Gas Chromatography-Quadrupole Time-of-Flight Mass Spectrometry

Subscriber access provided by AUSTRALIAN NATIONAL UNIV

Article

GC-QTOFMS based determination of isotopologue and tandem mass isotopomer fractions of primary metabolites for 13C-metabolic flux analysis Teresa Mairinger, Matthias Steiger, Justyna Nocon, Diethard Mattanovich, Gunda Koellensperger, and Stephan Hann Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b03173 • Publication Date (Web): 29 Oct 2015 Downloaded from http://pubs.acs.org on October 31, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

GC-QTOFMS based determination of isotopologue and tandem mass isotopomer fractions of primary metabolites for 13C-metabolic flux analysis 1,2

2,3

3

2,3

4

Teresa Mairinger , Matthias Steiger ,Justyna Nocon , Diethard Mattanovich , Gunda Koellensperger , 1,2 Stephan Hann 1

Department of Chemistry, University of Natural Resources and Life Sciences - BOKU Vienna, Muthgasse 18, 1190 Vienna, Austria 2 Austrian Centre of Industrial Biotechnology (acib), Muthgasse 11, 1190 Vienna, Austria 3 Department of Biotechnology, University of Natural Resources and Life Sciences - BOKU Vienna, Muthgasse 18, 1190 Vienna, Austria 4 Institute of Analytical Chemistry, Faculty of Chemistry, University of Vienna, Waehringerstrasse 38, 1090 Vienna, Austria * corresponding author: E-mail: [email protected]

Abstract For the first time an analytical work flow based on accurate mass GC‑QTOFMS with chemical ionization for analysis 13 providing a comprehensive picture of C distribution along the primary metabolism is elaborated. The method 13 provides a powerful new toolbox for C-based metabolic flux analysis which is an emerging strategy in metabolic 13 engineering. In this field, stable isotope tracer experiments based on, for example, C are central for providing characteristic patterns of labeled metabolites, which in turn give insights into the regulation of metabolic pathway kinetics. The new method enables the analysis of isotopologue fractions of 42 free intracellular metabolites within biotechnological samples, while tandem mass isotopomer information is also accessible for a large number of analytes. Hence, the method outperforms previous approaches in terms of metabolite coverage, while also providing rich isotopomer information for a significant number of key metabolites. Moreover, the established work flow includes novel evaluation routines correcting for isotope interference of naturally distributed elements, which is crucial following derivatization of metabolites. Method validation in terms of trueness, precision and limits of detection was performed, showing excellent analytical figures of merit with an overall maximum bias of 5.8%, very high precision for isotopologue and tandem mass isotopomer fractions representing >10% of total abundance, and absolute limits of detection in the fmol range. The suitability of the developed method is demonstrated on a flux 13 13 experiment of Pichia pastoris employing two different tracers, i.e. 1,6 C2- glucose and uniformly-labeled Cglucose.

Introduction Over the past quarter-century metabolic flux analysis (MFA) became a highly valuable tool for describing and 1 characterizing alteration within metabolic networks, especially in support of metabolic engineering . The assessment of metabolic fluxes allows the quantification of in vivo intracellular metabolic rates, typically expressed 2 as reaction rate on a per-unit cell volume or per-unit cell mass basis , and gives therefore an insight into metabolic pathway kinetics. As a matter of fact, determining fluxes allows inferring the extent of usage of certain metabolic 3 pathways in the overall metabolic network . In the beginning of metabolic flux analysis, stoichiometric metabolic flux analysis relied on directly measurable 4 extracellular rates for the calculation of intracellular fluxes . Assuming that the metabolism of interest is in a

