Metabolomics Biomarker Discovery Research Articles

Untargeted Metabolomic Biomarker Discovery for the Detection of Ectopic Pregnancy.

Ectopic pregnancy (EP) is the leading cause of maternal morbidity and mortality in the first trimester. Using an untargeted metabolomic approach, we sought to identify putative plasma biomarkers using tandem liquid chromatography-mass spectrometry for the detection of tubal EP. This case-control study included the prospective recruitment of 50 tubal EP cases and 50 early intrauterine pregnancy controls. To avoid over-fitting, logistic regression models were developed in a randomly selected discovery group (30 cases vs. 30 controls) and validated in the test group (20 cases vs. 20 controls). In total, 585 mass spectral features were detected, of which 221 molecular features were significantly altered in EP plasma (p < 0.05). Molecular networking and metabolite identification was employed using the Global Natural Products Social Molecular Networking (GNPS) database, which identified 97 metabolites at a high confidence level. Top significant metabolites include subclasses of sphingolipids, carnitines, glycerophosphocholines, and tryptophan metabolism. The top regression model, consisting of D-erythro-sphingosine and oleoyl-carnitine, was validated in a test group and achieved an area under receiving operating curve (AUC) (95% CI) = 0.962 (0.910-1) with a sensitivity of 100% and specificity of 95.9%. Metabolite alterations indicate alterations related to inflammation and abnormal placentation in EP. The validation of these metabolite biomarkers in the future could potentially result in improved early diagnosis.

Metabolomic Characterization of Acute Ischemic Stroke Facilitates Metabolomic Biomarker Discovery.

Acute ischemic stroke (AIS) is characterized by a sudden blockage of one of the main arteries supplying blood to the brain, leading to insufficient oxygen and nutrients for brain cells to function properly. Unfortunately, metabolic alterations in the biofluids with AIS are still not well understood. In this study, we performed high-throughput target metabolic analysis on 44 serum samples, including 22 from AIS patients and 22 from healthy controls. Multiple-reaction monitoring analysis of 180 common metabolites revealed a total of 29 metabolites that changed significantly (VIP > 1, p < 0.05). Multivariate statistical analysis unraveled a striking separation between AIS patients and healthy controls. Comparing the AIS group with the control group, the contents of argininosuccinic acid, beta-D-glucosamine, glycerophosphocholine, L-abrine, and L-pipecolic acid were remarkably downregulated in AIS patients. Twenty-nine out of 112 detected metabolites enriched in disturbed metabolic pathways, including aminoacyl-tRNA biosynthesis, glycerophospholipid metabolism, lysine degradation, phenylalanine, tyrosine, and tryptophan biosynthesis metabolic pathways. Collectively, these results will provide a sensitive, feasible diagnostic prospect for AIS patients.

Discovery of metabolomic biomarkers for discriminating platinum-sensitive and platinum-resistant ovarian cancer by using GC-MS.

This study aims to determine ovarian cancer (OC) patients with platinum resistance for alternative treatment protocols by using metabolomic methodologies. Urine and serum samples of platinum-resistant and platinum-sensitive OC were analyzed using GC-MS. After data processing of GC-MS raw data, multivariate analyses were performed to interpret complex data for biologically meaningful information and to identify the biomarkers that cause differences between two groups. The biomarkers were verified after univariate, multivariate, and ROC analysis. Finally, metabolomic pathways related to group separations were specified. The results of biomarker analysis showed that 3,4-dihydroxyphenylacetic acid, 4-hydroxybutyric acid, L-threonine, D- mannose, and sorbitol metabolites were potential biomarkers in urine samples. In serum samples, L-arginine, linoleic acid, L-glutamine, and hypoxanthine were identified as important biomarkers. R2Y, Q2, AUC, sensitivity and specificity values of platinum-resistant and sensitive OC patients' urine and serum samples were 0.85, 0.545, 0.844, 91.30%, 81.08 and 0.570, 0.206, 0.743, 77.78%, 74.28%, respectively. In metabolic pathway analysis of urine samples, tyrosine metabolism and fructose and mannose metabolism were found to be statistically significant (p < 0.05) for the discrimination of the two groups. While 3,4-dihydroxyphenylacetic acid, L-tyrosine, and fumaric acid metabolites were effective in tyrosine metabolism. D-sorbitol and D-mannose metabolites were significantly important in fructose and mannose metabolism. However, seven metabolomic pathways were significant (p < 0.05) in serum samples. In terms of p-value, L-glutamine in the nitrogen metabolic pathway from the first three pathways; L-glutamine and pyroglutamic acid metabolites in D-glutamine and D-glutamate metabolism. In the arginine and proline metabolic pathway, L-arginine, L-proline, and L-ornithine metabolites differed significantly between the two groups.

