Counting Missing Values in a Metabolite-Intensity Data Set for Measuring the Analytical Performance of a Metabolomics Platform

Salivary metabolomic identification of biomarker candidates for oral melanoma and oral squamous cell carcinoma in dogs

BACKGROUND

Oral melanoma (OM) and oral squamous cell carcinoma (OSCC) are frequently diagnosed in dogs, presenting a challenge in distinguishing them from benign oral tumors (BN). Salivary metabolomic biomarkers offer a practical solution because of saliva's direct contact with tumors and the noninvasive nature of collection.

OBJECTIVE

Assess the diversity and abundance of the salivary metabolome in dogs with BN, OM, and OSCC using amine/phenol submetabolome analysis and high-performance chemical isotope labeling liquid chromatography-mass spectrometry (CIL LC-MS).

ANIMALS

Study included 11 BN, 24 OM, 10 OSCC, and 20 healthy control dogs.

METHODS

Case-control cross-sectional study was conducted to assess salivary submetabolic profiles in dogs with BN, OM, and OSCC and healthy dogs. Samples were labeled with 12C-dansyl chloride and analyzed using CIL LC-MS targeted to amine- and phenol-containing metabolites for amine/phenol submetabolome analysis.

RESULTS

Distinct clusters and significant differences in metabolite concentrations were observed among the oral cancer, BN, and control groups. A total of 154 and 66 metabolites showed significantly altered concentrations, particularly in OM and OSCC, respectively, when compared with BN (Padj < .05). Potential metabolic biomarkers were identified for each cancer, including decreased concentrations of seryl-arginine and sarcosine in OSCC. Moreover, high-confidence putative metabolites were identified, including an increase in tryptophyl-threonine and a decrease in 1,2-dihydroxynapthalene-6-sulfonic acid and hydroxyprolyl-hydroxyproline for OM.

CONCLUSIONS AND CLINICAL IMPORTANCE

We identified high coverage of the amine/phenol submetabolome, including seryl-arginine, and sarcosine, in OSCC. Our findings emphasize the potential of these biomarkers for distinguishing between oral OSCC and BN in dogs.

Deciphering Aphanomyces euteiches-pea-biocontrol bacterium interactions through untargeted metabolomics

Aphanomyces euteiches causes root rot in pea, leading to significant yield losses. However, the metabolites involved in this pathosystem have not been thoroughly studied. This study aimed to fill this gap and explore mechanisms of bacterial suppression of A. euteiches via untargeted metabolomics using pea grown in a controlled environment. Chemical isotope labeling (CIL), followed by liquid chromatography-mass spectrometry (LC-MS), was used for metabolite separation and detection. Univariate and multivariate analyses showed clear separation of metabolites from pathogen-treated pea roots and roots from other treatments. A three-tier approach positively or putatively identified 5249 peak pairs or metabolites. Of these, 403 were positively identified in tier 1; 940 were putatively identified with high confidence in tier 2. There were substantial changes in amino acid pool, and fatty acid and phenylpropanoid pathway products. More metabolites, including salicylic and jasmonic acids, were upregulated than downregulated in A. euteiches-infected roots. 1-aminocyclopropane-1-carboxylic acid and 12-oxophytodienoic acid were upregulated in A. euteiches + bacterium-treated roots compared to A. euteiches-infected roots. A great number of metabolites were up- or down-regulated in response to A. euteiches infection compared with the control and A. euteiches + bacterium-treated plants. The results of this study could facilitate improved disease management.

Clinical metabolomics characteristics of diabetic kidney disease: A meta-analysis of 1875 cases with diabetic kidney disease and 4503 controls

AIMS

Diabetic Kidney Disease (DKD), one of the major complications of diabetes, is also a major cause of end-stage renal disease. Metabolomics can provide a unique metabolic profile of the disease and thus predict or diagnose the development of the disease. Therefore, this study summarises a more comprehensive set of clinical biomarkers related to DKD to identify functional metabolites significantly associated with the development of DKD and reveal their driving mechanisms for DKD.

MATERIALS AND METHODS

We searched PubMed, Embase, the Cochrane Library and Web of Science databases through October 2022. A meta-analysis was conducted on untargeted or targeted metabolomics research data based on the strategy of standardized mean differences and the process of ratio of means as the effect size, respectively. We compared the changes in metabolite levels between the DKD patients and the controls and explored the source of heterogeneity through subgroup analyses, sensitivity analysis and meta-regression analysis.

RESULTS

The 34 clinical-based metabolomics studies clarified the differential metabolites between DKD and controls, containing 4503 control subjects and 1875 patients with DKD. The results showed that a total of 60 common differential metabolites were found in both meta-analyses, of which 5 metabolites (p < 0.05) were identified as essential metabolites. Compared with the control group, metabolites glycine, aconitic acid, glycolic acid and uracil decreased significantly in DKD patients; cysteine was significantly higher. This indicates that amino acid metabolism, lipid metabolism and pyrimidine metabolism in DKD patients are disordered.

CONCLUSIONS

We have identified 5 metabolites and metabolic pathways related to DKD which can serve as biomarkers or targets for disease prevention and drug therapy.

Unraveling the potential of segment scan mass spectral acquisition for chemical isotope labeling LC-MS-based metabolome analysis: Performance assessment across different types of biological samples

BACKGROUND

Chemical isotope labeling (CIL) LC-MS is a powerful tool for metabolome analysis with high metabolomic coverage and quantification accuracy. In CIL LC-MS, the overall metabolite detection efficiency using Orbitrap MS can be further improved by employing a segment scan method where the full m/z range is divided into multiple segments for spectral acquisition with a significant increase in the in-spectrum dynamic range. Considering the metabolic complexity in different types of biological samples (e.g., feces, urine, serum/plasma, cell/tissue extracts, saliva, etc.), we report the development and evaluation of the segment scan method for metabolome analysis of different sample types.

RESULTS

It was found that sample complexity significantly influenced the performance of the segment scan method. In metabolically complex samples such as feces and urine, the method yielded a substantial increase (up to 94 %) in detected peak pairs or metabolites, compared to conventional full scan. Conversely, less complex samples like saliva exhibited more modest gains (approximately 25 %). Based on the observations, a 120-m/z segment scan method was determined as a routine approach for CIL LC-Orbitrap-MS-based metabolomics with good compatibility with different types of biological samples. For this method, a further investigation on relative quantification accuracy was done. The peak area ratios of 12C-/13-labeled metabolites were slightly reduced with 72%-84 % of peak pairs falling within the ±25 % range of the anticipated peak ratio of 1.0 among different samples, as opposed to 81%-90 % in the full scan, which was attributed to the inclusion of more low-abundance peak pairs within the narrow MS segments. However, the overall peak ratio measurement precision was not significantly affected by the segment scan.

SIGNIFICANCE AND NOVELTY

The segment scan method was found to be useful for CIL LC-Orbitrap-MS-based metabolome analysis of different types of samples with significant improvement in metabolite detectability (25-94 % increase), compared to the conventional full scan method.

HDPairFinder: A data processing platform for hydrogen/deuterium isotopic labeling-based nontargeted analysis of trace-level amino-containing chemicals in environmental water

The combination of hydrogen/deuterium (H/D) formaldehyde-based isotopic methyl labeling with solid-phase extraction and high-performance liquid chromatography-high resolution mass spectrometry (HPLC-HRMS) is a powerful analytical solution for nontargeted analysis of trace-level amino-containing chemicals in water samples. Given the huge amount of chemical information generated in HPLC-HRMS analysis, identifying all possible H/D-labeled amino chemicals presents a significant challenge in data processing. To address this, we designed a streamlined data processing pipeline that can automatically extract H/D-labeled amino chemicals from the raw HPLC-HRMS data with high accuracy and efficiency. First, we developed a cross-correlation algorithm to correct the retention time shift resulting from deuterium isotopic effects, which enables reliable pairing of H- and D-labeled peaks. Second, we implemented several bioinformatic solutions to remove false chemical features generated by in-source fragmentation, salt adduction, and natural 13C isotopes. Third, we used a data mining strategy to construct the AMINES library that consists of over 38,000 structure-disjointed primary and secondary amines to facilitate putative compound annotation. Finally, we integrated these modules into a freely available R program, HDPairFinder.R. The rationale of each module was justified and its performance tested using experimental H/D-labeled chemical standards and authentic water samples. We further demonstrated the application of HDPairFinder to effectively extract N-containing contaminants, thus enabling the monitoring of changes of primary and secondary N-compounds in authentic water samples. HDPairFinder is a reliable bioinformatic tool for rapid processing of H/D isotopic methyl labeling-based nontargeted analysis of water samples, and will facilitate a better understanding of N-containing chemical compounds in water.

