Targeted Metabolomics in the REGARDS Stroke Case-Cohort Study

Abstract

Targeted Metabolomics in the REGARDS Stroke Case-Cohort Study. A prospective epidemiological study with a case-cohort design. Targeted metabolomics was conducted on plasma samples from a nested case-cohort study within the bi-racial REasons for Geographic and Racial Differences in Stroke (REGARDS), a longitudinal prospective cohort. The REGARDS study was designed to investigate different health outcomes with a focus on stroke disparities across the USA. Participants were oversampled from the Southeastern regions of the "stroke belt," where stroke risk and mortality are 2-4 times higher in the African American population. The study recruited 30,239 community-dwelling Black and White participants aged 45 years or older. Plasma samples were collected during the baseline study visit between 2003 and 2007 and stored at -80°C. This dataset includes a total of 2,378 plasma samples representing the stroke case-cohort sub-study. Participants of the random cohort (N=1,056) were selected at the point of enrolment. Stroke cases were adjudicated up to April 2009 (N=1,322).

Background

Metabolomics provides a powerful tool for identifying biomarkers that can inform drug development and preventative strategies by monitoring metabolic changes across health and disease states. Extensive research has demonstrated the metabolic basis of stroke risk and pathology, with dietary patterns and nutrient intake influencing metabolic pathways linked to stroke. Metabolic syndrome components, including hypertension, diabetes, central obesity, and lipid imbalances, are strongly associated with increased stroke risk. Large-scale epidemiological studies have identified specific metabolites linked to stroke incidence. This study examines metabolite levels in participants from the REasons for Geographic and Racial Differences in Stroke (REGARDS) study, a large, biracial, longitudinal cohort designed to investigate health disparities and stroke risk in the United States, with an emphasis on Black Americans and individuals living in the high-risk "Stroke Belt" region.

Methods

Sample Collection Summary: K2EDTA plasma samples were collected at the time of the baseline study visit. All participants were asked to fast at least 8 hours before venous blood drawing in the early morning. Plasma was isolated in a timely manner and stored at -80 °C. Samples went through a freeze-thaw cycle for the purpose of sub-aliquoting and shipping for metabolomics assays. Samples went through a second freeze-thaw cycle immediately before extraction. All procedures requiring manual sample handling were conducted on wet ice. Sample Type: Blood (plasma) Storage Conditions: -80℃

Sample Preparation Summary: Samples were extracted over wet ice using protein precipitation. First, 30 μL of K2EDTA plasma was aliquoted onto a 96-well plate. Before quenching, each received 30 µL of internal solution mixture containing stable isotope-labeled proline (13C5, 15N), glutamine (13C5, 15N2), leucine-d10, and phenylalanine-d8 [(15 µM each in methanol: water (50:50)]. Samples were then briefly vortexed for 1 minute at 1400 rpm to ensure homogeneity. Proteins were precipitated with 110 µL of ice-cold acetonitrile: methanol (75:25), and samples were vortexed for a further 5 min at 1400 rpm. Samples were centrifuged at 4000 rpm for 10 min at 4°C, and 100 µL of the clean supernatants were transferred to clean well plates for LC-MS/MS injection. Processing Storage Conditions: 4℃ Extraction Method: Protein Precipitation (PPE)

MS Summary for negative ionization mode: Instrument Name: Agilent 6495 QQQ Instrument Type: Triple quadrupole Ionization source: ESI MS Comments: Metabolites were measured using an Agilent 6495 triple quadrupole mass spectrometer in dynamic MRM mode. The ESI source settings were as follows: polarity: negative, gas temperature: 230°C, gas flow: 15 L/min, nebulizer pressure: 30 psi, sheath gas temperature: 400°C, sheath gas flow: 12 L/min, capillary voltage: 2500 V, and nozzle voltage: 1000 V. Dynamic MRM transitions with collision energies are detailed in the attached protocol. Data were processed using Agilent MassHunter QQQ Quantitative Analysis utilizing the Agile2 integrator, Quartic/Quintic Savitzky-Golay, and the Root Mean Square (RMS) noise algorithms. Normalized peak areas were calculated by dividing the peak areas of the analytes by the corresponding nearest human pooled plasma (HPP) signal for each analyte. (The provided data included HPP samples; therefore, the values in the data table are pre-normalization).

