|Year : 2017 | Volume
| Issue : 1 | Page : 1-5
Mass spectrometry-based metabolomics in biomarker-assisted drug discovery and oxidative stress research
A Kiran Kumar1, Sagarika Devi2, V Sivaram1
1 Department of Biochemistry and Clinical Pharmacology, National Institute for Research in Tuberculosis, Chennai, Tamil Nadu, India
2 Department of Biotechnology, Laboratory of Bioorganic Chemistry, Indian Institute of Technology Madras, Chennai, Tamil Nadu, India
|Date of Web Publication||18-Oct-2017|
Department of Biochemistry and Clinical Pharmacology, National Institute for Research in Tuberculosis, Chetpet, Chennai - 600 031, Tamil Nadu,
Source of Support: None, Conflict of Interest: None
Biomarker-assisted drug discovery is a major step in identification of drug efficacy and disease proliferation of both communicable and noncommunicable diseases. Biomarker-assisted drug discovery using metabolomics approach is now a reliable “path of opportunity” in drug discovery, as it offers precise insight into the efficacy of tested drug in the biological setup. With the advent of recent advances in instrumentation, accuracy and specificity of biomarker-based drug discovery program is improved. A key challenge in biomarker discovery is the choice of detection method, as diverse metabolites present in the physiological system pose varying levels of detection response with different instrumentation platforms. In biomarker-assisted early diagnosis, mass spectrometry-based metabolomics has a major advantage due to its capability of wide metabolite range coverage with great specificity. This review also discusses the advantage of biomarker-assisted discovery using metabolome analysis.
Keywords: Biomarker, informatics, mass spectrometry, metabolomics, multivariate analysis
|How to cite this article:|
Kumar A K, Devi S, Sivaram V. Mass spectrometry-based metabolomics in biomarker-assisted drug discovery and oxidative stress research. Drug Dev Ther 2017;8:1-5
|How to cite this URL:|
Kumar A K, Devi S, Sivaram V. Mass spectrometry-based metabolomics in biomarker-assisted drug discovery and oxidative stress research. Drug Dev Ther [serial online] 2017 [cited 2018 May 23];8:1-5. Available from: http://www.ddtjournal.org/text.asp?2017/8/1/1/216931
| Introduction|| |
Conventional natural product-based drug discovery approach follows a reductionist workflow, which involves screening of therapeutic active principle from a natural source (phytotherapeutics, animal secondary metabolites, and microbial metabolites) proceeded by solvent extraction, biological screening, and finally compound characterization routine. The choice of the natural source depends on traditional wisdom or laboratory experiments on the medicinal properties of the particular source. In the recent years, with the advent of high-throughput technologies and automated screening platforms handled by robotic system, it is possible to screen and generate a number of drug leads (molecules that are used as starting point for drug discovery) in a shorter duration through in silico method. In this approach, the drug lead is generated through computer algorithm and further scrutinized for its ADMET properties through nonlinear kinetics. The potential drug-like molecule that exhibits no toxicological profile is further synthesized and screened in vitro andin vivo for its desired biological activity.