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 19

quasi-stationary state, the incoming fluxes of each intracellular metabolite pool will balance the outgoing fluxes. However, stoichiometric metabolic flux analysis shows some shortcomings in resolving comprehensively metabolic 4 pathways, such as parallel, cyclic or bidirectional steps . Due to these limitations, stable isotope tracer 13 experiments, mostly using C as tracer, were implemented in metabolic flux analysis with the purpose of indirectly 4 deducing the intracellular fluxes . As a result of introducing a stable isotope tracer into an organism, the labeled atoms are incorporated into downstream metabolites. This isotopic enrichment within the intracellular metabolite 5 pool can then be measured by nuclear magnetic resonance (NMR) or molecular mass spectrometry (MS) . By 4 combining information on labeling data and extracellular fluxes, the intracellular fluxes can be estimated . The 13 application of C as tracer isotope allows resolving alternative pathways, as differences in labeling patterns due to the participation in diverse reactions of the tracer’s carbon backbone can be observed. Apparently, the isotopic 13 tracer should be selected carefully regarding its C-labeling position(s), as this determines the formation of certain labeling patterns in a metabolic network and therefore also the flux resolution of certain pathways, as described 6,7,8 by Antoniewicz and coworkers . However, it has to be emphasized that when employing different isotopic tracers the flux solution stays the same, though the flux resolution varies. 13

As a matter of fact, C-MFA does not comprise the metabolites of the entire metabolic network, but rather 1 focuses on reactions of the central carbon metabolism which involves the major anabolic and catabolic fluxes . In general two approaches for estimation of the intracellular fluxes can be distinguished, either labeling data is 5 acquired from amino acids after protein hydrolysis or from labeling patterns of free intracellular metabolites, e.g. 9,10,11,12,13,14,15 . Even though challenging in terms of high turnover rate and low concentration, clearly labeling data 5 of free intracellular metabolites provides the richest source of information . For the analysis of amino acids after protein hydrolysis MS and NMR are suited, whereas when focusing on free intracellular metabolites MS based techniques are the method of choice due to sensitivity, throughput and the number of amenable analytes. Prior to detection of labeling patterns applying mass spectrometry, separation of the analytes of interest is necessary and is mostly performed either via gas (GC) or liquid chromatography (LC). Depending on the analytical approach (ionization mode and fragmentation strategy), LC-MS and GC-MS based methods deliver information on isotopologues, i.e. molecular entities that differ in their isotopic composition, or isotopomers, i.e. moieties having 16 the same number of each isotope but differ in their position . For a metabolite with n carbon atoms, n+1 n isotopologues and 2 isotopomers exist. In order to obtain partial information on the position of the tracer isotope, 1 fragmentation of the molecule of interest is necessary . This can be achieved employing either selective 15,17,18 fragmentation mechanisms like collision induced dissociation in tandem MS experiments or via harsh ionization conditions, like electron ionization (EI). In case of tandem MS experiments the term tandem mass isotopomer is introduced. This term refers to the carbon backbone fragment obtained via fragmentation and from 13 which conclusions on the position of the heavy isotopes, i.e. C, can be drawn. With regard to mass spectrometry, different set-ups and combinations can be applied. In hyphenation with LC, single stage instruments i.e. single quadrupoles or TOFs are useful for determination of isotopologue ratios. If tandem mass isotopomer distributions are desired, quadrupole tandem MS systems, QTOF systems or fragmenting ion traps (orbitraps) are needed. In case of GC, often single stage instruments are used with the possibility of generating in-source fragments via electron ionization. If EI is applied, comprehensive information is derived by merging information of various 5,19 fragments containing different parts of the carbon backbone . Due to selective breaks of the carbon backbone further positional information can be derived and generated tandem mass isotopomer distributions of certain 15 metabolites can be included into the model, increasing the precision on flux estimation calculation . The additional information on the position of the heavy isotopes is crucial for resolving certain metabolic branching points, e.g. for sample set of Bacillus subtilis it was shown that the reactions of upper glycolysis and pentose phosphate pathway and between lower glycolysis and the C4 intermediates of the TCA cycle were improved using 20 fragment data .


