image   Posters

Metabolomics Society 2016

Felice de Jong1, Chris Beecher1,2
[1] IROA Technologies, Ann Arbor, MI, [2] University of Florida, Gainesville, FL
The standard workflow for unbiased metabolomics matches peaks against a library for identification and finds a large number of unknowns since fragments are rarely identified as such. Source variations yield fragmentation
variation; therefore, spectral databases are of little use in the face of instrumentation variation, and data impurities stemming from artifacts, noise and ion-suppression. We have developed a general method for associating all
fragments, adducts, dimers, etc., that also makes the structure elucidation of unknowns easier. The IROA protocol incorporates stable-isotopes into metabolites, creating stable-isotopic Internal Standards (IS) for each and every metabolite measured so that specific alterations can be accurately measured and quantitated. Biochemically-complex IS are generated using growth media wherein all natural abundance 12C compounds (amino acids, sugars etc.) are replaced with randomly and universally 95% 13C-labelled compounds (95% 13C media) so when populations of cells are incubated in such media, all biological components in the cells, including metabolites, carry unique 13C signatures. When an IS, so created, is added to natural abundance samples and analyzed by MS, each metabolite peak carries a ready identifier of its origin, an enhanced M-1, M-2 etc. for the 95%13C IS and natural abundance M+1 for the 12C (Experimental) sample. Because of the presence of the IS all adducts and fragments may now be correctly identified even within areas of high co-elution, by the characteristics of the IROA peaks without the need of a fragmentation library. ClusterFinder software identifies IS parent ions and their collective fragments, adducts, and provides structural confirmation of metabolites since the masses and ratios between the IS and it’s NA analogues will be a unique determinant for each such cluster. Furthermore, for each fragment the number of carbons present and their monoisotopic masses provide accurate formulae which supports structure elucidation of the parent compound in an unknown. 
Chris Beecher1,2 , Tim Garrett2, Rick Yost2, Felice de Jong1,
[1] IROA Technologies, Ann Arbor, MI, [2] University of Florida, Gainesville, FL
All IROA protocols incorporate stable isotopes  into metabolites, creating unique isotopic patterns in all metabolites. The IROA Fluxomic protocol is used to first label every metabolic pool with a 5% 13C isotopic pattern, and then introduces a specific precursor whose flux is to be determined asa 95% to 99% 13C-labeled compound. Since every metabolic pool is labeled with a 5% 13C isotopic pattern the ClusterFinder software can easily identify all metabolic pools, differentiate them from artifacts and noise, and since their exact pattern is known, i.e. it is a 5% 13C isotopic pattern, it can automatically seek any perturbations in the expected 13C isotopic pattern that would indicate flux into that metabolic pool. Unlike other fluxomic approaches this process is completely automated, and examines every metabolic pool without the bias or the need to predetermine it as a target for investigation. We present a series of experiments in which the flux of glucose and glutamine are separately examined as the flux agent, in a HepG2 cell in a time-course experimental system. Our results are compared with similar experiments done using a more traditional approach. In the traditional approach, specific metabolic pools are queried and the label may be targeted to obtain highly specific information often detailing which carbon is transferred (i.e. positionally). In the IROA Fluxomic approach, the total number of carbons derived from the flux agent and transferred into every metabolic pool is easily determined; however positional effects are not available. These two techniques are therefore very complimentary, and increase the ability to understand flux as a function of physiological change. The IROA ClusterFinder software was specifically modified to complete the entireunbiased analysis with no preconceptions. It affords a new and unique fluxomic point of view.


