Source: UNIVERSITY OF CALIFORNIA, DAVIS submitted to NRP
UC DAVIS DIETARY BIOMARKERS DEVELOPMENT CENTER
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
ACTIVE
Funding Source
Reporting Frequency
Annual
Accession No.
1026941
Grant No.
2021-67017-35783
Cumulative Award Amt.
$4,100,000.00
Proposal No.
2021-08255
Multistate No.
(N/A)
Project Start Date
Aug 15, 2021
Project End Date
Aug 14, 2025
Grant Year
2024
Program Code
[A1342]- Food Specific Molecular Profiles and Biomarkers of Food and Nutrient Intake, and Dietary Exposure
Recipient Organization
UNIVERSITY OF CALIFORNIA, DAVIS
410 MRAK HALL
DAVIS,CA 95616-8671
Performing Department
Nutrition
Non Technical Summary
Memory-based dietary assessment methods are qualitative measures of diet that provide a subjective measurement that limits the ability to confidently link nutrition to health and disease risk in the greater population. New blood or urine markers of specific foods would be of tremendous help to assess diet patterns with greater precision. In this proposal, using state-of-the-art analytical and statistical tools, we propose to find biomarkers of dietary intake of various fruits and vegetables and validate these markers in controlled clinical studies. The long-term objective of this project and the UC Davis Dietary Biomarkers Development Center (UCD-DBDC) is to determine a set of blood plasma and urine biomarkers of food intake (BFI) for the fruits and vegetables food groups, that is applicable to the general population. To accomplish this objective, we will undertake 3 specific aims in the Biomarker Project, supported by 4 cores (Administrative, Intervention, Metabolomics, and Data Analysis cores). First, using metabolomics methods, the research team will determine the dose and time responses of metabolites in plasma and urine that are associated with acute exposure to increasing amounts of representative fruits or vegetables, either as a single food group (only fruit, or only vegetable) or as a combination of both food groups. We will employ a randomized controlled dietary intervention on clinically healthy individuals and collect and analyze fasting and postprandial blood and urine over the period of 24 h. Second, the research team will determine whether the BFIs identified in Aim 1 are predictive of acute and habitual consumption of MyPlate foods and food groups, and will also seek new BFIs reflective of a diet pattern rich in fruits and vegetables. We will recruit 40 volunteers and randomize them into one of two diets (either a typical American diet or a high-quality Dietary Guidelines for Americans diet) in a parallel design, and collect and analyze a series of serum and urine samples in the fasted state and after provision of the same test meal as in Aim 1. Finally, the research team will evaluate the reliability of the BFI in a cross-sectional study in a diverse cohort, comparing BFIs to traditional diet recall assessment tools. By accomplishing these aims the research team will discover and validate BFIs for fruits and vegetables in plasma and urine through the use of controlled dietary interventions and in the presence of variable background diets, and evaluate their utility to assess dietary intake relative to more traditional questionnaire-based modalities. The project will also establish novel strategies for the semi-quantitative determination of food-associated compounds for which commercially available standards are not available, a critical knowledge gap in the field.
Animal Health Component
0%
Research Effort Categories
Basic
100%
Applied
0%
Developmental
0%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
70160101010100%
Knowledge Area
701 - Nutrient Composition of Food;

Subject Of Investigation
6010 - Individuals;

