Inhibitory effect of Fruitow on platelet function: a randomized placebo-controlled trial in elderly subjects

Background: The elderly has a high risk of cardiovascular disease, which is often accompanied by platelet hyperactivity. Tomato extracts can inhibit platelet activation and have benecial health effects. We aimed to investigate the effect of Fruitow® (FF), a water-soluble tomato extract, on platelet function in elderly participants. Methods: This randomized group study was conducted with people over 50 years old. The participants were randomly divided into four groups: placebo (150 mg/day), FF (150 mg/day), acetylsalicylic acid (ASA; 100 mg/day), and FF (150 mg/day) + ASA (100 mg/day). These groups received the respective supplements after dinner daily for 7 days. Fasting blood was collected from the participants on days 0 and 8 to analyze platelet aggregation and the content of thromboxane B 2 (TXB 2 ), 6-keto-prostaglandin F 1α , and platelet factor 4 (PF4). Results: One hundred ninety elderly participants were recruited and completed this clinical trial. The results showed that the FF intervention for 7 days decreased platelet aggregation by 7.7% in adenosine diphosphate-stimulated platelets, which was similar to the effect of ASA, which decreased platelet aggregation by 9.4%. FF reduced platelet aggregation by 10.2% in collagen-stimulated platelets, and ASA reduced platelet aggregation by 38.3% in collagen-activated platelets. This suggests that ASA exerts a stronger inhibitory effect than FF on collagen-stimulated platelet aggregation. The combination of FF + ASA did not exert a synergistic inhibitory effect on platelet aggregation. Treatment with FF signicantly decreased plasma TXB2, 6-keto-PGF1α, and PF4 levels, and its effects were similar to ASA. Conclusion: Fruitow® suppressed platelet aggregation and decreased TXB2, 6-keto-PGF1α, and PF4 levels in elderly participants. These ndings indicate that FF might reduce the risk of thrombosis in cardiovascular diseases.


Introduction
Cardiovascular disease is associated with the highest rates of morbidity and mortality worldwide [1].
Elderly individuals usually have the highest risk of cardiovascular events [2,3]. Age-related vascular in ammation and platelet hyperactivity are linked to cardiovascular events [4,5]. Platelet hyperactivity has been associated with several chronic diseases, including hypertension, diabetes, and hyperlipidemia [6][7][8]. Platelet activation is involved in thrombosis and promotes chronic in ammation and vascular injury in atherosclerosis. Antiplatelet therapy is considered to effectively reduce the risk of cardiovascular events in these patients [9]. Acetylsalicylic acid (ASA; aspirin) is a classic antiplatelet medication that has been used clinically for decades, but its side effects are associated with a risk of bleeding and gastrointestinal damage after long-term intake [10][11][12]. Therefore, the discovery of new antiplatelet therapies with fewer side effects is still necessary.
Nutritional interventions may be promising treatment alternatives. Some foods have obvious health bene ts, such as reducing the incidence of cardiovascular disease in countries with a Mediterranean diet [13][14][15][16]. Studies have shown that tomatoes exert protective effects on the cardiovascular system, and the active ingredients of tomatoes can inhibit platelet function [17,18]. FF is a water-soluble tomato extract that mainly contains avone, adenosine, and chlorogenic acid [19,20]. FF was rst product in Europe to obtain an approved, proprietary health claim under Article 13 (5)

of the European Health Claims
Regulation 1924/2006 on nutrition and health claims made on foods [17]. The ingredients of FF were safe for human consumption [21][22][23]. O'Kennedy et. reported that FF have similar effect on inhibition of platelet function compared with aspirin, but the effect is weaker than aspirin. A water-soluble tomato extract was also reported to inhibit angiotensin-converting enzyme and relax the vascular endothelium [24].
The elderly has a high risk of cardiovascular disease that is associated with platelet hyperactivity and the dysfunction of endothelial cells [25]. Whether the FF intervention can improve platelet function in the elderly requires further investigation. The present study used randomized grouping and placebo-and ASA-controlled assessments in elderly participants to evaluate the effects of FF on platelet function.

