Postmenopausal women aged 50 to 70 years with BMI of 25 – 40 kg/m2 were enrolled. Postmenopausal status was defined as a lack of menses for at least two years or at least six months with a follicle-stimulating hormone level of 23 – 116.3 mIU/mL. Other inclusion criteria were an overall body weight equal to or greater than 100 pounds, and agreement to comply with all study procedures. Exclusion criteria included BMI greater than 40 kg/m2 , blood pressure greater than or equal to 140/90 mm Hg, abnormal values from a lipid panel, complete blood count , or comprehensive metabolic panel , use of prescription medications other than thyroid, daily use of anti-coagulation agents such as aspirin and non-steroidal anti-inflammatory drugs, or use of dietary supplements other than a general formula of multivitamins/minerals that provided up to 100% of the recommended dietary allowances. Additional exclusion criteria were vegetable consumption greater than or equal to 3 cups/day, fruit consumption greater than or equal to 2 cups/day, fatty fish intake greater than or equal to 3 times/week, dark chocolate intake greater than or equal to 3 oz/day, coffee and/or tea intake greater than or equal to 3 cups/day, alcohol intake greater than 3 drinks/week. Women were also excluded if they followed a non-traditional diet , engaged in routine high-intensity exercise, self-reported diabetes, renal or liver disease, malabsorption or gastrointestinal diseases, cancer within the last five years, or heart disease, including cardiovascular events or stroke. After determining initial eligibility through telephone screening, participants were further screened at the laboratory in the morning after an overnight fast. After informed consent was obtained, black plastic plant pots anthropometric measurements were taken, including body weight, height, and waist circumference.

Blood pressure and resting heart rate were measured three times, five minutes apart, after 15 minutes of sitting quietly. Volunteers also completed a diet and health habits questionnaire . A fasting blood sample was collected for a CMP, a CBC, and a lipid panel . If participants reported menses occurring within two years prior to the telephone screening, FSH was measured. Volunteers were excluded if their low-density lipoprotein value was greater than or equal to 190 mg/dL, or for those with zero to one major cardiovascular risk factors apart from high LDL cholesterol if their LDL was greater than or equal to 160 mg/dL, for those with two major cardiovascular risk factors apart from elevated LDL cholesterol greater than or equal to 130 mg/dL, or for those with two major cardiovascular risk factors apart from high LDL cholesterol and a Framingham 10-year risk score of 10 to 20% . Study I was a single-arm, four-week trial . Baseline values were collected at study visit 1 , which then began a run-in period of two weeks during which no mangos were consumed. At SV1, baseline anthropometry, blood pressure, PAT, and blood was collected, and taken again two hours later. At the end of two weeks, study visit 2 began with baseline measures taken, followed by ingestion of 330 gm of pre-packaged, fresh, frozen Ataulfo mangos, and data were collected two hours later. Participants then returned home with a 14-day supply of pre-packaged mangos and instructed to consume 330 gm of mangos daily, with 165 g eaten before noon, and the other half consumed in the evening. Two weeks later, study visit 3 ensued, which followed the same protocol as SV2 . Water was allowed ad libitum during all study visits. Prior to each study visit, participants were instructed to refrain from strenuous exercise for 24 hours before arriving at the laboratory to reduce the potential impact on PAT measurements.

Two 3-day food records were collected, once between SV1 and SV2, and again between SV2 and SV3. The records were analyzed using the Food Processor software . Study II was based on the findings from study I. This single-armed trial design is shown in Figure 2. After an overnight fast, at SV1, anthropometry, blood pressure, heart rate, and blood samples were collected at baseline and at one-hour and two-hour time points. At least 48 hours later, at SV2, baseline measures were taken, followed by ingestion of 330 gm of pre-packaged, fresh, frozen Ataulfo mangos, and data were collected 1h and 2h after intake. After at least two days, at SV3, baseline measures were taken, followed by ingestion of 113 g of white bread, which contained calories and carbohydrates similar to those found in 330 gm of mangos, and data were collected 1h and 2h after ingestion. The inclusion and exclusion criteria were the same for both study I and II. Participants were instructed to refrain from consuming additional mangos before SV1 and throughout their enrollment. Procedures were performed at the same time of the day to minimize circadian effects. The screening and interventions were conducted at the UC Davis Ragle Human Nutrition Research Center. The UC Davis Institutional Review Board approved the protocol, and the study was registered at ClinicalTrials.gov . Microvascular function was assessed by PAT . After resting in a supine position for 30 minutes, a non-invasive, sterile finger probe was fitted to each middle finger. A manual blood pressure cuff was placed on the distal forearm of the non-dominant arm. A baseline reading of peripheral arterial tone was recorded, and then the blood pressure cuff was inflated to a supra-systolic level approximately 60 mmHg above systolic blood pressure to induce occlusion of blood flow for five minutes. Then, the pressure was released, resulting in reactive hyperemia. Two consecutive blood pressure measures were taken immediately before and after the PAT assessment.

