Untargeted metabolomic analysis of urine samples in the diagnosis of some inherited metabolic disorders

Comparisons of different extraction methods and solvents for saliva samples

INTRODUCTION

Changes in the categories and concentrations of salivary metabolites may be closely related to oral, intestinal or systemic diseases. To study salivary metabolites, the first analytical step is to extract them from saliva samples as much as possible, while reducing interferences to a minimum. Frequently used extraction methods are protein precipitation (PPT), liquid-liquid extraction (LLE) and solid-phase extraction (SPE), with various organic solvents. The types and quantities of metabolites extracted with different methods may vary greatly, but few studies have systematically evaluated them.

OBJECTIVES

This study aimed to select the most suitable methods and solvents for the extraction of saliva according to different analytical targets.

METHODS

An untargeted metabolomics approach based on liquid chromatography-mass spectrometry was applied to obtain the raw data. The numbers of metabolites, repeatability of the data and intensities of mass spectrometry signals were used as evaluation criteria.

RESULTS

PPT resulted in the highest coverage. Among the PPT solvents, acetonitrile displayed the best repeatability and the highest coverage, while acetone resulted in the best signal intensities for the extracted compounds. LLE with the mixture of chloroform and methanol was the most suitable for the extraction of small hydrophobic compounds.

CONCLUSION

PPT with acetonitrile or acetone was recommended for untargeted analysis, while LLE with the mixture of chloroform and methanol was recommended for small hydrophobic compounds.

An evaluation of untargeted metabolomics methods to characterize inborn errors of metabolism

Inborn errors of metabolism (IEMs) encompass a diverse group of disorders that can be difficult to classify due to heterogenous clinical, molecular, and biochemical manifestations. Untargeted metabolomics platforms have become a popular approach to analyze IEM patient samples because of their ability to detect many metabolites at once, accelerating discovery of novel biomarkers, and metabolic mechanisms of disease. However, there are concerns about the reproducibility of untargeted metabolomics research due to the absence of uniform reporting practices, data analyses, and experimental design guidelines. Therefore, we critically evaluated published untargeted metabolomic platforms used to characterize IEMs to summarize the strengths and areas for improvement of this technology as it progresses towards the clinical laboratory. A total of 96 distinct IEMs were collectively evaluated by the included studies. However, most of these IEMs were evaluated by a single untargeted metabolomic method, in a single study, with a limited cohort size (55/96, 57%). The goals of the included studies generally fell into two, often overlapping, categories: detecting known biomarkers from many biochemically distinct IEMs using a single platform, and detecting novel metabolites or metabolic pathways. There was notable diversity in the design of the untargeted metabolomic platforms. Importantly, the majority of studies reported adherence to quality metrics, including the use of quality control samples and internal standards in their experiments, as well as confirmation of at least some of their feature annotations with commercial reference standards. Future applications of untargeted metabolomics platforms to the study of IEMs should move beyond single-subject analyses, and evaluate reproducibility using a prospective, or validation cohort.

Benchmarking Outlier Detection Methods for Detecting IEM Patients in Untargeted Metabolomics Data

Untargeted metabolomics (UM) is increasingly being deployed as a strategy for screening patients that are suspected of having an inborn error of metabolism (IEM). In this study, we examined the potential of existing outlier detection methods to detect IEM patient profiles. We benchmarked 30 different outlier detection methods when applied to three untargeted metabolomics datasets. Our results show great differences in IEM detection performances across the various methods. The methods DeepSVDD and R-graph performed most consistently across the three metabolomics datasets. For datasets with a more balanced number of samples-to-features ratio, we found that AE reconstruction error, Mahalanobis and PCA reconstruction error also performed well. Furthermore, we demonstrated the importance of a PCA transform prior to applying an outlier detection method since we observed that this increases the performance of several outlier detection methods. For only one of the three metabolomics datasets, we observed clinically satisfying performances for some outlier detection methods, where we were able to detect 90% of the IEM patient samples while detecting no false positives. These results suggest that outlier detection methods have the potential to aid the clinical investigator in routine screening for IEM using untargeted metabolomics data, but also show that further improvements are needed to ensure clinically satisfying performances.

