Gaining Insights Into Metabolic Networks Using Chemometrics and Bioinformatics: Chronic Kidney Disease as a Clinical Model.

Because of its ability to generate biological hypotheses, metabolomics offers an innovative and promising approach in many fields, including clinical research. However, collecting specimens in this setting can be difficult to standardize, especially when groups of patients with different degrees of disease severity are considered. In addition, despite major technological advances, it remains challenging to measure all the compounds defining the metabolic network of a biological system. In this context, the characterization of samples based on several analytical setups is now recognized as an efficient strategy to improve the coverage of metabolic complexity. For this purpose, chemometrics proposes efficient methods to reduce the dimensionality of these complex datasets spread over several matrices, allowing the integration of different sources or structures of metabolic information. Bioinformatics databases and query tools designed to describe and explore metabolic network models offer extremely useful solutions for the contextualization of potential biomarker subsets, enabling mechanistic hypotheses to be considered rather than simple associations. In this study, network principal component analysis was used to investigate samples collected from three cohorts of patients including multiple stages of chronic kidney disease. Metabolic profiles were measured using a combination of four analytical setups involving different separation modes in liquid chromatography coupled to high resolution mass spectrometry. Based on the chemometric model, specific patterns of metabolites, such as N-acetyl amino acids, could be associated with the different subgroups of patients. Further investigation of the metabolic signatures carried out using genome-scale network modeling confirmed both tryptophan metabolism and nucleotide interconversion as relevant pathways potentially associated with disease severity. Metabolic modules composed of chemically adjacent or close compounds of biological relevance were further investigated using carbon transfer reaction paths. Overall, the proposed integrative data analysis strategy allowed deeper insights into the metabolic routes associated with different groups of patients to be gained. Because of their complementary role in the knowledge discovery process, the association of chemometrics and bioinformatics in a common workflow is therefore shown as an efficient methodology to gain meaningful insights in a clinical context.

Network principal component analysis: a versatile tool for the investigation of multigroup and multiblock datasets.

MOTIVATION: Complex data structures composed of different groups of observations and blocks of variables are increasingly collected in many domains, including metabolomics. Analysing these high-dimensional data constitutes a challenge, and the objective of this article is to present an original multivariate method capable of explicitly taking into account links between data tables when they involve the same observations and/or variables. For that purpose, an extension of standard principal component analysis called NetPCA was developed. RESULTS: The proposed algorithm was illustrated as an efficient solution for addressing complex multigroup and multiblock datasets. A case study involving the analysis of metabolomic data with different annotation levels and originating from a chronic kidney disease (CKD) study was used to highlight the different aspects and the additional outputs of the method compared to standard PCA. On the one hand, the model parameters allowed an efficient evaluation of each group's influence to be performed. On the other hand, the relative relevance of each block of variables to the model provided decisive information for an objective interpretation of the different metabolic annotation levels. AVAILABILITY AND IMPLEMENTATION: NetPCA is available as a Python package with NumPy dependencies. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Exploring blood alterations in chronic kidney disease and haemodialysis using metabolomics.

Chronic kidney disease (CKD) is characterized by retention of uremic solutes. Compared to patients with non-dialysis dependent CKD, those requiring haemodialysis (HD) have increased morbidity and mortality. We wished to characterise metabolic patterns in CKD compared to HD patients using metabolomics. Prevalent non-HD CKD KDIGO stage 3b-4 and stage 5 HD outpatients were screened at a single tertiary hospital. Various liquid chromatography approaches hyphenated with mass spectrometry were used to identify 278 metabolites. Unsupervised and supervised data analyses were conducted to characterize metabolic patterns. 69 patients were included in the CKD group and 35 in the HD group. Unsupervised data analysis showed clear clustering of CKD, pre-dialysis (preHD) and post-dialysis (postHD) patients. Supervised data analysis revealed qualitative as well as quantitative differences in individual metabolites profiles between CKD, preHD and postHD states. An original metabolomics framework could discriminate between CKD stages and highlight HD effect based on 278 identified metabolites. Significant differences in metabolic patterns between CKD and HD patients were found overall as well as for specific metabolites. Those findings could explain clinical discrepancies between patients requiring HD and those with earlier stage of CKD.

Implementation of liquid chromatography-high resolution mass spectrometry methods for untargeted metabolomic analyses of biological samples: A tutorial.

Untargeted metabolomics is now widely recognized as a useful tool for exploring metabolic changes taking place in biological systems under different conditions. By its nature, this is a highly interdisciplinary field of research, and mastering all of the steps comprised in the pipeline can be a challenging task, especially for those researchers new to the topic. In this tutorial, we aim to provide an overview of the most widely adopted methods of performing LC-HRMS-based untargeted metabolomics of biological samples. A detailed protocol is provided in the Supplementary Information for rapidly implementing a basic screening workflow in a laboratory setting. This tutorial covers experimental design, sample preparation and analysis, signal processing and data treatment, and, finally, data analysis and its biological interpretation. Each section is accompanied by up-to-date literature to guide readers through the preparation and optimization of such a workflow, as well as practical information for avoiding or fixing some of the most frequently encountered pitfalls.