| Drug Discovery Through Systems Biology Prism|| |
As drug discovery moved into the “Omics” era, the process of drug screening further improved with the aid of other biologically relevant information, such as genomics, proteomics, metabolites, and their complex pathway interactions. Collectively, the biological systems and its complex interactions decide the fate of cellular response to a particular drug; or in other words, the knowledge on the systems biology for a particular organism (pathogen, host, and drug source) can give more grip on the drug discovery program in terms of assessing the impact of a certain drug on the regulatory networks of genes, proteins, metabolites, and their complex interactions. A drug meant to target the particular metabolic process, i.e., expression of a gene, functioning of a protein, or metabolite turnover, can also influence the other vital components of the host metabolome. This phenomenon is inevitable since, in all the biological systems, small changes in metabolism are well reflected in whole systems; thus, metabolic pathways are interlinked, and are thus capable of dynamically altering the behavior of other housekeeping pathways. Therefore, a wide range of diseases are basically the result of multiple metabolic abnormalities. Hence, the yield of integrated top–down approach like metabolomics is enormous and more accurate in terms of holistic coverage of drug behavior in systems framework. Omics datasets offer more accurate information to build a predictable model in the format of multivariate predictors, which can aid in silico drug discovery program for simulation purposes. The milieu of “systems” may vary with the context we deal with, such as host or pathogen or drug source. New era of the “omics” data comes from a variety of sources, but can be broadly classified into three major areas based on its chemical nature, namely, genomics (transcriptional and translational), proteomics, and metabolomics. Among these three, metabolomics data offer advantage over the others, as it enables direct readout of the metabolic status. The foundation of the successful metabolomics-based clinical pharmacological program lies on suitable pattern recognition algorithms that efficiently differentiate the metabolomics data coming from “normal” versus “experimental” condition. For instance, unsupervised multivariate analysis tool (MVA) like principal component analysis is used to assess the pattern variation between two groups. Likewise, supervised MVA procedures such as partial least squares, discriminant function analysis, and support vector machine are presently employed to extract putative biomarker features from metabolomics data. These features are compared between groups to identify the patterns and their differences, which lead to validation of relevant biomarkers. The choice of these algorithms depends on the nature of information to be extracted from the metabolomics data. In clinical pharmacology, metabolomics offers a wide range of possibilities such as assessment of patient response, ranging from population level to individual response toward a particular drug. Metabolomics-based biomarkers can facilitate the drug discovery program through evaluation of pathway interaction of the compound through metabolic network analysis. As significant as statistical strategies in metabolomics, laboratory-based analytical techniques are also important for a successful metabolomics approach in clinical pharmacology. Techniques, ranging from simple Ultraviolet-visible spectrophotometer “reads” to higher end spectrometric techniques such as nuclear magnetic resonance and mass spectrometry (MS) platforms, can generate clinically relevant metabolomics data. In particular, combining two or more spectrometric techniques can be more powerful than relying on a single platform. The metabolomics readouts can be either shotgun in nature or coupled with a separation source such as liquid chromatography (LC), gas chromatography (GC), and capillary electrophoresis. Each platform has its own unique advantage solely based on the metabolite coverage intended to validate, but in particular, MS offers a more robust and reliable platform for metabolomics analysis.
| Role of Liquid Chromatography-Mass Spectrometry Metabolomics in Drug Discovery|| |
MS is classified based on the choice of ionization source, along with the separation source. Two prominent separation sources coupled with mass spectrometers are natural choices for analysis, namely, LC-MS and GC-MS. Both techniques have different application areas, assisted with open-source and proprietary-based compound libraries readily available for precise metabolite identification. While GC-MS-based metabolome studies are dominant over other methods, LC-MS-based method can be a powerful tool for ab initio compound identification facilitated through soft ionization technique and tandem MS/MS (MSn) fragmentation. Soft ionization techniques such as electrospray ionization, atmosphere pressure chemical ionization, and atmospheric pressure photoionization coupled with triple quadrupole and time of flight provide LC-MS metabolomics with more flexible options depending on the chemical nature of the metabolite. [Table 1] summarizes the ionization modules available in LC-MS-based metabolomics studies. Early diagnosis of the disease certainly provides ample opportunity to cure it than late diagnosis which narrows down the chances of successful treatment and survival. As we stated earlier, metabolomics can offer comprehensive “snap shot” of the metabolic status in patient metabolome at early stages of disease progression, thus making it possible to discover and assign the biomarker molecules that indicate the disease prevalence. A typical LC-MS metabolomics-based biomarker discovery involves comparing the spectra of patient samples (plasma, urine, saliva, tissue biopsy, etc.) collected at early stage with normal healthy controls [Figure 1]. The pattern variance between “disease versus normal” spectra was further tested using multivariate statistical tools (such as MVA) to identify those possible features responsible for the potential disease positives.