The PAT software then automatically calculated the reactive hyperemia index , Framingham Reactive Hyperemia Index , augmentation index , and AI adjusted to 75 beats per minute . Whole blood was collected and rested at room temperature for 15 minutes before centrifugation at 200 x g for 10 minutes. Half of the serum was then aliquoted for use as platelet rich plasma , and the remaining serum was further centrifuged at 1500 x g for 15 minutes to provide platelet poor plasma . An average platelet count of PRP was measured with a hemocytometer. Depending on the platelet number in the sample, a specific ratio of PRP and PPP was combined to create a test sample with a final cell count of 250,000 platelets per µL. Then, the combined plasma was held at room temperature for 20 minutes, after which platelet aggregation was assessed . After calibration using sterile water, 500 µL of the previously prepared combined plasma was placed into glass cuvettes and incubated at 37 ºC for three minutes. Collagen was then added to the PRP to induce aggregation while the PPP was left untouched and served as a control. The collagen was added in separate cuvettes at either 1 or 3 µg collagen per 1 ml of PRP. The changes in aggregation were measured for amplitude , slope , lag time , and area under the curve . Microvascular function, calculated by the RHI, was the primary outcome for study I. Sample size calculation was determined based on a previous study from our laboratory assessing the effects of walnuts on vascular function.23 Microvascular function values were assumed to have a standard deviation of 0.5. Therefore, a sample size of 20 was needed to detect significant differences in RHI with 80% power at a 5% level of significance. Data were checked for normality and homogeneity of variance using the Shapiro-Wilk or Brown-Forsythe tests. The two-week differences in microvascular function, anthropometric and biochemical measures, and nutrient intake were analyzed using paired-t tests. The 2h change values for microvascular function, BP, platelet aggregation, and blood glucose were analyzed by one-way repeated measure Analysis of Variance using treatment as the main factor and participant ID as the random effect. For study II, the acute changes from baseline in BP, blood glucose, and insulin were analyzed by two-way RM ANOVA using time and treatment as the main factors and participant ID as the random effect. For main effects, Tukey’s tests were used for post-hoc analysis, with student t-tests used to determine significance within group pairs. A p < 0.05 was considered statistically significant. Statistical analyses were performed with JMP version 16 . During the two-week mango intake period, the estimated increases in soluble fiber, total sugar, monosaccharides, disaccharides, β-carotene, vitamin C, vitamin E, folate, were expected, black plastic garden pots compared to the reported intakes during the run-in, no-mango period. Despite these increases in carbohydrates during the mango feeding period, fasting glucose and plasma lipid levels, body weight, and waist circumference, did not change. Some animal and human studies suggest that mango intake may benefit blood glucose control. The blood glucose levels after an oral glucose tolerance test were significantly decreased in obese Wistar rats fed a high-fat diet and supplemented with 35 ml of mango juice with or without peel extract for seven days, compared to controls.

Another study reported a significant decrease in fasting blood glucose in diabetic but not normal male Wistar rats 30 days after consuming a diet mixed with dried Tommy Atkins mango powder at 5% of diet weight. The RHI, fRHI, AI, AI75, and platelet aggregation did not differ two weeks after daily mango intake, which may have been due to the relatively short intervention period. In a randomized double-masked, placebo-controlled, four-week trial among healthy individuals aged 40-70 years with a BMI of 19-30 kg/m2 , the RHI as measured by PAT was significantly increased after daily intake of 100 mg intake of unripe mango fruit powder made from the Kili-Mooku cultivar, compared to their baseline levels. When fed the same powder at 300 mg per day, the RHI was significantly increased, but only among individuals with compromised endothelial function. Another study reported that the daily intake of 400 g of fresh frozen Ataufo mango pulp for six weeks significantly decreased SBP in lean individuals aged 18-65 with BMI 18-26.2 kg/m2 , and significantly decreased hemoglobin A1C, plasminogen activator inhibitor-1, interleukin-8, and monocyte chemoattractant protein-1 in participants with BMI > 28.9 kg/m2 . While intriguing, the results need interpreted cautiously since the BMI numbers were not the standard values used for healthy, overweight, and obese criteria. The SBP was significantly reduced in the first two hours after the first mango intake in Study I, compared to baseline or run-in values. In contrast, the SBP was unchanged in Study II at one and two hours after mango intake. The discrepancy between values from Studies I and II may be due to a low number of participants in study II. However, the change in PP was significantly reduced 2h after mango intake in both study I and II. Importantly, the PP also changed after white bread intake, suggesting that the response might be due to a postprandial effect. In study II, although the postprandial changes of SBP and DBP were not significantly different between the mango and white bread groups, the HR changes 1h and 2h after white bread intake were significantly increased compared to no mango intake. This finding is consistent with a report that both supine and standing HRs were significantly increased 1h and 3h after a 790 kcal meal in the morning after an overnight fast. However, the calorie content in mango and white bread in this current study was only 298 kcal. Studies regarding the consumption of fruits and postprandial BP and HR are scarce. Future research is encouraged to investigate whether fruit intake will induce similar hemodynamics as meals. In study I, the 2h change in blood glucose was not different between mango or no mango intake, despite the difference in sugar intake from the fruit. This observation was reinforced further in study II, where the blood glucose was significantly increased 1h after white bread intake but not after eating an isocalorically-matched amount of mango. The insulin level was also significantly increased 1h after white bread intake compared to 1h after no mango or mango intake. In addition, although the 2h change in blood glucose after eating white bread returned to a level similar to baseline values, the 2h change of insulin was still significantly elevated compared to the 2h value seen in the no mango group. These data are consistent with other reports regarding mango consumption and glucose regulation. For example, in obesity-prone mice fed a high-fat diet, the fasting blood glucose, insulin, and homeostatic model assessment for insulin resistance score were significantly decreased after 10 weeks of mango fruit powder intake at each of three levels .