The present pilot study tested the efficiency of nanoTiO₂ sunscreen to prevent the oxidative stress/inflammation caused by ultraviolet (UV) radiation using biomarkers in subjects’ blood, urine, and exhaled breath condensate (EBC). In addition, the skin absorption of nanoTiO₂ was studied. Six identical subjects participated in three tests: (A) nanoTiO₂ sunscreen, (B) UV radiation, and (C) sunscreen + UV. The first samples were collected before the test and the second after sunscreen application and/or UV exposure. On day 4, the third samples were collected, and the sunscreen was washed off, and the fourth samples were collected on day 11. The following biomarkers were measured: malondialdehyde, 4-hydroxy-trans-hexenal, 4-hydroxy-trans-nonenal, aldehydes C6-C12, 8-iso-Prostaglandin F2α, o-tyrosine, 3-chlorotyrosine, 3-nitrotyrosine, 8-hydroxy-2-deoxyguanosine, 8-hydroxyguanosine, 5-hydroxymethyl uracil, and leukotrienes, using liquid chromatography-electrospray ionisation-tandem mass spectrometry. Titania was measured using inductively coupled plasma mass spectrometry and TiO₂ nanoparticles by transmission and scanning electron microscopy. Sunscreen alone did not elevate the markers, but UV increased the biomarkers in the plasma, urine, and EBC. The sunscreen prevented skin redness, however it did not inhibit the elevation of oxidative stress/inflammatory markers. Titania and nanoTiO₂ particles were found in the plasma and urine (but not in the EBC) in all sunscreen users, suggesting their skin absorption.

Mass spectrometry (MS) and nuclear magnetic resonance (NMR) have emerged as pivotal tools in biofluid metabolomics, facilitating investigation of disease mechanisms and biomarker discovery. Despite complementary capabilities, these techniques are rarely combined, although their integration is often beneficial. Typically, different sample preparation approaches are used, and compatibility challenges potentially arise due to the requirement for deuterated buffered solvents in NMR but not MS techniques. Additionally, MS-based approaches necessitate protein removal from samples whilst in NMR proteins can be potentially useful biomarkers. In this study, we developed a blood serum preparation protocol enabling sequential NMR and multi-LC-MS untargeted metabolomics analysis using a single serum aliquot in a research discovery setting. We analysed human serum samples using various untargeted NMR and multi-LC-MS platforms to assess the impact of deuterated solvents and buffers on detected compound-features. Employing multiple LC-MS profiling approaches, we observed no evidence of deuterium incorporation into metabolites following sample preparation with deuterated solvents. Furthermore, we demonstrated that buffers used in NMR were well tolerated by LC-MS. Protein removal, involving both solvent precipitation and molecular weight cut-off (MWCO) filtration, was identified as a primary factor influencing metabolite abundance. Our findings led to the development and validation of a serum sample preparation protocol enabling a combined NMR and multi-LC-MS analysis. Using a single clinical serum aliquot for simultaneous untargeted profiling via NMR and multi-LC-MS represents a highly efficient alternative to current methods. This approach reduces sample volume requirements and substantially expands the potential for broader metabolome coverage. Our study offers comprehensive insights into the impact of sample preparation on complex metabolic biofluid profiles, highlighting the compatibility and complementarity of LC-MS and NMR in metabolomics research.

Dicarbonyl stress plays an important role in the pathogenesis of microvascular complications that precede the formation of advanced glycation end products, and contributes to the development of renal dysfunction. In renal cells, toxic metabolites like methylglyoxal lead to mitochondrial dysfunction and protein structure modifications. In our study, we investigated the effect of methylglyoxal on metabolic, transcriptomic, metabolomic and proteomic profiles in the context of the development of kidney impairment in the model of metabolic syndrome. Dicarbonyl stress was induced by intragastric administration of methylglyoxal (0.5 mg/kg bw for 4 weeks) in a strain of hereditary hypertriglyceridaemic rats with insulin resistance and fatty liver. Methylglyoxal administration aggravated glucose intolerance (AUC_0–120 p < 0.05), and increased plasma glucose (p < 0.01) and insulin (p < 0.05). Compared to controls, methylglyoxal-treated rats exhibited microalbuminuria (p < 0.01). Targeted proteomic analysis revealed increases in urinary secretion of pro-inflammatory parameters (MCP-1, IL-6, IL-8), specific collagen IV fragments and extracellular matrix proteins. Urine metabolomic biomarkers in methylglyoxal-treated rats were mainly associated with impairment of membrane phospholipids (8-isoprostane, 4-hydroxynonenal). Decreased levels of glutathione (p < 0.01) together with diminished activity of glutathione-dependent antioxidant enzymes contributed to oxidative and dicarbonyl stress. Methylglyoxal administration elevated glyoxalase 1 expression (p < 0.05), involved in methylglyoxal degradation. Based on comparative transcriptomic analysis of the kidney cortex, 96 genes were identified as differentially expressed (FDR < 0.05). Network analysis revealed an over-representation of genes related to oxidative stress and pro-inflammatory signalling pathways as well as an inhibition of angiogenesis suggesting its contribution to renal fibrosis. Our results support the hypothesis that dicarbonyl stress plays a key role in renal microvascular complications. At the transcriptome level, methylglyoxal activated oxidative and pro-inflammatory pathways and inhibited angiogenesis. These effects were further supported by the results of urinary proteomic and metabolomic analyses.

