GSEs (grape seed extracts) which contain polyphenolic compounds cause an endothelium-dependent relaxation of blood vessels. The aim of the present study was to examine the mechanisms involved in this response. A well-characterized GSE was applied to rabbit aortic rings suspended in organ baths containing Krebs–Henseleit buffer maintained at 37 °C. In aortic rings pre-contacted with noradrenaline (norepinephrine), the extract produced a dose-dependent relaxation. The maximum relaxations elicited by the extract (71.9±1.0%) were similar to those elicited by acetylcholine (64.2±1.5%) (n=12 for each). As expected, the relaxations were abolished by removal of the endothelium and by prior incubation with L-NAME (NG-nitro-L-arginine methyl ester), confirming the essential role of eNOS (endothelial NO synthase) in the response. The responses to the GSE were also abolished by incubation with wortmannin and LY294002, which are inhibitors of PI3K (phosphoinositide 3-kinase). These compounds had no effect on the responses to acetylcholine. Using immunoblotting, we also demonstrated that the GSE induced the phosphorylation of both Akt and eNOS in HUVECs (human umbilical vein endothelial cells). Finally, the extract was modified by methylation of the hydroxy groups in the polyphenolic groups and was applied to the aortic rings. The modified extract failed to cause a relaxation. Taken together, these findings suggest that the endothelium-dependent relaxation induced by the GSE was mediated by activation of the PI3K/Akt signalling pathway through a redox-sensitive mechanism, resulting in phosphorylation of eNOS.

INTRODUCTION

There is evidence that a diet rich in vegetables and fruit has a beneficial effect on blood pressure. This effect has been attributed to phenolic compounds present in the plants. These compounds have also been shown to influence endothelial function in a variety of experimental situations [13]. In humans, extracts of fruits and vegetables have been shown to enhance flow-mediated vasodilation in the brachial artery [4].

Of all of the phenolic products, those derived from grapes appear to have received the most attention, possibly because of their involvement with the French paradox [5,6]. Grapes and grape products derived from the skin, seeds, pulp and stem are good sources of polyphenolic compounds; however, it has been found that >70% of polyphenolic compound are concentrated in the seeds [7]. GSEs (grape seed extracts) cause an EDR (endothelium-dependent relaxation) of aortic rings in vitro (for example, [2,8,9]). Similarly, these extracts have also been shown to activate eNOS [endothelial NOS (NO synthase)] [2,10] and up-regulate eNOS in cultured endothelial cells [11].

The mechanism mediating this response to GSEs has not been established with certainty. Grape juice [12] and extracts of red wine [13] have been shown to cause EDR, which was abolished by blocking the PI3K (phosphoinositide 3-kinase)/Akt pathway. In the present study, we have examined the effect of a well-characterized GSE, which has been shown previously to cause EDR in guinea pig aortic rings [14] and reduce blood pressure in humans [15], on the PI3K/Akt signalling pathway and phosphorylation of eNOS. The studies were undertaken on both rabbit aortic rings and HUVECs (human umbilical vein endothelial cells).

MATERIALS AND METHODS

Study design and procedures

This study was approved by Animal Use and Care Administrative Advisory Committee, University of California, Davis, CA, U.S.A. Male New Zealand rabbits, weighing 3–3.5 kg, were sedated by intramuscular injection of acepromazine. After 5 min, a lethal dose of sodium pentobarbitone (50 mg/kg of body weight; Abbott Laboratory) was administrated through the lateral ear vein. A thoracotomy was performed and the descending thoracic aorta was excised carefully. The aorta was flushed twice with fresh ice-cold KH (Krebs–Henseleit) buffer (118 mmol/l NaCl, 5.4 mmol/l KCl, 1.2 mmol/l, MgCl2 2.5 mmol/l CaCl2, 22 mmol/l NaHCO3, 1.2 mmol/l NaH2PO4 and 10.1 mmol/l glucose; using Sigma analytical grade reagents) and placed in a dissecting tray filled with the same buffer. All surrounding connective tissues and fat were removed carefully.

The GSE used in the present study is a water extract prepared by Polyphenolics Inc (Meganatural-BP®; patent pending). The extract is made up of polymers of catechin and has an average degree of polymerization of 2.3. The extract was dissolved in KH buffer, and the concentrations of the solution were based on a nominal Mr of 1000. The phenol content of the GSE solution (1 mg/ml) was measured using the Folin–Ciocalteu assay and was found to be 39.2±0.65 mmol/l gallic acid units (n=5). The characterization of the extract is given in Supplementary material available at http://www.clinsci.org/cs/114/cs1140331add.htm.

