs for nontargeting control and peroxisome proliferator-activated receptor-g (NM_005037) were purchased from Dharmacon Research. Transient transfection was done using INTERFERin siRNA transfection reagent (Polyplus-transfection SA, Illkirch, France), according to the manufacturer’s instructions. Two days after transfection, the cells were treated with the reagent, as indicated, for further experiments. Assay for MMP-1, MMP-2, and MMP-3 secretion. Human umbilical vein endothelial cells were pretreated with the indicated concentrations of GbE for 2 hours, followed by treatment with oxLDL for 24 hours. At the end of the oxLDL incubation period, cell supernatants were removed and assayed for MMP-1, MMP-2, and MMP-3 concentration using an enzyme-linked immunosorbent assay kit obtained from R&D Systems. MMPs activity. At the end of the oxLDL incubation period, cell supernatants were removed and assayed for MMP-1, MMP-2, and MMP-3 activity using activity kit obtained from AnaSpec (Freemont, Calif). PKC-a assay. At the end of the incubation period, cells were rinsed with ice-cold phosphate-buffered saline and lysed by the addition of reaction buffer (50 mM N-2-hydroxyethylpiperazine-N0 -2-ethanesulfonic acid [pH 7.2], 0.01% bovine serum albumin, 10 mM MgCl2, 1 mM dithiothreitol, and 1 lipid activator, provided in the kit). PKC-a activity in 10 mg lysate was measured with a PKC-a activity assay kit (Upstate Biotechnology, Lake Placid, NY), according to the manufacturer’s instructions. Measurement of antioxidant enzyme activity. To determine the effects of GbE after oxLDL exposure, superoxide dismutase and catalase activity in the homogenate was determined by an enzymatic assay method using a commercial kit according to the manufacturer’s instructions. Enzyme activity was converted to units per milligram of protein.