Metabolomic Biomarkers Differentiate Soy Sauce Freshness under Conditions of Accelerated Storage

Naturally fermented soy sauce is one of the few globally valued food condiments. It is complex in its substrate, manufacturing processes, and chemical profile of salts and organic compounds, resulting from spontaneous, enzymatic and biochemical reactions. The overall chemical character of soy sauce has a few rivals relative to its chemical and bioactive complexity. Resulting from this complexity are unique sensory attributes contributing to the characteristic soy sauce flavor as well as potentiating other sensory sensations. Soy sauce is susceptible to deterioration after bottling during storage. This work examined soy sauces over an eight-month period using descriptive sensory methods and the discovery of metabolomic biomarkers with high resolution mass spectrometry, wherein samples were derivatized to enable volatility and identification of polar analytes. While several thousand metabolites were detected, only organic acids, amino acids, and various glycosylated metabolites were statistically defensible biomarkers of storage time. The relationships between sensory and metabolomic data were assessed using Kendall rank-based correlations to generate Kendall Tau correlation coefficients. A second approach filtered the data based on correlation significance and grouped molecules based on hierarchical clustering. Mass spectrometry analyses discovered several thousand unique analyte peaks with relevant changes denoted as significant relative to the fresh samples using volcano depictions of p values versus changes in compound abundances. We present a metabolomic approach for the analysis of complex food systems capable of differentiating a quantifiable extrinsic variable, which is, in this case, storage time with a correlation coefficient of 0.99. We further demonstrate that changes in soy sauce resulting from storage are characterized by sensory decreases in fruity/grape and nutty/sesame aroma and increases in methional/potato aroma and astringent attributes with concomitant changes in the concentrations of several key biomarkers.

Rate normalization for sweat metabolomics biomarker discovery

As the demand for real-time exercise performance feedback increases, excreted sweat has become a biosource of interest for continuous human performance assessment. For sweat to truly fulfill this requirement, analyte concentrations must be normalized to adequately assess day-to-day differences within and among individuals. In this manuscript, data are presented highlighting the use of accurate localized sweat rate as a means for ion and global metabolomic data normalization. The results illustrate large sweat rate variability among individuals over the course of two distinct exercises protocols. Furthermore, the data show sweat rate is not symmetrical at similar locations among right and left forearms of individuals (p = 0.0007). Sweat ion conductivity analysis suggest overall sweat rate normalization reduces variability collectively among ion values and participants with principal component analysis showing 77.8% of variation in the data set attributable to sweat rate normalization. Global metabolomic analysis of sweat illustrated overall rate normalization increases the variability among test subjects with 72.7% of the variation explained by sweat rate normalization. Finally, overall rate normalized metabolomic features of sweat significantly correlated (ρ ≥ 0.7, ρ ≤ −0.7) with measured performance metrics of the individual, establishing the potential for sweat to be used as a biosource for performance monitoring. Collectively, these data illustrate the importance of accurate localized sweat rate determination, for analyte data normalization, in support for the use of sweat in biomarker discovery efforts to predict human performance.

Metabolomic Biomarkers for Detection, Prognosis and Identifying Recurrence in Endometrial Cancer.

Metabolic reprogramming is increasingly recognised as one of the defining hallmarks of tumorigenesis. There is compelling evidence to suggest that endometrial cancer develops and progresses in the context of profound metabolic dysfunction. Whilst the incidence of endometrial cancer continues to rise in parallel with the global epidemic of obesity, there are, as yet, no validated biomarkers that can aid risk prediction, early detection, prognostic evaluation or surveillance. Advances in high-throughput technologies have, in recent times, shown promise for biomarker discovery based on genomic, transcriptomic, proteomic and metabolomic platforms. Metabolomics, the large-scale study of metabolites, deals with the downstream products of the other omics technologies and thus best reflects the human phenotype. This review aims to provide a summary and critical synthesis of the existing literature with the ultimate goal of identifying the most promising metabolite biomarkers that can augment current endometrial cancer diagnostic, prognostic and recurrence surveillance strategies. Identified metabolites and their biochemical pathways are discussed in the context of what we know about endometrial carcinogenesis and their potential clinical utility is evaluated. Finally, we underscore the challenges inherent in metabolomic biomarker discovery and validation and provide fresh perspectives and directions for future endometrial cancer biomarker research.

