May 1, 2013 (Vol. 33, No. 9)
Kate Marusina Ph.D.
Metabolomics analyzes and quantifies all metabolites in a given biological sample. Powerful instrumentation along with sophisticated chemometric software allows for comparison of changes in thousands of chemical entities.
Metabolomics is becoming sufficiently mature to complement other systems biology tools and to deliver medically relevant results.
According to some metabolomics researchers, continued development of metabolomics platforms and technologies will inevitably lead to development of rapid and comprehensive disease diagnostics. This and other ideas were discussed at OMICS group’s “Metabolomics and Systems Biology” conference last month.
“Interpretations of metabolomics signatures will make rapid bedside diagnostics possible,” agreed Chris Beecher, Ph.D., CSO, NextGen Metabolomics. “In 5–10 years this information will transform our diagnostic capabilities. The meaningful metabolome interpretation will be available to a physician within minutes for a fraction of the cost of today’s analyses.”
NextGen is contributing to this vision by perfecting Isotopic Ratio Outlier Analysis™ (IROA), a mass-spectrometry-based protocol that enables efficient identification of all biological metabolites in a sample and their relative concentration, said Dr. Beecher.
The technology has its roots in a physiochemical phenomenon of naturally occurring C13 atoms. C13 has an extra neutron producing a pattern of additional peaks for each metabolite in a sample. Because of their low natural abundance and the presence of other elemental isotopes, these additional peaks (called M+1, M+2, etc.) are very small and not very informative.
“We simply boosted the C13 concentration by growing cells in a culture media engineered with precise balances of C13-bearing components,” continued Dr. Beecher. “When you compare a culture grown with 5% C13 with culture grown in 95% C13, each metabolite will be represented by a symmetrical pattern of C12 and C13 peaks. The parameters of this pattern tell us how many carbons the molecule contains and its mass.”
The mirror symmetry of MS peaks in IROA experiments enables rapid classification of all peaks as real cell products (have mirror counterparts) or artifacts and contaminants (no enhanced M+1, M+2, etc.). NextGen computer algorithms look for the same symmetry to perform phenotyping experiments.
To identify metabolites in tissue biopsies, the tissue cells are analyzed together with “standard” cells that were isotopically labeled with IROA C13 media. The tissue-derived metabolites can be easily identified by their mass and position relative to the standard.
The company demonstrated a significant proof of principle for toxicology analysis of drug candidates. Dr. Beecher’s team used IROA to show the response of every metabolite in the cell to flucytosine, a well-characterized inhibitor of DNA synthesis. The data showed clear response in metabolic pathways related to nucleotide biosynthesis, but not in other pathways.
“We plan to perfect this technology to answer fundamental questions in drug development,” commented Dr. Beecher. “What toxicities should we expect? What is the biological reason for this toxicity?” NextGen plans to set up internal biomarker and toxicology discovery departments based on IROA technology.
Global Signaling Pathway Analysis
Antibody-based proteomic methods deploy sample fractionation followed by separation by liquid chromatography (LC) and mass spectrometry (MS). Such enrichment strategies lower the “noise” from abundant proteins and focus the instrument time on the proteins of interest.
“Our motif antibodies recognize a particular post-translational modification (PTM), such as phosphorylation or ubiquitination,” said Matthew P. Stokes, Ph.D., scientist III, proteomics service group, Cell Signaling Technology.
“Motif antibodies and other generalized enrichment strategies produce large datasets but sometimes miss critical signaling proteins that are without the targeted motif or that are present at low levels. Our PTMScan™ Direct approach is unique; in addition to a PTM, our antibodies recognize the surrounding amino-acid context.”
Cell Signaling Technology generates hundreds of proprietary antibodies against specific post-translationally modified peptide targets in selected signaling pathways. Antibodies are formulated into one of the six PTMScan Direct Reagents. The DNA Damage/Cell Cycle Reagent, for instance, targets 263 sites on 137 proteins, and the Apoptosis/Autophagy Reagent targets 175 sites on 100 proteins.
“Each target peptide is rigorously validated using multiple strict criteria,” continued Dr. Stokes. “Moreover, we digest the starting material with trypsin, which helps to expose the binding site and ensures that single peptides are immunoprecipitated rather than whole proteins or protein complexes, further improving specificity of binding to our reagents. Our antibodies are capable of recognizing individual activation sites on a single protein.”
The company provides custom services for targeted screening and quantification of signaling pathways critical for drug and biomarker discovery. Added Dr. Stokes, “We are constantly innovating and developing new reagents and kits for our customers. But we also maintain an extensive R&D program identifying disease drivers in cancer.”
The company used its profiling technologies to demonstrate the aberrant activation of anaplastic lymphoma kinase (ALK) in ovarian cancer. The screening data was supported by further biological validation confirming ALK as a potential therapeutic target.
Further investigations shed light on the underlying mechanisms of acquired resistance to treatment with ALK tyrosine kinase inhibitors. PTMScan Direct Profiling of drug-resistant tumors revealed aberrations in multiple signaling pathways, providing the evidence for rational selection of combination therapies.
Energy Pathways in Alzheimer’s
“Our research focuses on evaluation of the dynamic changes of mitochondrial functioning in relation to the progression of Alzheimer’s disease (AD),” noted Eugenia Trushina, Ph.D., associate consultant, Mayo Clinic.
“Some assessments, such as mitochondrial trafficking and distribution in neurons, could only be done in animal models. However, metabolomic profiles of energy pathways could be developed by minimally invasive sampling of blood or cerebrospinal fluid. Moreover, these biochemical pathways are highly conserved between species, and therefore, easily translatable between mice and humans.”