Measurement of EDR

EDR was assessed as described previously [16]. Briefly, the aorta was segmented into rings (5 mm in length) which were mounted between two tungsten wire triangles. One triangle was attached to a strain-gauge transducer and the other to the bottom of an organ bath (20 ml) containing KH buffer maintained at 37 °C and oxygenated with a mixture of 95% O2/5% CO2. A pre-load of 8 g was applied to the rings, and the tissues were allowed to equilibrate for 60 min. The transducer was connected to a pen recorder (Gould-2400S recorder), and the changes in tensions were monitored using a Windaq computer program (2003 version; Dataq Instruments).

After equilibration for 60 min at a pre-load of 8 g, the aortic rings were pre-contracted with 10 μmol/l noradrenaline (norepinephrine; Sigma). Acetylcholine (Sigma) was added in an incremental manner to achieve bath concentrations from 0.1–10 μmol/l to obtain dose–response curves for EDR. The relaxations were expressed as a percentage of the contraction induced by noradrenaline.

GSE-induced EDR

After demonstrating EDR evoked by acetylcholine, the rings were treated with increasing concentrations of the GSE following pre-contraction with noradrenaline. In additional experiments, the effect of removing the endothelium on relaxation evoked by acetylcholine and the GSE were examined to establish the endothelium-dependent nature of the relaxation. In these experiments, after demonstrating the absence of relaxation, the rings were treated with SNP (sodium nitroprusside; Sigma) to establish the ability of the aortic smooth muscle to relax. As a further control, the effect of incubation with L-NAME (NG-nitro-L-arginine methyl ester; bath concentration, 1 mmol/l; Sigma), a competitive inhibitor of NOS, was examined to demonstrate the involvement of NOS in the relaxation of the rings.

Effect of blocking the PI3K/Akt pathway on EDR induced by the GSE

Previous studies have shown that the EDR evoked by polyphenolic compounds derived from grapes was abolished by inhibitors of the PI3K/Akt pathway [13,17], In the present study, the effect of the GSE was examined after incubating the aortic rings with wortmannin (30 nmol/l; Sigma) and LY294002 (30 μmol/l; Sigma) in KH buffer. Both wortmannin and LY294002 are potent and specific PI3K inhibitors. In testing the effect of each inhibitor, three aortic rings were tested simultaneously according to the sequence shown in Table 1. This protocol was based on a previous finding that prior exposure to the GSE and other phenolic compounds (e.g. cocoa) attenuated the effect of subsequent exposure [2,10]. Thus it is not possible to expose a ring to the same extract twice, before and after exposure to the inhibitor, and obtain meaningful data.

Table 1
Protocol for testing the effect of the PI3K inhibitors on EDR induced by the GSE
Step Ring 1 Ring 2 Ring 3 
Acetylcholine (10 μmol/l) Acetylcholine (10 μmol/l) Acetylcholine (10 μmol/l) 
Dose–response curve with acetylcholine Dose–response curve with acetylcholine Dose–response curve with acetylcholine 
Incubate with PI3K inhibitor for 30 min No incubation, KH buffer alone Incubate with PI3K inhibitor for 30 min 
Dose–response curve with the GSE Dose–response curve with the GSE Dose–response curve with acetylcholine 
Step Ring 1 Ring 2 Ring 3 
Acetylcholine (10 μmol/l) Acetylcholine (10 μmol/l) Acetylcholine (10 μmol/l) 
Dose–response curve with acetylcholine Dose–response curve with acetylcholine Dose–response curve with acetylcholine 
Incubate with PI3K inhibitor for 30 min No incubation, KH buffer alone Incubate with PI3K inhibitor for 30 min 
Dose–response curve with the GSE Dose–response curve with the GSE Dose–response curve with acetylcholine 

Step 1 was done to establish responsiveness of the rings to a standard concentration of acetylcholine, step 2 provided a baseline dose–response curve to acetylcholine, and steps 3 and 4 established the effect of the inhibitors.

Ring 1 was used to examine the effect of the extract after incubation with the inhibitor, and ring 2 was used to demonstrate the response to the extract without prior exposure to the inhibitor. It also showed that prior exposure to acetylcholine did not influence the response to the extract (i.e. the maximal responses were similar). Ring 3 was used to demonstrate that the response to acetylcholine was unaltered with time (time control) and that exposure to the blocker did not affect the ability of eNOS to be activated by acetylcholine. This protocol avoided the application of the extract twice in succession to a ring.