Current Status of Metabolomic Biomarker Discovery: Impact of Study Design and Demographic Characteristics.

Widespread application of omic technologies is evolving our understanding of population health and holds promise in providing precise guidance for selection of therapeutic interventions based on patient biology. The opportunity to use hundreds of analytes for diagnostic assessment of human health compared to the current use of 10–20 analytes will provide greater accuracy in deconstructing the complexity of human biology in disease states. Conventional biochemical measurements like cholesterol, creatinine, and urea nitrogen are currently used to assess health status; however, metabolomics captures a comprehensive set of analytes characterizing the human phenotype and its complex metabolic processes in real-time. Unlike conventional clinical analytes, metabolomic profiles are dramatically influenced by demographic and environmental factors that affect the range of normal values and increase the risk of false biomarker discovery. This review addresses the challenges and opportunities created by the evolving field of clinical metabolomics and highlights features of study design and bioinformatics necessary to maximize the utility of metabolomics data across demographic groups.

Evaluation of statistical techniques to normalize mass spectrometry-based urinary metabolomics data

Human urine recently became a popular medium for metabolomics biomarker discovery because its collection is non-invasive. Sometimes renal dilution of urine can be problematic in this type of urinary biomarker analysis. Currently, various normalization techniques such as creatinine ratio, osmolality, specific gravity, dry mass, urine volume, and area under the curve are used to account for the renal dilution. However, these normalization techniques have their own drawbacks. In this project, mass spectrometry-based urinary metabolomic data obtained from prostate cancer (n = 56), bladder cancer (n = 57) and control (n = 69) groups were analyzed using statistical normalization techniques. The normalization techniques investigated in this study are Creatinine Ratio, Log Value, Linear Baseline, Cyclic Loess, Quantile, Probabilistic Quotient, Auto Scaling, Pareto Scaling, and Variance Stabilizing Normalization. The appropriate summary statistics for comparison of normalization techniques were created using variances, coefficients of variation, and boxplots. For each normalization technique, a principal component analysis was performed to identify clusters based on cancer type. In addition, hypothesis tests were conducted to determine if the normalized biomarkers could be used to differentiate between the cancer types. The results indicate that the determination of statistical significance can be dependent upon which normalization method is utilized. Therefore, careful consideration should go into choosing an appropriate normalization technique as no method had universally superior performance.

The Search for Clinically Useful Biomarkers of Complex Disease: A Data Analysis Perspective.

Unmet clinical diagnostic needs exist for many complex diseases, which it is hoped will be solved by the discovery of metabolomics biomarkers. However, as yet, no diagnostic tests based on metabolomics have yet been introduced to the clinic. This review is presented as a research perspective on how data analysis methods in metabolomics biomarker discovery may contribute to the failure of biomarker studies and suggests how such failures might be mitigated. The study design and data pretreatment steps are reviewed briefly in this context, and the actual data analysis step is examined more closely.

Fabrication of homogenous three‐dimensional biomimetic tissue for mass spectrometry imaging

Reference samples are essential for mass spectrometric method optimization, data quality control, and target analyte quantitation. However, it is highly challenging to prepare an ideal homogeneous, standard-spiked tissue sample for mass spectrometry imaging (MSI) research. Herein, we present a standard-spiked 3D biomimetic tissue model fabricated with native cells, homogenate matrix, and biocompatible polymer. Unlike traditional homogenized tissue surrogates or those constructed with "on-tissue" or "under-tissue" micropipetting strategies, this simulated tissue shares both structural integrity of cells and homogeneous properties of matrix. As a result, analyte standards could undergo more in-depth incorporation and has a more comparable native status with a real tissue. Series of tissue sections made from the 3D tissue model were proven to be feasible and useful for the parameter optimization, analyte quantitation, and calibration curve fitting for the air-flow assisted desorption electrospray ionization MSI. Additionally, by analyzing the quality control model sections, we proposed a median principal component score calibration and demonstrated that this method can normalize instrumental fluctuations to stable levels in a large-scale untargeted MSI experiments for the reliable metabolomic biomarker discovery. Thus, these results indicated that the standard-spiked 3D biomimetic tissue has convincing significance in MSI analysis.