Field Of Science
1010 - Nutrition and metabolism;
Goals / Objectives
The objective assessment of dietary patterns is an unmet gap in the field of Nutrition. This project will determine a set of plasma and urine biomarkers of food intake (BFI) for fruits and vegetables through three aims. In Aim 1, we will determine how acute exposure to increasing amounts of fruits and vegetables, either as a single food group or as a combination of both food groups, impacts the dose and time response of metabolites appearing in plasma and urine. Using a randomized controlled dietary intervention design, fasting and postprandial blood and urine will be collected over 24 hours after a test meal. In Aim 2, we will determine whether the BFIs identified in Aim 1 are predictive of acute and habitual consumption of fruits and vegetables in volunteers randomized to consume one of two diets over a one-week period (either a typical American or high-quality Dietary Guidelines for Americans diet), with testing as described in Aim 1. Finally, in Aim 3, we will evaluate the reliability of the BFI in a cross-sectional study of a diverse cohort, comparing BFIs to traditional diet recall assessment tools. This will allow for discovery and validation of BFIs for fruits and vegetables through the use of controlled dietary interventions and in the presence of variable background diets, and evaluate their utility to assess dietary intake relative to more traditional questionnaire-based modalities. The project will also establish novel strategies for the semi-quantitative determination of food-associated compounds for which commercially available standards are not available.
Project Methods
Aim 1: Determine the dose and time response in plasma and urine associated with exposure to a mixture of MyPlate fruits and vegetables in individuals with different dietary patterns.We will employ a randomized controlled dietary intervention on adult males and females aged 18 and above, in which different servings of a mixture of fruits or vegetables will be provided as a dose (1 fruit / 3 vegetables, 2 fruit / 2 vegetables, and 3 fruit / 1 vegetables) within a standard mixed meal setting in an inverse dosing gradient (i.e. high to low fruit / low to high vegetables). The habitual diet of each subject will be assessed prior to the start of the study through FFQ. Recent diet will be assessed using 3-day ASA 24 assessments. A fasting blood sample will be collected, followed by 1, 2, 4, 6, and 8 h after finishing the test meal. Subjects will be required to stay at the WHRNC during this time and will be fed a standardized lunch and snacks. Urine will be collected and pooled between 0-2 hr, 2-4 hr, 4-6 hr and 6-8 hr. After the 8 hr blood/urine collection, subjects will be allowed to leave with their standardized dinner to eat at home and will complete collecting urine from 8-24 h. They will return to the WHNRC having fasted since 2100 the previous evening, return their urine, and blood will be drawn. A fecal sample will be taken sometime during the 24 h, and will be banked for future analysis. Subjects will be fed standardized meals and snacks throughout the day that are low in fruits and vegetables, and thus will not interfere with the determination of the BFI for fruits and vegetables. The test subjects will participate in a total of 4 dose-interventions that will be done in random order, with at least a 48-h washout in-between.After completion of all 4 arms by each subject, urine and plasma samples will be analyzed by the Metabolomics Core using a combination of LC-MS/MS and untargeted HILIC. To identify unknown metabolites, the Metabolomics Core will use a combination of exhaustive high-resolution MS/MS MS data collections with ramped collision energies using the LC-QTOF MS, and SWATH-based LC-TripleTOF MS. This will ensure that unknowns are identified and have associated high-quality retention time and accurate masses recorded. An extensive QA/QC strategy will be employed to ensure analytical precision and stability.The variability in kinetics of metabolite appearance in the blood and urine will be analyzed with assistance by the Data Analysis Core. Our data analysis approach will start with known dietary compounds of fruit and vegetable consumption, and will also include metabolite discovery through Mass Spec analysis of the raw food ingredients combined with predicted glucuronidated and sulfated products. We will integrate food composition databases to ensure proposed biomarkers are specific to the food groups. In addition, the Data Analysis Core will determine the limits of detection, and perform kinetic modeling in blood and urine, and compare inter-individual variability. These data will then be used to rank compounds as to their particular food group discriminating ability, allow us to determine optimal times for sample collection, and also allow us to stratify into markers of acute or habitual food or food group responses. Given that subjects are expected to be highly diverse with respect to genetics, lifestyle, environmental exposures, gut microbiome and ADME profiles, we expect that metabolite levels may be so diverse that typical statistical analyses will not be applicable. We will construct five generalized linear models (GLM), adjusting for subject metadata using Gaussian, log-link Gaussian, log-normal, log-link inverse Gaussian and log-link Gamma methods, and using the subjects as random effects, evaluate the influence of baseline metabolite levels on intervention-associated changes in BFI. Models with the lowest Bayesian information criterion will be selected. In addition, we will estimate effect sizes using Bayesian regression credible intervals of >95%.Aim 2: Determine whether biomarkers of fruit and vegetable intake are predictors of acute and habitual consumption of MyPlate foods and food groups.To validate our food exposure markers in a setting where diet is defined and will result in food exposure markers from low to high, we will recruit a total of 40 volunteers. Subjects will be randomized to one of two diets (either a TAD, or a DGA) in a parallel design. The DGA will, by definition, contain significantly more fruits and vegetables. All food and beverages will be provided as prepared meals.Subjects will provide an initial FFQ so that we can quantitate their historical diet intake and pattern, and they will provide a spot urine and fasted blood sample. For 2 weeks, we will ask that subjects continue their usual diet and they will fill out three automated ASA24 questionnaires. Subjects will then return to the WHNRC and provide a fasting blood sample. Subjects will be provided a test meal which will be the same as provided in Aim 1 containing 2 servings of fruit and 2 servings of vegetables, and the same collection schedule for blood and urine as described in Aim 1 will be performed. Subjects will begin their test diets and provided with a DGA or TAD meal patterns for 1 week. The DGA diet will be optimized to contain particular dietary components or nutrients of interest, without direct reference to the mixed test foods supplied in Aim 1. Compliance will be estimated by asking individuals to fill out a daily food checklist and they will be asked to record menu deviations. This documentation will be reviewed by the behavioral counselor/nutrition educator, and any serious adherence issues or diet deviations will be discussed at scheduled meetings. Compliance will also be evaluated post-hoc by measuring 24-h urine and red blood cells for urinary potassium, urinary nitrogen, red blood cell fatty acid profiles, and/or serum carotenoids. At the end of the feeding trial, subjects will repeat the meal challenge and sample collection protocol. We will use a variety of univariate and multivariate statistical methods outlined in the Data Analysis Core to determine if usual diet impacts the predicted markers in the fasted state and following an acute challenge.Aim 3: Evaluate the robustness and reliability of the food exposure markers within the range of typical and recommended dietary intakes and examine the associations of these biomarkers with dietary assessment instrumentsFinally, we plan to determine the ability of the biomarkers we developed in Biomarker Project Aims 1 and 2 to predict recent and habitual consumption of MyPlate foods and food groups in a heterogeneous and diverse population of 200 individuals. In this aim, each subject will come in for a single visit following an overnight fast. Subjects will participate in an informed consent process, and measurements of resting pulse and blood pressure, height, weight, and waist circumference will be measured. Subjects will complete demographic and health questionnaires, and complete the Block 2005 FFQ with the assistance of a trained interviewer. A fasting blood sample will be drawn and a spot urine collected. Prior to leaving the WHNRC, subjects will be provided with a urine collection kit to complete a 24-h urine collection, and a fecal collection kit to obtain a single sample for possible future microbiota and fecal metabolomics analysis. The subjects will return the next day with their urine and fecal samples.