MS Summary for positive ionization mode: Instrument Name: Agilent 6495 QQQ Instrument Type: Triple quadrupole Ionization source: ESI MS Comments: Metabolites were measured using an Agilent 6495 triple quadrupole mass spectrometer in dynamic MRM mode. The ESI source settings were as follows: polarity: positive, gas temperature: 230°C, gas flow: 15 L/min, nebulizer pressure: 30 psi, sheath gas temperature: 400°C, sheath gas flow: 12 L/min, capillary voltage: 2000 V, and nozzle voltage: 1000 V. Dynamic MRM transitions with collision energies are detailed in the attached protocol. Data were processed using Agilent MassHunter QQQ Quantitative Analysis utilizing the Agile2 integrator, Quartic/Quintic Savitzky-Golay, and the Root Mean Square (RMS) noise algorithms. Normalized peak areas were calculated by dividing the peak areas of the analytes by the corresponding nearest human pooled plasma (HPP) signal for each analyte. (The provided data included HPP samples; therefore, the values in the data table are pre-normalization).

Data Description

Raw files were acquired as Agilent .d files. Prior to data upload, files were converted to an open-source mzML format using ProteoWizard. Each data file was assigned a unique number reflecting the order of acquisition. Each study sample has one unique identifier and two associated data files: one acquired in negative ionization mode and one in positive ionization mode. Files with odd numbers contain data from negative ionization mode, while files with even numbers contain data from positive ionization mode. For example, files 19.mzML and 20.mzML contain information about the same sample extract for metabolites ionizing in negative and positive ionization modes, respectively. If a study sample represents an HPP (used for quality control or normalization), the assigned unique identifier includes both a number and a letter, which matches the HPP's negative ionization mode data file name. A separate tabulated Study Design table includes the unique identifiers, the corresponding data file names, and the case-cohort designation.
Outputs from the Agilent MassHunter processed analyte peak area values are provided in two separate files, one for the negative and one for the positive ionization mode compounds. Values may be normalized before further processing to account for systematic differences across each batch using the nearest pooled plasma normalization approach. Custom code to compute area ratio values from the deposited data is provided as a Stata code (v.17) and available for download from Figshare (https://doi.org/10.6084/m9.figshare.26755777).

Usage Notes

This dataset provides metabolomics data from the REasons for Geographic and Racial Differences in Stroke (REGARDS) Study, intended for research on metabolic biomarkers associated with stroke risk and related health disparities. Users are encouraged to explore the data for biomarker discovery, disease risk assessment, and metabolic pathway analyses.

Researchers should refer to the REGARDS study documentation for details on cohort design, participant recruitment, and data collection methods.

External Documentation: REGARDS Study website: https://www.uab.edu/soph/regardsstudy/

Related publications:

• Ament, Z. et al. Nucleosides Associated With Incident Ischemic Stroke in the REGARDS and JHS Cohorts doi 10.1212/WNL.0000000000200262
• Bhave, V. M. et al. Plasma Metabolites Link Dietary Patterns to Stroke Risk doi 10.1002/ana.26552
• Kijpaisalratana, N. et al. Association of Circulating Metabolites With Racial Disparities in Hypertension and Stroke in the REGARDS Study doi 10.1212/WNL.0000000000207264
• Ament, Z. et al. Gut microbiota-associated metabolites and risk of ischemic stroke in REGARDS. doi 10.1177/0271678X231162648
• Kijpaisalratana, N. et al. Trimethylamine N-Oxide and White Matter Hyperintensity Volume Among Patients With Acute Ischemic Stroke doi 10.1001/jamanetworkopen.2023.30446
• Kijpaisalratana, N. et al. Acetylglutamine Differentially Associated with First-Time Versus Recurrent Stroke doi 10.1007/s12975-023-01181-1
• Jones, A. C. et al. Metabolite profiles and DNA methylation in metabolic syndrome: a two-sample, bidirectional Mendelian randomization doi 10.3389/fgene.2023.1184661
• Kijpaisalratana, N. et al. Plasma Metabolites and Life’s Simple 7 in REGARDS doi 10.1161/STROKEAHA.123.044714.

Data files are available in .mzML fomat, Data results output is available in CSV format. Data normalization code is provided as Stata format (.do). Additonal method relsated documentation is recorded in Excel or PDF formats.  

Data Processing: Custom code created in Stata (v17) for nearest pooled plasma normalization and %CV calculation can be accessed from Figshare (https://doi.org/10.6084/m9.figshare.26755777)

Ethics

This study was approved by the Institutional Review Boards of all participating institutions, and informed consent was obtained from all participants. All procedures adhered to the ethical standards of institutional and national research committees and complied with the 1975 Helsinki Declaration and its later amendments. Metabolomics analysis was conducted under IRB approval by the Mass General Brigham Human Research Committee (MGBHRC, protocol number: IRB 2016P001801).

Conflicts of Interest

Conflict of Interest Statement: The authors declare no financial, commercial, legal, or professional relationships that could be construed as a potential conflict of interest in relation to this research.