Page 3 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Reflecting the number of published studies on this topic, especially the introduction of tandem MS experiments for deriving positional information on free intracellular metabolites is highly challenging. Moreover, exact information on the dissociation/ fragmentation of an analyte’s carbon backbone is indispensable. Additionally, dwell time/ acquisition time has to be considered, especially when metabolites with comparably high number of carbons in the backbone and consecutive fragmentation are analyzed. In this context, with a time-of-flight mass spectrometer, the information gain due to high mass resolution eases the unambiguous identification of non-selective fragments or tandem mass isotopomers. Besides, the possibility of recording all fragments for a given labeling state without restriction of time proved to be advantageous and can be used to its full capacity. The application of gas chromatography based techniques coupled to mass spectrometry has the advantage of broad metabolite coverage. Though this approach has the drawback of a necessary time-consuming derivatization step, the benefit of analyzing important compounds of the primary carbon metabolism, like organic acids, amino acids as well as sugars and sugar phosphates, within one chromatographic run, is convincing. Typically employed ionization techniques in GC-MS are either electron ionization or chemical ionization (CI). Electron ionization leads to a mass spectrum rich in fragments. Especially for analysis of the biologically highly relevant sugars and sugar phosphates, the highly abundant fragment ions present in the respective mass spectrum contain hardly carbon backbone information. In contrast to that, chemical ionization, as soft ionization technique, leads to a highly abundant protonated molecule. The benefit of CI has been recently demonstrated for analysis of sugar phosphates 21 with GC-CI-TOFMS yielding excellent chromatographic resolution and negligible in source fragmentation . In the present work, we established a comprehensive method for isotopologue and tandem mass isotopomer analysis of important compounds of the primary carbon metabolism, with the potential to discriminate alternative pathways due to available positional information. In general it has to be mentioned that a workflow for proteinogenic amino 5 acids is already well established . However, methods for metabolites of glycolysis, pentose phosphate pathway and tricarboxylic acid cycle are often missing, though these metabolites such as sugars and sugar phosphates, organic acids and free amino acids are of special interest from a modelling point of view. This represents also the scope of analysis of our developed method. In addition to isotopologue information of 42 analytes, tandem mass isotopomer information is also accessible for a large number of analytes using QTOFMS with isotopologue selective fragmentation via collision induced dissociation. Particularly due to derivatization, isotope interference 22 correction is indispensable and was performed with a novel object-oriented software tool , enabling isotopologue as well as tandem mass isotopomer correction. In a proof-of-concept study, our novel analytical approach was 13 evaluated within a C based MFA experiment aiming at the optimization of a selected strain of Pichia pastoris in the context of industrial biotechnology.

Materials and Methods Chemicals 6-Phosphogluconic acid barium salt (6PGA), 2-phosphoglyceric acid disodium salt hydrate (2PG), 3-phosphoglyceric acid disodium salt (3PG), DL-2-aminoadipic acid (AAA), cis-aconitic acid (Aco), α-ketoglutaric acid sodium salt (AKG), L-alanine (Ala), L-asparagine (Asn), dihydroxyacetone phosphate lithium salt (DHAP), erythrose-4-phosphate sodium salt (E4P), fructose-6-phosphate disodium salt hydrate (F6P), D-(-)-fructose (Fru), fumaric acid (Fum), D-(+)-glucose (Glc), glucose-6-phosphate dilithium salt (G6P), potassium D-gluconate (Glc-ON), glyceraldehyde-3phosphate solution (GAP), L-homoserine (H-Ser), DL-isocitric acid trisodium salt hydrate (I-Cit), L-isoleucine (Ile), L-cystathionine (LCT), L-lysine (Lys), D-mannitol (Man-OL), mannitol-1-phosphate lithium salt (Mt1P), manose-6-phosphate disodium salt hydrate (M6P), L-phenylalanine (Phe), L-proline (Pro), ribulose-5-phosphate



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 19

sodium salt (Rul5P), ribose-5-phosphate disodium salt hydrate (Ri5P), DL-serine (Ser), succinic acid (Suc), L-threonine (Thr), L-tyrosine( Tyr), were purchased at Sigma Aldrich (Sigma Aldrich, St.Louis, MO). L-aspartic acid (Asp), citric acid monohydrate (Cit), L-glutamine (Gln), L-glutamic acid (Glu), glycine (Gly), L-leucine (Leu), DL-malic acid (Mal), L-valine (Val) were purchased from Merck (Merck Millipore, Darmstadt, Germany) and sedoheptulose7-phosphate lithium salt (S7P), was purchased from Carbosynth (Carbosynth limited, UK). Stock standard solutions of all analytes were prepared by dissolving an appropriate amount of solid standard in LCMS-grade water or 0.1 M HCl, respectively. Analyte mixture solutions of all single standards with concentrations of 5, 25, 50 and 100 µM were prepared in LC-MS-grade water by appropriate dilution. All single stock standards and analyte mixture solutions were kept at -80 °C and were stable for a minimum of 4 weeks. Ethoxyamine hydrochloride was purchased from Sigma Aldrich, N-methyl-N(trimethylsilyl)trifluoroacetamide (MSTFA) in with 1% trimethylchlorosilane (TMCS) from Thermo Scientific (Waltham,MA, USA). The ethoxyamine solution was prepared daily by dissolving an appropriate amount of solid substances in water-free pyridine (Sigma Aldrich, St.Louis, MO, USA). After addition of an appropriate amount of ethoxylamine hydrochloride, the working standards were evaporated to complete dryness in a vacuum centrifuge operating at low pressure (below 1 mbar), crimped and stored at -80 °C. Dried working standards were stable for a minimum of 4 weeks. Prior to analysis, the prepared analyte mixtures were dried again for 0.5 h.

