Refine
Document Type
- Article (15)
- Doctoral Thesis (1)
- Preprint (1)
Has Fulltext
- yes (17)
Is part of the Bibliography
- no (17)
Keywords
- Endocannabinoids (2)
- Lipidomics (2)
- Metabolomics (2)
- Pre-analytics (2)
- lipidomics (2)
- metabolomics (2)
- (lyso)phospholipids (1)
- 2-AG (1)
- Biomarker Discovery Study (1)
- Blood sample handling (1)
The emerging disciplines of lipidomics and metabolomics show great potential for the discovery of diagnostic biomarkers, but appropriate pre-analytical sample-handling procedures are critical because several analytes are prone to ex vivo distortions during sample collection. To test how the intermediate storage temperature and storage period of plasma samples from K3EDTA whole-blood collection tubes affect analyte concentrations, we assessed samples from non-fasting healthy volunteers (n = 9) for a broad spectrum of metabolites, including lipids and lipid mediators, using a well-established LC-MS-based platform. We used a fold change-based approach as a relative measure of analyte stability to evaluate 489 analytes, employing a combination of targeted LC-MS/MS and LC-HRMS screening. The concentrations of many analytes were found to be reliable, often justifying less strict sample handling; however, certain analytes were unstable, supporting the need for meticulous processing. We make four data-driven recommendations for sample-handling protocols with varying degrees of stringency, based on the maximum number of analytes and the feasibility of routine clinical implementation. These protocols also enable the simple evaluation of biomarker candidates based on their analyte-specific vulnerability to ex vivo distortions. In summary, pre-analytical sample handling has a major effect on the suitability of certain metabolites as biomarkers, including several lipids and lipid mediators. Our sample-handling recommendations will increase the reliability and quality of samples when such metabolites are necessary for routine clinical diagnosis.
Small molecule biomarker discovery: Proposed workflow for LC-MS-based clinical research projects
(2023)
Mass spectrometry focusing on small endogenous molecules has become an integral part of biomarker discovery in the pursuit of an in-depth understanding of the pathophysiology of various diseases, ultimately enabling the application of personalized medicine. While LC-MS methods allow researchers to gather vast amounts of data from hundreds or thousands of samples, the successful execution of a study as part of clinical research also requires knowledge transfer with clinicians, involvement of data scientists, and interactions with various stakeholders.
The initial planning phase of a clinical research project involves specifying the scope and design, and engaging relevant experts from different fields. Enrolling subjects and designing trials rely largely on the overall objective of the study and epidemiological considerations, while proper pre-analytical sample handling has immediate implications on the quality of analytical data. Subsequent LC-MS measurements may be conducted in a targeted, semi-targeted, or non-targeted manner, resulting in datasets of varying size and accuracy. Data processing further enhances the quality of data and is a prerequisite for in-silico analysis. Nowadays, the evaluation of such complex datasets relies on a mix of classical statistics and machine learning applications, in combination with other tools, such as pathway analysis and gene set enrichment. Finally, results must be validated before biomarkers can be used as prognostic or diagnostic decision-making tools. Throughout the study, quality control measures should be employed to enhance the reliability of data and increase confidence in the results.
The aim of this graphical review is to provide an overview of the steps to be taken when conducting an LC-MS-based clinical research project to search for small molecule biomarkers.
Non-alcoholic steatohepatitis (NASH) and alcoholic steatohepatitis (ASH) are the leading causes of liver disease worldwide. To identify disease-specific pathomechanisms, we analyzed the lipidome, metabolome and immune cell recruitment in livers in both diseases. Mice harboring ASH or NASH had comparable disease severities regarding mortality rate, neurological behavior, expression of fibrosis marker and albumin levels. Lipid droplet size was higher in NASH than ASH and qualitative differences in the lipidome were mainly based on incorporation of diet-specific fatty acids into triglycerides, phosphatidylcholines and lysophosphatidylcholines. Metabolomic analysis showed downregulated nucleoside levels in both models. Here, the corresponding uremic metabolites were only upregulated in NASH suggesting stronger cellular senescence, which was supported by lower antioxidant levels in NASH as compared to ASH. While altered urea cycle metabolites suggest increased nitric oxide synthesis in both models, in ASH, this depended on increased L-homoarginine levels indicating a cardiovascular response mechanism. Interestingly, only in NASH were the levels of tryptophan and its anti-inflammatory metabolite kynurenine upregulated. Fittingly, high-content immunohistochemistry showed a decreased macrophage recruitment and an increased polarization towards M2-like macrophages in NASH. In conclusion, with comparable disease severity in both models, higher lipid storage, oxidative stress and tryptophan/kynurenine levels were seen in NASH, leading to distinct immune responses.
We recently described that monoacylglycerol lipase (MGL) is present in the tumor microenvironment (TME), increasing tumor growth. In this study we compare the implications of MGL deficiency in the TME in different tumor types.