|Table 1: Ionization modules available for liquid chromatography and their possible range of metabolite coverage in liquid chromatography-mass spectrometry metabolomics|
Click here to view
|Figure 1: Schematic workflow of liquid chromatography-mass spectrometry-based metabolomics experiment|
Click here to view
| Biomarker -Assisted Targeted Metabolomics In Drug Discovery|| |
Biomarkers assist the high throughput drug screening routine, as it can be used to rapidly gauge the organism's response towards drug molecule, in terms of toxicity efficacy and possible metabolic side effects. For example, in the early stages of cancer prevalence, accelerated incidence of glycolysis (Warburg's effect) can be a suitable indicator for cells that could turn potentially cancerous. Other upregulated pathways include pentose phosphate pathway and glutaminolysis that support the highly demanding energy budget of cancerous cells. In this case, glutamine breakdown and aerobic glycolysis metabolites can be a significant indicator for upregulated glutaminolysis in the potentially cancerous cells. In this case, multireaction monitoring (MRM)-based targeted metabolomics come as a handy tool to study the pathway interactions as whole than studying individual biomarkers as a diagnosis indicator.
| Oxidative Stress Biomarkers–mass Spectrometry-Based Metabolomics Approaches|| |
Oxygen, a Janus-faced molecule, which is the main component of life processes, can also cause significant damage if goes unchecked through proper redox mechanism. Oxidative stress is the starting point for many diseases stretching from simple muscular pain to cancer, Alzheimer's disease, premature aging, neurodegeneration, and multiple sclerosis. Early detection of free radical-mediated damage can give a great deal of assistance in devising a successful clinical program to tackle these life-threatening diseases. Biomarker-based diagnosis of earlier onset of disease through LC-MS by either single biomarker assessment or multibiomarker assessment (MBA) can serve as a handy tool, while MBA is to facilitate effective individualization of the treatment course with much specificity. These MBAs can give a holistic picture of the disease origin such as diet, habit, genetics, pathogen, and environment. [Table 2] summarizes the common LC-MS-based biomarkers to assess the rate of free radical-mediated damage in patients. Apart from redox biomarker assessment, LC-MS-based metabolomics assists the drug discovery program through creating new drug-lead inventories from wild sources with unique adaptation to oxidative stress. In our laboratory, initiatives are taken to identify/understand key metabolic adaptations in lower eukaryote extremophile organisms and to profile/reconstruct the pathways behind such adaptation through the LC-MS platform. The information extracted from metabolomics can be taken further to develop a “systems” framework through preexisting cross-species metabolome pathways such as Kyoto Encyclopedia of Genes and Genomes, MetaCyc, EcoCyc, and Pathway commons.
|Table 2: Common oxidative stress biomarkers for mass spectrometry-based metabolomics studies|
Click here to view
| Current Challenges in Liquid Chromatography-Mass Spectrometry Metabolomics|| |
Like any other “omics”-based instrumentation platforms, MS also requires reliable computational algorithms to process the signal sources, identify the compound arising from the signal, and model the role of the identified metabolite in the whole metabolome. Currently, two approaches are followed, i.e., targeted and untargeted [Figure 2]. In targeted metabolomic analysis, known metabolites that already resolved for their role as disease biomarker are analyzed using quantitative methods through MRM. This mode of analysis helps in determining the influence of drug on the overall patient metabolome during the treatment regimen. On the other hand, untargeted metabolomics focus on unknown metabolites and their qualitative identification through MS/MS fragment analysis. Alternatively referred as “discovery” mode, untargeted metabolomics play a major role in the discovery of novel biomarkers and drug targets. Untargeted metabolomics has unique advantages such as identification of novel metabolites and their role in the metabolic networks, pathway reconstruction analysis, and scope of a metabolite in whole cellular processes. Although both targeted and untargeted metabolome analyses have their own advantage and utility, both approaches posses their own limitation to overcome. In targeted metabolome analysis, it is important to devise a successful extraction strategy to cover the desired list of metabolites to be monitored, while in untargeted metabolome analysis, limitations include signal separation and discrimination of metabolite signal from overall noise. In particular, untargeted metabolomics data give a serious challenge to computational biology, such as the detection and identification of new metabolites, which does not preexist in compound libraries or any other literatures. Complexity of mass signals (adducts, fragmentation, and contamination), signal-to-noise ratio, and metabolite decomposition also need to be carefully addressed while identifying a metabolite during an LC-MS untargeted analysis.
|Figure 2: Comparative account on the approaches followed in “Targeted versus Untargeted” metabolomics analysis|
Click here to view
| Conclusion|| |
MS-based metabolomics needs significant efforts on the signal processing domain with the backing of effective computational algorithms to deal with complex redundant mass spectra. Nonetheless, MS-based metabolomics has been providing consistent insights into the metabolome snapshots during disease progression, biomarker-mediated successful treatment, and protocol validation. Most of these efforts are at laboratory scale, yet MS-based metabolomics find few real-time clinical applications such as neonatal screening for inborn diseases, oxidative stress biomarkers, early diagnosis of cancer, and nutritional-related disorders.