Combining the advantages of multilevel and orthogonal partial least squares data analysis for longitudinal metabolomics: Application to kidney transplantation.

Kidney transplantation is one of the renal replacement options in patients suffering from end-stage renal disease (ESRD). After a transplant, patient follow-up is essential and is mostly based on immunosuppressive drug levels control, creatinine measurement and kidney biopsy in case of a rejection suspicion. The extensive analysis of metabolite levels offered by metabolomics might improve patient monitoring, help in the surveillance of the restoration of a "normal" renal function and possibly also predict rejection. The longitudinal follow-up of those patients with repeated measurements is useful to understand changes and decide whether an intervention is necessary. The time modality, therefore, constitutes a specific dimension in the data structure, requiring dedicated consideration for proper statistical analysis. The handling of specific data structures in metabolomics has received strong interest in recent years. In this work, we demonstrated the recently developed ANOVA multiblock OPLS (AMOPLS) to efficiently analyse longitudinal metabolomic data by considering the intrinsic experimental design. Indeed, AMOPLS combines the advantages of multilevel approaches and OPLS by separating between and within individual variations using dedicated predictive components, while removing most uncorrelated variations in the orthogonal component(s), thus facilitating interpretation. This modelling approach was applied to a clinical cohort study aiming to evaluate the impact of kidney transplantation over time on the plasma metabolic profile of graft patients and donor volunteers. A dataset of 266 plasma metabolites was identified using an LC-MS multiplatform analytical setup. Two separate AMOPLS models were computed: one for the recipient group and one for the donor group. The results highlighted the benefits of transplantation for recipients and the relatively low impacts on blood metabolites of donor volunteers.

Choosing an Optimal Sample Preparation in Caulobacter crescentus for Untargeted Metabolomics Approaches.

Untargeted metabolomics aims to provide a global picture of the metabolites present in the system under study. To this end, making a careful choice of sample preparation is mandatory to obtain reliable and reproducible biological information. In this study, eight different sample preparation techniques were evaluated using Caulobacter crescentus as a model for Gram-negative bacteria. Two cell retrieval systems, two quenching and extraction solvents, and two cell disruption procedures were combined in a full factorial experimental design. To fully exploit the multivariate structure of the generated data, the ANOVA multiblock orthogonal partial least squares (AMOPLS) algorithm was employed to decompose the contribution of each factor studied and their potential interactions for a set of annotated metabolites. All main effects of the factors studied were found to have a significant contribution on the total observed variability. Cell retrieval, quenching and extraction solvent, and cell disrupting mechanism accounted respectively for 27.6%, 8.4%, and 7.0% of the total variability. The reproducibility and metabolome coverage of the sample preparation procedures were then compared and evaluated in terms of relative standard deviation (RSD) on the area for the detected metabolites. The protocol showing the best performance in terms of recovery, versatility, and variability was centrifugation for cell retrieval, using MeOH:H2O (8:2) as quenching and extraction solvent, and freeze-thaw cycles as the cell disrupting mechanism.

Toward a better understanding of chronic kidney disease with complementary chromatographic methods hyphenated with mass spectrometry for improved polar metabolome coverage.

The prevalence of chronic kidney disease (CKD) is increasing worldwide. New technical approaches are needed to improve early diagnosis, disease understanding and patient monitoring, and to evaluate new therapies. Metabolomics, as a prime candidate in the field of CKD research, aims to comprehensively analyze the metabolic complexity of biological systems. An extensive analysis of the metabolites contained in biofluids is therefore needed, and the combination of data obtained from multiple analytical platforms constitutes a promising methodological approach. This study presents an original workflow based on complementary chromatographic conditions, reversed-phase and hydrophilic interaction chromatography hyphenated to mass spectrometry to improve the polar metabolome coverage coupled with a univocal metabolite annotation strategy enabling a rapid access to the biological interpretation. This multiplatform workflow was applied in a CKD cohort study to assess plasma metabolic profile modifications related to renal disease. Multivariate analysis of 278 endogenous annotated metabolites enabled patient stratification with respect to CKD stages and helped to generate new biological insights, while also confirming the relevance of tryptophan metabolism pathway in this condition.

A scoring approach for multi-platform acquisition in metabolomics.

