Anal. Chem. 2008, 80, 9417–9425
High-Resolution Direct Infusion-Based Mass Spectrometry in Combination with Whole 13C Metabolome Isotope Labeling Allows Unambiguous Assignment of Chemical Sum Formulas Patrick Giavalisco,* Jan Hummel, Jan Lisec, Alvaro Cuadros Inostroza, Gareth Catchpole, and Lothar Willmitzer Max-Planck-Institut fu¨r Molekulare Pflanzenphysiologie, Am Mu¨hlenberg 1, 14476 Potsdam-Golm, Germany A new strategy for direct infusion-based metabolite analysis employing a combination of high-resolution mass spectrometry and 13C-isotope labeling of entire metabolomes is described. Differentially isotope labeled metabolite extracts from otherwise identically grown reference plants were prepared and infused into a Fourier transform ion cyclotron resonance mass spectrometer. The derived accurate mass lists from each extract were searched, using an in-house-developed database search tool, against a number of comprehensive metabolite databases. Comparison of the retrieved chemical formulas from both, the 12 C and 13C samples, leads to two major advantages compared to nonisotope-based metabolite fingerprinting: first, removal of background contaminations from the result list, due to the 12C/13C peak pairing principle and therefore positive identification of compounds of true biological origin; second, elimination of ambiguity in chemical formula assignment due to the same principle, leading to the clear association of one measured mass to only one chemical formula. Applying this combination of strategies to metabolite extracts of the model plant Arabidopsis thaliana therefore resulted in the reproducible identification of more than 1000 unambiguous chemical sum formulas of biological origin of which more than 80% have not been associated to Arabidopsis before. The central importance of the metabolome1 and metabolomics2 as an information source and readout for functional genomics approaches,3-7 for the classification of different cellular states,8-10 * To whom correspondence should be addressed. Telephone: 0049-3315678623. Fax: 0049-331-5678236. E-mail:
[email protected]. (1) Oliver, S. G.; Winson, M. K.; Kell, D. B.; Baganz, F. Trends Biotechnol. 1998, 16, 373–378. (2) Fiehn, O. Plant Mol. Biol. 2002, 48, 155–171. (3) Bino, R. J.; Hall, R. D.; Fiehn, O.; Kopka, J.; Saito, K.; Draper, J.; Nikolau, B. J.; Mendes, P.; Roessner-Tunali, U.; Beale, M. H.; Trethewey, R. N.; Lange, B. M.; Wurtele, E. S.; Sumner, L. W. Trends Plant Sci. 2004, 9, 418–425. (4) Fiehn, O.; Kopka, J.; Dormann, P.; Altmann, T.; Trethewey, R. N.; Willmitzer, L. Nat. Biotechnol. 2000, 18, 1157–1161. (5) Griffin, J. L.; Nicholls, A. W. Pharmacogenomics 2006, 7, 1095–1107. (6) Hall, R.; Beale, M.; Fiehn, O.; Hardy, N.; Sumner, L.; Bino, R. Plant Cell 2002, 14, 1437–1440. (7) Sumner, L. W.; Mendes, P.; Dixon, R. A. Phytochemistry 2003, 62, 817– 836. 10.1021/ac8014627 CCC: $40.75 2008 American Chemical Society Published on Web 11/17/2008
in crop trait development,11 in the mapping of metabolic quantitative trait loci,12,13 and as one of the four “omics” organizational strata (genomics, transcriptomics, proteomics, and metabolomics), enabling systems biology,14-18 has been widely demonstrated. In comparison to other “omics” technologies such as transcriptomics or proteomics, metabolomics does only very indirectly benefit from the increasing number of sequenced genomes. The reason for this is obvious: whereas RNA and proteins are essentially collinear with the information on the DNA level, this is not true for metabolites. Thus the increase in knowledge about the chemical structure or composition of metabolites detected by any profiling technology is still vastly lagging behind the progress made in the fields of proteomics or transcriptomics. A further problem specific to metabolite analysis is the fact that small molecules (metabolites in the mass range between 