Progress 08/15/23 to 08/14/24

Outputs
Target Audience:Fellow scientific investigators, nutritionists, academic and biomedical trainees Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The West Coast Metabolomics Center (WCMC) at UC Davis, led by Dr. Oliver Fiehn, continued to provide a selection of in-person, online and hybrid training courses covering a range of topics in metabolomics related to our project, with duration ranging from 3.5 hours to 2 weeks. In April 2024, Dr. Adams provided a CME-accredited presentation at UCLA entitled "Bioreactors in Our Gut: Xenometabolites as Signals and Markers of Health and Diet." This was part of The Future of Nutrition in Primary Care: "Food as Medicine" program hosted by the UCLA Center for Human Nutrition. Dr. Newman provided annual graduate level lectures in the UC Davis Dept of Nutrition on Metabolomics in Nutrition Research, which included coverage of the current state of dietary biomarker discovery and utilization. How have the results been disseminated to communities of interest?A scientific session on Food Biomarkers was held during the July 2023 annual meeting of American Society of Nutrition held in Boston MA. Each DBDC site (including UC Davis site - presented by Francene Steinberg) presented information about their study design and research Objectives. Session Title: "Paving the Way to Precision Nutrition: Using metabolomics to Discover & Validate Dietary Biomarkers". What do you plan to do during the next reporting period to accomplish the goals?We anticipate that we will complete recruitment and sample collection for all subjects for Aim/Phase 1 by August of 2024. We plan to start recruitment of individuals for Aim/Phase 2 andAim/Phase 3 in July 2024. Mass spectrometry (MS) analyses of blood and urine samples from Aim/Phase 1 will commence in September 2024,and we anticipate these analyses, including data analysis willbe completed by June 2025. Analyses of samples from Aims/Phases 2 and 3 will occur after putative biomarkers have been identified in Aim/Phase 1, and as data is collected. A workshop on Food Biomarkers is planned during upcoming annual conference of Metabolomics Society "Metabolomics 2024" in June 2024 in Osaka, Japan. Representatives from each of three DBDC sites (including Dr. Fiehn from our (UC Davis) site) are participating in developing materials on nutritional metabolomics for this workshop.