Participant
In the current trial, we included the participants with the age over 50 years old, with no medical history of serious disease, hemostatic disorders, or intolerance to ASA. The platelet aggregation rate in these participants reached higher than 50% in ADP-and collagen-stimulated platelets. The exclusion criteria were blood pressure ≥ 160/100 mmHg, platelet count < 100 × 10 9 /L, prothrombin time outside the normal reference range, the chronic use of medications that may affect platelet function and dietary supplements (e.g., sh oil and evening primrose oil), allergy to tomatoes or their ingredients, intolerance or allergy to ASA, contraindications for ASA, and positive fecal occult blood test. Moreover, according to the researchers' judgment, individuals with serious hematological diseases, digestive diseases, organic diseases of the heart, liver, or kidneys, severe liver or kidney dysfunction, or metabolic, endocrine, or nervous system diseases were also excluded. Eligible participants were recruited from February to December 2019. Participants undergone clinical examination on days 0 and 8. A owchart of the recruitment procedure and disposition of the participants is shown in Fig. 1.

Sample size calculation
On the basis of the sample size formula suggested for randomized clinical trials and by considering type I error of 5% (α = 0.05), type II error of 10% (β = 0.10, power =90%), and allowing for a 20 % drop-out rate, we determined a sample size of 40 individuals in each group. To get a more con dent result, we intended to recruit 100 participants for each group. However, we temporarily terminated the trial after the recruitment of 50 participants in each group because of COVID-19 crisis.

Ethical consideration
The present study was conducted according to the principles of the Declaration of Helsinki. All procedures that involved human participants were approved by the Ethics Committee of Beijing Hospital (no. 2018BJYYEC-195-02). Written informed consent was obtained from all of the participants. The study was retrospectively registered at the Chinese Clinical Trial Registry (www.clinicaltrials.gov; registration no. ChiCTR2000034647).

Study supplements and materials
Fruit ow®(FF) and placebo were provided by By-Health Co., Ltd (Zhuhai, Guangdong, China) in tablet formats. Each FF tablet contained 150 mg of water-soluble tomato extract. The main biological ingredients of FF are avone, adenosine, and chlorogenic acid (total > 8 mg/g). Each aspirin tablet contained 100 mg ASA (Bayer, purchased from Beijing Hospital, Beijing, China). The packages of FF and placebo tablets were same in the terms of color, appearance, and taste and were given to the participants at the study baseline.

Intervention
All study activities were carried out in the Beijing Hospital, Beijing, China. The study followed a randomized placebo-controlled trial. The participants were randomly allocated in a manner of 1:1:1:1 to four groups: placebo (150 mg/day), FF (150 mg/day), ASA (100 mg/day), and FF (150 mg/day) + ASA (100 mg/day). Simple randomization sequence was created using SPSS 25.0 (Armonk, NY, USA) statistical software. The participants in the placebo group did not know the aim of the study. The participants were asked to take their respective intervention after dinner daily for 7 days. The participants were asked to record their intervention form every day and to ensure their adherence in the present study. The duration of the interventions was based on O'Kennedy, who showed that the FF intervention for 7 days is su cient to evaluate its effectiveness [17]. The primary outcomes were post-supplement changes in platelet aggregation response to agonist, changes in TXB2, 6-keto-prostaglandin F 1α (6-keto-PGF 1α ) and platelet factor 4 (PF4) generation by platelets. Secondary outcomes included post-supplement changes in plasma clotting times and fecal occult blood test. Outcome variables were assessed at the study baseline and end of the trial (days 0 and 8). Fasting blood from participants was collected by a second researcher who was unaware of assignment of the groups.