Metabolomics profiling distinctively identified end-stage renal disease patients from chronic kidney disease patients

Chronic kidney disease (CKD) is a serious public health problem characterized by progressive kidney function loss leading to end-stage renal disease (ESRD) that demands dialysis or kidney transplantation. Early detection can prevent or delay progression to ESRD. The study aimed to gain new insights into the perturbed biochemical reactions and to identify novel distinct biomarkers between ESRD and CKD. Serum samples of 32 patients with ESRD (n = 13) and CKD (n = 19) were analyzed using chemical isotope labeling liquid chromatography-mass spectrometry metabolomics approach. A total of 193 metabolites were significantly altered in ESRD compared to CKD and were mainly involved in aminoacyl-tRNA biosynthesis, branched-chain amino acid (BCAA) biosynthesis, taurine metabolism, and tryptophan metabolism. Three kynurenine derivatives, namely, 2-aminobenzoic acid, xanthurenic acid, and hydroxypicolinic acid were upregulated in ESRD compared to CKD due to the significant decrease in glomerular filtration rate with the progression of CKD to ESRD. N-Hydroxy-isoleucine, 2-aminobenzoic acid, and picolinic acid yielded AUC > 0.99 when analyzed using Receiver Operating Characteristic (ROC) analysis. Our findings suggest that inhibiting the kynurenine pathway might be a promising target to delay CKD progression and that metabolites with high discriminative ability might serve as potential prognostic biomarkers to monitor the progression of CKD to ESRD or used in combination with current markers to indicate the status of kidney damage better.

Evaluating LC-HRMS metabolomics data processing software using FAIR principles for research software

BACKGROUND

Liquid chromatography-high resolution mass spectrometry (LC-HRMS) is a popular approach for metabolomics data acquisition and requires many data processing software tools. The FAIR Principles - Findability, Accessibility, Interoperability, and Reusability - were proposed to promote open science and reusable data management, and to maximize the benefit obtained from contemporary and formal scholarly digital publishing. More recently, the FAIR principles were extended to include Research Software (FAIR4RS).

AIM OF REVIEW

This study facilitates open science in metabolomics by providing an implementation solution for adopting FAIR4RS in the LC-HRMS metabolomics data processing software. We believe our evaluation guidelines and results can help improve the FAIRness of research software.

KEY SCIENTIFIC CONCEPTS OF REVIEW

We evaluated 124 LC-HRMS metabolomics data processing software obtained from a systematic review and selected 61 software for detailed evaluation using FAIR4RS-related criteria, which were extracted from the literature along with internal discussions. We assigned each criterion one or more FAIR4RS categories through discussion. The minimum, median, and maximum percentages of criteria fulfillment of software were 21.6%, 47.7%, and 71.8%. Statistical analysis revealed no significant improvement in FAIRness over time. We identified four criteria covering multiple FAIR4RS categories but had a low %fulfillment: (1) No software had semantic annotation of key information; (2) only 6.3% of evaluated software were registered to Zenodo and received DOIs; (3) only 14.5% of selected software had official software containerization or virtual machine; (4) only 16.7% of evaluated software had a fully documented functions in code. According to the results, we discussed improvement strategies and future directions.

Instrument-type effects on chemical isotope labeling LC-MS metabolome analysis: Quadrupole time-of-flight MS vs. Orbitrap MS

Chemical isotope labeling (CIL) LC-MS is a powerful tool for metabolome analysis with markedly improved metabolomic coverage and quantification accuracy over the conventional LC-MS technique. In addition, with differential isotope labeling, each labeled metabolite is detected as a peak pair in the mass spectra, offering the possibility of differentiating true metabolite peaks from the singlet noise or background peaks. In this study, we examined the effects of instrument type on the detectability of true metabolites with a focus on the comparison of quadrupole time-of-flight (QTOF) and Orbitrap mass spectrometers. Using the same ultra-high-performance liquid chromatography setup and optimized running conditions for QTOF and Orbitrap, we compared the total number of peak pairs detected and identified from the two instruments using human urine and serum as the test samples. Many common peak pairs were detected from the two instruments; however, there were a significant number of unique peak pairs detected in each type of instrument. By combining the datasets obtained using QTOF and Orbitrap, the total number of peak pairs detected could be significantly increased. We also examined the effect of mass resolving power on peak pair detection in Orbitrap (60,000 vs. 120,000 resolution). The observed differences in peak pair detectability were much less than those of QTOF vs. Orbitrap. However, the type of peak pairs detected using different resolutions could be somewhat different, offering the possibility of increasing the overall number of peak pairs by combining the two datasets obtained at two different resolutions. The results from this study clearly indicate that instrument type can have a profound effect on metabolite detection in CIL LC-MS. Therefore, comparison of metabolome data generated using different instruments needs to be carefully done. Moreover, future research (e.g., hardware modifications) is warranted to minimize the differences in order to generate more reproducible metabolome data from different types of instruments.

Segment Scan Mass Spectral Acquisition for Increasing the Metabolite Detectability in Chemical Isotope Labeling Liquid Chromatography-Mass Spectrometry Metabolome Analysis

We report a segmented spectrum scan method using Orbitrap MS in chemical isotope labeling (CIL) liquid chromatography-mass spectrometry (LC-MS) for improving the metabolite detection efficiency. In this method, the full m/z range is divided into multiple segments with the scanning of each segment to produce multiple narrow-range spectra during the LC data acquisition. These segmented spectra are separately processed to extract the peak pair information with each peak pair arising from a differentially labeled metabolite in the analysis of a mixture of ¹³C and ¹²C reagent-labeled samples. The sublists of peak pairs are merged to form the final peak pair list from the LC-MS run. Various experimental conditions, including automatic gain control (AGC) values, mass resolutions, segment m/z widths, number of segments, and total data acquisition time in the LC run, were examined to arrive at an optimal setting in the segment scan for increasing the number of detectable metabolites while maintaining the same analysis time as in the full scan. The optimal method used a segment width of 120 m/z with 60k resolution for a 16 min CIL LC-MS run. Using dansyl-labeled human urine samples as an example, we demonstrated that this method could detect 5867 peak pairs or metabolites (not features), compared to 3765 peak pairs detectable in a full scan, representing a 56% gain. Out of 5867 peak pairs, 5575 (95.0%) could be identified or mass-matched. The relative quantification accuracy was slightly reduced (81% peak pairs were within ±25% of the expected peak ratio of 1.0 in full, compared to 87% in the full scan) due to the inclusion of more low-abundance peak pairs in the segment scan. The peak ratio measurement precision was not significantly affected by the segment scan. We also showed the increase of the peak pair number detectable from 3843 in the full scan to 7273 (89% gain) using the Orbitrap operated at 120k resolution with a 60 m/z segment width when multiple repeat sample injections were used. Thus, segment scan Orbitrap MS is an enabling method for detecting coeluting metabolites in CIL LC-MS for increasing the metabolomic coverage.