We show that subcutaneous injection of KP (KrasLSL-G12D/p53fl/fl, mouse lung adenocarcinoma) or B16-F10 cells (mouse melanoma) induced tumor growth in MGL wild type (WT) and knockout (KO) mice. MGL deficiency in the TME attenuated the growth of KP cell tumors whereas tumors from B16-F10 cells increased in size. Opposite immune cell profiles were detected between the two tumor types in MGL KO mice. In line with their anti-tumorigenic function, the number of CD8+ effector T cells and eosinophils increased in KP cell tumors of MGL KO vs. WT mice whereas their presence was reduced in B16-F10 cell tumors of MGL KO mice. Differences were seen in lipid profiles between the investigated tumor types. 2-arachidonoylglycerol (2-AG) content significantly increased in KP, but not B16-F10 cell tumors of MGL KO vs. WT mice while other endocannabinoid-related lipids remained unchanged. However, profiles of phospho- and lysophospholipids, sphingomyelins and fatty acids in KP cell tumors were clearly distinct to those measured in B16-F10 cell tumors.
Our data indicate that TME-localized MGL impacts tumor growth, as well as levels of 2-AG and other lipids in a tumor specific manner.
Traumatic brain injury (TBI) is often complicated by long-lasting disabilities, including headache, fatigue, insomnia, hyperactivity, and cognitive deficits. In a previous study in mice, we showed that persistent non-goal-directed hyperactivity is a characteristic post-TBI behavior that was associated with low levels of endocannabinoids in the perilesional cortex. We now analyzed lipidome patterns in the brain and plasma in TBI versus sham mice in association with key behavioral parameters and endocannabinoids. Lipidome profiles in the plasma and subcortical ipsilateral and contralateral brain were astonishingly equal in sham and TBI mice, but the ipsilateral perilesional cortex revealed a strong increase in neutral lipids represented by 30 species of triacylglycerols (TGs) of different chain lengths and saturation. The accumulation of TG was localized predominantly to perilesional border cells as revealed by Oil Red O staining. In addition, hexosylceramides (HexCer) and phosphatidylethanolamines (PE and ether-linked PE-O) were reduced. They are precursors of gangliosides and endocannabinoids, respectively. High TG, low HexCer, and low PE/PE-O showed a linear association with non-goal-directed nighttime hyperactivity but not with the loss of avoidance memory. The analyses suggest that TG overload and HexCer and PE deficiencies contributed to behavioral dimensions of post-TBI psychopathology.
Endocannabinoids (ECs) are potent lipid mediators with high physiological relevance. They are involved in a wide variety of diseases like depression or multiple sclerosis and are closely connected to metabolic parameters in humans. Therefore, their suitability as a biomarker in different (patho-)physiological conditions is discussed intensively and predominantly investigated by analyzing systemic concentrations in easily accessible matrices like blood. Carefully designed pre-analytical sample handling is of major importance for high-quality data, but harmonization is not achieved yet. Whole blood is either processed to serum or plasma before the onset of analytical workflows and while knowledge about pre-analytical challenges in plasma handling is thorough they were not systematically investigated for serum.
Therefore, the ECs AEA and 2-AG, and closely related EC-like substances 1-AG, DHEA, and PEA were examined by LC-MS/MS in serum samples of nine healthy volunteers employing different pre-analytical sample handling protocols, including prolonged coagulation, and storage after centrifugation at room temperature (RT) or on ice. Furthermore, all analytes were also assessed in plasma samples obtained from the same individuals at the same time points to investigate the comparability between those two blood-based matrices regarding obtained concentrations and their 2-AG/1-AG ratio.
This study shows that ECs and EC-like substances in serum samples were significantly higher than in plasma and are especially prone to ex vivo changes during initial and prolonged storage for coagulation at RT. Storage on ice after centrifugation is less critical. However, storage at RT further increases 1-AG and 2-AG concentrations, while also lowering the already reduced 2-AG/1-AG ratio due to isomerization. Thus, avoidance of prolonged processing at RT can increase data quality if serum as the matrix of choice is unavoidable. However, serum preparation in itself is expected to initiate changes of physiological concentrations as standard precautionary measures like fast and cooled processing can only be utilized by using plasma, which should be the preferred matrix for analyses of ECs and EC-like substances.