Effect of wortmannin and LY294002 on phosphorylation of eNOS and Akt

HUVECs were grown in EGM-2 medium(Cambrex) with 10% (v/v) fetal bovine serum. Cells were grown to confluence (approx. 90%) and starved for 6 h in serum-free medium before the cells were treated with the GSE (10 μmol/l). Some wells were treated with LY294002 (30 μmol/l) or wortmannin (30 nmol/l) for 30 min before exposure to the GSE. Cells were incubated at 37 °C in a humidified atmosphere containing 5% CO2 for 10 min. The reaction was stopped by adding ice-cold PBS, washed twice with PBS and cell lysates were prepared in RIPA buffer [20 mmol/l Tris/HCl (pH 7.5), 150 mmol/l NaCl, 1 mmol/l EDTA, 1 mmol/l EGTA, 1% Nonidet P40, 1% sodium deoxycholate, 2.5 mmol/l sodium pyrophosphate, 1 mmol/l β-glycerophosphate, 1 mmol/l sodium orthovanadate and 1 μg/ml leupeptin]. Total proteins (30 μg) were separated by SDS/PAGE [7.5% (w/v) polyacrylamide gels] and were transferred electrophoretically on to nitrocellulose membranes (Amersham Biosciences). Membranes were blocked with blocking buffer containing 5% (w/v) non-fat milk in TBS-T (Tris-buffered saline containing 0.1% Tween 20) for 1 h. Phosphorylated Akt (at Ser473), phosphorylated eNOS (at Ser1177), Akt and eNOS were detected after the membranes were incubated with the respective primary antibodies {rabbit anti-[phospho-eNOS (Ser1177)], anti-eNOS, anti-[phospho-Akt (Ser473)] and anti-Akt; 1:1000 dilution; Cell Signaling Technology} overnight at 4 °C. Membranes were washed three times (10 min each) and incubated with the secondary antibody [HRP (horseradish peroxidase)-labelled anti-(rabbit IgG); 1:20000 dilution; Cell Signaling Technology] at room temperature (25 °C) for 60 min. Membranes were washed three times again (10 min each) and the specific protein bands were visualized using ECL® (Amersham Biosciences). All four proteins were detected on the same blot, and the membranes were washed with stripping buffer (Pierce Biotechnology) for 30 min in 37 °C before being incubated with the next primary antibody.

Effect of methylated GSE on EDR

An additional series of experiments were undertaken to study the effect of methylated GSE on rings of rabbit aorta. In each experiment, two rings were prepared as described above. Ring 1 was exposed to acetylcholine (10 μmol/l) and the GSE (100 μmol/l), and ring 2 was treated sequentially with acetylcholine (10 μmol/l), methylated GSE and the GSE (both 100 μmol/l). The methylation procedure is outlined in the Supplementary material available at http://www.clinsci.org/cs/114/cs1140331add.htm.

Statistical analysis

Group data are expressed as means±S.E.M. Comparisons between groups were compared using a paired Student's t test or ANOVA depending on the number of groups being examined. Dose–response curves were compared using repeated measures ANOVA. Data were analysed using Sigma Stat (version 3, 2003) statistical software. Statistical significance among treatments was determined as P<0.05.

RESULTS

Effect of the GSE on EDR

The GSE produced a dose-dependent relaxation of the aortic rings. The maximum relaxations observed were similar to those produced by acetylcholine (Figure 1). Removal of the endothelium abolished the responses evoked by acetylcholine and the GSE, confirming the obligatory role of the endothelium. Incubation with L-NAME, a competitive eNOS inhibitor, also abolished the relaxation responses to acetylcholine and the GSE. However, thereafter the rings remained responsive to SNP, which is a non-endothelium-dependent relaxant of smooth muscle (Figure 2). The maximum relaxations observed in the rings under the different conditions are summarized in Table 2. These results confirmed that the GSE causes EDR in rings of rabbit aorta.

Dose-dependent relaxation of aortic rings induced by acetylcholine and the GSE

Figure 1
Dose-dependent relaxation of aortic rings induced by acetylcholine and the GSE

Dose–response curves relating relaxation (as a percentage of contraction to 10 μmol/l noradrenaline) and concentration of the agonists in the organ bath. Dose–dependent relaxations were evoked by acetylcholine (◆) and the GSE (■). Values are means±S.E.M. (n=12).