Biomarker discovery in galactosemia: Metabolomics with UPLC/HRMS in dried blood spots

Introduction:Galactosemia (GAL) is a genetic disorder that results in disturbances in galactose metabolism and can lead to life-threatening complications. However, the underlying pathophysiology of long-term complications in GAL remains poorly understood. Methods: In this study, a metabolomics approach using ultra-performance liquid chromatography coupled with high-resolution mass spectrometry was used to investigate metabolomic changes in dried blood spots of 15 patients with GAL and 39 healthy individuals. Results: The study found that 2,819 metabolites underwent significant changes in patients with GAL compared to the control group. 480 human endogenous metabolites were identified, of which 209 and 271 were upregulated and downregulated, respectively. PA (8:0/LTE4) and ganglioside GT1c (d18:0/20:0) metabolites showed the most significant difference between GAL and the healthy group, with an area under the curve of 1 and 0.995, respectively. Additionally, the study identified potential biomarkers for GAL, such as 17-alpha-estradiol-3-glucuronide and 16-alpha-hydroxy DHEA 3-sulfatediphosphate. Conclusion: This metabolomics study deepened the understanding of the pathophysiology of GAL and presented potential biomarkers that might serve as prognostic biomarkers to monitor the progression or support the clinical diagnosis of GAL.

Untargeted metabolomic analysis of urine samples for diagnosis of inherited metabolic disorders

Metabolomics has become an important tool for clinical research, especially for analyzing inherited metabolic disorders (IMDs). The purpose of this study was to explore the performance of metabolomics in diagnosing IMDs using an untargeted metabolomic approach. A total of 40 urine samples were collected: 20 samples from healthy children and 20 from pediatric patients, of whom 13 had confirmed IMDs and seven had suspected IMDs. Samples were analyzed by Orbitrap mass spectrometry in positive and negative mode alternately, coupled with ultra-high liquid chromatography. Raw data were processed using Compound Discovery 2.0 ™ and then exported for partial least squares discriminant analysis (PLS-DA) by SIMCA-P 14.1. After comparing with m/zCloud and chemSpider libraries, compounds with similarity above 80% were selected and normalized for subsequent relative quantification analysis. The uncommon compounds discovered were analyzed based on the Kyoto Encyclopedia of Genes and Genomes to explore their possible metabolic pathways. All IMDs patients were successfully distinguished from controls in the PLS-DA. Untargeted metabolomics revealed a broader metabolic spectrum in patients than what is observed using routine chromatographic methods for detecting IMDs. Higher levels of certain compounds were found in all 13 confirmed IMD patients and 5 of 7 suspected IMD patients. Several potential novel markers emerged after relative quantification. Untargeted metabolomics may be able to diagnose IMDs from urine and may deepen insights into the disease by revealing changes in various compounds such as amino acids, acylcarnitines, organic acids, and nucleosides. Such analyses may identify biomarkers to improve the study and treatment of IMDs.

Application of Principal Component Analysis to Newborn Screening for Congenital Adrenal Hyperplasia

PURPOSE

Newborn screening laboratories are challenged to develop reporting algorithms that accurately identify babies at increased risk for congenital adrenal hyperplasia (CAH) due to 21-hydroxylase deficiency (21OHD). Screening algorithms typically use cutoff values for a key steroid(s) and include considerations for covariates, such as gestational age or birth weight, but false-positive and false-negative results are still too frequent, preventing accurate assessments. Principal component analysis (PCA) is a statistical method that reduces high-dimensional data to a small number of components, capturing patterns of association that may be relevant to the outcome of interest. To our knowledge, PCA has not been evaluated in the newborn screening setting to determine whether it can improve the positive predictive value of 21OHD screening.

