(160bf) Novel Biomarkers for Arterial Stiffness Identified with Untargeted Metabolomics Analysis

Several studies in children have shown a progressive increase in arterial stiffness as measured by pulse wave velocity (PWV) during childhood, indicating a structural change in the arterial wall that initiates early in life. These changes may accelerate under several conditions, the most well-known of which is obesity. Blood pressure (BP) in this context has been extensively investigated while there are less data by measuring 24-hour blood pressure (ABPM) and central BP. The aim of the present study was to identify novel metabolic markers in children suffering from arterial stiffness. Interaction of body mass index (BMI) with arterial stiffness, 24-hour recording and central BP in the pediatric population – were, also, taken into account.

The study included 50 children who came for random visits to a tertiary pediatric clinic. The children were divided into 2 groups: Group 1 = maximum PWV third, Group 2 = lower PWV third. The groups did not differ in age and gender and underwent measurements of ABPM, PWV, AIx as well as measurements of the central CA. We used t-test to investigate the differences between the two groups and linear regression analysis to investigate the determinants of PWV.

Serum samples were thawed and kept under control conditions at 4^οC throughout sample preparation. First, samples were vortexed to homogenize and then 200 μL of the sample were submitted to protein precipitation using 1:3 water:methanol. After centrifugation, at 15000rpm for 15 min, 300μl of the supernatant were evaporated to dryness through a sample concentrator using N₂. Dry extracts were reconstituted using 100μl of 100% water with internal standards, following by vortex for 5 min, and centrifugation at 15000rpm for 10 min. Supernatants were transferred to autosampler vials using inserts for analysis. QC samples were prepared by pooling an equal volume of all serum samples following by the aforementioned sample preparation.

Serum samples were analysed using an Agilent 1290 infinity HPLC LC System coupled to an Agilent Q-TOF 6540 MS system using a Dual AJS ESI probe in both positive and negative modes. A Zorbax Eclipse Plus C18 (50 x 2.1 mm, 1.8 μm, Agilent, USA) column was used kept at 40 ^oC. The flow rate and the injection volume were set at 250 μL/min and 5 μL respectively. The mobile phases for positive ionization (PI) and negative ionization mode (NI), were water (solvent A) and methanol (solvent B) both amended with 0.01% formic acid. The adopted gradient elution program was the same for both ionization modes, starting with 0% B with a flow rate of 0.35 mL/min for 2 min, increased to 100 % B in 17 min, kept constant for 5 min before the initial conditions were restored within 2 min, and kept for another 2 min. Data were acquired between 50 and 1000 m/z at a scan rate of 1.4 spectra/s in Full Scan centroid mode at a resolution of 40,000 FWHM. The source conditions are as follows: gas temperature 300 ^oC, drying gas 7 L/min, nebulizer 50 psig, fragmentor 150 V, skimmer 65 V, and capillary voltage 3500V or -3500V in positive or negative mode, respectively. Solvent and quality control samples were analyzed during the run sequence.

The UPLC-TOF-MS data of the analysed samples collected with the Agilent MassHunter Workstation Data Acquisition Software v.B.06.01. Spectral processing was performed using the Bioconductor R - based packages XCMS v.3.10.1 (Smith et al., 2006) and CAMERA v.1.44.0 (Kuhl et al., 2012), running under R version 3.6.1 (http://http://www.r-project.org/). Chromatographic peak detection was performed using the centWave algorithm. The Obiwarp method was used for alignment, and the retention time adjustment map was used as a diagnostic plot. The featureDefinitions function was used for the correspondence. We used the fillChromPeaks method to fill in intensity data for missing values from the original files due to false negatives. Finally, the PerformPeakAnnotation function used for isotope and adduct annotation using the CAMERA package. Normalization by median, mean centering scaling, and log transformation were performed to transform the data matrix into a more Gaussian-type distribution, thus, to reduce systematic error in experimental conditions. Alignment and filtering resulted in two data matrices of 1941 and 7529 metabolite features in positive and negative ionization mode respectively. The annotation R package xMSAnnotator developed by Uppal et al. (2017), was used to perform the accurate mass carries in online compound databases (HMDB, LipidMaps, and KEGG). The significantly differential metabolites were determined by using the non-parametric Kruskal-Wallis test in R followed by the Benjamini-Hochberg false discovery rate (FDR) correction for multiple comparisons to minimize false positives. A p-value cut-off of 0.05 was considered (Table 1). We calculated the area under the receiver-operating characteristic (ROC) curve to evaluate the predictive performance of the selected variables using the R package MetaboAnalystR 3.0 (Pang et al., 2020).

The performed analysis led to the detection of 15 novel biomarkers for the classification of the individuals with arterial stiffness. It is worth mentioning that the majority of the metabolites belong to the lipids category, and more specifically to fatty acyls. In vitro validation will be further performed by investigating the absolute levels using targeted quantitative double-antibody sandwich enzyme-linked immunosorbent assay (ELISA).

Our results describe the importance of childhood obesity in the progression of blood pressure and the acceleration of vascular aging. The mechanisms by which increased weight affects the vascular system beyond the rise in blood pressure remain to be elucidated.

Table 1. Summary statistics and annotation information on the potential biomarkers of hypertension. The biomarkers selection was based on the p-value (<0.05), and the log2 fold change (log2(FC) > 2). The selected biomarkers were evaluated using the area under the receiver operating characteristic curve (AUC). Metabolites annotation was performed the R package xMSAnnotator and confirmed by comparison with the RT, and fragmentation pattern of authentic analytical standards from the in-house or online libraries.

References

KUHL, C., TAUTENHAHN, R., BÖTTCHER, C., LARSON, T. R. & NEUMANN, S. 2012. CAMERA: An Integrated Strategy for Compound Spectra Extraction and Annotation of Liquid Chromatography/Mass Spectrometry Data Sets. Analytical Chemistry, 84, 283-289.

PANG, Z., CHONG, J., LI, S. & XIA, J. 2020. MetaboAnalystR 3.0: Toward an Optimized Workflow for Global Metabolomics. Metabolites, 10.

SMITH, C. A., WANT, E. J., O'MAILLE, G., ABAGYAN, R. & SIUZDAK, G. 2006. XCMS: processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching, and identification. Anal Chem, 78, 779-87.

UPPAL, K., WALKER, D. I. & JONES, D. P. 2017. xMSannotator: An R Package for Network-Based Annotation of High-Resolution Metabolomics Data. Anal Chem, 89, 1063-1067.