Figure 1
Dose-dependent relaxation of aortic rings induced by acetylcholine and the GSE

Dose–response curves relating relaxation (as a percentage of contraction to 10 μmol/l noradrenaline) and concentration of the agonists in the organ bath. Dose–dependent relaxations were evoked by acetylcholine (◆) and the GSE (■). Values are means±S.E.M. (n=12).

Effect of removal of endothelium on the maximum relaxation

Figure 2
Effect of removal of endothelium on the maximum relaxation

(A) Responses induced by acetylcholine (Ac). (B) Responses induced by the GSE. Removal of endothelium abolished the responses elicited by acetylcholine and the GSE. The rings remained responsive to SNP (Nitro), which is a non-endothelium-dependent relaxant of smooth muscle. Values are means±S.E.M. (n=4). **P<0.01 compared with the treatment with the endothelium present and SNP with the endothelium removed. Endo.(+), endothelium present; Endo.(−), endothelium removed.

Figure 2
Effect of removal of endothelium on the maximum relaxation

(A) Responses induced by acetylcholine (Ac). (B) Responses induced by the GSE. Removal of endothelium abolished the responses elicited by acetylcholine and the GSE. The rings remained responsive to SNP (Nitro), which is a non-endothelium-dependent relaxant of smooth muscle. Values are means±S.E.M. (n=4). **P<0.01 compared with the treatment with the endothelium present and SNP with the endothelium removed. Endo.(+), endothelium present; Endo.(−), endothelium removed.

Table 2
Summary of the maximum relaxations observed in the aortic rings

Values are means±S.E.M. of maximum percentage relaxation evoked by different treatments (n=6). Values with different superscripts are significantly different, as determined by ANOVA (P<0.05).

Agent Maximum relaxation (%) 
Acetylcholine (10 μmol/l) 64.2±1.5a 
GSE (100 μmol/l) 71.9±1.0b 
L-NAME (1 mmol/l)+GSE (100 μmol/l) 6.3±1.0c 
SNP (10 μmol/l) 81.4±1.6d 
Agent Maximum relaxation (%) 
Acetylcholine (10 μmol/l) 64.2±1.5a 
GSE (100 μmol/l) 71.9±1.0b 
L-NAME (1 mmol/l)+GSE (100 μmol/l) 6.3±1.0c 
SNP (10 μmol/l) 81.4±1.6d 

Effect of inhibitors of the PI3K/Akt pathway

Incubation of aortic rings which had been previously shown to be responsive to acetylcholine with wortmannin or LY294002 significantly attenuated the relaxation induced by the GSE. The responses evoked by acetylcholine were unaffected. The sequence of treatments described in the Materials and methods section were used in these experiments. An example of an experiment with each blocker is shown in Figure 3, showing that the GSE-induced dose-dependent EDR was significantly attenuated in rings exposed previously to a PI3K inhibitor. The responses induced by the highest concentration of the GSE in these experiments are shown in Figure 4. It was also confirmed that acetylcholine-induced EDR was unaffected by PI3K inhibitors. Therefore it is apparent that prior exposure to a PI3K inhibitor attenuated EDR evoked by the GSE, suggesting that EDR induced by the GSE is mediated by the activation of the PI3K/Akt pathway.

Dose–response curves evoked by the GSE after incubation with (A) wortmannin and (B) LY294002

Figure 3
Dose–response curves evoked by the GSE after incubation with (A) wortmannin and (B) LY294002

Treatment with the agonists and inhibitors are described in the Materials and methods section and Table 1. (A) Initial response to acetylcholine [curve (i)], response to the GSE after incubation with wortmannin (30 nmol/l) for 30 min [curve (ii)], initial response to acetylcholine [curve (iii)], and response to GSE without prior incubation with wortmannin [curve (iv)]. Dose–response curves (i) and (ii) were generated from ring 1, and curves (iii) and (iv) were generated from ring 2. (B) Initial response to acetylcholine [curve (i)], response to the GSE after incubation with LY294002 (30 μmol/l) for 30 min [curve (ii)], initial response to acetylcholine [curve (iii)], and response to the GSE without incubation with LY294002 [curve (iv)]. Dose–response curves (i) and (ii) were generated from ring 1, and curves (iii) and (iv) were generated from ring 2. Curve (ii) is different from other three in both (A) and (B). Values are means±S.E.M. (n=4) in both A and B. Results from ring 3 are not shown.