Development of a method for dansylation of metabolites using organic solvent-compatible buffer systems for amine/phenol submetabolome analysis

Metabolomics, which serves as a readout of biological processes and diseases monitoring, is an informative research area for disease biomarker discovery and systems biology studies. In particular, reversed-phase liquid chromatography-mass spectrometry (RPLC-MS) has become a powerful and popular tool for metabolomics analysis, enabling the detection of most metabolites. Very polar and ionic metabolites, however, are less easily detected because of their poor retention in RP columns. Dansylation of metabolites simplifies the sub-metabolome analysis by reducing its complexity and increasing both hydrophobicity and ionization ability. However, the various metabolite concentrations in clinical samples have a wide dynamic range with highly individual variation in total metabolite amount, such as in saliva. The bicarbonate buffer typically used in dansylation labeling reactions induces solvent stratification, resulting in poor reproducibility, selective sample loss and an increase in false-determined metabolite peaks. In this study, we optimized the dansylation protocol for samples with wide concentration range of metabolites, utilizing diisopropylethylamine (DIPEA) or tri-ethylamine (TEA) in place of bicarbonate buffer, and presented the results of a systemic investigation of the influences of individual processes involved on the overall performance of the protocol. In addition to achieving high reproducibility, substitution of DIPEA or TEA buffer resulted in similar labeling efficiency of most metabolites and more efficient labeling of some metabolites with a higher pK_a. With this improvement, compounds that are only present in samples in trace amounts can be detected, and more comprehensive metabolomics profiles can be acquired for biomarker discovery or pathway analysis, making it possible to analyze clinical samples with limited amounts of metabolites.

Time-series eco-metabolomics reveals extensive reshuffling in metabolome during transition from cold acclimation to de-acclimation in an alpine shrub

Recording environmentally induced variations in the metabolome in plants can be a promising approach for understanding the complex patterns of metabolic regulation and their eco-physiological consequences. Here, we studied metabolome-wide changes and eco-physiological adjustments occurring across the year at high elevation environments in the leaf tissue of Rhododendron anthopogon, an alpine evergreen shrub of the Himalaya. New leaves of R. anthopogon appear after the snow-melt and remain intact even when the plants get covered under snow (November-June). During this whole period, they may undergo several physiological and biochemical adjustments in response to fluctuating temperatures and light conditions. To understand these changes, we analyzed eco-physiological traits, that is, freezing resistance, dry matter content and % of nitrogen and the overall metabolome across 10 different time-points, from August until the snowfall in November 2017, and then from June to August 2018. As anticipated, the freezing resistance increased toward the onset of winters. The leaf tissues exhibited a complete reshuffling of the metabolome during the growth cycle and time-points segregated into four clusters directly correlating with distinct phases of acclimation: non-acclimation (August 22, 2017; August 14, 2018), early cold acclimation (September 12, September 29, October 11, 2017), late cold acclimation (October 23, November 4, 2017), and de-acclimation (June 15, June 28, July 14, 2018). Cold acclimation involved metabolic progression (101 metabolites) with an increase of up to 19.4-fold (gentiobiose), whereas de-acclimation showed regression (120 metabolites) with a decrease of up to 30-fold (sucrose). The changes in the metabolome during de-acclimation were maximum and were not just a reversal of cold acclimation. Our results provided insights into the direction and magnitude of physiological changes in Rhododendron anthopogon that occurred across the year.

Metabolomics Profiling of Cystic Renal Disease towards Biomarker Discovery

Simple Summary

Cystic renal disease (CRD) is a group of diseases characterized by abnormal sacs, or cysts, in the kidneys. CRD can be detected using certain imaging modalities (i.e., ultrasound). Patients with CRD might be symptoms-free, while others can show symptoms long after cysts development. Although these cysts represent structural changes, we hypothesized that they have an underlying biochemical alteration. If so, this would open the floor for potential biomarker discovery, which would aid in CRD diagnosis and, possibly, treatment. On that basis, this study focuses on identifying biomarkers for CRD. To achieve that, we employed a metabolomics-based approach to identify intermediate molecules inside the cells that are byproducts of biochemical reactions. We used dry blood spots and serum samples of CRD patients and healthy controls to study the differences in their metabolomic profile. Our results suggest that certain metabolites, including uridine diphosphate, cystine-5-diphosphate, and morpholine, are potential biomarkers for CRD. The affected biochemical pathways in CRD include aminoacyl-tRNA biosynthesis, purine, pyrimidine, glutathione, TCA cycle, and some amino acid metabolism. These preliminary results could be the starting point for possible diagnostic and therapeutic approaches for CRD in the future.

Abstract

Cystic renal disease (CRD) comprises a heterogeneous group of genetic and acquired disorders. The cystic lesions are detected through imaging, either incidentally or after symptoms develop, due to an underlying disease process. In this study, we aim to study the metabolomic profiles of CRD patients for potential disease-specific biomarkers using unlabeled and labeled metabolomics using low and high-resolution mass spectrometry (MS), respectively. Dried-blood spot (DBS) and serum samples, collected from CRD patients and healthy controls, were analyzed using the unlabeled and labeled method. The metabolomics profiles for both sets of samples and groups were collected, and their data were processed using the lab’s standard protocol. The univariate analysis showed (FDR p < 0.05 and fold change 2) was significant to show a group of potential biomarkers for CRD discovery, including uridine diphosphate, cystine-5-diphosphate, and morpholine. Several pathways were involved in CRD patients based on the metabolic profile, including aminoacyl-tRNA biosynthesis, purine and pyrimidine, glutathione, TCA cycle, and some amino acid metabolism (alanine, aspartate and glutamate, arginine and tryptophan), which have the most impact. In conclusion, early CRD detection and treatment is possible using a metabolomics approach that targets alanine, aspartate, and glutamate pathway metabolites.

Metabolic profile of irradiated whole blood by chemical isotope-labeling liquid chromatography-mass spectrometry

Irradiated blood is a new type of blood product used to prevent transfusion-associated graft-versus-host disease. However, the effects of irradiation on the metabolism of plasma, red blood cells (RBCs), and peripheral blood mononuclear cells (PBMCs) are largely unknown. We developed a workflow for testing metabolic changes in whole blood to determine the impact of irradiation by chemical isotope labeling liquid chromatography-mass spectrometry (CIL LC-MS). Blood parameters, PBMC proliferation and apoptosis were examined before and after irradiation. Next, the amine/phenol metabolites in the blood components were assayed by ¹²C- and¹³C-dansylation labeling LC-MS. We identified 1654, 1730, and 1666 peak pairs in plasma, RBCs, and PBMCs, respectively. We screened out 367, 177, and 219 significant metabolites in plasma, RBCs, and PBMCs, respectively, by principle component analyses, volcano plots, and Venn plots. Metabolic pathway analyses showed that irradiation modulated taurine and hypotaurine metabolism in plasma and purine metabolism in RBCs and PBMCs. Changes in potential biomarkers, including an increase in hypoxanthine level and a decrease in adenine level, may be related to the dysfunction of DNA synthesis in PBMCs. The decreased AMP level in RBCs may interfere with RBC storage lesions. Our research provides a more comprehensive perspective on blood metabolism associated with irradiation.

Computational Variation: An Underinvestigated Quantitative Variability Caused by Automated Data Processing in Untargeted Metabolomics

Computational tools are commonly used in untargeted metabolomics to automatically extract metabolic features from liquid chromatography-mass spectrometry (LC-MS) raw data. However, due to the incapability of software to accurately determine chromatographic peak heights/areas for features with poor chromatographic peak shape, automated data processing in untargeted metabolomics faces additional quantitative variation (i.e., computational variation) besides the well-recognized analytical and biological variations. In this work, using multiple biological samples, we investigated how experimental factors, including sample concentrations, LC separation columns, and data processing programs, contribute to computational variation. For example, we found that the peak height (PH)-based quantification is more precise when MS-DIAL was used for data processing. We further systematically compared the different patterns of computational variation between PH- and peak area (PA)-based quantitative measurements. Our results suggest that the magnitude of computational variation is highly consistent at a given concentration. Hence, we proposed a quality control (QC) sample-based correction workflow to minimize computational variation by automatically selecting PH or PA-based measurement for each intensity value. This bioinformatic solution was demonstrated in a metabolomic comparison of leukemia patients before and after chemotherapy. Our novel workflow can be effectively applied on 652 out of 915 metabolic features, and over 31% (206 out of 652) of corrected features showed distinctly changed statistical significance. Overall, this work highlights computational variation, a considerable but underinvestigated quantitative variability in omics-scale quantitative analyses. In addition, the proposed bioinformatic solution can minimize computational variation, thus providing a more confident statistical comparison among biological groups in quantitative metabolomics.

