Till Örebro universitet

Mass spectrometry-based methods have become fundamental to exposome research, providing the capability to explore a broad spectrum of chemical exposures. Liquid and gas chromatography coupled with low/high-resolution mass spectrometry (MS) are among the most frequently employed platforms due to their sensitivity and accuracy. However, these approaches present challenges, such as the inherent complexity of MS data and the expertise of biologists, chemists, clinicians, and data analysts to integrate and interpret MS data with other datasets effectively. The "omics" era advances rapidly, driven by developments of AI-based algorithms and an increase in accessible data; nevertheless, further efforts are necessary to ensure that exposomics outputs are comparable and reproducible, thus enhancing research findings. This review outlines the principles of MS-based methods for the exposome analytical pipeline, from sample collection to data analysis. We summarize and review both standard and cutting-edge strategies in exposome research, covering sample preparation, focusing on MS-based platforms, data acquisition strategies, and data annotation. The ultimate goal of this review is to highlight applications that enable the simultaneous analysis of endogenous metabolites and xenobiotics, which can help enhance our understanding of the impact of human exposure on health and disease and support personalized healthcare.

BACKGROUND: Traumatic brain injury (TBI) can come with long term consequences for functional outcome that can complicate return to work. OBJECTIVES: This study aims to make accurate patient-specific predictions on one-year return to work after TBI using machine learning algorithms. Within this process, specific research questions were defined: 1 How can we make accurate predictions on employment outcome, and does this require follow-up data beyond hospitalization? 2 Which predictors are required to make accurate predictions? 3 Are predictions accurate enough for use in clinical practice?

METHODS: This study used the core CENTER-TBI observational cohort dataset, collected across 18 European countries between 2014 and 2017. Hospitalized patients with sufficient follow-up data were selected for the current analysis (N = 586). Data regarding hospital stay and follow-up until three months post-injury were used to predict return to work after one year. Three distinct algorithms were used to predict employment outcomes: elastic net logistic regression, random forest and gradient boosting. Finally, a reduced model and corresponding ROC-curve was created.

RESULTS: Full models without follow-up achieved an area under the curve (AUC) of about 81 %, which increased up to 88 % with follow-up data. A reduced model with five predictors achieved similar results with an AUC of 90 %.

CONCLUSION: The addition of three-month follow-up data causes a notable increase in model performance. The reduced model - containing Glasgow Outcome Scale Extended, pre-injury job class, pre-injury employment status, length of stay and age - matched the predictive performance of the full models. Accurate predictions on post-TBI vocational outcomes contribute to realistic prognosis and goal setting, targeting the right interventions to the right patients.

Recent studies reveal that gut microbes produce diverse bile acid conjugates, termed microbially conjugated bile acids (MCBAs). However, their regulation and health effects remain unclear. Here, we analyzed early-life MCBA patterns and their link to islet autoimmunity. We quantified 110 MCBAs in 303 stool samples collected longitudinally (3-36 months) from children who developed one or more islet autoantibodies and controls who remained autoantibody-negative. Stool MCBAs showed distinct age-dependent trajectories and correlated with gut microbiome composition. Altered levels of ursodeoxycholic and deoxycholic acid conjugates were linked to islet autoimmunity as well as modulated monocyte activation in response to immunostimulatory lipopolysaccharide and Th17/Treg cell balance. These findings suggest MCBAs influence immune development and type 1 diabetes risk.

BACKGROUND: There is growing interest in the role of environmental factors (i.e., exposome) in the pathogenesis of Alzheimer's diseases. The exposome includes three categories: internal (e.g., metabolism, gut microbiome, inflammation), specific external (e.g., environmental pollutants, diet, drugs, occupational), and general external (e.g., socioeconomic status, education, climate). The metabolome provides a readout of the influences of the exposome, capturing the presence of a large number of chemical exposures and allowing an interrogation of the influences of these chemicals on cognition and brain imaging changes.

