Abstract P280: Urinary Hippurate and Proline Betaine Excretion relative to Fleshy Fruit Intake and Blood Pressure: The INTERMAP Study

Abstract

Background: Intakes of biologically complex fleshy fruits with substantial variations in nutritional composition may be reflected in urinary metabolite differences. We evaluated 24 hr urinary hippurate and proline betaine relative to individual fruit intake and blood pressure (BP) among free-living individuals in a population-based study. Methods: Cross-sectional epidemiological and metabonomic data from the International Study of Macro-/Micronutrients and Blood Pressure were used, available for 4,276 men and women ages 40-59 years from China, Japan, United Kingdom, and United States. Fleshy fruit intakes, according to Stern’s botanical classification, were calculated from four in-depth 24 hr dietary recalls. Urinary spectra from timed 24 hr urine collections were obtained by 1H nuclear magnetic resonance spectroscopy. Peak intensities of urinary hippurate (δ 7.84, doublet) and proline betaine (δ 3.11, singlet) were quantified. Mean urinary metabolite excretions across quartiles of energy-adjusted fruit intakes and corresponding P for trend were calculated. Country-specific multivariable linear regression coefficients between urinary metabolites and sum of metabolite-related fruits were estimated and pooled, weighted by inverse of their variance. Results: Age-sex-sample adjusted Pearson correlations (r) were: between hippurate and raw fruits 0.17, fruit juice 0.01, urinary potassium 0.19, and dietary fiber 0.15; between proline betaine and raw fruit 0.27, fruit juice 0.30, vitamin C 0.40, sugar 0.21, and dietary fiber 0.16; between hippurate and proline betaine 0.02. Significant multivariable adjusted P for trends (metabolome-wide significance level; P<4x10-6) across quartiles of intake were found between total raw fruit, apple, kiwi fruit, nectarine, peach, plum, dried apricot, dried prune and hippurate; similarly between raw fruit, fruit juice, banana, grapefruit, kumquat, orange, satsuma, tangerine, grapefruit juice, orange juice and proline betaine. Correlations between sum of metabolite-related fruits and hippurate were 0.24 (0.10 for sum of remaining fruits) and 0.64 with proline betaine (0.03 for remaining fruits). After multivariable adjustment including BMI, differences in systolic BP were for urinary hippurate higher by 2SD -1.53 mm Hg (95% CI: -2.32,-0.75) and for sum of hippurate-related fruits -0.60 (-1.32,0.12). Urinary proline betaine and sum of proline betaine-related fruits were not associated with BP. Conclusion: Consumption of several individual fruits was related to urinary hippurate and proline betaine excretion. Higher urinary hippurate excretion was associated with lower systolic BP, but this was not found for hippurate-related fruits. This suggests that other polyphenol-rich foods and/or quantity may be important in this relationship.