Platelet aggregation rate measurement
Venous blood was collected from the participants in the placebo, FF, ASA, and FF + ASA groups preintervention and post-intervention. The blood samples were immediately mixed with 3.2% sodium citrate.
The blood samples were centrifuged at 200 ´ g for 10 min to obtain platelet-rich plasma and then centrifuged again at 1500 ´ g for 10 min to obtain platelet-poor plasma. Platelet-rich plasma (300 μl) was used to analyze platelet aggregation using a Chrono-Log aggregometer. Platelet-rich plasma was set in the platelet aggregation assay channel with stirring for 1 min, and then ADP (2 μM) or collagen (2 μg/ml) was added to induce platelet aggregation. The rate of platelet aggregation was recorded using a Chrono-Log aggregometer.
2.7. Assessment of TXB2, 6-keto-PGF1α and PF4 Venous blood was collected from the participants on days 0 and 8. After centrifugation at 1500 ´ g for 15 min, plasma was collected and stored at -80°C. The levels of TXB 2 , 6-keto-PGF1α, and PF4 were determined using ELISA kits (Abcam, Boston, MA, USA) according to the manufacturer's instructions.

Biochemical determination
Biochemical testing was performed using a LABOSPECT 008 AS automatic biochemical analyzer (Hitachi, Tokyo, Japan).

Assessment of other variables
Required information on age, gender, having a history of other diseases, medication, body mass index (BMI), and supplement use was collected using a researcher-made questionnaire. BMI was calculated as weight in kilograms divided by height in square meters (kg/m 2 ). Height was measured in a standing position without shoes. Weight was determined with minimal clothing and without shoes by an analog scale.

Statistical analysis
The data are expressed as mean ± SD. All of the variables were tested for a normal distribution using the Kolmogorov-Smirnov test. Two-tailed paired Student's t-test was used to compare platelet aggregation and TXB 2 , 6-keto-PGF1α, and PF4 levels between pre-and post-intervention. The Kruskal-Wallis H test was used to analyze differences between groups. The statistical analyses were performed using SPSS The baseline characteristics of the 190 participants are presented in Table 1. Blood pressure in participants with hypertension was not higher than 160/100 mmHg. All of the smokers smoked no more than 5 cigarettes per day. A small number of participants had hypertension and diabetes and were taking related medications, but the platelet aggregation rate in these participants was higher than 50%, suggesting that platelet function in these participants was unaffected by their medications. No signi cant difference was found between groups in the baseline characteristics.
Plasma lipid levels, fasting plasma glucose (FPG), high-sensitivity C-reactive protein (HsCRP), and homocysteine (HCY) were evaluated pre-intervention and post-intervention. As shown in Table 2, no changes in plasma lipid levels, FPG, HsCRP, or HCY were found between pre-and post-intervention in any of the groups. Moreover, no signi cant differences were found in the biochemical testing indicators between groups.
Fecal occult blood and coagulation were evaluated pre-and post-intervention. As shown in Table 3, no changes in fecal occult blood were found between pre-and post-intervention in any of the groups. The coagulation tests indicated that prothrombin time, brinogen, activated partial thromboplastin time, and thrombin time were not different between pre-and post-intervention in any of the groups. None of these parameters were signi cantly different between groups or between pre-and post-intervention. Therefore, short-term (7 days) interventions with FF and ASA did not nd bleeding trend.