METHOD: Four centers of excellence in metabolomics have used mass spectrometry-based capabilities to provide broad coverage of the chemical exposome. A 'ring trial' was performed to determine the coverage of the metabolome/exposome using state-of-the-art analytical tools. Human serum/plasma standards from diseased individuals - Alzheimer's, COPD, IBS, osteoarthritis, and Type 2 diabetes (BioIVT), and NIST standards (1950 and 1958) were analyzed, using LC-MS/MS (targeted and untargeted), Direct Flow-MS/MS and/or ICP-MS. Similarly, we are measuring the exposome/metabolome in large studies of AD patients, including ADNI, ADRCs and the ROSMAP brain collection.

RESULT: Compounds were classified by chemical class (ClassyFire), toxicity/source groups (EPA CompTox DB)), and disease association (Comparative Toxicogenomics DB). Using the available InChiKey and CAS numbers available, ClassyFire sorted compounds into 184 chemical classes, EPA CompTox DB sorted 1179 unique InChiKey descriptors into 93 groups (toxicity class and/or exposure source), including DrugBank, Hazardous Substances DB 2019, BLOOD Toxic Substances Control Act, COSMO cosmetics and FDA Food Substances. These included 7 neurological groups with over 100 ring trial compounds. These compounds were classified as endogenous compounds, foods, medications, industrial chemicals, surfactants, plasticizers, personal care, and pesticides (Figure 1). Identical studies are being performed using large cohorts of brain and blood samples, including from ADRCs and the ROSMAP collection. We will present highlights of the brain chemical exposome and its links to cognition.

CONCLUSION: Big data is being generated to capture the influences of the exposome on brain health, connecting these peripheral influences on brain metabolic health and disease, facilitating the development of novel therapeutic approaches targeting the exposome and its effect on brain health.

Metabolites from feces provide important insights into the functionality of the gut microbiome. As immediate freezing is not always feasible in gut microbiome studies, there is a need for sampling protocols that provide the stability of the fecal metabolome and microbiome at room temperature (RT). Here, we investigated the stability of various metabolites and the microbiome (16S rRNA) in feces collected in 95% ethanol (EtOH) and commercially available sample collection kits with specific preservatives OMNImet•GUT/OMNIgene•GUT. To simulate field-collection scenarios, the samples were stored at different temperatures at varying durations (24 h + 4 °C, 24 h RT, 36 h RT, 48 h RT, and 7 days RT) and compared to aliquots immediately frozen at -80 °C. We applied several targeted and untargeted metabolomics platforms to measure lipids, polar metabolites, endocannabinoids, short-chain fatty acids (SCFAs), and bile acids (BAs). We found that SCFAs in the nonstabilized samples increased over time, while a stable profile was recorded in sample aliquots stored in 95% EtOH and OMNImet•GUT. When comparing the metabolite levels between aliquots stored at room temperature and at +4 °C, we detected several changes in microbial metabolites, including multiple BAs and SCFAs. Taken together, we found that storing samples at RT and stabilizing them in 95% EtOH yielded metabolomic results comparable to those from flash freezing. We also found that the overall composition of the microbiome did not vary significantly between different storage types. However, notable differences were observed in the α diversity. Altogether, the stability of the metabolome and microbiome in 95% EtOH provided results similar to those of the validated commercial collection kits OMNImet•GUT and OMNIgene•GUT, respectively.

In this community effort, we compare measurements between 34 laboratories from 19 countries, utilizing mixtures of labelled authentic synthetic standards, to quantify by mass spectrometry four clinically used ceramide species in the NIST (National Institute of Standards and Technology) human blood plasma Standard Reference Material (SRM) 1950, as well as a set of candidate plasma reference materials (RM 8231). Participants either utilized a provided validated method and/or their method of choice. Mean concentration values, and intra- and inter-laboratory coefficients of variation (CV) were calculated using single-point and multi-point calibrations, respectively. These results are the most precise (intra-laboratory CVs ≤ 4.2%) and concordant (inter-laboratory CVs < 14%) community-derived absolute concentration values reported to date for four clinically used ceramides in the commonly analyzed SRM 1950. We demonstrate that calibration using authentic labelled standards dramatically reduces data variability. Furthermore, we show how the use of shared RM can correct systematic quantitative biases and help in harmonizing lipidomics. Collectively, the results from the present study provide a significant knowledge base for translation of lipidomic technologies to future clinical applications that might require the determination of reference intervals (RIs) in various human populations or might need to estimate reference change values (RCV), when analytical variability is a key factor for recall during multiple testing of individuals.