Cultivation of P. pastoris 23

A P. pastoris strains producing human superoxide dismutase, designated as hSOD was used for this study . For the 13 -1 metabolic flux experiment employing 1,6 C2 glucose as tracer, cells were grown in 50 mL YNB medium (3.4 g L -1 -1 -1 13 YNB w/o amino acids and ammonium sulfate, 10 g L (NH4)2SO4, 400 mg L biotin,) with 2.5 g L 1,6- C-labelled 13 glucose (Cortecnet, Voisins-le-Bretonneux, France) as a single C-source. For the flux experiment using U C glucose 13 as substrate, a mixture of 20% uniformly labeled C glucose and 80% naturally labeled glucose was employed. The o cultures were grown in wide neck non-baffeled shake flasks at 25 C. The cells were inoculated at OD600 = 0.03 and grown for 20 h until exponential phase (around OD600 = 1). Glucose uptake and extracellular metabolites were determined in cultures grown on YNB medium supplemented with unlabeled glucose. Glucose, ethanol, acetate 24 and arabitol were quantified from supernatants by HPLC as described in . Optical density was measured at 600 nm wavelength in 1 mL of culture broth using a WPACO 8000 Cell Density Meter. For determination of yeast dry mass (YDM) 9 mL of culture broth was centrifuged, washed with 10mL of ddH2O and dried for 24 h at 105 °C in preweighted tubes. Unless described differently all cultures were repeated in triplicates. o For the analysis of intracellular metabolites, the cells were rapidly sampled into 60% methanol at -30 C and 25 filtered through cellulose acetate filters using a vacuum pump . The cell pellet on the filter was transferred to o precooled tubes and stored at -80 C until extraction. The metabolites were extracted by adding 4 mL of 75% o o ethanol at 85 C to the frozen cell pellets and incubating it for 3 min at 85 C. The samples were rapidly cooled and the extracts separated by centrifugation.

Automated just-in-time online derivatization For automated just-in-time online derivatization, a two-step reaction was performed. In the first step ethoxymation was employed in order to protect the carbonyl group. The second step comprised a trimethylsilylation using MSTFA. As a prerequisite of this derivatization procedure the samples have to be completely dry. In order to protect the carbonyl group of the metabolites already during the evaporation step, ethoxyamine hydrochloride was added and subsequently the mixture was evaporated to complete dryness in a vacuum centrifuge operating at low pressure (10% an average RSD of about 3.5% for IF, 9% in case of TMIF was assessed for technical replicates. Regarding biological repeatability of the tracer experiments, 8.5% in case of IF and 9% in case of TMIF was obtained. In contrast, lower fractions were affected by higher uncertainty of measurement. As a result, the authors consider that the application of a certain fraction threshold depending on signal intensity is recommended. Isotopologue analysis of free intracellular metabolites employing GC-CI-TOFMS For the hSOD strain of P. pastoris, isotopic interference corrected isotopologue distributions of 42 analytes are depicted as heat maps in Figure 3. As a matter of fact labeling patterns differ significantly between the different isotope tracer experiments, enabling the deduction of different pathway information, as described in detail in 6 37 Crown et al. and Metallo et al. . In terms of precision the two flux experiments as well as the pre- experiment do 13 13 not vary significantly. The average RSD for fractions > 10% were about 3.8%, 3.5% and 2.3% for the U C, 1,6 C2 glucose tracer experiments and the pre-experiment, respectively. The higher precision of the data obtained from the pre-experiment can be explained by the fact, that here higher fractions were obtained due to the natural isotope distribution of carbon.