Figure 3
Dose–response curves evoked by the GSE after incubation with (A) wortmannin and (B) LY294002

Treatment with the agonists and inhibitors are described in the Materials and methods section and Table 1. (A) Initial response to acetylcholine [curve (i)], response to the GSE after incubation with wortmannin (30 nmol/l) for 30 min [curve (ii)], initial response to acetylcholine [curve (iii)], and response to GSE without prior incubation with wortmannin [curve (iv)]. Dose–response curves (i) and (ii) were generated from ring 1, and curves (iii) and (iv) were generated from ring 2. (B) Initial response to acetylcholine [curve (i)], response to the GSE after incubation with LY294002 (30 μmol/l) for 30 min [curve (ii)], initial response to acetylcholine [curve (iii)], and response to the GSE without incubation with LY294002 [curve (iv)]. Dose–response curves (i) and (ii) were generated from ring 1, and curves (iii) and (iv) were generated from ring 2. Curve (ii) is different from other three in both (A) and (B). Values are means±S.E.M. (n=4) in both A and B. Results from ring 3 are not shown.

Effect of PI3K inhibitors on the maximum relaxation produced by GSE

Figure 4
Effect of PI3K inhibitors on the maximum relaxation produced by GSE

(A) All three rings responded to acetylcholine initially. Ring 1, which was incubated with wortmannin (30 nmol/l for 30 min) and tested with the GSE, had a significantly attenuated relaxation (**P<0.01 compared with GSE alone). Ring 2, which was not incubated with wortmannin, had a similar relaxation with the GSE. Ring 3, which was also incubated with wortmannin, had no significant change in the responses to acetylcholine. (B) Similar findings were observed with LY294002 (30 μmol/l) for the effect induced by the GSE. The relaxation evoked by the GSE was significantly decreased (**P<0.01 compared with GSE alone). All values are means±S.E.M. (n=4).

Figure 4
Effect of PI3K inhibitors on the maximum relaxation produced by GSE

(A) All three rings responded to acetylcholine initially. Ring 1, which was incubated with wortmannin (30 nmol/l for 30 min) and tested with the GSE, had a significantly attenuated relaxation (**P<0.01 compared with GSE alone). Ring 2, which was not incubated with wortmannin, had a similar relaxation with the GSE. Ring 3, which was also incubated with wortmannin, had no significant change in the responses to acetylcholine. (B) Similar findings were observed with LY294002 (30 μmol/l) for the effect induced by the GSE. The relaxation evoked by the GSE was significantly decreased (**P<0.01 compared with GSE alone). All values are means±S.E.M. (n=4).

GSE induces the phosphorylation of Akt and eNOS in HUVECs

EDR is caused by NO produced by the phosphorylation of eNOS. Therefore we investigated whether the GSE induced the phosphorylation of Akt (on Ser473) and eNOS (on Ser1177) in vitro in HUVECs. The GSE-induced phosphorylation of Akt and eNOS was shown by immunoblotting. Prior exposure to the PI3K inhibitor LY294002 abolished the phosphorylation of Akt and eNOS in HUVECs (Figure 5). These results suggested that the GSE phosphorylates eNOS through a PI3K/Akt pathway.

Effect of the GSE on eNOS and Akt phosphorylation

Figure 5
Effect of the GSE on eNOS and Akt phosphorylation

(A) HUVECs were treated with vehicle (control; lane 1), GSE (lane 2), LY294002 alone (lane 3) and GSE plus LY294002 (LY+GSE). In the controls, low levels of phosphorylated Akt and eNOS were observed, which were increased after incubation with GSE (lane 2). LY294002 alone (lane 3) had no effect on phosphorylation compared with the controls, but inhibited GSE-induced phosphorylation of Akt and eNOS when incubated with GSE (lane 4). (B and C) Quantification of the immunoblots of phosphorylated Akt (p-Akt) (B) and phosphorylated eNOS (p-eNOS) (C) using densitometry (n=4). The results are ratios of the phosphorylated and non-phosphorylated form of each enzyme. *P<0.05 compared with the control.

Figure 5
Effect of the GSE on eNOS and Akt phosphorylation

(A) HUVECs were treated with vehicle (control; lane 1), GSE (lane 2), LY294002 alone (lane 3) and GSE plus LY294002 (LY+GSE). In the controls, low levels of phosphorylated Akt and eNOS were observed, which were increased after incubation with GSE (lane 2). LY294002 alone (lane 3) had no effect on phosphorylation compared with the controls, but inhibited GSE-induced phosphorylation of Akt and eNOS when incubated with GSE (lane 4). (B and C) Quantification of the immunoblots of phosphorylated Akt (p-Akt) (B) and phosphorylated eNOS (p-eNOS) (C) using densitometry (n=4). The results are ratios of the phosphorylated and non-phosphorylated form of each enzyme. *P<0.05 compared with the control.