Impacts
What was accomplished under these goals? Project Administration, IRB, and Protocol Development. In the third year of the grant, the UCD-DBDC team continued to meet monthly to discuss ongoing activities related to this study, with bi-weekly meetings by the intervention core team. Dr. Slupsky presented our study designs for Aims/Phases 2 and 3 and progress to the DSMB on October 3, 2023, and March 12, 2024, respectively. UCD-DBDC team members participated in monthly meetings of the DBDC steering committee and three working group committees (Dietary Intervention, Combined Metabolomics and Data Analysis and Harmonization, Publications, Presentations and Ancillary Studies). In these meetings, the core team members reviewed study updates at all three DBDC sites as well as recruitment and monthly reports, upcoming DSMB, in-person and other meetings and continued to discuss and harmonize aspects of data and sample collection, sample analysis, and statistical analysis among the consortium sites. Based on discussions in consortium meetings, as well as in our local UCD-DBDC monthly (for PIs) and bi-weekly (for implementation team) meetings, we continued to update our IRB protocol, Aim/Phase 1 and 2 REDCap database, report shells for enrollment and sample collection, tentative timelines for our study Aims/Phases 1, 2 and 3, as well as review progress in recruitment of participants, sample collection and processing. Since our last report, Dr. Adams continued to co-chair the Publications, Presentations and Ancillary studies committee, and he also started to co-chair the DBDC Intervention working group from January 2024. Dr. Slupsky started to chair DBDC steering committee in November 2023 and Dr. Newman started to co-chair DBDC Metabolomics, Data Analysis, Harmonization working group in March 2024. Additionally, the project manager (Dr. Mishchuk) represented our (UC Davis) site at DBDC Consortium paper sub-working group meetings since August 2023 and continued to participate in the meetings with project managers from two other DBDC sites to ensure that sites stay on track with respect to DBDC action items. The latest minor modifications to our IRB protocol were approved on December 12, 2023. Our site submitted monthly Aim/Phase 1 recruitment updates and REDCap data cuts to the DCC starting in June 2023. The intervention core team at our site continued to recruit participants, perform the study intervention, collect, and process samples for our Aim/Phase 1 study. Currently they are actively working on finalizing our Aim/Phase 2 study diet and are getting ready to start recruiting participants for this Aim/Phase in July 2024. Successful implementation of recruitment. UC Davis DBDC continued recruitment for the Aim/Phase 1 study. As of May 14, 2024,20 participants competed all three test days for Aim/Phase 1 of our study (this is 67% out of a total 30 participants needed for this study), 2 who are on schedule to start the intervention, 2 individuals who are awaiting eligibility screening, and 2 who are scheduled for pre-screening, from a total of 144 individuals who have completed telephone pre-screening. We are planning to complete recruitment for our Aim/Phase 1 in July 2024. We plan to start recruitment of participants for our Aims/Phases 2 and 3 in July 2024. Untargeted metabolomics efforts. As of Apr 2024, we have optimized our LC-BinBase database-level metabolite identifications for both polar and lipophilic metabolites. We have demonstrated the method utility in association with a USDA-led 'pulse' feeding study that has a similar design as the UC Davis dietary biomarker study. In a temporal study of 18 individuals, we found a range of compounds that were previously absent in the LC-BinBase DB, including compounds that were directly related to different types of food (and bean) compositions. Specifically, we could show that LC-BinBase is now robust enough to distinguish peak isomers that elute as closely as 2 s from each other, even if those isomers are reflecting in-source fragmentations. To further improve the success of compound identifications, we have now completed data acquisition and retention index plus accurate mass MS/MS assignments from three commercial libraries of combined more than 1,200 authentic compounds: (a) An alkaloid compound library of 433 unique compounds across 64 different compound classes that yielded 806 spectra; (b) a polyphenol library with383 unique compounds that resulted in 932 MS/MS spectra, with 34 different chemical classes; and (c) a pharmaceutical compound library that included more than 500 compounds, including compounds of plant or food origin such as Chamomile. All spectra have been uploaded to the UC Davis public repository MassBank.us, hosted by the Fiehn laboratory. Targeted metabolomic efforts: As of Apr 2024, we have implemented a reverse phase urolithin metabolite quantification method for both plasma and urine. We have demonstrated the method utility in association with two independent walnut feeding studies. In a temporal study of 20 individuals, we showed time dependent elevation in urolithin and urolithin glucuronides in plasma after 3wks of daily ingestion that continued to increase to 4wks and captured the reported variance in urolithin metabotypes. In an independent randomized double-blind crossover of walnut feeding, our collaborators have shown that the urine and plasma profiles of these metabolites track each other. During these and other efforts we have demonstrated the general utility of nitrophenyl glucuronide as an analytical surrogate allowing for the accurate quantification of both endogenous and exogenous glucuronide conjugates in both plasma and urine. As the sample preparation parameters parallel those of Beckmann et al. 1, we will include this surrogate into the analytical suite going forward during the next year. We have evaluated the use of a 1.015 specific gravity target for urine dilutions prior to extraction and analysis. This target results in physical dilution of ~20% of samples and estimated recoveries were not correlated with the level of dilution. These results have been shared with the other consortium members and are under active discussion and evaluation in the Metabolomics/Data Analysis Harmonization Working Group. References 1 Beckmann, M. et al. A Standardized Strategy for Simultaneous Quantification of Urine Metabolites to Validate Development of a Biomarker Panel Allowing Comprehensive Assessment of Dietary Exposure. Mol Nutr Food Res, e2000517, doi:10.1002/mnfr.202000517 (2020).

Publications

  • Type: Journal Articles Status: Published Year Published: 2023 Citation: French CD, Arnold CD, Taha AY, Engle-Stone R, Schmidt RJ, Hertz-Picciotto I, Slupsky CM (2023) Assessing repeated urinary proline betaine measures as a biomarker of usual citrus intake during pregnancy: sources of within-person variation and correlation with reported intake. Metabolites 13:904. PMID: 37623848. doi: 10.3390/ metabo13080904.


