Blenderized watermelon consumption decreases body mass index, body mass index percentile, body fat and HbA1c in children with overweight or obesity

Childhood obesity increases risk factors related to metabolic diseases. Watermelon's bioactive components can help reduce these risk factors. However, no study has investigated the effects of whole watermelon including both the flesh and rind or have assessed the impacts of any form of watermelon on children with overweight or obesity. The goal of this study was to examine the effects of whole‐blenderized watermelon (BWM) consumption on cardiometabolic risk factors.


| INTRODUCTION
Childhood obesity has been a growing concern in the United States for many years. In 2018, about one third of children were affected by obesity or overweight, and an additional 6.1% were affected by severe obesity. 1 It is also evident that children with obesity had higher systolic blood pressure (SBP) and diastolic blood pressure (DBP), triacylglycerols, insulin, homeostatic model assessment for insulin resistance (HOMA-IR), inflammation, endothelial dysfunction, and cardiovascular disease (CVD) risk, as well as lower high-density lipoprotein cholesterol (HDL-C) levels. 2 Furthermore, a consequence of childhood obesity is a higher risk for adults to develop obesity and CVD. 2,3 A probable risk factor of childhood obesity may be the excess intake of unhealthful snacks. Children who are affected by overweight or obesity are more likely to consume more snacks daily and consume more added sugar, saturated fats, and sodium from snacking. 4 This poor dietary habit can be conducive to poorer health outcomes. Additionally, energy intake from snacking among children was found to be 459 kcal per day, which contributed to more than 25% of their energy from snack foods. 5 Therefore, interventions or strategies are needed to reduce the risk of obesity or help with weight management.
Consuming more nutrient-dense foods, rather than energy-dense snacks, could address this issue. The Dietary Guidelines for Americans recommend consuming more fruits to help children follow a healthier dietary pattern. Specifically, fruits contain vitamins, minerals, and health-promoting phytochemicals and are also prepared with little to no added sugars, saturated fats, and sodium. 6 Therefore, the switch from energy-to nutrient-dense foods should promote better health outcomes and reduce the risk of chronic disease.
Specifically, consuming watermelon is likely a healthy snack option for children; however, empirical data in children are lacking.
Animal studies have shown decreases in low-density lipoprotein cholesterol (LDL-C), total cholesterol, free fatty acids, body weight gain, fat mass, inflammation, oxidative stress, atherosclerosis, glucose, and liver function enzymes after consuming watermelon extract, powder, or juice. [7][8][9] Similarly, in human studies, the intake of watermelon by adults has been shown to reduce total cholesterol, LDL-C, triacylglycerols, waist-to-hip (W/H) ratios, body weight, satiety, SBP, DBP, and increased HDL-C, and total antioxidant capacity. [10][11][12][13][14][15] These effects of watermelon consumption are likely due to its high citrulline content as well as other bioactive components, such as lycopene. 16 Increased intake of citrulline reduced food intake, body weight gain, cholesterol, insulin, free fatty acids, and liver function markers in obese/diabetic mice. 17 Additionally, its most notable role is as a pre-cursor to arginine, from which nitric oxide (NO) is synthesized.
Increasing levels of NO have been observed to improve endothelial dysfunction, reduce blood pressure, and attenuate atherosclerosis. 18 In addition, lycopene scavenges free radicals, thus inhibiting lipid peroxidation and plays a role in inflammation and oxidative stress, which could help with obesity induced inflammation. [19][20][21] Yet, many watermelon studies use extracts and powders instead of fresh watermelons. In addition, the studies that use fresh watermelon do not include the rind, which is an underutilized part of the fruit. Interestingly, watermelon rind has a much higher citrulline content compared to the flesh. 16,18 Therefore, most people who consume watermelon may forgo the potential health benefits of higher citrulline content due to disposing of watermelon rind. Additionally, much of the watermelon studies focus on animals and adults. There are no studies that observe children, including children affected by obesity or overweight. The objective of this study was to examine the effects of blenderized whole watermelon consumption on body weight, body mass index (BMI), and cardiometabolic biomarkers. We hypothesized that whole watermelon consumption would improve BMI, body fat, blood glucose, insulin, HbA1c, inflammation, lipid profiles, liver function enzymes, and satiety. To our knowledge, this is the first study that investigates the possible effects of consuming whole watermelon containing the rind on cardiometabolic risk factors in children affected by obesity or overweight.