ASMS 2016

Chris Beecher1,3 , Timothy J. Garrett2 , Elizabeth S. Dhummakupt1, Vanessa Y. Rubio1
1 University of Florida, Gainesville, FL; 2 Department of Pathology, Immunology and Laboratory Medicine, University of Florida, Gainesville, FL, 3 IROA Technologies, Ann Arbor, MI
Mixtures are inevitable. In fact, most metabolomic samples are mixtures of one kind
or another, e.g. most tissue is a collection of different cell-types, most biopsies include both “normal” and disease, blood and plasma are mixtures of bodily functions, a cell population may include cells at different growth states, and even a single cell may represent a mixture of physiological states. Thus, the data we generate, rather than representing a single state, almost always represents the sum of data for all of the underlying tissue or cell types. Most of the time we attempt to solve this problem by trying to acquire tissues as pure as possible, and do not deal with the inherent mixed nature, assuming it will “all come out in the wash”. This study attempts to take a direct approach, namely to factorize the experimental dataset in an attempt to determine what the original mixtures were.
Non-negative Matrix Factorization (NMF) is a factorization technique that analyzes the correlation and partial correlation structure of the entire dataset to determine these underlying factors. For NMF to be successful it is necessary to have a number of samples; each one of which represents a slightly different mixture. It does not need to know the percentages of the mixed ingredients, nor does it have to have a pure species, i.e. any samples that are “unmixed”. This technique has received a great deal of use in image analysis, and text analysis, but has been rarely used in biology. In 2006, Golub pioneered the use of a NMF-based approach to create “Connectivity Maps” of gene expression data to great effect, but it has not yet received much use in metabolomics. In an effort to demonstrate the utility of NMF in metabolomics, this study created a number of artificial, and unknown to the researchers, mixtures of two meats, beef and pork. The question was “could NMF sort the resulting dataset and determine what a metabolomics signature for pure pork or pure beef would look like?”. If it can do this, then when presented with a collection of liver biopsies it can just as easily determine what a pure tumor metabolomics profile looks like even though the biopsies all represent mixtures of tumorous and normal cells. IROA[4] is a collection of protocols that dramatically strengthens the identification and quantification of metabolites. The IROA-based workflow (see below) inserts a mixture of 400+ internal standard compounds into a sample in order to measure 400+ compounds more accurately and reproducibly than otherwise possible. The IROA-based workflow is essentially noise-free, and therefore could generate a dataset of sufficient quality to be sorted by NMF. Here we introduce a novel approach based on the use of IROA to generate very clean complex targeted dataset devoid of artifacts and noise, and NMF to deconvolute mixtures of any kind into their pure component mixture. This is a general algorithm that may be successfully applied in most cases of such problems.

Isotopic Ratio Outlier Analysis (IROA) global metabolome interrogation of an actinomycete bacterium following introduction of a novel pathway

Felice de Jong2, Taylor A. Lundy1, Chris Beecher2,3, Amy L. Lane1
1 University of North Florida, Jacksonville, FL, USA, 2 IROA Technologies LLC, Bolton, MA, USA, 3 University of Florida, Gainsville, FL, USA
Microbial genomes harbor biosynthetic pathways for unknown natural products that are
potentially a rich source for medicinally important molecules. The key is to apply genetic and biochemical approaches to best activate and analyze these bioactive pathways and their products. The goal here was to evaluate global metabolic changes for a model actinomycete bacterium in response to the introduction of the ~55 kb cyanosporaside biosynthetic gene cluster (cyn) into the genome of Streptomyces lividans TK24, an actinomycete widely used as a host for producing small molecules. The addition of the cyanosporaside natural product biosynthetic pathway gives us the ability to interrogate metabolic response and track novel compounds introduced by this biosynthetic pathway. Comparative LC-hrMS profiling and IROA were used to quantify small-molecule readouts in response to the addition of the cyanosporaside pathway.

Metabolomics Society 2015

Metabolomic Analysis of Type 1 Diabetic Primary Cells using Isotopic Ratio Outlier Analysis (IROA) by LC-HRMS

Candice Z. Ulmer 1, Christopher Beecher 2, Timothy J. Garrett 3, Jing Chen 3, Clayton Matthews 3, Richard A. Yost 1,3
1Department of Chemistry, University of Florida, Gainesville, FL; 2IROA Technologies, Ann Arbor, MI; 3Department of Pathology, Immunology and Laboratory Medicine, College of Medicine, University of Florida, Gainesville, FL

Type 1 Diabetes (T1D) is an incurable, auto-immune disease that results from the destruction of insulin-producing pancreatic beta cells by pathogenic T lymphocytes. These defective T cells can differentiate into CD4+ T cells that correlate with T1D progression. Of the few experimental designs targeted to identifying the metabolic profile ofsolely T1D, many incorporate animal models that fail to account for pathophysiological differences in humans. There is a need to better understand the metabolic and lipidomic signature of this disease using human samples. This work employs isotopic labeling LC-HRMS methodologies to identify the metabolic and lipidomic trends of immune dysregulation using primary T cells obtained from T1D patients compared to 1st degree relatives and healthy controls.