Progress 08/15/22 to 08/14/23

Outputs
Target Audience:Fellow scientific investigators, nutritionists, academic and biomedical trainees. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The West Coast Metabolomics Center (WCMC) at UC Davis, led by Dr. Oliver Fiehn, provided a selection of in-person, online and hybrid training courses covering a range of topics in metabolomics related to our project, with duration ranging from 3.5 hours to 2 weeks (more information about these courses can be accessed using this link - A third professional course series, the "Bits & Bytes" series, had one topic per course event with duration of 3.5 hours. Four units were organized for Fall 2022 and additional four units were organized for Winter and Spring 2023. Each event was limited to a maximum number of 35 participants to enable direct assistance in training exercises by the instructor and the online helpers. Additionally on October 14, 2022, the MANA 2022 Fall Symposium "We are what we eat - metabolomics leading the way for nutrition research" was organized online by UC Davis with more than 500 participants. The list of speakers at this symposium included: Lars Dragsted (U Copenhagen), Christina Andres-Lacueva (U Barselona), Carlito Lebrilla (UC Davis), Susan Sumner (UNC), Francene Steinberg (UC Davis), Steven Watkins (The Periodic Table Food Initiative). How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?We anticipate that we will complete Aim 1 by April / May of 2024, and will have approximately 1/3 of subjects complete Aim 2. We also plan to start recruitment of individuals for Aim 3 starting in 2024. Mass spectrometry (MS) analysis of blood and urine will commence once 3 participants have fully completed each study and will proceed as participants fully complete the study in groups of 3, such that blood or urine samples will be analyzed in 10 batches. Statistical analyses will occur once all data has been collected for Aim 1. Analysis for Aims 2 and 3 will occur after putative biomarkers have been identified in Aim 1, and as data is collected. A session on Food Biomarkers is planned during upcoming annual meeting of American Society of Nutrition and is scheduled on July 23, 2023. Each DBDC site (including UC Davis site) will present the information about their study design in a special session.