Since the ultimate goal of untargeted metabolomics is the analysis of the broadest possible range of metabolites, some new metrics have to be used by researchers to evaluate and select different analytical strategies when multi-platform analyses are considered. In this context, we aimed at developing a scoring approach allowing to compare the performance of different LC-MS conditions for metabolomics studies. By taking into account both chromatographic and MS attributes of the analytes' peaks (i.e. retention, signal-to-noise ratio, peak intensity and shape), the newly proposed score reflects the potential of a set of LC-MS operating conditions to provide useful analytical information for a given compound. A chemical library containing 597 metabolites was used as a benchmark to apply this approach on two RPLC and three HILIC methods hyphenated to high resolution mass spectrometry (HRMS) in positive and negative ionization modes. The scores not only allowed to evaluate each analytical platform, but also to optimize the number of analytical methods needed for the analysis of metabolomics samples. As a result, the most informative combination of three LC methods and ionization modes was found, leading to a coverage of nearly 95% of the detected compounds. It was therefore demonstrated that the overall performance reached with three selected methods was almost equivalent to the performance reached when five LC-MS conditions were used.

Metabolomics in chronic kidney disease: Strategies for extended metabolome coverage.

Chronic kidney disease (CKD) is becoming a major public health issue as prevalence is increasing worldwide. It also represents a major challenge for the identification of new early biomarkers, understanding of biochemical mechanisms, patient monitoring and prognosis. Each metabolite contained in a biofluid or tissue may play a role as a signal or as a driver in the development or progression of the pathology. Therefore, metabolomics is a highly valuable approach in this clinical context. It aims to provide a representative picture of a biological system, making exhaustive metabolite coverage crucial. Two aspects can be considered: analytical and biological coverage. From an analytical point of view, monitoring all metabolites within one run is currently impossible. Multiple analytical techniques providing orthogonal information should be carried out in parallel for coverage improvement. The biological aspect of metabolome coverage can be enhanced by using multiple biofluids or tissues for in-depth biological investigation, as the analysis of a single sample type is generally insufficient for whole organism extrapolation. Hence, recording of signals from multiple sample types and different analytical platforms generates massive and complex datasets so that chemometric tools, including data fusion approaches and multi-block analysis, are key tools for extracting biological information and for discovery of relevant biomarkers. This review presents the recent developments in the field of metabolomic analysis, from sampling and analytical strategies to chemometric tools, dedicated to the generation and handling of multiple complementary metabolomic datasets enabling extended metabolite coverage to improve our biological knowledge of CKD.

Effective mobility as a robust criterion for compound annotation and identification in metabolomics: Toward a mobility-based library.

Capillary electrophoresis (CE) presents many advantageous features as an analytical technique in metabolomics, such as very low consumption of a sample or the possibility to easily detect very polar and ionizable compounds. However, CE remains an approach only used by a few research groups due to a relatively lower sensitivity and, higher analysis time compared to liquid chromatography. To circumvent these drawbacks, herein we propose a generic CE-mass spectrometry (MS) approach using positive electrospray ionization mode and performing normal- and reverse-polarity CE separations to analyze anionic and acidic compounds. Preliminary experiments showed better sensitivity using the ESI positive mode compared to the ESI negative mode on a set of representative anionic compounds from different biochemical families. This approach was applied to the investigation of an available library of metabolites. More than 450 compounds out of the 596 in the library were detected, with the possibility to monitor negatively ionizable compounds through their ammonium adducts. Migration time of each data point was converted to an effective mobility (µeff) scale and used for peak alignment in data pre-processing; µeff features were used as a robust migration index for peak annotation and identification criterion. For the first time, a large database based on experimental µeff was built, allowing for the straightforward annotation of detected features in biological samples and demonstrating how CE-MS can complement other analytical techniques commonly used in metabolomics.

Applicability of supercritical fluid chromatography - mass spectrometry to metabolomics. I - Optimization of separation conditions for the simultaneous analysis of hydrophilic and lipophilic substances.

The aim of this study was to evaluate the suitability of SFC-MS for the analysis of a wide range of compounds including lipophilic and highly hydrophilic substances (log P values comprised between -6 and 11), for its potential application toward human metabolomics. For this purpose, a generic unified chromatography gradient from 2 to 100% organic modifier in CO2 was systematically applied. In terms of chemistry, the best stationary phases for this application were found to be the Agilent Poroshell HILIC (bare silica) and Macherey-Nagel Nucleoshell HILIC (silica bonded with a zwitterionic ligand). To avoid system overpressure at very high organic modifier proportion, columns of 100 × 3 mm I.D. packed with sub-3 µm superficially porous particles were selected. In terms of organic modifier, a mixture of 95% MeOH and 5% water was selected, with 50 mM ammonium formate and 1 mM ammonium fluoride, to afford good solubility of analytes in the mobile phase, limited retention for the most hydrophilic metabolites and suitable peak shapes of ionizable species. A sample diluent containing 50%ACN/50% water was employed as injection solvent. These conditions were applied to a representative set of metabolites belonging to nucleosides, nucleotides, small organic acids, small bases, sulfated/sulfonated metabolites, poly-alcohols, lipid related substances, quaternary ammonium metabolites, phosphate-based substances, carbohydrates and amino acids. Among all these metabolites, 65% of the compounds were adequately analyzed with excellent peak shape, 23% provided distorted peak shapes, while only 12% were not detected (mostly metabolites having several phosphate or several carboxylic acid groups).