ASMS 2015

Structural Elucidation of the Metabolome using Isotopic Ratio Outlier Analysis (IROA) in combination with UHPLC-QTOF and Data-Independent Acquisition

Chris Beecher1; Felice de Jong1; Amrita Cheema2; Tyrone Dowdy2; Giuseppe Astarita3
1IROA Technologies LLC, Ann Arbor, MI, 2Georgetown University, Washington, DC, 3Waters Corporation, Milford, MA
Metabolite identification represents the bottleneck of most metabolomics studies. This is aggravated by the presence of noise signals, impurities due to sample collection and extraction procedures and other non-biological relevant information. Isotopic Ratio Outlier Analysis (IROA)1,2 protocol mitigates several of these commonly encountered sources of variance by using specific isotopic signature. Once the biological relevant analytes have been identified, the characterization of their structure often relies only on accurate mass and isotopic pattern. Here, we propose a metabolomics approach using IROA in combination with UHPLC-QTOF in data-independent acquisition (DIA) mode for a rapid screening of the metabolome and the simultaneously collection of both qualitative and quantitative information of known and unknown metabolites.

 Metabolic effect of drought stress during the grain filling growth stage in wheat measured by Isotopic Ratio Outlier Analysis (IROA)

Felice de Jong1, Chris Beecher2, Masum Akond3, John Ericson3, Md Ali Babar3
1IROA Technologies LLC, Bolton, MA, 2University of Florida, Gainsville, FL, Dept of Chemistry, 3University of Florida, Gainsville, FL, Dept of Agronomy   

Metabolomic approaches have been documented to have great value in phenotyping and diagnostic analyses in plants1. The IROA® protocol2,3 was applied to determine the biochemical response of wheat metabolomes to water-stress during the grain filling growth stage. SS8641, a high-yield soft-red winter wheat, was grown under well-watered and drought conditions. In this IROA phenotypic analysis, controlled greenhouse-grown leaves containing carbon at natural abundance werec ompared to Standard wheat leaves that were grown to contain universally-distributed ~97% 13C; namely, a targeted analysis using a biologically-relevant Internal Standard. The IROA patterns allowed the identification of the isotopically labeled peaks and their 12C isotopomers, and the removal of artifacts, noise and extraneous peaks.By pooling experimental and Standard samples, variances introduced during sample-preparation and analysis were controlled.

Untargeted metabolomic analysis of the yeast lipin phophatidate phosphatase deletion using IROA and LC-HRMS

Yu-Hsuan Tsai;1 Timothy J. Garrett;2 Yunping Qiu;3 Robyn Moir;5 Ian Willis;5 Chris Beecher;4 Richard A. Yost;1,2 Irwin Kurland3
1Department of Chemistry, University of Florida, Gainesville, FL; 2Department of Pathology, Immunology, and Laboratory Medicine, University of Florida, Gainesville, FL; 3 Department of Medicine, Albert Einstein College of  Medicine, Bronx, NY; 4 IROA Technologies, Ann Arbor, Michigan; 5 Department of Biochemistry, Albert Einstein College of Medicine, Bronx, NY
Lipins are phosphatidate phosphatases that generate diacylglycerol (DAG) from phosphatidic acid (PA), regulating a pathway key for production of triglycerides (PA->DAG->TAG). Absence of the mammalian lipin results in lipodystrophy, and yeast lipin (Pah1p)controls the formation of cytosolic lipid droplets. Depletion of PAH1 (pah1Δ) has been shown to result in a dramatic decrease in lipid droplet number. Metabolic pathways that might mediate this effect, and possibly have relevance to mammalian lipin-dependent lipodystrophy, were examined using Isotopic Ratio Outlier Analysis (IROA). IROA is a mass spectrometry based metabolomic profiling method using 13C labeling to eliminate sample-to-sample variance, discriminate against noise and artifacts, and improve compound identification. This work utilized IROA with LC-HRMS and investigated the metabolomic profiles from WT yeast vs pah1Δ.