50 and 2000 Da) are unavoidably present in any reagent or physical entity used during metabolite extraction and sample preparation, thus resulting in some ambiguity concerning the question of the origin of the compound detected, i.e., biological or contaminant. So far two main technologies, namely, nuclear magnetic resonance (NMR) spectroscopy and mass spectrometry (MS), have been employed in the field of metabolomics research.19-24 Compared to MS-based approaches, NMR requires very limited (8) Rabinowitz, J. D. Expert Rev. Proteomics 2007, 4, 187–198. (9) Hinkelbein, J.; Feldmann, R. E., Jr.; Peterka, A.; Schubert, C.; Schelshorn, D.; Maurer, M. H.; Kalenka, A. Curr. Neurovasc. Res. 2007, 4, 280–288. (10) Bhalla, R.; Narasimhan, K.; Swarup, S. Plant Cell Rep. 2005, 24, 562–571. (11) Dixon, R. A.; Gang, D. R.; Charlton, A. J.; Fiehn, O.; Kuiper, H. A.; Reynolds, T. L.; Tjeerdema, R. S.; Jeffery, E. H.; German, J. B.; Ridley, W. P.; Seiber, J. N. J. Agric. Food Chem. 2006, 54, 8984–8994. (12) Keurentjes, J. J.; Fu, J.; de Vos, C. H.; Lommen, A.; Hall, R. D.; Bino, R. J.; van der Plas, L. H.; Jansen, R. C.; Vreugdenhil, D.; Koornneef, M. Nat. Genet. 2006, 38, 842–849. (13) Meyer, R. C.; Steinfath, M.; Lisec, J.; Becher, M.; Witucka-Wall, H.; Torjek, O.; Fiehn, O.; Eckardt, A.; Willmitzer, L.; Selbig, J.; Altmann, T. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 4759–4764. (14) Schnackenberg, L. K. Expert Rev. Mol. Diagn. 2007, 7, 247–259. (15) van der Greef, J.; Hankemeier, T.; McBurney, R. N. Pharmacogenomics 2006, 7, 1087–1094. (16) Fernie, A. R.; Trethewey, R. N.; Krotzky, A. J.; Willmitzer, L. Nat. Rev. Mol. Cell. Biol. 2004, 5, 763–769. (17) Weckwerth, W. Annu. Rev. Plant Biol. 2003, 54, 669–689. (18) Nicholson, J. K.; Wilson, I. D. Nat. Rev. Drug Discovery 2003, 2, 668–676. (19) Pan, Z.; Raftery, D. Anal. Bioanal. Chem. 2007, 387, 525–527. (20) Dettmer, K.; Aronov, P. A.; Hammock, B. D. Mass Spectrom. Rev. 2007, 26, 51–78. (21) Glinski, M.; Weckwerth, W. Mass Spectrom. Rev. 2006, 25, 173–214.
Analytical Chemistry, Vol. 80, No. 24, December 15, 2008
9417
sample preparation and offers potential structural and quantitative annotation of the measured compound. The main problem with respect to metabolomics using NMR remains its low sensitivity if applied to highly complex biological samples.20,22,25,26 As a consequence, MS-based approaches with their higher sensitivity have begun to dominate the metabolomic literature. In the field of MS-based metabolomics, mostly hyphenated technologies, where a separation technique such as gas chromatography, capillary electrophoresis, or liquid chromatography (LC) is coupled to different kinds of mass spectrometers20-22 are used. The combination of chromatographic separation and mass spectrometric techniques is beneficial for many aspects of data acquisition such as the separation of isobaric compounds and the accurate quantification of chromatographic peaks.27-29 Furthermore, the chromatographic sample separation, prior to the MS analysis, helps to reduce disturbing ion suppression effects.30-33 Still, even though these chromatography-based metabolite analysis strategies offer many advantages compared to nonchromatography-based strategies, the major drawback of these hyphenated approaches lies in their limitation to screen a larger set of samples for changes in metabolite composition in a fast and simple way. For this purpose, direct infusion-based metabolite fingerprinting strategies present a very efficient tool to complement the more comprehensive chromatography-based approaches.34-39 In these fingerprinting experiments, complex samples are injected without sample separation steps, directly into the mass spectrometer, and the sum of the mass signals is used for data analysis. Statistical analysis often allows one to discriminate or differentiate organisms, tissues, or mutants from one another.20,23 However, although this fingerprinting approach provides a fast and sensitive method to get a first clue of different metabolite compositions, thus far this method lacks the ability to reach beyond pattern recognition. Thus, if low-resolution mass spectrometers are used, masses can be reliably identified but no actual metabolites.20 (22) Dunn, W. B.; Bailey, N. J.; Johnson, H. E. Analyst 2005, 130, 606–625. (23) Goodacre, R.; Vaidyanathan, S.; Dunn, W. B.; Harrigan, G. G.; Kell, D. B. Trends Biotechnol. 