Aβ1–40-Induced Platelet Adhesion Is Ameliorated by Rosmarinic Acid through Inhibition of NADPH Oxidase/PKC-δ/Integrin αIIbβ3 Signaling

In platelets, oxidative stress reportedly increases platelet adhesion to vessels, thus promoting the vascular pathology of various neurodegenerative diseases, including Alzheimer’s disease (AD). Recently, it has been shown that β-amyloid (Aβ) can increase oxidative stress in platelets; however, the underlying mechanism remains elusive. In the present study, we aimed to elucidate the signaling pathway of platelet adhesion induced by Aβ1–40, the major form of circulating Aβ, through Western blotting, immunofluorescence confocal microscopy, and fluorescence-activated cell sorting analysis. Additionally, we examined whether rosmarinic acid (RA), a natural polyphenol antioxidant, can modulate these processes. Our results show that Aβ1–40-induced platelet adhesion is mediated through NADPH oxidase/ROS/PKC-δ/integrin αIIbβ3 signaling, and these signaling pathways are significantly inhibited by RA. Collectively, these results suggest that RA may have beneficial effects on platelet-associated vascular pathology in AD.


Introduction
Accumulating evidence suggests a correlation between vascular pathology and Alzheimer's disease (AD) [1,2]. In numerous neuropathological studies, more than one-third of AD patients are accompanied by cerebrovascular lesions [3], and patients with vascular dementia also exhibit the hallmarks of AD, such as β-amyloid (Aβ) plaques and neurofibrillary tangles (NFTs) [4]. Several previous reports have suggested that vascular lesions induce Aβ deposition at the site of vascular damage, and elevated levels of circulating Aβ in the blood promote vascular lesions [5]. Additionally, various vascular risk factors such as atherosclerosis and hypercholesterolemia have been shown to increase the risk of AD [6].
Platelets are key cells that play a critical role in vascular pathology via thrombogenic activity, as well as by adhering to damaged vessels [1,7,8]. Recent studies have indicated a possible role of platelets in the pathology of vascular dementia and AD, given that platelets contain amyloid precursor proteins (APP) and α-, β-, and γ-secretases, which contribute to the production of circulating Aβ [9,10]. Abnormal platelet activity has also been reported in patients with AD [1]. In animal studies, Aβ injection enhanced platelet adhesion to injured blood vessels in a mouse model of carotid artery injury [5]. In platelets, reactive oxygen species (ROS), such as H 2 O 2 and O 2 − , regulate platelet functions such as platelet aggregation and adhesion [11,12]. Therefore, various antioxidants afford preventive effects on platelet activation and thrombosis [13][14][15][16]. Furthermore, several recent studies have reported that oxidative stress occurs early in the brain of patients with AD [17], and Aβ increases ROS levels in platelets, resulting in platelet aggregation [11]. Collectively, these results suggest that Aβ-induced ROS in platelets may play a role in the vascular pathology of AD [18,19].
Rosmarinic acid (RA) is one of the major compounds commonly found in species of the family Boraginaceae and the subfamily Nepetoideae (Lamiaceae), and it is known to exhibit various biological activities, including antioxidative and anti-inflammatory effects [20,21]. Previous studies have reported that RA has antioxidant and antiaggregating activities in platelets [22]. Additionally, RA has shown efficacy in vascular diseases, such as diabetes and hypertension, mediated via its antiplatelet activity [22,23]. However, there is little information about the effect of RA on Aβ-induced platelet activation. In the present study, we investigated the effect of RA on Aβ-induced platelet adhesion and integrin α IIb β 3 , a major adhesion molecule in platelets. We also investigated the underlying mechanism of RA in terms of nicotinamide adenine dinucleotide phosphate (NADPH) oxidase and protein kinase C (PKC), which are major signaling molecules involved in integrin activation in platelets.