Metabolomics of Hermaphroditic C. elegans via Isotopic Ratio Outlier Analysis using High-Resolution Accurate Mass LC/MS/MS

Caenorhabditis elegans is one of the best-studied animals in science. Despite this, metabolomic studies in C. elegans have only recently become active areas of research. The Isotopic Ratio Outlier Analysis (IROA) protocol uses 13C-isotopic signatures to identify and to quantitate metabolites. It reduces error introduced during sample preparation and analysis, including ionization suppression by the use of IROA standards. The marriage of IROA and high-resolution accurate mass (HRAM) LC/MS/MS with C. elegans metabolomics allows experiments which assess the biological response to stresses or stimuli. These experiments would conventionally be difficult due to interferences by metabolites of unlabeled organisms. With IROA labeling and HRAM detection, metabolites can be distinguished in an untargeted manner, quantitated and unambiguously identified to their chemical formulas.

Isotopic Ratio Outlier Analysis (IROA) of Myxobacteria using ultra high resolution mass spectrometry

Myxobacteria represent an important source of novel natural products exhibiting a wide range of biological activities. Some of these so-called secondary metabolites are investigated as potential leads for novel drugs. Traditional approaches to discovering natural products mainly employ bioassays and activity-guided isolation, but genomics-based strategies and “metabolome-mining” approaches become increasingly successful to reveal additional compounds. These newer methods hold great promise for uncovering novel secondary metabolites from myxobacterial strains, as the number of known compounds identified to date is often significantly lower than expected from genome sequence information. Analytical challenges for comprehensive MS-based profiling of myxobacteria include the need to reliably detect the significant differences between secondary metabolomes, e.g. as a consequence of gene knock-outs or regulatory effects, as well as the robust quantitation of known and unknown target compounds and their identification. The IROA protocol was applied to the analysis of myxobacterial secondary metabolomes.

Differential Metabolomic Profiling of Maize Genotypes under Drought-Stressed Conditions using IROA (Isotopic Ratio Outlier Analysis)

The IROA protocol has been applied in a phenotypic analysis of field grown maize (Zea mays) to understand the biochemical differences across selected genotypes when exposed to drought conditions. In this IROA phenotypic analysis, field-grown leaves containing carbon at natural abundance were compared to a standard maize leaf that was grown to contain universally-distributed ~97% 13C; becoming a targeted analysis using a biologically-relevant internal standard. At 97% 13C the IROA patterns were sufficient to find isotopically labeled peaks, identify their 12C isotopomers, and remove artifacts, noise and extraneous peaks. With accurate mass and IROA, the identification of observed component peaks to chemical formula is unambiguous. The benefit of IROA is it takes into account variances introduced during sample-preparation and analysis, including ion suppression.

Characterization and identification of unknown metabolites using Isotopic Ratio Outlier Analysis (IROA)

The identification of unknown metabolites is one of the biggest bottlenecks of metabolomics. The IROA protocol utilizes isotopically-defined media (in which all nutrients are labeled with either 5%13C, "C12 IROA media" (experimental), or 95%13C, "C13 IROA media" (control), to label all biological compounds with differing masses. Therefore, control and experimental samples can be analyzed as a single sample by LC-MS with all biological peaks uniquely paired. For any compound, the peak from the C12-media is mirrored by a second peak from the C13-media. The distance between these peaks is the number of carbons in the compound. The formula of the compound can be readily determined if the high-resolution mass and number of carbons is known.

Differential Metabolomic Profiling of Wheat Cultivars by IROA (Isotopic Ratio Outlier Analysis)

The interest in metabolomics to understand fundamental biology and applied biotechnology, especially in the field of plant science, has driven technology development. This study describes the use of a combined analytical and bioinformatic metabolomics technology applied to the understanding of plant metabolism. The diurnal metabolome changes exhibited in a cultivar of wheat, TX8544, were determined using the IROA protocol. Metabolomics plays an important role in how an organism adapts to change, in this case the diurnal pattern of heat and light. Here an isotopically-defined standard wheat sample is added to the experimental sample and is analyzed as a single sample, reducing suppression, and sample-to-sample variance, including variance introduced during preparation and analysis.