Effect of FF on ADP-and collagen-induced platelet aggregation
To determine the effect of FF on platelet aggregation in vivo, blood was collected from the participants in the placebo, FF, ASA, and FF + ASA groups. The platelet aggregation rate was compared between pre-and post-intervention for 7 days. Platelet aggregation was induced by ADP (2 mM) and collagen (2 mg/ml).
As shown in Fig. 2A, placebo treatment for 7 days did not alter the aggregation rate. The aggregation rate that was induced by ADP was 76.2% ± 16.6% and 76.2% ± 13.4% pre-and post-intervention, respectively. Treatment with FF (150 mg/day) for 7 days decreased the rate of ADP-induced platelet aggregation by 7.7%, and the aggregation rate was 73.7% ± 13.8% and 66.0% ± 16.3% pre-and post-intervention, respectively. A signi cant difference was found between pre-and post-intervention in the FF group (p < 0.01). A similar trend was found over 7 days in the ASA group, in which the rate of ADP-induced platelet aggregation decreased by 9.4%, and the aggregation rate was 74.3% ± 14.9% and 64.9% ± 18.9% pre-and post-intervention, respectively (p < 0.01). We then determined whether FF and ASA exert synergistic inhibitory effects on ADP-stimulated platelet aggregation. Treatment with the combination of FF + ASA for 7 days decreased the rate of ADP-induced platelet aggregation by 9.3% (72.8% ± 16.5% and 63.5% ± 19.7% pre-and post-intervention, respectively; p < 0.01). We analyzed differences in platelet aggregation between the placebo, FF, ASA, and FF + ASA groups. No difference was found between groups preintervention. Post-intervention, a difference was found between the FF, ASA, and FF + ASA groups and placebo group. Treatment with the combination of FF + ASA did not exert a synergistic inhibitory effect on ADP-induced platelet aggregation.
We also evaluated the effects of placebo, FF, ASA, and FF + ASA on collagen-induced platelet aggregation. As shown in Fig. 2B, the rate of platelet aggregation did not change between pre-and postintervention in the placebo group (80.4% ± 16.0% pre-intervention vs. 81.4% ± 13.3% post-intervention). Treatment with FF for 7 days decreased collagen-induced platelet aggregation by 10.2% (79.1% ± 12.4% pre-intervention vs. 68.9% ± 16.5% post-intervention; p < 0.001). Treatment with ASA decreased the rate of platelet aggregation by 39% (74.3% ± 14.9% pre-intervention vs. 35.3% ± 26.8% post-intervention; p < 0.001). Treatment with FF + ASA decreased the rate of platelet aggregation by 38.3% (75.3% ± 15.3% preintervention vs. 37.0% ± 29.9% post-intervention; p < 0.001). No differences were found between the four groups before treatment, whereas signi cant differences were found between the FF, ASA, and FF + ASA groups and the placebo group after treatment. Moreover, no difference in platelet aggregation was found between the ASA group and FF + ASA group. This suggests that FF did not enhance the inhibitory effect of ASA on collagen-induced platelet aggregation.

Effect of FF on TXB 2 and 6-keto-PGF1α levels
To investigate the effect of FF on TXB 2 and 6-keto-PGF 1α production, we measured plasma TXB 2 and 6keto-PGF 1α levels using ELISA kits. As shown in Fig. 3A, TXB 2 levels were not different between preintervention and post-intervention in the placebo group (799.5 ± 82.9 pg/ml vs. 835.5 ± 81.6 pg/ml, respectively; p > 0.05). In contrast, TXB 2 levels were signi cantly lower pre-intervention in the FF group after 7 days. TXB 2 content was 738.5 ± 70.4 pg/ml and 579.3 ± 37.1 pg/ml pre-and post-intervention, respectively (p < 0.01). TXB 2 levels also decreased post-intervention in the ASA group after 7 days (777.4 ± 93.1 pg/ml vs. 542.7 ± 55.6 pg/ml pre-and post-intervention, respectively; p < 0.01). Similarly, TXB 2 levels decreased post-intervention in the FF + ASA group after 7 days (704.1 ± 70.5 pg/ml vs. 531.6 ± 47.3 pg/ml pre-and post-intervention, respectively; p < 0.01). As shown in Fig. 3B, the analysis between groups showed no signi cant difference between groups pre-intervention. Post-intervention, the FF, ASA, and FF + ASA groups were signi cantly different from the placebo group. No signi cant difference was found between the FF, ASA, and FF + ASA groups.

Effect of FF on plasma PF4 levels
We investigated the effect of FF on PF4 levels, which is released from α-granules of platelets. Plasma PF4 levels were detected using an ELISA kit. As shown in Fig. 4A, PF4 levels did not signi cantly change in the placebo group between pre-and post-intervention. PF4 levels were 1012.2 ± 98.8 ng/ml and 984.7 ± 93.8 ng/ml in the placebo group pre-and post-intervention, respectively. Post-intervention, PF4 levels signi cantly decreased in the FF group. PF4 levels were 1028.0 ± 70.1 ng/ml and 656.9 ± 54.3 ng/ml preand post-intervention, respectively (p < 0.001). PF4 levels also decreased in the ASA group postintervention. Before and after treatment with ASA, PF4 levels were 963.8 ± 73.6 ng/ml and 658.8 ± 35.4 ng/ml, respectively (p < 0.001). PF4 levels were 1003.5 ± 84.0 ng/ml and 638.9 ± 46.1 ng/ml in the FF + ASA group pre-and post-intervention, respectively (p < 0.001). Before the intervention, PF4 levels were not signi cantly different between groups. After the intervention, signi cant differences were found between the FF, ASA, and FF + ASA groups and the placebo group.