| Participants
A total of 26 participants were recruited by posted flyers within the local San Diego community. Out of the 26 participants, two did not meet BMI criteria, and seven did not complete the study or were noncompliant. Thus, a total of 17 participants (nine males and eight females) were included in the analysis (Figure 1). Eligible participants were children with overweight or obesity (BMI ≥ 85th percentile based on the CDC growth chart) aged 10-17 years. 22 Female participants were excluded if they had abnormal menstrual cycles. Additionally, participants were excluded if children were using medications known to treat metabolic disease. Children with metabolic abnormalities such as higher blood lipids, glucose, or blood pressure and not taking medication for them were included. Participants were not excluded based on gender, race, ethnicity, or socioeconomic status. The study protocol was approved by the Institutional Review Board of San Diego State University. Each child signed a child assent form and each child's parent or guardian signed a parent consent form prior to participation in the study. The trial was registered at clinicaltrials.gov (#NCT04096586). BMW and SSB were provided to parents to give to their children once a day. Participating children were instructed to consume provided drinks once per day. It was suggested for parents to give the drinks as an afternoon snack after children finished school, but specific times were not assigned. Participating children or participants' parents were reminded to provide the snack through the study at each lab visit and in between visits via email or text reminders.

| Study design
They were also instructed not to consume watermelon during the sugar-sweetened beverage and washout period. Other than the provided snacks, participating children were instructed to maintain their usual diet and activity levels throughout the course of the study. Fasting blood samples (8 mL) were collected from each participant prior to and at the end of each trial. Blood samples were centrifuged at 1200Âg for 10 min at 4 C and serum was stored at À80 C until analysed.  Adiponectin was measured using an ELISA kit (RayBiotech, Norcross, GA). Incubations and washes followed manufacturer instructions. Wells were coated with a Biotinylated antibody and streptavidin were used to determine adiponectin levels. TMB substrate-based colour was measured at 450 nm.

| Satiety hormones
The appetite-regulating hormones ghrelin and leptin were assayed using ELISA kits (RayBiotech, Norcross, GA). Samples were added to antibody-specific or human ghrelin and leptin-precoated wells. Next, a biotinylated antibody and horseradish peroxidase-conjugated streptavidin solution were added to the wells, after their respective incubations and washes. A TMB substrate reagent was added to develop colour to the number of hormones bounded in the wells. After adding a stop solution, the assay plate was read in a spectrophotometer at 450 nm.

| Total antioxidant
Total antioxidant capacity was measured using the antioxidant assay kit (Cayman Chemical, Ann Arbor, MI). The assay plate was read at an absorbance of 405 nm using a spectrophotometer. The capacity of the antioxidants was compared with Trolox, a water-soluble tocopherol analogue.

| C-Reactive protein
C-reactive protein (CRP) was measured using an ELISA kit (Immundiagnostik AG, Bensheim, Germany). Samples were incubated with a rabbit-anti-CRP-antibody and then peroxidase-labelled conjugate.
Next, TMB substrate followed by a stop solution was added to each well.
Absorption was determined at 450 nm using a spectrophotometer.

| Lipid profiles
Blood lipid profiles were analysed using serum triglyceride (TG), total cholesterol (TC), and HDL-C (Stanbio Laboratory, Boerne, TX) colorimetric assay kits. LDL-C was calculated using the following formula: ). An ELISA kit was used to detect human oxidized LDL (ox-LDL) (MyBiosources, Norcross, GA). The absorbance based on the amount of ox-LDL was read at 450 nm.
Repeated measure ANOVA and paired t-tests were conducted to analyse the results and detect significance. Data are presented as mean ± standard deviation with a p ≤0.05 considered statistically significant.
Statistical difference of basal values was tested using paired t tests for baseline adjusted analysis of covariance (ANCOVA) follow ups. Pearson's correlations were conducted to analyse relationships between BMI or body fat and parameters.

| Anthropometrics
Among 17 participants, nine were males and eight were females, and the mean age was 12.9 ± 2 yr. Ethnicity of participants consisted of five White, 10 Hispanics, and two others (mixed races). Six participants were affected by overweight and 11 were affected by obesity.

| Glucose control markers
Fasting blood glucose was in the normal range. No significant differences in glucose were found between BWM and SSB (

| Satiety hormones
There was a trend of lower leptin in BWM week eight than SSB week eight (p = 0.081) ( Table 3). No significant difference was observed for ghrelin between and within trials.
3.5 | C-reactive protein, total antioxidant capacity, lipid profiles, and liver function markers No significant differences were observed between the two intervention trials regarding CRP (Table 4). Similarly, there was no significant difference for total antioxidant capacity, TG, TC, HDL-C, LDL-C, and ox-LDL. There was no significant difference for AST, but there was an increased trend after 8 weeks of SSB consumption compared to BWM week eight ( p = 0.081) ( Table 4). No significant differences were observed for the two trials regarding ALT, AP, CK, LDH, and γ-GT.

| Correlation between BMI or body fat and parameters
Significant positive correlations were observed for BMI regarding height, weight, BMIP, SBP, DBP, waist circumference, hip circumference, W/H ratio, and body fat amount ( p < 0.01) ( Table 5) There was a significant inverse relationship between body fat and adiponectin ( p < 0.001).