Obesity Connected Metabolic Changes in Type 2 Diabetic Patients Treated With Metformin

Metformin is widely used in the treatment of Type 2 Diabetes Mellitus (T2DM). However, it is known to have beneficial effects in many other conditions, including obesity and cancer. In this study, we aimed to investigate the metabolic effect of metformin in T2DM and its impact on obesity. A mass spectrometry (MS)-based metabolomics approach was used to analyze samples from two cohorts, including healthy lean and obese control, and lean as well as obese T2DM patients on metformin regimen in the last 6 months. The results show a clear group separation and sample clustering between the study groups due to both T2DM and metformin administration. Seventy-one metabolites were dysregulated in diabetic obese patients (30 up-regulated and 41 down-regulated), and their levels were unchanged with metformin administration. However, 30 metabolites were dysregulated (21 were up-regulated and 9 were down-regulated) and then restored to obese control levels by metformin administration in obese diabetic patients. Furthermore, in obese diabetic patients, the level of 10 metabolites was dysregulated only after metformin administration. Most of these dysregulated metabolites were dipeptides, aliphatic amino acids, nucleic acid derivatives, and urea cycle components. The metabolic pattern of 62 metabolites was persistent, and their levels were affected by neither T2DM nor metformin in obesity. Interestingly, 9 metabolites were significantly dysregulated between lean and obese cohorts due to T2DM and metformin regardless of the obesity status. These include arginine, citrulline, guanidoacetic acid, proline, alanine, taurine, 5-hydroxyindoleacetic acid, and 5-hydroxymethyluracil. Understanding the metabolic alterations taking place upon metformin treatment would shed light on possible molecular targets of metformin, especially in conditions like T2DM and obesity.

High-coverage quantitative liver metabolomics using perfused and non-perfused liver tissues

Comprehensive analysis of the liver metabolome can be very useful for discovering disease biomarkers and studying diseases, especially liver-related diseases. However, the presence of a relatively large amount of blood in liver tissue may have a profound effect on liver tissue metabolome analysis. We designed a study to address this issue in order to develop a liver metabolomics workflow based on high-coverage quantitative metabolome analysis using differential chemical isotope labeling (CIL) LC-MS. In the first set of experiments, we compared the metabolomes of mouse serum, non-perfused liver, and perfused liver without and with varying amounts of blood added. We found that there was a significant metabolome difference between the perfused liver and non-perfused liver. To illustrate the effects of perfusion conditions on tissue metabolome analysis, we analyzed the mouse livers that were subjected to perfusion under two different conditions. We found that ice-cold temperature perfusion led to less change of the liver metabolome, compared to room temperature perfusion; however, there was still a significant metabolome difference between the ice-cold-perfused liver and the non-perfused liver. Finally, we applied the method to a chemical (carbon tetrachloride) exposure liver injury model to examine the effects of blood in liver on the detection of significantly changed metabolites in two comparative groups of mice. Using multivariate and univariate analyses of the serum and liver metabolomes of control and diseased mice, we detected many unique significant metabolites in serum as well as in liver. This work demonstrates that perfusion can alter the liver metabolome significantly. Therefore, we recommend the use of non-perfused liver for high-coverage liver metabolomics.

Novel plasma metabolite markers of attention-deficit/hyperactivity disorder identified using high-performance chemical isotope labelling-based liquid chromatography-mass spectrometry

OBJECTIVES

Metabolites are the intermediate and final products of biological processes and ultimately reflect the responses of these processes to genetic regulation and environmental perturbations, including those involved in attention deficit/hyperactivity disorder (ADHD).

METHODS

We identified a quantitative profile of plasma metabolites in 58 ADHD patients (mean age 9.0 years, 77.6% males) and 38 healthy control subjects (mean age 10.2 years, 55.3% males) using the high-performance chemical isotope labelling (CIL)-based liquid chromatography-mass spectrometry (LC-MS). Using a volcano plot and orthogonal projections to latent structure-discriminant analysis (OPLS-DA), we determined nine metabolites with differentially expressed levels in ADHD plasma samples.

RESULTS

Compared to the control group, the plasma levels of guanosine, O-phosphoethanolamine, phenyl-leucine, hypoxanthine, 4-aminohippuric acid, 5-hydroxylysine, and L-cystine appeared increased in the ADHD patients, whilegentisic acid and tryptophyl-phenylalanine were down-regulated in the patients with ADHD. We found that the plasma levels of all nine metabolites were able to discriminate the ADHD group from the control group. Levels of O-phosphoethanolamine, 4-aminohippuric acid, 5-hydroxylysine, L-cystine, tryptophyl-phenylalanine, and gentisic acid were significantly correlated with clinical ADHD symptoms.

CONCLUSIONS

This study is the first to use the CIL-based LC-MS to profile ADHD plasma metabolites, and we identified nine novel metabolite biomarkers of ADHD.

Untargeted Urinary Metabolomics and Children's Exposure to Secondhand Smoke: The Influence of Individual Differences

Children’s exposure to secondhand smoke (SHS) is a severe public health problem. There is still a lack of evidence regarding panoramic changes in children’s urinary metabolites induced by their involuntary exposure to SHS, and few studies have considered individual differences. This study aims to clarify the SHS-induced changes in urinary metabolites in preschool children by using cross-sectional and longitudinal metabolomics analyses. Urinary metabolites were quantified by using untargeted ultra high-performance liquid chromatography-mass spectrometry (UPLC^(c)-MS/MS). Urine cotinine-measured SHS exposure was examined to determine the exposure level. A cross-sectional study including 17 children in a low-exposure group, 17 in a medium-exposure group, and 17 in a high-exposure group was first conducted. Then, a before–after study in the cohort of children was carried out before and two months after smoking-cessation intervention for family smokers. A total of 43 metabolites were discovered to be related to SHS exposure in children in the cross-sectional analysis (false discovery rate (FDR) corrected p < 0.05, variable importance in the projection (VIP) > 1.0). Only three metabolites were confirmed to be positively associated with children’s exposure to SHS (FDR corrected p < 0.05) in a follow-up longitudinal analysis, including kynurenine, tyrosyl-tryptophan, and 1-(3-pyridinyl)-1,4-butanediol, the latter of which belongs to carbonyl compounds, peptides, and pyridines. Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis indicated that 1-(3-pyridinyl)-1,4-butanediol and kynurenine were significantly enriched in xenobiotic metabolism by cytochrome P450 (p = 0.040) and tryptophan metabolism (p = 0.030), respectively. These findings provide new insights into the pathophysiological mechanism of SHS and indicate the influence of individual differences in SHS-induced changes in urinary metabolites in children.

Distinctive metabolic profiles between Cystic Fibrosis mutational subclasses and lung function

INTRODUCTION

Cystic fibrosis (CF) is a lethal multisystemic disease of a monogenic origin with numerous mutations. Functional defects in the cystic fibrosis transmembrane conductance receptor (CFTR) protein based on these mutations are categorised into distinct classes having different clinical presentations and disease severity.

OBJECTIVES

The present study aimed to create a comprehensive metabolomic profile of altered metabolites in patients with CF, among different classes and in relation to lung function.

METHODS

A chemical isotope labeling liquid chromatography-mass spectrometry metabolomics was used to study the serum metabolic profiles of young and adult CF (n = 39) patients and healthy controls (n = 30). Comparisons were made at three levels, CF vs. controls, among mutational classes of CF, between CF class III and IV, and correlated the lung function findings.