Figure 3: Heat maps of isotopologue fractions of 42 metabolites analyzed in P. pastoris, strain hSOD. a) shows isotopologue 13 13 fractions of P. pastoris fed with natural glucose. b) 1,6 C2 glucose was employed as stable isotope tracer c) U C glucose mixed with natural glucose (20:80) was applied as substrate


Page 13 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Tandem mass isotopomer analysis of free intracellular metabolites employing GC-CI-QTOFMS Beside of resolving certain nodes within the metabolic network, the information on the position of the stable 15 heavy isotopes may also improve flux estimate precision, as discussed by Rühl et al. as well as by Antoniewicz in 38 . As a proof of principle in Figure 4 we present here exemplarily the TMIDs of asparate, L-cystathionine and α-ketogluterate of 2 different labeling experiments. With regard to precision of the depicted tandem mass isotopomer fractions obtained from the two flux experiments, no significant difference of can be observed. However, a correlation between fraction and precision is clearly observable. Full information on all detectable tandem mass isotopomers of the two flux experiments is depicted in Table S-5d and S-5f.

Figure 4: Tandem mass isotopomer distributions of asparate (Asp), α-ketogluterate(AKG) and L-cystathionine (LCT) analyzed in P. pastoris strain hSOD employing GC-CI-QTOFMS. The TMIDs depicted in blue stem from a labeling experiment employing 13 13 1,6 C2 glucose as tracer, whereas TMIDs of the flux experiment using a mix of U C glucose and natural glucose (20:80) as substrate are depicted in red

Calculation of flux distribution In the presented case study the split ratio between glycolysis and pentose phosphate pathway (PPP) was of high 23 interest . Hence, labeling information on sugar phosphates was employed for calculation of flux distribution. Due to low concentration no isotopomer information was available for those metabolites. In Figure 5 the flux 13 distribution of the glycolysis and pentose pathways of P. pastoris hSOD strain, employing 1,6 C2 glucose as 13 substrate, is exemplarily shown. In the supplementary data, in Table S-6 the stoichiometric model for fitting the C flux measurements in OpenFLUX as well as the measured and simulated mass distribution values are depicted. The high data quality of extracellular rates allowed to constrain the model in terms of these rates significantly reducing degrees of freedom in the fitting process.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 19

Figure 5: Flux distribution of the glycolysis and pentose phosphate pathway of P. pastoris strain hSOD grown on glucose 13 -1 -1 using labelling information of 1,6- C2 glucose. The flux values are normalized to glucose uptake [Cmmol gCDW h ] and presented in [% C mol]. The upper value in the rectangular boxes represents the net flux of the forward reaction. The reaction reversibility is given with the lower value (f forward reaction, r reverse reaction). In case of a reversible reaction (r) the exchange flux is shown as ratio range between forward and net reaction. (The ratio range of the exchange flux is calculated by dividing the upper and lower flux boundaries of the forward reaction by the net flux of the reaction: 1 x –no flow through reverse reaction, 2 x forward reaction flux is increase by 2 fold (x)). 13

The use of 1,6 C2- glucose in combination with measuring free intracellular metabolites allows to determine accurately the flux through the PPP. Thus, the split ratio between glycolysis and pentose phosphate pathway can be calculated: 48.9 ± 3.0 % of the carbon is directly shuttled through glycolysis, whereas 51.1 ± 3.0 % is processed via the pentose phosphate pathway. The uncertainty of flux distributions was calculated using the gradient based 31 search algorithm for sensitivity analysis implemented in OpenFLUX . Especially the label information contained in 13 F6P together with the substrate 1, 6 C2 glucose allows accurate determination of this flux ratio. The exchange flux 7 is often difficult to predict for reversible reactions , which can be seen in the high flux range for the ratio between forward flux and net flux of such reactions, e.g. in the reaction from DHAP to G3P (Figure 5). For this reaction the -1 -1 forward flux can have absolute values ranging from 2.8 up to 15.0 [mmol gCDW h ] at a net reaction of 1.4 [mmol -1 -1 gCDW h ]. However, two reversible reactions in the PPP can be resolved using the labeling information of 11 metabolites (S7P + G3P -> E4P + F6P and XYL5P + E4P = F6P + G3P). It is noteworthy, that previous results obtained 13 on U C6- glucose via the analysis of proteinogenic amino acids had a significantly higher error rate regarding the 23 split ratio between glycolysis and PPP. Accordingly, the flux through glycolysis (78%) was overestimated , due to 13 missing labelling information. Summarizing, these results clearly demonstrate that the introduction of 1,6 C2glucose as tracer in combination with the analysis of free intracellular metabolites by GC-CI-TOFMS allows the accurate determination of the intracellular flux through the pentose phosphate pathway.