Impacts
What was accomplished under these goals? Project Administration, IRB, and Protocol Development. In the second year of the grant, the entire UCD-DBDC team continued to meet monthly to discuss ongoing activities related to this study, with weekly meetings by the intervention core team. Together with the intellectual property office at UC Davis, we finalized a Data Transfer Agreement between UC Davis and the Data Coordination Center (DCC) at Duke that was signed by all parties on March 27, 2023. Dr. Slupsky presented our study design and progress at an in-person DBDC Steering Committee meeting in Washington, DC on April 12, 2023 and to the DSMB on May 1, 2023. UCD-DBDC team members continued to participate in biweekly meetings of the DBDC steering committee and two working groups (Dietary Intervention and Combined Metabolomics and Data Analysis and Harmonization). In these meetings, the core team members discussed and harmonized aspects of data and sample collection, sample analysis, and statistical analysis amongst the consortium sites. Based on discussions in consortium meetings, as well as in our local UCD-DBDC team meetings, we updated and finalized our IRB protocol, Informed Consent Forms, REDCap database, report shells for enrollment and sample collection, tentative timelines for our study aims 1, 2 and 3, as well as protocols/SOPs for sample processing and storage. We also prepared and submitted proposals for supplemental funds from the NIH and USDA. UC Davis also participated in the launch of the Publications and Ancillary studies committee, with Dr. Adams as co-chair. Additionally, the project manager (Dr. Mishchuk) has been participating in separate biweekly meetings with project managers from other DBDC sites to ensure that sites stay on track with respect to DBDC action items. Our finalized IRB protocol was approved on December 15, 2022 and the information about our study was added to ClinicalTrials.gov on November 11, 2022. Our REDCap database was finalized and moved in production mode on May 28, 2023. The intervention core team finalized the design of our test diets that include all of the target fruits and vegetables, as well as the run-in diets, which included testing diets for visual appeal and palatability before the start of participant recruitment for our study. Challenge meal pilot studies were conducted among a diverse group of volunteers, helping ensure that taste, meal size, and other parameters could be finalized. We also developed SOPs and protocols for plasma, urine and fecal sample collection, processing, and storage as well as data collection and entry in our REDCap database. We are currently working on finalizing the Manual of Operations for our site that will be shared with NIDDK-CR together with the plasma and urine samples from our study that will be archived at NIDDK-CR. Our intervention team worked closely with USDA on shipping food sample aliquots from our study for sample processing and analysis at the USDA site. Successful implementation of recruitment. UC Davis DBDC started recruitment for our study in February 2023, and our first participant signed consent on March 14, 2023. The first participant completed the final of all three test days on May 24, 2023. Currently two participants have competed all three test days for aim 1 of our study, we have an additional 11 individuals who have been consented, from a total of 29 individuals who were pre-screened. We are working on scheduling participants over the summer. Untargeted metabolomics efforts. During the last period, we optimized the extraction process of plasma samples, decreasing the time and increasing the efficiency of metabolite extraction. Briefly, plasma samples are now being extracted using a Captiva-EMR 96-well plate (Agilent Technologies), allowing processing of 96 samples in ~3h. Both polar and nonpolar extracts are obtained for subsequent LC-MS analyses. Moreover, data acquisition LC-MS run time has been reduced from 15 min to 5 min. We can now generate untargeted LC-MS data for 750 injections per method and mode in approximately one month, a process that previously required several months. In addition, we have applied untargeted metabolomics to the characterization of the human upper intestinal tract 1. Of the ~1900 annotated features detected in >50% of samples in this study of 15 healthy individuals, ~350 metabolites differed between the proximal and distal intestine, with ~half higher in proximal samples. Notably, tests for associations with food intake logs validated a variety of previously reported plasma dietary biomarkers of fruits, alcohol, and previous caffeine consumption while identifying novel biomarkers in the intestinal lumen as well. Targeted metabolomic efforts: The development of targeted metabolomic assays for dietary biomarkers has proceeded, with an initial effort to implement the methods of Beckmann et al. 2. This method quantifies 62 urinary metabolites in one HILIC and one reverse phase analysis, providing a broad assessment of dietary exposures. This method will serve as the base platform for future targeted expansions. As of May 2023, 69 metabolites, including 8 glucuronides and 5 sulfates have been ordered and 90% have been received. In the meantime, we have begun by implementing an analysis for urolithins 3. Urolithins are ellagic acid metabolites and their glucuronide conjugates are putative plasma and urine biomarkers of ellagitannin-rich foods including walnuts, pecans, blackberries and black raspberries 4,5. Moreover, urolithin isomeric distributions are influenced by the gut microbiome 6. References 1 Folz, J. et al. Human metabolome variation along the upper intestinal tract. Nat Metab 5, 777-788, doi:10.1038/s42255-023-00777-z (2023). 2 Beckmann, M. et al. A Standardized Strategy for Simultaneous Quantification of Urine Metabolites to Validate Development of a Biomarker Panel Allowing Comprehensive Assessment of Dietary Exposure. Mol Nutr Food Res, e2000517, doi:10.1002/mnfr.202000517 (2020). 3 Garcia-Villalba, R., Espin, J. C. & Tomas-Barberan, F. A. Chromatographic and spectroscopic characterization of urolithins for their determination in biological samples after the intake of foods containing ellagitannins and ellagic acid. J Chromatogr A 1428, 162-175, doi:10.1016/j.chroma.2015.08.044 (2016). 4 Jeong Kang, M., Hyuk Suh, J., L. Guarneiri, L., A. Cooper, J. & M. Paton, C. LC-MS analysis of urolithin-related metabolites in human plasma reveals glucuronide conjugates as the primary species after 4-weeks of pecan consumption. Journal of Food Bioactives 21, doi:10.31665/jfb.2023.18336 (2023). 5 Wang, Y. H., Mondal, G., Khan, W., Gurley, B. J. & Yates, C. R. Development of a liquid chromatography-tandem mass spectrometry (LC-MS/MS) method for characterizing pomegranate extract pharmacokinetics in humans. J Pharm Biomed Anal 233, 115477, doi:10.1016/j.jpba.2023.115477 (2023). 6 Tomas-Barberan, F. A., Garcia-Villalba, R., Gonzalez-Sarrias, A., Selma, M. V. & Espin, J. C. Ellagic acid metabolism by human gut microbiota: consistent observation of three urolithin phenotypes in intervention trials, independent of food source, age, and health status. J Agric Food Chem 62, 6535-6538, doi:10.1021/jf5024615 (2014).

Publications

  • Type: Journal Articles Status: Published Year Published: 2023 Citation: Folz J, Culver RN, Morales JM, Grembi J, Triadafilopoulos G, Relman DA, Huang KC, Shalon D, Fiehn O. Human metabolome variation along the upper intestinal tract. Nat Metab 5, 777-788 (2023). doi:10.1038/s42255-023-00777-z.