Critical review of reporting of the data analysis step in metabolomics.

We present the first study to critically appraise the quality of reporting of the data analysis step in metabolomics studies since the publication of minimum reporting guidelines in 2007. The aim of this study was to assess the standard of reporting of the data analysis step in metabolomics biomarker discovery studies and to investigate whether the level of detail supplied allows basic understanding of the steps employed and/or reuse of the protocol. For the purposes of this review we define the data analysis step to include the data pretreatment step and the actual data analysis step, which covers algorithm selection, univariate analysis and multivariate analysis. We reviewed the literature to identify metabolomic studies of biomarker discovery that were published between January 2008 and December 2014. Studies were examined for completeness in reporting the various steps of the data pretreatment phase and data analysis phase and also for clarity of the workflow of these sections. We analysed 27 papers, published anytime in 2008 until the end of 2014 in the area or biomarker discovery in serum metabolomics. The results of this review showed that the data analysis step in metabolomics biomarker discovery studies is plagued by unclear and incomplete reporting. Major omissions and lack of logical flow render the data analysis' workflows in these studies impossible to follow and therefore replicate or even imitate. While we await the holy grail of computational reproducibility in data analysis to become standard, we propose that, at a minimum, the data analysis section of metabolomics studies should be readable and interpretable without omissions such that a data analysis workflow diagram could be extrapolated from the study and therefore the data analysis protocol could be reused by the reader. That inconsistent and patchy reporting obfuscates reproducibility is a given. However even basic understanding and reuses of protocols are hampered by the low level of detail supplied in the data analysis sections of the studies that we reviewed.

Combining strong sparsity and competitive predictive power with the L-sOPLS approach for biomarker discovery in metabolomics

In the context of metabolomics analyses, partial least squares (PLS) represents the standard tool to perform regression and classification. OPLS, the Orthogonal extension of PLS which has proved to be very useful when interpretation is the main issue, is a more recent way to decompose the PLS solution into predictive components correlated to the target Y and components pertaining to the data X but uncorrelated to Y. This predominance of (O)PLS can raise the question of the awareness of alternative multivariate regression and/or classification tools able to find biomarkers. Actually, the search for biomarkers remains a key issue in metabolomics as it is crucial to very accurately target discriminating features. Most of the time, (O)PLS methods perform well but a drawback often occurs: too many variables can be selected as potential biomarkers even using adapted statistical significance tests. However, for final users (in medical studies for instance), it can be advantageous to deal with only a small number of easily interpretable biomarkers. This drawback is approached in this paper via the use of sparse methods. The sparse-PLS (sPLS), an extension of PLS which promotes an inner variable/feature selection, is an interesting existing solution. But a new intuitive algorithm is proposed in this paper to combine sparsity and the advantages of an orthogonalization step: the “Light-sparse-OPLS” (L-sOPLS). L-sOPLS promotes sparsity on a previously optimized deflated matrix which implies the removal of the Y-orthogonal components. A discussion around the compromise between sparsity and predictive modelling performances is provided and it is shown that L-sOPLS produces convincing results, illustrated principally on the basis of $$^1$$ H-NMR spectral data but also on genomic RT-qPCR data. The L-sOPLS algorithm allows to reach better predictive performances than (O)PLS and sPLS while taking into account only a very small number of relevant descriptors.

Metabolic Pathway Extension Approach for Metabolomic Biomarker Identification.

