Urinary Biomarkers of Dietary Intake - Quantification of Sugars for Epidemiology Studies Using LC-MS/MS Nicola Gray (Presenter) University of Reading

Authors: Nicola Gray. Gunter GC Kuhnle
Department of Food and Nutritional Sciences, University of Reading, Reading, UK

For epidemiology studies that aim to investigate how diet influences health, accurate records of nutritional intake are essential. Nutritional assessment typically relies on self-reported food diaries or food-frequency questionnaires, known to be inaccurate and introduce bias. The development of accurate markers to assess dietary intake is therefore crucial to investigate associations between diet, health and disease. For example, urinary sucrose and fructose have been shown to correlate with total sugars intake and have been developed as a marker of sugar consumption. The application of such a biomarker to population studies require robust methods suitable for high-throughput analysis. Here, an LC-MS/MS method has been developed to distinguish and quantify sugars in urine for application to large clinical cohorts as an accurate measure of sugars consumption in free-living populations.

Introduction

Extensive epidemiological studies suggest that high consumption of refined sugars is associated with an increased risk of type 2 diabetes, cancer and cardiovascular diseases. In free-living populations, investigating the associations between disease risk and nutritional intake relies on self-reported dietary assessments, which are prone to errors such as underreporting and subjective estimations, and hence introduce bias in studies. Therefore, accurate dietary assessment methods are essential for the investigation of cause-effect relationships with regard to potential health risks. Nutritional biomarkers are an alternative assessment method which allow an objective and often unbiased account of dietary intake.

Urinary excretion of sucrose and fructose, for example, has been well described as a system to determine total sugars intake [1]. Currently, quantification methods mainly relay on enzymatic assays, which are time-consuming and labour intensive, or HPLC with optical detection. With the specificity offered by tandem quadrupole MS/MS, a fast and robust method for high-throughput analysis has been developed. The method presented illustrates the optimisation of an LC-MS/MS method for the quantification of sucrose and fructose in human urine. Chromatographic separation prior to MS/MS detection provides a platform to differentiate between various types of sugar, including sucrose and fructose, but also related substitutes often found in the diet such as trehalose. The application of the method demonstrates its suitability to large scale epidemiological studies in providing a tool for dietary assessment, and also aims to investigate exposure to alternative sweeteners including trehalose.

Methods

24-hour urine samples (100 µL) were prepared by addition of acetonitrile (1:1 v/v) containing internal standards (13C12-sucrose and 13C6-fructose). LC-MS/MS analysis was performed on a Waters Acquity UPLC coupled to a Quattro Ultima mass spectrometer. Separation was achieved utilising a 2.1 x 100mm Acquity BEH Amide column with isocratic conditions of 80% acetonitrile and 20% water containing 0.2 % NH4OH. The mass spectrometer was operated with positive ESI and data acquired in MRM mode. A three-day assay validation was performed to assess the precision and accuracy of the method and the robustness of the system. Following successful validation, application to clinical cohorts comprising adult and children samples demonstrated the applicability of the method to large scale analyses.

Results

The development of a robust method for the targeted analysis of sugars in urine is presented. Optimisation of chromatographic and mass spectrometry parameters was performed to ensure maximum sensitivity and reliability. Minimal sample pre-treatment was desirable for use as a high-throughput approach for large-scale epidemiological cohorts. The resulting method employing hydrophilic interaction liquid chromatography (HILIC) permits the retention and reproducible separation of sugars, including sucrose, fructose and trehalose. A variety of mobile phase additives were evaluated for chromatographic and ionisation efficiency, concluding that 0.2% NH4OH resulted in optimal performance.

The range of quantification spanned 0.1 - 500 µM. Stable isotope labelled standards were used to compensate for potential ion suppression or enhancement caused by the urine matrix components. This LC-MS/MS assay has been validated to indicate that it is linear, accurate and precise and suitable for large-scale epidemiological studies. Validation data indicated excellent linearity (r2 > 0.999) and accuracy and precision over thee QC levels (< 15 % error). The method was applied to several cohorts of samples from free-living individuals to provide an accurate measure of dietary exposure and support the interpretation of consumption in relation to health and disease.

Conclusions & Discussion

A robust LC-MS/MS method for the quantification of sugars in human urine has been developed and applied to samples from free-living large clinical cohorts. Sucrose and fructose data collected is reflective of sugar intake and is used as a tool to support the interpretation of clinical studies aiming of improve our understanding of the impact sugar consumption on disease risk.

1. Tasevska, N., et al., Urinary Sucrose and Fructose as Biomarkers for Sugar Consumption. Cancer Epidemiology Biomarkers & Prevention, 2005. 14(5): p. 1287-1294.