Conclusion A novel analytical approach based on accurate mass GC-CI-QTOFMS for analysis of isotopologue and tandem mass isotopomer fractions of free intracellular metabolites from central carbon metabolism, namely amino acids, organic acids, sugars and sugar phosphates, in biotechnological samples is presented.


Page 15 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


39,40

As a soft ionization technique, CI is mainly employed for identity confirmation and only in rare cases for quantitative or semi-quantitative approaches. However, for the presented approach of isotopologue-selective fragmentation for GC-QTOFMS, this ionization technique showed to be highly advantageous. Chemical ionization provides the protonated molecule as base peak in the mass spectrum and enables the selective fragmentation of isotopologues as well as the unambiguous assignment of isotopomers to the respective isotopologue. Due to the information gain when employing chemical ionization with consecutive collision induced dissociation, GC-CIQTOFMS proved to be most valuable for the determination of tandem mass isotopomer distributions of free intracellular primary metabolites. However, it has to be mentioned that this holds especially true for higher abundant metabolites, as this approach suffers from loss of sensitivity, compared to the TOFMS approach. One of the advantages of GC-CI-QTOFMS is that new compounds can be easily implemented in existing methods, as here the isotopologue represents the base peak and isotopologue selective fragmentation can be performed. Consecutively, all different labeling states and fragments are recorded, without the need of sophisticated input of bioinformatics. Besides, QTOFMS instrumentation allows the evaluation of various isotopomers without increasing the number of transition within one chromatographic peak, which is consequently beneficial in terms of dwell time/ acquisition time for each isotopologue as well as points per chromatographic peak. In terms of method development, the application of high-resolution mass spectrometry proved to be advantageous regarding selectivity as well as facilitating identity confirmation of fragments. Additionally, the possibility of retrospective data processing is highly valuable. The instrumental limitation of TOFMS regarding linear dynamic range is not of concern here, as 2- 3 orders of magnitude suffice for isotopologue analysis. The suitability of both, the developed analytical method and the isotope interference correction via ICT was demonstrated in a flux experiment comprising different isotopic tracers. Current analytical developments concern the improvement of sensitivity in QTOF-mode with collision induced dissociation, focusing on CID parameters and optimization of potential gradient between the collision cell and the TOF analyzer. In order to obtain also positional information on those analytes that are restricted by high limits of detection, future developments aim at selective enrichment strategies, especially of sugar phosphates.

Acknowledgment This work has been supported by the Federal Ministry of Science, Research and Economy (BMWFW), the Federal Ministry of Traffic, Innovation and Technology (bmvit), the Styrian Business Promotion Agency SFG, the Standortagentur Tirol, the Government of Lower Austria and ZIT - Technology Agency of the City of Vienna through the COMET-Funding Program managed by the Austrian Research Promotion Agency FFG. EQ VIBT GmbH is acknowledged for providing mass spectrometry instrumentation. The authors thank Christian Jungreuthmayer, Stefan Neubauer and Juergen Zanghellini for their contribution to the isotope interference correction.

Supporting Information Additional information as noted in text. References (1) (2) (3)

Sauer, U. Mol. Syst. Biol. 2006, 2. Sauer, U.; Lasko, D. R.; Fiaux, J.; Hochuli, M.; Glaser, R.; Szyperski, T.; Wuthrich, K.; Bailey, J. E. J. Bacteriol. 1999, 181, 6679–6688. Stephanopoulos, G. Metab. Eng. 1999, 1, 1–11.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23)

(24) (25) (26) (27) (28) (29) (30) (31) (32) (33) (34) (35) (36) (37) (38) (39) (40)