Discussion
Nutritional interventions can provide bene cial protection for health. This strategy has been widely accepted in the eld of disease prevention. In the present study, we rst investigated the effects of FF in a sample of elderly individuals in China. We conducted a randomized, placebo-controlled trial to evaluate the inhibitory effect of FF on platelet function in aging individuals and compared the effects of FF with ASA and the combination of FF + ASA. Our results showed that FF (150 mg/day) reduced ADP-induced platelet aggregation after the intervention, and the effect was compared with ASA (100 mg/day). The ASA group exhibited more potent suppression of collagen-stimulated platelet aggregation, and the effectiveness of ASA was four-times stronger than FF. However, the results did not show that FF acted synergistically with ASA to inhibit platelet aggregation. This suggests that the mechanism of the inhibitory effect of FF on platelet aggregation might be distinct from ASA. Moreover, our results showed that treatment with 150 mg/day FF or 100 mg/day ASA for 7 days signi cantly reduced TXB 2 and 6-keto-PGF1α levels. TXB 2 is a stable product of the platelet aggregation mediator TXA 2 . 6-Keto-PGF 1α is a stable product of the endogenous cyclooxygenase metabolite prostaglandin I 2 . Both TXB 2 and 6-keto-PGF 1α are metabolites of arachidonic acid. TXB 2 levels positively correlate with cardiovascular disease and thrombosis, whereas 6-keto-PGF 1α plays a negative role in anti-platelet activation [26,27]. Aspirin inhibits platelet function by inhibiting the arachidonic acid-metabolizing enzyme cyclooxygenase-1. Interestingly, although FF and ASA had different abilities in inhibiting collagen-stimulated platelet aggregation, FF effectively reduced plasma TXB 2 and 6-keto-PGF1α levels post-intervention, and the effectiveness of FF was the same as ASA. This suggests that FF might reduce the risk of thrombosis by lowering TXB 2 levels, but the mechanism of action of FF on TXB 2 is still unclear. A previous study reported that FF had no cumulative effect on cyclooxygenase-1 in platelets [19].
PF4, also referred to as CXCL4, belongs to the chemokine family. It is stored in α-granules of platelets and is the most abundant protein in α-granules. PF4 is a pleiotropic in ammatory chemokine that has been implicated in various in ammatory diseases, including atherosclerosis [28]. Endothelial cell damage occurs in early stages of atherosclerosis. PF4 promotes the local vascular in ammatory response by recruiting in ammatory cells to damaged endothelial cells. Our previous study found that PF4 levels signi cantly increased in apolipoprotein E-de cient mice, which is a model of atherosclerosis [29]. The present results showed that PF4 levels were higher before the intervention in elderly individuals, and treatment with FF, ASA, and FF + ASA decreased PF4 levels. This indicates that FF may reduce PF4 levels to improve the in ammatory state in the elderly body.
Epidemiological studies have found that the Mediterranean diet is bene cial for reducing the risk of cardiovascular disease [30,31]. Tomatoes are the main vegetable component in the Mediterranean diet. A European study reported that a daily intake of 65 mg FF can also partially inhibit platelet aggregation in health participants [17]. The present results are consistent with this previous study. Tomato extracts include various bene cial ingredients, such as lycopene, which has been considered to protect against benign prostatic hyperplasia [32]. This suggests that a tomato extract can provide various bene cial biological effects.
One limitation of the present study was the relatively short intervention time (i.e., 7 days). The side effects of ASA may only appear only after months or longer. Unknown is whether long-term FF administration has undesirable side effects in humans.

Conclusion
The present study found that FF inhibited platelet aggregation and reduced TXB 2 , 6-keto-PGF 1a , and PF4 levels in elderly participants. The combination of FF + ASA did not affect coagulation. Overall, the present results suggest that FF may be a suitable nutritional intervention strategy to reduce the risk of cardiovascular disease.