Discovery of metabolomic biomarkers represents an important task in disease diagnosis and therapy. Although the development of various analytical tools and online libraries facilitates the identification of biomarkers, the fast and reliable identification of new biomarkers that are not included in databases still represents a major bottleneck in the field of metabolomics. Here, we developed a metabolic pathway extension (MPE) approach to the fast characterization of metabolomic biomarkers. This approach was proposed based on a core concept that the whole metabolome is built from a limited number of initial metabolites via various kinds and multiple steps of metabolic reactions, and thus, theoretically, the whole metabolome might be mapped from the initial metabolites and metabolic reactions. Carnitine was used as an example of initial metabolites to validate this concept and the usefulness of MPE approach. The intragastric dosing of carnitine to mice induced a significant alternation of a total of 97 metabolites. Mass differences between each pair of metabolites were calculated and then matched with those of typical metabolic pathways automatically by an in-house developed program. Diagnostic ions and neutral losses were used for validating the matches. With this approach, 93 out of a total of 97 metabolites were putatively identified, while only half of them could be traced from the currently available online database. The MPE approach was further validated by applying to the identification of carnitine-associated biomarkers in a typical mice model of fasting, and extended to the development of bile acids submetabolome. Our study indicates that the MPE approach is highly useful for rapid and reliable identification of metabolically and structurally associated biomarkers.

Overlap in serum metabolic profiles between non-related diseases: Implications for LC-MS metabolomics biomarker discovery

Untargeted metabolic profiling has generated large activity in the field of clinical biomarker discovery. Yet, no clinically approved metabolite biomarkers have emerged with failure in validation phases often being a reason. To investigate why, we have applied untargeted metabolic profiling in a retrospective cohort of serum samples representing non-related diseases. Age and gender matched samples from patients diagnosed with pneumonia, congestive heart failure, lymphoma and healthy controls were subject to comprehensive metabolic profiling using ultra-performance liquid chromatography-mass spectrometry (UPLC-MS). The metabolic profile of each diagnosis was compared to the healthy control group and significant metabolites were filtered out using t-test with FDR correction. Metabolites found to be significant between each disease and healthy controls were compared and analyzed for overlap. Results show that despite differences in etiology and clinical disease presentation, the fraction of metabolites with an overlap between two or more diseases was 61%. A majority of these metabolites can be associated with immune responses thus representing non-disease specific events. We show that metabolic serum profiles from patients representing non-related diseases display very similar metabolic differences when compared to healthy controls. Many of the metabolites discovered as disease specific in this study have further been associated with other diseases in the literature. Based on our findings we suggest non-related disease controls in metabolomics biomarker discovery studies to increase the chances of a successful validation and future clinical applications.

Multiplexed separations for biomarker discovery in metabolomics: Elucidating adaptive responses to exercise training.

High efficiency separations are needed to enhance selectivity, mass spectral quality, and quantitative performance in metabolomic studies. However, low sample throughput and complicated data preprocessing remain major bottlenecks to biomarker discovery. We introduce an accelerated data workflow to identify plasma metabolite signatures of exercise responsiveness when using multisegment injection-capillary electrophoresis-mass spectrometry (MSI-CE-MS). This multiplexed separation platform takes advantage of customizable serial injections to enhance sample throughput and data fidelity based on temporally resolved ion signals derived from seven different sample segments analyzed within a single run. MSI-CE-MS was applied to explore the adaptive metabolic responses of a cohort of overweight/obese women (BMI > 25, n = 9) performing a 6-wk high-intensity interval training intervention using a repeated measures/cross-over study design. Venous blood samples were collected from each subject at three time intervals (baseline, postexercise, recovery) in their naïve and trained states while completing standardized cycling trials at the same absolute workload. Complementary statistical methods were used to classify dynamic changes in plasma metabolism associated with strenuous exercise and training status. Positive adaptations to exercise were associated with training-induced upregulation in plasma l-carnitine at rest due to improved muscle oxidative capacity, and greater antioxidant capacity as reflected by lower circulating glutathionyl-l-cysteine mixed disulfide. Attenuation in plasma hypoxanthine and higher O-acetyl-l-carnitine levels postexercise also indicated lower energetic stress for trained women.

Development of isotope labeling liquid chromatography mass spectrometry for mouse urine metabolomics: quantitative metabolomic study of transgenic mice related to Alzheimer's disease.