Washed Platelet Preparation
Freshly collected platelet-rich plasma (PRP) samples from healthy volunteers were procured from the Korean Red Cross Center, a blood donation facility for research purposes. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of Ajou University (Project No. 202002-HM-EX-001). PRP was centrifuged at 1000× g for 10 min at 22 • C without breaking platelet pellets. The supernatant obtained was termed platelet-poor plasma. The obtained platelet pellet was washed twice with Tyrode's buffer (pH 7.4).

Platelet Adhesion to Fibronectin
To observe platelet adhesion, stimulated platelets were stained with 10 µM CMFDA for 30 min at 37 • C. After washing with PBS, stained platelets (5 × 10 6 cells/mL) were added to a glass-bottom dish coated with fibronectin (Sigma-Aldrich, St. Louis, MO, USA). After incubation for 1 h, nonadherent platelets were removed by washing with PBS. Platelets were captured in five fields per dish using a confocal microscope (Nikon, Japan). Adherent cells were calculated by expressing the areas of adherent platelets as a percentage of the total area. This experiment was repeated a total of three or five times.

Measurement of Filopodia Length and Spread Area in Platelet
For the measurement of the filopodial length and spread area of platelets, a glassbottom dish was coated with fibronectin (Sigma-Aldrich, St. Louis, MO, USA). After washing with PBS, treated platelets (5 × 10 6 cells/mL) were added to a glass-bottom dish. After incubation for 1 h, the image of live cells was acquired through differential interference contrast (DIC) imaging using a confocal microscope at 400-1000× magnification (Nikon, Japan). Filopodia, membrane protrusions supported by bundles, were randomly selected per each platelet. The filopodia length and spread area (excluding filopodia) were measured from 20 randomly selected fields with ImageJ software. This experiment was repeated a total four times.

Surface Expression of Integrins α IIb and β 3
In brief, the washed human platelets (2 × 10 8 cells) were activated with 10 µM Aβ 1-40 for 1 h at room temperature and then washed twice with Tyrode's buffer and HEPES buffer. Next, the cells were immediately fixed with 2% paraformaldehyde on ice for 10 min. The cells were then washed twice in 1 × PBS, followed by the addition of 5 µL of FITC anti-human CD41a (integrin α IIb ) or 5 µL of APC anti-human CD61 (integrin β 3 ) to each sample and incubation for 30 min at room temperature in the dark. Finally, the cells were washed with 1 × PBS, resuspended in FACS sheath fluid, and analyzed using an FACSAria III flow cytometer (BD Biosciences, San Jose, CA, USA). This experiment was repeated a total of seven times.

Measurement of Free-Radical-Scavenging Activity by DPPH Assay
The free-radical-scavenging activity of RA was determined in vitro using the 2,2diphenyl-1-picrylhydrazyl (DPPH; Sigma-Aldrich, St. Louis, MO, USA) assay as previously described. The protocol to assess DPPH radical scavenging activity was adapted from Brand-Williams et al. (1995), with minor changes [24]. Briefly, 100 µL aliquots of methanolic solutions containing different RA concentrations (0.1-30 µM) were added to 100 µL of a 250 µM methanolic DPPH solution. After 30 min, the absorbance was measured at 517 nm, and the percentage inhibition activity was calculated. The percentage (%) of DPPH free radical scavenging was calculated using the formula (A0 − A1)/A0 × 100, where A0 is the absorbance of the control, and A1 is the absorbance of the extract/standard. This experiment was repeated a total of three times.