Financial support and sponsorship
Dr Sivaram thanks Science and Engineering Research Board (SERB), Government of India, for the financial support extended through grant No SB/FT/LS-296/2012.
Conflicts of interest
There are no conflicts of interest.
| References|| |
Swinney DC, Anthony J. How were new medicines discovered? Nat Rev Drug Discov 2011;10:507-19.
Dearden JC. In silico
prediction of drug toxicity. J Comput Aided Mol Des 2003;17:119-27.
Butcher EC, Berg EL, Kunkel EJ. Systems biology in drug discovery. Nat Biotechnol 2004;22:1253-9.
Butcher EC. Can cell systems biology rescue drug discovery? Nat Rev Drug Discov 2005;4:461-7.
Li S, Park Y, Duraisingham S, Strobel FH, Khan N, Soltow QA, et al.
Predicting network activity from high throughput metabolomics. PLoS Comput Biol 2013;9:e1003123.
Worley B, Powers R. Multivariate analysis in metabolomics. Curr Metabolomics 2013;1:92-107.
Kaddurah-Daouk R, Kristal BS, Weinshilboum RM. Metabolomics: A global biochemical approach to drug response and disease. Annu Rev Pharmacol Toxicol 2008;48:653-83.
Dettmer K, Aronov PA, Hammock BD. Mass spectrometry-based metabolomics. Mass Spectrom Rev 2007;26:51-78.
Zhou B, Xiao JF, Tuli L, Ressom HW. LC-MS-based metabolomics. Mol Biosyst 2012;8:470-81.
van der Kloet FM, Tempels FW, Ismail N, van der Heijden R, Kasper PT, Rojas-Cherto M, et al.
Discovery of early-stage biomarkers for diabetic kidney disease using ms-based metabolomics (FinnDiane study). Metabolomics 2012;8:109-19.
Castillo S, Gopalacharyulu P, Yetukuri L, Orešič M. Algorithms and tools for the preprocessing of LC–MS metabolomics data. Chemom Intell Lab Syst 2011;108:23-32.
Weljie AM, Jirik FR. Hypoxia-induced metabolic shifts in cancer cells: Moving beyond the Warburg effect. Int J Biochem Cell Biol 2011;43:981-9.
Beal MF. Aging, energy, and oxidative stress in neurodegenerative diseases. Ann Neurol 1995;38:357-66.
Zethelius B, Berglund L, Sundström J, Ingelsson E, Basu S, Larsson A, et al.
Use of multiple biomarkers to improve the prediction of death from cardiovascular causes. N
Engl J Med 2008;358:2107-16.
Maritim AC, Sanders RA, Watkins JB 3rd
. Diabetes, oxidative stress, and antioxidants: A review. J Biochem Mol Toxicol 2003;17:24-38.
Fiehn O, Kind T, Barupal DK, Data Processing, Metabolomic Databases and Pathway Analysis. Annual Plant Reviews; Vol 43; Biology of Plant Metabolomics; Edited by Robert Hall; Wiley-Blackwell. 2011;43;367-406.
Vuckovic D. Current trends and challenges in sample preparation for global metabolomics using liquid chromatography-mass spectrometry. Anal Bioanal Chem 2012;403:1523-48.
Ceglarek U, Leichtle A, Brügel M, Kortz L, Brauer R, Bresler K, et al.
Challenges and developments in tandem mass spectrometry based clinical metabolomics. Mol Cell Endocrinol 2009;301:266-71.
Xiao JF, Zhou B, Ressom HW. Metabolite identification and quantitation in LC-MS/MS-based metabolomics. Trends Analyt Chem 2012;32:1-4.
[Figure 1], [Figure 2]
[Table 1], [Table 2]