METHODS

PCA was applied to a data set of 920 newborns with measured concentrations of 5 key steroids that are known to be perturbed in patients with 21OHD. A decision tree for the known outcomes (confirmed 21OHD cases and unaffected individuals) was created with 2 principal components as predictors. The effectiveness of the PCA-derived decision tree was compared with the current algorithm.

RESULTS

PCA improved the positive predictive value of 21OHD screening from 20.0% to 66.7% in a retrospective study comparing the current algorithm to a tree-based algorithm using PCA-derived variables. The streamlined PCA-derived decision tree, comprising only 3 assessment points, greatly simplified the 21OHD reporting algorithm.

CONCLUSIONS

This first report of PCA applied to newborn screening for 21OHD demonstrates enhanced detection of affected individuals within the unaffected population.

Metabolomic approach to measuring quality of chilled chicken meat during storage

The metabolites of stored, chilled chicken meat were analyzed using liquid chromatograph-mass spectrometry and metabolomics. The results showed significant differences in the metabolites of chicken meat stored at 4°C for 0 D and meat stored for longer periods of 2 D, 4 D, 6 D, and 10 D, when analyzed based on a variable of importance >2 and P < 0.05. These changed metabolites included amino acids, amines, nucleosides, nucleotides, carbohydrates, organic acids, and other substances. The data from this study provide a holistic understanding of food quality changes in chicken meat during deterioration in storage.

Inborn Errors of Metabolism in the Era of Untargeted Metabolomics and Lipidomics

Inborn errors of metabolism (IEMs) are a group of inherited diseases with variable incidences. IEMs are caused by disrupting enzyme activities in specific metabolic pathways by genetic mutations, either directly or indirectly by cofactor deficiencies, causing altered levels of compounds associated with these pathways. While IEMs may present with multiple overlapping symptoms and metabolites, early and accurate diagnosis of IEMs is critical for the long-term health of affected subjects. The prevalence of IEMs differs between countries, likely because different IEM classifications and IEM screening methods are used. Currently, newborn screening programs exclusively use targeted metabolic assays that focus on limited panels of compounds for selected IEM diseases. Such targeted approaches face the problem of false negative and false positive diagnoses that could be overcome if metabolic screening adopted analyses of a broader range of analytes. Hence, we here review the prospects of using untargeted metabolomics for IEM screening. Untargeted metabolomics and lipidomics do not rely on predefined target lists and can detect as many metabolites as possible in a sample, allowing to screen for many metabolic pathways simultaneously. Examples are given for nontargeted analyses of IEMs, and prospects and limitations of different metabolomics methods are discussed. We conclude that dedicated studies are needed to compare accuracy and robustness of targeted and untargeted methods with respect to widening the scope of IEM diagnostics.

Novel two-step derivation method for the synchronous analysis of inherited metabolic disorders using urine

The aim of the present study was to conduct preliminary clinical screening and monitoring using a novel two-step derivatization process of urine in five categories of inherited metabolic disease (IMD). Urine samples (100 µl, containing 2.5 mmol/l creatinine) were taken from patients with IMDs. The collected urine was then treated using a two-step derivatization method (with oximation and silylation at room temperature), where urea and protein were removed. In the first step of the derivatization, α-ketoacids and α-aldehyde acids were prepared by oximation using novel oximation reagents. The second-step of the derivatization was that residues were silylated for analysis. Urine samples were examined using gas chromatography/mass spectrometry (GC/MS) and a retention time-locking technique. The simultaneous analysis and identification of >400 metabolites in >130 types of IMD was possible from the GC/MS results, where the IMDs included phenylketonuria, ornithine transcarbamylase deficiency, neonatal intrahepatic cholestasis caused by citrin deficiency, β-ureidopropionase deficiency and mitochondrial metabolic disorders. This method was demonstrated to have good repeatability. Considering α-ketoglutarate (α-KG) as an example, the relative standard deviations (RSDs) of the α-KG retention time and peak area were 0.8 and 3.9%, respectively, the blank spiked recovery rate was between 89.6 and 99.8%, and the RSD was ≤7.5% (n=5). The method facilitates the analysis of thermally non-stable and semi-volatile metabolites in urine, and greatly expands the range of materials that can be synchronously screened by GC/MS. Furthermore, it provides a comprehensive, effective and reliable biochemical analysis platform for the pathological research of IMDs.