RESULTS

A distinctive metabolic profile was observed in the three analyses. 78, 20, and 13 significantly differentially dysregulated metabolites were identified in the patients with CF, among the different classes and between class III and IV, respectively. The significantly identified metabolites included amino acids, di-, and tri-peptides, glutathione, glutamine, glutamate, and arginine metabolism. The top significant metabolites include 1-Aminopropan-2-ol, ophthalmate, serotonin, cystathionine, and gamma-glutamylglutamic acid. Lung function represented by an above-average FEV1% level was associated with decreased glutamic acid and increased guanosine levels.

CONCLUSION

Metabolomic profiling identified alterations in different amino acids and dipeptides, involved in regulating glutathione metabolism. Two metabolites, 3,4-dihydroxymandelate-3-O-sulfate and 5-Aminopentanoic acid, were identified in common between the three anlayses and may represent as highly sensitive biomarkers for CF.

Distinctive Metabolomics Patterns Associated With Insulin Resistance and Type 2 Diabetes Mellitus

Obesity is associated with an increased risk of insulin resistance (IR) and type 2 diabetes mellitus (T2DM) which is a multi-factorial disease associated with a dysregulated metabolism and can be prevented in pre-diabetic individuals with impaired glucose tolerance. A metabolomic approach emphasizing metabolic pathways is critical to our understanding of this heterogeneous disease. This study aimed to characterize the serum metabolomic fingerprint and multi-metabolite signatures associated with IR and T2DM. Here, we have used untargeted high-performance chemical isotope labeling (CIL) liquid chromatography-mass spectrometry (LC-MS) to identify candidate biomarkers of IR and T2DM in sera from 30 adults of normal weight, 26 obese adults, and 16 adults newly diagnosed with T2DM. Among the 3633 peak pairs detected, 62% were either identified or matched. A group of 78 metabolites were up-regulated and 111 metabolites were down-regulated comparing obese to lean group while 459 metabolites were up-regulated and 166 metabolites were down-regulated comparing T2DM to obese groups. Several metabolites were identified as IR potential biomarkers, including amino acids (Asn, Gln, and His), methionine (Met) sulfoxide, 2-methyl-3-hydroxy-5-formylpyridine-4-carboxylate, serotonin, L-2-amino-3-oxobutanoic acid, and 4,6-dihydroxyquinoline. T2DM was associated with dysregulation of 42 metabolites, including amino acids, amino acids metabolites, and dipeptides. In conclusion, these pilot data have identified IR and T2DM metabolomics panels as potential novel biomarkers of IR and identified metabolites associated with T2DM, with possible diagnostic and therapeutic applications. Further studies to confirm these associations in prospective cohorts are warranted.

Metabolomic profiling of parapneumonic effusion reveals a regulatory role of dipeptides in interleukin-8 production in neutrophil-like cells

Bacterial pneumonia is a lethal condition, and approximately 40% of bacterial pneumonia patients experience parapneumonic effusion (PPE). Based on the severity of inflammation, PPEs can be categorized as early-stage uncomplicated PPE (UPPE), advanced-stage complicated PPE (CPPE) and, most seriously, thoracic empyema. Appropriate antibiotic treatment at the early stage of PPE can prevent PPE progression and reduce mortality, indicating that understanding PPE generation and components can help researchers develop corresponding treatment strategies for PPE. To this end, metabolomes of 73 PPE (38 UPPE and 35 CPPE samples) and 30 malignant pleural effusion (MPE) samples were profiled with differential ¹²C₂-/¹³C₂-isotope dansylation labeling-based mass spectrometry. We found that PPE is characterized by elevated levels of dipeptides, especially for PPEs at advanced stages. Furthermore, with integrated proteomic and transcriptomic analyses of PPEs, the levels of dipeptides were strongly associated with the production of interleukin-8 (IL-8), an inflammation-associated cytokine. The production of IL-8 indeed increased upon the treatment of HL-60-derived neutrophilic cells with dipeptides, Gly-Val and Gly-Tyr. Our findings help to elucidate the metabolic perturbations present in PPE and indicate for the first time that dipeptides may be involved in the immune regulation observed during PPE progression.

Dexamethasone-Induced Perturbations in Tissue Metabolomics Revealed by Chemical Isotope Labeling LC-MS analysis

Dexamethasone (Dex) is a synthetic glucocorticoid (GC) drug commonly used clinically for the treatment of several inflammatory and immune-mediated diseases. Despite its broad range of indications, the long-term use of Dex is known to be associated with specific abnormalities in several tissues and organs. In this study, the metabolomic effects on five different organs induced by the chronic administration of Dex in the Sprague–Dawley rat model were investigated using the chemical isotope labeling liquid chromatography-mass spectrometry (CIL LC-MS) platform, which targets the amine/phenol submetabolomes. Compared to controls, a prolonged intake of Dex resulted in significant perturbations in the levels of 492, 442, 300, 186, and 105 metabolites in the brain, skeletal muscle, liver, kidney, and heart tissues, respectively. The positively identified metabolites were mapped to diverse molecular pathways in different organs. In the brain, perturbations in protein biosynthesis, amino acid metabolism, and monoamine neurotransmitter synthesis were identified, while in the heart, pyrimidine metabolism and branched amino acid biosynthesis were the most significantly impaired pathways. In the kidney, several amino acid pathways were dysregulated, which reflected impairments in several biological functions, including gluconeogenesis and ureagenesis. Beta-alanine metabolism and uridine homeostasis were profoundly affected in liver tissues, whereas alterations of glutathione, arginine, glutamine, and nitrogen metabolism pointed to the modulation of muscle metabolism and disturbances in energy production and muscle mass in skeletal muscle. The differential expression of multiple dipeptides was most significant in the liver (down-regulated), brain (up-regulation), and kidney tissues, but not in the heart or skeletal muscle tissues. The identification of clinically relevant pathways provides holistic insights into the tissue molecular responses induced by Dex and understanding of the underlying mechanisms associated with their side effects. Our data suggest a potential role for glutathione supplementation and dipeptide modulators as novel therapeutic interventions to mitigate the side effects induced by Dex therapy.

Lopata A, Li L, Dasouki M, Rahman AMA. Metabolomics Distinguishes DOCK8 Deficiency from Atopic Dermatitis: Towards a Biomarker Discovery

Bi-allelic mutations in the dedicator of cytokinesis 8 (DOCK8) are responsible for a rare autosomal recessive primary combined immunodeficiency syndrome, characterized by atopic dermatitis, elevated serum Immunoglobulin E (IgE) levels, recurrent severe cutaneous viral infections, autoimmunity, and predisposition to malignancy. The molecular link between DOCK8 deficiency and atopic skin inflammation remains unknown. Severe atopic dermatitis (AD) and DOCK8 deficiency share some clinical symptoms, including eczema, eosinophilia, and increased serum IgE levels. Increased serum IgE levels are characteristic of, but not specific to allergic diseases. Herein, we aimed to study the metabolomic profiles of DOCK8-deficient and AD patients for potential disease-specific biomarkers using chemical isotope labeling liquid chromatography-mass spectrometry (CIL LC-MS). Serum samples were collected from DOCK8-deficient (n = 10) and AD (n = 9) patients. Metabolomics profiling using CIL LC-MS was performed on patient samples and compared to unrelated healthy controls (n = 33). Seven metabolites were positively identified, distinguishing DOCK8-deficient from AD patients. Aspartic acid and 3-hydroxyanthranillic acid (3HAA, a tryptophan degradation pathway intermediate) were up-regulated in DOCK8 deficiency, whereas hypotaurine, leucyl-phenylalanine, glycyl-phenylalanine, and guanosine were down-regulated. Hypotaurine, 3-hydroxyanthranillic acid, and glycyl-phenyalanine were identified as potential biomarkers specific to DOCK8 deficiency. Aspartate availability has been recently implicated as a limiting metabolite for tumour growth and 3HAA; furthermore, other tryptophan metabolism pathway-related molecules have been considered as potential novel targets for cancer therapy. Taken together, perturbations in tryptophan degradation and increased availability of aspartate suggest a link of DOCK8 deficiency to oncogenesis. Additionally, perturbations in taurine and dipeptides metabolism suggest altered antixidation and cell signaling states in DOCK8 deficiency. Further studies examining the mechanisms underlying these observations are necessary.