ABTS Assay for Free-Radical-Scavenging Activity
The preformed radical monocation of 2,2'-azinobis(3-ethylbenzothiazoline-6-sulfonic acid (ABTS; Sigma-Aldrich, St. Louis, MO, USA) was generated as previously described [25]. The ABTS salt was weighed (19.3 mg) and dissolved in distilled water (5 mL). Then, 88 µL of K 2 S 2 O 8 solution (0.0378 g/mL) was added to the ABTS solution, and the mixture was left at room temperature for 12 h in the dark. To perform measurements, the ABTS solution was diluted with ethanol to an absorbance of 0.700 ± 0.05 at 734 nm. Subsequently, 270 µL of the free-radical solution was combined with 20 µL of RA. The absorbance was measured at 734 nm after incubation at room temperature for 30 min in the dark. Antioxidant activity, expressed as a percentage of inhibition, was calculated using a previously established equation. Additionally, the half-maximal inhibitory concentration (IC 50 ) was determined. This experiment was repeated a total of three times.

ROS Measurement
Briefly, washed human platelets (1 × 10 7 cells), pretreated with RA, Trolox, or VAS 2870 for 30 min were incubated with either Aβ 1-40 for 15 min, followed by incubation with 10 µM H 2 DCFDA or 10 µM DHE for 30 min. Fluorescence was immediately measured using an FACSAria III flow cytometer (BD Biosciences, San Jose, CA, USA) with an excitation wavelength of 488 nm and an emission wavelength of 530 nm. Gated cells (n = 10,000) were analyzed for each sample. This experiment was repeated a total of six times.

Measurement of NADPH Oxidase Activity
The activity of NADPH oxidase was determined in membrane fractions (50 µg of protein) incubated with 1 mM EGTA and 5 µM lucigenin in phosphate buffer (pH 7.0). The assay was initiated by adding 50 µM NADPH to the incubation mixture. Samples were immediately counted using a tabletop luminometer (Berthold Detection Systems FB Luminometer; Zylux Corp., Oak Ridge, Tennessee), with sampling performed every 6 s over a 5 min period; the fluorescence values were recorded for over 2 min of stable readings and averaged for each sample. This experiment was repeated a total of seven times.

Preparation of the PKC Membrane Fraction
Platelets were lysed using different lysis buffers, and lysates were extracted using different procedures. Briefly, the cells were incubated in lysis buffer A (20 mM Tris-HCl, pH 7.4, 250 mM sucrose, 1 mM EDTA, 0.1 mM NaF, 0.2 mM Na 3 VO 4 , 0.5 mM phenylmethylsulphonyl fluoride, 0.01 mM leupeptin, and 0.01 mg/mL aprotinin) for 30 min on ice and then centrifuged at 200,000× g in a Beckman Optima TL Ultracentrifuge (Beckman Coulter, Brea, CA, USA) at 4 • C for 30 min. The supernatants (cytosolic fractions, CFs) were removed, and the remaining pellets were resuspended in lysis buffer B (lysis buffer A containing 1% Triton X-100) and incubated on ice for 1 h. The suspension was then centrifuged as described above, and the supernatant (membrane fraction, MF) was obtained. This experiment was repeated a total of three times.

Statistical Analysis
All data are expressed as the mean ± standard error of mean (SEM). Two-tailed t-tests were performed to examine differences in continuous variables, overall and at each time point, investigated in different comparison groups. Differences were considered statistically significant at p < 0.05.

Antioxidant Activity of RA
We examined the in vitro antioxidant activity of RA using the DPPH and ABTS assays. As shown in Figure 3a,b, RA exhibited DPPH and ABTS radical-scavenging activities in a concentration-dependent manner. As shown in Figure 3c,

RA Inhibits Aβ1-40-Induced Platelet Activation in an NADPH Oxidase-Dependent Manner
We measured the NADPH oxidase activity to determine the source of Aβ peptideinduced ROS generation, as well as the effect of ROS downstream. Our results revealed

RA Inhibits Aβ1-40-Induced Integrin Activation via PKC-δ Activation in Platelets
The PKC family is an essential signaling mediator for platelet activation and aggregation [27]. Accordingly, we examined the potential role of the PKC family in RAmediated suppression of increased platelet adhesion by Aβ1-40 and assessed the expression