Page 16 of 19

Wiechert, W. Metab. Eng. 2001, 3, 195–206. Zamboni, N.; Fendt, S.-M.; Rühl, M.; Sauer, U. Nat. Protoc. 2009, 4, 878–892. Crown, S. B.; Antoniewicz, M. R. Metab. Eng. 2012, 14, 150–161. Crown, S. B.; Long, C. P.; Antoniewicz, M. R. Metab. Eng. 2015, 28, 151–158. Antoniewicz, M. R. Curr. Opin. Biotechnol. 2013, 24 , 973–978. Krömer, J. O.; Sorgenfrei, O.; Klopprogge, K.; Heinzle, E.; Wittmann, C. J. Bacteriol. 2004, 186, 1769–1784. Wahl, S. A.; Dauner, M.; Wiechert, W. Biotechnol. Bioeng. 2004, 85, 259–268. Iwatani, S.; Van Dien, S.; Shimbo, K.; Kubota, K.; Kageyama, N.; Iwahata, D.; Miyano, H.; Hirayama, K.; Usuda, Y.; Shimizu, K.; Matsui, K. J. Biotechnol. 2007, 128, 93–111. Nöh, K.; Grönke, K.; Luo, B.; Takors, R.; Oldiges, M.; Wiechert, W. J. Biotechnol. 2007, 129, 249–267. Winden, W. A. van; Dam, J. C. van; Ras, C.; Kleijn, R. J.; Vinke, J. L.; Gulik, W. M. van; Heijnen, J. J. FEMS Yeast Res. 2005, 5, 559–568. Yuan, J.; Fowler, W. U.; Kimball, E.; Lu, W.; Rabinowitz, J. D. Nat. Chem. Biol. 2006, 2, 529–530. Rühl, M.; Rupp, B.; Nöh, K.; Wiechert, W.; Sauer, U.; Zamboni, N. Biotechnol. Bioeng. 2012, 109, 763–771. McNaught, A. D.; Wilkinson, A. IUPAC. Compendium of Chemical Terminology, 2nd edition.; Blackwell Scientific Publications: Oxford, 1997. Jeffrey, F. M. H.; Roach, J. S.; Storey, C. J.; Sherry, A. D.; Malloy, C. R. Anal. Biochem. 2002, 300, 192–205. Choi, J.; Grossbach, M. T.; Antoniewicz, M. R. Anal. Chem. 2012, 84 , 4628–4632. Antoniewicz, M. R.; Kelleher, J. K.; Stephanopoulos, G. Anal. Chem. 2007, 79, 7554–7559. Rühl, M.; Coq, D. L.; Aymerich, S.; Sauer, U. J. Biol. Chem. 2012, 287, 27959–27970. Chu, D. B.; Troyer, C.; Mairinger, T.; Ortmayr, K.; Neubauer, S.; Koellensperger, G.; Hann, S. Anal. Bioanal. Chem. 2015, 1–11. Jungreuthmayer, C.; Neubauer, S.; Mairinger, T.; Zanghellini, J.; Hann, S. Bioinformatics 2015, btv514. Nocon, J.; Steiger, M. G.; Pfeffer, M.; Sohn, S. B.; Kim, T. Y.; Maurer, M.; Rußmayer, H.; Pflügl, S.; Ask, M.; Haberhauer-Troyer, C.; Ortmayr, K.; Hann, S.; Koellensperger, G.; Gasser, B.; Lee, S. Y.; Mattanovich, D. Metab. Eng. 2014, 24, 129–138. Pflügl, S.; Marx, H.; Mattanovich, D.; Sauer, M. Bioresour. Technol. 2012, 119, 133–140. Russmayer, H.; Troyer, C.; Neubauer, S.; Steiger, M. G.; Gasser, B.; Hann, S.; Koellensperger, G.; Sauer, M.; Mattanovich, D. FEMS Yeast Res. 2015, 15, fov049. Mistrik, R.; Lutisan, J.; Suchy, M. HighChem MassFrontier; Thermo Fisher Scientific. Loos, M.; Gerber, C.; Corona, F.; Hollender, J.; Singer, H. Anal. Chem. 2015, 87, 5738–5744. Moseley, H. N. BMC Bioinformatics 2010, 11, 139. Millard, P.; Letisse, F.; Sokol, S.; Portais, J.-C. Bioinformatics 2012, 28, 1294–1296. Carnicer, M.; Baumann, K.; Töplitz, I.; Sánchez-Ferrando, F.; Mattanovich, D.; Ferrer, P.; Albiol, J. Microb. Cell Factories 2009, 8, 65. Quek, L.-E.; Wittmann, C.; Nielsen, L. K.; Krömer, J. O. Microb. Cell Factories 2009, 8, 25. Baumann, K.; Carnicer, M.; Dragosits, M.; Graf, A. B.; Stadlmann, J.; Jouhten, P.; Maaheimo, H.; Gasser, B.; Albiol, J.; Mattanovich, D.; Ferrer, P. BMC Syst. Biol. 2010, 4, 141. Neubauer, S.; Haberhauer-Troyer, C.; Klavins, K.; Russmayer, H.; Steiger, M. G.; Gasser, B.; Sauer, M.; Mattanovich, D.; Hann, S.; Koellensperger, G. J. Sep. Sci. 2012, 35, 3091–3105. Millard, P.; Massou, S.; Portais, J.-C.; Létisse, F. Anal. Chem. 2014, 86, 10288–10295. Guerrasio, R.; Haberhauer-Troyer, C.; Steiger, M.; Sauer, M.; Mattanovich, D.; Koellensperger, G.; Hann, S. Anal. Bioanal. Chem. 2013, 405, 5133–5146. Magnusson, B.; Örnemark, U. Eurachem Guide: The Fitness for Purpose of Analytical Methods – A Laboratory Guide to Method Validation and Related Topics, 2nd ed.; 2014. Metallo, C. M.; Walther, J. L.; Stephanopoulos, G. J. Biotechnol. 2009, 144, 167–174. Antoniewicz, M. R. Curr. Opin. Biotechnol. 2013, 24, 48–53. Abate, S.; Ahn, Y. G.; Kind, T.; Cataldi, T. R. I.; Fiehn, O. Rapid Commun. Mass Spectrom. 2010, 24, 1172– 1180. Portolés, T.; Pitarch, E.; López, F. J.; Hernández, F.; Niessen, W. M. A. Rapid Commun. Mass Spectrom. 2011, 25, 1589–1599.