Because of a limited volume of urine that can be collected from a mouse, it is very difficult to apply the common strategy of using multiple analytical techniques to analyze the metabolites to increase the metabolome coverage for mouse urine metabolomics. We report an enabling method based on differential isotope labeling liquid chromatography mass spectrometry (LC-MS) for relative quantification of over 950 putative metabolites using 20 μL of urine as the starting material. The workflow involves aliquoting 10 μL of an individual urine sample for ¹²C-dansylation labeling that target amines and phenols. Another 10 μL of aliquot was taken from each sample to generate a pooled sample that was subjected to ¹³C-dansylation labeling. The ¹²C-labeled individual sample was mixed with an equal volume of the ¹³C-labeled pooled sample. The mixture was then analyzed by LC-MS to generate information on metabolite concentration differences among different individual samples. The interday repeatability for the LC-MS runs was assessed, and the median relative standard deviation over 4 days was 5.0%. This workflow was then applied to a metabolomic biomarker discovery study using urine samples obtained from the TgCRND8 mouse model of early onset familial Alzheimer's disease (FAD) throughout the course of their pathological deposition of beta amyloid (Aβ). It was showed that there was a distinct metabolomic separation between the AD prone mice and the wild type (control) group. As early as 15-17 weeks of age (presymptomatic), metabolomic differences were observed between the two groups, and after the age of 25 weeks the metabolomic alterations became more pronounced. The metabolomic changes at different ages corroborated well with the phenotype changes in this transgenic mice model. Several useful candidate biomarkers including methionine, desaminotyrosine, taurine, N1-acetylspermidine, and 5-hydroxyindoleacetic acid were identified. Some of them were found in previous metabolomics studies in human cerebrospinal fluid or blood samples. This work illustrates the utility of this isotope labeling LC-MS method for biomarker discovery using mouse urine metabolomics.

Development of a metabolomic radiation signature in urine from patients undergoing total body irradiation.

The emergence of the threat of radiological terrorism and other radiological incidents has led to the need for development of fast, accurate and noninvasive methods for detection of radiation exposure. The purpose of this study was to extend radiation metabolomic biomarker discovery to humans, as previous studies have focused on mice. Urine was collected from patients undergoing total body irradiation at Memorial Sloan-Kettering Cancer Center prior to hematopoietic stem cell transplantation at 4-6 h postirradiation (a single dose of 1.25 Gy) and 24 h (three fractions of 1.25 Gy each). Global metabolomic profiling was obtained through analysis with ultra performance liquid chromatography coupled to time-of-flight mass spectrometry (TOFMS). Prior to further analyses, each sample was normalized to its respective creatinine level. Statistical analysis was conducted by the nonparametric Kolmogorov-Smirnov test and the Fisher's exact test and markers were validated against pure standards. Seven markers showed distinct differences between pre- and post-exposure samples. Of those, trimethyl-l-lysine and the carnitine conjugates acetylcarnitine, decanoylcarnitine and octanoylcarnitine play an important role in the transportation of fatty acids across mitochondria for subsequent fatty acid β-oxidation. The remaining metabolites, hypoxanthine, xanthine and uric acid are the final products of the purine catabolism pathway, and high levels of excretion have been associated with increased oxidative stress and radiation induced DNA damage. Further analysis revealed sex differences in the patterns of excretion of the markers, demonstrating that generation of a sex-specific metabolomic signature will be informative and can provide a quick and reliable assessment of individuals in a radiological scenario. This is the first radiation metabolomics study in human urine laying the foundation for the use of metabolomics in biodosimetry and providing confidence in biomarker identification based on the overlap between animal models and humans.

Comparative analysis of volatile metabolomics signals from melanoma and benign skin: a pilot study

The analysis of volatile organic compounds (VOC) as biomarkers of cancer is both promising and challenging. In this pilot study, we used an untargeted approach to compare volatile metabolomic signatures of melanoma and matched control non-neoplastic skin from the same patient. VOC from fresh (non-fixed) biopsied tissue were collected using the headspace solid phase micro extraction method (HS SPME) and analyzed by gas chromatography and mass spectrometry (GCMS). We applied the XCMS analysis platform and MetaboAnalyst software to reveal many differentially expressed metabolic features. Our analysis revealed increased levels of lauric acid (C12:0) and palmitic acid (C16:0) in melanoma. The identity of these compounds was confirmed by comparison with chemical standards. Increased levels of these fatty acids are likely to be a consequence of up-regulated de novo lipid synthesis, a known characteristic of cancer. Increased oxidative stress is likely to cause an additional increase in lauric acid. Implementation of this study design on larger number of cases will be necessary for the future metabolomics biomarker discovery applications.Electronic supplementary materialThe online version of this article (doi:10.1007/s11306-013-0523-z) contains supplementary material, which is available to authorized users.