Discussion
In the present study, we, for the first time, demonstrated the involvement of NADPH oxidase/ROS/PKC-δ/integrin αIIbβ3 signaling in Aβ1-40-induced platelet adhesion. We

Discussion
In the present study, we for the first time demonstrated the involvement of NADPH oxidase/ROS/PKC-δ/integrin α IIb β 3 signaling in the mechanism of Aβ 1-40 -induced platelet adhesion. We further revealed that Aβ 1-40 -induced platelet adhesion is ameliorated by RA through inhibition of these signaling pathways.
The heterogeneous cleavage pattern of APP by βand γ-secretase results in the production of Aβ peptides of varying lengths [28]. Reportedly, Aβ 1-40 is the primary blood form of Aβ, contributing to vascular amyloid deposition in AD; Aβ 1-42 is the predominant form in neural plaques [29]. Aβ 1-40 accounts for more than 90% of Aβ produced from APP in the body and is the predominant type released from activated human platelets. Although Aβ 1-42 has been found to mediate platelet aggregation, the efficacy of Aβ 1-42 on vascular events is suggested to be far less than that mediated by Aβ 1-40 [30,31]. Despite numerous investigations examining the correlation between AD and vascular lesions, studies on Aβ 1-40 are relatively scarce when compared with those on Aβ 1-42 . Previously, we reported that the altered miRNA profile in platelets from patients with Alzheimer's pathology was similar to that of Aβ 1-40 -exposed platelets in vitro [32], suggesting that Aβ 1-40 -stimulated platelets could play an important role during the process of Alzheimer's pathology. Consistent with our previous report, the present study revealed that Aβ 1-40 increased integrin α IIb β 3 levels in human platelets, and that platelet adhesion to fibronectin was almost completely suppressed by a specific integrin α IIb β 3 inhibitor. Additionally, RA suppressed platelet adhesion and integrin α IIb β 3 activity in a concentration-dependent manner (Figures 1 and 2), indicating that the effect of RA on Aβ 1-40 -induced platelet adhesion may be mediated via integrin α IIb β 3 .
Numerous studies have reported that oxidative stress contributes to Aβ generation and NFT formation, suggesting a close association between amyloid plaques and ROS in the pathogenesis of AD [19,33,34]. Elevated ROS production reportedly increases the activity of βand γ-secretases, which leads to increased APP cleavage and Aβ generation. Furthermore, in patients with AD, lipid peroxidation markers, such as 4-hydroxynonenal and malondialdehyde, were found to be elevated in the peripheral tissues, possibly because of insufficient enzymatic/nonenzymatic antioxidants [35,36]. The present study revealed that Aβ 1-40 significantly increased platelet ROS levels, and the levels of O 2 − or H 2 O 2 began to increase rapidly at 5 min, peaking at 15 min after Aβ 1-40 exposure (Figure 3). RA significantly reduced elevated ROS levels in a concentration-dependent manner. Our results further revealed that Trolox, a powerful antioxidant, significantly inhibited both the Aβ 1-40 -induced increase in platelet adhesion and the integrin α IIb β 3 activation. These results suggest that Aβ 1-40 -induced ROS may activate platelet adhesion, consistent with the findings of a recent study indicating a potential role for platelets in the pathogenesis of AD [2,11,37]. Our results further indicate that the inhibitory effect of RA on Aβ 1-40 -induced platelet adhesion may occur through its antioxidant activity.