Page 17 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table of Contents artwork



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1: Assessment of trueness: a) distribution of isotopologue fractions versus bias employing GC-CITOFMS and GC-EI-TOFMS, respectively. The data obtained by GC-CI-TOFMS is depicted in red, where dark red corresponds to isotopologue fractions of the derivatives versus the respective bias and light red represents the isotope interference corrected data. The data obtained by GC-EI-TOFMS is depicted in blue, where dark blue corresponds to isotopologue fractions of the derivatives versus the respective bias and light blue represents the isotope interference corrected data. b) distribution of tandem mass isotopomer fractions versus bias employing GC-CI-QTOFMS and GC-EI-QTOFMS, respectively. The data obtained by GC-CIQTOFMS is depicted in orange, where dark orange corresponds to tandem mass isotopomer fractions of the derivatives versus the respective bias and light orange represents the isotope interference corrected data. The data obtained by GC-EI-QTOFMS is depicted in green, where dark green corresponds to tandem mass isotopomer fractions of the derivatives versus the respective bias and light green represents the isotope interference corrected data. 69x57mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 18 of 19

Page 19 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2: Assessment of precision: a) distribution of isotopologue fractions versus relative standard deviation (RSD), employing GC CI-TOFMS and GC-EI-TOFMS, respectively. The data obtained by GC-CITOFMS is depicted in red: dark red corresponds to isotopologue fractions of the derivatives versus the respective RSD; whereas light red represents the isotope interference corrected data. The data obtained by GC-EI-TOFMS is depicted in blue, where dark blue refers to isotopologue fractions of the derivatives versus the respective RSD and light blue shows the isotope interference corrected data. b) distribution of tandem mass isotopomer fractions versus precision employing GC-CI-QTOFMS and GC-EI-QTOFMS, respectively. The data obtained by GC-CI-QTOFMS is depicted in orange: dark orange corresponds to tandem mass isotopomer fractions of the derivatives versus the respective RSD, whereas light orange represents the isotope interference corrected data. The data obtained by GC-EI-QTOFMS is depicted in green, where dark green refers to tandem mass isotopomer fractions of the derivatives versus the respective RSD and light green shows the isotope interference corrected data. 69x57mm (300 x 300 DPI)

ACS Paragon Plus Environment

Gas Chromatography-Quadrupole Time-of-Flight Mass Spectrometry

Recommend Documents