Mass Accuracy Check Using Common Background Peaks for Improving Metabolome Data Quality in Chemical Isotope Labeling LC-MS

Chemical isotope labeling (CIL) LC-MS is a highly sensitive and quantitative method for metabolome analysis. Because of a large number of peaks detectable in a sample and the need of running many samples in a metabolomics project, any significant change in mass measurement accuracy during the whole period of running samples can adversely affect the downstream peak alignment and quantitative analysis. Herein, we report a rapid method to check the mass accuracy of individual spectra in each CIL LC-MS run in order to flag up any run containing spectra with accuracy drift that falls outside the expected error. The flagged run may be re-run or discarded before merging with other runs for peak alignment and analysis. This method is based on the observation that some background signals are commonly detected in almost all spectra collected in CIL LC-MS runs. A mass accuracy check (MAC) software program has been developed to first find the common background mass peaks and then use them as mass references to calculate any mass shifts over the course of multiple sample runs. Using a metabolome dataset of 324 human cerebrospinal fluid (CSF) samples and 35 quality control (QC) samples produced by CIL LC-MS, we show that this accuracy check method can streamline the initial raw data processing for downstream analysis in metabolomics.

Metabolomics Analyses of Saliva Detect Novel Biomarkers of Alzheimer's Disease

Using a non-invasive biofluid (saliva), we apply a powerful metabolomics workflow for unbiased biomarker discovery in Alzheimer's disease (AD). We profile and differentiate Cognitively Normal (CN), Mild Cognitive Impairment (MCI), and AD groups. The workflow involves differential chemical isotope labeling liquid chromatography mass spectrometry using dansylation derivatization for in-depth profiling of the amine/phenol submetabolome. The total sample (N = 109) was divided in to the Discovery Phase (DP) (n = 82; 35 CN, 25 MCI, 22 AD) and a provisional Validation Phase (VP) (n = 27; 10 CN, 10 MCI, 7 AD). In DP we detected 6,230 metabolites. Pairwise analyses confirmed biomarkers for AD versus CN (63), AD versus MCI (47), and MCI versus CN (2). We then determined the top discriminating biomarkers and diagnostic panels. A 3-metabolite panel distinguished AD from CN and MCI (DP and VP: Area Under the Curve [AUC] = 1.000). The MCI and CN groups were best discriminated with a 2-metabolite panel (DP: AUC = 0.779; VP: AUC = 0.889). In addition, using positively confirmed metabolites, we were able to distinguish AD from CN and MCI with good diagnostic performance (AUC > 0.8). Saliva is a promising biofluid for both unbiased and targeted AD biomarker discovery and mechanism detection. Given its wide availability and convenient accessibility, saliva is a biofluid that can promote diversification of global AD biomarker research.

MSC-triggered metabolomic alterations in liver-resident immune cells isolated from CCl4-induced mouse ALI model

Clinical trials testing mesenchymal stem cell (MSC) as a cellular remedy for acute liver injury (ALI) are underway, but its underlying mechanism has not been thoroughly scrutinized. We highlight that the metabolomic profile of the liver-resident immune cells is significantly altered after MSC administration; its potential correlation with ALI remission is discussed in this study. C57BL/6 mice are randomly divided into three groups: the sham group, MSC-treated ALI group and PBS-treated ALI group; acute liver injury is induced by intraperitoneal injection of carbon tetrachloride. A high-performance chemical isotope labeling liquid chromatography-mass spectrometry (CIL LC-MS) is exploited to profile amine, phenol and carbonyl submetabolome of the liver-resident immune cells in different treatments. 4295 peak pairs are quantified and 2461 peak pairs are further identified in zero-reaction and one-reaction libraries. Clear separation of the three groups is observed in the global PCA and OPLS-DA analyses. We identified 256 metabolites to be candidate biomarkers for ALI-activated immunity and 114 metabolites to be candidate biomarkers for MSC-modulated immunity. Ariginine, aspartate and glutamate metabolism are most affected in both cases, with eight significantly regulated metabolites as joints (glutamic-gamma-semialdehyde, aspartate acid, glutamate acid, gamma-Aminobutyric acidorinithine, 2-keto-glutaramic acid, N-acetylornithine, citrulline and ornithine). These findings shed new light on the therapeutic benefit of immune modulation during ALI rescue. It needs to be further investigated whether exogenous supply of certain metabolites will exert a profound impact on the metabolic network, crosstalking with immune responses and modulating ALI prognosis.

A Single Visualization Technique for Displaying Multiple Metabolite-Phenotype Associations

To assist with management and interpretation of human metabolomics data, which are rapidly increasing in quantity and complexity, we need better visualization tools. Using a dataset of several hundred metabolite measures profiled in a cohort of ~1500 individuals sampled from a population-based community study, we performed association analyses with eight demographic and clinical traits and outcomes. We compared frequently used existing graphical approaches with a novel ‘rain plot’ approach to display the results of these analyses. The ‘rain plot’ combines features of a raindrop plot and a conventional heatmap to convey results of multiple association analyses. A rain plot can simultaneously indicate effect size, directionality, and statistical significance of associations between metabolites and several traits. This approach enables visual comparison features of all metabolites examined with a given trait. The rain plot extends prior approaches and offers complementary information for data interpretation. Additional work is needed in data visualizations for metabolomics to assist investigators in the process of understanding and convey large-scale analysis results effectively, feasibly, and practically.

Metabolomic and Immunological Profiling of Respiratory Syncytial Virus Infection after Intranasal Immunization with a Subunit Vaccine Candidate

Respiratory syncytial virus (RSV) is a significant cause of mortality and morbidity in infants, the elderly, immunocompromised individuals, and patients with congenital heart diseases. Despite extensive efforts, a vaccine against RSV is still not available. We have previously reported the development of a subunit vaccine (ΔF/TriAdj) composed of a truncated version of the fusion protein (ΔF) and a polymer-based combination adjuvant (TriAdj). We compared inflammatory responses of ΔF/TriAdj-vaccinated and unvaccinated mice following intranasal challenge with RSV. Rapid and early inflammatory responses were observed in lung samples from both groups but modulated in the vaccinated group 7 days after the viral challenge. The underlying mechanism of action of ΔF/TriAdj was further studied through LC-MS-based metabolomic profiling by using ¹²C- or ¹³C-dansyl labeling for the amine/phenol submetabolome. RSV infection predominantly affected the amino acid biosynthesis pathways and urea cycle, whereas ΔF/TriAdj modulated the concentrations of almost all of the altered metabolites. Tryptophan metabolites were significantly affected, including indole, l-kynurenine, xanthurenic acid, serotonin, 5-hydroxyindoleacetic acid, and 6-hydroxymelatonin. The results from the present study provide further mechanistic insights into the mode of action of this RSV vaccine candidate and have important implications in the design of metabolic therapeutic interventions.

Chemical isotope labeling liquid chromatography mass spectrometry for investigating acute dietary effects of cow milk consumption on human urine metabolome

Biomarker discovery has been increasingly important in the field of metabolomics for the detection and understanding of diseases. Of the many biofluids available for metabolomics, urine is a preferred option as it is non-invasive to collect and contains a wide range of metabolites reflective of the health status of the testing individual. However, urine also contains many exogenous metabolites which are introduced through various sources such as diet. This complicates the data interpretation when searching the metabolome for disease-related endogenous metabolites. Since diet is difficult to control, this work aims to study the acute effects of diet (particularly cow milk) consumption on the human urine amine/phenol submetabolome by utilizing differential chemical isotope labeling (CIL) liquid chromatography mass spectrometry (LC-MS). LC-MS analysis of 62 urine samples collected before and after (1 hour and 2 hours) milk intake resulted in the detection of 4985 metabolites with an average of 3815 ± 206 (n = 62) detected per sample. The work aims to differentiate the exogenous “food” metabolites from the endogenous metabolite pool and to determine any dietary effects from milk intake on the human urine metabolome.