Steroid profiles in both blood serum and seminal plasma are not correlated and do not reflect sperm quality: Study on the male reproductive health of fifty young Swiss men.

Steroids play an important role in sperm production and quality. These hormones have been extensively studied in blood, but poorly investigated in semen. The purpose of our study was to evaluate the relationship between sperm quality and steroid profiles in blood and semen in a small cohort of young Swiss men. Another objective was to determine whether the presence of xenobiotics or drugs could influence these profiles. Semen analysis was performed according to WHO guidelines, and steroid profiles in blood serum and seminal plasma were determined by two complementary approaches: a targeted investigation involving the quantification of a limited number of relevant steroids for testing putative correlations with sperm parameters and a global "steroidomic" analysis highlighting their complex metabolic relationship. Results showed that steroid profiles are distinct within blood and seminal fluid. No significant correlation was found between individual steroids measured in blood and in semen, demonstrating the relevance of assessing hormone levels in both fluids. Moreover, testosterone and androstenedione levels were significantly correlated in semen but not in blood. None of the evaluated spermiogram parameters was linked to steroid levels measured in any medium. The steroidomic analyses confirmed that the steroids present in both fluids are different and that there is no correlation with spermiogram parameters. Finally, upon toxicological screening, we observed that all the three samples positive for tetrahydrocannabinol, which is known to act as an endocrine disruptor, displayed low seminal testosterone concentrations. In conclusion, we did not find any evidence suggesting using steroid profiles, neither in blood nor in semen, as surrogates for sperm analyses. However, steroid profiles could be useful biomarkers of individual exposure to endocrine disruptors.

ROMANCE: A new software tool to improve data robustness and feature identification in CE-MS metabolomics.

The use of capillary electrophoresis coupled to mass spectrometry (CE-MS) in metabolomics remains an oddity compared to the widely adopted use of liquid chromatography. This technique is traditionally regarded as lacking the reproducibility to adequately identify metabolites by their migration times. The major reason is the variability of the velocity of the background electrolyte, mainly coming from shifts in the magnitude of the electroosmotic flow and from the suction caused by electrospray interfaces. The use of the effective electrophoretic mobility is one solution to overcome this issue as it is a characteristic feature of each compound. To date, such an approach has not been applied to metabolomics due to the complexity and size of CE-MS data obtained in such studies. In this paper, ROMANCE (RObust Metabolomic Analysis with Normalized CE) is introduced as a new software for CE-MS-based metabolomics. It allows the automated conversion of batches of CE-MS files with minimal user intervention. ROMANCE converts the x-axis of each MS file from the time into the effective mobility scale and the resulting files are already pseudo-aligned, present normalized peak areas and improved reproducibility, and can eventually follow existing metabolomic workflows. The software was developed in Scala, so it is multi-platform and computationally-efficient. It is available for download under a CC license. In this work, the versatility of ROMANCE was demonstrated by using data obtained in the same and in different laboratories, as well as its application to the analysis of human plasma samples.

Metabolomic analysis of urine samples by UHPLC-QTOF-MS: Impact of normalization strategies.

Among the various biological matrices used in metabolomics, urine is a biofluid of major interest because of its non-invasive collection and its availability in large quantities. However, significant sources of variability in urine metabolomics based on UHPLC-MS are related to the analytical drift and variation of the sample concentration, thus requiring normalization. A sequential normalization strategy was developed to remove these detrimental effects, including: (i) pre-acquisition sample normalization by individual dilution factors to narrow the concentration range and to standardize the analytical conditions, (ii) post-acquisition data normalization by quality control-based robust LOESS signal correction (QC-RLSC) to correct for potential analytical drift, and (iii) post-acquisition data normalization by MS total useful signal (MSTUS) or probabilistic quotient normalization (PQN) to prevent the impact of concentration variability. This generic strategy was performed with urine samples from healthy individuals and was further implemented in the context of a clinical study to detect alterations in urine metabolomic profiles due to kidney failure. In the case of kidney failure, the relation between creatinine/osmolality and the sample concentration is modified, and relying only on these measurements for normalization could be highly detrimental. The sequential normalization strategy was demonstrated to significantly improve patient stratification by decreasing the unwanted variability and thus enhancing data quality.