Progress 08/15/21 to 08/14/22

Outputs
Target Audience:Dr. Francene Steinberg was interviewed for an Editor's Podcast with the Editor-in-Chief Dr. Linda Snetselaar of the Journal of the Academy of Nutrition and Dietetics (JAND). It was released on March 7, 2022 and is titled "Developing Dietary Biomarkers for the Future of Personalized Nutrition". This podcast can be accessed by the journal website under the Editor's Podcast link: https://www.jandonline.org/content/editorspodcast, or can be found on YouTube: https://www.youtube.com/watch?v=mptJcoNKP9s. This information was aimed at a target audience of dietitians and nutritionists. Dr Carolyn Slupsky and Dr. Francene Steinberg were interviewed for a campus web article titled "Diet Detectives: Identifying Blood Biomarkers that Track What Foods We Eat". It was posted on the UC Davis College of Agricultural and Environmental Sciences (CAES) website under News & Events on January 24, 2022 - https://caes.ucdavis.edu/news/diet-detectives-identifying-blood-biomarkers-track-what-foods-we-eat. The target audience is campus students and personnel, and the general public. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The West Coast Metabolomics Center (WCMC) at UC Davis, led by Dr. Oliver Fiehn, provided a selection of in-person, online and hybrid training courses covering a range of topics in metabolomics related to our project, with duration ranging from 3.5 hours to 2 weeks (more information about these courses can be accessed using this link - https://metabolomics.ucdavis.edu/courses-and-seminars/courses): · Both the Summer 2021 two-week course and the Winter 2022 one-week course were effectively and successfully completed with more than 20 trainees per course. · A third professional course series, the "Bits & Bytes" series, had one topic per course event with duration of 3.5 hours. Four units were organized for Fall 2021, six units were organized for Winter and Spring 2022. Each event was limited to a maximum number of 35 participants to enable direct assistance in training exercises by the instructor and the online helpers. Additionally on September 10, 2021, the MANA 2021 Summer Symposium "Novel analytical strategies for scoring confidence in compound identification" was organized online by UC Davis with more than 500 participants. The list of speakers at this symposium included: Erin Baker (UNC), Emma Schymanski (U Luxembourg), Kai Dührkop (U Jena, Germany), Joanna Godzien (Med U Bialystok, Poland), Corey Broeckling (UC Denver), Sara Yang (NIST), Tom Metz (PNNL). Among the eleven WHNRC online with seminars with 80-90 attendees, we added two seminar speakers with clear focus on this USDA Food Biomarkers grant: Dr. Lorraine Brennan, UC Dublin, Ireland: "Metabolomics: a powerful tool in human nutrition research" (February 2, 2022), and Dr. Justin van der Hooft, Wageningen University, the Netherlands: "Recent advances in mass spectral embedding and network based metabolomics approaches that enhance natural product discovery" (March 30, 2022). Other opportunities for training included CITI training renewal for Sean Adams and NDSR training for Aneeta Vedula, dietitian. How have the results been disseminated to communities of interest?Conferences and seminars: 1. Metabolomics Society conference, Valencia/Spain, June 2022 Dr. Oliver Fiehn et al., oral session keynote presentation: "Spatial-, temporal- and inter-person variation of metabolites across the upper and lower human gastrointestinal tract." 2. Recent Advances in Food Analysis conference, Prague/Czech Rep., November 2021 Dr. Oliver Fiehn et al., oral presentation (online), "Confidence in metabolite annotations from high-resolution MS/MS in food digestion along the human GI tract" 3. American Society of Mass Spectrometry conference, Minneapolis/MN, June 2022 Dr. Arpana Vaniya et al., oral presentation, "Comparison of CFM-ID, MS-FINDER, and SIRIUS+CSI:FingerID with LC-high resolution-MS/MS data from 500 new reference compounds" 4. American Society of Mass Spectrometry conference, Philadelphia/PA, November 2021 Dr. Arpana Vaniya et al., poster, "Extending the MassBank.us natural product reference mass spectral library with more than 4,400 new reference standards" What do you plan to do during the next reporting period to accomplish the goals? For the coming year, we plan to finalize the recruitment of lab personnel, submit our modified IRB for approval, finalize diets (including performing sensory analyses to ensure palatability), finalize our RedCap database, finalize SOPs, and run composite analysis of the diets to ensure macronutrient and micronutrient profiles are adequate and correct. As we are waiting to begin, we will perform mock test-days to ensure that once patients are recruited, our study will run smoothly. As soon as our modified and finalized IRB is approved, and our study is added to ClinicalTrials.gov, we will start recruiting participants for Aims 1 and 2 of the study. We anticipate that we will have approximately 10 - 15 individuals complete the study for Aim 1, and 20 to 25 individuals complete the study for Aim 2. Mass spectrometry (MS) analysis of blood and urine will commence once 3 participants have fully completed each study and will proceed as participants fully complete the study in groups of 3, such that blood or urine samples will be analyzed in 10 batches. Statistical analyses will occur once all data has been collected for Aim 1. Analysis for Aims 2 and 3 will occur after putative biomarkers have been identified in Aim 1, and as data is collected. In addition to starting the study, Dr. Oliver Fiehn is organizing the MANA 2022 Symposium which will occur on October 14, 2022, and is titled: "We are what we eat - metabolomics leading the way for nutrition research". This is an online symposium at UC Davis and it will include Lars Dragsted amongst other speakers.?