Bretschneider solution-induced alterations in the urine metabolome in cardiac surgery patients

The development of Bretschneider’s histidine-tryptophan-ketoglutarate (HTK) cardioplegia solution represented a major advancement in cardiac surgery, offering significant myocardial protection. However, metabolic changes induced by this additive in the whole body have not been systematically investigated. Using an untargeted mass spectrometry-based method to deeply explore the urine metabolome, we sought to provide a holistic and systematic view of metabolic perturbations occurred in patients receiving HTK. Prospective urine samples were collected from 100 patients who had undergone cardiac surgery, and metabolomic changes were profiled using a high-performance chemical isotope labeling liquid chromatography-mass spectrometry (LC-MS) method. A total of 14,642 peak pairs or metabolites were quantified using differential ¹³C-/¹²C-dansyl labeling LC-MS, which targets the amine/phenol submetabolome from urine specimens. We identified 223 metabolites that showed significant concentration change (fold change > 5) and assembled several potential metabolic pathway maps derived from these dysregulated metabolites. Our data indicated upregulated histidine metabolism with subsequently increased glutamine/glutamate metabolism, altered purine and pyrimidine metabolism, and enhanced vitamin B₆ metabolism in patients receiving HTK. Our findings provide solid evidence that HTK solution causes significant perturbations in several metabolic pathways and establish a basis for further study of key mechanisms underlying its organ-protective or potential harmful effects.

Alzheimer's Biomarkers From Multiple Modalities Selectively Discriminate Clinical Status: Relative Importance of Salivary Metabolomics Panels, Genetic, Lifestyle, Cognitive, Functional Health and Demographic Risk Markers

Background: Among the neurodegenerative diseases of aging, sporadic Alzheimer’s disease (AD) is the most prevalent and perhaps the most feared. With virtually no success at finding pharmaceutical therapeutics for altering progressive AD after diagnosis, research attention is increasingly directed at discovering biological and other markers that detect AD risk in the long asymptomatic phase. Both early detection and precision preclinical intervention require systematic investigation of multiple modalities and combinations of AD-related biomarkers and risk factors. We extend recent unbiased metabolomics research that produced a set of metabolite biomarker panels tailored to the discrimination of cognitively normal (CN), cognitively impaired and AD patients. Specifically, we compare the prediction importance of these panels with five other sets of modifiable and non-modifiable AD risk factors (genetic, lifestyle, cognitive, functional health and bio-demographic) in three clinical groups.

Method: The three groups were: CN (n = 35), mild cognitive impairment (MCI; n = 25), and AD (n = 22). In a series of three pairwise comparisons, we used machine learning technology random forest analysis (RFA) to test relative predictive importance of up to 19 risk biomarkers from the six AD risk domains.

Results: The three RFA multimodal prediction analyses produced significant discriminating risk factors. First, discriminating AD from CN was the AD metabolite panel and two cognitive markers. Second, discriminating AD from MCI was the AD/MCI metabolite panel and two cognitive markers. Third, discriminating MCI from CN was the MCI metabolite panel and seven markers from four other risk modalities: genetic, lifestyle, cognition and functional health.

Conclusions: Salivary metabolomics biomarker panels, supplemented by other risk markers, were robust predictors of: (1) clinical differences in impairment and dementia and even; (2) subtle differences between CN and MCI. For the latter, the metabolite panel was supplemented by biomarkers that were both modifiable (e.g., functional) and non-modifiable (e.g., genetic). Comparing, integrating and identifying important multi-modal predictors may lead to novel combinations of complex risk profiles potentially indicative of neuropathological changes in asymptomatic or preclinical AD.

Comprehensive evaluation of untargeted metabolomics data processing software in feature detection, quantification and discriminating marker selection

Data analysis represents a key challenge for untargeted metabolomics studies and it commonly requires extensive processing of more than thousands of metabolite peaks included in raw high-resolution MS data. Although a number of software packages have been developed to facilitate untargeted data processing, they have not been comprehensively scrutinized in the capability of feature detection, quantification and marker selection using a well-defined benchmark sample set. In this study, we acquired a benchmark dataset from standard mixtures consisting of 1100 compounds with specified concentration ratios including 130 compounds with significant variation of concentrations. Five software evaluated here (MS-Dial, MZmine 2, XCMS, MarkerView, and Compound Discoverer) showed similar performance in detection of true features derived from compounds in the mixtures. However, significant differences between untargeted metabolomics software were observed in relative quantification of true features in the benchmark dataset. MZmine 2 outperformed the other software in terms of quantification accuracy and it reported the most true discriminating markers together with the fewest false markers. Furthermore, we assessed selection of discriminating markers by different software using both the benchmark dataset and a real-case metabolomics dataset to propose combined usage of two software for increasing confidence of biomarker identification. Our findings from comprehensive evaluation of untargeted metabolomics software would help guide future improvements of these widely used bioinformatics tools and enable users to properly interpret their metabolomics results.

Impact of Oxygen Concentration on Metabolic Profile of Human Placenta-Derived Mesenchymal Stem Cells As Determined by Chemical Isotope Labeling LC-MS

The placenta resides in a physiologically low oxygen microenvironment of the body. Hypoxia induces a wide range of stem cell cellular activities. Here, we report a workflow for exploring the role of physiological (hypoxic, 5% oxygen) and original cell culture (normoxic, 21% oxygen) oxygen concentrations in regulating the metabolic status of human placenta-derived mesenchymal stem cells (hPMSCs). The general biological characteristics of hPMSCs were assessed via a variety of approaches such as cell counts, flow cytometry and differentiation study. A sensitive ¹³C/¹²C-dansyl labeling liquid chromatography-mass spectrometry (LC-MS) method targeting the amine/phenol submetabolome was used for metabolic profiling of the cell and corresponding culture supernatant. Multivariate and univariate statistical analyses were used to analyze the metabolomics data. hPMSCs cultured in hypoxia display smaller size, higher proliferation, greater differentiation ability and no difference in immunophenotype. Overall, 2987 and 2860 peak pairs or metabolites were detected and quantified in hPMSCs and culture supernatant, respectively. Approximately 86.0% of cellular metabolites and 84.3% of culture supernatant peak pairs were identified using a dansyl standard library or matched to metabolite structures using accurate mass search against human metabolome libraries. The orthogonal partial least-squares discriminant analysis (OPLS-DA) showed a clear separation between the hypoxic group and the normoxic group. Ten metabolites from cells and six metabolites from culture supernatant were identified as potential biomarkers of hypoxia. This study demonstrated that chemical isotope labeling LC-MS can be used to reveal the role of oxygen in the regulation of hPMSC metabolism, whereby physiological oxygen concentrations may promote arginine and proline metabolism, pantothenate and coenzyme A (CoA) biosynthesis, and alanine, aspartate and glutamate metabolism.

Missing Value Imputation Approach for Mass Spectrometry-based Metabolomics Data

Missing values exist widely in mass-spectrometry (MS) based metabolomics data. Various methods have been applied for handling missing values, but the selection can significantly affect following data analyses. Typically, there are three types of missing values, missing not at random (MNAR), missing at random (MAR), and missing completely at random (MCAR). Our study comprehensively compared eight imputation methods (zero, half minimum (HM), mean, median, random forest (RF), singular value decomposition (SVD), k-nearest neighbors (kNN), and quantile regression imputation of left-censored data (QRILC)) for different types of missing values using four metabolomics datasets. Normalized root mean squared error (NRMSE) and NRMSE-based sum of ranks (SOR) were applied to evaluate imputation accuracy. Principal component analysis (PCA)/partial least squares (PLS)-Procrustes analysis were used to evaluate the overall sample distribution. Student's t-test followed by correlation analysis was conducted to evaluate the effects on univariate statistics. Our findings demonstrated that RF performed the best for MCAR/MAR and QRILC was the favored one for left-censored MNAR. Finally, we proposed a comprehensive strategy and developed a public-accessible web-tool for the application of missing value imputation in metabolomics ( https://metabolomics.cc.hawaii.edu/software/MetImp/ ).