Results of FDG-PET scans of AD patients implicate early changes in energy metabolism. Dr. Trushina’s team analyzed multiple mitochondrial metabolic pathways, including Krebs cycle, amino-acid metabolism, and energy transfer, using transgenic animal models of familial AD.
In comparison with nontransgenic mice, metabolic profiles of the three types of FAD mice revealed significant alterations. Distinction in metabolic profiles was observed early in AD development, before typical behavioral manifestations and prior to formation of amyloid plaques.
“It is an exciting time in metabolomic analysis,” continued Dr. Trushina. “New statistical methods and quantitative algorithms now allow for analysis of thousands of metabolites. Even though the structure of some of them may not have been definitively identified, we can use the analysis to draw a map of correlative changes.”
Using Mayo Clinic Alzheimer’s Disease Research Center and Mayo Clinic Study of Aging tissue repository, Dr. Trushina and colleagues demonstrated similar changes in mitochondrial pathways, along with alterations in pathways involved in lipid trafficking and homeostasis, in cerebrospinal fluid and plasma from patients with early AD.
The team is planning to use targeted metabolomics to look at specific changes in mitochondrial metabolites in an attempt to identify predictive biomarkers of early AD, to compare the metabolic profiles of patients with various genetic backgrounds and of the patients transitioning from mild cognitive impairment to AD.
“We hope that our research leads to the development of an in vivo drug testing model that would aid in the development of drugs targeting early pathways,” concluded Dr. Trushina. “Most current drugs attempt to combat the amyloid plaques. Our research indicates that by then it may be too late.”
Essential Lipid Pathways Uncovered
Lipids are a major constituent of food and are vitally important in our diet as an essential source of energy. Lipids play a key role in cell signaling, endocrine actions, and membrane functions.
“Lipidomics is an emerging field in need of comprehensive analytical approaches,” commented Giorgis Isaac, Ph.D., senior research scientist, Waters. “Waters has developed state-of-the-art instrumentation that could accommodate both global profiling of intact molecules and determination of lipid structure.”
Since the position of double bonds within lipid molecules impacts their signaling and other biological functions, localizing the bonds may be critical for understanding lipid metabolism. Waters has established itself as a player the lipidomics field with instruments equipped with ion mobility separation technology (IMS).
IMS has the capability to separate ions by size, charge, and collisional cross sections. The latter roughly measures the “safe zones” around molecules before they collide, which helps to determine sizes of molecules in close proximity. Waters’ SYNAPT G2-S High Definition MS system performs fragmentation of a target molecule both before and after IMS, which helps determine the double bond position in the lipids. IMS is a very rapid method of lipid profiling, providing a greater degree of confidence for lipid identification, said Dr. Isaac.
Scientists at Waters in collaboration with Jing X. Kang, M.D., Ph.D., at Harvard Medical School used SYNAPT G2-S to study lipid metabolism of a transgenic fat-1 mouse. This animal model is capable of what no other mammal can do—convert omega-6 into omega-3 fatty acids. The only difference between these two fatty acids is that for omega-3, double bonds start at the third carbon atom, and for omega-6, at the sixth carbon atom.
TransOmics™, Waters’ data-processing tool, performed quantitative comparison and statistical analyses to find those lipids that significantly change between the samples and then identify the actual molecules from the MS data. “We completed detailed profiling of bioactive lipid species in plasma and liver samples from wild type and fat-1 mice,” said Dr. Isaac. “This research provides new clues to the pathways and mechanisms that can be associated with health benefits of omega-3 fatty acids.”
NMR for Epidemiological Screening
“Mass spectrometry and nuclear magnetic resonance (NMR) play complementary roles in metabolomics analysis,” said Manfred Spraul, Ph.D., director of NMR applications, Bruker BioSpin. Bruker has expertise in analytical NMR applications for metabolomics profiling. The company has developed a suite of NMR-based products to support both food quality control and clinical applications.
“NMR has several advantages that position this technology for analysis of biological fluids,” continued Dr. Spraul. “NMR needs minimal sample preparation, and it is extremely reproducible and fully quantitative over the full dynamic range of concentrations.”
NMR reproducibility is what attracted the scientists to utilize it in large multinational phenome project studies. As long as the samples are collected under standard operating procedures and analyzed using the same magnetic field strength, the resulting data is readily comparable and exchangeable on a worldwide basis. Moreover, when a new pattern is found it can be compared with all previous data obtained under the same standards.
Dr. Spraul’s team is the key partner in a project to establish NMR-based screening of newborn babies at 12 hospitals in Turkey. First, the team built an NMR profile of a “normal” newborn, using over 1,000 urine samples. The model captures all possible individual differences within normal range, claimed Dr. Spraul. In an untargeted approach, each new sample is compared with this spectral fingerprint.
In targeted mode, 64 compounds indicative of inborn errors of metabolism are checked for deviations from the norm. Although rare, these deviations can be readily identified with high statistical probability even if previously unknown.
“We are interested in creating models for each day of the newborn development,” said Dr. Spraul. By matching samples with the model reference spectrum, physicians will be able to detect the first signs of latent hereditary metabolic errors, he explained.
Bruker also is a funding partner of the Imperial Clinical Phenome Centre, based at St. Mary’s Hospital in London. There, the NMR technology will be used to assist physicians by providing real-time data for the patients, especially in intensive care units.
NMR spectra could predict whether or not a patient is recovering, experiencing organ rejection, or responding to the medication. “This technology has matured and is reaching the point when it can be a real partner in the clinical decision-making process,” concluded Dr. Spraul.