Background: The effects of cannabis are thought to be mediated by interactions between its constituents and the endocannabinoid system. Delta-9-tetrahydrocannabinol (THC) binds to central cannabinoid receptors, while cannabidiol (CBD) may influence endocannabinoid function without directly acting on cannabinoid receptors. We examined the effects of THC coadministered with different doses of CBD on plasma levels of endocannabinoids in healthy volunteers.

Methods: In a randomized, double-blind, four-arm crossover study, healthy volunteers (n=46) inhaled cannabis vapor containing 10 mg THC plus either 0, 10, 20, or 30 mg CBD, in four experimental sessions. The median time between sessions was 14 days (IQR=20). Blood samples were taken precannabis inhalation and at 0-, 5-, 15-, and 90-min postinhalation. Plasma concentrations of THC, CBD, anandamide, 2-arachidonoylglycerol (2-AG), and related noncannabinoid lipids were measured using liquid chromatography-mass spectrometry.

Results: Administration of cannabis induced acute increases in plasma concentrations of anandamide (+18.0%, 0.042 ng/mL [95%CI: 0.023-0.062]), and the noncannabinoid ethanolamides, docosatetraenylethanolamide (DEA; +35.8%, 0.012 ng/mL [95%CI: 0.008-0.016]), oleoylethanolamide (+16.1%, 0.184 ng/mL [95%CI: 0.076-0.293]), and N-arachidonoyl-L-serine (+25.1%, 0.011 ng/mL [95%CI: 0.004-0.017]) (p<0.05). CBD had no significant effect on the plasma concentration of anandamide, 2-AG or related noncannabinoid lipids at any of three doses used. Over the four sessions, there were progressive decreases in the preinhalation concentrations of anandamide and DEA, from 0.254 ng/mL [95%CI: 0.223-0.286] to 0.194 ng/mL [95%CI: 0.163-0.226], and from 0.039 ng/mL [95%CI: 0.032-0.045] to 0.027 ng/mL [95%CI: 0.020-0.034] (p<0.05), respectively.

Discussion: THC induced acute increases in plasma levels of anandamide and noncannabinoid ethanolamides, but there was no evidence that these effects were influenced by the coadministration of CBD. It is possible that such effects may be evident with higher doses of CBD or after chronic administration. The progressive reduction in pretreatment anandamide and DEA levels across sessions may be related to repeated exposure to THC or participants becoming less anxious about the testing procedure and requires further investigation. The study was registered on clinicaltrials.gov (NCT05170217).

Environmental toxicants (ETs) are associated with adverse health outcomes. Here we hypothesized that exposures to ETs are linked with obesity and insulin resistance partly through a dysbiotic gut microbiota and changes in the serum levels of secondary bile acids (BAs). Serum BAs, per- and polyfluoroalkyl substances (PFAS) and additional twenty-seven ETs were measured by mass spectrometry in 264 Danes (121 men and 143 women, aged 56.6 ± 7.3 years, BMI 29.7 ± 6.0 kg/m2) using a combination of targeted and suspect screening approaches. Bacterial species were identified based on whole-genome shotgun sequencing (WGS) of DNA extracted from stool samples. Personalized genome-scale metabolic models (GEMs) of gut microbial communities were developed to elucidate regulation of BA pathways. Subsequently, we compared findings from the human study with metabolic implications of exposure to perfluorooctanoic acid (PFOA) in PPARα-humanized mice. Serum levels of twelve ETs were associated with obesity and insulin resistance. High chemical exposure was associated with increased abundance of several bacterial species (spp.) of genus (Anaerotruncus, Alistipes, Bacteroides, Bifidobacterium, Clostridium, Dorea, Eubacterium, Escherichia, Prevotella, Ruminococcus, Roseburia, Subdoligranulum, and Veillonella), particularly in men. Conversely, females in the higher exposure group, showed a decrease abundance of Prevotella copri. High concentrations of ETs were correlated with increased levels of secondary BAs including lithocholic acid (LCA), and decreased levels of ursodeoxycholic acid (UDCA). In silico causal inference analyses suggested that microbiome-derived secondary BAs may act as mediators between ETs and obesity or insulin resistance. Furthermore, these findings were substantiated by the outcome of the murine exposure study. Our combined epidemiological and mechanistic studies suggest that multiple ETs may play a role in the etiology of obesity and insulin resistance. These effects may arise from disruptions in the microbial biosynthesis of secondary BAs.