Untargeted metabolic profiling of plasma and serum by liquid chromatography-mass spectrometry (LC-MS) is becoming increasingly important in clinical and translational research; however, sample preparation protocols can have a significant impact on study outcomes, and there is currently a lack of standardized approaches. In this study we demonstrate that pretreatment of serum and plasma samples with 1% formic acid (FA, v/v) prior to acetonitrile (MeCN)-induced protein precipitation significantly enhances analytical performance in untargeted metabolomics using reversed-phase liquid chromatography (RPLC)-MS. We show an increase in sample preparation reproducibility and signal intensity across both positive and negative ionization modes. In two independent serum cohorts (OPTIMA and VITACOG), FA-based extraction improved multivariate modeling (orthogonal partial least-squares discriminant analysis, OPLS-DA), with consistently higher classification accuracy, sensitivity, and specificity, alongside reduced variability and increased fold-changes in discriminatory compound-features. We investigated factors potentially involved in the enhanced performance and observed outcomes consistent with the disruption of noncovalent protein-metabolite interactions and the stabilization of labile species. We found no correlation with either protein depletion or differential adduct formation. The results were also not attributable to lowering pH after metabolite extraction. In summary, we demonstrate that FA pretreatment of plasma and serum, prior to protein precipitation, significantly improves sample reproducibility and detection sensitivity in untargeted RPLC-MS metabolomics. This optimized sample preparation strategy offers clear advantages for clinical and translational metabolomics, with the potential to enhance biomarker discovery and metabolic phenotyping.

INTRODUCTION: Elevated total homocysteine (tHcy) is a major predictor of brain atrophy, cognitive decline, and Alzheimer's disease (AD) progression. The VITACOG trial, a randomized, placebo‐controlled study in mild cognitive impairment (MCI), previously showed that B vitamin supplementation lowered tHcy, slowing brain atrophy and cognitive decline; however, the underlying mechanisms remained unclear. METHODS: We used untargeted, multi‐platform metabolomics, with nuclear magnetic resonance and liquid chromatography‐mass spectrometry to analyze serum samples from 89 B vitamin–treated and 84 placebo‐treated MCI participants over a 2 year follow‐up period. RESULTS: Multivariate modeling distinguished treated from placebo groups with 91.2 ± 1.8% accuracy. B vitamin supplementation induced significant metabolic reprogramming, lowering quinolinic acid, α‐ketoglutarate, α‐ketobutyrate, glucose, and glutamate. DISCUSSION: These findings reveal that B vitamins influence metabolic pathways beyond tHcy reduction, particularly the tricarboxylic acid cycle and glutamine–glutamate cycling, critical for brain energy homeostasis and neurotransmission. This metabolic signature supports B vitamin supplementation as a strategy for slowing MCI progression. Highlights: Nuclear magnetic resonance and multi‐platform liquid chromatography tandem mass spectrometry metabolomics were performed on serum samples from 89 B vitamin–treated and 84 placebo participants in the VITACOG trial. Multi‐platform metabolomics revealed B vitamin–driven metabolic reprogramming, achieving 91% classification accuracy. B vitamin supplementation modulates key neuroprotective metabolic pathways. Regulation of energy metabolism and neurotransmission by B vitamins contributes to brain health in elderly individuals. B vitamins demonstrate potential as an adjunct therapy in mild cognitive impairment, potentially mitigating progression to Alzheimer's disease.