Metabolic Signatures Associated with COVID-19 Vaccination in Serum from Healthy Individuals

Background: COVID-19 vaccines have proven effective in reducing severe disease and mortality from SARS CoV-2 infection. The underlying molecular mechanisms and alterations in the human serum metabolome influencing the effectiveness and development of immunity remain unclear. Methods: Serum samples were collected from 29 healthy individuals at three time points: prior to vaccination (A), post-first dose (B), and post-second dose (C). Untargeted high-resolution (HR) liquid chromatography coupled with mass spectrometry (LC-MS) was performed on these samples. Metabolites showing significant differential abundance at each time point were identified, and both multivariate and univariate statistical analyses were performed to determine changes associated with the pairwise comparisons, priming (A vs. B), booster (B vs. C), and the overall vaccine effect (A vs. C). Vaccination-specific features were determined after excluding metabolites associated with SARS-CoV-2 IgG seropositivity to better isolate vaccine-driven metabolic changes. Bioinformatics, pathway, and network analyses were conducted using Ingenuity Pathway Analysis (IPA) to identify relevant pathways. Results: Our study identified significant metabolic changes across the three time points. A total of 377 metabolites were identified, of which 59 metabolites, including prostaglandins, eicosanoids, and lipids, were shared across all three groups. The majority of these metabolites showed an initial decrease after the first dose, followed by broad upregulation after the second dose. We identified 1 (downregulated), 34 (26 upregulated and 8 downregulated), and 18 (2 upregulated and 16 downregulated) unique metabolites in the priming, booster, and the overall vaccine effect groups, respectively. L-3-hydroxykynurenine was observed to be significantly reduced by the priming dose effect. By contrast, the booster effect showed decreased myo-inositol 1,3,4,5-tetrakisphosphate, while levels of DL-DOPA, 3-methoxytyrosine, and prostaglandin-esterified phospholipids, including PC(P-16:0/PGF1α) and PE(PGF1α/18:0), increased. On the other hand, the overall vaccine effect revealed decreased cyclic AMP and increased 3'-O-methyladenosine levels. These changes were associated with perturbations in arachidonic acid metabolism, glycerophospholipid metabolism, arginine biosynthesis, and steroid hormone biosynthesis. IPA network analysis identified AKT, TP53, EGFR, and cAMP as key dysregulated nodes. Conclusions: Longitudinal metabolomic profiling demonstrated that COVID-19 vaccination induced distinct but interrelated biochemical changes throughout the vaccination course. The priming dose induced a limited set of early metabolic changes, whereas the booster was associated with more significantly changed metabolites that were involved in lipid, bile acid, steroid, amino acid, and nucleotide pathways. Together, these findings indicate that sequential vaccination is associated with dose-dependent systemic metabolic adaptation, with the booster dose having the largest number of dysregulated metabolites.