Effect of methylation of the GSE on EDR

It was found that methylated GSE failed to produce an ERD in the aortic rings. Subsequent exposure to the GSE (100 μmol/l) produced a significant relaxation, which was similar to that evoked by acetylcholine (10 μmol/l) (Figure 6).

Effect of methylation of the GSE on EDR

Figure 6
Effect of methylation of the GSE on EDR

The responses induced by acetylcholine (Ac), the GSE and methylated GSE (Met. GSE) are shown. Ring 1 had acetylcholine and the GSE applied in sequence, and Ring 2 had acetylcholine, methylated GSE and the GSE applied in sequence. No relaxation was evoked by methylated GSE (n=4).

Figure 6
Effect of methylation of the GSE on EDR

The responses induced by acetylcholine (Ac), the GSE and methylated GSE (Met. GSE) are shown. Ring 1 had acetylcholine and the GSE applied in sequence, and Ring 2 had acetylcholine, methylated GSE and the GSE applied in sequence. No relaxation was evoked by methylated GSE (n=4).

DISCUSSION

The present study has shown that the GSE used produced EDR in the rabbit aorta, which was significantly attenuated by prior incubation with the PI3K inhibitors wortmannin and LY294002. In these respects, the response is similar to that evoked by other derivatives of grapes which have been investigated extensively [12,13,18]. The novel aspects of the present study are the following: (i) the concurrent phosphorylation of both Akt and eNOS; (ii) modifying the antioxidant activity of the extract by methylation removed the ability to cause EDR; and (iii) we have used a compound that is very high in phenols (>90%), unlike GSEs used in other studies. Overall, the GSE used in the present study has been analysed in much greater detail than ones used by other investigators (see Supplementary material available at http://www.clinsci.org/cs/114/cs1140331add.htm).

Potential mechanism of action

Akt is a serine/threonine protein kinase that is recruited to the (endothelial) membrane by binding to PI3K-produced phosphoinositides. At the membrane, Akt is phosphorylated and activates eNOS (by phosphorylation at Ser1177 in humans), leading to the production of NO [17]. It has also been shown that the production of NO in response to fluid shear stress is controlled by Akt-dependent phosphorylation of eNOS [19]. However, recent studies performed in cell culture have established that polyphenolic compounds in red wine also affect the level of phosphorylation of Akt in a PI3K-dependent manner, which in turn phosphorylates eNOS, resulting in an increased formation of NO [13]. PI3K, which is a redox-sensitive protein kinase, appears to be activated by the redox sensitivity of polyphenols, leading to the production of NO. It has also been shown that, in endothelial cells, phosphorylation induced by polyphenols occurs on Ser1177 of eNOS and dephosphorylation at Thr495 within a few minutes of exposure. These changes in the phosphorylation level of eNOS were maintained for at least 30 min.

The GSE used in the present study is a relatively ‘pure’ one, with phenolic compounds forming 93% of its constituents. These compounds are mostly proanthocyanidins which occur as mixtures of oligomers and polymers of catechin and epicatechin (see Supplementary material available at http://www.clinsci.org/cs/114/cs1140331add.htm). Some of the larger polymeric compounds have the capacity to complex with proteins to form tannins. Plant tannins are divided into hydrolysable and condensed forms. The former contains gallic acid and a dimeric condensation product (hexahydroxydiphenic acid) that is esterified to a polyol such as glucose. The condensed tannins are high-molecular-mass oligomers and polymers of the monomeric unit flavanol-3-ol and their gallic acid esters. The monomeric units themselves are formed through oxidative condensation by carbon–carbon bonds, normally between carbon-4 of the heterocycle carbon ring and carbon-8 of the adjacent units (Figure 7). The GSE used in the present study consisted mainly of dimers and trimers (see Supplementary material available at http://www.clinsci.org/cs/114/cs1140331add.htm) and was devoid of gallic acid residues.

Basic structure of flavanoids

Figure 7
Basic structure of flavanoids

Proanthocyanidins are polymeric phenolic compounds characterized by a flavanoid with the basic three-ring structure.

Figure 7
Basic structure of flavanoids

Proanthocyanidins are polymeric phenolic compounds characterized by a flavanoid with the basic three-ring structure.