Immune-mediated inflammatory diseases (IMIDs), such as rheumatoid arthritis (RA), psoriatic arthritis (PsA), and psoriasis (Ps), represent autoinflammatory and autoimmune disorders, as well as conditions that have an overlap of both categories. Understanding the underlying pathogeneses, making diagnoses, and choosing individualized treatments remain challenging due to heterogeneous disease phenotypes and the lack of reliable biomarkers that drive the treatment choice. In this review, we provide an overview of the low-molecular-weight metabolites that might be employed as biomarkers for various applications, e.g., early diagnosis, disease activity monitoring, and treatment-response prediction, in RA, PsA, and Ps. The literature was evaluated, and putative biomarkers in different matrices were identified, categorized, and summarized. While some of these candidate biomarkers appeared to be disease-specific, others were shared across multiple IMIDs, indicating common underlying disease mechanisms. However, there is still a long way to go for their application in a routine clinical setting. We propose that studies integrating omics analyses of large patient cohorts from different IMIDs should be performed to further elucidate their pathomechanisms and treatment options. This could lead to the identification and validation of biomarkers that might be applied in the context of precision medicine to improve the clinical outcomes of these IMID patients.
Monoacylglycerol lipase (MGL) expressed in cancer cells influences cancer pathogenesis but the role of MGL in the tumor microenvironment (TME) is less known. Using a syngeneic tumor model with KP cells (KrasLSL-G12D/p53fl/fl; from mouse lung adenocarcinoma), we investigated whether TME-expressed MGL plays a role in tumor growth of non-small cell lung cancer (NSCLC).
In sections of human and experimental NSCLC, MGL was found in tumor cells and various cells of the TME including macrophages and stromal cells. Mice treated with the MGL inhibitor JZL184 as well as MGL knock-out (KO) mice exhibited a lower tumor burden than the controls. The reduction in tumor growth was accompanied by an increased number of CD8+ T cells and eosinophils. Naïve CD8+ T cells showed a shift toward more effector cells in MGL KOs and an increased expression of granzyme-B and interferon-γ, indicative of enhanced tumoricidal activity. 2-arachidonoyl glycerol (2-AG) was increased in tumors of MGL KO mice, and dose-dependently induced differentiation and migration of CD8+ T cells as well as migration and activation of eosinophils in vitro.
Our results suggest that next to cancer cell-derived MGL, TME cells expressing MGL are responsible for maintaining a pro-tumorigenic environment in tumors of NSCLC.
The evaluation of pharmacological data using machine learning requires high data quality. Therefore, data preprocessing, that is, cleaning analytical laboratory errors, replacing missing values or outliers, and transforming data adequately before actual data analysis, is crucial. Because current tools available for this purpose often require programming skills, preprocessing tools with graphical user interfaces that can be used interactively are needed. In collaboration between data scientists and experts in bioanalytical diagnostics, a graphical software package for data preprocessing called pguIMP is proposed, which contains a fixed sequence of preprocessing steps to enable reproducible interactive data preprocessing. As an R-based package, it also allows direct integration into this data science environment without requiring any programming knowledge. The implementation of contemporary data processing methods, including machine-learning-based imputation techniques, ensures the generation of corrected and cleaned bioanalytical data sets that preserve data structures such as clusters better than is possible with classical methods. This was evaluated on bioanalytical data sets from lipidomics and drug research using k-nearest-neighbors-based imputation followed by k-means clustering and density-based spatial clustering of applications with noise. The R package provides a Shiny-based web interface designed to be easy to use for non–data analysis experts. It is demonstrated that the spectrum of methods provided is suitable as a standard pipeline for preprocessing bioanalytical data in biomedical research domains. The R package pguIMP is freely available at the comprehensive R archive network (https://cran.r-project.org/web/packages/pguIMP/index.html).
Depletion of the enzyme cofactor, tetrahydrobiopterin (BH4), in T-cells was shown to prevent their proliferation upon receptor stimulation in models of allergic inflammation in mice, suggesting that BH4 drives autoimmunity. Hence, the clinically available BH4 drug (sapropterin) might increase the risk of autoimmune diseases. The present study assessed the implications for multiple sclerosis (MS) as an exemplary CNS autoimmune disease. Plasma levels of biopterin were persistently low in MS patients and tended to be lower with high Expanded Disability Status Scale (EDSS). Instead, the bypass product, neopterin, was increased. The deregulation suggested that BH4 replenishment might further drive the immune response or beneficially restore the BH4 balances. To answer this question, mice were treated with sapropterin in immunization-evoked autoimmune encephalomyelitis (EAE), a model of multiple sclerosis. Sapropterin-treated mice had higher EAE disease scores associated with higher numbers of T-cells infiltrating the spinal cord, but normal T-cell subpopulations in spleen and blood. Mechanistically, sapropterin treatment was associated with increased plasma levels of long-chain ceramides and low levels of the poly-unsaturated fatty acid, linolenic acid (FA18:3). These lipid changes are known to contribute to disruptions of the blood–brain barrier in EAE mice. Indeed, RNA data analyses revealed upregulations of genes involved in ceramide synthesis in brain endothelial cells of EAE mice (LASS6/CERS6, LASS3/CERS3, UGCG, ELOVL6, and ELOVL4). The results support the view that BH4 fortifies autoimmune CNS disease, mechanistically involving lipid deregulations that are known to contribute to the EAE pathology.