2004, 22, 245–252. (24) Lenz, E. M.; Wilson, I. D. J. Proteome Res. 2007, 6, 443–458. (25) Eisenreich, W.; Bacher, A. Phytochemistry 2007, 68, 2799–2815. (26) Ward, J. L.; Baker, J. M.; Beale, M. H. FEBS J. 2007, 274, 1126–1131. (27) Roy, S. M.; Becker, C. H. Methods Mol. Biol. 2007, 359, 87–105. (28) Silva, M. J.; Samandar, E.; Preau, J. L., Jr.; Reidy, J. A.; Needham, L. L.; Calafat, A. M. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 2007, 860, 106–112. (29) Freitas, L. G.; Gotz, C. W.; Ruff, M.; Singer, H. P.; Muller, S. R. J. Chromatogr., A 2004, 1028, 277–286. (30) Annesley, T. M. Clin. Chem. 2003, 49, 1041–1044. (31) Muller, C.; Schafer, P.; Stortzel, M.; Vogt, S.; Weinmann, W. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 2002, 773, 47–52. (32) Sterner, J. L.; Johnston, M. V.; Nicol, G. R.; Ridge, D. P. J. Mass Spectrom 2000, 35, 385–391. (33) Weckwerth, W.; Fiehn, O. Curr. Opin. Biotechnol. 2002, 13, 156–160. (34) Pope, G. A.; MacKenzie, D. A.; Defernez, M.; Aroso, M. A.; Fuller, L. J.; Mellon, F. A.; Dunn, W. B.; Brown, M.; Goodacre, R.; Kell, D. B.; Marvin, M. E.; Louis, E. J.; Roberts, I. N. Yeast 2007, 24, 667–679. (35) Vaidyanathan, S.; O’Hagan, S.; Goodacre, R. Rapid Commun. Mass Spectrom. 2006, 20, 21–30. (36) Scholz, M.; Gatzek, S.; Sterling, A.; Fiehn, O.; Selbig, J. Bioinformatics 2004, 20, 2447–2454. (37) Castrillo, J. I.; Hayes, A.; Mohammed, S.; Gaskell, S. J.; Oliver, S. G. Phytochemistry 2003, 62, 929–937. (38) Allen, J.; Davey, H. M.; Broadhurst, D.; Heald, J. K.; Rowland, J. J.; Oliver, S. G.; Kell, D. B. Nat. Biotechnol. 2003, 21, 692–696. (39) Goodacre, R.; Timmins, E. M.; Burton, R.; Kaderbhai, N.; Woodward, A. M.; Kell, D. B.; Rooney, P. J. Microbiology 1998, 144, 1157–1170.
9418
Analytical Chemistry, Vol. 80, No. 24, December 15, 2008
Some elegant strategies to overcome the shortcomings of unambiguous mass to molecule annotations have been developed in the field of protein/peptide analysis. So Pan et al., for example, used isotope-labeled amino acids to determine the amino acid code of peptides, measured by MALDI TOF MS.40 Another very interesting approach to determine peptide identity was employed by Shi et al.41 This group used isotope fine structure analysis by recording ultra-high-resolution FT-ICR MS spectra of peptides, allowing determination of the number of sulfur atoms present in the molecule. To close the gap between the comprehensive, chromatographybased methods and the low-resolution fingerprinting strategies in the analysis of metabolites, we developed a strategy employing high-resolution FT-ICR mass spectrometry in combination with 13 CO2-based isotope labeling of a whole metabolomes. Application of this strategy to extracts of Arabidopsis thaliana allows the annotation of more than 1000 distinct chemical sum formulas of biological origin. As the majority of these chemical sum formulas have not been represented by any compound present in specific databases of this organism (e.g., AraCyc), these findings suggest the existence of numerous novel compounds and henceforth undescribed pathways functional in this model plant. EXPERIMENTAL SECTION Chemicals. All chemicals and solvents employed were purchased from Sigma-Aldrich (Steinheim, Germany) with the highest purity grade available, unless otherwise stated. The 13CO2 (isotopic purity 99 atom % 13C,