Polyphenolic compounds are generally considered to be antioxidants [20,21]; however, under certain circumstances, they have pro-oxidant properties attributable to the hydroxy groups in the phenolic rings. For instance, treatment of cell cultures with polyphenolic compounds significantly increased the production of ROS (reactive oxygen species) such as H2O2 [22,23]. It has been proposed that H2O2 is generated by auto-oxidation of hydroxy groups present in phenolic compounds (see Figure 1 in [23]), which subsequently activate PI3K [13,23]. Ndaye et al. [13] have shown that removal of hydroxyl radicals derived from H2O2 by enhancing endogenous SOD (superoxide dismutase) abolished the EDR produced by GSEs. In the present study, we have demonstrated that the removal of the hydroxy groups from the GSE by prior methylation also abolished EDR, thus supporting the important role of the hydroxy groups in producing EDR.

Conclusions

In the present study, we provide evidence to suggest that EDR evoked by the GSE is mediated by the activation of the PI3K/Akt signalling pathway, resulting in the phosphorylation of eNOS. Previous studies have suggested that GSEs activate PI3K and downstream signalling via Akt and activate eNOS through a redox-sensitive mechanism [13]. Furthermore, we found that removal of the antioxidant activity from the GSE by methylation of the hydroxy groups abolished the EDR induced by GSEs. These results support the suggestion that ROS produced by GSEs can activate eNOS to produce NO and cause vasodilation.

Abbreviations

     
  • EDR

    endothelium-dependent relaxation

  •  
  • GSE

    grape seed extract

  •  
  • HUVEC

    human umbilical vein endothelial cell

  •  
  • KH buffer

    Krebs–Henseleit buffer

  •  
  • L-NAME

    NG-nitro-L-arginine methyl ester

  •  
  • NOS

    NO synthase

  •  
  • eNOS

    endothelial NOS

  •  
  • PI3K

    phosphoinositide 3-kinase

  •  
  • ROS

    reactive oxygen species

  •  
  • SNP

    sodium nitroprusside

The study was supported in part by Polyphenolics Inc., who also provided the GSE.

References

References
1
Aldini
G.
Carini
M.
Piccoli
A.
Rossoni
G.
Facino
R. M.
Procyanidins from grape seeds protect endothelial cells from peroxynitrite damage and enhance endothelium-dependent relaxation in human artery: new evidences for cardio-protection
Life Sci.
2003
, vol. 
73
 (pg. 
2883
-
2898
)
2
Cishek
M. B.
Galloway
M. T.
Karim
M.
German
J. B.
Kappagoda
C. T.
Effect of red wine on endothelium-dependent relaxation in rabbits
Clin. Sci.
1997
, vol. 
93
 (pg. 
507
-
511
)
3
Stoclet
J. C.
Chataigneau
T.
Ndiaye
M.
, et al. 
Vascular protection by dietary polyphenols
Eur. J. Pharmacol.
2004
, vol. 
500
 (pg. 
299
-
313
)
4
Plotnick
G. D.
Corretti
M. C.
Vogel
R. A.
Hesslink
R.
Jr
Wise
J. A.
Effect of supplemental phytonutrients on impairment of the flow-mediated brachial artery vasoactivity after a single high-fat meal
J. Am. Coll. Cardiol.
2003
, vol. 
41
 (pg. 
1744
-
1749
)
5
de Lorgeril
M.
Salen
P.
Paillard
F.
Laporte
F.
Boucher
F.
de Leiris
J.
Mediterranean diet and the French paradox: two distinct biogeographic concepts for one consolidated scientific theory on the role of nutrition in coronary heart disease
Cardiovasc. Res.
2002
, vol. 
54
 (pg. 
503
-
515
)
6
Chawla
R.
Regular drinking might explain the French paradox
Br. Med. J.
2004
, vol. 
329
 pg. 
1308
 