Background and aims: Individuals with metabolic dysfunction-associated steatotic liver disease (MASLD) and steatohepatitis (MASH) have higher prevalence of type 2 diabetes (T2D) and impaired fasting glycemia. Given the role of the liver as main contributor to endogenous glucose production (EGP), we investigated if hepatic glucose fluxes were increased with worsening of liver histology, from liver steatosis (SL) to MASH with advanced fibrosis, and the interplay with adipose tissue lipolysis.

Materials and methods: The cohort comprised n=260 subjects with liver biopsy from the EPoS cohort (92/168 Female/Male; 153/109 noT2D/T2D; 58/202 SL/MASH), of whom n=206 had transcriptomics of liver biopsy (EPoS-RNA) and n=68 (EPoS-flux) had measurement of fasting EGP and adipose tissue lipolysis by stable-isotope (6,6- 2 H 2 -glucose and U- 2 H 5 -glycerol infusion.). Intra-hepatic glucose fluxes were estimated using genome-scale metabolic modeling (GSMM) of transcriptomic data and results were validated vs tracer measurements.

Results: Tracer-measured EGP and hepatic insulin resistance were increased with histological severity (mainly with fibrosis and inflammation, Kruskal-Wallis’ test p< 0.05) and correlated with insulin resistance and tracer-measured adipose tissue lipolytic flux (Pearson’s test p<0.0001), even in absence of T2D. GSMM-analysis showed that high EGP was due to increased gluconeogenesis and anaplerosis and was associated with excess substrates uptake, rather than changes in the expression of glucogenic genes. However, hepatic expression of genes involved in insulin signaling (IRS1, IRS2, and AKT2) was reduced with the worsening of liver fibrosis and the presence of T2D (Kruskal-Wallis’ test p< 0.05).

Conclusion: In MASLD, not only lipids but also glucose metabolism is impaired. The dysregulation of EGP is due to the elevated release of gluconeogenic substrates from peripheral tissues and to the impairment in insulin signaling, and is associated with the severity of hepatic fibrosis and inflammation, rather than steatosis. These findings contribute to improve our understanding of the mechanisms underlying the increased risk of hyperglycemia and T2D in subjects with MASH.

Metabolic dysfunction-associated steatotic liver disease (MASLD) and steatohepatitis (MASH) are associated with a high prevalence of type 2 diabetes (T2D). Individuals with MASLD exhibit insulin resistance (IR) and hyperglycemia, but it is unclear whether hepatic glucose production (HGP) is increased with MASLD severity. We evaluated HGP in a cohort of histologically characterized individuals with MASL/MASH using stable isotope infusion (6,6-2H2-glucose, U-2H5-glycerol) and liver-specific genome-scale metabolic models (GEMs). Tracer-measured HGP is increased with liver fibrosis and inflammation, but not steatosis, and is associated with lipolysis and IR. The GEM-derived gluconeogenesis is elevated due to high glucogenic/energy metabolite uptakes (lactate, glycerol, and free fatty acid [FFA]), and the expression of insulin action genes (IRS1, IRS2, and AKT2) is reduced in MASH with fibrosis F2-F4, with/without T2D, suggesting these as putative mechanisms for increased fasting HGP and hyperglycemia. In conclusion, elevated HGP, lipolysis, and IR help to explain the mechanisms for the increased risk of hyperglycemia and T2D in MASH.