Impacts
What was accomplished under these goals? The objective assessment of dietary patterns is an unmet gap in the field of Nutrition. This project will determine a set of plasma and urine biomarkers of food intake for specific fruits and vegetables. Aim 1 will determine the pharmacokinetics and dose response resulting in the accumulation and excretion of biomarker metabolites in plasma and urine over a period of 24 hours, using 0, 1, or 2 servings of three fruits (strawberry, banana, and peach in a smoothie) and 0, 1, or 2 servings of three vegetables (green beans, tomatoes, and carrots in a soup). Aim 2 will determine whether the biomarkers generated in Aim 1 are predictive of habitual consumption in the context of low consumption (such as someone consuming them as part of a Typical American diet (TAD)) or high consumption (such as someone consuming them as part of a high quality Dietary Guidelines for Americans (DGA) diet). In Aim 3, we will evaluate the reliability of the biomarkers in a cross-sectional study of a diverse cohort, comparing these biomarkers to traditional diet recall assessment tools. This will allow for validation of biomarkers for fruits and vegetables in the presence of variable background diets, and evaluate their utility to assess dietary intake relative to more traditional questionnaire-based modalities. The project will also establish novel strategies for the semi-quantitative determination of food-associated compounds for which commercially available standards are not available. In this first year of the grant, our administrative core team set up regular monthly UCD-DBDC team meetings to discuss all ongoing activities related to this study, and additionally set up an internal secure collaborative site (through Confluence) where all documentation related to the study is stored for easy access at any time by team members. This has allowed (and will allow) new members to come up to speed quickly as they are onboarded. This internal website includes the original grant application, all minutes from internal and consortium meetings, as well as IRB protocols and related documents with version control. In the future, this website will host deidentified data as it is generated, analysis algorithms used for data analysis, final analyzed data, and will archive any data shared with the consortium members and manuscripts produced. This will allow us to archive all work done, and create a transparent way to share data amongst the group. This will also allow all members of the team access to raw and analyzed data generated through this grant and be a place where members can creatively generate new hypotheses for ancillary studies. Over the first year of this study, our UCD-DBDC team members participated in biweekly meetings of the DBDC steering committee and three working groups (Dietary Intervention, Metabolomics, and Data Analysis and Harmonization). Dr. Francene Steinberg has been serving as co-chair of the Dietary Intervention working group. In these meetings, the core teams have been focused on reviewing study design and harmonizing aspects of data and sample collection, sample analysis, and statistical analysis amongst the consortium sites. Based on discussions in consortium meetings, as well as in our local UCD-DBDC team meetings, we have made modifications to our original study design and developed an updated IRB protocol, Data Safety Management Plan (DSMP) and Statistical Analysis Plan (SAP) that are aligned with the other consortium sites. At the end of June all this work culminated in a meeting with the Data and Safety Monitoring Board (DSMB), comprised of individuals not associated with any of the consortium centers. At this meeting, we presented our study designs, shared our upated IRB protocols, as well as DSMPs and SAPs. Our initial IRB protocol was approved last year and currently we are finalizing modifications to it based on the suggestions from DSMB that we received at the end of July, anticipating that early in the second year of this project we will be able to begin recruiting subjects. In addition to study design, we have been working on designing the diets that includes all of the target fruits and vegetables that we are seeking biomarkers for. We developed the menus for the first aim of our study, and completed the diet design for the 2-day run-in diet. We have also designed the two day rotating menu for the second aim of our study. We have hired both a dietitian and a clinical research coordinator who will be responsible for finalizing the diets and setting up the recruitment process. Based on the guidance document for the Electronic Data Capture (EDC) System that was created through collaboration between sites in the DBDC Data Harmonization committee, we have begun to develop our inhouse REDCap database, and anticipate that it will be ready for the start of the study participant recruitment once our revised protocol has been approved by the IRB. The metabolomics core team has also made significant progress this year: (1) finalizing the new Fiehn Natural Products MS/MS library and; and (2) developing and finalizing the new Fiehn RP-LCMS/MS method for natural products analysis. For the natural products library, they purchased ~4,000 compounds from food products funded under a separate NIH grant and acquired retention times and mass spectra for many of these compounds using a pentafluorophenyl-RP liquid chromatography column and MS/MS analyses. Curating these spectra was more difficult than expected, so time and personnel from this project was used to generate "Libgen", a fully automated software for constructing, curating, optimizing, and sharing mass spectral libraries. The program was written in Python and the graphical user interface (GUI) was developed using the Kivy framework. The capability of the software generated in the metabolomics core was demonstrated by constructing a genuine mass spectral library using the data from a novel LC-MS system, the Sciex ZenoTof 7600, which utilizes both EAD and CID for fragmentation. In the EAD-CID library, the metabolomics core successfully identified 2,618 unique compounds and 627 compounds with duplicates. About 70% of duplicate compounds with over 0.7 entropy similarity were included in the library. After mass recalibration, the developed library contains 3,048 unique compound spectra. This novel automated library curation software for LC-MS system provides a highly reliable and robust new natural products database for food and food biomarker research. In addition, members of the metabolomics core developed a new Standard Operating Procedure (SOP) for a method to ensure measurement of compounds with varying lipophilicities (such as for different flavonoids and polyphenols found in foods). The corresponding SOP for the overall method was written up in a 15-page document for UC Davis internal use, until publication of the method, after which it will be available for the general scientific public.

Publications