Diagnosis of adenylosuccinate lyase deficiency by metabolomic profiling in plasma reveals a phenotypic spectrum

Adenylosuccinate lyase (ADSL) deficiency is a rare autosomal recessive neurometabolic disorder that presents with a broad-spectrum of neurological and physiological symptoms. The ADSL gene produces an enzyme with binary molecular roles in de novo purine synthesis and purine nucleotide recycling. The biochemical phenotype of ADSL deficiency, accumulation of SAICAr and succinyladenosine (S-Ado) in biofluids of affected individuals, serves as the traditional target for diagnosis with targeted quantitative urine purine analysis employed as the predominate method of detection. In this study, we report the diagnosis of ADSL deficiency using an alternative method, untargeted metabolomic profiling, an analytical scheme capable of generating semi-quantitative z-score values for over 1000 unique compounds in a single analysis of a specimen. Using this method to analyze plasma, we diagnosed ADSL deficiency in four patients and confirmed these findings with targeted quantitative biochemical analysis and molecular genetic testing. ADSL deficiency is part of a large a group of neurometabolic disorders, with a wide range of severity and sharing a broad differential diagnosis. This phenotypic similarity among these many inborn errors of metabolism (IEMs) has classically stood as a hurdle in their initial diagnosis and subsequent treatment. The findings presented here demonstrate the clinical utility of metabolomic profiling in the diagnosis of ADSL deficiency and highlights the potential of this technology in the diagnostic evaluation of individuals with neurologic phenotypes.

Metabolomic Profiling of Human Urine as a Screen for Multiple Inborn Errors of Metabolism

AIMS

We wished to determine the efficacy of using urine as an analyte to screen for a broad range of metabolic products associated with multiple different types of inborn errors of metabolism (IEMs), using an automated mass spectrometry-based assay. Urine was compared with plasma samples from a similar cohort analyzed using the same assay. Specimens were analyzed using two different commonly utilized urine normalization methods based on creatinine and osmolality, respectively.

METHODS

Biochemical profiles for each sample (from both affected and unaffected subjects) were obtained using a mass spectrometry-based platform and population-based statistical analyses.

RESULTS

We identified over 1200 biochemicals from among 100 clinical urine samples and identified clear biochemical signatures for 16 of 18 IEM diseases tested. The two diseases that did not result in clear signatures, X-linked creatine transporter deficiency and ornithine transcarbamylase deficiency, were from individuals under treatment, which masked biomarker signatures. Overall the process variability and coefficient of variation for isolating and identifying biochemicals by running technical replicates of each urine sample was 10%.

CONCLUSIONS

A single urine sample analyzed with our integrated metabolomic platform can identify signatures of IEMs that are traditionally identified using many different assays and multiple sample types. Creatinine and osmolality-normalized data were robust to the detection of the disorders and samples tested here.

Noninvasive metabolic profiling for painless diagnosis of human diseases and disorders

Metabolic profiling provides a powerful diagnostic tool complementary to genomics and proteomics. The pain, discomfort and probable iatrogenic injury associated with invasive or minimally invasive diagnostic methods, render them unsuitable in terms of patient compliance and participation. Metabolic profiling of biomatrices like urine, breath, saliva, sweat and feces, which can be collected in a painless manner, could be used for noninvasive diagnosis. This review article covers the noninvasive metabolic profiling studies that have exhibited diagnostic potential for diseases and disorders. Their potential applications are evident in different forms of cancer, metabolic disorders, infectious diseases, neurodegenerative disorders, rheumatic diseases and pulmonary diseases. Large scale clinical validation of such diagnostic methods is necessary in future.