Enzyme pathways known to induce ROS generation include NADPH oxidase, myeloperoxidase, xanthine oxidase, or uncoupled nitric oxide synthase. In particular, NADPH oxidase, an enzyme complex composed of several subunits and a small GTPase Rac, has been reported to play a role in some neurodegenerative diseases, including dementia, via ROS-induced neuronal death [38,39]. Additionally, NADPH oxidase is suggested to be a major source of Aβ-induced ROS in hippocampal cells [38,40]. Furthermore, growing evidence suggests that the enzymatic activity of NADPH oxidases plays a significant role in promoting platelet function [11,16]. Among the seven NADPH oxidase isotypes, only NADPH oxidases 1/2 are expressed in human platelets and are closely related to platelet activity. However, the molecular mechanism underlying Aβ 1-40 -induced platelet activation remains poorly understood [14,41]. According to our findings, Aβ 1-40 increased NADPH oxidase activity, and the NADPH oxidase inhibitor VAS2870 inhibited Aβ 1-40 -induced platelet adhesion and integrin α IIb β 3 activity. These results indicate that NADPH oxidase plays an important role in Aβ 1-40 -induced platelet activity, especially in terms of integrin α IIb β 3 activity ( Figure 5). Although Aβ 1-40 -induced NADPH oxidase isotype-specific activity was not detected in this study, it is speculated that Aβ 1-40 activated NADPH oxidases 1/2, as they are the primary forms present in platelets. In the present study, the Aβ 1-40 -induced increase in H 2 O 2 and O 2 − levels was blocked by inhibiting NADPH oxidase activity and vice versa. Alternatively, Trolox decreased NADPH oxidase activity, probably owing to the rapid scavenging of generated O 2 − following Trolox pretreatment, as the NADPH oxidase activity was measured by detecting O 2 − using lucigenin. Various cascades regulate integrin α IIb β 3 levels in platelets, among which PKC is a well-known factor [42]. However, the crosstalk between Aβ 1-40 and PKC isotypes in platelets has not been previously elucidated. In the present study, we, for the first time, reported that, among several isotypes of PKC (-α, -βI, -βII, and -γ), the level of PKC-δ was remarkably increased by Aβ 1-40 . Additionally, Aβ 1-40 -induced PKC-δ activation was suppressed by RA, Trolox, or VAS2870 ( Figure 6). Furthermore, we observed that rottlerin, a PKC-δ specific inhibitor, significantly inhibited the Aβ 1-40 -induced increase in integrin α IIb β 3 activity and platelet adhesion; however, Gö6976, a nonspecific inhibitor of PKCα, -β, and-γ, did not demonstrate this effect. These results suggest that Aβ 1-40 -induced integrin α IIb β 3 activity could be regulated by PKC-δ, which produces a representative G-protein-coupled receptor signaling cascade. PKC is divided into isotypes depending on whether calcium ions mediate their regulation. PKC-α, -β, and -γ are regulated by calcium ions, whereas PKC-δ acts independently of calcium ions. Thus, it is suggested that Aβ 1-40 -induced integrin α IIb β 3 activity may be regulated independently of calcium ions.
Overall, the underlying mechanisms for the inhibitory effect of RA on Aβ 1-40 -induced platelet adhesion may involve inhibition of NADPH oxidase/ROS/PKC-δ/integrin α IIb β 3 signaling. Although reports have previously indicated a relationship between NADPH oxidase and integrin α IIb β 3 activity in platelets [7,41], this study is the first to reveal a link between NADPH oxidase/PKC-δ and Aβ 1-40 -induced integrin α IIb β 3 in the process of platelet adhesion. Our study further showed that RA inhibits all these signaling pathways, consequently inhibiting platelet adhesion. These results indicate that integrin α IIb β 3 and NADPH oxidase can be potential therapeutic targets for platelet-associated vascular pathology in AD. Unsurprisingly, however, the most serious side-effects commonly observed with integrin α IIb β 3 antagonists include bleeding and thrombocytopenia. Therefore, more specific integrin inhibitors involved in the pathological mechanism are needed. In addition, NADPH oxidase may exhibit different roles in other cells, thus highlighting the need for platelet-specific NADPH oxidase inhibitors. As for RA, it has been reported to attenuate the pathological function of integrin α IIb β 3 , and possess antithrombotic effect against Aβ [21,43,44]. Therefore, RA could be developed as a therapeutic agent for platelet-associated vascular pathology in AD.