| DISCUSSION
Watermelon is a natural source of bioactive components, which can be conducive to better health outcomes. This is the first study to examine the potential effects of consuming BWM including the rind on anthropometrics, glucose control markers, satiety-related biomarkers, inflammation, lipid profiles, and liver function enzymes in children with overweight and obesity.
Weight management in children is one strategy to help reduced the risk of obesity in later life. Eight weeks of WMJ consumption reduced body fat, BMI, and BMIP compared to 8 weeks of SSB consumption. Poduri et al. also observed that LDL receptor-deficient mice that consumed watermelon extract for 12 weeks gained less body weight and fat mass without a change of lean body mass. 7 Similarly, adults with overweight or obesity that consumed fresh watermelon for 4 weeks exhibited decreased body weight and BMI. 7,11 Watermelon is rich in citrulline, a precursor to arginine. Citrulline supplementation reduced body weight gain and resulted in lower body fat mass in rats. 24 Meta-analysis based on human clinical trials showed arginine supplementation reduces BMI 25 and fat mass. 26 Watermelon consumption has increased blood arginine levels in postmenopausal women who were affected by overweight or obese. 27 It was also found that citrulline, arginine, nitrate, and nitrite (oxidation products of NO) increased after 4 weeks of watermelon pomace juice in Zucker diabetic fatty (ZDF) rats. 8  Several studies have shown the beneficial effects of watermelon consumption on improvements of blood pressure. [12][13][14][15] Specifically, adults who were prehypertensive, hypertensive, or obese and hypertensive that were supplemented with watermelon powder exhibited lower SBP and DBP. This may be linked to increased NO production, which can improve endothelial function. 8 We did not observe a significant effect of watermelon consumption on blood pressure in children; however, BP was normal at the start of the study, so watermelon intake was less likely to have a significant effect on BP.
Glucose dysregulation can characterize diabetes, and lifestyle modifications are recommended to reduce the risk of diabetes. 29 After 8 weeks, HbA1c levels were lower within the BWM trial. An animal study found that mice supplemented with watermelon fruit, rind, and seeds decreased levels of insulin and improved glucose metabolism. 30 Additionally, higher levels of arginine, from citrulline, may reduce glucose levels because of the increase of NO production. Higher NO production can promote the oxidation of glucose, which could increase energy expenditure and reduce blood glucose. 28 The watermelon-induced reduction in HbA1c in the current study suggests that blood glucose control was enhanced throughout the trial, which may have been secondary to improved insulin sensitivity or lower blood glucose excursions in children with obesity or overweight.
A feeling of satiety can possibly reduce food intake and help with weight management. No differences between satiety hormones at week eight of treatment trials were observed in the current study. Inflammation has been linked to obesity, type 2 diabetes, metabolic syndrome, and CVD risk. 2,20,21 In the present study, 8 weeks of BWM consumption did not change the level of CRP inflammation marker. In a similarly designed study of adults with overweight and obesity, it was also found that CRP did not change, but total antioxidant levels increased. 11 However, animal studies have demonstrated decreased circulating markers of inflammation. 7,9,31 Abnormal blood lipid profiles and liver functions are primary causal risk factors for CVD. 26 In our study, lipid markers and liver function markers did not significantly change after 8 weeks of BWM consumption. However, Massa et al. found that 6 weeks of watermelon extract consumption significantly reduced total cholesterol and LDL-C values in adults with dyslipidemia. 10 Since participants in this study were not dyslipidemic, null findings are not surprising.
Animal studies showed liver function enzymes decreased after consumption of watermelon powder for 9 weeks in rats. 9,31 Additionally, Kudo et al. 17 found that citrulline supplementation downregulated TNF-beta, which improved liver damage and reduced AST and ALT. Although our participants were young children affected by overweight or obeisty, their liver function enzyme activities were normal at baseline, which explains the reason for no positive changes from watermelon consumption.
Although our study observed the beneficial effects of BWM consumption for anthropometry and HbA1c, we acknowledge a few limitations. First, our study had a small sample size. Second, we studied the impacts of watermelon at only a single dose, and a higher dose of the watermelon should be considered for future studies since the previously mentioned studies detected significant effects on a broader range of risk factors after the consumption of higher-level amounts of watermelon. Additionally, measuring plasma concentrations of citrulline, arginine, nitrate and nitrite levels may add better understanding of watermelon's dose-dependent effect on these markers, and could elucidate optimal levels needed to see potential pathways activated.
To our knowledge, this was the first study to examine the effects of blenderized whole watermelon including flesh and rind on markers related to obesity and diabetes in children. This study demonstrated that whole watermelon consumption improved cardiometabolic risk factors including BMI, BMIP, body fat, and HbA1c with little change to other risk factors. Our study shows that watermelon is a potential alternative to unhealthful snacks for improving body composition and long-term glucose metabolism.