Missing Value Imputation Approach for Mass Spectrometry-based Metabolomics Data

Missing values exist widely in mass-spectrometry (MS) based metabolomics data. Various methods have been applied for handling missing values, but the selection can significantly affect following data analyses. Typically, there are three types of missing values, missing not at random (MNAR), missing at random (MAR), and missing completely at random (MCAR). Our study comprehensively compared eight imputation methods (zero, half minimum (HM), mean, median, random forest (RF), singular value decomposition (SVD), k-nearest neighbors (kNN), and quantile regression imputation of left-censored data (QRILC)) for different types of missing values using four metabolomics datasets. Normalized root mean squared error (NRMSE) and NRMSE-based sum of ranks (SOR) were applied to evaluate imputation accuracy. Principal component analysis (PCA)/partial least squares (PLS)-Procrustes analysis were used to evaluate the overall sample distribution. Student's t-test followed by correlation analysis was conducted to evaluate the effects on univariate statistics. Our findings demonstrated that RF performed the best for MCAR/MAR and QRILC was the favored one for left-censored MNAR. Finally, we proposed a comprehensive strategy and developed a public-accessible web-tool for the application of missing value imputation in metabolomics ( https://metabolomics.cc.hawaii.edu/software/MetImp/ ).

Chemical Isotope Labeling LC-MS for Human Blood Metabolome Analysis

Blood is a widely used biofluid in discovery metabolomic research to search for clinical metabolite biomarkers of diseases. Analyzing the entire human blood metabolome is a major analytical challenge, as blood, after being processed into serum or plasma, contains thousands of metabolites with diverse chemical and physical properties as well as a wide range of concentrations. We describe an enabling method based on high-performance chemical isotope labeling (CIL) liquid chromatography-mass spectrometry (LC-MS) for in-depth quantification of the metabolomic differences in comparative blood samples with high accuracy and precision.

The Impact of GFP Reporter Gene Transduction and Expression on Metabolomics of Placental Mesenchymal Stem Cells Determined by UHPLC-Q/TOF-MS

Introduction

Green fluorescent protein (GFP) is widely used as a reporter gene in regenerative medicine research to label and track stem cells. Here, we examined whether expressing GFP gene may impact the metabolism of human placental mesenchymal stem cells (hPMSCs).

Methods

The GFP gene was transduced into hPMSCs using lentiviral-based infection to establish GFP⁺hPMSCs. A sensitive ¹³C/¹²C-dansyl labeling LC-MS method targeting the amine/phenol submetabolome was used for in-depth cell metabolome profiling.

Results

A total of 1151 peak pairs or metabolites were detected from 12 LC-MS runs. Principal component analysis and partial least squares discriminant analysis showed poor separation, and the volcano plots demonstrated that most of the metabolites were not significantly changed when hPMSCs were tagged with GFP. Overall, 739 metabolites were positively or putatively identified. Only 11 metabolites showed significant changes. Metabolic pathway analyses indicated that three of the identified metabolites were involved in nine pathways. However, these metabolites are unlikely to have a large impact on the metabolic pathways due to their nonessential roles and limited hits in pathway analysis.

Conclusion

This study indicated that the expression of ectopic GFP reporter gene did not significantly alter the metabolomics pathways covered by the amine/phenol submetabolome.

Profiling novel metabolic biomarkers for Parkinson's disease using in-depth metabolomic analysis

OBJECTIVE

To profile the amine/phenol submetabolome to determine potential metabolite biomarkers associated with Parkinson's disease (PD) and PD with incipient dementia.

METHODS

At baseline of a 3-wave (18-month intervals) longitudinal study, serum samples were collected from 42 healthy controls and 43 PD patients. By wave 3 (year 3), 16 PD patients were diagnosed with dementia and were classified as PD with incipient dementia at baseline. Metabolomic profiling using dansylation isotope labeling liquid chromatography mass spectrometry was conducted to compare controls with the full PD, PD with no dementia, and PD with incipient dementia groups.

RESULTS

Metabolomic analyses detected 719 common metabolites in 80% of the samples. Some were significantly altered in pairwise comparison of different groups (fold change of >1.2 or <0.83 with q < 0.05). We discriminated PD and controls by using a 5-metabolite panel, vanillic acid, 3-hydroxykynurenine, isoleucyl-alanine, 5-acetylamino-6-amino-3-methyluracil, and theophylline. The receiver operating characteristic curve produced an area-under-the-curve value of 0.955 with 87.5% sensitivity and 93.0% specificity. In comparing PD with no dementia with PD with incipient dementia, we used an 8-metabolite panel, His-Asn-Asp-Ser, 3,4-dihydroxyphenylacetone, desaminotyrosine, hydroxy-isoleucine, alanyl-alanine, putrescine [-2H], purine [+O] and its riboside. This produced an area-under-the-curve value of 0.862 with 80.0% sensitivity and 77.0% specificity.

CONCLUSIONS

The significantly altered metabolites can be used to differentiate (1) PD patients from healthy controls with high accuracy and (2) the stable PD with no dementia group from those with incipient dementia. Following further validation in larger cohorts, these metabolites could be used for both discrimination and establishing prognosis in PD. © 2017 International Parkinson and Movement Disorder Society.

Overcoming Sample Matrix Effect in Quantitative Blood Metabolomics Using Chemical Isotope Labeling Liquid Chromatography Mass Spectrometry

Blood is widely used for discovery metabolomics to search for disease biomarkers. However, blood sample matrix can have a profound effect on metabolome analysis, which can impose an undesirable restriction on the type of blood collection tubes that can be used for blood metabolomics. We investigated the effect of blood sample matrix on metabolome analysis using a high-coverage and quantitative metabolome profiling technique based on differential chemical isotope labeling (CIL) LC-MS. We used ¹²C-/¹³C-dansylation LC-MS to perform relative quantification of the amine/phenol submetabolomes of four types of samples (i.e., serum, EDTA plasma, heparin plasma, and citrate plasma) collected from healthy individuals and compare their metabolomic results. From the analysis of 80 plasma and serum samples in experimental triplicate, we detected a total of 3651 metabolites with an average of 1818 metabolites per run (n = 240). The number of metabolites detected and the precision and accuracy of relative quantification were found to be independent of the sample type. Within each sample type, the metabolome data set could reveal biological variation (e.g., sex separation). Although the relative concentrations of some individual metabolites might be different in the four types of samples, for sex separation, all 66 significant metabolites with larger fold-changes (FC ≥ 2 and p < 0.05) found in at least one sample type could be found in the other types of samples with similar or somewhat reduced, but still significant, fold-changes. Our results indicate that CIL LC-MS could overcome the sample matrix effect, thereby greatly broadening the scope of blood metabolomics; any blood samples properly collected in routine clinical settings, including those in biobanks originally used for other purposes, can potentially be used for discovery metabolomics.

Exploring the human microbiome from multiple perspectives: factors altering its composition and function

Our microbiota presents peculiarities and characteristics that may be altered by multiple factors. The degree and consequences of these alterations depend on the nature, strength and duration of the perturbations as well as the structure and stability of each microbiota. The aim of this review is to sketch a very broad picture of the factors commonly influencing different body sites, and which have been associated with alterations in the human microbiota in terms of composition and function. To do so, first, a graphical representation of bacterial, fungal and archaeal genera reveals possible associations among genera affected by different factors. Then, the revision of sequence-based predictions provides associations with functions that become part of the active metabolism. Finally, examination of microbial metabolite contents and fluxes reveals whether metabolic alterations are a reflection of the differences observed at the level of population structure, and in the last step, link microorganisms to functions under perturbations that differ in nature and aetiology. The utilisation of complementary technologies and methods, with a special focus on metabolomics research, is thoroughly discussed to obtain a global picture of microbiota composition and microbiome function and to convey the urgent need for the standardisation of protocols.