7
Pastrana-Bonilla
E.
Akoh
C. C.
Sellappan
S.
Krewer
G.
Phenolic content and antioxidant capacity of muscadine grapes
J. Agric. Food Chem.
2003
, vol. 
51
 (pg. 
5497
-
5503
)
8
Fitzpatrick
D. F.
Fleming
R. C.
Bing
B.
Maggi
D. A.
O'Malley
R. M.
Isolation and characterization of endothelium-dependent vasorelaxing compounds from grape seeds
J. Agric. Food Chem.
2000
, vol. 
48
 (pg. 
6384
-
6390
)
9
Fitzpatrick
D. F.
Bing
B.
Maggi
D. A.
Fleming
R. C.
O'Malley
R. M.
Vasodilating procyanidins derived from grape seeds
Ann. N.Y. Acad. Sci.
2002
, vol. 
957
 (pg. 
78
-
89
)
10
Karim
M.
McCormick
K.
Kappagoda
C. T.
Effects of cocoa extracts on endothelium-dependent relaxation
J. Nutr.
2000
, vol. 
130
 (pg. 
2105S
-
2108S
)
11
Wallerath
T.
Deckert
G.
Ternes
T.
, et al. 
Resveratrol, a polyphenolic phytoalexin present in red wine, enhances expression and activity of endothelial nitric oxide synthase
Circulation
2002
, vol. 
106
 (pg. 
1652
-
1658
)
12
Anselm
E.
Chataigneau
M.
Ndiaye
M.
Chataigneau
T.
Schini-Kerth
V. B.
Grape juice causes endothelium-dependent relaxation via a redox-sensitive Src- and Akt-dependent activation of eNOS
Cardiovasc. Res.
2007
, vol. 
73
 (pg. 
404
-
413
)
13
Ndiaye
M.
Chataigneau
M.
Lobysheva
I.
Chataigneau
T.
Schini-Kerth
V. B.
Red wine polyphenol-induced, endothelium-dependent NO-mediated relaxation is due to the redox-sensitive PI3-kinase/Akt-dependent phosphorylation of endothelial NO-synthase in the isolated porcine coronary artery
FASEB J.
2004
, vol. 
19
 (pg. 
455
-
457
)
14
Edirisinghe
I.
Siva
B.
Randolph
J.
Steinberg
F.
Skepper
J.
Kappagoda
T.
A diet rich in saturated fat impairs endothelium dependent relaxation (EDR) in guinea pig (GP)
FASEB J.
2006
, vol. 
20
 pg. 
A287
 
15
Siva
B.
Edirisinghe
I.
Randolph
J.
Steinberg
F.
Kappagoda
T.
Effect of a polyphenoics extracts of grape seeds (GSE) on blood pressure (BP) in patients with the metabolic syndrome (MetS)
FASEB J.
2006
, vol. 
20
 pg. 
A305
 
16
Edirisinghe
I.
McCormick-Hallam
K.
Kappagoda
C. T.
Effect of fatty acids on endothelium-dependent relaxation in the rabbit aorta
Clin. Sci.
2006
, vol. 
111
 (pg. 
145
-
151
)
17
Kawasaki
K.
Smith
R. S.
Hsieh
C. M.
Chao
J.
Activation of the phosphatidylinositol 3-kinase/protein kinase Akt pathway mediates nitric oxide-induced endothelial cell migration and angiogenesis
Mol. Cell. Biol.
2003
, vol. 
23
 (pg. 
5726
-
5737
)
18
Ndiaye
M.
Chataigneau
T.
Andriantsitohaina
R.
Stoclet
J. C.
Schini-Kerth
V. B.
Red wine polyphenols cause endothelium-dependent EDHF-mediated relaxations in porcine coronary arteries via a redox-sensitive mechanism
Biochem. Biophys. Res. Commun.
2003
, vol. 
310
 (pg. 
371
-
377
)
19
Jin
Z. G.
Ueba
H.
Tanimoto
T.
Lungu
A. O.
Frame
M. D.
Berk
B. C.
Ligand-independent activation of vascular endothelial growth factor receptor 2 by fluid shear stress regulates activation of endothelial nitric oxide synthase
Circ. Res
2003
, vol. 
93
 (pg. 
354
-
363
)
20
Ariga
T.
The antioxidative function, preventive action on disease and utilization of proanthocyanidins
Biofactors
2004
, vol. 
21
 (pg. 
197
-
201
)
21
Koga
T.
Moro
K.
Nakamori
K.
, et al. 
Increase of antioxidative potential of rat plasma by oral administration of proanthocyanidin-rich extract from grape seeds
J. Agric. Food Chem.
1999
, vol. 
47
 (pg. 
1892
-
1897
)
22
Yang
G. Y.
Liao
J.
Li
C.
, et al. 
Effect of black and green tea polyphenols on c-jun phosphorylation and H2O2 production in transformed and non-transformed human bronchial cell lines: possible mechanisms of cell growth inhibition and apoptosis induction
Carcinogenesis
2000
, vol. 
21
 (pg. 
2035
-
2039
)
23
Akagawa
M.
Shigemitsu
T.
Suyama
K.
Production of hydrogen peroxide by polyphenols and polyphenol-rich beverages under quasi-physiological conditions
Biosci. Biotechnol. Biochem.
2003
, vol. 
67
 (pg. 
2632
-
2640
)

Supplementary data